1,]

 

3.vr3. a} F... $712. . . , . . . ,. ‘ .4, ,4. ....4.1f.!41
: égg} . ‘ ‘ : . . a? 3%;

a

 

 

:1 LIBRARY '
a 006 Michigan State
University ]

 

 

This is to certify that the
dissertation entitled

THE EFFECTS OF CAMPYLOBACTER JEJUNI AND TRICHURIS
SUIS ON LOCAL CYTOKINE EXPRESSION IN THE SWINE
COLON

presented by

Kathryn Marie Jones

has been accepted towards fulﬁllment
of the requirements for the

Doctoral degree in Microbiology and Molecular
Genetics

 

 

 
    

I l

k l 0 " v.41 ,
Major Professor’s Sign
c/aa/a 005

Date

ure

MSU is an Afﬁrmative Action/Equal Opportunity Institution

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

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

2mm

THE EFFECTS OF CAMPYLOBACTER JEJUNI AND TRICHURIS SUIS ON
LOCAL CYTOKINE EXPRESSION IN THE SWINE COLON

By

Kathryn Marie Jones

A DISSERTATION

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

DOCTOR OF PHILOSOPHY

Department of Microbiology and Molecular Genetics

2005

ABSTRACT
THE EFFECTS OF CAMPYLOBACTER JEJUNI AND TRICHURIS SUIS ON
LOCAL CYTOKINE EXPRESSION IN THE SWINE COLON
By

Kathryn Marie Jones

Campylobacterjejuni is a signiﬁcant enteric pathogen of humans, but it is a non-
pathogenic resident in the gastrointestinal tract of pigs. Upon infection with the nematode
Trichuris suis, C. jejuni opportunistically invades the mucosa in the distal colon, resulting
in severe disease and pathology. The central hypothesis of this dissertation project is that
natural resistance to C. jejuni infection in swine is due to a proinﬂammatory response in
the intestine. A related hypothesis is that T. suis stimulates an anti-inﬂammatory response
which alters the protective response to C. jejuni. Three experimental designs were used to
test these hypotheses. In the ﬁrst experiment, weaned piglets were inoculated rectally
with C. jejuni 81-176pWM1007, then samples of the distal colon and lymphoglandular
complexes were taken for quantitation of cytokine mRNA expression by real time PCR.
The data showed that C. jejuni stimulated a proinﬂammatory response in vivo. In the
second experiment, weaned piglets were orally inoculated with embryonated T. suis eggs,
C. jejuni 33292, or both, then samples of the jejunum, proximal colon, and distal colon
were taken for quantitation of cytokine mRNA expression by real time PCR. The data
showed that C. jejuni stimulated a proinﬂammatory response in the jejunum while T. suis
stimulated an anti-inﬂammatory response in the jejunum and proximal colon. When

dually infected, T. suis down regulated proinﬂammatory responses to C. jejuni in the

jejunum and proximal colon. In the third experiment, differentiated an undifferentiated
swine intestinal epithelial cells (IPEC-l) were infected with C. jejuni 81-176 or 33292
and IL-18 production was measured in culture supematants by ELISA. Only
differentiated cells produced signiﬁcant IL-18 in response to C. jejuni. Taken together,
these data indicate that C. jejuni stimulates a local proinﬂammatory response in vivo,
which can be attributed in part to the interaction Of C. jejuni with intestinal epithelial
cells. T. suis stimulates an anti-inﬂammatory response and down regulates
proinﬂammatory cytokines, abrogating the protective response to C. jejuni in vivo. This
immunomodulation could facilitate C. jejuni invasion into intestinal epithelial cells and

deeper tissues in vivo, resulting in destruction of the epithelium and severe pathology.

Cepyright by
KATHRYN MARIE JONES
2005

Acknowledgements

There were several people who were essential to completion Of this dissertation
project and I would like to acknowledge them here. Dr. Linda Mansﬁeld was my mentor
and gave me continued guidance and encouragement throughout my graduate studies. My
committee members, Drs. John Linz, Katheryn Meek, Richard Schwartz, and Charles
Mackenzie, were always willing to Offer guidance and shared their knowledge. My
labmates, Lakeisha Cunningham, Dr. Geetha Parthasarathy, and Stacey Wilder gave me
emotional support as well as assisted me in my experiments, helped me trouble shoot my
problems, and critiqued my writing. Dr. Julia Bell and David Wilson never failed to share
their expertise on all things pertaining to Campylobacter jejuni. Drs. Pawin Padungtod,
BO Norby, and Jean Gaymer shared their veterinary expertise and technical skills in
developing and executing the animal models. Dr. Matti Kiupel performed the staining
and microscopy on the animal tissues and shared his expertise in histopathology and
immunohistochemistry. Drs. Joseph Urban and Harry Dawson were our collaborators
from the USDA and trained me in real time PCR and graciously allowed me to analyze
all of my tissue samples in their facility. Drs. John Baker and Vilma Yuzbasiyan-Gurkan
assisted me in securing funding that was essential to completing my project. Dr. Shirley
Owens trained me in confocal microscopy and shared her expertise in ﬂuorescent
microscopy. Christopher Kar, my husband, not only provided unfailing emotional support
throughout my graduate studies, but he was also my 24 hour home technical support and
made it possible for me to write this dissertation. To all of you and many more who

helped me through my degree, I am eternally grateful.

TABLE OF CONTENTS

List of Tables ....................................................................................... viii
List of Figures ......................................................................................... ix

Chapter 1: Literature Review

C ampylobacter jejuni ........................................................................ 1
Trichuris suis ............................................................................... 15
Rationale for study ......................................................................... 24
References ................................................................................... 29

Chapter 2: Campylobacterjejuni invades distal colonic lymphoglandular complexes
causing upregulation of proinﬂammatory cytokines in a swine rectal challenge model

Abstract ....................................................................................... 37
Introduction ................................................................................. 38
Materials and Methods ..................................................................... 41
Results ....................................................................................... 5 1
Discussion ................................................................................... 55
References ................................................................................... 61
Figures ....................................................................................... 64

Chapter 3: T richuris suis down regulates proinﬂammatory cytokines in the proximal
colon of pigs concurrently infected with C. jejuni

Abstract ....................................................................................... 75
Introduction ................................................................................. 77
Materials and Methods ..................................................................... 79
Results ....................................................................................... 86
Discussion ................................................................................... 90
References ................................................................................... 98
Figures ...................................................................................... 101

Chapter 4: Campylobacterjejum' stimulates IL-18 secretion from differentiated intestinal
epithelial cells

Abstract ..................................................................................... 109
Introduction ................................................................................ 1 1 1
Materials and Methods ................................................................... 115
Results ...................................................................................... 122
Discussion ................................................................................. 124
References ................................................................................. 127

vi

Figures ...................................................................................... 131

Chapter 5: Summary and conclusions
Summary and conclusions .............................................................. 138
References ................................................................................. 1 64

vii

LIST OF TABLES

Chapter 2
Table 1: Summary Of Campylobacter detection and identiﬁcation ............... 64

Table 2:Primer and probe sequences for porcine speciﬁc cytokines, chemokines,

and iNOS .................................................................................. 73
Chapter 3
Table 1: Summary of Campylobacter detection and identiﬁcation ............... 102

viii

LIST OF FIGURES

Chapter 2
Figure l: Ampliﬁcation of the QRDR of C. jejuni gyrA from distal colon DNA
from experimentally infected pigs ....................................................... 65
Figure 2: Ala I digest of Campylobacter 23s rRNA gene ampliﬁed from pig
tissues by PCR .............................................................................. 66
Figrue 3: PCR testing for gfp ............................................................ 67

Figure 4*: Immunohistochemistry of Campylobacter infected IPEC-l cells ...... 68

Figure 5*: Immunohistochemistry of colonic tissues from Campylobacter infected

pigs ........................................................................................... 69
Figure 6: PCR testing DNA isolated from LCM samples ............................ 70
Figure 7: Ala I digest of Campylobacter 23s rRNA gene PCR product ampliﬁed
from microdissected crypts and submucosa ............................................ 71
Figure 8*: Cytokine expression in distal colonic tissues .............................. 72
Chapter 3
Figure 1: Inoculation timeline .......................................................... 101
Figure 2: RFLP analysis of the Campylobacter 23S rRNA gene .................. 103
Figrue 3*: H&E stained sections of colonic tissue ................................... 104
Figure 4*: H&E stained sections of colonic tissue .................................. 105
Figure 5*: H&E sections of LGCs from infected pigs ............................... 106
Figure 6*: Immunohistochemistry from colonic sections ........................... 107
Figure 7*: Cytokine expression in intestinal tissues ................................. 108
Chapter 4

Figure 1: Scanning electron micrograph of polarized IPEC-l cells (12 days) on
transwell membranes ..................................................................... 1 3 1

ix

Figure 2: IL-18 secretion from differentiated IPEC-l cells ......................... 132

Figrue 3: IL-18 secretion from differentiated IPEC-l cells ......................... 133
Figure 4: IL-1 8 secretion from undifferentiated IPEC-l cells ...................... 134
Figure 5: Optimization of ampliﬁcation parameters .................................. 135
Figure 6: Determination of optimal number of cycles for ampliﬁcation .......... 136
Figure 7: IL-18 expression from IPEC-l cells ....................................... 137

*Images in this dissertation are presented in color

Chapter 1: Literature Review

Campylobacterjejuni

Basic Microbiology

Campylobacter jejuni is a gram negative, slender, curved rod measuring 0.2-0.5um
wide and 0.5-5 um long [1]. It is motile due to the presence of uni- or bipolar ﬂagella. It
grows best at 42°C in a microaerophilic atmosphere of 3-15% oxygen and 2-5% carbon
dioxide [2]. C. jejuni is nonsaccharolytic and requires complex growth media for
laboratory culture [2]. The genome is small, approximately 1.6 to 1.7Mbp, and AT rich
with a GC ratio of approximately 30%. In addition, extrachromosomal elements have
been discovered, including conjugative plasmids and bacteriophages [2]. C. jejuni is
naturally transforrnable, able to take up DNA from the environment and incorporate it
into its genome [2-4]. Several genetic tools have been used to identify variants within the
species, including pulse ﬁeld gel electrophoresis, restriction fragment length
polymorphism typing of ribosomal and ﬂagellar genes, and ampliﬁed fragment length
polymorphism analysis [5-9]. These methods have revealed extensive genetic diversity

within the species.

Clinical Disease

C. jejuni is a signiﬁcant enteric pathogen of humans and is the leading cause of
foodborne bacterial illness worldwide. Ingestion of as few as 500 organisms can lead to
diarrhea] disease in susceptible humans [1]. Symptoms develop 1 to 7 days after ingestion

of bacteria, with an average 24 to 48 hour incubation period [10]. Clinically, fever,

abdominal pain, and general malaise precede diarrhea] disease. Diarrhea is profuse and
can range from watery to mucoid and bloody in character [11]. C. jejuni colonizes the
mucus overlying the epithelium of the distal ileum and colon [1] and colonic biopsies
from infected individuals show an acute inﬂammatory response with inﬁltration of the
epithelium and lamina propria with neutrophils and mononuclear cells [10]. Histologic
examination of colonic tissues from experimentally infected colostrum deprived piglets
that developed diarrhea] disease after C. jejuni inoculation showed ulcerative colitis with
denuded villus tips as well as inﬁltration of the submucosa with neutrophils [12]. C.
jejuni disrupts the absorptive capacity of epithelial cells by invasion, by the effect of
toxins, or through induction of an inﬂammatory response [2]. Gastrointestinal illness due
to C. jejuni infection usually resolves aﬂer an average of 10 days [1], but the bacteria can
be excreted in the feces for up to 16 days [10]. Due to the fact that illness is usually self
limiting, treatment is generally restricted to supportive care with ﬂuid therapy. However,
erythromycin is the drug of choice for severe cases or immunosuppressed individuals
[11].

The most common clinical manifestation of C. jejuni infection is diarrhea]
disease, however, in rare cases extraintestinal complications arise. Local spread of the
bacteria can induce cholecystitis, pancreatitis, peritonitis, and massive GI hemorrhage.
Systemic spread can cause meningitis, endocarditis, septic arthritis, osteomyelitis, and
neonatal sepsis [11]. One notable sequela of C. jejuni infection is Guillain-Barré
syndrome (CBS). CBS develops in 1 in every 1000 cases of C. jejuni infection,
developing approximately 1-3 weeks after the onset of diarrheal illness. CBS is an

ascending peripheral neuropathy that is most commonly associated with Penner serotype

0:19 strains in the US, and type 0:41 in South Africa [1], 13]. Molecular mimicry
between C. jejuni LPS and human gangliosides results in the generation of antibodies that
attack the myelin sheaths of peripheral nerves, causing nerve paralysis [11, 13]. The
combination of C. jejuni causing signiﬁcant diarrhea] disease and the debilitating

sequelae, though rare, warrants intense investigation of this pathogen.

Epidemiology

Epidemiologic studies have showed a high prevalence of C. jejuni in cases of
infectious diarrhea. Diarrhea] disease due to C. jejuni in the U. S. is more prevalent than
diarrhea] disease due to Salmonella, Shigella, and E. coli 0157:H7 [11]. The CDC
estimates 2.4 million cases of campylobacteriosis each year in the US [11]. Sporadic
illnesses account for the majority of reported cases and are generally attributed to
consumption of improperly prepared poultry products or other foods contaminated with
raw poultry ﬂuids [1]. Seasonal outbreaks occur with bimodal peaks in May and October,
and are usually due to consumption of contaminated raw milk or water sources
contaminated with animal feces [1]. Children and young adults, particularly young males,
are most commonly infected, with most deaths occurring in infants, elderly, and
immunosuppressed individuals [1]. The preferred environment for C. jejuni is the
intestinal tract of warm blooded animals. It is a commensa] in a variety of wild and
domestic mammals and birds, including chickens, cattle, sheep, and swine, and animals
can serve as a source of infection for humans [14, 15].

The clinical picture and epidemiology of C. jejuni infections is different in

developing countries. Diarrhea] disease due to C. jejuni in developing countries is usually

less severe than in developed countries and can range from watery diarrhea, without
blood or inﬂammation, to severe inﬂammatory, bloody diarrhea accompanied by fever
and abdominal pain [16]. Campylobacteriosis does not Show seasonal peaks in
developing countries and this could be attributed to climate variations, or to inadequate
surveillance for epidemics [16]. In developing countries, Campylobacteriosis is generally
a pediatric disease being the most frequently cause of diarrhea] disease in infants less
than 6 months of age [16]. In community based studies 40-60% of diarrheic children less
than 5 years of age were positive for the organisms on fecal cultures [17]. Additionally,
polymicrobial infection is common, with protozoa, viruses, and other enteric bacteria
coinfecting individuals [18, 19]. Sources for infection are commonly contact with

animals and environmental contamination from animal feces [17, 20, 21].

Virulence Factors

C. jejuni causes diarrhea] disease in humans by colonizing the intestinal tract and
invading the epithelium which results in decreased absorption due to loss of epithelial
function [2, 10]. The events that lead to invasion of the epithelium by C. jejuni include
colonization of the mucus layer overlying the epithelium, adherence to epithelial cells,
and invasion of epithelial cells. Several virulence factors have been described that play a

role in the pathogenesis of C. jejuni infection.

Motility:
C. jejuni are motile by means of one or two polar ﬂagella [2]. The ﬂagellin

protein is encoded by two adjacent genes on the chromosome designated ﬂaA and ﬂaB

[2]. Disruption of ﬂaB results in mutants that have ﬂagella of norrna] length, exhibit
normal motility, and invade INT-407 cells at wild type levels [22]. In contrast, disruption
of the ﬂaA gene results in mutants that are aﬂagellate or have truncated ﬂagella and are
unable to adhere to or invade epithelial cells [22, 23], or colonize the ceca of chicks [24].
Mutants with paralyzed ﬂagella are non motile and non-invasive, but are able to adhere to
cells [23]. However, non motile mutants are able to invade epithelial cells in vitro if
centrifugation is performed to bring the bacteria into contact with epithelial cells [22].
These data suggest that motility is necessary for invasion [22, 23].

C. jejuni motility is also affected by environmental factors. C. jejuni display
normal swarming ability on motility agar of neutral and basic pH (pH=7.3 and 8.5,
respectively). However, in an acidic pH of 5.0, swarming ability is decreased despite
normal morphology. C. jejuni displays smooth swimming patterns with longer paths and
greater velocity in viscous medium that mimics the viscosity of intestinal mucous.
Increased viscosity of culture medium also increases adherence to and invasion of Caco-2
cells [25]. The motility of C. jejuni is believed to allow penetration of the intestinal

mucous layer in vivo and successful colonization of the host [2].

Iron acquisition:

C. jejuni requires iron for optimal growth and prOper cell division. Under iron
limiting conditions, C. jejuni produces ﬁve proteins ranging in size from 52kDa to
89kDa. Three of these proteins with molecular weights of 74kDa, 76kDa, and 82kDa
proteins are surface expressed. Cjejum' is also capable of producing siderophores and of

using the exogenous siderophores enterochelin and ferrichrome for iron acquisition [26].

C. jejuni has evolved mechanisms for obtaining iron from its environment
necessary for growth. However, in the presence of oxygen iron is able to generate oxygen
radicals that can cause damage to DNA, proteins, and membrane lipids [27]. Therefore C.
jejuni has also evolved mechanisms to survive in conditions of environemental stress.
Alkyl hydroperoxide reductase C (AhpC) is transcriptionally regulated by iron
concentrations and has been shown to be involved in resistance to oxidative stress caused
by cumene hydroperoxide (CHP) and survival in atmospheric oxygen [28]. C. jejuni also
possesses two Fur homolog iron responsive elements, designated fur and perR. PerR is a
repressor of ahpC and katA, and perR mutants Show hyperresistance to CHP and
hydrogen peroxide [29]. An additional protein, designated Dps, has been identiﬁed in C.
jejuni. Dps of C. jejuni is structurally similar to the DNA-binding protein from starved
cells family of proteins. Dps does not bind DNA, but it does bind to iron. Dps confers

resistance to hydrogen peroxide [27].

Chemotaxis:

In vivo, C. jejuni colonizes the mucous blanket overlying the epithelium of the
intestines [30, 31]. This allows them to associate closely with intestinal epithelial cells
[31]. Chemotactic movement is one mechanism by which the bacteria reaches its target
niche and has shown to be essential for colonization in vivo [32]. In vitro studies have
identiﬁed several compounds found in the gastrointestinal environment that affect
chemotaxis of C. jejuni. Mucin, the major component of intestinal mucous and a
component of bile, attracts C. jejuni [33]. Amino acids and organinc acids, including L-

aspartate, L-glutamate, L-serine, L-fucose, pyruvate, succinate, fumarate, citrate, L-

malate, and OI-ketoglutarate also attract C. jejuni [32, 33]. These compounds as well as
low pH and high osmolarity have been shown in vitro to upregulate the ﬂaA promoter
028, while increased viscosity causes downregulation [34]. It is hypothesized that
increased production of FlaA in response to chemoattractants would facilitate
maintenance of ﬂagellar ﬁlaments, assisting in actions necessary for survival in the host.
These actions would include escaping the acid pH of the stomach and chemotaxis toward
mucin or amino acids that could be used as energy sources. Decreased motility in a

viscous environment simulating the mucous blanket overlying the epithelium would be

advantageous because it would signal that the desired niche had been reached [34].

Adherence:

Once C. jejuni successfully traverse the intestinal mucus layer and locate the
epithelial cells, they must adhere in order to enter the cells and translocate to deeper
tissues. Several factors have been identiﬁed that mediate adherence.

C. jejuni come into close apposition with the cell surface, particularly at the
intercellular spaces [35]. Fibronectin is a component of the basement membrane
underlying epithelial cells and is also present in regions of cell to cell contact in the GI
epithelium [36]. C. jejuni binds to the glycoprotein ﬁbronectin via a 37kDa surface
exposed outer membrane protein designated CadF [37]. CadF also mediates adherence to
intestinal epithelial cells in vitro, as mutation of CadF reduces adherence by 59% [36].
CadF has also been shown to be important for C. jejuni in vivo as it is necessary for
successful colonization of chicken ceca [38]. le is a surface lipoprotein that interacts

with Heat shock protein 90a, mediating adherence to HEp-2 epithelial cells. This binding

leads to activation of the transcription factors NF -kB and the signaling molecule p38
MAP kinase [39, 40]. The pebIA locus encodes a previously described protein, Cell
Binding Factor 1 (CBF -1) that mediates adherence to HeLa cells. Disruption of this locus
reduces adherence in vitro by 15 fold, and reduces colonization of mice intestines [41].
Flagella mediate adherence to intestinal epithelial cells in vitro. Aﬂagellated bacteria, due
to mechanical removal by shearing or by mutation, have decreased adherence to INT-407

cells [42].

Invasion:

Bacterial invasion into epithelial cells is proposed to contribute to the disease and
pathology caused by C. jejuni infection [2]. C. jejuni has been shown to invade intestinal
epithelial cells in vitro and in vivo [12, 35, 43, 44]. Studies have indicated a role for both
microﬁlarnents and microtubules in the invasion process. Oelschleger and colleagues
found that C. jejuni invasion into INT-407 cells is a microtubule-dependent endocytic
process that results in uptake into endosoma] vacuoles. Intemalization requires de novo
protein synthesis for the bacteria, but not for the host cell [35]. Hu and Kopecko also
showed that microtubules played a role in invasion as depolymerization inhibited C.
jejuni internalization into intestinal epithelial cells. Microtubule-based membrane
extensions were the initial point of contact between C. jejuni and the host cell. Once
internalized, the bacteria associated with microtubules and then migrated to the
perinuclear area via association with dynein [45]. Biswas and colleagues analyzed

internalization of several human and animal isolates of C. jejuni and found that

depolymerization of both microﬁlaments and microtubules inhibited invasion into INT-
407 cells depending on the strain tested [46].

In addition to host factors involved in invasion, bacteria] factors involved in
invasion have been identiﬁed. C. jejuni produces proteins that are essential for
internalization. Campylobacter invasion antigen (Cia) proteins are secreted by C. jejuni in
response to physiologic stimuli, including serum, bile salts, extracellular matrix
components, and contact with intestinal epithelial cells in vitro [47, 48]. Eight different
proteins have been detected in culture supematants. One of them, designated CiaB, is a
73kDa protein that is necessary for internalization of bacteria into TNT-407 cells. In
addition, CiaB itself is translocated into the host cell upon binding of the bacteria [47].

C. jejuni has been shown to encode virulence determinants on extrachromosomal
elements. Four virulence genes have been detected on a 35 kb plasmid, designated pVir,
found in some clinical isolates of C. jejuni. All four genes are homologous to type IV
secretion system components. Three genes, designated comBI, comBZ, and comB3,
display signiﬁcant identity to Helicobacter pylori genes involved in competence for
natural transformation and DNA uptake. The fourth gene displays identity to a type IV
secretion system component of Agrobacterium designated virBI 1. Mutation of comB3
reduces adherence and invasion of INT-407 cells, and decreases frequency of natural
transformation. Mutation of virBI I also reduced adherence and invasion in vitro and

reduced virulence in the ferret diarrhea] disease model [49].

Intracellular survival:

Intracellular survival of C. jejuni has been demonstrated in intestinal epithelial
cells and macrophages [35, 50, 51]. The bacteria can survive within macrophages for up
to 7 days [50]. Two genes have been identiﬁed to be necessary for survival within cells.
The gene katA encodes catalase, which confers resistance to hydrogen peroxide in vitro.
Mutation of katA does not affect intracellular survival in HEp-2 cells, but it does
eliminate recovery from peritoneal macrophages 72 hours after infection [52].
Superoxide dismutase serves to defend bacterial cells against oxidative damage by
catalyzing the breakdown of superoxide radicals to hydrogen peroxide and dioxygen. In
C. jejuni, a single gene, designated sodB, encodes this enzyme and has been shown to be
required for survival of invasion into INT-407 cells. Mutation of sodB decreases

intracellular survival 12 fold [51].

Lipopolysaccharide:

Lipopolysaccharide (LPS) is a constituent of the outer membrane of most gram
negative bacteria that is necessary for membrane integrity and function. It is composed of
3 components: the lipid A molecule that is embedded in the membrane, the
Oligosaccharide core, and the 0 chain. The lipid A molecule contains the endotoxic
activity of LPS; the core oligosaccharide is involved in immunomodulation, and the 0
chain is highly variable and confers antigenicity to the molecule [53]. C. jejuni strains
produce 2 types of LPS, one that contains all the components and one that lacks the 0
chain and is Similar to the lipooligosaccharides (LOS) of Neisseria spp. and Haemophilus

spp. LOS [53,54].

10

LPS and LOS function in C. jejuni virulence in vitro and in vivo. Carbohydrate
moieties of LPS mediate adherence to INT-407 and intestinal mucous in vitro [42]. GalE
is a UDP-glucose epimerase that catalyzes the interconversion between UDP-glucose and
UDPgalactose, which is a component of C. jejuni LPS. Mutation of galE inhibits
adherence and invasion of INT-407 cells and natural transformation efﬁciency [55].
Sialic acid moieties of C. jejuni LOS confer resistance to complement killing in vitro.
Mutation of neuC, a gene involved in the biosynthesis of the Sialic acid N-
acetylneuraminic acid (NeuNAc), results in loss of NeuNAc and increased serum
sensitivity of C. jejuni [56]. Carbohydrate moieties of LOS can also mimic human
gangliosides and play a role in the development of autoimmune disorders. The gene cgtA
encodes a N-Acetylgalactosaminyl transferase that is subject to phase variable expression
due to slip-strand mispairing in a homopolymeric G tract during DNA replication. This
gene plays a role in structural variation in the LOS of C. jejuni 81-176, which can rrrimic
GM; or GM3 gangliosides [57]. C. jejuni NCTC 11168 also has phase variable LOS that
mimics gangliosides GM; or GM; depending on expression of the wlaN gene. This gene
encodes a [3-1,3 galactosyltransferase and is also subject to slip-strand mispairing phase
variation , similar to cgtA [58]. In vivo, antibodies directed against C. jejuni LOS
molecules that mimic gangliosides can cause autoimmune mediated neuropathy. Rabbits
immunized with C. jejuni LOS develop nerve degeneration and paralysis which
resembles the clinical signs of GBS in humans [59]. Serum IgG from these rabbits reacts
with both GM1 ganglioside and LOS in vitro, indicating a direct role for ganglioside

mimicry by C. jejuni LOS in the development of autoimmune disease [59]. Taken

1]

together, these studies have demonstrated several roles for LOS in the pathogenesis of C.

jejuni infection.

Toxins:

Toxins have been identiﬁed from C. jejuni and are proposed to ﬁmction in cellular
destruction and clinical disease. An enterotoxin with immunologic similarity to cholera
toxin and the heat labile toxin of E. coli has been described in C. jejuni. Enterotoxin
causes elongation of Chinese hamster ovary (CHO) cells, increased intracellular cAMP,
and ﬂuid accumulation in a rat ligated intestinal loop model [60, 61]. However,
contradictory reports of researchers unable to replicate these results have confounded this
evidence [62, 63]. Cytotoxic activity has also been described by assessing rounding of
VerO cells or detachment of cells from the growth plate [60, 64]. Cytotoxic activity has
been detected in isolates from human patients with invasive diarrhea [64] and from
animal isolates [60]. Additionally, cytotoxic activity was detected from poultry isolates
tested on chicken lymphocytes using a chromium release assay [65]. Further
characterization of enterotoxins and cytotoxins is necessary to elucidate their speciﬁc

roles in clinical disease.

Perhaps the most characterized C. jejuni toxin to date is the cytolethal distending
toxin. Cytolethal distending toxin (CDT) causes distension and cell disintegration of a
variety of cell types in vitro [66-68]. In C. jejuni, the CDT holotoxin is composed of 3
subunits, Cth, CdtB, and CdtC which are produced from adjacent genes. The individual
subunit proteins are approximately 30kDa, 29kDa, and 21kDa, respectively. All three

genes are required for toxic activity [69]. The Cth and CdtC subunits are responsible for

binding to the host cell, while CdtB enters the host cell and exerts toxic effects [67]. CdtB
has signiﬁcant similarities to DNAse I enzymes and is believed to cause limited damage
to DNA resulting in arrest of the cell cycle in the G2 phase [70]. The arrest is mediated
by inhibition of CDC2, a kinase subunit required for entry into the M phase of the cell
cycle. This arrest results in cell enlargement, nuclear disintegration, and ultimately cell

death [68].

A few studies have suggested a function for CDT in induction of an immune
response. Mutant C. jejuni lacking the cdtB gene are capable of colonizing NF-kB
deﬁcient mice and producing enteritis, but the disease is less severe than that seen from
the wild type C. jejuni. In addition, serum IgG2a responses in mice infected with the
mutant were lower than mice infected with the wild type strain [7]]. Attenuated disease
and pathology with a CdtB mutant has also been demonstrated in a piglet model of
infection (Mansﬁeld et al, unpublished results). In vitro, CDT stimulates production of
lL-8 from intestinal epithelial cells [72]. These data support the hypothesis that CDT

plays a role in development of disease and stimulating an immune response in vivo.

Immune responses

Humoral immunity is necessary for clearance of C. jejuni from the intestine and
resolution of clinical disease. In human infections, C. jejuni stimulates secretion of IgA in
feces, urine, saliva, and mammary glands [73]. Flagellin is a strong stimulant for
antibody production, as evidenced by the fact that a large proportion of anti-C. jejuni IgG
antibodies found in convalescent serum is directed against ﬂagellin [73]. The relatively

low incidence Of disease in adults from endemic areas of developing countries is believed

l3

to be due to strong humoral immunity stimulated by repeated exposure during childhood
[16, 74, 75]. In a swine model of infection, lymphoglandular complexes have been
discovered to be a source of IgA in C. jejuni infection. Twenty seven days after oral
inoculation with a low dose of C. jejuni, B cell rich germinal centers develop in the LGCs
that stain positively for IgA and C. jejuni [76]. These studies suggest that the adaptive
immune response to C. jejuni is directed by Th 2 cytokines.

Several studies have also demonstrated a role for cytokines in C. jejuni infections.
Humans with infectious diarrhea caused by C. jejuni have increased levels Of nitric oxide
gas and increased secretion of IL-1 B in the stool during the acute phase of infection [77].
Intestinal colonization of C. jejuni infected mice is signiﬁcantly reduced when they are
given recombinant IL-5 or IL-6 prior to infection. In addition, recombinant IL—6 enhances
speciﬁc intestinal and systemic IgA production [78]. In vitro, secretion of IL-lB, IL-6,
IL-8, and TNFO. is stimulated from intestinal epithelial cells after infection with C. jejuni
[79] (Cunningham et al, unpublished results; Parthasarathy et al, unpublished results). In
addition, intracellular levels of IL-4, IL-10, IFNy, and TNFOI are increased in C. jejuni
infected intestinal epithelial cells [80]. Cultured human monocytes secrete IL-lB, IL-6,
IL-8, and TNFOI upon infection, and NF-KB is translocated to the nucleus [81]. Similar to
mammalian cells, chick kidney cells and avian macrophages have increased expression Of
iNOS with increased release of nitric oxide, as well as increased expression of IL-1 [3, IL-
6, IL-8 [82]. These in vitro studies suggest an important role for proinﬂammatory

cytokines in host resistance to C. jejuni infection.

T rich uris suis

T. suis life cycle

T richuris suis is a nematode parasite that resides in the cecum and colon of pigs
and it is closely related to T. muris, which infects mice, and T. trichuria which infects
humans. Morphologically it has a slender anterior end that is partially embedded in the
intestinal epithelium, and a thicker posterior end that is free in the lumen. Truchuris
worms are commonly referred to as whipworrns due to their whip like appearance. There
are 5 stages in the life cycle of T. suis from egg to mature adult. Non-infective barrel-
shaped, tan eggs with bipolar plugs are shed from the feces of infected pigs. Infective L1
larvae develop, or embryonate, within the egg in the environment. In vitro studies have
shown that eggs embryonate in as little as 19 days when cultured at 34°C. Embryonated
eggs are ingested by the host and hatch throughout the small intestine, cecum, and colon.
Hatched larvae have been detected in the distal small intestine as early as 9 hours after
oral infection. The larvae burrow into the mucosa via the crypts of Lieberkilhn and
invade the epithelial and goblet cells lining the crypts [83]. As the larvae grow they also
invade cells in the lamina propria. The ﬁrst moult, or shedding of the cuticle, begins
approximately 10 days after infection as the larvae enter the L2 stage. The larvae
continue to develop within the crypt cells and begin to migrate through the lamina propria
towards the surface mucosa] epithelium [83, 84]. The second moult to the L3 stage begins
16 days after infection and the tail begins to protrude into the lumen of the colon[84].
Throughout the moults the larvae are increasing in size and development of the

gastrointestinal and reproductive tracts can be observed. The third moult to the L4 stage

15

begins 20 days after infection. On day 32, larvae begin to undergo the last moult to the
L5 or adult stage. Sexual maturation completes and fully formed eggs can be detected in
the female uterus, and in the feces of the host, by 41 days, allowing the life cycle to

commence again [83, 84].

Prevalence

Recent studies have found a relatively low prevalence of T. suis in swine herds. T.
suis was detected in 5.6% of pigs from intensive pig farms in Guangdong Province,
China that do not use a strategic parasite control [85]. In Munsterland, Germany 8% of
sows sampled from a 144 breeding farms were found to have T. suis ova in their feces.
These farms routinely used anthelminthics in sows [86]. A study of Danish organic
swine herds detected T. suis from 4% of weaners, 13% of fatteners, and <1% of sows
[87]. Examination of a feral swine population in the US found that 10% of the pigs
harbored T. suis [88]. Despite the apparently low prevalence of T. suis in swine herds,
economic losses due to poor growth and death warrant further study of disease caused by

these nematodes [89].

Mucohemorrhagic diarrhea

Mucohemorrhagic diarrhea is a signiﬁcant health problem in the swine industry
that can result from infections with a variety of bacteria and parasitic worms. Treponema
hyodysenteriae, Salmonella typhimurium, Salmonella typhisuis, T richuris suis, and
Campylobacter spp. have all been found as causative agents. However, in many cases

two or more agents act synergistically, causing signiﬁcant enteric pathology [90, 9]].

l6

T richuris suis infection causes a disease commonly known as “21 day scours” due to the
fact that diarrhea] disease commonly occurs approximately 3 weeks after pigs are placed
in an area contaminated with infective T. suis eggs [90]. Diarrhea] disease associated with
severe infection is profuse, reddish-brown, mucoid, and bloody. In addition to diarrhea,
pigs become anorexic, anemic, and can have signiﬁcant weight loss. Penetration of the
nematode into the mucosa of the cecum and proximal colon causes inﬂammation,
hemorrhage, and in severe cases necrosis [44, 89-91]. T. suis was once believed to be
innocuous, but further study of these nematodes has revealed that they are an important
pathogen in the swine industry.

Histological lesions caused by T. suis infection are quite profound. Infection
induces thickening of all layers of the intestine, sloughing of degenerate epithelial cells,
and dilation of crypts with mucous. The lamina propria and submucosa become
inﬁltrated by inﬂammatory cells, including neutrophils, lymphocytes, plasma cells and
eosinophils [44, 89, 91]. In the distal colon, lymphoglandular complexes become
enlarged by increased numbers of lymphocytes and macrophages, and several species of
bacteria can be isolated from abscessed LGCs [44, 76, 89, 91]. Microscopic studies have
shown that T. suis causes signiﬁcant pathology both in the proximal colon where the
worms reside and in the distal colon far from sites of worm attachment.

Disease and pathology caused by T. suis infection is exacerbated by interaction
with the microbial component of the swine colon. In 1975 Rutter and Beer demonstrated,
through infections of conventionally reared, speciﬁc pathogen free and gnotobiotic pigs,
that some component of the normal gut ﬂora of pigs acted synergistically with T. suis to

cause severe mucohemorrhagic enteritis [92]. Further studies by Mansﬁeld and Urban

conﬁrmed Rutter and Beer’s ﬁndings, also showing that severe pathology caused by
secondary bacterial infection was ameliorated by antibiotic treatment [91]. Additionally,
fenbendazole treatment of both naturally and experimentally infected pigs removes all
adult T. suis and reverses gross pathology in 7 days (Mansﬁeld et al., unpublished
results). These studies demonstrate the complex nature of the polymicrobial infection

involving T. suis that results in severe disease.

T. suis excretory/secretory product (ESP)

The pathologic effects of T. suis infection are mediated both by direct interaction
of the worm with host tissues and through the actions of secreted worm products. Adult
T. suis produce an excretory/secretory product (ESP) composed of secreted products that
function in invasion and feeding in vivo, and excreted products that are byproducts of its
metabolism. Crude ESP can be collected from culture media of adult T. suis maintained
in vitro after removal from infected pigs [93] and several components have been isolated
and partially characterized. A 45kDa zinc metalloprotease was localized by
immunocytochemica] staining to the stichosome. The protease displayed proteolytic
activity against ﬁbrinogen and elastin and is proposed to have a role in feeding and tissue
penetration [93]. Phenol oxidase was localized to the reproductive tract of the female
worms and caused brown pigmentation of the worms in an oxygenated environment.
Phenol oxidase is proposed to function in tanning of the eggs [94, 95]. Thiol protease
activity, localized to the gut of the worm, is hypothesized to function in digestion and
absorption of nutrients [96]. Serine protease inhibitors were isolated from culture ﬂuids

and soluble worm extracts. These protease inhibitors showed activity against trypsin,

chymotrypsin, and elastase, and may ﬁmction in down-regulating host mast cell, and
other immune cell, effector mechanisms [97, 98]. Recently, antimicrobial activity which
inhibits the grth of C. jejuni, E. coli, and S. aureus in vitro in a dose-dependent
manner has been described and may play a role in defending the worm against bacteria in
the intestinal environment [99]. A 20kDa glycoprotein isolated from culture media was
detected in a Western blot assay using serum from T. suis infected pigs. Serum from
uninfected pigs, or pigs infected with the parasites 0esophagostomum dentatum,
T richuris spiralis, Ascaris suum, or T oxoplasma gondii did not cross react, indicating that
the 20kDa protein was speciﬁc for T. suis and could be used in diagnostic tests for
infection [100].

In addition to identifying components of ESP, two recent studies have described
its effects on host cells. Treatment of intestinal epithelial cells with crude ESP causes a
dose dependent cytotoxic effect, and decreased transepithelial electrical resistance of
differentiated cells. T. suis ESP treatment of swine intestinal epithelial cells prior to
infection with C. jejuni decreases invasion, most likely due in part to its antibacterial
activity [101]. Also, ESP has been shown to induce secretion of lL-6 and IL-10 from
both differentiated and undifferentiated swine intestinal epithelial cells (Parthasarathy et
al, unpublished results). These studies suggest a potential role for ESP in

immunomodulation and the development of pathology in vivo.

Immunity
Preliminary studies have demonstrated that T. suis infection induces a strong

immune response in conventionally reared pigs. Humoral immunity develops during

19

primary infection with the development of antibodies directed against a 20kDa
glycoprotein isolated from in vitro culture ﬂuids of adult worms. These T. suis speciﬁc
antibodies were detected in serum from pigs 21 days after natural or experimental
infection [100]. In addition, expression of IL-10 mRNA is increased 10 fold over
uninfected controls in the mesenteric lymph nodes of pigs experimentally infected with T.
suis. Pigs infected on a contaminated dirt lot have evidence of secondary bacterial
infection in the distal colon and expression of both IL-10 and IL-12p40 are signiﬁcantly
increased in draining mesenteric lymph nodes compared to uninfected controls [102].
Pigs exposed to low level trickle infections with T. suis eggs over 4 weeks have
decreased worm survival and worm fecundity during secondary challenge with 4000
infective eggs [103]. Taken together, these data suggest that T. suis stimulates a Th 2
directed immune response and that protective immunity can be achieved through low
level exposure. Additional research is necessary to further characterize the T. suis
stimulated immune response.

Studies in humans have also described a Th 2 directed immune response to the
human whipworm, T. trichuria. People that live in areas endemic for whipworm infection
have antigen speciﬁc serum antibodies that are predominantly of the IgG] and IgG4
isotypes, with lower levels of IgG2 and IgG3. Antigen speciﬁc IgA and IgE are also
present, with higher levels of IgE in less heavily infected individuals [104-106]. Whole
blood cultures stimulated with T. trichuris ESP antigens or whole worm extracts secrete
lL-4, IL-9, IL-10, IL-13, TNFOI, and to a lesser extent IFNy [104, 106]. Additionally,
whole blood cultures stimulated with whole worm extracts display a signiﬁcant

proliferative response [106]. Similar to the swine whipworm, only a few studies have

20

been conducted to characterized speciﬁc components of the immune response to this
nematode and further research is necessary.

In contrast to studies in swine and humans, extensive research in murine models
has characterized the roles of speciﬁc mediators in the immune response to Trichuris
infection. Inbred mouse strains have been developed that are either resistant or
susceptible to chronic infection with T. muris [107-110]. Resistant strains like BALB/K,
BALB/c and C57BL/6 mount a Th 2-directed immune response with increased
production of IL-4, IL-5, IL-9, IL-10, IL-13, Trichuris speciﬁc IgG], and increased
serum levels of IgE [107, 109, 111, 112]. These strains develop signiﬁcantly greater
tissue eosinophilia upon infection than susceptible mice and are able to expel the larval
stages of T. muris between 14 and 21 days after infection [107, 112]. Susceptible mouse
strains like AKR and B10.BR mount a Th 1 directed immune response with increased
production of IFNy and high levels of T richuris speciﬁc IgG2a as well as IgGl. These
strains are unable to expel the larval stages of T. muris and become chronically infected
[107, 111, 112]. These dichotomous strains have allowed researchers to deﬁne the roles
of speciﬁc immune mediators in the disease process.

Alteration of the polarized immune responses Of inbred mouse strains to T. muris
infection through administration of stimulatory antigens or recombinant cytokines have
elucidated the roles of speciﬁc cytokines in the immune response. Inoculation of
susceptible AKR mice with Schistosoma mansoni egg antigens prior to infection with T.
muris results in production of IL-4, IL-5, IL-10, IgG] and IgE and subsequent worm
expulsion [113]. Administration of recombinant IL-4 to AKR mice facilitates parasite

expulsion while blocking IL-4 activity in BALB/K mice with antibodies inhibits parasite

2]

expulsion [114]. Exogenous recombinant IL-9 causes increased intestinal mastocytosis,
increased serum IgE, and decreased worm burden in T. suis infected AKR mice [109]. In
addition, neutralization of IL-9 with antibodies decreases colonic muscle
hypercontractility and inhibits worm expulsion in normally resistant C57BL/6 mice
[110]. Exogenous recombinant IL-12 increases production of parasite speciﬁc IgG2a and
impaired worm expulsion in BALB/K mice [115]. Antibody depletion of IFNy activity in
normally susceptible AKR mice infected with T. muris results in parasite expulsion [114].

In addition to using exogenous agents, genetic manipulation of inbred mice to
abrogate production of certain cytokines has also proved to be a useful tool for studying
the effects of Th 2 cytokines in T. muris infection. IL-lO knockout mice are susceptible to
chronic infection with T. muris and have decreased IL-5 and lL-13 production, decreased
peripheral eosinophilia, and decreased intestinal mucous production [112]. IL-4
knockout mice infected with T. muris have decreased production of IL-5, IL-9, and IL-13
compared to wild type mice and are unable to generate parasite speciﬁc IgGl. IL-13
knockout mice also have impaired cytokine production initially, but by 21 days after
infection, IL-4, IL—5, and IL-9 production is at or above the levels produced by infected
wild type mice. This demonstrates an early role for IL-13 in promoting the development
of Th 2 responses stimulated by T. muris infection [116]. In addition to the roles of
speciﬁc cytokines and antigen speciﬁc antibody in T. muris infection, cell mediated
immunity is also important. SCID mice are normally susceptible to chronic T. muris
infection, but reconstitution with puriﬁed CD4+ T cells from T. muris infected BALB/c
mice resulted in expulsion of the parasite in the absence of antigen speciﬁc serum

antibodies [108]. These studies have demonstrated the critical roles of the Th 2 cytokines

22

IL-4, IL-5, IL-9, IL-10, and IL-13 in resolution of T. muris infection and helped elucidate

their speciﬁc effects in vivo.

23

Rationale for study

Campylobacter jejuni is a signiﬁcant enteric pathogen in humans causing severe
diarrhea] disease and in rare cases debilitating autoimmune disorders [2, 10, 11, 13].
Severe diarrhea] disease and colonic pathology in humans is due to invasion of the
intestine by C. jejuni with subsequent inﬂammation [2, 10]. Because pigs can become
colonized with C. jejuni within 24 hours after birth and can carry the bacteria without
clinical signs of disease, they serve as a potential source of infection for humans [15, 44].
T. suis infection of pigs harboring C. jejuni, either by natural or experimental infection,
results in mucohemorrhagic diarrhea and severe pathology that mimics the clinical
disease of Campylobacteriosis in humans [44, 91]. When T. suis is present concurrently,
C. jejuni invade epithelial cells, signiﬁcant inﬂammation develops, and
mucohemorrhagic diarrhea is evident clinically [44]. This suggests that in both human
and swine infections, bacterial invasion is the key event in the development of clinical
disease and pathology. In the case of dual infection in swine, it is believed that
immunomodulation induced by T. suis abrogates the natural resistance to C. jejuni,
allowing invasion of intestinal tissues.

Studies performed to date to elucidate the in vivo immune responses associated
with C. jejuni infection in humans and swine have focused mainly on immunoglobulin
production in serum and in lymphoid tissues [73, 76]. In vitro studies have focused
mainly on cytokine responses from cultured cells ([72, 79]; Parthasarathy et al,
unpublished results; Cunningham et al, unpublished results). The results indicate that
intestinal epithelial cells are an important source of proinﬂammatory cytokines which

may function in host resistance in vivo. Indeed, studies have shown a critical role for

24

epithelial cell derived IL-8 in control of experimental Shigellaﬂexneri infection in rabbits
[117].

Characterization of the swine immune response to T. suis is severely limited and
has focused on production of a relatively limited panel of cytokines in draining lymph
nodes and secreted in the feces ([102]; Parthasarathy et al, unpublished results;
Cunningham et al, unpublished results). Comparison of cytokine expression between
intestinal tissue at the site of infection and the draining lymph nodes in animal models of
nematode infection have shown that the cytokine proﬁles are not identical [112, 118]. T.
suis infection results in signiﬁcant pathology in the proximal colon due to direct
interaction of the worms with the intestinal mucosa, and in the distal colon due to
secreted worm products or immunomodulation [44, 91, 92]. Therefore, it is important to
study local cytokine responses in the region where the worms reside and also in distant
areas where worms are not present to determine the nature of the local immune response
and ascertain what effect those cytokines may have on the development of pathology. In
this dissertation, we have begun to characterize the swine local immune responses to C.
jejuni and T. suis in an effort to expand our knowledge of the immune response to T. suis

and to determine the nature of the immunomodulation that occurs during dual infection.

Hypothesis
The central hypothesis for this dissertation project is that a proinﬂammatory
response plays a role in swine natural resistance to C. jejuni. A related hypothesis is that a

T. suis stimulated anti-inﬂammatory response alters the protective proinﬂammatory

25

response. Thus T. suis infection would facilitate C. jejuni invasion in the colon and severe

pathology would develop.

Short term goals

The short term goals are to address the following questions:

1.

Does C. jejuni that expresses Gfp colonize the distal colons of pigs when inoculated
rectally?

Does rectally inoculated C. jejuni that expresses Gfp stimulate an immune response in
the distal colon?

What is the nature of the cytokine response in the distal colon stimulated by rectally
inoculated C. jejuni?

What is the nature of the local cytokine response in the small and large intestines Of
pigs infected orally with C. jejuni, T. suis, or both?

Is there evidence of immunomodulation in the colons of dually infected pigs?

Do epithelial cell derived cytokines play a role in swine resistance to C. jejuni

infection as determined by in vitro infection of swine intestinal epithelial cells?

In summary, the goal of this study was to determine the nature of the local cytokine

response to C. jejuni and T. suis and to determine whether cytokine dysregulation could

play a role in the pathology and disease caused by dual infection.

Chapter 2 summarizes the studies on the effects of rectal inoculation with C.

jejuni on colonization, localization of the bacteria in the distal colon, and on local

cytokine mRNA expression. Proinﬂammatory cytokines are upregulated in the intestines

26

of animals experimentally or naturally infected with enteric bacteria and they have been
shown to be critical for elimination of the bacteria and resolution of disease [117, 119-
121]. We measured mRNA expression of both proinﬂammatory and anti-inﬂammatory in
the distal colons and LGCS of pigs rectally inoculated with C. jejuni. The nature of the
cytokine response to C. jejuni in lymphoid and nonlymphoid tissue in the distal colon is
discussed.

Chapter 3 summarizes the study on the effect of oral infection with C. jejuni, T.
suis, or both on local cytokine expression in the intestines of pigs. Cytokine responses in
resistant mice infected with T. muris are predominantly Th 2 directed with increased
production of IL-4, IL-5, IL-10, and IL-13 [107, 109, 111, 112]. These cytokines can also
be characterized as anti-inﬂammatory as they downregulate proinﬂammatory cytokines
[122, 123]. We measured mRNA expression of both proinﬂammatory and anti-
inﬂammatory cytokines in the jejunum, proximal colon, and distal colons of pigs orally
inoculated with C. jejuni, T. suis, or both. The nature of the cytokine responses and the
differences in responses between the intestinal segments are discussed.

Chapter 4 summarizes the in vitro studies on proinﬂammatory cytokine
production from intestinal epithelial cells infected with C. jejuni. C. jejuni stimulates
production of IL-IB, IL-6, IL-8, and TNFOL from both swine and human intestinal
epithelial cells (Cunningham et al, unpublished results, Parthasarathy et al, unpublished
results; [72, 79, 80]. Intestinal epithelial cell derived IL-8 is critical for control of
infection in a rabbit ligated ileal loop model of Shigella ﬂexneri infection [117]. We
measured secreted lL-18 protein and IL-18 mRNA expression from C. jejuni infected

IPEC-l swine intestinal epithelial cells. The nature of the responses from differentiated

27

and undifferentiated IPEC-l cells and the role of transcriptional regulation of IL-18
production are discussed. A proposed model for epithelial cell derived IL-1 8 in resistance
to C. jejuni infection in vivo is also discussed.

Chapter 5 summarizes the previous chapters and discusses a proposed model for
disease and pathology during dual infection with C. jejuni and T. suis due to cytokine

dysregulation.

28

10.

ll.

l2.

l3.

14.

15.

16.

References

Altekruse, S.F., D.L. Swerdlow, and NJ. Stern, Microbial food borne pathogens. Campylobacter
jejuni. Vet Clin North Am Food Anim Pract, 1998. 14(1): p. 31-40.

Ketley, J.M., Pathogenesis of enteric infection by Campylobacter. Microbiology, 1997. 143 (Pt 1):
p. 5-21.

Wiesner, R.S., D.R. Hendrixson, and VJ. DiRita, Natural transformation of C ampylobacter jejuni
requires components of a type II secretion system. J Bacteriol, 2003. 185(18): p. 5408-18.

Wilson, D.L., et al., Variation of the natural transformation frequency of Campylobacter jejuni in
liquid shake culture. Microbiology, 2003. 149(Pt 12): p. 3603-15.

F ermer, C. and ED. Engvall, Speciﬁc PCR identiﬁcation and diﬂerentiation of the thermophilic
campylobacters, Campylobacterjejuni, C. coli, C. lari, and C. upsaliensis. J Clin Microbiol, 1999.
37(10): p. 3370-3.

Moore, J .E., et al., Molecular diversity of Campylobacter coli and C. jejuni isolated from pigs at
slaughter by ﬂaA-RF LP analysis and ribotyping. J Vet Med B Infect Dis Vet Public Health, 2002.
49(8): p. 388-93.

Cardarelli-Leite, P., et al., Rapid identification of Campylobacter species by restriction fragment
length polymorphism analysis of a PCR-ampliﬁed fragment of the gene coding for 1 6S rRNA. J
Clin Microbiol, 1996. 34(1): p. 62-7.

Vacher, S., et al., PCR-restrictionﬁagment length polymorphism analysis for detection of point
mutations associated with macrolide resistance in Campylobacter spp. Antimicrob Agents
Chemother, 2003. 47(3): p. 1125-8.

Wassenaar, T.M. and DC. Newell, Genotyping of Campylobacter spp. Appl Environ Microbiol,
2000. 66(1): p. 1-9.

Wassenaar, T.M. and MJ. Blaser, Pathophysiology of Campylobacter jejuni infections of humans.
Microbes Infect, 1999. 1(12): p. 1023-33.

Allos, B.M., Campylobacterjejuni Infections: update on emerging issues and trends. Clin Infect
Dis, 2001. 32(8): p. 1201-6.

Babakhani, F .K., G.A. Bradley, and LA. Joens, Newborn piglet model for campylobacteriosis.
Infect Immun, 1993. 61(8): p. 3466-75.

Goodyear, C.S., et al., Monoclonal antibodies raised against Guillain-Barre syndrome-associated
C ampylobacter jejuni lipopolysaccharides react with neuronal gangliosides and paralyze muscle-

nerve preparations. J Clin Invest, 1999. 104(6): p. 697-708.

Altekruse, S.F., et al., Food and animal sources of human Campylobacterjejuni infection. J Am
Vet Med Assoc, 1994. 204(1): p. 57-61.

Young, C.R., et al., Enteric colonisation following natural exposure to C ampylobacter in pigs.
Res Vet Sci, 2000. 68(1): p. 75-8.

Coker, A.O., et al., Human campylobacteriosis in developing countries. Emerg Infect Dis, 2002.
8(3): p. 237-44.

29

17.

l8.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

R30, M.R., et al., Pathogenicity and convalescent excretion ofCampylobacter in rural Egyptian
children. Am J Epidemiol, 2001. 154(2): p. 166-73.

Albert, M.J., et al., Case-control stuay of enteropathogens associated with childhood diarrhea in
Dhaka, Bangladesh. .1 Clin Microbiol, 1999. 37(11): p. 3458-64.

Bichile, L.S., et al., Acute Campylobacterjejuni enteritis in 385 hospitalised patients. J Assoc
Physicians India, 1992. 40(3): p. 164-6.

Adegbola, R.A., et al., Correlation between human and animal bio-serogroups of Campylobacter
isolates in Nigeria. J TrOp Med Hyg, 1990. 93(4): p. 280-3.

Coker, A.O., et al., Zoonotic infections in Nigeria: overview from a medical perspective. Acta
Trop, 2000. 76(1): p. 59-63.

Wassenaar, T.M., N.M. Bleumink-Pluym, and BA. van der Zeijst, Inactivation of Campylobacter
jejuniﬂagellin genes by homologous recombination demonstrates that ﬂaA but not ﬂaB is
required for invasion. Embo J, 1991. 10(8): p. 2055-61.

Yao, R., et al., Isolation of motile and non-motile insertional mutants of Campylobacter jejuni: the
role of motility in adherence and invasion of eukaryotic cells. Mol Microbiol, 1994. 14(5): p. 883-
93.

Nachamkin, I., X.H. Yang, and NJ. Stern, Role of Campylobacter jejuni flagella as colonization
factors for three-day-old chicks: analysis with ﬂagellar mutants. Appl Environ Microbiol, 1993.
59(5): p. 1269-73.

Szymanski, C.M., et al., Campylobacterjejuni motility and invasion of C ac0-2 cells. Infect
Immun, 1995. 63(11): p. 4295-300.

Field, L.H., et al., Inﬂuence of iron on growth, morphology, outer membrane protein composition,
and synthesis of siderophores in Campylobacterjejuni. Infect Immun, 1986. 54(1): p. 126-32.
lshikawa, T., et al., The iron-binding protein Dps confers hydrogen peroxide stress resistance to
Campylobacterjejuni. J Bacteriol, 2003. 185(3): p. 1010-7.

Baillon, M.L., et al., An iron-regulated alkyl hydroperoxide reductase (A hpC) confers
aerotolerance and oxidative stress resistance to the microaerophilic pathogen Campylobacter
jejuni. J Bacteriol, 1999. 181(16): p. 4798-804.

van Vliet, A.H., et al., Campylobacterjejuni contains two fur homologs: characterization of iron-
responsive regulation of peroxide stress defense genes by the PerR repressor. J Bacteriol, 1999.
181(20): p. 6371-6.

Beery, J.T., M.B. Hugdahl, and M.P. Doyle, Colonization of gastrointestinal tracts of chicks by
Campylobacterjejuni. Appl Environ Microbiol, 1988. 54(10): p. 2365-70.

Lee, A., et al., Mucus colonization as a determinant of pathogenicity in intestinal infection by
Campylobacterjejuni: a mouse cecal model. Infect Immun, 1986. 51(2): p. 536-46.

Takata, T., S. F uj imoto, and K. Amako, Isolation of nonchemotactic mutants of Campylobacter
jejuni and their colonization of the mouse intestinal tract. Infect Immun, 1992. 60(9): p. 3596-600.

Hugdahl, M.B., J.T. Beery, and M.P. Doyle, Chemotactic behavior of Campylobacter jejuni.
Infect Immun, 1988. 56(6): p. 1560-6.

30

34.

35.

36.

37.

38.

39.

40.

41.

42.

43.

44.

45.

46.

47.

48.

49.

50.

Allen, KJ. and M.W. Grifﬁths, Eﬂect of environmental and chemotactic stimuli on the activity of
the Campylobacterjejuni/lax! sigma(28) promoter. FEMS Microbiol Lett, 2001. 205(1): p. 43-8.

Oelschlaeger, T.A., P. Guerry, and DJ. Kopecko, Unusual microtubule-dependent endocytosis
mechanisms triggered by C ampylobacter jejuni and Citrobacterﬁ'eundii. Proc Natl Acad Sci U S
A, 1993. 90(14): p. 6884-8.

Monteville, M.R., J.E. Yoon, and ME. Konkel, Maximal adherence and invasion of [NT 407 cells
by Campylobacterjejuni requires the CadF outer-membrane protein and microﬁlament
reorganization. Microbiology, 2003. 149(Pt 1): p. 153-65.

Konkel, M.B., et al., Identiﬁcation and molecular cloning of a gene encoding a ﬁbronectin-
binding protein (CadF)ﬁ'om Campylobacterjejuni. Mol Microbiol, 1997. 24(5): p. 953-63.

Ziprin, R.L., et al., The absence of cecal colonization of chicks by a mutant of Campylobacter
jejuni not expressing bacterial fibronectin-binding protein. Avian Dis, 1999. 43(3): p. 586-9.

J in, 8., et al., leA, a novel surface-exposed lipoprotein specific to Campylobacterjejuni, mediates
adherence to host epithelial cells. Mol Microbiol, 2001. 39(5): p. 1225-36.

J in, S., et al., leA of Campylobacter jejuni interacts with surface-exposed heat shock protein
90alpha and triggers signalling pathways leading to the activation of NF -kappaB and p38 MAP
kinase in epithelial cells. Cell Microbiol, 2003. 5(3): p. 165-74.

Pei, 2., et al., Mutation in the peb IA locus of Campylobacter jejuni reduces interactions with
epithelial cells and intestinal colonization of mice. Infect Immun, 1998. 66(3): p. 938-43.

McSweegan, E. and RI. Walker, Identiﬁcation and characterization of two Campylobacterjejuni
adhesins for cellular and mucous substrates. Infect Immun, 1986. 53(1): p. 141-8.

Babakhani, F .K. and LA. Joens, Primary swine intestinal cells as a model for stuaying
Campylobacterjejuni invasiveness. Infect Immun, 1993. 61(6): p. 2723-6.

Mansﬁeld, L.S., et al., Enhancement of disease and pathology by synergy of Trichuris suis and
Campylobacterjejuni in the colon of immunologically naive swine. Am J Trop Med Hyg, 2003.
68(3): p. 70-80.

Hu, L. and DJ. Kopecko, Campylobacterjejuni 81-176 associates with microtubules and aynein
during invasion of human intestinal cells. Infect Immun, 1999. 67(8): p. 4171-82.

Biswas, D., K. Itoh, and C. Sasakawa, Uptake pathways of clinical and healthy animal isolates of
Campylobacterjejuni into INT-407 cells. FEMS Immunol Med Microbiol, 2000. 29(3): p. 203-1].

Konkel, M.E., eta1., Bacterial secreted proteins are required for the internaliztion of
Campylobacterjejuni into cultured mammalian cells. Mol Microbiol, 1999. 32(4): p. 691-701.

Rivera-Amill, V., et al., Secretion of the virulence-associated C ampylobacter invasion antigens
ﬁ'om Campylobacterjejuni requires a stimulatory signal. J Infect Dis, 2001. 183(11): p. 1607-16.

Bacon, DJ ., et al., Involvement of a plasmid in virulence of Campylobacter jejuni 81-176. Infect
Immun, 2000. 68(8): p. 4384-90.

Kiehlbauch, J.A., et al., Phagocytosis of Campylobacter jejuni and its intracellular survival in
mononuclear phagocytes. Infect Immun, 1985. 48(2): p. 446-5].

31

51.

52.

53.

54.

55.

56.

57.

58.

59.

60.

61.

62.

63.

64.

65.

66.

67.

68.

Pesci, E.C., D.L. Cottle, and CL. Pickett, Genetic, enzymatic, and pathogenic studies of the iron
superoxide dismutase of Campylobacter jejuni. Infect Immun, 1994. 62(7): p. 2687-94.

Day, W.A., Jr., et al., Role of catalase in Campylobacterjejuni intracellular survival. Infect
Immun, 2000. 68(11): p. 6337-45.

Moran, A.P., Structure and conserved characteristics of Campylobacter jejuni
lipopolysaccharides. J Infect Dis, 1997. 176 Supp12: p. 8115-21.

Aspinall, G.O., eta1., Serological diversity and chemical structures of Campylobacter jejuni low-
molecular-weight lipopolysaccharides. J Bacteriol, 1992. 174(4): p. 1324-32.

Fry, B.N., et al., The galE gene of Campylobacter jejuni is involved in lipopolysaccharide
synthesis and virulence. Infect Immun, 2000. 68(5): p. 2594-601.

Guerry, P., eta1., Sialylation of lipooligosaccharide cores affects immunogenicity and serum
resistance of Campylobacter jejuni. Infect Immun, 2000. 68(12): p. 6656-62.

Guerry, P., et al., Phase variation of C ampylobacter jejuni 81-176 lipooligosaccharide aﬂects
ganglioside mimicry and invasiveness in vitro. Infect Immun, 2002. 70(2): p. 787-93.

Linton, D., et al., Phase variation of a beta-1,3 galactosyltransferase involved in generation of the
ganglioside GMI-like lipo-oligosaccharide of C ampylobacter jejuni. Mol Microbiol, 2000. 37(3):
p. 501-14.

Yuki, N., et al., Carbohydrate mimicry between human ganglioside GM] and Campylobacter
jejuni lipooligosaccharide causes Guillain-Barre syndrome. Proc Natl Acad Sci U S A, 2004.
101(31): p. 11404-9.

Johnson, W.M. and H. Lior, Cytotoxic and cytotonic factors produced by Campylobacterjejuni,
Campylobacter coli, and Campylobacter laridis. J Clin Microbiol, 1986. 24(2): p. 275-8].

Klipstein, F.A. and RF. Engert, Properties of crude Campylobacterjejuni heat-labile enterotoxin.
Infect Immun, 1984. 45(2): p. 314-9.

Collins, S., et al., A cytotonic, cholera toxin-like protein produced by C ampylobacter jejuni. Comp
Biochem Physiol B, 1992. 103(2): p. 299-303.

Wassenaar, T.M., Toxin production by Campylobacter spp. Clin Microbiol Rev, 1997. 10(3): p.
466-76.

Klipstcin, F.A., et al., Pathogenic properties of C ampylobacter jejuni: assay and correlation with
clinical manifestations. Infect Immun, 1985. 50(1): p. 43-9.

Lam, K.M., A cytotoxicity test for the detection of Campylobacter jejuni toxin. Vet Microbiol,
1993. 35(1-2): p. 133-9.

Aragon, V., K. Chao, and LA. Dreyfus, Eﬂect of cytolethal distending toxin on F -actin assembly
and cell division in Chinese hamster ovary cells. Infect Immun, 1997. 65(9): p. 3774-80.

Lee, RB, et al., Interactions of Campylobacter jejuni cytolethal distending toxin subunits C th
and CdtC with HeLa cells. Infect Immun, 2003. 71(9): p. 4883-90.

Whitehouse, C.A., et al., Campylobacterjejuni cytolethal distending toxin causes a GZ-phase cell
cycle block Infect Immun, 1998. 66(5): p. 1934-40.

32

69.

70.

71.

72.

73.

74.

75.

76.

77.

78.

79.

80.

81.

82.

83.

84.

85.

Pickett, C.L., et al., Prevalence of cytolethal distending toxin production in Campylobacterjejuni
and relatedness of Campylobacter sp. cdtB gene. Infect Immun, 1996. 64(6): p. 2070-8.

Lara-Tejero, M. and J .E. Galan, A bacterial toxin that controls cell cycle progression as a
deoxyribonuclease I-like protein. Science, 2000. 290(5490): p. 354-7.

Fox, J.G., et al., Gastroenteritis in NF -kappaB-deﬁcient mice is produced with wild-type
Camplyobacterjejuni but not with C. jejuni lacking cytolethal distending toxin despite persistent
colonization with both strains. Infect Immun, 2004. 72(2): p. 1116-25.

Hickey, T.E., et al., Campylobacterjejuni cytolethal distending toxin mediates release of
interleukin-8 from intestinal epithelial cells. Infect Immun, 2000. 68(12): p. 6535-41.

Cawthraw, S.A., et al., Long-term antibody responses following human infection with
Campylobacterjejuni. Clin Exp Immunol, 2002. 130(1): p. 101—6.

Calva, J J ., et al., Cohort study of intestinal infection with campylobacter in Mexican children.
Lancet, 1988. 1(8584): p. 503-6.

Taylor, D.N., et al., Influence of strain characteristics and immunity on the epidemiology of
Campylobacter infections in Thailand J Clin Microbiol, 1988. 26(5): p. 863-8.

Mansﬁeld, LS. and D.T. Gauthier, Lymphoglandular complexes are important colonic sites for
immunoglobulin A induction against Campylobacterjejuni in a swine disease model. Comp Med,
2004. 54(5): p. 514-23.

Enocksson, A., et al., Rectal nitric oxide gas and stool cytokine levels during the course of
infectious gastroenteritis. Clin Diagn Lab Immunol, 2004. 11(2): p. 250-4.

Baqar, S., N.D. Pacheco, and FM. Rollwagen, Modulation of mucosal immunity against
Campylobacterjejuni by orally administered cytokines. Antimicrob Agents Chemother, 1993.
37(12): p. 2688-92.

Hickey, T.E., eta1., Campylobacterjejuni-stimulated secretion of interleukin-8 by IN T407 cells.
Infect Immun, 1999. 67(1): p. 88-93.

Al-Salloom, F.S., et al., Campylobacter-stimulated IN T407 cells produce dissociated cytokine
proﬁles. J Infect, 2003. 47(3): p. 217-24.

Jones, M.A., et al., Induction of proinflammatory responses in the human monocytic cell line
THP-l by Campylobacterjejuni. Infect Immun, 2003. 71(5): p. 2626-33.

Smith, C.K., et al., Campylobacterjejuni-induced cytokine responses in avian cells. Infect Immun,
2005. 73(4): p. 2094-100.

Beer, RJ., Studies on the biology of the life-cycle of Trichuris suis Schrank, I788. Parasitology,
1973. 67(3): p. 253-62.

Beer, RJ., Morphological descriptions of the egg and larval stages of Trichuris suis Schrank,
I788. Parasitology, 1973. 67(3): p. 263-78.

Wong, Y.B., et al., Survey of intestinal parasites in pigs from intensive farms in Guangdong
Province, People’s Republic of China. Vet Parasitol, 2005. 127(3-4): p. 333-6.

33

86.

87.

88.

89.

90.

91.

92.

93.

94.

95.

96.

97.

98.

99.

100.

101.

102.

Gerwert, S., K. Failing, and C. Bauer, Husbandry management, worm control practices and
gastro-intestinal parasite infections of sows in pig-breeding farms in Munsterland, Germany.
Dtsch Tierarztl Wochenschr, 2004. 111(10): p. 398-403.

Carstensen, L., M. Vaarst, and A. Roepstorff, Helminth infections in Danish organic swine herds.
Vet Parasitol, 2002. 106(3): p. 253-64.

Gipson, P.S., et al., Health status of a recently discovered population of feral swine in Kansas. J
Wildl Dis, 1999. 35(3): p. 624-7.

Hall, G.A., J .M. Rutter, and RJ. Beer, A comparative stuay of the histopathology of the large
intestine of conventionally reared, specific pathogen free and gnotobiotic pigs infected with
Trichuris suis. J Comp Pathol, 1976. 86(2): p. 285-92.

Tubbs, R.C., Four causes of mucohemorrhagic diarrhea you should recognize. Veterinary
Medicine, 1987. 82(1): p. 89-90.

Mansﬁeld, LS. and J .F . Urban, Jr., The pathogenesis of necrotic proliferative colitis in swine is
linked to whipworm induced suppression of mucosal immunity to resident bacteria. Vet Immunol
Immunopathol, 1996. 50(1-2): p. 1-17.

Rutter, J .M. and RJ. Beer, Synergism between Trichuris suis and the microbial ﬂora of the large
intestine causing dysentery in pigs. Infect Immun, 1975. 11(2): p. 395-404.

Hill, D.E., et al., Trichuris suis: a zinc metalloprotease from culture fluids of adult parasites. Exp
Parasitol, 1993. 77(2): p. 170-8.

Fetterer, RH. and DE. Hill, The occurrence of phenol oxidase activity in female Trichuris suis. J
Parasitol, 1993. 79(2): p. 155-9.

Fetterer, RH. and DE. Hill, Localization of phenol oxidase in female Trichuris suis. J Parasitol,
1994. 80(6): p. 952-9.

Hill, DE. and J.A. Sakanari, Trichuris suis: thiol protease activity ﬁom adult worms. Exp
Parasitol, 1997. 85(1): p. 55-62.

Rhoads, M.L., R.H. Fetterer, and DE. Hill, Trichuris suis: A secretory serine protease inhibitor.
Exp Parasitol, 2000. 94(1): p. 1-7.

Rhoads, M.L., et al., Trichuris suis: a secretory chymotrypsin/elastase inhibitor with potential as
an immunomodulator. Exp Parasitol, 2000. 95(1): p. 36-44.

Abner, S.R., eta1., T richuris suis: detection of antibacterial activity in excretory-secretory
products ﬁom adults. Exp Parasitol, 2001. 99(1): p. 26-36.

Hill, D.E., R.D. Romanowski, and IF Urban, Jr., A Trichuris specific diagnostic antigen ﬁom
culture ﬂuids of Trichuris suis adult worms. Vet Parasitol, 1997. 68(1-2): p. 91-102.

Abner, S.R., et al., Response of intestinal epithelial cells to Trichuris suis excretory-secretory
products and the influence on C ampylobacter jejuni invasion under in vitro conditions. J Parasitol,

2002. 88(4): p. 738-45.

Mansﬁeld, L.S., et al., Construction of internal cDNA competitors for measuring IL-10 and IL-12
cytokine gene expression in swine. Vet Immunol Immunopathol, 1998. 65(1): p. 63-74.

34

103.

104.

105.

106.

107.

108.

109.

110.

111.

112.

113.

114.

115.

116.

117.

118.

119.

Pedersen, S. and I. Saeed, Acquired immunity to Trichuris suis infection in pigs. Parasitology,
2001.123(Pt1):p.95-101.

Faulkner, H., et al., A ge- and infection intensity-dependent cytokine and antiboay production in
human trichuriasis: the importance of IgE. J Infect Dis, 2002. 185(5): p. 665-72.

Lillywhite, J .E., et al., Humoral immune responses in human infection with the whipworm
Trichuris trichiura. Parasite Immunol, 1991. 13(5): p. 491-507.

Turner, 1., et al., A comparison of cellular and humoral immune responses to trichuroid derived
antigens in human trichuriasis. Parasite Immunol, 2002. 24(2): p. 83-93.

Else, K.J. and R.K. Grencis, Cellular immune responses to the murine nematode parasite
Trichuris muris. l. Diﬂerential cytokine production during acute or chronic infection.
Immunology, 1991. 72(4): p. 508-13.

Else, KJ. and R.K. Grencis, Antibody-independent eﬂector mechanisms in resistance to the
intestinal nematode parasite Trichuris muris. Infect Immun, 1996. 64(8): p. 2950-4.

Faulkner, H., et al., Interleukin-9 enhances resistance to the intestinal nematode Trichuris muris.
Infect Immun, 1998. 66(8): p. 3832-40.

Khan, W.I., et al., Modulation of intestinal muscle contraction by interleukin-9 (IL-9) or IL-9
neutralization: correlation with worm expulsion in murine nematode infections. Infect Immun,

2003. 71(5): p. 2430-8.

Else, K.J., L. Hultner, and R.K. Grencis, Cellular immune responses to the murine nematode
parasite T richuris muris. II. Diﬂerential induction of TH-cell subsets in resistant versus
susceptible mice. Immunology, 1992. 75(2): p. 232-7.

Schopf, L.R., et al., IL-1 0 is critical for host resistance and survival during gastrointestinal
helminth infection. J Immunol, 2002. 168(5): p. 2383-92.

Curry, A.J., et al., Evidence that cytokine-mediated immune interactions induced by Schistosoma
mansoni alter disease outcome in mice concurrently infected with Trichuris muris. J Exp Med,
1995. 181(2): p. 769-74.

Else, K.J., et al., Cytokine-mediated regulation of chronic intestinal helminth infection. J Exp
Med, 1994. 179(1): p. 347-5].

Bancroft, A.J., et al., Interleukin-12 promotes a chronic intestinal nematode infection. Eur J
Immunol, 1997. 27(4): p. 866-70.

Bancroft, A.J., A.N. McKenzie, and R.K. Grencis, A critical role for IL-13 in resistance to
intestinal nematode infection. J Immunol, 1998. 160(7): p. 3453-61.

Sansonetti, RJ., et al., Interleukin-8 controls bacterial transepithelial translocation at the cost of
epithelial destruction in experimental shigellosis. Infect Immun, 1999. 67(3): p. 1471-80.

Dawson, H.D., et al., Localized multigene expression patterns support an evolving Thl/ThZ-like
paradigm in response to infections with Toxoplasma gondii and Ascaris suum. Infect Immun,
2005. 73(2): p. 1116-28.

Cho, W.S. and C. Chae, Expression of inﬂammatory cytokines (T NF -alpha, IL-1, IL-6 and IL-8) in

colon of pigs naturally infected with Salmonella typhimurium and S. choleraesuis. J Vet Med A
Physio] Pathol Clin Med, 2003. 50(10): p. 484-7.

35

120.

12].

122.

123.

Elhofy, A. and K.L. Bost, Limited interleukin-18 response in Salmonella-infected murine
macrophages and in Salmonella-infected mice. Infect Immun, 1999. 67(10): p. 5021-6.

Foss, D.L., M.J. Zilliox, and M.P. Murtaugh, Bacterially induced activation of interleukin-1 8 in
porcine intestinal mucosa. Vet Immunol Immunopathol, 2001. 78(3-4): p. 263-77.

Lugering, N., eta1., Interleukin (IL)-I3 and IL-4 are potent inhibitors of IL-8 secretion by human
intestinal epithelial cells. Dig Dis Sci, 1999. 44(3): p. 649-55.

Muchamuel, T., et al., IL-l3 protects mice ﬁ'om lipopolysaccharide-induced lethal endotoxemia:
correlation with down-modulation of TNF-alpha, IF N-gamma, and IL-12 production. J Immunol,
1997. 158(6): p. 2898-903.

36

Chapter 2: “Campylobacter jejuni invades distal colonic lymphoglandular complexes

causing upregulation of proinﬂammatory cytokines in a swine rectal challenge model”

Abstract

Campylobacter jejuni is the leading bacterial cause of foodborne illness
worldwide. Orally inoculated pigs have been used to study diarrhea] disease, intestinal
pathology, and humoral immune responses due to C. jejuni. These studies have shown
that C. jejuni mainly targets the distal colon. We used a rectal inoculation technique to
target the distal colon in order to assess pathogenesis of C. jejuni infection at speciﬁc
time points. Weaned piglets were inoculated rectally with 109 cfu of C. jejuni expressing
green ﬂuorescent protein (Gfp), and were euthanized 1 or 24 hours after inoculation.
Samples of the distal colon were taken for C. jejuni detection and localization, and
cytokine analysis. C. jejuni was detected in 93.75% of inoculated pigs. Immunostaining
detected organisms associated with surface epithelium, within crypts, and within crypts
of lymphoglandular complexes in the distal colon. Several proinﬂammatory cytokines,
including IL-8, IL-18, IL-IB, TNF-Ot, GM-CSF, iNOS, and MCP-l, were signiﬁcantly
upregulated in the distal colon and lymphoglandular complexes 1 hour after inoculation.
These results show that rectal inoculation of C. jejuni is a useﬁil technique for targeting
the distal colon and eliciting cytokine responses. Proinﬂammatory cytokines are likely

important in the pathogenesis of C. jejuni infections in pigs.

37

Introduction

Campylobacterjejuni is the most frequently reported bacterial cause of foodborne
illness worldwide. The CDC estimates 2.4 million cases of campylobacteriosis occur
each year in the US. [1]. Children and young adults are most commonly affected,
developing profuse, watery diarrhea with blood, mucus, and leukocytes 1 to 7 days aﬁer
ingestion of the organism [2, 3] . In most cases symptoms persist for 5 to 8 days and
include abdominal cramping, fever, and malaise [3]. In rare cases death occurs, most
commonly in infants, elderly, and immunosuppressed individuals [2]. Most
Campylobacter infections in humans result in self-limiting gastrointestinal disease.
However, extra intestinal infections can result in local complications including
cholecystitis, pancreatitis, peritonitis, and massive GI hemorrhage. Further spread can
result in meningitis, endocarditis, septic arthritis, osteomyelitis, and neonatal sepsis [1].
Notable sequelae of C. jejuni infection include Reiters syndrome and Guillain-Barré
syndrome (GBS) [1-4] . Reiters syndrome occurs in 1% of people affected with
campylobacteriosis. Symptoms commonly arise 7 to 10 days after onset of diarrheal
illness and include a sterile, reactive arthritis, urethritis, and conjunctivitis [2]. Guillain-
Barré syndrome, an ascending peripheral neuropathy, is an autoimmune disorder that
develops in 1 in every 1000 cases of C. jejuni infection and is most commonly associated
with Penner serotype 0:19 strains [1-4]. Food sources that have been associated with
infection include raw milk, poultry products, shellﬁsh, and fresh produce [2]. Live
animals are a signiﬁcant reservoir for C. jejuni and a source of infection for humans as C.
jejuni is a commensal in the gastrointestinal tract of a variety of wild and domestic

animal species including poultry, cattle, sheep, and swine [5-8]. Campylobacteriosis has

38

incited an abundance of research aimed at elucidating virulence mechanisms and immune
responses in an effort to develop therapies to treat and prevent infection with this
pathogen.

Immune responses to C. jejuni have been characterized in both human infections
and animal models. C. jejuni stimulates a humoral immune response in humans with
production of speciﬁc IgA in feces, urine, saliva, and mammary glands [9]. Flagellin is an
important antigen for stimulating immune responses, as serum taken from patients up to 2
months post infection contained a large proportion of antibodies directed against ﬂagellin
[9]. Humoral immunity affords protection against reinfection with homologous strains
[10], which is believed to account for the relatively low incidence of diarrhea] disease in
adults in developing countries compared to developed countries [11, 12]. Recently,
Mansﬁeld and Gauthier demonstrated that lymphoglandular complexes (LGCS) in the
distal colons of pigs are targeted by C. jejuni. The LGCS expand signiﬁcantly after
infection due to development of B cell rich germinal centers that produce IgA [13].

Cytokines also play an important role in C. jejuni infections. Mice orally
administered recombinant IL-5 or IL-6 have reduced intestinal colonization after
infection. In mice given IL-6, speciﬁc intestinal and systemic IgA responses are also
enhanced [14]. In vitro studies on human cells show that C. jejuni induces IL-8 secretion

and NF-KB translocation in intestinal epithelial cells [15-17]. C. jejuni infected

monocytes secrete IL-IB, IL-6, IL-8, and TNF-OI, and NF-KB is translocated to the

nucleus [18].
Several animal models have emerged to aid in the study of various aspects of C.

jejuni infection. Inbred mice have been used to study leukocyte responses, cytokine

39

responses, immunoglobulin responses, and colonization [19-22]. Ferrets have been used
to study diarrhea] disease, abortion, colonization, and antibody responses to C. jejuni
infection [23]. Pigs have been used to study diarrhea] disease, pathology, and immune
responses [13, 24, 25].

Rectal immunization protocols to study immune responses and vaccine efficacy
have shown that direct inoculation of infectious agents into the distal colon is an
effective, non-invasive method for eliciting and studying local immune responses [26,
27]. In this study we used rectal inoculation of C. jejuni to study the early immunological
response to initial invasion events in the distal colon. Our results indicate that rectal
inoculation of C. jejuni elicits measurable changes in cytokine expression in the distal
colon and lymphoglandular complexes (LGCS). Expression of several proinﬂammatory
cytokines was signiﬁcantly increased in the LGCS, which have previously been shown to

function as antigen sampling structures that are speciﬁcally targeted by C. jejuni [13, 25].

40

Materials and Methods

Animals

Four week old male and female outbred Landrace-Yorkshire cross pigs were
obtained from the Michigan State University (MSU) swine farm. Animals were housed in
the MSU Research containment facility in groups of 4 for social support. Individual
infection groups were housed in separate rooms to eliminate cross contamination. All
animal housing and handling complied with MSU animal use guidelines and National

Institutes of Health guidelines for humane use of laboratory animals.

Bacteria

C. jejuni strain 81-176 which had been electroporated with the plasmid pWM1007
[28] was used for all experiments. The plasmid pWM1007 contains the gin gene under
the inﬂuence of a constitutively active consensus Campylobacter promoter. Bacteria were
grown, from frozen stocks, on Bolton (Oxoid Inc., Basingstoke, Hampshire, England)
agar plates without antibiotics at 37°C, 5% C02 in humidiﬁed air for 48 to 72 hours.
Isolated colonies were subcultured onto Bolton agar with 50 ug/mL of kanamycin to
select for gfp-expressing clones. Four milliliter (4 mL) aliquots of Bolton broth without
antibiotics, in a 6-well polystyrene tissue culture plate (Costar®, Corning, Corning, NY),
were inoculated from isolated colonies from agar plates with kanamycin. The broth was
incubated at 37°C, 5% C02 in humidiﬁed air at 200 RPM until the early log phase of
growth, as determined by optical density (OD). The culture was diluted 1:100 in 75 mL

fresh Bolton broth without antibiotics and incubated under the previously described

41

conditions until an OD6oo of approximately 0.1 was reached. Serial dilutions and plate
counts from previous experiments have shown that this OD corresponds to the early log
phase of grth and a concentration of approximately 1x108 chL. The bacteria were
collected and pelleted by centrifugation and resuspended in serum free medium (SFM)
(Sigma Inc., St. Louis, M0) for rectal inoculation of pigs. A subsarnple of the inoculum
were examined microscopically using a Nikon Eclipse E600 with an Epi-ﬂuorescence
system (Mager Scientiﬁc, Dexter, M1) for darting motility and ﬂuorescence prior to
infecting pigs. Also, a sample from the prepared inocula was serially diluted and plated to
determine actual concentration. Each pig was inoculated with approximately 2x109 cfu C.

jejuni in a total volume of 10 mL SF M.

Experimental design and inoculation procedure

Pigs were divided into 4 groups as follows: Group 1: Control, euthanized after 1
hour; Group 2: Infected, euthanized after 1 hour; Group 3: Control, euthanized aﬂer 24
hours; Group 4: Infected, euthanized after 24 hours. All infected pigs received C. jejuni
resuspended in 10 mL SFM. All control pigs received 10 mL SFM without bacteria. All
pigs were tested for the presence of C. jejuni both before and aﬁer inoculation by culture
of rectal swabs or fecal samples. Rectal swabs or fecal samples were placed in Cary Blair
(Difco Laboratories, Detroit, MI) transport medium for later culture in the laboratory.
After samples were taken for C. jejuni culture, the pigs were anesthetized with
intramuscular injections of 4.4 mg/kg Telazol (Fort Dodge, Overland Park, KS) and 2.2
mg/kg xylazine (Butler, Columbus, OH). Once the pigs were fully anesthetized, they

were placed in lateral recumbency on heating pads and covered with towels to maintain

42

body temperature throughout the procedure. For each pig, the perineum was cleaned with
three scrubs, alternating betadine and isopropyl alcohol for each scrub, and a purse string
suture pattern was placed around the anus using 0 silk suture and left untied. A sterile 14
French feeding tube was inserted into the anus and advanced 10 to 15 cm into the distal
colon. The appropriate inoculum was slowly injected through the tube, which was
immediately removed. The purse string was tightened, and the hindquarters were elevated
slightly. All pigs remained anesthetized for at least 1 hour, after which they were either
euthanized or allowed to recover from anesthesia, depending on group assignment. Pigs
that were allowed to recover from anesthesia were reanesthetized, as previously
described, 24 hours after inoculation. All pigs were euthanized with an intravenous

injection of 86 mg/kg sodium pentobarbital (Fatal Plus®, Vortech Pharmaceuticals,

Dearbom, MI).

Sample collection

The colons of euthanized pigs were removed from the abdomen and Opened. A
separate set of sterile instruments was used for each group to eliminate cross
contamination between infection groups. Full thickness samples of the distal colon and
individual LGCS were collected. Samples were ﬁxed in 3.7% formaldehyde for histology
and irnmunohistochemistry, embedded in OCT compound for laser capture
microdissection, or placed in cryovials and snap frozen in liquid nitrogen for nucleic acid

isolation.

43

C. jejuni isolation and identification

Rectal swabs or fecal samples were streaked onto Preston selective agar (Oxoid
Inc., Basingstoke, Hampshire, England) and incubated at 42°C, 5% C02 in humidiﬁed air
for 48 hours. Colonies isolated from rectal swabs or fecal samples of experimentally
infected pigs were scored according to 5 criteria: 1) the bacteria grew on Campylobacter
selective media under microaerophilic conditions (5% C02 at 42°C); 2) colonies were
round, raised, had pink pigmentation, and were 1-2mm in diameter-morphology that was
consistent with Campylobacter spp.; 3) microscopic examination revealed curved motile
rods; 4) bacteria were positive for PCR ampliﬁcation of the quinolone resistance
determining region (QRDR) of the gyrA gene of C. jejuni [29]; and 5) bacteria
ﬂuoresced green when examined at 488 nm wavelength on a Nikon Eclipse E600 with an
Epi-ﬂuorescence system (Mager Scientiﬁc, Dexter, MI). Pigs were considered positive
for the strain of C. jejuni used for infections if all 5 criteria were met.

We found culture alone to be inadequate for detection of C. jejuni from
experimentally inoculated animals, therefore we used 3 different PCR techniques to
detect C. jejuni from tissue DNA. Total DNA was isolated from approximately 25 mg of
frozen tissue using a Qiagen DNEasy tissue kit (Qiagen, Valencia, CA) following the
manufacturer’s protocol. The DNA was quantiﬁed spectrophotometrically (Beckman,
Fullerton, CA). To detect C. jejuni from experimental animals, the C. jejuni QRDR target
was ampliﬁed by PCR from SOOng of total DNA isolated from full thickness tissue
samples using the previously designed primers JL238 (5‘ TGG GTG CTG TTA TAG
GTC GT 3') and JL239 (5' GCT CAT GAG AAA GT1" TAC TC 3') [29]. Amplicons

were electrophoresed on a 1.8% agarose gel in 1X Tris-Acetate-EDTA (TAE) buffer with

44

0.2 ug/mL ethidium bromide. The fragments were visualized by UV illumination on an
Alpha Imager Gel documentation system (Alpha Innotech Corporation, San Leandro,
CA).

A Restriction Fragment Length Polymorphism (RFLP) assay to detect and
distinguish C. jejuni, C. coli, C. lari, and C. upsaliensis by analysis of a portion of the
23S rRNA gene was used to detect thermophilic campylobacters in experimental animals
[30]. The target was ampliﬁed from 500ng of total pig tissue DNA using primers
THERMI (5'-TAT TCC AAT ACC AAC ATT AGT-3') and THERM4 (5'-CTT CGC
TAA TGC TAA CCC-3'). The amplicons were digested overnight with either AIuI or
T sp5091 following the manufacturers protocols (New England Biolabs, Beverly, MD),
and the resultant fragments were separated electrophoretically on a 2% agarose gel in 1X
Tris-Acetate-EDTA (TAE) buffer with 0.2 ug/mL ethidium bromide. The fragments were
visualized by UV illumination on an Alpha Imager Gel documentation system (Alpha
Innotech Corporation, San Leandro, CA).

Primers were designed to amplify a 543 bp fragment from the gfp gene (Genbank
Number AF 292556) using Squeb soﬂware (Genetics Computer Group, Madison, WI).
The forward primer sequence is 5'AGT GGA GAG GGT GAA GGT GAT G 3'; the
reverse primer sequence is 5' AAG GGC AGA TTG TGT GGA CAG G 3'. A 25 [IL PCR
reaction with 1X PCR buffer (Tris-HCL, pH 8.3, 50 mM KCl), 1.0 mM MgCl2, 0.05 mM
each dATP, dTTP, dCTP, dGTP, 37.5 ng of each primer, 0.625 U AmpliTaq Gold DNA
polymerase (Applied Biosystems, Foster City, CA), and 500ng total tissue DNA was
used to amplify the target. Ampliﬁcation parameters were initial denaturation at 94°C for

10 minutes, followed by denaturation for 1 minute at 94°C, annealing for 1 minute at

45

62°C, extension for 1 minute at 72°C for 39 cycles, and a ﬁnal extension at 72°C for 10
minutes. The amplicons were electrophoresed on a 1.8% agarose gel, stained, and

visualized as previously described.

Histology and Immunohistochemistry

Fixed tissue samples were embedded in parafﬁn and 5 pm sections were cut and
adhered to charged slides. Slides were routinely dehydrated in progressively higher
concentrations of ethanol, deparafﬁnized and rehydrated. For histological evaluation,
slides were stained with Mayer's hematoxylin and eosin and cover slipped. For
irnmunostaining, sections of selected tissues were incubated in preheated antigen retrieval
buffer (Dako, Carpinteria, CA) for 20 minutes at 100° C. Slides were removed from the
steamer and left in the retrieval solution at room temperature to cool down for 20
minutes, rinsed in distilled water and transferred to Tris-HCL buffer. Endogenous
peroxidase was blocked for 15 minutes with 3% hydrogen peroxide. Non-speciﬁc
immunoglobulin binding blocking was done by a 10 minute incubation with a protein-
blocking agent (Dako, Carpinteria, CA) prior to application of the primary antibody. A
rabbit polyclonal anti-C. jejuni antibody (Biogenesis, New Fields, UK) was used as the
primary antibody at a concentration of 1:100. Sections were stained with a Dako
autostainer using a labeled streptavidin-immunoperoxidase staining procedure (Dako
Corp., Carpinteria, CA). The immunoreaction was visualized with ABC (Dako Corp.,
Carpinteria, CA). Sections were counterstained with Mayer’s hematoxylin, then
dehydrated, cleared and mounted. Sections of distal colons from pigs in which infection

with C. jejuni had been conﬁrmed by PCR served as positive controls.

46

Determination of antibody speciﬁcity

Healthy, conventionally reared pigs have been shown to have more than one
species of Campylobacter residing as commensals in their intestines [7]. Due to the
likelihood that the pigs used in this study also harbored other Campylobacter species and
the possibility of cross reactivity with those species, we tested the speciﬁcity of the
polyclonal antibody for C. jejuni.

IPEC-l cells are a non-immortalized swine intestinal epithelial cell line derived
from neonatal pigs. All cell culture reagents were obtained from Invitrogen unless
otherwise indicated. IPEC-l cells were grown in DMEM/F12 media supplemented with
5% fetal bovine serum (FBS) and 1% Insulin-Transferrin-Selenium(ITS) in Costar
culture ﬂasks (Costar®, Corning, Corning, NY). Cells were washed in versene and
detached with 0.05% Trypsin-EDTA, then seeded at a density of 5x105 cells per well
onto glass chamber slides coated with 20ug/mL ﬁbronectin (Sigma, St. Louis, MO). The
cells were incubated at 37°C, 5% C02 until conﬂuent (2-3 days). C. jejuni 81-176
expressing gfp, C. coli 43134, C. lari 43675, and C. upsaliensis 43954 were streaked for
isolation onto Bolton agar plates (Oxoid Inc., Basingstoke, Hampshire, England) and
incubated for 72 hours at 37°C, 5% C02. Isolated colonies were streaked for lawn growth
and incubated for 20 to 24 hours. Lawns were harvested from plates with sterile swabs,
resuspended in DMEM/F12 supplemented with 5%F BS and 1%ITS, and adjusted to an
0Deoo of 0.1. IPEC-l cells were inoculated with an M01 100:] of each Campylobacter

species and incubated for 3 hours at 37°C, 5% C02, The supematants were removed and

47

the cells were ﬁxed in 10% formaldehyde, then stained as previously described with the

anti-C. jejuni antibody.

Laser Capture Microdissection

Five micron sections were cut from tissue samples embedded in OCT compound
on a Reichert-Jung cryotome (Reichert-Jung , Bensheim, Germany) and adhered to poly-
L-lysine (Sigma, St. Louis, MO) coated slides. Slides were rinsed in ethanol, stained with
Mayer’s Haemotoxylin, dehydrated in progressively higher concentrations of ethanol,
then rinsed in xylene. Crypt units and submucosa] samples, avoiding the surface mucosa,
were microdissected using a Pixcell II laser capture Microdissection system (Arcturus,
Mounrainview CA). Samples were adhered to CapSure HS LCM Caps (Arcturus,
Mounrainview CA) and stored at -20°C for later DNA isolation. For DNA extraction,
caps were incubated in Proteinase K digestion buffer (1mg/ml Proteinase K (Qiagen,
Valencia, CA) / 1% Tween 20 (Sigma) in TE Buffer pH.8.0) overnight at 37°C. Samples
were heated to 95°C for 10 minutes to inactivate the Proteinase K and stored at -20°C.
To determine if the inoculating strain was present in the crypts and submucosa, IOuL of
DNA were used as a template in a PCR reaction with primers designed in our laboratory
to detect the gp gene. RFLP analysis of the Campylobacter 23S rRNA gene was also
performed to determine if other Campylobacter strains were present in the crypts and

submucosa[3 O].

48

Cytokine Analysis

To determine the effect of rectal inoculation of C. jejuni on local immune
responses in the distal colon and LGCS, mRNA expression of 14 cytokines, chemokines
and iNOS was measured using a real time PCR assay. Total RNA was isolated from
frozen tissue samples from each pig using Trizol reagent (GibcoBRL Life Technologies).
Brieﬂy, 3mm3 tissue samples were homogenized in Trizol using a Polytron® tissue
homogenizer (Kinematica, Cincinnati, OH). RNA was extracted with chloroform,
precipitated with isopropanol, and washed with ethanol. Pellets were air dried, then
resuspended in DEPC treated H2O (GibcoBRL Life Technologies). DNA contamination
was removed by treatment with approximately 27 Kunitz units of DNase I for 15
minutes. Samples were repuriﬁed on a Qiagen RNEasy column (Qiagen, Valencia, CA).
RNA concentration and integrity were determined using an Agilent Bioanalyzer 2100
(Agilent Technologies, Palo Alto, CA). For each sample, 10 pg of total RNA was reverse
transcribed with random primers using a Stratagene First Strand RT-PCR kit (Stratagene,
La Jolla, CA). First strand cDNA was used as the template in real time PCR reactions.
Real time PCR primers and probes designed using the Primer Express (Applied
Biosystems, Foster City, CA) software package were used to analyze 14 cytokines,
chemokines, and iNOS (Table 2). The 18S rRNA probe was obtained from Applied
Biosystems and was used as a constitutively active gene for normalization. PCR was
performed using a commercially available kit (Brilliant kit, Stratagene, La Jolla, CA) on
an ABI PRISM 79OOHT (Applied Biosystems). Ampliﬁcation conditions were as
follows 50°C for 2 minutes; 95°C for 10 minutes; 40 cycles of 95°C 15 seconds and 60°C

for 1 minute. Fluorescence signals measured during ampliﬁcation were processed post

49

ampliﬁcation. Threshold cycle (Ct) values were determined using values of 0.2 for TET,

0.1 for VIC, and 0.05 for FAM reporter dyes.

Statistical Analysis

The cytokine mRNA expression data for each tissue was analyzed separately
using Statview 5.0 for Macintosh (Abacus Concepts, Berkeley, Calif). The CI value for
the 18S ribosomal subunit was subtracted from the C, value for each cytokine message to
normalize for RNA content prior to statistical analysis. One-way analysis of variance
(ANOVA) was performed to determine the statistically signiﬁcant differences between
treatments. A Fisher’s least squares difference test was performed to determine
differences between infected and control animals at each time point. P values of 0.05 or

less were considered statistically signiﬁcant.

50

Results

Campylobacter isolation and identification

C. jejuni was not isolated from any fecal samples taken prior to infection (data not
shown). One hour aﬂer infection, C. jejuni was not recovered from any infected or
control pigs (Table 1). Twenty four hours after inoculation, C. jejuni was recovered from
1 of 4 infected pigs and 0 of 4 control pigs. All C. jejuni isolates recovered from pigs
were conﬁrmed to be the inoculating strain by examination using ﬂuorescence
microscopy.

PCR analysis of total DNA isolated from pig tissues proved to be a more sensitive
assay for the detection of C. jejuni from experimentally inoculated pigs. One hour after
inoculation, C. jejuni was detected in all 4 infected pigs and none of the control pigs
using primers speciﬁc for the QRDR of the gyrA gene of C. jejuni. Twenty four hours
after inoculation, C. jejuni was detected in 1 of 4 infected pigs and 1 of 4 control pigs
(Figure l). The infected pig that was positive on culture 24 hours after inoculation was
the same pig that was positive by ampliﬁcation of the QRDR target.

Analysis of the Campylobacter 23S rRNA RFLP demonstrated that both C. jejuni
and C. coli were present in our experimental animals (Figure 2). One hour after
inoculation, 3 of 4 infected pigs had patterns indicative of C. jejuni and the remaining pig
had a pattern indicative of C. coli. All control pigs had patterns indicative of C. coli.
Additionally, 24 hours aﬁer inoculation, all infected and control animals had patterns
indicative of C. coli only. Neither C. lari nor C. upsaliensis were detected from any pig.

These data strongly suggest that all of the pigs were colonized by C. coli prior to

51

infection and at one hour after experimental infection C. jejuni is the predominant
species, but by 24 hours C. coli is predominant.

Detection of the gp gene proved to be the most sensitive assay for detection of C.
jejuni from experimentally inoculated pigs. One hour after inoculation, the mp target was
detected ﬁ'om all infected pigs and none of the controls (Figure 3). Twenty four hours
after inoculation, the target was detected from 2 Of 4 infected pigs and none of the

controls.

Histology and Immunohistochemistry

Distal colonic tissue samples were examined microscopically to determine if
lesions were induced by rectal inoculation with C. jejuni. Additionally, the location of
Campylobacter organisms in the samples was determined by immunohistochemistry
using a rabbit polyclonal anti-C. jejuni antibody. Control pigs showed no degenerative,
proliferative, or inﬂammatory lesions. Most experimentally infected pigs also showed no
degenerative, proliferative, or inﬂammatory lesions (data not shown).

Tests to determine the speciﬁcity of the polyclonal antibody used for
immunocytochemical staining showed strong cross reactivity with C. lari as well as C.
jejuni, but no cross reactivity with C. coli, C. hyointestinalis, or C. upsaliensis (Figure 4).
Microscopic examination of tissues stained with the polyclonal antibody showed
positively stained rods along the surface epithelium, within the mucus of the crypts and
the crypts of the LGCS, and associated with goblet cells (Figure 5A through D). Staining
patterns were similar at both 1 and 24 hours and were detected in both infected and

control animals.

52

Interestingly, one pig had more severe lesions at 1 hour post inoculation
characterized by multifocal distention of crypts with mucus, neutrophils, and small
curved rods that stained positively on IHC (Figure 5E). Crypts within the LGCS were also
distended with mucus, neutrophils, and positively staining curved rods. Lymphocytes,
neutrophils, and plasma cells had inﬁltrated the epithelium and superﬁcial propria, and
there was evidence of epithelial degeneration.

Immunocytochemical staining of rod shaped bacteria in the distal colonic crypts
suggested that Campylobacters were located in the crypts. However, cross reactivity of
the antibody with C. lari as well as C. jejuni necessitated a more speciﬁc technique to
determine if the bacteria in the crypts were the same strain used for inoculation. We used
laser capture microdissection to speciﬁcally isolate bacteria and enterocytes from the

crypts and submucosa to test by PCR for C. jejuni speciﬁc targets.

Laser Capture Microdissection

Laser capture microdissection was performed to isolate crypt units and
submucosa, avoiding the surface epithelium, to determine if the C. jejuni strain used for
inoculation was present. A PCR using primers speciﬁc for the gﬁ) gene ampliﬁed the
target only from the infected pig that was tested, while the control animal was negative
(Figure 6). Additionally, RFLP analysis of the Campylobacter 23s rRNA gene was
performed to determine if other Campylobacter species were present in crypts and
submucosa. The infected animal had banding patterns consistent with C. jejuni and C.
coli. A faint pattern that could be consistent with C. upsaliensis was also seen. The

control animal had banding patterns that were consistent with C. coli, and an additional

53

band at ~150bp that did not ﬁt the patterns Of C. jejuni, C. coli, C. lari, or C. upsaliensis

(Figure 7).

Cytokine mRNA expression

We measured mRNA expression of 14 cytokines, chemokines, and iNOS in both
the distal colon and LGCS to determine if C. jejuni had a signiﬁcant impact on cytokine
expression (Table 2). One hour after infection, expression of IL-18, TNFor, and IL-8 was
signiﬁcantly upregulated 2 to 3 fold over controls in the distal colon and IL-8 remained
upregulated 24 hours after infection. At the same time in the LGCS expression Of IL-18,
TNFOI, IL-8, GM-CSF, IL-IB, iNOS, MCP-l, IL-6, and IL-10 was upregulated 3 to 20
fold over controls (Figure 8). Downregulation of mRNA expression was not seen for any
target measured. These data indicates that a proinﬂammatory response was initiated by C.

jejuni infection, and the response was more robust in LGCS compared to the distal colon.

54

Discussion

Rectal inoculation of C. jejuni stimulates a proinﬂammatory cytokine response in
the colon characterized by increased expression of IL-IB, IL-6, IL-8, IL-18, GM-CSF,
iNOS, MCP-l, and TNFOI. In samples of the distal colon excluding LGCS, expression of
IL-18, TNFOL, and IL-8 was signiﬁcantly upregulated 1 hour after inoculation, and IL-8
remained upregulated 24 hours after inoculation. These data correlate with cytokine
production from in vitro infections of intestinal epithelial cells and monocytes with C.
jejuni. Swine intestinal epithelial cells, IPEC-l, secrete IL-IB, TNFOI, IL-6, IL-8 and IL-
18 upon infection with C. jejuni (Parthasarathy and Mansﬁeld, unpublished data; Jones et
al, unpublished data; Cunningham and Mansﬁeld, unpublished data). Human intestinal
epithelial cells, INT-407, have been shown to secrete IL-8 in response to C. jejuni [15,
16]. In addition, intracellular production of IFNy, TNFOI, IL-4, and IL-10 is increased
with C. jejuni infection [31]. In our model immunohistocherrristry detected
Campylobacter organisms associated with surface epithelial cells, goblet cells, and within
the mucus of distal colon crypts. Histologically, no inﬂammatory changes were seen in
infected pigs, therefore the signiﬁcantly increased cytokines are likely due to the
interaction of C. jejuni with intestinal epithelial cells. C. jejuni targets lymphoglandular
complexes in the distal colons of pigs after oral inoculation, stimulating an
immunoglobulin response [13]. LGCS from C. jejuni infected pigs are composed of the
follicle associated epithelium, B cell rich germinal centers, T cells, and cells with

macrophage morphology [13, 25]. In vitro human monocytes secrete IL-10, IL-6, IL-8,

and TNF-a upon infection with C. jejuni [18]. In our rectal challenge model, IHC

55

demonstrated Campylobacter organisms within the entrapped crypts of the LGCS and
within the follicle associated with leukocytes. In the LGCS IL-10, IL-6, IL-8, IL-18, GM-
CSF, iNOS, MCP-l, and TNFOL were signiﬁcantly upregulated. In addition, the response
was more robust than in the non-lymphoid portions of the distal colon with 3 to 20 fold
increases in cytokine expression compared to 2 to 3.5 fold increases in the distal colon.
In contrast to the distal colon, the increased cytokines in the LGC are likely due primarily
to the interaction of C. jejuni with leukocytes, which are capable of producing an
expanded cytokine proﬁle and greater levels of cytokines [15, 18]. Ultimately, both
tissues showed induction of a proinﬂammatory response. The enhanced response in
LGCS supports other data showing that C. jejuni speciﬁcally targets the lymphoid tissue
[13, 25].

Proinﬂammatory cytokines are induced by a variety of enteric pathogens.
Salmonella, Shigella, and Campylobacter increase rectal nitric oxide gas and stool IL-1 [3
levels in humans during the acute phase of infection [32]. Shigella stimulates production
of IL-IB, IL-8, and IL-18 in animal models of infection. These cytokines are critical for
initiating inﬂammation that limits invasion of the bacteria into the lamina propria and
ultimately eliminates the bacteria from the host [33, 34]. Salmonella stimulates
expression of IL-1, IL-6, IL-8, IL-18, and TNFOI from intestinal tissues of pigs [35, 36].
In mice, wild type Salmonella decreases expression of 1L-18 in Peyer's patches and
mesenteric lymph nodes, which is believed to be a mechanism for evading the immune
response and establishing an infection [37]. Helicobacter pylori stimulates increased
expression of IL-10, IL-6, IL-18 and TNFOI in the gastric mucosa of humans and

nonhuman primates [3 8-40]. Given that several enteric bacterial pathogens induce

56

proinﬂammatory cytokines in vivo, and C. jejuni induces proinﬂammatory cytokines in
vitro, it is not surprising that expression of several proinﬂammatory cytokines was
increased in the distal colons and LGCS of rectally infected pigs as these cytokines likely
play a role in resistance to C. jejuni infection.

In our model, rectally inoculated C. jejuni behaved in a similar manner to orally
inoculated bacteria. Oral inoculation studies have demonstrated C. jejuni within
entrapped crypts and in the follicle associated epithelium of distal colonic LGCS of germ
free piglets [25], and within superﬁcial mucosa] epithelial cells of the colons of
colostrums deprived piglets [24]. These bacteria have traversed the gut and encountered
the acidic pH of the stomach and bile in the small intestine, two environmental conditions
that can affect the virulence of the organism [41]. The C. jejuni flaA promoter can be
upregulated by acidic pH, bovine bile, and deoxycholate which suggests that these stimuli
enhance production of ﬂagella, an important virulence determinant [41, 42]. Additionally,
bile stimulates the synthesis of Cia proteins, which are essential for C. jejuni invasion
into intestinal epithelial cells [41]. However, these environmental stimuli are not
absolutely necessary for C. jejuni virulence. In vitro, routinely cultured bacteria invade
intestinal epithelial cells [15, 28, 43] and elicit cytokine responses [15-17]. In vivo, ferrets
rectally infected with C. jejuni become colonized and develop diarrhea similar to orally
infected animals [23]. Our data correlates with these ﬁndings, demonstrating
Campylobacter organisms associated with epithelial cells and goblet cells, within distal
colonic crypts, and within crypts of the LGCS at both 1 hour and 24 hours after
inoculation. PCR analysis of microdissected crypts conﬁrmed that C. jejuni 81-176

expressing gfp was present in infected pigs and not controls. Thus it was not entirely

57

unexpected that cytokine expression was signiﬁcantly increased in the distal colon and
LGCS

Positive staining with immunohistochemistry on the mucosal surface, in the
crypts, and in the entrapped crypts Of the LGCs of rectally infected pigs suggests that C.
jejuni is present in these areas. PCR analysis of DNA from full thickness tissue samples
and microdissected crypts and submocosa conﬁrm that the inoculating strain of C. jejuni
had invaded the crypts and submucosa of infected pigs. Positive staining was also seen in
control pigs but PCR analysis of DNA samples did not corroborate the presence of C.
jejuni speciﬁcally. Cross reaction of the polyclonal antibody with closely related
Campylobacter species could account for positive staining in control pigs. All control
pigs harbored C. coli, as determined by RFLP analysis, but our tests to determine
antibody speciﬁcity showed cross reaction only with C. lari (Figure 5). C. lari has been
detected by culture in the feces of healthy pigs [7] as well as from abscessed LGCs from
pigs with mucohemorrhagic diarrhea due to Trichuris suis infection [44]. All control and
infected animals from our rectal inoculation study were negative by PCR for the C. lari
Ipr gene (data not shown) [45]. Interestingly, Mansﬁeld and Urban isolated an
unidentiﬁed Campylobacter species from abscessed LGCs as well as C. jejuni, C. coli
and C. lari [44]. These bacteria were all part of the resident ﬂora of conventionally reared
pigs prior to T. suis infection, and could also have been present in the pigs used in our
experiment. In light of this, the changes in cytokine expression in the distal colon and
LGCS could also be due in part to perturbation of the resident bacterial population upon

introduction of exogenous Campylobacter.

58

Due to the low incidence of C. jejuni cultured from pigs after infection, we used
three different PCR methods to detect C. jejuni in DNA isolated from pig tissue samples:
ampliﬁcation of the C. jejuni QRDR, RFLP of Campylobacter 23S rRNA gene, and
ampliﬁcation of the gfp gene. The RF LP analysis gave valuable information about the
species of Campylobacter present in the distal colons of the experimentally infected
piglets, but it was not as sensitive for detection of the inoculation strain as was expected.
Therefore, we developed a PCR assay to detect the gfp gene that would only be present in
the samples if the inoculating strain was present in the tissue. The PCR protocols for
ampliﬁcation of C. jejuni QRDR and Campylobacter 23S rRNA gene were originally
designed for use on bacterial cultures, but we were able to adapt the protocols and
successfully amplify Campylobacter targets from mixed DNA samples containing both
pig and bacterial DNA. We found these assays to be more rapid and sensitive for
Campylobacter detection and identiﬁcation in infection studies than traditional culture
methods. Overall, we believe that culture alone is insufficient to detect and distinguish
Campylobacters that are present in the gut of conventionally reared pigs. PCR analyses
were more sensitive and speciﬁc for detection and identiﬁcation and did not rely on the
ability to recover viable bacteria, as would be needed for traditional speciation by
biochemical means.

The single pig that was infected for only one hour but had signiﬁcant microscopic
lesions was an unusual ﬁnding. Clearly there was an inﬂammatory response occurring in
this animal as was evident by inﬁltration of inﬂammatory cells into the lamina propria
and the existence of multifocal crypt abscesses (Figure 5E). It is unlikely that the

abscessation was caused by the C. jejuni inoculum as this pig was only infected for one

59

hour. However, it is interesting to note that there were Campylobacter organisms
associated with the abscessed crypts. Whether the organisms were from the inoculum or
from the resident population the pathology supports the hypothesis that Campylobacter
spp. are opportunistic pathogens in the distal colons of pigs. Campylobacter organisms
invade abscessed LGCs in the distal colons of both immunocompetent pigs infected with
the whipworm T suis [44] and germ free piglets infected with T. suis and C. jejuni [25].
In this dual infection model, it is hypothesized that the whipworm infection facilitates
secondary infection by resident gut ﬂora through dysregulation of the immune response,
or by direct effects on enterocytes that permit bacterial invasion [44] (Parthasarthy,
unpublished results). It is possible that in the rectally inoculated animal from our study,
some undetermined agent facilitated Campylobacter invasion similar to T. suis
facilitation of C. jejuni invasion.

Rectal inoculation proved to be a useful tool for studying the pathogenesis of C.
jejuni infection in a swine model. Rectal inoculation allowed targeted delivery of a
speciﬁc amount of bacteria to the distal colon and a speciﬁc time frame to study the
effects on cytokine expression in the distal colon. The fact that signiﬁcant changes in
cytokine expression were measured after rectal inoculation with C. jejuni conﬁrms that
the bacteria are capable of eliciting a response when delivered by a non conventional
route. This model allows for delivery of the agent of interest to a discrete area of the
intestine and assessment of responses at more accurate time points. We believe that this
model will continue to be useful for studying the complex interaction between pathogen

and host in vivo.

60

10.

ll.

12.

l3.

14.

15.

l6.

17.

References

Allos, B.M., Campylobacterjejuni Infections: update on emerging issues and trends. Clin Infect
Dis, 2001. 32(8): p. 1201-6.

Altekruse, S.F., D.L. Swerdlow, and NJ. Stern, Microbial food borne pathogens. C ampylobacter
jejuni. Vet Clin North Am Food Anim Pract, 1998. 14(1): p. 31-40.

Ketley, J.M., Pathogenesis of enteric infection by Campylobacter. Microbiology, 1997. 143 (Pt 1):
p. 5-21.

Hughes, R.A. and J .H. Rees, Clinical and epidemiologic features of Guillain-Barre syndrome. J
Infect Dis, 1997. 176 Suppl 2: p. 892-8.

Torre, E. and M. Tello, Factors inﬂuencing fecal shedding of C ampylobacter jejuni in dogs
without diarrhea. Am J Vet Res, 1993. 54(2): p. 260-2.

Wolfs, T.F., et al., Neonatal sepsis by Campylobacterjejuni: genetically proven transmission from
a household puppy. Clin Infect Dis, 2001. 32(5): p. E97-9.

Young, C.R., et al., Enteric colonisation following natural exposure to Campylobacter in pigs.
Res Vet Sci, 2000. 68(1): p. 75-8.

Altekruse, S.F., et al., Food and animal sources of human C ampylobacter jejuni infection. J Am
Vet Med Assoc, 1994. 204(1): p. 57-61.

Cawthraw, S.A., et al., Long-term antibody responses following human infection with
Campylobacterjejuni. Clin Exp Immunol, 2002. 130(1): p. 101-6.

Panigrahi, P., et al., Human immune response to Campylobacterjejuni proteins expressed in vivo.
Infect Immun, 1992. 60(11): p. 4938-44.

Calva, J .J ., et al., Cohort study of intestinal infection with campylobacter in Mexican children.
Lancet, 1988. 1(8584): p. 503-6.

Taylor, D.N., et al., Influence of strain characteristics and immunity on the epidemiology of
Campylobacter infections in Thailand. J Clin Microbiol, 1988. 26(5): p. 863-8.

Mansﬁeld, LS. and D.T. Gauthier, Lymphoglandular complexes are important colonic sites for
immunoglobulin A induction against C ampylobacter jejuni in a swine disease model. Comp Med,
2004. 54(5): p. 514-23.

Baqar, 8., ND. Pacheco, and F.M. Rollwagen, Modulation of mucosal immunity against
Campylobacterjejuni by orally administered cytokines. Antimicrob Agents Chemother, 1993.
37(12): p. 2688-92.

Hickey, T.E., et al., C ampylobacter jejuni-stimulated secretion of interleukin-8 by INT407 cells.
Infect Immun, 1999. 67(1): p. 88-93.

Hickey, T.E., et al., Campylobacterjejuni cytolethal distending toxin mediates release of
interleukin-8 from intestinal epithelial cells. Infect Immun, 2000. 68(12): p. 6535-41.

Mellits, K.H., et al., Activation of the transcription factor NF -kappaB by Campylobacterjejuni.
Microbiology, 2002. 148(Pt 9): p. 2753-63.

61

l8.

19.

20.

2].

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

Jones, M.A., et al., Induction ofproinﬂammatory responses in the human monocytic cell line
THP-l by Campylobacterjejuni. Infect Immun, 2003. 71(5): p. 2626-33.

Baqar, S., et al., Murine intranasal challenge model for the study of Campylobacter pathogenesis
and immunity. Infect Immun, 1996. 64(12): p. 4933-9.

Fox, J.G., et al., Gastroenteritis in NF -kappaB-deficient mice is produced with wild-type
Camplyobacterjejuni but not with C. jejuni lacking cytolethal distending toxin despite persistent
colonization with both strains. Infect Immun, 2004. 72(2): p. 1116-25.

Pancorbo, P.L., et al., Evaluation of cytokine production and phagocytic activity in mice infected
with Campylobacterjejuni. Curr Microbiol, 1999. 39(3): p. 129-33.

Pancorbo, P.L., et al., Potential intervention of Campylobacter jejuni in the modulation of murine
immune response. Curr Microbiol, 2001. 43(3): p. 209-14.

Bell, J .A. and DD. Manning, A domestic ferret model of immunity to Campylobacterjejuni-
induced enteric disease. Infect Immun, 1990. 58(6): p. 1848-52.

Babakhani, F.K., G.A. Bradley, and LA. Joens, Newborn piglet model for campylobacteriosis.
Infect Immun, 1993. 61(8): p. 3466-75.

Mansﬁeld, L.S., et al., Enhancement of disease and pathology by synergy of Trichuris suis and
C ampylobacter jejuni in the colon of immunologically naive swine. Am J Trop Med Hyg, 2003.
68(3): p. 70-80.

Abolhassani, M., et al., Mycobacterium bovis BCG induces similar immune responses and
protection by rectal and parenteral immunization routes. Infect Immun, 2000. 68(10): p. 565 7-62.

Zhao, J ., et al., Improved protection of rhesus macaques against intrarectal simian
immunodeficiency virus SI V (mac25 I ) challenge by a replication-competent Ad5hr-Sl V env/rev and
A d5hr-Sl Vgag recombinant priming/gpl20 boosting regimen J Virol, 2003. 77(15): p. 83 54-65.

Miller, W.G., et al., Detection on surfaces and in Caco-2 cells of Campylobacter jejuni cells
transformed with new gfp, yfp, and cfp marker plasmids. Appl Environ Microbiol, 2000. 66(12): p.
5426-36.

Wilson, D.L., et al., Identification of ciprofloxacin-resistant C ampylobacter jejuni by use of a
fluorogenic PCR assay. J Clin Microbiol, 2000. 38(11): p. 3971-8.

F ermer, C. and E0. Engvall, Specific PCR identification and differentiation of the thermophilic
campylobacters, Campylobacterjejuni, C. coli, C. lari, and C. upsaliensis. J Clin Microbiol, 1999.
37(10): p. 3370-3.

Al-Salloom, F.S., et al., Campylobacter-stimulated IN T407 cells produce dissociated cytokine
profiles. J Infect, 2003. 47(3): p. 217-24.

Enocksson, A., eta1., Rectal nitric oxide gas and stool cytokine levels during the course of
infectious gastroenteritis. Clin Diagn Lab Immunol, 2004. 11(2): p. 250-4.

Sansonetti, P.J., et al., Interleukin-8 controls bacterial transepithelial translocation at the cost of
epithelial destruction in experimental shigellosis. Infect Immun, 1999. 67(3): p. 1471-80.

Sansonetti, P.J., et al., Caspase-I activation of IL-1 beta and IL-18 are essential for Shigella
flexneri-induced inflammation Immunity, 2000. 12(5): p. 581-90.

62

35.

36.

37.

38.

39.

40.

41.

42.

43.

44.

45.

Cho, W.S. and C. Chae, Expression of inflammatory cytokines (T NF -alpha, IL- 1, IL-6 and IL-8) in
colon of pigs naturally infected with Salmonella typhimurium and S. choleraesuis. J Vet Med A
Physio] Pathol Clin Med, 2003. 50(10): p. 484-7.

Foss, D.L., M.J. Zilliox, and M.P. Murtaugh, Bacterially induced activation ofinterleukin-18 in
porcine intestinal mucosa. Vet Immunol Immunopathol, 2001. 78(34): p. 263-77.

Elhofy, A. and K.L. Bost, Limited interleukin-I8 response in Salmonella-infected murine
macrophages and in Salmonella-infected mice. Infect Immun, 1999. 67(10): p. 5021-6.

Fera, M.T., et al., Correlation between Helicobacter pylori infection and IL-18 mRNA expression
in human gastric biopsy specimens. Ann N Y Acad Sci, 2002. 963: p. 326-8.

Harris, P.R., et al., Inflammatory cytokine mRNA expression during early and persistent
Helicobacter pylori infection in nonhuman primates. J Infect Dis, 2000. 181(2): p. 783-6.

Tomita, T., et al., Expression of Interleukin-1 8, a Th1 cytokine, in human gastric mucosa is
increased in Helicobacter pylori infection. J Infect Dis, 2001. 183(4): p. 620-7.

Rivera-Amill, V., et al., Secretion of the virulence-associated Campylobacter invasion antigens
ﬁom Campylobacterjejuni requires a stimulatory signal. J Infect Dis, 2001. 183(11): p. 1607-16.

Allen, K.J. and M.W. Grifﬁths, Eﬂect of environmental and chemotactic stimuli on the activity of
the Campylobacterjejuni/MA sigma(28) promoter. FEMS Microbiol Lett, 2001. 205(1): p. 43-8.

Abner, S.R., et al., Response of intestinal epithelial cells to Trichuris suis excretory-secretory
products and the inﬂuence on C ampylobacter jejuni invasion under in vitro conditions. J Parasitol,
2002. 88(4): p. 738-45.

Mansﬁeld, LS. and J .F. Urban, Jr., The pathogenesis of necrotic proliferative colitis in swine is
linked to whipworm induced suppression of m ucosal immunity to resident bacteria. Vet Immunol

Immunopathol, 1996. 50(1-2): p. 1-17.

Wemo, A.M., eta1., Fatal case of Campylobacter lari prosthetic joint infection and bacteremia in
an immunocompetent patient. J Clin Microbiol, 2002. 40(3): p. 1053-5.

63

 

 

 

 

 

 

 

Inoculation Culture C. jejuni gfp PCR 23S rRNA RFLP
QRDR C. jejuni C. coli

C.jejuni 00f4 3of4 40f4 3of4 lof4

1hr

Control 0 of 4 0 of 4 O of 4 0 of 4 4 of 4

1hr

Cjejuni lof4 lof4 20f4 00f4 4of4

24 hr

Control 0 of 4 1 of 4 0 of 4 0 of 4 4 of 4

24hr

 

 

 

 

 

 

 

Table 1: Summary of Campylobacter detection and identiﬁcation. Campylobacters were
detected by culture of feces, or PCR for C. jejuni, gﬁ), or Campylobacter 23S rRNA
targets using total DNA isolated from distal colonic tissue samples as the template.

64

 

Figure l: Ampliﬁcation of the QRDR of C. jejuni gyrA from distal colon DNA from
experimentally infected pigs. Lane 1: 100bp DNA ladder; Lane 2: DNA from C.
jejuni expressing gfp; Lanes 3-6: 24 hour infected pigs; Lanes 7-10: 1 hour infected
pigs; Lanes 11-18: uninfected controls.

65

    

1 2 3 4 s 6 7 8 910111213141516171819

Figure 2: Alu I digest of Campylobacter 23s rRNA gene ampliﬁed from pig tissues
by PCR. Lane 1: 100bp DNA ladder; Lane 2: C. jejuni; Lane 3: C. coli ; Lanes 4-7 :
24 hour infected pigs ; Lanes 8-11 : 1 hour infected pigs ; Lanes 12-19 : uninfected
controls.

66

     

12 3 45 678 9101112131415161718

Figure 3: PCR testing for gfp. A 543 bp ampliﬁed fragment of the gﬂ gene was
ampliﬁed from total DNA isolated from distal colonic pig tissues. Lane 1: 100bp
DNA ladder; Lane 2: DNA from C. jejuni expressing gfp; Lanes 3-6: 24 hour
infected pigs; Lanes 7-10: 1 hour infected pigs; Lanes 11-18: uninfected controls.

67

 

Figure 4: Immunohistochemistry of Campylobacter infected IPEC-l cells.
Immunohistochemistry was performed on IPEC-l cells infected with one of ﬁve
Campylobacter species to determine antibody speciﬁcity. A: C. jejuni 81-176 expressing
gfp; B: C. coli 43134; C: C. lari 43675; D: C. upsaliensis 43954 ; E : C. hyointestinalis
35217. (100X magniﬁcation)

68

 

Figure 6: Immunohistochemistry of distal colon samples using the polyclonal antibody
detected Campylobacter spp. organisms associated with the brush border and in the
lumen (A), within the crypts (B), within the crypts of the LGCs and associated with
inﬂammatory cells (C), and associated with goblet cells, possibly within the cytoplasm
(D) (100X magniﬁcation). One pig had organisms within crypt abscesses (E) (20X
magniﬁcation).

69

 

12345

Figure 6: PCR testing DNA isolated from LCM samples. Primers designed to amplify a
fragment of the gfp gene were used to test DNA isolated from LCM samples. Lane 1:
100bp DNA ladder; Lane 2: DNA from C. jejuni expressing gfp; Lane 3: DNA from a
full thickness tissue sample taken from an infected animal; Lane 4: DNA from
microdissected crypts and submucosa from an infected animal; Lane 5: DNA from
microdissected crypts and submucosa from a control animal.

70

    

112345678'

Figure 7: Alu I digest of Campylobacter 23s rRNA gene PCR product ampliﬁed from
microdissected crypts and submucosa. Lane 1: 100bp DNA ladder; Lane 2: C. jejuni 81-
176 expressing gfp; Lane 3: C. lari 43675; Lane 4: C. upsaliensis 43954; Lane 5: C. coli
43134; Lane 6: DNA from a full thickness tissue sample taken from an infected animal;
Lane 7: DNA from microdissected crypts and submucosa from an infected animal; Lane
8: DNA from microdissected crypts and submucosa from a control animal.

71

GM-CSF
iNOS
IL-1B
lL-18
TNF-a
lL-8
MCP-1
IL-6
1L-12p35
lL-1 2p40
IFNy

IL-4

IL-5
IL-10

IL-13

Distal
Colon

 

LlLllJLlLllelLliﬂLlLlLl

 

 

 

 

 

9V

9@171F][TEN—11117FH‘IIJITVIFIHIT

94.—l

55..

Figure 8: Cytokine expression in distal colonic tissues. Expression of mRNA from a
panel of cytokines was measured by Real Time PCR. The data represents the average
fold change in expression from infected animals when compared to control animals.
Letters denote statistically signiﬁcant changes. A: p50.05; B: p500].

72

€253 N 2.3

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

2: of EEH-<OHOUUHEOOHUOOOUOOH-2<HE 38m

SE ooa <OEOH<O<OOUHO<OOHUU 8.85%

2: on HU<HO<UOH<UO<UU<OU<OHU pension 0339?. 2-»:

2: cm ~ <M§<HOOH<O<UUUOOPOUO<UU<<O-E<..E 02on

Ed com <<<O<<<OEOUEUOHOOEUH 855%

2: com OEO<UUH<UOPOO<U<<O<OP $83.8» Boom: 07.:

2: com <M2<HQ<OHUOHUHU<<OUH<OE<UUHUHUOHO-2<,.G 38m

Ed com HOE<OHO©EO<H<U©H<<UEU 8.63M

Ed com U<OEUU<O<O<UOOHOOHO<O ﬁssion ”woos—é. wk:

2: com §2<H-OO<U<UU<U<O<U<HUH<U<UOOUEO-5mm 0205

2: com OEUHUOO<OO<UHUOOUH 885%

2: com HOFO<OUHUUOOOUUO cSEom ommwox v-1:

2: com ﬂOmm-<UH<U<OOO<UH<UHHUEUO<OU<UU<émH 305

2: 00m OEOOHOHOOO/xOOOEUH 883%

Ed com <OH<H©<<UHUO<<U<UEOHUO<< Beach C 233 ova N71:

2: com g03m-UOO<<0HUOHOH<<<<UOUHU<UU<<U-HMH 38m

2d com <OHUHHUUOOEOEOOO 8.55M

2: com UCHEUEOH<<OO<UUOO “53.—om mohmmq mam—-1:

2: com ﬂOIm-<<<OOO<OPUUU<O<UUH<OHU<UO<FmH 3on

Ed com HUE<OOHUEOEUOOHOOH 3.85M

2: 90m E<O<OOHOHUOO<OOHOH<< Caution NNnowE ck:
eaten—H850 .m-.m 35.—vow 3.5.5 :o_mmouo< anew

 

 

 

Jean—SO

 

 

 

73

3.253 N 03:8

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

2: of <M2<H-<UHUOOHiOOHUOUODUOH-E<mo 38m

2: ooo <OHHOH<O<OOUHO<U©HUO 855%

2c om HU<HO<U©H<U©<UU<UU<OHU Beacon omomwma‘. m Ta:

Ed o2 «nigh-UOH<U<UUUUOHUUO<UO<<U-2<..G 38m

2: oom EOEOEOOEOUHOOHUH 355M

Ed oom OEU<UUH<OOHOO<U<<O<OH oEEom Soon: 2.4:

E: oom <M2<hF<OHO©HUHU<<OUH<OE<OUHUHUOHQE<mo 098m

2: oom HOE<OHOOEO<H<UOH<<UHHU 855%

2: oom U<OEUU<O<O<UOOHOOHO<O Emzcom wwooﬁo?‘ m5:

Ea oom <M§<H-©U<U<UU<U<O<U<HUH<U<UOOUEQE<mo 3.2m

2: oom OEUHUOO<OO<UHUOUOH 8.85%

E: oom HOPU<OUPUOOOOUOO Eanm ommwox Tu:

2: oom _02m-<DH<U<COO<UP<UEUEUO<UO<UU<-HE BEL

2: oom OEHOOHUHOOO<OOOEUH 355m

E: oom <OH<FO<<UHU©<<U<UEOHO©<< @835» A. ~ mwoD ova N g H:

2: SN ﬂOEm-UOOEOHUOHUHEﬁUOOHO<OU<<UFWH 028m

2: oom <OHUEUUOOEOEOOO 885%

SE oom DOPEUEOH<<00<UOUO Beacon mobmmq mam—-1:

E: oom _OEQ<<<OOO<OHUUU<O<UOH<OP”3004\th 090.5

E: ooo HUE<OOHUHOEOOOHOOH 355%

E: oom E<O<U©HOHUOO<OUHOH<< @820.» NNSmE on:
gouge—2.3.80 .m-.m 85.—wow hon—Em eommmoeu< 2.00

 

 

 

3:350

 

 

 

74

Chapter 3: “T richuris suis down regulates proinﬂammatory cytokines in the proximal

colon of pigs concurrently infected with C. jejuni”

Abstract

The swine whipworm Trichuris suis and the bacterial pathogen Campylobacter
jejuni synergize in the colons of pigs to cause mucohemorrhagic diarrhea and severe
pathology. We hypothesize that dysregulation of cytokine immune responses play a role
in pathogenesis of this disease. We analyzed cytokine expression in the jejunum,
proximal and distal colons of pigs infected with either or both of these pathogens to test
this hypothesis in the areas where the pathogens reside. Weaned piglets were orally
inoculated with 2500 embryonated T. suis eggs. Twenty one days after inoculation with
T. suis eggs pigs were orally inoculated with 2x108 cfu Of C. jejuni or sterile milk. All
pigs were euthanized and tissue samples were taken for analysis of cytokine expression
23 days after initial inoculation. C. jejuni was detected by PCR in proximal colon
samples from pigs dually infected with C. jejuni and T. suis. T. suis stimulated increased
expression of IL-13 in the proximal colon, while expression of IL-8, MCP-l, and IL-12
were decreased in pigs infected with T. suis alone, or concurrently with T. suis and C.
jejuni. In the distal colon, increased expression of IL-IB, IL-6, IFNy and IL-13 was
measured in response to T. suis alone. In the jejunum, C. jejuni infection stimulated
increased expression of MCP-l, TNFOI, IL-12p40, IFNy, IL-4, and IL-10, while infection
with C. jejuni and T. suis concurrently caused decreased expression of GM-CSF. These

data indicate that C. jejuni stimulates a proinﬂammatory response and T. suis down

75

regulates proinﬂammatory responses, both in single infection and concurrent infection

with C. jejuni.

76

Introduction

Trichuris suis is a worm parasite that invades the cecum and proximal colon of
swine causing inﬂammation, hemorrhage, and in severe cases necrosis [1]. The severity
of colonic pathology is increased by interaction with large intestinal ﬂora. Rutter and
Beer demonstrated this phenomenon when they showed that speciﬁc pathogen free and
gnotobiotic pigs developed signiﬁcantly milder lesions upon T. suis infection than
conventionally reared pigs with a normal population of intestinal ﬂora [2]. The role of
resident bacteria in Trichuriasis was conﬁrmed by Mansﬁeld and Urban, who showed
amelioration of disease and pathology by antibiotic treatment of T. suis infected pigs.
Additionally, several species of bacteria, including Campylobacter jejuni, were isolated
from abscessed lymphoglandular complexes in the distal colons of pigs that did not
receive antibiotics concurrently [3]. Further studies with germ free piglets showed that T.
suis and C. jejuni synergize to produce severe disease and colonic pathology, while
infection with either agent alone resulted in mild disease and pathology [4]. Severe
disease in this dual infection model is believed to result in part because of inappropriate
cytokine responses which ultimately cause dysregulation of the immune response
necessary to clear the bacteria in the short term.

Cytokines serve to initiate, direct, and sustain immune responses to infectious
agents. Helminths typically stimulate IL-4, IL-5, IL-9, IL-10, and IL-13 production [5-8].
Together these cytokines clear helminth infections by recruiting eosinophils, enhancing
IgE production, stimulating mucus production, and promoting intestinal smooth muscle
contractions that help to expel worms [8-11]. Enteric bacterial pathogens, like

Salmonella and Shigella, stimulate the production of IL-8, IL-IB, IL-18, and TNFOI [12-

77

16]. These cytokines initiate inﬂammation in the gut, control the spread of invading
bacteria, and facilitate elimination of the bacteria from the intestine [12, 13, 16].
Inappropriate cytokine responses hinder clearance of pathogens and resolution of disease.
Mice normally resistant to the intestinal nematodes Nippostrongylus brasiliensis and T.
muris develop chronic disease with suppression of Th 2 cytokines when given exogenous
IL-12 [17, 18]. Additionally, concurrent infections with the intracellular protozoan
parasite Eimeria ferrisi and the nematode IV. brasiliensis increases worm fecundity and
exacerbates disease in inbred mice normally resistant to this chronic nematode infection
[8]-

T. suis infected pigs reared in conﬁnement have increased expression of IL-10 in
mesenteric lymph nodes and mild evidence of secondary bacterial infection in the colon.
T. suis infected pigs reared on dirt lots have increased expression of both IL-10 and IL-12
in mesenteric lymph nodes and develop severe disease with evidence of extensive
secondary bacterial infection in the colon [5]. It is believed that IL-12 is stimulated by
invasive bacteria, and it exacerbates disease and pathology. We examined the local
cytokine responses in the jejunum, proximal colon, and distal colon of pigs singly or
dually infected with T. suis and C. jejuni. Our data show that IL-13 is signiﬁcantly
increased in the proximal colon of T. suis infected pigs while IL-12 and IL-8 are
signiﬁcantly decreased. The cytokine proﬁle in the proximal colons of T. suis infected
pigs did not appear to be signiﬁcantly altered by the presence of C. jejuni. In contrast, IL-
113, IL-6, and IFNy are up-regulated in the distal colon of pigs infected with T. suis only
which may indicate an inﬂammatory reaction to secondary infection from some

component of the resident microbial population.

78

Materials and Methods

Animals

Four week old male and female outbred Landrace-Yorkshire cross pigs were
obtained from the Michigan State University (MSU) swine farm. Animals were housed in
the MSU Research containment facility in groups of 6 for social support. Individual
infection groups were housed in separate rooms to minimize cross contamination. All
animal housing and handling complied with MSU animal use guidelines and National
Institutes of Health guidelines for humane use of laboratory animals. All animals were
tested for the presence of T. suis prior to inoculation by microscopic examination of fecal

ﬂoats for ova.

Bacteria

C. jejuni strain ATCC 33292 that had been rederived by passage through a pig
and stored after the second passage was used for experimental infections. The bacteria
were grown on Bolton Agar supplemented with 5% sheep’s blood for 48 hours at 37°C,
5% C02. Isolated colonies were subcultured for lawn growth for 18-20 hours. The lawns
were harvested with sterile swabs and resuspended in sterile milk to a ﬁnal concentration

of 1x108 cfu/mL.

79

T. suis embryonated eggs

T. suis eggs were collected from adult worms using previously described methods
[19]. Eggs were centrifuged 2-3 times in deionized distilled water at 10,000 rpm for 5
minutes to remove debris. The eggs were resuspended in deionized distilled water and
incubated at 34°C, 5% CO2 for 19 days to stimulate embryonation. After 19 days porcine
bile extract (Sigma Chemical Co., St. Louis, MO) was added to the eggs at a ﬁnal
concentration of 10mg/mL. The eggs were incubated for an additional 3-5 days, then
examined microscopically for L1 larval development and movement. Embryonated eggs
were collected, centrifuged at 10,000rpm for 10 minutes, and then resuspended in sterile
water. A subsample was loaded onto a hemocytometer and examined microscopically to
enumerate embryonated eggs, diluted to 1,250 eggs per mL, and stored at 4°C until

inoculation.

Experimental design and inoculation procedure

All pigs were tested for the presence of C. jejuni by culture of rectal swabs prior
to initial inoculation, and immediately prior to euthanasia. Rectal swabs were transported
to the laboratory in Cary Blair (Difco Laboratories, Detroit, MI) transport medium for
Campylobacter isolation and identiﬁcation.

For experimental inoculations (Figure l), weaned piglets were divided into 4
groups: Group 1- Uninfected control; Group 2- C. jejuni; Group 3- T. suis; Group 4- C.

jejuni and T. suis. On day 0, groups 3 and 4 were inoculated by oral gavage using a sterile

80

12 gauge steel ball feeding needle with 2500 embryonated T. suis eggs resuspended in 2
mLs sterile milk. Groups 1 and 2 were given 2 mLs of sterile milk only. Twenty one days
after inoculation with T. suis eggs, groups 2 and 4 were inoculated with 2x108 cfu of C.
jejuni by oral gavage. Groups 1 and 3 were given 2 mLs of sterile milk. Twenty three
days after initial inoculations, all pigs were euthanized with an intravenous injection of

86 mg/kg sodium pentobarbital (Fatal Plus®, Vortech Pharmaceuticals, Dearborn, MI).

Sample collection

To eliminate cross contamination, a separate set of sterile dissection instruments
was used for each group. For sample collection, the abdomen was opened by a midline
incision and the small and large intestines were exteriorized. Full thickness samples of
the jejunum, proximal colon, and distal colon were collected using sterile scissors and
forceps. One set of samples was ﬁxed in 3.7% formaldehyde for histology and
immunohistochemistry. A second set was placed in cryovials and snap frozen in liquid

nitrogen for nucleic acid isolation.

C. jejuni isolation and identification

All pigs were tested for the presence of C. jejuni by culture of rectal swabs prior
to inoculations, and immediately prior to euthanasia. Swabs were placed in Cary Blair
transport media (Difco Laboratories, Detroit, MI) and stored on ice for transport to the
laboratory where they were processed. Rectal swabs were streaked onto Preston selective
agar (Oxoid Inc., Basingstoke, Hampshire, England) and incubated at 42°C, 5% CO2 in

humidiﬁed air for 48 hours. Cells from isolated colonies were tested by PCR using the

81

previously designed primers JL238 (5' TGG GTG CTG TTA TAG GTC GT 3') and
JL239 (5' GCT CAT GAG AAA GTT TAC TC 3') which are speciﬁc for the quinolone
resistance-determining region (QRDR) of the gyrA gene of C. jejuni [20].

In addition to culture of rectal swabs, tissue samples from each segment of
intestine were tested for the presence of Campylobacter organisms from each pig. Total
DNA was isolated from approximately 25 mg of frozen tissue using a Qiagen DNEasy
tissue kit (Qiagen, Valencia, CA) following the manufacturer’s protocol. The DNA was
quantiﬁed spectrophotometrically (Beckman, Fullerton, CA). The C. jejuni QRDR target
was ampliﬁed by PCR from 500ng of total DNA isolated from full thickness tissue
samples using the previously designed primers JL238 and JL239 [20]. Amplicons were
electrophoresed on a 1.8% agarose gel in 1X Tris-Acetate-EDTA (TAE) buffer with 0.2
ug/mL ethidium bromide. The fragments were visualized by UV illumination on an
Alpha Imager Gel documentation system (Alpha Innotech Corporation, San Leandro,
CA).

A Restriction Fragment Length Polymorphism (RFLP) assay to detect and
distinguish C. jejuni, C. coli, C. lari, and C. upsaliensis by analysis of a portion of the
23S rRNA gene was used to detect thermophilic Campylobacters in experimental animals
[21]. The target was ampliﬁed from 500ng of total pig tissue DNA using primers
THERMI (5'-TAT TCC AAT ACC AAC ATT AGT-3') and THERM4 (5'-CTT CGC
TAA TGC TAA CCC-3'). The amplicons were digested overnight with either AluI or
Tsp5091 following the manufacturers protocols (New England Biolabs, Beverly, MD),
and the resultant fragments were separated electrophoretically on a 2% agarose gel in 1X

Tris-Acetate-EDTA (TAE) buffer with 0.2 ug/mL ethidium bromide. The fragments were

82

visualized by UV illumination on an Alpha Imager Gel documentation system (Alpha

Innotech Corporation, San Leandro, CA).

Histology and Immunohistochemistry

Tissues were examined microscopically to detect T. suis larvae, to evaluate
histological changes, and to localize speciﬁcally stained Campylobacter organisms. Fixed
tissue samples were embedded in parafﬁn and 5 pm sections were cut and adhered to
charged slides. Slides were routinely dehydrated in progressively higher concentrations
of ethanol, deparafﬁnized and rehydrated. For histological evaluation, slides were stained
with Mayer's hematoxylin and eosin and cover slipped. For irnmunostaining, sections of
selected tissues were incubated in preheated antigen retrieval buffer (Dako, Carpinteria,
CA) for 20 minutes at 100° C. Slides were removed from the steamer and left in the
retrieval solution at room temperature to cool down for 20 minutes, rinsed in distilled
water and transferred to Tris-HCL buffer. Endogenous peroxidase was blocked for 15
rrrinutes with 3% hydrogen peroxide. Non-speciﬁc immunoglobulin binding blocking was
done by a 10 minute incubation with a protein-blocking agent (Dako, Carpinteria, CA)
prior to application of the primary antibody. A rabbit polyclonal anti-C. jejuni antibody
(Biogenesis, New Fields, UK) was used as the primary antibody at a concentration of
1:100. Sections were stained with a Dako autostainer using a labeled streptavidin-
immunoperoxidase staining procedure (Dako Corp., Carpinteria, CA). The
immunoreaction was visualized with ABC (Dako Corp., Carpinteria, CA). Sections were

counterstained with Mayer’s hematoxylin, then dehydrated, cleared and mounted.

83

Cytokine Analysis

T. suis resides in the cecum and proximal colons of pigs [2], while C. jejuni can
colonize throughout the small and large intestine of pigs (Mansﬁeld et al, unpublished
results; [22]. To determine the effect of infection with T. suis, C. jejuni, or both on local
immune responses in areas where these pathogens are present, we took samples from the
jejunum, proximal colon, and distal colon. mRNA expression of 15 cytokines,
chemokines, and iNOS was measured from each tissue using a real time PCR assay. Total
RNA was isolated from frozen tissue samples from each pig using Trizol reagent
(GibcoBRL Life Technologies). Brieﬂy, 3mm3 tissue samples were homogenized in
Trizol using a Polytron® tissue homogenizer (Kinematica, Cincinnati, OH). RNA was
extracted with chloroform, precipitated with isopropanol, and washed with ethanol.
Pellets were air dried, then resuspended in DEPC treated H20 (GibcoBRL Life
Technologies). DNA contamination was removed by treatment with approximately 27
Kunitz units of DNase I for 15 minutes. Samples were repuriﬁed on a Qiagen RNEasy
column following the manufacturer’s protocol (Qiagen, Valencia, CA). RNA
concentration and integrity were determined using an Agilent Bioanalyzer 2100 (Agilent
Technologies, Palo Alto, CA). For each sample, 10 pg of total RNA was reverse
transcribed with random primers using a Stratagene First Strand RT-PCR kit (Stratagene,
La Jolla, CA). First strand cDNA was used as the template in real time PCR reactions.
Real time PCR primers and probes designed using the Primer Express (Applied
Biosystems, Foster City, CA) software package were used to analyze l4 cytokines
(Chapter 2, Table l). The 18S rRNA probe was obtained from Applied Biosystems and

was used as a constitutively active gene for normalization. PCR was performed using a

84

commercially available kit (Brilliant kit, Stratagene, La Jolla, CA) on an ABI PRISM
79OOHT (Applied Biosystems). Ampliﬁcation conditions were as follows: 50°C for 2
minutes; 95°C for 10 minutes; 40 cycles of 95°C 15 seconds and 60°C for 1 minute.
Fluorescence signals measured during ampliﬁcation were processed post-ampliﬁcation.
Threshold cycle (CI) values were determined using values of 0.2 for TET, 0.1 for VIC,

and 0.05 for FAM reporter dyes. .

Statistical Analysis

The cytokine mRNA expression data for each tissue was analyzed separately

 

using Statview 5.0 for Macintosh (Abacus Concepts, Berkeley, Calif). The C, value for ’7
the 18S ribosomal subunit was subtracted from the CI value for each cytokine message to
normalize for RNA content prior to statistical analysis. One-way analysis Of variance
(ANOVA) was performed to determine the statistically signiﬁcant differences between
treatments. A Fisher’s least squares difference test was performed to determine
differences between infected and control animals at each time point. P values of 0.05 or

less were considered statistically signiﬁcant.

85

 

Results

C. jejuni isolation and identification

C. jejuni was not isolated from fecal cultures of any pigs before or aﬁer infection
(Table 1). We previously found culture alone to be inadequate for detection and
identiﬁcation of C. jejuni from experimentally infected pigs. In addition, to correlate our
results from analysis of cytokine expression we needed to determine if C. jejuni was
present in all intestinal segments sampled. Therefore we used 2 PCR assays to detect C.
jejuni from the jejunum, proximal colon, and distal colon separately. C. jejuni was not
detected from any tissue samples from any group using primers speciﬁc for the QRDR of
the gyrA gene of C. jejuni. Analysis of the Campylobacter 23S rRNA RFLP
demonstrated that C. jejuni, C. coli, and a Campylobacter species that gave a unique
pattern were present in our experimental animals (Figure 2).

We had previously determined that the polyclonal antibody raised against C.
jejuni used for immunohistochemistry cross reacted with C. lari. We used primers
speciﬁc for the Ipr gene of C. lari to detect C. lari from tissue samples [23]. The target

was not ampliﬁed from any tissues of any pigs (data not shown).

Histopathology

Haemotoxylin and Eosin (H&E) stained sections of jejunum, proximal colon, and

distal colon of each pig were examined microscopically for the presence of T. suis larvae

86

and histological changes due to infection. T. suis larvae were seen in proximal colon
sections from 5 of 6 pigs in group 3, and 5 of 6 pigs in group 4. NO larvae were seen in
groups 1 or 2 (Figure 3).

Control animals (Group 1) had no proliferative, inﬂammatory, or degenerative
lesions in any of the tissues examined and were therefore used as a normal comparison
for all infection groups. All lesions were conﬁned to the colon. Colonic lesions due to C.
jejuni or T. suis infection were similar to those described in germ free piglets [4]. Pigs in
group 2 (C. jejuni only) developed focal mild crypt distension with sloughed epithelial
cells, mucus and bacteria in the crypts in the distal colon (Figure 4A) and LGCs (Figure 5
A). There was no signiﬁcant inﬂammatory inﬁltrate into the lamina propria on day 23
after inoculation. Pigs infected with T. suis only developed mild to moderate colitis with
goblet cell hyperplasia and crypt distension with mucus in superﬁcial lamina propria.
Lesions were found only in the areas where T. suis larvae were located. A mixed
inﬂammatory inﬁltrate was present in the lamina propria and included lymphocytes,
plasma cells, neutrophils, and eosinophils (Figure 4B). Where present crypts of LGCs
were not distended (Figure SB). Pigs infected with both C. jejuni and T. suis developed
mild colitis with goblet cell hyperplasia and crypt distension with mucus in superﬁcial
lamina propria in the areas where T. suis larvae were located. A mixed inﬂammatory
inﬁltrate in the lamina propria included lymphocytes, plasma cells, neutrophils, and
eosinophils. Entrapped crypts of the LGCs were distended with mucus and few sloughed

epithelial cells (Figure 5C).

87

Immunohistochemistry

Sections from jejunum, proximal colon, and distal colon were stained with a
polyclonal antibody directed against C. jejuni to determine the location of the bacteria.
We had previously determined that the antibody cross reacted with C. lari (Chapter 2,
ﬁgure 4) and therefore concluded that positive staining in tissue sections could be due to
either C. jejuni or C. lari. Positively stained organisms were not detected in jejunal
sections of any groups (data not shown). Positively stained organisms were detected in
the colonic crypts and on the mucosal surface from all groups, regardless of infection
status (Figure 6). Staining patterns were focal to multifocal in uninfected controls and
pigs infected singly with either C. jejuni or T. suis. Pigs dually infected with T. suis and

C. jejuni exhibited a diffuse staining pattern with more intense staining (Figure 6).

Cytokine expression

We measured expression of mRNA from 15 cytokines, chemokines, and iNOS in
each segment of intestine sampled in an effort to determine if T. suis or C. jejuni had a
signiﬁcant effect on cytokine expression, and to determine if the affected cytokines were

indicative of a speciﬁc response.

Cytokine expression patterns were distinct in each segment of the intestine tested
(Figure 7). In the proximal colon of the T. suis only and dual infection groups, lL-13
expression was increased 3 to 4 fold while IL-12 and IL-8 expression was decreased 2 to

4 fold when compared to uninfected controls. In the dual infection group, expression of

88

IL-5 and MCP-l were each decreased 2 fold compared to controls. These data indicate
that T. suis stimulates IL-13 in its local environment while decreasing expression of
proinﬂammatory cytokines. In the dual infection group this occurred despite the presence

of C. jejuni in the tissue as conﬁrmed by PCR analysis.

In the distal colon Of pigs infected with T. suis only, expression of IL-IB, IL-6,
IFNy, and lL-l3 was increased 3 to 4 fold over controls. Infection with C. jejuni only or
dual infection with C. jejuni and T. suis did not signiﬁcantly affect expression of any
cytokines, chemokines, or iNOS in the distal colon. These data indicate that a
predominately proinﬂammatory response has been initiated in the distal colons of pigs

infected with T. suis only.

In the jejunum, expression of MCP-l, TNFOI, IL-12p40, IFNy, IL-4, and IL-10
were increased 3 to 5 fold in the C. jejuni only group when compared to controls.
Expression of GM-CSF was decreased 5 fold in the dual infection group when compared
to controls. These data indicate that C. jejuni stimulates a predominately proinﬂammatory
response in the jejunum, while T. suis may be down regulating proinﬂammatory

cytokines.

89

 

Discussion

Oral infection of weaned piglets with T. suis and C. jejuni elicited signiﬁcant
changes in cytokine expression in both the small and large intestine. The effects of T. suis
were predominantly in the proximal colon, with some effects in the distal colon. C. jejuni
affected cytokine expression mainly in the jejunum, with few changes in the colon. T.
suis stimulated 3 to 4 fold increases in IL-13 expression in the proximal colon in groups 3
and 4, which is typical of a helminth infection. Concurrently, IL-8 was signiﬁcantly
decreased in both groups, and in the case of dual infection, MCP-l was also decreased.
Increased expression of IL-8 in the distal colon at both 1 and 24 hours after inoculation
with C. jejuni suggested an important role for this cytokine in host resistance to C. jejuni.
Additionally, IL-8 has been shown to be critical in controlling the spread of Shigella
ﬂexneri in a rabbit ligated intestinal loop model of infection [12]. Histological studies of
T. suis infection have shown that bacteria invade and proliferate in the lamina propria and
submucosa surrounding the site of worm attachment [2, 3], and C. jejuni was detected in
4 of 6 pigs (66%) in group 4 by RFLP analysis. Taken together, these data suggest that
decreased expression of IL-8 due to the presence Of T. suis may facilitate C. jejuni
invasion and proliferation in the lamina propria and submucosa, similar to the Shigella
model. Decreased expression of IL-8 can be attributed at least in part to IL-13, as IL-13
and IL-4 have been shown to inhibit IL-8 production from intestinal epithelial cells [24].

T. suis also signiﬁcantly decreased expression of both subunits of IL-12 in the
proximal colon. IL-12 is produced mainly by antigen presenting cells, including

monocytes, macrophages, and dendritic cells, and drives a Th 1 directed response by

90

inducing IFNy production from T cells [25, 26]. However, IL-l3, which is increased in
resistant mouse strains infected with T. muris [9], has been shown to downregulate
production of IL-12 [27, 28]. Thus the decreased expression of both subunits of IL-12 in
the proximal colons of pigs infected with T. suis can be attributed to the
immunomodulatory effects of IL-l3. Our data is in contrast to the data of Mansﬁeld and
colleagues who measured increased expression of IL-12 expression in the mesenteric
lymph nodes of T. suis infected pigs with evidence of secondary bacterial infection [5].
However, Mansﬁeld and colleagues used competitive RT-PCR, which is not as accurate
for quantitative analysis of cytokine expression as real time PCR. Another explanation
for this discrepancy is that they measured expression in a draining mesenteric lymph
node while we measured expression in the intestinal tissue where worms reside.
Comparison of cytokine expression in draining lymph nodes and intestinal tissues
directly in contact with the pathogen shows that the cytokine responses are not identical
[6, 29]. Detailed analysis of cell types and cytokine expression comparing intestinal
tissue to draining lymph nodes would enhance our knowledge of the swine immune
response to T. suis.

In addition to decreased expression of proinﬂammatory cytokines in the proximal
colon, IL-5 was decreased in the dual infection group. This was an unexpected ﬁnding.
IL-5 functions in eosinophil recruitment and it has been previously shown to be up
regulated in helminth infections along with other classical Th 2 cytokines [8]. However,
while treatment of helminth infected mice with anti-IL-5 antibody abolishes peripheral
and tissue eosinophilia, this does not enhance survival of Heligmosomoides worms [10].

In experimental C. jejuni infections, oral IL-5 pretreatment of mice reduces intestinal

91

colonization by C. jejuni, suggesting a protective effect of this cytokine for the host [30].
It follows that IL-5 may be down regulated by C. jejuni as a mechanism to evade the
immune response and establish infection. IL-5 expression was decreased in the proximal
colon of pigs infected with only C. jejuni, however the decrease was not statistically
signiﬁcant. Nevertheless, decreased expression in both groups that received C. jejuni
suggests that IL-5 plays an important role in pathogenesis of Campylobacteriosis in vivo.
In contrast to the proximal colon, IL-10, IL-6, IFNy and IL-13 were up regulated
in the distal colon of pigs infected with T. suis only. These effects can be attributed in
part to the effects of ESP in the distal colon. Up regulation of IL-IB, IL-6, IFNy is
indicative of an inﬂammatory reaction. In vitro, ESP treatment of differentiated IPEC-l
cells causes a dose dependent cytotoxic response and signiﬁcantly decreased TER [31].
ESP also stimulates secretion of IL-6 from both differentiated and undifferentiated IPEC-
1 cells in vitro (Parthasarathry et al, unpublished results). The destructive effects of ESP
on intestinal epithelial cells could stimulate an inﬂammatory reaction in the distal colon.
Another possibility for the inﬂammatory response seen is T. suis stimulated secondary
bacterial infection in the distal colon. Several resident bacterial species have been
cultured from abscessed lymphoglandular complexes in the distal colons of T. suis
infected pigs, including C. jejuni, C. coli, an unidentiﬁed Campylobacter species, and E.
coli [3]. C. jejuni and C. coli have been shown to induce IL-8 secretion from cultured
human INT-407 intestinal epithelial cells [32]. C. coli was detected in the distal colons of
2 of 6 pigs infected with T. suis only and an unidentiﬁed Campylobacter spp. was
detected in 3 of 6 pigs from this group by RFLP analysis of the Campylobacter 23S

rRNA gene. Thus it is possible that either of these species invaded tissues in the distal

92

colon and initiated an inﬂammatory response. Increased expression of IL-13 could also
be attributed to the effects of excretory/secretory product (ESP) produced by the 4th stage
larvae in the proximal colon that have traveled to the distal colon. A 20kDa glycoprotein
isolated from ESP collected from adult worms in vitro can be recognized by serum
antibodies from T. suis infected pigs as early as 21 days after inoculation [33]. This
indicates that the larval stages present in pigs 21 days after infection are stimulating an
immune response. Taken together, these data indicate that direct effects of ESP on
intestinal epithelial cells could account for increased cytokine production in vivo at sites
distal to worm attachment.

In the jejunum of pigs infected with C. jejuni only, expression of MCP-l, TNFOI,
IL-12p40, IFNy, IL-4, and IL-10 was increased. Although PCR testing and IHC did not
conﬁrm the presence of Campylobacters in this tissue, it is still likely that the response is
due to C. jejuni stimulation. C. jejuni stimulates intracellular production of MCP-l,
TNFOI, IFNy, IL-4, and IL-10 from INT-407 cells [34, 35], thus C. jejuni interaction with
intestinal epithelial cells in the jejunum could account for the cytokine changes seen. In
the jejunum of pigs dually infected with C. jejuni and T. suis, expression of GM-CSF is
decreased 5 fold compared to uninfected controls, suggesting that similar to the proximal
colon T. suis infection down regulates expression of proinﬂammatory cytokines. These
effects could be due to systemic IL-13 and IL-10 produced by lymphocytes stimulated by
worm antigens in draining lymph nodes.

In a separate study, secreted levels of IL-10, IL-8, TNFOI, IL-4, IL-6, and IL-10
were measured from the feces of these experimentally infected pigs (Cunningham et al,

unpublished results; Parthasarathy et al, unpublished results). On day 23, which

93

corresponds to the day tissues were taken for analysis Of cytokine expression, IL-1 0
secretion was increased in the feces of pigs infected with C. jejuni only or dually infected
with T. suis and C. jejuni when compared to controls. No other cytokines were increased
in the feces on day 23. Increased secretion of IL-1 [3 in the feces on day 23 correlates
with increased expression in the distal colon of the dual infected group and in the
jejunum of the C. jejuni only infected group. Interestingly, IL-1 [3 expression was also
signiﬁcantly increased in the distal colon in the T. suis only group, but there was not a
corresponding increase in fecal secretion. Taken together, these data suggest that C.
jejuni stimulated increases in IL-1 13, even in the presence of T. suis. It also suggests that
IL-IB production may be at least in part transcriptionally regulated, but a direct
correlation is not evident from these studies. On day 22, which corresponds to 24 hours
after inoculation with C. jejuni, secretion of lL-l B, TNFor, and IL-10 were increased in
all infection groups compared to controls, and IL-4 was increased in the dually infected
group. This suggests that the maximal effects of T. suis and C. jejuni infection on
production of these cytokines occurs at a time earlier that tissues were taken for
expression analysis. Additional experiments measuring both secreted protein in feces and
mRN A from tissues over time are necessary to clarify this issue.

We found that culture alone is not sensitive enough for detection of C. jejuni from
conventionally reared pigs. This correlates with infection studies of orally inoculated day
old chicks. C. jejuni was only recovered by culture of cecal contents from 64% to 72% of
infected animals [36, 37]. We found that to increase the sensitivity of detection in our
study, we needed to use PCR assays. RF LP analysis of the Campylobacter 23S rRNA

gene was able to detect C. jejuni in proximal colon samples from 66% (4 of 6) of pigs

94

dually infected with C. jejuni and T. suis, and 16% (1 of 6) from each of the control, T.
suis only, and C. jejuni only groups. Interestingly, RFLP analysis showed a distinct
banding pattern that was not consistent with C. jejuni, C. lari, C. coli, or C. upsaliensis in
some pigs (Figure 1, panel A, lanes 7 and 15). Fermer and Engvall noted that PCR
product of the correct size was ampliﬁed from one strain of C. mucosalis. This amplicon
was digested with the restriction enzymes and had a unique RFLP pattern, distinguishable
from C. jejuni, C. lari, C. coli, or C. upsaliensis [21]. Unfortunately the authors failed to
show the RF LP pattern for C. mucosalis and we were unable to compare the unique
pattern generated in this study to the pattern for C. mucosalis. In addition, positive
staining on immunohistochemistry was seen in colonic samples from all groups.
Although antibody validation tests showed that the antibody cross reacts with C. lari, all
pigs were negative when tested for the presence of C. lari by PCR (data not shown).
Taken together, these data indicate that multiple Campylobacter species were present in
the colons of our experimental animals and likely impacted the cytokine responses
elicited by experimental infections. Further investigation of Campylobacter species that
are present in our experimental groups is currently being pursued by cloning and
sequencing Campylobacter 16S rRNA genes ampliﬁed from tissue samples.
Histologically, the lesions observed in pigs infected singly or dually with T. suis
and C. jejuni were consistent with what has been demonstrated previously. Colostrum
deprived piglets orally inoculated with a large dose (4x108 cfu) of C. jejuni have normal
small intestines. In contrast, lesions in the colon include diffuse mild to moderate erosive
colitis and typhlitis with inﬁltration of the lamina propria with small numbers of

neutrophils [22]. Germ free piglets develop only mild colonic lesions in response to a low

95

dose (10’5 cfu) of C. jejuni with low level inﬁltration of the lamina propria and submucosa
with lymphocytes, macrophages, neutrophils, and eosinophils [4]. T. suis infection causes
goblet cell hyperplasia and a mixed inﬂammatory inﬁltrate in the proximal colons of both
germ free and conventionally reared pigs [3, 4]. The mild to moderate colitis with goblet
cell hyperplasia and mixed inﬂammatory inﬁltrate in the lamina propria and submucosa
elicited in our study conﬁrmed that pigs inoculated with T. suis eggs did become infected,
and allowed for correlation between the presence of the worm and cytokine changes seen
in the colon.

The balance between inﬂammatory cytokines and anti-inﬂammatory cytokines
allows for defense against invading pathogens, while preventing fatal outcomes by an
overwhelming inﬂammatory response. Inﬂammatory cytokines are produced in the early
stages of infections to combat bacteria and in some cases eliminate the threat before a
speciﬁc humoral or cell-mediated response is generated. In normal immunocompetent
individuals, anti-inﬂammatory cytokines are naturally produced after induction of
inﬂammatory cytokines to temper the response and prevent serious tissue destruction.
Based on time course studies in murine disease models, anti-inﬂammatory cytokines are
generally produced after the inﬂammatory cytokines have eliminated the pathogen.
Helminths generally stimulate a Th2 directed response with increased production of IL-4,
lL-5, IL-10, and IL-13 in gut associated lymphoid tissues [8, 38, 39]. These cytokines are
essential for modifying the enteric environment to be inhospitable for helminths,
ultimately resulting in expulsion of the pathogen and thus resolution of disease [8, 10, 11,
29]. A byproduct of these cytokines is downregulation of proinﬂammatory cytokines [24,

27]. In the case of dual infection, the anti-inﬂammatory response to the worm may

96

actually have a detrimental effect on the protective proinﬂammatory response toward
bacteria. Indeed, there has been one reported case of dual C. jejuni and T. suis infection in
a human with severe colitis, toxic megacolon, and acute renal failure [40]. The lack of an
appropriate level of proinﬂammatory cytokines could facilitate invasion of tissues by
resident bacterial populations, causing cellular destruction. The data presented in this

study support this hypothesis.

97

lO.

1].

12.

14.

15.

16.

17.

References

Batte, E.G., et al., Pathophysiology of swine trichuriasis. Am J Vet Res, 1977. 38(7): p. 1075-9.

Rutter, J .M. and RJ. Beer, synergism between T richuris suis and the microbial flora of the large
intestine causing dysentery in pigs. Infect Immun, 1975. 11(2): p. 395-404.

Mansﬁeld, LS. and J .F. Urban, Jr., The pathogenesis of necrotic proliferative colitis in swine is
linked to whipworm induced suppression of mucosal immunity to resident bacteria. Vet Immunol
Immunopathol, 1996. 50(1-2): p. 1-17.

Mansﬁeld, L.S., et al., Enhancement of disease and pathology by synergy of Trichuris suis and
Campylobacterjejuni in the colon of immunologically naive swine. Am J Trop Med Hyg, 2003.
68(3): p. 70-80.

Mansﬁeld, L.S., et al., Construction of internal cDNA competitors for measuring IL-10 and IL-12
cytokine gene expression in swine. Vet Immunol Immunopathol, 1998. 65(1): p. 63-74.

Dawson, H.D., et al., Localized multigene expression patterns support an evolving ThI/ThZ-like
paradigm in response to infections with Toxoplasma gondii and Ascaris suum. Infect Immun,
2005. 73(2): p. 1116-28.

Faulkner, H., et al., A ge- and infection intensity-dependent cytokine and antibody production in
human trichuriasis: the importance of IgE. J Infect Dis, 2002. 185(5): p. 665-72.

Urban, J .F., Jr., et al., Local TH I and TH2 responses to parasitic infection in the intestine:
regulation by [EN-gamma and IL-4. Vet Immunol Immunopathol, 1996. 54(1-4): p. 337-44.

Bancroft, A.J., A.N. McKenzie, and R.K. Grencis, A critical role for IL-13 in resistance to
intestinal nematode infection. J Immunol, 1998. 160(7): p. 3453-61.

F inkelman, F.D., et al., Regulation and biological ﬁmction of helminth-induced cytokine
responses. Parasitol Today, 1991. 7(3): p. 62-6.

Urban, J ., et al., IL-I3-mediated worm expulsion is B7 independent and IF N-gamma sensitive. J
Immunol, 2000. 164(8): p. 4250-6.

Sansonetti, P.J., eta1., Interleukin-8 controls bacterial transepithelial translocation at the cost of
epithelial destruction in experimental shigellosis. Infect Immun, 1999. 67(3): p. 1471-80.

Sansonetti, P.J., et al., Caspase-l activation of IL-1 beta and IL-18 are essential for Shigella
ﬂexneri-induced inflammation. Immunity, 2000. 12(5): p. 581-90.

Enocksson, A., et al., Rectal nitric oxide gas and stool cytokine levels during the course of
infectious gastroenteritis. Clin Diagn Lab Immunol, 2004. 11(2): p. 250-4.

F 053, D.L., M.J. Zilliox, and M.P. Murtaugh, Bacterially induced activation of interleukin-1 8 in
porcine intestinal mucosa. Vet Immunol Immunopathol, 2001. 78(3-4): p. 263-77.

Cho, W.S. and C. Chae, Expression of inflammatory cytokines (INF-alpha, IL-1, IL-6 and IL-8) in
colon of pigs naturally infected with Salmonella typhimurium and S. choleraesuis. J Vet Med A
Physio] Pathol Clin Med, 2003. 50(10): p. 484-7.

Bancroft, A.J., et al., Interleukin-12 promotes a chronic intestinal nematode infection. Eur J
Immunol, 1997. 27(4): p. 866-70.

98

l8.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

Finkelman, F .D., et al., Effects of interleukin 12 on immune responses and host protection in mice
infected with intestinal nematode parasites. J Exp Med, 1994. 179(5): p. 1563-72.

Hill, D.E., et al., Trichuris suis: a zinc metalloproteaseﬁ'om culture ﬂuids of adult parasites. Exp
Parasitol, 1993. 77(2): p. 170-8.

Wilson, D.L., et al., Identification of ciprofloxacin-resistant C ampylobacter jejuni by use of a
fluorogenic PCR assay. J Clin Microbiol, 2000. 38(1 1): p. 3971-8.

Fermer, C. and E0. Engvall, Specific PCR identification and diﬂerentiation of the thermophilic
campylobacters, Campylobacterjejuni, C. coli, C. lari, and C. upsaliensis. J Clin Microbiol, 1999.
37(10): p. 3370-3.

Babakhani, F.K., G.A. Bradley, and LA. Joens, Newborn piglet model for campylobacteriosis.
Infect Immun, 1993. 61(8): p. 3466-75.

Wemo, A.M., et al., Fatal case of Campylobacter lari prosthetic joint infection and bacteremia in
an immunocompetent patient. J Clin Microbiol, 2002. 40(3): p. 1053-5.

Lugering, N., et al., Interleukin (IL)-l3 and IL-4 are potent inhibitors of IL-8 secretion by human
intestinal epithelial cells. Dig Dis Sci, 1999. 44(3): p. 649-55.

Gately, M.K., et al., The interleukin-lZ/interleukin-l2-receptor system: role in normal and
pathologic immune responses. Annu Rev Immunol, 1998. 16: p. 495-521.

Sutterwala, PS. and D.M. Mosser, The taming of IL-1 2: suppressing the production of
proinflammatory cytokines. J Leukoc Biol, 1999. 65(5): p. 543-5].

Muchamuel, T., et al., IL-I3 protects mice from lipopolysaccharide-induced lethal endotoxemia:
correlation with down-modulation of TNF-alpha, IFN-gamma, and IL- 12 production. J Immunol,
1997. 158(6): p. 2898-903.

Murphey, E.D., et al., Diminished bacterial clearance is associated with decreased IL-12 and
interferon-gamma production but a sustained proinflammatory response in a murine model of
postseptic immunosuppression. Shock, 2004. 21(5): p. 415-25.

Schopf, L.R., et al., IL-10 is critical for host resistance and survival during gastrointestinal
helminth infection. J Immunol, 2002. 168(5): p. 2383-92.

Baqar, S., N.D. Pacheco, and F.M. Rollwagen, Modulation of mucosal immunity against
Campylobacterjejuni by orally administered cytokines. Antimicrob Agents Chemother, 1993.
37(12): p. 2688-92.

Abner, S.R., et al., Response of intestinal epithelial cells to Trichuris suis excretory-secretory
products and the inﬂuence on Campylobacterjejuni invasion under in vitro conditions. J Parasitol,

2002. 88(4): p. 738-45.

Hickey, T.E., et al., Campylobacterjejuni cytolethal distending toxin mediates release of
interleukin-8 from intestinal epithelial cells. Infect Immun, 2000. 68(12): p. 6535-41.

Hill, D.E., R.D. Romanowski, and J .F. Urban, Jr., A Trichuris specific diagnostic antigen ﬁom
culture ﬂuids of Trichuris suis adult worms. Vet Parasitol, 1997. 68(1-2): p. 91-102.

Al-Salloom, F.S., et al., C ampylobacter-stimulated IN T407 cells produce dissociated cytokine
profiles. J Infect, 2003. 47(3): p. 217-24.

99

35.

36.

37.

38.

39.

40.

Bakhiet, M., et al., Induction of alpha and beta chemokines by intestinal epithelial cells stimulated
with Campylobacterjejuni. J Infect, 2004. 48(3): p. 236-44.

Beery, J.T., M.B. Hugdahl, and M.P. Doyle, Colonization of gastrointestinal tracts of chicks by
Campylobacterjejuni. Appl Environ Microbiol, 1988. 54(10): p. 2365-70.

Ziprin, R.L., et al., The absence of cecal colonization of chicks by a mutant of Campylobacter
jejuni not expressing bacterial fibronectin—binding protein. Avian Dis, 1999. 43(3): p. 586-9.

Ramaswamy, K., D. Negrao-Correa, and R. Bell, Local intestinal immune responses to infections
with Trichinella spiralis. Real-time, continuous assay of cytokines in the intestinal (aﬂerent) and
eﬂerent thoracic duct lymph of rats. J Immunol, 1996. 156(11): p. 4328-37.

Svetic, A., eta1., A primary intestinal helminthic infection rapidly induces a gut-associated
elevation of Th2-associated cytokines and IL-3. J Immunol, 1993. 150(8 Pt 1): p. 3434-41.

Shin, J .L., et al., Does whipworm increase the pathogenicity of Campylobacter jejuni ? A clinical
correlate of an experimental observation. Can J Gastroenterol, 2004. 18(3): p. 175-7.

100

   

 

 

 

 

 

 

Day 0 Day 21 Day 23
Inoculation Inoculation Euthanasia

Group 1: Milk Group 1: Milk Group 1: Milk
Group 2: Milk Group 2: C. jejuni Group 2: C. jejuni
Group 3: T. suis Group 3: Milk Group 3: T. suis
Group 4: T. suis Group 4: C. jejuni Group 4: Dual

 

 

 

 

 

Figure 1: Inoculation timeline: Weaned piglets were divided into 4 groups of 6 pigs each
and inoculated with 2,500 embryonated T. suis eggs, 108 cfu C. jejuni 33292, or both as
shown. All inoculations were by oral gavage in 2mL of sterile milk.

101

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Campylobacter 23S
QRDR rRNA RFLP
Tissue Campy
Infection Culture Feces DNA Cj Cc spp.
Control 0/6 0/6 0/6 0/6 5/6 2/6
Jejunum C. jejuni 0/6 0/6 0/6 0/6 2/6 5/6
T. suis 0/6 0/6 0/6 0/6 3/6 2/6
Dual 0/6 0/6 0/6 1/6 4/6 1/6
Control 0/6 0/6 0/6 1/6 4/6 5/6
Proximal C. jejuni 0/6 0/6 0/6 1/6 2/6 4/6
colon T. suis 0/6 0/6 0/6 1/6 4/6 1/6
Dual 0/6 0/6 0/6 4/6 4/6 2/6
Control 0/6 0/6 0/6 0/6 0/6 6/6
Distal C. jejuni 0/6 0/6 0/6 0/6 0/6 6/6
colon T. suis 0/6 0/6 1/6 0/6 2/6 3/6
Dual 0/6 0/6 0/6 0/6 2/6 6/6

 

 

 

 

 

 

 

Table 1: Summary of Campylobacter detection and identiﬁcation. Campylobacters were
detected and identiﬁed from pigs by fecal culture, PCR for the C. jejuni QRDR on both
fecal and tissue DNA, and RF LP analysis of the thermophilic Campylobacter 23S rRNA

gene using tissue DNA.

102

 

l .2 ..3 4.5 .- 6.. 78 .9210 11._ 12 13 14 15 16

Figure 2: RFLP analysis of the Campylobacter 23S rRNA gene. Panel A: AluI digest of
proximal colon samples from infected pigs . Lane 1: 100bp DNA ladder; Lane 2: C.
jejuni; Lane 3: C. coli; Lanes 4 through 15: DNA from infected pigs. Panel B: Tsp5091
digest of proximal colon samples from infected pigs. Lane 1: 100bp DNA ladder; Lane 2:
C. jejuni; Lane 3: C. coli; Lane 4: C. hyoilei; Lanes 5 through 16: DNA from infected
pigs.

103

.3:on dance? 36 on“ boo Min H 05 mo :80 89a me
o me n E 388% can? @3053 853 deuce—MESS X3 dogma: 739 A— 58 ME. K "O Rio 33.5% U ”m ”—05:00 2‘ 35m
.832 Ea. H coo Banana... v.53 ammo mo 823 2: Eco acronym conga mum: .033 0.828 mo mucuoom 358m mdm an charm

\rr.
\

v

.x . U.
a. ..
..

 

104

.Eouaowiwaﬁ Novo 935 833V £3855: use A388 coo anaconda».

A395 026 £00 E83 .365 xoﬂno miqoﬁmoo woos—2: m=oo boﬁgcﬁ A388 cacao 82a— oﬁ 5T5 US$083

3505 «SEE 05 5 88:5: Doug—ma BEE 05 worsen.“ Am #053 ma 388%: MS». H a 93 A395 mom—no 3&8 05 5 £8
33530 Rim—8? mgonm 2 .288 mi @882 Exxex U a 89a 8308 mam .038: 3:28 Mo 30308 358m man: u v 0.5»;

.“,\, ..

an .,. .Iv.r .... o .. t . ... Id. . ‘I'
. ..u . . .. .~u ... .. Q¢Wﬁu M. . x r ‘ .0
. .. . , .. 4.. .. . .C. J I _ o... .0 I .
I... A... .. . ... .. .. I. . . I. .d. .r..
”so ..i. . , . ... .v a . .. ,(. if. . .. . .. .
.. .. . D .v A. 21 w. gnswae

. .i.t. uh” ..L/4110~‘W$

. ... V .. §I‘ A25 alh .qu

 

105

830 Sn

  

€233? Kev .333va you 80>» Am 35% m anew 89a
.6 353 v 28 2 355 N 339% E 808E 33 8658:. EEO .&E @8003 88m 801:0 833m mﬁm ”m 9.53%

.1»
“L

.. M., gum

 

106

aoﬁéews: xo: v anew ua
35$ w Anouaomiwaﬁ v83 m 955 ”0 35m MN 955 “m Baum : 9.80 u< ESE .mnsohm deacon“: :5 95 £838 MO 385 05 5
25:390 @0qu 3338a .53me U ”gamma 3&8 33:5 Eco—oboa 2: .8288 3:28 82m Fmaonoomeogg no 953..—

...4..... 3.. km".

 

107

 

Distal Colon

5
no
_
O
o
—

GM-CSF

iNOS

MCP-l

 

TNF-a
IL-1 [3
IL-6

IL-8

 

 

IL-18
IL—12p35
lL-lZp40
IFNy
lL—4

IL-5

ﬂDﬂﬂmDﬂﬂﬂl—H—H—lﬂﬂg
LJLJLJLJLJLILJLILJLJLILJLJLI
”MIDDDDDEDDDEDDED

IL-l0

 

 

 

 

 

 

IL-13

 

mjDDDDDDDDDDDDDD'
WTHUDDHHHEDUHDHI

MWDJIJ[JIJIJIJIJIJIJIJIJIJIJIJ[J3

mun[JE]D[]E]D[JE]D[]EJD[]Ef
WTHHHHHHHHHHHDDH

M119! '3
gimp! '3 ‘ :

Figure 7: Cytokine expression in intestinal tissues. Expression of mRNA from a panel
of cytokines was measured by Real Time PCR. The data represents the average fold
change in expression from infected animals when compared to control animals. Letters
indicate statistically signiﬁcant changes: A: p: 0.05; B: p: 0.01

108

Chapter 4: “Campylobacterjejuni stimulates IL-18 secretion from differentiated

intestinal epithelial cells”

Abstract

Campylobacter jejuni is the leading bacterial cause of foodborne illness
worldwide. Clinical signs of campylobacteriosis mimicking human disease have been
reproduced in piglets. Upon infection, the bacteria penetrate the intestinal mucus layer,
and adhere to and invade intestinal epithelial cells, resulting in inﬂammation. We have
previously shown that C. jejuni induces proinﬂammatory cytokines in the distal colon
and lymphoglandular complexes of rectally inoculated piglets one hour after infection.
Additionally, C. jejuni has been shown to stimulate secretion of the proinﬂammatory
cytokines IL-IB, TNFoc, IL-6, and lL-8 from intestinal epithelial cells and monocytes.
Interleukin 18 is a pleiotrophic cytokine that is constitutively expressed in intestinal
tissues and plays a critical role in initiating an immune response to pathogenic bacteria.
Because intestinal epithelial cells play an important role in the pathogenesis of C. jejuni
infection in vivo, we hypothesized that C. jejuni induces IL-18 from cultured swine
intestinal epithelial cells. Differentiated and undifferentiated IPEC-l cells were infected
apically with C. jejuni strain ATCC 33292 or 81-176. Supematants were collected at O, 1,
3, 6, 8, and 24 hours after infection and analyzed for IL—18 protein by ELISA. RNA was
isolated from cells 8 hours after infection to assay for IL-18 mRNA expression by RT-
PCR. Differentiated cells secreted signiﬁcantly higher levels of IL-18 protein as early as

3 hours after infection, compared to uninfected controls. Undifferentiated cells did not

109

secrete signiﬁcant levels of IL-18 at any time point. Expression of IL-18 mRNA was not
signiﬁcantly increased by C. jejuni 8 hours after infection from either differentiated or
undifferentiated cells. These results indicate that differentiated intestinal epithelial cells
are a signiﬁcant source of IL-l8. Epithelial cell derived IL-18 may play an important role

in the pathogenesis of C. jejuni infection in vivo.

110

Introduction

Campylobacterjejuni (C. jejuni) is the most frequent bacterial cause of foodborne
illness worldwide, with the CDC estimating 2.4 million cases of campylobacteriosis each
year in the US. [1]. Clinical signs include a prodrome of fever, abdominal pain, and
malaise followed by the onset of profuse, watery diarrhea with blood, mucus, and
leukocytes [2]. Microscopic examination of colonic tissues from humans and
experimentally infected animals demonstrates that C. jejuni infection stimulates an acute
inﬂammatory response with inﬁltration of the epithelium and lamina propria with
neutrophils and mononuclear cells [3-5]. Immunohistochemical staining of colonic
tissues shows C. jejuni associated with epithelial cells and goblet cells, within deep crypts
of the colon, and within lymphoglandular complexes (LGCs) in entrapped crypts and in
the follicle associated epithelium [3, 4]. The pathogenesis of C. jejuni infection involves
colonization of the intestinal mucous overlying the epithelium, adherence to intestinal
epithelial cells, and invasion of epithelial cells [2, 5]. C. jejuni is chemoattracted to
mucin, a major component of intestinal mucous, and polar ﬂagella combined with its
spiral morphology facilitate efficient movement through the mucous to come into contact
with the epithelium [2, 6, 7]. C. jejuni binds to epithelial cells, intercellular junctions, and
the basement membrane via adhesins, including CadF, le, CBF-l, and ﬂagella [8-13].
Bound bacteria are internalized by microtubule- and microﬁlament-dependent processes
[14-16]. Epithelial cell destruction can occur due to the effects of invasion and toxins

produced by C. jejuni, including enterotoxin, cytotoxins, and cytolethal distending toxin

111

[17-21]. Cellular destruction and loss of epithelial function leads to decreased absorption
which manifests clinically as diarrhea [2, 5].

Elimination of the bacteria and resolution of disease is dependent upon induction
of a humoral immune response and cytokine production. C. jejuni stimulates speciﬁc
serum IgG as well as secretory IgA in feces, urine, saliva, and mammary glands [22]. In a
swine model of infection, anti-C. jejuni IgA has been localized to B cell rich germinal
centers of colonic lymphoglandular complexes [23]. Oral pretreatment of mice with
recombinant IL-5 or IL-6 prior to oral challenge with C. jejuni signiﬁcantly reduces
intestinal colonization. In addition, recombinant IL-6 enhances speciﬁc intestinal and
systemic IgA production [24]. In vitro, C. jejuni stimulates secretion of IL-IB, TNFa,
IL-6, and IL-8 from intestinal epithelial cells and monocytes [25-27] (Parthasarathy and
Mansﬁeld, unpublished data; Cunningham and Mansﬁeld, unpublished data). These
studies indicate that a humoral immune response is necessary for resolution of infection
and that proinﬂammatory cytokines may be important for host resistance to C. jejuni
infection in viva.

IL-18 is a pleiotrophic cytokine capable of initiating and regulating both innate
and adaptive immune responses in viva [28-31]. Mouse peritoneal macrophages and

human peripheral blood mononuclear cells produce TNFa, IL-la, IL-1 [3, IL-6, IL-8, and

MCP-l upon stimulation with IL-18 [32, 33]. Naive T cells proliferate and produce IFNy
when stimulated with IL-12 and IL-18 [29, 34, 35]. NK cell cytotoxicity and Th 1 cell
differentiation are enhanced by stimulation with IL-12 and IL-18 [29, 34, 35]. In the
presence of lL-2, IL-18 stimulates IL-13 production from NK cells and T cells [35]. IL-

18 is produced by a variety of cell types and is widely distributed through tissues in viva

112

[35, 36]. IL-18 is both transcriptionally and post translationally regulated. The IL-18
promoter region is responsive to lipopolysaccharide (LPS) and increased mRNA
expression can be induced in LPS stimulated swine macrophages and splenocytes in vitro
[36, 37]. Immature IL-18 protein requires cleavage by IL-lB converting enzyme (ICE),
also known as Caspase-l, for activation and secretion from cells [38]. In addition, IL-18
binding protein regulates IL-18 activity in viva [39]. IL-l8 mRNA and protein are
constitutively expressed in a variety of cell types, including intestinal cells, suggesting a
role for IL-18 as a ﬁrst line of host defense against pathogens [3 5].

IL-18 is constitutively expressed in intestinal tissues and plays a critical role in
initiating an immune response to pathogenic bacteria [34, 35, 40, 41]. Also, as discussed
in Chapter 2, C. jejuni stimulates signiﬁcant expression of lL-18 and other
proinﬂammatory cytokines in the distal colon and lymphoglandular complexes 1 hour
after rectal inoculation. Because enterocytes are the ﬁrst line of defense against enteric
bacteria, we hypothesized that intestinal epithelial cells are a source of IL-18 during C.
jejuni infections. To test this hypothesis, differentiated and undifferentiated cultured
Intestinal Pig Epithelial Cells (IPEC-l) were infected with C. jejuni for O, l, 3, 6, 8, and
24 hours. IL-18 production was analyzed using two methods. IL-18 secretion in culture
supematants was measured by ELISA. IL-l8 mRNA expression in cells was measured by
RT-PCR. Differentiated cells secreted signiﬁcant levels of IL-18 protein as early as 3
hours after infection. Undifferentiated cells did not secrete signiﬁcant levels of IL-18 at
any time point. Expression of IL-1 8 mRNA was not signiﬁcantly increased by C. jejuni at

any time from either differentiated or undifferentiated cells. These results indicate that

113

differentiated intestinal epithelial cells are a signiﬁcant source of IL-l8. Epithelial cell-

derived IL-18 may be important for host resistance to C. jejuni infection in viva.

I“

‘ﬂiK.-.’- .l I.

 

114

Materials and Methods

Cells

IPEC-l is a non-immortalized undifferentiated cell line derived from the small
intestine of neonatal pigs. These cells will differentiate, developing tight junctions and
distinct apical and basolateral surfaces, when grown on porous supports for 10 to 14 days
[42]. The transepithelial electrical resistance (TER) can be measured across the
monolayer to indicate whether the cells have achieved a differentiated state.
Differentiated IPEC-l cells represent the mature enterocytes located on the villus tips in
the intestines.

When IPEC-l cells are seeded at the appropriate density onto ﬂat bottom
polystyrene plates and grown for only 2-3 days, the cells form conﬂuent monolayers that
remain undifferentiated. Undifferentiated cells represent the rapidly dividing cells found
in the base of the crypts in the intestine. We used IPEC-l cells in both differentiated and
undifferentiated states to determine if IL-18 production was stimulated upon C. jejuni
infection.

All culture media and growth supplements were obtained from Invitrogen unless
otherwise stated (Invitrogen, Frederick, MD). Growth ﬂasks and transwells were
obtained from Corning (Costar, Corning, NY). Cells were routinely cultured at 37°C, 5%
C02 in DMEM/F12 supplemented with 5% Fetal Bovine Serum (FBS) and 1% Insulin-
Transferrin-Selenium (ITS) in T75 ﬂasks. To seed cells for infection studies, the cells
were gently washed with versene and released from the ﬂask with 0.05% Trypsin-EDTA.

Detached cells were collected and pelleted by centrifugation at 1200rpm for 5 minutes.

115

 

 

The supernatant was removed and the cells were resuspended in RPMI supplemented
with 5%FBS. A subsarnple of the cells was mixed with a 1:1 volume of trypan blue
(Sigma, St. Louis, MO), then loaded onto a hemocytometer (Hausser Scientiﬁc,
Horsharn, PA) and viable cells were counted on a Nikon inverted microscope (Mager
Scientiﬁc, Dexter, MI).

Cells to be tested in a differentiated state were seeded at a density of 3x105 cells
onto 6.5mm diameter, 3.0um pore size transwell inserts which had been previously
coated with 20ug/mL ﬁbronectin (Sigma, St. Louis, MO). Cells were incubated for 12-
14 days to allow differentiation. Prior to infection TER was measured from each well
using an electrode to conﬁrm conﬂuency and tight junction formation (EVOMX, World
Precision Instruments, Sarasota, FL). TER values greater than 6OOQ/cm2 indicated that
the cells were differentiated.

Cells to be infected in the undifferentiated state were seeded at a density of 3x105
cells per well onto 24 well plates that had previously been coated with 20ug/mL of
ﬁbronectin (Sigma, St. Louis, MO). The cells were incubated for 2-3 days to allow

formation of a conﬂuent monolayer.

Scanning Electron Microscopy

Differentiated IPEC-l cells grown on transwells for 12 days were examined by
scanning electron microscopy to verify their state of differentiation. The cells were
seeded onto transwells and incubated for 12 days as previously described. The medium
was removed, the membrane was excised, and the cells were ﬁxed in 4% glutaraldehyde

in 0.1M sodium phosphate buffer. The cells were further processed and examined at the

116

 

Center for Microscopy at Michigan State University (East Lansing, MI). The cells were
ﬁxed in 1% osmium tetroxide, dehydrated through and ethanol gradient, then dried in a
critical point dryer (Balzers, Lichenstein). Dehydrated membranes were mounted on
aluminum stubs and sputter coated with 7nm of gold (Emscope SCSOO sputter coater,

UK). The cells were visualized on a scanning electron microscope (JEOL 64OOV, Japan).

Bacteria

Campylobacterjejuni strains ATCC 33292 and 81-176 were used for infections of
[PEG] cells. Both strains were originally isolated from humans with enteritis. The C.
jejuni strain ATCC 33292 used in these infections had been rederived by passage through
a pig in our laboratory and stored as second passage stocks. C. jejuni strain 81-176 was
kindly donated by Carol Pickett. Bacterial cultures were streaked for isolation onto
Bolton Agar plates and incubated for 48 to 72 hours at 37°C, 5% C02. Isolated colonies
were subcultured for lawn growth and grown for approximately 20 hours. Lawns were
harvested with sterile swabs and resuspended in RPMI without phenol red or FBS. The
OD560 was measured on a Beckman DU 530 UVN is spectrophotometer (Beckman
Coulter, Inc, Fullerton, CA) and subsequently adjusted to an OD560 of 0.1 in RPMI.
Previous studies in our lab have shown that this measurement corresponds to
approximately 5x108 cfu/mL. A subsarnple of each inoculum was examined

microscopically to assess morphology and darting motility of the bacteria.

117

 

Experimental Design

To determine the effect of C. jejuni on IL-18 secretion, IPEC-l cells were
exposed to each strain of bacteria apically at a multiplicity of infection (MOI) of
approximately 40:1. Previous studies in our laboratory have shown this to be the optimal
concentration of C. jejuni for stimulation of IL-6 and IL-10 secretion from IPEC-l cells.
RPMI without phenol red or F BS was used as the negative control and Concanavalin A
(Con A) was used at a concentration of lOOug/mL as the positive control. Supematants
were removed at O, l, 3, 6, 8, or 24 hours after infection then centrifuged for 10 minutes
at 6000rpm and 4°C to pellet bacteria. The supematants were transferred to
polypropylene tubes (DOT Scientiﬁc, Burton, MI) and stored at -80°C until ELISA was
performed to measure the concentration of IL-18 protein. For differentiated IPEC-l cells,
supematants from the upper and lower chamber were stored and analyzed separately to
determine if IL-18 was secreted from the apical or basolateral surface. Expression of IL-
18 mRNA was measured from IPEC-l cells in both differentiated and undifferentiated

states by RT-PCR of total RNA extracted from cells infected for 8 hours.

ELISA

Secreted IL-18 protein was measured from culture supematants by an indirect
ELISA developed in our laboratory. Recombinant porcine IL-18 and anti-swine IL-18
primary antibody were obtained from R and D systems (Minneapolis, MN). Biotin
conjugated anti-goat IgG secondary antibody, Extravidin®, bovine serum albumin
(BSA), and the ready to use peroxide containing tetramethyl benzidine (TMB) substrate

were obtained from Sigma (Sigma, St. Louis, MO). ELISA plates were washed 3 times

118

 

with Tris-buffered saline with 0.5% Tween-20 (TTBS) between incubations and all
incubations were for 1 hour at room temperature unless otherwise indicated. Prior to
measuring IL-18 from experimentally infected IPEC-l cells, the optimal concentrations
for the primary antibody, secondary antibody, and Extravidin were determined
experimentally (data not shown). For preparation of the standards, recombinant porcine
IL-18 was diluted in RPMI 1640 without phenol red to lOOng/mL, SOng/mL, 25ng/mL,
12.5ng/mL, 6.25ng/mL, 3.125ng/mL, and 1.56ng/mL ﬁnal concentrations. Supematants
and standards were coated in duplicate onto Nunc Immuno Maxisorp 96 well plates

(V WR, West Chester, PA) overnight at 4°C. Plates were blocked for 90 minutes with 3%

BSA at 37°C, then bound IL-18 was detected by 2 hour incubation with anti-swine IL-1 8
antibody diluted to 750ng/mL in 1%BSA in TTBS. Primary antibody was detected by
biotin conjugated secondary antibody diluted 1:10,000 in 1%BSA in TTBS. The biotin
conjugate was reacted with Extravidin® diluted to lug/mL in 1%BSA in TTBS, followed
by color development for approximately 15 minutes with the peroxide containing TMB
substrate. Color development was stopped by addition of 6N H2804 after approximately
15 minutes. Absorbance was measured at 450nm on an EL8OO Universal Microplate
Reader and the concentration of IL-18 protein in each sample was calculated by

KCjunior® software (Bio-Tek instruments, Winooski, Vermont).

Cytokine expression
Total RNA was isolated from IPEC-l cells using the Trizol® method. Brieﬂy,
cells were lysed in Trizol® reagent (Invitrogen, Frederick, MD), RNA was extracted with

chloroform, precipitated with isopropanol, and washed with ethanol. Pellets were air

119

 

17.3.,7

 

dried, then resuspended in DEPC-treated H20 (Invitrogen, Frederick, MD). Concentration
of RNA was determined spectrophotometrically (Beckman). Samples were checked for
degradation by electrophoresis on a 1.8% agarose gel in 1X TAE buffer with 0.2ug/mL
ethidium bromide. Gels were visualized by UV illumination on an Alpha Imager 2000
Documentation and Analysis system (Alpha Innotech Corporation, San Leandro, CA).
RNA samples were stored at -80°C. For each sample, 5ug of total RNA was reverse
transcribed with Oligo dT primers using a Stratagene (Stratagene, La Jolla, CA) First
Strand RT-PCR kit. First strand cDNA was used as the template in further PCR reactions

Swine IL-18 and Bz-microglobulin expression was measured from IPEC-l cDNA
samples using previously described primers [35, 43]. The primer sequences for IL-18
were forward 5' TAT GCC TGA TI‘C TGA CTG TT 3’ and reverse 5' ATG AAG ACT
CAA ACT GTA TCT 3' [3 5]. The primer sequences for Bz-microglobulin were forward
5' CTG CTC TCA CTG TCT GG 3', and reverse 5' ATC GAG AGT CAC GTG CT 3'
[43]. For each target, a 25 “L PCR reaction consisted of IX Perkin Elmer PCR buffer
(Tris-HCL, pH 8.3, 50 mM KCI), 1.0mM MgClz, 0.05mM each dNTP, 37.5ng of each
primer, 0.625U Perkin Elmer AmpliTaq Gold, and luL of ﬁrst strand cDNA in DEPC
treated water. Each sample was ampliﬁed in triplicate.

Prior to measuring expression in our sample set, the optimal number of
ampliﬁcation cycles for each target was determined for measurement of expression
during the exponential phase of ampliﬁcation. We measured expression from positive and
negative controls every 2 cycles from 20 to 40 cycles and plotted the values on a graph

(Figures 5 and 6). All samples were ampliﬁed with a 10 minute hot start at 94°C followed

120

 

i. 71“.“ haul

 

by 39 cycles of 94°C for 1 minute, 55°C for 1 minute, 72°C for 1 minute, with a ﬁnal
extension time of 10 minutes at 72°C.

The PCR product was electrophoresed on 1.8% agarose gels in 1X TAE with
0.2ug/mL ethidium bromide. The gels were visualized by UV illumination on an Alpha
Imager 2000 Documentation and Analysis system, and density of individual bands was
measured using the Line Densitometry ﬁmction. Bz-microglobulin was used as the
constitutively expressed gene for normalization of cytokine expression. Ratios of speciﬁc
cytokine to housekeeping gene were calculated by dividing the density values obtained

from the IL-18 target by those obtained from the Bz-microglobulin target [3 5].

Statistical Analysis

The data were analyzed in Excel (Microsoft, Redmond, WA). An F test for
sample variance was performed to compare means of infected cells to uninfected cells.
The appropriate T test was then used to determine the effect of treatment on IL-18
production at each time point. P values of 0.05 or less were considered statistically

signiﬁcant.

121

Results

Scanning Electron Microscopy (SEM)

A representative SEM image of IPEC-l cells grown on transwells for 12
days is shown in ﬁgure 1. TER values measured from 12 day old IPEC-l cells ranged
from 600 to 800 Q/cmz. The cell monolayers were conﬂuent and cells had developed

microvilli on their apical surfaces, indicating that they had achieved a differentiated state.

IL-l8 secretion

IPEC-l cells in a differentiated state secreted signiﬁcant amounts of IL-18 into
the culture supernatant by 3 hours after C. jejuni infection (Figure 2). In the upper
chamber, secretion was virtually undetectable immediately after infection (0 hr) and 1
hour after infection. Secretion peaked at approximately 12 ng/mL 3 hours after infection,
and remained elevated at both 6 and 8 hours after infection. By 24 hours, IL-18 was still
detectable, but the levels had decreased signiﬁcantly. In the lower chamber, lL-l8
secretion was very low and was not signiﬁcantly induced by C. jejuni infection at any
time point (Figure 3). IL-18 secretion in response to C. jejuni infection was not
signiﬁcantly different between strains.

In contrast, IPEC-l cells in an undifferentiated state did not secrete signiﬁcant

amounts of IL-18 in response to C. jejuni at any time (Figure 4).

122

'Aﬁ._ml l'

 

IL-18 mRNA expression

IL-18 expression was measured by RT—PCR from IPEC-l cells in both the
differentiated and undifferentiated state 8 hours after infection with C. jejuni. Prior to
measuring expression from experimental samples, the optimal cycle number to use was
determined by measuring the density of the IL-18 and BZ-microglobulin targets ampliﬁed
from stimulated (lOOng each of Calcium ionophore and phorbol myristate acetate) and
unstimulated cells (DMEM or RPMI medium only) over several cycles (Figure 5). The
optimal cycle number for IL-18 measurement from experimental samples was selected
from the linear portion of the ampliﬁcation curve for each target (Figure 6). The optimal
cycle numbers selected for measurement of target expression from experimental samples

were 31 for IL-18 and 27 for BZ-microglobulin.

IL-18 and BZ-microglobulin targets were ampliﬁed from IPEC-l cells in both
differentiation states with the exception of IPEC-l cells in the differentiated state treated
with Con A (data not shown). IL-18 expression was not signiﬁcantly impacted by C.

jejuni infection at any time point in IPEC-l cells in either differentiation state (Figure 7).

123

Discussion

We hypothesized that intestinal epithelial cells infected with C. jejuni are a source
of IL-18. Furthermore, early release of IL-18 from epithelial cells assaulted by C. jejuni
may serve to initiate a protective inﬂammatory response in viva. Our experiments show
that C. jejuni stimulates lL-18 secretion from IPEC-l cells that have achieved a
differentiated state. Secretion peaked at 3 hours and remained elevated at 6 and 8 hours
after infection before decreasing signiﬁcantly by 24 hours after infection. In vitro, C.
jejuni stimulates secretion of a panel of proinﬂammatory cytokines from intestinal
epithelial cells and monocytes, including IL-IB, TNFa, IL-6, and IL-8 [25-27]
(Parthasarathy and Mansﬁeld, unpublished data; Cunningham and Mansﬁeld,
unpublished data). Epithelial cell-derived cytokines play an important role in controlling
the spread of pathogenic bacteria in viva [44]. IL-18 facilitates neutrophil accumulation
in tissues of mice during lethal endotoxemia, and it has also been shown to activate
neutrophils, inducing granule release and enhancing the respiratory burst [45, 46].
Neutrophil inﬁltration of the lamina propria and submucosa is a consistent ﬁnding in
animal models of C. jejuni infection [3, 4, 47]. The effects of IL-18 on neutrophil
functions likely serve to kill invading C. jejuni and contain the infection. IL-18 also
stimulates production of IL-8, TNFa, and IL-IB, which would sustain the inﬂammatory
response and activate other mechanisms for eliminating bacterial pathogens [32, 33, 45,
46].

IL-18 secretion from differentiated IPEC-l cells was apparently not

transcriptionally regulated. IL-18 mRNA was constitutively expressed in these cells, but

124

expression was not signiﬁcantly increased by infection with C. jejuni. The IL-18
promoter sequence contains LPS response elements and IL-18 expression is induced in
LPS stimulated swine alveolar macrophages, adherent splenocytes, and non adherent
splenocytes [36, 37]. Differentiated IPEC-l cells stimulated with 200ug/mL of LPS ﬁom
E. coli 026:B6 secrete signiﬁcant amounts of IL-6 (Parthasarathy et al, unpublished
results), but not IL-18 (data not shown). This concurs with data from Okazawa and
colleagues who found that LPS stimulation of the human colonic cell line HT-29 did not
result in increased intracellular accumulation of IL-18 or increased secretion [48]. These
data indicate that LPS-induced transcriptional regulation does not increase production of
IL-18 from intestinal epithelial cells. Constitutive expression of IL-18 protein in the
intestine suggests that rapid production of IL-18 is important for early responses to
pathogens [35, 36]. Therefore, it is likely that post translational modiﬁcation of IL-l8, via
cleavage with Caspase-l, rather than increased expression of mRNA, is the main
mechanism for IL-18 secretion in the intestine. The data from differentiated IPEC-l cells
support that hypothesis.

In contrast to differentiated IPEC-l cells, C. jejuni infection did not stimulate
increased production of IL-18 protein or mRNA from undifferentiated IPEC-l cells at
any time assayed over 24 hours. Similar to differentiated cells, undifferentiated cells
express IL-18 mRNA constitutively (data not shown). Undifferentiated cells are capable
of secreting 7ng/mL of IL-18 after 8 hours of stimulation with Con A (Figure 4). It has
been demonstrated that undifferentiated IPEC-l cells secrete IL-10 in response to Con A,
but not in response to C. jejuni, while differentiated IPEC-l cells do secrete lL-10 in

response to C. jejuni infection (Parthasarathy et al, unpublished results). IL-1 [3 stimulated

125

differentiated HT-29 human intestinal epithelial cells have impaired production of IL-8
when compared to undifferentiated cells. This effect is due to decreased c-Jun N-terminal
kinase (JNK) and MB kinase (IKK) activity with subsequently reduced NF -KB DNA
binding after stimulation of the IL-1 [3 receptor [49]. Impaired secretion of IL-18 (and IL-
10) from undifferentiated IPEC-l cells may be due to altered intracellular signaling
pathways, but further study is necessary to elucidate this.

IL-l8 mRNA and immature protein are constitutively expressed in swine intestinal
epithelial cells and gut associated lymphoid tissues throughout the intestine [35, 36].
Salmonella, Shigella, Yersenia, and Helicobacter infections induce increased expression
of IL-18 mRNA and maturation of IL—18 protein to the active form, which are critical for
resolution of infection [34, 36, 41, 50-52]. We have also shown, as discussed in Chapter
2, that C. jejuni induces IL-18 and other proinﬂammatory cytokines in the distal colon
and lymphoglandular complexes of rectally inoculated pigs. We propose that in our swine
model of C. jejuni infection, IL-l8 released from C. jejuni infected epithelial cells
stimulates production of IL-IB, TNFOL, IL-6, IL-8, and MCP-l from epithelial cells and
leukocytes locally, inducing an inﬂammatory response. Also, neutrophil inﬁltration of the
epithelium and lamina propria with subsequent activation of their bacterial killing

mechanisms by IL-18 may be critical for reducing pathogen numbers.

126

 

10.

ll.

l2.

l3.

14.

15.

16.

References

Allos, B.M., Campylobacterjejuni Infections: update on emerging issues and trends. Clin Infect
Dis, 2001. 32(8): p. l201-6.

Ketley, J.M., Pathogenesis of enteric infection by Campylobacter. Microbiology, 1997. 143 (Pt I):
p. 5-21.

Babakhani, F.K., G.A. Bradley, and LA. Joens, Newborn piglet model for campylobacteriosis.
Infect Immun, 1993. 61(8): p. 3466-75.

Mansﬁeld, L.S., et al., Enhancement of disease and pathology by synergy of Trichuris suis and
Campylobacterjejuni in the colon of immunologically naive swine. Am J Trop Med Hyg, 2003.
68(3): p. 70-80.

Wassenaar, T.M. and M.J. Blaser, Pathophysialogy of C ampylobacter jejuni infections of humans.
Microbes Infect, 1999. 1(12): p. 1023-33.

Hugdahl, M.B., J.T. Beery, and M.P. Doyle, C hemotactic behavior of Campylobacter jejuni.
Infect Immun, 1988. 56(6): p. 1560-6.

Szymanski, C.M., et al., Campylobacterjejuni motility and invasion of Caca-2 cells. Infect
Immun, 1995. 63(11): p. 4295-300.

J in, S., et al., leA, a novel surface-exposed lipoprotein specific to Campylobacterjejuni, mediates
adherence to host epithelial cells. Mol Microbiol, 2001. 39(5): p. 1225-36.

J in, S., et al., leA of Campylobacter jejuni interacts with surface-exposed heat shock protein
90alpha and triggers signalling pathways leading to the activation of NF -kappaB and p38 MAP
kinase in epithelial cells. Cell Microbiol, 2003. 5(3): p. 165-74.

Konkel, M.B., et al., Identification and molecular cloning of a gene encoding a fibronectin-
binding protein (CadF) from Campylobacterjejuni. Mol Microbiol, 1997. 24(5): p. 953-63.

McSweegan, E. and RJ. Walker, Identification and characterization of two Campylobacterjejuni
adhesinsfor cellular and mucous substrates. Infect Immun, 1986. 53(1): p. 141-8.

Monteville, M.R., J .E. Yoon, and ME. Konkel, Maximal adherence and invasion of IN T 407 cells
by Campylobacterjejuni requires the CadF outer-membrane protein and microﬁlament
reorganization. Microbiology, 2003. 149(Pt l): p. 153-65.

Pei, Z., et al., Mutation in the peblA locus of Campylobacter jejuni reduces interactions with
epithelial cells and intestinal colonization of mice. Infect Immun, 1998. 66(3): p. 93 8—43.

Biswas, D., K. Itoh, and C. Sasakawa, Uptake pathways of clinical and healthy animal isolates of
Campylobacterjejuni into INT-407 cells. FEMS Immunol Med Microbiol, 2000. 29(3): p. 203-1 1.

Hu, L. and DJ. Kopecko, Campylobacterjejuni 81-1 76 associates with microtubules and afvnein
during invasion of human intestinal cells. Infect Immun, 1999. 67(8): p. 4171-82.

Oelschlaeger, T.A., P. Guerry, and DJ. Kopecko, Unusual microtubule-dependent endocytosis

mechanisms triggered by C ampylobacter jejuni and C itrabacter freundii. Proc Natl Acad Sci U S
A, 1993. 90(14): p. 6884-8.

127

17.

I8.

20.

21.

22.

23.

24.

25.

26.

27.

28.
29.

30.

31.

32.

33.

34.

Aragon, V., K. Chao, and LA. Dreyfus, Eﬂect of cytolethal distending toxin on F -actin assembly
and cell division in Chinese hamster ovary cells. Infect Immun, 1997. 65(9): p. 3774-80.

Johnson, W.M. and H. Lior, Cytotoxic and cytotonic factors produced by Campylobacterjejuni,
Campylobacter coli, and Campylobacter laridis. J Clin Microbiol, 1986. 24(2): p. 275-81.

Klipstein, F.A. and RF. Engert, Properties of crude Campylobacterjejuni heat-labile enterotoxin.
Infect Immun, 1984. 45(2): p. 314-9.

Lara-Tejero, M., A bacterial toxin that causes DNA damage to modulate cellular responses.
ScientiﬁcWorldJoumal, 2001. l: p. 190-1.

Lee, R.B., et al., Interactions of Campylobacter jejuni cytolethal distending toxin subunits C th
and CdtC with HeLa cells. Infect Immun, 2003. 71(9): p. 4883-90.

Cawthraw, S.A., et al., Long-term antibody responses following human infection with
Campylobacterjejuni. Clin Exp Immunol, 2002. 130(1): p. 101-6.

Mansﬁeld, LS. and D.T. Gauthier, Lymphoglandular complexes are important colonic sites for
immunoglobulin A induction against Campylobacterjejuni in a swine disease model. Comp Med,
2004. 54(5): p. 514-23.

Baqar, 8., ND. Pacheco, and F .M. Rollwagen, Modulation of mucosal immunity against
Campylobacterjejuni by orally administered cytokines. Antimicrob Agents Chemother, 1993.
37(12): p. 2688-92.

Hickey, T.E., et al., Campylobacterjejuni-stimulated secretion of interleukin-8 by IN T407 cells.
Infect Immun, 1999. 67(1): p. 88-93.

Hickey, T.E., et al., Campylobacterjejuni cytolethal distending toxin mediates release of
interleukin-8 from intestinal epithelial cells. Infect Immun, 2000. 68(12): p. 6535-41.

Jones, M.A., et al., Induction of proinﬂammatory responses in the human monocytic cell line
THP-I by Campylobacterjejuni. Infect Immun, 2003. 71(5): p. 2626-33.

Akira, S., The role of IL-1 8 in innate immunity. Curr Opin Immunol, 2000. 12(1): p. 59-63.
Dinarello, C.A., et al., Overview of interleukin-I 8: more than an interferon-gamma inducing
factor. J Leukoc Biol, 1998. 63(6): p. 658-64.

Gracie, J.A., S.E. Robertson, and LB. Mclnnes, Interleukin-I8. J Leukoc Biol, 2003. 73(2): p.
213-24.

Nakanishi, K., et al., Interleukin-I8 is a unique cytokine that stimulates both Th1 and Th2
responses depending on its cytokine milieu. Cytokine Growth Factor Rev, 2001. 12(1): p. 53-72.

Netea, M.G., et al., Interleukin-I8 induces production of proinﬂammatory cytokines in mice: no
intermediate role for the cytokines of the tumor necrosis factor family and interleukin-I beta. Eur J
Immunol, 2000. 30(10): p. 3057-60.

Puren, A.J., et al., Interleukin-I8 (IFNgamma-inducing factor) induces IL-8 and IL-I beta via
INF alpha production ﬁ'om non-CDI4+ human blood mononuclear cells. J Clin Invest, 1998.
101(3): p. 711-2].

Biet, F., C. Locht, and L. Kremer, Immunoregulatoryfunctions of interleukin I8 and its role in
defense against bacterial pathogens. .1 Mol Med, 2002. 80(3): p. 147-62.

128

35.

36.

37.

38.

39.

40.

41.

42.

43.

44.

45.

46.

47.

48.

49.

50.

51.

Fournout, S., et al., Cloning, chromosomal location, and tissue expression of the gene for pig
interleukin-I8. Immunogenetics, 2000. 51(4-5): p. 358-65.

Foss, D.L., M.J. Zilliox, and M.P. Murtaugh, Bacterially induced activation of interleukin-1 8 in
porcine intestinal mucosa. Vet Immunol Immunopathol, 2001. 78(3-4): p. 263-77.

Kim, Y.M., et al., Roles of IFN consensus sequence binding protein and PU. I in regulating IL-I8
gene expression. J Immunol, 1999. 163(4): p. 2000-7.

Dinarello, C .A., Interleukin-I beta, interleukin-I8, and the interleukin-I beta converting enzyme.
Ann N Y Acad Sci, 1998. 856: p. 1-1 I.

Dinarello, C.A., Novel targets for interleukin I8 binding protein. Ann Rheum Dis, 2001. 60 Suppl
3: p. iiil8-24.

Fox, J.G., et al., Gastroenteritis in NF -kappaB-deﬁcient mice is produced with wild-type
Camplyobacterjejuni but not with C. jejuni lacking cytolethal distending toxin despite persistent
colonization with both strains. Infect Immun, 2004. 72(2): p. 1116-25.

Sansonetti, P.J., et al., Caspase-l activation of IL-I beta and IL-18 are essential for Shigella
ﬂexneri-induced inflammation. Immunity, 2000. 12(5): p. 581-90.

Abner, S.R., et al., Response of intestinal epithelial cells to Trichuris suis excretory-secretory
products and the inﬂuence on Campylobacterjejuni invasion under in vitro conditions. J Parasitol,
2002. 88(4): p. 738-45.

Reddy, N.R., B.N. Wilkie, and B.A. Mallard, Construction of an internal control to quantitate
multiple porcine cytokine mRNAs by RT-PCR. Biotechniques, 1996. 21(5): p. 868-70, 872, 875.

Sansonetti, P.J., et al., Interleukin-8 controls bacterial transepithelial translocation at the cost of
epithelial destruction in experimental shigellosis. Infect Immun, 1999. 67(3): p. 1471-80.

Leung, B.P., et al., A role for IL-I8 in neutrophil activation. J Immunol, 2001. 167(5): p. 2879-86.

Netea, M.G., et al., Neutralization of IL-1 8 reduces neutrophil tissue accumulation and protects
mice against lethal Escherichia coli and Salmonella typhimurium endotoxemia. J Immunol, 2000.
164(5): p. 2644-9.

F ox, J .G., et al., Campylobacterjejuni infection in the ferret: an animal model of human
campylobacteriosis. Am J Vet Res, 1987. 48(1): p. 85-90.

Okazawa, A., et al., Human intestinal epithelial cell-derived interleukin (IL)-I8, along with IL-2,
IL-7 and IL-1 5, is a potent synergistic factor for the proliferation of intraepithelial lymphocytes.
Clin Exp Immunol, 2004. 136(2): p. 269-76.

Bocker, U., et al., Cellular diﬂerentiation causes a selective down-regulation of interleukin (IL)-
! beta-mediated NF -kappaB activation and IL-8 gene expression in intestinal epithelial cells. J
Biol Chem, 2000. 275(16): p. 12207-13.

Elhofy, A. and K.L. Bost, Limited interleukin-I8 response in Salmonella-infected murine
macrophages and in Salmonella-infected mice. Infect Immun, 1999. 67(10): p. 5021-6.

Fera, M.T., et al., Correlation between Helicobacter pylori infection and IL-I8 mRNA expression
in human gastric biopsy specimens. Ann N Y Acad Sci, 2002. 963: p. 326-8.

129

52. Tomita, T., et al., Expression of Interleukin-I 8, a Th] cytokine, in human gastric mucosa is
increased in Helicobacter pylori infection. J Infect Dis, 2001. 183(4): p. 620-7.

130

Figure 1: Scanning electron micrograph of polarized IPEC-l cells (12 days) on transwell
membranes. Top panel is overview of cells; bar represents 20 pm. The apical ﬁnger-like
microvilli are seen in the bottom panel; bar represents 2 pm. Micrographs are
representative of three samples each.

 

I31

 

lL-18 secretion from lPEC-1 cells- upper chamber

 

 

 

 

 

 

 

 

nglmL
a

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

IOI'Ir
I1hr
m3hr
56hr
EB”
.2417

 

 

 

81-176 3292

RPMI

 

 

 

 

Figure 2: IL-18 secretion from differentiated IPEC-l cells. Apical secretion of IL-18 in
response to apical infection with C. jejuni was measured by ELISA from the upper
chamber of differentiated cells. Secretion from C. jejuni infected cells was signiﬁcantly
greater than medium only controls at 3, 6, 8, and 24 hours aﬁer infection. (p50.05).

132

 

 

 

 

 

IL-18 secretion from IPEC-1 cells- lower chamber

30

 

25 - ~—

 

 

20

 

 

15 --————- —-——*ﬁ-~

 

nglmL

 

10 ~—- , ,,

 

 

 

 

 

81-176 3292

1.--. _ , J

 

 

II.
\

 

 

 

Figure 3: IL-18 secretion from differentiated IPEC-l cells. Basolateral secretion of IL-18
in response to apical infection with C. jejuni was measured by ELISA from the lower
chamber of differentiated cells. Secretion from C. jejuni infected cells was not
signiﬁcantly greater than medium only controls at any time after infection. (pS0.05).

133

30

25

ngImL

 

1

 

 

lL-18 secretion from lPEC-1 cells:
undifferentiated cells

 

 

2O «—

15~

 

1o

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

W IIIIII

 

S\\\\

   

 

ConA

  

81 -1 76 33292 RPMI

um]

 

Figure 4: IL-1 8 secretion from undifferentiated IPEC-l cells. IL-18 secretion in response

to C. jejuni infection of undifferentiated IPEC-l cells was measured by ELISA.

Undifferentiated IPEC-l cells secreted an average of 7ng/mL of IL-1 8 in response to

Concanavalin A, however IL-1 8 secretion from C. jejuni infected cells was not

signiﬁcantly greater than medium only controls at any time after infection. (p50.05).

134

 

1r—

 

Figure 5: Optimization of ampliﬁcation parameters. We titrated the cycle number needed
to measure IL-18 and B2-microglobulin expression within the linear portion of the
ampliﬁcation curve. Samples were ampliﬁed in triplicate, electrophoresed on an agarose
gel, and density of the bands was measured. Panel A: Cycle titration for IL-18
expression. Lane 1: 100bp DNA ladder; Lanes 2-4: 28 cycles; Lanes 5-7: 30 cycles;
Lanes 8-10: 32 cycles; Lanes 11-13: 34 cycles; Lanes 14-16: 36 cycles; Lanes 17-19: 38
cycles. Panel B : Cycle titration for BZ-microglobulin expression. Lane 1: 100bp DNA
ladder; Lanes 2-4: 24 cycles; Lanes 5-7: 26 cycles; Lanes 8-10: 28 cycles; Lanes 1 l-13:
30 cycles; Lanes 14-16: 32 cycles; Lanes 17-18: 34 cycles.

135

 

Cycle titration for IL-18 expression

1400

1200

1000

 

 

 

 

g 000
i
a 600
400
200
0
Cycle titration for 32- microglobulin
1600.0
1400.0
1200.0
1000.0 4 W
I 800 o 7 +CVPMA 1
‘ ‘ ,+, ,ﬂl

6000
4000

2000

 

0.0

24 26 20 30 32 34
Cycle Nunber

 

 

Figure 6: Determination of optimal number of cycles for ampliﬁcation. The average
density of the bands from each cycle was plotted on a graph. The optimal cycle number
for measurement of target expression from experimentally infected samples was chosen
from the linear portion of the ampliﬁcation curve. IL-18 expression was measured at 31
cycles while BZ-microglobulin expression was measured at 27 cycles.

136

 

lL-18 expression from differentiated lPEC-1

 

0.4 -* ~‘

lL-18zb2 ratio

0.2 - 7-7

 

 

0.8+ 7 ,ﬁ 7- w L, -7 ,, 7- ,7—

0.6 -- 7 - —— 7 if ,, .

 

Con A 81 -1 76 33292 RPM I

 

 

 

lL-18 expression from undifferentiated cells

 

 

 

 

Ratio lL-1B:82-M

 

 

 

CVHM B1-176bep 33292

 

Figure 7: IL-18 expression from IPEC-l cells. IL-18 expression was measured 8 hours
after infection from both differentiated and undifferentiated IPEC-l cells. The data from
each sample was normalized to BZ-microglobulin expression and is presented as the ratio
of IL-18 expression over to BZ-microglobulin expression. Neither IL-18 nor to [32-
microglobulin were measured from Con A stimulated differentiated cells. IL-1 8
expression was not signiﬁcantly changed in C. jejuni infected cells when compared to

medium only controls for either cell type.

137

 

 

 

g

Chapter 5: Summary and Conclusions

Pigs are an excellent animal model for studying C. jejuni infection in humans for
several reasons. Physiologically, the swine digestive tract is similar to humans and
outbred populations that are heterogeneous, similar to the human population, are readily
available at low cost from production facilities. Colostrum-deprived piglets become
colonized and develop mucohemorrhagic diarrhea and intestinal pathology that mimics
human disease after oral inoculation with C. jejuni [1, 2]. In addition, an extensive array
of swine speciﬁc reagents is commercially available for the analysis of immune
mediators. All together, these facts make pigs quite suitable for use in research on the
pathogenesis of C. jejuni infection. We used a dual infection model of concurrent C.
jejuni and T. suis infection in pigs to study the immune responses associated with
mucohemorrhagic diarrhea.

Campylobacter jejuni is a commensal in the colons of healthy pigs, but upon
infection with the whipworm Trichuris suis, these two agents synergize to produce
mucohemorrhagic diarrhea and severe pathology [2-4]. The long term goals of the
laboratory are to elucidate the mechanisms by which dual infection with these two agents
causes severe disease and pathology. One hypothesis is that T. suis infection facilitates C.
jejuni invasion by modulation of the host immune response. The central hypothesis for
this dissertation project is that a proinﬂammatory response plays a role in swine natural
resistance to C. jejuni. A related hypothesis is T. suis infection alters the protective
proinﬂammatory response resulting in C. jejuni invasion in the colon and severe
pathology. These hypotheses were tested, generally, by assessing local cytokine

expression in the intestines of C. jejuni and T. suis infected pigs, and more speciﬁcally by

138

assessing IL-18 secretion from C. jejuni infected IPEC-l cells, which serve as a model
system for the swine intestinal epithelium.

This dissertation project was designed to address three main objectives. The ﬁrst
objective was to determine the nature of the local immune response to C. jejuni in the
distal colon and LGC of pigs. To address this issue, weaned piglets were inoculated
rectally with C. jejuni 81-176 that produced Gfp. The pigs were killed 1 hour or 24 hours
after inoculation and samples from the distal colon and LGCs were taken for microscopic
examination for lesions, for immunohistochemical staining with anti-C. jejuni antibody to
localize the bacteria in the colon, and for measurement of cytokine mRNA expression by
real time PCR. The second objective was to determine the nature of the local immune
response in different segments of the intestine when pigs were infected with C. jejuni, T.
suis, or both. To address this issue, weaned piglets were inoculated by oral gavage with
embryonated T. suis eggs. Twenty one days after infection with T. suis, pigs were
inoculated by oral gavage with C. jejuni strain ATCC 33292. Two days after inoculation
with C. jejuni, all pigs were killed and samples from the jejunum, proximal colon, and
distal colon were taken for microscopic examination for lesions, for
immunohistochemical staining with anti-C. jejuni antibody to localize the bacteria, and
for measurement of cytokine mRNA expression by real time PCR. The third and ﬁnal
objective was to determine if epithelial cell derived cytokines play a role in swine
resistance to C. jejuni infection as determined by in vitro infection of swine intestinal
epithelial cells. To address this issue, IPEC-l cells in both a differentiated and
undifferentiated state were infected separately with C. jejuni strains 81-176 and ATCC

33292. Culture supematants were removed 0, 1, 3, 6, 8, and 24 hours aﬁer infection and

139

.— -4. -2' LI)“. .011

 

analyzed for IL-18 protein levels by ELISA. Total RNA was isolated from cells infected
for 8 hours and IL-18 mRNA expression was measured by RT-PCR. Collectively the
purpose of these experiments was to characterize the nature of the local immune
responses to C. jejuni and T. suis and to determine whether cytokine dysregulation could
play a role in the pathology and disease caused by dual infection.

Severe colonic pathology is characteristic of C. jejuni infection in humans, therefore
characterizing the immune response in the colon is critical for understanding the
pathogenesis of this disease. In a swine infection model C. jejuni targets distal colonic
LGCs, stimulating germinal center expansion with IgA secreting B cells [2, 5]. To assess
the effect of C. jejuni infection on cytokine expression in the distal colon and LGCs, I
developed a rectal challenge model of infection. Weaned piglets inoculated rectally with
2x109 cfu of C. jejuni had 2-3 fold increases in expression of IL-18, TNFoc, and IL-8 in
distal colon samples excluding lymphoid tissue 1 hour after infection, with a sustained
increase in IL-8 expression 24 hours after infection. Concurrently, 3 to 20 fold increases
in mRNA expression of IL-lB, IL-6, IL-8, IL-10, IL-18, GM-CSF, iNOS, MCP-l, and
TNFa were measured from LGCs 1 hour after infection. Expression of IL-4, IL-5, IL-
12p35, IL-l2p40, and IL-13 was unaffected by C. jejuni infection. These data show that
C. jejuni stimulates a proinﬂammatory response in vivo, and the more robust response in
LGCs supports the hypothesis that these secondary lymphoid tissues are important sites
for induction of an immune response directed against this pathogen.

C. jejuni colonizes the intestinal mucus in the colon, associating with epithelial
cells, goblet cells, and invading the deep crypts [1, 2]. In this thesis, I demonstrated

through immunohistochemistry and laser capture microdissection that rectally inoculated

140

C. jejuni is capable of migrating to the same locations as orally inoculated bacteria. This
occurs despite the fact that they did not travel through the acidic pH of the stomach or the
bile rich environment of the small intestine, two environmental conditions that have been
shown to stimulate C. jejuni virulence in vitro [6, 7]. These data correlate with those
found by Bell and Manning who showed that rectal inoculation of naive ferrets resulted
in colonization and diarrheal disease similar to their orally inoculated counterparts [8].
Although rectal inoculation in ferrets has already demonstrated the colonization and
disease causing potential of C. jejuni infection by this nontraditional route, we used this
technique in swine because they are more similar to humans physiologically and we were
able to analyze the local cytokine responses in LGCs which have been characterized in
pigs. Ultimately, these data show that rectal inoculation is a useful technique for studying
in vivo effects of infectious agents with better deﬁnition of the affected area and more
precise timing of responses.

The availability of a genetically modiﬁed strain of C. jejuni that behaved
identically to the wild type parent strain facilitated detection and identiﬁcation in a non
germ free or speciﬁc pathogen free animal infection model. In the rectal challenge study I
used a strain of C. jejuni that had been electroporated with a plasmid containing the gfp
gene under the inﬂuence of a constitutively active promoter. This strain was reisolated
from the feces of a pig 24 hours after rectal inoculation, and was quickly identiﬁed by
examination with an epiﬂuorescent microscope. The original goal was to use this strain to
allow for easy detection of the bacteria in frozen sections of tissue by ﬂuorescence
microscopy. Mixter and colleagues used a strain of C. jejuni expressing gfp to infect mice

intraperitoneally and were able to track the bacteria by ﬂow cytometry for Gfp. They

141

proposed this as a useful technique for detecting the bacteria in vivo [9]. While detection
of ﬂuorescent C. jejuni was successful in the mouse peritoneal infection model, in the
swine model examination of unstained tissue sections microscopically for the ﬂuorescent
bacteria was unsuccessful. The primary impediment to accurate visualization of the strain
used for inoculation was auto ﬂuorescence of the intestinal tissue that masked the
ﬂuorescent bacteria. I tried to overcome this problem by treating the tissues with
propidium iodide to stain nucleic acids to produce background staining of a different
wavelength than Gfp ﬂuorescence to facilitate visualization of the bacteria.
Unfortunately, due to the small size of the bacteria and the presence of debris in the
intestine that did not stain with propidium iodide, the bacteria could not be distinguished
deﬁnitively. In fact, small ﬂuorescent particles of the same size as the bacteria were
present in control pigs as well as infected animals. This phenomenon has also been
encountered in cecal and intestinal samples from chicks infected with C. jejuni with the
same gfp construct (Anna Bates, personal communication). The presence of ﬂuorescent
particles in the intestines severely limited the usefulness of ﬂuorescently tagged C. jejuni
for direct visualization in our system and necessitated the use of IHC, LCM, and DNA
tests to detect and localize the inoculating strain used in the rectal challenge experiment.
However, the gfp marker was very useful for distinguishing the inoculating strain from
naturally occurring Campylobacters in experimentally infected pigs.

IL-18 became a focal cytokine in this work because it has been shown to play an
important early role in the immune response to other enteric infections with similar
pathogenesis. IL-18 is a pleiotrophic cytokine that functions in both innate and acquired

immunity [10, 11]. In mammals studied to date, its mRNA and protein are both

142

constitutively expressed throughout intestinal tissues, including gut associated lymphoid
tissues and intestinal epithelial cells [12, 13]. IL-18 production early in the response to
Shigella and Salmonella infections is critical for stimulation of a Th 1 directed immune
response with increased production of IFNy, which is essential for resolution of disease
[14-16]. In vitro studies have shown that C. jejuni stimulates secretion of IL-1 B, IL-6, IL-
8, and TNFa from swine intestinal epithelial cells (Cunningham et al, unpublished data;
Parthasarathy et al, unpublished data). In addition to initiating a Th 1 directed immune
response, IL-18 stimulates proinﬂammatory cytokine production from leukocytes in vitro
[14-18]. Therefore, we hypothesized that C. jejuni stimulates secretion of IL-18 from
cultured swine intestinal epithelial cells and that this functions to stimulate further
immune responses that kill the bacteria. Differentiated IPEC-l cells secreted signiﬁcant
amounts of IL-18 as early as 3 hours after infection. IL-18 levels remained high at 6 and
8 hours after infection, and then decreased by 24 hours. In contrast, undifferentiated
IPEC-l cells did not secrete signiﬁcant amounts of IL-18 in response to C. jejuni at any
time, despite the fact that they secreted IL-18 in response to the positive control, Con A.
Additionally, mRNA expression of IL-18 was not signiﬁcantly altered by C. jejuni 8
hours after infection of IPEC-l cells in either differentiation state. Taken together, these
data show that C. jejuni does induce IL-18 production from differentiated cells. Increased
secretion in differentiated IPEC-l cells does not correlate with increased mRNA
expression, suggesting that transcriptional regulation does not play a role in the IL-18
response of these C. jejuni infected intestinal epithelial cells. The increased expression of
IL-18 seen in the rectal challenge model may be due to the intraepithelial lymphocytes,

macrophages, or other leukocytes present in the lamina propria and submucosa of the

143

distal colon and the follicles of the LGCs in live pigs with the full anatomic complexity
of the gastrointestinal tract. Increased secretion from differentiated IPEC-l cells is likely
due to increased cleavage of stored immature protein by Caspase-l , facilitating its release
from the cell. These data suggest that IL-18 functions in host defense against C. jejuni
and that intestinal epithelial cells may be an important source of IL-18 in combating this
infection in viva.

Immunomodulation during concurrent infections can determine whether host
pathology and disease are enhanced or abrogated. Pigs are naturally resistant to C. jejuni,
which is likely due in part to a proinﬂammatory response initiated early in life during a
primary infection. Upon infection with the swine whipworm T. suis in the proximal
colon, C. jejuni invade tissues surrounding the site of worm attachment and LGCs in the
distal colon [2, 3]. We hypothesized that C. jejuni invasion secondary to T. suis infection
is due to downregulation of a protective proinﬂammatory immune response. To address
this hypothesis, weaned piglets were orally inoculated with embryonated T. suis eggs.
Twenty one days after inoculation, pigs were orally infected with C. jejuni. Two days
after C. jejuni infection, pigs were killed and we measured cytokine mRNA expression in
the jejunum, proximal colon, and distal colon of pigs infected with T. suis, C. jejuni, or
both. In the proximal colons of both groups of pigs that received T. suis, IL-13 expression
was upregulated 3 to 4 fold over uninfected controls. Concurrently, expression of
cytokines known to be important for clearance of bacterial infections like IL-8 was
decreased 2 to 3 fold, and IL-12 expression was decreased 2 to 4 fold compared to
uninfected controls. Additionally, expression of both the chemokine MCP-l and the Th 2

cytokine IL-S decreased 2 fold in the dual infection group compared to mrinfected

144

controls. At the same time, there were no signiﬁcant differences in expression of any
cytokines in the proximal colons of pigs infected with C. jejuni only compared to
uninfected controls.

These data suggest a role for T. suis stimulated immunomodulation in the
proximal colon. T. suis infection increased expression of IL-13 while concurrently
decreasing the expression of proinﬂammatory cytokines in the proximal colon. IL-13 has
been shown to decrease secretion of IL-8 from stimulated human intestinal epithelial cells
in vitro [19]. Exogenous recombinant IL-13 administered to mice concurrently with a
lethal dose of LPS decreases production of IL-12p40 and protects the mice from lethal
toxemia [20]. Additionally, egg antigens from the intestinal helminth Schistosoma
mansoni stimulate IL-10 production from dendritic cells, while decreasing production of
IL-12 [21]. Based on T. muris infection studies in mice, we expected to see increased
expression of IL-10 in the tissues of these T. suis infected pigs along with increased
expression of IL-13. IL-10 expression was not increased in any tissue sampled from
orally infected pigs, however a concurrent study on secreted cytokines in the feces of
these pigs showed increased IL-10 in T. suis infected pigs on day 22 (Parthasarathy et al,
unpublished results). Increased expression of IL-10 may not have been measured because
we examined tissues 23 days after T. suis infection, which was after the increase of fecal
cytokine levels was measured. Also, T. muris infection studies in mice generally focus on
cytokine production in draining lymph nodes, which does not correlate exactly with local
cytokine expression in the tissues. Nevertheless, increased production of IL-10 on day 22
likely had immunomodulatory effects on cytokine expression on day 23. The combined

actions of T. suis stimulated IL-10 and IL-13 are likely responsible for decreased

145

expression of IL-8, IL-12, and possibly MCP-l as well. Based on data from previous dual
infection studies, we expected to see increased numbers of C. jejuni in the crypts and
tissues of dual infected pigs. C. jejuni was detected by RFLP analysis in 4 of 6 pigs from
the dual infection group, and l of 6 pigs from each of the single infection groups and the
controls. Increased prevalence of C. jejuni in the proximal colon of dual infected pigs
compared to pigs that received C. jejuni only suggests that T. suis facilitates its
proliferation and possibly invasion. Detection of C. jejuni from control animals suggests
that C. jejuni was likely present in low numbers as a commensal in these pigs. Taken
together, these data suggest that decreased proinﬂammatory cytokine expression resulting
from T. suis infection facilitates proliferation of C. jejuni.

In the distal colon, cytokine expression was only altered signiﬁcantly in the T. suis
only group. Three to four fold increases in expression of IL-10, IL-6 and IFNy were
measured, while expression of IL-13 was upregulated 3 fold compared to uninfected
controls. Increased expression of IL-lB, IL-6 and IFNy is indicative of an inﬂammatory
response, possibly due to secondary bacterial infection. C. jejuni was not detected by
RFLP in the distal colons of any pigs, however, it has previously been demonstrated that
T. suis facilitates secondary invasion of many species of resident bacteria, including C.
coli, an unidentiﬁed Campylobacter spp., and E. coli [3]. Invasive strains of C. coli
stimulate IL-8 secretion from cultured human intestinal epithelial cells [22]. Analysis of
distal colon tissues by RFLP of the Campylobacter 23S rRNA gene detected C. coli in 2
of 6 pigs, and an unidentiﬁed species of Campylobacter in 3 of 6 pigs. Thus it is possible
that either of these species invaded tissues in the distal colon and initiated an

inﬂammatory response. It has also been shown that oral inoculation of pigs with E. coli

146

DHSa causes increased secretion of proinﬂammatory cytokines in the feces
(Cunningham et al, unpublished results). Larval stages of T. suis produce ESP in vivo, as
is evidenced by the fact that serum antibodies directed against a 20kDa glycoprotein
isolated from adult worm culture ﬂuids can be detected as early as 21 days after
inoculation with embryonated eggs [23]. In addition, T. suis ESP stimulates production of
IL-6 and IL-10 from IPEC-l swine intestinal epithelial cells in vitro. Therefore, increased
expression of IL-13 in the distal colon of T. suis infected pigs could be attributed to the
effects of excretory/secretory product (ESP) produced by the 4th stage larvae in the
proximal colon that have traveled to the distal colon. Taken together, these data indicate
that ESP is highly stimulatory and likely has effects on intestinal epithelial cells which
would increase cytokine production in vivo at sites distal to worm attachment.
Interestingly, although the cytokine response in the proximal colon was similar
between the T. suis and dual infection groups, the responses diverged in the distal colon.
There were no signiﬁcant changes in cytokine expression in the distal colon of the dual
infection group. This was unexpected, as previous studies have shown that T. suis
facilitates invasion of C. jejuni and resident bacteria into the LGCs in the distal colon [2,
3], and distal colonic samples from our pigs likely included LGCs. In our rectal challenge
model we showed that C. jejuni stimulates proinﬂammatory cytokines in LGCs 1 hour
after infection, therefore we expected to see a strong cytokine response from the dual
infected pigs. Tissues for cytokine expression from orally inoculated pigs were taken 48
hours after C. jejuni infection, which corresponded to 23 days after T. suis infection. Our
rectal challenge model showed that proinﬂammatory cytokine responses were seen

primarily 1 hour after infection, therefore it is likely that proinﬂammatory cytokines had

147

returned to baseline levels by the time samples were taken from orally infected pigs. The
inability to recover C. jejuni from fecal culture of these pigs 48 hours after oral
inoculation indicated that low levels of C. jejuni were present. The oral inoculum was 108
cfu of C. jejuni, while the inoculum used in the rectal challenge experiment was 2x109
cfu of C. jejuni. Therefore the low level of C. jejuni present in the distal colon 48 hours
may have been too low to stimulate a signiﬁcant proinﬂammatory response. It is more
likely that the absence of a measurable proinﬂammatory cytokine response in the distal
colons of dual infected pigs is due to sampling after the response has waned. Indeed,
gnotobiotic piglets orally inoculated with only 106 cfu of C. jejuni 33292 developed
transient fever that lasted only for the ﬁrst 24 hours after infection, suggesting that C.
jejuni stimulated a proinﬂammatory response with increased production of IL-6 that
caused fever [2].

Another issue that may have affected distal colonic cytokine responses is the
larval stage present in the pigs. T. suis undergoes 4 molts from L1 larvae to L5 adults in
about 41 days [24, 25]. Mucohemorrhagic diarrhea associated with T. suis infection can
occur as early as 14 days in heavy infection, but on average occurs around 21 days [26,
27]. This corresponds to the emergence of the posterior end of L4 larvae into the lumen
[24, 25]. The severity of diarrheal disease and pathology increases as worms mature to
adults, reproducing and excreting signiﬁcant amounts of ESP [27] (Linda Mansﬁeld,
personal communication). Progression of disease and pathology likely correlates with
intensiﬁed cytokine responses. Though T. suis larvae were present in the proximal colons
of dual infected pigs, and signiﬁcant histologic changes were seen, lesions were conﬁned

to the areas where worms were present. The time that the pigs in our study were sampled

148

corresponded to the beginning of the molt from L3 to L4 larvae, and ESP production
would have been low. Therefore, it is possible that in the dual infection groups, distal
colonic effects of T. suis would not yet have been manifested. Analysis of cytokine
expression at later time points, including 30 days and 41 days after T. suis infection
would determine whether an inﬂammatory response is induced in the distal colon as
disease progresses.

C. jejuni infection alone stimulated signiﬁcant changes in the jejunum of orally
infected animals. In the jejunum of C. jejuni infected pigs expression of MCP-l, TNFat,
IL-12p40, IFNy, IL-4, and IL-10 were increased 3 to 5 fold over uninfected controls. C.
jejuni stimulates intracellular production of MCP-l, TNFoc, IFNy, IL-4, and IL-10 from
cultured human intestinal epithelial cells [28, 29]. Additionally, jejunal samples likely
contained samples of peyers patches as well as surrounding epithelium. It has been
demonstrated in our rectal challenge model, and in an oral challenge model of germ free
piglets [2] that C. jejuni targets LGCs and stimulates robust cytokine responses. It is
likely that C. jejuni stimulated a similar response in the jejunal Peyers patch, which
would have inﬂuenced our results. These results indicate that C. jejuni stimulates a
proinﬂammatory response in the jejunum in the absence of T. suis. In the jejunum of pigs
infected with C. jejuni and T. suis concurrently, GM-CSF is downregulated 5 fold
compared to uninfected controls, and the rest of the cytokines measured were not
signiﬁcantly different that uninfected controls. This suggests that T. suis exerts
immunomodulatory effects in the jejunum as well as the proximal colon, down regulating
C. jejuni induced proinﬂammatory cytokines. These effects could be due to systemic IL-

13 and IL-10 produced by lymphocytes stimulated by worm antigens in draining lymph

149

nodes. Examination of cytokine expression in draining lymph nodes and peripheral
leukocytes would be necessary to conﬁrm this.

Increased expression of proinﬂammatory cytokines in the distal colon and LGCs
of rectally inoculated pigs and in the jejunum of orally inoculated pigs indicated that C.
jejuni stimulated a proinﬂammatory response in viva. However, the fact that the pigs in
these studies were immunocompetent, conventionally reared animals with a complete
intestinal microbial population raises the question of whether some of the cytokine
responses are the result of direct actions of C. jejuni on intestinal tissues, or indirect
actions due to perturbation of intestinal ﬂora homeostasis by the introduction of
exogenous bacteria. Pigs orally infected with a non-pathogenic laboratory strain of E. coli
DHSa have increased secretion of IL-6 in the feces 48 hours after inoculation, but this
strain does not invade swine intestinal epithelial cells in vitro (Parthasarathy et al,
unpublished results). Introduction of exogenous bacteria could increase the total intestinal
bacterial load above normal levels. Additionally, any exogenous strain could compete
with resident strains for resources or provide an essential growth factor, causing an
imbalance in the intestinal environment and facilitating overgrowth of resident strains.
These factors could stimulate inﬂammatory responses in an effort by the host to reduce
total bacterial numbers and possibly eliminate the exogenous strain, restoring balance to
the microbial population. Taking this into account, exogenous C. jejuni may have
stimulated a host inﬂammatory response indirectly. However, in vitro data deﬁnitively
shows that C. jejuni infection alone stimulates secretion of a similar panel of

proinﬂammatory cytokines from cultured intestinal epithelial cells, suggesting that the

150

responses seen in viva can be attributed to both direct and indirect effects of C. jejuni on
intestinal tissues.

Collectively, the data from this dissertation project support the hypothesis that
cytokine dysregulation during dual infection with T. suis and C. jejuni is a major cause of
the severe disease and pathology observed. From this analysis of local cytokine
expression in pigs infected singly or concurrently with C. jejuni and T. suis, we propose
the following model for the roles of speciﬁc cytokines in disease progression in vivo. A
proinﬂammatory response is stimulated by C. jejuni early in the infection, from both
intestinal epithelial cells and LGCs. IL-1 8 released from intestinal epithelial cells early in
the immune response induces IL-lB, IL-8, and TNFa from leukocytes, amplifying and
sustaining inﬂammation. Additionally, epithelial cell derived IL-18 recruits and activates
neutrophils to engulf and eliminate extracellular bacteria. With the addition of T. suis, IL-
13 is upregulated while IL-8 and IL-12 are downregulated. IL-13 and IL-4 stimulate
mucin production from intestinal epithelial cells [30, 31] which provides a favorable
environment for proliferation of C. jejuni [32, 33]. Piglets infected with a low dose of C.
jejuni, 106 cfu, develop transient fever but no diarrhea, while piglets infected with high
doses of C. jejuni, 4-5x109 cfu, develop diarrheal disease and pathology [1, 2, 34].
Overgrowth of C. jejuni in a mucus rich environment could be responsible in part for the
mucohemorrhagic diarrhea seen in dual infection. Also, IL-13 downregulation of IL-8 is
likely to abrogate neutrophil recruitment and containment of bacteria in the epithelial
layer [35, 36], allowing invasion by signiﬁcantly increased numbers of C. jejuni into

deeper tissues.

151

Other cytokines may enhance invasion of C. jejuni. IL-4 has also been shown to
decrease transepithelial electrical resistance of differentiated IPEC-l cells and increase
invasion of C. jejuni (Parthasarathy et al, unpublished results). Although increased IL-4
expression was not detected in our oral challenge study, concurrent analysis of secreted
cytokines in the feces showed increased secretion on day 22 in dual infected pigs
(Parthasarathy et al, unpublished results). In vivo, T. suis induced IL-4 may weaken tight
junctions between epithelial cells allowing bacteria access to the basolateral surface of
cells. These bacteria that have breached the epithelial barrier could bind ﬁbronectin in the
extracellular matrix via CadF [37] and gain access to the lamina propria and submucosa.
Another factor to consider is the secreted worm products. ESP has a dose dependent
cytotoxic effect on undifferentiated epithelial cells. Additionally, ESP exposed to both
apical and basolateral surfaces of differentiated cells signiﬁcantly decreases
transepithelial electrical resistance and damages perijunctional complexes, making
cellular junctions more permissive [34]. The combined effects of ESP on differentiated
and undifferentiated epithelial cells would result destruction of the epithelial barrier in
viva, allowing invasion of C. jejuni into deeper tissues, increasing the severity of disease.
Ultimately, the combined actions of T. suis stimulated IL-4 and IL-13 would provide a
favorable environment for C. jejuni proliferation, while dampening the host response that

would protect against invasion by overwhelming amounts of C. jejuni.

152

Future Directions

The analysis of local cytokine expression in intestinal tissues from C. jejuni and
T. suis infected pigs expanded our knowledge of the swine immune response to these two
agents and shed some light on cytokine dysregulation that can occur during infection with
these two agents. These data would be enhanced by exploring several aspects of this dual
infection phenomenon. In chapter 2 we analyzed cytokine expression in the distal colon
and LGCs of pigs rectally inoculated with C. jejuni. Further analysis of cytokine
expression at different times, for example 4, 8, and 12 hours after infection, would be
instructive as to the in vivo kinetics of cytokine expression in response to C. jejuni.
Additionally, since our data showed that the cytokine response was more robust in the
LGCs compared to the distal colon, identiﬁcation of cytokine producing cells by in situ
hybridization or ﬂow cytometry in these two tissues would elucidate the speciﬁc cells
activated by C. jejuni. Concurrent analysis of intracellular protein within the cells by
immunohistochemistry, and of proteins secreted into the feces by ELISA would allow
comparison of changes in expression and changes in secretion.

In chapter 3 we analyzed cytokine expression in the jejunum, proximal colon, and
distal colon of pigs orally infected with C. jejuni, T. suis, or both. As with the rectally
inoculated pigs, further analysis of cytokine expression at different times with concurrent
analysis of intracellular and secreted proteins would expand our knowledge of in viva
cytokine kinetics. Analysis of cytokine expression 4, 8, 12, and 24 hours after infection
would allow us to determine if proinﬂammatory responses were induced in C. jejuni

infected pigs early in the response to oral infection. In addition, examination of cytokine

153

expression at later time points, for example on day 30, day 41, and day 48 after T. suis
infection, would help us to asses cytokine changes that may coincide with developmental
changes in T. suis. Concurrent examination of pathologic changes would correlate lesion
progression with cytokine changes.

Previous studies of immunoglobulin production from piglets orally inoculated
with C. jejuni and the rectal challenge study from this dissertation demonstrated the
importance of LGCs as sites for induction of immune responses to C. jejuni. To further
characterize the cytokine response to C. jejuni infection, identiﬁcation of speciﬁc cell
types that are reactive to C. jejuni and characterization of their cytokine responses should
be undertaken. To achieve this, multiple LGCs would be collected from pigs orally or
rectally infected with C. jejuni. Individually, each LGC would be gently disrupted to
make single cell suspensions and an aliquot would be tested for the presence of C. jejuni
by PCR ampliﬁcation of an appropriate gene, for example the QRDR of the gyrA gene.
LGCs that do not contain C. jejuni would serve as internal negative controls for each pig.
The cells would then be stimulated in vitro with C. jejuni and subsequently stained for
speciﬁc cell markers to detect CD4+ cells, CD8+ cells, B-cells, and macrophages. The
cells would be separated by ﬂow cytometry and assayed for expression of a panel of
proinﬂammatory and anti-inﬂammatory cytokines by real time PCR. This study would
determine which cell types are involved in the immune response to C. jejuni and also
provide information about the distribution of C. jejuni in vivo.

In chapter 4 we measured IL- 1 8 secretion from differentiated and undifferentiated
IPEC-l cells infected with C. jejuni. The difference in responses between the two

differentiation states raises the question of whether cellular maturation coincides with

154

intracellular signaling pathways that allow differentiated cells to secrete IL-18 in
response to C. jejuni, while undifferentiated cells do not. This question could be
addressed by examination of C. jejuni activated intracellular signaling molecules using
western blot analysis of intracellular proteins. We may ﬁnd that at some point in the
signal transduction cascade a key activator of signaling molecules speciﬁc to the IL-18
pathway is absent or repressed in undifferentiated cells, indicating that the differentiation
process includes synthesis of this activator or derepression of its expression. We also
found that E. coli LPS does not stimulate IL-l8 secretion from differentiated IPEC-l
cells. Previous studies have shown that adherence, invasion, and Cdt are virulence factors
of C. jejuni that stimulate IL-8 secretion from intestinal epithelial cells [22, 38]. The
effect of these and other speciﬁc virulence factors on IL-18 secretion from swine
intestinal epithelial cells should be explored to determine the mechanisms that C. jejuni
uses to stimulate host cytokine responses in the gut.

One factor that affected interpretation of the results from our in vivo studies,
discussed in chapters 2 and 3, was the presence of other Campylobacters in
conventionally reared pigs. It has been reported that healthy pigs can harbor C. jejuni, C.
coli, and C. lari as commensals [4]. In my studies C. coli was detected by RFLP analysis
of the Campylobacter 23S rRNA gene in both rectally and orally infected control pigs.
An unidentiﬁed Campylobacter species that had an RF LP pattern that was different from
all the other thermophilic Campylobacters detected by this assay was found in at least
one intestinal segment from virtually 100% of orally infected pigs.
Immunohistochemistry using antibody raised against whole C. jejuni bacteria positively

stained organisms in the colons of all experimental pigs. Antibody validation tests

155

revealed cross reactivity with C. lari, but preliminary PCR analysis of tissue DNA did not
detect C. lari in any pigs regardless of infection route. These data suggest that another
Campylobacter species was present in the colons of our experimental animals that
occupied the same niche as C. jejuni. A detailed analysis of the resident Campylobacter
strains in experimental animals would help to determine whether the changes in cytokine
expression were due to direct interaction of C. jejuni with colonic tissues, to the
interaction of C. jejuni with the unidentiﬁed Campylobacter resulting in it invading
tissues, or to the interaction of C. jejuni with the resident bacterial population causing a
perturbation of environmental homeostasis. To address this issue we would clone and
sequence l6S rRNA genes from as many Campylobacter species ampliﬁed from total pig
intestinal DNA as possible, then compare the sequences obtained from our pigs to
sequences in bacterial ribosomal databases. The results from this experiment may reveal
that C. lari is present in the pigs, that a variant of C. jejuni, C. coli, or C. lari is present
that gives a unique RFLP pattern using the 23S rRNA gene, or that an as yet
uncharacterized Campylobacter is present in the pigs. This analysis is currently
underway. The data gathered from these experiments could also be enhanced by
quantifying Campylobacters in the feces of experimental animals and comparing
bacterial loads between controls and infected animals. We may ﬁnd that the total
bacterial load is unchanged, but the ratios of different organisms are different and might
indicate a dose dependent effect of Campylobacter infection on local intestinal cytokine
production.

In summary, the data generated from this dissertation contributed to the body of

research deﬁning host immune responses to C. jejuni and T. suis infections. We

156

developed a swine rectal challenge model that proved to be a useful technique for
studying the complexity of the host immune response to C. jejuni, while allowing the
precise timing and speciﬁc targeting of tissues afforded by in vitro models. We began to
deﬁne the host local immune response to T. suis and the T. suis induced
immunomodulation that may decrease natural resistance to C. jejuni, facilitating invasion.
Finally, through in vitro infection studies we discovered that C. jejuni infected intestinal
epithelial cells are a source of early IL-18 production, which may be critical in initiating a
protective response to infection in viva. Collectively this data gave us evidence for T. suis
downregulation of proinﬂammatory responses that are detrimental to swine colonized by

the opportunistic pathogen C. jejuni.

157

The results obtained from this dissertation opened several avenues for further
research and generated new questions to be addressed. Experimental designs of studies to

address three of these questions will be discussed in detail.

Cytokine responses of intestinal epithelial cells to multi-species Campylobacter

infection

Healthy, conventionally reared pigs can harbor multiple Campylobacter species in
their colons without evidence of disease [4]. In our swine rectal challenge model,
inoculation of C. jejuni into the distal colons of conventionally reared pigs with naturally
acquired C. coli stimulated increased expression of proinﬂammatory cytokines. The
hypothesis for this study is that swine intestinal epithelial cells exposed to C. coli will
secrete proinﬂammatory cytokines when given a secondary challenge with C. jejuni.
Differentiated and undifferentiated IPEC-l cells will be exposed to C. coli for 0, 1, 3, 6,
8, and 24 hours. The supematants will be removed and saved for cytokine assay by
ELISA. Then the cells will be exposed to C. jejuni expressing Gfp, which are also
resistant to kanamycin, for l, 3, 6, 8, and 24 hours. The supematants will be removed and
saved for cytokine assay by ELISA, and then the cell monolayers will be treated with
100mg/mL gentamicin for 1 hour to kill extracellular bacteria. The gentamicin will be
removed, then the cells will by lysed in 0.1% sodium deoxycholate and the cell lysates
will be plated on Bolton agar with kanamycin to select for the inoculating strain of C.
jejuni. Supematants from C. coli only infected cells as well as cells infected with C. coli,

then subsequently infected with C. jejuni will be assayed for IL-IB, IL-6, IL-8, lL-18,

158

and TNFor by ELISA. These experiments will determine 1) if C. coli stimulates
proinﬂammatory cytokine production from IPEC-l cells; 2) if C. jejuni stimulates a
proinﬂammatory response in cells previously exposed to C. coli; and 3) if C. jejuni

invasion into IPEC-l cells is altered by prior exposure of the cells to C. coli.

159

The effects of IL-13 on proinflammatory cytokine production and C. jejuni invasion

in viva

T. suis induced IL-13 downregulates expression of proinﬂammatory cytokines in
the proximal colons of pigs concurrently infected with C. jejuni (Jones et al, discussed in
Chapter 3). In a rectal challenge model, expression of proinﬂammatory cytokines is
signiﬁcantly increased in pigs inoculated with C. jejuni. The hypothesis for this study is
that IL-13 will downregulate proinﬂammatory cytokines in the distal colon and facilitate
invasion of epithelial cells by C. jejuni. Using a rectal challenge model, the distal colons
of pigs will be pretreated with recombinant IL-13, serum free media, or no treatment for
5 hours, then all pigs will be infected with C. jejuni expressing Gfp rectally. Three hours
after C. jejuni infection, the pigs will be killed and the distal colons will be removed.
Intestinal epithelial cells will be isolated by the enzymatic digestion technique described
by Babakhani and Joens [1]. Brieﬂy, the colonic sections will be washed with PBS, then
the enterocytes will be detached by incubation with 0.25% trypsin-lmM EDTA at 37°C
on a rotator. Detached cells will be washed three times with MEM supplemented with 1%
FBS, then the cells will be incubated for 1 hour with lOOug/mL gentamicin in MEM
supplemented with 1% FBS to kill extracellular bacteria. The cells will then be ﬁxed with
paraforrnaldehyde, and cells containing internalized ﬂuorescent C. jejuni can be separated
and counted by ﬂow cytometry. The cells could then be stained for intracellular IL-8
protein with speciﬁc antibodies. These experiments will determine 1) if recombinant IL-

13 facilitates increased invasion of intestinal epithelial cells with C. jejuni in vivo; 2) if C.

160

jejuni invasion stimulates IL-8 production from intestinal epithelial cells; and 3) if C.

jejuni stimulated IL-8 secretion is decreased by IL-13 pretreatment.

161

The effects of IL-18 on C. jejuni invasion of intestinal epithelial cells in vitro

IL-18 is a pleiotrophic cytokine that is induced in viva during infections with
enteric bacteria, and is critical for host resistance to the pathogens. C. jejuni stimulates
IL-18 secretion from differentiated IPEC-l cells in vitro. In addition, IL-18 expression is
signiﬁcantly increased in the distal colons and LGCs of pigs one hour after rectal
inoculation with C. jejuni. The hypothesis of this study is that early production of IL-18
from intestinal epithelial cells after infection with C. jejuni is important for resistance to
invasion. Two experimental designs will be used to test this hypothesis. In the ﬁrst
experiment, differentiated IPEC-l cells will be pretreated with different doses of
recombinant IL-18 for 5 hours. The supernatant will be removed and saved for later
cytokine assay. The pretreated cells will then be infected with C. jejuni for 3 hours, and
then treated with lOOpg/mL gentamicin for 1 hour to kill extracellular bacteria. The
gentamicin will be removed, the cells will be lysed with 0.1% sodium deoxycholate, and
cell lysates will be plated to enumerate internalized bacteria. In the second experiment,
differentiated IPEC-l cells will be treated with anti-IL-18 antibody and concurrently
infected with C. jejuni. Supematants will be removed 1, 3, 6, 8, and 24 hours after
infection and saved for later cytokine assay. The infected monolayers will be treated with
gentamicin to kill extracellular bacteria, then lysed and plated to enumerate internalized
bacteria. The supematants from both experiments will be analyzed by ELISA for IL-IB,
IL-6, and IL-8 which can all be induced by IL-18 stimulation [17, 18]. These experiments
will determine 1) if pretreatment with recombinant IL-l8 enhances IPEC-l resistance to

invasion; 2) if blocking of IL-18 with antibodies during C. jejuni infection facilitates

162

invasion; and 3) if IL-l8, either exogenous or recombinant, induces increased secretion of

IL-1 [3, IL-6, and IL-8 during C. jejuni infection.

I63

10.

ll.

12.

l3.

I4.

15.

16.

References

Babakhani, F.K., G.A. Bradley, and LA. Joens, Newborn piglet model for campylobacteriosis.
Infect Immun, 1993. 61(8): p. 3466-75.

Mansﬁeld, L.S., et al., Enhancement of disease and pathology by synergy of Trichuris suis and
C ampylobacter jejuni in the colon of immunologically naive swine. Am J Trop Med Hyg, 2003.
68(3): p. 70-80.

Mansﬁeld, LS. and J .F. Urban, Jr., The pathogenesis of necrotic proliferative colitis in swine is
linked to whipworm induced suppression of mucosal immunity to resident bacteria. Vet Immunol

Immunopathol, 1996. 50(1-2): p. 1-17.

Young, C.R., et al., Enteric colonisation following natural exposure to Campylobacter in pigs.
Res Vet Sci, 2000. 68(1): p. 75-8.

Mansﬁeld, LS. and D.T. Gauthier, Lymphoglandular complexes are important colonic sites for
immunoglobulin A induction against Campylobacterjejuni in a swine disease model. Comp Med,

2004. 54(5): p. 514-23.

Allen, K.J. and M.W. Griffiths, Eﬂect of environmental and chemotactic stimuli on the activity of
the Campylobacterjejuni/MA sigma(28) promoter. FEMS Microbiol Lett, 2001. 205(1): p. 43-8.

Rivera-Ami“, V., et al., Secretion of the virulence-associated Campylobacter invasion antigens
from Campylobacterjejuni requires a stimulatory signal. J Infect Dis, 2001. 183(11): p. 1607-16.

Bell, J .A. and DD. Manning, A domestic ferret model of immunity to Campylobacterjejuni-
induced enteric disease. Infect Immun, 1990. 58(6): p. 1848-52.

Mixter, P.F., et al., In vivo tracking of C ampylobacter jejuni by using a novel recombinant
expressing green fluorescent protein. Appl Environ Microbiol, 2003. 69(5): p. 2864-74.

Akira, S., The role of IL-18 in innate immunity. Curr Opin Immunol, 2000. 12(1): p. 59-63.

Dinarello, C.A., et al., Overview of interleukin-1 8: more than an interferon-gamma inducing
factor. J Leukoc Biol, 1998. 63(6): p. 658-64.

Foss, D.L., M.J. Zilliox, and M.P. Murtaugh, Bacterially induced activation of interleukin-I 8 in
porcine intestinal mucosa. Vet Immunol Immunopathol, 2001. 78(34): p. 263-77.

Foumout, S., et al., Cloning, chromosomal location, and tissue expression of the gene for pig
interleukin-I8. Immunogenetics, 2000. 51(4-5): p. 358-65.

Biet, F ., C. Locht, and L. Kremer, Immunoregulatory functions of interleukin I 8 and its role in
defense against bacterial pathogens. J Mol Med, 2002. 80(3): p. 147-62.

Elhofy, A. and K.L. Bost, Limited interleukin-I8 response in Salmonella-infected murine
macrophages and in Salmonella-infected mice. Infect Immun, 1999. 67(10): p. 5021-6.

Sansonetti, P.J., et al., Caspase—I activation of IL-1 beta and lL-I8 are essential for Shigella
ﬂexneri-induced inflammation Immunity, 2000. 12(5): p. 581-90.

164

I7.

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

Netea, M.G., et al., Interleukin- I 8 induces production of proinﬂammatory cytokines in mice: no
intermediate role for the cytokines of the tumor necrosis factor family and interleukin-I beta. Eur J
Immunol, 2000. 30(10): p. 3057-60.

Puren, A.J., et al., Interleukin-I8 (IFNgamma-inducing factor) induces IL-8 and IL-I beta via
T NF alpha production from non-CD14 + human blood mononuclear cells. J Clin Invest, 1998.
101(3): p. 711-21.

Lugering, N., et al., Interleukin (IL)-I3 and IL-4 are potent inhibitors of IL-8 secretion by human
intestinal epithelial cells. Dig Dis Sci, 1999. 44(3): p. 649-55.

Muchamuel, T., et al., IL-I3 protects mice from lipopolysaccharide-induced lethal endotoxemia:
correlation with down-modulation of TNF-alpha, IF N-gamma, and IL- I 2 production. J Immunol,
1997. 158(6): p. 2898-903.

Kane, C.M., et al., Helminth antigens modulate TLR-initiated dendritic cell activation. J Immunol,
2004. 173(12): p. 7454-61.

Hickey, T.E., et al., Campylobacterjejuni cytolethal distending toxin mediates release of
interleukin-8ﬁ'am intestinal epithelial cells. Infect Immun, 2000. 68(12): p. 6535-41.

Hill, D.E., R.D. Romanowski, and J .F . Urban, Jr., A Trichuris speciﬁc diagnostic antigen from
culture ﬂuids of Trichuris suis adult worms. Vet Parasitol, 1997. 68(1-2): p. 91-102.

Beer, RJ., Morphological descriptions of the egg and larval stages of Trichuris suis Schrank,
I788. Parasitology, 1973. 67(3): p. 263-78.

Beer, RJ., Studies on the biology of the life-cycle of Trichuris suis Schrank, I788. Parasitology,
1973. 67(3): p. 253-62.

Batte, E.G., et al., Pathophysiology of swine trichuriasis. Am J Vet Res, 1977. 38(7): p. 1075-9.

Rutter, J .M. and RJ. Beer, Synergism between Trichuris suis and the microbial ﬂora of the large
intestine causing dysentery in pigs. Infect Immun, 1975. 11(2): p. 395-404.

Al-Salloom, F .S., et al., Campylobacter-stimulated IN T407 cells produce dissociated cytokine
proﬁles. J Infect, 2003. 47(3): p. 217-24.

Bakhiet, M., et al., Induction of alpha and beta chemokines by intestinal epithelial cells stimulated
with Campylobacterjejuni. J Infect, 2004. 48(3): p. 236-44.

Blanchard, C., et al., IL-4 and IL-I3 up—regulate intestinal trefoil factor expression: requirement
for STA T6 and de novo protein synthesis. J Immunol, 2004. 172(6): p. 3775-83.

Iwashita, J ., et al., mRNA of MUCZ is stimulated by IL-4, IL-I3 or TNF-alpha through a mitogen-
activated protein kinase pathway in human colon cancer cells. Immunol Cell Biol, 2003. 81(4): p.
275-82.

Beery, J .T., M.B. Hugdahl, and M.P. Doyle, Colonization of gastrointestinal tracts of chicks by
Campylobacterjejuni. Appl Environ Microbiol, 1988. 54(10): p. 2365-70.

McSweegan, E. and R.I. Walker, Identiﬁcation and characterization of two Campylobacterjejuni
adhesinsfor cellular and mucous substrates. Infect Immun, 1986. 53(1): p. 141-8.

165

34.

35.

36.

37.

38.

Abner, S.R., et al., Response of intestinal epithelial cells to Trichuris suis excretory-secretory
products and the influence on Campylobacterjejuni invasion under in vitro conditions. J Parasitol,
2002. 88(4): p. 738-45.

Netea, M.G., et al., Neutralization of lL-I 8 reduces neutrophil tissue accumulation and protects
mice against lethal Escherichia coli and Salmonella typhimurium endotoxemia. J Immunol, 2000.
164(5): p. 2644-9.

Sansonetti, P.J., et al., Interleukin-8 controls bacterial transepithelial translocation at the cost of
epithelial destruction in experimental shigellosis. Infect Immun, 1999. 67(3): p. 1471-80.

Konkel, M.B., et al., Identification and molecular cloning of a gene encoding a fibronectin-
binding protein (CadF) from Campylobacterjejuni. Mol Microbiol, 1997. 24(5): p. 953-63.

Hickey, T.E., et al., Campylobacterjejuni-stimulated secretion of interleukin-8 by IN T407 cells.
Infect Immun, 1999. 67(1): p. 88-93.

166