;1155§3§E},.
' “11131535;
21911’7‘ (‘1
’ , ”In“ .

‘I1
.L‘J“

V‘ffigy'q.

. 1
A (”*h“ «

v;

3.13%.

1;
Tu-‘H‘j‘
L‘W'

a..." n
:1“
—-

‘5.»w‘.
" At

I 15"“; MW!
11 .

‘1. . V '.'.}.. 4;...
U" -k“

.;

A

f
‘1
l
‘

- «:2:

.4. ...

'P‘~‘

.-

"'3 "1.5— fl. .

9 -o.

42+

A.
.741:

‘1 5;} 3:?

4‘-

.; 3}.“ ‘9';

A—Lv
”:-

w

1 «.31
r’!‘ 7.! 1-
:5? :2”

1:,

Le
a

1!
fig-'33“

iii-32;!“-
g?% [1"...

I.- ‘vh‘
5%7‘ .

rm

2..
be,

...-_

 

”m
3
212’)”.
é <0 3 5 "’7 57/19

This is to certify that the
dissertation entitled

ROLE OF TH-2 CYTOKINES IN CAMPYLOBACTER JEJUNI
AND TRICHURIS SUIS MEDIATED PATHOGENESIS IN
SWINE

presented by

GEETHA PARTHASARATHY

has been accepted towards fulfillment
of the requirements for the

Ph.D degree in Microbiology and Molecular
Genetics

 

 

  

 
    
 

. . ‘ l‘.,£“‘ 4’
Major Professor’s $19 ,I:

<32? 7/0200 {7/

Date

 

MSU is an Affinnative Action/Equal Opportunity Institution

 

LIBRARY
Michigan State
University

 

 

 

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

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/01 c10lFlC/DateDuo.pS§-p.15

 

ROLE OF TH-Z CYTOKINES IN CAMPYLOBA C TER JEJUNI AND T RICHURIS
S UIS MEDIATED PATHOGENESIS IN SWINE

By

Geetha Parthasarathy

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

2004

ABSTRACT

ROLE OF TH-Z CYTOKINES IN CAMPYLOBAC T ER JEJUNI AND TRICHURIS
SUIS MEDIATED PATHOGENESIS IN SWINE

By

Geetha Parthasarathy

Campylobacterjejuni, an enteric bacterium, is one of the leading causes of
bacterial enteritis in the world and T richuris suis is a whipworrn helrninth of the swine
intestinal tract. C. jejuni and T. suis acted synergistically in the colon of neo-natal genn-
free piglets to cause disease more severe than the additive effects of either of the single
infection. We hypothesized that cytokine dysregulation could be a contributing factor to
this phenomenon, and that Th-2 cytokines might play a role in this polymicrobial disease.
Thus the goal of this research was to evaluate the role of Th-2 cytokines such as IL-4, IL-
6 and IL-10 in immunity to C. jejuni and T. suis under in vitro and in vivo conditions, and
further evaluate the role(s) of individual cytokines in C. jejuni and T. suis mediated

pathogenesis in swine.

To my parents, Padma and Parthasarathy, my sister Radha, and my brother-in-law Ravi,

you are my real support and my strength through this unforgettable journey.

iii

ACKNOWLEDGEMENTS

The past six years have been a growing-up experience in many ways, and through
it all I was fortunate enough to meet some incredible people who will be milestones in
my life’s journey.

I would like to thank Dr. Linda Mansfield and Dr. John Linz for their support,
encouragement and valuable advice through out these past years. Many a times have I
stumbled, and you have been there to lend me a helping hand. I would like to thank my
committee members Dr. Richard Schwartz, Dr. Katheryn Meek and Dr. Martha Mulks for
their guidance and support through these years.

I would also like to thank the Dept. of Microbiology, Large Animal Clinical
Sciences, and the Graduate School for providing me with various fellowships and
assistantships over the years. Thanks to Dr. John Baker for believing in me and helping
me out at times of financial need.

I had the good fortune to work with many incredible people around me; Lakeisha
Cunningham, Dr. Kathryn Jones, Hany Elsheika, Dennis Shubitowski, and Sheila Abner—
thanks for listening to my problems; Dave Wilson -thanks for the intellectual and
philosophical conversations; Dr. Julia Bell- you are truly a God send to our lab.

My terrific friends; Shikha Gupta, Gauri J awdekar, Aarthi Rishi, Harini
Krishnamurthy and Shibani Mukherjee-you have all touched my life one way or another.
The past six years would not be possible but for you all.

Thanks to Pawin Padungtod, Bo Norby and Seth Walk for help with statistics; Dr.
Joe Leykam and staff of the Macromolecular Structure Facility for the sometime free-of

charge primers; Dr. Terese Best and staff at the Genomic Sequencing Facility for the

iv

sometime free-of-charge sequencing reactions; Russ Somers and Moses Fetters for
helping with my computer problems; Dr. Shirley Owens, Dr. Alicia Pastor and Dr. Ewa
Danielawicz for facilitating my microscopy experiments; Angie Zell, Jennifer Sysak,
Angela Corley, Tonie Brovont, Susan Dies and Margaret Nicholas, for all the
administrative support; the custodial staff at the National Food Safety and Toxicology
Center for all the night chats.

Last but not the least, thanks to my family and God for being at my side always.

TABLE OF CONTENTS

LIST OF TABLES ......................................................................................................... viii

LIST OF FIGURES ......................................................................................................... ix
CHAPTER 1

LITERATURE REVIEW ................................................................................................. l

Campylobacter ........................................................................................................ 2

Microbiology ..................................................................................................... 2

Epidemiology .................................................................................................... 6

Pathogenesis ...................................................................................................... 9

Immune responses ........................................................................................... 23

T richuris suis ......................................................................................................... 26

Morphology ..................................................................................................... 26

Life cycle ........................................................................................................ 27

Excretory Secretory Products ......................................................................... 29

Pathology ........................................................................................................ 3 1

Treatment ........................................................................................................ 34

Immune responses ..................................................................... 35

Rationale for this study .................................................................... 40

References ............................................................................................................. 44
CHAPTER 2

INTERLEUKIN 4 (IL-4), IL-6 AND IL-10 CYTOKINE SECRETION FROM
INTESTINAL EPITHELIAL CELLS (IPEC-l) IN RESPONSE TO
CAMPYLOBA C T ER JEJUNI AND TRICHURIS S UIS EXCRETORY

SECRETORY PRODUCTS ............................................................................................ 61
Abstract .................................................................................................................. 62
Introduction ............................................................................................................ 63
Materials and Methods ........................................................................................... 67
Results .................................................................................................................... 78
Discussion .............................................................................................................. 84
Acknowledgements ................................................................................................ 90
References .............................................................................................................. 91
Figures .................................................................................................................... 95

CHAPTER 3

IL-4, IL-6 AND IL-10 CYTOKINE RESPONSES IN SWINE INFECTED WITH

CAMPYLOBA CTER JEJUNI AND T RICH URIS SUIS. ............................................ 115
Abstract ................................................................................................................ 1 16
Introduction .......................................................................................................... l 18
Materials and Methods ......................................................................................... 121

vi

Results .................................................................................................................. 126

Discussion ............................................................................................................ 130
Acknowledgements .............................................................................................. 134
References ............................................................................................................ 135
Tables .................................................................................................................. 138
Figures .................................................................................................................. 142

CHAPTER 4

RECOMBINANT IL-4 (rIL-4) ENHANCES CAMPYLOBACT ER JEJUNI

INVASION OF PORCINE INTESTINAL EPITHELIAL CELLS (IPEC- 1) ........ 148
Abstract ................................................................................................................ 149
Introduction .......................................................................................................... 151
Materials and Methods ......................................................................................... 153
Results .................................................................................................................. 158
Discussion ............................................................................................................ 161
Acknowledgements .............................................................................................. 166
References ............................................................................................................ 167
Tables ................................................................................................................... 170
Figures .................................................................................................................. 171

CHAPTER 5

SUMMARY AND CONCLUSIONS ............................................................................ 182
References ............................................................................................................ 202

vii

Table 3. 1

Table 3. 2

Table 3. 3

Table 4. 1

LIST OF TABLES

Inoculations for the long-term experiments (A) and short-term
experiments (B) ............................................................ 138

Percentage of pigs showing an increase in a given cytokine by day
22—long-term experiment ................................................ 140

Percentage of pigs showing an increase in a given cytokine by day
24 or 48 hr—short-term experiment ..................................... 141

Translocation of C. jejuni to the baso-lateral chamber ............... 170

viii

Figure 2.

Figure 2.
Figure 2.
Figure 2.

Figure 2.

Figure 2.

Figure 2.

Figure 2.

Figure 2.

Figure 2.

Figure 2.

Figure 2.

Figure 2.

Figure 2.

Figure 2.
Figure 2.

Figure 2.

13

14

15

16

17

LIST OF FIGURES

Scanning electron microscopy of 12-day-old IPEC-l cells on

transwells .................................................................... 95
Standard curve for swine IL-6 ........................................... 96
Standard curve for swine IL-lO .......................................... 97
Standard curve for swine IL-4 ............................................ 99

Effect of different multiplicity of infection (MOI) of C. jejuni on [L-
6 secretion from lPEC-l cells ........................................... 100

IL-6 secretion from differentiated IPEC-l cells in response to apical
(A) and baso-lateral (B) infection with C. jejuni ..................... 101

IL-6 secretion from undifferentiated [PEC-l cells in response to C.
jejuni infection ............................................................ 103

IL-lO secretion from differentiated IPEC-l cells in response to apical
(A) and baso-lateral (B) infection with C. jejuni ..................... 104

IL-10 secretion from undifferentiated IPEC-l cells in response to C.
jejuni infection ............................................................ 106

IL-6 secretion from differentiated IPEC-l cells in response to T. suis

Excretory Secretory Products (ESP) .................................... 107
IL-6 secretion fi'om undifferentiated IPEC-l cells in response to T.
suis ESP .................................................................... 108
IL-lO secretion from differentiated IPEC-l cells in response to T.
suis ESP .................................................................... 109
IL-lO secretion fiom undifferentiated IPEC-l cells in response to T.
suis ESP .................................................................... 110
RT-PCR for IL-6 mRNA at 0 hr in response to C. jejuni infection of
[PEC-l cells ............................................................... 11 1
Effect of cycloheximide on IL-6 secretion at 0 hr .................... 112
Viability assays with cycloheximide ................................... 113
Western-blot analysis for pre-formed IL-6 protein ................... 114

Figure 3.

Figure 3.

Figure 3.

Figure 3.

Figure 3.

Figure 3.

Figure 4.

Figure 4.

Figure 4.
Figure 4.

Figure 4.

Figure 4.

Figure 4.

Figure 4.

Figure 4. 9

In vivo adaptive responses fi'om swine infected with C. jejuni and/ or
T. suis; fold-changes in IL-4 over day 0—long-term
experiments ................................................................ 142

In vivo adaptive responses from swine infected with C. jejuni and/ or
T. suis; fold-changes in IL-6 over day O—long-term
experiments ................................................................ 143

In vivo adaptive responses from swine infected with C. jejuni and] or
T. suis; fold-changes in IL-10 over day 0—long-term
experiments ................................................................ 144

In vivo responses fiom swine infected with C. jejuni and/ or T. suis;
fold-changes in IL-4 over 0 hr—short-term experiments ............. 145

In vivo responses from swine infected with C. jejuni and! or T. suis;
fold-changes in IL-6 over 0 hr—short-term experiments ............. 146

In vivo responses from swine infected with C. jejuni and/ or T. suis;
fold-changes in IL-lO over 0 hr—short-term experiments. . . . . . . . 147

Effect of exogenous IL-4 and IL-10 on C. jejuni invasion of IPEC-l

cells ......................................................................... 171
Effect of anti-IL—4 antibody on C. jejuni invasion of IPEC-l

cells ......................................................................... 172
Dose response curve of IL-4 on C. jejuni invasion .................. 173
Effect of IL-4 on E. coli DH50t invasion .............................. 174

Transepithelial resistance across the IPEC-l monolayer in response
to addition of exogenous IL-4 ........................................... 175

Light microscopy of IPEC-l cells following IL-4
or medium treatment ...................................................... 176

Cell-associated bacteria after pretreatment of IPEC-l with IL-
4 ............................................................................. 178

Transmission electron micrographs of IPEC-l cells following
gentamicin killing assays- A) after IL-4 pretreatment, B) after
medium pretreatment ..................................................... 179

Transmission electron micrographs of C. jejuni 33292 in IPEC-l
cells after IL-4 pretreatment ............................................. 180

Figure 4. 10 MTT assays on the viability of IPEC-l ................................ 181

Figure 5. 1 Potential role(s) of individual cytokines during dual-infection in
vivo ......................................................................... 201

xi

CHAPTER 1

LITERATURE REVIEW

I - CAMPYLOBA CTER

A) Microbiology
i) General Characteristics

Campylobacters are microaerophilic, gram-negative, non-spore forming spiral
bacteria with a unipolar or bipolar flagellum. Bacterial cells range from 0.5- 5.0 pm in
length and 0.2- 0.8 pm in width (122). There are 16 distinct species of Campylobacters
known. They include C. jejuni, C. coli, C. concisus, C. curvus, C. fetus, C. gracilis, C.
helveticus, C. hominis, C. hyointestinalis, C. lanienae, C. lari, C. mucosalis, C. rectus, C.
showae, C. sputorum and C. upsaliensis (54, 175). Most species of Campylobacter are
motile exhibiting corkscrew-like motion and darting motility. However, C. gracilis is
non—motile (169). The spiral structure is maintained through the exponential phase of
growth, and most cells become coccoid after 24 hr of growth in liquid culture. These
coccoid structures are deemed viable but non-culturable. However, some researchers
consider the coccoid form degenerate. Campylobacters are non-saccharolytic, i.e, they
cannot utilize carbohydrates as a source of energy. Instead, amino acids and tricarboxylic
acid cycle intermediates are used. On biochemical testing, most species are negative for
hippurate hydrolysis except C. jejuni subsp. jejuni and C. jejuni subsp. doylei. Also, most
can reduce nitrate and are oxidase positive (130, 131). Campylobacter genomes are
relatively small compared to many bacteria. The genome sizes of C. jejuni and C. coli are
around 1.6 — 1.7 Mb, while those of C. fetus subsp fetus and C. fetus subsp venerealz's are
1.1 and 1.3 Mb respectively (52, 155). The chromosome is circular, and the GC content
of Campylobacter DNA varies between 28 to 39% (51). Of all the Campylobacter

species, only the C. jejuni genome has been completely sequenced. There are 1654

predicted coding sequences in C. jejuni genome, and ~ 94.3% of the genome codes for
proteins, making it the most densely packed bacterial genome so far (135).
ii) Isolation of Campylobacter

Campylobacters are normally isolated from fecal samples, environment, food
samples, meat and tissue sections. Cary—Blair medium is commonly used as transport
medium for most species of Campylobacter. Once in the laboratory, they are routinely
cultured in enrichment broth cultures such as Preston broth, Campythio and
Campylobacter enrichment broth and others. Most strains grow optimally between 30 —
37° C, although isolates from some animals grow optimally at 42 °C. Typically, a gas
mixture of 5 — 10 % Oz, 5 — 10% C02, and 85% N2 is used. Some species like C.
sputorum, C. concisus, C. mucosalis, C. curvus, C. rectus and C. hyointestinalis might
additionally require hydrogen (3%) for optimal growth (122). Most strains of C. jejuni
and C. coli are inherently resistant to vancomycin, trimethoprim, penicillins, rifarnpin and
cephalosporins. Hence, some media like cefoperazone deoxycholate agar (CCDA),
charcoal-based selective medium (CSM), blood-containing Campy-CVA medium and
Skirrow medium might incorporate one or more of these antibiotics for selective growth
of these Campylobacters (75, 81, 122). For isolation from fecal samples, cefoperazone-
containing media have been suggested. However, this selective medium does not allow
growth of all species of Campylobacter. Hence, a newer method, dubbed “the Cape Town
protocol” has been developed, wherein stool samples are filtered through a membrane
onto antibiotic free blood plates and subsequently incubated in an Hz-containing

atmosphere (122).

iii - a) Identification

Campylobacters can be directly identified in stool samples by microscopic
examination and by using the Gram stain. Colony morphology (yellowish-gray,
glistening, round colonies) is also used to identify these organisms in a mixed culture.
Biochemical assays based on Campylobacter-specific traits like oxidase activity are
routinely used when pure cultures are available. PCR-based assays are more commonly
used in research laboratories for identification of this bacterium. Some of the genes used
for analysis include the fla gene, 23S rRNA gene and 16S rRNA gene. Both the
biochemical assays and PCR assays can distinguish Campylobacter at the genus and
species levels (24, 74, 121). Immunological assays for identification of C. jejuni and C.
coli are also commercially available (71).
iii — b) Subtyping

Campylobacters exhibit a high degree of genetic strain-to-strain variation even
within individual species. Subtyping of the species to identify individual strains is
critical, particularly during endemic outbreaks or even during epidemics, for
epidemiological investigations. A variety of assays are available to determine individual
type strains. Pulsed-field gel electrophoresis (PF GE) is commonly used to identify
different isolates of C. jejuni and C. coli. The assay uses specific restriction enzymes for
specific species with subsequent analysis of the pattern profile for each strain (25). PFGE
has also proved usefiil in typing strains of other species such as C. lari, C. fetus, C.
upsaliensis and others (22, 125). fla typing is yet another common genotypic method
employed to distinguish strains. After amplification of the flaA gene, the PCR product is

subjected to restriction enzyme digestion with subsequent analysis of the fragment

profiles. This assay is different from PFGE in that restriction analysis is performed using
a specific gene product rather than the whole genomic DNA. Also, fla typing is typically
used for C. jejuni and C. coli strains and not for other species. Yet another genotypic
method used to type strains is random amplification of polymorphic DNA (RAPD). This
method uses multiple random primers to generate fragments from the entire genome,
which are then subjected to agarose gel electrophoresis to visualize patterns. It is useful
in typing strains of C. jejuni, C. coli and C. lari (109). Apart from these techniques, two
recently introduced genotypic methods include amplified fi'agment length polymorphism
(AF LP) and multilocus sequence typing (MLST) (157, 161). In addition to these
genotypic methods, commonly used phenotypic methods include serotyping and phage
typing. For most epidemiological studies, a combination of methods is normally used to
allow for reliable identification of species and strains.
Genetic variations in Campylobacter

Subtyping of Campylobacter has revealed tremendous genetic diversity among
Campylobacter species. In one study, 168 ribotyping showed 77 ribotypes among 261 C.
jejuni isolates (45). Employment of PF GE in another study revealed 25 different banding
patterns among 80 isolates (132), while the use of restriction fragment length
polymorphism (RF LP) in another study showed 89 different polymorphism patterns in
120 C. jejuni isolates (96). Multiple mechanisms exist for generation of such genetic
diversity among bacterial populations, and include horizontal gene transfer within or
between bacteria (recombinations), point mutations, deletions, rearrangements,
duplications and inversions. Campylobacter strains have been shown to employ almost

all these mechanisms to generate diversity at multiple loci, although the role of

recombination is thought to be 50 times greater than that of mutation (157). Horizontal
gene transfer between heterologous and homologous strains of C. jejuni was
demonstrated both in vitra in liquid cultures and in viva in chickens through the
acquisition of antibiotic resistance markers (21, 180). In the in viva study, PFGE analysis
of the recombined strains also showed duplications of distinct regions as determined by
gain of a large DNA fragment, in addition to the markers (21). A single deletion frame
shift mutation in the polymeric T tract of the flhA gene, which controls the expression of
flaA and flaB, was shown to confer functional diversity in a C. coli strain. The resulting
mutants were non-motile, and motility was restored by reversible insertion of the thymine
residue (134). Sequence analysis of the lipooligosaccharide (LOS) loci from 11 C. jejuni
strains showed that gene rearrangements, deletions and single mutations played a role in
generation of genetic diversity among these strains (56). In accordance with these results,
whole genome comparisons using microarrays have shown that the most divergent genes
are the loci that code for lipooligosaccharide and flagellar synthesis, restriction-
modification systems, integrases, iron-acquisition, sialylation, sugar transport and
metabolism (34, 101, 145). Thus, Campylobacters as a group have been shown to be one
of the most divergent bacteria to date. This ability to generate genetic diversity confers
this organism a distinct advantage in survival in the host and environment as well as in

pathogenicity.

B) Epidemiology
Of the 16 different species of bacteria that comprise the genus Campylobacter

(family Campylobacteraceae), two species, C. jejuni subsp. jejuni and C. coli, are most

commonly associated with gastrointestinal infections in humans (84). Of the two, C.
jejuni infections are more prevalent (80 — 90% compared to 5 — 10 % by C. coli) and are
considered one of the leading causes of bacterial enteritis in the world, along with
Salmonella and Shigella (63, 84). It is estimated that around 2.4 million people are
affected every year in the United States alone (48). The incidence of Campylobacter
infections in the developed world is approximately 25—50/100,000 cases of diarrhea
reported for all persons and 300/ 100,000 cases for children between 1 and 4 years of age.
Campylobacter infections also show a sex-specific incidence pattern, with males being
affected more frequently than females. There is also a marked seasonal variation in
Campylobacter incidence, with infections peaking in the summer months. However,
while this is true for most of the developed world, there are marked differences in
incident rates and prevalence of Campylobacter infections in the developing world. The
incidence rate among children less than 4 years of age is much higher, ~ 40,000/ 100,000
cases reported. There is also no well-documented seasonal variation in infection rates in
the developing world. Additionally, the average number of infections /person/ lifetime in
the developed countries is 0—1 , while that in the developing countries is > 5.
Campylobacters also cause a different spectrum of gastrointestinal infections in
these two worlds. In the developed world, inflammatory diarrhea is more commonly
associated with C. jejuni infections, with symptoms including fever, vomiting, bile-filled
discharge, headaches, abdominal cramps and bloody diarrhea. On the other hand,
infections in the developing world are much milder, with profuse watery or secretory
diarrhea. The differences have been attributed to constant re-infection among people of

the developing world, which produces partial immunity (48, 127).

Campylobacters are essentially a food-home pathogen and are acquired through
contaminated raw milk, undercooked poultry, and other foods. Occasionally,
campylobacters may be spread through the community water supply (48). The infective
dose of C. jejuni is as low as 400-500 organisms, as determined from experimental
human infections (20). However, this dose might be hard to estimate in the developing
world, given the large rate of asymptomatic infections (127). C. jejuni infections usually
resolve within 1-3 weeks. However, other rare complications can accompany enteritis,
including appendicitis, neonatal meningitis, hepatitis, pancreatitis, peritonitis,
myocarditis, toxic megacolon, and others (162). Some late-onset syndromes include
reactive arthritis and a neuropathological disorder termed Guillain-Barré syndrome.
Mortality due to Campylobacter infection is estimated to be between 50 and 150 per year
in the United States (48). C. jejuni is also an opportunistic pathogen and is often isolated
from patients with Acquired Immunodeficiency Syndrome (AIDS),
hypogammaglobulinemia, and other immunocompromising conditions (82, 163).
Campylobacter infections are usually treated effectively with antibiotic therapy. The
antibiotics of choice include erythromycin and fluoroquinolones like ciprofloxacin.
Tetracyclines are normally contraindicated due to the rapid emergence of resistance to
this antibiotic (103).

In summary, Campylobacter infections are prevalent throughout the world. The
geographical area, immune status of the host, seasonal variations, and gender of the
individual can affect the rate and incidence of infections. Also, some strains are more
pathogenic than others, and the factors that affect pathogenesis will be discussed below.

While infections are contained effectively in most cases, the incidence is on the rise.

Also, other species of Campylobacter, such as C. upsaliensis, are poised to be new
human pathogens. With better public health care and public education, Campylobacter

infections can be mitigated significantly throughout the world.

C) Pathogenesis

C. jejuni is the best studied of all species of campylobacters, though its
pathogenic mechanisms are not well understood compared to those of other enteric
pathogens. The strain-to-strain variation, the source of isolates, and the variation in
eukaryotic cell lines used for analysis make it difficult to interpret results universally.
However, some of the pathogenic mechanisms and host factors that contribute to
pathogenicity have been documented.

Virulence factors
i) Campylobacter toxins

Multiple toxins have been described for C. jejuni, including cytolethal distending
toxin, cytotoxin, hemolysin, hepatotoxin, and porins.
a) Cytolethal distending toxin (CDT)

This is the most studied toxin of Campylobacter. The toxin causes mammalian
cells to be arrested at the G2 phase of the cell cycle, thereby causing the cells to become
distended in appearance (179). The holotoxin is composed of three subunits, Cth, CdtB
and CdtC (97). The enzyme activity is associated with CdtB, and was shown to be a
nuclease causing double-stranded breaks in DNA when microinj ected (62). While CdtB
is thought to be essential for the cytotoxic activity, the roles of Cth and CdtC are

controversial. One study with purified subunits showed that all three subunits are

required for cytotoxic activity on Henle-407 intestinal epithelial cells, while another
study showed that CdtB and CdtC are sufficient to induce similar activity on HeLa cells
(97, 100). All three proteins have leader sequences, suggesting that they are secreted from
the cell; however, this remains unproved, since most of the activity remains in bacterial
membrane fractions (144). A recent study showed that purified Cth and CdtC can bind
eukaryotic cells but that CdtB does not, suggesting that CDT might be an A: B toxin like
cholera toxin. The A subunit of cholera toxin binds eukaryotic cells while the B subunit is
translocated into the cells. In Campylobacter, the A subunit would include both Cth and
CdtC (100). The cdt genes are present in almost all strains of C. jejuni and C. coli, as well
as in the few isolates of C. fetus, C. upsaliensis and C. hyaintestinalis strains tested so far
(40, 41). However, the amount of cytolethal activity differs between isolates; the reason
for these differences remains unclear. Recent studies showed that CDT might play a role
in the induction of IL-8 from INT407 cells. Membrane fractions of mutant C. jejuni strain
81176 defective in any of the three membrane-associated subunits failed to induce IL-8
(66).

The role of CDT in C. jejuni pathogenesis is not clear. A few hypotheses have
been proposed based on what is known in other systems. In Actinabacillus
actinamycetecamitans, an irnmunosuppressive activity of CDT has been identified by its
ability to kill T-cells, and the CDT from Shigella dysenterz'ae has been shown to magnify
diarrheal symptoms in suckling mice in a dose-dependent manner (129, 159). In a very
recent study, C. jejuni strain 81176 CdtB mutant was shown to be defective in
colonization of C5 7BU129 mice 4 months post inoculation, while the wild type strain

was not. Also, in the same study, mice deficient for NF-KB did not develop

10

gastroenteritis when infected with the CdtB mutant suggesting a potential role for this
toxin in pathogenesis (47).
b) Cytotoxin

A 70 KDa cytotoxin has been described that causes HeLa cells and Chinese
hamster ovary cells to round up and eventually die. The toxin is heat-and trypsin-
sensitive and hence is thought to be a protein. It is also thought to be associated with the
bacterial cell membrane (58). Yet another cytotoxic activity has been described recently,
which is heat-stable, but sensitive to proteases. This toxin is fairly pH stable and retains
its activity after storage at —70°C and —20°C for over 2 years. However, these cytotoxic
activities have not been firrther characterized (99).
c) Hepatotoxin

A hepatotoxic acitivity has been identified in a few strains of C. jejuni, which
causes mouse liver lesions and hepatitis. The factor also causes massive infiltration of
mononuclear cells to the site of lesions. However, only two studies have been published
on this activity, and it awaits further characterization (86, 87)
d) Hemolysins

Hemolytic activity has been observed with a few strains of C. jejuni and C. cali.
Two different types of hemolytic activity have been described, one secreted by bacterial
cells and one based on physical contact with erythrocytes (144). This activity is not
generally seen in Campylobacter strains, and one of the reasons suggested is the
incubation temperature used for routine growth. In one study, the strains demonstrated
hemolytic activity at 42°C but not at 37 ° (3). Another study found blood agar plates to be

an unsuitable medium for detection of hemolysins and proposed the use of blood agarose

11

plates instead (170). In a recent study, a strain carrying a mutation in phospholipase A
gene (pldA), which encodes a protein found in the outer membrane of C. coli was shown
to be deficient in hemolysis; PldA was postulated to play a role in cell-mediated
hemolysis of erythrocytes (57).
e) Porins

Only one study has been conducted to date on a transient vacuole-fanning
cytotoxic activity of porins from C. jejuni, which causes rounding and eventual death of
HEp-2 cells. The activity is heat labile and trypsin resistant and was found to be
associated with lipopolysacharides (8).
ii) Motility

Motility has long been considered a virulence factor in Campylobacter
pathogenesis, with motile strains being considered more efficient in colonization and
more invasive than non-motile strains. In one study, non-motile strains were cleared from
the intestinal tract of suckling mice, while the parental wild type colonized efficiently
(120). Motility is conferred by a single flagellum or bipolar flagella that enable the
bacteria to swim through the mucus layer; motility is also thought to enable the bacteria
to seek receptor sites for effective adherence. The flagella are composed of two subunits,
FlaA and FlaB, and of the two, FlaA is thought to be essential for motility (59, 178).
Motility is also required for invasion, as non-flagellated, non-motile strains are invasion
deficient (184). Other factors that affect motility include pH and viscosity. While C.
jejuni exhibits normal motility at neutral pH and at a pH of 8.5, it is paralyzed at a pH of
5.0 (168). Also, as the viscosity of the environment increases, the motility increases as

well.

12

iii) Adherence and adhesins

Adherence is an important virulence factor in Campylobacter pathogenesis.
Adherence is required for successful colonization of the host gut, and prevents bacterial
cells fi'om being washed away by intestinal fluids and peristalsis. Also, prior to C. jejuni
invasion, the bacteria bind transiently to the host cells.

Many factors affect adherence. The fibronectin binding protein, CadF, and cell
binding factor-1 (CBFl or PEBl) are two well-known adhesins. Other putative adhesins
include the flagella, pili, outer membrane proteins, and lipopolysaccharides (LPS).

a) CadF and PEBl

CadF is a 37-KDa outer membrane protein present in all strains of C. jejuni and
C. coli; it binds the extracellular matrix fibronectin present on host cells (90). Targeted
deletion mutants of CadF showed a 100% reduction in colonization of newly hatched
Leghorn chickens (185). PEBl is a 28-KDa surface-exposed protein that is required for
adherence and colonization (83). C. jejuni with mutation in the pebl locus showed a 50 to
100-fold reduction in adherence of Campylobacter to HeLa cells and significant
inhibition in intestinal colonization of BALB/c mice (142). In addition, PEBl is thought
to be required for amino acid transport in C. jejuni (141) and to be an antigenic
determinant for production of serum antibodies (143).

b) Flagella and Pill

Apart from their role in motility, flagella also are considered putative adhesins.

Non-flagellated mutants have been shown to adhere less efficiently than the

corresponding flagellated wild type strains (72). Also, purified flagellar proteins have

13

been shown to bind INT407 cells (117). In another study, monoclonal antibodies raised
against the flagella inhibited adherence and colonization of suckling mice (172).

The role of pili in intestinal colonization or virulence is not clear. In one study, pilus-like
structures were reported to be induced in the presence of bile salts, and the corresponding
pilus-deficient mutants caused significantly decreased disease symptoms in a ferret
animal model (33). However, it was later shown that this pilus-like structure was in fact
an artifact (55).

c) OMPs

Multiple outer membrane proteins are present in Campylobacter, however, only a
few have been shown to play a role in adherence. In addition to PEBl and CadF, two
other OMPs of 26 and 30-KDa that bind HeLa cells are also thought to play a role in
attachment of Campylobacter to host cells.

(I) Lipopolysaccharides (LPS)

LPS are considered major antigenic determinants of most Gram- negative
bacteria. Aside from their role in the induction of serum antibodies, LPS/LOS of
campylobacters are also thought to play a role in adherence. In the only published study
of LPS, purified C. jejuni LPS was shown bind INT407 cells as well as intestinal mucus
(1 l 7).

LPS/L08 biosynthesis

LPS is a major outer surface component of Gram-negative bacteria, and is made
of three regions, the Lipid A molecule, the core region (inner and outer) and the O-
chains. The Lipid A molecule comprises the endotoxin part of the LPS and is attached to

the inner core through ketodeoxyoctulosonic acid or Kdo, which is specific for the inner

14

core. The outer core region is attached to the O-chains, which are made of 10 to 30 sugar
subunits comprised of one to five sugars. Lipooligosaccharides (LOS) unlike the LPS,
lack the O-chains (49). LPS biosynthesis is a complex process. The different regions of
the LPS are assembled in the inner membrane and then transported (or flipped) to the
outer membrane. The Lipid A moiety is synthesized initially, from N- acetylglucosamine
(GlcNAc) precursor. Kdo is added directly to the Lipid A molecule, followed by the
sugars that comprise the inner core. The outer core is attached next, subsequently
followed by the O-chains, and the entire molecule is transported to the outer membrane
of the bacterium. In Campylobacter, the outer core is thought to be assembled on a lipid
carrier, and then transferred to the inner core. A number of genes have been implicated in
LPS biosynthesis of Campylobacter. These include the waaC gene product, a
heptosyltransferase, that functions in the attachment of a heptose to Kdo; the wla gene
cluster, predominantly consisting of glycosyl and galactosyltransferases, and the galE
gene product, a UDP-glucose-4-epirnerse, that function in the synthesis of the core
molecules (88, 95, 102, 181). Campylobacter strains exhibit variability in the expression
of the O-chains. Some are LOS producers, while others produce LPS (49). However, the
attachment of the O-chains to the core region is controversial. Some C. jejuni strains have
been found to produce polysaccharides that are not attached to typical lipid A- core
molecules (6, 146). In these cases, it is thought that the O-chains may be attached to a
lipid molecule rather than to the core structure. However, this remains unresolved.

The LPS/LOS of Campylobacter also has been implicated in the pathogenesis of
Guillain Barré Syndrome (GBS). The core structures of Campylobacters contain sialic

acid residues, and in multiple serotypes these structures mimic human gangliosides,

15

particularly GM1, GM;, GD3, GDIa and GO“, (60). Gangliosides are glycosphingolipid
molecules found in nerve ganglia, and ganglioside mimicry in Campylobacter core
structure results in auto-immune antibodies directed against the gangliosides present in
peripheral nerves, resulting in neurological disease. In a recent study, the cgtA gene, that
encodes a putative N-acetylgalactosaminyl transferase involved in core formation, was
shown to undergo slip-strand mutagenesis, resulting in a GM; to GM; mimicry (61). The
cgtA gene contains a variable, homopolymeric G tract, and was shown to undergo slip-
strand mismatch recombination in this region, resulting in a differential number of G
nucleotides in the coding region. This was subsequently shown to transform the
ganglioside structure from GM; to GM3. Slip-strand recombination in the wlaN gene, also
involved in core structure synthesis, has been shown to occur as well. Recombination in
this gene affected the LOS core structure, and converted the surface coat from GM; to a
GM1 ganglioside (61, 105).
e) Host factors required for adherence

The knowledge of host factors that mediate adherence of C. jejuni is limited. One
of the well-known mediators mentioned previously is the extracellular matrix protein
fibronectin. Intestinal mucus also enhances adherence (30). In one study, the sugars
mannose, maltose, fucose, glucose, N—acetylglucosamine and galactose did not
significantly inhibit binding of campylobacters to Chinese Hamster ovary cells (CHO),
while lipids partly inhibited binding (167). However, in another study, purified LPS was
shown to be fucose-sensitive in binding to INT407 cells (117). Besides these putative

adhesins, host or-integn'n receptors (1461 and aSBl also have been implicated (92).

16

iv) Invasion

Invasion of host cells is considered as one of the most important virulent factors
for C. jejuni pathogenesis. Invasion has been associated with many models of C. jejuni
enteritis, including infant chickens, gnotobiotic piglets, chicken embryos, infant mice,
and infant monkeys; and most importantly, in patients with C. jejuni colitis (72). Clinical
strains have been shown to be more invasive in HeLa and Hep-2 cells than environmental
and non-clinical isolates (91 , 126). Also, isolates from patients with inflammatory
diarrhea are more invasive than the ones from non-inflarnmatory diarrhea. Invasion by C. '
jejuni occurs at low to moderate levels, between 0.01 to 5% of the inoculum (72), and
internalization occurs preferentially at the basolateral surface of epithelial cells, as was
recently demonstrated with T84 cells (119). Invasion requires de nova synthesis of ~ 14
bacterial proteins upon contact with host cells, but not synthesis of host proteins (72).
One of these proteins, a campylobacter invasion antigen B (Cia B) is translocated into
INT407 cells during invasion. CiaB is a 73 KDa protein with sequence similarity to other
secreted proteins such as YopB of Yersinia, SipB of Salmonella, and IpaB of Shigella.
Like these type III secreted proteins, CiaB does not have a signal sequence at its amino
terminus. Mutation in CiaB results in a 100-fold reduction in invasion and blocks
secretion of at least eight proteins, suggesting that CiaB is not only secreted itself, but is
also required for the secretion process. However, efforts to identify a Type III secretion
system in C. jejuni have not been successful so far. Analysis of the complete C. jejuni
chromosomal sequence (by homology) did not reveal any typical Type III secretion
systems other than the flagellar apparatus (135). In addition, Type III secretion genes in

other pathogenic organisms are usually organized as operons (93). Physical map

17

constructs of the coding regions adjacent to the CiaB gene have been shown to be
monocistronic. Also, the G + C content of the CiaB gene is not dissimilar to the C. jejuni
chromosome, suggesting that the gene is not on a pathogenicity island like other type III

secretion genes identified in Salmonella and Shigella (93).

a) Factors affecting internalization

The factors that affect adherence also affect invasion. Mutations in cadF and pebl
also result in reduced invasion of host cells. Motility is essential, since paralyzed but
flagellated mutants (pflA) exhibit impaired invasion of INT407 cells (184). However,
flagella have been shown to be required for invasion as well. Non-flagellated mutants of
C. jejuni CF84-340 show a 3-fold decrease in invasion compared to the parental wild
type (72). A study by Wassennaar et al., showed that FlaA is required for invasion but
not FlaB (177). Since Fla A is required for motility, and motility in turn is essential for
invasion, the FlaA effect might be through motility. A recent study by Konkel et al (94)
shows that flagella act as the export apparatus for the Cia proteins. They concluded that at
least one of the flagellar filament proteins (Fla A or B) is required for Cia protein
secretion. Taken together, these data suggest that motility is required for bacteria to
contact host cells, while flagellar proteins are involved in adhesion (along with other
factors) and in invasion through the secretion of the Cia proteins. In a recent study,
campylobacter glycosylated surface proteins have shown to be an additional factor
required for adherence and invasion. Inhibition of glycosylation by mutation in the ngH
gene resulted in significant inhibition of adherence to and invasion of CacoZ cells, as well

as chickens in viva (80).

18

b) Other factors
1) Growth factors

Temperature seems critical for optimal invasion. C. jejuni adherence (and
therefore invasion) is optimal at 37° C for human epithelial cells and suboptimal at 30
and 42° C. However, optimal temperatures might vary for cell lines of non-human origin.
Also, mid-log phase bacteria invade more efficiently than early stationary phase bacteria
(89). Growth of bacteria in the presence of mucin or bile salts (deoxycholate) also
enhances invasion (30). In addition, iron depletion increases the invasiveness of C. jejuni
and C. coli in vitro (158).
2) The Vir plasmid

A virulence plasmid identified in C. jejuni 81176 is thought to play a role in
virulence (7). This plasmid of approximately 35 Kbp contains 4 open reading frames that
are thought to encode components of a type IV secretion system. Mutation in 2 of these
genes, camB3 and virBI 1, causes three-fold and 11-fold reduction in C. jejuni invasion of
INT407 cells, respectively, as well as a 3-fold and 6-fold reduction in adherence,
respectively. A limited role in clinical significance was revealed by the identification of
virBI I in 6 of 58 clinical isolates.
3) Host factors

The host factors that are required for invasion are beginning to be explored in
detail. It is known that invasion is cell-line dependent. Human cell lines are much more
susceptible to invasion by C. jejuni or C. coli than are non-human cell lines. Also,

different C. jejuni strains do not have similar invasion potentials for multiple cell lines. In

19

addition, immature semiconfluent cell layers are more susceptible to invasion than are
confluent cell layers (72).

C. jejuni invasion is an energy-dependent process. Chemicals such as iodoacetate
or dinitrophenol, which inhibit glycolysis and the Kreb’s cycle respectively, also inhibit
invasion (29). The optimal multiplicity of infection (MOI) of C. jejuni for infection of
eukaryotic cells has been deduced, at least in one study. In this study, the number of
bacteria that invaded INT407 cells increased with increasing MOI, and reached maximal
numbers at around 200. However, the maximal invasion efficiency (percentage of
inoculum that was internalized) was achieved at an MOI of 0.02, suggesting that C. jejuni
can be efficient as a solitary invader (73). Since there is considerable laboratory variation
in assays and cell lines, whether these results are universally applicable remains to be
seen.

C. jejuni invasion is thought to be host-cell-cycle dependent, as peaks of invasion
occur at 2- 3 hr and after 7 hr of initial contact with the host cells. C. jejuni has been
localized near the junctional space of the host cells and inside vacuoles when internalized
or intracellular (73). The average number of internalized bacteria within each cell is
thought to be 2, however, as mentioned before, this might not be universally true.
Bacteria within the host cells are well separated in space, suggesting that that the invasion
event for each bacterium is independent of the other.

The host receptors that are required for invasion are not well understood.
Inhibition of coated pits and oaveolae also inhibit invasion, suggesting these host
membrane regions are involved- but no specific receptors have been found (128, 182).

Host microtubules (MT), microfilarnents (MF), or both are needed for the invasion

20

process, and the requirement for specific cyto-skeletal structure is strain dependent. C.
jejuni strain 81176 requires microtubules for internalization. Dyenin, which is associated
with MT, might be required for motor activity to transport endosomes containing C.
jejuni to perinuclear regions or across the cell (92). Other required host factors include
signal transduction proteins such as tyrosine kinases, phosphatidylionositol-3 -kinase
(PB-kinase) and calcium (19, 84).

Summary of invasion process

C. jejuni strains adhere to host cells through adhesins like CadF and PEB 1.

Strains that are adherent are not necessarily invasive, suggesting that adherence and
invasion are independent processes. Host membrane oaveolae and coated pits are
involved in invasion, although their roles are unclear. Binding of C. jejuni to the host cell
presumably causes aggregation of receptors, and C. jejuni secretes proteins, including
CiaB, into the host cell with the help of the flagellar apparatus, triggering signal
transduction events wherein tyrosine kinases phosphorylate host proteins. One of the
proteins that are phosphorylated is PI3-kinase; this phosphorylation causes the release of
calcium ions, leading to the recruitment of cytoskeletal structures such as microtubules,
rnicrofilaments or both, resulting in C. jejuni uptake into endosomes. For processes
involving microtubules, a dyenin motor is thought to power the endosomal trafficking
through the cell.

c) Intracellular survival

C. jejuni has been shown to survive inside HEp-2 cells and INT-407 cells for up

to 9 hr and 48 hr, respectively (29, 89). Biopsy specimens fiom infected macaques and

chickens have also been shown to contain Campylobacter (151). C. jejuni has been

21

shown to survive for up to 7 days in both human and mouse mononuclear phagocytes
(85). Intracellular survival enables Campylobacter to replicate and to subsequently
disseminate to other tissues. Two genes have been implicated in intracellular survival: the
sadB gene and the katA gene. SodB, superoxide dismutase, catalyzes the conversion of
superoxide to hydrogen peroxide and 0;, while KatA, catalase, breaks down hydrogen
peroxide to water and 0;. These two genes have been shown to protect C. jejuni against
the oxidative radicals inside the host cell (92).

A role of iron in intracellular survival of C. jejuni has been suggested.
Campylobacter possesses an enterochelin uptake system, can bind exogeneous
siderophores, and can synthesize bacterial ferritin. A ferritin-deficient mutant was shown
to be more sensitive to hydrogen peroxide than the parental wild type and had retarded
growth under iron deficient conditions, suggesting that ferritin plays a role in iron storage
and in the oxidative stress response (176). Recently, a novel iron binding protein, Dps,
that provides protection against hydrogen peroxide stress, has been identified. Dps is
constitutively expressed and binds free ferrous iron to prevent generation of oxygen
radicals. Mutation in this gene caused sensitivity to H;O; in vitro (76). In another study, a
novel 27-KDa oxidation sensitive protein was shown to degrade gradually in the presence
of hydrogen peroxide, with a concomitant decrease in viability (183). This protein shows
sequence similarity to non-heme iron proteins rubredoxin oxidoreductase and ruberythrin.
Also, in a recent study, after incubation with eukaryotic cells, a de nava synthesized 80
KDa protein in C. jejuni was identified. This gene is analogous to the flruA gene of E. coli

and is thought to be involved in iron acquisition - since its expression is repressed in the

22

presence of iron (53). However, a role for these iron regulation systems in intracellular
survival has not been demonstrated.
(1) Translocation

C. jejuni exhibits multiple “invasion types” for translocation. Some strains
translocate through transcellular pathways, i.e., through the host cell cytoplasm,
presumably in endosomal vacuoles. Other strains translocate paracellularly, i.e., between
the cell junctions, without disrupting the transepithelial resistance. Some strains exhibit
both types of translocation processes (92). A role for translocation as a virulence factor
and in pathogenesis has been suggested but has not been studied extensively. In one study
86% of clinical isolates from patients with colitis translocated across Caco-2 cell
monolayers. On the other hand, only 48% of the clinical isolates from patients with non-
inflarnmatory diarrhea exhibited a similar phenotype (39). Bacterial translocation through
the host cell is also thought to be essential for dissemination to other tissues.
Translocation through epithelial cells to underlying cell types like macrophages might
spread the organism to extraintestinal locations such as the liver or gall bladder, causing

hepatitis or cholecystitis respectively.

D) Immune responses

The role of the immune response in C. jejuni infections is known to some extent.
Two studies have been published within the last year an innate cytokine responses to C.
jejuni infection. However, most of the studies published so far have been limited to
identification of serum antibodies. Three main classes of antibodies are usually induced

in response to Campylobacter infection, including IgA, IgG, and IgM. The major surface

23

antigens that have been implicated in the induction of these serum antibodies include
flagella, LPS, and PEBl. Antibodies provide partial immunity to C. jejuni infection, since
application of anti-C. jejuni maternal antibodies been shown to delay colonization of
newly hatched chickens (154).

C. jejuni has been shown to induce IL-8 secretion from human intestinal INT407
cells, and live bacteria are required for this process (65). C. jejuni has also been shown to
induce IL-lot, IL-1 13, TNF-or, IL-6, and IL-8 secretion fiom THP-l, a human monocytic
cell line (79). Recently, INT407 cells were also shown to be immunopositive for
intracellular IL-4, IL-10, IFN-y and TNF-or in response to C. jejuni infection (2). In a
mouse model, a pathogenic strain of C. jejuni has been shown to induce TNF-or but not
IL-1 [3 production in mouse peritoneal cells (133). On the other hand, patients with C.
jejuni enteritis were shown to produce significant IL-10 in their stools along with
significant luminal nitric oxide (38). Thus, specific cytokine responses induced by C.
jejuni seems to be host species dependent.

Studies on the role of cell-mediated immunity in C. jejuni infections are limited.
Lymphocytes are recruited along with polymorphonuclear leukocytes during infection,
presumably due to the production of chemokines. Recently, clinical isolates of C. jejuni
were shown to increase the production of y5T cells in vitro (174). However, C. jejuni has
also been shown to induce apoptosis of chicken lymphocytes and of human Jurkat cell
(T-cell) line in vitro. In addition, it has also been shown to cause apoptosis of a mouse
macrophage cell line J 774A.1, and a human monocyte cell line THP-l. In all these cases

of apoptosis induction, the CiaB protein has been implicated, as CiaB mutants fail to

24

induce apoptosis of these cells. It has been postulated that apoptosis of immune cells is a

potential immune evasion mechanism for C. jejuni (92).

25

II — TRICHURIS SUIS

The genus T richuris (family T richuridae) consists of eight species of whipworms
whose hosts are distributed widely among the mammals. T. trichuira is a parasite of
primates; T. vulpz's is found in canids (dogs); T. muris, in rats; T. avis, in sheep; T.
discolor, in cattle; T. felis, in cats; and, T. suis, in swine.
A) Morphology

T richuris suis is a whipworrn helminth with a slender anterior end and a thicker
posterior end. The parasites are dioecious, that is, the sexes are separate. Males are
smaller than the females and have a distinct curved spicule on the posterior end. The
adult worms are ~5 centimeters long. They are also cylindrical and bilaterally
symmetrical. The outermost layer is called the cuticle and covers the entire body of the
worm. Beneath the cuticle is the hypodermis or subcuticle, and in the anterior end, the
hypodermis is comprised of columnar elongations called the bacillary band, which is
purported to have glandular activity. The bacillary band protrudes into the cuticle to form
pores. On either side of the bacillary bands are vesicular swellings of the cuticle called
the cephalic glands. Beneath the hypodermis is the muscle layer that runs longitudinally.
Underlying the muscle layer in the anterior section is a large cytoplasmic vacuolar body
called the stichosome that surrounds the narrow eSOphagus. T richuris spp have a
complete alimentary canal with a mouth and an anus. The male reproductory organs are
comprised of the testis, vas deferens, ejaculatory duct, spicule, and spicule muscle; and
the females possess vagina, uterus, oviduct and ovary. The worms also possess a
rudimentary nervous system with dorsal and ventral nerve cards that form part of the

hypodermis. There is no circulatory or respiratory system. Circulation occurs through

26

movement of fluids within the pseudocoel, while respiration occurs through the cuticle.

The life span of T. suis worms is around 4-5 months (77, 137).

B) Life cycle of T. suis

The life cycle of T. suis starts fiom eggs and culminates in adults. Sexual
reproduction results with the females laying ~ 3000-5000 eggs per day. The eggs are
unfertilized and unsegrnented and contain two distinct nuclei. They are barrel shaped
with two polar plugs and are covered by three thick outer layers and an inner thin
Vitelline membrane. The eggs exit the host in the feces and develop into an infective
larval stage within the egg, in the soil. The two nuclei firse and multiply resulting in an
embryo that subsequently forms the infective larva. The process takes about three to
fifteen weeks depending on temperature. The lower the temperature, the longer the
development time. The optimal temperature for the development of the embryo is around
34° C, beyond which the eggs begin to degenerate. The infective larva within the egg is
characterized by a poorly formed esophagus, an oral spear, and a rudimentary intestinal
tract made of granulated material. Infective larvae have been shown to infect pigs even
after storage at 5° C for over a year. In nature, the infective larvae can survive for 2—11
years and still retain infectivity. The infective larvae enter the host through the feeding
habits of the pig, and proceed to the next stage in life cycle, the infective larva within the
host (L1). This transition has been seen to occur only in the pig intestine and not in the
open environment. Once inside the host, the infective larvae within the eggs hatch to
release the L1 larvae, which may take between 9 hours to 3 days. Hatching takes place

predominantly in the cecum and the proximal colon. The L1 larvae have a stylet at the

27

anterior end, esophagus, a cell body, digestive tract, and rectum at the posterior end. By
means of the oral stylet, the L1 larva burrows into the crypt cells of the mucosa and
penetrates epithelial cells and goblet cells. By day 7, the larve are embedded deep in the
tissues, just under the mucosal epithelium. This marks the beginning of the histotrophic
phase in the development of the larvae. This stage can be recovered from the pig intestine
for around 10 days; on day 10 the L1 larva undergoes a second molt to form the L2 larva.
This stage lasts for 3-6 days, during which body organs are formed, including
rudimentary reproductive organs. The old cuticle is shed to expose the new cuticle
underneath. Histotrophic migration out of the mucosa and into the lumen begins by day
13. By day 16, the L2 larva undergoes another molt to form the L3 stage. The length of
the larvae is considerably longer, with further differentiation of the body organs. This
stage lasts up to 4 days, at the end of which the cloacae can be seen in the males. Since
the body length is considerable, the larvae are no longer contained beneath the
epithelium; the entire posterior end protrudes into the lumen of the large intestine. By day
20, the larvae undergo one more molt to form the L4 stage. The reproductive organs are
formed and differentiate into vagina, uterus, oviduct and ovary in the females and the
corresponding organs in the males. By day 28, most of the body of the larva is in the
lumen of the intestine with the anterior end inside a shallow tunnel covered by a thin
layer of mucosal epithelial cells. On day 35, the larvae molt for the final time to form the
adult stage. Complete reproductory organs are seen. The prepatency period of the females
starts around day 41, and eggs are seen shortly after. The proportion of females in the
swine gut is around 0.55-0.66 in natural infections (14, 16, 137). Recently, a phenol

oxidase activity has been identified in female worms (42, 43). The activity causes

28

browning of the female worms when exposed to air. This activity was subsequently
localized to the outer shell of the eggs and is activated when the eggs pass out of the host
digestive tract into an aerobic environment. Since the eggs remain in pastures for a long
time before being ingested, it is thought that this activity might help protect the eggs from

the harsh environment (42, 43).

C) Excretory Secretory Products (ESP) of T. suis

Adult T. suis worms excrete/secrete a variety of products into their environment
that are purported to play a vital role during their life cycle. Some of these products have
been identified and characterized, and include the zinc metalloprotease, a T richuris suis
specific antigen, a thiol protease, a serine protease inhibitor, and a chymotrypsin/elastase
inhibitor. A short description of these proteins and peptides follows. Since only the adult
worms can be cultured in vitro, the types of molecules in the excretions/secretions from
other stages of T. suis are not known.
i) Zinc metalloprotease

This 45 KDa protein degrades fibrinogen and elastin, two molecules that comprise
the extracellular matrix, and is thought to play a role in the burrowing activities of the
larva. The protease activity has a pH optimum of 7. 0, which is the same as the pH of the
tissue spaces in the intestines. The protease is localized to the stichocytes in the anterior
end and is thought to be secreted (67).
ii) T. suis specific antigen

This is a 20 KDa glycoprotein that is specific to T. suis infection in pigs and does

not crossreact with sera from pigs infected with T axoplasma gondii, Ascaris suum, or

29

T richinella spiralis (68). However, it does crossreact with sera from dogs infected with
another T richuris species, T. vulpis. The glycoprotein reacts with sera obtained fi'om pigs
infected with T. suis for 28 days and not with sera obtained early in infection. This
suggests that the glycoprotein is secreted/excreted from a larval stage of T. suis, most
likely L4 (68).
iii) Thiol protease

A thiol protease activity has also been found in the culture fluids of T. suis adults
(69). This activity has a pH range of 5.5 to 8.0 and was isolated from the gut extracts of
adult worms suggesting a role in feeding, nutrient absorption, and digestion. This activity
has not been further characterized (69).
iv) Serine protease inhibitor

This peptide of 6.6 KDa selectively inhibits serine proteases such as trypsin and
chymotrypsin, but has no effect on thrombin or elastases. Since neutrophils and
macrophages have been shown to release molecules with trypsin, and chymotrypsin-like
activities, it has been suggested that this inhibitor might have a role in
immunomodulation (148).
v) Chymotrypsinlelastase inhibitor

This activity is very similar to the serine protease inhibitor activity described
above. It is a peptide of 6.4 KDa that selectively inhibits proteases such as elastase and
chymotrypsin, but has no effect on thrombin or trypsin. This inhibitor shows a 48%
sequence similarity to the trypsin/chymotrypsin inhibitor activity, and is also purported to

play a similar role in immunomodulation (149).

30

vi) An antibacterial activity

An antibacterial activity of molecular weight < 10,000 has been identified (1).
Four different bacterial species were sensitive to this activity, including C. jejuni, C. coli,
Escherichia coli and Staphylococcus aureus. C. jejuni isolates are more susceptible to
this activity than C. coli isolates. The activity is heat stable and resistant to trypsin. Also,
preliminary data suggests that this activity might be bacteriocidal. This activity has not
been further characterized (l ).

In addition to the proteins and peptides from the ESP of T. suis, pore-forming
proteins of 47 KDa and 43 KDa in the ESP of T. trichiura and T. muris, respectively,
have also been identified. These proteins might have a significant role in the invasion of
mammalian intestinal cells, as they induce ion-conducting pores in lipid bilayers in vitro

(35).

D) Pathology

T. suis causes a clinical syndrome in pigs called trichuriasis, characterized by
anemia, anorexia, pronounced weight loss, watery and a bloody discharge that changes to
bloody mucoid diarrhea before death (13). Other clinical signs include cecitis, colitis,
inflammation, and hemorrhagic areas in the colon. Disruption of the mucosal epithelium
attributable to the parasite is seen, as are nodules in the cecum and colon. Other
pathophysiological signs include reduced albumin levels and erythrocyte and eosinophil
losses in the periphery. The symptoms begin 13-14 days post infection, coinciding with
the emergence of histotrophic larvae. T richurz's spp. infections are widespread through

the world. In the US and UK alone, is estimated that around 60—85% of the pigs in swine

31

farms may be infected with T. suis (13, 15). Economic losses to the swine industry are
considerable. The increase in feed-to-gain ratio due to weight loss is estimated to result in
losses of ~ $115 million annually (165). T. suis can infect humans as well. Experimental
infections in man with a single dose of 1000 infective larvae produced patent infection
(Beer RJ), and in another study, the eggs obtained subsequently from a human infection
were shown to be infective (15). Apart from potential cross infection with T. suis, an
estimated 1 billion people through the world are infected with the human T richuris
species T. trichiura, which also causes dysentery (15), (136, 137).

Multiple factors contribute to the development of patent T richuris infections.
Dose of infective larvae and parasite strains can be essential factors. In experimental
infections, clinical manifestations in swine have been seen with various doses ranging
from 25,000—40,000 infective larvae. Some infections do not produce clinical disease
even with 15,000 larvae, while others with moderate doses (2500-3000 (113)) have
resulted in disease. Strain differences might be an important factor, as might be the
method of inoculation (single vs trickle) (137). Host age also plays a role in the
establishment of infection. Weaned pigs are more susceptible to infection than sows
(138). In an experimental infection, sows (4 years of age) had less worm fecundity and
lower fecal egg excretion than weaned pigs (5 weeks of age). The worms were seen only
in the cecum and proximal colon of sows, while they were evenly distributed throughout
the colon in weaned pigs. The authors suggested that resistance to infection in sows
might be due to acquired immunity and nutritional status. In a follow-up study,
nutritional status indeed seemed important, as pigs fed a low protein diet had more T. suis

worms than pigs fed normal protein diet (140). However, a similar result was not seen in

32

another study; one of the reasons could be host and /or parasite strain difference(s) (78).
Another factor that affects pathophysiology is iron deficiency. In an experimental
infection, lO-week-old pigs fed a low iron diet developed more severe infections than
pigs fed a normal diet (139). The iron-deficient pigs also exhibited higher total fecal egg
count and warm fecundity compared to pigs fed a regular diet.

One additional important factor that contributes to pathology is secondary
bacterial infection. In his pioneering study on the life cycle of T. suis, Beer suggested that
the burrowing activities of T. suis into the mucosa might enable secondary bacterial
infections to occur, which in turn might enhance pathology (16, 152). In a study
conducted in 1996, 8-week-old weaned pigs were given a single dose of 2500
embryonated eggs with or without antibiotic treatments. Only the pigs that did not receive
antibiotics developed pathology similar to clinical trichuriasis. Liquid brown diarrhea
developed ~ 14 days post infection, as seen in natural infections (113). Since pathology
was seen with a moderate dose mimicking natural infective doses, it is likely that a
bacterial component is involved in natural disease processes as well. In support of this
view, many researchers have observed presence of spiral bacteria fiorn pigs with colitis
(17, 171). In addition, C. jejuni was isolated fiom the distal colon of pigs exhibiting
pathology in the 1996 study described above. To confirm this hypothesis, a follow-up
study was conducted, in which 3-day-old germ-free gnotobiotic piglets were infected
either with 106 colony forming units of C. jejuni, or 3000 T. suis infective larvae, or with
both. Only piglets that received both the pathogens developed clinical pathology with
bloody diarrhea (111). Recently, a human clinical case in Canada provided further

support to this dual pathogen hypothesis. A patient with C. jejuni infection developed

33

severe colitis with toxic megacolon and acute renal failure. In addition to C. jejuni, T.
suis ova also were recovered from the patient’s feces (160).

Studies on T. suis are limited; only a few studies have been conducted to date that
explore the various factors affecting pathology and infection. However, with increasing
awareness of the possible health hazards posed by T. suis, more studies should be

forthcoming.

E) Treatment

A variety of anti-helminthic treatments have been studied for efficacy against T.
suis infections. In earlier studies, two of the benzimidazole class of compounds,
Flubendazole and Fenbendazole, had varying effects against T. suis infections.
Flubendazole administered in the feed of pigs at 1.5 mg/kg for 5 days was 100% effective
as measured by fecal egg counts (23). Fenbendazole given at 3 mg/kg for 11 days in the
feed was 66% in one study (115) and 100% effective in another study (given the same
dose for 3 days in the feed). Drug efficacy in these studies was determined by fecal egg
counts and warm counts at necropsy (27). The ivermectin class of antihelminthics show
varying degrees of effectiveness. In one study, ivermectin given orally at 300ug/kg was
91.2% effective against T. suis, while it was only 78.1% effective when given
subcutaneously, as determined by necropsy worm counts (156). In another study,
ivermectin was poorly effective, with only 53.9% of the T. suis being cleared after 7 days
of ivermectin therapy (300 rig/kg, oral) (114). Doramectin, another anti-helminthic
belonging to the same class as ivermectin, has also been tested. It too has varying effects.

However, the lowest percentage efficacy obtained with doramectin was still higher than

34

that obtained with other anti-helrninthics. In four studies, doramectin given at 300 ug/kg
intramuscularly for 14—21days was found to be ~100% effective, while in another study it
was found to be 87% effective against adult worms, when administered at the same dose
for the same duration with a similar mode of injection (106, 118, 147, 153, 164). Thus, a
number of treatment options are available for eradication of T. suis infections. However,
the choice might depend on the geographical area, spectrum of activity, and the cost

effectiveness of the drug.

F) Immune responses

Immune responses to nematode infections have been studied extensively over the
last two decades. However, T. suis has not received much attention compared to its
human and rodent counterparts T. trichiura and T. muris. Only two studies exist to date
on immune responses to T. suis. In one study, pigs orally infected with moderate doses of
infective larva (2500 eggs) showed a significant increase in IL-10 in the mesenteric
lymph nodes compared to IL-12, which was undetectable (112). In another study, pigs
given trickle infection (250 eggs) twice weekly for 4 weeks and then challenged with
4000 infective larvae displayed resistance to infection as determined by fecal egg counts,
suggesting the presence of acquired immunity to T. suis (136). The paucity of
information is largely due to the lack of reagents available to test multiple aspects of the
pig immune response. Also, the expense of conducting swine experiments has proved
prohibitive for most laboratories. Therefore, it is not surprising that most studies on
Trichuris spp. have been limited to T. muris and T. trichiura. A review of immune

responses to both these gastrointestinal helrninthes will be presented here.

35

i) T. trichiura

Studies on populations highly endemic for T. trichiura show that antibody
responses play a strong role in infections with this parasite. In one of the earlier studies,
IgG and IgE antibody classes were elevated in the serum of patients with this parasite
(104). Also, IgG1 and IgG2 subclasses predominated compared to other IgG subtypes. In
another study, IgG4 and IgA were found to be additionally important, and there was also
an age-dependent persistence of these antibody levels. Children had a preponderance of
infection, and antibody levels correlated with the infection state. IgG4 and IgE levels
persisted into adulthood, and then declined, albeit gradually (123). IgA is thought to be
associated with acquired immunity, since secretory IgA in the saliva increases with age;
and the higher the IgA levels, the lower the infection intensity (124). With regard to the
cytokine response, the role of TNF-a was demonstrated in two studies, and IL-10 in one
study, as indicated by their upregulation in serum of patients with T. trichiura (107, 123).
The roles of other cytokines such as IL-4 in this intestinal disease are not known.
However, the presence of IgE, IgG1, and IgG4 suggests that Th-2 responses occur as
well.
ii) T. man's

Earlier studies indicated that host genetics might play a role in immunity to T.
muris infections. Certain strains of mice were susceptible to T. muris infections, while
others were resistant. The susceptible phenotype was characterized by a Th-l profile with
the production of IL-12, IL-18, IFN- y, and IgG2a (12). The production of these

cytokines was dependent on Toll Like Receptor 4 (TLR4) and MyD88 (an adaptor

36

molecule), as mice deficient in these molecules are highly resistant to T. muris infection
and develop a strong Th-2 profile (64).

The resistance phenotype on the other hand is characterized by a polarized Th-2
response with the production of IL-3, IL-4, IL-5, IL-9, IL-10, and IL-13 cytokines and
IgG1 and IgA antibodies (36, 37, 150). IgA seems more essential than IgGl for resistance
because transfer of IgA to susceptible mice provides protection against subsequent
infection, while IgGl does not (150). Eosinophils and mast cells were shown not to be
critical in resistance as well. Ablation of IL-5, which recruits eosinophils, and inhibition
of c—kit, a stem cell factor receptor involved in mast cell activation, still resulted in the
resistant phenotype (18). IL-4 activation requires the costirnulatory molecule B7, since
suppressing B7 suppresses IL-4 production (173). Also, in the absence of IL-4, IL-13 is
critical. IL-4 knockout mice can still expel worms successfully, however, neutralization
of IL-13 results in susceptibility (9). A role for TNF—a also has been suggested.
Disruption of the TNF-or receptor results in generation of a Th-l response and failure to
expel T. muris worms. TNF-or was also shown to be required for the IL-13 mediated
expulsion of worms in E-4 knockout mice (4). Another important factor that affects the
resistance phenotype is the transcription factor NF-KBI. Mice with a disruption of this
gene are completely susceptible to infection and develop polarized Th-l responses (5).
The role of chemokines in resistance has been elucidated recently. Mice lacking CCL2, a
CC chemokine ligand, do not mount resistance to T. muris (31). However, the role of
other chemokines is not known. In addition to the role of immune factors, non-immune
factors play a role. Low-level doses (< 40 eggs) induce a polarized Th-l response, while

higher doses (>200 eggs) induce a Th-2 profile in mice normally resistant to infection

37

(10, 11). When the doses are given as trickle infections, however, the doses eventually
reach a high enough threshold to induce Th-2 responses and resistance.

Thus, Th-2 cytokines play an important role in resistance to T richuris infections.
Apart from T richuris, Th—2 cytokines are required for resistance against a variety of other
nematode infections. Ascaris lumbricaides, a human parasite, induces higher levels of IL-
4 and IL-5 in humans harboring this helrninth (26). Its swine counterpart, A. suum, also
induces a distinct Th-2 profile including IL-2, IL-4, IL-5, and IL-10 in mice (108, 116).
Th-2 responses, particularly, IL-4 and 1L-13, are required for successful elimination of
T richinella spiralis, Heligosamaides palygyrus, and Nippastangylus brasiliensis
infections in mice (110). In addition, MCP-l and MIP-2 chemokines play a role in the
elimination of T. spiralis infection (50).

The role of individual Th-2 cytokines in the elimination of nematode infections
has been established to an extent. IL-4 has been shown to increase epithelial cell
permeability, and this in turn is thought to loosen the worms from the mucosa. IL-4 also
stimulates peristalsis in the gut for rapid worm expulsion. Furthermore, IL-4 induces IgE
and IgG1 class switching, and along with IL-13 and IL-9, increases mucosal mast cell
production. Finally, it also increases the fluid accumulation in the intestinal lumen. lL-5
plays a role in eosinophilia, while IL-13 is thought to play a supportive role for IL-4
(110).

In recent years, a novel use for helrninths is being investigated due to their ability
to induce a strong Th-2 response. They have been shown to be effective in suppressing
pathology associated with inflammatory bowel disease and Crohn’s disease, which have a

Th—l etiology in humans (46, 166). Whether this irnmunosuppressive therapy finds a

38

wider role or achieves commercial success like that of probiotics remains to be seen.
Also, since adverse effects of T. suis in both pigs and humans have been shown, this

avenue needs to he traveled with caution.

39

III - RATIONALE FOR THIS STUDY

C. jejuni and T. suis are pathogenic organisms that can infect humans and swine.
However, the immune responses to these organisms are poorly characterized. Th-2
immune responses like IL-4 and IL-10 have been shown to play a role in a number of
hehninth infections and IL-6 in a few bacterial infections (4, 9, 28). At the beginning of
this study, very little was known about the Th-2 cytokine responses to C. jejuni, or to T.
suis worms, or their ESP. While this is still true of T. suis, some advances have been
made in recent years with regard to C. jejuni infections. However, cytokine responses are
host species dependent, pathogen dependent, strain dependent and cell type dependent. In
addition, while numerous studies have been conducted using human cell lines in vitro
with multiple pathogens, studies on swine cells are few. Immune response studies on
enteric pathogens using defined cell lines from the animal host, like swine intestinal cells,
are even more rare. Hence, an attempt has been made in this study to elucidate the Th-2
cytokine responses from primary intestinal swine epithelial cells in response to both T.
suis ESP and C. jejuni.

While the microbial factors for C. jejuni pathogenesis have been elucidated in
some detail, the host factors that contribute to its pathogenesis are not very well
understood. We have observed that Trichuris can dramatically affect the outcome of C.
jejuni infections (111). However, other indirect or polymicrobial factors that can affect
the host responses, and subsequently contribute to C. jejuni pathogenesis are not known.
In addition, almost nothing is known about T. suis-mediated pathogenesis in swine. What
is the immune response to T. suis infection? What is the consequence of this immune

response? How does T. suis contribute to pathology? To provide some answers to these

40

questions, an existing swine model was used to study C. jejuni-and T. suis-mediated
pathogenesis. In this model, pathology is seen in swine only when both organisms are
present, and not during either of the single infections. Preliminary studies indicated that
the disease might be due to C. jejuni and that T. suis contributes to this process.
Therefore, this model provides an avenue to study the host factors that might be
modulated during dual infection and that may contribute to C. jejuni (and T. suis)
mediated pathology. Also, the cytokine immune responses to either infection can be
measured using this model, providing some insights to both C. jejuni and T. suis
infections in viva. Thus, this dissertation should make a substantial contribution to
understanding the mechanisms underlying these hitherto unknown phenomena.
Hypothesis
The primary hypothesis is that infection with T. suis induces a predominantly Th-2 type
of cytokine response fi'om the host, and infection with C. jejuni induces a predominantly
Th-l cytokine response, accompanied by a Th-2 component. Therefore, in dual
infections, synergy in Th-2 components occurs, resulting in upregulation of certain
cytokines like IL-4, IL-6 and IL-10 in the host, with one or more of these cytokines
playing a role in generating severe disease and pathology. An additional hypothesis is
that upregulation of Th-2 cytokines causes a down-regulation in Th-l cytokine responses
needed for resistance to C. jejuni.
Short-term goals- The short-term goals are to address the following questions.

1) Do T. suis excretory secretory products induce IL-4, IL-6 and IL-10 cytokine

responses from the swine intestinal cells in vitro?

41

2) Does C. jejuni induce IL-4, IL-6 and IL-10 cytokine responses from the swine
intestinal cells in vitro?

3) Can synergy occur in viva in any of these cytokines (IL-4, IL-6 and IL-10) as
determined from the in vitro epithelial cell culture system?

4) What is the Th-2 immune response to T. suis and C. jejuni infections in viva?

5) What is (are) the role (3) of Th-2 cytokines upregulated during the in viva
infection in C. jejuni mediated invasion of swine epithelial cells?

In summary, the goal of this study is to determine the role(s) of IL-4, IL-6 and IL-
10 in the colonic disease mediated by C. jejuni and T. suis in swine.

Chapter 2 summarizes studies on in vitro Th-2 cytokine responses from intestinal
swine epithelial cells in response to two different strains of C. jejuni and T. suis ESP. Th-
2 cytokines including IL-4 and IL-10, have been shown to be essential for resistance
against helrninth infections, while IL-6 is stimulated as part of the host defense against
bacterial infections (28, 32, 44). Additionally, IL-6 has been shown to downregulate anti-
leishmanial activity in human macrophages, while IL-4 and IL-10 inhibit Th-l cytokine
responses. However, far from being beneficial, an improper cytokine response can
enhance severity of disease as evidenced by IL-6 mediated malignancies and IL-4
mediated pathologies (70, 98). Since our swine dual infection model has both a helrninth
component and a bacterial component, we decided to analyze the production of IL-4, IL-
6, and IL-10 cytokines in response to these pathogens. These cytokine responses were
measured in both differentiated and undifferentiated epithelial cells after challenge with
C. jejuni and/ or T. suis ESP. The potential roles of these responses in the dual infection

are discussed.

42

Chapter 3 summarizes studies on in viva Th-2 cytokine responses (IL-4, IL-6,
and IL-10) from 4-5 week-old pigs infected with C. jejuni, T. suis eggs, or both.
Inoculations of pigs were carried out either concurrently (short-term experiments) or with
inoculation of T. suis 21 days prior to infection with C. jejuni, followed by C. jejuni
infection (long-term experiments). The potential roles of the cytokines induced at
significant levels during dual infection are discussed.

Chapter 4 summarizes studies on the specific roles of IL-4 and IL-10 in C. jejuni
invasion of IPEC-l cells. Invasive strains of C. jejuni are considered more virulent than
non-invasive strains. While the C. jejuni-secreted factors that contribute to invasion are
well documented, the host soluble factors that contribute to its invasion are not well
understood. In this chapter, evidence of a hitherto unknown role for IL-4 in C. jejuni
pathogenesis is presented, and its role during dual infection of swine with C. jejuni and T.
suis is discussed.

Chapter 5 provides a summary of all the prior chapters and discusses the possible
mechanism of pathogenesis in natural infections with C. jejuni and T. suis in swine in

view of these results.

43

10.

REFERENCES

Abner, S. R., G. Parthasarathy, D. E. Hill, and L. S. Mansfield. 2001.
T richuris suis: detection of antibacterial activity in excretory-secretory products
from adults. Exp Parasitol 99:26-36.

Al—Salloom, F. S., A. A] Mahmeed, A. Ismaeel, G. A. Botta, and M. Bakhiet.
2003. Campylobacter-stimulated INT407 cells produce dissociated cytokine
profiles. J Infect 47 :217-24.

Arimi, S. M., R. W. Park, and C. R. Fricker. 1990. Study of haemolytic
activity of some Campylobacter spp. on blood agar plates. J Appl Bacterial
69:384-9.

Artis, D., N. E. Humphreys, A. J. Bancroft, N. J. Rothwell, C. S. Potten, and
R. K. Grencis. 1999. Tumor necrosis factor alpha is a critical component of
interleukin 13-mediated protective T helper cell type 2 responses during helrninth

infection. J Exp Med 190:953-62.

Artis, D., S. Shapira, N. Mason, K. M. Speirs, M. Goldschmidt, J. Caamano,
H. C. Liou, C. A. Hunter, and P. Scott. 2002. Differential requirement for NF-

kappa B family members in control of helrninth infection and intestinal
inflammation. J Immunol 169:4481-7.

Aspinall, G. 0., C. M. Lynch, H. Pang, R. T. Shaver, and A. P. Moran. 1995.
Chemical structures of the core region of Campylobacterjejuni O:3
lipopolysaccharide and an associated polysaccharide. Eur J Biochem 231 :570-8.

Bacon, D. J., R. A. Alm, D. H. Burr, L. Hu, D. J. Kopecko, C. P. Ewing, T. J.
Trust, and P. Guerry. 2000. Involvement of a plasmid in virulence of
Campylobacterjejuni 81-176. Infect Irnmun 68:43 84-90.

Bacon, D. J., W. M. Johnson, and F. G. Rodgers. 1999. Identification and
characterisation of a cytotoxic porin-lipopolysaccharide complex from
Campylobacterjejuni. J Med Microbiol 48:139-48.

Bancroft, A. J., D. Artis, D. D. Donaldson, J. P. Sypek, and R. K. Grencis.
2000. Gastrointestinal nematode expulsion in IL-4 knockout mice is IL-13
dependent. Eur J Immunol 30:2083-91.

Bancroft, A. J., K. J. Else, and R. K. Grencis. 1994. Low-level infection with

T richuris muris significantly affects the polarization of the CD4 response. Eur J
Immun0124:3113-8.

44

11.

12.

l3.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

Bancroft, A. J., K. J. Else, N. E. Humphreys, and R. K. Grencis. 2001. The
effect of challenge and trickle Trichuris muris infections on the polarisation of the
immune response. Int J Parasitol 31 : 1627-37.

Bancroft, A. J., K. J. Else, J. P. Sypek, and R. K. Grencis. 1997. Interleukin-12
promotes a chronic intestinal nematode infection. Eur J Immunol 27 :866-70.

Batte, E. G., R. D. McLamb, K. E. Muse, S. D. Tally, and T. J. Vestal. 1977.
Pathophysiology of swine trichuriasis. Am J Vet Res 38:1075-9.

Beer, R. J. 1973. Morphological descriptions of the egg and larval stages of
T richuris suis Schrank, 1788. Parasitology 67 :263-78.

Beer, R. J. 1976. The relationship between T richuris trichiura (Linnaeus 1758)
of man and T richuris suis (Schrank 1788) of the pig. Res Vet Sci 20:47—54.

Beer, R. J. 1973. Studies on the biology of the life-cycle of Trichuris suis
Schrank, 1788. Parasitology 67:253-62.

Beer, R. J., and J. M. Rutter. 1972. Spirochaetal invasion of the colonic mucosa
in a syndrome resembling swine dysentery following experimental T richuris suis
infection in weaned pigs. Res Vet Sci 13:593-5.

Betts, C. J ., and K. J. Else. 1999. Mast cells, eosinophils and antibody-mediated
cellular cytotoxicity are not critical in resistance to T richuris muris. Parasite
Immunol 21 :45-52.

Biswas, D., H. Niwa, and K. Itoh. 2004. Infection with Campylobacterjejuni
induces tyrosine-phosphorylated proteins into 1N T-407 cells. Microbiol Immunol
48:221-8.

Black, R. E., M. M. Levine, M. L. Clements, T. P. Hughes, and M. J. Blaser.
1988. Experimental Campylobacterjejuni infection in humans. J Infect Dis
157 :472-9.

Boer, P., J. A. Wagenaar, R. P. Achterberg, J. P. Putten, L. M. Schouls, and
B. Duim. 2002. Generation of Campylobacterjejuni genetic diversity in viva.
Mol Microbiol 44:351-9.

Bourke, B., P. Sherman, H. Louie, E. Hani, P. Islur, and V. L. Chan. 1995.
Physical and genetic map of the genome of Campylobacter upsaliensis.
Microbiology 141 ( Pt 10):2417-24.

Bradley, R. E., J. Guerrero, H. N. Becker, B. F. Michael, and K. Newcomb.

1983. Flubendazole: dose range and efficacy studies against common internal
parasites of swine. Am J Vet Res 44:1329-33.

45

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

Cardarelli-Leite, P., K. Blam, C. M. Patton, M. A. Nicholson, A. G.
Steigerwalt, S. B. Hunter, D. J. Brenner, T. J. Barrett, and B. Swaminathan.
1996. Rapid identification of Campylobacter species by restriction fragment
length polymorphism analysis of a PCR-amplified fragment of the gene coding
for 16S rRNA. J Clin Microbial 34:62-7.

Chang, N., and D. E. Taylor. 1990. Use of pulsed-field agarose gel
electrophoresis to size genomes of Campylobacter species and to construct a SalI
map of Campylobacterjejuni UA580. J Bacteriol 172:5211-7.

Cooper, P. J., M. E. Chico, C. Sandoval, I. Espinel, A. Guevara, M. W.
Kennedy, J. F. Urban Jr, G. E. Griffin, and T. B. Nutman. 2000. Human
infection with Ascaris lumbricaides is associated with a polarized cytokine
response. J Infect Dis 182:1207-13.

Carwin, R. M., S. E. Pratt, and R. K. Muser. 1984. Evaluation of fenbendazale
as an extended anthelmintic treatment regimen for swine. J Am Vet Med Assoc
185:58-9.

de Man, P., C. van Kaaten, L. Aarden, I. Engberg, H. Linder, and C.
Svanbarg Eden. 1989. Interleukin-6 induced at mucosal surfaces by gram-
negative bacterial infection. Infect Irnmun 57:3383-8.

De Mela, M. A., G. Gabbiani, and J. C. Pechere. 1989. Cellular events and
intracellular survival of Campylobacterjejuni during infection of HEp-2 cells.
Infect Irnmun 57:2214-22.

de Mela, M. A., and J. C. Pechere. 1988. Effect of mucin an Campylobacter
jejuni association and invasion on HEp—2 cells. Microb Pathog 5:71-6.

deSchoalmeester, M. L., M. C. Little, B. J. Rollins, and K. J. Else. 2003.
Absence of CC chemokine ligand 2 results in an altered Thl/Th2 cytokine
balance and failure to expel Trichuris muris infection. J Immunol 170:4693-700.

Dimack, K. A., D. G. Addiss, M. L. Eberhard, and P. J. Lammie. 1994.
Differential praliferative and interleukin-10 responses to fractionated filarial

antigens: preferential recognition by patients with chronic lymphatic dysfunction.
J Infect Dis 170:403-12.

Dalg, P., R. Yao, D. H. Burr, P. Guerry, and T. J. Trust. 1996. An
environmentally regulated pilus-like appendage involved in Campylobacter
pathogenesis. Mol Microbiol 20:885-94.

Darrell, N., J. A. Mangan, K. G. Laing, J. Hinds, D. Linton, H. Al-Ghusein,

B. G. Barrel], J. Parkhill, N. G. Stoker, A. V. Karlyshev, P. D. Butcher, and
B. W. Wren. 2001. Whole genome comparison of Campylobacterjejuni human

46

35.

36.

37.

38.

39.

40.

41.

42.

43.

45.

isolates using a low-cost microarray reveals extensive genetic diversity. Genome
Res 11:1706-15.

Drake, L., Y. Karchev, L. Bashfard, M. Djamgaz, D. Wakelin, F. Ashall, and
D. Bundy. 1994. The major secreted product of the whipworrn, T richuris, is a
pore-forming protein. Proc R Soc Lond B Biol Sci 257 :255-61.

Else, K. J., G. M. Entwistle, and R. K. Grencis. 1993. Correlations between
warm burden and markers of Th1 and Th2 cell subset induction in an inbred strain

of mouse infected with Trichuris muris. Parasite Immunol 15:595-600.

Else, K. J., D. Wakelin, D. L. Wassam, and K. M. Hauda. 1990. The influence
of genes mapping within the major histocompatibility complex on resistance to
T richuris muris infections in mice. Parasitology 101 Pt 1:61-7.

Enackssan, A., J. Lundberg, E. Weitzberg, A. Norrby-Teglund, and B.
Sveuungssan. 2004. Rectal nitric oxide gas and stool cytokine levels during the
course of infectious gastroenteritis. Clin Diagn Lab Immunol 11:250-4.

Everest, P. H., H. Gaassens, J. P. Butzler, D. Lloyd, S. Knuttan, J. M. Ketley,
and P. H. Williams. 1992. Differentiated Caco-2 cells as a model for enteric
invasion by Campylobacterjejuni and C. coli. J Med Microbiol 37:319-25.

Eyigar, A., K. A. Dawson, B. E. Langlais, and C. L. Pickett. 1999. Cytolethal
distending toxin genes in Campylobacterjejuni and Campylobacter coli isolates:
detection and analysis by PCR. J Clin Microbiol 37:1646-50.

Eyigar, A., K. A. Dawson, B. E. Langlais, and C. L. Pickett. 1999. Detection of
cytolethal distending toxin activity and cdt genes in Campylobacter spp. isolated
from chicken carcasses. Appl Environ Microbiol 65:1501-5.

Fetterer, R. H., and D. E. Hill. 1994. Localization of phenol oxidase in female
T richuris suis. J Parasitol 80:952-9.

Fetterer, R. H., and D. E. Hill. 1993. The occurrence of phenol oxidase activity
in female T richuris suis. J Parasitol 79:155-9.

Finkelman, F. D., E. J. Pearce, J. F. Urban, J r., and A. Sher. 1991. Regulation
and biological function of helrninth-induced cytokine responses. Immunol Today
12:A62-6.

Fitzgerald, C., R. J. Owen, and J. Stanley. 1996. Comprehensive ribotyping

scheme for heat-stable serotypes of Campylobacterjejuni. J Clin Microbiol
34:265-9.

47

46.

47.

48.

49.

50.

51.

52.

53.

54.

55.

56.

Fax, J. G., P. Beck, C. A. Dangler, M. T. Whary, T. C. Wang, H. N. Shi, and
C. Nagler-Andersan. 2000. Concurrent enteric helrninth infection modulates
inflammation and gastric immune responses and reduces helicobacter-induced
gastric atrophy. Nat Med 6:536-42.

Fox, J. G., A. B. Rogers, M. T. Whary, Z. Ge, N. S. Taylor, S. Xu, B. H.
Harwitz, and S. E. Erdman. 2004. Gastroenteritis in NF-kappaB-deficient mice
is produced with wild-type Camplyabacterjejuni but not with C. jejuni lacking
cytolethal distending toxin despite persistent colonization with both strains. Infect
Irnmun 72:1116-25.

Friedman, C. R., Neimann, J., Wegener, H.C., and RV. Tauxe. 2000.
Epidemiology of Campylobacterjejuni infections in the United States and other
industrialized nations, p. 121-138. In I. Nachamkin, and Blaser, M.J (ed.),
Campylobacter, Second ed. American Society for Microbiology, Washington,
DC.

Fry, B., Oldfield, NJ, Karalik, V, Calae, PJ and JM Ketley. 2000. Genetics of
Campylobacter lipopolysaccharide biosyntheis, p. 381-404. In I. Nachamkin, and
Blaser, M.J (ed.), Campylobacter, Second ed. ASM Press, Washington DC.

Frydas, S., T. Rallis, I. Theodarides, M. N. Patsikas, C. Trakatellis, M. Di
Giaacchina, and M. Felaco. 2000. T richinella spiralis infection is mediated by
MCP-l and MIP-2, while echinococcus granulosus is strongly mediated by MCP-
1, but not MIP-2. Int J Immunapathol Pharrnacol 13:21-26.

Fuglsang, A. 2003. The genome of Campylobacterjejuni: codon and amino acid
usage. Apmis 111:605-18.

Fujita, M., T. Moraaka, S. Fujimata, T. Mariya, and K. Amaka. 1995.
Southern blotting analyses of strains of Campylobacter fetus using the conserved
region of sapA. Arch Microbial 164:444-7.

Galinda, M. A., W. A. Day, B. H. Raphael, and L. A. Jaens. 2001. Cloning and
characterization of a Campylobacterjejuni iron-uptake operon. Curr Microbiol
42: 139-43.

Garrity, G. M., Bell, J.A., and T.G. Lilburn. 2003. Taxonomic outline of the
procaryates, Second ed. Springer-Verlag, New York.

Gaynor, E. C., N. Ghari, and S. Falkaw. 2001. Bile-induced 'pili' in

Campylobacterjejuni are bacteria-independent artifacts of the culture medium.
Mol Microbiol 39: 1 546-9.

Gilbert, M., M. F. Karwaski, S. Bernatchez, N. M. Young, E. Tabaada, J.
Michniewicz, A. M. Cunningham, and W. W. Wakarchuk. 2002. The genetic

48

57.

58.

59.

60.

61.

62.

63.

64.

65.

66.

bases for the variation in the lipo-oligosaccharide of the mucosal pathogen,
Campylobacterjejuni. Biosynthesis of sialylated ganglioside mimics in the core
oligasaccharide. J Biol Chem 277 :327—37.

Grant, K. A., I. U. Belandia, N. Dekker, P. T. Richardson, and S. F. Park.
1997. Molecular characterization of pldA, the structural gene for a phospholipase
A from Campylobacter coli, and its contribution to cell-associated hemolysis.

Infect Irnmun 65:1172-80.

Guerrant, R. L., C. A. Wanke, R. A. Pennie, L. J. Barrett, A. A. Lima, and A.
D. O'Brien. 1987. Production of a unique cytotoxin by Campylobacterjejuni.
Infect Irnmun 55:2526-30.

Guerry, P., R. A. Alm, M. E. Power, S. M. Logan, and T. J. Trust. 1991. Role
of two flagellin genes in Campylobacter motility. J Bacteriol 173:4757-64.

Guerry, P., C. P. Ewing, T. E. Hickey, M. M. Prendergast, and A. P. Moran.
2000. Sialylation of lipooligosaccharide cores affects immunogenicity and serum
resistance of Campylobacterjejuni. Infect Irnmun 68:6656-62.

Guerry, P., C. M. Szymanski, M. M. Prendergast, T. E. Hickey, C. P. Ewing,
D. L. Pattarini, and A. P. Moran. 2002. Phase variation of Campylobacter
jejuni 81-176 lipooligosaccharide affects ganglioside mimicry and invasiveness in
vitro. Infect Irnmun 70:7 87-93.

Hassane, D. C., R. B. Lee, and C. L. Pickett. 2003. Campylobacterjejum’
cytolethal distending toxin promotes DNA repair responses in normal human
cells. Infect Irnmun 71 :541-5.

Healing, T. D. 1992. The surveillance of communicable disease in the European
Community. Commun Dis Rep CDR Rev 2:R73-7.

Helmby, H., and R. K. Grencis. 2003. Essential role for TLR4 and MyD88 in
the development of chronic intestinal nematode infection. Eur J Immunol
33:2974-9.

Hickey, T. E., S. Baqar, A. L. Bourgeois, C. P. Ewing, and P. Guerry. 1999.
Campylobacterjejuni-stimulated secretion of interleukin-8 by INT 407 cells.
Infect Irnmun 67 :88-93.

Hickey, T. E., A. L. McVeigh, D. A. Scott, R. E. Michielutti, A. Bixby, S. A.
Carroll, A. L. Bourgeois, and P. Guerry. 2000. Campylobacterjejuni cytolethal
distending toxin mediates release of interleukin-8 from intestinal epithelial cells.
Infect Immun 68:6535-41.

49

67.

68.

69.

70.

71.

72.

73.

74.

75.

76.

77.

78.

Hill, D. E., H. R. Gamble, M. L. Rhoads, R. H. Fetterer, and J. F. Urban, Jr.
1993. Trichuris suis: a zinc metalloprotease from culture fluids of adult parasites.
Exp Parasitol 77: 1 70-8.

Hill, D. E., R. D. Ramanowski, and J. F. Urban, Jr. 1997. A Trichuris specific
diagnostic antigen from culture fluids of T richuris suis adult worms. Vet Parasitol
68:91-102.

Hill, D. E., and J. A. Sakanari. 1997 . Trichuris suis: thiol protease activity fi'om
adult worms. Exp Parasitol 85:55-62.

Hirano, T., and T. Kishimata. 1989. Interleukin-6: possible implications in
human diseases. Ric Clin Lab 19:1-10.

Hadinka, R. L., and P. H. Gilligan. 1988. Evaluation of the campyslide
agglutination test for confirmatory identification of selected Campylobacter
species. J Clin Microbiol 26:47-9.

Hu, L., and Kopecko, DJ. 2000. Interactions of Campylobacter with eukaryotic
cells: Gut luminal colonization and mucosal invasion, p. 191-216. In I.
Nachamkin, and Blaser, M.J (ed.), Campylobacter, Second ed. American Society
for Microbiology, Washington, DC.

Hu, L., and D. J. Kapecka. 1999. Campylobacterjejuni 81-176 associates with
microtubules and dynein during invasion of human intestinal cells. Infect Irnmun
67:41 71 ~82.

Hurtada, A., and R. J. Owen. 1997 . A molecular scheme based on 23S rRNA
gene polymorphisms for rapid identification of Campylobacter and Arcabacter
species. J Clin Microbiol 35:2401-4.

Hutchinson, D. N., and F. J. Bolton. 1984. Improved blood free selective
medium for the isolation of Campylobacterjejuni from faecal specimens. J Clin
Pathol 37 :956-7 .

Ishikawa, T., Y. Mizunoe, S. Kawahata, A. Takade, M. Harada, S. N. Wai,
and S. Yashida. 2003. The iron-binding protein Dps confers hydrogen peroxide
stress resistance to Campylobacterjejuni. J Bacterial 185:1010-7.

Jenkins, T. 1970. A morphological and histochemical study of T richuris suis
(Schrank, 1788) with special reference to the host-parasite relationship.
Parasitology 61 :357-74.

Johansen, M. V., H. O. Bagh, H. Giver, L. Eriksen, P. Nansen, L.
Stephenson, and K. E. Knudsen. 1997. Schistasamajaponicum and T richuris

50

79.

80.

81.

82.

83.

84.

85.

86.

87.

88.

suis infections in pigs fed diets with high and low protein. Parasitology 115 ( Pt
3):257-64.

Jones, M. A., S. Tatemeyer, D. J. Maskell, C. E. Bryant, and P. A. Barrow.
2003. Induction of proinflarnmatory responses in the human monocytic cell line
THP-l by Campylobacter jejuni. Infect Irnmun 71:2626-33.

Karlyshev, A. V., P. Everest, D. Linton, S. Cawthraw, D. G. N ewell, and B.
W. Wren. 2004. The Campylobacterjejuni general glycosylation system is
important for attachment to human epithelial cells and in the colonization of
chicks. Microbiology 150:1957-64.

Karmali, M. A., A. E. Simar, M. Roscoe, P. C. Fleming, S. S. Smith, and J.
Lane. 1986. Evaluation of a blood-free, charcoal-based, selective medium for the
isolation of Campylobacter organisms from feces. J Clin Microbiol 23:456-9.

Kerstens, P. J., H. P. Endtz, J. F. Meis, W. J. Oyen, R. J. Koapmau, P. J. van
den Braek, and J. W. van der Meer. 1992. Erysipelas-like skin lesions
associated with Campylobacterjejuni septicemia in patients with
hypogammaglobulinemia. Eur J Clin Microbiol Infect Dis 11:842-7.

Kervella, M., J. M. Pages, Z. Pei, G. Grollier, M. J. Blaser, and J. L.
Fauchere. 1993. Isolation and characterization of two Campylobacter glycine-
extracted proteins that bind to HeLa cell membranes. Infect Irnmun 61 :3440-8.

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

Kiehlbauch, J. A., R. A. Albach, L. L. Baum, and K. P. Chang. 1985.
Phagocytosis of Campylobacterjejuni and its intracellular survival in
mononuclear phagocytes. Infect Irnmun 48:446-51.

Kita, E., F. N ishikawa, N. Kamikaidau, A. Nakana, N. Katsui, and S.
Kashiba. 1992. Mononuclear cell response in the liver of mice infected with
hepatotoxigenic Campylobacterjejuni. J Med Microbiol 37:326-31.

Kita, E., D. Oku, A. Hamura, F. Nishikawa, M. Emota, Y. Yagyu, N. Katsui,
and S. Kashiba. 1990. Hepatotoxic activity of Campylobacterjejuni. J Med
Microbiol 33:17 1-82.

Klena, J. D., S. A. Gray, and M. E. Konkel. 1998. Cloning, sequencing, and
characterization of the lipopolysaccharide biosynthetic enzyme

heptosyltransferase I gene (waaC) from Campylobacterjejuni and Campylobacter
coli. Gene 222:177-85.

51

89.

90.

91.

92.

93.

94.

95.

96.

97.

98.

99.

Konkel, M. E., M. D. Carwin, L. A. Jaens, and W. Cieplak. 1992. Factors that
influence the interaction of Campylobacterjejuni with cultured mammalian cells.
J Med Microbial 37:30—7.

Konkel, M. E., S. G. Garvis, S. L. Tiptan, D. E. Anderson, Jr., and W.
Cieplak, Jr. 1997. Identification and molecular cloning of a gene encoding a

fibronectin-binding protein (CadF) from Campylobacterjejuni. Mol Microbiol
24:953-63.

Konkel, M. E., and L. A. Jaens. 1989. Adhesion to and invasion of HEp-2 cells
by Campylobacter spp. Infect Irnmun 57:2984—90.

Konkel, M. E., Jaens, L.A., and Mixter, P.F. 2000. Molecular characterization
of Campylobacterjejuni virulence determinants, p. 217-240. In I. Nachamkin, and
Blaser, M.J (ed.), Campylobacter, Second ed. American Society for
Microbiology, Washington, DC.

Konkel, M. E., B. J. Kim, V. Rivera-Amill, and S. G. Garvis. 1999. Bacterial
secreted proteins are required for the intemaliztion of Campylobacterjejuni into
cultured mammalian cells. Mol Microbiol 32:691-701.

Konkel, M. E., J. D. Klena, V. Rivera-Amill, M. R. Monteville, D. Biswas, B.
Raphael, and J. Mickelson. 2004. Secretion of virulence proteins from
Campylobacterjejuni is dependent on a fimctional flagellar export apparatus. J
Bacteriol 186:3296-303.

Karalik, V., B. N. Fry, M. R. Alderton, B. A. van der Zeijst, and P. J. Colae.
1997. Expression of Campylobacter hyailei lipo-oligasaccharide (LOS) antigens
in Escherichia coli. Microbiology 143 ( Pt 11):3481-9.

Karalik, V., L. Moarthy, and P. J. Colae. 1995. Differentiation of
Campylobacterjejuni and Campylobacter coli strains by using restriction
endonuclease DNA profiles and DNA fragment polymorphisms. J Clin Microbial
33:1 136—40.

Lara-Tejera, M., and J. E. Galan. 2001. Cth, CdtB, and CdtC form a tripartite
complex that is required for cytolethal distending toxin activity. Infect Irnmun
69:4358-65.

Lawrence, C. E., J. C. Paterson, L. M. Higgins, T. T. MacDonald, M. W.
Kennedy, and P. Garside. 1998. IL-4-regulated enteropathy in an intestinal
nematode infection. Eur J Immunol 28:2672-84.

Lee, A., S. C. Smith, and P. J. Caloe. 2000. Detection of a novel campylobacter
cytotoxin. J Appl Microbiol 89:719-25.

52

100.

101.

102.

103.

104.

105.

106.

107.

108.

109.

110.

Lee, R. B., D. C. Hassane, D. L. Cattle, and C. L. Pickett. 2003. Interactions of
Campylobacterjejuni cytolethal distending toxin subunits Cth and CdtC with
HeLa cells. Infect Irnmun 71 :4883-90.

Leonard, E. E., 2nd, T. Takata, M. J. Blaser, S. Falkaw, L. S. Tompkins, and
E. C. Gaynor. 2003. Use of an open-reading frame-specific Campylobacterjejuni
DNA microarray as a new genotyping tool for studying epidemiologically related
isolates. J Infect Dis 187:691-4.

Lesse, A. J., A. A. Campagnari, W. E. Bittner, and M. A. Apicella. 1990.
Increased resolution of lipopolysaccharides and lipooligosaccharides utilizing
tricine-sodium dodecyl sulfate-polyacrylarrride gel electrophoresis. J Immunol
Methods 126:109-17.

Li, C. C., C. H. Chin, J. L. Wu, Y. C. Huang, and T. Y. Lin. 1998.
Antimicrobial susceptibilities of Campylobacterjejuni and coli by using E-test in
Taiwan. Scand J Infect Dis 30:39-42.

Lillywhite, J. E., D. A. Bundy, J. M. Didier, E. S. Cooper, and A. E. Bianca.
1991. Humoral immune responses in human infection with the whipworrn
T richuris trichiura. Parasite Immunol 13:491-507.

Linton, D., M. Gilbert, P. G. Hitchen, A. Dell, H. R. Morris, W. W.
Wakarchuk, N. A. Gregsan, and B. W. Wren. 2000. Phase variation of a beta-
l,3 galactosyltransferase involved in generation of the ganglioside GMl-like lipo-
oligosaccharide of Campylobacterjejuni. Mol Microbiol 37 :501-14.

Logan, N. B., A. J. Weatherley, and R. M. J ones. 1996. Activity of doramectin
against nematode and arthropod parasites of swine. Vet Parasitol 66:87-94.

MacDonald, T. T., J. Spencer, S. H. Murch, M. Y. Choy, S. Venugopal, D. A.
Bundy, and E. S. Cooper. 1994. Irnmunoepidemiology of intestinal helminthic
infections. 3. Mucosal macrophages and cytokine production in the colon of
children with T richuris trichiura dysentery. Trans R Soc Trap Med Hyg 88:265-
8.

Maceda, M. S., E. Faquim-Mauro, A. P. Ferreira, and I. A. Abrahamsohn.
1998. Immunomodulation induced by Ascaris suum extract in rrrice: effect of anti-
interleukin-4 and anti-interleukin-IO antibodies. Scand J Immunol 47:10-8.

Madden, R. H., L. Moran, and P. Scates. 1996. Sub-typing of animal and
human Campylobacter spp. using RAPD. Lett Appl Microbiol 23: 167-70.

Mansfield, L. 8., Cause, W.C.,Finkelman, F.D., and J.F. Urban, Jr. 2000.
Gastrointestinal nematodes and the immune system, p. 555-567. In M. W.

53

111.

112.

113.

114.

115.

116.

117.

118.

119.

120.

Cunningham, and Fuj inarrri, R.S. (ed.), Effects of microbes on the immune
system. Lippincott, Williams and Wilkins, Philadelphia.

Mansfield, L. S., D. T. Gauthier, S. R. Abner, K. M. Jones, S. R. Wilder, and
J. F. Urban. 2003. Enhancement of disease and pathology by synergy of

T richuris suis and Campylobacterjejuni in the colon of immunologically naive
swine. Am J Trop Med Hyg 68:70-80.

Mansfield, L. S., J. F. Urban, R. R. Halley-Shanks, M. P. Murtaugh, D. S.
Zarlenga, D. Foss, A. Canals, W. Cause, and J. K. Lunney. 1998. Construction
of internal cDNA competitors for measuring IL-10 and lL-12 cytokine gene
expression in swine. Vet Immunol Irnmunopathol 65:63-74.

Mansfield, L. S., and J. F. Urban, Jr. 1996. The pathogenesis of necrotic
praliferative colitis in swine is linked to whipwann induced suppression of
mucosal immunity to resident bacteria. Vet Immunol Irnmunopathol 50:1-17.

Marchionda, A. A., and J. Szanta. 1987. Efficacy of dichlorvos, fenbendazale,
and ivermectin in swine with induced intestinal nematode infections. Am J Vet
Res 48:1233-5.

Marti, O. G., T. B. Stewart, and O. M. Hale. 1978. Comparative efficacy of
fenbendazale, dichlorvos, and levarnisale HCI against gastrointestinal nematodes
of pigs. J Parasitol 64: 1028-31.

Matsumata, N., S. Katah, H. Mukae, T. Matsuo, K. Takatsu, and S.
Matsukura. 2003. Critical role of IL-5 in antigen-induced pulmonary
eosinophilia, but not in lymphocyte activation. Int Arch Allergy Immunol
130:209-15.

McSweegan, E., and R. 1. Walker. 1986. Identification and characterization of
two Campylobacterjejuni adhesins for cellular and mucous substrates. Infect
Irnmun 53:141-8.

Mehlharn, H., H. L. Jones, A. J. Weatherley, and B. Schumacher. 1993.
Doramectin, a new avermectin highly efficacious against gastrointestinal
nematodes and lungworrns of cattle and pigs: two studies carried out under field
conditions in Germany. Parasitol Res 79:603-7.

Monteville, M. R., and M. E. Konkel. 2002. F ibranectin-facilitated invasion of
T84 eukaryotic cells by Campylobacterjejuni occurs preferentially at the
basolateral cell surface. Infect Irnmun 70:6665-71.

Moraaka, T., A. Umeda, and K. Amaka. 1985. Motility as an intestinal

colonization factor for Campylobacterjejuni. J Gen Microbiol 131 ( Pt 8):]973-
80.

54

121.

122.

123.

124.

125.

126.

127.

128.

129.

130.

Nachamkin, 1., K. Bahachick, and C. M. Patton. 1993. Flagellin gene typing of
Campylobacterjejuni by restriction fragment length polymorphism analysis. J
Clin Microbiol 31 : 1531-6.

Nachamkin, 1., Engberg, J., and Aarestrup, FM. 2000. Diagnosis and
antimicrobial susceptibility of Campylobacter species, p. 45-66. In I. a. B.
Nachamkin, MJ (ed), Campylobacter, Second ed. American Society for
Microbiology, Washington, DC.

Needham, C. S., D. A. Bundy, J. E. Lillywhite, J. M. Didier, 1. Simmons, and
A. E. Bianca. 1992. The relationship between Trichuris trichiura transmission
intensity and the age-profiles of parasite-specific antibody isotypes in two
endemic communities. Parasitology 105 ( Pt 2):273-83.

Needham, C. S., J. E. Lillywhite, J. M. Didier, A. T. Bianca, and D. A. Bundy.
1994. Comparison of age-dependent antigen recognition in two communities with
high and low T richuris trichiura transmission. Acta Trop 58:87-98.

Newell, D. G., Frost, JA., Duim, B., Wagenaar, J .A., Madden, R.H., van der
Plas, J., and On, S.L.W. 2000. New Developments in the subtyping of
Campylobacter species, p. 27-44. In I. Nachamkin, and Blaser, M.J (ed.),
Campylobacter, Second ed. American Society for Microbiology, Washington,
DC.

Newell, D. G., H. McBride, F. Saunders, Y. Dehele, and A. D. Pearson. 1985.
The virulence of clinical and environmental isolates of Campylobacterjejuni. J
Hyg (Lond) 94:45-54.

Oberhelman, R. A., and Taylor, D.N. 2000. Campylobacter infections in the
developing countries, p. 139-154. In I. Nachamkin, and Blaser, M.J (ed.),
Campylobacter, Second ed. American Society for Microbiology, Washington,
DC.

Oelschlaeger, T. A., P. Guerry, and D. J. Kopecko. 1993. Unusual
microtubule-dependent endocytosis mechanisms triggered by Campylobacter
jejuni and Citrobacterfreundii. Proc Natl Acad Sci U S A 90:6884-8.

Okuda, J., M. Fukumata, Y. Takeda, and M. Nishibuchi. 1997. Examination
of diarrheagenicity of cytolethal distending toxin: suckling mouse response to the
products of the cthBC genes of Shigella dysenteriae. Infect Irnmun 65:428-33.

On, S. L. 1996. Identification methods for campylobacters, helicobacters, and
related organisms. Clin Microbiol Rev 9:405-22.

55

131.

132.

133.

134.

135.

136.

137.

138.

139.

140.

141.

On, S. L. W. a. B. H. 1995. Classification and identification of campylobacters,

helicobacters and allied taxa by numerical analysis of phenotypic characters. Syst.
Appl. Microbiol. 18:374-390.

Owen, R. J., K. Sutherland, C. Fitzgerald, J. Gibson, P. Barman, and J.
Stanley. 1995. Molecular subtyping scheme for serotypes HSl and HS4 of
Campylobacterjejuni. J Clin Microbiol 33:872-7.

Pancarba, P. L., M. A. de Pablo, E. Ortega, A. M. Gallega, C. Alvarez, and
G. Alvarez de Cienfuegas. 1999. Evaluation of cytokine production and

phagocytic activity in mice infected with Campylobacterjejuni. Curr Microbiol
39: 129-33.

Park, S. F., D. Purdy, and S. Leach. 2000. Localized reversible frameshift
mutation in the flhA gene confers phase variability to flagellin gene expression in
Campylobacter coli. J Bacteriol 182:207-10.

Parkhill, J., B. W. Wren, K. Mungall, J. M. Ketley, C. Churcher, D. Basham,
T. Chillingworth, R. M. Davies, T. Feltwell, S. Halrayd, K. J agels, A. V.
Karlyshev, S. Maule, M. J. Pallen, C. W. Penn, M. A. Quail, M. A.
Rajandream, K. M. Rutherford, A. H. van Vliet, S. Whitehead, and B. G.
Barrell. 2000. The genome sequence of the food-borne pathogen Campylobacter
jejuni reveals hypervariable sequences. Nature 403:665-8.

Pedersen, S., and I. Saeed. 2001. Acquired immunity to T richuris suis infection
in pigs. Parasitology 123:95-101.

Pedersen, S., and I. Saeed. 2000. Experimental infection of pigs with three dose
levels of T richuris suis. Parasite 7:275-81.

Pedersen, S., and I. Saeed. 2002. Host age influence on the intensity of
experimental T richuris suis infection in pigs. Parasite 9:75-9.

Pedersen, S., I. Saeed, H. Friis, and K. F. Michaelsen. 2001. Effect of iron
deficiency on Trichuris suis and Ascaris suum infections in pigs. Parasitology
122:589-98.

Pedersen, S., I. Saeed, K. F. Michaelsen, H. Friis, and K. D. Murrell. 2002.
Impact of protein energy malnutrition on Trichuris suis infection in pigs
concomitantly infected with Ascaris suum. Parasitology 124:561-8.

Pei, Z., and M. J. Blaser. 1993. PEBl, the major cell-binding factor of
Campylobacterjejuni, is a homolog of the binding component in gram-negative
nutrient transport systems. J Biol Chem 268: 1 8717-25.

56

142.

143.

144.

145.

146.

147.

148.

149.

150.

151.

152.

Pei, L, C. Burucaa, B. Grignan, S. Baqar, X. Z. Huang, D. J. Kopecko, A. L.
Bourgeois, J. L. Fauchere, and M. J. Blaser. 1998. Mutation in the pebIA locus
of Campylobacterjejuni reduces interactions with epithelial cells and intestinal
colonization of mice. Infect Irnmun 66:938-43.

Pei, Z. H., R. T. Ellison, 3rd, and M. J. Blaser. 1991. Identification,
purification, and characterization of major antigenic proteins of Campylobacter
jejuni. J Biol Chem 266:16363-9.

Pickett, C. L. 2000. Campylobacter toxins and their role in pathogenesis, p. 179-
190. In I. Nachamkin, and Blaser, M.J (ed.), Campylobacter, Second ed.
American Society for Microbiology, Washington, DC.

Poly, F., D. Threadgill, and A. Stintzi. 2004. Identification of Campylobacter
jejuni ATCC 4343l-specific genes by whole microbial genome comparisons. J
Bacteriol 186:4781-95.

Prendergast, M. M., A. J. Lastavica, and A. P. Moran. 1998.
Lipopolysaccharides from Campylobacterjejuni 0:41 strains associated with
Guillain-Barre syndrome exhibit mimicry of GM] ganglioside. Infect Irnmun
66:3649-55.

Reina, D., L. Anderson, M. Habela, A. J. Weatherley, and I. Navarrete. 2000.
Efficacy of doramectin against naturally acquired nematode infection in Iberian
swine. Vet Parasitol 89:139-47.

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

Rhoads, M. L., R. H. Fetterer, D. E. Hill, and J. F. Urban, Jr. 2000. Trichuris
suis: a secretary chymotrypsin/elastase inhibitor with potential as an
irnmunomodulator. Exp Parasitol 95:36-44.

Roach, T. 1., K. J. Else, D. Wakelin, D. J. McLaren, and R. K. Grencis. 1991.
T richuris muris: antigen recognition and transfer of immunity in mice by IgA
monoclonal antibodies. Parasite Immunol 13: 1-12.

Russell, R. G., M. O'Donnoghue, D. C. Blake, J r., J. Zulty, and L. J. DeTolla.
1993. Early colonic damage and invasion of Campylobacterjejuni in
experimentally challenged infant Macaca mulatta. J Infect Dis 168:210—5.

Rutter, J. M., and R. J. Beer. 1975. Synergism between Trichuris suis and the

microbial flora of the large intestine causing dysentery in pigs. Infect Irnmun
1 1 :395-404.

57

153.

154.

155.

156.

157.

158.

159.

160.

161.

162.

Saeki, H., T. Fujii, S. Fukumata, K. Kagota, A. Taneichi, S. Takeda, and M.
Tsukaguchi. 1997. Efficacy of doramectin against intestinal nematodes and
sarcoptic manage mites in naturally infected swine. J Vet Med Sci 59:129-32.

Sahin, 0., N. Luo, S. Huang, and Q. Zhang. 2003. Effect of Campylobacter-
specific maternal antibodies on Campylobacterjejuni colonization in young
chickens. Appl Environ Microbiol 69:5372-9.

Salama, S. M., M. M. Garcia, and D. E. Taylor. 1992. Differentiation of the
subspecies of Campylobacter fetus by genomic sizing. Int J Syst Bacteriol
42:446-50.

Schillhorn van Veen, T. W., and C. D. Gibson. 1983. Anthehnintic activity of
ivermectin in pigs naturally infected with Ascaris and T richuris. Am J Vet Res
44:1732-3.

Schouls, L. M., S. Reulen, B. Duim, J. A. Wagenaar, R. J. Willems, K. E.
Dingle, F. M. Colles, and J. D. Van Embden. 2003. Comparative genotyping of
Campylobacterjejuni by amplified fragment length polymorphism, multilocus
sequence typing, and short repeat sequencing: strain diversity, host range, and
recombination. J Clin Microbiol 41:15-26.

Schwartz, D., R. Perry, D. M. Dambraski, J. M. Merrick, and J. Galdhar.
1996. Invasive ability of C. jejuni/coli isolates from children with diarrhea and the
effect of iron-regulated proteins. Zentralbl Bakteriol 283:485—91.

Shenker, B. J., T. McKay, S. Datar, M. Miller, R. Chawhan, and D. Demuth.
1999. Actinobacillus actinamycetemcamitans irnmunosuppressive protein is a

member of the family of cytolethal distending toxins capable of causing a G2
arrest in human T cells. J Immunol 162:4773-80.

Shin, J. L., G. W. Gardiner, W. Deitel, and G. Kandel. 2004. Does whipworrn
increase the pathogenicity of Campylobacterjejuni? A clinical correlate of an
experimental observation. Can J Gastroenterol 18: 175-7.

Siemer, B. L., C. S. Harrington, E. M. Nielsen, B. Borck, N. L. Nielsen, J.
Engberg, and S. L. On. 2004. Genetic relatedness among Campylobacterjejuni
serotyped isolates of diverse origin as determined by numerical analysis of
amplified fi'agment length polymorphism (AF LP) profiles. J Appl Microbiol
96:795-802.

Skirrow, M., and Blaser, M.J. 2000. Clinical aspects of Campylobacter

infection, p. 69-88. In I. Nachamkin, and Blaser, M.J (ed.), Campylobacter,
Second ed. American Society for Microbiology, Washington, DC.

58

163.

164.

165.

166.

167.

168.

169.

170.

171.

172.

173.

174.

Sorvilla, F. J., L. E. Lieb, and S. H. Waterman. 1991. Incidence of
campylobacteriosis among patients with AIDS in Las Angeles County. J Acquir
Immune Defic Syndr 4:598-602.

Stewart, T. B., M. C. Fax, and S. E. Wiles. 1996. Dorarnectin efficacy against
gastrointestinal nematodes in pigs. Vet Parasitol 66: 101-8.

Stewart, T. B., and O. M. Hale. 1988. Losses to internal parasites in swine
production. J Anirn Sci 66:1548-54.

Summers, R. W., D. E. Elliott, K. Qadir, J. F. Urban, J r., R. Thompson, and
J. V. Weinstock. 2003. T richuris suis seems to be safe and possibly effective in
the treatment of inflammatory bowel disease. Am J Gastroenterol 98:2034-41.

Szymanski, C. M., and G. D. Armstrong. 1996. Interactions between
Campylobacterjejuni and lipids. Infect Irnmun 64:3467-74.

Szymanski, C. M., M. King, M. Haardt, and G. D. Armstrong. 1995.
Campylobacterjejuni motility and invasion of Caco-2 cells. Infect Irnmun
63:4295-300.

Tanner, A. C., J. L. Dzink, J. L. Ebersale, and S. S. Sacransky. 1987.
Walinella recta, campylobacter concisus, bacteraides gracilis. and Eikenella
carrodens from periodontal lesions. J Periodontal Res 22:327-30.

Tay, S. T., S. Devi, S. D. Puthucheary, and I. M. Kautner. 1995. Detection of
haemolytic activity of campylobacters by agarose haemolysis and microplate
assay. J Med Microbiol 42:175-80.

Taylor, D. J ., and W. F. Blakemore. 1971. Spirochaetal invasion of the colonic
epithelium in swine dysentery. Res Vet Sci 12: 177-9.

Ueki, Y., A. Umeda, S. Fujimata, M. Mitsuyama, and K. Amaka. 1987.
Protection against Campylobacterjejuni infection in suckling mice by anti-
flagellar antibody. Microbiol Immunol 31 :1 161-7 1.

Urban, J., H. Fang, Q. Liu, M. J. Ekkens, S. J. Chen, D. Nguyen, V. Mitra, D.
D. Donaldson, C. Byrd, R. Peach, S. C. Morris, F. D. Finkelman, L. Schapf,
and W. C. Gause. 2000. IL-l3-mediated warm expulsion is B7 independent and
IFN-gamma sensitive. J Immunol 164:4250-6.

Van Rhijn, I., L. H. Van den Berg, C. W. Aug, J. Admiraal, and T.

Logtenberg. 2003. Expansion of human gammadelta T cells after in vitro
stimulation with Campylobacterjejuni. Int Immunol 15:373-82.

59

175.

176.

177.

178.

179.

180.

181.

182.

183.

184.

185.

Vandamme, P. 2000. Taxonomy of the family Campylobacteraceae, p. 3-26. In I.
a. B. Nachamkin, M] (ed.), Campylobacter, Second ed. American Society for
Microbiology, Washington, DC.

Wai, S. N., K. Nakayama, K. Umene, T. Mariya, and K. Amaka. 1996.
Construction of a ferritin—deficient mutant of Campylobacterjejuni: contribution

of ferritin to iron storage and protection against oxidative stress. Mol Microbiol
20:1 127-34.

Wassenaar, T. M., N. M. Bleumink-Pluym, and B. A. van der Zeijst. 1991.
Inactivation of Campylobacterjejuni flagellin genes by homologous

recombination demonstrates that flaA but not flaB is required for invasion. Embo
J 10:2055-61.

Wassenaar, T. M., B. A. van der Zeijst, R. Ayling, and D. G. Newell. 1993.
Colonization of chicks by motility mutants of Campylobacterjejuni demonstrates
the importance of flagellin A expression. J Gen Microbial 139 ( Pt 6):] 171-5.

Whitehause, C. A., P. B. Balbo, E. C. Pesci, D. L. Cattle, P. M. Mirabita, and
C. L. Pickett. 1998. Campylobacterjejuni cytolethal distending toxin causes a
G2-phase cell cycle block. Infect Irnmun 66:1934-40.

Wilson, D. L., J. A. Bell, V. B. Young, S. R. Wilder, L. S. Mansfield, and J. E.
Linz. 2003. Variation of the natural transformation frequency of Campylobacter
jejuni in liquid shake culture. Microbiology 149:3603-15.

Wood, A. C., N. J. Oldfield, C. A. O'Dwyer, and J. M. Ketley. 1999. Cloning,
mutation and distribution of a putative lipopolysaccharide biosynthesis locus in
Campylobacterjejuni. Microbiology 145 ( Pt 2):379-88.

Waoldridge, K. G., and J. M. Ketley. 1997. Campylobacter-host cell
interactions. Trends Microbial 5:96-102.

Yamasaki, M., S. Igimi, Y. Katayama, S. Yamamato, and F. Amano. 2004.
Identification of an oxidative stress-sensitive protein from Campylobacterjejuni,
homologous to rubredoxin oxidoreductase/rubrerythrin. FEMS Microbiol Lett
235:57-63.

Yao, R., D. H. Burr, P. Daig, T. J. Trust, H. Niu, and P. Guerry. 1994.
Isolation of motile and non-motile insertional mutants of Campylobacterjejuni:
the role of motility in adherence and invasion of eukaryotic cells. Mol Microbiol

14:883-93.
Ziprin, R. L., C. R. Young, L. H. Stanker, M. E. Hume, and M. E. Konkel.

1999. The absence of cecal colonization of chicks by a mutant of Campylobacter
jejuni not expressing bacterial fibronectin-binding protein. Avian Dis 43:586-9.

60

CHAPTER 2
INTERLEUKIN 4 (IL-4), IL-6 AND IL-10 CYTOKINE SECRETION
FROM INTESTINAL EPITHELIAL CELLS (IPEC-l) IN RESPONSE TO
CAMPYLOBA C T ER JEJUNI AND T RICH URIS S UIS EXCRETORY

SECRETORY PRODUCTS

61

ABSTRACT

Swine with dual infections of Campylobacterjejuni and T richuris suis developed
disease more severe than the additive effect of either disease alone. In this polymicrobial
infection, resistance to either organism was altered by the presence of the other. Since
Th-2 cytokines play a vital role in resistance to helrninth infections, we hypothesized that
during dual infection, immune responses are biased to a predominantly Th-2 type and
synergy in cytokine production stimulated by these two pathogens results in Th-2-
mediated disease. In order to examine the role of C. jejuni and T. suis in Th-2 immune
modulation, Intestinal Pig Epithelial Cells-l (IPEC-l) were exposed to C. jejuni (strains
ATCC 33292 and 81176) and T. suis ESP, and Th-2 cytokine responses (IL-4, IL-6 and
IL-10) were measured. IL-6 was the predominant cytokine produced, followed by IL-1 0.
There was no secretion of IL-4 from IPEC-l cells. IL-6 was secreted within 24 hr in
response to C. jejuni, while IL-10 induction took longer (> 24 hr). In addition, IL-6 was
secreted from both differentiated and undifferentiated IPEC-l cells, while IL-10 was
secreted predominantly from differentiated cells, compared to negative controls. In all
cases, secretion of these cytokines occurred on the cell surface exposed to the bacteria.
T. suis ESP on the other hand, induced IL-6 and IL-10 within a 24 hr time frame, and
both the differentiated and undifferentiated cells secreted these cytokines in response to a
low level ESP (0.3 mg/ml) challenge. These data show that C. jejuni and T. suis ESP
induce Th-2 cytokines in swine epithelial cells and suggests that Th-2-mediated synergy
in cytokines might occur during concurrent infections, with IL-6 in particular. In

addition, preformed IL-6 was demonstrated in IPEC-l cells for the first time to date, and

62

this result suggests that swine intestinal cells play a role in immune response during

enteric infections.

63

INTRODUCTION

Campylobacterjejuni (C. jejuni) is one of the leading causes of bacterial enteritis
in hmnans, particularly children (7, 15, 26). It is a microaerophlic, gram-negative, spiral
or comma-shaped bacterium and is mainly acquired through contaminated food.
T richuris spp. are helrninth parasites that live in the cecum and proximal colon of many
mammals, including humans and swine, and are found commonly throughout the world
(4, 5). In an animal model of polymicrobial infection, C. jejuni and T richuris suis acted
synergistically in the colon of neonatal, gnotobiotic, and naive conventionally-reared pigs
to produce disease and pathological lesions (30, 31). Germ-free piglets infected
individually with low doses of either C. jejuni (106 cfu) or T. suis (3000 embryonated
eggs) displayed few clinical signs of disease. In contrast, severe colonic pathology was
observed in pigs with dual infections; pathology was marked by focal lesions, infiltration
of inflammatory cells, excessive mucus secretion, bloody diarrhea and bacterial invasion
into epithelial cells and macrophages (30). Lymphoid tissues, especially lymphoglandular
complexes (LGCs), appeared to be one important route for C. jejuni invasion. LGCs are
secondary lymphoid organs present in the distal colon of swine, humans, rats, cattle, and
other animals (25, 29, 34, 38) . They are structurally similar to the Peyer’s patches in the
small intestine and in the sensitized hast are composed of a follicle-associated epithelium
(FAE) overlying B-cell dependent germinal centers. IgA is generated here after C. jejuni
challenge and may also serve as sites where the bacterium is harbored or disseminated

(32).

64

This study was designed to understand the role of immunity and immuno-
modulation in this dual infection model. Since it is the specific immune reaction against
pathogens that keeps dangerous organisms at bay and determines the outcome of the
disease state, we hypothesized that cytokine dysregulation may be an important
contributor to this polymicrobial synergism. To begin to address this question, we
undertook to define the host Th-2 immune responses to these two pathogens. Th-2
cytokines, including IL-4 and IL-10, have been shown to be essential for resistance
against helrninth infections, while IL-6 is stimulated as part of the host defense against
bacterial infections (1 1-13, 18). Additionally, IL-6 has been shown to downregulate anti-
leishmanial activity in human macrophages, while IL-4 and IL-10 inhibit Th-l cytokine
responses (17, 22). However, far from being beneficial, an improper cytokine response
can enhance severity of disease, as evidenced by IL-6 mediated malignancies and IL-4
mediated pathologies (14, 21, 28). Since our swine dual infection model has both a
helrninth component and a bacterial component, we decided to analyze IL-4, IL-6 and IL-
10 cytokine levels, and in this study, we focused on C. jejuni and T. suis mediated in vitro
Th-2 responses. We posed the following hypotheses: 1) resistance to C. jejuni infection
requires a predominant Th-l immune response accompanied by a moderate Th-2
component, while resistance to T. suis requires a predominant Th-2 response, and 2) the
immune response is shifted towards a predominantly Th-2 type during polymicrobial
infections with T. suis, which interferes with the Th-l immmrity directed against the
bacteria. Since bacterial invasion of intestinal epithelial cells appears to be the first
interactive event in C. jejuni and T. suis infection of swine, and since epithelial cells have

been shown to be a significant source of the enteric immune response, IPEC-l cells were

65

chosen for the current study (19, 27, 30, 32, 36). lPEC-l is a non-immortalized,
undifferentiated small intestinal cell line from neonatal swine that differentiates when
grown on porous substrates for 10-14 days (6, 16). Because IPEC-l is a non-
immortalized cell line, it is a better model of the normal host intestine than a transformed
cell line. However, very little is known about the cytokine responses of IPEC-l cells.
Also, ESP from other helrninths has been shown to be irnmunogenic. Hence, in the
current study, IPEC-l monolayers were challenged with pathogenic C. jejuni strains and
T. suis ESP, and the IL-4, IL-6, and IL-10 cytokine responses from these cells were
measured. Secreted cytokines were quantified rather than intra-cellular cytokine levels, as
autocrine and paracrine effects of cytokines are more likely to interfere with immunity.
The data obtained suggests that intestinal epithelial cells (IPEC-l) could be a major
source of enteric IL-6 and some IL-10 and hence may play an important role in

immunomodulation during enteric C. jejuni and T. suis infections.

66

MATERIALS AND METHODS

Bacterial strains and culture

C. jejuni strains ATCC 33292 and 81-176 were used for the challenge studies.
Both were initially isolated from humans with enteritis. C. jejuni 81-176 was kindly
donated by Dr. Carol Pickett, University of Kentucky. Bacterial colonies were grown on
Mueller Hinton Agar supplemented with 5% sheep’s blood (Cleveland Scientific, Bath,
OH) for a period of 48-72 hr. An isolated colony was spread on a fresh plate and firrther
grown as a lawn for 20 hr. Bacterial growth was resuspended in RPMI 1640 (Gibco BRL,
Rockville, MD) and the optical density (OD560) adjusted to 0.1 to achieve 5 x 108
CFU/ml. Dilutions of the bacteria were made in RPMI 1640 as required for the

experimental design.

Cell culture

Both differentiated and undifferentiated IPEC-l cells were used to determine the
IL-4, IL-6 and IL-10 expression in response to C. jejuni in vitro. All media and
supplements for cell culture were obtained from Gibco BRL, Rockville, MD, unless
otherwise stated. transwell and 24-well culture plates were obtained from Corning Costar,
Corning, NY.
Differentiated cells

Differentiated cells were used to represent the intestinal villus tip cells and the
FAE of the LGCs. The culture protocol was kindly provided by Dr. Jerrold Turner of
University of Chicago (personal communication). Briefly, IPEC-l cells were cultured for

routine passage in DMEM/F-12 medium supplemented with 5% Fetal Bovine Serum

67

(F BS) and 1% Insulin-Transferrin-Selenium in cell culture flasks. After 5—6 days of

growth, the cells were washed in versene and trypisinized. Then, they were cultured at a
density of 3x105/ well on transwell inserts (6.5 mm diameter, 3 um pore size) that had

been previously coated with fibronectin (20 ug/ml; Sigma, St. Louis, MO). Cells were
allowed to differentiate for 10—14 days in RPMI 1640 medium containing phenol red
supplemented with 5% fetal bovine serum (F BS). Transepithelial electrical resistance
(TEER) across the monolayer was measured using an electrode (EVOMX, World
Precision Instruments, Sarasota, FL) to determine confluency and tight junction
formation.
Undifferentiated cells

IPEC-l cells in an undifferentiated state were used to mimic the epithelial cells in
the basilar crypt of the intestines. These cells, unlike the differentiated cells, do not have
tight junctions or distinct apical or baso-lateral surfaces. Since certain cytokines can be
induced only from differentiated cells and not from undifferentiated cells (33), we tested
for cytokine secretion from 2-day old undifferentiated IPEC-l cells. They were cultured
similarly to the differentiated cells on fibronectin-coated (20 ug/ml) 24-well plates, at 3x

105 cells per well and were grown to confluency (2-3 days).

Scanning Electron microscopy (SEM)

Electron microscopy was done on 12-day IPEC-l cells grown on transwells, to
verify their state of differentiation. SEM was carried out at Center for Microscopy,
Michigan State University. After 12 days of growth, media was removed and the

membranes were excised and fixed in 4% glutaraldehyde in 0.1M sodium phosphate

68

buffer. The specimens were subsequently fixed in 1% osmium tetroxide and dehydrated
in a graduated ethanol series. The dehydrated membranes were dried in a critical point
dryer (Balzers, Lichenstein), mounted on aluminum stubs, and sputter coated with 7 nm
of gold (Emscope SC500 sputter coater, UK). Visualization of the cells was canied out

with a scanning electron microscope (JEOL 6400V, Japan).

Establishing the optimal Multiplicity of Infection (MOI) for C. jejuni

Initially C. jejuni strain 33292 was used to determine the optimal bacterial
inoculum required to infect the IPEC-l cells, based an observed increases in IL-6 levels
in the medium. The bacterial strain was grown as described previously and adjusted to
0.1 OD560, yielding approximately 5 x 108 CFU/ml. This stock solution was further
diluted to yield the various MOI inocula (4:1, 40:1, 100:1 and 200:1). These dilution
stocks were then used to infect differentiated IPEC-l cells apically for 8, 12 and 24 hr. E.
coli LPS 026:B6 (200 ug/ml) was used as a positive control to induce IL-6 expression;
the optimal experimental dose was obtained from a dose response experiment. RPMI
1640 without phenol red was inoculated onto IPEC-l as the negative control. All inocula
were in 0.4 ml final volume; three replicates were used for each treatment group. After
incubation, supematants were collected, centrifuged, and analyzed for IL-6 protein by

enzyme-linked immunosorbant assay (ELISA).

Infection assays with C. jejuni

a) Measurement of secreted IL-6

69

Infection of IPEC-l with C. jejuni strains 33292 and 81176 was carried out for 8
and 24 hr time intervals with the optimum bacterial dose (40:1 MOI) obtained fi'om
previous results. To determine the direction of IL-6 secretion, differentiated IPEC-l on
transwells were challenged with C. jejuni strains apically or baso-laterally in separate
experiments. E. coli LPS 026:B6 served as the positive control and the RPMI 1640
medium without phenol red and PBS served as the negative control. Supematants were
collected as described previously and stored at -80° C until analysis by ELISA. Similarly,
undifferentiated IPEC-l cells growmg on 24-well plates were infected with the optimal
dose of C. jejuni strains for 0, 8, and 24 hr. Since these cells do not have distinct apical or
basolateral surfaces, treatments were simply exposed to cells rather than to a specific
surface as in differentiated cells. Supematants were removed from wells and analyzed for
IL-6. All treatments used for infection of both differentiated and undifferentiated cells
were performed in triplicate for each experiment.

b) Measurement of secreted IL-lO

To determine the effect of C. jejuni an IL-10 expression from IPEC-l, 10-14 day-
ald monolayers were challenged with C. jejuni strains 33292 and 81176 for 0 hr, 48 hr,
and 72 hrs. The time points were chosen based on preliminary experiments, which
showed delayed secretion of IL-10 (> 24 hr, data not shown) from C. jejuni infected
lPEC-l cells. As with IL-6, the specific direction of secretion (if any) was determined by
infecting the monolayers apically or baso-laterally. Concanavalin A (Can A; Sigma, St.
Louis, M0) at 100 ug/ml served as a positive control for IL-10 induction. The treatments
were suspended in RPMI 1640 media without phenol red and F BS, and the suspension

medium served as the negative control. All treatments were performed in triplicate.

70

Supematants were collected after the indicated intervals, processed as described
previously, and stored at -80° C until analysis. The same treatment design was used for
the undifferentiated IPEC-l cells. Supematants were removed from wells as indicated,
stored, and analyzed later for cytokine levels.
c) Measurement of secreted IL-4

Al-Salloom et al., (2) have recently demonstrated that epithelial cells are
immunopositive for IL-4 in response to C. jejuni infection. Hence, IL-4 was measured to
rule out the possibility that this Th-2 inducing cytokine was produced in our system.
Differentiated IPEC-l cells were exposed apically to the optimal C. jejuni inoculum
(40: 1) for 8 hr, 24 hr, 48 hr, and 72 hr. RPMI 1640 medium served as the negative
control. Undifferentiated cells grown on 24 well plates were challenged with the same
treatments as the differentiated cells and supematants processed as described previously.
In preliminary trials, differentiated IPEC-l cells were tested for IL-4 expression using
ConA (100ug/ml), phytohemagglutinin (PHA) (20 ug/ml, 50 ug/ml and 100 ug/ml), and
pokeweed mitogen (50% v/v) for 8 hr and 24 hrs and were shown not to induce

detectable levels of IL-4. Therefore, no positive cell treatment controls were available.

Extraction of ESP from T. suis adult worms

ESP was extracted according to the protocol described previously (20). Two to
four week old piglets were inoculated with 2500-5000 embryonated T. suis eggs. After
approximately 45 days, the pigs were humanely euthanized and their colons removed at

necropsy. The adult worms were extracted from the cecum and the proximal colon and

washed once in 0.85% saline solution, prewarmed to 37°C. The worms were then washed

71

four times in warm Hanks Balanced Salt Solution (HBSS), for 15 minutes each to remove
any fecal contamination and debris. They were then transferred to sterile culture dishes
containing RPMI 1640 with 5X antibiotics (penicillin (500 U/ml), streptomycin (0.5
mg/ml), and amphotericin B (1.25 ug/ml), Sigma Chemical Co. St.Louis, M0) at a
density of 4 warms/ml, and incubated overnight at 37°C, 5% CO;. The following
morning they were transferred to RPMI 1640 with 1X concentration of the same
antibiotics and again incubated overnight. Worms were washed fi'ee of the antibiotics in
sterile HBSS before being transferred to RPMI 1640 with 1% glucose. The culture
medium was collected every two days for 10 days. These supematants were frozen at -
80°C until required, pooled, and concentrated using an Amicon stirred cell with a 10,000
molecular weight cut-off. Different batches of ESP were standardized using Bradford

assay (8) and SDS PAGE gel of the total protein.

ESP challenge assays

Exposure of differentiated IPEC-l cells to T. suis ESP was carried out for 12 hr
and 24 hr at a concentration of 0.3 mg/ml. Higher doses of ESP have been shown to be
cytotoxic to IPEC-l (1); hence exposures were carried out at this low level dose. E. coli
LPS 026:B6 (200 ug/ml) served as a positive control for IL-6 secretion, while
Concanavalin A (ConA) (100 ug/ml) served as a positive control for IL-10 secretion. The
optimal dose of LPS was obtained from preliminary experiments (data not shown). As
mentioned before, none of the materials tested, ConA (100 ug/ml), phytohemoagglutinin
(PHA) (20 ug/ml, 50 ug/ml and 100 ug/ml) or pokeweed mitogen (50% v /v) induced

for IL-4 secretion. Hence no positive control for IL-4 was available. RPMI 1640 medium

72

without phenol red and PBS and Bovine Serum Albumin (BSA, 0.3 mg/ml) served as
negative controls. Supematants were collected after 12 and 24 hr, and processed as
before. All treatments were placed on the apical surface of the differentiated IPEC-l
cells, and experiments were performed in triplicate for each group.

Similar treatments were carried out for the undifferentiated IPEC-l cells.
Supematants were removed from wells 12 hr and 24 hr after treatment exposure,
processed as before, and analyzed later for IL-4, IL-6 and IL-10 cytokine secretions. All
treatments were in triplicate. E. coli LPS, ConA, and PHA were purchased from Sigma
Chemical Co., St. Louis, MO. Pokeweed mitogen was obtained from Gibco-BRL,

Rockville, MD.

IL-6 ELISA

Indirect ELISA was performed to measure secreted IL-6. The swine IL-6 cytokine
and primary anti-swine IL-6 antibody (rabbit polyclonal, IgG isotype) were purchased
from Biosaurce International, Carnarillo, CA. The secondary biotinylated antibody (anti-
rabbit IgG (whole molecule)), Bovine Serum Albumin (BSA), extravidin peroxidase, and
tetramethyl benzidine (TMB) were purchased from Sigma Chemical Co., St. Louis, MO.
Cell culture supematants removed from both the upper and lower chambers of the
transwell inserts (or wells in case of 24-well plates) were coated on 96-well plates and
stored overnight at 4°C. Each plate also contained IL-6 protein standard controls (in
duplicate), at 100, 50, 25, 12.5, 6.25, 3.125, 1.56, and 0 ng/ml, in RPMI 1640 without
phenol red and FBS. Following the overnight incubation, supematants were discarded

and the plates were blocked with 3% BSA solution (in Phosphate Buffered Saline (PBS,

73

0.01 M) containing 0.1% Tween 20) and incubated for 90 minutes at 37°C. Primary
rabbit anti-swine IL-6 antibody (1/ 100 titer) was used to capture the IL-6 protein (37°C,
90 rrrinutes) and a goat anti-rabbit antibody conjugated with biotin (1/5000) was used as a
secondary antibody (37°C, 90 minutes). The optimal titers for the primary and secondary
antibodies were determined from preliminary experiments (data not shown). The color
reaction was produced using extravidin peroxidase (l ug/ml, 45 minutes at room
temperature) and a ready-to-use peroxide-containing TMB substrate (30 minutes, room
temperature). The color reaction was stopped with 6N H;SO4. Plates were washed after
each step with PBS containing 0.1% Tween-20. A microplate reader (EL800 Universal
Microplate Reader, Bio-Tek instruments, Winooski, Vermont) was used to measure the
absorbance at 450 nm, and the concentration of IL-6 protein in each sample was

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

IL-10 and IL-4 ELISA

IL-10 and IL-4 were measured using sandwich ELISA kits obtained from
Biosource International (Camarillo, CA), and the assays were performed according to the
manufacturer’s protocol. Each assay contained the IL-4 and IL-10 standard protein
controls, and IL-10 and IL-4 protein were quantified (at 450 nm) as described previously

for IL-6.

Measurement of IL-6 mRNA at 0 hr by Reverse Transcriptase- Polymerase chain

Reaction (RT-PCR)

74

IPEC-l cells growing on transwells for 12 days were treated with LPS (200
pg/ml), C. jejuni strains 33292 and 81176, and RPMI 1640 media. Treatments were
removed almost immediately, and RNA was extracted using Trizol ® reagent using the
manufacturer’s instructions (Invitrogen, Frederick, MD). cDNA was constructed from 2
pg of total RNA for each sample, with the First Strand Pro-Star RT—PCR kit (Stratagene,
La Jolla, CA), according to the manufacturer’s protocol. PCR for IL-6 mRNA was
carried out using primers 5’ GGAACGCCTGGAAGAAGATG 3’, and 5’
ATCCACTCGTTCTGTGACTG 3’, and for the constitutively expressed beta-2 (0;)
microglobulin gene using primers 5’ CTGCTCTCACTGTCTGG 3’ and
5’ATCGAGAGTCACGTGCT 3’ (39). The assay conditions for the PCR were; 1 pl of
cDNA, 75ng/pl of each primer, 3 mM MgCl;, 10 mM of each deoxynucleoside
triphosphate (dNTP), and 0.625 U of AmpliTaq Gold (Perkin Elmer) in 25 pl reactions.
The PCR was run at the following therrnocycler conditions: 94°C, 10 minutes; 94°C, 1
minute for denaturing; 55°C, 1 minute for annealing; 72°C, 1 minutes for extension for
40 cycles. The PCR cycles were optimized prior to experimentation (data not shown).
The PCR products were run on a 1.8 % agarose gel, stained with ethidium bromide and
bands were visualized using the Alpha Irnager (Alpha Innotech Corporation, San

Leandro, CA).

Effect of cycloheximide (CHX) on IL-6 secretion from IPEC-l at 0 hr
The effect of inhibition of protein synthesis on IL-6 secretion was determined by
treating 12-day old IPEC-l cells on transwells with CHX. lPEC-l cells were infected

with C. jejuni strain 81176 at an MOI of 40:1 with or without different concentrations of

75

cycloheximide (100 pg/ml, 500 pg/ml and 1000 pg/ml). All treatments were suspended
in RPMI 1640 media. To determine the background levels of IL-6 in response to the
inhibitor, RPMI medium containing the different concentrations of CHX was included in
the assay. Each treatment group was in triplicate, and all treatments were removed from
the cells within 30 seconds of addition. The supematants thus removed were processed as
before and stored at -80° C until analysis by ELISA. Additional controls for CHX
inhibition were not added, as no swine reagents were available for detection of any
constitutively expressed proteins. Effect of CHX on viability of C. jejuni also was
determined by plating C. jejuni 33292 containing 100 pg/ml of CHX on Bolton agar
plates after limiting dilutions. In this viability experiment, bacteria were incubated with

CHX for ~l minute (0 hr) and two and a half hours.

Immunoblot Analysis for performed IL-6 protein

Whole cell extracts were made from 5-day old IPEC-l cells growing in tissue
culture flasks. Briefly, cell pellets were washed with Dulbeccos’ PBS (without calcium or
magnesium) (Cambrex Biasciences, Walkersville, MD), suspended in Jackson’s
extraction buffer (50 mM NaF, 20 mM HEPES pH 7.8, 450 mM NaCl, 25% glycerol, 0.2
mM EDTA) (pellet:buffer::1:1) and freeze thawed three times in liquid nitrogen-37°C
water bath. The suspension was centrifuged at 17,949 x g for 10 minutes. The supernatant
or the whole cell extract was separated from the pellet after centrifirgation, and total

protein was quantified using the Bradford assay. Different concentrations (100 pg/ml and

250 pg/ml) of the total protein were analyzed for preformed IL-6 alter electrophoresis on

76

a 15% SDS polyacrylamide gel and transfer to PVDF membrane (Millipore, Bedford,
MA). 80 ng of swine IL-6 protein (Biosource International, Carnarillo, CA) was included
in the assay as the positive control. A rabbit polyclonal antibody raised against IL-6 was
used as the primary antibody (1 :1000 dilution). An anti-rabbit antibody (Amersham
Biasciences, Piscataway, NJ) labeled with horseradish peroxidase (HRP) was used as the
secondary antibody (1:5000 dilution). The membrane was then incubated with a
chemiluminescent substrate (ECL, Amersham Biasciences, Piscataway NJ) according to

the manufacturer’s recommendations, and visualized.

Statistics
A mixed model AN OVA using pair-wise comparisons with Tukey adjustments
was used to determine the statistical significance of the cytokine concentrations obtained.

The p values were computed using the SAS 8.02 (SAS Institute Inc., Cary, NC) software.

P values S 0.05 were considered significant.

77

RESULTS

Scanning electron microscopy (SEM)

An SEM image of 12 day-old IPEC-l cells grown on transwells is shown in
Figure 2. 1. The TEER values for these cells were in the range of ~600- 800 (2. cm2.
Cells were confluent, and exhibited distinct microvilli on their apical surfaces, typical of

differentiated cells. Such cells were used in subsequent infection assays.

Sensitivity of IL-6, IL-10 and IL-4 ELISAs

The swine IL-6 ELISA was developed in the laboratory, and Figure 2. 2 shows
the sensitivity of the assay (~ 1.56 ng/ml) using the swine polyclonal anti-IL-6 antibody
(1: 100 titer). The regression curve obtained using KCjunior® software (Bio-Tek
instruments, Winooski, Vermont) showed a significant dose response curve (R = 0. 99) at
450 nm with the concentration range and antibody titer used. A similar result was seen
for the IL-10 (R= 0.99) and IL-4 (R= 1.0) standard curves as well (Figure 2. 3 and Figure

2. 4). Based on this standard curve, the assays detect 2 1 pg/ml of the IL-10 protein, and

2 2 pg/ml of the IL-4 protein.

Multiplicity of Infection assays

All MOI of C. jejuni 33292 (4:1, 40:1, 100:1 and 200:1) elicited a significantly
higher IL-6 response compared to both the positive and negative controls (Figure 2. 5).
However, statistical analysis showed that there was no significant increase in IL-6
expression between 40:1 and 100:1 MOI. Therefore a 40:1 MOI was used for all

subsequent in vitro infections. Also, there was significantly more IL-6 in the apical

78

chamber at 8 hr and 12 hr after infection in comparison to the lower chamber for most
treatment groups. However, by 24 hr after infection, there was no significant difference in

IL-6 levels between the apical and basolateral chambers.

Infection assays with C. jejuni
a) IL-6 ELISA

Figure 2. 6 shows the effect of apical and baso-lateral infection of IPEC-l cells
with C. jejuni. At 8 hr, after apical infection, IL—6 levels were significantly higher in the
apical chamber compared to the basolateral chamber in response to strain 33292 (p <
0.01), while a similar result was obtained at 24 hr with strain 81176 (p < 0.01) (Figure 2.
6A). Similarly, basolateral infections elicited a significant secretion towards the
basolateral side with both strains at 8 hr (p < 0.01) (Figure 2. 6B). There were no
statistically significant differences in IL-6 levels induced by either C. jejuni strain given
either as apical or basolateral challenges at any time point (p > 0.05).

The results of C. jejuni challenge of undifferentiated IPEC-l cells are shown in
Figure 2. 7. 0 hr (treatments removed as soon as possible; 30 seconds) was included at
this time to determine the earliest time point for IL-6 expression. Like the differentiated
cells, the undifferentiated cells also were capable of secreting IL-6 in response to C.
jejuni infection. Significant IL-6 secretion was seen within minutes to seconds of
infection (0 hr) with both strains of C. jejuni (p < 0.05) and remained elevated at 8 hr and
24 hr (p < 0.05) compared to the LPS and RPMI controls. As in bacterial challenge to
differentiated cells, there were no statistically significant differences in the IL-6 levels

from undifferentiated cells when challenged with C. jejuni strains 81176 and 33292.

79

b) IL-10 ELISA

The effect of C. jejuni an apical and baso-lateral secretion of IL-10 is seen in
Figure 2. 8. IL-10 secretion in response to apical C. jejuni infection was not seen until 24
hr (data not shown) but increased to significant levels (p < 0.01) by 72 hr (Figure 2. 8A;
later time points were not tested) after challenge with both strains of bacteria. Apical
infection produced an increase in IL-10 in comparison to the medium only control, in
both the upper and lower chambers by 72 hr (Figure 2. 8A). However, baso-lateral
infection elicited IL-10 secretion that was greater on the baso-lateral side than the apical
side, and that response was seen only with the C. jejuni strain 33292 (72 hr, p < 0.05),
Figure 2. 8B. The other strain, 81176, did not elicit a significant increase in IL-10 over
the negative medium only control when infected baso-laterally at any time.

The secretion of IL-10 from undifferentiated IPEC-l cells is shown in Figure 2. 9.
Unlike the differentiated cells, undifferentiated cells did not secrete significantly
increased amounts of IL-10 in comparison to the medium only controls at any time point,

and this result was obtained with both the C. jejuni strains tested.

c) IL-4 ELISA

There was no IL-4 secreted with any treatment from either differentiated or from

undifferentiated IPEC-l cells at all the time points tested. (Data not shown).

ESP challenge assays

IL-6 ELISA

80

IL-6 expression levels from differentiated and undifferentiated IPEC-l cells in
response to T. suis ESP are shown in Figure 2. 10 and Figure 2. 11 respectively. IL-6 was
produced fi'om both differentiated (Figure 2. 10) and undifferentiated IPEC-l cells
(Figure 2. 11), although at different levels, with the differentiated cells producing more
IL-6. IL-6 levels from differentiated cells were significantly higher in the apical chamber
than in the baso-lateral chamber at 12 hr (p < 0.001) and fell to comparable levels in both
chambers by 24 hr (p > 0.05, Figure 2. 10) in response to ESP. ESP-induced IL-6 levels
were also significantly higher than those induced by LPS, BSA and RPMI at both times
(p < 0.001). The latter two treatments did not elicit detectable amounts of IL-6 from the
differentiated IPEC-l cells.

Similarly, IL-6 levels from undifferentiated IPEC-l cells in response to ESP was
significantly higher compared to the BSA and RPMI induced levels at 12 hr (p < 0.01),
Figure 2. 11. However, unlike the results seen with differentiated cells, both BSA and
RPMI induced detectable amounts of IL-6. The high average level of IL-6 seen in

response to BSA at 24 hr from undifferentiated cells was due to a single high value.

IL-10 ELISA

The IL-10 responses from differentiated and undifferentiated IPEC-l cells due to
T. suis ESP treatment are shown in Figure 2. 12 and Figure 2. 13 respectively. Similar to
IL-6, both the differentiated and undifferentiated cells produced IL-10 in response to ESP
by 24 hrs. IL-10 levels from the differentiated cells were significantly higher in response
to both ESP and ConA compared to BSA and RPMI induced levels (p < 0.001, Figure 2.

12). However, there were no significant differences in IL-10 levels between the two

81

chambers. The undifferentiated IPEC-l cells also produced significant levels of IL-10 in
response to ConA and ESP but not in response to BSA and RPMI at both 12 hr and 24 hr

(p < 0.001, Figure 2. 13).

IL-4 ELISA
Unlike IL-6 and IL-10, there was no IL-4 produced either from differentiated or
from undifferentiated IPEC-l cells in response to T. suis ESP, at both time points tested

(data not shown). IL-4 production, if any, was below the detection limit of the assay.

RT-PCR for IL-6 mRNA at 0 hr

Since IL-6 secretion was seen within minutes of infection (if not seconds) in C.
jejuni infected cells, (Figure 2. 7), RT-PCR for IL-6 mRN A was conducted to determine
if the secretion was due to transcription. IL-6 mRNA was seen in samples treated with
RPMI, LPS, C. jejuni strains 33292 and 81176, and there were no differences in mRNA
bands irrespective of treatment groups (Figure 2.14). RT-PCR for 13; microglobulin
mRNA also showed sirrrilar band intensities for all treatment groups, indicating that total

mRN A concentrations were similar.

Effect of CHX on IL-6 secretion at 0 hr

Experiments were conducted to determine if the IL-6 secretion into the culture
supematants at 0 hr was due to rapid translation of mRNA. Cyclohexirnide, an inhibitor
of translation, was added in increasing concentrations to determine if there is a decrease

in secretion, and if so, whether it is dose dependent. As seen in Figure 2. 15, there was a

82

decrease in secretion of IL—6 into the apical chamber with the addition of CHX (100, 500
and 1000 pg/ml) with C. jejuni. However, there was no dose dependent effect, that is, no
significant suppression in IL-6 secretion in response to C. jejuni in cells treated with
increasing concentrations of CHX. Cyclaheximide alone did not elicit IL-6 secretion as
RPMI containing the various concentrations of CHX showed no significant increase in
IL-6. The decrease in IL-6 secretion in response to CHX treatment was determined not to
be a cytotoxic effect of CHX on C. jejuni, as validated by viability assays (Figure 2.16).
No significant effect of CHX on C. jejuni viability were seen either at 0 hr or 2.5hr. Also,
inhibition of protein synthesis did not completely abolish secretion as significant amounts
of IL-6 were seen in the culture supematants even after addition of 1000 pg/ml of CHX

with C. jejuni.

Immunoblot analysis

The results of imrnunoblot analysis for preformed IL-6 protein in IPEC-l cells not
exposed to any treatment, is seen in Figure 2. l7. Preformed IL-6 protein was seen in
whole cell extracts of IPEC-l cells at the two different total protein concentrations loaded
(100 and 250 pg/ml, lanes 3 and 4 respectively, Figure 2. 17). A faint band associated
with the cell pellet was also seen; however the band intensity was not as significant as the
ones from whole cell extracts (Figure 2. 17, lane 6). The lysis buffer, included as a

negative control for proteins, did not show any band as expected (lane 2).

83

DISCUSSION

Cytokines play a vital role in host defense mechanisms against a variety of
infections. The type and amount of cytokines produced by the host varies in response to
different infections and also varies within the host according to the tissue types infected
(35, 37, 41). Variations can also be host-specific, wherein different mammalian species
respond differently to infection with the same organism (10). Hence, it is vital to
determine the pattern of secretion of cytokines from a particular tissue in a host-parasite
combination, before conclusions about irnmunoregulation can be drawn. In this study, we
examined the ability of IPEC-l cells to secrete Th-2 cytokines IL-4, IL-6 and IL-10 in
response to C. jejuni infection and T. suis ESP. We hypothesized that infection with C.
jejuni would drive a predominantly Th-l cytokine response from the host. However, the
C. jejuni specific immune response should still have some Th-2 component because IgA
has been documented following the infection (32). This Th-2 response may be synergistic
with a similar response induced by T. suis, resulting in Th-2-mediated pathology. We
decided to examine the IL-4, IL-6 and IL-10 responses fi'om swine intestinal epithelial
cells, because host pathology is largely confined to the intestinal tract and because
epithelial cells are the first major barriers encountered by these enteric pathogens.

Since IPEC-l is a primary undifferentiated cell-line, we verified its state of
differentiation prior to experimentation. The SEM photographs indeed verified their
differentiated state when cultured according to our protocol (Figure 2. 1). Also, the
sensitivity of ELISA assays were validated by the use of regression curves, and we found
that IL-10 ELISA was much more sensitive in our estimate and could detect ~ 1 pg/ml

IL-lO compared to the 3 pg/ml sensitivity declared by the manufacturer (Figure 2. 3). IL-

84

6 and IL-4 ELISAs showed similar regression curves, suggesting that these assays are
very sensitive to low concentrations of IL-6 and IL-4 respectively (Figure 2. 2 and Figure
2. 4). Hence, the cytokine concentrations obtained from experimental results were used
without bias.

Both C. jejuni and T. suis ESP elicited Th2 cytokine production from [PEC-l
cells. IL-6 and IL-10 were both secreted by C. jejuni and T. suis ESP, though at different
levels and at different times after infection. IL-6 was the predominant cytokine produced
in terms of quantity, followed by IL-10. Although IL-4 is typically induced in T-cells by
helrninth infections (23), and has not been shown to be secreted by epithelial cells, we
decided to measure IL-4 to exclude the possibility that these cells had this potential.
However, not surprisingly, secretion levels of IL-4, if any, were below the detection
limits of our assay.

IL-6 was secreted in response to both C. jejuni and ESP, and IL-6 secretion levels
induced by the two agents, were not dissimilar. Also, IL-6 was secreted within a similar
time frame (i.e., < 24 hr) with respect to these treatments. While IL-6 levels secreted by
ESP were much more consistent, the levels secreted by C. jejuni varied between isolates
(Figure 2. 5, Figure 2. 6 and Figure 2. 7). However, there were no significant differences
in the IL-6 levels secreted by the two different strains of C. jejuni. Recently, Jones et a1.
(24) described infection of a TI-lP-l monocytic cell line with C. jejuni and found IL-6
induction at picogram/ml levels by 24 hr. In our study we found that IPEC-l cells were
an even greater source of IL-6, expressing this cytokine at nanogram levels within a few
minutes. In addition, the MOI of C. jejuni used in this study is likely more

physiologically relevant than the MOI of 100: lor more normally used in other studies (2,

85

24). However, since other cell lines studied so far have been of human origin, and IPEC-l
is of swine origin, the optimal dose of C. jejuni infections might vary between species.

Since C. jejuni is invasive, and since epithelial cells are likely to be exposed to
this bacterium on both the apical and/or baso-lateral surfaces, infection with C. jejuni was
carried out on both epithelial surfaces. We found that exposure of either the apical or the
baso-lateral surface to C. jejuni resulted in IL-6 secretion towards that particular surface,
similar to results obtained in another study. Kishikawa et a1 (27) found that application of
transmural pressure on either apical or baso-lateral side of epithelial cells resulted in IL-6
secretion towards that particular surface. This is an important determination for any
system, especially if autocrine and paracrine activities of cytokines in immunomodulation
are to be analyzed. Here, the ability of the IPEC-l cells to secrete IL-6 both apically and
baso-laterally suggests that while the apical IL-6 might recruit other mediators, the
basolaterally secreted IL-6 might have paracrine effects on the underlying tissues. Some
of the IL-6 seen in the baso-lateral chamber during apical infection (Figure 2. 6A) could
have been due to gravity, as cells seeded on the lower side of transwell inserts did not
show a significant level of IL-6 on the baso-lateral side when the infection was apical
(data not shown).

IL-10 was also secreted in IPEC-lcells in response to both C. jejuni and T. suis
ESP, and the levels were similar to each other, similar to their IL-6 comparison. While
IL-10 secretion by ESP was relatively fast (< 24 hr), IL-10 secretion by C. jejuni was
much slower, (> 24 hr) similar to that seen by Colgan et al., (9) who found that IL-10
levels peaked at 72 hr post CDld (a surface receptor) ligation. This suggests that the

mechanism of IL-10 induction by C. jejuni and T. suis ESP might follow different

86

pathways, and also suggests that while the primary immune response to ESP includes
both IL-6 and IL-10 (as evidenced by the time frame), for C. jejuni it is predominantly
IL-6, with a later secretion of H.-10. However, while C. jejuni 33292 induced IL-lO
secretion both apically and baso-laterally, the response to C. jejuni 81176 was
predominantly apical with no baso-lateral secretion, reinforcing the idea that variations
exist in host cell responses to different Campylobacter strains. The presence of this
predominantly apical IL-10 could lead to autocrine effects since epithelial cells have been
shown to possess IL-10 receptors (3, 9).

Since some cytokines are produced from cells in a differentiated state and not in
an undifferentiated state, we also tested to see if crypt-like cells could produce these
cytokines in response to C. jejuni and T. suis ESP challenge. We found that IL-6 was
secreted by undifferentiated IPEC-l cells in response to C. jejuni, but not IL-10 or IL-4.
This is contrary to the results reported by Al Salloom et al., (2) who found IL-4 and IL-10
intracellular cytokine production fi'om 3—5 day old INT407 cells, a human intestinal
epithelial cell line, by 24 hr in response to C. jejuni. However, this contrast might be due
to the fact that they measured intracellular cytokines rather than secreted cytokines as in
our study. Both IL-6 and IL-10 were induced from undifferentiated IPEC-l cells in
response to T. suis ESP, but not IL-4. In fact, we found no evidence of IL-4 secretion in
our system, at any of the times tested, with neither C. jejuni or T. suis, nor with either
states of cell differentiation.

Another interesting finding in this study was the short time span for IL-6 secretion
by the epithelial cells. IL-6 was seen in the culture supematants within minutes, if not

seconds of C. jejuni exposure, but not in the medium only controls (Figure 2. 7). While

87

the processing time between removal of supernatant and storage (10—15 min) could
account for some of the protein secretion, it is still relatively rapid compared to that
reported in other cell lines. We tested the possibility that this rapid secretion could be due
to transcription of IL-6 mRNA. However, as seen in Figure 2. 14, mRNA was seen in
RPMI medium treatments as well, and there were no difference in band intensities with
any treatments, suggesting no new transcription. This is in accordance with other results,
where IL-6mRNA transcription has been shown to occur only after 30 minutes or more
(40). Next, we tested the possibility that translation of this preexisting IL-6 mRNA could
account for this rapid secretion. However, while translation accounted for some of the
secretion, it could not account for the entire amount seen, as significant levels were still
secreted even after inhibition of translation with CHX (Figure 2. 15). And this inhibition
of translation was not due to a cytotoxic effect of CHX on C. jejuni (Figure 2. 16).
However, these levels could be accounted for by the presence of preformed IL-6, as
evidenced by Western blot analysis (Figure 2. 17). Significant quantities of IL-6 were
seen within lPEC-l cells, indicating that the rapid IL-6 secretion in response to C. jejuni
(and T. suis ESP) was due to a combination of translation and release of stored IL-6. This
appears to be the first demonstration of stored IL-6 cytokine in intestinal epithelial cells.
Since these intestinal cells are the first barriers encountered by intestinal pathogens, the
production and storage of this immune mediator might provide the host with an
immediate defense against enteric infections.

The role of epithelial cells as an important source and initiator of enteric immune
responses is a recent concept. Since epithelial cells cover a large surface area, secretion of

even picogram amounts of cytokines could make a significant contribution to the overall

88

immune response. IPEC-l is a non-immortalized cell line, hence the pattern of secretion
should be much more reflective of the physiology in the host than that represented by an
immortalized cell line. While IL-10 has been documented to act as an anti-inflammatory
cytokine, whether IL-6 behaves in a proinflammatory or anti-inflammatory manner in
terms of nitric oxide modulation remains to be tested in our system. Also, as IPEC-l cells
did not secrete IL-4, the question of whether C. jejuni or T. suis ESP elicits TL-4 from IL-
4 competent cells like macrophages or T-cells still remains to be investigated. However,
the production of IL-6 and IL-10 by IPEC-l in response to C. jejuni and T. suis ESP
challenge supports the hypothesis that the concurrent presence of C. jejuni and T. suis in
swine may lead to a Th-2-like environment with subsequent Th-2-mediated pathology.
This is also strengthened by the fact that concurrent studies conducted in the laboratory
show no induction of IL-1 [3 or IFNy secretion from (undifferentiated) IPEC-l cells in
response to C. jejuni infection by 24 hr (Cunningham L. D and LS. Mansfield,
unpublished data). Because IL-6 was secreted within a similar time fiame with respect to
these two pathogens, there might be synergy in IL-6 levels in viva. Furthermore, the
production of these cytokines and storage of preformed IL-6 within these cells suggest
that epithelial cells play a vital role in the induction of inflammation during pathogenic

infections in swine.

89

ACKNOWLEDGEMENTS
This work was supported by the NIH-grant, 61-0954, and the Center for Emerging

Diseases grant 61-6387. We thank Dr. Julia Bell for critical review of this article.

90

10.

11.

REFERENCES

Abner, S. R., D. E. Hill, J. R. Turner, E. D. Black, P. Bartlett, J. F. Urban,
and L. S. Mansfield. 2002. Response of intestinal epithelial cells to T richuris
suis excretory-secretory products and the influence on Campylobacterjejuni
invasion under in vitro conditions. J Parasitol 88:738-45.

Al-Salloom, F. S., A. A1 Mahmeed, A. Ismaeel, G. A. Botta, and M. Bakhiet.
2003. Campylobacter-stimulated INT407 cells produce dissociated cytokine
profiles. J Infect 47 :217-24.

Autschbach, F., J. Braunstein, B. Helmke, I. Zuna, G. Schurmann, Z. I.
Niemir, R. Wallich, H. F. Otto, and S. C. Meuer. 1998. In situ expression of

interleukin-10 in noninflarned human gut and in inflammatory bowel disease. Am
J Pathol 153:121-30.

Beer, R. J. 1973. Studies on the biology of the life-cycle of T richuris suis
Schrank, 1788. Parasitology 67:253-62.

Beer, R. J., B. F. Sansom, and P. J. Taylor. 1974. Erythrocyte losses from pigs
with experimental T richuris suis infections measured with a whole-body counter.
J Comp Path0184:331-46.

Berschneider, H. M. 1989. Development of normal cultured small intestinal
epithelial cell lines which transport Na and Cl. Gastroenterology 96:A4l.

Blaser, M. J., D. N. Taylor, and R. A. Feldman. 1983. Epidemiology of
Campylobacterjejuni infections. Epidemiol Rev 5:15 7-76.

Bradford, M. M. 1976. A rapid and sensitive method for the quantitation of
microgram quantities of protein utilizing the principle of protein-dye binding.
Anal Biochem 72:248-54.

Colgan, S. P., R. M. Hershberg, G. T. Furuta, and R. S. Blumberg. 1999.
Ligation of intestinal epithelial CDld induces bioactive IL-10: critical role of the
cytoplasmic tail in autocrine signaling. Proc Natl Acad Sci U S A 96:13938-43.

Crabtree, J. E., M. Court, M. A. Aboshkiwa, A. H. Jeremy, M. F. Dixon, and
P. A. Robinson. 2004. Gastric mucosal cytokine and epithelial cell responses to
Helicabacter pylori infection in Mongolian gerbils. J Pathol 202:197-207.

de Man, P., C. van Kooten, L. Aarden, I. Engberg, H. Linder, and C.

Svanbarg Eden. 1989. Interleukin-6 induced at mucosal surfaces by grarn-
negative bacterial infection. Infect Irnmun 57:3383-8.

91

12.

13.

14.

15.

l6.

17.

18.

19.

20.

21.

22.

23.

Dimack, K. A., D. G. Addiss, M. L. Eberhard, and P. J. Lammie. 1994.
Differential praliferative and interleukin-10 responses to fractionated filarial
antigens: preferential recognition by patients with chronic lymphatic dysfunction.
J Infect Dis 170:403-12.

Finkelman, F. D., E. J. Pearce, J. F. Urban, Jr., and A. Sher. 1991. Regulation
and biological function of helrninth-induced cytokine responses. Immunol Today
12:A62-6.

Fireman, P. 2003. Understanding asthma pathophysiology. Allergy Asthma Proc
24:79-83.

Friedman, C., Neimann,], Wegener,HC and Tauxe,RV. 2000. Epidemiology
of Campylobacterjejuni infections in the United States and other Industrialized
Nations, p. 121-138.11: N. I. a. B. M (ed.), Campylobacter. ASM.

Gonzalez-Vallina, R., H. Wang, R. Zhan, H. M. Berschneider, R. M. Lee, N.
0. Davidson, and D. D. Black. 1996. Lipoprotein and apolipoprotein secretion
by a newborn piglet intestinal cell line (IPEC-l). Am J Physiol 271:G249-59.

Hatzigeorgiou, D. E., S. He, J. Sobel, K. H. Grabstein, A. Hafner, and J. L.
Ho. 1993. IL-6 down-modulates the cytokine-enhanced antileishmanial activity in
human macrophages. J Immunol 151:3682-92.

Hedges, S., P. Anderson, G. Lidin-Janson, P. de Man, and C. Svanbarg. 1991.
Interleukin-6 response to deliberate colonization of the human urinary tract with
gram-negative bacteria. Infect Irnmun 59:421-7.

Hickey, T. E., S. Baqar, A. L. Bourgeois, C. P. Ewing, and P. Guerry. 1999.
Campylobacterjejuni-stimulated secretion of interleukin-8 by M407 cells.
Infect Irnmun 67:88-93.

Hill, D. E., H. R. Gamble, M. L. Rhoads, R. H. Fetterer, and J. F. Urban, Jr.
1993. T richuris suis: a zinc metalloprotease from culture fluids of adult parasites.
Exp Parasitol 77:170-8.

Hirano, T., and T. Kishimata. 1989. Interleukin-6: possible implications in
human diseases. Ric Clin Lab 19:1-10.

Janeway, C. T, Walport, M and Capra JD. 1999. T-cell mediated immunity, p.
7:1- 7:41. In M. a. L. Robertson, E (ed.), Immunobiology-The Immune system in
health and disease, Second ed. Current Biology Ltd and Garland Publishing Co

Janeway, C. T., P; Walport,M and Capra JD. 1999. The Humoral Immune
Response, p. 307-362. In P. a. L. Austin, E (ed.), Irnmuno Biology-The Immune

92

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

System in Health and Disease, Four ed. Current Biology Publications and Garland
Publishing.

Jones, M. A., S. Tatemeyer, D. J. Maskell, C. E. Bryant, and P. A. Barrow.
2003. Induction of proinflammatory responses in the human monocytic cell line
THP-l by Campylobacterjejuni. Infect Irnmun 71 :2626-33.

Kealy, W. F. 1976. Colonic lymphoid-glandular complex (microbursa): nature
and morphology. J Clin Pathol 29:241-4.

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

Kishikawa, H., S. Miura, H. Yoshida, M. Hirokawa, H. Nakamizo, H.
Higuchi, M. Adachi, R. C. Nakatsumi, H. Suzuki, H. Saito, and H. Ishii. 2002.
Transmural pressure induces IL-6 secretion by intestinal epithelial cells. Clin Exp
Immunol 129:86-91.

Lawrence, C. E., J. C. Paterson, L. M. Higgins, T. T. MacDonald, M. W.
Kennedy, and P. Garside. 1998. IL-4-regulated enteropathy in an intestinal
nematode infection. Eur J Immunol 28:2672-84.

Liebler, E. M., J. F. Pohlenz, and N. F. Cheville. 1988. Gut-associated
lymphoid tissue in the large intestine of calves. H. Electron microscopy. Vet
Pathol 25:509-15.

Mansfield, L. S., D. T. Gauthier, S. R. Abner, K. M. Jones, S. R. Wilder, and
J. F. Urban. 2003. Enhancement of disease and pathology by synergy of

T richuris suis and Campylobacterjejuni in the colon of immunologically naive
swine. Am J Trop Med Hyg 68:70-80.

Mansfield, L. S., and J. F. Urban, Jr. 1996. The pathogenesis of necrotic
praliferative colitis in swine is linked to whipworm induced suppression of
mucosal immunity to resident bacteria. Vet Immunol Immunopathol 50:1-17.

Mansfield, L. S. a. D. G. 2004. Lyrnphoglandular complexes are important
colonic sites for IgA induction against Campylobacterjejuni in a swine disease
model. Comparative medicine.

Marcelo, P., and F. Lefevre. 2002. Induction of the IFN-gamma gene and
protein is linked to the establishment of cell polarity in a porcine epithelial cell
line. Exp Cell Res 280:33-44.

Morfitt, D. C., and J. F. Pohlenz. 1989. Porcine colonic lymphoglandular
complex: distribution, structure, and epithelium. Am J Anat 184:41-51.

93

35.

36.

37.

38.

39.

40.

41.

Mukherjee, P., A. K. Ghosh, and A. C. Ghose. 2003. Infection pattern and
immune response in the spleen and liver of BALB/c mice intracardially infected
with Leishmania danavani amastigotes. Immunol Lett 86: l 3 1-8.

Napolitano, L. M., M. M. Buzdon, H. J. Shi, and B. L. Bass. 1997 . Intestinal
epithelial cell regulation of macrophage and lymphocyte interleukin 10
expression. Arch Surg 132:1271-6.

Nedrebo, T., R. K. Reed, R. Jonsson, A. Berg, and H. Wiig. 2004. Differential
cytokine response in interstitial fluid in skin and serum during experimental
inflammation in rats. J Physiol.

O'Leary, A. D., and E. C. Sweeney. 1986. Lymphoglandular complexes of the
colon: structure and distribution. Histopathology 10:267-83.

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

Zhang, J., K. Xu, B. Ambati, and F. S. Yu. 2003. Toll-like receptor 5-mediated
corneal epithelial inflammatory responses to Pseudamanas aeruginasa flagellin.
Invest Ophthalmol Vis Sci 44:4247-54.

Zhang, J. Z., M. Sinha, B. A. Luxon, and X. J. Yu. 2004. Survival strategy of

obligately intracellular Ehrlichia chajfeensis: novel modulation of immune
response and host cell cycles. Infect Irnmun 72:498-507.

94

Figure 2. 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.

 

95

Figure 2. 2. Standard curve for swine IL-6 used in the ELISA assays. Curve is

representative of multiple experiments (~10).

 

 

 

 

 

2.54- ' ///
11- '//
2.01- //
.'/
/’f
g 1.5" “///
I //6
1 01 //,,
./
05 // .
“ %
Mts+ir+rrsiireirre4
1O 20 30. 40 50 60 70 80 90 100
nglml
Ltnoar(y-Ax+B)
A=0.0272 B=-0.0521. R—Square 8 0.9934
0 Standards
0 Suppressedstda
Standard Curve

 

 

 

96

Figure 2. 3. Standard IL-10 curve used in ELISA assays. Figure 2. 3A is the entire
standard range, and Figure 2. 3B is the magnification of the range from 0 to 60 pg/ml in
Figure 2. 3A. Curve is representative of multiple experiments (~10).

Figure 2. 3A

 

I

2.5‘

//'I/
,/

2.0+ r
/

I
v
.0

1.5“ ./

nm
\\

0.5+ //

1%"

20 4o 60 so 100 120 140 160 180 200 220 240
pglml

 

 

4 Paramotar(y= (A- D)I(1+()dc.)"B)+ D)
A=0.1611 B=1.0217 C=580.5951 088.3169, R-Square = 0.9999

 

 

 

0 Standards
0 Suppressed Stds
Standard Curve

 

97

nm

Figure 2. 3B

 

0.9 t

0.8:

0.6"

0.4 "

0.3‘

0.7 r

0.5‘

 

0.2‘

20

palm!

4 Parameterbrs (A-D)I(1 + (x/C)"B) + D)
A=0.1437 B=O.&811 C=12098.715 D=80.0527. R—Square = 0.9981

 

 

 

0 Standards
0 Suppressed Stds
Standard Curve

 

98

 

Figure 2. 4. Standard curve for swine IL-4 used in ELISA. The curve is representative of

multiple experiments (3-4).

 

oe- 3

0.8 ‘

I
.‘\

0.7 " ,/"

ae~ /
e ’4'
41-

"m
p
‘i'
\

\
\

0.4" /

 

 

I
‘
fit 1
L
1L

 

O 100 200 300 400 500
pglml

4 Parameter (y = (A - D) I (1 4* (xlCY‘B) 7+ D)
A=0.0135 B=1 .0818 C=8877.6723 D=20.8354, R-Square = 1.0000

0 Standards
9 Suppressed Std
Standard Curve

 

 

 

 

99

Figure 2. 5. Effect of different MOI on IL-6 expression fi'om IPEC-l cells. 12-14-day-
old IPEC-l cells were infected apically with different MOI (4:1, 40:1, 100: l and 200:1)
of C. jejuni 33292 for 8, 12 and 24 hr, and supematants from both chambers were
analyzed for IL-6 by indirect ELISA. E. coli 026:B6 LPS (200 pg/ml) was used as a
positive control and RPMI medium as the negative control. All treatments were

performed in triplicate, and are representative of two identical experiments.

 

250 ..,

 

 

 

 

 

 

 

 

 

 

 

 

 

 

BCj 33292 4 to 1
no; 33292 40 101 i
__*M. “A ——-—- an 33292 100 101 i

le 33292 200101 '
A
E . LPS i
2:” L'l.RPM1__ b J
v
3 T
m

.1- 2
8 12 24
apical baso-lateral
Time (hr)

100

Figure 2. 6. The secretion of IL-6 due to apical (6A) and baso-lateral (6B) C. jejuni (Cj)
infections is shown. Differentiated IPEC-l cells were infected with the optimum MOI of
C. jejuni (40: 1) obtained from the experiment shown in Figure 2. 5. The results shown are
representative of three identical experiments for Figure 2. 6A, and two identical

experiments for Figure 2. 6B, with three replicates in each group. ND- not determined.

101

IL-6 (ng/ml)

IL-6 (ng/ml)

1 00
90
80
70
60
5O
40
30
20

100

 

 

Figure 2. 6A

"I RPM 1
, I LPS
UCj 33292 ‘
Ia 9131 17,55

124

l

Basolateral
Time (hr)

Figure 2. 6B

‘ITRPMT
1135
pg 33292
ICj 8171764

1
I

l gaical i basolateral

Time (hr)

102

Figure 2. 7. IL-6 secretion from 2-day old undifferentiated IPEC-l cells. 3 x 105 cells
were seeded on fibronectin-coated 24-well plates and grown to confluency for 2 days in
RPMI 1640 medium containing phenol red and supplemented with 5% FBS. The
treatments were suspended at optimal concentrations in RPMI 1640 medium without any
supplements, applied to the cells, and allowed to incubate for 0, 8, and 24 hr.
Supematants were removed, processed, and analyzed as described in the materials and

methods. Mean of 2 identical experiments is shown. The values shown are mean +/-

 

 

 

 

 

 

 

 

 

SEM.

100

90

80

70
E EIRPMI fl,
5’ 1213:3292 -
E may

 

 

 

 

 

 

 

 

 

 

Time (hr)

103

Figure 2. 8 shows IL-10 expression from IPEC-l cells in response to C. jejuni (Cj)
infection apically (A) and baso-laterally (B), respectively. Con A (100 pg/ml) served as
the positive control and RPMI 1640 medium served as the negative control. The values
are mean of two (apical infection) and three (basolateral infection) identical experiments

respectively, and mean +/- SEM is shown. ND- not determined.

104

Figure 2. 8A

 

l

 

 

 

 

 

 

 

liRPMI— i i
' Ele33292 F
(181176 l

3 I

on

E:

O

v-1

9.

 

 

 

 

 

 

 

 

 

 

 

I apical I baso-lateral
Time (hr)
Figure 2. 8B
30 -- _w _..
.RPMI !
25 -E__ _ A it ~*. M___E_‘ __ A _IConA in
ch33292;
le 81176,
20 -—~ ~—

 

 

IL-10 (pg/ml)
Ci

 

 

 

 

 

 

 

10
5 ..E__ __ .. ._ __
A i'
0 LT” ‘ r:
0 48 l 48
apical I baso-lateral

 

Time (hr)

105

Figure 2. 9. IL-10 expression from undifferentiated IPEC-l cells in response to C. jejuni
(Cj) infection. 2 day-old IPEC-l cells were challenged with optimal dose of C. jejuni for
0, 48 and 72 hr, supematants removed, and analyzed for IL-10 by a sandwich ELISA.

Results shown are the mean of three identical experiments. Values are mean +/- SEM.

 

 

 

 

 

: - g .
g DRPMI
g UConA
V

c ‘12le 33292
"‘ IC'81176i
A ., ,1, A4
H

 

 

 

 

 

Time (hr)

106

Figure 2. 10. IL-6 secretion from differentiated IPEC-l cells in response to Excretory
Secretory Products (ESP) of T. suis. 10-14 day old IPEC-l cells on porous supports were
exposed apically to 0.3 mgml of the ESP, E. coli LPS (200 pg/ml), BSA (0.3 mg/ml) and
RPMI medium. LPS served as the positive control, while BSA and RPMI medium served
as negative controls for IL-6 secretion. Supematants were extracted as described in the
materials and methods and analyzed for IL-6 cytokine expression by indirect ELISA. All
treatments were in triplicate. Values shown are mean +/- SEM, and are representative of

two identical experiments. ND- not determined.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

250 .
g I
200
E 150 T 'I— illlRPMl
in ’t iILPs l
' l
v IDES?
3 100 ~— i-BSA I
h! _ _____.
50 r“
i ND ND
0 _, , ._ , . .
l T . I I
12 , . 24 f i 12 3 g 24 l
i api al I baso-lateral
Time (hr)

107

Figure 2. 11. Two-to-three day-old IPEC-l cells grown to confluency were challenged
with 0.3 mg/ml of T. suis ESP, 200 pg/ml of LPS, BSA (0.3 mg/ml) and RPMI 1640
medium for 12 and 24 hr. The supematants were collected as described and analyzed by
an indirect ELISA for IL-6 secretion. Values shown are mean +/- SEM, and

representative of two identical experiments.

 

 

 

 

 

250 1

20° REM—Ii
I LPS 1
DES? i

150 __B_Sé___-

 

50

 

 

 

 

 

 

12

Time(hr)

108

Figure 2. 12. T. suis ESP (0.3mg/ml) induced IL-10 expression from differentiated IPEC-
1 cells. Confluent lPEC-l cells on Transwell membranes were exposed apically to T. suis
ESP (0.3 mg/ml), RPMI, Concanavalin A (ConA), and BSA for 12 and 24 hr. ConA

(100 pg/ml) served as the positive control for lL-10 expression while BSA (0.3 mg/ml)
and RPMI medium served as the negative controls. A sandwich ELISA was used to
analyze the supematants for IL-10 secretion. Values shown are mean +/- SEM, and

representative of two identical experiments.

 

 

 

 

 

 

 

25
20
E 15 I IIIIRPMI
g IConA
3 DESP
"' ZBSA
a 10 - ,
5 1'
0 T ' l ’ T ' " T T
12 24 l 12 . 24
I
apical , basolateral

 

Time (hr)

109

Figure 2. 13. The effect of T. suis ESP on IL-10 expression from undifferentiated (crypt-
like) IPEC-l cells is shown. As before, two-to-three day-old IPEC-l cells were
challenged with ConA (100 ug/ml), ESP (0.3 mg/ml), BSA (0.3 mg/ml) and RPMI 1640
medium for 12 and 24 hr, supematants removed after the indicated time intervals and
later analyzed for IL-10 secretion. All treatments were in triplicate, and are representative

of two identical experiments. ND- not determined.

 

 

 

 

 

 

 

 

25

20 j
: HEBRPMI‘
g 15 ————€IConA
g ‘UESP
o IBSA ‘
3 10- ,_
H

5 .4

ND ND ND ND .
0 - __§_l
12 24

Time (hr)

110

Figure 2. 14. RT-PCR for IL-6 mRNA at 0 hr. RNA was extracted from 12 day-old
IPEC-l cells on Transwells, at 0 hr. cDNA was made from 2 pg of total RNA for each
sample. PCR was performed for IL-6 cytokine and the constitutively expressed [32
microglobulin. Top gel is RT-PCR for IL-6 mRNA, bottom gel for [32 microglobulin
mRNA. Lane 1- 100 base pair ladder; lanes 2-3- RPMI treated IPEC-l; Lanes 4-5-LPS
treated cells; Lanes 6-7— C. jejuni 33292 treated cells; lanes 8-9— C. jejuni 81176 treated
cells. Lane 10— no RNA control. All treatments were in duplicate. Bands are

representative of two identical and independent experiments.

 

47ohp IL-s
Beta-2

254bp —mcm_
globulin

111

Figure 2. 15. Effect of cycloheximide on IL-6 secretion from IPEC-l cells, in response to
C. jejuni 33292 and RPMI 1640 medium. The bacteria were treated with or without
different concentrations of cycloheximide (100, 500 and 1000 uyml) at 0 hr.
Supematants were removed immediately, processed as described previously, and
analyzed for IL-6 by ELISA. RPMI medium was treated similarly and used as control.

Values shown are mean +/- SEM, and are representative of two identical experiments.

 

 

 

 

 

 

 

IL-6(nglml)

 

 

 

 

 

 

 

N "‘ X X X X X
a $2 55 £25 5 55 £5 a
s; ggé :3 3‘6 E :“5 2‘5 23
g .580 fig 5.: 3 SE SE SE
-. a; '.§ ~15 & at; H“ g?
o 0"; oo 0' o 8 8

In 0 ‘- ID 0

Treatmentgroups

112

Figure 2. 16. Effect of CHX on viability of C. jejuni. Strain 33292 was incubated with
100 11le of CHX for ~ a minute (0 hr) or 2.5 hrs. Bacterial cells were diluted by
limiting dilutions, plated and enumerated on Bolton agar plates for 48 hr at 37 C and 5%

C02. Results are data from a single experiment.

 

l

 

 

 

4.5 7 , 7 7 77* ,7. 7. ‘1
4 ECijz 7‘77 T
:ICj 3323mm IOOuflil CH3“
3.5

.L L“- _ L; j
l 7 1

‘ 4

 

 

colony count (x 10*7 cfu/ml)
N

 

CFU at 0 hr CFU after 2.5 hr
Time

113

Figure 2. 17. Western blots showing preformed IL-6 protein in IPEC-l cells prior to any

treatment. IL-6 standard was used at a concentration of 80 ng. Lane—1, IL-6 standard;
Lane 2, lysis buffer; lane 3, IPEC-l whole cell lysate (100 pg of total protein); lane 4,
IPEC-l whole cell lysate at (250 pg of total protein); lane 6- cell pellet alone (10 pl,

concentration not determined). Blots are representative of two identical experiments.

IPEC calla

 

.. a A. .3
a2 3’5 “is.
r—""l

. ‘9‘”

1 2 3 4 5 6

114

CHAPTER 3
IL-4, IL-6 AND IL-lO CYTOKINE RESPONSES IN SWINE INFECTED WITH

CAMPYLOBA C TER JEJUNI AND T RICH URIS S UIS

115

ABSTRACT

Campylobacterjejuni is one of the leading causes of bacterial enteritis in the
world. It is a gram-negative, spiral or comma-shaped bacterium generally acquired
through contaminated food. T richuris sp. is a helrninth that dwells in the cecum and
proximal colon of swine and humans and is commonly found in all parts of the world.
Germ-free piglets infected with either C. jejuni or T. suis display few clinical signs of
disease. In contrast, dual infection results in bloody diarrhea and severe pathology with
multifocal lesions, infiltration of inflammatory cells, excessive mucus secretion, and
bacterial invasion into epithelial cells and macrophages. In this study, we examined the
immune response in swine to single and dual C. jejuni and T. suis infections. Experiments
were conducted in two ways. In the long-term experiment, 4-5-week old conventionally
reared pigs were orally inoculated with T. suis (2500 eggs). After 21 days, half were
orally inoculated with C. jejuni, and half with milk. At this time, additional groups,
received either C. jejuni, E. coli DHSa or milk. In the short-term experiment, pigs were
inoculated concurrently with either C. jejuni, T. suis or with both. Control pigs received
E. coli DHSa or milk. Feces were collected daily, and fecal supematants were analyzed
for IL-4, IL-6 and IL-10 cytokines by ELISA. In the long-term experiment, 24 hours after
infection with C. jejuni, pigs previously infected with T. suis showed elevated levels of
IL-4 and IL-10 compared to pigs receiving milk alone. IL-lO, but not IL-4 was elevated
in pigs singly infected with T. suis. Pigs singly infected with C. jejuni did not exhibit
significantly elevated IL-4 and IL-10 responses by our evaluation methods. Also, there
were no significant differences in IL-6 cytokine levels in any of the groups. Conversely,

in the short-term experiment, pigs dually infected with C. jejuni and T. suis and pigs

116

singly infected with T. suis or C. jejuni, showed elevated IL-6 responses but not IL-4 or
IL-10 responses. However, pigs that received milk also showed elevated IL-6 responses
in addition to a significant IL-4 response, suggesting that IL-6 elevations in the dual
infected pigs are not important. Pigs that received E. coli DHSa had elevated IL-4
responses in their feces. Hence, no cytokines were considered significant during the
short-term infections for the dual infection pigs. Because in natural infections disease is
seen 21 days after introduction of T. suis in the lot, cytokine responses in the long-term
infections are likely more reflective of the disease state. Hence this study suggests that
IL-4 and possibly IL-lO play a role in progression of T. suis and C. jejuni induced disease
and mucohemorrhagic diarrhea. Also, we show that measurement of adaptive cytokines
like IL-4 and IL-10 from feces is a non-invasive tool to study anti-inflammatory cytokine

changes in the host.

117

INTRODUCTION

T richuris suis are parasites of the swine intestinal tract and cause a clinical
syndrome called trichuriasis in pigs marked by anemia, anorexia, pronounced weight
loss, bloody diarrhea and colitis (3). In his pioneering work on the life cycle of T. suis,
Beer suggested that the burrowing activities of T. suis into the cecum and colon might
enable secondary bacterial infections to occur, thereby enhancing the pathology (4).
Antibiotic experiments conducted in conventionally reared weaned piglets suggested that
a bacterial component might be involved in the pathology. 7—8 week old pigs given a low
dose of T. suis (3000 eggs) along with antibiotics did not develop disease symptoms (18).
On the other hand, swine that received T. suis eggs alone developed lesions and bloody
diarrhea similar to that seen in natural infections. Campylobacterjejuni, a gram-negative
spiral rod, was subsequently isolated from the distal colon of these pigs. In follow-up
experiments, gnotobiotic germ-free piglets infected with both T. suis (2500 embryonated
eggs) or C. jejuni (106 cfu) developed bloody diarrhea and colonic pathology, while
piglets infected with either agent singly did not. Numerous bacteria were seen around the
sites of worm attachment in the proximal colon and also in the lymphoglandular
complexes (LCGs) in the distal colon of dual infected pigs. LGCs are secondary
lymphoid organs present in the distal colon of swine, humans, rats, cattle, and other
animals (13, 15, 20, 22) . They are structurally similar to the Peyer’s patches in the small
intestine and in the sensitized host are composed of a follicle-associated epithelium
(F AE) overlying B-cell—dependent germinal centers. LGCs appear to be an important

route for C. jejuni invasion.

118

We hypothesized that cytokine dysregulation might be a contributing factor to this
polymicrobial disease phenomenon. Th-2 cytokines like IL-4, IL-5, and IL-10 are
induced in the host in response to hehninth infections (9). Bacterial infections typically
induce proinflarnmatory cytokines marked by IL- 12, TNF-a, IL-1 [3, IFN-y and others,
including IL-6 (12). In the dual infection model there is both a helrninth component and
a bacterial component. Also, manifestations of disease with clinical signs occur only in
the presence of bacteria (17, 18). Therefore we posed the following hypotheses. 1)
Resistance to T. suis infection involves a predominantly Th-2 immune response while
resistance to C. jejuni infection involves a predominantly Th-l immune response.
However, the host resistance to C. jejuni also has a Th-2 component, since IgA is
generated in the LGCs during C. jejuni infections (19). 2) The immune response is shified
towards a predominantly Th-2 type during polymicrobial infections with T. suis and C.
jejuni, contributing to the clinical symptoms observed.

To evaluate these hypotheses, two different experiments were conducted. In short-
terrn experiments 4—5 week old conventionally reared weaned piglets were infected with
either T. suis (2500 embryonated eggs) or C. jejuni (2 x 108 0111) or both. In long-term
experiments, pigs were infected with T. suis eggs for 21 days prior to infection with
either C. jejuni or milk. Additional pigs received C. jejuni alone, and IL-4, IL-6 and IL-
10 cytokine responses were measured in the feces. Control pigs received Escherichia coli
DHSa or milk in both experiments. IL-4 and IL-1 0 have been shown to inhibit Th-l
cytokine responses, while dendritic IL-6 has been recently shown to shifi cytokine
responses to a Th-2 type (6). Additionally, IL-6 is a switch factor for IgA production. The

results from this study show that Th-2 cytokines, IL-4 and IL-10, are likely to play a role

119

during long-term dual infection of T. suis and C. jejuni in swine, while no significant

cytokine responses were seen during short-term dual infections.

120

MATERIALS AND METHODS

Bacterial cultures

C. jejuni strain ATCC 33292 and the non-invasive control E. coli DHSa were
used for the experimental procedures. C. jejuni strain 33292 was initially obtained from a
human with enteritis, and passaged through a pig prior to experimental procedures. The
C. jejuni strain was grown at 37 °C, 5% C02 on Bolton agar supplemented with 5%
Sheep’s blood. The E. coli strain was grown on Bolton agar plates at 37°C without
Sheep’s blood. Both bacteria were grown for 18—20 hours, and resuspended in sterile

milk to a final concentration of 1 x 108 cfu/ml.

T. suis embryonated eggs

T. suis eggs were collected from adult worms according to previously published
methods (10), centrifuged 2—3 times at 10,000 X g for 5 minutes to remove the debris.
The eggs were then suspended in deionized distilled water and incubated at 34°C, 5%
C02 for nineteen days. On day 19, 10 mg/ml of porcine bile extract (Sigma Chemical
Co., St. Louis, MO) was added and the eggs were incubated for an additional 3—5 days.
Through this additional incubation process, the eggs were monitored for movement inside
the eggs indicative of embyronation. The embryonated eggs were centrifuged for 10

minutes at 10,000 X g, resuspended in sterile water and stored at 4°C until required.

Animals
Four-to-five—week-old weaned, conventionally reared Duroc/Yorkshire/Landrace

piglets were obtained from MSU swine farm, housed in groups of 5 or 6 pigs each in 5

121

separate rooms at the University Research Containment Facility, and provided with feed
and water ad libitum. The rooms were sterilized with bleach prior to housing of pigs. The
handling and management of the animals were performed in accordance with University

and NIH guidelines.

Identification of T. suis eggs and C. jejuni from the feces prior to inoculations

To ensure that swine were specific pathogen free, prior to inoculations feces were
collected fi'om the pigs and analyzed for T. suis eggs by fecal floats. Briefly, ~5 g of feces
were collected from each animal and transported in Cary-Blair medium (Difco
Laboratories, Detroit, M1) on ice. In the laboratory, the feces was mixed with ~ 10ml of
water and centrifuged in a 13 X 100 mm test tube for 4-5 minutes at 2000 rpm. The
supernatant was discarded and the fecal pellet was resuspended in 1.2% sucrose solution
containing 0.5% formaldehyde. The tube was centrifuged again at 2000 rpm for 4-5
minutes, filled with sucrose solution with a cover slip placed on top. After 5 minutes, the
coverslip was removed and observed under a compound microscope at 40X
magnification for the presence of T. suis eggs. Presence, or absence of C. jejuni was
analyzed by flaA gene PCR (21) and Taq Man® assays (25). Briefly, fecal swabs
obtained from pigs were plated on Bolton agar plates containing 5% sheep’s blood, and
incubated at 37° C, and 5% C02. DNA was obtained from bacterial colonies using Easy-
DNA® kit from Invitrogen (Carlsbad, CA) and PCR assays were carried out as described

(21, 25).

Animal Inoculation

122

The animals were inoculated orally with the different treatment groups as
indicated in Table 3. 1A and Table 3.13. All the treatments were suspended in sterile
milk reconstituted from dry powder.

Long-term experiment

Thirty pigs were divided into 5 treatment groups with 6 pigs in each group. The T.
suis group and the dual infection group received T. suis embryonated eggs on day 0. On
day 21 after T. suis infection, the rest of the treatment groups received their respective
infective doses (Table 3. 1A). Fecal samples were collected from all the pigs on day 0
and day 21 prior to inoculation. Feces were also collected on day 22 and day 23 post- T.
suis inoculation (or 24 hours and 48 hours post C. jejuni or E. coli inoculation).

The experiment was repeated once more with 5 pigs in each group. Rectal swabs
were also taken for C. jejuni isolation at each time point and transported in Cary—Blair
medium (Difco Laboratories, Detroit, M1) on ice.

Short— term experiment

Six pigs in each group were inoculated with their respective treatments on day 0
and fecal samples were collected from each pig prior to inoculation (Table 3. 13). Feces
were also collected twenty-four hours and forty-eight hours post-infection from all the
experimental animals. Feces were stored on ice in sterile containers during transportation
to the laboratory for processing. The experiment was repeated once more with 4 pigs in

each group.

Fecal Analysis

Identification of C. jejuni from feces-post inoculation

123

Bacterial DNA was extracted from the fecal samples using the DNA isolation kit
specific for fecal samples (Qiagen Inc. Valencia, CA). The DNA was then used to detect
the presence or absence of C. jejuni 33292 using Taq Man® assays and 238 rRNA gene
Random Fragment Length Polymorphism (RF LP) as described previously (11, 25).
Extraction of supematants for ELISA

5—10 g of each fecal sample was diluted 1:3 in sterile Phosphate Buffered Saline
(PBS) containing trypsin inhibitors (1 mg/ml) and phenyhnethylsulfonyl fluoride (PMSF,
1mg/ml), and centrifuged for 15—30 minutes at 16,500 rpm, 4°C. Supematants were
collected, aliquoted and stored at —80°C until analysis by ELISA. PBS was obtained from
Gibco BRL, Rockville, MD, and the trypsin inhibitor and PMSF were obtained from

Sigma, St. Louis, MO.

ELISA assays

The supematants fi'om fecal samples were analyzed for IL-4, and IL-10 cytokines
using swine ELISA kits according to the manufacturer’s protocol (Biosource
International Inc. Camarillo, CA). An H.-6 ELISA assay was developed in the laboratory
using swine reagents purchased from Biosource International, and carried out as

described previously in chapter 2.

Data analysis
The cytokine values obtained from individual pigs were normalized to day 0
values for each pig. Fold changes in cytokine levels over time were obtained with the

following formula: Fold change = (time x levels/ time 0 levels), where x = day 0, day 21,

124

22 or 23 for long-term experiments and 0 hr, 24 hr or 48 hr for short-term experiments.
The results were graphed as mean +/- SEM. The Student’s t test was performed on the
results obtained to determine if increases in cytokines were significant. P values S 0.05

were considered acceptable levels of statistical significance.

125

RESULTS
Fecal analysis
Identification of T. suis eggs and C. jejuni in the feces-preinoculation
We found no evidence of T. suis eggs or C. jejuni in the feces of pigs prior to
inoculations, and hence the animals were deemed free of these specific organisms (data

not shown).

ELISA—Iong-term experiment
IL-4 secretion

The changes in IL-4 levels, in response to the different treatments, are shown in
Figure 3. 1. The dual infection group showed a statistically significant increase in IL-4
secretion by day 22 compared to day 0 (p < 0.05). There was a 2-fold increase in IL-4, 24
hr after C. jejuni inoculation in the dual infection group. On the other hand, there were no
statistical increases in IL-4 secretion in pigs infected/inoculated with other treatments (p
> 0.05). 55% of pigs infected with both T. suis eggs and C. jejuni showed an increase in
IL-4 (Table 3. 2). A similar result was obtained in pigs infected with T. suis alone. 0n the
other hand, only 36% of pigs infected with C. jejuni and 10% of pigs given milk showed
any increase in IL-4. None of the pigs that received E. coli DHSor showed any increase in

IL-4 afier inoculation with this bacterium.

IL-6 secretion

None of the pig groups inoculated with any treatment showed a significant

increase in IL-6 in their feces (Figure 3. 2). Pigs that received T. suis eggs and the pigs

126

that received milk showed some moderate increase by day 21 , however, there were no
statistically significant increases in IL-6 for any of the treatment groups (p > 0.05). On an
individual basis, 55% of pigs that received T. suis eggs showed an increase in IL-6 in
their feces; however, 64% of the pigs that received milk alone also showed an increase in
IL-6. The rest of the groups had less than 50% of the pigs showing an increase in IL-6 in

their feces (Table 3. 2).

IL-10 secretion

Pigs singly infected with C. jejuni, T. suis, or both showed a significant increase
in IL-10 by day 22 (p < 0.05). There was nearly a 3-fold increase in IL-10 levels in the
feces of pigs in all the three treatment groups. The groups that received milk or E. coli did
not show any significant increase in this cytokine (Figure 3. 3, p > 0.05). 64% of the pigs
that received both pathogens, and 73% of the pigs that received T. suis alone showed an
increase in IL-lO while only 45% of the pigs given C. jejuni alone showed an increase
(Table 3. 2). In the control groups, pigs that received milk or E. coli, 36% and 33% of the

pigs respectively, showed an increase in IL-10.

ELISA—short—term experiments
IL-4 secretion

None of the groups that received C. jejuni, T. suis, or both showed any significant
increase in IL-4 by 48 hr. However, pigs in both the control groups, milk and E. coli,
showed a significant increase in IL-4 by 48 hr (p < 0.05, Figure 3. 4). Concurring with

this result, 70% of the pigs that received milk and 60% of the pigs that received E. coli,

127

showed an increase in IL-4 by 48 hr. However, only 30% of the pigs given C. jejuni, T.

suis, or both showed an increase in IL-4 (Table 3. 3).

IL-6 secretion

The pigs that received C. jejuni alone, milk alone, and both C. jejuni and T. suis
showed a significant increase in IL-6 by 48 hr (p < 0.05). The pigs that received T. suis
alone showed a similar statistically significant increase in IL-6 by 24 hr, while the group
that received E. coli did not (Figure 3. 5). The dual infection group showed a 2-fold
increase in IL-6 by 48 hr, while all the other groups except the E. coli group showed 1.5
fold increase in IL-6. On an individual basis, 70% of the pigs that received T. suis alone
or both C. jejuni and T. suis showed an increase in IL-6 in their feces by 48 hr, while 50%
of the pigs that received C. jejuni alone showed a similar result. In both the control

groups, only 40% of the pigs showed an increase in IL-6 (Table 3. 3).

IL-10 secretion

None of the pigs given any of the treatments showed a statistically significant
increase in IL-10 by 48 hr (Figure 3. 6, p > 0.05). Also, only 40% of the pigs that
received both C. jejuni and T. suis showed an increase in IL-10, while 50% of the pigs
that received T. suis had an increase in this cytokine (Table 3. 3). The other three groups,
given C. jejuni, E. coli, or milk had 5 30% of the population showing an increase in IL-

10 in their feces.

Identification of C. jejuni in the feces- post inoculation

128

In the groups that received C. jejuni, analysis of DNA obtained from feces did not
show the presence of the bacteria by 48 hr post-inoculation; this was true for both the
short-term and long-term experiments. However, proximal colon biopsies from the dual
infection pigs in the long-term experiment, obtained 48 hr post-inoculation showed the
presence of the bacteria in these tissues as determined by 23S rRNA RF LP. In addition,
hematoxylin eosin stain of the proximal colon sections showed the presence of T. suis

larvae (L4) from all the pigs tested (Dr. Kathryn Jones, personal communication).

129

DISCUSSION

In previous studies, differential development of pathology in swine infected with
T. suis eggs with or without antibiotics suggested the involvement of a bacterium such as
C. jejuni in clinical trichuriasis (18). This was borne out in subsequent studies, in which
dual infection of germ-free piglets with both T. suis and C. jejuni resulted in bloody
diarrhea and pathology but not single infections with C. jejuni or T. suis (17). In this
study we show that dual infection of swine with C. jejuni and T. suis results in increased
levels of IL-4 and IL-10 in the feces during long-term infections and of IL-6 during short-
term infections. Since in previous studies, pathology is seen 13—14 days after infection
with T. suis eggs, and since trichuriasis is known as a “21 day scour”, we believe that
long-term infections are more reflective of natural infections than short-term infections
(3, 4, 18).

We hypothesized that during dual infection with T. suis and C. jejuni, the cytokine
response could be shifted to a Th-2 type, down-regulating the response to the bacteria
resulting in clinical disease and pathology. Since the animals were free of T. suis eggs
and C. jejuni prior to inoculations, the responses were considered to be due to these
organisms alone. We saw up-regulation of IL-4 and IL-10 but not IL-6, following long-
term infection with C. jejuni and T. suis. IL-6 was up-regulated during short-term dual
infections alone. However, since the control pigs also showed increases in this cytokine,
it might not be a significant effect. Another interesting observation was that the cytokine
responses in individual pigs during dual infection (Tables 3. 2 & 3. 3) mirrored that from
pigs receiving only T. suis, suggesting that the responses are more hehninth driven. This

was also confirmed by the presence of worms in the proximal colon biopsies post-

130

infection. This suggested that innate responses might play a role during initial infections
with T. suis (and C. jejuni), while adaptive responses such as IL-4 come into play during
long-term infections.

T. suis infections alone induced a significant IL-10 response during long-term
infection and IL-6 responses during short-term infections. Although IL-4 responses were
increased in 55% of the pigs during long-term infections (Table 3. 2), it wasn’t a
significant increase on average (Figure 3. 1). This is probably due to genetic variability
between pigs. C. jejuni infections did not significantly induce any cytokines during
either short-term or long-term infections. This was a little surprising, since we expected
IL-6 responses to be up-regulated following infection with this bacterium. However, one
reason could be that although all pigs were given the same infective dose, different doses
might be required to establish infection and immune responses in individual pigs, since
the pigs are out-bred. Also, measurement of cytokines beyond 48 hr might have been
required to see significant response in some pigs.

IL-4 was significantly up-regulated in pigs given milk and E. coli during the
short-term infections. The reasons are unknown at this point. An evaluation of other
cytokines like IFN-y or IL-12 might provide a clue as to whether or not this effect seen is
simply a general stress response. Testing additional pigs might also help explain this
result.

Although we failed to detect C. jejuni in the feces 48 hr after infection, we did
detect it in proximal colon tissue samples obtained at the end of the study (day 23; data

not shown). Since C. jejuni has been detected in fecal material only 4-5 days afier

131

infection in other studies (24), and since this study was conducted for only 2 days afier
infection, we might have failed to detect this pathogen in the feces.

Recovery of cytokines from feces through artificial spiking showed that IL-4, IL-
6 and IL-10 levels seen in our results are 10%, 20% and 10% respectively, of the actual
cytokine values present (data not shown). This suggests that, when extrapolated,
enormous amounts of cytokines are produced in response to infections at tissue level, and
that the low recovery in feces might be due to dilution effects through the intestinal tract,
as well as due to the short half-life of these cytokines. However, inspite of these low
recovery levels, we were still able to see statistically significant differences with in
groups. Hence, cytokine measurements from feces might provide a safe, non-invasive
alternative for the diagnosis of immune status of the host. However, that said, it is still
incumbent on the researcher to evaluate these results depending on the host and specific
cytokines measured.

IL-4 and IL-10 are anti-inflammatory cytokines and are induced during a variety
of hehninth infections. IL-4 has been shown to increase epithelial cell permeability,
which in turn is thought to loosen the worms from the mucosa. IL-4 also stimulates
peristalsis in the gut for rapid worm expulsion. It also induces IgE, IgG1 and IgG4 class
switching and increases fluid accumulation in the intestinal lumen (16). IL-4 has been
shown to regulate enteropathy in Trichinella spiralis infections (8, 14), in IgE-mediated
hypersensitivity reactions (5), in asthma in humans and other pathologies (2, 7). IL-10 is
an anti-inflammatory cytokine, with a role in downregulating proinflammatory cytokines
such as IFN-y, IL-12, and others (1). The presence of IL-4 and IL-10 in the dually

infected pigs suggested that the pathology could be Th-2 mediated. However, other

132

studies in the laboratory on these same samples indicated the presence of IL-1 [3 and

TNFor as well, during dual infections of swine with T. suis and C. jejuni (Cunningham
LB and LS. Mansfield, unpublished data). Nevertheless, the production of H.-4 and IL-
10 suggests that these two cytokines have a role during the pathology mediated by dual
infection of T. suis and C. jejuni in pigs. The specific roles of these cytokines in
pathology will be the subject of subsequent studies.

Millions of people worldwide are infected with the human equivalent of T. suis, T.
trichiura, and C. jejuni is a well-known human pathogen. Hence this study has clinical
significance for both humans and pigs. Recently, a patient with severe colitis was shown
to have both C. jejuni and T. suis ova in the feces (23) —- a clinical correlate of our
experimental findings. With increasing awareness of the potential health hazard posed by
helrninth infections, more studies on hehninth—bacterial interactions should be

forthcoming.

133

ACKNOWLEDGEMENTS
This work was supported by the NIH-grant, 61-0954. We thank Dr. Pawin
Padungtod, and Dr. Jeanie Gaymer for their help with the animal experiments. Thanks are
also due to all the staff at the Containment Facility, Michigan State University. We thank

Dr. Julia Bell for the critical review of this article.

134

10.

ll.

12.

REFERENCES

Aggarwal, B. a. G., JU. 1992. Other cytokines, p. 427-428, Human Cytokines,
Handbook for basic and clinical research. Blackwell Scientific Publications,
Cambridge, MA.

Ajuebor, M. N., C. M. Hogaboam, T. Le, and M. G. Swain. 2003. C-C
chemokine ligand 2/monocyte chemoattractant protein-1 directly inhibits NKT
cell IL-4 production and is hepatoprotective in T cell-mediated hepatitis in the
mouse. J Immunol 170:5252-9.

Batte, E. G., R. D. McLamb, K. E. Muse, S. D. Tally, and T. J. Vestal. 1977.
Pathophysiology of swine trichuriasis. Am J Vet Res 38: 1075-9.

Beer, R. J. 1973. Studies on the biology of the life-cycle of Trichuris suis
Schrank, 1788. Parasitology 67:253-62.

Cooper, E. S., J. Spencer, C. A. Whyte-Alleng, O. Cromwell, P. Whitney, S.
Venugopal, D. A. Bundy, B. Haynes, and T. T. MacDonald. 1991. Immediate

hypersensitivity in colon of children with chronic T richuris trichiura dysentery.
Lancet 338:1 104-7.

Dodge, 1. L., M. W. Carr, M. Cernadas, and M. B. Brenner. 2003. IL-6
production by pulmonary dendritic cells impedes Th1 immune responses. J
Immunol 170:4457-64.

Fireman, P. 2003. Understanding asthma pathophysiology. Allergy Asthma Proc
24:79-83.

Garside, P., M. W. Kennedy, D. Wakelin, and C. E. Lawrence. 2000.
Immunopathology of intestinal hehninth infection. Parasite Immunol 22:605-12.

Gause, W. C., J. F. Urban, Jr., and M. J. Stadecker. 2003. The immune
response to parasitic helrninths: insights from murine models. Trends Immunol
24:269-77.

Hill, D. E., H. R. Gamble, M. L. Rhoads, R. H. Fetterer, and J. F. Urban, Jr.
1993. T richuris suis: a zinc metalloprotease fi'om culture fluids of adult parasites.
Exp Parasitol 77:170-8.

Iriarte, P., and R. J. Owen. 1996. PCR-RFLP analysis of the large subunit (23S)
ribosomal RNA genes of Campylobacterjejuni. Lett Appl Microbiol 23:163-6.

Janeway, C., Travers P., Walport M, and Capra JD. 1999. Host defense
against infection, p. 363-416. In A. P. a. L. E (ed), Immunobiology,The immune

135

l3.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

system in health and disease, Fourth ed. Current Biology Publications and
Garland Publishing, New York.

Kealy, W. F. 1976. Colonic lymphoid-glandular complex (microbursa): nature
and morphology. J Clin Pathol 29:241-4.

Lawrence, C. E., J. C. Paterson, L. M. Higgins, T. T. MacDonald, M. W.
Kennedy, and P. Garside. 1998. IL-4-regulated enteropathy in an intestinal
nematode infection. Eur J Immunol 28:2672-84.

Liebler, E. M., J. F. Pohlenz, and N. F. Cheville. 1988. Gut-associated
lymphoid tissue in the large intestine of calves. II. Electron microscopy. Vet
Pathol 25:509-15.

Mansfield, L. S., Gause, W.C.,Finkelman, F.D., and J.F. Urban, Jr. 2000.
Gastrointestinal nematodes and the immune system, p. 555-567. In M. W.
Cunningham, and Fujinami, R.S. (ed.), Effects of microbes on the immune
system. Lippincott, Williams and Wilkins, Philadelphia.

Mansfield, L. S., D. T. Gauthier, S. R. Abner, K. M. Jones, S. R. Wilder, and
J. F. Urban. 2003. Enhancement of disease and pathology by synergy of

T richuris suis and Campylobacterjejuni in the colon of immunologically naive
swine. Am J Trop Med Hyg 68:70-80.

Mansfield, L. S., and J. F. Urban, Jr. 1996. The pathogenesis of necrotic
proliferative colitis in swine is linked to whipworrn induced suppression of
mucosal immunity to resident bacteria. Vet Immunol Immunopathol 50:1-17.

Mansfield, L. S. a. D. G. 2004. Lymphoglandular complexes are important
colonic sites for IgA induction against Campylobacterjejuni in a swine disease
model. Comparative medicine.

Morfitt, D. C., and J. F. Pohlenz. 1989. Porcine colonic lymphoglandular
complex: distribution, structure, and epithelium. Am J Anat 184:41-51.

Nachamkin, I., K. Bahachick, and C. M. Patton. 1993. Flagellin gene typing of
Campylobacterjejuni by restriction fragment length polymorphism analysis. J
Clin Microbiol 31 :1531-6.

O'Leary, A. D., and E. C. Sweeney. 1986. Lymphoglandular complexes of the
colon: structure and distribution. Histopathology 10:267-83.

Shin, J. L., G. W. Gardiner, W. Deitel, and G. Kandel. 2004. Does whipworrn

increase the pathogenicity of Campylobacterjejuni? A clinical correlate of an
experimental observation. Can J Gastroenterol 18:175-7.

136

24.

25.

Skirrow, M., and Blaser, M.J. 2000. Clinical aspects of Campylobacter
infection, p. 69-88. In I. Nachamkin, and Blaser, M.J (ed.), Campylobacter,
Second ed. American Society for Microbiology, Washington, DC.

Wilson, D. L., S. R. Abner, T. C. Newman, L. S. Mansfield, and J. E. Linz.

2000. Identification of ciprofloxacin-resistant Campylobacterjejuni by use of a
fluorogenic PCR assay. J Clin Microbiol 38:3971-8.

137

 

 

 

 

 

 

 

 

 

Eu .2 x 3 :8 .m 652 E... .8 x 3 3:3 .6 ca: 30 .2 x s 33.3 o a an
Amwwo copes—20380
ca: 65: ca: 83 as H 9%» Baaobneo 83 as a o :5
Jill I
DE
8:: =8 .m 55 E33 U as» a mass U Ea as: .a $.80

 

 

A8839» in S 808596 €8.98— .8.“ meowaioofl ".3 .m 039“.

138

 

 

 

 

 

 

 

 

38 6%... 83:90:35 Emu «3 x S .szmx U 28
MW2 x a Run :8 .m. ca: 3% M2 x a Sex U 83 as a sumo Engines 8an as. H c an
08E.
dmflfl :8 .N :52 Eaxux U as“ H Fifi U can a?» H 9:80

 

 

@0833 E S Hogans Eurtofi no.“ 305385 um: .m oEfiH

139

 

 

 

 

 

 

 

 

 

 

 

Xmm $9.” $3 .xcmb $3 07.:
Xmm $3 $3 $mm $3. 0-..:
.xb $3 $3 $3 $3 Yd
0:336
525 :8 .m ca: 3.8.4 U as. s 5.3 U as as. a
25330 82» a E 08803 5 92305 mwa mo owfiaoocom 955

 

 

£585qu Shrugs

~

I macaw—=85 03838“ Sun NN ham .3 one—8.3 83m a E 88.85 5 9:30am mmfi mo owaueoobm “N .m 035.

140

 

 

 

 

 

 

 

 

 

 

 

 

. $o~ $om $om $om $2. 3.3
$3 $3 $3 $2. $2. 93
$8 $2. $2“ $om $om Yd
0586 1
M
ammo :8 .m “.52 .Eaxmx U ASM H .5:me U Ea n.3,. H
25.030 53» a 5 080.85 5 wgonm mmE mo owfinuobm 9:80

 

figfitoaxo

Sunfitonm I 323305 038258 .Baa E w». 8 vn .3 338.8 53% a 5 030.85 5 mgoam mmE mo omflqoouom "M .n 03:.

Figure 3. 1. Changes in IL-4 in feces of pigs inoculated with different groups during the

long-term experiment. Four-to-five—week-old piglets were inoculated with T. suis eggs on

day 0. Twenty-one days later, half of the pigs received C. jejuni, while the others

received milk. Additional groups received E. coli DHSa or C. jejuni at this point.

Control pigs received milk alone. Values shown are means of two identical experiments

with a total of 11 pigs in each group (n=6 for the E. coli group). Mean +/- SEM is

shown. (*, p < 0.05).

Fold changes in IL-4 over day 0

  
    

i ‘IDaya-j
”7 ‘IDay21l
_; ‘DDay223
lDayZi

jejuni only
group group

Treatment groups

142

Figure 3. 2. Swine inoculated with the different treannent groups were analyzed for IL-6
cytokine changes in the feces of pigs from respective groups (n = 11; n=6 for E. coli
group). Mean fold changes in IL—6 in different groups during the long-term experiment is

shown.

 

 

3 7 i ._7 7 , fl 7 _fi 7, 7 7 ,
EBay—(Tl
IDay21‘
. CIDay22
2" ”*" ’ ’ ’ " ’ " ”’ ’ mam

Fold changes in IL-6 over day 0
2
l
l
l
l
l
l
l
l
l

   

T. suis C. Control E. coli

group jejuni group group
group
Treatment groups

143

Figure 3. 3. Fold changes in IL-10 in the feces of pigs from long-term experiments.
Mean of two identical experiments is shown with a total of 11 pigs in each group (n= 6

for the E. coli group). Values shown are mean +/- SEM. (*, p < 0.05).

I“
a:
l

 

3(-

l

l
l

.JIDay 0
IDay 21
ClDay 22

j-Pafl
, , .J

 

  
    

7
l
l
l

 

N
u: o.)

 

 

 

 

  

Fold changes in IL-10 over day 0

C. . Control
jejuni group jejuni group
and T. group
suis
group

Treatment groups

144

Figure 3. 4. Mean fold changes in IL-4 cytokine in the feces of pigs from the short-term

experiments. Mean of two identical experiments is shown, with 10 pigs in each group.

Values are mean +/- SEM. C“, p < 0.05).

Fold changes in IL-4 over 0 hr

 

 

 

 

 

25p , *** * *‘ ' **** * ,-, ,, ,, liahrl
i * I24hr:
2. in , ---,--\ agha'
*
C. T.suis C.jejuni Control E.coli
jejuni Pigs pigs Pigs Pigs
and
T.suis
Pigs

Treatment groups

145

Figure 3. 5. Four to five week-old conventionally reared weaned piglets were inoculated
with the different treatments orally for 48 hr. Feces were collected prior to inoculation (0
hr), 24 hr and 48 hr afier inoculations. Mean of two identical experiments with 10 pigs in

each group is shown. Values shown are mean +/- SEM. (*, p < 0.05).

 

 

 

   

 

 

 

 

 

 

 

 

 

 

 

 

3 5 ~bwu “n_mmmm nm—q
5 l
c 3 T—vmwimw—mmmmm 1“ «l
i...
3
o
3zsw__—M%~fi~hh+2~~V*7+.w_m
i—I * .0 hr 3
E- I24 hr l
m 048 hr'
:5 2
a
a
.4:
o
1:
g- 1.5 ~
in

g
0.5 — g.
0 - ‘ _
C.jejuni T.suis C.jejuni E.ooli
and pigs pigs pigs
T.suis
pigs

Treatment groups

146

Figure 3. 6. IL-lO cytokine changes in pigs inoculated with different treatments during
the short-term experiment. Values are means from 2 identical experiments. Mean +/-

SEM is shown.

 

 

1.5

 

 

Fold changes in IL-10 over 0 hr

A
l

 

 

 

 

 

 

 

 

 

 

0.5 ~ r___
O . Lr—~ —.~— —--—
C.jejuni T.suis C.jejuni Control E.coli
and pigs pigs pigs pigs
T.suis
pigs

Treatment groups

147

CHAPTER 4
RECOMBINANT IL-4 (rIL-4) ENHANCES CAMPYLOBAC T ER JEJUNI

INVASION OF PORCINE INTESTINAL EPITHELIAL CELLS (IPEC-l)

148

ABSTRACT

Campylobacterjejuni (C. jejuni) is a food borne bacterial pathogen and is one
of the leading causes of bacterial enteritis in humans. It is commonly acquired through
ingestion of contaminated meat products or raw milk, and its pathogenesis is not
completely known. Helminth parasites such as T richuris suis (T. suis) have been
documented to modulate immune responses to intracellular pathogens. In previous
studies, germ-fi'ee piglets infected with either C. jejuni or T. suis alone display few
clinical signs of disease. In contrast, dual infection results in bloody diarrhea and severe
pathology with multifocal lesions, infiltration of inflammatory cells, excessive mucus
secretion, and bacterial invasion into epithelial cells and macrophages. In follow-up
studies, swine infected with both T. suis (2500 eggs) and C. jejuni (108 cfu) had elevated
IL-4 (and IL-10) responses in the feces. In this study, we hypothesized that IL-4 (or IL-
10) enhances invasion of C. jejuni in intestinal pig epithelial cells (IPEC). To test this
hypothesis, IPEC-l cells were pretreated with recombinant IL-4 (rIL-4) or rIL-lO for 5
hours followed by invasion assays with C. jejuni. Cells pretreated with rIL-4 showed ~ 6
fold increases in C. jejuni internalization, compared to cells that received no
pretreatment. Addition of anti-IL-4 antibody reversed this effect. Preincubation with rIL-
10 did not significantly alter the internalization of the bacteria. Also, the transepithelial
electrical resistance (TER) was significantly reduced following rIL-4 treatment but not
following rIL-lO treatment. Invasion increased with increasing doses of rIL-4. Light
microscopy showed more vacuoles within IPEC-l cells after rIL-4 treatment, while
transmission electron microscopy showed multiple bacteria within these cells. Most of

the bacteria were in the cytoplasm, and some were adjacent to vacuoles. We conclude

149

that rIL-4 alters the physiology of these epithelial cells allowing increased invasion of C.

jejuni.

150

INTRODUCTION

Campylobacteriosis caused by C. jejuni and C. coli is a serious illness in humans
characterized by a wide range of symptoms from acute diarrhea, fever, blood in stools,
vomiting and abdominal cramps- to a non-inflammatory secretory diarrhea (25).
Although rare, one of the serious complications is a neurological disorder termed Guillain
Barre Syndrome which occurs in roughly 1/ 1058 infected individuals (22). Deaths due to
Campylobacter infections are estimated to be in the hundreds in the United States alone
(8). The mechanisms by which Campylobacter species cause disease are not well
understood. Some of the virulence factors identified include toxins, adhesins, motility,
and invasion. Invasive strains of C. jejuni have been more commonly associated with the
inflammatory type of diarrhea and non-invasive strains with the secretory type of
diarrhea. Also, flagellated, motile strains are considered more invasive than non-
flagellated, non-motile strains (12). Recently, a campylobacter invasion antigen (Cia B)
has been identified that is thought to play a vital role in invasion (23). CiaB is thought to
be secreted into the host cells through a Type 1H secretion apparatus, where it triggers a
series of signal transduction events, resulting in C. jejuni uptake. However, the host
factors that contribute to invasion have not been studied very extensively. In a swine
animal model to study campylobacter pathogenesis, C. jejuni acted synergistically with a
whip-worm parasite T richuris suis in the colon of gnotobiotic piglets to produce severe
pathology marked by focal lesions, infiltration of inflammatory cells, excessive mucus
secretion, bloody diarrhea, and bacterial invasion into epithelial cells and macrophages,
similar to campylobacteriosis in humans (19). C. jejuni invasion into epithelial cells

occurred in the undifferentiated crypt epithelial cells surrounding the worms in the

151

proximal colon as well as in the differentiated follicle-associated epithelium lining the
secondary lymphoid organs in the distal colon (lymphoglandular complexes).
Conversely, pigs infected with either C. jejuni (1 x 106 cfu) or T. suis (3000 infective
larvae) alone did not show clinical signs of disease.

In similar studies with conventionally reared, weaned piglets, swine that received
T. suis (2500 infective larvae) followed 21 days later by C. jejuni (2 x 108 cfu) showed
enhanced IL—4 and IL-10 cytokine secretion in their feces compared to control pigs
(Parthasarathy, G, L. D. Cunningham and LS. Mansfield, unpublished data). IL-4 and
IL-10 are anti-inflammatory cytokines that are normally induced during helrninth
infections (9). IL-4 has been shown to down-regulate IL-1 or, IL-1 [3, TNFor and nitric
oxide in vitro (3). Most importantly, IL-4 has also been shown to play a critical role in
the enteropathology associated with T richinella spiralis infection in mice (17), asthma in
humans (6), T-cell mediated hepatitis (2) and other pathologies (5). IL-lO has been
shown to down-modulate Th-l cytokine responses in macrophages (1). In this study, we
hypothesized that IL-4 or IL-10 induced in response to dual infection enhances the
invasion of C. jejuni, and, hence, accounts for some of the pathology. To test this
hypothesis differentiated intestinal pig epithelial cells (IPEC-l) derived from a neonatal
piglet (4) were used to determine the effect of rIL-4 and rIL-lO on C. jejuni invasion. The
data from this study suggests that IL-4 is likely to play a dominant role in mediating

pathology in T. suis and C. jejuni infected swine.

152

MATERIALS AND METHODS

Bacterial strains and culture

C. jejuni strains ATCC 33292 and 81-176 were used for the invasion studies.
Both were initially isolated fi'om humans with enteritis. C. jejuni 81-176 was kindly
donated by Dr. Carol Pickett, University of Kentucky. Low passage bacterial, colonies
were grown for isolation on Bolton Agar supplemented with 5% sheep’s blood
(Cleveland Scientific, Bath, OH) for a period of 48—72 hr. The control strains Escherichia
coli DHSa and E. coli 0157zH43 (DEC7A) were grown similarly on Nutrient agar (Difco
Laboratories, Detroit, MI). DEC7A was initially isolated from swine and was a kind
contribution from Dr. Thomas Whittam, Michigan State University. An isolated colony
was further grown as a lawn on corresponding medium for 20 hr. Bacteria were swabbed
from plates and resuspended in RPMI 1640 supplemented with phenol red and 5% heat
inactivated fetal bovine serum (Gibco BRL, Rockville, MD). The optical density (OD560)
was adjusted to 0.1 to achieve ~ 108 CFU/ml. Bacterial suspensions were used

immediately after preparation.

Cell culture

Intestinal cells obtained from swine (IPEC-l) were used to elucidate the effect of
rIL-4 (or rIL—lO) on C. jejuni invasion of epithelial cells. IPEC-l is a non-immortalized
and undifferentiated neonatal small intestinal epithelial cell line that can be induced to
differentiate by growth on porous substrates for 10—14 days (10, 28). All media and
supplements for cell culture were obtained from Gibco BRL, Rockville, MD, unless

otherwise stated. The transwells plates were obtained from Corning Costar, Corning, NY.

153

The IPEC-l cells were cultured routinely in DMEM/F-12 medium supplemented
with 5% Fetal Bovine Serum (F BS) and 1% Insulin-Transferrin-Selenium in cell culture
flasks. After 5—6 days of growth, the cells were washed in versene and trypsinized. They
were then cultured at a density of 3x105/ well on transwell inserts (6.5 mm diameter; 3
pm pore size) that had been previously coated with fibronectin (20 pg/ml; Sigma, St.
Louis, MO). Cells were allowed to differentiate for 10 — 14 days in RPMI 1640 medium
containing phenol red and supplemented with 5% Fetal Bovine Serum (FBS).
Transepithelial electrical resistance (TER) across the monolayer was measured to
determine confluency and tight junction formation (EVOMX, World Precision
Instruments, Sarasota, FL). TER values between 150-600 Qcm2 were considered as

evidence of confluent monolayer.

Gentamicin killing assays

10—12 day old IPEC-l cells on transwell membranes were treated with either 500
pg of rIL-4 or 1000 pg of rIL-lO in 100 pl of RPMI 1640 supplemented with 5% F BS
(complete medium; CM) and incubated at 37° C, 5% C02. 100 pl of CM alone was
included as a negative control for the assay. The working concentrations of cytokines
were selected from optimal concentrations determined in previous studies in the swine
model, and the cytokine reagents were obtained from swine ELISA kits purchased from
Biosource International, Camarillo, CA. After 5 hr, the treatment solutions were
removed as completely as possible. The time point was chosen based on previously
published studies (29). IPEC-l cells were not washed in order to avoid damage to the

monolayers and were infected with 5 x 107 cfii of bacteria in 100 pl of CM. Invasion was

154

allowed to proceed for 3 hr at 37°C, 5% C02. The bacteria were then removed and
adherent bacteria on both sides of the membrane were killed with 100 pg/ml of
gentamicin (Gibco BRL, MD) suspended in CM, for 1 hr. The transwell membranes were
excised and the epithelial cells were lysed with a final concentration of 0.1% sodium
deoxycholate (Sigma Chemical Co., St. Louis, MO) in 1X Phosphate Buffered Saline
(0.01M). The internalized bacteria were enumerated by plating on Bolton agar plates. All
treatments were performed in triplicate.

Subsequently, gentamicin killing assays were also carried out 1) to determine the
effect of anti-IL-4 antibody on invasion, 2) to establish the dose response of rIL-4 to C.
jejuni internalization, and 3) to determine the effect of rIL-4 pretreatment on non-
invasive E. coli DHSa. In addition, the number of C. jejuni that had translocated to the
baso-lateral chamber also was determined. IPEC-l cells were pretreated with either rIL-4
(500 pg/well) or media for 5 hr, followed by incubation with C. jejuni 81176 for 3 hr.
Media in the lower chamber were sampled after the three-hours of incubation with C.

jejuni, plated on Bolton agar and bacterial colonies were enumerated.

Light microscopy

12-day old IPEC-l cells on transwells were treated with rIL-4 (500 pg/well) or
medium for 5 hr. The treatments were removed, the membranes were excised with a
sterile scalpel, and examined by light microscopy using a laser scanning microscope at

63x aperture (LSM Pascal 5, Zeiss International, Jana, Germany).

155

Adherence assays

Adherence assays were carried out similarly to the invasion assays. The control strain, E.
coli 01571H43 DEC 7A, exhibiting adherence characteristics was included to determine
the effect of IL-4 on adherence. Afier pretreatment with 500 pg of rIL-4 in CM or CM
alone for 5 hr, the treatments were removed and the bacteria (5 x 107 cfu in CM) were
allowed to adhere for 1 hr. The supematants containing bacteria which did not adhere
was then removed and the transwell membranes were excised with a sterile scalpel. The
epithelial cells were then lysed as described in the gentamicin assays, and the cell
associated (adherent) bacteria were serially diluted and enumerated on Bolton agar plates
for C. jejuni (or nutrient agar plates for E. coli). Invasion assays were conducted for the
C. jejuni strains concurrently as internal controls. However, invasion assays were not

done for the E. coli strain since its invasion potential is not known.

TEM

IPEC-l cells were subjected to invasion assays as described previously. After the
gentamicin treatment, medium was removed, the membranes were excised and fixed in
0.1M phosphate buffer containing 2.5% glutaraldehyde and 2.0% paraforrnaldehyde (pH
7.4). Membranes were washed three times in 0.1 M phosphate buffer (pH 7.4) for twenty
minutes each. Following the washes, the membranes were fixed in 2% osmium tetroxide
and dehydrated in increasing concentrations of acetone. The dehydrated membranes were
subsequently infiltrated with epoxy resin, embedded in silicone molds, polymerized for 2
days at 60° C, and then examined using the transmission electron microscope (J EOL

100CX, Japan). Complete gentamicin killing assays (plate assays) as described

156

previously, were conducted concurrently as internal controls; bacteria were identified by
colony morphology, microscopic examination and colonies enumerated. Seven samples
each for rIL-4 pretreatment and media pretreatment were processed for TEM. For each
treatment sample ~30 frames were examined at multiple magnifications. Cells were
scored for morphological changes and for location of C. jejuni (whether they were

adherent, intracellular, intravacuolar or intracytoplasmic).

MTT assays

MTT assays were performed to determine the cytotoxic effect of rIL-4, if any, on IPEC-l
cells. Briefly, IPEC-l cells on tranwells pretreated with rlL-4 or media were incubated
with 0.5mg/ml of MTT (suspended in RPMI media with 5% FBS), for 3 hr at 37° C, 5%
C02 in 50 p1 volumes. The treatment was removed, and cells were incubated with 0.1%
Triton-X-IOO for 10 minutes at 37° C and 5% C02. 100 pl of acidic isopropanol (0.1N
hydrochloric acid in isopropanol) was then added, mixed, and absorbance measured at

570nm. All treatments were performed in triplicate.

Data analysis

Data are presented as mean :t standard error of the mean (SEM). Student’s t-test
was performed on the results obtained using Microsoft Excel, unless otherwise stated. P
values 3 0.05 were considered significant. Regression analysis was also performed when

required using Microsoft Excel, with confidence set at 95% level.

157

RESULTS

rIL-4 pretreatment of IPEC-l caused a significant increase in the numbers of C.
jejuni invading IPEC-l cells (Figure 4. 1). There was a 6-10-fold increase in the number
of bacteria that invaded compared to IPEC-l cells pretreated with media alone (p < 0.05).
Both C. jejuni strains used invaded [PEC-l efficiently. There were no significant
differences in the ability to invade between these strains based on comparisons of
intracellular bacteria. rIL-lO pretreatment, unlike rIL-4, had no effect on invasion of C.
jejuni (Figure 4. 1).

To determine if rIL-4 treatment was necessary and sufficient for increased
invasion of lPEC-l by C. jejuni, cells were incubated with rIL-4 and anti-IL-4 antibody.
The results are shown in Figure 4. 2. When IPEC-l cells were pretreated with a mixture
of rIL-4 and anti-IL—4 antibody, the number of bacteria that invaded was similar to the
control levels, while pretreatment with rIL-4 alone showed a significant increase in
invasion (p < 0.01). The same results were seen for both C. jejuni strains tested.
Pretreatment with anti-IL-4 antibody alone had no effect on invasion (data not shown).

rIL-4 pretreatment had a dose dependent effect on C. jejuni invasion (Figure 4. 3).
There was a progressive increase in the number of intracellular bacteria with increasing
amounts of rIL-4 pretreatment. Regression analysis on the results obtained showed a
significant dose response effect for both C. jejuni strains (p < 0.05).

To determine if rIL-4 facilitated uptake without active participation by the
bacteria, a non-invasive strain of E. coli (DHSOL) was used in the invasion assay, both
with and without rIL-4 pretreatment of IPEC-l cells (Figure 4. 4). Unlike its effect on C.

jejuni, rIL-4 pretreatment had no effect on uptake of E. coli DHSa (p > 0.05).

158

rIL—4 treatment had a significant effect on the trans-epithelial electrical resistance
(TER) across the IPEC-l monolayer (Figure 4. 5). The TER fell to ~50% of its initial
values after treatment with rIL-4 (p < 0.01) while cells treated with rIL-10 and medium
alone showed no significant decrease in TER. This was corroborated by light
microscopy, which showed IPEC-l cells no longer adj oined after treatment with rIL-4 for
5 hr (Figure 4. 6C), unlike cells treated with medium alone which were confluent and
closely packed (Figure 4. 6A and 4. 6B). Increased vacuolation also was seen in the cells
that received rIL-4 treatment compared to cells that were treated with medium alone. The
presence of vacuoles in rIL-4 treated cells was not due to toxicity. MTT treatment assays
showed that these cells were viable, and not significantly different from medium alone
treated cells (Figure 4. 10). Also, translocation experiments were performed to determine
if there was an increase in the traffic of C. jejuni to the baso-lateral chamber after rIL-4
pretreatment. As seen in Table 4.1, there was no significant increase in the number of
bacteria to the baso-lateral surface when compared to media pretreated cells.

The effect of rIL-4 on C. jejuni adherence to IPEC-l cells was different than that
seen for invasion (Figure 4. 7). C. jejuni strain 81176 adhered equally to IPEC-l with or
without pretreatment of rIL-4. The other two bacteria, C. jejuni 33292 and E. coli
0157:H43 DEC7A, a pig pathogen with adherent capabilities, showed statistically
significant decrease in adherence after pretreatment of IPEC-l with rIL-4. Based on data
from the invasion assays and adherence assays, the percentages of C. jejuni that invaded
as a percentage of bacteria that adhered were ~0.2% (strain 81176) and 0.06% (strain
33292) for media pretreatment, and 0.8% (strain 81176) and 1.06% (strain 33292) for

rIL-4 pretreatment.

159

Transmission electron micrographs of IPEC-l cells showed bacteria inside cells
following rIL-4 pretreatment and after gentamicin-killing assays (Figure 4. 8A) but not
after medium pretreatment (Figure 4. SB). The bacteria were seen predominantly in the
cytoplasm just under the plasma membrane or adjacent to vacuoles (Figure 4. 9, upper
row). This was the case in almost all the ~ 30 fields analyzed; bacteria were seen largely
near the apical surface with a few on the baso-lateral side (Figure 4. 9 panels 3 and 4),
and in ~ 95% of the time free in the cytoplasm. Electron micrographic appearance of
intracellular C. jejuni were similar to those observed by Kiehlbauch et al.,(15) and
showed a dense cytoplasm surrounded by a lumen (Figure 4. 9, lower row, panels 1, 2,
and 4). Coccoid structures also were seen, similar to results seen by Mansfield et al., (19)

in vivo in pigs (Figure 4. 9, lower row, panel 3).

160

DISCUSSION

Host factors studied to date for C. jejuni invasion of epithelial cells have been
focused on signal transduction pathways and host cell surface receptors required for
binding. In this study, we show that a soluble factor produced by host cells can also play
a role in bacterial invasion. IL-4 is an anti-inflammatory cytokine normally induced in
response to helrninth infections. Although pro-inflammatory cytokines like TNF-a and
IFN-y have been traditionally associated with the production of inflammation and
pathology (13, 27), recently, IL-4 has been implicated as a source of pathology. In the
swine intestinal epithelial cells used in our study, rIL-4 caused an ~ 6-fold increase in
invasion by both the pathogenic strains of C. jejuni tested (Figure 4. 1). This effect was
seen only when IPEC-l cells were pretreated with rIL-4 for 5 hrs; simultaneous addition
of both rIL-4 and C. jejuni to IPEC-l cells had no effect on invasion (data not shown).
These results suggest that the effect of rIL-4 on C. jejuni invasion required additional
events to take place within the eukaryotic cells before invasion was enhanced. Also,
antibody to rIL-4 had a neutralizing effect on rIL-4 enhanced invasion of the bacterial
strains, presumably because the IL-4 antibody prevented rIL-4 fi'om binding its receptor
(Figure 4. 2). Together, these data suggest that rIL-4 might mediate its effect through
binding to its receptor on the IPEC-l cell surface, triggering signal cascades in the host
cell and eventually leading to enhanced internalization of C. jejuni.

rIL-4 mediated enhancement of bacterial invasion was dose-dependent (Figure 4.
3). Increasing concentrations of rIL-4 led to increased internalization of bacteria. This

suggests an increased cross-linking of IL-4 receptors, leading to increased signal strength,

161

(or increased recruitment of mediators in signaling pathways) leading to enhanced
invasion.

The effect of rIL-4 on invasion was observed only for pathogenic bacteria like C.
jejuni and not for a laboratory selected non- pathogenic species like E. coli DHSa (Figure
4. 4). It is possible that in live hosts, IL-4 may not influence uptake of non— invasive
commensals, rather only invasive bacteria. However, more non-invasive strains should be
tested before this conclusion can be drawn.

In previous studies with the nematode parasite, Heligmosomoides polygyrus,
Shea-Donohue et al., (24) demonstrated increased permeability in mouse intestinal
epithelial cells treated with IL-4. They suggested that the role of IL-4 might be to loosen
the junctions between cells so that the worms are no longer tightly embedded and hence
can be washed away by the intestinal fluids. Similarly, we found that pretreatment with
rIL-4 increased the permeability of the monolayer (Figure 4. 5 and 4. 6C). Also,
Monteville and Konkel concluded from their translocation experiments on T84
monolayers that C. jejuni might invade through the baso-lateral surface (21). We
performed translocation experiments, in which medium in the lower chambers were
sampled after incubation with bacteria for 3 hr to determine whether the increased
invasion seen was due to increased translocation to the baso-lateral surface. However, we
found no significant increase in translocation of bacteria to the lower chamber after
pretreatment with rIL-4 compared to translocation after media pretreatment (Table 4. 1).
Additionally, TEM analysis showed more bacteria just under the apical surface,
suggesting that invasion occurred largely through apical surfaces rather than through

baso-lateral surfaces (Figure 4. 8A).

162

Although C. jejuni has been reported to reside inside vacuoles in other studies
(12), and although increased vacuolation was seen within the epithelial cells after rIL—4
pretreatment (Figure 4. 6C), in these experiments more bacteria were seen in the
cytoplasm and adj acent to vacuoles than within them (Figures 4. 8 and 4. 9). However,
this does not negate the fact that C. jejuni might have escaped the vacuoles, since TEM at
earlier time points were not conducted. In addition, cytotoxicity assays with MTT showed
that this increased vacuolation after rIL-4 treatment was not due to a cytotoxic effect on
IPEC-lcells as there was no significant decrease in viability compared to medium-alone
treated cells (Figure 4. 10).

rIL—4 had differential effects on adherence of the two C. jejuni strains and the E.
coli strain 0157:H43 DEC7A (Figure 4. 7). The known adhesins for C. jejuni include the
CadF protein that binds fibronectin and the outer membrane protein PEB1(14, 16). Other
putative C. jejuni adhesins include the flagella and lipopoysaccharides (20). For the
ETEC strain the fimbriae K88 have been known to mediate adherence in pigs (7). C.
jejuni 33292 and E. coli 0157: H43 showed decreased adherence after pretreatment of
IPEC-l cells with rIL-4, while rIL-4 pretreatment had no effect on adherence of C. jejuni
strain 81176. This suggests that the binding sites on the host cell for C. jejuni 33292 and
E. coli 0157: H43 might be similar, but different from C. jejuni 81176 receptor, and that
rIL-4 might selectively down-regulate these receptors. One of the host receptors
implicated in K88 binding in vivo in pigs is an intestinal mucin-type glycoprotein (7), and
is a likely candidate for C. jejuni 33292 binding. Other explanations for the differential
binding of the two C. jejuni strains might be that these strains differentially express

certain adhesins or bind through different adhesins. However, even though adherence was

163

reduced by rIL-4 pretreatment, strain 33292 still showed enhanced invasion compared to
the cells that received media alone, since 1.06% of the adhered population invaded afier
rIL-4 pretreatment, while only 0.06% invaded afier medium pretreatment. Thus the
invasion potential of 33292 was higher after rIL-4 pretreatment, since the proportion of
the adhered population that invaded was higher. Also many of the cell-associated bacteria
could actually be internal, compared to media-treated cells. Results of an invasion assay
conducted at 1 hr with C. jejuni 33292 support this conclusion (data not shown).

Since this study was conducted in part to explore possible pathogenic mechanisms
in the dual infection model, we believe that the IL—4 previously shown to be induced in
response to both C. jejuni and T. suis in pigs might cause the enhanced invasion of
bacteria seen in the swine model (19). This is corroborated by the fact that Dr. Helene
Kringel at USDA found 3—6 fold elevation in IL-4 mRNA in swine infected with T. suis
in the proximal colon and the LGCs (personal communication)—the same areas where
enhanced invasion of C. jejuni was seen in dually infected pigs (19).

Our results are unlike those obtained by Hess et al., (11), who found no effect of
IL-4 pretreatment on internalization of Listeria monocytogenes or Salmonella
typhimurium into differentiated Caco-2 cells or HT-29 cells. However, Hess et al.,
pretreated cells with IL-4 for 48 hr or 72 hr, and the events we describe here occur much
earlier. In addition, the different results described here might be due to the different cell
line and bacteria used.

Previous studies have shown that IL-4 can downregulate TNF-or expression in
porcine macrophages by 6 hr afier exposure and in murine mast cells by 24 hr after

exposure (26, 29). IL-4 mediated down-regulation of TNF-or in the mast cells was

164

through STAT—6, a transcriptional activator, and occurred by destabilization of TNF-a
mRNA. Recent studies show that alteration of intestinal barrier permeability is also
mediated by STAT-6 (18). In addition, it is generally accepted that STAT-6 activation
occurs through IL-4 receptors (26). Therefore, we propose the following model for the
mechanism by which rIL-4 causes enhanced invasion of C. jejuni. rIL-4 binds to a
receptor, presumably IL-4Ra, and causes intracellular signal transduction events leading
to increased vacuolation and perhaps downregulation of TNF-or or other proinflammatory
cytokines, via STAT-6. These events are accompanied by changes to extracellular
surfaces marked by breakdown of tight junctions (and differential expression or exposure
of C. jejuni host receptors), facilitating increased invasion by C. jejuni.

rIL-10, unlike rIL-4, did not have a similar effect on C. jejuni invasion or on the
TER across the monolayer. Hence further studies were not conducted with this cytokine.
Also, corroborating our result, Dr. Kringel did not find any increase in IL-10 mRNA after
T. suis infection at the sites where increased bacterial invasion was seen. Thus the role of

IL-10 in this system remains to be determined.

165

ACKNOWLEDGEMENTS
This work was supported by the National Institute of Health grant 61-0954 and by
the Center for Emerging Infectious Diseases grant 61-6387, Michigan State University.

We sincerely thank Dr. Julia Bell for critical review of this article and valuable advice.

166

10.

REFERENCES

Aggarwal, B. and G., JU. 1992. Other cytokines, p. 427-428, Human Cytokines,
Handbook for basic and clinical research. Blackwell Scientific Publications,
Cambridge, MA.

Ajuebor, M. N., C. M. Hogaboam, T. Le, and M. G. Swain. 2003. C-C
chemokine ligand 2/monocyte chemoattractant protein-1 directly inhibits NKT
cell IL—4 production and is hepatoprotective in T cell-mediated hepatitis in the
mouse. J Immunol 170:5252-9.

Bancherean, J. and M.E. Rybak. 1994. Interleukin 4, p. 99-126. In A. Thomson
(ed.), The Cytokine Handbook, Second ed. The Academic Press.

Berschneider, H. M. 1989. Development of normal cultured small intestinal
epithelial cell lines which transport Na and Cl. Gastroenterology 96:A41.

Fallon, P. G., P. Smith, E. J. Richardson, F. J. Jones, H. C. Faulkner, J. Van
Snick, J. C. Renauld, R. K. Grencis, and D. W. Dunne. 2000. Expression of
interleukin-9 leads to Th2 cytokine-dominated responses and fatal enteropathy in
mice with chronic Schistosoma mansoni infections. Infect Irnmun 68:6005-11.

Fireman, P. 2003. Understanding asthma pathophysiology. Allergy Asthma Proc
24:79-83.

Francis, D. H., P. A. Grange, D. H. Zeman, D. R. Baker, R. Sun, and A. K.
Erickson. 1998. Expression of mucin-type glycoprotein K88 receptors strongly
correlates with piglet susceptibility to K88(+) enterotoxigenic Escherichia coli,
but adhesion of this bacterium to brush borders does not. Infect Irnmun 66:4050-
5.

Friedman, C. R., Neimann, J., Wegener, H.C., and RV. Tauxe. 2000.
Epidemiology of Campylobacter jejuni infections in the United States and other
industrialized nations, p. 121-138. In I. Nachamkin, and Blaser, M.J (ed.),
Campylobacter, Second ed. American Society for Microbiology, Washington,
DC.

Gause, W. C., J. F. Urban, Jr., and M. J. Stadecker. 2003. The immune
response to parasitic helrninths: insights from murine models. Trends Immunol
24:269-7 7 .

Gonzalez-Vallina, R., H. Wang, R. Zhan, H. M. Berschneider, R. M. Lee, N.

0. Davidson, and D. D. Black. 1996. Lipoprotein and apolipoprotein secretion
by a newborn piglet intestinal cell line (IPEC-l). Am J Physiol 271 :0249-59.

167

11.

12.

13.

14.

15.

16.

17.

18.

19.

20.

Hess, D. J., M. J. Henry-Stanley, E. A. Erickson, and C. L. Wells. 2002. Effect
of tumor necrosis factor alpha, interferon gamma, and interleukin-4 on bacteria-
enterocyte interactions. J Surg Res 104:88-94.

Hu, L., and Kopecko, DJ. 2000. Interactions of Campylobacter with eukaryotic
cells: Gut luminal colonization and mucosal invasion, p. 191-216. In I.
Nachamkin, and Blaser, M.J (ed.), Campylobacter, Second ed. American Society
for Microbiology, Washington, DC.

Ishida, Y., T. Maegawa, T. Kondo, A. Kimura, Y. Iwakura, S. Nakamura,
and N. Mukaida. 2004. Essential involvement of [FN-gamma in Clostridium
diflicile toxin A-induced enteritis. J Immunol 172:3018-25.

Kervella, M., J. M. Pages, Z. Pei, G. Grollier, M. J. Blaser, and J. L.
Fauchere. 1993. Isolation and characterization of two Campylobacter glycine-
extracted proteins that bind to HeLa cell membranes. Infect Irnmun 61 :3440-8.

Kiehlbauch, J. A., R. A. Albach, L. L. Baum, and K. P. Chang. 1985.
Phagocytosis of Campylobacterjejuni and its intracellular survival in
mononuclear phagocytes. Infect Irnmun 48:446-51.

Konkel, M. E., S. G. Garvis, S. L. Tipton, D. E. Anderson, Jr., and W.
Cieplak, Jr. 1997. Identification and molecular cloning of a gene encoding a
fibronectin-binding protein (CadF) from Campylobacterjejuni. Mol Microbiol
24:953-63.

Lawrence, C. E., J. C. Paterson, L. M. Higgins, T. T. MacDonald, M. W.
Kennedy, and P. Garside. 1998. IL-4-regulated enteropathy in an intestinal
nematode infection. Eur J Immunol 28:2672-84.

Madden, K. B., K. A. Yeung, A. Zhao, W. C. Gause, F. D. Finkelman, I. M.
Katona, J. F. Urban, Jr., and T. Shea-Donahue. 2004. Enteric nematodes

induce stereotypic STAT6-dependent alterations in intestinal epithelial cell
function. J Immunol 172:5616-21.

Mansfield, L. S., D. T. Gauthier, S. R. Abner, K. M. Jones, S. R. Wilder, and
J. F. Urban. 2003. Enhancement of disease and pathology by synergy of

T richuris suis and Campylobacterjejuni in the colon of immunologically naive
swine. Am J Trop Med Hyg 68:70-80.

McSweegan, E., and R. 1. Walker. 1986. Identification and characterization of

two Campylobacterjejuni adhesins for cellular and mucous substrates. Infect
Irnmun 53:141-8.

168

21.

22.

23.

24.

25.

26.

27.

28.

29.

Monteville, M. R., and M. E. Konkel. 2002. F ibronectin-facilitated invasion of
T84 eukaryotic cells by Campylobacterjejuni occurs preferentially at the
basolateral cell surface. Infect Irnmun 70:6665-71.

Nachamkin, I., Allos, B.M and Ho, T.W. 2000. Campylobacterjejuni infection
and the association with Guillain-Barre Syndrome, p. 155-178, Campylobacter,
Second ed. ASM press, Washington DC.

Rivera-Amill, V., B. J. Kim, J. Seshu, and M. E. Konkel. 2001. Secretion of
the virulence—associated Campylobacter invasion antigens from Campylobacter
jejuni requires a stimulatory signal. J Infect Dis 183:1607-16.

Shea-Donohue, T., C. Sullivan, F. D. Finkelman, K. B. Madden, S. C. Morris,
J. Goldhill, V. Pineiro-Carrero, and J. F. Urban, Jr. 2001. The role of IL-4 in

Heligmosomoides polygyrus-induced alterations in murine intestinal epithelial cell
function. J Immunol 167:2234-9.

Skirrow, M., and Blaser, M.J. 2000. Clinical aspects of Campylobacter
infection, p. 69-88. In I. Nachamkin, and Blaser, M.J (ed.), Campylobacter,
Second ed. American Society for Microbiology, Washington, DC.

Suzuki, K., H. N akajima, K. Ikeda, Y. Maezawa, A. Suto, H. Takatori, Y.
Saito, and I. Iwamoto. 2003. IL-4-Stat6 signaling induces tristetraprolin
expression and inhibits TNF-alpha production in mast cells. J Exp Med 198:1717-
27.

Tracey, K. 1994. Tumour Necrosis Factor-Alpha, p. 289-304. In A. Thomson
(ed.), The cytokine handbook, Second ed. Academic Press, San Diego.

Yao, Y., J. K. Eshun, S. Lu, H. M. Berschneider, and D. D. Black. 2002.
Regulation of triacyl glycerol and phospholipid trafficking by fatty acids in
newbom swine enterocytes. Am J Physiol Gastrointest Liver Physiol 282:G817-
24.

Zhou, Y., G. Lin, M. J. Baarsch, R. W. Scamurra, and M. P. Murtaugh.

1994. Interleukin-4 suppresses inflammatory cytokine gene transcription in
porcine macrophages. J Leukoc Biol 56:507-13.

169

Table 4. l. Translocation of C. jejuni to the baso-lateral surface after 3 hr incubation with
IPEC-l cells pretreated with either rIL-4 (500 pg/well) or RPMI medium for 5 hr. Results

shown are from a single experiment.

 

Number of bacteria translocated to the baso-lateral chamber

after 3 hours incubation (CFU)

 

rIL-4 pretreatment (500pg/well) for Medium pretreatment for 5 hr

5hr

 

C. jejuni 2 0

81176

 

 

 

 

 

170

Figure 4. 1. The effect of pretreatment of rIL-4 (500 pg/well) and rIL-lO (1000 pg/well)
on C. jejuni invasion of IPEC-l cells. Results shown are representative of five to seven
independent experiments. All treatments were in triplicate. Values shown are mean +/-

SEM. (*, p < 0.05).

1200

 

 

 

l
1000 i
l
l
l

if; jejuni 352952 {
‘IC.jejuni 81176 ‘

 

 

Internalized bacteria (cfu)

 

 

 

 

‘8 '\~ ”T *-
Efif Eégé’ Eaie
s— :25: g“o£
3.436 Ewa-T, “no;
2;“ REV; EEC-a
0.. n. a.

Treatment groups

171

Figure 4. 2. The effect of rIL-4 antibody on invasion of C. jejuni of IPEC-l cells. 12 day-
old IPEC-l cells on transwells were treated with rIL-4 +/- anti-IL-4 antibody or medium
alone for 5 hrs. Gentamicin assay was carried out as described in Materials and Methods.
All treatments were in triplicate. Results shown are representative of two identical,

independent experiments. C”, p < 0.01).

 

 

 

 

 

 

 

 

 

 

1200
1000 ‘
g l
9-1
3 800 j
a ,
.= ‘
8 .
g 600 “ " ’ ” ’ V
.9 1 ‘IC. jejuni 33292 ‘
g 6 egrerungsmé.
g 400 T .
h- l
3 l
= l
u l
200 .
l
l
o - 47* 77 ~- “ ”ans-7 4
Media rIL-4 (500 rIL-
pg/ well) 4(500pg/
well)and
lL-4 anti-
body
(100ul)

Pretreatment of IPEC-l for 5 hr

172

Figure 4. 3. Effect of increasing amounts of rIL-4 on C. jejuni invasion of epithelial
cells. The IPEC-l cells were treated with 100 pg/well, 200 pg/well, 300 pg/well, 400
pg/well, and 500 pg/well of rIL-4 for 5 hrs. Medium alone treatment was used as a
control. Gentamicin assay was carried out as described in the Materials and Methods.
Values shown are mean +/- SEM. All treatments were in triplicate. Results shown are

representative of two identical experiments.

1800 ,1, 7 7 ~~~\h ,, 3;;5; , ,
llC. jejuni 33292
ifli‘flmflllfl W 7

 

 

 

Internalised bacteria (cfu)

 

 

 

Pretreatment of IPEC-l for 5 hr

173

Figure 4. 4. Differential effect of rIL-4 on E. coli DHSa and C. jejuni invasion of IPEC-
1 cells. Gentamicin killing assays were carried out as described in Materials and
Methods. All treatments were in triplicate. Values shown are mean +/— SEM. (*, p <

0.05). Results are representative of two identical experiments.

1200

 

1000 ~77 a“ _,, 7 7 , 7777777

    

00

O

o
F

 

1*- Pretrnaagnentathimedia Eryhr
l

l I Pretreatment with rIL-
* l jQOQPE‘Y‘fl {0&5 his , , _l

O\
O
o
1f

l

l

l

l

l

l

l

l

l

l

l

l

c.
O
o

 

Internalized bacteria (cfu)

N
o
o

l

l

l

l

l

I

l

l

l

l

l

l

l

l

l

l

 

 

 

E. coli DHS alpha C. jejuni 81176

Treatments

174

Figure 4. 5. Transepithelial electrical resistance (TER) across the 12 day-old IPEC-l
monolayer before and afier treatment with rIL-4, rIL-lO or medimn for 5 hrs. All
treatments were in triplicate. Results shown are representative of three to four

experiments. Values shown are mean +/- SEM. (**, p < 0.01).

 

120

 

100 -

 

 

 

”LI—3.3? % }

 

 

 

TER %

 

 

 

 

 

 

 

 

 

 

 

- l " ‘7 . l
rIL-4 l rIL-lO I Media ‘
(SOOpg/well) | (IOOOpywell) i l
TER before TER after 5 hr treatment
l treatment

 

 

175

Figure 4. 6. Light microscopy of IPEC-l cells on transwell membranes as seen through
the baso-lateral surface. A) before any treatment, in medium, B) after treatment with
fresh medium for 5 hr, and C) after treatment with rIL—4 (500 pg/well) for 5 hours. Bars

represent 5 um. Micrographs are representative of two to three samples analyzed for each

panel.

176

 

177

Figure 4. 7. Effect of rIL-4 pretreatment on adherence of bacteria to IPEC-l. Adherence
assays were carried out as described in Materials and Methods. Values shown are mean
+/- SEM. Results are representative of two identical, independent experiments. (**, p <

0.01; ***, p < 0.001).

350

 

250 '"

 

 

llfiopretreatmgnt 7

 

 

 

Cell associated (adherent) bacteria x 1 0 ‘3

   

200 l
150 llPretreatment with rlL-4 ‘
1‘ 7(5QOEg/viell)ifor75i1r "l
100 ~
50 <
o .

c. jejuni 81176 c. jejuni 33292 E.co|i 0157:H43
DEC7A

Bacterial strains

178

Figure 4. 8. Transmission electron micrographs of IPEC-l cells pretreated with rIL-4
(8A) or medium (8B), followed by gentamicin killing invasion assays with C. jejuni
strain 33292. Multiple bacteria were seen just beneath the plasma membrane (arrows)
following rIL-4 pretreatment but not in the medium pretreated cells. Bars represent 1 pm.

Panels are representative of micrographs from seven samples each from two identical

experiments.

 

179

Figure 4. 9. Transmission electron micrographs of IPEC-l cells pretreated with rIL—4,
followed by gentamicin killing assays with C. jejuni 33292. Micrographs in the upper
row show bacteria in the cytoplasm or adjacent to vacuoles (arrows); bars represent 500

nm. Micrographs in the lower row are magnifications of the ones directly above; bars

represent 100 nm. Panels are from micrographs representative of seven samples.

 

180

Figure 4. 10. MTT assays afier IL-4 treatment and medium treatment. Assays were
carried out as described in the Materials and Methods. Values shown are mean +/- SEM.

All treatments were in triplicate and are representative of two independent experiments.

 

 

0.8
0.7
0.6
0.5
0.4
0.3
0.2

 

 

 

 

 

 

 

 

 

 

OD560nm

 

 

 

 

 

 

 

 

 

 

 

   

 

O +— ‘-t g ”f
rlL-4 Pretreatment (500 Media Pretreatment
pg/well)

Treatment groups

181

CHAPTER 5

SUMMARY AND CONCLUSIONS

182

The long-term goals of the laboratory are to determine the mechanisms of
pathogenesis in the dual infection of swine mediated by C. jejuni and T. suis. We
hypothesized that one of the mechanisms of pathogenesis could be an immune system
dysregulation in cytokine responses to C. jejuni and T. suis infections, resulting in the
pathology seen. Under this hypothesis, C. jejuni stimulates a predominantly Th-l
cytokine response accompanied by a moderate Th-2 component, while T. suis stimulates
a predominantly Th-2 response from the porcine host. The Th—2 responses due to
stimulation with both C. jejuni and T. suis act synergistically, causing a shift to a
predominantly Th-2 profile and downregulation of Th-l mediated immunity to the
bacteria. Since C. jejuni is a known opportunistic pathogen (12, 23), down-regulation of
the Th-l response could facilitate C. jejuni-mediated pathology. A primary swine
intestinal epithelial cell line, IPEC-l, was utilized to test these hypotheses, since
epithelial cells form the first barrier to this primarily intestinal disease.

The secretion of H.-4, IL-6 and IL-10 Th-2 cytokines from IPEC-l was tested in
response to C. jejuni and T. suis Excretory Secretory Products (ESP). Since T. suis ESP
has anti-microbicidal properties against C. jejuni (1), the cytokine responses to both these
challenges were tested separately. Both differentiated and undifferentiated epithelial cells
were tested for cytokine production to these pathogens to mimic the natural forms of
these eukaryotic cells that exist in vivo. IL-6 and IL—10 were both secreted in response to
C. jejuni infection. IL-6 was secreted within 24 hr in response to C. jejuni, while IL-lO
induction took > 24 hr. In addition, IL-6 was secreted fi'om both differentiated and
undifferentiated IPEC-l cells, while IL-10 was secreted predominantly from

differentiated cells compared to negative controls. In all cases, secretion of IL-6 and IL-

183

10 occurred on the cell surface exposed to the bacteria. Also, C. jejuni strain 33292
induced base-lateral induction of IL-10 while strain 81176 did not. T. suis ESP, on the
other hand, induced IL-6 and IL-10 within 24 hr, and both differentiated and
undifferentiated cells secreted these cytokines in response to a low level ESP (0.3 mg/ml)
challenge. There was no secretion of 11.4 from lPEC-l cells. These data suggest that
IPEC-l cells are a significant source of IL-6 and IL-10 innate immune responses to
enteric infections, but not adaptive responses like IL-4. It is also likely that C. jejuni LPS,
a major outer surface component, and an LPS-like substance in ESP can serve as inducer
of these cytokines. LPS has been shown to act through Toll—like receptors (TLR),
particularly TLR 4 and upregulate cytokine expression through the transcription factor
NFKB (10). Exposure of [PEG] cells to purified LPS from C. jejuni and subsequent
measurement of cytokines could test this hypothesis. Also, nuclear extracts from these
treated IPEC-l cells could be made and immuno-blot analysis for NFKB conducted to
determine if this transcription factor is recruited for induction of these cytokines. A study
to determine the presence of a LPS-like substance in ESP has been attempted, and
revealed an LPS-like substance capable of recruiting NFKB in Chinese Hamster Ovary
cells (CHO), and inducing TNFor secretion from RAW 264. 7, a mouse macrophage cell
line. In addition, T. suis ESP also potentiated LPS mediated endotoxic shock response in
mice, suggesting a role in immunomodulation similar to LPS (26).

The aim of these in vitro studies was to determine if there could be potential
synergy in any of these cytokines. The data suggests that there may be synergy in IL-6
(and perhaps IL-lO) innate cytokine responses in vivo during dual infections. The data

also suggests that intestinal epithelial cells could contribute substantially to the cytokine

184

responses in vivo, since they form a large continuum throughout the gut. The presence of
preformed IL-6 in IPEC-l cells also was demonstrated, suggesting that IL-6 secretion
could be a primary immune response to these enteric infections.

IL-4, IL-6 and IL-10 responses were tested during in viva infections of swine with
C. jejuni and T. suis. Experiments were conducted in two different ways. Short—term
experiments involved simultaneous oral infection of swine with C. jejuni and T. suis
eggs. In long-term experiments swine were infected orally with T. suis eggs 21 days prior
to infection with C. jejuni. H.-4 and IL-10 cytokine concentrations in feces were
significantly increased during the long-term infections in the dual infection, while IL—6
concentrations in feces increased in the short-term dual infections. Also, IL-4 and IL-1 0
levels were elevated 2-fold and 3-fold, respectively, during the long-term dual infections,
while IL-6 levels were elevated 2-fold during the short-term dual infections. Because
pathology is seen 21 days after introduction of T. suis during natural infections of swine,
it is reasonable to assume that long-term experiments are more reflective of the natural
infections that result in the observed disease and pathology than the short-term
experiments. Hence, IL-4 and IL-10, which were upregulated during the long-term
experiments, were hypothesized to play a role in pathology mediated by these two
organisms together. In addition, based on the analysis of singly and dually infected pigs,
the pattern of secretion was mainly reflective of T. suis-induced responses, suggesting
that the whipworrn was the predominant inducer of these cytokines. Also, as intestinal
epithelial cells predominantly secreted IL-6 in vitro, accompanied by a moderate IL-1 0
component, upregulation of IL-4 during long-term infections suggests that other cell

types like macrophages and lymphocytes likely secreted this cytokine. A previous

185

hypothesis from the in vitro experiments was that there might be synergy in innate IL-6
responses in viva during dual infections with C. jejuni and T. suis. IL-6 responses were
elevated 2 fold in 70% of pigs, compared to a slight induction in 40% of controls during
short-term concurrent infections. It is likely that intestinal epithelial cells contributed
significantly to this innate response.

In previous studies with swine, increased invasion of C. jejuni was seen in both
the follicle associated epithelium (F AE) of the lymphoglandular complexes (LGCs) and
in the crypt epithelial cells in the proximal colon around the worms (15). Since IL-4 and
IL-10 cytokine responses were elevated during long-term dual infections in swine, the
possibility that IL-4 or IL-10 might play a role in enhanced invasion of C. jejuni was
tested. Differentiated IPEC-l cells grown on transwell membranes to mimic the FAE
were pretreated with rIL-4 or rIL—lO, and infection assays with C. jejuni were carried out.
Pre-treatment with rIL-4 enhanced C. jejuni invasion ~6 fold, while pretreatment with IL-
10 had no effect. The rIL-4 mediated effect was dose dependent, and invasion increased
with increasing doses. Thus, although only moderate increases (2-fold) in IL-4 levels
were seen in pigs dually infected with C. jejuni and T. suis (chapter 3), this data suggests
that this moderate increase might be sufficient to enhance invasion of C. jejuni in vivo.
Because invasive strains of C. jejuni have been shown to be associated with inflammatory
diarrhea (13, 18), this enhanced invasion during dual infection might contribute to the
inflammatory diarrhea seen (15). rIL-4 also decreased the transepithelial resistance
across the IPEC-l monolayer by 5 hr. It has been suggested in other studies that IL-4
mediated increased permeability might loosen worms from the mucosa and enable worms

to be washed away (21). A similar result here suggests that IL-4 might have a similar role

186

in loosening T. suis from the intestines in viva. A study conducted at USDA by Dr.
Helene Kringel and others also corroborated these results. They found IL-4 mRNA to be
increased 3—6 fold, 21 days after T. suis infection in swine (personal communication).
This increase was seen in the proximal colon and the LGCs, but not in the distal colon.
Since the proximal colon and LGCs are the areas where enhanced C. jejuni invasion was
seen in previous studies (15), and since cytokine responses in viva seem more closely
associated with T. suis infection, this suggests that T. suis induces IL-4 during natural
infections, enabling enhanced invasion of C. jejuni. rIL-4 also enhanced vacuole
formation in IPEC-l cells. In previous studies, C. jejuni has been shown to be located
inside vacuoles, as well as demonstrated to invade through the baso-lateral surface of
human intestinal epithelial cells (9, 16). The presence of numerous vacuoles after rIL-4
pretreatment together with the increase in permeability, suggested that C. jejuni might
increasingly translocate to the base-lateral surface, and reside inside these vacuoles after
rIL-4 treatment. However, translocation experiments did not reveal increased
translocation of the C. jejuni strain 81176 to the baso-lateral side. In addition,
transmission electron microscopy (TEM) showed that bacteria were largely seen just
under the apical surface rather than near the base-lateral suface. Also, bacteria were
predominantly seen in the cytoplasm and adjacent to vacuoles rather than within them.
However, C. jejuni might have escaped the vacuoles. Also, it is possible that cell type
influenced this phenomenon since IPEC-l cells are of non-human origin, and the cell
lines used in other published studies were human cell lines, C. jejuni invasion
mechanisms might differ according to species. To test these hypotheses, TEM

examination of IPEC-l cells at more time points is needed. Also, invasion assays with or

187

without rIL-4 pretreatment in a few human and non-human epithelial cells need to be
done to compare and correlate invasion mechanisms by C. jejuni.

Since IL-6 is induced almost instantaneously with C. jejuni infection (Chapter 2),
its role in IL-4 enhanced invasion was also investigated. IPEC-l cells pretreated with rIL-
4 were subjected to invasion assays with C. jejuni, with the simultaneous addition of anti-
IL-6 antibody along with the bacteria. This was done to determine if concurrently
produced IL-6 potentiates or decreases invasion. However, mixed results were obtained
in vitro. In one experiment, addition of IL-6 antibody enhanced invasion of C. jejuni into
[PEC-l cells 2-fold, suggesting that IL-6 behaved in a pro-inflammatory manner, keeping
invasion levels low (data not shown). However, a repeat of this experiment showed no
change in invasion levels. Hence the role of IL-6 in invasion is inconclusive at this point,
and can be made clear through repeats of this experiment or with use of different cell
lines including those derived from humans.

IPEC-l is a primary epithelial cell line that we used for in vitro studies. The
immune responses are more likely reflective of in viva immune responses because these
cells are a primary cell line rather than a transformed cell line. The data from this thesis
study and concurrent studies in the laboratory suggests that the predominant cytokines
secreted from this cell line are IL-6, IL-8 and IL-10, IL-1 [3, War, but not IL-4 or IFNy
(L. D. Cunningham, personal communication). At the beginning of this study, almost
nothing was known about innate cytokine secretion responses from IPEC-l cells. At
present, we have made some headway in understanding the cytokine repertoire of this cell
line. However, other cytokine responses still need to be explored. Based on the current

data, this primary cell line makes a good model for understanding the patterns and

188

regulation of innate cytokine responses in swine intestinal epithelium to multiple enteric
infections.

At the beginning of this study, my hypothesis was that there is synergy of Th-2
cytokine responses induced in response to both C. jejuni and T. suis, and this
overproduction of Th-2 cytokines down-regulates the Th-l responses to the bacteria.
Indeed, we did see upregulation of two primary Th-2 cytokines IL-4 and IL-10 in the
long-term experiments in the dual infection. However, only part of this hypothesis was
accepted. Other studies done concurrently indicate that there is upregulation of IL-1 B and
TNF-at as well during long-term infections (Cunningham, LD and LS. Mansfield,
unpublished data). Therefore, the pathology seen during dual infection with C. jejuni and
T. suis may be due to simultaneous production of both proinflarnmatory and Th-Z
cytokines like TNF-a, IL-IB, IL-4 and IL-10, with individual cytokines playing a role in
clinical symptoms. IL-4 enhanced C. jejuni invasion of IPEC-l cells, while IL-10 and
TNF-a did not (Cunningham LD, personal communication). However, IL-lO has been
shown to downregulate IFN—y from macrophages (2), IL—lB in inflamed tissues and TNF-
(X. in colitis models (5, 25). In addition, IL-1 (3 has been shown to potentiate the effect of
TNF-a in mice (5). Since dual-infection with C. jejuni and T. suis in swine is
characterized by inflammation and colitis as well, this suggests that We and IL-1 B
may play a role in this aspect of pathology during dual-infection (Figure 5. 1).

In conclusion, a few novel results have been seen from this dissertation, that shed
new light on T. suis and C. jejuni mediated pathogenesis in swine. We show here for the
first time that rIL-4 enhances invasion of pathogenic bacteria such as C. jejuni into

epithelial cells, which suggests that, in natural dual infections of swine, T. suis indirectly

189

contributes to C. jejuni pathogenesis by induction of IL-4. This has enormous impact, not
only in the veterinary field with respect to trichuriasis, but also in humans. Millions of
people worldwide, particularly in the developing world, harbor the human counterpart of
T. suis, T. trichiura (19, 20). T. trichiura also has been shown to induce IgE, IgG4 and
IgGl responses, which are IL-4 mediated (14, 17). In addition, T. suis has been shown to
produce patent infections in humans (3, 4). C. jejuni is a well known human pathogen.
Although helrninth infections are uncommon in the developed world, it might be prudent
and prophylactic to analyze for helrninth infections in feces in addition to bacteria, during
enteritis. This was borne out in a recent case study in Canada, where T. suis ova was seen
along with C. jejuni in the feces of patient with toxic mega-colon and renal failure (22).
Also, asthma, and other allergic diseases, which have etiology in Th-2 cytokine
responses, affect close to 100 million people worldwide (11). A recent study showed that
incidence of Chlamydia pneumaniae, an intracellular pathogen, increased in patients with
asthma. It was suggested that the increase in incidence rate might be due to “modified
susceptibility to the organism, due to changes in host-cell physiology” (8). If we
extrapolate results from our study, this suggests that induction of Th-2 cytokines like IL-4
during asthma might cause a similar secondary bacterial infection to occur. Other
potential invasive pathogens that can cause respiratory infections and intestinal
infections, respectively, include Mycobacterium tuberculosis, Streptococcus pnuemaniae,
Pseudamanas aeruginosa and Salmonella spp., Intracellularis bacteria and others. Thus
this IL-4 mediated enhancement of bacterial invasion has the potential to be extended as a
predisposing mechanism to multiple pathogens as well. Also, it has been suggested that

when the immune profile is dominated by a Th-l response, the resistance to HIV and M.

190

tuberculosis is increased, and impaired when the profile is slanted toward a Th-2
phenotype (6).

Recently, T. suis and other helrninths are being advocated as ameliorative agents
for inflammatory bowel disease, including Crohn’s disease and ulcerative colitis (7, 24).
Since these diseases have an etiology based on a Th-l response, and since helrninths
induce a predominant Th-2 response, helrninth eggs were given to patients with these
conditions to downregulate these responses. This treatment was effective for most of the
patients. However, given the results in this thesis study, this avenue needs to be followed
with caution. T suis and other helrninths might not have any adverse effects on their own.
However, our work suggests that patients must be carefully scrutinized for any potential
infection with enteric pathogens that might elicit adverse reactions.

There are several other novel results from these studies. We show for the first
time in this study, that T. suis ESP induces innate responses in epithelial cells cultured in
vitro, as well as the innate and adaptive cytokine responses in feces in swine challenged
with T. suis and C. jejuni (chapter 2 and chapter 3). Another novel contribution to science
from this research was the identification of pre-formed IL-6 protein in IPEC—l cells. This
is the first demonstration of stored IL-6 cytokine in intestinal epithelial cells. This
reiterates the fact that primary cell lines are likely to be physiologically different fi'om
immortalized cell lines. It also opens up a new avenue for study of protein synthesis,

storage and secretion of IL-6 and other cytokines in defined intestinal cells.

191

Future directions

Host-pathogen interactions play a vital role in the outcome of a disease state. The
aim of most researchers worldwide in the field of infectious diseases is to understand the
mechanisms underlying these interactions in order to develop new avenues for therapy or
create awareness of potential health hazards. The goal of this study was to delineate some
of the mechanisms underlying the pathogenesis of dual infection mediated by C. jejuni
and T. suis in swine. Also, C. jejuni is a significant human pathogen. These studies also
were designed to understand how the host controls this pathogen. Additionally, because
T. suis is currently being advocated as therapy for intestinal diseases (7 , 24), the immune
responses to this helrninth and its consequences need to be explored in detail before its
widespread use. We have shown from our study, that IL-4 produced likely in reponse to
T. suis infections can potentiate the invasion of C. jejuni. However, it is likely that other
mechanisms also occur concurrently, that contributes to the disease process. There are
multiple avenues open for follow up experiments, in order to understand the immuno-
pathogenic or microbial pathogenic mechanisms mediated by these two organisms, either
together or alone. Also, other hypotheses need to be addressed to refine our in vitro and
in viva models. Furthermore, while some of the immune responses from the IPEC-l cell
line were characterized from this study, very little is known of other physiological
responses from these cells. As swine intestinal system mimics humans very closely, this
primary cell line makes an effective system to study innate immunity of swine to other
enteric infections.

In chapter 2, we showed that C. jejuni strains 33292 and 81176 differed in their

ability to induce IL-10 responses from differentiated cells. Also, only differentiated cells

192

induced any significant IL-lO response over the negative controls while the
undifferentiated cells did not. The mechanisms underlying the differences between cells
and different strains of bacteria should be followed up, as this would inform us about the
development of host immune response, and factors contributing to functional variations
in C. jejuni respectively. Understanding the antigenic basis of host response to T richuris
is also important. The immune responses to T. suis ESP were tested, but not to
constitutive products of the worm. Worm homogenates could provide a useful antigen for
these studies. Also, only immune responses to epithelial cells were tested in our study.
Cell types making up the intestinal tract are complex and varied. Macrophage or
lymphocyte responses will be important components to study. They are likely to drive
immunoregulation in the gastrointestinal tract. Another important finding is that pre-
formed IL-6 was seen for the first time. Its intracellular location, synthesis, gene
regulation and secretion are some important areas for follow-up study.

In chapter 3, IL-4, IL—6 and IL-10 cytokine responses were measured in the feces.
In another concurrent study, IL-8, IL-1 [5 and TNFa were measured. While measurement
of these cytokines gave us some insight into the pathogenic mechanisms during the dual
infection, other cytokines such as IL-5, TGFB, IFNy, GM-CSF, IL-13 and IL-12 also
need to be measured to completely understand the immunoregulation of T. suis and C.
jejuni infections. Also, comparison of these in viva induced cytokines with similar in
vitro responses can help us understand the cell types predominantly driving the immune
responses. A similar in viva study like the one just conducted can be undertaken, and
mRNA responses over time in different tissues could be measured. This would tell us the

specific tissues that contribute to immune responses, and help us correlate the findings

193

from this thesis study to areas of pathology seen during dual infection of swine with C.
jejuni and T. suis, and also corroborate Dr. Kringel’s Real time PCR cytokine data.

In chapter 4, rIL-4 enhanced C. jejuni invasion of epithelial cells and increased
the permeability of the cells. One question here is whether the increase in intestinal
permeability is necessary and sufficient for increased bacterial invasion or does it need to
be accompanied by intracellular changes in the IPEC-l cells?, This would inform us on
the specific changes that need to occur in IPEC-l cells that precede enhanced invasion of
C. jejuni. Another avenue is to determine whether the action of rIL-4 is receptor
mediated, which receptors are bound by rIL-4, and which signal transduction pathways
are activated thereafter. This provides an important avenue for intervention strategies
with the dual infection. The downstream cytoln'ne responses in IPEC-l cells following
rIL-4 treatment can be pursued, as this would tell us the immunomodulation that needs to
take place in IPEC-l cells to facilitate invasion. Effect of rIL-4 and rIL-lO on other tissue
types like macrophages also can be investigated, since C. jejuni invasion of macrophages
also occurred during dual infection of swine with T. suis and C. jejuni. This would give
information on differential requirement of cytokines, if any, on macrophage invasion
when compared to epithelial cell invasion. Differential adherence of the C. jejuni strains
to IPEC-l was seen after rIL-4 pretreatment. The microbial and eukaryotic factors
contributing to this difference need to be explored. Since adherence is an important
colonization factor for C. jejuni, this avenue would inform the researcher on differential
factors contributing to adherence in C. jejuni and hence form a basis for vaccine studies.
Since the enhanced invasion of C. jejuni into IPEC-l was shown in vitro, the in viva

effect of rIL-4 on C. jejuni invasion needs to be measured by use of removable intestinal

194

tie-adult rabbit diarrhea (RITARD) models. This would validate our findings in a
physiologically relevant setting.

In addition to host-pathogen interactions, interactions between T. suis and C.
jejuni also need to be studied, especially as numerous C. jejuni were seen around the
worms in the proximal colon during dual infection (15). The possible role of the warm or
its homogenates in the regulation of bacterial virulence genes is one area for study, which
provides another dimension to this complicated phenomenon.

Other relevant studies extending beyond the dual infection of swine are
epidemiological studies on human patients with enteric bacterial infections to determine
if these patients harbor helrninths. This study would help explain if complications seen
during enteric infections in humans could arise from other concurrent infections like the
one seen in our study. A similar study could be conducted in swine farms as well, to
determine the rate of C. jejuni infections in clinical trichuriasis, in order to correlate our
experimental findings with those occurring naturally. Another epidemiological study
would be to determine the susceptibility of patients with allergic diseases to bacterial
infections. As mentioned before, close to 100 million people suffer from asthma (11), and
such a study would help correlate our findings to extra-intestinal infections as well.
Hence, numerous avenues are available for further study from the results obtained in this
dissertation. However, the following avenues will be described in detail.

1) Modulation of cytokine responses in IPEC-l cells following rIL-4 treatment.
2) Identification of IL-4 receptors on IPEC-l cells, and signal transduction pathways

mediated by rIL—4 leading to increased internalization.

195

3) Molecular basis for differential adherence of C. jejuni strains 33292 and 81176 on

IPEC-l surface.

Modulation of cytokine responses in IPEC-l following rIL-4 pretreatment.

The hypothesis in this study is that rIL-4 treatment causes down-regulation of
certain cytokines such as TNF-at, IL-6, IL-8, IL-1 B, or IFNy, leading to increased
invasion of C. jejuni. To evaluate this hypothesis multiple techniques can be employed.
Real time RT-PCR: 12-day old IPEC-l cells on transwells will be exposed to rIL-4 or
media for 5 hr, then cells will be challenged with C. jejuni, and the treatment removed.
RNA will be extracted from the IPEC-l cells using Trizol® reagent or Qiagen RNAeasy
kit. cDNA will be constructed using first strand Pro-Star RT-PCR kit (Stratagene) and
PCR carried out for TNF-at, IL-6, IL-8, IL-1 B, IFNy and IL-10 using real-time porcine
primers. The control in this study will be cells exposed to media alone. Gentamicin
assays will be conducted concurrently as additional controls. Another technique is
ELISA: 12-day old IPEC-l cells on transwells will be exposed to rIL—4 for 5 hr, then
cells will be challenged with C. jejuni and the treatment removed. RPMI media will be
applied in 400,41 volumes and incubated for 10 minutes, 30 minutes, 1 hr, 2 hr and 3 hr.
The media will be removed afler regular intervals, centrifuged at 6000 rpm for 10
minutes, at 4° C, transferred to fresh centrifuge tubes, and analyzed for TNF-a, IL-6, IL-
8, IL-1 B, IFNy and IL-10 by ELISA using kits from Biosource International, Camarillo,
CA. Controls in this experiment will be cells exposed to media alone, rIL-4 and rIL-4
antibody. Another alternative technique is Western blot analysis for proteins. Whole cell
extracts will be made, blotted onto PVDF or NCP membranes, along with standards, and

probed with antibody to each cytokine. However, since the transwell size might not be

196

suitable for whole cell extracts (6.4 mm diameter), a bigger transwell size will be needed,
or alternatively, IPEC-l cells could be exposed to rIL-4 growing in tissue culture flasks.
A valid control would be cells exposed to media alone.

This study would inform the researcher on cytokine responses being altered
following rIL-4 treatment, and would give information on the cytokines that are potential
targets for immunomodulation by IL-4 during dual infection of swine with C. jejuni and

T. suis.

Identification of receptors bound by rIL-4.

Here, the hypothesis is that rIL-4 mediated enhancement of invasion occurs
through IL-4Ra. Some of the experimental techniques that can be used to validate this
hypothesis are 1) Application of IL-4Ra antibody: Antibody to IL-4Ra will be applied
along with rIL—4 to the IPEC-l monolayer, and gentamicin killing assays canied out as
described previously. Cross-reactivity to rIL-4 needs to be ruled out by ELISA prior to
use. If rIL-4 mediated enhancement of C. jejuni invasion is through IL-4Roc, then
application of antibody to the receptor will not enhance invasion, and the bacterial
numbers will be lower than in rIL-4 treated positive controls. If so, then further
experiments can be done to verify this. A) Membrane preparations can be made, and
incubated with rH.-4 beads for 5 hr. The bound proteins are then dissolved, resolved on
an SDS-PAGE gel and analyzed by western blot using antibody to IL-4R0L. B) Knock
down of the IL-4Rat gene using siRNA technology and subsequent invasion assays with
rIL-4. 2) Chemical cross-linking assays and affinity columns: Whole cells or

membrane preparations will be incubated with radiolabelled and unlabelled rIL—4 or with

197

radiolabelled rIL-4 alone, washed and crosslinked using disuccimidyl substrate
(crosslinking agent), and gel shift assays performed. The protein bands will be excised
from the gel, reverse cross-linked and mass spectrometry conducted on the radioactive
complex, and the protein identified. Gel shift assays without crosslinking will be run
concurrently. Membrane fractions will also be run through rIL-4 affinity columns so that
the columns elute out the proteins bound by the cytokine. The protein band will be
excised, and subjected to mass-spectrometry and protein identified.

Identification of the IL-4 receptor provides a potential target for intervention
therapies, and also a route for studying signal pathways mediated by this receptor in

IPEC-l cells, of which little is known.

Molecular basis for differential adherence of C. jejuni strains 33292 and 81176 on IPEC-
1 surface.

The hypothesis for this study is that C. jejuni strains 33292 and 81176 bind
through different adhesins, or have differential concentrations of the same adhesins.
In order to identify the adhesins on C. jejuni 33292 and 81176, again several techniques
can be employed. 1) Outer membrane protein (0MP) preparations: OMPs will be
extracted from individual C. jejuni strains using the EDTA buffer method. The
preparations will be run on an SDS-PAGE gel and visualized using Coomassie stain. If
the same adhesin is differentially expressed, the band intensity would be expected to be
stronger in one strain than the other. If there are distinct band differences then, the protein
bands will be excised, and partially sequenced by Edman degradation reaction, and

partial N-tenninal sequence obtained. This sequence can be compared against the

198

published C. jejuni genome sequence to identify the genes for adhesins. Alternatively,
mass-spectrometry could be done on the excised protein band and the protein identified.
2) Construction of a phage display library and immunoscreening: A phage library for
both strains will be constructed individually, with each phage containing a piece of
bacterial genomic DNA. Anti-sera will be raised against individual bands in the OMPs
and each serum will be screened for blocked adherence. The phage library will then be
screened with sera that inhibited adherence, and phage colonies that react with sera
further selected for analysis. The bacterial DNA in these colonies will be amplified by
PCR, and sequenced, and compared against the published genome sequence to identify
relevant genes. 3) Random transposon mutagenesis: Random transposon mutants could
be constructed for each strain and screened for decreased adherence on IPEC-l cells. The
mutated gene would then be identified by PCR with primers made from the known
transposon or antibiotic markers. 4) Targeted gene mutation: Since pebI and CadF are
known adhesins for C. jejuni, these genes could be mutated individually, and screened for
decreased adherence on IPEC-l. 5) LPS analysis: As LPS is also a known adhesin for C.
jejuni, the LPS of each strain could be analyzed to detemrine if the difference in
adherence is LPS mediated. For this study, LPS will be extracted by phenol-water
extraction, and anti-sera generated in rabbits. The anti-sera could then be used in
adherence assays to determine if LPS antibodies have an effect on adherence. Controls
will be adherence assays using a non-specific antibody isotype control. If an effect is
seen, the extracted LPS could be run on tricine SDS-PAGE gel and stained with silver

stain and the subsequent pattern profiles analyzed for differences. In addition, serotypes

199

of the strains could be determined, and additional strains of similar serotypes analyzed for
correlation in adherence patterns.

Campylobacter strains show enormous strain-to-strain variation. Identification of
genetic or molecular factors that contribute to functional and phenotypic variations in
Campylobacter infections will help the researcher understand the multiple mechanisms
employed by this versatile bacterium in colonization and enable suitable intervention

strategies for the control of its pathogenesis.

200

Figure 5. l. The potential role(s) for each cytokine in viva during long-term dual

infection of swine with C. jejuni and T. suis.

POTENTIAL ROLE(S) OF SIGNIFICANTLY INDUCED CYTOKINES IN

LONG-TERM DUAL INFECTION IN VIVO

 

 

 

 

 

 

 

 

THIS STUDY OTHER STUDIES

IL—4 IL-10 TNF-a IL-lB
Increases No effect on C. Potential role: Potential role:
invasion of C. jejuni invasion of Colitis? Inflammation?
jejuni into IPEC- IPEC-l cells.

1 cells.

Other roles: Role in

Other roles: downregulating

Loosening of T. IFN-y?

suis from the

mucosa by

increased

permeability?

 

 

 

 

 

 

201

10.

ll.

12.

REFERENCES

Abner, S. R., G. Parthasarathy, D. E. Hill, and L. S. Mansfield. 2001.
T richuris suis: detection of antibacterial activity in excretory-secretory products
from adults. Exp Parasitol 99:26-36.

Aggarwal, B. a. G., JU. 1992. Other cytokines, p. 427-428, Human Cytokines,
Handbook for basic and clinical research. Blackwell Scientific Publications,
Cambridge, MA.

Beer, R. J. 1971. Experimental infection of man with pig whipworrn. Br Med J
2:44.

Beer, R. J. 1976. The relationship between T richuris trichiura (Linnaeus 1758)
of man and Trichuris suis (Schrank 1788) of the pig. Res Vet Sci 20:47-54.

Dinarello, C. 1994. Interleukin-1, p. 31-56. In A. Thomson (ed.), The cytokine
handbook, Second ed. Academic Press Ltd., San Diego,CA.

Fincham, J. E., and M. B. Markus. 2001. Human immune response to Trichuris
trichiura. Trends Parasitol 17:121.

Fox, J. G., P. Beck, C. A. Dangler, M. T. Whary, T. C. Wang, H. N. Shi, and
C. N agler-Anderson. 2000. Concurrent enteric helrninth infection modulates

inflammation and gastric immune responses and reduces helicobacter-induced
gastric atrophy. Nat Med 6:536-42.

Gencay, M., and M. Roth. 2003. Chlamydia pneumoniae infections in asthma:
clinical implications. Am J Respir Med 2:31-8.

Hu, L., and D. J. Kopecko. 1999. Campylobacterjejuni 81-176 associates with
microtubules and dynein during invasion of human intestinal cells. Infect Irnmun
67:41 71 ~82.

Janeway, C., Travers, P, Walport, M and JD, Capra. 1999. Signaling through
lymphocyte receptors, p. 163-194.171 P. a. E. L. Austin (ed.), Immunobiology-
The immune system in health and disease, Fourth ed. Elsevier Science Ltd and
Garland Publishing, New York.

Kemp, J. P. 2003. Recent advances in the management of asthma using
leukotriene modifiers. Am J Respir Med 2:139-56.

Kerstens, P. J., H. P. Endtz, J. F. Meis, W. J. Oyen, R. J. Koapmau, P. J. van
den Brock, and J. W. van der Meer. 1992. Erysipelas-like skin lesions
associated with Campylobacterjejuni septicemia in patients with
hypogammaglobulinemia. Eur J Clin Microbiol Infect Dis 11:842-7.

202

 

 

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

Konkel, M. E., and L. A. Joens. 1989. Adhesion to and invasion of HEp-2 cells
by Campylobacter spp. Infect Irnmun 57:2984-90.

Lillywhite, J. E., D. A. Bundy, J. M. Didier, E. S. Cooper, and A. E. Bianco.
1991. Humoral immune responses in human infection with the whipworm
Trichuris trichiura. Parasite Immunol 13:491-507.

Mansfield, L. S., D. T. Gauthier, S. R. Abner, K. M. Jones, S. R. Wilder, and
J. F. Urban. 2003. Enhancement of disease and pathology by synergy of

T richuris suis and Campylobacterjejuni in the colon of immunologically naive
swine. Am J Trop Med Hyg 68:70-80.

Monteville, M. R., and M. E. Konkel. 2002. F ibronectin-facilitated invasion of
T84 eukaryotic cells by Campylobacterjejuni occurs preferentially at the
basolateral cell surface. Infect Irnmun 70:6665-71.

Needham, C. S., D. A. Bundy, J. E. Lillywhite, J. M. Didier, 1. Simmons, and
A. E. Bianca. 1992. The relationship between T richuris trichiura transmission
intensity and the age-profiles of parasite-specific antibody isotypes in two
endemic communities. Parasitology 105 ( Pt 2):273-83.

 

Newell, D. G., H. McBride, F. Saunders, Y. Dehele, and A. D. Pearson. 1985.
The virulence of clinical and environmental isolates of Campylobacterjejuni. J
Hyg (Lond) 94:45-54.

Pedersen, S., and I. Saeed. 2001. Acquired immunity to Trichuris suis infection
in pigs. Parasitology 123:95-101.

Pedersen, S., and I. Saeed. 2000. Experimental infection of pigs with three dose
levels of T richuris suis. Parasite 7 :275-81.

Shea-Donahue, T., C. Sullivan, F. 11). Finkelman, K. B. Madden, S. C. Morris,
J. Goldhill, V. Pineiro-Carrero, and J. F. Urban, Jr. 2001. The role of IL-4 in

Heligmasamoides palygyrus-induced alterations in murine intestinal epithelial cell
function. J Immunol 167:2234-9.

Shin, J. L., G. W. Gardiner, W. Deitel, and G. Kandel. 2004. Does whipworm
increase the pathogenicity of Campylobacterjejuni? A clinical correlate of an
experimental observation. Can J Gastroenterol 18:17 5-7 .

Sorvilla, F. J., L. E. Lieb, and S. H. Waterman. 1991. Incidence of
campylobacteriosis among patients with AIDS in Los Angeles County. I Acquir
Immune Defic Syndr 4:598-602.

203

 

24.

25.

26.

Summers, R. W., D. E. Elliott, K. Qadir, J. F. Urban, J r., R. Thompson, and
J. V. Weinstock. 2003. Trichuris suis seems to be safe and possibly effective in
the treatment of inflammatory bowel disease. Am J Gastroenterol 98:2034-41.

Tracey, K. 1994. Tumour Necrosis Factor-Alpha, p. 289-304. In A. Thomson
(ed.), The cytokine handbook, Second ed. Academic Press, San Diego.

Wilder, S. R. 2003. T richuris suis excretory secretory products modulate host
immune responses. Masters. Michigan State University, East Lansing.

204

 

   

ulililjiilliri1111er