ROLES OF THE MICROBIOTA IN INFLAMMATORY
BOWEL DISEASES
By
Nabeetha Aarti Nagalingam

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

DOCTOR OF PHILOSOPHY

Microbiology and Molecular Genetics
2010

Abstract

ROLES OF THE MICROBIOTA IN INFLAMMATORY BOWEL DISEASES
By
Nabeetha Aarti Nagalingam
Studying the roles of the microbial community structure as it relates to disease
can give important insight into development of inflammation. This study employs
concepts present in ecology to understand how altering bacterial communities in
mammalian intestines affect inflammation at different stages of disease. The
overall goal is to understand the role of the resident microbiota in initiation and
progression of inflammatory bowel disease (IBD). Changes in the community can
possibly act as a biomarker to diseases allowing preventative action to be taken
or development of novel therapies.
Work was done on two different models of IBD, dextran sodium sulfate (DSS)induced murine colitis and Helicobacter hepaticus triggered inflammation in IL-10/-

mice (model of Crohn’s disease). Through culture independent techniques,

changes in the microbiota were investigated, together with alterations in host
mediators.
In the first model, whether DSS-induced colitis was associated with changes in
the microbiota was investigated. DSS administration was associated with
microbial changes in intestinal communities in mice and these changes were

ii

present at onset of inflammation and persisted through later stages of disease. In
the second model, whether changes in microbial community structure affected
disease severity was investigated. The H. hepaticus triggered disease model was
used to explore this question, and it was found that H. hepaticus infection of mice
with different microbial community structures resulted in similar disease
outcomes. This indicated that disease severity was independent from antibioticinduced changes in the microbial community in this model. We also show that the
microbiota is essential in disease induction in mice in our colonies, as H.
hepaticus monoassociated mice do not develop disease.
These experiments demonstrate that the microbiota has important roles at several
stages of disease, and these roles can be affected by other factors that influence
IBD. In investigating stages at which the microbiota plays a role in IBD, we can
better understand stages along disease progression when prognosis or treatment
can be most effective.

iii

Acknowledgements
I would first and foremost like to acknowledge my advisor, Dr. Vincent B. Young
for giving me the opportunity to work in his lab. He has been a great mentor and I
would like to extend my deepest thanks to him for guiding me through the years.
I would also like to acknowledge my co-advisor, Dr. Thomas M. Schmidt, who
took on the role of co-mentor when I re-located to the University of Michigan to
continue my research. His support has made working across two campuses
possible.
The guidance of my advisory committee, Dr. Todd Ciche, Dr. Paul Coussens, Dr.
Linda Mansfield and Dr. Martha Mulks, as well as collaborators Drs. John Y. Kao,
Patrick D. Schloss and Gary Huffnagle, were invaluable.
I would also like to thanks members of the lab, both past and present, for their
support and friendship.

iv

Table of Contents
LIST OF TABLES................................................................................................vii
LIST OF FIGURES .............................................................................................viii
Chapter 1 ..............................................................................................................1
The mammalian intestinal microbiota ...............................................................1
Introduction ........................................................................................................1
Effects of the microbiota on the host..................................................................2
Analysis of microbial communities .....................................................................7
Resisting and rebounding from change: Stability of the microbiota .................14
Influences on the microbiota ............................................................................17
The microbiota and disease .............................................................................25
Conclusion .......................................................................................................32
REFERENCES ....................................................................................................33
Chapter 2 ............................................................................................................46
Models of IBD.....................................................................................................46
Value of animal models....................................................................................47
DSS-induced colitis ..........................................................................................49
Helicobacter hepaticus infected IL-10-/- murine model of IBD.........................56
Conclusions......................................................................................................61
REFERENCES ....................................................................................................63
Chapter 3 ............................................................................................................72
Microbial ecology of the murine gut associated with the development of
Dextran Sodium Sulfate-Induced colitis..........................................................72
Abstract and Key Words: .................................................................................72
Introduction: .....................................................................................................74
Materials and Methods.....................................................................................75
Results .............................................................................................................82
Discussion:.......................................................................................................93
SUPPLEMENTAL INFORMATION .....................................................................99
REFERENCES ..................................................................................................110
Chapter 4 ..........................................................................................................120
The effects of intestinal microbial community structure on disease
manifestation in IL-10-/- mice infected with Helicobacter hepaticus. .........120
Abstract and Key Words: ...............................................................................120
Introduction: ...................................................................................................122
Materials and methods:..................................................................................124
Results: ..........................................................................................................131
Discussion:.....................................................................................................147
v

Acknowledgements ........................................................................................152
SUPPLEMENTAL INFORMATION ...................................................................153
REFERENCES ..................................................................................................152
Chapter 5 ..........................................................................................................164
Summary, synthesis and future directions ...................................................164
Future Directions............................................................................................166
The role of the microbiota in initiation of IBD .................................................168
Conclusion .....................................................................................................174
REFERENCES ..................................................................................................175
APPENDICES ...................................................................................................179
Appendix A.......................................................................................................180
The effects of cefoperazone on Helicobacter hepaticus colonization. ......180
Purpose..........................................................................................................180
Methods and materials...................................................................................180
Results and conclusions ................................................................................182
REFERENCES ..................................................................................................190
Appendix B.......................................................................................................191
The effects of cefoperazone administration in H. hepaticus triggered
disease in IL-10 defiecient mice. ....................................................................191
Purpose..........................................................................................................191
Methods and materials...................................................................................191
REFERENCES ..................................................................................................197

vi

LIST OF TABLES
Table 3.1 Showing distribution of OTU’s across treatments from major groups
within the Firmicutes.....................................................................................92
Table 3.2. Changes in host gene expression following DSS treatment...............99
Table 3.3. Taxonomic distribution of rRNA genes clones from RDP classifier
analysis of treatment group ........................................................................102
Table 3.4. Comparison of common terminal restriction fragment lengths
generated by in silico digests of clones and the terminal restriction fragment
lengths from T-RFLP via Msp1 digests. .....................................................109
Table 4.3. Fold change in expressions of host mediators from cecum tissue. All
values are normalized to GAPDH expression and compared to cefoperazone
untreated, H. hepaticus uninfected control group. Asterisk (*) denotes
statistically significant changes (p<0.05). ...................................................154
Table 5.1 Roles of the microbiota in at various stages of inflammatory bowel
disease. ......................................................................................................167	
  

vii

LIST OF FIGURES
Figure 3.1. Histopathology in DSS-treated mice. Hemotoxylin and eosin (H&E)
stained sections were prepared from cecum sections of (a) untreated control
mice (b) mice after 3 days of DSS treatment and (c) after 14 days of DSS.
Arrow indicates submucosl edema. H&E sections were also prepared from
colon samples of (d) untreated controls (e) mice after 3 days of DSS
treatment (arrow points to inflammation) and (f) animals after 14 days of
DSS (arrow shows abscess). Histopathologic scores were calculated for
sections from all 31 animals for (g) cecum and (h) colon sections. Statistical
analysis done was Kruskal-Wallis test. *p< 0.05 **p<0.001. Initial
magnification 40X. ........................................................................................83
Figure 3.2. Changes in host gene expression after 3-days (dark bars) or 14-days
(grey bars) ± standard deviation of DSS administration. Expression levels
were compared to expression in tissue from animals not exposed to DSS,
after normalization with GAPDH. Statistical analysis done was Student’s ttest. Statistically significant differences (p<0.05) are denoted by asterisks (*).
......................................................................................................................85
Figure 3.3. Comparison of cecal communities in control animals (black triangles)
and in animals following 3 days of DSS treatment (white squares) and 14
days of treatment (black squares). Using an OTU definition of 97% similarity,
the Bray-Curtis similarity metric was calculated for each pair-wise
comparison and then results displayed in denrogram format. Analysis by
parsimony tests indicate that there is a statistically significant (*p<0.001)
difference between the communities in control animals and in both groups of
DSS-treated animals. ...................................................................................88
Figure 3.4 (A). Changes in the cecal gut microbial community following DSS
administration at the phylum level. Analysis was done by ANOVA and
statistical significance (p<0.05) is denoted by asterisks (*). For
interpretation of the references to color in this and all other figures, the
reader is referred to the electronic version of this dissertation. ...........90
(B) Changes in the cecal gut microbial community following DSS
administration within the phylum Firmicutes. Analysis was done by ANOVA
and statistical significance (p<0.05) is denoted by asterisks (*). For
interpretation of the references to color in this and all other figures, the
reader is referred to the electronic version of this dissertation. ...........91

viii

Figure 3.5. Comparison of the cecal community in control animals (black
triangles), and animals following 3 days of DSS treatment (white squares)
and 14 days of treatment (black squares). Using T-RFLP, the Bray-Curtis
similarity metric was calculated for each pair-wise comparison and then the
results displayed in dendrogram format. Communities that were further
analyzed, by construction of clone libraries, are indicated by arrows.
Analysis by the parsimony test indicates that there is a statistically significant
difference (*p<0.05 and **p<0.001). ...........................................................104
Figure 3.6. The rarefaction curves at 97% sequence similarity comparing the
microbial communities in mice after 3-days of DSS treatment (pink), 14-days
of DSS treatment (blue) and the controls (yellow). For interpretation of the
references to color in this and all other figures, the reader is referred to
the electronic version of this dissertation. ............................................105
Figure 3.7. Comparisons between T-RFLP traces and histograms of the terminal
restriction fragments (TRFs) of in silico digest of the clone library constructed
from the control communities. ....................................................................106
Figure 3.8. Comparisons between T-RFLP traces and histograms of the terminal
restriction fragments (TRFs) of in silico digest of the clone library constructed
from the 3-day DSS administered communities. ........................................107
Figure 3.9. Comparisons between T-RFLP traces and histograms of the terminal
restriction fragments (TRFs) of in silico digest of the clone library constructed
from the 14-day DSS administered communities. ......................................108
Figure 4.1 Schematic of experimental designs showing (A) cefoperazone (Cef)
and (B) vancomycin administration to IL10-/- C57BL/6 mice prior to or during
infection with H. hepaticus (Hh)..................................................................126

ix

Figure 4.2 Helicobacter hepaticus triggered inflammation in conventionally raised
IL10-/- C57BL/6 mice compared to germfree counterparts. Representative
sections of mouse cecum from experimental groups are depicted. (A) H.
hepaticus uninfected conventionally raised mice appear normal. (B) H.
hepaticus infected conventional (Conv) section shows edema, hyperplasia
and inflammatory infiltrate. (C) H. hepaticus uninfected germfree control, and
(D) H. hepaticus infected germfree cecum sections are both normal without
signs of inflammation. (E) H. hepaticus forms close association to the
intestinal epithelium, and in the crypts (yellow) compared to other microbes
(red) in the lumen. (F) In germfree mice monoassociated with H. hepaticus,
the bacteria are also found in close association with the epithelial layer (G)
Quantification of indicators of pathology showing significant inflammation
only in H. hepaticus infected conventionally raised mouse sections. Asterisk
(*) denotes statistically significant changes (p<0.05). For interpretation of
the references to color in this and all other figures, the reader is
referred to the electronic version of this dissertation. .........................134
Figure 4.3 H. hepaticus infection, as well as vancomycin treatment, significantly
alters intestinal community structure. Asterisk (*) denotes statistically
significantly different community structures calculated by parsimony test
(p<0.05). For interpretation of the references to color in this and all
other figures, the reader is referred to the electronic version of this
dissertation. ..............................................................................................136
Figure 4.4 Blinded scores showing significant inflammation in all H. hepaticus
infected sections regardless of vancomycin treatment compared to
uninfected controls. Asterisk (*) denotes statistically significant changes
(p<0.05). .....................................................................................................140
Figure 4.5 H. hepaticus infection significantly alters the community structure of
mice even after four weeks of termination of cefoperazone treatment.
Asterisk (*) denotes statistically significantly different community structures
calculated by parsimony test (p<0.05). For interpretation of the references
to color in this and all other figures, the reader is referred to the
electronic version of this dissertation....................................................142
Figure 4. 6 Blinded scores showing significant inflammation in all H. hepaticus
infected sections regardless of cefoperazone treatment compared to
uninfected controls. Asterisk (*) denotes statistically significant changes
(p<0.05). .....................................................................................................144

x

Figure 4.7 Changes in expression of host mediators in cecum tissue of mice
administered cefoperazone and infected with H. hepaticus (p<0.05).........146
Figure A1.2 Cefoperazone concentrations above 0.25 mg/ml cause significant
decreases in total bacterial loads and eradicates H. hepaticus. ................184
Figure A2.1 Experimental design showing H. hepaticus infection prior to
cefoperazone treatment. ............................................................................192
Figure A2.2 Photographs of ceca removed from mice from four treatment groups.
(Top left) Cecum is normal with luminal content of paste-like consistency,
unlike cecum from mice infected with H. hepaticus that have thick tissue (top
right). With cefoperazone administration (bottom panels), the cecum become
distended and tissue fragile, filled with dark luminal fluid. For interpretation
of the references to color in this and all other figures, the reader is
referred to the electronic version of this dissertation. .........................194
Figure A2.3 show significant inflammation in mice infected with H. hepaticus.
After administration of cefoperazone, inflammation was attenuated..........196	
  

xi

Chapter 1

The mammalian intestinal microbiota

Introduction
The human intestinal microbiota is comprised of seven of 70 known bacterial
phyla: Firmicutes, Bacteriodetes, Actinobacteria, Proteobacteria, Verrucomicrobia,
Fusobacteria and Cyanobacteria; and one of 13 archeal divisions: Euryarchaeota
(1, 2). Of these, Bacteriodetes and Firmicutes are most dominant. This may
suggest that communities are not very diverse (low richness and evenness),
especially when compared to those in soils or oceans that have over 12 major
phyla represented (3, 4), but this is far from the truth when genus or specieslevels are investigated. There are more than 400 species identified in the
mammalian intestines, which were found to contribute to host development (5, 6).
The participation of this microbial community is integral to our well-being and we
are sometimes referred to as a “super-organism,” as both the microbes and host
work in unison. This mutualism, however, is only maintained when the microbial
community possesses a particular structure, composition, richness and evenness
i.e. components of diversity. If this balance is perturbed, health can be
compromised (7-9). Fortunately, studies on the diversity of the intestinal
community demonstrate its resistance and resilience to external pressures (7, 10).

1

This review attempts to summarize roles of the mammalian gut microbiota in
health and disease. As previously mentioned, a stable microbial community
maintains health; therefore, how various factors change this structure will be
discussed, as well as techniques used to measure these changes. Also,
inflammatory bowel disease (IBD), a gastrointestinal condition, will be used as an
example of a disease linked to changes in resident microbial communities. But
first, I will address direct contributions of the microbiota to hosts in maintaining
health and promoting proper host development.

Effects of the microbiota on the host
The microbiota is very active in molding its host. Gut microbes function to
activate host genes in order to facilitate their own survival, as in the case of
Bacteroides thetaiotaomicron activating production of defensins to kill potential
pathogens and prevent competition for available space (11); or to benefit hosts
via metabolism of previously unavailable nutrients such as butyrate (12).
Microbial communities can also contribute to the host’s physiological and
immunological development.

Microbiota on host gut physiology
There is a wealth of information on the benefits of intestinal microbiota on host
development. A review by Wostmann summarizes a multitude of studies done on
mapping differences between germfree animals and conventionally raised ones
2

(13). Some of these differences include increased ceca size, decreased intestinal
weight and decreased colon surface area in germfree mice compared to their
conventionally raised counterparts (14, 15). This might be due to the absence of
butyrate-producing bacteria since this short chain fatty acid is the sole energy
source for colonocytes, and lack of butyrate would lead to poor cellular
development in germfree animals. Germfree mice also had less fat deposition
seen post-mortem, indicating bacteria were important in lipid metabolism (16).
Microbes contribute greatly to development and maintenance of healthy gut
tissue. Intestinal architecture of germfree animals shows poor development of
colonocyctes, as noted earlier. This is because these cells need the short-chain
fatty acid, butyrate, as an energy source to proliferate, which is produced by the
microbiota (17). A study also reported shorter crypt lengths and increased number
of goblet cells that secreted mucus associated with the presence of a microbial
community (18). Mucin is also very important in the barrier function of intestinal
epithelium. Presence of a resident microbiota can stimulate production of
polysaccharides and enhance the protective ability of the gut by providing a
thicker layer between potentially harmful bacteria, or toxic components, and the
host itself. Germfree mice also possess leaky tight junctions when compared to
mice colonized with bacteria. Tight junctions are composed of a protein mesh that
seals intercellular spaces in the epithelial layer. One study in particular shows that
colonization with the probiotic Escherichia coli Nissle 1917 caused increased
expression of the protein zonula occludens 1, involved in the formation of tight
junctions (19). This leads to improved barrier function and protection against
3

invasion by pathogens or antigens, and underscores the contribution of “beneficial
bacteria” to the host.

Microbiota and the immune system
The intestinal tract is an interface between hosts and their environments. As such,
it is a location where cues are presented to the immune system from the external
environment. Thus it is necessary that host’s defenses respond to appropriate
signals; otherwise we would be constantly inflamed. It is seen, through work with
germfree animals, that the intestinal microbiota is key in educating the immune
system.
Germfree mice were found to have fewer Peyer’s patches, thinner lamina propria,
fewer plasma cells in germinal centers and fewer lymphoid follicles as reviewed
by Round et al. (20). These mice also contained fewer T cells, paneth cells with
reduced gene expression and B cells that produced less IgA compared to
conventionally raised mice. All of these show importance of the resident
microbiota in immune system development. Paneth cells are located in crypts and
secrete defensins, small antimicrobial peptides. These cells also express RegIIIγ,
which is involved in killing gram-positive bacteria. Use of antibiotics was shown to
lower expression of this gene (21). This indicates that expression of this C-lectin
may depend on the gut microbiota and emphasizes the degree of cross-talk
between host and microbes in maintaining homoeostasis.

4

There are several mechanisms known for immunological responses elicited by
pathogenic bacteria, such as effector systems on pathogenic islands like the Type
III secretion system possessed by many pathogens such as Yersinia pestis, or
flagella of Salmonella (22, 23). However, limited information is present for why
resident bacteria do not trigger the host’s immune system and thereby can coexist with the host. Studies have cited various mechanisms that particular bacteria
have evolved to suppress immunological responses. Many bacteria act on
inhibiting NF-κB activation. Bacteroides thetaiotaomicron, for instance, is known
to act down stream of TLR receptor signaling and NF-κB activation (24). Intestinal
epithelial cells were incubated with Salmonella alone or Salmonella and resident
microbe, B. thetaiotaomicron. It was found that with Salmonella alone, there was
much less Rel A, the transcriptionally active subunit of NF-κB, in nuclei of host
cells. This indicates that B. thetaiotaomicron is involved in export of this subunit
and thus limits activity of NF-κB. Other studies show different mechanisms for
tolerance of the indigenous bacteria (25, 26). Some of these are described by
Cario et al. who reviewed tolerance of intestinal immunity through toll-like
receptors (TLRs) (27). Consequently, the host is not simply blind to the presence
of indigenous bacteria, but employs a two-way communication that results in
tolerance.

Microbiota on other locations besides the gut
As discussed above, gut microbiota is key in proper development of adaptive and
innate responses against pathogens and irritants. Therefore, it is not surprising
5

that the microbiota would also have far reaching effects in host organisms, and
not just on the intestines in which they are located.
It is known that signals from the brain have effects on the microbiota, but the
converse is also true, the microbiota has an effect on the brain’s function as well.
Vagus nerves, for instance, can trigger vomiting during food poisoning (28).
Studies with germfree mice showed that these animals had an increase in levels
of adrenocorticotrophic hormone (ACTH) and corticosterone, during a particular
stress experiments, compared to specific pathogen free controls. This study
indicates that the gut microbiota may have a role in suppression of ACTH and
corticosterone production (29). A study was also conducted investigating whether
there was a connection between the microbiota and pain. The authors reasoned
that the perception of pain was necessary for organisms to adapt to their
environment and external stresses. They conducted experiments using germfree
mice and found that a microbiota was indeed necessary for proper development
of the response to inflammatory pain (30).
The gut microbiota was also shown to play a role in liver functioning, as well as
with deconjugation of bile acids (by some Lactobacilli), and insulin resistance (31).
Dumas et al. conducted a study on BALB/c mice to study the metabolic profile
associated with insulin resistance and found that metabolites indicative of this
condition was related to host diet (31). They saw that the metabolic phenotype
associated with disease could be correlated to amounts of fat in the diet. By
extension, since the microbiota plays a large role in lipid available to the host, the

6

gut’s microbial community plays a role in altering host metabalomics and
occurrence of certain disease (32).
To underscore the far reaching effects of the gut microbiota, one recent paper
also reported effects of this community on lens of the eyes (33). Lipid content of
lens and retina of eyes, from mice with a gut microbiota, was measured together
with those from germfree animals. They found that conventionally raised mice had
less phosphatidylcholines in their lens indicating exposure to oxidative stress. It is
thus clear that the intestinal microbial community affects more than digestion.

Analysis of microbial communities
Earlier, the importance of community diversity and stability on the maintenance of
health was briefly mentioned. Before discussing the nature and impact of specific
changes in the microbial community, I will first summarize techniques used for
measuring these changes.
Tools and techniques
For a long time microbiologists studied organisms by growing them in culture.
However, more than 80% of intestinal bacteria cannot be cultivated using current
techniques, as described by “the great plate count anomaly” which addresses
differences between the greater number of microbes seen with a microscope than
those actually grown on culture plates (34).
When Carl Woese introduced the 16S rRNA-encoding gene as a target to study
microbial phylogeny, new doors opened for investigating microbial diversity (35).
7

Ashelford et al. have described features of the 16S rRNA gene that makes it
appropriate and useful for phylogenetic classification (36). This gene is highly
conserved given its function in protein synthesis; however, nine variable regions
interrupt this conserved gene. The first characteristic allows us to design broadrange primers that bind conserved regions so that amplicons can be retrieved that
include many groups of microbes. The second feature allows these amplicons to
be distinguished from each other since aligned sequence of these variable
regions will differ from gene to gene, and thus bacterium to bacterium. Many
molecular techniques used for studying diversity within the microbial community
depend on these features of the 16S rRNA-encoding gene. Some of these will be
described below.
Methods used to study the microbiota include denaturing gradient gel
electrophoresis (DGGE)/ temperature gradient gel electrophoresis (TGGE),
terminal restriction fragment length polymorphism (T-RFLP), fluorescent in-situ
hybridization (FISH), clone libraries, and more recently, high throughput
pyrosequencing.
In DGGE, the 16S rRNA genes from a sample are amplified through PCR and
16S rRNA gene amplicons are run across a polyacrylamide gel that has a vertical
gradient of denaturing substance. As the double stranded amplicons travel down
the gel, they pass through changing concentrations of this denaturing substance
and the double stranded helices dissociate at various points. Since each helix is
composed of DNA that spans a hypervariable region, these strands would have
different base pairs and hence separate at different points along the gradient. The
8

resulting gel would have a ladder-like appearance. If amplicons from two different
communities, of differing community structure, were run side-by-side, it would be
expected that patterns of “rungs” of the ladders would be different. Further
analyses are normally done to identify microbes according to patterns on the gel.
For example, hybridization analysis with species-specific probes can be used to
identify specific microbes (37). Bands from the gel can also be excised and
sequenced in order to ascertain microbial identities. An alternative approach
involves using a temperature gradient (TGGE) instead of a denaturing gradient, to
allow double stranded DNA helices to denature along the gel (38).
T-RFLP is another fingerprinting technique. Here PCR-amplified 16S rRNA genes
of a bacterial community are digested with restriction enzymes and cleaned
product is visualized using spectroscopy. This visualization is possible because
terminal fragments are fluorescent as one of the primers, either forward or
reverse, has an attached fluorophore (e.g. 8-carboxy-fluorescein label) (39).
Using T-RFLP, relative abundance of different operational taxonomic units
(OTUs) can be seen, as intensity of light emissions is proportional to abundance
of each OTU, and different OTUs are designated by peaks generated at cut sites
(39). This makes the use of traditional macro-ecological indices (e.g. Shannon or
Simpson Indices), employed to measure richness and evenness, easy to apply.
As mentioned briefly above, sequence-specific probes can be used to investigate
microbial identities using hybridization techniques. FISH is commonly used for a
quantitative measure of particular bacteria in an environmental community, such
as the gut, in situ (40, 41). In this technique, a fluorescently labeled
9

oligonucleotide is used to bind to genes of interest (frequently the 16S rRNA
gene) of microbes in a community. This oligonucleotide sequence would have a
certain degree of specificity according to its design, that is, compliment a variable
region for species-level detection or a more conserved region for higher
phylogenetic resolution such as phylum-level detection (40-43). The labeled
microbes can then be visualized by microscopy or flow cytometry.
Another method used to study the microbial communities is construction of clone
libraries. Clone libraries are typically constructed from full-length 16S rRNA gene
amplicons using primers that bind to conserved regions (10). The amplicons are
then ligated into vectors and transformed into a host cell, typically Escherichia
coli. These cells, carrying individual 16S rRNA gene fragments are cultured to
retrieve sufficient copies of the gene that are then sequenced. These sequences
are then aligned, and assigned identities by comparing to known bacterial
sequences that are already archived. This technique allows researchers to study
communities by comparing who is there and in what relative abundance. Although
this is a very powerful technique for analyzing a community, it is limited in that
there is relatively little sampling effort exerted (44, 45). Here, pyrosequencing has
bridged the gap with deep sampling, with hundreds of thousands of sequence
reads per run compared to only about a hundred sequences with clone libraries
(46, 47).
In pyrosequencing, a hypervariable section, about 300 base pairs, of the 16S
rRNA gene is amplified from a community. Adaptors are ligated on either end of
PCR fragments, which aid in community separation in future analyses, and
10

facilitate binding to beads. Beads are captured in droplets that house PCR
reactions so that there are millions of DNA strands associated with each bead.
Droplets are then broken and the beads placed into a microtitre plate, along with
enzymes. Sequencing occurs by measuring emissions generated when
deoxynucleotides wash over the plate (47, 48). This techniques generate massive
amounts of data and allows investigators the opportunity to observed the “rare
biosphere” (49).
These methods described above have all been invaluable in studying gut
communities. DGGE was first reported in a 1993 publication where it was used to
investigate microbial communities from various locations from microbial mats and
waste water samples (37). These authors used PCR to amplify the third variable
(V3) region of the 16S rRNA gene and used this amplicon to conduct the DGGE.
Zoetendal et al. also used this technique to study the microbiota of the gut by
comparing community composition of feces and tissues from different locations
along the colon (50). Likewise, through T-RFLP, a broad overview of the
community was observed when Keuhl et al. investigated whether there were
changes in community richness and evenness after infection with Helicobacter
hepaticus (39). The authors noted that presence of H. hepaticus had minimal
effects on other members of the community by comparing changes in resulting
traces. Although these techniques present information on general differences
between communities, they do not provide identities of members in the
community. Supporting techniques are used for this purpose, such as sequencespecific probe hybridization in DGGE that would bind to DNA of certain members
11

of the microbiota, or in silico digests of known bacteria to generate predictive
traces to compare with T-RFLP data.
FISH has the advantage of showing identities of certain groups in a particular
location. Franks et al. studied changes in abundance of certain bacteria in fecal
microbiota of nine participants over eight months (40). This hybridization
technique provided rapid results that were quantitative and allowed investigators
to track certain species of interest, Bacteriodes fragilis and Bacteriodes distasonis
among others. One constraint however, is since bacteria are measured using
emissions from fluorophores attached to probes, there is a limit to the number of
probes that can be used. This issue can be overcome using sequencing methods.
These methods allow generation of data that have components of relative
abundance, richness and identity of community members. With pyrosequencing,
communities can be sampled to such an extent that previously rare members are
observed. Dethlefsen et al. used this method to study changes in the human
microbiota induced by antibiotic treatment (49). They identified more than 5000
taxa associated with the human microbiota, and were then able to further
analyzed communities by focusing on specific groups, including the Bacteriodes,
Clostridiales and Lachnospiraceae. At the moment, this method is more
expensive than others previously mentioned, however, the volume of data
generated makes this technology very appealing.

12

Parameters in measuring microbial diversity
To begin to understand microbial communities, data need to be sorted and
interpreted. There are two important parameters that are measured in studying
diversity of intestinal microbial communities: richness and evenness. Richness
involves enumerating different bacteria that are present and is represented by
numbers of operational taxonomic units (OTUs) in communities; therefore,
defining what makes each OTU different is necessary. In T-RFLP, an OTU is
defined by each peak in the trace generated; or in clone libraries, the designation
can be applied after 16S rRNA gene sequences are aligned and compared.
Sequences that are 97% similar to each other are generally considered to be of
the same species (6). This is still subjective since some authors prefer to use
more stringent criteria for defining “species” such as 99% similarity (5). Once cutoff levels are defined though, analysis of communities is quite powerful and
represents an accurate and easy way of measuring diversity. Evenness refers to
relative abundance of groups within communities. This can be measured using
quantitative methods such as clone libraries, pyrosequencing, real-time PCR,
FISH, and microscopy. Fingerprinting methods, such as T-RFLP, also gives data
on relative abundance by noting relative fluorescent units associated with each
peak. This method is relatively cheap, and useful in retrieving a broad overview of
changes in a community. It is becoming clear that relative abundances of certain
groups, and not merely their presence or absence, is key in onset of disease (10,
51). Two previous parameters, richness and evenness, in part, describe structure
of the community. Another feature of microbial communities is that of function,

13

and depends on community composition. Composition refers to identities of the
various bacteria present and can be measured using sequencing methods.
Denoting membership allows investigators to potentially assign function and
thereby begin to understand mechanisms of community dynamics.

Resisting and rebounding from change: Stability of the microbiota
We have described how to measure community structure and composition. This
structure is subject to shifts that may be due to a number of factors. Ecologists
have been considering changes in terrestrial and aquatic communities for
decades and have generated concepts to explain and predict changes in these
communities. Microbiologists have now applied these concepts in studying the
microbial world, be it in soil communities or gut microbiota (3, 10). Intestinal
microbial communities experience many perturbations, including infection by
pathogens or treatment with antibiotics. In studying communities’ responses to
these disruptive forces, one must consider resistance and resilience, two key
ecologic concepts developed for studying communities of macroorganisms.
Resistance is the ability of a community to maintain its structure under stress,
while resilience is its ability to return to its pre-stress state after brief change in
structure (52). These two terms are associated with stability of a community, a
condition thought to be necessary for health since alterations in diversity of gut
microbial communities have been believed to play a role in some diseases such
as inflammatory bowel disease (IBD) and antibiotic associated diarrhea (10, 53).

14

The question of whether diversity begets stability has been ongoing for at least
the past 50 yrs (54). Ecologists first believed that more diverse communities are
more stable. This was supported by work on terrestrial communities from Odum
(55), Elton (56) and Tilman (57) who thought that simple communities were
susceptible to perturbations that left these communities open to invasion.
MacArthur had similar views after studying predator-prey interactions (58). More
recently, however, this view was challenged. Robert May used mathematical
models to show that diversity does not lead to stability in a community (59). In
fact, he showed that increased diversity destabilized the community. This issue
was further complicated when looking at food web relationships where strength of
associations also played a major role in the stability of community structure. In
these cases, it was seen that observations noted by Odum, Elton and MacArthur
did not apply since there was more than one trophic level involved (55, 56, 58,
59).
Microcosm experiments show that there is a positive correlation between diversity
and stability. This is seen at community-level (different groups of bacteria) and not
population-level (within a particular group, be it at phylum or genus-level). At the
population-level, it was seen that increasing diversity did not affect community
stability. Findings from these experiments led scientist to note that two
explanations were possible in explaining community stability. First, increasing
diversity increased the chances of variable species response to a perturbation
and hence some species can help maintain the normal functioning of the
ecosystem. Secondly, increasing diversity increased chances of functional
15

redundancy so if certain species are lost during a disturbance, communities still
function as all the tasks are still occurring (60, 61). These two explanations give
rise to the “insurance hypothesis.” It is also noteworthy that diversity is not the
“driver” of stability, but merely shows a positive correlation with it. Mechanisms by
which this occur are still under investigation. With this hypothesis, it is thought that
in the event of a perturbation, a community either shows resistance, in that it
generally does not show great change, or it is resilient, and bounces back after an
initial change because hardier members of the community compensate for
decrease of susceptible ones until they recover (61). The complexity of this issue
is emphasized by contradictions observed as one study showed that a less
diverse community was more resistant and resilient than highly diverse one (62).
Authors

explained

that

relationships

between

diversity

and

stability

in

communities depend on pre-stress states of those communities and mechanisms
that drive them.
In considering the gut microbial community, the same ideas apply. One would
expect that diversity correlates with stability at the community level. As seen with
some of the antibiotic work, communities can be resilient (10, 46). And as
suggested by the insurance hypothesis, this resilience may be due to functional
redundancy in communities. Functional redundancy is known to be present in our
gut microbiota, for instance several members are able to produce butyrate (12,
63).

16

Influences on the microbiota
There are many factors that sculpt microbial community structure in the intestine
and, on occasion, cause disruption. From birth, when colonization occurs, the
microbiota is shaped according to its anatomical location as conditions (pH,
oxygen concentration) vary, diet changes, pathogens infect and antimicrobials
are administered. It is not surprising then that there is great inter-individual
variation. However, there are also great similarities due to evolutionary
commonalities among vertebrates that give rise to the idea of a “core”
microbiome (6, 64).
Microbiota changes with age
The fetal environment is thought to be sterile and colonization starts during
birthing. Colonization, thus, depends on both the method of birthing (caesarian
section versus natural child birth) (65), and whether or not the child is breast-fed
(66, 67). This early microbiota then evolves into a mature community that begins
to stabilize soon after weaning. This community is then influenced by diet,
antibiotics and other environmental factors that contribute to inter-individual
variation, and persists into old-age.
Palmer et al. investigated development of the intestinal microbiota by following
infants through their first year of life (68). They assessed colonization through
stool sampling and comparisons with stool samples from other members of the
family. Mothers’ milk and vaginal swabs were also investigated in an attempt to
map route of colonization. The authors found that the gut microbiota in the first
days to months of an infant’s life is variable and is related to environmental
17

exposure, i.e. vaginal route, breast-feeding. The microbiota of these 12 infants,
however, converged toward the end of their first year. This community resembled
that of an adult, consisting of an abundance of Firmicutes, Bacteriodetes, with
some Verrucomicrobiae and Proteobacteria. Contrary to many publications (69,
70) though, there was a low prevalence of Bifidobacteria. The authors suggest
that previous papers may have exaggerated the importance of this group due to
its probiotic effects, and lost focus on other more prevalent groups (68).
Once the community reaches this stabilized adult structure, fluctuations from lifestyle, e.g., diet and disease, are its main effectors. This was demonstrated in
another recent investigation involving more than 35 000 adults (71). It was found
that bacterial communities, in stool samples, were generally the same over age
and gender. A few individual groups such as Enterococci, increased with age
while Bacteriodes decreased. Bifodobacteria and Lactobacillus appeared to be
stable. These data were retrieved by culture so do not represent species that may
be difficult to grow via methods used. Other limitations included the lack of
knowledge of antibiotics taken or diet of participants. Diet is known to play a role
in the structure of gut microbiota and these authors suggest that change in
bacterial communities with age could be due to changing diets and not the actual
process of aging (72, 73).

18

Microbiota changes with location
Studies often use fecal samples in analyses (68, 73). Although, stool is
convenient and non-invasive, it may not be representative of the gut microbiota
since microbial communities are not homogenous along the gastrointestinal tract,
or across the lumen. Gut microbiota is divided into two groups, autochthonous
(commonly referred to resident communities that are closely associated to the
host) and then allochthonous (transient bacteria that do not get permanently
associated with host and may include pathogens) (74). In studies investigating
resident microbiota, the autochthonous community can be investigated using
biopsy samples, which capture mucosal microbes that interact intimately with a
host. It was also found that certain resident bacteria associate with hosts at
particular locations along the intestine, at varying abundances (75).
The alimentary canal, although continuous, harbors various environments from
mouth to anus (51, 75). Focus will be placed on the intestinal section since much
emphasis has been placed on this community for its role in host development and
maintenance of health (1, 12, 76). Differences in anatomy between intestinal
sections are obvious with the cecum being a blind end, while the colon is tubular.
Other differences are detailed in a review by Snipes (77). There are also
differences in the immune elements among different regions of the intestines (78),
such as Paneth cells being present in small intestines and not colon. These
characteristics may contribute to the variation seen along different sections of the
intestinal tract. Wang et al. (75), through molecular techniques, saw that human
jejunum had the lowest microbial richness compared to distal ileum, ascending
19

colon and rectum. By comparing sequences against those of known bacteria in
the database of the National Center for Biotechnology Information (NCBI), they
also found that the main difference was that the jejunum had the least
Bacteriodetes but the most Streptococcus (68%) and Proteobacteria compared to
other

three

regions.

Another

group

suggests

that

acid-tolerance

by

Streptocococcus may select for this bacterium in the upper intestine since this
environment has a low pH (79). They also showed a complete lack of detection of
Clostridium XIVa group in the jejunum while other parts had between 20 and 33%
prevalence. The ileum and rectum both had a high prevalence of Bacteriodetes
(49 and 44%, respectively). Previous work by this group showed a gradient of
increasing load and richness from terminal ileum to colon in a single individual
(80). However, Zoetendal et al. (50), using DGGE, noted that bacteria along colon
mucosa of 10 individuals were evenly distributed. Wang et al. (75) also showed
similar distribution except in the case of Bacteriodetes. Since Wang et al. (75)
studied a single subject while Zoetendal et al. (50) had 10 participants, this
difference could be due to variation noted among individuals, as well as the
sampling methods employed.
Certain microbes may also be associated with different regions along the crosssection of the lumen since different microenvironments are present; for instance,
anoxic conditions exist at the center of the lumen particularly within stool (81),
whereas microaerophilic conditions are present near the vasularized epithelium
(82).

20

Communities retrieved also depend on samples and sampling techniques since
there are differences in dominant bacteria in feces and those attached to mucosa
as observed through TGGE, for instance (8). However, fecal and biopsy
communities look more similar to each other in one individual than two fecal
samples from two different individuals due to inter human variability (5, 83). We
can even go further in describing variation among mammals where it was seen
that grouping not only depends on genetics but also diet (64).

Microbiota changes with diet
Diet has been known to influence gut bacteria since it selects for certain groups
that can carry out certain functions, such as Bacteroides to break down protein or
Clostridia to act on complex carbohydrates in fiber degradation. Therefore, one
would expect a predominance of Clostridia in an individual consuming a
vegetarian diet. This was supported in a study done by Hayashi et al. (72) where
gut microbiota of a female vegetarian was investigated and compared to those of
three healthy men. It was found that the vegetarian had a high proportion of
Clostridium XIVa and Clostridium IV, as well as Clostridium ramosum from cluster
XVIII. This latter bacterium may be important in individuals on a vegetarian diet
since Finegold et al. also reported that this microbe was present in 52% of
vegetarians sampled in their study (84). Actual function, however, is still under
investigation. Researchers have also found that Bacteriodes were higher in
individuals with a greater meat intake. In addition, Bacteriodes convert bile into
products that are known carcinogens and thus a high proportion of this group has
21

been linked to colon cancer (85). This has been the subject of debate as a
vegetarian diet is proposed to reduce chances of developing colon cancer,
emphasizing a link between diet, microbiota and disease. Thus, diet is not only
associated with shaping gut communities, but also selecting microbes that are
associated with development of disease.

Microbiota changes with a pathogen
Pathogens are organisms that cause disease. Many, such as Clostridium difficile
and Salmonella enterica, are linked to intestinal disease in humans, and are
associated with shifts in resident microbial communities (7, 51). Studies on these
organisms show that infection normally occurs after shifts in community structure,
due to antibiotic treatment, and implies deterioration in colonization resistance.
Salmonella enterica Serovar Typhymurium infection will induce a typhoid-like
illness in mice, i.e. mice develop a disease resembling human typhoid fever, with
temperatures as high as 104°F and non-bloody diarrhea. As a human pathogen,
this microbe induces enteritis. FvB mice are susceptible to Salmonella infection
without antibiotic pretreatment. This provided a model to observe alterations of
gut microbiota in the presence of a pathogen. Upon infecting mice with
Salmonella, it was found that the diversity of the resident microbial community
changes (86). Salmonella colonized cecum and colon better than distal small
intestines, as shown by qPCR data. In the colon, numbers actually decreased
after 1 week compared to the 3 days post infection. This colonization coincided

22

with changes in load of other bacterial groups. Bacteriodetes decreased in the
cecum and colon. Lactobacillus/Enterococcus decreased significantly in the distal
small intestine after 3 days Salmonella infection; and the Eubacterium
rectale/Clostridium coccoides group decreased in all sites measured, with a
corresponding increase in Clostridium perfinges. However, these changes could
be due to host responses to infection. Actual onset of diarrhea can only be
associated with the 1 week post infection group since no diarrhea was seen 3
days post infection. Once Salmonella was cleared, the microbiota rebounded to it
previous set point. The authors observed only certain groups of bacteria to be
indicative of changes in the microbiota. It should be noted that there may be other
changes seen in groups that are not represented by the primer sets used. A more
wholistic approach, using “universal” primers, would be more informative in
probing community changes more thoroughly (10, 46). However, the study shows
subsequent shifts in the microbiota due to presence of a pathogen. Although this
change in the microbiota may not be directly caused by Salmonella invasion, it is
definitely associated with onset of diarrhea. Similarly, a study by Chang et al.
showed that patients recovering from C. difficile infections had gut microbial
communities more similar to healthy participants, than to those with active
disease (7), indicating a return of a “healthy microbiota” in recovering patients. C.
difficile infections are normally treated with an antibiotic regime, which does not
only target this pathogen, but also members of the resident community. This
introduces another agent of perturbation to gut microbial communities.

23

Microbiota changes with antibiotics
It has been shown that antibiotics can alter intestinal murine microbial
communities (51, 87). In so doing, colonization resistance is lowered and infection
by a pathogen such as Salmonella is very efficient (51). Antibiotics have long
been recognized to significantly altering the gut community (88).

Van der Waaij

et al. touted the use of antibiotics in “decontaminating” the digestive tract since
they saw dramatic reduction in aerobic and anaerobic bacteria after treatment
with various drugs (88). This survey was done by culture methods thus many
bacteria would have been undetectable (34). Never the less, antibiotics can
greatly impact gut communities. Antonopoulos et al. (46) showed that degree of
perturbation depended on specific treatments used. Researchers, through
pyrosequencing technology, showed that cefoperazone induced long-term
reduction in community diversity (up to six weeks after cessation of treatment).
Another drug combination, amoxicillin/metronidazole/bismuth (AMB), only showed
short-term reduction in community diversity with richness being restored after only
2 weeks after antibiotic treatment stopped. This exemplifies elasticity of
community diversity under certain conditions. Previous work from this group also
showed remarkable resilience of some members of the gut microbiota after a
perturbation caused by an antibiotic regime (10). Although the antibiotic
administered was for treatment of sinusitis, the gut microbiota of the patient in this
study showed shifts among different microbial groups. This change in diversity of
the gut community was associated with diarrhea. On completion of treatment,
diversity of the microbiota resembled that of the community before antibiotics.

24

Bowel movements also returned to normal. This suggests that certain microbial
diversity is associated with health, and that disturbing it can lead to disease.
Alternatively, health may be attributed to overall community structure with a predetermined balance perpetuating a healthy state.

The microbiota and disease
The microbiota contributes tremendously to health and development of its host,
as reviewed above. We have also seen that this community is very resilient. But
what happens when the microbiota is altered and cannot compensate for
associated changes? In some cases, this community disturbance has been
associated with onset of disease.
Inflammatory bowel disease
Inflammatory bowel disease (IBD) includes both Crohn’s disease (CD) and
ulcerative colitis (UC). The former is characterized by lesions, which may occur
anywhere along the gastrointestinal tract. These lesions can penetrate through
layers of the GI tract, and may even corrode through the gut wall and create
fissures into other organs such as the bladder. The latter is normally confined to
the lower intestines and rectum, and lesions are more superficial, affecting the
epithelial layer and lamina propria (89).
Since no specific bacteria can be linked to the development of IBD in humans,
Koch’s postulates cannot be fulfilled, leading to the current view that this disease

25

is a result of dysbiosis (changes in the microbial community structure). Koch’s
postulates are as follows:
1. The microorganism must be found in abundance in all organisms suffering
from the disease, but should not be found in healthy animals.
2. The microorganism must be isolated from a diseased organism and grown
in pure culture.
3. The cultured microorganism should cause disease when introduced into a
healthy organism.
4. The microorganism must be re-isolated from the inoculated, diseased
experimental host and identified as being identical to the original specific
causative agent.
Koch’s postulates are especially challenging to fulfill in the context of intestinal
disease since many microbes are difficult to culture by traditional methods (32),
therefore may not be easily isolated. Even if they are isolated, interactions with
other microbes or host may be required for growth, preventing pure culture.
However, many studies have been cited that link the microbiota with IBD (90, 91).
IL-10-/- mice have been convenient in showing roles of microbes in onset of IBD.
Since this anti-inflammatory cytokine is absent in these mice, inflammation
perpetuates without the inhibitory force of immune regulation that wild type
animals possess. Disease is then mediated from a Th1 type response. It was
found that wild type conventionally raised or germfree mice did not develop colitis,
neither did germfree IL-10-/- mice (92). However, IL-10-/- conventionally raised
mice developed disease leading authors to suggest that a resident intestinal
26

microbiota was necessary for development of colitis in these immune altered mice
(92). Also, it was found that disease was only present after mice were gavaged
with fecal slurry and not sterile filtrate of fecal slurry or sterile bacterial lysates,
indicating that microbes are responsible for immune stimulation (93).
Alternatively, it was postulated that one mechanism by which resident bacteria
cause colitis is by compromising the colonic epithelial barrier (94). Ohkusa et al.
showed that different indigenous bacteria were capable of inducing varying levels
of cytokine production from the host intestinal tissue (94). Activity of these
inflammatory molecules suggests that resident microbes could induce and
maintain inflammation and perhaps, did not cause disease in vivo since host
mediators kept them at bay. The host thus appears to control intestinal bacteria
from invading, emphasizing the tug-of-war that occurs in this dynamic system.
Nonetheless, resident microbes have an important role in onset of IBD.

Animal models of IBD
IBD is of unknown etiology. There are several animal models available that mimic
appropriate immunological and histopathological responses. Although these
models do not encompass all the complex parameters involved in human
disease, they have been used to elucidate pathophysiology of disease, and have
also been useful in determining possible therapies. There are two major types of
animal models used to study IBD: chemically induced (dextran sodium sulfate,
trinitro-benzene sulphonic acid) or immuno-deficient (T cell adoptive transfers, IL10-/-). In the following, the dextran sodium sulfate (DSS)- induced colitis model,
27

as well as the Helicobacter hepaticus infected IL-10-/-

mouse model are

discussed. These are frequently used models employed by our laboratory in
studying IBD related changes in the microbiota (39, 95).
Dextran sodium sulfate (DSS) is commonly used as a means to induce colitis
reliably (96). This chemical is heparin-like in nature and too large to transit
intestinal walls (97). This chemical can induce two types of disease according to
administration, acute or chronic. Acute disease occurs after one time dosage at
about 3.5% DSS, while chronic colitis occurs at a lower dosage that is repeated at
intervals (98). Mechanisms by which it causes diseases is still under investigation
(99).
There has been vast work done on effects of DSS on host tissues and immune
response as it relates to ulcerative colitis (UC) (100, 101). Using this model,
Medzhitov and colleagues have shown that TLR 4 knockout mice did not develop
colitic lesions on treatment with this chemical (102). MyD88 knockout mice also
fail to develop characteristic lesions of UC (25). More recently, Ungaro et al. found
that TLR 4 antagonist antibody ameliorates inflammation in colitic mice (103).
These studies suggest that intestinal bacteria can play a role in development of
UC via the MyD88 pathway.
Another model of IBD linked with the microbiota is infection of IL-10-/- mice by
Helicobacter hepaticus (104). There have been several studies showing
development of colitis in IL-10-/- mice infected with this microbe (95, 105, 106).
H. hepaticus induced colitis is a T cell response (104). Recent evidence has also
28

linked Th 17 that produces IL-23 and INFγ in development and modulation of this
infection (107). The mechanism by which this bacterium causes colitis is
unknown; however, it has been found that H. hepaticus has a pathogenicity island
(PAI) that is involved in induction of disease (106). This PAI was involved in
significantly increasing disease of the cecum, as seen through histology and
proinflammatory mediators, of B6.129-IL10tm1Cgn (IL10-/-) mice compared to
those infected with an isogenic mutant (106). The PAI region investigated showed
immuno-regulatory effects by showing lower IgG2c and IgG1 responses than wild
type, but did not affect colonization ability. Another paper from this group also
showed that the B component of cytolethal distending toxin (CdtB) was important
in intestinal colonization. Again, there were also lower levels of IgG2c and no
IgG1 response indicating that CdtB was involved in immuno-regulation (105).
Therefore, for this bacterium, PAI and CdtB are important in establishing
inflammation and causing colitis.
It is important to note that H. hepaticus induced colitis only occurs in mice
containing a conventional microbiota. Dieleman et al. has shown that disease only
ensues in the presence of a resident microbiota and not in H. hepaticus
monoassociated IL-10 deficient mice, on a mixed (C57BL/6 x 129/Ola) or inbred
(129/SvEv) genetic background (108). It should be noted that germfree animals
are not “normal” in that they do not have a developed immune system and would
not react as conventional animals. Also, disease does not occur in mice with an
intact immune system since several studies with this bacterium in wild type
animals have not resulted in inflammation (39). As previously discussed, a
29

pathogenicity island facilitates onset of disease, and CdtB is needed for
establishment (95, 105). In this model, disease then seems to depend on several
factors: resident community, host and H. hepaticus itself.

Probiotics
Non-resident bacteria can be used to enhance or rescue deficient functions of the
gut microbiota so that disease is prevented or ameliorated (109, 110). Probiotics
is defined as “living microorganisms which when administered in adequate
amount confer a health benefit on the host (17).” These benefits can be conferred
by several different ways. Probiotics can colonize the gut and occupy niches that
would have been available to potential pathogens as reviewed (111). This is
competitive exclusion. Probiotics can achieve this by using up nutritional
resources, or altering the environment (pH, mucus production) to make conditions
undesirable for other microbes (112). Beneficial bacteria can also alter gene
expression of potential pathogens so that virulence decreases (113).

These

microbes work synergistically with the intestinal community to maintain health.
Many of the genera used, such as Lactobacillus, are predominant members of the
intestinal community, and functionally important. Therefore, these bacteria
colonize and persist in the host long enough to convey benefits. The mechanism
of action of many probiotics is enhanced abilities to implement functions that are
done by the resident microbiota. Probiotics can act through modulation of the host
responses. They may induce the production of defensins or produce proteases

30

that

cleave

and

activate

this

antimicrobial

peptide

(114).

Bacteroides

thetaiotaomicron, a resident microbe in the human gut, also has this effect (11).
Probiotics can also activate expression of immuno-regulatory genes. Lactobacillus
rhamnosus GG activate NF-κB in macrophages (115). They activate the immune
response and allow hosts to clear potential pathogens. Conversely, another
species, Lactobacillus plantarum, act by blocking signaling to this gene and
suppresses inflammation (116). It was found that solution from L. plantarum
culture inhibited the NF-κB expression of murine intestinal epithelial cells. They
found that L. plantarum medium (without bacterium) inhibited NF-κB activation
from MyD88 dependent and independent pathways via TNF receptor.
Sometimes it is the bacterial metabolites, and not direct immunological effects of
the microbes themselves that have health benefits. Martin et al. studied effects of
two probiotics, Lactobacillus paracasei and L. rhamnosus in mice colonized with
human baby flora (HBF) (117). Presence of these bacteria altered metabolitic
profiles from blood and tissue of mice. L. paracasei changed energy metabolism
pathways of the host by allowing accumulation of bile acids (bacteria cannot
deconjugate this compound). This led to high lipid absorption in the intestine.
Efficient metabolism of soybean oil from mouse chow also allowed mice to reduce
plasma lipoprotein and liver cholesterol. Bacteria also changed metabolism of
amino acids via glucogenesis. Wostmann summarizes studies done on
contributions of the microbiota on biosynthetic pathways in the gut (13). Bacteria
change lipid composition of fecal content and vitamin and mineral uptake in the
gut. Also, complex carbohydrates are broken down into short chain fatty acids
31

with the aid of gut microbes. These functions indicate that it is a product of
bacterium and not the microbe itself that has beneficial effects exemplifying
various ways microorganisms affects the community.

Conclusion
Mammalian microbota is, without a doubt, a complex and dynamic ecosystem.
Not only do microbes interact with each other to exist harmoniously (among
resident microbes) or antagonistically (attack by parasites or predators such as
Bdellovibrio), but also communicate and cooperate with the host.
The gut microbiota is important in many aspects of host health. From discussions
above, it can be seen that it is necessary for the microbiota to be resistant and
resilient to disruptive forces in order for its function to be maintained. In the event
that diversity of the gut community is altered, diseases such as IBD may develop.
To study the relationship between the microbiota, host and disease, I intended to
use known models of IBD to induce disease in mice and investigate subsequent
changes in the intestinal microbiota. Studies have been done associating
changes in the microbiota with disease, but none investigating changes in the
DSS model in mice, or whether the relative abundance of H. hepaticus in the gut
of IL-10-/- mice impact on disease severity. I intended to address these questions
through experiments conducted in the upcoming chapters in order to attempt to
identify possible roles of the microbiota in influencing disease.

32

REFERENCES

33

REFERENCES
1.

Bäckhed F, Ley R, Sonnenburg J, et al. Host-bacterial mutualism in the
human intestine. Science. 2005;307:1915-1920

2.

Gill SR, Pop M, DeBoy RT, et al. Metagenomic Analysis of the Human
Distal Gut Microbiome. Science. 2006;312:1355-1359

3.

Dunbar J, Barns S, Ticknor L, et al. Empirical and theoretical bacterial
diversity in four Arizona soils. Appl Environ Microbiol. 2002;68:3035-3045

4.

Venter J, Remington K, Heidelberg J, et al. Environmental genome
shotgun sequencing of the Sargasso Sea. Science. 2004;304:66-74

5.

Eckburg PB, Bik EM, Bernstein CN, et al. Diversity of the Human Intestinal
Microbial Flora. Science. 2005;308:1635-1638

6.

Ley RE, Peterson DA, Gordon JI. Ecological and Evolutionary Forces
Shaping Microbial Diversity in the Human Intestine. Cell. 2006;24:837-848

7.

Chang JY, Antonopoulos DA, Kalra A, et al. Decreased Diversity of the
Fecal Microbiome in Recurrent Clostridium difficile-Associated Diarrhea. J
Infect Dis. 2008;197:435

8.

Lepage P, Seksik P, Sutren M, et al. Biodiversity of the mucosaassociated microbiota is stable along the distal digestive tract in healthy
individuals and patients with IBD. Inflamm Bowel Dis. 2005;11:473-480

9.

Dethlefsen L, McFall-Ngai M. & Relman, DA. An ecological and
evolutionary perspective on human–microbe mutualism and disease.
Nature. 2007;449:811-818

10.

Young VB, Schimdt TM. Antibiotic-Associated Diarrhea Accompanied by
Large-Scale Alterations in the Composition of the Fecal Microbiota. J Clin
Microbiol. 2004;42:1203-1206

34

11.

Stappenbeck T, Hooper L, Gordon J. Developmental regulation of
intestinal angiogenesis by indigenous microbes via Paneth cells. PNAS.
2002;99:15451-15455

12.

Cummings JH, Macfarlane GT. Role of intestinal bacteria in nutrient
metabolism. JPEN J Parenter Enteral Nutr. 1997;21:357-365

13.

Wostmann B. The germfree animal in nutritional studies. Ann Rev Nutr.
1981;1

14.

Wostmann B, Bruckner-Kardoss E. Development of cecal distention in
germ-free baby rats. Am J Physiol. 1959;197

15.

Gordon H, Bruckner-Kardoss E. Effect of the normal microbial flora on
various tissue elements of the small intestine. Acta Anat (Basel). 1961;44

16.

Levenson S. The influence of the indigenous microflora on mammalian
metabolism and nutrition. JPEN J Parenter Enteral Nutr 1978;2:78-107

17.

Ahmad MS, Krishnan S, Ramakrishna BS, et al. Butyrate and glucose
metabolism by colonocytes in experimental colitis in mice. Gut.
2000;46:493-499

18.

Hill R, Cowley H. The influence of colonizing micro-organisms on
development of crypt architecture in the neonatal mouse colon. Acta Anat
(Basel). 1990;137:137-140

19.

Ukena S, Singh A, Dringenberg U, et al. Probiotic Escherichia coli Nissle
1917 inhibits leaky gut by enhancing mucosal integrity. PLoS One.
2007;2:e1308

20.

Round J, Mazmanian S. The gut microbiota shapes intestinal immune
responses during health and disease. Nature Rev Immunol. 2009;9:313323

21.

Brandl K, Plitas G, Mihu C, et al. Vancomycin-resistant enterococci exploit
antibiotic-induced innate immune deficits. Nature. 2008;455:804-807

35

22.

Torruellas J, Jackson M, Pennock J, et al. The Yersinia pestis type III
secretion needle plays a role in the regulation of Yop secretion. Mol
Microbiol. 2005;57:1719-1733

23.

Stecher B, Hapfelmeier S, Müller C, et al. Flagella and chemotaxis are
required for efficient induction of Salmonella enterica serovar Typhimurium
colitis in streptomycin-pretreated mice. Infect Immun. 2004;72:4138-4150

24.

Kelly D, Campbell J, King T, et al. Commensal anaerobic gut bacteria
attenuate inflammation by regulating nuclear-cytoplasmic shuttling of
PPAR-gamma and RelA. Nature Immunol. 2004;5:104-112

25.

Rakoff-Nahoum S, Paglino J, Eslami-Varzaneh F, et al. Recognition of
Commensal Microflora by Toll-Like Receptors Is Required for Intestinal
Homeostasis. Cell. 2004;118:229-241

26.

Cario E, Gerken G, Podolsky DK. Toll-Like Receptor 2 Controls Mucosal
Inflammation by Regulating Epithelial Barrier Function. Gastroenterol.
2007;132:1359

27.

Cario E, Podolsky D. Intestinal epithelial TOLLerance
inTOLLerance of commensals. Mol Immunol. 2005;42:887-893

28.

Hu D, Zhu G, Mori F, et al. Staphylococcal enterotoxin induces emesis
through increasing serotonin release in intestine and it is downregulated
by cannabinoid receptor 1. Cell Microbiol. 2007;9:2267-2277

29.

Sudo N, Chida Y, Aiba Y, et al. Postnatal microbial colonization programs
the hypothalamic-pituitary-adrenal system for stress response in mice. J
Physiol. 2004;558:263-275

30.

Amaral F, Sachs D, Costa V, et al. Commensal microbiota is fundamental
for the development of inflammatory pain. PNAS. 2008;105:2193-2197

31.

Dumas M, Barton R, Toye A, et al. Metabolic profiling reveals a
contribution of gut microbiota to fatty liver phenotype in insulin-resistant
mice. PNAS. 2006;103:12511-12516

36

versus

32.

Ley RE, Backhed F, Turnbaugh P, et al. Obesity alters gut microbial
ecology. PNAS. 2005;102:11070-11075

33.

Orešič M, Seppänen-Laakso T, Yetukuri L, et al. Gut microbiota affects
lens and retinal lipid composition. Experimental Eye Research
2009;89:604-607

34.

Staley J, Konopka A. Measurement of in situ activities of
nonphotosynthetic microorganisms in aquatic and terrestrial habitats. Ann
Rev Microbiol. 2006;39:321-346

35.

Woese C, Kandler O, Wheelis M. Towards a natural system of organisms:
proposal for the domains Archaea, Bacteria, and Eucarya. PNAS.
1990;87:4576-4579

36.

Ashelford KE, Chuzhanova NA, Fry JC, et al. At Least 1 in 20 16S rRNA
Sequence Records Currently Held in Public Repositories Is Estimated To
Contain Substantial Anomalies. Appl Environ Microbiol. 2005;71:77247736

37.

Muyzer G, de Waal E, Uitterlinden A. Profiling of complex microbial
populations by denaturing gradient gel electrophoresis analysis of
polymerase chain reaction-amplified genes coding for 16S rRNA. Appl
Environ Microbiol. 1993;59:695-700

38.

Uenishi G, Fujita S, Ohashi G, et al. Molecular analyses of the intestinal
microbiota of chimpanzees in the wild and in captivity. Am J Primatol.
2007;69:367-376

39.

Keuhl CJ, Wood HD, Marsh TL, et al. Colonization of the Cecal Mucosa by
Helicobacter hepaticus Impacts the Diversity of the Indigenous Microbiota.
Infect Immun. 2005;73:6952-6961

40.

Franks AH, Harmsen HJM, Raangs GC, et al. Variations of Bacterial
Populations in Human Feces Measured by Fluorescent In Situ
Hybridization with Group-Specific 16S rRNA-Targeted Oligonucleotide
Probes. Appl Environ Microbiol. 1998;64:3336-3345

37

41.

Salzman NH, de Jong H, Paterson Y, et al. Analysis of 16S libraries of
mouse gastrointestinal microflora reveals a large new group of mouse
intestinal bacteria. Microbiol. 2002;148:3651-3660

42.

Ge Z, White D, Whary M, et al. Fluorogenic PCR-Based Quantitative
Detection of a Murine Pathogen, Helicobacter hepaticus. J Clin Microbiol.
2001 39:2598-2602

43.

Chan V, Crocetti G, Grehan M, et al. Visualization of Helicobacter species
within the murine cecal mucosa using specific fluorescence in situ
hybridization. Helicobacter. 2005;10:114-124

44.

Temperton B, Field D, Oliver A, et al. Bias in assessments of marine
microbial biodiversity in fosmid libraries as evaluated by pyrosequencing.
ISME J. 2009;3:792-796

45.

Neufeld J, Li J, Mohn W. Scratching the surface of the rare biosphere with
ribosomal sequence tag primers. FEMS Microbiol Lett. 2008;283:146-153

46.

Antonopoulos D, Huse S, Morrison H, et al. Reproducible community
dynamics of the gastrointestinal microbiota following antibiotic
perturbation. Infect Immun. 2009;77:2367-2375

47.

Margulies M, Egholm M, Altman W, et al. Genome sequencing in
microfabricated high-density picolitre reactors. Nature 2005;437:376-380

48.

Ronaghi M, Uhlén M, Nyrén P. A sequencing method based on real-time
pyrophosphate. Science. 1998;281:363-365

49.

Dethlefsen L, Huse S, Sogin M, et al. The pervasive effects of an antibiotic
on the human gut microbiota, as revealed by deep 16S rRNA sequencing.
PLoS Biol. 2008;6:e280

50.

Zoetendal E, von Wright A, Vilpponen-Salmela T, et al. Mucosaassociated bacteria in the human gastrointestinal tract are uniformly
distributed along the colon and differ from the community recovered from
feces. Appl Environ Microbiol. 2002;68:3401-3407

38

51.

Garner C, Antonopoulos D, Wagner B, et al. Perturbation of the small
intestine microbial ecology by streptomycin alters pathology in a
Salmonella enterica serovar typhimurium murine model of infection. Infect
Immun. 2009;77:2691-2702

52.

Allison S, Martiny J. Colloquium paper: resistance, resilience, and
redundancy in microbial communities. PNAS. 2008;105:11512-11519

53.

Swidsinski A, Ladhoff A, Pernthaler A, et al. Mucosal flora in inflammatory
bowel disease. Gastroenterol. 2002;122:44-54

54.

McCann. K. The diversity-stability debate. Nature. 2000;405:228-233

55.

Odum E. Fundamentals of ecology. 1953

56.

Elton C. Ecology of invasions by Animals and Plants. 1958

57.

Tilman D, Lehman C, Bristow C. Diversity-stability relationships: statistical
inevitability or ecological consequence? Am Nat. 1998;151:277-282

58.

MacArthur RH. Fluctuations of animal populations and measure of
community stability. Ecology. 1955;36:533-536

59.

May RM. Stability and complexity in model systems. 1973

60.

Naeem S. Species redundancy and ecosystem reliability. Conservation
Biology 1998;12:39-45

61.

Naeem S. Biodiversity:
2002;416:23-24

62.

Pfisterer A, Schmid B. Diversity-dependent production can decrease the
stability of ecosystem functioning. Nature. 2002;416:84-86

63.

Pryde SE, Duncan SH, Hold GL, et al. The microbiology of butyrate
formation in the human colon. FEMS Microbiol Lett. 2002;217:133-139

biodiversity

39

equals

instability?

Nature.

64.

Ley R, Lozupone C, Hamady M, et al. Worlds within worlds: evolution of
the vertebrate gut microbiota. Nat Rev Microbiol. 2008;6:776-788

65.

Biasucci G, Benenati B, Morelli L, et al. Cesarean delivery may affect the
early biodiversity of intestinal bacteria. J Nutr. 2008;138:1796S-1800S

66.

Nakamura N, Gaskins H, Collier C, et al. Molecular ecological analysis of
fecal bacterial populations from term infants fed formula supplemented
with selected blends of prebiotics. Appl Environ Microbiol. 2009;75:11211128

67.

Penders J, Thijs C, Vink C, et al. Factors influencing the composition of
the intestinal microbiota in early infancy. Pediatrics. 2006;118:511-521

68.

Palmer C, Bik E, DiGiulio D, et al. Development of the human infant
intestinal microbiota. PLoS Biol. 2007;5:e177

69.

Ouwehand A, Salminen S, Arvola T, et al. Microbiota composition of the
intestinal mucosa: association with fecal microbiota? Microbiol Immunol.
2004;48:497-500

70.

Sakata S, Tonooka T, Ishizeki S, et al. Culture-independent analysis of
fecal microbiota in infants, with special reference to Bifidobacterium
species. FEMS Microbiol Lett. 2005;243:417-423

71.

Enck P, Zimmermann K, Rusch K, et al. The effects of ageing on the
colonic bacterial microflora in adults. Z Gastroenterol. 2009;47:653-658

72.

Hayashi H, Sakamoto M, Benno Y. Fecal microbial diversity in a strict
vegetarian as determined by molecular analysis and cultivation. Microbiol
Immunol. 2002;46:819-831

73.

Mueller S, Saunier K, Hanisch C, et al. Differences in fecal microbiota in
different European study populations in relation to age, gender, and
country: a cross-sectional study. Appl Environ Microbiol. 2006;72:10271033

74.

Batt R, Rutgers H, Sancak A. Enteric bacteria: friend or foe? J Small Anim
Pract. 1996;37:261-267
40

75.

Wang M, Ahrne S, Jeppsson B, et al. Comparison of bacterial diversity
along the human intestinal tract by direct cloning and sequencing of 16S
rRNA genes. Volume 54: FEMS Microbiol Ecol. 2005:219-231

76.

Lundin A, Bok CM, Aronsson L, et al. Toll-like receptors and nuclear
receptors: a tripartite communication that tunes innate immunity in large
intestine. Cell Microbiol. 2008;10:1093-1103

77.

Snipes R. Anatomy of the cecum of the laboratory mouse and rat. Anat
Embryol. 1981;162:455-474

78.

Suzuki H. Differences in Intraepithelial Lymphocytes in the Proximal,
Middle, Distal Parts of Small Intestine, Cecum, and Colon of Mice.
Immunol Inves. 2009;38:780-796

79.

Takahashi N, Yamada T. Acid-induced acid tolerance and acidogenicity of
non-mutans streptococci. Oral Microbiol Immunol. 1999;14:43-48

80.

Wang X, Heazlewood S, Krause D, et al. Molecular characterization of the
microbial species that colonize human ileal and colonic mucosa by using
16S rDNA sequence analysis. J Appl Microbiol. 2003;95:508-520

81.

Swidsinski A, Loening-Baucke V, Vaneechoutte M, et al. Active Crohn's
disease and ulcerative colitis can be specifically diagnosed and monitored
based on the biostructure of the fecal flora. Inflamm Bowel Dis.
2008;14:147-161

82.

Marteyn B, West NP, Browning DF, et al. Modulation of Shigella virulence
in response to available oxygen in vivo. Nature 2010;465:355-358

83.

Suau A, Bonnet R, Sutren M, et al. Direct Analysis of Genes Encoding
16S rRNA from Complex Communities Reveals Many Novel Molecular
Species within the Human Gut. Appl Environ Microbiol. 1999;65:47994807

84.

Finegold S, Sutter V, Sugihara P, et al. Fecal microbial flora in Seventh
Day Adventist populations and control subjects. Am J Clin Nutri.
1977;30:1781-1792

41

85.

Kingston D, Van Tassell R, Wilkins T. The fecapentaenes, potent
mutagens from human feces. Chem Res Toxicol. 1990;3:391-400

86.

Barman M, Unold D, Shifley K, et al. Enteric salmonellosis disrupts the
microbial ecology of the murine gastrointestinal tract. Infect Immun.
2008;76:907-915

87.

Croswell A, Amir E, Teggatz P, et al. Prolonged impact of antibiotics on
intestinal microbial ecology and susceptibility to enteric Salmonella
infection. Infect Immun. 2009;77:2741-2753

88.

Van Der Waaij D, Strum C. Antibiotic Decontamination of the Digestive
Tract of Mice. Technical Procedures. Lab Anim Care. 1968;18:1-10

89.

Thoreson R, Cullen J. Pathophysiology of inflammatory bowel disease: an
overview. Surg Clin North Am. 2007;87:575-585

90.

Rath H, Schultz M, Freitag R, et al. Different Subsets of Enteric Bacteria
Induce and Perpetuate Experimental Colitis in Rats and MIce. Infect
Immun. 2001;69:2277-2285

91.

Rath HC, Ikeda JS, Linde HJ, et al. Varying cecal bacterial loads
influences colitis and gastritis in HLA-B27 transgenic rats. Gastroenterol.
1999;116:310-319

92.

Sellon R, Tonkonogy S, Schultz M, et al. Resident enteric bacteria are
necessary for development of spontaneous colitis and immune system
activation in interleukin-10-deficient mice. Infect Immun. 1998;66:52245231

93.

Sydora B, Martin S, Lupicki M, et al. Bacterial antigens alone can
influence intestinal barrier integrity, but live bacteria are required for
initiation of intestinal inflammation and injury. Inflamm Bowel Dis.
2006;12:429-436

94.

Ohkusa T, Yoshida T, Sato N, et al. Commensal bacteria can enter
colonic epithelial cells and induce proinflammatory cytokine secretion: a
possible pathogenic mechanism of ulcerative colitis. J Med Microbiol.
2009;58:535-545
42

95.

Young VB, Knox KA, Pratt JS, et al. In Vitro and In Vivo Characterization
of Helicobacter hepaticus Cytolethal Distending Toxin Mutants. Infec
Immun. 2004;72:2521-2527

96.

Okayasu I, Hatakeyama S, Yamada M, et al. A Novel Method in the
Induction of reliable Experimental Acute and Chronic Ulcerative Colitis in
Mice. Gastroenterol. 1990;98:694-702

97.

Kitajima S, Takuma S, Morimoto M. Changes in Colonic Mucosal
Permeability in Mouse Colitis Induced with Dextran Sulfate Sodium. Exper
Anim. 1999;48:137-143

98.

Wirtz S, Neufert C, Weigmann B, et al. Chemically induced mouse models
of intestinal inflammation. Nature Protoc. 2007;2:541-546

99.

Kitajima S, Morimoto M, Sagara E. A Model of Dextran Sodium Supfateinduced Mouse Colitis: Bacterial Degradation of DSS Does Not Occur
after Incubation with Mouse Cecal Contents. Exper Anim. 2002;51:203206

100.

Ito R, Kita M, Shin-Ya M, et al. Involvement of IL-17A in the pathogenesis
of DSS-induced colitis in mice. Biochem Biophys Res Commun.
2008;377:12-16

101.

Yan Y, Kolachala V, Dalmasso G, et al. Temporal and spatial analysis of
clinical and molecular parameters in dextran sodium sulfate induced
colitis. PLoS One 2009;4:e6073

102.

Rakoff-Nahoum S, Hao L, Medzhitov R. Role of Toll-like Receptors in
Spontaneous Commensal-Dependent Colitis. Immun. 2006;25:319-329

103.

Ungaro R, Fukata M, Hsu D, et al. A novel Toll-like receptor 4 antagonist
antibody ameliorates inflammation but impairs mucosal healing in murine
colitis. Am J Physiol, Gastroint Liver Physiol. 2009;296:G1167-79

104.

Cahill R, Foltz C, Fox J, et al. Inflammatory bowel disease: an immunitymediated condition triggered by bacterial infection with Helicobacter
hepaticus. Infect Immun. 1997;65:3126-3131

43

105.

Pratt JS, Sachen KL, Wood HD, et al. Modulation of host immune
responses by the Cytolethal Distending Toxin of Helicobacter hepaticus.
Infect Immun. 2006;74:4406-4503

106.

Ge Z, Sterzenbach T, Whary M, et al. Helicobacter hepaticus HHGI1 is a
pathogenicity island associated with typhlocolitis in B6.129-IL10 tm1Cgn
mice. Microbes Infect. 2008;10:726-733

107.

Kullberg M, Jankovic D, Feng C, et al. IL-23 plays a key role in
Helicobacter hepaticus-induced T cell-dependent colitis. J Exp Med.
2006;203:2485-2494

108.

Dieleman L, Arends A, Tonkonogy S, et al. Helicobacter hepaticus does
not induce or potentiate colitis in interleukin-10-deficient mice. Infect
Immun. 2000;68:5107-13

109.

Peran L, Camuesco D, Comalada M, et al. Lactobacillus fermentus, a
probiotic capable to release glutothione, prevents colonic inflammation in
the TNBS model of rat colitis. Int J Colorectal Dis. 2006;21:737-746

110.

Herias MV, Koninkx JF, Vos JG et al. Probiotic effects of Lactobacillus
casei on DSS-induced ulcerative colitis in mice. Int J Food Microbiol.
2005;103:143-155

111.

Callaway T, Edrington T, Anderson R, et al. Probiotics, prebiotics and
competitive exclusion for prophylaxis against bacterial disease. Anim
Health Res Rev. 2008;9:217-225

112.

Spurbeck R, Arvidson C. Inhibition of Neisseria gonorrhoeae epithelial cell
interactions by vaginal Lactobacillus species. Infect Immun. 2008;76:31243130

113.

Medellin-Peña M, Wang H, Johnson R, et al. Probiotics affect virulencerelated gene expression in Escherichia coli O157:H7. Appl Environ
Microbiol. 2007;73:4259-4267

114.

Wilson C, Ouellette A, Satchell D, et al. Regulation of intestinal alphadefensin activation by the metalloproteinase matrilysin in innate host
defense. Science. 1999;286:113-117
44

115.

Miettinen M, Lehtonen A, Julkunen I, et al. Lactobacilli and Streptococci
activate NF-kappa B and STAT signaling pathways in human
macrophages. J Immunol. 2000;164:3733-3740

116.

Petrof E, Claud E, Sun J, et al. Bacteria-free solution derived from
lactobacillus plantarum inhibits multiple NF-kappaB pathways and inhibits
proteasome function. Inflamm Bowel Dis. 2009;15:1537-1547

117.

Martin F-PJ, Dumas M-E, Wang Y, et al. A top-down systems biology view
of microbiome-mammalian metabolic interactions in a mouse model. Mol
Syst Biol. 2007;3

45

Chapter 2
Models of IBD

There are many models of inflammatory bowel disease (IBD) geared towards
understanding various aspects of this idiopathic disease. They address issues
that are difficult to investigate in humans, such as changes in early stages of
disease and effects of new therapies. Generally, these models are categorized
into three groups: chemical, immune dysregulation and environmental. However
these groups often overlap and some models involve more than just one factor,
for example in the case of immuno-compromised animals exposed to antigens
employ

both

immunological

and

environmental

triggers

(1,

2).

These

environmental triggers may include luminal stimulation from ingested material as
well as from the microbial community present in the gut. The modern view on the
role of luminal microbial communities extends from invasion of pathogens, to
alterations in resident community structure, also known as dysbiosis. However,
changes in the microbiota associated with disease in many of these models are
yet to be characterized. Moreover, roles of the microbiota at various stages of
disease, initiation and progression, as well as exacerbation or reduction of
inflammation, are still to be determined.
Here, we choose two well-known models of IBD, DSS-induced colitis and the
Helicobacter hepaticus triggered IL-10 deficient murine model, to investigate
46

various roles of the microbiota in IBD. The former causes disease in mice that
resembles human colitis, where inflammation is superficial, occurring in the
mucosa and submucosa, as well as being confined to the lower intestine.
Although there is some evidence that the microbiota plays a role in stimulating
immune dysregulation in this model, there has been limited evidence that
alterations in community structure occur after DSS administration, and no studies
characterizing changes that may occur over the course of disease progression.
The latter resembles Crohn’s disease in humans due to its Th1 immune
response and non-contiguous pattern of mucosal inflammation that occurs in
deep layers of the intestine. Importance of the role of the microbiota is inherent in
this model since a microbe invades the resident community and triggers disease.
However, the role of the microbiota in disease progression is unknown. Before
presenting experiments designed to address some of these questions at hand,
the following chapter aims to describe models that will be used in upcoming
studies.

Value of animal models
Animal models provide a means to test questions that may be difficult or
impossible to investigate in humans. An ideal model should be inexpensive, easy
to use and reproducible. There are typically several models designed to study a
single disease due to many aspects of physiology, immunology, microbiology
and genetics that interact to maintain a healthy biological system. A model is

47

chosen since it mimics, or is believed to mimic, a feature of the disease as is it is
occurs in humans.
Often times the condition studied in an animal model may not represent human
disease entirely, but rather only certain aspects are recapitulated, hence the use
of several models to answer questions about one disease. Each model is
typically used to answer specific questions and can be investigated at various
phase of disease where pathogenic components may be scrutinized.
Some conditions, such as IBD, are very complex and occur by interplay of
several mechanisms making prognosis and treatment challenging. As such,
several models have been developed to study disease. These systems have
proven to be very important in teasing apart various mechanisms by which
disease may be caused (3, 4). For example, models utilizing TLR, IL-10 and IL-2
deficient mice have been used to investigate various roles of the immune system
and genetic defects in disease (5-7). Furthermore, the role of the microbiota has
been studied using germfree and antibiotic treated animal models (5, 6); and
chemically induced disease models have been used to study loss of mucosal
tolerance and break-down of the epithelial barrier as potential mechanisms of
disease development (7, 8).

Two-widely used animal models in IBD
There are several types of animal models available to study various aspects of
IBD (3). They are generally divided into four groups: Those that are triggered by
48

dysfunction of innate immunity such as A20 deficient mice (9), by dysfunction of
adaptive immunity such as CD45RBHi transfer models (10), by the environment
such as H. hepaticus triggered disease in immuno-compromised mice and by
chemicals such as DSS and TNBD- induced disease (1, 8, 11). Here we focus on
two models investigated in following chapters: DSS-induced colitis and H.
hepaticus triggered disease in IL-10 deficient mice.

DSS-induced colitis
This murine model has been used since 1990 when Okayasu et al. modified it
from the original guinea pig system (8). Disease produced in this model is highly
dependent on mouse strain used as well as concentration of DSS administered
and duration of treatment (12-14). Variations of this model can also lead to acute
or chronic disease, and carcinomas (15).
DSS-induced colitis in mice can progress to dysplasia then carcinoma, similar to
the disease sequence in humans. This model of disease also responds to
therapies administered to humans such as steroids (16) and antibiotics (16), and
has an immunological response and genetic component (12). The model mimics
aspects of human disease, and its outcome is reproducible.

Origin of the model
In 1961, the bacteria Leuconostoc spp. and Streptococcus spp. were found to
produce a chemical polysaccharide called dextran from sucrose. Dextran sodium
49

sulfate (DSS) is produced by esterification of dextran with chlorosulfonic acid,
which produces a molecule with 17% sulfur. This chemical has a large range in
molecular weight, from 5000 to 1.4 million Da; however, usually 40 kDa is used in
inducing colitis (13).
Eight years after the discovery of DSS, the carrageenan model of IBD was
developed when rodents fed with seaweed developed disease which led to the
discovery of this sulfated polysaccharide that induced colitis (17). Later, due to its
similar characteristics to carrageenan, DSS was also used to induce disease in
hamsters (18), and then Okayasu and colleagues modified the model for mice
(8). DSS has been administered to hamsters, rabbits, rats and mice, with larger
rodents initially used because clinical signs could be easily monitored. Okayasu
et al. found that mice administered 3 to 10% DSS solutions quickly developed
inflammation resembling colitis in humans. They also found that changing the
treatment regimen by cycling the DSS treatment with periods of water, instead of
with continuous DSS treatment only, could induce chronic disease.

Mechanism of the model
The mechanism by which DSS-induced colitis develops is still unclear. However,
it is generally believed that inflammation is triggered after the epithelial barrier is
compromised by DSS, subsequently facilitating interactions between the host
and resident microbes, which then triggers a host immune response that
develops into inflammatory bowel disease.

50

DSS is thought to work by damaging the epithelial layers followed by
immunological stimulation by resident microbes (19). Researchers found that
DSS treatment led to reduction of zona occludens-1 protein production that is
involved in maintenance of tight junctions and these changes preceded
inflammation. There was also deposition of DSS in organs such as the liver and
uptake by macrophages, adding to evidence of DSS permeability. Others have
shown that DSS penetrates and disrupts the normally impervious mucin layer
allowing

bacteria

to

penetrate

the

barrier

within

12

hours

post-DSS

administration, before inflammatory changes are detected (20).
Another proposed mechanism for DSS action is by direct damage to intestinal
cells. Investigators found that the chemical caused cytotoxicity that increased
with time and concentration of DSS used (21). This group also found that
expression of various cell adhesion molecules were affected by DSS, indicating
that this alteration, particularly in β7 integrins, may lead to inflammation (21).
Araki et al. found that DSS inhibits generation of reactive oxygen species (ROS)
in Caco-2 cells, yet DSS administration in this murine model increases ROS,
leading authors to suggest that ROS production comes from deeper cells and not
the epithelial layer (22).
DSS also increases permeability of the epithelial layer, as seen in human IBD
(19), with significant increase in Evans Blue dye from the gut lumen after only 3
days of DSS administration, although no signs of inflammation or changes in
architecture were seen at this time point (19). This group also showed that DSS
may be responsible for basal crypt cell damage, and together with other cytotoxic
51

effects of DSS, it is clear that this chemical is involved in epithelial cell damage
(21, 23, 24), an outcome that may play an important role in induction of disease.
Another line of research suggests that the microbiota plays a major role in DSSinduced disease. Some have shown that germfree mice administered DSS
develop more severe disease than their conventionally raised counterparts,
suggesting a protective effect of the microbes (25).

This indicates that the

microbiota’s beneficial effects may be due to production of metabolites such as
short chain fatty acids (SCFA) that are necessary for colonocyte development
and epithelial cell health (26). However, others have shown that administration of
anaerobic bacterial antigens reduced severity of disease in mice treated with
DSS (27), leading these authors to speculate that ingested antigens increased
tolerance to intestinal bacteria. This indicates that altered homeostasis between
the microbiota and host immune response may be the driver for immune reaction
to DSS (27). These effects may be due to the antigens altering the microbial
community by competing for binding sites on epithelial surfaces, altering the
immune system that may have subsequently exerted effects of the microbiota.
Antibiotic studies show that certain antimicrobial therapies reduce inflammation
(16), indicating both a protective and detrimental effect of the microbiota on
colitis. Araki et al. treated Caco-2 cells in vitro with DSS and showed that H2S
(22), thought to be associated with DSS-induced colitis in mice and UC in
humans, was not produced by host reaction to the chemical, indicating that it
may be a by-product of metabolism by the microbiota (28). Others have tried to
explain the negative role of the microbiota by postulating that microbes act on
52

DSS and metabolize it to a form that causes inflammation. Kitajima et al.
incubated DSS with luminal contents for 24 hours before the components were
studied though spectrometry (29). They found that microbes did not metabolize
DSS, dispelling this potential mechanism of action. However, other evidence still
indicates an important role for the microbiota in this model of IBD (14, 20, 30).
Although there are many unknowns surrounding the mechanism of DSS disease
induction in mice, certain aspects of the model are known to be important. For
example molecular size of DSS and strain of mouse utilized can affect disease
severity. The molecular weight of the DSS administered to animals is relevant to
disease induction also (13). Investigators found that after testing the effects of
5% DSS solution using three different weights of DSS, mice fed with DSS
weights less than 5 kDa or over 500 kDa do not produce severe inflammation
while 40 kDa is typically used to trigger colitis (13). Others have also found that
the sulfur content was dose dependent on the ability of DSS to induce disease
with amounts lower than 16.6 % causing a significant reduction in inducing
inflammation (31). Authors speculated that larger molecular weights contained
more sulfur that was important in inducing disease when molecules permeate the
mucin layer, however DSS over 500 kDa are thought to be too large to penetrate
this layer, thus unable to induce inflammation.
This model is also dependent on strain of mice used with C57BL/6 mice being
very susceptible to disease and BALB/c being more resistant (12). There are also
strain dependent induction differences with BALB/c mice producing disease in
the distal colon while Swiss Webster mice show more inflammation in colon and
53

cecum (32, 33). It is suggested that these differences are due to genetically
controlled variation in inflammatory response or healing (34). Age and dose
dependence was also seen in rats with 4 week old Wistar rats being more
sensitive to DSS-induced colitis than those eight weeks old, and 4% DSS
producing more severe inflammation that 2% DSS over 7 days administration
(35).

Immunology of the model
Different phases of DSS-induced disease, acute and chronic, produce different
inflammatory responses. The acute state is normally characterized by an
increase in the Th1-Th17 response with up-regulation of TNFα, IL-6, IL-17 and
KC (40).

Other immunological characteristics of the acute state include up-

regulation of IL-12, INFγ and minor roles of Th2 mediators IL-10 and IL-4 (36). A
Th2 response is dominant in the chronic phase with a decrease in the
aforementioned mediators and increase in IL-4 and IL-10 (37). Egger et al. also
found that TNFα, IL-1 and IL-12 expression increased with increasing dose of
DSS, but not volume of DSS solution ingested.
Both Th1 and Th2 responses in Swiss Webster mice trigger DSS-induced colitis.
Dieleman et al. showed that the acute phase of DSS colitis is initiated by
neutrophils and macrophages and can lead to up-regulation of INFγ, IL-5 and IL4 expression (32), while the acute model requires macrophages predominantly
while the chronic disease model requires both macrophages and T cells (23).
54

Leung et al. showed that gamma/delta T cells interact with LPS and
macrophages before onset of inflammation and therefore likely to play an
important role in DSS-induced colitis in rats (38). However, findings that SCID
mice, which lack B and T cells, also developed disease after DSS treatment,
indicate that macrophages are the main drivers (23).

More recently, it was

reported that the adaptive response is required for both acute and chronic
phases of disease (39).

Variation in the model
This model was modified to be used in chronic colitis in hamsters in 1992 by
Yamada et al. (40), before being used in mice. DSS causes disease in several
strains of rodents with varying degrees of severity, allowing flexibility in choosing
an appropriate system for the biologic question at hand (12, 15, 41). Guinea pigs
develop severe inflammation within 3 days after start of DSS administration (42).
Among mice, C3H/HeJ are highly susceptible to treatment with BALB/c mice
being more resistant (12).
Long-term administration of DSS results in colorectal carcinoma, very similar to
dysplasia-carcinoma seen in cancer development in human ulcerative colitis (40,
43). Investigators treated hamsters with 1% DSS over 100 days and found that
the animals developed colorectal cancer. This was to provide a model of disease
seen in patients with long-term ulcerative colitis eventually developing colorectal
carcinomas.

55

Another variation of the DSS model is also used to study dysplasia and
carcinoma development by administering 3% of the chemical over seven days
followed by water for 14 days, for three cycles (44). Or mice are given a single
dose of azoxymethane (AOM) before three cycles of DSS prior to water
treatment.

Helicobacter hepaticus infected IL-10-/- murine model of IBD
Helicobacter species are a relatively newly identified group of bacteria that has
been documented to cause disease in humans and animals since the 1980’s
(45). Diseases are varied and dependent on infecting bacterial species. For
example, H. pylori causes peptic ulcers in humans, while H. rappini is associated
with intestinal disease in human and other mammals (46), and abortions in
guinea pigs (47). However, it was in 1992 that a new pathogen was found in mice
at the National Cancer Institute-Frederick Cancer Research and Development
Center (48), H. hepaticus, which was found in hepatic tissue of animals with liver
tumors. It is also triggers inflammation in immuno-compromised mice that
resembles human Crohn’s disease (1).
This model relies on an immuno-altered background since H. hepaticus does not
cause disease in wild-type mice (5, 49, 50). Disease resembles Crohn’s disease
in humans with a Th1 response and transmural inflammation, but like many other
models, it occurs in the lower intestine while human CD can occur anywhere
along the gastrointestinal tract. This model is very desirable as it is highly

56

reproducible, easy and cheap to perform. This model is also conducted in SCID
mice and severity of disease is dependent on mouse strains used (51).

Origin of the model
H. hepaticus is a microaerophilic, gram negative, flagellated (bipolar, sheathed),
spiral-shaped bacterium (may be curved or have several spirals), 1.5 to 5.0 µm
long and 0.2 to 0.3 µm wide. It has urease, oxidase and catalase activity,
produces H2S, and reduces nitrate to nitrite (52). It is associated with the
mucosal lining of the lower intestinal tract of mice and commonly occurs in the
cecum and colon (53, 54). It is not known as a human pathogen; however, recent
reports have indicated its presence in immuno-compromised children.
At one point 88% of mice in laboratory and animal facilities were infected with H.
hepaticus (55-57).

This bacterium can be a normal part of the resident

community of these mice and only in an immuno-altered state does it become
pathogenic. In wild-type mice this bacterium can represent up to 40% of the
microbiota (58). SCID mice and IL-10 deficient mice infected with H. hepaticus
then became a useful model to study human IBD since it produces a Th1-Th17
type immune response similar to Crohn’s disease in humans.

Mechanism of the model
The genome of H. hepaticus has been completely characterized and has been
shown to possess similarities to other pathogenic bacteria such as H. pylori
57

urease genes and the Campylobacter jejuni gene cluster that comprises
cytolethal distending toxin (59). Although urease is not necessary for intestinal
colonization, it plays a role in liver disease and triggers the immune response
(60). Additionally, the HHGI1 region, a pathogenicity island, contains 70 coding
regions which are mostly hypothetical proteins, three coding regions that are
similar to regions that code for the type IV secretion system, that also plays a role
in inducing disease. In a study with 12 strains of H. hepaticus it was observed
that the HHGI1 region was either completely in some strains or partially detected
in others which may be responsible for the varying virulence observed (29).
H. hepaticus produces cytolethal distending toxin (Cdt), which not only arrests
cell cycle in the G2 phase of cell division, but also has a role in colonization with
the Cdt mutant being cleared from infected animals within eight months of
infection (61). Researchers speculate that this may be due to the CdtB gene
functioning to suppress host’s immune response, as seen by lower colonic
inflammation in mice infected with the mutant, thereby allowing the bacterium to
avoid detection or clearance (61).
Disease does not occur in IL-10 deficient mice monoassociated with H. hepaticus
indicating an important role of the resident microbiota in development of disease
in this model of IBD (5). Furthermore, Helicobacter spp. have been found to
cause dysregulation of TLR2 and NOD1 in epithelial cells (62), receptors for
bacterial antigens. Investigation of the role of TLR4 and IL-10 in H. hepaticus
triggered colitis show that TLR/IL-10 deficient mice have elevated INFγ, IL-17,
TNFα, IL-1 and IL-6 when infected with this bacterium, emphasizing the role of
58

resident bacteria in induction of disease. Disease was more severe in double
knock-out mice than in IL-10 deficient only mice, which may be related to
increases the accumulation of apoptotic cells in the lamina propria in double
knock-outs.

Immunology of the model
This model elicits a Th1/Th17 type response with the production of TNFα, IL-17
and IL-23 (63). Kullberg et al. found that both IL-23 and IL-12 have p40 subunits,
which is an essential subunit in disease induction (63, 64). Previous work pointed
to IL-12 as having an important role in H. hepaticus triggered disease in IL-10
deficient mice after the use of anti-p40 monoclonal antibodies, however these
antibodies also targeted IL-23 (63). After researchers used more targeted
approaches to investigate these two cytokines it was determined that IL-23 and
not IL-12 is important in this model (63).

Furthermore, it was found that in the

absence of INFγ, IL-17 cannot produce severe disease. It was also found that
disease was inhibited only in IL-23 deficient mice. Authors suggest that when IL23 is down-regulated by IL-10, lowered IL-23 levels then suppress IL-17
production from TH17 cells.
This model hinges on the role of IL-10 in inhibiting diseases in mice. Wang et al.
examined the role of NFκB on IL-10 by testing the role of subunits of NFκB in IL10 expression with the use of IL-10 immunoglobulin (Ig) fusion proteins in RAG
mice (65). They found that the fusion protein prevented disease in H. hepaticus

59

infected RAG mice but not in mice deficient in p50. Radiation treatment further
revealed that p50/p105 subunit of this transcription factor in homopoietic cells of
the innate immune system is important in IL-10 expression and suppression of
disease (66). This indicates that there is an important role for the innate immune
system in H. hepaticus triggered colitis. Authors also found that c-Rel, a subunit
of NfκB, is needed for H. hepaticus triggered inflammation. This subunit is
necessary for proper functioning of IL-12/23 and absence prevented disease in
RAG-2 deficient mice and prevented the ability of adoptive transfer of
CD4+CD45RB high T cells to ameliorate disease (65).
The model works via the adaptive response with inflammation occurring in SCID
mice that had adoptive transfer of CD4+ T cells. Cahill et al. found that mice that
were infected with H. hepaticus developed severe inflammation if they had CD4+
T cells expressing high levels of CD45RB (1). Disease did not develop in mice
raised in germfree conditions indicating a role for resident bacteria, and a
dysfunctional immune response, in disease development.

Variations of this model
There are many lines of mice that develop colitis when infected with H. hepaticus
(51). SCID/NCr mice infected with H. hepaticus are used to demonstrate IBD.
Natural infection with H. hepaticus resulted in lesions in the liver and intestines
that progressed with age (49). Disease manifested was less severe than that
observed in A/JCr mice (67). This indicates that the T cell response caused

60

reduced disease as also suggested by IgG antibody levels being lower in SCID
mice compared to A/JCr mice. It is also possible that different microbiota present
in these mice strain play a role in the different disease manifestations observed.
Infection of H. hepaticus into BALB-Min and BALB-RagMin mice show that this
bacterium is involved with tumors of the colon and not small intestine. These
tumors are not dependent on severity of inflammation, but depend on organs on
which they occur (68).
Other species of Helicobacter have also been used in models of IBD, such as H.
bilis, H. cindai and H. fennili. Severity of disease that develops varies with the
species used. Co-infections of H. hepaticus and H. bilis produce colitis that later
develops into dysplasia in P-Glycoprotein-deficient mdr1a-/- mice. H. hepaticus
alone was found to delay colitis in these mice while mono-infection with H. bilis
accelerated colitis development. The co-infection is thought to progress to
dysplasia since the H. hepaticus slows progress of disease generated by the H.
bilis by inducing regulatory T-cells, and allowing abnormal growth as cells are
transformed by H. bilis activity (69). This system was proposed as a model to
study dysplasia, hyperplasia and neoplasia in humans.

Conclusions
By using the DSS-induced model of colitis and the H. hepaticus triggered IBD
model, roles of the microbiota in development of IBD can be assessed.
Upcoming experiments attempt to investigate roles of the microbiota in initiation
61

and progression of disease. With the DSS-induced model of colitis, changes in
the microbiota will be investigated at onset and progression of disease since
there is no reported data documenting whether dysbiosis is associated with
disease over time. Using the H. hepaticus triggered IBD model, where the
microbiota is essential in onset of disease, the role of changes in the microbiota
at initiation and during disease will also be investigated. By assessing how
changes in the microbial community during development of IBD affect disease
manifestation, advances towards prevention and therapy may be achieved.

62

REFERENCES

63

REFERENCES
1.
Cahill R, Foltz C, Fox J, et al. Inflammatory bowel disease: an immunitymediated condition triggered by bacterial infection with Helicobacter hepaticus.
Infect immun. 1997;65:3126-3131
2.
Hermiston ML, Gordon JI. Inflammatory Bowel Disease and Adenomas in
Mice Expressing a Dominant Negative N-Cadherin. Science. 1995;270:12031207
3.
Wirtz S, Neurath MF. Mouse models of inflammatory bowel disease. Adv
Drug Deliv Rev. 2007;59:1073-1083
4.
Warren S. Why study animal models of IBD? Inflamm Bowel Dis.
2008;14:S129-S131
5.
Dieleman L, Arends A, Tonkonogy S, et al. Helicobacter hepaticus does
not induce or potentiate colitis in interleukin-10-deficient mice. Infection and
Immunity. 2000;68:5107-5113
6.
Madsen KL, Doyle JS, Tavernini MM, et al. Antibiotic therapy attenuates
colitis in interleukin 10 gene-deficient mice. Gastroenterol. 2000;118:1094-1105
7.
Wirtz S, Neufert C, Weigmann B, et al. Chemically induced mouse models
of intestinal inflammation. Nat Protoc. 2007;2:541-546
8.
Okayasu I, Hatakeyama S, Yamada M, et al. A Novel Method in the
Induction of reliable Experimental Acute and Chronic Ulcerative Colitis in Mice.
Gastroenterol. 1990;98:694-702
9.
Hitotsumatsu O, Ahmad RC, Tavares R et al. The ubiquitin-editing
enzyme A20 restricts nucleotide-binding oligomerization domain containing 2triggered signals. Immunity. 2008;28:381-390
10.
Powrie F, Leach MW, Mauze S et al. Inhibition of Th1 responses prevents
inflammatory bowel disease in scid mice reconstituted with CD45RBhi CD4+ T
cells. Immunity. 1994;1:553-562

64

11.
Scheiffele F, Fuss I. Induction of TNBS colitis in mice. Curr Protoc
Immunol. 2002:15.19.11-15.19.14
12.
Mahler M, Bristol JI, Leiter HE, et al. Differential susceptibility of inbred
mouse strains to dextran sulphate sodium-induced colitis. Am J Physiol.
1998;274:544-551
13.
Kitajima S, Takuma S, Morimoto M. Histological analysis of murine colitis
induced by dextran sulfate sodium of different molecular weights. Exp Animals.
2000;49:9-15
14.
Hudcovic T, Stepankova R, Cebra J, et al. The Role of Microflora in the
Development of Intestinal Inflammation: Acute and Chronic Colitis Induced by
Dextran Sulfate in Germ-Free and Conventionally Reared Immunocompetent and
Immunodeficient Mice. Folia Microbiol (Praha). 2001;46:565-572
15.
Suzuki R, Kohno H, Sugie S, et al. Strain differences in the susceptibility
to azoxymethane and dextran sodium sulfate-induced colon carcinogenesis in
mice. Carcinogenesis. 2006;27:162-169
16.
Hans W, Schölmerich J, Gross V et al. The role of the resident intestinal
flora in acute and chronic dextran sulfate sodium-induced colitis in mice. Eur J
Gastroenterol Hepatol. 2000;12:267-273
17.
Watt J, Marcus R. Carrageenan-induced ulceration of the large intestine in
the guinea pig. Gut. 1971;12:164-171
18.
Ohkusa T. Production of experimental ulcerative colitis in hamsters by
dextran sulfate sodium and changes in intestinal microflora. Nippon Shokakibyo
Gakkai Zasshi. 1985;82:1327-1336
19.
Kitajima S, Takuma S, Morimoto M. Changes in Colonic Mucosal
Permeability in Mouse Colitis Induced with Dextran Sulfate Sodium. Exp Anim.
1999;48:137-143
20.
Johansson ME, Gustafsson JK, Sjöberg KE et al. Bacteria penetrate the
inner mucus layer before inflammation in the dextran sulfate colitis model. PloS
One. 2010;5:e12238

65

21.
Ni J, Chen SF, Hollander D. Effects of dextran sulphate sodium on
intestinal epithelial cells and intestinal lymphocytes. Gut. 1996;39:234-241
22.
Araki Y, Sugihara H, Hattori T. In vitro effects of dextran sulfate sodium on
a Caco-2 cell line and plausible mechanisms for dextran sulfate sodium-induced
colitis. Oncol Rep. 2006;16:1357-1362
23.
Dieleman LA, Ridwan BU, Tennyson GS et al. Dextran sulfate sodiuminduced colitis occurs in severe combined immunodeficient mice. Gastroenterol.
1994;107:1643-1652
24.
Mashimo H, Wu D-C, Podolsky DK, et al. Impaired Defense of Intestinal
Mucosa in Mice Lacking Intestinal Trefoil Factor. Science. 1996;274:262-265
25.
Kitajima S, Morimoto M, Sagara E, et al. Dextran Sodium Sufate-Induced
Colitis in Germ-Free IQI/Jic Mice. Exp Anim. 2001;50:387-395
26.
Ahmad MS, Krishnan S, Ramakrishna BS, et al. Butyrate and glucose
metabolism by colonocytes in experimental colitis in mice. Gut. 2000;46:493-499
27.
Verdu EF, Bercik P, Cukrowska B, et al. Oral administration of antigens
from intestinal flora anaerobic bacteria reduces the severity of experimemntal
acute colitis in BALB/c mice. Clin Exp Immunol. 2000;120:46-50
28.
Araki Y, Andoh A, Tsujikawa T et al. Alterations in intestinal microflora,
faecal bile acids and short chain fatty acids in dextran sulphate sodium-induced
experimental acute colitis in rats. Eur J Gastroenterol Hepatol. 2001;13:107-112
29.
Belzer C, Stoof J, Breijer S, et al. The Helicobacter hepaticus hefA gene is
involved in resistance to amoxicillin. Helicobacter. 2009;14:72-79
30.
Rath H, Schultz M, Freitag R, et al. Different Subsets of Enteric Bacteria
Induce and Perpetuate Experimental Colitis in Rats and MIce. Infect Immun.
2001;69:2277-2285
31.
Axelsson LG, Landström E, Goldschmidt TJ et al. Dextran sulfate sodium
(DSS) induced experimental colitis in immunodeficient mice: effects in CD4(+) cell depleted, athymic and NK-cell depleted SCID mice. Inflamm Res
1996;45:181-191
66

32.
Dieleman LA, Palmen MJ, Akol H, et al. Chronic experimental colitis
induced by dextran sulphate sodium (DSS) is characterized by Th1 and Th2
cytokines. Clin Expl Immunol. 1998;114:385-391
33.
Dieleman LA, Ridwan BU, Tennyson GS et al. Dextran sulfate sodiuminduced colitis occurs in severe combined immunodeficient mice. Gastroenterol.
1994;107:1643-1652
34.
Mähler M, Bristol IJ, Sundberg JP et al. Genetic analysis of susceptibility
to dextran sulfate sodium-induced colitis in mice. Genomics. 1999;55:147-156
35.
Shimizu T, Suzuki M, Fujimura J, et al. The Relationship Between the
Concentration of Dextran Sodium Sulfate and the Degree of Induced
Experimental Colitis in Weanling Rats. J Pediatr Gastroenterol Nutr.
2003;37:481-486
36.
Egger B, Bajaj-Elliott M, MacDonald TT et al. Characterisation of acute
murine dextran sodium sulphate colitis: cytokine profile and dose dependency.
Digestion. 2000;62:240-248
37.
Philip A, Nicholas CZ, Thuan N, et al. Distinct cytokine patterns identified
from multiplex profiles of murine DSS and TNBS-induced colitis. Inflamm Bowel
Dis. 2009;15:341-352
38.
Leung FW, Heng M, Allen S et al. Involvement of luminal bacteria, heat
shock protein 60, macrophages and gammadelta T cells in dextran sulfate
sodium-induced colitis in rats. Dig Dis Sci. 2000;45:142-1479
39.
Hall LJ, Faivre E, Quinlan A et al. Induction and Activation of Adaptive
Immune Populations During Acute and Chronic Phases of a Murine Model of
Experimental Colitis. Dig Dis Sci. 2010;Epub
40.
Yamada M, Ohkusa T, Okayasu I. Occurrence of dysplasia and
adenocarcinoma after experimental chronic ulcerative colitis in hamsters induced
by dextran sulphate sodium. Gut. 1992;33:1521-1527
41.
Stevceva L, Pavli P, Buffinton G, et al. Dextran sodium sulphate-induced
colitis activity varies with mouse strain but develops in lipopolysaccharideunresponsive mice. J Gastroentorol Hepatol. 1999;14:54-60
67

42.
Iwanaga T, Hoshi O, Han H et al. Morphological analysis of acute
ulcerative colitis experimentally induced by dextran sulfate sodium in the guinea
pig: some possible mechanisms of cecal ulceration. J Gastroenterol.
1994;29:430-438
43.
Isao O, Masahiro Y, Tetuo M, et al. Dysplasia and carcinoma development
in a repeated dextran sulfate sodium-induced colitis model. J Gastroenterol
Hepatol. 2002;17:1078-1083
44.
Okayasu I, Yamada M, Mikami T, et al. Dysplasia and carcinoma
development in a repeated dextran sulfate sodium-induced colitis model. J
Gastroenterol Hepatol. 2002;17:1078-1083
45.
Marshall BJ, Warren JR. Unidentified curved bacilli in the stomach of
patients with gastritis and peptic ulceration. Lancet. 1984;1:1311-1315
46.
Schauer DB, Ghori N, Falkow S. Isolation and characterization of
"Flexispira rappini" from laboratory mice. J Clin Microbiol. 1993;31:2709-2714
47.
Bryner JH, Ritchie AE, Pollet L et al. Experimental infection and abortion
of pregnant guinea pigs with a unique spirillum-like bacterium isolated from
aborted ovine fetuses. Am J Vet Res. 1987;48:91-95
48.
Ward JM, Fox JG, Anver MR, et al. Chronic Active Hepatitis and
Associated Liver Tumors in Mice Caused by a Persistent Bacterial Infection With
a Novel Helicobacter Species. J Natl Cancer Inst. 1994;86:1222-1227
49.
Li X, Fox JG, Whary MT, et al. SCID/NCr Mice Naturally Infected with
Helicobacter hepaticus Develop Progressive Hepatitis, Proliferative Typhlitis, and
Colitis. Infect Immun. 1998;66:5477-5484
50.
Chow J, Mazmanian S. A Pathobiont of the Microbiota Balances Host
Colonization and Intestinal Inflammation. Cell Host Microbe. 2010;7:265-276
51.
Foltz CJ, Fox JG, Cahill R, et al. Spontaneous Inflammatory Bowel
Disease in Multiple Mutant Mouse Lines: Association with Colonization by
Helicobacter hepaticus. Helicobacter. 1998;3:69-78

68

52.
Fox JG, Dewhirst FE, Tully JG, et al. Helicobacter hepaticus sp. nov., a
microaerophilic bacterium isolated from livers and intestinal mucosal scrapings
from mice. J Clin Microbiol. 1994;32:1238-1245
53.
Chan V, Crocetti G, Grehan M, et al. Visualization of Helicobacter species
within the murine cecal mucosa using specific fluorescence in situ hybridization.
Helicobacter. 2005;10:114-124
54.
Zhang L, Danon SJ, Grehan M et al. Natural colonization with
Helicobacter species and the development of inflammatory bowel disease in
interleukin-10-deficient mice. Helicobacter. 2005;10:223-230
55.
Bohr UR, Selgrad M, Ochmann C, et al. Prevalence and Spread of
Enterohepatic Helicobacter Species in Mice Reared in a Specific-Pathogen-Free
Animal Facility. J Clin Microbiol. 2006;44:738-742
56.
Goto K, Ohashi H, Takakura A, et al. Current status of Helicobacter
contamination of laboratory mice, rats, gerbils, and house musk shrews in Japan.
Curr Microbiol. 2000;41:161-166
57.
Nilsson HO, Ouis IS, Stenram U, et al. High Prevalence of Helicobacter
Species Detected in Laboratory Mouse Strains by Multiplex PCR-Denaturing
Gradient Gel Electrophoresis and Pyrosequencing. J Clin Microbiol.
2004;42:3781-3788
58.
Keuhl CJ, Wood HD, Marsh TL, et al. Colonization of the Cecal Mucosa by
Helicobacter hepaticus Impacts the Diversity of the Indigenous Microbiota. Infect
Immun. 2005;73:6952-6961
59.
Suerbaum S, Josenhans C, Sterzenbach T, et al. The complete genome
sequence of the carcinogenic bacterium Helicobacter hepaticus. PNAS.
2003;100:7901-7906
60.
Beckwith CS, McGee DJ, Mobley HL, et al. Cloning, Expression, and
Catalytic Activity of Helicobacter hepaticus Urease. Infect Immun. 2001;69:59145920
61.
Pratt JS, Sachen KL, Wood HD, et al. Modulation of host immune
responses by the Cytolethal Distending Toxin of Helicobacter hepaticus. Infect
Immun. 2006;74:4406-4503
69

62.
Chaouche-Drider N, Kaparakis M, Karrar A, et al. A commensal
Helicobacter sp. of the rodent intestinal flora activates TLR2 and NOD1
responses in epithelial cells. PLoS One. 2009;4:e5396
63.
Kullberg M, Jankovic D, Feng C, et al. IL-23 plays a key role in
Helicobacter hepaticus-induced T cell-dependent colitis. J Exp Med.
2006;203:2485-2494
64.
Kullberg MC, Rothfuchs AG, Jankovic D, et al. Helicobacter hepaticusInduced Colitis in Interleukin-10-Deficient Mice: Cytokine Requirements for the
Induction and Maintenance of Intestinal Inflammation. Infect Immun.
2001;69:4232-4241
65.
Wang Y, Rickman BH, Poutahidis T, et al. c-Rel Is Essential for the
Development of Innate and T Cell-Induced Colitis. J Immunol. 2008;180:81188125
66.
Tomczak M, Erdman S, Davidson A, et al. Inhibition of Helicobacter
hepaticus-induced colitis by IL-10 requires the p50/p105 subunit of NF-kappa B.
J Immunol. 2006;177:7332-7339
67.
Whary MT, Morgan TJ, Dangler CA, et al. Chronic Active Hepatitis
Induced by Helicobacter hepaticus in the A/JCr Mouse Is Associated with a Th1
Cell-Mediated Immune Response. Infect Immun. 1998;66:3142-3148
68.
Nagamine CM, Rogers AB, Fox JG, et al. Helicobacter hepaticus
promotes azoxymethane-initiated colon tumorigenesis in BALB/c-IL10-deficient
mice. Int J Cancer. 2008;122:832-838
69.
Maggio-Price L, Bielefeldt-Ohmann H, Treuting P, et al. Dual Infection with
Helicobacter bilis and Helicobacter hepaticus in P-Glycoprotein-Deficient mdr1a/- Mice Results in Colitis that Progresses to Dysplasia. Am J Pathol.
2005;166:1793-1806

70

This chapter has been published:
Nagalingam NA, Kao JK, Young VB. Microbial ecology of the murine gut
associated with the development of dextran sodium sulphate-induced colitis.
Inflamm Bowel Dis. 2010

I performed experiments, data retrieval and analyses. Dr. Kao provided custom
primers for detection of changes in expression of host mediators, scored
histology and provided discussion on observed host changes. I also prepared the
manuscript with contribution from two other authors.

71

Chapter 3

Microbial ecology of the murine gut associated with the development
of Dextran Sodium Sulfate-Induced colitis.

Abstract and Key Words:
Background: Dextran sodium sulfate (DSS) is used to induce murine colitis.
Although the exact mechanism by which DSS administration causes disease is
unknown, evidence suggests that resident bacteria play a role in the
development of murine DSS colitis, analogous to their role in human
inflammatory bowel diseases.

Methods: C57BL/6 mice received 5% DSS in drinking water, and were
euthanized 3 days and 14 days after initiation of DSS treatment. Cultureindependent methods were used to follow changes in community structure of the
gut’s microbiota following DSS treatment. Histologic evidence of disease and
changes in host gene expression were assessed.

Results: Histologic colitis was minimal in DSS-treated animals at three days, but
severe after 14 days. Analysis of 16S rRNA-encoding gene clone libraries
demonstrated that microbial communities in the ceca of DSS-treated mice were
distinct from those in control mice. The microbiota in the cecum of DSS-treated
72

animals was characterized by an overall decrease in microbial richness, an
increase in members of the phylum Verrucomicrobia, and decrease in
Tenericutes. Changes in the host’s inflammatory response and microbial
communities occurred before histologic appearance of severe disease in the
colon, but were seen concurrently in the cecum.

Conclusion: DSS administration is associated with reproducible changes in the
gut microbial diversity of mice. Changes in the microbial community and
increases in host mediators, such as TNFα and arginase, appeared before
development of severe inflammation in the cecum. This indicates that these
changes in the microbial community may play role in potentiation of the abnormal
inflammatory response seen in DSS-treated animals.
Key words: Microbiota, DSS, 16S rRNA-encoding gene, ecology, colitis

73

Introduction:
Microbes are essential to the health and well-being of their hosts (1, 2). However,
presence of bacteria alone is not enough to impart these benefits, rather
composition and relative abundance of specific microbes have an important role
in maintaining health (3). Alterations of community structure of the indigenous
microbiota have also been implicated in development of disease. One such
condition in which resident bacteria are thought to play a critical role in
pathogenesis is inflammatory bowel disease (IBD) (3, 4). Dysbiosis, defects in
immunoregulation and defects in the barrier function are all thought to contribute
to onset of disease (5-7). Evidence of a key role for the microbiota in
pathogenesis is provided by studies that demonstrate that antibiotics can reduce
or prevent inflammation both in patients and in murine models of disease (8, 9).
In IL-10-/- mice, which develop colitis after infection with certain bacteria or in the
setting of certain housing conditions, treatment with antibiotics is associated with
alterations in the microbial gut community (10), and prevention or amelioration of
disease (11).
Of the multiple murine models of IBD, one commonly used system involves
administration of dextran sodium sulfate (DSS) which induces disease very
similar to human ulcerative colitis (12). Antibiotic administration has been shown
to ameliorate DSS-induced colitis (13), and cathelicidin, an antimicrobial peptide,
was also found to have protective effects in this model of colitis (14), indicating
that the microbiota plays a role in this disease model system.

74

Multiple hypotheses have been postulated for the mechanism by which DSStriggers gut mucosal inflammation but the exact pathogenesis remains unclear
(14, 15). Given that human IBD is associated with an altered microbial diversity
(5, 16) and evidence for a role of the microbiota in DSS-induced colitis (14, 17),
we hypothesize that this model is also associated with an altered diversity of the
intestinal microbiota. To test whether DSS treatment can alter the microbial
community diversity, we employed molecular techniques targeting the 16S rRNAencoding gene to follow shifts in community structure of gut bacteria in animals
receiving DSS. We correlated changes in community structure with development
of disease.

Materials and Methods
Animals: C57BL/6 mice from a breeding colony initially established with
breeding stock from Jackson Laboratories (Bar Harbor, ME) were used for
experiments. Animal studies were conducted at Michigan State University and
were approved by the Michigan State University Animal Use Committee. Mice
were housed with autoclaved bedding, given sterile food and water ad libitum,
and exposed to 12:12 h light:dark cycles. Mice, between 12 to 16 wks old, were
assigned to cages according to gender. Dextran sodium sulfate (DSS) (36 00050 000 MW, ICN Biochemicals Inc., CA.) was administered as a 5% solution in
drinking water (8).
Three groups of mice were used in this experiment: ten control mice that were
maintained on sterile drinking water; ten mice that were placed on 5% DSS in
75

sterile drinking water for 3 days before being euthanized; and eleven mice that
were given 5% DSS for 14 days before euthanasia (18).

Necropsy and Histology: Mice were euthanized by CO2 asphyxiation and the
cecal tissue harvested as described previously (19). Tissues were gently washed
with 1X phosphate-buffered saline (PBS) to remove fecal contents, cut into
sections and snap-frozen in liquid nitrogen. One of these sections was used for
terminal restriction fragment length polymorphism (T-RFLP) analyses and clone
library analysis, and another for RNA extraction.
The remainder of the cecum was processed for histology as follows: Luminal
contents were removed, and the tissue was washed with PBS, placed in tissue
cassettes and submerged in 10% formalin for 24 hours. Tissue cassettes were
transferred to a 60% ethanol solution and then processed for paraffin embedding
and staining with haematoxylin and eosin (H&E).
Scoring was completed using the colitis index histological scoring system used by
Berndt et al. (20) and adapted from Rachmilewitz et al. (21). Briefly, sections
assessed on inflammation, transmural infiltration, cell wall thickening and
bleeding; and scored on a scale ranging from 0 to 40.

RNA extraction and PCR array analysis: Total RNA from cecal tissue was
isolated using TRIzol® (Invitrogen, Carlsbad, CA) as directed by the
manufacturer’s protocol. The High-Capacity cDNA Reverse Transcription Kit
(Applied Biosystems, Foster City, CA) was used to convert the total RNA to

76

cDNA. Changes in host gene expression were measured using an array of genespecific primers from Superarray® (Frederick, MD), designed for the following
targets (22): Jun, Cd80, Cd209a, Il12b, Irak3, Smad3, Arg1, Cd86, Mapk3, Il17a,
Mapk8, Tgfb1, Cd274, Creb1, Mapk1, Il1b, Tlr2, Tlr5, Vtcn1, Cx3cr1, Foxp3, Il2,
Tlr4, Tlr9, Ccr2, Cxcl1, Tnfrsf18, Il23a, H2-DMb1, Tnf, Ccr7, Cxcl10, Infg, Il4,
Nfkb1, Vegfa, Itgax, Cxcl5, Il10, Il6, Mapk14, Gapdh, Cd40, Cxcl2, Il12a, Irak4,
Pik3r1, ActB. The neutrophil marker, Ly-6G, was also quantified using the primer
set

from

Sasmono

et

al.

(23)

TGGACTCTCACAGAAGCAAAG-3'

and

(Ly-6G

forward

reverse

primer
primer

5'5'-

GCAGAGGTCTTCCTTCCAACA-3') as well as the Gapdh primer set from Cui et
al. (24) (forward primer 5'-ACCACAGTCCATGCCATCAC-3' and reverse primer
5'-TCCACCACCCTGTTGCTGTA-3).

Quantization

was

performed

using

LightCycler® 480 SYBR Green I Master and analyzed on a LightCycler® 480
system (Roche Diagnostics, Indianapolis, IN) as directed by manufacture’s
protocol, with the following cycling conditions: 95oC activation for 10 min; 40
cycles of 95oC denaturation for 15s, and 60oC annealing for 1 min. Resulting
threshold values were analyzed by calculating the 2-

ΔΔ

Ct

values, using GAPDH as

reference, to find fold regulation compared to no-DSS controls (25-27).

DNA extraction: Genomic DNA was extracted from tissue using the Qiagen
DNeasy® Blood & Tissue kit (Cat. No.69504, Valencia, CA). Briefly, tissue was
incubated overnight in lysis buffer and proteinase K at 56 oC. The following day,
this enzyme was denatured at 95 oC and DNA was purified using ethanol through

77

filter columns, as directed by the manufacturer’s protocol, and eluted in 30µl of
elution buffer. This genomic DNA was then used in preparation of clone libraries
and T-RFLP.

Clone library construction: For clone library construction, reaction was set up
as described previously (28), with illustra PuReTaqTM Ready To Go PCRTM beads
(GE Healthcare, Piscataway, NJ). Briefly, amplification by polymerase chain
reaction

was

performed

using

broad-ranged

primers,

(8F,

5'-

AGAGTTTGATCCTGGCTCAG-3'; 1492R, 5'-GGTTACCTTGTTACGACTT-3').
Amplicon purification was done using a commercial kit (GFX, GE Healthcare,
Piscataway, NJ) as directed by the manufacturer. Products were ligated into
TOPO

4®

vector

(Invitrogen

K4575-01,

Carlsbad,

CA)

according

to

manufacturer’s specifications, and transformed into Escherichia coli. Colonies
were picked into Luria Broth (LB) with carbenicillin (50 µg/ml)) and grown
overnight

at

37

o

C.

Vector

specific

CAGTCACGACGTTGTAAAACGACGGC-3';

primers
and

(M13F,
M13R,

5'5'-

CAGGAAACAGCTATGACCATG-3') were used to screen these colonies for
bacterial clones containing appropriate 1.5 KB amplicon insert. Partial 16S
sequences were determined by a single sequencing run using the 8F primer, at
the Genomic Core at Michigan State University. Raw sequence data were
processes through an automated “information pipeline” available through the
Ribosomal Database Project (RDP) Web site (http://rdp.cme.msu.edu/). Data
were screened using the Chimera Check Program before uploading into the

78

RDP. Following alignment of sequences via myRDP (29), distance matrices
representing each of the libraries were downloaded and then taxonomic
assignments designated (80% confidence cut-off) using Classifier through the
RDP website. These distance matrices were also input into mothur (30) to be
grouped into operational taxonomic units (OTUs). Analyses were performed
using a 97% sequence similarity to denote species level. An input table was also
generated for the EstimateS program via the RDP Pipeline. EstimateS (31) was
used to calculate ecological diversity indices from aligned sequences.
Dendrograms based on the Bray-Curtis similarity index were constructed using
the Mega3 program (32).
16S small subunit rRNA gene sequences obtained by clone library analysis were
subjected to in silico terminal restriction fragment length polymorphism (T-RFLP)
analysis using the TRF-cut program in the ARB suite of programs (33). The MspI
enzyme, a four base cutter, was used to calculate predicted terminal restriction
fragments (TRFs) from clone sequences. Histograms were then constructed
displaying relative abundances of these in silco generated TRFs and compared
to actual traces retrieved from T-RFLP analysis described below. Fragment sizes
obtained by T-RFLP anlaysis were compared to sizes of the TRFs generated in
silico from cloned 16S sequences. These fragments were identified within two
basepairs (± 2 bp) of predicted TRFs (34).

T-RFLP analysis: T-RFLP was performed as described elsewhere (35) using
primers 1492R and FAM labeled 8F. In summary, genomic DNA was amplified

79

using the aforementioned primers with PCR conditions: 94 oC denaturation for 2
min; 30 cycles of denaturation at 94 oC for 30 s, annealing at 58 oC for 45 s, and
extension at 72 oC for 90 s; and final extension at 72 oC for 4 min. The 1.5 kb
PCR product was verified and purified with GFX columns (GFX, GE Healthcare,
Piscataway, NJ), then subjected to digestion with MspI restriction enzyme for two
hours. Digested DNA was submitted for analysis to the Genome Technology
Support Facility (GTSF) at Michigan State University. Traces were visualized
using the program Genescan (Applied Biosystems, Foster City, CA).

The

T-RFLP

Stats

collection

of

programs

(http://styx.ibest.uidaho.edu/ibest/research.html) (36) were used to analyze
traces from T-RFLP output data. Briefly, peaks of each trace were binned into
operational taxonomic units (OTU) to produce data categorized by both OTUs
and

abundance.

This

was

then

used

as

input

for

the

EstimateS

(http://viceroyeebuconnedu/estimates) (31) program where the Bray-Curtis
values were calculated and used to compare the diversity among communities.
The Mega3 program (32) was used to construct dendrograms showing the
relationship between community structures using the Bray-Curtis values.
Significance differences between community structures were calculated using the
parsimony test from the mothur suite of programs.

qPCR analysis of microbial communities: The quantity of 16S rRNA operons
in samples relative to a single-copy host gene was measured using a

80

primer/probe set that targets a broad range of rRNA-encoding gene sequences
(5’-TCCTACGGGAGGCAGCAGT-3’),

reverse

GGACTACCAGGGTATCTAATCCTGTT-3’),

and

primer
probe

(5’-

(5’-[6-FAM]-

203CGTATTACCGCGGCTGCTGGCAC-[TAMRA]-3’) (37). A primer/probe set
targeting a 264-bp portion of the TNF alpha gene was used as a reference using
200 nanomoles of forward (TNFα_mu_se; 5’-GGCTTTCCGAATTCACTGGAG3’) and reverse primers (TNFα_mu_as; 5’- CCCCGGCCTTCCAAATAAA-3’), and
100

nanomoles

of

probe

(TNFα_mu_probe;

5’-[Cy5]-

ATGTCCATTCCTGAGTTCTGCAAAGGGA-[Iowa Black RQ™]-3’) 213 (38). The
reaction mix consists of LightCycler® 480 Probes Master reaction mix (Roche) at
1X concentration, and appropriate primer/probe pair. Amplification of each gene
was done under separate run conditions: cycling conditions for the 16S target
involved an activation step of 50 oC for 2 min followed by 95 oC for 10 min. Fortyfive cycles were done at 95 oC for 15 s and 60 oC for 1 min before holding. For
the TNF reference gene cycling conditions included an activation step of 50 oC
for 2 min followed by 95 oC for 10 min and forty-fives cycles of 95 oC for 20 s and
64 oC for 30 s. Calculations of 2-ΔΔCt were made to compare changes in the
amount of 16S from samples between treatment groups (21,22).

Statistical Analysis: The non-parametric Kruskal-Wallis test was used to
analyze histological scores. Differences in abundances of specific operational
taxonomic units among treatment groups were analyzed by ANOVA. Statistical
differences in dendrograms comparing distances between microbial communities

81

were calculated using the parsimony test function in mothur (30). Probability
values less than 0.05 were considered significantly different.

Results
DSS treatment leads to inflammation of the murine intestinal tract.
Wildtype mice were treated with 5% DSS in drinking water and gastrointestinal
tissue examined histologically after 3 and 14 days of treatment. All animals
survived DSS treatment and no significant clinical signs were noted. A mild
inflammatory infiltrate developed in the cecum 3 days after DSS administration
(figure 3.1 b), with notable progression of disease after 14 days characterized
development of a massive inflammatory cell infiltrate and severe edema (figure
3.1 c). These changes were significantly different from the control state (p<0.05),
as well as between 3 day and 14 day treatment outcomes (figure 3.1 g).
Similarly in the colon, a few animals developed morphological and inflammatory
changes after 3 days of DSS treatment, but as a group, these changes were not
significant when compared to controls. However, following 14 days of DSS
administration colonic tissue showed evidence of severe inflammation and
edema which was significantly greater than controls of animals after 3 days of
treatment (figure 3.1 h).

82

50 µm b

c

d

e

f

*

40
30
20
10
0

*

40

ns

**

*

20

Cecum

h

da
y

D
SS

14

3

da
y

no

D
SS

D
SS

da
y

14

3

da
y

D
SS

0

D
SS

10

N
o

g

**

30

D
SS

Inflammation

a

Colon

Figure 3.1. Histopathology in DSS-treated mice. Hemotoxylin and eosin (H&E)
stained sections were prepared from cecum sections of (a) untreated control
mice (b) mice after 3 days of DSS treatment and (c) after 14 days of DSS.
Arrow indicates submucosl edema. H&E sections were also prepared from
colon samples of (d) untreated controls (e) mice after 3 days of DSS
treatment (arrow points to inflammation) and (f) animals after 14 days of
DSS (arrow shows abscess). Histopathologic scores were calculated for
sections from all 31 animals for (g) cecum and (h) colon sections. Statistical
analysis done was Kruskal-Wallis test. *p< 0.05 **p<0.001. Initial
magnification 40X.

83

Changes in host gene expression following DSS treatment
Using a quantitative PCR array, we found significant changes in expression of
ten host genes in cecal tissue following DSS treatment (figure 3.2, supplemental
figure 1). Expression of arginase 1 began to increase after 3 days and was
significantly up-regulated ~seven fold (p<0.05) after 14 days of DSS
administration. A significant increase in tumor necrosis factor-alpha (TNFα) was
also seen after 3 days of DSS treatment, with a 12.9 fold up-regulation (p<0.05),
and remained significantly up-regulated at the 14-day time point.
Eight other genes were down-regulated, compared to untreated controls. They
were TLR 5, IL17, Ccr2, Cx3cr1, Cxcl2, Cd40, Cd80 and Cd209a. TLR 5 had
significant down-regulation after 14-days of DSS exposure, as did Cd40 and
Cxcl2. Cd80 showed significant change after 3 days of DSS only marginal
change after 14 days. Ccr2, Cx3cr1, Cd209a and IL17a showed significant downregulation both at the 3 day and 14 day time points.
Presence of neutrophils, as indicated by measuring the Ly-6G marker, increased
significantly with DSS treatment with a 7.7±1.2 fold increase after 3 days of DSS
administration, and 14.0±1.8 fold increase after 14 days of DSS (supplemental
table 3.1).

84

Fold regulation compared to No-DSS Control

15

*
*

10

*

3 Day DSS
14 Day DSS

5

0

-5

-10

TLR5 Ccr2 IL17a Cd80 Cd40 Cd209
TNF

Cx3cr1 Cxcl2
Arg1

*

*

*

*

*

*

*
*

*

*

*

*

Figure 3.2. Changes in host gene expression after 3-days (dark bars) or 14-days
(grey bars) ± standard deviation of DSS administration. Expression levels
were compared to expression in tissue from animals not exposed to DSS,
after normalization with GAPDH. Statistical analysis done was Student’s ttest. Statistically significant differences (p<0.05) are denoted by asterisks (*).

85

Shifts in gut microbial diversity following DSS treatment.
To obtain an overview of the status of the microbial community present in the
gastro-intestinal tract of each of the animals in this study, 16S terminal restriction
fragment length polymorphism (T-RFLP) analysis was performed on DNA
extracted from cecal tissue. Analysis of microbial communities in the cecum of all
31 animals in the study revealed that the structure of the microbial communities
from DSS treated mice were significantly different from the no-DSS controls
(p<0.001) (supplemental figures 3.2).
Since T-RFLP provides a broad overview of community structure without
information regarding changes in specific organisms, 16S clone libraries were
constructed to further investigate the microbial community in four representative
mice from each treatment group. A total of 328 partial 16S rRNA-encoding gene
sequences were retrieved from control animals (71, 84, 89 and 84 per animal),
364 sequences from animals 3 days after initiation of DSS treatment (88, 92, 93
and 91) and 370 from animals treated with 14 days of DSS (92, 94, 89 and 95).
Clone library analysis confirmed T-RFLP findings, indicating that community
compositions of the microbiota in DSS treated animals were more similar to each
other than they were to those of control mice (figure 3.3, supplemental figure
3.4). T-RFLP traces were compared to in silico TRFs of clone libraries to verify
that OTUs matched the traces (supplemental figures 3.5- 3.8). Communities in all
DSS-treated animals were significantly different from those in no-DSS controls
(p<0.001 by the parsimony test implemented in the analysis suite mothur (30)).
Rarefaction analysis of clone libraries revealed that DSS treatment resulted in a

86

decrease in phylotype richness (supplemental 3.4). This finding was also
supported by a decrease in the Simpson diversity index (1/D) following DSS
treatment from 43.0±12.1 to 21.3±8.3 (3 days) and 19.7±7.0 (14 days) (p<0.05).
However, quantitative PCR of the 16S gene from each treatment group showed
that there were no significant changes in overall bacterial biomass relative to noDSS controls (fold change for 3 day DSS= 0.68±0.92 SD, 14 day DSS=
0.97±0.82 SD).

87

*

0.40

0.30

14 Day DSS

0.20
3 Day DSS

0.10

0.00
No DSS

Figure 3.3. Comparison of cecal communities in control animals (black triangles)
and in animals following 3 days of DSS treatment (white squares) and 14
days of treatment (black squares). Using an OTU definition of 97% similarity,
the Bray-Curtis similarity metric was calculated for each pair-wise
comparison and then results displayed in denrogram format. Analysis by
parsimony tests indicate that there is a statistically significant (*p<0.001)
difference between the communities in control animals and in both groups of
DSS-treated animals.

88

We analyzed changes in the relative abundance of specific phylotypes by
classification of the 16S rRNA-encoding gene sequences. DSS treatment
resulted in a decrease in members of the phyla Bacteriodetes and Tenericutes.
Alterations in the microbial community were also apparent at lower taxonomic
levels. For example, there was an increase in unclassified genera under the
family Lachnospiraceae (figure 3.4, supplemental 3.3). The majority of classified
phylotypes in animals from all three experimental groups were members of the
phylum Firmicutes, which include the classes Clostridia and Bacilli. Bacteriodetes
was most abundant in control animals and represented to a lesser extent in DSS
treated groups (figure 3.4 A, supplemental 3.3). A striking observation was that
members of the phylum Verrucomicrobiae were detected only in DSS treated
mice and not in control mice (figure 3.4 A).
Among members of the phylum Bacteriodetes richness of the population did not
change, indicating that changes were mainly due to shift in abundances of
different bacteria within this phylum. Among the Firmicutes, unclassified genera
among families Ruminoccocaceae and Lachnospiraceae were most prevalent.
These groups showed treatment related shifts as seen in pie charts of figure 3.4
B. When analysis was done using 97% sequence similarity, there was high
richness among Lachnospiraceae, compared to other major groups within the
Firmicutes (table 3.1), indicating increased diversity within this category (p<0.05).
These groups also showed treatment related shifts (table 3.1).

89

Proportion of bacteria in each treatment group

A
98
96
94
92
90
88
86
84
82
80

Phylum Level Assignment
Tenericutes*
Crenarceota
Verrucomicrobiae*
Proteobacteria
Bacteriodetes
Firmicutes

0

No DSS Control

3 Day DSS

14 Day DSS

mean proportion of bacteria per treatment group
3.14 % ±1.82
0.30 % ±0.60
0.00 % ±0.00
0.55 % ±0.67
9.39 % ±5.00
86.90 % ±7.02

0.00 %
0.00 %
2.75 %
0.28 %
0.83 %
96.15 %

±0.00
±0.00
±1.10
±0.56
±0.55
±1.49

0.27 %
0.00 %
4.64 %
0.00 %
5.04 %
90.05 %

±0.54
±0.00
±3.42
±0.00
±4.17
±3.31

Figure 3.4 (A). Changes in the cecal gut microbial community following DSS
administration at the phylum level. Analysis was done by ANOVA and
statistical significance (p<0.05) is denoted by asterisks (*). For
interpretation of the references to color in this and all other figures, the
reader is referred to the electronic version of this dissertation.

90

B

No DSS Control
Assignments within Firmicutes

3 Day DSS

14 Day DSS

mean proportion of bacteria per treatment group

Lachnospiraceae (N/A) * 32.93 % ±9.22
Ruminococcaceae (N/A) 26.43 % ±9.58
Clostridiales (N/A)
10.30 % ±7.49
Lachnospiraceae
Incertae Sedis
10.10 % ±1.53

59.63 % ±6.27
16.43 % ±7.13
8.20 % ±5.62
8.32 %

±7.17

57.69 % ±11.64
17.39 % ± 5.74
8.95 % ± 4.03
8.75 % ± 3.46

* denotes statistical significance (p<0.05)

Figure 3.4. (B) Changes in the cecal gut microbial community following DSS
administration within the phylum Firmicutes. Analysis was done by ANOVA
and statistical significance (p<0.05) is denoted by asterisks (*). For
interpretation of the references to color in this and all other figures, the
reader is referred to the electronic version of this dissertation.

91

Shared OTUs among all groups
OTUs in control group only
OTUs in 3 day DSS group only
OTUs in 14 day DSS group only
Shared OTU in DSS treated groups

Clostridiales
N=17 (%)
6 (35)
4 (24)
2 (12)
2 (12)
3 (18)

Ruminococcus
N=36 (%)
16 (44)
8 (22)
4 (11)
5 (14)
3 (8)

Lachnospiraceae
N=65 (%)
15 (23)
13 (20)
8 (12)
12 (18)
17 (26)

Table 3.1 Showing distribution of OTU’s across treatments from major groups
within the Firmicutes.

92

Discussion:
Indigenous gut microbiota are felt to play a key role in pathogenesis of
inflammatory bowel disease. Much of the evidence for the involvement of
intestinal bacteria in IBD comes from studies with murine models of disease (39,
40). Many of these models have specific alterations in host defenses ranging
from altered epithelial cell function to altered innate or adaptive immunity (41-44).
These alterations in host defense result in abnormal interactions between the
host and indigenous microbiota that lead to disease (45).

Evidence for this

comes from the fact that antibiotic administration reduces severity of disease in
these models and rederivation of these mice to the germ-free state prevents
initiation of disease (13, 46).
Administration of DSS consistently triggers intestinal inflammation in rodents.
Similar to other models, antibiotic administration has a beneficial effect in the
system again pointing to a role for the indigenous microbiota in disease
development (47). Effects of DSS on germfree mice have been variable. In some
strains of mice more severe colitis is encountered while early death, without
development of significant inflammation, is observed in others (48, 49). Germfree IQI/Jic mice have a high mortality rate, compared to conventional mice,
when given a concentration of 5% DSS in drinking water (45). These mice died
secondary to hemorrhage within three days after DSS treatment, indicating a
protective function of the microbiota against toxic effects of DSS on the gut
epithelium. These results are consistent with the finding that antibiotic
93

administration or genetic defects in TLR signaling also increased mortality in
DSS-treated conventional mice (50, 51). Differences in these two studies
employing germ-free mice likely reflects differential effects of two doses of DSS
on the epithelium, but may also reflects differences in composition of the
microbiota of conventional control animals. Since specific subsets of the “normal”
microbiota have differing ability to initiate or sustain experimental colitis (47, 52)
these results emphasize the importance of characterizing changes in the gut
community associated with DSS-induced colitis across time.
Although it was originally suggested that IBD resulted from infection with an
unknown bacterial pathogen (53), the more contemporary view is that entire
shifts in the microbial community structure may be the trigger for disease (5). An
altered gut microbial community has been associated with disease as shown in
human studies and animal models (54, 55), suggesting that there is a change in
abundance of bacteria that are necessary for maintaining homeostasis (54, 55).
After administering DSS to mice, we show that treatment results in changes in
microbial community structure in the gut, also noted by others (51). Additionally,
we found that these changes occur early (within 3 days) and are characterized by
reduced overall diversity while bacterial load remains unchanged. Some
community members are below limits of detection after DSS treatment, while
others increase in abundance demonstrating a selective effect of DSS treatment
on specific members of the community. It has been shown that certain bacteria
can de-polymerization DSS and thus can grow in a DSS-rich environment better
than others (56). This may explain the bloom of Verrucomicrobia that we

94

observed in DSS treated mice. This group of microbes, which has been found in
mammalian intestine at very low abundances (57), metabolize sulfur and
degrade mucin (58). Therefore, perturbations of the community can cause
previously underrepresented members of the community to become dominant
(59), resulting in shifts in community structure, as seen in the phylum
Bacteriodetes and families Ruminococcaceae and Lachnospiraceae.
Another study has profiled effects of DSS treatment on the microbial community
(51). Similar to our study, it was noted that shifts in relative abundances of
specific phylotypes were observed following DSS treatment, although specific
changes were different from what we observed. One obvious reason for these
differences could be explained by the fact that different anatomic locations were
sampled (cecal mucosa used here versus distal colonic contents in study by
Heimesaat et al. (51)), as well as length and concentrations of DSS
administration. Additionally, it is likely that the baseline microbial communities
were quite different between the animals in that study and the animals we used.
Recent reports have demonstrated that the baseline microbiota can differ in
genetically identical animals obtained from different sources and these
differences can have a profound effect on host immune responses (60, 61).
In our study, disease severity in the cecum increased after 3 days and 14 days
post DSS administration, concurrent with changes observed in the microbial
community, indicating a link between the changes in diversity of the microbiota
and disease. Since shifts in the community structure can reduce beneficial
members of the indigenous microbiota that act to maintain epithelial health, such

95

as production of short-chain fatty acids and stimulation of mucin production (55,
62-65), it is likely that such changes can affect host inflammatory responses (45,
66).
Others have suggested that DSS-induced colitis occurs due to breaches in the
epithelial barrier with exposure to antigens produced by luminal microbes (5, 6773). However, there is evidence that detection of an altered microbial community
is an important factor in host response (74, 75). Our data indicate that DSS
induces changes in the microbiota, and together with a compromised barrier,
appears to facilitate disease initiation and progression. It has been shown that
the host responds to microbial-associated molecular patterns (MAMPS) via tolllike receptors (TLRs) in development of colitis (17, 40, 76). We found significant
a change in expression of TLR 5, expression of which was decreased as colitis
progressed. It has been previously demonstrated that TLR 5 is down-regulated in
patients with severe ulcerative colitis (77), in concurrence with our data.
However, the role of flagella in DSS colitis is not dependent on TLR5 (40, 78),
suggesting that down-regulation is perhaps a response from over-stimulation
from microbes through a deteriorating epithelial layer.
It is noteworthy that we saw an increase in arginase 1 after 14 days of DSS
treatment. This could be due to an increase of neutrophils after DSS
administration, as indicated by increased expression of the neutrophil marker
Ly6-G. Arginase 1 also has a protective role in the Citrobacter rodentium model
of colitis where this enzyme inhibition aggravated disease (79). It has been seen
that arginase expression increases with TNFα/lipopolysaccharides (LPS)

96

stimulation in human intestinal microvascular endothelial cells (80). With
breaches in the epithelial layer that allow increased stimulation by LPS, together
with an increase in TNFα as seen here and in other studies with DSS (70, 81), it
follows that arginase expression is likely to be up-regulated. Interestingly,
although TNFα appeared to be at lower levels at 14 days compared to 3 days,
arginase was up-regulated at 14 days compared to 3 days, consistent with the
presumed role for this gene product in the epithelial repair response. This is not
unexpected since DSS administration disrupts the epithelial barrier (15). To our
knowledge, this study is the first to document the significant increase of this
enzyme in DSS-induced colitis.
Although others have noted changes in the gut microbiota following DSS
treatment (51) after development of disease, this is the first study to monitor
DSS-induced changes in the intestinal microbiota over time. In this study, we not
only corroborate that changes in the gut microbial community are associated with
DSS-induced disease, but we also show that diversity of the microbiota changed
as early as after 3 days and continued to 14 days of DSS administration. Early
shifts in the microbial community were characterized with increases in
abundances of microbes previously undetectable by methods used. These
altered communities are present at onset of inflammation, and maintained as
inflammation severity increased, reinforcing the idea that disease occurs, and
persists, in the presence of an altered intestinal community. These early changes
in the microbiota support the view that disturbance of the resident bacterial

97

community structure may have a very important role in onset of diseases, such
as IBD.

Acknowledgment
I would like to extend thanks to Jason Pratt for technical assistance, and Dr.
Courtney Robinson, Dr. Grace Chen and Dr. Gary Huffnagle for helpful
discussions and review of the manuscript. I am especially grateful for the funding
provided by the Center for Microbial Pathogenesis at MSU and the NIH. I would
also like to acknowledge the Genomics Core at Michigan State University where
all the sequencing and T-RFLP data were generated, as well as the RDP staff for
facilitating the work.

98

SUPPLEMENTAL INFORMATION
Table 3.2. Changes in host gene expression following DSS treatment.
TARGET

FUNCTION

FOLDREGULATION
3-Day 14-Day
DSS
DSS

Th1 response
Interleukin 12a (Il12a)

differentiation of naïve T cells

Interleukin 12b (Il12b)

differentiation of naïve T cells

Interleukin 2 (Il2)

leukocytotrophic hormone

Tumor Necrosis Factor
alpha (Tnf)
Interferon gamma (Infg)

proinflammatory cytokine

Toll-like Receptor 2 (Tlr2)

pathogen recogition

Toll-like Receptor 4 (Tlr4)

pathogen recognition

Toll-like Receptor 5 (Tlr5)

pathogen recognition

Toll-like Receptor 9 (Tlr9)

pathogen recognition

Chemokine (C-X-C motif)
ligand 10 (Cxcl10)

chemoattractant

2.35 ±
0.81
1.42 ±
0.33
1.79 ±
0.00
12.91*
± 1.23
2.26 ±
0.20
-1.22 ±
0.91
1.21 ±
1.28
-2.38 ±
1.15
1.05 ±
1.41
2.76 ±
2.90

1.57 ±
1.66
-2.53 ±
0.40
-2.45 ±
0.10
10.51*
± 1.62
-2.05 ±
0.12
-2.51 ±
0.38
1.74 ±
1.31
-3.60*
± 1.51
-1.58 ±
1.53
2.64 ±
1.31

1.82 ±
0.00
-8.19*
± 1.29

-2.26 ±
0.26
-4.33*
± 1.28

-3.60*
± 0.58

-9.00*
± 0.89

2.48 ±
0.80

-2.02 ±
0.13

Th2 response
Interleukin 4 (Il4)
chemokine (C-C motif)
receptor 2 (Ccr2)
Th17 response
Interleukin 17a (Il17a)
Interleukin 23a (Il23a)

proinflammatory cytokine

differentiation of naïve T cells
chemoattractant

induction and mediation of
proinflammatory responses
(allergic)
differentiation of naïve T cells

99

Table 3.2 (cont’d)
CD4+ T cell marker
Interleukin 6 (Il6)
Mitogen-activated protein
kinase 8 (Mapk8)
Cluster of Differentiation
80 (Cd80)
Cluster of Differentiation
86 (Cd86)
chemokine (C-C motif)
receptor 7 (Ccr7)
V-set domain-containing
T-cell activation inhibitor 1
(Vtcn1)
Tumor necrosis factor
receptor superfamily
member 18 (Tnfrsf18)
"Integrin, alpha X
(Itgax/CD11c)"
Tregs
Transforming growth
factor beta (Tgfb1)
Interleukin 10 (Il10)
forkhead box P3 (Foxp3)
Transcription factors
nuclear factor kappa-lightchain-enhancer of
activated B cells (Nfkb1)
Jun

pro-inflammatory and antiinflammatory
"regulation of cellular activities
(gene expression, mitosis,
differentiation, proliferation, cell
survival/apoptosis)"
receptor

-1.41 ±
0.26
-1.83 ±
0.35

-2.94 ±
0.97
-1.75 ±
1.03

-3.59*
± 0.13
1.41 ±
2.10
-2.71 ±
0.43
1.61 ±
0.00

-2.45 ±
1.38
2.23 ±
1.73
-2.42 ±
1.97
-2.8 ±
0.00

receptor

1.21 ±
0.66

1.42 ±
1.69

receptor

1.47 ±
0.44

1.26 ±
2.05

differentiation of naïve T
cells/anti-inflammatory cytokine
anti-inflammatory cytokine
differentiation of naïve T
cells/anti-inflammatory cytokine

-1.4 ±
0.53
-2.85 ±
0.00
-1.6 ±
1.24

1.01 ±
1.24
-2.78 ±
1.06
-2.68 ±
1.02

regulation of the immune
response to infection

-1.19 ±
0.56

-1.1 ±
1.92

early response factor to AP-1

1.43 ±
1.47
-1.33 ±
0.90
-1.24 ±
0.90

1.62 ±
0.91
-1.1 ±
1.91
-1.17 ±
1.46

1.65 ±
1.84

1.94 ±
1.36

receptor
receptor
ligand on antigen presenting
cells

Smad3
Interleukin-1 receptorassociated kinase 3
(Irak3)
Interleukin-1 receptorassociated kinase 4
(Irak4)

signalling innate response
signalling innate response

100

Mitogen-activated protein
kinase 14 (Mapk14)

"regulation of cellular activities
(gene expression, mitosis,
differentiation, proliferation, cell
survival/apoptosis)"
Mitogen-activated protein
"regulation of cellular activities
kinase 1 (Mapk1)
(gene expression, mitosis,
differentiation, proliferation, cell
survival/apoptosis)"
Mitogen-activated protein
"regulation of cellular activities
kinase 3 (Mapk3)
(gene expression, mitosis,
differentiation, proliferation, cell
survival/apoptosis)"
CAMP responsive element binds DNA to regulate
binding protein 1 (Creb1)
transcription
B cell
Phosphatidylinositol 3kinase regulatory subunit
alpha (Pik3r1)
Cluster of differentiation
40 (Cd40)
Other
Cluster of differentiation
209a (Cd209a)
Arginase 1 (Arg1)

Table 3.2 (cont’d)
-1.29 ± -1.35 ±
0.38
1.84
-1.27 ±
0.31

-1.03 ±
2.32

-1.49 ±
1.61

-1.57 ±
1.85

-1.45 ±
0.25

-1.32 ±
1.76

insulin receptor

-1.38 ±
0.64

-1.14 ±
1.16

receptor

-2.87 ±
1.28

-4.98*
± 1.31

receptor

-3.03*
± 0.30
2.59 ±
0.45
2.58 ±
1.13
-6.31*
± 1.30
2.23 ±
0.00
2.23 ±
1.60
-2.4 ±
0.18
1.71 ±
0.20
-1.08 ±
1.49
7.7*±
1.20

-5.86*
± 1.05
7.22* ±
1.46
1.23 ±
1.11
-3.00*
± 1.79
-1.93 ±
0.13
2.94 ±
2.13
-2.65 ±
1.93
-2.42 ±
0.29
-3.46*
± 0.84
14.0*±
1.80

wound healing

Cluster of differentiation
274 (Cd274)
Cx3cr1

receptor

Chemokine (C-X-C motif)
ligand 1 (Cxcl1)
MHC class II beta chain
(H2-DMb1)
Vascular endothelial
growth factor A (Vegfa)
Chemokine (C-X-C motif)
ligand 5 (Cxcl5)
Chemokine (C-X-C motif)
ligand 2 (Cxcl2)
Lymphocyte antigen 6
complex (Ly-6G)

chemoattractant

receptor

antigen processing and
presentation
cell differentiation, maturation,
proliferation, anti-apoptosis
chemoattractant
chemoattractant
Neutrophil marker

101

Table 3.3. Taxonomic distribution of rRNA genes clones from RDP classifier
analysis of treatment group
3
14
Day
Day
No
PHYLUM CLASS
FAMILY
GENUS
DSS DSS
DSS
Total Bacteroidetes
33
3
19
Bacteroid
etes

Bacteroid
etes

Porphyrom
onadaceae

Tannerella

13

2

10

Parabacteroides

1
2
17

0
1
0

0
2
7

1

0

0

282

351

334

7
2

4
0

4
1

4
9
0
78
1

6
0
1
58
0

4
0
0
58
0

30
2
0
1
0
95

29
0
1
0
6
209

29
0
0
0
6
193

1
2

0
1

0
4

2
26
3

0
30
1

0
30
0

N/A
Crenarcha
eota
N/A
Total
Firmicute
s
Firmicutes

Clostridia

Ruminococ
caceae
Anaerotruncus
Papillibacter
Ruminococcace
ae Incertae
Sedis
Ruminococcus
Fastidiosipila
N/A
Sporobacter
Lachnospir Lachnospiracea
aceae
e Incertae Sedis
Roseburia
Lachnospira
Catonella
Butyrivibrio
N/A
Peptococc
aceae
Peptococcus
N/A
Peptostreptococ
Peptostrep caceae Incertae
tococca
Sedis
N/A

102

Table 3.3 (cont’d)

Bacilli

Lactobacill
aceae
N/A

Erysipelotr Erysipelotri
chaceae
ichi

Lactobacillus

2
2

1
0

0
0

Allobaculum
Turicibacter
Erysipelotrichac
eae Incertae
Sedis
Holdemania
Erysipelotrichac
eae Incertae
Sedis

12
4

2
0

0
3

0
0

1
1

1
0

0
6
2

0
0
1

1
0
0

Herbaspirillium

1
1
0

0
0
1

0
0
0

Anaeroplasma

11

0

1

Akkermansia

0
1
328

10
0
365

17
0
371

N/A
Total Proteobacteria
Proteobac
teria

Betaprote
obacteria

Oxalobacte
raceae
N/A

N/A
Tenericut
es

Mollicute
s

Anaeropla
smatacea
e

Verrucom Verrucom Verrcomic
icrobiae
icrobia
robiaceae
N/A
TOTAL

103

*

**

0.25

0.20

14 Day DSS

0.15 0.10 0.05

3 Day DSS

0.00

No DSS

Figure 3.5. Comparison of the cecal community in control animals (black
triangles), and animals following 3 days of DSS treatment (white squares)
and 14 days of treatment (black squares). Using T-RFLP, the Bray-Curtis
similarity metric was calculated for each pair-wise comparison and then the
results displayed in dendrogram format. Communities that were further
analyzed, by construction of clone libraries, are indicated by arrows.
Analysis by the parsimony test indicates that there is a statistically significant
difference (*p<0.05 and **p<0.001).

104

140
120

OTU

100
80
60
40

14 day
3 day
Control

20
0

0

50

100

150

200

250

300

350

400

Collection effort/ Sequences

Figure 3.6. The rarefaction curves at 97% sequence similarity comparing the
microbial communities in mice after 3-days of DSS treatment (pink), 14-days
of DSS treatment (blue) and the controls (yellow). For interpretation of the
references to color in this and all other figures, the reader is referred to
the electronic version of this dissertation.

105

in silico TRFs

Relative flourescence units

T-RFLP traces of control communities

size (base pair)
Figure 3.7. Comparisons between T-RFLP traces and histograms of the terminal
restriction fragments (TRFs) of in silico digest of the clone library constructed
from the control communities.

106

Relative flourescence units

T-RFLP traces of 3 day DSS communities in silico TRFs

size (base pair)
Figure 3.8. Comparisons between T-RFLP traces and histograms of the terminal
restriction fragments (TRFs) of in silico digest of the clone library constructed
from the 3-day DSS administered communities.
107

Relative flourescence units

T-RFLP traces of 14 day DSS communities in silico TRFs

size (base pair)
Figure 3.9. Comparisons between T-RFLP traces and histograms of the terminal
restriction fragments (TRFs) of in silico digest of the clone library constructed
from the 14-day DSS administered communities.
108

TREATMENT PHYLUM

GENUS

Control

Tannerella (1)
Coprococcus (1)
Roseburia
Syntrophococcus
Coprococcus (2)
Coprococcus (3)
Oscillibacter (1)
Oscillibacter (2)
Tannerella (2)
Coprococcus (3)
Coprococcus (6)
Dorea
Peptococcus
Akkermansia

3 Day DSS

14 Day DSS

Bacteriodetes
Firmicutes
Firmicutes
Firmicutes
Firmicutes
Firmicutes
Firmicutes
Firmicutes
Bacteriodetes
Firmicutes
Firmicutes
Firmicutes
Firmicutes
Verrucomicrobia

TRF from
clone (size/
base pairs)
83
261
76
115
197
236
259
252
58
176
142
175
258
223

TRF from
T-RFLP
trace (size/
base
pairs)
83
260
78
116
197
235
260
253
56
174
142
174
258
223

Table 3.4. Comparison of common terminal restriction fragment lengths
generated by in silico digests of clones and the terminal restriction fragment
lengths from T-RFLP via Msp1 digests.

109

REFERENCES

110

REFERENCES
1.
Dethlefsen L, McFall-Ngai, M. & Relman, D. A. An ecological and
evolutionary perspective on human–microbe mutualism and disease. Nature.
2007;449:811-818
2.
Ley RE, Peterson DA, Gordon JI. Ecological and Evolutionary Forces
Shaping Microbial Diversity in the Human Intestine. Cell. 2006;24:837-848
3.
Round J, Mazmanian S. The gut microbiota shapes intestinal immune
responses during health and disease. Nat Rev Immunol. 2009;9:313-323
4.
Subramanian S, Campbell BJ, Rhodes JM. Bacteria in the pathogenesis
of inflammatory bowel disease. Curr Opin Infect Dis. 2006:475-484
5.
Packey C, Sartor R. Interplay of commensal and pathogenic bacteria,
genetic mutations, and immunoregulatory defects in the pathogenesis of
inflammatory bowel diseases. J Intern Med. 2008;263:597-606
6.
Swidsinski A, Ladhoff A, Pernthaler A, et al. Mucosal flora in inflammatory
bowel disease. Gastroenterol. 2002;122:44-54
7.
Ogura Y, Bonen DK, Inohara N, et al. A frameshift mutation in NOD2
associated with susceptibility to Crohn's disease. Nature. 2001;411:603-606
8.
Kang SS, Bloom SM, Norian LA, et al. An antibiotic-responsive mouse
model of fulminant ulcerative colitis. PLoS Med. 2008;5:e41
9.
Swidsinski A, Weber J, Loening-Baucke V, et al. Spatial Organization and
Composition of the Mucosal Flora in Patients with Inflammatory Bowel Disease. J
Clin Microbiol. 2005;43:3380-3389
10.
Sellon RK, Tonkonogy S, Schultz M, et al. Resident enteric bacteria are
necessary for development of spontaneous colitis and immune system activation
in interleukin-10-deficient mice. Infect Immun. 1998;66:5224-5231
11.
Madsen KL, Doyle JS, Tavernini MM, et al. Antibiotic therapy attenuates
colitis in interleukin 10 gene-deficient mice. Gastroenterol. 2000;118:1094-1105
111

12.
Okayasu I, Hatakeyama S, Yamada M, et al. A Novel Method in the
Induction of reliable Experimental Acute and Chronic Ulcerative Colitis in Mice.
Gastroenterol 1990;98:694-702
13.
Rath H, Schultz M, Freitag R, et al. Different Subsets of Enteric Bacteria
Induce and Perpetuate Experimental Colitis in Rats and MIce. Infect Immun.
2001;69:2277-2285
14.
Tai EK, Wu WK, Wong HP, et al. A New Role for Cathelicidin in Ulcerative
Colitis in Mice. Exp Biol Med (Maywood). 2007;232:799-808
15.
Kitajima S, Takuma S, Morimoto M. Changes in Colonic Mucosal
Permeability in Mouse Colitis Induced with Dextran Sulfate Sodium. Exp Anim.
1999;48:137-143
16.
Frank DN, St. Amand AL, Feldman RA, et al. Molecular-phylogenetic
characterization of microbial community imbalances in human inflammatory
bowel diseases. PNAS. 2007;104:13780-13785
17.
Rakoff-Nahoum S, Paglino J, Eslami-Varzaneh F, et al. Recognition of
Commensal Microflora by Toll-Like Receptors Is Required for Intestinal
Homeostasis. Cell. 2004;118:229-241
18.
Vowinkel T, Kalogeris T, Mori M, et al. Impact of Dextran Sulfate Sodium
Load on the Severity of Inflammation in Experimental Colitis. Dig Dis Sci.
2004;49:556-564
19.
Young VB, Knox KA, Pratt JS, et al. In Vitro and In Vivo Characterization
of Helicobacter hepaticus Cytolethal Distending Toxin Mutants. Infect Immun.
2004;72:2521-2527
20.
Berndt BE, Zhang M, Chen G-H, et al. The Role of Dendritic Cells in the
Development of Acute Dextran Sulfate Sodium Colitis. J Immunol.
2007;179:6255-6262
21.
Rachmilewitz D, Karmeli F, Takabayashi K, et al. Immunostimulatory DNA
ameliorates experimental and spontaneous murine colitis. Gastroenterol.
2002;122:1428-1441

112

22.
Lees CW, Zacharias WJ, Tremelling M, et al. Analysis of Germline GLI1
Variation Implicates Hedgehog Signalling in the Regulation of Intestinal
Inflammatory Pathways. PLoS Med. 2008;5:1761-1775
23.
Sasmono RT, Ehrnsperger A, Cronau SL, et al. Mouse neutrophilic
granulocytes express mRNA encoding the macrophage colony-stimulating factor
receptor (CSF-1R) as well as many other macrophage-specific transcripts and
can transdifferentiate into macrophages in vitro in response to CSF-1. J Leukoc
Biol. 2007;82:111-123
24.
Cui W, Cuartas E, Ke J, et al. CD200 and its receptor, CD200R, modulate
bone mass via the differentiation of osteoclasts. PNAS. 2007;104:14436-14441
25.
Pfaffl MW. A new mathematical model for relative quantification in realtime RT-PCR. Nucleic Acids Res. 2001;29:e45
26.
Guide to Performing Relative Quantitation of Gene Expression Using
Real-Time Quantitative PCR. Applied Biosystems. 2004
27.
Schmittgen TD, Livak KJ. Analyzing real-time PCR data by the
comparative C(T) method. Nat Protoc. 2008;3:1101-1108
28.
Young VB, Schimdt TM. Antibiotic-Associated Diarrhea Accompanied by
Large-Scale Alterations in the Composition of the Fecal Microbiota. J Clin
Microbiol. 2004;42:1203-1206
29.
Cole JR, Chai B, Farris RJ, et al. The ribosomal database project (RDP-II):
introducing myRDP space and quality controlled public data. Nucleic Acids Res.
2007;35:D169-172
30.
Schloss P, Westcott S, Ryabin T, et al. Introducing mothur: Open Source,
Platform-independent, Community-supported Software for Describing and
Comparing Microbial Communities. Appl Environ Microbiol. 2009;75:7537-7541
31.
Colwell RK. EstimateS: Statistical estimation of species richness and
shared species from samples. User's Guide and application published at:
http://purloclcorg/estimates. 2006

113

32.
Kumar S, Tamura K, Nei M. MEGA3: integrated software for molecular
evolutionary genetics analysis and sequence alignment. Brief Bioinform.
2004;5:150-163
33.
Ludwig W, Strunk O, Westram R, et al. ARB: a software environment for
sequence data. Nucleic Acid Res. 2004;32:1363-1371
34.
Liu WT, Marsh TL, Cheng H, et al. Characterization of microbial diversity
by determining terminal restriction fragment length polymorphisms of genes
encoding 16S rRNA. Appl Environ Microbiol. 1997;63:4516-4522
35.
Keuhl CJ, Wood HD, Marsh TL, et al. Colonization of the Cecal Mucosa by
Helicobacter hepaticus Impacts the Diversity of the Indigenous Microbiota. Infect
Immun. 2005;73:6952-6961
36.
Abdo Z, Schuette UME, Bent SJ, et al. Statistical methods for
characterizing diversity of microbial communities by analysis of terminal
restriction fragment length polymorphisms of 16S rRNA genes. Environmental
Microbiology. 2006;8:929-938
37.
Martin FE, Nadkarni MA, Jacques NA, et al. Quantitative Microbiological
Study of Human Carious Dentine by Culture and Real-Time PCR: Association of
Anaerobes with Histopathological Changes in Chronic Pulpitis. J Clin Microbiol.
2002;40:1698-1704
38.
Nitsche A, Becker M, Junghahn I, et al. Quantification of human cells in
NOD/SCID mice by duplex real-time polymerase-chain reaction. Haematologica.
2001;86:693-699
39.
Wirtz S, Neurath MF. Mouse models of inflammatory bowel disease. Adv
Drug Deliv Rev. 2007;59:1073-1083
40.
Erridge C, Duncan SH, Bereswill S, et al. The Induction of Colitis and
Ileitis in Mice Is Associated with Marked Increases in Intestinal Concentrations of
Stimulants of TLRs 2, 4, and 5. PLoS. 2010;5:e9125
41.
Tang Y, Clayburgh DR, Mittal N, et al. Epithelial NF-{kappa}B Enhances
Transmucosal Fluid Movement by Altering Tight Junction Protein Composition
after T Cell Activation. Am J Pathol.176:158-167
114

42.
Shea-Donohue T, Thomas K, Cody MJ, et al. Mice deficient in the CXCR2
ligand, CXCL1 (KC/GRO-{alpha}), exhibit increased susceptibility to dextran
sodium sulfate (DSS)-induced colitis. Innate Immun. 2008;14:117-124
43.
Blumberg RS, Saubermann LJ, Strober W. Animal models of mucosal
inflammation and their relation to human inflammatory bowel disease. Curr Opin
Immunol. 1999;11:648-656
44.
Belardelli F, Ferrantini M. Cytokines as a link between innate and adaptive
antitumor immunity. Trends Immunol. 2002;23:201-208
45.
Abt MC, Artis D. The intestinal microbiota in health and disease: the
influence of microbial products on immune cell homeostasis. Curr Opin
Gastroenterol. 2009;25:496-502
46.
Sartor RB. Microbial influences in inflammatory bowel diseases.
Gastroenterol. 2008;134:577-594
47.
Rath HC, Schultz M, Freitag R, et al. Different subsets of enteric bacteria
induce and perpetuate experimental colitis in rats and mice. Infect Immun.
2001;69:2277-2285
48.
Hudcovic T, Stepankova R, Cebra J, et al. The Role of Microflora in the
Development of Intestinal Inflammation: Acute and Chronic Colitis Induced by
Dextran Sulfate in Germ-Free and Conventionally Reared Immunocompetent and
Immunodeficient Mice. Folia Microbiol. 2001;46:565-572
49.
Kitajima S, Morimoto M, Sagara E, et al. Dextran sodium sulfate-induced
colitis in germ-free IQI/Jic mice. Exp Anim. 2001;50:387-395
50.
Rakoff-Nahoum S, Paglino J, Eslami-Varzaneh F, et al. Recognition of
commensal microflora by toll-like receptors is required for intestinal homeostasis.
Cell. 2004;118:229-241
51.
Heimesaat M, Fischer A, Siegmund B, et al. Shift towards proinflammatory intestinal bacteria aggravates acute murine colitis via Toll-like
receptors 2 and 4. PLoS ONE. 2007;25:e662

115

52.
Hoentjen F, Harmsen HJ, Braat H, et al. Antibiotics with a selective
aerobic or anaerobic spectrum have different therapeutic activities in various
regions of the colon in interleukin 10 gene deficient mice. Gut. 2003;52:17211727
53.
Hansen R, Thomson JM, El-Omar EM, et al. The role of infection in the
aetiology of inflammatory bowel disease. J Gastroenterol. 2010;45:266-276
54.
Maukonen J, Satokari R, Matto J, et al. Prevalence and temporal stability
of selected clostridial groups in irritable bowel syndrome in relation to
predominant faecal bacteria. J Med Microbiol. 2006;55:625-633
55.
Cummings JH, Macfarlane GT. Role of intestinal bacteria in nutrient
metabolism. JPEN J Parenter Enteral Nutr. 1997;21:357-365
56.
Araki Y, Mukaisho K, Sugihara H, et al. Proteus mirabilis sp. intestinal
microflora grow in a dextran sulfate sodium-rich environment. Int J Mol Med.
2010;25:203-208
57.
Wang M, Ahrne S, Jeppsson B, et al. Comparison of bacterial diversity
along the human intestinal tract by direct cloning and sequencing of 16S rRNA
genes. FEMS Microbiol Ecol; 2005:219-231
58.
Derrien M, Vaughan EE, Plugge CM, et al. Akkermansia muciniphila gen.
nov., sp. nov., a human intestinal mucin-degrading bacterium. Int J Syst Evol
Microbiol. 2004;54:1469-1476
59.
Antonopoulos D, Huse S, Morrison H, et al. Reproducible community
dynamics of the gastrointestinal microbiota following antibiotic perturbation. Infect
Immun. 2009;77:2367-2375
60.
Ivanov II, Frutos Rde L, Manel N, et al. Specific microbiota direct the
differentiation of IL-17-producing T-helper cells in the mucosa of the small
intestine. Cell Host Microbe. 2008;4:337-349
61.
Ivanov II, Atarashi K, Manel N, et al. Induction of intestinal Th17 cells by
segmented filamentous bacteria. Cell 2009;139:485-498

116

62.
Ahmad MS, Krishnan S, Ramakrishna BS, et al. Butyrate and glucose
metabolism by colonocytes in experimental colitis in mice. Gut. 2000;46:493-499
63.
Barcenilla A, Pryde SE, Martin JC, et al. Phylogenetic Relationships of
Butyrate-Producing Bacteria from the Human Gut. Appl Environ Microbiol.
2000;66:1654-1661
64.
Arijs I, De Hertogh G, Lemaire K, et al. Mucosal gene expression of
antimicrobial peptides in inflammatory bowel disease before and after first
infliximab treatment. PLoS One. 2009;24:e7984
65.
Wostmann B. The germfree animal in nutritional studies. Annu Rev Nutr.
1981;1: 257-279
66.
Clarke TB, Davis KM, Lysenko ES, et al. Recognition of peptidoglycan
from the microbiota by Nod1 enhances systemic innate immunity. Nature
Medicine. 2010;16:228-231
67.
Johansson ME, Phillipson M, Petersson J, et al. The inner of the two Muc2
mucin-dependent mucus layers in colon is devoid of bacteria. PNAS.
2008;105:15064-15069
68.
Joerg DS, Svenja P, Maren A, et al. Epithelial Tight Junctions in Intestinal
Inflammation. Ann NY Acad Sci. 2009;1165:294-300
69.
Nenci A, Becker C, Wullaert A, et al. Epithelial NEMO links innate
immunity to chronic intestinal inflammation. Nature. 2007;446:557
70.
te Velde A, de Kort F, Sterrenburg E, et al. Comparative analysis of
colonic gene expression of three experimental colitis models mimicking
inflammatory bowel disease. Inflamm Bowel Dis. 2007;13:325-330
71.
Podolsky D, Gerken G, Eyking A, et al. Colitis-associated variant of TLR2
causes impaired mucosal repair because of TFF3 deficiency. Gastroenterol.
2009;137:209-220
72.
Yan Y, Kolachala V, Dalmasso G, et al. Temporal and spatial analysis of
clinical and molecular parameters in dextran sodium sulfate induced colitis.
PLOS One. 2009;4:e6073
117

73.
Ito R, Kita M, Shin-Ya M, et al. Involvement of IL-17A in the pathogenesis
of DSS-induced colitis in mice. Biochem Biophys Res Commun. 2008;377:12-16
74.
Stefan L, Dick SD, Eva J, et al. The role of the Lps gene in experimental
ulcerative colitis in mice. APMIS. 1996;104:823-833
75.
Salzman NH, Bevins CL. Negative interactions with the microbiota: IBD.
Adv Exp Med Biol. 2008;635:67-78
76.
Rakoff-Nahoum S, Hao L, Medzhitov R. Role of Toll-like Receptors in
Spontaneous Commensal-Dependent Colitis. Immun. 2006;25:319-329
77.
Stanislawowski M, Wierzbicki PM, Golab A, et al. Decreased Toll-like
receptor-5 (TLR-5) expression in the mucosa of ulcerative colitis patients. J
Physiology Pharmacol. 2009;60:71-75
78.
Sabine MI, Megan EH, Gijs H, et al. TLR5 is not required for flagellinmediated exacerbation of DSS colitis. Inflamm Bowel Dis.16:401-409
79.
Gobert AP, Cheng Y, Akhtar M, et al. Protective Role of Arginase in a
Mouse Model of Colitis. J Immunol. 2004;173:2109-2117
80.
Horowitz S, Binion DG, Nelson VM, et al. Increased arginase activity and
endothelial dysfunction in human inflammatory bowel disease. Am J Physiol
Gastrointest Liver Physiol. 2007;292:G1323-1336
81.
Philip A, Nicholas CZ, Thuan N, et al. Distinct cytokine patterns identified
from multiplex profiles of murine DSS and TNBS-induced colitis. Inflamm Bowel
Dis. 2009;15:341-352

118

This chapter is in preparation for publication in Gut Microbes.
Nagalingam NA, Robinson CJ, Young VB. The effects of intestinal microbial
community structure on disease manifestation in IL10-/- mice infected with
Helicobacter hepaticus.

I performed experiments and processed samples for the cefoperazone study,
and analyzed sequence data for both cefoperazone and vancomycin studies. Dr.
Robinson conducted the vancomycin study and also processed these samples. I
prepared the manuscript with contribution from two other authors.

119

Chapter 4

The effects of intestinal microbial community structure on disease
manifestation in IL-10-/- mice infected with Helicobacter hepaticus.

Abstract and Key Words:
Backgroud: Inflammatory bowel disease (IBD) is an idiopathic disease where
inflammation is associated with several factors, including changes in the
intestinal microbiota. Here, we investigated the role of the microbiota in
Helicobacter hepaticus infected IL-10-/- mice, a commonly used IBD model, in
disease manifestation. We hypothesized that alterations in the microbiota affect
disease severity during disease progression.
Method: We altered the microbiota in mice by infecting animals with H. hepaticus
and treating with vancomycin. We also infected animals with H. hepaticus after
inducing alteration in the microbiota with cefoperazone. Through surveys of the
16S rRNA encoding-gene, analyses of histology and changes in host mediator
expression, we measured changes in the microbiota and host response,
Results: We found that severity of disease is independent of antibiotic-induced
changes in the microbial community structure. However, we also documented
differences in expression of host inflammatory mediators associated with shifts in
the microbiota, though these differences led to the same pathology.
120

Conclusion: In conclusion, we demonstrated that disease is initiated and
progresses in the presence of several different microbial communities. Also, H.
hepaticus is the main driver of inflammation in this model, with changes in the
microbiota playing a minor role in disease manifestation once inflammation is
triggered. Finally, cefoperazone-induced changes in the microbiota may have
long-term effects on the expression of host mediators.
Key words: microbiota, IBD, Helicobacter hepaticus, IL-10-/- mice, 16S rRNA

121

Introduction:
Inflammatory bowel disease (IBD) is most often represented by Crohn’s disease
(CD) and ulcerative colitis (UC), but is a collective term referring to an idiopathic
condition that results in inflammation of the intestinal mucosa. Although incidence
of this disease is increasing (1, 2), its cause is yet to be determined. However,
advances in the field point to genetic factors, dysregulated immune responses
and altered intestinal microbial communities as playing key roles in the onset of
disease (3).
Dysbiosis, or alteration of microbial community structure, has been associated
with intestinal diseases, both in patients and animal models of IBD (4-8). For
example, fecal and biopsy samples from IBD patients have different microbial
communities than those from healthy controls (9). Furthermore, patients with CD
contain communities that are distinct from those in UC patients (9). Animal
models have been key in elucidating the importance of not only the interaction
between the immune system and the microbiota, but also the structure of the
microbiota in IBD. Mice that are deficient in both Tbet, a transcriptional factor, and
Rag2, recombination activating gene, contain colitogenic communities that can be

transferred into healthy recipients and subsequently result in disease (10).
Additionally, studies utilizing animals raised in germfree or gnotobiotic conditions
provide further evidence for the necessity of the resident community and the
significance of community structure in enteric diseases (11-13). The results of
these studies suggest that changes in the balance among members of the
indigenous microbial community can trigger an inflammatory response.
122

To further explore the role of the microbiota in IBD, several animal models are
available used to monitor changes in these communities and their effects on
disease, in a controlled environment (14, 15). A common model of IBD employs
the infection of Helicobacter hepaticus in interleukin 10 (IL-10) deficient mice. In
this model, infection elicits a Th1/Th17 response that most resembles Crohn’s
disease in humans. Infection with H. hepaticus does not cause disease in wildtype mice and only triggers inflammation in immune altered hosts (16, 17). Onset
of disease is MyD88 dependent (18). Additionally, H. hepaticus only causes
disease in the presence of the resident microbiota with no disease observed in
germfree mice (19), indicating the importance of the microbiota in the onset of
IBD in this model. However, apart from these findings there is limited knowledge
of the role that the resident microbiota plays in this model. Therefore, the aim of
this study is to investigate the role of the microbiota in this model of IBD by
altering community structure and observing disease manifestation.
Previously we showed that treatment with cefoperazone significantly alters the
microbial community of the murine gut (20), as does vancomycin (21). We used
these antibiotics to modify the gut community structure in IL10-/- C57BL/6 mice
before and during infection with H. hepaticus, and then monitored disease. We
hypothesized that the structure of the microbiota is related to disease severity in
IL-10 deficient mice infected with H. hepaticus.

123

Materials and methods:
Animals: C57BL/6 IL-10-/- mice from a breeding colony initially established with
breeding stock from Jackson Laboratories (Bar Harbor, ME) were use for all
experiments. Mice were housed with autoclaved bedding, given sterile food and
water ad libitum, and exposed to 12:12 h light:dark cycles. Germfree mice were
housed in sterile microisolators. All animal studies were conducted at the
University of Michigan and were approved by the University Committee on Use
and Care of Animals (UCUCA).
Murine infection with H. hepaticus. The wild-type H. hepaticus strain 3B1 (type
strain, ATCC 51488) was incubated at 37!°C under microaerobic conditions for
48 h, on Tryptic Soy Agar plates (TSA), supplemented with 5% sheep’s blood,
and then re-suspended in 5 ml of tryptic soy broth (TSB). Mice were inoculated
with a suspension of bacteria with an optical density of 1.0 at 600 nm (~108 CFU)
in a volume of 0.2 to 0.3 ml. Bacteria were introduced directly into the stomach
with a 24-gauge ball-tipped gavage needle. Control mice were inoculated with
sterile TSB.
Three experimental studies were conducted to investigate roles of the microbiota
in disease development: 1) To investigate whether disease can develop in H.
hepaticus monoassociated IL-10 deficient C57BL/6 mice in our colonies, two
groups of germfree mice were used: five mice were gavaged with TSB only and
five were gavaged with H. hepaticus. Tissues were collected four weeks postinfection. 2) To investigate roles of an altered microbiota in disease development
in this model, we changed microbial communities using two different antibiotic
124

treatments. Cefoperazone was administered, ad libitum, as a 0.5 mg/ml solution
in the drinking water of 10 weeks old, conventionally raised C57Bl/6 IL-10
deficient mice. Mice were assigned to six groups, as shown in figure 4.1A: one
group was infected with H. hepaticus after 1 day recovery from cefoperazone,
another group infected after 6 weeks recovery from cefoperazone, and a third
group infected without prior cefoperazone treatment. Three other groups were
used as the uninfected controls for respective cefoperazone treatments 3) The
second antibiotic treatment involved ad libitum administration of 0.1 mg/ml
vancomycin in the drinking water of 6-8 week old, conventionally-raised C57Bl/6
IL-10 deficient mice (21). Mice were assigned to six experimental groups as
depicted in figure 4.1B: One group of mice was given vancomycin then infected
with H. hepaticus as antibiotic administration continued. A second group had
similar treatment but vancomycin treatment was ceased 10 days post infection as
mice recovered fro 11 days from antibiotic treatment. A third group of mice was
used as vencomycin untreated controls. Three H. hepaticus uninfected control
groups were used for respective antibiotic treatments. Mice were grouped into
cages according to gender and treatment.

125

A

C57BL/6 IL10-/- mice
(n=36)
Cef 10 days
(n=12)

Control
(n=14)
Hh uninfected
4 weeks
(n=7)

Hh 4 weeks
(n=7)

Hh uninfected
4 weeks
(n=7)

Cef 10 days
(n=10)

1 day off drug

6 weeks off drug

Hh
4 weeks
(n=5)

Hh uninfected 4
weeks
(n=5)

Hh
4 weeks
(n=5)

Harvest cecum for
microbial ecology
analyses
B
C57BL/6 IL10-/- mice
(n=48)

Vancomycin
14 days (n=34)

Control
(n=14)

Hh uninfected
10 days
(n=6)
Hh uninfected
10 days
(n=6)

Hh10
days
(n=8)

Hh uninfected
10 days
(n=6)

Hh
10 days
(n=10)

Hh
10 days
(n=12)
11 days
off drug

Harvest cecum for
microbial ecology
analyses

Figure 4.1 Schematic of experimental designs showing (A) cefoperazone (Cef)
and (B) vancomycin administration to IL10-/- C57BL/6 mice prior to or during
infection with H. hepaticus (Hh).

126

Necropsy and Histology: Mice were euthanized by CO2 asphyxiation and
tissues harvested (22). The cecum sections were gently washed with 1X
phosphate-buffered saline (PBS) to remove luminal contents, cut into sections
and snap-frozen in liquid nitrogen and stored at -80oC until future use.
The remaining portion of the cecum was processed for histology by removing the
luminal contents before washing with PBS, and placing the tissue in tissue
cassettes. Cassettes were then submerged in 10% formalin for 24 hours and
transferred to a 60% ethanol solution prior to processing for paraffin embedding
and staining with hematoxylin and eosin (H&E).
Histologic sections were scored by a veterinary pathologist who was blind to the
sample source. The following scoring system was used: 0, no significant
inflammation; 1, small multifocal lamina proprial or transepithelial leukocyte
aggregates; 2, coalescing lamina proprial leukocyte aggregates, may have
submucosal involvement; 3, frequent coalescing leukocyte aggregates with
submucosal involvement, may have follicle formation; 3.5, strong submucosal
component, infrequent extension to muscularis externa or mesocolon; 4, diffuse
or regionally extensive transmural involvement.
DNA extraction: Genomic DNA was extracted from tissue using the MagNa
Pur® system (Roche, Indianapolis, IN). Briefly, the tissue was incubated in lysis
buffer and proteinase K at 56 oC, before being placed in the MagNa Pur®
Compact machine for purification of DNA. The resulting DNA was then used in
preparation of the clone libraries and quantitative PCR (qPCR).
127

Clone library construction: PCRs were set-up as described in Young et al.
(23). Briefly, illustra PuReTaqTM Ready To Go PCRTM beads (GE Healthcare,
Piscataway,

NJ).

and

AGAGTTTGATCCTGGCTCAG-3';

broad-ranged
1492R,

primers,

(8F,

5'-

5'-GGTTACCTTGTTACGACTT-3')

were used to amplify 16S rRNA genes. GE illustraTM MicrospinTM S-400 HR
columns were used to purify amplicons (GFX, GE Healthcare, Piscataway, NJ)
as directed by the manufacturer. Purified PCR products were ligated into the
TOPO

4®

vector

(Invitrogen

K4575-01,

Carlsbad,

CA)

according

to

manufacturer’s specifications, and transformed into Escherichia coli TOP10 cells.
Clones were grown overnight at 37 oC in Luria Broth (LB) amended with
carbenicillin

(50

µg/ml)).

Vector

CAGTCACGACGTTGTAAAACGACGGC-3';

specific
and

primers

(M13F,
M13R,

5'5'-

CAGGAAACAGCTATGACCATG-3') were used to screen for clones containing
the appropriately sized inserts and to amplify inserts for sequencing. Partial 16S
rRNA gene sequences were generated by sequencing with primer 8F, at the
Sequencing Core at the University of Michigan. Raw sequence data were
processed through an automated “information pipeline” available through the
Ribosomal Database Project (RDP) Web site (http://rdp.cme.msu.edu/). Distance
matrices containing data from each library were calculated from alignments
generated by the RDP aligner (24). The program mothur (25) was used to assign
sequences to operational taxonomic units (OTUs) based on values in the
distance matrices.

Analyses were done using a 95% sequence similarity to

denote genus level and at a 80% sequence similarity to approximate phylum

128

level (25). EstimateS (26) and mothur were used to calculate ecological
measures from the distance matrices. The Morisita Horn index was then used to
measure the similarity among samples taking into account both abundance and
richness of the community. Phylogenetic trees were calculated using the Mega3
program and the parsimony test from the mothur suite of programs was used to
denote significant differences between treatment groups (25, 27). Additionally,
taxonomic assignments were determined using Classifier through the RDP
website (80% confidence cut-off).
qPCR analysis of microbial communities: The quantity of 16S rRNA operons
in the samples relative to a single-copy host gene was measured using a
primer/probe set that targets a broad range of rRNA-encoding gene sequences
(forward

primer

(5’-TCCTACGGGAGGCAGCAGT-3’);

GGACTACCAGGGTATCTAATCCTGTT-3’);

and

reverse
probe

primer

(5’-

(5’-[6-FAM]-

CGTATTACCGCGGCTGCTGGCAC-[TAMRA]-3’)) (28). A primer/probe set
targeting a 264-bp portion of the TNF alpha gene was used as a reference using
200

nanomoles

of

the

forward

(TNFα_mu_se;

5’-

GGCTTTCCGAATTCACTGGAG-3’) and reverse primers (TNFα_mu_as; 5’CCCCGGCCTTCCAAATAAA-3’),
(TNFα_mu_probe;

and

100

nanomoles

of

the

probe

5’-[Cy5]-ATGTCCATTCCTGAGTTCTGCAAAGGGA-[Iowa

Black RQ™]-3’) (29). The reaction mix consisted of LightCycler® 480 Probes
Master reaction mix (Roche) at 1X concentration, and appropriate primer/probe
pair. Amplification of each gene was done under separate run conditions: the
16S rRNA gene target had an activation step of 50 oC for 2 min followed by 95 oC
129

for 10 min. Forty-five cycles were done at 95 oC for 15 s and 60 oC for 1 min
before holding at 4 oC. The TNF reference gene was also amplified with an
activation step of 50 oC for 2 min followed by 95 oC for 10 min. Forty-fives cycles
of 95 oC for 20 s and 64 oC for 30 s were done before holding at 4 oC.
Calculations of 2-ΔΔCt were made to compare changes in the amount of 16S from
samples between treatment groups (21,22).
RNA extraction and determination for cytokine expression: Total nucleic
acids were extracted by MagNaPur® (Roche) as directed by the manufacturer’s
protocol. Total RNA was then isolated by treating total nucleic acids with DNaseI
recombinant, RNase-free (Roche, Indianapolis, IN). The High-Capacity cDNA
Reverse Transcription Kit (Applied Biosystems, Foster City, CA) was used to
transcribe total RNA to cDNA. Cytokine expression was determined by applying
the cDNA to a custom Superarray® (Frederick, MD) PCR plate which contained
gene-specific primers designed for the following targets (30): IFNγ, IL-12a, TNFα,
IL-4, IL-13, IL-5, TGFβ, IL-10, IL-6, FOXp3, IL-17a, IL-23a, CXCL2, CCL5,
NOS2, Arg1, Chi3/4, PTGS2, INDO, GAPDH, βActin, a reverse-transcription
control and a genomic DNA control. Quantization was done with LightCycler®
480 SYBR Green I Master and analyzed on a LightCycler® 480 system (Roche
Diagnostics, Indianapolis, IN) as directed by manufacture’s protocol, with the
following program: 95 oC activation for 10 min; 40 cycles of 95 oC denaturation
for 15s, and 60 oC annealing for 1 min. Resulting threshold values were analyzed
by calculating the 2-ΔΔCt values, using GAPDH as the reference, to calculate fold

130

regulation compared to the antibiotic untreated, H. hepaticus uninfected control
(31-33).
Statistical Analysis: Statistical significances were determined using KruskalWallis tests for the histological scores, and ANOVA tables for the microbial
abundances among the treatment groups using the Metastats, software for
metagenomic analysis (34). The differences in the dendrograms were calculated
using the parsimony test function in mothur (25). Probability values less than
0.05 were considered significantly different.

Results:
Germfree mice infected with H. hepaticus do not get colitis. To investigate
whether H. hepaticus can colonize germfree mice from our colonies, we infected
IL-10 deficient animals and monitored them for 4 weeks. Absence of disease
when mice are monoassociated with H. hepaticus would indicate that the
resident microbiota has an important role in disease manifestation. Detection of
H. hepaticus was done through qPCR, and revealed that the loads in germfree
mice were 2.94 ±1.90 fold less than loads in conventional mice, a level that was
not significantly different (p=0.446) (table 4.1). This indicates that not only can H.
hepaticus colonize germfree mice, as indicated by others (19), but that the
bacterium may have a particular niche that is unaffected by the presence of
resident microbes.

131

Treatment
Cefoperazone treatment
No cef
1 day recovery from cef
1 day recovery from cef
before infection
No cef
1 day recovery from cef
before infection
6 wk recovery from cef
before infection
6 wk recovery from cef
before infection
Vancomycin treatment
No vanc
Vanc
No vanc
Vanc
Vanc + 11 days recovery
Vanc + 11 days recovery

H. hepaticus
infection
No
No

Fold change ± SD
Total bacterial
H. hepaticus
load
calibrator
N/A
-3333.00 ± 0.34
N/A

No

2.16 ± 1.18

N/A

Yes

1.25 ± 0.75

calibrator

Yes

4.48 ± 2.55

-1.12 ± 2.79

Yes

-1.35 ± 0.71

3.94 ± 6.30

No

1.14 ± 0.04

N/A

No
No
Yes
Yes
Yes
No

1.17 ± 0.68
1.95 ± 1.41
1.90 ± 1.91
1.98 ± 1.62
1.14 ± 1.16
1.41 ± 0.91

N/A
N/A
1.04 ± 0.30
-1.59 ± 0.52
1.28 ± 0.56
N/A

N/A

-2.94 ± 1.90

Germfree mice infected with H. hepaticus

Table 4.1. Quantitative PCR measures of fold changes in the total bacterial and
in H. hepaticus loads from genomic DNA extracted from cecum tissue. The fold
change reported (2-ΔΔCt) is normalized to GAPDH and in comparison to the
samples indicated as “calibrator” in the respective columns.

132

Through histological analysis, no inflammation was detected in cecum tissues
from germfree mice that were infected with H. hepaticus (figure 4.2). This result
is in agreement with a previous study (19). However, unlike previous reports (19),
IL-10 deficient mice raised in specific pathogen-free conditions did not develop
inflammation. This indicates that spontaneous disease is only associated with
mice containing specific microbial communities, again emphasizing the important
role of the microbiota in disease development.

133

A

B

E

C

D

F

50 m

**
Inflammation Score

E

Mouse
H. hepaticus

4
3

ns

2
1
0

Germfree Germree
uninfected infected

Conv
Conv
uninfected infected

Figure 4.2 Helicobacter hepaticus triggered inflammation in conventionally raised
IL10-/- C57BL/6 mice compared to germfree counterparts. Representative
sections of mouse cecum from experimental groups are depicted. (A) H.
hepaticus uninfected conventionally raised mice appear normal. (B) H.
hepaticus infected conventional (Conv) section shows edema, hyperplasia
and inflammatory infiltrate. (C) H. hepaticus uninfected germfree control, and
(D) H. hepaticus infected germfree cecum sections are both normal without
signs of inflammation. (E) H. hepaticus forms close association to the
intestinal epithelium, and in the crypts (yellow) compared to other microbes
(red) in the lumen. (F) In germfree mice monoassociated with H. hepaticus,
the bacteria are also found in close association with the epithelial layer (G)
Quantification of indicators of pathology showing significant inflammation
only in H. hepaticus infected conventionally raised mouse sections. Asterisk
(*) denotes statistically significant changes (p<0.05). For interpretation of
the references to color in this and all other figures, the reader is
referred to the electronic version of this dissertation.
134

Vancomycin alters communities in uninfected mice.

Vancomycin was

administered to mice infected with H. hepaticus prior to infection in order to
investigate the effects of altering community structure on the development of
inflammation. Through clone library construction, we found that vancomycin
treated communities were significantly altered and these changes persisted even
after administration ceased. Although bacterial load, via qPCR, was not altered
by vancomycin treatment (table 4.1), this antibiotic caused a decrease in overall
microbial diversity and an increase in relative abundance of members of the
Proteobacteria, namely E.coli/Shigella (figure 4.3). Also, total community
richness was reduced in vancomycin treated mice (Shannon index= 2.10± 0.55)
compared to untreated controls (Shannon index= 3.55± 0.15) (supplemental
figure 4.3). These results indicate that vancomycin alone causes changes in the
microbial community. Communities did not cluster according to gender or
housing.

135

Changes in vancomycin treated communities infected with H. hepaticus

*
vancomycin

*
H. hepaticus

0.4

0.3

0.2

0.1

0.0

Anaeroplasma
Akkermansia
Escherichia/Shigella
Helicobacter
Barnesiella
unclassif Porphyromonadaceae
Clostridium
Dorea
Coprococcus
unclassif Lachnospiraceae

Anaerotruncus
Oscillibacter
Ruminococcus
Butyricicoccus
unclassif Ruminococcacea
Peptococcus
unclassified Peptococcaceae
Anaerovorax
unclassif Clostridiales
Allobaculum
Holdemania
Coprobacillus
Turicibacter
unclassif Erysipelotrichaeae
Lactobacillus
NA

Control
Recovered from vancomycin
Vancomycin
control + H. hepaticus
Recovered from vancmycin + H. hepaticus
Vancomycin + H. hepaticus

Figure 4.3 H. hepaticus infection, as well as vancomycin treatment, significantly
alters intestinal community structure. Asterisk (*) denotes statistically
significantly different community structures calculated by parsimony test
(p<0.05). For interpretation of the references to color in this and all
other figures, the reader is referred to the electronic version of this
dissertation.
136

Vancomycin and H. hepaticus alter communities in infected mice.
Investigations on the effects of H. hepaticus infection on the microbial community
structure, during vancomycin treatment, showed reduction in community richness
(Shannon index= 1.35± 0.35) compared to microbiota in infected, vancomycinuntreated mice (Shannon index= 2.47± 1.01) (supplemental figure 4.3). We also
found that mice that were continuously administered vancomycin and infected
with H. hepaticus (VancHh) contained the same load of H. hepaticus as mice that
were not administered antibiotics (Hh) (p=0.06).

However, VancHh mice

contained fewer H. hepaticus than mice that had been administered vancomycin
previous to H. hepaticus infection, but not during infection (VancRecHh)
(p=0.009). There was no difference between the loads in Hh and VancRecHh
mice (p=0.28).
Structurally, the communities in VancRecHh mice were more like the
communities in mice that were only infected with H. hepaticus and no antibiotic
(Hh) (Morisita-Horn distance= 0.36) than those in VancHh mice (Morisita-Horn
distance= 0.43). As compared to the communities in VancRecHh mice, VancHh
mice contained communities with decreased population sizes of Oscillibacter
(p=0.005), Lachnospiraceae (p=0.003), Coprocococcus (0.011) and Dorea
(p=0.038). Additionally, the VancHh communities contained more E. coli/Shigella
(p=0.001), Akkermansia (p=0.001) and Anaeroplasma (p<0.001). Shannon
diversity index indicates that communities in VancRecHh mice are more diverse
than communities in VancHh mice (p= 0.001). Taxonomic analyses indicated that
communities

in

VancRecHh

also

exhibited
137

a

significant

increase

in

Lachnospiraceae (p=0.001) and decrease in E. coli/Shigella (p=0.005) as
compared to communities in VancHh mice (figure 4.3).
Inflammation occurs in vancomycin treated mice infected with H.
hepaticus.

To investigate whether changes in the microbiota affect disease

manifestation, cecum sections of vancomycin treated mice were scored. All
animals infected with H. hepaticus developed inflammation, regardless of the
shifts in the microbial community structures (figure 4.4).
H. hepaticus infection alters communities in infected mice. Since
vancomycin treatment did not result in differences in disease manifestation in H.
hepaticus infected mice, we administered cefoperazone to animals since it is
known to cause significant, broad-spectrum changes to the intestinal microbiota.
The community 1 day following cefoperazone treatment has over a 1000 fold
lower bacterial load (table 4.1).
After cefoperazone treatment, mice were administered untreated water and
allowed to recover for one day, or for six weeks before infecting with H.
hepaticus. Four weeks post infection, all communities had similar total bacterial
loads compared to the cefoperazone untreated mice that were H. hepaticus
uninfected, indicating that the communities had recovered during the four weeks.
This was also supported by diversity measures (supplemental figure 4.3).
Communities in infected mice contained similar amounts of H. hepaticus
regardless of cefoperazone treatment or recovery period (table 4.1). It is

138

noteworthy that there was an exception, with one infected mouse having 20-fold
less detectable CdtB target, indicating a low H. hepaticus load.

139

*

Inflammation Score

4
3
2
1
0
no

yes

yes

11
va
nc
+

va
nc
+

yes

va
da
nc
ys
re
co
ve
ry

va
da
nc
ys
re
co
ve
ry
No
ne

no

11

Vancomycin treatment

no
No
ne

H. hepaticus infection

Figure 4.4 Blinded scores showing significant inflammation in all H. hepaticus
infected sections regardless of vancomycin treatment compared to
uninfected controls. Asterisk (*) denotes statistically significant changes
(p<0.05).

140

Clone libraries were built from three individual mice from each treatment group
(supplemental 4.1). OTU-based analyses show that the communities in H.
hepaticus-infected mice were significantly different from those in mice that did not
harbor H. hepaticus (p<0.001). In communities of infected mice, the relative
abundance of H. hepaticus varied between 25% and 70% (figure 4.5). This
difference in relative abundance of H. hepaticus was independent of
cefoperazone treatment.
The mouse that contained low numbers of H. hepaticus, as determined via
qPCR, contained a community that produced only 2 of 90 clones that affiliated
with Helicobacter. This supports the finding that H. hepaticus did not dominate
this community as was typically seen in the other mice in this treatment group.
There were reductions in the genera Clostridia (p=0.032) and family
Lachnospiraceae (p=0.016) in communities infected with H. hepaticus after only
one day off cefoperazone compared to those infected after six-weeks recovery.
This indicates differences in recovery dynamics in the two treatment groups.
There were no gender or housing effects on communities’ relatedness.

141

Changes in cefoperazone treated communities infected with H. hepaticus

* H. hepaticus

0.4

0.3

0.2

0.1

0.0

Anaeroplasma
Akkermansia
Escherichia/Shigella
Helicobacter
Barnesiella
unclassif Porphyromonadaceae
Clostridium
Dorea
Coprococcus
unclassif Lachnospiraceae

Anaerotruncus
Oscillibacter
Ruminococcus
Butyricicoccus
unclassif Ruminococcacea
Peptococcus
unclassified Peptococcaceae
Anaerovorax
unclassif Clostridiales
Allobaculum
Holdemania
Coprobacillus
Turicibacter
unclassif Erysipelotrichaeae
Lactobacillus
NA

Control
Recovered from cefoperazone for 1 day
Recovered from cefoperazone for 6wks
control + H. hepaticus
Recovered from cefoprazone 1 day + H. hepaticus
Recovered from cefoprazone 6 wks + H. hepaticus

Figure 4.5 H. hepaticus infection significantly alters the community structure of
mice even after four weeks of termination of cefoperazone treatment.
Asterisk (*) denotes statistically significantly different community structures
calculated by parsimony test (p<0.05). For interpretation of the references
to color in this and all other figures, the reader is referred to the
electronic version of this dissertation.

142

Inflammation occurs in cefoperazone treated mice infected with H.
hepaticus. To investigate whether a major perturbation would affect disease
manifestation in H. hepaticus infected IL-10 deficient mice, cefoperazone was
administered to mice before infecting with the bacterium. Cecum sections were
then scored, and it was found that all animals infected with H. hepaticus
developed inflammation (figure 4.6). There was no correlation between disease
scores and relative abundance of H. hepaticus (Spearman’s ρ = 0.242). Despite
changes to the microbiota, cefoperazone did not alter disease severity in H.
hepaticus infected mice as compared to infected mice that were not administered
antibiotics.

143

*

2
1

no

yes

ce
f
co
v

w

ks

re

v
y
da
1
no

6

re
co

re
c
ks
w

6
no

ce
f

ne
no

ov

ce
f
ov
re
c

1
H. hepaticus infection

ce
f

0

da
y

Cefoperazone treatment
prior to infection

3

no
ne

Inflammation Score

4

yes

yes

Figure 4. 6 Blinded scores showing significant inflammation in all H. hepaticus
infected sections regardless of cefoperazone treatment compared to
uninfected controls. Asterisk (*) denotes statistically significant changes
(p<0.05).

144

Infection with H. hepaticus and/or cefoperazone induces host response.
Although development of inflammation was not affected by cefoperazone
treatment, qPCR showed that antibiotic treated mice that were infected 1 day
after treatment ceased, had a lower load of H. hepaticus than animals that were
allowed to recover from treatment for six weeks prior to infection. To investigate
possible host effects due to this difference, we measure changes in expression of
host mediators. This analysis indicated that changes were elicited by
administration of cefoperazone in communities containing H. hepaticus.
After cefoperazone treatment, there was a significant up-regulation of several
host mediators including TNFα, CCL5, CCL2, Nos2 and Arg1, or down-regulation
in CHI3/4 and IL-23a (figure 4.7, supplemental 4.2). Even when mice were
allowed to recover for 6 weeks from cefoperazone treatment, expression of some
mediators remained significantly different from the no-cefoperazone, H.
hepaticus infected control, indicating that treatment with cefoperazone causes an
altered immune response in the presence of this bacterium. These changes do
not mediate lesions.

145

*

Fold regulation compared to
no cefoperazone, H. hepaticus infected control

40
Cefoperazone 1 day recovery, H. hepaticus
Cefoperazone 6 wk recovery, H. hepaticus

25
20

*

15
*

10

*

5
IL-23a

0
-5

*

*

*

CHI3/4

*

*

-10 IL-12a
-15

*

*

TGFb
TNF

*

*
CCL5

CCL2

-50

NOS2

ARG1

INDO
PTGS2

*

Figure 4.7 Changes in expression of host mediators in cecum tissue of mice
administered cefoperazone and infected with H. hepaticus (p<0.05).

146

Discussion:
Intestinal microbes have been credited with many beneficial functions including
break-down of complex carbohydrates, protection against pathogens and
education of the immune system (35). These are important interactions between
the host and its resident microbes as seen by deficiencies in the host
development in animals raised in germfree conditions (36-38). Collapse of this
delicate balance has been implicated in the onset of diseases, such as
inflammatory bowel disease (IBD), a condition triggered by an inappropriate
response by the hosts’ own immune system. It is also associated with an altered
microbiota, a condition commonly referred to as dysbiosis. Patients with IBD
often exhibit a reduction in diversity of the microbial community structure
compared to healthy participants (39-42). Additionally, some investigators noted
a relative increase in particular bacterial groups such as the Bacteriodetes in
patients with IBD compared to healthy counterparts (43, 44). There is also
evidence that the microbial community is important in the persistence of disease
with antibiotic studies showing amelioration of IBD after treatment (45, 46). Apart
from this we have little insight into the role of the microbiota in IBD. We therefore
used a murine model of IBD, Helicobacter hepaticus infected IL-10 deficient
C57BL/6 mice, to further investigate how changes in the microbial community
associated with disease by testing whether the severity of disease is affected by
different microbial community structures.
Helicobacter hepaticus is sometimes found as part of the resident intestinal
community of mice (47-49), and triggers inflammation in immuno-compromised
147

mice such as Rag-/-, IL-10-/- and SCID (50-52). Therefore, this bacterium is a
pathobiont, a member of the resident bacteria with the potential to cause disease
(53). Furthermore, H. hepaticus-induced inflammation can be ameliorated by the
use of antibiotics (54, 55). However, Helicobacter causes disease in IL-10
deficient mice only in the presence of the resident microbiota indicating an
essential role of the indigenous community towards inducing inflammation in this
model (19). Although other groups have shown the development of spontaneous
disease in IL-10 deficient mice (11, 19), we report less than 5% of mice in our
breeding colony develop disease over a 2 year period (17), suggesting that
disease is associated with certain microbial communities. These data collectively
suggest that the microbiota is an important driver of inflammation in IBD,
particularly in this model, making it an appropriate system in which to study the
effects of an altered microbial community structure on severity of inflammation.
To investigate the extent of the role of the microbiota, we infected mice
containing various communities with H. hepaticus and assessed whether there
were differences in disease severity. Two antibiotics, vancomycin and
cefoperazone, were used to alter the microbial communities. Each drug causes
different changes in the gut community: vancomycin, a drug that targets the
gram-positive bacteria, caused shifts in the microbial structure without reducing
the bacterial load, and cefoperazone, a broad-spectrum antibiotic, reduced the
total bacterial load almost 1000-fold as we have shown here and previously (20,
21). With vancomycin, these shifts were mainly due to the reduction in the
Firmicutes, a phylum that typically comprises more than 50% of the gut
148

community (23, 56). After a recovery period communities exposed to vancomycin
began to resemble control communities, but also exhibited lasting effects with a
reduction of the phylum Firmicutes and an increase in the Proteobacteria.
Increases in the Proteobacteria after vancomycin treatment have also been
observed by other investigators (57), and are perhaps due to the reduction in
gram-positive bacteria, thereby making formerlly occupied niches available to
members of the Proteobacteria.

Although we did not detect them in control

groups in this study, previously we showed that members of the Proteobacteria
inhabit the ceca of IL-10-/- mice using 454 pyrosequencing analysis of the
microbiota (20). It is likely that they were present in the communities analyzed in
the current study, but were below the detection limits of clone library
construction. Like other communities (23, 58) after a recovery period,
cefoperazone treated communities began to resemble the communities in
untreated controls, demonstrating resilience, even after a significant reduction in
bacterial loads immediately after treatment.
Perturbation of the microbiota can enhance colonization by pathogens as seen
with increased loads of Salmonella in mice treated with antibiotics (57, 59-61). In
this study, cefoperazone and vancomycin treatments caused different changes in
the microbiota, and although the relative proportions of the H. hepaticus were
different among the various communities, the bacterium’s loads were comparable
to that in the antibiotic-untreated group. These data suggest that H. hepaticus
may have a specialized niche as its colonization is unaffected by the alterations
in these resident microbiota. One study shows that Salmonella exhibits similar
149

colonization abundance when introduced into mice containing two different
microbial communities, one perturbed by multi-drug cocktail and another that had
recovered (61). The observation suggested to investigators that since the
communities were unlikely to be restricting the Salmonella colonization, it is
plausible that host innate immune system was enforcing the resistance (61).
Future ecological studies can contribute to understanding the mechanism by
which H. hepaticus interacts with the microbiota in induction of disease.
From the histological analyses, H. hepaticus infection caused infiltration by
inflammatory cells indicative of IBD (62). Disease occurrence was independent
of community structure with the only common significant association, between
disease and community change, being the presence of H. hepaticus. Even then,
the relative abundance of this bacterium did not correlate with disease score.
This finding also resonates with the outcome that severe disease occurs even at
very low colonization loads of H. hepaticus, just as it does with relatively high
colonization levels. Many factors contribute to the development of IBD (3),
however these contributions may not be equal. There may be a skew towards
one factor being a predominant contributor to disease development as seen in
monozygotic twin studies where environment plays an important part, while
genetics’ role is to a lesser degree (63); or where fecal transplants or probiotics
lead to the amelioration of disease demonstrating an important role of the
microbiota over other factors (64, 65). Likewise, this model indicates that the
mere presence of H. hepaticus is a strong trigger for inflammation and changes
in the resident community plays a comparatively minor role, if any.
150

The host immune response to H. hepaticus, such as increase in arginase and
Nos2, was consistent with the work of others which revealed increases in iNOS
(66) These enzymes are important in wound healing and are associated with
increases in neutrophil activity, which was observed here. Alterations in the
microbial community by antibiotic treatment have been associated with altered
immune responses (67, 68), and since cefoperazone treatment is known to
significantly reduce bacterial load in the intestine, it was surprising to observe
that final inflammation exhibited was unaffected by antibiotic treatment. We
decided to further investigate whether changes in other host response were
altered by the treatment regime. We discovered that after recovery from
cefoperazone treatment, and H. hepaticus infection, TNFα increased significantly
while INFγ decreased, which may be a compensatory act for the lack of INFγ
response in IL-10 deficient mice (69). Also, the decrease in chitinase (chi3/4)
expression in cefoperazone treated mice may be a response triggered by the
decrease in microbial load after antibiotic treatment. Chitinase is highly
expressed in chemically-induced IBD and has a role in enhanced bacterial
adhesion (70). With lower bacterial loads, chitinase expression may also be
reduced through lack of microbial stimulation. Likewise, the increases in certain
mediators such as TNFα and chemokines CCL5 and CCL2 seen in the
cefoperazone treated mice, compared to untreated mice infected with H.
hepaticus may be due to changes in the microbial stimuli due to shifts in the
microbiota after antibiotic treatment (71). These findings not only demonstrate
that cefoperazone treatment can have long-term effects on host response, but

151

that different expression of host mediators can result in similar disease
manifestation in H. hepaticus infected mice.
Perhaps the most significant finding of this work was that disease can be initiated
in several different microbial communities. Also, after initiation, disease
progressed to exhibit similar manifestations suggesting that microbiota structure
may not be a strong factor in exacerbating or reducing disease in this model. The
actual mechanism whereby the Helicobacter interacts with the microbiota during
inflammation is yet to be determined. It is possible that the resident microbes
prime the host immune system so that a response can be generated to the
Helicobacter, or cross-talk with other microbes are needed for the H. hepaticus to
trigger a host response. The microbiota may contribute to disease in multiple
ways, making studies of potential patterns in the community important in
unraveling the complexities of IBD perhaps leading to novel therapies or the
development of prognostic tools for early detection of this debilitating condition.

Acknowledgements
I would like to extend thanks to Judith Opp and Kristen Frey for technical
assistance, and to laboratory colleagues and Dr. Gary Huffnagle for advice and
discussion of the manuscript. We would also like to especially thank the NIH for
funding.

152

SUPPLEMENTAL INFORMATION
Table 4.2. Clones generated from genomic DNA extracted from cecum samples.
Treatment
Cefoperazone treatment
No cef
1 day recovery from cef
before infection
No cef
1 day recovery from cef
before infection
6 wk recovery from cef
before infection
6 wk recovery from cef
before infection
Vancomycin treatment
No
Vanc
No
Vanc
Vanc + 11 days recovery
Vanc + 11 days recovery

H. hepaticus
infection
No
No

Clones/sample (N)

Yes
Yes

83, 92, 86 (3)
150, 163, 76 (3)

Yes

93, 72, 58 (3)

93, 54, 93 (3)
90, 75, 92 (3)

No

No
No
Yes
Yes
Yes
No

153

95, 95, 95 (3)
94, 84, 94 (3)
89, 89, 92, 93, 94 (5)
94, 95, 93, 90, 92, 94 (6)
94, 91, 92, 93, 91, 90, 96
(7)
95, 93, 92 (3)

Table 4.3. Fold change in expressions of host mediators from cecum tissue. All
values are normalized to GAPDH expression and compared to cefoperazone
untreated, H. hepaticus uninfected control group. Asterisk (*) denotes statistically
significant changes (p<0.05).
1 day recovery from cef,
6 wk recovery from cef,
H. hepaticus
H. hepaticus
Host mediator
IFNγ
1.94 ± 1.12
-2.70 ± 1.09
1.50 ± 1.57 *
-3.85 ± 1.53 *
IL-12a
TNFα
1.06 ± 1.03 *
8.84 ± 1.50 *
1.77 ± 1.53
-1.85 ± 0.62
IL-4
-1.37 ± 1.21
-1.03 ± 1.97
IL-13
1.77 ± 1.53
-1.85 ± 0.62
IL-5
TGFβ
-2.86 ± 0.68 *
4.77 ± 3.14 *
1.77 ± 1.53
-1.85 ± 0.62
IL-6
1.77 ± 1.53
-1.85 ± 0.62
FOXp
1.77 ± 1.53
-1.85 ± 0.62
IL-17a
-1.52 ± 1.29 *
-6.67 ± 1.01 *
IL-23a
1.35 ± 1.85
-2.44 ± 0.94
CXCL2
1.59 ± 0.61 *
9.42 ± 0.56 *
CCL5
-1.19 ± 1.18 *
25.00 ± 1.04 *
CCL2
9.36 ± 2.24 *
41.5 ± 1.32 *
NOS2
6.57 ± 3.16 *
4.17 ± 6.28 *
ARG1
-7.69 ± 1.39 *
-50.00 ± 1.05 *
CHI314
2.38 ± 2.29 *
-5.56 ± 1.25 *
PTGS2
6.74 ± 1.82 *
3.26 ± 3.12 *
INDO

154

Rarefaction curves (at 95% sequence similarity) showing decrease
in diversity of microbial communities treated with vancomycin
A

90
80

H. hepaticus uninfected

OTU

70
60
50
40
30
20
10
0

0

B80

50

100

150

200

250

300

H. hepaticus infected

70
60

OTU

50
40
30
20
10
0

0

50 100 150 200 250 300 350 400 450 500
collection effort/sequences

vancomycin control
cefoperzone control

vancomycin
vancomycin recovered

cefoperazone 4 wks recovered
cefoperazone 10 wks recovered

Figure 4. 8 (A) Rarefaction curves show a reduction in richness in
communities treated with vancomycin, with or without the presence of H.
hepaticus. (B) Communities recovering in the presence of H. hepaticus have
lower community richness than communities in mice infected with H. hepaticus
after 6 weeks off cefoperazone, or those in mice that were not administered
cefoperazone. For interpretation of the references to color in this and all
other figures, the reader is referred to the electronic version of this
dissertation.

155

REFERENCES

156

REFERENCES
1.
Lakatos P. Recent trends in the epidemiology of inflammatory bowel
diseases: up or down? World J Gastroenterol. 2006;12:6102-6108
2.
Gismera CS, Aladren B. Inflammatory bowel diseases: a disease (s) of
modern times? Is incidence still increasing? World J Gastroenterol. 2008;4:54915494
3.
Packey C, Sartor R. Interplay of commensal and pathogenic bacteria,
genetic mutations, and immunoregulatory defects in the pathogenesis of
inflammatory bowel diseases. J Intern Med. 2008;263:597-606
4.
Chang JY, Antonopoulos DA, Kalra A, et al. Decreased Diversity of the
Fecal Microbiome in Recurrent Clostridium difficile-Associated Diarrhea. J Infect
Dis. 2008;197:435
5.
Swidsinski A, Weber J, Loening-Baucke V, et al. Spatial Organization and
Composition of the Mucosal Flora in Patients with Inflammatory Bowel Disease. J
Clin Microbiol. 2005;43:3380-3389
6.
Thompson-Chagoyan OC, Maldonado J, Gil A. Aetiology of inflammatory
bowel disease (IBD): Role of intestinal microbiota and gut-associated lymphoid
tissue immune response. Clin Nutri. 2005;24:339-352
7.
Araki Y, Andoh A, Tsujikawa T, et al. Alterations in intestinal microflora,
faecal bile acids and short chain fatty acids in dextran sulphate sodium-induced
experimental acute colitis in rats. Eur J Gastroenterol Hepatol. 2001;13:107-112
8.
Heimesaat M, Fischer A, Siegmund B, et al. Shift towards proinflammatory intestinal bacteria aggravates acute murine colitis via Toll-like
receptors 2 and 4. PLoS ONE. 2007;25:e662
9.
Swidsinski A, Loening-Baucke V, Vaneechoutte M, et al. Active Crohn's
disease and ulcerative colitis can be specifically diagnosed and monitored based
on the biostructure of the fecal flora. Inflamma Bowel Dis. 2008;14:147-161
10.
Garrett WS, Lord G, Punit S, et al. Communicable ulcerative colitis
induced by T-bet deficiency in the innate immune system. Cell. 2007;131:33-45
157

11.
Sellon R, Tonkonogy S, Schultz M, et al. Resident enteric bacteria are
necessary for development of spontaneous colitis and immune system activation
in interleukin-10-deficient mice. Infect Immun. 1998;66:5224-5231
12.
Sadlack B, Merz H, Schorle H et al. Ulcerative colitis-like disease in mice
with a disrupted interleukin-2 gene. Cell. 1993;75:203-261
13.
Kim SC, Tonkonogy SL, Karrasch T et al. Dual-association of gnotobiotic
IL-10-/- mice with 2 nonpathogenic commensal bacteria induces aggressive
pancolitis. Inflamm Bowel Dis. 2007;13:1457-1466
14.
Warren S. Why study animal models of IBD? Inflamm Bowel Dis.
2008;14:S129-S131
15.
Wirtz S, Neufert C, Weigmann B, et al. Chemically induced mouse model
of intestinal inflammation. Nat Protoc. 2007;2:541-546
16.
Keuhl CJ, Wood HD, Marsh TL, et al. Colonization of the Cecal Mucosa by
Helicobacter hepaticus Impacts the Diversity of the Indigenous Microbiota. Infect
Immun. 2005;73:6952-6961
17.
Pratt JS, Sachen KL, Wood HD, et al. Modulation of host immune
responses by the Cytolethal Distending Toxin of Helicobacter hepaticus. Infect
Immun. 2006;74:4406-4503
18.
Asquith MJ, Boulard O, Powrie F et al. Pathogenic and protective roles of
MyD88 in leukocytes and epithelial cells in mouse models of inflammatory bowel
disease. Gastroenterol. 2010;139:519-529
19.
Dieleman L, Arends A, Tonkonogy S et al. Helicobacter hepaticus does
not induce or potentiate colitis in interleukin-10-deficient mice. Infect Immun.
2000;68:5107-5113
20.
Antonopoulos D, Huse S, Morrison H, et al. Reproducible community
dynamics of the gastrointestinal microbiota following antibiotic perturbation. Infect
immun. 2009;77:2367-2375
21.
Robinson C,, Young VB. Antibiotic administration alters the community
structure of the gastrointestinal microbiota. Gut Microbes. 2010;1:1-6
158

22.
Young VB, Knox KA, Pratt JS, et al. In Vitro and In Vivo Characterization
of Helicobacter hepaticus Cytolethal Distending Toxin Mutants. Infect Immun.
2004;72:2521-2527
23.
Young VB, Schimdt TM. Antibiotic-Associated Diarrhea Accompanied by
Large-Scale Alterations in the Composition of the Fecal Microbiota. J Clin
Microbiol. 2004;42:1203-1206
24.
Cole JR, Chai B, Farris RJ, et al. The ribosomal database project (RDP-II):
introducing myRDP space and quality controlled public data. Nucl Acids Res.
2007;35:D169-172
25.
Schloss P, Westcott S, Ryabin T, et al. Introducing mothur: Open Source,
Platform-independent, Community-supported Software for Describing and
Comparing Microbial Communities. Appl Environ Microbiol. 2009;23:7537-7541
26.
Colwell RK. EstimateS: Statistical estimation of species richness and
shared species from samples. User's Guide and application published at:
http://purloclcorg/estimates. 2006
27.
Kumar S, Tamura K, Nei M. MEGA3: integrated software for molecular
evolutionary genetics analysis and sequence alignment. . Brief Bioinform
2004;5:150-163
28.
Martin FE, Nadkarni MA, Jacques NA, et al. Quantitative Microbiological
Study of Human Carious Dentine by Culture and Real-Time PCR: Association of
Anaerobes with Histopathological Changes in Chronic Pulpitis. J Clin Microbiol.
2002;40:1698-1704
29.
Nitsche A, Becker M, Junghahn I, et al. Quantification of human cells in
NOD/SCID mice by duplex real-time polymerase-chain reaction. Haematologica.
2001;86:693-699
30.
Lees CW, Zacharias WJ, Tremelling M, et al. Analysis of Germline GLI1
Variation Implicates Hedgehog Signalling in the Regulation of Intestinal
Inflammatory Pathways. PLoS Med. 2008;5:1761-1775
31.
Pfaffl MW. A new mathematical model for relative quantification in realtime RT-PCR. Nucleic Acids Res. 2001;29:e45
159

32.
Guide to Performing Relative Quantitation of Gene Expression Using
Real-Time Quantitative PCR. Applied Biosystems. 2004
33.
Schmittgen TD, Livak KJ. Analyzing real-time PCR data by the
comparative C(T) method. Nature Protoc. 2008;3:1101-1108
34.
White JR, Nagarajan N, Pop M. Statistical methods for detecting
differentially abundant features in clinical metagenomic samples. PLoS Comput
Biol. 2009;5:e1000352
35.
Maslowski KM, Vieira AT, Ng A, et al. Regulation of inflammatory
responses by gut microbiota and chemoattractant receptor GPR43. Nature.
2009;461:1282-1286
36.
Wostmann B. The germfree animal in nutritional studies. Ann Rev Nutri.
1981;1:257-279
37.
Smith K, McCoy KD, Macpherson AJ. Use of axenic animals in studying
the adaptation of mammals to their commensal intestinal microbiota. Semin
Immunol. 2006;19:59-69
38.
Round J, Mazmanian S. The gut microbiota shapes intestinal immune
responses during health and disease. Nat Rev Immunol. 2009;9:313-323
39.
Ott SJ, Musfeldt M, Wenderoth DF, et al. Reduction in diversity of the
colonic mucosa associated bacterial microflora in patients with active
inflammatory bowel disease. Gut. 2004;53:685-693
40.
Nishikawa J, Kudo T, Sakata S, et al. Diversity of mucosa-associated
microbiota in active and inactive ulcerative colitis. Scandinavian J Gastroenterol.
2009;44:180-186
41.
Manichanh C, Rigottier-Gois L, Bonnaud E, et al. Reduced diversity of
faecal microbiota in Crohn's disease revealed by a metagenomic approach. Gut.
2006:205-211
42.
Scanlan PD, Shanahan F, O'Mahony C, et al. Culture-Independent
Analyses of Temporal Variation of the Dominant Fecal Microbiota and Targeted
Bacterial Subgroups in Crohn's Disease. J Clin Microbiol. 2006;44:3980-3988
160

43.
Rehman A, Lepage P, Nolte A, et al. Transcriptional activity of dominant
gut mucosal microbiota in chronic inflammatory bowel diseases. J Med Microbiol.
2010;59:1114-1122
44.
Takaishi H, Matsuki T, Nakazawa A, et al. Imbalance in intestinal
microflora constitution could be involved in the pathogenesis of inflammatory
bowel disease. Int J Med Microbiol. 2008;298:463-472
45.
Muniyappa P, Gulati R, Mohr F, et al. Use and safety of rifaximin in
children with inflammatory bowel disease. J Pediatr Gastroenterol Nutr.
2009;49:400-404
46.
Dickinson RJ, O'Connor HJ, Pinder I, et al. Double blind controlled trial of
oral vancomycin as adjunctive treatment in acute exacerbations of idiopathic
colitis. Gut. 1985;26:1380-1384
47.
Nilsson H-O, Ouis I-S, Stenram U, et al. High Prevalence of Helicobacter
Species Detected in Laboratory Mouse Strains by Multiplex PCR-Denaturing
Gradient Gel Electrophoresis and Pyrosequencing. J Clin Microbiol.
2004;42:3781-3788
48.
Li Z, Stephen JD, Martin G, et al. Natural Colonization with Helicobacter
species and the Development of Inflammatory Bowel Disease in Interleukin-10deficient Mice. Helicobacter. 2005;10:223-230
49.
Taylor NS, Xu S, Nambiar P, et al. Enterohepatic Helicobacter Species
Are Prevalent in Mice from Commercial and Academic Institutions in Asia,
Europe, and North America. J Clin Microbiol. 2007;45:2166-2172
50.
Mazmanian S, Round J, Kasper D. A microbial symbiosis factor prevents
intestinal inflammatory disease. Nature. 2008;453:620-625
51.
Cahill R, Foltz C, Fox J, et al. Inflammatory bowel disease: an immunitymediated condition triggered by bacterial infection with Helicobacter hepaticus.
Infect Immun. 1997;65:3126-3131
52.
Li X, Fox JG, Whary MT, et al. SCID/NCr Mice Naturally Infected with
Helicobacter hepaticus Develop Progressive Hepatitis, Proliferative Typhlitis, and
Colitis. Infect Immun. 1998;66:5477-5484
161

53.
Chow J, Mazmanian S. A Pathobiont of the Microbiota Balances Host
Colonization and Intestinal Inflammation. Cell host Microbe. 2010;7:265-276
54.
Foltz CJ, Fox JG, Yan L, et al. Evaluation of various oral antimicrobial
formulations for eradication of Helicobacter hepaticus. Lab Anim Sci.
1996;46:193-197
55.
Madsen KL, Doyle JS, Tavernini MM, et al. Antibiotic therapy attenuates
colitis in interleukin 10 gene-deficient mice. Gastroenterol. 2000;118:1094-1105
56.
Eckburg PB, Bik EM, Bernstein CN, et al. Diversity of the Human Intestinal
Microbial Flora. Science. 2005;308:1635-1638
57.
Sekirov I, Tam NM, Jogova M, et al. Antibiotic-Induced Perturbations of
the Intestinal Microbiota Alter Host Susceptibility to Enteric Infection. Infect
Immun. 2008;76:4726-4736
58.
De La Cochetiere MF, Durand T, Lepage P, et al. Resilience of the
Dominant Human Fecal Microbiota upon Short-Course Antibiotic Challenge. J
Clin Microbiol. 2005;43:5588-5592
59.
Garner C, Antonopoulos D, Wagner B, et al. Perturbation of the small
intestine microbial ecology by streptomycin alters pathology in a Salmonella
enterica serovar typhimurium murine model of infection. Infect Immun.
2009;77:2691-2702
60.
Stecher B, Macpherson AJ, Hapfelmeier S, et al. Comparison of
Salmonella enterica Serovar Typhimurium Colitis in Germfree Mice and Mice
Pretreated with Streptomycin. Infect Immun. 2005;73:3228-3241
61.
Croswell A, Amir E, Teggatz P, et al. Prolonged impact of antibiotics on
intestinal microbial ecology and susceptibility to enteric Salmonella infection.
Infect Immun. 2009;77:2741-2753
62.
Chin AC, Parkos CA. Neutrophil transepithelial migration and epithelial
barrier function in IBD: potential targets for inhibiting neutrophil trafficking. Ann
NY Acad Sci. 2006;1072:276-287

162

63.
Spehlmann ME, Begun AZ, Burghardt J, et al. Epidemiology of
inflammatory bowel disease in a German twin cohort: results of a nationwide
study. Inflamm Bowel Dis. 2008;14:968-976
64.
Borody TJ, Warren EF, Leis S, et al. Treatment of ulcerative colitis using
fecal bacteriotherapy. J Clin Gastroenterol. 2003;37:42-47
65.
Haskey N, Dahl WJ. Synbiotic therapy: a promising new adjunctive
therapy for ulcerative colitis. Nutr Rev. 2006;64:132-138
66.
Erdman SE, Rao VP, Poutahidis T, et al. Nitric oxide and TNF-alpha
trigger colonic inflammation and carcinogenesis in Helicobacter hepaticusinfected, Rag2-deficient mice. PNAS. 2009;106:1027-1032
67.
Brandl K, Plitas G, Mihu C, et al. Vancomycin-resistant enterococci exploit
antibiotic-induced innate immune deficits. Nature. 2008;455:804-807
68.
Ivanov II, Frutos RL, Manel N, et al. Specific microbiota direct the
differentiation of IL-17-producing T-helper cells in the mucosa of the small
intestine. Cell host Microbe. 2008;4:337-349
69.
Kullberg MC, Rothfuchs AG, Jankovic D, et al. Helicobacter hepaticusInduced Colitis in Interleukin-10-Deficient Mice: Cytokine Requirements for the
Induction and Maintenance of Intestinal Inflammation. Infect Immun.
2001;69:4232-4241
70.
Mizoguchi E. Chitinase 3-like-1 exacerbates intestinal inflammation by
enhancing bacterial adhesion and invasion in colonic epithelial cells.
Gastroenterol. 2006;130:398-411
71.
Wlodarska M, Finlay BB. Host immune response to antibiotic perturbation
of the microbiota. Mucosal Immunol. 2010;3:100-103

163

Chapter 5

Summary, synthesis and future directions

Understanding the changes that occur during onset and progression of disease
can also allow for predictive diagnoses from biopsies or fecal samples and help
in managing disease. Since it is difficult to investigate changes in IBD in patients
due to ethical considerations and inability to predict onset of disease, we must
rely on animal models. However, studies on changes in the microbiota
associated with IBD in murine models are very sparse. The purpose of the work
present here was to investigate the role of the microbiota at various stages of
development of IBD through use of mouse models of disease.
In order to investigate the extent to which the microbiota is involved in disease
development, two models were used to address two different questions: First,
DSS-induced colitis is commonly used to investigate onset of IBD, however,
information of microbiota changes that may occur was limited (1, 2). Previously,
there was one study, using non-culture based methods, that documented an
altered microbiota during DSS-induced colitis (2), however, the role of the
microbiota at early stages of disease were unknown. We therefore conducted a
temporal study to document changes that occur during early and late stages of
disease. The current study demonstrates that community changes are not only
associated with later stages of disease, but these changes also appear very
164

early. Alterations in the microbial community occurred as soon as three days post
DSS administration, just as host inflammatory responses were also observed.
This altered structure persisted to 14 days post DSS administration as
inflammatory changes progressed and became more severe. Microbial changes
in community structure were associated with changes in host response, linking
disease with shifts in microbial community structure at various stages of disease
progression. These results indicate that microbiota changes are indeed
associated with DSS-induced colitis, and these changes may also be associated
with maintenance of inflammation.
In the second model, it is known that H. hepaticus alters the microbiota as it
becomes a predominant member of the microbial community (3), thus, the model
inherently involves changes in the microbiota that triggers disease. However, we
did not know whether changes in the resident community would exacerbate or
reduce disease once initiated, therefore, the role of the microbiota in this system
was further probed by finding by inducing shifts in the resident community by
antibiotic treatments. We found that shifts in the resident microbiota, or changes
in prevalence of H. hepaticus in the community, did not influence severity of
disease. We thus demonstrated that the role of the resident microbiota might be
minor once disease is initiated in this model. Also, we investigated whether the
microbiota was important in disease induction in mice in our colonies through
infection of germfree mice with H. hepaticus. Results revealed that the cecum
sections in mice monoassociated with H. hepaticus were similar to uninfected
controls indicating that the microbiota was essential in initiating disease. Here, it
165

appeared that the microbiota is essential for initiating disease but not in
exacerbating or reducing inflammation during disease progression.
It is thus apparent that there are different roles for the microbiota in different
stages of IBD, and these roles may be dependent on other factors associated
with IBD such as environmental triggers as in the case of H. hepaticus infection.
Elucidating these roles of the microbiota associated to IBD will be important in
developing methods for prognosis and/or treatment of human disease.

Future Directions
From our work and that of others, we know that the microbiota can be essential in
initiation of disease as seen in H. hepaticus triggered disease in IL-10 deficient
mice, but play a lesser role in exacerbation/reduction of disease as demonstrated
here. We also show alterations in the microbiota can play a role at the onset and
persistence of disease, as shown by microbiota changes associated with DSSinduced colitis. However, even with these two models, several aspects by which
the microbiota is involved in disease development are yet to be determined (table
5.1). The following attempts to present studies that may fill these gaps in
knowledge.

166

Stages of IBD
Initiation

DSS-induced colitis
Maybe [(4-6)]

H. hepaticus triggered IBD
Yes [Chapter 4, (7)]

Progression

Maybe [Chapter 3]

Unknown

Exacerbation/
Reduction

Yes [(8)]

No [Chapter 4]

Table 5.1 Roles of the microbiota in at various stages of inflammatory bowel
disease.

167

The role of the microbiota in initiation of IBD
In H. hepaticus triggered IBD, there is a clear role for the microbiota in initiation
of disease since germfree mice monoassociated with this bacterium do not
develop inflammation. However, the role microbes play in DSS-induced colitis is
unclear. In some studies, DSS administration resulted in increased inflammation
in germfree mice (5), indicating a protective function of the microbiota in onset of
disease, while others show more severe disease phenotype in conventional
animals compared to germfree counterparts (4), indicating a detrimental role of
the microbiota leading to onset of disease. Nonetheless, differences in disease
development in germfree and conventional animals demonstrate a role of the
microbiota in affecting DSS-induced disease. However, due to conflicting results,
more studies are necessary in order to understand the role of the microbiota in
this model.
Some of this conflicting results in germfree mice investigations stem from the use
of different strains of mice (9), ages and even diet. To investigate the role of the
microbiota in disease initiation, several strains of germfree mice, at similar ages,
can be given DSS and then monitored for disease development. It will be
important to provide elemental diets (minerals, vitamins, amino acids, fats and
sugars), which are devoid of potential antigenic material such as bacterial
components, plant material and other proteins. If inflammation does not develop
in these germfree mice, then they can be conventionalized before, after and
during DSS administration to observe effects of introduced microbiota on disease
development. If disease manifestations were altered in any way after introduction
168

of the microbiota, this would indicate that microbes play a role in initiating
disease. Similar experimental designs have been used in testing the effects of
probiotics and antibiotics in IBD (8, 10). Alternatively, inflammation may be worse
in germfree mice given DSS suggesting that the microbiota has a protective role
against disease induction. Testing whether live bacteria or heat-killed bacteria
impart these benefits can then be done.
In conventionally raised mice, DSS-induced disease is known to be highly strain
dependent (9, 11, 12). This may be related to the microbiota harbored by various
strains since it is known that the microbial community differs according to many
factors, including the host (13, 14). Therefore, characterizing the microbiota of
various strains that are associated with different disease manifestations may
reveal particular community structures linked to disease. If a particular microbial
community is associated with severe inflammation, transfer of this microbiota into
another strain can be done to test if the community is truly involved in IBD in this
model. Others have performed similar experiments in the T-bet Rag-/- ulcerative
colitis (TRUC) model where the microbiota of inflamed mice were transmitted to
wild type mice that subsequently developed disease (15), indicating that genetic
background can alter the microbiota to a community that can induce IBD.

The role of the microbiota in disease progression of IBD
There has been some evidence that the microbiota in DSS-induced colitis is
important in sustaining inflammation since antibiotics have been successful in

169

ameliorating disease (8). However, the extent to which microbes are important in
disease in H. hepaticus triggered infection is unknown.
In H. hepaticus triggered disease, we know that H. hepaticus is important in
maintaining inflammation since eradication of H. hepaticus ameliorates disease.
Here, we also showed that changes in the community do not affect disease
manifestation. However, whether the resident community is at all necessary, after
disease is already induced, is unknown. Cefoperazone (0.5 mg/ml) can provide a
means to test whether the microbiota is important in potentiating disease since it
significantly reduces the intestinal bacteria more than 1000 fold (16).
Unfortunately, this antibiotic concentration also eradicates H. hepaticus (see
appendices 1 and 2). To circumvent this issue, heat-killed Helicobacter may be
useful.
Stimulation from Helicobacter is necessary in this disease model, but whether live
bacteria are needed has yet to be determined. By administering heat-killed
Helicobacter to conventional mice can address this question. Subsequently, if
disease is initiated using the heat-killed bacteria, cefoperazone can be
administered to mice in conjunction with killed bacteria to investigate effects of
significant reduction in microbiota during disease progression. With this
experimental design, the resident microbiota can be significantly reduced while
maintaining stimulation by H. hepaticus. Amelioration of disease by cefoperazone
treatment would indicate that the microbiota is essential in reducing or maintaining
disease in this model. If, however, disease progresses unaltered during
cefoperazone treatment, it may be because the microbiota is not essential for
170

disease perpetuation once it has been induced. Testing whether the microbiota is
essential in disease perpetuation would be challenging since removing the
microbiota from a conventional animal is impossible. However, by performing
gnotobiotic experiments and finding out whether a simple community is sufficient
to trigger disease in this model, other experiments may be possible (see
upcoming section).

For instance, if disease can be initiated in gnotobiotic

communities consisting of bacteria that are transient in the intestinal community
such as those present in probiotic mixtures (17), we can investigate whether H.
hepaticus triggered disease perpetuates once the transient microbes have been
cleared.

Testing the importance of microbiota complexity in IBD
It is not known whether a complex microbial community is necessary to induce
disease, or whether a simple community is sufficient. Murine association with the
Altered Schaedler Flora (ASF) can be used as a simple community to investigate
changes in the microbiota that may accompany disease in DSS-induced colitis
and H. hepaticus triggered disease in IL-10 deficient mice. Development of
disease in mice harboring simple communities would indicate that complex
communities are unnecessary; a single microbe may even be enough for
inflammation to occur.
Bacteroides thetaiotaomicron is part of the normal human microbiota (18), and
thought to stimulate intestinal development in germfree mice so it resembles that

171

of conventional animals (6). Due to its non-pathogenic nature, as well as its
benefits in attenuating some characteristics of germfree animals, including
vascularization, architecture and immune modulation (6, 19), this bacterium can
be used to infect germfree mice before DSS treatment, or H. hepaticus infection,
to test whether a simple system of bacterial-host interaction is enough to trigger
disease comparable to that seen in conventionally raised mice. If these
monoassociated mice develop disease that is similar to that observed in their
conventionalized counterparts in each IBD model system, this would imply that
simple microbial interaction or even luminal antigenic stimulation from killed
bacteria, or bacterial components, might be enough to cause inflammation in the
DSS-induced model, as observed in other systems (20). The next step would
involve inoculation of mice with heat-killed bacteria and observing whether
disease occurs in these models.

DSS-induced colitis: Do alterations in the microbiota precede immune
changes?
To attempt to tease apart changes induced by DSS administration, a study can be
conducted where frequent samples are collected between day zero and day three
of DSS administration. Recent evidence indicates that intimate interactions
between the microbiota and host occur within the first 24 hours of DSS
administration (21). Bacteria were present in the normally impenetrable mucosa
layer indicating DSS increases permeability of the mucin matrix allowing close
association with the host within a day, before host inflammatory responses were
172

observed (21). By investigating whether changes in the microbial community
structure occurred within this time would be helpful in tracing the sequence of
events involving host response and microbiota changes after exposure to DSS.
If changes in host response occur before alteration in the microbiota, it is likely
that DSS acts directly on the host in the presence of luminal bacteria and
community shifts are not involved in onset of disease. Subsequent changes in the
microbial community may simply be a reaction to host’s inflammatory response
(22), as it known that components of the immune system can alter the microbiota.
This is exemplified by increases in segmented filamentous bacteria in mice with
deficient expression of intestinal IgA (23). Subsequent reconstitution of IgA in
intestinal tissue resulted in a return to a community resembling that in mice
expressing IgA (23).
Alternatively, if changes in the microbial community structure occur before
changes in the host immune system are detected, this may be an indication that
the altered microbiota plays a role in disease causation. Effects of different
microbial structures on development of disease were tested by antibiotic
treatment (8), but using gnotobiotic models would also be useful in understanding
the role of certain members of the community, as discussed above.
However, due to the almost simultaneous changes in the microbiota and host
immune response, teasing apart the sequence of event may be challenging.
Germfree and gnotobiotic experiments proposed at the beginning of this section

173

may be more prudent approaches to investigating the role of the microbiota in
onset of disease.

Conclusion
Over the past two decades, studies on the microbiota and its influence on
disease development has grown exponentially (24). Due to the complexities
involved in understanding both the role of the hosts’ immune response and
changes in diversity and function of the microbiota, studies that lay ahead will be
challenging. Nonetheless, development of new tools for investigation of these
changes, and continued vigorous research into the field, provide a promising
future for patients affected with diseases such as IBD.

174

REFERENCES

175

REFERENCES
1.
Araki Y, Andoh A, Tsujikawa T, et al. Alterations in intestinal microflora,
faecal bile acids and short chain fatty acids in dextran sulphate sodium-induced
experimental acute colitis in rats. Eur J Gastroenterol Hepatol. 2001;13:107-112
2.
Heimesaat M, Fischer A, Siegmund B, et al. Shift towards proinflammatory intestinal bacteria aggravates acute murine colitis via Toll-like
receptors 2 and 4. PLoS ONE. 2007;25:e662
3.
Keuhl CJ, Wood HD, Marsh TL, et al. Colonization of the Cecal Mucosa by
Helicobacter hepaticus Impacts the Diversity of the Indigenous Microbiota. Infect
Immun. 2005;73:6952-6961
4.
Hudcovic T, Stepankova R, Cebra J, et al. The Role of Microflora in the
Development of Intestinal Inflammation: Acute and Chronic Colitis Induced by
Dextran Sulfate in Germ-Free and Conventionally Reared Immunocompetent and
Immunodeficient Mice. Folia Microbiol. 2001;46:565-572
5.
Kitajima S, Morimoto M, Sagara E, et al. Dextran Sodium Sufate-Induced
Colitis in Germ-Free IQI/Jic Mice. Exp Anim. 2001;50:387-395
6.
Stappenbeck T, Hooper L, Gordon J. Developmental regulation of
intestinal angiogenesis by indigenous microbes via Paneth cells. PNAS.
2002;99:15451-15455
7.
Dieleman L, Arends A, Tonkonogy S, et al. Helicobacter hepaticus does
not induce or potentiate colitis in interleukin-10-deficient mice. Infect Immun.
2000;68:5107-5113
8.
Rath H, Schultz M, Freitag R, et al. Different Subsets of Enteric Bacteria
Induce and Perpetuate Experimental Colitis in Rats and MIce. Infect Immun.
2001;69:2277-2285
9.
Mahler M, Bristol JI, Leiter HE, et al. Differential susceptibility of inbred
mouse strains to dextran sulphate sodium-induced colitis. Am J Physiol.
1998;274:544-551

176

10.
Herias MV, Koninkx JF, Vos JG, et al. Probiotic effects of Lactobacillus
casei on DSS-induced ulcerative colitis in mice. Int J Food Microbiol.
2005;103:143-155
11.
Suzuki R, Kohno H, Sugie S, et al. Strain differences in the susceptibility
to azoxymethane and dextran sodium sulfate-induced colon carcinogenesis in
mice. Carcinogenesis. 2006;27:162-169
12.
Stevceva L, Pavli P, Buffinton G, et al. Dextran sodium sulphate-induced
colitis activity varies with mouse strain but develops in lipopolysaccharideunresponsive mice. J Gastroentorol Hepatol. 1999;14:54-60
13.
Rawls JF, Samuel BS, Gordon JI. Gnotobiotic zebrafish reveal
evolutionarily conserved responses to the gut microbiota. PNAS. 2004;101:45964601
14.
Ley RE, Peterson DA, Gordon JI. Ecological and Evolutionary Forces
Shaping Microbial Diversity in the Human Intestine. Cell. 2006;24:837-848
15.
Garrett WS, Lord G, Punit S, et al. Communicable ulcerative colitis
induced by T-bet deficiency in the innate immune system. Cell. 2007;131:33-45
16.
Antonopoulos D, Huse S, Morrison H, et al. Reproducible community
dynamics of the gastrointestinal microbiota following antibiotic perturbation. Infect
Immun. 2009;77:2367-2375
17.
Gaudier E, Michel C, Segain J-P, et al. The VSL#3 Probiotic Mixture
Modifies Microflora but Does Not Heal Chronic Dextran-Sodium Sulfate-Induces
Colitis or Reinforce the Mucus barrier in Mice. J Nutr. 2005;135:2753-2761
18.
Duerden BI. The isolation and identification of Bacteroides spp. from the
normal human faecal flora. J Med Microbiol. 1980;13:69-78
19.
Peterson DA, McNulty NP, Guruge JL, et al. IgA response to symbiotic
bacteria as a mediator of gut homeostasis. Cell Host Microbe. 2007;2:328-339
20.
Hoffmann M, Kim SC, Sartor RB, et al. Enterococcus faecalis strains
differentially regulate Alix/AIP1 protein expression and ERK 1/2 activation in

177

intestinal epithelial cells in the context of chronic experimental colitis. J Proteome
Res. 2009;8:1183-1192
21.
Png CW, Linden SK, Gilshenan KS, et al. Mucolytic Bacteria With
Increased Prevalence in IBD Mucosa Augment In Vitro Utilization of Mucin by
Other Bacteria. Am J Gastroenterol. 2010;e-pub
22.
Salzman NH, Underwood MA, Bevins CL. Paneth cells, defensins, and the
commensal microbiota: a hypothesis on intimate interplay at the intestinal
mucosa. Semin Immunol. 2007;2007:2
23.
Suzuki K, Meek B, Doi Y, et al. Aberrant expansion of segmented
filamentous bacteria in IgA-deficient gut. PNAS. 2004;101:1981-1986
24.
Sekirov I, Russell SL, Antunes LC, et al. Gut Microbiota in Health and
Disease. Physiol Rev.90:859-904

178

APPENDICES

179

Appendix A

The effects of cefoperazone on Helicobacter hepaticus
colonization.
Purpose
Cefoperazone (0.5 mg/ml) was previously used to alter microbial community
structure in the intestine (1), and is commonly used in selective media to screen
for H. hepaticus growth in vitro (2). We therefore initially used this antibiotic to
alter the microbial community in mice infected with H. hepaticus in order to study
effects of changes in the community during disease. Surprisingly however,
cefoperazone treatment eradicated H. hepaticus in vivo. After reducing the
concentration administered to mice by ten fold, there seemed to be negligible
effects on total bacterial and H. hepaticus loads in vivo. We therefore suspected
that bacterial colonization maybe dependent on concentration of cefoperazone
administered. We thus tested this hypothesis that changes in the microbiota are
associated with cefoperazone concentration administered. Here, effects of
various concentrations of cefoperazone on total bacterial load and H. hepaticus
loads are reported.

Methods and materials
To investigate the effects of cefoperazone treatment on H. hepaticus in vivo
eleven groups of mice were infected with H. hepaticus for ten days before
180

antibiotic was administered. There were three mice in each group. Dosage
ranged from 0.5 mg/ml to 0.05 mg/ml. Previously, 0.05 mg/ml was seen to have
no effect of either H. hepaticus colonization or total bacterial load. Mice were
sacrificed seven days after the start of cefoperazone administration as shown in
figure A1.1.
Quantitative PCR was performed and clone libraries constructed from mice from
each experimental group as described in previous chapters 3 and 4.

181

C57BL/6 mice

gavage with H.
hepaticus

0 mg/ml

0.05 mg/ml

0 mg/ml

0.20 mg/ml

10 days
0.30 mg/ml 0.40 mg/ml 0.50 mg/ml

0.10 mg/ml 0.25 mg/ml 0.35 mg/ml 0.45 mg/ml
7 days
Harvest cecum for microbial ecology analysis

Figure A1.1 Experimental design depicting administration of various
cefoperazone concentrations to H. hepaticus infected mice.

182

Results and conclusions
Quantitative data shows that 0.5 mg/ml to 0.30 mg/ml significantly reduced total
bacterial loads over 1000 fold and eradicated H. hepaticus. At about 0.25 mg/ml,
the antibiotic did not affect bacterial load or presence of H. hepaticus, indicating a
threshold for cefoperazone activity against intestinal bacteria.
Eradication of H. hepaticus was confirmed by allowing the intestinal communities
to recover after cefoperazone administration. It was found that H. hepaticus was
not detectable after four weeks cessation of antibiotic administration.

183

Fold change compared to no cefoperazone treatment

1000
0.50 0.45

0.40

0.35

0.30 0.25

0.20

0.10

0.05

0

cefoperazone (mg/ml)
-1000

H. hepaticus
Total bacteria

-2000
-3000
-4000
-5000
-6000
-7000

Figure A1.2 Cefoperazone concentrations above 0.25 mg/ml cause significant
decreases in total bacterial loads and eradicates H. hepaticus.

184

To investigate these affects more closely, clone libraries were constructed. Clone
libraries also show presence of H. hepaticus in communities in mice treated with
up to 0.25 mg/ml cefoperazone. The community in mice treated with 0.05 mg/ml
cefoperazone appeared to be more similar to communities in untreated mice
(figure A1.2 A). Furthermore, although bacterial load in mice treated with 0.05
mg/ml was comparable to that in mice treated with 0.10 mg/ml cefoperazone,
there was a decrease in richness in the latter community (figure A1.2 B),
indicating that this concentration affects colonization of some members of the
community more than others.
With 0.30 mg/ml and more, decrease in bacterial loads made it very difficult to
construct clone libraries due to poor primary amplification of the bacterial 16S
rRNA gene. This indicates that these higher concentrations affect many
members of the intestinal community than 0.25 mg/ml cefoperazone. Perhaps
the machinery designed to remove cefoperazone from the bacterial cells are
overwhelmed at these concentrations causing the bacteria to be susceptible to
the antibiotic.

185

100%
90%
Clostridiaceae
unclassified_Peptococcaceae
Peptococcaceae
Incertae Sedis XIII
Erysipelotrichaceae
Helicobacteraceae
unclassified_Porphyromonadaceae
Porphyromonadaceae
unclassified_"Ruminococcaceae"
Ruminococcaceae
Anaeroplasmataceae
unclassified_"Lachnospiraceae"
Lachnospiraceae

80%

Prevalence of bacteria

70%
60%
50%
40%
30%
20%
10%
0%

0
0
uninfected

0.05
0.1
0.2
0.25
cefoperazone (mg/ml)

60
0 mg/ml cef, no H. hepaticus
0 mg/ml cef, H. hepaticus
0.20 mg/ml cef, H. hepaticus
0.25 mg/ml cef, H. hepaticus
0.05 mg/ml cef, H. hepaticus
0.10 mg/ml cef, H. hepaticus

50

OTU

40
30
20
10
0

0

20

40

60

80

100

Observed sequences

Figure A1.3. (A) H. hepaticus were present in the microbiota of communities
treated with cefoperazone concentrations below 0.25 mg/ml. Some communities
showed significant changes due to H. hepaticus infection and cefoperazone
treatment. (B) There was also a reduction in intestinal community richness when
concentrations of 0.20 mg/ml cefoperazone and above were administered to
mice. For interpretation of the references to color in this and all other
figures, the reader is referred to the electronic version of this dissertation.

186

At 0.03 mg/ml cefoperazone, H. hepaticus is absent from microbial communities.
Below 0.25 mg/ml, however, H. hepaticus is present, and together with the
finding that total bacterial load is also unaffected at this concentration, it can be
postulated that cefoperazone is not active at lower concentrations. This may be
due to break down of the compound at low concentrations during in vivo passage
since typical signs of cefoperazone treatment, such as dark cecal contents and
distended cecum, were absent from these mice (see photographs in appendix B).
To test this, cefoperazone levels can be measures from the cecal contents by
mass spectrometry.
Another possibility is that at 0.30 mg/ml, H. hepaticus is eradicated from the
intestinal community suggesting that the bacterium is susceptible to this
concentration of cefoperazone. H. hepaticus is known to have efflux pumps that
functions to remove antibiotics from the bacterial cell (3). It is easy to imagine
that by increasing cefoperazone levels, that the pump may not efficiently remove
the molecules before they act.
To test the hypothesis that an efflux pump may be responsible for susceptibility
of H. hepaticus to high concentrations of cefoperazone, mutants with increased
pumping activity can be used to test whether they are resistant to higher
concentrations of cefoperazone treatment. Alternatively, efflux pump inhibitors
may be used, in a dose-dependent manner, to investigate whether there are
effects on the ability of H. hepaticus to grow at concentrations below 0.25 mg/ml
(4). If pumps are indeed important in conferring resistance to cefoperazone in
this bacterium, increasing levels of the inhibitor increase susceptibility.
187

Dosages of 0.10 mg/ml to 0.25 mg/ml were associated with higher relative
abundance of H. hepaticus. More experiments to investigate how H. hepaticus
colonizes a community is needed to understand the low relative abundance of
the bacterium in communities associated with 0.05 mg/ml cefoperazone
administration compared to those administered 0.10, 0.20 and 0.25 mg/ml
cefoperazone. It would be expected that H. hepaticus would be more prevalent at
lower antibiotic concentrations, with communities resembling those of either
communities from mice not administered cefoperazone; or more like those
associated with 0.10 mg/ml. Perhaps at 0.05 mg/ml cefoperazone treatment,
resident microbes shifts so certain members of particular families increase in
abundance. This may explain lowered relative abundances of H. hepaticus while
total loads are unchanged, as well as richness (Shannon index=3.27) of the
microbiota similar to that in cefoperazone untreated mice (Shannon index= 3.37),
compared to the low richness of antibiotic treated mice (Shannon index= 1.59
with 0.10 mg/ml cefoperazone, 1.52 with 0.20 mg/ml cefoperazone and 1.51 with
0.25 mg/ml cefoperazone) at 3% cutoff.

188

REFERENCES

189

REFERENCES
1.
Antonopoulos D, Huse S, Morrison H, et al. Reproducible community
dynamics of the gastrointestinal microbiota following antibiotic perturbation. Infect
Immun. 2009;77:2367-2375
2.
Pratt JS, Sachen KL, Wood HD, et al. Modulation of host immune
responses by the Cytolethal Distending Toxin of Helicobacter hepaticus. Infect
Immun. 2006;74:4406-4503
3.
Belzer C, Stoof J, Breijer S, et al. The Helicobacter hepaticus hefA gene is
involved in resistance to amoxicillin. Helicobacter. 2009;14:72-79
4.
Lomovskaya O, Warren MS, Lee A, et al. Identification and
Characterization of Inhibitors of Multidrug Resistance Efflux Pumps in
Pseudomonas aeruginosa: Novel Agents for Combination Therapy. Antimicrob
Agents Chemother. 2001;45:105-116

190

Appendix B

The effects of cefoperazone administration in H.
hepaticus triggered disease in IL-10 defiecient mice.
Purpose
Others have shown effects of antibiotic treatment on H. hepaticus infected mice
and reported that some are effective in ridding colonies of this bacterium (1). We
showed that cefoperazone, an antibiotic normally administered through
intravenously in humans, was found to eradicate H. hepaticus in wild type mice
(appendix A). However, effect of cefoperazone treatment in H. hepaticus triggered
disease was unknown. We therefore tested the hypothesis that cefoperazone
administration was attenuate disease in IL-10 deficient mice infected with H.
hepaticus.
Methods and materials
Four groups of mice were used in this experiment as indicated in figure A2.1.
After euthanasia, the ceca of the mice were removed and used for genomic DNA
extraction and histologic preparation (see chapters 3 and 4 of this thesis).

191

Figure A2.1 Experimental design showing H. hepaticus infection prior to
cefoperazone treatment.

192

Results and Discussion
Photographs were taken to document changes that occur with different
treatments (figure A2.2). Cefoperazone treatment resulted in distension of the
cecum with dark, fluid luminal content, making tissue fragile. This indicates that
cefoperazone is associated with alterations in the host, which may be related to
changes in the microbiota. Changes in host response mediators were observed
after cefoperazone treatment and documented in experiments described in
chapter 4 of this thesis. Some of these changes in during immune regulation
may affect permeability of the epithelial wall, which may be associated fluid
luminal content and distension (3).
H. hepaticus infection was associated with thickening of the cecal tissue making
it more rigid, with a reduction in luminal content. This is a common sign of
disease as scarring occurs during IBD and was also observed in experiments
described in chapter 4. Cefoperazone administration, following H. hepaticus
infection, appeared similar to tissue from cefoperzone treated, H. hepaticus
uninfected mice. This may be due to effects of cefoperazone-induced alterations
in microbiota. Whether cefoperazone alone can induce host immune changes is
yet to be determined. This can be investigated by administering cefoperazone to
germfree mice.

193

Figure A2.2 Photographs of ceca removed from mice from four treatment groups.
(Top left) Cecum is normal with luminal content of paste-like consistency,
unlike cecum from mice infected with H. hepaticus that have thick tissue (top
right). With cefoperazone administration (bottom panels), the cecum become
distended and tissue fragile, filled with dark luminal fluid. For interpretation
of the references to color in this and all other figures, the reader is
referred to the electronic version of this dissertation.

194

After cefoperazone administration, H. hepaticus infected mice appeared to have
less inflammation. This corresponded to a reduction in total bacterial load and
eradication of H. hepaticus (figure A1.3). This finding is supported by reports that
eradication of H. hepaticus can ameliorate disease in rodents (1). More animals
would have to be used in this experiment to attain more statistical power to dram
convincing conclusions.

195

Figure A2.3 show significant inflammation in mice infected with H. hepaticus.
After administration of cefoperazone, inflammation was attenuated.

196

REFERENCES

197

REFERENCES
1.
Foltz CJ, Fox JG, Yan L, et al. Evaluation of various oral antimicrobial
formulations for eradication of Helicobacter hepaticus. Lab Anim Sci.
1996;46:193-197
2.
Abdo Z, Schuette UME, Bent SJ, et al. Statistical methods for
characterizing diversity of microbial communities by analysis of terminal
restriction fragment length polymorphisms of 16S rRNA genes. Environ Microbiol.
2006;8:929-938
3.
Ménard S, Cerf-Bensussan N, Heyman M. Multiple facets of intestinal
permeability and epithelial handling of dietary antigens. Mucosal Immunol.
2010;3:247-259

198