Intestinal microRNA Response to Citrobacter rodentium in ... · Lastly, I would like to take this...
Transcript of Intestinal microRNA Response to Citrobacter rodentium in ... · Lastly, I would like to take this...
Intestinal microRNA Response to Citrobacter rodentium in the Presence and Absence of
Bifidobacterium bifidum MIMBb75
by
Bijun Wen
A thesis submitted in conformity with the requirements for the degree of Mater of Science
Graduate Department of Nutritional Sciences University of Toronto
© Copyright by Bijun Wen 2015
ii
Intestinal microRNA Response to Citrobacter rodentium in the
Presence and Absence of Bifidobacterium bifidum MIMBb75
Bijun Wen
Master of Science
Graduate Department of Nutritional Sciences
University of Toronto
2015
Abstract
The continuous crosstalk between gut microbiota and the host epithelium is essential for
intestinal homeostasis. Presence of harmful or beneficial bacteria can impinge host gene
expression and affect homeostasis. Yet, the molecular basis underlying host-bacterial crosstalk is
unclear. The objectives of this project were to determine if pathogen infection impacts
microRNA-mediated posttranscriptional regulation of gene expression in mouse intestine and if
probiotic treatment improves pathogen-induced pathologies and microRNA alterations. It was
found that Citrobacter rodentium infection alters murine colonic microRNA signature with
implications in modulating host gene expression involved in the apoptosis pathway, contributing
to the epithelial hyperplasia in response to Citrobacter rodentium pathogenicity. Probiotic
Bifidobacterium bifidum MIMBb75 supplementation did not attenuate intestinal pathology nor
normalize microRNA alterations. Overall, these findings indicate that host-bacterial crosstalk is
potentially mediated by microRNA modulation, and this particular strain of Bifidobacterium
bifidum did not confer apparent benefits in Citrobacter rodentium infection.
iii
Acknowledgments
I would like to express my sincere gratitude to many individuals for their help and support
throughout my Master’s study.
First and foremost, I would like to thank my supervisor, Dr. Elena Comelli, for giving me the
opportunity to work on this intriguing research project. The knowledge and experience that I
have gained from this project are invaluable for a lifetime. All of these would not have been
possible if it were not for Elena’s continuous support. She has immense knowledge that helped
guide me throughout this research study. Elena, I would like to thank you for being such an
understanding, attentive and inspiring supervisor. I am especially grateful for the countless time
you spent providing prompt and insightful feedback on my questions, writings and presentations
even when you had a busy schedule. Although English is not my first language, you have always
been patient listening to my ideas and believed in my ability to present our data whether in your
undergraduate class or conferences. This is really motivating and important for me to build up
confidence for public speaking. I would also like to thank you for encouraging me to stay calm
and positive when I felt nervous and overwhelmed.
I would also like to thank my advisory committee members, Dr. Ward and Dr. Kim. Their advice
and enlightening questions at each committee meeting had provided me with new perspectives
and helped me in advancing my understanding about the project.
This project would not have been possible without the help from my lab members and I really
treasure the friendships that I have made with them. I would like to give a special thanks to Jim
Chen for helping me with the implementation of the intestinal permeability test and teaching me
how to make histology slides step by step. In addition, his kindness in offering suggestions
related to practical issues based on his profound experience in conducting animal studies will not
soon be forgotten. I would also like to thank our postdoctoral fellow, Amel Taibi, for teaching
me everything that I needed to know about microbiology and performing gene expression
experiments. She was such an attentive and caring person offering help whenever needed on her
own accord. I am especially delighted to have made a great friendship with Christopher Villa
over the years and I really appreciate his kind advice and comforting words when I was going
through the ups and downs of a Master’s student. His help with my project, improving my
iv
English, as well as setting up the computer before my seminars and presentations just reinforces
his caring and generous nature. I would also like to thank Monica Ponta, Sofia Sagaidak and
Alex Lee for their help on the project and being good friends when I needed it most. Moreover, I
would like to thank the department staff, Louisa and Emeliana for helping me with
administrative issues.
Lastly, I would like to take this opportunity to thank my family. Thank you, mom and dad, for
being so understanding and supportive throughout my life, especially for giving me all the
essentials required to ensure my successful completion of my education and career goals. The
unconditional love that my parents and my sister have given me has always contributed to my
desire of becoming a better person. I cannot use words to express how grateful I am and how
much I love you all.
v
Table of Contents
Acknowledgments .......................................................................................................................... iii
Table of Contents ............................................................................................................................ v
List of Tables ................................................................................................................................ vii
List of Figures .............................................................................................................................. viii
List of Abbreviations ..................................................................................................................... ix
Chapter 1 Introduction .................................................................................................................... 1
1 Introduction ................................................................................................................................ 1
Chapter 2 Literature review ............................................................................................................ 4
2 Literature Review ....................................................................................................................... 4
2.1 Intestinal Homeostasis ........................................................................................................ 4
2.1.1 Intestinal Epithelium Turnover ............................................................................... 5
2.1.2 Physical and Chemical Barriers .............................................................................. 7
2.1.3 Immune Barrier ....................................................................................................... 7
2.1.4 Gut Microbiota ........................................................................................................ 8
2.2 Foodborne Enterohemorrhagic E. coli .............................................................................. 11
2.2.1 EHEC Infection ..................................................................................................... 11
2.2.2 Citrobacter rodentium .......................................................................................... 12
2.2.3 EHEC in IBD ........................................................................................................ 14
2.2.4 EHEC in CRC ....................................................................................................... 15
2.3 Probiotics .......................................................................................................................... 16
2.3.1 Bifidobacterium bifidum ....................................................................................... 19
2.3.1.1 Adhesive Factors and Interaction with the Host ..................................... 20
2.3.1.2 Health Benefits ....................................................................................... 22
2.4 MicroRNA ........................................................................................................................ 31
vi
2.4.1 Response of Intestinal miRNA to Bacteria ........................................................... 33
2.4.2 MiRNA deregulations in IBD and CRC ............................................................... 42
2.5 Animal Model ................................................................................................................... 44
Chapter 3 Rationale, Hypothesis and Objectives .......................................................................... 46
Chapter 4 Study 1- Citrobacter rodentium Infection Alters Murine Colonic microRNA
Signature .................................................................................................................................. 48
4.1 Abstract .............................................................................................................................. 49
4.2 Introduction ........................................................................................................................ 50
4.3 Materials and Methods ....................................................................................................... 51
4.4 Results ................................................................................................................................ 56
4.5 Discussion .......................................................................................................................... 79
Chapter 5 Study 2- Effects of Bifidobacterium bifidum on Citrobacter rodentium Infection
via microRNA Modulation ...................................................................................................... 85
5.1 Abstract .............................................................................................................................. 86
5.2 Introduction ........................................................................................................................ 87
5.3 Materials and Methods ....................................................................................................... 88
5.4 Results ................................................................................................................................ 92
5.5 Discussion .......................................................................................................................... 99
Chapter 6 General Discussion ..................................................................................................... 103
6.1 Strengths, Limitations and Future Directions .................................................................. 105
6.2 Implications ...................................................................................................................... 108
6.3 Conclusions ...................................................................................................................... 110
References ................................................................................................................................... 111
vii
List of Tables
Chapter 2
Table 2.1 Clinical Health Benefits of B. bifidum .......................................................................... 26
Table 2.2 Impacts of Gastrointestinal Pathogens on Host microRNA Expression ....................... 36
Table 2.3 Impacts of Probiotics on Host Intestinal microRNA Expression ................................. 40
Chapter 4
Table 4.1 Differentially Expressed miRNAs ................................................................................ 60
Table 4.2 GO Biological Process (total number of genes: 824; total number process hits: 2158).
....................................................................................................................................................... 63
Table 4.3 GO Molecular Function (total number of genes: 824; total number process hits: 1062).
....................................................................................................................................................... 64
Table 4.4 GO Cellular Component (total number of genes: 824; total number process hits: 487).
....................................................................................................................................................... 65
Table 4.5 Panther Pathways (total number of genes: 824; total number process hits: 895). ........ 66
viii
List of Figures
Chapter 2
Figure 2.1 Colonic Cell Turnover and Transmissible Murine Hyperplasia .................................... 6
Figure 2.2 Models of miRNA-dependent Regulatory Network. ................................................... 32
Chapter 4
Figure 4.1 Study Design ............................................................................................................... 52
Figure 4.2 C. rodentium infection kinetics and effect on mouse body and organ weights. .......... 57
Figure 4.3 C. rodentium induced intestinal lesions, crypt hyperplasia and barrier dysfunction. .. 58
Figure 4.4 Loss of barrier integrity of C. rodentium infected mice on day 10 p.i. ....................... 59
Figure 4.5 C. rodentium infected mice exhibit distinct miRNA signature. .................................. 61
Figure 4.6 Enriched signaling pathways among miRNA-regulated gene targets. ........................ 73
Figure 4.7 Putative regulatory network of selected miRNAs. ...................................................... 75
Figure 4.8 Expression of selected genes in distal colon of sham and infected mice. ................... 76
Figure 4.9 Action of 11 differentially expressed miRNAs on Bim. ............................................. 78
Chapter 5
Figure 5.1 Study Design. .............................................................................................................. 89
Figure 5.2 C. rodentium infection kinetics and effect on mouse body and organ weights. .......... 93
Figure 5.3 Fecal B. bifidum load before and after infection. ........................................................ 94
Figure 5.4 B. bifidum effects on intestinal crypt hyperplasia and tissue damage at day 10 p.i.. .. 96
Figure 5.5 B. bifidum effects on intestinal barrier at day 10 p.i. ................................................... 97
Figure 5.6 Distal colon miRNA expression at day 10 p.i.. ........................................................... 98
ix
List of Abbreviations
A/E lesions-attaching and effacing lesions
ABC-ATP-binding cassette
Abcc3-ATP-Binding Cassette, Sub-Family C, Member 3
AIEC-adherent-invasive Escherichia coli
AJ-adherence junctions
B. bifidum-Bifidobacterium bifidum
Bim-B cell leukemia/lymphoma-2 interacting mediator
BopA-bifidobacterial outer protein
C. rodentium-Citrobacter rodentium
C1galt1-core 1 synthase, glycoprotein-N-acetylgalactosamine 3-beta-
galactosyltransferase 1
CD-Crohn’s disease
Cdkn1a-cyclin-dependent kinase inhibitor
CFU-colony forming units
CRC-colorectal cancer
CT- cycle threshold
Cxcl1/2-chemokine (C-X-C motif) ligand 1/2
DMH-1,2-dimethylhydrazine
DSS-dextran sodium sulfate
E. coli-Escherichia coli
EAEC-enteroaggregative E. coli
EHEC-enterohemorrhagic E. coli
EIEC-enteroinvasive E. coli
EMT-epithelial–mesenchymal transition
Epas1-endothelial PAS domain protein 1
EPEC-enteropathogenic E. coli
EPK-Extracellular signal-regulated kinases
ETEC-enterotoxigenic E. coli
FASEB-Federation of American Societies for Experimental Biology
x
FOS-fructooligosaccharides
Gb3-globotriaosylceramides
GOS-galactooligosaccharides
GRO-growth-related oncogene
H&E-hematoxylin and eosin
H. pylori -Helicobacter pylori
HMOs-human milk oligosaccharides
IBS-irritable bowel diseases
IgE-immunoglobulin E
IL- interleukin
IBD-inflammatory bowel disease
IRAK1 – interleukin-1 receptor-associated kinase 1
L.-Lactobacillus
LB-Luria-Bertani
LEE-Locus of Enterocyte Effacement
LPS-lipopolysaccharide
M-CSF-macrophage colony-stimulating factor
MAMPs-microbial associated molecular patterns
MAPK-Mitogen-activated protein kinase
miR-microRNA (mature form)
miRNA-microRNA
mRNA-messenger RNA
NEC-necrotizing enterocolitis
NLRs-Nod-like receptors
p.i.-post-infection
PDCD4 -programmed cell death 4
pre-miRNAs-precursor miRNAs
pri-miRNAs-primary miRNAs
Prkcz -protein kinase C zeta isoform a
PRRs-pattern recognition receptors
PTEN -phosphatase and tensin homolog
RAC2-Ras-related C3 botulinum toxin substrate 2
xi
RCTs-randomized controlled trials
REST-Relative Expression Software Tool
RhoB -Ras Homolog Family Member B
RISC-RNA-induced silencing complex
Rnd3- Rho family guanosine triphosphate-ase 3
S.-Saccharomyces
SCFAs-short chain fatty acids
Ship1 -Src homology2 domain-containing inositol phosphatase
Stx-Shiga toxin
T3SS-type III secretion system
Tab2 -TGF-beta activated kinase 1/MAP3K7 binding protein 2
Tad-tight-adherence
TA-transit amplifying
Th-T helper
Tir-translocated intimin receptor
TJ-tight junctions
TLRs-Toll-like receptors
TMCH-transmissible murine colonic hyperplasia
TNBS-trinitrobenzene sulfonic acid
TRAF6-Tumor necrosis factor receptor-associated factor 6
Treg-regulatory T
UC-ulcerative colitis
UTR-untranslated region
Wnt-Wingless
Zeb1/2 -zinc finger E-box binding homeobox 1/2
1
Chapter 1
Introduction
1 Introduction
Gastrointestinal diseases, including both acute and chronic conditions such as infectious diarrhea
and inflammatory bowel disease (IBD), account for considerable economic and healthcare
burden in Canada. According to the Canadian Communicable Disease Report, there are
approximately 1.3 cases per-person year of acute gastrointestinal diseases , with symptoms of
vomiting and diarrhea, translating to an annual economic cost of 3.7 billion dollars [1].
Foodborne enteric pathogen infection is the major cause of acute gastrointestinal diseases with
the estimated occurrence of 11 million episodes every year in Canada [1]. Verotoxigenic
Escherichia coli (E. coli), Salmonella, and Campylobacter are the most common bacterial agents
responsible for foodborne diseases [2]. In contrast, IBD is a chronic recurrent condition affecting
over 250,000 Canadians, which represents one of the highest prevalence rates around the world
[3]. In 2012, the economic costs of IBD were about 2.8 billion dollars in Canada. Patients suffer
from both direct and indirect long-term consequences, including work losses and lower quality of
life compared to the general population [4]. Notably, enteric infection episodes and IBD are
linked to the development of colorectal cancer (CRC) [5-7], which is the third most common
diagnosed cancer and the fourth most common cause of cancer death worldwide [8, 9].
The gut microbiota is a complex community of trillions of microorganisms. Alteration of the gut
microbiota composition (dysbiosis) and disruption of its crosstalk with the host have pivotal
implications in these diseases. Fecal microbiota analysis revealed that IBD patients experience
dysbiosis with increased abundance of Proteobacteria and decreased abundance of Bacteroidetes
at the phylum level [10, 11], as well as reduced counts of bacteria belonging to the
Bifidobacterium genus [12]. Pathogens have capacity to breach the host defense mechanisms and
disrupt the healthy host-microbial crosstalk. Enterohemorrhagic E. coli (EHEC) O157:H7 is one
of the most common foodborne enteric pathogens responsible for infectious diarrhea outbreaks
and acute hemorrhagic colitis around the world [13]. Citrobacter rodentium (C. rodentium), a
2
murine-specific pathogen adopts the same virulence mechanisms as EHEC and is commonly
used to model EHEC infection. Colonization of C. rodentium can cause prominent dysbiosis as
showed by the reduction of overall microbiota diversity. Up to 90% of the microbiota can be
replaced by the pathogen at the peak of infection [14, 15]. This is accompanied by intestinal
barrier dysfunction, robust immune responses, as well as colonic crypt hyperplasia, which is the
hallmark of C. rodentium infection. As these features resemble some of the pathological
manifestation of IBD and colonic tumorigenesis, C. rodentium infection has also been
extensively used to study IBD and colorectal cancer development [15].
Probiotics, live microorganisms which when administered in adequate amounts confer a health
benefit on the host [16], contribute to maintenance and restoration of intestinal homeostasis.
Bifidobacterium bifidum, an indigenous member of the human microbiota and a common
probiotic, was shown to confer health-promoting effects in many conditions upon consumption
including, but not restricted to, infectious diarrhea [17-19], necrotizing enterocolitis (NEC) [20],
and irritable bowel syndrome [21-23]. This withstanding, results have not been always consistent
across clinical studies hence recommendation in clinical practice cannot be made yet. Functional
studies also revealed that B. bifidum use can improve gut microbiota composition [24], enhance
barrier function [25-27], modulate host immunity [28-30] and exert antimicrobial activity. B.
bifidum has also been implicated in EHEC infection and IBD. B. bifidum ATCC 29521 has been
shown to interfere with EHEC attachment and colonization in vitro [31]; and oral administration
of B. bifidum S17 alleviated intestinal inflammation in mice with chemically-induced IBD [32].
However, no study has examined the effect of B. bifidum in C. rodentium infection.
Although the host-microbial crosstalk is essential for intestinal homeostasis, the molecular basis
underlying both harmful and beneficial bacteria-host interaction is not fully understood. It is
known that the interaction between host and microorganisms can significant impact host gene
expression [33]. Genome-wide transcriptome analyses revealed that both C. rodentium infection
and B. bifidum colonization can impinge host gene expression in the intestine [30, 34, 35]. In
addition, recent reports show that bacteria can also affect the expression of intestinal microRNAs
(miRNAs). MiRNAs are non-coding RNA molecules, which function post-transcriptionally to
fine-tune gene expression. Currently, 2,588 miRNAs in humans and 1,915 miRNAs in mice have
been identified (miRBase release 21, June 2014) [36]. Previous studies in our research group
have discovered that the presence of microbiota can impact murine miRNA with potential
3
implications in intestinal barrier gene regulation [37]. Administration of B. bifidum MIMBb75 to
mice for two days also affected miRNA expression in the caecum. Meanwhile, recent studies
have demonstrated that various pathogen infections, such as Listeria monocytogenes [38]
Salmonella [39] and Campylobacter concisus [40], can impinge host mucosal miRNA signature.
Interestingly, a study used sterile (germ-free) and conventional (harboring a normal microbiota)
mice and showed that the presence of the microbiota influences the intestinal miRNA response
to Listeria infection [38]. This suggests that the interaction between host and bacteria in the gut
is regulated by miRNAs. However, molecular mechanisms of host-bacteria crosstalk via miRNA
modulation have not been comprehensively investigated in the frame of pathogen-host and
probiotic-host interaction.
The aim of the present study was to investigate if intestinal miRNA are altered in response to C.
rodentium infection and if this response is modulated by probiotic B. bifidum MIMBb75
resulting in mitigation of intestinal inflammation.
4
Chapter 2
Literature review
2 Literature Review
2.1 Intestinal Homeostasis
The intestinal tract comprises the small (duodenum, jejunum, ileum) and large (caecum, colon,
rectum) intestine and is the primary site of nutrient, water, and electrolyte transport; this entails a
series of processes including digestion, absorption, motility, and secretion. The intestinal tract
wall consists of four layers, including the mucosa, submucosa, muscularis externa, and serosa
along the transversal axis. The intestinal mucosa comprises three layers, the epithelium
monolayer facing the lumen, the subepithelial lamina propria and the muscularis mucosae.
Numerous folds are formed in the inner surface of the mucosa along the lumen. Villi (finger-like
luminal protrusions) and microvilli (finger-like structure on brush border membrane of epithelial
cells) are formed in the small intestine, which enhances effectiveness of nutrient absorption;
crypts (flask-shaped submucosal invaginations) are formed in both the small and large intestine
and function as proliferative and secretory compartments of the intestinal epithelium. The
submucosa is a layer of connective tissue, where nerves, blood and lymph vessels are located.
The muscularis externa and serosa are made up of smooth muscle and connective tissue
constituting the outer wall of the intestinal tract, which is responsible for motor activity. Besides
its role in nutrient absorption, the intestinal tract is also the largest immune organ of the body.
The intestinal tract is constantly confronted with dietary and bacterial antigens from the external
environment and is in a symbiotic relationship with the resident microbiota. In order to maintain
homeostasis, the intestinal tract has developed the capacity to distinguish self from non-self and
plays a critical function in preventing harmful compounds or pathogens from entering the
internal system. This is achieved through a complex line of defense that includes epithelium
turnover as well as physical, chemical, and immunological barriers.
5
2.1.1 Intestinal Epithelium Turnover
The intestinal epithelium has a turnover rate of about 5 days in humans and 3 days in mouse,
which is thought to be essential for the timely removal of harmful substances and pathogens [41].
This continuous self-renewal is sustained by proliferation and differentiation of multipotent stem
cells located at the base of the crypt (Figure 2.1, panel A). Upon stimulation, multipotent stem
cells divide into daughter cells known as transit amplifying (TA) cells, which can proliferate
rapidly. Proliferation is driven by the Wingless (Wnt)/β-catenin signalling pathway, which is
initiated by the ligand binding of the secreted glycoprotein Wnt to its membrane-bound
receptors. The activation of this signalling cascade leads to accumulation of β-catenin in the
cytoplasm by sequestering proteins responsible for β-catenin degradation, including APC and
Auxin. As cytoplasmic β-catenin increases, it translocates into the nucleus acting as a
transcriptional factor that promotes proliferation [41]. After a limited number of divisions, the
TA cell population expands to the midway of the crypt and gradually differentiates into epithelial
cells. Absorptive enterocytes represent 90% of the differentiated cells; other cell types include:
goblet cells (secreting mucus), paneth cells (producing antimicrobial peptides and enyzmes),
enteroendocrine cells (secreting hormones), and microfold cells (M cells, sensing antigens). The
distribution and abundance of the diverse cell types is region specific along the intestine. For
example, paneth cells and M cells are predominately found in the small intestine, whereas goblet
cells are abundant in the large intestine [42]. The self-renewal process is completed as
differentiated epithelial cells migrate toward the top of the villus (small intestine) / crypt (large
intestine), where programmed cell death and shedding take place. Therefore, epithelial turnover
requires a tight balance between proliferation, differentiation, apoptosis and exfoliation
processes, and ensures maintenance of epithelial barrier and absorptive functions, contributing to
intestinal defense and homeostasis.
6
Figure 2.1 Colonic Cell Turnover and Transmissible Murine Hyperplasia
This figure illustrates (A) cell turnover in the colon under homeostasis and (B) during C. rodentium infection. (A)
Stem cells, located at the base of the colonic crypt, divide into hyperproliferative TA cells under stimulation of the
signaling molecule Wnt. After a few divisions, TA cells differentiate into various epithelial cells such as absorptive
enterocytes and goblet cells, which produce a thick mucus layer separating the epithelium from the colonic lumen.
When epithelial cells migrate to the top of the crypt, they undergo programmed cell-death and are removed by cell-
shedding. (B) C. rodentium infection is marked by dramatic crypt hyperplasia in the colon known as transmissible
murine colonic hyperplasia (TMCH). The expression of the type III secretion system allows C. rodentium to form
tight attachment onto the host epithelial cell, resulting in host mucosal damage known as attaching and effacing
lesions (A/E lesions), characterized by effacement of microvilli on enterocytes and formation of pedestal-like
structure beneath the attachment site. A/E lesion prevents epithelial cells from shedding. Inflammation is triggered
by the activation of membrane-bound Toll-like receptors (Tlr2/4) and the cytosolic Nod-like receptors (Nod1/2),
which in turn activate the NF-kB signaling cascade, leading to proinflammatory cytokine production that drives Th1
and Th17-dominated responses. Although not completely clear, NF-kB activation leads to over-activation of Wnt
along with other uncharacterized signals, which inhibit hyperproliferative TA cells from differentiating. The
accumulation of undifferentiated cells at the base of the crypt and inhibition of cell-shedding at the top of the crypt
lead to dramatic crypt elongation. The increase of TA cell population eliminates goblet cells, impairing barrier
function of the mucus layer. The impairment of epithelial barrier integrity allows paracellular translocation of
bacterial cells in the system, modified from [15].
7
2.1.2 Physical and Chemical Barriers
In addition to epithelium turnover, several lines of defense to physically and chemically protect
the internal environment from foreign insults are in place to maintain homeostasis. The first is a
physical barrier created by the mucus covering the epithelium. This comprises two layers; an
inner layer, devoid of bacteria and adhering tightly to the epithelium, and an outer layer which is
thicker (830 vs. 116 μm in the human colon) and harbors commensal microorganisms [43].
Mucins produced by goblet cells are the main constituent of the mucus layers. They are heavily
glycosylated proteins that contribute to the viscoelastic properties of the mucus. Oligosaccharide
side chains on mucins may serve as attachment sites and carbon source for bacteria, especially
for commensal bacteria [44].
Intestinal epithelial cells are kept together by tight junctions (TJ), adherence junctions (AJ) and
desmosomes. TJs are intercellular protein complexes made up by transmembrane claudins,
occludin, tricellulin, and intracellular zona occludens and F-actin. They are anchored at the
apical side of the epithelium, sealing the gap between adjacent cells and preventing paracellular
translocation of luminal substances. A dynamic model of TJs has been proposed, indicating that
they can open and close upon stimulation [45]. AJs are made up of transmembrane E-cadherin,
and intracellular β-catenin and α-catenin. Both AJs and desmosomes are located beneath the TJ
and are responsible for strengthening cell-to-cell adhesion [46].
Chemical protection is provided by antimicrobial peptides secreted by paneth cells (for example,
α-defensins and lysozyme) and enterocytes (for example, β-defensins and cathelicidins). These
compounds have cationic regions that can interact with negatively charged phospholipids on the
membrane of both Gram-positive and Gram-negative bacteria, resulting in depolarization and
structural damage of the bacterial membrane [47].
2.1.3 Immune Barrier
Some bacteria can breach the physical and chemical barriers and come in close contact with the
epithelium. The intestinal epithelium has an immunological surveillance system to recognize and
respond to these challenges in order to maintain homeostasis. The activation of the innate
immune response via pattern recognition receptors (PRRs) plays a pivotal role in this
surveillance system [48]. PRRs are glycoprotein receptors specialized in recognizing
8
evolutionarily conserved structures of microbes, known as microbial associated molecular
patterns (MAMPs). Examples of MAMPs include lipopolysacharide (LPS) of Gram-negative
bacteria, flagellin, and peptidoglycans components of the bacterial cell wall [49]. There are two
classes of PRRs, the membrane-bound Toll-like receptors (TLRs) and the cytosolic Nod-like
receptors (NLRs). Ligand binding of MAMPs to PRRs activates NF-κB signalling cascades,
triggering production of proinflammatory cytokines and chemokines for bacterial elimination.
Interestingly, although MAMPs are common in both resident and pathogenic bacteria, the
epithelium has the ability to discriminate between the two. Unlike pathogenic bacteria, which
can invade deep into the host tissue, commensals are localized on the mucosal surface [48]. It has
been found that TLR5 (recognizing flagellin) expression is lower in the apical than the
basolateral membranes of the epithelial cells [50]; TLR2 (recognizing peptidoglycan) and TLR4
(recognizing LPS) are generally down-regulated in epithelial cells [51, 52]. Thus, direct contact
of commensal-derived MAMPs with PRRs is reduced. Nevertheless, immune response toward
commensal bacteria is critical for gut homeostasis. This is well-evidenced by studies showing
that inhibition of NF-κB activation in epithelial cells leads to spontaneous colitis in conventional
mice [53]; also dextran sodium sulfate (DSS)-treated mice deficient in TLR4 or Myd88, an
adaptor necessary for TLR activation, have reduced inflammatory response but increased colitis
and bacterial translocation across the epithelium [54]. NF-κB-induced cytokine production plays
an important part in the polarization of cell-mediated immune response, which involves
differentiation of naive T-lymphocytes into functionally distinct subtypes, such as
proinflammatory T helper (Th) 1, Th2, and Th17 cells, and anti-inflammatory regulatory T (Treg)
cells [55]. It is believed that constant interaction between commensal bacteria and host
epithelium promotes a balance of Treg versus Th1 or Th17 response [56-58]. Therefore, under
homeostasis the gut attains tolerance toward resident bacteria, and immunological readiness
against pathogenic attack.
2.1.4 Gut Microbiota
The human gastrointestinal tract harbours a complex community of microorganisms from all
three domains of life including bacteria, archaea and eukarya, collectively known as the gut
microbiota. The gut microbiota is a diverse and dynamic ecosystem, with approximately 1014
microbe cells encompassing over 100 times more genes (microbiome) than the human genome
[59, 60]. The gut microbiota is dominated by the phyla of Bacteroidetes (B), Firmicutes (F),
9
Proteobacteria (P), and Actinobacteria (A), and by species of the genera Bacteroides (B),
Clostridium (F), Lactobacillus (F), Escherichia (P) and Bifidobacterium (A) [60, 61]. Microbial
density and diversity vary along the horizontal and longitudinal gut axes. Specifically, bacterial
numbers, expressed as colony forming units (CFU)/gram of gut content, increase from 103 in the
stomach and 104-10
7 in the jejunum and ileum to 10
11-10
12 in the colon, which is the most
heavily colonized region of the gastrointestinal tract [60]. The upper gastrointestinal regions are
predominantly occupied by aciduric bacteria such as lactobacilli and streptococci, whereas
colonic regions are preferential sites for facultative and obligate anaerobes such as
Enterobacteria, Bacteroides and Bifidobacterium [62]. Despite the diverse and dynamic nature
of the gut microbiota, it is relatively resilient to changes. Alteration of microbiota, known as
dysbiosis, is associated with intestinal disorders, including IBD [12, 63] and CRC [64], although
it remains unclear whether dysbiosis is the cause of effect of these disorders. Specifically,
pyrosequencing analysis of fecal microbiome from twin pairs, where only one twin developed
IBD, disclosed significant microbial composition alterations in IBD (Crohn’s disease subtype)
patients, which showed an increase abundance of Firmicutes and Proteobacteria at the phylum
level compared to their healthy twin. The increase of Proteobacteria is primarily attributable to
increased representation of the Enterobaceriaceae family, which mainly consists of E. coli [10].
In line with this, presence of adherent-invasive E. coli (AIEC) was more frequently identified in
IBD patients than healthy controls [65]. Likewise, another twin-study revealed that IBD
(ulcerative colitis subtype) patients had a general reduction of mucosal microbiota diversity with
a higher proportion of Proteobacteria and less Bacteroidetes compared to their IBD-free twin
[11]. In CRC, the microbiota composition at genus level showed clear separation between CRC
patients and healthy controls, and the Bacteroides/Prevotella group was significantly increased
in patients [64].
Maintenance of a stable microbiota is essential for intestinal homeostasis. Indeed, the continuous
interaction between gut microbiota and the host imparts many beneficial effects, such as,
providing nutrients and vitamins [66], regulating host’s energy balance [67], protecting against
pathogens [68], and educating the intestinal immune system [69]. For example, bacterial
biosynthesis and fermentation can result in production of vitamins (folic acid, vitamin B and K)
as well as short chain fatty acids (SCFAs), mainly acetate, propionate and butyrate, which can be
utilized by the host as nutrient and energy sources [66, 67]. Gut microbiota resilience is critical
10
for colonization resistance to pathogens, providing protection to the host intestine. It has been
shown that germ-free and antibiotic-treated mice were more susceptible to pathogen infections,
such as C. rodentium [70], Listeria monocytogenes [71], and Salmonella enterica [72] than
conventional mice. Commensal bacteria are able to exclude pathogens from the mucosa via
secretion of antimicrobial peptides, competition for nutrients and attachment sites, and quorum
sensing [73]. These same mechanisms are also utilized by probiotic (beneficial) bacteria. For
example, E. coli Nissle 1907 can compete with Salmonella for iron and reduce Salmonella
colonization in mice [74]. Microorganisms utilize quorum sensing to regulate gene expression
based on population density via the secretion of compounds referred to as autoinducers [75]. It
has been shown that autoinducers produced by resident Ruminococcus obeum reduce
colonization of a diarrhea-causing pathogen, Vibrio cholera, in gnotobiotic mice associated with
an artificial bacterial community resembling the healthy human microbiota [76]. Some pathogens
have evolved colonization strategies that take advantage of quorum sensing. The expression of
the pathogenicity island Locus of Enterocyte Effacement (LEE) of EHEC is activated by a
quorum sensing molecule produced by the microbiota resulting in the formation of tight
attachment to intestinal cells [77]. Furthermore, commensal bacteria are able to reinforce the host
defense barrier by constantly shaping and priming the mucosal immune system. It has been
shown that resident microbes have both proinflammatory and anti-inflammatory effects on the
host mucosa to keep activities of different subsets of T-cells in balance. For example, the
resident microbe Candidatus Savagella (also known as segmented filamentous bacteria),
stimulates a pro-inflammatory Th17 response, which is crucial in fighting against pathogen
infection and is associated with colitis when overly activated [78, 79]. In contrast, other
commensal bacteria, such as Bacteroides fragilis, can downregulate Th17 response and induce
Treg response, thereby promoting immune tolerance [80]. Hence, the gut microbiota is essential
in maintaining physiological inflammation of the host mucosa.
11
2.2 Foodborne Enterohemorrhagic E. coli
Pathogenic bacteria have evolved strategies to escape host defenses and come in close contact
with host epithelium in order to survive and replicate. Although E. coli is the predominant non-
pathogenic facultative anaerobe found in the normal gut microbiota, some strains of E. coli have
acquired pathogenic properties to cause diarrheal disease [81]. These diarrheagenic E. coli strains
are categorized into five classes depending on their virulence. These include enteropathogenic E.
coli (EPEC), enterotoxigenic E. coli (ETEC), enteroaggregative E. coli (EAEC), enteroinvasive
E. coli (EIEC), and EHEC. E. coli O157:H7, a serotype belonging to EHEC, is a foodborne
enteric bacterial pathogen responsible for most cases of infectious diarrhea outbreaks around the
world [13]. This infection can also lead to hemorrhagic colitis, hemolytic uremic syndrome, and
even death due to renal failure in children [13]. It is highly transmissible between persons via the
oral-fecal route and is usually acquired through consumption of ruminant feces-contaminated
drinking water and food, such as undercooked ground beef, fruits and vegetables. The largest
outbreak in Canada took place in Walkerton, ON in 2000 as a consequence of contaminated
drinking water resulting in over two thousand cases of infection and 7 deaths [82]. It was found
that infected individuals had an increased risk of developing a chronic condition known as post-
infectious irritable bowel syndrome (IBS) two years after the Walkerton outbreak [83]. Although
no correlation was observed between EHEC infection and other chronic disorders such as IBD
and CRC, viral and bacterial gastrointestinal infection have been suggested to play a role in the
development of these chronic disorders [84, 85]. The following sections will focus on addressing
pathogenesis and animal model of EHEC infection and its potential links to IBD and CRC.
2.2.1 EHEC Infection
EHEC infection causes a characteristic damage to the host intestine, known as A/E lesions [86].
A/E lesions are characterized by the intimate bacterial attachment onto the host epithelium,
effacement of intestinal brush border microvilli and formation of pedestal-like structures beneath
the attachment site [15, 87]. Virulence genes controlling the establishment of A/E lesions of
EHEC are located on the pathogenicity island, called LEE, which encodes a type III secretion
system (T3SS) [87]. The T3SS allows the bacteria to secret effector proteins into the host cell via
the formation of a needle-like channel breaching through the host membrane. One of the central
secreted effectors is the translocated intimin receptor (Tir), which can be incorporated onto the
host membrane surface, serving as a receptor for the bacterial cell surface adhesin protein called
12
intimin. This ligand-receptor binding ensures firm adhesion between the bacteria and the host
epithelium [15]. Other effector proteins such as EspG, EspF and MAP stay in the epithelial
cytosol and are able to hijack the host cell machinery to cause damages to the host cell, including
actin-cytoskeleton rearrangement [88, 89], tight junction defects[90, 91], solute transport
dysfunction [92, 93], innate immunity activation [94, 95], and cell-cycle disruption [96-99]. The
rearrangement of actin-cytoskeleton allows formation of the pedestal-like structure beneath the
attachment site, increasing the proximity between the bacterial cell and host epithelium. The
consequences of these pathologic actions are manifested as increased intestinal permeability,
electrolytes imbalance, crypt hyperplasia (details in section 2.2.2), and inflammation (details in
section 2.2.3) [15]. Moreover, E. coli O157:H7 has the ability to release the potent hemolytic
Shiga toxin (Stx) upon infection [100]. Stx is an AB toxin originated from bacteriophages. The
globotriaosylceramides (Gb3) receptor expressed on the intestinal epithelial cell surface can bind
to the B subunit of Stx triggering epithelial internalization of the toxin. Once the A subunit is
inside the host cell, it inhibits protein synthesis, causing local cell death in the intestine and
kidney damages via blood circulation. It has been suggested that Stx can also enter the host
system in a Gb3-receptor independent manner [101]. Treatment of EHEC remains challenging,
since antibiotic treatment augments Stx release upon bacterial cell death.
2.2.2 Citrobacter rodentium
C. rodentium is a Gram-negative murine-restricted pathogen. With 67% shared genetic
similarities with EHEC, C. rodentium also harbours and expresses LEE, and thereby, adopts the
same virulence mechanisms as EHEC [102]. Although C. rodentium does not process the Stx
gene, a genetically engineered Stx-expressing C. rodentium strain was recently constructed
[103]. C. rodentium infection has been extensively used to model EHEC infection. The
establishment of this animal model is usually achieved by orally inoculation of mice with live
cells of C. rodentium, which results in a highly reproducible infection cycle in mice of the same
strain [104-106]. In C57Bl/6 mice, the initial site of C. rodentium colonization is in the caecum;
but about 2-3 days post-infection (p.i.) C. rodentium re-localizes and predominantly colonizes
the colon, especially the distal colon. The peak of bacteria load is reached by day 8-10 with over
109 CFU/g of feces. However, complete bacterial clearance takes about 2 – 3 weeks after
inoculation [104, 105]. Colonization of C. rodentium causes significant dysbiosis. C. rodentium
infected mice had a 3-fold reduction of resident bacteria counts and percentages of Bacteroides
13
and γ-Proteobacteria were significant decreased and increased, respectively, on day 7 p.i. [14].
However, infection kinetics can vary between different mouse strains. NIH Swiss and C57Bl/6
mice are resistant strains that can recover from the infection, while C3H/HeJ mice are a
susceptible strain that suffers 100% mortality by day 10 p.i [107]. Compared to resistant strains,
C3H/HeJ mice were rapidly and heavily colonized by C. rodentium (109 vs. 10
4 CFU/g on day 4
p.i.), with greater mucosal damage and bacterial systemic translocation. Recent studies suggested
that transferring microbiota from resistant mice to C3H/HeJ mice reduces C. rodentium
colonization, colonic pathology and mortality [108, 109]. Ghosh et al. reported that the
microbiota of C3H/HeJ had a lower abundance of Bacteroidetes compared to that of C57Bl/6
mice, and receiving C57Bl/6 microbiota significantly increased C3H/HeJ Bacteroidetes load and
attenuated infection pathology, implying that the protective effects of resistant microbiota might
be attributable to Bacteroidetes abundance [108]. Using similar study approaches, Willing et al.
suggested that the protective effects of resistant microbiota was associated with modulating host
immunity toward Th17 response, as transferring susceptible mouse microbiota to resistant host
reduced this defensive response [109]. One study on germ-free C57Bl/6 mice found that germ-
free mice had a 10-fold increase of peak C. rodentium load with a delayed bacterial eradication
until after 42 days p.i; however there is no significant difference in tissue pathology or mortality
between germ-free and convention mice [110]. Interestingly, another study found that antibiotic-
induced dysbiosis exacerbated colon pathology but has no impact on C. rodentium colonization
in conventional mice [70]. These findings indicate that microbiota is involved in regulating host
susceptibility to C. rodentium infection, but not necessarily via pathogen exclusion.
The hallmark of C. rodentium infection is TMCH (Figure 2.1, Panel B) [15]. During C.
rodentium infection, there is an excessive proliferation of undifferentiated enterocytes (ie. TA
cells) in the colonic crypt, as well as inhibition of cell shedding at the top of the crypt due to the
tight bacterial attachment, which together lead to dramatic colonic crypt elongation, goblet cell
depletion and mucosal thickening, collective known as TMCH [15]. The initiation of TMCH is
believed to be associated with NF-κB activation, as MyD88-deficient mice did not develop
TMCH upon C. rodentium infection [111, 112]. Indeed, microarray analyses of colonic
transcriptome revealed pronounced alterations in expression of genes related to immune response
and cell-cycle regulation in C. rodentium-infected mice compared to controls, indicating C.
rodentium pathogenicity can impact host gene expression [34, 35]. For example, Spehlmann et
14
al. found that C. rodentium infection resulted in a significant induction of genes encoding for
chemokines production in the colon, such as chemokine (C-X-C motif) ligand 1/2 (Cxcl1/2),
which are potent chemokines invoking mucosal neutrophil influx and contribute to host defense
against pathogens [35]. Borenshtein et al. found that Aquaporin 8, encoding for a water channel
protein, was one of the prominent downregulated genes during C. rodentium infection [34]; it
was showed before that this gene was highly expressed in differentiated epithelial cells but
weakly in colorectal tumor tissue[113].
2.2.3 EHEC in IBD
IBD is a chronic intestinal disorder, characterized by sustained bowel inflammation, intestinal
barrier dysfunction, abdominal pain and diarrhea [114]. There are two main forms of IBD,
Crohn’s disease (CD) and ulcerative colitis (UC), distinguished mainly by histopathological
features. Notably, CD is associated with a patchy and transmural intestinal inflammation
dominated by Th1 responses, while UC demonstrates a continuous and shallow inflammation
dominated by Th2 responses [115]. There is an increase incidence and prevalence of IBD
worldwide, especially in North America, where an estimated 1.4 million people are affected by
IBD [116]. Although genetic, immunologic, and environmental factors have been identified, IBD
is a multifactorial disorder of unknown etiology [115]. There is no direct evidence showing
association between EHEC infection and IBD; epidemiology studies have suggested a link
between acute enteric infections and IBD, where the risk of developing IBD is increased after
enteric bacterial infection episodes [84, 117]. In addition, Qiu et al. revealed that IBD patients
have a higher serum level of EHEC-derived bacterial toxins compared to healthy controls [118].
It has been hypothesized that enteric infection may act as an initiation factor triggering chronic
inflammation in people genetically susceptible to IBD [119].
Because IBD is characterized by intestinal inflammation, common models of IBD are established
by using genetic modification (IL-10 knock out) and chemical reagents, such as DSS and
trinitrobenzene sulfonic acid (TNBS) to induce colitis in mice [115]. C. rodentium infection is
also extensively used as a biological reagent to model IBD [120, 121]. Mice infected with C.
rodentium elicit a Th1 profile that resembles the immune response seen in CD [121].
Specifically, upon colonization of the intestine C. rodentium-derived MAMPs are recognized by
TLR2, TLR4, as well as NOD1 and NOD2, followed by activation of NF-κB signalling, which in
15
turn induces production of proinflammatory cytokines, TNF-α, IFN-γ, IL-12 that characterize the
Th1 response [120, 122, 123]. A robust Th1 response can recruit lymphocytes and macrophages
to the infection site for pathogen control, resulting in significant immune cell infiltration into the
host mucosa. Besides sharing immunologic similarities with IBD, dysbiosis caused by C.
rodentium infection is also comparable to IBD. For instance, the increase of the proportion of
Proteobacteria in the microbiota found in C. rodentium infected mice is commonly seen in IBD
patients [124-126]. In fact, it has been suggested that infection-induced dysbiosis may lead to
hyperactive T-cell mediated immune response in genetically predisposed individuals,
contributing to persistent inflammation [127, 128]. Intriguingly, Fattouh et al. showed that
knockout of Ras-related C3 botulinum toxin substrate 2 (RAC2), a gene associated with IBD in
humans, attenuates C. rodentium-induced colitis in mice, which implies a role of RAC2 in IBD
pathogenesis [129]. Thus, understanding C. rodentium-induced colitis may contribute to
elucidation of the pathogenesis of IBD.
2.2.4 EHEC in CRC
CRC is the third most common diagnosed cancer and the fourth most common cause of cancer
death worldwide [8, 9]. Similar to IBD, it is more prevalent in Western countries. Genetic
predisposition [130], diet [131], and history of IBD [5] have been identified as some of the risk
factors associated with CRC. Recently, enteric pathogen infections have been implicated in the
etiology of CRC [6, 7]. Being one of the most common causes of enteric infection, EHEC has
been shown to be associated with CRC pathogenesis [97, 132-135]. Two independent studies
demonstrated that E. coli O157:H7-derived components have genotoxic effects and can lead to
genomic instability [133, 134]. This is important because chromosomal instability is a
characteristic observed in about 80% of CRC [136]. Specifically, Tyrer et al. reported that
exposure of Caco-2 and HCT116 cell lines to E. coli O157:H7-derived components including Stx
causes DNA double-strand breaks associated with cell cycle arrest, leading to an increase of cell
proliferation [133]. Similarly the expression of a tumor suppressor gene p53 was increased in
heat-killed E. coli O157:H7 treated mouse intestine, suggesting an activation of repair
mechanisms of DNA double-strand break [134] and dysfunction of p53 is associated with cancer
development including CRC in humans [137]. However, currently there is a lack of human
studies substantiating a link between EHEC and CRC. In fact, C. rodentium has recently
emerged as a model for studying colonic tumorigenesis. It has been shown that C. rodentium
16
infection promotes tumorigenesis in carcinogen-treated and genetically susceptible (ApcMin/+
heterozygous) mice [138, 139]. As discussed above, the hallmark of C. rodentium infection is
TMCH, which is believed to be associated with NF-κB signaling activation with the precise
mechanisms uncharacterized [15, 111, 112]. Tight regulation of the Wnt/β-catenin signalling
pathway is critical to the equilibrium of epithelial cell proliferation and differentiation, where
over-activation contributes to CRC development [41]. Mutation in Apc gene, encoding protein
that inactivates β-catenin, is seen in 80% of sporadic CRC [140]. Recent study by Chandrakesan
et al. revealed that elevation of active β-catenin and reduction of APC levels coincide with
increase of crypt hyperplasia, while inhibition of β-catenin activity leads to decrease hyperplasia
in C. rodentium infected mice, indicating a central role of Wnt/β-catenin activation in bacteria-
driven hyperplasia [141]. Interestingly, the interplay between NF-κB and Wnt/β-catenin in
controlling TMCH has been evidenced, although not fully understood [141, 142]. Moreover,
intestinal hypoxia and activation of the Wnt/β-catenin are essential for triggering epithelial–
mesenchymal transition (EMT), which is a key feature of cancer metastasis. It has been
demonstrated that C. rodentium infection induces a hypoxic state, and transforms cultured
enterocyte into fibroblast-like mesenchymal cells, which undergo significant epigenetic profile
changes such as histone modification and chromatin remodeling [143]. In summary, C.
rodentium-induced hyperplasia is an important model for studying the molecular events
underlying tumorigenesis and the early-onset of EMT.
2.3 Probiotics
Probiotics are defined as “live microorganisms which when administered in adequate amounts
confer a health benefit on the host” by the World Health Organization and Food and Agriculture
Organization [16]. Based on this definition, in order for a microorganism to be considered as a
probiotic, it must be capable to survive through the gastrointestinal tract, colonize and proliferate
in the gut, be safe and effective in humans, as well as remain viable and effective during product
shelf-life [144]. Probiotics can be found in various foods, including some fermented products,
nutritional supplements, and medical foods constituting a considerable part of the functional food
market [145]. Bacteria of the Bifidobacterium and L. (Lactobacillus) genera are the most
commercially available probiotics for humans [146]. Bifidobacterium species including B.
adolescentis, B. animalis, B. bifidum, B. breve and B. longum, and L. species including L.
acidophilus, L. casei, L. fermentum, L. gasseri, L. johnsonii, L. paracasei, L. plantarum, L.
17
rhamnosus and L. salivarius with a dosage of 109 CFU per food serving are the species for which
general probiotics claims are currently accepted by Health Canada [147]. Other bacterial species
from Enterococcus, Escherichia, Streptococcus genera and yeast Saccharomyces (S.) boulardii
have also been used as probiotics. They are usually isolated from fermented foods, intestinal
contents of humans and animals, and more recently breast milk and non-fermented meat and
fruits [148]. In addition to general benefits of supporting healthy gut microbiota, digestive tract,
and immune system [149], probiotics use may confer strain-specific beneficial effects in a broad
spectrum of conditions. These include infectious diarrhea, antibiotic-associated diarrhea,
pouchitis, atopic eczema, and upper respiratory tract infections [150].
It is believed that probiotics may exert beneficial effects against pathogen infections via their
antimicrobial activity. L. salivarius UCC118 has been shown to protect mice from Listeria
monocytogenes infection via the production of antimicrobial bacteriocin; while the mutant strain
defect of producing bacteriocin failed to reduce pathogen translocation to liver and spleen [151].
Similar to the gut microbiota, probiotics can exert pathogen exclusion effects to protect the host.
Johnson-Henry et al. showed that pre-treating mice with the probiotic mixture of L. rhamnosus
R0011 and L. acidophilus R0052 reduced C. rodentium colonization of the intestine and
attenuated TMCH and colitis; they further suggested that these probiotic effects are as a result of
colonization resistance, as co-culturing the probiotics with C. rodentium inhibits C. rodentium
growth on human colonic epithelial cells in a time-dependent manner [152]. Attachment to the
epithelium is essential for the pathogenicity of enteric pathogens; probiotics may out-compete
pathogens from adhering to the epithelium. Using in vitro competition assays, L. casei DN-114
001 and E. coli Nissle 1917 were showed to be able to prevent epithelial adhesion and invasion
of AIEC that are abnormally abundant in CD patients [153, 154]. It was suggested that the strong
binding ability of E. coli Nissle1917 to epithelial cells may allow the establishment of a biofilm
that physically blocks pathogens from accessing the epithelium [154]. Moreover, probiotics use
may also provide protection against pathogens by enhancing mucosal barrier integrity and host
immunity. Increased intestinal permeability as a result of redistribution and reduction of TJ
proteins is commonly observed in active IBD patients [155, 156]. Administration of the probiotic
mixture VSL#3 (L. casei, L. plantarum, L. acidophilus, L. bulgaricus, B. longum, B. infantis, B.
breve, and Streptococcus salivarius subspecies thermophilus) for 7 days to mice with DSS-
induced colitis, has been shown to mitigate inflammation and restore barrier integrity via
18
stabilizing expression and apical localization of TJ proteins, occluding, ZO-1 and claudins [157].
In line with this, in vitro studies further demonstrated that this effect of VSL#3 is achieved by the
activation of Extracellular signal-regulated kinases (EPK) in the Mitogen-activated protein
kinase (MAPK) signaling pathway that regulates TJ proteins expression [158, 159]. Gareau et al.
found that C. rodentium infection resulted in a decrease Treg response and increase mortality in
neonatal mice, but administration of the probiotic mixture L. rhamnosus R0011 and L. helveticus
R0052 normalized Treg levels and prevent neonatal mice from death. However, these protective
effects were abolished in T-cell deficient neonatal mice, indicating the critical role of T-cell
response in mediating probiotic function [160]. Similarly, probiotic B. breve was found to be
able to modulate T-cell polarization toward a Th2 and Treg biased profile in DSS-induced IBD
model [161].
It is worth noting that although encouraging results have been obtained in conditions like
infectious diarrhea and IBD from animal studies, the efficacy of probiotic use remains
controversial in human studies. For instance, although early findings from randomized control
trials (RCTs) suggested that the use of probiotic yeast S. boulardii can prevent recurrent of
Clostridium difficile-induced diarrhea [162, 163], a more recent systematic review of eleven
clinical trials revealed that S. boulardii treatment is ineffective in preventing C. difficile-induced
diarrhea [164]. Similarly, insufficient evidence was observed in terms of beneficial effects of
probiotics in IBD, especially for CD. According to systematic review studies, the use of VSL#3
was showed to significantly promote remission of active UC patients compared to placebo [165,
166], and all four RCTs showed positive effective of VSL#3 in maintaining remission of
pouchitis [167-170]. However, there is a lack of evidence for the efficacy of probiotic use in CD
patients. Indeed, the uses of VSL#3 [171] and several other probiotics such as E. coli Nissle
1917[172] and L. rhamnosus GG [173, 174] in clinical trials were all shown to be ineffective in
preventing relapses of CD compared to placebo. To date, recommendation of probiotics use for
IBD in clinical practice can only be made for pouchitis [150].
The following sections will focus on a probiotic bacterium, B. bifidum, and its health-promoting
effects.
19
2.3.1 Bifidobacterium bifidum
B. bifidum is a Y-shaped Gram-positive, anaerobic, non-motile, non-gas-producing bacterium
belonging to the Actinobacteria phylum. B. bifidum was first isolated from the feces of breast-fed
infants by Henri Tissier in 1899 [175] and it is one of the 47 species in the Bifidobacterium
genus [176].
B. bifidum is one of the first autochthonous inhabitants that colonize the human lower
gastrointestinal tract shortly after birth [177]. Although other bifidobacterial species are
widespread along the intestine, B. bifidum is believed to localize distally [178, 179]. Yet, the
colonization patterns of this bacterium are largely unknown. A study revealed that the number of
B. bifidum cells is greater in the proximal colon and the cecum than in the distal colon or feces in
mice administered with B. bifidum [180]; nevertheless, other studies detected similar abundance
of B. bifidum at about 108 – 10
9 CFU/g in both ceacal lumen and feces of human adults [181-
183]. In addition to the digestive system, B. bifidum is also the most frequently detected species
in breast milk [184].
Gut microorganisms are capable to metabolize complex carbohydrates that are not readily
digestible by the host enzymes in intestine. It was found that bifidobacterial genomes contain
about 30% more genes involved in carbohydrate metabolism than most of other gut microbes
[176]. Although bifidobacterial species can metabolize a variety of carbohydrates (such as
dextrin, maltose, and lactose), the fermentation ability is believed to be dependent on species and
strains [185]. Compared to other bifidobacteria, B. bifidum strains degrade a relatively small
number of carbohydrates, which implicates a strategy for efficient niche-specific colonization
and adaptation [186]. What sets B. bifidum apart from other bacteria of the genus is the ability to
utilize host-derived carbohydrates, especially mucin-associated glycans and human milk
oligosaccharides (HMOs). Early studies showed that B. bifidum was the only species of the 29
bifidobacteria tested that can metabolize porcine gastric mucin [187]. It is known that the
capacity of bifidobacteria to degrade intestinal mucin depends on the presence of genes, afcA and
engBF, encoding two different extracellular glycosidases. An in vitro study of different
representative strains of bifidobacteria species demonstrated that only strains of B. bifidum, L22
and D119, possess both genes with the highest degradation competence [188]. In contrast to
other species, B. bifidum strains, especially PRL2010, can grow on mucin-base medium by using
20
mucin as their sole carbon source [19]. Only Bifidobacterium and Bacteroides were reported to
have the capacity utilize HMOs. It was previously shown that exploiting its special enzymes a-
fucosidases and lacto-N-biosidase B. bifidum breaks down HMOs more efficiently than other
species such as B. adolescentis [189]. The capability to ferment HMOs implies a strong selective
advantage of this species in very early life (i.e. during breastfeeding), accounting for their
predominance in infant gut microbiota. Moreover, members of B. bifidum can also ferment plant-
derived oligosaccharides, such as fructooligosaccharides (FOS) and galactoologosaccharides
(GOS), which are recognized as bifidogenic factors. For example, it was shown that strain
NCIMB 41171 can produce four different β-galactosidases; oral intake of metabolites from these
enzymes significantly increased bifidobacterial counts in feces of healthy humans [190].
2.3.1.1 Adhesive Factors and Interaction with the Host
The ability to adhere to the intestinal epithelial cells may play a pivotal role for bacterial
establishment in the gut. It has been shown that B. bifidum strains may possess adherent
properties. For example, a probiotic strain, B. bifidum MIMBb75, was shown to bind most
tightly to Caco-2 cell monolayer than other probiotic species such as B. longum NCC2705 and L.
rhamnosus GG [29]. B. bifidum MIMBb75 can also adhere to mucus producing HT-29 epithelial
cells and display autoaggregating phenotype. The finding that strong autoaggregation takes place
under low pH conditions might indicate a protective strategy for gastric transit. It is plausible that
bacterial cells self-aggregate in the stomach hindering adhesins to interact with host epithelium,
whereas they disassociate in the intestine as pH rises, permitting attachment and colonization
[179]. Two surface adhesion-like molecules, bifidobacterial outer protein (BopA) and hair-like
appendages have been proposed to be involved in assisting B. bifidum to adhere to the
epithelium.
BopA is a cell wall-associated lipoprotein that is unique to the B. bifidum species [29]. It was
first identified in B. bifidum MIMBb75, where BopA was chromatographically purified from
bacterial cell wall extracts. Competitive adhesion assays showed that treatment of purified BopA
significantly inhibited adhesion of B. bifidum strains MIMBb75 to Caco-2 cells [29]. Likewise,
another study showed that adhesion of B. bifidum S17 was reduced after incubating T84 and
HT29 cells with purified BopA protein, suggesting BopA may compete with B. bifidum for
binding sites on epithelial cells [191]. Overexpressing BopA in B. longum resulted in a
21
prominent increase of adhesion to epithelial cells compared to wild-type B. longum, which does
not possess BopA and has weak adherent ability [191]. However, the role of BopA in B. bifidum
adhesion remains controversial. A more recent study revisited BopA and showed that blockage
of BopA did not cause significant reduction of bacterial adhesion, arguing that there may be
cooperative mechanisms involved [23]. It has been proposed that BopA is a moonlighting protein
for it is thought to exercise multiple biological functions besides being an adhesion factor [29,
191]. Gene and protein sequence analyses revealed that BopA is located on an operon encoding a
putative ATP-binding cassette (ABC) transporter, and the BopA protein contains a
tripeptide/oligopeptide solute-binding domain, implying its potential role in peptide nutrient
uptake [29, 191]. Interestingly, incubation of Caco-2 cells with B. bifidum MIMBb75 or purified
BopA resulted in elevation of IL-8 levels but did not impact IL-6 production, indicating that B.
bifidum MIMb75 and BopA may be able to modulate the host immune response stimulating
transient subclinical inflammation in the gut [29].
Like other bifidobacteria, B. bifidum also possesses surface hair-like appendages, known as pili
and fimbriae. They are considered as strong adhesion factors contributing to intimate interaction
between microbes and host epithelial cells [192]. For example, tight-adherence (Tad) pili
identified in B. breve UCC2003 were shown in vivo to be essential for colonization in the mouse
gut [193, 194]. The gene of Tad pili is strictly conserved within bifidobacteria, such that it is also
present in B. bifidum S17 and PRL2010 [195, 196]. However, the number and sequence of gene
clusters encoding for sortase-dependent pili differ among strains, which might confer strain
specific functions. For instance, it is known that B. bifidum PRL2010 displayed the highest
binding ability to extracellular matrix proteins, especially fibrinogen, than other bifidobacteria
possessing these sortase-dependant pili, such as B. breve and B. adolescentis [196]. The two
sortase-dependent pili identified on strain PRL2010 can induce TNF-α response in vivo; when
transgenically expressed these pili genes in the initially nonpiliated bacteria strain, TNF-α
response was also induced upon intestinal colonization. Stimulation of TNF-α response might be
an important feature of PRL2010 as a probiotic strain, since TNF-α associated cytokines not only
represent major proinflammatory mediators, but may also play a crucial role in exerting
antitumor and anti-infection effects [196-198]. These observations imply that sortase-dependent
pili not only promote adhesion but also modulate host immune responses. Indeed, exposing
human epithelial cell lines and mouse intestinal cells to PRL2010 resulted in significant changes
22
of intestinal cell transcriptome affecting an array of immune-related genes; this includes
pronounced downregulation of heat shock proteins and chemokines such as Cxcl2 and Cxcl3,
along with upregulation of antimicrobial β-defensins and TJ proteins [30]. Importantly, heat
shock proteins are usually expressed in response to cellular stress which may have implications
in tumorigenesis linked to infection and chronic inflammation [199]; overexpression Cxcl2 and
Cxcl3 have also been associated with colorectal carcinoma [200]. These suggest that B. bifidum
PRL2010 impacts host gene expression which may presumably enhance host innate defense
response.
2.3.1.2 Health Benefits
Probiotic effects of different B. bifidum strains have been implicated in various disease
conditions as summarized in Table 2.1. Most clinical studies use B. bifidum in a mixture with
other probiotic species from the Lactobacillus genus. Although beneficial effects observed from
a probiotic mix cannot be extrapolated to B. bifidum alone, these studies can still be interesting
because they suggest complementary effects of B. bifidum with other probiotics and provide
directions for mechanistic studies. In infants and children, the use of B. bifidum has shown
beneficial effects in infectious diarrhea and NEC. A large multicentre single-blind RCT showed
that oral administration of a probiotic mixture containing B. bifidum for 5 days significantly
reduced duration and severity of acute diarrhea in children aged 3-36 months compared to the
placebo [17]. Similar results were found in two other clinical trials in hospitalized infants and
children between 3 to 120 months old with different probiotic formulas containing B. bifidum
[18, 19]. Animal studies showed that B. bifidum strains, such as ATCC15696 and CIDCA5310,
improved diarrhea and enterocolitis caused by rotavirus and Clostridium difficile, which are two
common infectious agents responsible for hospital-acquired diarrhea in children [201-203]. The
use of B. bifidum in a probiotic mixture has been found to reduce incidence and severity of NEC
in very low birth weight preterm infants in a retrospective cohort study [20]; but two other
double-blind RCTs with different B. bifidum-containing probiotic mixtures found only reduction
in NEC incidence and morbidity, with no difference in severity [204, 205]. Feeding NEC
neonatal rats with B. bifidum strain OLB6378 attenuated NEC-induced mucosal apoptosis and
inflammation, via normalizing the expression and localization of TJs and AJs as well as
proinflammatory cytokine IL-6 level [25, 26, 206]. Furthermore, B. bifidum consumption has
also been evidenced to have systemic benefits outside of the intestinal system, such as in atopic
23
eczema and asthma among children, plausibly via reducing serum levels of antibody
immunoglobulin E (IgE) [207-209]. B. bifidum use in adults is mainly associated with beneficial
effects in IBS. IBS is a chronic gastrointestinal functional disorder, characterized by abdominal
pain and bloating in absence of detectable structural abnormalities or obvious inflammation
[210]. It has been shown by two clinical trials that B. bifidum BGN4 and NCIMB 30153 when
used in a mixture with Lactobacillus species improved IBS symptoms with no adverse effects
[21, 22]. Interestingly, a recent prospective RCT showed that the use of B. bifidum MIMBb75 for
4 weeks was able to significantly alleviate pain, discomfort and digestive disorder in IBS patients
compared to the placebo controls [23].
Moreover, B. bifidum has also been implicated in pathogen infections, IBD and CRC. Ingestion
of fermented milk containing strain BF-1 for 12 weeks resulted in reduction of serum biomarkers
associated with H. pylori (Helicobacter pylori) infection compared to baseline in H. pylori-
positive patients [211]. Bayoumi et al. recently showed that B. bifidum ATCC 29521 was able to
interfere with EHEC and Salmonella enterica serovar Typhimurium attachment and colonization
by downregulating the pathogens’ virulence factors intimin expression in vitro, thereby attaining
pathogen exclusion effects [31]. Although there is insufficient clinical evidence showing B.
bifidum efficacy in IBD to date, beneficial effects of B. bifidum have been implicated in animal
and cell culture studies in intestinal inflammation. For instance, Kim et al. reported that dietary
supplement of B. bifidum BGN4 alleviates lymphocytes infiltration and Th1-type cytokines
production in mice with induced IBD [212]. Similarly, ingestion of strain S17 before and after
induction of TNBS-colitis reduced Th1-driven intestinal inflammation [32]. One study using
non-invasive ETEC to induce Th1 inflammation in rats demonstrated that the supplementation of
a probiotic mix composed of B. bifidum R0071, B. longum subsp. infantis R0033 and L.
helveticus R0052, modulated Th1 immune response by supressing expression of
proinflammatory cytokines IL-6 and TNF-α [213]. In DSS-model of colitis, the probiotic mixture
LaBb Dahi containing L. acidophilus LaVK2 and B. bifidum BbVK3 was found to be able to
inhibit activity of β-glucuronidase, which is an intestinal enzyme elevated in colitis and
associated with carcinogenic and cytotoxic metabolites production [214]. As a matter of fact,
links have been drawn between B. bifidum and CRC. Specifically, an early in vitro study
demonstrated that in a screen of 30 different bifidobacterial strains, B. bifidum BGN4 elicited the
highest inhibitory effect on colon cancer cells HT-29 growth; treatment of BGN4-derived
24
polysaccharides inhibits DNA synthesis of various colon cancer cell lines in a dose-dependent
manner, suggesting their antitumor effects via repressing cellular proliferation [215]. A recent
study showed that feeding LaBb Dahi for 32 weeks in the 1, 2-dimethylhydrazine (DMH)
induced colorectal carcinogenesis rat model was able to protect rats from developing CRC
evidenced by significant attenuation of pre-neoplastic lesions formation in treated rats [216]. It
was postulated by the investigators that the molecular basis underlying the anti-carcinogenic
effects of LaBb Dahi could be attributed to upregulation of carcinogen-detoxifying activities,
downregulation of β-glucuronidase activity, and decrease expression of anti-apoptotic bcl-2 and
c-myc genes in colon tissue [216].
Most studies to date have focused on examining beneficial effects of B. bifidum in diseases
states, while only a few explored health maintaining effects of B. bifidum. These include two
RCTs reporting that the use of B. bifidum in healthy children and adults for over 3 months was
linked to a reduction and shortening of common cold episodes [217, 218]. An early study by
Schiffrin et al. demonstrated that intake of B. bifidum Bb12 was able to increase phagocytosis
activity against E. coli species in healthy adult blood, suggesting an enhancement of host anti-
infectious mechanism in response to B. bifidum [219]. Immunomodulation effects were also seen
in studies with healthy animals, where colonization of B. bifidum given in a probiotic mixture
increased anti-inflammatory cytokine IL-10 production of the large intestine [220]. Another
beneficial effect of B. bifidum is associated with improvement of gut microbiota composition.
Specifically, one small clinical trial showed that ingestion of B. bifidum containing probiotics
significantly increased the abundance and diversity of fecal bifidobacterial population compared
to baseline in healthy elderly subjects [24]. Inoculation of healthy mice with B. bifidum
MIMBb75 for two weeks, altered microbiota composition at different intestinal loci including
reduction of Clostridium coccoides in the ceacum, increase of total bacteria in the proximal
colon and increase of bifidobacteria in the proximal and distal colon [180]. Moreover,
colonization of B. bifidum may impart nutritional benefits to healthy individuals. Fermentation of
non-digestible oligosaccharides by B. bifidum can result in production of SCFAs such as
butyrate, which is a principle energy source supporting colonocyte growth [221-223]. Another
nutritional benefit of B. bifidum springs from its biosynthesis of vitamins that cannot be
endogenously generated by human, especially B group vitamins. In fact, B. bifidum is one of the
25
most effective folate producers within the genus [224, 225], and administration of B. bifidum was
found to significantly improve folate status in animals and humans [226-228].
26
Table 2.1 Clinical Health Benefits of B. bifidum
Conditions Subjects Study Design Agents and Administration
(strains are indicated when provided)
Main Findings Ref.
Disease-related
Acute diarrhea
Hospitalized
infants
Double-blind RCT1,
n=55.
B. bifidum and S. thermophilus
supplemented in infant formula; once
daily for 4447 patient-days
Reduce the incidence of acute
diarrhea.
[18]
Hospitalized
children
Prospective, single-
blind, multicenter
trial, n=209.
B. bifidum, B. longum, L. acidophilus, L.
rhamnosus, Enterococcus faecium, with
fructooligosaccharides; once daily for 5
days.
Reduced duration of diarrhea and
hospitalization.
[229]
Children Single-blind,
multicentre RCT,
n=571.
B. bifidum, L. delbrueckii var bulgaricus,
Streptococcus thermophilus, and L.
acidophilus, once daily for 5 days.
Reduced duration and severity of
diarrhea.
[17]
Pathogen
Infection
Candida infection
in young patients
with broad
spectrum
antibiotics therapy.
Double-blind RCT,
n=150
B. bifidum, B. longum, L. acidophillus, L.
rhamnosus, Saccharomyces boulardi,
Saccharomyces thermophilus, with
fructooligosaccharides; once daily for 7
days.
Reduced Candida albicans and
Candida tropicalis colonization in
the intestine and candiduria.
[230]
27
Children infected
by H. pylori
RCT, n=100
B. bifidum-12 and L. acidophilus; once
daily for 6 weeks.
Improved H. pylori eradication
and restored altered gut
microbiota; ineffective in
reducing eradication therapy-
associated adverse effects;
[231]
Adult patients with
H. pylori infection
Double-blind RCT,
n=107
B. bifidum, L. acidophilus, L. rhamnosus,
and Streptococcus faecium; once daily for
30 days.
Ineffective in improving efficacy
or reducing adverse effects of
eradication therapy
[232]
Adult patients with
H. pylori infection
Double-blind RCT,
n=30.
B. bifidum CUL17 and Rhodia and
L.acidophilus; once daily starting 7 days
before H. pylori eradication therapy
eradication therapy for 15 days.
A trend to improve microbiota
composition altered by antibiotic
treatment, but not statistically
significant.
[233]
Functional
gastrointestinal
disorders
Adult patients with
IBS2
Prospective, double-
blind, multi-centre
RCT, n=122.
B. bifidum MIMBb75; once daily for 4
weeks.
Improved the IBS symptoms and
digestive disorder.
[23]
Adult patients with
IBS
Double-blind
placebo-controlled
trial, n=52.
B. bifidum CUL20, B. lactis CUL34, L.
acidophilus CUL60 and L. acidophilus
CUL21; once daily for 8 weeks.
Improved IBS symptoms. [22]
Adult patients with Prospective, double- B. bifidum BGN4, B. lactis AD011, L. Improved IBS symptoms with no [21]
28
IBS blind RCT, n=70
acidophilus AD031, and L. casei IBS041;
twice daily for 8 weeks.
adverse effects.
Adult patients with
gastric and lower
abdominal
symptoms
Double-blind,
controlled, crossover
trial, n=27
B. bifidum YIT 10347 and Streptococcus
thermophilus YIT 2021 fermented milk;
once daily for 2 weeks, followed by
crossover for 3 weeks after washout
period of 3 weeks.
Reduced gastric symptoms. [234]
Children and adult
patients with
functional
gastrointestinal
disorders
Open-label clinical
trial, n=37
B. bifidum YIT 10347 and Streptococcus
thermophilus YIT 2021 fermented milk;
once daily for 4 weeks.
Improved gastrointestinal
symptoms and reduced
psychological stress.
[235]
NEC3
Preterm infants
with very low birth
weight
Prospective, double-
blind, multicenter
RCT, n=434.
B. bifidum NCDO 1453 and L.
acidophilus NCDO 1748; twice daily for
6 weeks.
Reduced incidence of death and
NEC.
[204]
Preterm infants
with very low birth
weight
Double-blind RCT,
n=145.
B. bifidus, B. infantis, and Streptococcus
thermophilus; once daily until 36 weeks
postconception.
Reduced both the incidence and
severity of NEC.
[20]
Preterm infants
with very low birth
weight
Prospective, double-
blind RCT, n=186.
B. bifidum, B. infantis, B. longum and L.
acidophilus; twice daily.
Reduce NEC-associated
morbidity and duration of
hospitalization, no significant
difference in NEC severity.
[205]
Allergy
Pediatric patients
with atopic
Double-blind RCT,
n=40.
B. bifidum, L. acidophilus, L. casei, and L.
salivarius; once daily for 8 weeks.
Reduced atopic dermatitis
severity and serum IL-5, IL-6,
[207]
29
dermatitis IFN-γ, and IgE levels.
Infants born to
mothers with a
family history of
allergic diseases
Double-blind RCT,
n=68.
B. bifidum BGN4, B. lactis AD011, and L.
acidophilus AD031; once daily to
pregnant women starting at 4–8 weeks
before delivery and continuing until 6
months after delivery.
Reduced incidence of eczema in
infants at 1 year-old; no
significant difference in serum
IgE level.
[209]
Children with mild
to moderate
asthma
Placebo-controlled
trial, n= 46.
B. bifidum, L. acidophilus, and L.
delbrueckii subsp. bulgaricus; once daily
for 12 weeks.
Reduced episodes of asthma
exacerbations and improved lung
function.
[208]
Others
HIV-infected
children
Double-blind RCT,
n=77.
B. bifidum and Streptococcus
thermophilus; once daily for 2 months. Ineffective in improving diarrhea
episodes.
[236]
Adult patients with
non-alcoholic
steatohepatitis
Open-label RCT,
n=20
B. bifidum, L. plantarum, L. acidophilus,
L. rhamnosus; once daily for 6 months.
Reduced liver fat (ie. intrahepatic
triglyceride content), and serum
aspartate aminotransferase level.
[237]
Institutionalized
elderly
Open-label RCT,
n=25
B. bifidum; once daily for 28 days.
Reduced colonic inflammatory
infiltration, chronic inflammation
in the sigmoid colon.
[238]
30
Health
Pathogen
Infection
(prevention)
Healthy
schoolchildren
Double-blind RCT,
n=80.
B. bifidum and L. acidophilus; twice daily
for 3 months.
Reduced episodes of common
cold and related symptoms (fever,
cough, rhinorrhea), but no
significant benefit on diarrhea and
vomiting.
[217]
University students
with academic
stress
Prospective, double-
blind RCT, n=583
B. bifidum R0071
B.longum ssp. infantis R0033,
Lactobacillus helveticus R0052; once
daily for 6 weeks.
Reduced cold/flu episodes during
acute stress.
[239]
Healthy adults Double-blind RCT,
n=179.
B. bifidum MF, B. longum SP, and L.
gasseri PA; once daily for 3-5 months.
Reduced the severity of symptoms
and shortened common cold
episodes.
[218]
Healthy adult
volunteers
Double-blind RCT,
n=80.
B. bifidum BF-1 and Streptococcus
thermophilus YIT 2021 fermented milk;
once daily for 12 weeks.
Reduced values of biomarkers for
inflammation associated with H.
pylori infection and improved
upper gastrointestinal health.
[240]
1RCT, randomized controlled trial;
2IBS, irritable bowel syndrome;
3NEC, necrotizing enterocolitis.
31
2.4 MicroRNA
MicroRNAs (miRNAs) are 20-22 nucleotides long, single-stranded, non-coding RNA molecules
involved in posttranscriptional gene regulation by targeting mRNA transcripts [241]. In animals,
genes encoding miRNAs are transcribed by RNA polymerase II into primary precursor miRNAs
(pri-miRNAs), which are then processed by the enzyme Drosha into hairpin-shaped precursor
miRNAs (pre-miRNAs). Through Exportin-5, pre-miRNAs are then exported to the cytoplasm,
where they are further cleaved by the enzyme Dicer into a miRNA duplex. Only one of the
strains of the duplex becomes a mature miRNA (miR). miR can be incorporated into the RNA-
induced silencing complex (RISC) and recognize complementary binding sites on the 3’
untranslated region (UTR) of target messenger RNA (mRNA) transcripts. Depending on the
degree of sequence complementarity, this may lead to mRNA translational repression or
degradation [242]. To date, 2,588 miRNAs have been identified in humans and 1,915 miRNAs in
mice (miRBase release 21, June 2014) [36]. In fact, up to 60% of the human transcriptome is
regulated by miRNAs [243]. This potent impact of miRNA in regulating gene expression is
attributable to the fact that each miRNA can have hundreds of mRNA targets while the same
target can be regulated by multiple miRNAs. Different regulatory network models have been
proposed to illustrate how miRNAs may exert cooperative actions to fine-tune gene expression
[244]. These include the linear model where one miRNA determines one phenotype, the
divergent model where one miRNA regulates genes involved in various functions, and the
network model where multiple miRNAs can target multiple genes in a cooperative manner to
determine one physiological outcome (Figure 2.2) [244]. A wide range of biological processes,
such as cell cycle [245] and immune responses [246], can be influenced by miRNA regulation.
Given the important role of miRNAs, it is not surprising that the expression of miRNA itself is
tightly regulated in the context of tissue type and developmental stage [242]. It is also known
that various stimuli can influence miRNA expression in a rapid and time-dependent fashion [247,
248]. For instance, exposing mouse lung tissue to the proinflammatory endotoxin LPS resulted in
transient global miRNA expression changes, which peaked at 3 hours and drops at 6 hours post-
LPS exposure and correlated with the intensity of the inflammatory response [248]. The
importance of miRNAs in intestinal homeostasis was demonstrated in Dicer1-deficient mice,
32
which exhibited significant intestinal barrier dysfunction and inflammation evidenced by the
decrease of goblet cell and increase of crypt apoptosis and lymphocyte infiltration compared to
the wild-type [249]. It is known that various gut bacteria and pathogens can impinge miRNA
expression, and deregulation of miRNA expression is associated with different pathological
conditions including IBD and colonic malignancy, which will be discussed in the following
sections.
Figure 2.2 Models of miRNA-dependent Regulatory Network.
A representation of different miRNA-mediated regulatory models proposed including the linear model
(miRNA1:Target1, determining phenotype 1), the divergent model (miRNA1:Target 1,2…n, determining
phenotype1,2,3), and the network model (miRNA1,2…n:Target2,3…n, determining phenotype2), modified from
[244].
33
2.4.1 Response of Intestinal miRNA to Bacteria
Two recent studies have demonstrated that the resident microbiota impacts the intestinal miRNA
signature. Singh et al. found that the presence of the gut microbiota influences the mouse caecal
miRNA signature with 16 miRNAs being differentially expressed between germ-free and
conventional mice. These include higher expression of miR-148a and reduced expression of
miR-150 in conventional versus germ-free mice. Putative targets of these microbiota-modulated
miRNAs were found to be involved in the regulation of the barrier function [37], which is in line
with the Dicer1-deficient study discussed above. For example, barrier related genes, C1galt1
(core 1 synthase N-acetylgalactosamine 3-beta-galactosyltransferase 1) and Prkcz (protein kinase
C zeta isoform a), which are putative targets of miR-148a, were deregulated in conditional
Dicer1 knockout mice, suggesting a potential role of microbiota-dependent epigenetic regulation
of host barrier function [37, 249]. Dalmasso et al. found 1 and 8 differentially expressed
miRNAs in the ileum and colon, respectively, of germ-free mice following colonization with
microbes derived from pathogen-free mice; by crossing gene expression data with potential
targets of these miRNAs, the upregulated Abcc3 (ATP-Binding Cassette, Sub-Family C, Member
3), a multidrug resistance-associated gene regulating xenobiotics and endogenous toxins
metabolism, was identified to be a direct target of miR-665, which was downregulated in
response to microbiota colonization in the colon [250]. Together, these studies show that the gut
microbiota is an important contributor to the homeostatic intestinal miRNA signature. Further
substantiation to this concept comes from a study performed with the pathogen Listeria
monocytogenes in germ-free and conventional mice. By comparing miRNA profiles between
germ-free and conventional mice challenged with Listeria monocytogenes, Archambaud et al.
identified 5 microbiota-dependent miRNAs alterations; namely, the decrease of miR-143, miR-
148a, miR-200b, miR-200c, and miR-378 were only observed in the ileum of conventional but
not in germ-free mice infected with Listeria monocytogenes [38]. Crossing bioinformatically
predicted targets of miR-143 and miR-378 with differentially expressed genes altered in response
to Listeria monocytogenes, led to the identification of a miRNA-mRNA network, which infers
possible miRNA-mediated regulation of host gene expression in response to pathogen infection.
Other gastrointestinal pathogen infections have been reported to alter host miRNA expression
including Helicobacter, Salmonella, and Campylobacter, as summarized in Table 2.2.
34
Many of these pathogen-induced miRNA alterations are involved in regulating host immune
responses, especially miR-146a and miR-155. These two miRNAs are consistently upregulated
upon Salmonella [39], Listeria [251], Campylobacter [40], and Helicobacter infection [252,
253], and are recognized as important negative regulators of innate immune response. It is
known that both miR-146a and miR-155 are induced upon activation of NFκB signalling
pathway as LPS-stimulated TLR4 activation resulted in increase expression of these miRNAs in
human monocytes [254]. MiR-146a was later found to be important in promoting immune
tolerance via targeting IRAK1 (IL-1R-associated kinase 1), which is an important adaptor
protein of MyD88, to repress TLR-induced NFκB activity [255-257]. Low IRAK1 protein
expression is essential for the mucosa to develop tolerance to initial colonization of resident
microbe during neonatal life. Chassin et al. showed that oral administration of miR-146a
inhibitor to vaginally delivered neonatal mice rescued mucosal IRAK1 protein expression and
resulted in increased susceptibility to epithelial apoptosis caused by bacterial colonization [257].
MiR-155 has been shown to have both pro- and anti-inflammatory effects as it targets both
suppressors such as Ship1 (Src homology2 domain-containing inositol phosphatase) and inducers
such as Tab2 (TGF-beta activated kinase 1/MAP3K7 binding protein 2) of inflammatory
signalling in immune cells [258-260]. The biological function of miR-155 augmentation during
pathogen infection is unclear but miR-155 deficiency has been shown to compromise host
immunity against infection. Specifically, miR-155 knockout mice were unable to develop
vaccine-induced immunity against H. pylori due to impaired Th1 and Th17 response [261]. In
addition to miR-146a and miR-155, a few other miRNAs were also constantly deregulated
during pathogen infections. Particularly, miR-16 and miR-200a/b/c were found to be consistently
up and down-regulated, respectively, in response to both H. pylori and Listeria infections [38,
251, 262-265]. The increase of miR-16 has recently been suggested to play a role in attenuating
TNBS-induced damages by suppressing TNF-α and Th1/Th17 responses [266], while the
decrease of miR-200a/b/c expression is linked to epithelial-to-mesenchymal transition [267].
Collectively, these findings may depict a plausible consensus of miRNA alterations in various
gastrointestinal pathogen infections. Nevertheless, there is only a few studies that looked at
miRNA alterations in response to pathogen infection, and they are limited to pathogens of the
stomach (H. pylori) and small intestine (Salmonella, Listeria, Campylobacter). Currently, no
data is available with respect to pathogen infections of the colon, such as EHEC. Interestingly,
one study showed that miR-155-deficient mice experienced a prolonged eradication of C.
35
rodentium and had impaired humoral immune response [268], unveiling the potential role of
miRNAs in regulating host response to the infection of this colonic pathogen.
Even fewer studies have looked at the effect of probiotic bacteria on miRNA expression (Table
2.3). One study showed that administration of two Lactobacillus species (L. paracasei and L.
casei) to Listeria monocytogenes infected mice attenuated infection severity as well as reductions
of miR-192, miR-200b and miR-215 levels caused by Listeria monocytogenes [264].
Interestingly, after exposing human dendritic cells to L. rhamnosus GG for 12 hours, expression
levels of miR-155 level was increased while miR-146a was decreased, implying
immunomodulation effects of this probiotic strain [269]. Very recently, a study demonstrated
that supplementing L. rhamnosus GG culture supernatant to mice increases TJ protein occludin
expression by inhibiting miR-122a expression, which protects mice from alcohol-induced
intestinal permeability augmentation and liver injury [270]. Another in vitro study showed that
incubating T84 cells with probiotic E. coli Nissle 1907 enhanced TJ protein expression by
downregulating miR-203, miR-483 and miR-595 [271]. Moreover, a previous study from our
group found that supplementation of probiotic strain B. bifidum MIMBb75 for two days resulted
in significant miRNA signature changes with an increased level of miR-148a observed in mouse
ceacum [272]. Overall, the current data suggests that probiotic bacteria may confer beneficial
effects in restoring pathogen-induced miRNA perturbation and enhancing host defense
mechanism presumably by shaping miRNA expression.
36
Table 2.2 Impacts of Gastrointestinal Pathogens on Host microRNA Expression
Bacterium Study Design Altered miRNA Physiological Outcome Ref.
H. pylori miRNA microarray profiling
of endoscopic gastric mucosa
biopsies from H. pylori-
positive patients and controls
(n=N/A)
↑: miR-223
↓: let-7a/b/d/e/f, miR-101, miR-103, miR-
106b, miR-125a, miR-130a, miR-141, miR-
200a/b/c, miR-203, miR-204, miR-210, miR-
214, miR-31, miR-32, miR-320, miR-375,
miR-377, miR-379, miR-429, miR-455, miR-
491-5p, miR-500, miR-532, miR-652
miRNA alterations are associated
with the degree of neutrophil and
mononuclear cell infiltration with
implications in chronic
inflammation.
[262]
Gastric biopsy from H.
pylori-positive patients and
controls (n=30)
↑: miR-155 miR-155 modulates inflammatory
response during the infection.
[263]
H. pylori 26695 infected with
various gastric epithelial cell
lines (GES-1, AGS and
MKN45) for 24 hours
↑: miR-155 (GES-1, AGS, MKN45); miR-
146a, miR-16 (GES-1)
miR-155 suppresses inflammation
by reducing IL-8 and growth-
related oncogene (GRO)-α
production.
[263]
Gastric biopsy from H.
pylori-positive patients and
controls (n=48)
↑: miR-146a miR-146a modulates
inflammatory response during
infection.
[273]
37
gastric epithelial cell lines
(MNK-45, GES-1, HGC-27,
AGS) infected with H. pylori
26695 for 24 hours
↑: miR-146a miR-146a represses NF-κB
activity by targeting IRAK1 and
TRAF6.
[273]
Listeria
monocytogenes
C57BL/6J female 9-12 week-
old conventional mice
infected with Listeria
monocytogenes EGDe for 24
and 72 hours (ileum)
↓: miR-143, miR-148a, miR-194, miR-200b/c,
and miR-378
Deregulations of miR-143, miR-
148a, miR-200b, miR-200c, and
miR-378 are microbiota-
dependent, in which miR-143 and
miR-378 are involved in a
putative regulatory network
controlling genes associated with
immune response, cell
differentiation, and enzymatic
activities.
[38]
C57BL/6J female 9-12 week-
old humanized mice infected
with Listeria monocytogenes
EGDe for 24 hours (ileum)
↓: miR-192, miR-200b, miR-215 miRNA alterations are involved in
a putative regulatory network
encompassing hundreds of
differentially expressed genes
upon the infection and may play
a role in epithelial cell
differentiation.
[264]
38
Caco-2 cells infected with
Listeria monocytogenes
EGDe for 1 hour
↑: miR-146b, miR-16, miR-155
↓: let-7a, miR-145
miRNA alterations are involved in
modulating inflammatory
response.
[251]
Mouse bone marrow derived
macrophages infected with
Listeria monocytogenes
EGDe for 6 hours
↑: miR-155, miR-146a, miR-125a-3p/5p, miR-
149, miR-147, miR-191, miR-132, miR-497,
miR-200c, miR-139-5p
miRNA alterations are involved in
innate immune defense against the
pathogen.
[265]
Salmonella
enterica
Phagocytic (Mouse RAW
264.7) and non-phagocytic
(Human HeLa) cell lines
infected with Salmonella
enterica serovar
Typhimurium SL1344 for 24
hours
↑: miR-146a/b, miR-155, miR-21(RAW264.7);
miR-1308 (Hela)
↓: let-7a/c/d/f/g/i, miR-98 (RAW264.7); let-
7a/c/d/f/g/i (Hela)
miRNA alterations are involved in
inflammatory response;
downregulation of let-7 family
induces IL-6 and IL-10
production in macrophages and
epithelial cells.
[39]
Human HT-29 infected with
Salmonella enterica serovar
enteritidis SE2472
↑: miR-128, miR-196a, miR-330-3p, miR-20a miR-128 suppresses macrophage
recruitment to infection site by
inhibiting the expression of
macrophage colony-stimulating
factor.
[274]
39
Campylobacter
concisus
Human monocytic leukemia
THP-1 cell line infected with
Campylobacter concisus for
6 hours
↑: miR-146a, miR-29b, miR-3687, miR-3648,
miR-7162, miR-6080, miR-3916, miR-221
↓: miR-4477b, miR-621
miRNA alterations are involved in
immune response and
tumorigenesis.
[40]
40
Table 2.3 Impacts of Probiotics on Host Intestinal microRNA Expression
Bacteria Study Design Regulated miRNA Physiological Outcome Ref.
L. rhamnosus
GG
supernatant
C57BL/6N male 8-10 week-old
mice administered with alcohol and
bacteria-free L. rhamnosus GG
supernatant
↓: miR-122a Downregulation of miR-122a
potentiates TJ protein expression,
improving alcohol-induced barrier
dysfunction.
[270]
L. rhamnosus
GG
Human dendritic cells incubated
with heat-killed Lactobacillus
rhamnosus GG for 12 hours
↑: miR-155
↓: miR-146a
miRNA alterations are involved in
modulating inflammatory response.
[269]
L casei BL23 C57Bl/6J female 9-12 week-old
humanized mice infected with
Listeria monocytogenes EGDe for
24 hours (ileum) after orally treated
with Lactobacillus for 3 days
↑: miR-192, miR-200b, miR-215 Lactobacillus treatments normalize
aberrant miRNA expression induced
by Listeria infection.
[264]
L. paracasei
CNCM I-3689
C57Bl/6J female 9-12 week-old
humanized mice infected with
Listeria monocytogenes EGDe for
24 hours (ileum) after oral probiotic
administration for 3 days
↑:miR-192 Lactobacillus treatments normalize
aberrant miRNA expression induced
by Listeria infection.
[264]
41
E. coli Nissle
1917
Intestinal epithelial cell line T84
incubated with E. coli Nissle 1917
for 3 hours
↓: miR-203, miR-483-3p, miR-595 miRNA downregulations improve
barrier function by modulating the
expression of regulatory and
structural components of TJs.
[271]
B. bifidum
MIMBb75
C57Bl/6J male 7 week-old mice
administered with B. bifidum
MIMBb75 for 2 days (Ceacum)
↑: miR-148a, miR-455 miRNA modulations may have
implications in improving barrier
function.
[272]
42
2.4.2 MiRNA deregulations in IBD and CRC
MiRNA deregulation has also been linked to chronic intestinal disorders. Alterations in miRNA
expression are common in the colon tissue of IBD patients. For example, by comparing miRNA
profiles between active UC and healthy subjects, Wu et al. identified 11 differentially expressed
miRNAs including the increase of miR-21 and decrease of miR-192 [275]. More miRNA
alterations were identified in other studies in colonic biopsy of IBD patients, such as
upregulation of miR-146a in CD [276] and miR-155 in UC [277] and downregulation of miR-
200b in both UC and CD patients [278]. Many of these miRNAs have been shown to play a role
in regulating immune response and barrier function. As mentioned before, NLRs are important
activator of NFκB innate immune response and NOD2 has been shown to be a strong genetic
locus contributing to IBD susceptibility [279]. It was found that miR-192, which is
downregulated in IBD [275], targets NOD2 while overexpressing miR-192 in HCT116 cells was
able to suppress NFκB activation and subsequent proinflammatory cytokine production [280].
MiR-192 has also been found to target macrophage inflammatory peptide-2 alpha, which is a
chemokine expressed by epithelial cells and has elevated levels in IBD [275]. MiRNAs may also
contribute to T-cell differentiation. For example, members in the miR-17-92 cluster (miR-17,
miR-20, and miR-18a, miR-19a, miR-19b and miR-92a) were shown to promote Th1 while
suppress Treg differentiation [281]. Moreover, alteration of miRNAs may contribute to intestinal
barrier dysfunction, a common feature in IBD [282]. The increase of miR-21 has been validated
by several studies on IBD patient colon biopsy [275, 277, 283]. Shi et al. showed that miR-21
knockout mice have improved severity of inflammation and survival rate in DSS-induced colitis
[284]. They further illustrated that overexpressing miR-21 in Caco-2 cells impaired TJ protein
expression accompanied with increased barrier permeability by directly targeting RhoB (Ras
Homolog Family Member B), which is a GTPase important in regulating actin cytoskeleton
organization [285]. In addition, one study found that miR-150 was upregulated in active UC
patients, and overexpression of miR-150 in HT29 cells potentiates apoptosis via inhibiting c-
Myb expression. C-Myb is an important regulator of cell cycle progression, and its
downregulation can lead to cell death, which may account for the loss of epithelium integrity
seen in IBD [286]. However, a recent study found that miR-21 expression varies in different
models of IBD, and miR-21deletion resulted in exacerbation of colitis in TNBS-treated mice.
43
Further work is needed to unravel these divergent functions of miR-21 in different IBD models
[287].
Aberrant miRNA expression patterns have been described in various cancers including CRC, as
evidenced by screening data of miRNA expression profile between tumor and nontumor tissues
using microarray or deep sequencing [288-290]. A recent review of over twenty studies on
miRNA expression in CRC concluded that there is a total of 160 miRNAs differentially
expressed in CRC, highlighting the downregulation of miR-143 and miR-145, and upregulation
of miR-20a and miR-92a as potential biomarkers for CRC [291]. Functional validations of these
findings revealed that miRNAs exert both oncogenic and tumor-suppressive effects in CRC
development and progression of CRC by modulating gene expression. MiRNAs that process
oncogenic effects are commonly denoted as oncomirs [292]. MiR-21 is an oncomir in CRC and
its expression level has been shown to increase in accordance with cancer development stages,
with greater degrees of elevation in advanced carcinomas than precancerous adenomas based on
in situ hybridization results [293]. Increased expression of miR-21 has been functionally verified
to inhibit tumor-suppressor gene expression, such as PTEN (phosphatase and tensin homolog)
[294] and PDCD4 (programmed cell death 4) [295]. Members in the miR-17-92 cluster have also
been shown to be consistently upregulated during the transition from adenomas to carcinomas
and are collectively referred to as oncomir-1 [296, 297]. In CRC, oncomir-1 and its homolog
miR-106a/b are overexpressed and intriguingly, elevations of these miRNAs in fecal samples
containing exfoliated colonocytes have been shown to be sensitive biomarkers for distal colon
cancer detection [298]. It is known that the activation of oncogene c-myc induces miR-17-92
expression, which in turn suppresses the expression of an array of anti-proliferative genes, such
as the cyclin-dependent kinase inhibitor (Cdkn1a) that controls cell cycle progression [299],
Bcl2-like 11 (Bim) [300] that promotes apoptosis and Rho family GTPase 3 (Rnd3) that controls
cytoskeletal dynamics [301]. Hence, increase expression of these oncomirs potentiates
colonocyte proliferation. On the contrary, miRNAs exerting tumor suppressive effects are
constantly downregulated during malignancy. EMT is an important process in the maintenance
of stemness, which allows epithelial cells to acquire mesenchymal cell phenotypes including
losses of cell-cell adhesion, cellular polarity and epithelial cell marker E-cadherin expression
accompanied by increased capacity to migrate and proliferate. However, uncontrolled activation
of EMT is a trigger of cancer progression and metastasis [302]. Low expression of miR-200
44
family (including miR-200a/b/c, miR-141 and miR-429) has recently been demonstrated to be
crucial in triggering CRC metastasis [303-305]. It has been functionally validated in many cell
types that miR-200s target Zeb1/2 (zinc finger E-box binding homeobox 1/2), two potent
transcription factors that promote EMT by repressing E-cadherin transcription [267]. Chen et al.
showed that transfecting miR-200c mimics into SW620 colon cancer cells resulted in lower
expression of Zeb1 mRNA and protein levels, as well as reduced number of migrating cells
[303]. Nonetheless, the upstream events causing the downregulation of miR-200s remain
unclear, but studies revealed that the expression of Zeb1 reciprocally suppresses miR-200s in
colon cancer cell lines. These findings imply that the balance expression between miR-200s and
Zeb1/2 could be a critical switch for metastasis [267]. Many other miRNAs have been
implicated in regulating various aspects of CRC development, and complex miRNA regulatory
networks are likely in place to fine regulate the phenotype. Hence, it is crucial and biologically
relevant to study miRNA fine-tuning effects in the context of a regulatory network.
2.5 Animal Model
The murine model is one of the most commonly used for gastrointestinal studies, especially
when investigating microbiota, pathological disorders and using functional genomics approaches
[199, 306, 307]. There are some anatomical and physiological differences between mice and
humans. Although the mouse gastrointestinal tract is relatively smaller in size than humans’, the
relative size of mouse caecum is larger than humans; the ceacum is the major site of fermentation
in mouse, but for humans fermentation takes place in the colon [308, 309]. In addition, gastric
pH is higher in mouse (about 4.0 in fasted animals vs 2.0 in humans) there is also more water
content per kg body weight in the mouse gastrointestinal tract than in humans [310]. These
differences may have implications in the transit time of food digestion and therefore duration of
interaction with the microbiota. With respect to microbiota composition, both man and mouse
have the same dominant bacterial phyla: Firmicutes, Bacteroidetes and Actinobacteria [311].
This withstanding, at a lower taxonomical level, the genus of Mucispirillum and some selected
bacterial species are exclusive to mice, such as Lactobacillus murinus, while others are of human
origin such as B. bifidum and B. breve [312]. By inoculating a human microbiota into germ-free
mice the humanized gnotobiotic mouse model has been developed and has been shown to
45
recapitulate about 100% and 88% of the human gut microbiota at phylum and genus level,
respectively [309]. With respect to pathological diseases like IBD and infectious diseases, upon
induction mice undergo some degrees of microbiota shift and share similar histological features
and inflammatory profiles as humans, as discussed in previous chapters [307, 313]. Moreover,
mouse and human share significant genetic homology. The protein-coding regions of human and
mouse genomes share approximately 85% similarity [314]. These regions are evolutionarily
conserved for their biological functions. Likewise, the majority of miRNA genes are also
conserved between human and mouse [315]. In fact, evolutionarily conserved miRNA homologs
across species are assigned to the same numerical identifier following the prefixes that
differentiate species such as, hsa-miR-21 in human and mmu-miR-21 in mouse [316].
C57Bl/6J is the most extensively used inbred mouse strain with its complete genome sequenced
[314, 317]. Given that inbred mice exhibit less genetic variability, most of the genetically
engineered mouse models are generated in the C57Bl/6 background. For example, the IL-10-
knock out colitis model was constructed using C57Bl/6 mice [318]. This would be an advantage
for the future studies for implementing genetically manipulated models. More importantly, with
respect to the present study, this strain of mice is resistant to C. rodentium infection compared to
C3H/HeJ [107]. Infection in this strain does not result in extensive mortality and the mice
recover within 3 weeks. This moderate degree of disease severity may help detecting potential
probiotic effects of B. bifidum and would allow studies of C. rodentium clearance. In fact, C57
Bl/6 mice have been used in the literature for probiotic studies in the context of C. rodentium
infection, as described previously [152, 160, 319-323].
46
Chapter 3
Rationale, Hypothesis and Objectives
Rationale:
The continuous crosstalk between the host epithelium and gut microorganisms is essential for
intestinal homeostasis. Presence of harmful bacteria can significantly disrupt host homeostasis.
C. rodentium is a murine-specific enteric pathogen used to model EHEC infection, IBD and CRC
in humans [15]. Colonization of C. rodentium peaks at day 10 post-infection in mouse distal
colon resulting in dysbiosis, intestinal barrier dysfunction, colitis and colonic crypt hyperplasia
[15]. In contrast to pathogen-host interaction, beneficial bacteria-host interaction contributes to
maintenance and restoration of intestinal homeostasis. B. bifidum is an indigenous member of the
human microbiota and a common probiotic. In vitro and in vivo studies have revealed that some
strains of B. bifidum may be beneficial in EHEC infection and IBD. Specifically, B. bifidum
ATCC 29521 limits EHEC attachment and colonization in intestinal cell culture [31]. B. bifidum
S17 [32] and BGN4 [212] were shown to be able to attenuate intestinal pathology in mice with
chemically- and immunologically-induced IBD, respectively . However, no study currently has
looked at the effect of B. bifidum on C. rodentium infection. Based on genome-wide
transcriptome microarray analyses, both host-C. rodentium and host-B. bifidum interaction can
impact of host gut physiology by modulating gene expression [30, 34, 35]. MiRNAs are potent
post-transcriptional regulators of gene expression. Increasing evidence has suggested that host-
microbial interaction can influence mucosal miRNA response. For instance, various enteric
pathogen infections can alter host gastric (H. pylori [262]) and small intestinal (Listeria [38,
264], Salmonella [39], and Campylobacter concisus [40]) miRNA signature. Our research group
previously demonstrated that conventional mice supplemented with B. bifidum MIMBb75
exhibit a different miRNA signature in the caecum [272]. The strain B. bifidum MIMBb75 has
been shown to have strong adhesive ability [29, 179] and improve IBS symptoms in a clinical
trial [23]. Nevertheless, whether the presence of this probiotic strain confers beneficial effects
and modulates miRNA response in the context of C. rodentium infection has yet to be explored.
47
Overall hypothesis:
B. bifidum MIMBb75 administration attenuates C. rodentium infection and modulates the
intestinal microRNA response associated with the infection in mouse.
Objectives:
(1) To determine if C. rodentium infection alters murine colonic miRNA signature and if the
alterations are involved in the host response to the infection (Study 1).
(2) To examine if B. bifidum MIMBb75 administration alleviates intestinal damage and
normalizes miRNA alterations associated with C. rodentium infection (Study 2).
48
Chapter 4
Study 1- Citrobacter rodentium Infection Alters Murine Colonic
microRNA Signature
Study Contributions:
All aspects of the in vivo study and analyses were conducted by the author.
This chapter has been partially reported through poster and oral presentations at the Federation of
American Societies for Experimental Biology (FASEB) Annual Conference (Boston, MA,
March 28-30, 2015), and at the Microbiology and Infectious Diseases Research Day at the
University of Toronto (June 15, 2015).
49
4.1 Abstract
MicroRNAs have been suggested to play a part in the interaction between pathogenic bacteria
and host cells. Citrobacter rodentium is a murine pathogen causing transmissible colonic
hyperplasia and colitis with similar pathogenicity as the foodborne enterohaemorrhagic
Escherichia coli O157:H7 in humans. This study aimed to examine if colonic miRNA signature
is altered during C. rodentium infection. C57Bl6/J male mice were randomized to C. rodentium-
infected or control group, and sacrificed at the peak of infection (10 days post-infection). Crypt
hyperplasia and intestinal inflammation were confirmed by histology and in vivo permeability
test. Colonic RNA was used to profile 578 miRNAs by NanoString technology. Gene targets of
the differentially expressed miRNAs were identified in silico by prediction algorithms and cross-
matching with experimentally verified targets databases. Ninety-one miRNAs were differentially
expressed (p<0.05), with 42 downregulated and 49 upregulated versus control (0.2-8.6 fold), and
infected samples clustered together and separately from the controls based on unsupervised
hierarchical clustering. Moreover, prediction analysis revealed that 865 genes targeted by these
miRNAs are mainly involved in cell cycle and immune response. Key genes relevant to C.
rodentium pathogenesis were examined and an inverse correlation was found between miR-17-
92 related clusters and their experimentally validated gene target Bim, which is an apoptosis
facilitator. This study shows that C. rodentium infection alters the intestinal miRNA signature;
differentially expressed miRNAs may be involved in host hyperplasia response during the
infection.
50
4.2 Introduction
Citrobacter rodentium is a murine-specific enteric pathogen. Similarly to foodborne EHEC in
humans, C. rodentium causes A/E lesions in the host. During infection, mice develop TMCH and
colitis, which are believed to be triggered by the activation of the Wingless (Wnt)/β-catenin and
the Tlr signalling pathways, respectively [15]. Because of the pathological similarities, including
crypt hyperplasia and Th1-dominated inflammation, C. rodentium infection is also used to model
colon tumorigenesis and IBD [15]. The host responds to C. rodentium infection via alteration of
gene expression, which is triggered by the recognition of the pathogen-associated molecular
patterns. Microarray studies found that genes involved in ion transport, epithelial cell
development, and inflammatory responses are altered, such as the decrease of solute carrier
family 26 member 3 (Slc26a3) and B cell leukemia/lymphoma 2-like 11 (Bcl2l11/Bim), and the
increases of Cxcl2 and Cxcl5 expression in mouse colon. As a consequence, there is an
electrolyte imbalance, hyperproliferation, and neutrophil infiltration in the colonic mucosa
during infection [34, 35]. In addition, gene-specific approaches identified that genes endothelial
PAS domain protein 1 (Epas1) and solute carrier family 15 member 1 (Slc15a1) involved in
intestinal barrier function were deregulated [324-326]. However, post-transcriptional regulation
of gene expression has not been thoroughly investigated. MiRNAs are non-coding RNA
molecules working in a cooperative manner to fine-tune gene expression. To date, 2,588
miRNAs have been identified in humans and 1,915 in mice (miRBase release 21, June 2014)
[36] and they are believed to regulate up to 60% of the transcriptome [243]. Recent studies have
suggested that various pathogenic bacteria can influence host miRNA expression in the GI tract.
For instance, H. pylori-positive patients exhibit a different miRNA signature in gastric mucosa
compared to controls [262]. The expression of the miR-200 family is downregulated in Listeria
infected mouse ileal tissue [38, 264], while miR-146 and miR-155 were upregulated in Listeria-
infected human intestinal epithelial caco-2 cells [251] and in Salmonella-infected mouse
phagocytes [39]. Intriguingly, miR-155-deficient mice have been shown to be more susceptible
to C. rodentium infection than wild-type mice [268], suggesting that miRNAs may also play a
role in the colonic infection by C. rodentium. However, a comprehensive analysis of the host
response to this pathogen at the miRNA level is lacking. Thus, the aim of this study was to
examine if C. rodentium infection alters miRNA signature of the mouse colon and if the
alterations are involved in the host response to the infection.
51
4.3 Materials and Methods
Mice
Animal study design and procedures were approved by the animal ethics committee at the
University of Toronto (Animal Use Protocol Number: 20010228) and were in accordance with
the Regulations of the Animals for Research Act in Ontario and the Guidelines of the Canadian
Council on Animal Care. Forty conventional C57Bl6/J male mice, six weeks of age, were
obtained from Jackson Laboratories (Bar Harbor, ME), and housed in filtered cages with sterile
bedding, and sterile chow diet and water ad libitum. Mice were randomized into two groups
(n=20/group), sham infected and C. rodentium infected. Sample size was calculated based on a
priori sample size calculation performed in G*Power 3.1.9.2 [327] (n=9 based on outcomes of
intestinal permeability and crypt hyperplasia in a published paper [323]). Infection was
performed by intra-gastric gavage of 100 µl Luria-Bertani (LB)-cultured Citrobacter rodentium
(16 hours overnight culture, 109
colony forming units (CFU)/ml) or an equal volume of sterile
LB (sham) as previously described by Bhinder et al [328]. Body weights were measured and
freshly passed fecal pellets were collected every other day and just before sacrifice. All mice
were sacrificed on day 10 post-infection (p.i.), which is the peak of infection as discussed in
chapter 2, by cervical dislocation after a brief exposure to carbon dioxide (Figure 4.1); distal
colon (the major site of infection on day 10 p.i. and defined as the distal 3.5 cm of the colon after
excision of the rectum), kidneys, spleen, liver, and adipose tissue were dissected on ice and fixed
in 10% formalin or snap-frozen in liquid nitrogen and stored at -80°C.
52
Figure 4.1 Study Design
Forty male C57BL/6J mice were randomized into a sham infected and a C. rodentium infected group. Infection was
performed by intra-gastric gavage on day 0. Body weight and fecal C. rodentium load were monitored every other
day. All mice were sacrificed at 10 days post infection where their intestinal integrity and intestinal miRNA
expression were examined.
53
C. rodentium culturing and quantification
For gavage, C. rodentium DBS100 (kindly provided by Dr. Philpott, Department of
Immunology, University of Toronto) was grown in LB broth aerobically at 37°C for 16 hours, as
previously described [328]. Viable counts of C. rodentium in liquid culture, fecal and liver
samples were determined by plating on MacConkey Agar (BioShop) after aerobic incubation for
24 hours at 37°C. Colonies were identified based on morphology (round shape with red color at
the centre and white color at the edges) [152].
Histological analysis
Paraffin blocks of the distal colon (the distal 1 cm of the colon after excision of the rectum) were
used for hematoxylin and eosin (H&E) staining. Five μm sections were assessed microscopically
for scoring the severity of mucosal damages using a scale 0-4 (0, no signs of inflammation; 1,
minimal evidence of inflammatory cell infiltration; 2, significant evidence of inflammatory cell
infiltration; 3, significant evidence of inflammation with goblet cell depletion; 4, sever
inflammation characterized by widespread inflammatory cells infiltration, goblet cell depletion
and destruction of the mucosal architecture) as previously described [329]. The depths of at least
5 well-oriented far-separated crypts per section were measured to determine hyperplasia. A
minimum of 1 transverse and 2 longitudinal sections per distal colon and 5 fields per section
(total 15 measurements) were assessed to determine the mean mucosal damage score and crypt
hyperplasia for each mouse. To reduce bias, samples were coded such that the examiner was not
aware of which group a sample belongs to.
RNA extraction
Total RNA was isolated from one-half longitudinal segment of the distal colon tissues using the
mirVana miRNA Isolation Kit (Ambion, Austin, TX, USA), as per the manufacturer’s
instructions, eluted in 50 μl of RNAse-free water and stored at -80°C. Recovered total RNA
concentration and purity were assessed using Thermoscientific’s Nanodrop 1000
Spectrophotometer.
54
NanoString nCounter miRNA profiling and data analysis
The expression of 578 miRNAs was assessed using 100 ng of total RNA (n=6/group) and the
nCounter Mouse miRNA Assay (NanoString Technologies, Seattle, WA) (miRbase built version
15) at the Princess Margaret Genomics Centre, Toronto, Canada. Data were processed with the
nSolver Analysis Software v1.1 (NanoString). Code counts were normalized to the geometric
mean of the top 100 miRNAs and background (Mean negative controls + 2 SD) was subtracted; only
probes that were above the background in 80% of the samples of any of the two groups were
retained. For the purpose of this study, probes detecting more than one miRNAs simultaneously
are counted as one miRNA. The final dataset was log2-transformed before statistical analysis.
The relative expression between groups was expressed as Infected/Sham. Student’s t-test was
used to calculate statistical significance (p<0.05). Unsupervised hierarchical clustering was
performed using MultiExperiment Viewer software (MeV 4.9; Institute of Genomic Research,
Rockville, MD) operating under the R statistical computing environment.
MiRNAs target prediction and pathway analysis
Experimentally validated target genes of the differentially expressed miRNAs were identified in
silico using the miRWalk database [330] with a complementary manually curated search. These
gene targets were used for pathway analysis using the PANTHER Classification System version
10.0 [331]. Regulatory networks of miRNA-mRNA targets were visualized using Cytoscape
3.2.1 [332].
Real-Time quantitative PCR (qPCR)
Total RNA (10ng) was reverse transcribed with the Taqman® MicroRNA Reverse Transcription
Kit and primers specific for miR-148a (Assay ID: 000470), miR-200a (Assay ID: 000502), miR-
200b (Assay ID: 002251), and the endogenous control snoRNA135 (Assay ID: 001230) (Applied
Biosystems, Foster City, CA). Real-time PCR was then conducted using undiluted cDNA, and
the TaqMan 2X Universal PCR Master Mix, No AmpEraseUNG (Applied Biosystems) in a 10 μl
PCR reaction. Each reaction was run in triplicates in a 384-well optical plate in Applied
Biosystems’ 7900 HT thermocycler using the 9600 emulation mode with an initial hold at 95°C
for 10 minutes followed by 40 cycles of 95°C for 15 seconds, and 60°C for 60 seconds.
55
For gene expression analysis, 2µg of purified total RNA were reverse transcribed with the High
Capacity cDNA RT kit (Applied Biosystems) according to the manufacturer’s instructions. Real-
time PCR was carried out using 100ng of cDNA product in 10µl PCR reaction with TaqMan
Gene Expression Assay (Hprt assay ID: Mm01545399_m1, Zeb1 assay ID: Mm00495564_m1,
Bcl2l11 assay ID: Mm00437796_m1, Ctnnb1 assay ID: Mm00483039_m1, Epas1 assay ID:
Mm01236112_m1, Slc15a1 assay ID: Mm04209483_m1) and TaqMan Gene Expression Master
Mix (Applied Biosystems). Each reaction was run in triplicates in a 384-well optical plate in
Applied Biosystems’ 7900 HT Real-Time PCR machine with default thermocycling conditions.
The 2−ΔΔ
CT method was used to calculate relative expression levels [333] for both miRNA and
mRNA using sno-135 and Hprt as normalizers, respectively. Significance was determined using
QIAGEN 2009 Relative Expression Software Tool (REST) [334].
Fluorescein Isothiocyanate–Dextran in vivo Permeability Assay
On day 10 p.i. overnight fasted mice (10/group) were gavaged with 4 kDa isothiocyanate
(FITC)-dextran (Sigma – Aldrich) at 88 mg/mL in 100 µL of sterile PBS. Four hours after
gavage, blood was collected by cardiac puncture and serum FITC-dextran concentration was
quantified by fluorometry (FusionTM
, PerkinElmer) with excitation wavelength of 485 nm and
emission wavelength of 535 nm as previously described [328].
Statistical Analysis
Student’s t-test was use to determine statistically significant differences in fecal C. rodentium
load and body and organ weights, crypt hyperplasia, colon damage scores, intestinal
permeability, and bacterial translocation between the infected and the sham group. Bacterial
counts were expressed as log10/g of feces. A p-value of p<0.05 was considered as statistically
significant, with an exception in PANTHER pathway analysis where p<0.0001 was considered
as statistically significant. All statistical analyses were performed in GraphPad Prism 6 software
(La Jolla, CA, USA). Analysis of NanoString and qPRC miRNA expression is described above.
56
4.4 Results
C. rodentium infection kinetics and characteristics
C. rodentium was detectable in the feces two days after infection. Viable count continued to
increase until reaching a plateau (109 CFU/g) at day 8-10 p.i. (Figure 4.2a). Infection did not
result in significant body weight change (Figure 4.2b). Spleen weights of the infected mice were
significantly higher than the sham (0.53 ± 0.007 vs. 0.29 ± 0.071 % body weight; p<0.01); there
was no significant difference in kidney, liver and adipose tissue weights between the two groups
(Figure 4.2c).
Intestinal histology and barrier integrity
C. rodentium infected mice exhibited a significant increase in distal colon crypt lengths
compared to the sham (262.3 ± 11.5 vs.186.8 ± 5.2 μm; p<0.0001) (Figure 4.3a). The infection
resulted in significant histological damages to the distal colon (damage score 3.0 ± 0.17 vs. 0.4 ±
0.04; p<0.0001) (Figure 4.3b), as evidenced by significant signs of lymphocyte infiltration,
goblet cell depletion, and even loss of mucosal architecture (Figure 4.3c,d,e,f). C. rodentium
infected mice also experienced loss of intestinal barrier integrity. Intestinal permeability was
significantly higher in the infected mice compared to the sham mice (1042.5 ± 233.4 vs. 443 ±
71.6 ng/ml; p<0.05) on day 10 p.i. (Figure 4.4a). Moreover, C. rodentium was detectable in the
liver of infected but not control mice (102 CFU/g, Figure 4.4b).
Distal colon microRNA signature
One hundred and forty-six probes were detectable in the distal colon. Of these, 91 were
differentially expressed between C. rodentium-infected and sham-infected mice (p<0.05), where
42 were downregulated and 49 were upregulated in response to C. rodentium (0.2-8.6 fold)
(Table 4.1). Some probes can detect more than one miRNAs (eg. mmu-miR-20a+mmu-miR-
20b); one probe is counted as one miRNA in this project. Unsupervised hierarchical clustering
analysis revealed that infected samples clearly clustered together and separately from the
controls based on the expression profiles of these 91 miRNAs (Figure 4.5a). The expression of 3
selected miRNAs (miR-148a, miR-200a, and miR-200b), which were statistically significant
down-regulated by C. rodentium, was validated by qPCR (Figure 4.5b).
57
a.
b.
c.
Figure 4.2 C. rodentium infection kinetics and effect on mouse body and organ weights.
(a) Viable C. rodentium counts in infected mice feces collected every other day post-infection (p.i.); (b) body
weights of sham and infected mice p.i., n=20/group; (c) percent of organ weight per gram of body weight between
sham and infected mice on day 10 p.i., n=10/group. Statistical significance was determined by Student’s t-test, **P
< 0.01.
58
a.
b.
c.
d.
e.
f.
Figure 4.3 C. rodentium induced intestinal lesions, crypt hyperplasia and barrier
dysfunction.
(a) Tissue damage score and (b) crypt length of the distal colon were assessed blindly (n=10/group). Representative
hematoxylin and eosin stained histological slides were taken on day 10 p.i. (c) score 0: no inflammation with normal
crypt and goblet cells (black arrows), (d) Score 1: minimal evidence of inflammatory infiltrate (black circle), (e)
Score 3: significant evidence of inflammatory infiltrate (black circle) with goblet cell depletion, (f) Score 4: severe
inflammation with widespread lymphocyte infiltration, goblet cell depletion, and destruction of architecture (red
arrow). m: mucosa; s: submucosa; ms: muscularis. 400X. Results are expressed as mean ± SE. Statistical
significance was determined by Student’s t-test, ***P < 0.001.
s
m
ms
m s ms
m
s
m
s
59
a.
b.
Figure 4.4 Loss of barrier integrity of C. rodentium infected mice on day 10 p.i.
(a) Infected mice had an increased serum concentration of the orally administered macromolecule FITC-dextran (4-
kDa), n=8-10. Barrier intergrity was also assessed based (b) translocation of C. rodentium to the liver, n=10/group.
Results are expressed as mean ± SE. Statistical significance was determined by Student’s t-test, *P < 0.05. N.D, not
detectable.
60
Table 4.1 Differentially Expressed miRNAs
61
a.
b.
Figure 4.5 C. rodentium infected mice exhibit distinct miRNA signature.
(a) Heat map shows unsupervised hierarchical clustering of the 91 differentially expressed miRNAs between
infected and sham samples, n=6/group; statistical significance was determined by Student’s t-test. The red color
denotes log2 expression level above the mean, and green color denotes log2 expression level below the mean. (b)
qPCR validation of expression levels of selected miRNAs, n=8-9/group (for 2 of the samples in the Sham group and
1 in the CR group the amount of RNA was not sufficient for qPCR). Results are expressed as mean ± SE. The three
miRNAs were significantly decreased in the CR group (*P<0.05, **P < 0.01, ***P < 0.001 based on Student’s t-
test), and there are no statistical significant differences between the qPCR and Nanostring techniques.
62
Identification of gene targets of differentially expressed miRNAs and pathway analysis
Fifty-eight differentially expressed miRNAs with fold change > 1.5 or < 0.67 were carried
forward for gene target analysis. In silico analysis identified 11579 putative and 865
experimentally validated gene targets of these 58 miRNAs. Of the 865 genes, 824 were mapped
to the Gene ontology PANTHER classification system. Gene ontology classification analysis
revealed that several categories were significantly enriched (p<0.0001) among these validated
targets: (1) biological processes, including 55, 53, and 36% of the annotated genes classified into
the metabolic process, cellular process (eg. cell cycle and communication), and biological
regulation categories, respectively (Table 4.2); (2) molecular functions, including 48, 31, and
18% of the annotated genes classified as protein and nucleic acid binding, catalytic activity, and
transcription factor activity genes, respectively (Table 4.3); (3) cellular component, with 37 and
13% of the annotated genes classified to the cell part, and organelle (eg. cytoskeleton) categories
(Table 4.4). The majority of the genes were annotated to distinct pathways (Table 4.5), and
PANTHER Pathway Overrepresentation test revealed that many of the pathways involved in
inflammatory response and cell cycle regulation such as the Gastrin and cholecystokinin receptor
(CCKR), Tlr, TGF-β, apoptosis, and Wnt signalling pathways, are significantly enriched
(p<0.0001) (Figure 4.6).
63
Table 4.2 GO Biological Process (total number of genes: 824; total number process hits: 2158).
Category name (Accession)
# of
genes
Percent of gene hit
against total # genes
Percent of gene hit against
total # Process hits
cellular component organization or
biogenesis (GO:0071840) 74 9.00% 3.40%
cellular process (GO:0009987) 434 52.70% 20.10%
localization (GO:0051179) 116 14.10% 5.40%
apoptotic process (GO:0006915) 59 7.20% 2.70%
reproduction (GO:0000003) 34 4.10% 1.60%
biological regulation (GO:0065007) 300 36.40% 13.90%
response to stimulus (GO:0050896) 177 21.50% 8.20%
developmental process (GO:0032502) 232 28.20% 10.80%
multicellular organismal process
(GO:0032501) 115 14.00% 5.30%
locomotion (GO:0040011) 11 1.30% 0.50%
biological adhesion (GO:0022610) 39 4.70% 1.80%
metabolic process (GO:0008152) 451 54.70% 20.90%
growth (GO:0040007) 1 0.10% 0.00%
immune system process (GO:0002376) 115 14.00% 5.30%
64
Table 4.3 GO Molecular Function (total number of genes: 824; total number process hits:
1062).
Category name (Accession)
# of
genes
Percent of gene hit
against total # genes
Percent of gene hit against
total # Function hits
transporter activity (GO:0005215) 41 5.00% 3.90%
translation regulator activity
(GO:0045182) 6 0.70% 0.60%
protein binding transcription factor activity
(GO:0000988) 8 1.00% 0.80%
enzyme regulator activity (GO:0030234) 50 6.10% 4.70%
catalytic activity (GO:0003824) 256 31.10% 24.10%
receptor activity (GO:0004872) 101 12.30% 9.50%
nucleic acid binding transcription factor
activity (GO:0001071) 145 17.60% 13.70%
antioxidant activity (GO:0016209) 5 0.60% 0.50%
structural molecule activity (GO:0005198) 55 6.70% 5.20%
binding (GO:0005488) 395 47.90% 37.20%
65
Table 4.4 GO Cellular Component (total number of genes: 824; total number process hits:
487).
Category name (Accession)
# of
genes
Percent of gene hit
against total # genes
Percent of gene hit against
total # Component hits
synapse (GO:0045202) 5 0.60% 1.00%
cell junction (GO:0030054) 4 0.50% 0.80%
membrane (GO:0016020) 62 7.50% 12.70%
macromolecular complex (GO:0032991) 40 4.90% 8.20%
extracellular matrix (GO:0031012) 23 2.80% 4.70%
cell part (GO:0044464) 180 21.80% 37.00%
organelle (GO:0043226) 110 13.30% 22.60%
extracellular region (GO:0005576) 63 7.60% 12.90%
66
Table 4.5 Panther Pathways (total number of genes: 824; total number process hits: 895).
Category name (Accession)
# of
genes
Percent of gene hit
against total # genes
Percent of gene hit against
total # Pathway hits
Toll_pathway_drosophila (P06217) 1 0.10% 0.10%
SCW_signaling_pathway (P06216) 1 0.10% 0.10%
DPP_signaling_pathway (P06213) 1 0.10% 0.10%
DPP-SCW_signaling_pathway (P06212) 1 0.10% 0.10%
BMP_signaling_pathway-drosophila
(P06211) 1 0.10% 0.10%
Axon guidance mediated by netrin
(P00009) 1 0.10% 0.10%
Axon guidance mediated by Slit/Robo
(P00008) 4 0.50% 0.40%
Axon guidance mediated by semaphorins
(P00007) 1 0.10% 0.10%
Apoptosis signaling pathway (P00006) 31 3.80% 3.50%
Gonadotropin releasing hormone receptor
pathway (P06664) 63 7.60% 7.00%
Angiogenesis (P00005) 37 4.50% 4.10%
Ornithine degradation (P02758) 1 0.10% 0.10%
Alzheimer disease-presenilin pathway
(P00004) 22 2.70% 2.50%
Alzheimer disease-amyloid secretase
pathway (P00003) 8 1.00% 0.90%
Alpha adrenergic receptor signaling
pathway (P00002) 2 0.20% 0.20%
67
Methylmalonyl pathway (P02755) 1 0.10% 0.10%
Adrenaline and noradrenaline biosynthesis
(P00001) 1 0.10% 0.10%
CCKR signaling map (P06959) 51 6.20% 5.70%
Lysine biosynthesis (P02751) 1 0.10% 0.10%
Ubiquitin proteasome pathway (P00060) 3 0.40% 0.30%
p53 pathway (P00059) 14 1.70% 1.60%
Wnt signaling pathway (P00057) 36 4.40% 4.00%
Heme biosynthesis (P02746) 1 0.10% 0.10%
VEGF signaling pathway (P00056) 14 1.70% 1.60%
Transcription regulation by bZIP
transcription factor (P00055) 2 0.20% 0.20%
Toll receptor signaling pathway (P00054) 18 2.20% 2.00%
Formyltetrahydroformate biosynthesis
(P02743) 2 0.20% 0.20%
T cell activation (P00053) 17 2.10% 1.90%
Folate biosynthesis (P02742) 1 0.10% 0.10%
TGF-beta signaling pathway (P00052) 28 3.40% 3.10%
Plasminogen activating cascade (P00050) 5 0.60% 0.60%
De novo pyrimidine deoxyribonucleotide
biosynthesis (P02739) 2 0.20% 0.20%
Parkinson disease (P00049) 13 1.60% 1.50%
De novo purine biosynthesis (P02738) 3 0.40% 0.30%
PI3 kinase pathway (P00048) 14 1.70% 1.60%
68
PDGF signaling pathway (P00047) 21 2.50% 2.30%
Oxidative stress response (P00046) 9 1.10% 1.00%
Notch signaling pathway (P00045) 10 1.20% 1.10%
Nicotinic acetylcholine receptor signaling
pathway (P00044) 10 1.20% 1.10%
Muscarinic acetylcholine receptor 2 and 4
signaling pathway (P00043) 5 0.60% 0.60%
Muscarinic acetylcholine receptor 1 and 3
signaling pathway (P00042) 7 0.80% 0.80%
Metabotropic glutamate receptor group I
pathway (P00041) 3 0.40% 0.30%
Asparagine and aspartate biosynthesis
(P02730) 1 0.10% 0.10%
Metabotropic glutamate receptor group II
pathway (P00040) 4 0.50% 0.40%
Synaptic_vesicle_trafficking (P05734) 1 0.10% 0.10%
GABA-B_receptor_II_signaling (P05731) 5 0.60% 0.60%
Endogenous_cannabinoid_signaling
(P05730) 2 0.20% 0.20%
Ascorbate degradation (P02729) 1 0.10% 0.10%
Arginine biosynthesis (P02728) 1 0.10% 0.10%
Metabotropic glutamate receptor group III
pathway (P00039) 6 0.70% 0.70%
JAK/STAT signaling pathway (P00038) 6 0.70% 0.70%
Ionotropic glutamate receptor pathway
(P00037) 5 0.60% 0.60%
69
Interleukin signaling pathway (P00036) 24 2.90% 2.70%
Interferon-gamma signaling pathway
(P00035) 7 0.80% 0.80%
Xanthine and guanine salvage pathway
(P02788) 1 0.10% 0.10%
Adenine and hypoxanthine salvage
pathway (P02723) 1 0.10% 0.10%
Integrin signalling pathway (P00034) 23 2.80% 2.60%
Insulin/IGF pathway-protein kinase B
signaling cascade (P00033) 10 1.20% 1.10%
Insulin/IGF pathway-mitogen activated
protein kinase kinase/MAP kinase cascade
(P00032) 14 1.70% 1.60%
p53 pathway feedback loops 2 (P04398) 10 1.20% 1.10%
Inflammation mediated by chemokine and
cytokine signaling pathway (P00031) 40 4.90% 4.50%
p53 pathway by glucose deprivation
(P04397) 4 0.50% 0.40%
Hypoxia response via HIF activation
(P00030) 3 0.40% 0.30%
Vitamin D metabolism and pathway
(P04396) 2 0.20% 0.20%
Thyrotropin-releasing hormone receptor
signaling pathway (P04394) 4 0.50% 0.40%
Thiamine metabolism (P02780) 1 0.10% 0.10%
Ras Pathway (P04393) 17 2.10% 1.90%
70
Oxytocin receptor mediated signaling
pathway (P04391) 3 0.40% 0.30%
Huntington disease (P00029) 19 2.30% 2.10%
Heterotrimeric G-protein signaling
pathway-rod outer segment
phototransduction (P00028) 3 0.40% 0.30%
Heterotrimeric G-protein signaling
pathway-Gq alpha and Go alpha mediated
pathway (P00027) 10 1.20% 1.10%
p38 MAPK pathway (P05918) 14 1.70% 1.60%
Heterotrimeric G-protein signaling
pathway-Gi alpha and Gs alpha mediated
pathway (P00026) 10 1.20% 1.10%
Opioid proopiomelanocortin pathway
(P05917) 4 0.50% 0.40%
Hedgehog signaling pathway (P00025) 4 0.50% 0.40%
Opioid prodynorphin pathway (P05916) 2 0.20% 0.20%
Glycolysis (P00024) 1 0.10% 0.10%
Opioid proenkephalin pathway (P05915) 2 0.20% 0.20%
General transcription regulation (P00023) 1 0.10% 0.10%
Nicotine pharmacodynamics pathway
(P06587) 3 0.40% 0.30%
Enkephalin release (P05913) 5 0.60% 0.60%
FGF signaling pathway (P00021) 21 2.50% 2.30%
Dopamine receptor mediated signaling
pathway (P05912) 5 0.60% 0.60%
71
FAS signaling pathway (P00020) 7 0.80% 0.80%
Angiotensin II-stimulated signaling
through G proteins and beta-arrestin
(P05911) 8 1.00% 0.90%
Histamine H2 receptor mediated signaling
pathway (P04386) 1 0.10% 0.10%
Pyruvate metabolism (P02772) 1 0.10% 0.10%
Histamine H1 receptor mediated signaling
pathway (P04385) 3 0.40% 0.30%
Gamma-aminobutyric acid synthesis
(P04384) 1 0.10% 0.10%
Cortocotropin releasing factor receptor
signaling pathway (P04380) 2 0.20% 0.20%
Endothelin signaling pathway (P00019) 15 1.80% 1.70%
EGF receptor signaling pathway (P00018) 21 2.50% 2.30%
DNA replication (P00017) 2 0.20% 0.20%
Cytoskeletal regulation by Rho GTPase
(P00016) 11 1.30% 1.20%
Purine metabolism (P02769) 1 0.10% 0.10%
Circadian clock system (P00015) 3 0.40% 0.30%
Cholesterol biosynthesis (P00014) 1 0.10% 0.10%
Cell cycle (P00013) 4 0.50% 0.40%
Cadherin signaling pathway (P00012) 15 1.80% 1.70%
Blood coagulation (P00011) 6 0.70% 0.70%
Beta2 adrenergic receptor signaling 1 0.10% 0.10%
72
pathway (P04378)
B cell activation (P00010) 12 1.50% 1.30%
Beta1 adrenergic receptor signaling
pathway (P04377) 1 0.10% 0.10%
5HT4 type receptor mediated signaling
pathway (P04376) 1 0.10% 0.10%
5HT3 type receptor mediated signaling
pathway (P04375) 1 0.10% 0.10%
5HT2 type receptor mediated signaling
pathway (P04374) 5 0.60% 0.60%
5HT1 type receptor mediated signaling
pathway (P04373) 4 0.50% 0.40%
5-Hydroxytryptamine degredation
(P04372) 1 0.10% 0.10%
73
Figure 4.6 Enriched signaling pathways among miRNA-regulated gene targets.
In silico analysis identified 865 genes as experimentally validated targets of the top differentially expressed miRNA
(fold-change > 1.5 or < 0.67). Listed pathways were significantly (p<0.0001) enriched among these genes based on
the PANTHER Pathways Overrepresentation Test version 10.0. s.p., signalling pathway.
74
Regulatory networks of miRNA-gene targets
A regulatory network of 6 miRNAs of interest and their experimentally verified targets involved
in the Tlr and Wnt signalling pathways was constructed (Figure 4.7). These selected miRNAs are
highly expressed with large fold-changes based on NanoString data, and have been previously
implicated to play a role in host-microbial interaction, inflammatory response and cell cycle
regulation based on the manual curated search and as discussed in Chapter 2.4. Each connection
in the network represents a putative regulatory relationship between miRNA and mRNA during
C. rodentium infection in mouse colon. The network suggests that the selected miRNAs
cooperate to impact the two important pathways in C. rodentium infection. In particular, miR-
146a targets largely Tlr genes, whereas miR-21 and miR-200s predominately target genes of the
Wnt pathway. Four selected miRNA:mRNA pairs (miR-200a/b:Zeb1; miR-148a/152:Ctnnb1,
200a/b:Ctnnb1; miR-146a:Tlr4) did not show inverse correlation, since the gene targets were
downregulated (Zeb1: 0.37 fold, Ctnnb1: 0.58 fold) or unchanged (Tlr4) in infected mice (Figure
4.8). Because the currently available databases are incomplete, additional pairs of interest were
identified by manually curated search. These include: miR-148a:Epas1 and miR193:Slc15a1
(fold-change 0.30 and 0.07) (Figure 4.8) but these pairs again did not show inverse correlation.
75
Figure 4.7 Putative regulatory network of selected miRNAs.
A regulatory network linking selected miRNAs to their experimentally verified gene targets. Purple ovals are genes
associated with the Wnt pathway, and yellow circles are genes associated with the TLR pathway. Red and green
rectangles are upregulated and downregulated miRNAs by C. rodentium, respectively.
76
Figure 4.8 Expression of selected genes in distal colon of sham and infected mice.
Gene expression was evaluated by qPCR (n=7-9/group, for 3 of the samples in the Sham group and 1 in the CR
group the amount of RNA was not sufficient for qPCR). Data are presented as relative fold change in infected mice
versus sham mice. Delta CTs were normalized to Hprt1 (except for Tlr4 that was normalized to β-actin). The tested
genes were all significantly downregulated in C. rodentium infected mice based on the 2−ΔΔ
CT method. Statistical
significance was determined by REST, ***P < 0.001.
77
Finally, crossing the 865 experimentally validated gene targets of the deregulated miRNAs in the
present study with published DNA microarray data of differentially expressed genes in C.
rodentium-infected mouse colon from two independent studies [34, 35] revealed 103 overlapping
genes. Of these genes, Bim was of interest for its expression correlated inversely with a number
of highly upregulated miRNAs targeting it. More importantly, Bim is annotated to apoptosis
process under the biological process category; although it is not currently annotated into any
signaling pathways on PANTHER, Bim is a potential downstream effector of the enriched CCK
[335-337] and the apoptosis [338] signaling pathways, based on literature search. The
downregulation of Bim (0.49 fold) was validated in the present study by qPCR. Based on
expression levels of miRNAs targeting Bim a regulatory network was defined (Figure 4.9). The
network illustrates miRNA regulation of Bim expression and reveals the simultaneous impact of
the differentially expressed miRNAs on the expression of this gene with a net suppressive effect
on Bim expression during C. rodentium infection. Specifically, miR-92a (8.6 fold), miR-106/17
(3.0 fold), miR-19a (2.9 fold), miR-20a/b (2.9 fold), miR-93 (2.0 fold), miR-19b (1.5 fold), miR-
130b (1.5 fold), miR-32 (1.5 fold), miR-25 (1.3 fold) were upregulated and miR-148a (0.3 fold)
and miR-181a (0.6 fold) were downregulated by C. rodentium, suggesting that Bim is a key
candidate gene target of miRNA-mediated response of the host to C. rodentium infection.
Notably, most of these overexpressed miRNAs belong to the miR-17-92 cluster (miR-17, -18a, -
19a, -20a, -19b-1, -92a-1), and its two paralogues miR-106a-363 cluster (miR-106a, -18b, -20b, -
19b-2, -92a-2, -363) and miR-106b-25 cluster (miR-106b, -93, -25), which together will be
referred to as miR-17-92-related clusters from this point on.
78
Figure 4.9 Action of 11 differentially expressed miRNAs on Bim.
Differentially expressed miRNAs that have been experimentally verified to directly target Bim are shown in
rectangles. The color intensity represents miRNA fold-change in infected mice versus sham mice, where red and
green denote over- and under-expression, respectively, in C. rodentium infected mice.
79
4.5 Discussion
Alteration of the transcriptome underlies the response of the host to C. rodentium infection. This
study explored the impact of this pathogen at the post-transcriptional level. It was found that C.
rodentium infection alters the colonic miRNA signature with implications for several pathways
including Wnt, Tlr and apoptosis.
There are only a limited number of studies investigating miRNA expression in response to an
enteric pathogen infection. These include Salmonella [39], Listeria [251] and Campylobacter
infection [40], which, unlike C. rodentium, are pathogens of the small and but not large intestine.
Though, comparing results from the present study with these studies, miR-146 and miR-16 (all
up-regulated), and miR-200a/b and miR-148a (all down-regulated) are also been found to be
differentially expressed in a consistent manner in at least one other enteric pathogen infection. In
particular, miR-146 and miR-16 are upregulated in H. pylori-infected gastric epithelial cells
[263, 273] and Listeria-infected intestinal epithelial Caco-2 cells [251]; miR-200a/b are
downregulated in H. pyori-infected patients’ gastric mucosa [262] and Listeria-infected mouse
ileal tissue [38, 264]; miR-148a is downregulated in Listeria-infected mouse ileal tissue[38].
Thus, there might be a consensus of pathogen-induced miRNA alterations at different regions of
the GI tract. However, this requires further examination, since most of the current published
studies did not examine miRNA profile in a comprehensive manner and many of our
differentially expressed miRNAs were not included in those analyses. Indeed, it might not be
surprising that miR-146 is consistently up-regulated in response to all of these pathogens, as it is
a well-established negative regulator of innate immune response and its expression is induced
upon Tlr signal activation. The increase of miR-146 is associated with improved immune
tolerance by suppressing immune-related genes like interleukin 1 receptor-associated kinase
(Irak1) and Tumor necrosis factor receptor-associated factor 6 (Traf6) [254, 256, 257, 339, 340].
Although in the present study the increase of miR-146a did not significantly suppress Tlr4
expression, it does not exclude that this gene is regulated by the miRNA. It is known that C.
rodentium activates host innate immune response via Tlr4 signaling cascade [120, 122, 123], and
Tlr4 expression has protective effect against Gram-negative pathogens such as Salmonella and
uropahogenic E. coli [341, 342]. However, C. rodentium infection did not induce Tlr4 expression
changes in the present study. This is in line with Khan et al. who previously suggested that Tlr4
expression in C. rodentium was a maladaptive response that is dispensable for the host defense
80
against the pathogen [122]. This was based on the finding that Tlr4-deficient mice have
attenuated disease severity including C. rodentium colonization, hyperplasia, and weight loss
compared to the wild-type. They also observed that Tlr4 mRNA was expressed at comparable
levels on day 6 and 10 p.i. [122]. This is in line with our results, indicating that Tlr4 expression
might be tightly regulated as its expression does not seem to confer benefit to the host during
infection. Hence, it can be postulated that miR-146a is perhaps involved in this tight regulation
of Tlr4 level. Furthermore, some of these miRNA alterations might be microbiota-dependent, as
it was found that the decreases of miR-148a and miR-200b expression were only observed in
conventional, but nor germ-free, mice upon L. monocytogenes infection [38]. Similarly, it is
known that the expression of miR-148a in the intestine depends on the presence of the
microbiota [37]. A few studies showed that C. rodentium infection can be mitigated by probiotic
administration [152, 319-323] and one of the hallmarks of probiotic action is their effect on
microbiota composition. Thus, miRNA regulation may be one of the mechanisms underlying
beneficial effects of probiotics in C. rodentium infection. However, it is important for future
studies to consider the incorporation of control groups with chemically induced inflammation,
such as DSS-induced colitis, to further isolate miRNA response specifically due to C. rodentium
infection.
C. rodentium causes intestinal inflammation and hyperplasia that resemble IBD and colon
tumorigenesis [343]. Interestingly, among the 58 top-regulated miRNAs, 8 (miR-200b, miR-192,
miR-16, miR-21, miR-146a, miR-93, miR-132 and mir-106a+miR-17) [344, 345] and 16
miRNAs (miR-26a, miR-200a, miR-200b, miR-148a, miR-152, miR-192, miR-194, miR-150,
miR-21, miR-146a, miR-223, miR-92a, miR-106a+miR-17, miR-20a+miR-20b, miR-19a, miR-
93) [346-352] are consistently modulated in these two pathologies, respectively. Among the
above, miR-21 is an oncomir associated with IBD. Overexpression of miR-21 was seen in IL-10
knockout mice[287], DSS-treated mice[284] and mucosal biopsies of UC patients[285]. Shi et al.
showed that knocking out miR-21 in DSS-treated mice improves survival rate and attenuates
inflammation [284]. Intestinal barrier dysfunction is a marked characteristic of IBD as well as C.
rodentium infection. Yang et al. found in Caco-2 cells that increased expression of miR-21 can
impair the tight junction gene, RhoB, leading to increased intestinal permeability [285]. In this
study, paracellular permeability was increased in response to C. rodentium, suggesting that miR-
21 may play a role in this phenotypic outcome. MiR-93 and miR-106a, together with miR-17,
81
miR-19a, miR-20a/b, miR-92a and miR-93 belong to the miR-17-92-related clusters, which are
highly up-regulated clusters by C. rodentium. Overexpression of miR-93 and miR-106a are
believed to disrupt autophagy events and pathogen clearance, resulting in chronic inflammatory
state seen in IBD [353, 354]. The miR-17-92 cluster is also known as oncomir-1 [355] and its
overexpression promotes proliferation and reduces apoptosis of epithelial cells in CRC [356-
358]. Thus, the increased expression of these clusters may have implications in inflammatory
response as well as crypt hyperplasia during C. rodentium infection. Moreover, miR-200 family
is one of the top downregulated miRNAs in our study. This family inhibits epithelial-
mesenchymal-transition (EMT) in CRC by directly targeting the zinc finger E-box-binding
homeobox (Zeb) to maintain expression of the epithelial marker E-cadherin [359]. The
downregulation of miR-200 family will result in loss of inhibition on Zeb, followed by loss of
epithelial phenotype and cell-cell adhesion, contributing to tumorigenesis and metastasis [305,
360, 361]. Hence, reduced levels of miR-200a/b in C. rodentium infection may also contribute to
the hyperproliferation of TA cells, which do not possess epithelial phenotype and can proliferate
without being restricted by cell-cell contact [15]. Generally speaking, the fact that some miRNA
alterations induced by C. rodentium are commonly seen in both IBD and CRC may reflect the
fundamental pathological similarities between these conditions and may infer a regulatory role of
miRNAs in the pathogenesis of these conditions.
Pathway analysis of the miRNA targets revealed a number of enriched pathways, with apoptosis,
Tlr, and Wnt signaling pathways being of particular interest in the context of C. rodentium
pathology. Based on the findings discussed above, miR-21, miR-146a, miR-148a, miR-200a/b,
and miR-17-92 cluster were identified as candidate miRNAs involved in regulating these
pathways. Potential regulatory networks of the selected miRNAs were generated for the three
enriched pathways based on bioinformatics in order to identify differentially expressed genes
corresponding to miRNA alterations. In these networks, genes, including Bim, Bcl2, Zeb1/2,
Ctnnb1, and Apc, have been previously shown to be deregulated in tumorigenesis. It is worth
noticing that pathway analysis based on the available databases is often limited due to the
inherent bias in the amount of data available for different biological pathways. Some pathways,
for example cancer-related, are better studied than others; thus, genes in those pathways might be
more represented in the current databases. In addition, since the current databases are still being
optimized, genes can often be misclassified. For example, Bim belongs to the Bcl2 family of
82
apoptosis-related genes and functions as an apoptosis facilitator [338], but it has not yet been
categorized into the apoptosis regulator pathway on PANTHER Classification System.
Therefore, one limitation of using bioinformatics analysis in this study is that confounding
results can be produced while important information can be lost. With that said, the current study
incorporates an extensive manual literature search to complement these disadvantages. Several
miRNA:mRNA pairs are of interest in the context of C. rodentium infection. These include miR-
200a/b:Zeb1; miR-148a/152:Ctnnb1, 200a/b:Ctnnb1; miR-146a:Tlr4; miR-148a:Epas1 and miR-
193:Slc15a1; miR-17-92-related clusters:Bim. While Zeb1, Ctnnb1, EPAS1 and Slc15a1 are
experimentally validated targets of their corresponding miRNA, they were not inversely
regulated upon C. rodentium infection.
C. rodentium infection can also induce hypoxia response. One study by Xue et al. showed that
expression of endothelial PAS domain protein 1(Epas1), a hypoxia-inducible factor, is increased
upon C. rodentium infection, and genetic deletion of Epas1 attenuates inflammation in mouse
colon [324]. On the contrary, the present study revealed that Epas1 was significantly down-
regulated in response to C. rodentium infection. This could be due to fundamental differences
between the two studies. In the study of Xue et al. Epas1 expression was examined 7 days p.i. at
protein level compared to 10 days p.i. at mRNA level in the present study. In the microarray
study by Borenshtein et al. many genes significantly increased on day 4 post-C. rodentium
infection were significantly reduced on day 9 p.i.[34]. Although it requires further validation, it
is possible that Epas1 expression is time-dependent as infection progresses, where by day 10 p.i.
the expression of Epas1 at mRNA level might have been reduced after the initial increase
responsible for the elevation of protein expression observed on day 7 p.i.. It has been shown that
miR-148a and miR-20b directly regulate Epas1 expression in human embryonic stem cells; and
underexpression of both miR-148a and miR-20b increases Epas1 resulting in mesenchymal stem
cell phenotype [362]. Notably, as opposed to miR-148a, miR-20b (a member of miR-106a-363
cluster) is highly up-regulated in the current study. As a matter of fact, Epas1 is a predicted
target of the miR-17-92, miR-106a-363 and miR-106b-25, which are all highly up-regulated.
Hence, miRNA regulation may account for the low expression of Epas1 at mRNA level 10 days
after C. rodentium infection.
Another gene of interest is Slc15a1 (solute carrier family 15 member 1), which functions to
transport bacterial peptide products into epithelial cells to induce inflammation [325]. It has been
83
shown that being a target of miR-193a, Slc15a1 is induced at mRNA and protein levels in mouse
colon 6 hours and 7 days after C. rodentium infection [325, 326]. However, there is a robust
reduction of Slc15a1 mRNA on day 10 p.i. with a decrease of miR-193a expression in the
current study. This could be a consequence of the activation of inflammation resolving
mechanisms at the peak of infection (day 10 p.i.) to prevent further Slc15a1 protein elevation. In
fact, the discrepancy between Slc15a1 mRNA and protein levels has been observed at later
stages of inflammation. Dai et al. showed that expression of miR-193a correlated inversely with
Slc15a1 protein but not with mRNA in both UC tissues and 12 weeks after DSS-induced colitis
in mice [326]. Therefore, future investigation can examine Slc15a1 expression at both mRNA
and protein levels and at different time points as inflammation progresses. This is a future
direction for other genes, including Zeb1, and Ctnnb1, whose mRNA levels did not show an
inverse relationship with miRNA expression, to elucidate the timing effects of miRNA
regulation on gene expression.
On the contrary the miR-17-92-related clusters:Bim pair displayed an inverse correlation (Figure
4.9). Bim is a potent downstream effector of the TGF-β pathway that promotes physiological
apoptosis in various tissues. Down-regulation of Bim has been associated with resistance to TGF-
β-induced apoptosis and thereby tumorigenesis [363-365]. Sinicrope et al. showed that loss of
Bim expression is linked to poor overall survival rates in patients with colon carcinomas [366].
Low expression of Bim at mRNA and protein levels has also been implicated in skin and renal
carcinoma; therefore, Bim is recognized as a tumor suppressor gene [367, 368]. Here, Bim is
significantly down-regulated. This is in line with published genome-wide microarray data, which
revealed that Bim is one of the significantly down-regulated genes in mouse colon 9-day post C.
rodentium infection [34]. The mechanisms underlying Bim suppression in cancers and C.
rodentium infection are unclear. Several studies have verified that Bim is a direct target of
members in the miR-17-92, miR-106a-363 and miR-106b-25 clusters, as well as miR-130b and
miR-32 [369-371], In particular, Tsuchida et al. determined that high expression of miR-92a
promotes malignant transformation in CRC development by directly suppressing Bim expression
[300]; other studies showed that over-expressing miR-25 or miR-130b in gastric cancer cell lines
impairs TGF-β-induced apoptosis via targeting Bim [252, 370]. Remarkably, all of these
miRNAs targeting Bim were up-regulated in C. rodentium infection. Moreover, Bim has also
been shown to regulate immune responses, where Bim-deficient mice were not able to terminate
84
T-cell-mediated responses after viral infection [372, 373]. A recent study found that high
expression of miR-148a in T-lymphocytes upon viral infection reduces Bim expression and
contributes to persistent immune responses in chronic inflammation [374]. In contrast to this
study, miR-148a was down-regulated in response to C. rodentium infection. Yet, C. rodentium
infection does not provoke chronic inflammation and it is unknown whether miR-148a also
regulates Bim in colonic epithelial cells. As Figure 4.9 demonstrates, the effect of the up-
regulated miRNAs is likely to overrule that of the down-regulated miRNAs based on the number
and fold-change of the up- and down-regulated miRNAs targeting Bim, resulting in an overall
down-regulation of Bim in C. rodentium infection. Therefore, these findings suggest that the loss
of Bim-induced apoptosis may contribute to hyperproliferation of the colonic cells in response to
C. rodentium infection and this process is regulated post-transcriptionally via the coordination of
multiple miRNAs.
Taken together, this study shows that C. rodentium–infected mice displayed an altered miRNA
signature. The miRNA alterations are involved the host response to the infection; particularly,
physiological apoptosis is disrupted as a result of miRNA post-transcriptional regulation of the
apoptosis facilitator Bim.
85
Chapter 5
Study 2- Effects of Bifidobacterium bifidum on Citrobacter
rodentium Infection via microRNA Modulation
Study Contributions:
The quantitative data for fecal C. rodentium load in Figure 5.2a titled “C. rodentium infection
kinetics and effect on mouse body and organ weights”, and the data for C. rodentium
translocation to liver and spleen in Figure 5.5b titled “B. bifidum effects on intestinal barrier at
day 10 p.i.” were provided by a fourth-year summer research project student, Sofia Sagaidak.
The quantitative data for mouse body weights and organ weights in Figure 5.2b,c titled “C.
rodentium infection kinetics and effect on mouse body and organ weights”, and the data for
caecal crypt lengths and damage score in Figure 5.4a,c titled “B. bifidum effects on intestinal
crypt hyperplasia and tissue damage at day 10 p.i.” were partially contributed by a fourth-year
summer research project student, Alex Lee.
86
5.1 Abstract
B. bifidum is a common probiotic. Various strains of B. bifidum have been shown to confer
beneficial effects against pathogen infection and inflammation in animal studies. Administration
of B. bifidum MIMBb75, a highly adhesive strain, was found to be able to influence miRNA
expression in mouse caecum. This study aimed to examine if administration of B. bifidum
MIMBb75 attenuates pathology and miRNA alterations associated with C. rodentium infection.
C57Bl6/J male mice were randomized into four groups, sham, C. rodentium infection, C.
rodentium infection with daily B. bifidum administration initiated on the same days as infection,
and C. rodentium infection with daily B. bifidum administration initiated 7 days before infection.
Mice were sacrificed at day 10 post-infection. Fecal C. rodentium and B. bifidum fecal load were
monitored throughout the study. Crypt hyperplasia and intestinal inflammation were examined
by histology and in vivo intestinal permeability test. Expressions of selected miRNAs were
quantified by real-time PCR in the proximal colon. Presence of B. bifidum did not prevent or
reduce C. rodentium colonization. Intestinal permeability was not different among the groups. B.
bifidum treatment did not attenuate pathology nor normalize miRNA alterations associated with
C. rodentium infection.
87
5.2 Introduction
B. bifidum is a Gram-positive anaerobe belonging to the Actinobacteria phylum. B. bifidum is
one of the first colonizers of the infant gut [177] and members of this species have many health-
promoting effects, including microbiota composition improvement [375], immunomodulation
[376], bacteriocin production [377] and pathogen exclusion [211]. These beneficial effects may
have implications in enteric infection and inflammation. For instance, Bayoumi et al. recently
showed that B. bifidum ATCC 29521 interferes with EHEC attachment and colonization in vitro
[31]. López et al. revealed that HT29 cell cultures treated with B. bifidum LMG13195 display
improved monolayer integrity [27]. It was also reported that oral administration of B. bifidum
BGN4 alleviates lymphocytes infiltration and Th1-type cytokines production in the naive T-cell
transfer mouse model of IBD [212]. B. bifidum MIMBb75 exhibits strong adhesive ability to
intestinal epithelial cells [29, 179] and it was found to have beneficial effects in mitigating
intestinal discomfort of IBS patients [23]. The molecular basis for these beneficial effects
attributed to B. bifidum is largely unknown. It is known that microorganisms influence gut
physiology through host gene expression modulation [33]. For instance, global transcriptome
analysis of mouse intestinal cells revealed that B. bifidum PRL2010 colonization was able to
modulate host innate immune response with enhanced production of IL-6 and IL-8 cytokines
[30]. Recent evidence suggests that some probiotic bacteria, such as Lactobacillus casei [264]
and E. coli Nissle 1907 [271], can influence the expression of host epithelial miRNAs, which are
important posttranscriptional regulators of gene expression. Our research group previously found
that mice supplemented with B. bifidum MIMBb75 exhibit a different miRNA signature in the
caecum, where miR-148a was up-regulated in treated mice [272]. Intriguingly, study 1 in the
present project has demonstrated that C. rodentium infection, a model for IBD and colonic
tumorigenesis, substantially altered miRNA signature in mouse colon, with miR-148a being one
of the top down-regulated miRNAs among other differentially expressed miRNAs such as miR-
200a/b, miR-146a, miR-21 and miR-17-92 cluster. However, whether miRNA-mediated
posttranscriptional regulation play a part in the amelioration effects of B. bifidum MIMBb75 has
not been investigated, especially in the context of enteric inflammation induced by C. rodentium
infection. The aim of study 2 was to explore if daily B. bifidum MIMBb75 administration in mice
before and after C. rodentium infection would alleviate C. rodentium colonization and intestinal
barrier dysfunction and inflammation via modulating host miRNA expression.
88
5.3 Materials and Methods
Mice
Animal study design and procedures were approved by the animal ethics committee at the
University of Toronto (Animal Use Protocol Number: 20010228) and were in accordance with
the Regulations of the Animals for Research Act in Ontario and the Guidelines of the Canadian
Council on Animal Care. Eighty C57Bl6/J male mice, six weeks of age, were obtained as two
staggered batches (n=40/batch) from Jackson Laboratories (Bar Harbor, ME), and housed in
filtered cages with sterile bedding, and sterile chow diet and water ad libitum. Mice in each batch
were randomized into four groups (total n=20/group): (1) sham infected, (2) C. rodentium
infected (CR), (3) C. rodentium infected with daily administration of B. bifidum initiated on the
same day as, but temporally after, the infection (BB post-CR), (4) C. rodentium infected with
daily B. bifidum administration initiated one week (7 days) before the infection (BB pre-post-
CR) (Figure 5.1). Infection was performed by intra-gastric gavage of 100 µl LB-cultured C.
rodentium (16 hours overnight culture, 109
CFU/ml) or an equal volume of sterile LB (Sham) as
previously described [328]. Daily B. bifidum treatment was performed by intra-gastric gavage of
200 µl B. bifidum suspension (109
CFU/ml) or an equal volume of sterile PBS. C. rodentium
infection and sham infection were performed at 9 until 11 AM on the day of infection; gavage of
B. bifidum and PBS (control) were performed at 2 until 4 PM everyday throughout the study.
Body weights were measured and freshly passed fecal pellets were collected on p.i. days 2, 4, 6,
8, 9 and just before sacrifice. A subset of mice (n=16/group) were sacrificed on day 10 p.i.,
which is the peak of infection as previously discussed, by cervical dislocation after a brief
exposure to carbon dioxide; caecum (the initial site of infection about 2-3 days p.i.) and distal
colon (the major site of infection at day 10 p.i. and defined as the distal 3.5 cm of the colon after
excision of the rectum), kidneys, spleen and liver were dissected on ice and fixed in 10%
formalin or snap-frozen in liquid nitrogen and stored at -80°C. To collect preliminary evidence,
four mice from each group were retained to monitor C. rodentium clearance after day 10 p.i. and
thereafter, freshly passed fecal pellets were collected every other day up to day 28 p.i. based on
the infection cycle of C. rodentium previously described where clearance generally takes place
within 2-3 weeks p.i. [104, 105] . Mice were sacrificed when C. rodentium was completely
cleared (C. rodentium load under the detection limit of 3x103 CFU/ml for two consecutive days).
89
Figure 5.1 Study Design.
Eighty male C57BL/6J mice were randomized into four groups after one week of acclimatization. (1) sham infected
(Sham), (2) C. rodentium infected (CR), (3) C. rodentium infected with B. bifidum daily administration initiated on
the same day after the infection (BB post-CR), (4) C. rodentium infected with daily B. bifidum administration
initiated one week (day -7 p.i.) before the infection (BB pre-post-CR). B. bifidum (yellow bars) was administered
from 2 to 4 PM every day by intra-gastric and sterile PBS (blue bars) was used to control for gavage. C. rodentium
(arrows) infection was performed from 9 to 11 AM on day 0 p.i. by intra-gastric gavage. On day 10 p.i., 16
mice/group were sacrificed with a subset (n=8/group) used to perform FITC-dextran intestinal permeability test, and
the rest were used for histology and gene expression analyses. The remaining mice (n=4/group) were kept alive and
used to monitor C. rodentium clearance. Body weights were recorded and fecal samples were collected every other
day throughout the study.
90
Bacteria culturing and quantification
For gavage, C. rodentium DBS100 (kindly provided by Dr. Philpott, Department of
Immunology, University of Toronto) was grown in LB broth aerobically at 37°C for 16 hours, as
previously described [328]. Viable counts of C. rodentium in liquid culture (gavage), fecal and
liver samples were determined by classical culturing on MacConkey Agar (BioShop). After
aerobic incubation for 24 hours at 37°C, C. rodentium colonies were identified based on
morphology (round shape with red color at the centre and white color at the edges) [152].
B. bifidum MIMBb75 was grown anaerobically at 37°C in Man Rogosa Sharpe broth
supplemented with 0.05% L-cysteine hydrochloride (cMRS) for 24 hours. Culture was washed
and re-suspended in sterile pre-reduced PBS (200 µl) immediately before gavage. Viable counts
of B. bifidum in gavage culture were enumerated microscopically and by classical culturing on
L-cysteine supplemented MRS agar. To assess B. bifidum colonization in mice, fecal pellets were
collected from uninfected mice 2 days post-B. bifidum initiation (i.e. BB pre-post-CR) and all
groups on day 9 p.i.. DNA was extracted from fecal pellets using the Omega E.Z.N.A.TM Stool
DNA Isolation Kit as per the manufacturer’s instructions. Quantitative Real-Time PCR was
performed with 50 ng of DNA using the SYBR Green Master Mix (Applied Biosystems,
Carlsbad, CA) with specific primers targeting the BopA gene, which is specific to the B. bifidum
species[29].(Forward: 5’ACCGAATTCGCCTGTCACTT3’; Reverse:
5’ACGGCGCGGATTCGT3’) at optimized concentrations (F-R: 100-100 nM/reaction). To
determine absolute B. bifidum counts, each reaction (10 μl) was run in triplicates in a 384-well
optical plate using a 7900 HT Real-Time PCR machine (Applied Biosystems) with default
conditions. Bacterial counts were determined using pre-constructed standard curves and data
were expressed as log10 CFU per gram of wet feces.
Fluorescein Isothiocyanate–Dextran in vivo Permeability Assay
This assay was conducted as described in Study 1.
Histological analysis
These analyses were conducted as described in Study 1.
91
RNA extraction
Total RNA was isolated from one-half longitudinal segment of the distal colon tissues using the
mirVana miRNA Isolation Kit (Ambion, Austin, TX, USA), as per the manufacturer’s
instructions, eluted in 50 μl of RNAse-free water and stored at -80°C. Recovered total RNA
concentration and purity were assessed using Thermoscientific’s Nanodrop 1000
Spectrophotometer.
Real-Time quantitative PCR (qPCR)
Total RNA (10ng) was reverse transcribed with the Taqman® MicroRNA Reverse Transcription
Kit and primers specific for miR-146a (Assay ID: 000468), miR-148a (Assay ID: 000470), miR-
19a(Assay ID: 000395), miR-200a (Assay ID: 000502), miR-200b (Assay ID: 002251), miR-21
(Assay ID: 000397), and the endogenous control snoRNA135 (Assay ID: 001230) (Applied
Biosystems, Foster City, CA). Real-time PCR was then conducted using undiluted cDNA, and
the TaqMan 2X Universal PCR Master Mix, No AmpEraseUNG (Applied Biosystems) in a 10 μl
PCR reaction. Each reaction was run in triplicates in a 384-well optical plate in Applied
Biosystems’ 7900 HT thermocycler using the 9600 emulation mode with an initial hold at 95°C
for 10 minutes followed by 40 cycles of 95°C for 15 seconds, and 60°C for 60 seconds. The
2−ΔΔ
CT method was used to calculate relative expression levels [333] using sno-135 as a
reference gene. Significance was determined based on ΔCT values using one-way ANOVA,
followed by Bonferroni’s post-hoc test.
Statistical Analysis
One-way ANOVA followed by the Bonferroni’s post-hoc test was used to determine statistically
significant differences in fecal C. rodentium load and body and organ weights, crypt hyperplasia,
colon damage scores, intestinal permeability, and bacterial translocation. Difference of fecal B.
bifidum counts before and after infection, between B. bifidum-treated groups, and between
batches were assessed by Student’s t-test. Bacterial counts were expressed as log10/g of feces.
MiRNA expression among groups was assessed based on ΔCt values using one-way ANOVA,
followed by Bonferroni’s post-hoc test. Outliers were determined by the Grubbs’ Outlier Test. A
p-value of p<0.05 was considered as statistically significant. All statistical analysis were
performed in GraphPad Prism 6 software (La Jolla, CA, USA).
92
5.4 Results
C. rodentium infection kinetics and characteristics
C. rodentium was detectable in the feces two days after infection in all three C. rodentium
infected groups. In all infected groups viable count continued to increase until reaching a plateau
(1010
CFU/g) at day 10 p.i. (Figure 5.2a). Four mice from each group were retained for
monitoring C. rodentium clearance. C. rodentium counts decreased rapidly after the peak and
were below detection limit (103 CFU/g) after day 24 p.i. in most mice, although in a few mice
fecal C. rodentium counts fluctuated after 28 days of infection. However, there was no
significant difference on daily C. rodentium count or bacterial clearance rate among the groups.
Infection did not result in significant body weight change among the groups; growth rate was
slowed down from day 2 to day 8 p.i. in the infected groups compared to the sham, though not
significant (Figure 5.2b). Spleen weights were significantly higher in the infected groups
compared to the sham but did not differ among the infected groups (CR: 0.54 ± 0.075, BB post-
CR: 0.61 ± 0.06, BB pre-post-CR: 0.58 ± 0.07, Sham: 0.28 ± 0.02 % body weight; p<0.01); there
was no significant difference in kidney and liver weights among all groups (Figure 5.2c).
B. bifidum colonization
Before infection, B. bifidum load in feces of the BB pre-post-CR mice after 2 day of
administration was about 107cells/g and there was no batch difference (batch 1: 7.4 ± 0.20, batch
2: 6.9 ± 0.15 log cell/g) (Figure 5.3a). On day 9 p.i., B. bifidum remained detectable in the B.
bifidum-treated groups and there was no significant difference in B. bifidum load between the
two groups (BB post-CR: 7.0 ± 0.4, BB pre-post-CR: 6.7 ± 0.4 log cell/g) (Figure 5.3b). There
was no significant difference of B. bifidum load in group BB pre-post-CR before (-5 p.i.) and
after infection (9 p.i.) (Figure 5.3b). B. bifidum was not detected in CR and Sham, which did not
receive B. bifidum treatment.
93
a.
b.
c.
Figure 5.2 C. rodentium infection kinetics and effect on mouse body and organ weights.
(a) Viable C. rodentium counts in infected mice feces were quantified every other day p.i., n=9-13/group (day 2-10
p.i.), and n=1-4/group (day 12-28 p.i.); (b) body weights among sham and infected groups, including C. rodentium-
infected only (CR), C. rodentium-infected with B. bifidum treatment 10 days post-infection (BB post-CR), and B.
bifidum treatment one week prior to C. rodentium infection and 10 days after infection (BB pre-post-CR),
n=20/group (except for day 10 p.i., only non-fasted mice (n=8/group) used for sacrifice were measured); (c) percent
of organ weight per gram of body weight among sham and infected groups on day 10 p.i., n=8/group. Data are
presented as mean ± SE. Statistical significance was determined by one-way ANOVA followed by Bonferroni’s
post-hoc test. Different superscripts (a, b) indicate statistic significance among groups, p<0.05.
94
a.
b.
Figure 5.3 Fecal B. bifidum load before and after infection.
(a) QPCR quantification of fecal B. bifidum load two days after B. bifidum initiation in the group receiving B.
bifidum prior infection, n=10/batch. (b) qPCR quantification of fecal B. bifidum in all groups of mice after C.
rodentium (CR) infection (n=9/group), and in BB pre-post-CR group before CR infection (n=20/group). Statistical
significance was determined by Student’s t-test, data presented as mean ± SE. N.D, not detectable.
95
Intestinal histology and barrier integrity
C. rodentium infected mice exhibited a significant increase of caecal crypt length compared to
the sham (CR: 117 ± 5.57, BB post-CR: 120 ± 2.90, BB pre-post-CR: 109.7 ± 4.39, vs. Sham:
79.9 ± 2.28 μm; p<0.01) (Figure 5.4a). This is also the case in the distal colon (CR: 232.9 ±
14.61, BB post-CR: 242.2 ± 7.19, BB pre-post-CR: 224.4 ± 14.48, vs. Sham: 160.6 ± 6.59 μm;
p<0.0001) (Figure 5.4b). B. bifidum-treatment did not yield significant difference in caecal or
distal colon crypt lengths compared to the infected CR group without the treatment. Histological
analysis of the ceacum and distal colon tissue revealed significant difference in tissue damage
scores between infected mice and sham mice, but no difference between B. bifidum-treated and
infection-only mice (caecum damage score CR: 1.5 ± 0.2, BB post-CR: 1.3 ± 0.2, BB pre-post-
CR: 1.2 ± 0.2, vs. Sham: 0.3 ± 0.1 μm; distal colon damage score CR: 2.6 ± 0.4, BB post-CR: 3.3
± 0.2, BB pre-post-CR: 3.0 ± 0.3, vs. Sham: 0.6 ± 0.2, p<0.0001) (Figure 5.4c,d). There was no
difference in intestinal permeability measured by the FITC-dextran assay among the four groups
on day 10 p.i. (Figure 5.5a). C. rodentium was detectable in the liver and spleen of infected but
not control mice; but there was no difference in the degree of bacterial translocation with or
without B. bifidum treatment (Figure 5.5b).
Distal colon microRNA expression
Based on results in Study 1, the expression of 6 relevant miRNAs (miR-148a, miR-200a, miR-
200b, miR-19a, miR-21, and miR-146a) was examined (Figure 5.6). MiR-148a, miR-19a were
numerically, but not significantly, reduced and increased, respectively, in the C. rodentium
infected groups. MiR-21 was differentially expressed among the four groups (one-way ANOVA:
p<0.05). Post-hoc Bonferroni’s test revealed that the expression of miR-21 was significantly
elevated in groups BB post-CR (fold=3.4 ± 0.1, p<0.05) and BB pre-post-CR (fold=3.1 ± 0.2,
p<0.05), but not in the CR (fold=2.1 ± 0.2).
96
a.
b.
c.
d.
Figure 5.4 B. bifidum effects on intestinal crypt hyperplasia and tissue damage at day 10
p.i..
(a) Distal colon and (b) ceacal crypt lengths were measured with the examiner unaware of the group of the samples
(n=10/group; 2 additional samples collected from mice used for intestinal permeability test were included); infected
groups showed significantly greater crypt lengths than the uninfected group (p<0.001); B. bifidum treatment did not
influence crypt hyperplasia induced by C. rodentium. (c) Distal colon and (d) ceacal damage score were evaluated in
the same manner as for distal colon (n=10/group; 2 additional samples collected from mice used for intestinal
permeability test were included); infected groups had significantly higher tissue damage scores compared to the
uninfected control (p<0.001); B. bifidum treatment did not influence intestinal colitis induced by C. rodentium.
Results are expressed as mean ± SE. Statistical significance was determined by one-way ANOVA, followed by the
Bonferroni’s post-hoc test, p<0.05. Different superscripts (a, b) indicate statistic significance between groups.
97
a.
b.
Figure 5.5 B. bifidum effects on intestinal barrier at day 10 p.i.
(a) No statistical significance was observed in intestinal permeability among the groups, despite of C. rodentium
infection or B. bifidum treatment, n=8/group. (b) Barrier integrity was also assessed based on C. rodentium
translocation to secondary organs (n=8/group); no difference in C. rodentium load was observed in spleen and liver
among the infected groups, regardless of B. bifidum treatment. Results are expressed as mean ± SE. Statistical
significance was determined by one-way ANOVA. N.D, not detectable.
98
Figure 5.6 Distal colon miRNA expression at day 10 p.i..
Real-time PCR quantification of 6 selected miRNA expression; data were normalized to snoRNA135, n=7-8/group
(one statistical outlier was taken out from the Sham group, with a Ct value comparable to the negative control).
Results are expressed as mean ± SE. Statistical significance was determined based on the ΔCt values using one-way
ANOVA, followed by Bonferroni’s post-hoc test. Different superscripts (a, b) indicate statistical significance
between groups.
99
5.5 Discussion
The current study investigated the impact of B. bifidum MIMBb75 on C. rodentium infection in
the framework of miRNA modulation. It was found that administration of B. bifidum MIMBb75
before and after C. rodentium infection did not improve intestinal pathology (C. rodentium
colonization, crypt hyperplasia, inflammation, and barrier dysfunction) or miRNA alterations
associated with the infection.
Several probiotic strains belonging to the Lactobacillus and Bifidobacterium species and their
fermented products have been implicated in improving C. rodentium induced colitis in vitro and
in vivo [152, 319-323]. Some studies found that the use of probiotic bacteria resulted in a
reduction in bacterial load during C. rodentium infection [152, 319, 322]. For example, treatment
of L. rhamnosus R0011 and L. acidophilus R0052 reduced C. rodentium attachment to T84
epithelial cells in vitro, and reduced C. rodentium load in colonic luminal contents on day 9 p.i.
in C57Bl/6 mice [152]. A 3-day pretreatment of B. breve UCC2003 in BALB/c mice reduced
fecal C. rodentium load from day 3-14 p.i.[319]. On the contrary, other studies revealed that
probiotics may prevent C. rodentium-induced pathology independently from colonization
resistant effects. Particularly, Collins et al. showed that treatment of B. breve UCC2003 for 3
days prior to infection attenuated crypt hyperplasia without affecting C. rodentium colonization
or A/E lesion formation in C57Bl/6 mice [320]. A more recent study conducted by this group
also found that administration of fermented product from Lactobacillus species to C57Bl/6 mice
improved infection but did not alter C. rodentium colonization kinetics [321]. In the present
study, B. bifidum MIMBb75 administration did not suppress C. rodentium colonization nor
improved clearance. The discrepancies in colonization resistance conferred by different
probiotics could be attributed to strain-specific effects of probiotics and variations in study
conditions such as treatment duration and animal strains used. Also, it was of interest to examine
if C. rodentium would in turn compete with B. bifidum MIMBb75 intestinal counts in particular
taking into consideration that B. bifidum MIMBb75 is a human-restricted strain while C
rodentium is only pathogenic in murine. It was found that fecal B. bifidum counts in BB pre-post-
CR group before and after infection (Figure 5.3) were not significantly different and the number
was comparable to data previously obtained in our group where fecal B. bifidum load 24 hours
post-gavage was about 6.8 ± 0.3 log cell/g [378]. This indicates that infection did not interfere
with B. bifidum fecal load. Knowing that C. rodentium can replace almost 90% of the resident
100
microbiota, the fact that B. bifidum MIMBb75 was not displaced indicates that this
bifidobacterium strain is a good colonizer [14]. A previous study from our group also revealed
that B. bifidum predominately resides in the ceacum instead of the distal colon [378], where C.
rodentium preferentially localizes. Hence, it might be of interest for future studies to examine
colonization competition between the two bacteria in a region-specific manner.
The use of selected probiotic bacteria, especially when given before infection, has been shown to
alleviate TMCH and colitis caused by C. rodentium in mice [152, 160, 320-323]. For instance, in
the study with B. breve UCC2003, Collins et al. found a significant reduction of colonic crypt
hyperplasia and mucosal infiltration of immune cells in mice treated with the probiotic strain 3
days before and 8 days after infection [320]. Rodrigues et al. demonstrated that C. rodentium-
induced Th17 and Th1 response and crypt hyperplasia were mitigated when probiotic treatment
(L. rhamnosus R0011 and L. helveticus R0052) was initiated one week before or concurrently
with the infection but not after the infection[323]. In the present study, TMCH and histological
damage of the ceacum and distal colon were not attenuated in mice treated with B. bifidum
MIMBb75 whether before or after C. rodentium challenge. This probiotic strain was chosen
based on its adhesion properties and ability to modify intestinal miRNAs such as miR-148a. It is
possible that this particular strain, B. bifidum MIMBb75, is ineffective in preventing or treating
infectious colitis, or the conditions in the study are not optimal for detecting benefits. Indeed,
even the same strain of probiotic may impose different influence on the host under different
conditions. In the study by Rodrigues et al. the same probiotic intervention prevented mortality
but did not reduce TMCH in neonatal mice as opposed to adult mice [323]. Interestingly, an
earlier study by the group, however, observed crypt reduction in probiotic-treated neonatal mice
[160]. These findings imply that experimental conditions may be an indispensable factor
determining the efficacy of a probiotic; namely, age of animals, dosage of C. rodentium
challenges, time and duration of intervention may all potentially mask or interfere with probiotic
effects.
One unexpected observation in the present study was the general increase of intestinal
permeability in all mice including the uninfected ones. It is known that C. rodentium infection
results in intestinal barrier dysfunction and therefore, increased paracellular translocation of
macromolecular tracer FITC-dextran, as seen in Study 1. Uninfected control mice were expected
to have marginal permeability based on Study 1 and the current literature [152, 160, 323, 328].
101
For example, using the in vivo intestinal permeability assay Rodrigues et al. showed that on day
10 p.i. serum FITC-dextran level of the infected adult mice was two-times higher than that of the
sham mice, and this high level was normalized to be comparable to the sham when treated with
probiotics [323]. Surprisingly, serum FITC-dextran concentration of the sham mice in the present
study was as high as the infected groups on day 10 p.i., suggesting impaired barrier integrity in
healthy uninfected mice. This could be a consequence of the daily intra-gastric gavage treatment
on the mice for a prolonged period of time. Although the employment of intra-gastric gavage
allows for a better control on the amount of probiotic administered to each mouse, it imposes
physical and psychological stress to the animals. It was suggested that chronic psychological
stress can trigger stress glucocorticoid hormone release leading to mucosal barrier dysfunction,
thereby increased intestinal permeability and host defense mechanism impairments [379]. As a
matter of fact, in the studies that showed positive effects of probiotic treatment with normal
ranges of intestinal permeability in the sham controls, probiotics were administered daily in
drinking water instead of intra-gastric gavage [152, 160, 323]. Other studies, such as with B.
breve UCC2003, used prolonged daily gavage of probiotics, but did not measure permeability
[320]. The present study is actually the first study to examine intestinal permeability under the
condition of daily gavage. Therefore, barrier integrity in all groups may have been disrupted due
to the daily gavage event. In line with this, addition, C. rodentium translocation to secondary
organs, spleen and liver, was similar among the infected groups, indicating similar alteration of
the barrier integrity. Although it is still possible for a probiotic to confer other health benefits
even when given via daily gavage, based on our data this administration method may increase
variability within experimental groups and thereby, a larger sample size may be needed to attain
statistical power for detecting any potential significant effects. Future studies should consider
exploiting other administration methods such as by drinking water or perorally behind the mouse
incisors, to minimize perturbation caused to animals. Importantly, data generated from studies
where probiotics were administered by daily gavage should be interpreted with caution in the
future, considering that the data may be affected by the altered barrier function.
The field of bacteria-host interaction at the level of miRNAs is still in its infancy. To date, there
is only one study that investigated the impact of probiotic Lactobacillus species on shaping
pathogenic Listeria-induced miRNA aberrancy [264]. The present study examined the
expression of six selected miRNAs (miR-148a, miR-200a, miR-200b, miR-19a, miR-21, and
102
miR-146a) that were deregulated during C. rodentium infection based on findings from Study 1,
and found that B. bifidum MIMBb75 treatment did not result in normalization of these miRNAs.
Importantly, there are inconsistencies between expression data of Study 1 and Study 2. Unlike
Study 1, which was based on NanoString and qPCR, miR-148a and miR-200a/b were not
significantly down-regulated upon infection and miR-19a, miR-21 and miR-146a were not
significantly up-regulated, even though a numerical increase was observed. There seems to be a
trend that the expression of miR-19a, miR-21 and miR-146a were induced upon C. rodentium
infection, which are in line with those in Study 1, and the increase of miR-21 was significant in
the probiotic-treated groups The statistical insignificance observed in Study 2 might be attributed
to a relatively small sample size (n=8 vs n=10 in study 1) and high variability. Furthermore, one
of the limitations of this study was the lack of an uninfected healthy group receiving daily B.
bifidum treatment. Therefore, it is unclear what would be the impact of B. bifidum alone on
shaping the expression of these miRNA. A previous study in our group found that miR-148a is
up-regulated in healthy mice treated with B. bifidum MIMBb75 for two days [272]. In the current
study, expression of miR-148 was decreased in the CR group, and miR-148a expression levels
were slightly elevated in the probiotic-treated groups. Although none of these attained
statistically significance, most likely because of the high variability, future studies may further
focus on this specific miRNA. Finally, even though B. bifidum MIMBb75 treatment did not
normalize the selected miRNA alterations associated with C. rodentium, it will be important to
analyze and compare genome-wide miRNA expression profiles between the experimental groups
in order to decipher potential miRNA modulations mediated by B. bifidum MIMBb75.
In summary, this is the first study looking at effects of B. bifidumMIMBb75 on infectious colitis
and host miRNA responses. It was found that administration of this strain did not attenuate C.
rodentium colonization, intestinal inflammation and crypt hyperplasia, barrier dysfunction, or
miRNA alterations associated with C. rodentium infection. Although the expression of the
selected miRNAs was not influenced by B. bifidum treatment, follow-up studies are needed to
examine genome-wide miRNA signature and transcriptome among the experimental groups in
the distal colon as well as the caecum. Findings from this study may provide insights for future
studies with respect to the administration method of probiotics as well as probiotics of choice for
mitigating intestinal inflammation.
103
Chapter 6
General Discussion
This research project includes two studies that investigated host-microbial crosstalk at the level
of posttranscriptional regulation of host gene expression. The first study (Chapter 4) showed that
C. rodentium infection alters the murine colonic miRNA signature, which might play a role in
the deregulation of apoptosis pathways upon infection as shown by bioinformatic and gene
expression analyses. In the second study (Chapter 5), administration of the probiotic strain B.
bifidum MIMBb75 before and after infection did not attenuate C. rodentium-induced pathology.
Moreover, based on the data of six selected miRNAs (miR-148a, miR-200a/b, miR-146a, miR-
21 and miR-19a of the miR-17-92 cluster), the presence of B. bifidum MIMBb75 did not have
significant impact on colonic miRNA alterations in response to C. rodentium infection.
To date, only a limited number of studies have explored miRNA expression in response to an
intestinal pathogen infection. Most of these studies focused on pathogens of the stomach
(Helicobacter) [252, 253], and small intestine (Listeria [251] and Salmonella [39]), and only
examined selected miRNAs. The present work reports for the first time a comprehensive analysis
of miRNA signature in the context of a colon-specific pathogen infection. Although there is a
lack of comprehensive data available in the literature, some of the differentially expressed
miRNAs in Study 1, including miR-146a/b, miR-21, miR-16, miR-200a/b and miR-148a, were
identified to be deregulated in a consistent manner in at least one other pathogen infection. This
is indicative of a plausible consensus of pathogen responsive miRNAs along the gastrointestinal
tract. The fact that these miRNAs have been implicated in regulating immune responses [254,
266] and cell cycle [267, 295, 380] highlights the critical role of miRNA in the host response to
pathogen infections. Meanwhile, some other miRNA could be deregulated in a pathogen-specific
manner. Exploiting bioinformatic approaches with gene expression analysis, this study proposes
the existence of a miRNA regulatory network, involving 9 upregulated and 2 downregulated
miRNAs, repressing the expression of the apoptosis factor Bim during infection. A low level of
Bim expression is associated with colon carcinomas [300, 366] and thereby, may contribute to
the development of crypt hyperplasia, which is a hallmark of C. rodentium pathogenicity.
Recent evidence has also revealed that miRNA expression patterns can be influenced by the
presence of gut microbiota [37, 38, 250] or probiotic bacteria [264, 269, 271, 272]. The probiotic
104
strain B. bifidum MIMBb75 has been previously demonstrated in our group by Singh et al. to be
able to modulate ceacal miRNA signature when administered to conventional mice for two days
[272]. It was found that this strain of B. bifidum induced expression of miR-148a, which was
downregulated in C. rodentium-infected colon based on Study 1. In light of this, Study 2 in this
project attempted to explore colonic miRNA responses in the interplay between the host, a
pathogen, and a probiotic bacterium. Six C. rodentium responsive miRNAs in Study 1 were
further investigated in Study 2. They are miR-148a, miR-200a/b, miR-146a, miR-21 and miR-
19a of the miR-17-92 cluster. Interestingly, in Study 2, none of the selected miRNAs were
significantly deregulated in mice infected with C. rodentium without probiotic treatment. These
findings are inconsistent with those in Study 1. Though, it should be noted that the expression of
miR-21, miR-19a and miR-146a were numerically increased in the all C. rodentium infected
groups, and miR-21 was significantly increased in the probiotic-treated infected groups. The lack
of significant differences of the C. rodentium group could be attributed to the relatively small
sample size with a high sample variability that lowered the statistical power for detecting
significance. Administration of B. bifidum MIMBb75 did not have a significant impact on these
C. rodentium-induced miRNA alterations. It is possible that the influence of a pathogen on the
host may outcompete that of a probiotic or that the host exhibits a time-dependent response to the
two bacteria. Indeed, in the study of Singh et al. mir-148a was only differentially expressed after
B. bifidum MIMBb75 was given for 2 days but not 14 days [272]. This may explain the lack of
influence of probiotic administration in Study 2, where B. bifidum was given for 17 or 10 days.
The postulation that host may respond early to probiotic colonization was also shown in Listeria
infection. Infection-induced miRNA alterations were attenuated when mice were treated with
probiotic Lactobacillus strains for 3 days in advance of the 24 hours long infection [264].
Furthermore, B. breve UCC2003 has been shown previously to exert beneficial effects against C.
rodentium infection [319, 320]. A few strains of B. bifidum have also been implicated in
preventing EHEC colonization in vitro and improving IBD in an animal model [31, 212].
Nonetheless, B. bifidum MIMBb75 administration was ineffective in preventing or mitigating C.
rodentium associated pathology in the current study. The ineffectiveness of B. bifidum
MIMBb75 could be attributed to the implementation of daily intra-gastric gavage, which is a
stressful event to animals and can lead to intestinal barrier dysfunction, as evidenced by the
abnormally high intestinal permeability in health control mice [379]. Another possibility is that
105
B. bifidum MIMBb75 does not confer beneficial effects on infectious colitis. A double-blind,
placebo-controlled RCT showed that supplementation of this probiotic strain improved IBS
symptoms [23]. Unlike IBD or infectious colitis, IBS is a functional disorder with no apparent or
a low grade chronic inflammation. It is possible that B. bifidum MIMBb75 has strain-specific
benefits on functional disorders but not in prominent inflammatory pathological conditions.
Generally speaking, findings in this project reveal a potential mechanism underlying the
crosstalk between the host and the pathogen C. rodentium such that miRNA-mediated post-
transcriptional regulation may contribute to the host hyperplasic response during the infection.
The study also shows that daily gavage, a procedure commonly used to study probiotic-host
interaction may not be ideal since it alters the intestinal barrier function. B. bifidum did not
mitigate C. rodentium-induced colitis and did not interfere with the expression of intestinal
miRNA affected by C. rodentium at the time point considered. Time-dependent effects should be
evaluated. Also, it would be of interest to investigate the miRNA response in mice infected with
C. rodentium and treated with probiotics proven to be beneficial in this context, such as B. breve
UCC2003 [320], L. rhamnosus R0011 and L. acidophilus R0052 [160, 323].
6.1 Strengths, Limitations and Future Directions
This is the first study using a highly sensitive and reproducible expression profiling technique,
NanoString, to study pathogen-host interaction at the level of posttranscriptional regulation. With
this advanced technology, miRNA response to a colonic pathogen was examined for the first
time in a comprehensive manner. This is important as miRNAs work cooperatively to regulate
gene expression. Moreover, a highly adhesive probiotic strain, which preferentially colonizes the
large intestine, was utilized for studying the effect of bifidobacteria species on pathogen
infection of the colon.
Though, there are also some limitations that need to be considered. First of all, in both Study 1
and 2 whole thickness tissues were used for gene expression analysis. Therefore it is not possible
to isolate the cellular origins of the expression signals detected. Taking into account that there is
a substantial mucosal infiltration of immune cells, it is uncertain whether the differentially
expressed miRNAs or mRNAs represent the expression pattern of epithelial cells or immune
106
cells during infection. Indeed, some deregulated miRNAs found in Study 1 such as miR-146a
and miR-155, have also been shown to be differentially expressed in macrophages and other
immune cells upon various enteric pathogen infections [39, 40, 265]. Nevertheless, studies
currently available in the literature also used whole-thickness tissues, such that RNA was
extracted from the whole-thickness of ileal tissue in the studies of miRNA expression upon
Listeria infection [38, 264]. To pinpoint epithelial-specific responses, future studies could
consider employing other sampling approaches such as laser capture microdissection that allow
direct examination of a specific cell type. To isolate C. rodentium-specific responses, future
studies should also compare baseline miRNA expression before infection to that at day 10 p.i..
Another potential limitation of both studies pertains to the use of a non-purified standard chow
diet. It was shown previously that diet formulation can have significant impact on gut microbiota
composition and even diseases outcome [381]. Specifically, Ooi et al. compared the effects of
three diets on DSS and C. rodentium induced colitis and found that DSS-induced colitis was
most severe in mice fed with the high calcium containing Teklad diet (TD) designed for breeding
vitamin D receptor knockout mice, followed by the ones fed with standard chow, and was least
severe in mice fed with a synthetic purified diet (PD); notably, the TD-fed mice experienced a
delayed eradication of both primary and secondary C. rodentium infection with a persisted higher
fecal load compared to PD-fed mice. Although both PD and TD are purified diets, it was
suggested that the protective effect of PD could be due to its high percentages of simple sugars
(glucose and sucrose) which has implications in CD [382] and was shown to be favoured by the
commensal bacteria Bacteriodes thetaiotaomicron to outcompete C. rodentium [110, 381].
Standard Chow diet on the other hand is not a synthetic diet; it contains a relatively high level of
folic acid [383], and compared to sterile purified AIN-93G diet it contains about 20 times more
bacteria-derived endotoxin lipopolysaccharide which has proinflammatory effects [384].
Knowing the importance of diet composition, it is important to utilize a purified diet with well-
defined composition to ensure that the study is conducted in the most controlled setting. Thus,
future studies would be benefit from employing a purified diet to control for potential variability
in nutritional composition of a non-synthetic diet.
Furthermore, Study 1 evaluated the expression of several genes that were postulated to be
regulated by miRNAs based on bioinformatic analysis with complementary literature search as
107
well as cross-matching with published microarray data. In order to have a more comprehensive
view of miRNA-mediated gene expression regulation during C. rodentium infection, global
transcriptome expression profiling could be performed and the data could be used for cross-
matching with the bioinformatically identified putative targets. This would increase the chances
of identifying relevant genes for follow up. In addition, expression of the putative miRNA-
modulating gene targets was evaluated at the mRNA level. Discrepancy could exist between a
gene and its corresponding protein levels. This is particularly relevant in miRNA studies because
a miRNA can suppress gene expression by either translational repression or degradation [242].
Therefore, the action of a miRNA does not necessarily reduce the number of transcripts of its
target. Thus, it is important to examine target expression at the protein level using protein assays
such as Western blot analysis, in situ hybridization, and ELISA. Future studies should also
consider functionally confirming that Bim is a direct target of miR-17-92-related clusters by
performing in vitro gain-of-function transfection experiments in Caco-2 cells.
Last but not least, Study 2 examines the effect of a probiotic strain on host miRNA response
during C. rodentium infection, which includes four experimental groups: sham control, C.
rodentium-infection, infection with B. bifidum pretreatment, and infection with B. bifidum post-
treatment. However, the inclusion of a group that received probiotic intervention without being
infected might be important for the isolation of effects solely attributable to B. bifidum. For
example, a previous study showed that miR-148a was upregulated after 2 days but not 14 days of
B. bifidum treatment in healthy mice [272]. In the present study miR-148a was not differentially
expressed in B. bifidum-treated infected mice. Without the B. bifidum-only group, it is unclear
whether the inhibitory effect of the infection outcompetes the stimulatory effect of B. bifidum, or
if B. bifidum did not have an impact on miR-148a at the time point considered. In addition, the
probiotic was administered to mice daily by intra-gastric gavage. While this is a commonly used
protocol, this administration procedure likely resulted in increased intestinal permeability,
masking the effects of C. rodentium. Future studies should consider implementation of other
administration methods to reduce this type of stress, such as administering the probiotics via the
drinking water or perorally. Moreover, both study 1 and 2 would benefit from the addition of a
inflammation positive control group with chemically-induced colitis (such as DSS-treatment),
which would allow the identification of miRNA responses specific to C. rodentium that are not
caused by mucosal inflammation or damages in general. Since miRNA alterations in response to
108
other colon-specific infectious agents are unknown, the inclusion of an infectious positive
control group is currently not applicable. Furthermore, it might also be essential to monitor
intestinal permeability throughout the course of infection in the future. The in vivo FITC-dextran
assay requires performance of cardiac puncture and it is impractical to utilize for monitoring
permeability at various time points, since it will enormously increase the number of animals to
be used [385]. Other assays, such as quantification of fecal albumin, have been shown to produce
comparable results as the FITC-dextran test and could thus be preferred [386].
6.2 Implications
The first miRNA was discovered in 1993 in Caenorhabditis elegans [387] but the term miRNA
was not coined until 2001[388]. Since then, a growing body of research has been conducted to
understand the biological role of miRNA and recent research suggests that they may be
important in regulating host-bacteria interaction. Though, this field is still in its infancy.
Currently, only a limited number of studies have demonstrated the impacts of presence of
microbiota, probiotic bacteria and pathogens on host miRNA expression. Findings in the present
project provide the first comprehensive analysis of miRNA expression patterns in response to a
colonic pathogen and extend the current knowledge of miRNA regulation being a potential
mechanism underlying host-microbial crosstalk.
MiRNAs are constantly deregulated in pathological conditions, and have been recognized as
biomarkers for diagnosis and prognosis of a disease. Given the biological relevance of C.
rodentium-induced colitis to IBD and tumorigenesis, the differentially expressed miRNAs
identified in this study may serve as biomarkers for these conditions. As a matter of fact,
circulating and colonic levels of miR-19a, miR-21 and miR-146a, which were deregulated by C.
rodentium, have recently been suggested to be candidate biomarkers for differentiating subtypes
of IBD, CD and UC, based on genome-wide miRNA signature analysis [389]. A high serum
level of miR-21 was also found to be associated with poor prognosis in CRC [390], and
increased levels of miR-21 and miR-17-92a cluster in exfoliated colonocytes were suggested to
be promising fecal miRNA biomarkers for CRC [298].
109
The differentially expressed miRNAs may also serve as targets of therapeutic or nutritional
interventions. The identification of the putative miRNA regulatory network on an apoptosis
facilitator gene, Bim, may have implications in elucidating pathogenicity mechanisms of C.
rodentium-induced TMCH via miRNAs. From a therapeutic perspective, understanding the
pathogenicity of C. rodentium is important for developing potential treatments for EHEC
infection, where the use of antibiotics is ineffective and discouraged. Deciphering the regulatory
role of miRNAs in TMCH shed light for developing miRNA-based anti-tumor therapies for
CRC, aiming to reprogram the aberrant miRNA networks. From a disease prevention standpoint,
dietary interventions that aim to promote homeostasis by miRNA modulation could also benefit
from the findings in this study. For example, increased expression of miR-106, which is
overexpressed in C. rodentium infection, is associated with CRC; and an in vitro study revealed
that microbe-derived SCFA butyrate can suppress miR-106 expression in a human colon cancer
cell line, suggesting that the differentially expressed miRNAs found in the current study may be
relevant candidate targets for dietary interventions.
Moreover, the discovery of increased intestinal permeability associated with daily gavage
implicates that cautions should be taken when interpreting studies that administer probiotics
through daily gavage, as the data might be influenced by alteration of the barrier function.
Finally, findings pertaining to the efficacy of B. bifidum MIMBb75 on C. rodentium infection
may provide some insights with respect to recommendation of probiotic strains in functional and
inflammatory diseases.
110
6.3 Conclusions
(1) C. rodentium infection alters murine colonic miRNA signature and the alterations are
involved in the anti-apoptotic response of host epithelium associated with the
development of crypt hyperplasia during the infection.
(2) B. bifidum MIMBb75 administration does not alleviate intestinal damage nor normalize
selected miRNA alterations associated with C. rodentium infection.
111
References
1. Thomas, M.K., et al., Burden of acute gastrointestinal illness in Canada, 1999-2007:
interim summary of NSAGI activities. Can Commun Dis Rep, 2008. 34(5): p. 8-15.
2. PublicHealthAgencyofCanada, National Enteric Surveillance Program (NESP), Annual
Summary 2009. Available from:
http://publications.gc.ca/collections/collection_2012/aspc-phac/HP37-15-2009-eng.pdf.
3. Fedorak, R.N., K. Wong, and R. Bridges, Canadian Digestive Health Foundation Public
Impact Series. Inflammatory bowel disease in Canada: Incidence, prevalence, and direct
and indirect economic impact. Canadian Journal of Gastroenterology, 2010. 24(11): p.
651-655.
4. Rocchi, A., et al., Inflammatory bowel disease: a Canadian burden of illness review. Can
J Gastroenterol, 2012. 26(11): p. 811-7.
5. Kim, E.R. and D.K. Chang, Colorectal cancer in inflammatory bowel disease: the risk,
pathogenesis, prevention and diagnosis. World J Gastroenterol, 2014. 20(29): p. 9872-81.
6. Hasan, N., A. Pollack, and I. Cho, Infectious causes of colorectal cancer. Infect Dis Clin
North Am, 2010. 24(4): p. 1019-39, x.
7. Vogelmann, R. and M.R. Amieva, The role of bacterial pathogens in cancer. Curr Opin
Microbiol, 2007. 10(1): p. 76-81.
8. Ferlay, J., et al., Estimates of worldwide burden of cancer in 2008: GLOBOCAN 2008.
Int J Cancer, 2010. 127(12): p. 2893-917.
9. Center, M.M., et al., Worldwide variations in colorectal cancer. CA Cancer J Clin, 2009.
59(6): p. 366-78.
10. Willing, B.P., et al., A pyrosequencing study in twins shows that gastrointestinal
microbial profiles vary with inflammatory bowel disease phenotypes. Gastroenterology,
2010. 139(6): p. 1844-1854.e1.
11. Lepage, P., et al., Twin study indicates loss of interaction between microbiota and
mucosa of patients with ulcerative colitis. Gastroenterology, 2011. 141(1): p. 227-36.
12. Sokol, H., et al., Low Counts of Faecalibacterium prausnitzii in Colitis Microbiota.
Inflammatory Bowel Diseases, 2009. 15(8): p. 1183-1189.
13. Ho, N.K., et al., Pathogenicity, host responses and implications for management of
enterohemorrhagic Escherichia coli O157:H7 infection. Can J Gastroenterol, 2013.
27(5): p. 281-5.
112
14. Lupp, C., et al., Host-mediated inflammation disrupts the intestinal microbiota and
promotes the overgrowth of Enterobacteriaceae. Cell Host Microbe, 2007. 2(2): p. 119-
29.
15. Collins, J.W., et al., Citrobacter rodentium: infection, inflammation and the microbiota.
Nat Rev Microbiol, 2014. 12(9): p. 612-23.
16. FAO/WHO, Evaluation of health and nutritional properties of powder milk and live
lactic acid bacteria, Food and Agriculture Organization of the United Nations and World
Health Organization Expert Consultation Report. 2001: Cordoba, Argentina
17. Canani, R.B., et al., Probiotics for treatment of acute diarrhoea in children: randomised
clinical trial of five different preparations. Bmj, 2007. 335(7615): p. 340.
18. Saavedra, J.M., et al., Feeding of Bifidobacterium bifidum and Streptococcus
thermophilus to infants in hospital for prevention of diarrhoea and shedding of rotavirus.
Lancet, 1994. 344(8929): p. 1046-9.
19. Dinleyici, E.C., et al., The effect of a multispecies synbiotic mixture on the duration of
diarrhea and length of hospital stay in children with acute diarrhea in Turkey: single
blinded randomized study. Eur J Pediatr, 2013. 172(4): p. 459-64.
20. Bin-Nun, A., et al., Oral probiotics prevent necrotizing enterocolitis in very low birth
weight neonates. J Pediatr, 2005. 147(2): p. 192-6.
21. Hong, K.S., et al., Effect of Probiotics on Symptoms in Korean Adults with Irritable
Bowel Syndrome. Gut and Liver, 2009. 3(2): p. 101-107.
22. Williams, E.A., et al., Clinical trial: a multistrain probiotic preparation significantly
reduces symptoms of irritable bowel syndrome in a double-blind placebo-controlled
study. Alimentary Pharmacology & Therapeutics, 2009. 29(1): p. 97-103.
23. Guglielmetti, S., et al., Randomised clinical trial: Bifidobacterium bifidum MIMBb75
significantly alleviates irritable bowel syndrome and improves quality of life - a double-
blind, placebo-controlled study. Alimentary Pharmacology & Therapeutics, 2011. 33(10):
p. 1123-1132.
24. Bartosch, S., et al., Microbiological effects of consuming a synbiotic containing
Bifidobacterium bifidum, Bifidobacterium lactis, and oligofructose in elderly persons,
determined by real-time polymerase chain reaction and counting of viable bacteria. Clin
Infect Dis, 2005. 40(1): p. 28-37.
25. Khailova, L., et al., Bifidobacterium bifidum improves intestinal integrity in a rat model
of necrotizing enterocolitis. American Journal of Physiology-Gastrointestinal and Liver
Physiology, 2009. 297(5): p. G940-G949.
26. Khailova, L., et al., Bifidobacterium bifidum reduces apoptosis in the intestinal
epithelium in necrotizing enterocolitis. American Journal of Physiology-Gastrointestinal
and Liver Physiology, 2010. 299(5): p. G1118-G1127.
113
27. Lopez, P., et al., Interaction of Bifidobacterium bifidum LMG13195 with HT29 Cells
Influences Regulatory-T-Cell-Associated Chemokine Receptor Expression. Applied and
Environmental Microbiology, 2012. 78(8): p. 2850-2857.
28. Ko, E.J., et al., Bifidobacterium bifidum exhibits a lipopolysaccharide-like mitogenic
activity for murine B lymphocytes. J Dairy Sci, 1999. 82(9): p. 1869-76.
29. Guglielmetti, S., et al., Implication of an outer surface lipoprotein in adhesion of
Bifidobacterium bifidum to Caco-2 cells. Applied and Environmental Microbiology,
2008. 74(15): p. 4695-4702.
30. Turroni, F., et al., Bifidobacterium bifidum PRL2010 modulates the host innate immune
response. Appl Environ Microbiol, 2014. 80(2): p. 730-40.
31. Bayoumi, M.A. and M.W. Griffiths, In vitro inhibition of expression of virulence genes
responsible for colonization and systemic spread of enteric pathogens using
Bifidobacterium bifidum secreted molecules. International Journal of Food Microbiology,
2012. 156(3): p. 255-263.
32. Philippe, D., et al., Treatment with Bifidobacterium bifidum 17 partially protects mice
from Th1-driven inflammation in a chemically induced model of colitis. International
Journal of Food Microbiology, 2011. 149(1): p. 45-49.
33. Hooper, L.V., et al., Molecular analysis of commensal host-microbial relationships in the
intestine. Science, 2001. 291(5505): p. 881-4.
34. Borenshtein, D., et al., Diarrhea as a cause of mortality in a mouse model of infectious
colitis. Genome Biol, 2008. 9(8): p. R122.
35. Spehlmann, M.E., et al., CXCR2-dependent mucosal neutrophil influx protects against
colitis-associated diarrhea caused by an attaching/effacing lesion-forming bacterial
pathogen. J Immunol, 2009. 183(5): p. 3332-43.
36. Kozomara, A. and S. Griffiths-Jones, miRBase: annotating high confidence microRNAs
using deep sequencing data. Nucleic Acids Res, 2014. 42(Database issue): p. D68-73.
37. Singh, N., et al., The murine caecal microRNA signature depends on the presence of the
endogenous microbiota. Int J Biol Sci, 2012. 8(2): p. 171-86.
38. Archambaud, C., et al., The intestinal microbiota interferes with the microRNA response
upon oral Listeria infection. MBio, 2013. 4(6): p. e00707-13.
39. Schulte, L.N., et al., Analysis of the host microRNA response to Salmonella uncovers the
control of major cytokines by the let-7 family. Embo j, 2011. 30(10): p. 1977-89.
40. Kaakoush, N.O., et al., Transcriptomic and proteomic analyses reveal key innate immune
signatures in the host response to the gastrointestinal pathogen Campylobacter concisus.
Infect Immun, 2015. 83(2): p. 832-45.
114
41. Pinto, D. and H. Clevers, Wnt, stem cells and cancer in the intestine. Biol Cell, 2005.
97(3): p. 185-96.
42. van der Flier, L.G. and H. Clevers, Stem cells, self-renewal, and differentiation in the
intestinal epithelium. Annu Rev Physiol, 2009. 71: p. 241-60.
43. Atuma, C., et al., The adherent gastrointestinal mucus gel layer: thickness and physical
state in vivo. Am J Physiol Gastrointest Liver Physiol, 2001. 280(5): p. G922-9.
44. McGuckin, M.A., et al., Mucin dynamics and enteric pathogens. Nat Rev Microbiol,
2011. 9(4): p. 265-78.
45. Weber, C.R., Dynamic properties of the tight junction barrier. Ann N Y Acad Sci, 2012.
1257: p. 77-84.
46. Neunlist, M., et al., The digestive neuronal-glial-epithelial unit: a new actor in gut health
and disease. Nat Rev Gastroenterol Hepatol, 2013. 10(2): p. 90-100.
47. Zasloff, M., Antimicrobial peptides of multicellular organisms. Nature, 2002. 415(6870):
p. 389-95.
48. Magalhaes, J.G., I. Tattoli, and S.E. Girardin, The intestinal epithelial barrier: how to
distinguish between the microbial flora and pathogens. Semin Immunol, 2007. 19(2): p.
106-15.
49. Philpott, D.J. and S.E. Girardin, The role of Toll-like receptors and Nod proteins in
bacterial infection. Mol Immunol, 2004. 41(11): p. 1099-108.
50. Gewirtz, A.T., et al., Cutting edge: bacterial flagellin activates basolaterally expressed
TLR5 to induce epithelial proinflammatory gene expression. J Immunol, 2001. 167(4): p.
1882-5.
51. Abreu, M.T., et al., Decreased expression of Toll-like receptor-4 and MD-2 correlates
with intestinal epithelial cell protection against dysregulated proinflammatory gene
expression in response to bacterial lipopolysaccharide. J Immunol, 2001. 167(3): p.
1609-16.
52. Naik, S., et al., Absence of Toll-like receptor 4 explains endotoxin hyporesponsiveness in
human intestinal epithelium. J Pediatr Gastroenterol Nutr, 2001. 32(4): p. 449-53.
53. Nenci, A., et al., Epithelial NEMO links innate immunity to chronic intestinal
inflammation. Nature, 2007. 446(7135): p. 557-61.
54. Fukata, M., et al., Toll-like receptor-4 is required for intestinal response to epithelial
injury and limiting bacterial translocation in a murine model of acute colitis. Am J
Physiol Gastrointest Liver Physiol, 2005. 288(5): p. G1055-65.
55. Gerondakis, S., et al., NF-kappaB control of T cell development. Nat Immunol, 2014.
15(1): p. 15-25.
115
56. Khoo, U.Y., I.E. Proctor, and A.J. Macpherson, CD4+ T cell down-regulation in human
intestinal mucosa: evidence for intestinal tolerance to luminal bacterial antigens. J
Immunol, 1997. 158(8): p. 3626-34.
57. Sansonetti, P.J. and R. Medzhitov, Learning tolerance while fighting ignorance. Cell,
2009. 138(3): p. 416-20.
58. Pasare, C. and R. Medzhitov, Toll-dependent control mechanisms of CD4 T cell
activation. Immunity, 2004. 21(5): p. 733-41.
59. Backhed, F., et al., Host-bacterial mutualism in the human intestine. Science, 2005.
307(5717): p. 1915-20.
60. Sekirov, I., et al., Gut microbiota in health and disease. Physiol Rev, 2010. 90(3): p. 859-
904.
61. Nicholson, J.K., E. Holmes, and I.D. Wilson, Gut microorganisms, mammalian
metabolism and personalized health care. Nat Rev Microbiol, 2005. 3(5): p. 431-8.
62. Sartor, R.B., Microbial influences in inflammatory bowel diseases. Gastroenterology,
2008. 134(2): p. 577-94.
63. Manichanh, C., et al., The gut microbiota in IBD. Nat Rev Gastroenterol Hepatol, 2012.
9(10): p. 599-608.
64. Sobhani, I., et al., Microbial Dysbiosis in Colorectal Cancer (CRC) Patients. Plos One,
2011. 6(1).
65. Darfeuille-Michaud, A., et al., High prevalence of adherent-invasive Escherichia coli
associated with ileal mucosa in Crohn's disease. Gastroenterology, 2004. 127(2): p. 412-
421.
66. Gill, S.R., et al., Metagenomic analysis of the human distal gut microbiome. Science,
2006. 312(5778): p. 1355-1359.
67. Cummings, J.H. and G.T. Macfarlane, Role of intestinal bacteria in nutrient metabolism
(Reprinted from Clinical Nutrition vol 16, pg 3, 1997). Journal of Parenteral and Enteral
Nutrition, 1997. 21(6): p. 357-365.
68. Gilmore, M.S. and J.J. Ferretti, Microbiology: The thin line between gut commensal and
pathogen. Science, 2003. 299(5615): p. 1999-+.
69. Mazmanian, S.K., et al., An immunomodulatory molecule of symbiotic bacteria directs
maturation of the host immune system. Cell, 2005. 122(1): p. 107-118.
70. Wlodarska, M., et al., Antibiotic treatment alters the colonic mucus layer and predisposes
the host to exacerbated Citrobacter rodentium-induced colitis. Infect Immun, 2011.
79(4): p. 1536-45.
116
71. Zachar, Z. and D.C. Savage, Microbial interference and colonization of the murine
gastrointestinal tract by Listeria monocytogenes. Infect Immun, 1979. 23(1): p. 168-74.
72. Ferreira, R.B., et al., The intestinal microbiota plays a role in Salmonella-induced colitis
independent of pathogen colonization. PLoS One, 2011. 6(5): p. e20338.
73. Yurist-Doutsch, S., et al., Gastrointestinal microbiota-mediated control of enteric
pathogens. Annu Rev Genet, 2014. 48: p. 361-82.
74. Deriu, E., et al., Probiotic bacteria reduce salmonella typhimurium intestinal
colonization by competing for iron. Cell Host Microbe, 2013. 14(1): p. 26-37.
75. Miller, M.B. and B.L. Bassler, Quorum sensing in bacteria. Annu Rev Microbiol, 2001.
55: p. 165-99.
76. Hsiao, A., et al., Members of the human gut microbiota involved in recovery from Vibrio
cholerae infection. Nature, 2014. 515(7527): p. 423-6.
77. Sperandio, V., et al., Quorum sensing controls expression of the type III secretion gene
transcription and protein secretion in enterohemorrhagic and enteropathogenic
Escherichia coli. Proc Natl Acad Sci U S A, 1999. 96(26): p. 15196-201.
78. Ivanov, II, et al., Induction of intestinal Th17 cells by segmented filamentous bacteria.
Cell, 2009. 139(3): p. 485-98.
79. Thompson, C.L., A. Mikaelyan, and A. Brune, Immune-modulating gut symbionts are not
"Candidatus Arthromitus". Mucosal Immunol, 2013. 6(1): p. 200-1.
80. Round, J.L., et al., The Toll-like receptor 2 pathway establishes colonization by a
commensal of the human microbiota. Science, 2011. 332(6032): p. 974-7.
81. Croxen, M.A. and B.B. Finlay, Molecular mechanisms of Escherichia coli pathogenicity.
Nat Rev Microbiol, 2010. 8(1): p. 26-38.
82. Schuster, C.J., et al., Infectious disease outbreaks related to drinking water in Canada,
1974-2001. Can J Public Health, 2005. 96(4): p. 254-8.
83. Marshall, J.K., et al., Incidence and epidemiology of irritable bowel syndrome after a
large waterborne outbreak of bacterial dysentery. Gastroenterology, 2006. 131(2): p.
445-50; quiz 660.
84. Ternhag, A., et al., Short- and long-term effects of bacterial gastrointestinal infections.
Emerging Infectious Diseases, 2008. 14(1): p. 143-148.
85. Collins, D., A.M. Hogan, and D.C. Winter, Microbial and viral pathogens in colorectal
cancer. Lancet Oncol, 2011. 12(5): p. 504-12.
86. Schauer, D.B. and S. Falkow, Attaching and effacing locus of a Citrobacter freundii
biotype that causes transmissible murine colonic hyperplasia. Infect Immun, 1993. 61(6):
p. 2486-92.
117
87. Deng, W.Y., et al., Locus of enterocyte effacement from Citrobacter rodentium: Sequence
analysis and evidence for horizontal transfer among attaching and effacing pathogens.
Infection and Immunity, 2001. 69(10): p. 6323-6335.
88. Campellone, K.G., D. Robbins, and J.M. Leong, EspFU is a translocated EHEC effector
that interacts with Tir and N-WASP and promotes Nck-independent actin assembly. Dev
Cell, 2004. 7(2): p. 217-28.
89. Shaw, R.K., et al., Enteropathogenic Escherichia coli type III effectors EspG and EspG2
disrupt the microtubule network of intestinal epithelial cells. Infect Immun, 2005. 73(7):
p. 4385-90.
90. Guttman, J.A., et al., Attaching and effacing pathogen-induced tight junction disruption
in vivo. Cell Microbiol, 2006. 8(4): p. 634-45.
91. Guttman, J.A., et al., Evidence that tight junctions are disrupted due to intimate bacterial
contact and not inflammation during attaching and effacing pathogen infection in vivo.
Infect Immun, 2006. 74(11): p. 6075-84.
92. Munera, D., et al., Recruitment and membrane interactions of host cell proteins during
attachment of enteropathogenic and enterohaemorrhagic Escherichia coli. Biochem J,
2012. 445(3): p. 383-92.
93. Martinez, E., et al., Binding to Na(+) /H(+) exchanger regulatory factor 2 (NHERF2)
affects trafficking and function of the enteropathogenic Escherichia coli type III secretion
system effectors Map, EspI and NleH. Cell Microbiol, 2010. 12(12): p. 1718-31.
94. Sham, H.P., et al., Attaching and effacing bacterial effector NleC suppresses epithelial
inflammatory responses by inhibiting NF-kappaB and p38 mitogen-activated protein
kinase activation. Infect Immun, 2011. 79(9): p. 3552-62.
95. Gao, X., et al., NleB, a bacterial effector with glycosyltransferase activity, targets
GAPDH function to inhibit NF-kappaB activation. Cell Host Microbe, 2013. 13(1): p. 87-
99.
96. Hemrajani, C., et al., NleH effectors interact with Bax inhibitor-1 to block apoptosis
during enteropathogenic Escherichia coli infection. Proc Natl Acad Sci U S A, 2010.
107(7): p. 3129-34.
97. Marches, O., et al., Enteropathogenic and enterohaemorrhagic Escherichia coli deliver a
novel effector called Cif, which blocks cell cycle G2/M transition. Mol Microbiol, 2003.
50(5): p. 1553-67.
98. Zhao, S., et al., The N-terminal domain of EspF induces host cell apoptosis after infection
with enterohaemorrhagic Escherichia coli O157:H7. PLoS One, 2013. 8(1): p. e55164.
99. Pearson, J.S., et al., A type III effector antagonizes death receptor signalling during
bacterial gut infection. Nature, 2013. 501(7466): p. 247-51.
118
100. Obrig, T.G., Escherichia coli Shiga Toxin Mechanisms of Action in Renal Disease.
Toxins (Basel), 2010. 2(12): p. 2769-2794.
101. Maluykova, I., et al., Latrunculin B facilitates Shiga toxin 1 transcellular transcytosis
across T84 intestinal epithelial cells. Biochim Biophys Acta, 2008. 1782(6): p. 370-7.
102. Petty, N.K., et al., The Citrobacter rodentium genome sequence reveals convergent
evolution with human pathogenic Escherichia coli. J Bacteriol, 2010. 192(2): p. 525-38.
103. Mallick, E.M., et al., A novel murine infection model for Shiga toxin-producing
Escherichia coli. J Clin Invest, 2012. 122(11): p. 4012-24.
104. Wiles, S., et al., Organ specificity, colonization and clearance dynamics in vivo following
oral challenges with the murine pathogen Citrobacter rodentium. Cell Microbiol, 2004.
6(10): p. 963-72.
105. Wiles, S., et al., In vivo bioluminescence imaging of the murine pathogen Citrobacter
rodentium. Infect Immun, 2006. 74(9): p. 5391-6.
106. Collins, J.W., et al., 4D multimodality imaging of Citrobacter rodentium infections in
mice. J Vis Exp, 2013(78).
107. Vallance, B.A., et al., Host susceptibility to the attaching and effacing bacterial pathogen
Citrobacter rodentium. Infect Immun, 2003. 71(6): p. 3443-53.
108. Ghosh, S., et al., Colonic microbiota alters host susceptibility to infectious colitis by
modulating inflammation, redox status, and ion transporter gene expression. Am J
Physiol Gastrointest Liver Physiol, 2011. 301(1): p. G39-49.
109. Willing, B.P., et al., Altering host resistance to infections through microbial
transplantation. PLoS One, 2011. 6(10): p. e26988.
110. Kamada, N., et al., Regulated virulence controls the ability of a pathogen to compete with
the gut microbiota. Science, 2012. 336(6086): p. 1325-9.
111. Wang, Y., et al., Citrobacter rodentium-induced NF-kappaB activation in
hyperproliferating colonic epithelia: role of p65 (Ser536) phosphorylation. Br J
Pharmacol, 2006. 148(6): p. 814-24.
112. Gibson, D.L., et al., MyD88 signalling plays a critical role in host defence by controlling
pathogen burden and promoting epithelial cell homeostasis during Citrobacter
rodentium-induced colitis. Cell Microbiol, 2008. 10(3): p. 618-31.
113. Fischer, H., et al., Differential expression of aquaporin 8 in human colonic epithelial
cells and colorectal tumors. BMC Physiol, 2001. 1: p. 1.
114. Halpin, S.J. and A.C. Ford, Prevalence of Symptoms Meeting Criteria for Irritable Bowel
Syndrome in Inflammatory Bowel Disease: Systematic Review and Meta-Analysis.
American Journal of Gastroenterology, 2012. 107(10): p. 1474-1482.
119
115. Xavier, R.J. and D.K. Podolsky, Unravelling the pathogenesis of inflammatory bowel
disease. Nature, 2007. 448(7152): p. 427-34.
116. Molodecky, N.A., et al., Increasing incidence and prevalence of the inflammatory bowel
diseases with time, based on systematic review. Gastroenterology, 2012. 142(1): p. 46-
54.e42; quiz e30.
117. Garcia Rodriguez, L.A., A. Ruigomez, and J. Panes, Acute gastroenteritis is followed by
an increased risk of inflammatory bowel disease. Gastroenterology, 2006. 130(6): p.
1588-94.
118. Qiu, H., et al., Serum bacterial toxins are related to the progression of inflammatory
bowel disease. Scand J Gastroenterol, 2014. 49(7): p. 826-33.
119. Cario, E., Toll-like receptors in inflammatory bowel diseases: a decade later. Inflamm
Bowel Dis, 2010. 16(9): p. 1583-97.
120. Gibson, D.L., et al., Toll-like receptor 2 plays a critical role in maintaining mucosal
integrity during Citrobacter rodentium-induced colitis. Cell Microbiol, 2008. 10(2): p.
388-403.
121. Higgins, L.M., et al., Citrobacter rodentium infection in mice elicits a mucosal Th1
cytokine response and lesions similar to those in murine inflammatory bowel disease.
Infection and Immunity, 1999. 67(6): p. 3031-3039.
122. Khan, M.A., et al., Toll-like receptor 4 contributes to colitis development but not to host
defense during Citrobacter rodentium infection in mice. Infect Immun, 2006. 74(5): p.
2522-36.
123. Geddes, K., et al., Identification of an innate T helper type 17 response to intestinal
bacterial pathogens. Nat Med, 2011. 17(7): p. 837-44.
124. Gophna, U., et al., Differences between tissue-associated intestinal microfloras of
patients with Crohn's disease and ulcerative colitis. J Clin Microbiol, 2006. 44(11): p.
4136-41.
125. Martinez-Medina, M., et al., Abnormal microbiota composition in the ileocolonic mucosa
of Crohn's disease patients as revealed by polymerase chain reaction-denaturing
gradient gel electrophoresis. Inflamm Bowel Dis, 2006. 12(12): p. 1136-45.
126. Andoh, A., et al., Terminal restriction fragment length polymorphism analysis of the
diversity of fecal microbiota in patients with ulcerative colitis. Inflamm Bowel Dis, 2007.
13(8): p. 955-62.
127. Strober, W., I. Fuss, and P. Mannon, The fundamental basis of inflammatory bowel
disease. J Clin Invest, 2007. 117(3): p. 514-21.
128. Balfour Sartor, R., Bacteria in Crohn's disease: mechanisms of inflammation and
therapeutic implications. J Clin Gastroenterol, 2007. 41 Suppl 1: p. S37-43.
120
129. Fattouh, R., et al., Rac2-deficiency leads to exacerbated and protracted colitis in
response to Citrobacter rodentium infection. PLoS One, 2013. 8(4): p. e61629.
130. Stigliano, V., et al., Early-onset colorectal cancer: a sporadic or inherited disease?
World J Gastroenterol, 2014. 20(35): p. 12420-30.
131. Song, M., W.S. Garrett, and A.T. Chan, Nutrients, foods, and colorectal cancer
prevention. Gastroenterology, 2015. 148(6): p. 1244-60.e16.
132. Magdy, A., et al., Enteropathogenic Escherichia coli (EPEC): Does it have a role in
colorectal tumourigenesis? A Prospective Cohort Study. Int J Surg, 2015. 18: p. 169-73.
133. Tyrer, P.C., F.A. Frizelle, and J.I. Keenan, Escherichia coli-derived outer membrane
vesicles are genotoxic to human enterocyte-like cells. Infect Agent Cancer, 2014. 9(1): p.
2.
134. Koturbash, I., et al., Heat-killed bacteria induce genome instability in mouse small
intestine, liver and spleen tissues. Cell Cycle, 2009. 8(12): p. 1935-9.
135. Khan, S., Potential role of Escherichia coli DNA mismatch repair proteins in colon
cancer. Crit Rev Oncol Hematol, 2015.
136. Pino, M.S. and D.C. Chung, The chromosomal instability pathway in colon cancer.
Gastroenterology, 2010. 138(6): p. 2059-72.
137. Tachibana, M., et al., Dysfunction of p53 pathway in human colorectal cancer: analysis
of p53 gene mutation and the expression of the p53-associated factors p14ARF,
p33ING1, p21WAF1 and MDM2. Int J Oncol, 2004. 25(4): p. 913-20.
138. Barthold, S.W. and A.M. Jonas, Morphogenesis of early 1, 2-dimethylhydrazine-induced
lesions and latent period reduction of colon carcinogenesis in mice by a variant of
Citrobacter freundii. Cancer Res, 1977. 37(12): p. 4352-60.
139. Newman, J.V., et al., Bacterial infection promotes colon tumorigenesis in Apc(Min/+)
mice. J Infect Dis, 2001. 184(2): p. 227-30.
140. Segditsas, S. and I. Tomlinson, Colorectal cancer and genetic alterations in the Wnt
pathway. Oncogene, 2006. 25(57): p. 7531-7.
141. Chandrakesan, P., et al., Differential effects of beta-catenin and NF-kappaB interplay in
the regulation of cell proliferation, inflammation and tumorigenesis in response to
bacterial infection. PLoS One, 2013. 8(11): p. e79432.
142. Umar, S., et al., Dual alterations in casein kinase I-epsilon and GSK-3beta modulate
beta-catenin stability in hyperproliferating colonic epithelia. Am J Physiol Gastrointest
Liver Physiol, 2007. 292(2): p. G599-607.
143. Chandrakesan, P., et al., Utility of a bacterial infection model to study epithelial-
mesenchymal transition, mesenchymal-epithelial transition or tumorigenesis. Oncogene,
2014. 33(20): p. 2639-54.
121
144. FAO/WHO, Guidelines for the evaluation of probiotics in food, Food and Agriculture
Organization of the United Nations and World Health Organization Working Group
Report. 2002: London, Ontario.
145. Stanton, C., et al., Market potential for probiotics. Am J Clin Nutr, 2001. 73(2 Suppl): p.
476s-483s.
146. Senok, A.C., A.Y. Ismaeel, and G.A. Botta, Probiotics: facts and myths. Clinical
Microbiology and Infection, 2005. 11(12): p. 958-966.
147. HealthCanada. Accepted Claims about the Nature of Probiotic Microorganisms in Food
2009; Available from: http://www.hc-sc.gc.ca/fn-an/label-etiquet/claims-
reclam/probiotics_claims-allegations_probiotiques-eng.php.
148. Fontana, L., et al., Sources, isolation, characterisation and evaluation of probiotics. Br J
Nutr, 2013. 109 Suppl 2: p. S35-50.
149. Hill, C., et al., Expert consensus document. The International Scientific Association for
Probiotics and Prebiotics consensus statement on the scope and appropriate use of the
term probiotic. Nat Rev Gastroenterol Hepatol, 2014. 11(8): p. 506-14.
150. Taibi, A. and E.M. Comelli, Practical approaches to probiotics use. Appl Physiol Nutr
Metab, 2014. 39(8): p. 980-6.
151. Corr, S.C., et al., Bacteriocin production as a mechanism for the antiinfective activity of
Lactobacillus salivarius UCC118. Proc Natl Acad Sci U S A, 2007. 104(18): p. 7617-21.
152. Johnson-Henry, K.C., et al., Amelioration of the effects of Citrobacter rodentium
infection in mice by pretreatment with probiotics. J Infect Dis, 2005. 191(12): p. 2106-17.
153. Ingrassia, I., A. Leplingard, and A. Darfeuille-Michaud, Lactobacillus casei DN-114 001
inhibits the ability of adherent-invasive Escherichia coli isolated from Crohn's disease
patients to adhere to and to invade intestinal epithelial cells. Appl Environ Microbiol,
2005. 71(6): p. 2880-7.
154. Boudeau, J., et al., Inhibitory effect of probiotic Escherichia coli strain Nissle 1917 on
adhesion to and invasion of intestinal epithelial cells by adherent-invasive E. coli strains
isolated from patients with Crohn's disease. Aliment Pharmacol Ther, 2003. 18(1): p. 45-
56.
155. Schmitz, H., et al., Altered tight junction structure contributes to the impaired epithelial
barrier function in ulcerative colitis. Gastroenterology, 1999. 116(2): p. 301-9.
156. Zeissig, S., et al., Changes in expression and distribution of claudin 2, 5 and 8 lead to
discontinuous tight junctions and barrier dysfunction in active Crohn's disease. Gut,
2007. 56(1): p. 61-72.
122
157. Mennigen, R., et al., Probiotic mixture VSL#3 protects the epithelial barrier by
maintaining tight junction protein expression and preventing apoptosis in a murine
model of colitis. Am J Physiol Gastrointest Liver Physiol, 2009. 296(5): p. G1140-9.
158. Dai, C., D.-H. Zhao, and M. Jiang, VSL#3 probiotics regulate the intestinal epithelial
barrier in vivo and in vitro via the p38 and ERK signaling pathways. International
Journal of Molecular Medicine, 2012. 29(2): p. 202-208.
159. Ewaschuk, J.B., et al., Secreted bioactive factors from Bifidobacterium infantis enhance
epithelial cell barrier function. Am J Physiol Gastrointest Liver Physiol, 2008. 295(5): p.
G1025-34.
160. Gareau, M.G., et al., Probiotics prevent death caused by Citrobacter rodentium infection
in neonatal mice. J Infect Dis, 2010. 201(1): p. 81-91.
161. Zheng, B., et al., Bifidobacterium breve attenuates murine dextran sodium sulfate-
induced colitis and increases regulatory T cell responses. PLoS One, 2014. 9(5): p.
e95441.
162. Tung, J.M., L.R. Dolovich, and C.H. Lee, Prevention of Clostridium difficile infection
with Saccharomyces boulardii: a systematic review. Can J Gastroenterol, 2009. 23(12): p.
817-21.
163. McFarland, L.V., Meta-analysis of probiotics for the prevention of antibiotic associated
diarrhea and the treatment of Clostridium difficile disease. Am J Gastroenterol, 2006.
101(4): p. 812-22.
164. Szajewska, H. and M. Kolodziej, Systematic review with meta-analysis: Saccharomyces
boulardii in the prevention of antibiotic-associated diarrhoea. Aliment Pharmacol Ther,
2015. 42(7): p. 793-801.
165. Ghouri, Y.A., et al., Systematic review of randomized controlled trials of probiotics,
prebiotics, and synbiotics in inflammatory bowel disease. Clin Exp Gastroenterol, 2014.
7: p. 473-87.
166. Jonkers, D., et al., Probiotics in the management of inflammatory bowel disease: a
systematic review of intervention studies in adult patients. Drugs, 2012. 72(6): p. 803-23.
167. Gionchetti, P., et al., Oral bacteriotherapy as maintenance treatment in patients with
chronic pouchitis: a double-blind, placebo-controlled trial. Gastroenterology, 2000.
119(2): p. 305-9.
168. Gionchetti, P., et al., Prophylaxis of pouchitis onset with probiotic therapy: a double-
blind, placebo-controlled trial. Gastroenterology, 2003. 124(5): p. 1202-9.
169. Mimura, T., et al., Once daily high dose probiotic therapy (VSL#3) for maintaining
remission in recurrent or refractory pouchitis. Gut, 2004. 53(1): p. 108-14.
123
170. Pronio, A., et al., Probiotic administration in patients with ileal pouch-anal anastomosis
for ulcerative colitis is associated with expansion of mucosal regulatory cells. Inflamm
Bowel Dis, 2008. 14(5): p. 662-8.
171. Madsen, K., et al., M1207 A Randomized Controlled Trial of VSL#3 for the Prevention of
Endoscopic Recurrence Following Surgery for Crohn's Disease. Gastroenterology, 2008.
134(4, Supplement 1): p. A-361.
172. Malchow, H.A., Crohn's disease and Escherichia coli. A new approach in therapy to
maintain remission of colonic Crohn's disease? J Clin Gastroenterol, 1997. 25(4): p. 653-
8.
173. Schultz, M., et al., Lactobacillus GG in inducing and maintaining remission of Crohn's
disease. BMC Gastroenterol, 2004. 4: p. 5.
174. Prantera, C., et al., Ineffectiveness of probiotics in preventing recurrence after curative
resection for Crohn's disease: a randomised controlled trial with Lactobacillus GG. Gut,
2002. 51(3): p. 405-9.
175. Apas, A.L., et al., Probiotic administration effect on fecal mutagenicity and microflora in
the goat's gut. J Biosci Bioeng, 2010. 110(5): p. 537-40.
176. Pokusaeva, K., G.F. Fitzgerald, and D. van Sinderen, Carbohydrate metabolism in
Bifidobacteria. Genes and Nutrition, 2011. 6(3): p. 285-306.
177. Biavati, B., et al., Bifidobacteria: history, ecology, physiology and applications. Annals
of Microbiology, 2000. 50(2): p. 117-131.
178. Turroni, F., et al., Exploring the diversity of the bifidobacterial population in the human
intestinal tract. Appl Environ Microbiol, 2009. 75(6): p. 1534-45.
179. Guglielmetti, S., et al., Study of the Adhesion of Bifidobacterium bifidum MIMBb75 to
Human Intestinal Cell Lines. Current Microbiology, 2009. 59(2): p. 167-172.
180. Singh, N., et al., Impact of Bifidobacterium bifidum MIMBb75 on mouse intestinal
microorganisms. Fems Microbiology Ecology, 2013. 85(2): p. 369-375.
181. Matsuki, T., et al., Quantitative PCR with 16S rRNA-gene-targeted species-specific
primers for analysis of human intestinal bifidobacteria. Appl Environ Microbiol, 2004.
70(1): p. 167-73.
182. Matsuki, T., et al., Use of 16S rRNA gene-targeted group-specific primers for real-time
PCR analysis of predominant bacteria in human feces. Appl Environ Microbiol, 2004.
70(12): p. 7220-8.
183. Marteau, P., et al., Comparative study of bacterial groups within the human cecal and
fecal microbiota. Appl Environ Microbiol, 2001. 67(10): p. 4939-42.
184. Tuzun, F., et al., Breast milk jaundice: effect of bacteria present in breast milk and infant
feces. J Pediatr Gastroenterol Nutr, 2013. 56(3): p. 328-32.
124
185. Masco, L., et al., Polyphasic taxonomic analysis of Bifidobacterium animalis and
Bifidobacterium lactis reveals relatedness at the subspecies level: reclassification of
Bifidobacterium animalis as Bifidobacterium animalis subsp animalis subsp nov and
Bifidobacterium lactis as Bifidobacterium animalis subsp lactis subsp nov. International
Journal of Systematic and Evolutionary Microbiology, 2004. 54: p. 1137-1143.
186. Turroni, F., et al., Analysis of Predicted Carbohydrate Transport Systems Encoded by
Bifidobacterium bifidum PRL2010. Applied and Environmental Microbiology, 2012.
78(14): p. 5002-5012.
187. Crociani, F., et al., DEGRADATION OF COMPLEX CARBOHYDRATES BY
BIFIDOBACTERIUM SPP. International Journal of Food Microbiology, 1994. 24(1-2):
p. 199-210.
188. Ruas-Madiedo, P., et al., Mucin degradation by Bifidobacterium strains isolated from the
human intestinal microbiota. Applied and Environmental Microbiology, 2008. 74(6): p.
1936-1940.
189. LoCascio, R.G., et al., A versatile and scalable strategy for glycoprofiling bifidobacterial
consumption of human milk oligosaccharides. Microbial Biotechnology, 2009. 2(3): p.
333-342.
190. Depeint, F., et al., Prebiotic evaluation of a novel galactooligosaccharide mixture
produced by the enzymatic activity of Bifidobacterium bifidum NCIMB 41171, in healthy
humans: a randomized, double-blind, crossover, placebo-controlled intervention study.
Am J Clin Nutr, 2008. 87(3): p. 785-91.
191. Gleinser, M., et al., Improved adhesive properties of recombinant bifidobacteria
expressing the Bifidobacterium bifidum-specific lipoprotein BopA. Microb Cell Fact,
2012. 11: p. 80.
192. Kline, K.A., et al., Bacterial Adhesins in Host-Microbe Interactions. Cell Host &
Microbe, 2009. 5(6): p. 580-592.
193. Foroni, E., et al., Genetic analysis and morphological identification of pilus-like
structures in members of the genus Bifidobacterium. Microbial Cell Factories, 2011. 10.
194. Motherway, M.O., et al., Functional genome analysis of Bifidobacterium breve UCC2003
reveals type IVb tight adherence (Tad) pili as an essential and conserved host-
colonization factor. Proceedings of the National Academy of Sciences of the United
States of America, 2011. 108(27): p. 11217-11222.
195. Westermann, C., et al., Exploring the genome sequence of Bifidobacterium bifidum S17
for potential players in host-microbe interactions. Symbiosis, 2012. 58(1-3): p. 191-200.
196. Turroni, F., et al., Role of sortase-dependent pili of Bifidobacterium bifidum PRL2010 in
modulating bacterium-host interactions. Proceedings of the National Academy of
Sciences of the United States of America, 2013. 110(27): p. 11151-11156.
125
197. Lebeer, S., J. Vanderleyden, and S.C. De Keersmaecker, Host interactions of probiotic
bacterial surface molecules: comparison with commensals and pathogens. Nat Rev
Microbiol, 2010. 8(3): p. 171-84.
198. Wajant, H., K. Pfizenmaier, and P. Scheurich, Tumor necrosis factor signaling. Cell
Death Differ, 2003. 10(1): p. 45-65.
199. Lecuit, M., et al., Functional genomic studies of the intestinal response to a foodborne
enteropathogen in a humanized gnotobiotic mouse model. J Biol Chem, 2007. 282(20): p.
15065-72.
200. Friederichs, J., et al., Gene expression profiles of different clinical stages of colorectal
carcinoma: toward a molecular genetic understanding of tumor progression. Int J
Colorectal Dis, 2005. 20(5): p. 391-402.
201. Duffy, L.C., et al., Effectiveness of Bifidobacterium bifidum in mediating the clinical
course of murine rotavirus diarrhea. Pediatr Res, 1994. 35(6): p. 690-5.
202. Qiao, H., et al., Immune responses in rhesus rotavirus-challenged BALB/c mice treated
with bifidobacteria and prebiotic supplements. Pediatr Res, 2002. 51(6): p. 750-5.
203. Trejo, F.M., G.L. De Antoni, and P.F. Perez, Protective effect of bifidobacteria in an
experimental model of Clostridium difficile associated colitis. Journal of Dairy Research,
2013. 80(3): p. 263-269.
204. Lin, H.C., et al., Oral probiotics prevent necrotizing enterocolitis in very low birth weight
preterm infants: a multicenter, randomized, controlled trial. Pediatrics, 2008. 122(4): p.
693-700.
205. Samanta, M., et al., Prophylactic Probiotics for Prevention of Necrotizing Enterocolitis
in Very Low Birth Weight Newborns. Journal of Tropical Pediatrics, 2009. 55(2): p. 128-
131.
206. Underwood, M.A., et al., Bifidobacterium bifidum in a rat model of necrotizing
enterocolitis: antimicrobial peptide and protein responses. Pediatric Research, 2012.
71(5): p. 546-551.
207. Yesilova, Y., et al., Effect of probiotics on the treatment of children with atopic
dermatitis. Ann Dermatol, 2012. 24(2): p. 189-93.
208. Gutkowski, P., et al., Effect of orally administered probiotic strains Lactobacillus and
Bifidobacterium in children with atopic asthma. Central European Journal of
Immunology, 2010. 35(4): p. 233-238.
209. Kim, J.Y., et al., Effect of probiotic mix (Bifidobacterium bifidum, Bifidobacterium lactis,
Lactobacillus acidophilus) in the primary prevention of eczema: a double-blind,
randomized, placebo-controlled trial. Pediatric Allergy and Immunology, 2010. 21(2): p.
E386-E393.
126
210. Saha, L., Irritable bowel syndrome: pathogenesis, diagnosis, treatment, and evidence-
based medicine. World J Gastroenterol, 2014. 20(22): p. 6759-73.
211. Miki, K., et al., Effect of Bifidobacterium bifidum fermented milk on helicobacter pylori
and serum pepsinogen levels in humans. Journal of Dairy Science, 2007. 90(6): p. 2630-
2640.
212. Kim, N., et al., Oral feeding of Bifidobacterium bifidum (BGN4) prevents CD4(+)
CD45RB(high) T cell-mediated inflammatory bowel disease by inhibition of disordered T
cell activation. Clin Immunol, 2007. 123(1): p. 30-9.
213. Cazzola, M., T.A. Tompkins, and M.G. Matera, Immunomodulatory impact of a synbiotic
in T(h)1 and T(h)2 models of infection. Therapeutic advances in respiratory disease,
2010. 4(5): p. 259-70.
214. Jadhav, S.R., U.K. Shandilya, and V.K. Kansal, Exploring the ameliorative potential of
probiotic Dahi containing Lactobacillus acidophilus and Bifidobacterium bifidum on
dextran sodium sulphate induced colitis in mice. Journal of Dairy Research, 2013. 80(1):
p. 21-27.
215. You, H.J., D.K. Oh, and G.E. Ji, Anticancerogenic effect of a novel chiroinositol-
containing polysaccharide from Bifidobacterium bifidum BGN4. FEMS Microbiol Lett,
2004. 240(2): p. 131-6.
216. Mohania, D., et al., Probiotic Dahi containing Lactobacillus acidophilus and
Bifidobacterium bifidum modulates the formation of aberrant crypt foci, mucin-depleted
foci, and cell proliferation on 1,2-dimethylhydrazine-induced colorectal carcinogenesis
in Wistar rats. Rejuvenation Res, 2014. 17(4): p. 325-33.
217. Rerksuppaphol, S. and L. Rerksuppaphol, Randomized controlled trial of probiotics to
reduce common cold in schoolchildren. Pediatrics International, 2012. 54(5): p. 682-687.
218. de Vrese, M., et al., Effect of Lactobacillus gasseri PA 16/8, Bifidobacterium longum SP
07/3, B. bifidum MF 20/5 on common cold episodes: a double blind, randomized,
controlled trial. Clin Nutr, 2005. 24(4): p. 481-91.
219. Schiffrin, E.J., et al., Immune modulation of blood leukocytes in humans by lactic acid
bacteria: criteria for strain selection. Am J Clin Nutr, 1997. 66(2): p. 515s-520s.
220. Vinderola, C.G., M. Medici, and G. Perdigon, Relationship between interaction sites in
the gut, hydrophobicity, mucosal immunomodulating capacities and cell wall protein
profiles in indigenous and exogenous bacteria. J Appl Microbiol, 2004. 96(2): p. 230-43.
221. Devries, W., Gerbrand.Sj, and Stoutham.Ah, CARBOHYDRATE METABOLISM IN
BIFIDOBACTERIUM BIFIDUM. Biochimica Et Biophysica Acta, 1967. 136(3): p. 415-
&.
222. Macfarlane, G.T. and S. Macfarlane, Bacteria, colonic fermentation, and gastrointestinal
health. J AOAC Int, 2012. 95(1): p. 50-60.
127
223. Siew-Wai, L., et al., Fermentation of Metroxylon sagu resistant starch type III by
Lactobacillus sp. and Bifidobacterium bifidum. J Agric Food Chem, 2010. 58(4): p.
2274-8.
224. Deguchi, Y., T. Morishita, and M. Mutai, COMPARATIVE STUDIES ON SYNTHESIS
OF WATER-SOLUBLE VITAMINS AMONG HUMAN SPECIES OF BIFIDOBACTERIA.
Agricultural and Biological Chemistry, 1985. 49(1): p. 13-19.
225. D'Aimmo, M.R., et al., The potential of bifidobacteria as a source of natural folate.
Journal of Applied Microbiology, 2012. 112(5): p. 975-984.
226. Pompei, A., et al., Folate production by bifidobacteria as a potential probiotic property.
Appl Environ Microbiol, 2007. 73(1): p. 179-85.
227. Pompei, A., et al., Administration of folate-producing bifidobacteria enhances folate
status in Wistar rats. J Nutr, 2007. 137(12): p. 2742-6.
228. Strozzi, G.P. and L. Mogna, Quantification of folic acid in human feces after
administration of Bifidobacterium probiotic strains. J Clin Gastroenterol, 2008. 42 Suppl
3 Pt 2: p. S179-84.
229. Dinleyici, E.C., et al., The effect of a multispecies synbiotic mixture on the duration of
diarrhea and length of hospital stay in children with acute diarrhea in Turkey: Single
blinded randomized study. European Journal of Pediatrics, 2013. 172(4): p. 459-464.
230. Kumar, S., et al., Evaluation of Efficacy of Probiotics in Prevention of Candida
Colonization in a PICU-A Randomized Controlled Trial. Critical Care Medicine, 2013.
41(2): p. 565-572.
231. Wang, Y.H. and Y. Huang, Effect of Lactobacillus acidophilus and Bifidobacterium
bifidum supplementation to standard triple therapy on Helicobacter pylori eradication
and dynamic changes in intestinal flora. World Journal of Microbiology &
Biotechnology, 2014. 30(3): p. 847-853.
232. Navarro-Rodriguez, T., et al., Association of a probiotic to a Helicobacter pylori
eradication regimen does not increase efficacy or decreases the adverse effects of the
treatment: a prospective, randomized, double-blind, placebo-controlled study. Bmc
Gastroenterology, 2013. 13.
233. Madden, J.A.J., et al., Effect of probiotics on preventing disruption of the intestinal
microflora following antibiotic therapy: A double-blind, placebo-controlled pilot study.
International Immunopharmacology, 2005. 5(6): p. 1091-1097.
234. Gomi, A., et al., Health benefits of fermented milk containing Bifidobacterium bifidum
YIT 10347 on gastric symptoms in adults. J Dairy Sci, 2015. 98(4): p. 2277-83.
235. Urita, Y., et al., Continuous consumption of fermented milk containing Bifidobacterium
bifidum YIT 10347 improves gastrointestinal and psychological symptoms in patients
128
with functional gastrointestinal disorders. Biosci Microbiota Food Health, 2015. 34(2): p.
37-44.
236. Trois, L., E.M. Cardoso, and E. Miura, Use of probiotics in HIV-infected children: a
randomized double-blind controlled study. J Trop Pediatr, 2008. 54(1): p. 19-24.
237. Wong, V.W., et al., Treatment of nonalcoholic steatohepatitis with probiotics. A proof-of-
concept study. Ann Hepatol, 2013. 12(2): p. 256-62.
238. De Simone, C., et al., Effect of Bifidobacterium bifidum and Lactobacillus acidophilus on
gut mucosa and peripheral blood B lymphocytes. Immunopharmacol Immunotoxicol,
1992. 14(1-2): p. 331-40.
239. Langkamp-Henken, B., et al., Bifidobacterium bifidum R0071 results in a greater
proportion of healthy days and a lower percentage of academically stressed students
reporting a day of cold/flu: a randomised, double-blind, placebo-controlled study. Br J
Nutr, 2015. 113(3): p. 426-34.
240. Miki, K., et al., Effect of Bifidobacterium bifidum fermented milk on Helicobacter pylori
and serum pepsinogen levels in humans. J Dairy Sci, 2007. 90(6): p. 2630-40.
241. Bartel, D.P., MicroRNAs: genomics, biogenesis, mechanism, and function. Cell, 2004.
116(2): p. 281-97.
242. Siomi, H. and M.C. Siomi, Posttranscriptional regulation of microRNA biogenesis in
animals. Mol Cell, 2010. 38(3): p. 323-32.
243. Friedman, R.C., et al., Most mammalian mRNAs are conserved targets of microRNAs.
Genome Res, 2009. 19(1): p. 92-105.
244. Chen, C.Z., et al., Regulation of immune responses and tolerance: the microRNA
perspective. Immunol Rev, 2013. 253(1): p. 112-28.
245. Bueno, M.J., I. Perez de Castro, and M. Malumbres, Control of cell proliferation
pathways by microRNAs. Cell Cycle, 2008. 7(20): p. 3143-8.
246. O'Connell, R.M., et al., Physiological and pathological roles for microRNAs in the
immune system. Nat Rev Immunol, 2010. 10(2): p. 111-22.
247. Koufaris, C., et al., Time and dose-dependent effects of phenobarbital on the rat liver
miRNAome. Toxicology, 2013. 314(2-3): p. 247-53.
248. Moschos, S.A., et al., Expression profiling in vivo demonstrates rapid changes in lung
microRNA levels following lipopolysaccharide-induced inflammation but not in the anti-
inflammatory action of glucocorticoids. BMC Genomics, 2007. 8: p. 240.
249. McKenna, L.B., et al., MicroRNAs control intestinal epithelial differentiation,
architecture, and barrier function. Gastroenterology, 2010. 139(5): p. 1654-64, 1664.e1.
129
250. Dalmasso, G., et al., Microbiota modulate host gene expression via microRNAs. PLoS
One, 2011. 6(4): p. e19293.
251. Izar, B., et al., microRNA response to Listeria monocytogenes infection in epithelial cells.
Int J Mol Sci, 2012. 13(1): p. 1173-85.
252. Petrocca, F., et al., E2F1-regulated microRNAs impair TGFbeta-dependent cell-cycle
arrest and apoptosis in gastric cancer. Cancer Cell, 2008. 13(3): p. 272-86.
253. Lario, S., et al., microRNA profiling in duodenal ulcer disease caused by Helicobacter
pylori infection in a Western population. Clin Microbiol Infect, 2012. 18(8): p. E273-82.
254. Taganov, K.D., et al., NF-kappaB-dependent induction of microRNA miR-146, an
inhibitor targeted to signaling proteins of innate immune responses. Proc Natl Acad Sci
U S A, 2006. 103(33): p. 12481-6.
255. Nahid, M.A., et al., miR-146a is critical for endotoxin-induced tolerance: IMPLICATION
IN INNATE IMMUNITY. J Biol Chem, 2009. 284(50): p. 34590-9.
256. Chassin, C., et al., MicroRNA-146a-mediated downregulation of IRAK1 protects mouse
and human small intestine against ischemia/reperfusion injury. EMBO Mol Med, 2012.
4(12): p. 1308-19.
257. Chassin, C., et al., miR-146a mediates protective innate immune tolerance in the neonate
intestine. Cell Host Microbe, 2010. 8(4): p. 358-68.
258. O'Connell, R.M., et al., Inositol phosphatase SHIP1 is a primary target of miR-155. Proc
Natl Acad Sci U S A, 2009. 106(17): p. 7113-8.
259. Ceppi, M., et al., MicroRNA-155 modulates the interleukin-1 signaling pathway in
activated human monocyte-derived dendritic cells. Proc Natl Acad Sci U S A, 2009.
106(8): p. 2735-40.
260. Schulte, L.N., A.J. Westermann, and J. Vogel, Differential activation and functional
specialization of miR-146 and miR-155 in innate immune sensing. Nucleic Acids Res,
2013. 41(1): p. 542-53.
261. Oertli, M., et al., MicroRNA-155 is essential for the T cell-mediated control of
Helicobacter pylori infection and for the induction of chronic Gastritis and Colitis. J
Immunol, 2011. 187(7): p. 3578-86.
262. Matsushima, K., et al., MicroRNA signatures in Helicobacter pylori-infected gastric
mucosa. Int J Cancer, 2011. 128(2): p. 361-70.
263. Xiao, B., et al., Induction of microRNA-155 during Helicobacter pylori infection and its
negative regulatory role in the inflammatory response. J Infect Dis, 2009. 200(6): p. 916-
25.
264. Archambaud, C., et al., Impact of lactobacilli on orally acquired listeriosis. Proc Natl
Acad Sci U S A, 2012. 109(41): p. 16684-9.
130
265. Schnitger, A.K., et al., Listeria monocytogenes infection in macrophages induces
vacuolar-dependent host miRNA response. PLoS One, 2011. 6(11): p. e27435.
266. Huang, Z., et al., Dual TNF-alpha/IL-12p40 interference as a strategy to protect against
colitis based on miR-16 precursors with macrophage targeting vectors. Mol Ther, 2015.
267. Brabletz, S. and T. Brabletz, The ZEB/miR-200 feedback loop--a motor of cellular
plasticity in development and cancer? EMBO Rep, 2010. 11(9): p. 670-7.
268. Clare, S., et al., Enhanced susceptibility to Citrobacter rodentium infection in microRNA-
155-deficient mice. Infect Immun, 2013. 81(3): p. 723-32.
269. Giahi, L., et al., Regulation of TLR4, p38 MAPkinase, I kappa B and miRNAs by
inactivated strains of lactobacilli in human dendritic cells. Beneficial Microbes, 2012.
3(2): p. 91-98.
270. Zhao, H., et al., Inhibition of miR122a by Lactobacillus rhamnosus GG culture
supernatant increases intestinal occludin expression and protects mice from alcoholic
liver disease. Toxicol Lett, 2015. 234(3): p. 194-200.
271. Veltman, K., et al., Identification of specific miRNAs targeting proteins of the apical
junctional complex that simulate the probiotic effect of E. coli Nissle 1917 on T84
epithelial cells. Int J Biochem Cell Biol, 2012. 44(2): p. 341-9.
272. Singh, N., et al., Early Response of the Murine Caecum to the Probiotic Bifidobacterium
Bifidum MIMBb75 Through MicroRNA Modulation. Gastroenterology, 2012. 142(5): p.
S394-S394.
273. Liu, Z., et al., Up-regulated microRNA-146a negatively modulate Helicobacter pylori-
induced inflammatory response in human gastric epithelial cells. Microbes Infect, 2010.
12(11): p. 854-63.
274. Zhang, T., et al., Salmonella enterica serovar enteritidis modulates intestinal epithelial
miR-128 levels to decrease macrophage recruitment via macrophage colony-stimulating
factor. J Infect Dis, 2014. 209(12): p. 2000-11.
275. Wu, F., et al., MicroRNAs are differentially expressed in ulcerative colitis and alter
expression of macrophage inflammatory peptide-2 alpha. Gastroenterology, 2008.
135(5): p. 1624-1635.e24.
276. Fasseu, M., et al., Identification of restricted subsets of mature microRNA abnormally
expressed in inactive colonic mucosa of patients with inflammatory bowel disease. PLoS
One, 2010. 5(10).
277. Takagi, T., et al., Increased expression of microRNA in the inflamed colonic mucosa of
patients with active ulcerative colitis. J Gastroenterol Hepatol, 2010. 25 Suppl 1: p.
S129-33.
131
278. Chen, Y., et al., miR-200b inhibits TGF-beta1-induced epithelial-mesenchymal transition
and promotes growth of intestinal epithelial cells. Cell Death Dis, 2013. 4: p. e541.
279. Cuthbert, A.P., et al., The contribution of NOD2 gene mutations to the risk and site of
disease in inflammatory bowel disease. Gastroenterology, 2002. 122(4): p. 867-74.
280. Chuang, A.Y., et al., NOD2 expression is regulated by microRNAs in colonic epithelial
HCT116 cells. Inflamm Bowel Dis, 2014. 20(1): p. 126-35.
281. Jiang, S., et al., Molecular dissection of the miR-17-92 cluster's critical dual roles in
promoting Th1 responses and preventing inducible Treg differentiation. Blood, 2011.
118(20): p. 5487-97.
282. Salim, S.Y. and J.D. Soderholm, Importance of Disrupted Intestinal Barrier in
Inflammatory Bowel Diseases. Inflammatory Bowel Diseases, 2011. 17(1): p. 362-381.
283. Wu, F., et al., Identification of microRNAs associated with ileal and colonic Crohn's
disease. Inflamm Bowel Dis, 2010. 16(10): p. 1729-38.
284. Shi, C., et al., MicroRNA-21 knockout improve the survival rate in DSS induced fatal
colitis through protecting against inflammation and tissue injury. PLoS One, 2013. 8(6):
p. e66814.
285. Yang, Y., et al., Overexpression of miR-21 in patients with ulcerative colitis impairs
intestinal epithelial barrier function through targeting the Rho GTPase RhoB. Biochem
Biophys Res Commun, 2013. 434(4): p. 746-52.
286. Bian, Z., et al., Role of miR-150-targeting c-Myb in colonic epithelial disruption during
dextran sulphate sodium-induced murine experimental colitis and human ulcerative
colitis. J Pathol, 2011. 225(4): p. 544-53.
287. Wu, F., et al., Divergent influence of microRNA-21 deletion on murine colitis phenotypes.
Inflamm Bowel Dis, 2014. 20(11): p. 1972-85.
288. Cummins, J.M., et al., The colorectal microRNAome. Proc Natl Acad Sci U S A, 2006.
103(10): p. 3687-92.
289. Schetter, A.J., et al., MicroRNA expression profiles associated with prognosis and
therapeutic outcome in colon adenocarcinoma. Jama, 2008. 299(4): p. 425-36.
290. Lin, M., et al., MicroRNA expression profiles in human colorectal cancers with liver
metastases. Oncol Rep, 2011. 25(3): p. 739-47.
291. Luo, X., et al., MicroRNA signatures: novel biomarker for colorectal cancer? Cancer
Epidemiol Biomarkers Prev, 2011. 20(7): p. 1272-86.
292. Esquela-Kerscher, A. and F.J. Slack, Oncomirs - microRNAs with a role in cancer. Nat
Rev Cancer, 2006. 6(4): p. 259-69.
132
293. Yamamichi, N., et al., Locked nucleic acid in situ hybridization analysis of miR-21
expression during colorectal cancer development. Clin Cancer Res, 2009. 15(12): p.
4009-16.
294. Xiong, B., et al., MiR-21 regulates biological behavior through the PTEN/PI-3 K/Akt
signaling pathway in human colorectal cancer cells. Int J Oncol, 2013. 42(1): p. 219-28.
295. Asangani, I.A., et al., MicroRNA-21 (miR-21) post-transcriptionally downregulates
tumor suppressor Pdcd4 and stimulates invasion, intravasation and metastasis in
colorectal cancer. Oncogene, 2008. 27(15): p. 2128-36.
296. Xiang, J. and J. Wu, Feud or Friend? The Role of the miR-17-92 Cluster in
Tumorigenesis. Curr Genomics, 2010. 11(2): p. 129-35.
297. Diosdado, B., et al., MiR-17-92 cluster is associated with 13q gain and c-myc expression
during colorectal adenoma to adenocarcinoma progression. Br J Cancer, 2009. 101(4):
p. 707-14.
298. Koga, Y., et al., MicroRNA expression profiling of exfoliated colonocytes isolated from
feces for colorectal cancer screening. Cancer Prev Res (Phila), 2010. 3(11): p. 1435-42.
299. Ivanovska, I., et al., MicroRNAs in the miR-106b family regulate p21/CDKN1A and
promote cell cycle progression. Mol Cell Biol, 2008. 28(7): p. 2167-74.
300. Tsuchida, A., et al., miR-92 is a key oncogenic component of the miR-17-92 cluster in
colon cancer. Cancer Sci, 2011. 102(12): p. 2264-71.
301. Luo, H., et al., Up-regulated miR-17 promotes cell proliferation, tumour growth and cell
cycle progression by targeting the RND3 tumour suppressor gene in colorectal
carcinoma. Biochem J, 2012. 442(2): p. 311-21.
302. Tania, M., M.A. Khan, and J. Fu, Epithelial to mesenchymal transition inducing
transcription factors and metastatic cancer. Tumour Biol, 2014. 35(8): p. 7335-42.
303. Chen, M.L., L.S. Liang, and X.K. Wang, miR-200c inhibits invasion and migration in
human colon cancer cells SW480/620 by targeting ZEB1. Clin Exp Metastasis, 2012.
29(5): p. 457-69.
304. Edvardsson, K., et al., Estrogen receptor beta expression induces changes in the
microRNA pool in human colon cancer cells. Carcinogenesis, 2013. 34(7): p. 1431-41.
305. Tian, Y., et al., MicroRNA-200 (miR-200) cluster regulation by achaete scute-like 2
(Ascl2): impact on the epithelial-mesenchymal transition in colon cancer cells. J Biol
Chem, 2014. 289(52): p. 36101-15.
306. Gootenberg, D.B. and P.J. Turnbaugh, Companion animals symposium: humanized
animal models of the microbiome. J Anim Sci, 2011. 89(5): p. 1531-7.
307. Lu, L., et al., Animal models of gastrointestinal inflammation and cancer. Life Sci, 2014.
108(1): p. 1-6.
133
308. Kararli, T.T., Comparison of the gastrointestinal anatomy, physiology, and biochemistry
of humans and commonly used laboratory animals. Biopharm Drug Dispos, 1995. 16(5):
p. 351-80.
309. Nguyen, T.L., et al., How informative is the mouse for human gut microbiota research?
Dis Model Mech, 2015. 8(1): p. 1-16.
310. McConnell, E.L., A.W. Basit, and S. Murdan, Measurements of rat and mouse
gastrointestinal pH, fluid and lymphoid tissue, and implications for in-vivo experiments. J
Pharm Pharmacol, 2008. 60(1): p. 63-70.
311. Dewhirst, F.E., et al., Phylogeny of the defined murine microbiota: altered Schaedler
flora. Appl Environ Microbiol, 1999. 65(8): p. 3287-92.
312. Lamendella, R., et al., Bifidobacteria in feces and environmental waters. Applied and
Environmental Microbiology, 2008. 74(3): p. 575-584.
313. DeVoss, J. and L. Diehl, Murine models of inflammatory bowel disease (IBD):
challenges of modeling human disease. Toxicol Pathol, 2014. 42(1): p. 99-110.
314. Waterston, R.H., et al., Initial sequencing and comparative analysis of the mouse
genome. Nature, 2002. 420(6915): p. 520-62.
315. Landgraf, P., et al., A mammalian microRNA expression atlas based on small RNA
library sequencing. Cell, 2007. 129(7): p. 1401-14.
316. Griffiths-Jones, S., et al., miRBase: microRNA sequences, targets and gene
nomenclature. Nucleic Acids Res, 2006. 34(Database issue): p. D140-4.
317. Church, D.M., et al., Lineage-specific biology revealed by a finished genome assembly of
the mouse. PLoS Biol, 2009. 7(5): p. e1000112.
318. Berg, D.J., et al., Enterocolitis and colon cancer in interleukin-10-deficient mice are
associated with aberrant cytokine production and CD4(+) TH1-like responses. J Clin
Invest, 1996. 98(4): p. 1010-20.
319. Fanning, S., et al., Bifidobacterial surface-exopolysaccharide facilitates commensal-host
interaction through immune modulation and pathogen protection. Proc Natl Acad Sci U
S A, 2012. 109(6): p. 2108-13.
320. Collins, J.W., et al., Pre-treatment with Bifidobacterium breve UCC2003 modulates
Citrobacter rodentium-induced colonic inflammation and organ specificity.
Microbiology, 2012. 158(Pt 11): p. 2826-34.
321. Collins, J.W., et al., Fermented dairy products modulate Citrobacter rodentium-induced
colonic hyperplasia. J Infect Dis, 2014. 210(7): p. 1029-41.
322. Chen, C.C., et al., Effect of probiotics Lactobacillus acidophilus on Citrobacter
rodentium colitis: the role of dendritic cells. Pediatr Res, 2009. 65(2): p. 169-75.
134
323. Rodrigues, D.M., et al., Probiotics Are Effective for the Prevention and Treatment of
Citrobacter rodentium-Induced Colitis in Mice. Journal of Infectious Diseases, 2012.
206(1): p. 99-109.
324. Xue, X., et al., Endothelial PAS domain protein 1 activates the inflammatory response in
the intestinal epithelium to promote colitis in mice. Gastroenterology, 2013. 145(4): p.
831-41.
325. Nguyen, H.T., et al., Pathogenic bacteria induce colonic PepT1 expression: an
implication in host defense response. Gastroenterology, 2009. 137(4): p. 1435-47.e1-2.
326. Dai, X., et al., MicroRNA-193a-3p Reduces Intestinal Inflammation in Response to
Microbiota via Down-regulation of Colonic PepT1. J Biol Chem, 2015. 290(26): p.
16099-115.
327. Faul, F., et al., Statistical power analyses using G*Power 3.1: tests for correlation and
regression analyses. Behav Res Methods, 2009. 41(4): p. 1149-60.
328. Bhinder, G., et al., The Citrobacter rodentium mouse model: studying pathogen and host
contributions to infectious colitis. J Vis Exp, 2013(72): p. e50222.
329. Rennick, D.M. and M.M. Fort, Lessons from genetically engineered animal models. XII.
IL-10-deficient (IL-10(-/-) mice and intestinal inflammation. Am J Physiol Gastrointest
Liver Physiol, 2000. 278(6): p. G829-33.
330. Dweep, H., et al., miRWalk--database: prediction of possible miRNA binding sites by
"walking" the genes of three genomes. J Biomed Inform, 2011. 44(5): p. 839-47.
331. Thomas, P.D., et al., PANTHER: a library of protein families and subfamilies indexed by
function. Genome Res, 2003. 13(9): p. 2129-41.
332. Shannon, P., et al., Cytoscape: a software environment for integrated models of
biomolecular interaction networks. Genome Res, 2003. 13(11): p. 2498-504.
333. Livak, K.J. and T.D. Schmittgen, Analysis of relative gene expression data using real-
time quantitative PCR and the 2(-Delta Delta C(T)) Method. Methods, 2001. 25(4): p.
402-8.
334. Pfaffl, M.W., G.W. Horgan, and L. Dempfle, Relative expression software tool (REST)
for group-wise comparison and statistical analysis of relative expression results in real-
time PCR. Nucleic Acids Res, 2002. 30(9): p. e36.
335. He, H., et al., Involvement of G proteins of the Rho family in the regulation of Bcl-2-like
protein expression and caspase 3 activation by Gastrins. Cell Signal, 2008. 20(1): p. 83-
93.
336. He, H. and G.S. Baldwin, Rho GTPases and p21-activated kinase in the regulation of
proliferation and apoptosis by gastrins. Int J Biochem Cell Biol, 2008. 40(10): p. 2018-
22.
135
337. Tripathi, S., et al., The gastrin and cholecystokinin receptors mediated signaling network:
a scaffold for data analysis and new hypotheses on regulatory mechanisms. BMC Syst
Biol, 2015. 9(1): p. 40.
338. Thomadaki, H. and A. Scorilas, BCL2 family of apoptosis-related genes: functions and
clinical implications in cancer. Crit Rev Clin Lab Sci, 2006. 43(1): p. 1-67.
339. Lu, L.F., et al., Function of miR-146a in controlling Treg cell-mediated regulation of Th1
responses. Cell, 2010. 142(6): p. 914-29.
340. Saba, R., D.L. Sorensen, and S.A. Booth, MicroRNA-146a: A Dominant, Negative
Regulator of the Innate Immune Response. Front Immunol, 2014. 5: p. 578.
341. Vazquez-Torres, A., et al., Toll-like receptor 4 dependence of innate and adaptive
immunity to Salmonella: importance of the Kupffer cell network. J Immunol, 2004.
172(10): p. 6202-8.
342. Schilling, J.D., et al., Toll-like receptor 4 on stromal and hematopoietic cells mediates
innate resistance to uropathogenic Escherichia coli. Proc Natl Acad Sci U S A, 2003.
100(7): p. 4203-8.
343. Luperchio, S.A. and D.B. Schauer, Molecular pathogenesis of Citrobacter rodentium and
transmissible murine colonic hyperplasia. Microbes Infect, 2001. 3(4): p. 333-40.
344. Chen, W.X., L.H. Ren, and R.H. Shi, Implication of miRNAs for inflammatory bowel
disease treatment: Systematic review. World J Gastrointest Pathophysiol, 2014. 5(2): p.
63-70.
345. Chapman, C.G. and J. Pekow, The emerging role of miRNAs in inflammatory bowel
disease: a review. Therap Adv Gastroenterol, 2015. 8(1): p. 4-22.
346. Xuan, Y., et al., MicroRNAs in colorectal cancer: small molecules with big functions.
Cancer Lett, 2015. 360(2): p. 89-105.
347. Chen, Y., et al., Altered expression of MiR-148a and MiR-152 in gastrointestinal cancers
and its clinical significance. J Gastrointest Surg, 2010. 14(7): p. 1170-9.
348. Sundaram, P., et al., p53-responsive miR-194 inhibits thrombospondin-1 and promotes
angiogenesis in colon cancers. Cancer Res, 2011. 71(24): p. 7490-501.
349. Zhao, H.J., et al., MiR-194 deregulation contributes to colorectal carcinogenesis via
targeting AKT2 pathway. Theranostics, 2014. 4(12): p. 1193-208.
350. Chiang, Y.P., et al., microRNA-192,-194 and-215 are frequently downregulated in
colorectal cancer. Experimental and Therapeutic Medicine, 2012. 3(3): p. 560-566.
351. Feng, J., et al., miR-150 functions as a tumour suppressor in human colorectal cancer by
targeting c-Myb. J Cell Mol Med, 2014. 18(10): p. 2125-34.
136
352. Takahashi, M., et al., The clinical significance of MiR-148a as a predictive biomarker in
patients with advanced colorectal cancer. PLoS One, 2012. 7(10): p. e46684.
353. Zhai, Z., et al., miR-106b fine tunes ATG16L1 expression and autophagic activity in
intestinal epithelial HCT116 cells. Inflamm Bowel Dis, 2013. 19(11): p. 2295-301.
354. Lu, C., et al., MIR106B and MIR93 prevent removal of bacteria from epithelial cells by
disrupting ATG16L1-mediated autophagy. Gastroenterology, 2014. 146(1): p. 188-99.
355. He, L., et al., A microRNA polycistron as a potential human oncogene. Nature, 2005.
435(7043): p. 828-33.
356. Numata, M., et al., Relationship between RegIV gene expression to outcomes in
colorectal cancer. J Surg Oncol, 2011. 104(2): p. 205-9.
357. Jiang, H., et al., Quantitatively controlling expression of miR-17~92 determines colon
tumor progression in a mouse tumor model. Am J Pathol, 2014. 184(5): p. 1355-68.
358. Humphreys, K.J., et al., Histone deacetylase inhibition in colorectal cancer cells reveals
competing roles for members of the oncogenic miR-17-92 cluster. Mol Carcinog, 2013.
52(6): p. 459-74.
359. Gregory, P.A., et al., An autocrine TGF-beta/ZEB/miR-200 signaling network regulates
establishment and maintenance of epithelial-mesenchymal transition. Mol Biol Cell,
2011. 22(10): p. 1686-98.
360. Gregory, P.A., et al., The miR-200 family and miR-205 regulate epithelial to
mesenchymal transition by targeting ZEB1 and SIP1. Nat Cell Biol, 2008. 10(5): p. 593-
601.
361. Korpal, M., et al., The miR-200 family inhibits epithelial-mesenchymal transition and
cancer cell migration by direct targeting of E-cadherin transcriptional repressors ZEB1
and ZEB2. J Biol Chem, 2008. 283(22): p. 14910-4.
362. Giraud-Triboult, K., et al., Combined mRNA and microRNA profiling reveals that miR-
148a and miR-20b control human mesenchymal stem cell phenotype via EPAS1. Physiol
Genomics, 2011. 43(2): p. 77-86.
363. Ohgushi, M., et al., Transforming growth factor beta-dependent sequential activation of
Smad, Bim, and caspase-9 mediates physiological apoptosis in gastric epithelial cells.
Mol Cell Biol, 2005. 25(22): p. 10017-28.
364. Gross, A., J.M. McDonnell, and S.J. Korsmeyer, BCL-2 family members and the
mitochondria in apoptosis. Genes Dev, 1999. 13(15): p. 1899-911.
365. Pinon, J.D., et al., Bim and Bmf in tissue homeostasis and malignant disease. Oncogene,
2008. 27 Suppl 1: p. S41-52.
366. Sinicrope, F.A., et al., Prognostic impact of bim, puma, and noxa expression in human
colon carcinomas. Clin Cancer Res, 2008. 14(18): p. 5810-8.
137
367. Qin, J.Z., et al., Proteasome inhibitors trigger NOXA-mediated apoptosis in melanoma
and myeloma cells. Cancer Res, 2005. 65(14): p. 6282-93.
368. Kojima, T., et al., FOXO1 and TCF7L2 genes involved in metastasis and poor prognosis
in clear cell renal cell carcinoma. Genes Chromosomes Cancer, 2010. 49(4): p. 379-89.
369. Mogilyansky, E. and I. Rigoutsos, The miR-17/92 cluster: a comprehensive update on its
genomics, genetics, functions and increasingly important and numerous roles in health
and disease. Cell Death Differ, 2013. 20(12): p. 1603-14.
370. Lai, K.W., et al., MicroRNA-130b regulates the tumour suppressor RUNX3 in gastric
cancer. Eur J Cancer, 2010. 46(8): p. 1456-63.
371. Yang, J., et al., Oridonin triggers apoptosis in colorectal carcinoma cells and
suppression of microRNA-32 expression augments oridonin-mediated apoptotic effects.
Biomed Pharmacother, 2015. 72: p. 125-34.
372. Pellegrini, M., et al., Shutdown of an acute T cell immune response to viral infection is
mediated by the proapoptotic Bcl-2 homology 3-only protein Bim. Proc Natl Acad Sci U
S A, 2003. 100(24): p. 14175-80.
373. Hildeman, D.A., et al., Activated T cell death in vivo mediated by proapoptotic bcl-2
family member bim. Immunity, 2002. 16(6): p. 759-67.
374. Haftmann, C., et al., miR-148a is upregulated by Twist1 and T-bet and promotes Th1-cell
survival by regulating the proapoptotic gene Bim. Eur J Immunol, 2015. 45(4): p. 1192-
205.
375. Chow, W.L. and Y.K. Lee, Free fucose is a danger signal to human intestinal epithelial
cells. British Journal of Nutrition, 2008. 99(3): p. 449-454.
376. Ohno, H., et al., Oral administration of Bifidobacterium bifidum G9-1 suppresses total
and antigen specific immunoglobulin E production in mice. Biol Pharm Bull, 2005.
28(8): p. 1462-6.
377. Yildirim, Z., D.K. Winters, and M.G. Johnson, Purification, amino acid sequence and
mode of action of bifidocin B produced by Bifidobacterium bifidum NCFB 1454. Journal
of Applied Microbiology, 1999. 86(1): p. 45-54.
378. Singh, N., et al., Impact of Bifidobacterium bifidum MIMBb75 on mouse intestinal
microorganisms. FEMS Microbiol Ecol, 2013. 85(2): p. 369-75.
379. Teitelbaum, A.A., et al., Chronic peripheral administration of corticotropin-releasing
factor causes colonic barrier dysfunction similar to psychological stress. Am J Physiol
Gastrointest Liver Physiol, 2008. 295(3): p. G452-9.
380. Yan, J., et al., MiR-148a regulates MEG3 in gastric cancer by targeting DNA
methyltransferase 1. Med Oncol, 2014. 31(3): p. 879.
138
381. Ooi, J.H., et al., Dominant effects of the diet on the microbiome and the local and
systemic immune response in mice. PLoS One, 2014. 9(1): p. e86366.
382. King, T.S., J.T. Woolner, and J.O. Hunter, Review article: the dietary management of
Crohn's disease. Aliment Pharmacol Ther, 1997. 11(1): p. 17-31.
383. Leamon, C.P., et al., Impact of high and low folate diets on tissue folate receptor levels
and antitumor responses toward folate-drug conjugates. J Pharmacol Exp Ther, 2008.
327(3): p. 918-25.
384. Hrncir, T., et al., Gut microbiota and lipopolysaccharide content of the diet influence
development of regulatory T cells: studies in germ-free mice. BMC Immunol, 2008. 9: p.
65.
385. Wang, L., et al., Methods to determine intestinal permeability and bacterial translocation
during liver disease. J Immunol Methods, 2015. 421: p. 44-53.
386. Hartmann, P., et al., Deficiency of intestinal mucin-2 ameliorates experimental alcoholic
liver disease in mice. Hepatology, 2013. 58(1): p. 108-19.
387. Lee, R.C., R.L. Feinbaum, and V. Ambros, The C. elegans heterochronic gene lin-4
encodes small RNAs with antisense complementarity to lin-14. Cell, 1993. 75(5): p. 843-
54.
388. Lagos-Quintana, M., et al., Identification of novel genes coding for small expressed
RNAs. Science, 2001. 294(5543): p. 853-8.
389. Schaefer, J.S., et al., MicroRNA signatures differentiate Crohn's disease from ulcerative
colitis. BMC Immunol, 2015. 16: p. 5.
390. Toiyama, Y., et al., Serum miR-21 as a diagnostic and prognostic biomarker in
colorectal cancer. J Natl Cancer Inst, 2013. 105(12): p. 849-59.