DSC8262 compressed


My research program addresses how the innate immune system functions within the mammalian gastrointestinal tract, primarily studying how it recognizes and responds to enteric microbes. To address these areas, I have developed significant expertise in the generation and study of mouse models of intestinal inflammation, induced in response to either commensal or pathogenic microbes. My studies are thus highly relevant to enteric infectious diseases as well as to idiopathic conditions thought to involve bacteria, such as Inflammatory Bowel Disease (IBD). Although there are many microbiologists and immunologists in Canada, I am one of the few Canadian researchers that possess expertise in both fields. Moreover, my focus on the use of in vivo models to study intestinal inflammation makes my research program unique in Canadian gastroenterology. Unlike many other researchers who study bacterial infections in order to understand the pathogen, my primary goal is to use these models to understand how the immune system functions within the mammalian gastrointestinal tract. While this has significant relevance to a number of gastrointestinal diseases, we have also begun to assess the potential role of enteric microbes in causing type 1 diabetes, through the disruption of intestinal barrier function and the activation of previously quiescent yet autoreactive T lymphocytes.

Inflammatory Bowel Disease

My research on IBD has three main goals. The first is to understand how inflammation in the GI tract is regulated. The second is to use bacterial infections to model tissue pathology seen in IBD, such as tissue repair, intestinal fibrosis and goblet cell depletion. The last is to understand how commensal bacteria can trigger inflammation during Crohn’s disease and Ulcerative colitis.

Innate Immunity and Inflammatory Bowel Disease

Numerous studies have shown that colonic epithelial cells are hypo-responsive to most bacterial products. In recent collaborative studies with Dr. Xiaoxia Li (Cleveland Clinic, USA) we determined that this hypo-responsiveness is in part due to the expression of single immunoglobulin IL-1 receptor related molecule (SIGIRR), a negative regulator of TLR4 and IL-1β signalling. Expressed primarily by colonic epithelial cells, we recently showed that SIGIRR is critical in limiting inflammatory and epithelial proliferative responses to commensal bacteria during dextran sodium sulfate (DSS) induced colitis. In additional studies we have shown that SIGIRR negatively regulates most TLRs expressed by colonic epithelial cells and through the use of human biopsy samples, we have shown that SIGIRR is expressed primarily by mature epithelial cells at the apex of colonic crypts. We are currently assessing what factors induce SIGIRR expression in the GI tract and studying whether its expression or function is aberrant in the intestines of patients with IBD.

Intestinal Fibrosis/Stricturing in IBD

So far, my demonstration that the innate immune response to intestinal bacteria helps promote tissue repair in the gut has identified novel mechanisms of host susceptibility to bacterial pathogens. Moreover these studies are also relevant to intestinal fibrosis, a condition affecting many of the 200,000 Canadians with Inflammatory Bowel Disease. Using a Salmonella typhimurium driven model of intestinal fibrosis, we have found that Toll-like receptors play a critical role in the development of fibrosis, contributing to both fibroblast recruitment as well as collagen deposition. We are currently studying how bacterial products activate immune cells to precipitate fibrosis. We are also using both mouse models and human tissues to examine the mechanisms underlying the recruitment/proliferation of fibroblasts in the inflamed gut, as well as the activation of these fibroblasts to cause chronic fibrosis.

Goblet Cells Prevent Bacterial Induced Colitis Including IBD

I have also developed a research program studying the role of goblet cells (a subtype of intestinal epithelial cell) during this infection. We have found significant changes in the expression of goblet cell mediators during both Citrobacter rodentium (Bergstrom et al., Infect & Immun., 2008) and S typhimurium (in preparation) infection. Moreover we are writing a manuscript showing that goblet cell derived mucins play a critical role in controlling pathogen burdens during C. rodentium infection. Future studies will address the roles played by mucins in limiting commensal bacterial interactions with the mucosal immune system, and whether changes in mucin expression during IBD may be a contributing causal factor in these diseases.

Enteric infectious diseases

My infectious disease research focuses on two types of enteric bacterial pathogens, (1) attaching/effacing (A/E) bacteria including enteropathogenic E. coli and Citrobacter rodentium and (2), the invasive bacterial pathogen Salmonella typhimurium. We have studied the roles of innate receptors in providing host defense against these pathogens. In addition, we have also developed approaches to image these bacterial infections in live hosts as well as assess bacterial virulence gene expression within the mammalian intestine. Moreover, we have identified bacterial virulence proteins that suppress normally protective host inflammatory responses. Our goal is to define the complex interactions between pathogen and host that determine the relative success of the infectious process.

Innate Defenses That Protect Against Luminal Bacterial Pathogens

To investigate the role of the innate immune system in the host response to C. rodentium infection, I began by assessing which toll-like receptors were activated by this pathogen. As expected for a non-motile Gram-negative microbe, we found C. rodentium activates toll-like receptors 2, 4 and 9 but not 5. Assessment of mice lacking most TLR signaling (MyD88 deficient mice) found significant deficiencies in host inflammatory responses, control of pathogen burdens and in controlling mucosal homeostasis, as demonstrated by the development of widespread ulcerations (Gibson et al, Cell Micro., 2008). We have determined that TLR4 plays the primary pro-inflammatory role in this infection (Khan et al, Infect & Immun., 2006) and we determined that TLR2 plays a critical role in maintaining mucosal integrity (Gibson et al, Cell Micro., 2008). Our current studies are addressing the mechanisms involved; however the severity of the phenotype in the MyD88 deficient mice suggests other pathways participate in the host response. We are now assessing whether TLR2 and TLR4 compensate for each other, such that mice lacking both receptors would recapitulate the phenotype seen in MyD88 deficient mice. Alternatively, TLR9, IL-1β and IL-18 (all signal through MyD88) may also be involved in the control of mucosal function and host defense.

Subversion of Intestinal Host Defenses

During my post-doctoral studies, I found that intestinal epithelial cells directly infected by C. rodentium did not express the pro-inflammatory enzyme iNOS, whereas nearby uninfected cells did express this enzyme. This finding led to my hypothesis that C. rodentium and other attaching/effacing (A/E) bacterial pathogens injected specific translocated effector proteins into epithelial cells to suppress their inflammatory and anti-microbial response, creating a protected niche within the infected gastrointestinal tract. We as well as other groups have confirmed that epithelial inflammatory responses are suppressed by the related human A/E pathogen enteropathogenic E. coli (EPEC) in a manner dependent on their type 3 secretion system (T3SS), however the specific effector responsible was unknown. In the last 6 months, we have identified the effector that is responsible for this phenotype and have confirmed that it suppresses epithelial inflammatory responses to A/E pathogens, both in vitro and in vivo. We are currently studying what role it plays in the infectious process and in bacterial virulence.

Imaging of Infection and Bacterial Virulence Gene Expression In Vivo

In collaborative studies, we have developed imaging approaches to study the temporal-spatial course of in vivo C. rodentium and S. typhimurium infections, as well as the in vivo expression of specific bacterial virulence genes. We plan to use both in vivo imaging as well as multiphoton microscopy to define the complex interactions between pathogens and hosts that determine the success of the infectious process, as well as the susceptibility/resistance of the host to infection.

Role of Enteric Bacteria in Causing Type 1 Diabetes

Another original aspect of my research is my exploration of the putative role played by enteric microbes in the pathogenesis of type 1 diabetes (T1D). Performed in collaboration with diabetes researcher Dr. Jan Dutz, this research was recently funded by a 3 year grant from the Juvenile Diabetes Research Foundation (JDRF). We have found that C. rodentium infection of mice that are susceptible to T1D leads to an acceleration of pancreatic inflammation and the increased activation of diabetogenic T cells in these mice. We are currently studying whether this is due to the disruption of intestinal epithelial barrier function, or the inflammation that accompanies infection. We hope that further investigation of the mechanisms involved will help decipher the potential role played by pathogenic and commensal microbes in modulating T1D.



Figure 1. Intestinal gene transfer using adenoviral vectors. Intestinal serosal cells inected by adenovirus and expressing the transgene beta-galactosidase stain blue.


Figure 2. A thick layer of mucus (purple) covering epithelial cells provides the first protective barrier in the colon, preventing bacteria in the lumen of the gut from entering the underlying tissue. Note the mucin filled goblet cells (also purple) that make up approximately 25% of the cells in the colonic epithelium.


Figure 3. Immunolocalization of C. rodentium Map in the infected colonic epithelium. Panel A shows merged immunofluorescence staining of the C. rodentium effectors Tir (green) and Map (red), while host cell nuclei are stained in blue. Note that Tir is found on the epithelial surface, while Map is found in a punctuate pattern in the cytoplasm, indicating localization with mitochondria.


Figure 4. Salmonella infection induced intestinal fibrosis as a model of Crohn’s stricturing. Note that uninfected (cntrl) caecal tissues exhibit very little collagen staining (blue), however the caecum of mice at day 21 post-infection show dramatic mucosal thickening, inflammation, edema and collagen staining (blue). Note most of the collagen is found in the submucosa (SM).


C. rodentium infection leads to significant epithelial proliferation. Distal colonic tissue sections were stained for Ki67 (red), and host cell nuclei by DAPI (blue), and C. rodentium tir (green). Infection leads to a dramatic and significant increase in Ki67 positive epithelial cells in the distal colons of C57BL/6 mice.

Claudin-3 staining in the mouse colon. Claudin-3 (red) was found laterally distributed in the mid region of the colon from C57BL/6 both before and during C. rodentium infection. Host cell nuclei are stained in blue.

Figure 5. Citrobacter rodentium infection leads to disruption of mitochondrial structure in infected colonic epithelial cells. Panels A and B – uninfected colon, note the dark, amorphous shaped mictochondria. Panels C and D – C. rodentium infected colonic epithelial cells, note the swollen and distorted mitochondria. Panels E and F. This disruption of mictochondrial structure depends on the translcoated effector Map, as cells infected by Δmap C. rodentium do not suffer the mitochondrial disruption.