Microbial Factors in the Pathogenesis of IBD

Abstract

The etiopathogenesis of both forms of inflammatory bowel disease (IBD), ulcerative colitis (UC) and Crohn's disease (CD), is still unclear, but it is now generally believed that IBD results from the interplay of genetic, environmental, microbial, and immune factors. Ever since UC and CD were recognized as distinct clinical entities, the notion that microbial agents were involved in the causation of chronic intestinal inflammation has been continuously considered. Because of its similarities with infectious colitis, UC was initially considered as an infectious disease. When CD was first described, because of its resemblance to intestinal tuberculosis, it was also suspected to have an infectious cause. Unsuccessful early attempts to isolate common infectious agents from IBD tissues were followed by the notion that IBD could be caused by an abnormal reaction against the enteric flora in the 60's through the 70's, a period of time when evidence emerged that an abnormal immune response can lead to autoimmunity. Thus, a hypothesis was entertained that patients with UC or CD were reacting against antigens on gut epithelial cells cross-reating with enterobacterial antigens. The presence of circulating antibodies against E. coli in IBD patients was initially demonstrated and followed by evidence of cell-mediated immunity against an "enterobacterial common antigen". Because of lack of specificity and concrete evidence that the antibodies or the T-cells reacting against bacterial antigens were pathogenic, both responses were eventually dismissed as simple epiphenomena. The search for novel or atypical microbial agents went on, and in the 80's two new putative infectious agents received considerable attention: M. paratuberculosis and the measles virus. A group of investigators in the USA recovered atypical mycobacteria from a few CD tissues, and a group in the United Kingdom claimed that the CD granulomas containded viral particle resembling the measles virus. These findings were followed by a wave or research worldwide, but the bulk of the evidence collected so far does not support that these two infectious agents are responsible for any sizable portion of CD cases. In the 90's the concept that microbes could be involved in IBD pathogenesis was drastically re-emphasized by the advent of animal models of IBD. A variety of models were created and, in spite of vastly differente underlying defects or mechanisms of induction, they shared one crucial feature: practically all failed to develop colitis when animals were kept in a germ-free environment. This observation immediately emphasized that commensal bacterial were indispensable to experimental colitis, opening a whole new way of thinking about microbes in IBD. At the same time, a series of clinical observations provided complementary impetus to the importance of the flora in IBD, and CD in particular. Loss of tolerance against the endogenous flora was reported in both experimental and human IBD. Antibiotics were shown to have some efficacy in IBD. Loops of bowel that reamain excluded from the normal fecal stream in CD fail to develop inflammation, but they do so almost immediately after reanastomosis. The development of inflammation (pouchitis) in the reservoir constructed after proctocolectomy in UC patients, where an ileal mucosa suddenly becomes exposed to colon-like flora, also supports the importance of resident bacteria in the development of chronic intestinal inflammation. Finally, with the advent of probiotics in the treatment of pouchitis, the idea that interfering with the resident flora somehow beneficially modifies an abnormal balance with the host consolidated the concept that the gut flora is centrally involved in IBD pathogenesis. At the moment there is little doubt that microbes represent an integral component of IBD pathogenesis, but many challenges remain to understand why and how some individuals react abnormally against their own flora and end up with UC or CD. The recent discovery of mutations in the NOD2 gene in a subgroup of CD patients is giving some insight into how the host responds pathologically to endogenous bacteria. A cricital task is to define the composition of the normal flora in humans and detect quantitative or qualitative variations that may be present in IBD patients. Finally, it must be determined whether the flora in IBD patients is truly normal, putting the burden of pathogenesis on the host immune response, or subtle changes in microbe composition or antigenicity may exist that trigger inflammation under the influence or environmental and genetic factors.