Experimental mouse models of inflammatory bowel disease: New insights into pathogenic mechanisms

Abstract

Inflammatory bowel disease (IBD) is a term that comprises two chronic inflammatory diseases of humans, namely ulcerative colitis and Crohn's disease. Despite many years of study the exact etiology and pathogenesis of these disorders have remained elusive. Various infections can mimic the tissue pathology seen in each of these diseases; however, no organism or infectious cause has yet been identified. As with chronic inflammatory disorders in other organs, these diseases appear to involve complex interactions among immunologic, environmental, and genetic components. The early inductive phases of these diseases are particularly difficult to study in humans because patients come to clinic only after their symptoms are established. Experimental models in animals have a number of advantages in this regard, in that the environmental conditions and genetics can be either controlled or defined. No animal model exactly reproduces human IBD, nor could it. There cannot be precise models of imprecise, ill-defined diseases. The value of the models, as is illustrated in this chapter, is the insight they allow into the complex, multifaceted processes and mechanisms that can result in chronic intestinal inflammation. In recent years quite a number of new experimental models of chronic intestinal inflammation have been described. The focus here will be mainly on these new models. The reader is referred to a previous review for others not discussed here [1]. Knowledge gained from these experimental models has already resulted in the development of new hypotheses and therapies that are being tested in patients with IBD. The growth in this area is remarkable considering that the first such model of IBD was reported only as recently as 1994. Most of these new models involve some form of genetic manipulation, either insertion (transgenic) or selective deletion (knockout) of a gene. Mice resulting from such genetic manipulation are now collectively referred to as 'induced mutants' to distinguish them from mice with spontaneously occurring mutations. The induced mutants that develop IBD, usually in the absence of any further manipulation, represent a small subset of the total number of immune system genes that have been mutated. This argues that the mutations that have resulted in disease must represent genes involved in pathways critical to the maintenance of mucosal homeostasis. The results that have been obtained from these models to date provide strong support for the immunologic hypothesis that a dysregulated mucosal immune response, particularly a CD4+ T cell response, to antigens of the enteric bacteria in a genetically susceptible host result in chronic intestinal inflammation. This hypothesis has been advanced in various forms over the past several decades as knowledge of the immune system grew, as did the recognition that many microbial products are intensely stimulatory for immune cells. Given the large number and variety of microbes resident in the intestine, which outnumber the cells in the body by 10 to 1, the mystery has been why all of us do not have IBD. The answer to that mystery is now emerging through study of these new models. These studies have shown that the host interaction with the flora is complex, but that there are a select number of cells and molecules that are critical to this effort. When these key pathways are impaired, the host response to the bacterial flora results in IBD. That the bacterial flora can cause IBD under these circumstances has been demonstrated in multiple models in which animals rendered germfree do not contract IBD unless they are reconstituted with an enteric bacterial flora. There is clearly a genetic influence to this host response, in that certain strains are much more susceptible to develop colitis than others. This has been shown to be due to the presence of modifier genes that confer susceptibility in some strains. (Figure presented) Another important finding from these models is that the key effector cell responsible for disease in most instances is the CD4+ T cell. CD4 T cells can be placed into several functional subsets based on the types of cytokines that they produce (Fig. 1). Thl cells produce IFN-γ, IL-2, and TNF-β, cytokines that are important in delayed hypersensitivity and cellular immunity. Th2 cells produce instead IL-4, IL-5, IL-6, IL-10, and IL-13, cytokines important in humoral immunity. The type of CD4+ T cell response to a given pathogen can be critical for the host in that inbred mouse strains that respond to a microbial pathogen such as Leishmania major with a Th2 response succumb to the infection and die, whereas inbred strains that respond to the same infection with a predominant Thl response recover and are resistant to reinfection [2]. Although there are data in some systems showing that Thl and Th2 subsets can reciprocally regulate one another via IFN-γ inhibition of Th2 responses and IL-10 inhibition of Thl responses, each of these CD4 T cell effector CD4 T cell subsets has been found to mediate colitis in various mouse models. There are no data at present that demonstrate that Th2 cells regulate Thl cells in the intestine or vice versa; thus, experimental colitis is not explained as an imbalance between Thl and Th2 subsets. To the contrary, administration of exogenous IL-4 has been shown to exacerbate Thlmediated colitis in one model [3]. At present, the data are compatible with the concept that excessive responses of either the Thl or Th2 effector subsets are detrimental and can result in IBD. The tissue damage resulting from the CD4 + T cell in either case is likely to be indirectly mediated through cytokines rather than by direct cytotoxicity, and a critical molecule in the Thl pathway is likely to be TNF-α [4]. In regard to application of the Thl /Th2 paradigm to human disease, Crohn's disease (CD) does appear to be mediated by Thl effector cells. Mucosal lesions in CD have increased numbers of IFN-y-producing T cells, increased IL-12 mRNA levels, and increased levels of the STAT-4 transcription factor which is the intracellular messenger of IL-12 signaling. Ulcerative colitis has been called Th2-like' because of increased T cell production of IL-5 by cells isolated from the mucosa [5], but does not fit the Th2 pattern in that there is no increase in IL-4, the hallmark Th2 cytokine. Nevertheless, the data being generated in these models are highly germane to our understanding of IBD in humans. Additional CD4+ T cell subsets with regulatory activity for CD4+ effector T cells have been identified in recent years. The effects of regulatory cells have been seen in various experimental systems for many years, but because they have been difficult to isolate and culture, their existence has been questioned. These cells are now being isolated and characterized. It is unclear how many distinct subsets of regulatory CD4 T cells there are, but three deserve special mention. T-regulatory-1 (Trl) cells produce high levels of IL-10, variable levels of TGF-β1, low levels of IFN and IL-4, and no IL-2. They are generated by chronic activation of CD4+ T cells in the presence of IL-10. Trl cells were originally identified in humans [6] and subsequently in mice [7]. Trl cells have been shown to be capable of suppressing the induction of colitis in the CD45RB transfer model, which is discussed below. Another regulatory subset that has been found to be important in intestinal regulation are T-helper-3 cells (Th3) that produce high amounts of active TGF-β1 when stimulated with specific antigen. Th3 cells have been identified in experiments examining mechanisms of oral tolerance induced by autoantigen feeding in mice [8, 9] and in the peripheral blood of humans with multiple sclerosis [10]. Th3 cells mediate protective oral tolerance in mice fed TNP-colon proteins prior to induction of TNBSinduced colitis [11]. Lastly, there is evidence that deficient mucosal TGF-β1 production is associated with the development of TNP-KLH-induced colitis in IL-2-deficient mice [12], suggesting that this form of experimental colitis is due to lack of development of the Th3 subset. A third important T regulatory subset is the CD4+, CD25+ lineage that is generated in the thymus early in life. This subset comprises some 10% of peripheral CD4+ T cells in adults. In mice this subset has been shown to maintain peripheral tolerance to autoantigens. It may play a similar role in regulating the mucosal immune response to commensal bacterial antigens but that is not yet estabhshed. It is also unknown whether this subset is the precursor for Trl and/or Th3 regulatory CD4+ T cells. Although the experimental data indicate that CD4 + T cells are the key regulatory cells in the intestine, other cell types very likely also contribute to maintaining mucosal homeostasis. Data supporting such a role for CD8+ T cells [13, 14], NK cells [15], NK-T cells [16], and B cells/antibody [17] have been generated in various experimental model systems. © 2005 Springer Science+Business Media, Inc.