When expression is not enough

Two articles in this edition of Inflammatory Bowel Diseases advance the debate over whether Fas/FasL interactions underlie intestinal epithelial cell apoptosis during colitis. Chen et al present a state-of-the-art opinion on the evidence for a role for Fas/FasL in apoptosis and cancer but also on other possible mechanisms. Then the same group provides an original research article suggesting that the Fas/FasL mechanism in the dextran sulfate sodium (DSS)-induced colitis model in mice is in fact cytoprotective for epithelial cells. Cells in the healthy colon epithelium undergo apoptosis in the normal course of differentiation and migration to the surface epithelium, presumably making room for the progeny of the dividing stem cells in the crypt. This understanding has made the intestinal epithelium a popular model for dissecting mechanisms of apoptosis, but whether mechanisms of homeostatic apoptosis versus during pathological conditions such as inflammatory bowel disease (IBD) are the same is not entirely clear. This is despite evidence for increasing numbers of apoptotic cells comparing the healthy colon crypts with uninvolved crypts and involved crypts from ulcerative colitis patients1 and involved areas in the Crohn's disease patient colon.2 It is unclear whether this increase in apoptosis is an intrinsic response of the cells to a direct assault or a consequence of extrinsic stimuli from cells such as lymphocytes or neutrophils infiltrating the mucosa. One extrinsic means of effecting apoptosis that has drawn the research community's attention in colitis is the Fas (CD95) / FasL (CD95 ligand) pathway. Two articles in this edition advance the debate over whether Fas/FasL interactions underlie intestinal epithelial cell (IEC) apoptosis during colitis.3,4 Chen et al3 present their state-of-the-art opinion on the evidence for a role for Fas/FasL in apoptosis and cancer but also on other possible mechanisms. Then the same group provides an original research article exploring the contribution of the Fas/FasL mechanism in the dextran sulfate sodium (DSS)-induced colitis model in mice.4 Fas is a transmembrane tumor necrosis factor (TNF) receptor family protein possessing a death domain that assembles a multimeric death-inducing signaling complex (DISC). The product of DISC assembly is the cleavage of procaspase 8 into a heterotetramer constituted with 2 p18 and 2 p10 subunits. Activated caspase 8 cleaves downstream caspases, including caspase 3, to effect apoptosis. Two pathways have been described in different cell types depending on the relative amount of the heterotetramer made. One pathway, Type I, is invoked by membrane-bound FasL and relatively copious amounts of the heterotetramer are released into the cytoplasm. The Type II pathway can be invoked by membrane-bound FasL or soluble FasL, liberates less heterotetramer that cleaves Bid, which then recruits the mitochondria mechanistically in apoptosis. XIAP can bridge the two responses since deficiency converts Type II responding cells into Type I cells. Mice with deficiencies in Fas (lpr) and FasL (gld) are available and have been used in models of colitis. The Park et al4 research is the first to use these mice combined with the DSS-induced colitis model. The lpr mutation results in loss of Fas expression. These mice are reported to experience severe tri-nitrobenzene sulphonic acid (TNBS) colitis that becomes lethal and includes high levels of interferon-γ.5 The gld mice express a mutated FasL and reportedly heightened TNBS colitis, including greater ulcers, but only when wildtype mice were already recovering (7 days following intracolonic TNBS).6 Crossing lpr with IL-2 deficiency reversed the spontaneous colitis and lethality due to the cytokine deficiency.7 This was concomitant with reduced apoptosis of the IEC. Mice made deficient in both fas and fasL became diseased with the hapten di-nitrobenzene sulphonic acid (DNBS).8 Notwithstanding the differences in how the hypersensitivity model is applied, these published findings do not result in a unifying model, with different deficiencies resulting in an inconsistent response to colitis. When examined, IEC constitutively express fas but also fasL during inflammation (IEC with heightened expression of fasL led to the engaging idea that tumor cells may be armed for a “Fas counter-attack” against infiltrating T cells, see Ref. 9). The different outcomes underscore an emerging opinion that soluble versus membrane-bound Fas may have unique biological roles and possibly tissue-specific functions, and a recent observation introduced the concept that membrane-bound FasL may in fact signal.3 Thus, there is a need to examine the different mutations in the same protocol and to directly address the question of whether Fas expression by IEC is critical to the severity of colitis. The study by Park et al4 fills this criteria, using DSS-induced colitis. As DSS-induced pathology typically includes considerable ulceration, this model is widely thought to be the product of direct epithelial cell cytotoxicity by the DSS. This idea is difficult to reconcile in view of the site selectivity for colitis to develop in the mid-colon when mice are ingesting DSS. More important, there is evidence to the contrary, that the ulcers are a product of the host response. For example, interferon-γ gene-deleted mice and mice lacking the CXCR2 gene or treated with blocking antibodies to CXCR2 develop few if any ulcers.10,11 Additionally, IEC resilience to apoptosis during DSS colitis can be manipulated. One example is that IEC-specific Raf knockout mice experience heightened apoptosis early during DSS treatment,12 arguing that the mechanism of apoptosis is highly regulated. Park et al4 used the DSS-induced colitis model in mice and utilized an additional mutant strain, the lprcg, which expresses Fas that does not signal. Challenged with 2% DSS, gld mice proved less sensitive, while lpr mice were equally sensitive as wildtype (C57BL/6). These results contrast the published experience using the hapten model. The lprcg mice (on the C3H genetic background) had similar disease to the wildtype controls, both receiving 2.5% DSS. On the other hand, irradiated lprcg mice reconstituted with bone marrow from wildtype mice proved to be more sensitive to acute DSS than wildtype mice reconstituted with wildtype bone marrow. The alternate chimera, lprcg bone marrow into irradiated wildtype mice, was not reported. Their findings immediately challenge the notion that expression of Fas on IEC underlies disease. To further refine the focus on IEC they established IEC-specific CD95 knockout mice by crossing villin/cre animals with CD95loxP mice. These mice again proved to be more sensitive to DSS, including a failure to recover after the first DSS cycle. Considering the increased sensitivity to DSS of the IEC CD95 knockout mice they next assessed whether the lack of expression increased the animals' risk of developing colon carcinoma. Using the azoxymethane plus DSS regime there was no difference in dysplastic lesions comparing the 2 strains, leading to the conclusion that IEC CD95 does not contribute to the emergence of carcinoma in this model. The overall conclusion is expressed in the provocative title, “CD95 is cytoprotective for intestinal epithelial cells in colitis.” This apparent reversal in thinking about the role of Fas/FasL in IEC apoptosis during colitis is certain to garner further attention.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.