We thank Wen et al. (1) for their comments regarding our recent article (2). Although SAMP1/YitFc mice have many features that closely resemble human Crohn’s disease including pathological and histologic features, cytokine profile activation, and beneficial response to treatments that are effective in Crohn’s disease patients (3, 4), we agree that it is possible that resistin-like molecule β (RELM-β) could be expressed differently in humans and mice. In our discussion, we raise the possibility that RELM-β could be a mediator of intestinal inflammation in humans as a focus for stimulating further studies in this area. We are aware of no published studies in humans that directly address the role of RELM-β in human intestinal inflammation. However, our data suggest the possibility that RELM-β could have a role in human inflammatory bowel disease (IBD) and what that role may be is worthy of further examination.
Wen et al. also assert that their studies revealed no differences in RELM-β mRNA expression between patients with either Crohn’s disease or ulcerative colitis and controls and therefore “an altered expression level of RELM-β is not associated with IBD in humans.” It is difficult for us to comment specifically on the conclusions of Wen et al. in this regard since their study has not been published. However, it is important to note that many variables could affect the outcome of such a clinical study in humans. For example, the clinical characteristics of the patient population examined, disease heterogeneity, concomitant therapy, adequacy of the methodology used, and sufficiency of the sampling size to allow for detection of differences could all have important effects on the outcome and interpretation of such studies. Therefore, in our view the relationship of RELM-β to the pathogenesis of human intestinal inflammation and IBD remains an open question.
Finally, the participation of RELM-β in initiating, mediating, or exacerbating inflammation may not necessarily require up-regulation of its expression. A recent study by our group (5) demonstrated that a vigorous inflammatory response of the colon could be induced by dextran sodium sulfate in the presence of constitutive levels of colonic RELM-β that were not different from those found in untreated control mice, whereas inflammation was attenuated in mice lacking this protein. Dextran sodium sulfate induced damage to the barrier function of the colon could allow RELM-β access to the lamina propria from which it is normally excluded. This observation suggests that it may not be the absolute levels of RELM-β that are important but rather the localization of RELM-β and its access to cells of the innate and/or adaptive immune system within the intestine.
It is clear from recently published studies that much more work is needed, utilizing both animal models and samples from IBD patients and their appropriate controls, before any firm conclusions can be drawn regarding the role of RELM-β in human IBD. Nevertheless, our current study does present some intriguing possibilities and food for thought.