Our understanding of the mechanisms and pathways involved in human inflammatory bowel disease (IBD) has grown exponentially in recent decades, leading to the identification of a seemingly never-ending list of potential ‘drugable’ targets. We have learned from the past, however, that all that glitters is not gold and we should exercise some caution when translating results from the bench to the bedside. The most recent example of a promising target gone wrong in IBD concerns IL-17.1

We tend to assume that molecules that are up-regulated during inflammation should constitute potential targets to block. While this may seem logical at first, a more in depth understanding of the mechanisms at play reveals that: (1) during inflammation many regulated genes actually exert protective, regenerative or anti-inflammatory actions; (2) a proportion of those inflammation-dependent genes remain altered even during remission in IBD patients.2 Therefore, in addition to its over-expression, obtaining evidence of the in vivo function of a given pathway is essential before considering the development of a blocking strategy for therapeutic purposes. Unfortunately, the effects of blocking cytokines in vivo have proven highly unpredictable. Cytokines are small proteins well known for their multiple cellular sources, pleiotropic nature and highly context-dependent behaviour, making the consequences of their blockage especially unpredictable. One should avoid rushing into clinical development, especially when different models yield contradictory results. Again, this proved to be the case with IL-17, the blockade of which resulted in protection or no effect or even worsening of the disease, depending on the experimental model of colitis employed.3

The case for IL-33/ST2 in IBD

In spite of various difficulties encountered, or actually because of them, the search for new and improved therapeutic targets in IBD is far from over. Ideal targets consist of upstream drivers of the disease whose blockade effectively induces mucosal healing, the gold standard of disease management. In that respect, data accumulated thus far on the IL-33/ST2 axis in driving mucosal inflammation warrants its further study.

IL-33 is a cytokine produced under un-inflamed conditions by non-hematopoietic cells. This protein can act as a nuclear factor. In response to stress, IL-33 is released and promotes Th2 responses by binding to its receptor ST2, an IL-1 receptor-related protein.4 Indeed, IL-33 administration in vivo induces an increase in IgA and IgE serum levels, eosinophilia, and respiratory and intestinal mucosa hyperplasia, all of which are IL-13-dependent.4

More recent studies have shown a role for IL-33 in mucosal inflammation. In that sense, IL-33, ST2 mRNA, and protein have been consistently shown to be up-regulated in the inflamed intestinal mucosa of ulcerative colitis5–9 and, to a lesser degree, in Crohn's disease.7 ,10 Importantly, IL-33 serum levels, which were also found to be elevated in patients with ulcerative colitis, were significantly reduced after anti-TNF-α treatment.7

Immunohistochemistry analysis points to epithelial cells as the main IL-33 producers in a healthy colon, while both epithelial and lamina propria cells become positive for IL-33 antibodies in the inflamed IBD mucosa. The receptor for IL-33, ST2, is primarily expressed by epithelial cells in healthy mucosa whereas, according to one report, epithelial cell expression is lost in IBD with lamina propria cells becoming strongly ST2 positive.7

While this is all descriptive data, the functional role of IL-33 signalling through its receptor has been described in experimental models, specifically in a paper by Sedhom and colleagues.10 These experiments using IL-33- or ST2-deficient mice, bone marrow chimeras, or blocking antibodies, unanimously support a role for the IL-33 pathway in promoting inflammation, particularly by delaying disease resolution.10 ,11 The exact mechanisms, however, remain unclear.

What could make IL-33 an interesting target in IBD is its potential role in the early acute, as well as in the more chronic, phases of the inflammatory response. IL-33 is initially released by epithelial cells as an ‘alarmin’ in response to tissue injury and can act on innate cells.11 Later on, IL-33 is produced by lamina propria cells in response to inflammatory cytokines such as TNF-α and IL-1β.4 In addition, IL-33 can itself induce production of mediators such as TNF-α, IL-1β and IFNγ in different cell types, further contributing to amplifying the inflammatory cascade and potentially antagonising disease resolution. On its own, IL-33 may act by promoting epithelial cell damage, as suggested by Sedhom et al,10 and by impairing recovery of the epithelial layer, a key step in achieving mucosal healing. In this regard, the paper by Sedhom and coworkers sheds new light on the relevance of IL-33/ST2 signals in the pathogenesis of tissue damage, in proving that extracellular IL-33 plays a role in increasing intestinal epithelial cell permeability. Importantly, epithelial cell regeneration appears to be impaired in ST2−/− mice. In addition, data on other tissues predicts a role for IL-33 in promoting tissue fibrosis, potentially by inducing Th2 responses.12 Given that no effective pharmacological treatment is available to reverse the development of intestinal fibrosis, exploring anti-IL-33/ST2 therapies may be a worthwhile pursuit.

Finally, another important issue is whether the blockade of IL-33 or ST2 is equally efficacious and results in the same biological effects. Available information on the in vivo effects of the different IL-33 isoforms indicates that the mature IL-33 form can promote ST2-independent effects,13 thus suggesting that blocking of the cytokine or its receptor should be independently tested.

In summary, the available data conclusively shows that IL-33 is up-regulated during intestinal inflammation, both in patients and in models of intestinal inflammation. The therapeutic basis and potential for the IL-33/ST2 blockade may reside in the fact that IL-33 is released by epithelial cells as an ‘alarmin’ in response to injury, while other cell sources in the lamina propria produce IL-33 in chronic IBD patients. This could make the IL-33 pathway a desirable target for both the acute and chronic phases of the response.

As with other immunosuppressive therapies, the effect of blocking this pathway on immune defence against infection, particularly of parasitical or bacterial origin, needs to be evaluated. Furthermore, IL-33 has been reported to exert cardioprotective effects, and extreme caution should be exercised with patients with cardiovascular risk factors.