Finally, function for “IL-1Fs”

So, another cytokine called “interleukin” followed by a high number is important in immunology! Before you look away, this is no ordinary “new” cytokine system; in fact, the 4 proteins of the IL-36 system (sometimes collectively nicknamed the “IL-1Fs”) have spent the past decade grossly underperforming in vitro despite looking frustratingly like their vigorous cousin IL-1.¹ 

In this issue of Blood, Vigne and colleagues are the first to obtain results in a biologically relevant system with plausible concentrations of IL-36.²  They show that the IL-36 system interacts powerfully with both dendritic and T cells and is likely to be fundamental to the regulation of the immune response in the mammalian skin.

The 3 agonistic cytokines IL-36α, IL-36β, and IL-36γ are products of separate genes that (in humans) sit inside the 400 000 bp of the IL-1 gene cluster.³  A fourth gene in the IL-36 group encodes a competitive antagonist for the IL-36 receptor, IL-36 receptor antagonist (IL-36Ra), which is approximately 50% identical to IL-1Ra (which has the same function on the IL-1 receptor). The genes encoding the receptors for IL-36 and IL-1 are next-door neighbors too, both physically and in evolutionary space. When bound to an agonist ligand, the IL-36 receptor associates with its signaling partner, IL-1 receptor accessory protein (IL-1RAcP), which is exactly the same molecule that associates with the IL-1 receptor as illustrated above. The association is required to recruit the signaling complex^4,5  (see figure). Like IL-1, IL-36 tends to generate an unequivocal and vigorous response in any of the rare cell lines that have constitutive receptors for it, without requiring the influence of other cytokines.^4,5  It may seem startling, therefore, that the entire set of IL-36 cytokines was reported so long ago yet results like those of Vigne et al have been so slow to arrive.

Most cells have constitutive receptors for IL-1 but not IL-36. When added to almost any cell in tissue culture at concentrations of 100 picomolar, the response to IL-1 is virtually saturated and inflammatory cytokines, chemokines, and small molecule mediators are secreted abundantly. Distressingly, the recombinant full-length protein products of all of the IL-36 system have gram-for-gram efficacies, compared with IL-1, of around 1/100 000,^4,5  and it has been very difficult to countenance (or obtain funding to research) a cytokine that needs to be added in the micromolar concentration range! A piece was missing from the puzzle.

IL-1 enthusiasts are not startled by very specialized pathways of proteolytic processing that can lead to activation because they are seen in IL-1β and IL-18 and maybe others. It was not obvious, though, how IL-36 could be processed. All 4 IL-36 group members are more or less the same size as fully processed IL-1α and IL-1β (and the unprocessed intracellular form of IL-1Ra) and the extra portions of the N- and C-terminus of the full-length IL-36 proteins do not easily suggest a consensus cleavage site. The logjam was broken when Towne and colleagues showed that all 3 IL-36 agonists and IL-36Ra need to be processed at one particular position (with apparently zero tolerance), close to the N-terminus of the precursor protein.⁶  Once supplied to cells that have IL-36 receptors in this “polished” form, the vast functional discrepancy disappears between IL-36 and IL-1. What we still do not know is which proteases in nature can “polish” IL-36. Knowing this will probably reveal much about the biologic function of IL-36.

In their paper, Vigne et al argue that responses to IL-1–like cytokines have more to do with the cell types that express them than to intrinsic differences surrounding the receptors² ; a very sensible position given that the receptors themselves are so similar and that IL-1RacP is reused for IL-1, IL-33, and IL-36. They show that both dendritic cells and CD4⁺ T lymphocytes from mice have IL-36 receptors and that both cell types respond to “polished” IL-36. Dendritic cells responded more strongly to IL-36 than to IL-1 and the response was a barrage of cytokines and the up-regulation of cell-activation and antigen-presentation markers, suggesting that IL-36 should be better than IL-1 at promoting antigen presentation. In other experiments, Vigne and colleagues showed that IL-36 functions as an adjuvant in vivo.

There are differences between the response to IL-36 and IL-1 in CD4⁺ T cells too. It is not IL-1–like that IL-36 strongly induces interferon-γ and induces IL-4 significantly. These are activities that would be expected from IL-18 and IL-33, respectively. In isolation too, in these authors' hands, IL-1 induced IL-17 production strongly while IL-36 did not. Vigne and colleagues come to the conclusion that stimulation by IL-36 might favor the differentiation of the TH1 subset of T-helper cells rather than the TH17 subset, which are heavily implicated in autoimmune diseases. IL-1, by contrast, is essential for the TH17 response in vivo and drives the production of substantial amounts of the highly pro-inflammatory cytokine IL-17 from cultured spleen T cells.⁷  However, the story might not end this simply because strong IL-17 production in this type of assay in response to IL-1 usually requires IL-23, and Vigne et al have not yet reported the effect of co-stimulation of T cells with IL-36 and IL-23.

Complementing the work in the current paper, there is now strong genetic evidence that proper regulation of IL-36 in the skin is vital in humans.⁸  Individuals who have no functional IL-36Ra gene develop life-threatening episodic pustular inflammation; a clear case of immune dysregulation in the skin as well as more generalized episodic fever.

Now that we have the reagents, there will be plenty to investigate about the IL-36 system. We can now do so in the faith that it is actually functional!