Interleukin-33 Is Biologically Active Independently of Caspase-1 Cleavage

The new interleukin (IL)-1 family cytokine IL-33 is synthesized as a 30-kDa precursor. Like pro-IL-1β, human pro-IL-33 was reported to be cleaved by caspase-1 to generate an 18-kDa fragment, which is sufficient to activate signaling by the IL-33 receptor T1/ST2. However, the proposed caspase-1 cleavage site is poorly conserved between species. In addition, it is not clear whether caspase-1 cleavage of pro-IL-33 occurs in vivo and whether, as for IL-1β, this cleavage is a prerequisite for IL-33 secretion and bioactivity. In this study, we further investigated caspase-1 cleavage of mouse and human pro-IL-33 and assessed the potential bioactivity of the IL-33 precursor. We observed the generation of a 20-kDa IL-33 fragment in cell lysates, which was enhanced by incubation with caspase-1. However, in vitro assays of mouse and human pro-IL-33 indicated that IL-33 is not a direct substrate for this enzyme. Consistently, caspase-1 activation in THP-1 cells induced cleavage of pro-IL-1β but not of pro-IL-33, and activated THP-1 cells released full-length pro-IL-33 into culture supernatants. Finally, addition of full-length pro-IL-33 induced T1/ST2-dependent IL-6 secretion in mast cells. However, we observed in situ processing of pro-IL-33 in mast cell cultures, and it remains to be determined whether full-length pro-IL-33 itself indeed represents the bioactive species. In conclusion, our data indicate that pro-IL-33 is not a direct substrate for caspase-1. In addition, our results clearly show that caspase-1 cleavage is not required for pro-IL-33 secretion and bioactivity, highlighting major differences between IL-1β and IL-33.Interleukin (IL)² -33, the most recently described cytokine of the IL-1 family, is synthesized as a 30-kDa precursor. Human pro-IL-33, like pro-IL-1β, was reported to be cleaved by caspase-1 in vitro to generate an 18-kDa fragment, termed mature IL-33, which is sufficient to activate signaling by the IL-33 receptor T1/ST2 (1).Caspase-1 is an endoproteinase that specifically cleaves Asp-Xaa bonds, where Xaa typically refers to a small, often hydrophobic residue (2–4). Caspase-1 activity absolutely requires the presence of an Asp residue at position −1 of the cleavage site. Consistently, replacement of Asp¹¹⁶ by other amino acids, such as Ala, was previously demonstrated to prevent caspase-1 cleavage of pro-IL-1β (2). Recombinant (r) mature IL-33 starts at Ser¹¹² for human (h) IL-33 and at Ser¹⁰⁹ for mouse (m) IL-33, neither of which corresponds exactly to the position of a potential caspase-1 cleavage site. Indeed, the N-terminal moiety of human pro-IL-33 sequence contains a single Asp at position 110, and the N-terminal portion of mouse pro-IL-33 contains an Asp at positions 88 and 106. In fact, the region located between amino acids 80 and 110 of pro-IL-33 is rather poorly conserved between species (5). In particular, no Asp residues can be consistently found at an identical position across species to hint at the presence of a conserved caspase-1 cleavage site. So far, caspase-1 cleavage of pro-IL-33 has not been investigated in any species other than human.Expression of endogenous IL-33 has been described most extensively in endothelial cells, where essentially nuclear, full-length 30-kDa pro-IL-33 is detected (5–7). To date, only two studies have examined potential effects of caspase-1 activation on the processing and secretion of pro-IL-33 in living cells. In one study, stimulation of murine glial cultures with caspase-1 activators induced secretion of bioactive IL-33 into culture supernatants, but the size of the secreted protein was not assessed (8). It is thus not clear whether caspase-1 cleavage of pro-IL-33 occurs in mouse cells. In a second study, Western blot analysis revealed the presence of a 32-kDa protein and minor 17 and 20 kDa bands reacting with anti-IL-33 antibodies in the supernatants of THP-1 cells upon caspase-1 activation, suggesting secretion of full-length pro-IL-33 and of two potential cleavage products (9). Although this last observation suggests that some pro-IL-33 may be secreted, it not known to what extent IL-33 secretion is dependent on caspase-1 cleavage. Finally, so far all studies reporting T1/ST2-mediated effects of IL-33 were performed using the recombinant mature form of IL-33, whereas potential biological activity of the full-length precursor form has not been tested. It thus remains to be shown whether, as for IL-1β, caspase-1 cleavage is indeed required for IL-33 bioactivity. In the present study, we thus further investigated caspase-1 cleavage of mouse and human pro-IL-33 in vitro and in cultured cells and assessed the potential bioactivity of the IL-33 precursor.