Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 7-Day Trial for You or Your Team.

Learn More →

Intracellular Interleukin-1 Receptor 2 Binding Prevents Cleavage and Activity of Interleukin-1α, Controlling Necrosis-Induced Sterile Inflammation

Intracellular Interleukin-1 Receptor 2 Binding Prevents Cleavage and Activity of Interleukin-1α,... Immunity Article Intracellular Interleukin-1 Receptor 2 Binding Prevents Cleavage and Activity of Interleukin-1a, Controlling Necrosis-Induced Sterile Inflammation 1,3 1,3 2 1 1, Yue Zheng, Melanie Humphry, Janet J. Maguire, Martin R. Bennett, and Murray C.H. Clarke * Division of Cardiovascular Medicine Clinical Pharmacology Unit Department of Medicine, University of Cambridge, Addenbrooke’s Hospital, Cambridge CB2 0QQ, UK These authors contributed equally to this work *Correspondence: [email protected] http://dx.doi.org/10.1016/j.immuni.2013.01.008 Open access under CC BY license. SUMMARY (Clarke et al., 2010; Kamari et al., 2007), graft rejection (Rao et al., 2007, 2008), toxic liver insults (Chen et al., 2007), and ischemia- Necrosis can induce profound inflammation or be reperfusion injury (Cohen et al., 2010; Luheshi et al., 2011). clinically silent. However, the mechanisms underlying The prototypic IL-1 family is ancient, with homologs identified back to echinoderms (Beck and Habicht, 1986). IL-1a, one of the such tissue specificity are unknown. Interleukin-1a principal ligands, is expressed by most lineages as a signal (IL-1a) is a key danger signal released upon necrosis peptide-less protein, is not readily secreted (Dinarello, 2009), that exerts effects on both innate and adaptive and is actively retained during apoptosis (Cohen et al., 2010). immunity and is considered to be constitutively Once released into the extracellular milieu, IL-1a ligation of the active. In contrast, we have shown that necrosis- type 1 IL-1 receptor (IL-1R1) leads to multiple proinflammatory induced IL-1a activity is tightly controlled in a cell effects (Dinarello, 2009), including cytokine secretion, neutrophil type-specific manner. Most cell types examined ex- recruitment, and upregulation of major histocompatibility pressed a cytosolic IL-1 receptor 2 (IL-1R2) whose complex (MHC) and costimulatory molecules on antigen- binding to pro-IL-1a inhibited its cytokine activity. In presenting cells. IL-1a also has powerful effects on adaptive cell types exhibiting a silent necrotic phenotype, immunity by enhancing expansion and survival of T cells, differ- entiation of T helper 17 (Th17) cells, and effector T cell prolifera- IL-1R2 remained associated with pro-IL-1a. Cell tion in the presence of regulatory T cells (Sims and Smith, 2010). types possessing inflammatory necrotic phenotypes These potent effects mean that activity is tightly controlled at either lacked IL-1R2 or had activated caspase-1 multiple levels. Mice deficient in IL-1a or IL-1b exhibit no before necrosis, which degraded and dissociated phenotype. However, mice lacking the IL-1 receptor antagonist IL-1R2 from pro-IL-1a. Full IL-1a activity required (IL-1RA) have small litters and retarded growth and develop cleavage by calpain after necrosis, which increased spontaneous arthritis-like polyarthropathy, arteritis, and cancer its affinity for IL-1 receptor 1. Thus, we report a cell (Dinarello, 2009). Indeed, increased IL-1 activity is a hallmark type-dependent process that fundamentally governs of many chronic inflammatory conditions, including rheumatoid IL-1a activity postnecrosis and the mechanism arthritis, gout, diabetes, atherosclerosis, and psoriasis (Dinar- allowing conditional release of this blockade. ¨ ello, 1996, 2009; Duewell et al., 2010; Rajamaki et al., 2010). IL-1 family members are synthesized as inactive precursors INTRODUCTION unable to bind their receptor, providing an initial level of control. IL-1b and IL-18 are activated by caspase-1, which requires inflam- masome formation. In contrast, IL-33 processing by caspase-3 Understanding why the immune system responds to necrosis and how this is controlled is critical in unraveling multiple human or caspase-1 results in inactivation (Cayrol and Girard, 2009; Lu¨ thi diseases. The ‘‘danger’’ model proposes that immunity responds et al., 2009). Pro-IL-1a (p33) is processed to mature IL-1a (p17) by to nonphysiological cell death, damage, or stress (Matzinger, calpain (Kobayashi et al., 1990), but the biological consequences 1994). Accordingly, necrotic death releases damage-associated of cleavage are unknown given that p33 is reported to be fully molecular patterns (DAMPs), which are sensed as danger and active. This finding is credited to two papers, but one only act as universal signals to activate immunity (Chen and Nun˜ ez, discusses p33 activity (March et al., 1985), whereas activity within 2010; Kono and Rock, 2008; Rock et al., 2010). DAMPs are the second study may be compromised by p33 degradation retained in healthy cells and during apoptosis (Basu et al., 2000; (Mosley et al., 1987). Interestingly, calpain is activated upon loss Cohen et al., 2010; Scaffidi etal., 2002), whereas necrosis releases of plasma membrane integrity (Wang, 2000), suggesting that them into the extracellular milieu. Interleukin-1a (IL-1a)isan calpain cleavage of IL-1a could be a control point for activity post- important DAMP that activates immunity postnecrosis (Chen necrosis. Although a recent study reports increased IL-1a activity et al., 2007; Clarke et al., 2010; Cohen et al., 2010; Eigenbrod after granzyme B cleavage (Afonina et al., 2011), differential et al., 2008; Kono et al., 2010; Rao et al., 2007), driving pathologies efficacy of p33 and p17 IL-1a is still controversial (Gross et al., as diverse as tumorigenesis (Sakurai et al., 2008), atherosclerosis 2012), and no mechanism to explain this has been reported. Immunity 38, 285–295, February 21, 2013 ª2013 Elsevier Inc. 285 Immunity IL-1R2 Controls IL-1a Activity Postnecrosis Figure 1. Necrosis-Induced Sterile Inflam- mation Is Cell Type Specific (A) IL-6 concentrations in conditioned media of macrophages (Macs), Jurkat cells, or VSMCs incubated with lysates from their respective necrotic cells or with IL-1a. (B and C) Immunoblots of IL-1a content and pro- cessing in whole cell (WC) or necrotic lysates (NL) (B) or in necrotic cell lysates pretreated with protease inhibitors as indicated (C). (D and E) IL-6 and MCP-1 concentrations in conditioned media of VSMCs incubated with necrotic VSMC lysates made in the presence of D protease inhibitors (D), or with IL-1a alone, or with calpeptin or EGTA (E). Data represent mean ± SD; *p % 0.007 versus control, n = 3; **p % 0.03 (MCP-1), p % 0.002 (IL- 6), n R 4. NS, not significant. See also Figure S1. inflammation, we analyzed IL-6 release from VSMCs, Jurkat cells, and primary macrophages treated with lysates from E these cell types undergoing necrosis. Only necrotic VSMCs induced significant IL-6 release from viable VSMCs, previ- ously shown to be IL-1a dependent (Clarke et al., 2010), and only VSMCs re- sponded to IL-1a (Figure 1A). Little IL-1a activity was found in membrane fractions of necrotic macrophages or Jurkat cells (Figure S1A available online) or in lysates from cells undergoing hypoxia-induced necrosis (Figure S1B). Although compa- rable amounts of IL-1a were found in all three cell types, only VSMCs processed We report that necrosis-induced IL-1a-dependent responses p33 to p17 IL-1a (Figure 1B). Calpain cleaves p33 to p17 are highly cell type dependent and correlate with calpain (Kobayashi et al., 1990); however, the relevance of cleavage cleavage of IL-1a during necrosis. Contrary to current under- to IL-1a activity is unknown. Several reports indicate that pro- standing, p33 requires calpain processing for full biological cessing is not required for secretion (Brough et al., 2003; activity, which increases its affinity for IL-1R1. Cell type depen- Prudovsky et al., 2003), and cytokine activity is supposedly inde- dency occurs due to expression of an intracellular form of IL- pendent of cleavage (March et al., 1985; Mosley et al., 1987). 1R2 that binds IL-1a, preventing calpain cleavage and cytokine VSMCs rapidly cleaved p33 to p17 during necrosis, which activity. After inflammasome activation, caspase-1 specifically was prevented by the calpain inhibitors calpeptin and EGTA but not the proteasomal inhibitor lactacystin (Figure 1C). cleaves IL-1R2, which abrogates IL-1a binding, allows calpain Necrotic VSMC lysates prepared with calpain inhibition cleavage, and completely restores IL-1a-dependent responses. Regulated secretion of IL-1a also requires IL-1R2 cleavage. produced significantly reduced responses (Figure 1D), whereas Thus, we report an important cell type-dependent mechanism calpeptin or EGTA treatment did not inhibit cytokine production that fundamentally governs IL-1a activity postnecrosis and the in response to IL-1a (Figure 1E). However, prolonged incubation mechanism allowing conditional release of this blockade. at 37 C in the presence of calpeptin resulted in some processing of p33 to p17 (Figure S1C), which could contribute to the RESULTS apparent IL-1a activity seen with calpeptin-treated VSMCs (Figure 1D). Nevertheless, taken together this suggests that, Necrosis-Induced Sterile Inflammation Is Cell Type contrary to the accepted literature, p33 is not fully active until Specific and Correlates with Calpain Cleavage cleaved. of pro-IL-1a We have recently shown that IL-1a released during vascular Calpain-Cleaved p33 and Recombinant p17 Are More smooth muscle cell (VSMC) necrosis is a powerful DAMP that Active than p33 IL-1a induces local vessel inflammation (Clarke et al., 2010). To deter- To directly compare p17 and p33 activity, we purified HIS- mine which other cell types could drive IL-1a-dependent sterile tagged recombinant proteins. Although we could express and 286 Immunity 38, 285–295, February 21, 2013 ª2013 Elsevier Inc. Immunity IL-1R2 Controls IL-1a Activity Postnecrosis Figure 2. Calpain-Cleaved p33 and Recom- AB binant p17 Is More Active than p33 IL-1a (A) Cytokine concentrations in conditioned media of VSMCs treated with p33 or calpain-cleaved p33, ± IL-1a neutralizing antibody (a pAb). (B) VSMCs were also treated with calpain sham reactions (no p33) or commercial recombinant p17 IL-1a (cr17) ± calpain sham. (C) IL-2 concentration in conditioned media of murine EL4 cells treated with p33 ± calpain. (D) Immunoblot of in vitro cleavage of p33 by calpain. (E) Coomassie stain of purified soluble p17- and p33-GST fusion proteins. (F and G) IL-2 (F) or IL-6 (G) concentrations in DE conditioned media of EL4 or VSMCs, respectively, treated with 1 nM p17- or p33-GST ± a pAb. (H) IL-2 concentrations in conditioned media of EL4 cells incubated with p33-GST ± calpain, and ± a pAb. (I) GR1 cells recruited intraperitoneally in wild- type or Il1r1 mice injected with saline or 29 fmol/g p17 or p33. Data represent mean ± SD or mean ± SEM (I); *p % 0.0003, n = 4 (A), n = 3 (B, C) protein prepa- rations and cleavage reactions; **p% 0.007, n = 4. NS, not significant. F G indicating that the increased activity is due to processing of p33 to p17. These data were also reproduced cross-species with the classic IL-1-sensitive murine cell line EL4 (Figure 2C), suggesting a conserved requirement for calpain processing of IL-1a. Although calpain HI cleavage of IL-1a increased activity, these differences represent an underesti- mate because of the nonstoichiometric conversion of p33 to p17 during in vitro cleavage (Figure 2D). By fusing IL-1a to GST, we purified small amounts of p17 and p33 as soluble proteins (145 mg/l) (Figure 2E). Compar- ison at equal molarities demonstrated significant increases in IL-2 release from EL4 cells (Figure 2F) and IL-6 release from VSMCs for p17 compared to p33 (Figure 2G). IL-1a neutralization reduced purify p17 in E. coli, p33 was very insoluble and prone to responses to control indicating IL-1a-dependent stimulation, aggregation (not shown)—a finding reported by others excluding effects from copurified bacterial PAMPs (Figures 2F (Tokunaga et al., 2010). After denaturation in urea, p33 could and 2G). To eliminate whether GST-p33 was ‘‘functionally be purified and renatured to a soluble protein after sequential dead’’ as a result of a purification artifact or intrinsic instability, dialysis, but this made comparison between p17 and p33 we cleaved it with calpain or incubated it at 37 C for 16 hr, impossible because of the inability to control for refolding effi- respectively. Cleavage of GST-p33 significantly restored ciency between different proteins. To circumvent this, we activity (Figure 2H), whereas incubation did not result in protein cleaved p33 in vitro with calpain, which increased cytokine degradation (not shown), again supporting an inherent lower release compared to uncleaved p33 (Figure 2A). Importantly, activity of p33. Lastly, injection of p17 into the peritoneum a neutralizing antibody to IL-1a reduced responses to control resulted in significantly increased neutrophil recruitment (Figure 2A), whereas a ‘‘calpain sham’’ reaction without p33 compared to injection of p33 in wild-type mice, a result that neither promoted nor inhibited cytokine release (Figure 2B), was not seen in Il1r1 mice (Figure 2I). Immunity 38, 285–295, February 21, 2013 ª2013 Elsevier Inc. 287 Immunity IL-1R2 Controls IL-1a Activity Postnecrosis Figure 3. p33 IL-1a Shows Minimal Activity A B at Physiological Concentrations because of Lower Receptor Affinity (A and B) Concentration-response curves to p17 and p33, minus control, in VSMCs (50 pM to 8 nM) (A) or EL4 cells (5 pM to 8 nM) (B). (C and D) IL-2 concentrations in conditioned media from EL4 cells treated with p17 or p33 at 4 nM (C), and 0.1 nM p17 or 4 nM p33 (D), with increasing concentrations of IL-1RA. Data represent mean ± SD; *p % 0.03, n = 3 (A), n R 4 (B); for difference in IC p < 0.0001 (C), C D 50 p = 0.52 (D), n = 3. See also Figure S2. IL-1RA completely inhibited cytokine release and gave an IC of 10.3 ± 2.9 ng/ml (Figure 3D). Thus, a 40-fold difference in concentration gave near superimposable curves (p = 0.52), corre- lating well with the value above. Together these data show that, in contrast to previous reports, IL-1a is not fully active p33 IL-1a Shows Minimal Activity and Behaves as until processed by calpain and that cell types unable to cleave a Partial Agonist due to a Lower Receptor Affinity IL-1a induce much smaller IL-1a-dependent responses after To understand why p33 is less active than p17, we generated necrosis. cytokine concentration-response curves. VSMCs responded to p17 at 50 pM and saturated at 2 nM (Figure 3A). In contrast, An Intracellular Form of IL-1R2 Protects p33 IL-1a p33 generated responses only at supraphysiological amounts from Calpain Processing and failed to saturate (Figure 3A). At physiological concentra- We hypothesized that cell type-specific cleavage of IL-1a could tions of 1 nM, p17 was 50 times more active than p33. EL4 cells occur through modification of enzyme or substrate, excess of are very sensitive to IL-1a, with responses detectable with fM a calpain inhibitor, or a binding partner that protects IL-1a from amounts. p17 activity was detectable at 5 pM and rapidly satu- processing. Jurkat and macrophage necrotic lysates incubated rated at 0.5 nM (Figure 3B). However, p33 gave a linear concen- at 37 C for extended periods still showed no processing of tration response, such that 65% of the maximal response was endogenous p33 (Figure 4A), whereas recombinant His-p33 obtained at 8 nM but did not reach saturation (Figure 3B). At spiked into Jurkat lysates also remained uncleaved (Figure 4A), 1 nM p17 was 10 times more active than p33. To exclude steric excluding a modification to the cell-derived cytokine. Similar hindrance of fused GST only on p33, we specifically cleaved off results were also found with primary T lymphocytes (Fig- GST. Comparison of p33 with or without GST revealed identical ure S3A). An excess of calpain inhibitor was excluded, because efficacies at a range of concentrations (Figure S2). Together, this high calpain activity was found in all necrotic cell types (Fig- suggests that p33 can bind IL-1R1 at higher concentrations and ure S3B). Addition of purified calpain to Jurkat necrotic lysates thus acts as a partial agonist. also failed to cleave p33 (Figure 4B), whereas analysis of spec- To investigate whether p33 binds IL-1R1 but fails to induce trin, an alternative calpain substrate, revealed calpain-depen- downstream signaling, we conducted competition experiments dent processing in both Jurkat and VSMCs necrotic lysates with a fixed concentration of p17 (1 nM) and increasing concen- (Figure 4C). trations of p33 (up to 8 nM). However, even 8 nM p33 failed to Protection of recombinant p33 by Jurkat necrotic lysates inhibit cytokine responses (data not shown), suggesting suggested that an excess of a binding partner must be present, a stronger affinity of p17 for the receptor. IL-1RA binds IL-1R1 and therefore that it could be transferred. Indeed, necrotic but does not induce signaling (Dinarello, 2009). Therefore the lysates made from mixed Jurkat and VSMCs displayed no difference in concentration of IL-1RA required to inhibit 50% p33 cleavage (Figure 4D). Known binding partners of IL-1a of the cytokine response induced by p17 or p33 (the IC ) include the type 1 and 2 IL-1 receptors; therefore we treated directly relates to the difference in affinity of the two ligands Jurkat necrotic lysates with a large excess of IL-1RA, which for IL-1R1. Responses to 4 nM p33 were completely inhibited resulted in calpain-specific cleavage of p33 (Figure S3C). The by 100 ng/ml IL-1RA (IC of 13.7 ± 3.8 ng/ml) (Figure 3C). In only known receptors for IL-1RA are IL-1R1 and IL-1R2, sug- contrast, 1,600 ng/ml IL-1RA failed to inhibit responses to gesting that IL-1a may be bound to an intracellular IL-1R. 4 nM p17, but 50% inhibition occurred at 633.3 ± 16.7 ng/ml (Fig- RT-PCR revealed that VSMCs express only IL-1R1, whereas ure 3C). This indicates that p33 has an affinity for IL-1R1 46.3 Jurkat cells (Figure 4E) and macrophages (data not shown) times lower than does p17 (p < 0.0001). To exclude p17 express IL-1R2. Immunoblots confirmed IL-1R2 expression in responses from another receptor insensitive to IL-1RA blockade, Jurkat cells (Figure 4F), macrophages, and T cells (Figure S3D), we repeated with 0.1 nM p17. Under these conditions 100 ng/ml whereas addition of recombinant IL-1R2 to a cell-free cleavage 288 Immunity 38, 285–295, February 21, 2013 ª2013 Elsevier Inc. Immunity IL-1R2 Controls IL-1a Activity Postnecrosis AB GH IJ Figure 4. An Intracellular Form of IL-1R2 Protects IL-1a from Calpain Processing (A and B) Immunoblot for endogenous IL-1a processing in necrotic lysates of primary macrophages (Macs) and Jurkat cells (A), for exogenous His-p33 in necrotic Jurkat lysates (A), or for Jurkat necrotic lysates treated with increasing amounts of purified calpain (B). (C and D) Immunoblot of whole cell (WC) and necrotic (NL) Jurkat and VSMC lysates for a-spectrin (C) or IL-1a after mixing of lysates from both cell types (D). (E and F) RT-PCR (E) or immunoblot (F) for IL-1 receptors in VSMCs and Jurkat cells. (G and H) Immunoblot for IL-1a cleavage in a cell-free system with purified calpain (G) and in necrotic VSMC lysates (H), ± 250 ng IL-1R2. (I and J) Coimmunoprecipitation of IL-1R2 with p33-GST (I) and p33 with IL-1R2-HIS (J) in transfected HEK cell lysates. See also Figure S3. reaction (Figure 4G) or necrotic VSMC lysates (Figure 4H) pre- HIS (Figures 4I and 4J), whereas a proximity ligation assay vented calpain-dependent p33 processing in a dose-depen- demonstrated association of endogenous proteins in situ (Fig- dent manner (Figure S3E). IL-1R2 also antagonized p33 in ure S3G), supporting a direct interaction. Finally, IL-1R2 also a dose-dependent manner (Figure S3F). Importantly, IL-1R2 protected IL-1a from cleavage by the inflammatory proteases coimmunoprecipitated with p33-GST and p33 with IL-1R2- granzyme B, chymase, and elastase (Figure S3H). Immunity 38, 285–295, February 21, 2013 ª2013 Elsevier Inc. 289 Immunity IL-1R2 Controls IL-1a Activity Postnecrosis Figure 5. Silencing of IL-1R2 Enables IL-1a Processing and Restores Necrotic Cell- Induced Inflammation (A–C) siRNA-mediated silencing reduces IL-1R2 mRNA (A) and protein in whole cell (WC) and necrotic lysates (NL) (B), which leads to calpain- dependent IL-1a cleavage upon necrosis (C) that is inhibited with calpeptin (+C). (D and E) IL-2 concentrations in conditioned media of EL4 cells incubated with necrotic lysates from C HeLa cells treated with siRNA to IL-1R2 or scrambled, either alone (D) or with IL-1b treatment (E), ± a pAb. Data represent mean ± SD; *p = 0.0007, n = 3 independent silencings and treatments. NS, not significant. (F) GR1 cells recruited intraperitoneally in wild- type or Il1r1 mice injected with saline or 8.3 3 10 control or IL-1R2-silenced necrotic HeLa cells. Data represent mean ± SEM; **p % 0.006, n R 4. NS, not significant. (G) Immunoblot for IL-1R2 and IL-1a in necrotic lysates of empty vector or soluble IL-1R2-trans- fected VSMCs. See also Figure S4. receptor that limits bioavailability of IL-1 (Colotta et al., 1993), we examined whether IL-1R2 silencing had reversed ‘‘general’’ IL-1 antagonism within necrotic lysates. By neutralizing all IL-1a activity within necrotic lysates, we com- pared responsiveness to IL-1b with or without scrambled or IL-1R2-silenced necrotic lysates present (Figure 5E). This revealed no significant difference in IL-2 production, implying that IL-1R2 silencing hadn’t simply reduced general IL-1 antagonism. Sterile peritonitis induced with scrambled control necrotic lysates recruited equal numbers of neutrophils in wild-type and Il1r1 mice, therefore representing IL-1-inde- pendent responses (Figure 5F). However, IL-1R2-silenced necrotic lysates re- cruited 3-fold more cells, which was IL-1 dependent because of loss of response in Il1r1 mice (Figure 5F). Because IL-1R2 Silencing Enables Calpain Cleavage of p33 IL-1a previously described forms of IL-1R2 retain a signal peptide and Restores Inflammatory Response to Necrotic Cells (Liu et al., 1996), they would be expected to be in the exocytic Low transfection efficiency and suspension growth made Jurkat pathway. Therefore we determined whether cell surface-shed cells unsuitable for siRNA silencing. Therefore, necrotic HeLa IL-1R2 (Orlando et al., 1997) protects IL-1a by inhibiting metallo- cells were tested for p33 cleavage (Figure S4A), calpain activity proteases with BB-94; p33, however, remained uncleaved (Fig- (data not shown), and IL-1R2 expression (Figures 5A and 5B), ure S4B). Furthermore, immunofluorescence revealed a large which indicated that they responded similarly. siRNA to IL-1R2 amount of IL-1R2 to be intracellular (Figure S4C) in multiple cell reduced both IL-1R2 mRNA (Figure 5A) and protein (Figure 5B), types (Figure S4D) and endogenous p33 and IL-1R2 to be highly which was not seen with scrambled control. After IL-1R2 colocalized (Figure S4E) throughout the cytoplasm (Figure S4F). silencing, necrotic HeLa lysates displayed calpain-dependent In addition, subcellular fractionation (Figure S4G) and protease K processing of p33 to p17 (Figure 5C) and restoration of IL- protection assays (Figure S4H) revealed a native pool of IL-1R2 1a-dependent responses (Figure 5D), whereas scrambled in the cytosol without internal membrane disruption, as evi- control cells did not. Because IL-1R2 functions as a decoy denced by a lack of calreticulin in cytosolic fractions. Finally, 290 Immunity 38, 285–295, February 21, 2013 ª2013 Elsevier Inc. Immunity IL-1R2 Controls IL-1a Activity Postnecrosis Figure 6. Caspase-1 Cleavage of IL-1R2 A B Restores Necrotic Cell-Induced IL-1a- Dependent Inflammation (A) Immunoblot for IL-1a in necrotic lysates from LPS- and ATP-treated macrophages ± calpeptin (during lysis) or ± caspase inhibitor Z-VAD (during LPS and ATP treatment). (B) IL-2 concentrations in conditioned media of EL4 cells incubated with necrotic lysates from control or LPS- and ATP-treated macrophages ± a and/or b pAb, or ± Z-VAD (during LPS and ATP treatment). Data represent mean ± SEM; *p < 0.00004, **p % 0.02, n R 3. (C and D) Immunoblot for IL-1R2 cleavage after incubation with active caspase-1 ± Z-YVAD (C) or a panel of active caspases (D). (E) Edman degradation of cleaved IL-1R2 detected two sequences (italic underlined) corresponding to processing at two tetrapeptide sites (large bold). (F) Immunoblot for cleaved IL-1R2 after mutation of both caspase-1 sites. EF (G) IL-2 concentrations in conditioned media of EL4 cells incubated with IL-1a, and IL-1R2 ± caspase-1 cleavage. Data represent mean ± SD; *p = 0.009, n = 3. (H) IL-1a concentration in conditioned media of GH Wild-type activated THP-1 cells transfected with caspase site mutant IL-1R2, or as indicated. Data repre- sentative of mean ± SD from n = 2. See also Figure S5. IL-1a- and IL-1b-neutralizing antibodies (Figure 6B). IL-1R2 contains many pre- dicted caspase sites (Figure S5A) and cell-free cleavage resulted in caspase-1- Wild-type specific processing (Figure 6C). IL-1R2 could be cleaved by inflammatory caspase-1 and caspase-5, but not by caspase-4 or the apoptotic caspase-3 expression of exogenous soluble IL-1R2 in VSMCs prevented (Figure 6D). To confirm the cleavage site, we mutated the P1 p33 cleavage (Figure 5G). Together these results demonstrate position Asp to Ala, resulting in a nonconsensus sequence for that a cytosolic complex of p33 and IL-1R2 exists in many cell caspases. Individual mutation of all sites that could give the types. During necrosis this prevents calpain cleavage of p33 to correct sized products did not prevent IL-1R2 processing (data the fully active p17 form, which reduces IL-1a-dependent not shown). However, Edman degradation of the C-terminal responses. IL-1R2 fragment sequenced two products at equal abundance corresponding to cleavage at two separate Asp residues 12 Caspase-1 Specifically Cleaves IL-1R2, which Restores amino acids apart (Figure 6E), and subsequent mutation of IL-1a-Dependent Inflammation Postnecrosis both these sites prevented cleavage (Figure 6F). The crystal Macrophages are suggested to be the sensors of necrotic- structure of IL-1R2 complexed to IL-1 (Wang et al., 2010) re- derived DAMPs, which then release IL-1a to drive sterile inflam- vealed that cleavage at these sites would remove the D1 mation (Chen et al., 2007; Kono et al., 2010). However, our domain vital for binding. Indeed, antagonistic activity toward current data indicate that macrophage IL-1a is bound to IL- IL-1a was lost upon processing (Figure 6G), whereas sham 1R2, and therefore is nonfunctional. Because recent work cleavage reactions neither promoted nor inhibited IL-2 release demonstrates that IL-1a secretion requires inflammasome acti- (Figure S5B), confirming that cleaved IL-1R2 cannot bind IL- vation (Fettelschoss et al., 2011), we investigated whether this 1a. Finally, caspase-1 cleavage of IL-1R2 appears to be a pre- could overcome IL-1R2 blockade. Necrotic lysates made from requisite for physiological IL-1a secretion, as indicated by the LPS and ATP-stimulated macrophages displayed processing fact that expression of the noncleavable IL-1R2 mutant reduced of p33, which could be prevented by the caspase inhibitor IL-1a release (Figure 6H). Therefore, after exposure to stimuli Z-VAD-fmk or calpeptin (Figure 6A). Furthermore, after LPS that activate inflammasomes, intracellular IL-1R2 is cleaved by and ATP treatment, macrophage necrotic lysates induced IL-2 caspase-1, causing it to dissociate from IL-1a. Subsequently, release from EL4 cells, which was blocked with Z-VAD-fmk or during necrosis calpain can now cleave IL-1a to the fully active Immunity 38, 285–295, February 21, 2013 ª2013 Elsevier Inc. 291 Immunity IL-1R2 Controls IL-1a Activity Postnecrosis form, which allows the generation of robust IL-1a-dependent in both wild-type and Il1r1 mice (representing the net effect of responses. all other DAMPS, independent of IL-1), necrotic cells without IL-1R2 induced a 3-fold greater response, which was all IL-1 DISCUSSION dependent. This key observation implies that without IL-1R2 expression, the single most powerful DAMP within necrotic Although an immune response to necrosis may resolve the orig- cells is IL-1a. inal pathology and initiate repair, recruited leukocytes can Our studies also identify multiple mechanisms that could give damage the surrounding tissue. This sterile inflammatory a tissue- and cell type-dependent response to necrosis. Given response can also lead to unwanted activation of adaptive that p33 has some activity, dependent on its concentration immunity. Many human diseases are driven by activation and sensitivity of the responding cells, extensive necrosis in an of these pathways, including ischemia-reperfusion injury, IL-1R1-rich tissue (e.g., liver) could activate immunity regardless Alzheimer’s disease, atherosclerosis, and toxic insults to liver of IL-1a cleavage. Conversely, minimal necrosis in an IL-1R1-low and lung (Chen and Nun˜ ez, 2010; Kono and Rock, 2008; Stewart tissue (e.g., kidney) that cannot cleave p33 may not respond et al., 2010). Therefore, tight regulation of both release and to IL-1a at all. If necrosis occurs with a stimulus that activates activity of DAMPs is critical for host defense. caspase-1 (e.g., infection), IL-1R2 blockade is abrogated and The current understanding is that p33 is fully active and there- cleaved fully active p17 can signal to any IL-1R1-expressing fore requires only cell lysis to signal after necrosis. In contrast, cell. Necrosis of cells that can cleave IL-1a within IL-1R1-dense we find that activity is kept under tight control at multiple levels tissues would give extensive immune activation (e.g., vascular tissue). Where the amount of IL-1R2, IL-1R1, and necrosis in a cell type-dependent and conditional manner. Blocking preclude IL-1a activity, the response to necrosis could be calpain processing of endogenous p33 during necrosis signifi- cantly decreases activity, and comparison of recombinant deferred to other DAMPs and specialist immune cells, perhaps proteins demonstrates up to 50-fold increased activity of p17 giving more control over reactions. Additionally, activation of compared to p33. This difference occurs at the receptor, with IL-1a during necrosis could act as a ‘‘tipping point’’ that drives a 50-fold greater affinity of p17 for IL-1R1. IL-1a cleavage an inflammatory response toward adaptive immunity. Indeed, and activity after necrosis is cell type dependent because of this is the case in graft-versus-host disease, where injured expression of a cytosolic IL-1R2, which binds and protects endothelial cell-derived IL-1a induces intimal T cell recruitment IL-1a from cleavage and prevents IL-1a activity. IL-1R2 also pro- and IL-17 production, driving human artery allograft rejection tected IL-1a from granzyme B, chymase, and elastase cleavage, (Rao et al., 2007, 2008). suggesting that the primary control over IL-1a activity postnec- Necrosis is induced by many stimuli including physical trauma, rosis is by IL-1R2 binding. Lastly, active caspase-1 specifically chemical stress, and bacterial toxins and is characterized by cleaves IL-1R2, causing dissociation from IL-1a, calpain pro- loss of plasma membrane integrity (Kono and Rock, 2008). 2+ cessing, and complete restoration of IL-1a activity after necrosis Calpains are activated by Ca entry after loss of membrane or during regulated secretion. integrity (Wang, 2000), and in VSMCs p33 cleavage occurs DAMPs need to be retained on apoptosis, released on immediately upon necrosis. Granzyme B is proposed to activate necrosis, and able to activate the immune system. IL-1a is IL-1a extracellularly (Afonina et al., 2011); but given its ineffi- near universally expressed (Dinarello, 2009), lacks a signal ciency to process IL-1a relative to calpain (Afonina et al., peptide, is rarely extracellular under normal physiology, and 2011), cleavage within the necrotic body by calpain pro- also associates with chromatin during apoptosis, resulting in bably predominates when IL-1R2 is absent. IL-1a and IL-1b retention (Cohen et al., 2010). Other identified DAMPS include arose from ancient gene duplications (Dinarello, 2009) but have heat shock proteins, uric acid, HMGB1, ATP, and dsDNA diverged molecularly. IL-1a is a DAMP that is activated by (Kono and Rock, 2008). These DAMPs utilize specific cognate a protease intrinsically linked to necrosis, whereas IL-1b is acti- receptors, such as RAGE, or pattern recognition receptors, vated in response to PAMPs by a protease family intrinsically such as TLRs, which have a limited expression pattern that linked to apoptosis and inflammation. However, these current data indicate that in many cells IL-1a function requires both restricts cell types able to respond. In contrast, IL-1R1 is widely calpain and caspase-1. Secretion of IL-1a from macrophages expressed throughout many lineages (Dinarello, 2009), enabling IL-1a to act as a ‘‘universal’’ DAMP. Although this may provide (Fettelschoss et al., 2011) or dendritic cells (DCs) (Gross et al., more efficient detection of necrosis, it could predispose the 2012) requires inflammasome activation, but given that IL-1a is immune system to overreact, perhaps explaining the extensive not a caspase substrate, this has been puzzling (Keller et al., control over IL-1a activity via release, IL-1R2 binding, and 2008). We suggest that IL-1R2 is the likely target of caspase-1 calpain cleavage. This cell type and conditional dependency during physiological IL-1a release. Indeed, expression of an for IL-1a activity after necrosis perhaps suggests an immune uncleavable IL-1R2 mutant reduces IL-1a release, whereas advantage in allowing only some cell types to utilize IL-1a as a Casp1 mice have long been known to be deficient in IL-1a DAMP. Previous studies concluded that IL-1a is only a secondary secretion (Kuida et al., 1995; Li et al., 1995). signaling molecule released by macrophages during sterile An important finding of this work is that IL-1R2 binds p33 in inflammation (Chen et al., 2007; Kono et al., 2010). However, the cytosol. Although the previously described splice variant of this finding might have occurred due to the use of necrotic cell IL-1R2 (Liu et al., 1996) loses the transmembrane domain, it still types that express IL-1R2 and therefore cannot cleave and utilize codes for a signal peptide and should be secreted. Despite this, IL-1a (Y.Z., M.H., and M.C.H.C.; data not shown). Indeed, our data show that a large pool of IL-R2 exists in the cytosol. One whereas control necrotic lysates induce an equivalent peritonitis explanation for this is that the IL-1R2 signal peptide is short 292 Immunity 38, 285–295, February 21, 2013 ª2013 Elsevier Inc. Immunity IL-1R2 Controls IL-1a Activity Postnecrosis overnight hypoxic exposure. To activate inflammasomes, cells were treated with (13 amino acids) and is predicted to be relatively weak (Signal P; LPS (1 mg/ml; 4 hr), followed by ATP (5 mM) or Nigericin (20 mM) for 30 min. Y-max = 0.316) compared to other secreted proteins (e.g., IL-8; Calpain activity was determined with Calpain-Glo (Promega). VSMCs, HEK, Y-max = 0.790), and many proteins with signal peptides can and THP-1 cells were transfected with pcDNA3 (Invitrogen) with nucleofection be found in the cytosol (Davis et al., 2006). Indeed, multiple (Amaxa) or FugeneHD (Promega). factors determine how much of a signal peptide-containing protein ends up nontargeted, including the signal peptide and Protein Expression and Purification flanking amino acid sequences and cell type and growth Human p33 (1-271) or p17 (119-271) was cloned into pET15b (Novagen) or condition effects (Levine et al., 2005). Clearly further work is pGEX-4T-3 (GE). Human IL-1R2 (1-296) was cloned into pGEX-4T-3. For His needed to elucidate the specific form of IL-1R2 that binds p33, purification, IPTG-induced cultures were lysed in BugBuster (Novagen) with the mechanisms that generate it, and how this is controlled. benzonase, lysozyme, and protease inhibitors (10 min, RT). Urea (6 M) was 2+ added and incubated (10 min) before clarification and purification on a Ni Cell type-specific effects of IL-1R2 expression may occur in column. Columns were washed (40 mM imidazole) and eluted (250 mM imid- a number of diseases. Induction of antitumor immunity in vivo azole). Concentrated protein (Vivaspin) was dialyzed against decreasing urea is dependent upon the mode of cell death, whereby necrosis (4 M, 2 M, 0 M) in 10 mM Tris (pH 8.0), 50 mM NaCl. For GST purification, is more immunogenic than apoptosis (Melcher et al., 1998) IPTG-induced cultures were lysed in 50 mM Na HPO (pH 7.5), 150 mM 2 4 and local IL-1 activity (Ghiringhelli et al., 2009; Michaud NaCl, 1 mM DTT, 1 mM EDTA, with benzonase, lysozyme, and protease inhib- et al., 2011). Necrosis can induce maturation of DCs and expres- itors (30 min, RT), clarified, applied to glutathione agarose (QIAGEN), washed, eluted (50 mM reduced glutathione), concentrated, and dialyzed against sion of costimulatory molecules and can stimulate T cells 10 mM Tris (pH 8.0), 50 mM NaCl. GST tag was removed with biotinylated (Basu et al., 2000; Sauter et al., 2000)—all processes critical thrombin (Novagen). Protein concentration was determined (660 nm; Thermo for antitumor immunity and all known activities of IL-1. Indeed, Scientific), checked by Coomassie staining, and if necessary adjusted and differences are reported between the necrotic cell type and rechecked. Proteins were stored in 10% glycerol (80 C). ability to induce DC maturation (Sauter et al., 2000), perhaps reflecting IL-1R2 expression and thus ability of necrotic cell- Protease Cleavage derived IL-1a to modulate antigen-presenting cell function. His- or GST-p33 was incubated in 10 mM Tris (pH 7.5), 150 mM NaCl, 1 mM Intriguingly, IL-1R2 is upregulated in some tumors including DTT, 2 mM CaCl , with calpain (Calbiochem) ± calpeptin or IL-1R2 (RT). pancreatic ductal adenocarcinoma (Ru¨ ckert et al., 2010) and IL-1R2 was incubated with active caspase-1, caspase-3, caspase-4, or ovarian cancer, where it provides a powerful distinction between caspase-5 (Promokine) in 50 mM HEPES (pH 7.2), 50 mM NaCl, 0.1% CHAPS, 10 mM EDTA, 5% glycerol, and 10 mM DTT (37 C). Jurkat lysates were primary and recurrent tumors (Laios et al., 2008). Finally, athero- also incubated with granzyme B (100 nM; Cambridge Bioscience), chymase, sclerosis is highly dependent on IL-1 (Chamberlain et al., 2009; or elastase (100 nM; both Enzo). Chi et al., 2004; Duewell et al., 2010), although recent work casts doubt on IL-1b’s role (Menu et al., 2011), suggesting that IL-1a Cytokine Release Assay could be the major ligand affecting atherosclerotic plaques. VSMCs were plated, adhered overnight, and incubated in SF DMEM (24 hr). Indeed, necrotic VSMC-released IL-1a is a potent inducer of Fresh SF DMEM and treatments were added and incubated (6 hr). EL4 cells local vessel inflammation in vivo (Clarke et al., 2010), and given were washed and plated in SF DMEM with treatments and incubated (24 hr). that VSMCs express high amounts of IL-1a and IL-1R1 but little 3 4 17.5 3 10 necrotic VSMCs or macrophages or 5.25 3 10 necrotic Jurkat IL-1R2, their death is highly inflammatory. or HeLa cells were used per 500 ml. Supernatants were clarified and cytokines In summary, in addition to acting as a decoy receptor, IL-1R2 assayed by ELISA (PeproTech) or Cytomix (eBioscience). also plays a hitherto unreported role that fundamentally controls IL-1a activity postnecrosis. Cells expressing intracellular IL-1R2 Immunoblotting, Edman Degradation, and Co-IP 5 5 release little active IL-1a after necrosis without prior activation of Whole cell or necrotic lysates from 1 3 10 VSMCs or macrophages, or 3 3 10 Jurkat, HeLa, or T cells were loaded per lane. Antibodies used were IL-1a caspase-1, which processes IL-1R2. In contrast, necrotic cells (PeproTech), His (GE), a-spectrin (Millipore), b-actin (Sigma), calreticulin (Cell devoid of IL-1R2 are inherently powerful inducers of sterile Signaling), HDAC1 (Santa Cruz), and IL-1R2 (R&D). Caspase-1-cleaved IL-1R2 inflammation able to fully activate IL-1a upon calpain cleavage. was separated, electroblotted to PVDF, and stained and bands were excised Thus, changes in intracellular IL-1R2 expression may underlie for Edman degradation (ABI Procise 494HT). For co-IP, transfected HEK cells and modulate many chronic inflammatory diseases or other were incubated on ice (50 mM Tris [pH 8], 150 mM NaCl, 1% Triton X-100, pathologies involving cellular necrosis. protease inhibitors), freeze thawed, clarified, and incubated with 2 mg of anti-His or anti-GST (GE) (16 hr, 4 C). Immunocomplexes were precipitated with magnetic EXPERIMENTAL PROCEDURES protein-G beads (Dynal) and beads washed before elution with Laemmli buffer. All materials are from Sigma-Aldrich unless otherwise stated. RT-PCR RNA was extracted with TRI reagent, DNase treated (Ambion), and reverse Cell Culture transcribed (Promega) before PCR with the following primers: IL-1R1, AGGA VSMCs, EL4, HEK, and HeLa cells were cultured in DMEM and Jurkat and THP-1 GACGGAGGACTTGTGT and GCGTCATAGGTCTTTCCATC; total IL-1R2, cells in RPMI 1640, all supplemented with penicillin, streptomycin, L-glutamine, CATTACAAGCGGGAGTTCAG and TAGTGCAGACGTAGGTGCCA; soluble and 10% FCS. Human monocyte-derived macrophages were differentiated as IL-1R2, TGGCACCTACGTCTGCACTA and TGTCTCCAAAAGGAAGAGCGA; described previously (Brown et al., 2000). Cells were treated as indicated with GAPDH, TGTTGCCATCAATGACCCCTT and CTCCACGACTGACTCAGCG. Calpeptin (30 mM), Lactacystin (10 mM; both Biomol), IL-1a pAb (1 mg/ml), IL-1a and IL-1b (all PeproTech), IL-1RA (Amgen), IL-1R2 (250 ng; R&D), Z-YVAD, Z-VAD-fmk (both 10 mm; Bachem), LPS (1 mg/ml), EGTA (5 mM), and siRNA Silencing BB-94 (1 mg/ml; Tocris). Cells in serum-free (SF) DMEM were disrupted by freeze IL-1R2 silencing was performed with SMARTpool siRNA and controls thaw in liquid N , clarified, and stored at 80 C. Cells were also made necrotic (Dharmacon). In brief, HeLa cells were transfected with 20 nM of siRNA via by incubation with 7-BIO (25 mM; Enzo) or digitonin (0.1%) (data not shown) or by HiPerFect (QIAGEN), retransfected after 48 hr, and harvested 48 hr later. Immunity 38, 285–295, February 21, 2013 ª2013 Elsevier Inc. 293 Immunity IL-1R2 Controls IL-1a Activity Postnecrosis Immunofluorescence Chamberlain, J., Francis, S., Brookes, Z., Shaw, G., Graham, D., Alp, N.J., For IL-1R2, cells were fixed in 2% formaldehyde and permeabilized with 0.5% Dower, S., and Crossman, D.C. (2009). Interleukin-1 regulates multiple athero- NP-40. For IL-1R2 and p33, cells were fixed in 2% formaldehyde, washed, genic mechanisms in response to fat feeding. PLoS ONE 4, e5073. and fixed with methanol. All were blocked in 1% BSA, incubated with IL-1R2 Chen, G.Y., and Nunez, G. (2010). Sterile inflammation: sensing and reacting to mAb (R&D) and anti-N-terminal IL-1a (Aviva Systems Biology), washed, damage. Nat. Rev. Immunol. 10, 826–837. incubated with Alexa Fluor antibodies (Molecular Probes), washed, and Chen, C.J., Kono, H., Golenbock, D., Reed, G., Akira, S., and Rock, K.L. (2007). imaged on a BX51 (Olympus) or a TCS SP2 AOBS (Leica) microscope. PLA Identification of a key pathway required for the sterile inflammatory response assays with the primary antibodies as above were conducted according triggered by dying cells. Nat. Med. 13, 851–856. to the manufacturer (Olink). For flow cytometry, cells were fixed in 2% formaldehyde, washed, blocked in 1% BSA and 0.05% sodium azide, stained Chi, H., Messas, E., Levine, R.A., Graves, D.T., and Amar, S. (2004). Interleukin-1 receptor signaling mediates atherosclerosis associated with with anti-IL-1R2, washed, stained with secondary antibodies, washed, and analyzed on a C6 flow cytometer (Accuri). Intracellular staining steps had bacterial exposure and/or a high-fat diet in a murine apolipoprotein E hetero- zygote model: pharmacotherapeutic implications. Circulation 110, 1678–1685. 0.5% saponin present. Clarke, M.C., Talib, S., Figg, N.L., and Bennett, M.R. (2010). Vascular smooth muscle cell apoptosis induces interleukin-1-directed inflammation: effects of Animal Protocols hyperlipidemia-mediated inhibition of phagocytosis. Circ. Res. 106, 363–372. Experiments were conducted under UK Home Office licensing. Mice were in- jected intraperitoneally with saline, 0.29 fmol/g body weight of p17 or p33, or Cohen, I., Rider, P., Carmi, Y., Braiman, A., Dotan, S., White, M.R., Voronov, E., 8.3 3 10 necrotic HeLa cells. After 6 hr, the peritoneal cavity was lavaged with Martin, M.U., Dinarello, C.A., and Apte, R.N. (2010). Differential release of chro- 5 ml of PBS. GR-1 cells were counted by FACS after anti-GR-1-FITC staining matin-bound IL-1alpha discriminates between necrotic and apoptotic cell (Biolegend). death by the ability to induce sterile inflammation. Proc. Natl. Acad. Sci. USA 107, 2574–2579. Statistics Colotta, F., Re, F., Muzio, M., Bertini, R., Polentarutti, N., Sironi, M., Giri, J.G., Parametric tests were employed for analysis of continuous ELISA and Dower, S.K., Sims, J.E., and Mantovani, A. (1993). Interleukin-1 type II peritonitis data, conducted with a one-way, two-tailed ANOVA (Excel). receptor: a decoy target for IL-1 that is regulated by IL-4. Science 261, 472–475. Davis, M.J., Hanson, K.A., Clark, F., Fink, J.L., Zhang, F., Kasukawa, T., Kai, SUPPLEMENTAL INFORMATION C., Kawai, J., Carninci, P., Hayashizaki, Y., and Teasdale, R.D. (2006). Differential use of signal peptides and membrane domains is a common occur- Supplemental Information includes Supplemental Experimental Procedures rence in the protein output of transcriptional units. PLoS Genet. 2, e46. and five figures and can be found with this article online at http://dx.doi.org/ 10.1016/j.immuni.2013.01.008. Dinarello, C.A. (1996). Biologic basis for interleukin-1 in disease. Blood 87, 2095–2147. ACKNOWLEDGMENTS Dinarello, C.A. (2009). Immunological and inflammatory functions of the inter- leukin-1 family. Annu. Rev. Immunol. 27, 519–550. We gratefully acknowledge monocyte donations from E. Chilvers’ laboratory Duewell, P., Kono, H., Rayner, K.J., Sirois, C.M., Vladimer, G., Bauernfeind, and J. Skepper for help with confocal. This study was supported by British F.G., Abela, G.S., Franchi, L., Nun˜ ez, G., Schnurr, M., et al. (2010). NLRP3 in- Heart Foundation Grants FS/09/005/26845 (M.C.H.C.), PG/06/024/20354, flammasomes are required for atherogenesis and activated by cholesterol and RG 04/001 (M.R.B.), and NIHR Cambridge BRC. crystals. Nature 464, 1357–1361. Eigenbrod, T., Park, J.H., Harder, J., Iwakura, Y., and Nu´ n˜ ez, G. (2008). Cutting Received: April 11, 2012 edge: critical role for mesothelial cells in necrosis-induced inflammation Accepted: January 18, 2013 through the recognition of IL-1 alpha released from dying cells. J. Immunol. Published: February 7, 2013 181, 8194–8198. Fettelschoss, A., Kistowska, M., LeibundGut-Landmann, S., Beer, H.D., REFERENCES Johansen, P., Senti, G., Contassot, E., Bachmann, M.F., French, L.E., Oxenius, A., and Ku¨ ndig, T.M. (2011). Inflammasome activation and IL-1b Afonina, I.S., Tynan, G.A., Logue, S.E., Cullen, S.P., Bots, M., Lu¨ thi, A.U., target IL-1a for secretion as opposed to surface expression. Proc. Natl. Reeves, E.P., McElvaney, N.G., Medema, J.P., Lavelle, E.C., and Martin, Acad. Sci. USA 108, 18055–18060. S.J. (2011). Granzyme B-dependent proteolysis acts as a switch to enhance the proinflammatory activity of IL-1a. Mol. Cell 44, 265–278. Ghiringhelli, F., Apetoh, L., Tesniere, A., Aymeric, L., Ma, Y., Ortiz, C., Vermaelen, K., Panaretakis, T., Mignot, G., Ullrich, E., et al. (2009). Basu, S., Binder, R.J., Suto, R., Anderson, K.M., and Srivastava, P.K. (2000). Activation of the NLRP3 inflammasome in dendritic cells induces IL-1beta- Necrotic but not apoptotic cell death releases heat shock proteins, which dependent adaptive immunity against tumors. Nat. Med. 15, 1170–1178. deliver a partial maturation signal to dendritic cells and activate the NF-kappa B pathway. Int. Immunol. 12, 1539–1546. Gross, O., Yazdi, A.S., Thomas, C.J., Masin, M., Heinz, L.X., Guarda, G., Quadroni, M., Drexler, S.K., and Tschopp, J. (2012). Inflammasome activators Beck, G., and Habicht, G.S. (1986). Isolation and characterization of a primitive induce interleukin-1a secretion via distinct pathways with differential require- interleukin-1-like protein from an invertebrate, Asterias forbesi. Proc. Natl. ment for the protease function of caspase-1. Immunity 36, 388–400. Acad. Sci. USA 83, 7429–7433. Kamari, Y., Werman-Venkert, R., Shaish, A., Werman, A., Harari, A., Gonen, A., Brough, D., Le Feuvre, R.A., Wheeler, R.D., Solovyova, N., Hilfiker, S., Voronov, E., Grosskopf, I., Sharabi, Y., Grossman, E., et al. (2007). Differential Rothwell, N.J., and Verkhratsky, A. (2003). Ca2+ stores and Ca2+ entry differ- role and tissue specificity of interleukin-1alpha gene expression in atherogen- entially contribute to the release of IL-1 beta and IL-1 alpha from murine esis and lipid metabolism. Atherosclerosis 195, 31–38. macrophages. J. Immunol. 170, 3029–3036. Brown, S.B., Clarke, M.C., Magowan, L., Sanderson, H., and Savill, J. (2000). Keller, M., Ru¨ egg, A., Werner, S., and Beer, H.D. (2008). Active caspase-1 is a regulator of unconventional protein secretion. Cell 132, 818–831. Constitutive death of platelets leading to scavenger receptor-mediated phagocytosis. A caspase-independent cell clearance program. J. Biol. Kobayashi, Y., Yamamoto, K., Saido, T., Kawasaki, H., Oppenheim, J.J., and Chem. 275, 5987–5996. Matsushima, K. (1990). Identification of calcium-activated neutral protease as a processing enzyme of human interleukin 1 alpha. Proc. Natl. Acad. Sci. USA Cayrol, C., and Girard, J.P. (2009). The IL-1-like cytokine IL-33 is inactivated after maturation by caspase-1. Proc. Natl. Acad. Sci. USA 106, 9021–9026. 87, 5548–5552. 294 Immunity 38, 285–295, February 21, 2013 ª2013 Elsevier Inc. Immunity IL-1R2 Controls IL-1a Activity Postnecrosis Kono, H., and Rock, K.L. (2008). How dying cells alert the immune system to Orlando, S., Sironi, M., Bianchi, G., Drummond, A.H., Boraschi, D., Yabes, D., danger. Nat. Rev. Immunol. 8, 279–289. and Mantovani, A. (1997). Role of metalloproteases in the release of the IL-1 type II decoy receptor. J. Biol. Chem. 272, 31764–31769. Kono, H., Karmarkar, D., Iwakura, Y., and Rock, K.L. (2010). Identification of the cellular sensor that stimulates the inflammatory response to sterile cell Prudovsky, I., Mandinova, A., Soldi, R., Bagala, C., Graziani, I., Landriscina, M., Tarantini, F., Duarte, M., Bellum, S., Doherty, H., and Maciag, T. (2003). death. J. Immunol. 184, 4470–4478. The non-classical export routes: FGF1 and IL-1alpha point the way. J. Cell Kuida, K., Lippke, J.A., Ku, G., Harding, M.W., Livingston, D.J., Su, M.S., and Sci. 116, 4871–4881. Flavell, R.A. (1995). Altered cytokine export and apoptosis in mice deficient in Rajama¨ ki, K., Lappalainen, J., Oo¨ rni, K., Va¨ lima¨ ki, E., Matikainen, S., Kovanen, interleukin-1 beta converting enzyme. Science 267, 2000–2003. P.T., and Eklund, K.K. (2010). Cholesterol crystals activate the NLRP3 inflam- Laios, A., O’Toole, S.A., Flavin, R., Martin, C., Ring, M., Gleeson, N., D’Arcy, T., masome in human macrophages: a novel link between cholesterol metabolism McGuinness, E.P., Sheils, O., Sheppard, B.L., and O’ Leary, J.J. (2008). An and inflammation. PLoS ONE 5, e11765. integrative model for recurrence in ovarian cancer. Mol. Cancer 7,8. Rao, D.A., Tracey, K.J., and Pober, J.S. (2007). IL-1alpha and IL-1beta are Levine, C.G., Mitra, D., Sharma, A., Smith, C.L., and Hegde, R.S. (2005). The endogenous mediators linking cell injury to the adaptive alloimmune response. efficiency of protein compartmentalization into the secretory pathway. Mol. J. Immunol. 179, 6536–6546. Biol. Cell 16, 279–291. Rao,D.A.,Eid, R.E.,Qin,L.,Yi,T.,Kirkiles-Smith,N.C.,Tellides,G.,and Li, P., Allen, H., Banerjee, S., Franklin, S., Herzog, L., Johnston, C., McDowell, Pober, J.S. (2008). Interleukin (IL)-1 promotes allogeneic T cell intimal infiltra- J., Paskind, M., Rodman, L., Salfeld, J., et al. (1995). Mice deficient in IL-1 tion and IL-17 production in a model of human artery rejection. J. Exp. Med. beta-converting enzyme are defective in production of mature IL-1 beta and 205, 3145–3158. resistant to endotoxic shock. Cell 80, 401–411. Rock, K.L., Latz, E., Ontiveros, F., and Kono, H. (2010). The sterile inflamma- Liu, C., Hart, R.P., Liu, X.J., Clevenger, W., Maki, R.A., and De Souza, E.B. tory response. Annu. Rev. Immunol. 28, 321–342. (1996). Cloning and characterization of an alternatively processed human Ru¨ ckert, F., Dawelbait, G., Winter, C., Hartmann, A., Denz, A., Ammerpohl, O., type II interleukin-1 receptor mRNA. J. Biol. Chem. 271, 20965–20972. Schroeder, M., Schackert, H.K., Sipos, B., Klo¨ ppel, G., et al. (2010). Luheshi, N.M., Kovacs, K.J., Lopez-Castejon, G., Brough, D., and Denes, A. Examination of apoptosis signaling in pancreatic cancer by computational (2011). Interleukin-1a expression precedes IL-1b after ischemic brain injury signal transduction analysis. PLoS ONE 5, e12243. and is localised to areas of focal neuronal loss and penumbral tissues. Sakurai, T., He, G., Matsuzawa, A., Yu, G.Y., Maeda, S., Hardiman, G., and J. Neuroinflammation 8, 186. Karin, M. (2008). Hepatocyte necrosis induced by oxidative stress and IL-1 Lu¨ thi, A.U., Cullen, S.P., McNeela, E.A., Duriez, P.J., Afonina, I.S., Sheridan, alpha release mediate carcinogen-induced compensatory proliferation and C., Brumatti, G., Taylor, R.C., Kersse, K., Vandenabeele, P., et al. (2009). liver tumorigenesis. Cancer Cell 14, 156–165. Suppression of interleukin-33 bioactivity through proteolysis by apoptotic Sauter, B., Albert, M.L., Francisco, L., Larsson, M., Somersan, S., and caspases. Immunity 31, 84–98. Bhardwaj, N. (2000). Consequences of cell death: exposure to necrotic tumor March, C.J., Mosley, B., Larsen, A., Cerretti, D.P., Braedt, G., Price, V., Gillis, cells, but not primary tissue cells or apoptotic cells, induces the maturation of S., Henney, C.S., Kronheim, S.R., Grabstein, K., et al. (1985). Cloning, immunostimulatory dendritic cells. J. Exp. Med. 191, 423–434. sequence and expression of two distinct human interleukin-1 complementary Scaffidi, P., Misteli, T., and Bianchi, M.E. (2002). Release of chromatin protein DNAs. Nature 315, 641–647. HMGB1 by necrotic cells triggers inflammation. Nature 418, 191–195. Matzinger, P. (1994). Tolerance, danger, and the extended family. Annu. Rev. Sims, J.E., and Smith, D.E. (2010). The IL-1 family: regulators of immunity. Nat. Immunol. 12, 991–1045. Rev. Immunol. 10, 89–102. Melcher, A., Todryk, S., Hardwick, N., Ford, M., Jacobson, M., and Vile, R.G. Stewart, C.R., Stuart, L.M., Wilkinson, K., van Gils, J.M., Deng, J., Halle, A., (1998). Tumor immunogenicity is determined by the mechanism of cell death Rayner, K.J., Boyer, L., Zhong, R., Frazier, W.A., et al. (2010). CD36 ligands via induction of heat shock protein expression. Nat. Med. 4, 581–587. promote sterile inflammation through assembly of a Toll-like receptor 4 and 6 heterodimer. Nat. Immunol. 11, 155–161. Menu, P., Pellegrin, M., Aubert, J.F., Bouzourene, K., Tardivel, A., Mazzolai, L., and Tschopp, J. (2011). Atherosclerosis in ApoE-deficient mice progresses Tokunaga, H., Saito, S., Sakai, K., Yamaguchi, R., Katsuyama, I., Arakawa, T., independently of the NLRP3 inflammasome. Cell Death Dis. 2, e137. Onozaki, K., Arakawa, T., and Tokunaga, M. (2010). Halophilic beta-lactamase as a new solubility- and folding-enhancing tag protein: production of native Michaud, M., Martins, I., Sukkurwala, A.Q., Adjemian, S., Ma, Y., Pellegatti, P., human interleukin 1alpha and human neutrophil alpha-defensin. Appl. Shen, S., Kepp, O., Scoazec, M., Mignot, G., et al. (2011). Autophagy- Microbiol. Biotechnol. 86, 649–658. dependent anticancer immune responses induced by chemotherapeutic agents in mice. Science 334, 1573–1577. Wang, K.K. (2000). Calpain and caspase: can you tell the difference? Trends Neurosci. 23, 20–26. Mosley, B., Urdal, D.L., Prickett, K.S., Larsen, A., Cosman, D., Conlon, P.J., Gillis, S., and Dower, S.K. (1987). The interleukin-1 receptor binds the human Wang, D., Zhang, S., Li, L., Liu, X., Mei, K., and Wang, X. (2010). Structural interleukin-1 alpha precursor but not the interleukin-1 beta precursor. J. Biol. insights into the assembly and activation of IL-1b with its receptors. Nat. Chem. 262, 2941–2944. Immunol. 11, 905–911. Immunity 38, 285–295, February 21, 2013 ª2013 Elsevier Inc. 295 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Immunity Unpaywall

Intracellular Interleukin-1 Receptor 2 Binding Prevents Cleavage and Activity of Interleukin-1α, Controlling Necrosis-Induced Sterile Inflammation

Download PDF
11 pages

Loading next page...
 
/lp/unpaywall/intracellular-interleukin-1-receptor-2-binding-prevents-cleavage-and-kj3yVAAKgg

References

References for this paper are not available at this time. We will be adding them shortly, thank you for your patience.

Publisher
Unpaywall
ISSN
1074-7613
DOI
10.1016/j.immuni.2013.01.008
Publisher site
See Article on Publisher Site

Abstract

Immunity Article Intracellular Interleukin-1 Receptor 2 Binding Prevents Cleavage and Activity of Interleukin-1a, Controlling Necrosis-Induced Sterile Inflammation 1,3 1,3 2 1 1, Yue Zheng, Melanie Humphry, Janet J. Maguire, Martin R. Bennett, and Murray C.H. Clarke * Division of Cardiovascular Medicine Clinical Pharmacology Unit Department of Medicine, University of Cambridge, Addenbrooke’s Hospital, Cambridge CB2 0QQ, UK These authors contributed equally to this work *Correspondence: [email protected] http://dx.doi.org/10.1016/j.immuni.2013.01.008 Open access under CC BY license. SUMMARY (Clarke et al., 2010; Kamari et al., 2007), graft rejection (Rao et al., 2007, 2008), toxic liver insults (Chen et al., 2007), and ischemia- Necrosis can induce profound inflammation or be reperfusion injury (Cohen et al., 2010; Luheshi et al., 2011). clinically silent. However, the mechanisms underlying The prototypic IL-1 family is ancient, with homologs identified back to echinoderms (Beck and Habicht, 1986). IL-1a, one of the such tissue specificity are unknown. Interleukin-1a principal ligands, is expressed by most lineages as a signal (IL-1a) is a key danger signal released upon necrosis peptide-less protein, is not readily secreted (Dinarello, 2009), that exerts effects on both innate and adaptive and is actively retained during apoptosis (Cohen et al., 2010). immunity and is considered to be constitutively Once released into the extracellular milieu, IL-1a ligation of the active. In contrast, we have shown that necrosis- type 1 IL-1 receptor (IL-1R1) leads to multiple proinflammatory induced IL-1a activity is tightly controlled in a cell effects (Dinarello, 2009), including cytokine secretion, neutrophil type-specific manner. Most cell types examined ex- recruitment, and upregulation of major histocompatibility pressed a cytosolic IL-1 receptor 2 (IL-1R2) whose complex (MHC) and costimulatory molecules on antigen- binding to pro-IL-1a inhibited its cytokine activity. In presenting cells. IL-1a also has powerful effects on adaptive cell types exhibiting a silent necrotic phenotype, immunity by enhancing expansion and survival of T cells, differ- entiation of T helper 17 (Th17) cells, and effector T cell prolifera- IL-1R2 remained associated with pro-IL-1a. Cell tion in the presence of regulatory T cells (Sims and Smith, 2010). types possessing inflammatory necrotic phenotypes These potent effects mean that activity is tightly controlled at either lacked IL-1R2 or had activated caspase-1 multiple levels. Mice deficient in IL-1a or IL-1b exhibit no before necrosis, which degraded and dissociated phenotype. However, mice lacking the IL-1 receptor antagonist IL-1R2 from pro-IL-1a. Full IL-1a activity required (IL-1RA) have small litters and retarded growth and develop cleavage by calpain after necrosis, which increased spontaneous arthritis-like polyarthropathy, arteritis, and cancer its affinity for IL-1 receptor 1. Thus, we report a cell (Dinarello, 2009). Indeed, increased IL-1 activity is a hallmark type-dependent process that fundamentally governs of many chronic inflammatory conditions, including rheumatoid IL-1a activity postnecrosis and the mechanism arthritis, gout, diabetes, atherosclerosis, and psoriasis (Dinar- allowing conditional release of this blockade. ¨ ello, 1996, 2009; Duewell et al., 2010; Rajamaki et al., 2010). IL-1 family members are synthesized as inactive precursors INTRODUCTION unable to bind their receptor, providing an initial level of control. IL-1b and IL-18 are activated by caspase-1, which requires inflam- masome formation. In contrast, IL-33 processing by caspase-3 Understanding why the immune system responds to necrosis and how this is controlled is critical in unraveling multiple human or caspase-1 results in inactivation (Cayrol and Girard, 2009; Lu¨ thi diseases. The ‘‘danger’’ model proposes that immunity responds et al., 2009). Pro-IL-1a (p33) is processed to mature IL-1a (p17) by to nonphysiological cell death, damage, or stress (Matzinger, calpain (Kobayashi et al., 1990), but the biological consequences 1994). Accordingly, necrotic death releases damage-associated of cleavage are unknown given that p33 is reported to be fully molecular patterns (DAMPs), which are sensed as danger and active. This finding is credited to two papers, but one only act as universal signals to activate immunity (Chen and Nun˜ ez, discusses p33 activity (March et al., 1985), whereas activity within 2010; Kono and Rock, 2008; Rock et al., 2010). DAMPs are the second study may be compromised by p33 degradation retained in healthy cells and during apoptosis (Basu et al., 2000; (Mosley et al., 1987). Interestingly, calpain is activated upon loss Cohen et al., 2010; Scaffidi etal., 2002), whereas necrosis releases of plasma membrane integrity (Wang, 2000), suggesting that them into the extracellular milieu. Interleukin-1a (IL-1a)isan calpain cleavage of IL-1a could be a control point for activity post- important DAMP that activates immunity postnecrosis (Chen necrosis. Although a recent study reports increased IL-1a activity et al., 2007; Clarke et al., 2010; Cohen et al., 2010; Eigenbrod after granzyme B cleavage (Afonina et al., 2011), differential et al., 2008; Kono et al., 2010; Rao et al., 2007), driving pathologies efficacy of p33 and p17 IL-1a is still controversial (Gross et al., as diverse as tumorigenesis (Sakurai et al., 2008), atherosclerosis 2012), and no mechanism to explain this has been reported. Immunity 38, 285–295, February 21, 2013 ª2013 Elsevier Inc. 285 Immunity IL-1R2 Controls IL-1a Activity Postnecrosis Figure 1. Necrosis-Induced Sterile Inflam- mation Is Cell Type Specific (A) IL-6 concentrations in conditioned media of macrophages (Macs), Jurkat cells, or VSMCs incubated with lysates from their respective necrotic cells or with IL-1a. (B and C) Immunoblots of IL-1a content and pro- cessing in whole cell (WC) or necrotic lysates (NL) (B) or in necrotic cell lysates pretreated with protease inhibitors as indicated (C). (D and E) IL-6 and MCP-1 concentrations in conditioned media of VSMCs incubated with necrotic VSMC lysates made in the presence of D protease inhibitors (D), or with IL-1a alone, or with calpeptin or EGTA (E). Data represent mean ± SD; *p % 0.007 versus control, n = 3; **p % 0.03 (MCP-1), p % 0.002 (IL- 6), n R 4. NS, not significant. See also Figure S1. inflammation, we analyzed IL-6 release from VSMCs, Jurkat cells, and primary macrophages treated with lysates from E these cell types undergoing necrosis. Only necrotic VSMCs induced significant IL-6 release from viable VSMCs, previ- ously shown to be IL-1a dependent (Clarke et al., 2010), and only VSMCs re- sponded to IL-1a (Figure 1A). Little IL-1a activity was found in membrane fractions of necrotic macrophages or Jurkat cells (Figure S1A available online) or in lysates from cells undergoing hypoxia-induced necrosis (Figure S1B). Although compa- rable amounts of IL-1a were found in all three cell types, only VSMCs processed We report that necrosis-induced IL-1a-dependent responses p33 to p17 IL-1a (Figure 1B). Calpain cleaves p33 to p17 are highly cell type dependent and correlate with calpain (Kobayashi et al., 1990); however, the relevance of cleavage cleavage of IL-1a during necrosis. Contrary to current under- to IL-1a activity is unknown. Several reports indicate that pro- standing, p33 requires calpain processing for full biological cessing is not required for secretion (Brough et al., 2003; activity, which increases its affinity for IL-1R1. Cell type depen- Prudovsky et al., 2003), and cytokine activity is supposedly inde- dency occurs due to expression of an intracellular form of IL- pendent of cleavage (March et al., 1985; Mosley et al., 1987). 1R2 that binds IL-1a, preventing calpain cleavage and cytokine VSMCs rapidly cleaved p33 to p17 during necrosis, which activity. After inflammasome activation, caspase-1 specifically was prevented by the calpain inhibitors calpeptin and EGTA but not the proteasomal inhibitor lactacystin (Figure 1C). cleaves IL-1R2, which abrogates IL-1a binding, allows calpain Necrotic VSMC lysates prepared with calpain inhibition cleavage, and completely restores IL-1a-dependent responses. Regulated secretion of IL-1a also requires IL-1R2 cleavage. produced significantly reduced responses (Figure 1D), whereas Thus, we report an important cell type-dependent mechanism calpeptin or EGTA treatment did not inhibit cytokine production that fundamentally governs IL-1a activity postnecrosis and the in response to IL-1a (Figure 1E). However, prolonged incubation mechanism allowing conditional release of this blockade. at 37 C in the presence of calpeptin resulted in some processing of p33 to p17 (Figure S1C), which could contribute to the RESULTS apparent IL-1a activity seen with calpeptin-treated VSMCs (Figure 1D). Nevertheless, taken together this suggests that, Necrosis-Induced Sterile Inflammation Is Cell Type contrary to the accepted literature, p33 is not fully active until Specific and Correlates with Calpain Cleavage cleaved. of pro-IL-1a We have recently shown that IL-1a released during vascular Calpain-Cleaved p33 and Recombinant p17 Are More smooth muscle cell (VSMC) necrosis is a powerful DAMP that Active than p33 IL-1a induces local vessel inflammation (Clarke et al., 2010). To deter- To directly compare p17 and p33 activity, we purified HIS- mine which other cell types could drive IL-1a-dependent sterile tagged recombinant proteins. Although we could express and 286 Immunity 38, 285–295, February 21, 2013 ª2013 Elsevier Inc. Immunity IL-1R2 Controls IL-1a Activity Postnecrosis Figure 2. Calpain-Cleaved p33 and Recom- AB binant p17 Is More Active than p33 IL-1a (A) Cytokine concentrations in conditioned media of VSMCs treated with p33 or calpain-cleaved p33, ± IL-1a neutralizing antibody (a pAb). (B) VSMCs were also treated with calpain sham reactions (no p33) or commercial recombinant p17 IL-1a (cr17) ± calpain sham. (C) IL-2 concentration in conditioned media of murine EL4 cells treated with p33 ± calpain. (D) Immunoblot of in vitro cleavage of p33 by calpain. (E) Coomassie stain of purified soluble p17- and p33-GST fusion proteins. (F and G) IL-2 (F) or IL-6 (G) concentrations in DE conditioned media of EL4 or VSMCs, respectively, treated with 1 nM p17- or p33-GST ± a pAb. (H) IL-2 concentrations in conditioned media of EL4 cells incubated with p33-GST ± calpain, and ± a pAb. (I) GR1 cells recruited intraperitoneally in wild- type or Il1r1 mice injected with saline or 29 fmol/g p17 or p33. Data represent mean ± SD or mean ± SEM (I); *p % 0.0003, n = 4 (A), n = 3 (B, C) protein prepa- rations and cleavage reactions; **p% 0.007, n = 4. NS, not significant. F G indicating that the increased activity is due to processing of p33 to p17. These data were also reproduced cross-species with the classic IL-1-sensitive murine cell line EL4 (Figure 2C), suggesting a conserved requirement for calpain processing of IL-1a. Although calpain HI cleavage of IL-1a increased activity, these differences represent an underesti- mate because of the nonstoichiometric conversion of p33 to p17 during in vitro cleavage (Figure 2D). By fusing IL-1a to GST, we purified small amounts of p17 and p33 as soluble proteins (145 mg/l) (Figure 2E). Compar- ison at equal molarities demonstrated significant increases in IL-2 release from EL4 cells (Figure 2F) and IL-6 release from VSMCs for p17 compared to p33 (Figure 2G). IL-1a neutralization reduced purify p17 in E. coli, p33 was very insoluble and prone to responses to control indicating IL-1a-dependent stimulation, aggregation (not shown)—a finding reported by others excluding effects from copurified bacterial PAMPs (Figures 2F (Tokunaga et al., 2010). After denaturation in urea, p33 could and 2G). To eliminate whether GST-p33 was ‘‘functionally be purified and renatured to a soluble protein after sequential dead’’ as a result of a purification artifact or intrinsic instability, dialysis, but this made comparison between p17 and p33 we cleaved it with calpain or incubated it at 37 C for 16 hr, impossible because of the inability to control for refolding effi- respectively. Cleavage of GST-p33 significantly restored ciency between different proteins. To circumvent this, we activity (Figure 2H), whereas incubation did not result in protein cleaved p33 in vitro with calpain, which increased cytokine degradation (not shown), again supporting an inherent lower release compared to uncleaved p33 (Figure 2A). Importantly, activity of p33. Lastly, injection of p17 into the peritoneum a neutralizing antibody to IL-1a reduced responses to control resulted in significantly increased neutrophil recruitment (Figure 2A), whereas a ‘‘calpain sham’’ reaction without p33 compared to injection of p33 in wild-type mice, a result that neither promoted nor inhibited cytokine release (Figure 2B), was not seen in Il1r1 mice (Figure 2I). Immunity 38, 285–295, February 21, 2013 ª2013 Elsevier Inc. 287 Immunity IL-1R2 Controls IL-1a Activity Postnecrosis Figure 3. p33 IL-1a Shows Minimal Activity A B at Physiological Concentrations because of Lower Receptor Affinity (A and B) Concentration-response curves to p17 and p33, minus control, in VSMCs (50 pM to 8 nM) (A) or EL4 cells (5 pM to 8 nM) (B). (C and D) IL-2 concentrations in conditioned media from EL4 cells treated with p17 or p33 at 4 nM (C), and 0.1 nM p17 or 4 nM p33 (D), with increasing concentrations of IL-1RA. Data represent mean ± SD; *p % 0.03, n = 3 (A), n R 4 (B); for difference in IC p < 0.0001 (C), C D 50 p = 0.52 (D), n = 3. See also Figure S2. IL-1RA completely inhibited cytokine release and gave an IC of 10.3 ± 2.9 ng/ml (Figure 3D). Thus, a 40-fold difference in concentration gave near superimposable curves (p = 0.52), corre- lating well with the value above. Together these data show that, in contrast to previous reports, IL-1a is not fully active p33 IL-1a Shows Minimal Activity and Behaves as until processed by calpain and that cell types unable to cleave a Partial Agonist due to a Lower Receptor Affinity IL-1a induce much smaller IL-1a-dependent responses after To understand why p33 is less active than p17, we generated necrosis. cytokine concentration-response curves. VSMCs responded to p17 at 50 pM and saturated at 2 nM (Figure 3A). In contrast, An Intracellular Form of IL-1R2 Protects p33 IL-1a p33 generated responses only at supraphysiological amounts from Calpain Processing and failed to saturate (Figure 3A). At physiological concentra- We hypothesized that cell type-specific cleavage of IL-1a could tions of 1 nM, p17 was 50 times more active than p33. EL4 cells occur through modification of enzyme or substrate, excess of are very sensitive to IL-1a, with responses detectable with fM a calpain inhibitor, or a binding partner that protects IL-1a from amounts. p17 activity was detectable at 5 pM and rapidly satu- processing. Jurkat and macrophage necrotic lysates incubated rated at 0.5 nM (Figure 3B). However, p33 gave a linear concen- at 37 C for extended periods still showed no processing of tration response, such that 65% of the maximal response was endogenous p33 (Figure 4A), whereas recombinant His-p33 obtained at 8 nM but did not reach saturation (Figure 3B). At spiked into Jurkat lysates also remained uncleaved (Figure 4A), 1 nM p17 was 10 times more active than p33. To exclude steric excluding a modification to the cell-derived cytokine. Similar hindrance of fused GST only on p33, we specifically cleaved off results were also found with primary T lymphocytes (Fig- GST. Comparison of p33 with or without GST revealed identical ure S3A). An excess of calpain inhibitor was excluded, because efficacies at a range of concentrations (Figure S2). Together, this high calpain activity was found in all necrotic cell types (Fig- suggests that p33 can bind IL-1R1 at higher concentrations and ure S3B). Addition of purified calpain to Jurkat necrotic lysates thus acts as a partial agonist. also failed to cleave p33 (Figure 4B), whereas analysis of spec- To investigate whether p33 binds IL-1R1 but fails to induce trin, an alternative calpain substrate, revealed calpain-depen- downstream signaling, we conducted competition experiments dent processing in both Jurkat and VSMCs necrotic lysates with a fixed concentration of p17 (1 nM) and increasing concen- (Figure 4C). trations of p33 (up to 8 nM). However, even 8 nM p33 failed to Protection of recombinant p33 by Jurkat necrotic lysates inhibit cytokine responses (data not shown), suggesting suggested that an excess of a binding partner must be present, a stronger affinity of p17 for the receptor. IL-1RA binds IL-1R1 and therefore that it could be transferred. Indeed, necrotic but does not induce signaling (Dinarello, 2009). Therefore the lysates made from mixed Jurkat and VSMCs displayed no difference in concentration of IL-1RA required to inhibit 50% p33 cleavage (Figure 4D). Known binding partners of IL-1a of the cytokine response induced by p17 or p33 (the IC ) include the type 1 and 2 IL-1 receptors; therefore we treated directly relates to the difference in affinity of the two ligands Jurkat necrotic lysates with a large excess of IL-1RA, which for IL-1R1. Responses to 4 nM p33 were completely inhibited resulted in calpain-specific cleavage of p33 (Figure S3C). The by 100 ng/ml IL-1RA (IC of 13.7 ± 3.8 ng/ml) (Figure 3C). In only known receptors for IL-1RA are IL-1R1 and IL-1R2, sug- contrast, 1,600 ng/ml IL-1RA failed to inhibit responses to gesting that IL-1a may be bound to an intracellular IL-1R. 4 nM p17, but 50% inhibition occurred at 633.3 ± 16.7 ng/ml (Fig- RT-PCR revealed that VSMCs express only IL-1R1, whereas ure 3C). This indicates that p33 has an affinity for IL-1R1 46.3 Jurkat cells (Figure 4E) and macrophages (data not shown) times lower than does p17 (p < 0.0001). To exclude p17 express IL-1R2. Immunoblots confirmed IL-1R2 expression in responses from another receptor insensitive to IL-1RA blockade, Jurkat cells (Figure 4F), macrophages, and T cells (Figure S3D), we repeated with 0.1 nM p17. Under these conditions 100 ng/ml whereas addition of recombinant IL-1R2 to a cell-free cleavage 288 Immunity 38, 285–295, February 21, 2013 ª2013 Elsevier Inc. Immunity IL-1R2 Controls IL-1a Activity Postnecrosis AB GH IJ Figure 4. An Intracellular Form of IL-1R2 Protects IL-1a from Calpain Processing (A and B) Immunoblot for endogenous IL-1a processing in necrotic lysates of primary macrophages (Macs) and Jurkat cells (A), for exogenous His-p33 in necrotic Jurkat lysates (A), or for Jurkat necrotic lysates treated with increasing amounts of purified calpain (B). (C and D) Immunoblot of whole cell (WC) and necrotic (NL) Jurkat and VSMC lysates for a-spectrin (C) or IL-1a after mixing of lysates from both cell types (D). (E and F) RT-PCR (E) or immunoblot (F) for IL-1 receptors in VSMCs and Jurkat cells. (G and H) Immunoblot for IL-1a cleavage in a cell-free system with purified calpain (G) and in necrotic VSMC lysates (H), ± 250 ng IL-1R2. (I and J) Coimmunoprecipitation of IL-1R2 with p33-GST (I) and p33 with IL-1R2-HIS (J) in transfected HEK cell lysates. See also Figure S3. reaction (Figure 4G) or necrotic VSMC lysates (Figure 4H) pre- HIS (Figures 4I and 4J), whereas a proximity ligation assay vented calpain-dependent p33 processing in a dose-depen- demonstrated association of endogenous proteins in situ (Fig- dent manner (Figure S3E). IL-1R2 also antagonized p33 in ure S3G), supporting a direct interaction. Finally, IL-1R2 also a dose-dependent manner (Figure S3F). Importantly, IL-1R2 protected IL-1a from cleavage by the inflammatory proteases coimmunoprecipitated with p33-GST and p33 with IL-1R2- granzyme B, chymase, and elastase (Figure S3H). Immunity 38, 285–295, February 21, 2013 ª2013 Elsevier Inc. 289 Immunity IL-1R2 Controls IL-1a Activity Postnecrosis Figure 5. Silencing of IL-1R2 Enables IL-1a Processing and Restores Necrotic Cell- Induced Inflammation (A–C) siRNA-mediated silencing reduces IL-1R2 mRNA (A) and protein in whole cell (WC) and necrotic lysates (NL) (B), which leads to calpain- dependent IL-1a cleavage upon necrosis (C) that is inhibited with calpeptin (+C). (D and E) IL-2 concentrations in conditioned media of EL4 cells incubated with necrotic lysates from C HeLa cells treated with siRNA to IL-1R2 or scrambled, either alone (D) or with IL-1b treatment (E), ± a pAb. Data represent mean ± SD; *p = 0.0007, n = 3 independent silencings and treatments. NS, not significant. (F) GR1 cells recruited intraperitoneally in wild- type or Il1r1 mice injected with saline or 8.3 3 10 control or IL-1R2-silenced necrotic HeLa cells. Data represent mean ± SEM; **p % 0.006, n R 4. NS, not significant. (G) Immunoblot for IL-1R2 and IL-1a in necrotic lysates of empty vector or soluble IL-1R2-trans- fected VSMCs. See also Figure S4. receptor that limits bioavailability of IL-1 (Colotta et al., 1993), we examined whether IL-1R2 silencing had reversed ‘‘general’’ IL-1 antagonism within necrotic lysates. By neutralizing all IL-1a activity within necrotic lysates, we com- pared responsiveness to IL-1b with or without scrambled or IL-1R2-silenced necrotic lysates present (Figure 5E). This revealed no significant difference in IL-2 production, implying that IL-1R2 silencing hadn’t simply reduced general IL-1 antagonism. Sterile peritonitis induced with scrambled control necrotic lysates recruited equal numbers of neutrophils in wild-type and Il1r1 mice, therefore representing IL-1-inde- pendent responses (Figure 5F). However, IL-1R2-silenced necrotic lysates re- cruited 3-fold more cells, which was IL-1 dependent because of loss of response in Il1r1 mice (Figure 5F). Because IL-1R2 Silencing Enables Calpain Cleavage of p33 IL-1a previously described forms of IL-1R2 retain a signal peptide and Restores Inflammatory Response to Necrotic Cells (Liu et al., 1996), they would be expected to be in the exocytic Low transfection efficiency and suspension growth made Jurkat pathway. Therefore we determined whether cell surface-shed cells unsuitable for siRNA silencing. Therefore, necrotic HeLa IL-1R2 (Orlando et al., 1997) protects IL-1a by inhibiting metallo- cells were tested for p33 cleavage (Figure S4A), calpain activity proteases with BB-94; p33, however, remained uncleaved (Fig- (data not shown), and IL-1R2 expression (Figures 5A and 5B), ure S4B). Furthermore, immunofluorescence revealed a large which indicated that they responded similarly. siRNA to IL-1R2 amount of IL-1R2 to be intracellular (Figure S4C) in multiple cell reduced both IL-1R2 mRNA (Figure 5A) and protein (Figure 5B), types (Figure S4D) and endogenous p33 and IL-1R2 to be highly which was not seen with scrambled control. After IL-1R2 colocalized (Figure S4E) throughout the cytoplasm (Figure S4F). silencing, necrotic HeLa lysates displayed calpain-dependent In addition, subcellular fractionation (Figure S4G) and protease K processing of p33 to p17 (Figure 5C) and restoration of IL- protection assays (Figure S4H) revealed a native pool of IL-1R2 1a-dependent responses (Figure 5D), whereas scrambled in the cytosol without internal membrane disruption, as evi- control cells did not. Because IL-1R2 functions as a decoy denced by a lack of calreticulin in cytosolic fractions. Finally, 290 Immunity 38, 285–295, February 21, 2013 ª2013 Elsevier Inc. Immunity IL-1R2 Controls IL-1a Activity Postnecrosis Figure 6. Caspase-1 Cleavage of IL-1R2 A B Restores Necrotic Cell-Induced IL-1a- Dependent Inflammation (A) Immunoblot for IL-1a in necrotic lysates from LPS- and ATP-treated macrophages ± calpeptin (during lysis) or ± caspase inhibitor Z-VAD (during LPS and ATP treatment). (B) IL-2 concentrations in conditioned media of EL4 cells incubated with necrotic lysates from control or LPS- and ATP-treated macrophages ± a and/or b pAb, or ± Z-VAD (during LPS and ATP treatment). Data represent mean ± SEM; *p < 0.00004, **p % 0.02, n R 3. (C and D) Immunoblot for IL-1R2 cleavage after incubation with active caspase-1 ± Z-YVAD (C) or a panel of active caspases (D). (E) Edman degradation of cleaved IL-1R2 detected two sequences (italic underlined) corresponding to processing at two tetrapeptide sites (large bold). (F) Immunoblot for cleaved IL-1R2 after mutation of both caspase-1 sites. EF (G) IL-2 concentrations in conditioned media of EL4 cells incubated with IL-1a, and IL-1R2 ± caspase-1 cleavage. Data represent mean ± SD; *p = 0.009, n = 3. (H) IL-1a concentration in conditioned media of GH Wild-type activated THP-1 cells transfected with caspase site mutant IL-1R2, or as indicated. Data repre- sentative of mean ± SD from n = 2. See also Figure S5. IL-1a- and IL-1b-neutralizing antibodies (Figure 6B). IL-1R2 contains many pre- dicted caspase sites (Figure S5A) and cell-free cleavage resulted in caspase-1- Wild-type specific processing (Figure 6C). IL-1R2 could be cleaved by inflammatory caspase-1 and caspase-5, but not by caspase-4 or the apoptotic caspase-3 expression of exogenous soluble IL-1R2 in VSMCs prevented (Figure 6D). To confirm the cleavage site, we mutated the P1 p33 cleavage (Figure 5G). Together these results demonstrate position Asp to Ala, resulting in a nonconsensus sequence for that a cytosolic complex of p33 and IL-1R2 exists in many cell caspases. Individual mutation of all sites that could give the types. During necrosis this prevents calpain cleavage of p33 to correct sized products did not prevent IL-1R2 processing (data the fully active p17 form, which reduces IL-1a-dependent not shown). However, Edman degradation of the C-terminal responses. IL-1R2 fragment sequenced two products at equal abundance corresponding to cleavage at two separate Asp residues 12 Caspase-1 Specifically Cleaves IL-1R2, which Restores amino acids apart (Figure 6E), and subsequent mutation of IL-1a-Dependent Inflammation Postnecrosis both these sites prevented cleavage (Figure 6F). The crystal Macrophages are suggested to be the sensors of necrotic- structure of IL-1R2 complexed to IL-1 (Wang et al., 2010) re- derived DAMPs, which then release IL-1a to drive sterile inflam- vealed that cleavage at these sites would remove the D1 mation (Chen et al., 2007; Kono et al., 2010). However, our domain vital for binding. Indeed, antagonistic activity toward current data indicate that macrophage IL-1a is bound to IL- IL-1a was lost upon processing (Figure 6G), whereas sham 1R2, and therefore is nonfunctional. Because recent work cleavage reactions neither promoted nor inhibited IL-2 release demonstrates that IL-1a secretion requires inflammasome acti- (Figure S5B), confirming that cleaved IL-1R2 cannot bind IL- vation (Fettelschoss et al., 2011), we investigated whether this 1a. Finally, caspase-1 cleavage of IL-1R2 appears to be a pre- could overcome IL-1R2 blockade. Necrotic lysates made from requisite for physiological IL-1a secretion, as indicated by the LPS and ATP-stimulated macrophages displayed processing fact that expression of the noncleavable IL-1R2 mutant reduced of p33, which could be prevented by the caspase inhibitor IL-1a release (Figure 6H). Therefore, after exposure to stimuli Z-VAD-fmk or calpeptin (Figure 6A). Furthermore, after LPS that activate inflammasomes, intracellular IL-1R2 is cleaved by and ATP treatment, macrophage necrotic lysates induced IL-2 caspase-1, causing it to dissociate from IL-1a. Subsequently, release from EL4 cells, which was blocked with Z-VAD-fmk or during necrosis calpain can now cleave IL-1a to the fully active Immunity 38, 285–295, February 21, 2013 ª2013 Elsevier Inc. 291 Immunity IL-1R2 Controls IL-1a Activity Postnecrosis form, which allows the generation of robust IL-1a-dependent in both wild-type and Il1r1 mice (representing the net effect of responses. all other DAMPS, independent of IL-1), necrotic cells without IL-1R2 induced a 3-fold greater response, which was all IL-1 DISCUSSION dependent. This key observation implies that without IL-1R2 expression, the single most powerful DAMP within necrotic Although an immune response to necrosis may resolve the orig- cells is IL-1a. inal pathology and initiate repair, recruited leukocytes can Our studies also identify multiple mechanisms that could give damage the surrounding tissue. This sterile inflammatory a tissue- and cell type-dependent response to necrosis. Given response can also lead to unwanted activation of adaptive that p33 has some activity, dependent on its concentration immunity. Many human diseases are driven by activation and sensitivity of the responding cells, extensive necrosis in an of these pathways, including ischemia-reperfusion injury, IL-1R1-rich tissue (e.g., liver) could activate immunity regardless Alzheimer’s disease, atherosclerosis, and toxic insults to liver of IL-1a cleavage. Conversely, minimal necrosis in an IL-1R1-low and lung (Chen and Nun˜ ez, 2010; Kono and Rock, 2008; Stewart tissue (e.g., kidney) that cannot cleave p33 may not respond et al., 2010). Therefore, tight regulation of both release and to IL-1a at all. If necrosis occurs with a stimulus that activates activity of DAMPs is critical for host defense. caspase-1 (e.g., infection), IL-1R2 blockade is abrogated and The current understanding is that p33 is fully active and there- cleaved fully active p17 can signal to any IL-1R1-expressing fore requires only cell lysis to signal after necrosis. In contrast, cell. Necrosis of cells that can cleave IL-1a within IL-1R1-dense we find that activity is kept under tight control at multiple levels tissues would give extensive immune activation (e.g., vascular tissue). Where the amount of IL-1R2, IL-1R1, and necrosis in a cell type-dependent and conditional manner. Blocking preclude IL-1a activity, the response to necrosis could be calpain processing of endogenous p33 during necrosis signifi- cantly decreases activity, and comparison of recombinant deferred to other DAMPs and specialist immune cells, perhaps proteins demonstrates up to 50-fold increased activity of p17 giving more control over reactions. Additionally, activation of compared to p33. This difference occurs at the receptor, with IL-1a during necrosis could act as a ‘‘tipping point’’ that drives a 50-fold greater affinity of p17 for IL-1R1. IL-1a cleavage an inflammatory response toward adaptive immunity. Indeed, and activity after necrosis is cell type dependent because of this is the case in graft-versus-host disease, where injured expression of a cytosolic IL-1R2, which binds and protects endothelial cell-derived IL-1a induces intimal T cell recruitment IL-1a from cleavage and prevents IL-1a activity. IL-1R2 also pro- and IL-17 production, driving human artery allograft rejection tected IL-1a from granzyme B, chymase, and elastase cleavage, (Rao et al., 2007, 2008). suggesting that the primary control over IL-1a activity postnec- Necrosis is induced by many stimuli including physical trauma, rosis is by IL-1R2 binding. Lastly, active caspase-1 specifically chemical stress, and bacterial toxins and is characterized by cleaves IL-1R2, causing dissociation from IL-1a, calpain pro- loss of plasma membrane integrity (Kono and Rock, 2008). 2+ cessing, and complete restoration of IL-1a activity after necrosis Calpains are activated by Ca entry after loss of membrane or during regulated secretion. integrity (Wang, 2000), and in VSMCs p33 cleavage occurs DAMPs need to be retained on apoptosis, released on immediately upon necrosis. Granzyme B is proposed to activate necrosis, and able to activate the immune system. IL-1a is IL-1a extracellularly (Afonina et al., 2011); but given its ineffi- near universally expressed (Dinarello, 2009), lacks a signal ciency to process IL-1a relative to calpain (Afonina et al., peptide, is rarely extracellular under normal physiology, and 2011), cleavage within the necrotic body by calpain pro- also associates with chromatin during apoptosis, resulting in bably predominates when IL-1R2 is absent. IL-1a and IL-1b retention (Cohen et al., 2010). Other identified DAMPS include arose from ancient gene duplications (Dinarello, 2009) but have heat shock proteins, uric acid, HMGB1, ATP, and dsDNA diverged molecularly. IL-1a is a DAMP that is activated by (Kono and Rock, 2008). These DAMPs utilize specific cognate a protease intrinsically linked to necrosis, whereas IL-1b is acti- receptors, such as RAGE, or pattern recognition receptors, vated in response to PAMPs by a protease family intrinsically such as TLRs, which have a limited expression pattern that linked to apoptosis and inflammation. However, these current data indicate that in many cells IL-1a function requires both restricts cell types able to respond. In contrast, IL-1R1 is widely calpain and caspase-1. Secretion of IL-1a from macrophages expressed throughout many lineages (Dinarello, 2009), enabling IL-1a to act as a ‘‘universal’’ DAMP. Although this may provide (Fettelschoss et al., 2011) or dendritic cells (DCs) (Gross et al., more efficient detection of necrosis, it could predispose the 2012) requires inflammasome activation, but given that IL-1a is immune system to overreact, perhaps explaining the extensive not a caspase substrate, this has been puzzling (Keller et al., control over IL-1a activity via release, IL-1R2 binding, and 2008). We suggest that IL-1R2 is the likely target of caspase-1 calpain cleavage. This cell type and conditional dependency during physiological IL-1a release. Indeed, expression of an for IL-1a activity after necrosis perhaps suggests an immune uncleavable IL-1R2 mutant reduces IL-1a release, whereas advantage in allowing only some cell types to utilize IL-1a as a Casp1 mice have long been known to be deficient in IL-1a DAMP. Previous studies concluded that IL-1a is only a secondary secretion (Kuida et al., 1995; Li et al., 1995). signaling molecule released by macrophages during sterile An important finding of this work is that IL-1R2 binds p33 in inflammation (Chen et al., 2007; Kono et al., 2010). However, the cytosol. Although the previously described splice variant of this finding might have occurred due to the use of necrotic cell IL-1R2 (Liu et al., 1996) loses the transmembrane domain, it still types that express IL-1R2 and therefore cannot cleave and utilize codes for a signal peptide and should be secreted. Despite this, IL-1a (Y.Z., M.H., and M.C.H.C.; data not shown). Indeed, our data show that a large pool of IL-R2 exists in the cytosol. One whereas control necrotic lysates induce an equivalent peritonitis explanation for this is that the IL-1R2 signal peptide is short 292 Immunity 38, 285–295, February 21, 2013 ª2013 Elsevier Inc. Immunity IL-1R2 Controls IL-1a Activity Postnecrosis overnight hypoxic exposure. To activate inflammasomes, cells were treated with (13 amino acids) and is predicted to be relatively weak (Signal P; LPS (1 mg/ml; 4 hr), followed by ATP (5 mM) or Nigericin (20 mM) for 30 min. Y-max = 0.316) compared to other secreted proteins (e.g., IL-8; Calpain activity was determined with Calpain-Glo (Promega). VSMCs, HEK, Y-max = 0.790), and many proteins with signal peptides can and THP-1 cells were transfected with pcDNA3 (Invitrogen) with nucleofection be found in the cytosol (Davis et al., 2006). Indeed, multiple (Amaxa) or FugeneHD (Promega). factors determine how much of a signal peptide-containing protein ends up nontargeted, including the signal peptide and Protein Expression and Purification flanking amino acid sequences and cell type and growth Human p33 (1-271) or p17 (119-271) was cloned into pET15b (Novagen) or condition effects (Levine et al., 2005). Clearly further work is pGEX-4T-3 (GE). Human IL-1R2 (1-296) was cloned into pGEX-4T-3. For His needed to elucidate the specific form of IL-1R2 that binds p33, purification, IPTG-induced cultures were lysed in BugBuster (Novagen) with the mechanisms that generate it, and how this is controlled. benzonase, lysozyme, and protease inhibitors (10 min, RT). Urea (6 M) was 2+ added and incubated (10 min) before clarification and purification on a Ni Cell type-specific effects of IL-1R2 expression may occur in column. Columns were washed (40 mM imidazole) and eluted (250 mM imid- a number of diseases. Induction of antitumor immunity in vivo azole). Concentrated protein (Vivaspin) was dialyzed against decreasing urea is dependent upon the mode of cell death, whereby necrosis (4 M, 2 M, 0 M) in 10 mM Tris (pH 8.0), 50 mM NaCl. For GST purification, is more immunogenic than apoptosis (Melcher et al., 1998) IPTG-induced cultures were lysed in 50 mM Na HPO (pH 7.5), 150 mM 2 4 and local IL-1 activity (Ghiringhelli et al., 2009; Michaud NaCl, 1 mM DTT, 1 mM EDTA, with benzonase, lysozyme, and protease inhib- et al., 2011). Necrosis can induce maturation of DCs and expres- itors (30 min, RT), clarified, applied to glutathione agarose (QIAGEN), washed, eluted (50 mM reduced glutathione), concentrated, and dialyzed against sion of costimulatory molecules and can stimulate T cells 10 mM Tris (pH 8.0), 50 mM NaCl. GST tag was removed with biotinylated (Basu et al., 2000; Sauter et al., 2000)—all processes critical thrombin (Novagen). Protein concentration was determined (660 nm; Thermo for antitumor immunity and all known activities of IL-1. Indeed, Scientific), checked by Coomassie staining, and if necessary adjusted and differences are reported between the necrotic cell type and rechecked. Proteins were stored in 10% glycerol (80 C). ability to induce DC maturation (Sauter et al., 2000), perhaps reflecting IL-1R2 expression and thus ability of necrotic cell- Protease Cleavage derived IL-1a to modulate antigen-presenting cell function. His- or GST-p33 was incubated in 10 mM Tris (pH 7.5), 150 mM NaCl, 1 mM Intriguingly, IL-1R2 is upregulated in some tumors including DTT, 2 mM CaCl , with calpain (Calbiochem) ± calpeptin or IL-1R2 (RT). pancreatic ductal adenocarcinoma (Ru¨ ckert et al., 2010) and IL-1R2 was incubated with active caspase-1, caspase-3, caspase-4, or ovarian cancer, where it provides a powerful distinction between caspase-5 (Promokine) in 50 mM HEPES (pH 7.2), 50 mM NaCl, 0.1% CHAPS, 10 mM EDTA, 5% glycerol, and 10 mM DTT (37 C). Jurkat lysates were primary and recurrent tumors (Laios et al., 2008). Finally, athero- also incubated with granzyme B (100 nM; Cambridge Bioscience), chymase, sclerosis is highly dependent on IL-1 (Chamberlain et al., 2009; or elastase (100 nM; both Enzo). Chi et al., 2004; Duewell et al., 2010), although recent work casts doubt on IL-1b’s role (Menu et al., 2011), suggesting that IL-1a Cytokine Release Assay could be the major ligand affecting atherosclerotic plaques. VSMCs were plated, adhered overnight, and incubated in SF DMEM (24 hr). Indeed, necrotic VSMC-released IL-1a is a potent inducer of Fresh SF DMEM and treatments were added and incubated (6 hr). EL4 cells local vessel inflammation in vivo (Clarke et al., 2010), and given were washed and plated in SF DMEM with treatments and incubated (24 hr). that VSMCs express high amounts of IL-1a and IL-1R1 but little 3 4 17.5 3 10 necrotic VSMCs or macrophages or 5.25 3 10 necrotic Jurkat IL-1R2, their death is highly inflammatory. or HeLa cells were used per 500 ml. Supernatants were clarified and cytokines In summary, in addition to acting as a decoy receptor, IL-1R2 assayed by ELISA (PeproTech) or Cytomix (eBioscience). also plays a hitherto unreported role that fundamentally controls IL-1a activity postnecrosis. Cells expressing intracellular IL-1R2 Immunoblotting, Edman Degradation, and Co-IP 5 5 release little active IL-1a after necrosis without prior activation of Whole cell or necrotic lysates from 1 3 10 VSMCs or macrophages, or 3 3 10 Jurkat, HeLa, or T cells were loaded per lane. Antibodies used were IL-1a caspase-1, which processes IL-1R2. In contrast, necrotic cells (PeproTech), His (GE), a-spectrin (Millipore), b-actin (Sigma), calreticulin (Cell devoid of IL-1R2 are inherently powerful inducers of sterile Signaling), HDAC1 (Santa Cruz), and IL-1R2 (R&D). Caspase-1-cleaved IL-1R2 inflammation able to fully activate IL-1a upon calpain cleavage. was separated, electroblotted to PVDF, and stained and bands were excised Thus, changes in intracellular IL-1R2 expression may underlie for Edman degradation (ABI Procise 494HT). For co-IP, transfected HEK cells and modulate many chronic inflammatory diseases or other were incubated on ice (50 mM Tris [pH 8], 150 mM NaCl, 1% Triton X-100, pathologies involving cellular necrosis. protease inhibitors), freeze thawed, clarified, and incubated with 2 mg of anti-His or anti-GST (GE) (16 hr, 4 C). Immunocomplexes were precipitated with magnetic EXPERIMENTAL PROCEDURES protein-G beads (Dynal) and beads washed before elution with Laemmli buffer. All materials are from Sigma-Aldrich unless otherwise stated. RT-PCR RNA was extracted with TRI reagent, DNase treated (Ambion), and reverse Cell Culture transcribed (Promega) before PCR with the following primers: IL-1R1, AGGA VSMCs, EL4, HEK, and HeLa cells were cultured in DMEM and Jurkat and THP-1 GACGGAGGACTTGTGT and GCGTCATAGGTCTTTCCATC; total IL-1R2, cells in RPMI 1640, all supplemented with penicillin, streptomycin, L-glutamine, CATTACAAGCGGGAGTTCAG and TAGTGCAGACGTAGGTGCCA; soluble and 10% FCS. Human monocyte-derived macrophages were differentiated as IL-1R2, TGGCACCTACGTCTGCACTA and TGTCTCCAAAAGGAAGAGCGA; described previously (Brown et al., 2000). Cells were treated as indicated with GAPDH, TGTTGCCATCAATGACCCCTT and CTCCACGACTGACTCAGCG. Calpeptin (30 mM), Lactacystin (10 mM; both Biomol), IL-1a pAb (1 mg/ml), IL-1a and IL-1b (all PeproTech), IL-1RA (Amgen), IL-1R2 (250 ng; R&D), Z-YVAD, Z-VAD-fmk (both 10 mm; Bachem), LPS (1 mg/ml), EGTA (5 mM), and siRNA Silencing BB-94 (1 mg/ml; Tocris). Cells in serum-free (SF) DMEM were disrupted by freeze IL-1R2 silencing was performed with SMARTpool siRNA and controls thaw in liquid N , clarified, and stored at 80 C. Cells were also made necrotic (Dharmacon). In brief, HeLa cells were transfected with 20 nM of siRNA via by incubation with 7-BIO (25 mM; Enzo) or digitonin (0.1%) (data not shown) or by HiPerFect (QIAGEN), retransfected after 48 hr, and harvested 48 hr later. Immunity 38, 285–295, February 21, 2013 ª2013 Elsevier Inc. 293 Immunity IL-1R2 Controls IL-1a Activity Postnecrosis Immunofluorescence Chamberlain, J., Francis, S., Brookes, Z., Shaw, G., Graham, D., Alp, N.J., For IL-1R2, cells were fixed in 2% formaldehyde and permeabilized with 0.5% Dower, S., and Crossman, D.C. (2009). Interleukin-1 regulates multiple athero- NP-40. For IL-1R2 and p33, cells were fixed in 2% formaldehyde, washed, genic mechanisms in response to fat feeding. PLoS ONE 4, e5073. and fixed with methanol. All were blocked in 1% BSA, incubated with IL-1R2 Chen, G.Y., and Nunez, G. (2010). Sterile inflammation: sensing and reacting to mAb (R&D) and anti-N-terminal IL-1a (Aviva Systems Biology), washed, damage. Nat. Rev. Immunol. 10, 826–837. incubated with Alexa Fluor antibodies (Molecular Probes), washed, and Chen, C.J., Kono, H., Golenbock, D., Reed, G., Akira, S., and Rock, K.L. (2007). imaged on a BX51 (Olympus) or a TCS SP2 AOBS (Leica) microscope. PLA Identification of a key pathway required for the sterile inflammatory response assays with the primary antibodies as above were conducted according triggered by dying cells. Nat. Med. 13, 851–856. to the manufacturer (Olink). For flow cytometry, cells were fixed in 2% formaldehyde, washed, blocked in 1% BSA and 0.05% sodium azide, stained Chi, H., Messas, E., Levine, R.A., Graves, D.T., and Amar, S. (2004). Interleukin-1 receptor signaling mediates atherosclerosis associated with with anti-IL-1R2, washed, stained with secondary antibodies, washed, and analyzed on a C6 flow cytometer (Accuri). Intracellular staining steps had bacterial exposure and/or a high-fat diet in a murine apolipoprotein E hetero- zygote model: pharmacotherapeutic implications. Circulation 110, 1678–1685. 0.5% saponin present. Clarke, M.C., Talib, S., Figg, N.L., and Bennett, M.R. (2010). Vascular smooth muscle cell apoptosis induces interleukin-1-directed inflammation: effects of Animal Protocols hyperlipidemia-mediated inhibition of phagocytosis. Circ. Res. 106, 363–372. Experiments were conducted under UK Home Office licensing. Mice were in- jected intraperitoneally with saline, 0.29 fmol/g body weight of p17 or p33, or Cohen, I., Rider, P., Carmi, Y., Braiman, A., Dotan, S., White, M.R., Voronov, E., 8.3 3 10 necrotic HeLa cells. After 6 hr, the peritoneal cavity was lavaged with Martin, M.U., Dinarello, C.A., and Apte, R.N. (2010). Differential release of chro- 5 ml of PBS. GR-1 cells were counted by FACS after anti-GR-1-FITC staining matin-bound IL-1alpha discriminates between necrotic and apoptotic cell (Biolegend). death by the ability to induce sterile inflammation. Proc. Natl. Acad. Sci. USA 107, 2574–2579. Statistics Colotta, F., Re, F., Muzio, M., Bertini, R., Polentarutti, N., Sironi, M., Giri, J.G., Parametric tests were employed for analysis of continuous ELISA and Dower, S.K., Sims, J.E., and Mantovani, A. (1993). Interleukin-1 type II peritonitis data, conducted with a one-way, two-tailed ANOVA (Excel). receptor: a decoy target for IL-1 that is regulated by IL-4. Science 261, 472–475. Davis, M.J., Hanson, K.A., Clark, F., Fink, J.L., Zhang, F., Kasukawa, T., Kai, SUPPLEMENTAL INFORMATION C., Kawai, J., Carninci, P., Hayashizaki, Y., and Teasdale, R.D. (2006). Differential use of signal peptides and membrane domains is a common occur- Supplemental Information includes Supplemental Experimental Procedures rence in the protein output of transcriptional units. PLoS Genet. 2, e46. and five figures and can be found with this article online at http://dx.doi.org/ 10.1016/j.immuni.2013.01.008. Dinarello, C.A. (1996). Biologic basis for interleukin-1 in disease. Blood 87, 2095–2147. ACKNOWLEDGMENTS Dinarello, C.A. (2009). Immunological and inflammatory functions of the inter- leukin-1 family. Annu. Rev. Immunol. 27, 519–550. We gratefully acknowledge monocyte donations from E. Chilvers’ laboratory Duewell, P., Kono, H., Rayner, K.J., Sirois, C.M., Vladimer, G., Bauernfeind, and J. Skepper for help with confocal. This study was supported by British F.G., Abela, G.S., Franchi, L., Nun˜ ez, G., Schnurr, M., et al. (2010). NLRP3 in- Heart Foundation Grants FS/09/005/26845 (M.C.H.C.), PG/06/024/20354, flammasomes are required for atherogenesis and activated by cholesterol and RG 04/001 (M.R.B.), and NIHR Cambridge BRC. crystals. Nature 464, 1357–1361. Eigenbrod, T., Park, J.H., Harder, J., Iwakura, Y., and Nu´ n˜ ez, G. (2008). Cutting Received: April 11, 2012 edge: critical role for mesothelial cells in necrosis-induced inflammation Accepted: January 18, 2013 through the recognition of IL-1 alpha released from dying cells. J. Immunol. Published: February 7, 2013 181, 8194–8198. Fettelschoss, A., Kistowska, M., LeibundGut-Landmann, S., Beer, H.D., REFERENCES Johansen, P., Senti, G., Contassot, E., Bachmann, M.F., French, L.E., Oxenius, A., and Ku¨ ndig, T.M. (2011). Inflammasome activation and IL-1b Afonina, I.S., Tynan, G.A., Logue, S.E., Cullen, S.P., Bots, M., Lu¨ thi, A.U., target IL-1a for secretion as opposed to surface expression. Proc. Natl. Reeves, E.P., McElvaney, N.G., Medema, J.P., Lavelle, E.C., and Martin, Acad. Sci. USA 108, 18055–18060. S.J. (2011). Granzyme B-dependent proteolysis acts as a switch to enhance the proinflammatory activity of IL-1a. Mol. Cell 44, 265–278. Ghiringhelli, F., Apetoh, L., Tesniere, A., Aymeric, L., Ma, Y., Ortiz, C., Vermaelen, K., Panaretakis, T., Mignot, G., Ullrich, E., et al. (2009). Basu, S., Binder, R.J., Suto, R., Anderson, K.M., and Srivastava, P.K. (2000). Activation of the NLRP3 inflammasome in dendritic cells induces IL-1beta- Necrotic but not apoptotic cell death releases heat shock proteins, which dependent adaptive immunity against tumors. Nat. Med. 15, 1170–1178. deliver a partial maturation signal to dendritic cells and activate the NF-kappa B pathway. Int. Immunol. 12, 1539–1546. Gross, O., Yazdi, A.S., Thomas, C.J., Masin, M., Heinz, L.X., Guarda, G., Quadroni, M., Drexler, S.K., and Tschopp, J. (2012). Inflammasome activators Beck, G., and Habicht, G.S. (1986). Isolation and characterization of a primitive induce interleukin-1a secretion via distinct pathways with differential require- interleukin-1-like protein from an invertebrate, Asterias forbesi. Proc. Natl. ment for the protease function of caspase-1. Immunity 36, 388–400. Acad. Sci. USA 83, 7429–7433. Kamari, Y., Werman-Venkert, R., Shaish, A., Werman, A., Harari, A., Gonen, A., Brough, D., Le Feuvre, R.A., Wheeler, R.D., Solovyova, N., Hilfiker, S., Voronov, E., Grosskopf, I., Sharabi, Y., Grossman, E., et al. (2007). Differential Rothwell, N.J., and Verkhratsky, A. (2003). Ca2+ stores and Ca2+ entry differ- role and tissue specificity of interleukin-1alpha gene expression in atherogen- entially contribute to the release of IL-1 beta and IL-1 alpha from murine esis and lipid metabolism. Atherosclerosis 195, 31–38. macrophages. J. Immunol. 170, 3029–3036. Brown, S.B., Clarke, M.C., Magowan, L., Sanderson, H., and Savill, J. (2000). Keller, M., Ru¨ egg, A., Werner, S., and Beer, H.D. (2008). Active caspase-1 is a regulator of unconventional protein secretion. Cell 132, 818–831. Constitutive death of platelets leading to scavenger receptor-mediated phagocytosis. A caspase-independent cell clearance program. J. Biol. Kobayashi, Y., Yamamoto, K., Saido, T., Kawasaki, H., Oppenheim, J.J., and Chem. 275, 5987–5996. Matsushima, K. (1990). Identification of calcium-activated neutral protease as a processing enzyme of human interleukin 1 alpha. Proc. Natl. Acad. Sci. USA Cayrol, C., and Girard, J.P. (2009). The IL-1-like cytokine IL-33 is inactivated after maturation by caspase-1. Proc. Natl. Acad. Sci. USA 106, 9021–9026. 87, 5548–5552. 294 Immunity 38, 285–295, February 21, 2013 ª2013 Elsevier Inc. Immunity IL-1R2 Controls IL-1a Activity Postnecrosis Kono, H., and Rock, K.L. (2008). How dying cells alert the immune system to Orlando, S., Sironi, M., Bianchi, G., Drummond, A.H., Boraschi, D., Yabes, D., danger. Nat. Rev. Immunol. 8, 279–289. and Mantovani, A. (1997). Role of metalloproteases in the release of the IL-1 type II decoy receptor. J. Biol. Chem. 272, 31764–31769. Kono, H., Karmarkar, D., Iwakura, Y., and Rock, K.L. (2010). Identification of the cellular sensor that stimulates the inflammatory response to sterile cell Prudovsky, I., Mandinova, A., Soldi, R., Bagala, C., Graziani, I., Landriscina, M., Tarantini, F., Duarte, M., Bellum, S., Doherty, H., and Maciag, T. (2003). death. J. Immunol. 184, 4470–4478. The non-classical export routes: FGF1 and IL-1alpha point the way. J. Cell Kuida, K., Lippke, J.A., Ku, G., Harding, M.W., Livingston, D.J., Su, M.S., and Sci. 116, 4871–4881. Flavell, R.A. (1995). Altered cytokine export and apoptosis in mice deficient in Rajama¨ ki, K., Lappalainen, J., Oo¨ rni, K., Va¨ lima¨ ki, E., Matikainen, S., Kovanen, interleukin-1 beta converting enzyme. Science 267, 2000–2003. P.T., and Eklund, K.K. (2010). Cholesterol crystals activate the NLRP3 inflam- Laios, A., O’Toole, S.A., Flavin, R., Martin, C., Ring, M., Gleeson, N., D’Arcy, T., masome in human macrophages: a novel link between cholesterol metabolism McGuinness, E.P., Sheils, O., Sheppard, B.L., and O’ Leary, J.J. (2008). An and inflammation. PLoS ONE 5, e11765. integrative model for recurrence in ovarian cancer. Mol. Cancer 7,8. Rao, D.A., Tracey, K.J., and Pober, J.S. (2007). IL-1alpha and IL-1beta are Levine, C.G., Mitra, D., Sharma, A., Smith, C.L., and Hegde, R.S. (2005). The endogenous mediators linking cell injury to the adaptive alloimmune response. efficiency of protein compartmentalization into the secretory pathway. Mol. J. Immunol. 179, 6536–6546. Biol. Cell 16, 279–291. Rao,D.A.,Eid, R.E.,Qin,L.,Yi,T.,Kirkiles-Smith,N.C.,Tellides,G.,and Li, P., Allen, H., Banerjee, S., Franklin, S., Herzog, L., Johnston, C., McDowell, Pober, J.S. (2008). Interleukin (IL)-1 promotes allogeneic T cell intimal infiltra- J., Paskind, M., Rodman, L., Salfeld, J., et al. (1995). Mice deficient in IL-1 tion and IL-17 production in a model of human artery rejection. J. Exp. Med. beta-converting enzyme are defective in production of mature IL-1 beta and 205, 3145–3158. resistant to endotoxic shock. Cell 80, 401–411. Rock, K.L., Latz, E., Ontiveros, F., and Kono, H. (2010). The sterile inflamma- Liu, C., Hart, R.P., Liu, X.J., Clevenger, W., Maki, R.A., and De Souza, E.B. tory response. Annu. Rev. Immunol. 28, 321–342. (1996). Cloning and characterization of an alternatively processed human Ru¨ ckert, F., Dawelbait, G., Winter, C., Hartmann, A., Denz, A., Ammerpohl, O., type II interleukin-1 receptor mRNA. J. Biol. Chem. 271, 20965–20972. Schroeder, M., Schackert, H.K., Sipos, B., Klo¨ ppel, G., et al. (2010). Luheshi, N.M., Kovacs, K.J., Lopez-Castejon, G., Brough, D., and Denes, A. Examination of apoptosis signaling in pancreatic cancer by computational (2011). Interleukin-1a expression precedes IL-1b after ischemic brain injury signal transduction analysis. PLoS ONE 5, e12243. and is localised to areas of focal neuronal loss and penumbral tissues. Sakurai, T., He, G., Matsuzawa, A., Yu, G.Y., Maeda, S., Hardiman, G., and J. Neuroinflammation 8, 186. Karin, M. (2008). Hepatocyte necrosis induced by oxidative stress and IL-1 Lu¨ thi, A.U., Cullen, S.P., McNeela, E.A., Duriez, P.J., Afonina, I.S., Sheridan, alpha release mediate carcinogen-induced compensatory proliferation and C., Brumatti, G., Taylor, R.C., Kersse, K., Vandenabeele, P., et al. (2009). liver tumorigenesis. Cancer Cell 14, 156–165. Suppression of interleukin-33 bioactivity through proteolysis by apoptotic Sauter, B., Albert, M.L., Francisco, L., Larsson, M., Somersan, S., and caspases. Immunity 31, 84–98. Bhardwaj, N. (2000). Consequences of cell death: exposure to necrotic tumor March, C.J., Mosley, B., Larsen, A., Cerretti, D.P., Braedt, G., Price, V., Gillis, cells, but not primary tissue cells or apoptotic cells, induces the maturation of S., Henney, C.S., Kronheim, S.R., Grabstein, K., et al. (1985). Cloning, immunostimulatory dendritic cells. J. Exp. Med. 191, 423–434. sequence and expression of two distinct human interleukin-1 complementary Scaffidi, P., Misteli, T., and Bianchi, M.E. (2002). Release of chromatin protein DNAs. Nature 315, 641–647. HMGB1 by necrotic cells triggers inflammation. Nature 418, 191–195. Matzinger, P. (1994). Tolerance, danger, and the extended family. Annu. Rev. Sims, J.E., and Smith, D.E. (2010). The IL-1 family: regulators of immunity. Nat. Immunol. 12, 991–1045. Rev. Immunol. 10, 89–102. Melcher, A., Todryk, S., Hardwick, N., Ford, M., Jacobson, M., and Vile, R.G. Stewart, C.R., Stuart, L.M., Wilkinson, K., van Gils, J.M., Deng, J., Halle, A., (1998). Tumor immunogenicity is determined by the mechanism of cell death Rayner, K.J., Boyer, L., Zhong, R., Frazier, W.A., et al. (2010). CD36 ligands via induction of heat shock protein expression. Nat. Med. 4, 581–587. promote sterile inflammation through assembly of a Toll-like receptor 4 and 6 heterodimer. Nat. Immunol. 11, 155–161. Menu, P., Pellegrin, M., Aubert, J.F., Bouzourene, K., Tardivel, A., Mazzolai, L., and Tschopp, J. (2011). Atherosclerosis in ApoE-deficient mice progresses Tokunaga, H., Saito, S., Sakai, K., Yamaguchi, R., Katsuyama, I., Arakawa, T., independently of the NLRP3 inflammasome. Cell Death Dis. 2, e137. Onozaki, K., Arakawa, T., and Tokunaga, M. (2010). Halophilic beta-lactamase as a new solubility- and folding-enhancing tag protein: production of native Michaud, M., Martins, I., Sukkurwala, A.Q., Adjemian, S., Ma, Y., Pellegatti, P., human interleukin 1alpha and human neutrophil alpha-defensin. Appl. Shen, S., Kepp, O., Scoazec, M., Mignot, G., et al. (2011). Autophagy- Microbiol. Biotechnol. 86, 649–658. dependent anticancer immune responses induced by chemotherapeutic agents in mice. Science 334, 1573–1577. Wang, K.K. (2000). Calpain and caspase: can you tell the difference? Trends Neurosci. 23, 20–26. Mosley, B., Urdal, D.L., Prickett, K.S., Larsen, A., Cosman, D., Conlon, P.J., Gillis, S., and Dower, S.K. (1987). The interleukin-1 receptor binds the human Wang, D., Zhang, S., Li, L., Liu, X., Mei, K., and Wang, X. (2010). Structural interleukin-1 alpha precursor but not the interleukin-1 beta precursor. J. Biol. insights into the assembly and activation of IL-1b with its receptors. Nat. Chem. 262, 2941–2944. Immunol. 11, 905–911. Immunity 38, 285–295, February 21, 2013 ª2013 Elsevier Inc. 295

Journal

Immunity – Unpaywall

Published: Feb 1, 2013

There are no references for this article.