Mutations in cryopyrinBypassing roadblocks in the caspase 1 inflammasome for interleukin-1 secretion and disease activity.код для вставкиСкачать
ARTHRITIS & RHEUMATISM Vol. 56, No. 9, September 2007, pp 2817–2822 DOI 10.1002/art.22841 © 2007, American College of Rheumatology EDITORIAL Mutations in Cryopyrin: Bypassing Roadblocks in the Caspase 1 Inflammasome for Interleukin-1␤ Secretion and Disease Activity Charles A. Dinarello Unlike the release of tumor necrosis factor ␣ (TNF␣), there are several roadblocks to the release of interleukin-1␤ (IL-1␤), beginning with the transcription of the IL1B gene and ending with the exit of the active cytokine from the cell. There is also the additional roadblock of the high level of synthesis and secretion of the naturally occurring IL-1 receptor antagonist (IL1Ra), with its tight binding to the IL-1 receptor type I and an affinity higher than that of either IL-1␤ or IL-1␣. If these roadblocks are not enough to limit the activity of IL-1␤, there is nature’s decoy mechanism, which consists of the preferential binding of the cell surface IL-1 receptor type II to IL-1␤, preventing it from triggering the signal-transducing type I receptor. The soluble form of the type II receptor also prevents active IL-1␤ from engaging the signaling receptor by forming an inactive complex with the soluble IL-1 receptor accessory protein, which is constitutively produced in the liver. The facilitation of IL-1␤ processing by the caspase 1 inflammasome through ATP activation of the P2X7 receptor can also be viewed as a potential roadblock to the activity of IL-1␤. It seems clear that in the evolution of cytokines, some cytokines, such as IL-6 and IL-1Ra, are quickly driven from the cell by the Golgi machinery, whereas others encounter a series of roadblocks that limit their activity. The most revealing justifications for nature’s wisdom in placing several roadblocks to IL-1␤ activity are the consequences of the effects of IL-1␤ in the loss of insulin-producing pancreatic beta cells (1) and the manifestations of autoinflammatory syndromes, diseases that carry a portfolio of destructive inflammation (2–4). Familial Mediterranean fever (5) and hyper-IgD syndrome (6) are also classified as autoinflammatory diseases because of the poor control of the processing and secretion of IL-1␤. In this issue of Arthritis and Rheumatism (7), the study reported by Gattorno and coworkers offers important new lessons about these roadblocks. The investigators studied the release of IL-1␤ from primary blood monocytes obtained from patients with chronic infantile neurologic, cutaneous, articular (CINCA) syndrome (known in North America as neonatal-onset multisystem inflammatory disease [NOMID]) and from a patient with Muckle-Wells syndrome (MWS). These 2 syndromes, as well as familial cold-induced autoinflammatory syndrome (FCAS), are the primary examples of dysregulated processing and secretion of IL-1␤. The investigators also studied the effects of in vivo treatment with anakinra, a recombinant human IL-1Ra. Specifically, the study examined the secretion of IL-1␤ from peripheral blood monocytes obtained from the patients before and during treatment with anakinra. The study focused on the function of the IL-1␤ “inflammasome” (more appropriately termed the caspase 1 inflammasome) and the effect of mutations in cryopyrin, the protein of the NALP3 gene (8). The lessons learned from these studies provide new insights into the tight control of the processing and secretion of IL-1␤. While the syndromes studied are rare, the mechanisms revealed by these studies have broad implications for how evolution favored the restriction of biologic responses to IL-1␤ by placing several roadblocks to its egress from the cell. The clinical manifestations of these syndromes are characterized by recurrent fevers, neutrophilic leukocytosis, rashes, deforming arthritis, and high serum levels of acute-phase reactants. Progressive deafness, Supported by NIH grant AI-15614. Charles A. Dinarello, MD: University of Colorado Health Sciences Center, Denver. Dr. Dinarello has received consulting fees (less than $10,000) from Amgen and has given expert testimony for which he received a fee (less than $10,000) from Chiron. Address correspondence and reprint requests to Charles A. Dinarello, MD, University of Colorado Health Sciences Center, 4200 East Ninth Avenue B168, Denver, CO 80262. E-mail: email@example.com. Submitted for publication May 13, 2007; accepted in revised form May 18, 2007. 2817 2818 leptomeningitis, and amyloidosis may be present in some patients. These syndromes are part of a larger group of diseases that are called “autoinflammatory” to distinguish them from those considered to be “autoimmune,” although there is clearly overlap between the 2 groups. Both the adult-onset and the juvenile-onset forms of Still’s disease (systemic-onset juvenile idiopathic arthritis) fall into the autoinflammatory group. The best criterion for identifying an autoinflammatory disease is that the clinical, biochemical, and hematologic manifestations are rapidly and impressively reversed upon initiation of treatment with IL-1Ra, with the IL-1 trap, or with monoclonal anti–IL-1␤ antibodies. Hence, the culprit in these diseases is IL-1␤ and not IL-1␣; essentially, autoinflammatory diseases are diseases that occur as a result of poor control of the release of active IL-1␤. Compared with monocytes from healthy subjects, freshly isolated peripheral blood monocytes from patients with autoinflammatory diseases secrete more IL-1␤ when incubated in vitro. For example, peripheral blood mononuclear cells (PBMCs) from patients with NOMID secreted ⬎1 ng/ml more IL-1␤ during a 24hour incubation in the absence of any exogenous stimulus, as compared with ⬍50 pg/ml secreted by cells from healthy controls (2). When stimulated with endotoxin, PBMCs from NOMID patients secreted 3–4 times more IL-1␤ than PBMCs from controls. In the Gattorno study (7), monocytes from patients with CINCA syndrome secreted even greater amounts of IL-1␤ when stimulated with lipopolysaccharide as compared with monocytes from controls. Although patients with FCAS exhibit dramatic reversal of disease manifestations upon treatment with anakinra, elevated levels of IL-1␤ in the circulation following cold exposure is not observed (4). Rather, the circulating cytokine is IL-6, which is under the control of IL-1. These observations reinforce an important clinical lesson in cytokine-mediated diseases: causation is established with specific receptor blockade or specific cytokine neutralization, and not with elevated circulating levels. For example, many diseases are associated with elevated serum levels of IL-6 because this cytokine is readily secreted, but few are due to a pathologic role of IL-6. Although increased secretion of IL-1␤ by stimulated blood monocytes in vitro can be helpful in the diagnosis of an IL-1␤–mediated disease, a few days of treatment with anakinra is safe and will unambiguously reveal the role of IL-1 in the disease in question. Despite this rather “easy IL-1 test of causation,” in the report by Gattorno, there is a lesson about the control of active DINARELLO IL-1␤ secretion and this lesson may apply to other cytokines, such as IL-18 and IL-33. Although IL-18 likely contributes to several autoinflammatory diseases, macrophage activation syndrome appears to be a disease that is attributable to IL-18 (9). IL-33 may mediate mast cell–related and other T helper type 2 diseases. Both cytokines are members of the IL-1 family and are caspase 1 dependent for secretion of the active cytokine. The first lesson from the Gattorno study is that monocytes obtained from their patients secreted IL-1␤ without a second signal. What is meant by the expression “second signal”? There are 2 ways to study the secretion of IL-1␤ from fresh human blood monocytes. Most studies stimulate the cells with endotoxin (usually at concentrations of 1 ng/ml or lower) and, after 24 hours of incubation, measure the “mature” cytokine released into the supernatant, using a specific enzyme-linked immunosorbent assay. In the absence of cell leakage or cell death, starting after 4 hours and over the course of 24 hours, IL-1␤ is steadily released into the supernatant medium. The IL-1␤ that is released is called “mature” because it represents the “active” cytokine following cleavage of the inactive IL-1␤ precursor by caspase 1. As described below, activation of the intracellular cysteine protease caspase 1 is the primary function of the inflammasome. In general, most, if not all, Toll-like receptor (TLR) ligands, as well as microorganisms themselves, stimulate the secretion of IL-1 ␤ via a caspase 1–dependent processing of the IL-1␤ precursor. Like endotoxins from gram-negative bacteria (via TLR-4), microbial products from gram-positive bacteria (via TLR-2) will result in the progressive release of IL-1␤ from monocytes starting at 4 hours and continuing for the next 20–36 hours. The vast majority of studies on the secretion of IL-1␤ use a 24-hour stimulation period. Importantly, in animals or humans injected with endotoxins, the earliest detectable level of IL-1␤ in the circulation is at 3 hours, and peak levels are present between 4 and 8 hours. The second way to study the secretion of mature IL-1␤ is to accelerate the process by incubating the monocytes with endotoxin for a short period of time (3 hours) and then triggering the rapid release of mature IL-1␤ within 15 minutes. This experimental approach is called the “two-signal” method. As shown in Figure 1A, 3 hours of exposure to endotoxin provides the first signal and results in transcription and translation of the IL-1␤ precursor. A second signal results in the rapid release of mature IL-1␤. This so-called second signal is triggered by activation of the purinergic ion channel called P2X7 Figure 1. Steps in the activation of the caspase 1 inflammasome and the secretion of interleukin-1␤ (IL-1␤). A, Monocytes from unaffected, healthy controls. Activation of Toll-like receptor (TLR) (step 1) results in transcription (step 2) and translation of the IL-1␤ precursor (step 3). The IL-1␤ precursor remains diffusely in the cytosol (14). Upon activation of the P2X7 receptor (P2X7 R) (step 4), there is a rapid efflux of potassium from the cell (step 5a), resulting in a fall in intracellular potassium levels (step 5b). The fall in intracellular potassium levels triggers the assembly of the components of the caspase 1 inflammasome (step 6) and its association with procaspase 1. The caspase 1 inflammasone is composed of cryopyrin plus ASC (apoptosis-associated speck-like protein containing a caspase 1 recruitment domain [CARD]) and Cardinal (CARD inhibitor of NF-B–activating ligands). Cryoprin is a large protein with 4 domains: PYR (a pyrin domain), NACHT (a domain found in NAIP [neuronal apoptosis inhibitor protein], CIITA [class II major histocompatibility complex transcription activator], HET-E [incompatibility locus protein from Podospora anserina, a bacterial mucleotide triphosphatase protein], and TP-1 [telomerase-associated protein 1]), NAD (NALP-associated domain), and LRR (leucine-rich repeats). The processing of procaspase 1 results in the formation of the active caspase 1 heterodimer and the cleavage of the IL-1␤ precursor. The enzymatic processing of the IL-1␤ precursor by caspase 1 may take place in the cytosol (step 7a) or in the secretory lysosome (step 7b) or both. An influx of calcium into the cell (step 8), with an increase in intracellular calcium levels, provides a mechanism by which mature IL-1␤ is released from the cell (10,15) (step 9). Calcium influx activates 3 phospholipases; phosphatidylcholine-specific phospholipase C and calcium-dependent phospholipase A2 are required for secretion of IL-1␤, with exocytosis of the lysosomal contents (10). B, Monocytes from subjects with mutations in cryopyrin. TLR activation and the synthesis of the IL-1␤ precursor take place (steps 1, 2, and 3) as in A. TLR triggering may also result in the assembly of the inflammasome (step 4). However, there is constitutive activation of caspase 1 in monocytes from subjects with mutations in cryopyrin, and activation of the inflammasome by a fall in intracellular potassium is not required. Similar to monocytes from healthy controls, the processing of the IL-1␤ precursor by caspase 1 may take place in the cytosol (step 5a) or in the secretory lysosome (step 5b) or both. Also similar to monocytes from healthy controls, an influx of calcium into the cell (step 6), with an increase in intracellular calcium levels, provides a mechanism by which mature IL-1␤ is released from the cell (10,15) (step 7). TIR ⫽ Toll/interleukin-1 receptor. 2820 with millimolar concentrations of ATP. The activation of the P2X7 receptor by ATP results in the nearsimultaneous efflux of potassium from the cell (see Figure 1A). The fall in intracellular potassium levels provides the signal for the cleavage of the IL-1␤ precursor by caspase 1 (10). The cleavage may take place in cellular organelles called secretory lysosomes, but also possibly in the cytosol. The cleavage likely takes place near the cell membrane. Within 15 minutes of ATP activation of the P2X7 receptor, potassium levels fall, and mature IL-1␤ is released from the cell. Activation of the P2X7 receptor and the fall in intracellular K⫹ concentrations, with the subsequent events leading to IL-1␤ processing and secretion, likely also take place in the “long incubation” method, but at lower levels, resulting in lower secretion of IL-1␤. Naturally occurring peptides that activate the P2X7 receptor have been described (11). The expected finding in the study by Gattorno was that after 3 hours of exposure to endotoxin, monocytes from the CINCA syndrome or MWS patients secreted significantly more IL-1␤ than did those from the healthy controls (7). The unexpected finding in the study was that monocytes from the CINCA syndrome and MWS patients did not respond to ATP with a burst of IL-1␤ release. The rapid release of IL-1␤ by monocytes from healthy donors is triggered by a short exposure to ATP. Unlike monocytes from healthy donors, the release of larger amounts of IL-1␤ did not require the second signal from ATP. Thus, the ATP activation step (step 4 in Figure 1A) was not needed in monocytes from patients with the cryopyrin mutations. One interpretation is that the caspase 1 inflammasome is constitutively active in these cells during the 3-hour incubation period and does not require the ATP-driven efflux of potassium to cleave the precursor (Figure 1B). But, there was yet another unexpected finding. In monocytes from healthy controls, ATP activated the processing and the rapid secretion not only of mature IL-1␤, but also of caspase 1. However, caspase 1 release did not take place following the addition of ATP to cells from the patients. Hence, the 2 ATP-driven effects are not needed in monocytes from these patients. Procaspase 1 is present in resting monocytes, whereas synthesis of the IL-1␤ precursor requires TLR activation. Therefore, one would expect to detect spontaneous secretion of caspase 1 in monocytes from patients with a constitutively activated caspase 1 inflammasome. The observation that caspase 1 secretion by the patients’ monocytes occurred only after TLR triggering suggests DINARELLO that the caspase 1 inflammasome is partially, but not fully, active in cells from patients with autoinflammatory syndromes. Therefore, one can interpret the effect of mutations in cryopyrin as bypassing one of the roadblocks that control the exit of active IL-1␤ from the cell. Although all patients with CINCA syndrome exhibit the same spectrum of clinical, biochemical, and hematologic abnormalities, not all patients carry mutations in the CIAS1 gene. For example, in a study of 18 patients with NOMID (CINCA syndrome), only 67% carried the mutation (2). Nevertheless, monocytes from all of the patients released significantly more IL-1␤, whether resting or after stimulation with endotoxin for 24 hours, as compared with monocytes from healthy controls. The study by Gattorno and colleagues (7) went one step further and discovered that the failure to require ATP for the 3-hour release of IL-1␤ was only observed in the CINCA syndrome and MWS patients who had CIAS1 mutations. Monocytes from the CINCA syndrome patient without CIAS1 mutations responded to ATP, as did monocytes from healthy donors. We can assume that this differential response to ATP also existed in the cohort of 18 NOMID patients whose cells were studied only for the release of IL-1␤ over 24 hours (2). The Gattorno study draws attention to a unique property of the mutations, since one can now assign a molecular mechanism to the increase in IL-1␤ secretion relevant to the ATP-driven activation of caspase 1 by the complex of interacting proteins that comprise the caspase 1 inflammasome. It is a unique example of how a single mutation reveals a great deal of new information about a complex molecular mechanism. The assembly of proteins of the mutated inflammasome, although not requiring the fall in intracellular potassium, nevertheless still requires an initiating signal, and in the laboratory, endotoxin is used. In the patient, IL-1␤ itself likely functions as the initiator. The function of the caspase 1 inflammasome is primarily to convert inactive procaspase 1 into the active enzyme. In the rapid release method, this activation is triggered by the efflux of potassium and is rapidly followed by the appearance of mature IL-1␤ in the supernatant together with the processed caspase 1. Gattorno provides evidence that the mutation in cryopyrin results in activation of caspase 1 without requiring a sudden fall in the level of intracellular potassium in order to activate the inflammasome. The authors propose that mutated cryopyrin allows for the assembly of the complex of interacting proteins that comprise the inflammasome in the presence of normal EDITORIAL intracellular levels of potassium. This explanation would, in fact, provide a molecular mechanism for the ATP trigger in the wild-type cryopyrin. In the wild-type inflammasome, ATP activation of the P2X7 receptor opens the potassium channel, and as potassium levels fall, caspase 1 is simultaneously activated by the inflammasome. Thus, cryopyrin, in which any one of several mutations may occur, is the molecular target and results in the activation of caspase 1. The model provides an explanation for the consistent observation that monocytes from patients with CINCA/NOMID or MWS secrete more IL-1␤ than do those from healthy controls. Although studied using an exogenous stimulant such as endotoxin, monocytes from these patients more likely experience background stimulation from endogenous inflammation in vivo. In fact, as discussed below, the most likely endogenous stimulant is IL-1␤ itself. The model also supports the concept that the rate-limiting step in the secretion of IL-1␤ is activation of caspase 1. Any disease process that includes an increase in the steady-state levels of caspase 1, components of the inflammasome, or the IL-1␤ precursor carries the potential to be an “autoinflammatory” disease. Gattorno and colleagues report that in vivo therapy with anakinra in patients with CIAS1 mutations was associated with an in vitro reduction in the secretion of IL-1␤ from monocytes as compared with secretion before treatment. Monocytes from the patient without mutations, although secreting more IL-1␤ than those from controls, secreted the same amount of IL-1␤ before, as well as after, anakinra treatment. The reduction in the secretion of IL-1␤ following initiation of anakinra treatment was also reported by GoldbachMansky et al (2). What accounts for these observations, whether assessed by the long (24-hour) release method (2) or the short (3 hours plus 15 minutes) release method (7)? There is no magic about these observations. IL-1␤ stimulates both its own gene expression as well as the translation of the messenger RNA into the precursor. The original observations were made 20 years ago (12,13). The most likely explanation for a downregulation of secretion with IL-1 receptor blockade is that IL-1␤ stimulates the synthesis of caspase 1 and IL-1␤ precursor. In fact, in NOMID patients treated with anakinra, there was a down-regulation of both caspase 1 and the IL-1␤ precursor (2), suggesting that blocking the activity of IL-1␤ brings about a reduction in the further release of IL-1␤. Thus, the rate-limiting step is IL-1␤ activity itself. In support of this concept is the 2821 observation that a single dose of a neutralizing anti– IL-1␤ monoclonal antibody in patients with MWS resulted in a rapid amelioration of disease that lasted 6 months, but within 24 hours after the infusion of the antibody, steady-state levels of IL-1␤ messenger RNA from PBMCs were decreased (Lachmann H: personal communication). It would therefore be of considerable interest to activate the caspase 1 inflammasome in healthy donor monocytes stimulated with IL-1␣ for 3 hours and then activate the P2X7 receptor with ATP. Such a study would address a model closer to the clinical reality, since IL-1 itself, rather than endotoxin, is the challenge encountered by patients who have autoinflammatory diseases. REFERENCES 1. Larsen CM, Faulenbach M, Vaag A, Volund A, Ehses JA, Seifert B, et al. Interleukin-1–receptor antagonist in type 2 diabetes mellitus. N Engl J Med 2007;356:1517–26. 2. Goldbach-Mansky R, Dailey NJ, Canna SW, Gelabert A, Jones J, Rubin BI, et al. Neonatal-onset multisystem inflammatory disease responsive to interleukin-1␤ inhibition. N Engl J Med 2006;355: 581–92. 3. Hawkins PN, Lachmann HJ, Aganna E, McDermott MF. Spectrum of clinical features in Muckle-Wells syndrome and response to anakinra. Arthritis Rheum 2004;50:607–12. 4. Hoffman HM, Rosengren S, Boyle DL, Cho JY, Nayar J, Mueller JL, et al. Prevention of cold-associated acute inflammation in familial cold autoinflammatory syndrome by interleukin-1 receptor antagonist. Lancet 2004;364:1779–85. 5. Chae JJ, Komarow HD, Cheng J, Wood G, Raben N, Liu PP, et al. Targeted disruption of pyrin, the FMF protein, causes heightened sensitivity to endotoxin and a defect in macrophage apoptosis. Mol Cell 2003;11:591–604. 6. Simon A, van der Meer JW. Pathogenesis of familial periodic fever syndromes or hereditary autoinflammatory syndromes. Am J Physiol Regul Integr Comp Physiol 2007;292:R86–98. 7. Gattorno M, Tassi S, Carta S, Delfino L, Ferlito F, Pelagatti MA, et al. Pattern of interleukin-1␤ secretion in response to lipopolysaccharide and ATP before and after interleukin-1 blockade in patients with CIAS1 mutations. Arthritis Rheum 2007;56:3138–48. 8. Hoffman HM, Mueller JL, Broide DH, Wanderer AA, Kolodner RD. Mutation of a new gene encoding a putative pyrin-like protein causes familial cold autoinflammatory syndrome and MuckleWells syndrome. Nat Genet 2001;29:301–5. 9. Mazodier K, Marin V, Novick D, Farnarier C, Robitail S, Schleinitz N, et al. Severe imbalance of IL-18/IL-18BP in patients with secondary hemophagocytic syndrome. Blood 2005;106:3483–9. 10. Andrei C, Margiocco P, Poggi A, Lotti LV, Torrisi MR, Rubartelli A. Phospholipases C and A2 control lysosome-mediated IL-1␤ secretion: implications for inflammatory processes. Proc Natl Acad Sci U S A 2004;101:9745–50. 11. Elssner A, Duncan M, Gavrilin M, Wewers MD. A novel P2X7 receptor activator, the human cathelicidin-derived peptide LL37, induces IL-1␤ processing and release. J Immunol 2004;172: 4987–94. 12. Dinarello CA, Ikejima T, Warner SJ, Orencole SF, Lonnemann G, 2822 Cannon JG, et al. Interleukin 1 induces interleukin 1. I. Induction of circulating interleukin 1 in rabbits in vivo and in human mononuclear cells in vitro. J Immunol 1987;139:1902–10. 13. Warner SJ, Auger KR, Libby P. Interleukin-1 induces interleukin-1. II. Interleukin-1 induces production of interleukin-1 by adult human vascular endothelial cells in vitro. J Immunol 1987; 139:1911–7. DINARELLO 14. Andersson J, Bjork L, Dinarello CA, Towbin H, Andersson U. Lipopolysaccharide induces human interleukin-1 receptor antagonist and interleukin-1 production in the same cell. Eur J Immunol 1992;22:2617–23. 15. Kahlenberg JM, Dubyak GR. Mechanisms of caspase-1 activation by P2X7 receptor-mediated K⫹ release. Am J Physiol Cell Physiol 2004;286:C1100–8.