Therapeutic modulation of inflammasome pathways

1Our innate immune system orchestrates a first and rapid response against microbial pathogens and sterile damage. Innate immune cells respond to these threats by activation of non‐specific germline‐encoded pattern recognition receptors (PRRs). 1 Some PRRs directly bind evolutionary conserved moieties from pathogens that are commonly referred to as pathogen‐associated molecular patterns (PAMPs) or respond to danger‐associated molecular patterns (DAMPs) that are generated or dislocated upon cellular damage. Other PRRs indirectly guard against infections by detecting the activity of virulence factors used by a suite of pathogens to manipulate the host cell machinery in order to survive in a hostile intracellular environment. Toll‐like receptors (TLRs) and c‐type lectin receptors (CLRs) are examples of membrane‐bound PRRs that survey the extracellular environment and the endosomal compartment. Other PRRs are expressed intracellularly or face the cytosolic compartment, such as the nucleotide‐binding domain, leucine‐rich repeat‐containing proteins (NLRs), the retinoic acid‐inducible gene‐I (RIG‐I)‐like receptors (RLRs), absent in melanoma 2 (AIM2)‐like receptors (ALRs) and members of the tripartite motif (TRIM) receptors. 1

2Most PRRs engage signaling pathways that result in extensive transcriptional reprogramming of the cell, including upregulation of additional PRRs and production of pro‐inflammatory cytokines, chemokines and type I interferons that orchestrate the early inflammatory and host defense responses and help to shape downstream activation of adaptive immunity. 2 In contrast, a subset of NLRs (NLRP3, NLRC4, and NLRP1), ALRs (AIM2), and Pyrin (a TRIM family member) regulate innate immune responses at the post‐translational level by assembling inflammasome complexes. 3, 4 Although these inflammasome sensors differ in their expression patterns and the upstream signaling pathways that regulate their activation, they all oligomerize and recruit the adapter molecule apoptosis‐associated speck‐like protein containing a CARD (ASC) to seed a fibrillary structure named the “ASC speck” that recruits caspase‐1 through homotypic pyrin‐pyrin (PYD‐PYD) and CARD‐CARD interactions. 3, 4 These inflammasome complexes serve to facilitate proximity‐induced autoactivation of caspase‐1, which proceeds to cleave the pro‐inflammatory cytokines interleukin (IL)‐1β and IL‐18 into bioactive, secreted cytokines. Besides the amino‐terminal CARD domain by which it is recruited into inflammasomes, caspase‐1 consists of a carboxy‐terminal protease domain that harbors the evolutionary conserved catalytic residues in its p20 and p10 subunits. 5 Interestingly, naturally occurring genetic variants in human caspase‐1 that decrease its protease activity and the levels of IL‐1β production have been associated with the development of autoinflammatory disease in patients. 6 Recent studies in mice expressing a catalytically inactive caspase‐1‐C284A mutant suggest that inactive caspase‐1 may recruit receptor‐interacting protein kinase 2 (RIPK2) to enhance NF‐κB‐mediated IL‐6 and TNF‐α production. 7 Additionally, inflammasome activation promotes ASC‐ and caspase‐8‐mediated apoptosis in macrophages expressing catalytically inactive caspase‐1, suggesting that nature evolved mechanisms to safeguard inflammatory responses when caspase‐1 activation is compromised. 8

3Concomitant with cleavage of IL‐1β and IL‐18, caspase‐1 activation results in a lytic regulated cell death mode, termed pyroptosis, through proteolytic activation of gasdermin D (GSDMD). 9, 10, 11 Upon cleavage, the pore‐forming amino‐terminal domain of GSDMD inserts into the plasma membrane, where it assembles into higher order oligomers that perforate the plasma membrane and eventually result in osmotic swelling and cell rupture. 12, 13, 14 GSDMD‐mediated cell lysis promotes the extracellular release of IL‐1β and IL‐18 from macrophages along with other alarmins such as IL‐1α and high mobility group protein B1 (HMGB1) that contribute to shaping innate immune and inflammatory responses. 9, 10, 11, 15, 16, 17, 18

4In this review, we will briefly introduce the different inflammasome pathways followed by a discussion of IL‐1‐targeted biologics that are currently approved for the treatment of specific autoinflammatory diseases. Finally, we will provide an overview of experimental biologics and small molecule inflammasome agonist and antagonist therapies that are under development for modulating inflammasome activation in inflammatory and malignant diseases.

5The first member of the NLR family shown to form an inflammasome is NLRP1. 19 Humans encode a single NLRP1 gene, whereas mice carry three paralogous genes named Nlrp1a, Nlrp1b, and Nlrp1c. The latter is considered a pseudogene, whereas both NLRP1a and NLRP1b assemble inflammasomes. 20, 21 Bacillus anthracis lethal toxin (LeTx) specifically activates the NLRP1b inflammasome, while agents that selectively engage the NLRP1a inflammasome remain to be discovered. The cytosolic postproline dipeptidyl peptidases (DPP)8 and DPP9 suppress activation of human NLRP1 and its murine orthologs through mechanisms that are incompletely understood. In agreement, boronic acid peptides such as Val‐boroPro (also known as Talabostat) that inhibit DPP8/9 as well as several other DPP family members, and the DDP8/9‐selective inhibitor 1G244 activate human NLRP1 and the murine NLRP1b inflammasomes. 22, 23, 24, 25 Activation of human NLRP1 and murine NLRP1b is thought to involve proteasomal degradation of the auto‐inhibitory amino‐terminal region, which enables the recruitment of ASC and caspase‐1 to their carboxy‐terminal UPA‐CARD‐containing regions (Figure 1). 26, 27

6A homozygous NLRP1 gain‐of‐function mutation was recently reported to cause an autosomal recessive form of juvenile‐onset recurrent respiratory papillomatosis in two siblings with mild dermatologic abnormalities. 28 However, dominantly inherited gain‐of‐function mutations in human NLRP1 predispose patients to the development of Mendelian diseases characterized by skin inflammation and dyskeratosis named multiple self‐healing palmoplantar carcinoma (MSPC), NLRP1‐associated autoinflammation with arthritis and dyskeratosis (NAIAD), and autoinflammation with arthritis and dyskeratosis (AIADK). 29, 30

7Due to its broad role in anti‐microbial immunity and sterile inflammation, NLRP3 is subjected to multilayered regulation at the transcriptional, post‐translational, and activation levels. The protein has a tripartite domain architecture composed of an amino‐terminal PYD, a central NACHT domain, and a C‐terminal region containing leucine‐rich repeat (LRR) motifs. The NACHT domain features an (d)ATP binding pocket with extended Walker A/B motifs that supports ATP hydrolysis, a prerequisite for activation of the NLRP3 inflammasome. 31 In naive macrophages, NLRP3 expression levels are insufficient to support inflammasome activation, and NLRP3 activation therefore first requires a transcriptional priming signal (signal 1) to increase NLRP3 protein expression levels (Figure 1). 32 Priming may additionally involve post‐translational modifications such as phosphorylation and deubiquitination to license rapid subsequent NLRP3 activation by signal 2 agents. 33, 34, 35, 36, 37 Moreover, both the chaperone Hsp90 and the cell cycle‐regulated kinase NIMA‐related kinase 7 (NEK7) physically interact with and regulate NLRP3 activation, although TAK1‐dependent post‐translational priming may also license NEK7‐independent NLRP3 activation in human macrophages. 38, 39, 40, 41, 42

8Signal 2 agents are chemically and structurally diverse, ranging from extracellular ATP, microbial pore‐forming protein toxins, and organic ionophores to fine particulate matter such as silica and uric acid crystals. It is therefore thought that NLRP3 senses these agents indirectly through their effect on disruption of the plasma membrane, and the concomitant efflux of intracellular potassium levels. 43, 44, 45 NLRP3‐activating agents such as crystalline monosodium urate (MSU), calcium pyrophosphate dehydrate and cholesterol, and protein aggregates like amyloid‐β and islet amyloid polypeptide (IAPP) are implicated in the development of inflammatory pathology in patients. Correspondingly, a rich body of preclinical studies showed that Nlrp3 deficiency in mice protects against inflammatory pathology in models of gout, pseudogout, atherosclerosis, inflammatory arthritis, non‐alcoholic steatohepatitis (NASH), Alzheimer's disease (AD), and the experimental autoimmune encephalomyelitis (EAE) model of multiple sclerosis. 46, 47, 48, 49, 50, 51, 52, 53, 54 Of note, NLRP3 inflammasome activation by influenza A virus (IAV) infection is regulated by the Z‐DNA binding protein 1 (ZBP1)‐containing PANoptosome, a recently identified multi‐protein complex that senses IAV and controls induction of pyroptosis, apoptosis, and necroptosis in IAV‐infected cells. 55 TAK inhibition and a host of additional microbial pathogens also induce NLRP3 inflammasome‐mediated pyroptosis concomitant with induction of necroptosis and extrinsic apoptosis in the targeted cell population, a process that was recently coined as PANoptosis. 56, 57, 58, 59

9Moreover, gain‐of‐function mutations in NLRP3 cause autosomal dominantly inherited autoinflammatory diseases that are collectively named Cryopyrin‐Associated Periodic Syndrome (CAPS), and which comprise—in increasing order of clinical severity—familial cold‐induced autoinflammatory syndrome (FCAS), Muckle‐Wells syndrome (MWS), and neonatal‐onset multisystem inflammatory disease (NOMID) (also known as chronic infantile neurologic, cutaneous, articular syndrome (CINCA)). Mechanistic studies in CAPS mouse models and clinical experience with IL‐1‐targeted therapies both suggest that excessive IL‐1β production is the key mediator of systemic inflammatory pathology in CAPS, with evidence for additional roles of IL‐18 and pyroptosis. 18, 60, 61

10The NLRC4 inflammasome assembles upon physical binding of bacterial flagellin and type III secretion systems (T3SS) components by NLR family apoptosis inhibitory proteins (NAIPs). Human NAIP recognizes the T3SS needle protein, whereas murine NAIP5 and NAIP6 detect flagellin, and NAIP1 and NAIP2 detect needle and rod proteins of T3SS, respectively (Figure 1). 62, 63, 64 Gain‐of‐function mutations in NLRC4 cause potentially lethal periodic fever syndromes that are characterized by high circulating IL‐18 levels and increased risk for the development of macrophage activation syndrome (MAS). 65, 66

11The ALR family member AIM 2 mediates inflammasome responses upon recognition of double‐stranded DNA (dsDNA) in the cytosol of murine macrophages. It does not appear to select for specific nucleic acid sequences and the detected dsDNA fragments may originate from the host's nuclear genome, or may originate from infection with specific viruses (vaccinia virus and cytomegalovirus) or bacteria (Francisella tularensis and Listeria monocytogenes) (Figure 1). 67, 68, 69, 70, 71 GTPases of the guanylate‐binding proteins (GBP) family mediate activation of the AIM2 inflammasome by F. tularensis and other bacterial pathogens. 72, 73 The AIM2 inflammasome has also been shown to promote irradiation‐induced hematopoietic failure and gastrointestinal syndrome in response to genomic dsDNA breaks in mice. 74 Notably, the NLRP3 inflammasome promotes secretion of IL‐1β in human primary monocytes following recognition of cytosolic dsDNA by the cGAS‐STING pathway. 75 This may explain why contrary to other inflammasome sensors, gain‐of‐function mutations in AIM2 have not been reported to cause monogenic autoinflammatory disorders in patients.

12The Pyrin inflammasome indirectly senses inhibition of RhoA GTPase activity by Clostridium difficile toxin A (TcdA) and toxin B (TcdB) and other bacterial toxins. 76, 77, 78 Pyrin phosphorylation (at S208/S242 in human Pyrin; S205/S241 in murine Pyrin) keeps the protein in an inactive state in naïve macrophages. 77, 79 RhoA GTPase inhibition results in dephosphorylation of Pyrin, releasing it from 14‐3‐3 proteins and allowing ASC recruitment and subsequent caspase‐1 activation (Figure 1). In addition, microtubules are required for inflammasome activation downstream of Pyrin dephosphorylation. 77, 78 Gain‐of‐function mutations in the MEFV gene that encodes Pyrin cause Familial Mediterranean Fever (FMF), a monogenic autoinflammatory disease that affects an estimated 150,000 patients worldwide. 60 FMF patients present with recurrent fevers associated with serositis and may develop serum amyloid A (SAA) amyloidosis if colchicine therapy is not initiated in time. Notably, disease‐penetrant FMF mutations were recently shown to render Pyrin inflammasome activation independent of microtubules. 78, 80

13Whereas caspase‐1 is the central effector protease in canonical inflammasome pathways, caspases 4 or 5 in humans, and the orthologous caspase‐11 in rodents, exert this role in the non‐canonical inflammasome pathway that is engaged upon cytosolic detection of lipopolysaccharides (LPS)—a major component of the outer membrane of Gram‐negative bacteria. 81, 82, 83, 84, 85, 86 These inflammatory caspases cleave GSDMD to induce pyroptosis, and the resulting potassium efflux is thought to result in secondary activation of the NLRP3 inflammasome (Figure 1). 10, 81, 85, 86 Inhibition of the non‐canonical inflammasome pathway has gained considerable traction as a potential therapeutic strategy in Gram‐negative infections and sepsis 9, 12, 82, 86, 87, 88, 89

14Therapeutic targeting of IL‐1 family cytokines, and IL‐1β in particular, has gained significant traction due to its central role in a host of autoinflammatory diseases. 90, 91 Moreover, results from both clinical trials and preclinical studies implicate IL‐1β in a suite of metabolic, malignant, and neurodegenerative diseases. 90, 91, 92 IL‐1β is a highly inflammatory and pyrogenic cytokine, and its production and activity are tightly regulated. Cells of the myeloid lineage—the primary source of IL‐1β in the body—do not express IL‐1β in the absence of inflammatory cues, and first need to transcriptionally upregulate its production in the context of infection or tissue damage. Rather than being secreted through the conventional secretory pathways that govern secretion of most other cytokines, activated myeloid cells produce IL‐1β as a biologically inert precursor (proIL‐1β) that is stored in the cytosol. 93 It gains biological activity following caspase‐1‐mediated removal of its amino‐terminal pro‐peptide. Once released into the extracellular space, mature IL‐1β binds to IL‐1 receptor 1 (IL‐1R1) on effector cells, which promotes heterodimerization of the receptor with IL‐1R accessory protein (IL‐1RAcP) and the subsequent recruitment of intracellular signaling components such as myeloid differentiation primary response gene 88 (MyD88). This process initiates a signaling cascade ultimately resulting in the activation of NF‐κB and mitogen‐activated protein kinase (MAPK) signaling (Figure 2). Because of its extensive functions and potency, IL‐1 activity in the extracellular environment is further regulated by two mechanisms. Firstly, the IL‐1R antagonist (IL‐1Ra) binds to IL‐1R1 to block binding of IL‐1α and IL‐1β to IL‐1R1. Binding of IL‐1Ra to IL‐1R1 hampers the heterodimerization with IL‐1RAcP, which prevents initiation of intracellular signaling. 94 Secondly, the decoy receptor IL‐1R type II (IL‐1R2) that lacks the cytoplasmic TIR domain also binds to IL‐1α and IL‐1β, thereby preventing IL‐1 signaling. 95

15Dependent on the cell type, IL‐1 signaling will result in the production of pro‐inflammatory cytokines, such as TNF‐α, IL‐8, and IL‐6; and chemokines that stimulate the attraction of macrophages and other immune cells, such as monocyte chemoattractant protein 1 (MCP1) and the neutrophil attractant chemokines CXCL1 and MIP‐2. Moreover, IL‐1β is a potent driver of IL‐6‐dependent production of acute phase proteins in hepatocytes, of which C‐reactive protein (CRP) and SAA are two major representatives. 96 In addition to its central role in innate immune and inflammatory responses, IL‐1β drives polarization of CD4+ T cells toward T‐helper type (Th) 1 and Th17 cells, and induces expansion and differentiation of antigen‐specific CD8+ T cells. 97, 98 IL‐1β has been attributed a largely beneficial role in resolving acute inflammation. However, several autoinflammatory syndromes are driven by excessive or chronically dysregulated IL‐1 signaling, and IL‐1‐targeted biologics have proven highly effective in treating inflammasomopathies. 60 Three biologic anti‐IL‐1 agents are currently approved for clinical use: anakinra, canakinumab, and rilonacept (Figure 2).

16Anakinra is a nonglycosylated recombinant version of human IL‐1Ra, and functions as a competitive inhibitor of both IL‐1α and IL‐1β for binding to IL‐1R1. It has a good safety profile, but its short half‐life of 4‐6hours makes daily subcutaneous injections necessary to maintain high target engagement levels. Anakinra was initially approved for the treatment of rheumatoid arthritis with methotrexate in patients that failed treatment on methotrexate alone, but these patients are nowadays prioritized to TNF‐blocking agents. Anakinra is nowadays primarily used to treat CAPS and colchicine‐resistant FMF. 60 Recent clinical studies also support its efficacy in gout showing that the drug is equally effective in relieving pain associated with acute gout flares compared to the standard care of treatment. 99 A consensus guideline issued in 2012 also recommended anakinra as a treatment for Schnitzler's syndrome, an autoinflammatory disorder with CAPS‐like symptoms that include urticarial, periodic fever and joint and bone pain. Moreover, in a small multicentre, open‐label, randomized clinical trial that compared anakinra with a host of TNF‐α blockers, IL‐1 inhibition with anakinra also proved effective in improving glycemic and inflammatory parameters in patients affected with rheumatoid arthritis and type 2 diabetes. 100

17Canakinumab is a human monoclonal antibody with an extended half‐life of 26 days that specifically binds and neutralizes IL‐1β. It has been approved by the FDA for the treatment of CAPS. Canakinumab is also being used for treating autoinflammatory diseases such as FMF, TNF receptor‐associated periodic syndrome (TRAPS) and hyperimmunoglobulin D syndrome (HIDS)/mevalonate kinase deficiency (MKD), systemic juvenile idiopathic arthritis, and adult onset Still's disease. 101 More recently, canakinumab was evaluated in the Canakinumab Anti‐inflammatory Thrombosis Outcome Study (CANTOS) (ClinicalTrials.gov Identifier: NCT01327846). This multi‐center, randomized, double‐blind phase 3 trial involved more than 10,000 atherosclerosis patients with previous myocardial infarction and a persistent pro‐inflammatory response, defined by circulating levels of the acute phase protein CRP of 2mg/L or higher. 102 High sensitivity CRP measurements are approved as a clinical biomarker of cardiovascular risk, and CANTOS was designed to evaluate the therapeutic potential of canakinumab in reducing the risk of cardiovascular events and secondary stroke in this cohort. 103 Notably, results from the study showed that IL‐1β inhibition with canakinumab reduced serum levels of CRP in a subset of patients. Moreover, a 25% reduction in major adverse cardiovascular events (MACE) and 31% reduction in cardiovascular mortality and all‐cause mortality was observed in patients who achieved CRP levels lower than 2 mg/L, whereas no significant benefit was observed in patients that retained CRP levels of 2 mg/L or above on canakinumab. 103 In a separate analysis of the CANTOS cohort, the investigators detected a remarkable dose‐dependent reduction in the incidence of lung cancer and lung cancer‐associated mortality with a relative risk reduction of 67% for lung cancer (HR 0.33 [95% CI: 0.18‐0.59]) and 77% for lung cancer mortality (HR 0.23 [95% CI: 0.10‐0.54]) in patients receiving a 300 mg dose of canakinumab every three months. 104 These exploratory results suggest that IL‐1β neutralization may contribute importantly to intercepting the health trajectory of lung cancer patients and warrant further confirmation in independent studies.

18A third clinically approved IL‐1‐targeted therapy for the treatment of CAPS is rilonacept or IL‐1 Trap. 105 Rilonacept is a human dimeric fusion protein consisting of the extracellular domains of IL‐1R1 and IL‐1RAcP fused to the Fc fragment of IgG. It functions as a soluble decoy receptor that prevents binding of both IL‐1α and IL‐1β to the IL‐1R1 (Figure 2). This chimeric protein has a half‐life of approximately 1week, bridging those of anakinra and canakinumab. Rilonacept also proved effective in controlling symptoms in a cohort of patients suffering from Schnitzler's syndrome (ClinicalTrials.gov Identifier: NCT01045772). 106

19Although proIL‐18 is constitutively expressed in myeloid cells, production of bioactive IL‐18—akin to IL‐1β—requires caspase‐1‐mediated cleavage following inflammasome activation. Once secreted into the extracellular environment, binding of IL‐18 to IL‐18 receptor α (IL‐18Rα) and IL‐18 receptor β (IL‐18Rβ) results in downstream activation of inflammatory signaling in effector cells. Given its pleiotropic immune roles, several mechanisms exist that tightly regulate IL‐18 activity in the extracellular milieu. 60, 94 One such regulatory mechanism that is being exploited for diagnosis and therapy of patients is the neutralization of IL‐18 by the natural IL‐18 binding protein (IL‐18BP), which precludes binding of IL‐18 to its cognate receptors (Figure 2).

20Given that the expression ratio of IL‐18 and IL‐18BP is often disturbed in inflammatory diseases, circulating levels of IL‐18BP‐unbound IL‐18 (free IL‐18) have been proposed as a biomarker for disease activity in adult onset Still's disease and systemic‐onset juvenile idiopathic arthritis. 107, 108 Moreover, a recombinant version of human IL‐18BP named “tadekinig alfa” demonstrated a favorable safety profile with early signs of efficacy in a recent phase II clinical trial of patients with adult onset Still's disease (ClinicalTrials.gov Identifier: NCT02398435). 109 Based on promising data showing successful treatment of life‐threatening inflammation in an NLRC4‐MAS patient 110, the therapeutic potential of anti‐IL‐18 therapy with tadekinig alfa is currently being investigated in a single‐arm, open‐label phase 3 trial in patients afflicted with NLRC4‐MAS or autoinflammation due to monogenic XIAP mutations (ClinicalTrials.gov Identifier: NCT03113760) given that both syndromes feature high circulating IL‐18 levels.

21GSK1070806 is another biologic that was developed to neutralize IL‐18‐mediated inflammation. Unlike tadekinig alfa, GSK1070806 is a recombinant human IL‐18‐neutralizing antibody with an extended half‐life in circulation. It is currently being tested in a phase 2 clinical study for the treatment of patients with moderate to severe Crohn's disease (ClinicalTrials.gov Identifier: NCT03681067) (Figure 2). If proven effective, these clinical studies may lay the foundation to expand the use of IL‐18‐inhibiting biologics to additional autoinflammatory and autoimmune diseases and expand our understanding of how IL‐18 promotes destructive and potentially life‐threatening inflammation in patients.

22IL‐1‐ and IL‐18‐targeted biologics have dramatically improved therapeutic outcomes for patients suffering from a subset of autoinflammatory diseases, but these protein‐based products require patients to subject themselves to frequent subcutaneous injections. More importantly, chronic suppression of a key arm of the innate immune system inevitably comes with an increased risk for serious infections. Clinical experience with canakinumab and anakinra provides comforting real‐world evidence that the risk of Mycobacterium tuberculosis reactivation and tuberculosis development is significantly lower with IL‐1‐targeted biologics compared to TNF‐α inhibitors, but it is evident that chronic IL‐1 blockade does increase risk of opportunistic infections of the upper airways along with some evidence of increased urinary tract infections with Escherichia coli and Streptococci. 102, 111 Additional adverse events that are associated with chronic IL‐1 inhibition include neutropenia, low platelet counts, headaches, abdominal pain, diarrhea, urticarial lesions, and inflammation at the site of injection. 100, 101, 102

23Therapies acting upstream of IL‐1β secretion might show an improved safety and efficacy profile because selective inflammasome inhibition would still allow IL‐1‐driven host defense through non‐targeted inflammasomes, for example, when the patient faces an acute infection, while preventing chronic inflammatory pathology that originate from the silenced inflammasome. At the same time, selective inflammasome targeting may increase therapeutic potency through the simultaneous blockade of IL‐1β, IL‐18 and by preventing pyroptosis‐mediated release of DAMPs such as IL‐1α, HMGB1, and ATP. This notion is supported by empirical evidence from a host of mouse disease models that point to GSDMD‐driven pyroptosis as the key driver of inflammasome‐mediated pathology. 17, 18, 21, 61, 112, 113 Finally, small molecule inflammasome inhibitors that can be dosed orally would relief patients from the need of lifelong injections. Hence, a suite of tool compounds and natural compounds have been reported to inhibit inflammasome activation in recent years, and some molecules are currently being evaluated in early clinical studies.

24A vast body of studies has implicated the NLRP3 inflammasome in a host of inflammatory, metabolic, and neurodegenerative diseases, rendering this pathway an attractive target for pharmacological modulation across therapeutic areas. 60, 91, 92, 114 The first demonstration that selective pharmacological inhibition of the NLRP3 inflammasome is feasible arrived in 2009 with the demonstration that the sulfonylurea‐containing compound glyburide inhibited NLRP3‐induced pyroptosis and IL‐1β secretion without affecting activation of the NLRC4 and NLRP1b inflammasomes. 115 The half‐inhibitory concentration (IC50) of glyburide is in the low μM range, whereas the related diarylsulfonylurea compound MCC950/CRID3 inhibits NLRP3 activation with nM potency. 116 Like glyburide, MCC950/CRID3 exerts high target selectivity as it inhibits Nlrp3 activation without interfering with the AIM2, NLRC4, Pyrin, and NLRP1b inflammasomes, or altering TLR4‐induced cytokine secretion. 78, 116 More recently, MCC950/CRID3 and related tool compounds were shown to directly target the NACHT domain of NLRP3 in the vicinity of its ATP/dATP binding pocket, and they represent the most potent tool compounds that are currently available for selective NLRP3 inflammasome inhibition in human cells and preclinical rodent studies. 117, 118, 119 Consequently, MCC950/CRID3‐mediated inhibition of NLRP3 has shown beneficial effects in mice and other preclinical species that have been subjected to disease models, among which models of atherosclerosis, myocardial infarction, colitis, and skin and airway inflammation. 47, 116, 120, 121, 122, 123 Interestingly, MCC950/CRID3 and related diarylsulfonylurea compounds may be substantially less effective in inhibiting CAPS‐related NLRP3 mutants because in vivo MCC950/CRID3 concentrations that potently inhibited NLRP3‐driven inflammation in wildtype mice were shown to be ineffective in two in vivo mouse models of CAPS disease. 119 Concordantly, MCC950/CRID3‐derived photoaffinity probes labeled the NACHT domain of wildtype NLRP3, but not CAPS‐associated NLRP3 mutants. 119 These results raise the possibility that MCC950/CRID3‐based therapies may need to be dosed at higher concentrations in CAPS patients to effectively treat inflammatory pathology than is needed to curb inflammatory pathology driven by wildtype NLRP3.

25In addition to inflammatory disease, MCC950/CRID3 also improved clinical and molecular disease markers in models of neurodegenerative disease such as Parkinson's disease, Alzheimer's disease, and the EAE model of multiple sclerosis. 52, 92, 114, 124 While EAE is associated with breakdown of the blood‐brain barrier and enhanced vascular permeability, 125 it is currently unclear whether MCC950/CRID3 also crosses the blood‐brain barrier in Alzheimer's and Parkinson's disease models to reach clinically relevant concentrations in the central nervous system, or whether it dampens neuroinflammatory pathology indirectly in these models by modulating NLRP3 activity in the periphery.

26Based on these promising results, several MCC950/CRID3 analogs are currently being developed for clinical applications. Two such NLRP3 antagonists, namely IZD334 (ClinicalTrials.gov Identifier: NCT04086602) and Inzomelid (ClinicalTrials.gov Identifier: NCT04015076), have recently completed early clinical testing to assess the safety, tolerability, and pharmacological properties of the drugs in healthy adult participants and to obtain first evidence of clinical efficacy in adult CAPS patients (Figure 3). As a secondary outcome, these studies analyzed NLRP3 activity in stimulated whole blood and monitored clinical symptoms in adult CAPS patients.

27Several other small molecule NLRP3 inhibitors have been reported, although the potency and selectivity profile of most are less stringent than that of MCC950/CRID3. For instance, Bay 11‐7082 and parthenolide are anti‐inflammatory compounds that inhibit ATPase activity of NLRP3 and also interfere with NF‐κB‐mediated cytokine production. 126 Moreover, parthenolide was shown to be a direct inhibitor of recombinant caspase‐1 in vitro 126; and Bay 11‐7082 and parthenolide were recently also reported to bind and inhibit GSDMD. 127 These molecules therefore appear less suitable for clinical development as NLRP3‐targeted therapies given the risks associated with broad immunosuppression.

28The tryptophan derivative tranilast also inhibits NLRP3 by direct binding to its NACHT domain (Figure 3). 128 This molecule demonstrated therapeutic efficacy in mouse models of NLRP3‐driven gouty arthritis and CAPS. In addition, tranilast blunted spontaneous caspase‐1 activation and IL‐1β release from isolated synovial fluid mononuclear cells of arthritis patients. 128 Since its identification in 1982 as an anti‐allergic agent that suppresses histamine release from mast cells, tranilast has been associated with a growing list of anti‐inflammatory effects, including suppression of TGF‐β, NF‐κB, and MAPK signaling (reviewed in 129), suggesting that it may not solely inhibit inflammation by suppressing NLRP3 activation. However, considering its clinical safety, tolerability, and efficacy profile in different indications, tranilast is currently being evaluated in a phase 2 open‐label clinical trial for efficacy and safety in CAPS patients (ClinicalTrials.gov Identifier: NCT03923140).

29The β‐sulfonyl nitrile compound OLT1177 also was shown to inhibit NLRP3‐mediated IL‐1β release from stimulated PBMCs of CAPS patients and primary human neutrophils (Figure 3). 130 Studies in mice suggest additional NLRP3‐independent anti‐inflammatory mechanisms of actions for the compound, but in a phase 1 clinical trial, OLT1177 showed a good pharmacokinetic profile in healthy individuals with long half‐life and no hematological or organ toxicity at any of the tested doses. 130 A recently completed phase 2 proof of concept study suggests that OLT1177 may have significant anti‐inflammatory activity in patients suffering from acute gout. 131

30CY‐09 is an analog of a cystic fibrosis transmembrane conductance regulator (CFTR) channel inhibitor that was found to inhibit the NLRP3 inflammasome with μM IC50 values without affecting NLRP3 priming. 132 Unlike the parental molecule, CY‐09 failed to inhibit CFTR activity and was shown to specifically target the ATPase activity of NLRP3 by binding at the walker A motif in the central NACHT domain (Figure 3). Concordantly, CY‐09 significantly decreased serum levels of IL‐1β and neutrophil influx in a murine model of peritoneal MSU‐induced inflammation, and it protected mice from neonatal lethality upon in vivo expression of an MWS‐associated NLRP3 mutant. 132

313,4‐methylenedioxy‐β‐nitrostyrene (MNS), a known inhibitor of tyrosine kinases, emerged as a hit for NLRP3 from a focused kinase inhibitor library screen. 133 It interferes with NLRP3's ATPase activity by binding to the central NACHT and carboxy‐terminal LRR domains of NLRP3 (Figure 3). The acrylamide derivative INF58 irreversibly inhibits NLRP3 by covalently modifying Cys residues in its target protein. It was shown to inhibit ATPase activity of human recombinant NLRP3 and IL‐1β release from primary and immortalized macrophages. 134

32Unlike the previously mentioned synthetic molecules, oridonin is a natural compound derived from the medicinal plant Rabdosia rubescens. It was shown to inhibit NLRP3 activation by preventing the interaction between NLRP3 and NEK7 through covalent binding to Cys279 in the NACHT domain of NLRP3 (Figure 3). Furthermore, oridonin inhibited IL‐1β production in NLRP3‐dependent mouse models of peritonitis and gouty arthritis. 135 However, in addition to interfering with NLRP3 inflammasome activation, the molecule was shown to inhibit inflammation through several additional mechanisms of action, including inhibition of NF‐κB and MAPK signaling, induction of apoptosis, cell cycle arrest and it was also shown to have anti‐tumor activity. 136, 137, 138, 139

33In addition to molecules with a direct mechanism of action on NLRP3, several molecules that inhibit NLRP3 activation indirectly by targeting post‐translational modifications have also been reported. Two recent studies suggest that acetylation of the amino‐terminal PYD in NLRP3 is required for recruitment of ASC and optimal downstream inflammasome signaling. 140, 141 The authors go on to show that this mechanism was specific to NLRP3 because the AIM2 and NLRC4 inflammasomes were not affected by inhibition of this post‐translational modification. Furthermore, c‐Jun N‐terminal kinase1 (JNK1)‐mediated phosphorylation of NLRP3 Ser194 was shown to be a critical priming event that supports optimal NLRP3 activation. 34 Although selective modulation of NLRP3 activation through these indirect mechanisms may prove challenging, potent pharmacological inhibitors that target specific post‐translational events might provide additional avenues to suppress NLRP3‐driven inflammation.

34In contrast to the breath of molecules that inhibit the NLRP3 inflammasome, selective small molecule inhibitors of other inflammasome sensors remain to be discovered. Nevertheless, the microtubule polymerization inhibitor colchicine selectively inhibits activation of the Pyrin inflammasome in murine macrophages and human monocytes. 78 Interestingly, however, disease‐penetrant FMF mutations in MEFV render activation of the Pyrin inflammasome resistant to colchicine inhibition. This feature appeared highly specific for FMF‐associated mutations and was recently leveraged to develop a functional diagnostic assay for rapidly stratifying FMF patients. 80 Significantly lower doses of colchicine are being used in the clinic since the 1970' to effectively control FMF symptoms, but the therapeutic mode of action of colchicine in FMF and gout patients remains a mystery.

35Human NLRP1 and its murine paralogs Nlrp1a and Nlrp1b are unique in their requirement for proteasomal degradation to promote inflammasome activation. 26, 27, 142, 143 In line herewith, proteasome inhibitors such as MG‐132 and bortezomib selectively inhibit activation of the NLRP1 inflammasome without modulating activation of NLRP3 and other inflammasome pathways.

36As the protease that matures IL‐1β and the common denominator of all canonical inflammasome pathways, developing clinical‐grade molecules that selectively inhibit caspase‐1 protease activity has been a focal effort of the pharmaceutical industry for many years. Like in other caspases, the evolutionary conserved Cys285 in the active site of caspase‐1 employs a nucleophilic attack. An electrophilic warhead that reacts with and inhibits Cys285 is therefore one of the strategies used by reversible (aldehyde‐containing) and non‐reversible (ketone‐based) tetrapeptide inhibitors to inhibit caspase‐1. 144 Nonetheless, due to their toxic byproducts, poor stability, solubility and selectivity, tetrapeptide‐based inhibitors are not suitable for therapeutic application. 145 Fully synthetic peptide‐mimetic prodrugs such VX‐740 (pralnacasan) and VX‐765 (belnacasan) were generated to overcome the toxicity and low bioavailability issues associated with peptidic caspase inhibitors. These prodrugs are metabolically converted in the cytosol by esterase activity to, respectively, VRT‐18858 and VRT‐043198, both of which act as reversible inhibitors of caspase‐1 and the related inflammatory caspases 4 and 5 (Figure 3). 146, 147

37VX‐740 was shown to reduce joint damage in two mouse models of osteoarthritis and it attenuated dextran sulfate sodium (DSS)‐induced colitis. 147, 148 However, a phase 2 clinical trial with VX‐740 in rheumatoid arthritis patients was discontinued after hepatotoxicity was observed in long‐term follow‐up animal studies. 145 VX‐765, on the other hand, was shown to reduce acute seizures and chronic epileptic activity in two mouse models of epilepsy, and it significantly suppressed the production of inflammatory mediators in models of rheumatoid arthritis and skin inflammation. 149 It was also shown to improve cognition in a mouse model of Alzheimer's disease by inhibiting amyloid beta deposition and neuroinflammation. 150 Although VX765 has a favorable safety profile, a phase 2 clinical trial in patients with treatment‐resistant partial epilepsy failed to reach its primary endpoints (ClinicalTrials.gov Identifier: NCT01501383). 151 These challenges have spurred investigators to refocus efforts on inhibiting inflammasome sensors that act upstream of caspase‐1, and more recently, on targeting the pyroptosis effector GSDMD.

38GSDMD is expressed as a soluble, inactive precursor protein that consists of an amino‐terminal pore‐forming domain (GSDMD‐N) and a carboxy‐terminal auto‐inhibitory domain (GSDMD‐C) that are separated by a flexible interdomain linker. 9, 10 Inflammatory caspases (ie, caspases 1, 4, and 5 in humans; and caspases 1 and 11 in mice) cleave GSDMD in the interdomain linker, thus allowing GSDMD‐N to translocate into the inner plasma membrane leaflet. It subsequently assembles into oligomers that pinch both plasma membrane leaflets to form large GSDMD pores allowing osmotic swelling and pyroptotic cell death. 152 In addition to pyroptosis, GSDMD was shown to induce the release of web‐like chromatin structures called “neutrophil extracellular traps” from neutrophils as part of a specialized cell death mode known as NETosis. 153, 154

39Mice deficient in GSDMD are significantly protected against LPS‐induced endotoxic shock and lethality. 9 Moreover, GSDMD was shown to play a critical role in driving IL‐1β‐dependent inflammatory pathology in mouse models of CAPS and FMF. 17, 18 Moreover, preventing GSDMD activation in peripheral myeloid cells was recently shown to protect mice from neuroinflammation and demyelination in the EAE model of multiple sclerosis. 113 Another recent study found a key role for GSDMD in non‐alcoholic steatohepatitis in humans and mice. 155

40These findings raised significant interest in the discovery of GSDMD‐targeted drugs, and several molecules have already been reported to inhibit GSDMD‐mediated pore formation. The alkylating compound necrosulfonamide (NSA) binds directly to full‐length GSDMD via Cys191, thereby preventing GSDMD‐N dimers from assembling higher order oligomers that drive pyroptotic pore formation and cell lysis (Figure 3). NSA similarly modifies a reactive Cys residue in mixed lineage kinase domain‐like pseudokinase (MLKL), thus preventing its recruitment to RIPK3 and inhibiting induction of another lytic cell death mode termed necroptosis. 156, 157

41Disulfiram and Bay 11‐7082 were recently discovered in a screen for inhibitors of GSDMD using a fluorogenic liposome leakage assay. 127 Notably, both molecules inhibited GSDMD by covalently modifying the GSDMD Cys191 residue that is also targeted by NSA (Figure 3). 127 As discussed above, Bay 11‐7082 is a potent inhibitor of inhibitor κB kinase (IKK) that also targets NLRP3 and caspase‐1. 126 As an inhibitor of acetaldehyde dehydrogenase, disulfiram is used in the clinic for treating alcohol addiction. 158 Disulfiram also inhibited TLR‐induced priming at higher concentrations, and both Disulfiram and Bay 11‐7082 attenuated ASC speck formation and the protease activities of caspases 1 and 11. 126, 127, 159 Disulfiram was shown to inhibit pyroptosis in human and murine myeloid cells with µM potency, and it prevented secretion of IL‐1β from these cells. 127 When dosed intraperitoneally in LPS‐challenged mice, it significantly delayed lethality at a concentration of 50 mg/kg. 127

42LDC7559 emerged as a direct inhibitor of GSDMD from a small molecule screen for inhibitors of phorbol ester‐induced NETosis in human neutrophils (Figure 3). 153. Whereas disulfiram, NSA and Bay 11‐7082 inhibit GSDMD through covalent modification of Cys191, the mechanism of action of LDC7559 is unclear. LDC7559 inhibited pyroptosis by the NLRP3 and AIM2 inflammasomes, 153 but other findings suggest that it may lack activity on recombinant GSDMD in the liposome leakage assay. 160 Future studies should shed light on this apparent paradox.

43Altogether, these findings highlight GSDMD as an attractive novel target to modulate pathological inflammation driven by pyroptosis, including the release of inflammasome‐dependent cytokines and DAMPs. While the current GSDMD tool compounds may still be a long way from the clinic, they may set the stage for the development of selective GSDMD‐targeted therapies by helping to shed light on the mechanisms that regulate GSDMD pore formation.

44As discussed above, results from CANTOS suggest that canakinumab‐mediated blockage of IL‐1β activity protects against the development of lung cancer, although the mechanism involved remains unclear. However, IL‐1β may have opposing roles in established tumors. 91 Evidence that IL‐1β may have anti‐tumor activity was first presented 30 years ago, when intraperitoneal injection of recombinant human IL‐1β induced T‐cell‐dependent regression of subcutaneous SA1 sarcoma and L5178Y lymphomas in mice. 161 The cancer‐suppressive effect of IL‐1β was confirmed in a mouse model for myeloma showing that IL‐1β (as well as IL‐1α) neutralization by anakinra severely impaired myeloma clearance by tumor‐specific Th1 cells and tumor‐infiltrating macrophages. 162 Additionally, IL‐1β release from dendritic cells was shown to be essential for priming of interferon gamma (IFNγ)‐producing CD8+ T cells and elimination of tumor cells. 163

45Researchers are also investigating whether inflammasome activation and the induction of pyroptosis upstream of IL‐1β release can help render tumors more sensitive to therapy. Inhibitors of immune checkpoint modulators such as programmed death 1 (PD‐1) and cytotoxic T‐lymphocyte antigen 4 (CTLA‐4) invigorate CD8+ T cell‐mediated killing of tumor cells and are revolutionizing immunotherapy of cancer. However, only a minority of patients responds to checkpoint blockade due to a lack of CD8+ T‐cells in the tumor microenvironment. Researchers are exploring combination therapies that facilitate CD8+ T‐cell intratumoral infiltration in order to sensitize tumors to checkpoint therapy. In line with this strategy, an investigational phase I clinical trial in patients with solid cancers has recently been initiated with the NLRP3 agonist BMS‐986299 to study its therapeutic potential as a pyroptosis‐stimulating compound alone or when dosed to patients in combination with the checkpoint inhibitors Nivolumab and Ipilimumab (ClinicalTrials.gov Identifier: NCT03444753) (Figure 3).

46Another clinical study aims to clarify the therapeutic effect of pyroptosis induction through the NLRP1 inflammasome in patients diagnosed with acute myeloid leukemia (AML), a heterogenous hematological malignancy driven by uncontrolled expansion of myeloid progenitor cells in bone marrow and in circulation. This study centers on the experimental broad‐spectrum DPP inhibitor BXCL701 (also named VbP, PT‐100, or Talabostat) that was recently assigned orphan drug designation for the treatment of AML by the FDA, and is based on recent findings showing that VbP/BXCL701 and more selective DPP8/DPP9 inhibitors are potent inducers of pyroptosis in AML cell lines and in primary AML samples from patients. 164

47Several recent studies showed that enforced activation of Gasdermin family members in cancer cells enhanced immune cell‐mediated tumor clearance in mouse models, 165, 166, 167 adding credence to the notion that pyroptosis induction may enhance cancer immunotherapies to the benefit of patients.

48It has now become evident that inflammasomes play a crucial role in the immune response during microbial infections and non‐infectious diseases. A wealth of information regarding the complexity of inflammasome signaling and their roles in diseases has emerged during the past two decades. This has spurred interest in translating basic research findings into potential novel therapies for inflammasome‐driven autoinflammatory diseases. Moreover, intriguing findings from the CANTOS trial suggest that inflammasomes may contribute importantly to cardiovascular risk in atherosclerosis patients and to the development of lung cancer. Without doubt, the ongoing development of clinical inflammasome agonists and inhibitors will help tremendously to gain further insights into inflammasome‐driven autoinflammatory pathologies in the future and may expand treatment options for patients suffering from a host of common diseases with high unmet need, including metabolic and neurodegenerative diseases as well as certain cancers.

49D. Chauhan is an employee of Janssen Pharmaceutica N.V. The authors declare no conflict of interest.