Inflammatory caspase-related pyroptosis: mechanism, regulation and therapeutic potential for inflammatory bowel disease.

As an essential part of programmed cell death, pyroptosis is an inflammatory response that is elicited upon infection by intracellular pathogens. Metabolic diseases, atherosclerosis and vital organ damage occur if pyroptosis is over-activated. Macrophages are the main cells that induce pyroptosis with the help of intracellular pattern-recognition receptors stimulated by danger signals and pathogenic microorganisms in the cytosol of host cells. Activated inflammatory caspases induce pyroptosis and produce pro-inflammatory cytokines, such as interleukin-1β and interleukin-18. Inflammatory programmed cell death is classified as canonical or non-canonical based on inflammatory caspases, which includes caspase-1 (in human and mouse) and caspase-11 (in mouse) or caspase-4 and -5 (in humans). Activated inflammatory caspases cleave the pore-forming effector protein, gasdermin-D, inducing osmotic pressure deregulation of internal fluids and subsequently rupturing the cell membranes. Inflammatory caspases could be attractive therapeutic targets for inflammatory bowel disease (IBD) in which pyroptosis may play an important role. This article reviews the current understanding of the mechanism of pyroptosis, focusing on the regulation of inflammatory caspases and therapeutic strategies for IBD.

Introduction

Pyroptosis is a form of programmed cell death and plays a critical role in both tissue homeostasis and immune response. Dysregulation of pyroptosis promoted by microbial infection and danger signals is often associated with immune dysregulation and excessive inflammation response can result in a number of inflammatory diseases. Pyroptosis is characterized by macrophage cell membrane rupture brought about by cell swelling and lysis. Furthermore, these cells release pro-inflammatory cytokines and intracellular contents [1]. Pyroptosis was first presented as an unusual caspase-1-dependent mechanism of necrosis in macrophages infected with Salmonella typhimurium, and was different from traditional necrosis and apoptosis [2]. The newly caspase-1-dependent pro-inflammatory form of programmed cell death was named pyroptosis and arises from microbial infection and danger signals, including crystalline substances, monosodium urate, asbestos, silica crystals toxins and extracellular ATP [3, 4]. It is well known that pyroptosis plays a critical role in resisting microbial colonization [5]. In addition, pyroptosis contributes to the pathogenesis of autoinflammatory and autoimmune diseases, such as kidney diseases [6], neurodegenerative diseases [7, 8], ischemia-reperfusion [9], liver fibrosis [10], colon cancer [11] and atherosclerosis [12].

Although different inflammatory caspases can cause canonical or non-canonical inflammasome pathway activation, the co-substrate gasdermin-D (GSDMD) causes them to have identical morphological changes. Pyroptosis forms pores in the plasma membrane using a necrotic N-terminal fragment [13]. A study of Salmonella-infected macrophages found that functional pyroptotic pores were around the 1.1–2.4-nm range in diameter based on the size of osmoprotectant molecules that prevented cell lysis [14, 15]. Recent studies suggest that DNFA5 is cleaved by caspase-3 into Asp270, which generates a necrotic DFNA5-N fragment that could induce secondary necrosis or pyroptosis by forming cell membrane pores [16, 17]. DNFA5 induces the signaling pathways of pyroptosis. The relationship between pyroptosis and other gasdermin family members, including GSDMA, GSDMB and GSDMA3, on the ability to form gasdermin N-terminal domains by specific inducers needs to be deciphered. Caspase-3, a distinctive apoptotic caspase, may play an important role between apoptosis and pyroptosis.

Inflammatory bowel disease (IBD) is a chronic, relapsing and incurable disease of the intestinal tract, which includes Crohn’s disease (CD) and ulcerative colitis (UC). Although research has shown that the pathogenesis of IBD is closely related to mucosal barrier damage, dysbacteriosis and immunity dysregulation, the mechanisms underlining the pathogenesis have not been fully understood [18]. There is increasing evidence that pattern-recognition receptors may play an important role in maintaining immune-system homeostasis, modulating the gastrointestinal microflora and facilitating intestinal epithelium regeneration and repair [19]. Numerous studies have indicated the role of NOD-like receptors (NLRs) in IBD on forming signaling complexes called inflammasomes. These complexes cleave and activate pro-cytokines interleukin-1β (IL-1β) and IL-18 into mature IL-1β and IL-18, to induce pyroptosis [20].

In this review, we aim to explore of the regulatory mechanism of pyroptosis and inflammatory caspases in immune dysregulation. Of particular importance is the therapeutic potential for IBD through targeting pyroptosis signaling.

Regulation of inflammatory caspases in pyroptosis

Caspases are a family of highly conserved aspartate-specific cysteines with 15 mammalian members. The majority can be grouped into inflammatory caspases and apoptotic caspases [21]. The former includes caspase-1, -4, -5 and -11, and the latter consists of caspase-2, -3, -6, -7, -8, -9 and -10 [22]. The initiation of pyroptosis and apoptosis relies on specific caspases to induce their respective programmed cell death pathways [23]. Unlike apoptotic caspases, inflammatory caspases are not categorized as initiators or effectors, but induce a form of necrotic or inflammatory programmed cell death, namely pyroptosis [24]. Previous studies have characterized the various contributions of inflammatory caspases in murine models of salmonellosis [25]. Pyroptosis was first described as a mediator of innate immune responses in macrophages and is initiated by caspase-1, resulting in the secretion of IL-1β and IL-18 [2, 14]. IL-1β is an inducer of vasodilation, immune-cell extravasation and inflammation [26]. It also has a significant role in shaping adaptive immunity [26]. IL-18 induces interferon (IFN)-γ production in Th1 cells, NK cells and cytotoxic T cells, participates in Th2 development and enhances local inflammation [27]. Subsequent studies have demonstrated that pyroptosis also participates in eliminating other bacterial infections, not just Salmonella, even in the absence of IL-1β and IL-18 [5].

Studies have demonstrated that all inflammatory caspases can be activated by the inflammasome [28, 29]. Caspases associate with inflammasome-initiating sensors (NLRP1, NLRP3, NLRC4, AIM2 or pyrin) to form macromolecular protein complexes even in the absence of the inflammasome adapter protein ASC (apoptosis-associated speck-like protein containing a caspase recruitment domain) [28, 30]. Necroptosis, defined as a necrotic and inflammatory form of programmed cell death, in the absence of caspase signaling, can be distinguished from pyroptosis [31, 32]. Pyroptosis can be induced by different inflammatory caspases and presents morphological characteristics such as cell swelling, positivity for Annexin V and TUNEL staining, chromatin condensation and membrane pore formation, but with the absence of DNA laddering [14, 15, 33]. Consequently, the cells rupture and release cytoplasmic contents, pro-inflammatory cytokines, endogenous ligands, alarmins and other toxins [34-37].

In the following sections, the canonical and non-canonical inflammasome pathway and the two major pyroptosis signaling pathways will be discussed. The canonical inflammasome pathway is activated by caspase-1 and assembled by cytoplasmic sensors, such as NLRs. The non-canonical inflammasome pathway is related to caspase-4/5/11 and is stimulated by immune-system activators, such as lipopolysaccharide (LPS), which are the components of the bacterial cell wall [38, 39].

Canonical inflammasome pathway in pyroptosis

Depending on the inflammation-initiating sensors to recognize pathogen-associated molecular patterns (PAMPs) or danger-associated molecular patterns (DAMPs), caspase-1 activation can induce the canonical inflammation pathway in pyroptosis [31, 40]. The C-terminal caspase recruitment domain is considered the ‘bridge’ between pro-caspase-1 and NLRs containing pyrin domains. ASC has an important role in multiple inflammatory diseases and autoimmune diseases associated with inflammasome dysfunction [41-43]. As mentioned previously, Pyrin, Aim2 and NLRs have been associated with pyroptosis. NLR is one of the inflammation-initiating sensors that consists of a tripartite structure: a C-terminal leucine-rich repeat that detects PAMP; a central nucleotide-binding oligomerization domain that mediates self-oligomerization important for activation; and an N-terminal effector domain, which acts as an N-terminal caspase recruitment domain, pyrin domain or baculovirus inhibitor repeat [44]. Inflammation-initiating sensors commit themselves to recognizing danger signals, such as pathogens, foreign bodies and host molecules [45, 46]. They invade the host cell cytoplasm and trigger the activation of caspase-1 [45, 46], which leads to inflammation assembly and inflammatory cytokines IL-18 and IL-1β secretion.

Although there are slight differences in the activation modes among the different microbial infections, a series of experiments have demonstrated similar molecular mechanisms. For instance, inflammasome sensors of NLRP1b and NLRP1 in mouse are activated by Bacillus anthracis poisons [47]. AIM2 is activated by direct binding to the cytoplasmic dsDNA [48]. NLRP3 is activated primarily by cytosolic flagellin and the NEK7 kinase has recently been demonstrated to play a role in upstream signaling [49, 50]. Pyrin responds to bacterial toxin-induced modifications of Rho GTPases and requires microtubule assembly for its activation [51]. Unlike other inflammation-initiating sensors, NLRC4 does not have an N-terminal pyrin domain and binds directly to the inner rod, flagellin or needle proteins of the Type III secretion system of bacteria, which then activates pyroptosis accompanied by the secretion of IL-1β [52]. Conversely, abundant IL-1β would increase ASC speck formation and synergize with cofactors to promote pattern recognition to assist NLRC4 signaling [53]. Increasing evidence suggests that interactions between multiple NLRs may contribute to inflammasome formation [1].

Host cells interact with caspase-1-activating ligands and sensors to trigger common downstream signaling to induce pyroptosis [54]. In addition, the adapter protein ASC can self-associate and form analogous complexes without an NLR to form inflammasomes [5]. For instance, Salmonella-mediated caspase-1 activation is downregulated in ASC-deficient macrophages, suggesting that NLRC4 facilitates caspase-1 activation, rather than completely relying on ASC [52]. In general, caspase-1 can be activated by different inflammation-initiating sensors mentioned above, such as inflammasomes, which are the cytosolic multiprotein complexes assembled by intracellular nucleotide-binding oligomerization NLRs. Caspase-1 induces pro-inflammatory cytokines (pro-IL-1β, pro-IL-18) and gasdermin-D N-terminal domains to form pores in the plasma membrane of an infected cell (details discussed below), which then causes infected cells to release bacteria and inflammatory cytokines (Figure 1).

Figure 1.

Inflammatory caspases (caspase-1 and caspase-4/5/11) are activated through canonical or non-canonical inflammasomes, triggered by microorganism infection. In the non-canonical pathway, procaspase-4/5/11 directly binds to lipopolysaccharide (LPS), resulting in their activation. In the canonical pathway, inflammasome-initiating sensors (NLRP1, NLRP3, NLRC4, AIM2 or pyrin) recognize pathogens and danger signals, and form multimerization of inflammasomes binding with caspase-1. Upon activation, inflammatory caspases cleave gasdermin-D to produce gasdermin-D N-terminal (NT) fragments. Gasdermin-D NT as a pore-forming protein contributes to pyroptosis pore-forming in the infected cell. This then releases cytokines and debris of pyroptotic macrophages, which had previously trapped intracellular bacteria to further promote uptake into neutrophils by efferocytosis (secondary phagocytes).

Non-canonical inflammasome pathway in pyroptosis

Caspase-1-independent pyroptosis can be activated by non-canonical inflammasomes—to be precise, by human caspase-4, human caspase-5 or mouse caspase-11 [55]. Unlike the canonical inflammasome pathway, these caspases recognize LPS from gram-negative bacteria with caspase recruitment domains in the host cytoplasm [39, 56]. A previous study demonstrated that Casp1-/-Casp11-/- mice with orogastric infection caused by S. typhimurium have more bacteria in their systemic organs compared to Casp1-/-Casp11Tg mice, while both strains of mice have more bacteria than wild-type mice [57]. This suggests that caspase-11 increases susceptibility to Salmonella infection. Moreover, intracellular Salmonella could induce host cell death via caspase-11 in the absence of caspase-1 [57, 58]. Knodler et al. [59] demonstrated that IL-18 secretion in intestinal tissue in response to S. typhimurium infection was dependent on caspase-11 to control bacterial numbers in the cecum. Kayagki et al. [60] showed that caspase-11 can trigger pyroptosis, although weaker in directly cleaving pro-IL-1β and pro-IL-18, and Knodler et al. [59] demonstrated that caspase-4 has the ability to cleave pro-IL-1β and pro-IL-18. Numerous pieces of research have suggested that non-canonical inflammasome pathways play a dominant role during acute inflammation, whereby activation of caspase-11 brings about IL-1β and IL-18 cleavage, which is a typical mechanism of the canonical inflammasome pathway [61-63] (Figure 1).

GSDMD N-terminal domains induce pyroptosis

GSDMD is a common substrate for inflammatory caspases and has recently been determined to be the mechanistic link between caspase signaling and pyroptosis [60, 64, 65]. Numerous studies have demonstrated that the N-terminal fragment of GSDMD is the key player in inducing pyroptosis, not merely in vitro, but also in vivo. In addition, only in the presence of both GSDMD and caspase-1 were liposomes able to release dye, suggesting membrane pore formation [15, 40, 60, 66]. Additional studies have demonstrated that GSDMD was cleaved by caspase-1, 4, 5 or 11, but not with apoptotic caspases (caspase-2, -8 and -9) in 293 T cells. The cleavage products (31 kDa) of GSDMD contain 276GVPAEG281 or 272FLTD275 in human and 273LLSD276 in mouse [67]. The cleavage of GSDMD is required for pyroptosis. The D/A mutant of GSDMD (Asp276 in mouse gasdermin-D converted to A) is resistant to cleavage by caspase-1 and caspase-4/5/11 [67]. When these were over-expressed in 293 T cells, they were unable to induce pyroptosis [67]. A current hypothesis on the mechanism by which GSDMD induces pyroptosis is that the GSDMD N-terminal domain binds to phosphoinositides and cardiolipin, then oligomerizes to form pores on the bacterial cell membrane [64, 66] (Figure 2). Specifically, cleaved GSDMD is found to selectively bind to plasma membranes containing lipids, such as the mitochondrial and bacterial lipid cardiolipin, and phosphatidylinositol phosphates that are located on the inner leaflet of the mammalian cell membrane [64]. Binding results in the formation of oligomeric pores that subsequently kill mammalian cells, which is restricted to activated cells and does not kill bystander cells [64]. However, the relationship between GSDMD and secretion of IL-1β and IL-18 remains to be elucidated.

Figure 2.

Gasdermin-D, containing 276GVPAEG281 or 272FLTD275 in human and 273LLSD276 in mouse, is cleaved by caspase-1, -4, -5 or -11. The cleavage products result in a 31-kDa product of gasdermin-D. The activated Gasdermin-D N-terminal domain (31 kDa) then binds to phosphoinositides or cardiolipin in the plasma membrane to form multiple monomers that oligomerize to form a gasdermin pore.

Gasdermin-D: a direct executor of the canonical inflammatory pathway

In the past two years, researchers have demonstrated that GSDMD functioned downstream of inflammatory caspases [13, 68–70] and formed large permeability circular pores with ring diameters around 20 nm [69]. Several studies showed that caspase-1-mediated pyroptosis in iBMDM cells was induced by the very potent agonist of NAIP–NLRC4 inflammasome, namely Burkholderia thailandensis (LFn–BsaK), while GSDMD-/- iBMDM cells were resistant [40, 71, 72]. That suggests that GSDMD plays a critical role in caspase-1-mediated pyroptosis downstream of the NAIP–NLRC4 inflammasome. The N-terminal fragment of GSDMD can also drive the activation of the NLRP3-dependent caspase-1 inflammasome [40, 60, 73], possibly requiring potassium efflux caused by GSDMD-induced membrane pores [74]. GSDMD-mediated pyroptosis plays an important role in mature IL-1β release but does not affect its maturation [65]. In summary, caspase-1-mediated pyroptosis is induced by GSDMD, which may or may not involve the secretion of cytokines in innate immunity.

Gasdermin-D: a direct executor of the non-canonical inflammatory pathway

Yang et al. [71] and Zhao et al. [72] performed an unbiased genome-wide genetic screen using CRISPR-Cas9 technology to identify new components in LPS/caspase-4/11-induced pyroptosis. Similarly to mice lacking caspase-11, mice lacking GSDMD were remarkably resistant to LPS-induced endotoxemia compared to wild-type mice [60]. The study found that GSDMD-/- iBMDM cells regained the sensitivity to LPS electroporation when complemented with either human or mouse GSDMD. Both caspase-11 and caspase-4/5 were activated by direct binding of LPS to trigger pyropotosis via GSDMD, which would be the likely mechanism against cytosolic-invasive gram-negative bacteria.

Pyroprosis triggers pore-induced intracellular trap (PIT) and neutrophil extracellular trap (PET)

Viable bacteria can remain trapped within the cellular debris of pyroptotic macrophages, termed PITs. Jorgensen et al. [75] showed that pyroptosis had specific immunological functions to further promote uptake into neutrophils by efferocytosis in vivo. A more popular view is that both unbound and trapped bacteria are released by pyroptotic cells, which can be captured by phagocytic cells in tissues [31]. An advantage of the capture of pyroptosis-released bacteria by neutrophils is that these host cells are relatively resistant to pyroptosis in response to S. typhimurium infection and other inflammasome activators [76]. Pyroptosis can trigger the release of intracellular bacteria residing in macrophages, including S. typhimurium, Legionella pneumophila and Burkholderia thailandensis. These newly released bacteria can then be phagocytosed and killed by neutrophils via a mechanism dependent on the production of reactive oxygen species but independent of IL-1β and IL-18 (Figure 1). In mice that lack NADPH oxidase NOX2, neutrophils can undergo pyroptosis infected with Pseudomonas aeruginosa, suggesting that pyroptosis can be activated to compensate for lack of anti-microbial pathway activity [77]. NETs and PITs act independently of each other, as NET-defective mice (Mpo−/− and Elane−/−) are fully competent to clear bacteria trapped in PITs, and NET is not inhibited by the pan-caspase inhibitor z-VAD-fmk [75, 77–79].

The role of the gasdermin-family in pyroptosis

DFNA5 was discovered as a physiological substrate for caspase-3 and the N-terminal domain of DFNA5, which induces cell death by attacking the plasma membrane [16, 17, 66]. Similar pathways were observed in other gasdermin family members, including GSDMA, GSDMB, GSDMC, DFNA5 and DFNB59 [16, 66]. N-terminal domains of the gasdermin family, but not the full-length proteins or N-terminal domains, are efficiently and specifically precipitated by cardiolipin liposomes [66]. Whether other gasdermin family members exacerbate programmed cell death or mediate different cell death pathways is currently being investigated. Several diseases have been linked to gasdermin and whether pyroptosis plays an important role in these diseases requires further study (Table 1). Chao et al. [82] showed that the crystal structure of the GSDMB N-terminal domain was linked to IBD. Interestingly, caspase-3, -6 and -7, rather than the inflammatory caspases, cleaved GSDMB [40]. Rogers et al. [16] demonstrated that caspase3 and GSDMD together induced apoptosis, rather than pyroptosis, but pyroptosis could be induced when GSDMD was substituted for GSDME. Apoptosis mediated by apoptotic caspases (such as caspase-2, -3, -6, -7, -8 and -9 [87]), results in poly (ADP-ribose) polymerase and inhibitor of caspase-activated DNase to undergo proteolysis to cleave DNA into oligo nucleosomal DNA fragments [14]. This pathway differs from pyroptosis, although DNA cleavage could sometimes be triggered [14, 88]. Apoptosis has generally been considered an immunologically silent process, although emerging evidence indicates that apoptosis can be inflammatory when induced under certain circumstances and has roles in host defense against infection [89-91]. There may be possible interplays between apoptosis and pyroptosis.

Table 1.

Function of gasdermin family members

Gene name			Induces extensive pyroptosis	Activates caspases	Associated disease	References
Human	Mouse	Domains
GSDMA (gasdermin 1)	Gsdma1-3	Gasdermin N- and C-terminal domains	Yes	Unknown	Aalopecia and excoriation	[80, 81]
GSDMB (gasdermin-like /GSDML)	Gsdmc1-4	Gasdermin N- and C-terminal domains	Yes	Caspase-3/6/7	Asthma and inflammatory bowel disease	[82, 83]
GSDMD (DFNA5L)	Gsdmd	Gasdermin N- and C-termianl domains	Yes	Caspase-1/4/5/11	Infection diseases and sepsis	[31, 84]
GSDME (DFNA5)	Dfna5	Gasdermin N- and C-terminal domains	Yes	Caspase-3	Autosomal-dominant congenital deafness	[17]
DFNB59 (pejvakin)	Dfnb59	Gasdermin N-terminal domain	No	Unknown	Autosomal-recessive congenital deafness	[85, 86]

Pyroptosis of intestinal epithelial cells in IBD

With the expanding urbanization of newly industrialized countries, such as India and China, IBD, ‘ a Western disease’, is gradually growing into a global disease [92, 93]. Although the use of glucocorticoids, immunosuppressants, biological reagents and antibiotics reduces the symptoms of IBD, a minority of patients will eventually need surgical treatment. This has led to a considerable rise in health-care costs over time [94-97]. Prevention, diagnosis and effective therapy are the challenges. IBD is one of the most common abnormal inflammation diseases of the gastrointestinal tract. Intestinal epithelial cells (IEC) serve as a barrier between the host cell and the intestinal microbiota [98]. Excessive inflammasome expression in IEC results in an imbalance between the immune system and gut microbiota. Immune treatment strategies targeting pyroptosis in IEC of IBD may provide an alternative treatment and thereby reduce suffering and health-care costs. System homeostasis is maintained by pattern-recognition receptors (PRRs) [98, 99] and NLRs are important components that are dysregulated in IBD patients [100]. NLR inflammasomes are macromolecular platforms that sense cytosolic PAMPs and DAMPs, resulting in the maturation of IL-1β and IL-18, which is similar to the process of pyroptosis, as mentioned previously. A great deal of interest on whether or not pyroptosis plays a pivotal role in the development of IBD is currently being investigated.

Dextran sulfate sodium (DSS) mouse (ASC-/-) colitis models have a more severe phenotype than NLR knockout colitis mice [101-104], implying that loss of all inflammasomes is more detrimental to IBD pathogenesis than the loss of any individual sensor. NLR inflammasomes can attenuate gastrointestinal inflammation during experimental colitis [19, 105]. In addition, several studies have shown a significant increase in morbidity and mortality, weight loss, rectal bleeding and other pathological features of disease progression in mice lacking caspase-1 [101, 106, 107]. A series of experiments have shown that over-activation of inflammasome-initiating sensors results in intestine damage and intestinal inflammation. Strong upregulation of AIM2 and IFI16 inflammasomes in the mucosa of patients with active IBD [108] and A20-deficiency in macrophages significantly enhancing Nlrp3 inflammasome-mediated caspase-1 activation [109] demonstrates that the canonical inflammasome pathway and likely resulting over-activation of pyropotosis play a pivotal role in generation, progression and prognosis of IBD.

Similarly to the caspase-1 studies, caspase-11 has also been shown to attenuate gastrointestinal inflammation during experimental colitis in mice [103]. Research has demonstrated that caspase-11-/- mice appear to have more severe inflammation and epithelial cell damage in the colon than wild-type mice [103]. Caspase-11-/- mice were highly sensitive to acute DSS, independently of microbiome shifts [108]. However, the protective effects of caspase-11 appears to be only restricted to acute experimental colitis, as a lack of caspase-11 had only a minimal impact in chronic DSS and AOM + DSS models [103]. A similar phenomenon during canonical inflammasome pathway happens on non-canonical inflammasome pathway. In humans, the caspase-11 orthologues caspase-4 and caspase-5 were dramatically up-regulated in both CD and UC patient samples [109]. Further experiments revealed that caspase-1/5 expression was significantly higher in inflamed colonic but not ileal tissues. In summary, caspase-4/5/11 in the non-canonical inflammasome pathway may influence the development of inflammatory bowel disease [110] and pyroptosis induced by non-canonical inflammasomes in IBD should not be overlooked.

Pyroptosis presumably occurs in IBD, which is based on IEC interactions with immune cells located in the intestinal epithelium to affect cellular communication [111]. Inflammasomes, especially NLRP3, activated by inflammatory caspases play important roles in pyroptosis of IEC [50, 112–114]. Several lines of evidence have demonstrated that deficiencies in NLRP3 inflammasome components can protect mice from DSS-induced colitis [46, 115, 116]. The interaction between IEC integrin receptors and bacterial adhesin invasin provides the first signal for NLRP3 inflammasome activation in response to Yersinia enterocolitica infection, which is followed by the release of IL-18 and inflammatory responses at mucosal sites [117]. IL-1β secretion is not present in macrophages lacking NLRP3 and pharmacological inhibition of caspase-1 with pralnacasan can achieve a level of mucosal protection comparable to NLRP3 deficiency, indicating that DSS activates caspase-1 via the NLRP3 inflammasome pathway [118] (Figure 3). Other disease models of IBD also support the important role of NLRP3 in enteritis [119, 120] (Table 2).

Figure 3.

NLRP3 inflammasome activation is induced by bacteria and danger signals in intestinal epithelial cells (IEC). Nek7 is thought to be upstream of Nlrp3 but detailed mechanisms are unclear. Activated NLRP3 binds with cysteine protease caspase-1 to form a macromolecular complex, which directly cleaves gasdermin-D/B to initiate pyroptosis and inflammation. In addition, caspase-1 functions in the maturation and release of IL-1β and IL-18.

Table 2.

NLRP3 in disease models of inflammatory bowel disease

	Pathogenic factors	Caspases	Cytokines	References
NLRP3	Clostridium	Caspase-1	IL-1β	[119]
	Helicobacter pylori		IL-18	[121]
	Mycobacterium		IL-1	[120]
	Enterobacteriaceae		IL-1β	[122]
	2, 4, 6-trinitrobenzene sulfonic acid (TNBS)		IL-18, IL-1β	[123]
	Dextran sulfate sodium (DSS)		IL-18, IL-1β	[123]
	Bacterial muramyl dipeptide		–	[124]

In colonic macrophages, enterobacteriaceae induce intestinal inflammation via release of IL-1β mediated by NLRP3, which is similar to the pathogenic mechanism of Salmonella [125]. Our experiments demonstrated that the expression of caspase-1, NLRP3 and GSDMD was significantly higher in human inflammatory intestine tissues compared to normal intestine tissue (unpublished data). Schmid-Burgk et al. [49] found that macrophages with NLRP3 inflammasome activation would be rescued by targeting Nek7, which was thought to be upstream of Nlrp3 [126]. This study raises the hypothesis that Nek7 may be a key therapeutic target for pyroptosis. Inhibition of pyroptosis provides a potential therapeutic avenue to cure intestinal inflammation. Liu et al. [83] demonstrated that cholecalciterol cholesterol emulsion (CCE) was able to attenuate the degree of colitis in rat models by inhibiting pyroptosis, and observed the decrease of inflammatory responses (ASC, caspase-1, IL-1β, IL-18 and IL-6). CCE also provides an alternative therapeutic target for IBD patients and clinical trials have demonstrated its potential [83]. IEC infected with S. typhimurium undergo activation of the caspase-1 or caspase-11 inflammasome, resulting in physical entry of infected enterocytes from the intestine [119, 120, 127]. Hence, elimination of infected cells from tissues would be another mechanism by which pyroptosis eliminates invasive intestinal bacteria. Davis et al. [20] suggested that inhibition of IEC pyroptosis may be the therapeutic mechanism of action of mesalamine and corticosteroids.

Conclusions and future directions

Pyroptosis is a form of programmed cell death and takes place in monocytes, macrophages and epithelial cells. The typical catalytic action of inflammatory caspases (caspase-1/4/5/11) during pyroptosis differs from that of apoptosis, and the morphological characteristics of membrane rupture are also different [128]. Despite the morphological features being similar to necroptosis, pyroptosis has its own characteristics of inducing a specific size of membrane pores and does not depend on receptor interacting protein (RIP) 1 and 3 signaling pathway for its actions [129, 130]. Pyroptosis depends on GSDMD to form specifically sized pores on the cell membrane, although the exact mechanism on how it does this has not been fully elucidated [66]. It is speculated that inflammatory caspases and GSDMD play important roles in pyroptosis. Future studies to find inhibitors of the pyroptosis signaling pathway may provide therapeutic potential for relevant diseases induced by pyroptosis.

The etiology and pathogenesis of IBD still need to be fully characterized. However, numerous studies have demonstrated that alteration of intestinal flora affects the intestinal mucosal immunity and leads to the occurrence of IBD [131]. In various models of IBD, inflammatory caspases were found to be significantly elevated, suggesting that pyroptosis may be involved in the development of IBD [110, 132]. Several studies have demonstrated the existence of pyroptosis in IBD. Xiong et al. [133] demonstrated that CCE could attenuate experimental colitis by suppressing pyroptosis signaling. Therapeutic drugs for IBD such as mesalamine and corticosteroids may relate to the inhibition of pyroptosis in IEC [20]. The signaling mechanism of pyroptosis has yet to be fully deciphered when compared to apoptosis. In addition, whether pyroptosis plays a significant role in IBD is still unclear. Further research into pyroptosis signaling pathways is necessary to support its therapeutic target for IBD.

Conflict of interest statement: none declared.