Inflammasomes in epithelial innate immunity: front line warriors

Our epithelium represents a battle ground against a variety of insults including pathogens and danger signals. It encodes multiple sensors that detect and respond to such insults, playing an essential role in maintaining and defending tissue homeostasis. One key set of defense mechanisms is our inflammasomes which drive innate immune responses including, sensing and responding to pathogen attack, through the secretion of pro‐inflammatory cytokines and cell death. Identification of physiologically relevant triggers for inflammasomes has greatly influenced our ability to decipher the mechanisms behind inflammasome activation. Furthermore, identification of patient mutations within inflammasome components implicates their involvement in a range of epithelial diseases. This review will focus on exploring the roles of inflammasomes in epithelial immunity and cover: the diversity and differential expression of inflammasome sensors amongst our epithelial barriers, their ability to sense local infection and damage and the contribution of the inflammasomes to epithelial homeostasis and disease.

The epithelial innate immune system Our epithelium covers the surfaces of all our organs exposed to the external environment.As such the epithelial innate immune system is the first-in-line defense that our body employs to protect against infections, foreign invaders, and tissue damages [1][2][3][4].Traditionally, our epithelium was thought to act as a nonspecific defense mechanism; however, this dogma is increasingly being challenged.Our epithelial cells, once thought to be inclusive of just physical and chemical barriers, perform a key role in host immune responses inclusive of innate and even adaptive memory responses, enabling resistance to reinfection [5][6][7][8].These epithelial cells are more than a barrier, and are involved in sensing and responding to a wide range of Abbreviations ALR, AIM2-like receptor; ASC, apoptosis-associated speck-like protein containing caspase activation and recruitment domain; DAMPs, danger-associated molecular patterns; ESCRT, endosomal sorting complexes required for transport; FIIND, function-to-find domain; GBPs, guanylate-binding proteins; GI, gastro-intestinal; GoF, gain-of-function; GSDMD, gasdermin-D; HMGB1, high mobility group box 1; HPK, human primary keratinocytes; IL-1R1, type 1 IL-1 receptor; KSHV, Kaposi sarcoma-associated virus; LDH, lactate dehydrogenase; LTA, lipoteichoic acids; MAPKs, mitogen-activated protein kinases; NINJ1, Ninjurin-1; NLR, nucleotide-binding domain and leucine-rich repeatcontaining; NLRP6, NOD-like receptor family pyrin domain containing 6; OMVs, outer-membrane vesicles; PAMPs, pathogen-associated molecular patterns; pro-IL-18, pro-interleukin 18; pro-IL-1b, pro-interleukin 1b; PRRs, pattern recognition receptors; RLR, RIG-I-like receptors; RSR, ribotoxic stress response; SCC, squamous cell carcinomas; SNPs, single nucleotide polymorphisms; TLR, Toll-like receptors; UVB, ultraviolet B rays. potential threats, helping to contain or eliminate them.To do so, they employ a number of germline-encoded innate immune sensors, also known as pattern recognition receptors (PRRs) [9,10].PRRs are specialized proteins that play a crucial role in detecting potential threats to epithelial homeostasis by recognizing them directly or indirectly or by detecting alterations in cellular processes [11].PRRs can be broadly subdivided into two major classes based on their subcellular localization.Toll-like receptors (TLR 1, 2, 4-6) and the Ctype lectin receptors are transmembrane receptors that recognize specific molecular patterns found at the cell surface [1,[12][13][14].Intracellular receptors such as TLR 3, 7-9, RIG-I-like receptors (RLR), the AIM2-like receptor (ALR) and nucleotide-binding domain and leucine-rich repeat-containing (NLR) proteins detect threats inside the cell [15,16].A subset of these PRRs are capable of forming large multiprotein structures termed inflammasomes.Over the last few years, research has shown the inflammasome to be a wellregulated and important platform present in nonmyeloid cells, such as our epithelium.As such dysregulation of inflammasomes contributes to a spectrum of epithelial diseases.In this review, we will focus on the current understanding of the function of inflammasome sensors expressed in epithelial barrier tissues, guided by human genetics.

Epithelial inflammasomes
Tschopp and Martinon were the first to describe the inflammasome in monocytes, as a large complex capable of controlling the activation of inflammatory caspases within the cytosol [17].Since then, multiple lines of evidence have shown inflammasomes to play a crucial role in the innate immune response against most pathogens.Inflammasomes are primarily responsible for acting as a scaffold for the rapid conversion of caspase zymogens into their active species.This includes the well-studied caspase-1, but also the less-studied inflammatory caspases; caspase-11 (in mice) and -4/-5 (in humans) into enzymatically active proteases [17,18].Caspase-8, an apoptotic caspase, has also been shown to be activated by inflammasomes and to directly modulate inflammatory signaling in gut epithelial cells, subsequently contributing to an inflammatory response [19][20][21][22][23][24][25][26].These caspases are monomeric under their resting state and become activated by dimerization on the inflammasomes, which increases the local concentration of caspases [27].Based on artificial dimerization systems, dimerization is believed to drive their activation, although dimerization on their activating platform has not been fully investigated [28][29][30].Once dimerized, inflammatory caspases cleave themselves into their fully active forms and cleave defined substrates, including interleukin 1 (IL-1) family members (pro-interleukin 1b (pro-IL-1b), pro-interleukin 18 (pro-IL-18)), and gasdermin-D (GSDMD) [31].Both mature IL-1b and IL-18 are extremely potent cytokines that require gasdermin (GSDM) pores or cell lysis for release due to lack of the signal peptide targeting them to the conventional endoplasmic reticulum-golgi secretory pathway [32][33][34][35].
In the epithelium, GSDMD is the most studied mediator of inflammasome-derived IL-1 release.However it must be highlighted that mutations in murine gasdermin a1/gasdermin a3 and human gasdermin B genes have been linked to skin defects and asthma, respectively, suggesting their role in controlling epithelial inflammation [36][37][38].Furthermore, GSDMA, which is highly expressed in skin, is activated by the major skin pathogen Streptococcus pyogenes [39][40][41].GSDMD is processed at its linker region that connects the functional N-terminal fragment (GSDMD-NT) to an auto-inhibitory domain [42,43].Processing of GSDMD releases the GSDMD-NT fragment which subsequently forms pores in the plasma membrane through oligomerization [21,44].Once activated, GSDMD pores allow ions and water to enter the cell causing the cells to swell [45,46].Importantly, GSDMD pores also mediate electrostatic filtering which preferentially allows mature IL-1s to selectively be released over the proforms [47,48].Endosomal sorting complexes required for transport (ESCRT) and acid shingomyelinase control GSDMD pores through unique mechanisms [49].For the former, Ca 2+ influx associated with GSDMD pore formation recruits the ESCRT complex to the membrane to facilitate membrane repairs and negatively regulate pyroptosis.The latter mechanism is controlled by caspase-7, an apoptotic caspase also activated by caspase-1 [50].This mechanism is believed to delay cell death, enabling efficient cellular extrusion of infected epithelial cells while supporting tissue repair and integrity [51,52].
In general, GSDMD-activated cells undergo plasma membrane rupture, which enhance the release of alarmins and promote inflammation, leading to a lytic form of cell death termed pyroptosis [53].However, recent studies showed that GSDMD-dependent release of IL-1 cytokines can occur independently of pyroptotic cell death in macrophage cells, but the underlying mechanisms of this protection remain unknown [32,47,54].Another recent discovery showed that the transmembrane protein Ninjurin 1 (NINJ1) is required for pyroptotic plasma membrane rupture and subsequent release of the larger danger-associated molecular patterns (DAMPs) such as high mobility group box 1 (HMGB1) and lactate dehydrogenase (LDH) in macrophages [55,56].Currently, the involvement of NINJ1 in regulating epithelial DAMPs and cell death is unexplored.Membrane rupture is a double-edged sword, indeed while it can be beneficial for removal of pathogens or damaged cells, it could destabilize the epithelium by destruction of the structural integrity of the epithelium.
The purpose of activation of epithelial cell lysis and coupled pyroptotic cell death has been the focus of much discussion.Mechanistically, activation of pyroptosis in infected epithelial cells is thought to halt the replication of intracellular pathogens through the elimination of the infected cell [51,52].Moreover, pyroptosis could influence adaptive immunity, by releasing antigens into the extracellular milieu and through the impact of the cytokines released on adaptive cells (e.g.IL-18) [57,58].Pyroptotic cell death can also release other key inflammatory molecules such as IL-33, HMGB1, IL-1ɑ and eicosanoid lipid mediators [52,53,59,60].IL-1ɑ is thought to function as an alarmin and is constitutively expressed at barrier tissues such as our skin, lung, and gastro-intestinal tract [61].Notably, the IL-1ɑ precursor is biologically active and does not require caspase-1 processing for its activity, unlike IL-1b and its direct release triggers essential local inflammatory responses which promote epithelial wound healing and protective barrier function [61][62][63][64].Interestingly, deficiency of the type 1 IL-1 receptor (IL-1R1) leads to the development of an autoinflammatory skin disorder broadly termed as neutrophilic dermatosis, which is associated with cutaneous inflammatory lesions and granulocyte infiltration [65,66].More recent work shows that IL-1ɑ can additionally be cleaved into a smaller and biologically active protein in macrophages by proteases such as caspase-5, calpain, cathepsin G or granzyme B, yet the importance of cleaved IL-1ɑ in human epithelium is unexplored [67][68][69][70].Another important IL-1 family member shown to be secreted by GSDMD pores is IL-33.Secretion of IL-33 in airway epithelium promotes type 2 immune responses and allergic responses in lung epithelium and has been suggested to be released by an uncharacterized GSDMD cleavage product (p40 instead of p30) [71,72].To summarize, work to date highlights the importance of inflammasomes and downstream pyroptosis in protecting barrier tissues such as our skin, lungs and gastro-intestinal tract through removal of infected or damaged cells and alerting further immune cells to clear the risk.Mechanistically, pyroptosis can be executed in an inflammasome-dependent manner through canonical or non-canonical pathways; these pathways will be discussed briefly below (Fig. 1).

Canonical inflammasome
In terms of epithelial immunity, canonical inflammasomes CARD8, NLRP1, NLRP3, NLRC4, NLRP6, NLRP7, NLRP10, and AIM2 all have important roles in pathogen defense which are discussed further in the sections below.The assembly of the canonical inflammasome converts caspase-1 into a catalytically active enzyme through autoproteolytic, proximity-induced activation [31,73].Active caspase-1 is capable of cleaving the precursor cytokines pro-IL-1b, pro-IL-18, and GSDMD to induce pyroptotic cell death [48].Recruitment of apoptosis-associated speck-like protein containing caspase activation and recruitment domain (ASC) plays a key role for most canonical inflammasomes by helping to recruit caspase-1 [17,74].Structurally, ASC acts as an adapter molecule that bridges PYD-containing inflammasome receptors or CARD-containing inflammasome receptors to the CARD-containing caspase-1 using its own PYD or CARD domains [75][76][77].However, upon receptor activation, ASC also clusterizes to form a structure called the ASC speck, which is a macromolecular structure [77][78][79][80].Formation of this ASC speck is independent of caspase-1 activity but requires the oligomerization of ASC into a large insoluble aggregate [79,81].These ASC prion-like polymerizations act as a signal amplification mechanism for inflammasome-dependent cytokine production, providing a "signalosome" for robust and rapid activation of caspase-1 [77].A subset of canonical inflammasomes (containing a CARD domain) have been shown to activate caspase-1 directly in the absence of ASC [31,82,83].However, this has been demonstrated in a context of animal and cell lines deficient for ASC, and biological scenarios leading to this are currently unknown.

Non-canonical inflammasome
In addition to the set of canonical inflammasomes existing in epithelial cells, a less described noncanonical pathway, which utilizes caspase-11 in mice and caspase-4 and -5 in human cells has been shown to be critical during gram-negative bacteria, such as Escherichia coli, Salmonella, Pseudomonas, Helicobacter and Legionella.Similar to caspase-1, these caspases are initiator-like caspases and are recruited onto the inflammasome through their N-terminal CARD domain, allowing their dimerization and activation [84].Although structural information on the noncanonical inflammasome is lacking, biochemical studies suggest that dimerization is sufficient for the activation of caspase-4 and -11 [29].Caspase-5 biochemical features have been poorly studied.In general, inducible expression of non-canonical inflammasome components is a prerequisite for its activation [85,86].While caspase-4 is expressed constitutively in most human cells, caspase-5 and -11 require priming for full expression [86].Caspase-4/-11 have important documented roles in epithelial immunity [87][88][89][90].However, the involvement and function of caspase-5 in general, especially in epithelial cells is less clear and subject to investigations.Whereas the canonical inflammasome recognizes a range of DAMPs and pathogen-associated molecular patterns (PAMPs), the non-canonical inflammasome is specialized in recognizing lipopolysaccharides (LPS; a component of the gram-negative bacterial cell wall), from intracellular bacteria that reaches the cytosol or from bacterial structures called outer-membrane vesicles (OMVs) [90][91][92][93][94].In addition, OMVs can be released from extracellular gram-negative bacteria such as E. coli and be detected by the non-canonical inflammasome intracellularly [95][96][97].Initially, caspase-4/-11 has been shown to directly recognize the lipid A counterpart of LPS, in spite of its highly hydrophobic nature [90].Recent evidence suggests that, at least in epithelial cells, recognition of LPS is supported by a family of large GTPases called Guanylate-binding proteins (GBPs) which leads to the recruitment of caspase-4 [98][99][100][101].These IFN-inducible GTPases are novel inflammasomes Much of the research performed to date explores inflammasome function in traditional immune cells of myeloid tissue origin, such as monocytes, macrophages, and dendritic cells.However, if we examine the expression of inflammasome sensors in the epithelium, in the Human Protein Atlas RNA-seq database [111], we find that the majority of expressed inflammasome sensors have restricted human epithelial tissue distributions (Fig. 2).For instance, NLRP1 and NLRP10 are found to have the highest expression level in skin, NLRP6 and AIM2 are highest in the gut and NLRC4 and GBP1 are highest in the lung epithelium.NLRP3, NLRP7 and NLRP12 have very little expression in the developed epithelial organ and IFI16 and CARD8 are widely expressed at a high level and low level, respectively, in the majority of epithelial organs.In the next sections, we will focus on the current understanding of inflammasome activation at different barrier sites and the genetic contribution of inflammasomes to disease pathologies and in maintaining cellular homeostasis (Table 1).Given the complexity of inflammasome activation and the number of components involved, it's not surprising that mutations in inflammasome components have been associated with human diseases and can contribute to a range of chronic inflammatory diseases and autoinflammatory disorders (Table 2).

Genetic evidence from mutations and SNPs
The skin is an essential outer barrier of the human body, highlighted by rare genetic skin fragility conditions, where the patients experience multiple complications including susceptibility to infection as a result of reduced barrier function [112].Its outermost compartment consists of the epidermis which is largely made from densely packed keratinocytes that have undergone a specific differentiation program, termed cornification, and contains few traditional immune cells [113,114].However, the keratinocytes of the skin express genes involved in immune regulation, furthermore and perhaps strategically, terminal differentiation increases the number of innate immune-related genes to include genes such as PYDC1, Caspase-14 and NLRP10 whose function in the epidermis remains elusive [115,116].That said, under homeostatic conditions, NLRP1 inflammasome appears to be the dominant inflammasome sensor in skin epithelial cells, exemplified by multiple lines of evidence.Gain-of-function (GoF) mutations in the NLRP1 inflammasome sensor cause a rare skin inflammatory and cancer susceptibility disease in patients [117,118].These documented GoF mutations lead to spontaneous NLRP1 inflammasome activation and paracrine IL-1 signaling in keratinocytes from these patients [117,118].In addition to rare skin diseases, NLRP1 has also been linked to common inflammatory skin diseases; atopic dermatitis and psoriasis [119][120][121].Evidence suggests that IL-1 is responsible for maintaining a pro-inflammatory response in the keratinocytes of atopic dermatitis and psoriasis patients, which prompted a clinical trial to determine the efficacy of anakinra for a rare form of psoriasis [122][123][124][125].In humans, a number of NLRP1 single nucleotide polymorphisms (SNPs) associated with autoimmune disease also affect the skin, including vitiligo which is an autoimmune disease characterized by loss of melanin, and Addison's disease which is characterized by skin hyperpigmentation [126][127][128][129][130].Although the mechanism connecting NLRP1 to vitiligo is unknown, it is believed that activation of NLRP1 and subsequent secretion of IL-1b contribute to the loss of melanin in skin.Furthermore, haplotypes of NLRP1 which have been shown to secrete increased levels of IL-1b, are associated with vitiligo, leprosy and malignant melanoma [131,132].
As well as skin inflammatory diseases, NLRP1 has also been connected to skin cancers.NLRP1 GoF mutations predispose patients to the development of a common type of skin cancer, squamous cell carcinomas (SCC) yet paradoxically NLRP1 expression is suppressed in other SCC tumors and immortalized cell lines [117,133,134].Kaposi sarcoma-associated virus (KSHV) is the causative agent of Kaposi's sarcoma (KS) which is a type of cancer that can form masses in the skin, and was recently shown to activate human NLRP1 inflammasome [135].KSHV patients expressing a non-functional NLRP1 variant exist suggesting that NLRP1 may be required for KSHV immunity [135,136].These data suggest that NLRP1 could play a dual role as both a tumor suppressor and an oncogene during skin cancer development.
Aside from NLRP1, other inflammasomes have been linked to skin disease however these sensors are typically not constitutively expressed in keratinocytes and require a priming step to become activated.For instance, type II interferon-primed AIM2 inflammasome activation has been linked with psoriasis, through its ability to sense cytosolic DNA [137,138].Interestingly, AIM2 inflammasome has also been shown to participate in immune memory of epithelial cells [58,139].Thus evidence of our epithelium keeping memories of inflammation are starting to be discovered.By remembering inflammatory breaches, epithelial cells would be better equipped to deal with the next infectious episode [140,141].

Mechanisms of inflammasome activation
Due to the expression of NLRP1 being so high in the epidermis, there has been a search for activators of NLRP1 in the skin.Ultraviolet B rays (UVB) were first identified by Fenini et al., to induce NLRP1 inflammasome activation in human primary keratinocytes (HPK) however only recently was the mechanism by which UVB activated NLRP1 understood [142,143].Structurally, NLRP1 contains a function-tofind domain (FIIND) which undergoes autoproteolysis resulting in the N-terminal and C-terminal domain being covalently bound when inactive.Activation occurs when the N-terminal fragment is degraded by the proteasome, liberating the C-terminal fragment to form the active inflammasome complex [144].UVB was discovered to indirectly activate NLRP1 through phosphorylation of the NLRP1 disordered linker region, resulting in NLRP1 N-terminal fragment degradation thus liberating the C-terminal for inflammasome activation.NLRP1 was shown to be phosphorylated by mitogen-activated protein kinases (MAPKs) ZAKa and p38 which leads to NLRP1 inflammasome activation and IL-1 release in response to UVB [145,146].For a long time, it has been known that nucleic acids can absorb UVB causing photolesions in the RNA triggering the ribotoxic stress response [147][148][149][150].However, only recently in 2020 was the kinase responsible for detecting and responding to ribotoxic stress discovered as ZAKa, which is a MAPK activated by the ribotoxic stress response (RSR) pathway [151][152][153].ZAKa was found to sense the stalling and collisions of ribosomes by direct binding of its ribosome-binding domains [151,152].Once ZAKa is activated, it auto-phosphorylates itself and phosphorylates NLRP1 and MAPKs p38 and c-Jun N-terminal kinase (JNK) and downstream proteins involved in apoptosis and inflammation [151,152].More recently, bacterial toxins, dsRNA and the NLRP3 agonist nigericin are shown to induce NLRP1 activation through ribosome stalling and activation of the ZAKa/p38 pathway, suggesting that targeting this pathway may serve as a target for a broad spectrum of diseases [154][155][156][157][158][159].More recently, cytosolic peptide accumulation in combination with reductive stress was found to strongly activate NLRP1 activation.It will be important to develop assays to assess reductive stress and peptide accumulation during the maintenance of skin homeostasis while under pathogen attack [160,161].NLRP10 is a new inflammasome sensor in the skin that has been recently described.Pr ochnicki et al. [162] identify a compound, a phospholipase C activator called 3m3-FBS, which causes activation of NLRP10 inflammasome in keratinocytes.NLRP10 is recruited to 3m3-FBS destabilized mitochondria leading to the formation of a canonical inflammasome with ASC, caspase-1 and GSDMD leading to IL-1b release, in a similar manner to mouse NLRP10 in the gut [162,163].Interestingly, in the skin NLRP10 expression is induced by terminal differentiation and could potentially contribute to barrier function of skin tissue.In line with this hypothesis, a missense variant in NLRP10 (R243W) was identified in a genome-wide association study to be associated with atopic dermatitis [164].Given its connection with atopic dermatitis, it is tempting to speculate that NLRP10 is a bacterial sensor of pathogenic skin bacteria such as Staphylococcus aureus or S. pyogenes which are commonly found in patients with atopic dermatitis.In addition to NLRP10, the AIM2 inflammasome has also been shown to cause IL-1b and IL-18 release in keratinocytes infected with human papilloma viruses (HPV), a double-stranded DNA virus that can infect skin epithelia [165].

Genetic evidence and mechanisms of activation
The airway epithelium constitutes a stratified layer of cells that lines the respiratory tract, encompassing the nasal passages, trachea, and bronchi [166].Functionally, it serves as a pivotal interface and fulfills essential roles in immunological defense safeguarding the respiratory system against the intrusion of exogenous particles, pathogens, and pollution [167,168].A growing body of literature suggests that respiratory pathogens, in particular respiratory viruses, activate inflammasomes that contribute to viral clearance and tissue regeneration [169].As such, a SNP has recently been identified in the non-canonical inflammasome effector caspase-4 (R344W) that increases the susceptibility to Burkholderia pseudomallei and leads to reduced bacterial clearance, causing a persistent infection in the lungs [170].This mutation impacts caspase-4 dimerization and dampens its activation, resulting in a decreased release of the cytokine IL-18.The release of IL-18 is essential to bridge antibacterial adaptive immune responses mediated by NK and T cells during gram-negative infections, such as nosocomial Melioidosis associated with ECMO [170].Administration of IFNg reverted the caspase-4 mutation's effect on the infection and enabled full clearance of the infection [170].The data suggest a role for the non-canonical inflammasome in airway defense.
As well as being highly expressed in skin, the canonical inflammasome NLRP1 is also expressed in airway epithelia.As such, homozygous gain-of-function NLRP1 mutations have been identified in patients with recurrent growth of papillomas in the airways [171].These patients also have a history of infection with HPV, suggesting a role for NLRP1 in viral infection and respiratory disease.A common NLRP1 SNP (M1184V) that has been associated with asthma was described to increase cleavage and subsequent activation within the NLRP1 FIIND domain, further suggesting a role from NLRP1 in airway defense [172][173][174][175].The airway epithelium and in particular the lung epithelial cells in the respiratory tract activate the inflammasome in response to a diverse range of stimuli including both pathogens and sterile agents such as SARS-COV-2, bacteria, fungus, and particulate matter [176][177][178][179].These studies have clearly shown that IL-1b secretion occurs in response to these stimuli, likely through an inflammasome-specific inflammatory component, however modeling these complex respiratory tract pathologies is challenging to perform and study.In 2020, the first physiological activators of NLRP1 were described as the 3C proteases from enteroviruses that were reported to cleave NLRP1, such that the Nterminal fragment is degraded so that the C-terminal domain can be liberated to form the inflammasome [180][181][182][183]. Later, Plan es et al. [184] show that human NLRP1 is cleaved at the Q333 site by multiple coronavirus 3CL proteases, triggering inflammasome assembly and cell death.The same group also shows IL-1 secretion and hypersensitivity to Exotoxin A from Pseudomonas aeruginosa through ribotoxic stress-dependent NLRP1 inflammasome activation in nasal epithelial cells obtained from Cystic Fibrosis patients [155].These observations connect NLRP1 to the deleterious effects of P. aeruginosa infections of Cystic Fibrosis patients and could also extend to patients with Chronic Obstructive Pulmonary Disease and their increased pulmonary inflammation and epithelial disruption.

Mechanisms of inflammasome activation
The gastro-intestinal (GI) tract is a complex tubular structure that encompasses the oral cavity, esophagus, stomach, and intestine and is the primary site for digestion and absorption of nutrients.Additionally, the GI tract plays a critical role in immune defense and microbial balance which protects against pathogen invasion from food-borne pathogens but also maintains a healthy gut microbiota.The GI tract is subject to infections by numerous enteric pathogens, including Listeria, Salmonella, enterovirus and Candida.Inflammasomes within the GI tract are instrumental in responding to such pathogens using a range of mechanisms and mediate host protection against gramnegative bacteria including the secretion of IL-1b and IL-18 during pyroptosis [81,185,186].Activation of inflammasomes in the GI tract has a range of key consequences that limits the spread of infection including pyroptosis, maturation of cytokines and cellular extrusion of dying cells [51,52].These events limit the pathogens infiltration to the lamina propria of the intestine.Expression patterns of inflammasomeforming PRR across the GI tract impact the immune responses.The oral epithelium is heavily exposed to particulate matter, as well as bacterial and fungal agents.One of the major pathogens of the oral cavity is Porphromonas gingivalis which is responsible for periodontitis [187,188].Periodontitis has been linked to increased expression of NLRP3 and AIM2 in the oral epithelium which leads to increased IL-1b gene expression [189,190].Interestingly, in human periodontium and gingiva, NOD-like receptor family pyrin domain containing 6 (NLRP6) seems to also impact on the homeostasis of the oral cavity [191,192].These increased levels of IL-1b due to inflammasome activation have been hypothesized to contribute to oral and esophageal carcinogenesis [189,193].Currently, most studies of inflammasomes in the GI tract have focused on the innate immune responses of the intestinal epithelial cells, of which the NAIP-NLRC4 and NLRP3 inflammasome has been studied in the most detail [52,[194][195][196][197][198][199].In response to Salmonella, NAIP-NLRC4 inflammasome is activated and results in expulsion of the pyroptotic body out of the intestinal epithelial cell layer into the lumen [52,195,197].Timelapse organoid studies show that intestinal epithelial cells undergo lysis within the epithelium before expulsion, suggesting that inflammatory signals are relayed to the underlying tissue [52].Gain-of-function mutations in NLRC4 have been identified in newborns and children with symptoms ranging from skin inflammation to enterocolitis [200][201][202][203].The pro-inflammatory impact of these mutations was mainly associated with high levels of IL-18.In the gut, maturation of IL-18 by inflammatory response is important for pathogen's clearance.IL-18 is expressed endogenously and does not require priming, making it an important cytokine across the GI tract.IL-18 can be directly cleaved and activated by both canonical and non-canonical inflammasome components caspase-1 and -4 [100,204].Additionally, the NLRP3 inflammasome has been shown to be activated in the intestinal epithelium by a wide array of gram-positive and gram-negative bacteria, including E. coli which mainly triggers NLRP3 through potassium (K+) efflux [198,199].
In addition to NAIP-NLRC4 and NLRP3 expression, in the intestinal epithelial cells, there is high expression of the NLRP6 inflammasome.NLRP6 is expressed in the goblet cells where it is suggested to play a role in regulating the gut microbiome by impacting the goblet cell mucus secretion [205,206].NLRP6 is thought to contribute to GI inflammatory conditions but also may regulate extra-intestinal diseases.The NLRP6 has been shown to bind directly lipoteichoic acids (LTA) from gram-positive pathogens (e.g., Listeria) and to trigger ASC recruitment and caspase-11 activation [207].Caspase-4, which is part of the non-canonical inflammasome, has been shown to indirectly detect Salmonella Typhimurium through the LPS receptor GBP1 [99][100][101].As such, GBP1 binds to the bacterial surface of Salmonella Typhimurium and further recruits GBP2-4 to create an oligomeric structure which is then capable of activating caspase-4.Certain gram-negative bacteria have evolved mechanisms to avoid non-canonical inflammasome detection.These include modification of LPS acyl chains, the usage of effector targeting caspase-4 directly and effectors that impact its signaling [208][209][210][211][212][213].In mice, the dependence on the non-canonical inflammasome has been demonstrated, as caspase-11 knockout mice are more susceptible to infection from intracellular bacteria such as Burkholderia thailandensis, Shigella flexneri and Salmonella Typhimurium [88,93,214,215].Furthermore, in mice GBP2 and GBP3 are also thought to be capable of directly sensing LPS, expanding the repertoire of bacterial sensors in the GI tract [216].

Outlook and perspective
Pioneer work by Janeway and colleagues opened the door to an impressive array of mechanisms controlling defense to pathogens and initiating adaptive immune responses [3].Although discovered more than 20 years ago, the complexity of inflammasomes and their roles in innate immune responses is being unraveled through increasing scientific efforts [17].Inflammasome sensors and their effectors have been characterized extensively in myeloid cells but much effort remains to achieve a similar understanding in epithelia.As cell-specific and tissue-specific functions have emerged for defined inflammasomes, it is highly likely that inflammasome sensors and their caspases have unique mode of regulation in epithelial tissues, unveiling new strategies to target them in diseases.For example, NINJ1 has emerged as an exciting pharmaceutical target against inflammasomes, however its role in epithelial tissues is currently unclear [56].The identification and characterization of novel inflammasome-forming PRR in the skin, such as NLRP10, is likely to identify novel mechanisms controlling innate immune responses in epithelium [162].Furthermore, recent advances in human genomics have increased our understanding of human inflammasomes and highlighted differences between murine and human inflammasomes.Learning how tissue-specific mutations and a dysregulated innate immune system impact infections and drive autoinflammatory diseases will further our understanding of inflammasomes in health and disease.
In addition to genetic and cellular host determinants, new microbial communities have emerged as strong modulators of tissue homeostasis and provide physical barriers that impact pathogens' access to epithelium.Microbiomes vary drastically between the skin, gut, and the airway and have a strong impact on the epithelial immune environments [217].Modulation of inflammasomes by microbes in the skin is an evolving area of research.In the skin, microbiota impact inflammatory behavior of bacteria such as Cutibacterium acnes, by secreting chemicals (e.g., Porphyrin, propionate) that can directly modulate inflammasomes such as the NLRP3 present in the sebocytes [218].These changes in microbiota occurring in individuals with acne have been shown to have a direct impact on skin inflammation.Similarly, the gut microbiota secrete a range of metabolites that modulate the inflammatory network with the subsequent activation of inflammasomes potentially impacting the composition of the microbiome and modulating the secretion of antimicrobial peptides.Furthermore, inflammasome sensors, such as NLRP6, have also been suggested to modulate mucus secretion in the gut, directly impacting the microbiome biophysical environment [218,219].Future research into interactions between the host microbiota and our inflammasomes will teach us how our immune system differentiates between dangerous pathogens and beneficial symbiotic microbes [206].Finally, as inflammasomes are crucial effectors against infections, the identification of everevolving pathogen defense strategies will continue to emerge and translate toward the development of novel inflammasome-targeted interventions.

1336FEBS
Letters 598 (2024) 1335-1353 ª 2024 The Authors.FEBS Letters published by John Wiley & Sons Ltd on behalf of Federation of European Biochemical Societies.

Fig. 1 .
Fig. 1.Schematic overview of the human canonical and non-canonical inflammasome.The canonical inflammasome is centralized around the activation of caspase-1 which cleaves GSDMD for it to assemble pores in response to a broad range of intracellular PAMPS and DAMPS sensed by inflammasome-forming PRR.The non-canonical inflammasome pathway specifically senses intracellular LPS, resulting in the activation of caspase-4/5 which cleaves GSDMD for it to assemble pores.In humans, GBPs act as an intracellular LPS sensor that recruits caspase-4/5 to activate non-canonical inflammasome signaling.DAMPS, damage-associated molecular patterns; GBP, guanylatebinding protein; GSDMD, gasdermin D; PAMPS, pathogen-associated molecular patterns; PRR, pattern recognition receptor.Figure created with BioRender.com.

Fig. 2 .
Fig. 2. Relative expression levels of different inflammasome sensors in epithelial tissues.Dataset consists of normalized expression (nTPM) levels for each tissue created by combining the internally generated Human Protein Atlas RNA-seq data and Genotype-Tissue Expression (GTEx) transcriptomics datasets from the Human Protein Atlas [111].

Table 1 .
Table of known pathogen blocking strategies performed by epithelial inflammasomes.Table of epithelial inflammasomes and their known pathogen detection strategies used to mediate epithelial host protection, as discussed in the text.

Table 2 .
Table of patient mutations/SNPs in epithelial inflammasome.Table of disease, associated sensor and epithelial organ affected which are discussed in the text.FIIND, function-to-find; LRR, leucine-rich repeat; PYD, pyrin domain.
FEBS Letters 598 (2024) 1335-1353 ª 2024 The Authors.FEBS Letters published by John Wiley & Sons Ltd on behalf of Federation of European Biochemical Societies.