Interleukin‐1 family cytokines at the crossroads of microbiome regulation in barrier health and disease

Recent advances in understanding how the microbiome can influence both the physiology and the pathogenesis of disease in humans have highlighted the importance of gaining a deeper insight into the complexities of the host‐microbial dialogue. In tandem with this progress, has been a greater understanding of the biological pathways which regulate both homeostasis and inflammation at barrier tissue sites, such as the skin and the gut. In this regard, the Interleukin‐1 family of cytokines, which can be segregated into IL‐1, IL‐18 and IL‐36 subfamilies, have emerged as important custodians of barrier health and immunity. With established roles as orchestrators of various inflammatory diseases in both the skin and intestine, it is now becoming clear that IL‐1 family cytokine activity is not only directly influenced by external microbes, but can also play important roles in shaping the composition of the microbiome at barrier sites. This review explores the current knowledge surrounding the evidence that places these cytokines as key mediators at the interface between the microbiome and human health and disease at the skin and intestinal barrier tissues.

Appropriate colonization with commensal microbiota in early life is critical in promoting healthy physiological development [5].This underscores the hypothesis that there are key developmental windows during which healthy microbial colonization is important for human health and any perturbations that may arise in this process could play an important role in the development of complex diseases later in life.The timing of such developmental windows is somewhat controversial.Although there is clear evidence that the maternal microbiome can influence offspring physiology [6][7][8], the idea that the microbiome can directly influence foetal development has recently been challenged [9].Most recent evidence indicates that initial exposure takes place at birth and it is interesting that, the mode of delivery is also known to shape the microbial landscape of the new born [9][10][11].Regardless of timing, a key aspect of this interplay between the host and external microbes is the appropriate education of the immune system to tolerate commensal microorganisms, while retaining the ability to fight infection.Indeed, exposure to commensals is a critical factor in shaping both the innate and adaptive immune responses not just at tissue barrier sites, but also systemically [12,13].As a result, any changes in 'normal' healthy colonization, such as a reduction in the diversity of microbe exposure, could likely have consequences for the regulation of inflammatory responses later in life [14].Added to this developmental role, it is becoming increasingly evident that the microbiome is also significantly altered in many chronic inflammatory disease settings.This is likely of particular relevance to diseases affecting major barrier sites such as the skin or gut.

The skin
The skin is the largest barrier site acting as a physical barrier against external threats, while also regulating body temperature and hydration.The microbial composition of healthy human skin is dominated by bacteria (~70%), followed by fungi (1-5%), and is the most diverse during early life, before stabilizing during the transition to adulthood [15][16][17].In order to preserve a healthy symbiosis between the host and skin resident commensal microbes, a tightly regulated immune tolerance is maintained, involving both innate (keratinocyte and dendritic cell [DC]) and adaptive (T cell) immune cell subsets [18].Defined skin resident DCs sample antigens derived from commensal species at the epidermis.This may occur through direct sampling through dendrite exposure at the skin surface, or through the diffusion of antigensat locations where the keratinocyte barrier is more permissive such as hair follicles.Under homeostatic conditions, skin resident DC continuously migrate towards draining lymph nodes whereupon they stimulate commensal specific T cell response which return to skin and reinforce barrier function [19].Pertubations in this host-microbe dialogue, such as may occur under conditions of dysbiosis of the skin microbiome, may have profound implications on this finely tuned balance and accumulating evidence also suggests that dysbiosis of the microbiota is associated with various skin disease conditions.For example, acne vulgaris is a complex multifactorial skin disease in which the skin microbiome is considered a major factor.Cutibacterium acnes, a commensal bacteria normally found in healthy human skin, is also implicated in the pathophysiology of acne [20].It is now generally accepted that rather than an increased abundance of this pathogen, one of the triggers for this disease is an alteration in the C. acnes strains that are present in the skin, together with a general microbiome dysbiosis [21][22][23].Similarly, Atopic dermatitis (AD) is a chronic inflammatory disease of the skin characterized by an impairment of the skin barrier, increased type 2 inflammation and dysregulation of the skin microbiome [24,25].Skin bacterial communities differ between healthy and patients with AD, with an overrepresentation of Stahpylococcus aureus, together with decreased overall bacterial diversity [26,27].Interestingly, disease severity, in AD, also positively correlates with a decrease in the alpha-and beta-diversity of skin bacterial communities [28][29][30][31] suggesting that microbial dysbiosis contributes to disease pathogenesis.
Psoriasis is a further common chronic inflammatory disease of the skin in which the microbiome has been implicated [32,33].While it is accepted that the skin microbiota is profoundly altered in psoriasis, these changes are not as well defined as they are in AD, where, as described above the importance of S. aureus colonization is well established [26,27].Conflicting results have been observed in studies that have analysed the microbiome in the psoriatic skin.While some of these studies found that Actinobacteria species were underrepresented in the psoriatic skin in comparison to healthy controls [15], some others observed that the abundance of these bacteria was increased in the lessional skin of psoriasis patients [34,35].Such conflicting results likely reflect the complexities of skin microbiome analysis, including historical differences in methodologies used, as well as sample sites analysed.

The gut
Similar to the skin, the gastrointestinal tract (GIT) has evolved to allow the colonization of micro-organisms that form the intestinal microbiota.The vast majority of the gut microbiota is composed of bacteria, specifically those belonging to the Firmicutes, Bacteriodetes, Actinobacteria and Proteobacteria phyla [36].These organisms are in constant dialogue with the intestinal barrier, which is primarily composed of a mucus layer, intestinal epithelial cells (IECs) and subepithelial intestinal immune cells.A cross-talk is established between the intestinal barrier and gut resident microbes, which in part takes place through the secretion of bacterialderived products such as short-chain fatty acids (SCFAs), trimethylamine N-oxide (TMAO) or indole-3-acetic acid (IAA), among others [37,38].Direct recognition of the bacteria by intestinal epithelial cells, and innate immune cells, is also an important feature in this host-microbiome dialogue.Intestinal DCs are key players in the establishment of immune tolerance in the healthy gut, in part by mediating the expansion of anti-inflammatory regulatory T cells (Tregs) [39][40][41][42].As IECs are in constant contact with luminal microbes, a finely tuned regulation is required to favour tolerogenic mechanisms in response to commensal bacteria.Compartmentalization of surface pathogen recognition receptor (PRR) expression to the basal surface of IECs is one such strategy through which this is achieved [43,44].Apart from immune tolerance, IECs play a major role in forming the intestinal epithelial barrier.While facilitating the absorption of water, ions and nutrients, tight junctions (TJs) between these cells provide an impermeable barrier to restrict the entry of gut microbes and microbe-derived products to the submucosal layers of the GI tract [45].
Disruption of intestinal barrier integrity is a central characteristic of the pathophysiology of gut diseases such as inflammatory bowel disease (IBD), celiac disease or diarrheic infection, but also systemic metabolic diseases such as obesity, non-alcoholic fatty liver disease (NAFLD) or diabetes [46][47][48].When the permeability of the intestinal barrier is compromised, luminal microbes can penetrate into the lamina propria, disrupting immune tolerance and initiating tissue inflammatory responses [49].As a consequence, bacterial pathogen-associated-molecular patterns (PAMPs), such as lipopolysaccharide (LPS), access the bloodstream establishing low-grade systemic inflammation, a phenomenon known as endotoxemia [50][51][52].The link between endotoxemia and increased intestinal barrier permeability is further reinforced by the observation that LPS, as well as markers of endotoxin-signalling cascade activation, are found in the serum of patients with IBD with both Crohn's disease (CD) or ulcerative colitis (UC) [53][54][55][56][57][58].Moreover, in CD, LPS levels are positively correlated with disease severity [59].This is perhaps unsurprising, given that increased intestinal barrier permeability is a hallmark of IBD.However, the primary cause for intestinal leakage in IBD, and other inflammatory intestinal diseases is not clear.
It is generally accepted that patients with IBD present with an altered composition of bacterial communities in the gut compared with healthy individuals [60][61][62].The major reported changes observed among intestinal bacterial communities in IBD include higher levels of Proteobacteria and an under-representation of Bacteroides, Eubacterium and Firmicutes.Firmicutes include SCFA-producing bacteria, such as Faecalibacterium prausnitzii, which have an anti-inflammatory effect [63,64] indicating that the bacterial landscape in IBD is directed towards "pro-inflammatory" phenotype.A similar picture has been observed in rodent models of IBD, wherein experimentally induced colitis caused a shift in the composition of gut bacterial communities in mice and rats towards a pro-inflammatory microbial signature [65][66][67][68].Although resident intestinal immune cells are relatively hyporesponsive to bacterial-derived products, such as LPS, in order to avoid a hyperactivated inflammatory response against commensal bacteria [69], it has been also demonstrated that the grade of TLR4 activation by bacterial LPS is species-specific.Stephen et al. compared the effect of LPS derived from different gram-negative bacteria commonly found in the gut, and demonstrated that the grade of TLR4 activation, and intestinal barrier breakdown, was different depending on the bacterial strain of LPS origin.From the studied species, the greatest impact on the intestinal barrier was associated with Serratia marcescens-derived LPS [70].Interestingly, S. marcescensis over-represented in the gut of IBD patients [71].Such observations add to the evidence indicating that gut microbial dysbiosis plays an important role in driving the pathogenesis of IBD.

IL-1 family cytokines and microbiome regulation at barrier sites
The exact mechanisms through which microbial dysbiosis at barrier sites is implicated in various diseases is not well understood, but is thought to involve dysregulated inflammatory responses in the barrier tissue microenvironment [72].Inflammation at barrier sites is a crucial response to external stimuli (pathogens, injury and toxins), coordinated by immune and parenchymal cells, in a dialogue involving soluble mediators such as cytokines.One such subset of cytokines, which have emerged as critically important mediators of inflammation and homeostasis at the skin and intestinal barrier sites, are the Interleukin-1 family of cytokines [73].Recent discoveries have identified specific roles of the broader IL-1 family, which comprises the IL-1, IL-18 and IL-36 subfamilies, in the interplay between the host and external microbiome particularly in the context of inflammatory diseases of the skin and gut.

IL-1 subfamily
The IL-1 subfamily includes the proinflammatory cytokines IL-1a, IL-1b and IL-33, and the antiinflammatory IL-1 Receptor antagonist (IL-1Ra).The two isoforms of IL-1, IL-1a and IL-1b, bind to the same transmembrane receptor complex, formed by IL-1R1 and the IL-1R accessory protein (IL-1RAcp) and have similar biological properties, however, they also share some dissimilarities [74][75][76].IL-1a is constitutively expressed by epithelial and endothelial cells and it is produced as a functional cytokine ready to be secreted from necrotic cells, functioning as an "alarmin".Furthermore, it can be upregulated under cellular stress or inflammatory conditions [77].Although the full-length form of IL-1a accounts for signalling functions itself, the processing of IL-1a by calpain results in a 17-kDa fragment that shows an increased affinity for its receptor [78,79].IL-1a is considered a dual cytokine, due to its role as an extracellular cytokine but also as an intranuclear transcriptional regulator [80].In contrast, IL-1b is expressed mainly by haematopoietic cells under a proinflammatory stimulus.The precursor form of IL-1b accumulates in the cytosol and needs to be cleaved by the proteolytic enzymes caspase-1/À11, as part of a signalling cascade orchestrated by the inflammasomes [81,82].The other member of the IL-1 subfamily, IL-33, signals through the ST2 receptor, which associates with IL-1RAcP and induces MyD88-dependent signalling [83].IL-33 is expressed both constitutively, and can be induced, predominantly in stromal cells at barrier sites.Similar to IL-1a, IL-33 is a dual alarmin that can act extracellularly but also inside the nucleus as a gene transcription regulator [84].In contrast to IL-1b, IL-33 is deactivated by caspases, and its biologically active form is the full-length protein [85,86].

IL-1 subfamily and the skin microbiome
The first step towards the establishment of immune tolerance in the skin is bacterial recognition by keratinocytes through TLRs and NLRs.As keratinocyte PRRs can respond to both commensal and pathogenic microorganisms, fine-tuned regulation is required to distinguish between them.A primary mechanism through which this is achieved is through the differentiation and expansion of differential T cell subsets upon antigen presentation by dermal DCs.In the skin, contact with the commensal Staphylococcus epidermidis mediates CD4 + regulatory T cell (Treg) differentiation and influx to the skin, concomitant with low activation of effector CD8 + T cells (Teff).In contrast, Staphylococcus aureus induces a Teff-skewed response.This is accomplished by the local production of IL-1b by skin myeloid cells in response to S. aureus toxins, which drives the expansion of effector T cells [87][88][89][90].Interestingly, S. epidermidis is also able to induce IL-1b, however, the extent of IL-1b signalling in response to this commensal is tightly regulated by the induction of the TNF alpha-induced protein 3 (TNFAIP3) which limits IL-1b signalling thereby avoiding inflammation [91].This mechanism not only favours commensal tolerance but also regulates the ability of the immune system to respond against infection (Fig. 1).For example, Naik et al. demonstrated that the presence of S. epidermidis in the skin also confers a protective phenotype against parasitic infection with Leishmania major.This protection was abolished in IL1R1 À/À mice and upon IL1Ra administration, indicating that S. epidermidis-mediated protection requires IL-1 signalling [92].
As well as playing central roles in the maintenance of tolerance in dermal tissues, IL-1 cytokines are also involved in the underlying mechanisms of several skin inflammatory diseases in which the microbiome is altered.Several cutaneous inflammatory disorders have been associated with dysbiosis of the skin microbiota, including psoriasis, AD or acne vulgaris, among others [90,93].Many of these skin diseases manifest with flares that are associated with the outgrowth of normally commensal bacterial species which can become pathogenic when the microbiome equilibrium is altered.IL-1 cytokines and specifically IL-1b, are important mediators of skin inflammation upon recognition of these pathogenic bacteria.Although the precise role of IL-1 cytokines in the pathogenesis of acne has not been fully elucidated, in several studies it has been shown that C. acnes can induce IL-1b release.C. acnes drives IL-1b production in vivo, in the skin as well as in vitro in mouse macrophages [94].Similar effects have also been reported in human peripheral neutrophils, monocytes and sebocytes [95][96][97].Moreover, high levels of IL-1b are found in the skin lesions of acne patients [94,98].Interestingly, it has also been recently proposed that only specific acneic strains of C. acnes are capable of triggering NLRP3-induced IL-1b production.Mechanistically, pathogenic C. acnes strains, but not their commensal counterparts, produce porphyrins that provoke K + leakage in keratinocytes, and consequently activate IL-1b processing and release dependent on the activation of the NLRP3 inflammasome in these cells [99].This may explain why some phylotypes elicit an inflammatory response and are involved in acne pathology whereas others are natural commensals found on healthy skin.
In keratinocytes from patients with AD, the expression of all the IL-1 subfamily members, IL-1a, IL-1b and IL-33, (as well as IL-18), have been found to be increased [100][101][102].However, how these cytokines regulate inflammation in AD and the role that S. aureus plays in this scenario is not well defined.Interestingly, while IL-1b has been shown to promote S. aureus growth [103], S. aureus has also been reported to induce IL-1b production by neutrophils in the skin, airway and lungs [89,104,105] indicating that a positive feed forward loop may be established between this bacteria and IL-1b in AD skin.
A major risk factor for AD among some populations is loss-of-function mutations in the gene encoding filaggrin protein, which is key in maintaining the integrity of the skin epithelial barrier [106].Moreover, independently of these mutations, filaggrin can also be downregulated in AD patients, downstream of Th2 and Th22 signalling [107].These observations have Fig. 1.IL-1 family cytokines regulate the host-microbiome interactions in the skin.Under homeostasis, immune tolerance to commensal bacteria is achieved though several mechanisms in the skin.Contact with commensal bacteria, such as S. epidermidis, induce low levels of IL-1b production.Upon recognition of this bacteria, skin resident dendritic cells mediate antigen transport to the skin draining lymph nodes (SDLNs), which induces the priming of antigen specific regulatory CD4 + T cells (T regs ) and its subsequent migration into the skin.Changes in the skin microbial communities are described in several inflammatory diseases and include a reduction in the diversity of some commensals, together with the increased presence of pathogenic bacteria.Recognition of these pathogens, such as S. aureus, induce the migration of effector T cells (T effs ) with a minor T regs response.This mechanism is dependent on IL-1b production by skin myeloid cells and IL-1R signalling.S. aureus also induces the release of IL-1a and IL-33 by the KCs, which promotes neutrophil responses in the skin, including chemokine production and formation of neutrophil extracellular traps (NETs).IL-18 is an important player in skin inflammation.KCs constitutively express IL-18, and its production is enhanced upon pro-inflammatory stimuli in the skin such as pathogenic bacteria.IL-18 acts as a chemoattractant, recruiting DCs that express IL-18R in their surface.It also promotes KCs production of the skin chemokines CXCL9, CXCL10 and CXCL11, which initiate the recruitment of Th1 cells.been extensively phenocopied in murine studies, wherein filaggrin deficiency led to the development of spontaneous AD [108].Subsequently, these same authors demonstrated that filaggrin-mutant (FLG ft/ft ) mice exhibited an altered skin microbiome during the neonatal period, with a marked overrepresentation of bacteria from the genus Staphyolococus.In these mice, skin inflammation was mediated by IL-1R1 signalling, but occurred independently of NLRP1, NLRP3, NLRP6 or AIM2 inflammasomes, suggesting that alternative modes of IL-1 processing, such as neutrophil and mast cell-derived proteases, may be important in this model [109].Alternatively, other groups have also reported that IL-1b mediated skin inflammation is mediated by the NLRP3 inflammasome.Simanski et al. [110] showed that S. aureus induced IL-1b and IL-1a production in human keratinocytes in a manner dependent on caspase-1 and ASC components of the inflammasome.In an effort to reconcile these observations, it was later reported that IL-1b production in the skin can occur through both inflammasomedependent and independent mechanisms [105].
Although most evidence concerning the role of the skin microbiome in the development of AD inflammation is focused on IL-1b, IL-1a, and in particular IL-33, also likely play important roles (Fig. 1).IL-1a is also produced by keratinocytes upon S. aureus stimulation [110,111].Interestingly, IL-1a, and not IL-1b, was reported to mediate neutrophil-derived chemokine expression in this context [111].IL-33, whose expression has been shown to be increased in response to S. aureus, is of particular interest in the context of AD due to its Th2 polarizing properties [101].Wang et al. have demonstrated that in response to S. aureus [112], IL-33 mediates the formation of neutrophil extracellular traps (NETs), which are a crucial defence mechanism against bacterial pathogens.Interestingly, IL-33 production has been suggested to be specific to S. aureus and not induced by other Staphylococcal species [113].Most recently, it has been reported that a microbial metabolite, propionate, can suppress ADlike inflammation by inhibiting IL-33-induced responses, further reinforcing the idea that the altered composition of the skin microbiome in AD may precede the onset of the disease [114,115].
A link between microbiome alterations and IL-1b signalling in psoriatic inflammation has also been suggested.Levels of IL-1b expression were found to be elevated in both human psoriatic skin, as well as in imiquimod (IMQ)-treated mice, where increased expression was dependent on the presence of commensal bacteria and required to mediate dermal inflammation [35].Although elevated IL-33 activity has also been implicated in dermal inflammation in psoriasis [116,117], whether it plays a role in the host microbial interactions in this setting, or indeed in the establishment of immune tolerance to skin commensals, has not been extensively investigated and warrants further investigation.

IL-1 subfamily and gut microbiome
In several studies, a role for IL-1b in intestinal barrier disruption has been suggested.Even at physiological concentrations, IL-1b can induce an increase in intestinal barrier permeability [118,119], allowing the permeation of luminal antigens and causing intestinal inflammation (Fig. 2) [120,121].One of the proposed mechanisms for the IL-1b-meditated disruption of the intestinal barrier is an increase in the expression and activity of the myosin light chain kinase (MLCK), which causes an opening of the TJ that maintains the binding between the IECs [121].It has also been reported that IL-1b causes a decrease in occludin expression, by up-regulating the transcription of the microRNA, mir-200c-3p in enterocytes [122].
Several studies have also suggested that the link between gut dysbiosis and intestinal epithelial barrier dysfunction can be mediated by IL-1R-dependent mechanisms.In an elegant study by Seo et al., it was found that commensal bacteria, specifically Proteus mirabilis, induced NLRP3-dependent IL-1b production in the intestine, worsening inflammation observed dextran sulfate sodium (DSS)-treated mice.This effect was abolished in NLRP3 À/À mice and by treating the animals with recombinant IL-1Ra, [123].Similarly, Shaw et al. [124], also reported that commensal microbiota induced IL-1b production by intestinal resident macrophages, which was critical for the development of CD4 + Th17 cells in the intestine.
As described earlier, IL-1b cytokine maturation and release can be regulated through inflammasome activation [81,82].However, the role of these signalling platforms in intestinal inflammation and homeostasis remains controversial.In the majority of the studies, deficiency of nlrp3 or nlrp6 has been found to be deleterious in the context of intestinal infection or inflammation, and is associated with changes in the intestinal microbiome [125][126][127][128][129][130].However, in other studies, deficiency of NLRPs or other inflammasome components has resulted in reduced levels of intestinal inflammation [123,[131][132][133].In many instances, such discordances may be explained by the different experimental models used to investigate colitis.In addition, different approaches aimed at investigating the role of the microbiome in influencing disease progression such as, germ-free (GF) mice, broad-spectrum antibiotics treatment, cohousing or faecal transplantation, are likely an important consideration.While several questions surrounding the interplay between the microbiome and IL-1 in intestinal inflammation remain to be answered, it is likely that alterations in the microbial community composition can weaken the intestinal epidermal barrier and promote intestinal inflammation at least in part through IL-1b dependent mechanisms.
IL-1a has been implicated in the development of intestinal inflammation in the context of IBD but may play a different role than its counterpart, IL-1b.In response to intestinal epithelial cell necrosis, IL-1a is rapidly released by intestinal fibroblasts, and can act to amplify gut inflammation by inducing cytokine secretion in the surrounding myeloid cells (Fig. 2) [134].In mice undergoing DSS-induced colitis, IL-1a expression was detected during the earliest phases of disease progression, followed by a subsequent increase in IL-1b, suggesting that IL-1a is an early mediator of colitogenic inflammation [134].Similar results were noted by Bersudsky et al. [135], who also demonstrated that early IL-1a expression by intestinal epithelial cells can act as a proinflammatory signal, whereas later expression of IL-1b, by gut resident myeloid cells, can mediate intestinal resolution and repair.Interestingly, such data indicate that specifically targeting IL-1b in the context of IBD may not prove beneficial and is consistent with studies demonstrating that genetic ablation of nlrp3, nlrp6 or other inflammasome components can result in increased susceptibility to intestinal inflammation and worsens epithelial intestinal barrier integrity [125][126][127][128][129][130].
Fig. 2. IL-1 family cytokines regulate the host-microbiome interactions in the gut.The gut microbiome is mainly composed of bacteria, with Firmicutes and Bacteroides the most abundant phyla under homeostatic conditions.Specific subsets of dendritic cells (DCs) mediate tolerogenic responses to these commensal microbes.Resident CD103 + DCs recognize luminal bacteria and migrate to the surrounding mesenteric lymph nodes (MLNs).Here, DCs induce the expansion of antigen-specific regulatory T cells (Tregs) and induce the migration of these cells to the gut (gut homing).IL-33, produced by the intestinal epithelial cells (IECs), plays a key role in maintaining the homeostatic intestinal microbiome by mediating the secretion of anti-microbial peptides (AMPs), such as Reg3c.A reduction in bacterial diversity, and specifically in anti-inflammatory short-chain fatty acids (SCFAs)-producing bacteria, is associated with the progression of several intestinal inflammatory diseases.This can in turn lead to increased expression and production of IL-18 which can promote inflammatory processes in the gut.The alarmin IL-1a is rapidly produced by intestinal fibroblasts under recognition of inflammatory stimuli and mediates IL-1b production by the intestinal innate immune cells, amplifying gut inflammation.Moreover, IL-1a is associated with an increase in Proteobacteria species such as Helicobacter pylori.IL-1b increases the permeability in the intestinal epithelial barrier, which allows the penetration of the luminal antigens into the lamina propria.As a consequence, bacterial products such as lipopolysaccharide can enter into the bloodstream inducing endotoxemia.
Two recent studies have explored the link between IL-1a and bacterial communities in the gut.Interestingly, Menghini et al. reported that specific neutralization of IL-1a with a monoclonal antibody ameliorated colitis through modulating the components of the microbiota towards an anti-inflammatory profile (Fig. 2).IL-1a blockade resulted in a decreased abundance of Proteobacteria, specifically Helicobacter species, while increasing Bacteriodetes and Lactobacillus.Furthermore, the protective effect of anti-IL-1a was abolished in GF mice [136].Interestingly it has also been reported that IL-1a À/À mice exhibited a lower abundance of Akkermansia muciniphila, a mucusdegrading commensal bacteria which may be associated with colitis severity [67].In association with such changes, IL-1a À/À mice were found to exhibit relative protection from colitis which was reversed upon cohousing of mice with their wt counterparts and microbiome equalization [137].Collectively, these data suggest microbiome dysbiosis as being an important mediator of IL-1a driven inflammation in the gut.
There is also accumulating evidence to suggest that IL-33 can play an important role in mediating host microbial interactions in the gut.In contrast to IL-1a, IL-33 deficiency has been reported to lead to a specific outgrowth of A. muciniphila, as well as segmented filamentous bacteria (SFB), as a consequence of reduced IgA secretion [138].These changes were also associated with increased susceptibility to DSS-induced colitis, and colitis-associated cancer, indicating that IL-33 plays a key role in mediating intestinal homeostasis by regulating the composition of the intestinal microbiome (Fig. 2) [138].As well as through regulating IgA secretion, it has been proposed that IL-33 can act to alter the composition of the intestinal microbiome through regulating the expression of the antimicrobial peptides (AMPs), Reg3c by IECs [139].Indeed, loss of IL-33-induced Reg3y was also associated with increased susceptibility to infectious colitis [139].Collectively, these data indicate an important role of IL-33 in the homeostatic maintenance of a healthy intestinal barrier.

IL-18 subfamily
IL-18 and IL-37 cytokines form the IL-18 subfamily.Similar to IL-1b, IL-18 is synthesized in precursor form, and its processing and secretion are inflammasome-dependent.The main cellular sources of IL-18 are endothelial cells, keratinocytes and intestinal epithelial cells, which constitutively express the precursor form.IL-18 shares some functional similarities with other IL-1 family members.Similar to IL-1a and IL-33, it can act as an alarmin and can also facilitate the differentiation of Th17 cells, similar to IL-1b [75,76].However, IL-18 also has unique and nonredundant functions.For example, IL-18 can act, together with IL-12 and IL-15, to enhance Th1-type responses, inducing IFNc expression in T cells and NK cells.In the absence of IL-12, if IL-2 or IL-4 are present, IL-18 can also induce Th2 responses, enhancing the production of IL-4, IL-13 and the generation of immunoglobulin E (IgE) responses.[140,141].IL-18 signalling is mediated by a heterodimeric receptor consisting of the IL-18 alpha chain (IL-18Ra or IL-1R5) and the IL-18 receptor beta chain (IL-18Rb or IL-1R7).Receptor complex formation, upon IL-18 binding, initiates a pro-inflammatory cascade orchestrated by MyD88 and Nf-jB downstream signalling [74,142].Excessive IL-18 signalling is regulated by the expression of a soluble decoy receptor, the IL-18 binding protein (IL-18Bp).IL-18Bp is a secreted protein that is constitutively expressed in healthy serum and exhibits a relatively high affinity for IL-18, thereby, neutralizing its activity under steady-state conditions.During inflammatory disease, excess levels of induced IL-18 can overcome these neutralizing effects [143].
Less is known about the other member of the IL-18 subfamily, IL-37, in part due to its lack of a homologue in rodent species.IL-37 has been described to be elevated in human inflammatory diseases and is historically considered to uniquely exhibit antiinflammatory properties.Interestingly, it has been suggested that IL-37 also binds to IL-18Ra, with SIGIRR, in place of IL-18Rb, as a co-receptor.In this context, IL-37 can initiate downstream antiinflammatory signalling pathways.IL-37 is also considered a dual cytokine due to its capacity to migrate to the nucleus and act as a transcriptional regulator.[74][75][76].More recently, it has been reported that upon specific processing by neutrophil-derived proteases, IL-37 can also exhibit proinflammatory signalling through the IL-36R [144].

IL-18 subfamily and the skin microbiome
As a pleiotropic cytokine mainly expressed in epithelial and endothelial cells, IL-18 has been associated with several skin inflammatory diseases.Human keratinocytes constitutively express IL-18, and they can secrete active IL-18 in response to different pro-inflammatory stimuli [145][146][147].In the context of skin inflammation, IL-18 regulates the production of chemokines such as CXCL9, CXCL10 and CXCL11, and can act itself as a chemoattractant recruiting DC that expresses IL-18R to the skin (Fig. 1) [148,149].
In psoriasis, high levels of IL-18 are found, not only in psoriatic skin lesions, but also systemically in the serum of patients [150][151][152].Moreover, elevated IL-18 has been associated with disease progression in terms of severity and response to treatment, leading to its proposal as a potential disease biomarker [153,154].In in vivo models of psoriasis, it has been demonstrated that IL-18 may play a role in potentiating immune cell infiltration to the skin [155,156].IL-18 deficient mice displayed a milder disease phenotype in the IMQinduced model of disease associated with reduced immune cell infiltration [156].To date, no reports have examined in detail whether IL-18 can either act to influence, or have its activity modulated by, skin microbial communities in the context of psoriasis.This is an area which warrants further investigation and indeed, some authors have begun to explore the possible link between AMPs associated with psoriatic inflammation and IL-18 in human skin.Human bdefensins (hBD)-2,-3,-4 and cathelicidin, LL-37, have been shown to induce the production of IL-18 in vitro in human keratinocytes perhaps indicating an interplay between IL-18 signalling and the skin microbiome in psoriatic inflammation [157,158].
In the context of AD, IL-18 has been proposed to participate in the inflammatory response to S. aureus in the skin of patients.IL-18 levels are increased, both in the serum and epidermal lesions of patients with AD, and are significantly correlated with dermatitis severity [100,159,160], indicating a pathogenic role.S. aureus-derived protein A (SpA) induced IL-18 expression by mouse keratinocytes in vitro and S. aureus-derived extracellular vesicles can enhance both IL-18, and IL-1b, expression and activation through inflammasome activation in macrophages [161,162].In vivo, the potential role of IL-18 in S. aureus-dependent AD inflammation has been explored using the SDS/SpA model wherein dermal inflammation is induced through first compromising the skin barrier with detergent (SDS), followed by topical administration SpA.Interestingly, neutralizing IL-18 in this model was shown to inhibit skin inflammation [161,163].Such observations indicate that IL-18 may play an important role in mediating S. aureusdependent inflammation in AD.
The functional role of the IL-18 subfamily member, IL-37 in the context of dermal inflammation remains relatively ill-defined.Several recent studies have demonstrated that IL-37 expression is decreased in lesional skin from psoriasis, AD, and Hidradenitis suppurativa (HS) patients [164][165][166] suggesting that loss of its immunosuppressive activity may be important.In addition, several studies have suggested that IL-37 may play an important role in regulating the expression of epidermal differentiation complex proteins and in maintaining a healthy skin barrier [165,167].In contrast, it has also been suggested that specific proteolytic processing leads to the expression of a proinflammatory form of IL-37 which can elicit psoriatic-like inflammatory responses in the skin, at least in part through engaging the IL-36R.While much remains to be learned about the specific roles of IL-37 at barrier sites, such data raise the intriguing prospect that IL-37 may also play a role in mediating dialogue between the host and the skin microbiome.

IL-18 subfamily and the gut microbiome
Due to its primary expression in intestinal epithelial cells, it has been suggested that IL-18 may play a role in the pathogenesis of intestinal diseases such as IBD.In this regard, early observations established that intestinal levels of IL-18 expression are upregulated in CD, and positively correlate with disease severity in these patients [168].Specific deletion of IL-18 in IECs protected mice from DSS-induced colitis, while IL-18Bp deficiency resulted in an exacerbated colitis phenotype associated with the defective maturation of intestinal goblet cells (Fig. 2) [169].Similarly, mice deficient in either IL-18 or IL-1b, or both cytokines, were found to exhibit reduced intestinal inflammation in a trinitrobenzene sulfonic acid (TNBS) colitis model.Interestingly, enhanced protection was observed in the absence of both cytokines, suggesting an additive role in the context of intestinal inflammation [170].
A proinflammatory role for IL-18 in the intestine has been specifically associated with NLRP6 inflammasome activation.Elinav et al. demonstrated that, under steady-state conditions, NLRP6 À/À mice showed reduced levels of IL-18 in the serum in comparison to WT mice.In addition, NLRP6 À/À mice developed more severe colitis than their WT counterparts, and this exacerbated disease phenotype was transferred to WT mice upon co-housing.These effects were associated with dysbiosis of the gut microbiome driven by nlrp6 deficiency, and mediated by IL-18 [125].Subsequent reports have supported the idea of NLRP6 being protective in intestinal inflammation, in association with changes in the gut microbiome [128,129].However, additional studies, using littermate and same sex controls, have indicated that nlrp6 deficiency does not affect either the composition of bacterial communities in the intestine, or susceptibility to colitis.[171,172].Considering these conflicting results, the role of NLRP6 in regulating the gut microbial composition and intestinal inflammation through IL-18 remains unresolved.
Several important studies have indicated that the IL-18 subfamily cytokine, IL-37, plays an important role in regulating intestinal homeostasis and suppressing inflammation.Initial observations, using a humanized transgenic mouse model demonstrated that IL-37b can play a protective role in the DSS colitis model through suppressing intestinal inflammation [173].More recently, it has been reported that a homozygous loss of function mutation in IL37 resulted in the onset of infantile IBD [174].Collectively, these data strongly indicate that IL-37 plays an important immune suppressive role in the intestine.While a link between IL-37 and the composition of the intestinal microbiome has yet to be fully investigated in this context, it is interesting to note that several reports have suggested that IL-37 activity can influence gut microbial dysbiosis [175,176].

IL-36 subfamily
The IL-36 subfamily consists of three cytokine ligands IL-36a, IL-36b and IL-36c, as well as two specific antagonists, IL-36Ra and IL-38.Similar to related IL-1 family members, IL-36 activity is enhanced upon proteolytic processing, largely by neutrophil-derived proteases [177,178].IL-36 cytokines initiate signalling through binding to the specific IL-36R which, in conjunction with IL-1RAcP, activates downstream signalling cascades to drive proinflammatory gene transcription.When present in excess, the specific antagonist IL-36Ra preferentially binds to the IL-36R chain without recruiting IL-1RAcP and thereby preventing ligand binding and receptor activation.Expression of IL-36 family members is predominantly found at barrier sites such as the skin and intestine, where they play important roles in regulating homeostasis and inflammation [73,179,180].
IL-36 subfamily and the skin microbiome IL-36 cytokines are recognized as major players in psoriatic inflammation stemming from the discovery that mutations in the gene encoding the IL-36R antagonist (il36rn), which reduce its inhibitory function, can result in the onset of generalized pustular psoriasis [181].In addition, several studies have confirmed the orchestrating role of IL-36 cytokines in mediating psoriatic inflammation in mice [182,183].Such observations have prompted significant interest in gaining a deeper understanding of how IL-36 activity regulates dermal homeostasis and inflammation, not just in psoriatic disease, but also in conditions such as AD and HS.Recent data indicate that IL-36 cytokines, and IL-36a and IL-36c, in particular, play important roles in driving neutrophil infiltration and the development of pathogenicT-cell responses in psoriatic skin [184,185].Ongoing clinical studies, evaluating strategies aimed at inhibiting IL-36 cytokine activity in inflammatory skin conditions, have led to the recent regulatory approval of one such approach for the treatment of generalized pustular psoriasis GPP [186].
Whether IL-36 activity is regulated by, or itself influences, the altered composition of the skin microbiome in psoriasis remains to be investigated.However, recent confirmation that IL-36 receptor activity in keratinocytes is critical to the development of psoriatic inflammation in mice [184], alongside the reported ability of IL-36 cytokines to modulate the expression of skin AMPs such as REG3c and human betadefensins (HBD2/3) indicate a potential role [187,188].Such observations may also have relevance to HS which is also characterized by elevated IL-36 cytokine expression [189] and dysbiosis of the skin microbiome [190].
Although IL-36 subfamily expression is also elevated in AD skin, its role in promoting dermal inflammation in this setting is less well-defined.It has however been reported that IL-36 activity can mediate Th17dependent skin inflammation in response to epicutaneous exposure to S. aureus in mice.Interestingly, these effects were specific to IL-36 and were evident even in the absence of IL-1a/b, IL-18 or IL-33 signalling [191].
IL-38 appears to act in a similar fashion to the IL-36Ra in the context of psoriatic inflammation.Like IL-36Ra deficient mice, IL-38 À/À mice have been shown to exhibit exacerbated disease in the IMQ model of psoriatic inflammation [192].Interestingly, and in contrast to IL-36Ra, IL-38 expression is decreased in psoriatic lesional skin indicating that loss of its immunosppressive activity may contribute to dermal inflammation [193,194].It is unclear what role, if any, the skin microbiome might play in this context.

IL-36 subfamily and the gut microbiome
IL-36a and IL-36c expression are also elevated in the intestinal mucosa of patients with IBD, where they appear to play dichotomous roles, promoting both intestinal inflammation and resolution in different models of disease [179].In contrast to the skin, there are several reports which indicate that IL-36 signalling can play an instructive role in shaping the intestinal microbiome.Similar to IL-33 deficiency, mice-lacking expression of the IL-36 receptor antagonist, were found to have a relative outgrowth of A. muciniphila.Although the significance of this observation in the context of colitis was not investigated, it was demonstrated that these alterations occurred in association with improved intestinal barrier integrity upon exposure to high-fat diet [195].In addition, it has also been reported that deficiency of the IL36 receptor resulted in alterations in the constituents of the intestinal microbiota, as well as reduced IL-23, IL-22 and AMPs expression in colitogenic mice [196].These data support earlier observations surrounding the influence of the IL-23/22 axis in modulating the intestinal microbiome and suggest a potential role IL-36 in these effects [197,198].As well as influencing the composition of the microbiome, it has also been demonstrated that the gut microbiota is required to induce the increased expression of IL-36 cytokines observed in the intestines of colitogenic mice [199].These observations suggest a bidirectional dialogue, between the IL-36 cytokine subfamily and the gut microbiome, which may play an important role in intestinal inflammation and homeostasis.
Expression of IL-38 is also elevated in the colonic mucosa of UC patients.In support of an antiinflammatory role, several recent reports have demonstrated that IL-38 À/À mice suffer from more severe DSS-induced colitis [200].Interestingly, this increased severity has been associated with dysregulated nlrp3 activity resulting in enhanced IL-1b processing suggesting a novel mode of regulation between different IL-1 family members [200], whether IL-38 activity impacts the gut microbiota has not been investigated.

Regulation of IL-1 family cytokine activity by microbe-derived proteases
As well as playing pivotal instructive roles in regulating the composition of the microbiome at barrier sites, there is significant evidence that microbes themselves can have a profound influence in shaping the reciprocal activity of IL-1 family cytokines [201].Extracellular proteases from Group A Streptococcus (GSA) and S. aureus have been demonstrated to cleave full-length IL-1b into its more active form and indeed pharmacological inhibition of IL-1b in patients is associated with increased rates of invasive GSA infection [202][203][204].Similarly, recent studies by MacLeod et al. have reported that epithelial pathogenic microbes, such as GSA, induce not only the expression and release of IL-36c, but also its activation through cleavage by microbe-derived proteases.In contrast, commensal species such as S. epidermidis did not elicit IL-36c secretion and activation [205].Such observations have led to the suggestion that IL-1 family cytokines may act as sentinal innate immune sensors of proteolytic enzymes specifically expressed by pathologic microbes, as opposed to harmless commensal species.They also serve to highlight the central role of IL-1 family cytokines in the bidirectional dialogue between the host and microbial species at barrier sites.

Summary and future perspectives
It is becoming increasingly evident that IL-1 family cytokines play important roles in regulating hostmicrobiome dialogue at barrier sites by mediating immune cell and parenchymal cell responses (Fig. 3).As data continues to emerge concerning how IL-1 family cytokines can orchestrate these events, some key questions concerning the relevance of these findings in the context of human health and disease will likely be answered.While much of the data generated to date in this area is derived from preclinical, and especially murine, models, it will be of particular interest to determine how such findings translate to human disease states, where microbiome dysbiosis is best characterized, including IBD and dermatoses such AD and psoriasis.
Several strategies aimed at inhibiting IL-1 cytokine family activity, have either entered, or are under current evaluation in the clinic in recent times.Those already approved include recombinant receptor antagonists such as Anakinra (IL-1RA), specific monoclonal antibodies such as canakinumab (Anti-IL-1b) and spesolimab (Anti-IL-36R) and soluble receptor fusion proteins such as rilonacept (IL-1Trap).While to date, these strategies have been most demonstrably efficacious in treating monogenic autoinflammatory disorders, they are also being evaluated across more common chronic inflammatory diseases of the intestines and skin [206][207][208][209][210][211].Such clinical studies would provide an ideal opportunity to effectively translate findings from murine studies, on the relevance of the microbiome to disease states across relevant human cohorts.As well as providing a deeper mechanistic insight into the role of IL-1 cytokines in disease, the analysis of how therapeutic inhibition of these cytokines may impact microbiome dysbiosis will also provide potential further opportunities to improve patient outcomes.If a link between IL-1 family cytokine activity and dysbiosis in human disease can be established, such findings would open up significant possibilities for further investigation into the potential influence of the microbiome in driving disease pathogenesis.In addition, the impact of therapeutic inhibition of IL-1 family members on the composition of microbial communities resident in the skin and/or gut could potentially identify powerful biomarkers of disease progression and drug responsiveness.In these aspects, the incorporation of microbiome analyses into ongoing and future clinical investigations would be extremely valuable.
It is noteworthy that much of our existing depth of knowledge concerning the influence of IL-1 family cytokines at the interface with the microbiome centres around more established family members such as the prototypical IL-1 subfamily.Much less is known about what role, if any, more novel family members such as IL-37 and IL-38 might play.As further insights are gained into the importance of these cytokines in the pathogenesis of inflammatory disease at barrier sites, it is likely that whether they too can play a role in this dialogue will be revealed.

Fig. 3 .
Fig. 3. Mechanisms involved in microbiome regulation by IL-1 family cytokines.Mechanisms by which IL-1 family cytokines regulate the microbiome involve both immune cell and parenchymal cell responses.IL-1 family cytokines regulate the activation, differentiation, migration and secretory function of these cells leading to alterations in downstream effectors (antibodies; cytokines/chemokines; tight junction proteins; antimicrobial peptides; mucus layer) which are directly influence microbiome regulation in skin and gut.HBD2/3, human beta defensin 2/3; IECs, intestinal epithelial cells; IgA, immunoglobulin A; REG3c, regenerating islet-derived protein 3 gamma.