The Role of Inflammasomes in Intestinal Inflammation

View current table in a new window

Tables index

Veiw figure View Table

View next table

3.1.2. Toll-like Receptors (TLRs)

Toll like receptors in general sense PAMPs, but can occasionally sense DAMPs, for example the heat shock proteins released by dead or dying epithelial cells is recognized by TLR4 ^{[118, 131]}. TLRs are transmembrane glycoproteins containing; a multiple leucine-rich regions and one or two cysteine rich regions in the extracellular domain, a short transmembrane region, and an intracellular domain containing a toll/interleukin-1 receptor (TIR) domain ^[118]. Epithelial cells, macrophages, dentritic cells, neutrophils, monocytes, T cells and B cells are all known to express TLRs ^{[3, 11, 82]}, however the expression levels of TLRs are typically lower in epithelial cells than hematopoietic cells, and the pattern of TLR expression is modulated by activation and cell maturation ^[43]. To date, 10 functional TLRs have been identified in humans ^{[3, 11, 61, 118]} and 13 functional TLRs have been identified in mice ^[118]. TLRs are sequestered in the Endoplasmic reticulum (ER) and the heat-shock protein GP96 acts as a general chaperoned and delivers most of the TLRs to the appropriate endosome compartment or plasma membrane ^[61]. TLRs 1,2,4,5 and 6 are located on cell membrane surfaces and generally recognize microbial membrane components, whereas TLRs 3, 7, 8 and 9 are located on intracellular endosome membranes and recognize nucleic acids from pathogens or infected cells ^{[61, 118]}. As examples; TLR2 recognizes peptidoglycan (a bacterial cell wall component) and constituents of gram-negative bacteria, mycobacteria and fungi, TLR3 sensors double stranded RNA,TLR4 recognizes bacterial lipopolysaccharides (LPS) and TLR5 sensors bacterial flagellin ^[124] (Table 1). The TLRs can act alone, in concert with one another or with other pattern recognition receptors, and although the diversity of PAMPs recognized by TLRs is limited to some degree, it is sufficient to detect most microbes that invade the host through the mucosal barrier.

Upon ligand recognition TLRs undergo conformational changes at the TIR domain which allow for homo- or heterophilic interactions of TLRs and the recruitment of a single adaptor molecule or a specific combination of adaptor molecules ^[11]. Most TLRs recruit the MyD88 adaptor molecule however “MyD88 independent” pathways which recruit the MAL, TRIF, TIRAP and TRAM, are possible and are utilized by TLR3, TLR4 and TLR5 ^{[3, 11, 18, 61]} (Table 1). The dominant MyD88 dependent pathway leads to the phosphorylation of I_kB-kinases which releases the NF-_kB transcription factor sequestered in the cystol, and at the same time activates the mitogen activated protein kinase (MAPK) pathway causing nuclear translocation and the activation of inflammatory cytokines such as TNF-α, IL-1, IL-6 and IL-18 ^{[43, 81]}. The alternative MyD88 independent pathways results in late-phase NF-_kB nuclear translocation and activation of IRF transcription factors (similar in structure to NF-_kB), which together act to induce type 1 IFN and inflammatory cytokine production ^[61].

TLR expression levels vary in disease states, for example in CD and UC TLR4 is reportedly upregulation in intestinal epithelial cells (IECs) and lamina propria mononuclear cells ^{[11, 76]}. Szebeni et al ^[124] also observed higher TLR2 and TLR4 mRNA and protein levels in the inflamed colonic mucosa of children with IBD compared with controls and non-inflamed mucosa and similar TLR3 levels across controls, active disease and non-inflamed colonic mucosa. In mouse models TLR4 is involved in regeneration of IECs ^[76]. Both TLR4 KO mice and MyD88 KO mice subjected to dextran sulphate sodium (DSS) induced colitis, showed less neutrophil recruitment to injury site and increased gram-negative bacteria translocation to the mesenteric lymph nodes (MLN) than wild-type littermates, indicating diminished bacterial clearance and suggesting that TLR4 and MyD88 are required for neutrophil recruitment to the damaged mucosa ^[44]. Studies by Cario et al ^{[12, 37]} using TLR2 deficient mice with DSS induced colitis demonstrated ameliorated mucosal inflammation and accelerated wound healing in intestinal epithelial cells when mice were given the TLR ligand Pam3CSK orally.

3.1.3. NOD-like Receptors (NLRs)

The NLR family comprises 23 human and 34 mouse members ^{[3, 106, 118]}, and based on the N-terminal region can be grouped into 5 distinct subgroups; the NLRC/NLRA/NLRX1 (contain a CARD domain), NLRPs (contain a PYD domain)and NLRB subfamilies (contain a BIR domain) ^{[35, 118]}. All NRLs contain a central nucleotide-binding and oligomerization NACHT (NACHT) domain, flanked by a C-terminal leucine-rich repeat region (LRR) and N-terminal effector domain ^{[21, 35, 114]}. The NACHT domain is involved in activation of the NLR via an ATP dependent oligomerization process. The LRR region of the NLR, is similar in structure to the TLR leucine-rich region, and is thought to be responsible for ligand sensing either by direct or indirect means ^{[21, 114]}. The N-terminal effector region consists of a CARD, PYD or BIR domain which participates in protein-protein interactions and recruitment of downstream effector molecules ^[35].

The most extensively studied NLRs; NOD1 and NOD2, belong to the NLRC receptor family, and share similar structural homology with the exception of the amino terminus where NOD2 contains 2 CARDs and NOD1 has one CARD ^[118]. In the resting state, NOD2 is maintained in an inactive conformation by auto-inhibition of the NACHT domain by the LRR ^[8]. NOD1 and NOD2 recognize intracellular bacteria by recognizing the breakdown products of peptidoglycans shed during normal bacterial cell growth and cell division ^[114].

NOD1 senses muro-peptides (iE-DAPS) found mainly in Gram-negative bacteria and some Gram-positive bacteria, and NOD2 senses muramyl dipeptide (MDP) from Gram-negative and Gram-positive bacteria ^{[21, 35, 114]} and has been shown to recognise viral ssRNA ^[110]. Transport systems such as Pannexin-1, PepT1 and PepT2 and endocytosis have been suggested as possible mechanisms whereby bacteria products infect and enter the cystol ^{[35, 73]}.

Both NOD1 and NOD2 are expressed in antigen presenting cells, however NOD2 expression is higher in the specialized cells, such as paneth cells located in the small intestine ^{[8, 94]}, whereas NOD1 is mainly expressed in tissue and epithelial cell subsets ^[22]. The expression of NOD1 is regulated by IFN-ɣ (interferon-ɣ) acting through IRF1 (IFN-regulatory factor 1) at the CARD4 promoter, whereas NOD2 expression is regulates by LPS, TNF-α (tumor necrosis factor-α) and IFN-ɣ ^[118]. Noteworthy, is NOD2 ability to upregulate itself upon LPS or TNF-α stimulation by activating NF-_kB, which binds to the CARD15 promoter site inducing NOD2 gene expression ^{[120, 121]}.

Upon ligand sensing of the LRR, both NOD1 and NOD2 undergo conformational changes and oligomerization of the NACHT domain, allowing engagement of the CARD domain with RIP2 (receptor-interacting serine/threonine kinase 2) via CARD-CARD interactions ^[118]. RIP2 activates the IKK complex through IKKɣ and NEMO (NF-_kB essential modulator) ^[118] and also interacts with cIAP1 and cIAP2 (apoptosis inhibitor proteins) leading to the release of the NF-_kB sequestered in the cystol for translocation into the nucleus and the initiation of proinflammatory gene transcription which includes the production of TNF-α, IL-1β, and IL-6 cytokines ^[6]. In a similar manner, ligand sensing by NOD2 and NOD1 can activate the MAPK pathway via RIP2, leading to the upregulation of proinflammatory cytokines ^[21].

A series of experiments have confirmed the importance of NOD1 and NOD2 in regulating the intestinal microbiota. Firstly, Petnicki et al ^[99] showed that mice in germ free conditions had lower NOD2 expression levels and introducing commensal bacteria resulted in increased NOD2 expression. Secondly, Geddes et al ^[45] showed that mice deficient in both NOD1 and NOD2 featured decreased inflammation but increased bacterial colonization of the colon.

4. Inflammasomes

The inflammasome is a multimolecular, possibly hexamers or heptamers complex ^[114], which serves as a platform for the activation of caspase-1 and the maturation of inflammatory cytokines such as IL-1β and IL-18 ^{[15, 60, 82, 106]}. Several of the NOD-like receptors; NLRP1,NLRP2, NLRP3, NLRC4, NLRP6, NLRP7 and NLRP12 ^{[33, 114]}, and two members of the PYHIN (haematopoietic interferon-inducible nuclear protein) family; AIM2 and IFI16 are able to form inflammasome complexes ^[106] (Table 2). Most inflammasome complexes comprise of NLR or PYHIN proteins, apoptosis-associated speck-like protein containing a CARD (ASC) adaptor molecules, and procaspase-1. The structure of the inflammasome has been likened to that of the apoptosome, with nuclear magnetic resonance spectroscopy (NMR) and electron microscopy (EM) showing striking similarities between the NLRP1 inflammasome and the central hub of the Apaf-1 apoptosome ^[38].

Inflammasomes play a critical role in orchestrating multiple innate and adaptive immune responses ^[114] in response to pathogens and cellular stress, and in the recruitment of neutrophils into damaged tissue ^[46]. Inflammasome formation is essential for the production of the mature IL-1β and IL-18 cytokines and is seen as a two-step process ^[1]. Step one, is the priming step and involves TLR signaling to induce NF-kB transcription of proIL-1β, proIL-18 and inflammasome component such as; NLRP3 and NLRP1 ^{[3, 82]}. Step two; another stimulus leads to inflammasome oligomerization, caspase-1 auto activation, cytokine cleavage and mature IL-1β and Il-18 production and cellular release ^[46]. It is worth noting that one pathogen can activate multiple inflammasomes, for example; Listeria monocytogenes is able to activate NLRP3, AIM2 and NLRC4 ^{[65, 86, 138]}, while Candida albicans can activate NLRP3 and NLRC4 ^{[51, 129]}.

4.1. The ASC adaptor protein

The apoptosis-associated speck-like protein containing a CARD (ASC), also known as PYCARD ^[55], is as an essential adaptor protein and key mediator in apoptosis and inflammation ^{[9, 27]}. Structurally ASC consists of a CARD domain connected to a PYD domain by a flexible 23- residue linker ^[27]. Both domains are six- helix bundle folds, which are orientated back to back by the linker residue, therefore avoiding steric interference of binding sites, while enhancing the prospect of finding a binding partner^[27]. The absence of interdomain NMR-derived contacts (NOEs) confirms that PYD and CARD domains do not interact with one another ^[27].

ASC is localized to the nucleus and upon inflammatory stimulation is rapidly translocated to the cystol where it participates in inflammasome assembly ^[9]. Inflammasome formation can be abolished by preventing the redistribution of ASC, thereby affirming how important this adaptor protein is to inflammasome assembly ^[10].

Studies by Bryan et al ^[9] have identified 3 isoforms of ASC; ASC-b, ASC-c, and ASC-d, each showing the potential to regulate the inflammasome. The full length ASC and isoform ASC-b were able to co-localize with NLRP3 and were able to function as adaptor proteins, while ASC-c was regarded as inhibitory and only co-localized with caspase-1 not NLRP3, while ASC-d showed no co-localization and therefore did not impact on inflammasome function.

Table 2. The Pattern recognition receptors that form inflammasomes complexes, including their structural component and the most common alternative name used

Download as

View current table in a new window

Tables index

Veiw figure View Table

View previous table

View next table

4.2. The AIM2 and the Nuclear IFII6 Inflammasome

The cytosolic AIM2 inflammasome and the nuclear IFI16 inflammasome contain a N terminal PYD domain and a HIN200 domain ^[82] (Figure 1) and are known as true receptors, because they can directly bind their activating ligand; dsDNA (double stranded DNA) ^{[64, 106]}. AIM2 binds viral, bacterial, mammalian and synthetic dsDNA at multiple binding sites on the dsDNA, via the C-terminal HIN domain ^{[39, 52, 113]}. Binding occurs irrespective of DNA sequence or size, and once bound recruits the ASC adaptor molecule which binds to the PYD domain of AIM2 or IFI16, via PYD-PYD interactions ^[113]. The CARD domain of ASC then binds the CARD of procaspase-1, leading to caspase-1 activation and IL-1β maturation in proIL-1β presenting cells ^[113]. The mechanism by which the IFI16 inflammasome is able to sense viral/microbial DNA while remaining unresponsive to self DNA in the nucleus is unclear ^[106].

4.3. The NLRP1 Inflammasome

The NLRP1 inflammasome complex was one of the first to be identified, and is the only NLR to have a carboxy-terminal function-to-find domain (FIIND) and two signal-transduction domains: the CARD domain and the PYD domain ^[42] (Figure 1).The FIIND domain is a highly conserved protein region with striking structural similarities to the ZU5-UPA domain found in the autoproteolytic PIDD protein ^[25]. Based on amino acid sequencing the FIIND domain is present in only two human proteins; the NLRP1 and CARD8 ^[130]. The CARD8 protein is also known as TUCAN, CARDINAL or NPP1, and is often overexpressed in many types of cancers ^[130].

The NLRP1 protein structure is not conserved across species ^[40], with the mouse NLRP1 gene being polymorphic with three homologs: NLRP1a, NLRP1b and NLRP1c, whereas the human NLRP1 gene is singular ^[106]. The human NLRP1 protein is expressed in gut epithelial cells, granulocytes, dentritic cells and B cells, with weak expression level in monocytes and high expression levels in T cells and Langerhans cells ^{[35, 66]}.

Figure 1. The Nod-like receptors; NLRP1, NLRP3, NLRC4, NLRP6 and AIM2/IFI16 all form inflammasome that are, to date, relevant in inflammatory bowel processes.

Download as

Veiw figure View Figure

View next figure

So far, bacterial muramyl dipeptide (MDP) and lethal toxin (LeTx) produced by Bacillus anthracis have been shown to activate the NALP1 inflammasome ^{[35, 54, 82]}. B. anthracis, like other gram-positive bacteria is able to activate TLR2 and the anthrolysin O protein of B.anthracis is known to potently activate TLR4 ^[95].

Even though NLRP1 was one of the first inflammasomes to be identified, the mechanisms of NLRP1 activation are still being unraveled, several models have been proposed but many lack conclusive experimental evidence. Early studies by Faustin et al ^[38] using purified recombinant proteins defined the minimal requirements for caspase-1 activation in the NLRP1 inflammasome as being; MDP, NLRP1, ATP and procaspase-1, a process not requiring ASC, but enhanced in the presence of ASC.

More recent studies by D’Osualdo et al ^[25] have shown the FIIND domain to be a previously unrecognized type of ZU5-UPA domain which undergoes post-translational auto cleavage at phenylalanine and serine residues located at the C-terminus of the Zu5 domain. The glutamic acid and histidine residues, located in proximity of the cleavage site are involved in regulating the autoprocessing efficiency.

Only recently, has Finger et al ^[40] confirmed that the NLRP1 inflammasome is dependent on autolytic proteolysis within the FIIND domain, and after cleavage the two fragments remain associated producing a processed NLRP1. Dimers of ASC bound by PYD-PYD interactions are then recruited to the C-terminal CARD domain, which is in contrast to NLPR3 which recruits the ASC to the N-terminal PYD domain during inflammasome formation. The free remaining CARD domain binds to pro-caspase-1, leading to inflammasome formation, and production of mature IL-1β and IL-18 ^[40] (Figure 2).

Another study by Hsu et al ^[54] showed the sensing of B. anthracis and MDP, is linked to an association between NOD2 (structure; N-terminal-CARD-CARD-NACHT-LRR) and NLRP1 in activated cells, however it was unclear if direct sensing of MPD occurred between NOD2 or NLRP1 or both. The association of NOD2 and NLRP1 allows NOD2 to directly bind caspase-1 through N-terminal CARD domains resulting in the processing of IL-1β and IL-18. Interestingly, while MDP induced interactions were observed between NOD2 and NLRP1, no similar interactions were observed between NOD2 and NLRP3 during MDP induction.

Figure 2. NLRP1 inflammasome activation as proposed by Finger et al . The FIIND domain undergoes autolytic proteolysis resulting in a processed NLRP1. Dimers of ASC bind NLRP1 and procaspase- 1 via CARD interactions leading to mature IL-1β and IL-18 production

Download as

Veiw figure View Figure

View previous figure

View next figure

4.4. The NLRP3 Inflammasome

The NLRP3 inflammasome is by far the best characterized NLRP containing inflammasome and is often referred to as a global sensor because of its ability to activate upon exposure to a wide range of whole pathogens, environmental irritants and structurally diverse DAMPs and PAMPs ^{[1, 82, 114]} (Table 3). The NLRP3 protein contains 1016 amino acids, ^[70] and consists of a central NACHT domain flanked by an N-terminal PYD domain, and a C-terminal LRR domain ^{[70, 114]}.

Figure 3. Activation of the NLRP3 inflammasome begins with the dissociation of the chaperones;SGT1 and HSP90 from the LRR region. The addition of ASC via PYD-PYD interactions allows the binding of procaspase-1 and generation of mature IL-1β and IL-18.

Download as

Veiw figure View Figure

View previous figure

View next figure

In the gut, the NLRP3 protein is expressed in granulocytes, dentritic cell, monocytes, epithelial cells, T cells and B cells, with subcellular distribution localization mainly in the cytoplasm ^[66].

NLRP3 is held in an inactive, but signal competent state by two chaperone proteins located in the LRR namely; ubiquitin ligase-associated protein (SGT1) and heat-shock protein (HSP90) ^[85]. Ligand sensing causes dissociation of SGT1 and HSP90 resulting in a conformational change within NLRP3 which favors the recruitment of the ASC adaptor protein ^[85]. The NLRP3 protein binds to the ASC adaptor protein via PYD-PYD interactions, the CARD domain of the ASC binds the CARD domain of procaspase-1 leading to the proteolytic cleavage of procaspase-1 and the generation of mature IL-1β and IL-18 (Figure 3).

Lack of experimental evidence demonstrating direct binding of the NLRP3 inflammasome has led to the general hypothesis that the NLRP3 inflammasome senses changes in the cellular environment ^{[15, 26, 69, 106, 139]}. Assuming the mechanisms of NLRP3 activation are not mutually exclusive, three modelshave been proposed for the activation of NLRP3; the generation of mitochondrial-derived reactive oxygen species (mROS), ATP induced potassium efflux and phagolysosomal destabilization ^{[57, 74, 106]}.

4.4.1. Common Pathway of ROS Generation Activates the NLRP3 Inflammasome

Reactive oxygen species (ROS) are oxygen derived molecules that readily participate in the oxidation of other molecules. Intracellular ROS are usually derived from the one electron reduction of O₂, which give rise to the superoxide (O₂^-). Superoxide dismutases (SODs) are then responsible for converting O₂^- to hydrogen peroxide (H₂O₂) ^[115]. ROS are mainly produced in the mitochondria, however many other organelles produce ROS ^[141].

Mitochondrial ROS (mROS) are generally short lived molecules produced in response to cell stress, and by acting as a signal intermediate mROS evoke an appropriate cellular response to the stress ^{[92, 115]}. Stresses can include; hypoxia, starvation, pathogen infection (or molecular harbingers of pathogen infection), growth factor stimulation or membrane damage ^[115]. Cells with dysfunctional autophagy and mitophagy (autophagy of mitochondria) secrete high levels of ROS due to the accumulation of dysfunctional and damaged mitochondria ^[126].

Studies by Zhou et al ^[141] in concordance with previous studies proposed that NLRP3 senses mitochondrial dysfunction and activate accordingly. By inhibiting the electron transport of the mitochondria and inhibiting mitophagy, mROS production increased and this was shown to activate the NLRP3 inflammasome.

4.4.2. ATP Induced Potassium Efflux Activates the NLRP3 Inflammasome

ATP is used for many metabolic cellular processes and is therefore often released in cellular degradation. High levels of extracellular ATP activate the non-selective cation channel purinergic P2X₇ receptor causing it to open ^{[50, 98]}. Binding of ATP to the P2X₇ receptor induces potassium efflux and the formation of a larger accessory pore mediated by the hemichannel protein pannexin-1. The pannexin-1 pore is permeable to molecules 900kDa or greater ^{[46, 98]}.

Table 3. Most common activators of the NLRP3 inflammasome include a variety of self-derived and environmental activators. Pathogenic activators are mainly of bacteria origin however fungal and viral activation of the NLRP3 inflammasome have been demonstrated.

Download as

View current table in a new window

Tables index

Veiw figure View Table

View previous table

Studies by Petrilli et al ^[100] using LPS primed mouse macrophages has demonstrated that inflammasome inhibition can be achieved by the addition of 130mM extracellular potassium in response to nigericin, ATP, MSU and R837 (a synthetic imidazoquinoline compound mimicking ssDNA or ssRNA). Therefore indicating potassium efflux is essential for NLRP3 activation and suggesting the NLRP3 inflammasome senses the drop in intracellular potassium and this causes inflammasome activation ^{[80, 100]}.

More recent studies by Riteau et al ^[108] in human macrophages demonstrated that silica, uric acid and Alum crystals trigger the extracellular delivery of endogenous ATP a process which depends on the purinergic receptor and pannexin-1 channels. This demonstrates a role for ATP and purinergic receptor/pannexin-1 in formation of the NLRP3 inflammasome.

In light of these results, Riteau et al ^[108] proposed a new mechanism for NLRP3 activation whereby seemingly unrelated activators could be linked together via ATP release which depended on cell surface purinergic receptor P2X₇activation. It was also suggested the mechanistic link between eATP, purinergic receptors, pannexin-1 channels and inflammasome activation could be extended to include intracellular ATP produced by phagolysosomal destabilization and autophagic ATP release from dying cells ^{[46, 108]}.

4.4.4. Phagolysosomal Destabilization and Release of Cathespins B Activates the NLRP3 Inflammasome

Phagocytosis is a means whereby crystalline particles and pathogens can be absorbed into the cell ^[53]. The fusion of the phagosome with the lysosome during maturation forms the phagolysosome, and the destabilization of the phagolysosomal membrane and release of cysteine protease cathespin B into the cystol is thought to be sensed by NLRP3 ^{[14, 53, 57, 139]}. Sterile crystal activators such as aluminium ^[53], silica ^[53], MSU ^[83], calcium pyrophosphate dihydrate (CPPD) crystals ^[83], cholesterol ^[31] and amyloid β ^[49] are all known to activate NLRP3, a process dependent on lysosomal acidification and release of cathespin B. Supporting the theory of lysosomal destabilization Hornung et al ^[53] showed that artificial lysosomal damage was sufficient to activate the NLRP3 inflammasome, indicating the inflammasome senses the lysosomal disruption not the crystal activator.

4.5. NLRC4 Inflammasome

The NLRC4 inflammasome like the NLRP3 inflammasome is well characterised in the mouse system ^[106]. The NLRC4 protein consists of a central NACHT domain flanked by an N-terminal CARD domain and a C-terminal leucine rich region ^[97]. The NLRC4 inflammasome is activated by sequence motifs in basal body rod components of the type III (T3SS) and type IV secretion systems, and a similar sequence motif found in flagellin, a component of the flagellum apparatus ^{[41, 87]}.

Several gram-negative bacteria use secretion systems to deliver virulence factors or effector proteins to the cytosol where they modify the cellular environment to favorcolonization of the pathogen ^{[15, 89]}. NLRC4 detects basal body rod components of the T3SS secretory system in Salmonella typhimurium (PrgJ), Legionella pneumophila, Burkholderia pseudomallei (BsaK), Escherichia coli (EprJ and Esci), Shingella flexneri (Mxil) and Pseudomonas aeruginosa (Pscl) ^[89]. NLRC4 also detects leaked cytosolic flagellin from Listeria monocytogenes, Salmonella typhimurium, Pseudomonas aeruginosa and Legionella pneumophila ^{[87, 138, 140]}

Like NLRP3, it is proposed that NLRC4 is held in an inactive, but signal competent state by two chaperone proteins located in the LRR namely; ubiquitin ligase-associated protein (SGT1) and heat-shock protein (HSP90) ^[85]. Studies by Zhao et al ^[140] in mice have shown that bacterial sensing in the NLRC4 inflammasome is unique and involves the direct interaction of two accessory proteins namely; NAIP5 and NAIP2. NAIP5 is a BIR domain containing NLR protein which directly and specifically interacts with bacterial flagellin and promotes a physical NAIP5-NLRC4 interaction. Similarly, the BIR domain containing NLR protein; NAIP2 acts as a specific inflammasome receptor for secretory system rod proteins like PrgJ in Salmonella typhimurium and promotes a NAIP2- NLRC4 interaction ^[140]. In humans only one NAIP family member exists which suggests other proteins could be involved in a similar sensing mechanism ^{[34, 106]}. Zhao et al ^[140] also showed that in humans U937 monocyte derive macrophages, human NAIP recognized the TTSS needle subunit Cprl from Chromobacterium violaceum and was able to activate the NLRC4 inflammasome.

Following bacterial sensing the NLRC4 is phosphorylated between the NACHT domain and the LRR at Ser533 by kinases like PKCδ and this drives a conformation change that necessitates inflammasome formation ^[103]. Unlike the other inflammasome forming NLRs, NLRC4 contains a CARD domain and can theoretically bind procaspase-1 independently of ASC via CARD-CARD interaction, however the ASC adaptor protein enhances IL-1β production as shown by partial defects in IL-1β secretion in response to cytoplasmic flagellin by ASC deficient bone marrow derived macrophages ^[87]. Binding of procaspase-1 to the NLRC4 causes proteolytic cleavage of procaspase-1 and the generation of mature IL-1β and IL-18 ^[103].

4.6. NLRP6 Inflammasome

The NLRP6 is a more recently discovered inflammasome and as such is still being characterised ^[34]. The NLRP6 protein consists of a central NACHT domain flanked by a N-terminal PYD domain and a C-terminal leucine rich region ^[15] (Figure 1). Early in vitro studies have showed that NLRP6 colocalizes with ASC via PYD-PYD interactions, and the co-expression of NLRP6 with ASC induces a caspase-1 dependent processing of IL-1β ^[47] therefore indicating that procaspase-1 more than likely binds to the CARD domain of NLRP6 via CARD-CARD interactions.

In the gut, NLRP6 is highly expressed in colon epithelial cells and to a lesser extent in the lamina propria cells of the granulocyte and monocyte lineage ^[13]. The signal that activates the NLRP6 inflammasome is still unknown and the function of the NLRP6 is only recently beginning to be unraveled ^{[36, 133]}.

Recent studies by Elinav et al ^[36] have shown the NLRP6 inflammasome plays a role in intestinal homeostasis by regulating the microbial flora within the gut. The genetically ablation of NLRP6 in mice results in altered fecal microbiota (with a predominance of TM7 and Prevotellaceae species), reduced inflammasome activation and increased susceptibility to colitis and tumorigenesis. By cohousing NLRP6 deficient mice with wild type mice the colitis forming microbiota of TM7 and Prevotellaceae could be transferred to mice with normal immune systems and from one generation to the next ^[36]. In addition, NLRP6 -/-, ASC-/- and caspase-1 -/- mice all failed to upregulate the IL-18 cytokine necessary for stimulating epithelial cell barrier function and epithelial cell regeneration after DSS administration ^[36]. Furthermore, untreated NLRP6-/-, ASC-/- and IL-18-/- mice all showed epithelial cell upregulation of the chemotactic cytokine CCL5 which plays an pivotal role in recruiting proinflammatory immune cells and chronic inflammation ^[133] (Figure 4). Interestingly NRLP6, ASC and caspase-1 in rats is absent prior to embryonic age day 20 and processed IL-18 emerges in the postnatal period which coincides with gut colonization ^[63].

4.7. Inflammatory Caspases

Caspases are a group of cysteine-aspartic proteases which play essential roles in the cellular processes of apoptosis (programmed cell death), necrosis and inflammation ^{[32, 139]}. Caspase-1 is a prominent member of the human inflammatory caspase subgroup which includes; caspase-1, caspase-4, caspase-5, caspase-11 and caspase-12 ^{[84, 139]} structurally caspase-1 consists of a central p20 subunit flanked by a single CARD domain at the N-terminus and a small catalytic subunit at the C-terminus. Caspase-1 is synthesized as an inactive 45 kDa zymogen pro-caspase-1, and requires inflammasome activation to become proteolytically active in order to cleave proIL-1β and proIL-18 at recognition sites adjacent to aspartic acid residues, resulting in mature IL-1β and IL-18 production ^[139]. Processed caspase-1 can also cleave the precursors of IL-33 and IL-I7 ^[30].

Figure 4. When the NLRP6 inflammasome is present mature IL-18 is secreted and acts to stimulate the production of IFNɣ in subset of T cells which promotes macrophage activity and bacterial clearance. When NLRP6 is absent the intestinal flora stimulates epithelial cells to produce the chemotactic cytokine CCL5 which attacts immune cells like neutrophils causing a chronic inflammatory response, culminating in colitis,

Download as

Veiw figure View Figure