Apoptosis

, 14:522

Inflammasomes in infection and inflammation

Authors

  • Christian R. McIntire
    • Department of BiochemistryMcGill University
  • Garabet Yeretssian
    • Department of Medicine, Division of Critical Care, and Centre for the Study of Host ResistanceMcGill University
    • Department of BiochemistryMcGill University
    • Department of Medicine, Division of Critical Care, and Centre for the Study of Host ResistanceMcGill University
Cell Death and Disease

DOI: 10.1007/s10495-009-0312-3

Cite this article as:
McIntire, C.R., Yeretssian, G. & Saleh, M. Apoptosis (2009) 14: 522. doi:10.1007/s10495-009-0312-3

Abstract

Two of the main challenges that multicellular organisms faced during evolution were to cope with invading microorganisms and eliminate and replace dying cells. Our innate immune system evolved to handle both tasks. Key aspects of innate immunity are the detection of invaders or tissue injury and the activation of inflammation that alarms the system through the action of cytokine and chemokine cascades. While inflammation is essential for host resistance to infections, it is detrimental when produced chronically or in excess and is linked to various diseases, most notably auto-immune diseases, auto-inflammatory disorders, cancer and septic shock. Essential regulators of inflammation are enzymes termed “the inflammatory caspases”. They are activated by cellular sensors of danger signals, the inflammasomes, and subsequently convert pro-inflammatory cytokines into their mature active forms. In addition, they regulate non-conventional protein secretion of alarmins and cytokines, glycolysis and lipid biogenesis, and the execution of an inflammatory form of cell death termed “pyroptosis”. By acting as key regulators of inflammation, energy metabolism and cell death, inflammatory caspases and inflammasomes exert profound influences on innate immunity and infectious and non-infectious inflammatory diseases.

Keywords

InflammasomeInfectionInflammationCaspasesCell death

Abbreviations

ICE

Interleukin-1β Converting Enzyme

IL-1

Interleukin-1

Asp

Aspartic acid

Ala

Alanine

Cys

Cysteine

His

Histidine

Gly

Glycine

Ser

Serine

Arg

Arginine

Glu

Glutamic acid

IGIF

Interferon-γ-inducing factor

ST2

Interleukin 1 receptor-like 1

TH1

Type 1 helper T cells

TH2

Type 2 helper T cells

TIM

Triose-phosphate isomerase

GAPDH

Glyceraldehyde-3-phosphate dehydrogenase

SREBP

Sterol regulatory element binding protein

iTRAQ

Isobaric tag for relative and absolute quantitation

FGF2

Fibroblast growth factor 2

Bid

BH3-interacting domain death agonist

AIP-1/WDR-1

Atrophin-interacting-protein 1/WD repeat domain 1

LD

Lethal dose

SIRS

Systemic inflammatory response syndrome

LPS

Lipopolysaccaride

TNFα

Tumor necrosis factor α

SAP

Severe acute pancreatitis

DSS

Dextran sodium sulfate

ARF

Acute renal failure

CTCL

Cutaneous T-cell lymphoma

VCAM-1

Vascular cell adhesion molecule-1

MS

Multiple sclerosis

EAE

Autoimmune encephalomyelitis

TLR

Toll-like receptor

NLR

Nod-like receptor

PAMP

Pathogen-associated molecular pattern

Nod

Nucleotide binding and oligomerization domain

NALP

NACHT domain- leucine-rich repeat-, and PYD-containing protein

ASC

Apoptosis-associated speck-like protein containing a CARD

Cardinal

CARD inhibitor of NF-kB-activating ligands

LRR

Leucine-rich repeat

CARD

Caspase recruitment domain

IPAF

ICE protease-activating factor

BIR

Baculovirus IAP Repeat

NAIP

Neuronal apoptosis inhibitory protein

PYD

Pyrin domain

NBS

Nucleotide binding site

DF

Death-fold

CIITA

Class II trans-activator

LCV

Legionella containing vacuole

Icm/Dot

Intracellular multiplication/defective organelle trafficking

Lgn

Legionella

IFN

Interferon

SipB

Salmonella invasion protein B

TTSS

Type III secretion system

IpaB

Invasin B

DAMP

Danger-associated molecular pattern

ATP

Adenosine triphosphate

iPLA2

Ca2+-independent phospholipase A2

MSU

Monosodium urate

CPPD

Calcium pyrophosphate dihydrate

ROS

Reactive oxygen species

NADPH

Nicotinamide adenine dinucleotide phosphate-oxidase

P22phox

Protein 22 phagocyte and oxidase

OVA

Ovalbumin

MDP

Muramyl dipeptide

HSP90

Heat shock protein 90

NF-κB

Nuclear factor-kappa B

COP

CARD-only protein

INCA

Inhibitory CARD

RIP2

Receptor-interacting protein 2

FMF

Familial Mediterranean fever

vPYD

Viral PYD

Bcl-2

B-cell lymphoma 2

Bcl-XL

Basal cell lymphoma-extra large

IKKβ

I kappa B kinase β

LOF

Loss of function

GOF

Gain of function

BLS-II

Type II bare lymphocyte syndrome

SCID

Severe combined immunodeficiency

MHC-II

Major histocompatibility complex class II

FCU

Familial cold urticaria

MWS

Muckle–Wells syndrome

NOMID

Neonatal-onset mutli-system inflammatory disease

CIAS1

Cold-induced auto-inflammatory syndrome 1

CD

Crohn’s disease

Caspase-1: an overview

The biochemistry of caspase-1

Caspase-1, previously known as Interleukin-1β Converting Enzyme (ICE), is a member of a large family of intracellular aspartate specific cysteine proteases. Caspase-1 was discovered during attempts to elucidate the enzyme responsible for the processing of pro-IL-1β. Synthesized as a cytosolic 31 kDa precursor, pro-IL-1β is processed by caspase-1 into a 17.5 kDa product, which represents its mature biologically active form [14]. Based on the activity of various class-specific inhibitors, caspase-1 was characterized as a cysteine protease with a high degree of substrate specificity, cleaving pro-IL-1β between amino acid residues Asp116 and Ala117 [2, 3]. It was later purified from monocytic cells, and was found to be a heterodimer composed of 20 kDa and 10 kDa subunits (p20 and p10) present in an equimolar ratio and derived from a 45 kDa (p45) precursor [57]. Cloning of caspase-1 indicated that the residues around the active site had limited homology with several serine proteases [5, 8] and that its catalytic fold was novel [9]. X-ray crystallography of caspase-1 with an inhibitor substantiated that caspase-1 was multimeric, with two p20 and p10 polypeptides, and showed that caspase-1 has a tertiary structure of the form (αβ)2 [9, 10]. Selective labeling with iodoacetic acid revealed that the active site of the enzyme is composed of residues from both the p20 and p10 subunits, the catalytic cysteine being Cys285 in the C-terminus of the p20 and the major determinants of substrate specificity being found in the p10 [5]. The catalytic site of the enzyme includes residues His237 and Cys285, as well as Gly238 and Ser339 that stabilize the catalytic site orientation, and the deep substrate-binding pocket is comprised of residues Arg179 and Arg341, as well as Glu283 and Ser347 [9, 10]. Caspase-1 is activated via oligomerization and its active conformation is then stabilized through two autocatalytic processing events, one between the small and large subunits and the second cleaving the prodomain [1013]. Using synthetic peptide substrates and recombinant pro-IL-1β, it was demonstrated that caspase-1 has an absolute requirement for Asp in the P1 position of the substrate immediately N-terminal to the scissile bond, while it can tolerate conservative substitutions in the P2 and P3 positions [14]. Based on substrate specificity, determined by the nature of the amino acid at position P4, caspases can be subgrouped into different subfamilies. Caspase-1 belongs to the inflammatory caspase subfamily, which includes caspases-1, -4, -5, (-11, in rodents), and -12 [15]. With the exception of caspase-12, the inflammatory caspases have a preference for a hydrophobic amino acid, such as a tryptophan or tyrosine, at the P4 position and are able to cleave synthetic substrates with the sequence WEHD or YVAD [16]. Among caspase-1’s natural substrates are members of the IL-1 family, which include pro-IL-1β, pro-IL-18, also known as interferon-γ-inducing factor (IGIF) [17], pro-IL-33 [18], and the related IL-1 family member IL-1F7b (also known as IL-1H4) [19]. IL-1β and IL-18 are potent proinflammatory cytokines that regulate immune and inflammatory responses. At physiological concentrations, IL-1β is essential for the activation of mononuclear and endothelial cells, which renders it a crucial effector of the host response to pathogens. However, at high systemic concentrations, its effects become potentially lethal. Deregulated production of IL-1β is associated with severe pathological conditions [20], including rheumatoid arthritis, inflammatory bowel disease, sleep sickness, acute and chronic myelogenous leukemia, insulin-dependent diabetes mellitus, atherosclerosis, asthma, and septic shock [21]. IL-18 induces interferon-γ production and promotes the differentiation of type 1 helper T cells (Th1) [20]. In comparison, IL-33, a ligand of the ST2 receptor, plays a role in allergic type diseases via mediation of type 2 helper T cell (Th2) responses [18].

Functions of caspase-1

Caspase-1 is not merely an ICE, as its functions are not restricted to processing of IL-1 family cytokines. Caspase-1 is implicated in the induction of macrophage cell death by certain bacteria. Earlier studies with caspase-1 reported that when overexpressed in fibroblasts, caspase-1 can induce apoptosis [22]. However, caspase-1-deficient mice show no overt defect in apoptosis [23, 24]. Nonetheless, macrophages infected with intracellular pathogens die within minutes following infection, stimulating a unique form of host cell death distinct from apoptosis that is characterized by rapid cell lysis with release of proinflammatory intracellular contents [2537]. This type of caspase-1 dependent cell death is termed “pyroptosis”, from the Greek roots “pyro”, meaning fire, which denotes the release of proinflammatory mediators, and “ptosis”, meaning falling, which is commonly used to describe cell death [38]. In addition to roles involving proinflammatory cytokine processing, caspase-1 cleaves a number of cellular substrates including actin [39], the kinase PITSLRE [40], parkin [41], pyrin [42], and caspase-7 [43]. More recently, the caspase-1 digestome was determined [44], and caspase-1 was shown to target multiple cellular pathways by processing more than 40 different proteins. Most notably, caspase-1 cleaves the glycolysis enzymes aldolase, triose-phosphate isomerase (TIM), glyceraldehyde-3-phosphate dehydrogenase (GAPDH), α-enolase, and pyruvate kinase [44]. The enzymatic activity of GAPDH is dampened by caspase-1 processing, and Salmonella typhimurium infection impairs glycolysis in wild-type cells but not in caspase-1-deficient cells. Because glycolysis is essential for macrophage activation and survival [45] the cleavage of glycolysis substrates and reduction of cellular glycolysis may be an essential step toward cell death. A role for caspase-1 has also been defined in activating sterol regulatory element binding proteins (SREBPs) to promote lipid biogenesis in response to microbial pore-forming toxins [46]. More recently, caspase-1 has also been implicated in unconventional protein secretion [47]. iTRAQ proteomic analysis identified multiple proteins that are secreted upon activation of caspase-1, including IL-1α, FGF2, uncleaved Bid, Aip-1/Wdr-1, annexin A2, and peroxiredoxin-1. These results come to confirm earlier findings, which showed that although IL-1α is not a caspase-1 substrate, its secretion depended on a functional caspase-1. Indeed, casp-1−/− mice, which are deficient in IL-1β production, also fail to produce normal levels of IL-1α in response to endotoxemia [4, 23, 24]. Caspase-1 is shown to bind directly to pro-IL-1α, as well as other leaderless proteins that lack a signal peptide, and mediate their non-classical secretion. Altogether, these functions suggest that caspase-1 activation leads to the processing and release of various proteins with functions in inflammation, cytoprotection and tissue repair to help restore cell homeostasis after tissue damage or stress or in response to infection.

Caspase-1 in the host response to pathogens

Caspase-1-deficient mice are susceptible to bacterial infections with Escherichia coli, Shigella flexneri, Salmonella typhimurium and Francisella tularensis, but have partial defects in their response to Listeria monocytogenes or the fungal pathogen Candida albicans. In an experimental model of sepsis induced by the intra-peritoneal injection of live E. coli, caspase-1-knockout mice succumbed to infection with a three- to fourfold lower LD50 (50% lethal dose) compared to wild-type or caspase-1 heterozygous mice [48]. Similarly, wild-type mice treated with the caspase-1 inhibitor Ac-YVAD-CHO are more sensitive to E. coli infection than untreated mice, and administration of recombinant murine IL-1β and IL-18 to caspase-1-null mice restores resistance to infection. Intranasal infection with S. flexneri, the agent of bacillary dysentery, leads to lethality of caspase-1-null mice but not their wild-type counterparts. Deficiency in caspase-1 leads to a more severe infection with 10- and 670-fold more bacteria at 24 and 48 h post-infection, respectively, in casp-1−/− compared to wild-type mice [49]. As with E. coli and S. flexneri, caspase-1-deficient mice show increased susceptibility to oral infection with S. typhimurium, a shorter time to death than control animals, and display significantly higher bacterial loads in the Peyer’s patches, mesenteric lymph nodes and spleens [32, 50]. They also exhibit increased bacterial burdens in the spleen and liver following intraperitoneal infection with S. typhimurium [50]. Caspase-1 functions appear to be equally crucial for the host resistance to F. tularensis, the etiological agent of tularemia. Caspase-1-deficient mice infected with F. tularensis show markedly increased bacterial burdens and mortality compared to wild-type mice [31]. Casp-1−/− mice also have an impaired ability to clear L. monocytogenes early in infection; however, they successfully clear the pathogen following secondary infection. On the contrary, caspase-1 has not proven critical in the response to primary systemic infections with C. albicans, but its deficiency greatly impairs the Th1-mediated resistance upon reinfection [51].

Caspase-1 and inflammatory diseases

While caspase-1 is essential for the host defense to pathogens, its activity is associated with a variety of inflammatory conditions including, in the most severe cases, systemic inflammatory response syndrome (SIRS) and septic shock. Caspase-1-deficient mice are resistant to multiple inflammatory models including endotoxemia [23, 24], peritonitis [52, 53], pancreatitis [54], and colitis [55]. Caspase-1 has important roles in the pathogenesis of various other inflammation-promoted diseases including acute renal failure (ARF), metastatic melanoma, cutaneous T-cell lymphoma (CTCL), multiple sclerosis (MS), arthritis and asthma. Caspase-1 protein levels and activity are markedly upregulated following ischemic and hypoxia-induced ARF [56, 57]. Genetic or pharmacologic depletion of caspase-1 protects the kidney from these failures and results in decreased IL-18 and neutrophil infiltration as well as less severe tubular necrosis [5861]. In metastatic melanoma, overexpression of human caspase-1 in the keratinocytes of transgenic mice results in spontaneous development of recalcitrant dermatitis and skin ulcers, which correlates with increased levels of mature IL-1β and IL-18 [62]. Caspase-1 null mutations were also found to significantly reduce hepatic melanoma metastases, by reducing the expression of vascular cell adhesion molecule-1 (VCAM-1) and cell adherence [63]. Moreover, serum caspase-1 levels are significantly increased in human metastatic melanoma patients compared to normal cohorts, particularly in biochemorefractory patients compared to responding ones [64]. Similarly, plasma levels of caspase-1 and IL-18 are elevated in patients with CTCL, a malignancy of skin homing Th2 T-cells, and mRNA for these factors are also upregulated in skin lesions from these patients [62]. In MS, a chronic inflammatory demyelinating disease of the central nervous system, caspase-1-deficient mice show a reduction in the incidence and severity of disease as observed in experimental models of autoimmune encephalomyelitis (EAE) [65]. Caspase-1 is induced in EAE with its levels correlating with the transcriptional rate of proinflammatory cytokines. Moreover, studies involving human cohorts identified that caspase-1 transcriptional activity was significantly upregulated in the peripheral blood mononucleocytes of patients with MS compared to healthy controls [65, 66]. Similarly, elevated caspase-1 levels are found in the synovial fluids of patients with idiopathic juvenile arthritis, and inhibition of caspase-1 using irreversible inhibitors blocks the progression of type II collagen-induced arthritis in mice [67, 68]. Altogether, these results indicate that levels of caspase-1 may be a reliable marker of ongoing immune-inflammatory responses.

Discovery of the inflammasome

The oligomerization of initiator caspases into close proximity induces their activation. In 2002, the molecular platform that oligomerizes and activates caspase-1 was characterized. Based on its function in triggering pro-IL-1β processing, this complex was appropriately named the “inflammasome” [69]. The inflammasome was first described as being composed of the intracellular receptor NALP1 and its adaptor protein ASC. NALPs belong to the large NOD-like receptor (NLR) family of pattern recognition receptors which sense “danger” signals and microbial motifs, and activate inflammatory and anti-microbial responses [70, 71]. Based on phylogenetic analysis, the NLR family is subdivided into several distinct subfamilies including the CIITA, nucleotide binding and oligomerization domain (NOD), IPAF and NALP subfamilies. Most NLR members are characterized by 3 functional domains, including a central oligomerization domain (NACHT), a C-terminal ligand sensing leucine-rich repeats (LRRs) domain, and an N-terminal caspase binding region, which is a CARD (Caspase recruitment domain) in IPAF and NOD proteins, three BIRs (Baculovirus IAP Repeats) in NAIP, or a PYD (pyrin domain) in NALP proteins. The NACHT domain, also termed a nucleotide binding site (NBS) or NOD, facilitates the formation of oligomers. LRRs are short motifs 22–28 residues in length that provide a versatile framework for the formation of protein-protein or protein-carbohydrate/lipid interactions [72]. LRRs are also found in Toll-like receptors (TLRs), which survey the extracellular and endosomal environments for pathogenic or danger signals [73, 74]. The CARD and PYD domains are members of the death-fold (DF) family, and are characterized by 6 α-helices that are tightly packed in a Greek key fold [75]. Proteins containing a DF domain bind to other members of the same family via homotypic interactions [76]. In the NALP1 inflammasome, NALP1 recruits caspase-1 through the adaptor molecule ASC. ASC, also known as PYCARD, is a bimodular adaptor composed of a N-terminal PYD and a C-terminal CARD, that mediates interaction with NALPs through a PYD–PYD interaction and with caspase-1 through a CARD–CARD interaction [77]. A C-terminal CARD domain in NALP1 binds and activates caspase-5, bringing it into close proximity to caspase-1, thereby facilitating cross-activation. The NALP3 inflammasome is akin to NALP1 in that it requires ASC for caspase-1 activation. However, NALP3 and NALP2 (and possibly all other NALPs) also require a second adaptor known as CARDINAL. CARDINAL is needed to recruit a second molecule of caspase-1, which forms a dimer with the caspase-1 engaged by the ASC adaptor [78]. No murine homolog of CARDINAL has been identified as of yet.

The inflammasome in the host response to infection

The investigation of the inflammasome in the host response to infection with various pathogens has been widely reported in the last few years. Here we present few representative examples of pathogens sensed by NLRs and fought by inflammasomes.

The NAIP5 inflammasome senses L. pneumophila

Legionella pneumophila, the causative agent of Legionnaire’s disease, is a facultative Gram-negative bacteria that can invade and replicate in amoebae and human macrophages. Following macrophage internalization, L. pneumophila “hides” and multiplies in an ER membrane-bound phagosome termed the Legionella containing vacuole (LCV) [79]. The bacteria use a type IV secretion system, known as Icm/Dot, to inject virulence factors into the host cytosol. These effector proteins interfere with the phagosome-lysosome fusion hence, increasing the ability of the bacteria to survive in host cells. Human cells as well as A/J mouse strain macrophages are permissive to L. pneumophila replication. In mice, resistance to infection is controlled by the recessive Lgn1 locus found on mouse chromosome 13 [8082]. Genetic and physical mapping studies indicated that the Lgn1 genomic region contains multiple copies of the Naip/Birc1 gene and provided the first evidence that NAIP proteins might be important in the host resistance to L. pneumophila [83, 84]. Many polymorphisms in the Naip5/Birc1e gene were subsequently found, that modify the Lgn1 phenotype, and account for the enhanced susceptibility to infection in A/J mice [85, 86]. Several studies have demonstrated the formation of a NAIP5-caspase-1 inflammasome in L. pneumophila non-permissive cells, which leads to pyroptosis and elimination of infected cells. In permissive cells, and because of NAIP5 polymorphisms, inflammasome assembly and caspase-1 activation occur inefficiently. Inflammasome activation by L. pneumophila is triggered by bacterial flagellin reaching the cytosol via the Dot-Icm secretion system [33, 34, 8789]. In addition to NAIP5, IPAF, which is involved in sensing bacterial flagellin, is required to limit bacterial growth in macrophages [90, 91]. Consistently, Ipaf−/− macrophages are permissive to intracellular L. pneumophila replication. It was recently suggested that restriction of bacterial growth requires IPAF-dependent activation of caspase-1 and caspase-1-independent NAIP5 signaling [92]. However, the recent characterization of Naip5−/− mice indicated critical functions of NAIP5 in inflammasome activation. Using retrovial transduction systems to express L. pneumophila flagellin, without bacterial contaminants or secretion systems, this study also demonstrated that NAIP5 recognizes a conserved domain of flagellin 35 amino acid of the carboxyl terminus. These results strengthen earlier findings that both NAIP5 and IPAF are necessary for inflammasome activation [93]. Nonetheless, the pathways and mechanisms involved in the restriction of Legionella growth are far more complex than anticipated and involve crosstalk among different innate immunity pathways. Consistently, it was recently shown that the inflammasome synergizes with signaling pathways involving TNF and type I IFN to control L. pneumophila growth [94].

The IPAF inflammasome stimulates the host response to S. typhimurium, S. flexneri, P. Aeruginosa in flagellin-dependent and -independent manners

For a number of pathogens, activation of caspase-1 is an important host defense mechanism. Caspase-1 deficiency renders mice susceptible to infection with the enteropathogen S. typhimurium (see above). Initially, it was described that caspase-1 activation by S. typhimurium was mediated via direct binding of the Salmonella invasin SipB to caspase-1 [27, 95]. However, this idea was quickly abandoned with the finding that Salmonella flagellin reaching the host cell cytosol was the agonist stimulating caspase-1 activation. Macrophages lacking the flagellin sensing NLR, IPAF, are defective in S. typhimurium-induced cell death and IL-1β release [29, 90, 91]. In contrast, Nalp3−/− macrophages exhibited an intact response, activating caspase-1 and secreting normal levels of IL-1β and IL-18 [96, 97].

As with S. typhimurium infection, activation of caspase-1 by the related enteric pathogen S. flexneri was originally thought to involve the type III secretion system (TTSS) translocator IpaB, a protein biochemically-related to Salmonella’s SipB. We now know that these proteins do not activate caspase-1 directly but are needed for the translocation of flagellin into the host cell cytosol. Interestingly, the activation of caspase-1 in S. flexneri-infected macrophages is IPAF-dependent but flagellin-independent [98]. S. flexneri induces inflammation and inflammation-associated cell death by engaging multiple NLRs. In macrophages, the IPAF inflammasome induces caspase-1 activation and caspase-1-dependent pyroptosis, but at higher bacterial levels, NALP3 is additionally triggered and induces necrosis (pyronecrosis) in a cathepsin B-dependent fashion [99]. As in S. flexneri infection, flagellin-independent activation of the IPAF inflammasome has been recently reported in response to P. aeruginosa [35], suggesting that other PAMPs, or alternatively other bacterial effector proteins, might have an impact on IPAF activation.

Inflammasome in danger

NLR activation and subsequent caspase-1 activation and IL-1β release from macrophages in response to bacterial invasion or danger signals constitute important events in inflammatory and innate immune responses. However, whether NLRs act as direct receptors of PAMPs or DAMPs remain uncertain. It is hypothesized that rather than being bonafide “receptors”, NLRs might function as “guard” proteins, similarly to R proteins in plants, associating with receptors and transducing the danger signal into effector mechanisms such as inflammation and cell death.

ATP and pore-forming toxins

The NALP3 inflammasome activates caspase-1 and induces IL-1β release in response to a wide range of structurally unrelated molecules, including among others microbial toxins such as nigericin, maitotoxin and aerolysin, as well as high concentrations of ATP (Fig. 1). Through different molecular mechanisms, these triggers lead to K+ efflux, which appears to be the common determinant for caspase-1 activation in the cell. ATP activates the purinergic receptor P2X7 resulting in plasma membrane depolarization, cell swelling, loss of cytoskeletal organization and K+ efflux [100102]. The role of P2X7 in mediating caspase-1 activation is evidenced by the inability of P2X7-deficient cells to release IL-1 in response to ATP [102]. Additionally, cells treated with P2X7 antagonists are similarly defective in IL-1 release [103]. Potassium efflux leads to calcium influx and activation of various phospholipases involved in the sequence of events culminating in IL-1 maturation and secretion [104, 105]. While Ca2+-independent phospholipase A2 (iPLA2) has been reported to be involved in pro-IL-1β processing after P2X7 engagement, other phospholipases are also required for lysosome exocytosis and IL-1β secretion [103, 105]. Following rapid opening of the potassium channel, P2X7 activation leads to the formation of a larger pore mediated by the protein pannexin-1 [106]. Pannexin-1 was shown to be essential for caspase-1 activation in LPS-stimulated macrophages pulsed with ATP or stimulated with the pore-forming toxins nigericin and maitotoxin [107, 108]. Altogether, it appears that low intracellular K+ concentrations are sensed as a danger signal triggering NALP3 inflammasome assembly and caspase-1 activation. However, the molecular mechanism by which a drop in intracellular K+ results in NALP3 inflammasome formation and activation remains unclear.
https://static-content.springer.com/image/art%3A10.1007%2Fs10495-009-0312-3/MediaObjects/10495_2009_312_Fig1_HTML.gif
Fig. 1

Pathways for activation of the inflammasome by PAMPs and DAMPs. Multiple pathways are responsible for the activation of the caspase-1 inflammasomes. Pores formed in the host cell membrane by pore-forming toxins and ATP-P2X7-activated pannexin-1 lead to K+ efflux, Ca2+ influx and entry of extracellular PAMPs into the host cytosol, which activate Nalp inflammasomes. S. typhimurium, S. flexneri and L. pneumophila require a functional type III or type IV secretion system to secrete flagellin into the host cell cytosol, which activates the IPAF inflammasome. Phagocytosis of MSU, CPPD, silica, asbestos and aluminum salts leads to phagocytosis of the crystals followed by lysosomal rupture, which results in the release of cathepsin B into the cytosol and activates the NALP3 inflammasome. Crystals that are too large to be ingested remain at the cell surface where they induce membrane perturbations or “frustrated phagocytosis”. Reactive oxygen species (ROS), produced through the actions of membrane-bound NADPH oxidase, are implicated in transducing the signal from the membrane to the Nalp3 inflammasome. Inflammasome assembly stimulates the activity of caspase-1, which mediates processing of cytokines, glycolysis enzymes and other substrates, pyroptosis, unconventional protein secretion, and promotion of cell survival through activation of SREBPs in response to pore-forming toxins

Crystalline structures, adjuvants and fibers

In addition to ATP and pore-forming toxins, crystalline structures as well as protein aggregates have been recently reported to activate NALP3 signaling. Monosodium urate (MSU) and calcium pyrophosphate dihydrate (CPPD) crystals, which are linked to the auto-inflammatory disorders gout and pseudogout, respectively, were first shown to activate the NALP3 inflammasome [109]. Caspase-1, ASC or NALP3-null mice are resistant to MSU crystal-induced peritonitis and macrophages derived from these strains are deficient in IL-1β production. Mechanistically, caspase-1 activation occurs following endocytosis of the crystals, as confirmed by the use of the endocytosis inhibitor colchicine. The airborne polluants, asbestos and silica, which are implicated in inflammation of the lung, fibrosis, and lung cancer trigger inflammation through activation of the NALP3 inflammasome [110112]. Macrophages derived from NALP3 or ASC deficient mice are defective in caspase-1 activation and IL-1β production in response to physiological amounts of these crystalline structures [111113]. It was hypothesized that because some of these crystals are too large to be fully phagocytosed, they remain trapped at the cell surface and signal to the inflammasome through membrane perturbations and reorganization of the actin filaments, a process termed “frustrated phagocytosis” [111] (Fig. 1). In support of this hypothesis, disruption of the actin cytoskeleton with cytochalasin D inhibits the ability of MSU and asbestos to induce IL-1β release. Crystal-induced NALP3 activation also involves reactive oxygen species (ROS) production and NADPH oxidase activity [111, 112]. Indeed, asbestos and silica crystals were previously shown to spontaneously catalyze the formation of ROS in aqueous solutions or after uptake by cells [110]. Moreover, pharmacological inhibitors of NADPH oxidase, ROS detoxifying proteins or knockdown of p22phox by short hairpin RNA all result in reduced caspase-1 activation and diminished IL-1β production [111, 112]. In a model of asbestos inhalation, Nalp3−/− mice are resistant to asbestosis, showing impaired recruitment of inflammatory cells to the lungs and lower cytokine production [111].

More recently, aluminum salts, and the amyloid-β peptide were also shown to induce the activation of the NALP3 inflammasome [114, 115]. Aluminum adjuvants or “Alum” are the most commonly used vaccine additives worldwide. When combined with soluble antigens in vaccines, adjuvants boost innate immunity required for efficient adaptive immune responses. It was demonstrated that alum does not require intact TLR signaling to activate the immune system [116, 117]. Using mice deficient in NALP3, ASC or caspase-1 supported the idea that all inflammasome components are required for alum adjuvanticity in an ovalbumin (OVA) immunization model [114, 118, 119]. Mechanistically, lysosomal permeabilization and rupture following alum particle uptake occurred upstream of inflammasome activation. It was shown that the release of cathepsin B into the cytosol was necessary for inflammasome activation [113]. Similarly, phagocytosis of amyloid-β fibers, which are the main constituent of amyloid plaques in the brains of Alzheimer’s disease patients [120] triggers NALP3-dependent caspase-1 activation and production of pro-inflammatory cytokines [115].

Other DAMPs and PAMPs

Keratinocytes are an important physical barrier and as such are armed with the necessary molecules needed to defend the host against invading pathogens or environmental assaults such as UVB irradiation. NALP1, ASC and caspase-1 are constitutively expressed in keratinocytes, and UVB activates caspase-1 in keratinocytes through the NALP1 inflammasome [121, 122]. Depletion of the different inflammasome components (NALP1, ASC or pro-caspase-1) by siRNA strongly suppresses IL-1β release from UV irradiated keratinocytes [122].

As previously described, PAMPs generally activate TLRs and stimulate pro-IL-1β accumulation in cells; but when they reach the cytosol, some PAMPs behave like danger signals and provide the necessary information essential to stimulate inflammasome assembly. MDP is such an example. It induces activation of the inflammasome via NALP1, NOD2 and NALP3 [121, 123, 124]. MDP challenge induces the formation of a NOD2/NALP1 complex that mediates caspase-1-dependent IL-1β secretion [125].

Molecular regulation of the inflammasome

As for other cellular processes, inflammation is under the stringent control of a network of positive and negative regulators that quickly stimulate or terminate the response when needed. A number of inflammasome regulators have been identified. The inflammatory caspase-5 and its murine ortholog, caspase-11, act as co-activators of caspase-1 in select inflammasomes. Caspase-5 is required for caspase-1 activation in the NALP1 inflammasome [69] and casp-11−/− macrophages are defective in caspase-1 activation and IL-1Β maturation in response to LPS [126]. In addition, heat-shock protein 90 (HSP90) and the ubiquitin ligase-associated protein SGT1 regulate the inflammasome by maintaining the complex in an activation-competent configuration [127] (Fig. 2). Interestingly, this mode of regulation is evolutionarily conserved, as R proteins in plants are similarly modulated [128]. Several NLRs such as NALP3, NOD2, IPAF and NALP12 (Monarch-1) interact with SGT1 and/or HSP90 [127, 129, 130]. These interactions are required for inflammasome activation and NOD-1-induced NF-κB activation [127, 130]. Multiple factors negatively regulate the inflammasome pathway by either impacting inflammasome assembly or inflammatory caspase activation. Both endogenous proteins and pathogen derived effector proteins have been recently described to blunt inflammasome activation. CARD-only proteins (COPs) and PYD-only proteins (POPs) have emerged as important modulators of the inflammasome in infection and tissue damage [131]. The first group comprises six members characterized by the presence of a CARD highly similar to that of caspase-1. Among these, four encode decoy caspase-1 genes present in the human caspase-1 locus, including Iceberg, INCA, COP1/Pseudo-ICE and caspase-12 [132, 133]. Through CARD–CARD interactions, these proteins interact with caspase-1 and prevent the recruitment of inflammasome adaptors or scaffolding molecules [134136]. Other COPs, such as CARD8 and NOD2-S are also involved in the inhibition of the inflammatory response. CARD-8 interacts with caspase-1, ICEBERG and pseudo-ICE, and negatively regulates caspase-1-dependent signaling [137]. It also dampens NF-κB activation in response to TNFα stimulation. On the other hand, the short isoform of NOD2 termed NOD2-S, which encodes a COP, competes with NOD2 for RIP2 binding, inhibiting NOD2/RIP2-induced signaling pathways and cytokine/chemokine production [138]. POPs constitute another group of inflammasome inhibitors, which are characterized by the presence of a PYD and are believed to interfere with PYD–PYD interactions between ASC and NALP proteins [139]. This group has 3 main regulators including Pyrin, POP (DASC) and viral PYDs (vPYDs). Pyrin, the protein mutated in familial Mediterranean fever (FMF), negatively regulates the inflammasome by blocking the caspase-1-ASC interaction, and mice expressing a truncated form of Pyrin are sensitive to endotoxin [42]. POP1 and POP2 possess PYD domains highly related to those of ASC and NALPs, respectively, and have been also shown to regulate inflammasome assembly and IL-1β secretion, [140142]. Several poxvirus strains modulate the inflammatory response to virus infection through viral POPs [143, 144]. The poxviral gene product M13L-PYD interacts with ASC and inhibits caspase-1 activation and processing of IL-1β and IL-18 induced by diverse stimuli [144]. In addition to POPs and COPs, endogenous inhibitors of the inflammasome also include the anti-apoptotic factors Bcl-2 and Bcl-XL, which bind to NALP1 and suppress caspase-1 activation and IL-1β production [145]. This recent finding highlights the intricate crosstalk and evolutionary conservation between the apoptosis and innate immunity signaling pathways. Another level of crosstalk exists between the NF-κB and inflammasome pathways. It was recently shown that the central NF-κB activating kinase, IKKβ, is a negative regulator of the inflammasome and of caspase-1-dependent IL-1β secretion [146] (Fig. 2). It was suggested that this mechanism, rather than inhibiting inflammation, might be beneficial for host defense. As multiple pathogens have evolved the ability to inhibit NF-κB signaling, the enhanced production of IL-1β in the absence of NF-κB may represent a compensatory mechanism that activates a second host defense pathway.
https://static-content.springer.com/image/art%3A10.1007%2Fs10495-009-0312-3/MediaObjects/10495_2009_312_Fig2_HTML.gif
Fig. 2

Molecular regulation of the inflammasome. Inflammasome activity must be tightly regulated to prevent the development of inflammatory diseases. Inflammasomes are assembled by NLR proteins and are positively regulated by the inflammatory caspases-5 and -11 as well as SGT-1 and HSP-90. Conversely, inflammasome signaling is inhibited by a variety of molecules. CARD-only proteins (COPs) including ICEBERG, INCA, COP1/Pseudo-ICE, and caspase-12 interact directly with caspase-1, adaptors or scaffolding proteins to inhibit inflammasome assembly. Pyrin-only proteins (POPs) including pyrin, POP1, POP2 and the viral POP, M13L-PYD also inhibit inflammasome assembly through competing with caspase-1 binding to ASC. The serpin like protein PI-9, the NF-κB pathway as well as the anti-apoptotic proteins Bcl-2 and Bcl-XL have been recently shown to inhibit the inflammasome. Pathogen-derived inflammasome/caspase-1 inhibitors include the Yersinia proteins YopE, YopT, the Pseudomonas effector molecules ExoS and ExoU, as well as the cowpox virus protein crmA

NLR-related diseases

Genetic variations in the NLR genes CIITA, NOD2 and NALP3 are implicated in autoinflammatory diseases. Loss of function (LOF) mutations in the gene encoding CIITA cause type II bare lymphocyte syndrome (BLS-II), a severe primary immunodeficiency, clinically similar to severe combined immunodeficiency (SCID), and characterized by a total lack of major histocompatibility complex class II (MHC-II) expression [147, 148]. Familial periodic fever syndromes are inherited autoinflammatory disorders characterized by recurrent episodes of fever and inflammation [149]. Three of these disorders, namely Familial Cold Urticaria (FCU) [150], Muckle–Wells syndrome (MWS) [151] and neonatal-onset mutli-system inflammatory disease (NOMID) [152], are linked to mutations in the Nalp3 (Cias1) gene. These disorders represent a spectrum of clinical symptoms with FCU being the mildest and NOMID the most severe [153]. The disease-causing mutations occur throughout the entire NACHT domain of NALP3 and lead to conformational changes in the protein structure. Functional studies have revealed that the Cias1 mutations result in a gain of function (GOF) phenotype, as the ensuing mutant proteins are constitutively active and able to induce NF-κB activation and IL-1β release [78, 154]. Mutations in another NLR gene, Nod2, are associated with two inflammatory conditions, Crohn’s disease (CD) and Blau syndrome. CD is a lifelong inflammatory disorder of the digestive tract associated with granuloma formation. The mutations in NOD2 that are associated with CD are mostly SNPs or frameshift mutations affecting the LRR domain, leading to a defect in the ability of NOD2 to sense bacteria [155, 156]. Unlike CD, mutations that cause Blau syndrome are GOF mutations located in the NACHT domain of NOD2 that result in MDP-independent constitutive activation of NF-κB [157]. Other chronic inflammatory diseases are caused by the local production of host-derived danger signals. In gout and pseudogout, MSU and CPPD crystals trigger excessive activation of the NALP3 inflammasome resulting in chronic sterile inflammation of the joints and toes [109]. In the clinic, anti-IL-1β therapies including the recombinant IL-1R antagonist (IL-1Ra) Anakinra are especially effective in these diseases and have proven highly beneficial for patients with hereditary periodic fevers and gout [158160].

Copyright information

© Springer Science+Business Media, LLC 2009