Pyroptosis

Injury and physical trauma may inflict accidental cell death, but we have come to realize during the past four decades that cells may also actively engage cell death when needed. These regulated cell death forms are intrinsically connected with human embryonic development, homeostatic maintenance and disease pathology. For instance, the human body is composed of approximately 1014 cells, millions of which are removed daily by apoptosis and replaced with newly differentiated cells in order to secure organ functionality. Apoptotic cells are orderly packed in ‘apoptotic bodies’ for uptake by neighboring cells and professional phagocytes, thereby avoiding deleterious inflammatory responses by circulating leukocytes. Unlike apoptosis, however, more recently identified forms of regulated cell death — such as necroptosis and pyroptosis — are characterized by an early breach of the plasma membrane integrity, which results in extracellular spilling of the intracellular contents. Here, we will describe and discuss this and other features of pyroptosis. Injury and physical trauma may inflict accidental cell death, but we have come to realize during the past four decades that cells may also actively engage cell death when needed. These regulated cell death forms are intrinsically connected with human embryonic development, homeostatic maintenance and disease pathology. For instance, the human body is composed of approximately 1014 cells, millions of which are removed daily by apoptosis and replaced with newly differentiated cells in order to secure organ functionality. Apoptotic cells are orderly packed in ‘apoptotic bodies’ for uptake by neighboring cells and professional phagocytes, thereby avoiding deleterious inflammatory responses by circulating leukocytes. Unlike apoptosis, however, more recently identified forms of regulated cell death — such as necroptosis and pyroptosis — are characterized by an early breach of the plasma membrane integrity, which results in extracellular spilling of the intracellular contents. Here, we will describe and discuss this and other features of pyroptosis. By definition, pyroptosis is a proinflammatory form of regulated cell death that relies on the enzymatic activity of inflammatory proteases that belong to the cysteine-dependent aspartate-specific protease (caspase) family. The term ‘pyroptosis’ was first coined in 2001 and stems from the Greek roots pyro — relating to fire or fever — and ptosis (to-sis) — to denote a falling — thus reflecting the inflammatory nature of this form of cell death. Of note, the early permeabilization of the plasma membrane of pyroptotic cells is shared by other caspase-independent immunogenic forms of cell death, such as necroptosis and accidental necrosis. Although apoptosis is not associated with cell lysis and is generally considered immunologically silent, cells undergoing pyroptosis nevertheless share some features with apoptotic cells. Annexin V staining, DNA fragmentation, chromatin condensation, maturation of caspases 3 and 7, and cleavage of poly(ADP-ribose) polymerase 1 (PARP1) are all considered hallmarks of apoptosis. However, these parameters may not be apoptosis-selective biomarkers, as they have also been observed in macrophages undergoing pyroptosis in response to infection with Salmonella enterica serovar Typhimurium (Salmonella Typhimurium) or treatment with lethal toxin of Bacillus anthracis. For instance, pyroptosis in S. Typhimurium-infected macrophages is accompanied by DNA fragmentation, which can be detected by terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL), as in apoptotic cells. Moreover, flippase activity in living cells actively restricts distribution of the phospholipid phosphatidylserine (PS) to the inner leaflet of the plasma membrane. Apoptotic cells become annexin-V-positive because caspase-mediated inactivation of PS flippases and the concomitant activation of a phospholipid scramblase promote early and active translocation of PS to the outer plasma membrane leaflet. On the other hand, the early membrane rupture of pyroptotic cells exposes the inner leaflet of the plasma membrane to the extracellular surface, therefore allowing annexin V to bind to PS in the inner leaflet. Pyroptosis is also thought to be accompanied by the formation of caspase-1 activity-dependent pores of 1–2 nm in the plasma membrane that lead to transmembrane ion fluxes, cytoplasmic swelling and, finally, osmotic lysis of the cell. These pores allow small molecular weight molecules, like propidium iodide (PI), to readily enter and stain pyroptotic cells. Thus, annexin V staining does not distinguish between apoptotic and pyroptotic cell death, but combined staining with both annexin V and PI may be used to discriminate between these differentially regulated cell death modes. Pyroptosis is mainly documented to occur in professional phagocytes of the myeloid lineage, such as macrophages, dendritic cells and neutrophils, although it has also been observed in CD4+ T cells, keratinocytes, epithelial cells, endothelial cells and neurons. A possible explanation is that these cell types may express higher levels of the inflammatory caspases that drive pyroptosis (i.e. caspases 1 and 11 in mice, and their orthologs caspases 1, 4 and 5 in humans). Research during the past decade has shown that caspase-1 is expressed as an inert cytosolic zymogen that becomes active after its recruitment to inflammasomes — cytosolic scaffolds that are assembled by particular pattern recognition receptors (PRRs). These receptors are proteins that scan the cellular environment for the presence of conserved microbial structures, termed pathogen-associated molecular patterns (PAMPs), or host-derived molecules that are released by damaged host cells and are collectively referred to as danger-associated molecular patterns (DAMPs). PRRs have a diversity of roles in the immune system and include transmembrane Toll-like receptors (TLRs) and C-type lectin receptors (CLRs), as well as cytosolic proteins such as the retinoic acid-inducible gene (RIG)-I-like receptors (RLRs) and NOD-like receptors (NLRs). Whereas most of these PRRs initiate signaling pathways culminating in complex transcriptional reprogramming of the cell, a subset of these PRRs assembles inflammasomes that activate caspase-1. The latter group comprises some members of the NLR family, the cytosolic double-stranded DNA (dsDNA) sensor absent in melanoma 2 (AIM2), and pyrin, a protein that is mutated in patients suffering from familial Mediterranean fever (FMF). The inflammasomes assembled by these scaffold proteins recruit caspase-1 either directly or through the adaptor protein apoptosis-associated speck-like protein containing a CARD (ASC), and will be briefly discussed in the following paragraphs (Figure 1). To date, three members of the NLR family, Nlrp3, Nlrp1b and Nlrc4, have been shown to engage well-defined inflammasomes. These NLRs are characterized by the presence of a central nucleotide-binding and oligomerization (NACHT) domain that is commonly flanked by carboxy-terminal leucine-rich repeats (LRRs) and amino-terminal caspase recruitment (CARD) or pyrin (PYD) domains. The LRR motifs fold back onto the NACHT to lock the NLR in a resting state. Recognition of an activating ligand relieves auto-inhibition and enables ATP-dependent self-oligomerization of the NACHT domain and subsequent induction of downstream signaling cascades via homotypic interactions involving the CARD or PYD domain. The Nlrp3 inflammasome responds to a variety of bacterial, viral and fungal agents that broadly includes Gram-positive bacteria such as Staphylococcus aureus, Gram-negative bacteria like enteropathogenic Escherichia coli and Citrobacter rodentium, RNA and DNA viruses such as influenza virus and adenovirus, and yeasts such as Candida albicans and Saccharomyces cerevisiae. In addition, some extracellular indicators of cell damage, such as ATP, monosodium urate crystals and amyloid-β aggregates, as well as environmental and industrial particles, such as silica and asbestos, activate the Nlrp3 inflammasome. Given the extraordinary number of pathophysiologically and chemically unrelated stimuli, activation of the Nlrp3 inflammasome is believed to be relayed by a common host-derived secondary messenger that is elicited by the above agents. Multiple mechanisms have been proposed to account for Nlrp3 inflammasome activation, including K+ efflux, lysosomal destabilization, pore formation in the plasma membrane, mitochondrial damage, the production of reactive oxygen species, Ca2+ influx, and cell swelling. However, most of the proposed secondary messengers are still being debated because they do not appear to be shared by all Nlrp3-activating triggers or because they are not unique to Nlrp3 stimuli. K+ efflux is probably the best candidate so far, were it not for the fact that it has also been implicated in activation of the Nlrp1b inflammasome by anthrax lethal toxin. Regardless, what has become clear is that, given such a broad range of Nlrp3 stimuli, its activation occurs through a tightly-controlled two-step mechanism. An NF-κB-stimulating priming signal is first required to upregulate Nlrp3 expression to levels that suffice to promote efficient inflammasome assembly in the presence of a second signal that then activates the inflammasome (Figure 2). Unlike Nlrp3, the Nlrc4 inflammasome responds to a more restricted set of bacterial components, namely flagellin and components of bacterial type III secretion systems expressed by some facultative intracellular pathogens such as S. Typhimurium, Shigella flexneri, Pseudomonas aeruginosa, Bacillus thailandensis, and Legionella pneumophila. For the actual detection of these conserved bacterial structures, Nlrc4 relies on other members of the NLR family, the NLR apoptosis inhibitory protein (NAIP) subfamily. The Nlrp1b inflammasome currently has only one clearly defined activator, namely Bacillus anthracis lethal toxin (LeTx), a key virulence factor of the pathogen that causes anthrax. Interestingly, instead of detecting the presence of LeTx, Nlrp1b appears to sense its metalloprotease activity in the cytosol of intoxicated macrophages because catalytically inactive LeTx mutants do not engage the Nlrp1b inflammasome. However, the precise mechanisms by which this is accomplished remain vague. In addition to the inflammasome-assembling NLR family members listed above, two other intracellular receptors assemble inflammasomes: the HIN-200 family member AIM2, and the tripartite motif (TRIM) family member pyrin/TRIM20. The cytosolic DNA sensor AIM2 contributes to host defense against Franciscella tularensis, the causative agent of tularaemia, and DNA viruses like cytomegalovirus. Besides an amino-terminal PYD domain, AIM2 consists of a HIN-200 domain that directly binds cytosolic dsDNA. Finally, the pyrin inflammasome is indirectly activated by Rho GTPase-modifying bacterial toxins, including Clostridium botulinum C3 toxin and the major virulence factor toxin B of Clostridium difficile, a leading cause of hospital-acquired infectious diarrhea in immunocompromised patients. To initiate inflammation and defense, the aforementioned inflammasomes converge and rely on activation of caspase-1 to release mature interleukin (IL)-1β and IL-18 and to induce pyroptosis (Figure 2). Recently, caspase-11 and its human orthologs caspases 4 and 5 have also emerged as caspases that initiate pyroptotic cell death of macrophages infected with vacuolar Gram-negative bacteria such as E. coli and Citrobacter rodentium. Guanylate-binding protein (GBP)-mediated lysis of these pathogen-containing vacuoles allows leakage of lipopolysaccharide (LPS) into the cytoplasm, which directly binds to the CARD domain of caspase-11 to trigger its oligomerization and activation. Interestingly, pyroptosis induction by internalized LPS is directly promoted by caspase-11 (and -4 and -5 in humans) without requiring caspase-1. Nevertheless, caspase-11 engages the Nlrp3 inflammasome in parallel to promote caspase-1-dependent cytokine production since caspase-11 itself cannot directly promote the maturation of IL-1β and IL-18. This pathway is therefore referred to as the ‘non-canonical’ Nlrp3 inflammasome pathway (Figure 2). As discussed above, activation of caspase-1 by canonical inflammasomes or the non-canonical Nlrp3 inflammasome pathway results in the proteolytic maturation and extracellular release of its cytokine substrates IL-1β and IL-18. IL-1β, which signals through IL-1 receptor type 1 (IL-1R1), is a pyrogenic cytokine that induces fever and recruits and activates immune cells to inflamed tissues. Extracellular IL-18 binds and activates the IL-18R to promote either T helper 1 (Th1) or T helper 2 (Th2) immune responses, depending on the cytokine milieu. While Th1 responses play critical roles in eliminating intracellular pathogens, a Th2 response is more effective towards clearance of extracellularly replicating pathogens. Moreover, Th2 immunity is considered detrimental in allergies. The precise mechanisms by which IL-1β and IL-18 reach the extracellular space have been debated for decades, but pyroptosis is increasingly regarded as a mechanism promoting the passive release of these highly inflammatory cytokines. In this regard, single-cell imaging of transgenic macrophages expressing a fluorescence resonance energy transfer (FRET) sensor that detects caspase-1 activity suggested that IL-1β secretion fully coincided with pyroptosis induction in the same cell. Due to its lytic nature, pyroptosis further leads to spilling of additional inflammatory factors, including the abundant nuclear protein high mobility group box 1 (HMGB1), several S100 proteins — a vertebrate-specific family of 24 members with diverse intra- and extracellular functions — and IL-1α. Remarkably, each of these DAMPs — also known as alarmins — also appears to have an intracellular homeostatic role. In the nucleus of resting cells, HMGB1 regulates gene transcription by modifying the chromatin structure, thereby facilitating the binding of other proteins. However, when cells undergo pyroptosis, HMGB1 is passively released into the extracellular space, where it functions in conjunction with PAMPs to signal the onset and sustainment of inflammatory and chemotactic responses by engaging receptor for advanced glycation end products (RAGE) and TLRs, respectively. Similar to HMGB1, in many cell types IL-1α is predominantly intranuclear where it may modulate transcription. During pyroptosis, IL-1α is released into the extracellular milieu where it binds to IL-1R1. Although IL-1α and IL-1β act through a shared receptor, IL-1α does not require caspase-1-dependent proteolysis, as it is fully immunologically active in its uncleaved form. By controlling the extracellular release of IL-1β, IL-18 and the DAMPs described above, pyroptosis is thought to make important contributions to inflammatory and anti-microbial responses in the resolution of infections (Figure 2). In addition, pyroptotic cells release micrometer-sized particles that are rich in the inflammasome adaptor ASC and are therefore named ‘ASC specks’. These extracellular ASC specks have been suggested to further amplify the inflammatory response following their phagocytosis by professional phagocytes and neighboring cells. Thus, by promoting the release of ASC specks, IL-1β, IL-18 and DAMPs, pyroptosis is increasingly regarded as a major effector mechanism by which inflammasomes contribute to inflammatory and host defense responses. Although the molecular mechanisms of pyroptosis are largely obscure, it was recently shown that cleavage of gasdermin D (Gsdmd) by the inflammatory caspases 1 and 11 in murine macrophages, and by their human orthologs caspases 1, 4 and 5, is critical for pyroptosis to occur (Figure 2). Gsdmd-deficient macrophages failed to induce pyroptosis after exposure to cytosolic LPS and other known inflammasome stimuli, consequently identifying Gsdmd as a shared component of caspase-1 and -11-mediated pyroptosis. Strikingly, although caspase-1 was able to cleave IL-1β in Gsdmd-deficient macrophages, IL-1β could not be detected in the culture medium of these cells, supporting the notion that pyroptosis mediates extracellular release of IL-1β. The question arises whether Gsdmd has direct pore-forming properties causing membrane permeabilization or whether it signals to downstream effector mechanisms. Therefore, further investigation is required to elucidate the precise mechanism of how Gsdmd engages pyroptosis. Strikingly, while hundreds of caspase-dependent cleavage events coordinate apoptotic cell death, pyroptosis appears to rely on caspase-mediated cleavage of Gsdmd and perhaps only a handful of additional substrates. Both pyroptosis and apoptosis are programmed cell death modes that require caspase activity; however, these distinct modes of cell death are orchestrated by different subsets of caspases. Caspase-1-deficient mice have been shown to be defective in the processing of IL-1β and IL-18 in an extensive set of in vivo disease models. Until quite recently, caspase-1 was believed to mediate sensitivity to LPS-induced lethal shock. However, the contribution of caspase-1 to LPS-induced lethality was revisited when it was discovered that previously generated caspase-1-deficient mice also lack a functional caspase-11 allele, rendering these mice effective caspase-1/11 double knockout mice. By re-introducing a functional caspase-11 allele from a bacterial artificial chromosome (BAC) into these mice, it was shown that caspase-11 was the major culprit that promoted LPS-induced shock. In agreement, caspase-11-deficient mice have long been known to be resistant to LPS-induced shock. Moreover, the observation that mice lacking both IL-1β and IL-18 also are largely susceptible to LPS-induced endotoxemia further supports a key role for caspase-11-driven pyroptosis and the associated release of a variety of inflammatory factors as a major event in endotoxic shock. Neutralization of HMGB1 effectively protected against lethality during LPS-induced endotoxemia. In addition, mice lacking Gsdmd also were shown to be protected from a lethal dose of LPS. A growing body of evidence further suggests that pyroptosis may act as an effective antimicrobial host defense response during infections. For example, caspase-1/11 double knockout mice were shown to be more resistant to E. coli-induced septic shock than mice lacking both the caspase-1 substrates IL-1β and IL-18. Additionally, clearance of L. pneumophila and Burkholderia thailandensis, B. pseudomallei and F. tularensis was controlled by both IL-1β/IL-18-dependent as well as -independent activities of caspase-1 and -11. It is thought that caspase-1- and -11-induced pyroptosis may not only release DAMPs to stimulate immune responses, but also eject bacteria away from their intracellular replicative niches and into the extracellular space where they are taken up and killed by neutrophils and other phagocytes. These studies highlight pyroptosis as an effective host defense mechanism against bacterial infections that otherwise may be lethal. However, the absence of well-defined and selective pyroptosis biomarkers seriously hampered a detailed analysis of the in vivo roles of pyroptosis. The recent identification of Gsdmd as a pyroptosis-inducing substrate of caspase-1 and -11 offers a potentially suitable biomarker to monitor pyroptosis in vivo. Pyroptosis is neither observed under homeostatic conditions nor thought to contribute to embryonic development. However, it is increasingly associated with differential pathophysiological outcomes in infectious and chronic inflammatory diseases. Most evidently, it has been implicated in the pathology of several human autoinflammatory diseases. Cryopyrin-associated periodic syndromes (CAPS) are caused by gain-of-function mutations in Nlrp3 that lead to increased inflammasome activation and result in increased pyroptosis and excessive secretion of IL-1β and IL-18. CAPS is composed of three related chronic inflammatory diseases of increasing severity, namely familial cold autoinflammatory syndrome (FCAS), Muckle–Wells syndrome (MWS) and neonatal onset multisystem inflammatory disease (NOMID). CAPS patients suffer from systemic inflammation, including fever, rashes, joint pain and conjunctivitis. IL-1β-blocking therapies are impressively effective in CAPS patients and have confirmed the key role of IL-1β in the pathogenesis of CAPS. Unlike the situation in humans, FCAS mutations in the murine Nlrp3 gene result in death in the perinatal period. Abrogating IL-1R/IL-18R signaling altogether in these mice only modestly improved life expectancy, whereas caspase-1 deletion resulted in a complete phenotypic rescue. This suggests that pyroptosis-related DAMPs may have contributed to pathology in this autoinflammatory disease model. Inflammasome-associated cell death is also suspected of driving pathological features in autoinflammatory patients with gain-of-function mutations in NLRC4. In all studied cases, recurrent fevers began early in life, but other symptoms were more variable and included skin rash, joint pains, gastro-intestinal manifestations and macrophage activation syndrome. Regardless of this variable clinical presentation, patients with NLRC4-activating mutations presented with remarkably high levels of IL-18 in the circulation, by far exceeding the levels seen in CAPS patients. Despite limited efforts so far, IL-1β neutralization appears to improve a subset of symptoms, but it does not rescue the high levels of circulating IL-18 seen in these patients. This suggests that the excessive Nlrc4-induced cell death that is associated with excessive IL-18 secretion and macrophage activation syndrome may be an important cause of pathology in Nlrc4-associated autoinflammation. These findings undoubtedly warrant a further investigation of inflammasome-induced cell death and its roles in CAPS and Nlrc4-associated autoinflammation. Detecting and responding to micro-organisms that invade host cells is crucial for clearing dangerous infections. Pyroptosis is a lytic and inflammatory mode of regulated cell death by which the intracellular pathogen is expelled from its replicative niche by rapid plasma membrane rupture and is exposed to the extracellular space where it becomes a target for immune effector mechanisms. In parallel, pyroptosis releases pro-inflammatory cytokines and danger signals from the infected cell to recruit additional immune cells to the site of infection that may help in eradicating the pathogen. However, uncontrolled pyroptosis can become detrimental in the context of chronic autoinflammatory diseases and sepsis. Significant work is still required to understand the mechanisms of pyroptosis and its precise roles in vivo. Only by gaining detailed insight into this phenomenon can we start to exploit effective novel treatments to remedy sepsis and chronic inflammatory diseases associated with overactive pyroptosis.