Encyclopedia of Signaling Molecules

2018 Edition
| Editors: Sangdun Choi

Caspase Family

  • Alexandre Desroches
  • Dave Boucher
  • Jean-Bernard Denault
Reference work entry
DOI: https://doi.org/10.1007/978-3-319-67199-4_176


The importance of peptidases in cell signaling is well established. Unlike many pathways controlled by phosphorylation, glycosylation, ubiquitination, or other types of post-translational modifications, steps governed by proteases are essentially irreversible because there is no efficacious mechanism for peptide bond ligation. This chapter presents the peptidase family of caspases, which performs limited proteolysis on a wide range of substrates with molecular consequences ranging from inactivation to gain-of-function to accelerated degradation of their targets. It is important to emphasize that caspases are signaling peptidases and not degrading enzymes akin to lysosomal cathepsins or digestive enzymes. The caspases that are principally implicated in inflammation and apoptosis will be the focus of this chapter; we will not discuss the roles of caspase 14.

In 1842, Karl Christoph Vogt, a German scientist, recognized the presence of cell death during the neuronal development of Alytes obstetricans tadpoles. Later in 1885, Walther Flemming, another German biologist who later became the father of cytogenetic, provided a more precise description of cell death based on the study of rabbit ovarian follicular development. The term “programmed cell death” (PCD) was coined in 1964 by Lockshin and coworkers and originally designated the breakdown of tissues during development, especially in insects. Subsequently, in 1972, Kerr and colleagues described apoptosis for the first time. The first genetic evidence of “programming” in apoptosis came with works by H. Robert Horvitz’s group in 1986. They reported the identification of genes required for the execution of cell death in the roundworm Caenorhabditis elegans. In this nematode, exactly 131 cells die from apoptosis to form the 959-cell adult hermaphrodite. It took the work of Black and coworkers (Black et al. 1989), who identified the mammalian cysteine peptidase caspase 1 (ICE, interleukin (IL)-1beta converting enzyme) as responsible for the maturation of pro-IL-1β, to realize that one of the genes found by Horvitz’s group, CED-3 (C. elegans cell death), encoded a cysteinyl peptidase orthologous to caspase 1. These groundbreaking studies paved the way for the identification of caspases responsible for apoptosis in mammals and other family members with roles in inflammation, keratinization of epithelia, and many more functions.


Caspase stands for cysteine aspartyl-specific protease. They are members of clan CD peptidases, i.e., with a cysteine nucleophile and catalytic residue in the order His-Cys in the protein primary structure. They form the C14 family of cytosolic cysteine peptidases, which is defined by strict specificity for the hydrolysis of aspartyl bonds, although a recent study has identified many substrates cleaved following a glutamate residue (Seaman et al. 2016). Caspase structure is conserved among family members and is composed of two principal domains: the N-terminal domain that varies in size and features and a C-terminal catalytic domain (Figs. 1 and 2). The catalytic domain, or unit, is defined as having one large and one small subunit separated by a flexible interdomain linker of variable lengths (23–46 residues; hereafter referred to as the linker). This segment is sensitive to proteolysis and is always cleaved during or following caspase activation or during recombinant expression in bacteria with the notable exception of caspase 14 that does not self-process during E. coli expression. Cleavage of the catalytic domain gives rise to fragments often referred to as p20 (~20 kDa) and p10 (~10 kDa). The catalytic unit is structured by 6 β-strands and 5 α-helices assembled in an open α/β barrel fold (Fig. 2). The large subunit contains the catalytic dyad, cysteine 285 (the nucleophile; residues numbered according to human caspase 1 structural nomenclature), and histidine 237 that controls the hydrolysis of the peptide bond. The small subunit possesses most residues that form the substrate binding pocket, including a critical arginine 341, a residue involved in aspartate recognition by the S1 substrate binding pocket. Based on the available literature, catalysis follows the same steps as other cysteine and serine peptidases. Fuentes-Prior and Salvesen have described in great depth the topology of the substrate binding site and catalytic cycle of caspases (Fuentes-Prior and Salvesen 2004a). Importantly, active caspases are obligate dimers, and this feature is at the root of the activation mechanism of inflammatory and initiator caspases.
Caspase Family, Fig. 1

Schematic representation of the human caspase family. Human caspases are grouped according to their main role: inflammatory caspases and initiator and executioner apoptotic caspases. Caspase 14 is involved in epithelial differentiation. The five different domains of caspases are color coded: CARD (yellow) and DED (orange) are involved in recruitment and activation of the peptidase; the N-peptide of executioner caspases (red) that is removed during apoptosis; the large subunit (blue) and small subunit (light blue) forming the catalytic domain are separated by a linker of variable length. The catalytic histidine and cysteine dyad residues are indicated with their residue numbers below each large subunit. Caspases are drawn to scale, and the number of amino acids of each protein is shown on the right with the corresponding UniProt database entry used (www.uniprot.org)

Caspase Family, Fig. 2

Caspase domain structures. (a) General structure of caspases. Caspase structure is highly conserved and composed of two domains: the N-terminal domain and the catalytic domain. The catalytic domain (which active site is in red/orange), or unit, comprises one large subunit (blue/gray) and one small subunit (light blue/light gray) separated by a flexible linker of variable length. This segment is sensitive to proteolysis and is always cleaved during activation. The large subunit contains the catalytic His-Cys dyad necessary for peptide bond hydrolysis. Cleavage of the catalytic domain gives rise to fragments often referred to as p20 and p10. The cleaved linker separating the two subunits of the catalytic domain interacts with the other end of the other catalytic unit and stabilizes the active site. The catalytic unit is structured by six β-strands and five α-helices assembled in an open α/β barrel fold. Active caspase 3: RCSB (www.rcsb.org) PDB entry 1CP3. (b) Caspase structure is highly conserved. Superimposition of 7 structures of different active caspases in standard orientation (left) and after a 90° rotation along the vertical axis (right). Only one catalytic unit is presented for simplicity. Caspase-1: PDB 1ICE; Caspase-2: PDB 3R5J; Caspase-3: PDB 1CP3; Caspase-6: PDB 30D5; Caspase-7: PDB 1F1J; Caspase-8: PDB 1QDU; Caspase-9: PDB 1JXQ

Human caspases are classified in three main categories based on their main functions (Fig. 1): apoptotic caspases (caspases 2, 3, 6, 7, 8, 9, and 10), inflammatory caspases (caspases 1, 4, and 5), and caspase 14 involved in epithelial differentiation. However, accumulating evidence has clearly demonstrated roles of some caspases outside their traditional realm (Yi and Yuan 2009). For instance, Lamkanfi and colleagues (Lamkanfi et al. 2008) have shown the activation of the apoptotic caspase 7 during the inflammatory response of macrophages to strong stimuli. Additionally, caspase 8 has been reported to be activated on the inflammasome (see below) and to contribute to its signaling (Sagulenko et al. 2013).

Inflammatory Caspases

In humans, three inflammatory caspases coexist, caspase 1 being the prototypical member of this group (in mice, there are only two inflammatory caspases, caspases 1 and 11). The similar domain organization and the fact that their genes are all located on chromosome 11 suggest a rather recent evolution. These caspases are expressed in most, if not all, tissues, but mRNA levels are higher in inflammatory cells, such as macrophages and monocytes (Martinon and Tschopp 2007).


In healthy cells, inflammatory caspases rest as cytosolic monomeric inactive enzymes (zymogens). It is through recruitment by their N-terminal CARD (caspase-recruitment domain) to multimeric platforms named inflammasomes that they gain activity. Similar to other homotypic interaction domains, DD (death domain) and DED (death effector domain), CARD is a member of the death domain superfamily and shares a similar fold (Park et al. 2007). The function of an inflammasome is to bring enough caspase molecules in close proximity to provoke their dimerization and activation (the so-called induced-proximity model (Salvesen and Dixit 1999)). Subsequently, cleavage of the linker occurs, which further stabilizes the dimer.

Classical inflammasomes are multimeric complexes comprised of several different proteins (Schroder and Tschopp 2010) that enable the activation of inflammatory caspases and subsequent downstream signaling (Fig. 3). At their core, they contain a NLR (NOD (nucleotide-binding oligomerization domain)-like receptor) or an ALR (absent in melanoma 2 (AIM2)-like receptors). The complex also contains caspase 1 and may require the adaptor ASC (apoptosis-associated speck-like protein containing a CARD) if no CARD is present (Bryant and Fitzgerald 2009). Caspase 4 (Sanders et al. 2015) and caspase 5 (Martinon et al. 2002) have also been reported to interact with classical inflammasomes, but the relevance of such interactions remain to be characterized.
Caspase Family, Fig. 3

Canonical inflammasomes. The minimal unit of the canonical inflammasomes are depicted with their various domains. Upon recognition of a DAMP or a PAMP by the LRRs (blue discs) of NLRs, a platform called the inflammasome assembles driven by the NACHT domain (gray) found in the NLRP3, murine NLRP1b, and NLRC4, and this will lead to the recruitment and activation of caspase 1. Whereas the murine NLRP1b has its own CARD (yellow), the human NLRP1 has a PYD (green) and, like NLRP3, requires the adaptor ASC. The central FIIND (function to find domain; orange) is cleaved via autocatalysis. The AIM2 inflammasome recognizes foreign DNA via a HIN200 domain (purple). Adapted from Lamkanfi and Dixit (2014)

In humans, 22 NLRs have been identified. So far, conclusive work has demonstrated that only NLRP3 (NLR protein 3), NLRP1, NLRC4 (NLR family CARD domain-containing protein 4) as well as NLRP2, NLRP6, NLRP7, and NLRP12 can initiate the formation of inflammasome complexes (Lamkanfi and Dixit 2014). With the exception of NLRP10, all NLRs contain a LRR (leucine rich-repeat), a central NACHT (NACHT: NAIP (neuronal apoptosis inhibitor protein), C2TA (MHC class 2 transcription activator), HET-E (incompatibility locus protein from Podospora anserina), and TP1 (telomerase-associated protein)) domain and one to three variable homotypic protein-protein interaction domains (BIR (baculovirus inhibitory repeat), PYD (pyrin domain), or CARD domain). NLRs recognize a wide range of pathogen-associated molecular patterns (PAMPs), such as lipopolysaccharide (LPS), muramyl dipeptide, lethal toxins and crystalline factors, and endogenous damaged-associated molecular patterns (DAMPs). The recognition of such ligands is barely understood, and the NLRC4 inflammasome provide the best example of this kind of recognition so far. Ligands like flagellin are not sensed directly by NLRC4 but rather by NAIP (neuronal apoptosis inhibitor protein), which forms a seed to a hub of 10–12 NLR molecules (Hu et al. 2015). It is suggested that other inflammasomes use a similar mechanism, but the nature of ligands and activators remain to be clarified. Ligand sensing triggers an ATP-dependent oligomerization of NLR and the nucleation of the adaptor protein ASC into a large structure called the ASC-speck (Lu and Wu 2015). Subsequently, caspase 1 monomers are recruited on that speck, increasing their local concentration and enabling their dimerization and activation (Bryant and Fitzgerald 2009). Certain NLRs, like NLRC4 and NLRP1, contain a CARD domain and can directly dimerize inflammatory caspases in the absence of ASC.

ALR receptors, mainly AIM2 and IFI16 (interferon-inducible protein 16), recognize cytoplasmic nucleic acids and dimerize on them. They contain a HIN200 (hematopoietic expression, interferon-inducible proteins with 200 amino acid repeats) DNA-binding domain and a PYD. Once bound to DNA, they trigger ASC-speck formation and caspase 1 recruitment and activation (Lamkanfi and Dixit 2014). Conversely, caspases 4 and 5 have been recently shown to form a novel kind of inflammasome complexes coined unconventional inflammasomes (Shi et al. 2014) in which caspases 4 and 5 directly recognize LPS, but further studies are needed to flesh out this rather unusual molecular platform.

To further add to the complexity surrounding inflammasome formation, several decoy CARD-only proteins (COPs) and PYD-only proteins (POPs) have been described that can compete with caspases at activation platforms (Dorfleutner et al. 2015).

Inflammatory Caspases Functions

Active caspase 1 cleaves cytosolic cytokine precursors to their mature form. Based on small peptidic substrate libraries, caspase 1 prefers (W/Y)XXD↓ motifs with a small residue in P1’ (Thornberry et al. 1997; Stennicke et al. 2000), a motif that is found in caspases 1 (WFKD↓S), 4 (WVRD↓S) and 5 (WVRD↓S), and in pro-IL-1β (YVHD↓S). Pro-IL-1β and pro-IL-18 are the principal caspase 1-cleaved inflammatory mediators. Proteomic data suggest that caspase 1 is responsible for the bulk of cleavage events versus other inflammatory caspases (Agard et al. 2010). Caspase 4 cleaves caspase 3, IL-1β, and IL-1F7B (Luthi and Martin 2007). The transcription factor Max is a substrate of caspase 5 and cleavage occurs at a glutamate residue instead of aspartate (Luthi and Martin 2007), one of the very few validated example of cleavage at a nonaspartate residue by a caspase. In the unconventional inflammasomes, activated caspases 4 and 5 likely process an as yet-unknown substrate (likely gasdermin D; see below) to enable NLRP3-dependant activation of caspase 1 downstream of LPS recognition.

Other substrates of caspase 1 include the apoptotic caspase 7 during pyroptosis (Lamkanfi 2011). Pyroptosis is a pro-inflammatory form of cell death occurring in macrophages, dendritic cells, and monocytes stimulated with specific stimuli, such as some intracellular pathogens, DAMPs, and PAMPs (Jorgensen and Miao 2015). Recently, few groups have shown that inflammatory caspases must cleave gasdermin D to enable the execution of pyroptosis (Broz 2015). The generated N-terminal fragment of gasdermin D leads to the formation of pores at the plasma membrane that are necessary to release cytosolic inflammatory contents during pyroptosis. Conversely, pannexin-1 cleavage by inflammatory caspases has also been reported to be critical for cell death downstream of unconventional inflammasomes (Yang et al. 2015).

Apoptotic Caspases: The Initiators

Apoptotic caspases are expressed in all tissues and cell types. Initiator caspases are integrators of apoptotic stimuli, and their activity is required for the activation of executioner caspases, their main substrates (Fig. 6a). Like inflammatory caspases, initiators possess a protein-protein interaction domain at the N-terminus (Fig. 1) and are activated by dimerization on macromolecular platforms (Mace and Riedl 2010a) (Fig. 6a). For a long time, cleavage of an initiator caspase has been wrongly viewed as synonymous to activation. Cleavage of the linker, which gives rise to the characteristic large and small subunits, always occurs during apoptosis. However, in a reconstituted system, uncleavable caspase 9 sustains caspase activation to the same extent as the fully cleaved peptidase, and in vitro proteolysis is not necessary for initiator activation (Boatright and Salvesen 2003). Consequently, cleavage of an initiator caspase does not equate to its activation. Nevertheless, studies have shown that cleavage is necessary for caspase 8-driven apoptosis in cellulo (Oberst et al. 2010). Therefore, cleavage of initiator caspases seems to play a role not in activation per se but rather in modulating the activity of the active peptidase. Interestingly, executioner caspase 6 can cleave caspase 8 and caspase 3 cleaves caspase 9. The role for the cleavage of caspase 9 by caspase 3 is to relieve it from inhibition set by the endogenous XIAP (X-linked IAP (inhibitor of apoptosis protein)) caspase inhibitor. In contrast, cleavage of caspase 8 produces a more stable dimer (Pop et al. 2007). Because mice do not have caspase 10, little attention has been paid to this caspase in humans. Nevertheless, it is known that, similar to other initiators, this caspase is activated by dimerization (Wachmann et al. 2010). Importantly, unprocessed caspase 10 possesses high activity on Bid (BH3-interacting domain death agonist), but efficacious cleavage of its other targets requires linker proteolysis (Wachmann et al. 2010). Initiator caspases are at the apex of two major intracellular pathways to drive apoptosis: the extrinsic and the intrinsic pathways.

The Extrinsic Pathway

The extrinsic pathway (Fig. 6a, left) drives the activation of caspases 8 and 10. This pathway is initiated by ligation of preassembled trimeric death receptors (DRs) by their cognate ligand (Guicciardi and Gores 2009). These ligands are membrane bound in their more potent form and are normally produced by cytotoxic lymphocytes and natural killer cells, which play a crucial role in immunosurveillance against viral infection and cancer. Based on X-ray crystallography studies, ligation of DRs enables the opening of the cytoplasmic region of the receptor and exposes the DDs. This signals the recruitment of the adaptor FADD (Fas (fibroblast-associated) associated with DD) by homophilic interactions between the newly exposed DDs and FADD’s DD. On the contrary, NMR studies have proposed a model that does not involve structural opening, but instead the sandwiching of Fas and FADD’s DDs into a pentameric organization of Fas/FADD heterodimers. Interestingly, this latter arrangement is similar to the one observed for the proposed caspase 2 activation platform (Hymowitz and Dixit 2010). No matter which model prevails, both highlight the importance of receptor clustering for death-inducing signaling complex (DISC) formation. The other domain found on FADD is a DED that is also present in initiator caspases 8 and 10. This four-component complex, ligand-receptor-adaptor-caspase, constitutes the basis of the apoptotic extrinsic pathway. DISC composition and signaling vary depending on the initiating ligand-receptor pair (reviewed in Guicciardi and Gores 2009a). All assemblies use FADD adaptor protein to link the DR to caspase 8, but the TNF-R1 (tumor necrosis factor receptor type 1)-associated DD (TRADD) is also used as an adaptor by TNFα receptor.

Bound receptors cluster in lipid rafts and are internalized. Subsequently, the cytosolic side of the DISC becomes a microdomain of DEDs and recruits caspases 8 and 10. The locally high concentration of initiator caspase molecules causes their dimerization through induced proximity (Salvesen and Dixit 1999) activation and cleavage of the caspase. A layer of regulation is provided by cFLIP (cellular FLICE (full-length ICE) inhibitory protein) (Thome et al. 1997). The long form of cFLIP (cFLIPL) is a catalytically disabled caspase 8-like protein that competes with initiator caspases at the DISC, thus damping the death signal. Alternatively, low levels of cFLIPL can serve as a caspase 8 dimer partner and support caspase activation. In general, death signaling is strong during or after internalization, but weak when receptors remain at the plasma membrane where it is likely to generate pro-survival signals through the NF-κB (nuclear factor κB) and mitogen-activated protein kinase (MAPK) pathways (Fig. 6b, left; see text below) (Varfolomeev et al. 2005). A key mediator of the pro-survival effect is RIPK1 (receptor-interacting serine/threonine-protein kinase 1) that binds to all DRs and FADD using its own DD. Post-translational modifications, such as phosphorylation, proteolytic cleavage by caspase 8, and ubiquitination, dictate the various effects of this kinase (Cho et al. 2009). Some particularities of each DISC complex are briefly summarized below (reviewed in (Guicciardi and Gores 2009a)).

The Fas/CD95 DISC

Fas/CD95 was discovered as the antigen to an antibody recognizing T lymphocytes and B cells (Oehm et al. 1992) and is considered a tumor suppressor. Compared to TNF-R1 and to a lesser extent Trail-R1/2 (TNF-related apoptosis-inducing ligand receptor 1/2; see APO2L/TRAIL), Fas does not display strong pro-survival effects unless its proapoptotic signaling is masked. For example, tumor cells with disabled Fas-mediated apoptosis will activate the pro-survival NF-κB and MAPK pathways (Shikama et al. 2003; Barnhart et al. 2004). In nonpathological conditions, Fas signaling via extracellular signal-regulated kinase (ERK) is critical for neuronal regeneration and stellate cell proliferation after hepatic injury and is required for activated T-cell and thymocyte proliferation. However, despite these few examples, Fas is primarily a death receptor.

The mechanism by which Fas is activated is well understood. After ligation by FasL (Fas ligand), the adaptor FADD is recruited and the receptor is palmitoylated, constituting the signal for receptor clustering in a lipid raft. Subsequently, Fas is internalized in endosomes in a clathrin-dependent manner where it continues to recruit and promote caspases 8 and 10 activation. In some cells (so-called type II cells, e.g., hepatocytes and pancreatic β-cells), direct Fas-driven caspase activation is insufficient to cause apoptosis. Consequently, the intrinsic pathway (see text below) also contributes following the cleavage of Bid by caspases 8 and 10 (Wachmann et al. 2010), which will engage the mitochondrion-dependent pathway (Fig. 6a). The discriminating difference between type I and II cells is the level of XIAP, which needs to be overcome for apoptosis to proceed (Jost et al. 2009). Although this dichotomy was initially described for Fas, it also applies to the other DRs.


The TNF receptor system is comprised of two receptors, TNF-R1 and TNF-R2, but only TNF-R1 is considered a true DR because TNF-R2 does not have a DD to recruit FADD. Three ligands bind TNF-R1: the membrane bound form of TNFα, the soluble form of TNFα, and TNFβ, also known as the lymphotoxin- α. However, the focus will be on TNFα signaling because it is the most studied. TNFα is produced by many immune cell types, hepatocytes, and fibroblasts and plays a crucial role in inflammation, proliferation, and differentiation. The complex formed after TNFα ligation to TNF-R1 is peculiar in that its primary signaling is pro-survival rather than apoptotic. It involves the recruitment of the adaptor TRADD, RIPK1, TRAF2/5 (TNF receptor-associated factor 2/5), and cIAP1/2 (cellular inhibitors of apoptosis protein 1/2). The role of the latter two proteins is to promote RIPK1 poly-ubiquitination via an E3 ubiquitin ligase activity. RIPK1 poly-ubiquitination serves as a platform for the assembly of two complexes: (1) a NEMO (NF-κB essential modulator)/IKKα (inhibitors of κB kinase) and IKKβ complex, which allows the release of NF-κB to promote differentiation, proliferation, and inflammation (Wajant and Scheurich 2001) and (2) a TAK1 (TGFβ-activated kinase 1), TAB2 (TGFβ-activated kinase 1/MAP3K7 binding protein 2), and TAB3 complex, which is able to stimulate signaling via the MAP kinase pathways (JNK (c-Jun N-terminal kinase), ERK1/2, and p38). Together, both routes counteract each other because despite the fact that at a low level of activity the JNK pro-death signal is counteracted by NF-κB activation, sustained JNK activity leads to the activation of caspases 8 and 10 and apoptosis.

A second important function of the TNFα DISC that has arisen in recent years is the induction of necroptosis (Fig. 6b), a genetically programmed form of necrosis (reviewed in Vanden Berghe et al. 2014; Pasparakis and Vandenabeele 2015; Newton and Manning 2016). The decision to go down the necroptosis route is decided by a balance of RIPK1 poly-ubiquitination by the E3 ligase activity of cIAP1/2 and the de-ubiquitinating activity of CYLD (cylindromatosis). Upon de-ubiquitination, RIPK1 leaves the membrane and provokes the assembly of a cytosolic complex comprising phosphorylated RIPK3 molecules, FADD, and inactive caspase 8. RIPK3 then phosphorylates MLKL (mixed lineage kinase domain-like) to form pores. In order to avoid necroptosis, the RIPK1/RIPK3/MLKL cascade must be shut down by caspase-mediated proteolytic cleavage of RIPK1/3 and other DISC components (see Caspases 8 and 10 Substrates below). If the cell chooses apoptosis, a pro-death DISC will efficiently assemble after internalization of the receptor, following which FADD and both initiator caspases 8 and 10 are recruited and activated.

The Trail-R1/2 DISC

Signaling via Trail-R1/2 seems to follow a mechanism similar to that of Fas to activate initiator caspases 8 and 10. However, internalization is dispensable for efficient apoptosis induction in type I cells, but is necessary for type II cells (Guicciardi and Gores 2009a). Trail can also initiate pro-survival signaling by a secondary complex that is initiated at the receptor but works independently afterward. It contains FADD and the initiator caspases 8 and 10, but this complex does not support caspase activation. The complex also contains RIPK1, TRAF2, the adaptor TRADD, and the NEMO/IKKα/IKKβ complex in a manner reminiscent of TNF-R1 signaling. Furthermore, this complex can lead to MAP kinase pathway activation. The role of this complex is not clear because it antagonizes the cell death pathway, but it may explain why many cancer cells retain Trail receptor expression.

Caspases 8 and 10 Substrates

Initiator caspases 8 and 10 have few established substrates, which is in line with the institution of a proteolytic cascade between the initiators and executioner caspases to amplify the apoptotic signal. Caspases 8 and 10 cleave themselves at the DISC and proteolyze executioner pro-caspases 3 and 7 and Bid (Luthi and Martin 2007), their main substrates. Caspases 8 and 10 also cleave arrestin-2 and generate a fragment which then translocates to the mitochondrion and enhances tBid-dependent cytochrome c (cyt c) release (Kook et al. 2014). Because both caspases are recruited to the DISC, it is not surprising that some of their substrates are part of this complex. Indeed, these initiators cleave proteins that are part of pro-survival signaling assemblies with a net result of promoting death. For example, caspase 8 proteolyzes RIPK1/3, thus shifting TNFα signaling toward cell death. FLIPL is also cleaved by caspase 8. The ability of FLIPL to compete with the initiators at the DISC, to heterodimerize and activate caspases 8 and 10 (Yu et al. 2009), and even alter caspase 8 substrate preference make it a key player in DISC signaling. To illustrate this, work by Oberst and colleagues showed that uncleaved caspase 8-FLIPL heterodimer abrogates RIPK3-dependent necroptosis without inducing apoptosis (Oberst et al. 2011), suggesting a different activity of this dimer compared to the cleaved version. Taken together, both protein composition of the DISC and the specific form of each of its constituents are crucial in determining the substrate repertoire of caspases 8 and 10 and, therefore, cell fate.

Few other caspase 8- and 10-activating platforms have been proposed in recent years. One of them promotes limited activation of caspase 8 and is controlled by the paracaspase MALT1 (mucosa-associated lymphoid tissue; see “ MALT1”) during lymphocyte proliferation (Kawadler et al. 2008). Independently of its peptidase domain, MALT1 can associate with caspase 8 and direct its activation. However, the activity of caspase 8 in this complex seems limited to FLIP.

Several proteins regulate extrinsic initiator caspases, among them kinases such as src (sarcoma) and ERK1/2 (Kurokawa and Kornbluth 2009). The specifics of each phosphorylation event will not be discussed here, but the general effect of phosphorylation is the inhibition of caspase activity or their activation.

The Intrinsic Pathway

Caspase 9 drives an intracellular apoptotic pathway called the intrinsic/mitochondrial pathway (Fig. 6a, right). This route is activated in response to various stresses and developmental cues. Indeed, the critical role of caspase 9 during mouse development is exemplified by the fact that knockout mice die perinatally and show severe brain malformation. The intrinsic pathway integrates signals such as DNA damage, reactive oxygen species (ROS) generation, and metabolic cues and can also be activated by the extrinsic pathway through the cleavage of Bid. The detailed mechanisms by which these stimuli lead to caspase activation are out of the scope of this chapter, but a basic understanding of the major players is necessary. The mitochondrion is central to the intrinsic pathway in that it amalgamates cues on its outer membrane through a delicate balance of pro- and anti-apoptotic proteins of the Bcl-2 (B-cell lymphoma 2) family. Basically, at the resting state, the anti-apoptotic subgroup (A1, Bcl-2, Bcl-W, Bcl-xL, and Mcl-1) keeps the pro-apoptotic members Bak, Bax, and Bok in check by forming heterodimers with them. A third group, named BH3 (Bcl-2 homology 3)-only proteins (Bad, Bid, Bik, Bim, Bmf, BNIP3, HRK, Noxa, and Puma), works to destabilize the equilibrium set by the other two groups by either binding to anti-apoptotic members, thus freeing pro-apoptotic proteins, or by directly binding to pro-apoptotic members to activate them (Tait and Green 2010). Either model leads to mitochondrial outer membrane permeabilization (MOMP) and leakage of key proteins including cyt c and Smac (second mitochondria-derived activator of caspases). The latter protein neutralizes IAPs (Fuentes-Prior and Salvesen 2004) including XIAP, a direct inhibitor of caspases 3, 7, and 9.

Caspase 9 Activation and Regulation

Once released into the cytosol, cyt c binds to Apaf-1 (apoptotic peptidase activating factor 1). In summary, Apaf-1 contains a CARD domain at its N-terminus, a central NOD, and a C-terminal WD-40 repeats. The latter domains inhibit the oligomerization of Apaf-1, which is relieved by cyt c binding, enabling the formation of a heptameric wheel-shaped structure called the apoptosome (Mace and Riedl 2010). This complex also requires dATP/ATP ligation to the NOD, a feature also found in NLRs. Subsequently, the apoptosome recruits caspase 9 through homotypic CARD interaction into a central hub where it dimerizes. Studies have proposed that only a single dimer of caspase 9 is at the apoptosome at any given time and that it is active only when bound to Apaf-1 (Malladi et al. 2009) creating a molecular timer for executioner caspase activation regulated in part by caspase 9 on the apoptosome. Caspase 9 auto-proteolysis is dispensable for the proper execution of apoptosis.

The crystal structure of dimeric caspase 9 has been published. It reveals only one active catalytic unit whereas the other domain has a distorted active site. The biological implication of this half-active dimer remains unclear.

Bratton and Salvesen have surveyed the regulators of caspase 9 activation and activity in detail (Bratton and Salvesen 2010). Therefore, only the most relevant and best-characterized ones will be addressed below. First and foremost, cleaved caspase 9 is inhibited by XIAP, which prevents dimerization via its BIR3 domain. The strength of inhibition relies partly on the N-terminus of the cleaved small subunit of caspase 9 that contains a motif reminiscent of the mature Smac N-terminus. Interestingly, caspase 3, but not caspase 7, can remove this epitope by cleaving downstream of it, thus promoting caspase 9 activity by removing the influence of XIAP.

Another way by which caspase 9-driven apoptosis is modulated is through direct phosphorylation (Kurokawa and Kornbluth 2009). Representative examples include phosphorylation by ERK2 and Akt, both of which prevent caspase 9 activation. Notably, cancer cells rely heavily on the MAPK pathway led by ERK1/2 and growth factor-phosphatidylinositol 3-kinase (PI3K)-Akt pathway to survive, thus showing a clear path to damper the suicidal tendencies of tumor cells. Interestingly, and contrary to most kinases, c-Abl (Abelson murine leukemia viral oncogene homolog 1)-mediated phosphorylation promotes caspase 9 cleavage in response to DNA damage.

Caspase 9 Substrates

There are only three well-established caspase 9 substrates: itself and pro-caspases 3 and 7 (Luthi and Martin 2007). By analogy to DISC proteins that are substrates of caspases 8 and 10, proteins that are in close proximity to the apoptosome may be cleaved by caspase 9. Few other substrates have been proposed for caspase 9, but for many of them experimental evidence is circumstantial at best.

Based on peptidic substrate library screens, caspase 9 cleaves the preferred LEHD↓ motif (Thornberry et al. 1997); there is no data on residue preference at the prime sites of the scissile bond. Remarkably, none of the reported substrate cleavage sites, including those on pro-caspases 3 and 7 and caspase 9 itself, match the preferred site. Similar to most caspases, the rule of a “good-enough site” seems to prevail. Indeed, no more than twofold improvement can be attained by substituting the primary pro-caspase 7 activation site for the preferred motif recognized by caspase 9 or 8. Even the workhorse of apoptosis, caspase 3, does not have most of its substrates matching its preferred motif, i.e., DXXD↓(G/A/S).

Caspase 2

Caspase 2 is the most conserved member across species and the closest homolog to the unique C. elegans ortholog ced-3 (Vakifahmetoglu-Norberg and Zhivotovsky 2010). However, it is also the least understood of the caspase. Opinions over its roles are controversial, with demonstrated pro-survival and apoptotic functions. It is mainly present in the cytoplasm and on the Golgi apparatus, and it is found in the nucleus of some cell types.

Caspase 2 is rapidly activated in response to acute DNA damage by a mechanism involving p53. In this context, p53 promotes the expression of the CARD-containing adaptor RAIDD (RIP1 domain-containing adaptor with DD) and PIDD (p53-induced protein with a DD). Following a two-step auto-proteolysis process, which converts PIDD from a pro-survival to a proapoptotic protein, PIDD relocalizes to the nucleus and forms an activation complex with RAIDD called the PIDDosome. However, Bouchier-Hayes and colleagues have localized the PIDDosome only in the cytosol, contradicting previous findings. Irrespective of the definitive location of this complex, it is responsible for the recruitment and activation of caspase 2 by dimerization in a manner reminiscent of other initiator caspase-activating platforms. X-ray crystallography showed that the PIDDosome has a ring-like shape composed of 5 PIDD and 7 RAIDD molecules (Mace and Riedl 2010). Caspase 2 activation seems to be upstream of mitochondria, but the real requirement for this pathway remains unknown. Interestingly, PIDD-deficient cells display normal caspase 2 cleavage and no abnormalities, which suggests that another pathway can activate this caspase.

Contrary to other caspases, the caspase 2 substrate binding pocket extends beyond the S4 pocket, because it is relatively inactive on tetrapeptidic substrates, but fairly active on pentapeptides. Caspase 2 substrates are limited compared to other caspases. Unlike the other initiators, caspase 2 cannot process executioner caspases, but it processes itself and cleaves Bid. There used to be only a handful of caspase-2 substrates, but 235 substrates were recently identified using mass spectrometry (Julien et al. 2016). Those substrates are involved in different cellular functions such as RNA processing, regulation of cell death, intracellular transport, and chromosome and cytoskeleton organization.

Apoptotic Caspases: The Executioners

Caspases 3, 6, and 7 are classified as executioner caspases based on their position in the activation cascade, i.e., downstream of initiator caspases (Fig. 6a). Caspase 6 is often omitted from this group, but because of its crucial role in cleaving nuclear lamins and other important death substrates, it should be of equal status with the two other executioners. Executioner caspases are responsible for cell demise by cleaving many cellular substrates critical for the apparition of apoptotic features (Table 1). These caspases are expressed in all cell types and exist as obligate dimers at intracellular concentration.
Caspase Family, Table 1

Death substrates cleaved by apoptotic caspases that result in an apoptotic hallmark

Death substrates


Main role



Intermediate filament networking protein

Structural, cytoskeleton component

Intermediate filament dismantling/blebbing


Calcium-independent phospholipase A2

Generation of arachidonic acid and lysophosphatidyl-choline

Find-me signal, healing response, phagocyte recruitment


Human serine/threonine kinase

Phosphorylate many substrates like JNK, p38 MAPK and H2B histone, Akt1 inhibition

Chromatin condensation via histone H2B phosphorylation


Proapoptotic Bcl-2 family member

Transmit extrinsic death signal to the intrinsic pathway

Intrinsic pathway activation


ADP-ribose transferase activity

Signal breaks in double-stranded DNA; gene repression

PARP-1 cleavage, parthanatos inhibition


Inhibitor of caspase-activated DNAse

Inhibits and assists the folding of caspase-activated DNAse

Internucleosomal DNA fragmentation


Serine/threonine kinase, Rho effector kinase

Phosphorylation of myosin light chain

Membrane blebbing, apoptotic bodies formation, nuclear fragmentation


Protein kinase C, isoform delta

Serine/threonine kinase involved in cell signaling

Amplify apoptosis signal by promoting degradation of Mcl-1

p75 (NDUFS1)

p75 subunit of complex I

Participate in the respiratory chain of mitochondria

Loss of mitochondrial potential and ROS generation

Nuclear lamins, lamin receptors

Nuclear structural proteins

Nuclear morphology

Chromatin condensation, nuclear fragmentation


Calcium-dependent regulator of actin filament dynamic

Regulator of actin organization

Cell round-up, matrix detachment


Ion channel

ATP and UTP release

Find-me signal


Lipid membrane transporter

Membrane phospholipid scrambling

Eat-me signal


Lipid membrane transporter

Membrane phospholipid asymmetry

Eat-me signal

aIntracellular phospholipase A2

bMammalian sterile 20-like 1

cBH3 interacting domain death agonist

dPoly(ADP ribose) polymerase 1

eInhibitor of CAD

fRho-associated protein kinase

gProtein kinase C δ

hXK (X-linked/Kell)-related protein 8

Unlike the others, executioner caspases possess a short 23–28 amino acid N-terminal peptide instead of protein-protein interaction domains. Although its role remains elusive, this segment is always cleaved during apoptosis. However, whereas for caspases 3 and 6 the N-peptide seems to silence the enzyme in cellulo, caspase 7’s N-peptide removal allows efficient activation by the serine peptidase granzyme B (GrB), at least in some cells (Fig. 5). Recently, an exosite formed of four lysine residues was identified in the N-terminal domain of caspase-7 (Boucher et al. 2012). This exosite enhances the cleavage rate of the poly(ADP ribose) polymerase 1 (PARP-1) and that of the Hsp90 cochaperone p23. The N-peptide of caspase 7 seems to silence the exosite.

Cleavage of the linker that separates the large and small subunit of the catalytic domain is the driving force of executioner caspase activation. Cleavage occurs at conserved aspartate residues and is performed by initiator caspases, GrB, or caspase-3 that activates caspase-6. Other peptidases can activate executioner caspases, further demonstrating the sole requirement for cleavage of the linker for activation. Proteolysis allows the formation of the substrate binding pocket and the reorientation of the catalytic cysteine. Cleavage is required for executioners possibly due to the relatively short length of the linker. However, this can hardly be reconciled with the linker of caspase 6, which is longer than that of caspase 8, and still requires cleavage for activation.

Caspase 7 is the best-understood executioner caspase at a structural level. Many structures provided information on executioner activation mechanism and substrate catalysis. The first reported structure of caspase 7 revealed a globular arrangement of two catalytic units and the presence of interdigitation between units by the so-called L2-L2’ loops (Fig. 4b). The zymogen and the active form of caspase 7 adopt a similar fold. The main differences between them implicate three flexible loops surrounding the unformed catalytic site and substrate binding pocket (Fig. 4a). The unliganded form of caspase 7 exhibits an unformed substrate binding pocket, suggesting that substrate recognition is key to the formation of the fully active enzyme (induced fit process). However, these data remain conflicting as some groups have reported no differences between unbound and inhibitor-bound caspases 3 and 7. Later, many studies on caspase-6 activation mechanism were published. Globally, conformational changes occurring during the passage from the zymogen to the active state resemble the changes described for caspase 7 (Wang et al. 2010).
Caspase Family, Fig. 4

Conformational changes during caspase activation. (a) Initiator caspase activation: caspase 8. The NMR structure of pro-caspase 8 catalytic domain shows a monomeric (left) catalytic unit and has the general fold of an active catalytic unit (right). However, the dimer interface provided by the small subunit (orange) is disorganized. Interestingly, and contrary to executioner pro-caspase 7, the rather long linker is structured and sits over the unformed catalytic domain where it contacts several residues from loop L1. Recruitment at the DISC brings molecules in close proximity and caspase 8 dimerizes. The activity gained by this process allows cleavage of caspase 8, which further stabilizes the active form. However, it is not clear whether proteolysis occurs in cis (self-cleavage) or trans (cleavage by a neighboring molecule). Caspase 8 zymogen: PDB 2K7Z; active caspase 8: PDB 1QDU. (b) Executioner caspase activation: caspase 7. The zymogen (left and middle) and the active form (right) of caspase 7 adopt a similar fold, a globular arrangement of two catalytic units (one blue, one gray). The main differences between them involve three flexible loops (L2–L4, yellow) surrounding the unformed catalytic site and substrate binding pocket of the zymogen. The transition to the active form is driven by proteolysis at one of the two cleavage sites (magenta) found in the L2 loop. It is noteworthy that in the zymogen dimer, L2 loops adopt a different conformation. The L3 loop, which is implicated in the formation of substrate binding pocket, is reoriented following cleavage of the caspase linker, allowing the formation of the substrate binding pocket in the mature form. Together with loop L2, loop L4 forms a loop bundle that stabilizes the active site in the active form of caspase 7. The L2 loop changes drastically upon cleavage and interacts with the same loop (L2’) of the other catalytic unit stabilizing the active site. The cleavage of L2 loops leads to the reorientation of the catalytic cysteine in a solvent accessible conformation. Caspase 7 zymogen: PDB 1GQF; active caspase 7: PDB 1F1J

As mentioned, the removal of their respective N-terminal peptides seems an important step during in cellulo activation of executioners (Fig. 5). Once activated, caspase 3 cleaves its own N-terminal peptide and then that of caspase 7. Because caspase 3 has higher enzymatic activity than caspase 7 and is present at a higher concentration in cells (∼100 nM), it is considered the major executioner. Caspase 6 is directly activated by caspase 3, not by initiators.
Caspase Family, Fig. 5

Executioner caspases activation. Within the executioners, a proteolytic cascade exists and is initiated by the cleavage of pro-caspase 3 by the initiator caspases or GrB (step 1). Caspase 3 gains full activity after self-removal of its N-peptide (step 2). Subsequently, caspase 3 cleaves pro-caspase 6 in its linker and the N-terminal peptide of pro-caspase 7 (step 3). Caspase 6 removes its own N-peptide (step 4); alternatively, caspase 8 can perform this step. Removal of caspase 7 N-peptide renders this caspase available to initiators and GrB for activation (step 5). GrB-mediated cleavage of pro-caspases 3 and 7 is one of the fastest proteolytic event known. Afterward, active executioners cleave their repertoire of death substrates (step 6) and cause apoptosis. Trans cleavage events are indicated by black arrows whereas gray arrows indicate cis proteolysis. The activity status of each peptidase is indicated by color-coded stars (inactive, red; silenced, yellow; active, green)

Executioner caspase activation and activity are regulated by IAPs. Executioner caspases 3 and 7 (and initiator caspase 9) are directly inhibited by XIAP. This IAP, which is often upregulated in cancer cells, provides a unique way to adjust sensitivity to apoptosis and is responsible for defining type I and type II cells (Jost et al. 2009). Based on work by Scott et al. (2005), the second BIR domain of XIAP uses two binding sites to achieve low nanomolar affinity toward caspase 7. Interestingly, the region N-terminal to BIR2 binds in reverse orientation over the substrate binding pocket, whereas the BIR2 domain binds further away from the catalytic site. Through its C-terminal RING (really interesting new gene) E3 ligase domain, XIAP can ubiquitinate caspases and hasten their proteasomal degradation. Moreover, some studies suggest that executioner caspases and proteasome activity levels are strongly linked (Gray et al. 2010).

Cell Death by Caspases

Many phenotypic hallmarks resulting from caspase activation can be recapitulated, at least in some form, without caspase activity (e.g., DNA fragmentation, mitochondrial depolarization and cleavage of some death substrates). Caspase activation is thought by some to be a means to an end more than it is thought to be a part of apoptosis, and this is still being debated. Some have argued that apoptosis should be defined as “caspase-mediated cell death,” which contradicts formulations such as “caspase-independent apoptosis” as it is often suggested in the literature. We personally adhere to the former definition, because the series of events leading to the swift execution of apoptosis require caspase activity in normal, unaltered cells.

Activation of any of the three executioner caspases is sufficient to induce apoptosis (Gray et al. 2010). For example, overexpression of caspase 7 results in cell rounding and detachment, PARP-1 and protein kinase C δ (PKC δ) cleavage, and DNA fragmentation without the activation of any other caspases and cyt c release from the mitochondria. The “good” cleavage motif recognized by caspases 3 and 7 is relatively simple (i.e., DXXD↓(G/A/S)) and is commonly encoded in mammalian genomes. Indeed, ∼2000 caspase substrates have been identified covering the whole gamut of protein types and cellular functions (Julien et al. 2016; Luthi and Martin 2007). Interestingly, only a handful of them other than caspases themselves have been shown necessary to produce full-blown apoptosis (Table 1). Surprisingly, genetic studies in mice on the requirement for DNA fragmentation by CAD (caspase-activated DNAse), a hallmark of apoptosis, revealed that this endonuclease is dispensable for apoptosis and proper cell removal by professional phagocytes. Conversely, mice carrying a caspase 3- and 7-resistant PARP-1, whose cleavage is also a characteristic hallmark of apoptosis, are resistant to endotoxic shock and to intestinal and renal ischemia reperfusion, but not to apoptosis. PKCδ cleavage seems essential for apoptosis mediated by some genotoxic reagents only. Despite these few examples, it is difficult to clearly establish that a given protein is a genuine death substrate. We can divide caspase substrates into three categories: death substrates whose cleavage (1) is absolutely required; (2) hastens/contributes to the process; or (3) is accessory to apoptosis (bystander substrates). Also, for each stimulus, cell type, or tissue examined, specificities may exist. For example, cleavage of Bid illustrates group 1 proteins in type II cells, but not in type I cells, in which proteolysis only accelerates apoptosis (group 2). With the exception of caspases themselves, required cleavage events are quite rare and most cleavage events are probably of the bystander-type, especially if they occur later during apoptosis. Timmer and Salvesen have addressed this issue and proposed simple steps to discriminate true death substrates from bystanders (Timmer and Salvesen 2007).

The timeline of hallmark’s appearance may be different from one cell to another, may depend on the stimulus that initiates apoptosis, and may be dictated by sensitivity of the assay used. However, it is clear that there are early, intermediate, and late events. The earliest event is likely the exposure of phosphatidyl-serine (PS) on the external leaflet of the plasma membrane, occurring within minutes in some cells. This “eat-me” signal originates from both the caspase-mediated activation of the XKr8 (XK-related protein 8) scramblase, which promotes the exposure phosphatidyl-serine on the extracellular side by scrambling phospholipids on both sides of the plasma membrane, and the caspase-mediated inactivation of the ATP11c (ATPase class VI Type 11c) phospholipid flippase, which normally transfers phosphatidyl-serine from the outer to the cytosolic side of the plasma membrane, thus setting the normal membrane asymmetry (Suzuki et al. 2013, Segawa et al. 2014). At the other end of the timescale, the packaging of cell components into small apoptotic bodies is a late event.

Apoptotic Caspases Beyond Death

Acute attention has been paid to nonapoptotic roles of caspases (Yi and Yuan 2009). Caspase 8 is implicated in the immune network by modeling the proliferation and the differentiation of many cell types (e.g., monocytes, B and T cells). Aside from DISC components, caspase 8 cleave few other proteins. HDAC7, a class II histone deacetylase, is cleaved by caspase 8 in the absence of apoptosis, which changes its subcellular localization from the cytosol to the nucleus where it represses gene transcription (Scott et al. 2008). Reports also ascribe a role to caspase 8 in cell migration and metastasis. Furthermore, caspase 3 activation has been linked to differentiation of some cell types. Skeletal muscles need cleavage of Mst-1 (mammalian sterile 20-like 1) and ICAD (inhibitor of CAD) for myoblast differentiation (Larsen et al. 2010). In mice, caspases 3 and 7 are implicated in the production of factors promoting wound healing and tissue regeneration. This role is concomitant with the execution of apoptosis and seems important for compensatory proliferation during normal tissue apoptosis. Besides, caspase 3 has a role in the maturation of erythrocytes and embryonic stem cell differentiation driven by Nanog proteolysis. Many nonapoptotic roles of caspases are probably hidden by the crude phenotypes of caspase knockout mice, and many other nonapoptotic roles are possibly masked by the compensatory phenotype observed in viable mice lineages.

Caspases and Diseases

Multiple attempts to identify mutations in genes encoding caspases in cancers have been made, but no mutation hotspot has emerged. However, in some cancers, mutation rates are relatively high, with many of them having an effect on caspase protein levels. Intuitively, any mutation diminishing caspase activity or its expression should contribute to tumorigenesis. However, there is currently little evidence suggesting that caspase gene mutations are early events in tumorigenesis, and the impact of mutations to establish their relevance in promoting cancer development has not been studied or published. Nevertheless, at least two publications have suggested that caspases 2 and 8 are tumor suppressors.

Human genomic mutations in extrinsic initiator caspases have been reported and cause divergent yet related diseases. For example, whereas an inactivating mutation in human caspase 8 (Arg248Trp in the large subunit) leads to immunodeficiency, mutation in caspase 10 (Ile406Leu in the small subunit) produces autoimmune lymphoproliferative syndrome (ALPS). In addition to caspase mutations, cancer cells have many other ways to control caspase activation and activity to escape apoptosis. Indeed, caspase phosphorylation and the upregulation of XIAP and anti-apoptotic Bcl-2 family members are well-known mechanisms by which caspase regulation is altered in tumor cells.

On the other hand, many pathogens use proteins that target caspases to counteract their action and alter the normal cellular response (Best 2008; Faherty and Maurelli 2008). For example, the viral serpin CrmA (cytokine response modifier A) encoded by the cowpox virus is a potent inhibitor of caspases 1 and 8 and granzyme B (a serine peptidase injected by cytotoxic lymphocytes and natural killer cells into the cytosol of target cells; Fig. 6a), thus preventing the inflammatory response and apoptosis. Viral FLIP-like protein (vFLIP) encoded by many viruses (e.g., Kaposi’s sarcoma-associated herpesvirus) binds the DED-containing proteins FADD and initiator caspases 8 and 10, thus hindering on the extrinsic pathway. Viral FLIP also contributes to tumor development by promoting NF-κB pathway activation.
Caspase Family, Fig. 6

Apoptotic pathways. (a) Extrinsic, intrinsic, and GrB-mediated apoptosis. There are two fundamental pathways leading to caspase activation. The intrinsic pathway relies on the cell’s internal control mechanisms, which are integrated at the mitochondrion by Bcl-2 family members (blue ovals). If proapoptotic members dominate, the mitochondrion releases, among others, cyt c (red) and Smac (lime green). With dATP (pale green), cyt c causes Apaf-1 protein oligomerization into an heptameric structure, the apoptosome, that regroups CARD (yellow) at its apex. This allows the gathering of at least two initiator caspase 9 molecules by virtue of their CARD and their activation by dimerization. Smac, which is released along with cyt c, relieves the inhibition of caspases by XIAP (the BIR2 domain inhibits caspases 3 and 7, BIR3 inhibits dimerization of caspase 9) and amplifies the signal initiated by cyt c. The second pathway, called extrinsic, originates from outside the cell and requires the ligation of DRs such as Trail-Rs, CD95/Fas, or TNF-R1. Their ligands are produced by natural killer cells, cytotoxic T lymphocytes, and few other cell types. DRs activation provokes the assembly of a DISC that will recruit the adaptor protein FADD and initiator caspase 8 and 10 (and also FLIP) by virtue of DEDs (green) and DDs (magenta) homotypic interactions. The high concentration of initiator caspases recruited at the DISC allows their activation through dimerization, which can be prevented by excess of FLIP. In the absence of significant RIPK1 ubiquitination (see text; see panel b), a cytosolic complex constituted of at least FADD, active caspase 8 and RIPK1 will detach from the initial DISC. This latter complex is the major contributor to executioner caspases 3 and 7 activation. The Bcl-2 family member Bid is proteolyzed by initiator caspases 8 and 10 into truncated Bid, which provides a way for the extrinsic pathway to activate the intrinsic pathway that is necessary in the so-called type II cells. Some immune cells use perforin pores to inject in the cytosol of target cells the serine peptidase GrB to complement the extrinsic pathway by directly activating executioner caspases. Finally, some regulating steps are provided by executioner caspases through their ability to cleave the initiator caspases, although these events do not activate the initiators. Black arrows indicate direct action whereas gray arrows indicate component or translocation. (b) Nonapoptotic signaling of DRs. Contrary to the FasL and Trail-induced DISC assembly, the TNFα DISC has strong inflammatory and survival signaling that rely on RIPK1, the adaptor TRADD, TRAF2/5 (orange), and the cIAP1/2 E3 ligases (white). It is following RIPK1 poly-ubiquitination (yellow chain) that two signaling cascades (i.e., TAK1/TAB2/TAB3 (gray) and NEMO/IKKα/IKKβ (blue)) promote inflammation and survival. Additionally, the TNFα DISC is able to activate necroptosis via the de-ubiquitination of RIPK1 by CYLD (black), the formation of poly-RIPK3 (light green) filaments, and MLKL (dark green) phosphorylation. This kinase complex (necrosome) also contains the adaptor FADD and inactive caspase 8. Avoidance of necroptosis by the induction of apoptosis requires the proteolytic cleavage of RIPK1/3 by caspase 8.

Some neurological illnesses are linked to deregulation of caspase activation. In mice, huntingtin, which is a caspase 6 substrate, needs cleavage for Huntington disease to occur. Caspase 7 is implicated in spinocerebellar ataxia type 7, a neurological poly-glutamine disease. Finally, a mutation in the cleavage site of DJ-1, normally recognized by caspase 6, is responsible for a subset of familial Parkinson’s disease.


Since their discovery more than 25 years ago, caspases have drawn intense scrutiny. The role of caspases as mediator of inflammation, apoptosis, and skin maturation is relatively well understood, but their involvement in nontraditional processes remains unclear. The mechanisms by which other activities of caspases are selected over the traditionally recognized ones remain elusive.

Despite all the work done, many questions linger. An exciting topic in caspase biology is currently to delineate the substrate repertoire of each of them during apoptotic (Gray et al. 2010) and nonapoptotic processes. A corollary to this is to understand how caspases “know” which subset of proteins needs cleavage in a particular setting. Timmer and Salvesen have proposed that caspases use exosites to “select” their substrates (Timmer and Salvesen 2007). Accordingly, the first caspase exosite was identified in caspase-7 (Boucher et al. 2012). Furthermore, Hill et al. (2016) presented evidence that the cleavage of lamin C by caspase 6 also involves an exosite. Irrespective of the death substrate repertoire of each caspase, another enduring question regards the relevance for apoptosis of a given cleavage event: Is there a need to cleave ∼2000 proteins for apoptosis to work? It is unlikely; but which cleavage events are important? And in which situations?

Finally, caspase therapeutical potential is largely untapped. Although it is not expected that caspase inhibitors will be used for chronic treatments in many diseases, their acute use can potentially be useful, for instance, in ischemia reperfusion. Nevertheless, at the time of writing, Emricasan (Conatus Pharmaceuticals), a multicaspase inhibitor, is in phase II clinical trial for the treatment of nonalcoholic steatohepatitis (NASH), fibrosis, and liver diseases. Direct caspase activators are, at least for now, more problematic in their design but may become attractive for cancer treatment.

See Also


  1. Agard NJ, Maltby D, Wells JA. Inflammatory stimuli regulate caspase substrate profiles. Mol Cell Proteomics. 2010;9:880–93.CrossRefPubMedPubMedCentralGoogle Scholar
  2. Barnhart BC, Legembre P, Pietras E, Bubici C, Franzoso G, Peter ME. CD95 ligand induces motility and invasiveness of apoptosis-resistant tumor cells. EMBO J. 2004;23:3175–85.CrossRefPubMedPubMedCentralGoogle Scholar
  3. Best SM. Viral subversion of apoptotic enzymes: escape from death row. Annu Rev Microbiol. 2008;62:171–92.CrossRefPubMedPubMedCentralGoogle Scholar
  4. Black RA, Kronheim SR, Merriam JE, March CJ, Hopp TP. A pre-aspartate-specific protease from human leukocytes that cleaves pro-interleukin-1 beta. J Biol Chem. 1989;264:5323–6.PubMedGoogle Scholar
  5. Boatright KM, Salvesen GS. Mechanisms of caspase activation. Curr Opin Cell Biol. 2003;15:725–31.CrossRefPubMedGoogle Scholar
  6. Boucher D, Blais V, Denault JB. Caspase-7 uses an exosite to promote poly(ADP ribose) polymerase 1 proteolysis. Proc Natl Acad Sci U S A. 2012;109:5669–74.CrossRefPubMedPubMedCentralGoogle Scholar
  7. Bratton SB, Salvesen GS. Regulation of the Apaf-1-caspase-9 apoptosome. J Cell Sci. 2010;123:3209–14.CrossRefPubMedPubMedCentralGoogle Scholar
  8. Broz P. Immunology: caspase target drives pyroptosis. Nature. 2015;526:642–3.CrossRefPubMedGoogle Scholar
  9. Bryant C, Fitzgerald KA. Molecular mechanisms involved in inflammasome activation. Trends Cell Biol. 2009;19:455–64.CrossRefPubMedGoogle Scholar
  10. Cho YS, Challa S, Moquin D, Genga R, Ray TD, Guildford M, et al. Phosphorylation-driven assembly of the RIP1-RIP3 complex regulates programmed necrosis and virus-induced inflammation. Cell. 2009;137:1112–23.CrossRefPubMedPubMedCentralGoogle Scholar
  11. Dorfleutner A, Chu L, Stehlik C. Inhibiting the inflammasome: one domain at a time. Immunol Rev. 2015;265:205–16.CrossRefPubMedPubMedCentralGoogle Scholar
  12. Faherty CS, Maurelli AT. Staying alive: bacterial inhibition of apoptosis during infection. Trends Microbiol. 2008;16:173–80.CrossRefPubMedPubMedCentralGoogle Scholar
  13. Fuentes-Prior P, Salvesen GS. The protein structures that shape caspase activity, specificity, activation and inhibition. Biochem J. 2004a;384:201–32.CrossRefPubMedPubMedCentralGoogle Scholar
  14. Gray DC, Mahrus S, Wells JA. Activation of specific apoptotic caspases with an engineered small-molecule-activated protease. Cell. 2010;142:637–46.CrossRefPubMedPubMedCentralGoogle Scholar
  15. Guicciardi ME, Gores GJ. Life and death by death receptors. FASEB J. 2009a;23:1625–37.CrossRefPubMedPubMedCentralGoogle Scholar
  16. Hill ME, MacPherson DJ, Wu P, Julien O, Wells JA, Hardy JA. Reprogramming caspase-7 specificity by regio-specific mutations and selection provides alternate solutions for substrate recognition. ACS Chem Biol. 2016;11:1603–12.Google Scholar
  17. Hu Z, Zhou Q, Zhang C, Fan S, Cheng W, Zhao Y, et al. Structural and biochemical basis for induced self-propagation of NLRC4. Science. 2015;350:399–404.CrossRefPubMedGoogle Scholar
  18. Hymowitz SG, Dixit VM. Unleashing cell death: the Fas-FADD complex. Nat Struct Mol Biol. 2010;17:1289–90.CrossRefPubMedGoogle Scholar
  19. Jorgensen I, Miao EA. Pyroptotic cell death defends against intracellular pathogens. Immunol Rev. 2015;265:130–42.CrossRefPubMedPubMedCentralGoogle Scholar
  20. Jost PJ, Grabow S, Gray D, McKenzie MD, Nachbur U, Huang DC, et al. XIAP discriminates between type I and type II FAS-induced apoptosis. Nature. 2009;460:1035–9.CrossRefPubMedPubMedCentralGoogle Scholar
  21. Julien O, Zhuang M, Wiita AP, O’Donoghue AJ, Knudsen GM, Craik CS, et al. Quantitative MS-based enzymology of caspases reveals distinct protein substrate specificities, hierarchies, and cellular roles. Proc Natl Acad Sci U S A. 2016;113:E2001–10.CrossRefPubMedPubMedCentralGoogle Scholar
  22. Kawadler H, Gantz MA, Riley JL, Yang X. The paracaspase MALT1 controls caspase-8 activation during lymphocyte proliferation. Mol Cell. 2008;31:415–21.CrossRefPubMedPubMedCentralGoogle Scholar
  23. Kook S, Zhan X, Cleghorn WM, Benovic JL, Gurevich VV, Gurevich EV. Caspase-cleaved arrestin-2 and BID cooperatively facilitate cytochrome C release and cell death. Cell Death Differ. 2014;21:172–84.CrossRefPubMedGoogle Scholar
  24. Kurokawa M, Kornbluth S. Caspases and kinases in a death grip. Cell. 2009;138:838–54.CrossRefPubMedPubMedCentralGoogle Scholar
  25. Lamkanfi M. Emerging inflammasome effector mechanisms. Nat Rev Immunol. 2011;11:213–20.CrossRefPubMedGoogle Scholar
  26. Lamkanfi M, Dixit VM. Mechanisms and functions of inflammasomes. Cell. 2014a;157:1013–22.CrossRefPubMedGoogle Scholar
  27. Lamkanfi M, Kanneganti TD, Van Damme P, Vanden Berghe T, Vanoverberghe I, Vandekerckhove J, et al. Targeted peptidecentric proteomics reveals caspase-7 as a substrate of the caspase-1 inflammasomes. Mol Cell Proteomics. 2008;7:2350–63.CrossRefPubMedPubMedCentralGoogle Scholar
  28. Larsen BD, Rampalli S, Burns LE, Brunette S, Dilworth FJ, Megeney LA. Caspase 3/caspase-activated DNase promote cell differentiation by inducing DNA strand breaks. Proc Natl Acad Sci U S A. 2010;107:4230–5.CrossRefPubMedPubMedCentralGoogle Scholar
  29. Lu A, Wu H. Structural mechanisms of inflammasome assembly. FEBS J. 2015;282:435–44.CrossRefPubMedGoogle Scholar
  30. Luthi AU, Martin SJ. The CASBAH: a searchable database of caspase substrates. Cell Death Differ. 2007;14:641–50.CrossRefPubMedGoogle Scholar
  31. Mace PD, Riedl SJ. Molecular cell death platforms and assemblies. Curr Opin Cell Biol. 2010a;22:828–36.CrossRefPubMedPubMedCentralGoogle Scholar
  32. Malladi S, Challa-Malladi M, Fearnhead HO, Bratton SB. The Apaf-1*procaspase-9 apoptosome complex functions as a proteolytic-based molecular timer. EMBO J. 2009;28:1916–25.CrossRefPubMedPubMedCentralGoogle Scholar
  33. Martinon F, Burns K, Tschopp J. The inflammasome: a molecular platform triggering activation of inflammatory caspases and processing of proIL-beta. Mol Cell. 2002;10:417–26.CrossRefPubMedGoogle Scholar
  34. Martinon F, Tschopp J. Inflammatory caspases and inflammasomes: master switches of inflammation. Cell Death Differ. 2007;14:10–22.CrossRefPubMedGoogle Scholar
  35. Newton K, Manning G. Necroptosis and inflammation. Annu Rev Biochem. 2016;85:743–63.CrossRefPubMedGoogle Scholar
  36. Oberst A, Dillon CP, Weinlich R, McCormick LL, Fitzgerald P, Pop C, et al. Catalytic activity of the caspase-8-FLIP(L) complex inhibits RIPK3-dependent necrosis. Nature. 2011;471:363–7.CrossRefPubMedPubMedCentralGoogle Scholar
  37. Oberst A, Pop C, Tremblay AG, Blais V, Denault JB, Salvesen GS, et al. Inducible dimerization and inducible cleavage reveal a requirement for both processes in caspase-8 activation. J Biol Chem. 2010;285:16632–42.CrossRefPubMedPubMedCentralGoogle Scholar
  38. Oehm A, Behrmann I, Falk W, Pawlita M, Maier G, Klas C, et al. Purification and molecular cloning of the APO-1 cell surface antigen, a member of the tumor necrosis factor/nerve growth factor receptor superfamily. Sequence identity with the Fas antigen. J Biol Chem. 1992;267:10709–15.PubMedGoogle Scholar
  39. Park HH, Lo YC, Lin SC, Wang L, Yang JK, Wu H. The death domain superfamily in intracellular signaling of apoptosis and inflammation. Annual Rev Immunol. 2007;25:561–86.CrossRefGoogle Scholar
  40. Pasparakis M, Vandenabeele P. Necroptosis and its role in inflammation. Nature. 2015;517:311–20.CrossRefPubMedGoogle Scholar
  41. Pop C, Fitzgerald P, Green DR, Salvesen GS. Role of proteolysis in caspase-8 activation and stabilization. Biochemistry. 2007;46:4398–407.CrossRefPubMedGoogle Scholar
  42. Sagulenko V, Thygesen SJ, Sester DP, Idris A, Cridland JA, Vajjhala PR, et al. AIM2 and NLRP3 inflammasomes activate both apoptotic and pyroptotic death pathways via ASC. Cell Death Differ. 2013;20:1149–60.CrossRefPubMedPubMedCentralGoogle Scholar
  43. Salvesen GS, Dixit VM. Caspase activation: the induced-proximity model. Proc Natl Acad Sci U S A. 1999;96:10964–7.CrossRefPubMedPubMedCentralGoogle Scholar
  44. Sanders MG, Parsons MJ, Howard AG, Liu J, Fassio SR, Martinez JA, et al. Single-cell imaging of inflammatory caspase dimerization reveals differential recruitment to inflammasomes. Cell Death Disease. 2015;6:e1813.CrossRefPubMedPubMedCentralGoogle Scholar
  45. Schroder K, Tschopp J. The inflammasomes. Cell. 2010;140:821–32.CrossRefPubMedGoogle Scholar
  46. Scott FL, Denault JB, Riedl SJ, Shin H, Renatus M, Salvesen GS. XIAP inhibits caspase-3 and -7 using two binding sites: evolutionarily conserved mechanism of IAPs. EMBO J. 2005;24:645–55.CrossRefPubMedPubMedCentralGoogle Scholar
  47. Scott FL, Fuchs GJ, Boyd SE, Denault JB, Hawkins CJ, Dequiedt F, et al. Caspase-8 cleaves histone deacetylase 7 and abolishes its transcription repressor function. J Biol Chem. 2008;283:19499–510.CrossRefPubMedPubMedCentralGoogle Scholar
  48. Seaman JE, Julien O, Lee PS, Rettenmaier TJ, Thomsen ND, Wells JA. Cacidases: caspases can cleave after aspartate, glutamate and phosphoserine residues. Cell Death Differ. 2016;23:1717–26.Google Scholar
  49. Segawa K, Kurata S, Yanagihashi Y, Brummelkamp TR, Matsuda F, Nagata S. Caspase-mediated cleavage of phospholipid flippase for apoptotic phosphatidylserine exposure. Science. 2014;344:1164–8.CrossRefPubMedGoogle Scholar
  50. Shi J, Zhao Y, Wang Y, Gao W, Ding J, Li P, et al. Inflammatory caspases are innate immune receptors for intracellular LPS. Nature. 2014;514:187–92.CrossRefPubMedGoogle Scholar
  51. Shikama Y, Yamada M, Miyashita T. Caspase-8 and caspase-10 activate NF-kappaB through RIP, NIK and IKKalpha kinases. Eur J Immunol. 2003;33:1998–2006.CrossRefPubMedGoogle Scholar
  52. Stennicke HR, Renatus M, Meldal M, Salvesen GS. Internally quenched fluorescent peptide substrates disclose the subsite preferences of human caspases 1, 3, 6, 7 and 8. Biochem J. 2000;350:563–8.CrossRefPubMedPubMedCentralGoogle Scholar
  53. Suzuki J, Denning DP, Imanishi E, Horvitz HR, Nagata S. Xk-related protein 8 and CED-8 promote phosphatidylserine exposure in apoptotic cells. Science. 2013;341:403–6.CrossRefPubMedGoogle Scholar
  54. Tait SW, Green DR. Mitochondria and cell death: outer membrane permeabilization and beyond. Nat Rev Mol Biol. 2010a;11:621–32.CrossRefGoogle Scholar
  55. Thome M, Schneider P, Hofmann K, Fickenscher H, Meinl E, Neipel F, et al. Viral FLICE-inhibitory proteins (FLIPs) prevent apoptosis induced by death receptors. Nature. 1997;386:517–21.CrossRefPubMedGoogle Scholar
  56. Thornberry NA, Rano TA, Peterson EP, Rasper DM, Timkey T, Garcia-Calvo M, et al. A combinatorial approach defines specificities of members of the caspase family and granzyme B. Functional relationships established for key mediators of apoptosis. J Biol Chem. 1997;272:17907–11.CrossRefPubMedGoogle Scholar
  57. Timmer JC, Salvesen GS. Caspase substrates. Cell Death Differ. 2007;14:66–72.CrossRefPubMedGoogle Scholar
  58. Vakifahmetoglu-Norberg H, Zhivotovsky B. The unpredictable caspase-2: what can it do? Trends Cell Biol. 2010;20:150–9.CrossRefPubMedGoogle Scholar
  59. Vanden Berghe T, Linkermann A, Jouan-Lanhouet S, Walczak H, Vandenabeele P. Regulated necrosis: the expanding network of non-apoptotic cell death pathways. Nat Rev Mol Cell Biol. 2014a;15:135–47.CrossRefPubMedGoogle Scholar
  60. Varfolomeev E, Maecker H, Sharp D, Lawrence D, Renz M, Vucic D, et al. Molecular determinants of kinase pathway activation by Apo2 ligand/tumor necrosis factor-related apoptosis-inducing ligand. J Biol Chem. 2005;280:40599–608.CrossRefPubMedGoogle Scholar
  61. Wachmann K, Pop C, van Raam BJ, Drag M, Mace PD, Snipas SJ, et al. Activation and specificity of human caspase-10. Biochemistry. 2010;49:8307–15.CrossRefPubMedPubMedCentralGoogle Scholar
  62. Wajant H, Scheurich P. Tumor necrosis factor receptor-associated factor (TRAF) 2 and its role in TNF signaling. Int J Biochem Cell Biol. 2001;33:19–32.CrossRefPubMedGoogle Scholar
  63. Wang XJ, Cao Q, Liu X, Wang KT, Mi W, Zhang Y, et al. Crystal structures of human caspase 6 reveal a new mechanism for intramolecular cleavage self-activation. EMBO Rep. 2010;11:841–7.CrossRefPubMedPubMedCentralGoogle Scholar
  64. Yang D, He Y, Munoz-Planillo R, Liu Q, Nunez G. Caspase-11 requires the pannexin-1 channel and the purinergic P2X7 pore to mediate pyroptosis and endotoxic shock. Immunity. 2015;43:923–32.CrossRefPubMedPubMedCentralGoogle Scholar
  65. Yi CH, Yuan J. The Jekyll and Hyde functions of caspases. Dev Cell. 2009;16:21–34.CrossRefPubMedPubMedCentralGoogle Scholar
  66. Yu JW, Jeffrey PD, Shi Y. Mechanism of procaspase-8 activation by c-FLIPL. Proc Natl Acad Sci U S A. 2009;106:8169–74.CrossRefPubMedPubMedCentralGoogle Scholar

Suggested Reading

  1. Fuentes-Prior P, Salvesen GS. The protein structures that shape caspase activity, specificity, activation and inhibition. Biochem J. 2004b;384:201–32.CrossRefPubMedPubMedCentralGoogle Scholar
  2. Galluzzi L, Aaronson SA, Abrams J, Alnemri ES, Andrews DW, Baehrecke EH, Bazan NG, Blagosklonny MV, Blomgren K, Borner C, Bredesen DE, Brenner C, Castedo M, Cidlowski JA, Ciechanover A, Cohen GM, De Laurenzi V, De Maria R, Deshmukh M, Dynlacht BD, El-Deiry WS, Flavell RA, Fulda S, Garrido C, Golstein P, Gougeon ML, Green DR, Gronemeyer H, Hajnoczky G, Hardwick JM, Hengartner MO, Ichijo H, Jaattela M, Kepp O, Kimchi A, Klionsky DJ, Knight RA, Kornbluth S, Kumar S, Levine B, Lipton SA, Lugli E, Madeo F, Malomi W, Marine JC, Martin SJ, Medema JP, Mehlen P, Melino G, Moll UM, Morselli E, Nagata S, Nicholson DW, Nicotera P, Nunez G, Oren M, Penninger J, Pervaiz S, Peter ME, Piacentini M, Prehn JH, Puthalakath H, Rabinovich GA, Rizzuto R, Rodrigues CM, Rubinsztein DC, Rudel T, Scorrano L, Simon HU, Steller H, Tschopp J, Tsujimoto Y, Vandenabeele P, Vitale I, Vousden KH, Youle RJ, Yuan J, Zhivotovsky B, Kroemer G. Guidelines for the use and interpretation of assays for monitoring cell death in higher eukaryotes. Cell Death Differ. 2009;16:1093–107.CrossRefPubMedPubMedCentralGoogle Scholar
  3. Guicciardi ME, Gores GJ. Life and death by death receptors. FASEB J. 2009b;23:1625–37.CrossRefPubMedPubMedCentralGoogle Scholar
  4. Lamkanfi M, Dixit VM. Mechanisms and functions of inflammasomes. Cell. 2014b;157:1013–22.CrossRefPubMedGoogle Scholar
  5. Mace PD, Riedl SJ. Molecular cell death platforms and assemblies. Curr Opin Cell Biol. 2010b;22:828–36.CrossRefPubMedPubMedCentralGoogle Scholar
  6. Tait SW, Green DR. Mitochondria and cell death: outer membrane permeabilization and beyond. Nat Rev Mol Cell Biol. 2010b;11:621–32.CrossRefPubMedGoogle Scholar
  7. Vanden Berghe T, Linkermann A, et al. Regulated necrosis: the expanding network of non-apoptotic cell death pathways. Nat Rev Mol Cell Biol. 2014b;15:135–47.CrossRefPubMedGoogle Scholar

Copyright information

© Springer International Publishing AG 2018

Authors and Affiliations

  • Alexandre Desroches
    • 1
  • Dave Boucher
    • 2
  • Jean-Bernard Denault
    • 1
  1. 1.Department of Pharmacology and Physiology, Faculty of Medicine and Health SciencesUniversité de SherbrookeSherbrookeCanada
  2. 2.Institute for Molecular BioscienceUniversity of QueenslandSt LuciaAustralia