Shortcut to death


  • Potential conflict of interest: Nothing to report.

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;460:1035–1039. (Reprinted with permission.)


FAS (also called APO-1 and CD95) and its physiological ligand, FASL, regulate apoptosis of unwanted or dangerous cells, functioning as a guardian against autoimmunity and cancer development. Distinct cell types differ in the mechanisms by which the ‘death receptor’ FAS triggers their apoptosis. In type I cells, such as lymphocytes, activation of ‘effector caspases’ by FAS-induced activation of caspase-8 suffices for cell killing, whereas in type II cells, including hepatocytes and pancreatic β-cells, caspase cascade amplification through caspase-8-mediated activation of the pro-apoptotic BCL-2 family member BID (BH3 interacting domain death agonist) is essential. Here we show that loss of XIAP (X-chromosome linked inhibitor of apoptosis protein) function by gene targeting or treatment with a second mitochondria-derived activator of caspases (SMAC, also called DIABLO; direct IAP-binding protein with low pI) mimetic drug in mice rendered hepatocytes and β-cells independent of BID for FAS-induced apoptosis. These results show that XIAP is the critical discriminator between type I and type II apoptosis signalling and suggest that IAP inhibitors should be used with caution in cancer patients with underlying liver conditions.


Two Types of Programmed Cell Death.

Tissue homeostasis in a metazoan organism is based on constantly ongoing decisions between life and death at a cellular level. This tightly orchestrated principle also governs the response to exogenous insults or unphysiological proliferation in specified tissues, for example, the liver. Programmed cell death, apoptosis, is an energy-dependent process which leads to elimination of a target cell. Activation of death receptors of the tumor necrosis factor (TNF) receptor superfamily is the most extensively studied mechanism, and was shown to be a key player in acute and chronic liver injury. Additionally, it is well accepted that resistance to the physiological elimination of transformed cells paves the way of hepatocarcinogenesis.1 The signaling pathways of the most prominent death receptors in hepatocytes, the CD95 receptor (Apo-1/Fas) and the TNF receptor 1 (CD120a, p55), comprise numerous adapters and enzymatically active signaling proteins that act as checkpoints of the deadly machinery. Proapoptotic key factors such as the caspases are controlled among others by members of the bcl-2 family and the group of “inhibitors of apoptosis” (IAPs), which were initially described in 1995 in association with TNF-R2.2 IAPs contain characteristic domains (BIR [baculoviral inhibitor of apoptosis protein repeat] domains and RING domains) which are highly conserved in both viral and mammalian isoforms. Among these, the X-chromosome linked inhibitor of apoptosis protein (XIAP) was described as a caspase-3 and caspase-9 substrate, which inhibits caspase activity through substrate competition and is degraded as the cell death process commences.3, 4 However, XIAP does not solely act through its BIR domains, but also exerts an E3 ubiquitin ligase activity involving the RING domain and thereby modulates other pathways such as transforming growth factor-β induced nuclear factor-κB activation.5, 6

Cell-specific control starts at the very top of the cascade: Upon ligand-induced aggregation of the CD95 receptor, an intracellular death-inducing signaling complex (DISC) forms, which among others comprises the adapter molecule mediator of receptor-induced toxicity 1/Fas-associated protein with death domain (Mort1/FADD) and caspase-8.7 Based on the amount of caspase-8 activation at the DISC, cells can be divided into two groups. This distinction of type I and type II cells was initially proposed in 1998 by Scaffidi and colleagues8: the authors observed that cells can be distinguished on the basis of their dependency of mitochondrial factors to fully execute the apoptotic process. In hepatocytes, which are considered type II cells, DISC formation and caspase-8 activation in response to stimulation of the CD95 receptor occurred to a lesser extent compared to type I cells, and overexpression of Bcl-2 or Bcl-xL prevented activation of effector caspases. However, this resistance could be overcome by overexpression of the more downstream effector caspase-3.8 Subsequently, Bid was identified as the crucial molecule linking CD95 activation to the release of proapoptotic cytochrome c from mitochondria, and deletion of Bid was shown to protect hepatocytes from CD95-induced apoptosis.9 Although some differences based on the studied models existed, it was generally accepted that type I cells exhibited a “strong” activation of the DISC and caspase-8, which by itself was sufficient to induce apoptosis, whereas type II cells required amplification of the apoptosis signal through the “mitochondrial loop”.10, 11 Recent findings supported this notion; however, the requirement of the mitochondrial amplification was shown to be overcome through changes in the cellular microenvironment, e.g., extracellular matrix, or stimulation of CD95 with multimerized ligand, as published with CD95 hexamers in HEPATOLOGY.12, 13

Role of XIAP.

The study recently published by Jost and colleagues expands our insights into the regulation of apoptosis in hepatocytes and the critical role of XIAP in this setting (Fig. 1A).14 In accordance with published data, the authors found that apoptosis in hepatocytes required Bid to augment the apoptotic signal and to activate the effector caspase-9, caspase-3, and caspase-7, whereas thymocytes which are considered type I cells undergo apoptosis independent of Bid (Fig. 1B). Additionally, the authors observed an increase in XIAP protein in type II cells following activation of CD95, whereas type I cells showed decreased levels (Fig. 1A). To further clarify the role of XIAP in vivo, the impact of (1) the deletion of XIAP and (2) combined deletion of Bid and XIAP were studied. Mice lacking XIAP succumbed from CD95 activation due to hepatocyte apoptosis even more rapidly than did wild-type mice (Fig. 1C). Both wild-type and XIAP−/− mice exhibited comparable activation of caspases and Bid cleavage. In contrast, loss of XIAP did not affect rates of apoptosis in type I cells. Double deficiency for XIAP and Bid restored the sensitivity toward CD95 activation in hepatocytes that was observed in the absence of Bid alone (Fig. 1D). This implies that, although cleavage of Bid is required for apoptosis in wild-type hepatocytes, a type I cell–like “short cut” to activate effector caspases without involvement of mitochondria exists and commences when loss of XIAP occurs. Previous observations implied that activation of caspase-3 and caspase-7, but not caspase-9, is essential for CD95-induced apoptosis in hepatocytes.15, 16 In agreement with this, the authors of the current article found that Bid/XIAP double-deficient hepatocytes exhibited only low levels of caspase 9 activity while caspase-3 and caspase-7 activation occurred to the full extent. Consequently, XIAP exerts a pivotal checkpoint role, preventing hepatocellular apoptosis. In wild-type and double knockout mice (Bid−/− XIAP−/−) apoptosis could be prevented through a pan-caspase inhibitor. Interestingly, this was not sufficient to rescue XIAP−/− mice that succumbed despite application of a pan-caspase inhibitor, thus pointing to a so-far not clearly defined caspase-independent mode of cell death.

Figure 1.

(A) Schematic of CD95-induced apoptosis in type I (thymocyte) and type II (hepatocyte) cells. Interaction of the CD95 ligand and receptor at the cell surface causes a strong apoptotic signal in type I cells that is independent of the proapoptotic Bcl-2 family member Bid (left side). Activation of caspase-8 occurs canonically, and XIAP is down-regulated as apoptosis commences. Type II cells require amplification of the apoptotic signaling pathway involving Bid and the release of proapoptotic factors form the mitochondria (Bax, Bad). Jost and coauthors found that XIAP is up-regulated (↑) in type II cells in response to CD95 activation, and contributes to the observed phenotype that caspase activation independent of the “mitochondrial loop” is not strong enough to induce apoptosis (right side). This effect in type II cells was dependent on caspase activity and Bid wild-type protein. (B-E) Deletion of Bid and/or XIAP in hepatocytes. (B) Deletion (⊗) of Bid in hepatocytes abrogates the hepatocellular apoptosis and activation of caspase-9, caspase-7, and caspase-3 involving XIAP, despite caspase-8 activation. (C) Deletion of XIAP propagates apoptosis which then commences more rapidly than in wild-type mice. (D) Deletion of Bid and XIAP promotes CD95-dependent apoptosis with increased activation of effector caspase-3 and caspase-7. (E) In the absence of Bid, SMAC mimetics are sufficient to inhibit (⊣) XIAP and allow for CD95-mediated caspase-3 and caspase-7 activation.

Very little is known about the role of XIAP in necrosis or autophagy. In transformed cell types that are resistant to drug-induced apoptosis due to high levels of XIAP, autophagy can commence in response to oxidative stress or disruption of the mitochondrial membrane, indicating that XIAP specifically relates to apoptosis.17

Although deletion of Bid was sufficient to block CD95-induced apoptosis, TNF-mediated liver injury induced by D-galactosamine and lipopolysaccharide occurred despite this deletion. Previous reports have established that in contrast to CD95, TNF-mediated injury in mice lacking Bid occurs at a slower rate and involves activation of mitochondria independent of Bid.18 Also, TNF-neutralizing antibodies, which could prevent potentially occurring autocrine-mediated injury from TNF in response to SMAC mimetics, did not alter liver injury. Thus, although CD95 and TNF signaling pathways share some redundant signaling molecules, TNF-mediated apoptosis activates the mitochondrial loop even in the absence of Bid. This implies that apoptosis which involves Bid occurs more rapidly and potently, whereas Bid-independent mechanisms progress gradually and involve multiple players.

Therapy by Targeted Modulation of Apoptosis.

SMAC/Diablo is a modulatory protein that can inhibit XIAP and other IAPs. The regulation of IAPs involves ubiquitin-dependent degradation through the proteasome.19 This spiked an interest in applying SMAC mimetics to treat apoptosis-resistant cancer, especially hepatocellular carcinoma (HCC).20, 21 Inhibition of XIAP could potentially improve the poor chemosensitivity of HCC, in which high levels of XIAP are associated with advanced disease.22 In light of the recent progress that was achieved for HCC treatment in the SHARP (SorafenibHCCAssessmentRandomizedProtocol)Trial, application of sorafenib to HCC cell lines reduced cellular levels of XIAP protein.23 Likewise, in colorectal cancer cell lines (type II cells) the combined inhibition of XIAP, which affects the effector caspase-9, caspase-3, and caspase-7, and FLICE-inhibitory protein (FLIP), which affects the initiator caspase-8, promoted TNF-mediated apoptosis independent of a mitochondrial pathway.24 The current study tested the SMAC mimetic BV6 and found sensitization toward CD95-induced apoptosis comparable to XIAP deficiency in mice, which was sensitive toward inhibition with a pan-caspase inhibitor (Fig. 1E).

In summary, the article by Jost and colleagues supports and extends the concept of type I and type II cells and contributes an important piece of information to the understanding of the underlying mechanisms. In hepatocytes, apoptosis can only commence if the integrity of the mitochondrial membrane is lost and proapoptotic factors tip the balance that is maintained through the anti-apoptotic factor XIAP. On the other hand, the pathways that contribute to rapid cell death in type I cells are intact in hepatocytes and sufficient to elucidate an apoptotic response in the absence of XIAP even if Bid has been deleted (“shortcut to death”). Whether future therapies for HCC will include a specific antagonist of XIAP function will be dictated by the degree of toxicity through cell death induced in nonmalignant parenchymal tissue.