The death receptors FAS, TRAIL‐Rs and TNFR1 play critical roles in programmed cell death, particularly in the immune system. Upon ligation of death receptors, caspase‐8 is activated within the so‐called ‘Death Induced Signalling Complex’ (DISC) but the mechanisms that mediate and modulate the activation of caspase‐8 are still not fully understood. This is an important issue because caspase‐8 is essential for apoptosis induced by death receptors. In this issue of The EMBO Journal, Kranz and Boutros (2014) describe their findings from a whole genome siRNA screen for the identification of novel regulators of death receptor induced apoptosis signalling. They identified the atypical cadherin FAT1 as a negative regulator of TRAIL‐R‐mediated caspase‐8 activation and consequent induction of apoptosis, although it had no impact on NF‐κB activation. The authors also show that FAT1 depletion substantially increased TRAIL‐induced killing of glioblastoma‐derived cell lines, suggesting a potential novel approach for treatment of this highly aggressive cancer.
The death receptor‐induced apoptotic pathway is a mechanism to eliminate unwanted and/or potentially dangerous (e.g. pathogen infected) cells (Strasser et al, 2009). Ligand‐induced aggregation with consequent activation of pre‐assembled trimers of death receptors (e.g. FAS, TNFR1 or TRAIL‐Rs) leads to the formation of a so‐called ‘death‐inducing signalling complex’ (DISC) (Kischkel et al, 1995) at the inside of the plasma membrane. This DISC is comprised of the death receptor, the adaptor protein FADD (and for some but not all death receptors also TRADD), caspase‐8 and its inhibitor c‐FLIP. This leads to the activation of caspase‐8, an aspartate‐specific cysteine protease, which triggers apoptosis by proteolytic activation of the effector caspases (caspase‐3 and ‐7) and the pro‐apoptotic BH3‐only protein BID, the latter to further amplify the caspase cascade by engaging the BCL‐2‐regulated apoptotic pathway (Jost et al, 2009).
Mechanistically, activation of caspase‐8 was initially believed to be explained by the ‘induced proximity model’ (Salvesen & Dixit, 1999), whereby two caspase‐8 molecules are brought together in the DISC via FADD, which connects the death receptors via homotypic death domain‐death domain (DD) and death effector domain‐death effector domain (DED) interactions with the caspase‐8 molecules (Fig 1). It is, however, now established that the mere proximity of two caspase‐8 molecules with the resulting conformational change (Hughes et al, 2009) is not sufficient for full enzymatic activation. This form of caspase‐8 activation at the plasma membrane, now known to involve a complex with c‐FLIP (Oberst et al, 2011), is sufficient to prevent the embryonic lethality (~E9.5) caused by complete loss of caspase‐8 (Kang et al, 2004), which was revealed to be due to RIPK3‐ and MLKL‐mediated necroptosis (Kaiser et al, 2011; Oberst et al, 2011; Murphy et al, 2013). For induction of apoptosis, caspase‐8 must cleave itself at specific sites to allow formation of a (p202p102) hetero‐tetrameric protease that must be released into the cytosol to gain access to its critical apoptotic substrates, caspase‐3, ‐7 and BID (Kang et al, 2004). Moreover, the recruitment of caspase‐8 into the DISC does not follow a 1:1 stoichiometry with the FADD molecules, but many more caspase‐8 molecules (than FADD molecules) assemble as chains at the underside of the plasma membrane (Schleich et al, 2013; Dickens et al, 2012; Fig 1). Cullin3‐mediated ubiquitination and subsequent p62 recruitment was reported to stabilise these large caspase‐8 containing complexes (Jin et al, 2009).
Functional genomics using RNAi is a powerful approach for the identification of novel or previously unrecognised components of signalling pathways of interest. In this issue of The EMBO Journal, Kranz and Boutros (2014) report on a whole mouse genome siRNA screen to identify regulators that when knocked down enhance TRAIL‐induced apoptosis (in a so‐called ‘synthetic lethal’ screen). Surprisingly they identified the atypical cadherin FAT1 as a negative regulator of caspase‐8, although this protein has never been associated with death receptor signalling before. FAT1 belongs to a family of cadherins that was originally identified in Drosophila and can be subdivided into the FAT and FAT‐like members. Mammalian FAT4 shows the largest extent of homology to Drosophila FAT, whereas mammalian FAT1, FAT2 and FAT3 are more closely related to Drosophila FAT‐like. The Drosophila gene Fat encodes a tumour suppressor that restricts cellular proliferation during development. In human glioblastoma, head‐and‐neck as well as colorectal cancers, somatically acquired mutations or deletion of FAT1 were found to be associated with enhanced tumour growth, and this was attributed to increased WNT signalling. Interestingly, Kranz and Boutros (2014) showed that RNAi‐mediated knock‐down of FAT1 sensitises primary glioblastoma cells to TRAIL‐induced apoptosis. They used the novel and very exciting CRISPR/Cas9 genome editing technology to generate FAT1‐deficient glioblastoma lines to confirm this observation. These results indicate that glioblastomas harbouring inactivating mutations or deletions of FAT1 might respond to low‐dose TRAIL, a treatment that would be expected to have little impact on healthy tissues, in part due to their expression of FAT1.
How does FAT1 inhibit caspase‐8 activity? The authors showed that upon TRAIL‐R stimulation FAT1 interacts with caspase‐8 to prevent the formation of the high‐molecular‐weight complexes (HMWC) that contain active caspase‐8 (Fig 1). Accordingly, FAT1 knock‐down caused increased accumulation of caspase‐8 molecules in HMWC at the plasma membrane, probably representing the recently described activating assembly chains of caspase‐8 (Dickens et al, 2012; Schleich et al, 2012). It is therefore tempting to hypothesise that FAT1 somehow interrupts the assembly of caspase‐8 chains at the DISC, thereby dampening activation of the downstream effector caspases. However, death receptors do not only trigger apoptotic cell death but can under certain conditions also activate NF‐κB signalling, which promotes cell survival, proliferation, migration and inflammatory cytokine production. Interestingly, FAT1 knock‐down did not appear to affect TNFα‐induced NF‐κB signalling (reflected by no impact on cytokine production). Nevertheless, it remains possible that FAT1 modulation of caspase‐8 activity could affect the balance between pro‐apoptotic versus NF‐κB signalling, triggered by death receptors, simply by keeping cells alive for a longer time, thereby facilitating increased overall production of inflammatory cytokines.
In conclusion, the paper by Kranz and Boutros (2014) in this issue of The EMBO Journal presents FAT1 as a novel and unexpected regulator of caspase‐8 activation and hence the death receptor apoptotic pathway. This work nicely exemplifies the power of siRNA library screening for the discovery of novel regulators and the utility of the CRISPR/Cas9 genome editing technology for functional validation. The findings raise several exciting questions: How does FAT1 regulate recruitment and activation of caspase‐8 within the DISC? What is the impact of FAT1 loss in non‐cancerous cells? Are different FAT and FAT‐like family members involved in different apoptotic processes or do they have no role in cell death? Finally, it will be interesting to examine whether FAT1 also regulates caspase‐8 activity in other cell death processes, particularly necroptosis, which only occurs when caspase‐8 is deleted or its enzymatic activity blocked (Murphy et al, 2013).
Work in our laboratories is supported by the National Health and Medical Research Council, Australia (Program Grant 1016701 and Fellowship 1020363) and the Leukemia & Lymphoma Society (Specialized Center of Research 7001‐13) and was made possible through Victorian State Government Operational Infrastructure Support and Australian Government NHMRC IRIISS.
- © 2014 The Authors