The nucleolus: A new home for the PIDDosome

Members of the caspase family of proteases play essential roles in apoptosis functioning in both the initiation and execution of cell death. Despite being the second caspase ever identified, the role of caspase-2 in apoptosis has been notoriously difficult to unravel. So much so, that it has frequently been given monikers like the orphan caspase, the Cinderella caspase, or the enigmatic caspase.¹ Much of this confusion has stemmed from the inability to resolve disparate and often contradictory observations regarding caspase-2 activation and function as well as the lack of molecular tools to specifically and accurately measure caspase-2 activation.¹ Fortunately, the field has benefitted from several recent papers that have sought to unravel both the physiologic roles and the mechanisms of this misunderstood caspase.

Caspase-2 has long been considered to be activated by the multi-protein complex called the PIDDosome, comprised of the p53-inducible protein PIDD (p53-inducible protein with a death domain) and the adaptor protein RAIDD (RIP-associated Ich-1/CED homologous protein with death domain).² However, the requirement for PIDD in caspase-2 activation has been controversial. While it has been shown essential for caspase-2-induced apoptosis in the absence of p53 and Chk1 (checkpoint kinase 1) activity, caspase-2 activation has been shown to proceed in the absence of PIDD in many contexts.¹ Our recent report in the Journal of Cell Biology seeks to resolve this controversy.³ We identified a novel site for caspase-2 activation: the nucleolus. Therefore, caspase-2 activation can be performed in the nucleolus or in the cytoplasm. Surprisingly, the nucleolar complex appears to be the traditional PIDDosome requiring both PIDD and RAIDD while caspase-2 activation in the cytoplasm was PIDD independent.

This study arose from a successful collaboration between 2 laboratories, based on 2 independent discoveries. Using bimolecular fluorescence complementation (BiFC) imaging-based approaches, the Bouchier-Hayes laboratory at Baylor College of Medicine was able to visualize the nucleolar caspase-2 complex forming in response to DNA damage in live cells. At the same time, the Sidi laboratory at Mt Sinai used proteomics to identify the nucleolar protein nucleophosmin (NPM1) as a novel PIDD interacting protein. By joining forces, we were able to elucidate a novel mechanism for caspase-2 activation where caspase-2 can be activated by 2 distinct activation complexes: one in the cytoplasm that requires RAIDD and one in the nucleolus that requires NPM1, PIDD and RAIDD (Fig. 1). The latter complex assembled primarily in response to DNA damaging agents including the topoisomerase I inhibitor, camptothecin, and the combination of irradiation and Chk1 inhibition. Interestingly, agents like the cytoskeletal disruptor vincristine exclusively induced caspase-2 activation in the cytoplasm. This suggests that the mode of activation of caspase-2 is stimulus specific.

Figure 1.

Model of differential caspase-2 activation platform assembly in the cytosol and nucleolus. DNA damage leads to the assembly of 2 distinct caspase-2 activation platforms (represented by yellow circles): one in the cytosol and one in the nucleolus. The nucleolar complex comprises NPM1 (nucleophosmin), PIDD (p53-inducible protein with a death domain) and RAIDD (RIP-associated Ich-1/CED homologous protein with death domain) while the cytoplasmic complex only requires RAIDD. Other stimuli (e.g. cytoskeletal damage) only engage the cytosolic platform.

The role of the nucleolus is primarily to regulate rRNA polymerization. However, it is emerging that the nucleolus has numerous non-ribosomal roles that range from tumor suppression regulation to cell cycle and DNA repair. NPM1 in particular has known roles in tumor suppression, apoptosis, and protecting from genomic instability.⁴ In addition, it is the most frequently mutated gene in cytogenically normal AML (acute myelogenous leukemia), a change that causes NPM1 to mislocalize to the cytoplasm. The essential role for NPM1 in nucleolar PIDDosome assembly indicates that it may provide an important decision point in where caspase-2 is activated and this may even decide the functional outcomes of caspase-2 activation. Indeed, a recent publication from Villunger and colleagues shows that PIDD is required to protect from aneuploidy.⁵ Given that NPM1 has a proven role in regulating centriole duplication, it is possible that this function and other non-apoptotic functions attributed to caspase-2 are directed from the nucleolus. Consistent with this idea, we showed that blocking NPM1 increased cellular proliferation in a caspase-2 dependent manner. NPM1 and caspase-2 driven apoptosis from the nucleolus could also play a key role in this process, since Kumar and colleagues recently showed that caspase-2 induces apoptosis to remove aneuploid cells.⁶ Continued efforts to unravel the contributions of the nucleolar complex to apoptosis or regulation of proliferation will be imperative to determining how caspase-2 protects from aneuploidy.

In summary, the nucleolus appears to be the primary site of PIDDosome formation in response to DNA damage and may even be the mechanism to remove damaged or polyploid cells through either cell death or cell cycle arrest. This observation provides a crucial missing piece to the puzzle that has been caspase-2 function and may even underlie the observed tumor suppression role of caspase-2. Given the recent observation that caspase-2 expression is impaired in colon cancers where BcL9L is mutated,⁷ it will be important to determine the physiologic implications of this novel complex for caspase-2s role as a safeguard against cancer.

Disclosure of potential conflicts of interest

No potential conflicts of interest were disclosed.