Got PIDD1? Natural killer cells clear polyploid cells to ensure a balanced genome

Plain language summary

Removal of polyploid cells is essential to preventing cancer and restricting tumor growth. A new study published in The EMBO Journal shows assembly of the NEMO‐PIDDosome on extra centrioles. Activation of this protein complex leads to NF‐κB activation that, in turn, induces NK cell‐mediated cell clearance.

Recent work shows that extra centrosomes assemble the PIDDosome, whose activation leads to NF‐κB activation and subsequent NK‐mediated cell clearance.

Polyploidy arises when a cell undergoes whole‐genome duplication such that it contains one or more full extra sets of chromosomes. The generation of polyploid cells is a natural scheduled process in many events at the cellular, tissue, and organismal levels. For example, megakaryocytes become polyploid to allow these cells to increase in size and produce platelets, and polyploidy allows mammary tissue growth during lactation. Increases in polyploidy can be a mechanism to promote tissue repair, wound healing, and organ regeneration. However, when polyploidy occurs in an unscheduled fashion, the results are detrimental. Over 30% of solid tumors show evidence of polyploidy (Bielski et al, 2018). This leads to increased genomic instability and progression of cancer in ways that are not fully understood. On the contrary, it is thought that polyploidy buffers the loss of tumor suppressor gene function to control tumor formation and growth (Zhang et al, 2018). Therefore, polyploidy can both promote and suppress tumor formation and growth. Polyploidy arises from cell fusion or impaired mitosis giving rise to extra centrioles.§ Extra centrioles compromise the assembly of the bipolar spindle leading to the activation of a p53‐dependent tetraploidy checkpoint, subsequent arrest, and often apoptosis. Therefore, polyploidy often results in cell removal or senescence. However, if p53 is absent, cells can undergo multipolar mitosis and become aneuploid. As a result, aneuploid cells with multiple centrioles often form the most aggressive tumors (Storchova & Pellman, 2004).

Clearly, cells need to develop means to prevent a cell that undergoes unscheduled polyploidy from becoming aneuploid. Aside from cell‐intrinsic mechanisms like the tetraploidy checkpoint, there is evidence that polyploid cells become immunogenic and are targeted for removal by the immune system. Polyploidy has been shown to induce constitutive ER‐stress resulting in cell surface expression of calreticulin, which acts as an “eat‐me” signal for phagocytes and stimulates cytotoxic T cells (Senovilla et al, 2012). However, how and if polyploidy engages other mechanisms for immune surveillance or additional components of the immune system has, until now, been unclear. In this issue, Andreas Villunger and collaborators (Garcia‐Carpio et al, 2023) show that polyploid cells become immunogenic through activation of a molecular complex known as the PIDDosome, resulting in clearance mediated by natural killer (NK) cells.

The PIDDosome is a large molecular‐weight complex best known for its ability to activate the pro‐apoptotic protein caspase‐2. Villunger and colleagues previously showed that extra centrioles cause the protein PIDD1 to assemble the PIDDosome on centrioles leading to the activation of caspase‐2, cleavage of MDM2, and p53‐dependent cell cycle arrest (Fava et al, 2017). Thus, caspase‐2 appears to be a major trigger of the tetraploidy checkpoint. PIDD1 can form a second complex by recruiting the proteins RIPK1 and NEMO to activate the NF‐κB pathway (Janssens et al, 2005). Although the NEMO‐PIDDosome was originally described as being able to induce NF‐κB in response to genotoxic stress, to date, it has been difficult to ascribe a physiological role for this pathway. Garcia‐Carpio et al (2023) now present a compelling case for the involvement of the NEMO‐PIDDosome in NF‐κB activation in response to unscheduled ploidy increases. The mechanism presented illustrates a novel way how cells can prevent unwanted consequences of polyploidy, through induction of NK cell cytotoxic activity as a key mechanism to induce clearance of potentially cancerous cells.

Garcia‐Carpio et al (2023) demonstrated this by examining the transcriptional profile of cells subjected to cytokinesis failure, revealing innate immune system changes consistent with an NF‐κB‐mediated transcriptional program (as well as responses in cell cycle arrest genes consistent with their prior work). They validated that treatment with ZM447439, which induces polyploidy, and induction of PLK4, which increases centriole number without inducing polyploidy, both induced NF‐κB activity, with minimal detectable DNA damage. Thus, rather than polyploidy or the DNA damage that often results, the presence of extra centrioles is responsible for inducing the NF‐κB response. Strikingly, PIDD1‐deficient cells did not mount an NF‐κB response to supernumerary centrioles, and the authors went on to show this NF‐κB activation required PIDD1 to be recruited to distal centriole appendages via the proteins SCLT1 and ANKRD26. These interactions serve to bring multiple PIDD1 molecules together to form a functional PIDDosome in the same way that enables PIDD1 to recruit and activate caspase‐2 via the adaptor protein RAIDD. What is novel about this study is that PIDD1 assembly on centrioles is also able to form the NEMO‐PIDDosome. Binding studies showed that the NEMO‐PIDDosome formed as a transient complex concomitant with the increase in ploidy and supernumerary centrioles, and loss of either NEMO or RIPK1 dampened the NF‐κB response.

Importantly, the authors next asked what could be the outcome of this NF‐κB activation. They showed that macrophages cultured with media from cells that had undergone centriole amplification were polarized to a pro‐inflammatory M1‐like phenotype. This was because the polyploid cells upregulated several cytokines and chemokines in a PIDD1‐dependent fashion. Using an established NK cell co‐culture model, they showed that NK cells were more likely to target and kill polyploid cells with an intact PIDD1 pathway. This provides strong evidence that polyploidy induces PIDD‐1‐dependent NF‐κB activation to enhance NK cell‐mediated cell clearance.

The model presented by this paper presents a remarkable conservation of function where the same protein, PIDD1, carries out seemingly opposing functions depending on the proteins it recruits, yet leading to the same outcome—removal of a potentially cancerous cell. First, through activation of caspase‐2, the canonical PIDDosome induces the p53‐dependent tetraploid checkpoint leading to cell cycle arrest. Second, through activation of the NEMO‐PIDDosome, activation of NF‐κB leads to NK cell‐mediated cell clearance (Fig 1). The description of the role of the NEMO‐PIDDosome in this context hints at an answer to a troublesome question. If PIDD1 can form complexes with different functions, how are these roles balanced? When expressed, PIDD1 undergoes sequential autocleavage forming PIDD‐C, which recruits RIPK1 and NEMO, and then PIDD‐CC, which recruits RAIDD and caspase‐2. Previous work has suggested that at low doses of DNA damage, PIDD1 preferentially forms PIDD1‐C to form the NEMO‐PIDDosome, while at higher doses, PIDD‐1CC is the dominant cleavage product leading to RAIDD recruitment (Tinel et al, 2007). In another study, phosphorylation of PIDD by ATM appears to be the decision point determining the complex formed, but this appears to occur only when Chk1 is inhibited (Ando et al, 2012). In contrast to these studies, Garcia‐Carpio et al (2023) show that, at least when exogenously expressed, PIDD1 binds RAIDD nearly constitutively, whereas the NEMO‐PIDDosome assembles only after induction of extra centrioles. This may argue that the RAIDD complex forms first, and, as p53 activity increases, the concomitant increase in PIDD1 levels leads to NEMO‐PIDDosome assembly. If this were the case, loss of caspase‐2 would be expected to impair NEMO‐PIDDosome function by reducing p53 activity, but the authors show no difference. Therefore, the model presented implies that the complexes form near‐simultaneously in response to extra centrioles. Consistent with this, the caspase‐2‐dependent effects of MDM2 cleavage leading to p53 stabilization and cell cycle arrest occur in parallel to PIDD‐NEMO activation. Nevertheless, it would be interesting to see whether any synergy exists between these two pathways to promote NK cell cytotoxicity.

Figure 1. PIDD1 removes polyploid cells by intrinsic and extrinsic mechanisms.

PIDDosome assembly on supernumerary centrioles results in the assembly of the canonical PIDDosome resulting in cell cycle arrest, and the NEMO‐PIDDosome, inducing NF‐κB activation and expression of immune response genes leading to the activation of NK cells and clearance of the polyploid cells.

Unfortunately, Garcia‐Carpio et al have not yet recapitulated this phenomenon in an established cancer or precancer model. The same group previously showed that in the Eμ‐Myc lymphoma mouse model, PIDD1—unlike caspase‐2, which acts as a tumor suppressor—appears to be pro‐oncogenic, since its loss was associated with increased lymphoma‐free survival (Manzl et al, 2012).ǂ This result appears counter to the ability of PIDD1 to induce immune surveillance. However, NK cell development is impaired in mice with Myc‐driven lymphomas (Swaminathan et al, 2020), underscoring the need to test the impact of the NEMO‐PIDDosome in solid tumor models known to undergo polyploidy. It would be of particular interest to perform these experiments in the context of p53 loss, an exceedingly common aberration in cancers of many types. Although Garcia‐Carpio et al (2023) show that the NEMO‐PIDDosome‐induced release of cytokines and chemokines is sufficient to polarize macrophages to a pro‐inflammatory phenotype, it will be important to see how this competes with the often‐immunosuppressive effects of the tumor microenvironment. Determining whether there is a place for the PIDDosome in discussions of the tumor microenvironment would launch these pathways into places of far greater clinical relevance. Indeed, with the emergence of NK cell‐based therapies for cancer treatment, would there be a way to co‐opt this mechanism to enhance this therapeutic approach? For this to work, providing a safe way to mimic the effects of extra centrioles without actually inducing them would be key.

In addition to the important role NK cells play in removing tumor cells, they are key mediators of the innate immune response to infection. A number of primary immune deficiency diseases are characterized by impaired NK cell function. In particular, diseases resulting from gain‐ or loss‐of‐function mutations in the STAT1 gene severely impair NK cell cytotoxicity (Vargas‐Hernandez & Forbes, 2019). Patients with these rare diseases are highly susceptible to severe infections. Since many viral infections can induce polyploidy, is it possible that the PIDD1 pathway is compromised in diseases with similar clinical features? Ultimately, Garcia‐Carpio et al (2023) present the PIDDosome field with an exciting new direction of research. The implication of an immune function for the less studied NEMO‐PIDDosome will no doubt serve to drive re‐examinations of this pathway in both cancer and immune deficiency.

The EMBO Journal (2023) 42: e115307