The Mitochondrial Protease Clpp Is a Metabolic Vulnerability and an Immunogenic Trigger Against Multiple Myeloma

Introduction: On top of their established essential bioenergetic role, mitochondria are responsible for critical cellular processes that cancers may coopt to gain heightened fitness. Multiple myeloma (MM) cells show abundant and intensely turned-over mitochondria, with heightened mitochondrial biogenesis and OXPHOS gene expression associated with poor prognosis. We thus hypothesized that mitochondria may offer unexploited vulnerabilities in MM and, prompted by its distinctive expression, focused on the mitochondrial protease ClpP, a proteasome-like proteolytic structure responsible for mitochondrial matrix homeostasis.

Methods: We manipulated ClpP expression in MM cell lines, by both stable and inducible shRNA-mediated knockdown (KD), and assessed tumor growth in immunocompromised and immunocompetent hosts. We evaluated the sequelae of ClpP silencing by Seahorse bioenergetic profiling, transcriptomics, proteomics, metabolomics and flow cytometry.

Results: We found higher expression of ClpP in malignant plasma cells (PC) as compared to normal counterparts, with intermediate ClpP abundance in premalignant conditions and further increased expression at relapse. Attesting to its functional relevance, constitutive or inducible ClpP silencing was toxic for MM cells, both in vitro and in vivo. Surprisingly, we found that ClpP established role in OXPHOS maintenance was dispensable for MM. Rather, Seahorse bioenergetic profiling demonstrated that ClpP silencing has no effect on OXPHOS and energy production in MM cells. Conversely, by unlabeled metabolomics and arginine tracing upon ClpP silencing we identified a deranged processing of ornithine leading to a significant depletion of polyamines. In line with the role of polyamines in sustaining autophagy and PC fitness, pharmacologic inhibition of this pathway phenocopied the toxicity induced by ClpP silencing. By expressing a catalytically inactive ClpP mutant for proteomic identification of captured substrates, we were then able to unveil a tonic degradation of the enzyme ornithine aminotransferase (OAT) by ClpP. Indeed, in ClpP silenced MM cells, OAT accumulation diverted ornithine away from polyamine biosynthesis towards glutamate and proline, thus suggesting a causal role of this enzyme in mediating the addiction of MM cells to ClpP.

In parallel, by RNA-sequencing and dedicated experiments we also unveiled a cGAS-dependent activation of a type-I interferon (IFN) response in ClpP silenced cells, likely due to the leakage of mitochondrial DNA in the cytosol. The downstream changes in MM secretome proved able to stimulate dendritic cells in vitro and to boost their ability to promote IFNγ production by CD4⁺ and CD8⁺ T cells. Importantly, when we injected ClpP silenced 5TGM1 MM cells in immunocompetent KaLwRij mice we observed a significantly prolonged survival as compared to mock 5TGM1 recipients. ClpP silenced cells were also able to reshape the bone marrow microenvironment, inducing higher proportions of IFNγ-producing, central memory (CD44⁺CD62L⁺), and effector memory (CD44⁺CD127⁺) CD4⁺ T cells and lower percentages of exhausted CD4⁺ T cells and myeloid-derived suppressor cells. Overall, these findings suggest that ClpP inhibition in MM cells may afford tumor containment through stimulation of a more effective immune response.

Conclusions: We found that ClpP is essential to MM cells as it ensures ornithine supply for polyamine biogenesis, required to maintain PC fitness, and that ClpP ablation triggers a cGAS-STING-dependent immune response. Besides offering a new twofold anti-myeloma target, ClpP may epitomize the notion that mitochondria are an unexplored source of vulnerabilities of MM, which integrate essential non-oncogene metabolic addictions with the possibility to unleash mitochondrial molecules to reignite an otherwise indolent immune microenvironment.