BNIP3 AND ITS ROLE IN MODULATING THE INTERFACE BETWEEN MELANOMA CELLS AND IMMUNE CELLS

The hypoxia responsive protein BNIP3 plays an important role in promoting cell death and/or autophagy, ultimately resulting in a cancer type-dependent, tumour-enhancer or tumour-suppressor activity. We previously reported that in melanoma cells, BNIP3 regulates cellular morphology, mitochondrial clearance, cellular viability and maintains protein expression of CD47, a pro-cancerous, immunosuppressive 'don't eat me' signal. Surface-exposed CD47 is often up-regulated by cancer cells to avoid clearance by phagocytes and to suppress ICD (Immunogenic Cell Death) elicited by anti-cancer therapies. However, whether melanoma-associated BNIP3 modulates CD47-associated immunological effects or ICD has not been properly explored. To this end, here we evaluated the impact of the genetic ablation of BNIP3 (i.e. BNIP3KD) in melanoma cells, on macrophage-based phagocytosis, polarization and chemotaxis both in vitro and in vivo, by means of intra peritoneum (IP) injection in mice. We also assessed the ability of BNIP3 to affect the in vivo growth of B16-F10 melanoma cells transplanted in syngeneic immune-competent mice, as well as the polarization status of the recruited TAMs (Tumour Associated Macrophages). Additionally, we tested its effects on crucial determinants of chemotherapy-induced ICD (i.e. danger signals), as well as in vivo anti-cancer vaccination effect. Interestingly, loss of BNIP3 reduced the expression of CD47 in both normoxic and hypoxic conditions while macrophage-phagocytosis and -chemotaxis were accentuated only when BNIP3KD melanoma cells were exposed to hypoxia. Furthermore, BNIP3-deficient tumours displayed a significantly delayed tumour growth compared to the BNIP3-proficient counterparts.  Although loss of BNIP3 in B16-F10 melanoma impaired the recruitment of TAMs it did not affect their polarization. Finally, when exposed to the ICD inducer mitoxantrone, the loss of melanoma cell-associated BNIP3 did not alter apoptosis-induction, but significantly prevented ATP secretion and reduced phagocytic clearance of dying cells. In line with this, prophylactic vaccination experiments showed that the loss of BNIP3 tends to increase the intrinsic resistance of B16-F10 melanoma cells to ICD-associated anti-cancer vaccination effect in vivo. Thus, normoxic vs. hypoxic and alive vs. dying cell contexts influence the ultimate immunomodulatory roles of melanoma cell-associated BNIP3.