Project Summary Recently, it was shown by Dr. Ben Boursi, Sheba Medical Center, that some metastatic melanoma patients who are refractory to anti-PD-1 immunotherapy can be converted to responders by fecal microbiota transfer (FMT) from a melanoma patient that had a complete response to immunotherapy. Unfortunately, other donor-recipient combinations were unsuccessful implying that an additional uncontrolled factor may determine the effects of microbiota modulation of immunotherapy. Concurrently, we have been using congenic C57BL/6 mice harboring different naturally occurring mitochondrial DNAs (mtDNAs) (mtDNAB6, mtDNA129, and mtDNANZB) to test melanoma sensitivity and anti-PD-L1 therapy. We discovered that the mtDNANZB mice are highly resistant to melanoma progression and strongly respond to anti-PD-L1 therapy, while mtDNA129 mice are permissive for melanoma growth and refractory to immunotherapy, with mtDNAB6 mice being in between. These mice also differ in their gut microbiota and metabolomic analysis of the mtDNANZB mice revealed impaired fatty acid oxidation of relevance to the elaboration of short chain fatty acids (SCFAs) by the gut microbiota. When we expressed the mitochondrially-targeted antioxidant enzyme catalase (mCAT) in the mitochondria of the mouse hematopoietic cells, we diminished the anti-tumor immune response of the mtDNANZB mice and changed the gut microbiota of both the mtDNAB6 and mtDNANZB mice. These observations led us to the hypothesis that: Both the gut microbiota and the immune system are modulated by the mitochondrial genome, in part through mitochondrial reactive oxygen species (mROS) production in immune cells linking the gut microbiota, tumor progression, and immunotherapy. To test this hypothesis, we propose three specific aims. First, we will evaluate mitochondrial function and mROS production in our three congenic strains and correlate this with their immune cell repertoire and function. Then, we will determine if these congenic strains show the same range of responses to other tumor types. Second, we will determine which subclass of hematopoietic cells are responsible for the anti-tumor and pro-immunotherapy response by using adoptive cell transfer (ACT) to replace mtDNA129 immune cells with mtDNANZB cells. We will then express mCAT in the functional immune cells to determine if this negates the anti- tumor and pro-immunotherapy response and changes their microbiota. Third, we will use FMT to replace the gut microbiota of the mtDNA129 and mtDNANZB mice with that of the three congenic strains to determine if mtDNANZB microbiota enhances the mtDNA129 anti-tumor and pro-immunotherapy phenotype and if mtDNA129 microbiota diminish the mtDNANZB phenotype. To confirm that this is mediated by mROS production, we will express mCAT in the responsible immune cells of the mtDNA129 mice and confirm that this blocks the induction of any anti-tumor and pro-immunotherapy phenotype induced by FMT from mtDNANZB mice. To expeditiously extend these findings to human mtDNA lineages and clinical service, Dr. Boursi has agreed to be a collaborator and Dr. Yardeni has arranged positions at both CMEM and Sheba.

Project Narrative Cancer immunotherapy and fecal microbiota transplantation (FMT) can augment natural immune response against several tumor types, but not all patients respond, thus it is important to understand the reasons behind this clinical observation in order to achieve better outcomes. We have observed, in mouse studies, that naturally occurring differences in the mitochondrial DNA can drastically alter gut microbial composition and immune responses against melanoma, both mediated by the mitochondrial generation of reactive oxygen species (mROS) in immune cells. We now propose to determine if this mitochondrial modulation of tumorigenicity is applicable to other cancers, the immune cell type responsible, the metabolic basis of the mitochondrial- microbiota interaction, and the importance of immune cell mROS in modulating the efficacy of FMT.