Mitochondria transfer via glioma-astrocyte network microtubes reprograms tumor cells for enhanced tumorigenicity

Abstract Glioblastoma (GBM) interaction with neural cells is critical to its pathobiology. Emerging evidence suggests that GBM cells form an interconnected network with astrocytes, facilitating tumor persistence. Given reports of intercellular transfer of mitochondria in ischemic stroke and other pathologic disease states outside the CNS, we hypothesized that this network facilitates mitochondria transfer from astrocytes to GBM with protumorigenic sequelae. Employing transgenic mice and intracranial viral vector transductions in rats, we found that mitochondria transfer from the TME to GBM occurs in intracranial mouse and patient-derived xenograft models (in nude rats) of GBM. Mitochondria transfer from bone marrow-derived immune cells was minimal in bone marrow chimera mouse models of orthotopic GBM, suggesting that neural cells were the primary mitochondria donors. We confirmed this in vitro, where mouse astrocytes were the major mitochondria donors, followed by microglia and to a much smaller extent bone marrow-derived macrophages. Immortalized human astrocytes transduced with mitochondria-localized mCherry (mito-mCherry) also transferred their mitochondria to numerous patient-derived glioma stem cell (GSC) models at rates of ~5-20%, assessed by flow cytometry and confocal microscopy. Mitochondria were visualized along intercellular actin bridges, structurally resembling tumor microtubes. Blocking actin polymerization or knocking down GAP43 (previously linked to microtube formation) decreased mitochondria transfer from astrocytes to GBM in vitro. Functionally, sorted mito-mCherry+ patient-derived GSCs displayed higher mitochondrial respiration, metabolomic reprogramming and proliferation-promoting phospho-signaling. Mito-mCherry+ GBM cells were more likely to be in the proliferative G2/M phases of the cell cycle, and when sorted from co-cultures had high self-renewal (in vitro) and tumor-initiating capacity (in vivo xenograft mouse model). In ongoing work, we are investigating the role of retrograde GBM to astrocyte transfer of mitochondria by dual-color labeling of the organelle, as well as further delineating the protein machinery involved in this fundamental protumorigenic process, with the goal of identifying novel therapeutic targets.