Ovarian cancer drives mitochondrial dysfunction via WT1 in tumor associated stroma

Ovarian cancer (OvCa) is the deadliest gynecologic cancer with most of its lethality attributed to late diagnosis and early metastasis. Prior work demonstrates that carcinoma associated mesenchymal stem cells (CA-MSC) enhance OvCa metastasis by donating their mitochondria to metabolically vulnerable OvCa cells thus increasing OvCa cell oxidative phosphorylation (OXPHOS). Although we have shown a crucial role for these donated mitochondria in OvCa progression and metastasis, the functionality of CA-MSC mitochondria and how they differ from normal MSC (nMSC) mitochondria is not known. The purpose of this study is to characterize differences in mitochondrial form and function in CA-MSC compared to nMSC with the goal of targeting CA-MSC mitochondria to decrease OvCa progression and metastasis. We discovered that CA-MSC derived mitochondria persist in OvCa cells over multiple passages but fail to incorporate into the host mitochondrial matrix and instead take a donut and punctate shaped morphology which is indicative of mitochondrial stress. Interestingly, mitochondrial ATACseq revealed that CA-MSC, compared to their nMSC counterparts, are enriched in pathogenic mitochondrial mutations. GSEA analysis on RNAseq dataset comparing CA-MSC to nMSC show OXPHOS as one of the top altered pathways. Here we demonstrate that CA-MSC have altered mitochondrial functionality and morphology compared to nMSC. Using 11 patient derived CA-MSC lines and 7 patient derived nMSC lines in a cell mito stress test assay, we found that CA-MSC have increased mitochondrial respiration compared to nMSC. Although CA-MSC mitochondria consume more oxygen, we demonstrate that they are not as efficient as their normal counterparts in coupling this to ATP production. Using confocal and TEM microscopy, we demonstrate that CA-MSC have more networked mitochondria consistent with increased mitochondrial respiration, but we also observe increase in donut and punctate mitochondria, morphologies linked to oxidative stress. Importantly, CA-MSC have increased mitochondrial ROS and decreased mitochondrial membrane potential compared to nMSC, suggesting mitochondrial dysfunction. Using a fluorescent reporter protein, MitoTimer, we demonstrate that CA-MSC accumulate more oxidatively stressed mitochondria and preferentially donate these dysfunctional mitochondria to OvCa cells. Finally, using knockdown and overexpression models, we demonstrate that expression of the Wilm's Tumor 1 (WT1) gene mediates oxidative stress in CA-MSC. Our current study elucidates the functional consequences of the unique mitochondrial phenotype of CA-MSC in OvCa to enable novel targeted strategies to improve outcomes in OvCA.