Mitochondrial haplotype affects tumorigenesis and metastatic efficiency through cell-autonomous and non-cell autonomous mechanisms

Abstract

Metastatic disease is responsible for the vast majority of cancer related deaths, yet remains therapeutically elusive. Recent evidence has shown there are genetic underpinnings to the cascade of events that result in fully formed metastases. We utilized Mitochondrial Nuclear Exchange (MNX) mice, a unique model created by transferring a nucleus isolated from a fertilized oocyte of strain x into an enucleated oocyte of strain y, to directly test mitochondrial haplotype’s role in tumorigenesis and metastasis. Through a series of genetic crosses we show that mitochondrial haplotype alters tumor latency, total metastatic number, and metastatic size, but does not affect growth of primary mammary tumors. By comparison to previous work, we show that these mitochondrial effects in mammary cancer are oncogene dependent. We next paired metastatic mammary and melanoma cell lines with nuclear matched MNX mice. We show that pulmonary metastatic burden of both mammary and melanoma cells is altered by mitochondrial haplotype of the host environment. We found that scavenging of mitochondrial superoxide in highly metastatic backgrounds was able to decrease the total number of metastases to the same level as that in low metastatic backgrounds. We show that both mitochondrial haplotype as well as superoxide scavenging alters the expression of select nuclear encoded genes. Preliminary data shows that mitochondrial load, membrane potential, copy number, and metabolic and glycolytic flux profiles vary slightly between strains, but do not correlate with metastatic data. Taken together, we conclude that mitochondria from within tumor cells as well as mitochondria located in the secondary environment can alter metastasis. We posit that mitochondrial-nuclear crosstalk through reactive oxygen species signaling alter nuclear gene expression allowing permissive and restrictive metastatic environments.