A new mutation-independent approach to cancer therapy: Inhibiting oncogenic RAS and MYC, by targeting mitochondrial biogenesis.

Abstract

Here, we used MCF7 cells as a model system to interrogate how MYC/RAS co-operativity contributes to metabolic flux and stemness in breast cancer cells. We compared the behavior of isogenic MCF7 cell lines transduced with c-Myc or H-Ras (G12V), either individually or in combination. Cancer stem cell (CSC) activity was measured using the mammosphere assay. c-Myc augmented both mammosphere formation and mitochondrial respiration, without any effects on glycolytic flux. In contrast, H-Ras (G12V) synergistically augmented both mammosphere formation and glycolysis, but only in combination with c-Myc, directly demonstrating MYC/RAS co-operativity. As c-Myc is known to exert its effects, in part, by stimulating mitochondrial biogenesis, we next examined the effects of another stimulus known to affect mitochondrial biogenesis, i.e. ROS production. To pharmacologically induce oxidative stress, we used Rotenone (a mitochondrial inhibitor) to target mitochondrial complex I. Treatment with Rotenone showed bi-phasic effects; low-dose Rotenone (1 to 2.5 nM) elevated mammosphere formation, while higher doses (10 to 100 nM) were inhibitory. Importantly, the stimulatory effects of Rotenone on CSC propagation were blocked using a mitochondrial-specific anti-oxidant, namely Mito-tempo. Thus, "mild" mitochondrial oxidative stress, originating at Complex I, was sufficient to pheno-copy the effects of c-Myc, effectively promoting CSC propagation. To validate the idea that mitochondrial biogenesis is required to stimulate CSC propagation, we employed Doxycycline, a well-established inhibitor of mitochondrial protein translation. Treatment with Doxycycline was indeed sufficient to block the stimulatory effects of H-Ras (G12V), c-Myc, and Rotenone on CSC propagation. As such, Doxycycline provides a strong rationale for designing new therapeutics to target mitochondrial biogenesis, suggesting a new "mutation-independent" approach to cancer therapy. In support of this notion, most currently successful anti-cancer agents therapeutically target "cell phenotypes", such as increased cell proliferation, rather than specific genetic mutations. Remarkably, we demonstrated that Doxycycline inhibits the effects of diverse oncogenic stimuli, of both i) genetic (MYC/RAS) and ii) environmental (Rotenone) origins. Finally, we discuss the advantages of our "Proteomics-to-Genomics (PTG)" approach for in silico validation of new biomarkers and novel drug targets. In this context, we developed a new Myc-based Mito-Signature consisting of 3 mitochondrial genes (HSPD1; COX5B; TIMM44) for effectively predicting tumor recurrence (HR=4.69; p=2.4e-08) and distant metastasis (HR=4.94; p=2.8e-07), in ER(+) in breast cancer patients. This gene signature could serve as a new companion diagnostic for the early prediction of treatment failure in patients receiving hormonal therapy.