Mechanisms of acquired resistance to riluzole in melanoma cells in vitro and in vivo

The last two decades in melanoma treatment have seen significant progress in patient outcomes. However, loss of response and onset of therapeutic resistance remain the biggest therapeutic challenges in malignant melanoma. Recently, our group completed an 18-week longitudinal study using our melanoma-prone transgenic mice treated with troriluzole, a prodrug of riluzole and an inhibitor of glutamatergic signaling, and/or an anti-PD-1 agent. In this study, therapeutic responses gradually decreased over time, suggesting the emergence of a resistant tumor cell population. To understand the mechanisms underlying this resistance, we isolated (tro)riluzole-resistant mouse and human melanoma cells using both in vitro and in vivo approaches. Cell viability assays showed the “resistant” cells were less sensitive to riluzole compared to cell cultures established under similar conditions with vehicle. Using mass-spectrometry analysis, we were able to identify several candidate proteins as potential culprits for the onset of resistance in each model. A second complementary method, western immunoblot, was used to verify the alterations in these candidate proteins.Two candidate proteins, SMYD2 and IRF9, were confirmed in the mouse model. Pharmacological inhibition of SMYD2 in combination with riluzole led to a substantial decrease in the viability of cells, especially in the riluzole-resistant cell line. Exogenous overexpression of SMYD2 in the riluzole-sensitive cells conveyed an increased cell viability in the presence of riluzole, and exogenous overexpression of IRF9 in the riluzole-sensitive cells produced inconclusive results on modulation of their cell viability. However, double transfection of SMYD2 and IRF9 concomitantly into the same cells did not lead to an increase in the cell viability. Together, these results indicate a possibility that these proteins are involved in some way in the onset of resistance to riluzole.In the human model, GPX4 and CPT1A were identified as candidate proteins to drive therapeutic resistance to riluzole using a similar approach as in the mouse model. Pharmacological inhibition of CPT1A in combination with riluzole decreased the growth of riluzole-resistant cells more effectively than riluzole monotherapy. And GPX4 inhibition revealed a decrease in ferroptosis in riluzole-resistant cells in the presence of riluzole and showed that ferroptosis suppression might be one of the mechanisms of therapeutic resistance to riluzole. Different utilization of glutamate in riluzole-resistant cells was identified as another alteration that might have been caused by onset of resistance to riluzole.Exogenous overexpression of GPX4 or CPT1A in the riluzole-sensitive cells was unable to change their viability in the presence of riluzole. However, in paired patient tumor samples from completed riluzole single-agent clinical trials, a comparable altered expression in all four candidate proteins identified in our cell models (SMYD2, IRF9, GPX4, CPT1A) was noted in about 50% of the patient samples. Together, our data suggest that these proteins might be involved in the response to riluzole in some patients.