Abstract An effective therapy for glioblastoma (GBM), the most lethal brain tumor, is lacking1-11. Several antipsychotic drugs, including pimozide, have recently been shown to have anti-GBM potential 12,13, but the development of tumor resistance diminishes their efficacy. Identifying the mechanisms for such resistance will help repurpose these brain-penetrant drugs for GBM therapy. Here, we found that pimozide inhibits cholesterol and fatty acids (FAs) release via lysosome-mediated lipid droplets (LDs) and lipoprotein hydrolysis, starving GBM cells of these essential building blocks. However, we identified that GBM cells become resistant to pimozide-based therapy due to an elevation of glutamine metabolism. Mechanistically, reductions in cholesterol and the promotion of signaling via sterol regulatory element-binding protein 1 (SREBP-1), a classical lipogenic transcription factor14, stimulate glutamine transporter ASCT2 expression, increasing glutamine uptake and consumption. Enhanced glutamine uptake leads to the release of ammonia that further stimulates SREBP-1 activation, thereby forming a feedforward loop to concurrently increase both glutamine metabolism and lipid synthesis, resulting in drug resistance. Pharmacological targeting of either ASCT2, or glutaminase (GLS) to limit glutaminolysis, abolishes this loop, and in combination with pimozide result in marked killing of GBM cells. Our study demonstrates that combinatorial targeting of glutamine metabolism and lysosomal function is a promising approach to treat GBM, which may be applicable to other aggressive cancers that are also addictive to glutamine and lipids.
Read full abstract