Abstract
Prostate apoptosis response-4 (Par-4) is an endogenous tumor suppressor that selectively induces apoptosis in a variety of cancers. Although it has been the subject of intensive research in other cancers, less is known about its significance in gliomas, including whether it is regulated by key driver mutations, has therapeutic potential against glioma stem cells (GSCs), and/or is a prognostic marker. We found that patient-derived gliomas with mutant isocitrate dehydrogenase 1 have markedly lower Par-4 expression (P < 0.0001), which was validated by The Cancer Genome Atlas dataset (P = 2.0 E-13). The metabolic product of mutant IDH1, D-2-hydroxyglutarate (2-HG), can suppress Par-4 transcription in vitro via inhibition of promoter activity as well as enhanced mRNA degradation, but interestingly not by direct DNA promoter hypermethylation. The Selective for Apoptosis induction in Cancer cells (SAC) domain within Par-4 is highly active against glioma cells, including orthotopic xenografts of patient-derived primary GSCs (P < 0.0001). Among high-grade gliomas that are IDH1 wild type, those that express more Par-4 have significantly longer median survival (18.4 vs. 8.0 months, P = 0.002), a finding confirmed in two external GBM cohorts. Together, these data suggest that Par-4 is a significant component of the mutant IDH1 phenotype, that the activity of 2-HG is complex and can extend beyond direct DNA hypermethylation, and that Par-4 is a promising therapeutic strategy against GSCs. Furthermore, not every effect of mutant IDH1 necessarily contributes to the overall favorable prognosis seen in such tumors; inhibition of Par-4 may be one such effect.