Analysis of the X-chromosome cancer driver gene Lysine-specific demethylase 5C (KDM5C) in Glioblastoma Multiforme (GBM) reveals novel molecular signatures

Glioblastoma multiforme (GBM) is the most aggressive tumour of brain without an effective pharmacological treatment. Very few data have been reported on the identification of GBM biomarkers making the development of personalized treatments even more difficult. We analyzed the levels of the cancer driver gene Lysine (K)-specific demethylase 5C (KDM5C) in a pilot series of tumour tissues isolated from an Italian cohort of GBM patients. KDM5C belongs to the KDM5 subfamily of Jumonji C domain-containing histone demethylases involved in various types of cancers including breast, colon, ovarian and prostate cancers. This chromatin enzyme catalyzes the removal of the methyl groups from di- and tri-methylated lysine 4 on histone H3 in a Fe (II)- and ?-ketoglutarate-dependent manner. By using real-time quantitative PCR and Western blotting analysis, we found an altered abundance of KDM5C transcript and protein in GBM samples identifying patients with higher (KDM5CHigh) and lower (KDM5CLow) levels compared to control samples. Analysis of KDM5C levels in GBM series of TCGA and GTEx databases revealed two distinct GBM cohorts with high- and low KDM5C expression associated to a different DNA methylation profile. By exploring the impact of the defective KDM5C quantity, a positive and negative relationship respectively with hypoxia-inducible transcription factor-1? (HIF-1?) and BDNF levels were found in KDM5CHigh patients. KDM5C overexpression and hypoxic studies performed in glioblastoma cell line (T98G) suggest that the stimulation of KDM5C expression is preceded by the induction of HIF-1?. High levels of HIF1?-KDM5C axis was also found associated with high levels of NANOG, SOX2 and NESTIN in GBM tissues isolated from conventional and 5-aminoleveulinic acid (5-ALA) fluorescence-guided surgery (FGS). A proinflammatory condition was also detected in 5-ALA FGS highlighting differences across the GBM microenvironment. Taken together, our study reveals for the first time a correlation between the HIF-1?-KDM5C axis and GBM opening a new field of investigation to validate KDM5C as a new GBM biomarker.