Pharmacologic inhibition of FTO using FB23-2 reduces tumor growth rates in intracranial IDH1wt gliomasphere xenografts. A, Pharmacokinetic analysis of FB23-2 permeability through the BBB showing drug elimination dynamics in plasma and brain (2 mice/timepoint). B, Schematic of in vivo experiments whereby intracranial gliomasphere xenografts were established before undergoing randomization to daily intraperitoneal injections of either FB23-2 treatment or DMSO control, with injection volumes based on individual weights (20 mg/kg). In vivo monitoring of tumor growth measured via plasma Gaussia Luciferase activity (tail vein) revealed differences at 7–8 weeks (GS187) and 6–7 weeks (XDS4130; at week 8, only 1 mouse remained in DMSO group), respectively [ANOVA, (C) treatment: F(1,16) = 4.7, P ≤ 0.05; time: F(7,106) = 79.1, P ≤ 0.0001; interaction: F(7,106) = 4.3, P ≤ 0.0003; (D) treatment: F(1,18) = 16.7, P ≤ 0.0007; time: F(1.6,23.6) = 41.7, P ≤ 0.0001; interaction: F(7,104) = 5.3, P ≤ 0.0001; asterisks indicate post hoc Newman–Keuls on day 7 and 8]. Kaplan–Meier curves showing overall survival differences in xenograft mice treated with FB23-2 (3 µmol/L) or DMSO control, evaluated via a log-rank (Mantel–Cox) test for GS187 (E, P < 0.005) and XDS4130 (F, P ≤ 0.05). G, Activated Cas3 immunostaining (G, 10X magnification) of representative XDS4130 xenografts (surgically extracted and flash-frozen post-mortem). “MS” refers to the mouse identification number. H, Increase in number of Cas3 positive cells in tumors from mice treated with FB23-2 relative to DMSO controls (P ≤ 0.005). *, P ≤ 0.05; **, P ≤ 0.01; ***, P ≤ 0.001; and ****, P ≤ 0.0001 compared with relevant controls. Unless otherwise stated, P values indicate unpaired Student t test comparisons with the control, or between two groups as indicated by the horizontal line.
Funding
U.S. Department of Defense (DOD)
HHS | NIH | National Cancer Institute (NCI)
UC | UCLA | Jonsson Comprehensive Cancer Center (JCCC)
HHS | NIH | National Institute of General Medical Sciences (NIGMS)
ARTICLE ABSTRACT
IDH1mut gliomas produce high levels of D-2-hydroxyglutarate (D-2-HG), an oncometabolite capable of inhibiting α-ketoglutarate–dependent dioxygenases critical to a range of cellular functions involved in gliomagenesis. IDH1mut gliomas also exhibit slower growth rates and improved treatment sensitivity compared with their IDH1wt counterparts. This study explores the mechanism driving apparent reduced growth in IDH1mut gliomas. Specifically, we investigated the relationship between IDH1mut and the RNA N6-methyladenosine (m6A) demethylases FTO and ALKBH5, and their potential for therapeutic targeting. We investigated the role of D-2-HG and m6A in tumor proliferation/viability using glioma patient tumor samples, patient-derived gliomaspheres, and U87 cells, as well as with mouse intracranial IDH1wt gliomasphere xenografts. Methylation RNA immunoprecipitation sequencing (MeRIP-seq) RNA sequencing was used to identify m6A-enriched transcripts in IDH1mut glioma. We show that IDH1mut production of D-2-HG is capable of reducing glioma cell growth via inhibition of the m6A epitranscriptomic regulator, FTO, with resultant m6A hypermethylation of a set of mRNA transcripts. On the basis of unbiased MeRIP-seq epitranscriptomic profiling, we identify ATF5 as a hypermethylated, downregulated transcript that potentially contributes to increased apoptosis. We further demonstrate how targeting this pathway genetically and pharmacologically reduces the proliferative potential of malignant IDH1wt gliomas, both in vitro and in vivo. Our work provides evidence that selective inhibition of the m6A epitranscriptomic regulator FTO attenuates growth in IDH1wt glioma, recapitulating the clinically favorable growth phenotype seen in the IDH1mut subtype.
We show that IDH1mut-generated D-2-HG can reduce glioma growth via inhibition of the m6A demethylase, FTO. FTO inhibition represents a potential therapeutic target for IDH1wt gliomas and possibly in conjunction with IDH1mut inhibitors for the treatment of IDH1mut glioma. Future studies are necessary to demonstrate the role of ATF5 downregulation in the indolent phenotype of IDH1mut gliomas, as well as to identify other involved gene transcripts deregulated by m6A hypermethylation.
