Mutant Isocitrate Dehydrogenase 1 Product, 2-Hydroxy Glutarate, Activates MutT Homolog 1 in Glioma Cells via Augmentation of Reactive Oxygen Species


 MutT Homolog1 (MTH1) is an enzyme responsible for removing oxidized nucleotides from cells. Activation of MTH1 is reported in many cancer cells and is thought to be responsible for imparting resistance towards anticancer drugs. While there are several mechanisms for the activation of MTH1 in cancer cells, this study aimed to evaluate the role of mutant Isocitrate Dehydrogenase1 (mIDH1) - mediated reactive oxygen species (ROS) in the activation of MutT Homolog1 in glioma cells. MTH1 was found to be upregulated in both mIDH1 expressed and 2- HG treated cells. mIDH1 and its product, 2-HG, increased the ROS levels in the cultured glioblastoma cells. Further, increased expression and activity of MTH1 was observed in glioma tissues harboring mIDH1 compared to tissues with wild-type IDH1. Our study unveils a novel mechanism of activation of MTH1 in cells harboring mutant IDH1.


Introduction
Isocitrate dehydrogenase (IDH) enzymes are key players in various metabolic processes such as the Tri Carboxylic Acid (TCA) cycle, lipogenesis, glutamine metabolism and redox regulation (Han et al. 2020). These NAD(P)+-dependent enzymes catalyses the conversion of isocitrate to α-ketoglutarate (α-KG) via the oxidative decarboxylation and generates NAD(P)H (Shi et al. 2013). Mutations in IDH1 were rst identi ed in human glioma tissue biopsies by whole genome sequencing, wherein it was observed that some mutations are commonly seen in most of the low-grade glioma (LGG) and secondary glioblastomas (GBM) (Parsons et al. 2008). The most important of these being a point mutation where the arginine gets replaced by histidine (R132H) at the 132nd position. IDH1 mutation status has now been included as an essential criteria for the glioma classi cation as per the WHO classi cation of the Central Nervous System Tumors of 2016 (Louis et al. 2016).
It has been earlier reported that IDH1 mutation generates increased ROS in cells via its enzymatic product, 2-Hydroxyglutarate (2-HG) (Garrett et al. 2018). ROS thus generated can interact with several biomolecules leading to activation of various signaling events leading to robust responses like antioxidant and DNA damage repair pathways. Among these, ROS mediated activation of various sanitation enzymes such as MutT Homolog1 (MTH1) has been reported (Qing et al. 2018). MTH1 belongs to a superfamily of enzymes called the nucleoside diphosphates linked to moiety-X (NUDIX) hydrolases and is potentially the only one enzyme involved in preventing mutations in DNA (Gad et al. 2014). In our previous study, we had reported that silencing of MTH1 can affect glioma cell migration, invasion and inhibits the regulators of angiogenesis (Bhavya et al. 2020b). In the present study, we hypothesized that mutant-IDH1 (mIDH1) and its product, 2-HG, could be responsible for the activation of MTH1, via the production of ROS. Two human glioblastoma cell lines (U87 and U251) were transfected with mIDH1 plasmid and were analyzed for MTH1 expression. mIDH1 expressed human glioblastoma cells showed an increased MTH1 expression. When 2-HG, a product of mutant IDH1 enzyme, was exogenously supplied to the cells, MTH1 expression was found to be signi cantly high. Our results also suggest that the 2-HG is a major contributor of increased ROS generation in mIDH1 cells. In order to verify the above ndings originated from cell lines, we checked for MTH1 expression levels/activity in IDH1 wild type gliomas and with IDH1 mutation. We found a positive correlation between MTH1 expression and IDH1 mutation in glioma patient biopsies. Consistent with this, 8-oxo-dG levels, which are indicative of the MTH1 activity were found to be high in mIDH1 harboring glioma biopsies than the wild-type IDH1 (wt-IDH1) patients. This study provides evidence of mIDH1/2-HG mediated ROS in activating MTH1 in both glioma cell lines and in mIDH1 gliomas.

Intracellular ROS measurement
The intracellular ROS levels were measured in cultured cells after various treatments using Dichlorodihydro uorescein diacetate (DCF-DA). Cells were cultured in a 96-well black plate at a seeding density of 1x10 4 per well. The wells were washed with HBSS after treatment and incubated with 10 µM DCFH-DA at 37°C in the dark for 1 h. Cells treated with H 2 O 2 were used as a positive control. The wells were washed 2 times with HBSS to remove excess dye, and the DCF uorescence developed was measured using Fluorimeter (BioTek instruments, Winooski, USA) at 530 nm (excitation 488 nm), with Gen5 software. The relative uorescence of the treated groups to the control was calculated using the uorescence intensities from triplicates.

Immunoblot
Glioblastoma cells and glioma tissues were processed for protein expression analysis and probed for desired proteins along with loading controls as described elsewhere (Bhavya et al. 2020b

Enzyme Immuno Assays
Universal 8-oxo-dG ELISA kit (ImmunoTag, St. Louis, MO, USA) was used in order to measure the levels of 8-oxo-dG which is an indicator of MTH1 activity. The cell/tissue lysates were prepared as per the kit protocol and added in triplicates into the 96-well plate coated with antibodies against 8-oxo-dG. Brie y, the biotinylated antibodies were added which was followed by Streptavidin-HRP to label them. Then the substrate (provided with the kit) for HRP was added for color development and the reaction was terminated after 10 min. The absorbance was then measured at 450 nm using ELISA plate reader (Bio- Tek Instruments, VT, USA) and the 8-oxo-dG concentrations were extrapolated from their respective standard curves.  (Fig. 1A). Similar results were obtained with the U251 cells as well (Fig. 1B). Subsequently, U87MG cells treated with 2-HG showed signi cant upregulation of MTH1 (p=0.0308, 1.65±0.20) compared to the untreated control (Fig. 1C). When checked for the expression of OGG1, a base excision repair enzyme was signi cantly upregulated in mIDH1 expressed and 2-HG treated U251 cells, but upon inhibiting mIDH1 with its speci c inhibitor, reduced its expression (Fig. 1D).
Increased MTH1 expression and activity in mIDH1 harboring glioma tissues The DNA sequencing data of a subset of the patient samples (n=21) denoting the IDH1 status shows that ve of them had R132H mutation (Fig. 3A). Out of the 57 glioma patient samples, MTH1 expression was found to be elevated in patients harboring mIDH1 when compared to patients with wild-type IDH1. The Western blot analyses of glioma tissues showed that MTH1 expression was signi cantly upregulated (p=0.0249) in patients harboring IDH1 mutation (n=27, 1.51±0.15) when compared to patients carrying wild-type IDH1 (n=30, 1.04±0.13) ( Fig. 3B and 3C). In patients with mIDH1, there is a moderate positive correlation with MTH1 expression (n=27; r= 0.4354, p= 0.0232) (Fig. 3D).
Next, 8-oxo-dG levels, which are indicative of activity of MTH1 enzyme was measured in glioma tissue biopsy extracts by immunoassay. We found increased 8-oxo-dG levels in those patients harboring mIDH1 (n= 15; 855.3 ± 242.6 ng/mg) compared to patients with wtIDH1 (n=9; 332.6 ± 114.1 ng/mg) (p=0.0148) (Fig. 3E), which is in concordance with the correlation found between MTH1 and mIDH1 expression. This indicates that concurrent with the high MTH1 expression pattern observed in mIDH1 glioma samples; there is a relative increase in MTH1 activity too. Since increased ROS is associated with greater propensity for oxidant mediated DNA damage in cells, we hypothesized that mIDH1/2-HG may induce the expression of MTH1, an enzyme responsible for sanitizing oxidant nucleotide pool in cells. Our results were in line with our hypothesis and 2-HG treated and mIDH1 expressed U87MG as well as U251 cells showed high level of MTH1 expression. The increased MTH1 expression in mIDH1 expressed cells was signi cantly reduced upon treatment with a speci c mIDH1 inhibitor AGI-5198, which establishes the role of mIDH1/2-HG in regulating MTH1 levels in cells.

Discussion
Augmented ROS levels seen in both mIDH1 expressed and 2-HG treated cells are in concordance with an earlier report (Gilbert et al. 2014). This shift in the redox status of the cells is likely to trigger the escalation of the antioxidant enzymes, GPx4 and SOD2, as seen from our results. In order to check, whether obliteration of basal ROS would signi cantly affect the prooxidant in uences imparted by mIDH1/2-HG, cells were pretreated with NAC and probed for ROS levels. Intriguingly, NAC pretreatment had little or no in uence on the ROS inducing capabilities of mIDH1/2-HG. When probed for MTH1 expression under NAC pretreatment in mIDH1/2-HG treated cells, there was nearly 2-fold increase in protein levels and in the MTH1 enzyme activity, determined by the 8-oxodG levels in cells. These results indicate the augmented MTH1 activity in the treated cells, even after scavenging the basal ROS that clearly suggests ROS generated via mIDH1/2-HG is responsible for the activation MTH1 in glioma cells.
Interestingly, in glioma patient biopsies with IDH1 mutation, we observed a signi cantly higher MTH1 expression compared to the wtIDH1 samples. Concomitant increase in the activity (as assessed by the increased 8-oxo-dG levels) supported the protein expression results in glioma tissues.
The highlight of the current study is that, we found a link between mIDH1 and MTH1 activation, mediated by ROS. This forms the basis for the increased MTH1 expression seen in mIDH1 glioma tissues and is the rst study linking IDH1 mutation, ROS and activation of MTH1 in cells. As it was found that mIDH1/2-HG causes increased DNA damage (more OGG1 expression) in U251 cells, this might yet be another factor that played a role in the upregulation of MTH1 in gliomas. Further experiments are required in order to decipher the mechanistic molecules involved in relation to MTH1 and mIDH1.