Exploitation of DNA repair defects has enabled major advances in treating specific cancers. Recent work discovered that the oncometabolite 2-hydroxyglutarate (2-HG), produced by neomorphic isocitrate dehydrogenase 1/2 (IDH1/2) mutations, confers a homology-directed repair (HDR) defect through 2-HG–induced histone hypermethylation masking HDR signaling. Here, we report that IDH1-mutant cancer cells are profoundly sensitive to the histone deacetylase inhibitor (HDACi) vorinostat, by further suppressing the residual HDR in 2-HG–producing cells. Vorinostat downregulates repair factors BRCA1 and RAD51 via disrupted E2F-factor regulation, causing increased DNA double-strand breaks, reduced DNA repair factor foci, and functional HDR deficiency even beyond 2-HG's effects. This results in greater cell death of IDH1-mutant cells and confers synergy with radiation and PARPi, both against cells in culture and patient-derived tumor xenografts. Our work identifies HDACi's utility against IDH1-mutant cancers, and presents IDH1/2 mutations as potential biomarkers to guide trials testing HDACi in gliomas and other malignancies.

Implications:

IDH1-mutant cells show profound vulnerability to HDACi treatment, alone and with PARPi and radiation, via HDR suppression, presenting IDH1/2 mutations as biomarkers for HDACi use in gliomas and other malignancies.

Recurrent metabolic gene mutations have been identified across various cancers, prompting the study of oncometabolites, metabolites whose accumulation is associated with cancer. Most notable are mutations in isocitrate dehydrogenase (IDH) genes, IDH1 and IDH2, which convert isocitrate to α-ketoglutarate (α-KG). Cancer-associated IDH mutations occur as heterozygous missense mutations that confer the neomorphic function of converting α-KG to 2-hydroxyglutarate (2-HG; ref. 1). IDH1/2 mutations are found in multiple cancers, including 80% of low-grade gliomas, 20% of higher-grade glioblastomas (2), 20% of acute myeloid leukemias (3), and 20% of cholangiocarcinomas (4).

2-HG's oncogenic role has been linked to its inhibition of α-KG–dependent dioxygenases, outcompeting α-KG due to similar chemical structure and highly elevated concentrations. These enzymes include the JmjC domain-containing histone demethylases (KDM), whose inhibition is linked to 2-HG–induced histone hypermethylation (5). Investigation of epigenetic aberrations in 2-HG–producing cells has largely focused on altered gene expression (6). Recent work presented a new dimension, identifying a DNA repair defect (7), consistent with improved survival of patients with IDH-mutant glioblastoma, (8) and superior response to radiation and chemotherapy (9, 10). An initial investigation of DNA repair in IDH1/2-mutant cells found a defect in homology directed repair (HDR), correlating with sensitivity to PARP inhibitors (PARPi; ref. 11). PARPi vulnerability as a potential therapeutic strategy was first demonstrated in BRCA1/2-mutant cancers, capitalizing on HDR deficiency by inhibiting single-strand break repair (12, 13). Mechanistic studies have revealed that IDH1/2 mutations impact HDR upstream from BRCA1/2 by disrupting early chromatin signaling at a double-strand break (DSB) site. Specifically, histone demethylase KDM4B inhibition results in aberrant histone 3 lysine 9 trimethylation (H3K9me3) that masks an H3K9 methylation signal essential for HDR factor recruitment. However, this inhibition is not complete, as residual HDR can still be detected in IDH1-mutant cells at sites that remain hypomethylated despite elevated 2-HG levels (7). This prompted us to hypothesize that further suppression of HDR in IDH1-mutant cells by histone deacetylase inhibitors (HDACi) might present another mechanism for IDH-mutant–specific targeting.

HDACi are known to inhibit HDR by downregulating the expression of repair factors RAD51 and BRCA1 (14–18). These effects are reported for vorinostat, also known as SAHA, an FDA-approved HDACi that inhibits Class I and II HDACs and crosses the blood–brain barrier (19, 20). Targeting RAD51 and BRCA1 expression presents an attractive approach for suppressing residual HDR in IDH-mutant cells, as reducing factor availability acts orthogonally to impaired recruitment. Additionally, HDACi are established radiosensitizers and synergize with PARPi in various cancer types (18, 21–27). Based on preclinical studies against wild-type (WT) IDH gliomas, alone and in combination with PARPi and radiation (28–34), SAHA has been advanced into clinical trials for glioma (35). However, these trials were conducted without selection by IDH1/2 status and have yielded only modest results.

Here, we have sought to test SAHA's effect on the DNA repair and survival of IDH-mutant cells and tumors, with the ultimate goal of informing future trials. We demonstrate SAHA's ability to downregulate HDR factor expression, specifically through E2F-factor transcriptional regulation. Downregulation of these factors further decreases the residual HDR in IDH1-mutant cells, as measured by reporter gene HDR assays, DNA repair factor foci formation, and DNA DSB levels via the comet assay. This HDR suppression mediates single-agent activity of SAHA against IDH1-mutant cells and provides for enhanced effectiveness of SAHA in combination with radiation or PARPi, yielding marked tumor growth delay in patient-derived, IDH1-mutant human xenografts in mice. These findings identify SAHA as a potential agent for specific targeting of IDH1/2-mutant cancers.

Cell culture

HT1080, U2OS EJDR, maintained in DMEM + 10% FBS, HCT116 IDH1+/+ and IDH1 R132H/+, maintained in McCoy 5A + 10% FBS, and U87 IDH1+/+ and IDH1 R132H/+, maintained in DMEM/F12 + 10% FBS, were obtained from R. Bindra (Yale University School of Medicine). Immortalized astrocytes were obtained from T. Chan (Cleveland Clinic; ref. 36) and maintained in DMEM + 10% FBS. Cells were passaged twice after thawing before use in experiments, regularly tested for Mycoplasma (MycoAlert, Lonza) and authenticated using short tandem repeat profiling (ATCC).

Drugs and reagents

SAHA, olaparib, BMN-673, AGI-5198, and temozolomide were all obtained from SelleckChem, and dissolved in DMSO for cell culture experiments. For animal tumor studies, BMN-673 was diluted in a vehicle containing 10% dimethylacetamide, 6% Kolliphor HS 15, and 84% PBS. SAHA was diluted in PBS. Both were administered to mice via oral gavage 5 days a week, for 3 weeks. Octyl-2-HG [(2R)-octyl-α-HG;] Cayman Chemical) and Octyl-α-KG (Sigma-Aldrich) were dissolved in DMSO. Horizon Dharmacon ON-TARGETplus Human BRCA1 and RAD51 siRNA were used at 20 nmol/L for 72 hours.

Western blotting

Cells were lysed in AZ lysis buffer. Band intensity was quantified by densitometry via ImageJ software, and normalized to Vinculin or β-actin as endogenous controls. Antibodies are provided in Supplementary Materials.

qRT-PCR analysis

RNA was isolated via RNA Miniprep kit (Qiagen) and used for cDNA synthesis via High-Capacity cDNA Reverse Transcription Kit (Applied Biosystems). cDNA was used in qRT-PCR assays using Taqman Universal PCR Master Mix and Gene Expression Assays (Applied Biosystems) and a StepOnePlus RT-PCR system (Thermo Fisher Scientific). 18S was used as an endogenous control.

EJDR HDR assay

U2OS EJDR cells were pretreated with SAHA for 24 hours and 2-HG for 3 days before adding ligands, triamcinolone (0.5 μmol/L, Sigma), and Shield1 (100 nmol/L, Clontech), to media for 24 hours. Cells were analyzed by flow cytometry after 72 hours for GFP signal.

Neutral comet assays

Neutral comet assays were performed as previously described (11). Imaging was completed with an EVOS FL microscope (Advanced Microscopy Group) and analyzed via OpenComet software (37).

Immunofluorescent foci analysis

Cells were seeded in 8-chamber slides, with or without SAHA, and irradiated 24 hours later using an X-RAD 320 X-Ray Biological Irradiator (Precision X-Ray, Inc). Nonirradiated controls were completed in parallel. Cells were fixed, permeabilized, and incubated in blocking buffer overnight at 4°C. Cells were incubated in RAD51 primary for 30 minutes at room temperature, and overnight at 4°C. After washes, samples were incubated with secondary in blocking for 75 minutes at room temperature. After washes, samples were incubated with γH2AX primary for 90 minutes at room temperature, and overnight at 4°C. Antibodies are provided in Supplementary Materials. DNA was stained with DAPI for 15 minutes at room temperature. After washes, chambers were removed and slides were covered with coverslips using DAKO Fluorescence Mounting Medium (Dako NA, Inc.) and sealed with nail polish. Images were taken using a Nikon Eclipse TiE inverted fluorescence microscope (Nikon Corp.), Plan Apo 60X/1.40 Oil DIC h objective, CSU-W1 confocal spinning disk unit (Yokogawa Corp. of America) with the iXon Ultra888 EMCCD (Andor Technology), MLC 400B laser unit (Agilent Technologies), and NIS Elements 4.30 software (Nikon Corp.). Images were saved as multi-channel .nd2 files with no further editing. Representative images of 55 × 55 microns were prepared with ImageJ software and saved as 8-bit .tiff images. For γ-H2AX and RAD51, cells with greater than 10 foci/nuclei were considered foci-positive.

Cell viability and clonogenic assays

Cell viability assays were performed using CellTiter-Glo (Promega). One thousand cells/well were seeded in a 96-well plate, then treated with SAHA, olaparib, or BMN-673 for 7 days. Luminescence was quantified using a Synergy HR plate reader (BioTek). Five-hundred cells per well were seeded in 6-well plates in drug-free media for clonogenic survival assays, then changed to drugged medium or irradiated after 24 hours. Cells were treated with SAHA for 2 days, olaparib, BMN-673, and Temozolomide for 5 days. Plates were kept in an incubator for 10 to 14 days, depending on colony size. Cells were washed in PBS, stained with crystal violet, and manually counted. For combinations, cells were seeded into SAHA-drugged media, then irradiated or switched to combined treatment after 24 hours. Synergy was calculated via Combenefit software's HSA model (38).

Mouse tumor xenograft studies

Human HT1080 cells (5 × 106) and HCT116 cells (1 × 106) were implanted subcutaneously in the flanks of athymic nu/nu mice (Harlan). Three times a week, tumors were measured using calipers to calculate tumor volume [V = ½(4π/3)(length/2)(width/2)(height)]. Protocol was approved by Yale University Institutional Animal Care and Use Committee.

Statistical analysis

Statistical analysis was completed in GraphPad Prism. Unless otherwise stated, data is presented as means ± SEM and n = 3 replicates. Significance was determined using unpaired t tests or ANOVA, all two-sided with a testing level of 0.05.

SAHA suppresses HDR factor expression in IDH1-mutant cells

To first assess SAHA's impact on DNA repair in IDH1-mutant cells, we tested changes in the expression of repair factors BRCA1 and RAD51 after treatment with SAHA. We chose the HT1080 line, a fibrosarcoma line carrying an endogenous IDH1 R132C/+ mutation, as well as an engineered U87 isogenic glioma pair, mutant IDH1 R132H/+, and WT IDH1+/+. We observed dose-dependent decreases in BRCA1 and RAD51 protein, to greater effect in the U87 glioma cell background, and even greater decreases in respective mRNA levels (Fig. 1A and B). We also probed SAHA's epigenetic effect on IDH1-mutant cells, looking at H3K9me3 for 2-HG–induced hypermethylation as well as H3K9 acetylation (H3K9ac) and pan histone 4 acetylation (H4ac) for HDACi effects. We observed similarly elevated acetylation in all three lines, while seeing larger decreases in H3K9me3 in IDH1-mutant cells, due to elevated baselines (Supplementary Fig. S1A). These results reveal that SAHA reduces 2-HG–induced hypermethylation, while suppressing HDR factor expression, informing our study of DNA repair function.

Figure 1.

SAHA downregulates HDR factor expression in IDH1 mutant cells. A, Western blot of HDR factors in glioma pair (U87) and patient-derived mutant IDH1 fibrosarcoma cells (HT1080) 24 hours after treatment with increasing SAHA doses (n = 3 independent experiments). B, Relative gene expression of HDR factors measured at the mRNA level by qRT-PCR in U87 and HT1080 cells treated with increasing SAHA doses (two-way ANOVA, effect of SAHA: U87+/+, U87 R132H/+, HT1080, P< 0.001; n = 3 independent experiments). C, Western bot of HDR factors in HT1080 cells at baseline and after radiation (6 Gy), with and without 24 hours SAHA pretreatment (5 μmol/L; n = 3 independent experiments). Data are represented as means ± SEM. Values below Western blot images represent relative quantification normalized to loading control. hr, hour.

Figure 1.

SAHA downregulates HDR factor expression in IDH1 mutant cells. A, Western blot of HDR factors in glioma pair (U87) and patient-derived mutant IDH1 fibrosarcoma cells (HT1080) 24 hours after treatment with increasing SAHA doses (n = 3 independent experiments). B, Relative gene expression of HDR factors measured at the mRNA level by qRT-PCR in U87 and HT1080 cells treated with increasing SAHA doses (two-way ANOVA, effect of SAHA: U87+/+, U87 R132H/+, HT1080, P< 0.001; n = 3 independent experiments). C, Western bot of HDR factors in HT1080 cells at baseline and after radiation (6 Gy), with and without 24 hours SAHA pretreatment (5 μmol/L; n = 3 independent experiments). Data are represented as means ± SEM. Values below Western blot images represent relative quantification normalized to loading control. hr, hour.

Close modal

Additionally, we investigated whether SAHA's suppression of HDR factor expression was sustained during a radiation-induced DNA damage response, in patient-derived HT1080 cells. Radiation induced a time-dependent increase in BRCA1 and RAD51 in control DMSO-treated HT1080 cells, whereas there was a sustained downregulation of BRCA1 and RAD51 in SAHA-treated cells, even after radiation (Fig. 1C).

SAHA-induced HDR down-regulation is mediated via E2F1 loss

Seeing strong suppression of RAD51 and BRCA1 expression at the mRNA level by SAHA treatment, we sought to identify the mechanism of the transcriptional downregulation. It has been proposed that HDACi HDR suppression may be due to disrupted E2F transcription factor regulation, specifically E2F1 activator loss, in prostate cancer models (16). E2F-factor dysregulation is also more broadly established as an HDR suppression mechanism in hypoxia and specific drug treatments (39–42). Hence, we first tested SAHA's effect on E2F1 activator and E2F4 repressor expression, in both HT1080 and U87 matched-pair cells. SAHA decreased both E2F1 and E2F4 protein and mRNA expression, but with greater impact on E2F1 (Fig. 2A and B). This differential effect prompted us to perform chromatin immunoprecipitation (ChIP) assays for the two factors to interrogate occupancy at RAD51 and BRCA1 promoter sites, with and without treatment, in HT1080 and U87 matched-pair cells. E2F1 and E2F4 ChIP revealed occupancy decreases in both factors but, importantly, there was a much greater fold decrease in E2F1 activator occupancy, as previously reported (Fig. 2C; ref. 16). This large reduction in E2F1 occupancy at RAD51 and BRCA1's promoter sites provides a mechanistic explanation for their transcriptional downregulation.

Figure 2.

SAHA-induced HDR downregulation is mediated via E2F1 loss. A, Western blot of E2F factors, E2F1 and E2F4, in U87 and HT1080 cells 24 hours after treatment with increasing SAHA doses (n = 3 independent experiments). B, Relative gene expression of E2F factors by qRT-PCR of HDR factors in U87 and HT1080 cells 24 hours after treatment with increasing SAHA doses (two-way ANOVA, effect of SAHA: U87+/+, U87 R132H/+, HT1080, P < 0.001; effect of E2F1 vs. E2F4: U87+/+, U87 R132H/+, HT1080, P < 0.001; n = 3 independent experiments). C, ChIP analysis of E2F1 and E2F4 occupancy at HDR promoters in U87 and HT1080 after 24 hours of SAHA treatment (5 μmol/L) represented as fold change from untreated controls (two-way ANOVA, effect of E2F1 vs. E2F4: P < 0.001; n = 3 technical replicates). Data are represented as means ± SEM. Values below Western blot images represent relative quantification normalized to loading control.

Figure 2.

SAHA-induced HDR downregulation is mediated via E2F1 loss. A, Western blot of E2F factors, E2F1 and E2F4, in U87 and HT1080 cells 24 hours after treatment with increasing SAHA doses (n = 3 independent experiments). B, Relative gene expression of E2F factors by qRT-PCR of HDR factors in U87 and HT1080 cells 24 hours after treatment with increasing SAHA doses (two-way ANOVA, effect of SAHA: U87+/+, U87 R132H/+, HT1080, P < 0.001; effect of E2F1 vs. E2F4: U87+/+, U87 R132H/+, HT1080, P < 0.001; n = 3 independent experiments). C, ChIP analysis of E2F1 and E2F4 occupancy at HDR promoters in U87 and HT1080 after 24 hours of SAHA treatment (5 μmol/L) represented as fold change from untreated controls (two-way ANOVA, effect of E2F1 vs. E2F4: P < 0.001; n = 3 technical replicates). Data are represented as means ± SEM. Values below Western blot images represent relative quantification normalized to loading control.

Close modal

SAHA suppresses residual HDR in IDH1-mutant cells

After confirming that SAHA downregulates HDR factor expression, we next investigated whether this has a functional effect on DNA repair in IDH1-mutant cells. Because of SAHA's ability to decrease H3K9me3 levels by enhancing H3K9ac, we were concerned that this might mitigate 2-HG's effects on DNA repair, which result from H3K9 hypermethylation. To address this, we tested whether intermediate 2-HG and SAHA doses, that alone cause moderate reductions in HDR, might worsen DNA repair in the HDR pathway in combination. We used the U2OS EJDR reporter cell line, a human osteosarcoma cell line that carries a ligand-inducible, chromosomally integrated GFP-based HDR reporter (43). This system employs a widely used approach for gauging HDR activity: restoration of a mutated GFP reporter when a neighboring donor region is used as a repair template due to sequence homology. Repair is triggered by site-specific cutting within the reporter gene by a ligand-responsive I-SceI nuclease, following the addition of the ligands shield-1 and triamcinolone which enable protein stabilization and nuclear localization. U2OS EJDR cells were pretreated with SAHA and/or cell-permeable (2R)-Octyl-2-HG (here-on referred as Octyl-2-HG) before ligand treatment and scored for HDR by quantifying GFP positivity. By itself, Octyl-2-HG treatment elevated H3K9me3 levels and decreased HDR activity, as expected (11). Octyl-2-HG's suppression of HDR activity was dose-dependent, with a moderate effect seen with 0.5 mmol/L and a larger decrease with 1 mmol/L. This decrease in HDR treatment is specific to Octyl-2-HG, as no significant reduction was seen when cells were treated with a high dose of Octyl-α-KG. SAHA treatment alone decreased HDR, also in a dose-dependent manner, as well as downregulating BRCA1 and RAD51 expression while increasing H3K9ac. High Octyl-2-HG and SAHA doses lead to significant HDR suppression, though not to the extent of full siRNA knockdown of RAD51 and BRCA1. Importantly, we found that the combination of moderate Octyl-2-HG and SAHA doses had an additive effect in suppressing HDR, resulting in a reduction similar to high single-agent doses. This result demonstrates that SAHA worsens the overall DNA repair deficiency in cells with high 2-HG (although by a different mechanism) and does not reverse it (Fig. 3A and B). Notably, addition of SAHA overrode Octyl-2-HG treatment's increase in histone hypermethylation, seen at H3 lysines 9, 27, and 36, with HDACi forcing robust global acetylation seen via Western blot (Fig. 3B; Supplementary Fig. S2A). 2-HG intracellular concentrations were confirmed by a 2HGDH-mediated enzyme assay after treatment (Supplementary Fig. S2B), and siRNA knockdown of BRCA1 and RAD51 was confirmed by Western blot analysis (Supplementary Fig. S2C). The additive effect of combined SAHA and Octyl-2-HG treatment in suppressing HDR shows that SAHA's effect of increasing H3K9 acetylation while decreasing methylation is not sufficient to reverse the HDR defect caused by 2-HG.

Figure 3.

SAHA suppresses residual HDR in IDH1 mutant cells. A, Representative sorting and quantification of HDR via percent GFP-positive U2OS EJDR cells after treatment with DMSO, SAHA (3, 5 μmol/L), and/or Octyl-2-HG (0.5, 1 mmol/L), Octyl-α-KG (1 mmol/L), siBRCA1, or siRAD51 (t test, DMSO vs. SAHA 3 μmol/L, vs. SAHA 5 μmol/L, vs. Octyl-2-HG 0.5 mmol/L, vs. Octyl-2-HG 1 mmol/L: P< 0.001; DSMO vs. SAHA + Octyl-2-HG: P = 0.0015; SAHA 3 μmol/L vs. SAHA + Octyl-2-HG: P = 0.0118; Octyl-2-HG 0.5 mmol/L vs. SAHA+ Octyl-2-HG: P = 0.0322; SAHA 3 μmol/L vs. SAHA 5 μmol/L, Octyl-2-HG 0.5 mmol/L vs. Octyl-2-HG 1 mmol/L: P < 0.001; SAHA 5 μmol/L vs. SAHA + Octyl-2-HG: P = 0.4083; Octyl-2-HG 1 mmol/L vs. SAHA + Octyl-2-HG: P = 0.3923; Octyl-α-KG vs. Octyl-2-HG 1 mmol/L: P < 0.001; DMSO vs. Octyl-α-KG: 0.1613; n = 3 biological replicates). B, Western blot of HDR factors, H3K9 trimethylation and acetylation in U2OS EJDR cells after SAHA and Octyl-2-HG treatment (n = 3 independent experiments). C, Representative images and quantification of neutral comet assays for HT1080 cells completed 24 hours after SAHA, olaparib, AGI-5198, or combined treatment (Scale bar 400 μmol/L; t test, DMSO vs. AGI-5198: P < 0.001; DMSO vs. SAHA: P = 0.0011; DMSO vs. olaparib: P< 0.001; SAHA vs. SAHA olaparib: P = 0.042; olaparib vs. SAHA olaparib: P = 0.0412; AGI-5198 vs. AGI-5198 SAHA, vs. AGI-5198 olaparib: P< 0.001; AGI-5198 SAHA vs. AGI-5198 olaparib SAHA: P = 0.0014; AGI-5198 olaparib vs. AGI-5198 olaparib SAHA: P = 0.0065). D, Quantification of neutral comet assays for U87 cells completed after SAHA and/or olaparib, representative images in Supplementary Fig. S3A (t test, U87+/+ vs. U87 R132H/+: P = 0.0054; DMSO vs. SAHA, DMSO vs. olaparib, SAHA vs. SAHA olaparib, olaparib vs. SAHA olaparib: U87+/+, U87 R132H/+, P< 0.001). E, Percentage of RAD51 and F, γ-H2AX positive cells (threshold: 10 foci/nuclei) of HT1080 and U87 cells at indicated times postirradiation (2 Gy), with or without SAHA (5 μmol/L; two-way ANOVA, RAD51: HT1080, U87+/+, Pinteraction < 0.001; U87 R132H/+, Pinteraction = 0.0021; γH2AX: HT1080, Pinteraction < 0.001; U87+/+, Pinteraction = 0.0013; U87 R132H/+, Pinteraction = 0.0087; n = 3 biological replicates). G, Representative images of RAD51 and γ-H2AX foci in HT1080 cells with or with SAHA, at baseline and 4 hours postirradiation (green, RAD51; far-red, γH2AX; blue, DAPI; images represent 55 × 55 μmol/L). Full HT1080 and U87 course in Supplementary Fig. S4A. Data are represented as means ± SEM. Values below Western blot images represent relative quantification normalized to loading control.

Figure 3.

SAHA suppresses residual HDR in IDH1 mutant cells. A, Representative sorting and quantification of HDR via percent GFP-positive U2OS EJDR cells after treatment with DMSO, SAHA (3, 5 μmol/L), and/or Octyl-2-HG (0.5, 1 mmol/L), Octyl-α-KG (1 mmol/L), siBRCA1, or siRAD51 (t test, DMSO vs. SAHA 3 μmol/L, vs. SAHA 5 μmol/L, vs. Octyl-2-HG 0.5 mmol/L, vs. Octyl-2-HG 1 mmol/L: P< 0.001; DSMO vs. SAHA + Octyl-2-HG: P = 0.0015; SAHA 3 μmol/L vs. SAHA + Octyl-2-HG: P = 0.0118; Octyl-2-HG 0.5 mmol/L vs. SAHA+ Octyl-2-HG: P = 0.0322; SAHA 3 μmol/L vs. SAHA 5 μmol/L, Octyl-2-HG 0.5 mmol/L vs. Octyl-2-HG 1 mmol/L: P < 0.001; SAHA 5 μmol/L vs. SAHA + Octyl-2-HG: P = 0.4083; Octyl-2-HG 1 mmol/L vs. SAHA + Octyl-2-HG: P = 0.3923; Octyl-α-KG vs. Octyl-2-HG 1 mmol/L: P < 0.001; DMSO vs. Octyl-α-KG: 0.1613; n = 3 biological replicates). B, Western blot of HDR factors, H3K9 trimethylation and acetylation in U2OS EJDR cells after SAHA and Octyl-2-HG treatment (n = 3 independent experiments). C, Representative images and quantification of neutral comet assays for HT1080 cells completed 24 hours after SAHA, olaparib, AGI-5198, or combined treatment (Scale bar 400 μmol/L; t test, DMSO vs. AGI-5198: P < 0.001; DMSO vs. SAHA: P = 0.0011; DMSO vs. olaparib: P< 0.001; SAHA vs. SAHA olaparib: P = 0.042; olaparib vs. SAHA olaparib: P = 0.0412; AGI-5198 vs. AGI-5198 SAHA, vs. AGI-5198 olaparib: P< 0.001; AGI-5198 SAHA vs. AGI-5198 olaparib SAHA: P = 0.0014; AGI-5198 olaparib vs. AGI-5198 olaparib SAHA: P = 0.0065). D, Quantification of neutral comet assays for U87 cells completed after SAHA and/or olaparib, representative images in Supplementary Fig. S3A (t test, U87+/+ vs. U87 R132H/+: P = 0.0054; DMSO vs. SAHA, DMSO vs. olaparib, SAHA vs. SAHA olaparib, olaparib vs. SAHA olaparib: U87+/+, U87 R132H/+, P< 0.001). E, Percentage of RAD51 and F, γ-H2AX positive cells (threshold: 10 foci/nuclei) of HT1080 and U87 cells at indicated times postirradiation (2 Gy), with or without SAHA (5 μmol/L; two-way ANOVA, RAD51: HT1080, U87+/+, Pinteraction < 0.001; U87 R132H/+, Pinteraction = 0.0021; γH2AX: HT1080, Pinteraction < 0.001; U87+/+, Pinteraction = 0.0013; U87 R132H/+, Pinteraction = 0.0087; n = 3 biological replicates). G, Representative images of RAD51 and γ-H2AX foci in HT1080 cells with or with SAHA, at baseline and 4 hours postirradiation (green, RAD51; far-red, γH2AX; blue, DAPI; images represent 55 × 55 μmol/L). Full HT1080 and U87 course in Supplementary Fig. S4A. Data are represented as means ± SEM. Values below Western blot images represent relative quantification normalized to loading control.

Close modal

Next, we directly assessed DNA DSB levels in IDH1-mutant cells treated with or without SAHA by measuring DSBs using the neutral comet assay. In patient-derived HT1080 IDH1 R132C/+ cells, SAHA treatment elevated the comet phenotype, suggesting increased DNA damage in the form of DSBs (Fig. 3C). To test the relative contribution of 2-HG on SAHA treatment's effect, HT1080 cells were treated with AGI-5198, an inhibitor of IDH1 R132C's neomorphic 2-HG production (44), with or without SAHA. As expected, AGI-5198 significantly reduced the comet phenotype (11), demonstrating 2-HG's role in increased DNA damage at baseline. With AGI-5198 treatment, SAHA still elevated the level of DSB damage from this lower baseline (Fig. 3C). SAHA's effect of increasing DSBs was also seen in both the IDH1 WT and matched IDH1 R132H/+ U87 cells, with IDH1-mutant cells having increased DSB damage at baseline (Fig. 3D; Supplementary Fig. S3A). These results demonstrate that SAHA's repair effect occurs independent of 2-HG production, and is additive in IDH1-mutant cells, causing greater DNA damage. Next, we tested the impact on DNA DSBs when SAHA is combined with the PARPi olaparib. We found that olaparib increased the comet tail moments in HT1080 cells, with and without AGI-5198 treatment, and in the U87 pair, relative to baseline, in keeping with prior work (11). Importantly, we found that combination with SAHA increased the level of DSBs above and beyond the effect of olaparib alone (Fig. 3C and D; Supplementary Fig. S3A). IDH1-mutant U87 cells ultimately experienced higher levels of cellular DNA damage following SAHA treatment, likely resulting from their baseline repair deficiency. This differential effect was also seen with treatment of siBRCA1 and siRAD51, correlating with the impact of HDR downregulation caused by SAHA. HDR factor loss via siRNA knockdown elevated WT U87 cells to the mutant baseline and IDH1-mutant cells to a much higher level of cellular DNA damage (Supplementary Fig. S3B and S3C). These results demonstrate that SAHA elevates constitutive DSBs to greater effect in IDH1-mutant cells and presents a rational for possible combination with PARPi.

Lastly, to further test the functional HDR effect caused by SAHA, we performed immunofluorescent DNA repair factor foci analyses for RAD51 (indicative of HDR) and γ-H2AX (a marker for DSBs) post irradiation in HT1080 and the U87 pair. Specifically, cells were scored foci-positive with a count of 10 or more foci per nucleus. Across the 3 cell lines, SAHA treatment decreased RAD51 foci at timepoints both before and after radiation, indicative of reduced HDR capacity, consistent with the gene expression and comet assay data (Fig. 3E and G; Supplementary Fig. S4A). We did note more diffuse staining and greater background in SAHA-treated cells, potentially resulting from reduced RAD51 abundance and reduced signal-to-noise. In accord with reduced RAD51 foci, increased γ-H2AX foci formation postradiation persisted at higher levels and for longer times over 24 hours in the SAHA-treated cells, indicating unresolved DSBs (Fig. 3F and G; Supplementary Fig. S4A). SAHA-treated cells had both more and larger γ-H2AX foci than untreated cells, as seen in representative images, potentially indicative of more complex DSB sites (Fig. 3G; Supplementary Fig. S4A). Foci formation patterns of the matched U87 glioma pair demonstrates the expected DNA repair defect in IDH1-mutant cells, with greater damage measured by γH2AX foci and reduced HDR measured by RAD51 foci. SAHA's effects, while seen in both IDH1 WT and mutant U87 glioma cells, ultimately results in a greater repair crisis in IDH1-mutant cells.

IDH1-mutant cells have a greater vulnerability to SAHA treatment

To test the implication of this SAHA-induced repair crisis on cell survival, we investigated whether IDH1-mutant cells have a greater sensitivity to SAHA treatment. Using the U87 pair, we performed long-term clonogenic survival assays (CSA) and short-term ATP-based viability assays following SAHA treatment. We saw greater SAHA sensitivity in IDH1-mutant glioma cells with both methods (Fig. 4A), to an extent similar to the sensitivity seen via CSA of glioma cells treated with the PARPi olaparib and BMN-673 (talazoparib; Supplementary Fig. S5A). IDH1-mutant sensitivity was further supported by a greater increase of cleaved PARP, an established apoptosis marker. The mutant U87 line experienced a greater than 10-fold increase in cleaved PARP from baseline following SAHA treatment versus a 2.4-fold increase in WT (Fig. 4B).

Figure 4.

IDH1 mutant cells have a greater vulnerability to SAHA treatment. Clonogenic survival and cell viability, measured by ATP-based CellTiter-Glo assay, of A, U87 pair and C, IDH1 R132H-inducible M70 astrocyte line ± doxycycline (dox), treated with increasing SAHA doses (two-way ANOVA, U87: survival, Pinteraction = 0.0014; viability, Pinteraction < 0.001; M70: survival, Pinteraction = 0.0012; viability, Pinteraction < 0.001; viability n = 4, survival n = 3 biological replicates). B, Western blot of Cleaved PARP in U87 pair after 24 hours of increasing SAHA doses (n = 3 independent experiments). D, Western blot of Cleaved PARP, HDR factors, and IDH1 R132H in M70 cells ± doxycycline after 24 hours of increasing SAHA doses (n = 3 independent experiments). Data are represented as means ± SEM. Values below Western blot images represent relative quantification normalized to loading control.

Figure 4.

IDH1 mutant cells have a greater vulnerability to SAHA treatment. Clonogenic survival and cell viability, measured by ATP-based CellTiter-Glo assay, of A, U87 pair and C, IDH1 R132H-inducible M70 astrocyte line ± doxycycline (dox), treated with increasing SAHA doses (two-way ANOVA, U87: survival, Pinteraction = 0.0014; viability, Pinteraction < 0.001; M70: survival, Pinteraction = 0.0012; viability, Pinteraction < 0.001; viability n = 4, survival n = 3 biological replicates). B, Western blot of Cleaved PARP in U87 pair after 24 hours of increasing SAHA doses (n = 3 independent experiments). D, Western blot of Cleaved PARP, HDR factors, and IDH1 R132H in M70 cells ± doxycycline after 24 hours of increasing SAHA doses (n = 3 independent experiments). Data are represented as means ± SEM. Values below Western blot images represent relative quantification normalized to loading control.

Close modal

To further demonstrate IDH1-mutant cells' heightened vulnerability to SAHA, we utilized an immortalized human astrocyte line with a stably integrated, doxycycline-inducible IDH1 R132H expression construct. Astrocytes were treated with or without doxycycline for multiple passages, then tested for SAHA sensitivity. As seen in the U87 matched pair, mutant IDH1-expressing astrocytes had greater SAHA sensitivity (Fig. 4C), similar in extent to their sensitivity to olaparib and BMN-673 (Supplementary Fig. S5B). SAHA treatment also resulted in greater DNA damage as measured via the comet assay in the doxycycline-induced astrocytes, consistent with the cell survival and viability data (Supplementary Fig. S3D). Furthermore, there was a 5.0-fold increase in cleaved PARP response in doxycycline-induced cells versus 1.3-fold in the uninduced cells, while BRCA1 and RAD51 suppression was comparable in both contexts (Fig. 4D). These results further demonstrate SAHA's strong effect against IDH1-mutant cells.

SAHA combination with radiation and PARPi has greater effects against IDH1-mutant cells

Next, we sought to identify combinations that capitalized on this HDR suppression, first investigating SAHA's combination with radiation against IDH1-mutant cells via CSAs in HT1080 and U87 matched-pair cells. Combination of SAHA and radiation treatment caused an additive effect across all 3 lines, but this combined effect was greater in IDH1-mutant cells (U87+/+P = 0.1836, U87 R132H/+ P < 0.001; Fig. 5A). HSA model synergy of this combination showed a synergistic effect across all 3 lines, though to a much greater degree in IDH1-mutant HT1080 (R132C) and U87 (R132H) cells (Supplementary Fig. S6A). The sensitivity of the IDH1-mutant U87 glioma line to this combination is particularly significant given the modest results reported from clinical trials of SAHA in glioma that were conducted without IDH1/2 status selection, specifically, those of a Phase I/II study (NCT00731731) testing combination of SAHA, radiation, and temozolomide (45). Pertinent to this, we also tested the combination of SAHA and temozolomide in the U87 matched pair. We found increased temozolomide sensitivity in the IDH1-mutant cells, as expected (46), as well as a stronger synergistic effect in combination with SAHA (Fig. 5B; Supplementary Fig S6B). These results present a further rationale for using IDH1/2 mutations as biomarkers for future clinical trials testing HDACi in combination therapies in patients with gliomas.

Figure 5.

SAHA combination with radiation, temozolomide (TMZ), and PARPi is effective against IDH1 mutant cells in cell culture and tumor xenografts. Clonogenic survival of HT1080 and U87 cells treated with increasing doses of A, SAHA and radiation, (two-way ANOVA, HT1080, Pinteraction = 0.0296; U87 R132H/+, interaction P< 0.001; U87+/+, Pinteraction = 0.1836; n = 3 technical replicates), B, SAHA and temozolomide (two-way ANOVA, U87 R132H/+, Pinteraction < 0.001; U87+/+, Pinteraction = 0.0011; n = 3 technical replicates), C, SAHA and PARPi, BMN-673, and olaparib, (two-way ANOVA, BMN-673 & Olaparib: HT1080, U87 R132H/+, U87+/+, Pinteraction < 0.001; n = 3 technical replicates). D, Growth curves of HT1080 xenograft tumors in control mice and mice treated with SAHA (30 mg/kg) and/or BMN-673 (0.33 mg/kg; two-way ANOVA, control vs. BMN-673, Pinteraction = 0.1765; control vs. SAHA, Pinteraction = 0.0132; control vs. SAHA BMN-673, Pinteraction < 0.001; SAHA vs. SAHA BMN-673, Pinteraction = 0.3245; n = 10 mice). Toxicity studies of SAHA, BMN-673 and combination via E, mouse body weight (two-way ANOVA, vehicle vs. BMN-673: P = 0.0693; vehicle vs. SAHA BMN-673: P = 0.1933; vehicle vs. SAHA: P = 0.0142) and F, bone-marrow cell counts (t test, vehicle vs. SAHA: P = 0.8223; vehicle vs. BMN-673: P = 0.0016; vehicle vs. SAHA BMN-673: P = 0.0017; BMN-673 vs. SAHA BMN-673: P = 0.8535). G, Growth curves of HCT116 IDH1 R132H/+ xenograft tumors in control mice and mice treated with SAHA (30 mg/kg) and/or BMN-673 (0.33 mg/kg; two-way ANOVA, control vs. BMN-673, Pinteraction = 0.0257; control vs. SAHA, Pinteraction >0.9999; control vs. SAHA BMN-673, Pinteraction < 0.001; SAHA vs. SAHA BMN-673, Pinteraction < 0.001; BMN-673 vs. SAHA BMN-673, Pinteraction = 0.0332; n = 7 mice). Data are represented as means ± SEM.

Figure 5.

SAHA combination with radiation, temozolomide (TMZ), and PARPi is effective against IDH1 mutant cells in cell culture and tumor xenografts. Clonogenic survival of HT1080 and U87 cells treated with increasing doses of A, SAHA and radiation, (two-way ANOVA, HT1080, Pinteraction = 0.0296; U87 R132H/+, interaction P< 0.001; U87+/+, Pinteraction = 0.1836; n = 3 technical replicates), B, SAHA and temozolomide (two-way ANOVA, U87 R132H/+, Pinteraction < 0.001; U87+/+, Pinteraction = 0.0011; n = 3 technical replicates), C, SAHA and PARPi, BMN-673, and olaparib, (two-way ANOVA, BMN-673 & Olaparib: HT1080, U87 R132H/+, U87+/+, Pinteraction < 0.001; n = 3 technical replicates). D, Growth curves of HT1080 xenograft tumors in control mice and mice treated with SAHA (30 mg/kg) and/or BMN-673 (0.33 mg/kg; two-way ANOVA, control vs. BMN-673, Pinteraction = 0.1765; control vs. SAHA, Pinteraction = 0.0132; control vs. SAHA BMN-673, Pinteraction < 0.001; SAHA vs. SAHA BMN-673, Pinteraction = 0.3245; n = 10 mice). Toxicity studies of SAHA, BMN-673 and combination via E, mouse body weight (two-way ANOVA, vehicle vs. BMN-673: P = 0.0693; vehicle vs. SAHA BMN-673: P = 0.1933; vehicle vs. SAHA: P = 0.0142) and F, bone-marrow cell counts (t test, vehicle vs. SAHA: P = 0.8223; vehicle vs. BMN-673: P = 0.0016; vehicle vs. SAHA BMN-673: P = 0.0017; BMN-673 vs. SAHA BMN-673: P = 0.8535). G, Growth curves of HCT116 IDH1 R132H/+ xenograft tumors in control mice and mice treated with SAHA (30 mg/kg) and/or BMN-673 (0.33 mg/kg; two-way ANOVA, control vs. BMN-673, Pinteraction = 0.0257; control vs. SAHA, Pinteraction >0.9999; control vs. SAHA BMN-673, Pinteraction < 0.001; SAHA vs. SAHA BMN-673, Pinteraction < 0.001; BMN-673 vs. SAHA BMN-673, Pinteraction = 0.0332; n = 7 mice). Data are represented as means ± SEM.

Close modal

Next, given IDH1-mutant cells' strong PARPi sensitivity and SAHA's HDR suppression, we tested SAHA in combination with olaparib and BMN-673, in HT1080 and U87 matched-pair cells. We found very strong combined effects with both olaparib and BMN-673 in all three lines, though survival of the IDH1 mutant cells was more highly affected (Fig. 5C). The PARPi plus SAHA combination had strong synergy, as expected, in both IDH1-mutant and wild-type U87 lines, though importantly IDH1-mutant cells showed greater cell death (Supplementary Fig. S6C and S6D).

To extend our findings to an in vivo model, we performed a tumor xenograft study in mice using HT1080 cells, with four arms testing vehicle, SAHA alone, BMN-673 alone, and SAHA plus BMN-673 combination. Interestingly, moderate SAHA dosing strongly delayed tumor growth of the IDH1-mutant HT1080 xenografts, with a greater growth inhibition than seen with aggressive BMN-673 dosing. SAHA's effect was further enhanced in combination with BMN-673, yielding the greatest tumor growth delay (Fig. 5D). There was no added toxicity in the combination treatment arm beyond single-agent effects as measured by mouse body weight throughout the study and bone-marrow cell counts at the study's end (Fig. 5E and F).

We replicated these results with a second tumor xenograft model using an engineered IDH1 R132H/+ HCT116 colorectal carcinoma line. In cell culture, SAHA had the same effects on HCT116 cells as seen in glioma and patient-derived cells: downregulation of BRCA1 and RAD51 expression (Supplementary Fig. S7A), increased constitutive DSBs measured via neutral comet assay (Supplementary Fig. S7B), and stronger effects alone and in combination with PARPi against IDH1-mutant cells (Supplementary Fig. S7C and S7D). PARPi plus SAHA combination had strong synergy in both IDH1-mutant and WT cells (Supplementary Fig. S7E). This synergy was demonstrated in mutant IDH1 HCT116 tumor xenografts, with SAHA plus BMN-673 combination resulting in the greatest tumor growth delay (Fig. 5G). These results strongly demonstrate the benefit of SAHA plus PARPi combination against IDH1-mutant tumors in vivo, highlighting HDACi's potential clinical utility in the context of IDH-mutant malignancies.

In this work, we demonstrate the HDACi SAHA's ability to suppress residual HDR in IDH1-mutant cells, through downregulation of repair factors BRCA1 and RAD51 mediated by disrupted E2F-factor transcriptional regulation. This mechanism occurs independent of 2-HG, and displays an additive effect resulting in a greater DNA repair crisis due to compounding losses in DNA repair-factor abundance and DNA-repair signaling. This repair crisis underlies SAHA's strong effects against IDH1-mutant cells alone and in combination with radiation and PARPi. We demonstrate the effectiveness of SAHA alone and in combination with PARPi BMN-673 against patient-derived, IDH1-mutant HT1080 mouse tumor xenografts. The benefit of HDACi use in fibrosarcoma was previously identified with a large SAHA dose (1,050 mg/kg/week) against HT1080 xenografts, though without exploring the role of IDH1-mutant status or DNA repair (47). Here, we present strong tumor growth delay with much smaller dosing (150 mg/kg/week), and data demonstrating SAHA's suppression of residual HDR in IDH1-mutant cells.

HDACi's effects on HDR factor expression (14–18) and its utility against WT gliomas (31–34) has been previously explored, but not in the IDH1/2-mutant context. Our work provides new insight into the utility of HDACi in the subset of gliomas that are IDH1/2 mutant, demonstrating HDACi suppression of residual HDR, based on reporter gene HDR assays and repair foci analyses. A remaining question is the extent to which elevated H3K9ac as a result of HDAC inhibition may impede methylation-based epigenetic signaling and factor recruitment at a DSB, which we plan to address in future studies as another potential mechanistic contribution. Nonetheless, our focus in this work centered on SAHA's functional impact on IDH1-mutant cells' residual HDR, and its resulting therapeutic utility.

We further present multiple treatment combinations that capitalize on the DNA repair crisis in IDH1-mutant cells following SAHA treatment. SAHA's IDH1-mutant–dependent synergy with radiation and temozolomide, seen in glioma cells, provides clear support for consideration of IDH1/2 status in the design of clinical trials using these agents. Furthermore, the effectiveness of the SAHA plus PARPi combination was demonstrated in vivo against 2 IDH1-mutant xenograft models, presenting support for testing this strategy in trials of PARPi in IDH1/2-mutant cancers. SAHA's utility against IDH1-mutant cells, alone or in combination, has major relevance to glioma, though the application of PARPi and SAHA will depend on the advancement of suitable brain-penetrant PARPi for clinical use. Overall, the work presented here highlights a new strategy for targeting IDH1/2-mutant cancers based on suppression of residual HDR with HDACi.

P.M. Glazer reports grants from NIH during the conduct of the study; personal fees from Cybrexa Therapeutics; and personal fees from Gennao Bio outside the submitted work; in addition, P.M. Glazer has a patent for “Compositions and methods for targeting and treating homologous recombination-deficient tumors” pending and a patent for “Compositions and Methods For Targeting Cancers” pending. No disclosures were reported by the other authors.

J. Dow: Conceptualization, formal analysis, validation, investigation, visualization, methodology, writing–original draft, project administration, writing–review and editing. A. Krysztofiak: Formal analysis, validation, investigation, visualization, methodology. Y. Liu: Formal analysis, investigation. D.A. Colon-Rios: Investigation. F.A. Rogers: Supervision. P.M. Glazer: Conceptualization, resources, supervision, funding acquisition, visualization, writing–original draft, project administration, writing–review and editing.

This work was supported by the NIH grants R35CA197574 and R01ES005775 (to P.M. Glazer). J. Dow was supported by the National Institute of General Medical Science training grant T32GM007499. We thank D. Hegan and A. Dhawan for their assistance. We thank Dr. Joerg Nikolaus from the Yale West Campus Imaging Core for technical support.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1.
Dang
L
,
White
DW
,
Gross
S
,
Bennett
BD
,
Bittinger
MA
,
Driggers
EM
, et al
Cancer-associated IDH1 mutations produce 2-hydroxyglutarate
.
Nature
2009
;
462
:
739
44
.
2.
Yan
H
,
Parsons
DW
,
Jin
G
,
McLendon
R
,
Rasheed
BA
,
Yuan
W
, et al
IDH1 and IDH2 mutations in gliomas
.
N Engl J Med
2009
;
360
:
765
73
.
3.
Mardis
ER
,
Ding
L
,
Dooling
DJ
,
Larson
DE
,
McLellan
MD
,
Chen
K
, et al
Recurring mutations found by sequencing an acute myeloid leukemia genome
.
N Engl J Med
2009
;
361
:
1058
66
.
4.
Jiao
Y
,
Pawlik
TM
,
Anders
RA
,
Selaru
FM
,
Streppel
MM
,
Lucas
DJ
, et al
Exome sequencing identifies frequent inactivating mutations in BAP1, ARID1A and PBRM1 in intrahepatic cholangiocarcinomas
.
Nat Genet
2013
;
45
:
1470
3
.
5.
Xu
W
,
Yang
H
,
Liu
Y
,
Yang
Y
,
Wang
P
,
Kim
SH
, et al
Oncometabolite 2-hydroxyglutarate is a competitive inhibitor of α-ketoglutarate-dependent dioxygenases
.
Cancer Cell
2011
;
19
:
17
30
.
6.
Losman
JA
,
Kaelin
WG
 Jr
. 
What a difference a hydroxyl makes: mutant IDH, (R)-2-hydroxyglutarate, and cancer
.
Genes Dev
2013
;
27
:
836
52
.
7.
Sulkowski
PL
,
Oeck
S
,
Dow
J
,
Economos
NG
,
Mirfakhraie
L
,
Liu
Y
, et al
Oncometabolites suppress DNA repair by disrupting local chromatin signalling
.
Nature
2020
;
582
:
586
91
.
8.
Parsons
DW
,
Jones
S
,
Zhang
X
,
Lin
JCH
,
Leary
RJ
,
Angenendt
P
, et al
An integrated genomic analysis of human glioblastoma multiforme
.
Science
2008
;
321
:
1807
12
.
9.
Tran
AN
,
Lai
A
,
Li
S
,
Pope
WB
,
Teixeira
S
,
Harris
RJ
, et al
Increased sensitivity to radiochemotherapy in IDH1 mutant glioblastoma as demonstrated by serial quantitative MR volumetry
.
Neuro Oncol
2014
;
16
:
414
20
.
10.
Cairncross
JG
,
Wang
M
,
Jenkins
RB
,
Shaw
EG
,
Giannini
C
,
Brachman
DG
, et al
Benefit from procarbazine, lomustine, and vincristine in oligodendroglial tumors is associated with mutation of IDH
.
J Clin Oncol
2014
;
32
:
783
90
.
11.
Sulkowski
PL
,
Corso
CD
,
Robinson
ND
,
Scanlon
SE
,
Purshouse
KR
,
Bai
H
, et al
2-Hydroxyglutarate produced by neomorphic IDH mutations suppresses homologous recombination and induces PARP inhibitor sensitivity
.
Sci Transl Med
2017
;
9
:
eaal2463
.
12.
Bryant
HE
,
Schultz
N
,
Thomas
HD
,
Parker
KM
,
Flower
D
,
Lopez
E
, et al
Specific killing of BRCA2-deficient tumours with inhibitors of poly(ADP-ribose) polymerase
.
Nature
2005
;
434
:
913
7
.
13.
Farmer
H
,
McCabe
N
,
Lord
CJ
,
Tutt
AN
,
Johnson
DA
,
Richardson
TB
, et al
Targeting the DNA repair defect in BRCA mutant cells as a therapeutic strategy
.
Nature
2005
;
434
:
917
21
.
14.
Adimoolam
S
,
Sirisawad
M
,
Chen
J
,
Thiemann
P
,
Ford
JM
,
Buggy
JJ
. 
HDAC inhibitor PCI-24781 decreases RAD51 expression and inhibits homologous recombination
.
Proc Natl Acad Sci
2007
;
104
:
19482
7
.
15.
Weberpals
JI
,
O'Brien
AM
,
Niknejad
N
,
Garbuio
KD
,
Clark-Knowles
KV
,
Dimitroulakos
J
. 
The effect of the histone deacetylase inhibitor M344 on BRCA1 expression in breast and ovarian cancer cells
.
Cancer Cell Int
2011
;
11
:
29
.
16.
Kachhap
SK
,
Rosmus
N
,
Collis
SJ
,
Kortenhorst
MS
,
Wissing
MD
,
Hedayati
M
, et al
Downregulation of homologous recombination DNA repair genes by HDAC inhibition in prostate cancer is mediated through the E2F1 transcription factor
.
PLoS One
2010
;
5
:
e11208
.
17.
Lai
TH
,
Ewald
B
,
Zecevic
A
,
Liu
C
,
Sulda
M
,
Papaioannou
D
, et al
HDAC inhibition induces microRNA-182, which targets RAD51 and impairs HR repair to sensitize cells to sapacitabine in acute myelogenous leukemia
.
Clin Cancer Res
2016
;
22
:
3537
49
.
18.
Ha
K
,
Fiskus
W
,
Choi
DS
,
Bhaskara
S
,
Cerchietti
L
,
Devaraj
SGT
, et al
Histone deacetylase inhibitor treatment induces ‘BRCAness’ and synergistic lethality with PARP inhibitor and cisplatin against human triple negative breast cancer cells
.
Oncotarget
2014
;
5
:
5637
50
.
19.
Villar-Garea
A
,
Esteller
M
. 
Histone deacetylase inhibitors: understanding a new wave of anticancer agents
.
Int J Cancer
2004
;
112
:
171
8
.
20.
Palmieri
D
,
Lockman
PR
,
Thomas
FC
,
Hua
E
,
Herring
J
,
Hargrave
E
, et al
Vorinostat inhibits brain metastatic colonization in a model of triple-negative breast cancer and induces DNA Double-strand breaks
.
Clin Cancer Res
2009
;
15
:
6148
57
.
21.
Munshi
A
,
Tanaka
T
,
Hobbs
ML
,
Tucker
SL
,
Richon
VM
,
Meyn
RE
. 
Vorinostat, a histone deacetylase inhibitor, enhances the response of human tumor cells to ionizing radiation through prolongation of gamma-H2AX foci
.
Mol Cancer Ther
2006
;
5
:
1967
74
.
22.
Chen
X
,
Wong
JY
,
Wong
P
,
Radany
EH
. 
Low-dose valproic acid enhances radiosensitivity of prostate cancer through acetylated p53-dependent modulation of mitochondrial membrane potential and apoptosis
.
Mol Cancer Res
2011
;
9
:
448
61
.
23.
Chen
X
,
Wong
P
,
Radany
EH
,
Stark
JM
,
Laulier
C
,
Wong
JYC
. 
Suberoylanilide hydroxamic acid as a radiosensitizer through modulation of RAD51 protein and inhibition of homology-directed repair in multiple myeloma
.
Mol Cancer Res
2012
;
10
:
1052
64
.
24.
Min
A
,
Im
SA
,
Kim
DK
,
Song
SH
,
Kim
HJ
,
Lee
KH
, et al
Histone deacetylase inhibitor, suberoylanilide hydroxamic acid (SAHA), enhances anti-tumor effects of the poly (ADP-ribose) polymerase (PARP) inhibitor olaparib in triple-negative breast cancer cells
.
Breast Cancer Res
2015
;
17
:
33
.
25.
Konstantinopoulos
PA
,
Wilson
AJ
,
Saskowski
J
,
Wass
E
,
Khabele
D
. 
Suberoylanilide hydroxamic acid (SAHA) enhances olaparib activity by targeting homologous recombination DNA repair in ovarian cancer
.
Gynecol Oncol
2014
;
133
:
599
606
.
26.
Chao
OS
,
Goodman
OB
 Jr
. 
Synergistic loss of prostate cancer cell viability by coinhibition of HDAC and PARP
.
Mol Cancer Res
2014
;
12
:
1755
66
.
27.
Yin
L
,
Liu
Y
,
Peng
Y
,
Peng
Y
,
Yu
X
,
Gao
Y
, et al
PARP inhibitor veliparib and HDAC inhibitor SAHA synergistically co-target the UHRF1/BRCA1 DNA damage repair complex in prostate cancer cells
.
J Exp Clin Cancer Res
2018
;
37
:
153
.
28.
Eyüpoglu
IY
,
Hahnen
E
,
Buslei
R
,
Siebzehnrübl
FA
,
Savaskan
NE
,
Lüders
M
, et al
Suberoylanilide hydroxamic acid (SAHA) has potent anti-glioma properties in vitro, ex vivo and in vivo
.
J Neurochem
2005
;
93
:
992
9
.
29.
Yin
D
,
Ong
JM
,
Hu
J
,
Desmond
JC
,
Kawamata
N
,
Konda
BM
, et al
Suberoylanilide hydroxamic acid, a histone deacetylase inhibitor: effects on gene expression and growth of glioma cells in vitro and in vivo
.
Clin Cancer Res
2007
;
13
:
1045
52
.
30.
Cornago
M
,
Garcia-Alberich
C
,
Blasco-Angulo
N
,
Vall-Llaura
N
,
Nager
M
,
Herreros
J
, et al
Histone deacetylase inhibitors promote glioma cell death by G2 checkpoint abrogation leading to mitotic catastrophe
.
Cell Death Dis
2014
;
5
:
e1435
.
31.
Hsu
CC
,
Chang
WC
,
Hsu
TI
,
Liu
JJ
,
Yeh
SH
,
Wang
JY
, et al
Suberoylanilide hydroxamic acid represses glioma stem-like cells
.
J Biomed Sci
2016
;
23
:
81
.
32.
Yang
XF
,
Zhao
ZJ
,
Liu
JJ
,
Yang
XH
,
Gao
Y
,
Zhao
S
, et al
SAHA and/or MG132 reverse the aggressive phenotypes of glioma cells: An in vitro and vivo study
.
Oncotarget
2017
;
8
:
3156
69
.
33.
Barazzuol
L
,
Jeynes
JC
,
Merchant
MJ
,
Wéra
AC
,
Barry
MA
,
Kirkby
KJ
, et al
Radiosensitization of glioblastoma cells using a histone deacetylase inhibitor (SAHA) comparing carbon ions with X-rays
.
Int J Radiat Biol
2015
;
91
:
90
8
.
34.
Rasmussen
RD
,
Gajjar
MK
,
Jensen
KE
,
Hamerlik
P
. 
Enhanced efficacy of combined HDAC and PARP targeting in glioblastoma
.
Mol Oncol
2016
;
10
:
751
63
.
35.
Chen
R
,
Zhang
M
,
Zhou
Y
,
Guo
W
,
Yi
M
,
Zhang
Z
, et al
The application of histone deacetylases inhibitors in glioblastoma
.
J Exp Clin Cancer Res
2020
;
39
:
138
.
36.
Turcan
S
,
Rohle
D
,
Goenka
A
,
Walsh
LA
,
Fang
F
,
Yilmaz
E
, et al
IDH1 mutation is sufficient to establish the glioma hypermethylator phenotype
.
Nature
2012
;
483
:
479
83
.
37.
Gyori
BM
,
Venkatachalam
G
,
Thiagarajan
PS
,
Hsu
D
,
Clement
MV
. 
OpenComet: an automated tool for comet assay image analysis
.
Redox Biol
2014
;
2
:
457
65
.
38.
Di Veroli
GY
,
Fornari
C
,
Wang
D
,
Mollard
S
,
Bramhall
JL
,
Richards
FM
, et al
Combenefit: an interactive platform for the analysis and visualization of drug combinations
.
Bioinformatics
2016
;
32
:
2866
8
.
39.
Bindra
RS
,
Gibson
SL
,
Meng
A
,
Westermark
U
,
Jasin
M
,
Pierce
AJ
, et al
Hypoxia-induced down-regulation of BRCA1 expression by E2Fs
.
Cancer Res
2005
;
65
:
11597
604
.
40.
Hegan
DC
,
Lu
Y
,
Stachelek
GC
,
Crosby
ME
,
Bindra
RS
,
Glazer
PM
. 
Inhibition of poly(ADP-ribose) polymerase down-regulates BRCA1 and RAD51 in a pathway mediated by E2F4 and p130
.
Proc Natl Acad Sci U S A
2010
;
107
:
2201
6
.
41.
Bindra
RS
,
Glazer
PM
. 
Repression of RAD51 gene expression by E2F4/p130 complexes in hypoxia
.
Oncogene
2007
;
26
:
2048
57
.
42.
Kaplan
AR
,
Gueble
SE
,
Liu
Y
,
Oeck
S
,
Kim
H
,
Yun
Z
, et al
Cediranib suppresses homology-directed DNA repair through down-regulation of BRCA1/2 and RAD51
.
Sci Transl Med
2019
;
11
:
eaav4508
.
43.
Bindra
RS
,
Goglia
AG
,
Jasin
M
,
Powell
SN
. 
Development of an assay to measure mutagenic non-homologous end-joining repair activity in mammalian cells
.
Nucleic Acids Res
2013
;
41
:
e115
.
44.
Li
L
,
Paz
AC
,
Wilky
BA
,
Johnson
B
,
Galoian
K
,
Rosenberg
A
, et al
Treatment with a small molecule mutant IDH1 inhibitor suppresses tumorigenic activity and decreases production of the oncometabolite 2-hydroxyglutarate in human chondrosarcoma cells
.
PLoS One
2015
;
10
:
e0133813
.
45.
Galanis
E
,
Anderson
SK
,
Miller
CR
,
Sarkaria
JN
,
Jaeckle
K
,
Buckner
JC
, et al
Phase I/II trial of vorinostat combined with temozolomide and radiation therapy for newly diagnosed glioblastoma: results of Alliance N0874/ABTC 02
.
Neuro Oncol
2018
;
20
:
546
56
.
46.
Lu
Y
,
Kwintkiewicz
J
,
Liu
Y
,
Tech
K
,
Frady
LN
,
Su
YT
, et al
Chemosensitivity of IDH1-mutated gliomas due to an impairment in PARP1-Mediated DNA repair
.
Cancer Res
2017
;
77
:
1709
18
.
47.
Sampson
ER
,
Amin
V
,
Schwarz
EM
,
O'Keefe
RJ
,
Rosier
RN
. 
The histone deacetylase inhibitor vorinostat selectively sensitizes fibrosarcoma cells to chemotherapy
.
J Orthop Res
2011
;
29
:
623
32
.