Abstract
Small molecule imipridones including ONC201, ONC206 and ONC212 have anti-cancer activity mediated in part through the integrated stress response, induction of TRAIL and its receptor DR5, and activation of mitochondrial caseinolytic protease ClpP with impaired oxidative phosphorylation. ONC201 provides clinical benefit in a subset of patients with histone H3K27M-mutated diffuse glioma (DG). We hypothesized that EZH2 inhibitors (EZH2i) may sensitize tumors to imipridones by mimicking H3K27M mutation. EZH1 is a homolog and alternative for EZH2 in assembling PRC2 complex. We combined ONC201, ONC206 or ONC212 plus dual EZH1/2i in tumors and observed synergy. We observed synergies with imipridones combined with HDACi or triple combination of ONC201/ONC206, EZH2i and HDACi in DG, GBM, prostate cancer and SCLC cells. Our observations implicate EZH1/2 suppression in mechanism of anti-cancer effect of imipridones. We investigated effects of imipridones on EZH1/2 in DG cells and solid tumor cells including GBM, CRC, PDAC, SCLC, prostate cancer, gastric cancer, HCC and breast cancer cells and found inhibition of EZH1/EZH2 expression across tumor types and cell viability suppression by imipridones is correlated with EZH1/2 reduction. Imipridone or EZH2i-treated tumor cells showed similar cytokine profile changes. RNA-seq showed ONC201 and EHZ2i tazemetostat-treated cells have similar transcriptional profiles and share overlap of top regulated genes. Thus, imipridones inhibit EZH1/2 in tumor cells in a manner that mimics H3K27M mutation supporting their role in anti-cancer efficacy. ONC201 and EZH2i share similar targets and actions on tumors. Synergistic combinations of imipridones plus EZH1/2i or imipridones, EZH2i and HDACi merit further investigation.
Keywords: ONC201, ONC206, ONC212, EZH1, EZH2, glioma, DMG, H3K27M, tazemetostat, ISRIB, vorinostat, Panobinostat, valemetostat, cytokine profiles, epigenetic therapy, solid tumor
Introduction
ONC201 was discovered as an anti-cancer compound in the screen for p53-independent inducers of the TNF-related-apoptosis-inducing ligand (TRAIL) gene transcription [1-3]. It is the founding member of the imipridone family including another two members currently under development, ONC206 and ONC212. Imipridones exert a tumoricidal function through induction of TRAIL [2,3], activation of an HRI-dependent eIF2-alpha- and ATF4/CHOP-mediated integrated stress response with downstream induction of TRAIL receptor, DR5 [4]. These events are triggered by imipridone agonist action and activation of mitochondrial caseinolytic protease ClpP with consequent impairment of oxidative phosphorylation [5,6].
DG (diffuse glioma) is typically an inoperable tumor with a poor prognosis requiring the development of novel therapies for cure. More than 80% of DG patients carry the driver H3K27M mutation leading to epigenomic reprogramming and oncogenesis [7]. ONC201 was observed to provide clinical benefit in a subset of DG patients with histone H3K27M mutation in some early clinical data [8,9].
H3K27M mutation inhibits the function of EZH2 in methylating H3K27, thus reshaping the epigenetic profile [10]. EZH1 is a homolog of EZH2 and forms an alternative for EZH2 in assembling the PRC2 complex. Inhibition of EZH1/2 downregulates the H3K27me3 and derepressed target gene transcription, while histone acetylation leads to a more relaxed chromatin structure and ultimately gene transcription activation.
Based on the effect of imipridones on DG tumors carrying H3K27M mutation, we hypothesized that EZH2 inhibitor may sensitize such tumors to imipridones by mimicking H3K27M mutation. EZH2 inhibitors are effective in H3K27M-mutant DG, and the underlying mechanism may be a reduction of H3K27me3 in the tumors [11]. Dual inhibition of EZH1 and EZH2 was reported effective in some hematological malignancies in several preclinical studies [12-14].
These phenomena led us to investigate the action of imipridones on EZH1 and EZH2 protein expression and provide the rationale for the combination of imipridones plus EZH inhibitors or the triple combination of imipridones, EZH inhibitors and HDAC inhibitors in the treatment of tumors, especially given that we already observed synergy of ONC201 plus EZH2 inhibitor, ONC201 plus HDAC inhibitors or ONC201 plus EZH2 inhibitor and HDAC inhibitors in a panel of tumors in our previous studies [15].
We found in the present studies that imipridones reduce both EZH1 and EZH2 protein expression levels. This novel observation raised the question as to whether there are similarities between imipridones and EZH inhibitors with respect to their actions and molecular mechanisms in tumors. We performed RNA-seq and cytokine profile assays using tumor cells following treatment with imipridones or EZH2 inhibitor to look for the similarities.
Our results provide novel insights into the mechanism of action of imipridones and support the rationale for their combination with epigenetic drugs some of which are already clinically approved as glioma therapies.
Methods
Cell culture and reagents
All DG cell lines including SU-DIPG-4, SU-DIPG-13, SU-DIPG-25, SU-DIPG-29 and SU-DIPG-36 cell lines were provided by Dr. Michelle Monje at Stanford University and were generously shared by Dr. Tapinos with our group. The cells were maintained in Tumor Stem Medium (TSM) made of TSM Base enriched with B-27 supplement minus vitamin A, 0.2% Heparin (from STEMCELLTM Technologies, Vancouver, BC, Canada), human PDGF-BB, human PDGF-AA, human FGF-basic 154 aa (FGF2), and human EGF (from Shenandoah Biotechology Inc., Warwick, PA, USA). TSM base was made of 1:1 mixture of Neurobasal-A Medium and D-MEM/F-12 with Antibiotic-Antimycotic liquid, MEM Sodium Pyruvate Solution, non-essential amino acids solution, glutaMAX, HEPES Buffer solution (all purchased from Thermo Fisher Scientific Inc., Invitrogen brand, Carlsbad, CA, USA).
The human prostate cancer, PDAC, CRC, gastric cancer, SCLC, HCC, breast cancer and GBM cell lines were purchased from the American Type Culture Collection (ATCC). H1048 and H1882 SCLC cells were cultured in HITES medium [DMEM: F12 Medium supplemented with 0.005 mg/ml Insulin, 0.01 mg/ml Transferrin, 30 nM Sodium selenite (final conc.), 10 nM Hydrocortisone (final conc.), 10 nM beta-estradiol (final conc.), extra 2 mM L-glutamine (for final conc. of 4.5 mM) and 5% fetal bovine serum]. All the other cell lines were cultured in their ATCC-recommended media supplemented with 10% (v/v) fetal bovine serum and 1% Penicillin/Streptomycin. All cell lines were confirmed to be mycoplasma-free using PCR testing methods and cultured with or without chemotherapy agents at 37°C within a 95% humidified atmosphere containing 5% carbon dioxide in an incubator.
Tazemetostat was purchased from Selleckchem and was solubilized in DMSO at a storage concentration of 20 mM. Valemetostat was purchased from Selleckchem and was solubilized in DMSO at a storage concentration of 10 mM. Vorinostat was purchased from MedKoo Biosciences and was solubilized in DMSO at a storage concentration of 50 mM. Panobinostat was purchased from MedKoo Biosciences Inc. and was solubilized in DMSA at a storage concentration of 20 mM.
ONC201, ONC206 and ONC212 were supplied by Chimerix, Inc. and reconstituted in DMSO at a storage concentration of 20 mM. Integrated stress response inhibitor (ISRIB) with the formula of C22H24Cl2N2O4 and the IUPAC name of trans-N, N’-(Cyclohexane-1,4-diyl)bis(2-(4-chlorophenoxy)acetamide), was purchased from Selleckchem and was solubilized in DMSO at a storage concentration of 20 mM.
Immunoblotting
Cells were seeded in 6-well plates at a density of 4-7×105 cells per well depending on the doubling time of the cells and incubated overnight in culture media before the addition of ONC201/ONC206/ONC212 alone or combination of ONC201/ONC206/ONC212 plus tazemetostat, ONC201/ONC206/ONC212 plus valemetostat, ONC201 plus vorinostat/panobinostat and tazemetostat or ONC206 plus panobinostat and tazemetostat. The culture was continued for 48 hours. Then, the cells were washed with PBS and lysed in lysis buffer [150 mM NaCl, 1% Triton X-100, 0.5% Sodium deoxycholate, 0.1% SDS, 50 mM Tris-HCl (pH 8.0)]. The proteins were quantified with the Bio-Rad protein assay. LDS Sample Buffer (4X) and reducing reagent were added to the lysates. For the immunoblotting of histone, the cells were lysed directly with lysis buffer [150 mM NaCl, 1% Triton X-100, 0.5% Sodium deoxycholate, 0.1% SDS, 50 mM Tris-HCl (pH 8.0)] plus LDS Sample Buffer (4X). The lysates were loaded equally onto 4 to 12% NuPAGE SDS-polyacrylamide gels (Thermo Fisher Scientific). Standard procedures were performed to transfer proteins to polyvinylidene difluoride (PVDF) membranes. After blocking with 5% milk, the PVDF membranes were incubated with primary antibody overnight and subsequently appropriate secondary antibodies labeled with horseradish peroxidase for 1 hour. The membranes were developed using an ECL reagent. The primary antibodies used in this study were as follows: antibodies against EZH1 (D7D5D) (cat. no. 42088, Cell Signaling), EZH2 (D2C9) XP (cat. no. 5246, Cell Signaling), Cleaved PARP (Asp214) (19F4) (cat. no. 9546S, Cell Signaling), Histone H3 (tri-methyl K27) antibody (mAbcam 6002), β-Actin (A5441, Millipore Sigma), and Ran (cat. no. 610341, BD Bioscience). Secondary antibodies were acquired from Pierce (cat. nos. 31430 and 31460) (horseradish peroxide-conjugated).
Cell viability and apoptosis assays
Cells were seeded in opaque-walled 96-well plates at a density of 25,000 cells per well for H1882 and 5000 cells per well for all the other cells and incubated overnight in 100 µL culture medium before the addition of ONC201, ONC206 or ONC212 alone or combination of ONC201, ONC206 or ONC212 plus tazemetostat, ONC201, ONC206 or ONC212 plus valemetostat, ONC201 plus vorinostat or panobinostat and tazemetostat or ONC206 plus panobinostat and tazemetostat. After treatment for 72 hours. 20 µL CellTiterGlo bioluminescence agent (Promega Corporation, Madison, WI) was added to each well. The content was mixed for 2 minutes on a plate shaker to induce cell lysis. Cell viability was determined by the CellTiterGlo assay. Combination indices (CI) were calculated by the method of Chou and Talalay using the CompuSyn software. CI < 1.0 indicates drug synergy.
Reverse-transcription polymerase chain reaction (RT-PCR)
MDA-MB-468 breast cancer cells were seeded in 6-well plates at a density of 5×105 cells per well and incubated overnight in culture media before the addition of ONC201. Treatment was continued for 24 hours. RNA was extracted using the RNeasy Plus Mini Kit (Qiagen) according to manufacturers’ instructions and quantitated using a Nanodrop spectrophotometer. Genes of interest were amplified using 2 μg of total RNA reverse-transcribed to cDNA using a Superscript II kit (Invitrogen) with random hexamer primer. In the RT-PCR step, PCR reactions were performed with 1 ml cDNA/20 ml reaction and primers specific for EZH1 (F: 5’-caattcaagctggcgaagag-3’, R: 5’-caagacagtgccgctacca-3’), EZH2 (F: 5’-gccaagagagccatccagac-3’, R: 5’-ccgacatacttcagggcatca-3’), and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) (F: 5’-gcatcttcttttgcgtc-3’, R: 5’-tgtaaaccatgtagttgaggt-3’) with AccuPrime™ Pfx DNA Polymerase kit (ThermoFisher). Thermal cycling was initiated at 95°C for 10 min followed by 35 cycles of PCR (95°C for 15 s and 60°C for 1 min). GAPDH was used as an endogenous control. The PCR product was detected by electrophoresis on 3% agarose gels.
Cytokine profile analysis
Cells were plated at 2.5×104 cells in a 48-well plate in a complete medium and incubated at 37°C with 5% CO2 overnight. The complete medium was replaced with a drug-containing medium until almost all the tumor cells were adherent to the bottom of the plates. Subsequently, the culture supernatants were collected after 48 hours of incubation and were frozen at -80°C until the measurement of cytokines was performed. On the day of analysis, samples were thawed and centrifuged to remove cellular debris. An R&D systems Human Premixed Multi-Analyte Kit (R&D Systems, Inc., Minneapolis, MN) was run on a Luminex 200 Instrument (LX200-XPON-RUO, Luminex Corporation, Austin, TX) according to the manufacturer’s instructions. Cell culture supernatant levels of TNF-alpha, 4-1BB/TNFRSF9/CD137, IL-8/CXCL8, Ferritin, IFN-beta, IL-10, CCL2/JE/MCP-1, VEGF, CXCL13/BLC/BCA-1, IFN-gamma, CCL20/MIP-3 alpha, CCL3/MIP-1 alpha, CCL22/MDC, CCL4/MIP-1 beta, Fas Ligand/TNFSF6, IL-17/IL-17A, IL-2, BAFF/BLyS/TNFSF13B, GM-CSF, CXCL5/ENA-78, TRANCE/TNFSF11/RANK L, CXCL9/MIG, G-CSF, IFN-gamma R1/CD119, VEGFR3/Flt-4, C-Reactive Protein/CRP, CXCL11/I-TAC, IL-21, CXCL14/BRAK, IL-6, Fas/TNFRSF6/CD95, TRAIL R3/TNFRSF10C, IL-4, CCL5/RANTES, PD-L1/B7-H1, CCL7/MCP-3/MARC, Chitinase 3-like 1, CXCL10/IP-10/CRG-2, IL-1 beta/IL-1F2, IL-7, Prolactin, CCL8/MCP-2, TRAIL R2/TNFRSF10B, M-CSF, IL-15, Granzyme B, IFN-alpha, TREM-1, IL-12/IL-23 p40, TRAIL/TNFSF10, CCL11/Eotaxin, IL-18/IL-1F4 were measured. The means of triplicate values of treatment samples were compared with untreated samples by t-test. The cytokine profiles of different treated groups were clustered and depicted with R 4. 2. 2.
RNA-sequencing
SU-DIPG-13 and SU-DIPG-25 cells were seeded in 6-well plates at a density of 7×105 cells per well and incubated for 24 hours in culture media before treatment. All treatment conditions were collected in biological triplicate. The cells were treated with ONC201, Panobinostat or tazemetostat alone, or combination of ONC201 plus Panobinostat or ONC201 plus tazemetostat, or a triple combination of ONC201 plus Panobinostat and tazemetostat. Subsequently, the cells were collected after incubation for 12 hours and total RNA was extracted using the RNeasy Plus Mini Kit (Qiagen) according to manufacturers’ instructions and quantitated using a Nanodrop spectrophotometer. The RNA-sequencing was performed by AZENTA US Inc. Poly A selection was performed before sequencing. cDNA was synthesized. Sequencing depth was 20-30 million reads per sample. Standard RNA analysis package includes mapping, and differential gene expression. Using DESeq2, a comparison of gene expression between the groups of samples was performed. The Wald test was used to generate p-values and log2 fold changes. Genes with an adjusted p-value < 0.05 and absolute log2 fold change > 1 were called differentially expressed genes. The transcriptional profile of ONC201, tazemetostat and Panobinostat treated DMG cells were clustered with R 4. 2. 2.
Results
Imipridones downregulate the expression of EZH1 and EZH2 in a panel of human cancer cell lines across tissue origins
To investigate the role of EZH1/2 in the mechanism of action of anti-cancer imipridone therapeutic agents, we performed immunoblotting for EZH1 and EZH2 after treating multiple human cancer cell lines including prostate cancer, PDAC, CRC, gastric cancer, SCLC, breast cancer, DMG and GBM cells with ONC201 or ONC206 or ONC212 for 48 hours. Both EZH1 and EZH2 were downregulated by ONC201 or ONC206 or ONC212 (Figures 1 and S1). EZH1 and EZH2 are histone methyltransferases. They catalyze the addition of methyl groups on histone H3 at lysine. The decrease of EZH1 and EZH2 by imipridones brought up the question of whether the impridones can reduce methylation levels of H3K27. Thus, we tested the trimethylation level in MCF7 cells treated with ONC201. EZH1 and EZH2 proteins were reduced as expected and trimethylation of H3K27 can’t be reduced by ONC201 (Figure 1F).
Figure 1.
ONC201, ONC206 and ONC212 reduce EZH1 and EZH2 protein levels in tumor cells, and ONC201 can’t reduce H3K27 trimethylation as tazemtostat does. Immunoblotting of EZH1 and EZH2 in (A) LNCaP prostate cancer, (B) MDA-MB-468 TNBC, (C) H1048 SCLC, (D) U251 GBM and (E) SU-DIPG-29 DMG cells upon the treatment with ONC201, ONC206 or ONC212. (F) Immunoblotting of EZH2 and H3K27me3 in MCF7 cells treated with ONC201 or tazemetostat for 72 hours showed that ONC201 reduce EZH2 but can’t reduce H3K27me3, whereas tazemetostat reduce H3K27me3.
Given that imipridones induce the integrated stress response (ISR) in tumor cells with ATF4 upregulation as the marker and reduce global protein production, we suspected that the decrease of EZH1/2 may be attributable to the ISR. Thus, we treated U251 GBM cells with the combination of ISRIB and ONC201 or ONC206 to rescue the decrease of EZH1/2 by inhibiting ISR. However, the ISRIB could not rescue the decrease of EZH1/2 induced by ONC201 or ONC206 (Figure 2). Thus, the decrease of EZH1/2 expression induced by imipridones cannot be attributed to the ISR.
Figure 2.
ISRIB does not rescue reduction of EZH1/2 following imipridone treatment. Immunoblotting of EZH1/2, ATF4 and CHOP in U251 cells upon treatment of ONC201, ISRIB or the combination of ONC201 and ISRIB for (A) 24 hours and (B) 48 hours, and ONC206, ISRIB or the combination of ONC206 and ISRIB for (C) 24 hours and (D) 48 hours.
We assessed the suppression of cell viability by ONC201, ONC206, or ONC212 alone in the panel of tumor cells by CellTiterGlo assay after treating the cells with ONC201, ONC206, or ONC212 for 72 hours (Figure S2) and picked cell viability at doses with treatment of ONC201, ONC206 or ONC212 for each cell line (Table S1) to perform linear regression. The linear regression showed that the cell viability suppression is correlated with EZH1 reduction and showed a trend to correlate with EZH2 reduction (Figure 3). In the assessment of EZH1 and EZH2 mRNA levels in MDA-MB-468 cells treated with ONC201 for 24 hours, ONC201 did not appear to reduce transcription of EZH1 or EZH2 (Figure S1O), while it decreased EZH1 and EZH2 protein expression in MDA-MB-468 (Figure 1B). RNA-sequencing of SU-DIPG-13 and SU-DIPG-25 cell lines treated with ONC201 showed no alteration of transcription of EZH1 or EZH2 (Figure S1P). Thus, EZH1 and EZH2 are downregulated by imipridones at the protein level.
Figure 3.
Cell viability suppression induced by imipridones correlates with EZH1 reduction and shows a trend to correlate with EZH2 reduction in tumor cells. Cell viability with treatment of certain doses of ONC201, ONC206 or ONC212 was obtained from dose-response relationship curves shown in Figure S2. EZH1 and EZH2 protein levels from immunoblotting shown in Figure 1 normalized with loading control. The cell lines, doses of ONC201, ONC206 or ONC212, the normalized EZH1 or EZH2 protein level and the cell viability are displayed in Table S1.
H3K27 acetylation and reduction of EZH1 and EZH2 correlate with synergy when imipridones are combined with EZH1/2 inhibitors or HDAC inhibitors
EZH1/2 inhibition downregulates methylation of H3K27 and thereby leaves the H3K27 accessible to acetylation. EZH1 and EZH2 inhibition by imipridones provides a rationale for combining imipridones with EZH1/2 inhibitors and the addition of HDAC inhibitors in the combination of imipridones plus EZH1/2 inhibitors in the treatment of tumors.
We treated DMG cells with imipridones, tazemetostat, or HDAC inhibitors alone or combinations of imipridones plus tazemetostat, imipridones plus HDAC inhibitors, or the triple combination of imipridones plus tazemetostat and HDAC inhibitor, assessed the cell viability by CellTiterGlo assay after treatment, and quantified the synergy with combination index (CI). CI < 1 indicates synergy, and the lower the CI, the more potent the synergy. We observed synergies in DMG (Figures 4A-D, S3A-G; Table S2) GBM (Figure S3H-K; Table S2), prostate cancer (Figures 4E-G, S3L-N; Table S2), liver cancer (Figure S3O; Table S2) and SCLC (Figure S3P-S; Table S2) cells. The cell line experiments revealed synergies with the combinatorial treatments and the synergies are listed in Table 1. Immunoblotting of cleaved-PARP demonstrated more apoptosis with a combination of valemetostat plus ONC201, ONC206, or ONC201 (Figure 4H-J) in 22Rv1 prostate cancer cells, combination of ONC206 plus valemetostat in H1048 SCLC cells (Figure 4K), combination of ONC206 plus Panobinostat and tazemetostat in SU-DIPG-25 DMG cells (Figure S3T), combination of ONC201 plus valemetostat in LNCaP prostate cancer cells (Figure S3U).
Figure 4.
Combination of Imipridones with EZH inhibitors or imipridones with EZH2 inhibitor and HDAC inhibitors synergizes in suppressing cell viability and inducing apoptosis in tumor cells. Cell viability of SU-DIPG-13 cells by CellTiterGlo assay and the plots of the combination index (CI) as a function of cell viability upon 72H treatment with combination of (A) ONC201 plus tazemetostat or valemetostat, (B) ONC201 plus tazemetostat and vorinostat, (C) ONC201 plus Panobinostat and tazemetostat, (D) ONC206 plus Panobinostat and tazemetostat. Cell viability of 22Rv1 cells by CellTiterGlo assay and the plots of the CI as a function of cell viability upon 72 H treatment with combination of (E) ONC201 plus valemetostat, (F) ONC206 plus valemetostat, (G) ONC212 plus valemetostat. Immunoblotting of cleaved-PARP in 22Rv1 cells upon 48 H treatment of combination of (H) ONC201 plus valemetostat, (I) ONC206 plus valemetostat and (J) ONC212 plus valemetostat, in H1048 SCLC cells (K) treated with combination of ONC206 plus valemetostat for 48 H. Drug doses are as indicated.
Table 1.
Synergies of the combinations of imipridones with EZH inhibitors with or without HDAC inhibitors
| Cell lines | Combinatorial treatment with synergies |
|---|---|
| SU-DIPG-13 | ONC201 + tazemetostat (Figure 4A; Table S2) |
| ONC201 + tazemetostat + vorinostat (Figure 4B; Table S2) | |
| ONC201 + tazemetostat + panobinostat (Figure 4C; Table S2) | |
| ONC206 + tazemetostat + Panobinostat (Figure 4D; Table S2) | |
| SU-DIPG-4 | ONC201 + tazemetostat (Figure S3A; Table S2) |
| ONC201 + valemetostat (Figure S3A; Table S2) | |
| ONC201 + tazemetostat + vorinostat (Figure S3B; Table S2) | |
| SU-DIPG-25 | ONC201 + tazemetostat (Figure S3C; Table S2) |
| ONC201 + tazemetostat + vorinostat (Figure S3D; Table S2) | |
| ONC201 + tazemetostat + Panobinostat (Figure S3E; Table S2) | |
| ONC206 + tazemetostat + Panobinostat (Figure S3F; Table S2) | |
| SU-DIPG-29 | ONC201 + tazemetostat (Figure S3G; Table S2) |
| ONC201 + valemetostat (Figure S3G; Table S2) | |
| U251 | ONC201 plus valemetostat (Figure S3H; Table S2) |
| SNB19 | ONC201 + tazemetostat (Figure S3I; Table S2) |
| ONC201+ valemetostat (Figure S3J; Table S2) | |
| ONC206+ valemetostat (Figure S3K; Table S2) | |
| 22Rv1 | ONC201+ valemetostat (Figure 4E; Table S2) |
| ONC206+ valemetostat (Figure 4F; Table S2) | |
| ONC212+ valemetostat (Figure 4G; Table S2) | |
| LNCaP | ONC201 + tazemetostat (Figure S3L; Table S2) |
| ONC201+ valemetostat (Figure S3L; Table S2) | |
| ONC206+ valemetostat (Figure S3M; Table S2) | |
| ONC212+ valemetostat (Figure S3N; Table S2) | |
| Hep3B | ONC201 + tazemetostat (Figure S3O; Table S2) |
| ONC201+ valemetostat (Figure S3O; Table S2) | |
| H1882 | ONC201 + tazemetostat (Figure S3P; Table S2) |
| ONC201+ valemetostat (Figure S3P; Table S2) | |
| H1048 | ONC201 + tazemetostat (Figure S3Q; Table S2) |
| ONC201+ valemetostat (Figure S3R; Table S2) | |
| ONC206+ valemetostat (Figure S3S; Table S2) |
We previously reported the synergistic cell death noted with an increase in PARP cleavage following treatment with ONC201 + vorinostat or ONC201 + tazemetostat or the triple combination of ONC201 + vorinostat + tazemetostat in SU-DIPG-13, SU-DIPG-25 and SU-DIPG-29 cells, and an increase in H3K27 acetylation with combinations containing vorinostat in SU-DIPG-25 and SU-DIPG-29 cells [15].
In the present study, we observed an increase in PARP cleavage in SU-DIPG-25 treated with a combination of ONC206 plus tazematostat and panobinostat and an increase in H3K27 acetylation with combinations containing Panobinostat (Figure S3T). We observed reduction of EZH1 and EZH2 by imipridones in all these cell lines (Figures 1 and S1). Thus, H3K27 acetylation and reduction of EZH1 and EZH2 correlate with synergy when imipridones are combined with EZH inhibitors or HDAC inhibitors. SU-DIPG-4 and Hep3B are the only two cell lines in which EZH1/2 dual inhibitor valemetostat demonstrated more synergy than EHZ2 inhibitor tazemetostat when combined with ONC201 (Figure S3A and S3O).
In SU-DIPG-29 and LNCaP cells, the synergy of tazemetostat combined with ONC201 was comparable to that of valemetostat combined with ONC201 (Figure S3G and S3I). In SU-DIPG-13 and H1882 cells, synergy with tazemetostat plus ONC201 is more than that of valemetostat plus ONC201 (Figures 3A and S3P). Thus, it is not a universal phenomenon that dual inhibition of both EZH1 and EZH2 leads to more synergy than an EZH2-only inhibitor when they are combined with ONC201.
Imipridones share similar targets and actions in tumor cells with EZH2 inhibitors
We performed cytokine profiling assays to investigate the impact of imipridones, EHZ2i or HDACi alone or a combination of imipridones plus an EZH2 inhibitor or an HDAC inhibitor or the triple combination of imipridone plus EZH2 inhibitor and HDAC inhibitor on the tumor microenvironment and tumor immunity. We treated DMG, GBM and HCC cells with the drugs alone or the combinations and performed cytokine profiling assays using the cell culture supernatants. Hierarchical clustering showed that ONC201 or ONC206 induce similar cytokine profile changes with tazemetostat in T98G (Figure 5A) and U251 (Figure S4A) GBM, SU-DIPG-29 DMG (Figure 5B) and Hep3B HCC (Figure S4B) cells. Thus, based on these experiments, imipridones and tazemetostat would appear to have similar effects on the tumor microenvironment and anti-tumor immunity.
Figure 5.
ONC201 or ONC206 have similar effects as EZH2 inhibitor on cytokine profile and transcriptional profile in tumor cells. (A, B) Cytokine profile assay of (A) T98G GBM cells upon treatment with ONC201, tazemetostat or Panobinostat alone or the combinations indicated, and (B) SU-DIPG-29 DMG cells upon the treatment with ONC206, tazemetostat or Panobinostat alone or the combinations indicated. (C) Clustering heatmap of expression profile of SU-DIPG-13 cells treated with ONC201, tazemetostat or Panobinostat alone by plotting their log2FoldChange transformed expression values. (D) Cluster heatmap of mRNA expression profiles of differentially expressed genes with adjusted p-value < 0.05 in SU-DIPG-13 cells treated with ONC201, tazemetostat or Panobinostat alone by plotting their log2FoldChange transformed expression values. (E, F) The principal component analysis revealing the similarities between untreated SU-DIPG-13 cells and SU-DIPG-13 cells treated with ONC201 (E) or tazemetostat (F). (G-I) Bi-clustering heatmaps indicating the expression profile of the top 30 differentially expressed genes sorted by their adjusted p-value by plotting their log2 transformed expression values in SU-DIPG-13 cells treated by drugs indicated in the plot. Each group was conducted with triplicates.
Based on our finding that imipridones reduce EZH1 and EZH2 in tumor cells, we hypothesized that imipridones have similar targets and actions with EZH1/2 inhibitors in tumor cells. We further investigated transcriptional profiles of ONC201- and tazemetostat-treated DMG cells to test the hypothesis. We treated SU-DIPG-13 and SU-DIPG-25 cells with ONC201, tazemetostat or Panobinostat and performed RNA-seq. Hierarchical clustering of the transcriptional profiles showed the similarity of ONC201 and tazemetostat treatment of SU-DIPG-13 (Figure 5C) and SU-DIPG-25 cells (Figure S4C).
Cluster heatmap of mRNA expression profiles of differentially expressed genes with adjusted P value < 0.05 in SU-DIPG-13 and SU-DIPG-25 cells treated with ONC201, tazemetostat or Panobinostat alone also showed similarities between ONC201 and tazemetostat treated cells (Figures 5D, S4D). The principal component analysis showed that the expression profiles of cells treated with ONC201 or tazemetostat were well differentiated from the control (Figures 5E, 5F, S4E and S4F).
The overlaps of the top differentially expressed genes further showed the similarities of mRNA expression profile alterations between ONC201 and tazemetostat treated cells. We observed an overlap of eight genes among the top 30 differentially expressed genes sorted by adjusted p-value in ONC201 or tazemetostat treated SU-DIPG-13 cells (Figure 5G, 5H and Table 2). The 8 shared genes are ALDH1B1, ARHGEF2, ATF3, ATF4, CHAC1, DDIT3, DDR2 and SARS (Table 2). By contrast, there was no overlap between the top 30 differentially expressed genes of cells treated with ONC201 or Panobinostat (Figure 5G, 5I and Table 2).
Table 2.
The top 30 differentially expressed genes in tazemetostat, ONC201 or Panobinostat treated SU-DIPG-13 cells
| Top 30 differentially expressed genes in tazemetostat treated SU-DIPG-13 cells | Top 30 differentially expressed genes in ONC201 treated SU-DIPG-13 cells | Top 30 differentially expressed genes in panobinostat treated SU-DIPG-13 cells |
|---|---|---|
| ACAT1 | ALAS1 | ABCA7 |
| ADM2 | ALDH1B1 | AHNAK2 |
| ALDH1B1 | ANKRD52 | APLP1 |
| ARHGEF2 | ARHGEF2 | CADM3 |
| ARRDC3 | ATF3 | CILP2 |
| ASNS | ATF4 | CLMP |
| ATF3 | BBC3 | COL1A1 |
| ATF4 | CARS | ELOVL6 |
| ATP5F1A | CEBPG | EML2 |
| CCDC9B | CHAC1 | GPR17 |
| CCND1 | DDIT3 | H1F0 |
| CHAC1 | DDR2 | HLA-E |
| CLPB | DGCR8 | IL16 |
| DDIT3 | FASTKD5 | INAFM1 |
| DDR2 | GAB2 | KCNC3 |
| EIF1 | JDP2 | KCNN3 |
| GOT2 | KLF15 | KIRREL2 |
| MRPS28 | KLF4 | NFKB2 |
| NDUFS1 | MAT2A | NMNAT2 |
| NFS1 | MT-ND5 | NXPH4 |
| NUPR1 | MT-ND6 | PBXIP1 |
| PDHA1 | MT-TC | PGPEP1 |
| SARS | MT-TN | PLPPR2 |
| SDHA | MT-TP | PRRX1 |
| SESN2 | MT-TY | REEP6 |
| SLC7A11 | SARS | RUNDC3A |
| TAGLN2 | SLC1A5 | SEZ6L2 |
| TNFRSF12A | SLC6A9 | SLC8A2 |
| TXNRD2 | SRRT | TJP3 |
| ULBP1 | TFRC | TTC9B |
There was an overlap of ten genes among the top 30 differentially expressed genes in ONC201 or tazemetostat treated SU-DIPG-25 cells (Figure S4G, S4H and Table S3). The 10 shared genes are ASNS, ATF4, DDR2, INHBE, SLC7A11, STC2, TUBE1, ULBP1, VEGFA and VLDLR (Table S3). There was no overlap between the top 30 differentially expressed genes of SU-DIPG-25 cells treated with ONC201 or Panobinostat (Figure S4G, S4I and Table S3).
Based on the significant overlaps of the top differentially expressed genes, we conclude that ONC201 which reduces EZH1 and EZH2 proteins share similar targets and actions with EZH2 inhibitors on tumor cells.
Discussion
TIC10/ONC201 was discovered as a TRAIL-inducing compound [1-3]. The tumoricidal effect of ONC201 and its potent analogs ONC206 and ONC212 on cancer cells was demonstrated to be related to the induction of the integrated stress response marked by upregulation of ATF4 [4] as a consequence of activation of mitochondrial caseinolytic protease ClpP [5,6]. Clinical data showed that a subset of DMG patients benefit from ONC201 [8,9]. Given that more than 80% of DG patients carry the driver mutation H3K27M that impairs the function of EHZ2 methylating H3K27 [10], we hypothesized that H3K27M mutation sensitization of tumors to ONC201 may be mimicked by EZH inhibitors in inhibiting methylation of H3K27. We observed synergy of imipridones plus EZH2 inhibitor, tazemetostat or dual EZH1/2 inhibitor, valemetostat in a panel of tumors from different tissue origins. Thus, inhibition of H3K27 methylation makes the tumor cells more sensitive to imipridones as we hypothesized.
Our current results suggest that EZH1/2 may play a role in the action of imipridones and ONC201. In our studies, EZH1 and EZH2 protein expression is reduced by imipridones in all the tested tumor cell lines and the suppression of cell viability by imipridones correlates with the EZH1 and EZH2 reduction. Thus, it’s reasonable to raise the question of whether ONC201 can reduce H3K27 trimethylation by decreasing EZH1 and EZH2 proteins. Our results showed that H3K27 methylation function of EZH2 was not reduced by ONC201, so the residual EZH proteins still fully conduct their histone methyltransferase function for the cells.
Cytokine profiles show similarity between ONC201 or ONC206 and tazemetostat. RNA-sequencing shows the overlap of the top differentially expressed genes in ONC201 and tazemetostat treated DMG cells whereas there was no overlap between ONC201 and Panobinostat treated cells. Therefore, ONC201 or ONC206 and EZH2 inhibitors share similar targets and actions on tumor cells, and the similarities are not dependent on the reduction of H3K27 methylation which can’t be achieved by ONC201 treatment.
The findings of the action of imipridones on EZH1/2 provide new clues for elucidating the tumoricidal effect of imipridones with links to H3K27M and epigenetic alteration of gene expression and cytokine profiles.
Acknowledgements
W.S.E-D. is an American Cancer Society Research Professor and is supported by the Mencoff Family University Professorship at Brown University. This work was supported by an NIH grant (CA173453) to W.S.E-D. This work was presented in part at the 2023 meeting of the American Association for Cancer Research.
Disclosure of conflict of interest
W.S.E-D. is a co-founder of Oncoceutics, Inc., a subsidiary of Chimerix. Dr. El-Deiry has disclosed his relationship with Oncoceutics/Chimerix and potential conflict of interest to his academic institution/employer and is fully compliant with NIH and institutional policy that is managing this potential conflict of interest.
Supporting Information
References
- 1.Allen JE, Kline CL, Prabhu VV, Wagner J, Ishizawa J, Madhukar N, Lev A, Baumeister M, Zhou L, Lulla A, Stogniew M, Schalop L, Benes C, Kaufman HL, Pottorf RS, Nallaganchu BR, Olson GL, Al-Mulla F, Duvic M, Wu GS, Dicker DT, Talekar MK, Lim B, Elemento O, Oster W, Bertino J, Flaherty K, Wang ML, Borthakur G, Andreeff M, Stein M, El-Deiry WS. Discovery and clinical introduction of first-in-class imipridone ONC201. Oncotarget. 2016;7:74380–74392. doi: 10.18632/oncotarget.11814. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Allen JE, Krigsfeld G, Mayes PA, Patel L, Dicker DT, Patel AS, Dolloff NG, Messaris E, Scata KA, Wang W, Zhou JY, Wu GS, El-Deiry WS. Dual inactivation of Akt and ERK by TIC10 signals Foxo3a nuclear translocation, TRAIL gene induction, and potent antitumor effects. Sci Transl Med. 2013;5:171ra117. doi: 10.1126/scitranslmed.3004828. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Allen JE, Krigsfeld G, Patel L, Mayes PA, Dicker DT, Wu GS, El-Deiry WS. Identification of TRAIL-inducing compounds highlights small molecule ONC201/TIC10 as a unique anti-cancer agent that activates the TRAIL pathway. Mol Cancer. 2015;14:99. doi: 10.1186/s12943-015-0346-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Kline CL, Van den Heuvel AP, Allen JE, Prabhu VV, Dicker DT, El-Deiry WS. ONC201 kills solid tumor cells by triggering an integrated stress response dependent on ATF4 activation by specific eIF2α kinases. Sci Signal. 2016;9:ra18. doi: 10.1126/scisignal.aac4374. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Ishizawa J, Zarabi SF, Davis RE, Halgas O, Nii T, Jitkova Y, Zhao R, St-Germain J, Heese LE, Egan G, Ruvolo VR, Barghout SH, Nishida Y, Hurren R, Ma W, Gronda M, Link T, Wong K, Mabanglo M, Kojima K, Borthakur G, MacLean N, Ma MCJ, Leber AB, Minden MD, Houry W, Kantarjian H, Stogniew M, Raught B, Pai EF, Schimmer AD, Andreeff M. Mitochondrial ClpP-mediated proteolysis induces selective cancer cell lethality. Cancer Cell. 2019;35:721–737. e729. doi: 10.1016/j.ccell.2019.03.014. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Ferrarini I, Louie A, Zhou L, El-Deiry WS. ONC212 is a novel mitocan acting synergistically with glycolysis inhibition in pancreatic cancer. Mol Cancer Ther. 2021;20:1572–1583. doi: 10.1158/1535-7163.MCT-20-0962. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Chan KM, Fang D, Gan H, Hashizume R, Yu C, Schroeder M, Gupta N, Mueller S, James CD, Jenkins R, Sarkaria J, Zhang Z. The histone H3.3K27M mutation in pediatric glioma reprograms H3K27 methylation and gene expression. Genes Dev. 2013;27:985–990. doi: 10.1101/gad.217778.113. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Arrillaga-Romany I, Odia Y, Prabhu VV, Tarapore RS, Merdinger K, Stogniew M, Oster W, Allen JE, Mehta M, Batchelor TT, Wen PY. Biological activity of weekly ONC201 in adult recurrent glioblastoma patients. Neuro Oncol. 2020;22:94–102. doi: 10.1093/neuonc/noz164. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Hall MD, Odia Y, Allen JE, Tarapore R, Khatib Z, Niazi TN, Daghistani D, Schalop L, Chi AS, Oster W, Mehta MP. First clinical experience with DRD2/3 antagonist ONC201 in H3 K27M-mutant pediatric diffuse intrinsic pontine glioma: a case report. J Neurosurg Pediatr. 2019;23:719–725. doi: 10.3171/2019.2.PEDS18480. [DOI] [PubMed] [Google Scholar]
- 10.Harutyunyan AS, Krug B, Chen H, Papillon-Cavanagh S, Zeinieh M, De Jay N, Deshmukh S, Chen CCL, Belle J, Mikael LG, Marchione DM, Li R, Nikbakht H, Hu B, Cagnone G, Cheung WA, Mohammadnia A, Bechet D, Faury D, McConechy MK, Pathania M, Jain SU, Ellezam B, Weil AG, Montpetit A, Salomoni P, Pastinen T, Lu C, Lewis PW, Garcia BA, Kleinman CL, Jabado N, Majewski J. H3K27M induces defective chromatin spread of PRC2-mediated repressive H3K27me2/me3 and is essential for glioma tumorigenesis. Nat Commun. 2019;10:1262. doi: 10.1038/s41467-019-09140-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Mohammad F, Weissmann S, Leblanc B, Pandey DP, Højfeldt JW, Comet I, Zheng C, Johansen JV, Rapin N, Porse BT, Tvardovskiy A, Jensen ON, Olaciregui NG, Lavarino C, Suñol M, de Torres C, Mora J, Carcaboso AM, Helin K. EZH2 is a potential therapeutic target for H3K27M-mutant pediatric gliomas. Nat Med. 2017;23:483–492. doi: 10.1038/nm.4293. [DOI] [PubMed] [Google Scholar]
- 12.Fujita S, Honma D, Adachi N, Araki K, Takamatsu E, Katsumoto T, Yamagata K, Akashi K, Aoyama K, Iwama A, Kitabayashi I. Dual inhibition of EZH1/2 breaks the quiescence of leukemia stem cells in acute myeloid leukemia. Leukemia. 2018;32:855–864. doi: 10.1038/leu.2017.300. [DOI] [PubMed] [Google Scholar]
- 13.Ito J, Yamagata K, Shinohara H, Shima Y, Katsumoto T, Aikawa Y, Kitabayashi I. Dual inhibition of EZH1/2 induces cell cycle arrest of B cell acute lymphoblastic leukemia cells through upregulation of CDKN1C and TP53INP1. Int J Hematol. 2023;117:78–89. doi: 10.1007/s12185-022-03469-8. [DOI] [PubMed] [Google Scholar]
- 14.Honma D, Kanno O, Watanabe J, Kinoshita J, Hirasawa M, Nosaka E, Shiroishi M, Takizawa T, Yasumatsu I, Horiuchi T, Nakao A, Suzuki K, Yamasaki T, Nakajima K, Hayakawa M, Yamazaki T, Yadav AS, Adachi N. Novel orally bioavailable EZH1/2 dual inhibitors with greater antitumor efficacy than an EZH2 selective inhibitor. Cancer Sci. 2017;108:2069–2078. doi: 10.1111/cas.13326. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Zhang Y, Zhou L, Safran H, Borsuk R, Lulla R, Tapinos N, Seyhan AA, El-Deiry WS. EZH2i EPZ-6438 and HDACi vorinostat synergize with ONC201/TIC10 to activate integrated stress response, DR5, reduce H3K27 methylation, ClpX and promote apoptosis of multiple tumor types including DIPG. Neoplasia. 2021;23:792–810. doi: 10.1016/j.neo.2021.06.007. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.