Oncogenic KRAS Sensitizes Lung Adenocarcinoma to GSK-J4–Induced Metabolic and Oxidative Stress
Beom-Jin Hong1, Woo-Yong Park2, Hwa-Ryeon Kim3, Jin Woo Moon1, Ho Yeon Lee4, Jun Hyung Park1, Seon-Kyu Kim5, Youngbin Oh1, Jae-Seok Roe3, and Mi-Young Kim1,6
Abstract
Research
Genetic and epigenetic changes (e.g., histone methylation) contribute to cancer development and progression, but our understanding of whether and how specific mutations affect a cancer’s sensitivity to histonedemethylase (KDM)inhibitors is limited. Here, we evaluated the effects of a panel of KDM inhibitors on lung adenocarcinomas (LuAC) with various mutations. Notably, LuAC lines harboring KRAS mutations showed hypersensitivity to the histone H3K27 demethylase inhibitor GSK-J4. Specifically, GSK-J4 treatment of KRAS mutant–containing LuAC downregulated cell-cycle progres- sion genes with increased H3K27me3. In addition, GSK-J4 upregulated expression of genes involved in glutamine/
glutamate transport and metabolism. In line with this, GSK- J4 reduced cellular levels of glutamate, a key source of the TCA cycle intermediate a-ketoglutarate (aKG) and of the antiox- idant glutathione, leading to reduced cell viability. Supple- mentation with an aKG analogue or glutathione protected
KRAS-mutant LuAC cells from GSK-J4–mediated reductions in viability, suggesting GSK-J4 exerts its anticancer effects by inducing metabolic and oxidative stress. Importantly, KRAS knockdown in mutant LuAC lines prevented GSK-J4–induced decrease in glutamate levels and reduced their susceptibility to GSK-J4, whereas overexpression of oncogenic KRAS in wild- type LuAC lines sensitized them to GSK-J4. Collectively, our study uncovers a novel association between a genetic mutation and KDM inhibitor sensitivity and identifi es the underlying mechanisms. This suggests GSK-J4 as a potential treatment option for cancer patients with KRAS mutations.
Signifi cance: This study not only provides a novel asso- ciation between KRAS mutation and GSK-J4 sensitivity but also demonstrates the underlying mechanisms, suggesting a potential use of GSK-J4 in cancer patients with KRAS mutations.
Introduction
Lung adenocarcinoma (LuAC) is the most common subtype of non–small cell lung carcinoma (NSCLC; ref. 1). The 5-year survival rate for LuAC is about 18%, owing mainly to a lack of effective treatment options. Current treatment options include
surgery, chemo- and radiotherapies, and targeted therapies against EGFR and ALK (anaplastic lymphoma kinase) mutant– containing LuAC (1, 2). Although several strategies to target LuAC with KRAS mutations, the most prevalent type of mutation in LuAC (3, 4), have been suggested (5–7), no effective treatments for this subtype are currently available, necessitating novel approaches.
1Department of Biological Sciences, Korea Advanced Institute of Science and Technology (KAIST), Daejeon, South Korea. 2Laboratory of Cancer Biology and Genetics, National Cancer Institute, Bethesda, Maryland. 3Department of Bio- chemistry, College of Life Science and Biotechnology, Yonsei University, Seoul, Korea. 4SK Biopharmaceuticals, Seoul, Korea. 5Personalized Genomic Medicine Research Center, Korea Research Institute of Bioscience & Biotechnology (KRIBB), Daejeon, South Korea. 6KAIST Institute for the BioCentury, Cancer Metastasis Control Center, Daejeon, South Korea.
Note: Supplementary data for this article are available at Cancer Research Online (http://cancerres.aacrjournals.org/).
B.-J. Hong and W.-Y. Park are co-first authors. H.-R. Kim, J.W. Moon, and H.Y. Lee are co-second authors.
Corresponding Authors: Mi-Young Kim, Korea Advanced Institute of Science and Technology, 291 Gwahakro, Daejeon, 34141, Republic of Korea. Phone: 824- 2350-2615; Fax: 824-2350-2610; E-mail: [email protected]; and Jae-Seok Roe, Department of Biochemistry, Yonsei University, 50 Yonsei-ro, Seodaemun- gu, Seoul, 03722, Republic of Korea. Phone: 822-2123-2700; E-mail: [email protected]
Cancer Res 2019;79:5849–59
doi: 10.1158/0008-5472.CAN-18-3511
ti2019 American Association for Cancer Research.
In addition to genetic mutations, epigenetic deregulation also contributes to LuAC (8–10). Thus, epigenetic inhibitors have been explored as alternative anticancer drugs for LuAC. For example, inhibitors against DNA methyltransferases and histone deacetylases areusedastherapeuticagentsforseveralcancers (11). In the case of LuAC, therapeutic strategies combining these epigenetic inhibitors with EGFR inhibitors or other chemother- apeutic agents are currently in clinical trials (11).
Furthermore, inhibitors targeting histone lysine methyltrans- ferases (KMT) and demethylases (KDM) are being tested as cancer therapeutics. For example, inhibitors against H3K27 methyltrans- ferase EZH2 are under clinical trial for solid tumors (12). In lung cancer, EZH2 inhibition sensitizes BRG1 and EGFR mutant NSCLC to TopoII inhibitors in a pre-clinical model (13). In addition, inhibitors targeting H3K4 demethylase LSD1 and H3K79 methyltransferase DOT1L are being evaluated for leuke- mia (12, 14). Inhibitors targeting KDM5 and KDM6, demethy- lases for H3K4 and H3K27, respectively, have even been tested in some preclinical cancer models (15, 16).
The importance of targeting KMTs and KDMs in cancer is also supported by an interplay between these enzymes and
www.aacrjournals.org 5849
Hong et al.
metabolism (17). Cancer cells often undergo oncogene-induced metabolic rewiring, which then leads to changes in cellular levels of a number of metabolites (17, 18). Because KMTs and KDMs use metabolic intermediates as cofactors [e.g., SAM for EZH2 and a-ketoglutarate (aKG) for Jumonji C-domain–containing (JMJD) enzymes], changes in cellular levels of these cofactors can alter the activities of KMTs and KDMs in cancer cells (17, 18). Reciprocally, these enzymes can regulate the expression of met- abolic genes (19–21). For example, EZH2 reprograms metabo- lism in glioblastoma and leukemia by regulating metabolic genes (19, 20). In addition, KDM6B regulates genes involved in b-oxidation and anabolism in hepatocytes, although their roles in cancer metabolism are unknown (21).
Despite the recent development of KMT and KDM inhibitors, their effects and mechanisms of action remain unclear. This emphasizes the need for an in-depth investigation of the associ- ation between genetic mutations, their underlying mechanisms, and their susceptibility to epigenetic modification.
Here, we report a novel association between oncogenic KRAS and sensitivity to GSK-J4 in LuAC and the underlying mechanism.
Materials and Methods
Cell culture
The human LuAC and pancreatic ductal adenocarcinoma (PDAC) cell lines were cultured in RPMI1640 containing 10% FBS. WI38 and IMR90 were cultured in MEM containing 10% FBS.
HCC827, H23, H2009, and SNU-324 were purchased from Korean cell line bank. H1975, A549, MIA-Paca-2, and PANC-1 were originally obtained from the ATCC. All cell lines were recently authenticated by DNA fi ngerprinting analysis and regu- larly tested for Mycoplasma contamination by e-Myco plus Myco- plasma PCR Detection Kit. All cell lines were used within 15 passages.
RNAi-mediated gene knockdown
SiRNA were synthesized from Bioneer. ON-TARGETplus SMARTpool siRNA targeting for KRAS were purchased from Dharmacon. Cells were transfected with siRNAs using Lipofecta- mine 2000. siRNA sequences used in this study are summarized in Supplementary Materials and Methods.
Cell viability assay
Viabilities of LuAC cells treated with KDM inhibitors or 50 nmol/L siRNAs against KDM6A and B were measured by using CellTiter-Glo Assay Kit (Promega). For analysis of GSK-J4 sensi- tivity upon overexpression and knockdown of KRAS or EGFR and upon dimethyl-2-oxoglutarate (D2OG) treatment or Vit E, see Supplementary Methods.
Colony formation assay
Cells were treated with GSK-J4 for 3 days and the media were replaced with GSK-J4–free media. After 3 days, cells were stained with 0.2% crystal violet and intensities of crystal violet staining were quantifi ed by Image J software (NIH, Bethesda, MD).
Animal studies
All animal studies were conducted in accordance with a protocol approved by the Institutional Animal Care and Use Committee at KAIST. LuAC cells were injected into the lower fl anks of 6- to 7-week-old BALB/c nude mice (OrientBio). When tumor volume reached approximately 70 mm3, the mice
were treated with DMSO or GSK-J4 (100 mg/kg) via intraper- itoneal injection every two days for 5 times. Tumor volume was calculated as width2 ti length ti 0.5.
Cell-cycle and apoptosis analysis
Cells were treated with 10 mmol/L GSK-J4 for 24 hours and stained with 30 mg/mL 7-AAD (Sigma Aldrich) or Annexin V – Alexa 488 (Life Technologies) and 7-AAD solution (BD) for 30 minutes. Flow cytometry was conducted using BD LSRFortessa (BD). Cell-cycle distribution and percent apoptotic cells were analyzed by using FlowJo software v10 (FlowJo, LLC).
Genome-wide RNA sequencing analysis
Total RNA was extracted from LuAC cells treated with DMSO or 10 mmol/L GSK-J4 for 24 hours or siKDM6B. Libraries were prepared using the TruSeq Stranded mRNA Sample Prepa- ration Kit (Illumina) according to the manufacturer’s instructions. The resulting RNA sequencing (RNA-seq) libraries were then sequenced on an Illumina NextSeq 500 using the single-end protocol with a read length of 75 nt. For detailed information on analysis, see Supplementary Materials and Methods.
Gene set enrichment analysis
Gene set enrichment analysis (GSEA) was performed according to the instructions. Briefl y, GSK-J4 UP and GSK-J4 DN were generated on the basis of the genes defi ned in Fig. 3A. For GSEA preranked analyses of KDM6B knockdown samples, log2 of the fold changes calculated from RNA-seq analysis were provided.
Genome-wide ChIP sequencing analysis
To align chromatin immunoprecipitation (ChIP) sequencing (ChIP-seq) reads, raw reads from the fastq fi les were aligned to the reference human genome assembly hg19 using Bowtie with default settings except for allowing two-mismatches for unique alignments (i.e., ‘-v 2 -m 1’). To avoid clonal artifacts introduced in the library amplifi cation steps, duplicated mapped reads were further removed using Samtools. BAM fi les were subjected to other analyses with the HOMER suite. Peaks were identifi ed in the processed BAM fi les and ChIP-seq tags were calculated using the fi ndPeaks tool set to fi nd histone-enriched regions (i.e., “-style histone”). Detailed information is provided in Supple- mentary Materials and Methods.
Gene ontology
A total of 246 up- and 591 downregulated genes were subjected to Gene Ontology (GO) term enrichment with Metascape (http://
metascape.org; ref. 22).
Quantification of cellular glutamine, glutamate, and glutathione levels
The cellular concentrations of glutamine/glutamate and glutathione from cells treated with DMSO or 5 mmol/L GSK- J4 for 12 hours were measured by using the Glutamine/
Glutamate-Glo and GSH-Glo Glutathione Assay Kit (Promega), respectively.
Quantification of reactive oxygen species
Cells were plated on 96-well plates and treated with 5 mmol/L GSK-J4 with or without 100 mmol/L VitE for 48 hours. Reactive oxygen species (ROS) levels were quantified using ROS-Glo H2O2 Assay Kit (Promega).
5850 Cancer Res; 79(22) November 15, 2019 Cancer Research
KRAS Mutation Determines GSK-J4 Sensitivity in LuAC
Statistical analysis
Results are reported as means ti SEM and statistical significance was determined by two-tailed unpaired Student t test, unless indicated otherwise. P values of <0.05 were considered statisti- cally signifi cant.
Data availability
RNA-seq and ChIP-seq data are available in NCBI GEO: GSE133970.
Additional information is described in the Supplementary Materials and Methods.
Results
LuAC cell lines exhibit differential GSK-J4 sensitivity
To determine whether LuAC cell lines show differential sensi- tivity toward various KDM inhibitors, we analyzed the viability of a panel of LuAC cell lines following treatment with JIB-04 (a pan-JMJC inhibitor), SD70 (a KDM4C inhibitor), IOX1 [a 2-oxoglutarate (2OG)-dependent KDM inhibitor], PBIT (a KDM5 inhibitor), GSK-LSD1 (a LSD1 inhibitor), and GSK-J4 (a KDM5 and 6 subfamily inhibitor; refs. 23–26). Notably, at concentra- tions with minimal effects on the normal lung fibroblast cell lines WI38 and IMR90, only GSK-J4 treatment segregated LuAC lines into two groups in cell viability and colony formation assays: "GSK-J4 sensitive" and "GSK-J4 resistant" lines (Fig. 1A–C; Sup- plementary Fig. S1A–S1C). These differences in GSK-J4 sensitivity were recapitulated by KDM6B but not KDM6A knockdown (Fig. 1D; Supplementary Fig. S1D and S1E), suggesting KDM6B as a major GSK-J4 target. Further supporting this, GSK-J4 atten- uated tumor growth by the sensitive (H2030 and A549), but not by the resistant (PC9 and HCC827), lines (Fig. 1E and F; Sup- plementary Fig. S1F–S1H). Similar results were observed upon KDM6B knockdown (Supplementary Fig. S1I).
GSK-J4 attenuates cell-cycle progression of GSK-J4–sensitive LuAC
On the basis of differential GSK-J4 sensitivities observed in LuAC, we investigated the underlying cellular mechanisms. GSK-J4 and KDM6B knockdown induced cell-cycle arrest only in GSK-J4–sensitive lines (Fig. 2A; Supplementary Fig. S2A and S2B). GSK-J4 also increased apoptosis in the sensitive line H23 (Fig. 2B) without affecting resistant lines (Fig. 2C; Supplementary Fig. S2C and S2D). In addition, GSK-J4 did not induce apoptosis in the other sensitive lines even at a later time point (Supple- mentary Fig. S2E–S2G), which is consistent with H23 being the most sensitive to GSK-J4 (refer to Fig. 1A–C). Whether or not GSK-J4 induces apoptosis in the sensitive lines are probably dependent on differences in their intrinsic biological properties. Together, GSK-J4's antitumor effects on the sensitive lines occur via cell-cycle arrest and/or induction of apoptosis.
GSK-J4 affects specifi c transcriptional programs in sensitive LuAC
Next, we investigated the molecular mechanisms underlying the differential GSK-J4 sensitivities of LuAC. Because KDM6B knockdown mimicked GSK-J4's effect on cell viability, we first looked for a correlation between GSK-J4 sensitivity and basal levels of H3K27me3 and its modifiers EZH2, KDM6A, and KDM6B, but found none (Supplementary Fig. S3A and S3B). We hypothesized that GSK-J4 sensitivity may arise from GSK-J4–
mediated modulation of specifi c transcriptional programs in sensitive LuAC lines. To test this, we performed RNA-seq exper- iment and selected genes exhibiting GSK-J4–induced expression changes (fold change > 1.5, P < 0.05) in two sensitive lines (H23 and A549) but not the resistant line (HCC827). We defined the resulting list of 246 downregulated and 591 upregulated genes as a "GSK-J4 signature" (Fig. 3A; Supplementary Data S1). GSEA then revealed GSK-J4 signature genes show significant overlap with KDM6B-regulated genes in GSK-J4–sensitive lines (Fig. 3B; Supplementary Fig. S3C), again supporting preferential targeting of KDM6B by GSK-J4.
Most of the genes downregulated by GSK-J4 and KDM6B knockdown in sensitive lines were cell-cycle progression genes (Fig. 3C; Supplementary Data S2; Supplementary Fig. S3D). Furthermore, promoters of downregulated genes exhibited an enrichment of binding sites for cell cycle–related transcription factors (Supplementary Fig. S3E; refs. 27–30). This suggests GSK- J4 attenuates the cell cycle in sensitive LuAC by downregulating cell cycle–related genes. On the other hand, a majority of GSK-J4 upregulated genes belong to the stress-related pathways that respond to external stimuli, endoplasmic reticulum (ER) stress, and ROS (Fig. 3D; Supplementary Table S1; Supplementary Fig. S3F; Supplementary Data S3). We also observed significant enrichment for genes that function in amino acid transport and metabolism (Fig. 3D; Supplementary Data S3), which will be further discussed below.
Next, we investigated whether GSK-J4–mediated gene expres- sion changes are associated with changes in histone H3 methyl- ation. Consistent with our Western blotting results, ChIP-seq analysis revealed that GSK-J4–induced enrichment of H3K27me3 in the sensitive A549 line without affecting expression of its target KDMs (Fig. 3E; Supplementary Fig. S4A–S4C). Moreover, about half of the chromatin regions exhibiting GSK-J4–induced H3K27me3 increase were common to KDM6B knockdown including cell-cycle genes like E2F8 region (Fig. 3F and G; Sup- plementary Fig. S4D and S4E). We also observed H3K4me3 increase in a subset of GSK-J4–upregulated genes (Supplementary Fig. S4F and S4G). Thus, our data support that GSK-J4 affects several transcriptional programs in sensitive LuAC and this is associated with changes in histone methylation.
GSK-J4 induces oxidative and metabolic stress in sensitive LuAC
As described in Fig. 3D, we found GSK-J4 treatment enhanced the expression of stress-related genes. Consistent with this, the promoters of most of these genes contained binding sites for activating transcription factor 4 (ATF4), an important effector in the integrative stress response (ISR; Supplementary Fig. S5A). ATF4 is induced by oxidative and metabolic stress (31, 32) and upregulates transcription of target genes (31, 32) that lead to cell- cycle arrest and/or apoptosis (32). GSK-J4 increased ATF4 expres- sion at the mRNA and protein levels as well as eIF2a phosphor- ylation, which promotes ATF4 translation (Supplementary Fig. S5B and S5C; ref. 32). Increased ATF4 was accompanied by upregulation of ATF4 target genes (Supplementary Fig. S5D), suggesting GSK-J4 specifically activates the ATF4 pathway in sensitive LuAC lines.
On the basis of this, we hypothesized that GSK-J4–mediated ATF4 induction may contribute to an antitumor effect of GSK-J4 in sensitive LuAC. Contrary to our expectation, however, ATF4 knockdown had no effect on GSK-J4 sensitivity (Supplementary
www.aacrjournals.org Cancer Res; 79(22) November 15, 2019 5851
Hong et al.
Figure 1.
GSK-J4 treatment segregates LuAC cell lines into GSK-J4–sensitive and resistant groups. A, Relative viability of the LuAC cell lines 48 hours post GSK-J4 treatment. Cell viability was normalized to DMSO-treated cells. Resistant lines: HCC827, PC9, H1975, and HCC4006. Sensitive lines: H2030, H23, A549, and H2009. The highest P values between the resistant and sensitive groups at the indicated concentration of GSK-J4 are shown [n ¼ 7 (HCC827, H23, A549, H2030); n ¼ 6 (PC9); n ¼ 4 (H2009); n ¼ 3 (H1975, HCC4006)]. B and C, Colony formation by indicated LuAC cell lines upon GSK- J4 treatment. Cells were treated
with GSK-J4 for 3 days and grown for another 3 days in fresh media without GSK-J4. Quantifi cation of relative colony formation (B) is shown with same color codes used in A (n ¼ 3). Representative images
are shown (C). D, Comparison of cell viability 96 hours posttransfection with 50 nmol/L of siNT (nontargeting), siKDM6A, or siKDM6B (n ¼ 6 except for siKDM6B #1, n ¼ 4). E and F, Tumor growth rate of PC9 (E) and H2030 (F) upon DMSO or GSK-J4 treatment in xenograft assays. Cells were subcutaneously injected into the immunocompromised mice. When tumor sizes reached 50 to 70 mm3, mice were treated with DMSO or GSK-J4 (100 mg/kg) every 2 days for fi ve times. D0, the initial day of the drug treatment.
Arrows on the x-axis indicate the completion of drug treatment. Data, means ti SEM. ti , P < 0.05; titi , P < 0.01; titititi , P < 0.0001. n.s., nonsignifi cant.
Fig. S5E–S5H). Thus, rather than directly contributing to GSK-J4– mediated reductions in cell viability, ATF4 induction may simply refl ect the activation of the ISR pathway in sensitive lines.
On the basis of this, we searched for other pathway(s) that could contribute to GSK-J4 sensitivity. Interestingly, GSK-J4 changed the expression of genes associated with amino acid transport and metabolism and with responses to nutrient deprivation (refer to Fig. 3D; Supplementary Table S1). Further analysis of RNA-seq data revealed that GSK-J4 altered the expression of genes involved in glutamine (Gln)/glutamate (Glu) transport and metabolism only in the sensitive lines (Supplementary Fig. S6A). Basal levels of Gln and Glu showed no correlation with GSK-J4 sensitivity (Supplementary Fig. S6B). GSK-J4, however, led to accumulation and reduction of cellular Gln and Glu levels only in the sensitive lines, respectively (Fig. 4A and B). Similar effects were observed upon KDM6B knockdown (Supplementary Fig. S6C).
Glutamate is a major source for the cellular antioxidant gluta- thione as well as aKG, a major intermediate in the TCA cycle (33, 34). Thus, we hypothesized that GSK-J4–mediated reduction in glutamate may cause a subsequent decrease in glutathione and aKG, leading to increased oxidative and meta- bolic stress, respectively. Although we found no correlation between basal levels of ROS and glutathione and GSK-J4 sensi- tivity (Fig. 4C and D), GSK-J4 reduced cellular glutathione levels and increased ROS levels in the sensitive lines (Fig. 4D and E). In addition, changes in Glu/Gln and ROS levels in H23 were satu- rated at 5 mmol/L GSK-J4, while A549 showed a GSK-J4 dose- dependent response (Supplementary Fig. S6D and S6E). This is consistent with colony formation assays in which the antigrowth effect of GSK-J4 was saturated at 5 mmol/L in H23, while A549 cells showed further decrease at 10 mmol/L (refer to Fig. 1B and C). This supports a direct association between GSK-J4 sensitivity and GSK-J4–mediated induction of oxidative stress.
5852 Cancer Res; 79(22) November 15, 2019 Cancer Research
KRAS Mutation Determines GSK-J4 Sensitivity in LuAC
Figure 2.
GSK-J4 attenuates cell-cycle progression and/or induces apoptosis in GSK-J4–sensitive cell lines. A, Cell-cycle analysis of the indicated cell lines upon GSK-J4
treatment (10 mmol/L, 24 hours). Cells were stained with 7-AAD following the drug treatment and their cell- cycle distributions were analyzed by
fl ow cytometry (n ¼ 3). B and C, Apoptosis analysis in H23 (B) and H1975 (C) upon GSK-J4 treatment (10 mmol/L, 48 hours). Cells were stained with Annexin-V and 7-AAD and the percent apoptotic cells were analyzed by fl ow cytometry. Representative FACS profi les and percent apoptotic cells (bar graph) are shown (H23, n ¼ 6; H1975, n ¼ 2). Data, means ti SEM for A and B, SD for C. titi , P < 0.01. n.s., nonsignificant.
Furthermore, relieving oxidative stress by supplementation of glutathione and the ROS scavenger a-tocopherol (VitE) reduced the GSK-J4 susceptibility of the sensitive lines, albeit not completely (Fig. 4F–H). Finally, supplementation of D2OG, a cell-permeable aKG analogue, protected GSK-J4–sensitive lines from GSK-J4 and this was accompanied by a prevention of the GSK-J4–mediated induction of ATF4 (Fig. 5A–D; Supplementary Fig. S6F). This indicates that GSK-J4–induced glutamate depri- vation accounts for the activation of the ISR pathway in sensitive LuAC. Collectively, our data indicate that GSK-J4–induced reduc- tions in glutamate reduce cellular glutathione and aKG levels, which together contribute to the GSK-J4 susceptibility of sensitive LuAC cells.
An activating KRAS mutation sensitizes LuAC to GSK-J4
AlthoughourstudyidentifiedthemechanismunderlyingGSK-J4 sensitivity of LuAC, we were curious how mutations in oncogenes and/or tumor suppressors would affect GSK-J4 sensitivity. Inter- estingly, GSK-J4–sensitive and resistant lines harbored activating KRAS and EGFR mutations, respectively (SupplementaryTable S2).
Neither overexpression of EGFR in GSK-J4–sensitive lines nor EGFR knockdown in the resistant line affected their GSK-J4 sensitivity (Supplementary Fig. S7A–S7C). However, KRAS knockdown in GSK-J4–sensitive lines signifi cantly reduced their sensitivity to GSK-J4 (Fig. 6A; Supplementary Fig. S7D and S7E). As reported, the viability of KRAS mutant LuAC cells was sensitive to KRAS knockdown (35). Thus, we performed these experiments with a suboptimal KRAS knockdown that only caused a minimal reduction in cell viability. Overexpres- sion of activating KRAS mutants (KRASG12V and KRASG12C) in the resistant H1975 line increased GSK-J4 susceptibility in vitro and in vivo (Fig. 6B and C; Supplementary Fig. S7F). Further- more, KRAS knockdown in H1975 reduced GSK-J4 sensitivity at a higher concentration of GSK-J4 (Supplementary Fig. S7G), demonstrating that KRAS dependence of GSK-J4 sensitivity in isogenic cell lines. We noted that mutant KRAS-containing PDAC cell lines also showed GSK-J4–mediated increase in ROS and higher sensitivity to GSK-J4, compared with wild type– containing lines (Supplementary Fig. S7H and S7I). Further- more, KRAS knockdown protected mutant KRAS-containing
www.aacrjournals.org Cancer Res; 79(22) November 15, 2019 5853
Hong et al.
Figure 3.
GSK-J4 affects specifi c transcriptional programs in sensitive LuAC. A, Venn diagram showing genes down- (left) and upregulated (right) upon GSK-J4 treatment in sensitive (H23, A549) and resistant (HCC827) lines according to an RNA-seq analysis. Genes subjected to further gene ontology analyses are indicated by asterisks. B, GSEA plots showing enrichment of GSK-J4 signature gene sets downregulated (left) or upregulated (right) upon KDM6B knockdown in A549. NES, normalized enrichment score. C and D, Gene Ontology (GO) analysis of 246 GSK-J4-DN (C) and 591 GSK-J4-UP (D) genes in sensitive lines. The top 20 enriched biological processes ranked by their –log10(P) values. E and F, Box plots showing genome-wide H3K27me3 enrichment upon GSK-J4 (E) or siKDM6B (F) treatment. P values were obtained from Welch t test. Data, means ti Max/Min. G, Venn diagram showing common H3K27me3 ChIP-seq signals upon GSK-J4 and KDM6B knockdown.
5854 Cancer Res; 79(22) November 15, 2019 Cancer Research
KRAS Mutation Determines GSK-J4 Sensitivity in LuAC
Figure 4.
GSK-J4 induces oxidative stress in sensitive LuAC cell lines. A and B, Comparison of glutamine (A) and glutamate (B) levels in GSK-J4–sensitive and
-resistant lines upon GSK-J4 treatment [5 mmol/L, 12 hours; n ¼ 6 (A549); n ¼ 5 (H23); n ¼ 4 (PC9, H2009); n ¼ 3 (HCC4006, HCC827, H1975)]. C, Comparison of basal ROS levels in GSK-J4–resistant and -sensitive lines in the absence of GSK-J4. The ROS level in H23 was set to 1 [n ¼ 5 (H23, A549); n ¼ 4 (HCC4006, HCC827, H2030); n ¼ 3 (PC9, H1975, H2009)].
D, Basal levels (DMSO) and GSK-J4–mediated changes (GSK-J4, 5 mmol/L, 12 hours) in total glutathione levels in resistant and sensitive lines
(n ¼ 3 except for A549, n ¼ 4). E, GSK-J4 induced fold changes in ROS levels in resistant and sensitive cell lines (5 mmol/L, 48 hours). ROS levels were
normalized to DMSO-treated samples. No cells: ROS levels in GSK-J4-containing media (n ¼ 3 except for H23 and A549, n ¼ 4). F, Comparison of A549 cell viability upon GSK-J4 treatment (5 mmol/L, 48 hours), with or without 5 and 10 mmol/L of supplemental glutathione (GSH; n ¼ 4). G and H, Comparison of
ROS levels (left) and viability (right) of A549 cells (G) and H23 (H) upon GSK-J4 treatment (5 mmol/L,
48 hours), with or without Vit E (100 mmol/L; n ¼ 4). Data, means ti SEM. ti , P < 0.05; titi , P < 0.01;
tititi , P < 0.001; titititi , P < 0.0001. n.s., nonsignificant.
cell line from GSK-J4's effect on cell viability (Supplementary Fig. S7J and S7K). Thus, oncogenic KRAS renders LuAC and PDAC more susceptible to GSK-J4.
Because GSK-J4 reduces the viability of sensitive lines by inducing oxidative and metabolic stress, we asked whether oncogenic KRAS sensitizes LuAC to GSK-J4 via these mechan- isms. KRAS knockdown in A549 impaired GSK-J4-mediated ROS induction (Fig. 6D) while H1975-KRASG12V increased ROS levels upon GSK-J4 treatment (Fig. 6E). In addition, KRAS knockdown in A549 reversed the GSK-J4-mediated
changes in Gln/Glu levels (Fig. 6F) while H1975-KRASG12V showed increased Gln and decreased Glu levels upon GSK-J4 (Fig. 6G). Consistent with this, KRAS knockdown in H1975 impaired GSK-J4-mediated increases in ROS and Gln (Supple- mentary Fig. S7L and S7M).
Furthermore, D2OG restored the viability of H1975-KRASG12V in the presence of GSK-J4 (Fig. 6H), which is similar to what we observed with the sensitive lines (refer to Fig. 5A), Finally, KRAS knockdown in the sensitive line prevented GSK-J4-mediated induction of ATF4, while KRAS overexpression in the resistant
www.aacrjournals.org Cancer Res; 79(22) November 15, 2019 5855
Hong et al.
Figure 5.
GSK-J4 induces metabolic stress in sensitive LuAC lines. A and B, Comparison of GSK-J4 sensitivity with or without dimethyl-2- oxoglutarate (D2OG). GSK-J4– sensitive cells were pretreated with
1mmol/L (H23) or 2 mmol/L (A549 and H2009) D2OG for 2 hours, followed by GSK-J4 treatment for 24 hours (n ¼ 3). Then, their cell viabilities (A) and ATF4 proteins levels (B) were analyzed. C and D, Similar to A and B, except GSK-J4– resistant lines were used (n ¼ 3). Data, means ti SEM. ti , P < 0.05;
titi , P < 0.01; tititi , P < 0.001. n.s., nonsignifi cant.
line stimulated it (Fig. 6I). These results are consistent with GSK-J4–mediated activation of ISR in the sensitive lines. Together, our data indicate oncogenic KRAS sensitizes LuAC to GSK-J4 by inducing oxidative and metabolic stress.
Discussion
Deregulated histone methylation contributes to the progres- sion of several types of cancers, suggesting KMT and KDM inhi- bitors as anticancer drug candidates. This promising concept, however, has not been as straightforward as it fi rst appeared mainly because of the low specificity of these inhibitors and
insufficient information on their mechanisms of action. More importantly, a lack of appropriate biomarkers often makes it diffi cult to identify the patient groups who will benefit most from epigenetic inhibitors.
Here, we demonstrated KRAS mutations render LuAC sus- ceptible to GSK-J4 and identifi ed the underlying mechanisms. To our knowledge, this is the fi rst report of an association between a genetic mutation in a non-histone gene and a cancer's susceptibility to GSK-J4. Our data indicate GSK-J4 confers its antitumor effect on LuAC mainly by inhibiting KDM6B activity. Consistent with our study, other reports have suggested KDM6B as a main target of the GSK-J4 in high-risk
5856 Cancer Res; 79(22) November 15, 2019 Cancer Research
KRAS Mutation Determines GSK-J4 Sensitivity in LuAC
Figure 6.
An activating KRAS mutation sensitizes LuAC to GSK-J4. A, Relative GSK-J4 response in A549 cells upon KRAS knockdown. A549 cells were transfected with 3 nmol/L siNT, siKRAS #1, or #2. Seventy-two hours posttransfection, cells were treated with GSK-J4 for 48 hours, and relative viability was analyzed (n ¼
5). B, Similar experiments to A except that H1975-expressing control vector
(Vec) or oncogenic KRASG12V or KRASG12C were used (n ¼ 4). C, Tumor growth rates of the indicated cell lines upon DMSO (D) or GSK-J4 (G) treatment in xenograft assays (n ¼ 12 except for H1975-KRASG12V DMSO, n ¼ 11). D and E, Relative ROS levels in cells from A and B upon GSK-J4 treatment (5 mmol/L, 48 hours; n ¼ 3 for D, 4 for E). F and G, Relative Gln and Glu levels in the indicated A549 (F) and H1975 (G) cell lines upon GSK-J4 treatment (5 mmol/L, 12 hours; n ¼ 3). H, Comparison
of GSK-J4 sensitivity in H1975 cells with
or without D2OG. The indicated cells were pretreated with 2 mmol/L D2OG for
2hours, followed by GSK-J4 treatment for 24 hours (n ¼ 3). I, Changes in ATF4 protein levels in the same lines used in F and G upon GSK-J4 treatment (5 mmol/L, 12 hours). Representative images are
shown (n ¼ 3). D, DMSO; G, GSK-J4. Data, means ti SEM. ti , P < 0.05; titi , P < 0.01;
tititi , P < 0.001, titititi , P < 0.0001.
neuroblastoma, leukemia, and H3.3 K27M mutant–containing glioma (36–38). Although GSK-J4 seems to function in sensi- tive LuAC primarily via inhibition of KDM6B, it may also act via KDM5 family members.
Our study suggests that GSK-J4–mediated glutamate reduction as an important contributor to LuAC GSK-J4 sensitivity. This glutamate reduction then induces oxidative and metabolic stress by limiting cellular levels of the anti-oxidant glutathione and the TCA cycle intermediate aKG. Consistent with our results, recent
studies reported glutaminase inhibitor shows antitumor effects in hepatocarcinoma, triple-negative breast cancer, IDH1 mutant– containing glioma, and KRAS/KEAP1/NRF2 mutant–containing LuAC (39–41).
In addition to being a TCA cycle intermediate, aKG is a cofactor for JMJC domain–containing KDM family members. Thus, changes in aKG can affect KDM activities. In BRAF-mutant mel- anoma, for example, tumor areas with low glutamate (and low aKG) showed increased H3K27me3 owing to reduced KDM6B
www.aacrjournals.org Cancer Res; 79(22) November 15, 2019 5857
Hong et al.
activity (42). Thus, we speculate GSK-J4-mediated reduction of aKG may amplify the effects of GSK-J4 by further inhibition of KDM6B function. This may explain why GSK-J4's antitumor effect in vivo lasts so long after drug removal, but this warrants further investigation.
The oxidative and metabolic stress caused by GSK-J4 acti- vates the ISR in LuAC as demonstrated by ATF4 induction, but ATF4 knockdown did not affect GSK-J4 sensitivity. We specu- late that ATF4 activation is a consequence of GSK-J4-induced oxidative and nutrient stress rather than the direct cause of reduced cell viability. Interestingly, Lochmann and colleagues recently reported GSK-J4 induces ATF4 in neuroblastoma (36) and this mediates the effects of GSK-J4 (36). The difference between their study and ours may depend on inherent biolog- ical differences between the two cancer types.
Importantly, we found GSK-J4–induced oxidative and meta- bolic stress depends on the presence of oncogenic KRAS. Previous studies showed that KRAS-mutant cancer cells are more sensitive to nutrient and oxidative stress (43, 44). Furthermore, another recent study showed in NSCLC that oncogenic KRAS alters expression of genes involved in amino acid transport and metab- olism upon glutamine deprivation (45). Thus, our study links the role of oncogenic KRAS in the metabolic stress response to GSK-J4 sensitivity. In addition to having KRAS mutation, three of four GSK-J4–sensitive LuAC lines contained LKB1 comutations. Because KRAS/LKB1 comutations have been reported to confer hypersensitivity to a DNA methyltransferase inhibitor in LuAC and PDAC (46), we also examined the contribution of LKB1 mutation in GSK-J4 sensitivity, but found no association. This suggests that oncogenic KRAS is a major determinant of GSK-J4 sensitivity.
To data, association between histone H3.3 K27M mutations and GSK-J4 sensitivity in glioma has been validated (47). A potential association of GSK-J4 sensitivity with PTEN loss and CEBP/p300 mutations has been suggested, but experimental data was insuffi cient (48, 49). In addition, consistent with our results, a recent study reported A549 is more sensitive to GSK-J4 than the PC9 line (50), but this was not investigated further. Thus, our study is the first to experimentally validate an association between
a nonhistone gene mutation and GSK-J4 sensitivity. Collectively, we have demonstrated an unprecedented synthetic lethal inter- action between oncogenic KRAS and GSK-J4 and discovered its novel mechanism of action. Further studies on this synthetic lethality may expand the use of GSK-J4 in treating various cancers with KRAS mutations.
Disclosure of Potential Conflicts of Interest
No potential conflicts of interest were disclosed.
Authors' Contributions
Conception and design: B.-J. Hong, W.-Y. Park, M.-Y. Kim Development of methodology: W.-Y. Park, J.W. Moon, J.H. Park
Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): B.-J. Hong, W.-Y. Park, H.-R. Kim, J.W. Moon, H.Y. Lee, J.H. Park, Y. Oh, J.-S. Roe, M.-Y. Kim
Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): B.-J. Hong, W.-Y. Park, J.W. Moon, S.-K. Kim, J.-S. Roe Writing, review, and/or revision of the manuscript: B.-J. Hong, J.-S. Roe, M.-Y. Kim
Study supervision: W.-Y. Park, M.-Y. Kim
Acknowledgments
We thank Drs. J. Massague for PC9, H2030, AsPC-1, and BxPC-3, G-.H. HA for HCC4006, D.S-LIM for WI38 and IMR90, J.-H. Cho for pBABE EGFR WT and EGFR L858R constructs, H.-S. Cheong for the pBABE KRASG12V mutant construct, and Z. Luo for pLenti-LKB1 construct. This work was supported by Ministry of Health & Welfare (HI17C2049), awarded to W.Y. Park; KIB CMCC (N1018001), KC30 (N11180008), NRF-2016M3A9B4915818, NRF- 2019R1A2C2007207), Intelligent Synthetic Biology Center ISBC (2011- 0031955), awarded to M.-Y. Kim; The Yonsei University Future-leading Research Initiative of 2018 (2018-22-0051), NRF grant funded by MSIT (NRF-2018R1C1B6003133), Brain Korea 21 Plus program, and the T.J. Park Science Fellowship, awarded to J.-S. Roe; BK21 Plus Program, and WISET funded by MSIT under the Program for Returners into R&D to H.-R. Kim.
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.
Received November 6, 2018; revised July 29, 2019; accepted September 6, 2019; published first September 10, 2019.
References
1.Herbst RS, Morgensztern D, Boshoff C. The biology and management of non-small cell lung cancer. Nature 2018;553:446–54.
2.Zappa C, Mousa SA. Non-small cell lung cancer: current treatment and future advances. Transl Lung Cancer Res 2016;5:288–300.
3.Feng H, Wang X, Zhang Z, Tang C, Ye H, Jones L, et al. Identification of Genetic Mutations in Human Lung Cancer by Targeted Sequencing. Cancer Inform 2015;14:83–93.
4.Greulich H. The genomics of lung adenocarcinoma: opportunities for targeted therapies. Genes Cancer 2010;1:1200–10.
5.Leung AW, de Silva T, Bally MB, Lockwood WW. Synthetic lethality in lung cancer and translation to clinical therapies. Mol Cancer 2016;15:61.
6.Shin SM, Choi DK, Jung K, Bae J, Kim JS, Park SW, et al. Antibody targeting intracellular oncogenic Ras mutants exerts anti-tumour effects after sys- temic administration. Nat Commun 2017;8:15090.
7.Spencer-Smith R, Koide A, Zhou Y, Eguchi RR, Sha F, Gajwani P, et al. Inhibition of RAS function through targeting an allosteric regulatory site. Nat Chem Biol 2017;13:62–8.
8.Park WY, Hong BJ, Lee J, Choi C, Kim MY. H3K27 Demethylase JMJD3 Employs the NF-kappaB and BMP Signaling Pathways to Modulate the Tumor Microenvironment and Promote Melanoma Progression and Metastasis. Cancer Res 2016;76:161–70.
9.Dawson MA, Kouzarides T. Cancer epigenetics: from mechanism to ther- apy. Cell 2012;150:12–27.
10.Ansari J, Shackelford RE, El-Osta H. Epigenetics in non-small cell lung cancer: from basics to therapeutics. Transl Lung Cancer Res 2016;5: 155–71.
11.Tanaka M, Roberts JM, Qi J, Bradner JE. Inhibitors of emerging epigenetic targets for cancer therapy: a patent review (2010–2014). Pharm Pat Anal 2015;4:261–84.
12.Morera L, Lubbert M, Jung M. Targeting histone methyltransferases and demethylases in clinical trials for cancer therapy. Clin Epigenetics 2016;8: 57.
13.Fillmore CM, Xu C, Desai PT, Berry JM, Rowbotham SP, Lin YJ, et al. EZH2 inhibition sensitizes BRG1 and EGFR mutant lung tumours to TopoII inhibitors. Nature 2015;520:239–42.
14.Daffl on C, Craig VJ, Mereau H, Grasel J, Schacher Engstler B, Hoffman G, et al. Complementary activities of DOT1L and Menin inhibitors in MLL-rearranged leukemia. Leukemia 2017;31: 1269–77.
15.Vinogradova M, Gehling VS, Gustafson A, Arora S, Tindell CA, Wilson C, et al. An inhibitor of KDM5 demethylases reduces survival of drug-tolerant cancer cells. Nat Chem Biol 2016;12:531–8.
5858 Cancer Res; 79(22) November 15, 2019 Cancer Research
KRAS Mutation Determines GSK-J4 Sensitivity in LuAC
16.Chen Y, Liu X, Li Y, Quan C, Zheng L, Huang K. Lung cancer therapy targeting histone methylation: opportunities and challenges. Comput Struct Biotechnol J 2018;16:211–23.
17.Miranda-Goncalves V, Lameirinhas A, Henrique R, Jeronimo C. Metabo- lism and epigenetic interplay in cancer: regulation and putative therapeutic targets. Front Genet 2018;9:427.
18.Wong CC, Qian Y, Yu J. Interplay between epigenetics and metabolism in oncogenesis: mechanisms and therapeutic approaches. Oncogene 2017; 36:3359–74.
19.Gu Z, Liu Y, Cai F, Patrick M, Zmajkovic J, Cao H, et al. Loss of EZH2 Reprograms BCAA Metabolism to Drive Leukemic Transformation. Cancer Discov 2019.
20.Pang B, Zheng XR, Tian JX, Gao TH, Gu GY, Zhang R, et al. EZH2 promotes metabolic reprogramming in glioblastomas through epigenetic repression of EAF2-HIF1alpha signaling. Oncotarget 2016;7:45134–43.
21.Seok S, Kim YC, Byun S, Choi S, Xiao Z, Iwamori N, et al. Fasting-induced JMJD3 histone demethylase epigenetically activates mitochondrial fatty acid beta-oxidation. J Clin Invest 2018;128:3144–59.
22.Tripathi S, Pohl MO, Zhou Y, Rodriguez-Frandsen A, Wang G, Stein DA, etal. Meta-andOrthogonal Integration ofInfl uenza"OMICs" DataDefi nes a Role for UBR4 in Virus Budding. Cell Host Microbe 2015;18:723–35.
23.Heinemann B, Nielsen JM, Hudlebusch HR, Lees MJ, Larsen DV, Boesen T, et al. Inhibition of demethylases by GSK-J1/J4. Nature 2014;514:E1–2.
24.Kruidenier L, Chung CW, Cheng Z, Liddle J, Che K, Joberty G, et al. A selective jumonji H3K27 demethylase inhibitor modulates the proinflam- matory macrophage response. Nature 2012;488:404–8.
25.Wang L, Chang J, Varghese D, Dellinger M, Kumar S, Best AM, et al. A small molecule modulates Jumonji histone demethylase activity and selectively inhibits cancer growth. Nat Commun 2013;4:2035.
26.Dalvi MP, Wang L, Zhong R, Kollipara RK, Park H, Bayo J, et al. Taxane- platin-resistant lung cancers co-develop hypersensitivity to JumonjiC demethylase inhibitors. Cell Rep 2017;19:1669–84.
27.Bertoli C, Skotheim JM, de Bruin RA. Control of cell cycle transcription during G1 and S phases. Nat Rev Mol Cell Biol 2013;14:518–28.
28.Zhu W, Giangrande PH, Nevins JR. E2Fs link the control of G1/S and G2–M transcription. EMBO J 2004;23:4615–26.
29.Ly LL, Yoshida H, Yamaguchi M. Nuclear transcription factor Y and its roles in cellular processes related to human disease. Am J Cancer Res 2013;3: 339–46.
30.Haugwitz U, Wasner M, Wiedmann M, Spiesbach K, Rother K, Mossner J, et al. A single cell cycle genes homology region (CHR) controls cell cycle- dependent transcription of the cdc25C phosphatase gene and is able to cooperate with E2F or Sp1/3 sites. Nucleic Acids Res 2002;30:1967–76.
31.Dey S, Baird TD, Zhou D, Palam LR, Spandau DF, Wek RC. Both tran- scriptional regulation and translational control of ATF4 are central to the integrated stress response. J Biol Chem 2010;285:33165–74.
32.Pakos-Zebrucka K, Koryga I, Mnich K, Ljujic M, Samali A, Gorman AM. The integrated stress response. EMBO Rep 2016;17:1374–95.
33.Kornberg H. Krebs and his trinity of cycles. Nat Rev Mol Cell Biol 2000;1: 225–8.
34.Mailloux RJ, Beriault R, Lemire J, Singh R, Chenier DR, Hamel RD, et al. The tricarboxylic acid cycle, an ancient metabolic network with a novel twist. PLoS One 2007;2:e690.
35.Sunaga N, Shames DS, Girard L, Peyton M, Larsen JE, Imai H, et al. Knockdown of oncogenic KRAS in non-small cell lung cancers suppresses tumor growth and sensitizes tumor cells to targeted therapy. Mol Cancer Ther 2011;10:336–46.
36.Lochmann TL, Powell KM, Ham J, Floros KV, Heisey DAR, Kurupi RIJ, et al. Targeted inhibition of histone H3K27 demethylation is effective in high- risk neuroblastoma. Sci Transl Med 2018;10. doi: 10.1126/scitranslmed. aao4680.
37.Li Y, Zhang M, Sheng M, Zhang P, Chen Z, Xing W, et al. Therapeutic potential of GSK-J4, a histone demethylase KDM6B/JMJD3 inhibitor, for acute myeloid leukemia. J Cancer Res Clin Oncol 2018;144:1065–77.
38.Ntziachristos P, Tsirigos A, Welstead GG, Trimarchi T, Bakogianni S, Xu L, et al. Contrasting roles of histone 3 lysine 27 demethylases in acute lymphoblastic leukaemia. Nature 2014;514:513–7.
39.Gross MI, Demo SD, Dennison JB, Chen L, Chernov-Rogan T, Goyal B, et al. Antitumor activity of the glutaminase inhibitor CB-839 in triple-negative breast cancer. Mol Cancer Ther 2014;13:890–901.
40.Seltzer MJ, Bennett BD, Joshi AD, Gao P, Thomas AG, Ferraris DV, et al. Inhibition of glutaminase preferentially slows growth of glioma cells with mutant IDH1. Cancer Res 2010;70:8981–7.
41.Galan-Cobo A,Sitthideatphaiboon P, Qu X,Poteete A,PisegnaMA, Tong P, et al. LKB1 and KEAP1/NRF2 pathways cooperatively promote metabolic reprogramming with enhanced glutamine dependence in KRAS-mutant lung adenocarcinoma. Cancer Res 2019;79:3251–67.
42.Pan M, Reid MA, Lowman XH, Kulkarni RP, Tran TQ, Liu X, et al. Regional glutamine defi ciency in tumours promotes dedifferentiation through inhibition of histone demethylation. Nat Cell Biol 2016;18:1090–101.
43.Iskandar K, Rezlan M, Yadav SK, Foo CH, Sethi G, Qiang Y, et al. Synthetic lethality of a novel small molecule against mutant KRAS-expressing cancer cells involves AKT-dependent ROS production. Antioxid Redox Signal 2016;24:781–94.
44.Gaglio D, Metallo CM, Gameiro PA, Hiller K, Danna LS, Balestrieri C, et al. Oncogenic K-Ras decouples glucose and glutamine metabolism to support cancer cell growth. Mol Syst Biol 2011;7:523.
45.Gwinn DM, Lee AG, Briones-Martin-Del-Campo M, Conn CS, Simpson DR, Scott AI, et al. Oncogenic KRAS regulates amino acid homeostasis and asparagine biosynthesis via ATF4 and alters sensitivity to L-asparaginase. Cancer Cell 2018;33:91–107e6.
46.Kottakis F, Nicolay BN, Roumane A, Karnik R, Gu H, Nagle JM, et al. LKB1 loss links serine metabolism to DNA methylation and tumorigenesis. Nature 2016;539:390–5.
47.Hashizume R, Andor N, Ihara Y, Lerner R, Gan H, Chen X, et al. Pharma- cologic inhibition of histone demethylation as a therapy for pediatric brainstem glioma. Nat Med 2014;20:1394–6.
48.Daures M, Idrissou M, Judes G, Rifai K, Penault-Llorca F, Bignon YJ, et al. A new metabolic gene signature in prostate cancer regulated by JMJD3 and EZH2. Oncotarget 2018;9:23413–25.
49.Mathur R, Sehgal L, Havranek O, Kohrer S, Khashab T, Jain N, et al. Inhibition of demethylase KDM6B sensitizes diffuse large B-cell lympho- ma to chemotherapeutic drugs. Haematologica 2017;102:373–80.
50.Watarai H, Okada M, Kuramoto K, Takeda H, Sakaki H, Suzuki S, et al. Impact of H3K27 demethylase inhibitor GSKJ4 on NSCLC cells alone and in combination with metformin. Anticancer Res 2016;36:6083–92.
www.aacrjournals.org Cancer Res; 79(22) November 15, 2019 5859
Oncogenic KRAS Sensitizes Lung Adenocarcinoma to GSK-J4– Induced Metabolic and Oxidative Stress
Beom-Jin Hong, Woo-Yong Park, Hwa-Ryeon Kim, et al.
Cancer Res 2019;79:5849-5859. Published OnlineFirst September 10, 2019.
Updated version
Access the most recent version of this article at: doi:10.1158/0008-5472.CAN-18-3511
Supplementary Material
Access the most recent supplemental material at: http://cancerres.aacrjournals.org/content/suppl/2019/09/10/0008-5472.CAN-18-3511.DC1
Cited articles
This article cites 48 articles, 11 of which you can access for free at: http://cancerres.aacrjournals.org/content/79/22/5849.full#ref-list-1
E-mail alerts Sign up to receive free email-alerts related to this article or journal.
Reprints and Subscriptions
To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at [email protected].
Permissions
To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/79/22/5849.
Click on “Request Permissions” which will take you to the Copyright Clearance Center’s (CCC) Rightslink site.