Abstract
Gliomas, the most common primary brain tumors, are characterized by isocitrate dehydrogenase 1 mutation (IDH1-M). High mutation frequency of IDH1 indicates it’s promoting role in tumorgenesis. However, the observation that patients with IDH1-M have better survival comparing with patients with IDH1 wild-type (IDH1-W) suggests that this alteration has other significant beneficial features for patients. Currently, temozolomide (TMZ) is a standard of care for patients which play a major role in DNA methylation that is similar with the role of IDH1-M in genome-wide methylation. In this study, we collected 323 gliomas samples with genome-wide methylation microarray, 502 samples with genome-wide mRNA expression microarray and 295 samples with RNA-seq. By significance analysis of microarray (SAM), we identified 18 genes which are hypermethylation and low expression in samples with IDH1-M comparing with IDH1-W (FDR<0.01). Furthermore, 18 candidate genes were downregulated in TMZ-treated samples. Finally, we obtained two candidate genes, F3 and RBP1. Survival analysis showed that hypermethylation or low expression of the two genes indicated a favorable prognosis, which was consistent with IDH1-M and administration of TMZ in glioma patients. F3 and RBP1 were further validated by qPCR on an independent validation cohort containing 145 samples. Our data suggest that these candidate genes were suppressed by TMZ or IDH1-M induced hypermethylation, resulting in the favorable prognosis of patients with gliomas.
Keywords: Gliomas, IDH1 mutation, temozolomide, RNA-Seq, RNA microarray, methylation microarray
Introduction
Gliomas are the most common primary brain tumors and important cause of cancer-related mortality among adults. Aggressive surgery followed by adjuvant radiation and/or chemotherapy is considered the standard of care, but provides limited benefits [1]. Temozolomide (TMZ), a new oral alkylating agent, is currently the gold standard for adjuvant chemotherapy in patients with gliomas, and has been shown to significantly improve drug efficacy and safety. DNA methylation and failure of mismatch repair play a major role in TMZ’s cytotoxicity. Addition of a methyl group to the O6 position of guanine in genomic DNA results in the incorporation of thymine residue complementing O6-methylguanine instead of the normal cytosine residue. The abnormal guanine-thymine pair leads to a pause in the DNA replication fork and triggers the DNA mismatch repair pathway, which eventually leads to cell cycle arrest and cell death [2]. However, many genes, like MGMT, can cause TMZ resistance in cancer cells by repairing the TMZ-induced DNA methylation [3]. Similarly, tumor cells present other genetic alterations (mutation, hypermethylation and overexpression) and do not undergo TMZ-induced G2 arrest; such cells are resistant to TMZ-induced cell death [4]. Moreover, patient survival is only prolonged by two months of TMZ administration (12.1 months with radiotherapy alone vs 14.6 months with radiotherapy plus TMZ) [1]. Therefore, TMZ-based chemotherapy must be improved in order to overcome the reduced sensitivity resulting from aberrant genes in a subgroup of glioma.
Recently, IDH1 was shown to be mutated in up to 70% of low grade gliomas (grades II and III) and secondary glioblastomas (grade IV), making it an important marker to guide treatment decisions [5,6]. Patients with IDH1-M have a favorable prognosis compared with those harboring IDH1-W [7]. However, most studies have proposed IDH1-M to be an oncogene in glioma development [8,9], which is inconsistent with its prognostic value. Conversely, studies have found that expression of mutant IDH1 causes increased hypermethylation of a large number of genes [10], which might reasonably explain the prognostic value of IDH1-M. IDH1 catalyzes the oxidative decarboxylation of isocitrate to α-ketoglutarate (α-KG) [11]. IDH1-M is heterozygous and exclusively affects arginine at position 132. This mutation occurs at the arginine residue of the enzyme’s active site and causes reduction of α-KG to D-2-hydroxyglutarate. Accumulation of this metabolite induces DNA hypermethylation, leading to genome-wide epigenetic changes [12]. Therefore, it is plausible that IDH1-M occurrence and TMZ administration play a similar role in gene methylation. It is well accepted that gene methylation and expression are closely related: methylation affects prognosis by regulating the expression of some genes. Consequently, identifying such methylation associated genes is important for decision-making when applying alkylating drugs to patients with gliomas.
In this study, we collected 323, 502, and 295 samples with methylation microarray, mRNA microarray, and RNA-seq, respectively. By comparing samples with and without IDH1-M, and samples treated with TMZ and control untreated samples, we obtained 2 candidate genes associated with IDH1-M or TMZ induced hypermethylation. Survival analysis showed that hypermethylation or low expression of the two genes indicated a favorable prognosis, consistently with IDH1-M and TMZ treatment. Finally, the expression levels were further validated on an independent validation cohort. Overall, these findings provide a further basis for understanding the roles of IDH1-M and TMZ treatment in gliomas. The molecular mechanism of methylation may well be represented by these suppressed genes, which are novel potential interfering targets for treatment of the subgroup of patients with gliomas without IDH1-M.
Materials and method
Samples
Samples obtained from four datasets were listed in Table S1. The characteristics of patients from The Chinese Glioma Genome Atlas (CGGA, http://www.cgga.org.cn) are detailed in Table 1. These samples were used to perform whole genome expression profiling, genome-wide methylation profiling, RNA-sequencing, and survival analysis. All patients underwent surgical resection from January 2005 through December 2012. Patients were eligible for the study if their diagnosis was established histologically by 2 neuropathologists according to the 2007 WHO classification guidelines. Only samples with 80% tumor cells were selected for analysis. This study was approved by the institutional review boards, and wrote informed consent was obtained from each patient. The independent sample cohorts The Cancer Genome Atlas (TCGA) and GSE16011 are well described in public databases (http://cancergenome.nih. gov and http: //www.ncbi.nlm. nih.gov/geo/query).
Table 1.
Clinical characteristics of the IDH1 status in CGGA samples
| Variables | IDH1-M | IDH1-W | Total |
|---|---|---|---|
| Samples | 241 (48.0%) | 261 (52.0%) | 502 |
| Age (average) | 46.4 (9-81) | 38.8 (10-66) | 42.8 (9-81) |
| Sex | |||
| Male | 137 (45.8%) | 162 (54.2%) | 299 |
| Female | 104 (51.2%) | 99 (48.8%) | 203 |
| Histology | |||
| A | 67 (75.3%) | 22 (24.4%) | 89 |
| O | 27 (79.4%) | 7 (20.6%) | 34 |
| OA | 46 (66.7%) | 23 (33.3%) | 69 |
| AA | 13 (39.4%) | 20 (60.6%) | 33 |
| AO | 16 (84.2%) | 3 (15.8%) | 19 |
| AOA | 26 (51.0%) | 25 (49.0%) | 51 |
| GBM | 46 (22.2%) | 161 (81.8%) | 207 |
| Location* | |||
| Left hemisp | 110 (46.6%) | 126 (53.4%) | 236 |
| Right hemisp | 110 (48.8%) | 116 (51.2%) | 226 |
| Mixed | 16 (57.1%) | 12 (42.9%) | 28 |
12 samples lost the information.
DNA pyro-sequencing for IDH1 mutation
Genomic DNA was isolated from frozen tumor tissues by using the QIAamp DNA Mini Kit (Qiagen). The genomic region spanning wild-type R132 of IDH1 was analyzed by pyrophosphate sequencing using the following primers: 5’-GCTTGTGAGTG-GATGGGTAAAAC-3’ and 5’-Biotin-TTGCCAACATG ACT-TACTTGATC-3’. The PCR analysis was performed in duplicate in 40 µl reaction containing 1 µl of 10 µM each primer, 4 µl of 10× buffer, 3.2 µl of 2.5 mM dNTPs, 2.5 U hotstart Taq (Takara) and 2 μl of 10 µM DNA. The PCR was carried out on an ABI PCR system 9700 (Applied Biosystems) with the following program: 95°C for 3 min; 50 cycles of 95°C for 15 s, 56°C for 20 s, and 72°C for 30 s; 72°C for 5 min. Single-stranded DNA was purified from the total PCR product and subjected to pyrosequencing on PyroMark Q96 ID System (QIAGEN) using the primer 5’-TGGATGGG TAAAACCT-3’ and EpiTect Bisulfite Kit (QIAGEN).
Whole genome expression profiling (CGGA)
Microarray analysis was performed using an Agilent Whole Human Genome Array according to the manufacturer’s instructions. The integrity of total RNA was checked with an Agilent 2100 Bioanalyzer (Agilent). cDNA and biotinylated cRNA were synthesized and hybridized to the array. Data were acquired using the Agilent G2565BA Microarray Scanner System and Agilent Feature Extraction Software (version 9.1). Probe intensities were normalized using GeneSpring GX 11.0. GSE16011 whole genome expression profiling data were downloaded from public database.
Genome-wide DNA methylation profiling (CGGA)
For methylation profiling, we used the Illumina Infinium Human Methylation 27 Bead-Chips (Illumina Inc.) as described previously [13]. The BeadChip contains 27,578 highly informative CpG sites covering more than 14,000 human RefSeq genes. This allows researchers to interrogate all these sites in samples at a single nucleotide resolution. Bisulfite modification of DNA, chip processing and data analysis were performed following the manufacturer’s manual at the Wellcome Trust Centre for Human Genetics Genomics Lab, Oxford, UK. The results were analyzed with the BeadStudio software (Illumina). TCGA methylation profiling data were downloaded from the TCGA database.
RNA-seq (CGGA)
Total RNA was isolated using RNeasy Mini Kit (Qiagen) according to the manufacturer’s instructions. A pestle and a QIAshredder (Qiagen) tube were used to disrupt and homogenize the frozen tissues. RNA intensity was assessed using 2100 Bioanalyzer (Agilent Technologies); only high quality samples with RNA Integrity Number (RIN) values greater than or equal to 7.0 were used to construct the sequencing library. The subsequent steps included end repair, adapter ligation, size selection and polymerase chain reaction enrichment. The DNA fragment length was measured using a 2100 Bioanalyzer, and median insert sizes were 200 nucleotides. The libraries were sequenced on the Illumina HiSeq 2000 platform using the 101-bp pair-end sequencing strategy. Short sequence reads were aligned to the human reference genome (Hg19 Refseq) using the Burrows-Wheeler Aligner (BWA, Version 0.6.2-r126).
Real time quantitative PCR (qPCR)
Gene expression levels of F3 and RBP1 in 145 samples were analyzed by real-time quantitative PCR using the SYBR Supermix Kit (Bio-Rad, Hercules, CA). PCR reactions included the following components: 100 nM each primer, diluted cDNA templates, and iQ SYBR Green supermix. The PCR efficiency was examined by assessing serial dilutions of the template cDNA, and melting curve data were collected to evaluate specificity. Each cDNA sample was analyzed in triplicate. ACTIN was used for normalization, and relative mRNA level was determined as 2-[Ct (ACTIN)-Ct (gene of interest)]. The primer sequences: F3-F: 5’CCGA CGAGATTGTGAAGGATGTGA-3’, F3-R: 5’-TCCGAGG TTTGTCTCC AGGTAAGG-3’; RBP1-F: 5’-TCCAGT ACTCCCCGAAATG-3’, RBP1-R: 5’-AGGTACTCCTCGAAATTCTCGTT-3’.
Statistical analysis
Differentially expression genes in microarray and RNA-seq were detected by significance analysis of microarray (SAM) (FDR<0.01). Differentially expression genes in TMZ-treated samples and untreated samples were detected by unpaired Student’s t-test. Kaplan-Meier survival analysis was used to estimate the survival distributions. The log-rank test was applied to assess the statistical significance between stratified survival groups using the GraphPad Prism version 4.0 statistical software. KEGG pathway analysis was performed using DAVID (http://david.abcc.ncifcrf.gov/) [14]. Heat maps of different grade ODs were constructed by Gene Cluster 3.0 and Gene Tree View software using the differentially expression genes. Multivariate Cox models were used after univariate analysis using SPSS, version 13.0 (SPSS). A two-sided p value <0.05 was regarded as significant.
Results
Gene expression levels in samples with IDH1-M
IDH1 mutation increases gene methylation levels and results in downregulation of these genes [10,12]. To identify the downregulated genes induced by IDH1-M, we compared gene expression levels in 276 samples with and without IDH1-M from the CGGA dataset (FDR<0.01). By SAM analysis, a total of 4543 genes (5952 probes) were identified with lower expression in 126 samples with IDH-M compared with 150 IDH1-W samples. We projected these genes into 295 samples with RNA-seq from the CGGA dataset and 226 samples with mRNA microarray from the GSE16011 dataset (Figure 1A). Finally, 1568 out of 4543 genes were validated on 232 samples with IDH1-M and 289 samples with IDH-W (Figure 1B). As expected, the majority of samples with IDH1-M (53.2% and 84.1%) were classified into proneural subtype and lower grade gliomas (II and III). Among these genes, MGMT has been described in previous studies by us or others: its high methylation and low expression indicate a favorable outcome in patients with gliomas [15].
Figure 1.
IDH1-M specific gene expression profiling based on 1562 genes. A. By overlapping the data from the CGGA mRNA microarry and RNA-seq, and GSE16011 mRNA microarry dataset, a total of 1562 genes were identified by comparing the gene expression levels between samples with and without IDH1-M (FDR<0.01). B. Heat map was created using the 1562 genes using CGGA dataset; IDH1 status, glioma grade of samples and TCGA subtypes were indicated with different colors. C. Kaplan-Meier plots of the overall survival based on IDH1 status are shown. The overall survival data were analyzed using log-rank tests. W-IDH1 wild-type, M-IDH1 mutation; Ne-Neural, Pr-Proneural, Cl-Classical, Me-Mesenchymal.
Hypermethylation status in samples with IDH1-M
To identify the hypermethylated genes induced by IDH1-M, we compared the genome-wide methylation levels in 118 samples with and without IDH1-M from the CGGA dataset (FDR<0.01). A total of 1055 genes (1211 probes) showed higher methylation levels in 72 samples with IDH-M in comparison with 46 samples with IDH1-W. The list was projected into 205 samples with methylation microarray from the TCGA dataset (FDR<0.01) (Figure 2A). Finally, 606 out of 1055 genes showed higher methylation levels in 22 samples with IDH1-M compared with 183 IDH-W samples (Figure 2B). As expected, the majority of samples with G-CIMP+ were concentrated in the IDH1-M group. Among the genes, ALDH1A3 has been reported in our previous study with higher methylation levels in patients with prolonged survival time compared with those with short survival. A significant negative correlation was observed between ALDH1A3 methylation status and protein expression [13].
Figure 2.
IDH1-M specific hypermethylation profiling based on 604 genes. A. By overlapping the data from CGGA methylation profiling and TCGA methylation profiling, a list of 604 hypermethylatied genes were identified by comparing the methylation levels between samples with and without IDH1-M (FDR<0.01). B. Heat map was created using the 604 genes using CGGA dataset which had high methylation in samples with IDH-M; IDH1 status, grade of samples and G-CIMP subtypes were indicated with different colors. C. Kaplan-Meier plots of the overall survival based on IDH1 status are shown.
Candidate genes were downregulated in samples from patients treated with TMZ
To identify the genes downregulated as a result of hypermethylation, we overlapped the hypermethylated genes and downregulated genes based on the IDH1-M status. Finally, 156 genes were obtained with higher methylation and lower expression in samples with IDH1-M in comparison with the IDH1-W group. TMZ play a major role in DNA methylation that leads to cytotoxicity. To further confirm that these genes were affected by methylation, we collected 23 samples (recurrent gliomas) from patients who received at least one course of TMZ and 114 samples without any treatment on the CGGA dataset. We compared the gene expression levels and found that 18 out of 156 candidate genes were downregulated after treatment with TMZ (Figure 3A and 3B). Downregulation of these 18 candidate genes might result from the hypermethylation status and contribute to the favorable prognosis observed in the IDH1-M group and TMZ treatment (Figures 1C, 2C and 3C).
Figure 3.
18 candidate genes were downregulated in TMZ-treated samples. A and B. After overlapping the above two IDH1-M specific gene profiling, 18 genes were found to be downregulated in TMZ-treated samples (P<0.05). These patients were treated with TMZ chemotherapy or radiation after the first surgery and underwent second operation, and samples were collected. C. TMZ-treated patients with gliomas have significantly longer survival than untreated patients in the CGGA dataset.
High methylation and low expression confer a better clinical outcome
Increasing evidence has showed that IDH1-M and administration of TMZ are beneficial for glioma patients [1,16]. However, whether the candidate genes obtained by analysis of IDH1-M and methylation status are associated with patient survival is unknown. Multivariate Cox analyses showed that two candidate genes, F3 (P=0.037; 95% CI: 0.986-1.026) and RBP1 (P=0.003; 95% CI: 1.063-1.339), were significantly associated with OS independent of tumor grades and patients age (Tables S2, S3 and S4). Kaplan-Meier survival analysis showed that high methylation or low expression of F3 and RBP1 conferred a better clinical outcome that was similar with IDH1-M and TMZ administration of gliomas (Figure 4A). The result was further validated in TCGA methylation data and GSE16011 gene expression data (Figure 4B).
Figure 4.
Candidate genes were tightly associated with prognosis in both CGGA cohort and validation cohort. A. Low expression or hypermethylation of candidate genes indicates longer survival than high expression or hypomethylation based on CGGA dataset. B. The prognostic value of candidate genes was validated on GSE16011 mRNA microarray dataset and TCGA methylation microarray dataset.
Candidate genes were validated on an independent validation cohort
Semi-quantitative RT-PCR was performed to assess F3 and RBP1. The expression levels of the two genes were significantly lower in 75 samples with IDH-M than 70 samples with IDH1-W (Figure 5). These results provided further evidence supporting the downregulated genes resulted from IDH1-M. These data demonstrated that the candidate genes were affected by IDH1-M or TMZ, their alterations contributing to the prognostic value of IDH1-M and TMZ. These findings provide new targets for subgroups of patients without IDH1-W or those insensitive to TMZ.
Figure 5.
Candidate genes were validated by qPCR in 145 additional samples. The expression levels of F3 and RBP1 were lower in samples with IDH1-M than in those with IDH1-W on an independent validation cohort containing 145 samples (P<0.05).
Discussion
Gliomas are the most common primary brain tumors with various genetic alterations, such as TP53 mutation, PTEN mutation, EGFR amplification and 1p/19q loss. Currently, IDH1-M is a hot topic in glioma research. Direct sequencing in a series of 685 brain tumors revealed the highest frequencies of IDH1-M in gliomas (68-88%) [5]. The high frequency of IDH1-M suggests a role in early tumor development. Therefore, IDH1-M was thought to have protumorigenic potential [8,9]. However, glioma patients with IDH1-M have a better survival outcome compared with those harboring wild-type IDH1. IDH1-M acts as a predictive biomarker in gliomas and exhibits a better response to TMZ [17]. These data suggest that IDH1-M is a double-edged sword in gliomas. Recent studies have found that IDH1-M resulted in increased levels of D-2-hydroxyglutarate that induces DNA hypermethylation, leading to genome-wide epigenetic changes [11]. Researchers have found that glioma-CpG island methylator phenotype (G-CIMP+) is tightly associated with IDH1-M. This alteration along with G-CIMP+, defines a relatively favorable subtype (Proneural) [10]. Therefore, IDH1-M increases gene methylation levels and results in downregulation of some genes that might contribute to the favorable value of IDH1-M. However, the role of IDH1-M on the gene methylation might be similar to TMZ effect in gliomas. TMZ is a standard of care for patients which play a major role in DNA methylation.
In our study, we analyzed the genome-wide methylation profiling, whole genome expression profiling and RNA-seq to identify genes suppressed by the IDH1-M induced hypermethylation. A total of 18 genes were identified with higher methylation and lower expression in samples with IDH1-M compared with those without IDH1-M. Moreover, these candidate genes were also downregulated in TMZ-treated samples: ADPRH, ARSD, F3, FBXO6, FUCA2, GALM, MGST2, MOXD1, MT1E, MT1F, RAB3D, RBP1, RGN, RNF135, SLC12A7, SLC25A20, SSBP4 and TRIP4. Among the 18 genes, F3 and RBP1 were independent prognostic markers for patients with gliomas and hypermethylation or low expression of the two genes indicated a favorable prognosis.
RBP1 had been reported before that hypermethylation was described in nearly all IDH1 and IDH2 mutated gliomas (79/82) and shown to be associated with improved patient survival [17,18]. Indeed, decreased RBP1 expression was noted in glioma patients with long term survival. More importantly, reduction of methylation was closely related to increasing mRNA expression in the demethylated cell lines. These data suggest that RBP1 is regulated by methylation results from IDH1-M or administration of TMZ. But the exact functional effect of RBP1 hypermethylation in gliomas has yet to be elucidated. F3 encodes the coagulation factor III, a cell surface glycoprotein. It functions as a high-affinity receptor for coagulation factor VII and enables cells to initiate the blood coagulation cascades [19]. F3 mRNA and protein expression levels had proven to be increased in tumors from patients with clear cell carcinoma. Full-length F3 is overexpressed in breast cancer, and alternatively, spliced F3 promotes breast cancer growth in a β1 integrin-dependent manner. These findings show that F3 is an oncogene that leads to tumor development. However, these two genes have been rarely reported in human gliomas and their exact regulatory mechanisms remain to be elucidated. In the present study, these genes for the first time were reported by their association with IDH1-M and methylation in gliomas. Our results indicate that gene methylation induced by IDH1-M or TMZ administration might lead to low expression of these candidate genes, which contributes to favorable survival outcome.
Overall, these findings provide new insights for understanding the fundamental basis of IDH1-M and TMZ roles in gliomas. The molecular mechanism of methylation may well be represented by these genes. Finally, these genes constitute novel interfering targets for glioma patients.
Acknowledgements
This work was supported by grants from Beijing science and technology plan (No. Z131100006113018), National Science Foundation of China (No. 91229121) and National Natural Science Foundation of China (No. 81201993).
Disclosure of conflict of interest
None.
Supporting Information
References
- 1.Stupp R, Mason WP, van den Bent MJ, Weller M, Fisher B, Taphoorn MJ, Belanger K, Brandes AA, Marosi C, Bogdahn U, Curschmann J, Janzer RC, Ludwin SK, Gorlia T, Allgeier A, Lacombe D, Cairncross JG, Eisenhauer E, Mirimanoff RO. Radiotherapy plus concomitant and adjuvant temozolomide for glioblastoma. N Engl J Med. 2005;352:987–996. doi: 10.1056/NEJMoa043330. [DOI] [PubMed] [Google Scholar]
- 2.Roos WP, Batista LF, Naumann SC, Wick W, Weller M, Menck CF, Kaina B. Apoptosis in malignant glioma cells triggered by the temozolomide-induced DNA lesion O6-methylguanine. Oncogene. 2007;26:186–197. doi: 10.1038/sj.onc.1209785. [DOI] [PubMed] [Google Scholar]
- 3.Friedman HS, McLendon RE, Kerby T, Dugan M, Bigner SH, Henry AJ, Ashley DM, Krischer J, Lovell S, Rasheed K, Marchev F, Seman AJ, Cokgor I, Rich J, Stewart E, Colvin OM, Provenzale JM, Bigner DD, Haglund MM, Friedman AH, Modrich PL. DNA mismatch repair and O6-alkylguanine-DNA alkyltransferase analysis and response to Temodal in newly diagnosed malignant glioma. J. Clin. Oncol. 1998;16:3851–3857. doi: 10.1200/JCO.1998.16.12.3851. [DOI] [PubMed] [Google Scholar]
- 4.Johannessen TC, Bjerkvig R. Molecular mechanisms of temozolomide resistance in glioblastoma multiforme. Expert Rev Anticancer Ther. 2012;12:635–642. doi: 10.1586/era.12.37. [DOI] [PubMed] [Google Scholar]
- 5.Balss J, Meyer J, Mueller W, Korshunov A, Hartmann C, von Deimling A. Analysis of the IDH1 codon 132 mutation in brain tumors. Acta Neuropathol. 2008;116:597–602. doi: 10.1007/s00401-008-0455-2. [DOI] [PubMed] [Google Scholar]
- 6.Yan H, Parsons DW, Jin G, McLendon R, Rasheed BA, Yuan W, Kos I, Batinic-Haberle I, Jones S, Riggins GJ, Friedman H, Friedman A, Reardon D, Herndon J, Kinzler KW, Velculescu VE, Vogelstein B, Bigner DD. IDH1 and IDH2 mutations in gliomas. N Engl J Med. 2009;360:765–773. doi: 10.1056/NEJMoa0808710. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Yan W, Zhang W, You G, Bao Z, Wang Y, Liu Y, Kang C, You Y, Wang L, Jiang T. Correlation of IDH1 mutation with clinicopathologic factors and prognosis in primary glioblastoma: a report of 118 patients from China. PLoS One. 2012;7:e30339. doi: 10.1371/journal.pone.0030339. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Watanabe T, Nobusawa S, Kleihues P, Ohgaki H. IDH1 mutations are early events in the development of astrocytomas and oligodendrogliomas. Am J Pathol. 2009;174:1149–1153. doi: 10.2353/ajpath.2009.080958. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Dang L, White DW, Gross S, Bennett BD, Bittinger MA, Driggers EM, Fantin VR, Jang HG, Jin S, Keenan MC, Marks KM, Prins RM, Ward PS, Yen KE, Liau LM, Rabinowitz JD, Cantley LC, Thompson CB, Vander HM, Su SM. Cancer-associated IDH1 mutations produce 2-hydroxyglutarate. Nature. 2010;465:966. doi: 10.1038/nature09132. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Turcan S, Rohle D, Goenka A, Walsh LA, Fang F, Yilmaz E, Campos C, Fabius AW, Lu C, Ward PS, Thompson CB, Kaufman A, Guryanova O, Levine R, Heguy A, Viale A, Morris LG, Huse JT, Mellinghoff IK, Chan TA. IDH1 mutation is sufficient to establish the glioma hypermethylator phenotype. Nature. 2012;483:479–483. doi: 10.1038/nature10866. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Zhao S, Lin Y, Xu W, Jiang W, Zha Z, Wang P, Yu W, Li Z, Gong L, Peng Y, Ding J, Lei Q, Guan KL, Xiong Y. Glioma-derived mutations in IDH1 dominantly inhibit IDH1 catalytic activity and induce HIF-1alpha. Science. 2009;324:261–265. doi: 10.1126/science.1170944. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Figueroa ME, Abdel-Wahab O, Lu C, Ward PS, Patel J, Shih A, Li Y, Bhagwat N, Vasanthakumar A, Fernandez HF, Tallman MS, Sun Z, Wolniak K, Peeters JK, Liu W, Choe SE, Fantin VR, Paietta E, Lowenberg B, Licht JD, Godley LA, Delwel R, Valk PJ, Thompson CB, Levine RL, Melnick A. Leukemic IDH1 and IDH2 mutations result in a hypermethylation phenotype, disrupt TET2 function, and impair hematopoietic differentiation. Cancer Cell. 2010;18:553–567. doi: 10.1016/j.ccr.2010.11.015. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Zhang W, Yan W, You G, Bao Z, Wang Y, Liu Y, You Y, Jiang T. Genome-wide DNA methylation profiling identifies ALDH1A3 promoter methylation as a prognostic predictor in G-CIMP- primary glioblastoma. Cancer Lett. 2013;328:120–125. doi: 10.1016/j.canlet.2012.08.033. [DOI] [PubMed] [Google Scholar]
- 14.Huang DW, Sherman BT, Lempicki RA. Systematic and integrative analysis of large gene lists using DAVID bioinformatics resources. Nat Protoc. 2009;4:44–57. doi: 10.1038/nprot.2008.211. [DOI] [PubMed] [Google Scholar]
- 15.Hegi ME, Diserens AC, Gorlia T, Hamou MF, de Tribolet N, Weller M, Kros JM, Hainfellner JA, Mason W, Mariani L, Bromberg JE, Hau P, Mirimanoff RO, Cairncross JG, Janzer RC, Stupp R. MGMT gene silencing and benefit from temozolomide in glioblastoma. N Engl J Med. 2005;352:997–1003. doi: 10.1056/NEJMoa043331. [DOI] [PubMed] [Google Scholar]
- 16.Yan H, Parsons DW, Jin G, McLendon R, Rasheed BA, Yuan W, Kos I, Batinic-Haberle I, Jones S, Riggins GJ, Friedman H, Friedman A, Reardon D, Herndon J, Kinzler KW, Velculescu VE, Vogelstein B, Bigner DD. IDH1 and IDH2 mutations in gliomas. N Engl J Med. 2009;360:765–773. doi: 10.1056/NEJMoa0808710. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Minniti G, Scaringi C, Arcella A, Lanzetta G, Di Stefano D, Scarpino S, Bozzao A, Pace A, Villani V, Salvati M, Esposito V, Giangaspero F, Enrici RM. IDH1 mutation and MGMT methylation status predict survival in patients with anaplastic astrocytoma treated with temozolomide-based chemoradiotherapy. J Neurooncol. 2014;118:377–383. doi: 10.1007/s11060-014-1443-0. [DOI] [PubMed] [Google Scholar]
- 18.Chou AP, Chowdhury R, Li S, Chen W, Kim AJ, Piccioni DE, Selfridge JM, Mody RR, Chang S, Lalezari S, Lin J, Sanchez DE, Wilson RW, Garrett MC, Harry B, Mottahedeh J, Nghiemphu PL, Kornblum HI, Mischel PS, Prins RM, Yong WH, Cloughesy T, Nelson SF, Liau LM, Lai A. Identification of retinol binding protein 1 promoter hypermethylation in isocitrate dehydrogenase 1 and 2 mutant gliomas. J Natl Cancer Inst. 2012;104:1458–1469. doi: 10.1093/jnci/djs357. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Kocaturk B, Van den Berg YW, Tieken C, Mieog JS, de Kruijf EM, Engels CC, van der Ent MA, Kuppen PJ, Van de Velde CJ, Ruf W, Reitsma PH, Osanto S, Liefers GJ, Bogdanov VY, Versteeg HH. Alternatively spliced tissue factor promotes breast cancer growth in a beta1 integrin-dependent manner. Proc Natl Acad Sci U S A. 2013;110:11517–11522. doi: 10.1073/pnas.1307100110. [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.