Integrative transcriptome analysis identified a BMP signaling pathway‑regulated lncRNA AC068643.1 in IDH mutant and wild‑type glioblastomas
- Authors:
- Published online on: April 15, 2020 https://doi.org/10.3892/ol.2020.11542
- Pages: 75-84
-
Copyright: © Huang et al. This is an open access article distributed under the terms of Creative Commons Attribution License.
Abstract
Introduction
Glioblastomas (GBMs) are the most common and deadliest tumors of the central nervous system (CNS) (1). According to multiple studies from different regions of the world, despite radical therapy involving maximal tumor resection and chemo/radiotherapy, the median survival time is only 14.6 months, and the 5-year survival rate is only 9.8% for patients with GBM (2–4). At present, the development of targeted molecular therapy for GBMs is unsatisfactory; therefore, investigation of GBM pathogenesis at the genetic and molecular levels is required.
Somatic mutations in isocitrate dehydrogenases (IDH) have been commonly identified in lower grade gliomas (LGG; WHO grade II and III) and secondary GBMs (5). Development of IDH mutant (IDHMT) and IDH wild-type (IDHWT) gliomas is driven by different oncogenic processes, thus these subtypes present distinct molecular and clinical features (6). IDHWT GBM is more aggressive compared with IDHMT GBM, and the median survival time of patients with IDHMT GBM is 31 months, whereas that of patients with IDHWT GBM is <15 months (4,5). IDHMT GBM only constitutes a small proportion of total GBM cases (5), suggesting that research investigating the underlying mechanisms of the disease should focus on IDHWT GBM. However, tumor behaviors are closely associated with the altered gene expression patterns (7), and thus it is necessary to compare the transcriptomes of IDHMT and IDHWT GBM tissues, which may provide more valuable information than comparing IDHWT GBM and normal brain tissue transcriptomes.
Long noncoding RNAs (lncRNAs), which are >200 nucleotides long, have attracted attention due to their multiple gene regulation functions at the transcriptional, post-transcriptional and epigenetic levels (8). Previous studies have indicated that lncRNAs are predominantly expressed in the CNS (9,10) and are spatio-temporally regulated to serve functions in the CNS development (11). Therefore, dysregulation of lncRNA expression may contribute to a number of CNS diseases, including brain injuries, neurodegenerative diseases (12) and GBM (13,14). For example, Han et al (15) identified 1,308 lncRNAs dysregulated in GBM tissues compared with the normal brain tissues. LncRNAs serve a number of roles in GBM, including the involvement in the NEAT-1-regulated EGFR pathway (16), Hox transcript antisense intergenic RNA HOTAIR-regulated GBM proliferation (17,18) and stem-related lncRNA HIF1A-AS2-regulated GBM stemness maintenance (19). The present study aimed to identify differentially expressed lncRNAs (DE-lncRNAs) and mRNAs (DE-mRNAs) between IDHMT and IDHWT GBM by mining RNA sequencing data from The Cancer Genome Atlas database (TCGA), exploring their interactions and potential functions in mediating the different tumor behaviors observed in IDHMT and IDHWT GBM.
Materials and methods
GBM datasets and clinical samples
The present study was approved by The Medical Ethics Committee of The Second Affiliated Hospital (Xinqiao Hospital) of the Third Military Medical University, and written informed content was obtained from the patients. mRNA and lncRNA expression data and corresponding clinical information, including patient sex, age at diagnosis and Karnofsky Performance Status (KPS), for 132 GBM samples (122 IDHWT and 10 IDHMT cases) were obtained from TCGA database (cancer.gov/tcga). The datasets contained 19,676 protein-coding mRNAs and 9,599 lncRNAs. A total of 128 GBM samples (107 IDHWT cases and 21 IDHMT cases) with mRNA microarray data and corresponding clinical information were obtained from the Chinese Glioma Genome Atlas (CGGA; cgga.org.cn/download.jsp) database for further validation. In addition, 45 GBM samples (37 IDHWT and 8 IDHMT cases) and 12 normal peritumor tissues were obtained from patients in Xinqiao Hospital (Chongqing, China) for validating the bioinformatics analysis results (XQ cohort). The age of patients from XQ cohort ranged from 22 to 76 years old and the median age was 52.3 years old and the patients' male: female sex ratio was 1.67. The tumor and peritumor tissues were collected at the time of surgery and then immediately stored at −80°C until use. Patients' clinicopathological characteristics are detailed in Table SI.
Cell culture and subcellular RNA extraction
SNB19 (authenticated by STR profiling) and SF295 (20) human GBM cell lines with <5 total passage times were obtained from the National Cancer Institute-60 (NCI-60) Cancer Cell Line Panel (dtp.cancer.gov/discovery_development/nci-60/default.htm), whereas LN229 and T98G glioma cells were obtained from ATCC. All cell lines were cultured as previously described (21). The total RNA was extracted using TRIzol® reagent (Invitrogen; Thermo Fisher Scientific Inc.), and subcellular RNA was extracted using RNA Subcellular Isolation kit (Active Motif, Inc.) according to the manufacturer's protocol.
Bone morphogenetic protein 2 (BMP2) and Myostatin (MSTN) cytokine treatment of T98G cells
T98G glioma cells were cultured in Minimum Essential Medium (Gibco; Thermo Fisher Scientific, Inc.) supplemented with 10% FBS (Gibco; Thermo Fisher Scientific, Inc.) at 37°C in a humified 5% CO2 atmosphere. Recombinant human BMP2 and MSTN (PeproTech, Inc.) cytokines were added to the T98G glioma cell line culture medium to achieve 10 and 50 ng/ml, respectively. The BMP2/MSTN medium was used for 4–5 days to culture T98G cells prior to harvesting the cells for the subsequent experiments.
Reverse transcription-quantitative (RT-q)PCR
Total RNA in cells or tissues was extracted by TRIzol® reagent (Invitrogen; Thermo Fisher Scientific Inc.) to perform (RT-q)PCR. For tissue RNA extraction, 50–100 mg tissues were added into 1 ml TRIzol® reagent and homogenized using Precellys 24 (Bertin Technologies). Then, 0.2 ml chloroform was added into the homogenized tissue-TRIzol mixture and incubated for 2–3 min and centrifuged at 12,000 g at 4°C for 15 min for phase separation. After that, the RNA-contained upper aqueous phase was transferred to a new RNase/DNase-free Tube and 500 µl isopropanol was added to precipitate the RNA. After centrifugation at 12,000 g at 4°C for 10 min, the supernatant was discarded and 1 ml 75% ethanol was added to wash the RNA. Finally, after the RNA pellet was air dried for 5–10 min, 20 µl RNase-free water was added to dissolve RNA pellet.
RNA reverse transcription was performed using a PrimeScript™ RT Master mix (Takara Bio, Inc.). The reaction condition was reverse transcript at 37°C for 30 min and inactivation of reverse transcriptase at 85°C for 5s. First-strand cDNA was detected using a Bio-Rad CFX96™ Real Time system using the iTaq™ universal SYBR® Green Supermix (both Bio-Rad Laboratories, Inc.) according to the manufacturer's protocol. The PCR conditions were as follows: 40 cycles of degeneration at 95°C for 5 sec, and annealing and extension at 60°C for 30 sec. mRNA relative expression levels were analyzed using the 2−ΔΔCq method (22), and GAPDH was used as a housekeeping gene for baseline expression. For absolute qPCR, lncRNA AC068643.1 qPCR product was retrieved using an AxyPrep DNA Gel Extraction kit (Axygen; Corning, Inc.) to establish a standard curve. Small nucleolar RNA SNORA16A was used as an internal reference. Sequences of PCR primers were as follows: GAPDH forward, 5′-GACTCATGACCACAGTCCATGC-3′ and reverse, 5′-AGAGGCAGGGATGATGTTCTG-3′; lncRNA AC068643.1 forward, 5′-GAAACTACAAGAAGATGCATCTGTCTT-3′ and reverse, 5′-TATCTCTCTCTTGTGTTGCACCTTTTA-3′; BMP2 forward, 5′-CAGCTTCCACCATGAAGAATCTTTGG-3′ and reverse, 5′-ATTCGGTGATGGAAACTGCTATTGTTT-3′; MSTN forward, 5′-GAAACAGCTCCTAACATCAGCAAAGAT-3′ and reverse, 5′-AGACTCTGTAGGCATGGTAATGATTGT-3′; ATPase (ATP)13A5 forward, 5′-GAGTTTGGAAAGAAACTGGACTGGAAA-3′ and reverse, 5′-GGCTTTCCCTTAGAAAAGATGAATGCT-3′; SNORA16A forward, 5′-CTTCCGCATAGCTGCTGTGGTCAA-3′ and reverse, 5′-AGTTACAACAAACAGAACGGCGACC-3′.
Identification of DE-lncRNAs/mRNAs between IDHWT and IDHMT GBMs
GBM RNASeq HTSeq Level 3 (23) Agilent-based lncRNA/mRNA profiling datasets were obtained from TCGA database. DE-lncRNA/mRNAs (DERs) were identified using R 3.4.3 (24) software and the R/Bioconductor package ‘edgeR’ (version 3.20.6). The significant DERs were filtered based on log2 fold-change ≥2/≤-2 and -log10 adjusted P-value ≥2.
Survival analysis
Survival analysis was performed in the R studio (25) and R version 3.4.3 statistical environments. R package ‘survival’ was used to determine the prognostic DERs. For univariate Cox proportional hazards regression analysis, patient survival time, status (dead or alive) and lncRNA/mRNA expression levels were combined and processed using the ‘coxph’ function to explore the association between DERs and patient overall survival (OS). DERs with a hazard ratio (HR) value >1 were defined as risk RNAs, whereas those with HR <1 were defined as protective RNAs. DE-lncRNAs with P<0.001 and DE-mRNAs with P<0.01 were selected as survival-associated DERs. Similarly, for multivariate Cox regression analysis, the ‘coxph’ and ‘step’ functions from the R package ‘survival’ were used to determine the key lncRNAs/mRNAs associated with the survival of patients with GBM (survival-related DERs). Risk scores were established using the survival-related DERs, which were weighted using the regression coefficients in the multivariate Cox regression model, and a prognostic lncRNA/mRNA signature was constructed. Kaplan-Meier curves (using GraphPad Prism version 6.00 for Windows, graphpad.com/) were used to estimate the survival for patients between low/high-risk lncRNA/mRNA groups.
Integrative prediction analysis of lncRNA function
Pearson's correlation analysis was used to evaluate the co-expression relationship between prognostic lncRNAs and mRNAs. Correlation coefficient r>0.4 was considered as co-expressed, and co-expressed mRNAs were used to predict the function of lncRNAs using The Database for Annotation, Visualization and Integrated Discovery (david.ncifcrf.gov/). Functional enrichment results were limited to Gene Ontology (GO) terms, including biological process, cell components and molecular function, and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway categories. The presented results include the top 10 significantly enriched GO terms and top five enriched KEGG terms.
Statistical analysis
Data are presented as the mean ± standard deviation. Unpaired Students' t-test (two-tailed) was used to analyze the differences between the means of two groups. Pearson correlation analysis was used to establish the correlation between RNA expression levels. A Kruskal-Wallis test with Bonferroni's post-hoc correction was used for comparisons among multiple groups. P<0.05 was considered to indicate a statistically significant difference.
Results
Identification of prognostic DERs between IDHMT and IDHWT GBMs
To determine the key factors mediating different behaviors between IDHMT and IDHWT GBMs, DE-lncRNAs between the 122 IDHWT and 10 IDHMT GBM samples from TCGA database were investigated. Heatmap clustering analysis revealed that IDHWT and IDHMT GBMs presented distinct lncRNA expression patterns (Fig. 1A), demonstrating that IDHWT and IDHMT were distinct GBM subtypes. Using screening criteria log2 fold-change ≥1/≤-1 and -log10 adjusted P-value ≥2, 62 upregulated and 78 downregulated lncRNAs in IDHMT GBMs were identified (Fig. 1B). To identify DE-lncRNAs that served roles in GBM, the DE-lncRNAs were subjected to an univariate Cox model. The results revealed that four lncRNAs were significantly associated with the patients' OS (P<0.001), among which AC068643.1 and AC022148.1 were protective lncRNAs (HR <1), whereas Linc01776 and Linc02036 were risk lncRNAs (HR >1) (Table SII). These OS-related lncRNAs were further subjected to a multivariate Cox model and the results revealed a three-lncRNA prognostic signature that best able to predict patients' survival. The three-lncRNA prognostic signature includes one protective lncRNA (AC068643.1) and two risk-associated lncRNAs (Linc01776 and Linc02036) (Table I; Fig. 1C). Using the three-lncRNA prognostic signature, a risk score was calculated for each patient with GBM. All patients were then divided into two groups: Low-risk (n=66) and high risk (n=66) using the median risk score as the cutoff value (Fig. 1C). Kaplan-Meier survival analysis revealed that the survival time of patients in the high-risk group was significantly shorter compared with the low-risk group [P=0.004, HR=1.779, 95% confidence interval (CI), 1.246–2.852; Fig. 1D]. Therefore, the lncRNA profile, COX and survival analyses identified three prognostic lncRNAs that may serve roles in determining the distinct properties of IDHWT and IDHMT GBM.
 | Table I.Detailed information of the prognostic three-lncRNA signature from The Cancer Genome Atlas database. |
Previous studies have demonstrated that lncRNAs act through mRNA interaction (8,26); therefore, the present study sought to identify the target mRNAs that interacted with the three prognostic signature lncRNAs. The DE-mRNAs in the 122 IDHWT and 10 IDHMT GBM samples were also explored. Heatmap clustering revealed that IDHMT GBMs exhibited distinct mRNA expression patterns compared with IDHWT GBMs (Fig. 2A). A total of 142 upregulated and 219 downregulated lncRNAs were identified in IDHMT GBMs (Fig. 2B). Univariate Cox model analysis of DE-mRNAs revealed 47 survival-related mRNAs, including 10 protective and 37 risk mRNAs (P<0.01; Table SIII). In addition, multivariate Cox regression analysis identified a fifteen-mRNA prognostic signature that predicted the survival of patients with GBM, including four protective mRNAs: BMP2, MSTN, ATP13A5, Histone cluster 3 H2a and 11 risk-associated mRNAs: Carboxylesterase 1 (CES1), G protein-coupled receptor 1 (GPR1), Testis expressed 261, WAP four-disulfide core domain 2 (WFDC2), Salute carrier family 16 member 11, Desmin, AHNAK nucleoprotein 2 (AHNAK2), DNA damage inducible transcript 4 like, SH3 and cysteine rich domain 2 and Pentraxin 3 (PTX3), as presented in Fig. 2C and Table II. Patients were also divided into low/high-risk groups according to the median risk score. Kaplan-Meier survival analysis revealed that the OS time of patients in the high-risk group was significantly shorter compared with the low-risk group (P<0.001; HR=3.284; 95% CI, 2.972–7.072; Fig. 2D). Therefore, the mRNA profile analysis revealed 15 prognostic mRNAs that may serve roles in determining the distinct properties of IDHWT and IDHMT GBMs.
| Table II.Detailed information of the prognostic fifteen-mRNA signature from The Cancer Genome Atlas database. |
Cox regression analysis was performed to determine if the prognostic capacities of the three-lncRNA and fifteen-mRNA signatures were independent of the other clinical variables in patients with GBM, including age, sex, KPS and IDH mutation status. As presented in Table III, patient IDH mutation status combined with the three-lncRNA and fifteen-mRNA signature significantly predicted the survival of patients with GBM in the univariate Cox model. In the multivariate Cox model, the three-lncRNA and the fifteen-mRNA signatures remained significant prognostic markers for patients with GBM (Table III). These results suggested that the three-lncRNA and fifteen-mRNA signatures may serve roles in mediating the different behaviors observed in IDHMT and IDHWT GBMs and contribute to the distinct prognosis of these patients.
| Table III.Univariate and Multivariate Cox Regression analyses of the prognostic three-lncRNA and fifteen-mRNA signatures. |
Association between the three-lncRNA and the fifteen-mRNA signatures
First, the correlations between the expression of the three lncRNAs and the 19,676 mRNAs from the whole genome were investigated. Pearson's correlation analysis identified 337 mRNAs co-expressed with the three lncRNAs (r>0.4), 187 of which were co-expressed with AC068643.1, 145 with Linc02036 and 5 with Linc01776 (Fig. 3A-C). Functional enrichment analyses revealed that these co-expressed mRNAs were significantly enriched in 19 GO terms and 5 KEGG pathways (P<0.05; Fig. 3D and E), which were primarily involved in regulating stem cell pluripotency, Hippo and Wnt signaling pathways and cell differentiation.
The mRNAs co-expressed with the fifteen signature DE-mRNAs were analyzed; the protective DE-lncRNA AC068643.1 was significantly correlated with three protective signature mRNAs: BMP2 (r=0.403; P<0.001), MSTN (r=0.415; P<0.001) and ATP13A5 (r=0.471; P<0.001) as presented in Figs. 3A and S1A-C. BMP2 and MSTN both belong to the BMP-signaling pathway (27), thus it was predicted that AC068643.1 may function through interaction with the BMP signaling pathway. qPCR analysis of GBM samples from the XQ cohort demonstrated that protective lncRNA AC068643.1, BMP2 and MSTN were upregulated in IDHMT GBMs (Fig. 4A-C), and AC068643.1 was significantly positively correlated with BMP2 (r=0.427; P=0.004) and MSTN (r=0.395; P=0.010; Fig. 4D and E). RNA profiling data from the CGGA database confirmed that BMP2 and MSTN expression levels were upregulated in IDHMT compared with IDHWT GBMs, predicting a favorable prognosis for these patients (Fig. S1D-G). These results indicated an association between AC068643.1, BMP2 and MSTN.
Following treatment with BMP2 and MSTN cytokines in cell culture medium, AC068643.1 expression levels were significantly upregulated in T98G cells treated with different concentrations (0, 10 and 50 ng/ml) of BMP2 or MSTN (Fig. 4F). These results suggested that upregulated BMP signaling may promote IDHMT GBM differentiation through promoting AC068643.1 expression.
Functional analysis of lncRNA AC068643.1
To investigate the function of AC068643.1, GO and KEGG analysis of the 187 co-expressed mRNAs of AC068643.1 were performed. The results suggested that AC068643.1 was associated with ‘mitochondrial matrix’, ‘regulation of transcription, DNA-templated’, ‘nucleic acid binding’ and ‘regulation of pluripotency of stem cells’ (Fig. S2A and B). In addition, subcellular expression of AC068643.1 was detected using different glioma cell lines, including SF295, SF268, LN229 and T98G, and the results demonstrated that AC068643.1 exhibited low expression levels and was primarily expressed in the glioma cell nucleus (Fig. 4G), which was consistent with the gene enrichment analysis.
Discussion
The discovery of IDH mutations changed the classification system of glioma, leading to the acknowledgement of the IDHWT and IDHMT subtypes (6). Studies investigating the role of IDH mutations in IDHMT gliomas have aimed to treat this glioma subtype through IDH mutation-targeted treatment (28,29), whereas the mechanisms underlying IDHWT glioma, which exhibits a higher degree of malignancy and accounts for >70% of all GBMs (5), have been somewhat neglected. The present study aimed to explore the altered lncRNA/mRNAs expression levels between IDHWT and IDHMT gliomas, the results of which may provide insight into understanding the biological behaviors of IDHWT GBM.
The present study utilized comprehensive Agilent RNA-profiling data containing ~10,000 lncRNAs in each sample. IDHMT GBMs only constitute a small proportion of GBMs cases (~5.6%) (5), and thus data for only 10 IDHMT GBM cases were obtained from TCGA, compared with 122 cases of IDHWT GBM, which is a limitation of the present study. A total of 140 DE-lncRNAs and 361 DE-mRNAs between IDHWT and IDHMT GBMs were identified. Three-lncRNA and fifteen-mRNA signatures with prognostic value were identified, suggesting that these lncRNAs and mRNAs may function in determining the distinct properties between IDHWT and IDHMT GBMs. Gene enrichment analysis revealed that the three-lncRNA signature was associated with cell differentiation-associated KEGG pathways and GO terms. By contrast, seven mRNAs from the fifteen-mRNA signature, including: five risk mRNAs, CES1 (30), GPR1 (31), WFDC2 (32), AHNAK2 (33) and PTX3 (34), and two protective mRNAs, BMP2 (35) and MSTN (36), were associated with cell stemness or cell differentiation. These results suggested that cell differentiation/stemness status may be the primary distinct property between IDHWT and IDHMT GBMs.
Previous studies have demonstrated that tumor stem cells are responsible for tumor growth, chemoresistance and relapse, and that tumor stemness is associated with poor patient survival outcomes (37,38). The results of the present study indicated that the distinct tumor behaviors and clinical features between IDHWT and IDHMT GBMs may be partially due to the different cell differentiation status. A recent study has reported that the stem cell markers nestin and CD133 are significantly upregulated in IDHWT compared with IDHMT gliomas (39). Overexpression of IDH mutant protein in glioma stem cells downregulates nestin and CD133 expression levels and promotes glioma stem cell differentiation via the Wnt/β-catenin pathway (39), providing further evidence that varied differentiation/stemness status between IDHWT and IDHMT GBMs may be a contributing factor for their distinct behaviors.
BMP2 and MSTN belong to the BMP signaling pathway, which can prevent cell self-renewal and enhance differentiation (27). In the present study, the expression levels of AC068643.1 were significantly positively correlated with those of BMP2 and MSTN, suggesting that there may be a potential positive regulatory axis between AC068643.1 and the BMP signaling pathway responsible for promoting glioma differentiation. In vitro treatment of T98G cells with BMP2 and MSTN significantly upregulated AC068643.1 expression, suggesting that AC068643.1 expression level is regulated by the BMP signaling pathway. Future studies should focus on exploring the effects of AC068643.1 knockdown or overexpression in the BMP signaling pathway and glioma differentiation.
In conclusion, by comparing the transcriptome profiling data between IDHMT and IDHWT GBMs, the present study identified stemness-related three-lncRNA and fifteen-mRNA signatures, suggesting that cell differentiation or stemness status may be a primary property which is distinct between IDHWT and IDHMT GBMs. Protective lncRNA AC068643.1 was significantly upregulated in IDHMT GBMs and was positively regulated by BMP2 and MSTN, which suggested that AC068643.1 overexpression may prevent glioma self-renewal and enhance cell differentiation. These results may aid our understanding of the more differentiated status and the more favorable prognosis of IDHMT compared with IDHWT GBMs and may provide a basis for IDHWT GBM treatment by inducing lncRNA AC068643.1 overexpression.
Supplementary Material
Supporting Data
Acknowledgements
The authors would like to thank Miss. Qing-Rui Li (Department of Pathology Southwest Hospital, Third Military Medical University) for her work in collecting clinical samples and patients' follow-up and Professor Shuli Xia (Department of Neurology, Johns Hopkins School of Medicine) for her supervison and revision of this study.
Funding
The present study was supported by the National Natural Science Foundation of China (grant no. NSFC:81972360).
Availability of data and materials
The datasets used and/or analyzed during the present study are available from the corresponding author on reasonable request.
Authors' contributions
GHH, SQL and JL conceived and designed the experiments. GHH, YCP, LY, KJM and YX performed the experiments and data statistics. GHH and JHT discussed and interpreted the data. GHH wrote the manuscript. SQL supervised the study. All authors read and approved the final manuscript.
Ethics approval and consent to participate
The present study was approved by The Medical Ethics Committee of the Second Affiliated Hospital of the Third Military Medical University.
Patient consent for publication
Not applicable.
Competing interests
The authors declare that they have no competing interests.
References
|
Louis DN, Perry A, Reifenberger G, von Deimling A, Figarella-Branger D, Cavenee WK, Ohgaki H, Wiestler OD, Kleihues P and Ellison DW: The 2016 World Health organization classification of tumors of the central nervous system: A summary. Acta Neuropathol. 131:803–820. 2016. View Article : Google Scholar : PubMed/NCBI | |
|
Stupp R, Mason WP, van den Bent MJ, Weller M, Fisher B, Taphoorn MJ, Belanger K, Brandes AA, Marosi C, Bogdahn U, et al: Radiotherapy plus concomitant and adjuvant temozolomide for glioblastoma. N Engl J Med. 352:987–996. 2005. View Article : Google Scholar : PubMed/NCBI | |
|
Gilbert MR, Dignam JJ, Armstrong TS, Wefel JS, Blumenthal DT, Vogelbaum MA, Colman H, Chakravarti A, Pugh S, Won M, et al: A randomized trial of bevacizumab for newly diagnosed glioblastoma. N Engl J Med. 370:699–708. 2014. View Article : Google Scholar : PubMed/NCBI | |
|
Vredenburgh JJ, Desjardins A, Herndon JE II, Dowell JM, Reardon DA, Quinn JA, Rich JN, Sathornsumetee S, Gururangan S, Wagner M, et al: Phase II trial of bevacizumab and irinotecan in recurrent malignant glioma. Clin Cancer Res. 13:1253–1259. 2007. View Article : Google Scholar : PubMed/NCBI | |
|
Yan H, Parsons DW, Jin G, McLendon R, Rasheed BA, Yuan W, Kos I, Batinic-Haberle I, Jones S, Riggins GJ, et al: IDH1 and IDH2 mutations in gliomas. N Engl J Med. 360:765–773. 2009. View Article : Google Scholar : PubMed/NCBI | |
|
Turkalp Z, Karamchandani J and Das S: IDH mutation in glioma: New insights and promises for the future. JAMA Neurol. 71:1319–1325. 2014. View Article : Google Scholar : PubMed/NCBI | |
|
Wang RA: MTA1-a stress response protein: A master regulator of gene expression and cancer cell behavior. Cancer Metastasis Rev. 33:1001–1009. 2014. View Article : Google Scholar : PubMed/NCBI | |
|
Wang KC and Chang HY: Molecular mechanisms of long noncoding RNAs. Mol Cell. 43:904–914. 2011. View Article : Google Scholar : PubMed/NCBI | |
|
Carninci P, Kasukawa T, Katayama S, Gough J, Frith MC, Maeda N, Oyama R, Ravasi T, Lenhard B, Wells C, et al: The transcriptional landscape of the mammalian genome. Science. 309:1559–1563. 2005. View Article : Google Scholar : PubMed/NCBI | |
|
Mercer TR, Dinger ME, Sunkin SM, Mehler MF and Mattick JS: Specific expression of long noncoding RNAs in the mouse brain. Proc Natl Acad Sci USA. 105:716–721. 2008. View Article : Google Scholar : PubMed/NCBI | |
|
Qureshi IA, Mattick JS and Mehler MF: Long non-coding RNAs in nervous system function and disease. Brain Res. 1338:20–35. 2010. View Article : Google Scholar : PubMed/NCBI | |
|
Cuevas-Diaz Duran R, Wei H, Kim DH and Wu JQ: Invited Review: Long non-coding RNAs: Important regulators in the development, function and disorders of the central nervous system. Neuropathol Appl Neurobiol. 45:538–556. 2019. View Article : Google Scholar : PubMed/NCBI | |
|
Zhang XQ and Leung GK: Long non-coding RNAs in glioma: Functional roles and clinical perspectives. Neurochem Int. 77:78–85. 2014. View Article : Google Scholar : PubMed/NCBI | |
|
Ramos AD, Attenello FJ and Lim DA: Uncovering the roles of long noncoding RNAs in neural development and glioma progression. Neurosci Lett. 625:70–79. 2016. View Article : Google Scholar : PubMed/NCBI | |
|
Han L, Zhang K, Shi Z, Zhang J, Zhu J, Zhu S, Zhang A, Jia Z, Wang G, Yu S, et al: LncRNA profile of glioblastoma reveals the potential role of lncRNAs in contributing to glioblastoma pathogenesis. Int J Oncol. 40:2004–2012. 2012.PubMed/NCBI | |
|
Chen Q, Cai J, Wang Q, Wang Y, Liu M, Yang J, Zhou J, Kang C, Li M and Jiang C: Long Noncoding RNA NEAT1, Regulated by the EGFR pathway, contributes to glioblastoma progression through the WNT/β-catenin pathway by scaffolding EZH2. Clin Cancer Res. 24:684–695. 2018. View Article : Google Scholar : PubMed/NCBI | |
|
Pastori C, Kapranov P, Penas C, Peschansky V, Volmar CH, Sarkaria JN, Bregy A, Komotar R, St Laurent G, Ayad NG, et al: The Bromodomain protein BRD4 controls HOTAIR, a long noncoding RNA essential for glioblastoma proliferation. Proc Natl Acad Sci USA. 112:8326–8331. 2015. View Article : Google Scholar : PubMed/NCBI | |
|
Zhang JX, Han L, Bao ZS, Wang YY, Chen LY, Yan W, Yu SZ, Pu PY, Liu N, You YP, et al: HOTAIR, a cell cycle-associated long noncoding RNA and a strong predictor of survival, is preferentially expressed in classical and mesenchymal glioma. Neuro Oncol. 15:1595–1603. 2013. View Article : Google Scholar : PubMed/NCBI | |
|
Mineo M, Ricklefs F, Rooj AK, Lyons SM, Ivanov P, Ansari KI, Nakano I, Chiocca EA, Godlewski J and Bronisz A: The Long Non-coding RNA HIF1A-AS2 facilitates the maintenance of mesenchymal glioblastoma stem-like cells in hypoxic niches. Cell Rep. 15:2500–2509. 2016. View Article : Google Scholar : PubMed/NCBI | |
|
Rutka JT, Giblin JR, Dougherty DY, Liu HC, McCulloch JR, Bell CW, Stern RS, Wilson CB and Rosenblum ML: Establishment and characterization of five cell lines derived from human malignant gliomas. Acta Neuropathol. 75:92–103. 1987. View Article : Google Scholar : PubMed/NCBI | |
|
Tang JH, Ma ZX, Huang GH, Xu QF, Xiang Y, Li N, Sidlauskas K, Zhang EE and Lv SQ: Downregulation of HIF-1a sensitizes U251 glioma cells to the temozolomide (TMZ) treatment. Exp Cell Res. 343:148–158. 2016. View Article : Google Scholar : PubMed/NCBI | |
|
Livak KJ and Schmittgen TD: Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) method. Methods. 25:402–408. 2001. View Article : Google Scholar : PubMed/NCBI | |
|
Zhu Y, Qiu P and Ji Y: TCGA-assembler: Open-source software for retrieving and processing TCGA data. Nat Methods. 11:599–600. 2014. View Article : Google Scholar : PubMed/NCBI | |
|
R Core Team, . R: A language and environment for statistical computing. R Foundation for Statistical Computing Vienna, Austria: 2012, ISBN 3-900051-07-0, URL http://www.R-project.org/. | |
|
RStudio Team, . RStudio: Integrated Development for R. RStudio, Inc.; Boston, MA: 2015, URL http://www.rstudio.com/. | |
|
Zhang Y, He W and Zhang S: Seeking for correlative genes and signaling pathways with bone metastasis from breast cancer by integrated analysis. Front Oncol. 9:1382019. View Article : Google Scholar : PubMed/NCBI | |
|
Caja L, Bellomo C and Moustakas A: Transforming growth factor beta and bone morphogenetic protein actions in brain tumors. FEBS Lett. 589:1588–1597. 2015. View Article : Google Scholar : PubMed/NCBI | |
|
Rohle D, Popovici-Muller J, Palaskas N, Turcan S, Grommes C, Campos C, Tsoi J, Clark O, Oldrini B, Komisopoulou E, et al: An inhibitor of mutant IDH1 delays growth and promotes differentiation of glioma cells. Science. 340:626–630. 2013. View Article : Google Scholar : PubMed/NCBI | |
|
Pusch S, Krausert S, Fischer V, Balss J, Ott M, Schrimpf D, Capper D, Sahm F, Eisel J, Beck AC, et al: Pan-mutant IDH1 inhibitor BAY 1436032 for effective treatment of IDH1 mutant astrocytoma in vivo. Acta Neuropathol. 133:629–644. 2017. View Article : Google Scholar : PubMed/NCBI | |
|
Seno A, Kasai T, Ikeda M, Vaidyanath A, Masuda J, Mizutani A, Murakami H, Ishikawa T and Seno M: Characterization of gene expression patterns among artificially developed cancer stem cells using spherical self-organizing map. Cancer Inform. 15:163–178. 2016. View Article : Google Scholar : PubMed/NCBI | |
|
Tummler C, Snapkov I, Wickstrom M, Moens U, Ljungblad L, Maria Elfman LH, Winberg JO, Kogner P, Johnsen JI and Sveinbjørnsson B: Inhibition of chemerin/CMKLR1 axis in neuroblastoma cells reduces clonogenicity and cell viability in vitro and impairs tumor growth in vivo. Oncotarget. 8:95135–95151. 2017. View Article : Google Scholar : PubMed/NCBI | |
|
Hofman VJ, Moreilhon C, Brest PD, Lassalle S, Le Brigand K, Sicard D, Raymond J, Lamarque D, Hébuterne XA, Mari B, et al: Gene expression profiling in human gastric mucosa infected with Helicobacter pylori. Mod Pathol. 20:974–989. 2007. View Article : Google Scholar : PubMed/NCBI | |
|
Wang M, Li X, Zhang J, Yang Q, Chen W, Jin W, Huang YR, Yang R and Gao WQ: AHNAK2 is a novel prognostic marker and oncogenic protein for clear cell renal cell carcinoma. Theranostics. 7:1100–1113. 2017. View Article : Google Scholar : PubMed/NCBI | |
|
Thomas C, Henry W, Cuiffo BG, Collmann AY, Marangoni E, Benhamo V, Bhasin MK, Fan C, Fuhrmann L, Baldwin AS, et al: Pentraxin-3 is a PI3K signaling target that promotes stem cell-like traits in basal-like breast cancers. Sci Signal. 10(pii): eaah46742017. View Article : Google Scholar : PubMed/NCBI | |
|
Persano L, Pistollato F, Rampazzo E, Della Puppa A, Abbadi S, Frasson C, Volpin F, Indraccolo S, Scienza R and Basso G: BMP2 sensitizes glioblastoma stem-like cells to Temozolomide by affecting HIF-1α stability and MGMT expression. Cell Death Dis. 3:e4122012. View Article : Google Scholar : PubMed/NCBI | |
|
Di Fiore R, Fanale D, Drago-Ferrante R, Chiaradonna F, Giuliano M, De Blasio A, Amodeo V, Corsini LR, Bazan V, Tesoriere G and Vento R: Genetic and molecular characterization of the human osteosarcoma 3AB-OS cancer stem cell line: A possible model for studying osteosarcoma origin and stemness. J Cell Physiol. 228:1189–1201. 2013. View Article : Google Scholar : PubMed/NCBI | |
|
Esparza R, Azad TD, Feroze AH, Mitra SS and Cheshier SH: Glioblastoma stem cells and stem cell-targeting immunotherapies. J Neurooncol. 123:449–457. 2015. View Article : Google Scholar : PubMed/NCBI | |
|
Sen Z, Zhan XK, Jing J, Yi Z and Wanqi Z: Chemosensitizing activities of cyclotides from Clitoria ternatea in paclitaxel-resistant lung cancer cells. Oncol Lett. 5:641–644. 2013. View Article : Google Scholar : PubMed/NCBI | |
|
Yao Q, Cai G, Yu Q, Shen J, Gu Z, Chen J, Shi W and Shi J: IDH1 mutation diminishes aggressive phenotype in glioma stem cells. Int J Oncol. 52:270–278. 2018.PubMed/NCBI |
