The oligodendrocyte lineage transcription factor 2 (OLIG2) is epigenetically regulated in acute myeloid leukemia
2017; Elsevier BV; Volume: 55; Linguagem: Inglês
10.1016/j.exphem.2017.07.009
ISSN1873-2399
AutoresArzu Yalcin, Marlon Kovarbasic, Julius Wehrle, Rainer Claus, Heiko Becker, Mahmoud Abdelkarim, Verena I. Gaidzik, Andrea Schmidts, Ralph Wäsch, Heike L. Pahl, Konstanze Döhner, Lars Bullinger, Justus Duyster, Michael Lübbert, Björn Hackanson,
Tópico(s)Acute Lymphoblastic Leukemia research
Resumo•DNA methylation of oligodendrocyte lineage transcription factor 2 (OLIG2) results in downregulation of OLIG2 mRNA/protein in acute myeloid leukemia (AML).•OLIG2 mRNA/protein can be re-expressed with demethylating treatment.•Overexpression of OLIG2 leads to moderate growth inhibition in AML cells.•AML M3 harboring t(15;17) expresses OLIG mRNA and protein. DNA methylation differences between normal tissue and cancerous tissue resulting in differential expression of genes are a hallmark of acute myeloid leukemia (AML) and can provide malignant cells with a growth advantage via silencing of specific genes, for example, transcription factors. Oligodendrocyte lineage transcription factor 2 (OLIG2) was reported to be differentially methylated and associated with prognosis in AML and, as reported for acute lymphoblastic leukemia and malignant glioma, may play a role in malignant transformation. We report that DNA methylation of OLIG2 is associated with decreased expression of mRNA in AML cell lines and patients. Moreover, in cell lines, decreased mRNA expression also translated into decreased OLIG2 protein expression. Treatment of non-expressing cell lines PL-21 and U-937 with the demethylating agent decitabine resulted in robust re-expression of OLIG2 on mRNA and protein levels. Furthermore, stable overexpression of OLIG2 in non-expressing cell lines Kasumi-1 and U-937, using a lentiviral vector system, led to moderate growth inhibition after 4 days and resulted in signs of differentiation in U-937 cells. Interestingly, although CD34 + cells from healthy donors and 10 of 12 AML patients exhibited no protein expression, OLIG2 was expressed in two patients, both bearing the translocation t(15;17), corresponding to OLIG2 expression in NB-4 cells, also harboring t(15;17). In conclusion, we provide first evidence that OLIG2 is epigenetically regulated via DNA methylation and expressed in a subset of AML patients. OLIG2 may exert antiproliferative activity in leukemia cell lines, and its potential leukemia-suppressing role in AML warrants further investigation. DNA methylation differences between normal tissue and cancerous tissue resulting in differential expression of genes are a hallmark of acute myeloid leukemia (AML) and can provide malignant cells with a growth advantage via silencing of specific genes, for example, transcription factors. Oligodendrocyte lineage transcription factor 2 (OLIG2) was reported to be differentially methylated and associated with prognosis in AML and, as reported for acute lymphoblastic leukemia and malignant glioma, may play a role in malignant transformation. We report that DNA methylation of OLIG2 is associated with decreased expression of mRNA in AML cell lines and patients. Moreover, in cell lines, decreased mRNA expression also translated into decreased OLIG2 protein expression. Treatment of non-expressing cell lines PL-21 and U-937 with the demethylating agent decitabine resulted in robust re-expression of OLIG2 on mRNA and protein levels. Furthermore, stable overexpression of OLIG2 in non-expressing cell lines Kasumi-1 and U-937, using a lentiviral vector system, led to moderate growth inhibition after 4 days and resulted in signs of differentiation in U-937 cells. Interestingly, although CD34 + cells from healthy donors and 10 of 12 AML patients exhibited no protein expression, OLIG2 was expressed in two patients, both bearing the translocation t(15;17), corresponding to OLIG2 expression in NB-4 cells, also harboring t(15;17). In conclusion, we provide first evidence that OLIG2 is epigenetically regulated via DNA methylation and expressed in a subset of AML patients. OLIG2 may exert antiproliferative activity in leukemia cell lines, and its potential leukemia-suppressing role in AML warrants further investigation. DNA hypermethylation has been established as a key event in malignant transformation and is associated with transcriptional gene silencing [1Esteller M. Epigenetics in cancer.N Engl J Med. 2008; 358: 1148-1159Crossref PubMed Scopus (2897) Google Scholar]. Particularly in acute myeloid leukemia (AML), genomewide and gene-specific methylation analyses have revealed numerous genes, including transcription factors that are altered by aberrant DNA methylation [2Figueroa M.E. Lugthart S. Li Y. et al.DNA methylation signatures identify biologically distinct subtypes in acute myeloid leukemia.Cancer Cell. 2010; 17: 13-27Abstract Full Text Full Text PDF PubMed Scopus (668) Google Scholar, 3Bullinger L. Ehrich M. Döhner K. et al.Quantitative DNA methylation predicts survival in adult acute myeloid leukemia.Blood. 2010; 115: 636-642Crossref PubMed Scopus (130) Google Scholar, 4Yalcin A. Kreutz C. Pfeifer D. et al.MeDIP coupled with a promoter tiling array as a platform to investigate global DNA methylation patterns in AML cells.Leuk Res. 2013; 37: 102-111Abstract Full Text Full Text PDF PubMed Scopus (14) Google Scholar]. Given the pharmacologic reversibility of DNA methylation by hypomethylating agents, such as 5-azacytidine and decitabine, and the comparably low toxicity rate of these drugs, epigenetic therapy harbors considerable significance among novel AML treatment strategies. In fact, today, both drugs are a standard of care for older AML patients [5National Comprehensive Cancer Network NCCN clinical practice guidelines in oncology. Acute myeloid leukemia V1.http://www.nccn.org/professionals/physician_gls/pdf/aml.pdfDate: 2015Google Scholar, 6Döhner H. Estey E.H. Amadori S. et al.Diagnosis and management of acute myeloid leukemia in adults: recommendations from an international expert panel, on behalf of the European Leukemia Net.Blood. 2010; 115: 453-474Crossref PubMed Scopus (2613) Google Scholar]. Recently, several groups, including ours, have reported on differential DNA methylation of the central nervous system-restricted oligodendrocyte lineage transcription factor 2 (OLIG2) in myelodysplasia and AML patients and on OLIG2 hypermethylation's potentially adverse effect on outcome [7Kroeger H. Jelinek J. Estécio M.R. et al.Aberrant CpG island methylation in acute myeloid leukemia is accentuated at relapse.Blood. 2008; 112: 1366-1373Crossref PubMed Scopus (129) Google Scholar, 8Shen L. Kantarjian H. Guo Y. et al.DNA methylation predicts survival and response to therapy in patients with myelodysplastic syndromes.J Clin Oncol. 2010; 28: 605-613Crossref PubMed Scopus (304) Google Scholar, 9Hiller J.K. Schmoor S. Gaidzik V.I. et al.Evaluating the impact of genetic and epigenetic aberrations on survival and response in acute myeloid leukemia patients receiving epigenetic therapy.Ann Hematol. 2017; 96: 559-565Crossref PubMed Scopus (19) Google Scholar]. OLIG2 is a basic helix–loop–helix (bHLH) transcription factor that is crucial for differentiation and proliferation of neuroepithelial progenitor cells. OLIG2 also bears substantial oncogenic potential, as it has been found that overexpression of the protein is a key factor in the development and maintenance of malignant glioma [10Mehta S. Huillard E. Kesari S. et al.The central nervous system-restricted transcription factor Olig2 opposes p53 responses to genotoxic damage in neural progenitors and malignant glioma.Cancer Cell. 2011; 19: 359-371Abstract Full Text Full Text PDF PubMed Scopus (125) Google Scholar]. Interestingly, aberrant expression of OLIG2 outside the central nervous system has rarely been reported, but was detected in a few cases of T-cell acute lymphoblastic leukemia (ALL) and on the mRNA level, in AML cell lines HEL and HL60 [11Wang J. Jani-Sait S.N. Escalon E.A. et al.The t(14;21)(q11.2;q22) chromosomal translocation associated with T-cell acute lymphoblastic leukemia activates the BHLHB1 gene.Proc Natl Acad Sci USA. 2009; 97: 3497-3502Crossref Scopus (76) Google Scholar, 12Lin Y.W. Deveney R. Barbara M. et al.OLIG2 (BHLHB1), a bHLH transcription factor, contributes to leukemogenesis in concert with LMO1.Cancer Res. 2005; 65: 7151-7158Crossref PubMed Scopus (29) Google Scholar]. Moreover, Teneng et al. recently reported that OLIG2 is epigenetically silenced in lung cancer cell lines and that re-expression led to a significant growth reduction of non-small cell lung cancer cells [13Teneng I. Tellez C.S. Picchi M.A. et al.Global identification of genes targeted by DNMT3b for epigenetic silencing in lung cancer.Oncogene. 2015; 34: 621-630Crossref PubMed Scopus (28) Google Scholar]. In AML, however, functional studies on the impact of OLIG2, which is not expressed in normal hematopoiesis, are still lacking. Therefore, in this study, we investigated the role of OLIG2 in AML, implementing functional studies of cell lines and investigating primary patient samples. Pretreatment bone marrow (n = 70) and peripheral blood (n = 35) cells (Ficoll–Hypaque density gradient-separated mononuclear cells) from 105 AML patients were examined. DNA for methylation analysis was derived from 93 patients who had been treated within the German 00331 decitabine trial (DRKS00000069); protein for OLIG2 Western blot was derived from the bone marrow of 12 patients who had been treated with conventional induction chemotherapy [9Hiller J.K. Schmoor S. Gaidzik V.I. et al.Evaluating the impact of genetic and epigenetic aberrations on survival and response in acute myeloid leukemia patients receiving epigenetic therapy.Ann Hematol. 2017; 96: 559-565Crossref PubMed Scopus (19) Google Scholar, 14Lübbert M. Rüter B.H. Claus R. et al.A multicenter phase II trial of decitabine as first-line treatment for older patients with acute myeloid leukemia judged unfit for induction chemotherapy.Haematologica. 2012; 97: 393-401Crossref PubMed Scopus (194) Google Scholar]. Control CD34 + bone marrow cells and mononuclear cells from peripheral blood samples were derived from healthy volunteer donors. The Freiburg University Medical Center institutional review board and ethics committee approved the study. All patients had given written informed consent for data and specimen collection and treatment in compliance with the Declaration of Helsinki. KG-1, U-937, Kasumi-1, Jurkat, PL-21, HL-60, MV4-11, THP-1, and NB-4 cells were purchased from DSMZ (Braunschweig, Germany), and HEL cells were kindly provided by Professor Heike Pahl, University Medical Center Freiburg. All cell lines were cytogenetically identified and cultured in RPMI-1640 medium (Gibco Life Technologies, Germany) supplemented with 100 U/ml penicillin/streptomycin and 10% to 20% fetal calf serum (PAA Laboratories) at 37°C and 5% CO2. For OLIG2 re-expression analyses, cell lines were treated with non-cytotoxic concentrations (viability >85%) of 5-aza-2′-deoxycytidine (DAC, Sigma-Aldrich, Taufkirchen, Germany) dissolved in phosphate-buffered saline. Cells were cultured for 7 days, the first 3 days of which included concentrations of DAC of 200 and 400 nmol/L, respectively. Cell count and viability were determined with the trypan blue (Biochrom AG, Berlin, Germany) exclusion test. Bisulfite pyrosequencing-based quantitative promoter DNA methylation analysis of OLIG2 was performed as previously described [4Yalcin A. Kreutz C. Pfeifer D. et al.MeDIP coupled with a promoter tiling array as a platform to investigate global DNA methylation patterns in AML cells.Leuk Res. 2013; 37: 102-111Abstract Full Text Full Text PDF PubMed Scopus (14) Google Scholar]. In summation, DNA was treated with bisulfite using an EZ DNA Methylation-Gold kit (Zymo Research, Freiburg, Germany). Primers for pyrosequencing were designed using Pyrosequencing Assay Design Software and in keeping with the literature [4Yalcin A. Kreutz C. Pfeifer D. et al.MeDIP coupled with a promoter tiling array as a platform to investigate global DNA methylation patterns in AML cells.Leuk Res. 2013; 37: 102-111Abstract Full Text Full Text PDF PubMed Scopus (14) Google Scholar, 8Shen L. Kantarjian H. Guo Y. et al.DNA methylation predicts survival and response to therapy in patients with myelodysplastic syndromes.J Clin Oncol. 2010; 28: 605-613Crossref PubMed Scopus (304) Google Scholar]. Sequences of polymerase chain reaction (PCR) and sequencing primers and conditions are listed in Supplementary Table E1 (online only, available at www.exphem.org). Pyrosequencing analyses were performed in triplicate, and data were analyzed using the Pyro Q-CpG software. All assays were controlled with fully methylated (100%) and unmethylated (0%) genomic DNA. RNA was isolated from mononuclear AML cells using Trizol reagent, and RNA quality was assessed by gel electrophoresis. RNA samples were hybridized to Affymetrix U133 plus 2.0 arrays according to Affymetrix standard protocol. The raw data were processed and normalized with the robust multichip average (RMA) algorithm using R software in conjunction with the Affymetrix library and log2 transformed. RNA from cell lines was extracted using the RNAeasy Mini Kit (Qiagen, Hilden, Germany), according to the manufacturer's instructions. First-strand cDNA synthesis was carried out using SuperScriptIII First-Strand Synthesis System (Life Technologies). Sequences of OLIG2 gene-specific and β-actin control primers are listed in Supplementary Table E1. Quantitative PCR (qPCR) was performed using LightCycler 480 SYBR Green I Master (Roche, Switzerland); amplification and detection were carried out using a LightCycler 480 (Roche). The PCR cycling program consisted of 45 three-step cycles of 10 s/95°C, 15 s/60°C, and 20 s/72°C. To confirm signal specificity, a melting curve program was implemented after the PCR cycles were completed. Quantitative PCR experiments were performed in triplicate. For Western blot analyses, cells lines were lysed using a nuclear extraction kit (Active Motif), per the manufacturer's instructions. Bone marrow samples of AML patients and CD34 + cells of healthy donors were lysed in radio-immunoprecipitation buffer (50 mmol/L Tris-HCl, pH 8.0, 150 mmol/L NaCl, 1% Nonidet P-40, 0.1% sodium dodecyl sulfate [SDS], 0.5% deoxycholate, and complete Protease Inhibitor Cocktail Tablet [Roche]). Protein concentrations were determined with the Pierce BCA Protein Assay kit (Thermo Scientific). A Spectra Multicolor Broad Range Protein Ladder was used to determine molecular weights. Thirty micrograms of protein per sample were loaded on a 10% SDS gel and electro-transferred to 0.45-µm nitrocellulose membranes. After 1 hour of blocking with 5% non-fat dry milk in Tris-buffered saline and Tween 20 (TBS-T), membranes were incubated overnight at 4°C with OLIG2 antibody (1:500; sc-48817, Santa Cruz Biotechnology). After being washed with TBS-T, the blot was incubated for 1 hour at room temperature with the secondary antibody (anti-mouse [sc-2005]) diluted at 1:10,000 in 5% non-fat dry milk/TBS-T. After a final TBS-T wash, the blots were developed with the Advansta WesternBright Quantum Chemilumineszenz substrate and exposed to film or visualized with an Intas Imager (Intas, Göttingen, Germany). Western blot experiments were performed in duplicate. Densitometry to quantify protein expression in NB4 cells after OLIG2 siRNA knockdown (Supplementary Figure E2, online only, available at www.exphem.org) was performed using LabImage 1D (Kapelan Bio-Imaging). First, GAPDH and OLIG2 bands were measured, and then, OLIG2 densitometry results were divided by GAPH densitometry results. To generate OLIG2-overexpressing lentiviral constructs, RNA of OLIG2-expressing cell line NB-4 was extracted and first-strand cDNA-synthesis was performed as explained above. OLIG2 cDNA was amplified using primers for generating restriction sites (Supplementary Table E1). PCR products were purified with the GeneJET Gel Extraction kit (Thermo Scientific) and ligated into the pLeGO-iG-hU6 vector using EcoRI and StuI restriction sites [15Weber K. Bartsch U. Stocking C. et al.Multicolor panel of novel lentiviral "gene ontology" (LeGO) vectors for functional gene analysis.Mol Ther. 2008; 16: 698-706Abstract Full Text Full Text PDF PubMed Scopus (253) Google Scholar, 16Roelz R. Pilz I. Mutschler M. et al.Of mice and men: Human RNA polymerase III promoter U6 is more efficient than its murine homologue for shRNA expression from a lentiviral vector in both human and murine progenitor cells.Exp Hematol. 2010; 38: 792-797Abstract Full Text Full Text PDF PubMed Scopus (32) Google Scholar]. The ligated plasmid construct was transformed into One Shot Stbl3TM Chemically Competent Escherichia coli (Invitrogen), and the insert sequence was verified by Sanger sequencing (GATC Biotech AG). Viral particles were generated by transfecting 293T cells with the cDNA vector construct, Gag/Pol (pCMV-dr8.74), and Envelope (pMD2 VSVG) plasmids via calcium phosphate transfection as described [17Kingston R.E. Chen C.A. Okayama H. Calcium phosphate transfection.Curr Protoc Cell Biol. 2001; 19 (3.1-20.3.8): 20Google Scholar]. Viral supernatants were collected on days 2, 3, and 4 after transfection, filtered (0.22 µm), ultracentrifuged (50,000g for 2.5 h), and stored at −80°C [18Cooray S. Howe S.J. Thrasher A.J. Retrovirus and lentivirus vector design and methods of cell conditioning.Methods Enzymol. 2012; 507: 29-57Crossref PubMed Scopus (48) Google Scholar]. For stable transfection, Kasumi-1 and U-937 cells were transfected using a multiplicity of infection of 0.2 without a transfection facilitator. After 5 days, cells were subjected to fluorescence-activated cell sorting (FACS) for Green Fluorescent Protein (GFP)-positive cells and kept in culture for another 5–10 days; functional analysis was subsequently performed. For growth assays, Kasumi-1 and U-937 cells were seeded on day 0 (105 cells/mL) as triplicates in 1.5 mL of growth medium according to the manufacturer's recommendations and counted daily over the course of 5 days. Viability and apoptotic cells were measured by flow cytometry using 7-AAD Viability Solution and APC Annexin V (BioLegend). NB-4 cells (1 × 105/mL) were transfected with 41.67 nmol/L of either non-targeting RNA control oligonucleotides or OLIG2 siRNA (Stealth RNAi, Thermo Fisher Scientific) using lipofectamine RNAiMAX technology (Thermo Fisher Scientific), according to the manufacturer's instructions. At 48 to 96 hours after transfection, whole-cell lysates were prepared, and mRNA and protein expression for OLIG2 was determined as described above. Data are expressed as the mean ± standard deviation of three independent experiments. Statistical analysis of data was performed using GraphPad Prism 5.0 software with Student's one-sample t-test and paired t-test to detect significant (p < 0.05) differences between the two groups. To investigate the DNA methylation status of the previously reported OLIG2 promoter region, we used quantitative pyrosequencing in leukemia cell lines U-937, KG-1, Kasumi-1, HL-60, PL-21, THP-1, Jurkat, K562, and HEL [4Yalcin A. Kreutz C. Pfeifer D. et al.MeDIP coupled with a promoter tiling array as a platform to investigate global DNA methylation patterns in AML cells.Leuk Res. 2013; 37: 102-111Abstract Full Text Full Text PDF PubMed Scopus (14) Google Scholar, 8Shen L. Kantarjian H. Guo Y. et al.DNA methylation predicts survival and response to therapy in patients with myelodysplastic syndromes.J Clin Oncol. 2010; 28: 605-613Crossref PubMed Scopus (304) Google Scholar]. Although KG-1, U-937, K562, Kasumi-1, and Jurkat cells exhibited high DNA methylation levels >80%, PL-21, HEL, and HL60 cells had DNA methylation levels between 60% and 70%, and THP-1 and NB-4 cells had very low DNA methylation levels 0.002), with the highest OLIG2 mRNA expression in THP-1 and NB4 cells; the highly methylated cell lines KG-1, U-937, K562, Kasumi-1, and Jurkat exhibited no expression (Fig. 1A,B). Importantly, this inverse correlation between DNA methylation of OLIG2 and mRNA expression translated to protein expression as demonstrated by Western blot, where OLIG2 protein expression was strong in low-methylated cell lines THP-1 and NB4 and to a lesser extent in medium-methylated cell lines HEL and HL-60 (Fig. 1C). Highly methylated cell lines KG-1, U-937, Kasumi-1, and Jurkat as well as PL-21 exhibited almost no protein expression.Figure 1DNA methylation and mRNA and protein expression of OLIG2 in leukemia cell lines and AML patient samples. Pyrosequencing and qPCR analyses reveal that cell lines with lower DNA methylation exhibit higher mRNA expression of OLIG2 and vice versa (A). This inverse correlation of DNA methylation/mRNA expression in leukemia cell lines is significant in the regression analysis (B). Lower DNA methylation of OLIG2 is associated with OLIG2 protein expression in Western blot analyses and vice versa in leukemia cell lines (C). DNA methylation level of bone marrow and peripheral blood specimen of 93 AML patients compared with bone marrow, CD34 + cells, and PBLs of healthy donors using pyrosequencing. Thirty-seven patients exhibited hypermethylation of OLIG2 ≥25% in the OLIG2 promoter region (D). Methylation levels <25% were detected in 56 patients (E). In a subset of 13 patients (for whom RNA was available), an inverse correlation between OLIG2 DNA methylation and mRNA expression was significant (F). Western blot analyses reveal that OLIG2 protein is not expressed in the bone marrow of normal karyotype AML patients and CD34 + cells of healthy donors (G). Corresponding to the t(15;17) cell line NB4, OLIG2 protein is expressed in two of three AML patients with t(15;17); low protein input for the second patient was due to very limited patient material (H). (Color version available online.)View Large Image Figure ViewerDownload Hi-res image Download (PPT) We further investigated OLIG2 DNA methylation status in 93 AML patients from the recently published German 00331 decitabine trial and for whom we had previously described a potentially adverse impact of higher OLIG2 methylation [9Hiller J.K. Schmoor S. Gaidzik V.I. et al.Evaluating the impact of genetic and epigenetic aberrations on survival and response in acute myeloid leukemia patients receiving epigenetic therapy.Ann Hematol. 2017; 96: 559-565Crossref PubMed Scopus (19) Google Scholar, 14Lübbert M. Rüter B.H. Claus R. et al.A multicenter phase II trial of decitabine as first-line treatment for older patients with acute myeloid leukemia judged unfit for induction chemotherapy.Haematologica. 2012; 97: 393-401Crossref PubMed Scopus (194) Google Scholar]. Corresponding well to previous reports by Shen et al., we found substantial DNA methylation (≥25%) of OLIG2 in 39.8% (37/93) of AML patients, whereas 60.2% (56/93) of patients had methylation levels <25% (Fig. 1D,E) [8Shen L. Kantarjian H. Guo Y. et al.DNA methylation predicts survival and response to therapy in patients with myelodysplastic syndromes.J Clin Oncol. 2010; 28: 605-613Crossref PubMed Scopus (304) Google Scholar, 9Hiller J.K. Schmoor S. Gaidzik V.I. et al.Evaluating the impact of genetic and epigenetic aberrations on survival and response in acute myeloid leukemia patients receiving epigenetic therapy.Ann Hematol. 2017; 96: 559-565Crossref PubMed Scopus (19) Google Scholar]. DNA methylation was more heterogeneously distributed throughout the five CpG sites than in cell lines, with CpG1 and CpG5 having the highest methylation levels. Furthermore, we sought to investigate the correlation between OLIG2 DNA methylation and mRNA expression. However, as this was a retrospective study, RNA material was available for only 13 patients. Nonetheless, as for leukemia cell lines, a negative correlation (r = −0.74, p < 0.004) between methylation and expression could be observed in these patients (Fig. 1F). On a protein level, we did not see OLIG2 expression in nine AML patients with normal karyotype and CD34 + cells of healthy donors (Fig. 1G). Interestingly, however, OLIG2 protein expression was detected in two of three AML M3 patients harboring the translocation t(15;17), corresponding to the cell line data of NB-4 cells (t(15;17) positive) (Fig. 1H) and revealing—to our knowledge for the first time—OLIG2 protein expression in AML patients. This specific finding was further supported when analyzing reports of OLIG2 mRNA expression in AML M3 t(15;17) patients from two publicly available databases: bloodspot (http://servers.binf.ku.dk/bloodspot/) and TCGA (https://cancergenome.nih.gov/), both revealing significantly higher OLIG2 mRNA expression than other AML subtypes (Supplementary Figure E1A,B, online only, available from www.exphem.org). To further augment DNA methylation of OLIG2 as a regulatory epigenetic silencing mechanism, we treated OLIG2 hypermethylated and non-expressing cell lines PL-21 and U-937 with the demethylating agent 5-aza-2′-deoxycytidine (decitabine) at doses of 200 and 400 nmol/L for 3 days. DNA methylation and mRNA and protein expression were analyzed on days 3 and 6. First, we investigated the effect of decitabine doses on the well-studied, non-cytotoxic, and growth-inhibitory effect of this drug on leukemia cell lines and detected a dose-dependent growth reduction in both cell lines, which was substantive when compared with untreated cells (Fig. 2A,D). Cell viability remained >90% in all experiments (data not shown), proving the non-cytotoxic, growth-inhibiting effect of decitabine.Figure 2Cell growth inhibition and re-expression of OLIG2 in AML cell lines on treatment with decitabine. Three-day treatment with doses of 200 and 400 nmol is effective decitabine treatment, with growth inhibition in cell lines PL-21 (A) and U-937 (D). Three-day decitabine treatment of PL-21 caused a dose-dependent decrease in OLIG2 promoter DNA methylation in pyrosequencing analyses resulting in mRNA (qPCR) and protein re-expression (Western blot) on days 3 and 6 (B, C). In U-937, the decrease in promoter demethylation and mRNA re-expression of OLIG2 was less pronounced (E). However, OLIG2 protein was re-expressed on days 3 and 6 (F). KG-1 and Kasumi-1 cells served as negative controls, and NB4 as a positive control. ***p < 0.05 compared with control. (Color version available online.)View Large Image Figure ViewerDownload Hi-res image Download (PPT) With decitabine treatment, DNA methylation of OLIG2 was reduced on day 3 by 20.1% at 200 nmol/L and by 38.9% at 400 nmol/L, respectively, in PL-21 cells as compared with untreated cells, whereas on day 6, OLIG2 DNA methylation had begun to increase again (Fig. 2B). DNA demethylation of OLIG2 was less pronounced in U-937 cells, with a maximum reduction of 17.8% at 400 nmol/L on day 3, and reduction remained stable on day 6 at both doses (Fig. 2E). Quantitative PCR analysis of OLIG2 mRNA revealed re-expression beginning on day 3 and increasing in a dose-dependent manner in PL21 and U-937 cells as well as in a time-dependent manner in PL21 cells (Fig. 2B,E). Most importantly, re-expression of OLIG2 mRNA translated into robust re-expression of OLIG2 protein in both cell lines, again suggesting time-dependent re-expression in PL-21 cells (Fig. 2C,F). The finding that demethylating treatment with decitabine results in re-expression of OLIG2 on the transcriptional and translational levels in PL-21 and U-937 cells suggests that OLIG2 hypermethylation may be an epigenetic silencing mechanism for OLIG2 expression. Growth inhibition without significant cytotoxicity with low doses of demethylating substances is a well-known mechanism in AML cells, and we investigated whether OLIG2 contributes to cell growth inhibition in AML cell lines. Therefore, we used a lentiviral vector construct, pLeGO-iG-OLIG2, to stably transduce Kasumi-1 and U-937 cells (while generation of overexpressed PL-21 cells unfortunately was repeatedly unsuccessful), resulting in no endogenous OLIG2 expression in the cell lines. Five days after transduction, Western blot analysis revealed robust expression of OLIG2 protein in Kasumi-1 and U-937 cells, whereas untreated cells and cells transduced with an empty vector exhibited no expression (Fig. 3A). OLIG2 expression was maintained after freezing and thawing of Kasumi-1 and U-937 cells and after 14 days of cell culture, supporting stable transduction. When cell growth kinetics of pLeGO-iG-OLIG2-transduced cells were compared with that of empty vector-transduced cells, we observed growth inhibition of pLeGO-iG-OLIG2–transduced Kasumi-1 cells beginning on the second day of culture and reaching a maximum of 27.0% on day 5 (Fig. 3B). This effect of OLIG2 was less pronounced in U-937 cells; however, on day 4, a growth reduction of 21.9% was observed (Fig. 3C). Although the observed growth inhibitory effect of OLIG2 overexpression is modest, it still supports a potential role in steering cell proliferation in AML cells. We then sought to determine the functional consequences of OLIG2 re-expression after pLeGO-iG-OLIG2 transduction and investigated whether transduced Kasumi-1 and U-937 cells manifested signs of differentiation. Although we did not detect signs of differentiation on a morphological level using Giemsa staining in Kasumi-1 cells (data not shown), U-937 cells exhibited a significant increase in CD11b expression—an established marker of granulocytic differentiation—on days 3 and 5 (Fig. 3D). On investigation of initiation of apoptosis using Annexin-V staining and flow cytometry, we could detect a moderate increase in apoptotic cells in OLIG2-transduced Kasumi-1 cells as compared with empty vector, whereas this effect was not observed in U-937 cells (Fig. 3E,F). To further investigate whether a knockdown of OLIG2 results in increased cell growth, we transfected NB-4 cells with siRNA against OLIG2. However, although OLIG2 mRNA and protein expression could be reduced substantially after 72 and 96 hours, respectively, cell growth was not changed (Supplementary Figure E2A,B). OLIG2's crucial role in differentiating and controlling proliferation in the neuroepithelial compartment is well established. Importantly, when overexpressed, OLIG2 has been demonstrated to bear strong oncogenic potential in malignant glioma [10Mehta S. Huillard E. Kesari S. et al.The central nervous system-restricted transcription factor Olig2 opposes p53 responses to genotoxic damage in neural progenitors and malignant glioma.Cancer Cell. 2011; 19: 359-371Abstract Full Text Full Text PDF PubMed Scopus (125) Google Scholar]. After several groups reported on differential DNA methylation of OLIG2 in AML, we investigated the potential role of this gene, which is not physiologically expressed outside the central nervous system, and we provide here the first evidence that OLIG2 is epigenetically regulated in AML cell lines and exerts antiproliferative activity. We recently confirmed the observation by Shen et al. and Kroeger et al. that OLIG2 is differentially methylated in myelodysplasia and AML and that hypermethylation of OLIG2 may be associated with worse outcome [7Kroeger H. Jelinek J. Estécio M.R. et al.Aberrant CpG island methylation in acute myeloid leukemia is accentuated at relapse.Blood. 2008; 112: 1366-1373Crossref PubMed Scopus (129) Google Scholar, 8Shen L. Kantarjian H. Guo Y. et al.DNA methylation predicts survival and response to therapy in patients with myelodysplastic syndromes.J Clin Oncol. 2010; 28: 605-613Crossref PubMed Scopus (304) Google Scholar, 9Hiller J.K. Schmoor S. Gaidzik V.I. et al.Evaluating the impact of genetic and epigenetic aberrations on survival and response in acute myeloid leukemia patients receiving epigenetic therapy.Ann Hematol. 2017; 96: 559-565Crossref PubMed Scopus (19) Google Scholar]. We now report that in leukemia cell lines and primary patient samples, DNA methylation of OLIG2 and that of mRNA are inversely correlated, hinting at an epigenetic regulatory mechanism. In addition, in leukemia cell lines, the inverse correlation between DNA methylation and expression also translated into expression of OLIG2 protein, which to our knowledge, is the first report OLIG2 protein expression in the myeloid compartment. Thus far, only Lin et al. have reported detection of OLIG2 RNA in leukemia cell lines [12Lin Y.W. Deveney R. Barbara M. et al.OLIG2 (BHLHB1), a bHLH transcription factor, contributes to leukemogenesis in concert with LMO1.Cancer Res. 2005; 65: 7151-7158Crossref PubMed Scopus (29) Google Scholar]. In a sample set of primary normal karyotype AML samples and CD34 + cells from healthy donors, we observed no OLIG2 protein expression. However, because we had seen strong OLIG2 expression in the NB4 cell line carrying the translocation t(15;17), we specifically investigated three AML patients with translocation t(15;17) and observed OLIG2 protein expression in two patients. Although interesting, interpretation of this result is limited by the small sample size, and a larger study is required to draw robust conclusions. Nonetheless, we here describe for the first time patients with OLIG2 protein expression in their AML cells. Previously, RNA expression was observed in a patient with ALL harboring the translocation t(14;21)(q11.2;q22) [11Wang J. Jani-Sait S.N. Escalon E.A. et al.The t(14;21)(q11.2;q22) chromosomal translocation associated with T-cell acute lymphoblastic leukemia activates the BHLHB1 gene.Proc Natl Acad Sci USA. 2009; 97: 3497-3502Crossref Scopus (76) Google Scholar]. To further augment our hypothesis of epigenetic regulation of OLIG2 in AML, we treated leukemia cell lines PL-21 and U-937, which do not express OLIG2, with the demethylating agent decitabine and found that DNA demethylation of OLIG2 resulted in re-expression not only of mRNA, but also of OLIG2 protein. Although we did not observe this effect in two other leukemia cell lines (Kasumi-1 and KG-1), we conclude that DNA methylation contributes at least in part to the regulation of OLIG2 in a subset of leukemia cell lines. To elucidate the impact of OLIG2 re-expression on cell proliferation, we stably transduced U-937 and Kasumi-1 cell lines (both not expressing OLIG2 protein) with a pLeGO-iG-OLIG2 cDNA vector. Transduction resulted in robust OLIG2 protein expression and led, although only modestly, to growth inhibition of 27.0% in Kasumi-1 cells and 21.9% in U-937 cells after 4 and 5 days, respectively. In following functional analysis focusing on proliferation, we did not observe a morphological change in leukemia cells over-expressing OLIG2; however, in U-937 cells, CD11b as an established marker of granulocytic differentiation was significantly upregulated and Kasumi-1 cells exhibited an increase in apoptotic cells. Transient knockdown of OLIG2 with siRNA in NB-4 cells did not result in increased cell growth, most likely because of residual OLIG2 protein; stable knockdown experiments in future studies, however, could further elucidate the role of OLIG2 in AML. Nonetheless, our results, in context with the data from re-expression experiments, point toward a tumor-inhibiting role of OLIG2 expression in AML. This would be in line with our previous observation that OLIG2 hypermethylation portends a negative prognosis in AML patients [9Hiller J.K. Schmoor S. Gaidzik V.I. et al.Evaluating the impact of genetic and epigenetic aberrations on survival and response in acute myeloid leukemia patients receiving epigenetic therapy.Ann Hematol. 2017; 96: 559-565Crossref PubMed Scopus (19) Google Scholar]. In lymphoid malignancies, however, ectopic expression of OLIG2 may have oncogenic potential, as in a mouse model co-expressing OLIG2 and LMO1 and for one patient with ALL; other groups have reported a leukemia/lymphoma-enhancing effect [11Wang J. Jani-Sait S.N. Escalon E.A. et al.The t(14;21)(q11.2;q22) chromosomal translocation associated with T-cell acute lymphoblastic leukemia activates the BHLHB1 gene.Proc Natl Acad Sci USA. 2009; 97: 3497-3502Crossref Scopus (76) Google Scholar, 12Lin Y.W. Deveney R. Barbara M. et al.OLIG2 (BHLHB1), a bHLH transcription factor, contributes to leukemogenesis in concert with LMO1.Cancer Res. 2005; 65: 7151-7158Crossref PubMed Scopus (29) Google Scholar]. Taken together, these findings suggest a leukemia cell type-specific role for OLIG2. In summary, we confirm previous studies of differential methylation of OLIG2 in AML patients and provide first evidence of an epigenetic regulatory mechanism of OLIG2 expression in AML. Moreover, our data demonstrate that OLIG2 is ectopically expressed in AML patients and that expression of OLIG2 protein could act as a leukemia cell growth-inhibiting factor. Additional studies of combinations of decitabine and differentiation-inducing drugs, such as ATRA, as well as functional assays, including knockdown strategies and ideally a mouse model, are now warranted to further investigate the role of OLIG2 in AML. This work was supported by grants from the German Cancer Aid, Bonn, Germany (BH 110213, RC and LB 110530) and the German Research Foundation, Bonn, Germany (DFG; SPP 1463, ML 429/8-1 and CRC 992 MEDEP, C4; Heisenberg-Professur BU 1339/8-1). We are thankful to Gregor Klaus for technical assistance with experiments. Supplementary Figure E1Differential mRNA expression of OLIG2 in AML patients using publicly available data sources. Analyzing OLIG2 mRNA expression levels of AML patients from two publicly available databases, bloodspot (http://servers.binf.ku.dk/bloodspot/) (A) and TGCA (https://cancergenome.nih.gov/, analyzed by Oncomine) (B) revealed increased OLIG2 mRNA expression in AML M3 t(15;17) patients as compared to other leukemia subtypes. ***: p < 0.05View Large Image Figure ViewerDownload Hi-res image Download (PPT)Supplementary Figure E2siRNA knockdown of OLIG2 in NB4 cells. OLIG2 mRNA expression by RT-PCR (normalized to GAPDH and compared to scrambled = control) shows a significant decrease 72h after transfection with sIRNA against OLIG2, while this effect regresses at 96h (A). Western blot of OLIG2 and its densitometry shows substantially reduced (45%) expression at 96h for OLIG2 siRNA #2 and #3 (B). OLIG2 siRNA transfection did not result in a change of cell growth (data not shown) *, **, ***: p < 0.05.View Large Image Figure ViewerDownload Hi-res image Download (PPT)Supplementary Table E1Primer design and conditions for quantitative bisulfite-pyrosequencing analyses, qPCR and OLIG2 overexpressionPyrosequencing:Region(bp)Primer sequences and annealing temperatureProduct lengthSequencing primersOLIG2:ForwardTTTTAAAGGTGAGGATGTTTATTAT229AGGTGAGGATGTTTATTATAReverseGGGACACCGCTGATCGTTTAAAAAATCCAAACCCCCTATATuniversal: GGGACACCGCTGATCGTTTA(5′-Biotin)50°CqPCR:OLIG2-Forward: 5-AGATCGACGCGACACCAGCG-3OLIG2-Reverse: 5-TCGGACCCGAAAATCTGGATGCG-3β-actin-Forward: 5- AGCCTCGCCTTTGCCGATCC-3β-actin-Reverse: 5-ACATGCCGGAGCCGTTGTCG-3Primers for OLIG2 overexpression:fwd 5'-GATACAGAATTCAAGGAGGACCCTGCGAAAGC-3'rev 5'- GTTATTAGGCCTAACATCCCCTCACTCCCCA-3' Open table in a new tab
Referência(s)