Berberine And Decitabine Act As Demethylation Modulators In Acute Myeloid Leukemia For Reversal Of Drug Resistance

Background and objectivesAcute myeloid leukemia (AML) is one of the common hematological malgiance in adult. In recent years, combination therapy of chemotherapy and stem cell transplantation, greatly increases the rate of complete remission AML. However, improvements were less pronounced in relapsed patients. The unmet need in this population may be addressed through optimal combination of chemotherapy and novel agents. Epigenetic changes result in alterations in phenotype or gene expression through mechanisms other than changes in the underlying DNA sequence. Epigenetic changes include DNA methylation, histone modifications and microRNAs and other small RNAs regulators. Cancer cells are adept at altering these normal developmental and cellular processes to ensure their own survival and advantage, and myeloid malignancies represent fertile ground for elucidation of epigenetic cancer mechanisms and the development of effective targeted therapies based on this knowledge.The DNA methyltransferase inhibitors decitabine represent archetypal drugs for epigenetic cancer therapy. One alternative approach for the treatment of AML is the use of hypomethylating agents, including the5-aza-2-deoxycytidine (decitabine; DAC) and other epigenetic regulators. Berberine, an isoquinoline plant alkaloid, has a wide range of biochemical and pharmacological effects. However, the exact mechanism of these bio-activities remains poorly understood. Our experiments independently confirmed berberine’s activities on growth suppression and cell cycle cessation, and its ability to augment ROS level and to inhibit IL6production in multiple myeloma cells. And based on the bioinformatics analyses, predictions and experimental validations, our study firstly discovered the possibility of berberine being an epigenetic modulator in U266cells. Many gene sets associated with polycomb complex1/2epigenetic regulation were enriched. Reverse-docking using berberine as a ligand identified lysine-N-methyltransferase as a putative target of berberine. These findings suggested a possible role of berberine in epigenetic modulation. The goal of this study was to identify the some important epigenetic changes induced by DAC and berberine which could cause myeloid differentiation, and depress downstream pathway wnt/β-catenin.Methods and Materials1. Cell line and reagentsThe cell line Kasumi-1was kindly provided by Professor Liu Qiang (Hematology Department of The Third Affiliated Hospital, Medicine College, Zhongshan University). HL-60/ADR, HL-60and KG1-α cell lines were obtained from Institutes for Biological Sciences Cell Resource Center, Chinese Academy of Sciences, Shanghai, China. The cells were maintained in RPMI1640(Gibco, USA) supplemented with10%fetal bovine serum (FBS; Gibco), at37℃in5%CO2incubator at37℃. High concentration of berberine (Sigma-Aldrich, USA) was diluted in DMSO (Sigma-Aldrich, USA). DAC was kindly provided by Johnson&Johnson Ltd (America).2. We used MTS (promega, America) to determine the time and dose dependent effect on proliferative inhibition of berberine and DAC to Kasumi-1, HL-60/ADR, HL-60and KG1-a cells.3. Gene expression microarrayTotal RNA was separately harvested and purified from HL-60and HL-60/ADR cells seperatively. RNA sample amplification and labeling were performed using an the Amino Allyl MessageAmpTM Ⅱ aRNA Amplification Kit (Ambion, Inc.) according to the manufacturer’s instructions. Samples (4μg) labeled with Cy3/Cy5were hybridized and processed on Chipscreen operon human oligo microarray U133. Scanning was performed with the LuxScanTM10K Microarray Scanner using under the settings recommended by CapitalBio. After removal of negative-flagged probes and data normalization using R’s "limma" package, differential genes were selected based on the fold-change for functional enrichment with the help ofusing R programming (www.cran.r-project.org) and the Nextbio platform (www.nextbio.com). Gene set enrichment was done performed with using our in-house web tool, in which the whose main data sources are were provided by MSigDB (Broad Institute), Reactome, Biocarta and Gene Ontology Consorti μM, in addition to the gene sets we collected from literature mining.4. Flow cytometry and Wright stainingCells apoptosis and differentiation was detected by flow cytometry (AnnexinV-FITC/PI, KeyGEN) and Wright staining. Intracellular concentration of Adriamycin (ADM) in HL-60/ADR and Kasumi-1cells was determined also by flow cytometry.5. RNA extractionHL-60/ADR and KGl-a cells were seeded in75cm2plates at the concentration of1×107cells/plate and were treated for48hours with berberine at final concentrations of0,11,22,44μM. RNA was extracted using Trizol (Invitrogen, USA) according to the manufacturer’s recommended protocol. The quantity and purity of RNA were determined by measuring A260and A280using NanoDrop ND-1000(NanoDrop). Integrity of RNAs was determined by1.5%agarose gel electrophoresis.6. Real-time PCRThe cDNAs were synthesized by reverse transcription from0.5μg total RNAs using oligo (dT) primer. Real Time PCR reactions were carried out using SYBR Green PCR Master Mix (Takara) on a LightCycler480System (Roche).7. Western blotting analysisBerberine was added to HL-60/ADR and KGl-a cells at final concentrations of0,11,22μM for48h. The concentration of DAC was1μM for treatment. Nuclear and whole-cell lysates were prepared as previously described. Protein concentrations of the SDS lysates were determined by the bicinchoninic acid method. Primary antibodies used in this study were the following:anti-H3K27me3monoclonal antibody (CST, No.9733), anti-Histone3monoclonal antibody (CST, No.3377), anti-EZH2monoclonal antibody (CST, No.5246), anti-DNMT1monoclonal antibody (CST, No.5032), anti-β-catenin monoclonal antibody (CST, No.8480), and anti-GAPDH monoclonal antibody (CST, No.5174).8. Methylation specific PCR (MSP)DNA of all samples were extracted and modified by sodium bisulfite treatment converting unmethylated, but not methylated. Modified DNA was used as a template for PCR using primers specific for either methylated or unmethylated sequences. MSP was performed to observe the methylation of sFRP1, sFRP2, sFRP4, sFRP5. The amplification products were separated on2%agarose gel and visualized by ethidium bromide staining and UV transilluminafion.9. Statistical analysisEach experiment was performed in triplicate and repeated at least three times. All data from these experiments are expressed as the means+/-S.E.M using SPSS13.0software. The difference among the means of multiple groups was analyzed by one-way analysis of variance (ANOVA) followed by the LSD test when equal variances assumed, while Dunnett’s T3test were used when equal variances not ass umed. A difference was defined as significant at P<0.05.Results1. The differences of genes expression between HL-60/ADR and HL-60cellsThere are many differences in genes expression between the two cell lines. Many gene sets associated with the polycomb complex1/2and DNA methylation, which could regulate downstream pathways, such as Wnt, hedgehog, TNF-mediated apoptosis, NF-kB, Jak-STAT, and ERK/AKT.2. Proliferation inhibition and apoptosisProliferative activity was detected by MTS assay. Different dose of DAC and berberine had proliferative inhibition effect on HL-60, KGl-α, Kasumi-1and HL-60/ADR cells, inhibition ratio was elevated accompanied with time extension and dose increase, it was a time-dose dependent relationship. HL-60/ADR, KGl-α cells were treated by different concentration of berberine for24,48,72and96hours, and48h was the best time. HL-60, KG1-α, Kasumi-1and HL-60/ADR cells were treated with DAC for48,72and96hours, and72h was the best time. HL-60/ADR and KG1-α cells were treated with the combination of berberine and DAC and the proliferative inhibition ratio was more than the single drug. Berberine and DAC did not significant induce apoptosis at the concentration of1,11μM and1μM respectively.3. Cell differentiation1μM DAC significantly induced expression of CD11b on HL-60/ADR and KG1-α cells. Berberine did not has this effect. Cell morphology had great changes too. Treated by DAC makes cells more mature.4. Effect on drug resistance in HL-60/ADR cells treated with DAC and BBRThe intracellular accumulation of adriamycin was detected by flow cytometer, HL-60/ADR cells were cultured with0.5μg/ml adriamycin after treatment with1μM DAC and11.22μM Bortezomib for72,48hours. Adriamycin positive in DAC and BBR group were higher than control group. It demonstrated that DAC and BBR could improve intracellular accumulation of adriamycin and reverse drug resistance.5. Real-time PCR resultsIn HL-60/ADR and KG1-α cells, the mRNA expression of CREBBP, EP300and SIRT3, all of which are the members of the class III histone acetyltransferases (HD AC) in berberine-treated group was significantly higher than that of control group. The expression of histone methyltransferase and demethylase mRNA were also measured. The data showed that SETD7, WHSC1I and KDM6A were up-regulated while WHSC1Ⅱ and SMYD3were down-regulated upon treatment. There was no difference in HDAC8(Figure8).6. Berberine and DAC affects level of H3K27me3, EZH2, DNMT1and β-catenin proteins The protein levels of methylated histone H3K27me3, EZH2, DNMT1and β-catenin were measured semi-quantitatively by Western blot analysis. These proteins expressions trended lower as concentrations of berberine increased, but there is no difference in EZH2after been treated by DAC.7. BBR and DAC can decrease methylation level of SFRPsAt a concentration of11and22μM, the BBR induced decrease methylation level in the SFRPs. And as reported, DAC at a concentration of1μM can significantly decrease the methylation.Conclusion1.BBR has dose and time dependent antiproliferation effect on HL-60, HL-60/ADR, KG1-α leukemia cells in vitro, but the effect of DAC was not so significant in HL-60/ADR.2. DAC can induce differentiation of HL-60/ADR and KG1-α.3. BBR can inhibit EZH2and DNMT1expression.4.And BBR and DAC can depressed the wnt/β-catenin pathway through decreasing methylation of SFRPs genes.5. We conclude that the antitumor effects of Berberine may result from the modulation of key epigenetic regulators. This offers a novel explanation of how berberine initiates so many therapeutic activities in clinical. Although our combined bioinformatics and experimental study supports this working hypothesis, more experiments should be carried out to justify this view, for example, to dissect the interplays between berberine and its target(s) and the downstream signaling cascade.