The Roles And Mechanisms Of MicroRNA Involved In TRAIL Resistance Of Acute Myeloid Leukemia

Background:Acute myeloid leukemia （AML） is a kind of highly heterogeneous hematopoietic malignancy, characterized by infiltration of the bone marrow, peripheral blood and other tissues by clonal, proliferative and abnormally differentiated cells of the hematopoietic system. AML is the most common adult acute leukemia. Although advances have been achieved in stem-cell transplantation and standard chemotherapy of acute myeloid leukemia （AML） over the past few decades, chemoresistance and therapy-related mortality still play a vital role in therapeutic failure and poor outcomes. Therefore, it’s very needed to explore for novel therapeutic methods.TNF-related apoptosis-inducing ligand or Apo2L （Apo2L/TRAIL） is a member of the TNF family expressed mainly by cells of the immune system. By binding to death receptors TRAIL-R1 and TRAIL-R2, TRAIL is able to trigger’extrinsic and intrinsic apoptosis’which specifically kill tumor cells while sparing normal tissues, rendering it an anti-tumor agent of great value. However, the clinical use of TRAIL for cancer therapy is now limited since quite a few human cancer cells, such as leukemia cells, are resistant to TRAIL-induced apoptosis either primarily or aquiredly. Many researches have addressed this issue and put forward several possible mechanisms including the non-functional binding to death decoy receptor 1 or 2 （DcR1,2）, reduced Caspase expression, dysfunctions of death receptor 4 （DR4） and death receptor 5 （DR5） or overexpression of anti-apoptotic proteins cellular FLICE inhibitory protein （cFLIP）, Mcl-1, Bcl2, etc. Still a large portion of TRAIL resistance awaits to be unveiled.MicroRNAs （miRNAs） are a large group of small non-coding RNAs that negatively regulate the expression of target genes by binding to the 3’untranslated region （UTR） of their mRNAs, resulting in translation inhibition or mRNA degradation. To date, tremendous studies have expounded the versatile roles that miRNAs play in proliferation, differentiation, apoptosis, metastasis, angiogenesis and carcinogenesis. Meanwhile aberrant expression of miRNAs has been constantly identified in pathogenesis and chemoresistance of hematopoietic malignancies. In addition, Garofalo et al. reported that compared with less invasive and/or normal lung and liver cells, miR-221 & 222 were overexpressed in aggressive non-small cell lung cancer and hepatocarcinoma, which facilitated their functions on inducing TRAIL resistance by targeting PTEN and TIMP3 tumor suppressors. These data indicate that miRNAs exerted effects on both AML development and TRAIL resistance of tumors. Whether they influence AML TRAIL resistance remains unknown. In conclusion, identification of the functional miRNAs and underlying molecular mechanisms involved in AML TRAIL resistance may provide novel therapeutic targets for AML TRAIL treatment.Objective:The aim of this study was to screen the differentially expressed miRNAs in AML TRAIL resistance process and reveal the underlying molecular mechanisms by identifying their direct target genes.Materials and Methods:1. AML cells were cultured in 96-well plates and treated with serial dilutions of TRAIL （Peprotech, USA） at concentrations of 0,20,50,100,200ng/ml for 48h. Cell Counting 8-assay （Beyotime, China, JS） was added to each well and cells were incubated for 4h at 37℃. The absorbance was measured at 450 nm and 650 nm simultaneously and we calculated inhibition rate of cell growth and IC50 values （the concentration of drug at which 50% of cell proliferation inhibited）. Each test was performed in triplicates.2. Patient samples:After informed consent, bone marrow （BM） samples were obtained from 78 AML patients and 10 healthy donors at Qilu Hospital, Shandong University. The study protocols were approved by the Medical Ethical Committee of Qilu Hospital, Shandong University. Mononuclear cells were isolated from the bone marrow samples using density-gradient centrifugation with Ficoll-Paque Plus （Ficoll, Pharmacia LKB Biotechnology, Piscataway, NY） and stored at -80℃. We classified AML patients into newly diagnosed group and complete remission group. Several matched-pair BM samples were acquired from the same patients both at the time of diagnosis and at the complete remission state.3. Total cellular RNA was extracted from cell lines using Trizol （Invitrogen, Carlsbad, CA,USA） according to the manufacturer’s protocols. TaqMan MicroRNA Assay （ABI, Foster City, CA, USA） was used to detect the expression levels of mature miR-424 and miR-27a. As for PLAG1, Bcl2 expression, quantitative real-time PCR was performed by a Applied Biosystem 7900HT System （ABI, Foster City, CA, USA） using SYBR Green PCR Master Mix （Toyobo, Osaka, Japan）. Each sample was run in triplicates and amplified in a 20μl reaction volume under the manufacturer’s instructions. The house-keeping gene GAPDH, which has relatively constant expression level in AML cell lines, was used as an internal control.4. Functional studies of miR-424, miR-27a and PLAG1 in AML TRAIL resistance4.1 Cell Transfection:The synthetic mimics of miR-424 and miR-27a along with scrambled oligonucleotides were purchased from GenePharma （Shanghai, GenePharma Co.Ltd）. SiRNAs of PLAG1 and its negative control were synthesized by Ribobio （Guangzhou Ribo Bio Co. Ltd）. Lipofectamine 2000 reagent （Invitrogen, Carlsbad, CA） was used to transfect cells with miRNAs or siRNAs at a final concentration of 50nM according to the manufacturer’s instructions.48h after transfection, qRT-PCR or Western Blot were conducted to detect the expression levels of miR-424, miR-27a and PLAG1.4.2 CCK8 assays were performed to examine TRAIL sensitivity of AML cells after transfection:AML cells with transfection of miRNAs or siRNAs were cultured in 96-well plates and treated with serial dilutions of TRAIL at concentrations of 0, 20,50,100,200ng/ml or 0,50,100,200,500ng/ml for 48h. Cell Counting 8-assay was added to each well and cells were incubated for 4h at 37℃. The absorbance was measured at 450nm and 650nm simultaneously and we calculated inhibition rate of cell growth and IC50 values. Each test was performed in triplicates.4.3 After 48h of tranfection, the apoptosis assay was conducted using Annexin V/propidium iodide （PI） Apoptosis Detection Kit （Invitrogen, Carlsbad, CA） under the guidence of the manufacturer’s instructions. Apoptosis rate of cells was measured by Becton Dickinson FACS Calibur flow cytometer, detecting the fluorescence of at least 10000 cells each sample.5. Identification of miR-424 and miR-27a direct target genes5.1 We used bioinformatic methods （TargetScan, Pictar） to predict potential targets of miR-424 and miR-27a. Upon transfection of miR-424 or miR-27a mimic, PLAG1 expression was evaluated by qRT-PCR and Western Blot.5.2 Full length of the 3’UTR of PLAG1 with or without mutations was cloned from human genomic cDNA and inserted into pMIR-REPORT vector （Ambion Inc., Austin, TX, USA） upstream the firefly luciferase coding sequence, which formed a construct carrying the wildtype 3’UTR （pPLAGl_WT） and constructs carrying 3’UTR sequences with mutations at the putative recognition sites of miR-424 and miR-27a （pPLAG1₄24M and pPLAG1₂7aM respectively）. Wildtype and mutant inserts were confirmed by DNA sequencing.293T cells were cotransfected with 0.5ug reporter gene constructs and miR-424 mimic, miR-27a mimic or scrambled control using Lipofectamine 2000. Then firefly and renilla luciferase activities were measured by the dual-luciferase reporter assay system under the manufacturer’s instructions.5.3 After transfection of either miR-424/27a mimic or siPLAG1, Western Blot analysis was conducted to examine the protein expression of several classical molecules that are known to favor or counteract TRAIL-induced cell death, including TRAIL-R1, TRAIL-R2, cFLIP, Mcl-1, CD44, TRAF2 and Bcl2.5.4 MiR-424, miR-27a overexpressing or PLAG1 knockdown K562 cells were treated with （+） or without （-） 200ng/ml TRAIL for 3h and then analyzed by Western Blot for the protein expression of Caspase3, Caspase8 and PARP.5.5 For stable expression of PLAG1, full length of its CDS was PCR-amplified from human genomic cDNA and inserted between the Kpnl and Xbal sites of p3xFLAG-CMV-10 vector （kindly provided by Dr Xiulian Sun, Qilu Hospital of Shandong University）. Similarly the CDS of Bcl2 cloned from Bcl2-pCEP4 （Addgene, Plasmid# 16461） was inserted into pLenti-C-GFP vector. To package Bcl2-Lenti-C-GFP or control lentiviral particles, Hek293T cells were co-transfected with a mixture of 10μg pLenti-C-GFP-Bcl2 or pLenti-C-GFP, and 6.67μg psPAX2,3.3μg pMD2.G utilizing Lipofectamine 2000 transfection reagent.5.6 Plasmid LB322 containing Bcl2 promoter from ATG to-3934 （Addgene, Plasmid#15381） and p3xFLAG-CMV-10-PLAG1 were co-transfected into 293T cells and we conducted dual-luciferase reporter assay as mentioned above. After transfection of p3xFLAG-CMV-10-PLAG1 into 293T cells, Bcl2 mRNA expression was measured by qRT-PCR.5.7 Western Blot analysis validated Bcl2 overexpression in K.562 cells which were infected with Bcl2-Lenti-C-GFP. K562 cells stably expressing Bcl2-Lenti-C-GFP were transfected with either miR-424/27a mimic or siPLAG1 and their respective controls.48h later cells were treated with or without 200ng/ml TRAIL for an additional 48h. Cell viability was quantified with the CCK8 assays.6. Statistical analysis:Each experiment was performed at least three times and all the data were presented as mean±S.E. Comparison between control and experimental groups was conducted using Student’s t-test. To measure the discrepancy between clinical AML samples, Mann-Whitney test was used. All the tests were two tailed and computed by SPSS （version 17.0） software. p<0.05 was considered statistically significant.Results:1. The IC50 values of HL-60, HL-60/ADM, K562, K562/A02 and NB4 cells were 699.34 ng/ml,76.71 ng/ml,2057.05 ng/ml,333.70 ng/ml and 476.33 ng/ml respectively. Based on the above results, we classified AML cell lines into TRAIL-resistant （K562）, semi-resistant （HL-60, NB4, K562/A02） and TRAIL-sensitive （HL-60/ADM） groups.2. The expression levels of miR-424, miR-27a and PLAG1 in AML cell lines.2.1 MiR-424 and miR-27a were markedly downregulated in TRAIL-resistant cells while upregulated in TRAIL-sensitive cells. Semi-resistant HL-60 and NB4 exhibited intermediate levels of miR-424 and miR-27a expression. Eventhough K562/A02 was semi-resistant to TRAIL, its miR-424 and miR-27a expression levels were only slightly higher than that of K562.2.2 PLAG1 upregulation was observed in TRAIL-resistant （K562） and semi-resistant （HL-60 and NB4）, versus TRAIL-sensitive cells （HL-60/ADM）. Despite of the moderate sensitivity to TRAIL, K562/A02 possessed PLAG1 expression level inferior only to K562. By Western Blot analysis, we discovered that endogenous PLAG1 protein levels of TRAIL-resistant and semi-resistant cells were distinguishably higher than that of the TRAIL-sensitive cells.3. The expression levels of miR-424, miR-27a, PLAG1 and Bcl2 in AML patient samples.3.1 Upregulation of miR-424 and miR-27a was observed in AML patients when they attained complete remission （CR） in contrast to that of the same patients at the time of diagnosis.3.2 PLAG1 was notably overexpressed in newly diagnosed patients with respect to either CR patients or controls. In addition, BM samples obtained from the same patients at both the newly diagnosed and the CR state were analyzed for their PLAG1 expression. Most patients exhibited a sharp decrease in PLAG1 mRNA level once they achieved the CR state.3.3 Concurrent with PLAG1, Bcl2 was evidently upregulated in the newly diagnosed AML patients versus CR patients as well as controls. Similarly, Bcl2 expression of newly diagnosed AML patients dropped down from a relatively higher level when they achieved the CR state.4. Functional studies of miR-424, miR-27a and PLAG1 in AML TRAIL resistance.4.1 Increased expression of miR-424 and miR-27a upon transfection were confirmed by qRT-PCR.4.2 qRT-PCR and Western Blot showed that after transfection of siRNA, PLAG1 expression in AML cell lines was prominently reduced.4.3 After 48h of transfection, proliferation assays showed that overexpression of miR-424 and miR-27a both led to enhanced inhibition of proliferation in comparison with controls. Apoptosis assays demonstrated that enforced miR-424 and miR-27a expression raised the proportion of apoptotic cells evidently with regard to their respective controls.4.4 AML cells with PLAG1 knockdown showed a much lower proliferation rate compared with the control group. AnnexinV-FITC assays on TRAIL-resistant and semi-resistant cell lines revealed a significant increase in TRAIL-induced apoptosis after PLAG1 silencing.5. MiR-424 and miR-27a both targeting PLAG1 directly5.1 Bioinformatic methods （TargetScan, Pictar） predicted that human PLAG1 （NM₀01114634） contained regions （nucleotides 5218-5225; 4565-4571） that matched the seed sequences of miR-424 and miR-27a respectively. Upregulation of miR-424 and miR-27a scarcely influenced mRNA expression of PLAG1. However, endogenous PLAG1 protein levels of AML cell lines had been significantly downregulated after transfection with miR-424 or miR-27a mimic.5.2 Dual luciferase reporter assays showed that transfection of 293T cells with miR-424 or miR-27a mimic reduced the pPLAG1_WT luciferase activity by 4.21-fold and 3.11-fold respectively. This repression could be undermined with mutations at the predicted binding sites, suggesting that miR-424 and miR-27a were able to specifically target the putative regions in the 3’UTR of PLAG1.5.3 The protein levels of DR5, Mcl-1, cFLIP, TRAF2 and CD44 that are known to favor or counteract TRAIL-induced cell death remained unchanged after miR-424&27a upregulation or PLAG1 downregulation. However, anti-apoptotic protein Bcl2 expression was found negatively correlated with miR-424 and miR-27a while positively associated with PLAG1 expression.5.4 Increased miR-424 and miR-27a facilitated processing of Caspase8, Caspase3 and poly（ADP-ribose）ploymerase （PARP） in response to TRAIL treatment. PLAG1 silencing by siRNA also augmented cleavage of Caspase8, Caspase3 and PARP, indicating that miR-424 and miR-27a might enhance TRAIL sensitivity by targeting PLAG1 through activating apoptotic pathways.5.5 qRT-PCR confirmed that PLAG1 upregulation caused about 41.9%elevation of Bcl2 mRNA expression in 293T cells. Cotransfection of 3xFLAG-CMV-10-PLAG1 and LB322, a luciferase reporter vector containing full length of Bcl2 promoter, into 293T cells increased Bcl2 promoter activity by up to 90%.5.6 Forced expression of Bcl2 in K562 cells was verified by Western Blot. Upregulation of Bcl2 partially rescued the proliferation inhibition aroused by miR-424 and miR-27a mimic transfection in the presence of TRAIL. Likewise, overexpressed Bcl2 also impaired the TRAIL sensitizing effects resulting from PLAG1 downregulation.Conclusions:1. MiR-424 and miR-27 were upregulated in AML TRAIL-sensitive cells while downregulated in TRAIL-resistant cells, indicating that decreased expression of miR-424 and miR-27a may contribute to TRAIL resistance of AML cells.2. MiR-424 and miR-27a increase TRAIL sensitivity of AML cells by directly targeting PLAG1.3. PLAG1 exerts its anti-apoptotic functions on AML cells by transcriptional activation of Bcl2 which interferes with both intrinsic and extrinsic apoptotic pathways via cleavage blockade of Caspase8, Caspase3 and PARP.