Study Of Gene Expression Profiles In Refractory Acute Myeloid Leukemia (M2a)

Background Although marked progress has been made in the treatment of acute myeloid leukemia (AML), about 20%⁴0% patients with AML fail to enter complete remission (CR) after standard induction chemotherapy. Furthermore, approximately 70% patients in CR receiving post remission chemotherapy will experience relapse and be refractory to therapy. In the end, these patients will die from disease. Therefore, it is urgent to elucidate the mechanisms of relapsed/refractory AML and to search for indications for early diagnosis of refractory AML. Previous prognostic factors comprise mainly of age, leukocyte count, secondary AML, cytogenetic abnormalities in leukemic cells. Of these factors, karyotype is crucial. But among different age groups there are many patients with normal karyotype. Furthermore, the patients with the same karyotype have evidently different response to the same therapy. Due to technical restriction, most previous studies merely simultaneously determined one or a few genes expression at transcriptional or protein level. Results of these studies could not commendably explain clinical phenomena. It is probable that some undiscovered genes are associated with AML prognosis or/and the development of relapsed/refractory AML. Objective Different gene expression profiles have been studied between AML-M2a patients with evidently different duration of first continuous complete remission (CCR₁) in order to find out the genes relevant to the prognosis of theAML-M2a. Different gene expression profiles have also been studied between self-paired patients with AML-M2a at presentation and at relapse in order to find out the genes related to recurrence of AML-M2a. The present study aim at providing basis for further investigating early diagnosis of refractory leukemia and designing new therapeutic strategy. Methods1. Patients and samples Bone marrow mononuclear cells at primary diagnosis and at relapsed disease were collected and used as different experimental groups respectively. Group A composed of three AML-M2a patients with CCRi> 12 months at primary diagnosis, Group B composed of three AML-M2a patients with CCRi<6 months at primary diagnosis and Group C consisted of Group B patients at relapse. The Group A and Group B samples contained 78 to 93 percent blast cells. The Group B and Group C samples contained 65 to 92 percent blast cells. Gender, clinical signs, immunophenotype and median of age and leukocyte count between Group A and Group B were almost identical. The karyotype was normal in six patients with AML-M2a at presentation. Induction chemotherapy and post remission chemotherapy were almost same.2. Gene expression profiling analysis Total RNA was extracted using standard Trizol RNA isolation protocol (Invitrogen) . The quality, purity and concentration of RNA were assessed with the use of RNA Lab On Chip using Agilent BioAnalyzer 2100. We used 2ug of total RNA from each sample to perform fluorescent linear amplification and two-color fluorescence labeling with the use of Low Input RNA Fluorescent Linear Amplification Kit (Agilent). Each sample in Group B was divided into two samples. Samples in Group B were labeled with Cy3, while samples in Group A were labeled with Cy5 in order to study different gene expression profiles between Group A and Group B. Samples in Group C were labeled with Cy3, whilesamples in Group B were labeled with Cy5 to study different gene expression profiles between Group B and Group C. Mixed probes were hybridized to the 60mer Human IB Oligo Microarray (Agilent). The hybridized Oligo Microarray was scanned using 2565BA gene chip scanner (Agilent). Gene expression data were analyzed with the use of G 2567 AA Feature extraction software (v 7.2) (Agilent). Up-regulated genes were determined by processed signal ratio (Cy3/Cy5)>2.0, while down-regulated genes were determined by processed signal ratio (Cy3/Cy5)<0.5. Genes whose mRNA levels were up/down regulated in each Group B patient in comparison with each Group A patient were identified as commonly differentially expressed genes. The common genes with different expression were regarded as the genes associated with prognosis in AML-M2a. Genes whose mRNA levels were up/down regulated in each Group C patient were identified as commonly differentially expressed genes as compared with each patient in Group B. The common genes with different expression were regarded as the genes associated with the recurrence of AML-M2a. 3. Semi-quantitative RT-PCR First strand of cDNA was synthesized using Oligo (dT) is (Promega) primer. Specific primers for CCNA1, DAPK1 and the housekeeping gene GAPDH were designed by Premier Primer 5.0 software. Forward and reverse primers for CCNA1, DAPK1 and GAPDH were as follows respectively: 5'-CAGTATTGAGAAAGGCTGGTC3',5'-ACGTGCAGAAGCCTATGA3'; 5'-ATCCTAGACGTGGTCCGGTAT3\ 5'-GTTCTCGCAGCCTGGGTA3' and 5'-CGGGAAACTGTGGCGTGAT3',5'-TGGCAACTGTGAGGAGGG3' .The length of amplified section of CCNA1, DAPK1 and GAPDH was 402bp, 402bp and 553bp, respectively. Samples of CCNA1, DAPK1 and GAPDH were amplified for 19 cycles, 19 cycles and 13 cycles, respectively. PCR products were added to 1.5 percent agarose gel for electrophoresis. The gel was scanned and then band density was qualified by Quantity One software (BIO-RAD).Results1. Results of quality control of RNA The median of ratio of A260 to A280 was 1.815 (range: 1.740 to 1.973) and the median of ratio of 28s to 18s ribosomal RNA was 1.716 (range: 1.668 to 1.804) in Group A and Group B RNA samples. The ratio of A260 to A280 was at least 1.878 and the 28s/18s ratio was at least 1.702 in Group C RNA samples. None of the RNA samples showed obvious degradation or contamination by DNA. These high-quality total RNA samples were used for microarray experiments.2. Results of quality control of hybridization against microarray Signal intensity of positive controls on the microarray was very high, while signal intensity of negative controls on the microarray was very low. Microarray analysis showed that over 99 percent of spots were consistently hybridized using G 2567 AA Feature extraction software (v 7.2).3. Microarray hybridization result There were twenty-two commonly differently expressed genes between Group A and Group B. Of these genes, ten genes were up-regulated in each Group B patient, while twelve genes were down-regulated in each Group B patient as compared with each patient in Group A. The ten up-regulated genes comprised of HBG2, GGTLA1, I₉30827, I₁109295, FLJ13310, FAM3A, FLJ13324, HSD3B2, I₉65618 and I₉63615. The twelve down-regulated genes composed of I₉64508, I₉30336, I₉30066, I₁109518, FLJ12448, PDCD7, CGI-135, CA6, KIAA1872, GABARAPL1, KLRF1 and APR No recent studies have been reported that any of these genes is associated with prognosis of leukemia. There were ten commonly differentially expressed genes between Group B and Group C. Among these genes, seven genes were the common up-regulated genes in Group C composed of FLJ40773, DAPK1, ARG1, CXorfl, MMRN, CCNA1 and CRTAP. Only three genes were the common down-regulated genes in Group C comprised ofFLJ31564, I₉36798 and I₉5633. Similarly, no recent reports have shown that theten commonly differentially expressed genes were studied in relapsed leukemia.4. Results of semi-quantitative RT-PCR mRNA levels of DAPK1 and CCNA1were up 5.16, 8.25, 2.01 and 5.60, 2.14, 2.24 times in each self-paired AML-M2a atrelapse as compared with presentation, respectively. These validated the microarrayresults..Conclusions1. The microarray results are validated by serai-quantitative RT-PCR.2. The new discovered twenty-two commonly differentially expressed genes between Group A and Group B may be related to CCRi of AML-M2a and be important for prognosis in AML-M2a. These genes may be indications for early diagnosis of refractory AML-M2a.3. The new discovered ten commonly differentially expressed genes between Group B and Group C reflect that gene expression changes of AML-M2a with adverse prognostic factors at relapse and may be associated with the development of refractory AML-M2a.4. MDR1 is up-regulated in Group B. There is no MDR1 expression change in Group C as compared with Group B.