Differential Expression of miRNAs during the Course of Chronic Myeloid Leukemia.

Abstract

Despite the high efficiency of CML therapy with tyrosine kinase inhibitor imatinib (Gleevec®, STI571), there are 20–30% of patients with suboptimal response or therapy failure. There are several known mechanisms of resistance (e.g. mutation in BCR-ABL kinase domain, BCR-ABL or hOCT-1 overexpression), while others are still uncovered. Aberrant expression of specific miRNAs might contribute to the resistance and disease progression, or could be the consequence of the disease status. Our aim was to screen the expression of known miRNAs in patient samples of different stages of CML treated with imatinib: 1. Major molecular response (MMR), 2. Suboptimal response/failure (not achieved cytogenetic response after 12 months of STI); 3. Start of hematological relapse; 4. Blast crisis (BC); 5. Diagnosis. Groups were created according to defined parameters (ELN definition) and were precisely characterized by BCR-ABL transcript level, cytogenetic, hematological and clinical examinations. Each group was represented by pooled leukocytes from peripheral blood of 5 patients, who contributed to the pool with the same number of cells; total pool concentration was 10x106 leukocytes/ml. Altogether 25 patient samples were tested. A pool of total leukocytes from 13 healthy donors was used as a control. Total RNA was isolated from the homogenate (5x106 leukocytes) of pooled samples. The RNA from the diagnosis pool was isolated in duplicate to check the technical variability of the assay. The miRNA expression was measured using the PIQORTM miRXplore arrays (Miltenyi Biotech) with 780 probes. Mean signal and mean local background intensities were obtained for each spot of the microarray images using the ImaGene® (Biodiscovery). Raw data derived from ImaGene software were processed with the PIQOR Analyzer (Miltenyi Biotech). As expected, most differentially expressed miRNAs (25 miRNAs with more than 2-fold change to control) were found in blast crisis. In BC we found the overexpression of polycistronic and oncogenic miR-17-92 cluster (miR-17, miR-19a, miR- 20a, miR-19b, and miR-92), which has been described in CD34+ cells from CML patients (Venturini et al.2007, Blood, 109 (10): 4399). Other important miRNAs that play roles in hematopoiesis were overexpressed in BC: e.g. miR-181a (16 fold), miR-222 (6 fold), miR-106a (3.5 fold), miR-221 (3 fold). Interestingly, the quantity of specific miRNAs differed in the context of a disease status. MiR-150 was downregulated at diagnosis, which was not observed in MMR and suboptimal response/failure. The downregulation appeared again at the start of hematological relapse (3 fold) and BC (10 fold). The miR-150 target is Myb oncogene encoding transcription factor required for proliferation, differentiation and survival of normal and leukemic hematopoietic cells. C-Myb has a longer half-life in BCR-ABL expressing cells than in normal ones. At the start of the hematological relapse and blast crisis let7c was upregulated but in MMR and suboptimal response/failure let7c level was equal to the control sample. The known target of let7c is c-Myc that in CML contributes to tumorigenesis. MiR-126 was downregulated in MMR, suboptimal response/ failure and start of hematological relapse, but in blast crisis it was overexpressed. The predicted miR-126 target is the antiapoptotic heat shock protein hsp70. It was shown that BCR-ABL expression results in the upregulation of hsp70. Attenuation of hsp70 levels either by imatinib or hsp70 antisense sensitized K562 cells to e.g. cytarabine and etoposide (Ossenkoppele 2005, Blood, 105 (3): 917). In conclusion, it seems that miR-150, 126 and let7c reflect the disease status and could become new possible markers in CML. Supported by GACR GA301/08/P154.