Abstract
Introduction:
Primary myelofibrosis (PMF) belongs to the heterogeneous group of chronic myeloproliferative neoplasms (MPN) together with essential thrombocytosis (ET) and polycythemia vera (PV). It has been suggested that these neoplasms represent a biological continuum from early cancer stage (ET,PV) to advanced MF. Multiple studies report frequent mutations in epigenetic regulators. However, the association to epigenetic changes and the role of epigenetic aberrations in different cell populations is still unknown. We therefore performed DNA methylation profiling of sorted cells from MF patients to unravel pathways contributing to disease phenotype and gain insight into MF pathogenesis. As an aberrant DNA methylation pattern may be an early event in tumorigenesis and may be crucial for progression of the malignant clone towards the more aggressive forms of MPN, we further aimed to identify methylated candidate driver genes.
Material and methods:
Peripheral blood samples from 16 MF patients were together with BM (bone marrow) and peripheral blood from 3 healthy age matched controls sorted in CD34+ cells, granulocytes and mononuclear cells, and analysed for differential methylated regions using Illumina Infinium HumanMethylation 450K BeadChip. Candidate genes were validated by pyrosequencing in a second cohort of 30 MF patients. To identify potential driver genes the DNA methylation status of candidate genes were likewise analyzed in a larger cohort consisting of 60 ET and PV patients.
Results:
The number of differential methylated CpG sites between MF cells and the healthy counterparts differed extensively among the three cell populations analyzed. In MF CD34+ cells 1628 CpG sites were differential methylated compared to normal CD34+ cells, and 519 and 213 differential methylated CpG sites were observed in MF granulocytes and MF mononuclear cells, respectively (Δβ was set to 0.2 with an adjusted p-value < 0.05, T-test). Differentially methylated genes were mainly involved in cancer and embryogenic pathways in both the MF CD34+ and mononuclear cells, while mononuclear cells also showed aberrant methylation of genes involved in the inflammatory disease pathways. MF granulocytes showed significant aberrations in pathways involving immunological diseases, cell death and survival.
Candidate genes have been identified and validation is ongoing. Interestingly, a gradual increase of the DNA methylation level of TRIM59 was observed from the healthy controls (31%) over ET (53%) to PV (64%) and MF (65%). ET patients could be distinguished from both healthy controls (P= 0.0004, Mann-Whitney test) and from the more progressed stages PV and MF (P=0.0132, Mann-Whitney test) based on the TRIM59 DNA methylation level. TRIM59 promoter methylation could, however, not discriminate between PV and MF (P=0.4721, Mann-Whitney test).
Conclusion:
Genome-wide DNA methylation profiling of sorted MF blood cells provided an exclusive insight into which pathways that are contributing to MF disease phenotype at a cell specific level. The MF CD34+ cells had the highest number of differential methylated CpG sites (n=1628) when comparing to granulocytes (n=519) and mononuclear cells (n=213) and should be cells of choice when exploring new treatment strategies. Interestingly, the mononuclear compartment show aberrant methylation of inflammatory genes supporting a role of aberrant immune regulation in the pathogenesis of MPN. Earlier studies have failed to identify aberrant methylation in early ET and PV, however, the preliminary data on the methylation of individual genes (TRIM59 promoter methylation) shows that it might be possible to identify early driver genes, and that it may be possible to select a panel of genes that can discriminate early MPN from the late MF stage.
Disclosures
No relevant conflicts of interest to declare.
The DNA methylation Profile of Undifferentiated Arthritis Patients Anticipates their Subsequent Differentiation to Rheumatoid Arthritis
