EZH2 gene is located on chromosome 7q, a frequently affected genomic area in myeloid neoplasms including in MDS. The SET domain protein encoded by EZH2 is a component of the polycomb repressive complex-2 (PRC2) that has histone lysine 27 residue of histone 3 (H3K27) methyltransferase activity, whose methylation forms a repressive epigenetic marker for gene regulation. A spectrum of loss-of-function mutations of EZH2 have been identified in myeloid neoplasms including in MDS, underlying the biological and pathological importance of EZH2 gene product in MDS pathogenesis. To further characterize molecular alterations of EZH2 in the hematopoietic progenitor cell compartment of MDS, we have undertaken the study to assess EZH2 expression in bone marrow CD34+ cells of patients with MDS. We used Q-RTPCR to compare EZH2 RNA levels between the CD34+ cells that were isolated from patients with MDS (N=74) and healthy individuals (N=9). Q-RTPCR results indicate that 51% of the patients (N=38) had reduced EZH2 expression. Although the whole patient cohort did not show a significantly difference of EZH2 expression compared controls, the subset of patients bearing chromosome 7 (7-) or 7q deletion (7q-) (N=16) presented a significant reduction of EZH2 RNA expression (0.4 fold of control, p=0.04). Thirteen of the 16 patients (81%) of the 7-/7q- subset had an over 50% decrease of EZH2 RNA expression level in bone marrow CD34+ cells. These results suggest that haploinsufficiency plays a key role in the molecular regulation underlying reduced EZH2 expression in hematopoietic progenitor cells of MDS. In order to further evaluate molecular alterations of EZH2 in patients without 7/7q deletion, we subsequently performed capture deep sequencing to survey potential EZH2 mutations in the subset of diploid patients (N=32) of the main cohort using their bone marrow mononuclear cells (BM-MNNC). Sequencing of all coding exons of EZH2 gene reveals that three of the 32 diploid patients (9%) carry EZH2 mutations, including one missense (C590Y), one nonsense and one splicing mutation. We then took off these three patients with EZH2 mutations and analyzed potential sole impact of EZH2 RNA expression level in the diploid/EZH2-wildtype background in the remaining 29 patient subset. Fourteen (48%) of these patients have an over 50% reduction of EZH2 expression in bone marrow CD34+ cells. Furthermore, compared to other diploid patients that are without EZH2 reduction, patients with reduced EZH2 expression (<50%) have a significant increase for the expression of a panel of innate immune regulatory genes. These genes include JMJD3 (p=0.05), TLR2 (p=0.02), IL-8 (p=0.03), IL-1B (p=0.02), and S100A9 (p=0.02). Of importance, overexpression of these innate immune genes in bone marrow CD34+ cells have been demonstrated to be implicated in the pathogenesis of MDS (Wei et al. Leukemia 2013). Of interest, survival analysis revealed that patients with reduced EZH2 RNA expression in bone marrow CD34+ cells had significant longer survival in the diploid/EZH2-wildtype subset (N=29, 42 months v.s. 22 months p=0.04). The same result also applies to the whole patient cohort of this study (N=74, 33 months v.s. 15 months, p=0.004). In contrast to worse prognosis associated with EZH2 gene mutation, this data suggest that down-regulation of EZH2 expression may not have such impact on survival. Taken together, results of current study indicate that, besides genomic mutation, down-regulation of EZH2 expression in the hematopoietic progenitor cell compartment also occurs in a subset of patients with MDS, and more frequently in patients with 7/7q deletions. Furthermore, this study also suggests that in patients with diploid karyotype and wildtype EZH2 gene, reduced EZH2 expression and its co-occurrence with overexpression of innate immune genes may have potential pathogenic and prognostic implications. Further investigation of EZH2 gene regulation and its interaction with innate immune signals should be performed in MDS.

Disclosures:

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution