UC San Diego

Myelodysplastic syndromes (MDS) are hematologic malignancies characterized by impaired differentiation of hematopoietic stem and progenitor cells (HSPC) causing peripheral blood cytopenias and an increased propensity for developing acute myeloid leukemia. MDS arises as a consequence of genetic and/or epigenetic lesions in HSPCs. Dysfunctional epigenetic regulation is a hallmark of MDS, where a majority of MDS patients harbor mutations in genes regulating DNA methylation (5mC) or chromatin modification. Aberrant 5mC patterns are common in MDS and may correlate with disease progression.

Disease progression in MDS is highly variable, making prognostication critical for decisions on treatment options and intensity. In Chapter 2 we interrogated 5mC patterns at ~500,000 CpGs across the genome in 141 MDS patient tumor samples and applied an unsupervised classification to define 5mC sub-types of MDS. We identified five patient clusters with distinct 5mC patterns, and these groups were enriched for distinct patterns of genetic lesions related to prognostic risk. The 5mC sub-types displayed differences in survival that were independent of all known prognostic variables, including genetics. Differentially methylated genes between clusters included those with known prognostic impact, as well as novel gene associations which were validated in external MDS cohorts. Our findings highlight the importance of 5mC in disease progression and its utility as a prognostic biomarker.

DNA methyltransferase inhibitors, such as 5-Azacitidine (5-Aza), are the only therapy approved for treating higher-risk MDS, yet only half of patients respond to therapy. Mutations in the 5mC regulator, TET2, are associated with increased response rates, however the mechanisms involved are unknown. In Chapter 3 we modeled TET2 loss in isogenic erythroleukemia cell lines and studied differences in 5mC, 5-hydroxymethylcytosine (5hmC), and gene expression during 5-Aza exposure. We show that TET2 loss particularly influences 5mC and 5hmC patterns at erythroid gene enhancers, and is associated with down-regulation of gene expression. 5-Aza disproportionately induces expression of erythroid genes in TET2KO cells through 5mC reduction at erythroid enhancers. This work highlights the role of 5mC and 5hmC changes at enhancers in altering differentiation-associated gene expression signatures, and sheds light on how 5-Aza may be more effective in patients harboring TET2 mutations.