A role for histone H3, histone H4 and histone associating proteins DNMT3A and PHF6 in JAK2V617F positive myeloproliferative neoplasms

Abstract

The Philadelphia chromosome negative myeloproliferative neoplasms (MPN); polycythemia vera, essential thrombocythemia and primary myelofibrosis, are clonal disorders harbouring the specific Janus kinase 2 (JAK2) lesion (JAK2V617F) at a high frequency. Accumulating evidence from pedigrees of MPN together with the identification of a plethora of heterogeneous lesions identified in sporadic MPN patients suggest that JAK2V617F, and other acquired changes in JAK2, cooperate with mutations in other genes to generate clonal disease. The nature of the other mutations dictates the disease phenotype and contributes to the potential for transformation to acute leukaemia. Emerging research is focussed on understanding the contribution of these other changes to MPN pathogenesis. As many of the other recurrent mutations reported in MPN affect genes involved in epigenetic regulation, studies have focused on identifying the role of epigenetic changes. Many epigenetic regulators mediate their effects via interaction with post-translationally modified histone H3 and H4 and, given the findings that pathogenic mutations are present in histone H3 in other tumours, the focus here was on the role of histone H3 and H4 variants in MPN pathogenesis. Thus, a key aim of this PhD project was to identify pathogenic coding variants in the histone H3 and histone H4 genes in MPN. In the first study, MPN peripheral blood mononuclear cells or granulocyte patient samples were screened using Sanger sequencing for histone H4 coding region variants. The screen identified previously unidentified sequence variants in several of the 15 histone H4 genes. A coding variant of HIST1H4C, resulting in the substitution of cysteine for arginine [R, (HIST1H4C:p.R4C)], affects a known key residue involved in epigenetic regulation (R3 residue on the mature protein). This gene was also shown to make a major contribution to the histone H4 mRNA pool in several haemopoietic cell types, further indicating a potential for this variant to confer functional consequences. This was tested using enforced expression of the variant in two cell lines, HEK293 and a myeloid cell line (FDM cells). Further, it was demonstrated by RNA microarray and QPCR analyses of FDM cells expressing HIST1H4C:p.R4C that this variant conferred selective differential expression of five genes. We extended this analysis of Histone H4 genes to screen for disease-associated variants in histone H3 genes (n=17) and the histone-H3 interacting protein PHF6 (consisting of 9 coding exons). For this, we used an amplicon-based next generation sequencing (NGS) approach and the Roche 454 sequencing platform. This identified a coding region variant in HIST1H3E (HIST1H3E:p.A96V), the presence of which was confirmed by Sanger sequencing. A number of other changes identified by the NGS approach were not confirmed by Sanger re-sequencing, however the possibility that these are present in the original patient sample at a level below the detection limit for Sanger sequencing cannot be excluded. Sanger sequencing of the PHF6 terminal exons 9 and 10 identified a somatic mutation (PHF6R335fs) in a PV patient. Finally a Sanger-based sequencing screen of the terminal exon of gene encoding DNA (cytosine-5-)-methyltransferase 3 alpha (DNMT3A) identified somatic R882C and M880V substitutions in two PV patients. Clonal analysis for these mutations in DNMT3A indicated that their acquisition can either precede or follow the acquisition of JAK2V617F.