Studies of histone demethylase JARID1B in hematopoiesis and leukemogenesis

Abstract

Post-translational modifications of histone proteins serve as one of the key epigenetic regulatory mechanisms in the development of organisms. It is well-known that methylation on histone lysine residues is an important epigenetic modification for the transcriptional regulation of normal hematopoiesis and leukemogenesis. JARID1B, a member of the JARID1 histone H3 lysine 4 (H3K4) demethylases, was found essential for the self-renewal of both embryonic stem cell and melanoma stem-like cell, and was involved in regulating genes, such as Egr1, Bmi-1 and p27, during embryo development. In addition, JARID1B is involved in the differentiation of neural cells and macrophages. Although JARID1B is believed to have important functions in stem cell biology, its role in hematopoiesis and leukemogenesis has not been systematically studied.

We therefore examined the expression profile of JARID1B in different hematopoietic lineage cells. We observed an up-regulation of JARID1B in differentiated hematopoietic cells by comparing with hematopoietic stem cells and progenitor cells, suggesting that the enhanced cellular level of JARID1B is associated with hematopoietic lineage commitment. Interestingly, JARID1B expression is generally low in human leukemia cell lines and in CML (Chronic Myeloid leukemia) patient samples compared to 〖CD34〗^+ cord blood cells and normal peripheral white blood cells, which indicates the down-regulation of JARID1B is associated with leukemia development. We further modulated the expression of JARID1B in human leukemia cell lines, K562 and SEM, and in mouse hematopoietic stem/progenitor cells (〖Lin〗^-/〖Sca〗^(-1+)/c-〖Kit〗^+, LSK cells). We found that knockdown of Jarid1b in LSK cells did not alter their cell-cycle pattern. However, total colony formation number was reduced in serial re-plating assays, suggesting Jarid1b is required for the maintenance of colony-forming ability and self-renewal property. Knockdown of JARID1B in K562 cells did not change their cell proliferation and cell-cycle pattern, but did consistently inhibit their colony-forming ability during serial re-plating assays. On the other hand, overexpression of JARID1B in K562 and SEM leukemic cells inhibited cell proliferation and colony formation, but with no significant changes on cell-cycle patterns. Furthermore, apoptosis staining did not show any correlations between JARID1B overexpression and apoptosis.

Previously, JARID1A, another JARID1 family member, was found as a fusion partner in AML (Acute Myeloid Leukemia); its third PHD domain, which locates at C-terminus, is associated with leukemogenesis. By amino acid sequence alignment, the differences between JARID1A and JARID1B protein are mainly occurred at their C-terminal regions, after the second PHD domain. Therefore, GST fusion protein pull-down experiments for this region was performed. Preliminary results showed that the C-terminus of JARID1B protein interacts with proteins of RNA transcriptional machinery complex. However, further investigation is needed to demonstrate these interactions are directly associated with JARID1B inhibitory effects on gene expression. To conclude, our results suggest that JARID1B plays an essential role in the biology of hematopoietic stem cells and leukemic cells. Investigation on its interacting partners and downstream target genes would lead us a detailed understanding of JARID1B function in hematopoietic cells.