We isolated the human homologue of the Drosophila l(3)mbt gene by exon trapping, based on its location on the long arm of chromosome 20 (20q) within a region frequently deleted in hematologic malignancies (most commonly polycythemia vera, MDS, and AML).
The l(3)mbt gene encodes a member of the polycomb group (PcG) of transcriptional regulatory proteins, which play a vital role in maintaining the proper expression of key genes, such as homeobox-containing and cell cycle regulatory genes. Evidence linking abnormalities in the expression or function of PcG genes with the development or progression of several types of human cancer is now emerging (e.g., lymphoma, Hodgkin's disease, and multiple myeloma).
The human l(3)mbt protein is highly homologous to the lethal malignant brain tumor [l(3)mbt] protein, which functions as a tumor suppressor gene in Drosophila. To more completely define the role of the human l(3)mbt gene in hematologic cancers, we are assessing its genomic integrity and the relative levels of l(3)mbt expression in primary, human myeloid malignancy samples. L3MBT contains 300 amino acid MBT domains and with the Patel Lab at MSKCC, we crystallized the structure identifying a three-bladed propeller-like structure with ligan-binding pockets in each of the domains. Subsequent studies have shown that L3MBTL1 compacts chromatin based on its ability to bind to histone H4K20 mono and dimethyl and H1BK26 mono and dimethyl. These studies identified the MBT domain as the only chromatin binding module that preferentially recognizes the mono and dimethyl over the trimethyl form of lysines. We are in the midst of analyzing the L3MBTL1 conditional knock-out mice that we have generated, in the interim we have been studying the effects of knocking down of L3MBTL1 using Sh or SiRNAs. These studies have implicated L3MBT in erythroid differentiation and in antagonizing Ras transferation. Additional studies are underway to identify how downregulation of L3MBTL1 can interface with overexpression of oncogenic tyrosine kinases, such as the JAK2V617F mutation. We continue to define the biological effects of l(3)mbt by expressing wild type and mutant forms in hematopoietic cells. We are also characterizing l(3)mbt-interacting proteins and are generating l(3)mbt conditioned knock-out mice to analyze its phenotype. These studies will define whether l(3)mbt functions as a classic tumor suppressor gene or can be implicated in human cancer by haploinsufficiency.
