Modulation of RUNX2 Activity by ERa in Osteoblasts

摘要：ER对成骨细胞RUNX2活性的调节

ABSTRACT: Modulation of RUNX2 Activity by ER in Osteoblasts The bone-sparing properties of estrogens are mediated by ER in a variety of cell types, including cells of the osteoblast and osteoclast lineages, but the underlying molecular mechanisms are poorly understood. In particular, there are contrasting reports on how estradiol (E2) affects RUNX2, an osteoblast master regulator that has been implicated in human bone mass control through GWAS. RUNX2 plays critical roles in the osteoblast lineage to stimulate both autonomous cellular differentiation (and thus bone formation) and osteoblast-driven osteoclastogenesis (and thus bone resorption). While this project initiated with a focus on inhibition of RUNX2 by E2, the present renewal application aims to take genome-wide approaches to understand why E2 does not inhibit RUNX2-driven transcription uniformly. We show that RUNX2 activity upon various targets is modulated differently, with most targets inhibited, but others not inhibited and some even cooperatively stimulated, to various extents, by RUNX2 and E2. Locus-dependent differential modulation of RUNX2 is expected to ultimately change the balance between RUNX2-mediated osteoblast differentiation and RUNX2-mediated osteoblast-driven osteoclastogenesis. Our preliminary data also demonstrate that both raloxifen and lasofoxifene poorly mimic E2 in modulating activity of RUNX2 at different loci. Mechanisms underlying the locus- and ligand-dependent modulation of RUNX2 by ER are completely unknown. Based on our preliminary results, we hypothesize that ER differentially modulates RUNX2 across the osteoblast genome depending on local relative positions of sites occupied by RUNX2 and ER and the ER ligand. Additionally, we hypothesize that local collaborating transcription factors (TFs), as well as specific cognate motif sequences for RUNX2, ER and collaborating TFs shape local effects of ER on RUNX2. We will first investigate by ChIP-seq analysis of histone marks how E2 modulates RUNX2-driven changes to the chromatin activation status at every genomic locus. The mutual effects of ER and RUNX2 on occupying their target loci will also be determined genome-wide by ChIP-seq. Computational models will then be developed to explain the locus-specific combinatorial transcriptional regulation by RUNX2 and ER. These models will be tested, first computationally and then experimentally, for their ability to predict effects of E2 on RUNX2 activity. Finally, since ER in pre-osteoblasts protects female cortical bone in mice, and since raloxifen and lasofoxifene, SERMs commonly prescribed in the US and in Europe, respectively, only reduce the risk for vertebral (predominantly trabecular) osteoporotic fractures, but not non-vertebral fractures, we will also decode how raloxifene- and lasofoxifene-bound ER modulate RUNX2 activity genome wide. Principal genomic determinants will be identified, which explain the differential modulation of RUNX2 activity as a function of the locus (different targets) and the ligand (E2 vs. raloxifen vs. lasofoxifene). This project will help understand how E2 differentially modulates RUNX2 activity at different loci to potentially alter the balance between bone formation and bone resorption. Gene sets will be identified where SERMs mimic, or do not mimic E2. While unveiling the underlying genomic codes, tools will be generated for future investigation of how E2, or any SERM, modulate RUNX2 activity at prototypic loci to differentially regulate specific gene sets.