p62-ZZ Domain Inhibition Prevents MM Cell-Induced Epigenetic Changes at the Runx2 and C/EBPb Promoters

Multiple myeloma bone disease (MMBD) is a paradigm for uncoupled bone remodeling and is characterized by non-healing lytic bone lesions. Osteoblast (OB) function is highly suppressed or absent, persists in the absence of MM cells, and remains a significant cause of skeletal-related events. The persistence of OB suppression in MMBD suggests that MM cells induce repressive epigenetic changes at the Runx2 gene, the key transcription factor required for OB differentiation of bone marrow stromal cells (BMSC), which are preOB. We reported that TNFα is a major suppressor of OB in MMBD that reduces Runx2 levels by inducing Gfi1, a transcriptional repressor of Runx2. Using ChIP analysis of MM-exposed BMSC, we showed that Gfi1 directly binds the Runx2 promoter and recruits the chromatin corepressor HDAC1 to the Runx2 promoter in MM-exposed BMSC, reducing transcriptionally permissive euchromatin marks such as H3K9ac.

Recently, we reported that p62 (sequestosome-1) in BMSC is critical for the formation of MM cell-induced signaling complexes that mediate OB suppression and IL-6 production. We found that XRK3F2, a novel inhibitor of the p62 ZZ domain (p62-ZZ), blunted MM cell-induced repression of Runx2 and induction of Gfi1 in BMSC, and induced new bone formation and remodeling in mice with established MMBD, but did not alter normal bone. In the current study, we further evaluated the specificity of XRK3F2's inhibition of p62-ZZ interactions and investigated the mechanism by which blocking p62-ZZ prevents MM-induced suppression of key osteogenic transcription factors in BMSC.

We previously showed that TNFα and MM cell-induced IL-6 production by BMSC are both p62 dependent and independent. XRK3F2 blocked TNFα-induced NFκB signaling in BMSC and MM cells, and partially inhibited TNFα-induced IL-6 production by BMSC. We now report that XRK3F2 blunted the TNFα upregulation observed in BMSC after MM cell coculture but did not alter TNFα production in p62 knockout (p62KO) BMSC following coculture with MM cells. TNFα treatment of p62KO BMSCs resulted in minimal induction of IL-6, which was not altered by XRK3F2. Transfection of p62KO BMSC with full-length p62 constructs restored TNFα-induced IL-6 production and XRK3F2’s capacity to reduce TNFα-induced IL-6. In contrast, XRK3F2 had no effect on TNFα-induced IL-6 production by p62KO BMSC transfected with p62 constructs lacking p62-ZZ.

To further investigate XRK3F2’s mechanism of action, we tested if XRK3F2 prevents Gfi1-induced epigenetic suppression of Runx2 by preventing Gfi1’s upregulation and binding to Runx2. ChIP analysis of MM exposed BMSC treated with XRK3F2 demonstrated that XRK3F2 reduced MM-induced Gfi1 occupancy of the Runx2 promoter and prevented MM-induced reduction of H3K9ac. Since the region proximal to the Runx2 promoter contains putative C/EBPβ binding sites, we tested if MM-induced p62-ZZ signaling activates C/EBPβ, a transcription factor that regulates OB lineage maturation and IL-6 production in BMSC and plays a role in the upregulation of Gfi1 expression. MM-exposed BMSC had increased enrichment of C/EBPβ at both the Gfi1 gene and the C/EBPβ-regulated Il6 gene, and XRK3F2 reduced upregulation of MM-induced C/EBPβ binding at both the Gfi1 and Il6 genes. We conclude that XRK3F2’s reduction of the MM-induced C/EBPβ binding in BMSC results in decreased Gfi1 mRNA expression, and thereby reduces MM-induced Gfi1 occupancy and HDAC1 recruitment at the Runx2 promoter. These results suggest that targeting p62-ZZ in MMBD may reverse C/EBPβ mediated Gfi1 activation, resulting in rescue of the MM cell-induced epigenetic suppression of Runx2 in BMSC, and that next-generation derivatives of XRK3F2 should impact BMSC-supported MM cell survival.