Abstract 3520: PBRM1 inactivation in renal cell cancer alters master transcription factor hub composition to repress instead of activate epithelial-differentiation

Clear cell renal cell cancers (RCC) universally inactivate PBRM1 (single allele deletion 100%, bi-allelic deletion/mutation ~40%), however, the mechanisms by which PBRM1 loss contributes to RCC-genesis are unknown. Of ~4000 genes consistently less expressed in RCC than normal renal cortex, >30% were kidney differentiation genes by gene ontology analysis, e.g., 394 (~10%) suppressed genes had specialized renal epithelial functions of cation transport, cell adhesion and excretion (TCGA). Suppression of hundreds of epithelial-differentiation genes suggested disruption to the core transcription factor circuit (protein network hub) activating this program. This master transcription factor hub contains PAX2, PAX8, GATA3 and LHX1, shown by murine knock-outs and human pedigrees with renal defects, and confirmed by our protein interactome studies. Of these master transcription factors, PAX2 and PAX8 expression was preserved in RCC compared to normal kidney, but downstream GATA3 and LHX1 expression was markedly decreased (2 to 4-fold) even though PAX8 localized at these target genes by chromatin immunoprecipitation - this suggested compromised PAX8 transcription activating function. PAX8 protein interactome in RCC cells indicated that PBRM1 is a PAX8 coactivator, and introduction of PBRM1 into PBRM1-haploinsufficient RCC cells (ACHN) by transfection increased activation of GATA3, WT1 and HNF4A, decreased MYC protein, increased p27/CDKN1B protein that mediates cell cycle exits by terminal differentiation, produced morphological changes of epithelial differentiation, and decreased proliferation. Coactivator versus corepressor recruitment by master transcription factors is a dynamic interchange, and the protein interactome studies indicated PBRM1 loss unbalanced stoichiometry of the PAX8-associated protein hub towards corepressors including DNMT1. Depletion of DNMT1 by non-cytotoxic concentrations of decitabine (0.5 µM) produced a remarkable shift from corepressors to coactivators in the PAX8 protein interactome, accompanied by terminal epithelial-differentiation as observed with PBRM1 reintroduction. Cell cycle exits by differentiation do not require p53/p16, and this effect was also produced in vivo in a subcutaneous xenograft model of TP53-mutated refractory/relapsed RCC, especially when decitabine in vivo pharmacology was optimized for non-cytotoxic DNMT1-depletion by avoiding a high Cmax and increasing its half-life and distribution into solid tissue, by administration of low dose in combination with an inhibitor of the enzyme cytidine deaminase that degrades decitabine. PBRM1 loss in RCC thus alters composition of a master transcription factor hub to repress instead of activate terminal epithelial-differentiation, an effect that can be reversed pharmacologically for novel, p53/p16-independent differentiation-restoring therapy. Citation Format: Xiaorong Gu, Francis Enane, Reda Mahfouz, Tomas Radivoyevitch, Bartlomiej Przychodzen, Yvonne Parker, Daniel Lindner, Brian Rini, Yogen Saunthararajah. PBRM1 inactivation in renal cell cancer alters master transcription factor hub composition to repress instead of activate epithelial-differentiation [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr 3520. doi:10.1158/1538-7445.AM2017-3520