Abstract 106: Metastatic outcome of early-stage estrogen receptor-positive (ER+) primary breast cancers predicted by differential expression of a subset of wild-type p53 target genes

Abstract

Abstract The p53 tumor suppressor regulates a plethora of biological processes including cell cycle, apoptosis, DNA repair and cellular senescence. Gene expression signatures associated with loss-of-function p53 mutations have previously been linked to poor prognosis primary breast cancers, even among estrogen receptor (ER)-positive breast cancers the majority of which expresses wildtype (wt) p53. However, the proportion of DNA-bound (direct) or non-bound (indirect) p53 target genes within these gene expression signatures remains unknown. Therefore, we utilized chromatin immunopreciptiation-sequencing (ChIP-seq) (FactorPath, Genpathway) and expression microarrays (Affymetrix U133Av2) to characterize genome-wide direct and indirect p53 target genes associated with transcriptome changes induced in ER+ MCF7 cells after upregulating wt-p53 protein by the selective MDM2 inhibitor MI-63 (Ascenta Therapeutics). ChIP-seq and expression data were mapped together using gene symbols to yield 10653 unique genes for the differential binding and expression analysis. 2239 showed differential p53 binding (MACS; p<1e-10) mapping within +/- 10kb of their UCSC gene coordinates. Of the 1903 unique genes found differentially expressed (SAM; FDR p <0.005), 539 were shown by ChIP-seq to be direct p53-bound targets and 1364 were shown to be indirect p53 target genes. Gene Ontology (GO) analysis was performed to identify enriched functional categories among the direct and indirect p53 target genes, finding 62 GO biological processes significantly enriched among direct p53 target genes (FDR p<0.05) most of which were related to apoptosis, in contrast to the 147 GO biological processes enriched within the indirect p53 target genes including cell cycle, DNA repair, and telomere maintenance. The average expression of these direct and indirect p53 target genes were then computed in a pooled expression microarray dataset of 303 adjuvant naïve, node-negative ER+ breast cancer cases from two independent sources (GSE2034, GSE7390) annotated for distant metastasis-free survival (DMFS), Cox analysis revealed a significant association between overall p53-regulated gene expression (direct + indirect targets) and DMFS (p=0.001), but this prognostic association was significant only for the indirect p53 targets (p < 0.001; with p = 0.08 for direct p53 targets) These findings suggest that even among early-stage primary ER+ breast cancers in which p53 mutations are rare, differentially expressed p53 target genes are prognostic, with those not directly bound but nonetheless transcriptionally linked to critical wt-p53 functions (e.g. cell cycle control, DNA repair) most strongly associated with subsequent metastatic outcome. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 101st Annual Meeting of the American Association for Cancer Research; 2010 Apr 17-21; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2010;70(8 Suppl):Abstract nr 106.