摘要：Sequence motifs are short, recurring patterns in DNA that can mediate sequence-specific binding for proteins such as transcription factors or DNA modifying enzymes. The androgen response element (ARE) is a palindromic, dihexameric motif present in promoters or enhancers of genes targeted by the androgen receptor (AR). Using chromatin immunoprecipitation sequencing (ChIP-Seq) we refined AR-binding and AREs at a genome-scale in androgen-insensitive and androgen-responsive prostate cancer cell lines. Model-based searches identified more than 120,000 ChIP-Seq motifs allowing for expansion and refinement of the ARE. We classified AREs according to their degeneracy and their transcriptional involvement. Additionally, we quantified ARE utilization in response to somatic copy number amplifications, AR splice-variants, and steroid treatment. Although imperfect AREs make up 99.9% of the motifs, the degree of degeneracy correlates negatively with validated transcriptional outcome. Weaker AREs, particularly ARE half sites, benefit from neighboring motifs or cooperating transcription factors in regulating gene expression. Taken together, ARE full sites generate a reliable transcriptional outcome in AR positive cells, despite their low genome-wide abundance. In contrast, the transcriptional influence of ARE half sites can be modulated by cooperating factors.
