Sequence Decay in Alu Subfamilies Governs Transcription Factor Access Within the ERα Signaling Architecture

Yusuf Qavi

Mentor: Dr. Sreejith Nair, Department of Oncology, Georgetown University Medical Center.

Date/Time: August 25th, 2026 at 2:45 PM.

Abstract: Estrogen receptor alpha (ERα) drives estrogen-responsive gene programs in ER+ breast cancer, but its genome-wide binding pattern depends heavily on the pioneer factor FOXA1, which must first open chromatin for ERα to access it. Alu elements, the most abundant repetitive element in the human genome, have been proposed as a structural influence on this regulatory landscape, but whether Alu's own sequence content shapes transcription factor access has not been directly tested at the level of individual genomic copies.

Using independently validated ChIP-seq, ATAC-seq, and PRO-seq data in MCF-7 cells across four estradiol doses, we show that ERα is significantly depleted at young Alu subfamilies, and that this exclusion is substantially inherited from FOXA1 rather than intrinsic to ERα itself. Direct motif scanning of every genomic copy of three Alu subfamilies reveals a sequence-level cause: FOXA1's own binding motif is measurably rarer in young Alu than in old Alu (~8-fold), a pattern independently replicated in a second, unrelated transcription factor, GATA3 (~15-fold). This relationship resolves to individual DNA copies, motif-carrying copies of young Alu sit significantly closer to real ERα binding sites than copies that have lost the motif, and to a single base-pair difference between old and young Alu at the site of ERα's own response element. Critically, this intact sequence directly predicts real, measured ERα occupancy, not merely computational motif similarity.

These results establish Alu subfamily age as a genuine, mechanistically resolved determinant of transcription factor access within the ER signaling network, generalized across independent proteins and validated against real binding data. Whether this sequence-level architecture extends to downstream gene expression, pathway activity, or clinical outcome was tested and found unsupported at current resolution.