Exploring the Genomic Binding Landscape of the Oncogenic Transcription Factor TCF3-PBX1 in Acute Lymphoblastic Leukemia

Abstract

Treatment of acute lymphoblastic leukemia (ALL) has become quite effective over the last two decades. However, further progress is limited by a lack of understanding of the biological processes that drive this disease. This understanding will be crucial for developing effective targeted therapies. The oncogenic transcription factor TCF3-PBX1 is expressed in 5% of ALL cases consequent to a somatic chromosomal translocation between chromosomes 1 and 19. It is broadly accepted that TCF3-PBX1 acts as a transcriptional activator resulting in dysregulated proliferation, survival and/or differentiation of hematopoietic cells. Studies have implicated the histone acetyltransferase p300 as an important player in mediating these transcriptional effects. However, the particular binding sites of TCF3-PBX1 and the genes affected have not been identified. We hypothesized that TCF3-PBX1 binds to B-lymphopoietic cis-regulatory regions, such as enhancers and promoters, via the PBX1 homeodomain and affects gene transcription by recruiting p300 in a manner that leads to leukemogenesis. Through the use of chromatin immunoprecipitation followed by DNA sequencing (ChIP-seq), this project identified over 3500 genomic regions where TCF3-PBX1 binds in a t(1;19)-bearing, ALL-derived cell line that expresses endogenous TCF3-PBX1. Most binding sites have characteristics associated with cis-regulatory elements, such as particular histone modifications, and are commonly occupied by B-lymphopoietic transcription factors and PBX family members. TCF3-PBX1 additionally recruits p300 to novel sites, which plays a role in increasing the expression of nearby genes. One hundred and sixty-three genes had nearby TCF3-PBX1 binding sites and were differentially expressed (determined using RNA-sequencing) upon short-hairpin RNA-mediated knockdown of TCF3-PBX1 in two cell lines, suggesting that they are direct transcriptional targets of TCF3-PBX1. Fifty-eight of these genes were additionally found to be differentially expressed between primary, patient-derived ALL samples that likely do, versus do not, harbor the t(1;19) translocation. Together, these results support our hypothesis described above and address the need to identify the transcriptional targets of the TCF3-PBX1 fusion oncoprotein. These target genes and their associated cellular pathways will inform future studies to elucidate the molecular mechanisms by which TCF3-PBX1 disturbs lymphoid development in ALL and will contribute to the eventual development of novel treatments that target these mechanisms.