Study Maps Focused Transcription Factor Binding in Human Cells

Published in Epigenetics & Chromatin, the study leverages extensive collections of ChIP-seq data from more than 1,000 transcription factors in two human cell lines, HepG2 and K562. The team located these highly focused binding sites, termed FBSs, as single-nucleosome depleted regions within cis-regulatory elements. These narrow regions accommodate the majority of total transcription factor binding events, challenging previous assumptions about the physical distribution of regulatory proteins on DNA.

The central finding of the research is that transcription factor binding is not evenly distributed across regulatory elements but is instead concentrated in very narrow zones. In both cell types, approximately 25,000 strong focused binding sites were identified. For more than 90% of the transcription factors analyzed, including pioneer factors such as GATA1 and JUN, binding sites located outside of these focused regions show little to no nucleosome depletion. This suggests that the open chromatin state typically associated with active gene regulation is tightly linked to the collective presence of many transcription factors in a small physical space.

To understand the physical nature of these binding sites, the researchers analyzed the distribution of transcription factor peaks within merged genomic regions. They found that the vast majority of individual transcription factor peaks overlap with other peaks, with 93.6% of peaks in HepG2 cells located less than 100 base pairs from their nearest neighbor. This high degree of overlap indicates that different transcription factors tend to bind together at specific genomic locations. The study demonstrates that these collective binding events are concentrated in the top 30,000 most transcription-factor-enriched merged peaks in HepG2 cells and the top 50,000 in K562 cells.

The focused binding sites are characterized by a significant reduction in nucleosome occupancy, a state known as nucleosome depletion. Using MNase-seq data, the team observed that these sites correspond to the depletion of a single nucleosome. At distal regulatory sites, the nucleosome-free region is flanked by well-phased nucleosomes on both sides. In contrast, promoter regions show a deeper valley of nucleosome occupancy but lack the well-positioned flanking nucleosomes, likely due to the directional positioning of nucleosomes around transcription start sites.

The study highlights a critical connection between the focused binding of a large number of transcription factors and the sustained nucleosome-depleted status of DNA in vivo. The researchers propose that the short DNA residence time of most transcription factors underlies the requirement for a large number of these proteins to maintain an open chromatin state. While a few pioneer factors may initially displace a nucleosome, the rapid turnover of transcription factors means that a continuous presence of many binding partners is necessary to prevent the nucleosome from re-occupying the site.

Sequence analysis of these focused binding sites revealed strong enrichment for motifs associated with essential cellular functions. In HepG2 cells, motifs for Forkhead and Nuclear Receptor families were most significant, while in K562 cells, GATA and Jun/Fos related motifs were prominent. These motifs are concentrated within the focused binding sites but are not significantly enriched in the immediate flanking regions. This specificity suggests that the focused sites are the primary loci for functional gene regulation, distinct from the broader regulatory elements previously identified.

The researchers also examined the evolutionary conservation of these sites, finding that sequence conservation is peaked at the center of the focused binding sites. This conservation is remarkably higher than in the middle areas of the corresponding merged peaks. Furthermore, phenotypically causal genetic variants, including eQTLs, raQTLs, and GWAS SNPs, are strongly enriched within these focused sites. For instance, the enrichment fold for raQTLs in focused binding sites ranged from 2.8 to 3.7 compared to the broader merged peaks, indicating that these narrow regions are likely to harbor variants that affect gene expression and disease traits.

Co-binding analysis demonstrated a high degree of promiscuity among transcription factors at these sites. On average, more than 120 transcription factors from over 25 families are bound to a single focused site in HepG2 cells. Despite this large number of binding partners, the majority of these transcription factors do not contain their canonical binding motifs at the site. This suggests that many transcription factors are recruited through non-canonical mechanisms, such as protein-protein interactions or DNA shape recognition, rather than direct motif recognition.

The study provides a high-resolution map of chromatin accessibility at distal regulatory elements, offering a new perspective on the formation of high-occupancy target loci. By pinpointing the essential functional and sequence features to single-nucleosome depleted regions, the researchers resolved a long-standing paradox regarding how hundreds of transcription factors can bind to a long stretch of DNA while maintaining a nucleosome-depleted state. The findings suggest that the focused binding of a large number of transcription factors is a necessary requirement to sustain an open chromatin environment.

Limitations of the study include the reliance on ChIP-seq data from the ENCODE database, which may be subject to antibody-dependent biases and lab-specific artifacts. The researchers acknowledge that while the basic pattern of focused binding likely reflects true biological mechanisms, specific quantifications of transcription factor enrichment may require further validation with advanced techniques. Additionally, the study is limited to two human cell lines, and the cell-type specific differences in nucleosome occupancy patterns observed suggest that the mechanisms may vary across different cellular contexts.

This research contributes to the understanding of how chromatin accessibility is regulated through the interplay between transcription factors and nucleosomes. By identifying the physical and functional characteristics of focused binding sites, the study provides a framework for interpreting the complex regulatory landscape of the human genome. The proposed model of transcription factor co-binding offers insights into the dynamic nature of gene regulation and the role of collective protein binding in maintaining genomic states.

The implications of these findings extend to the interpretation of genetic variants associated with disease. Since phenotypically causal variants are concentrated in these focused sites, future studies may prioritize these regions for functional validation. The high resolution of the maps generated in this study could aid in the identification of regulatory mechanisms underlying complex traits and diseases, providing a more precise view of the genomic architecture that controls gene expression.

Beyond linear sequences, the researchers investigated spatial genome architecture by mapping distal focused binding sites to high-throughput chromosome conformation capture contacts from ENCODE. Filtering out looping interactions wider than two kilobases revealed that distal focused sites are enriched roughly sixty-fold within three-dimensional chromatin loop anchors relative to random genomic positions, yielding 5,583 unique promoter-distal loops in HepG2 cells and 3,603 in K562 cells. These physical bridges connect distant regulatory elements directly to their putative target genes. While roughly 60 percent of promoter-associated focused sites were shared between the two cell lines, only about 15 percent of distal sites overlapped, illustrating that distal focused sites drive lineage-specific genomic regulation.

The transient nature of regulatory proteins explains why sustained chromatin opening requires dense clusters of factors. Nucleosomes at active chromatin turn over within 30 minutes to two hours, whereas most transcription factors remain bound to DNA for only seconds. This brief residence time allows nucleosomes to readily reclaim exposed sequences unless a high concentration of dynamic factors repeatedly occupies the space. By contrast, structural insulator proteins such as CTCF and cohesin demonstrate fundamentally different dynamics. Although insulator sites exhibit central nucleosome depletion comparable to enhancer focused sites, CTCF maintains an average residence time of 29 minutes, and interphase cohesin binds for several hours. This persistent occupancy physically excludes other regulatory proteins and stabilizes rigid nucleosomes across neighboring flanking sequences.

Functional variant analyses further highlighted the specialized role of these regions. When assessing 3,123 liver-derived chromatin accessibility quantitative trait loci, distal focused sites in HepG2 cells demonstrated a 3.2-fold enrichment compared to broader merged peaks. Promoter regions showed weaker enrichment, likely because baseline nucleosome depletion at transcription starts reduces sensitivity to single mutations. Furthermore, trait-specific genome-wide association variants reflected cellular lineage, showing 1.5-fold to 4-fold enrichment in focused sites over merged peaks for liver-related traits like triglyceride levels in HepG2 cells and immune traits like hemoglobin concentration in K562 cells. Across 100 vertebrate species, focused sites linked to these specific traits displayed significantly higher sequence conservation scores than general focused sites, a distinction absent from expression or reporter quantitative trait loci.

Structural profiling of factor families revealed broad co-occupancy alongside rare individual exceptions. In both cell types, representatives from widespread structural families including C2H2 zinc finger, basic leucine zipper, and basic helix-loop-helix factors were detected at virtually all focused binding sites. Across 804 analyzed transcription factors, however, only 36 proteins, representing roughly 4.5 percent, exhibited strong nucleosome depletion outside of focused sites, and a mere 3.4 percent maintained open chromatin when bound with fewer than ten partners. These solitary clearing capabilities were often strictly cell-type specific. While isolated binding of ARID4B cleared nucleosomes in HepG2 cells, it failed in K562 cells; conversely, ZNF143 maintained depletion in K562 cells but showed none in HepG2 cells, reflecting distinct cellular cofactor partnerships.

Subject of Research: High-resolution mapping of transcription factor binding and chromatin accessibility in human cells

Article Title: High-resolution map of chromatin accessibility – insights into the focused binding of a large number of transcription factors

Article References: Zhu, I., & Landsman, D. (2026). High-resolution map of chromatin accessibility – insights into the focused binding of a large number of transcription factors. Epigenetics & Chromatin, 19(1), Article 14. https://doi.org/10.1186/s13072-026-00665-2

Image Credits: AI Generated

DOI: 10.1186/s13072-026-00665-2

Keywords: chromatin accessibility, transcription factors, nucleosome depletion, HepG2 cells, K562 cells, ChIP-seq, MNase-seq, cis-regulatory elements, gene regulation, epigenetics, high-occupancy targets, DNA residence time, pioneer factors, genomic variants, eQTLs, GWAS, ENCODE project, co-binding, nucleosome positioning, regulatory elements