During development, transcription factors (TF) control cell fate decisions by binding to DNA “enhancer” sequences to drive gene expression. However, the mechanism by which TFs contend with the millions of nucleosomes that tightly wrap DNA and act as barriers to TF binding to enhancers remains a mystery. “Pioneer factors”, TFs that can physically bind to DNA wrapped in nucleosomes, offer one potential solution.
Although there is in vitro evidence that TF-nucleosome interactions occur, it is not known if these interactions take place regularly at natural enhancers in vivo or whether TFs bind opportunistically when nucleosomes get displaced or the DNA is partially unwrapped. TFs can access their targets either by binding DNA directly or by nucleosome binding and displacement; nevertheless, these mechanisms are difficult to distinguish during development.
Dr. Steven Henikoff from the Basic Sciences division and members of his laboratory tackled this challenge using existing and new technologies from the Henikoff lab. The authors employed CUT&RUN, an in situ chromatin profiling method that uses antibody-targeted micrococcal nuclease (MNase) to liberate DNA fragments bound by a target protein. Small fragments of < 120 base pairs represent direct TF contacts with DNA whereas larger fragments indicate nucleosomal protection. CUT&RUN was used during the differentiation of human embryonic stem cells (hESCs) to definitive endoderm (DE) to profile several TFs that engage inactive enhancer chromatin. To analyze the structure of TF-chromatin interactions, the Henikoff lab devised a computational strategy named Enhanced Chromatin Occupancy (EChO), which can infer TF-chromatin interactions at each individual TF binding site across the genome, as well as the likelihood of TFs co-binding simultaneously in a complex versus binding in a mutually exclusive manner to the same locus. The results of this work were published in a recent issue of the journal Molecular Cell.
Led by Dr. Michael Meers, a postdoc in the Henikoff lab, the authors used Sparse Enrichment Analysis for CUT and RUN (SEACR), a recently published methodology from the Henikoff lab to call enriched regions from their CUT&RUN data. The results showed that multiple factors collaborate differentially across developmental time to regulate enhancer usage. EChO was used to characterize TF binding within chromatin using average fragment size. Using this approach, the authors triangulated the positions of TFs on nucleosomes or exposed DNA in vivo during the course of development for the first time.
Dr. Meers describes their unexpected results: “The most surprising finding was that pioneer factors in fact have both mechanisms in their toolkit—some enhancers, typically the ones with strong binding sites, are bound directly by the pioneer factor from the outset without interference from nucleosomes, while other, weaker enhancers need the nucleosome-binding activity of the pioneer factor to wriggle free. Even more interesting, the enhancers that need that extra push from pioneer factors appear to actually have multiple factors working together to achieve their goal in many cases.”
The authors’ finding that TF motif strength is anti-correlated with the incidence of pioneer nucleosome binding suggests that binding via a nucleosome may stabilize interactions at low-affinity motif targets. This supposes a model in which TFs engage low-affinity motifs through a combination of direct DNA binding and nucleosomal anchoring to ensure access, whereas high-affinity motifs are bound during opportunistic windows of DNA unwrapping from the nucleosomal core.
Dr. Meers explains the significance of this work: “The pioneer factor hypothesis has been around for decades, but it’s been nearly impossible to confirm its presence in natural developmental systems, so it’s really exciting to contribute some insight to a debate that’s at the heart of how we understand transcription factor activity. It also gives us a crucial window into how pioneer factors might function in diseases such as cancer, where the enhancer binding activity of pioneer factors such as FoxA1 can “fill in” for the gene regulatory activity of other DNA binding proteins in treatment-resistant cancers. In many cases cancer is accompanied by a reversion to a more naive developmental state to which pioneer factors are the gatekeepers, so it’s a “know your enemy” kind of thing as well.”
In the immediate term, Dr. Meers and his team are interested in following up on the major hypothesis stemming from their findings: that pioneer factors need this dual activity as a “failsafe” to ensure that crucial enhancer switches get flipped, regardless of nucleosomes being in the way. “We can get at this by specifically removing one of the activities, the nucleosome binding, and observing how enhancer activation responds” explains Dr. Meers.
Developmental enhancers do a lot of heavy lifting in determining cell fate; understanding how they play a role in the context of the binding proteins that coevolved with them is a phenomenal task. “There are a lot of other fascinating directions to go from here though,” says Dr. Meers. “For instance, the partitioning of pioneer factor binding strategies based on strong and weak binding sites drives at the heart of what makes an enhancer in the first place, and how they might have evolved. Could pioneer factors have played a role in giving “proto-enhancers” with weak binding sites that first nudge towards becoming full-fledged members of the gene regulatory network? Or could it be the opposite, that the nucleosome binding activity of pioneer factors has allowed formerly strong binding sites to ‘let themselves go,’ while being confident that they’ll still be activated? These are questions that I’d love to tackle.”
Meers MP, Janssens DH, Henikoff S. 2019. Pioneer factor-nucleosome binding events during differentiation are motif encoded. Mol Cell Jun 10. Pii: S1097-2765(19)30397-1.
This work was supported by the National Institutes of Health and the Howard Hughes Medical Institute.