Assessing genome-wide dynamic changes in enhancer activity during early mESC differentiation by FAIRE-STARR-seq

Abstract

Embryonic stem cells (ESCs) can differentiate into any given cell type and therefore represent a versatile model to study the link between gene regulation and differentiation. To quantitatively assess the dynamics of enhancer activity during the early stages of murine ESC differentiation, we analyzed accessible genomic regions using STARR-seq, a massively parallel reporter assay. This resulted in a genome-wide quantitative map of active mESC enhancers, in pluripotency and during the early stages of differentiation. We find that only a minority of accessible regions is active and that such regions are enriched near promoters, characterized by specific chromatin marks, enriched for distinct sequence motifs, and modeling shows that active regions can be predicted from sequence alone. Regions that change their activity upon retinoic acid-induced differentiation are more prevalent at distal intergenic regions when compared to constitutively active enhancers. Further, analysis of differentially active enhancers verified the contribution of individual TF motifs toward activity and inducibility as well as their role in regulating endogenous genes. Notably, the activity of retinoic acid receptor alpha (RARα) occupied regions can either increase or decrease upon the addition of its ligand, retinoic acid, with the direction of the change correlating with spacing and orientation of the RARα consensus motif and the co-occurrence of additional sequence motifs. Together, our genome-wide enhancer activity map elucidates features associated with enhancer activity levels, identifies regulatory regions disregarded by computational prediction tools, and provides a resource for future studies into regulatory elements in mESCs.

ATAC-STARR-seq reveals transcription factor–bound activators and silencers within chromatin-accessible regions of the human genome

“…A recent study in mouse embryonic stem cells made the same observation using an orthogonal approach, suggesting this phenomenon is present in other mammalian species (Glaser et al 2021). We note that a small percentage of accessible peaks (4.4%) contain both active and silent regions, demonstrating that there can be competing regulatory regions within the same accessible peak.…”

“…To narrow the scope of the assay, recent methods have combined STARR-seq with techniques that capture accessible chromatin to specifically test the regulatory potential of accessible DNA (Buenrostro et al 2013;Wang et al 2018;Chaudhri et al 2020;Glaser et al 2021). As a result, these methods only sample a fraction of the human genome (~2%) while assaying nearly all regulatory elements capable of driving transcription endogenously, because they are derived from open chromatin.…”

Identifying transcription factor-bound activators and silencers in the chromatin accessible human genome using ATAC-STARR-seq

“…This suggests that the majority of accessible DNA has no regulatory potential in this context or, alternatively, that ATAC-STARR-seq is not sensitive enough to measure weakly active or weakly silent regions. A recent study in mouse embryonic stem cells made the same observation using an orthogonal approach, suggesting this phenomenon is preserved in other mammalian species (Glaser et al 2021). Interestingly, a small percentage of accessible peaks (4.4%) contain both active and silent regions, demonstrating that there can be competing regulatory regions within the same accessible peak.…”

“…To narrow the scope of STARR-seq while assaying nearly all endogenously active regulatory elements, recent methods, such as assay for transposase-accessible chromatin coupled to STARR-seq (ATAC-STARR-seq), have combined STARR-seq with assays that capture accessible chromatin to extract and specifically assay the regulatory potential of accessible DNA (Buenrostro et al 2013; Wang et al 2018; Chaudhri et al 2020; Glaser et al 2021). In this way, these assays only sample a fraction of the human genome (∼2%) while assaying nearly all regulatory elements capable of driving transcription endogenously, because they are derived from open chromatin.…”

Challenges and considerations for reproducibility of STARR-seq assays

“…To assess and quantify reproducibility for STARR-seq based studies, we systematically compared 24 studies 10,11,12,13,14,15,16,17,18,19,30,26,31,33,32,40,43,44,45,46,47,48,49,50 that performed STARRseq experiments on mammalian or Drosophila cells (Fig. 4).…”