Dissecting genetic and epigenetic contribution to chromatin opening through a chromatin-integrated reporter assay

Chromatin accessibility at cis-regulatory elements (CREs) is influenced by both cis- encoded and trans-acting mechanisms. This includes short DNA motifs recognized by transcription factors (TFs), as well as chromatin post-translational modifications on histones (PTMs). A challenge in identifying determinants of chromatin accessibility is that TF binding and deposition of activating chromatin marks are often concomitant at endogenous CREs, making it difficult to disentangle their individual contribution to chromatin accessibility and transcription activation. To overcome this limitation, I established a chromatin-integrated reporter assay to test whether regulatory sequences (~250bp) can recruit TFs, open chromatin, and activate transcription at an ectopic site. I created libraries containing hundreds of regulatory sequences and inserted them at the same landing pad located at a chromatin locus devoid of activating or repressing chromatin marks. I profiled whether these short genetic elements could recapitulate their endogenous chromatin accessibility using a parallelized chromatin accessibility assay based on targeted Single Molecule Footprinting. I analyzed ~400 fragments containing promoters and enhancers active in mouse embryonic stem cells (mESCs), and compared the chromatin accessibility at ectopic sites with their endogenous levels. As expected, CTCF-containing sequences (used as positive control) autonomously re-establish chromatin accessibility at identical levels when inserted at the ectopic locus. In contrast, sequences bound by other TFs, including NRF1, NFY and pluripotency factors such as KLF4, OCT4 show only partial recapitulation of their endogenous accessibility when isolated at the ectopic site; and this reduction was not explained by CpG content alone. To investigate whether sequence context beyond the core ~250 bp CRE could explain reduced accessibility, I performed a fragment-length experiment on a single CRE. Although preliminary, the results suggest that accessibility at the ectopic site reached endogenous levels at ≥1 kb. This suggests that, although the regulatory sequences are genetically the same of their endogenous counterparts, the ~250 bp fragments inserted at the ectopic locus may lack additional sequence features required for full activity – such as histone PTMs, whose deposition depends on cooperative interactions between multiple TFs. To further explore this possibility, I next investigated the role of H3K27Ac and found that the extent to which CREs establish chromatin accessibility is anti-correlated with the presence of H3K27Ac at the endogenous locus, suggesting that H3K27Ac may enhance accessibility in its native context. To test this hypothesis, I globally perturbed acetylation by chemically inhibiting the histone acetyltransferase p300, that deposits acetylation at a subset of enhancers in mESCs. As predicted, chromatin accessibility was reduced at endogenous loci, and the amplitude of the loss scaled with the degree of H3K27Ac reduction, but almost unaffected at the ectopic locus, which is located within a chromatin neutral environment. Finally, I implemented a proof-of-concept reporter assay to simultaneously quantify chromatin accessibility and transcriptional output for multiple ectopically inserted CREs, demonstrating the feasibility of parallel measurements at scale.