Nucleosomes are an essential player in gene regulation, dictating DNA accessibility for most transcription factors. Yet our current understanding of the importance of nucleosome positions for transcription factor binding is limited, partly due to a lack of methods for precisely mapping their positions. The recent advent of SMAC-seq, which couples m6A methylation of accessible regions with long-read sequencing, provided a solution. We set out to develop a computational tool to translate SMAC-seq accessibility marks into precise maps of nucleosome positions genome-wide. We validated our methodology against traditional MNase-seq and show that our approach faithfully reproduces published nucleosome positions at increased resolution.
With this new genomic lens, the next question became – where should we look? The y-globin genes, whose identical promoters are unresolvable by short-read MNase-seq, presented an interesting case study. These genes are developmentally switched off by two well characterised repressors, but how this repression is maintained remains poorly understood. One hypothesis is that the repressors recruit chromatin remodelers to shift nucleosomes and package the genes in an inactive state.
Our tool revealed the nucleosome landscape of these genes. We observed a remarkably tight array of well-positioned nucleosomes at the repressed promoters. Knockouts of both repressors as well as base editing of their key binding sites disrupted this repression array, indicating that both repressors play a role in this positioning. Surprisingly, we also saw marked differences in the nucleosome landscape between the two y-globin genes, challenging current conceptions that regulation of both genes is identical due to their genomic similarity.
We present a new methodology to precisely map nucleosome positions from SMAC-seq reads, and using it generate data to suggest that permanent silencing of the y-globin genes is due to repressors influencing the chromatin landscape.