Understanding how the same DNA sequence instructs an abundance of transcriptional programmes associated with different cell states cannot be complete without considering epigenetics. The ability of cells to commit and memorize specific epigenetic programmes is essential for the construction of complex tissues and organs during development. However, because transcriptional and epigenetic changes go hand in hand, determining the roles of epigenetic mechanisms in cell state acquisition remains a notoriously difficult task. During my undergraduate studies, I stumbled upon a landmark paper from the lab of Howard Cedar, which sparked my long-term interest in this problem.
This study aimed to address how genome-wide DNA methylation patterns are established and instructed by the DNA sequence. In vertebrates, 5-methylcytosine occurs mainly in the context of CG dinucleotides, and somatic cells present a bimodal landscape of methylation: the bulk of the genome comprises low-density CG sequences and is uniformly decorated by methylation, whereas high-density CG regions, termed CG islands (CGIs), are largely hypomethylated. The coupling of methylation maintenance to DNA replication provides a simple and robust mechanism for epigenetic memory across cell divisions. Yet, an unresolved question was how the bimodal methylation landscape is initially installed in the early embryo. Utilizing the ability of mouse embryonic stem cells to methylate and demethylate exogenously introduced genomic sequences, the authors showed that an isolated 5′ region of the Aprt gene contains all the necessary information to be recognized as an endogenous CGI. When introduced unmethylated to the cells, the sequence remained hypomethylated. But when the same sequence was introduced fully methylated, it underwent substantial demethylation. A major finding of this study was the identification of sequence motifs that recruit the Sp1 transcription factor to the Aprt CGI. In turn, the binding of Sp1 was shown to protect the CGI from acquiring de novo methylation in stem cells and embryos. This principle proved to be a general mechanism by which trans-acting factors establish the stereotypic methylation landscape observed in differentiated cells.
This is a preview of subscription content, access via your institution