The activity of genes is coordinated by proteins known as transcription factors, which selectively bind specific regulatory DNA sequences and activate or inhibit gene expression. However, these factors do not operate in isolation, and scientists are increasingly discovering that it can be difficult to predict a gene’s behavior based simply on the presence or absence of a given transcription factor’s binding site. Larry Stanton and co-workers at the A*STAR Genome Institute of Singapore have now unraveled part of the complexity inherent to this system.
The researchers studied the RE1-silencing transcription factor (REST), which inhibits target genes by binding ‘repressor element 1’ (RE1) sites throughout the genome and plays an important role in directing embryonic stem (ES) cells to develop into more specialized cells. REST is a poor inhibitor on its own, and relies on an array of protein partners. Accordingly, Stanton and his team performed a genome-wide search to find target genes that specifically recruit REST in conjunction with various known co-repressors of REST.
To their surprise, they observed striking variability in the functional recruitment of REST on different genomic targets. “We have shown that not all REST-binding sites are the same,” says Stanton. Some sites assemble REST and multiple cofactors into complexes, while other sites recruit REST by itself. The extent of complex formation was largely dependent on the strength of interaction between REST and a given site, which is partly determined by the extent to which these sites resemble the ‘ideal’ RE1 sequence (see image).
The strength of interaction also appears to be a key determinant of the potency of its effects on gene regulation. “In general, ‘strong’ sites recruit REST complexes, which leads to repression of those target genes,” explains Stanton. These findings could explain confusing outcomes from previous attempts to understand gene regulation based on maps of transcription factor binding. “Our results reflect a more general phenomenon,” says Stanton, “where co-assembly of transcription factors and cofactors—not any one factor alone—is required to control gene expression.”
REST only represses a relative minority of the genes that it binds, and Stanton’s team is still working to untangle this protein’s complicated role in ES cell fate determination. “We have found that ES cells can maintain themselves in an undifferentiated state without any REST, contrary to claims that it is required to maintain pluripotency,” says Stanton, “but its absence does have a significant impact on these cells’ differentiation potential.”
The A*STAR-affiliated researchers contributing to this research are from the Genome Institute of Singapore.