Building the enhancer activity rulebook

Broad researchers use a large-scale synthetic biology approach to explore how different genomic features affect enhancer function.

Lauren Solomon, Broad Communications
Credit: Lauren Solomon, Broad Communications

Noncoding DNA that regulates gene expression is becoming ever more relevant to genomic studies, but researchers can’t yet predict the activity of a regulatory element simply by looking at its sequence. Towards this goal, a team led by Broad Institute researchers used the transcription factor PPARγ as a case study to uncover a set of rules governing gene enhancer activity. The ultimate goal is to create a predictive model that uses DNA sequence alone to determine enhancer activity in a given cell type.

According to the textbook model of genetic enhancer function, a noncoding stretch of DNA is bound by a complementary sequence on a transcription factor protein to increase expression of a downstream gene. But it’s not always that simple — for example, the mouse genome contains approximately 1.5 million sites for PPARγ, but the protein only binds to a small percentage of these sites and their regulatory activity varies substantially.

In a series of synthetic biology experiments published in PNAS, the research team — led by graduate student Shari Grossman and Broad Institute director Eric Lander — manipulated thousands of PPARγ binding sites, varying a number of elements, and cloned the sites into plasmids with uniquely barcoded genes for massively parallel reporter assays. The researchers introduced the plasmids into mouse adipocytes and observed PPARγ binding and the related gene expression levels to tease apart the effects of chromatin structure, PPARγ’s propensity to bind to a certain site, and interactions with other transcription factors on binding and regulatory activity. The barcoding allowed the researchers to identify how each artificial change affected enhancer activity.

All of the potential binding sites taken directly from the genome successfully bound PPARγ in plasmids, indicating that unbound sites in the genome have the potential to bind PPARγ but are likely closed off due to chromatin structure. However, only the enhancers matching those bound in the genome were active regulatory elements.

To determine the additional factors at play, researchers tweaked the DNA flanking either side of the binding sites, and found that certain groups of transcription factors were binding nearby and interacting with PPARγ to either increase or decrease its enhancer’s activity level in a predictable manner. The results suggested a consistent series of rules for how different families of transcription factors interact with one another. Going forward, the researchers plan to test their model in other cell types to determine if these rules apply more generally.

Paper(s) cited

S. Grossman et al. Systematic dissection of genomic features determining transcription factor binding and enhancer function. PNAS. Online January 30, 2017. DOI: 10.1073/pnas.1621150114

A. Melnikov et al. Systematic dissection and optimization of inducible enhancers in human cells using a massively parallel reporter assay. Nature Biotechnology. Online February 26, 2012. DOI: 10.1038/nbt.2137