Enhancer-promoter distance is a potent modulator of gene expression
To boost transcription of a target gene, enhancer sequences must make contact with the gene’s promoter. This crucial meeting is mediated by interacting proteins and the formation of chromatin loops that bring distant enhancers and promoters together. Although it’s clear that enhancers increase transcription this way, the primary mechanisms by which an enhancer’s target genes are defined remain a matter of debate. A report by Quintero-Cadena et al. in the December issue of G3 presents a step toward unravelling this mystery.
There are two general schools of thought about how enhancer-promoter (EP) pairs are defined. One is that enhancers select particular promoters through specific protein-protein and protein-DNA interactions. The other model is that the EP interactions are essentially non-specific, and the key factor in determining the relationship is the distance between the enhancer and promoter. In this view, closer enhancers lead to more transcription. If the distance between enhancers and promoters is indeed a major factor in their pairing, the authors reasoned, there should be a correlation between the expression levels of genes near one another, with nearby genes being up- or down-regulated together. By analyzing the previously published transcriptomes of five eukaryotic species, the researchers found that such a relationship does exist—and that the correlation between the expression of nearby genes decays exponentially with distance.
In existing in situ hybridization data from fruit flies, the researchers found further evidence of a link between EP distance and transcription. Nearby genes were likely to be expressed in the same tissue types—and this correlation also decays exponentially with distance between the genes. These data, the authors conclude, support a model in which EP distance plays a major role in determining which genes are activated by a given enhancer. This implies that rather than enhancers and promoters interacting in highly specific pairings, they may instead be broadly compatible with each other.
To test their model, the authors inserted DNA constructs into the C. elegans genome. These constructs consisted of one of two enhancers plus a gfp expression reporter with or without a 2 kb spacer between the enhancer and reporter. In agreement with their model, the researchers observed more gfp transcription without the spacer, when the enhancer was closer to the promoter. They also noticed that transcription of the two native genes that flanked their foreign insert was markedly decreased when the spacer was present. But the transcription of these genes was unaffected when the construct without the spacer was introduced. This also supports their model because the spacer not only increased the distance between the native genes and the new enhancer, it also increased the distance to the genes’ corresponding natural enhancers. When the inserted sequence lacked the spacer, the inserted enhancer could compensate for this separation from natural enhancers.
Altogether, these findings imply that most enhancers are not specific to promoters. However, work by other groups does suggest that EP pairs are often specific. One interpretation of these seemingly conflicting findings, supported by some other research, is that enhancers could have specificity for classes of promoters rather than individual promoters. The limited number of promoter types could then confer a degree of EP specificity, while also allowing room for a EP pairing to be dependent on distance. This work thus does not necessarily contradict the prior research, but instead provides a step toward integrating these findings into a comprehensive model.
Quintero-Cadena, P.; Sternberg, P. Enhancer Sharing Promotes Neighborhoods of Transcriptional Regulation Across Eukaryotes.
G3, 6(12), 4167-4174.
We found a similar relationship in a human cell line: “The relationship between enhancer distance and gene expression fold-change followed an exponential decay function. We found that the relationship is well described by the function of enhancer to gene-start site distance: F(dist) = e ∗ ln(dist) / dist.”
Neurobiol Dis. 2018 Jun;114:53-64.