Compatibility or specificity? The interaction between enhancers and promoters in human
Figure 1 – Effect of different enhancers on distinct promoters. Created in BioRender.com
Abstract
Gene regulation in the human genome is mediated by complex mechanisms involving the interaction between distal enhancers and specific promoters. However, the understanding of the nature and extent of this interaction remains incomplete. Recent studies have suggested a “biochemical compatibility” model, which has not been tested on the human genome. In a recent study published in Nature, the authors, through a new approach called enhancer x promoter self-transcribing active regulatory region sequencing (ExP STARR-seq), proposed a model in which the intrinsic activities of enhancers and promoters multiplicative combine to regulate gene expression in humans.
Review
Introduction
Enhancers are regulatory genomic regions located upstream, downstream, and within a gene, which enhance its transcription. An enhancer can interact with multiple genes, and a gene can be controlled by multiple enhancers. Promoters are sequences found upstream of the transcription start site (TSS), recognized by transcription factors (TFs) due to the presence of consensus sequences, allowing RNA polymerase to initiate the transcription process.
Since their initial discovery, enhancers have demonstrated the ability to activate various promoters in a non-specific manner1. However, recent studies, such as those conducted in Drosophila, have revealed that promoters of housekeeping genes are more responsive to certain enhancers2. The findings of these studies suggest the existence of a “biochemical compatibility” wherein different enhancers activate specific promoters by recruiting specific transcription factors (TFs). However, this model has not yet been extensively tested in higher eukaryotes, such as humans.
In this study3, the authors analyzed the interaction between 1000 enhancers and 1000 promoters in K562 human erythroleukemia cells using a novel approach called ExP STARR-seq. This approach aimed to investigate the compatibility between all possible combinations of selected enhancer (E) and promoter (P) sequences.
Design of the ExP STARR-seq technology
To analyze all possible pairs between the selected enhancer and promoter sequences, the authors designed a modified STARR-seq4 plasmid, in which each promoter and enhancer had a total length of 264 bp. In particular, they modified both the cloning technique and RNA-seq steps to allow the analysis of different enhancer sequences combined with different promoter sequences. The enhancer and promoter sequences were synthesized in an oligonucleotide array format. Subsequently, the sequences were separately amplified via PCR and inserted into the STARR-seq plasmid vector, in the enhancer and promoter positions, in a single cloning step by Gibson assembly. Moreover, since the reporter mRNA of the original STARR-seq protocols does not contain the full promoter sequence, the authors included a random 16bp barcode upstream the enhancer sequence to precisely identify the enhancer-promoter pair that produces a specific reporter RNA. This was achieved building a promoter-barcode dictionary via sequencing of the PCR amplicon containing the promoter and plasmid barcode sequences. The authors then transfected the obtained plasmid pool in K562 cell, and sequenced the STARR-seq RNA and input DNA libraries. Finally, by computational analysis of sequencing data, they quantified the RNA abundance generated by each enhancer-promoter pair.
Figure 2 – Design of ExP STARR-seq plasmid with the type of E and P selected from the authors. Created in BioRender.com
Discussion
The results obtained from the ExP STARR-seq indicate that there is not a particular specificity between enhancers and promoters, as most enhancers activated promoters similarly, and most promoters showed similar levels of activation by enhancers. This suggests the existence of broad compatibility between enhancers and promoters. Additionally, by mathematical modelling, researchers proved that the intrinsic activities of enhancers and promoters multiply to produce RNA output in the ExP STARR-seq.
Despite this broad compatibility of the analyzed sequences of E and P, the authors attempted to identify classes that exhibited more specific preferences. They identified two groups of sequences for both enhancers (referred to as E1 and E2) and promoters (referred to as P1 and P2). The two classes of enhancers distinguished by the authors exhibit different characteristics. E1 class is made by strong enhancers, whereas enhancers belonging to the E2 class reflect weak enhancers. Moreover, they observed that E1 more strongly activate P1 compared to P2, while E2 activate both classes of promoters similarly. The promoter classes are distinguished based on transcriptional activity. Specifically, promoters belonging to the P1 class are associated with genes expressed more variably (for example erythroid genes), showing low intrinsic activity, and exhibiting a stronger response to E1. As for promoters in the P2 class, they are primarily found in housekeeping genes and are characterized by: i) high intrinsic activity; ii) a weaker response to E1; iii) binding of transcription factors that play significant roles in transcriptional activation, such as GABPA and YY1. To further understand the properties of promoter classes, the authors wondered whether the binding of these transcription factors influenced the low reactivity that P2 promoters show towards E1 enhancers. Through ChIP-seq analysis, they confirmed that GABPA and YY1 also bind to E1 enhancers. Moreover, by inserting or scrambling GABPA or YY1 motifs in promoter sequences and measuring their activity using ExP STARR-seq, they showed that the presence or absence of such motifs decrease or increase promoter responsiveness to E1 enhancers. This suggests the presence of built-in enhancers in P2 promoters, making them less responsive to E1 enhancers.
Conclusions
In conclusion, the effect of enhancers on RNA expression is determined by various factors that multiplicatively combine: P activity, E activity, E-P contact, and compatibility between E-P classes. E-P contact has previously been studied by the same authors using the ABC model5 which assumes that all enhancers and all promoters are equally compatible and that enhancer activity and 3D contact frequencies between enhancer and promoter regulate the relative effect of an enhancer on gene expression.
It is important to consider some limitations of this study. First, these experiments were not sufficiently powered to quantify possible compatibility among the weakest enhancers and promoters. Moreover, the episomal STARR-seq assay does not capture all mechanisms that might influence transcriptional activation in the genome.
Finally, the authors used only one cell line. It could be beneficial to understand whether the compatibility between enhancers and promoters is due to the specific cell type used in this analysis or represents a general mechanism underlying human gene expression.
To advance in the understanding of the interaction between enhancers and promoters, future studies could link the sequences that control enhancer and promoter activities with effects on particular biochemical steps in transcription (for example transcription initiation or elongation).
This study provides the groundwork for understanding the mechanisms of gene regulation in humans. Indeed, comprehending the interaction between enhancers and promoters can have implications in biomedical research, contributing to the development of more precise and efficient genetic therapies.
References
- Nguyen, T. A. et al. High-throughput functional comparison of promoter and enhancer activities. Genome Res. 26, 1023–1033 (2016).
- Zabidi, M. A., Arnold, C. D., Schernhuber, K. & Pagani, M. Enhancer–core-promoter specificity separates developmental and housekeeping gene regulation. Nature 518, 556–559 (2015).
- Bergman D.T., Jones T.R., Liu V., Ray J., Jagoda E., Siraj L., Kang H.Y., Nasser J., Kane M.F., Antonio Ray Rios, Nguyen T.H., Grossman S.R., Fulco C.P., Lander E.S. & Engreitz J.M. Compatibility rules of human enhancer and promoter sequences. Nature 607, 176–184 (2022).
- Arnold, C. D. et al. Genome-wide quantitative enhancer activity maps identified by STARR-seq. Science 339, 1074–1077 (2013).
- Fulco, C. P. et al. Activity-by-contact model of enhancer-promoter regulation from thousands of CRISPR perturbations. Nat. Genet. 51, 1664–1669 (2019).