People can have some pretty strange tastes. From mayonnaise on fries to syrup on spaghetti, there’s no accounting for personal preference. Proteins can have some interesting preferences too, where some regions of the genome look better than others. To get at these preferences, a recent publication from the Laboratory of Bing Ren at UC San Diego sought to find the histone modification tastes of proteins that prefer to bind enhancers.
Enhancers upregulate specific genes during development, but we’re still unclear on exactly how they work. Mono-methylation of lysine 4 on histone H3 (H3K4me1) is a dynamic modification that specifically marks both active and primed enhancers, while trimethylation (H3K4me3) marks promoters. The presence of H3K4me1 has been used to generate enhancer maps. However, the actual mechanistic role of H3K4me1 at enhancers remains unclear. H3K4me1/3 are both suspected to recruit chromatin modifiers/transcription factors, but identifying enhancer-specific proteins preferring H3K4me1 over H3K4me3 is no easy task.
With that goal in mind, the talented team sought to identify H3K4me1-associated proteins and their binding characteristics at enhancers. The group used SILAC (Stable Isotope Labeling by Amino acids in Cell culture) to label all proteins in HeLa cell nuclear extract prior to pull down with either H3K4me1- or H3K4me3-marked nucleosomes. By using heavy isotopes for one methylation state and light for the other, they were able to identify putative H3K4me1-prefering proteins with differential abundance in each pull down.. They also used various ChIP-seq and X-ray crystallography experiments in mouse embryonic stem cells (mESCs) to validate their findings.
Using these approaches, they found that:
- A subset of proteins prefer binding to H3K4me1 over H3K4me3, including known enhancer-associated chromatin modifiers such as BAF (SWI/SNF) complex members
- Knockout of the H3K4me1-specific methyltransferases KMT2C and KMT2D in mESCs result in dramatic reduction of H3K4me1 and several known chromatin modifiers at the same loci, but not H3K4me3
- Catalytic inactivation of KMT2C/2D results in fewer changes in H3K4me1, but regions that do lose methylation also lose specific chromatin modifiers
- Using a nucleosome remodeling assay, they found that H3K4me1-marked nucleosomes were more efficiently remodeled by the BAF complex
- After solving the crystal structure of a specific BAF protein (BAF45C), they found that it may prefer H3K4me1 over H3K4me3 due to a small binding pocket that could not accommodate H3K4 tri- or di-methylation.
Overall, this work suggests that H3K4me1, but not H3K4me3, has an active role in recruiting the BAF complex and other chromatin modifiers to enhancers. This binding is likely important during development and differentiation. Follow up on the other proteins identified in this paper may provide further insight into enhancer function, which is still not well understood. So just like people, finding out a protein’s preferences can really help you get to know them.
Check out the full article at Nature Genetics, January 2018