A gene’s precise chromatin context profoundly affects its regulation. For example, genes located near nucleosomes containing acetylated histones are generally actively transcribed. By contrast, genes located near nucleosomes containing histones methylated at specific lysine residues are usually transcriptionally repressed. But is the opposite also true? Do specific gene sequences affect the surrounding epigenetic landscape when placed into a new chromatin context? Furthermore, would different genes be regulated in the same manner if placed in the same chromatin context? In the January, 2013 issue of Cell Reports, Chen et al., attempted to answer these questions using an innovative approach.
The authors reasoned that the Saccharomyces cerevisiae gene deletion library represented an untapped resource for studying gene position effects upon the surrounding epigenetic landscape. In the deletion library, a Rap1-driven kanamycin resistance cassette (kanMX) systematically replaced nearly every gene in the yeast genome. Therefore, the exact same gene was inserted into multiple different chromatin contexts. The authors thus identified an experimental model that effectively controls for gene sequence while simultaneously varying the surrounding epigenetic environment.
Chen et al. found that compared to the corresponding wild type gene, insertion of kanMX at different loci did not affect epigenetic modifications in the surrounding chromatin landscape. However, kanMX expression significantly varied depending on where it was located in the genome. Given that gene insertion did not alter nearby epigenetic modifications, the authors next asked whether or not the existing epigenetic marks are predictive of kanMX expression. Correlating published data for the position of seven different types of histone modifications and the observed levels of differential kanMX expression in the yeast deletion libraries, the researchers found that histone H3 trimethylated on lysine 36 (H3K36me3) marks were found at sites with lower kanMX expression. The H3K36me3 modification was not associated with repression of wild type gene expression, suggesting that this mark can specifically regulate expression of certain genes, such as kanMX. The authors then showed that the molecular mechanism governing the H3K36me3 modification was dependent upon Rpd3 histone deacetylase complex-mediated recruitment of Set2 methyltransferase, the enzyme responsible for adding the H3K36me3 modification. Accordingly, transcription of kanMX increased in yeast strains with either Rpd3 or Set2 deletions. The H3K36me3 epigenetic modification seems to be interfering with Rap1 transcription factor binding in the kanMX cassette, thereby repressing its transcription. Finally, the authors showed that H3K36me3 modification repressed kanMX expression at some insertion positions, which likely prevented the recovery of select gene deletion strains in the past.
The idea of using the yeast gene deletion library represents a simple, yet highly effective, means to study gene position effects. In their report, Chen et al. demonstrated a novel regulatory interaction between the kanMX cassette and the H3K36me3 epigenetic mark via inhibition of Rap1 transcription factor-mediated gene activation. In the future, it will be interesting to see how different gene deletion libraries respond to novel gene insertions at diverse chromatin contexts. For example, further studies will show whether or not different reporter genes respond to H3K36me3 marks in the same way as kanMX does. In addition, gene deletion libraries are either already available, or soon will be, for more complex organisms including worms, flies, plants, and mice. It will be interesting to see how the more complex epigenetic mechanisms that exist in those organisms, such as DNA methylation and RNAi pathways, influence the expression of genes placed into those different epigenetic and chromatin contexts.
Chen M, Licon K, Otsuka R, Pillus L, & Ideker T (2013). Decoupling epigenetic and genetic effects through systematic analysis of gene position. Cell reports, 3 (1), 128-37 PMID: 23291096