Genome-wide Maps of Histone Modifications in the Human Retina

Abstract

Purpose : Histone modifications have been implicated in controlling the development and maintenance of retinal neurons. However, an extensive in-depth analysis of their profile and function in the human retina is still lacking. Here we systematically characterized its histone modification patterns on a genome-wide scale using chromatin immunoprecipitation accompanied by sequencing (ChIP-Seq).

Methods : We performed genome-wide, epigenetic analyses of human retina samples from non-visually impaired post-mortem donors. Using ChIP-Seq, we mapped a core set of seven histone modification marks, including methylation marks H3K4me1, H3K4me3, H3K36me3, H3K27me3 and H3K9me3 as well as acetylation marks H3K27ac and H3K9ac. After completing our basic quality control, we compared two independent experiments carried out using the same histone mark antibody by correlation analysis, and next performed the irreproducible discovery rate (IDR) analysis for assessing reproducibility of ChIP-Seq data sets after peak calling analysis. We then selected genes from a retina gene expression database (http://retina.tigem.it) and integrated their RNA sequencing data with our histone profiling data in order to discover the genes affected by the histone marks in the human retina.

Results : Our ChIP-Seq analysis produced a total of 16 data sets with more than 600 million aligned reads. The high degree of correlation from a pair of biological replicates indicates that our ChIP-Seq data is high-quality with good reproducibility and reliability. Typical patterns of histone methylations exhibited at promoters, enhancers, intergenic regions and gene bodies are determined. Methylation of H3K4 is positively correlated with transcriptional levels, while H3K27me3 and H3K9me3 have higher signals at promoters of silent genes, suggesting its correlation with gene repression. A number of genes including retina related genes such as MADD, LZTS3, AIPL1 and DDB among those most highly expressed in the human retina, are affected by these histone modification marks.

Conclusions : Our genome-wide histone modification patterns as well as integration with gene expression profiling from a retina database provide a more comprehensive picture of transcriptional and epigenetic regulatory mechanisms in the human retina.

This is an abstract that was submitted for the 2018 ARVO Annual Meeting, held in Honolulu, Hawaii, April 29 - May 3, 2018.