DNA modifications in specific nuclei populations purified from human brain tissue

Abstract

Epigenetic variation has been implicated in brain development and the progression of brain disorders including neuropsychiatric and neurodegenerative diseases. The human cortex is highly complex, comprising a mix of different neural cells. We sought to generate reference maps of DNA modifications in each cell type using a Fluorescence Activated Nuclei Sorting (FANS) method developed by our group to separate cell-specific nuclei from bulk cortex tissue using cell-specific markers. DNA from each nuclei population was isolated and sequenced using Oxford Nanopore technology. DNA sequence was base-called and base modification (5-methyl-cytosine and 5-hydroxymethyl-cytosine) detected using Dorado. Modkit was used to aggregate modification calls at CpG sites. Mean coverage of ~30x against the reference genome was obtained, and each sample had more than 28 million CpG sites with at least 10x coverage. DNA methylation data was compared to that obtained using the Illumina EPIC array for the same cell types, with a high overall correlation between sites profiled using both methods. There were striking differences in the pattern of DNA hydroxymethylation across the genome in different cell types, with much greater levels found in neuronal nuclei, consistent with previous studies. We compared DNA hydroxymethylation profiles identified using nanopore sequencing with data generated via the sequencing of immunoprecipitated (IP) 5-hydroxymethylcytosine. In summary, using FANS and nanopore sequencing we have developed a reliable pipeline for studying epigenetic changes in purified neural cell types from the human brain. The additional coverage obtained using nanopore sequencing compared to array- and IP-based approaches, and the ability to directly phase DNA modification data with DNA sequence variation, highlights the utility of this approach for identifying epigenetic variation associated with human brain disease.

Biography

After graduating from Oxford, Darren trained to be a science teacher and then spent four years teaching. In 1996, he moved to Exeter to study for a PhD, with Professor Nick Talbot, entitled ‘Regulation of the pathogenicity gene MPG1 in the rice blast fungus Magnaporthe grisea’. Darren’s subsequent post-doctoral positions have been in the field of bioinformatics, including several projects involving comparative genomics of fungal phytopathogens, as well as transcriptional regulation in M. grisea. He has also been a bioinformatician at the Exeter Sequencing service and in the Complex Disease Epigenomics group. In 2023, Darren joined the NIHR Exeter Biomedical Research Centre as the main bioinformatician within the neurodegeneration theme.