Enhanced Reduced Representation Bisulfite Sequencing Overview
Enhanced Reduced Representation Bisulfite Sequencing (ERRBS) provides quantitative base-pair resolution of cytosine methylation status at approximately 3 million CpGs genome-wide, located at promoters, CpG islands and shores, and intergenic regulatory regions. This is accomplished through the use of a restriction enzyme that cuts within CCGG sites, followed by bisulfite conversion. The restriction digestion generates low-molecular-weight fragments amenable to NGS library preparation. During bisulfite conversion, unmethylated cytosines are converted to uracil and later as amplified as “T” in PCR, while 5mC and 5hmC are resistant to conversion and are amplified as “C”.
Enhanced Reduced Representation Bisulfite Sequencing Procedure
Genomic DNA (gDNA) is spiked with unmethylated Lambda phage DNA, which is a control to assess bisulfite conversion efficiency. DNA is first digested with MspI, then fragments are end-repaired, A-tailed, and ligated to a methylated Illumina-compatible adapter. Ligation reactions are purified with magnetic beads. Size-selection is performed on an agarose gel, and DNA libraries are bisulfite-converted. PCR is performed with dual-indexed primers to barcode the sample libraries and increase the amount of templates for sequencing. DNA libraries are purified with magnetic beads and QC’ed with the Qubit quantification system and Agilent TapeStation. Libraries are quantified via qPCR and pooled before submission to the Advanced Genomics Core for sequencing.
Enhanced Reduced Representation Bisulfite Sequencing Sample Guidelines
Because of the use of a restriction enzyme, this assay requires gDNA of high molecular weight. It also works best when starting with 50 ng of genomic DNA. Lower input is possible, but we have seen a higher rate of sample failure at lower starting inputs. If you have lower amounts, please contact the Epigenomics Core Managing Director. Typically, ERRBS libraries are sequenced with paired-end 50 cycle sequencing on the NovaSeq 6000, and we aim to obtain 60M to 80M reads per library.