DNA methylation at CpG dinucleotides is the most well-characterized epigenetic modification. Bisulfite conversion combined with sequencing provides a quantitative, genome-wide readout of this mark.
Bisulfite Chemistry
Treatment of genomic DNA with sodium bisulfite (NaHSO₃) at low pH (pH 5.0, 50–70 °C, 4–16 hours) deaminates unmethylated cytosines to uracil, while 5-methylcytosine (5mC) and 5-hydroxymethylcytosine (5hmC) are resistant to deamination. After PCR amplification, uracil is read as thymine, and 5mC is read as cytosine.
The conversion rate should be >99% to be reliable. An unconverted lambda phage DNA spike-in is often used to measure conversion efficiency.
Methods
Whole-genome bisulfite sequencing (WGBS): sequencing the entire genome after bisulfite conversion. This is the gold standard for comprehensive methylation analysis at single-base resolution across the entire genome. Computational costs are high: 30× coverage requires ~900 million reads for the human genome. As bisulfite-converted DNA is largely C→T (and G→A on the complementary strand), alignment requires a specialized three-letter genome index (C→T converted on both strands).
Reduced representation bisulfite sequencing (RRBS): genomic DNA is first digested with a methylation-insensitive restriction enzyme (MspI, C↓CGG), which cuts at CpG-rich regions. Fragments are size-selected (40–220 bp), bisulfite-converted, and sequenced. RRBS enriches for CpG islands, promoters, and other regulatory regions at ~10–20% of the cost of WGBS.
Targeted bisulfite sequencing: PCR primers are designed to amplify specific regions of interest from bisulfite-converted DNA. The PCR product is Sanger-sequenced or NGS-sequenced. Targeted multiplexed panels covering hundreds of regions are available commercially. This approach is cost-effective for projects focused on specific gene panels.
Pyrosequencing: a quantitative, real-time sequencing-by-synthesis method for short amplicons (100–200 bp) from bisulfite-converted DNA. The peak height at each C position reflects the ratio of C (methylated) to T (unmethylated), giving a quantitative methylation percentage at each CpG site. Pyrosequencing is the gold standard for validation of methylation marks identified by array or NGS.
Data Analysis
Methylation at each CpG is reported as a β-value: β = M / (M + U + 100), where M is methylated and U is unmethylated reads. β-values range from 0 (fully unmethylated) to 1 (fully methylated).
Differential methylation analysis identifies differentially methylated positions (DMPs) and regions (DMRs) between conditions. Most methods use a beta-binomial model to account for the variable coverage across CpG sites.
Distinguishing 5mC from 5hmC
Standard bisulfite sequencing cannot distinguish 5-methylcytosine from 5-hydroxymethylcytosine, both are protected from deamination. To specifically detect 5hmC, additional methods are required:
- TAB-Seq: T4 β-glucosyltransferase protects 5hmC with glucose, then TET1 oxidizes 5mC to 5caC, which is then converted to uracil by bisulfite. Only 5hmC reads as C.
- oxBS-Seq: potassium perruthenate (KRuO₄) oxidizes 5hmC to 5fC, which is converted to uracil by bisulfite. 5mC remains as C. Comparing oxBS-Seq with standard BS-Seq gives the 5hmC level by subtraction.
