Abstract
BACKGROUND
DNA methylation is a potential source of disease biomarkers. Typically, methylation levels are measured at individual cytosine/guanine (CpG) sites or over a short region of interest. However, regions of interest often show heterogeneous methylation comprising multiple patterns of methylation (epialleles) on individual DNA strands. Heterogeneous methylation is largely ignored because digital methods are required to deconvolute these usually complex patterns of epialleles. Currently, only single-molecule approaches, such as next generation sequencing (NGS), can provide detailed epiallele information. Because NGS is not yet feasible for routine practice, we developed a single-molecule–like approach, named for epiallele quantification (EpiQ).
METHODS
EpiQ uses DNA ligases and the enhanced thermal instability of short (≤19 bases) mismatched DNA probes for the relative quantification of epialleles. The assay was developed using fluorescent detection on a gel and then adapted for electrochemical detection on a microfabricated device. NGS was used to validate the analytical accuracy of EpiQ.
RESULTS
In this proof of principle study, EpiQ detected with 90%–95% specificity each of the 8 possible epialleles for a 3-CpG cluster at the promoter region of the CDKN2B (p15) tumor suppressor gene. EpiQ successfully profiled heterogeneous methylation patterns in clinically derived samples, and the results were cross-validated with NGS.
CONCLUSIONS
EpiQ is a potential alternative tool for characterizing heterogeneous methylation, thus facilitating its use as a biomarker. EpiQ was developed on a gel-based assay but can also easily be adapted for miniaturized chip-based platforms.
Preferential binding to branched DNA strands and strand-annealing activity of the human Rad51B, Rad51C, Rad51D and Xrcc2 protein complex
