Global DNA methylation measured by liquid chromatography-tandem mass spectrometry: analytical technique, reference values and determinants in healthy subjects

BACKGROUND: Alterations in global DNA methylation are implicated in various pathobiological processes. We describe a liquid chromatography-electrospray ionization-tandem mass spectrometry (LC-ESI-MS/MS) method for determination of cytosine and 5-methylcytosine in DNA. METHODS: DNA was hydrolyzed using formic acid. Cytosine and 5-methylcytosine were separated by gradient-elution reversed-phase chromatography with a mobile phase containing nonafluoropentanoic acid (NFPA) as ion-pairing reagent and quantified using stable isotope dilution LC-ESI-MS/MS. The method was applied to DNA isolated from leukocytes of healthy volunteers. RESULTS: Linear calibration curves were obtained in the range 0.111-4.422 ng/microL [mean correlation co-efficient 0.9983 (SD=0.0011), n=9] for cytosine and 0.0048-0.1936 ng/microL [mean correlation coefficient 0.9991 (SD=0.0010), n=9] for 5-methylcytosine. The intra- and inter-assay CVs for the 5-methylcytosine/total cytosine ratio (mCyt/tCyt) was 1.7% (n=9) and 3.5% (n=8) for calf thymus DNA (mean mCyt/tCyt ratio 6.5%), and 4.5% (n=6) and 6.5% (n=14), respectively for pBR322 DNA (mean mCyt/tCyt ratio 0.48%). The limit of detection (signal-to-noise ratio 3) was 2 pg on-column for cytosine and 5-methylcytosine. In healthy subjects (n=109), the mCyt/tCyt ratio varied from 2.6% to 4.8% (median 4.1%). DNA methylation was negatively correlated to age, but only in subjects with the methylenetetrahydrofolate reductase (MTHFR) 677 TT genotype (p=0.046). No association with B-vitamin status was observed. CONCLUSIONS: This LC-ESI-MS/MS method is easy to perform and offers reproducibility, selectivity and sensitivity for studying DNA methylation. The method allows a sample throughput of approximately 200 samples/week. The MTHFR C677T genotype influences age-related changes in DNA methylation.