THE DEVELOPMENT, VALIDATION AND APPLICATION OF A P-32 POSTLABELING ASSAY TO QUANTIFY O-6-METHYLGUANINE IN HUMAN DNA

In this study, a combined immunoaffinity purification/ P-32-postlabelling procedure has been used to quantify O-6-methyldeoxyguanosine-3'-monophosphate (O-6-MedGp) in human DNA. DNA digests are subjected to a two-stage immunopurification in which the acetone-eluted fraction from the first stage is reapplied to a second immuno-column, and the O-6-MedGp specifically eluted using O-6-methylguanosine (O-6-MerG). O-6-MedGp is then P-32-postlabelled in the presence of deoxyinosine-3'-monophosphate (dIp) as internal standard, separated by two-dimensional TLC and levels of the adduct quantified using storage phosphor technology. The recovery of O-6-MedGp at levels between 0.4 and 500 fmol was 61%. Analysis of human DNA samples indicated that <1 fmol O-6-methyldeoxyguanosine-5'-monophosphate (O-6-MepdG) could be detected with a high degree of precision (coefficient of variation <12%) during a 2 h exposure to a storage phosphor screen. The assay was then applied to 25 human samples from three separate populations, one of which was exposed to methylating agent chemotherapy, for which O-6-methyldeoxyguanosine (O-6-MedG) levels had already been quantified by HPLC/radioimmunoassay. The results indicated a high degree of correlation between the two assays (r = 0.99). O-6-MedGp was detected in all the samples analysed with levels ranging from 0.026 to 23.2 mu mol O-6-MedGp/mol dG. The minimum amount of O-6-MepdG detected was 0.2 fmol. As there was no detectable signal in the area to which O-6-MepdG maps in negative control samples, a detection limit based upon the signal/noise ratio was impossible to quantify, However the limit of detection of the storage phosphor technology itself was estimated by quantifying a visually identifiable compound, which mapped to the same region. The amount of this compound was determined to be 32 +/- 27 amol (n = 5). If a similar amount of O-6-MepdG was detected from 50 mu g of DNA, and assuming that the labelling efficiency and recovery was similar to that found in this study, then this would correspond to an adduct level of similar to 3 nmol O-6-MedGp/mol dG.