The trace detection of DNA adducts by accelerator mass spectrometry and 14C-postlabelling
Accelerator mass spectrometry (AMS) is currently the most sensitive method for trace DNA adduct detection. O6-Mehtyldeoxyguanosine (O 6-MedG) is a strongly mutagenic lesion formed by a variety of alkylating agents. For this reason this was the adduct of choice for use in this study, the aim of which was to develop a 14C-postlabelling technique, involving incorporation of radiolabel onto O6-MedG adducts after isolation, thus enabling exploitation of AMS to detect low levels of adducts without the need to administer a 14C-labelled compound. A method was developed and optimised for acetylating O6-MedG, in < 90% yields. This method was then used to acetylate the adduct with 14C-acetic anhydride, but changes required for the safe handling of radiolabelled compound altered the reaction product profile, resulting in the major derivative being 14C-di-acetyl O6-MedG, (38% yield). This pure standard was used to determine detection limits of 1.4 pmoles of adduct using HPLC and liquid scintillation counting and 79 attomoles using AMS. A 3H-O6MedG standard was used to determine the efficiency of each stage of the 14C-postlabelling procedure. Overall assay efficiency was calculated to be 29% and taking this into account the practical limit of detection for the 14C-postlabelling procedure was calculated to be 272 attomoles of adduct. The assay was then utilised to quantitate O6-MedG adducts formed by exposure of DNA with MNU. The theoretical limit of detection was 3 adducts/1012 nucleosides, based on background 14C levels in blank HPLC fractions, which indicates the assay should ultimately be the most sensitive for detection of endogenous O6-MedG adducts in control human DNA. The techniques developed for this assay were utilised to quantitate O6-MedG adducts formed in DNA exposed to 14C-S-adenosylmethionine. Unfortunately no adducts were conclusively detected by AMS due to contamination of the samples, highlighting the problem of contamination in AMS analysis. However, with the addition of further purification steps, this assay could be sensitive enough to detect O6-MedG formed by the suspected endogenous carcinogen.