Comparison of different methods of measuring 8-oxoguanine as a marker of oxidative DNA damage

We are attempting to resolve some of the problems encountered in measuring 8-hydroxy-2'-deoxyguanosine (8-oxodG) in human cellular DNA as a marker of oxidative stress. Samples of authentic 8-oxodG were distributed, and participating laboratories undertook to analyse this material within a specified period. Most HPLC procedures gave values for 8-oxodG within ±40% of the target, as did two of four GC-MS procedures, and both LC-MS-MS methods. Calf thymus DNA samples containing increasing amounts of 8-oxodG were also distributed for analysis. Fewer than half the procedures tested were able to detect the dose response; those that were successful tended to be procedures with low coefficients of variation. For the analysis of 8-oxodG in human cells, where it is likely to be present at much lower concentrations than in the calf thymus DNA, it is crucial to reduce analytical variation to a minimum; a coefficient of variation of less than 10% should be the aim, to give reasonable precision. HPLC with amperometric electrochemical detection is not recommended, as it is less sensitive than coulometric detection. Immunological detection, ³²P-postlabelling and LC-MS-MS are alternative approaches to measurement of 8-oxodG in DNA that, on the grounds of precision and detection of dose response, cannot at present be recommended.