Is high resolution a strict requirement for mass spectrometry-based cellular DNA adductomics?

Abstract

Exposure to endogenous and exogenous factors can result in the formation of a wide variety of DNA adducts, and these may lead to gene mutations, thereby contributing to the development of cancer. DNA adductomics, a novel tool for exposomics, aims to detect the totality of DNA adducts. Liquid chromatography-high resolution mass spectrometry (LC-HRMS) is the state-of-the-art method for DNA adductomic analysis, although its high cost has limited widespread use. In this study, we compared the analytical performance between HRMS and the more popular/accessible triple-quadrupole MS (QqQ-MS). We initially developed and optimized a hybrid quadrupole-linear ion trap-orbitrap MS (Q-LIT-OT-MS) method, considering the detection of both purine and pyrimidine adducts. LC-Q-LIT-OT-MS and LC-QqQ-MS methods were compared by non-targeted screening of formaldehyde-induced DNA adducts. Using the results from Q-LIT-OT-MS as the gold standard, QqQ-MS successfully detected 12 out of 18 formaldehyde-induced DNA adducts/inter-strand crosslinks (ICLs). QqQ-MS however also produced nine false-positive results owing to the inherent instrumental mass resolution limits. To discriminate the false-positive results from the accurate ones, we firstly introduced a statistical analysis, partial least squares-discriminant analysis, which efficiently excluded the false signals. Six DNA adducts/ICLs were not detected by QqQ-MS, due to insufficient sensitivity. This could be overcome by employing a selected reaction monitoring scan mode with multiple injections. Overall, this study demonstrated that high resolution may not be a strict requirement for MS-based DNA adductomics. LC-QqQ-MS with statistical analysis, could also provide a comparable performance as HRMS for pre-screening purposes.