Abstract
The N(2) position of guanine is a preferential reaction site in DNA for numerous dietary and environmental carcinogens or their electrophilic metabolites, aldehydes arising from lipid peroxidation as well as reactive by-products of normal metabolism. However, DNA repair mechanisms of the resulting covalent adducts in mammalian cells are not well understood, with nucleotide excision repair (NER), base excision repair, and a dioxygenase-mediated damage reversal being discussed as likely pathways. Considering fundamentally different damage recognition principles between the global genome NER and the transcription-coupled (TC)-NER, we here assessed transcription blocking capacities of four synthetic deoxyguanosine (dGuo) adducts of variable size and geometry, using a transfection-based reporter assay. Notably, adducts as different as the aliphatic N(2)-ethylguanine, the exocyclic 1,N(2)-ethenoguanine, and the bulky polycyclic 3-(deoxyguanosin-N(2)-yl)-2-acetylaminofluorene, displayed robust DNA strand-specific transcription-blocking properties. The specific TC-NER components ERCC8/CSA and ERCC6/CSB were consistently required for the removal of all transcription-blocking N(2)-dGuo adducts, whereas the absence of XPC or DDB2/XPE (both specific to global genome NER) did not compromise the repair capacities in the isogenic human cell models. In contrast, no inhibition of the gene expression was detected for reporter constructs carrying N(2)-methylguanine even in the NER-deficient XP-A cell line, suggesting that this adduct is either bypassed with very high efficiency during transcription or repaired by a mechanism different from NER. Collectively, the results identify N(2)-dGuo adducts bigger than methylguanine as a structural subclass of transcription-blocking DNA lesions whose repair heavily relies on the TC-NER pathway.