DNA phosphate oxygens are sites for alkylation leading to DNA phosphotriester adduct (PTE) formation. Previously, we have reported that the manifestation of PTEs was nonrandom in mouse liver DNA treated in vivo [Guichard et at. (2000) Cancer Res. 60, 1276-1282], and while further studies revealed possible PTE repair, this was determined not to play a role in the observed nonrandom manifestation in vivo [Le Pla et al. (2004) Chem. Res. Toxicol. 17, 1491-1500]. In the present study, to determine whether the nonrandom manifestation of PTEs in vivo was specifically due to their nonrandom formation, we have compared the in vitro formation of diethylsulfate (DES)-induced PTEs in h2E1/OR human B-lymphoblastoid cells, their isolated nuclei, and their isolated DNA, using the 5' nearest neighbor analysis postlabeling procedure developed by Le Pla et al.. Furthermore, to determine the role of electrophile character in PTE manifestation, prepared oligonucleotides ([dT](20)[dG](20):[dC](20)[dA](20)) were treated with three alkylating agents of differing electrophilic character (DES, methylnitrosourea, and ethylnitrosourea), and PTE manifestation was assessed by postlabeling. The formation of PTEs was determined to be nonrandom in the whole cells, nuclei, and DNA, with PTEs being formed to a greater extent 3' to pyrimidine moieties than 3' to purine moieties. The studies with the oligonucleotides confirm these observations and demonstrate that the nonrandom formation of PTEs is primarily determined by DNA sequence, and not by DNA packaging/chromatin factors, and that the extent of the nonrandom formation of PTEs is also governed by electrophile reactivity, with the more reactive electrophiles yielding a more random formation of PTEs. From our observations, we propose a model for the nonrandom formation of PTEs, which is governed by (i) the phosphate oxygens having to compete with adjacent nucleophilic sites for the alkylating electrophile and (ii) the electrophile's inherent reactivity.
