DNA adduction by phenol, hydroquinone, or benzoquinone in vitro but not in vivo: nuclease P1-enhanced 32P-postlabeling of adducts as labeled nucleoside bisphosphates, dinucleotides and nucleoside monophosphates.

The carcinogenicity of benzene has been considered to be in part mediated by its chemically reactive metabolic product benzoquinone (BQ), which is formed from the intermediary metabolites phenol and hydroquinone (HQ). We have evaluated the DNA-binding capability of these chemicals in vitro and in vivo by postlabeling. Treatment of rat Zymbal glands in culture with phenol and HQ or direct reaction of BQ with DNA produced DNA adducts, which were detectable by the nuclease P1-enhanced 32P-postlabeling assay as 5'-32P-labeled 3',5'-bisphosphate products. The enhancement of sensitivity in this assay is based on the previous finding that nuclease P1 hydrolyzes the phosphate attached to the 3' side of normal nucleotides but not the corresponding phosphate of most aromatic/bulky adducted nucleotides. Also based on this hydrolytic property of nuclease P1, we developed an additional sensitive procedure that permitted the detection of DNA lesions as 5'-32P-labeled products of dinucleotides, pXpN, or of nucleoside monophosphates, pX, where X and N indicate an adducted nucleoside and a normal nucleoside respectively. In the latter assay, adducted DNA was first digested with nuclease P1 and acid phosphatase to yield XpN and N. The latter were then 32P-labeled to yield [5'-32P] pXpN or 32P-labeled and treated with venom phosphodiesterase to obtain [5'-32P]pX. After optimization of enzymatic conditions, the modified nuclease P1 assay yielded adduct recoveries similar to those obtained by the bisphosphate assay for in vitro phenol-, HQ- and BQ-DNA adducts. Neither of the nuclease P1-enhanced postlabeling procedures showed exposure-specific adducts in vivo in the bone marrow, Zymbal gland, liver and spleen of female Sprague-Dawley rats at 24 h after the last of four single, daily p.o. doses of 75 mg/kg phenol or 150 mg/kg phenol/HQ (1:1). Our results show that phenol, HQ and BQ produce adducts in vitro, but corresponding adducts are not detected in vivo with phenol and phenol/HQ, even when measured by the standard and modified nuclease P1 postlabeling methods capable of detecting 1 adduct in 10(9-10) DNA bases.