Fluoranthene-DNA adducts: identification and quantification by an HPLC-32P-postlabeling method.

A 32P-postlabeling procedure for identifying and quantifying fluoranthene (FA)-DNA adducts has been developed through modifications of the method of Randerath and collaborators. In this modified procedure, labeled adducts are separated chromatographically by high-pressure liquid chromatography (HPLC) and quantified by liquid scintillation counting. FA-modified DNA is digested to nucleotide 3'-monophosphates and nucleotide 3'-monophosphate adducts; unmodified nucleotides are then separated from adducts using a disposable C18 cartridge. Residual unmodified nucleotides, which reduce the efficiency of 32P-postlabeling of FA adducts, are removed by brief digestion with nuclease P1. This treatment selectively dephosphorylates unmodified nucleotide 3'-monophosphates, while FA adducts are minimally affected. FA adducts are then 5'-phosphorylated with polynucleotide kinase and [gamma-32P]ATP. Prolonged treatment with nuclease P1 then is employed to remove the unlabeled 3'-phosphate from adducted diphosphate nucleotides, following which adducts are separated by HPLC and quantified by liquid scintillation counting. Postlabeled microsomally-activated FA-modified DNA contained adducts derived from anti- and syn-2,3-dihydroxy-1,10b-epoxy-1,2,3-trihydrofluoranthene. The identity of the major adduct in DNA-bound microsomally-activated FA was confirmed by this HPLC-32P-postlabeling method as an anti-2,3-dihydroxy-1,10b-epoxy-fluoranthene nucleotide adduct. Each step in the procedure was optimized with respect to experimental conditions, and the recovery of adducts was determined by analysis of DNA modified with [3H]FA. In repeated analyses of 2-50 micrograms DNA containing 1.8 adducts per 10(8) nucleotides, 10-15% of total DNA-bound [3H]FA was recovered as the major adduct; recovery was greater from DNA containing higher levels of adducts. The reproducibility of multiple analyses of the same sample was approximately 5%, and multiple analyses at different times were reproducible within experimental error. The limit of detection of the method was approximately 0.1 fmol adduct, representing a binding level of approximately 3 adducts per 10(8) nucleotides in 1 microgram DNA or approximately 1 adduct per 10(10) nucleotides in 500 micrograms DNA. Because the method is not limited with respect to the amount of DNA that can be subjected to analysis, the inherent sensitivity for adduct detection can be greatly enhanced by analysis of larger quantities of DNA.