Appearance of artefacts when using 32P-postlabelling to investigate DNA adduct formation by bile acids in vitro: lack of evidence for covalent binding.

Abstract

We reported (Scates et al. Carcinogenesis 1994, 15, 2945-2948) that incubating a range of bile acids with DNA in vitro, with or without exogenous metabolic activation, gave no evidence of DNA adduct formation as judged by the nuclease P1 method of 32P-postlabelling. In contrast Hamada et al. (Carcinogenesis 1994, 15, 1911-1915), also using postlabelling, claimed that chenodeoxycholic acid, lithocholic acid, glycolithocholic acid and taurolithocholic acid bound covalently to DNA in vitro. To investigate this discordance we incubated solutions of salmon sperm DNA for 1 h at 37 degrees C with 1 mg/ml of cholic acid, chenodeoxycholic acid, lithocholic acid, glycolithocholic acid or taurolithocholic acid. Each incubate was extracted extensively with diethyl ether after which a sample of DNA was taken and 32P-postlabelled using the nuclease P1 method. The DNA in the remaining incubate was precipitated from high salt solution with ethanol. Aliquots of this DNA were postlabelled. The remainder of the DNA was purified with proteinase-K, ribonuclease, phenol-chloroform, precipitated and postlabelled. Parallel incubates were made with the same bile acids, under the same conditions but in the absence of DNA and were then extracted, precipitated and postlabelled as described above. When DNA was present in the incubate but was not precipitated, chenodeoxycholic acid, lithocholic acid, glycolithocholic acid and taurolithocholic acid, but not cholic acid, produced spots similar to those reported by Hamada et al. No such spots were seen when DNA was postlabelled after precipitation, or after precipitation and purification. These same bile acids produced spots when postlabelled in the absence of DNA, but spots were absent when these incubates were precipitated and purified before postlabelling. We conclude that the spots obtained when bile acids are incubated with DNA which is not precipitated from high salt before it is postlabelled are technical artefacts, and cannot be regarded as evidence that bile acids bind covalently to DNA to form adducts. We also confirm reports (Vulimiri et al. Carcinogenesis 1994, 15, 2061-2064) that bile acids alone can produce spots when incubated with T4 polynucleotide kinase and [gamma-32P]ATP.