Investigation of the potential for binding of di(2-ethylhexyl) phthalate (DEHP) and di(2-ethylhexyl) adipate (DEHA) to liver DNA in vivo

Abstract

It was the aim of this investigation to determine whether covalent binding of di(2-ethylhexyl) phthalate (DEHP) to rat liver DNA and of di(2-ethylhexyl) adipate (DEHA) to mouse liver DNA could be a mechanism of action contributing to the observed induction of liver tumors after lifetime feeding of the respective rodent species with high doses of DEHP and DEHA. For this purpose, DEHP and DEHA radiolabeled in different parts of the molecule were administered orally to female rats and mice, respectively, with or without pretreatment for 4 weeks with 1% unlabeled compound in the diet. Liver DNA was isolated after 16 hr and analyzed for radioactivity. The data were converted to a covalent binding index, CBI = (micromoles of substance bound per mole of DNA nucleotides)/(millimoles of substance applied per kilogram body weight), in order to allow a quantitative comparison also with other carcinogens and noncarcinogens. Administration of [ 14C]carboxylate-labeled DEHP to rats resulted in no measurable DNA radioactivity. The limit of detection, CBI < 0.02 was about 100 times below the CBI of compounds where an observable tumor-inducing potential could be due to genotoxicity. With [ 14C]- and [ 3H]DEHP labeled in the alcohol moiety, radioactivity was clearly measurable in rat liver DNA. HPLC analysis of enzyme-degraded or acid-hydrolyzed DNA revealed that the natural nucleosides or purine bases were radiolabeled whereas no radioactivity was detectable in those fractions where the carcinogen-modified nucleoside or base adducts are expected. The respective limits of detection were at 0.07 and 0.04 CBI units for the 14C and 3H labels, respectively. The experiments with [ 14C]- and [ 3H]DEHA, labeled in the alcohol moiety and administered to mice, revealed a minute radioactivity of <50 dpm/mg liver DNA, too little to allow a nucleoside analysis to determine that fraction of the radioactivity which had been incorporated via biosynthesis. Expressed in the CBI units, values of 0.05 to 0.15 for 14C and 0.01 to 0.12 for 3H resulted. Determination of the level of 14CO 2 expiration revealed a linear correlation with the specific activity of DNA. Experiments with 2-ethyl[1- 14C]hexanol performed with both rats and mice allowed the conclusion that most if not all DEHA radioactivity in mouse liver DNA was due to biosynthetic incorporation. A maximum possible true DNA binding by DEHA must be below CBI 0.01. Pretreatment of the animals with unlabeled compound had no effect on the DNA radioactivities in either species. The present negative data, in conjunction with other negative short-term tests for mutagenicity, strongly indicate that covalent interaction with DNA is highly unlikely to be the mode of tumorigenic action of DEHP and DEHA in rodents.