Dose-Response Curves for Liver DNA Fragmentation Induced in Rats by Sixteen 7V-Nitroso Compounds as Measured by Viscometric and Alkaline Elution Analyses1

A new viscometric technique, capable of detecting DNA strand breaks and alkali-labile sites by monitoring time-dependent changes of DNAreduced viscosity, has been used to analyze dose-response curves for the induction of DNA damage in liver of rats treated with single p.o. doses of sixteen A-nit roso compounds. Statistically significant changes of DNA viscometric parameters, which are considered indicative of DNA frag mentation, were produced by A'-nitrosodimethylamine (0.022 mg/kg), ,'Vnitrosomethylethylamine (0.025 mg/kg), V-nitrosodiethylamine (0.067 mg/kg), jV-nitrosodiethanolamine (1.03 mg/kg), Ar-nitrosodi-/i-propylamine (031 mg/kg), Nnitrosodi-n-butylamine (0.083 mg/kg), /V-nitroso-A1methylurea (0.56 mg/kg), A'-nirroso-A'-ethylurca (037 mg/kg), /V-nitrosoA'-butylurca (0.16 mg/kg), streptozotocin (20 mg/kg), A'-nitrosomorpholine (0.4 mg/kg), JV-nitrosopiperidine (2.22 mg/kg), A'-nitrosopyrrolidine (5.0 mg/kg), l-nitroso-2-imidazolidinone (031 mg/kg), and A'-methylA"-nitro-A'-nitrosoguanidine (5.57 mg/kg). The contemporary measurement of liver DNA fragmentation by the alkaline elution technique revealed that in our experimental conditions higher doses are needed to produce a statistically significant increase of DNA elution rate. This suggests that the viscometric method is capable of detecting smaller levels of /V-nitroso compound-induced DNA frag mentation, but it does not exclude that the sensitivity of alkaline elution can be improved by appropriate modifications of the experimental pro cedure. With both techniques DNA damage was undetectable in liver of rats treated with 540 mg/kg of the non-hepatocarcinogen A'-nitrosodiphenylamine. With the exception of A-nitrosodiethanolamine, that ex hibited a plateau effect, all the other A'-nitroso compounds examined displayed a linear dose-response curve over the entire wide range of doses tested. Consequently, a nonlinearity of the relationship between dose and tumor response cannot be attributed to a nonlinearity of the pharmacokinetic processes involved in the formation of DNA damage. INTRODUCTION NOC3 have been recognized as potent carcinogens, active in a large variety of animal species (1-3). Since they are wide spread in the human environment and can be formed endogenously in humans from precursors ingested separately (4, 5), their importance cannot be overrated. However, the relevance of NOC to a possible carcinogenic risk to humans is difficult to establish, although some epidemiolÃ3gica! studies have indi rectly suggested their possible role in the etiology of a number of human cancers (6-9). In addition to interspecies differences, the source of an erroneous risk estimate may be the linear extrapolation to the low doses encountered in human exposure of the results obtained in the experimental animals at doses many orders of magnitude higher. In fact, the relation between Received 10/20/86; revised 2/18/87; accepted 3/23/87. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. 1This work was supported by the Consiglio Nazionale delle Ricerche. Finalized Projects "Oncology" (contract 85.02051.44) and "Preventive and Rehabilitative Medicine" (contract 85.00687.56), and by funds from the Public Instruction Ministry. 1To whom requests for reprints should be addressed, at Istituto di Farmaco logia dell'Università , Viale Benedetto XV, no. 2,1-16132 Genova, Italy. 3The abbreviations used are: NOC, Ai-nitroso compounds; LD», 50% lethal dose; all the other abbreviations are listed in Table I. applied dose and tumor response may be nonlinear (10). This nonlinearity often reflects a nonlinearity of the pharm ucokine ticprocesses, absorption, distribution, metabolism, and elimina tion from which the effective concentration of the ultimate carcinogen at the level of the target organ is dependent (10). There is indeed evidence of a frequent linear relation between the concentration of carcinogen-DNA adducts and tumor inci dence (11-15), and this correlation may be improved when the levels of specific DNA adducts and the rates of their removal are included in the evaluation (16-18). A limitation for a DNAbinding assay in vivo is the need of radioactively labeled chem icals. Recent evidence has shown that a new viscometric tech nique provides a reliable estimate of the amounts of DNA fragmentation produced in the intact rat by very low doses of some chemical carcinogens (19-22). It is reasonable to assume that for a preliminary study of the relationship between admin istered dose and biologically active concentration the measure ment of the amount of DNA fragmentation may represent an alternative approach to the determination of DNA binding. It was therefore decided to evaluate the relationship between dose and induction of DNA fragmentation in the liver of rats treated with a wide spectrum of single p.o. doses of 16 NOC of various chemical structure. This evaluation also was done in order to assess the sensitivity of the above-mentioned viscometric method in detecting DNA damage induced by this family of carcinogens. The frequency of DNA single-strand breaks and/ or alkali-labile sites was measured either by the alkaline elution technique (23) or by the viscometric method (21). In our exper imental conditions alkaline elution should allow the detection of one DNA lesion/5 x IO8 to IO9 daltons and the viscometric method of approximately 0.3 lesions/1010 daltons (19). MATERIALS AND METHODS Chemicals. The chemical name, abbreviation, molecular weight, LDM for single p.o. administration, source, and chemical purity of the NOC investigated are reported in Table 1. Tc true*thy (ammonium hydroxide was purchased from E. Merck, Darmstadt, Federal Republic of Ger many. All other chemicals were reagent grade. Treatment of Animals. Random-bred male albino Sprague-Dawley rats (160 to 180 g) fasted for 12 h were used. NBU, NDPHA, NIMIO, and MNNG were suspended in distilled water with 1% carboxymethyl cellulose as suspending agent, and the other NOC were dissolved in distilled water immediately before use. A single dose of each NOC was administered by gavage in 0.01 ml of vehicle/g of body weight. Controls were given by gavage the same amount of the same vehicle. Rats were sacrificed 3 h after treatment. Viscometric Analysis of DNA Damage. A detailed description of the theory and technique of DNA viscometric analysis has already been published (19-21). Briefly, the viscometer, used at constant temperature (22 ±0.1'C) consists of a perspex crucible oscillating around a nickel/ chromium wire from which it is suspended and containing the DNA solution under study in a half-filled U-shaped concentrica! channel dug into the disc. The parameter measured is the harmonic damped oscil lation of the crucible, which depends on the frictional torque exerted by the liquid on the internal walls of the circular channel. In order to 3485 on July 20, 2017. © 1987 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from LIVER DNA DAMAGE BY W-NITROSO COMPOUNDS Table 1 N-Nitroso compounds tested NameAT-nitrosodimethylamineiV-nitrosomethylethylamine.V nilniMiiliclhylanlincjV-nitrosodiethanolamine/V-nitrosodi-n-propylamine,Y-nitrosodi-n-butylamineA'-nitroso-A'-methylureaJV-nitroso-yV-ethylurea/V-nitroso-A'-butylureaStreplozotocin/V-nitrosomorpholine.V nitre isc ipipcriili H-W-nitrosopyrrolidine1 -Nitroso-2-imidazolidinoneJV-Methyl-N'-nitroWV-nitrosoguanidineiV-nitrosodiphenylamineAbbreviationNDMANMEANDEANDELANDPANDBANMUNEUNBUSTRNMORNPIPNPYRNIMIDMNNGNDPHAMolecular wt74.188.1102.1134.1130.2158.2103.1117.1145.2265.2116.1114.2100.2115.1147.1198.2LDM(mg/kg)°2690280>75004801200110300400Not