Rat liver subcellular fractions catalyze aerobic binding of 1-nitro[14C]pyrene to DNA.

The recently characterized environmental mutagen and potential carcinogen 1-nitropyrene (NP) is known to bind DNA in Salmonella typhimurium, and also in anaerobic incubations catalyzed by purified xanthine oxidase. In this study we show that rat liver S9 supernatant, microsomal and cytosolic subcellular fractions are also able to catalyze the binding of 1-nitropyrene labeled with 14C to calf thymus DNA in vitro. In incubations conducted under air, S9 and microsomes from Charles River CD rats were the most active fractions, and NADPH was required for maximum activity (25-100 pmole NP bound/mg DNA/mg protein in 1 hr). S9 and microsomes had about one-fourth the activity under nitrogen, although less of this activity was NADPH-dependent. Binding in cytosolic incubations was generally low (1 to 5 pmole NP/mg DNA/mg protein in 1 hr), was somewhat enhanced under N2, and was more extensive in the absence of NADPH. Treatment of rats (Harlan Sprague-Dawley) with the inducing agents phenobarbital (PB), Aroclor 1254 (A), or 3-methylcholanthrene (3-MC) enhanced NADPH-dependent binding in aerobic S9 (2- to 5-fold) and microsomal (10- to 20-fold) incubations. The effects of induction regimen on binding assays conducted under N2 were more equivocal: 3-MC produced a 2-fold increase in binding in both S9 and microsomes, while the other two agents decreased binding from 50 to 75%. These results indicate that classic cytochrome P-450 inducers were able to stimulate activation of NP, but that this activation is not mediated solely by cytochrome P-450.


Introduction
1-Nitropyrene (NP), found in photocopier toners (1,2) and in particulate matter from diesel vehicle emissions (3) and ambient air (4), is a mutagen in bacterial and mammalian assay systems (5)(6)(7). NP has also been reported to cause mammary gland tumors in rats (8). Covalently bound adducts formed by interaction of NP with DNA have been detected (9) and characterized in Salmonella typhimurium (10) and in cell-free incubations catalyzed under anaerobic conditions by purified xanthine oxidase (10). The principal adduct formed under these conditions was identified as N-(deoxyguanosin-8yl)-1-aminopyrene, thought to be generated from a hydroxylamine intermediate during the process of reductive metabolism of NP (10). On the other hand, Salmeen et al. (11) observed that although the mutagenicity of NP differed widely between nitroreductase-deficient and -proficient Salmonella strains, and V79 Chinese hamsters and primary rat hepatocytes, no corresponding variations were seen in the ability of all these cell types * to reduce NP to 1-aminopyrene, and therefore nitroreduction might not be the only determinant of mutagenicity. NP was also shown to bind to endogenous DNA and protein when incubated with rabbit pulmonary tissues under aerobic conditions such that metabolism could occur by oxidative pathways (12).
The liver is the major site of oxidative metabolism in mammals and has been shown to convert NP to oxidized products which are themselves mutagenic (13,14). Although mutagenicity and DNA binding are not necessarily directly causally linked, both phenomena arise from the type of interactions with cellular genetic material which may ultimately lead to genotoxic damage to the organism. We have previously attempted to elucidate the role of oxidative metabolism in the activation of NP as measured by mutagenicity in the Salmonella plate incorporation assay (14,15).
The present study was undertaken to evaluate the ability of hepatic enzyme systems to catalyze the formation of NP metabolites capable of interacting with DNA by forming covalent adducts, and if so, whether these reactions could be linked to oxidative rather than to reductive metabolism. We also were interested to see whether the terminal enzyme of the mixed-function oxidase system, cytochrome P-450, known to be particularly active in the metabolism of foreign compounds, appears to be involved in this activation process. Chemicals 1-Nitro [4,5,9,10-14C]pyrene (10 mCi/mmole, :99.9% radiochemical and chemical purity) was obtained from the Midwest Research Institute (Kansas City, MO) (14,15). NADPH and calf thymus DNA (Type I) were purchased from Sigma Chemical Co (St. Louis, MO). Laboratory reagents and solvents were obtained at the highest degree of purity from commercial suppliers and were used as supplied.

Animals
Livers of adult male Charles River CD rats (150-180 g) were excised and homogenized in sterile 1.15% KCI-0.02 M HEPES, pH 7.4 (4 vol). S9 supernatant, cytosol and microsomes were prepared by differential centrifugation. The microsomal pellet was resuspended in 0.1 M HEPES buffer, pH 7.4. Groups of five male Sprague-Dawley rats (body weight 100-200 g; Harlan Sprague-Dawley, In., Madison, WI) were treated by intraperitoneal injection with the following cytochrome P-450 inducers: phenobarbital (Mallinkrodt Chemical Co., St. Louis, MO; 20 mg in 0.5 mL of 0.85% NaCl daily for 4 days), Aroclor 1254 (Monsanto, St. Louis, MO) 120 mg in 0.5 mL corn oil 4 days before sacrifice), or 3-methylcholanthrene (Eastman Organic Chemicals, Rochester, NY; 1 mg in 0.2 mL corn oil four days before sacrifice (16). A further group of five rats served as untreated controls. Rats were starved for 20 hr prior to sacrifice; hepatic microsomes and S9 fraction were prepared as described (17). Protein concentration was assayed (18) with bovine serum albumin (Sigma Chemical Co., St. Louis, MO) as the protein standard. Cytochrome P-450 content was determined by the dithionite difference method (19). All incubations were carried out on freshly prepared tissues. Incubations S9, cytosol or microsomes (1-1.4 mg protein) were suspended in 0.95 mL of 0.1 M HEPES buffer, pH 7.4, containing 2 mg calf thymus DNA. The incubation mixture was warmed to 37°C for 3 min, then 100 ,uM 14C-NP was added in 10 ,uL of DMSO. The reaction was started by addition of NADPH (1 mg in 50 puL of 0.1 M HEPES buffer, pH 7.4), and terminated after 60 min by addition of an equal volume of CHCl3/isoamyl alcohol/ phenol (CIP; 24:1:25 by volume, saturated with HEPES buffer) and vortexed. Where incubations were conducted under nitrogen, the incubation mix was purged with N2 as it was equilibrating to 37°C, purged again after addition of 14C-NP and NADPH, then sealed for the duration of the incubation.

Determination of Radioactivity Bound to DNA
The protein was removed by CIP extraction (3 x 1 volume), and the DNA precipitated from the aqueous phase by addition of NaCI and ice-cold EtOH. The DNA was washed twice with EtOH, then harvested by winding onto glass pipettes, dried, and redissolved in 0.1 x standard saline/citrate. The '4C content of the DNA solution was determined by liquid scintillation counting (Model 2660, Packard Instrument Co., Downers Grove, IL). The UV spectrum of the DNA solution was scanned (240-320 nm; DW-2, American Instrument Co., Silver Spring, MD); the DNA content was assayed by measurement of the OD260 and the purity of the DNA was verified by examination of the OD260/0D280 ratio.

Results
After incubation of 14C-NP with rat liver subcellular fractions and calf thymus DNA, radioactivity was detected in the isolated DNA fraction. DNA isolated from similar incubations conducted in the absence of tissue protein contained little or no 14C detectable above background, which corresponded to limits of detectability of 0.1 to 0.2 pmole NP bound/mg DNA (Table 1). In incubations where tissues were derived from untreated Charles River CD rats, the highest rates of binding were seen in microsomes in the presence of air and NADPH (Table 1). This reaction was verified to be linear with respect to time up to 60 min and with respect to protein up to 2 mg of protein (results not shown). Although NADPH was required for maximum activity in S9 and microsomes, measurable rates of binding were also found in the absence of this co-factor. Both S9 and microsomes were about one-fourth as active under nitrogen as under air, although the activity under nitro- gen was somewhat less dependent on NADPH. The activity of the cytosol was low under air, and slightly higher under nitrogen. When the cytosol was incubated with '4C-NP and NADPH, but without DNA, under the conditions which had resulted in uniformly low DNA binding, and the incubates analyzed by HPLC for NP metabolites as described previously (14), this tissue was readily able to reduce NP to 1-aminopyrene (results not shown). Treatment of the animals used as sources of hepatic tissues with inducing agents duly increased the P-450 content in the S9 and microsomal fractions (Table 2), and also the extent of DNA binding catalyzed by these fractions, although the increase in binding was not proportional to the enhancement in P-450 content. Aroclor 1254 produced the highest increase in cytochrome P-450 content in both S9 and microsomes and the highest specific binding in S9 under air, whereas phenobarbital and 2-methylcholanthrene were more effective than Aroclor in enhancing binding under nitrogen in S9 and in microsomes under both air and nitrogen. Although in this study 3-methylcholanthrene was by far the least potent inducer of cytochrome P-450, our phenobarbital and 3-methylcholanthrene treatments were almost equivalent in their ability to induce binding of 14C-NP to DNA.
Differences in activity between strains were also apparent, since the microsoMal fraction from the Harlan Sprague-Dawley rats bound 14C-NP under air at very low rates, in contrast to the more extensive binding seen in the microsomes from Charles River CD rats. These low figures for aerobic untreated microsomal binding resulted in consequently high induction ratios for this parameter (Table 2).

Discussion
DNA binding of 14C-NP in hepatic tissues was largely (though not entirely) dependent on air and NADPH. This binding was enhanced by cytochrome P-450 induc-ers: 3-methylcholanthrene and phenobarbital proved more effective than Aroclor 1254 in inducing binding in microsomes, whereas Aroclor was more effective in S9. This pattern was seen under both air and nitrogen. In contrast, much more enhancement in cytochrome P-450 content was seen here following Aroclor 1254 and phenobarbital treatment, indicating that though 3-methylcholanthrene was the least effective inducer of total P-450 here, it was the agent most effective at inducing binding of NP to DNA. The correlation between percentage increase in binding and in cytochrome P-450 content was also closer in the incubations catalyzed by tissues from 3-methylcholanthrene-treated animals than in those from the other two agents. Binding in the previously studied systems (9,10) is thought to be due to the reductive intermediate N-hydroxy-1-aminopyrene. In this study we have shown that incubations conducted under conditions most conducive to NP reduction resulted in only low levels of binding, thus confirming the report that in mammalian cells the mutagenicity of NP does not necessarily correlate with the reduction of NP (11). In rabbit liver microsomes NP can be reduced by both cytochrome P-450 and aldehyde oxidase, and NADPH was reported to be the most effective electron donor (20). Mammalian microsomal nitroreductase activity appears to occur via the oxygensensitive one-electron pathway, which generates a nitro anion free-radical intermediate, whereas some strains of bacteria can metabolize nitro compounds by both the one-electron pathway and the two-electron pathway, which does not form the free radical (21). The latter pathway may therefore be more efficient in generating hydroxylamine intermediates capable of interacting with macromolecules, whereas the former can produce cytotoxicity by means of the superoxide arising from reoxidation of the nitro free radical.
Our findings that maximal levels of NADPH-dependent binding of NP to DNA occur under air suggest that the active intermediates responsible for binding are being generated either by oxidative routes, or by some previously unsuspected oxygen-insensitive mammalian nitroreductase. The enzyme(s) involved would also appear to respond to inducing agents (especially 3-methylcholanthrene) in a manner similar to cytochrome P-450. The effects of mixed-function oxidase inducers are not limited to cytochrome P-450 (22,23), so that this response is not diagnostic for P-450; moreover, the specific binding activity does not increase consistently on going from S9 to microsomes as would be the case for reactions due entirely to cytochrome P-450. The overall pattern of activity indicates that both soluble and microsomal enzymes contribute to the overall process, but that the microsomal components are capable of functioning independently of the cytosolic portions, while the converse does not hold. In contrast, activation of NP in the Ames assay was clearly shown to reside in the microsomal fraction (24). Further studies are in progress with specific inhibitors and co-factors in order to clarify the pathways responsible, singly or in conjunction, for catalysis of the in vitro binding of NP to DNA.
L.M.B. is grateful to the National Research Council, Washington, DC, for a Resident Research Associateship Award. We thank R.
Easterling, Carcinogenesis and Metabolism Branch, Genetic Toxicology Division, for preparing tissue fractions from Harlan Sprague-Dawley rats.
This report has been reviewed by the Health Effects Research Laboratory, US Environmental Protection Agency, and approved for publication. Approval does not signify that the contents necessarily reflect the views and policies of the Agency nor does mention of trade names or commercial products constitute endorsement or recommendation for use.