Detection and comparison of DNA adducts after in vitro and in vivo diesel emission exposures.

Development of methods to evaluate certain classes of polycyclic aromatic compounds (PAC) detected in complex mixtures to which humans are exposed would greatly improve the diagnostic potential of 32P-postlabeling analysis. Identification of DNA adduct patterns or specific exposure-related marker adducts would strengthen associations between observed DNA adducts and exposures to different environmental pollutants (e.g., kerosene, cigarette smoke, coke oven, and diesel). We have compared diesel-modified DNA adduct patterns in various in vitro and in vivo rodent model systems and compared them to DNA reactive oxidative and reductive metabolites of 1-nitropyrene. The formation of nitrated polycyclic aromatic hydrocarbon (nitrated PAH) DNA adducts, derived from the metabolism of diesel extract constituents, was enhanced relative to other PAH-derived DNA adducts via xanthine oxidase-catalyzed nitroreduction. These adducts were detectable only by the butanol extraction version of the postlabeling analysis. Five major DNA adducts were detected in human lymphocytes treated in vitro with diesel extract. A major adduct detected in human lymphocytes treated in vitro with diesel extract comigrated with a major adduct detected in lymphocyte DNA treated with benzo[a]pyrene (BaP) alone. Other adducts that co-migrated with the major BaP-derived adducts were detected in skin and lung DNA isolated from rodents topically treated with (50 mg) diesel extract and the major adduct detected in calf thymus DNA treated with rat liver S9 and diesel particle extract. Postlabeling of lung DNA isolated from rodents exposed via lung inhalation for 24 months to diesel combustion emissions resulted in the formation of a major nuclease-P1-sensitive DNA adduct that did not co-migrate with the major BaP-diol epoxide adduct.(ABSTRACT TRUNCATED AT 250 WORDS) ImagesFIGURE 1. aFIGURE 1. bFIGURE 1. cFIGURE 2.FIGURE 3.


Introduction
Nitrated polycyclic aromatic hydrocarbons (nitrated PAH) are emitted from several combustion emission processes (e.g., diesel). They are also formed in ambient air through atmospheric reaction (1). Thus, there is interest in studying the role nitrated PAH play in human cancer. Nitrated PAH are present in conjunction with other PAH, therefore it is difficult to ascertain the importance of nitrated-PAH relative to other PAH in the cancer process. Refinement of DNA adduct methods (e.g., 32P-postlabeling analysis) to provide information about the identity of nitrated PAH DNA reactive intermediates relative to other PAH intermediates would substantially improve the diagnostic potential of 32P-postlabeling methods. In this paper, we describe the diesel-derived DNA adducts resulting from several in vitro and in vivo model systems. The detection of specific adducts that are unique to diesel exposure will be useful when analyzing DNA adducts detected in human tissues and cells exposed to complex emission sources containing nitrated PAH. Diesel combustion emissions or particle extracts ofthese emissions are used as a model because these emissions contain relatively high concentrations of nitrated PAH compared to other PAH (e.g., 1-nitropyrene; 1-NP). Calf thymus (CT) DNA adduct patterns observed after reductive or combinations of reductive and oxidative metabolism of 1-NP are compared to DNA adduct patterns resulting from modifying CT DNA and human lymphoctyes in vitro with diesel particle extracts. DNA adduct patterns resulting from topical application ofdiesel extract or inhalation of diesel combustion emissions are compared.
We also present methods that enhance the formation ofnitrated PAH-derived adducts from the array of other PAH reactive metabolites through xanthine oxidase-mediated nitroreduction of the nitrated PAH constituents in the diesel organic extracts. Based on the marked differences in sensitivity for the detection ofN-substituted arylamine adducts for the two most commonly used versions of the 32P-postlabeling methodology (2,3), we have tentatively identified a number ofN-substituted arylamine adducts.

Materials and Methods
In Vitro Assays Isolation of Reductive/Oxidative Metabolites of 1-NP. 1-Nitropyrene (20 ,uM) was incubated for 1.5 hr at 37°C in the presence of xanthine oxidase (XO) (0.5 units/mL) and hypoxanthine 0.6 mg/mL in an anaerobic chamber. To retain the reductive metabolites generated by the xanthine oxidase-catalyzed reduction of 1-NP, the incubate was passed through two primed (10 mL of MeOH and 10 mL water) C-18 Sep-Pak cartridges. The metabolites were eluted from the column using 10 mL of HPLCgrade methanol. The extract containing the metabolites was evaporated to dryness under a stream of nitrogen. The dried extract was redissolved in a 16 mM phosphate buffer (pH 7.41) and incubated in a shaking water bath under aerobic conditions for 1.5 hr at 370°C in the presence of 1 mL of a 2-mg/mL CT DNA solution and rat liver S9 (0.5 mg protein/mL). The rat liver S9 was prepared as previously reported (4). The reaction was terminated by quick freezing the reaction mixture in liquid nitrogen. Samples were stored at -70°C before DNA extraction. Diesel Modified-CalfThymus DNA. Preparation of the diesel particle extracts used in the study have been previously described (5). To obtain oxidative metabolites, CT DNA (1 mg of a 2mg/mL DNA solution) was modified with diesel combustion particle extract (100 ug/mL), and rat liver S9 (0.5 mg/mL) under aerobic conditions for 1.5 hr at 37 'C. Reductive metabolites were generated by adding XO (0.5 units/mL) and hypoxanthine (0.6 mg/mL) under anaerobic conditions for 1.5 hr at 37°C in the presence of 1 mg of CT DNA.

In Vivo Treatments
Skin Painting. Female C-57 mice were topically treated with 50 mg diesel extract as previously described (6). Twenty-four hours after the last exposure, the mice were sacrificed and skin removed and stored at -70°C until subsequent DNA isolation.
Inhalation. Cdl(WI)Br rats were exposed to diesel combustion emissions containing 7.5 mg/m3 soot for a period of24 months.
DNA Extraction and 32P-Postlabeling Analysis. DNA was isolated from CT DNA and rodent tissue essentially as described by Gupta (7). Human lymphocyte DNA was isolated as described by Maniatis et al. (8). Twelve micrograms of DNA was digested to mononucleotides at 37°C for 3.5 hr with micrococcal endonuclease and spleen phosphodiesterase (6). Selected samples were divided into two aliquots and DNA adducts were enriched from one aliquot by the butanol extraction version of the 32ppostlabeling assay (9) and the other by the nuclease P1 treatment essentially as previously described (10). Intensifying screen-enhanced autoradiography at -70°C was used to detect the presence of radiolabeled adducts on the thin layer chromatography plates. Exposure times for autoradiograms are shown in the figure legends.

Results and Discussion
It is often difficult to correlate DNA adduct levels detected in human tissue and cells with occupational or environmental exposure due to the differences in interindividual responses to various carcinogens. Moreover, the exposures under investigation are typically confounded by unrelated exposures.
32P-postlabeling methods are commonly used to assess the overall DNA adduct burden as a result of exposure to complex air pollution mixtures. With respect to DNA adduct identification, improvement in 32P-postlabeling methods will be necessary if advances are to be made with respect to distinguishing the important PAH in the mixture that play a key role in the cancer process.
White blood cells (WBC) have been proposed to be useful surrogates for monitoring exposures  lymphocytes were treated in vtro with 1 pg/mL diesel particle extract or dimethylsulfoxide (DMSO). Background DNA adducts, of unknown origin, were observed in the DMSO-treated human lymphocytes (Fig. la, adducts 1-3). The presence of background levels of adducts in blood cell DNA is consistent with studies where DNA adducts were observed in nonsmokers with no known occupational exposures (11,12). Five diesel-derived DNA adducts were reproducibly detected in lymphocyte DNA (Fig. Ib,  adducts 3-7). One major DNA adduct detected in diesel-treated human lymphocyte DNA (Fig. Ib, adduct 4) co-migrated with the major BaP-derived DNA adduct detected in human lymphocytes treated with 1 pM BaP (Fig. lc, adduct4). Previous studies indicate that human lymphocytes can metabolize single chemicals as well as complex air pollution particle extracts (13,14). Future studies to evaluate the differences in labeling efficiency by comparing DNA adduct patterns in the nuclease P1 and butanol extraction versions of the postlabeling assay may reveal evidence for the presence ofadditional N-substituted arylamine adducts resulting from the nitroreduction ofnitrated PAH present in diesel extracts. 1,6-Dinitropyrene, for example, when administered via lung implantation has been shown to be metabolically activated by nitroreduction to form DNA adducts in spleen lymphocytes (15).
In vitro assays using CT DNA and rat liver S9 treated with diesel particle extract resulted in the formation of one major DNA adduct (Fig. 2a, adduct 1) detectable by the butanol and nuclease P1 version ofthe assay. Adduct 1 also co-migrated with the major BaP-diol epoxide DNA adduct detected in skin DNA (Fig. 3d, adduct 3) and with the major DNA adduct detected in DNA isolated from rodents topically treated with 50 mg diesel particle extract (Fig. 3a, adduct 1). Based on the low concentration of BaP in diesel extracts, relative to other complex air pollution mixtures that we have analyzed in this laboratory, it is not likely that the diesel-derived DNA adducts arise from BaP only (6). The formation of nitrated PAH-derived DNA adducts relative to other PAH-derived DNA adducts detected in our CT DNA model systems was achieved by XO-mediated nitroreduction of nitrated PAH constituents in the diesel extract. CT DNA treated with diesel extract and XO resulted in the formation of one major DNA adduct (Fig. 2b, adduct 2). This adduct was chromatographically distinct from the major DNA adduct observed in the rat liver S9-treated incubations (Fig. 2a, adduct 1) and the major adduct detected in rodent skin DNA treated with diesel particle extract (Fig. 3a, adduct 1). Adduct no. 2 was detectable only with the butanol extraction version of the 32P-postlabeling assay (Fig. 2b). Based on its nuclease P1 sensitivity and similarity in chromatographic properties compared to a major 1-NPderived adduct (Fig. 2c, adduct 7), adduct 2 has been tentatively identified as an N-substituted arylamine adduct.
Two adducts with different sensitivity to nuclease P1 monophosphatase activity were observed (Fig 2d, adducts 3,4) when 1-NP was metabolized to reductive and oxidative metabolites in the presence of CT DNA. Adduct 4 was unique in as much as it has not been detected in any other in vitro or in vivo experiments employed in our study. 1-NP-derived DNA adduct (adducts 3 and 4) were not observed in DNA adduct profiles detected in DNA isolated from rodents exposed via skin to diesel extract or to diesel particles via inhalation.
When rodents were exposed to diesel combustion emission for 24 months, a major nuclease-PI-sensitive DNA adduct was detected in lung DNA (Fig. 3c, adduct 2). Adduct 2 was slightly more polar compared to the major BaP-derived epoxide adduct detected in rodent skin DNA treated with 1-BaP (Fig. 3d, adduct  3).
The differences in DNA adduct patterns between the in vitro and in vivo (skin and lung) exposure likely reflect differences in metabolic pathways as influenced by route ofexposure differences (skin versus lung) or specific enzyme systems unique, for example, to rat liver S9, xanthineoxidase, and human lymphocytes. Many nitrated PAH are tumorigenic via inhalation but not via skin application (16,17), and this is consistent with differences in the type ofadducts that weobserved forthe two exposure routes. These observations emphasize the potential importanceofnitrated PAH found in diesel emissions or in air-containing nitrated aromatics. In fact, we have preliminary evidence ofsuch nitrated PAH (or other Nsubstituted arylamine adducts in human autopsy lung samples (18). Future studies will focus on othercombinationsofmetabolic pathways to further characterize-the nature ofDNA adducts as detected in in vivo experimental systems. This manuscript was presented as a poster at the Conference on Biomonitoring and Susceptibility Markers in Human Cancer: Applications in Molecular Epidemiology and Risk Assessment that was held in Kailua-Kona, Hawaii, 26 October-l November 1991.
The excellent technical assistance of E. Charlet, J. Scott, and J. Inmon is ap-preciated. This document has been reviewed in accordance with the U.S. Environmental Protection Agency policy and approved for publication. Mention of trade names or commercial products does not constitute endorsement or recommendation for use.