Nitropyrene: DNA binding and adduct formation in respiratory tissues.

Binding of 1-nitro (14C)pyrene (NP) or its metabolites to cellular DNA and protein in cultures of rabbit alveolar macrophages, lung tissue, and tracheal tissue was examined. DNA binding in tracheal tissue (136 +/- 18.3 pmole NP/mg DNA) was four to five times the levels measured in either lung tissue (38 +/- 9.4 pmole NP/mg DNA) or macrophages (26 +/- 7.5 pmole NP/mg DNA). Adduct analysis of DNA isolated from lung tissue incubated with 1-nitro[H3]pyrene in vitro resulted in the identification of 2 to 5% of the NP adducts as C8-deoxyguanosine 1-aminopyrene. NP was also bound to cellular protein in tracheal tissue and lung tissue, and at a lower level in macrophages. Cocultivation of the macrophages with lung and tracheal tissue decreased the DNA binding in tracheal tissue by 45%. Following intratracheal instillation of diesel particles (5 mg) vapor-coated with 14C-NP (380 ppm, 0.085 muCi/mg) particles into rats, 5-8% of the radioactivity remained in the lungs after 20 hr. Most of the diesel particles were also deposited in the lung. Examination of DNA and protein binding in this tissue showed 5 to 12% of the pulmonary 14C bound to protein and no detectable levels of 14C bound to DNA.

Among these NO2-PAH, 1-nitropyrene (1-NP) has been found to be a potent bacterial mutagen (12) and potentially an animal carcinogen (13). This compound has been identified and quantitated in diesel particle extracts (3,4) and may be responsible for up to 30% of the mutagenicity of these extracts (4,14). The bacterial mutagenicity of 1-NP is believed to result from reduction to reactive electrophiles which bind to DNA (12,15).
Howard and co-workers (1) have identified one DNA adduct formed by Salmonella typhimurium from the reduction of 1-NP as an N-(deoxyguanosin-8-yl)-1-aminopyrene, i.e., C8-deoxyguanosine-1-aminopyrene adduct (C8-dG-AP). They have also shown that some mammalian enzymes are capable of reducing 1 form this same adduct. In addition, mammalian tissues are capable of oxidizing 1-NP both in vivo (16) and in vitro (17,18) to metabolites that are themselves mutagenic. We have shown in an earlier study (19) that rabbit pulmonary macrophages and tissues activated 1-NP to DNA-bound adducts. In this work we evaluate DNA adducts resulting from the metabolism of 1-NP by mammalian cells and tissues to see whether and to what extent oxidative metabolism of 1-NP forms the C8-dG-AP adduct previously characterized as a product of reductive metabolism. Respiratory tissue has been selected for this evaluation because of potential for exposure to 1-NP associated with diesel particles. In vivo studies have shown that inhaled diesel particles are readily deposited in the respiratory tract (20,21) in particular, in the alveolar macrophages (21,22). Although many of the particles are rapidly cleared from the lung, a substantial portion remain in the lung for a long period (20,22), thus increasing exposure to compounds associated with the particles. It has also been shown that both 1-NP and mutagenic activity were lost from diesel particles after incubation with pulmonary alveolar macrophages in vitro (23). We have now used diesel particles vapor-coated with 14Clabeled 1-NP to study in vivo the fate of 1-NP absorbed onto diesel particles and evaluate interactions with pulmonary macromolecules.
After incubation, the attached macrophages were detached by trypsinization and recovered by centrifugation at 300g, then lysed with 25 mL lysing buffer (50 mM Tris-HCl, 10 mM EDTA, 1% SDS, pH 8.0). The lung or tracheal tissues were separated from the culture medium by filtration through nylon mesh, then homogenized in 20 mL of the lysing buffer. A full description of these procedures has been given by King et al. (19).
Lung tissue (29.0 ± 2.0 ,ug/mL) was also incubated with 3H-1NP (2 ,ug/mL) as described above to provide modified DNA of higher specific radioactivity for identification of the adducts formed.

In Vivo Studies
Male rats (300 g; Sprague-Dawley, Charles River CD-1, Wilmington, MO) received 14C-NP-coated diesel particles by intratracheal instillation of 5 mg of particles in 0.2 mL phosphate-buffered saline with 1% Tween 80 under light halothane anesthesia. The animals were maintained individually in metabolism cages (Nalgene Co., Rochester, NY) for collection of urine and feces and received food and water ad libitum.
After 24 hr each animal was anesthetized with ether and the thoracic cavity opened to expose the lungs. The lungs were removed en bloc following total exsanguination of the animal. All vascular and connective tissues were removed and the lung tissue was placed into a 50-mL tube on ice. The tissue was thoroughly minced, than an aliquot (1 g) was weighed out and homogenized in 10 mL lysing buffer.
Quantitation of DNA and Protein Binding DNA and protein were isolated according to an extraction scheme previously described (25) and illustrated in Figure 1. Briefly, protein was extracted from cellular nucleic acid in the lysates or homogenates with chloroform:isoamyl alcohol:phenol (CIP, 24:1:25) and precipitated with acetone. DNA was removed from the aqueous portion by precipitation with cold ethanol. Residual protein and RNA were removed by incubation with proteinase K (1 mg/mL) and NaOH, respectively. The radioactivity associated with the DNA was determined by liquid scintillation counting after acid precipitation onto glass fiber filters. Radioactivity associated with protein was determined by liquid scintillation counting of NaOH digested protein neutralized with HCI. Total DNA content was measured by the diphenylamine colorimetric assay (26) and total protein by the Lowry method (27).

Identification of DNA Adducts
The isolated DNA was redissolved in water and precipitated with ethanol in presence of sodium acetate until at least 95% of the radioactivity was precipitable. The DNA (<1 mg/mL) was dissolved in 10 mM Tris-HC1-0. 1 M NaCl-5 mM MgCl2 (pH 7.9) and was treated consecutively with 200 units of DNase I for 2 hr, 5 units of alkaline phosphatase for 2 hr, 2 units of phosphodiesterase I for 4 hr, 2 units phosphodiesterase II for 2 hr, and 5 units of alkaline phosphatase for 2 hr.
The resulting deoxyribonucleosides were then separated on a Sephadex LH-20 column. After application of the sample, the column was washed with water to remove unmodified deoxyribonucleosides. The modified adducts were then eluted with methanol and the solvent evaporated under reduced pressure. The sample was dissolved in 10 ,uL of methanol containing unlabeled 1nitropyrene-modified DNA adducts [prepared according to Howard et al. (1)] and was analyzed by HPLC on a Dupont Instruments Model 850 high-pressure liquid chromatograph fitted with a Cl8-puBondapak column (Waters Associates) using a concave gradient (number 2) from 50 to 100% methanol in water at 50°C and a flow rate of 1   Protein in the CIP portion was precipitated with acetone and DNA in the aqueous portion was precipitated with cold ethanol. After residual protein and RNA were hydrolyzed with proteinase K and NaOH, respectively, DNA was acid precipitated onto glass fiber filters and associated radioactivity measured in a liquid scintillation spectrometer. Samples used for identification of DNA adducts were not taken through the hyrolysis procedure, but were prepared directly as described in "Materials and Methods." at 254 nm, and 1-min fractions were collected then assayed for radioactivity by liquid scintillation counting.

Results
Binding levels of 14C-lNP metabolites to DNA and protein from respiratory tissues exposed in vivo are presented in Figure 2. Tracheal tissue exhibited the highest binding to DNA ( Fig. 2A) and protein (Fig. 2B) in both the separate 20-hr cultures of isolated macrophages and lung and tracheal tissue and the cocultured system.
DNA from lung tissue incubated for 20 hr with 3H-1NP was hydrolyzed and the resulting deoxyribonucleosides prepared for identification of modified adducts by high-pressure liquid chromatography as described in "Materials and Methods." A chromatogram from this analysis is shown in Figure 3. From 2 to 5% of the total DNA adducts co-eluted with the synthetic C8-dG-AP adduct in fraction number 60 (retention time 60 min). However, the majority (over 90%) of the modified DNA was not retained on the Sephadex LH-20 column and therefore was not amenable to HPLC analysis. Of the material retained on the LH20 column, 30% or more eluted with the solvent front, and the other two peaks observed (35 and 94 min) did not correspond to any synthetic standard currently available.
'4C-lNP vapor-coated onto diesel particles was administered to rats by intratracheal instillation to investigate the binding of 1-NP to respiratory tissue in vivo in an attempt to more closely model realistic exposure routes. Results are presented in Figure 4. After 24 hr, 6-8% of the total 14C dose as well as a major portion of the particles (based on visual inspection) remained in the lungs. Analysis of DNA and protein binding in the lung tissue showed that 5 to 13% of the residual pulmonary 14C (representing 0.6% of the total dose) was bound to protein, and no detectable 14C was bound to DNA.

Discussion
Although macrophages and respiratory tissue are known to be able to metabolize PAH to reactive inter-  . Chromatogram of lung DNA adducts fr( pyrene. This chromatogram represents the HPL( modified deoxyribonucleosides obtained from the hyc isolated from lung tissue exposed in vitro to 3H-11 ,uCi/mL). The major peak (fraction 60) is identif deoxyguanosine-1-aminopyrene adduct and represer total activity bound to DNA.
mediates that bind DNA (28), the metabolis ing of NO2-PAH have only recently begu attention. We have previously presente( clearly show that macrophages and tissu respiratory tract are capable of binding 1-N tabolites to DNA and protein (19). We hav tified one adduct bound to DNA from lung tie to 3H-1NP. The N-(deoxyguanosin-8-yl)-1-aminopyrene (C8-dG-AP) adduct, which had previously been identified as the major DNA adduct formed from 1-NP by Salmonella typhimurium TA 1538 and by xanthine oxidase (1), was found to represent a small portion (2-5%) of the total DNA adducts formed by lung tissue. The majority of the NP-DNA adducts either did not respond to the digestive enzyme treatments, or yielded mononucleoside products that were not hydrophobic enough to be retained on the Sephadex LH-20 column. This is in marked contrast to the xanthine oxidase-catalyzed adducts, of which more than 80% were extractable and resolved into the C8-dG-AP adduct, and to the adducts formed in TA 1538, where the C8-dG-AP adduct was 80 100 the only peak detected by HPLC (1). However, the presence of even small quantities of C8-dG-AP in mixtures of NP-DNA adducts from lung tissue is signifiom [ .1-nitrocant, in that it indicates that some portion of pulmonary c profile of the metabolism and activation of 1-NP is by reduction of irolysis of DNA the nitro function, thus forming a C8-dG type of adduct NP (8.1 ,uM; 48 that appears to be a characteristic of arylamines acti-Red as the C8-vated by N-hydroxylation (1). its 2-5% of the Analysis of binding to DNA and protein from lung tissue exposed in vivo to 14C-lNP vapor-coated onto diesel particles showed that a very small portion of the total radioactivity instilled in the lungs was bound to 3m and bindprotein and that there was no detectable DNA binding. d to receive This is significant, however, because less than 10% of d data that the total activity administered remained in the lung es from the tissue even though most of the particulate matter was [P or its mestill present. This indicates that not only is respiratory se now identissue able to metabolize 1-NP, but that it is capable of ssue exposed removing it from diesel particles in vivo, as had previously been demonstrated for pulmonary alveolar macrophages in vitro (23). The majority of NP originally present on the diesel particles is therefore fully available for distribution throughout the body and for further metabolism and possible activation. t The information we have presented here strongly suggests that additional studies need to be conducted lat J on the metabolism and binding of the NO2 PAH found lat C in diesel exhaust and urban air. More information is needed for defining the genotoxic burden of NO2 PAH in ambient air. Further studies are currently being con--24 hours ducted in our laboratories to elucidate better the DNA and protein binding of 1-NP to tracheal epithelial cells. In addition, new DNA hydrolysis techniques are being evaluated to increase total DNA adduct recovery and identification.
Part of the research described in this article is conducted under contract to the Health Effects Institute (HEI, #83-23, with the Columbia University Health Science Center), an organization that supports the conduct of independent research and is jointly funded by the United States Environmental Protection Agency (EPA) and auie vapor-coated tomotive manufacturers. Publication here implies nothing about the iP vapor-coated view of the contents by HEI of its research sponsors. HEI's Health th the particles. Review Committee may comment at any time and will evaluate the then the lungs final report of the project.
NA and protein ules was meas- The work described in this document has also been funded in part by the U. S. Environmental Protection Agency under Contract 68-02-4031 with Environmental Research and Testing, Inc., under cooperative agreement CR-810849-010 with the Department of Environmental Sciences and Engineering, University of North Carolina, and under assistance agreement X808859 with HEI. The contents, however, do not reflect the views and policies of the Agency; nor does mention of trade names or commercial products constitute endorsement or recommendations for use.