Photochemical Reaction of 7,12-Dimethylbenz[a]anthracene (DMBA) and Formation of DNA Covalent Adducts.

DMBA, 7,12-dimethylbenz[a]anthracene, is a widely studied polycyclic aromatic hydrocarbon that has long been recognized as a probable human carcinogen. It has been found that DMBA is phototoxic in bacteria as well as in animal or human cells and photomutagenic in Salmonella typhimurium strain TA102. This article tempts to explain the photochemistry and photomutagenicity mechanism. Light irradiation converts DMBA into several photoproducts including benz[a]anthracene-7,12-dione, 7-hydroxy-12-keto-7-methylbenz[a]anthracene, 7,12-epidioxy-7,12-dihydro-DMBA, 7-hydroxymethyl-12-methylbenz[a]anthracene and 12-hydroxymethyl-7-methylbenz[a]anthracene. Structures of these photoproducts have been identified by either comparison with authentic samples or by NMR/MS. At least four other photoproducts need to be assigned. Photo-irradiation of DMBA in the presence of calf thymus DNA was similarly conducted and light-induced DMBA-DNA adducts were analyzed by 32P-postlabeling/TLC, which indicates that multiple DNA adducts were formed. This indicates that formation of DNA adducts might be the source of photomutagenicity of DMBA. Metabolites obtained from the metabolism of DMBA by rat liver microsomes were reacted with calf thymus DNA and the resulting DNA adducts were analyzed by 32P-postlabeling/TLC under identical conditions. Comparison of the DNA adduct profiles indicates that the DNA adducts formed from photo-irradiation are different from the DNA adducts formed due to the reaction of DMBA metabolites with DNA. These results suggest that photo-irradiation of DMBA can lead to genotoxicity through activation pathways different from those by microsomal metabolism of DMBA.


Introduction
Polycyclic aromatic hydrocarbons (PAHs) are widespread genotoxic and tumorigenic environmental pollutants. It has long been known that PAHs require metabolic activation in order to exert their biological activities, including carcinogenicity [1][2][3][4].
Since skin is the largest organ in human and when concomitantly exposed to environmental chemicals and sunlight, these chemicals may be activated by photoirradiation and exert adverse health effects [12,13].
3-Methylcholanthene-induced female Sprague-Dawley rat liver microsomes were purchased from In Vitro Technologies (Baltimore, MA). Protein concentrations were determined using a protein assay based on the Bradford method using a Bio-Rad protein detection kit (Bio-Rad Laboratories, Hercules, CA).

Light Source
The UVA light box was custom made with a 4-lamp unit using UVA lamps (National Biologics). The irradiance of light was determined using an Optronics OL754 Spectroradiometer (Optronics Laboratories, Orlando, FL), and the light dose was routinely measured using a Solar Light PMA-2110 UVA detector (Solar Light Inc., Philadelphia, PA). The maximum emission of the UVA is between 340 -355 nm. The light intensities at wavelengths below 320 nm (UVB light) and above 400 nm (visible light) are about two orders of magnitude lower than the maximum at 340-355 nm.

Photo-Irradiation of DMBA
A solution (2-3 mL) of 0.4 mM DMBA dissolved in 90% ethanol was placed in a UV-transparent cuvette and photo-irradiated under UVA light to receive a light dose of 2.6 J/cm 2 /min for a period of 40, 90, and 360 min, respectively. The reaction mixture was then concentrated to about 200 µL under reduced pressure. Reversed-phase HPLC separation of the resulting photodecomposition products was accomplished using a Prodigy 5 µ ODS column (4.6 x 250 mm, Phenomenex, Torrance, CA) eluted isocratically with 90% methanol in water (v/v) at 1 mL/min. For isolation of photodecomposition products in larger amounts, a Prodigy 5 µ ODS column (10 x 250 mm, Phenomenex, Torrance, CA) eluted isocratically with 90% methanol in water (v/v) at 5 mL/min was used. Following a published procedure [27], approximately 10 µg of DNA was 32 P-postlabeled using nuclease P1 enrichment.
Adducts were separated by thin layer chromatography performed on 0.1 mm Machery Nagel 300 polyethylene imine cellulose plates (Alltech, Deerfield, IL) using the following solvent directions, D1: 0.9 M sodium phosphate, pH 6. In vitro metabolism was carried out by incubation of DMBA solution (0.8 mM dissolved in 200 µl of acetone) with shaking at 37° for 30 min in a l0 ml reaction mixture containing 0.5 mM of Tris-HCl (pH 7.5), 30 µM of MgCl 2 , l unit of glucose-6-phosphate dehydrogenase (type XII, Sigma), l mg NADP + , 6 mg of glucose-6-phosphate, l0 mg of microsomal protein, and 10 mg of purified calf thymus DNA. After incubation, the reaction was cooled with ice-water, then sequentially extracted with 5.0 mL phenol, 5.0 mL phenol/chloroform/isoamyl alcohol (v/v/v, 25/24/1), and 5.0 mL chloroform/isoamyl alcohol (v/v, 24/1). The DNA in the aqueous phase was precipitated by adding 1 mL 5 M sodium chloride followed by equal volume of cold ethanol and washed three times with 70% ethanol. After redissolving in 3 mL distilled water, the DNA concentration and purity were determined spectrophotometrically, and DNA adducts were analyzed by 32 P-postlabeling/TLC analysis with the method described above.
Instrumentation A Waters HPLC system consisting of a Model 600 controller, a Model 996 photodiode array detector, and pump was used for separation and purification of DMBA photodecomposition products. Direct exposure probe (DEP) mass spectrometry (MS) was performed on a ThermoFinnigan TSQ 700 triple quadrupole mass spectrometer operated in the electron ionization (EI) mode.

DMBA Photoproduct Analysis.
Photo-irradiation of DMBA in ethanol/water (v/v, 90/10) by UVA light at a light dose of 2.6 J/cm 2 /min for a period of 40, 90, and 360 min, respectively was conducted and the reaction mixture was separated by reversed phase HPLC (Figure 1). Based on comparison of the HPLC retention time, UV-absorption spectrum, and mass spectrum with those of DMBA, the material contained in the chromatographic peak eluting at 19.0 min was identified as the recovered DMBA. As shown in Figure 1A, the amount of DMBA decreased and the amounts of photodecomposition products increased rapidly. For collection of sufficient amount of the photodecomposition products for structural identification, the products formed after 360 min of photo-irradiation were separated by repeated preparative HPLC ( Figure 2). Based on mass ( Figure 3A) and NMR ( Figure 4A) spectral analysis, the material in the chromatographic peak eluting at 5.3 min (P5 in Figure 1C) was tentatively identified as 7-hydroxy-12keto-7-methylbenz[a]anthracene (7-OH-12-keto-7-MBA). The chromatographic peak eluting at 6.6 min (P8) was identified as 7,12-epidioxy-7,12-dihydro-DMBA. This is based on the comparison of its UV-visible absorption spectrum, HPLC retention time, mass spectrum ( Figure 3B), and NMR spectrum ( Figure 4B) with those of the authentic sample (data not shown) [26]. The material in chromatographic peak eluting at 5.8 min (P6) (Figure 2) had a mass spectrum with a molecular ion M + at m/z 272 (data not shown), suggesting this is an oxygenated DMBA. This compound has the mass spectrum, UV-visible absorption spectrum ( Figure 5A insert) and HPLC retention time ( Figure  5A) identical to those of the synthetic standard for 7-HOCH 2 -12-MBA ( Figure 5B). Thus, it confirms that this photodecomposition product is 7-HOCH 2 -12-MBA. The material in chromatographic peak (P7) eluting at 6.2 min was similarly identified as 12-HOCH 2 -7-MBA using a synthetic standard. Based on comparison of HPLC retention time and UV-visible absorption spectrum ( Figure 5C and insert) with those of the authentic BA-7,12-dione ( Figure 5D and insert), the chromatographic peak (P9) eluting at 9.4 min was identified as BA-7,12-dione.

Kinetics for the Photodecomposition of DMBA and Photoproduct Formation
The formation and decomposition of the five identified photodecomposition products, P5-P9, were studied. As shown in Figure 6, while DMBA completely decomposed at about 260 min under light irradiation, the photodecomposition products P5 and P8 reached the highest yield at about 400 min of irradiation time. Compound P9 kept increasing, suggesting that the other decomposition products gradually converted into BA-7,12-dione (P9). Compounds P6 and P7 also increased during the whole course of irradiation.
Therefore, for facilitating in mechanistic understanding, the DNA adduct formation from these two compounds was also pursued.
As shown in Figure 7, the 3',5'-bisphosphate deoxyribonucleosides of DMBA ( Figure 7A) and the four oxidized derivatives ( Figure 7C-7F) were separated by thin-layer chromatography (TLC) into multiple spots. Analysis of these resulting TLC spots indicated that the spot profiles are nearly identical. Figure 8 shows the autoradiogram of 32 P-postlabeled nuclease P1-treated calf thymus DNA from (A) photoirradiation of the DNA in the presence of DMBA by UVA light (14 J/cm 2 ) and (B) the metabolite mixture from metabolism of DMBA by rat liver microsomes in the presence of calf thymus DNA.

Reaction of Metabolites of DMBA with Calf thymus DNA and DNA Adducts Analyzed by 32 P-postlabeling/TLC
For comparison of DNA adduct profile, the metabolites obtained from metabolism of DMBA by rat liver microsome in the presence of calf thymus DNA were also analyzed by 32 P-postlabeling/TLC under identical conditions. Comparison of the DNA adduct profiles indicates that the DNA adducts formed from photo-irradiation of DMBA are different from those from reaction of DMBA metabolites (Figure 8).

Discussion
In this study, photo-irradiation of DMBA under UVA light resulted in the formation of multiple photodecomposition products, of which four products were identified, including benz[a]anthracene-7,12-dione, 7,12epidioxy-7,12-dihydro-DMBA, 7-HOCH 2 -12-MBA, and 12-HOCH 2 -7-MBA, and one product tentatively assigned as 7hydroxy-12-keto-7-MBA ( Figure 1). Although Wood et al. [26] reported that 7-CHO-12-MBA and 12-CHO-7-MBA were produced from photo-oxidation of DMBA, these two compounds were not formed under our experimental conditions. This discrepancy illustrates that the formation of photo-oxidation products from DMBA is highly dependent on experimental conditions, particularly the light wavelength.
The results of 32 P-postlabeling/TLC analysis indicate that photo-irradiation of DMBA and the four oxidized derivatives      These results highly suggest that the photo-induced DNA adduct formation from DMBA is mediated through these oxidized derivatives by two distinct pathways: (i) oxidation of DMBA at the 7-methyl group to 7-HOCH 2 -12-MBA, then 7-CHO-12-MBA, then to the reactive species; and (ii) oxidation of DMBA at the 12-methyl group to 12-HOCH 2 -7-MBA, then 12-CHO-7-MBA, then to the reactive species. Also the kinetic study of photo-irradiation of DMBA, as shown in Figure 6, indicates that BA-7,12-dione is the final and stable product. Consequently, we conclude that the reactive photodecomposition product that can bind to DNA and form DNA adducts is from 7-CHO-12-MBA or 12-CHO-7-MBA, and is further oxidized to the inert BA-7,12dione. Comparison of the DNA adduct profiles indicates that the DNA adducts formed from photo-irradiation of DMBA and from metabolism of DMBA by 3methylcholanthrene-induced rat liver microsomes are different. These results suggest that photo-irradiation of DMBA can lead to genotoxicity through activation pathways different from those by microsomal metabolism of DMBA.
Thus, our study indicates that photo-irradiation of DMBA generates genotoxic photo-oxidation products that can lead to DNA adduct formation. Besides, Boyland et al. reported that 7-HOCH 2 -12-MBA is able to cause adrenal apoplexy and mammary cancer in rats [27]. 7,12-Epidioxy-7,12-dihydro-DMBA has been shown to be toxic to chicken fibroblast cells [28]. Consequently, these results suggest that photoirradiation of PAHs can generate genotoxic products and can be highly harmful to human health. This warrants further investigation.