Quinone Methide Formations in the Cu2+-Induced Oxidation of a

bacteria, fungi, and cancerous cell lines (1-7). The biological activity has been ..... exclusion of the oxygen from the air during the workup of this...
0 downloads 0 Views 236KB Size
Download PDF
382

Chem. Res. Toxicol. 2005, 18, 382-388

Quinone Methide Formations in the Cu2+-Induced Oxidation of a Diterpenone Catechol and Concurrent Damage on DNA Qibing Zhou* and Miguel A. Zuniga Department of Chemistry, Virginia Commonwealth University, 1001 West Main Street, Richmond, Virginia 23284 Received October 28, 2004

Terpene quinone methides have been isolated from natural resources and exhibit broad biological activities against bacteria, fungi, and tumor cells through the reactive quinone methide (QM) moiety. The biological potential of the oxidation of terpene QM precursors, however, has not been assessed even though Cu2+-induced oxidation of catechol shows detrimental effects on cells. In this study, a diterpenone catechol was investigated as a precursor of terpene QM under aqueous conditions in the presence of Cu2+. Direct QM formation was implied in the Cu2+-induced oxidation through the study of thiol addition using HPLC and ESI-MS analysis. In addition, oxidation of the initial QM adduct to a second-QM intermediate was observed. The direct QM oxidation pathway may be unique for diterpenone catechol in the Cu2+-induced oxidation and is an addition to the reported isomerization pathway of o-quinones to QMs. The DNA damage by the Cu2+-induced oxidation of diterpenone catechol was assessed on a short duplex DNA target. Both direct DNA cleavage and nucleobase oxidation were observed extensively by in situ-generated hydroxyl radicals.

Introduction Terpene quinone methides (QMs)1 have been isolated from a variety of plants and show broad activities against bacteria, fungi, and cancerous cell lines (1-7). The biological activity has been attributed to a reactive QM moiety in the structure (Scheme 1, 1-7). QM has been widely used as an alklyating agent in biological applications (8-11). The QM moieties of natural terpene QMs share high structural similarities including a hydroxyl group adjacent to the carbonyl carbon and the extended conjugation of the exocyclic alkene and thus reflect similar reaction mechanisms. Although terpene QMs and their analogues have been studied for medicinal applications (12, 13), the oxidative potential of terpene QM precursors, homoconjugated catechols, has not been investigated under biological conditions. Under aqueous conditions, the in situ oxidation of catechol derivatives has been shown to have detrimental effects on cells. For example, the oxidation of 2-hydroxyphenol in the presence of Cu2+ causes nonselective DNA cleavages and nucleobase oxidation by in situ-generated hydroxyl radicals (14-16). In addition to oxidative stress, catechol oxidation is responsible for the generation of carcinogenic metabolites of estrogen, which form covalent DNA nucleobase adducts (17-21). The Cu2+-induced oxidation of catechol proceeds through a two-step one-electron transfer process, and o-quinone is the sole oxidation product (14-16). The oxidation mechanism involves the disproportionation of the Cu(II)/ * To whom correspondence should be addressed. Tel: 804-828-3520. Fax: 804-828-8599. E-mail: [email protected]. 1 Abbreviations: QM, quinone methide; EDCI, 1-(3-dimethylaminopropyl)-3-ethylcarbodiimide hydrochloride; COSY, correlated spectroscopy; NOESY, nuclear overhauser and exchange spectroscopy; ESI-MS, electrospray ionization mass spectroscopy.

Scheme 1. Examples of Natural Di- and Triterpene QMs

Scheme 2. Possible Oxidation Pathways of a Diterpenone Catechol

Cu(I) redox cycle and the transfer of electrons to O2 generating reactive hydroxyl radical in situ. For a homoconjugated terpene catechol 1, the Cu2+-induced oxidation may undergo three different pathways (Scheme 2). Either terpene o-quinone 2 or terpene QM 3 is directly formed in the oxidation process, or terpene QM 3 is formed through the isomerization of terpene o-quinone 2 (19). In contrast to o-quinone, there is no report on the direct QM formation in the Cu2+-induced oxidation under aqueous conditions although QMs have been generated by other oxidants under organic conditions (22-25). On the other hand, isomerization of o-quinone to QM has been demonstrated with a homoconjugated catechol and the rate of the isomerization was found to be slower than the initial o-quinone formation (19, 26, 27). In this study, we present the studies of QM formations from a homo-

10.1021/tx049703a CCC: $30.25 © 2005 American Chemical Society Published on Web 02/04/2005

Quinone Methide Formations and Concurrent DNA Damage

conjugated diterpenone catechol 1 in the Cu2+-induced oxidation under aqueous conditions and the concurrent oxidative damage on DNA.

Experimental Procedures All chemicals were purchased from Fisher Scientific (Pittsburgh, PA) or Sigma-Aldrich (Milwaukee, WI) and used without further purification. The NMR spectra of the synthesized compounds were obtained by Variant NMR spectrometers. HPLC analysis was carried out on a Jasco 2000 series station (Easton, MD) using a Microsorb MV C18 column (2.1 mm × 250 mm, 8 µm) from Varian (Walnut Creek, CA). Electrospray ionization mass spectroscopy (ESI-MS)/MS analysis was carried out on a Thermo Finnigan LCQ Deca XP Plus mass spectrometer (Waltham, MA). Oligonucleotides were purchased from Nucleic Acids Research Facilities at Virginia Commonwealth University (Richmond, VA). Aqueous solutions of [γ-32]ATP (250 µCi) were purchased from ICN Pharmaceuticals (Costa Mesa, CA). Gel images of isotopic 32P were obtained by a Molecular Dynamics Typhoon 8600 Variable Mode Imager (Sunnyvale, CA), and water was purified through a Barnstead E-pure 4-Module Deionization System (Dubuque, IA). Isopropyl (4E)-2-(3,4-Dimethoxyphenyl)-5,9-dimethyldeca-4,8-dienoate (5). To a solution of fresh distilled diisopropylamine (3.1 mL, 22 mmol) in dry THF (10 mL) at -78 °C under N2 was added a solution of n-butyllithium (2.5 M, 8.8 mL). After 10 min at -78 °C, a solution of 3,4-dimethoxyphenylacetic acid (2.0 g, 10 mmol) in THF (10 mL) was slowly added. The resulting solution was stirred for 1 h at -78 °C, and then, geranyl chloride (2.0 mL, 11 mmol) was added. The reaction solution was stirred under N2 for 18 h and allowed to slowly warm to room temperature. The reaction solution was then quenched with 1 N HCl (250 mL) and extracted with CH2Cl2 (250 mL × 2). The organic layers were collected, washed with brine, dried with MgSO4, and concentrated. Flash chromatographic separation (0.5-7% MeOH in CH2Cl2) afforded the free acid as a yellow oil (3.0 g) in 89% yield. 1H NMR (CDCl3, 300 MHz): δ 6.87-6.79 (m, 3H), 5.06-4.99 (m, 2H), 3.87(s, 3H), 3.85 (s, 3H), 3.49 (t, J ) 7.5 Hz, 1H), 2.79-2.69 (m, 1H), 2.50-2.40 (m, 1H), 2.00-1.93 (m, 4H), 1.64 (s, 3H), 1.56 (s, 3H), 1.56 (s, 3H). 13C NMR (CDCl3, 75 MHz): δ 179.4, 149.1, 148.5, 138.0, 131.6, 130.9, 124.2, 120.7, 120.5, 111.2, 111.2, 56.0, 51.4, 39.9, 31.9, 26.7, 25.8, 17.8, 16.3. HRMS calcd for C20H28O4, 332.1988; found, 332.2003. To a solution of the resulting acid (3.0 g, 9.0 mmol) in CH2Cl2 (25 mL) were added 4-(dimethylamino)pyridine (110 mg, 0.9 mmol), 1-(3-dimethylaminopropyl)-3-ethylcarbodiimide hydrochloride (EDCI) (1.7 g, 1.1 mmol), and 2-propanol (0.76 mL, 9.9 mmol). The resulting reaction mixture was stirred under N2 for 18 h at room temperature. The reaction solution was diluted with a solution of 0.1 N HCl (250 mL) and extracted with CH2Cl2 (250 mL × 3). The organic layers were collected, washed with brine, dried with MgSO4, and concentrated. Flash chromatographic separation (5-10% EtOAc in hexanes) afforded product 5 as a colorless oil (2.5 g) in 76% yield. 1H NMR (CDCl3, 300 MHz): δ 6.87-6.77 (m, 3H), 5.06-4.96 (m, 3H), 3.87 (s, 3H), 3.85 (s, 3H), 3.43 (t, J ) 7.9 Hz, 1H), 2.77-2.67 (m, 1H), 2.452.35 (m, 1H), 2.00-1.93 (m, 4H), 1.66 (s, 3H), 1.57 (s, 6H), 1.21 (d, J ) 6.3 Hz, 3H), 1.14 (d, J ) 6.3 Hz, 3H). 13C NMR (CDCl3, 75 MHz): δ 173.7, 148.9, 148.2, 137.6, 132.0, 131.6, 124.3, 121.1, 120.0, 111.1, 111.0, 68.0, 56.0, 51.9, 39.9, 32.6, 26.8, 25.9, 22.0, 21.8, 17.8, 16.4. HRMS calcd for C23H34O4, 374.2457; found, 374.2463. 1,2,3,4,4a,9,10,10a-Octahydro-6,7-dimethoxy-1,1,4a-trimethylphenanthrene-9-carboxylic Acid (6). To a solution of 5 (2.5 g, 6.8 mmol) in dry CH3NO2 (20 mL) at -15 °C was added BF3‚Et2O (1.6 mL, 13 mmol), and the resulting reaction solution was stirred under N2 at -15 °C for 2 h and then room temperature for 0.5 h. The reaction solution was diluted with a solution of saturated NaHCO3 (150 mL) and extracted with CH2Cl2 (150 mL × 3). The organic layers were collected,

Chem. Res. Toxicol., Vol. 18, No. 2, 2005 383 washed with brine, dried with MgSO4, and concentrated. Flash chromatographic separation (10-12% EtOAc in hexanes) afforded the diastereoisomeric mixture of isopropyl ester as a colorless oil (2.1 g) in 82% yield. This diastereoisomers were not further separated since one of the chiral centers was removed during the decarboxylation step. 1H NMR (CDCl3, 300 MHz) for the major isomer: δ 6.76 (s, 1H), 6.63 (s, 1H), 5.12-5.00 (m, 1H), 3.84 (s, 3H), 3.80 (s, 3H), 3.80-3.79 (m, 1H), 2.26-1.99 (m, 3H), 1.77-1.41 (m, 6H), 1.29-1.16 (m, 10H), 0.97 (s, 3H). 13C NMR (CDCl , 75 MHz) for the isomers: δ 175.2, 174.9, 148.0, 3 147.0, 142.9, 123.9, 112.1, 110.3, 107.6, 67.9, 55.8, 55.6, 48.9, 46.9, 41.5, 39.0, 37.3, 33.3, 33.1, 24.8, 23.1, 21.8, 21.7, 21.4, 19.1. HRMS calcd for C23H34O4, 374.2457; found, 374.2446. To a solution of the resulting ester (2.0 g, 5.3 mmol) in ethanol (20 mL) was added crushed NaOH pellet (2.1 g), and the resulting mixture was refluxed for 3 h. The reaction solution was acidified with 2.5 N HCl (150 mL), and the aqueous solution was then extracted with CH2Cl2 (150 mL × 3). The organic layers were collected, washed with brine, dried with MgSO4, and concentrated to afford the diastereoisomeric mixture of 6 as a viscous oil (1.7 g) in 96% yield. 1H NMR (CDCl3, 300 MHz) for the major isomer: δ 6.77 (s, 1H), 6.67 (s, 1H), 3.84 (s, 3H), 3.81 (s, 3H), 3.81 (m, 1H), 2.34-2.17 (m, 2H), 2.07-1.98 (m, 1H), 1.76-1.16 (m, 9H), 0.94 (s, 3H), 0.94 (s, 3H). 13C NMR (CDCl3, 75 MHz) for the major isomer: δ 182.0, 148.3, 147.1, 143.3, 122.9, 111.0, 107.8, 56.1, 56.0, 49.2, 46.5, 41.7, 39.2, 37.5, 33.6, 33.3, 24.9, 23.6, 21.6, 19.5. HRMS calcd for C20H28O4, 332.1988; found, 332.1985. 1,2,3,4,4a,10a-Hexahydro-6,7-dimethoxy-1,1,4a-trimethylphenanthrene (7). To a mixture of 6 (1.6 g, 4.8 mmol), lead(IV) acetate (4.8 g, 11 mmol), and copper(II) acetate (0.98 g, 5.4 mmol) was added quinoline (30 mL). The resulting dark solution was degassed under vacuum and then heated under N2 at 130 °C for 3 h. After it was cooled to room temperature, the reaction solution was diluted with 1 N HCl (200 mL) and extracted with CH2Cl2 (150 mL × 3). The organic layers were collected, washed with brine, dried with MgSO4, and concentrated. Quinoline from the resulting residue was removed by vacuum distillation. Flash chromatographic separation (5-10% EtOAc in hexanes) afforded product 7 as a colorless oil (720 mg) in 51% yield. 1H NMR (CDCl3, 400 MHz): δ 6.72 (s, 1H), 6.60 (s, 1H), 6.45 (d, J ) 2.8 Hz, 1H), 5.91 (d, J ) 2.8 Hz, 1H), 3.87 (s, 3H), 3.84 (s, 3H), 2.14-2.08 (m, 2H), 1.77-1.49 (m, 4H), 1.26-1.19 (m, 1H), 1.03 (s, 3H), 1.00 (s, 3H), 0.96 (s, 3H). 13C NMR (CDCl3, 75 MHz): δ 148.3, 146.8, 141.4, 128.7, 127.1, 126.1, 110.0, 106.3, 56.2, 56.1, 51.5, 41.2, 37.9, 36.4, 33.0, 32.8, 22.7, 20.2, 19.2. HRMS calcd for C19H26O2, 286.1933; found, 286.1944. 4b,5,6,7,8,8a-Hexahydro-2,3-dimethoxy-4b,8,8-trimethylphenanthren-9(10H)-one (8). To a solution of 7 (184 mg, 0.64 mmol) in CH2Cl2 (10 mL) at 0 °C was added m-chloroperoxybenzoic acid (70-75%, 220 mg). The resulting reaction solution was stirred at 0 °C under N2 for 3 h. The reaction solution was quenched with a solution of 5% Na2S2O3 and extracted with CH2Cl2 (100 mL × 2). The organic layers were collected, washed with brine, dried with MgSO4, and concentrated. The residue was dissolved in CHCl3 (10 mL), and trifluoroacetic acid (1 mL) was added. The resulting reaction solution was stirred under N2 for 18 h and then diluted in CH2Cl2 (100 mL). The organic solution was washed with solutions of saturated NaHCO3 and brine, dried with MgSO4, and concentrated. Flash chromatographic separation (10-15% EtOAc in hexanes) afforded product 8 as a colorless oil (88 mg) in 45% yield. 1H NMR (CDCl3, 400 MHz): δ 6.83 (s, 1H), 6.53 (s, 1H), 3.88 (s, 3H), 3.85 (s, 3H), 3.55 (s, 2H), 2.41 (s, 1H), 2.31-2.27 (m, 1H), 1.79-1.68 (m, 3H), 1.45-1.39 (m, 1H), 1.31 (s, 3H), 1.31-1.16 (m, 1H), 1.14 (s, 3H), 1.08 (s, 3H). 13C NMR (CDCl3, 75 MHz): δ 209.8, 147.9, 147.7, 124.4, 111.0, 107.4, 63.0, 56.3, 56.1, 44.9, 42.9, 40.6, 39.0, 33.1, 32.7, 24.9, 21.7, 18.9. HRMS calcd for C19H26O3, 302.1882; found, 302.1898. 4b,5,6,7,8,8a-Hexahydro-2,3-dihydroxy-4b,8,8-trimethylphenanthren-9(10H)-one (1). To a mixture of 8 (50 mg, 0.17 mmol) in CH2Cl2 (3.0 mL) under N2 was added BBr3 (1.0 M in

384

Chem. Res. Toxicol., Vol. 18, No. 2, 2005

heptane, 0.5 mL, 0.5 mmol). The resulting dark solution was stirred at room temperature for 2 h. The reaction solution was quenched with a solution of brine (100 mL) and extracted with degassed EtOAc (100 mL × 2). The organic layers were collected, dried with MgSO4, and concentrated. Flash chromatographic separation (25-35% EtOAc in hexanes) under N2 afforded product 1 as a brown oil (38 mg) in 82% yield. 1H NMR (CD3CN, 400 MHz): δ 6.80 (s, 1H), 6.58 (br s, 1H), 6.51 (s, 2H), 3.52 (d, J ) 21 Hz, 1H), 3.38 (d, J ) 21 Hz, 1H), 2.41 (s, 1H), 2.26-2.23 (m, 2H), 1.76-1.69 (m, 1H), 1.63-1.60 (m, 2H), 1.401.36 (m, 1H), 1.27 (s, 3H), 1.21-1.16 (m, 1H), 1.08 (s, 3H), 1.04 (s, 3H). 13C NMR (CD3CN, 75 MHz): δ 209.9, 143.3, 142.9, 141.3, 124.3, 114.7, 110.9, 62.4, 44.3, 42.5, 40.1, 38.7, 32.4, 32.3, 24.3, 21.2, 18.7. ESI-MS calcd for C34H44O6-H+ (dimer), 547, and for C17H22O3-H+, 273; found, 547 and 273. HPCL Analysis of Diterpenone Catechol under Aqueous Conditions in the Absence or Presence of CuCl2. Stock solutions of compound 1 (10 mM in acetonitrile), sodium phosphate buffer (50 mM, pH 7.0), CuCl2 (10 mM), and mercaptoethanol (10 mM) were prepared for the following experiments. A gradient eluting method was used for the HPLC analysis starting from 20% CH3CN in water to 100% CH3CN over 30 min at a flow rate of 1.0 mL/min. The ESI-MS analysis of the HPLC signals was carried out in the negative mode with an ionizing voltage of 4.5 keV. All of the experiments were repeated at least twice. The stability of 1 under aqueous conditions was monitored by HPLC analysis over 24 h at ambient temperature. The reaction solution contained 0.25 mM of compound 1, 5.0 mM phosphate buffer (pH 7.0), and 10% acetonitrile. The amount of compound 1 after 24 h was calculated based on the area integration vs that of an internal standard (1,3,5-trimethoxybenzene). The oxidation of compound 1 in the presence of Cu2+ was initiated by the addition of a solution of CuCl2 to a final concentration of 1.0 mM. Formation of Diterpenone QM under Organic Conditions and Thiol Addition. To a solution of compound 1 in acetonitrile-d3 (7.5 mM, 0.80 mL) was added Ag2O (5 mg), and the resulting suspension was stirred vigorously at ambient temperature for 20 min. The solids were removed by filtration with a 0.2 µm filter (Acrodisc, 13 CR, PTFE). The resulting yellow solution was confirmed as the desired QM by 1H NMR analysis. Because of the high reactivity of QM, attempts of 13C NMR and MS analysis were not successful. 1H NMR for diterpenone QM (CD3CN, 400 MHz): 7.43 (br 1H), 6.51 (s 1H), 6.40 (s 1H), 6.30 (s 1H), 2.60 (s 1H), 1.80-1.60 (m, 4H), 1.401.20 (m, 2H), 1.24 (s, 6H), 1.15 (s, 3H). Thiol addition was achieved by adding mercaptoethanol (1 µL) to the reaction solution, and the disappearance of diterpenone QM was monitored by 1H NMR analysis over 15 h. The formation of the diterpenone QM-thiol adduct was confirmed by a HPLC analysis and ESI-MS analysis. ESI-MS for compound 9: calcd for C38H52O8S2-H+ (dimer), 699, and for C19H26O4S-H+, 349; found, 699, 349, 287, and 271. HPLC Analysis of Thiol Addition on Diterpenone Catechol in the Presence of CuCl2 under Aqueous Conditions. The oxidation of 1 by Cu2+ in the presence of mercaptoethanol was initiated by the addition of 1. The final reaction solution contained 0.25 mM of compound 1, 5.0 mM phosphate buffer (pH 7.0), 1.0 mM CuCl2, 0.75 mM HSCH2CH2OH, and 10% acetonitrile. The progress of the reaction was monitored by HPLC analysis over 9 h at ambient temperature. Mercaptoethanol Adduct 10. To a solution of phosphate buffer (50 mM, pH 7.0, 10 mL) were added water (62.5 mL), CuCl2 (10 mM, 10 mL), mercaptoethanol (10 mM, 7.5 mL), and compound 1 (5 mg in acetonitrile, 10 mL). The resulting solution was stirred at room temperature for 1 h. The reaction solution was then extracted with EtOAc (50 mL × 2), and the organic layer was collected and then concentrated. The resulting residue was dissolved in 50% CH3CN in H2O, and adduct 10 was purified by a reverse phase HPLC separation as a white solid (2.1 mg); mp 205 °C (decomp.). 1H NMR (MeOH-d4, 400 MHz):

Zhou and Zuniga Scheme 3. Synthesis of Diterpenone Catechol 1

6.94 (s, 1H), 6.63 (s, 1H), 4.50 (t, J ) 6 Hz, 2H), 3.37-3.32 (m, 1H), 3.20-3.14 (m, 1H), 2.93 (s, 1H), 2.22 (br d, J ) 13 Hz, 1 H), 1.82-1.63 (m, 4H), 1.38 (s, 3H), 1.22-1.20 (m, 1H), 1.18 (s, 3H), 1.00 (s, 3H). 13C NMR (MeOH-d4, 75 MHz): 203.7, 146.8, 144.5, 141.6, 125.7, 115.1, 109.7, 94.1, 73.5, 59.4, 42.6, 42.1, 39.0, 33.9, 32.3, 31.1, 25.5, 21.4, 18.8. ESI-MS calcd for C38H48O8S2-H+ (dimer), 695, and for C19H24O4S-H+, 347; found, 695, 347, and 287. Oxidative Damage of DNA by Diterpenone Catechol in the Presence of CuCl2. The 30-mer oligonucleotide of the target DNA was radiolabeled with 32P-phosphate at the 5′position by T4 polynucleotide kinase (Fisher Scientific) according to the manufactory instructions. Hybridization of the complimentary strands was achieved by heating a stock solution of oligonucleotides (4.0 µM each, 0.28 µCi/µL) in 20 mM phosphate buffer (pH 7.0) with a 90 °C water bath and then allowed to cool to room temperature slowly. A series of reaction solutions were prepared with different concentrations of CuCl2, and the DNA lesion was initiated by the addition of a solution of compound 1 in acetonitrile. The final reaction solutions contained 1.0 µM duplex DNA (0.07 µCi/µL), 5.0 mM phosphate buffer (pH 7.0), CuCl2 (0, 750, 0, 750, 250, 83, 150, 50, or 17 µM), compound 1 (0, 0, 250, 250, 250, 250, 50, 50, or 50 µM, respectively), and 10% acetonitrile. The resulting reaction solutions were incubated at 37 °C for 2 h. A portion of each reaction solution (0.1 µCi) was then mixed with formamide and directly separated by a 20% denatured PAGE. The piperidine treatment was achieved by mixing the reaction solution (10 µL each) with a 10% piperidine solution (100 µL) and subsequently heating at 90 °C for 20 min. The resulting solutions were lyophilized, and the residues were dissolved in 90% formamide loading buffer. Each reaction solution (0.15 µCi) was separated by 20% denatured PAGE and analyzed by gel image analysis software. As a control, the radiolabeled single strand was included in the above investigation under similar conditions. On the basis of the gel analysis, there is no DNA alkylation product observed while extensive DNA cleavages and oxidative damages were observed.

Results Synthesis of Diterpenone Catechol 1. The homoconjugated diterpenone catechol 1 was designed as a simplified model of the triterpene QMs that show antitumor activity possibly through DNA intercalation and alkylation (1, 2). Diterpenone catechol was synthesized based on published procedures with modifications (Scheme 3, 12, 28). Homoveratric acid 4 was first coupled with

Quinone Methide Formations and Concurrent DNA Damage

Chem. Res. Toxicol., Vol. 18, No. 2, 2005 385 Scheme 4. Potential Thiol Adducts of Diterpenone o-Quinone and p-QM

Scheme 5. Diterpenone QM Formation and Thiol Addition under Organic Conditions

Figure 1. Stability of diterpenone catechol 1 in the absence and presence of Cu2+ by HPLC analysis at room temperature. The reaction solution contained compound 1 (0.25 mM) and phosphate buffer (5.0 mM, pH 7.0) without/with CuCl2 (1.0 mM) in 10% acetonitrile.

geranyl chloride, and then, the carboxylic acid group was converted to the isopropyl ester. The polyene cyclization of 5 by BF3‚OEt2 afforded a mixture of diastereoisomers, which were not separated since one of the chiral centers was removed during the decarboxylation step. Subsequent hydrolysis of the ester group and decarboxylation resulted in styrene 7 in 51% yield. The trans conformation of compound 7 formed in the cyclization step was confirmed at this stage by correlated spectroscopy (COSY) and nuclear overhauser and exchange spectroscopy (NOESY) NMR analysis. The 1H NMR chemical shifts were also consistent with those of similar trans derivatives reported (1-7). Epoxidation of styrene 7 with m-chloroperoxybenzoic acid was followed by an acidcatalyzed rearrangement to ketone 8 in 45% yield. Finally, the methyl protecting groups on 8 were removed with BBr3 at room temperature (29). We found that exclusion of the oxygen from the air during the workup of this step was necessary for the isolation of pure racemic diterpenone catechol 1. Stability of Diterpenone Catechol in the Absence and Presence of Cu2+. The stability of diterpenone catechol in solutions was determined to assess the potential of spontaneous oxidation. We found that a solution of diterpenone catechol in 100% acetonitrile could be stored for at least 1 month at -5 °C, and no degradation of the diterpenone catechol was detected by HPLC analysis. Under aqueous conditions (5 mM phosphate buffer, pH 7.0, 10% acetonitrile), HPLC analysis showed that more than 80% of diterpenone catechol 1 remained unchanged after 24 h at ambient temperature (Figure 1a,b) and the integrity of 1 was confirmed by ESMS analysis. However, upon the addition of a solution of Cu2+ (1.0 mM), diterpenone catechol 1 was completely consumed within 2 min (Figure 1c) and broad-tailed signals were observed in the HPLC chromatogram starting from a retention time of 20 min. Direct ESI-MS analysis of the HPLC fractions of these regions failed to provide any structural information or identity of possible oxidized species. Although the Cu2+-induced oxidation of diterpenone catechol 1 observed was similar to that of catechol, it was unclear whether p-QM or o-quinone was formed during the oxidation and thus an indirect approach was necessary. QM Formations in the Cu2+-Induced Oxidation of Diterpenone Catechol. The identity of the oxidized

product was revealed indirectly by the analysis of thiol adducts since nucleophilic additions to similar p-QMs and o-quinones have been reported (26, 27, 30-32). For the diterpenone catechol, either the quinone or the p-QM should afford two isomeric thiol adducts (Scheme 4). If the isomerization of o-quinone to p-QM occurred at a comparable rate as the thiol addition, four possible thiol adducts should be formed. A direct and convenient approach was to first investigate the thiol addition on the preformed p-QM of diterpenone catechol under organic conditions and then the thiol addition on the oxidation product of diterpenone catechol in the presence of Cu2+ under aqueous conditions. This was based on that if p-QM was the common reactive intermediate, the same thiol adducts should be observed regardless how p-QM was formed and if p-QM was not formed in the presence of Cu2+, different adducts should be observed. The oxidation of diterpenone catechol to p-QM was achieved in anhydrous acetonitrile-d3 by Ag2O (22, 33), and the formation was confirmed by 1H NMR analysis (Scheme 5). After the oxidation, both of the aromatic and benzylic proton signals of diterpenone catechol disappeared completely while three new proton signals were observed at chemical shifts of 6.30, 6.40, and 6.51 ppm, respectively, which were similar to those of p-QMs reported (1-7). The disappearance of the benzylic protons of diterpenone catechol was fully consistent with the p-QM formation and indicated that o-quinone was not the oxidized product by Ag2O. In contrast to natural terpene QM, diterpenone QM was only stable in anhydrous acetonitrile at a diluted concentration (7.5 mM) and red precipitation was observed upon concentration. The low stability of diterpenone QM was probably due to the lack of the m-methyl group on the p-QM structure, which may contribute enough steric hindrance to prevent polymerization. The addition of mercaptoethanol to the diterpenone QM solution resulted in the gradual disappearance of QM signals over 15 h. The HPLC analysis of this reaction solution afforded one dominant signal at a retention time of 16 min. The UV absorbance spectrum of this signal was similar to that of diterpenone catechol 1 with a λmax at 292 nm. Direct ESI-MS analysis of this HPLC fraction detected four mass signals at 699, 349, 287, and 271 m/z, which were assigned as the noncovalently associated homodimer (calcd as 699 for C38H52O8S2-H+), the molecular ion M+ (calcd as 349 for

386

Chem. Res. Toxicol., Vol. 18, No. 2, 2005

Zhou and Zuniga Scheme 6. Thiol Adducts of Diterpenone Catechol in the Presence of Cu2+

Figure 2. HPLC analysis of the thiol addition on diterpenone catechol 1 in the absence and presence of Cu2+ at room temperature. (a) Compound 1 (0.25 mM) in phosphate buffer (5.0 mM, pH 7.0) without Cu2+; (b) diterpenone QM-mercaptoethanol adduct under organic conditions; and (c) compound 1 (0.25 mM) in phosphate buffer (5.0 mM, pH 7.0) and mercaptoethanol (0.75 mM) in the presence of Cu2+ (1.0 mM).

C19H26O4S-H+), M-62, and M+-mercaptoethanol, respectively. The formation of the noncovalently associated homodimer was confirmed through the direct fragmentation of the dimeric mass ion signal by MS/MS analysis. The homodimer formation was also observed for the diterpenone catechol in the ESI-MS analysis. Thus, it was concluded that the signal at the retention of 16 min was the direct mercaptoethanol adduct of the diterpenone p-QM. The thiol addition in the Cu2+-induced oxidation under aqueous conditions was investigated similarly by HPLC and ESI-MS analysis. The oxidation was initiated by the addition of diterpenone catechol, and the progress of the reaction was monitored by HPLC analysis over 9 h (Figure 2c). The HPLC analysis showed that a new signal A formed dominantly after 1 min at a retention time of 16 min, concurrent with the disappearance of diterpenone catechol 1 (Figure 2c). This initially formed signal was then converted completely to a signal B at a retention time of 20 min over a period of 50 min (Figure 2c). The UV spectra of both signals A and B had similar absorbance curves with λmax at 292 nm and resembled that of diterpenone catechol. The retention time of signal A was identical to that of the terpene QM-thiol adduct under organic conditions (Figure 2b,c). ESI-MS analysis further confirmed the identity of this species with an identical molecular ion and fragmentation pattern as that of diterpenone p-QM thiol adduct. The signal at the retention time of 20 min gave a different mass signal and fragmentation pattern than that of the p-QM thiol adduct in ESI-MS analysis. Only three mass signals were observed at 695, 347, and 287 m/z, which were assigned as the noncovalently associated homodimer (calcd as 695 for C38H48O8S2-H+), the molecular ion M+ (calcd as 347 for C19H24O4S-H+), and M+-60, respectively. The molecular ion mass suggested that the initial p-QM thiol adduct A was further oxidized by Cu2+ under aqueous conditions and trapped intramo-

lecularly. The thiol adduct B was then isolated from a scaled-up reaction and characterized by 1H and 13C NMR analysis in MeOH-d4. The chemical shifts in both proton and carbon-13 spectrum and the integration ratio of proton signals were fully consistent with the structure of the intramolecular adduct 10, which was formed through a p-QM intermediate (Scheme 6). Thus, a second QM was formed by the oxidation of the initial QM-thiol adduct under aqueous conditions in the presence of Cu2+. DNA Damages in the Cu2+-Induced Oxidation of Diterpenone Catechol. The QM formations in Cu2+induced oxidization of diterpenone catechol 1 implied that multiple DNA lesions could result from the oxidative damage by the hydroxyl radical and nucleobase alkylation by QMs. The extent of DNA damage with diterpenone catechol 1 in the presence of Cu2+ was investigated on a short duplex DNA target selected from the P-53 mutated region (34). The target DNA duplex contains a five-nucleobase overhang at the 3′-position, which allowed us to assess the extent of lesions of the single strand vs the duplex DNA on the same target. The longer strand (30-mer) was radiolabeled with radioactive 32P and then hybridized with the complementary strand (25-mer). The Cu2+-induced oxidation was initiated by the addition of diterpenone catechol, and the resulting reaction solutions were incubated at 37 °C for 2 h. The extent of DNA modification was assessed by comparing the amount of modified radiolabeled DNA vs that of unmodified DNA, which were separated by denatured gel electrophoresis (35). As shown in Figure 3, extensive DNA cleavage was observed in both nonpiperidine- and piperidine-treated assays. DNA cleavage with no piperidine treatment was due to the direct fragmentation of the DNA backbone by the hydroxy radicals while the cleavage of DNA after piperidine treatment was the result of the oxidative damages on the nucleobases (14-16, 36). As expected, no DNA damage was detected with either Cu2+ or diterpenone catechol alone. In the presence of both Cu2+ and diterpenone catechol, the oxidative DNA damage was clearly nonsequence selective and in proportion to the concentrations of diterpenone catechol. Importantly, similar levels of DNA lesion were observed at the same concentration of diterpenone catechol with decreased concentrations of Cu2+ even at a one-third of the concentration of diterpenone catechol, which was fully consistent with the disproportionation of the Cu(II)/Cu(I) redox cycle in the oxidation process. In addition, the DNA damage at the low concentrations of Cu2+ (16.7 µM) reflected the potential toxicity of diterpenone catechol under biological conditions. The double helix DNA was found to be less susceptible toward the oxidative lesions than the single strand DNA, which was also observed on the fragmentation pattern of 3′-overhang of the duplex DNA target. The extent of direct DNA cleavage was comparable to that of the oxidative nucleobase damage

Quinone Methide Formations and Concurrent DNA Damage

Chem. Res. Toxicol., Vol. 18, No. 2, 2005 387 Scheme 7. Mechanism for Oxidation of Diterpenone Catechol in the Presence of Cu2+

Figure 3. PAGE analysis of the oxidative damage on DNA by diterpenone catechol 1 in the presence of Cu2+. Each reaction solution contained 1.0 µM oligonucleotides, phosphate buffer (5.0 mM, pH 7.0), and 10% CH3CN. The reactions were initiated by adding a solution of 1 (250 or 50 µM), and the resulting solutions were incubated at 37 °C for 2 h. The piperidine treatment was achieved by adding a solution of 10% piperidine, and the resulting mixtures were subsequently heated at 90 °C for 20 min. The concentration of Cu2+ in the lanes were as follows: 1, 0 µM; 2, 750 µM; 3, 0 µM; 4, 750 µM; 5, 250 µM; 6, 83.3 µM; 7, 150 µM; 8, 50 µM; 9, 16.7 µM; 10, 750 µM; 11, 250 µM; and 12, 83.3 µM.

revealing the nonselective nature of the hydroxyl radial damage on the DNA. In the gel analysis, no higher molecular band was observed than the original radiolabeled oligonucleotide, indicating that no DNA alkylation product was formed. One rationale is that water addition to the p-QM might outcompete the DNA alkylation. It was also possible that DNA alkylation adduct might be formed in the reaction yet was not detected or directly cleaved by the hydroxyl radicals in situ, which is currently being investigated.

Discussion The Cu2+-induced oxidation of diterpenone catechol 1 under aqueous conditions afforded only p-QM adducts through multiple nucleophilic additions while no oquinone adducts were observed. Although diterpenone p-QM is thermodynamically more stable than o-quinone by 1.92 kcal/mol based on a semiempirical calculation (AM1, 37), the isomerization of o-quinone to p-QM proceeds through a keto-enol tautomerization of which the activation energy was estimated between 10 and 30 kcal/mol for β-diketones under aqueous conditions by ab initio calculations (38). Experimentally, Bolton et al. reported that the rate of the isomerization of a homoconjugated catechol was estimated to be 1.9 × 10-3 s-1 (t1/2 ) 9 min) while the pseudo-first-order rate of thiol addition was 1.2 × 102 s-1 (26, 27). It was also found that the oxidation of catechols in the presence of thiols only afforded the o-quinone adducts, which indicated that the o-quinone was fully depleted by thiol addition and not available for isomerization to QM (26, 31). Similarly in our study, the Cu2+-induced oxidation of diterpenone catechol was investigated in the presence of thiol and

afforded only the QM adduct. In addition, the intramolecular trapping of the second oxidation species resulted in QM adduct 10 as the sole adduct. The time-dependent formation of trapped QM adducts by HPLC analysis also indicated that it was unlikely that QMs were trapped on the HPLC column during the analysis. Thus, it is implied that p-QM was formed directly in the Cu2+mediated oxidization of diterpenone catechol under aqueous conditions. This direct QM formation may be unique for diterpenone catechol in the Cu2+-induced oxidation process and is an addition to the reported isomerization pathway of o-quinones to QMs. On the basis of these results and reported mechanisms of Cu2+-induced oxidation (14-17, 39), we propose the following Cu2+-induced oxidation/addition mechanism for diterpenone catechol under aqueous conditions (Scheme 7). Similar to the Cu2+-induced oxidation of 2-hydroxyphenol, the oxidation of diterpenone catechol to p-QM involves the disproportionation of the Cu(II)/Cu(I) redox cycle and the transfer of electrons to oxygen generating hydroxyl radicals as a side product. The first nucleophilic addition of the p-QM forms an initial QM adduct, which is then oxidized to p-QM again through the Cu2+-induced oxidation. The oxidation of the initial QM adduct to another QM is also consistent with the reported oxidation of the ammonium adduct of a natural diterpene QM (40). The second nucleophilic addition of p-QM affords a stable bis-nucleophile adduct, which does not contain an oxidizable benzylic proton. This proposed mechanism suggests that in the presence of Cu2+, diterpenone catechol could cause extensive DNA lesions through multiple steps of oxidative damage by the hydroxyl radicals and potential DNA alkylations by p-QMs, which was partially demonstrated by the DNA study. Finally, further oxidation of the bis-nucleophile catechol product by Cu2+ will proceed via the reported o-quinone mechanism similar to that of 2-hydroxyphenol (14-16). In conclusion, the multiple p-QM formations from diterpenone catechol 1 in the Cu2+-induced oxidation demonstrate the potential of high toxicity on DNA by terpene QM precursors under biological conditions and imply that natural terpene QMs may exist in the form of catechol precursors in the plants prior to the isolation. The antitumor activity of these catechol precursors is currently under investigation.

Acknowledgment. We thank Virginia Commonwealth University for financial support (startup fund)

388

Chem. Res. Toxicol., Vol. 18, No. 2, 2005

and Dr. Lambert M. C. Ngoka for the help on the ESIMS analysis. Supporting Information Available: NMR spectra of compounds 1, 3, 4-8, and 10 and MS spectra of 1, 9, and 10. This material is available free of charge via Internet at http:// pubs.acs.org.

References (1) Topcu, G., Altiner, E. S., Gozcu, S., Halfon, B., Aydogmus, Z., Pezzuto, J. M., Zhou, B.-N., and Kingston, D. G. I. (2003) Studies on di- and triterpenoids from Salvia Staminea with cytotoxic activity. Planta Med. 69, 464-467. (2) Setzer, W., Holland, M., Bozeman, C. A., Rozmus, G. F., Setzer, M. C., Moriarity, D. M., Reeb, S., Vogler, B., Bates, R. B., and Haber, W. A. (2001) Isolation and frontier molecular orbital investigation of bioactive quinone-methide triterpenoids from the bark of Salcia Petenensis. Planta Med. 67, 65-69. (3) Ulubelen, A., O ¨ Kso¨z, S., Kolak, U., Bozok-Johansson, C., Celik, C., and Voelter, W. (2000) Antibacterial diterpenes from the roots of Salvia viridis. Planta Med. 66, 458-462. (4) Cha´vez, H., Este´vez-Braum, A., Ravelo, A. G., and Gonza´lez, A. G. (1999) New phenolic and quinone-methide triterpenes from Maytenus amazonia. J. Nat. Prod. 62, 434-436. (5) Alvarenga, N. L., Vela´zquez, C. A., Go´mez, R., Canela, N. J., Bazzocchi, I. L., and Ferro, E. A. (1999) A new antibiotic nortriterpene quinone methide from Maytenus Catingarum. J. Nat. Prod. 62, 750-751. (6) Gonza´lez, A. G., Alvarenga, N. L., Bazzocchi, I. L., Ravelo, A. G., and Moujir, L. (1998) A new bioactive norquinone methide triterpene from Maytenus scutioides. Planta Med. 64, 769771. (7) Kupchan, S. M., Karim, A., and Marcks, C. (1968) Taxodione and taxodone, two novel diterpenoid quinone methide tumor inhibitors from Taxodium Distichum. J. Am. Chem. Soc. 90, 5923-5924. (8) Zhou, Q., and Rokita, S. E. (2003) A general strategy for targetpromoted alkylation in biological systems. Proc. Natl. Acad. Sci. U.S.A. 26, 15452-15457. (9) Thopmson, D. C., Thopmson, J. A., Sugumaran, M., and Moldeus, P. (1992) Biological and toxical consequences of quinone methide formation. Chem.-Biol. Interact. 86, 129-162. (10) Peter, M. G. (1989) Chemical modifications of biopolymers by quinone methide. Angew. Chem., Int. Ed. Engl. 28, 555-570. (11) Moore, H. W., and Czerniak, R. (1981) Naturally occurring quinones as potential bioreductive alkylating agents. Med. Res. Rev. 1, 249-280. (12) Yang, Z., Kitano, Y., Chiba, K., Shibata, N., Kurokawa, H., Doi, Y., Arakawa, Y., and Tada, M. (2001) Synthesis of various oxidized abietane diterpenes and their antibacterial activities against MRSA and VRE. Bioorg. Med. Chem. 9, 347-356. (13) Schaller, F., Wolfender, J.-L., Hostettmann, K., and Mavi, S. (2001) New antifungal “quinone methide” diterpenes from Bobgunnia madagascariensis and study of their interconversion by LC/NMR. Helv. Chim. Acta 84, 222-229. (14) Flowers, L., Ohnishi, T., and Penning, T. M. (1997) DNA strand scission by polycylic hydrocarbon o-quinone: Role of reactive oxygen species, Cu(II)/(I) redox cycling, and o-semiquinone anion radicals. Biochemistry 36, 8640-8648. (15) Oikawa, S., Hirosawa, I., Hirakawa, K., and Kawanishi, S. (2001) Site specificity and mechanism of oxidative DNA damage induced by carcinogenic catechol. Carcinogenesis 22, 1239-1245. (16) Hirakawa, K., Oikawa, S., Hirosawa, I., Hirosu, I., and Kawanishi, S. (2002) Catechol and hydroquinone have different redox properties responsible for their differential DNA-damaging ability. Chem. Res. Toxicol. 15, 76-82. (17) Liu, X., Yao, J., Pisha, E., Yang, Y., Hua, Y., van Breeman, R. B., and Bolton, J. L. (2002) Oxidative DNA damage induced by equinine estrogen metabolites: Role of estrogen receptor R. Chem. Res. Toxicol. 15, 512-519. (18) Zhang, F., Swanson, S. M., van Breeman, R. B., Liu, X., Yang, Y., Gu, C., and Bolton, J. L. (2001) Equine estrogen metabolite 4-hydroxyequilenin induced DNA damage in the rat mammary tissues: Formation of single strand breaks, apurinic sites, stable adducts and oxidized bases. Chem. Res. Toxicol. 14, 16541659.

Zhou and Zuniga (19) Bolton, J. L., Pisha, E., Zhang, F., and Qiu, S. (1998) Role of quinoids in estrogen carcinogenesis. Chem. Res. Toxicol. 10, 11131127. (20) Lin, P.-H., Nakamura, J., Yamaguchi, S., Asakura, S., and Swenberg, J. A. (2003) Aldehydic DNA lesions induced by catechol estrogen in calf thymus DNA. Carcinogenesis 24, 1133-1141. (21) Akanni, A., Tabakvoic, K., and Abul-Hajj, Y. (1997) Estrogennucleic acid adducts: Reaction of 3,4-estrone o-quinone with nucleic acid bases. Chem. Res. Toxicol. 10, 477-481. (22) Filar, L. J., and Winstein, S. (1960) Preparation and behavior of simple quinone methides. Tetrehedron Lett. 25, 9-25. (23) Volod’kin, A. A., and Ershov, V. V. (1988) Stable methylenequinones. Russ. Chem. Rev. 57, 336-349. (24) Gru¨nanger, P. (1979) Chinomethide. In Methoden der Organisch Chemie (Muller, E., and Bayer, O., Eds.) Vol. VII/3b, pp 395521, Houben-Weyl, Stuttgart. (25) Wagner, H. U., and Gompper, R. (1974) Quinone methides. In The Chemistry of Quinonoid Compound (Patai, S., Ed.) Vol. 1, pp 1145-1178, John Wiley & Sons, New York. (26) Bolton, J. L., Acay, N. M., and Vukomanovic, V. (1994) Evidence that 3-allyl-o-quinone spontaneously rearrange to their more electrophilic quinone methides: Potential bioactivation mechanism for the hepatocarcinogen safrole. Chem. Res. Toxicol. 7, 443450. (27) Iverson, S. L., Hu, L. Q., Vukomanovic, V., and Bolton, J. L. (1995) The influence of the p-alkyl substituent on the isomerization of o-quinones to p-quinone methides: Potential bioactivation Mechanism for catechols. Chem. Res. Toxicol. 8, 537-544. (28) Tada, M., Nishiiri, S., Yang, Z., Imai, Y., Tajima, S., Okazaki, N., Kitano, Y., and Chiba, K. (2000) Synthesis of (+)- and (-)ferruginol via asymmetric cyclization of a polyene. J. Chem. Soc., Perkin Trans. 1, 2657-2664. (29) Vickery, E. H., Pahler, L. F., and Eisenbraun, E. J. (1979) Selective O-demethylation of catechol ethers. Comparison of boron tribromide and iodotrimethylsilane. J. Org. Chem. 44, 4444-4446. (30) Yao, J., Chang, M., Li, Y., Pisha, E., Liu, X., Yan, D., Elguindi, E. C., Blond, S. Y., and Bolton, J. L. (2002) Inhibition of cellular enzymes by equine catechol estrogen in human breast cancer cells: Specificity for glutathione S-transferase P1-1. Chem. Res. Toxicol. 15, 935-942. (31) Sugumaran, M., Saul, S., and Semensi, V. (1989) Trapping of transiently formed quinone methide during enzymatic conversion of N-acetyldopamine to N-acetylnorepinephrine. FEBS Lett. 252, 135-138. (32) Bolton, J. L., Turnipseed, S. B., and Thompson, J. A. (1997) Influence of quinone methide reactivity on the alkylation of thiol and amino groups in proteins: Studies utilizing amino acid and peptide models. Chem.-Biol. Interact. 107, 185-200. (33) Zhou, Q., and Turnbull, K. D. (2000) Trapping phosphodiesterquinone methide adducts through in situ lactonization J. Org. Chem. 65, 2022-2029. (34) Anderson, C. W., and Allalunis-Turner, M. J. (2000) Human TP53 from the malignant glioma-derived cell lines M059J and M059K has a cancer-associated mutation in exon 8′′. Radiat. Res. 154, 473-476. (35) Zhou, Q., Pande, P., Johnson, A. E., and Rokita, S. E. (2001) Sequence specific delivery of a quinone methide intermediate to the major groove of DNA. Bioorg. Med. Chem. 9, 2347-2354. (36) Schweigert, N., Acero, J. L., von Gunten, U., Canonica, S., Zehnder, A. J. B., and Eggen, R. I. L. (2000) DNA degradation by the mixture of copper and catechol is caused by DNA-copperhydroperoxo complexes, probably DNA-Cu(I)OOH. Environ. Sci. Mol. Mutagen. 36, 5-12. (37) Spartan (2002) Essential, Wave Function Inc., Irvine, CA. (38) Yamabe, S., Tsuchida, N., and Miyajima, K. (2004) Reaction paths of keto-enol tautomerization of β-diketones. J. Phys. Chem. A 108, 2750-2757. (39) Seike, K., Murata, M., Oikawa, S., Hiraku, Y., Hirakawa, K., and Kawanishi, S. (2003) Oxidative DNA damage induced by benz[a]anthracene metabolites via redox cycles of quinone and unique nonquinone. Chem. Res. Toxicol. 16, 1470-1476. (40) Kupchan, S. M., Karim, A., and Marcks, C. (1969) Tumor inhibitors. XLVIII. Taxodione and taxodone, two novel diterpenoid quinone methide tumor inhibitors from Taxodium Distichum. J. Org. Chem. 34, 3912-3918.

TX049703A