1594
Chem. Res. Toxicol. 2008, 21, 1594–1599
Combined Activation of Methyl Paraben by Light Irradiation and Esterase Metabolism toward Oxidative DNA Damage Yoshinori Okamoto,† Tomohiro Hayashi, Shinpei Matsunami, Koji Ueda, and Nakao Kojima* Faculty of Pharmacy, Meijo UniVersity, 150 Yagotoyama, Tempaku-ku, Nagoya 468-8503, Japan ReceiVed February 19, 2008
Methyl paraben (MP) is often used as a preservative in foods, drugs, and cosmetics because of its high reliability in safety based on the rapid excretion and nonaccumulation following administration. Light irradiation sometimes produces unexpected activity from chemicals such as MP; furthermore, there is ample opportunity for MP to be exposed to sunlight. Here, we investigated whether MP shows DNA damage after sunlight irradiation. Two major photoproducts, p-hydroxybenzoic acid (PHBA) and 3-hydroxy methyl paraben (MP-3OH), were detected after sunlight irradiation to an aqueous MP solution. Both photoproducts were inactive in the in vitro DNA damage assay that measures oxidized guanine formed in calf thymus DNA in the presence of divalent copper ion, a known mediator of oxidative DNA damage. Simulated MP metabolism using dermal tissues after light irradiation produced these two photoproducts, which reacted with a microsomal fraction (S9) of the skin. A metabolite from MP-3OH, not PHBA, caused distinct DNA damage in the in vitro assay. This active metabolite was identified as protocatechuic acid, a hydrolyzed MP-3OH product. In addition, NADH, a cellular reductant, enhanced DNA damage by approximately five times. These results suggest that reactive oxygen species generated by the redox cycle via metal ion and catechol autoxidation are participating in oxidative DNA damage. This study reveals that MP might cause skin damage involving carcinogenesis through the combined activation of sunlight irradiation and skin esterases. Introduction 1
Methyl paraben (MP) is a methyl ester of p-hydroxybenzoic acid (PHBA). Parabens are used singly or in combination to exert the intended antimicrobial effect against molds and yeasts as well as bacteria (1). The cytotoxicity of MP is considered to be due to the inhibitory effects on membrane transport and mitochondrial functions (2). Parabens meet several criteria as an ideal preservative; that is, they have broad spectrum antimicrobial activity and high reliability in safety for use in addition to excellent chemical stability over practical pH and temperature (3). The antimicrobial activity of parabens increases as the chain length of the ester group increases. However, MP use is actually based on the short chain esters, since water solubility decreases with increasing chain length. MP has been used as a preservative in foods, drugs, and cosmetics for over 50 years. The metabolic fate of MP has been revealed using experimental animals. MP is hydrolyzed to PHBA, conjugated, and rapidly excreted in the urine. Almost 86% of MP was excreted within 24 h after oral administration of a single dose (0.8 g/kg) to rabbits (4). No evidence of accumulation of MP and its metabolites was reported. Acute toxicity studies indicate that MP is practically nontoxic irrespective of the various routes of administration (5). Genotoxicity studies such as the Ames test * To whom correspondence should be addressed. Tel: +81-52-839-2676. Fax: +81-52-834-8090. E-mail:
[email protected]. † Present address: Laboratory of Chemical Biology, Department of Pharmacological Sciences, State University of New York at Stony Brook, Stony Brook, New York 11794-8651. 1 Abbreviations: MP, methyl paraben; PHBA, p-hydroxybenzoic acid; 8-oxodG, 8-oxo-7,8-dihydro-2′-deoxyguanosine; PC, protocatechuic acid; BSTFA, bis(trimethylsilyl)trifluoroacetamide; dG, 2′-deoxyguanosine; DTPA, diethylenetriaminepentaacetic acid; MP-3OH, 3-hydroxy methyl paraben.
Figure 1. MP photoproducts. (A) PHBA and MP-3OH were eluted at 7.0 and 11.5 min, respectively. (B) An aqueous solution of MP was irradiated with UV or sunlight, as described in the Experimental Procedures. The sunlight irradiation figure is a representative finding. MP was eluted as a large peak at a 22.0 min retention time, although it has been omitted from this figure.
provided mostly negative results, with the exception of several chromosome tests involving Chinese hamster cells (6). No in vivo evidence of MP carcinogenicity has been noted (7). Overall, the genotoxicity of MP is predominantly negative. When parabens are used in cosmetics and sunscreens, photochemical decomposition appears to be one of the important clearance routes along with metabolism in dermal tissues (8).
10.1021/tx800066u CCC: $40.75 2008 American Chemical Society Published on Web 07/26/2008
OxidatiVe DNA Damage Induced by Methyl Paraben
Chem. Res. Toxicol., Vol. 21, No. 8, 2008 1595
Figure 2. UV spectra of MP photoproducts. UV spectra of p-1 and p-2 were analyzed by a diode array detector equipped with HPLC and overlapped those of PHBA and MP-3OH, respectively.
Figure 3. GC/MS spectra of MP photoproducts. p-1 and p-2 were derivatized with BSTFA and analyzed by GC/MS, as described in the Experimental Procedures.
In spite of the extensive investigation on metabolic fates of parabens, photoproducts formed by sunlight and UV exposure have not been determined. We reported previously that several compounds listed below produce adverse effects such as endocrine disruption or DNA damage due to environmental sunlight exposure. Benzophenone, a basal compound of sunscreens, protects skin from UV, but its hydroxylated photoproducts exert estrogenic activity (9). Phthalate esters, a controversial environmental endocrine disruptor, acquire the unequivocal endocrine-disrupting activity by sunlight irradiation (10). Ethylbenzene, an additive of fuel or a putative causal chemical of sick building syndrome, induces oxidative DNA damage after sunlight exposure (11). Nonylphenol, an environmental endocrine disruptor, exerts DNA-damaging activity under sunlight (12). Recently, MP was reported to potentiate UVinduced damage of keratinocytes (13). This indicates the involvement of reactive oxygen species leading to apoptosis via an increase of transcription factors such as NF-κB, although the origin of radicals has not yet been elucidated. We hypothesized that certain MP photoproducts might be responsible for this free radical generation. In this report, we discuss whether MP shows DNA-damaging activity after sunlight exposure and metabolism using skin S9 fraction. DNA-damaging activity was
determined as the in vitro formation of 8-oxo-7,8-dihydro-2′deoxyguanosine (8-oxodG) in calf thymus DNA. Our results suggest that MP is activated via a novel combined activation pathway by sunlight and skin enzymes.
Experimental Procedures Chemicals. MP (99.0%), PHBA, and protocatechuic acid (PC) were purchased from Wako Pure Chemical (Osaka, Japan). Bis(trimethylsilyl)trifluoroacetamide (BSTFA), 2′-deoxyguanosine (dG), 8-oxodG, and calf thymus DNA were from Sigma Chemical (St. Louis, MO). Diethylenetriaminepentaacetic acid (DTPA) was from Dojindo (Kumamoto, Japan). Calf intestinal alkaline phosphatase was from Roche Diagnostics (Mannheim, Germany). Nuclease P1 was from Yamasa Shoyu (Choshi, Chiba, Japan). 3-Hydroxy methyl paraben (MP-3OH; synonym, methyl protocatechuate) was synthesized in our laboratory from PC and methanol according to the published procedures (14). The structure of MP-3OH was confirmed by UV, 1H/13C NMR, and MS spectra. Light Irradiation. MP aqueous solution (10 mM) was light irradiated in a quartz cuvette sealed with a Teflon stopper (GL Science, Tokyo, Japan). UV was from an Hg arc (254 nm, 0.194
1596
Chem. Res. Toxicol., Vol. 21, No. 8, 2008
Okamoto et al.
Figure 4. Metabolites of MP photoproducts and the DNA-damaging activities. (A and D) HPLC chromatograms of S9-treated PHBA and MP-3OH are shown. (B and E) 8-oxodG formation activity of each fraction of PHBA and MP-3OH metabolites are indicated. 8-OxodG formation activity is shown as a molar ratio of 8-oxodG vs dG in calf thymus DNA. (C) PC was eluted at 4.9 min.
J/min) equipped with UV cross-linker (DNA-FIX, Atto Corp., Tokyo, Japan) for 2 h. Sunlight irradiation was carried out by setting a sample cuvette in a bright location for 8 h from 9 a.m. to 5 p.m. on October 6, 2005 (total amount of global solar radiation was approximately 11.7 mJ/m2 according to the Japan Meteorological Agency). Light-irradiated aliquots were dried and dissolved in methanol for HPLC analysis. Preparation of Skin S9 and Metabolic Reaction. Fiveweek-old male Sprague-Dawley rats (Clea Japan, Tokyo, Japan) were shaved on the dorsal surface and abdominal areas, and then, skin was obtained for S9 according to the published method (15). The protein concentration was measured with the Bradford method using a reagent from Biorad (Hercules, CA). Absorbance at 595 nm was measured by a microplate reader (Biorad). Standard curves were obtained from bovine serum albumin. The protein concentration of the obtained S9 was 11.7 mg protein/mL. PHBA or MP-3OH (4% DMSO solution, 1.2 mM) was used as a substrate solution. A reaction mixture containing 750 µM substrate and 1.2 mg of skin S9 in 400 µL of 100 mM phosphate buffer (pH 7.4) was incubated for 1 h at 37 °C. After 20 µL of 4 M HCl was added, metabolites were extracted with 1.5 mL of ethyl acetate. The ethyl acetate layer was dried and dissolved in methanol before HPLC analysis. HPLC Analysis. Photoproducts or metabolites were subjected to a Cosmosil ODS column maintained at 40 °C and eluted at a flow rate 1.0 mL/min with 30% methanol containing 0.1% trifluoroacetic acid. HPLC was performed on an LC-VP system (Shimadzu, Kyoto, Japan) equipped with diode array detector (SPD-M10A VP), system controller (SCL-10A VP), and ClassVP workstation. To prepare samples for both GC/MS analysis and 8-oxodG formation, photoproducts or metabolites were fractionated every 30 s with the HPLC. The aliquots were dried and stored at -80 °C until use.
GC/MS Analysis. Photoproducts and metabolites were derivatized with BSTFA for 60 min at room temperature. GC/ MS analysis was performed on a gas chromatograph (HP 6890 GC System Plus, Agilent Technologies, Palo Alto, CA) equipped with a mass spectrometer (JMS-700 MStation, JEOL, Tokyo, Japan) using electron impact ionization at 70 eV. Helium was used as a carrier gas at a flow rate of 1 mL/min. The temperature of the injector, interface, and ion source was maintained at 250 °C. The temperature program for the DB-1 column (0.25 mm i.d. × 30 m, film thickness 0.25 µm, J&W Scientific, Folsom, CA) was as follows: 70 (2 min isothermal), 70-280 (10 °C/ min), and 280 °C (2 min isothermal). Measurement of 8-OxodG Formation in Calf Thymus DNA. HPLC fractions and PC were dissolved in DMSO and used for determining 8-oxodG formation. Test solution, 100 µM calf thymus DNA (calculated by 6600 M-1 cm-1 at 258 nm), 20 µM CuCl2, and 100 µM NADH (if indicated) were mixed in 4 mM phosphate buffer (pH 7.8) containing 5 µM DTPA and incubated for 1 h at 37 °C. After the addition of DTPA up to 100 µM and ethanol precipitation, DNA was digested into nucleosides with nuclease P1 (10 U) and calf intestinal alkaline phosphatase (1.3 U) in the appropriate solutions. Amounts of dG and 8-oxodG were measured using an HPLC equipped with UV and electrochemical detectors (Coulochem II, ESA, Chelmsford, MA) as previously reported (11). HPLC conditions were as follows: column, ODS-80Ts 150 mm (Tosoh, Tokyo, Japan); column temperature, 25 °C; mobile phase, 10 mM NaH2PO4 containing 8% methanol; flow rate, 1 mL/min; and detection wavelength for dG, 254 nm. Electrochemical detector conditions: guard cell, 400 mV; channel 1, E 150 mV/R 100 µA/filter 2/output 1 V; and channel 2, E 300 mV/R 200 nA/filter 10/ output 1 V.
OxidatiVe DNA Damage Induced by Methyl Paraben
Chem. Res. Toxicol., Vol. 21, No. 8, 2008 1597
Figure 5. UV and GC/MS spectra of DNA-damaging metabolites of MP photoproducts. (A) UV spectra of the active metabolite (m-1) and PC. (B) GC/MS spectrum of m-1.
Figure 6. DNA damage by active MP photoproduct metabolites. The formation of 8-oxodG induced by various amounts of PC, an MP photoproduct metabolite, was measured as described in the Experimental Procedures. The 8-oxodG formation activity is shown as a molar ratio of 8-oxodG vs dG in calf thymus DNA.
Results Identification of MP Photoproducts. Aqueous MP solution was exposed to sunlight as described in the Experimental Procedures. HPLC analysis of the irradiated MP solution detected two major peaks (designated as p-1 and p-2 in the order of elution) at earlier retention times than MP (20.1 min) (Figure 1B). Approximately the same profile was obtained when artificial UV (254 nm) was used. Peaks p-1 and p-2 were eluted at 7.0 and 11.4 min, respectively. Maximum absorptions of UV spectra were observed at 255 nm for p-1 and at 260 and 295 nm for p-2 (Figure 2). In GC/MS analysis, these photoproducts were trimethylsilylated by BSTFA to achieve good volatility and separation (Figure 3). A molecular ion of p-1-TMS was observed at m/z 282 with fragment ions at m/z 267 (base peak), 223, and 193. A molecular ion of p-2-TMS was also observed at m/z 312 with a fragment ion at m/z 193 (base peak). The p-1 and p-2 data were consistent with those of PHBA and MP3OH, respectively (Figures 1–3). Additional 1H NMR performance supported the above structure identification.
Identification of Metabolite from MP Photoproducts and 8-OxodG Formation Activity. MP photoproducts, PHBA and MP-3OH, reacted with S9 fraction from rat skin, assuming a metabolic reaction in the dermal tissues after light irradiation. One metabolite (designated as m-1) was detected from MP-3OH at a retention time of 5.0 min, whereas no metabolite was observed from PHBA (Figure 4). While MP, PHBA, and MP-3OH were negative in 8-oxodG formation, the fraction containing m-1 showed distinct activity. Peak m-1 showed maximum absorptions of UV spectrum at 260 and 295 nm (Figure 5). The GC/MS spectrum of m-1-TMS showed molecular ion at m/z 370 with fragment ions at m/z 355, 311, 281, 193 (base peak), and 73. Both UV and MS spectra of m-1 were consistent with those of PC (Figures 4 and 5), indicating that a DNA-damaging metabolite was formed by certain esterase(s) contained in the skin S9. The 8-oxodG formation activity of PC increased in a dose-dependent manner and was enhanced approximately five times by the addition of NADH, a cellular reductant (Figure 6).
Discussion When parabens are used in cosmetics and sunscreens, photochemical decomposition appears to be one of the important clearance routes along with dermal tissue metabolism (8). In the present study, MP photoproducts and metabolites were characterized, and their DNA-damaging activities were evaluated based on the formation of 8-oxodG in calf thymus DNA. The present study has demonstrated that MP is converted to DNAdamaging compounds by the combined activation with sunlight irradiation and skin esterase metabolism. This activation occurs with the use of MP-containing products such as cosmetics and sunscreens, because the source of light used in the experiment is natural sunlight, and the concentration of cosmetic MP (
