18 JOURNAL OF COSMETIC SCIENCE UV A exposure to dermal fibroblasts leads to the reduction of collagen synthesis (8) and the excess elevation of matrix metalloproteinase-1 (MMP-1)/interstitial collagenase (9). MMP-1 is a member of the MMPs, a superfamily of endopeptidase that is capable of degrading extracellular matrix components (10). Excess expression of MMP-1 by skin fibroblasts causes subsequent damage of dermal connective tissue. The imbalance be­ tween the synthesis and degradation of collagen critically contributes to the process of matrix alteration (11) and leads to photoaging. UVB causes acute damage in the skin, such as DNA damage and apoptosis of keratino­ cytes, even in dermal cells (12). In addition, UVB induces the production of cytokines, hormones and chemical messengers, IL-1, TNF-a, propiomelanocortin-derived hor­ mones, and prostaglandin E2, which consequently leads to erythema and inflammation in the dermis (13). UVB radiation creates superoxide anion (0 2 -) (type I photosenitization) due to reaction with water, activation of mitochondrial function, and release of peroxides by inflamma­ tory cells (14), while UV A radiation generates singlet oxygen (10 2 ) (type II photosen­ sitization) through photosensitization reactions with several intracellular chromophors, such as NADH, NADPH, and flavine protein (15). It has been reported that 1 0 2 generated by UV A mediates the induction of MMP-1 through the pathway of IL-6 and IL-1 (16,17). Furthermore, a much higher level of oxidative protein was observed at the papillary dermis of photoaged skin compared to younger skin (18). This evidence indicates that the process of skin photoaging is in part mediated by oxidative stress, including reactive oxygen species (ROS). Therefore, an antioxidant possessing a wide spectrum of ROS scavenging should prevent UV-induced skin damage. Ergothioneine (EGT) is a natural antioxidant, and an amino acid not incorporated into protein, whose sulfur is predominantly in the thione form (Figure 1). EGT is a fungal metabolite that cannot be endogenously synthesized by mammals and must be taken up in the diet (19). It is found in many mammalian tissues in millimolar quantities (19). EGT is generally regarded as an antioxidant, although results are conflicting. Some regard it as a scavenger of hydrogen peroxide (20), while others contend that it does not readily react with hydrogen peroxide but does scavenge hydroxyl radicals (21). Also, some data indicate that EGT quenched 1 02 by monitoring 1270-nm phosphorescence derived from 1 02 (22). In this study, we examined the scavenging abilities of EGT against •02 - and 1 0 2 using chemical and biological systems to identify antioxidative characters. In addition, the effects of EGT on UV-induced cellular responses such as expression of both TNF-a and MMP-1 were evaluated. Figure l. Chemical structure of L-ergothioneine (EGT).

ANTIOXIDANT ACTIVITY OF EGT 19 EXPERIMENT PROCEDURES REAGENT AND CELL CULTURE Dulbecco's modified Eagles medium (DMEM) was purchased from Nikken Bio Medical Laboratory (Kyoto, Japan). Fetal calf serum (FCS), TRizol reagent, and the Super Script first-strand system for RT-PCR were purchased from Invitrogen (Carlsbad, California). Alloxan, hematoporphyrin, hypoxanthine, nitroblue tetrazolium, rose bengal, tyrpsin, tyrpsin inhibitor, 2,2,6,6-tetramaethyl-piperidone hydrochloride (TMPD), and xanthine oxidase, were purchased from Sigma-Aldrich Chemical Company (St. Louis, MO). Takara Tag was purchased from Takara shuzou (Shiga, Japan). FITC-labeled collagen type I was purchased from Yagai Co. Ltd. (Yamagata, Japan). All other chemicals used were of the purest grade commercially available. The normal human fibroblasts were purchased from Kurabo Industries Ltd. (Osaka, Japan), and these cells were maintained with DMEM with 5% FCS. Cells were grown in a humidified incubator at 37°C under a 5% CO 2 atmosphere. SCAVENGING ABILITY AGAINST •02 - Superoxide was generated using a hypoxanthine/xanthine oxidase system. EGT or a blank was added to 1 ml of a superoxide buffer consisting of 50 mM KH2PO4 (pH 7.4), 1 mM NaEDTA, 1 mM hypoxanthine, and 100 mM nitroblue tetrazolium. Xanthine oxidase (0.33 units) was added, the sample was held at 25°C, and the OD 5600m was read at 10 min. The scavenging ability was calculated as the percent difference between the antioxidant sample and control. Alloxan was also used to generate superoxide. Alloxan forms a redox cycle with its reduced form, dialuric acid, and generates superoxides (23). It was prepared fresh at 100 mM in phosphate-buffered saline with 1 % sodium citrate, diluted, and used immedi­ ately. Phosphatidylcholine liposomes in phosphate-buffered saline with 1 % sodium citrate were mixed with the 50-mM alloxan solution in the presence of various antioxi­ dants, all at 20 µM. Oxidation products in the liposomes were assayed after 60 minutes using the K-Assay LPO-CC lipid peroxide kit from Kamiya Biochemicals (Seattle, WA). QUENCHING ACTIVITY AGAINST 1 02 The 1 0 2 quenching activity of EGT was estimated both by ESR-spin trapping (24) and lipid peroxidation using liposomes. TMPD was used as a 1 O 2 -trapping reagent. Various concentrations of the compounds were added to a solution containing 0.05 mM hema­ toporphyrin and 50 mM TMPD in 100 mM Tris-HCI buffer (pH 8.0). The resulting solutions were transferred into an ESR quartz flat cell and set in the cavity of the ESR. After UV A irradiation (0.65 J/cm2 , 1 Kw Xenon lamp USHIO Inc.) filtered with a YV-3 3 filter (Toshiba Glass Co.), the ESR spectra were recorded with a field modulation frequency of 100 kHz and a modulation amplitude of 0.1 mT at an output power of 5 mW. Mn2 + doped in MnO was used as a standard. All experiments were carried out at 21 °C. Singlet oxygen was also generated by mixing phosphatidylcholine liposomes in a 10-mM sodium phosphate buffer (pH 7.4) with 10 µM rose bengal, with and without antioxi-