M. HETEROPHYLLA AND INHIBITION OF MMP-1 281 (40:60, 50:50, 60:40, 80:20, 100:0, v/v) to give 20 fractions (1-20). Compound 1 (370 mg) was obtained by recrystallization in EtOAc from fraction 15. Fraction 12 was further separated by a Sephadex LH-20 CC (35 x 2.3 cm) eluted with hexane-EtOAc­ MeOH (7:3:1, 5:5:2, 5:5:3, 4:6:4) to yield nine subfractions (I-IX). Compound 2 (105 mg) was obtained from subfraction VIII. 1,2,4,6-tetra-O-galloyl-13-0-glucopyranose (1): Amorphous pale brown powder FeC13 positive FAB-MS m/z: 789 [M+ +1} UV (MeOH) "-max: 217, 277 nm 1 H-NMR (500 MHz, acetone-d6) 3: 7.32, 7.30, 7.28, 7.26 (each 2H, s, H-2', 6'), 6.10 (lH, d,]=9.3 Hz, H-1), 5.45-5.52 (2H, m, H-2, 4), 4.62-4.67 (lH, m, H-6), 4.41-4.47 (3H, m, H-3, 5, 6) 13C-NMR (125 MHz, acetone-d 6 ) 3: 167.32 (lC), 167.13 (2C), 166.94 (lC) of galloyl carbonyl carbon atoms, 146.27-146.37 (8C, C-3', 5'), 139.29, 139.15, 139.07, 139.00 (each lC, C-4'), 122.34, 122.18, 121.95, 121.77 (each lC, C-1'), 110.26-110.42 (8C, C-2', 6'), 102.93 (lC, C-1), 74.02, 73.80, 73.30, 71.68 (each lC, C-2, 3, 4, 5), 63.08 (lC, C-6). 3,4,5-trihydroxybenzoic acid (2:) Amorphous white powder FeC1 3 positive FAB-MS m/z: 171 [M+ + 1} UV (MeOH) "-max: 220, 272 nm 1 H-NMR (500 MHz, acetone-d6) 3: 7.15 (2H, s, H-2, 6) 13C-NMR (125 MHz, acetone-d6) 3: 167.86 (lC, carbonyl carbon atom), 146.07 (2C, C-3, 5), 138.72 (lC, C-4), 122.13 (lC, C-1), 110.31 (2C, C-2, 6). MEASUREMENT OF DPPH RADICAL SCAVENGING ACTIVITY The DPPH radical scavenging effect was evaluated according to the method of Hatano et al. (9) with minor modification. DPPH solution (0.1 mM) was added to the same volume of sample solution and allowed to react for 10 min at room temperature. The optical density was measured at 565 nm using an automated microplate reader (ELx800, Bio-Tek Instruments, USA). MEASUREMENT OF SUPEROXIDE RADICAL SCAVENGING ACTIVITY The scavenging activity on the ROS was measured by monitoring the reduction of nitroblue terazolium (NBT) (10). Briefly, the samples, 0.05 M Na2CO 3 buffer (pH 10.2), 3 mM xanthine, 3 mM ethylene diamine tetraacetic acid (EDTA), 0.75 mM NBT, and bovine serum albumin (BSA) solution were mixed, and the reactant was incubated at 25°C for 10 min. Xanthine oxidase (0.25 U/ml) enzyme solution was then added, and further incubation was conducted at 25°C for 25 min. The reaction was quenched with 6 mM CuC1 2 . The scavenging activity was calculated by comparing the optical density at 565 nm of the control with that of the samples. COLLAGENASE (MMP-1) INHIBITION ASSAY The in vitro collagenase inhibition assay, which is based upon fluorescence measurement of collagen fragments upon cleavage by MMP-1, was performed using EnzChek colla­ genase/ gelatinase kits (Molecular Probes Inc., USA) according to the supplier's instruc­ tion. The enzymes were mixed with quenched fluorescent substrates (250 µg/ml) in a final volume of 200-µl reaction buffer in 96-well microplates. Digested products from

282 JOURNAL OF COSMETIC SCIENCE DQ collagen substrates had an absorption maximum at -495 nm and a fluorescence emission maximum at -515 nm in a LS55 fluorescence microplate reader (Perkin Elmer, USA). For all the MMPs tested, the activities under these conditions were linear for at least 15 min. Each time point was corrected for background fluorescence by subtracting the values derived from the non-enzyme control. CULTURE OF HUMAN DERMAL FIBROBLASTS Human dermal fibroblasts (HDFs), isolated from human neonatal foreskin, were pur­ chased from Modern Tissue Technologies Inc. (Korea). HDF were maintained in Dul­ becco's modified Eagle's medium/Ham's F-12 nutrient mixture (DMEM/F-12 3:1, v/v, Sigma) supplemented with 10% heat-inactivated fetal bovine serum (FBS), penicillin (100 IU/ml), and streptomycin (100 µg/ml). HDFs were grown at 37°C in a CO2 incubator. Fibroblast cultures were sub-cultured by trypsinization and used between the sixth and tenth passages. UV A IRRADIATION HDFs (1.5 x 105 cells/ml) were seeded into 35:p plates (Corning Inc., USA) and cultured overnight. Prior to irradiation, when cells were subconfluent, they were washed twice with phosphate-buffered saline (PBS). UVA simulator (F15T8.BLB, Sankyo Denki, Japan), filtered for the emission of UVA (320-400 nm), was used at a tube-to-target distance of 15 cm. The dose of UV A radiation, determined with a UV radiometer (International Light Inc., USA) was set at 6.3 J/cm2 • During irradiation, control cells were treated identically, except for exposure to UV light. After irradiation, fresh serum­ free medium, with or without samples at different concentrations, was added to cells at 3 7°C for 24 h. RNA ISOLATION AND RT-PCR RNA was extracted using an RNeasy Mini Kit (Qiagen, Germany) according to the supplier's instructions. First, a reverse-transcriptase polymerase chain reaction (RT-PCR) was performed to synthesize cDNA using an Omniscript RT-PCR Kit (Qiagen, Ger­ many) according to the manufacturer's instructions. PCR was then performed with each cDNA of MMP-1, �-actin fragments, primers, and Tag DNA polymerase. The primers used were as follows: MMP-1: sense 5' -AAAGGGAATAAGTACTGGGC-3', antisense 5'-AATTCCAGGAAAGTCATGTG-3' �-actin: sense 5'-ATGCAGAAGGAGAT­ CACTGC-3 ', antisense 5 '-CTGCGCAAGTTAGGTTTTGT-3'. The primer sets yielded PCR products of 237 and 248 bp for MMP-1 and �-actin, respectively. Reac­ tions were carried out in an automatic heat-block DNA thermal cycler (ASTEC PC801, ASTEC Inc, Japan). Denaturation, annealing, and elongation were carried out at 94°C, 50°C, and 72°C for 30 s, 30 s, and 60 s, respectively, for 25 cycles. Electrophoresis of the PCR products was performed on a 1.5% agarose gel in TAE (40 mM Tris acetate, 1 mM EDTA) containing 1 mg/ml ethidium bromide. The level of each mRNA gene expression was expressed as the ratio of the intensity of each PCR gene product to the corresponding �-actin PCR product as a reference molecule for the measuring of mRNA stability and normalized to the control sample.