20 JOURNAL OF COSMETIC SCIENCE dant, in a cuvette, and irradiating it on ice with a Sylvania 150 W slide projector spot tungsten filament two inches from the lens for 60 minutes. Oxidation products in the liposomes were assayed using the K-Assay™ LPO-CC lipid peroxide kit from Kamiya Biochemicals (Seattle, WA). TNF-a EXPRESSION OF FIBROBLASTS AFTER UVB IRRADIATION The fibroblast line XPTNF2 is an SV 40-transformed fibroblast line deficient in DNA repair and carrying the TNF-a promoter chloramphenicol acetyltransferase (CAT) re­ porter gene (25). The assay of TNF-a promoter activity using fluorescent chloramphen­ icol was as described. The assays were in duplicate, and the background was subtracted and the results averaged. MMP-1 OF HUMAN FIBROBLASTS AFTER UV A IRRADIATION MMP-1 activity. Fibroblasts were seeded at a density of 2 x 104 cells per well in a 96-well plate. After one day, cells were exposed to 20 mJ/cm2 UVA in Hank's buffer (Ca, Mg free) containing various concentrations of EGT. Following UVA irradiation, cells were cultured with DMEM-supplemented 0.1 % BSA for 24 h, and the culture medium was assayed for MMP-1 activity. MMP-1 is secreted into the medium in inactive from, and so it is necessary to convert it to the active form using a proteinase such as trypsin. The medium (0.05 ml) was collected, and 0.01 ml of trypsin was added, followed by 15-min incubation at 3 7 ° C. The resulting solution was incubated with FITC-labeled collagen type I, which is a substrate for MMP-1, for 2 h at 37°C. A 40-mM o-phenanthrolin was added and incubated for 30 min at 37°C. The degraded collagen was extracted with 0.05 ml of 0.17 M Tris-HCI (pH 9.5) containing 70% EtOH, and centrifuged for 15 min at 12000 rpm. The MMP-1 activity was estimated by measuring the fluorescence intensity of the FITC-peptide cleaved by MMP-1 (excitation: 495 nm emission: 520 nm). MMP-1 activity was expressed as type I collagen degraded per minute (unit/min/µg collagen). RT-PCR. Fibroblasts were seeded at a density of 1 x 108 cells per 35-cm dish. After 1-day culture, cells were applied to Hank's buffer (Ca, Mg free) containing various concentrations of EGT and then exposed to 20 mJ/cm2 UV A light (Toshiba FL-20S BLB). Following UVA irradiation, cells were cultured in DMEM-supplemented 0.1 % BSA for 24 h, and then cells were submitted for the determination of the MMP-1 mRN A expression level. Total RNA from the cells was extracted with TRizol reagent according to the manu­ facturer's protocol. The first-strand cDNA synthesis containing 1.5 µg of total RNA was primed with oligo(dT) and M-MLV reverse transcriptase. Primers used for PCR ampli­ fication of MMP-1 and the glycerolaldehyde-3-phosphate dehydrogenase (G3PDH) housekeeping gene were designed as follows: for MMP-1 forward primer, 5'-CGACTCTAGAAACACAAGAGCAAGA-3' and reverse primer, 5'-AAGGTT­ AGCTTACTGTCACACGCTT-3' for G3PDH forward primer, 5 '-ACCAC­ AGTCCATGCCATCAC-3' and reverse primer, 5'-TCCACCACCCTGTTGCTGTA-3'. The cycling conditions in the first-strand cDNA system consisted of denaturation at 94°C for 1 min, annealing at 60°C for 2 min, and extension at 72°C for 2 min. PCR was

ANTIOXIDANT ACTIVITY OF EGT 21 carried out for 3 5 cycles. PCR products were analyzed by 1 % agarose gel electrophoresis and visualized by ethidium bromide staining under ultraviolet light. RESULTS SCAVENGING ABILITY AGAINST •0 2 - The scavenging ability of EGT against •0 2 - was evaluated using the hypoxanthine and xanthine oxidase system as a source of •02 - . EGT showed scavenging against •02 - in a dose-dependent manner in the micromolar range (Figure 2). In addition, we examined the effects of EGT on lipid peroxidation (LPO) of liposomes initiated by •02 - generated by alloxan. The basal level of LPO in the control liposome was 17 .3 7 nmol/ml, and the addition of alloxan to the system was increased to 50.31 nmol/ml. EGT (20 µM) reduced LPO to 22.12 nmol/ml, an 85% reduction, and exhibited superior effects among other sulfur-containing antioxidants that were tested at the same concentration (Figure 3 ). QUENCHING ACTIVITY AGAINST 1 02 The quenching activity of EGT was measured by using the ESR spin-trapping method and lipid peroxidation (LPO) initiated by 1 0 2 . In general, hematoporphyrin produces 1 0 2 during UV A irradiation. As a source of 1 0 2 , the hematoprophyrin and UV A system was used. The ESR spectrum of the 1 0 2 is shown in Figure 4. The addition of EGT showed a decrease of 1 Orderived TEMP radicals in a dose-dependent manner. These results indicated that EGT effectively quenched 1 0 2 . The results for LPO initiated by 30 .-. 25 � en 20 C en 15 C 10 UJ 5 0 0 20 40 60 Cone. of EGT (µM) Figure 2. Scavenging effect of EGT against superoxide anion generated by hypoxanthine-xanthine oxidase system. The OD56onrn in the absence of EGT was set as 0% scavenging. The scavenging percent was calculated as the reduction in OD divided by the starting OD, x 100.