286 160 140 120 (1) ..... I 100 a.. - � _g � C 80 - 0 0 u 2 - C 0 60 2 ':::R. C � 0 40 0 20 0 JOURNAL OF COSMETIC SCIENCE Non-UVA UVA tRA 3 6 6 15 30 Cone. of eomp.1 Cone. of eomp.2 (µM) (µM) 100 80 60 40 20 0 J:I - � 0 .c C -� 0 u - = 0 Q) ':::R. 0 0 - Figure 3. Effects of compounds 1 and 2 isolated from M. heterophylla on the production of MMP-1 in UVA-irradiated human dermal fibroblasts. The cells were treated with various concentrations of these compounds for 24 h. aMMP-1, matrix metalloproteinase-1 (■). The MMP-1 contents in culture media were determined by ELISA as detailed under Materials and Methods. bCytotoxicity was measured by MTT assay (0). The results are expressed as the average of triplicate experiments with SD. *p 0.05 compared with UVA-treated control cell. UVA dosage was 6.3 J/cm2 tRA refers to 4.0 µM of all-trans retinoic acid. We further investigated the inhibitory effects of these compounds on MMP-1 expression in UVA-irradiated HDFs (6.3 J/cm2). HDF were treated with 1.0, 3.0, 6.0, 15 .0, and 30.0 µM for 24 h, and then the MMP-1 contents in the culture media were determined by ELISA. The treatment of UVA-irradiated HDFs with these compounds significantly suppressed MMP-1 production at the protein level in a dose-dependent manner, by an average of 24.5 ± 5.0% (p 0.05) at 1.0 µM, 38.8 ± 4.6% (p 0.05) at 3.0 µM, and 56.4 ± 7.5% (p 0.05) at 6.0 µM of compound 1, and an average of 30.3 ± 10.8% (p 0.05) at 6.0 µM, 43.8 ± 14.7% (p 0.05) at 15.0 µM, and 64.1 ± 7.5% (p 0.05) at 30.0 µM of compound 2, compared with UV A-treated control cells, respectively (Figure 3). Interestingly, the inhibitory effect of compound 1 on MMP-1 production at the protein level was slightly higher than that of, or appeared to be similar to that of, all-trans retinoic acid (tRA, 33.0 ± 5.0% at 4.0 µM), which is well known as an inhibitor of UV A-induced MMPs. Fisher et al. (7) have reported that tRA, which transrepresses transcription factor (AP)-1, substantially reduces AP-1 and MMPs induction in human skin. Ramos et al. (26) have reported that UV A irradiation results in a rapid transcriptional up-regulation of two members of the MMP family, MMP-1 and MMP-10. Therefore, in this study, we investigated in UV A-irradiated HDFs whether these compounds modu­ late the expression of MMP-1 steady-state mRNA levels. As shown in Figure 4, these compounds did not suppress MMP-1 production at the mRNA level, whereas tRA significantly reduced production. These results suggest that these compounds suppress MMP-1 production at the protein level, not at the mRNA level.

M. HETEROPHYLLA AND INHIBITION OF MMP-1 287 (A) 1 2 3 4 5 MMP-1 237 bp �-actin ------- 248 bp (B) 140 * 120 Q) 100 � - 80 0 z C: 0:: 0 60 E CJ - Q) 0 � 40 � Q) 0:: 20 0 Non-UVA UVA tRA Comp. 1 Comp. 2 Figure 4. Effects of compounds 1 and 2 isolated from M. heterophylla on MMP-1 mRNA level in UVA­ irradiated human dermal fibroblasts. (A) After irradiation by UVA, HDFs were treated with various concentrations of these compounds for 12 h. Total RNA extracted from HDFs was analyzed by RT-PCR, and each band, which is quantirated by a densitometer using 13-actin transcripts as references, in (A) corresponds to each bar in (B). Lane 1: UV A non-irradiated. Lane 2: UV A (6.3 J/cm2 ) irradiated. Lane 3: UV A + 4.0 µM of tRA. Lane 4: UV A + 6.0 µM of compound 1. Lane 5: UV A + 30.0 µM of compound 2. CONCLUSIONS Two compounds isolated from the roots of M. heterophylla exhibited significant DPPH and superoxide anion radical scavenging activities. Also, these compounds significantly reduced the activity and expression of MMP-1 at the protein level in a dose-dependent manner. Therefore, we suggest that these compounds may be potential candidates to protect the skin from the aging effects of UV stimulation. REFERENCES (1) M. Wlaschek, I. Tantcheva-Poor, L. Naderi, W. Ma, L.A. Schneider, Z. Razi-Wolf, J. Schuller, and K. Scharffetter-Kochanek, Solar UV irradiation and dermal photoaging, J. Photochem. Photobiol., 63, 41-51 (2001).