366 JOURNAL OF COSMETIC SCIENCE Figure 4. Stability of kojic acid in the aqueous base held at room temperature. Untreated tissues (A) were treated with vehicle control and kojic acid in the aqueous base held at room temperature for one week (B) and one month (C). Tyrosinase activity was detected in melanocytes as described in Materials and Methods. method to measure kojic acid in different products. Surprisingly, HPLC analysis showed that kojic acid in an aqueous non-ionic base was not stable (Table I). Based on these data, a new anhydrous base was developed and tested on Melanoderm. Kojic acid in this base showed 41% inhibition in tyrosinase activity (data not shown). Moreover, HPLC data showed that only 2% of kojic acid was lost after 26 weeks in an anhydrous base (Table I). In contrast, 87% of the kojic acid was lost from the aqueous base after five weeks (Table I). To confirm in vitro data, clinical testing was conducted with both anhydrous and aqueous bases containing kojic acid. There was a gradual increase in lightening of the skin over a period of three months with the anhydrous base containing kojic acid (Table II). The non-ionic aqueous base containing kojic acid was less effective (Table II), presumably due to loss of kojic acid over a period of three months. In conclusion, the kojic acid inhibition of tyrosinase in Melanoderm, the kojic acid quantification by HPLC, and the relative clinical efficacy of aqueous and anhydrous bases containing kojic acid showed good agreement. Thus, our present study indicates that Melanoderm can be used as a quick, inexpensive, and reliable in vitro model for screening whitening agents and helping to develop whitening products before clinical testing. ACKNOWLEDGMENTS The authors wish to thank Melanie Smith and Bruce Smith for their editorial support. Susan Cooke is gratefully acknowledged for her expert assistance with the photoshop. Table I HPLC Analysis of Kojic Acid Time Lost (%) Lost (%) Base (weeks) 23øC 37øC Aqueous Anhydrous 0.00 0.00 -- 2.00 0.00 14.79 5.00 87.83 86.35 0.00 0.00 -- 8.00 -- 21.22 12.00 -- 19.20 26.00 2.03 --

SKIN-WHITENING PRODUCTS 3 67 Table II Clinical Study Type of base One Two Three with kojic acid month months months Aqueous 0.00 0.90* 0.20 Anhydrous 0.80 1.6' 2.6' Hyd roquinone 1.10* 0.2 2.8 * Caused significant reduction in skin darkness (L*) as compared to baseline value (P • 0.05) REFERENCES (1) K. S. Stenn and L. Weiss. "The Skin" in Histology O Tissue Biology (Elsevier Biomedical, New York, 1983). (2) D. Tobin, A. Quinn, S. Ito, and A. Thody, The presence of tyrosinase and related proteins in human epidermis and their relationship in melanin type, Pigment Ce//Res., 7, 204-209 (1994). (3) V.J. Hearing and K. Tsukamoto, Enzymatic control of pigmentation in mammals, FASEB J., 5, 2902-2909 (1991). (4) S.J. Orlow, R. E. Boissy, D.J. Mortan, and S. Pifkohirst, Subcellular distribution of tyrosinase and tyrosinase related protein. 1: Implication for melanosomal biogenesis. J. Invest. DermatoL, 100, 55-64 (1993). (5) K. Nakazwa, F. Sahue, O. Damour, C. Cellombel., and H. Nakazawa, Regulatory effects of heat on normal human melanocyte growth and melanogenesis comparative study with UVB, J. Invest. Der- matoL, 110, 972-977 (1998). (6) M. Archambault, M. Yaar, and B.A. Gilchrest, Keratinocytes and fibroblasts in a human skin equivalent model enhance melanocytes' survival and melanin synthesis after ultraviolet irradiation, J. Invest. DermatoL, 104, 859-867 (1995). (7) R. Halaban, R. Langdon, and N. Birchall, Basic fibroblasts growth factor from human keratinocytes is a natural mitogen for melanocytes, J. Cell. Biol., 107, 1611-1619 (1988). (8) K. Nakazawa, O. Damour, and C. Collombel, Modulation of normal human melanocytes' dendricity by growth promoting agents, Pigment Cell Res., 6, 406-416 (1993). (9) O. Lee and E. Kim, Skin lightening, Cosmet. Toiletr., 110, 51-56 (1995). (10) E. Rafal, C. E. M. Griffiths, C. M. Ditre, L.J. Finkel, T. A. Hamilton, and J.J. Voorhees, Topical tretinoin (retinoic acid) treatment for liver spots associated with photodamage, New. Engl. J. Med., 326, 368-374 (1992). (11) A.M. Fishman, Skin lighteners, Happi., 42, (February 1998). (12) B.J. Kim, J. M. Kim, H. P. Kim, and M. Y. Heo, Biological screening of 100 plant extracts for cosmetic use: Inhibitory activities of tyrosinase and DOPA auto-oxidation, Int. J. Cosmet. Sci., 19, 291-298 (1997). (13) P.S. Friedmann and B. A. Gilchrest, Ultraviolet radiation directly induced pigment production by cultured human melanocytes,J. Cell. Physiol., 133, 88-94 (1987). (14) K. Kameyama, C. Sakai, S. Kondoh, K. Yonemoto, S. Nishiyama, and K. Blanock, Inhibitory effect of magnesium ascorbyl phosphate on melanogenesis in vitro and in vivo, J. Am. Acad. Dermatol., 34, 29-33 (1996). (15) G. Majmudar and M. Smith, In vitro screening technique in dermatology: A review of the tests, models and markers, Cosmet. Toiletr., 113, 69-76 (1998).