JOURNAL OF COSMETIC SCIENCE 372 EFFECT OF ARBUTIN, KOJIC ACID, α-LIPOIC ACID, AND AZELAIC ACID AND THEIR COMBINATIONS ON MELANIN SYNTHESIS To assess the effect of these agents and their combinations on melanogenesis, the inhibition of melanin production in HSMs was examined. The concentrations used were the maxi- mum non-cytotoxic for each agent and combination. As shown in Figure 5, arbutin, kojic acid, azelaic acid, and α-lipoic acid produced an inhibition of 27%, 16%, 3%, and 46% at their maximum non-cytotoxic concentrations, respectively. Arbutin + azelaic acid, arbutin + kojic acid, arbutin + α-lipoic acid, azelaic acid + α-lipoic acid, kojic acid + α-lipoic acid, and azelaic acid + kojic acid produced an inhibition of 25%, 33%, 35%, 41%, 47%, and 22%, respectively. Hydroquinone inhibition on melanin synthesis was used as a reference compound. These results suggest that there is not an observable synergistic effect between these agents. However, a potentiation effect is observed in arbutin + α-lipoic acid and azelaic acid + kojic acid combinations in which one of the agents has no effect individually, but the combination effect is higher than the effect of the other agent. DISCUSSION In this study, combination effects between arbutin, kojic acid, azelaic acid, and α-lipoic acid were investigated. In the complex synthesis pathway of melanin, the key enzyme is tyrosinase, which regulates the fi rst two steps of the pathway and is a common target of depigmenting agents. However, depigmenting agents can act at different levels in the production of melanin. This is why the most common classifi cation for these agents is based on their mechanism of action. Because of these different ways to inhibit melanin synthesis, it has been proposed that combining these agents can lead to an increase in the inhibition of melanin synthesis, making the effect of the combination higher than the sum of the two agents individually, which is called a synergistic effect. To evaluate these possible synergistic effects, we fi rst performed a WST-1 assay to fi nd the maximum non-cytotoxic concentrations of each inhibitor and combination between them, which were used after in the following experiments. To study the mechanism of action of these agents, we performed a kinetic analysis on the inhibition of mushroom tyrosinase. Afterwards, we evaluated possible synergies between different combinations in mushroom tyrosinase using the Chou–Talalay method (23). Finally, we tested the Figure 5. Effect of arbutin, kojic acid, azelaic acid, α-lipoic acid, and their combinations on melanin syn- thesis in human skin melanocytes. Cells were incubated at the maximum non-cytotoxic concentrations for 72 h. Results are expressed as percentage relative to control.

COMBINATION OF DEPIGMENTING AGENTS IN VITRO 373 inhibition of melanin production of these agents and combinations in human melano- cytes to detect possible synergistic or increased effects. In the WST-1 assay, we saw that arbutin, kojic acid, azelaic acid, and α-lipoic acid were not cytotoxic up to 1000, 100, 100, and 10 μg/ml, respectively. As observed, all these agents are less cytotoxic than hydroquinone, as it has been previously described (1,2,30). Moreover, we observed that arbutin (1000 μg/ml), kojic acid (100 μg/ml), and α-lipoic acid (10 μg ml) showed a very signifi cant increase in cell viability (p 0,001), and azelaic acid was also but with less signifi cance (p 0.05). A cytoprotective and antioxidant activ- ity of arbutin has been reported by Seyfi zadeh et al. and Takebayashi et al. (25,26), al- though they used liver cells and fi broblasts, respectively, instead of melanocytes. Antioxidant effect of kojic acid, azelaic acid, and α-lipoic acid has also been reported (10,12,19,20). So, it might explain this proliferative effect (25). This effect is also seen in some of the combinations between agents. Regarding the combinations tested, hydroqui- none was not included because as mentioned in the introduction, it is very cytotoxic and has been forbidden in cosmetics. Mushroom tyrosinase inhibition kinetics was studied to fi nd out the mechanism of inhi- bition of these agents over mushroom tyrosinase. Our results confi rm what is proposed by other authors: arbutin and azelaic acid are competitive and kojic acid is a mixed-type inhibitor (1,8,9). Besides, in this project, we have studied the type of inhibition of α-lipoic acid, which as far as we know has not been described before. According to our results, α-lipoic acid seems to be a competitive inhibitor of diphenolase activity of mush- room tyrosinase. Mushroom tyrosinase assay was performed to study synergistic effects of agent combina- tions on tyrosinase activity. Dose/effect curves for each compound and combination were performed to obtain the CI, which indicates synergism (1), additive effect (=1), or antago- nism (1). As expected, an additive effect was observed between α-lipoic acid and azelaic acid as they have the same type of inhibition (competitive) on mushroom tyrosinase. Kojic acid is a mixed-type inhibitor of mushroom tyrosinase, so synergistic effects with the other agents could be expected. Indeed, kojic acid + α-lipoic acid combination showed to be synergistic (CI = 0.70). However, kojic acid with arbutin or azelaic acid combinations showed to have an antagonistic relationship. This might be due to the fact that although arbutin, azelaic acid, and α-lipoic acid bind to the same site of the enzyme (same type of inhibition), their binding can be directed to different mechanistic forms (different enzyme– substrate complexes) (27). So this could explain why there is a synergy between α-lipoic acid and kojic acid and an antagonistic effect between arbutin or azelaic acid and kojic acid. Melanin content measurement was carried out to study synergistic effects on melanin synthesis. The individual values are in concordance with that of Lajis et al. (13), Tai et al. (28), and Lee et al. (29). Hydroquinone inhibition value was very low because the concentration used was smaller than the other compounds due to its high cytotoxicity. Arbutin + azelaic acid combination had similar values than arbutin alone (27% vs. 25%), indicating that there is neither synergy nor antagonism between them. The same happens with the azelaic acid + α-lipoic acid combination (41% of inhibition in combination and 45% of inhibition by α-lipoic acid alone) and kojic acid + α-lipoic acid combination (47% of inhibition in combination and 46% by α-lipoic acid alone). However, kojic acid + α-lipoic acid combination showed to be synergistic over mushroom tyrosinase. This mismatch between mushroom tyrosinase inhibition and melanin synthesis inhibition