j. Cosmet. Sci., 53, 237-240 (July/August 2002) A test for antioxidant activity in cosmetic formulations E. PELLE, T. MAMMONE, K. MARENUS, D. DICANIO, and D. MAES, Estee Lauder Research Laboratories, 125 Pindawn Road, Me/ville, NY l 1747. Accepted for publication March 15, 2002. Synopsis The aim of this study was to develop a technique to assay for the activity of antioxidants in a finished cosmetic product. This was accomplished by adapting the Randox Assay for Total Antioxidant Status kit so that diluted samples could be evaluated by kinetic as well as end-point determinations. Using this technique, we found that a finished product had an IC5o of 0.07 gm of product and a relative antioxidant activity concentration of 52.7 nmoles/mg. INTRODUCTION Environmental insult to human skin by ultraviolet (UV) radiation, as well as by cigarette smoke and air pollution, generates reactive oxygen intermediates that contribute to both acute and chronic skin damage (1,2). For example, immediately after overexposure to sunlight, an erythemal response is induced that is associated with epidermal inflamma- tory oxidative reactions. Moreover, in terms of chronic exposure, the involvement of oxygen free radicals has also been implicated in actinic skin damage that manifests itself in elastosis, collagen disorganization, and most notably in the appearance of wrinkles (3). Due to increased outdoor leisure activities, these visible signs of photodamage and premature aging have become widespread in our society. To address this problem, the cosmetics industry has devoted much research toward the development of various skin care products. Although protective sunscreen products that absorb UV and diffuse photonic energy are widely used, cosmetic products that contain antioxidants, which scavenge deleterious reactive oxygen species produced in skin after environmental trauma, have also become standard for a healthy skin care regimen. Although analytical techniques are available to measure the level of antioxidants in cosmetic products, in general, they do not provide any information regarding their potential activity. Further, due to the complex nature of cosmetic formulations, extract- ing and determining biochemical activity in a finished product can be a challenging task. Previously, we evaluated the antioxidant potential of certain cosmetic ingredients (4) and also the antioxidant efficacy of finished products on skin (5). In this study, we 237

238 JOURNAL OF COSMETIC SCIENCE now report on a novel technique to measure antioxidant activity directly in a complete cosmetic product. MATERIALS AND METHODS SAMPLE PREPARATION A typical cosmetic formulation containing a blend of emulsifiers was prepared as either a control with no antioxidants or as a complete formula with a mixture of antioxidants. The following antioxidants were used in the formulation: 2.0% tocopheryl acetate (Hoffman-LaRoche, Parsippany, NJ), 0.1% butylated hydroxy toluene (Rhone-Poulenc, Cranbury, NJ), 1.0% magnesium ascorbyl phosphate (Barnet, Englewood, NJ), 0.1% ubiquinone 50 and 0.5% N-acetyl-L-cysteine (Seizer, Carlsbad, CA), 0.1% rosemary (Robertet, Oakland, CA), and 0.1% tocopherol cysteamine (Mercier, S. Plainfield, NJ). ASSAY The Randox Assay for Total Antioxidant Status kit (Randox, Antrim, UK) was adapted for use in cosmetic products by diluting the formulations to be tested to 1% in isopropyl alcohol. At 1% in isopropyl alcohol, the samples are sufficiently clarified and the antioxidants solubilized to allow the reaction to proceed without interference. Briefly, 2,2'-azino-di-(3-ethylbenzthiazoline sulphonate) (ATBS) is reacted with a peroxidase and H20 2 to convert ATBS into a radical cation. In this state, ATBS forms a chromogen that can be measured spectrophotometrically at 600 nm. In the presence of antioxidants, this color formation is inhibited. Typically, 50-100 pl of the 1% sample is diluted in water up to 250 pl. Then, 1.5 ml of chromogen solution is added, followed by the addition of 0.3 ml of substrate solution. The absorbance (A) of the samples is then measured immediately in a Beckman DU-7500 spectrophotometer using the kinetics/ time program. CALCULATIONS Percent inhibition was calculated as (dAvehicle-dAproduct/dA,•ehicle) x 100 and used to quantitate an IC5o value. Also, a range of 15 to 85 nanomoles of an antioxidant standard (6-hydroxy-2,5,7,8-tetramethylchroman-2-carboxylic acid) was used to determine the relative activity of a product. RESULTS One hundred microliters (1 mg) of a 1% cosmetic sample dilution was assayed, and a typical three-minute kinetic plot of the data is shown in Figure 1. The sample without antioxidants had a dA/min of 0.126 whereas the sample containing antioxidants had a dA/min of 0.011. Thus, there was an 86.8% decrease from the control sample. From similar kinetic plots, the average value that would inhibit the reaction by 50% was then determined to be 0.7 mg of the sample. Since the sample was a 1% dilution of the