80 JOURNAL OF COSMETIC SCIENCE In the final analysis, i, vivo human testing is the most definitive method. Several procedures have been tried, and these include the immersion of treated sites on the forearms in a water or whirlpool bath for a measured period of time followed by exposure to UV to determine the degree of residual photoprotection (3,4). However, the most commonly used test for measuring the washout resistance of sunscreens in the United States is one that has been proposed by the FDA (5). In this method the sunscreen is declared water-resistant if it retains substantial UV absorbance after two 20-minute periods of water activity. The main objective of the present study was to develop simple i, vitro methods for studying the water resistance of sunscreens using micro-Yucatan pig skin. Certain polymers were included in prototype sunscreen formulations to evaluate their effective- ness in enhancing water resistance. METHODS MATERIALS Freshly harvested excised micro-Yucatan pig skin was purchased from Charles River Laboratories, Wilmington, MA. Benzophenone-3, OMC, diisopropyl adipate, propylene glycol (and) diazolidinyl urea (and) methyl paraben (and) propyl paraben (Germaben © II), imidazolidinyl urea, isocetyl alcohol, polyquaternium-28, PVP/eicosene copolymer, PVP/VA copolymer, and tricontanyl PVP were obtained from ISP, Wayne, NJ. Radio- isotope labeling of sunscreens, [•4C] benzophenone-3 with a specific activity of 30.5 mCi/mmol and [3H] octyl methoxycinnamate (573 mCi/mmol), were obtained from Chemsyn Science Laboratories, Lenexa, KS. Adipic acid/diethylene glycol/glycerin (ADG) crosspolymer (Inolex Chemical Co., Philadelphia, PA), Carbomer 940 (BF Goodrich Specialty Chemicals, Brecksville, OH), DEA-cetyl phosphate (Hoffmann-La Roche Inc., Paramus, NJ), PEG-15 cocamine (Akzo Nobel Chemical Inc., Chicago, IL), Poloxamine 704 (BASF Corp, Parsippany, NJ), PPG-12 SMDI copolymer (polyolpre- polymer-2 or PP-2) and PPG-51 SMDI copolymer (polyolprepolymer-14 or PP-14) (Barnet Products Corp., Englewood Cliffs, NJ), polyquaternium-10 (Amerchol, Edison, NJ), polyquaternium-41 (Phoenix Chemical, Somerville, NJ), SD alcohol 40 (Eastman Chemical Co., Kingsport, TN), and PEG-20 oleyl ether (Croda Inc., Parsippany, NJ) were obtained as research samples. Oily Vegetol Marigold WL1072 ©, octyl dodecyl myristate, and polyglycerol-3-DI-IS were received as research samples from Gattefosse, Westwood, NJ. HPLC-grade acetonitrile, beeswax, glacial acetic acid, glycerin, mag- nesium sulfate, methyl paraben, mineral oil, reagent alcohol, scintillation fluid, sodium chloride, and triethanolamine were purchased from Fisher Scientific, Fair Lawn, NJ. Tissue-solubilizing fluid (Solvable TM) was purchased from Packard Instrument Co., Meriden, CT. Transparent tape #800 (Scotch TM) was purchased from 3M Packaging Systems Division, St. Paul, MN. All other materials were of high quality and purity and obtained from standard sources. SUNSCREEN FORMULATIONS The in vitro studies to develop a procedure for measuring water resistance were performed using hydroalcoholic and oil prototype formulations and o/w and w/o emulsions. Dii-

WATER RESISTANCE OF SUNSCREENS 81 sopropyl adipate was chosen as solvent for the prototype oil formulation with benzo- phenone-3 and octyl methoxycinnamate concentrations at 6% and 7.5% w/w, respec- tively. The other formulations are presented in Tables I through III. IN VITRO MODELING TECHNIQUE Radiolabeling of formulations. [•4C] benzophenone-3 was obtained as solid material and was solubilized in ethanol to give a specific activity of 1.05 pCi/pl of solution. Sunscreen preparations were spiked with the radiolabeled solution so that each microgram of benzophenone-3 applied to the skin surface had approximately 1600 dpm (disintegra- tions per minute) for the hydroalcoholic preparations, 600 dpm for the diisopropyl adipate oil formulations, and 450 dpm for emulsions. [3HI octyl methoxycinnamate was obtained as a solution in toluene with a specific activity of 1.93 pCi/pl. Sunscreen preparations were spiked with the radiolabeled solution so that each microgram of octyl methoxycinnamate applied to the skin surface for in vitro studies had 1600 dpm for hydroalcoholic preparations, 450 dpm for diisopropyl adipate oil formulations, and 250 dpm for emulsions. Selection of receptor fluid. The receptor fluid was aqueous solution of 0.5 % polyoxyethylene oleyl ether (Volpo © 20), a nonionic surfactant with an HLB of 16. Polyoxyethylene oleyl ether increased the solubility of both sunscreens by the mechanism of micellar solubi- lization. This receptor fluid, though non-physiologic, increases the solubility of sun- screen agents without affecting skin barrier function (6). Preparation of skin membrane. Upon receipt the freshly excised skin was washed gently with 1% (w/w) aqueous dishwashing detergent, rinsed with aleionized water, and patted dry with a paper towel. A 250-300 lam thick layer of the skin was cut from the surface with a Padgett © Electrodermatome (Padgett Instrument, Kansas City, MO). The skin pieces were then rinsed and dried with paper towels before storage in plastic bags at 4øC. The skin was removed from the refrigerator and kept in isotonic solution to hydrate at room temperature one hour before starting the experiment. The dermatomed skin was then cut into 10-mm circular pieces with a brass punch and placed epidermis-side up in the diffusion cells. The skin treated in this fashion from the stage of receipt until use retained its original permeability characteristics for four weeks after dermatoming (7). After mounting the skin in diffusion cells the receptor fluid was permitted to flow for 15 minutes followed by application of 0.5 ml of aleionized water to the skin surface for 30 minutes. After ensuring that the skin was intact (no leakage or drainage), the water Table I Composition of Hydroalcoholic Prototype Formulation Ingredient % W/W Benzophenone-3 3.0 Octyl p-methoxycinnamate 7.0 SD alcohol 40 (anhydrous) 68.0 Poloxamine 704 9.0 Polymer (if present) 5.0 Water 8.0