DELIVERY AND METABOLISM OF ot-TAc 227 Jersey) carbomer Carbopol ©, (BF Goodrich, Ohio) Bis (2-hydroxyethyl)-ammonium hexadecyl hydrogen phosphate, DEA-cetyl phosphate, Amphisol © (Roche Vitamins and Fine Chemicals, New Jersey) diazolidinyl urea, Germall © (Sutton Laboratories, New Jersey) Ethomeen C/25 (Akzonobel, Illinois) and hydroxypropyl cellulose, Klucel © (Hercules Co., Germany). Tween © 20, polyoxyethylene (20) sorbitan monolaurate, and Tween © 80, polyoxyethylene (20) sorbitan monooleate were from ICI surfactants (Dela- ware). Transcutol, ethoxy diglycol, was obtained from Gattefoose (France). Special 30-ml centrifuge tubes were obtained from Corning Inc. The solvents acetonitrile, chloroform, and n-hexane of HPLC grade were obtained from Fisher Scientific (Springfield, NJ). Water refers to freshly deionized water. The centrifuge used was Fisher Centrific. Or- ganic solvents were evaporated using a Savant Speed Vac ©, SC 110. The wrist-action shaker used was from Burrell (Ohio). Vortexing was done with a Vortex-Genie (New York). Particle size analysis was carried out with a Bookhaven Instruments Corporation particle size analyzer where 20 ul of the microemulsion was diluted with 480 lal of water before size determination. The mechanical shaker bath used was a Dubnoff metabolic shaking incubator (Precision Scientific, Illinois). TOPICAL FORMULATIONS All formulations, listed in Tables I and II, were prepared on a weight/weight basis. A simple isopropyl myristate (IPM) solution of tx-TAc (5%) was also prepared. The sta- bility of formulations was assessed by HPLC immediately after preparation and after storage for three weeks at room temperature. Gels were prepared by shaking with the gelling agent for eight hours on a shaker bath at room temperature. The aqueous phase and oily phases of emulsions were heated to 55 øC, after which the former was added to the latter and agitated to produce creamy emulsions. Clear microemulsions were made using a simple vortexing procedure. RECEPTOR FLUID In the metabolism study the receptor media must serve two important functions, viz., maintaining the viability of skin tissue and ensuring adequate solubility of the prodrug and its metabolite. Dulbecco's modified phosphate-buffered saline (DMPBS) was used to maintain tissue viability (8,9). As compounds that are essentially insoluble in water (like the tx-T derivatives) may not partition freely from excised skin into an aqueous receptor fluid, bovine serum albumin (3%) was added to DMPBS (8,10). Fresh deionized water Table I Gel Formulations Used in This Study Gel 1 Gel 2 Gel 3 Ingredients (% w/w) (% w/w) (% w/w) c•-TAc 5 5 5 SD alcohol 92 87 67 Isocetyl alcohol -- -- 12 PEG-15 cocamine 5 5 Hydroxypropyl cellulose 3 3 3 Water -- 8

228 JOURNAL OF COSMETIC SCIENCE Table II Emulsion Formulations Used in This Study Emulsion 1 • Ingredients Quantity (% w/w) o•-TAc Diisopropyl adipate Mineral oil DEA-CP Water Carbomer Diazolidinyl urea Emulsion 2 b 5 7.5 7.5 2 77.4 0.3 0.3 Ingredients Quantity (% w/w) o•-TAc Isopropyl myristate Polysorbate 80 Sorbitol Water 5 10 12 30 43 Emulsion 3 • Ingredients Quantity (% w/w) o•-TAc 5 Benzyl alcohol 12.3 Water 42.7 Diethylene glycol monoethyl ether 15.9 Taurocholate sodium 1 Polysorbate 20 18.1 o/w macroemulsion. Microemulsions containing IPM or benzyl alcohol as oily phases. was used to prepare the buffer, and the solutions were adjusted to pH 7.4 with 10% w/v NaOH prior to use. ANIMAL TISSUE AND PREPARATION Fresh viable micro-Yucatan pig skin was obtained from Charles River Laboratories (Wilmington, MA). The pig skin was cut into squares of 10 x 10 cm 2 and placed in Tupperware © containers filled with DMPBS. The skin was then sent in cool packs and used for the metabolism experiments within one day of its arrival. Upon receipt, the fresh skin was gently washed with a 1% (w/w) mild soap and deionized water. A 250-300-pm-thick layer of the skin was cut from the surface with a Padgett Electro- dermatome TM instrument (Padgett Instrument, Kansas City, MO). The dermatomed skin was used the same day and was cut into 10-mm circular pieces with a brass punch and placed epidermis-side up in Bronaugh diffusion cells. DOSING Finite dosing was used to simulate actual use conditions in all the i, vitro permeation and