DELIVERY AND METABOLISM OF o•-TAc 229 metabolism experiments. The smallest volume of the formulation required, to obtain complete and uniform coverage of the diffusion cell surface area (0.636 cm2), was determined to be 5 pl, corresponding to a weight of about 4 mg. After application, the preparation was uniformly spread on the stratum corneum side of the skin with the help of a glass rod, and the tip of the rod was washed into a vial containing 2 ml ofacetonitrile in order to account for the material lost on spreading. With this technique, the exact amount of material applied on the skin surface was determined. STATISTICAL DESIGN OF THE EXPERIMENT As there were three important factors to be considered, viz., the formulations, interday variability, and replicates on both days, it was decided to set up an experimental design that could incorporate all the factors. A randomized complete block design was chosen as the statistical design for the experiment. The statistical model used is shown in Table III. Using this table, each formulation was applied in duplicate on each day, such that at the end of two days there were four replicates per formulation. IN VITRO SKIN PERMEATION/METABOLISM METHODOLOGY A flow-through system was used for conducting i, vitro permeation experiments. The total system consisted of a receptor fluid reservoir a variable flow rate peristaltic pump, Cassette © (Manostat, New York, NY) a circulating water bath, Lauda © (Brickman Instrument, Westbury, NY) two cell-holding blocks 14 Teflon © flow-through diffu- sion cells and a Retriever IV fraction collector (ISCO Inc., Lincoln, NE) to collect effluent fractions over the adjusted time period. Each diffusion cell had an inner diameter of 9 mm and a surface area of 0.636 cm 2 exposed to the receptor fluid. The receptor fluid was pumped at the flow rate of 1.5 ml/h from the reservoir to the diffusion cells placed Table III Statistical Randomized Complete Block Design for the Application of Formulations Formulation Day 1 a Day 2 IPM solution W1 b Y1 •l X1 • Z1 e Gel 1 W2 Y2 X2 Z2 Gel 2 W3 Y3 X3 Z3 Gel 3 W4 Y4 X4 Z4 Emulsion 1 W5 Y5 X5 Z5 Emulsion 2 W6 Y6 X6 Z6 Emulsion 3 W7 Y7 X7 Z7 a Two consecutive days, days 1 and 2. b,c Replicates on day 1 for the IPM solution. a,• Replicates on day 2 for the IPM solution.

230 JOURNAL OF COSMETIC SCIENCE in the holding blocks. The skin surface temperature was maintained at 32øC by adjust- ing the circulating water bath temperature to 39øC (11). The effluent from the diffusion cells was collected directly into glass scintillation vials every eight hours for 24 hours. Experiments were conducted in quadruplicate. SKIN TREATMENT At the conclusion of the experiment, the donor compartment was washed three times with 1 ml of acetonitrile. The washes were collected and analyzed by HPLC for the amount of active remaining on the surface. The washed skin samples were removed from the cells. The tape-stripping technique was used to separate the stratum corneum from the rest of the epidermis to get an estimate of material remaining in the barrier layer of the skin. In this technique, seventeen strips of the active-treated side of the skin, using a 3M Scotch TM tape, were taken as two + fifteen strips. The first two strips represented the active superficially adhering to the surface (and so included in the wash), and the next 15 strips represented the active recovered from the stratum corneum. To each of the combined strips, 10 ml of ,-hexane was added exactly. Both these strips were shaken in a wrist-action shaker for 45 minutes, at the end of which the mixture was filtered and injected into the HPLC. The remainder of the skin was placed in plastic culture tubes, 5 ml of ice cold DMPBS was added, and the skin was homogenized (Polytron homogenizer, Switzerland) for five minutes until a buff-colored suspension was obtained. The absence of chunks of skin was ensured. This suspension was extracted three times, each time with 5 ml of ,-hexane. Each extraction involved shaking the mixture on a wrist-action shaker for 45 minutes. After extraction, a 45-minute centrifugation process helped to separate the ,-hexane and DMPBS layers, and the upper ,-hexane layer was carefully removed with a pipette and pooled together into 30-ml glass tubes. This procedure was repeated three times, each time shaking for 45 minutes and centrifuging for 45 minutes for each skin sample. The pooled hexane layer was evaporated under vacuum. Acetonitrile (2 ml) was added to this mixture, vortexed to ensure complete mixing, and the solution was filtered and injected into the HPLC column. Prior experiments were conducted (unpublished observations) to establish the adequacy of hexane extraction for both o•-TAc and o•-T from both strips and pig skin. In both cases greater than 99% recovery of the active was obtained (12). RECEPTOR TREATMENT Receptor solutions were collected in glass scintillation vials every eight hours, and the 3% BSA was precipitated using 4 ml of ACN. The active was extracted into the organic ,-hexane layer (5 ml) by shaking the above mixture for 45 minutes on a wrist-action shaker, followed by a 45-minute centrifugation process. This procedure was repeated two times. Centrifugation was followed by careful removal of the ,-hexane layer into 30-ml glass tubes and evaporation of the organic phase. Acetonitrile (1.5 ml) was added and vortexed to ensure good mixing. A sufficient quantity of this mixture was filtered and injected into the HPLC. Thus, the amount of o•-TAc and its metabolite was estimated in the following four locations in each i, vitro permeation experiment: (a) receptor fluid, (b) washes, (c) stratum corneum (from strippings), and (d) viable tissues of the skin.