j. Soc. Cosmet. Chem., 45, 229-238 (September/October 1994) Effect of salicylic acid encapsulation by phospholipid vesicles on transport through an inert membrane ERIC P. GUI•NIN* and JOEL L. ZATZ, Department of Pharmaceutics, Rutgers University College of Pharmacy, P.O. Box 789, Piscataway, NJ 08855-0789. Received October 12, 1993. Synopsis Salicylic acid (SA) was incorporated into phospholipid vesicles based on Phospholipon © 80. Entrapment of SA and flux through an inert membrane (Silastic ©) were a function of pH. From mathematical treatment of data obtained at more than one value of pH, the relative contributions of ionized and unionized species were determined. Both entrapment and flux were dominated by the uncharged form of the compound. An increase in lipid concentration resulted in an increase in entrapment and corresponding reduction in the steady-state flux of SA through the membrane. The SA permeation pattern following liposome application under finite dose conditions was qualitatively different from that in buffer. Penetration from liposomes was slower, and the amount crossing the membrane at 24 hours was significantly less. INTRODUCTION Topical administration of active compounds has been accomplished in the past using traditional ointments, creams, and lotions. An alternative means of delivery is offered with vesicular systems, which have been the subject of several investigations. The components of liposomal transport were studied with hairless mouse skin and involved three hypothetical mechanisms: transfer of the drug directly from liposome to skin, transfer to the skin after passage in the aqueous compartment, or a combination of both (1). Liposome preparations have been investigated with in vivo and in vitro studies for possible therapeutic applications (2-10). The results suggested a decrease of systemic absorption relative to control formulations and an increase of skin concentration, espe- cially in the stratum corneum and epidermis. In these studies, liposome concentrations up to 5 % (w/w) were used, but none of these reports indicates a relationship between the liposome concentration and the extent of delivery. The release of drug compounds from formulation into the skin can be decomposed into two major components: the rate of release from the vehicle and the permeability through the skin, which are physicochemical and physiological processes, respectively (11). In a first part (12), it was reported that phospholipids and nonionic surfactant involved in vesicle formulation had minor effect on skin integrity (monitored by water permeabil- ity). The focus of this second study is to derive the physicochemical parameters con- * Presently at IFF R&D, 1515 State Highway 36, Union Beach, NJ 07735. 229

230 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS trolling the release. The purpose of this work is to determine the relationship between liposomal uptake and kinetics of delivery of an ionizable compound, salicylic acid (SA), through a model membrane, Silastic ©. This constitutes a stepwise approach to derive an explanation for the enhancement of drug levels into the skin reported in the literature. MATERIALS AND METHODS CHEMICALS The chemicals for formulations or analysis were salicylic acid, certified A.C.S. (Fisher Scientific, Fair Lawn, NJ), 1,1,1-trichloro-2-methyl-2-propanol, cholesterol primary standard (Eastman Kodak, Rochester, NY), Phospholipon © 80 (Nattermann Phospho- lipid GMBH, Cologne, Germany), and •4C salicylic acid, with a specific activity of 56.1 mCi/mmol and a concentration of 0.1 mCi/ml (NEN, Wilmington, DE). All other chemicals were reagent grade. MEMBRANE Silastic ©, medical grade sheeting, non-reinforced, 0.005-inch-thick (500-1), was pur- chased from Dow Corning Medical (Midland, MI). The membrane was cut to the dimensions of the diffusion cell. Then the disk of Silastic © was washed with soap and water and dried in an oven at 45øC overnight prior to use. LIPOSOME MANUFACTURE Phospholipid vesicles were prepared using Phospholipon © 80 as a lipid source. Phos- pholipon © 80 is a commercially available raw material derived from soy lecithin. It is composed of phophatidylcholine (76 -+ 3%), but it also contains lyso- phosphatidylcholine (3 -+ 3%) and fatty acids. The dried film method, described by Bangham (13), was used to manufacture the liposomes. Briefly, the phospholipids are deposited from chloroform/methanol (10:1) in a thin film on the wall of a round-bottom flask by rotary evaporation under reduced pressure. This film was further dried by a gentle flow of nitrogen overnight. 200 mmol pH 4.5 phosphate buffer, or 200 mmol pH 2 HCI/KCI buffer containing 0.5 mg/ml of SA (spiked with •4C salicylic acid for Silastic © permeation experiments), was added, and the lipids were hydrated at about 50øC. The manufacture was completed by a brief period of sonication to provide ho- mogeneity of the preparation (Fisher Sonic Dismembrator model 300, Fisher Scientific, Fair Lawn, NJ). Formation of multilamellar liposomes was verified by freeze-etching scanning electronic microscopy. In preliminary experiments, salicylic acid was also incorporated by adding a solution to preformed vesicles. The degree of entrapment of this solute (see methodology below) was unaffected by the variation in manufacturing procedure. ENTRAPMENT DETERMINATION The entrapment efficiency was determined by ultracentrifugation of the preparation at