376 JOURNAL OF COSMETIC SCIENCE been encapsulated in liposomes for imaging agents (5-7). Some clinical trials for tumor imaging using labeled liposomes have already been initiated (7-9). Tranexamic acid (TA) is a hydrophilic drug used as an antifibrinolytic agent (10-15). It is a trans-4-aminomethyl cyclohexane carboxylic acid (CsH•5NO2, mw 157.21) (16). It has been claimed that TA has anti-inflammatory (16) and whitening effects for topical use (17). The current commercially available preparations of TA are tablets and injec- tions (18). We have previously developed liposome formulations for TA from various lipid compositions, containing neutral or positively or negatively charged lipids, which may potentially reduce the irritation and allergy caused by TA and improve the mois- turizing effect (19). It was found that the charged liposome composed of hydrogenated soya phosphatidylcholine/cholesterol/stearylamine (+) or dicetyl phosphate (-) at a molar ratio of 7:2:1 for TA demonstrated 13-16% of drug entrapment. Thus, these liposome formulations were selected for further stability and release studies. It is expected that the developed liposome formulations can be applied for the future development of TA, not only as a sustained release preparation but for topical whitening cosmetics as well. The present study reports the characterization of stability and release of TA from various multilamellar liposome formulations, prepared by a chloroform film method with soni- cation. The release rates of TA from liposomes were compared with those of free TA from solutions (5% and 10% in deionized water). Various factors (TA content and charges of liposomes) that may affect the stability and release properties of TA are discussed. EXPERIMENTAL MATERIALS TA was obtained commercially from Asahi Chemical Industry Co., Ltd. (Japan). Hy- drogenated soya phosphatidylcholine (Emulmetik 90 ©) (HSC) was kindly donated by JJ-Degussa (T) Ltd., Bangkok. Cholesterol (CHL), dicetyl phosphate (DCP), stearyl- amine (SA), 2,4,6-trinitrobenzosulfonic acid, and boric acid were purchased from Sigma Chemical Company (St. Louis, MO). Triton-X 100 was obtained from BDH Ltd. (Poole, England). Chloroform and potassium dihydrogen orthophosphate were of analytical reagent grade. PREPARATION OF LIPOSOMES Liposome dispersion samples were prepared by a chloroform film method with sonica- tion as previously described (19). The lipids used were neutral (HSC and CHL) and charged lipids (SA for positively charged lipids and DCP for negatively charged lipids), with a total lipid concentration of 25 mg/ml. Liposome formulations composed of HSC/CHL/SA = 7:2:1(+) and HSC/CHL/DCP = 7:2:1(-) at molar ratios, with the entrapped TA (5% and 10% solutions in DI water), were prepared. For each lot, an amount of 60 ml of each liposome dispersion sample was prepared. The liposome dispersion samples were kept at 4 ø + løC and protected from light prior to use.

TRANEXAMIC ACID LIPOSOMES 377 LIPOSOME SIZE DETERMINATION The particle size and size distribution of liposome dispersion samples were measured by a light-scattering particle analyzer (Mastersizer S Long Bed Ver. 2.11, Malvern Instru- ments Ltd., Malvern, UK), ten days after sample preparation (kept at 4øC). The particle size range was set between 0.05 and 800 t•m, with the beam length at 2.40 mm and the dispersant refractive index at 1.3300. Polydisperse model analysis was employed. STABILITY STUDY Each liposome formulation (5 ml for physical and 20 ml for chemical stability studies) was transferred into a vial with stopper prior to storage at three different temperatures, namely 4 ø, 30 ø, and 45 ø (+ iø)C for 90 days. At predetermined time intervals (0, 14, 30, 60, and 90 days), 1.5 g of sample was withdrawn and mixed with 1.5 g of DI water, then centrifuged at 150,000 g (4øC) for 90 min in a Centrikron T-1180 ultracentrifuge (Kontron Instruments, Milan, Italy). The supernatant was removed and diluted (100 times) with DI water prior to the determination of the amount of the unentrapped TA. The pellet was dissolved in 5 ml of 10% Triton-X 100 solution and sonicated for 20 min. This solution was further diluted (ten times) with Triton-X 100 solution prior to the determination of the amount of entrapped TA in liposomes. This study was per- formed to estimate the leakage rate of entrapped TA from various liposome formulations. To characterize the chemical stability of total TA in liposomes, samples (0.25 g) with- drawn at predetermined time intervals (see above) were dissolved in 12.25 ml of 10% Triton-X 100 solution and sonicated for 20 min. These solutions were further diluted (3.33 times) with Triton-X 100 solution prior to the determination of the total amount of TA in liposomes. An amount of 0.1 ml of the above samples (unentrapped, entrapped, and total TA) was withdrawn and derivatized, following the procedures described in the Analytical Method section below, prior to analysis. This study was conducted in six replicates. RELEASE STUDY Various liposome formulations, 5% and 10% TA solutions in DI water, were used as samples in the release study. The vertical Franz diffusion cells (Crown Bio Scientific, Inc., Somerville, NJ) were set at 37 ø + iøC, and the receiver chamber was filled with 12 ml of DI water. A molecular porous membrane tubing with molecular weight cutoff of 12,000-14,000 was fixed in the diffusion cell (contact area, 1.77 cm2). The membrane was soaked prior to use overnight in DI water. Each liposome formulation (3.0 g) was centrifuged at 150,000 g (4øC) for 90 min in an ultracentrifuge, and the supernatant was discarded. The pellet containing liposomes was resuspended in 6.0 ml of DI water, and the dispersion sample (2.0 ml) was loaded into the donor chamber. At predetermined time intervals (0.25, 0.5, 1, 2, 4, 6, 8, and 24 h), 0.5 ml of sample in the receiver chamber was withdrawn and then assayed spectrophotometrically for TA content, fol- lowing the procedures described in the Analytical Method section below. After each sampling, 0.5 ml of DI water was added to the receiver chamber to replace the loss of receiver medium. The correction factor of this dilution was used to calculate the TA concentration of the next sample. This study was performed in six replicates.