- Email: [email protected]
Future drug delivery research in South Korea
Journal of Controlled Release 62 (1999) 73–79 www.elsevier.com / locate / jconrel
Future drug delivery research in South Korea Suk-Jae Chung Department of Pharmaceutics, Seoul National University, San 56 -1, Shinlim-dong, Kwanak-gu, Seoul 151 -742, South Korea
Abstract According to the Science Citation Index database, over 50 papers related to drug delivery have been published from South Korea during the last 3 years. For the purpose of this presentation, some of the recently carried out research in our department will be introduced and future research directions presented. Proliposomes are free flowing particles which are composed of drugs, phospholipids and a water soluble porous powder, and immediately form a liposomal dispersion upon hydration. The preparation can be stored sterilized in a dried state and, by controlling the size of the porous powder in proliposomes, relatively narrow range of reconstituted liposome size can be obtained. Because of these properties, proliposomes appear to be a potential alternative to liposomes in design and fabrication of liposomal dosage forms. Thus, we tested the feasibility of this preparation as a sustained transdermal dosage form. Proliposomes containing varying amount of nicotine, a model drug, were prepared using sorbitol and lecithin. Microscopic observation revealed that this preparation is converted to liposomes almost completely within minutes following contact with water. The release pattern of the model drug from this preparation was apparently similar to that of the Exodus patch, a commercially available transdermal nicotine formulation. Compared with nicotine powder, the nicotine flux across rat skin from proliposomes was initially retarded, and appeared to remain constant. This observation indicated that sustained transdermal delivery of nicotine is feasible using proliposomal formulations under occluded condition. In addition to the investigation of the potential application of proliposomes, we were also interested in the role of the stratum corneum (SC) in the enhanced delivery of drugs from liposomes. Thus, liposome–gel formulation containing hydrocortisone (HC), a model hydrophobic drug, was prepared and used in this study. The study was carried out on both normal and stratum corneum removed skins. Percutaneous absorption of HC across SC removed skin was significantly faster than that across normal skin, suggesting that SC behaves as a penetration barrier for the liposome-bound drugs. Interestingly, the liposome–gel in this study reduced the skin absorption of HC, compared with the conventional ointment formulation. The amount of HC absorbed from the liposome–gel after 8 h into the SC-removed skin was less than one third of that from the conventional ointment. Despite the reduced absorption, a higher and sustained skin concentrations of HC were achieved for the liposome–gel. Drug concentration in both viable and deep skin reached a maximum within 0.5 h. However, drug concentrations in these tissues declined as a function of time for conventional ointment, while those from the liposome–gel were greatly sustained, resulting in a 5-fold higher viable skin drug level was obtained at 8 h after the application. In contrast, plasma concentration of HC at 4 h from the liposome–gel was only one-fourth the value from the conventional ointment in the SC-removed skin. Therefore, the higher and sustained drug concentration in the viable skin appeared not to be due to the enhanced percutaneous absorption but due to retarded diffusion of the drug from the skin. 1999 Published by Elsevier Science B.V. All rights reserved.
Keywords: Drug delivery; South Korea; Proliposomes; Review; Recent advances
0168-3659 / 99 / $ – see front matter 1999 Published by Elsevier Science B.V. All rights reserved. PII: S0168-3659( 99 )00025-5
74
S.-J. Chung / Journal of Controlled Release 62 (1999) 73 – 79
1. Introduction A broad range of research that are related to drug delivery area is currently undergoing in South Korea. As a result, over 50 papers, according to the Science Citation data base, has been published from South Korea for researches related to drug delivery during the last 3 years. These papers covered topics ranging from the material sciences of delivery systems to biopharmaceutics and, thus, it is difficult to project a precise direction of future drug delivery research in South Korea. In this paper, therefore, some of the recently carried out research in our department will be introduced and future research directions presented. Proliposomes are free-flowing particles, composed of drug, phospholipid and a water soluble porous powder, which immediately form a liposomal dispersion upon hydration [1,2]. When proliposomes are applied on mucosal membranes, they are expected to form liposomes upon hydration by mucosal fluids. The resulting liposomes may act as a sustained release dosage form of the loaded-drugs. In this presentation, we examined whether proliposomes can be used as a transdermal delivery system for drugs. Thus, nicotine was selected as a model drug to be delivered in a sustained fashion since it is absorbed fairly well and eliminated rapidly when applied topically [3] and transdermal nicotine therapy is widely used to aid smoking cessation. In the second part of the presentation, we were interested in the mechanism of enhanced delivery of drugs from liposomes. Many investigators [4] have suggested that liposomes enhance percutaneous penetration of drugs through an interaction of liposomes with stratum corneum [5,6] and that this may explain the higher drug concentrations in the skin. This explanation, however, is not valid when higher drug concentrations are achieved by retarded diffusion of the skin-absorbed drug to the systemic body than by the enhanced percutaneous drug absorption. This study was carried out to verify whether the higher drug concentration in the skin following application of liposomes is due to enhanced drug absorption to the skin or to retarded systemic diffusion of the skin-absorbed drug. Hydrocortisone (HC) was selected in this study as a model drug which requires localized topical delivery beneath the skin with
minimal systemic absorption [7]. A hydrogel formulation of liposomes (liposome–gel) was prepared and the result was compared with parallel studies using a conventional ointment formulation.
2. Materials and methods
2.1. Materials Nicotine was purchased from Fluka AG (Switzerland). Egg lecithin (type X-E from dried egg yolk, the phosphatidylcholine content of approximately 60% w / w), HC and cholesterol were purchased from Sigma (St. Louis, MO). Sorbitol was purchased from Junsei Chemical Co. (Tokyo, Japan). 1,2,6,7- 3 H-bHC (78.4 Ci / mmol) was obtained from NEN Research Product (Boston, MA). Carbopol 934 was obtained from BF Goodrich (Cleveland, OH).
2.2. Preparation of nicotine-loaded proliposomes A modified rotary evaporation unit (Eyela, Tokyo Rikakikai Co., Tokyo, Japan) was used for the preparation as described earlier [1] using sorbitol (10 g, particle size of 105–350 mm), egg lecithin (1 g) and nicotine (83 mg or 162 mg). The proliposomes fraction having a 105–350-mm diameter was collected using appropriate sieves. The proliposome particles were almost free-flowing.
2.3. Release of nicotine across the semipermeable membrane Release of nicotine across a semipermeable membrane from the nicotine-containing proliposomes was determined using a USP dissolution apparatus (DST200, Fine Instrument, Seoul, South Korea) equipped with a rotating paddle. Of nicotine-loaded proliposomes, 1 g was put into a clean dialysis bag (Spectra / Por 2 membrane, M.W. cut off of 12 000– 14 000; Spectra Medical Ind., Los Angeles, CA). The bag was secured with two clamps at each end to yield a rectangular shape of 3.935.0 cm and placed into the release apparatus containing 500 ml phosphate buffer (pH 7.4). The buffer was kept at 37618C and stirred with the paddle at 100 rpm. One-milliliter aliquots of the medium were sampled
S.-J. Chung / Journal of Controlled Release 62 (1999) 73 – 79
at various times (fresh medium was added as a replacement). For comparison, nicotine powder (1 g), instead of proliposomes, was placed in the dialysis bag and the release of nicotine across the bag was tested in the same manner. Release of nicotine from a commercial nicotine patch, Exodus (Elan Pharm. Ireland, nicotine 30 mg, transpassing area: 7 cm 2 ), was also determined by the paddle over disk method of USP for transdermal patches [8].
2.4. In vitro skin permeation study 2.4.1. Preparation of rat skin Abdominal skin of Wistar male rats was used in the study [9]. The skin was stored at 2208C until permeation study. Before the permeation study, the skin was hydrated in normal saline (contained 200 ppm gentamicin) at 48C and adipose tissue layer of the skin was removed by rubbing with a cotton swab. 2.4.2. Permeation of nicotine from proliposomes under occlusive condition To simulate in vivo situation where proliposomes are applied onto the surface of skin under occlusive condition, nicotine powder (2.0 mg) or proliposome particles equivalent to 2.0 mg of nicotine were evenly spread on the skin surface of circular area of 2.16 cm 2 and covered tightly with an occlusive film (Transpaseal , Porcupine Canvas, Ontario, Canada) [9]. The proliposome-loaded skin was mounted carefully on the Keshary–Chien diffusion cells (K.C. Scientific Co., South Korea). The stratum corneum side and dermal side of the skin were located in order to face the donor compartment and receptor compartment, respectively. Effective transpassing area of the diffusion cell was 2.16 cm 2 . Receptor compartment was filled with 10-ml of phosphate buffer (pH 7.4), and the buffer was stirred by a magnetic stirrer rotating at 600 rpm and kept at 37618C. No buffer was added into the donor compartment and, thus, the stratum corneum side of the skin was exposed to atmosphere. One hundred-microliter aliquot was sampled (fresh medium was replaced) various times from the receptor compartment. The samples were stored in a freezer prior to analysis of nicotine.
75
2.5. Preparation of liposome–gel containing HC Liposomes containing radiolabelled HC were prepared by the conventional solvent evaporation method using egg phosphatidylcholine (20 mg), cholesterol (1 mg), HC (1 mg) and 3 H-HC (2 mCi in 2 ml) [10]. This process resulted in liposomes of mean particle size distribution (mean6S.E. of 223.269.7 nm; n53 batches). Liposome-containing Carbopol gel was prepared according to Kim et al. [10]. The homogeneity of the blending was confirmed by the even distribution of radioactivity in the gel. The approximate concentration of HC and radioactivity in the final liposome– gel were 1.0% (w / w) and 5310 7 dpm per gram, respectively.
2.6. Preparation of conventional HC ointment A commercial ointment of HC was utilized instead for the comparison study. A 100-mg amount of commercial HC ointment (1.0% w / w, HC; Jinro Co. Seoul, South Korea) was blended homogeneously with 2 ml of 3 H-HC (1.0 mCi / ml). The major components of the ointment were polyethylene glycol 6000 (10% w / w), propylene glycol (10% w / w) and paraffin (6% w / w). The final concentration of HC and radioactivity in the ointment were identical to those of the liposome–gel.
2.7. Removal of stratum corneum layer by a stripping method Female hairless mice (6–7 weeks old, 2065 g; Charles River, USA) fed on a standard laboratory diet and allowed tap water ad libitum were used. Each mouse was anesthetized with 0.5% pentobarbital saline solution injected intraperitoneally at a dose of 60 mg / kg. The stratum corneum layer of abdominal skin was removed according to the reported stripping method [11].
2.8. Application of liposome–gel and conventional ointment to the skin Each preparation (liposome–gel or ointment) was spread on area of 3.0 cm 2 (1.532 cm) of the normal or stripped skin. The dose of each preparation was
76
S.-J. Chung / Journal of Controlled Release 62 (1999) 73 – 79
adjusted to 0.1 g in weight, which contained 1 mg of HC and 2 mCi of radioactivity.
2.9. Sampling of viable tissue, deep skin and plasma Approximately 0.1 g of the viable tissue (epidermis plus dermis) and deep skin (subcutaneous tissue plus muscle) were sampled at various times after the dosing. Upon completion of the viable tissue collection, the mouse was sacrificed by abdominal artery bleeding. Blood was collected through the abdominal artery and plasma sample was obtained from a centrifugation of blood sample. Skin and plasma samples were weighed and placed in scintillation vials containing 0.5 ml of Solvable (NEN Research Product, Boston). Since the extent of skin metabolism for HC was less than 10% of the dose [12], the drug concentration in the samples could be estimated, with reasonable accuracy by measurement of total radioactivity. The radioactivity in the samples was measured by liquid scintillation counting (System 1400, Wallac, Finland).
transmission electron microscopy of the liposomes formation from the proliposomes is indicated that particles having approximate diameter of 100 nm are observed (data not shown). Therefore, we conclude that proliposomes are converted to liposomes in the presence of water. The entrapment efficiency (EE) of nicotine in liposomes varied from 45.1 (63.8) to 57.9 (63.4)% as the composition (lecithin / sorbitol / nicotine) of the proliposomes changed from 1:10:0.162 and 1:10:0.083, respectively
3.2. Release of nicotine across the semipermeable membrane Fig. 1 shows the percentage of nicotine released across the semipermeable membrane from the nicotine powder and nicotine-loaded proliposomes into a phosphate buffer (pH 7.4) as a function of time. Release of nicotine from the powder was rapid and reached approximately 90% of dose in 1 h. But that from the proliposomes was significantly retarded, indicating proliposomes can be a sustained release dosage form of nicotine. Despite of the sustained release, no significant time lag was observed for the release from the proliposomes. It may be due to
3. Results and discussion
3.1. Characteristics of nicotine-containing proliposomes and reconstituted liposomes Nicotine contents in the proliposomes were 1.37 and 0.69% (w / w) for the proliposomes of different lecithin / sorbitol / nicotine ratio (1:10:0.162 and 1:10:0.083, respectively). This represents that more than 90% (94.4 and 92.1%, respectively) of added nicotine was recovered in the proliposomes. Surface morphology of proliposomes of 105–350mm fraction was similar to that of the untreated sorbitol of the same particle size by scanning electron microscopy (data not shown). Observation under an optical microscopy revealed that, upon hydration, proliposome particles are progressively, but rapidly (i.e. in less than 30 s), converted to form a semi-transparent mixture in water. Liposome formation was supported by the size of the reconstituted particles in the mixture which was consistent with the size of liposomes [9]. In addition to the particle size determination study,
Fig. 1. Release of nicotine through a cellulose membrane (M.W. cut-off, 12 000–14 000) to the receptor fluid (phosphate buffer, pH 7.4, 10 ml, 37618C) from 1 g of nicotine powder (filled square), proliposomes (filled circle: lecithin / sorbitol / nicotine5 1:10:0.162; empty circle: lecithin / sorbitol / nicotine51:10:0.083) by USP paddle method (100 rpm). Release of nicotine from Exodus patch (filled triangle) was also tested for comparison using USP paddle over disk method (100 rpm). Each point represents mean6S.E. of three different determinations.
S.-J. Chung / Journal of Controlled Release 62 (1999) 73 – 79
rapid release of nicotine from the proliposomes into the medium which is consistent with the rather low value of EE. Although sustained release of nicotine could be achieved from the proliposomes, only 60% of dose was recovered from the release medium. It can be explained by the fact that liposome particles cannot permeate across the semipermeable membrane. Then, part of nicotine dose (approximately 40% of dose in this study) is likely to be retained in the liposomes probably as dissolved or entrapped. The release patterns from both proliposomes were quite similar to that from Exodus , a commercial transdermal delivery system of nicotine. Although the similarity of the release pattern should not be overestimated since the experimental conditions were different for the proliposomes and Exodus , it is noteworthy that release of nicotine from the Exodus also showed a plateau at approximately 60% of dose. Entrapment of nicotine, probably in vehicles composing the patch, may be responsible for the incomplete release.
3.3. In vitro skin permeation study Fig. 2 shows the cumulative amount of nicotine
77
penetrated from nicotine powder and proliposomes to the receptor compartment (pH 7.4 phosphate buffer) of the Keshary–Chien diffusion cell. Throughout the experimental period, the powder showed much larger amount of penetration than the proliposomes. Initial flux of nicotine (i.e. the slope of Fig. 2) from the powder was more than twice of the proliposome preparations (i.e. 172 mg / cm 2 / h for powder vs. 73 mg / cm 2 / h for the proliposomes). However, the difference in flux was apparently reduced after 4 h. For the proliposomes, the nicotine flux was kept almost constant in all sampling times except initial period (between 0 and 1 h). This result suggests a feasibility of proliposomes as a sustained transdermal delivery system.
3.4. Properties of liposomes and liposome–gel The entrapment efficiency (EE) of HC in the liposomes was 44.563.6% (n55). There were no significant changes in vesicle size of the liposomes by incorporation into Carbopol gel. The liposomes in the gel were 276.4621.5 nm (mean6S.D.) in size of the vesicles. These liposomes and the liposome–gels exhibited no changes in vesicle size distribution when stored at room temperature for over 6 months.
3.5. Drug concentration in viable skin
Fig. 2. Cumulative amount (mean6S.E.) of nicotine penetrated to the receptor fluid (phosphate buffer, pH 7.4, 10 ml, 37618C) across the rat skin (2.16 cm 2 ) from nicotine powder (2 mg) or proliposomes (equivalent to 2 mg of nicotine) in the Keshary– Chien diffusion cell. Drug was loaded on the stratum corneum side of the skin, which was exposed to the atmosphere during the experiment. The receptor fluid was stirred at 600 rpm. Each point represents mean6S.E. of three different determinations. Key: nicotine powder, empty square; proliposomes, lecithin / sorbitol / nicotine51:10:0.162, filled circle; nicotine-containing lecithin / sorbitol / nicotine51:10:0.083, empty circle.
Fig. 3 shows the drug concentration or amount– time profile in viable skin (epidermis plus dermis) following application of the liposome–gel and conventional ointment on the normal and stripped skins. Drug concentrations across the stripped skin were much higher than those across the normal skin again implicating the barrier function of stratum corneum for the percutaneous absorption of the drug. Drug concentration, after the application of the ointment, reached its maximum from the first sampling point, 0.5 h, and, decreased with time, while that of the liposome–gel continued to increase for over 8 h. As a result, the drug concentration in the viable skin from the liposome–gel became 2.5-fold (P,0.05) and 5-fold higher (P,0.01) than that from ointmenttreated skin at 4 and 8 h respectively. Drug concentrations in the viable skin from liposome–gel was almost identical to that from the ointment at the first sampling point (0.5 h). This
78
S.-J. Chung / Journal of Controlled Release 62 (1999) 73 – 79
Fig. 3. Hydrocortisone concentration–time profiles in viable skin following topical application liposome–gel (filled circle) and conventional ointment (empty circle) to the normal (? ? ?) and stratum corneum-removed (—) skin. All data are expressed as mean6S.E. of four experiments.
suggests that the rate of initial partition of the drug from both formulations into the viable skin is similar. Since drug absorption from the liposome– gel into the skin was much smaller than that from the ointment [10], the sustained drug concentration from the liposome–gel could be attributed to the retarded diffusion of the drug in the viable skin into the deep skin or blood stream.
3.6. Drug concentration in deep skin Fig. 4 shows the drug concentration–time profile in the deep skin (subcutaneous tissue plus muscle) following application of the liposome–gel and conventional ointment on stripped skins. The concentrations following topical application on normal skin were below the detection limit for both formulations and, thus, are not shown in the figure. The concentration decreased almost linearly with time for both formulations, but the rate of decrease from the liposome–gel was much lower than that from the ointment. The rate of decrease was nearly the same
Fig. 4. Hydrocortisone concentration–time profiles in deep skin following application of liposome–gel (filled circle) and conventional ointment (empty circle) on the stripped skin. All data are expressed as mean6S.E. of four experiments.
for both viable and deep skin which had been treated with ointment, but, for the case of the liposome–gel treatment, it was much slower in the viable skin (Fig. 3) than in the deep skin (Fig. 4). This may indicate that the liposome–gel retards drug diffusion more in the viable skin than in the deep skin.
3.7. Systemic absorption of the drug In addition to higher and sustained drug concentration in the skin, the absence of significant systemic absorption of the drug is desirable, in order to treat skin pathology without significant side-effects. Fig. 5 shows the plasma drug concentration– time profiles following topical application of the two formulations on the stripped skin. No detectable HC was found in the plasma following application to the normal skin. The concentration from the ointment decreased almost log-linearly with time, while that from the liposome–gel initially increased and then decreased with time, showing a maximum at 4 h. At 0.5 h, the concentration from the ointment was 4-fold higher (P,0.01) than that from the liposome–gel.
S.-J. Chung / Journal of Controlled Release 62 (1999) 73 – 79
79
References
Fig. 5. Plasma hydrocortisone profiles following application of liposome–gel (filled circle) and conventional ointment (empty circle) on the stripped skin. All data are expressed as mean6S.E. of four experiments.
4. Conclusion For drug delivery system involving proliposomes, applicability of the system in various routes of administration (e.g. nasal delivery) is our short-term research interest. For liposome delivery system, the mechanism of the retarded diffusion of drugs from the liposome is currently under investigation. However, because government spending on basic researches is expected to decrease because of the current economic difficulty in South Korea, research on drug delivery is likely to be affected to a certain extent. As a result, private funding will play a crucial role for drug delivery research in the next few years in South Korea.
[1] N.I. Payne, I. Browning, C.A. Hynes, Characterization of proliposomes, J. Pharm. Sci. 75 (1986) 330–333. [2] N.I. Payne, P. Timmis, C.V. Ambrose, M.D. Warel, F. Ridgway, Proliposomes: a novel solution to an old problem, J. Pharm. Sci. 75 (1986) 325–329. [3] J.H. Jaffe, Drug addiction and drug abuse, in: A.G. Goodman, T.W. Rall, A.S. Nies, P. Taylor (Eds.), The Pharmacological Basis of Therapeutics, 8th ed., Maxwell MacMillan, London, 1991, pp. 522–573. [4] C. Michel, T. Purmann, E. Mentrup, E. Seiller, J. Kreuter, Effect of liposomes on percutaneous penetration of lipophilic materials, Int. J. Pharm. 84 (1992) 93–105. [5] A. Blume, M. Jansen, M. Ghyczy, J. Gareiss, Interaction of phospholipid liposomes with lipid model mixtures for stratum corneum lipids, Int. J. Pharm. 99 (1993) 219–228. [6] J.A. Bouwstra, H.E. Hofland, G.S. Gooris, F. Spies, H.E. Junginger, Structural changes in human stratum corneum induced by liposomes, Proc. Int. Symp. Control. Release Bioact. Mater. 18 (1991) 305–306. [7] A. Goodman, L.S. Gilman, Adrenocortical steroids, in: A. Goodman, L.S. Gilman, T.W. Rall, F. Murad (Eds.), The Pharmacological Basis of Therapeutics, Macmillan, New York, 1985, pp. 1471–1485. [8] USP XXII, The United States Pharmacopeial Convention, Inc., Mack Printing Company, Easton, PA, 1990, pp. 1581– 1583. [9] B.-Y. Y Hwang, B.-W. Jung, S.-J. Chung, M.-H. Lee, C.-K. Shim, In vitro skin permeation of nicotine from proliposomes, J. Control. Release 49 (1997) 177–184. [10] M.K. Kim, S.-J. Chung, M.-H. Lee, A.-R. Cho, C.-K. Shim, Targeted and sustained delivery of hydrocortisone to normal and stratum corneum-removed skin without enhanced skin absorption using a liposome gel, J. Control. Rel. 46 (1997) 243–251. [11] A. Rougier, C. Lotte, The stripping technique, in: V.P. Shah, H.I. Mailbach (Eds.), Topical Drug Bioavailability, Bioequivalence and Penetration, Plenum Press, New York, 1993, p. 180. [12] S.L. Hsia, A.J. Mussallem, V.H. Witten, Further metabolic studies of hydrocortisone- 14 C in human skin: Requirement for pyridine nucleotides and sites of metabolism, J. Invest. Dermatol. 45 (1965) 384–390.
Future drug delivery research in South Korea Suk-Jae Chung Department of Pharmaceutics, Seoul National University, San 56 -1, Shinlim-dong, Kwanak-gu, Seoul 151 -742, South Korea
Abstract According to the Science Citation Index database, over 50 papers related to drug delivery have been published from South Korea during the last 3 years. For the purpose of this presentation, some of the recently carried out research in our department will be introduced and future research directions presented. Proliposomes are free flowing particles which are composed of drugs, phospholipids and a water soluble porous powder, and immediately form a liposomal dispersion upon hydration. The preparation can be stored sterilized in a dried state and, by controlling the size of the porous powder in proliposomes, relatively narrow range of reconstituted liposome size can be obtained. Because of these properties, proliposomes appear to be a potential alternative to liposomes in design and fabrication of liposomal dosage forms. Thus, we tested the feasibility of this preparation as a sustained transdermal dosage form. Proliposomes containing varying amount of nicotine, a model drug, were prepared using sorbitol and lecithin. Microscopic observation revealed that this preparation is converted to liposomes almost completely within minutes following contact with water. The release pattern of the model drug from this preparation was apparently similar to that of the Exodus patch, a commercially available transdermal nicotine formulation. Compared with nicotine powder, the nicotine flux across rat skin from proliposomes was initially retarded, and appeared to remain constant. This observation indicated that sustained transdermal delivery of nicotine is feasible using proliposomal formulations under occluded condition. In addition to the investigation of the potential application of proliposomes, we were also interested in the role of the stratum corneum (SC) in the enhanced delivery of drugs from liposomes. Thus, liposome–gel formulation containing hydrocortisone (HC), a model hydrophobic drug, was prepared and used in this study. The study was carried out on both normal and stratum corneum removed skins. Percutaneous absorption of HC across SC removed skin was significantly faster than that across normal skin, suggesting that SC behaves as a penetration barrier for the liposome-bound drugs. Interestingly, the liposome–gel in this study reduced the skin absorption of HC, compared with the conventional ointment formulation. The amount of HC absorbed from the liposome–gel after 8 h into the SC-removed skin was less than one third of that from the conventional ointment. Despite the reduced absorption, a higher and sustained skin concentrations of HC were achieved for the liposome–gel. Drug concentration in both viable and deep skin reached a maximum within 0.5 h. However, drug concentrations in these tissues declined as a function of time for conventional ointment, while those from the liposome–gel were greatly sustained, resulting in a 5-fold higher viable skin drug level was obtained at 8 h after the application. In contrast, plasma concentration of HC at 4 h from the liposome–gel was only one-fourth the value from the conventional ointment in the SC-removed skin. Therefore, the higher and sustained drug concentration in the viable skin appeared not to be due to the enhanced percutaneous absorption but due to retarded diffusion of the drug from the skin. 1999 Published by Elsevier Science B.V. All rights reserved.
Keywords: Drug delivery; South Korea; Proliposomes; Review; Recent advances
0168-3659 / 99 / $ – see front matter 1999 Published by Elsevier Science B.V. All rights reserved. PII: S0168-3659( 99 )00025-5
74
S.-J. Chung / Journal of Controlled Release 62 (1999) 73 – 79
1. Introduction A broad range of research that are related to drug delivery area is currently undergoing in South Korea. As a result, over 50 papers, according to the Science Citation data base, has been published from South Korea for researches related to drug delivery during the last 3 years. These papers covered topics ranging from the material sciences of delivery systems to biopharmaceutics and, thus, it is difficult to project a precise direction of future drug delivery research in South Korea. In this paper, therefore, some of the recently carried out research in our department will be introduced and future research directions presented. Proliposomes are free-flowing particles, composed of drug, phospholipid and a water soluble porous powder, which immediately form a liposomal dispersion upon hydration [1,2]. When proliposomes are applied on mucosal membranes, they are expected to form liposomes upon hydration by mucosal fluids. The resulting liposomes may act as a sustained release dosage form of the loaded-drugs. In this presentation, we examined whether proliposomes can be used as a transdermal delivery system for drugs. Thus, nicotine was selected as a model drug to be delivered in a sustained fashion since it is absorbed fairly well and eliminated rapidly when applied topically [3] and transdermal nicotine therapy is widely used to aid smoking cessation. In the second part of the presentation, we were interested in the mechanism of enhanced delivery of drugs from liposomes. Many investigators [4] have suggested that liposomes enhance percutaneous penetration of drugs through an interaction of liposomes with stratum corneum [5,6] and that this may explain the higher drug concentrations in the skin. This explanation, however, is not valid when higher drug concentrations are achieved by retarded diffusion of the skin-absorbed drug to the systemic body than by the enhanced percutaneous drug absorption. This study was carried out to verify whether the higher drug concentration in the skin following application of liposomes is due to enhanced drug absorption to the skin or to retarded systemic diffusion of the skin-absorbed drug. Hydrocortisone (HC) was selected in this study as a model drug which requires localized topical delivery beneath the skin with
minimal systemic absorption [7]. A hydrogel formulation of liposomes (liposome–gel) was prepared and the result was compared with parallel studies using a conventional ointment formulation.
2. Materials and methods
2.1. Materials Nicotine was purchased from Fluka AG (Switzerland). Egg lecithin (type X-E from dried egg yolk, the phosphatidylcholine content of approximately 60% w / w), HC and cholesterol were purchased from Sigma (St. Louis, MO). Sorbitol was purchased from Junsei Chemical Co. (Tokyo, Japan). 1,2,6,7- 3 H-bHC (78.4 Ci / mmol) was obtained from NEN Research Product (Boston, MA). Carbopol 934 was obtained from BF Goodrich (Cleveland, OH).
2.2. Preparation of nicotine-loaded proliposomes A modified rotary evaporation unit (Eyela, Tokyo Rikakikai Co., Tokyo, Japan) was used for the preparation as described earlier [1] using sorbitol (10 g, particle size of 105–350 mm), egg lecithin (1 g) and nicotine (83 mg or 162 mg). The proliposomes fraction having a 105–350-mm diameter was collected using appropriate sieves. The proliposome particles were almost free-flowing.
2.3. Release of nicotine across the semipermeable membrane Release of nicotine across a semipermeable membrane from the nicotine-containing proliposomes was determined using a USP dissolution apparatus (DST200, Fine Instrument, Seoul, South Korea) equipped with a rotating paddle. Of nicotine-loaded proliposomes, 1 g was put into a clean dialysis bag (Spectra / Por 2 membrane, M.W. cut off of 12 000– 14 000; Spectra Medical Ind., Los Angeles, CA). The bag was secured with two clamps at each end to yield a rectangular shape of 3.935.0 cm and placed into the release apparatus containing 500 ml phosphate buffer (pH 7.4). The buffer was kept at 37618C and stirred with the paddle at 100 rpm. One-milliliter aliquots of the medium were sampled
S.-J. Chung / Journal of Controlled Release 62 (1999) 73 – 79
at various times (fresh medium was added as a replacement). For comparison, nicotine powder (1 g), instead of proliposomes, was placed in the dialysis bag and the release of nicotine across the bag was tested in the same manner. Release of nicotine from a commercial nicotine patch, Exodus (Elan Pharm. Ireland, nicotine 30 mg, transpassing area: 7 cm 2 ), was also determined by the paddle over disk method of USP for transdermal patches [8].
2.4. In vitro skin permeation study 2.4.1. Preparation of rat skin Abdominal skin of Wistar male rats was used in the study [9]. The skin was stored at 2208C until permeation study. Before the permeation study, the skin was hydrated in normal saline (contained 200 ppm gentamicin) at 48C and adipose tissue layer of the skin was removed by rubbing with a cotton swab. 2.4.2. Permeation of nicotine from proliposomes under occlusive condition To simulate in vivo situation where proliposomes are applied onto the surface of skin under occlusive condition, nicotine powder (2.0 mg) or proliposome particles equivalent to 2.0 mg of nicotine were evenly spread on the skin surface of circular area of 2.16 cm 2 and covered tightly with an occlusive film (Transpaseal , Porcupine Canvas, Ontario, Canada) [9]. The proliposome-loaded skin was mounted carefully on the Keshary–Chien diffusion cells (K.C. Scientific Co., South Korea). The stratum corneum side and dermal side of the skin were located in order to face the donor compartment and receptor compartment, respectively. Effective transpassing area of the diffusion cell was 2.16 cm 2 . Receptor compartment was filled with 10-ml of phosphate buffer (pH 7.4), and the buffer was stirred by a magnetic stirrer rotating at 600 rpm and kept at 37618C. No buffer was added into the donor compartment and, thus, the stratum corneum side of the skin was exposed to atmosphere. One hundred-microliter aliquot was sampled (fresh medium was replaced) various times from the receptor compartment. The samples were stored in a freezer prior to analysis of nicotine.
75
2.5. Preparation of liposome–gel containing HC Liposomes containing radiolabelled HC were prepared by the conventional solvent evaporation method using egg phosphatidylcholine (20 mg), cholesterol (1 mg), HC (1 mg) and 3 H-HC (2 mCi in 2 ml) [10]. This process resulted in liposomes of mean particle size distribution (mean6S.E. of 223.269.7 nm; n53 batches). Liposome-containing Carbopol gel was prepared according to Kim et al. [10]. The homogeneity of the blending was confirmed by the even distribution of radioactivity in the gel. The approximate concentration of HC and radioactivity in the final liposome– gel were 1.0% (w / w) and 5310 7 dpm per gram, respectively.
2.6. Preparation of conventional HC ointment A commercial ointment of HC was utilized instead for the comparison study. A 100-mg amount of commercial HC ointment (1.0% w / w, HC; Jinro Co. Seoul, South Korea) was blended homogeneously with 2 ml of 3 H-HC (1.0 mCi / ml). The major components of the ointment were polyethylene glycol 6000 (10% w / w), propylene glycol (10% w / w) and paraffin (6% w / w). The final concentration of HC and radioactivity in the ointment were identical to those of the liposome–gel.
2.7. Removal of stratum corneum layer by a stripping method Female hairless mice (6–7 weeks old, 2065 g; Charles River, USA) fed on a standard laboratory diet and allowed tap water ad libitum were used. Each mouse was anesthetized with 0.5% pentobarbital saline solution injected intraperitoneally at a dose of 60 mg / kg. The stratum corneum layer of abdominal skin was removed according to the reported stripping method [11].
2.8. Application of liposome–gel and conventional ointment to the skin Each preparation (liposome–gel or ointment) was spread on area of 3.0 cm 2 (1.532 cm) of the normal or stripped skin. The dose of each preparation was
76
S.-J. Chung / Journal of Controlled Release 62 (1999) 73 – 79
adjusted to 0.1 g in weight, which contained 1 mg of HC and 2 mCi of radioactivity.
2.9. Sampling of viable tissue, deep skin and plasma Approximately 0.1 g of the viable tissue (epidermis plus dermis) and deep skin (subcutaneous tissue plus muscle) were sampled at various times after the dosing. Upon completion of the viable tissue collection, the mouse was sacrificed by abdominal artery bleeding. Blood was collected through the abdominal artery and plasma sample was obtained from a centrifugation of blood sample. Skin and plasma samples were weighed and placed in scintillation vials containing 0.5 ml of Solvable (NEN Research Product, Boston). Since the extent of skin metabolism for HC was less than 10% of the dose [12], the drug concentration in the samples could be estimated, with reasonable accuracy by measurement of total radioactivity. The radioactivity in the samples was measured by liquid scintillation counting (System 1400, Wallac, Finland).
transmission electron microscopy of the liposomes formation from the proliposomes is indicated that particles having approximate diameter of 100 nm are observed (data not shown). Therefore, we conclude that proliposomes are converted to liposomes in the presence of water. The entrapment efficiency (EE) of nicotine in liposomes varied from 45.1 (63.8) to 57.9 (63.4)% as the composition (lecithin / sorbitol / nicotine) of the proliposomes changed from 1:10:0.162 and 1:10:0.083, respectively
3.2. Release of nicotine across the semipermeable membrane Fig. 1 shows the percentage of nicotine released across the semipermeable membrane from the nicotine powder and nicotine-loaded proliposomes into a phosphate buffer (pH 7.4) as a function of time. Release of nicotine from the powder was rapid and reached approximately 90% of dose in 1 h. But that from the proliposomes was significantly retarded, indicating proliposomes can be a sustained release dosage form of nicotine. Despite of the sustained release, no significant time lag was observed for the release from the proliposomes. It may be due to
3. Results and discussion
3.1. Characteristics of nicotine-containing proliposomes and reconstituted liposomes Nicotine contents in the proliposomes were 1.37 and 0.69% (w / w) for the proliposomes of different lecithin / sorbitol / nicotine ratio (1:10:0.162 and 1:10:0.083, respectively). This represents that more than 90% (94.4 and 92.1%, respectively) of added nicotine was recovered in the proliposomes. Surface morphology of proliposomes of 105–350mm fraction was similar to that of the untreated sorbitol of the same particle size by scanning electron microscopy (data not shown). Observation under an optical microscopy revealed that, upon hydration, proliposome particles are progressively, but rapidly (i.e. in less than 30 s), converted to form a semi-transparent mixture in water. Liposome formation was supported by the size of the reconstituted particles in the mixture which was consistent with the size of liposomes [9]. In addition to the particle size determination study,
Fig. 1. Release of nicotine through a cellulose membrane (M.W. cut-off, 12 000–14 000) to the receptor fluid (phosphate buffer, pH 7.4, 10 ml, 37618C) from 1 g of nicotine powder (filled square), proliposomes (filled circle: lecithin / sorbitol / nicotine5 1:10:0.162; empty circle: lecithin / sorbitol / nicotine51:10:0.083) by USP paddle method (100 rpm). Release of nicotine from Exodus patch (filled triangle) was also tested for comparison using USP paddle over disk method (100 rpm). Each point represents mean6S.E. of three different determinations.
S.-J. Chung / Journal of Controlled Release 62 (1999) 73 – 79
rapid release of nicotine from the proliposomes into the medium which is consistent with the rather low value of EE. Although sustained release of nicotine could be achieved from the proliposomes, only 60% of dose was recovered from the release medium. It can be explained by the fact that liposome particles cannot permeate across the semipermeable membrane. Then, part of nicotine dose (approximately 40% of dose in this study) is likely to be retained in the liposomes probably as dissolved or entrapped. The release patterns from both proliposomes were quite similar to that from Exodus , a commercial transdermal delivery system of nicotine. Although the similarity of the release pattern should not be overestimated since the experimental conditions were different for the proliposomes and Exodus , it is noteworthy that release of nicotine from the Exodus also showed a plateau at approximately 60% of dose. Entrapment of nicotine, probably in vehicles composing the patch, may be responsible for the incomplete release.
3.3. In vitro skin permeation study Fig. 2 shows the cumulative amount of nicotine
77
penetrated from nicotine powder and proliposomes to the receptor compartment (pH 7.4 phosphate buffer) of the Keshary–Chien diffusion cell. Throughout the experimental period, the powder showed much larger amount of penetration than the proliposomes. Initial flux of nicotine (i.e. the slope of Fig. 2) from the powder was more than twice of the proliposome preparations (i.e. 172 mg / cm 2 / h for powder vs. 73 mg / cm 2 / h for the proliposomes). However, the difference in flux was apparently reduced after 4 h. For the proliposomes, the nicotine flux was kept almost constant in all sampling times except initial period (between 0 and 1 h). This result suggests a feasibility of proliposomes as a sustained transdermal delivery system.
3.4. Properties of liposomes and liposome–gel The entrapment efficiency (EE) of HC in the liposomes was 44.563.6% (n55). There were no significant changes in vesicle size of the liposomes by incorporation into Carbopol gel. The liposomes in the gel were 276.4621.5 nm (mean6S.D.) in size of the vesicles. These liposomes and the liposome–gels exhibited no changes in vesicle size distribution when stored at room temperature for over 6 months.
3.5. Drug concentration in viable skin
Fig. 2. Cumulative amount (mean6S.E.) of nicotine penetrated to the receptor fluid (phosphate buffer, pH 7.4, 10 ml, 37618C) across the rat skin (2.16 cm 2 ) from nicotine powder (2 mg) or proliposomes (equivalent to 2 mg of nicotine) in the Keshary– Chien diffusion cell. Drug was loaded on the stratum corneum side of the skin, which was exposed to the atmosphere during the experiment. The receptor fluid was stirred at 600 rpm. Each point represents mean6S.E. of three different determinations. Key: nicotine powder, empty square; proliposomes, lecithin / sorbitol / nicotine51:10:0.162, filled circle; nicotine-containing lecithin / sorbitol / nicotine51:10:0.083, empty circle.
Fig. 3 shows the drug concentration or amount– time profile in viable skin (epidermis plus dermis) following application of the liposome–gel and conventional ointment on the normal and stripped skins. Drug concentrations across the stripped skin were much higher than those across the normal skin again implicating the barrier function of stratum corneum for the percutaneous absorption of the drug. Drug concentration, after the application of the ointment, reached its maximum from the first sampling point, 0.5 h, and, decreased with time, while that of the liposome–gel continued to increase for over 8 h. As a result, the drug concentration in the viable skin from the liposome–gel became 2.5-fold (P,0.05) and 5-fold higher (P,0.01) than that from ointmenttreated skin at 4 and 8 h respectively. Drug concentrations in the viable skin from liposome–gel was almost identical to that from the ointment at the first sampling point (0.5 h). This
78
S.-J. Chung / Journal of Controlled Release 62 (1999) 73 – 79
Fig. 3. Hydrocortisone concentration–time profiles in viable skin following topical application liposome–gel (filled circle) and conventional ointment (empty circle) to the normal (? ? ?) and stratum corneum-removed (—) skin. All data are expressed as mean6S.E. of four experiments.
suggests that the rate of initial partition of the drug from both formulations into the viable skin is similar. Since drug absorption from the liposome– gel into the skin was much smaller than that from the ointment [10], the sustained drug concentration from the liposome–gel could be attributed to the retarded diffusion of the drug in the viable skin into the deep skin or blood stream.
3.6. Drug concentration in deep skin Fig. 4 shows the drug concentration–time profile in the deep skin (subcutaneous tissue plus muscle) following application of the liposome–gel and conventional ointment on stripped skins. The concentrations following topical application on normal skin were below the detection limit for both formulations and, thus, are not shown in the figure. The concentration decreased almost linearly with time for both formulations, but the rate of decrease from the liposome–gel was much lower than that from the ointment. The rate of decrease was nearly the same
Fig. 4. Hydrocortisone concentration–time profiles in deep skin following application of liposome–gel (filled circle) and conventional ointment (empty circle) on the stripped skin. All data are expressed as mean6S.E. of four experiments.
for both viable and deep skin which had been treated with ointment, but, for the case of the liposome–gel treatment, it was much slower in the viable skin (Fig. 3) than in the deep skin (Fig. 4). This may indicate that the liposome–gel retards drug diffusion more in the viable skin than in the deep skin.
3.7. Systemic absorption of the drug In addition to higher and sustained drug concentration in the skin, the absence of significant systemic absorption of the drug is desirable, in order to treat skin pathology without significant side-effects. Fig. 5 shows the plasma drug concentration– time profiles following topical application of the two formulations on the stripped skin. No detectable HC was found in the plasma following application to the normal skin. The concentration from the ointment decreased almost log-linearly with time, while that from the liposome–gel initially increased and then decreased with time, showing a maximum at 4 h. At 0.5 h, the concentration from the ointment was 4-fold higher (P,0.01) than that from the liposome–gel.
S.-J. Chung / Journal of Controlled Release 62 (1999) 73 – 79
79
References
Fig. 5. Plasma hydrocortisone profiles following application of liposome–gel (filled circle) and conventional ointment (empty circle) on the stripped skin. All data are expressed as mean6S.E. of four experiments.
4. Conclusion For drug delivery system involving proliposomes, applicability of the system in various routes of administration (e.g. nasal delivery) is our short-term research interest. For liposome delivery system, the mechanism of the retarded diffusion of drugs from the liposome is currently under investigation. However, because government spending on basic researches is expected to decrease because of the current economic difficulty in South Korea, research on drug delivery is likely to be affected to a certain extent. As a result, private funding will play a crucial role for drug delivery research in the next few years in South Korea.
[1] N.I. Payne, I. Browning, C.A. Hynes, Characterization of proliposomes, J. Pharm. Sci. 75 (1986) 330–333. [2] N.I. Payne, P. Timmis, C.V. Ambrose, M.D. Warel, F. Ridgway, Proliposomes: a novel solution to an old problem, J. Pharm. Sci. 75 (1986) 325–329. [3] J.H. Jaffe, Drug addiction and drug abuse, in: A.G. Goodman, T.W. Rall, A.S. Nies, P. Taylor (Eds.), The Pharmacological Basis of Therapeutics, 8th ed., Maxwell MacMillan, London, 1991, pp. 522–573. [4] C. Michel, T. Purmann, E. Mentrup, E. Seiller, J. Kreuter, Effect of liposomes on percutaneous penetration of lipophilic materials, Int. J. Pharm. 84 (1992) 93–105. [5] A. Blume, M. Jansen, M. Ghyczy, J. Gareiss, Interaction of phospholipid liposomes with lipid model mixtures for stratum corneum lipids, Int. J. Pharm. 99 (1993) 219–228. [6] J.A. Bouwstra, H.E. Hofland, G.S. Gooris, F. Spies, H.E. Junginger, Structural changes in human stratum corneum induced by liposomes, Proc. Int. Symp. Control. Release Bioact. Mater. 18 (1991) 305–306. [7] A. Goodman, L.S. Gilman, Adrenocortical steroids, in: A. Goodman, L.S. Gilman, T.W. Rall, F. Murad (Eds.), The Pharmacological Basis of Therapeutics, Macmillan, New York, 1985, pp. 1471–1485. [8] USP XXII, The United States Pharmacopeial Convention, Inc., Mack Printing Company, Easton, PA, 1990, pp. 1581– 1583. [9] B.-Y. Y Hwang, B.-W. Jung, S.-J. Chung, M.-H. Lee, C.-K. Shim, In vitro skin permeation of nicotine from proliposomes, J. Control. Release 49 (1997) 177–184. [10] M.K. Kim, S.-J. Chung, M.-H. Lee, A.-R. Cho, C.-K. Shim, Targeted and sustained delivery of hydrocortisone to normal and stratum corneum-removed skin without enhanced skin absorption using a liposome gel, J. Control. Rel. 46 (1997) 243–251. [11] A. Rougier, C. Lotte, The stripping technique, in: V.P. Shah, H.I. Mailbach (Eds.), Topical Drug Bioavailability, Bioequivalence and Penetration, Plenum Press, New York, 1993, p. 180. [12] S.L. Hsia, A.J. Mussallem, V.H. Witten, Further metabolic studies of hydrocortisone- 14 C in human skin: Requirement for pyridine nucleotides and sites of metabolism, J. Invest. Dermatol. 45 (1965) 384–390.