- Email: [email protected]
Transdermal iontophoretic delivery of enoxacin from various liposome-encapsulated formulations
Journal of Controlled Release 60 (1999) 1–10
Transdermal iontophoretic delivery of enoxacin from various liposome-encapsulated formulations Jia-You Fang a
a,b ,
*, K.C. Sung c , Hung-Hong Lin c , Chia-Lang Fang d
Graduate Institute of Pharmaceutical Sciences, Taipei Medical College, 250 Wu-Hsing Street, Taipei, Taiwan b Research Center for Clinical Pharmacy, Taipei Municipal Wanfang Hospital, Taipei, Taiwan c School of Pharmacy, Chia Nan College of Pharmacy and Science, Tainan Hsien, Taiwan d Department of Pathology, Taipei Medical College, Taipei, Taiwan Received 9 November 1998; accepted 19 December 1998
Abstract The major purpose of this work was to study the effect of various liposome formulations on the iontophoretic transport of enoxacin through excised rat skin. The electrochemical stability of these liposomes was also evaluated. The encapsulation percentage of enoxacin was significantly enhanced after 6 h incubation in an electric field; whereas the fusion of liposomes was inhibited by application of electric current. The results of iontophoretic drug transport showed that the permeability of enoxacin released from liposomes was higher compared with that of free drug. The iontophoretic permeability of enoxacin released from liposomes increased with a decrease in the fatty acid chain length of the phospholipid, which may be due to the different phase transition temperatures of the phospholipids. Incorporation of charged phospholipid resulted in an alteration of the transdermal behavior of enoxacin: the iontophoretic permeation as well as the amount of enoxacin partitioned in skin was greatly reduced after incorporation of stearylamine in liposomes, which can be attributed to the competitive ion effect. The enoxacin released from stratum corneum-based liposomes showed the highest amount of enoxacin partitioned into skin depot. The results of employing cathodal iontophoresis on negative charged liposomes suggested that the liposomal vesicles or phospholipids may carry enoxacin into deeper skin strata via the follicular route. 1999 Elsevier Science B.V. All rights reserved. Keywords: Enoxacin; Transdermal iontophoresis; Liposome; Stability
1. Introduction Enoxacin is a newly synthesized antimicrobial fluoquinolone structurally related to nalidixic acid. It has a broad spectrum of activity against both Grampositive and Gram-negative bacteria [1]. However, several drug interactions and adverse drug effects *Corresponding author. Fax: 1886-2-2707-4804. E-mail address: [email protected] (J. Fang)
after oral administration have limited its use. Some GI symptoms, including anorexia, nausea, vomiting, diarrhea and metallic taste have been reported; the drug interaction between enoxacin and antiacids containing magnesium or aluminum may result in the decrease of enoxacin oral bioavailability [2]. Moreover, the plasma half-life of enoxacin is only 3–6 h [3], suggesting frequent dosing is needed. Accordingly, the transdermal drug delivery system (TDDS) may be suitable for enoxacin in order to reduce the
0168-3659 / 99 / $ – see front matter 1999 Elsevier Science B.V. All rights reserved. PII: S0168-3659( 99 )00055-3
2
J. Fang et al. / Journal of Controlled Release 60 (1999) 1 – 10
adverse effects caused by oral administration and to prolong the therapeutic duration. The feasibility of enoxacin transdermal iontophoretic delivery for systemic and topical application has been investigated and reported in a previous publication [4]. Increasing evidence has shown that a liposomebased drug delivery system is not only suitable for parenteral administration, but also for topical applications [5]. Recently, the combination of iontophoresis and liposome technologies in drug delivery has been reported [6]: the liposome may act as a drug reservoir to continuously release drug to the skin surface; iontophoresis may pump released drug into the dermis for systemic or topical applications. The combination of both systems may provide a constant rate of drug input without the constraints of traditional diffusion-based transdermal devices, such as molecular size and lipophilicity. Nevertheless, in order to optimize this technique, the effect of various liposome formulations on transdermal iontophoretic drug delivery has to be studied in a systematic way. There were two major goals in the present study. The first goal was to assess the electrochemical stability of enoxacin-loaded liposome formulations under application of electric current. The second goal was to study the effect of various liposome formulations on enoxacin iontophoretic transport through rat skin.
2. Materials and methods
2.1. Materials Enoxacin, dimyristoyl-L-a-phosphatidylcholine (DMPC), dipalmitoyl-L-a-phosphatidylcholine (DPPC), distearoyl-L-a-phosphatidylcholine (DSPC), dicetyl phosphate, bovine brain ceramide, cholesterol (CH), cholesterol 3-sulfate (CS), palmitic acid (PA) were obtained from Sigma Chemical Co. (USA). Stearylamine was purchased from Acros Chemical Co. (USA). All the other chemicals and solvents were analytical grade and used as received.
2.2. Preparation of liposomes
directly added to an alcoholic solution (5.1% methanol, 90.3% ethanol and 4.6% isopropyl alcohol) containing phospholipids, the solution was then rapidly injected into five times the volume of deionized water with magnetic stirring. The liposomal dispersions were then frozen in a dry ice–acetone bath and dried in a freeze-dryer (Labconco Co., USA) under 10 microHg vacuum at 2508C for 24 h. The lyophilized liposome powder was redispersed and diluted with pH 5 McIlvaine buffer to give a total lipid concentration of 3.9 mM and an enoxacin concentration of 2.5310 24 g / ml. The various liposome formulations used in this study are listed in Table 1.
2.3. Electrochemical stability of liposomes The in vitro system for studying liposome stability in an electric field was modified from the method of Nakakura et al. [8]. Two electrodes of Ag /AgCl wire (15 mm in effective length) were perpendicularly immersed in a cylindrical device (silicone, 25 mm I.D.) with 4.5 ml liposome dispersion. The distance between the two electrodes was 15 mm. The electrodes were connected to a power supplier (Yokogawa Co., Japan). A constant current of 0.5 mA was passed through this system for 6 h at 258C. The system was agitated with a magnetic stirrer at 600 rpm. The encapsulation percentage of enoxacin and the vesicle size of liposomes were examined before and after this experiment to determine the electrochemical stability of liposomes.
Table 1 Liquid constituents of the various liposomes No.
Liposome code
Composition (molar ratio)
1 2 3 4 5 6 7
DMPC DPPC DSPC Positive (1) Negative (2) DMPC / CH SC (stratum corneum)a
DMPC DPPC DSPC DMPC: stearyl amine59:1 DMPC: dicetyl phosphate59:1 DMPC:CH57:3 ceramide:CH:PA:CS54:2.5:2.5:1 a
a
Liposomes were prepared by a combination of ethanol injection and freeze-drying techniques modified from the methods of Lin et al. [7]. Enoxacin was
The ratio of SC liposome composition is weight ratio. DMPC, dimyristoyl-L-a-phosphatidylcholine; DPPC, dipalmitoyl-L-aphosphatidylcholine; DSPC, distearoyl-L-a-phosphatidylcholine; CH, cholesterol.
J. Fang et al. / Journal of Controlled Release 60 (1999) 1 – 10
2.4. Determination of enoxacin encapsulation percentage The method for determining encapsulation percentages of enoxacin was reported previously [7,9]. Briefly, the enoxacin-containing liposomes were separated from untrapped drug by filtering the liposome dispersion under vacuum (5 mm Hg) through a 0.025 mm filter (Millipore Co., USA); the liposomes were then washed with buffer to completely remove the free drug. The liposome dispersion was then dissolved in an equal volume of ethanol to give a clear solution and enoxacin concentrations were measured by the HPLC method [9].
2.5. Vesicle size analysis Light scattering measurements were performed with a Coulter submicron particle-size analyzer (Model N4MD, Coulter Co., USA). The liposome dispersion was diluted with four times the volume of distilled water to determine the vesicle distribution. The instrumental settings were: temperature 208C; viscosity 0.01 poise; refractive index 1.333; reference angle 908; run time 200 s; and range 0–3000 nm.
after careful shaving. Skin pieces were soaked in the receptor buffer solution for 45 min prior to the beginning of experiment. The rat skin was mounted between the cell compartments with the side of stratum corneum facing towards the donor cell. The donor cell was filled with 8 ml of liposome dispersion. The receptor phase contained 8 ml of pH 7.4 McIlvaine buffer (0.06 M). The available diffusion surface area was 0.785 cm 2 . The stirring rate and temperature were kept at 600 rpm and 378C, respectively. A pair of Ag /AgCl wires with effective length of 15 mm were used as electrodes. The anode was immersed in the donor and the cathode was immersed in the receptor. The electrodes were connected to a current power supplier. The current density was 0.3 mA / cm 2 . Samples (0.2ml) were withdrawn from the receptor compartment at regular intervals and immediately replaced by an equal volume of fresh buffer to maintain a constant volume. The samples were analyzed by the HPLC method [9]. At the end of the permeation experiment, the skin was removed and homogenized in 0.1 M HCl solution. The homogenized skin was then centrifuged and filtered before enoxacin analysis [4].
2.6. Separation of unencapsulated enoxacin
3. Results
The transdermal delivery of enoxacin from liposome was conducted after removing the unencapsulated drug. The liposome dispersion was centrifuged at 100,000 rpm and 48C for 20 min in order to separate the incorporated enoxacin from the free enoxacin. The liposomes and supernatant were then separated and the liposome was redispersed in pH 5 McIlvaine buffer to obtain approximately 31.2 mM lipid concentration and 2310 23 g / ml enoxacin concentration before applying to the skin.
3.1. Electrochemical stability of liposomes
2.7. Transdermal iontophoretic delivery of liposome-encapsulated enoxacin The in vitro transdermal iontophoretic delivery of enoxacin was conducted using a side-by-side glass diffusion cell. Excised Wistar rat skin was used as the skin barrier. The rat hair was removed with electric clippers and the abdominal skin was excised
3
The choice of phospholipids in preparing liposomes is often limited to the phosphatidylcholine family because of toxicological considerations and the availability of pure compounds. Accordingly, the phospholipids of phosphatidylcholine with saturated fatty acids, C 14 (DMPC), C 16 (DPPC) and C 18 (DSPC) were selected for the stability study. The enoxacin encapsulation percentage and vesicle size of liposomes after 6 h incubation in pH 5 buffer with and without applying electric current are shown in Tables 2 and 3, respectively. During the 6 h incubation in buffer without applying electric current, no significant enoxacin loss from liposomes was observed except for the DMPC / CH-based composition. The loss of enoxacin from DMPC / CH-based liposomes was similar to a previous report that the encapsulation percentage of enoxacin was reduced
J. Fang et al. / Journal of Controlled Release 60 (1999) 1 – 10
4
Table 2 Encapsulation (%) of enoxacin liposomes after preparation (Initial) and 6 h incubation with or without 0.5 mA current density (n53)a Liposome code
Initial
After 6 h in buffer without current
After 6 h in buffer with 0.5 mA current
DMPC DPPC DSPC Positive (1) Negative (2) DMPC / CH SC (stratum corneum)
61.5566.79 54.5869.66 44.1165.23 66.83610.70 61.2069.97 68.9864.22 88.67615.71
53.4062.13 52.4865.44 44.15614.16 63.12624.62 65.3863.23 49.8767.81 88.5065.72
78.1464.35 72.35619.52 64.04613.03 79.9365.31 59.1163.44 79.6068.92 87.9564.09
a
Each value represents the mean6SD (n53).
after incorporation of CH, and this phenomenon may be due to the occurrence of hydrolysis [9]. After 6 h treatment with 0.5 mA current in buffer, the encapsulation percentages of enoxacin were slightly increased, except for the SC and Negative liposomes (Table 2). The incubation of liposomes in buffer with and without applying electric field were both associated with the fusion of liposomes and an increase in vesicle size, suggesting that the fusion form of liposome is a thermodynamically more favorable state (Table 3). The size of the current-treated liposomes was generally smaller than that of untreated liposomes after incubation.
DMPC-encapsulated form was higher (P,0.05, ttest) than that of free drug (Fig. 2). Fig. 1 also shows that the permeation of enoxacin was increased significantly after applying iontophoresis, and the DMPC-encapsulated enoxacin had a higher skin
3.2. Transdermal iontophoretic delivery of liposome-encapsulated enoxacin Fig. 1 shows the permeation of free and DMPCencapsulated enoxacin with and without applying electric current. There was essentially no enoxacin permeated through the skin via passive diffusion. At the end of the passive diffusion studies, the amount of enoxacin partitioned into the skin reservoir for the
Fig. 1. Cumulative amount of permeated enoxacin from various formulations versus time with and without iontophoresis: (d) free drug without iontophoresis; (s) DMPC-based liposomes without iontophoresis; (j) free drug with iontophoresis; (h) DMPCbased liposomes with iontophoresis. Each value represents the mean6SD (n53).
Table 3 Vesicle size (nm) of enoxacin liposomes after preparation (Initial) and 6 h incubation with or without 0.5 mA current density Liposome code
Initial
After 6 h in buffer without current
After 6 h in buffer with 0.5 mA current
DMPC DPPC DSPC Positive (1) Negative (2) DMPC / CH SC (stratum corneum)
198 657 912 451 148 388 364
1050 1326 2250 1570 337 1142 872
668 924 1997 593 330 968 926
J. Fang et al. / Journal of Controlled Release 60 (1999) 1 – 10
5
Fig. 2. Enoxacin amount partitioned in skin reservoir from various formulations with and without iontophoresis (ITP). Each value represents the mean6SD (n53).
permeation rate compared to that of free drug. However, the partition of enoxacin in the skin reservoir was higher for the free drug (Fig. 2). The transdermal iontophoretic delivery of enoxacin from liposomes based on various saturated fatty acids, including C 14 (DMPC), C 16 (DPPC) and C 18 (DSPC), are compared in Fig. 3 (A). The flux of enoxacin decreased with an increase in fatty acid chain length of the phospholipid. A similar trend was observed in the permeation of enoxacin from the liposomes through a cellulose membrane using iontophoresis (Fig. 3 (B)). Although the permeation rates of enoxacin from various liposome formulations were different, no significant difference (P, 0.05, ANOVA test) was observed for the amounts of enoxacin partitioned in skin for these three liposome formulations (Fig. 4). The incorporation of charged phospholipids in the liposome may result in an alteration of the transdermal behavior of enoxacin (Figs. 4 and 5). Under anodal iontophoresis, the permeation rate and skin residue of enoxacin were greatly reduced after incorporation of stearylamine in the liposomes (the Positive liposomes); moreover, both the permeation rate and skin residue of enoxacin decreased with an increase of stearylamine concentration in liposomes (Figs. 6 and 7). Incorporation of dicetyl phosphate in liposomes (the Negative liposomes) did not significantly change the permeation behavior of enoxacin by anodal iontophoresis (Figs. 4 and 5). The
Fig. 3. Cumulative amount of permeated enoxacin from liposomes with various phospholipid chain lengths through (A) rat skin; (B) cellulose acetate membrane versus time following iontophoresis: (d) free drug; (s) DMPC liposomes; (j) DPPC liposomes; (h) DSPC liposomes. Each value represents the mean6SD (n53).
Fig. 4. Enoxacin amount partitioned in skin reservoir from various liposome formulations following iontophoresis (ITP). Each value represents the mean6SD (n53).
6
J. Fang et al. / Journal of Controlled Release 60 (1999) 1 – 10
Fig. 5. Cumulative amount of permeated enoxacin from liposomes with various charges versus time following anodal iontophoresis: (d) DMPC liposomes (neutral); (s) Positive liposomes; (j) Negative liposomes. Each value represents the mean6SD (n53).
permeation of negative liposomes may be driven by cathodal iontophoresis due to the negative charge in the phospholipid bilayers. This phenomenon is shown in Table 4. In comparison to the anodal iontophoresis, there was a significantly lower but detectable flux and skin residues of enoxacin from negative liposomes via cathodal iontophoresis. However, for the permeation of free enoxacin in pH 5 buffer, in which enoxacin behaves as a positively charged molecule, it is seen that no detectable amounts of enoxacin were transported across rat skin when the cathode was placed in the donor and the anode was placed in the receptor.
Fig. 6. Cumulative amount of permeated enoxacin from positive liposomes with various stearylamine amounts versus time following iontophoresis: (d) 0.195 mM; (s) 0.390 mM; (j) 0.780 mM. Each value represents the mean6SD (n53).
Fig. 7. Enoxacin amount partitioned in skin reservoir from positive liposomes with various stearylamine amounts following iontophoresis. Each value represents the mean6SD (n53).
The iontophoretic permeation of enoxacin released from SC and DMPC / CH liposomes was slower than that from DMPC liposomes, which may be due to the presence of CH in the first two formulations (Fig. 8). However, the SC liposomes resulted in the highest amount of enoxacin partitioned in skin among the various liposome formulations studied (Fig. 4).
4. Discussion
4.1. Electrochemical stability of liposomes It is a general phenomenon that pore widening in phospholipid bilayers can be induced by application of electric field pulses, resulting in the irreversible breakdown of liposomes and the leakage of drug [10]. However, the previous study also suggests that an electric field may inhibit pore opening and stabilize pores of lipid bilayers under certain conditions [11]. In the present study, the encapsulation percentage of enoxacin was similar with and without application of electric current, and increased in some formulations (Table 2). A previous report has indicated that the opening and closing of liposome pores depend on the forces acting on the pore boundary [11]. In principle, the effective values of the energy (r eff ) and the lateral tension (seff ) acting on the pore boundary can be either negative or positive depending on the electric fields. If both are negative, the pore radius will be stable. In this study, the enoxacin
J. Fang et al. / Journal of Controlled Release 60 (1999) 1 – 10
7
Table 4 Comparison of flux and skin residual of enoxacin from anodal and cathodal iontophoresis for Negative (2) liposome as well as cathodal iontophoresis for free enoxacin without (liposome)a
Anodal ITP for (2) Cathodal ITP for (2) Cathodal ITP for free drug a
Flux (mg / cm 2 / h)
Enoxacin in skin (mg / g)
53.9364.90 0.7860.59 0
265.10681.20 74.8769.32 23.02612.68
Each value represents the mean6SD (n53).
liposomes in the electric field may fit with this condition, contributing to the stabilization or sealing of the liposomal pores. However, the mechanism by which this occurs is not completely understood, and further studies are needed to explore this mechanism. Another explanation for the observed electrochemical stability of enoxacin liposomes is that, in order for the pore widening and irreversible breakdown of the liposome to happen, a higher electric field pulse must be introduced [12]. The 0.5 mA current density used in our study may be too low to induce the rupture of liposomes. Except for the SC-based liposomes, the vesicle size of enoxacin liposomes subjected to a 0.5 mA current was smaller than untreated liposomes after 6 h incubation in pH 5 buffer (Table 3). This indicates that the electric field inhibited fusion of phosphatidylcholine based liposomes. Electric fields can cause motion of the electrolyte as well as the lipid. Moreover, the movement induced by electric field may have a similar effect on vesicle size as sonication [10]; large liposomes subjected to sonication
Fig. 8. Cumulative amount of permeated enoxacin from various liposome formulations versus time following iontophoresis: (d) DMPC liposomes; (s) DMPC / CH liposomes; (j) SC liposomes. Each value represents the mean6SD (n53).
often reduce vesicle size and form small liposomes [13,14]. The results in Table 3 are consistent with the above inferences and previous findings. SC liposomes have been used as a model to predict the lipid fluidizing effect of SC in skin [15,16]. In the present study, SC liposomes showed minor changes in the enoxacin encapsulation percentage and vesicle size after treatment with electric current. The results suggest that the SC structure of skin may remain intact after iontophoresis. The results are also in accordance with our previous histological finding that there was no significant change in anatomical structure of SC for rat skin after in vitro iontophoresis [17].
4.2. Comparison of iontophoretic enoxacin permeation between liposome-encapsulated form and free drug form Fig. 1 shows that the iontophoretic permeation of enoxacin across rat skin was significantly increased after encapsulation of enoxacin in liposomes. This phenomenon can be well explained by the following inferences and findings: the addition of phosphatidylcholine to dermal dosage forms has been reported to enhance transdermal absorption [18]. The liposomes may be adsorbed by and fuse with the surfaces of the skin, which can alter the lipid barrier and result in a more permeable structure. It has also been shown that mixing of liposomes with intercellular lipid domains in the SC may contribute to the enhancement of skin permeability [19–21]. Moreover, the observed fusion of phospholipids on the skin surface may facilitate accumulation of water in the SC [13,22], which subsequently contributes to the increase of conductivity in skin during iontophoresis and enhancement of drug permeation [17,23]. Fig. 2 shows that the amount of enoxacin partitioned in skin was lower for the enoxacin from
8
J. Fang et al. / Journal of Controlled Release 60 (1999) 1 – 10
DMPC liposomes compared to that of free drug. The results indicate that the free drug accumulated in the skin to a greater extent than the enoxacin from DMPC-based liposomes. These results are consistent with permeation results since the formation of a skin depot can be reduced when the total drug permeation increases during iontophoresis [24].
4.3. Transdermal iontophoretic delivery of enoxacin from liposomes with various fatty acid chain lengths The rate of transdermal drug delivery depends on the rate of drug release from the formulation to the skin surface and the rate of drug permeation through the skin. Both processes can affect the rate of drug delivery and the slower one will be the rate-limiting step. In order to identify the predominant rate process, the permeation of enoxacin through the skin membrane and a cellulose acetate membrane were compared (Fig. 3(A) and (B)). The permeation study clearly showed that the permeation rate of free enoxacin through skin was dramatically lower than that of enoxacin through the cellulose membrane. These results suggest that the rate-limiting transdermal process for enoxacin was the skin permeation step rather than the drug release step. Fig. 3 (B) showed that the release rate of enoxacin was significantly reduced after encapsulation in liposomes. The release rate of enoxacin from liposomes (especially DPPC and DSPC) approached the skin permeation rate of enoxacin released from liposomes (Fig. 3(A)), demonstrating that the rate-limiting process has gradually shifted to the drug release step. Fig. 3(A) shows that the release rate of enoxacin from DMPC liposome is higher than that from DPPC and DSPC-based liposomes. This can be explained by the phase transition temperatures of the phospholipids. The temperature (378C) used in the skin permeation experiments was higher than the phase transition temperature of DMPC (238C) where the hydrocarbon chains are randomly oriented and fluid (liquid crystalline phase); whereas the temperature was below the transition temperature of DPPC (418C) and DSPC (558C) where the hydrocarbon chains are fully extended and closely packed (gel phase) [25]. The enoxacin will diffuse more easily
through the random hydrocarbon chain than the closed packed gel phase [26].
4.4. Transdermal iontophoretic delivery of enoxacin from charged liposomes Stearylamine and dicetyl phosphate were used as charged components to produce positive and negative liposomes, respectively. The iontophoretic permeation of enoxacin was dramatically reduced after incorporation of stearylamine. The low enoxacin permeability of positive liposomes may be attributed to the competition of stearylamine and positively charged enoxacin (pKa 57.5) for the applied current density. Due to the smaller molecular size of stearylamine relative to enoxacin, most current density is expected to be carried by stearylamine during iontophoretic delivery [27]. Another possible mechanism is the shielding effect of stearylamine. Stearylamine may associate strongly with the negatively charged portion of the skin at pH 5. As the concentration of stearylamine increases, more inherent fixed charge of the skin would be effectively shielded or reversed in the ion-conducting pathways. Accordingly, the level of electroosmotic flow may be decreased or driven in the opposite (cathode-toanode) direction and result in a diminished flux of enoxacin [4]. In order to verify these inferences, different concentrations of stearylamine were incorporated in the liposomes and a series of iontophoretic permeation experiments were performed. Both the enoxacin permeation rate and the amount partitioned in skin decreased with an increase of stearylamine (Figs. 6 and 7), which is consistent with competitive ion effect described above. Cathodal iontophoresis of negative liposomes of enoxacin was utilized in the present study to determine how deeply liposome may penetrate into the skin. Cathodal iontophoresis of free enoxacin at pH 5 showed that no detectable amounts of enoxacin were transported across the skin (Table 4). This result indicates that enoxacin itself with the positive charge cannot permeate across the skin during cathodal iontophoresis. However, there was a low but detectable amount of positively charged enoxacin present in the receptor after cathodal iontophoresis of enoxacin encapsulated in negative liposomes. Previous reports have shown that neither liposomes nor phos-
J. Fang et al. / Journal of Controlled Release 60 (1999) 1 – 10
9
Table 5 Comparison of flux and skin residual of enoxacin from SC liposome and free enoxacin molecules pretreated with palmitic acid (PA) on rat skin a
SC liposome Free drug with PA pretreatment Free drug without PA pretreatment a
Flux (mg / cm 2 / h)
Enoxacin in skin (mg / g)
46.8466.95 90.1667.33 41.9463.86
471.10627.50 466.56692.23 361.80625.90
Each value represents the mean6SD (n53).
pholipid molecules diffuse across intact skin [22]; some reports have indicated that the liposomal drug transport into strata may occur via a follicular route [28]. The above experimental results and previous findings both support the idea that the liposomal vesicles or phospholipids may carry enoxacin into deeper skin strata where receptor sites reside. Moreover, the use of furry rat skin with a higher number of hair follicles in the present study may amplify this effect.
drug form after PA pretreatment, which may be due to the presence of CH in SC liposomes preventing the release of enoxacin from liposomes. There was no significant difference (P.0.05, t-test) between the amount of enoxacin in the skin reservoir of SC liposomes and free drug after PA pretreatment, suggesting that PA has a significant effect on the partition of enoxacin in the skin reservoir for both free enoxacin and enoxacin from SC liposomes.
4.5. Transdermal iontophoretic delivery of enoxacin from liposomes with CH and SC
Acknowledgements
The iontophoretic permeation of enoxacin from DMPC / CH and SC liposomes was lower than that from DMPC liposomes (Fig. 8). This result may be because the CH-incorporated phospholipids are more cohesive and compressible in the electric field, which prevents the release of drug from liposomes [29]. Moreover, the incorporation of CH into drug-laden liposomes may hamper the penetration of lipid vesicles into the skin [30]. Among the formulations studied, Fig. 4 shows that the enoxacin from SCbased liposomes had the greatest amount of drug partitioned into the skin reservoir. Previous results have shown that integrating SC liposomes with the lipid layer of skin facilitates the release of drug from SC liposomes and the permeation of drug into skin [31]. Moreover, the PA in SC liposomes may act as a penetration enhancer and modify the lipid components of skin [32]. In order to verify this point, an iontophoretic experiment on free enoxacin permeation was performed for the skin pretreated with 0.529% (w / v) PA (the same PA concentration as SC liposomes). The results (Table 5) showed that both the permeation flux and the amount of enoxacin in the skin increased. The permeation rate of enoxacin from SC-based liposomes was one half of the free
The authors are grateful to the National Science Council, Taiwan, for the financial support of this study (NSC-87-2314-B-041-006).
References [1] S.H. Zinner, Clinical overview of enoxacin, Clin. Pharmacokinet. 16 (Suppl.) (1989) 59–64. [2] F. Sorgel, U. Jaehde, K. Naber, U. Stephan, Pharmacokinetic disposition of quinolones in human body fluids and tissues, Clin. Pharmacokinet. 16 (Suppl.) (1989) 5–24. [3] M. Schulz, A. Schmoldt, Therapeutic and toxic blood concentrations of more than 500 drugs, Pharmazie 52 (1997) 895–911. [4] J.Y. Fang, H.H. Lin, H.I. Chen, Y.H. Tsai, Development and evaluation on transdermal delivery of enoxacin via chemical enhancers and physical iontophoresis, J. Control. Release 54 (1998) 293–304. [5] T. Ogiso, N. Niinaka, M. Iwaki, Mechanism for enhancement effect of lipid disperse system on percutaneous absorption, J. Pharm. Sci. 85 (1996) 57–64. [6] N.B. Vulta, G.V. Betageri, A.K. Banga, Transdermal iontophoretic delivery of enkephalin formulated in liposomes, J. Pharm. Sci. 85 (1996) 5–8. [7] H.H. Lin, S.M. Ko, L.R. Hsu, Y.H. Tsai, The preparation of norfloxacin-loaded liposomes and their in-vitro evaluation in pig’s eye, J. Pharm. Pharmacol. 48 (1996) 801–805. [8] M. Nakakura, Y. Kato, E. Hayakawa, K. Ito, T. Kuroda,
10
[9]
[10]
[11] [12]
[13]
[14]
[15]
[16]
[17]
[18]
[19]
[20]
J. Fang et al. / Journal of Controlled Release 60 (1999) 1 – 10 Electrochemical stability of desmopressin acetate during iontophoresis, Chem. Pharm. Bull. 44 (1996) 1238–1241. J.Y. Fang, H.H. Lin, L.R. Hsu, Y.H. Tsai, Characterization and stability of various liposome-encapsulated enoxacin formulations, Chem. Pharm. Bull. 45 (1997) 1504–1509. M. Winterhalter, Liposomes in electric fields, in: Y. Barenholz, D.D. Lasic (Eds.), Handbook of Nonmedical Applications of Liposomes, CRC Press, London, 1996, pp. 285–307. M. Winterhalter, W. Helfrich, Effect of voltage on pores in membranes, Phys. Rev. 36 (1987) 5874–5876. C. Wilhelm, M. Winterhalter, U. Zimmermann, R. Benz, Kinetics of pore size during irreversible electrical breakdown of lipid bilayer membranes, Biophys. J. 64 (1993) 121–128. E. Touitou, H.E. Juginger, N.D. Weiner, T. Nagai, M. Mezei, Liposomes as carriers for topical and transdermal delivery, J. Pharm. Sci. 83 (1994) 1189–1203. S. Vemuri, C.T. Rhodes, Preparation and characterization of liposomes as therapeutic delivery systems: a review, Pharm. Acta Helvet. 70 (1995) 95–111. 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. T.M. Suhonen, L. Pirkanen, M. Raisanen, K. Kosonen, J.H. Rytting, P. Paronen, A. Urtti, Transepidermal delivery of b-blocking agents: evaluation of enhancer effects using stratum corneum lipid liposomes, J. Control. Release 43 (1997) 251–259. J.Y. Fang, C.L. Fang, Y.B. Huang, Y.H. Tsai, Transdermal iontophoresis of sodium nonivamide acetate; III: Combined effect of pretreatment by penetration enhancers, Int. J. Pharm. 149 (1997) 183–193. Y. Kurosaki, N. Nagahara, T. Tanizawa, H. Nishimura, T. Nakayama, T. Kimura, Use of lipid disperse systems in transdermal delivery: comparative study of flufenamic acid permeation among rat abdominal skin, silicon rubber membrane and stratum corneum sheet isolated from hamster cheek pouch, Int. J. Pharm. 67 (1991) 1–9. Y. Yokomizo, Effect of phosphatidylcholine on the percutaneous penetration of drugs through the dorsal skin of guinea pigs in vitro; and analysis of the molecular mechanism, using attenuated total reflectance–fourier transform infrared (ATR– FTIR) spectroscopy, J. Control. Release 42 (1996) 249–262. L. Coderch, M. Oliva, M. Pons, A. de la Maza, A.M. Manich, J.L. Parra, Percutaneous penetration of liposomes
[21]
[22]
[23]
[24]
[25]
[26]
[27]
[28]
[29]
[30]
[31]
[32]
using the tape stripping technique, Int. J. Pharm. 139 (1996) 197–203. T. Ogiso, N. Niinaka, M. Iwaki, T. Tanino, Mechanism for enhancement effect of lipid disperse system on percutaneous absorption Part II, Int. J. Pharm. 152 (1997) 135–144. H. Schreier, J. Bouwstra, Liposomes and niosomes as topical drug carriers: dermal and transdermal drug delivery, J. Control. Release 30 (1994) 1–15. V.U. Rao, A.N. Misra, Enhancement of iontophoretic permeation of insulin across human cadaver skin, Pharmazie 49 (1994) 538–539. M.C. Heit, N.A. Monterio-Riviere, F.L. Jayes, J.E. Riviere, Transdermal iontophoretic delivery of luteinizing hormone releasing hormone (LHRH): effect of repeated administration, Pharm. Res. 11 (1994) 1000–1003. S.W. Burgess, J.D. Moore, W.A. Shaw, Lipids as raw materials for pharmaceutical products: properties and characteristics, in: Y. Barenholz, D.D. Lasic (Eds.), Handbook of Nonmedical Applications of Liposomes, CRC Press, London, 1996, pp. 5–21. P. Saarinen-Savolainen, T. Jarvinen, H. Taipale, A. Urtti, Method for evaluating drug release from liposomes in sink conditions, Int. J. Pharm. 159 (1997) 27–33. N.H. Yoshida, M.S. Roberts, Solute molecular size and transdermal iontophoresis across excised human skin, J. Control. Release 25 (1993) 177–195. J. du Plessis, K. Egbaria, C. Ramachandran, N. Weiner, Topical delivery of liposomally encapsulated gamma-interferon, Antiviral Res. 18 (1992) 259–265. D. Needham, R.M. Hochmuch, Electro-mechanical permeabilization of lipid vesicles, Biophys. J. 55 (1989) 1001– 1009. G. Cevc, G. Blume, A. Schatzlein, D. Gebauer, A. Paul, The skin: a pathway for the systemic treatment with patches and lipid-based agent carriers, Adv. Drug Deliv. Rev. 18 (1996) 349–378. H.Y. Yu, H.M. Liao, Triamcinolone permeation from different liposome formulations through rat skin in vitro, Int. J. Pharm. 27 (1996) 1–7. N. Weiner, N. Williams, G. Birch, C. Ramachandran, C. Shipman, G. Flynn, Topical delivery of liposomally encapsulated interferon evaluated in a cutaneous herpes guinea pig model, Antimicrob. Agents Chemother. 33 (1989) 1217– 1221.
Transdermal iontophoretic delivery of enoxacin from various liposome-encapsulated formulations Jia-You Fang a
a,b ,
*, K.C. Sung c , Hung-Hong Lin c , Chia-Lang Fang d
Graduate Institute of Pharmaceutical Sciences, Taipei Medical College, 250 Wu-Hsing Street, Taipei, Taiwan b Research Center for Clinical Pharmacy, Taipei Municipal Wanfang Hospital, Taipei, Taiwan c School of Pharmacy, Chia Nan College of Pharmacy and Science, Tainan Hsien, Taiwan d Department of Pathology, Taipei Medical College, Taipei, Taiwan Received 9 November 1998; accepted 19 December 1998
Abstract The major purpose of this work was to study the effect of various liposome formulations on the iontophoretic transport of enoxacin through excised rat skin. The electrochemical stability of these liposomes was also evaluated. The encapsulation percentage of enoxacin was significantly enhanced after 6 h incubation in an electric field; whereas the fusion of liposomes was inhibited by application of electric current. The results of iontophoretic drug transport showed that the permeability of enoxacin released from liposomes was higher compared with that of free drug. The iontophoretic permeability of enoxacin released from liposomes increased with a decrease in the fatty acid chain length of the phospholipid, which may be due to the different phase transition temperatures of the phospholipids. Incorporation of charged phospholipid resulted in an alteration of the transdermal behavior of enoxacin: the iontophoretic permeation as well as the amount of enoxacin partitioned in skin was greatly reduced after incorporation of stearylamine in liposomes, which can be attributed to the competitive ion effect. The enoxacin released from stratum corneum-based liposomes showed the highest amount of enoxacin partitioned into skin depot. The results of employing cathodal iontophoresis on negative charged liposomes suggested that the liposomal vesicles or phospholipids may carry enoxacin into deeper skin strata via the follicular route. 1999 Elsevier Science B.V. All rights reserved. Keywords: Enoxacin; Transdermal iontophoresis; Liposome; Stability
1. Introduction Enoxacin is a newly synthesized antimicrobial fluoquinolone structurally related to nalidixic acid. It has a broad spectrum of activity against both Grampositive and Gram-negative bacteria [1]. However, several drug interactions and adverse drug effects *Corresponding author. Fax: 1886-2-2707-4804. E-mail address: [email protected] (J. Fang)
after oral administration have limited its use. Some GI symptoms, including anorexia, nausea, vomiting, diarrhea and metallic taste have been reported; the drug interaction between enoxacin and antiacids containing magnesium or aluminum may result in the decrease of enoxacin oral bioavailability [2]. Moreover, the plasma half-life of enoxacin is only 3–6 h [3], suggesting frequent dosing is needed. Accordingly, the transdermal drug delivery system (TDDS) may be suitable for enoxacin in order to reduce the
0168-3659 / 99 / $ – see front matter 1999 Elsevier Science B.V. All rights reserved. PII: S0168-3659( 99 )00055-3
2
J. Fang et al. / Journal of Controlled Release 60 (1999) 1 – 10
adverse effects caused by oral administration and to prolong the therapeutic duration. The feasibility of enoxacin transdermal iontophoretic delivery for systemic and topical application has been investigated and reported in a previous publication [4]. Increasing evidence has shown that a liposomebased drug delivery system is not only suitable for parenteral administration, but also for topical applications [5]. Recently, the combination of iontophoresis and liposome technologies in drug delivery has been reported [6]: the liposome may act as a drug reservoir to continuously release drug to the skin surface; iontophoresis may pump released drug into the dermis for systemic or topical applications. The combination of both systems may provide a constant rate of drug input without the constraints of traditional diffusion-based transdermal devices, such as molecular size and lipophilicity. Nevertheless, in order to optimize this technique, the effect of various liposome formulations on transdermal iontophoretic drug delivery has to be studied in a systematic way. There were two major goals in the present study. The first goal was to assess the electrochemical stability of enoxacin-loaded liposome formulations under application of electric current. The second goal was to study the effect of various liposome formulations on enoxacin iontophoretic transport through rat skin.
2. Materials and methods
2.1. Materials Enoxacin, dimyristoyl-L-a-phosphatidylcholine (DMPC), dipalmitoyl-L-a-phosphatidylcholine (DPPC), distearoyl-L-a-phosphatidylcholine (DSPC), dicetyl phosphate, bovine brain ceramide, cholesterol (CH), cholesterol 3-sulfate (CS), palmitic acid (PA) were obtained from Sigma Chemical Co. (USA). Stearylamine was purchased from Acros Chemical Co. (USA). All the other chemicals and solvents were analytical grade and used as received.
2.2. Preparation of liposomes
directly added to an alcoholic solution (5.1% methanol, 90.3% ethanol and 4.6% isopropyl alcohol) containing phospholipids, the solution was then rapidly injected into five times the volume of deionized water with magnetic stirring. The liposomal dispersions were then frozen in a dry ice–acetone bath and dried in a freeze-dryer (Labconco Co., USA) under 10 microHg vacuum at 2508C for 24 h. The lyophilized liposome powder was redispersed and diluted with pH 5 McIlvaine buffer to give a total lipid concentration of 3.9 mM and an enoxacin concentration of 2.5310 24 g / ml. The various liposome formulations used in this study are listed in Table 1.
2.3. Electrochemical stability of liposomes The in vitro system for studying liposome stability in an electric field was modified from the method of Nakakura et al. [8]. Two electrodes of Ag /AgCl wire (15 mm in effective length) were perpendicularly immersed in a cylindrical device (silicone, 25 mm I.D.) with 4.5 ml liposome dispersion. The distance between the two electrodes was 15 mm. The electrodes were connected to a power supplier (Yokogawa Co., Japan). A constant current of 0.5 mA was passed through this system for 6 h at 258C. The system was agitated with a magnetic stirrer at 600 rpm. The encapsulation percentage of enoxacin and the vesicle size of liposomes were examined before and after this experiment to determine the electrochemical stability of liposomes.
Table 1 Liquid constituents of the various liposomes No.
Liposome code
Composition (molar ratio)
1 2 3 4 5 6 7
DMPC DPPC DSPC Positive (1) Negative (2) DMPC / CH SC (stratum corneum)a
DMPC DPPC DSPC DMPC: stearyl amine59:1 DMPC: dicetyl phosphate59:1 DMPC:CH57:3 ceramide:CH:PA:CS54:2.5:2.5:1 a
a
Liposomes were prepared by a combination of ethanol injection and freeze-drying techniques modified from the methods of Lin et al. [7]. Enoxacin was
The ratio of SC liposome composition is weight ratio. DMPC, dimyristoyl-L-a-phosphatidylcholine; DPPC, dipalmitoyl-L-aphosphatidylcholine; DSPC, distearoyl-L-a-phosphatidylcholine; CH, cholesterol.
J. Fang et al. / Journal of Controlled Release 60 (1999) 1 – 10
2.4. Determination of enoxacin encapsulation percentage The method for determining encapsulation percentages of enoxacin was reported previously [7,9]. Briefly, the enoxacin-containing liposomes were separated from untrapped drug by filtering the liposome dispersion under vacuum (5 mm Hg) through a 0.025 mm filter (Millipore Co., USA); the liposomes were then washed with buffer to completely remove the free drug. The liposome dispersion was then dissolved in an equal volume of ethanol to give a clear solution and enoxacin concentrations were measured by the HPLC method [9].
2.5. Vesicle size analysis Light scattering measurements were performed with a Coulter submicron particle-size analyzer (Model N4MD, Coulter Co., USA). The liposome dispersion was diluted with four times the volume of distilled water to determine the vesicle distribution. The instrumental settings were: temperature 208C; viscosity 0.01 poise; refractive index 1.333; reference angle 908; run time 200 s; and range 0–3000 nm.
after careful shaving. Skin pieces were soaked in the receptor buffer solution for 45 min prior to the beginning of experiment. The rat skin was mounted between the cell compartments with the side of stratum corneum facing towards the donor cell. The donor cell was filled with 8 ml of liposome dispersion. The receptor phase contained 8 ml of pH 7.4 McIlvaine buffer (0.06 M). The available diffusion surface area was 0.785 cm 2 . The stirring rate and temperature were kept at 600 rpm and 378C, respectively. A pair of Ag /AgCl wires with effective length of 15 mm were used as electrodes. The anode was immersed in the donor and the cathode was immersed in the receptor. The electrodes were connected to a current power supplier. The current density was 0.3 mA / cm 2 . Samples (0.2ml) were withdrawn from the receptor compartment at regular intervals and immediately replaced by an equal volume of fresh buffer to maintain a constant volume. The samples were analyzed by the HPLC method [9]. At the end of the permeation experiment, the skin was removed and homogenized in 0.1 M HCl solution. The homogenized skin was then centrifuged and filtered before enoxacin analysis [4].
2.6. Separation of unencapsulated enoxacin
3. Results
The transdermal delivery of enoxacin from liposome was conducted after removing the unencapsulated drug. The liposome dispersion was centrifuged at 100,000 rpm and 48C for 20 min in order to separate the incorporated enoxacin from the free enoxacin. The liposomes and supernatant were then separated and the liposome was redispersed in pH 5 McIlvaine buffer to obtain approximately 31.2 mM lipid concentration and 2310 23 g / ml enoxacin concentration before applying to the skin.
3.1. Electrochemical stability of liposomes
2.7. Transdermal iontophoretic delivery of liposome-encapsulated enoxacin The in vitro transdermal iontophoretic delivery of enoxacin was conducted using a side-by-side glass diffusion cell. Excised Wistar rat skin was used as the skin barrier. The rat hair was removed with electric clippers and the abdominal skin was excised
3
The choice of phospholipids in preparing liposomes is often limited to the phosphatidylcholine family because of toxicological considerations and the availability of pure compounds. Accordingly, the phospholipids of phosphatidylcholine with saturated fatty acids, C 14 (DMPC), C 16 (DPPC) and C 18 (DSPC) were selected for the stability study. The enoxacin encapsulation percentage and vesicle size of liposomes after 6 h incubation in pH 5 buffer with and without applying electric current are shown in Tables 2 and 3, respectively. During the 6 h incubation in buffer without applying electric current, no significant enoxacin loss from liposomes was observed except for the DMPC / CH-based composition. The loss of enoxacin from DMPC / CH-based liposomes was similar to a previous report that the encapsulation percentage of enoxacin was reduced
J. Fang et al. / Journal of Controlled Release 60 (1999) 1 – 10
4
Table 2 Encapsulation (%) of enoxacin liposomes after preparation (Initial) and 6 h incubation with or without 0.5 mA current density (n53)a Liposome code
Initial
After 6 h in buffer without current
After 6 h in buffer with 0.5 mA current
DMPC DPPC DSPC Positive (1) Negative (2) DMPC / CH SC (stratum corneum)
61.5566.79 54.5869.66 44.1165.23 66.83610.70 61.2069.97 68.9864.22 88.67615.71
53.4062.13 52.4865.44 44.15614.16 63.12624.62 65.3863.23 49.8767.81 88.5065.72
78.1464.35 72.35619.52 64.04613.03 79.9365.31 59.1163.44 79.6068.92 87.9564.09
a
Each value represents the mean6SD (n53).
after incorporation of CH, and this phenomenon may be due to the occurrence of hydrolysis [9]. After 6 h treatment with 0.5 mA current in buffer, the encapsulation percentages of enoxacin were slightly increased, except for the SC and Negative liposomes (Table 2). The incubation of liposomes in buffer with and without applying electric field were both associated with the fusion of liposomes and an increase in vesicle size, suggesting that the fusion form of liposome is a thermodynamically more favorable state (Table 3). The size of the current-treated liposomes was generally smaller than that of untreated liposomes after incubation.
DMPC-encapsulated form was higher (P,0.05, ttest) than that of free drug (Fig. 2). Fig. 1 also shows that the permeation of enoxacin was increased significantly after applying iontophoresis, and the DMPC-encapsulated enoxacin had a higher skin
3.2. Transdermal iontophoretic delivery of liposome-encapsulated enoxacin Fig. 1 shows the permeation of free and DMPCencapsulated enoxacin with and without applying electric current. There was essentially no enoxacin permeated through the skin via passive diffusion. At the end of the passive diffusion studies, the amount of enoxacin partitioned into the skin reservoir for the
Fig. 1. Cumulative amount of permeated enoxacin from various formulations versus time with and without iontophoresis: (d) free drug without iontophoresis; (s) DMPC-based liposomes without iontophoresis; (j) free drug with iontophoresis; (h) DMPCbased liposomes with iontophoresis. Each value represents the mean6SD (n53).
Table 3 Vesicle size (nm) of enoxacin liposomes after preparation (Initial) and 6 h incubation with or without 0.5 mA current density Liposome code
Initial
After 6 h in buffer without current
After 6 h in buffer with 0.5 mA current
DMPC DPPC DSPC Positive (1) Negative (2) DMPC / CH SC (stratum corneum)
198 657 912 451 148 388 364
1050 1326 2250 1570 337 1142 872
668 924 1997 593 330 968 926
J. Fang et al. / Journal of Controlled Release 60 (1999) 1 – 10
5
Fig. 2. Enoxacin amount partitioned in skin reservoir from various formulations with and without iontophoresis (ITP). Each value represents the mean6SD (n53).
permeation rate compared to that of free drug. However, the partition of enoxacin in the skin reservoir was higher for the free drug (Fig. 2). The transdermal iontophoretic delivery of enoxacin from liposomes based on various saturated fatty acids, including C 14 (DMPC), C 16 (DPPC) and C 18 (DSPC), are compared in Fig. 3 (A). The flux of enoxacin decreased with an increase in fatty acid chain length of the phospholipid. A similar trend was observed in the permeation of enoxacin from the liposomes through a cellulose membrane using iontophoresis (Fig. 3 (B)). Although the permeation rates of enoxacin from various liposome formulations were different, no significant difference (P, 0.05, ANOVA test) was observed for the amounts of enoxacin partitioned in skin for these three liposome formulations (Fig. 4). The incorporation of charged phospholipids in the liposome may result in an alteration of the transdermal behavior of enoxacin (Figs. 4 and 5). Under anodal iontophoresis, the permeation rate and skin residue of enoxacin were greatly reduced after incorporation of stearylamine in the liposomes (the Positive liposomes); moreover, both the permeation rate and skin residue of enoxacin decreased with an increase of stearylamine concentration in liposomes (Figs. 6 and 7). Incorporation of dicetyl phosphate in liposomes (the Negative liposomes) did not significantly change the permeation behavior of enoxacin by anodal iontophoresis (Figs. 4 and 5). The
Fig. 3. Cumulative amount of permeated enoxacin from liposomes with various phospholipid chain lengths through (A) rat skin; (B) cellulose acetate membrane versus time following iontophoresis: (d) free drug; (s) DMPC liposomes; (j) DPPC liposomes; (h) DSPC liposomes. Each value represents the mean6SD (n53).
Fig. 4. Enoxacin amount partitioned in skin reservoir from various liposome formulations following iontophoresis (ITP). Each value represents the mean6SD (n53).
6
J. Fang et al. / Journal of Controlled Release 60 (1999) 1 – 10
Fig. 5. Cumulative amount of permeated enoxacin from liposomes with various charges versus time following anodal iontophoresis: (d) DMPC liposomes (neutral); (s) Positive liposomes; (j) Negative liposomes. Each value represents the mean6SD (n53).
permeation of negative liposomes may be driven by cathodal iontophoresis due to the negative charge in the phospholipid bilayers. This phenomenon is shown in Table 4. In comparison to the anodal iontophoresis, there was a significantly lower but detectable flux and skin residues of enoxacin from negative liposomes via cathodal iontophoresis. However, for the permeation of free enoxacin in pH 5 buffer, in which enoxacin behaves as a positively charged molecule, it is seen that no detectable amounts of enoxacin were transported across rat skin when the cathode was placed in the donor and the anode was placed in the receptor.
Fig. 6. Cumulative amount of permeated enoxacin from positive liposomes with various stearylamine amounts versus time following iontophoresis: (d) 0.195 mM; (s) 0.390 mM; (j) 0.780 mM. Each value represents the mean6SD (n53).
Fig. 7. Enoxacin amount partitioned in skin reservoir from positive liposomes with various stearylamine amounts following iontophoresis. Each value represents the mean6SD (n53).
The iontophoretic permeation of enoxacin released from SC and DMPC / CH liposomes was slower than that from DMPC liposomes, which may be due to the presence of CH in the first two formulations (Fig. 8). However, the SC liposomes resulted in the highest amount of enoxacin partitioned in skin among the various liposome formulations studied (Fig. 4).
4. Discussion
4.1. Electrochemical stability of liposomes It is a general phenomenon that pore widening in phospholipid bilayers can be induced by application of electric field pulses, resulting in the irreversible breakdown of liposomes and the leakage of drug [10]. However, the previous study also suggests that an electric field may inhibit pore opening and stabilize pores of lipid bilayers under certain conditions [11]. In the present study, the encapsulation percentage of enoxacin was similar with and without application of electric current, and increased in some formulations (Table 2). A previous report has indicated that the opening and closing of liposome pores depend on the forces acting on the pore boundary [11]. In principle, the effective values of the energy (r eff ) and the lateral tension (seff ) acting on the pore boundary can be either negative or positive depending on the electric fields. If both are negative, the pore radius will be stable. In this study, the enoxacin
J. Fang et al. / Journal of Controlled Release 60 (1999) 1 – 10
7
Table 4 Comparison of flux and skin residual of enoxacin from anodal and cathodal iontophoresis for Negative (2) liposome as well as cathodal iontophoresis for free enoxacin without (liposome)a
Anodal ITP for (2) Cathodal ITP for (2) Cathodal ITP for free drug a
Flux (mg / cm 2 / h)
Enoxacin in skin (mg / g)
53.9364.90 0.7860.59 0
265.10681.20 74.8769.32 23.02612.68
Each value represents the mean6SD (n53).
liposomes in the electric field may fit with this condition, contributing to the stabilization or sealing of the liposomal pores. However, the mechanism by which this occurs is not completely understood, and further studies are needed to explore this mechanism. Another explanation for the observed electrochemical stability of enoxacin liposomes is that, in order for the pore widening and irreversible breakdown of the liposome to happen, a higher electric field pulse must be introduced [12]. The 0.5 mA current density used in our study may be too low to induce the rupture of liposomes. Except for the SC-based liposomes, the vesicle size of enoxacin liposomes subjected to a 0.5 mA current was smaller than untreated liposomes after 6 h incubation in pH 5 buffer (Table 3). This indicates that the electric field inhibited fusion of phosphatidylcholine based liposomes. Electric fields can cause motion of the electrolyte as well as the lipid. Moreover, the movement induced by electric field may have a similar effect on vesicle size as sonication [10]; large liposomes subjected to sonication
Fig. 8. Cumulative amount of permeated enoxacin from various liposome formulations versus time following iontophoresis: (d) DMPC liposomes; (s) DMPC / CH liposomes; (j) SC liposomes. Each value represents the mean6SD (n53).
often reduce vesicle size and form small liposomes [13,14]. The results in Table 3 are consistent with the above inferences and previous findings. SC liposomes have been used as a model to predict the lipid fluidizing effect of SC in skin [15,16]. In the present study, SC liposomes showed minor changes in the enoxacin encapsulation percentage and vesicle size after treatment with electric current. The results suggest that the SC structure of skin may remain intact after iontophoresis. The results are also in accordance with our previous histological finding that there was no significant change in anatomical structure of SC for rat skin after in vitro iontophoresis [17].
4.2. Comparison of iontophoretic enoxacin permeation between liposome-encapsulated form and free drug form Fig. 1 shows that the iontophoretic permeation of enoxacin across rat skin was significantly increased after encapsulation of enoxacin in liposomes. This phenomenon can be well explained by the following inferences and findings: the addition of phosphatidylcholine to dermal dosage forms has been reported to enhance transdermal absorption [18]. The liposomes may be adsorbed by and fuse with the surfaces of the skin, which can alter the lipid barrier and result in a more permeable structure. It has also been shown that mixing of liposomes with intercellular lipid domains in the SC may contribute to the enhancement of skin permeability [19–21]. Moreover, the observed fusion of phospholipids on the skin surface may facilitate accumulation of water in the SC [13,22], which subsequently contributes to the increase of conductivity in skin during iontophoresis and enhancement of drug permeation [17,23]. Fig. 2 shows that the amount of enoxacin partitioned in skin was lower for the enoxacin from
8
J. Fang et al. / Journal of Controlled Release 60 (1999) 1 – 10
DMPC liposomes compared to that of free drug. The results indicate that the free drug accumulated in the skin to a greater extent than the enoxacin from DMPC-based liposomes. These results are consistent with permeation results since the formation of a skin depot can be reduced when the total drug permeation increases during iontophoresis [24].
4.3. Transdermal iontophoretic delivery of enoxacin from liposomes with various fatty acid chain lengths The rate of transdermal drug delivery depends on the rate of drug release from the formulation to the skin surface and the rate of drug permeation through the skin. Both processes can affect the rate of drug delivery and the slower one will be the rate-limiting step. In order to identify the predominant rate process, the permeation of enoxacin through the skin membrane and a cellulose acetate membrane were compared (Fig. 3(A) and (B)). The permeation study clearly showed that the permeation rate of free enoxacin through skin was dramatically lower than that of enoxacin through the cellulose membrane. These results suggest that the rate-limiting transdermal process for enoxacin was the skin permeation step rather than the drug release step. Fig. 3 (B) showed that the release rate of enoxacin was significantly reduced after encapsulation in liposomes. The release rate of enoxacin from liposomes (especially DPPC and DSPC) approached the skin permeation rate of enoxacin released from liposomes (Fig. 3(A)), demonstrating that the rate-limiting process has gradually shifted to the drug release step. Fig. 3(A) shows that the release rate of enoxacin from DMPC liposome is higher than that from DPPC and DSPC-based liposomes. This can be explained by the phase transition temperatures of the phospholipids. The temperature (378C) used in the skin permeation experiments was higher than the phase transition temperature of DMPC (238C) where the hydrocarbon chains are randomly oriented and fluid (liquid crystalline phase); whereas the temperature was below the transition temperature of DPPC (418C) and DSPC (558C) where the hydrocarbon chains are fully extended and closely packed (gel phase) [25]. The enoxacin will diffuse more easily
through the random hydrocarbon chain than the closed packed gel phase [26].
4.4. Transdermal iontophoretic delivery of enoxacin from charged liposomes Stearylamine and dicetyl phosphate were used as charged components to produce positive and negative liposomes, respectively. The iontophoretic permeation of enoxacin was dramatically reduced after incorporation of stearylamine. The low enoxacin permeability of positive liposomes may be attributed to the competition of stearylamine and positively charged enoxacin (pKa 57.5) for the applied current density. Due to the smaller molecular size of stearylamine relative to enoxacin, most current density is expected to be carried by stearylamine during iontophoretic delivery [27]. Another possible mechanism is the shielding effect of stearylamine. Stearylamine may associate strongly with the negatively charged portion of the skin at pH 5. As the concentration of stearylamine increases, more inherent fixed charge of the skin would be effectively shielded or reversed in the ion-conducting pathways. Accordingly, the level of electroosmotic flow may be decreased or driven in the opposite (cathode-toanode) direction and result in a diminished flux of enoxacin [4]. In order to verify these inferences, different concentrations of stearylamine were incorporated in the liposomes and a series of iontophoretic permeation experiments were performed. Both the enoxacin permeation rate and the amount partitioned in skin decreased with an increase of stearylamine (Figs. 6 and 7), which is consistent with competitive ion effect described above. Cathodal iontophoresis of negative liposomes of enoxacin was utilized in the present study to determine how deeply liposome may penetrate into the skin. Cathodal iontophoresis of free enoxacin at pH 5 showed that no detectable amounts of enoxacin were transported across the skin (Table 4). This result indicates that enoxacin itself with the positive charge cannot permeate across the skin during cathodal iontophoresis. However, there was a low but detectable amount of positively charged enoxacin present in the receptor after cathodal iontophoresis of enoxacin encapsulated in negative liposomes. Previous reports have shown that neither liposomes nor phos-
J. Fang et al. / Journal of Controlled Release 60 (1999) 1 – 10
9
Table 5 Comparison of flux and skin residual of enoxacin from SC liposome and free enoxacin molecules pretreated with palmitic acid (PA) on rat skin a
SC liposome Free drug with PA pretreatment Free drug without PA pretreatment a
Flux (mg / cm 2 / h)
Enoxacin in skin (mg / g)
46.8466.95 90.1667.33 41.9463.86
471.10627.50 466.56692.23 361.80625.90
Each value represents the mean6SD (n53).
pholipid molecules diffuse across intact skin [22]; some reports have indicated that the liposomal drug transport into strata may occur via a follicular route [28]. The above experimental results and previous findings both support the idea that the liposomal vesicles or phospholipids may carry enoxacin into deeper skin strata where receptor sites reside. Moreover, the use of furry rat skin with a higher number of hair follicles in the present study may amplify this effect.
drug form after PA pretreatment, which may be due to the presence of CH in SC liposomes preventing the release of enoxacin from liposomes. There was no significant difference (P.0.05, t-test) between the amount of enoxacin in the skin reservoir of SC liposomes and free drug after PA pretreatment, suggesting that PA has a significant effect on the partition of enoxacin in the skin reservoir for both free enoxacin and enoxacin from SC liposomes.
4.5. Transdermal iontophoretic delivery of enoxacin from liposomes with CH and SC
Acknowledgements
The iontophoretic permeation of enoxacin from DMPC / CH and SC liposomes was lower than that from DMPC liposomes (Fig. 8). This result may be because the CH-incorporated phospholipids are more cohesive and compressible in the electric field, which prevents the release of drug from liposomes [29]. Moreover, the incorporation of CH into drug-laden liposomes may hamper the penetration of lipid vesicles into the skin [30]. Among the formulations studied, Fig. 4 shows that the enoxacin from SCbased liposomes had the greatest amount of drug partitioned into the skin reservoir. Previous results have shown that integrating SC liposomes with the lipid layer of skin facilitates the release of drug from SC liposomes and the permeation of drug into skin [31]. Moreover, the PA in SC liposomes may act as a penetration enhancer and modify the lipid components of skin [32]. In order to verify this point, an iontophoretic experiment on free enoxacin permeation was performed for the skin pretreated with 0.529% (w / v) PA (the same PA concentration as SC liposomes). The results (Table 5) showed that both the permeation flux and the amount of enoxacin in the skin increased. The permeation rate of enoxacin from SC-based liposomes was one half of the free
The authors are grateful to the National Science Council, Taiwan, for the financial support of this study (NSC-87-2314-B-041-006).
References [1] S.H. Zinner, Clinical overview of enoxacin, Clin. Pharmacokinet. 16 (Suppl.) (1989) 59–64. [2] F. Sorgel, U. Jaehde, K. Naber, U. Stephan, Pharmacokinetic disposition of quinolones in human body fluids and tissues, Clin. Pharmacokinet. 16 (Suppl.) (1989) 5–24. [3] M. Schulz, A. Schmoldt, Therapeutic and toxic blood concentrations of more than 500 drugs, Pharmazie 52 (1997) 895–911. [4] J.Y. Fang, H.H. Lin, H.I. Chen, Y.H. Tsai, Development and evaluation on transdermal delivery of enoxacin via chemical enhancers and physical iontophoresis, J. Control. Release 54 (1998) 293–304. [5] T. Ogiso, N. Niinaka, M. Iwaki, Mechanism for enhancement effect of lipid disperse system on percutaneous absorption, J. Pharm. Sci. 85 (1996) 57–64. [6] N.B. Vulta, G.V. Betageri, A.K. Banga, Transdermal iontophoretic delivery of enkephalin formulated in liposomes, J. Pharm. Sci. 85 (1996) 5–8. [7] H.H. Lin, S.M. Ko, L.R. Hsu, Y.H. Tsai, The preparation of norfloxacin-loaded liposomes and their in-vitro evaluation in pig’s eye, J. Pharm. Pharmacol. 48 (1996) 801–805. [8] M. Nakakura, Y. Kato, E. Hayakawa, K. Ito, T. Kuroda,
10
[9]
[10]
[11] [12]
[13]
[14]
[15]
[16]
[17]
[18]
[19]
[20]
J. Fang et al. / Journal of Controlled Release 60 (1999) 1 – 10 Electrochemical stability of desmopressin acetate during iontophoresis, Chem. Pharm. Bull. 44 (1996) 1238–1241. J.Y. Fang, H.H. Lin, L.R. Hsu, Y.H. Tsai, Characterization and stability of various liposome-encapsulated enoxacin formulations, Chem. Pharm. Bull. 45 (1997) 1504–1509. M. Winterhalter, Liposomes in electric fields, in: Y. Barenholz, D.D. Lasic (Eds.), Handbook of Nonmedical Applications of Liposomes, CRC Press, London, 1996, pp. 285–307. M. Winterhalter, W. Helfrich, Effect of voltage on pores in membranes, Phys. Rev. 36 (1987) 5874–5876. C. Wilhelm, M. Winterhalter, U. Zimmermann, R. Benz, Kinetics of pore size during irreversible electrical breakdown of lipid bilayer membranes, Biophys. J. 64 (1993) 121–128. E. Touitou, H.E. Juginger, N.D. Weiner, T. Nagai, M. Mezei, Liposomes as carriers for topical and transdermal delivery, J. Pharm. Sci. 83 (1994) 1189–1203. S. Vemuri, C.T. Rhodes, Preparation and characterization of liposomes as therapeutic delivery systems: a review, Pharm. Acta Helvet. 70 (1995) 95–111. 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. T.M. Suhonen, L. Pirkanen, M. Raisanen, K. Kosonen, J.H. Rytting, P. Paronen, A. Urtti, Transepidermal delivery of b-blocking agents: evaluation of enhancer effects using stratum corneum lipid liposomes, J. Control. Release 43 (1997) 251–259. J.Y. Fang, C.L. Fang, Y.B. Huang, Y.H. Tsai, Transdermal iontophoresis of sodium nonivamide acetate; III: Combined effect of pretreatment by penetration enhancers, Int. J. Pharm. 149 (1997) 183–193. Y. Kurosaki, N. Nagahara, T. Tanizawa, H. Nishimura, T. Nakayama, T. Kimura, Use of lipid disperse systems in transdermal delivery: comparative study of flufenamic acid permeation among rat abdominal skin, silicon rubber membrane and stratum corneum sheet isolated from hamster cheek pouch, Int. J. Pharm. 67 (1991) 1–9. Y. Yokomizo, Effect of phosphatidylcholine on the percutaneous penetration of drugs through the dorsal skin of guinea pigs in vitro; and analysis of the molecular mechanism, using attenuated total reflectance–fourier transform infrared (ATR– FTIR) spectroscopy, J. Control. Release 42 (1996) 249–262. L. Coderch, M. Oliva, M. Pons, A. de la Maza, A.M. Manich, J.L. Parra, Percutaneous penetration of liposomes
[21]
[22]
[23]
[24]
[25]
[26]
[27]
[28]
[29]
[30]
[31]
[32]
using the tape stripping technique, Int. J. Pharm. 139 (1996) 197–203. T. Ogiso, N. Niinaka, M. Iwaki, T. Tanino, Mechanism for enhancement effect of lipid disperse system on percutaneous absorption Part II, Int. J. Pharm. 152 (1997) 135–144. H. Schreier, J. Bouwstra, Liposomes and niosomes as topical drug carriers: dermal and transdermal drug delivery, J. Control. Release 30 (1994) 1–15. V.U. Rao, A.N. Misra, Enhancement of iontophoretic permeation of insulin across human cadaver skin, Pharmazie 49 (1994) 538–539. M.C. Heit, N.A. Monterio-Riviere, F.L. Jayes, J.E. Riviere, Transdermal iontophoretic delivery of luteinizing hormone releasing hormone (LHRH): effect of repeated administration, Pharm. Res. 11 (1994) 1000–1003. S.W. Burgess, J.D. Moore, W.A. Shaw, Lipids as raw materials for pharmaceutical products: properties and characteristics, in: Y. Barenholz, D.D. Lasic (Eds.), Handbook of Nonmedical Applications of Liposomes, CRC Press, London, 1996, pp. 5–21. P. Saarinen-Savolainen, T. Jarvinen, H. Taipale, A. Urtti, Method for evaluating drug release from liposomes in sink conditions, Int. J. Pharm. 159 (1997) 27–33. N.H. Yoshida, M.S. Roberts, Solute molecular size and transdermal iontophoresis across excised human skin, J. Control. Release 25 (1993) 177–195. J. du Plessis, K. Egbaria, C. Ramachandran, N. Weiner, Topical delivery of liposomally encapsulated gamma-interferon, Antiviral Res. 18 (1992) 259–265. D. Needham, R.M. Hochmuch, Electro-mechanical permeabilization of lipid vesicles, Biophys. J. 55 (1989) 1001– 1009. G. Cevc, G. Blume, A. Schatzlein, D. Gebauer, A. Paul, The skin: a pathway for the systemic treatment with patches and lipid-based agent carriers, Adv. Drug Deliv. Rev. 18 (1996) 349–378. H.Y. Yu, H.M. Liao, Triamcinolone permeation from different liposome formulations through rat skin in vitro, Int. J. Pharm. 27 (1996) 1–7. N. Weiner, N. Williams, G. Birch, C. Ramachandran, C. Shipman, G. Flynn, Topical delivery of liposomally encapsulated interferon evaluated in a cutaneous herpes guinea pig model, Antimicrob. Agents Chemother. 33 (1989) 1217– 1221.