Dyphylline Liposomes for Delivery to the Skin E. TOUITOU'~, N. SHACO-EZRA*, N. DAYAN*,M.JUSHYNSKI*, R. RAFAELOFF*, AND R. &OURY* Received September 10, 1990, from the 'Department of Pharmacy, School of Pharmacy, Hebrew University of Jerusalem, P.O. Box 12065, Jerusalem, Israel, and the Qepaflment of Applied Chemistry, Soreq, Nuclear Research Center, Yavne, Israel. Accepted for publication March 7, 1991. Abstract 0 Delivery of dyphylline to the skin using liposomes was
investigated. Xanthines are inhibitors of cAMP phosphodiesterase and have been considered for treatment of psoriasis. Dyphylline was chosen because of its solubility in water, which should allow for incorporation of higher concentrations within the liposomes. Liposomes containing dyphylline were prepared by a method using sonication. Transmission electron micrograph (TEM) visualization showed small particles ranging from 40 to 100 nm, and particle size distribution determined by light scattering showed the vesicles to have an average diameter of 360 nm. The transdermal delivery of free dyphylline and dyphylline incorporated in unilamellar liposomes was measured from polyethylene glycol (PEG), Carbopol gel, a PEG enhacer base, and water. For comparison, similar experiments were carried out with theophylline as well. When the drugs were incorporated in Carbopol gel, a large difference was seen between their fluxes, with free dyphylline having the highest permeation, followed by liposomal dyphylline, and then theophylline. With the PEG enhancer base, a very high permeation of theophylline was observed relative to dyphylline and liposomal dyphylline. From the PEG base, liposomal dyphylline exhibited the lowest skin permeation flux relative to other bases. Using the PEG base for dyphylline incorporated in liposomes, a high skin partitioning of the drug, along with low transdermal permeation, was measured. These results may indicate that the drug is localized in the skin.
A number of works have shown that cyclic adenosine 3',5'-monophosphate (CAMP)is involved in differentiation of cells and a cessation of cell proliferation. Psoriasis, a chronic recurrent inflammatory disease of the skin, is characterized by an accelerated epidermal cell cycle; therefore, this disease may be caused by a defect in the cAMP cascade.1.2 Theophylline, a n inhibitor of the phosphodiesterase which inactivates CAMP,was therefore suggested a s a possible treatment for psoriasis. Its oral administration has met with limited success,3 most likely because a high systemic level of the drug is necessary in order for the local skin concentration to reach therapeutic levels. Topical delivery of effective concentrations of theophylline derivatives directly into the skin may be one way of overcoming this drawback. In a previous work, we showed that the skin permeation of theophylline can be modulated by using chemical enhancers and selected carriers. An increase of two orders of magnitude of theophylline flux through the hairless mouse skin was obtained in vitro by using a hydrophilic enhancing base containing oleic acid.4 With the final goal being topical administration of the drug, we sought a dosage form which allows for drug accumulation in the skin with low transderma1 permeation. In recent years, liposomes (microscopicvesicles composed of phospholipid bilayers) have become increasingly important as a vehicle for delivery of active compounds which are used for topical management. In a number of investigations, Mezei et a1.M showed that liposomes can be used to deliver drugs such as corticosteroids, econazole, progesterone, and tetracaine into the skin in greater quantities than conventional 0022-3549/92'0200-0 131$02.50/0 0 1992, American Pharmaceutical Association
vehicles, with localization at the desired site of action. In animal experiments, the liposomal form provided higher drug concentration in the skin and lower concentration in the internal organs than the conventional topical dosage forms. The present study was undertaken in order to determine the effectivenessof delivering a theophylline derivative to the skin via liposomes, for further use as a topical treatment for psoriasis. On the assumption that a high water solubility of the drug would allow for drug incorporation within the liposome, dyphylline, a water-soluble theophylline derivative, was chosen for our studies. The effect of the carrier base on dyphylline skin permeation was investigated. Similar experiments with theophylline, for comparison with the results using dyphylline, were carried out.
Experimental Sectidn Materials-Dyphylline, theophylline, and the formulation carriers [polyethylene glycol (PEG) 400, PEG 4000, oleic acid, Carbopol 934, and propylene glycol] were all analytical grade or conformed to B.P. requirements and were all purchased from Sigma. Diglycol was a gift of Gattefosse (France). The liposome ingredients were soya phosphatidylcholine (Phospholipon 901, a gift from Nattermann phospholipid GmbH, Germany, and cholesterol, purchased from Sigma. Semisolid Carriers for Dyphylline, Dyphylline in Liposomes, and Theophylline-The semisolid bases used for the drug incorporation were: (1)polyethylene glycol (PEG)base (a mixture of PEG 400 and PEG 4000); (2) gel base (a Carbopol934 hydrophilic gel); and (3) PEG enhancer base4 (PEG base with oleic acid and diglycol). The drug concentration in all the formulations was 10 mglg. Preparation of Liposomes-In the present study, dyphyllinecontaining liposomes were prepared by dissolving Phospholipon 90 and cholesterol in a 5:l ratio by weight in chloroform. The solvent was removed under reduced pressure in a rotary evaporator until a smooth dry lipid film was obtained. One milliliter of 5 8 (w/v) dyphylline in saline was added for each 100 mg of dry lipid film. The lipiddyphylline suspension was allowed to stand at room temperature for 2 h with occasional shaking, followed by 10 min of vigorous mixing using a Vibrofix VF, shaker. The suspension was sonicated in an ice bath using a Braun sonicator with a 1-cm-diameter titanium probe for 5 bursts of 2 min each, with a cooling period of 2 min in between, in the presence of N2.The final liposome suspension contained 5% (w/w) dyphylline. The entrapment efficiency of dyphylline in liposomes was determined by dialysis using the method described by Szoka and Rapahadjopoulos.9 Liposomes were tested by particle size distribution (see below) after 4 weeks of storage at 4 "C and found to be stable. Identification of Liposornes by Electron Microscopy-Negative stain electron micrographs were prepared on copper grids which were covered with a 0.5% solution of parlodion (Sigma) in amyl acetate and then coated with carbon. The grids were subjected to glow discharge for 20 seconds. Liposomes were applied to the grid for 30 seconds and drawn off with filter paper. A drop of a 1%solution of uranyl acetate (Sigma) was immediately applied to the grid and drawn off after 30 seconds with a piece of filter paper and allowed to dry for at least 30 minutes. A Philips EA 300 electron microscope at 80 KV was used. Particle Size Distribution Determinations-The particle size distribution (PSD) of liposomes was determined optically using a computerized inspection system (Malvern Instruments, USA). In this Journal of Pharmaceutical Sciences I 131 Vol. 81, No. 2, February 1992
system, a fixed volume of fluid-containing vesicles was scanned by a Ne-He laser beam. A computer program was used to correlate the light scattering signals to the vesicle size. The determination of sizes in the ranges 0.1-0.5 and 0.55.0pm was accomplished in one step. Skin Permeation Measurements-The permeation kinetics were determined on the hairless mouse skin using the Franz cell assembly. Full-thickness abdominal skin excised from six-week-old male hairless mice was mounted in the cell with a surface area of 1.77 cm2 and receiver volume of -8 mL. The formulation was applied on the stratum corneum of the skin. The experiment was carried out as previously described10 and run for 24 h. The dyphylline concentration in the samples waa assayed by HPLC. A computer program was used for the calculation of kinetic parameters.11 Each system was tested in three cells and duplicated. The significance of the results was determined using the two-tailed, unpaired or paired (depending on the set of experiments compared)t test. For this analysis, the "Balance" (IBM) computer program was used. Determination of the Partition Coefticient (K,) (Skin-Carriers) of the Drug-The experiment was carried out in diffusion cells at 37 "C, with the receiver compartment empty. Formulations of various compositions containing 10 mglmL of drug were applied to the skin in the donor compartment of the cells. "he concentration of the donor was measured at 0 time (C,) and at the end of the 24-h experiment (C). The dyphylline concentration was determined by HPLC as described below. The K, was calculated from:
(1) Control experiments with the liposomal preparation kept in the donors of the diffusion cells, where the skin was replaced by a glass disk (Crown Glass Company), were also run. It was found that the dyphylline concentration remained constant at the end of the 24-h experiment at 37 "C. Moreover, autoradiographic studies with dyphylline liposomes containing [3Hlcholesterol show that the liposome8 did penetrate the upper strata of the skin.'* Assay-The dyphylline concentration was determined using a Merck Hitachi HPLC with a 655A variable wavelength UV monitor. The determination was made at 275 nm on a reversed-phase C,, column. The mobile phase was composed of 8%acetonitrile in water and had a flow rate of 1 mumin. The internal standard used was /3-hydroxyethyltheophylline.l3There was no overlap between the peaks of the drug or standard and those of the metabolites and other skin componenta released into the receiver compartment during the experiment.
Results and Discussion Characterization of Dyphylline Liposomes-Active compound-carrying liposomes are prepared by a variety of techniques, such as sonication, vortex, ether injection, French press, or reversed-phase evaporation.'* By varying the method of preparation and lipid composition, liposomes with specific characteristics can be designed. Liposomes containing dyphylline were prepared by the sonication method described in the Experimental Section. The entrapment efficiency of dyphylline was 18%(w/w). Figure 1 is a transmission electron micrograph (TEM)of the liposomes, which shows vesicles with sizes ranging between 40 and 100 nm. In Figure 2, a differential number density distribution of the liposome population is shown. In this histogram, it can be seen that the average mean of the PSD is 363 nm, with 60% of the vesical population having a diameter falling between 189 and 739 nM. The literature describes liposomes which are obtained upon sonication of phospholipid dispersions and which have diameters ranging from 25 to 50 nm as small unilamellar vesicles (SUVs).Those unilamellar liposomes with diameters ranging from 100 to 500 nm are referred to as large unilamellar vesicles (LUVs).14From the PSD data, it would appear that, using the method of sonication to prepare the liposomes as described in this study, a mixture of LUVs and SUVs are 132 I Journal of Pharmaceutical Sciences Vol. 81, No.2, February 1992
.
*:+.5J.,:.
'
I
~
*
Flgure l-Liposomes of dyphylline visualized by E M ( x 40 OOO).
-
24.5 30.0
--i 8
-
95.8 120.2
-
298.0 374.0
iij
-
927.2 1163.5
-
4542.4 5699.9 Signal intensity (%)
Flgure 2-Size distribution of dyphylline liposornes. Key: (-) imrnediately after preparation;(- - -) after 4 weeks of storage at 4 "C. Samples were analyzed using a Malvern autosizer II c.
obtained. However, the electron micrograph (Figure 1)shows that the vesicles are mostly small. The relatively large mean average of the differential number density distribution can be attributed to agglomerations of vesicles, which may have been measured as single units using the Malvern autosizer, as is known to occur with light scattering methods.14 Such clusters of vesicles can be seen in Figure 1. Drug Delivery to the Skin-Liposomal dyphylline incorporated in the three bases (PEG, Carbopol gel, and PEG enhancer) and in aqueous suspension was tested for skin permeation, and the kinetic parameters were compared with those obtained in systems containing free dyphylline, as well as with systems containing theophylline. The skin permeation patterns a t steady state were similar for the various systems regardless of the dosage form or of the state of the drug (free or liposomal). However, the skin permeation flux did vary by almost three orders of magnitude accordingto the carrier, the dru , and its form of incorporation, and ranged from 2.7 x 10-8to 1.1 x lo-' mg cm-' h-'. The effect of the carrier on the drug permeation from dyphylline liposomes is shown in Figure 3. The results show that the carriers had different effects on the liposomal dymg cm-' h-') was phylline; the smallest flux (2.7 x obtained when the liposomes were incorporated into the PEG base, a larger flux was evident when in the Carbopol gel carrier (2.6 x 10-3mg * cm-' h-l) or in aqueous suspension (4.2 x 10-3mg.cm-2-h-'), and the highest value was
-
e
-
80
400
330
J
70
-
60
-
50
-
0 r
5 & r! 200
5
40-
E"
30-
3 2 U
100
20
-
10
-
0 4
0 PEG BASE
c€l
PEGENHANCER
DYPHYLLINE DYPHYLLINE LIPOSOMES
A 0 SUSPENSION
Flgure SEffect of carrier on the skin permeation fluxes of dyphylline incorporated in liposomes.
THEOPHYLLNE
Flgure &Skin permeation fluxes of dyphylline, liposomal dyphylline, and theophylline from a Carbopol gel base.
obtained when the PEG enhancer base was the carrier used for delivery (2.1 x lo-' mg * cm-' * h-'). Further, it was interesting to compare the permeation behavior of liposomal dyphylline with free dyphylline or theophylline incorporated in the three semisolid bases used in this study. Figures 4-6 illustrate the effect of the three bases on the permeation fluxes of the drugs. In the PEG base, dyphylline and liposomal dyphylline exhibited very similar permeation fluxes (Figure 4). A larger variation between the permeation fluxes of the drugs was seen when the drugs were incorporated in the Carbopol gel, with the highest value obtained for free dyphylline, followed by liposomal dyphylline, and the lowest value for 2.6 x and 7.2 x free theo hylline (5.7 x mg * a n-'. h-', respectively) (Figure 5). When the PEG enhancer base was used (Figure 61, a very high permeation of theophylline was observed (1.1 x lo-' mg * cm-' * h-'). The flux values for liposomal dyphylline and dyphyllinepermeating the skin from this base were much lower (2.1 x lo-' and 3.4 x lo-' mg cm-' * h-l, respectively).
T
DYPHYLLINE
DYPHYLLINE LIPOSOMES
THEWHYLLINE
Flgure +Skin permeation fluxes of dyphylline, liposomal dyphylline, and theophylline from a PEG enhancer base. w
4-
Since the goal of this work was to administer the drug locally into the skin without having significant transderma1 permeation, the formulations with the PEG base were further tested for drug partitioning (K,) between the skin and the carrier. As seen in Table I, the highest K,,,value was obtained for liposomal dyphylline. Although lower than the liposomal form, the skin partition coefficient of free dyphylline is higher relative to that of free theophyl-
W
0 r
x
-
r
S
3-
Y
5 . F
x-
a -I u
*.
Table CDrug Partltlonlng between the Skln and Carrler for Dyphylline and Theophylllne In Polyethylene Glycol
1 -
Drug 0 4 DYPHYLLINE OYPHYI I INE I IPOSOMCS
TIIEOPHYLLINE
Flgure &Skin permeation fluxes of dyphylline, liposomal dyphylline, and theophylline from a PEG base.
Dyphylline in liposomes Dyphylline Theophyllineb
& (mean 5 SD)" 0.98 2 0.01 0.59 2 0.08 0.10 5 0.02
The & values were statistically significant at p < 0.01. The theophylline in the PEG enhancer base was 0.29 2 0.08.
& of
Journal of Pharmaceutical Sciences I 133 Vol. 81, No. 2, February 1992
line. This high partitioning of dyphylline into the skin can be explained by the dihydroxypropyl side-chain of dyphylline, which may alter the solubility of the drug in the skin relative to theophylline.
Conclusions The results obtained in the present work show that the carrier may affect the degree of permeation of the drug incorporated in liposomes by up to two orders of magnitude, the lowest permeation being obtained in the PEG base. Therefore, by delivering dyphylline in liposomes incorporated in a PEG base, it is possible to obtain good skin partitioning with low transdermal delivery, so that the drug is localized in the skin. It can also be concluded that dyphylline is a good choice for preparing xanthine-derivative liposomes for local delivery to the skin due to its physical properties which enable drug incorporation into the liposome and a favorable skin partitioning from a chosen carrier.
References and Notes 1. Voorhees, J. J.; Duel1 E. A. Arch. Dermatol. 1971,104, 352358.
134 I Journal of Pharmaceutical Sciences Vol. 81, No. 2, February 7992
2. Voorhees, J.J. Arch. Dermntol. 1982,118,869-874. 3. Iancu, L.; Shneur, A.; Cohen, H. Dermatologica 1979, 159, 55-61. 4. Touitou, E.; Levy-Shaeffer. F.; Shaco-Ezra, N.; Ben-Yosef, R.; Fabin, B. Znt. J. Phurm. 1991,70,159-166. 5. Mezei, M.; Gulasekharam, V. J. Phurm. Pharmacol. 1982,34, 473-474. 6. Mezei, M. In Topics in Pharmaceutical Sciences; Breimer, D. D.; Speiser, P., Eds.; Elsevier Science: New York, 1985;pp 345-358. 7. Mezei, M. In Liposomes as Drug Carriers: Recent Trend and Progress; Gregoriadis, G., Ed.; John Wiley & Sons: New York, 1988;pp 663-617. 8. Gesztes, A.; Mezei, M. Anesth. Amlg. 1988,67, 1079-1981. 9. Szoka, F. C.;Papahadjopoulos, D. Proc. Natl. Acad. Sci., U.SA. 1978,75,4194-4198. 10. Touitou, E.; Fabin, 33.;Dany, S.; Almog, S.; Int. J. Pharm. 1988, 43,9-15. 11. Touitou, E. Znt. J. Pharm. 1986,33,3743. 12. ‘l’ouitou, E., unpublished results. 13. Giaclon, L.; Rowse, K.; Ayres, J. Res. Commun. Chem. Pathol. P h u r m o l . 1979,23,523-531. 14. Riaz, M.; Weiner, N.; Martin, F. In Pharmaceutical Dosage Forms. Dis rse S stems, Vol. 2; Lieberman, H.A.; Rieger, M. M.;B a n g r , G d , Eds.; Marcel Dekker: New York, 1988;pp ,567402.
investigated. Xanthines are inhibitors of cAMP phosphodiesterase and have been considered for treatment of psoriasis. Dyphylline was chosen because of its solubility in water, which should allow for incorporation of higher concentrations within the liposomes. Liposomes containing dyphylline were prepared by a method using sonication. Transmission electron micrograph (TEM) visualization showed small particles ranging from 40 to 100 nm, and particle size distribution determined by light scattering showed the vesicles to have an average diameter of 360 nm. The transdermal delivery of free dyphylline and dyphylline incorporated in unilamellar liposomes was measured from polyethylene glycol (PEG), Carbopol gel, a PEG enhacer base, and water. For comparison, similar experiments were carried out with theophylline as well. When the drugs were incorporated in Carbopol gel, a large difference was seen between their fluxes, with free dyphylline having the highest permeation, followed by liposomal dyphylline, and then theophylline. With the PEG enhancer base, a very high permeation of theophylline was observed relative to dyphylline and liposomal dyphylline. From the PEG base, liposomal dyphylline exhibited the lowest skin permeation flux relative to other bases. Using the PEG base for dyphylline incorporated in liposomes, a high skin partitioning of the drug, along with low transdermal permeation, was measured. These results may indicate that the drug is localized in the skin.
A number of works have shown that cyclic adenosine 3',5'-monophosphate (CAMP)is involved in differentiation of cells and a cessation of cell proliferation. Psoriasis, a chronic recurrent inflammatory disease of the skin, is characterized by an accelerated epidermal cell cycle; therefore, this disease may be caused by a defect in the cAMP cascade.1.2 Theophylline, a n inhibitor of the phosphodiesterase which inactivates CAMP,was therefore suggested a s a possible treatment for psoriasis. Its oral administration has met with limited success,3 most likely because a high systemic level of the drug is necessary in order for the local skin concentration to reach therapeutic levels. Topical delivery of effective concentrations of theophylline derivatives directly into the skin may be one way of overcoming this drawback. In a previous work, we showed that the skin permeation of theophylline can be modulated by using chemical enhancers and selected carriers. An increase of two orders of magnitude of theophylline flux through the hairless mouse skin was obtained in vitro by using a hydrophilic enhancing base containing oleic acid.4 With the final goal being topical administration of the drug, we sought a dosage form which allows for drug accumulation in the skin with low transderma1 permeation. In recent years, liposomes (microscopicvesicles composed of phospholipid bilayers) have become increasingly important as a vehicle for delivery of active compounds which are used for topical management. In a number of investigations, Mezei et a1.M showed that liposomes can be used to deliver drugs such as corticosteroids, econazole, progesterone, and tetracaine into the skin in greater quantities than conventional 0022-3549/92'0200-0 131$02.50/0 0 1992, American Pharmaceutical Association
vehicles, with localization at the desired site of action. In animal experiments, the liposomal form provided higher drug concentration in the skin and lower concentration in the internal organs than the conventional topical dosage forms. The present study was undertaken in order to determine the effectivenessof delivering a theophylline derivative to the skin via liposomes, for further use as a topical treatment for psoriasis. On the assumption that a high water solubility of the drug would allow for drug incorporation within the liposome, dyphylline, a water-soluble theophylline derivative, was chosen for our studies. The effect of the carrier base on dyphylline skin permeation was investigated. Similar experiments with theophylline, for comparison with the results using dyphylline, were carried out.
Experimental Sectidn Materials-Dyphylline, theophylline, and the formulation carriers [polyethylene glycol (PEG) 400, PEG 4000, oleic acid, Carbopol 934, and propylene glycol] were all analytical grade or conformed to B.P. requirements and were all purchased from Sigma. Diglycol was a gift of Gattefosse (France). The liposome ingredients were soya phosphatidylcholine (Phospholipon 901, a gift from Nattermann phospholipid GmbH, Germany, and cholesterol, purchased from Sigma. Semisolid Carriers for Dyphylline, Dyphylline in Liposomes, and Theophylline-The semisolid bases used for the drug incorporation were: (1)polyethylene glycol (PEG)base (a mixture of PEG 400 and PEG 4000); (2) gel base (a Carbopol934 hydrophilic gel); and (3) PEG enhancer base4 (PEG base with oleic acid and diglycol). The drug concentration in all the formulations was 10 mglg. Preparation of Liposomes-In the present study, dyphyllinecontaining liposomes were prepared by dissolving Phospholipon 90 and cholesterol in a 5:l ratio by weight in chloroform. The solvent was removed under reduced pressure in a rotary evaporator until a smooth dry lipid film was obtained. One milliliter of 5 8 (w/v) dyphylline in saline was added for each 100 mg of dry lipid film. The lipiddyphylline suspension was allowed to stand at room temperature for 2 h with occasional shaking, followed by 10 min of vigorous mixing using a Vibrofix VF, shaker. The suspension was sonicated in an ice bath using a Braun sonicator with a 1-cm-diameter titanium probe for 5 bursts of 2 min each, with a cooling period of 2 min in between, in the presence of N2.The final liposome suspension contained 5% (w/w) dyphylline. The entrapment efficiency of dyphylline in liposomes was determined by dialysis using the method described by Szoka and Rapahadjopoulos.9 Liposomes were tested by particle size distribution (see below) after 4 weeks of storage at 4 "C and found to be stable. Identification of Liposornes by Electron Microscopy-Negative stain electron micrographs were prepared on copper grids which were covered with a 0.5% solution of parlodion (Sigma) in amyl acetate and then coated with carbon. The grids were subjected to glow discharge for 20 seconds. Liposomes were applied to the grid for 30 seconds and drawn off with filter paper. A drop of a 1%solution of uranyl acetate (Sigma) was immediately applied to the grid and drawn off after 30 seconds with a piece of filter paper and allowed to dry for at least 30 minutes. A Philips EA 300 electron microscope at 80 KV was used. Particle Size Distribution Determinations-The particle size distribution (PSD) of liposomes was determined optically using a computerized inspection system (Malvern Instruments, USA). In this Journal of Pharmaceutical Sciences I 131 Vol. 81, No. 2, February 1992
system, a fixed volume of fluid-containing vesicles was scanned by a Ne-He laser beam. A computer program was used to correlate the light scattering signals to the vesicle size. The determination of sizes in the ranges 0.1-0.5 and 0.55.0pm was accomplished in one step. Skin Permeation Measurements-The permeation kinetics were determined on the hairless mouse skin using the Franz cell assembly. Full-thickness abdominal skin excised from six-week-old male hairless mice was mounted in the cell with a surface area of 1.77 cm2 and receiver volume of -8 mL. The formulation was applied on the stratum corneum of the skin. The experiment was carried out as previously described10 and run for 24 h. The dyphylline concentration in the samples waa assayed by HPLC. A computer program was used for the calculation of kinetic parameters.11 Each system was tested in three cells and duplicated. The significance of the results was determined using the two-tailed, unpaired or paired (depending on the set of experiments compared)t test. For this analysis, the "Balance" (IBM) computer program was used. Determination of the Partition Coefticient (K,) (Skin-Carriers) of the Drug-The experiment was carried out in diffusion cells at 37 "C, with the receiver compartment empty. Formulations of various compositions containing 10 mglmL of drug were applied to the skin in the donor compartment of the cells. "he concentration of the donor was measured at 0 time (C,) and at the end of the 24-h experiment (C). The dyphylline concentration was determined by HPLC as described below. The K, was calculated from:
(1) Control experiments with the liposomal preparation kept in the donors of the diffusion cells, where the skin was replaced by a glass disk (Crown Glass Company), were also run. It was found that the dyphylline concentration remained constant at the end of the 24-h experiment at 37 "C. Moreover, autoradiographic studies with dyphylline liposomes containing [3Hlcholesterol show that the liposome8 did penetrate the upper strata of the skin.'* Assay-The dyphylline concentration was determined using a Merck Hitachi HPLC with a 655A variable wavelength UV monitor. The determination was made at 275 nm on a reversed-phase C,, column. The mobile phase was composed of 8%acetonitrile in water and had a flow rate of 1 mumin. The internal standard used was /3-hydroxyethyltheophylline.l3There was no overlap between the peaks of the drug or standard and those of the metabolites and other skin componenta released into the receiver compartment during the experiment.
Results and Discussion Characterization of Dyphylline Liposomes-Active compound-carrying liposomes are prepared by a variety of techniques, such as sonication, vortex, ether injection, French press, or reversed-phase evaporation.'* By varying the method of preparation and lipid composition, liposomes with specific characteristics can be designed. Liposomes containing dyphylline were prepared by the sonication method described in the Experimental Section. The entrapment efficiency of dyphylline was 18%(w/w). Figure 1 is a transmission electron micrograph (TEM)of the liposomes, which shows vesicles with sizes ranging between 40 and 100 nm. In Figure 2, a differential number density distribution of the liposome population is shown. In this histogram, it can be seen that the average mean of the PSD is 363 nm, with 60% of the vesical population having a diameter falling between 189 and 739 nM. The literature describes liposomes which are obtained upon sonication of phospholipid dispersions and which have diameters ranging from 25 to 50 nm as small unilamellar vesicles (SUVs).Those unilamellar liposomes with diameters ranging from 100 to 500 nm are referred to as large unilamellar vesicles (LUVs).14From the PSD data, it would appear that, using the method of sonication to prepare the liposomes as described in this study, a mixture of LUVs and SUVs are 132 I Journal of Pharmaceutical Sciences Vol. 81, No.2, February 1992
.
*:+.5J.,:.
'
I
~
*
Flgure l-Liposomes of dyphylline visualized by E M ( x 40 OOO).
-
24.5 30.0
--i 8
-
95.8 120.2
-
298.0 374.0
iij
-
927.2 1163.5
-
4542.4 5699.9 Signal intensity (%)
Flgure 2-Size distribution of dyphylline liposornes. Key: (-) imrnediately after preparation;(- - -) after 4 weeks of storage at 4 "C. Samples were analyzed using a Malvern autosizer II c.
obtained. However, the electron micrograph (Figure 1)shows that the vesicles are mostly small. The relatively large mean average of the differential number density distribution can be attributed to agglomerations of vesicles, which may have been measured as single units using the Malvern autosizer, as is known to occur with light scattering methods.14 Such clusters of vesicles can be seen in Figure 1. Drug Delivery to the Skin-Liposomal dyphylline incorporated in the three bases (PEG, Carbopol gel, and PEG enhancer) and in aqueous suspension was tested for skin permeation, and the kinetic parameters were compared with those obtained in systems containing free dyphylline, as well as with systems containing theophylline. The skin permeation patterns a t steady state were similar for the various systems regardless of the dosage form or of the state of the drug (free or liposomal). However, the skin permeation flux did vary by almost three orders of magnitude accordingto the carrier, the dru , and its form of incorporation, and ranged from 2.7 x 10-8to 1.1 x lo-' mg cm-' h-'. The effect of the carrier on the drug permeation from dyphylline liposomes is shown in Figure 3. The results show that the carriers had different effects on the liposomal dymg cm-' h-') was phylline; the smallest flux (2.7 x obtained when the liposomes were incorporated into the PEG base, a larger flux was evident when in the Carbopol gel carrier (2.6 x 10-3mg * cm-' h-l) or in aqueous suspension (4.2 x 10-3mg.cm-2-h-'), and the highest value was
-
e
-
80
400
330
J
70
-
60
-
50
-
0 r
5 & r! 200
5
40-
E"
30-
3 2 U
100
20
-
10
-
0 4
0 PEG BASE
c€l
PEGENHANCER
DYPHYLLINE DYPHYLLINE LIPOSOMES
A 0 SUSPENSION
Flgure SEffect of carrier on the skin permeation fluxes of dyphylline incorporated in liposomes.
THEOPHYLLNE
Flgure &Skin permeation fluxes of dyphylline, liposomal dyphylline, and theophylline from a Carbopol gel base.
obtained when the PEG enhancer base was the carrier used for delivery (2.1 x lo-' mg * cm-' * h-'). Further, it was interesting to compare the permeation behavior of liposomal dyphylline with free dyphylline or theophylline incorporated in the three semisolid bases used in this study. Figures 4-6 illustrate the effect of the three bases on the permeation fluxes of the drugs. In the PEG base, dyphylline and liposomal dyphylline exhibited very similar permeation fluxes (Figure 4). A larger variation between the permeation fluxes of the drugs was seen when the drugs were incorporated in the Carbopol gel, with the highest value obtained for free dyphylline, followed by liposomal dyphylline, and the lowest value for 2.6 x and 7.2 x free theo hylline (5.7 x mg * a n-'. h-', respectively) (Figure 5). When the PEG enhancer base was used (Figure 61, a very high permeation of theophylline was observed (1.1 x lo-' mg * cm-' * h-'). The flux values for liposomal dyphylline and dyphyllinepermeating the skin from this base were much lower (2.1 x lo-' and 3.4 x lo-' mg cm-' * h-l, respectively).
T
DYPHYLLINE
DYPHYLLINE LIPOSOMES
THEWHYLLINE
Flgure +Skin permeation fluxes of dyphylline, liposomal dyphylline, and theophylline from a PEG enhancer base. w
4-
Since the goal of this work was to administer the drug locally into the skin without having significant transderma1 permeation, the formulations with the PEG base were further tested for drug partitioning (K,) between the skin and the carrier. As seen in Table I, the highest K,,,value was obtained for liposomal dyphylline. Although lower than the liposomal form, the skin partition coefficient of free dyphylline is higher relative to that of free theophyl-
W
0 r
x
-
r
S
3-
Y
5 . F
x-
a -I u
*.
Table CDrug Partltlonlng between the Skln and Carrler for Dyphylline and Theophylllne In Polyethylene Glycol
1 -
Drug 0 4 DYPHYLLINE OYPHYI I INE I IPOSOMCS
TIIEOPHYLLINE
Flgure &Skin permeation fluxes of dyphylline, liposomal dyphylline, and theophylline from a PEG base.
Dyphylline in liposomes Dyphylline Theophyllineb
& (mean 5 SD)" 0.98 2 0.01 0.59 2 0.08 0.10 5 0.02
The & values were statistically significant at p < 0.01. The theophylline in the PEG enhancer base was 0.29 2 0.08.
& of
Journal of Pharmaceutical Sciences I 133 Vol. 81, No. 2, February 1992
line. This high partitioning of dyphylline into the skin can be explained by the dihydroxypropyl side-chain of dyphylline, which may alter the solubility of the drug in the skin relative to theophylline.
Conclusions The results obtained in the present work show that the carrier may affect the degree of permeation of the drug incorporated in liposomes by up to two orders of magnitude, the lowest permeation being obtained in the PEG base. Therefore, by delivering dyphylline in liposomes incorporated in a PEG base, it is possible to obtain good skin partitioning with low transdermal delivery, so that the drug is localized in the skin. It can also be concluded that dyphylline is a good choice for preparing xanthine-derivative liposomes for local delivery to the skin due to its physical properties which enable drug incorporation into the liposome and a favorable skin partitioning from a chosen carrier.
References and Notes 1. Voorhees, J. J.; Duel1 E. A. Arch. Dermatol. 1971,104, 352358.
134 I Journal of Pharmaceutical Sciences Vol. 81, No. 2, February 7992
2. Voorhees, J.J. Arch. Dermntol. 1982,118,869-874. 3. Iancu, L.; Shneur, A.; Cohen, H. Dermatologica 1979, 159, 55-61. 4. Touitou, E.; Levy-Shaeffer. F.; Shaco-Ezra, N.; Ben-Yosef, R.; Fabin, B. Znt. J. Phurm. 1991,70,159-166. 5. Mezei, M.; Gulasekharam, V. J. Phurm. Pharmacol. 1982,34, 473-474. 6. Mezei, M. In Topics in Pharmaceutical Sciences; Breimer, D. D.; Speiser, P., Eds.; Elsevier Science: New York, 1985;pp 345-358. 7. Mezei, M. In Liposomes as Drug Carriers: Recent Trend and Progress; Gregoriadis, G., Ed.; John Wiley & Sons: New York, 1988;pp 663-617. 8. Gesztes, A.; Mezei, M. Anesth. Amlg. 1988,67, 1079-1981. 9. Szoka, F. C.;Papahadjopoulos, D. Proc. Natl. Acad. Sci., U.SA. 1978,75,4194-4198. 10. Touitou, E.; Fabin, 33.;Dany, S.; Almog, S.; Int. J. Pharm. 1988, 43,9-15. 11. Touitou, E. Znt. J. Pharm. 1986,33,3743. 12. ‘l’ouitou, E., unpublished results. 13. Giaclon, L.; Rowse, K.; Ayres, J. Res. Commun. Chem. Pathol. P h u r m o l . 1979,23,523-531. 14. Riaz, M.; Weiner, N.; Martin, F. In Pharmaceutical Dosage Forms. Dis rse S stems, Vol. 2; Lieberman, H.A.; Rieger, M. M.;B a n g r , G d , Eds.; Marcel Dekker: New York, 1988;pp ,567402.
