Liquid crystalline pharmacogel based enhanced transdermal delivery of propranolol hydrochloride

Journal of Controlled Release 82 (2002) 223–236 www.elsevier.com / locate / jconrel

Liquid crystalline pharmacogel based enhanced transdermal delivery of propranolol hydrochloride Alok Namdeo, N.K. Jain* Novel Drug Delivery Research Laboratory, Department of Pharmaceutical Sciences, Dr. H.S. Gour University, Sagar ( M.P.) 470 003, India Received 20 January 2002; accepted 4 March 2002

Abstract A novel pharmacogel was developed for the enhanced transdermal delivery of propranolol hydrochloride (PH). The synthesized prodrugs, propranolol palmitate hydrochloride (PPH) and propranolol stearate hydrochloride (PSH) selfassembled to form gel simply upon mixing alcoholic solution of prodrug with an aqueous solution in a specified ratio. By varying the ratio of prodrug, alcohol and water, three-component phase diagram was constructed which revealed isotropic–gel–vesicular dispersion regions, respectively concomitant to increasing the ratio of water. The gel phase is termed ‘Pharmacogel’ and exhibits birefringence under plane-polarized light corroborating the presence of lamellar liquid crystals. The pharmacogel by virtue of high chemical potential gradient and improved physicochemical properties showed the enhanced in-vitro skin permeation flux of 51.563.7 and 42.563.1 mg / cm 2 / h from PPH and PSH gel, respectively, as compared to 1.960.1 mg / cm 2 / h for control; and decrease in lag time (1.8 and 2.8 h for PPH and PSH gel, respectively) compared to control (7.6 h) was observed. The admixing of egg lecithin (EL) in increasing ratio concomitantly decreased the flux values to 31.762.1 mg / cm 2 / h (at a mole ratio of 50:50 PPH:EL) and increased the lag time. In the gel containing 50% EL, the addition of span 40 and cholesterol slightly reduced the permeation while sodium deoxycholate and Tween-80 improved it. The plasma drug levels following transdermal application of control were low (Cmax 523 ng / ml) while in PPH gel, it increased with time reaching Cmax of 94 ng / ml at 8 h post-application of PPH gel (Cmax of 75 ng / ml at 12 h post application of PL5 gel) and maintained for longer times. The AUC 0 – 32 h for PPH gel was much higher (1968 ng h / ml) than control (AUC 0 – 18 h was 239 ng h / ml), while EL mixed gel also showed better absorption (AUC 0 – 32 h was 1707 ng h / ml). The gel formulations also caused less irritation than control, while mixed gel showed least irritation. This novel self-assembled pharmacogel providing high transdermal permeation with many variables to regulate the delivery is therefore having a great potential in percutaneous delivery.  2002 Elsevier Science B.V. All rights reserved. Keywords: Prodrug; Liquid crystals; Propranolol hydrochloride; Egg lecithin; Transdermal delivery

1. Introduction

*Corresponding author. Fax: 191-7582-23712. E-mail address: [email protected] (N.K. Jain).

Propranolol, one of the most widely prescribed b-blockers in the long-term treatment of hypertension and in psychotherapy is usually taken orally,

0168-3659 / 02 / $ – see front matter  2002 Elsevier Science B.V. All rights reserved. PII: S0168-3659( 02 )00106-2

224

A. Namdeo, N.K. Jain / Journal of Controlled Release 82 (2002) 223 – 236

although an intravenous form is available for acute administration. Following oral administration, propranolol is rapidly and completely absorbed from gastrointestinal tract (less than 5% recovered in feces) [1], still the oral bioavailability is low (30%) [2] because of significant first pass hepatic metabolism by CYP2C19 and 2D6 (urinary recovery as unchanged propranolol is less than 1% of the administered dose) [3]. Moreover, in healthy individuals and in patients with various disease states, plasma concentration varies as much as 10–20-fold following oral and not intravenous administration due to difference in the first pass effect. Available sustained release dosage of oral propranolol is not as effective as the equivalent amount of single dosages because slower oral absorption leads to greater hepatic metabolism. Administration of drugs via transdermal route effectively bypasses first pass metabolism, along with providing sustained delivery and therapy can easily be terminated by removing the patch if toxicity ensues. The patch can be applied nearer to the heart where it is needed and also increases patient compliance. Therefore, percutaneous application is ideal for propranolol, which exhibits short biological half-life of 3–4 h [4]. The success of a transdermal drug delivery system however depends on the ability of the drug to permeate the skin in sufficient quantity to maintain the therapeutic levels. But, the transdermal absorption of propranolol is poor and several approaches have been undertaken to improve the permeation [5–8]. Several methods have been reported in the literature to enhance the skin permeation such as incorporation of permeation enhancers, iontophoresis, liposomal / niosomal incorporation, prodrug formation, etc. Ahead of conventionals, an attractive sum of literature has surged concerning the use of liposome / niosome vesicles and organogels, acting as vehicle and / or as permeation enhancer. The liposomes and organogels are distinct structures formed by altering the thermodynamic variables (i.e., concentration of phospholipid) in the system. The efficiency of delivery from vesicles is primarily governed by their fusion with skin; and vesicle–skin interaction studies have concluded that the phospholipids per se apparently do not penetrate deeper into the skin, therefore researchers have focused mainly on dermal delivery of corticosteroids using

this carrier [9]. Both the system however achieve limited efficacy in transdermal delivery. The prodrug strategy involving bioreversible chemical modification to modify the physicochemical property of a drug is also being increasingly used to optimize dermal and transdermal delivery because regeneration of parent compound occurs by the enzymatic processes. A promising prodrug strategy is to derivatize the drug with a promoiety that possesses inherent ability to enhance the permeation [10]. Moreover, for successful transdermal delivery from topical formulation, drug should dissolve in vehicle to reach vehicle–skin interface, then partition and diffuse through the lipophilic stratum corneum and again partition and diffuse through the hydrophilic viable epidermis and dermis before uptake by the cutaneous microcirculation. This requires optimal balance of solubility. Propranolol hydrochloride, the commercially available form of propranolol is hydrophilic in nature. We synthesized its prodrug by covalently conjugating with free fatty acids, i.e., stearic acid and palmitic acid (lipophilic). This prodrug in combination with a definite ratio of water and ethanol could be formulated as a smooth transparent gel. The drug (pharmakon) is part of the gel forming amphipath; therefore we termed the system as pharmacogel. Examination of gel under polarized light microscope revealed birefringence, which is characteristic of lamellar liquid crystal, implying that the prodrug assembles in liquid crystalline lattice. The liquid crystalline materials possess the crystalline aspect of periodicity but the molecules are not locked rigidly in a lattice, instead they exhibit liquid like diffusion. The major barrier in skin permeation is outer stratum corneum due to presence of the intercellular lipids, arranged in lamellar sheets. Prodrug might intercalate and fluidize the intercellular lipids and increase its own permeation through this layer of skin. Prodrug bearing hydrophilic and lipophilic portion acquires amphiphilic character, therefore is supposed to permeate better through both the lipophilic and hydrophilic layers of the skin. Hydrolysis of prodrug in the skin will produce fatty acids, which are known to act as penetration enhancers [5]. Moreover, the pharmacogel can very well accommodate other vehicles such as phospholipids allowing modulation in the release behavior. The distinguishing feature of

A. Namdeo, N.K. Jain / Journal of Controlled Release 82 (2002) 223 – 236

this novel pharmacogel therefore becomes that the formulation can be developed without any additional ingredients or vehicles thereby provides high chemical potential gradient enabling enhanced permeation and the drug is available for complete absorption on placing in direct contact with the skin. Additives like phospholipids, etc., can be incorporated for further control of release pattern and further addition of water transforms the gel into vesicles without any precipitation, ensuring stability of the system in presence of skin hydration. Unlike liposome, wherein leakage of entrapped cargo may result in failure of effective permeation, the pharmacogel acts at molecular level where partitioning and absorption of a single prodrug molecule also provides transdermal delivery. The present study therefore explores the assembling of the synthesized amphiphilic prodrug into gel and vesicular phases; potential of the gel phase (termed pharmacogel) in the transdermal delivery and modification of the flux by additives and subsequent in-vivo studies.

2. Materials Egg lecithin was purchased from Sigma (USA) and palmitoyl / stearoyl chloride from E. Merck (India). Propranolol hydrochloride was generously supplied as gift sample by Cipla (Mumbai, India). The other chemicals and organic solvents were of synthesis / analytical-reagent grade.

3. Methods

3.1. Synthesis of prodrug Prodrug was synthesized as reported previously [11] by refluxing propranolol hydrochloride and stearoyl / palmitoyl chloride in ethanol-free chloroform and washed with water; the chloroform was dried over anhydrous sodium sulfate and evaporated to get the compound which was purified on column packed with silica gel (E. Merck). The elemental analysis, IR and nuclear magnetic resonance (NMR) confirmed the formation of prodrug. The partition coefficient (PC) was determined by equilibrating the prodrug between chloroform–water and estimating

225

the prodrug content in both the phases by highperformance liquid chromatography (HPLC).

3.2. Preparation of the phase diagram The three-component phase diagram of prodrug– ethanol–water was constructed in order to define mesomorphic assemblage of amphiphilic prodrug in viscous gel, vesicular dispersion or isotropic solution depending upon the thermodynamic variable, i.e., concentration of the components. For this, alcoholic solutions of prodrug was prepared in a glass tube covered with a lid to prevent loss of solvent and warmed on a water bath at 40–50 8C to dissolve the content. The measured quantity of aqueous phase (water) was then added and warmed on a water bath till clear solution results. The solution was allowed to cool with shaking at room temperature, and kept without shaking for 30 min to equilibrate. The procedure was repeated until a phase change was observed. The boundary of different regions of clear isotropic–viscous gel–vesicular region was examined initially macroscopically and then under the microscope. The gel was also prepared by admixing egg lecithin and permeation enhancers as shown in Table 1.

3.3. Microscopic evaluation and vesicle characterization A thin layer of gel was spread in a cavity slide and after placing a cover slip, was observed under microscope with and without polarizing light (Leica DMLB, Germany). The birefringence emanated due to the presence of anisotropic lamellar liquid crystals was observed and photomicrographs were taken. The gel (100 mg) was hydrated in a small glass tube using 10 ml of water with manual shaking for 5 min; size of the resulting vesicles was determined by Malvern Mastersizer (UK). To verify the presence of vesicles, the gel was hydrated with solution of 6carboxyfluorecein (6-CF). In the resulting dispersion, free 6-CF was removed by gel permeation chromatography (G-25; Pharmacia, Sweden) using HEPES buffer as eluent. Fractions containing vesicles were collected and disrupted by adding ethanol; released free 6-CF was determined by spectrofluorimeter at an

226

A. Namdeo, N.K. Jain / Journal of Controlled Release 82 (2002) 223 – 236

Table 1 Effect of formulation composition on mean size of vesicular dispersion Code

Composition

Mean vesicle size (nm)

Skewness (v.s.)6s.d. (nm)

PDI

PSH PPH PL1 PL2 PL3 PL4 PL5 PLD PLT PLS PLC

PSH PPH PPH:EL (90:10) PPH:EL (80:20) PPH:EL (70:30) PPH:EL (60:40) PPH:EL (50:50) PPH:EL:SD (47.5:47.5:5) PPH:EL:Tw80 (47.5:47.5:5) PPH:EL:Sp60 (45:45:10) PPH:EL:CHL (45:45:10)

2616258 2846244 3146267 3936316 4636342 5436375 6126318 5736374 5856413 6386496 6516520

0.409 0.387 0.365 0.315 0.285 0.212 0.185 0.205 0.212 0.185 0.121

0.958 0.859 0.857 0.816 0.767 0.755 0.742 0.685 0.715 0.764 0.638

*Results are mean of three observations. v.s. shows mean vesicle size; s.d. indicates standard deviation; PDI is polydispersity index calculated as v.s. / s.d.; skewness is Karl Pearsons coefficient of skewness obtained using formula5(mean2mode) / S.D. SD is sodium deoxycholate; Tw80 is Tween-80; Sp60 is Span60 and CHL is cholesterol.

excitation wavelength of 490 nm and an emission wavelength of 520 nm.

3.4. In vitro skin permeation studies 3.4.1. Fabrication of transdermal patch In a flexible plastic sheet of 1.0 mm thickness, a circular hole of 1.369 cm 2 was cut in the center. This was stuck with adhesive onto a circular aluminum foil of diameter 2.5 cm, to act as a backing membrane. The gel or ointment was uniformly spread over this area and covered with a fine nylon mesh. 3.4.2. Preparation of rat skin The hair of the back area of albino rats (Sprauge– Dawley strain) were carefully removed 24 h prior to use, first by clipping with scissors and then with depilatory (Anne French, India) applied for 10 min; and then wiped off with cotton. A portion (about 333 cm) of the full thickness skin was carefully excised. The dermal side of the skin was carefully cleared of any adhering subcutaneous tissues and blood vessels. The prepared skin was then washed with saline and used afresh. 3.4.3. Skin permeation studies In-vitro skin permeation studies were performed using Keshary–Chein type diffusion cell. The patch bearing gel (containing 40 mg prodrug) or control

(40 mg drug in carbopol gel) was applied over the stratum corneum surface of the skin and mounted in the cell (reservoir volume 10 ml, area exposed to donor compartment 1.369 cm 2 ). The receptor medium was 50% ethanol in water (hydro–alcoholic) solution to maintain the sink condition. The receptor solution was maintained at 3761 8C by surrounding water-jacket and stirred by magnetic bead operated on a magnetic stirrer (Expo, India). The aliquots of the receptor fluid were withdrawn periodically up to 48 h, every time replacing with fresh solution and samples were analyzed by HPLC using the method reported by Irwin and Belaid [12] for propranolol prodrugs. The mobile phase consisted of acetonitrile–diethylamine–88% phosphoric acid–water (88:0.6:0.2:11.2, pH 2.8) and detection was at 290 nm. The studies were performed for different formulations and to determine the effect of phospholipids, etc., on permeation.

3.4.4. Analysis of data The in-vitro penetration parameters were calculated from in-vitro permeation data using the following equations: DKm Cs Q Jss 5 ] 5 ]] 5 Kp Cs AT d

(1)

d2 ] 6t

(2)

A. Namdeo, N.K. Jain / Journal of Controlled Release 82 (2002) 223 – 236

Kpd Km 5 ]] D

(3)

where Jss is steady state flux measured as the slope of the profile after regression analysis. The X-intercept of the extrapolated linear region of the curve gives t (lag time); Kp is apparent permeability coefficient through the skin calculated by dividing Jss by the employed concentration Cs using Eq. (1). The diffusion constant within the skin (D) was calculated from t with known thickness of skin (d ) taken as 0.092 cm using Eq. (2). The partition coefficient (Km ) was calculated using Eq. (3).

3.5. In-vivo evaluation The hair on the dorsal side of albino rats (Sprauge–Dawley strain) of either sex weighing 150–175 g were removed as described above. The transdermal patch bearing PPH Gel, PL5 Gel or control (propranolol in carbopol gel) was then applied and fixed in position using dressing tape. The blood samples were collected from retro-orbital plexus of eye at defined time interval and analyzed by HPLC by the reported method of Drummer et al. [13]. Briefly, after separating the plasma (50 ml) in a small tube and adding labetalol as internal standard, 100 ml of sodium carbonate was added and drug was extracted into ether, dried and residue reconstituted in mobile phase containing 50% methanol in 10 mM monobasic potassium phosphate buffer adjusted to pH 3.5 and detected fluorometrically at excitation and emission wavelengths of 295 and 360 nm, respectively. For skin irritation studies, albino rabbits were maintained in a controlled room, hair of the back was removed using Anne French and clipping; next day transdermal patch of control, PPH gel and PL5 gel was applied. After 32 h, patch was removed and skin scores awarded by visual observations.

4. Results and discussion The covalent attachment of hydrophilic drug with lipophilic carrier resulted in amphiphilic compound, which is not only ideal for transdermal delivery but also exhibits lyotropic mesomorphism. The word

227

‘lyotropic’ conveys the information that the system has more than one component (e.g., lipid and water) and that the resulting structure depends on the ratios of the three components. The word mesomorphism denotes the fact that the system may assume several structurally distinct phases. The phenomenon has been studied for phospholipids [14,15], which forms the semisolid organogel or bilayer vesicles depending on the variables in the system. The prodrug was also tensioactive and forms gel (termed pharmacogel) in a specified ratio of prodrug, ethanol and water. However, if thermodynamic variables are varied too far, there appears a distinctly different vesicular phase, representative of co-operative structural rearrangement in the system. In this paper we report on the formulation and optimization of unique pharmacogel, a smooth transparent gel for direct application on the skin. The topical application in general requires incorporation of drug into vehicle, which then determines diffusion of the drug to vehicle skin interface and vehicle skin partitioning. These factors substantially limit the percutaneous absorption. In case of pharmacogel, prodrug assembled in gel phase remains in direct contact with the skin allowing immediate transfer into the skin. The study comprised optimization of gel phase by threecomponent phase diagram, extensive in-vitro skin permeation studies and in-vivo evaluation.

4.1. Phase diagram studies The triangular phase diagram for lipid–ethanol– water system is shown in Fig. 1a, b, wherein, the number and types of phases were examined and the percentage composition at which a phase change occurred was calculated. The phase diagram has been divided into three fields namely, isotropic solution (I), viscous gel region (G), and vesicular dispersion region (V). As apparent from the diagram, initial addition of water forms solution of prodrug in aqueous ethanol, whereas further increase in water content forms viscous gel which transforms into vesicular dispersion at higher dilutions. Perret et al. [16] constructed the phase diagram for phospholid–ethanol–water at room temperature and reported that continuous addition of water to ethanolic solution of phospholipid follows isotropic to precipitated bilayer to liposomal dispersion region,

228

A. Namdeo, N.K. Jain / Journal of Controlled Release 82 (2002) 223 – 236

concentrations as shown in Fig. 1b for the phase diagram of prodrug:EL (1:1 mole ratio) lipid system. The position of boundaries between different regions was difficult to determine as the region themselves are not completely homogeneous at the boundaries. Dissolved lipid and hydrated lipid coexisted in the neighborhood of the border between the clear solution and the viscous region. The same was more difficult between viscous gel region and vesicular dispersion because saturation of polar head region had allowed stripping of some of the vesicles while rest of the mass was still in loosely held liquid crystalline region therefore giving the mass of considerably lower viscosity. The gel composition possessing consistency ascribed to middle phase as described by Rosevear [17] was selected for the preparation of gel. Vora et al. [18] drew schematics for transformation of liquid crystalline proniosomal gel into niosomal vesicles upon hydration proposing different degree of hydration of surfactant molecules. Initially, due to limited solvent system present, the gel contains mixture of lamellar liquid crystals resembling palisades and vesiculating lamellae linked together, further addition of water leads to swelling of bilayers as well as vesicles due to interaction of water with polar groups of the surfactant and above a limiting concentration of solvent, gives rise to vesicles.

4.2. Microscopic and size distribution studies

Fig. 1. Three-component phase diagram for lipid–water–ethanol system. (a) For prodrug (PPH)–ethanol–water system, (b) PPH:EL (1:1 molar ratio)–ethanol–water system; (I) isotropic region; (G) viscous gel phase region; (V) vesicular dispersion region.

respectively without presence of gel region. The same pattern however was observed with prodrug if treated at room temperature. Thus the water was added to the hot alcoholic solution of prodrug followed by cooling to form gel. The pharmacogel could only be formulated at higher concentration of prodrug probably due to single fatty acid chain than double chain in phospholipid, therefore the incorporation of egg lecithin formed the gel at lower

The various regions of phase diagram could also be differentiated by observing under cross polarizers. In the isotropic region darkness was prevailing, while gel region could be recognized by prominently displayed birefringence, which is characteristic of anisotropic lamellar liquid crystals. Fig. 2a shows the photomicrograph of birefringence emanated from PL5 gel while the transformation into vesicles is shown in Fig. 2b. The exhibited birefringence is widespread as observed in lecithin liquid crystals [15] and also forms the positive radiating units (appear as flower like in photomicrograph) as classified by Rosevear [17]. These are classic features of lyotropic lamellar phases. For preparing the gel, dilution was arrested in this liquid crystal region. Further dilution transformed the gel into vesicular dispersion and the size distribution profile of PPH

A. Namdeo, N.K. Jain / Journal of Controlled Release 82 (2002) 223 – 236

229

Fig. 3. Size distribution profile of PPH type vesicular dispersion.

Fig. 2. (a) Photomicrograph showing birefringence form PL5 gel formulation (magnification 453); (b) photomicrograph showing transformation of gel into vesicles (magnification 703).

type vesicular dispersion obtained on hydration with agitation is shown in Fig. 3. This shows a bimodal distribution with most of the vesicles having diameter below 0.4 mm and smaller proportion of larger vesicles was present; mean vesicle size was found to be 2846244 nm. The distribution profile of PS did not vary much and mean vesicle size (MVS) was found to be 2616258 nm. The mean vesicle size and statistical interpretations of pharmacosomes of different compositions are shown in Table 1. The degree of hydration changed the vesicle size as shown in Fig. 4. At a fixed concentration of lipid, an increase in volume of water in the system resulted in a decrease in particle size of lipid vesicles. This may be because of stronger hydrophobic bonding among molecules in presence of water than in ethanol. The formation of vesicles by increased water content ensured that prodrug would not precipitate in

higher concentration of water. When we centrifuged the vesicular dispersion, lysed the pelleted vesicles and analyzed by HPLC, nearly 100% prodrug was present indicating that the vesicles formed are of prodrug and is not of drug lipid mixture. In one more experiment, the free fatty acid, i.e., palmitic acid or stearic acid was included in the formulation, above 20% ratio of free acids aggregates were observed under the microscope. The combination of drug (propranolol hydrochloride) and fatty acid again did not form the vesicles and only aggregates were observed under the microscope. These studies verify that the vesicles are supromolecular organization of prodrug and not the assembly of drug–lipid mixture.

Fig. 4. Effect of dilution on mean vesicle size of PL5 type formulation.

230

A. Namdeo, N.K. Jain / Journal of Controlled Release 82 (2002) 223 – 236

In one more experiment, prodrug was hydrated with solution of 6-carboxyfluorecein as per hand-shaking method, the resulting dispersion was passed through a Sephadex G-50 column, eluted vesicles lysed and analyzed for 6-carboxyfluorescein on a fluorometer. Presence of 6-carboxyfluorscein was detected; indicating the encapsulation of water-soluble dye thereby verifies the presence of lamellar structures. The effect of change in composition by including the components, i.e., phospholipids, bile salts etc. on vesicle size is shown in Table 1. Phospholipid themselves form liquid crystals thus can be successfully incorporated as additives, the hydrated vesicles showed increase in size as the ratio of egg lecithin (EL) was increased. The gel of lecithin mixed prodrug is supposed to bear intercalated prodrug and phospholipid molecules. Upon addition of bile salts, the vesicles formed were smaller in size and only up to 10% molar ratio could be successfully incorporated. Incorporation of span 60 and cholesterol increased the vesicle size. The formulation containing pure prodrug readily transformed into vesicles as compared to EL mixed formulation. The comparatively higher polarity of prodrug must have allowed faster hydration and easy stripping of vesicles from liquid crystalline mass while the presence of EL must have slowed this process. Addition of sodium deoxycholate and Tween-80 further increased the formation of vesicles.

4.3. In-vitro skin permeation studies The developed formulation must be evaluated by performing actual studies in humans. This is however costly, time consuming and requires efficient safeguards and approved clinical protocols [19]. Instead, the in-vitro skin permeation studies correlate with in-vivo performance as they dictate the amount of drug available for absorption. Further, these are economic, readily available and remarkable similarity in absorption between the skin of human and skin of hairless mouse has been shown by Stroughton [20]. Skin is a highly metabolic organ containing various esterases. Yu et al. [21] have used hairless mouse skin as a model for the study of esterase activity in transdermal delivery and found that esterase activity is much higher in outer layer half

than the other half of the skin. Due to high metabolic activity of skin, transdermal delivery through prodrug has been successfully employed for many drugs. The palmitate and stearate prodrugs of propranolol might also hydrolyze by non-specific esterases present in the skin. The in-vitro skin permeation studies were therefore carried out in detail to evaluate the effect of adjuvants and scrutinize the optimized formulation. The freshly removed full thickness skin was used to estimate complete transfer from both lipophilic and hydrophilic portions of stratum corneum and dermis layer, respectively. Hydro–alcoholic (50% ethanol in water) solution was used in receiver compartment to maintain the sink conditions, which has earlier been used successfully for skin permeation studies on indomethacin prodrug [22,23]. Fig. 5 shows the cumulative permeated amount of propranolol and the prodrug (sum of intact prodrug and delivered propranolol) plotted against time. The resulting curve indicates much higher release from pharmacogel than control. The control containing propranolol hydrochloride in gel showed negligible release after 24 h, therefore curve has been shown for release from gel containing propranolol base, this released merely 81 mg in 32 h, which is in accordance with earlier findings. This low permeation transformed theoretically ideal transderaml delivery becoming a formidable task, and among others Ogiso and Shintani [5] used lipids and Chesnoy et al. [7] applied iontophoresis in the presence of enhancers to improve the transdermal delivery. Addition of palmitic acid or stearic acid to the control increased the flux value, however it was substantially less than the pharmacogel indicating that the permeation enhancement by pharmacogel is not due to the effect of free fatty acids on the skin. Various factors might be contributing for higher flux from pharmacogels such as the prodrug shows log PC of 2.18 (PPH) and 2.22 (PSH) and possess low melting point (37 8C and 39 8C for PSH and PPH, respectively), which are favorable as log PC between 1 and 3 is optimum for dermal delivery and decrease in melting point leads to concomitant increase in transdermal absorption [24]; high thermodynamic activity from pharmacogel results in greater constant driving force; and the prodrug being amphiphilic, may act as penetration enhancer.

A. Namdeo, N.K. Jain / Journal of Controlled Release 82 (2002) 223 – 236

231

Fig. 5. Effect of addition of stearic acid (SA) or palmitic acid (PA) in control and type of prodrug in pharmacogel on permeation across the rat skin (n53).

Nonhomogeneous distribution of esterases must have allowed greater hydrolysis in outer skin forming fatty acids which are known to improve the absorption while hydrolyzed less lipophilic drug will tend to permeate faster through underlined hydrophilic viable epidermis and dermis. The mathematical interpretation of curves in terms of flux values, enhancement ratio, lag time, permeability coefficient, partition coefficient and diffusion constant are given in Table 2. The flux value for control of 1.960.1 has increased considerably in pharmacogel showing enhancement ratios of 27.1 for PPH and 22.4 for PSH pharmacogel. There were 71- and 59-fold increases in permeability and 16.9- and 21.7-fold increases in partition coefficient for PPH and PSH gel, respectively, as compared to control. Due to higher permeability and greater partitioning from pharmacogel, lag time had reduced as compared to control. A lag period of 7.16 h was observed for control, which had reduced to 2.8 h for PPH and 1.8 h for PSH gel. The PSH is therefore partitioning faster into the skin supposedly due to greater lipophilicity while PPH is permeating faster therefore showing greater flux value. Hence the data suggest that both the prodrugs are promis-

ing, especially if the goal is to achieve greater permeation with a lower amount of drug. Greater penetration with a lower concentration of drug is always preferable as Kobayashi et al. [25] has shown increase in skin irritation on increasing drug concentration in the transdermal device. Furthermore, previous reports on permeation enhancement were conducted by placing solution of drug and enhancer in donor solution under stirring condition. The observed higher fluxes in such cases do not estimate drug vehicle interaction and vehicle skin transfer of drug. The considerably higher flux values of hydrophilic PH were reported in-vitro from solution of drug containing terpenes as penetration enhancer, however same when formulated in hydrogel could not show much difference [8]. Thus pharmacogel by virtue of its ability to improve the absorption from its applicable formulation is more encouraging. Fig. 6 shows the graph indicating the amount of permeated prodrug and drug with time. PSH showed 21.34 and 12.4% of hydrolyzed free drug at 4 and 32 h, respectively, while PPH showed 26.9 and 17.9% of hydrolyzed free drug at 4 and 32 h, respectively. The percent of hydrolyzed drug is higher at initial hours but decreases with time. The low hydrolysis

A. Namdeo, N.K. Jain / Journal of Controlled Release 82 (2002) 223 – 236

232

Table 2 Mathematical interpretation of in-vitro skin permeation data D 310 24 f (cm 2 / h)

0.09 0.47 0.39

1.85 5.2 4.5

12.860.9 10.660.7

1.5 1.9

7.79 4.98

1.7 1.9 2.1 2.3 2.8

12.260.9 10.960.8 9.960.6 8.960.6 7.960.5

1.4 1.4 1.3 1.3 2.3

8.21 7.12 6.85 6.21 4.98

2.6 2.7 2.9 3.1

8.560.5 8.260.5 7.860.5 7.760.5

1.5 1.4 1.5 1.5

5.32 5.32 4.72 4.62

ic (h)

1.960.1 8.961.9 7.861.4

– 4.7 4.1

7.6 2.7 3.1

Effect of type of prodrug PPH 51.563.7 PSH 42.563.1

27.1 22.4

1.8 2.8

Effect of egg lecithin PL1 PL2 PL3 PL4 PL5

48.963.5 43.963.1 39.762.5 35.662.3 31.762.1

25.7 23.1 20.9 18.7 16.7

Effect of adjuvant PLD PLT PLS PLC

33.962.1 32.762.1 31.162.0 30.661.9

17.8 17.2 16.4 16.1

Control Ct-PA

Jss a

Km 310 21 e

ERb

Formulation

Kp 310 24 d (cm / h) 0.1860.01 2.260.5 1.960.4

a

Steady state transdermal flux. Enhancement ratio calculated as Jss (Forml) /Jss (Control). c Lag time. d Permeability coefficient. e Partition coefficient. f Diffusion constant. b

can be attributed to the small area of skin in contact with the formulation as well as the deterioration in enzyme activity with time. However, it includes the possibility of hydrolysis of prodrug both in the skin and in the plasma. PPH gel showing higher flux values and greater hydrolysis were therefore selected for further studies. Liquid crystal system of lecithin has been studies earlier [14,15]. Vora et al. [18] successfully formulated proniosomes from mixed lecithin non-ionic surfactant system. Thus, lecithin was selected as adjuvant and mixed prodrug lecithin gel was prepared. The lecithin could be mixed in prodrug in different molar ratios, the release has been studied for systems containing 10–50% (mole) of lecithin and graph is shown in Fig. 7. With increase in the ratio of egg lecithin, there was proportionate decrease in the flux values. Similarly, there was increase in lag time and decrease in permeability coefficient, diffusion constant on increasing the egg

lecithin ratio. The decrease in chemical potential gradient on increasing the egg lecithin ratio along with the formation of matrix of egg lecithin must have been responsible for the decrease in permeation. This however introduces an important parameter to obtain the desired flux by optimizing the combination. The sodium deoxycholate (SD), Tween-80, span 40 and cholesterol were incorporated in PL5 formulation and effect on permeation is shown in Fig. 8. The addition of span and cholesterol slightly reduced the flux while SD and Tween slightly improved it. Sodium deoxycholate acts as penetration enhancer hence it must have improved the permeation while Tween was found to reduce the viscosity and by virtue of improved hydrophilicity might have improved the permeation. The cholesterol and span by increasing the compactness of the matrix could have reduced the permeation. The formulations thus show the flux values in following order: PPH.PSH.

A. Namdeo, N.K. Jain / Journal of Controlled Release 82 (2002) 223 – 236

Fig. 6. Cumulative amount of prodrug and drug permeated across the skin (a) PSH gel (b) PPH gel.

PL1.PL2.PL3.PL4.PLD.PLT.PL5.PLS. PLC. Therefore, by manipulating the composition of lipid system, the desired flux values can be obtained.

4.4. In-vivo evaluation The PPH and PL5 gels were selected for in-vivo studies and compared with control. The plasma drug level profile in rats following transdermal application of control, PPH or PL5 formulation is shown in Fig.

233

9. The plasma levels of propranolol after application of control were low (Cmax 523 ng / ml) while the plasma levels after application of the PPH gel gradually increased and reached a peak plasma drug level of 94 ng / ml at 8 h post-administration and maintained for longer times. The area under the curve (AUC 0 – 32 h ) was found to be 1968 ng h / ml and was much larger than 239 ng h / ml being AUC 0 – 18 h of control. The plasma drug level from gel must have been contributed by absorbed drug and from hydrolysis of absorbed prodrug vis-a-vis the sustained transdermal delivery must have lead to much higher and sustained plasma drug concentration. The EL incorporated PL5 gel also showed higher drug levels than control, however they were low as compared to PPH gel. The peak plasma level achieved was 75 ng / ml in 12 h. The AUC 0 – 32 h was 1708 ng h / ml. The plasma drug level at 32 h was however higher in case of PL5 gel than PPH gel. Therefore, this formulation is providing comparatively slow and sustained absorption. Thus the enhanced permeation from gel formulation is clearly demonstrated in-vivo. There have been reports of concentration dependent skin irritation reaction for b-blockers. Kobayashi et al. [25] correlated the amount of propranolol absorbed from patch with skin irritation. The skin irritation scores as defined by the code of federal regulations [26] following application of different formulations at 32 h is compiled in Table 3. The edema was almost absent in all cases. The mild to well-defined erythema was however observed in case of control showing mean score of 1.33. The gel formulation caused comparatively less irritation showing mean scores of 0.66 and 0.33 for PPH and PL5 gel, respectively. This decrease in irritation may be due to change of irritation properties after prodrug formation. National institute for occupational safety and health [27] interprets the score of 0–0.9 as non-irritant and safe for intact human skin contact; while score of 1–1.9 is interpreted as mild irritant requiring protective measure. Thus we considered the pharmacogel as safer and causing less irritation than control, although this conclusion should be confirmed by a larger sample size. Therefore, the gel formulations delivering much higher concentration of drug in blood with reduced skin irritation attests the utility of the system.

234

A. Namdeo, N.K. Jain / Journal of Controlled Release 82 (2002) 223 – 236

Fig. 7. Effect of egg lecithin on permeation across the rat skin. (Results are mean of three observations; error bars are not included to avoid overlapping).

5. Conclusion In summary, the study explored the supramolecu-

lar organization of prodrug and recognition of different phases led us to exploit the gel as a novel formulation for transdermal application. The

Fig. 8. Effect of adjuvants on permeation across the rat skin. (Results are mean of three observations; error bars have are not included to avoid overlapping).

A. Namdeo, N.K. Jain / Journal of Controlled Release 82 (2002) 223 – 236

235

India, for providing financial assistance in the form of a Senior Research Fellowship.

References

Fig. 9. Plasma drug level profile following trandsermal application of the formulation / control (n53).

physicochemical properties of prodrug and self-assembly enabled higher transdermal permeation, which could be suitably modified by inclusion of adjuvants. The blood level profiles following transdermal application similarly showed greater permeation. This approach can be extended to the preparation of different prodrugs and using different spacers to modulate the in-vivo bioconversion. Therefore, a new array of application can be envisaged in the successful transdermal delivery of drugs using this innovative formulation and prodrug approach.

Acknowledgements One of the authors, A.N. is thankful to the Council of Scientific and Industrial Research, New Delhi,

[1] D.J. Kazierad, K.D. Schlanz, M.B. Bottorf, Beta blockers, in: W.E. Evans, J.J. Schlentag, W.J. Jusko (Eds.), Applied Pharmacokinetics, Principle of Therapeutic Drug Monitoring, 3rd Edition, Applied Therapeutics, Vancouver, 1992, pp. 24–41. [2] J.G. Riddel, D.W.G. Harron, R.G. Shanks, Clinical pharmacokinetics of b-adrenoceptor antagonists, Clin. Pharmacokin. 12 (1987) 305–320. [3] A. Parkinson, Biotransformation of xenobiotics, in: C.P. Klaassen (Ed.), Casarett and Doulls Toxicology, The Basic Science of Poisons, International Edition, 1st Edition, McGraw-Hill, New York, 1996, pp. 152–153. [4] D.G. Shand, F.M. Nuckolls, J.A. Oates, Plasma propranolol levels in adults with observations in four children, Clin. Pharmacol. Ther. 11 (1970) 112–120. [5] T. Ogiso, M. Shintani, Mechanism for effect of fatty acids on the percutaneous absorption of propranolol, J. Pharm. Sci. 79 (12) (1990) 1065–1071. [6] S. Ahmed, T. Imai, M. Otagiri, Stereoselective hydrolysis and penetration of propranolol prodrugs: in-vitro evaluation using hairless mouse skin, J. Pharm. Sci. 84 (7) (1995) 877–885. [7] S. Chesnoy, D. Durand, J. Doucet, G. Gouarraze, Structural parameters involved in the permeation of propranolol HCl by iontophoresis and enhancers, J. Control. Rel. 58 (1999) 163–175. [8] J.R. Kunita, V.R. Gaskonda, H.O. Brotherton, M.A. Khan, I.K. Reddy, Effect of menthol and related terpenes on the percutaneous absorption of propranolol across excised hairless mouse skin, J. Pharm. Sci. 86 (12) (1997) 1369–1373. [9] H. Schreier, J. Bouwstra, Liposomes and niosomes as topical

Table 3 Skin irritation scores following application of different formulations Rabbit No.

Control

PPH gel

Mixed gel (PL5)

Erythema

Edema

Erythema

Edema

Erythema

Edema

1 2 3 4 5 6

2 0 1 2 2 1

0 0 0 0 1 0

0 1 1 1 0 1

0 0 1 0 0 0

1 0 0 2 0 0

0 0 0 0 0 0

Average

1.33

0.13

0.66

0.16

0.33

0

Scores are as defined by code of federal regulation: 0—no erythema, 1—very slight erythema, 2—well defined erythema, 3—moderate to severe erythema. Similarly defined is edema.

236

[10]

[11]

[12]

[13]

[14]

[15]

[16]

[17]

[18]

A. Namdeo, N.K. Jain / Journal of Controlled Release 82 (2002) 223 – 236 drug carriers: dermal and transdermal drug delivery, J. Control. Rel. 30 (1994) 1–5. R.H. Guy, J. Hadgraft, Percutaneous penetration enhancement: physicochemical considerations and implications for prodrug design, in: K.B. Sloan (Ed.), Prodrugs: Topical and Ocular Drug Delivery, Marcel Dekker, New York, 1992, pp. 1–16. A. Namdeo, N.K. Jain, Spontaneous vesiculation from dry granular propharmacosome bearing propranolol hydrochloride, in: 4th National Conference of Indian Society of Chemists and Biologists on Drug Research Scenario of 21st century, Lucknow, 2000. W.J. Irwin, K.A. Belaid, Drug-delivery by ion-exchange hydrolysis and rearrangement of ester prodrug of propranolol, Int. J. Pharm. 46 (1988) 57. O.H. Drummer, J. McNeil, F. Pritchard, J. Louis, Combined high-performance liquid chromatographic procedure for measuring 4-hydroxypropranolol and propranolol in plasma: pharmacokinetic measurements following conventional and slow release propranolol administration, J. Pharm. Sci. 70 (9) (1981) 1030. P. Walde, P.L. Luisi, A. Gazzaniga, F. Stroppolo, Lecithin organogel as matrix for transderaml transport of drugs, J. Pharm. Sci. 81 (9) (1992) 871–874. S. Wahlgren, A.L. Lundstorm, S.E. Friberg, Liquid crystals as potential ointment vehicle, J. Pharm. Sci. 73 (10) (1984) 1484–1486. S. Perret, M. Golding, W.P. Williams, A simple method for the preparation of liposomes for pharmaceutical applications: characterization of the liposomes, J. Pharm. Pharmacol. 43 (1991) 154–161. F.B. Rosevear, The microscopy of the liquid crystalline neat and middle phases of soaps and synthetic detergents, J. Am. Chem. Soc. 31 (1954) 628–639. B. Vora, A.J. Khopade, N.K. Jain, Proniosome based trans-

[19]

[20]

[21]

[22]

[23]

[24]

[25]

[26] [27]

dermal delivery of levonorgestrel for effective contraception, J. Control. Rel. 54 (1998) 149–165. H. Durrheim, G.L. Flynn, W.I. Higuchi, C.R. Behl, Permeation of hairless mouse skin I: experimental methods and comparison with human epidermal permeation by alkaloids, J. Pharm. Sci. 69 (1980) 781–786. R.B. Stroughton, in: H.I. Maibach (Ed.), Animal Models in Dermatology, Churchill Livingston, New York, 1975, pp. 121–132. C.D. Yu, N.A. Gordin, L. Fox, W.L. Higuchi, N.F.H. Ho, Physical model evaluation of topical prodrug delivery– simultaneous transport and bioconversion of vidarabine-59valerate V: mechanistic analysis of influence of nonhomogeneous enzyme distribution in hairless mouse skin, J. Pharm. Sci. 69 (1980) 775–780. L.G. Mueller, Novel anti-inflammatory esters, pharmaceutical composition and method for reducing inflammation, UK Pat., GB 2 204 869 A, 23 November 1988. F. Palagiano, L. Arenare, F. Barbato, M.I.L. Rotonda, F. Quaglia, F.P. Bonina, L. Montenegro, P.D. Caprariis, In-vitro and in-vivo evaluation of terpenoid esters of indomethacin as dermal prodrugs, Int. J. Pharm. 149 (1997) 171–182. B.C. Finnin, T.M. Morgan, Transdermal penetration enhancers: applications, limitations and potential, J. Pharm. Sci. 88 (10) (1999) 955–958. I. Kobayashi, K. Hosaka, T. Ueno, H. Maruo, M. Kamiyama, C. Konno, M. Gemba, Relationship between the amount of propranolol permeating through the stratum corneum of guinea pig skin and after application of propranolol adhesive patches and skin irritation, Biol. Pharm. Bull. 19 (6) (1996) 839–844. Code of Federal Regulations, Title 16, Part 173.210, 1976. K.I. Campbell, E.L. George, L.L. Hall, J.F. Stara, Dermal irritancy of metal compounds, Arch. Environ. Health 30 (1975) 168–170.