Drug permeation across the skin: effect of penetrant hydrophilicity

Transdermal absorption of bupranolol in rabbit skin in vitro and in vivo

This study was designed to clarify the percutaneous penetration of bupranolol (BP), a beta-adrenoceptor antagonist, through rabbit skin and to compare the in vitro penetration with the in vivo absorption. BP penetrated across the skin slowly in the absence of enhancers in vitro. Isopropyl myristate and N-methyl-2-pyrrolidone enhanced the in vitro penetration, with a 3.6 times higher flux compared with that without enhancers. However, in the in vivo percutaneous absorption, the maximal penetration was obtained with the formulation added dlimonene, with a 3.0 times higher area under the concentration-time curve (AUC) than that for the formulation without enhancers. The plasma levels of BP determined, however, were extremely lower than the theoretical plasma steady-state concentrations predicted. The plasma levels of BP after application of these formulations were maintained in the range of 7-22 ng/ml for 30 h, of which concentrations were above the therapeutically effective concentration (1.5-4 ng/ml). Therefore, the transdermal systems will offer an efficient drug delivery system for the treatment of angina pectoris and tachycardia.

Physiologically relevant two-compartment pharmacokinetic models for skin

Pharmacokinetic (compartment) models for skin have been used to predict or analyze absorption of chemical into and through skin. For highly lipophilic chemicals, the stratum corneum (sc) and the viable epidermis (v.e.) both contribute a significant resistance to chemical penetration and thus, both should be included in the model. This paper describes two-compartment models that represent the sc and the ve separately by extending the procedures previously developed for one-compartment models. The two-compartment models described here were developed by matching characteristics of a two-membrane model of skin. These compartment models were compared with membrane representations of the s.c. and v.e. for several different dermal exposure scenarios. When valid, which it is for many chemical exposure scenarios, the two-compartment model developed using characteristic times of the membrane model (model B2) more closely represents the two-membrane model than the model developed with equilibrium conditions of the membrane model (model B1). When model B2 is invalid, then model B1 is recommended. Criteria are provided for choosing from the various one- or two-compartment model options.

The effect of terpene enhancer lipophilicity on the percutaneous permeation of hydrocortisone formulated in HPMC gel systems

The percutaneous permeation of hydrocortisone (HC) was investigated in hairless mouse skin after application of an alcoholic hydrogel using a diffusion cell technique. The formulations contained one of 12 terpenes, the selection of which was based on an increase in their lipophilicity (log P 1.06-5.36). Flux, cumulative receptor concentrations, skin content, and lag time of HC were measured over 24 h and compared with control gels (containing no terpene). Furthermore, HC skin content and the solubility of HC in the alcoholic hydrogel solvent mixture in the presence of terpene were determined, and correlated to the enhancing activity of terpenes. The in vitro permeation experiments with hairless mouse skin revealed that the terpene enhancers varied in their ability to enhance the flux of HC. Nerolidol which possessed the highest lipophilicity (log P = 5.36+/-0.38) provided the greatest enhancement for HC flux (35.3-fold over control). Fenchone (log P = 2.13+/-0.30) exhibited the lowest enhancement of HC flux (10.1-fold over control). In addition, a linear relationship was established between the log P of terpenes and the cumulative amount of HC in the receptor after 24 h (Q(24)). Nerolidol, provided the highest Q(24) (1733+/-93 microg/cm(2)), whereas verbenone produced the lowest Q(24) (653+/-105 microg/cm(2)). Thymol provided the lowest HC skin content (1151+/-293 microg/g), while cineole produced the highest HC skin content (18999+/-5666 microg/g). No correlation was established between the log P of enhancers and HC skin content. A correlation however, existed between the log P of terpenes and the lag time. As log P increased, a linear decrease in lag time was observed. Cymene yielded the shortest HC lag time, while fenchone produced the longest lag time. Also, the increase in the log P of terpenes resulted in a proportional increase in HC solubility in the formulation solvent mixture.

Permeation of several drugs through keratinized epithelial-free membrane of hamster cheek pouch

The hamster cheek pouch mucosa was selected as a substitute for the human buccal mucosa in an in vitro permeation study. Considering that a keratinized layer is not present in the human buccal mucosa, keratinized epithelial-free hamster cheek pouch (KEF-membrane) was prepared by chemical splitting. To confirm the usefulness of the KEF-membrane, a permeation study was conducted using several drugs with different lipophilicities. The permeability coefficient of hydrophilic drugs through the KEF-membrane (Pkef) was significantly greater than that through a viable KEF-membrane (Pkef-viable), which was estimated by using the permeability coefficient of the viable full-thickness membrane and that of the keratinized layer. On the other hand, the Pkef values of lipophilic drugs were comparable with the Pkef-viable values. Furthermore, the ratio of these P values (Pkef/Pkef-viable) decreased with increasing lipophilicity of drugs. These findings indicate that the KEF-membrane may be useful for buccal permeation studies of lipophilic drugs.

Effect of cod-liver oil extract on the buccal permeation of ergotamine tartrate

Ergotamine tartrate (ET) is used clinically in the treatment of migraines. However, the bioavailability of ET is rather poor following oral administration. Therefore, we tried to improve ET delivery using buccal administration. The purpose of this study was to investigate the characteristics of the permeation of ET through the hamster cheek pouch in vitro using a two-chamber diffusion cell, and to evaluate the effect of permeation enhancers on the transbuccal delivery of ET. Cod-liver oil extract (CLOE), polyoxyethylene hydrogenated castor oil (HCO 60), sodium glycocholate (GC), and sodium caprate (CA) were selected as premeation enhancers considering their low irritancy of the mucosa. When the enhancers were added to the donor cell at a 5% concentration each, the ET permeation rate markedly increased compared with that in a control not containing enhancer. Among these enhancers, CLOE exhibited the greatest effect. Because CLOE is composed of 16 kinds of fatty acids, the enhancement action of each of the major components was separately determined. As major fatty acids, palmitic acid, oleic acid, eicosapentaenoic acid (EPA), and docosahexaenoic acid (DHA) were selected and their enhancing effects were studied. The enhancing effect of each fatty acid was significantly lower than that of CLOE.

On mathematical modeling of dermal and transdermal drug delivery

This paper deals with two extensions of diffusion models for the drug delivery process into human skin in order to give a more realistic approach. As one extension several penetrating substances formulated within a vehicle are considered for modeling the case of an applied drug and some penetration modifiers (enhancers and reducers, respectively). A coupling via concentration-dependent diffusivities between the diffusion equations of the involved substances is used to model the dependencies between them. Furthermore, a moving boundary problem for the diffusion equation of the drug delivery process is developed to describe the time-dependent maximum penetration depth of each penetrant marked by a movingboundary. On this basis a model is developed that can predict both the concentration profile and the position of the penetration boundary depending on time. Both concepts are described on a two-dimensional multilayered domain representing a cross section through human skin. The model equations are solved by exploiting a suitable numerical discretization method.

Mechanism for enhancement effect of lipid disperse system on percutaneous absorption

To clarify the mechanism involved in the enhancement effect of lipid disperse systems (LDS) on percutaneous absorption, the effect of the LDSs of betahistine (BH), prepared using egg phosphatidylcholine (EPC, phase transition temperature, tau m, -15 to -17 degrees C) or hydrogenated soybean phosphatidylcholine (HSPC, tau m, 50 to 60 degrees C), cholesterol, and dicetylphosphate, on the percutaneous absorption of BH, the amount of skin lipids (ceramides, triglycerides, and phospholipids), the fluidity of skin lipids, and the partitioning of LDS-BH into the skin layers were investigated using Wistar and hairless rats. Also examined was whether the LDS penetrated through the stratum corneum (SC) or follicles, using a fluorescent probe (Nile Red). The plasma concentrations of BH were much higher and more sustained after application of a gel formulation containing EPC-LDS and D-limonene (prep. 2) than those after the non-LDS formulation containing D-limonene (prep. 1), whereas the plasma levels after application of a formulation containing HSPC-LDS (prep. 5) were not largely increased compared with those after prep. 1. The content of ceramides (intercellular lipids) and triglycerides (sebaceous gland lipides) in the SC were dramatically decreased by the treatment with prep. 1 and prep. 2, with the more decreased levels of these lipids by the treatment with prep. 2. The phospholipid content of the SC was enhanced by 2-fold following the prep. 2 treatment, indicating the extensive incorporation of LDS lipids into the SC. The histochemical examination of the skin, following application of EPC-LDS with a fluorescent probe, indicated that the LDS lipids penetrated rapidly through the SC and follicles into the viable skins. The fluidity of the SC lipids was dramatically increased following the treatment with the fluid EPC-LDS, whereas the fluidity was significantly decreased by the solid HSPC-LDS. The BH in each skin layer was also significantly increased by the treatment with prep. 2. These results surely demonstrated that the fluid LDS permeated rapidly into the SC and the viable epidermis through the intercellular domains and the follicles in intact vesicles or lipid mixtures, thus ensuring the facilitated transport of LDS-drug through the skin.

Anti-inflammatory effect of flurbiprofen tape applied percutaneously to rats with adjuvant-induced arthritis

The anti-inflammatory effect of flurbiprofen tape (FP-T) by topical application was investigated, and the findings were compared with the results of oral administration of flurbiprofen to adjuvant arthritic rats. The topical application of FP-T significantly suppressed both applied and non-applied hind paw edema, with a potency similar to that seen with the oral administration of flurbiprofen. Body weight also increased with these treatments. Plasma levels of flurbiprofen differed little between topical application of FP-T and oral administration of flurbiprofen. Gastric damage induced by topical application of FP-T was significantly less than that seen in case of oral administration of flurbiprofen. These results suggest that the anti-inflammatory effects of FP-T cannot be entirely explained by flurbiprofen permeating inflamed tissue below the application site; rather, flurbiprofen penetrating into the systemic circulation may explain these actions.

Effect of various enhancers on transdermal penetration of indomethacin and urea, and relationship between penetration parameters and enhancement factors

The enhancing capacity of various chemicals, which are widely recognized as enhancers, for the transdermal penetration into full-thickness rat skin of a model lipophilic drug [indomethacin (IND)] and a hydrophilic permeant (urea) was estimated by an in vitro technique. In addition, the fluidity of the stratum corneum lipids, the partitioning of IND into skin, the lipid (ceramides) extraction from the stratum corneum by enhancers, and the IND solubility in enhancer vehicle were measured and related to the enhancing capacity. In vitro permeation experiments with hairless rat skin unequivocally revealed that the enhancers varied in abilities to enhance the fluxes of both agents. Laurocapram, isopropylmyristate (IPM), sodium oleate, and cineol increased fluxes of both agents to a great extent, but N-methyl-2-pyrrolidone (NMP), N,N-diethyl-m-tolamide (DEET), and oleyl oleate were less effective acclerants. Many enhancers increased the fluidity of the lipids [with a threshold of approximately 0.6-0.8 ns at 37 degrees C in the rotational correlation time (tau c)], the skin partitioning of IND, the extraction of ceramides from the cornified cells, and the thermodynamic activity of IND in vehicle (calculated from the solubility) to varying extents. A good correlation was observed between the increase in the fluidity of stratum corneum lipids and the partitioning of IND into skin, between the increase in the fluidity and the flux or the decrease in lag time for IND, between the removal of ceramides and the skin partitioning of IND, and between the removal of ceramides and the flux of urea (p < 0.05 in all cases).(ABSTRACT TRUNCATED AT 250 WORDS)

Diffusion and metabolism of prednisolone farnesylate in viable skin of the hairless mouse

The diffusion and metabolism of prednisolone 21-farnesylate were investigated in viable skin of the hairless mouse in vitro. The prodrug ester was extensively metabolized in viable skin, while it was stable in the donor and receptor solutions. The rate of appearance of the prodrug and its metabolite prednisolone was markedly influenced by the direction of the skin placed between the in vitro diffusion half-cells. The rate of bioconversion of the prodrug was determined as a function of the distance from the surface of the skin. The prodrug was increasingly metabolized with the distance from the surface of the skin, indicating that the responsible enzymes are enriched in the lower layers of the viable skin. A model with linearly increasing enzyme activity in the viable skin accounts for the in vitro profiles of the diffusion/metabolism of the prodrug in the viable skin of hairless mouse.

In vivo and in vitro analysis of skin penetration enhancement based on a two-layer diffusion model with polar and nonpolar routes in the stratum corneum

In vitro and in vivo skin penetration of three drugs with different lipophilicities and the enhancing effects of 1-geranylazacycloheptan-2-one (GACH) were studied in rats. In vivo drug absorption profiles obtained by deconvolution of urinary excretion profiles were compared to the corresponding in vitro data obtained with a diffusion experiment. In vivo skin penetration of lipophilic butylparaben was considerably greater than that observed in vitro, while hydrophilic mannitol and acyclovir showed low penetration in both systems without GACH pretreatment. On the other hand, GACH enhanced mannitol and acyclovir penetration, especially in the in vivo system. Analysis of absorption profiles, using a two-layer skin model with polar and nonpolar routes in the stratum corneum, suggested that the diffusion length of a viable layer (viable epidermis and dermis) was shorter in vivo than in vitro and the effective area of the polar route in the stratum corneum was larger in vitro without GACH pretreatment. GACH increased the partitioning of acyclovir into the nonpolar route to the same extent in both systems. In addition, GACH increased the effective area of the polar route in vivo, probably because of enhanced water permeability; however, this effect was smaller in vitro since the stratum corneum was already hydrated even without GACH pretreatment.

Comparison of the partition coefficient and skin penetration of a marine algal toxin (lyngbyatoxin A)

Lyngbyatoxin A is produced by marine algae, and causes local cutaneous toxicity in swimmers. The purpose of this research was (1) to determine the partition coefficient of lyngbyatoxin A in octanol/water and (2) to use methods in vitro to measure the penetration and distribution of lyngbyatoxin A in guinea pig and human skin. Discs of excised guinea pig and human skin were mounted in diffusion chambers that exposed the epidermal surface to air and bathed the dermis with HEPES-buffered Hanks' balanced salt solution with gentamicin sulphate. The epidermal surfaces were dosed with 26 micrograms lyngbyatoxin A/cm2 dissolved in 13 microliters dimethyl sulphoxide/cm2. The diffusion chambers were incubated at 36 degrees C for varying periods (1.0-24 hr). HPLC was used to quantify lyngbyatoxin A. Skin penetration was calculated by summing the amount of lyngbyatoxin A recovered from the dermis and receptor fluid. The mean partition coefficient for lyngbyatoxin A was 1.53. Penetration of lyngbyatoxin A (expressed as a percentage of dose, n = 3) in guinea pig and human skin was 23 and 6.2 (respectively) after 1 hr of topical exposure. The amount of lyngbyatoxin A in the dermis and receptor fluid did not change significantly over time.

A numerical approach to study the effect of binding on the iontophoretic transport of a series of amino acids

The objective of this investigation was to quantitate the effect of binding on the iontophoretic transport of a series of amino acids. The diffusivity and concentration in the stratum corneum and viable skin were estimated from the passive permeation profiles through whole and stripped skin. Using these parameters and the Langmuir binding parameters from equilibrium binding studies, the passive permeation and desorption profiles were simulated using a bilayer skin permeation model. The parameters were adjusted until convergence with experimental data was achieved. Then the iontophoretic profiles were simulated, using the parameters from the simulated passive permeation studies, an estimate of convective flow from the flux of tritiated water, and measurement of the potential across the skin. The overall effect of binding on the iontophoretic profiles was found to dampen the effect of the iontophoresis treatment; the profiles appear flatter, and the transition to passive diffusion less distinct compared to profiles which do not include the binding parameters. It also appear that the degree of skin hydration has a substantial effect on the shape of the iontophoretic profile, such that approximately 50% of the enhancement in the iontophoretic profile of freshly excised skin continues after treatment.

Analysis of drug penetration through the skin by the two-layer skin model

A diffusion model for the skin penetration of drug in the finite-dose system was developed considering the skin to be composed of two layers, the outermost layer (stratum corneum) and the lower layer (viable epidermis and dermis). Based on this skin model, the Laplace transforms of the equations for the drug amounts in the receptor, the vehicle, and the skin were derived. The penetration profiles of 6-mercaptopurine (6-MP) through the intact and stripped guinea pig skin were obtained from in vitro diffusion experiments. The computer fitting of those profiles to the Laplace-transformed equations by a nonlinear least-squares program based on a fast inverse Laplace transform algorithm [MULTI-(FILT)] gave parameters such as diffusion coefficients of 6-MP and thicknesses of both layers. The mean transit time (MTT) for each diffusion process was defined based on statistical moment concept and calculated using the obtained parameters. Under the present condition, the process to move from the vehicle to the stratum corneum is demonstrated to have the longest mean time in overall processes of 6-MP penetration.

A method for predicting steady-state rate of skin penetration in vivo

A simple in vivo method was proposed for predicting the steady-state rate of penetration of drugs across the stratum corneum. Both the diffusion coefficient and the partition coefficient in the stratum corneum can be determined by the amounts of drug in the stratum corneum at two time intervals under transient conditions after transdermal drug application. The amount of drug entering the stratum corneum is determined by 20 strippings with an adhesive tape. The steady-state rate of penetration was then calculated for the thickness of the stratum corneum and the concentration of the donor solution. The steady-state rates of penetration of ascorbic acid and estradiol across hairless mouse skin were evaluated from this in vivo approach and compared with those obtained from in vitro penetration experiment using excised hairless mouse skin. The data confirmed that the proposed in vivo method can predict the steady-state rate of penetration of these drugs across the stratum corneum in normal skin.