A comparative ex vivo drug permeation study of beta-blockers through porcine buccal mucosa

A comparative ex vivo drug permeation study of beta-blockers through porcine buccal mucosa

International Journal of Pharmaceutics 468 (2014) 50–54 Contents lists available at ScienceDirect International Journal of Pharmaceutics journal hom...

415KB Sizes 2 Downloads 19 Views

International Journal of Pharmaceutics 468 (2014) 50–54

Contents lists available at ScienceDirect

International Journal of Pharmaceutics journal homepage: www.elsevier.com/locate/ijpharm

A comparative ex vivo drug permeation study of beta-blockers through porcine buccal mucosa Sonia Amores a , Jacinto Lauroba a, *, Ana Calpena a , Helena Colom a , Alvaro Gimeno b , José Domenech a a Pharmacy Pharmaceutical Technology Department, Biopharmaceutical Pharmacokinetics Unit, Faculty of Pharmacy, Av. Joan XXIII, s/n, Barcelona 08028, Spain b Animal Facility at Bellvitge-University of Barcelona, Spain



Article history: Received 22 October 2013 Received in revised form 26 March 2014 Accepted 26 March 2014 Available online 13 April 2014

Apparent permeability coefficients (kp) of a series of beta-blockers: acebutolol, atenolol, labetalol, metoprolol, oxprenolol and propranolol, through porcine buccal mucosa were determined. The aim of the study was to determine the permeation parameters (apparent permeability coefficient, kp; flux, J; and lag time, TL) as a measure of the intrinsic permeability of porcine buccal mucosa to these drugs, in order to predict the efficacy of their possible administration through human buccal mucosa. A positive linear correlation was observed between the apparent permeability coefficient, kpand the partition coefficient, P. Oxprenolol and propranolol are the drugs that presented the highest values of kp: 0.3231 102 cm/h and 0.5666  102 cm/h, respectively. Multiple linear regression (MLR) using least square estimation was performed on the data set with log kp as dependent variable and the descriptors as predictor

Keywords: Buccal permeation Beta-blockers Porcine buccal mucosa Ex vivo drug permeation

variables. The potential systemic capacity after a buccal administration was predicted by estimating the plasma concentrations at steady-stated (Css). Considering the entire process of permeation ex vivo, propranolol and oxprenolol would seem to be the best candidates for administration through the buccal mucosa. ã 2014 Published by Elsevier B.V.

1. Introduction In recent years, most biopharmaceutical and pharmacokinetic research have focused either on the use of new routes for drug administration or on new drug delivery systems, with the aim of obtaining improved therapeutic activity, fewer adverse effects or better patient compliance (Harris and Robinson, 1992; Squier and Wertz, 1996; Shojaei, 1998). Buccal drug delivery is a logical alternative delivery route for drugs which undergo extensive degradation in the stomach and the liver. Drugs delivered by the buccal route gain direct entry into systemic circulation (Rathbone et al., 1994; Şenel and Hincal, 2001; Hao and Heng, 2003). The permeability of buccal mucosa is between 4 to 4000 times greater than that of skin. As a result, faster onset of action for several drugs has been reported (Galey et al., 1976). Since human buccal mucosa is not widely available, animal oral mucosa is routinely used for studies in vitro. In recent studies, porcine buccal mucosa has been chosen as an animal model due to its close resemblance to human

* Corresponding author. Tel.: +34 93 402 45 60; fax: +34 93 402 45 63. E-mail address: [email protected] (J. Lauroba). http://dx.doi.org/10.1016/j.ijpharm.2014.03.050 0378-5173/ ã 2014 Published by Elsevier B.V.

buccal mucosa in both ultra structure and enzyme activity (de Vries et al., 1990; Wertz and Squier, 1991; Xiang et al., 2002; Diaz del Consuelo et al., 2005). The permeability characteristics of porcine buccal mucosa are also similar to those of human buccal mucosa (Lesch et al., 1989). The objective of this study was to compare the intrinsic buccal permeability characteristics of a series of beta-blockers (acebutolol, atenolol, labetalol, metoprolol, oxprenolol and propranolol) through porcine buccal mucosa. Since the same experimental conditions were maintained throughout the study, we are able to compare the possible buccal permeation of each individual drug assayed. We also attempt to establish correlations between buccal permeability and the various structural and the physicochemical descriptors of the drugs studied. 2. Materials and methods 2.1. Materials Acebutolol chlorhydrate, atenolol base, labetalol chlorhydrate and metoprolol tartrate were supplied by Sigma–Aldrich (Madrid, Spain). Oxprenolol chlorhydrate and propranolol chlorhydrate

S. Amores et al. / International Journal of Pharmaceutics 468 (2014) 50–54

were provided by Novartis and Acofarma, respectively (Barcelona, Spain). Hank’s balanced salt solution (HBSS) (Composition in g/L: CaCl2 = 0.14; KCl = 0.14; KH2PO4 = 0.06; MgSO4 = 0.1; MgCl2 = 0.1; NaCl = 8.0; NaHCO3 = 0.35; Na2HPO4 = 0.09; glucose = 1) was obtained from Biological Industries (Barcelona, Spain). Phosphate buffered saline (PBS) was obtained from Dulbecco’s, Life technologies (Barcelona, Spain). Acetonitrile (ACN), acetic acid, disodium hydrogen phosphate (anhydrous) and potassium phosphate (monobasic) were purchased from Panreac (Barcelona, Spain). All the chemicals were of analytical grade and used without further purification. 2.2. Analytical method A Waters HPLC (model LC Module Plus) was used to analyse the drugs. The analyses were performed at room temperature with a C18 Atlantis reverse phase column (5 mm particle size, 4.6  150 mm) purchased from Waters (Barcelona, Spain). UV detection of the drug at a wavelength of 280 nm was used. The specific conditions of the mobile phase and the flow rate for each drug assayed are shown in Table 1. Various concentrations of each drug in the buffer solution (pH 7.4) were used to construct the calibration curves. The concentration ranges were between 0.05 and 30 mg/mL (atenolol, labetalol and propranolol) and between 0.2 and 30 mg/mL (acebutolol, metoprolol and oxprenolol). The correlation coefficient that relates the theoretical and the real concentrations of the beta-blockers was, r = 0.999 in all cases. Validation of the analytical methods (n = 6) indicated that they were exact and precise. Accuracy, expressed as a relative error, ranged from 11.9% to 10.3%. Precision, expressed as a relative standard deviation, ranged from 0.5% to 14.3%. The limit of detection was between 0.102 mg/mL for atenolol and 0.532 mg/mL for acebutolol. The limit of quantification was estimated to be between 0.308 mg/mL for atenolol and 1.611 mg/mL for acebutolol. These results allow us to quantify the amount of drug in each of the samples taken at the pre-established times. 2.3. Solubility determination Drug solubility (C0) was determined under the same conditions as the permeation studies (pH 6.8). An excess of the drug was added to the buffer, the mixture was incubated in a shaking water bath maintained at 37  1  C for 24 h. After centrifugation at 4000 rpm, the supernatant was filtered (nylon, 0.45 mm). When the solution was appropriately diluted, the concentration of each drug was determined by HPLC. The solubility was measured in triplicate. 2.4. Determination of n-octanol-buffer solution (pH 6.8) partition coefficient (P) The partition coefficient (P) is a characteristic physicochemical constant of drugs that indicates how lipophilic they are and it is Table 1 Chromatographic conditions used in each case. Drug

Acebutolol Atenolol Labetalol Metoprolol Oxprenolol Propranolol a

Mobile phase Buffera




10 10 10 10 10 75

25 5 25 25 25 25

65 85 65 65 65 –

2 2 2 2 2 1

Ammonium acetate pH 3.


very closely related to their capacity to penetrate lipid membranes, such as the stratum corneum of the skin and the buccal mucosa. It is defined as the relation between the concentration of nonionized drug in an organic solvent and the concentration of the non-ionized active ingredient in an aqueous solvent. When we consider drugs that can be ionized at physiological pH, as is the case with beta-blockers, it is better to estimate the distribution coefficient (D) (Hadgraft and Valenta, 2000; Scott and Clymer, 2002) which is a parameter that is better suited to calculating the partition coefficient of the non-ionized compound. In the present work the logarithm of the distribution coefficient (log DpH 6.8) was obtained from the literature (ChemIDplus, 2007). By applying the corresponding equation, we arrived at the logarithm of the partition coefficient (log P); and then, taking the antilogarithm of log P gave us the corresponding partition coefficient (P). 2.5. Buccal mucosa permeation procedure The porcine buccal mucosa, obtained immediately after the pigs (3–4-month-old females) had been slaughtered, was obtained from the Animal Facility at the Bellvitge Campus, University of Barcelona. The animals were slaughtered using an overdose of sodium thiopental anaesthesia. The buccal tissues were transferred from the hospital to the laboratory in containers filled with Hank’s liquid. The study was approved by the Animal Experimentation Ethics Committee of the University of Barcelona and the Animal Experimentation Committee of the regional authorities (Generalitat of Catalonia). For the permeation studies, the porcine buccal mucosa was cut to a thickness of 500  50 mm (Dermatom Aesculap, Tuttlingen, Germany). This thickness corresponds to the buccal epithelial thickness, which contributes to the diffusion barrier (Sudhakar et al., 2006). The membranes were then mounted on specially designed membrane holders with a permeation orifice diameter of 1.2 cm (diffusion area: 1.1 cm2). Using the membrane holder, each porcine buccal membrane was mounted between the donor (1.5 mL) and the receptor (6 mL) compartments of six Franz-type diffusion cells (Franz, 1975). In the donor compartment we placed 250 mL of the solution of the drug to be studied. A pH of 6.8 was used in the donor as it represents a mean value of the physiological oral cavity pH (Le Brun et al., 1989). The receptor pH was fixed at 7.4 to simulate in vivo plasma pH. Prior to conducting the experiments, the diffusion cells were incubated for 1 h in a water bath to equalize the temperature in all the cells at 37  1  C by means of an insulating jacket. The receptor solution was PBS, pH 7.4, which was continuously stirred at 600 rpm with a Teflon-coated bar magnet placed inside the cell. Sink conditions were ensured in all experiments after initial testing of the drug saturation concentration in the receptor medium. Samples (300 mL) were withdrawn via syringe from the centre of the receptor compartment of the six cells at the following time intervals: 0.25, 0.5, 1, 2, 3, 4, 5 and 6 h. The sample volume was immediately replaced with the same volume of fresh receptor medium. The cumulative amounts of the drug (mg) that had penetrated per unit surface area of the mucosa membrane (cm2) were corrected for this sample removal and plotted versus time (h). The drug concentrations in the samples taken from the receptor compartment were assayed by HPLC. 2.6. Data analysis The permeation profiles were analysed on the basis of a diffusion model for an infinite dose system (Okamoto et al., 1986). The permeation parameters were calculated from experimental data in steady state (straight section of the curve). The P1 (related


S. Amores et al. / International Journal of Pharmaceutics 468 (2014) 50–54

with the partition coefficient) and the P2 (related with the diffusion parameter) parameters were estimated by fitting the Laplace program (Micromath Scientific Software, 1991) to the experimental data amounts of drug vs time. The apparent permeability coefficient (kp), the flux (J) and the lag time (TL) were calculated from the following equations: kP ¼ P1  P2


J ¼ C 0  kp


1 6  P2


TL ¼

where C0 is the initial concentration of the drug in the donor solution. When examining the fundamental permeation behaviour of a material, flux data are usually obtained by application of an infinite dose to the tissue surface. In general terms, when an infinite dose is applied to the tissue, it is assumed that there is no change in drug concentration (or more accurately, in its thermodynamic activity) during the experiment. Therefore, a saturated solution of the drug was used to obtain maximal thermodynamic activity. Under the conditions used, the permeate concentration in the donor phase does not fall by more than 10% from saturation during the experimental period (Williams, 2003). The potential systemic capacity after a buccal administration can be predicted by estimating the plasma concentrations in steady-stated (Css) from the following equation: C ss ¼



where S is the surface area of permeation (Franz cell, 1.1 cm2), J is the flux determined in this study and ClP is the plasma clearance taken from the literature (Frishman and Alwarshetty, 2002). The majority of the theoretical models described in the literature for prediction of drug permeability through mucosa include the permeability as response variable and physicochemical and the various structural descriptors as the predictor variables. Molecular weight (MW), molecular volume (MV), (MV is a better indicator for molecular size than MW) (Hou et al., 2004), octanol– water partition coefficient (log P), log DpH 6.8 (logarithm of distribution coefficient at pH 6.8, which corresponds to salivary pH). Topological polar surface area (TPSA), number of hydrogen bond acceptors (HBA) and donors (HBD) and number of rotatable bonds (nRotB). Solubility (C0) was determined experimentally using the shake-flask method. log DpH 6.8 were obtained from the literature (ChemIDplus, 2007). The remaining descriptors were calculated using an online cheminformatics service (Molinspiration, 2007).

Table 2 Physicochemical parameters corresponding to the beta-blockers assayed. Drug

C0 (mg/mL)


Log DpH

Acebutolol Atenolol Labetalol Metoprolol Oxprenolol Propranolol

71005 13921 1126 32131 10006 82357

9.1 9.54 9.2 9.56 9.6 9.45

1.56 1.30 0.03 0.56 0.81 1.20

a 6.8

Log P


1.478 0.22 2.326 1.95 3.082 3.48

30.06 1.65 211.84 89.12 1207.81 3019.95

Drug solubility (C0), ionization constant (pKa), logarithm distribution coefficient (log D pH 6.8), logarithm partition coefficient (log P) and partition coefficient (P). a Values from literature (ChemIDplus, 2007).

3. Results and discussion As commented above, the anatomical characteristics of porcine buccal mucosa, in terms of structure and composition, are similar to those of human buccal mucosa; this is due to the fact that it is not keratinized. For this reason, results from studies of how drugs permeate porcine buccal mucosa may be similar to those that would be obtained in humans. Furthermore, the surface area of the porcine buccal mucosa is relatively large, which means that interindividual variability can be minimized. To study the intrinsic capacity of beta-blockers to pass through the buccal mucosa, a transmucosal permeation study of beta-blockers, in a saturated and filtered solution at pH 6.8, was performed; using porcine buccal mucosa as the permeation membrane. Table 2 shows the concentrations of the saturated solutions (C0) for each of beta- blockers used in the permeation tests, and also shows the values of the following physicochemical parameters: pKa, for each drug; the logarithms of the distribution coefficient at pH 6.8 (log DpH 6.8), the logarithm of the partition coefficient (log P) for each drug assayed and the corresponding partition coefficient (P). Based on the values of the distribution coefficients and, through applying the corresponding equation, the value of log P was calculated for each drug taking into account the value of pKa. The beta-blockers permeate the mucosa via passive diffusion of the non-ionized fraction, independently of the initial drug concentration (Liu et al., 2011). As the value of pKa is similar for all the betablockers studied and the pH of the medium is constant (pH 6.8), the non-ionized fraction of the drug will always be the same under the working conditions used in this study. Time course of mean cumulative amounts permeated (mg) of each beta-blocker assayed is shown in Fig. 1. Significant differences (p = 0.000279) (Kruskal–Wallis test) were found in the amounts permeated through buccal mucosa for the beta-blockers assayed.

2.7. Statistical analysis In studies of this kind, the permeability do not display normal distribution (permeability constant frequently display log-normal distribution) (Williams et al., 1992). Therefore, it is advisable to perform statistical studies via non-parametric methods (The Kruskal–Wallis test). However, in the present study, to consider two factors of variability (drug and buccal mucosa) when the same number of replications were not available for every mucosa (due to the failure of mucosa integrity), a two-way ANOVA was performed, after logarithmic transformation of the data. Statistical analysis was performed using different programs (SAS, 2006; NCSS, 2006). A significance level of p < 0.05 was adopted in all cases. However, the results corresponding to the parameters of permeation are expressed as median and range of the values considered.

Fig. 1. Time course of mean cumulative amounts permeated of the drugs assayed.

S. Amores et al. / International Journal of Pharmaceutics 468 (2014) 50–54


Table 3 Values of the permeability parameters (P1 related with the partition coefficient and P2 related with the diffusion parameter), permeability coefficient (kP), flux (J) and lag time (TL) obtained for the series of drugs assayed. Values are presented as median and maximum and minimum values in parenthesis. Drug

P1 (cm)

P2 (h–1)

kp 102 (cm/h)

J (mg/h cm2)

TI (h)


0.0030 (0.00002–0.0071) 0.0020 (0.0008–0.0203) 0.2569 (0.0070–0.4670) 0.00005 (0.00002–0.00017) 0.0172 (0.0023–0.0246) 0.0545 (0.00622–0.1142)

0.0345 (0.0250–0.8503) 0.2724 (0.0474–2.2886) 0.1078 (0.0066–0.2369) 0.2056 (0.0586–0.4834) 0.1572 (0.0593–0.2867) 0.1032 (0.0743–0.8583)

0.0097 (0.0012–0.0179) 0.0473 (0.0043–2.0576) 0.1959 (0.1667–4.0730) 0.0009 (0.0005–0.0012) 0.3231 (0.0135–0.3891) 0.5666 (0.4601–0.8698)

6.6 (0.9–12.4) 6.6 (0.6–277.4) 19.2 (1.9–44.7) 299.4 (169.4–380.4) 323.3 (13.5–389.3) 466.6 (378.9–716.3)

3.482 (0.408–3.630) 1.014 (0.046–2.492) 2.070 (1.167–2.313) 0.997 (0.448–10762) 1.164 (0.446–2.194) 1.439 (0.186–1.912)

Atenolol Labetalol Metoprolol Oxprenolol Propranolol

Using fresh porcine mucosa, the permeation parameters corresponding to the beta- blockers assayed in the study were calculated. Table 3 shows the permeability parameters (P1 related with the partition coefficient and P2 related with the diffusion parameter), permeability coefficient (kp), flux (J) and lag time (TL) obtained via an infinite dose model for each beta-blocker in a saturated solution. The median values of the permeation parameters of the betablockers studied indicate that propranolol presents the greatest flux (466.6 mg/h cm2) and the greatest apparent permeability coefficient (0.5666 cm/h). This could be due to the fact that it is the beta-blocker with the largest C0 (82357 mg/mL), since flux is given by: J ¼ C 0  kp . Alternatively, it may be because the values of P1 and P2 are relatively high, 0.0545 cm and 0.1032 h1, respectively (kP ¼ P1  P2 ). The value of P2 for propranolol is practically double that of P1, which indicates that the diffusion parameter for propranolol through the mucosa is more important in permeation by this beta-blocker than its partition coefficient, which is related to P1. The beta-blockers with the lowest flux are acebutolol and atenolol (both with 6.6 mg/h cm2) while the former is also the one that presents the greatest lag time (3.482 h). In the case of atenolol, the value of P2 (0.2724 h1) is greater than that of acebutolol (0.0345 h1); and for both these drugs, the value of P1 is of the same order: 0.0030 cm and 0.0020 cm, respectively. This indicates that the mechanisms of permeation of these two drugs through the mucosa are different: in the case of atenolol the diffusion parameter (P2) has a greater influence; while for acebutolol, the influence of the diffusion parameter and of the partition coefficient (P) is similar. Although the value of kp for atenolol is greater than that for acebutolol, since the former has a larger value of P2 and the two have similar values of P1, the value of the flux of the two drugs is similar; due to the value of C0 for acebutolol being far higher (71,005 mg/mL) than that for atenolol (13,921 mg/mL). Labetalol is the least soluble (1126 mg/mL); however, it has a greater flux than acebutolol and atenolol, which are both more soluble: 71,005 mg/ mL and 13,921 mg/mL, respectively. This may be due to labetalol having greater diffusion through the membrane as a consequence of its partition coefficient (P) (211.84) being greater than those of acebutolol and atenolol (30.06 and 1.65, respectively) and, therefore labetalol also has a greater value of P1. In contrast, the value of P2 of labetalol (0.1078 h1) is of the same order as that of atenolol (0.2724 h1) and greater than that of acebutolol (0.0345 h1). The ultimate consequence is that the value of kp (kP ¼ P1  P2 ) for labetalol (0.1959  102 cm/h) is greater than that of acebutolol (0.0097  102 cm/h) and atenolol (0.0473  102 cm/h). Given that flux is given by: J ¼ C 0  kp , although the value of C0 for labetalol is less than that for acebutolol and atenolol, the value of the flux for labetalol is higher.

These results allow us to order the beta-blockers studied in terms of their flux of permeation through the mucosa as follows: propranolol > oxprenolol > metoprolol > labetalol > acebutolol > atenolol The values of the parameters that are representative of the process of permeation through fresh porcine mucosa of the betablockers studied, with the exception of P2, do not have normal distributions and neither have homogeneity of variance. Therefore, the experimental data were subjected to a non-parametric test: the Kruskal–Wallis test. The test demonstrates that there are statistically significant differences (p < 0.0001), except for P2 (p = 0.5823) and TL (p = 0.2959), in the amounts of drug that pass through fresh porcine mucosa. The fact that there are no statistically significant differences for the diffusion parameter, P2, indicates that, in all cases, the diffusion of the drug through the membrane intervenes in the process. The lag time is likewise shown to be similar in the process of permeation for all the beta-blockers studied. The values for the various molecular descriptors for each drug are given in Table 4. Multiple lineal regression using least square estimation was performed (IBM SPSS Statistics 20, 2011) on the data set with log kp as dependent variable and the predictor variables. Stepwise regression showed that solubility was not an important descriptor. This descriptor was therefore excluded from subsequent analyses. Stepwise MLR analysis of buccal permeability with the various descriptors resulted in the following model: Log kp ðcm=hÞ ¼ 3:885 ð3:770Þ þ 1:626 ð0:625Þ logDpH 6:8  0:521ð0:297Þ  nRotB þ2:236ð1:023Þ  HBA

r2 ¼ 0:857

The 95% confidence limits for each regression are given in parentheses. Log DpH 6.8, nRotB and HBA were the most important parameters describing log kp. Table 4 Molecular descriptor values for the various drugs used in this study. Drug







Acebutolol Atenolol Labetalol Metoprolol Oxprenolol Propranolol

336 266 328 267 265 259

331 261 315 273 267 258

88 85 96 51 51 41

5 5 4 4 4 3

3 4 5 2 2 2

10 8 8 9 9 6

Values from literature (Drug Bank Database and Molinspiration (MV)) Molecular weight (MW), molecular volume (MV), topological polar surface area (TPSA), number hydrogen bond acceptors (HBA), number hydrogen bond donors (HBD), number rotatable bonds (nRotB).


S. Amores et al. / International Journal of Pharmaceutics 468 (2014) 50–54

Table 5 Values of plasmatic clearance (Clp), predicted plasma concentration at steady stated (Css) and therapeutic concentration (Ct). Drug

Clp (L/h)

Css (mg/L)

Ct (mg/L)

Acebutolol Atenolol Labetalol Metoprolol Oxprenolol Propranolol

0.36 7.8 162 66 22.8 60

0.02031 0.00093 0.00013 0.00499 0.01560 0.00856

0.2–2 0.2–5 0.7–3 50–100 0.08–0.1 0.05–0.1

Clp and Ct from literature (Frishman and Alwarshetty, 2002).

This equation indicates that the distribution coefficient between the liquid solvent and the membrane lipids of the drug is positively associated with the process of permeation by the betablockers studied. The topological descriptor related to the number of rotatable bonds (nRotB) is a measure of molecular flexibility (Hou et al., 2004). This means that decreased rotatable bond count has a positive effect on the permeation rate. However, when the distribution coefficient of the drugs increase, the value of nRotB also decrease, which results in a positive effect on the permeation of the drugs through the mucosa. Number of hydrogen bond acceptors (HBA) could potentially impede skin permeation, but not more than 10 hydrogen bond acceptors (nitrogen or oxygen atoms) (the beta-blockers studied) are predicted to have good bioavailability. These results are in agreement with those published by other authors (Veber et al., 2002; Ajay et al., 1998). The potential systemic capacity after a buccal administration can be predicted by estimating the plasma concentration in steadystated (Css) from the Eq. (4). The values obtained for Css are presented in Table 5. The plasmatic clearance (Clp) and the therapeutic concentrations (Ct) taken from literature (Frishman and Alwarshetty, 2002), are also shown. We need to consider the importance of area of permeation when calculating steady stated plasma concentrations. Taking into account the area for permeation (Franz cell) used in the experimental test was of 1.1 cm2, it wouldn’t be possible to achieve concentrations within the therapeutic range for the beta-blockers studied (see Table 5). However, for example, if the drug is formulated in a patch of 10 cm2 or in a mucoadhesive gel applied to an area of 10 cm2 of buccal mucosa, it would be obtained predicted plasma concentrations at steady stated of 0.077 mg/L for propranolol (therapeutic concentration, Ct: 0.05–0.1) and 0.142 mg/L for oxprenolol (therapeutic concentration, Ct: 0.08–0.1) (see Table 5). 4. Conclusion The ex vivo buccal permeation studies performed on acebutolol, atenolol, labetalol, metoprolol, oxprenolol and propranolol demonstrate that these drugs have low intrinsic buccal permeation capacities, considering the area for permeation (Franz cell, 1.1 cm2) used in the experimental test. However, formulations of propranolol or oxprenolol in a patch of 10 cm2 or in a mucoadhesive gel applied to an area of 10 cm2 of buccal mucosa, it would be obtained predicted plasma concentrations at steady stated. Regression analysis showed that 85.7% of variability in permeability data can be explained by a model that includes distribution coefficient at pH 6.8, number of rotatable bonds and number of hydrogen bond acceptors. High lipophilicity, greater

flexibility and hydrogen bond capability were important for permeability of beta-blockers studied. References Ajay, A., Walters, W.P., Murcko, A.M., 1998. Can we learn to distinguish between drug-like and nondrug-like molecules? Journal of Medical Chemistry 41, 3314– 3327. ChemIDplus, 2007. http://chem.sis.nlm.nih.gov/chemidplus/. de Vries, M.E., Bodde, H.E., Junginger, H., 1990. The in vitro penetration of beta blocking agents through porcine bucal epithelium. Proceedings of International Symposium Control Release Bioact 17. Diaz del Consuelo, I., Jacques, Y., Pizzolato, G.P., Guy, R.H., Falson, F., 2005. Comparison of the lipid composition of porcine buccal and esophageal permeability barriers. Archives of Oral Biology 50, 981–987. Franz, T.J., 1975. Percutaneous absorption. On the relevance of in vitro data. Journal of Investigative Dermatology 64, 190–195. Frishman, W.H., Alwarshetty, M., 2002. Beta-adrenergic blockers in systemic hypertension. Pharmacokinetic considerations related tothe current guidelines. Clinical Pharmacokinetics 41, 505–516. Galey, W.R., Lonsdale, H.K., Nacht, S., 1976. The in vitro permeability of skin and buccal mucosa to selected drugs and tritiated water. Journal of Investigative Dermatology 67, 713–717. Hadgraft, J., Valenta, C., 2000. pH, pK(a) and dermal delivery. International Journal of Pharmaceutics 200, 243–247. Hao, J., Heng, P.W.S., 2003. Buccal delivery systems. Drug Development and Industrial Pharmacy 29, 821–832. Harris, D., Robinson, J.R., 1992. Drug delivery via the mucous membranes of the oral cavity. Journal of Pharmaceutical Sciences 81, 1–10. Hou, T.J., Zhang, W., Xia, K., Qiao, X.B., Xu, X.J., 2004. ADME evaluation in drug discovery. 5. Correlation of Caco-2 permeation with simple molecular properties. Journal of Chemical Information and Computer Sciences 44, 1585–1600. IBM SPSS Statistics 20. IBM Corporation 1989, 2011. Le Brun, P.P.H., Fox, P.L.A., de Vries, M.E., Bodde, H.E., 1989. In vitro penetration of some beta-adrenoreceptor blocking drugs through porcine buccal mucosa. International Journal of Pharmaceutics 49, 141–145. Lesch, C.A., Squier, C.A., Cruchley, A., Williams, D.M., Speight, P., 1989. The permeability of human oral mucosa and skin to water. Journal of Dental Research 68, 1345–1349. Liu, X., Testa, B., Fahr, A., 2011. Lipophilicity and its relationship with passive drug permeation. Pharmaceutical Research 28, 962–977. Molinspiration. 2007. http://www.molinspiration.com. Micromath Scientific Software, 1991. NCSS, 2006. User’s Guide: Regression and Curve Fitting. NCSS, Kaysville, Utah. Okamoto, H., Komatsu, H., Hashida, M., Sezaki, H., 1986. Effects of beta-cyclodextrin and di-o-methyl-beta-cyclodextrin on the percutaneous absorption of butylparaben, indomethacin and sulfanilic acid. International Journal of Pharmaceutics 30, 35–45. Rathbone, M., Drummond, B., Tucker, I., 1994. The oral cavity as a site for systemic drug delivery. Advance Drug Delivery Reviews 13, 1–22. SAS Enterprise Guide 4.1, SAS Institute Inc., 2006. Scott, D.C., Clymer, J.W., 2002. Estimation of distribution coefficients: from the partition coefficient and pKa. Pharmaceutical Technology North America 26, 30–40. Şenel, S., Hincal, A.A., 2001. Drug permeation enhancement via buccal route: possibilities and limitations. Journal of Controlled Release 72, 133–144. Shojaei, A.H., 1998. Buccal mucosa as a route for systemic drug delivery: a review. Journal of Pharmacy and Pharmaceutical Science 1, 15–30. Squier, C.A., Wertz, P.W., 1996. Structure and function of the oral mucosa and implications for drug delivery. In: Rathbone, M.J. (Ed.), Oral Mucosa Drug Delivery. Marcel Dekker, New York, pp. 1–26. Sudhakar, Y., Kuotsu, S.K., Bandyopadhyay, A.K., 2006. Buccal bioadhesive drug delivery–a promising option for orally less efficient drugs. Journal of Controlled Release 114, 15–40. Veber, D.F., Johnson, S.R., Cheng, H.Y., Smith, B.R., Ward, K.W., Kopple, K.D., 2002. Molecular properties that influence the oral bioavailability of drug candidates. Journal of Medicinal Chemisty 45, 2615–2623. Wertz, P.W., Squier, C.A., 1991. Cellular and molecular basis of barrier function in oral epithelium. Critical Reviews in Therapeutic Drug Carrier Systems 8, 237–269. Williams, A.C., 2003. Transdermal and Topical Drug Delivery, 1st ed. Pharmaceutical Press, Royal Pharmaceutical Society of Great Britain, pp. 44–64. Williams, A.C., Cornwell, P.A., Barry, B.W., 1992. On the non-Gaussian distribution of human skin permeabilities. International Journal of Pharmaceutics 86, 69–77. Xiang, J., Fang, X., Li, X., 2002. Transbuccal delivery of 20 ,30 -dideoxycytidine: in vitro permeation study and histological investigation. International Journal of Pharmaceutics 231, 57–66.