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

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: 0.3231×10(2) cm/h and 0.5666×10(2) cm/h, respectively. Multiple linear regression (MLR) using least square estimation was performed on the data set with logkpas dependent variable and the descriptors as predictor 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.

Interactions of beta-blockers with model lipid membranes: molecular view of the interaction of acebutolol, oxprenolol, and propranolol with phosphatidylcholine vesicles by time-dependent fluorescence shift and molecular dynamics simulations.

Since pharmacokinetic and pharmacodynamic activities of drugs are often related to their interactions with biomembranes, it is of high interest to establish an approach for the characterization of these interactions at the molecular level. For the present study, beta-blockers (oxprenolol, propranolol, and acebutolol) were selected due to their well described nonspecific membrane effects (NME). Their interactions with model lipid membranes composed of palmitoyloleoylphosphatidylcholine (POPC) were studied using Time-Dependent Fluorescence Shift (TDFS) and Generalized Polarization (GP) as well as molecular dynamics (MD) simulations. Liposomal vesicles were labeled with fluorescent membrane polarity probes (Laurdan, Prodan, and Dtmac). Increasing beta-blocker concentrations (0-10 mM for acebutolol and oxprenolol, and 0-1.5 mM for propranolol) significantly rigidifies the lipid bilayer at the glycerol and headgroup level, which was detected in the steady-state and in the time-resolved fluorescence data. The effects of propranolol were considerably stronger than those of the two other beta-blockers. The addition of fluorescent probes precisely located at different levels within the lipid bilayer revealed the insertion of the beta-blockers into the POPC bilayer at the glycerol backbone level, which was further confirmed by MD simulations in the case of propranolol.

FTIR spectroscopic investigation and modeling of solute/polymer interactions in the hydrated state.

Attenuated total reflectance infrared spectroscopy was used to investigate possible interactions during transport of oxprenolol x HCl, bovine serum albumin, alpha-chymotrypsin, and fibrinogen through poly(acrylic acid) and its random copolymeric gels. Carbonyl and carboxylate ion peak shifts were used to identify drug/gel binding due to electrostatic and hydrogen bond interactions between the polymer carrier and the drugs tested. These findings were used to interpret the decrease in calculated diffusion coefficients of drugs diffusing through these gels and the associated hindering of drug transport. A model was developed to analyze this transport process as a function of the binding heat of the drug with the polymer.

In vitro permeability of eight beta-blockers through Caco-2 monolayers utilizing liquid chromatography/electrospray ionization mass spectrometry.

It is demonstrated that the apparent permeability (P(app)) coefficients of beta-adrenoceptor antagonist drugs can easily be determined for Caco-2 cell culture intestinal models utilizing liquid chromatography/mass spectrometry (LC/MS). The LC/MS method with electrospray ionization in the single ion monitoring mode showed an increased sensitivity of 1000-fold compared with LC/UV detection and enhanced selectivity with respect to both LC/UV and radioactivity assays. The P(app) coefficients of beta-adrenoceptor antagonists determined by LC/MS have the same ranking order as those determined by LC/UV and radioactivity assays. However, the P(app) coefficients determined in this study showed significant discrepancies from those determined in other laboratories. There are several experimental factors that directly affect the absolute value of the P(app) coefficients, including pH gradients, additional diffusion barriers (i.e. unstirred water layer and type of filter support), analyte concentration, detection method and possibly cell culture variations. These parameters should be controlled when generating Caco-2 P(app) coefficients for different compounds.

Relationship between amount of beta-blockers permeating through the stratum corneum and skin irritation after application of beta-blocker adhesive patches to guinea pig skin.

We evaluated the relationship between the cumulative amounts of 5 kinds of beta-blockers (alprenolol, oxprenolol, timolol, acebutolol and atenolol) permeating through the stratum corneum and a* values obtained by measuring the formation of erythema, a skin irritation reaction, with a chromameter after transdermal application of adhesive patches containing 2 beta-blocker to the skin of guinea pigs. The cumulative amount of beta-blocker released from each adhesive patch to the skin increased with the increase in application time. The contents of alprenolol, oxprenolol and timolol in the stratum corneum and in the stripped skin increased markedly up to 4 h after application and thereafter were maintained at high levels up to 24 h. The contents of acebutolol and atenolol, on the other hand, increased up to 24 h, but these values were low. a* values of all adhesive patches 24 h after application were higher than those before application. The correlation coefficients between the cumulative amounts of alprenolol, oxprenolol, timolol, acebutolol or atenolol permeating through the stratum corneum and (delta a* -delta a*Placebo) values were 0.739, 0.717, 0.722, 0.551 and 0.633, respectively. The correlation coefficient calculated by averaging the cumulative amounts of 6 kinds of beta-blockers permeating through the stratum corneum [including propranolol which was reported previously (Kobayashi I., et al., Biol. Pharm. Bull., 19, 839-844 (1996))] was 0.731, higher than the correlation coefficient between contents of these beta-blockers in the stripped skin and (delta a* -delta a*Placebo) values (r = 0.552). This suggests that there was a high correlation between the cumulative amounts of beta-blockers permeating through the stratum corneum and (delta a* -delta a*Placebo) values.

Concentration-dependent binding of the chiral beta-blocker oxprenolol to isoelectric or negatively charged unilamellar vesicles.

Large unilamellar vesicles (LUVs) of different lipid compositions were used to study the type of binding of the beta-blocking cationic agent oxprenolol to the lipid matrix of biological membranes at a physiologic pH value of 7.4. When isoelectric membranes of pure egg lecithin or egg lecithin/cholesterol (7:3 mol/mol) were used, a linear relationship between membrane-bound and free oxprenolol indicated a constant molar partition coefficient of 54 or 44 between the liposomal and the aqueous phase over a wide concentration range of the drug up to 25 mM. This pointed to deep insertion of the drug molecules into the hydrophobic membrane interior. Drug binding to membranes of negatively charged phosphatidylserine from bovine brain was cooperative with a Hill coefficient h of 3.4 at concentrations below 0.5 mM and a molar ratio Re of bound drug per lipid of 1:10. Above drug concentrations of 2.5 mM and Re = 1:5, a constant molar partition coefficient of 33 could be estimated. R-oxprenolol or S-oxprenolol, as well as the racemic drug, showed no differences in membrane binding, even with egg lecithin LUVs containing 20 mol% of the negatively charged (2S, 4R)-N-(hexadecanoyl)-4-hydroxyproline, which has a pronounced chiral headgroup. Our results suggest that enantioselective interactions of the chiral oxprenolol with the chiral lipids of biological membranes can be excluded. Furthermore, surface adsorption of the drug is probable only on the negatively charged cytosolic side of biological plasma membranes, whereas on the isoelectric exterior the cationic drug is inserted deeply into the membrane.

Effects of turpentine oil pretreatment on beta-blocker pharmacokinetic parameters in rats.

Turpentine oil treatment (0.2 mL kg-1, s.c.) was used to increase the plasma concentration of alpha 1-acid glycoprotein (0.13 mg mL-1 in control rats) to 1.72 mg mL-1 after 2 days, and allow assessment of its effects on the pharmacokinetics and stereoselective binding of three beta-blockers. Racemates (5 mg kg-1) were administered intravenously to control and turpentine oil-pretreated rats and the plasma concentrations were determined up to 90 min. Stereoselective analysis showed the apparent distribution volume and the area under plasma concentration-time curves (AUC) of R-(+)-propranolol to be, respectively, one-quarter and twice those of the S-(-)-enantiomer and differences in pharmacokinetic parameters between the two were magnified by turpentine oil pretreatment. Pharmacokinetic parameters of oxprenolol enantiomers were essentially similar for the controls but after turpentine oil pretreatment, a higher affinity of the R-(+)-enantiomer for plasma was observed. Acebutolol enantiomers behaved non-stereospecifically throughout. These results were consistent with predictions from the in-vitro stereospecific binding properties of these agents to purified rat alpha 1-acid glycoprotein.

Formation of the N-nitroso derivatives of six beta-adrenergic-blocking agents and their genotoxic effects in rat and human hepatocytes.

Six beta-adrenergic-blocking drugs, atenolol, metoprolol, nadolol, oxprenolol, propranolol and sotalol, were found to react with sodium nitrite in HCl solution, yielding the corresponding N-nitrosamines. The genotoxic activity of the six nitrosamines was evaluated in primary cultures of both rat and human hepatocytes; DNA fragmentation was measured by the alkaline elution technique, and DNA repair synthesis by quantitative autoradiography. Positive dose-related responses were produced in cells of both species after 20 h of exposure to the following subtoxic concentrations: NO-propranolol, 0.01-0.1 mM; NO-oxprenolol, 0.03-1 mM; NO-atenolol and NO-metoprolol, 0.1-1 mM; and NO-nadolol and NO-sotalol, 0.3-3 mM. Modest but statistically significant differences between the DNA-damaging potencies for the two species were observed with NO-atenolol and NO-oxprenolol, which were both more active against rat hepatocytes, and with NO-propranolol, which was more active against human hepatocytes. At equal or higher concentrations, the six N-nitrosamines did not produce DNA fragmentation in Chinese hamster lung V79 cells; this indicates that they behave as indirectly acting compounds, which need to be transformed into reactive metabolites in order to exert a genotoxic effect.

Influence of inflammation on serum concentration, molecular heterogeneity and drug binding properties of canine alpha-1-acid glycoprotein.

The concentration and the heterogeneity of alpha-1-acid glycoprotein (alpha-1-AGP) and oxprenolol binding were determined in serum of healthy dogs and dogs with inflammatory disease. In inflammation, an increase in the mean alpha-1-AGP concentration from 0.47 to 2.85 g/l was accompanied by a reduction in the mean free oxprenolol fraction from 25% to 6%. alpha-1-AGP concentration and oxprenolol binding were inversely correlated. The heterogeneity of canine alpha-1-AGP remained essentially unchanged in dogs with inflammation and, in both these dogs and the controls, between five and seven forms with different isoelectric points and one single concanavalin A-reactive form were detected. It is concluded that in dogs, as in humans, oxprenolol binds to serum alpha-1-AGP. Changes in serum binding of oxprenolol during inflammation result from a change in the serum concentration of alpha-1-AGP rather than a change of molecular heterogeneity.

Influence of continuous ambulatory peritoneal dialysis on serum alpha 1-acid glycoprotein concentration and drug binding.

The influence of continuous ambulatory peritoneal dialysis (CAPD) on the concentrations of alpha 1-acid glycoprotein in serum and dialysate and on the serum binding of oxprenolol, propranolol and phenytoin has been studied. Before starting CAPD treatment, the serum binding of oxprenolol and propranolol was higher and that of phenytoin lower than in healthy volunteers, and the serum alpha 1-AGP concentration was higher. During the first days to weeks after starting CAPD, the serum alpha 1-AGP concentration rose with a concomitant increase in the binding of oxprenolol and propranolol. Subsequently, the alpha 1-AGP level and the binding of oxprenolol and propranolol decreased to the values found before starting CAPD. The binding of phenytoin showed little change. The concentration of alpha 1-AGP in dialysate was 2 to 5% of that in serum.

Dog alpha-1-acid glycoprotein: purification and biochemical characterization.

Dog alpha-1-acid glycoprotein was purified to homogeneity from dog serum in a three-step procedure involving precipitation with sulphosalicylic acid, isoelectric focusing and size exclusion chromatography. The molecular heterogeneity in the peptide part and in the carbohydrate part of the molecule was investigated with analytical isoelectric focusing in a narrow pH range and crossed immunoaffinity electrophoresis with concanavalin A (con A) in the first-dimension gel. Up to seven molecular forms with different isoelectric points were found, whereas only a single con A-dependent molecular form was detected.

The pharmacokinetics of oxprenolol following oral and rectal dosing--a comparison of delivery systems and routes of administration.

Plasma oxprenolol concentrations were measured in eight healthy volunteers who received equivalent oral doses of the drug in the form of an aqueous solution and a 10/170 oxprenolol Oros drug delivery system. Absorption from the lower gastrointestinal tract was assessed by measurement of plasma concentrations after rectal administration of the pre-equilibrated Oros systems. Because three of the first four volunteers suffered local irritation, however, the other four volunteers received Slow Trasicor 160 mg orally as a comparative preparation. The rate of in vivo absorption after oral administration of the Oros system closely mirrored its in vitro release rate. Drug availability from Oros was reduced, however, and was equivalent to 77% of that from the oral solution. Oxprenolol was well absorbed from the rectum while the system was present in this segment of the gut. The reduced systemic availability in three of the volunteers could be accounted for largely by drug loss when the system was expelled. Slow Trasicor produced higher peaks but lower 24 h plasma concentrations than the orally administered Oros system. As judged from the relative areas under the plasma concentration-time curve, however, the availability of the drug from the two dosage forms was comparable.

Alpha 1-acid glycoprotein and serum binding of drugs in healthy and diseased dogs.

Inter-individual variation in drug serum protein binding was studied in healthy dogs and in dogs with inflammatory diseases for lidocaine, oxprenolol and propranolol, which bind mainly to alpha 1-acid glycoprotein (alpha 1-AGP), and for diazepam, digitoxin and phenytoin, which bind mainly to albumin. For the drugs mostly bound to alpha 1-AGP, in both groups of dogs binding varied considerably, and it was markedly higher in dogs with inflammatory disease. For the other drugs, the variation in binding was smaller and did not differ between the two groups of dogs. In both groups of dogs, the alpha 1-AGP concentration varied widely; it was higher in the serum of the dogs with inflammation, while the concentration of albumin was lower in these animals. There was a significant negative correlation between percentage free lidocaine, oxprenolol or propranolol and alpha 1-AGP concentration, suggesting that the inter-individual variation in binding of these drugs is due to the variation in alpha 1-AGP concentration. There was a marked intra-individual variation in lidocaine binding and in serum alpha 1-AGP concentration, studied over a period of 3 weeks in healthy dogs; a significant negative correlation between percentage free lidocaine and alpha 1-AGP concentration was obtained.

Gastric and intestinal absorption of oxprenolol in humans.

The gastrointestinal absorption of the beta blocker oxprenolol was investigated in four healthy subjects by an intubation technique. Oxprenolol was introduced into the stomach, dissolved in a homogenized meal containing the marker 14C-polyethylene glycol (PEG) 4000. Unlabeled PEG 4000 was perfused during the whole experiment into the duodenum at the ampulla of Vater. Samples of luminal contents were collected at regular intervals over four hours in the stomach, at the angle of Treitz, and 30 cm below this point. Blood was also collected. Oxprenolol was not absorbed in the stomach. About 80% of the drug emptied from the stomach was absorbed in the duodenum, and 80% of that released from the duodenum was absorbed in a 30-cm segment of the jejunum. The amounts absorbed in these two intestinal segments were directly proportional to the amounts delivered. The areas under the plasma concentration-time curves were not related to the amounts absorbed. A single dose of oxprenolol taken with an homogenized meal did not modify the gastric emptying and secretory response.

Comparison of pharmacokinetic profiles of single and multiple doses of a slow release Oros oxprenolol delivery system in young normotensive and elderly hypertensive subjects.

Oxprenolol in an Oros 8/130 sustained release osmotic pump system (equivalent to 120 mg oxprenolol hydrochloride in a conventional formulation and releasing 8 mg h-1) was given to eight normal young subjects (mean age 23 years) and eight elderly hypertensive patients (mean age 77 years). The plasma concentration-time profiles of oxprenolol were determined over 32 h using gas liquid chromatography after the initial dose and following seven doses. The elderly patients had a significantly higher AUC and maximum plasma oxprenolol concentration following both the first and final doses studied. It is unlikely that this difference is due to a prolonged absorption phase in the elderly patients. Reduced drug clearance seems the most probable explanation.

The disposition and metabolism of 14C-oxprenolol.HCl in man.

The disposition and metabolism of oxprenolol have been investigated in two healthy male volunteers, following a single 160 mg oral dose of racemic 14C-labelled oxprenolol. Absorption was rapid and complete. Peak blood concentrations of total radioactivity were 8.83 and 8.21 nmol X g-1 after 1 and 1.5 h in the two subjects. After 4 days 93.4 and 81.9% of the dose was excreted in urine, and a total of 96.6 and 84.5% found in the excreta. Mean peak blood concentrations of unchanged R(+)- and S(-)- oxprenolol were 0.83 and 0.81 nmol X g-1. Maximal concentrations of the glucuronides of the R(+)- and S(-)- isomers were 1.98 and 3.51 nmol X g-1. The mean half-lives of both oxprenolol enantiomers were 1.8 h, those of their glucuronides were 3.2 h (R(+] and 4.6 h (S(-]. Unchanged oxprenolol and the oxprenolol glucuronides constituted 11.4 and 66.5% of the area under the blood concentration-time curve (AUC, 0-24 h) of total radioactivity. The AUC-ratio of R(+) to S(-) was 1.19 for free oxprenolol and 0.36 for the glucuronides. Free metabolites II-X represented together 4.3% of 14C-AUC, and their glucuronides 15.2%. In urine, 1.8 and 1.0% of the total radioactivity was present as unchanged R(+)- and S(-)- oxprenolol, respectively. The glucuronides of the enantiomers accounted for 24.5 and 26.5%. The percentages of free 4- and 5-hydroxy oxprenolol were 0.7 and 2.4% while those of their glucuronides were 12.3 and 7.5%. Metabolites IV-X constituted together 6.2% in free form and 5.3% in conjugated form. In conclusion, the good mass balances in blood and urine has enabled the comprehensive and quantitative description of the metabolic fate of oxprenolol in man. Oxprenolol is extensively metabolized, direct O-glucuronidation being the major metabolic pathway and oxidative reactions minor ones. The disposition of the oxprenolol enantiomers revealed no remarkable stereoselective differences.

The pharmacokinetics and dynamics of oxprenolol: a simulation study with six subjects.

Experimental results of plasma concentration determinations and lowering of exercise heart rate for six subjects taking a conventional tablet and a sustained release preparation of oxprenolol have been analysed by a comprehensive computer simulation model. Individual plasma values were simulated using a lest squares procedure and the results were applied to evaluate individual release patterns following dosage with the sustained release preparation. Application of the model to the lowering of exercise heart rate indicated that the response is in a steady state with the plasma values and that the response-concentration relation is of the saturable, Emax, type. The parameters for this were evaluated for each subject for the results after a dose of a conventional tablet. These parameters were applicable to the results after dosage with a sustained release preparation. The method should be applicable to other sustained release preparations.

A comparison of the pharmacokinetics of atenolol, metoprolol, oxprenolol and propranolol in elderly hypertensive and young healthy subjects.

Six elderly patients with established hypertension and six young healthy subjects were studied after 8 days of treatment with atenolol 50 mg day-1, metoprolol 50 mg day-1, oxprenolol 80 mg day-1 and propranolol 80 mg day-1. The area under the blood concentration-time curve was increased in the elderly group for each drug, but the difference was statistically significant only for atenolol. The lower serum albumin concentrations in the elderly group did not result in a decrease in the percentage of propranolol or oxprenolol bound to serum proteins.

Beta-blocker brain concentrations in man.

Patients with subarachnoid haemorrhage (SAH) have been shown to benefit from beta-blockade. SAH patients who came to surgery were investigated if they had been receiving chronic (approximately one week) oral treatment with either hydrophilic atenolol (100 mg/day) or one of the following lipophilic beta-blockers: propranolol (80 mg b.i.d.), oxprenolol (80 mg b.i.d.), or metoprolol (100 mg b.i.d.). Cerebrospinal fluid concentrations of beta-blockers did not reflect their concentrations in the brain. Brain concentrations of the three lipophilic beta-blockers were 10-20 times higher than those of atenolol. The approximate brain/plasma concentration ratios were 26 for propranolol, 50 for oxprenolol, 12 for metoprolol, and 0.2 for atenolol. The brain is thus buffered from peak blood concentrations of atenolol, and this may account for the low incidence of CNS-related side-effects with this hydrophilic beta-blocker.

N-dealkylation of oxprenolol: formation of 3-aryloxypropane-1,2-diol, 3-aryloxylactic acid, and 2-aryloxyacetic acid metabolites in the rat.

Oxprenolol (1), like related beta-adrenergic antagonists, undergoes oxidative N-dealkylation to form the corresponding 3-aryloxypropane-1,2-diol (2), 3-aryloxylactic acid (3), and 2-aryloxyacetic acid (4) metabolites. Compounds 3 and 4 were synthesized by conversion of 2-allyloxyphenol (5) to the aryloxyacetaldehyde 6 and subsequent elaboration to the desired acids. Both acids (3 and 4) and glycol 2 were confirmed as metabolites formed from 1 in vivo in the rat and in vitro in the rat liver 9000 X g supernatant fraction. Incubation of a pseudoracemate of 1, made up of equal molar amounts of (2S)-1-d0 and (2R)-1-d2, showed that 2 and 3 arise principally from (2S)-1 by S/R ratios of approximately 5:1 and 2:1, respectively. On the other hand, acetic acid derivative 4 arises about equally from both enantiomers of 1.