Studies on the interaction of diflunisal ion with cyclodextrins using ion-selective electrode potentiometry.

The interaction of diflunisal ion (DF) with beta-cyclodextrin (betaCD), gamma-cyclodextrin (gammaCD), and hydroxypropyl-beta-cyclodextrin (HPbetaCD) was studied in phosphate buffer, pH 7.4, at 5-37 degrees C and various CD concentrations using a home-made diflunisal ion-selective electrode. Typical direct binding plots and Scatchard plots were obtained with HPbetaCD. The Scatchard model for one class of binding sites was used for the estimation of binding parameters for the DF/HPbetaCD interaction. The estimates for n (number of binding sites per CD molecule) were in all cases very close to unity, indicating 1:1 complexation. The association constant (K) estimates decrease with increasing temperature. Sigmoidal direct binding plots and concave-downwards Scatchard plots were obtained with various betaCD or gammaCD concentrations. The Hill model was used for the estimation of the binding parameters for the DF/betaCD and DF/gammaCD interactions. Both the Hill coefficients and the binding constants were markedly dependent on the CD concentration. These findings indicate the cooperative character of DF/betaCD and DF/gammaCD interactions. The free energy change, DeltaG, and the thermodynamic parameters, DeltaH and DeltaS, were estimated for each of the interactions studied using the Van't Hoff equation.

The effect of polyethylene glycol on the charcoal adsorption of chlorpromazine studied by ion selective electrode potentiometry.

BACKGROUND: This investigation was undertaken to study: a) the adsorption characteristics of chlorpromazine to activated charcoal and its formulations Carbomix powder and Ultracarbon tablets at gastric pH; b) the effect on chlorpromazine adsorption of polyethylene glycol and its combination with electrolyte lavage solution; c) the effect of the order of addition of polyethylene glycol-electrolyte lavage solution. METHOD: Ion selective electrode potentiometry, based on the selective, direct and continuous response of a chlorpromazine-ion selective electrode to the concentration of the free drug, was used. Successive additions of microvolumes of a chlorpromazine solution were made into a charcoal slurry in acidic medium of pH 1.2 with measurement of the chlorpromazine-ion selective electrode potential at equilibrium. RESULTS: The maximum adsorption capacity values of activated charcoal, Carbomix and Ultracarbon, were 297, 563, and 382 mg/g respectively, while the affinity constant values were 40.2, 70.4, and 40.5 L/g, respectively. The adsorption of chlorpromazine to each of the Ultracarbon and Carbomix components was compared to the total adsorption of the formulations. The addition of polyethylene glycol-electrolyte lavage solution causes a slight desorption of chlorpromazine from activated charcoal at gastric pH, more pronounced when polyethylene glycol-electrolyte lavage solution follows the addition of activated charcoal, suggesting the possibility of a nonspecific binding of chlorpromazine to polyethylene glycol. The amount of chlorpromazine absorbed to Carbomix and Ultracarbon was not significantly affected at gastric pH by the presence of polyethylene glycol or polyethylene glycol-electrolyte lavage solution added either concurrently or sequentially to these formulations.

Construction of a diflunisal ion sensor and its use in automated flow-injection methods for assay, content uniformity, and dissolution studies of formulations.

A diflunisal ion selective electrode of the PVC membrane type with an ion-exchanger consisting of the tetraheptylammonium-diflunisal ion pair is described. The sensor exhibits a rapid, near-Nernstian, selective response to diflunisal anion in the pH range 7-10, with a (batch) detection limit of 1 x 10(-5) M. The ion sensor was used as a flow detector in an automated flow-injection analyzer to develop routine methods for assays (concentration range 1-50 x 10(-4) M, (flow) detection limit 2.6 x 10(-5) M), content uniformity, and dissolution studies of diflunisal formulations. No serious interference from common ions and tablet excipients was found, and the drug can be directly determined in colored samples without separation steps. Fourty measurements can be performed automatically per hour with a precision of 0.5-1.8% relative standard deviation. The automated method for the dissolution test provides a complete dissolution profile by the end of the experiment. Using the constructed ion sensor, the intramolecular hydrogen bonding of the diflunisal anion was studied, thereby revealing a new application of ion sensor potentiometry.

General treatment of competitive binding as applied to the potentiometric ion probe technique: application to the interaction of nonsteroidal anti-inflammatory drugs with bovine serum albumin.

The binding of naproxen, ketoprofen, phenylbutazone, salicylic acid, azapropazone, and indobufen to bovine serum albumin was studied by applying the potentiometric ion probe technique. An ion-selective electrode for the ion probe 1-anilino-8-naphthalene-sulfonate was utilized for the purposes of this study. A modified site-oriented competitive binding model was used for the estimation of the drugs' binding parameters, considering different number of binding sites on the competing binding class(es) for the probe and the drug. Calculations were based exclusively on the concentration data of the free probe. The model's ability for accurate estimations of binding parameters was evaluated by simulation studies. The following values of binding parameters were found at 25 degrees C for the drugs under study; naproxen, n1 = 9.1, k1 = 9.4 x 10(5) M-1; ketoprofen, n1 = 8.8, k1 = 10.8 x 10(5) M-1; phenylbutazone, n1 = 3.2, k1 = 1.4 x 10(5) M-1; salicylic acid, n1 = 2.6, k1 = 1.8 x 10(5) M-1, n2 = 21.5, k2 = 1.0 x 10(4) M-1; azapropazone, n1 = 0.5, k1 = 7.8 x 10(5) M-1, n2 = 26.3, k2 = 1.9 x 10(4) M-1; indobufen, n1 = 5.8, k1 = 5.8 x 10(5) M-1, n2 = 19.9, k2 = 3.8 x 10(5) M-1, where ni the number of binding sites of the i class and ki the corresponding association constant.

Automated flow-injection technique for use in dissolution studies of sustained-release formulations: application to iron(II) formulations.

The application of flow-injection analysis (FIA) to automated dissolution studies of sustained-release formulations is described. The long-term stability of the dissolution-FIA analyser was checked during unattended operation for 42 h. The construction of multiple calibration curves with the so-called electronic dilution FIA procedure was used to extend the linear range of the determination. The computer-controlled FIA system and the principles of associated software are described and applied to dissolution studies of sustained-release formulations of iron(II) using its sensitive reaction with the colour reagent, ferrozine. The extended linear range of the determination is 1-130 ppm iron(II) and the precision (RSD) better than 3% (n = 3).

Effect of cyclodextrins on protein binding of drugs: the diflunisal/hydroxypropyl-beta-cyclodextrin model case.

The binding of diflunisal to hydroxypropyl-beta-cyclodextrin (HP beta CD), bovine serum albumin (BSA), human serum albumin (HSA), normal human plasma, and mixed solutions of HP beta CD/protein was studied at 25 degrees C, pH 7.4, by potentiometry using an electrode selective to diflunisal. The experimental data for diflunisal/HP beta CD fit well to the 1:1 binding model. The binding of diflunisal with each of the studied proteins was compatible with a model having two independent classes of binding sites. The binding of diflunisal in mixed solutions HP beta CD/BSA, HP beta CD/HSA, and HP beta CD/plasma increased considerably when the HP beta CD concentration was increased. The binding behavior of the two biomolecules in the mixed solutions of HP beta CD/BSA or HP beta CD/HSA was described with an "additive" model formulated on the basis of the estimates of the binding parameters of diflunisal derived from the separate experiments with each one of the binders tested. The lower than theoretical binding observed in HP beta CD/plasma solutions was ascribed to the competitive displacement of diflunisal from the HP beta CD cavity by plasma cholesterol.

Determination of association constants in cyclodextrin/drug complexation using the Scatchard plot: application to beta-cyclodextrin-anilinonaphthalenesulfonates.

The appropriate Scatchard equation was developed for a system involving the formation of 1:1 and 1:2 substrate:cyclodextrin complexes. Simulation of this system was performed under the most common experimental conditions encountered in this type of study. The use of the equation allows for nonlinear least-squares estimation of the association constants. The interaction of the model compounds 1-anilino-8-naphthalenesulfonate (1,8-ANS) and 2(p-toluidinyl)-6-naphthalenesulfonate (2,6-TNS) with beta-cyclodextrin (beta-CD) was used to evaluate the theoretical model. Binding experiments were performed using either potentiometric titration or fluorimetric detection. The experimental data for 1,8-ANS/beta-CD fit well to the 1:1 binding model, with an association constant of 87 +/- 1 M-1. The association constants of the 1:1 and 1:2 2,6-TNS/beta-CD complexes utilizing direct potentiometry were 3737 +/- 6 and 149 +/- 2 M-1. It is shown that fluorimetry can give biased estimates for the association constants of the complexation 2,6-TNS/beta-CD, since the assumption of an equivalent quantum yield of bound species is not valid.

Complexation studies of cyclodextrins with tricyclic antidepressants using ion-selective electrodes.

The complexation of six tricyclic antidepressant drugs [amitriptylin (AMN), nortriptylin (NRN), imipramin (IMN), doxepin (DXN), protriptylin (PTN), and maprotilin (MPN)] with alpha- and beta-cyclodextrins (CDs) using ion-selective electrodes (ISEs) as drug ion sensors is described. Binding parameters were calculated by nonlinear fitting of the model described by the Scatchard equation, to the experimental data of a titration of a CD solution with the ion of interest. One binding site (the CD cavity) was found in all cases with both CDs. The calculated association constants at 25 degrees C using CD concentrations in the range of 0.0100-0.0010 M, varied from 4.81 x 10(3) M-1 (MPN) to 23.9 x 10(3) M-1 (AMN) in the case of beta-CD and from 50 M-1 (DXN) to 123 M-1 (MPN) in the case of alpha-CD. The precision for the estimation of the binding parameters was 0.1-5.0% (within-run RSD%) and 8-10% (between-run RSD%; n = 3). The complexation of the drugs with beta-CD was also examined as a function of temperature in the range of 5-37 degrees C; it was found to decrease by increasing temperature. Van't Hoff analysis gave good correlations (r greater than or equal to 0.989) for all drug ions studied. The estimates of the thermodynamic parameters indicate that the formation of inclusion complexes is enthalpy driven. A compensation plot based on the thermodynamic parameters delta H and delta S resulted in a linear relationship, which is indicative of a common type of force involved in the complexation of drugs to beta-CD.

Binding study of the fluorescence probe 1-anilino-8-naphthalensulfonate to human plasma and human and bovine serum albumin using potentiometric titration.

The binding of 1-anilino-8-naphthalenesulfonate (ANS) to bovine serum albumin (BSA), human serum albumin (HSA), and human plasma has been studied by potentiometric titration utilizing a laboratory constructed ion selective electrode (ISE) of ANS. Three classes of ANS binding sites were found on BSA, HSA, and plasma at 25 and 37 degrees C. Computer analysis of the data resulted in estimates for the association constants, number of binding sites (HSA, BSA), and binding capacity of each class. The association constants for the first class of binding sites at 25 degrees C were found to be 7.53 (+/- 0.59) x 10(5), 2.70 (+/- 0.20) x 10(5), and 2.64 (+/- 0.26) x 10(5) M-1 for BSA, HSA, and plasma, respectively. Lower values for the association constants of all binding classes were estimated at the higher temperature (37 degrees C). The binding capacity for ANS decreased in the order BSA, plasma, HSA.

Binding studies of ions with cyclodextrins using ion-selective electrodes.

The use of ion-selective electrodes (ISEs) to study the binding of various organic ions to alpha- and beta-cyclodextrins (CDs) is described. The ISEs respond selectively, directly, and continuously to the activity of the free ion of interest in solution. Binding parameters (intrinsic binding constants and number of binding sites) are calculated with a computer program which performs a nonlinear fit of the Scatchard model to the experimental data (electrode potential versus total ion concentration) of a titration of a CD solution with the ion of interest. Electrodes selective to chloramine-T, naproxenate, p-nitrophenolate, and chlorpromazine ions were used as models. One binding site was found for all cases, with binding constants varying from 0.67 x 10(3) M-1 (naproxenate to beta CD) to 1.27 x 10(4) M-1 (chlorpromazine to beta CD) at 25 degrees C and a CD concentration of 0.010 M. Precision for the estimations of the binding parameters was 2.0-7.5% within run and 8-10% between run (n = 3). Nonspecific binding was observed for CAT: beta CD and CPM: beta CD interactions. The effects of pH, temperature, CD concentration, and ionic strength on the binding are also presented. The overall binding phenomenon is discussed in light of the estimates of the thermodynamic parameters in relation to the size of the guest molecule, the possible mechanism(s) involved, and the forces participating in the interaction of ions with CDs.

Drug binding and solubility in milk.

The binding and solubility of nitrofurantoin, piroxicam, indomethacin, prednisolone, diazepam, dicumarol, and griseofulvin in milk were determined at 15, 25, and 37 degrees C in bovine milk samples with fat contents of 0.75 and 3.50%. Drug binding to milk components was independent of drug concentration over the drug concentration studied, and the fat content of milk strongly affected binding values of most of the listed drugs. Further, drug binding increased with decreasing temperatures for most of the drugs examined. The solubility of all drugs is greatly enhanced in milk compared to their aqueous solubility (pH 6.5 phosphate buffer). The high solubility cannot be accounted for solely on the basis of drug binding to milk components. An attempt is made to correlate the binding and solubility data with physicochemical properties of the drugs (logP, pKa, aqueous solubility). The potential significance of these findings is discussed with regard to preparation and in vivo delivery of drugs from drug-milk formulations.

Effect of temperature and fat content on the solubility of hydrochlorothiazide and chlorothiazide in milk.

The solubility of hydrochlorothiazide and chlorothiazide in milk has been studied. Experiments were carried out at 5, 15, 25, and 37 degrees C on a buffer solution of pH 6.5, a 2.6% solution of casein, bovine skim milk samples, and bovine milk samples with fat contents of 0.75, 1.70, and 3.50%. The "total" solubility of both drugs in the media studied was higher than the buffer solubility. The highest "total" solubility for both drugs was observed in skim milk. Based on binding data of thiazides to milk, the "total" solubility was split into "free" and "bound" solubility. The increases of solubility noted cannot be explained on the basis of drug-milk binding data. The enhancement of solubility was attributed to the increase of intrinsic solubility of drugs in milk. Results of the thermodynamic analysis of solubility data showed that a different solubilization process of hydrochlorothiazide may be responsible for the high solubility values found in skim milk for this drug. In contrast, the thermodynamic parameters of chlorothiazide in all types of milk are similar, indicating a common solubilization mechanism. The biopharmaceutical significance of the findings is discussed in light of the freeze-dried drug-milk formulations and coadministration of drugs with milk in general.

Construction of a naproxen ion-selective electrode and its application to pharmaceutical analysis.

A naproxenate-selective electrode with a liquid membrane consisting of a tetraheptylammonium-naproxenate ion pair dissolved in p-nitrocumene is described. The electrode exhibits a rapid and near-Nernstian response to naproxenate activity from 10(-1) to 10(-4) M at pH 9.0 (borate buffer). No serious interference from common ions and tablet excipients was found and the electrode was used for the direct assay of naproxen tablets by means of the calibration graph technique and of suppositories using the standard additions technique. A dissolution study of naproxen tablets was also carried out and the results compared favourably with those given by the USP XXI methods.

Toward a quantitative approach for the prediction of the fraction of dose absorbed using the absorption potential concept.

An equation based on the absorption potential concept was developed. This enabled us to establish an approach for the quantitative prediction of the fraction of dose absorbed. Classification of drugs into three broad categories, according to their absorption potential values in relation to the fraction of dose absorbed, was attempted. The new approach was tested using literature data with very good results.

Effect of temperature and fat content on the binding of hydrochlorothiazide and chlorothiazide to milk.

The binding of hydrochlorothiazide and chlorothiazide to milk has been measured. Experiments were carried out at 5, 15, 25, and 37 degrees C on bovine milk samples with fat contents of 0.75, 1.70, and 3.50%, using a wide range of drug concentrations to mimic concentrations encountered when a drug-milk freeze-dried system is utilized. Binding experiments with a 2.6% solution of casein were also carried out at the same temperature and concentration range of drugs. The binding to milk and casein was found to be not dependent on the concentration of drugs. The fat content of milk had no significant effect on the binding of both drugs. Higher binding was observed at lower temperatures than at higher temperatures for both drugs examined. The binding of both drugs to casein at 37 degrees C agrees fairly well with the corresponding binding to all types of milk at 37 degrees C. The potential significance of the findings in respect to preparation and in vivo delivery of drugs from drug-milk formulations is discussed.

Studies on freeze-dried drug-milk formulations. II: Effect of regenerated fluid volume on nitrofurantoin bioavailability.

The effect of different milk volumes on the extent and consistency of nitrofurantoin (1-[(5-nitrofurfurylidene)amino]hydantoin) absorption from freeze-dried nitrofurantoin-milk formulations was studied in four male volunteers in three separate crossover designs. Each volunteer received six single-dose treatments (one 100-mg nitrofurantoin capsule with 100, 200, and 400 mL of milk and 100 mg of nitrofurantoin as a freeze-dried nitrofurantoin milk formulation regenerated with 100, 200, and 400 mL of water). Analysis of the urine data revealed superiority of the nitrofurantoin-milk formulations regenerated with 200 and 400 mL of milk over the corresponding capsule formulations in the rates and extents of nitrofurantoin excretion. The binding of nitrofurantoin to casein and bovine serum albumin and its solubility in the presence of the proteins were measured in vitro. The presence of both proteins caused increases in the solubility of nitrofurantoin. Normal protein binding is responsible for the increase of nitrofurantoin solubility in the presence of bovine serum albumin, whereas the increase of nitrofurantoin solubility in the presence of casein is attributed to the formation of aggregates in casein solution at 37 degrees C. The in vivo data were discussed in light of the in vitro data. The freeze-dried nitrofurantoin-milk formulation regenerated with 200 mL of water has a potential for use as a nitrofurantoin delivery system.

Studies on drug-milk freeze-dried formulations. I: Bioavailability of sulfamethizole and dicumarol formulations.

In this study, solid dispersion formulations of dicumarol (3,3'-methylenebis[4-hydroxycoumarin]) and sulfamethizole (N'-(5-methyl-1,3, 4-thiadiazol-2-yl)sulfanilamide) in defatted milk were prepared by freeze-drying. X-ray crystallographic data showed that both drugs were dispersed in the formulations in an amorphous state. Bioequivalency comparisons between freeze-dried formulations, after regeneration with water, and control capsules containing the pure drug substances were studied in four male volunteers. Determination of the plasma dicumarol levels indicated superiority of the dicumarol-milk formulation. Statistically significant differences were found between area under the curve, maximum plasma concentration, and apparent elimination rates. Analysis of the urine sulfamethizole data revealed that the two formulations exhibit statistically equivalent rates and extents of excretion of unchanged sulfamethizole. The binding of both drugs to casein and their solubility in the presence of casein were measured in vitro. The presence of casein caused an increase in the solubility of dicumarol, while it had no effect on the solubility of sulfamethizole. Normal protein binding cannot be responsible for the effects noted. Extrapolation of the in vitro data to the in vivo situation was attempted. Drug-milk freeze-dried formulations are promising for the enhancement of the bioavailability of sparingly water soluble drugs.

Graphical approach for determining whether absorption and elimination rate constants are equal in the one-compartment open model with first-order processes.

An equation was developed which enables blood level data to be utilized for determining whether or not the first-order absorption and elimination rate constants are equal in the one-compartment open model. This equation was tested using simulated data with excellent results.