The pH-dependence in the partitioning behaviour of (RS)-[3H]propranolol between MDCK cell lipid vesicles and buffer

Dynamic Protonation Dramatically Affects the Membrane Permeability of Drug-like Molecules

Permeability (P_m) across biological membranes is of fundamental importance and a key factor in drug absorption, distribution, and development. Although the majority of drugs will be charged at some point during oral delivery, our understanding of membrane permeation by charged species is limited. The canonical model assumes that only neutral molecules partition into and passively permeate across membranes, but there is mounting evidence that these processes are also facile for certain charged species. However, it is unknown whether such ionizable permeants dynamically neutralize at the membrane surface or permeate in their charged form. To probe protonation-coupled permeation in atomic detail, we herein apply continuous constant-pH molecular dynamics along with free energy sampling to study the permeation of a weak base propranolol (PPL), and evaluate the impact of including dynamic protonation on P_m. The simulations reveal that PPL dynamically neutralizes at the lipid-tail interface, which dramatically influences the permeation free energy landscape and explains why the conventional model overestimates the assigned intrinsic permeability. We demonstrate how fixed-charge-state simulations can account for this effect, and propose a revised model that better describes pH-coupled partitioning and permeation. Our results demonstrate how dynamic changes in protonation state may play a critical role in the permeation of ionizable molecules, including pharmaceuticals and drug-like molecules, thus requiring a revision of the standard picture.

Preferential Adsorption of l-Histidine onto DOPC/Sphingomyelin/3β-[N-(N',N'-dimethylaminoethane)carbamoyl]cholesterol Liposomes in the Presence of Chiral Organic Acids

We investigated the effect of organic acids such as mandelic acid (MA) and tartaric acid (TA) on the adsorption behavior of both histidine (His) and propranolol (PPL) onto liposomes. A cationic and heterogeneous liposome prepared using 1,2-dioleoyl-sn-glycero-3-phosphocholine (DOPC)/sphingomyelin (SM)/3β-[N-(N',N'-dimethylaminoethane)carbamoyl]cholesterol (DC-Ch) in a ratio of (4/3/3) showed the highest adsorption efficiency of MA and TA independent of chirality, while neutral liposome DOPC/SM/cholesterol = (4/3/3) showed low efficiency. As expected, electrostatic interactions were dominant in MA or TA adsorption onto DOPC/SM/DC-Ch = (4/3/3) liposomes, suggesting that organic acids had adsorbed onto SM/DC-Ch-enriched domains. The adsorption behaviors of organic acids onto DOPC/SM/DC-Ch = (4/3/3) were governed by Langmuir adsorption isotherms. For adsorption, the membrane polarities slightly decreased (i.e., membrane surface was hydrophilic), but no alterations in membrane fluidity were observed. In the presence of organic acids that had been preincubated with DOPC/SM/DC-Ch = (4/3/3), the adsorption of l- and d-His onto those liposomes was examined. Preferential l-His adsorption was dramatically prevented only in the presence of l-MA, suggesting that the adsorption sites for l-His and l-MA on DOPC/SM/DC-Ch = (4/3/3) liposomes are competitive, while those for l-His and d-MA, l-TA, and d-TA are isolated.

In Vitro and in Silico Tools To Assess Extent of Cellular Uptake and Lysosomal Sequestration of Respiratory Drugs in Human Alveolar Macrophages

Accumulation of respiratory drugs in human alveolar macrophages (AMs) has not been extensively studied in vitro and in silico despite its potential impact on therapeutic efficacy and/or occurrence of phospholipidosis. The current study aims to characterize the accumulation and subcellular distribution of drugs with respiratory indication in human AMs and to develop an in silico mechanistic AM model to predict lysosomal accumulation of investigated drugs. The data set included 9 drugs previously investigated in rat AM cell line NR8383. Cell-to-unbound medium concentration ratio (K_p,cell) of all drugs (5 μM) was determined to assess the magnitude of intracellular accumulation. The extent of lysosomal sequestration in freshly isolated human AMs from multiple donors (n = 5) was investigated for clarithromycin and imipramine (positive control) using an indirect in vitro method (±20 mM ammonium chloride, NH₄Cl). The AM cell parameters and drug physicochemical data were collated to develop an in silico mechanistic AM model. Three in silico models differing in their description of drug membrane partitioning were evaluated; model (1) relied on octanol-water partitioning of drugs, model (2) used in vitro data to account for this process, and model (3) predicted membrane partitioning by incorporating AM phospholipid fractions. In vitro K_p,cell ranged >200-fold for respiratory drugs, with the highest accumulation seen for clarithromycin. A good agreement in K_p,cell was observed between human AMs and NR8383 (2.45-fold bias), highlighting NR8383 as a potentially useful in vitro surrogate tool to characterize drug accumulation in AMs. The mean K_p,cell of clarithromycin (81, CV = 51%) and imipramine (963, CV = 54%) were reduced in the presence of NH₄Cl by up to 67% and 81%, respectively, suggesting substantial contribution of lysosomal sequestration and intracellular binding in the accumulation of these drugs in human AMs. The in vitro data showed variability in drug accumulation between individual human AM donors due to possible differences in lysosomal abundance, volume, and phospholipid content, which may have important clinical implications. Consideration of drug-acidic phospholipid interactions significantly improved the performance of the in silico models; use of in vitro K_p,cell obtained in the presence of NH₄Cl as a surrogate for membrane partitioning (model (2)) captured the variability in clarithromycin and imipramine K_p,cell observed in vitro and showed the best ability to predict correctly positive and negative lysosomotropic properties. The developed mechanistic AM model represents a useful in silico tool to predict lysosomal and cellular drug concentrations based on drug physicochemical data and system specific properties, with potential application to other cell types.

Spectroscopic investigation of fluorinated phenols as pH-sensitive probes in mixed liposomal systems

The pK_a values of three fluorinated phenols, 2,4,6-trifluorophenol (3FP), 2,3,5,6-tetrafluorophenol (4FP) and 2,3,4,5,6-pentafluorophenol (5FP) have been measured by using UV-vis and ¹⁹F-NMR spectroscopy at 25 °C in water and in the presence of pure POPC, pure DDAB and mixed POPC–DDAB liposomes.

Alternative measures of lipophilicity: from octanol-water partitioning to IAM retention

This review describes lipophilicity parameters currently used in drug design and QSAR studies. After a short historical overview, the complex nature of lipophilicity as the outcome of polar/nonpolar inter- and intramolecular interactions is analysed and considered as the background for the discussion of the different lipophilicity descriptors. The first part focuses on octanol-water partitioning of neutral and ionisable compounds, evaluates the efficiency of predictions and provides a short description of the experimental methods for the determination of distribution coefficients. A next part is dedicated to reversed-phase chromatographic techniques, HPLC and TLC in lipophilicity assessment. The two methods are evaluated for their efficiency to simulate octanol-water and the progress achieved in the refinement of suitable chromatographic conditions, in particular in the field of HPLC, is outlined. Liposomes as direct models of biological membranes are examined and phospolipophilicity is compared to the traditional lipophilicity concept. Difficulties associated with liposome-water partitioning are discussed. The last part focuses on Immobilised Artificial Membrane (IAM) chromatography as an alternative which combines membrane simulation with rapid measurements. IAM chromatographic retention is compared to octanol-water and liposome-water partitioning as well as to reversed-phase retention and its potential to predict biopartitioning and biological activities is discussed.

The Thermodynamics of the Partitioning of Ionizing Molecules Between Aqueous Buffers and Phospholipid Membranes

PURPOSE

To study the thermodynamics of partitioning of eight ionising dual D2-recepto beta2-adrenoceptor agonists between vesicles of L-alpha-dimyristoylphosphatidylcholine (DMPC) and aqueous buffers.

METHODS

The thermodynamics of partitioning have been studied by isothermal titration calorimetry (ITC).

RESULTS

Compounds which are predominantly cationic at pH 7.4 (designated as class 1 compounds) have a more exothermic partitioning than those which are predominantly in the electronically neutral form (designated as class 2 compounds) at pH 7.4, and less positive standard entropies of partitioning. Under acidic conditions (pH 4.0), class compounds 2 (predominantly electronically neutral at pH 7.4) are almost completely cationic and accordingly have a more exothermic partitioning than at pH 7.4. The standard entropies of partitioning also depend on the pH. When the compounds are predominantly cationic, the standard entropy change is less positive (less favourable) than under conditions where the compounds are predominantly electronically neutral.

CONCLUSIONS

The observations are consistent with the notion of there being a favourable electrostatic interaction (enthalpically) between the positively charged amino-group of predominantly cationic compounds and the negatively charged phosphate group of the vesicle.

Interaction of Basic Drugs with Lipid Bilayers Using Liposome Electrokinetic Chromatography

PURPOSE

This study explores factors influencing the interactions of positively charged drugs with liposomes using liposome electrokinetic chromatography (LEKC) for the development of LEKC as a rapid screening method for drug-membrane interactions.

METHODS

Liposomes were prepared and the retention factors were measured for a series of basic drugs under a variety of buffer conditions, including various buffer types, concentrations, and ionic strengths as well as using different phospholipids and liposome compositions. LEKC retention is compared with octanol-water partitioning.

RESULTS

The interaction of ionizable solutes with liposomes decreased with increasing ionic strength of the aqueous buffer. The type of buffer also influences positively charged drug partitioning into liposomes. Varying the surface charge on the liposomes by the selection of phospholipids influences the electrostatic interactions, causing an increase in retention with increasing percentages of anionic lipids in the membrane. Poor correlations are observed between LEKC retention and octanol-water partitioning.

CONCLUSIONS

These studies demonstrate the overall buffer ionic strength at a given pH is more important than buffer type and concentration. The interaction of positively charged drugs with charged lipid bilayer membranes is selectively influenced by the pKa of the drug. Liposomes are more biologically relevant in vitro models for cell membranes than octanol, and LEKC provides a unique combination of advantages for rapid screening of drug-membrane interactions.

Drug permeation in biomembranes

In the past decades, it has become increasingly apparent that in addition to therapeutic effect, drugs need to exhibit favourable absorption, distribution, metabolism and excretion (ADME) characteristics to produce a desirable response in vivo. As the recent progress in drug discovery technology enables rapid synthesis of vast numbers of potential drug candidates, robust methods are required for the effective screening of compounds synthesized within such programs, so that compounds with poor pharmacokinetic properties can be rejected at an early stage of drug development. Furthermore, a viable in silico method would save resources by enabling virtual screening of drug candidates already prior to synthesis. This review gives a general overview of the approaches aimed at predicting biological permeation, one of the cornerstones behind the ADME behaviour of drugs. The most important experimental and computational models are reviewed. Physicochemical factors underlying the permeation process are discussed.

Liposome/water lipophilicity: methods, information content, and pharmaceutical applications

This review discusses liposome/water lipophilicity in terms of the structure of liposomes, experimental methods, and information content. In a first part, the structural properties of the hydrophobic core and polar surface of liposomes are examined in the light of potential interactions with solute molecules. Particular emphasis is placed on the physicochemical properties of polar headgroups of lipids in liposomes. A second part is dedicated to three useful methods to study liposome/water partitioning, namely potentiometry, equilibrium dialysis, and (1)H-NMR relaxation rates. In each case, the principle and limitations of the method are discussed. The next part presents the structural information encoded in liposome/water lipophilicity, in other words the solutes' structural and physicochemical properties that determine their behavior and hence their partitioning in such systems. This presentation is based on a comparison between isotropic (i.e., solvent/water) and anisotropic (e.g., liposome/water) systems. An important factor to be considered is whether the anisotropic lipid phase is ionized or not. Three examples taken from the authors' laboratories are discussed to illustrate the factors or combinations thereof that govern liposome/water lipophilicity, namely (a) hydrophobic interactions alone, (b) hydrophobic and polar interactions, and (c) conformational effects plus hydrophobic and ionic interactions. The next part presents two studies taken from the field of QSAR to exemplify the use of liposome/water lipophilicity in structure-disposition and structure-activity relationships. In the conclusion, we summarize the interests and limitations of this technology and point to promising developments.

Electron Cryo‐Microscopy Reveals Mechanism of Action of Propranolol on Artificial Membranes

The pharmacological activity of several amphiphilic drugs is often related to their ability to interact with biological membranes. Propranolol is an efficient multidrug resistance (MDR) modulator; it is a nonselective beta-blocker and is thought to reduce hypertension by decreasing the cardiac frequency and thus blood pressure. It is used in drug delivery studies in order to treat systemic hypertension. We are interested in the interaction of propranolol with artificial membranes, as liposomes of controllable size are used as biocompatible and protective structures to encapsulate labile molecules, such as proteins, nucleic acids or drugs, for pharmaceutical, cosmetic or chemical applications. We present here a study of the interaction of propranolol, a cationic surfactant, with pure egg phosphatidylcholine (EPC) vesicles. The gradual transition from liposome to micelle of EPC vesicles in the presence of propranolol was monitored by time-resolved electron cryo-microscopy (cryo-EM) under different experimental conditions. The liposome-drug interaction was studied with varying drug/lipid (D/L) ratios and different stages were captured by direct thin-film vitrification. The time-series cryo-EM data clearly illustrate the mechanism of action of propranolol on the liposome structure: the drug disrupts the lipid bilayer by perturbing the local organization of the phospholipids. This is followed by the formation of thread-like micelles, also called worm-like micelles (WLM), and ends with the formation of spherical (globular) micelles. The overall reaction is slow, with the process taking almost two hours to be completed. The effect of a monovalent salt was also investigated by repeating the lipid-surfactant interaction experiments in the presence of KCl as an additive to the lipid/drug suspension. When KCl was added in the presence of propranolol the overall reaction was the same but with slower kinetics, suggesting that this monovalent salt affects the general lipid-to-micelle transition by stabilizing the membrane, presumably by binding to the carbonyl chains of the phosphatidylcholine.

Absorption prediction from physicochemical parameters

Modern drug design not only focuses on the pharmacological activity of a compound but also considers its ability to be absorbed and to reach its site of action. The prediction of in vivo barrier permeabilities, particularly intestinal absorption and blood-brain barrier passage, are substantial concerns in the development of new drug compounds. For several decades, n-octanol-water partition coefficients dominated absorption prediction. In recent years, the basic physicochemical parameters describing both membrane permeability and lipophilicity have been established. This review provides an outline of some selected absorption-prediction models with emphasis on intestinal absorption.

Determination of liposome partitioning of ionizable drugs by titration

Drug partitioning to liposomes has been suggested as a model for partitioning to biomembranes but has been lacking a rapid analytical assay useful for drug screening. A fast pH-metric titration method for the determination of liposome partitioning of ionizable drugs using small unilamellar phosphatidylcholine vesicles prepared by sonic homogenization has been successfully developed, enabling the use of high lipid-to-drug ratios. Liposome-water partition coefficients of diclofenac and propranolol were determined to study the impact of varying titration parameters, temperature, equilibration time, lipid, and liposome types on the partitioning. To validate this method, the results were compared to literature values generated with different techniques and to pH-metric titration results with large unilamellar vesicles. The rapid pH-metric assay gave liposome partitioning data for the two model compounds which were consistent with other analytical techniques and liposome types.

Immobilized artificial membrane (IAM)-HPLC for partition studies of neutral and ionized acids and bases in comparison with the liposomal partition system

PURPOSE

To study the partitioning of model acids ((RS)-warfarin and salicylic acid), and bases (lidocaine, (RS)-propranolol and diazepam), with immobilized artificial membrane (IAM)-HPLC, as compared to partitioning in the standardized phosphatidylcholine liposome/buffer system.

METHODS

The pH-dependent apparent partition coefficients D were calculated from capacity factors (k'IAM) obtained by IAM-HPLC, using a 11-carboxylundecylphosphocholine column. For lipophilic compounds k'IAM, values were determined with organic modifiers and extrapolation to 100% water phase (k'IAMw) was optimized. Temperature dependence was explored (23 to 45 degrees C), and Gibbs free energy (deltaG), partial molar enthalpy (deltaH) and change in entropy (deltaS) were calculated. Equilibrium dialysis was used for the partitioning studies with the liposome/buffer system.

RESULTS

For extrapolation of k'IAMw, linear plots were obtained both with the respective dielectric constants and the mole fractions of the organic modifier. All tested compounds showed a similar pH-D diagram in both systems; however, significant differences were reproducibly found in the pH range of 5 to 8. In all cases, deltaG and deltaH were negative, whereas deltaS values were negative for acids and positive for bases.

CONCLUSIONS

In both partitioning systems, D values decreased significantly with the change from the neutral to the charged ionization state of the solute. The differences found under physiological conditions, i.e. around pH 7.4, were attributed to nonspecific interactions of the drug with the silica surface of the IAM column.

Interactions of macrolide antibiotics (Erythromycin A, roxithromycin, erythromycylamine [Dirithromycin], and azithromycin) with phospholipids: computer-aided conformational analysis and studies on acellular and cell culture models

The potential of 14/15 membered macrolides to cause phospholipidosis has been prospectively assessed, and structure-effects examined, using combined experimental and conformational approaches. Biochemical studies demonstrated drug binding to phosphatidylinositol-containing liposomes and inhibition of the activity of lysosomal phospholipase A1 toward phosphatidylcholine included in the bilayer, in close correlation with the number of cationic groups carried by the drugs (erythromycin A </= roxithromycin < erythromycylamine </= azithromycin). In cultured cells (fibroblasts), phospholipidosis (affecting all major phospholipids except sphingomyelin) was observed after 3 days with the following ranking: erythromycin A </= roxithromycin < erythromycylamine < azithromycin (roxithromycin could, however, not be studied in detail due to intrinsic toxicity). The difference between erythromycylamine and azithromycin was accounted for by the lower cellular accumulation of erythromycylamine. In parallel, based on a methodology developed and validated to study drug-membrane interactions, the conformational analyses revealed that erythromycin A, roxithromycin, erythromycylamine, and azithromycin penetrate into the hydrophobic domain of a phosphatidylinositol monolayer through their desosamine and cladinose moieties, whereas their macrocycle is found close to the interface. This position allows the aminogroups carried by the macrocycle of the diaminated macrolides (erythromycylamine and azithromycin) to come into close contact with the negatively charged phosphogroup of phosphatidylinositol, whereas the amine located on the C-3 of the desosamine, common to all four drugs, is located at a greater distance from this phosphogroup. Our study suggests that all macrolides have the potential to cause phospholipidosis but that this effect is modulated by toxicodynamic and toxicokinetic parameters related to the drug structure and mainly to their cationic character.

Potential of immobilized artificial membranes for predicting drug penetration across the blood-brain barrier

PURPOSE

The present study evaluates immobilized artificial membrane (IAM) chromatography for predicting drug permeability across the blood-brain barrier (BBB) and outlines the potential and limitations of IAMs as a predictive tool by comparison with conventional methods based on octanol/water partitioning and octadecylsilane (ODS)-HPLC.

METHODS

IAM-and ODS-HPLC capacity factors were determined in order to derive the hydrophobic indices log kIAM nad log kW for two sets of compounds ranging from very lipid soluble (steroids) to more hydrophilic agents (biogenic amines). The uptake of the compounds across the in vivo BBB expressed as brain uptake index (BUI) has been correlated with these HPLC capacity factors as well as octanol/ water partition (ClogP) and distribution coefficients (log D7.4).

RESULTS

For both test groups log kIAM correlates significantly with the respective log BUI of the drug (r2 = 0.729 and 0.747, p < 0.05), whereas with log kW, log D7.4 and ClogP there is only a correlation for the group of steroids (r2 = 0.789, 0.659 and 0.809, p < 0.05) but not for the group of biogenic amines. There is a good correlation between log kIAM and log kW. ClogP or log D7.4 for the group of steroids (r2 = 0.945.0867 and 0.974, p < 0.01) but not for the biogenic amines.

CONCLUSIONS

All physico-chemical descriptors examined in this study equally well describe brain uptake of lipophilic compounds, while log kIAM is superior over log D7.4, ClogP and log kW when polar and ionizable compounds are included. The predictive value of IAMs, combined with the power of HPLC holds thus great promise for the selection process of drug candidates with high brain penetration.

The effect of ionic strength on liposome-buffer and 1-octanol-buffer distribution coefficients

The distribution of salmeterol and proxicromil between unilamellar vesicles of dioleoylphosphatidylcholine (DOPC) and aqueous buffer at pH 7.4 has been studied, using an ultrafiltration method, as a function of compound concentration, DOPC concentration, and buffer ionic strength. The binding of these ionized lipophilic compounds to neutral DOPC vesicles induces a surface charge, which causes the observed membrane distribution coefficient D(mem)obs to vary significantly with bound compound to DOPC ratio and with ionic strength. This variability is shown to be well-described with use of the Gouy-Chapman theory of the ionic double layer and is contrasted with the ideal behavior shown by the neutral compound clofibrate. Increasing ionic strength is also shown to increase the observed 1-octanol-buffer distribution coefficients D(o/w)obs of proxicromil but through a very different mechanism involving the extraction of ion pairs. This study highlights the experimental difficulty in determining concentration-independent liposome distribution coefficients of ionized lipophilic compounds and describes when deviations will be significant and how observed values may be corrected for such effects. The general effect of ionic strength on membrane-buffer distribution and 1-octanol-buffer distribution is discussed with particular reference to the very different propensity for ion pair formation shown by the two systems, and the most suitable experimental conditions that should be used with each system.

Towards the predictability of drug-lipid membrane interactions: the pH-dependent affinity of propanolol to phosphatidylinositol containing liposomes

PURPOSE

Prediction of the pH-dependent affinity of (RS)-[3H]propranolol to mixed phosphatidylcholine (PhC)/phosphatidylinositol(Phl) membranes from the partitioning in the single lipid liposome/buffer systems.

METHODS

Partition studies in liposome/buffer systems were performed by means of equilibrium dialysis at 37 degrees C between pH 2 and 11 at a molar propranolol to lipid ratio of 10(-6) to 10(-5) in the membrane. Results. The Phl membrane more strongly attracts the protonated (RS)-[3H]propranolol than the neutral solute, i.e. the partition coefficient of the protonated base (Pi) is 17'430+/-1320, P of the neutral compound (Pn) is 3110+/-1650. In the PhC-liposome system Pi is 580+/-17, Pn 1860+/-20. The partition coefficients show an exponential dependence on the molar Phl fraction in mixed liposomes. The partitioning in mixed PhC/Phl membranes is predictable from Pn and Pi in the single lipid liposome systems.

CONCLUSIONS

The negative charge of biological lipid membranes causes strong electrostatic interactions with positively charged solutes. This strong attraction is not predictable from the octanol/buffer partition system, but it is important regarding drug accumulation in the tissue and drug attraction by certain lipids in the vicinity of membrane proteins.

Free fatty acids cause pH-dependent changes in drug-lipid membrane interactions around physiological pH

PURPOSE

The influence of oleic acid (OA) and phosphatidylethanolamine (PhE) as membrane constituents on the partition behavior of (RS)-[3H]propranolol between unilamellar liposomes and buffer was studied as a function of pH.

METHODS

Partition studies were performed by means of equilibrium dialysis at 37 degrees C over a broad pH range at a molar propranolol to lipid ratio in the membrane of 10(-6).

RESULTS

As compared to the standard phosphatidylcholine (PhC)-liposome/buffer partition system PhE and OA have an enhancing effect on the apparent partition coefficient (D) of (RS)-[3H]propranolol between pH 6 and 11. Data analysis with Henderson-Hasselbalch equations revealed that the neutral propranolol has a higher affinity to membranes containing net neutral charged PhE than to pure PhC-liposomes. Net negatively charged PhE seems to have no significant influence on the partitioning. Deprotonated OA caused an increase in the true partition coefficient (P) of the protonated propranolol. Neutral OA showed no influence on the partitioning. From the fit D vs pH curves and from zeta potential measurements of the liposomes the intrinsic pKa values of the membranous lipids were calculated as 7.5 to 7.8 for OA and 9.7 to 9.8 for PhE.

CONCLUSIONS

Since the pKa of membranous OA is close to the physiological pH and D depends on the ionisation state of OA, small pH changes around the physiological pH may cause large differences in drug-lipid membrane interactions.