Prediction of drug absorption based on immobilized artificial membrane (IAM) chromatography separation and calculated molecular descriptors

Recent developments in the application of immobilized artificial membrane (IAM) chromatography to drug discovery

INTRODUCTION

Immobilized artificial membrane (IAM) chromatography is widely used in many aspects of drug discovery. It employs stationary phases, which contain phospholipids combining simulation of biological membranes with rapid measurements.

AREAS COVERED

Advances in IAM stationary phases, chromatographic conditions and the underlying retention mechanism are discussed. The potential of IAM chromatography to model permeability and drug-membrane interactions as well as its use to estimate pharmacokinetic properties and toxicity endpoints including ecotoxicity, is outlined. Efforts to construct models for prediction IAM retention factors are presented.

EXPERT OPINION

IAM chromatography, as a border case between partitioning and binding, has broadened its application from permeability studies to encompass processes involving tissue binding. Most IAM-based permeability models are hybrid models incorporating additional molecular descriptors, while for the estimation of pharmacokinetic properties and binding to off targets, IAM retention is combined with other biomimetic properties. However, for its integration into routine drug discovery protocols, reliable IAM prediction models implemented in relevant software should be developed, to enable its use in virtual screening and the design of new molecules. Conversely, preparation of new IAM columns with different phospholipids or mixed monomers offers enhanced flexibility and the potential to tailor the conditions according to the target property.

Modeling of small molecule's affinity to phospholipids using IAM-HPLC and QSRR approach enhanced by similarity-based machine algorithms

Immobilized artificial membrane chromatography (IAM) has been proposed as a more biosimilar alternative to classical lipophilicity measurement. Determination of small molecule's affinity to phospholipids can be supported for predicting their behavior in the human body. Therefore, a better understanding of the molecular interaction mechanism between small xenobiotics and phospholipids can accelerate drug discovery. Here, the quantitative structure-retention relationships (QSRR) approach was integrated with mechanistic descriptors calculated using Chemicalize software to propose an easy-to-interpretation QSRR model. Considering the heterogeneous character of the data set, locally weighted least squares kernel regression belonging to similarity-based machine learning methods have been applied. The results showed that lipophilicity, charge, and maximum projection area determine molecule binding to phospholipids. Full validation of the obtained model based on OECD recommendations has been performed and the applicability domain was defined using the probability-oriented distance-based approach. The high values of predictive squared correlation coefficient (Q2), and small root mean square error of prediction (RMSEP), 0.812 and 6.739, respectively, confirmed that the obtained QSRR model is not well-fitted to the training data but also showed prediction power. Additionally, only 1.5% of molecules from the training set and 2.8% from the validation test are outside the applicability domain, confirming great predictive abilities.

Permeability Data of Organosulfur Garlic Compounds Estimated by Immobilized Artificial Membrane Chromatography: Correlation Across Several Biological Barriers

Among healthy vegetables, those of the genus Allium stand out. Antioxidant and anti-inflammatory properties have been associated with these vegetables, attributed mainly to organosulfur compounds (OSCs). In turn, they are linked to a protective effect counteracting cardiovascular disease development. Now, to really ensure the bioactive efficacy of the said compounds once consumed, it is necessary to previously evaluate the ADME (absorption, distribution, metabolism, and excretion) profile. Alternatively, in vitro and in silico methods attempt to avoid or reduce experimental animals' use and provide preliminary information on drugs' ability to overcome the various biological barriers inherent in the ADME process. In this sense, in silico methods serve to provide primary information on drugs' bioavailability mechanisms. High-performance liquid chromatography (HPLC) using a stationary phase composed of phospholipids, the so-called immobilized artificial membrane (IAM), has been widely recognized as a valuable alternative method to extract and quantify information about the structure and physicochemical properties of organic compounds which are extensively used in studies of quantitative structure-activity relationships (QSARs). In the present study, the chromatographic capacity factors (log k' (IAM)) for 28 OSCs were determined by IAM-HPLC. In order to evaluate the ability of the IAM phase in assessing lipophilicity of the compounds under study, several quantitative structure-retention relationships (QSRRs) were derived from exploring fundamental intermolecular interactions that govern the retention of compounds under study on IAM phases. As expected, the hydrophobic factors are of prime importance for the IAM retention of these compounds. However, the molecular flexibility and specific polar interactions expressed by several electronic descriptors (relative negative charge, RNCG, and Mulliken electronegativity) are also involved. We also evaluated the IAM phase ability to assess several ADME parameters for the OSCs under study obtained using the SwissADME web tool integrated into the SwissDrugDesign workspace and the PreADMET web tool. The human gastrointestinal absorption (HIA), blood-brain barrier (BBB) permeation, and skin permeability were investigated through QSAR modeling, using several chemometric approaches. The ADME properties under study are strongly dependent on hydrophobic factors as expressed by log k'(IAM), which provide evidence for the great potential of the IAM phases in the development of QSAR models.

Phosphatidylethanolamine functionalized biomimetic monolith for immobilized artificial membrane chromatography

In this research, a new phospholipid based monolith was fabricated by in situ co-polymerization of 1-dodecanoyl-2-(11-methacrylamidoundecanoyl)-sn-glycero-3-phosphoethanolamine and ethylene dimethacrylate to mimick bio-membrane environment. Excellent physicochemical properties of this novel monolith that were achieved included column efficiency, stability, and permeability. Moreover, the biomimetic monolith showed outstanding separation capability for a series of intact proteins and small molecules. In particular, it exhibited good potential as an alternative to the commercial immobilized artificial membrane (IAM) column (IAM.PC.DD2) for studying drug-membrane interactions. This study not only enriched the types of IAM stationary phases, but also provided a simple model for the prediction of phosphatidylethanolamine related properties of drug candidates.

•

A novel phosphatidylethanolamine (PE) functionalized immobilized artificial membrane (IAM) monolithic column was developed.

•

The composition of the polymerization mixtures used for the preparation of poly(MDSPE-co-EDMA) monolithic columns was optimized for micro-LC applications.

•

The biomimetic monolith exhibited good separation selectivity for both intact proteins and small molecules.

•

The biomimetic monolith exhibited great potential as a replacement of commercial IAM columns for studying drug-membrane interactions.

In Silico Prediction of PAMPA Effective Permeability Using a Two-QSAR Approach

Oral administration is the preferred and predominant route of choice for medication. As such, drug absorption is one of critical drug metabolism and pharmacokinetics (DM/PK) parameters that should be taken into consideration in the process of drug discovery and development. The cell-free in vitro parallel artificial membrane permeability assay (PAMPA) has been adopted as the primary screening to assess the passive diffusion of compounds in the practical applications. A classical quantitative structure–activity relationship (QSAR) model and a machine learning (ML)-based QSAR model were derived using the partial least square (PLS) scheme and hierarchical support vector regression (HSVR) scheme to elucidate the underlying passive diffusion mechanism and to predict the PAMPA effective permeability, respectively, in this study. It was observed that HSVR executed better than PLS as manifested by the predictions of the samples in the training set, test set, and outlier set as well as various statistical assessments. When applied to the mock test, which was designated to mimic real challenges, HSVR also showed better predictive performance. PLS, conversely, cannot cover some mechanistically interpretable relationships between descriptors and permeability. Accordingly, the synergy of predictive HSVR and interpretable PLS models can be greatly useful in facilitating drug discovery and development by predicting passive diffusion.

Lipophilicity of porphyrins and their retention in IAM, C8-C18 and HILIC chromatographic systems

Porphyrins are a class of photosensitizers used in photodynamic therapy (PDT). Understanding the interaction of porphyrins with membrane cells components is important in order to improve this therapy. Many analytical methods can be used for this aim. High performance liquid chromatography (HPLC) was used for the separation of porphyrins on RP and HILIC stationary phases as well as on a biomimetic membrane IAM phase. Twenty-six tetraphenyl porphyrins (TPP) were successfully separated on an IAM column, a C18 Gravity RP column, a C8 Gravity RP column, a PolarTec RP column and a HILIC column. Stationary phases were chosen as the most appropriate to cover the study of different types of interactions. Elution was performed with a 45 min linear gradient. Obtained gradient retention times were converted to gradient chromatography hydrophobicity index (CHI) and to an apparent retention factor (kapp). The partition coefficients (logP) of the 26 compounds were measured in a 2-octanol/PBS system and estimated in silico. Correlation between kapp values was studied. Moreover, a multivariate analysis was performed to explain columns relationships. Obtained results show that porphyrins are separated mainly according to hydrophobic interactions that are relative to their structure (sugar number and the disposition around the porphyrin macrocycle).

Modeling the drugs' passive transfer in the body based on their chromatographic behavior

One of the most challenging aims in modern analytical chemistry and pharmaceutical analysis is to create models for drugs' behavior based on simulation experiments. Since drugs' effects are closely related to their molecular properties, numerous characteristics of drugs are used in order to acquire a model of passive absorption and transfer in the human body. Importantly, such direction in innovative bioanalytical methodologies is also of stressful need in the area of personalized medicine to implement nanotechnological and genomics advancements. Simulation experiments were carried out by examining and interpreting the chromatographic behavior of 113 analytes/drugs (400 observations) in RP-HPLC. The dataset employed for this purpose included 73 descriptors which are referring to the physicochemical properties of the mobile phase mixture in different proportions, the physicochemical properties of the analytes and the structural characteristics of their molecules. A series of different software packages was used to calculate all the descriptors apart from those referring to the structure of analytes. The correlation of the descriptors with the retention time of the analytes eluted from a C4 column with an aqueous mobile phase was employed as dataset to introduce the behavior models in the human body. Their evaluation with a Partial Least Squares (PLS) software proved that the chromatographic behavior of a drug on a lipophilic stationary and a polar mobile phase is directly related to its drug-ability. At the same time, the behavior of an unknown drug in the human body can be predicted with reliability via the Artificial Neural Networks (ANNs) software.

Keeping a critical eye on the science and the regulation of oral drug absorption: a review

This review starts with an introduction on the theoretical aspects of biopharmaceutics and developments in this field from mid-1950s to late 1970s. It critically addresses issues related to fundamental processes in oral drug absorption such as the complex interplay between drugs and the gastrointestinal system. Special emphasis is placed on drug dissolution and permeability phenomena as well as on the mathematical modeling of oral drug absorption. The review ends with regulatory aspects of oral drug absorption focusing on bioequivalence studies and the US Food and Drug Administration and European Medicines Agency guidelines dealing with Biopharmaceutics Classification System and Biopharmaceutic Drug Disposition Classification System.

Rank-based ant system method for non-linear QSPR analysis: QSPR studies of the solubility parameter

The solubility parameter (δ) plays a unique role in the development of stable pharmaceutical formulations for assessing phase segregation during product synthesis. Understanding this parameter helps to determine how a drug substance will behave when processed or when dosed in vivo. The aim of this work was to develop a novel comprehensive yet rapid and accurate Quantitative Structure-Property Relationship (QSPR) method based on the rank-based ant system feature selection. The method was coupled with the multiple linear regression and support vector regression and applied to the assessment of solubility parameters for a diverse dataset of 1804 chemical compounds. The models were validated by solubility prediction of 360 test set compounds which were not used in building models. The developed models have high prediction power characterized by r (2) values 0.75 and 0.82, and RMSE values 1.96 and 1.65 (J/(cm(3)))(0.5) for the external test set. Various validation techniques and comparison results with the novel optimized support vector regression indicate that the developed models can be used to determine the solubility parameters for a diverse set of chemicals with an acceptable accuracy. The developed models can be beneficial for designing new chemical materials with desired solubility parameter values.

Database of published retention factors for immobilized artificial membrane HPLC and an assessment of the effect of experimental variability

A database was collated of published experimental logarithmic values for the relative retention factors (log k(IAM)) measured using an immobilized artificial membrane column and high-performance liquid chromatography (IAM HPLC). Log k(IAM) is an alternative measure of hydrophobicity to the octanol/water partition coefficient (log K(OW)). While there are several accepted methods to measure log K(OW), no standardized method exists to determine log k(IAM). The database of collated log k(IAM) values includes 13 key experimental parameters and contains 1,686 values for 555 compounds, which are predominantly polar organic compounds and include drug molecules and surfactants. These compounds are acidic, basic, and neutral and both ionized and un-ionized under the conditions of analysis. The data compiled demonstrated experimental variability for each experimental parameter considered, including column stationary phase, pH, temperature, and mobile phase. Reducing the experimental variability allowed for greater consistency in the datasets.

Engineering and validation of a novel lipid thin film for biomembrane modeling in lipophilicity determination of drugs and xenobiotics

Background

Determination of lipophilicity as a tool for predicting pharmacokinetic molecular behavior is limited by the predictive power of available experimental models of the biomembrane. There is current interest, therefore, in models that accurately simulate the biomembrane structure and function. A novel bio-device; a lipid thin film, was engineered as an alternative approach to the previous use of hydrocarbon thin films in biomembrane modeling.

Results

Retention behavior of four structurally diverse model compounds; 4-amino-3,5-dinitrobenzoic acid (ADBA), naproxen (NPX), nabumetone (NBT) and halofantrine (HF), representing 4 broad classes of varying molecular polarities and aqueous solubility behavior, was investigated on the lipid film, liquid paraffin, and octadecylsilane layers. Computational, thermodynamic and image analysis confirms the peculiar amphiphilic configuration of the lipid film. Effect of solute-type, layer-type and variables interactions on retention behavior was delineated by 2-way analysis of variance (ANOVA) and quantitative structure property relationships (QSPR). Validation of the lipid film was implemented by statistical correlation of a unique chromatographic metric with Log P (octanol/water) and several calculated molecular descriptors of bulk and solubility properties.

Conclusion

The lipid film signifies a biomimetic artificial biological interface capable of both hydrophobic and specific electrostatic interactions. It captures the hydrophilic-lipophilic balance (HLB) in the determination of lipophilicity of molecules unlike the pure hydrocarbon film of the prior art. The potentials and performance of the bio-device gives the promise of its utility as a predictive analytic tool for early-stage drug discovery science.

A molecular chromatographic approach to analyze the cell diffusion of arginase inhibitors

Our group demonstrated that arginase inhibition reduces endothelial dysfunction in spontaneously hypertensive rats [C. Demougeot, A. Prigent-Tessier, C. Marie, A. Berthelot, J. Hypertens. 23 (2005) 971; C. Demougeot, A. Prigent-Tessier, T. Bagnost, C. Andre, Y. Guillaume, M. Bouhaddi, C. Marie, A. Berthelot, Life Sci. 80 (2007) 1128] which opens perspectives in the development of drugs against hypertension. In previous papers [T. Bagnost, Y.C. Guillaume, M. Thomassin, J.F. Robert, A. Berthelot, A. Xicluna, C. Andre, J. Chromatogr. B: Analyt. Technol. Biomed. Life Sci. 856 (2007) 113; T. Bagnost, Y.C. Guillaume, M. Thomassin, A. Berthelot, C. Demougeot, C. Andre, J. Chromatogr. B: Analyt. Technol. Biomed. Life Sci. 873 (2008) 37], we developed a biochromatographic column for studying the binding of an arginase inhibitor with this enzyme and the effect of magnesium on this binding. In this paper, the interaction of arginase inhibitors with an immobilized artificial membrane (IAM) has been studied using a biochromatographic approach. This IAM provided a biophysical model system to study the inhibitor passive transport across cells. It was demonstrated that more the inhibitor cross the cell membrane by passive diffusion more it is potent. As well, an analysis of the thermodynamics of the interaction of the arginase inhibitors with the IAM showed that van der Waals, hydrogen and ionic bonds were the main forces between the arginase inhibitors and the polar head groups of the IAM surface.

Structure-activity models of oral clearance, cytotoxicity, and LD50: a screen for promising anticancer compounds

BACKGROUND

Quantitative structure-activity relationship (QSAR) models have become popular tools to help identify promising lead compounds in anticancer drug development. Few QSAR studies have investigated multitask learning, however. Multitask learning is an approach that allows distinct but related data sets to be used in training. In this paper, a suite of three QSAR models is developed to identify compounds that are likely to (a) exhibit cytotoxic behavior against cancer cells, (b) exhibit high rat LD50 values (low systemic toxicity), and (c) exhibit low to modest human oral clearance (favorable pharmacokinetic characteristics). Models were constructed using Kernel Multitask Latent Analysis (KMLA), an approach that can effectively handle a large number of correlated data features, nonlinear relationships between features and responses, and multitask learning. Multitask learning is particularly useful when the number of available training records is small relative to the number of features, as was the case with the oral clearance data.

RESULTS

Multitask learning modestly but significantly improved the classification precision for the oral clearance model. For the cytotoxicity model, which was constructed using a large number of records, multitask learning did not affect precision but did reduce computation time. The models developed here were used to predict activities for 115,000 natural compounds. Hundreds of natural compounds, particularly in the anthraquinone and flavonoids groups, were predicted to be cytotoxic, have high LD50 values, and have low to moderate oral clearance.

CONCLUSION

Multitask learning can be useful in some QSAR models. A suite of QSAR models was constructed and used to screen a large drug library for compounds likely to be cytotoxic to multiple cancer cell lines in vitro, have low systemic toxicity in rats, and have favorable pharmacokinetic properties in humans.

Determination of liposome permeability of ionizable carbamates of zidovudine by steady state fluorescence spectroscopy

In the present paper the relative permeabilities of AZT-Pyp and AZT-Ethy across a phospholipid bilayer were estimated by the means of fluorescence spectroscopy. The center of spectral mass of both non-encapsulated AZT-derivatives (AZT-der) emission spectra increased as a function of the illumination time inside the spectrofluorimeter cell. This phenomenon was even more evident when drugs were incubated under an UV mercury lamp, suggesting its photolytic origin. AZT-der were protected from photolysis inside liposomes and decomposed upon irradiation when they were free in the aqueous phase. The time-dependent decrease in the fluorescence intensity at a constant wavelength was fitted to a two-exponential equation and the values of rate constants for permeability and photolysis were calculated. It was concluded that AZT-Pyp but not AZT-Ethy diffused across the bilayer. This behavior correlated with the molecular volumes of AZT-Pyp (379.6A(3)) and AZT-Ethy (450.5A(3)), determined from the minimum energy conformations but not with previously reported logP values. These results reinforce the concept that not only lipophilicity but also membrane structure and AZT-der molecular size had a critical influence in passive diffusion across bilayers and may help in future refinements of other AZT-der molecular design.

Evaluation of electronic, lipophilic and membrane affinity effects on antiproliferative activity of 5-substituted-2-(2,4-dihydroxyphenyl)-1,3,4-thiadiazoles against various human cancer cells

The QSAR studies of 5-substituted-2-(2,4-dihydroxyphenyl)-1,3,4-thiadiazoles set of antiproliferative activity against human cancer cell lines have been performed. Electronic properties of compounds were estimated by the Hartree-Fock method at 6-31G** level. Lipophilicity and membrane affinity parameters were determined by the chromatographic methods RP-8 OPLC and IAM HPLC, respectively. Mono- and multivariable regression analyses were performed. The principle factor for determination of activity of compounds is partial charge of nitrogen (q(N3), q(N4)) and carbon (q(C5)) atoms of the 1,3,4-thiadiazole ring. Biological effect is also connected with molar refractivity (CMR) and lipophilicity of derivatives obtained by RP-8 chromatography. The analysis of the QSAR equations for individual cell lines indicates both similarities and differences of electron, steric factors and hydrophobic-hydrophilic character of the analogues of the tested set affecting the antiproliferative activity.

Evaluation of chromatographic descriptors for the prediction of gastro-intestinal absorption of drugs

The use of chromatographic descriptors in QSAR was evaluated. Therefore, retentions were measured on an immobilized artificial membrane system, 2 micellar liquid chromatography systems and 17 orthogonal or disimilar reversed-phase liquid chromatographic systems. It was investigated whether it was possible to model gastro-intestinal absorption as a function of chromatographic retentions applying two linear and one non-linear multivariate modeling technique. In a second step it was evaluated if models built with theoretical descriptors could be improved by adding the measured retention factors to the data set of descriptive variables. It was seen that gastro-intestinal absorption could be modelled in function of chromatographic retention using the non-linear modeling technique multivariate adaptive regression splines (MARS). The best models were obtained using a combination of theoretical and chromatographic descriptors with MARS as modeling technique.

Quantitative structure-retention relationship studies with immobilized artificial membrane chromatography II: partial least squares regression

We aimed to establish quantitative structure-retention relationship (QSRR) with immobilized artificial membrane (IAM) chromatography using easily understood and obtained physicochemical molecular descriptors and to elucidate which descriptors are critical to affect the interaction process between solutes and immobilized phospholipid membranes. The retention indices (logk(IAM)) of 55 structurally diverse drugs were determined on an immobilized artificial membrane column (IAM.PC.DD2) directly or obtained by extrapolation method for highly hydrophobic compounds. Ten simple physicochemical property descriptors (clogP, rings, rotatory bond, hydro-bond counting, etc.) of these drugs were collected and used to establish QSRR and predict the retention data by partial least squares regression (PLSR). Five descriptors, clogP, rotatory bond (RotB), rings, molecular weight (MW) and total surface area (TSA), were reserved by using the Variable Importance for Projection (VIP) values as criterion to build the final PLSR model. An external test set was employed to verify the QSRR based on the training set with the five variables, and QSRR by PLSR exhibited a satisfying predictive ability with R(p)=0.902 and RMSE(p)=0.400. Comparison of coefficients of centered and scaled variables by PLSR demonstrated that, for the descriptors studied, clogP and TSA have the most significant positive effect but the rotatable bond has significant negative effect on drug IAM chromatographic retention.

Pharmacophore-based discovery of ligands for drug transporters

The ability to identify ligands for drug transporters is an important step in drug discovery and development. It can both improve accurate profiling of lead pharmacokinetic properties and assist in the discovery of new chemical entities targeting transporters. In silico approaches, especially pharmacophore-based database screening methods have great potential in improving the throughput of current transporter ligand identification assays, leading to a higher hit rate by focusing in vitro testing to the most promising hits. In this review, the potential of different in silico methods in transporter ligand identification studies are compared and summarized with an emphasis on pharmacophore modeling. Various implementations of pharmacophore model generation, database compilation and flexible screening algorithms are also introduced. Recent successful utilization of database searching with pharmacophores to identify novel ligands for the pharmaceutically significant transporters hPepT1, P-gp, BCRP, MRP1 and DAT are reviewed and the challenges encountered with current approaches are discussed.

Prediction of solubility parameters using partial least square regression

The total solubility parameter (delta) values were effectively predicted by using computed molecular descriptors and multivariate partial least squares (PLS) statistics. The molecular descriptors in the derived models included heat of formation, dipole moment, molar refractivity, solvent-accessible surface area (SA), surface-bounded molecular volume (SV), unsaturated index (Ui), and hydrophilic index (Hy). The values of these descriptors were computed by the use of HyperChem 7.5, QSPR Properties module in HyperChem 7.5, and Dragon Web version. The other two descriptors, hydrogen bonding donor (HD), and hydrogen bond-forming ability (HB) were also included in the models. The final reduced model of the whole data set had R(2) of 0.853, Q(2) of 0.813, root mean squared error from the cross-validation of the training set (RMSEcv(tr)) of 2.096 and RMSE of calibration (RMSE(tr)) of 1.857. No outlier was observed from this data set of 51 diverse compounds. Additionally, the predictive power of the developed model was comparable to the well recognized systems of Hansen, van Krevelen and Hoftyzer, and Hoy.

Relationship between immobilized artificial membrane chromatographic retention and human oral absorption of structurally diverse drugs

Capacity factors are determined for a set of drugs for which human oral absorption (HOA) data are available, using immobilized artificial membrane (IAM) chromatography. The compound set represented acidic, basic, neutral and amphoteric drugs from various structure classes and having low to high human oral absorption. Effect of mobile phase pH on retention was investigated to determine the optimal condition for better correlation with HOA. The retention (capacity factor, k'(IAM) of each drug was measured by reverse phase HPLC using an IAM.PC.DD2 (1 cm x 3 mm i.d., 12 microm) column with an eluent of acetonitrile - 0.01 M phosphate buffer at pH 4.5-7.4. The pH dependent k'(IAM) was in accordance with pH partition theory. Using non-linear regression analysis the obtained log k'(IAM) values were compared with published data on HOA in order to establish correlation. The better correlation with HOA was observed when the highest log k'(IAM) value selected among pH 4.5-7.4 (R(2)=0.8566) for each drug rather than obtained at more traditional pH 7.4 (R(2)=0.7403). Finally, it was confirmed by Cook's D outlier test that there was no influential observation in the model that affect the relationship between IAM capacity factor and HOA. The assay conditions were optimized and validated to make it suitable for routine analysis and for compound characterization in early discovery where permeability may be an issue.