Toward predictive permeabilities: Experimental measurements and multiscale simulation of methanol transport in Nafion

A polymer membrane's permeability to solutes determines its suitability for various applications: a permeability value is essential for predicting performance in diverse contexts. Using aqueous methanol permeation through Nafion as an example, we describe a methodology for determining membrane permeability that accounts for boundary layer effects and the possibility of swelling. For the materials and apparatus used herein, analysis of a permeance measurement and computational fluid dynamics simulations show that the mass transfer boundary layer is on the order of ones to tens of microns. The data are used to develop and validate a multiscale model describing solute permeation through a hydrated membrane as a series of physical mechanistic steps: reversible adsorption from solution at the membrane interface, diffusion driven by a concentration gradient within the membrane, and reversible desorption into solution at the opposite membrane interface. The validated model is used to predict methanol transport across a solar-driven CO₂ reduction device and to assess the impact of polymer changes on the measured value. The approach of combining experimental data, computational fluid dynamics, and the mechanistic multiscale model is expected to provide more accurate analysis of membrane permeation data in cases with polymer swelling or unusual device geometries, among others.     

Permeation of CO2 and N2 through glassy poly(dimethyl phenylene) oxide under steady- and presteady-state conditions

Glassy polymers are often used for gas separations because of their high selectivity. Although the dual-mode permeation model correctly fits their sorption and permeation isotherms, its physical interpretation is disputed, and it does not describe permeation far from steady state, a condition expected when separations involve intermittent renewable energy sources. To develop a more comprehensive permeation model, we combine experiment, molecular dynamics, and multiscale reaction–diffusion modeling to characterize the time-dependent permeation of N₂ and CO₂ through a glassy poly(dimethyl phenylene oxide) membrane, a model system. Simulations of experimental time-dependent permeation data for both gases in the presteady-state and steady-state regimes show that both single- and dual-mode reaction–diffusion models reproduce the experimental observations, and that sorbed gas concentrations lag the external pressure rise. The results point to environment-sensitive diffusion coefficients as a vital characteristic of transport in glassy polymers.     

Rationalizing the generation of broad spectrum antibiotics with the addition of a positive charge

Antibiotic resistance of Gram-negative bacteria is largely attributed to the low permeability of their outer membrane (OM). Recently, we disclosed the eNTRy rules, a key lesson of which is that the introduction of a primary amine enhances OM permeation in certain contexts. To understand the molecular basis for this finding, we perform an extensive set of molecular dynamics (MD) simulations and free energy calculations comparing the permeation of aminated and amine-free antibiotic derivatives through the most abundant OM porin of E. coli, OmpF. To improve sampling of conformationally flexible drugs in MD simulations, we developed a novel, Monte Carlo and graph theory based algorithm to probe more efficiently the rotational and translational degrees of freedom visited during the permeation of the antibiotic molecule through OmpF. The resulting pathways were then used for free-energy calculations, revealing a lower barrier against the permeation of the aminated compound, substantiating its greater OM permeability. Further analysis revealed that the amine facilitates permeation by enabling the antibiotic to align its dipole to the luminal electric field of the porin and form favorable electrostatic interactions with specific, highly-conserved charged residues. The importance of these interactions in permeation was further validated with experimental mutagenesis and whole cell accumulation assays. Overall, this study provides insights on the importance of the primary amine for antibiotic permeation into Gram-negative pathogens that could help the design of future antibiotics. We also offer a new computational approach for calculating free-energy of processes where relevant molecular conformations cannot be efficiently captured.

A rapid pathway sampling method combining Monte Carlo and graph theory, developed to describe permeation pathways through outer membrane porins, can distinguish between structurally similar analogs with different permeabilities.     

An atomistic model of passive membrane permeability: application to a series of FDA approved drugs

We apply an atomistic model of passive membrane permeability to a series of weakly basic drugs. The computational model uses conformational sampling in combination with an all-atom force field and implicit solvent model to estimate relative passive membrane permeabilities. The model does not require the use of training data for rank-ordering compounds, and as such represents a different approach from the more commonly employed QSPR models. We compare the computational results to previously published experimental PAMPA and Caco-2 permeabilities.     

Predicting Caco-2 cell permeation coefficients of organic molecules using membrane-interaction QSAR analysis

A methodology termed membrane-interaction QSAR (MI-QSAR) analysis has been developed in order to predict the behavior of organic compounds interacting with the phospholipid-rich regions of biological membranes. One important application of MI-QSAR analysis is to estimate ADME properties including the transport of organic solutes through biological membranes as a computational approach to forecasting drug intestinal absorption. A training set of 30 structurally diverse drugs, whose permeability coefficients across the cellular membranes of Caco-2 cells were measured, was used to construct significant MI-QSAR models of Caco-2 cell permeation. Cellular permeation is found to depend primarily upon aqueous solvation free energy (solubility) of the drug, the extent of drug interaction with a model phospholipid (DMPC) monolayer, and the conformational flexibility of the solute within the model membrane. A test set of eight drugs was used to evaluate the predictivity of the MI-QSAR models. The permeation coefficients of the test set compounds were predicted with the same accuracy as the compounds of the training set.     

Adsorbent-filled polymeric membranes

Assuming three-dimensional isotropy for the suspended particles of cubes or spheres in a polymer matrix, resistance models are developed for adsorbent-filled polymer membranes. These are compared with the existing models in the literature and they agree very well. Monotonically increasing or decreasing permeabilities of composite membrane are well described by these models. The model calculations are compared with several experimental sets of data for alcohols and gases in the literature and proven to be very good. However, there are unexplained experimental observations of permeability behavior as the amount of fillers increase in polymer. Further studies are recommended to directly measure the zeolite permeabilities by measuring the actual throughput. This information is vital in interpreting the data of permeations and separations through filled membranes.     

Comparison of charge models for fixed-charge force fields: small-molecule hydration free energies in explicit solvent

In molecular simulations with fixed-charge force fields, the choice of partial atomic charges influences numerous computed physical properties, including binding free energies. Many molecular mechanics force fields specify how nonbonded parameters should be determined, but various choices are often available for how these charges are to be determined for arbitrary small molecules. Here, we compute hydration free energies for a set of 44 small, neutral molecules in two different explicit water models (TIP3P and TIP4P-Ew) to examine the influence of charge model on agreement with experiment. Using the AMBER GAFF force field for nonbonded parameters, we test several different methods for obtaining partial atomic charges, including two fast methods exploiting semiempirical quantum calculations and methods deriving charges from the electrostatic potentials computed with several different levels of ab initio quantum calculations with and without a continuum reaction field treatment of solvent. We find that the best charge sets give a root-mean-square error from experiment of roughly 1 kcal/mol. Surprisingly, agreement with experimental hydration free energies does not increase substantially with increasing level of quantum theory, even when the quantum calculations are performed with a reaction field treatment to better model the aqueous phase. We also find that the semiempirical AM1-BCC method for computing charges works almost as well as any of the more computationally expensive ab initio methods and that the root-mean-square error reported here is similar to that for implicit solvent models reported in the literature. Further, we find that the discrepancy with experimental hydration free energies grows substantially with the polarity of the compound, as does its variation across theory levels.     

Conformational flexibility, internal hydrogen bonding, and passive membrane permeability: successful in silico prediction of the relative permeabilities of cyclic peptides

We report an atomistic physical model for the passive membrane permeability of cyclic peptides. The computational modeling was performed in advance of the experiments and did not involve the use of "training data". The model explicitly treats the conformational flexibility of the peptides by extensive conformational sampling in low (membrane) and high (water) dielectric environments. The passive membrane permeabilities of 11 cyclic peptides were obtained experimentally using a parallel artificial membrane permeability assay (PAMPA) and showed a linear correlation with the computational results with R(2) = 0.96. In general, the results support the hypothesis, already well established in the literature, that the ability to form internal hydrogen bonds is critical for passive membrane permeability and can be the distinguishing factor among closely related compounds, such as those studied here. However, we have found that the number of internal hydrogen bonds that can form in the membrane and the solvent-exposed polar surface area correlate more poorly with PAMPA permeability than our model, which quantitatively estimates the solvation free energy losses upon moving from high-dielectric water to the low-dielectric interior of a membrane.     

Comparison of permeability coefficients of organic vapors through non-porous polymer membranes by two different experimental techniques

The permeability coefficients of saturated and non-saturated vapors of benzene, hexane and cyclohexane through flat polymer membranes (low density polyethylene BRALEN FB2-30 and polyether-block-amide PEBA 4033-PE) by two different experimental techniques at 298.15 K are reported. The permeation data have been obtained using the differential flow permeameter and sorption ones by glass sorption apparatus with McBain’s spiral balance. The so-called stationary (steady) diffusion theory has been applied for evaluating the permeability coefficients from sorption (equilibrium) data and obtained values have been compared with the permeability coefficients from permeation (steady-state) measurements. In the case of relative lower vapors sorption in polymers (hexane and cyclohexane) good agreement between permeability coefficients from sorption and permeation is obtained. Hence, this paper proves the possibility to estimate the permeability coefficients of organic vapors from sorption data without need of performing the permeation experiments.     

Improvement of "hit-to-lead" optimization by integration of in vitro HTS experimental models for early determination of pharmacokinetic properties

Development of predictive in vitro surrogate methods for traditional approaches assessing bioavailability and pharmacokinetics of lead compounds must be made to both keep pace with high-throughput (HT) lead identification and to mitigate the high costs associated with progression of compounds with poor chances of developmental success. Indeed opportunities for improvement still exist in the lead optimization phase versus the lead identification phase, where HT methodologies have been nearly optimized. Review of examples, limitations, and development of high-throughput microtiterplate-based assays for evaluating metabolic liabilities, such as in vitro radiometric and fluorometric assays for inhibition of cytochrome p450 (CYP) activity, determination of stability of a compound in liver microsomes, or cloned CYPs coupled to reconstituting systems are described. Parallel approaches to improve speed, resolution, sample preparation, as well as data analysis using LC/MS and LC/MS/MS approaches and technologies to assess compound integrity and biotransformation by automation and multiplexing are also discussed. Realization of the benefits in automation of cell-based models for determining drug permeability to predict drug absorption are still hampered by bottlenecks in analytical analysis of compounds. The implementation and limitations of surrogate physiochemical methods for passive adsorption such as immobilized artificial membranes (IAM) and parallel artificial membrane permeation assays (PAMPA), and compound solubility by laser nephelometry are reviewed as well. Additionally, data from a high-throughput 96-well equilibrium dialysis device, showing good correlation to classical methods, is presented. Finally, the impact of improvements in these downstream bottlenecks in lead optimization and preclinical drug discovery are discussed in this review.