Solute Descriptors Research Articles

Applications of polyparameter linear free energy relationships in environmental chemistry.

Partitioning behavior of organic chemicals has tremendous influences on their environmental distribution, reaction rates, bioaccumulation, and toxic effects. Polyparameter linear free energy relationships (PP-LFERs) have been proven to be useful to characterize the equilibrium partitioning of organic chemicals in various environmental and technical partitioning systems and predict the respective partition coefficients. Over the past decade, PP-LFER solute descriptors for numerous environmentally relevant organic chemicals and system parameters for environmentally important partitioning systems have been determined, extending substantially the applicability of the PP-LFER approaches. However, the information needed for the use of PP-LFERs including descriptors and parameters is scattered over a large number of publications. In this work, we review the state of the art of the PP-LFER approaches in environmental chemical applications. The solute descriptors and system parameters reported in the literature and the availability of their database are summarized, and their calibration and prediction methods are overviewed. We also describe tips and pitfalls associated with the use of the PP-LFER approaches and identify research needs to improve further the usefulness of PP-LFERs for environmental chemistry.

Ionic liquids as stationary phases in gas chromatography--an LSER investigation of six commercial phases and some applications.

The separation properties of six novel stationary phases for gas chromatography, commercially available from Sigma-Aldrich (Supelco) and based on ionic liquids (ILs), were investigated. The linear solvation energy relationship model (LSER) was used to describe the molecular interactions between these stationary phases and 30 solutes. The solutes belong to different groups of compounds, like haloalkanes, alcohols, ketones, aromatics, aliphatics, and others. A good description of different interactions, as described by the LSER model, could be achieved. The calculated values of system constants for the ionic liquid phases were compared with constants of commonly used standard phases like a 5% phenyl/95% dimethyl siloxane and a polyethylene glycol phase. The solute descriptors are in good agreement with those found by previous authors who have used the LSER model for 44 different ionic liquids as stationary phase. The experiments were carried out at two temperatures to evaluate the influence on the phase parameters and separation characteristics. The interactions of different functional groups with the IL phases are discussed. These novel IL phases are a promising replacement of or an addition to common polar phases. Based on the evaluated phase properties, several possibilities for applications of these novel phases are shown.

Cross-column prediction of gas-chromatographic retention indices of saturated esters

We combine computational molecular descriptors and variables related with the gas-chromatographic stationary phase into a comprehensive model able to predict the retention of solutes in external columns. To explore the quality of various approaches based on alternative column descriptors, we analyse the Kováts retention indices (RIs) of 90 saturated esters collected with seven columns of different polarity (SE-30, OV-7, DC-710, OV-25, XE-60, OV-225 and Silar-5CP). Cross-column retention prediction is evaluated on an internal validation set consisting of data of 40 selected esters collected with each of the seven columns, sequentially excluded from calibration. The molecular descriptors are identified by a genetic algorithm variable selection method applied to a large set of non-empirical structural quantities aimed at finding the best multi-linear quantitative structure–retention relationship (QSRR) for the column OV-25 having intermediate polarity. To describe the columns, we consider the sum of the first five McReynolds phase constants and, alternatively, the coefficients of the corresponding QSRRs. Moreover, the mean RI value for the subset of esters used in QSRR calibration or RIs of a few selected compounds are used as column descriptors. For each combination of solute and column descriptors, the retention model is generated both by multi-linear regression and artificial neural network regression.

Partitioning of polychlorinated biphenyls into human cells and adipose tissues: evaluation of octanol, triolein, and liposomes as surrogates.

Whereas octanol, triacylglycerides, and liposomes have all been proposed as surrogates for measuring the affinity of hydrophobic organic contaminants to human lipids, no comparative evaluation of their suitability exists. Here we conducted batch sorption experiments with polyoxymethylene passive samplers to determine the partition coefficients at 37 °C of 18 polychlorinated biphenyls (PCBs) from water into (i) triolein (Ktriolein/water), (ii) eight types of liposomes (Kliposome/water), (iii) human abdominal fat tissues (KAFT/water) from seven individuals, and (iv) human MCF-7 cells cultured in vitro (Kcell/water). Differences between KAFT/water among individuals and between Kliposome/water among liposome types were very small and not correlated to structural attributes of the PCBs. Similarly, the length and degree of saturation of the phospholipid carbon chains, the headgroup, and the composition of the liposome did not affect the partitioning of PCBs into the studied liposomes. Whereas Kliposome/water values were similar to literature values of Koctanol/water adjusted to 37 °C, they both were lower than KAFT/water and Kcell/water by a factor of 3 on average. Partitioning of PCBs into triolein on the other hand closely mimicked that into human lipids, for which triolein is thus a better surrogate than either octanol or liposomes. Previously published polyparameter linear free energy relationships for partitioning from water into storage lipids and liposomes predicted the measured partition coefficients with a root-mean-square error of less than 0.15 log units, if the chosen equations and solute descriptors do not allow chlorine substitution in the ortho-position to influence the prediction. By guiding the selection of (i) a surrogate for the experimental determination and (ii) a method for the prediction of partitioning into human lipids, this study contributes to a better assessment of hydrophobic organic contaminant bioaccumulation in humans.

Solubility prediction of pharmaceuticals in dioxane + water mixtures at various temperatures: Effects of different descriptors and feature selection methods

Solubility of drugs in dioxane+water mixtures at different temperatures was mathematically described by the Jouyban–Acree model employing different solute descriptors; i.e. 1) Abraham solvation, 2) partial solubility, 3) HyperChem and 4) Dragon descriptors in the solute–solvent interactions terms to represent the effects of solute structure on the solubility of drugs in dioxane+water as a model solvent mixture system. Trained versions of these models using collected solubility data in dioxane+water mixtures were proposed and their accuracies were evaluated employing the individual percentage deviation (IPD) and the mean of IPD (MPD) for each drug. Different feature selection methods were applied to select the relevant descriptors and the accuracy of the different descriptors and these approaches were compared. The validity of each model was evaluated using the leave-one-drug out validation method. The results of Dragon descriptors selected by the enhanced replacement method gave a significantly improved model for predicting solubility in dioxane+water. The results confirm that solute–solvent interaction is a critical factor in predicting the solubility of drugs in solvent mixtures. In addition, the feature selection is a critical issue in developing a quantitative structure–property relationships (QSPR) model.

Predicting Partition Coefficients of Polyfluorinated and Organosilicon Compounds using Polyparameter Linear Free Energy Relationships (PP-LFERs)

The environmental behavior, fate, and effects of polyfluorinated compounds (PFCs) and organosilicon compounds (OSCs) have received increasing attention in recent years. In this study, polyparameter linear free energy relationships (PP-LFERs) were evaluated for predicting partition coefficients of neutral PFCs and OSCs, using experimental data for fluorotelomer alcohols (FTOHs) and cyclic volatile methylsiloxanes (cVMS) reported in the literature and measured newly for this work. It was found that the recently proposed PP-LFER model that uses the McGowan characteristic volume (V), the logarithmic hexadecane-air partition coefficient (L), and three polar interaction descriptors can accurately describe partition coefficients of PFCs and OSCs. The prediction errors were <1 log unit when literature descriptors were used, and the errors were reduced to <0.2 log units on average by further optimization of the descriptors. Surprisingly, the conventional forms of PP-LFERs that include the excess molar refraction (E) sometimes led to substantial errors (>1 log unit) even with optimized parameters. The system parameters for octanol-water, air-water, octanol-air, oil-water, liposome-water, and organic carbon-water partition coefficients as well as the solute descriptors for FTOHs and cVMS were recalibrated in this work, which should provide even more reliable predictions of partition coefficients. The results also confirm the consistency of the published experimental partition coefficients for FTOHs and cVMS.

In silico modelling for predicting the cationic hydrophobicity and cytotoxicity of ionic liquids towards the Leukemia rat cell line, Vibrio fischeri and Scenedesmus vacuolatus based on molecular interaction potentials of ions

In this study we present prediction models for estimating in silico the cationic hydrophobicity and the cytotoxicity (log [1/EC50]) of ionic liquids (ILs) towards the Leukemia rat cell line (IPC-81), the marine bacterium Vibrio fischeri and the limnic green algae Scenedesmus vacuolatus using linear free energy relationship (LFER) descriptors computed by COSMO calculations. The LFER descriptors used for the prediction model (i.e. excess molar refraction (E), dipolarity/polarizability (S), hydrogen-bonding acidity (A), hydrogen-bonding basicity (B) and McGowan volume (V)) were calculated using sub-descriptors (sig2, sig3, HBD3, HBA4, MR, and volume) derived from COSMO–RS, COSMO and OBPROP. With the combination of two solute descriptors (B, V) of the cation we were able to predict cationic hydrophobicity values (log ko ) with r 2 = 0.987 and standard error (SE) = 0.139 log units. By using the calculated log k o values, we were able to deduce a linear toxicity prediction model. In the second prediction study for the cytotoxicity of ILs, analysis of descriptor sensitivity helped us to determine that the McGowan volume (V) terms of the cation was the most important predictor of cytotoxicity and to simplify prediction models for cytotoxicity of ILs towards the IPC-81 (r 2 of 0.778, SE of 0.450 log units), Vibrio fischeri (r 2 of 0.762, SE of 0.529 log units) and Scenedesmus vacuolatus (r 2 of 0.776, SE of 0.825 log units). The robustness and predictivity of the two models for IPC-81 and Vibrio fischeri were checked by comparing the calculated SE and r 2 (coefficient of determination) values of the test set.

An Extensible Framework for Capturing Solvent Effects in Computer Generated Kinetic Models

Detailed kinetic models provide useful mechanistic insight into a chemical system. Manual construction of such models is laborious and error-prone, which has led to the development of automated methods for exploring chemical pathways. These methods rely on fast, high-throughput estimation of species thermochemistry and kinetic parameters. In this paper, we present a methodology for extending automatic mechanism generation to solution phase systems which requires estimation of solvent effects on reaction rates and equilibria. The linear solvation energy relationship (LSER) method of Abraham and co-workers is combined with Mintz correlations to estimate ΔG(solv)°(T) in over 30 solvents using solute descriptors estimated from group additivity. Simple corrections are found to be adequate for the treatment of radical sites, as suggested by comparison with known experimental data. The performance of scaled particle theory expressions for enthalpic-entropic decomposition of ΔG(solv)°(T) is also presented along with the associated computational issues. Similar high-throughput methods for solvent effects on free-radical kinetics are only available for a handful of reactions due to lack of reliable experimental data, and continuum dielectric calculations offer an alternative method for their estimation. For illustration, we model liquid phase oxidation of tetralin in different solvents computing the solvent dependence for ROO• + ROO• and ROO• + solvent reactions using polarizable continuum quantum chemistry methods. The resulting kinetic models show an increase in oxidation rate with solvent polarity, consistent with experiment. Further work needed to make this approach more generally useful is outlined.

Analysis of amino acid–water interactions by partitioning in aqueous two-phase systems. I—Amino acids with non-polar side-chains

Partition ratios of 10 l-amino acids with non-polar side chains (Gly, Ala, Val, nor-Val, Ile, Leu, nor-Leu, Phe, Trp and Pro) were measured in ten different polymer/polymer aqueous two-phase systems (ATPS) containing 0.15M NaCl in 0.01M phosphate buffer, pH 7.4. The solute-specific coefficients representing the solute dipole–dipole, hydrogen bonding and electrostatic interactions with aqueous environment for the compounds examined were determined by the multiple linear regression analysis using the modified linear solvation energy relationship equation. It is shown that linear combinations of these coefficients are correlated with the amino acid lipophilicity/hydrophobicity scales reported in the literature. The results obtained imply that the solute-specific coefficients may be used as solute descriptors for quantitative structure–property relationship (QSPR) analysis.

Using linear solvation energy relationship model to study the retention factor of solute in liquid chromatography

This review highlights the use of linear solvation energy relationships (LSERs) to evaluate liquid chromatography systems. From the systems investigated, this paper discusses the descriptors of solutes and the coefficients of various descriptors in detail. Descriptors, such as the solute size, polarity/polarizability, hydrogen bond acidity, and hydrogen bond basicity, in the LSER were examined. The effects of the stationary phase and mobile phase properties on the coefficients of these descriptors were reviewed. The LSER coefficients were evaluated in terms of the corresponding properties of the stationary phase and mobile phase. Copyright © 2012 John Wiley & Sons, Ltd.

Partition coefficients of furan derivative compounds in 1-n-butyl-3-methylimidazolium chloride ([bmim][Cl])–supercritical CO2 biphasic systems and their correlation and prediction with the LSER-δ model

Infinite dilution partition coefficients of furan and furan derivative compounds (2-methylfuran, 2-ethylfuran, furfural and furfuryl alcohol) between ionic liquid ([bmim][Cl]: 1-n-butyl-3-methylimidazolium chloride) and supercritical carbon dioxide (scCO2) were measured at temperatures from 313 to 353K and at pressures from 9 to 21MPa with a chromatographic method. A linear solvation energy relationship-solubility parameter (LSER-δ) model was applied to available data and was compared with a proposed entropic term and with the Sanchez–Lacombe equation of state (SL EoS). The average relative deviations (ARDs) of furan derivative compounds, which were newly measured, for SL EoS model, LSER-δ model and modified LSER-δ model were within 3.2%, 14.9% and 14.7%, respectively. The LSER-δ model can be applied as prediction model for Kw3∞ of ionic liquid–CO2 systems for which solute descriptors are available.

Liver Specificity of the Carcinogenicity of NOCs: A Chemical–Molecular Perspective

This study aimed to determine the most significant molecular features associated with the liver specificity of the carcinogenicity of N-nitroso compounds (NOCs). Accordingly, quantitative structure-activity relationship (QSAR) analysis was performed to extract molecular information from NOCs using a topological substructural molecular descriptor (TOPS-MODE) approach. A linear discriminant analysis (LDA) model of a series of NOCs for rat liver was developed using TOPS-MODE descriptors to predict nonliver- and liver-carcinogenic NOCs. Two descriptors exclusively calculated from the molecular structures of the compounds were selected by a genetic algorithm. The descriptors were then weighted with bond distances as well as the Abraham solute descriptor partition between water and aqueous solvent systems to indicate the importance of their roles in liver specificity. The performances of the LDA model were rigorously validated by leave-one-out cross-validation and external validation, with the prediction accuracy reaching 88.3% and 80.0%, respectively. The contributions of the different molecular fragments to rat-liver specificity were computed. The results served as important information related to liver specificity and were analyzed from the chemical-molecular perspective. The resulting model can provide an efficient method to discriminate between as well as extrapolate nonliver- and liver-carcinogenic NOCs. The contribution of the entire nitrosamine molecule was determined as being responsible for the liver specificity of nitrosamine carcinogenicity. Although the QSAR showed limitations in complex hepatocarcinogenicity, the proposed method may considerably help elucidate the role of nitrosamines in liver specificity from the chemical-molecular perspective. The nature of these enzyme-substrate interactions is characterized. Insight into the chemical-structural and biological factors related to the liver-specific biological activity of NOCs is also provided.

QSAR study of liver specificity of carcinogenicity of N-nitroso compounds

The quantitative structure–activity relationship (QSAR) of N-nitroso compounds (NOCs) for rat liver was developed by a topological sub-structural molecular-descriptors (TOPS-MODE) approach to predict non-liver-carcinogenic and liver-carcinogenic N-nitroso compounds based on a data set of 108 NOCs. Three descriptors calculated solely from the molecular structures of the compounds were selected by enhanced replacement method (ERM) and were weighted, respectively, with atomic weight, bond dipole moments and Abraham solute descriptor partition between water and aqueous solvent systems to indicate the importance of their roles in liver specificity. A detailed discussion on these three descriptors was carried out, and the contributions of different fragments to rat-liver specificity and the interactions among fragments were analyzed. Such results can offer some useful theoretical references for understanding the chemical structural and biological factors related to the liver-specific biological activity of NOCs.

Human Skin Permeation of Neutral Species and Ionic Species: Extended Linear Free Energy Relationship Analyses

The permeability, K(p), of some ionized solutes (including nine acids and nine bases) through human epidermis membrane was measured in this work. Combined with the experimental K(p) data set for neutral species created by Abraham and Martins and reliable K(p) data for ionic species from the literature, a linear free-energy relationship (LFER) analysis was conducted. The values of log K(p) for 118 compounds have been correlated with solute descriptors to yield an LFER equation that incorporates neutral species and ionic species, with R(2) = 0.861 and SD = 0.462 log units. The equation can be used to predict K(p) for neutral species and ionic species, as well as partly ionized solutes. Predicted values for the passive permeation of the sodium ion and the tetraethylammonium ion are in good accord with the experimental values. It was observed that neutral acids and bases are more permeable than their ionized forms, and that the ratio depends on the actual structure. The correlation between human skin permeation and water-organic solvent/artificial membrane partitions was investigated by comparison of the coefficients in the LFER equations. Partition into cerasome is a reasonable model for partition into skin, and using cerasome as a surrogate for the partitioning process, we separate permeation into partition and diffusion processes. We show that the poor permeability of ionic species is largely due to slow diffusion through the stratum corneum. This is especially marked for a number of protonated base cations.

Determination of Abraham model solute descriptors for benzoin based on measured solubility ratios

Experimental solubilities are reported for benzoin dissolved in 1 chloroalkane, 15 alcohols, 5 aromatic hydrocarbons, 5 alkyl acetates, 5 alkoxyalcohols and 3 ethers at 298.15 K. The measured solubility data were mathematically correlated with the Abraham solvation parameter model. The solute descriptors that were calculated for benzoin describe the experimental molar solubility data to within an overall average standard deviation of 0.097 log units.

Prediction and Experimental Evaluation of Soil Sorption by Natural Hormones and Hormone Mimics

Surface runoff from manure-fertilized fields is a significant source of endocrine-disrupting compounds (EDCs) in the environment. Sorption by soils may play a major role in the environmental fate of manure-borne EDCs, including 17α- and 17β-estradiol (17α-E2 and 17β-E2), estrone (E1), melengestrol acetate (MGA), 17α- and 17β-trenbolone (17α-TB and 17β-TB), trendione (TND), and zeranol (α-ZAL). As a measure of sorption behavior, the organic carbon-normalized partition coefficients (K(OC)) of 17β-E2, E1, MGA, and α-ZAL were experimentally determined for three agricultural soils with initial EDC concentrations spanning from ∼0.01 to >1 μM. Sorption isotherms were linear for most solute-soil combinations. Measured K(OC) values were compared to those predicted using a suite of single-parameter and polyparameter linear free energy relationships (sp- and pp-LFERs). Sp-LFER models were based on experimentally determined octanol-water partition coefficients (K(OW)), whereas pp-LFER solute descriptors were calculated indirectly from experimentally determined solvent-water partition coefficients or the program ABSOLV. Log K(OC) predictions by sp-LFERs were closest to the experimentally determined values, whereas pp-LFER predictions varied considerably due to uncertainties in both solute and sorbent descriptors determined by ABSOLV or estimates using the partition coefficient approach.

Evaluation of Acidity Constants of Anthraquinone Derivatives in Methanol/Water Mixtures Using Real Quantum Descriptors

This work presents an evaluation of the acidity constants of 1-hydroxy-9,10-anthraquinone derivatives using electronic descriptors in four methanol/water mixtures with different volume percents and in isolation. An Onsager reaction field model was applied for charges, orbital energies, and dipole descriptors of solutes in the four different solvent mixtures with different volume percents. Each solute was placed in a cavity surrounded by a continuous dielectric constant medium, and the quantum chemical descriptors of the solute were calculated. Electrostatic potential energy descriptors of the isolated solutes were obtained using density functional theory. Relevant descriptors affecting the acidities of the anthraquinones were selected using a genetic algorithm (GA) method. Contributions of the selected variables to the prediction of the experimental acidity constant values were determined by applying a multi-layer perceptron-based feedforward artificial neural network (ANN). A proper model, reproducing experimental acidity constants within average absolute errors of less than 1.10% was achieved. The model proposed relatively higher contributions of dipoles (insolvent) and electrostatic potentials (in isolation) than the orbital energies, revealing the role of strong dipole–dipole interactions such as hydrogen bonds as well as halogen bonds in the proton dissociation process. The gaps between the HOMO and LUMO energies were found to contribute to the stabilization of anions leading to higher dissociation constants.

Effect of NaCl additive on properties of aqueous PEG–sodium sulfate two-phase system

The concentrations of all components in the phases of aqueous two-phase polyethylene glycol-sodium sulfate system of a fixed composition with different concentrations of NaCl additive were determined. Solvatochromic solvent features of aqueous media in the phases of all the systems were characterized in terms of solvent dipolarity/polarizability, solvent hydrogen bond donor acidity and hydrogen bond acceptor basicity. Partitioning of a homologous series of dinitrophenylated amino acids with aliphatic alkyl side chain was examined in all the systems, and the differences between the relative hydrophobicity and electrostatic properties of the phases were quantified. These differences were described in terms of solvatochromic solvent features of the phases. The previously reported partition coefficients of twelve different nonionic compounds in all the systems were expressed in terms of solute descriptors. It is demonstrated that two solvatochromic solvent descriptors (solvent dipolarity/polarizability, and solvent hydrogen bond donor acidity) could adequately describe the partitioning of the solutes in all the systems employed.

Predicting Carcinogenicity and Understanding the Carcinogenic Mechanism of N-Nitroso Compounds Using a TOPS-MODE Approach

A linear discriminant analysis (LDA) coupled with an enhanced replacement method (ERM) was used as an alternative method to predict the carcinogenicity of N-nitroso compounds (NOCs) in rats. This presented LDA based on the topological substructural molecular descriptors (TOPS-MODE) approach was developed to predict the carcinogenic and noncarcinogenic activity on a data set of 111 NOCs with a good classification value of 90.1%. The predictive power of the LDA model was validated through an external validation set (37 compounds) with a prediction accuracy of 94.6% and a leave-one-out cross-validation procedure (LOOCV) with a good prediction of 86.5%. This methodology showed that the TOPS-MODE descriptors weighted, respectively, by bond dipole moment and Abraham solute descriptor dipolarity/polarizability affected the NOC carcinogenicity. The contributions of certain bonds and fragments to carcinogenicity were used to assess biotransformation and carcinogenic mechanisms. The positive contribution of the carbon-nitrogen single bond (between the N-nitroso group and α-carbon to the N-nitroso group) indicated that the α-hydroxylation reaction could occur at the α-carbon or otherwise not occur. Similarly, the contributions from the molecular fragment could be applied to indicate whether the fragments generated an alkylating agent. These results suggested that this approach could discriminate between carcinogenic and noncarcinogenic NOCs, thereby providing insight into the structural features and chemical factors related to NOC carcinogenicity.

Serum Albumin Binding of Structurally Diverse Neutral Organic Compounds: Data and Models

Binding to serum albumin has a strong influence on freely dissolved, unbound concentrations of chemicals in vivo and in vitro. For neutral organic solutes, previous studies have suggested a log-log correlation between the albumin-water partition coefficient and the octanol-water partition coefficient (K(ow)) and postulated highly nonspecific binding that is mechanistically analogous to dissolution into solvents. These relationships and concepts were further explored in this study. Bovine serum albumin (BSA)-water partition coefficients (K(BSA/w)) were measured for 83 structurally diverse neutral organic chemicals in consistent experimental conditions. The correlation between log K(BSA/w) and log K(ow) was moderate, with R(2) = 0.76 and SD = 0.43. The log K(BSA/w) of low-polarity compounds including a series of chlorobenzenes and polycyclic aromatic hydrocarbons increased with log K(ow) linearly up to log K(ow) = 4-5, but then the linear relationship apparently broke off, and the increase became gradual. The fitting of polyparameter linear free energy relationship models with five solute descriptors was just comparable to that of the log K(ow) model (R(2) = 0.78-0.79, SD = 0.41-0.42); the relatively high SD obtained suggests that solvent dissolution models are not capable of modeling albumin binding accurately. A size limitation of the binding site(s) of albumin is suggested as a possible reason for the high SD. An equilibrium distribution model indicates that serum albumin generally has high contributions to the binding in the serum of polar compounds and relatively small low-polarity compounds, whereas albumin binding for large low-polarity compounds is outcompeted by the strong partitioning into lipids due to low relative affinity of albumin for these compounds.