Prediction of Human Lethal Doses and Concentrations of MEIC Chemicals from Rodent LD50 Values: An Attempt to Make Some Reparation.

The prediction of human toxicities from animal toxicity tests is often poor, and is now discouraged and in some cases banned, especially those involving the LD50 test. However, there is a vast number of historical LD50 data in both public and in-house repositories that are being put to little use. This study examined the correlations between human lethality (doses and concentrations) of 36 MEIC chemicals and the median values of a large number of mouse and rat LD50 values obtained for four different routes of administration. The best correlations were found with mouse and rat intraperitoneal LD50 values (r2 = 0.838 and 0.810 for human lethal dose, and r2 = 0.753 and 0.785 for human lethal concentration). The results show that excellent prediction of human lethal dose and concentration can be made, for this series of chemicals at least, by using uncurated rodent LD50 values, thus offering some reparation for the millions of rodent lives sacrificed in LD50 testing.

Calculation of aqueous solubility of crystalline un-ionized organic chemicals and drugs based on structural similarity and physicochemical descriptors.

Solubilities of crystalline organic compounds calculated according to AMP (arithmetic mean property) and LoReP (local one-parameter regression) models based on structural and physicochemical similarities are presented. We used data on water solubility of 2615 compounds in un-ionized form measured at 25±5 °C. The calculation results were compared with the equation based on the experimental data for lipophilicity and melting point. According to statistical criteria, the model based on structural and physicochemical similarities showed a better fit with the experimental data. An additional advantage of this model is that it uses only theoretical descriptors, and this provides means for calculating water solubility for both existing and not yet synthesized compounds.

Aqueous Drug Solubility: What Do We Measure, Calculate and QSPR Predict?

Detailed critical analysis of publications devoted to QSPR of aqueous solubility is presented in the review with discussion of four types of aqueous solubility (three different thermodynamic solubilities with unknown solute structure, intrinsic solubility, solubility in physiological media at pH=7.4 and kinetic solubility), variety of molecular descriptors (from topological to quantum chemical), traditional statistical and machine learning methods as well as original QSPR models.

2-Heteroarylimino-5-benzylidene-4-thiazolidinones analogues of 2-thiazolylimino-5-benzylidene-4-thiazolidinones with antimicrobial activity: synthesis and structure-activity relationship.

2-Heteroarylimino-5-benzylidene-4-thiazolidinones, unsubstituted or carrying hydroxy, methoxy, nitro and chloro groups on the benzene ring, were synthesised and assayed in vitro for their antimicrobial activity against gram positive and gram negative bacteria, yeasts and mould. The antimicrobial activity of the 2-benzo[d]thiazolyl- and of the 2-benzo[d]isothiazolyl-imino-5-benzylidene-4-thiazolidinones is, on the whole, lower in comparison with the high activity detected for the derivatives of the 2-thiazolylimino-5-benzylidene-4-thiazolidinone class. Nevertheless most of the benzo[d]thiazole analogues display good inhibition of the growth of gram positive bacilli and staphylococci, including methicillin-resistant Staphylococcus strains. Among the 2-benzo[d]isothiazole analogues a few derivatives show a strong and selective activity against bacilli. Moreover, it is worth noting that the replacement of the thiazole nucleus for the benzo[d]thiazole bicyclic system in the parent 2-(benzo[d]thiazol-2-ylimino)thiazolidin-4-one leads to significant antifungal properties against both yeasts and moulds, properties not shown by the analogous 2-thiazolyl- and 2-benzo[d]isothiazolyl-imino)thiazolidin-4-ones. The structure-activity relationship of 33 analogues possessing the 2-heteroarylimino-4-thiazolidinone structure is analysed through QSAR models.

Classification (Agonist/Antagonist) and Regression "Structure-Activity" Models of Drug Interaction with 5-HT6.

INTRODUCTION: One promising target for novel psychotropic drugs is the 5-HT6 receptor, GProtein- Coupled Receptor (GPCR) family, displaying seven transmembrane domains. There is considerable interest in how both 5-HT6 receptor agonist and antagonist compounds can have marked procognitive effects. METHODS: An exact structure of the 5-HT6 receptor is not available, so application of powerful methods of (Q)SAR and molecular modelling, which play an essential role in modern drug design, are currently limited to structure-based homology models. The present study is devoted to a detailed QSAR analysis of 61 drugs (26 agonists and 35 antagonists) acting on the 5-HT6 receptor (rattus norvegicus and homo sapiens). Five classification methods were used: k-Nearest Neighbors (k-NN), Logistic Regression (LG), Linear Discriminant Analysis (LDA), Random Forest (RF), and Support Vector Machine (SVM). Multiple Regression Analysis (MRA) was involved also for regression analysis. Spectra of Inter Atomic Interactions (SIAI) were applied in the search for ligand centres interacting with the 5- HT6 receptor. RESULTS & CONCLUSION: SAR and QSAR models based on the use of HYBOT, MOLTRA, VolSurf+, and SYBYL programs, and having cross-validated coefficients of determination of at least 0.80, show a predominant influence of H-bond acceptor ability and hydrophobicity on the type of ligand activity and degree of inhibition.

QSPR model of Henry's law constant for a diverse set of organic chemicals based on genetic algorithm-radial basis function network approach.

Six quantitative structure-property relationship (QSPR) models for a diverse set of experimental data of Henry's law constant (H) of organic chemicals under environmental condition (T=25 degrees C; water-air system) have been developed based on four different molecular descriptor sets. Three different models based on the descriptors of CODESSA (Comprehensive Descriptors for Structural and Statistical Analysis), Tsar, and Dragon software and a model based on a combined descriptor set from these packages, and in addition from HYBOT software, have been established using the stepwise regression method. The combined descriptors set model gave the best results. Furthermore, a genetic algorithm was used for descriptor selection from a combined set of descriptors, and a radial basis function network was utilized to establish a model with a low root mean square error (RMSE). The results of this study were compared with the well-known bond contribution and group contribution methods. The group contribution method failed to predict Henry's law constant of 170 from all 940 compounds in the data-set. RMSEs of 0.693, 0.798, and 0.564 were achieved for bond contribution, group contribution and the best QSPR model of this study, respectively, based on logarithm of H. Analysis of different QSPR models showed that hydrogen bonding between the organic solute and water as a solvent has the greatest influence on this partitioning phenomenon.

In silico prediction of ADMET properties: how far have we come?

There have been considerable advances in the last few years in both the quantity and the quality of in silico ADMET property predictions. Most ADMET properties are now computable, and the accuracy of some of the software predictions for physicochemical properties in particular is close to that of measured data. There is, however, universal agreement that more good experimental ADMET data are needed for use in in silico model development, for models are only as good as the data on which they are based. Many data remain confidential but it is to be hoped that, with projects such as the Vitic toxicity database, being developed by Lhasa Limited, pharmaceutical companies will be prepared to release data to an 'honest broker' on a confidential basis, so that better in silico models can be developed. Incorporation of calculated ADMET properties into drug discovery and development is a multi-factorial problem and really needs a multi-factorial solution. Some progress is being made in this direction and it is hoped that within the foreseeable future software will be available for this purpose.

Defining the toxic mode of action of ester sulphonates using the joint toxicity of mixtures.

Previously we have presented aquatic toxicity data for the class of anionic surfactant ester sulphonates (ES) to Daphnia magna. We now present toxicity data for binary mixtures of ES substances with reference substances of known mode of action. Using a toxic unit (TU) approach, data indicated that ES substances exhibit concentration addition with linear alkylbenzene sulphonate (LAS) and phenols and response addition with alcohols. This suggests that ES behave with a similar mode of action to phenol and LAS which are known polar narcotics and with a dissimilar mode of action to alcohols which are known baseline narcotics.

The aquatic toxicity of anionic surfactants to Daphnia magna--a comparative QSAR study of linear alkylbenzene sulphonates and ester sulphonates.

This paper develops quantitative structure activity relationships (QSARs) for the acute aquatic toxicity of the anionic surfactants linear alkylbenzene sulphonates (LAS) and ester sulphonates (ES) to Daphnia magna, the aim being to investigate the modes of action by comparing the QSARs for the two types of surfactant. The generated data for ES have been used to develop a QSAR correlating toxicity with calculated log P values: log(1/EC50)= 0.78 log P+1.37. This equation has an intercept 1.1 log units lower than a QSAR for linear alkylbenzene sulphonates (LAS). The findings suggest that either ES surfactants act by a different mode of action to LAS and other anionic surfactants or the log P calculation method introduces a systematic overestimate when applied to ES.

Quantitative structure-activity-activity and quantitative structure-activity investigations of human and rodent toxicity.

Acute toxicity in different biological systems, including humans and rodents in vivo, and human and rodent cell lines in vitro, was investigated. The data were taken from the MEIC (Multicentre Evaluation of In Vitro Cytotoxicity) programme. Quantitative structure-activity-activity relationship (QSAAR) models were developed for the in vivo human and rodent toxicity including a combination of toxicity endpoint and structural descriptors as predictor variables. The human peak blood/serum LC(50) concentrations were most strongly related to human liver cell toxicity, while the in vivo oral human lethal doses were most closely related to the in vivo rodent LD(50) values. The QSAARs included structural descriptors encoding electronic/reactivity properties, presence of H-bond donors, compound aromaticity, and size/shape properties. Quantitative structure-activity relationships (QSARs) were derived by using structural descriptors accounting for molecular hydrophobicity, size and shape, and electronic properties. These models have the potential to provide useful insights in the development of non-animal (in vitro and in silico) methods for predicting human and mammalian toxicity.

Physicochemical property profile for brain permeability: comparative study by different approaches.

A comparative study of classification models of brain penetration by different approaches was carried out on a training set of 1000 chemicals and drugs, and an external test set of 100 drugs. Ten approaches were applied in this work: seven medicinal chemistry approaches (including "rule of 5" and multiparameter optimization) and also three SAR techniques: logistic regression (LR), random forest (RF) and support vector machine (SVM). Forty-one different medicinal chemistry descriptors representing diverse physicochemical properties were used in this work. Medicinal chemistry approaches based on the intuitive estimation of preference zones of CNS or non-CNS chemicals, with different rules and scoring functions, yield unbalanced models with poor classification accuracy. RF and SVM methods yielded 82% and 84% classification accuracy respectively for the external test set. LR was also successful in CNS/non-CNS (denoted in this study as CNS+/CNS-) classification and yielded an overall accuracy equivalent to that of SVM and RF. At the same time, LR is especially valuable for medicinal chemists because of its simplicity and the possibility of clear mechanistic interpretation.

Computer-aided rodent carcinogenicity prediction.

The potential of the computer program PASS (Prediction Activity Spectra for Substances) to predict rodent carcinogenicity for chemical compounds was studied. PASS predicts carcinogenicity of chemical compounds on the basis of their structural formula and of structure-activity relationship analysis of known carcinogens and non-carcinogens. The data on structures and experimental results of 2-year carcinogenicity assays for 412 chemicals from the NTP (National Toxicological Program) and 1190 chemicals from the CPDB (Carcinogenic Potency Database) were used in our study. The predictions take into consideration information about species and sex of animals. For evaluation of the predictive accuracy we used two procedures: leave-one-out cross-validation (LOO CV) and leave-20%-out cross-validation. In the last case we randomly divided the studied data set 20 times into two subsets. The data from the first subset, containing 80% of the compounds, were added to the PASS training set (which includes about 46,000 compounds with about 1500 biological activity types collected during the last 20 years to predict biological activity spectra), the second subset with 20% of the compounds was used as an evaluation set. The mean accuracy of prediction calculated by LOO CV is about 73% for NTP compounds in the 'equivocal' category of carcinogenic activity and 80% for NTP compounds in the 'evidence' category of carcinogenicity. The mean accuracy of prediction for the CPDB database is 89.9% calculated by LOO CV and 63.4% calculated by leave-20%-out cross-validation. Influence of incorporation of species and sex data on the accuracy of carcinogenicity prediction was also investigated. It was shown that the accuracy was increased only for data on male animals.

Use of artificial neural networks in the QSAR prediction of physicochemical properties and toxicities for REACH legislation.

With the introduction of the REACH legislation in the European Union, there is a requirement for property and toxicity data on chemicals produced in or imported into the EU at levels of 1 tonne/year or more. This has meant an increase in the in silico prediction of such data. One of the chief predictive approaches is QSAR (quantitative structure-activity relationships), which is widely used in many fields. A QSAR approach that is increasingly being used is that of artificial neural networks (ANNs), and this chapter discusses its application to the range of physicochemical properties and toxicities required by REACH. ANNs generally outperform the main QSAR approach of multiple linear regression (MLR), although other approaches such as support vector machines sometimes outperform ANNs. Most ANN QSARs reported to date comply with only two of the five OECD Guidelines for the Validation of (Q)SARs.

In silico Prediction of Aqueous Solubility: a Comparative Study of Local and Global Predictive Models.

32 Quantitative Structure-Property Relationship (QSPR) models were constructed for prediction of aqueous intrinsic solubility of liquid and crystalline chemicals. Data sets contained 1022 liquid and 2615 crystalline compounds. Multiple Linear Regression (MLR), Support Vector Machine (SVM) and Random Forest (RF) methods were used to construct global models, and k-nearest neighbour (kNN), Arithmetic Mean Property (AMP) and Local Regression Property (LoReP) were used to construct local models. A set of the best QSPR models was obtained: for liquid chemicals with RMSE (root mean square error) of prediction in the range 0.50-0.60 log unit; for crystalline chemicals 0.80-0.90 log unit. In the case of global models the large number of descriptors makes mechanistic interpretation difficult. The local models use only one or two descriptors, so that a medicinal chemist working with sets of structurally-related chemicals can readily estimate their solubility. However, construction of stable local models requires the presence of closely related neighbours for each chemical considered. It is probable that a consensus of global and local QSPR models will be the optimal approach for construction of stable predictive QSPR models with mechanistic interpretation.

Design of new cognition enhancers: from computer prediction to synthesis and biological evaluation.

To discover new cognition enhancers, a set of virtually designed synthesizable compounds from different chemical series was investigated using two computer-aided approaches. One of the approaches is prediction of biological activity spectra for substances (PASS) and the second is prediction of toxicity, mutagenicity, and carcinogenicity (DEREK). To increase the probability of finding new chemical entities, we investigated a heterogeneous set of highly diverse chemicals including different types of heterocycles: five-membered (thiophenes, thiazoles, imidazoles, oxazoles, pyrroles), six-membered (pyridines, pyrimidines), seven-membered (diazepines, triazepines), fused five+six-membered heterocycles (indoles, benzothiazoles, purines, indolizines, neutral, mesoionic, and cationic azolopyridines). A database including 5494 structures of compounds was created. On the basis of the PASS and DEREK prediction results, eight compounds with the highest probability of cognition-enhancing effect were selected. The cognition-enhancing activity testing showed that all of the selected compounds had a pronounced antiamnesic effect and were found to reduce significantly scopolamine-induced amnesia of passive avoidance reflex (PAR). The action of compounds at doses of 1 and 10 mg/kg caused a statistically significant increase in latent time of reflex and in the number of animals, which did not enter the dark chamber when testing the PAR. Therefore, on the basis of computer prediction, new cognition-enhancing agents were discovered within the chemical series, in which this activity was not known previously.

Quantitative structure-property relationships for prediction of boiling point, vapor pressure, and melting point.

Boiling point, vapor pressure, and melting point are important physicochemical properties in the modeling of the distribution and fate of chemicals in the environment. However, such data often are not available, and therefore must be estimated. Over the years, many attempts have been made to calculate boiling points, vapor pressures, and melting points by using quantitative structure-property relationships, and this review examines and discusses the work published in this area, and concentrates particularly on recent studies. A number of software programs are commercially available for the calculation of boiling point, vapor pressure, and melting point, and these have been tested for their predictive ability with a test set of 100 organic chemicals.

Quantitative structure-property relationships for predicting Henry's law constant from molecular structure.

Various models are available for the prediction of Henry's law constant (H) or the air-water partition coefficient (Kaw), its dimensionless counterpart. Incremental methods are based on structural features such as atom types, bond types, and local structural environments; other regression models employ physicochemical properties, structural descriptors such as connectivity indices, and descriptors reflecting the electronic structure. There are also methods to calculate H from the ratio of vapor pressure (p(v)) and water solubility (S(w)) that in turn can be estimated from molecular structure, and quantum chemical continuum-solvation models to predict H via the solvation-free energy (deltaG(s)). This review is confined to methods that calculate H from molecular structure without experimental information and covers more than 40 methods published in the last 26 years. For a subset of eight incremental methods and four continuum-solvation models, a comparative analysis of their prediction performance is made using a test set of 700 compounds that includes a significant number of more complex and drug-like chemical structures. The results reveal substantial differences in the application range as well as in the prediction capability, a general decrease in prediction performance with decreasing H, and surprisingly large individual prediction errors, which are particularly striking for some quantum chemical schemes. The overall best-performing method appears to be the bond contribution method as implemented in the HENRYWIN software package, yielding a predictive squared correlation coefficient (q2) of 0.87 and a standard error of 1.03 log units for the test set.

Quantitative cationic-activity relationships for predicting toxicity of metals.

Developing and validating quantitative cationic-activity relationships or (Q)CARs to predict the toxicity metals is challenging because of issues associated with metal speciation, complexation and interactions within biological systems and the media used to study these interactions. However, a number of simplifying assumptions can be used to develop and validate (Q)CARs to predict the toxicity of metals: The ionic form is the most active form of a metal; the bioactivity of a dissolved metal is correlated with its free ion concentration or activity; most metals exist in biological systems as cations, and differences in metal toxicity result from differences in metal ion binding to biological molecules (ligand-binding). In summary, it appears that certain useful correlations can be made between several physical and chemical properties of ions (mostly cations) and toxicity of metals. This review provides a historical perspective of studies that have reported correlations between physical and chemical properties of cations and toxicity to mammalian and nonmammalian species using in vitro and in vivo assays. To prepare this review, approximately 100 contributions dating from 1839 to 2003 were evaluated and the relationships between about 20 physical and chemical properties of cations and their potential to produce toxic effects were examined.

Guidelines for developing and using quantitative structure-activity relationships.

Numerous quantitative structure-activity relationships (QSARs) have been developed to predict properties, fate, and effects of mostly discrete organic chemicals. As the demand for different types of regulatory testing increases and the cost of experimental testing escalates, there is a need to evaluate the use of QSARs and provide some guidance to avoid their misuse, especially as QSARs are being considered for regulatory purposes. This paper provides some guidelines that will promote the proper development and use of QSARs. While this paper uses examples of QSARs to predict toxicity, the proposed guidelines are applicable to QSARs used to predict physical or chemical properties, environmental fate, ecological effects and health effects.

Quantitative structure-activity relationships for human health effects: commonalities with other endpoints.

This article describes the use of quantitative structure-activity relationships (QSARs) to predict toxicity endpoints for ecologically relevant and human-surrogate species. The interrelationships between the endpoints, and the possibilities of exploring the commonalities of chemical action from one species to another as well as from one endpoint to another, are evaluated. A number of toxic endpoints are discussed including mutagenicity and carcinogenicity; developmental toxicity (teratogenicity); acute toxicity; skin sensitization; skin, eye, and sensory irritation; and the modeling of membrane permeability. A number of electrophilic molecular substructures have been identified that are common to a number of toxicities. It is postulated that if such a substructure is observed in a molecule, it may exhibit a range of toxicities. Further, there appear to be relationships between the toxicity to ecologically relevant and human-surrogate species, which may allow for appreciation and possible extrapolation in both directions. Overall, however, QSARs are limited by the paucity of available toxicological data and information.