Development of predictive QSAR models for Vibrio fischeri toxicity of ionic liquids and their true external and experimental validation tests

Implementing comprehensive machine learning models of multispecies toxicity assessment to improve regulation of organic compounds

Machine learning has made significant progress in assessing the risk associated with hazardous chemicals. However, most models were constructed by randomly selecting one algorithm and one toxicity endpoint towards single species, which may cause biased regulation of chemicals. In the present study, we implemented comprehensive prediction models involving multiple advanced machine learning and end-to-end deep learning to assess the aquatic toxicity of chemicals. The generated optimal models accurately unravel the quantitative structure-toxicity relationships, with the correlation coefficients of all training sets from 0.59 to 0.81 and of the test sets from 0.56 to 0.83. For each chemical, its ecological risk was determined from the toxicity information towards multiple species. The results also revealed the toxicity mechanism of chemicals was species sensitivity, and the high-level organisms were faced with more serious side effects from hazardous substances. The proposed approach was finally applied to screen over 16,000 compounds and identify high-risk chemicals. We believe that the current approach can provide a useful tool for predicting the toxicity of diverse organic chemicals and help regulatory authorities make more reasonable decisions.

Review of the toxic effects of ionic liquids

Interest in ionic liquids (ILs), called green or designer solvents, has been increasing because of their excellent properties such as thermal stability and low vapor pressure; thus, they can replace harmful organic chemicals and help several industrial fields e.g., energy-storage materials production and biomaterial pretreatment. However, the claim that ILs are green solvents should be carefully considered from an environmental perspective. ILs, given their minimal vapor pressure, may not directly cause atmospheric pollution. However, they have the potential to cause adverse effects if leaked into the environment, for instance if they are spilled due to human mistakes or technical errors. To estimate the risks of ILs, numerous ILs have had their toxicity assessed toward several micro- and macro-organisms over the past few decades. Since the toxic effects of ILs depend on the method of estimating toxicity, it is necessary to briefly summarize and comprehensively discuss the biological effects of ILs according to their structure and toxicity testing levels. This can help simplify our understanding of the toxicity of ILs. Therefore, in this review, we discuss the key findings of toxicological information of ILs, collect some toxicity data of ILs to different species, and explain the influence of IL structure on their toxic properties. In the discussion, we estimated two different sensitivity values of toxicity testing levels depending on the experiment condition, which are theoretical magnitudes of the inherent sensitivity of toxicity testing levels in various conditions and their changes in biological response according to the change in IL structure. Finally, some perspectives, future research directions, and limitations to toxicological research of ILs, presented so far, are discussed.

Ionic Liquids-A Review of Their Toxicity to Living Organisms

Ionic liquids (ILs) were initially hailed as a green alternative to traditional solvents because of their almost non-existent vapor pressure as ecological replacement of most common volatile solvents in industrial processes for their damaging effects on the environment. It is common knowledge that they are not as green as desired, and more thought must be put into the biological consequences of their industrial use. Still, compared to the amount of research studying their physicochemical properties and potential applications in different areas, there is a scarcity of scientific papers regarding how these substances interact with different organisms. The intent of this review was to compile the information published in this area since 2015 to allow the reader to better understand how, for example, bacteria, plants, fish, etc., react to the presence of this family of liquids. In general, lipophilicity is one of the main drivers of toxicity and thus the type of cation. The anion tends to play a minor (but not negligible) role, but more research is needed since, owing to the very nature of ILs, except for the most common ones (imidazolium and ammonium-based), many of them are subject to only one or two articles.

Multiproduct Microalgae Biorefineries Mediated by Ionic Liquids

Ionic liquids (ILs) are salts with low melting points that can be used as solvents for mild extraction and selective fractionation of biomolecules (e.g., proteins, carbohydrates, lipids, and pigments), enabling the valorisation of microalgal biomass in a multiproduct biorefinery concept, while maintaining the biomolecules' structural integrity and activity. Aqueous biphasic systems and emulsions stabilised by core-shell particles have been used to fractionate disrupted microalgal biomass into hydrophobic (lipids and pigments) and hydrophilic (proteins and carbohydrates) components. From nondisrupted biomass, the hydrophobic components can be directly extracted using ILs from intact cells, while the most fragile hydrophilic components can be obtained upon further mechanical cell disruption. These multiproduct biorefinery concepts will be discussed in an outlook on future separations using IL-based systems.

Norm index in QSTR work for predicting toxicity of ionic liquids on Vibrio fischeri

Ionic liquids have been becoming new 'green solvent' because of the low saturation vapor pressure, less volatilization and more recycling utilization. Since most ILs are soluble in water, it should be indispensable to evaluate the ecotoxicology effect of ILs on aquatic environment before using them widely. Based on the concept of norm index, a set of norm descriptors were proposed for anions, cations and ILs. The whole IL structure optimization method has been used to build a predictive norm index-based quantitative structure-toxicity relationship model for the toxicity of ILs on Vibrio fischeri. Statistical results indicated that norm descriptors were reliable and robust in expressing the relationship between structural information and toxicity of ILs. Meanwhile, a series of ILs without experimental values were predicted based on this stable QSTR model. The results indicated that for imidazole-based ILs, an increase in the length of substituent in the branch could enhance the toxicity of ILs on Vibrio fischeri, and the branch contains hydroxyl group, double bond or triple bonds might reduce the toxicity of ILs. Results obtained in this present work would be valuable for the molecular design and the toxicity evaluation toward aquatic organism of ILs.

Application of general toxic effects of ionic liquids to predict toxicities of ionic liquids to Spodoptera frugiperda 9, Eisenia fetida, Caenorhabditis elegans, and Danio rerio

Modeling for the toxicity of ionic liquids (ILs) is necessary to fill data gaps for untested chemicals and to understand the relevant mechanisms at the molecular level. In order for many researchers to easily predict toxicity and/or develop some prediction model, simple method(s) based on a single parameter should be proposed. Therefore, previously our group developed a comprehensive toxicity prediction model with unified linear free-energy relationship descriptors to address the single parameter for predicting the toxicities, as follows (Cho et al., 2016b). Log 1/toxicity in the unit of mM= (2.254 E_c - 2.545 S_c + 0.646 A_c - 1.471 B_c + 1.650 V_c + 2.917 J⁺ - 0.201 E_a + 0.418 V_a + 0.131 J^-) - 0.709. It is considered that the model can calculate the general toxicological effect of ILs in parenthesis, as it was developed on the basis of numerous toxic effects i.e., 58 toxicity testing methods and about 1600 data points. In order to check the hypothesis, the values calculated by the model were correlated with four different datasets from insect cell line (Spodoptera frugiperda 9), earthworm (Eisenia fetida), nematode (Caenorhabditis elegans), and fish (Danio rerio). The results clearly showed that the calculated values are in good agreement with each dataset. In the case of S. frugiperda 9 cells, the calculated parameters were correlated with log1/LC₅₀ values, measured after 24 h and 48 h incubation, in R² of 0.67 and 0.88, respectively. The R² values for the earthworm, nematode, and fish were 0.88, 0.96, and 0.94-0.95, respectively. This study confirmed that the comprehensive model can be simply and accurately used to predict toxicity of ILs.

Prediction of Terpenoid Toxicity Based on a Quantitative Structure-Activity Relationship Model

Terpenoids, including monoterpenoids (C10), norisoprenoids (C13), and sesquiterpenoids (C15), constitute a large group of plant-derived naturally occurring secondary metabolites with highly diverse chemical structures. A quantitative structure-activity relationship (QSAR) model to predict terpenoid toxicity and to evaluate the influence of their chemical structures was developed in this study by assessing in real time the toxicity of 27 terpenoid standards using the Gram-negative bioluminescent Vibrio fischeri. Under the test conditions, at a concentration of 1 µM, the terpenoids showed a toxicity level lower than 5%, with the exception of geraniol, citral, (S)-citronellal, geranic acid, (±)-α-terpinyl acetate, and geranyl acetone. Moreover, the standards tested displayed a toxicity level higher than 30% at concentrations of 50-100 µM, with the exception of (+)-valencene, eucalyptol, (+)-borneol, guaiazulene, β-caryophellene, and linalool oxide. Regarding the functional group, terpenoid toxicity was observed in the following order: alcohol > aldehyde ~ ketone > ester > hydrocarbons. The CODESSA software was employed to develop QSAR models based on the correlation of terpenoid toxicity and a pool of descriptors related to each chemical structure. The QSAR models, based on t-test values, showed that terpenoid toxicity was mainly attributed to geometric (e.g., asphericity) and electronic (e.g., maximum partial charge for a carbon (C) atom (Zefirov's partial charge (PC)) descriptors. Statistically, the most significant overall correlation was the four-parameter equation with a training coefficient and test coefficient correlation higher than 0.810 and 0.535, respectively, and a square coefficient of cross-validation (Q2) higher than 0.689. According to the obtained data, the QSAR models are suitable and rapid tools to predict terpenoid toxicity in a diversity of food products.

Toxicity of aromatic pollutants and photooxidative intermediates in water: A QSAR study

Extensive commercial use of aromatic hydrocarbons results with significant amounts of these chemicals and related by-products in waters, causing a severe ecological and health threat, thus requiring an increased attention. This study was aimed at developing models for prediction of the initial toxicity of the aromatic water-pollutants (expressed as EC₅₀ and TU₀) as well as the toxicity of their intermediates at half-life of the parent pollutant (TU_1/2). For that purpose, toxicity toward Vibrio fischery was determined for 36 single-benzene ring compounds (S-BRCs), diversified by the type, number and position of substituents. Quantitative structure-activity relationship (QSAR) methodology paired with genetic algorithm optimization tool and multiple linear regression was applied to obtain the models predicting the targeted toxicity, which are based on pure structural characteristics of the tested pollutants, avoiding thus additional experimentation. Upon derivation of the models and extensive analysis on training and test sets, 4-, 4- and 5-variable models (for EC₅₀ and TU₀, TU_1/2, respectively) were selected as the most predictive possessing 0.839<R²< 0.901 and 0.789<Q²< 0.859. The analysis of the selected descriptors indicated three major structural characteristics influencing the toxicity: electronegativity, geometry and electrotopological states of the molecule. Degradation kinetics determining as well the pathways of intermediates formation, reflected over ionization potential, was found to be an important parameter determining the toxicity in half-life.

Modeling adsorption of organic pollutants onto single-walled carbon nanotubes with theoretical molecular descriptors using MLR and SVM algorithms

Prediction of adsorption equilibrium coefficients (K) of organic compounds onto single walled carbon nanotubes (SWNTs) from in silico molecular descriptors is of importance for probing potential applications of SWNTs as well as for evaluating environmental behavior and ecological risks of organic pollutants and SWNTs. In this study, two models for predicting logK were developed with multiple linear regression (MLR) and support vector machine (SVM) algorithms. The two models have satisfactory goodness-of-fit, robustness and predictive ability, and the SVM model performs slightly better than the MLR model. The two models are based on the up-to-date experimental dataset consisting of 61 logK values, and the applicability domains cover diverse organic compounds with functional groups > CC<, CC, C₆H₅, >CO, COOH, C(O)O, OH, O, F, Cl, Br, NH₂, NH, >N, >NN<, NO₂, >NC(O)NH₂, >NC(O)NH, S and S(O)(O). The adsorption of organic compounds toward SWNTs is mainly determined by van der Waals forces and hydrophobic interactions. Since only in silico molecular descriptors were employed for the modeling, the developed models are beneficial for prediction purposes.

Understanding the interactions of imidazolium-based ionic liquids with cell membrane models

IL–phospholipid interactions were studied using Langmuir monolayers and molecular simulations.

Ecotoxicological evaluation of magnetic ionic liquids

Although magnetic ionic liquids (MILs) are not yet industrially applied, their continued development and eventual commercial use may lead to their appearance into the aquatic ecosystem through accidental spills or effluents, consequently promoting aquatic contaminations. Furthermore, the deficient information and uncertainty surrounding the environmental impact of MILs could be a major barrier to their widespread industrial application and international registration. Thus, in the present work, a range of cholinium salt derivatives with magnetic properties was synthesized and their ecotoxicity was evaluated towards the luminescent bacteria Vibrio fischeri. The results suggest that all MILs structures tested are moderately toxic, or even toxic, to the bacteria. Furthermore, their toxicity is highly dependent on the structural modifications of the cation, namely the alkyl side chain length and the number of hydroxyethyl groups, as well as the atomic number of the metal anion. Finally, from the magnetic anions evaluated, the [MnCl₄]^2- is the less toxic. In order to improve the knowledge for the prospective design of environmentally safer MILs, it is important to expand this study to other aquatic organisms at different trophic levels.

Modeling from Theory and Modeling from Data: Complementary or Alternative Approaches? The Case of Ionic Liquids

In the field of ionic liquids (ILs), theory-driven modeling approaches aimed at the best fit for all available data by using a unique, and often nonlinear, model have been widely adopted to develop quantitative structure-property relationship (QSPR) models. In this context, we propose chemoinformatic and chemometric data-driven procedures that lead to QSPR soft models with local validity that are able to predict relevant physicochemical properties of ILs, such as viscosity, density, decomposition temperature, and conductivity. These models, which use readily available and easily interpretable VolSurf+ descriptors, represent an unexploited opportunity for experimentalists to model and predict the physicochemical properties of ILs in industrial R&D design.

Comprehensive approach for predicting toxicological effects of ionic liquids on several biological systems using unified descriptors

The challenge and opportunity for design of environmentally-benign ionic liquids (ILs) would start from prediction of their toxicological effects on several endpoints solely based on the structural formulas. Especially, a comprehensive yet simple equation able to predict several biological responses to IL toxicity is of much advantage. Therefore, based on 50 toxicity testing systems on ILs a comprehensively approachable prediction method was developed. For the modelling, approximately 1600 toxicity values measured by several biological systems and an amended linear free energy relationship (LFER) model were used. Since the toxicological activities of an IL could be differently described according to sensitivity of toxicity testing systems, the sensitivity of each of toxicity testing systems was also estimated in the modelling. By statistical analysis with the calculated descriptors, a LFER model was built. Also the sensitivity value of each system on the basis of the comprehensively approachable model was numerically estimated. In results, it was observed that the combination of single model and sensitivity terms was able to predict each of 50 toxicological effects of ILs with R(2) of 0.593~0.978, and SE of 0.098~0.699 log unit, and the total data set with R(2) of 0.901 and SE of 0.426 log unit.