Modeling and predicting pKa values of mono-hydroxylated polychlorinated biphenyls (HO-PCBs) and polybrominated diphenyl ethers (HO-PBDEs) by local molecular descriptors

Prediction study on the distribution of polycyclic aromatic hydrocarbons and their halogenated derivatives in the atmospheric particulate phase

Polycyclic aromatic hydrocarbons (PAHs) and their halogenated derivatives (X-PAHs), which generally produced from photochemical and thermal reactions of parent PAHs, widely exist in the environment. They are semi-volatile organic chemicals (SVOCs) and the partitioning between gas/particulate phases affects their environmental migration, transformation and fate, which further impacts their toxicity and health risk to human. However, there is a large data missing of the experimental distribution ratio in the atmospheric particulate phase (f), especially for X-PAHs. In this study, we first checked the correlation between experimental f values of 53 PAH derivatives and their octanol-air partitioning coefficients (log K_OA), which is frequently used to characterize the distribution of chemicals in organic phase, and yielded R² = 0.803. Then, quantum chemical descriptors derived from molecular structural optimization by M06-2X/6-311 +G (d,p) method were further employed to develop Quantitative Structure-Property Relationship (QSPR) model. The model contains two descriptors, the average molecular polarizability (α) and the equilibrium parameter of molecular electrostatic potential (τ), and yields better performance with R² = 0.846 and RMSE = 0.122. The mechanism analysis and validation results by different strategies prove that the model can reveal the molecular properties that dominate the distribution between gas and particulate phases and it can be used to predict f values of other PAHs/X-PAHs, providing basic data for their environmental ecological risk assessment.

Hydroxylated Polychlorinated Biphenyls Are Emerging Legacy Pollutants in Contaminated Sediments

We measured the concentrations of 837 hydroxylated polychlorinated biphenyls (OH-PCBs, in 275 chromatographic peaks) and 209 polychlorinated biphenyls (PCBs, in 174 chromatographic peaks) in sediments from New Bedford Harbor in Massachusetts, Altavista wastewater lagoon in Virginia, and the Indiana Harbor and Ship Canal in Indiana, USA and in the original commercial PCB mixtures Aroclors 1016, 1242, 1248, and 1254. We used the correlation between homologues and the peak responses to quantify the full suite of OH-PCBs including those without authentic standards available. We found that OH-PCB levels are approximately 0.4% of the PCB levels in sediments and less than 0.0025% in Aroclors. The OH-PCB congener distributions of sediments are different from those of Aroclors and are different according to sites. We also identified a previously unknown compound, 4-OH-PCB52, which together with 4'-OH-PCB18 made up almost 30% of the OH-PCBs in New Bedford Harbor sediments but less than 1.2% in the Aroclors and 3.3% in any other sediments. This indicates site-specific environmental transformations of PCBs to OH-PCBs. We conclude that the majority of OH-PCBs in these sediments are generated in the environment. Our findings suggest that these toxic breakdown products of PCBs are prevalent in PCB-contaminated sediments and present an emerging concern for humans and ecosystems.

Polarizability and aromaticity index govern AhR-mediated potencies of PAHs: A QSAR with consideration of freely dissolved concentrations

Polycyclic aromatic hydrocarbons (PAHs) are ubiquitous environmental pollutants associated with adverse human effects including cancer, and the aryl hydrocarbon receptor (AhR) is a key ligand-activated transcription factor mediating their toxicity. However, there is presently a lack of data on AhR potencies of PAHs. Simple, transparent, interpretable and predictive quantitative structure-activity relationship (QSAR) models are helpful, especially with the consideration of freely dissolved concentrations linked to bioavailability. Here, QSAR models on AhR-mediated luciferase activity of PAHs were developed with nominal median effect concentrations (EC₅₀, _nom) and freely dissolved concentration (EC_{50, free}) as endpoints, and quantum chemical and Dragon descriptors as predictor variables. Results indicated that only the EC_{50, free} model met the acceptable criteria of QSAR model (determination coefficient (R²) > 0.600, leave-one-out cross validation (Q_LOO²) > 0.500, and external validation coefficient (Q_EXT²) > 0.500), implying that it has good goodness-of-fit, robustness and external predictive power. Molecular polarizability and aromaticity index reflecting the partition behavior and intermolecular interactions can effectively predict AhR-mediated potencies of PAHs. The results highlight the necessity of adoption of the freely dissolved concentration in the QSAR modeling and more in silico models need to be further developed for different animal models (in vivo or in vitro).

QSPR models for predicting the adsorption capacity for microplastics of polyethylene, polypropylene and polystyrene

Microplastics have become an emerging concerned global environmental pollution problem. Their strong adsorption towards the coexisting organic pollutants can cause additional environmental risks. Therefore, the adsorption capacity and mechanisms are necessary information for the comprehensive environmental assessments of both microplastics and organic pollutants. To overcome the lack of adsorption information, five quantitative structure–property relationship (QSPR) models were developed for predicting the microplastic/water partition coefficients (log K_d) of organics between polyethylene/seawater, polyethylene/freshwater, polyethylene/pure water, polypropylene/seawater, and polystyrene/seawater. All the QSPR models show good fitting ability (R² = 0.811–0.939), predictive ability (Q²_ext = 0.835–0.910, RMSE_ext = 0.369–0.752), and robustness (Q_cv² = 0.882–0.957). They can be used to predict the K_d values of organic pollutants (such as polychlorinated biphenyls, chlorobenzene, polycyclic aromatic hydrocarbons, antibiotics perfluorinated compounds, etc.) under different pH conditions. The hydrophobic interaction has been indicated as an important mechanism for the adsorption of organic pollutants to microplastics. In sea waters, the role of hydrogen bond interaction in adsorption is considerable. For polystyrene, π–π interaction contributes to the partitioning. The developed models can be used to quickly estimate the adsorption capacity of organic pollutants on microplastics in different types of water, providing necessary information for ecological risk studies of microplastics.

Understanding and predicting the diffusivity of organic chemicals for diffusive gradients in thin-films using a QSPR model

The diffusion coefficient (D) is a key physicochemical parameter for the diffusive gradients in thin films technique (DGT) for environmental sampling, which has been extended to organic chemicals (so called o-DGT). D can be measured in the laboratory, although for organic chemicals this parameter might be predicted based on chemical structure. Here we developed for the first time a Quantitative Structure-Property Relationship (QSPR) model to predict the D values. Twenty quantum chemical descriptors that quantify the electronic and energy properties of 120 organic compounds were selected together with molecular mass, solubility and hydrophobicity. The best QSPR model was established by using genetic algorithm and multiple linear regression (GA-MLR). The results indicated that the model derived from the average molecular polarizability (α), the chemical potential (ξ) and the global electrophilicity index (ω) could explain the diffusion of organics in o-DGT and had good statistical performance (R² = 0.767, RMSE = 0.101). Different validation strategies confirmed that the developed model was robust and predictive. 93% of tested compounds were within the applicability domain (AD) and predicted accurately. We concluded that the proposed QSPR model can serve as an efficient predictive tool for new chemicals in the AD, would be useful to cross validate measured D values and provide a better the understanding of the diffusive behaviour of organics in o-DGT and measurements in the environment. It might also be useful in the non-target analysis with o-DGT for chemicals without measured D values.

Prediction of pKa Values for Neutral and Basic Drugs based on Hybrid Artificial Intelligence Methods

The pKa value of drugs is an important parameter in drug design and pharmacology. In this paper, an improved particle swarm optimization (PSO) algorithm was proposed based on the population entropy diversity. In the improved algorithm, when the population entropy was higher than the set maximum threshold, the convergence strategy was adopted; when the population entropy was lower than the set minimum threshold the divergence strategy was adopted; when the population entropy was between the maximum and minimum threshold, the self-adaptive adjustment strategy was maintained. The improved PSO algorithm was applied in the training of radial basis function artificial neural network (RBF ANN) model and the selection of molecular descriptors. A quantitative structure-activity relationship model based on RBF ANN trained by the improved PSO algorithm was proposed to predict the pKa values of 74 kinds of neutral and basic drugs and then validated by another database containing 20 molecules. The validation results showed that the model had a good prediction performance. The absolute average relative error, root mean square error, and squared correlation coefficient were 0.3105, 0.0411, and 0.9685, respectively. The model can be used as a reference for exploring other quantitative structure-activity relationships.

Mechanistic insights into alginate fouling caused by calcium ions based on terahertz time-domain spectra analyses and DFT calculations

Fouling mechanisms underlying the filtration behaviors of alginate solution caused by calcium addition were investigated by Terahertz time-domain spectroscopy (THz-TDS) and density functional theory (DFT) techniques. Filtration tests showed that specific filtration resistance (SFR) of alginate solution (0.75 g L^-1) monotonously increased with calcium addition at a relatively low range of calcium concentration (0-1.0 mM), and SFR (2.61 × 10¹⁵ m kg^-1) of alginate solution with 1.0 mM calcium addition was extremely high as compared with sludge suspension. Characterizations by X-ray photoelectric spectroscopy (XPS), Fourier transform infrared spectroscopy (FTIR) and Thermogravimetric analysis (TGA) showed that the composition of functional groups, elements and thermal stability of alginate was not apparently affected by calcium concentration. Howbeit, THz-TDS spectra showed that calcium addition caused structural variation of alginate polymer in solution. DTF calculation results showed that initial binding of alginate chains induced by calcium ions preferentially occurred in intermolecular other than intramolecular, and moreover, the two alginate chains bridged by a calcium atom tend to stretch in a tetrahedron structure (cross to each other) other than parallel to each other. According to these results, "chemical potential gap" depicted by Flory-Huggins theory was suggested to be responsible for the filtration behaviors of alginate solution caused by calcium addition. This study provided the mechanistic insights into membrane fouling.

Developing predictive models for toxicity of organic chemicals to green algae based on mode of action

Organic chemicals in the aquatic ecosystem may inhibit algae growth and subsequently lead to the decline of primary productivity. Growth inhibition tests are required for ecotoxicological assessments for regulatory purposes. In silico study is playing an important role in replacing or reducing animal tests and decreasing experimental expense due to its efficiency. In this work, a series of theoretical models was developed for predicting algal growth inhibition (log EC₅₀) after 72 h exposure to diverse chemicals. In total 348 organic compounds were classified into five modes of toxic action using the Verhaar Scheme. Each model was established by using molecular descriptors that characterize electronic and structural properties. The external validation and leave-one-out cross validation proved the statistical robustness of the derived models. Thus they can be used to predict log EC₅₀ values of chemicals that lack authorized algal growth inhibition values (72 h). This work systematically studied algal growth inhibition according to toxic modes and the developed model suite covers all five toxic modes. The outcome of this research will promote toxic mechanism analysis and be made applicable to structural diversity.

In silico investigation of gas/particle partitioning equilibrium of polybrominated diphenyl ethers (PBDEs)

Polybrominated diphenyl ethers (PBDEs), a group of typical brominated flame retardants (BFRs), have drawn an increasing concern due to their widespread manufacture, usage and disposal around the world and the frequent detection in a variety of environmental media. In the present study, we investigated the molecular mechanism of the partitioning equilibrium of PBDEs between gas and atmospheric particles, and developed a new temperature-dependent predictive model for the gas/particle partition coefficient (K_P) of these chemicals. Quantum chemical computations were implemented at B3LYP/6-31G (d,p) level of theory based on the neutral electronic ground state of PBDE congeners by Gaussian 09 software package. The model performance was assessed by different validation strategies and the application domain was defined by Williams Plot. Mechanism analysis indicated that the interactions of dispersion, electrostatic and hydrogen bond play crucial roles in the partitioning of PBDEs between the two phases. The developed model can be used to estimate the K_P values of PBDEs for which experimental measurements are restricted. Therefore, this work provides an alternative method in a regulatory context of PBDEs.

Understanding the microscopic binding mechanism of hydroxylated and sulfated polybrominated diphenyl ethers with transthyretin by molecular docking, molecular dynamics simulations and binding free energy calculations

Polybrominated diphenyl ethers (PBDEs), one typical type of persistent environmental contaminant, have toxicological effects such as disrupting thyroid homeostasis in the human body. The high binding affinities of hydroxylated metabolites of PBDEs (OH-PBDEs) with transthyretin (TTR) were considered to be one major reason for their extraordinary capacity of passing through the blood-brain barrier via competitive thyroid hormone (T4) transport protein binding. Recent findings showed that sulfated PBDEs can be formed in human liver cytosol as phase-II metabolites. However, experimentally determined data for the TTR binding potential of the sulfated PBDEs are still not available. Therefore, molecular docking and molecular dynamics (MD) simulations were employed in the present study to probe the molecular basis of TTR interacting with hydroxylated and sulfated PBDEs at the atomic level. The docking scores of LeDock were used to construct the structure-based predictive model. The calculated results showed that the sulfated PBDEs have stronger affinity for TTR than the corresponding OH-PBDEs. Further analysis of structural characteristics based on MD simulations indicated that upon the binding of PBDE metabolites, the stability of TTR was enhanced and the dissociation rate of the tetrameric protein structure was potentially decreased. Subsequent binding free energy calculations implied that van der Waals interactions are the dominant forces for the binding of these metabolites of PBDEs at the T4 site of TTR. The residues Ser117/Ser117' and Lys15/Lys15' were identified, by both residue energy decomposition and computational alanine-scanning mutagenesis methods, as key residues which play an important role in determining the binding orientations of the -OSO₃^- group of sulfated PBDEs by formation of either hydrogen bonds or electrostatic interactions, respectively. In general, the combination of docking calculations with MD simulations provided a theoretically toxicological assessment for the metabolites of PBDEs, deep insight into the recognition mechanism of TTR for these compounds, and thus more comprehensive understanding of the thyroid-related toxic effects of PBDEs as well.