Structure-activity Relationship Study Of Emerging Contaminants In Water

ObjectiveComputational toxicology is a new subject for predicting the toxicity and risk assessment of compounds.The quantitative structure-activity relationship（QSAR）method as a branch of computational toxicology has been applied to the screening and activity prediction of many kinds of compounds.In order to expand the application of QSAR,this study selected the emerging disinfection by-products and microplastics（MPs）as the research object,and used the QSAR method to carry out computational toxicological application research on their protein binding characteristics or environmental toxicological parameters.1.MPs have become an emerging global pollution problem with high abundance and wide distribution in the environment,especially in the oceans,and due to the large amount of plastic production and use and difficult to degrade,resulting in increasingly serious pollution by MPs.MPs are strong carriers of organic pollutants,which can adsorb and enrich organic pollutants,change the distribution and attribution of organic pollutants in the environment,and through MPs are strong carriers of organic pollutants and can adsorb and enrich organic pollutants,change the distribution and fate of organic pollutants in the environment,and enter the human body through the food chain,drinking water and skin contact,which may lead to an increase in the concentration of organic pollutants in the human body and increase their toxic effects,threatening human health.Therefore,in order to gain insight into the sorption of organic pollutants by MPs and their mechanisms,it is proposed to construct a QSAR model of the sorption capacity of polyethylene（PE）MPs in seawater for sorption of organic pollutants to rapidly predict the sorption capacity of PEs and to explore the possible sorption mechanisms.2.The emerging disinfection by-products of drinking water disinfection,halogenated phenols,and organic pollutants emitted during chemical production,halogenated benzenethiophenols,are halogenated aromatic compounds that have attracted widespread attention and research because of their multiple biotoxic effects.Some of these compounds can bind to human thyroxine protein（h TTR）and interfere with the endocrine function of the thyroid gland,endangering human health.Their endocrine disrupting effects are highly correlated with their binding affinity to h TTR,and the chemical structure of compound molecules may be the decisive factor affecting the binding affinity.Therefore,QSAR models for the binding affinity of such organic pollutants and h TTR,and the analysis of the interaction mechanism of their binding,can provide modeling support for rapid prediction,assessment and screening of thyroid-disrupting organic pollutants in the environment.MethodsThirty-five halogenated aromatic compounds bound to h TTR and relative binding affinity data,37 organic pollutants adsorbed on PE-MPs and adsorption capacity data were collected separately by reviewing the literature,and the data sets were organized and formed and randomly divided into training and test sets.Then molecules were constructed,molecular structures were optimized using SYBYL-X,molecular alignment was performed,and 3D-QSAR models were constructed for each of the two components using the comparative molecular field analysis（Co MFA）and comparative molecular similarity index analysis（Co MSIA）methods,and additionally the HQSAR method was used to construct QSAR models of the adsorption capacity of PE-MPs.All QSAR models were rigorously validated internally and externally using various validation parameters to ensure the quality of the models.Contribution plots of the best HQSAR model and contour plots of the best 3D-QSAR model were generated and analyzed,together with the MOLCAD module,to analyze the possible molecular mechanisms involved.Results1.HQSAR and 3D-QSAR models of the adsorption capacity of seawater PE-MPs for organic pollutantsThe q² of the optimal HQSAR,Co MFA and Co MSIA models were 0.924,0.510and 0.806,r² were 0.988,0.996 and 0.983,respectively,and 0.862,0.862 and 0.906,respectively.External validation parameters such asQ_F1² ,Q_F2 ²,Q_F2 ³,were greater than 0.7,and the consistency correlation coefficients CCC were greater than 0.85.The steric and electrostatic field contributions of the Co MFA model are 0.460 and 0.560,respectively.The contribution rates of steric and electrostatic fields are 0.460 and 0.560,respectively,and the contribution rates of the five fields of the best Co MSIA model are14.5%for steric field,29.6%for electrostatic field,38.9%for hydrophobic field,12.3%for hydrogen bond donor field,and 4.7%for hydrogen bond acceptor field,respectively.2.3D-QSAR model of the binding affinity of halogenated aromatic organics to h TTRThe cross-validation coefficient q² of the Co MFA model is 0.525 and the non-cross-validation coefficient r² is 0.997.The q² and r² of the best Co MSIA model are0.679 and 0.994,respectively.The Pearson prediction correlation coefficients of the test set for the Co MFA and Co MSIA models are 0.841 and 0.823,both of which are greater than 0.6.The contributions of the steric and electrostatic fields of the Co MFA model were 38.1%and 61.9%,respectively;the contributions of the five fields of the best Co MSIA model were 4.5%of the steric field,26.3%from the electrostatic field,28.8%of the hydrophobic field,23.3%of the hydrogen bond donor field,and 17.0%of the hydrogen bond acceptor field.Conclusions1.HQSAR,Co MFA and Co MSIA models for the adsorption capacity of seawater PE-MPs for organic pollutant adsorption were successfully constructed with significant correlation,good fitting ability,robust internal and external prediction ability,which can be used to characterize the quantitative relationship of microplastic/water partition coefficients for organic pollutant adsorption by PE microplastics in seawater,and can be used to predict the above-mentioned categories of organic pollutants Log Kd values for PE-MPs adsorption in seawater.The Log Kd values of organic pollutants adsorbed to PE-MPs in seawater are mainly influenced by hydrophobic interactions and,to a lesser extent,by electrostatic fields,and substances with high hydrophobicity have a higher affinity for adsorption to PE-MPs.2.The constructed best Co MFA and Co MSIA models for the binding affinity of halogenated aromatic organics to h TTR have significant correlation,good fitting ability,robust internal and external prediction ability to characterize the quantitative relationship between the relative binding affinity of the studied organic pollutants and h TTR,and can be used to predict the relative binding affinity of the above-mentioned classes of organic pollutants to h TTR.The relative binding affinity of halogenated aromatic organic compounds to h TTR is related to the benzene ring and the substituent groups on the benzene ring,and the position,number,and type of halogen substituents affect the Log RP value.The binding of the two is related to electrostatic interaction,hydrophobic interaction and hydrogen bonding interaction.