Search

The evaluation and prediction of mixture toxicity is a hot topic of toxicology and environmental sciences. The Concentration Addition (CA) and Independent action (IA) models are the primary reference models. CA is suitable for predicting the toxicity of the mixture where the components have same mode of action (MOA) and the IA for toxicity of mixture where components have different MOA. Nonetheless, it is very difficult to obtain the information about the MOA of chemicals. Therefore, it is important to explore a feasible and efficient method to identify MOA. In this study, to reveal the mechanisms of toxic action of ionic liquids (ILs) to firefly luciferase and further build the prediction model for mixture toxicity, luciferase is treated as the biomolecular receptor and four 1-alkyl- 3-methyl-imidzole chloride ILs ([Cnmim]Cl, ILn, n =2, 6, 8, 12) are treated as the ligands. The binding modes of ligands (ionic liquids) and receptor (luciferase) are identified by molecular docking and molecular dynamics simulations. Molecular dynamics simulations revealed that the imidazolium ring of [C2mim] is bound at the bottom of the luciferin (D-LH2) pocket which is surrounded by Arg218, Phe247 Thr251, Leu286, Arg337, Gly339 and Ile351 and the alkyl-chain extends from the bottom of the pocket to the entrance. The [C6mim] is bound at the pocket which is surrounded by Phe247, Thr251, Gly315, Gly339, Leu342, Ala348 and Ile351. In contrast, the imidazolium ring of [C8mim] and [C12mim] is bound at the entrance of the D-LH2 pocket and the alkyl-chain inserts into the bottom of the D-LH2 pocket, surrounded by Gly200, Arg218, His245, Phe247, Thr251, Gly316, Arg337, Thr343, Leu526 and Thr527. According to the information of the binding site, binding pose, hydrogen bonds and hydrophobic contacts, we can determine whether the ILs have similar binding pattern (BP). The results show that [C2mim]Cl belongs to BP1 and [C6mim]Cl belongs to BP2. [C8mim]Cl and [C12mim]Cl belong to BP3. We used the firefly luciferase-microplate toxicity analysis to determine the luminescence inhibition effect of the four ILs on luciferase and direct equipartition ray design (EquRay) to design 18 binary mixture rays. We evaluated the toxicities of the mixtures by CA and IA. The results show that the mixture toxicities of [C2mim]Cl-[C8mim]Cl and [C2mim]Cl-[C12mim]Cl where components have dissimilar BP can be predicted by IA. The mixture toxicities of [C2mim]Cl-[C6mim]Cl and [C8mim]Cl-[C12mim]Cl where components have similar BP could be predicted by CA. This study demonstrates that CA and IA are suitable for predicting the toxicity of mixtures according to the different BPs of the individual ILs to luciferase.

Experimental and epidemiological studies suggest that some man-made and naturally occurring chemicals related to the environment have the potential to interrupt normal functioning of the endocrine systems of humans and wildlife. These chemicals, termed EDCs (Endocrine disrupting Chemicals), pose serious threats to the reproductive capability of humans and wildlife. Because of the structural diversity and various types, development of structure-based rapid screening methodologies is important and necessary for the assessment of the environmental pollutants. In this paper molecular hologram based QSAR models were developed with the combinatory application of partial least square (PLS) regression for a large diverse set of 105 environmental estrogens. Quantitatively predictive models were developed based on only molecular structures, which can be used for the accurate prediction of estrogenicity to rapidly screen potential environmental endocrine disrupting chemicals.

Nitroaromatics are typical toxic organic pollutants and are ubiquitous in environment with diverse structures. They are widely used in many industries and formed during many natural and anthropogenic processes. Most of these pollutants are potentially carcinogenic and the assessment and prediction of the mutagenicity of nitroaromatics are of great interest. In this paper the structure-mutagenicity relationships of 219 nitroaromatics are investigated by molecular orbital theory based classic structure-activity relationships and comparative molecular field analysis (CoMFA). A comparison is undertaken in respect of the interpretation of mechanism and predictive ability for these two categories of QSAR approaches and highly predictive QSAR models have been developed.