Molecular determinants of thyroid hormone receptor selectivity in a series of phosphonic acid derivatives: 3D-QSAR analysis and molecular docking

• Effective QSAR models for agonists of thyroid hormone receptor have been developed.
• Molecular docking was developed to explore the binding features between the agonists and the receptors.
• The results of QSAR models and molecular docking will be helpful in the design of more potent agonists.

A mathematical study was performed on a set of phosphonic acid derivatives that are substrates for thyroid hormone receptor β (TRβ) and thyroid hormone receptor α (TRα), three-dimensional quantitative structure-activity relationship (3D-QSAR) models using comparative molecular field analysis (CoMFA) and comparative molecular similarity indices analysis (CoMSIA) methods were employed to investigate the structural requirements for this series of compounds with improved activity. Some descriptors were also employed to significantly improve the performance of the derived models. The CoMFA model for TRβ exhibited Rcv2 of 0.612, Rpred2 of 0.7218, whereas CoMSIA model showed Rcv2 of 0.621, R2pred of 0.7358; the CoMFA model for TRα displayed Rcv2 of 0.678, Rpred2 of 0.6424, and the CoMSIA model had Rcv2 of 0.671, Rpred2 of 0.6932, which indicate that the constructed models are statistically significant. The derived contour maps further pointed out the regions where interactive fields may influence the activity. In order to validate the QSAR models and explore the origin of the selectivity at the amino acid level, molecular docking was developed, and the results indicate that Arg282, Arg320, Asn331, Gly332, Thr329 and His435 for TRβ, but Ala225, Arg228, Met259, Arg262 and His381 for TRα, respectively are important residues. The information obtained from the QSAR models can be used in the design of more potent TR agonists.