Investigation of the structure requirement for 5-HT₆ binding affinity of arylsulfonyl derivatives: a computational study

Abstract

5-HT(6) receptor has been implicated in a series of diseases including anxiety, depression, schizophrenia and cognitive dysfunctions. 5-HT(6) ligands have been reported to play a significant role in the treatment for central nervous system (CNS) diseases. Presently, a large series of 223 5-HT(6) ligands were studied using a combinational method by 3D-QSAR, molecular docking and molecular dynamics calculations for further improvement of potency. The optimal 3D models exhibit satisfying statistical results with r(2) (ncv), q(2) values of 0.85 and 0.50 for CoMFA, 0.81 and 0.53 for CoMSIA, respectively. Their predictive powers were validated by external test set, showing r(2) (pred) of 0.71 and 0.76. The contour maps also provide a visual representation of contributions of steric, electrostatic, hydrophobic and hydrogen bond fields as well as the prospective binding models. In addition, the agreement between 3D-QSAR, molecular docking and molecular dynamics simulation proves the rationality of the developed models. These results, we hope, may be helpful in designing novel and potential 5-HT(6) ligands.

Keywords: 3D-QSAR; 5-HT6; CoMFA; CoMSIA; molecular dynamics.

Figure 1.

The correlation plots of the predicted versus the actual pK_i values using the training set (black dot) and the test set (red asterisk) based on the (A) CoMFA and (B) CoMSIA of ligand-based model.

Figure 2.

CoMFA StDev*Coeff contour plots for 5-HT₆ receptor ligands in combination with compound 198. (A) Steric (green/yellow) contour map. Green contours indicate regions where bulky groups increase activity (favored level 80%); yellow contours indicate regions where bulky groups decrease activity (disfavored level 20%); (B) Electrostatic contour map (blue/red). Blue contours indicate regions where positive charges increase activity (favored level 80%); red contours indicate regions where negative charges increase activity (disfavored level 20%).

Figure 3.

Ligand-based CoMSIA StDev*Coeff contour plots combined with compound 132. (A) Steric (green/yellow) contour map. Green contours indicate regions where bulky groups increase activity (favored level 80%); yellow contours indicate regions where bulky groups decrease activity (disfavored level 20%); (B) Electrostatic contour map (blue/red). Blue contours indicate regions where positive charges increase activity (favored level 80%); red contours indicate regions where negative charges increase activity (disfavored level 20%); (C) Hydrophobic contour map (yellow/white). Yellow contours indicate regions where hydrophobic substituents enhance activity (favored level 80%); white contours indicate regions where hydrophobic substituents decrease activity (disfavored level 20%); (D) CoMSIA contour maps illustrating hydrogen-bond acceptor features (magenta/red). The magenta contour indicates regions where H-bond acceptor groups increase activity (favored level 80%); red contour indicates the disfavored region for H-bond acceptor groups (disfavored level 20%).

Figure 4.

Homology modeling results. (A) Sequence alignment of 2RH1 (Chain A) and the 5-HT₆ receptor model, where the important transmembrane domains (TM1-TM7) are marked in red frame; (B) Superposition of the template protein 2RH1 chain A (red ribbon) and the 5-HT₆ receptor model structure (yellow ribbon) from homology modeling. Compound 198 with bottle-green stick is binding into the pocket.

Figure 5.

The binding result of compound 198 with 5-HT₆. The ligand are colored in bottle-green and key amino acid residues in red labels around compound 198 within 4 Å. H-bonds are shown in yellow dash lines.

Figure 6.

MD results. (A) RMSD of the docked complex versus MD simulation time in the MD-simulated structure; (B) Superimposition of backbone atoms in the average structure of MD simulations (light-blue) and the initial docking structure (bottle-green) for compound 198-5-HT₆ complex. Herein, Compound 198 is represented as carbon-chain in bottle-green for the docking complex and light-blue for the average complex, respectively.

Figure 7.

Alignment of all molecules in the dataset. (A) Compound 198 is used as the template. The common substructure (shown in bold) for the alignment of all molecules; (B) Ligand-based; and (C) receptor-based alignment of all the compounds.