In silico quantitative structure-activity relationship studies on P-gp modulators of tetrahydroisoquinoline-ethyl-phenylamine series

Abstract

Background: Multidrug resistance (MDR) is a major obstacle in cancer chemotherapy. The drug efflux by a transport protein is the main reason for MDR. In humans, MDR mainly occurs when the ATP-binding cassette (ABC) family of proteins is overexpressed simultaneously. P-glycoprotein (P-gp) is most commonly associated with human MDR; it utilizes energy from adenosine triphosphate (ATP) to transport a number of substrates out of cells against concentration gradients. By the active transport of substrates against concentration gradients, intracellular concentrations of substrates are decreased. This leads to the cause of failure in cancer chemotherapy.

Results: Herein, we report Topomer CoMFA (Comparative Molecular Field Analysis) and HQSAR (Hologram Quantitative Structure Activity Relationship) models for third generation MDR modulators. The Topomer CoMFA model showed good correlation between the actual and predicted values for training set molecules. The developed model showed cross validated correlation coefficient (q2) = 0.536 and non-cross validated correlation coefficient (r2) = 0.975 with eight components. The best HQSAR model (q2 = 0.777, r2 = 0.956) with 5-8 atom counts was used to predict the activity of test set compounds. Both models were validated using test set compounds, and gave a good predictive values of 0.604 and 0.730.

Conclusions: The contour map near R1 indicates that substitution of a bulkier and polar group to the ortho position of the benzene ring enhances the inhibitory effect. This explains why compounds with a nitro group have good inhibitory potency. Molecular fragment analyses shed light on some essential structural and topological features of third generation MDR modulators. Fragments analysis showed that the presence of tertiary nitrogen, a central phenyl ring and an aromatic dimethoxy group contributed to the inhibitory effect. Based on contour map information and fragment information, five new molecules with variable R1 substituents were designed. The activity of these designed molecules was predicted by the Topomer CoMFA and HQSAR models. The novel compounds showed higher potency than existing compounds.

Figure 1

Fragmentation pattern (R1 and R2) for all molecules of dataset in topomer CoMFA analysis. R1 fragment is represented by the blue color and R2 fragment is denoted by the red color.

Figure 2

Scatter plot diagram for Topomer CoMFA analysis. Predicted versus actual activity of the training set (blue diamond) and the test set (red square) compounds.

Figure 3

Topological alignment of R1 and R2 fragments generated by Topomer CoMFA analysis.

Figure 4

Steric and electrostatics stdev* coefficient contour map for compound 28 by Topomer CoMFA analysis. (A) Steric contour map for the R1 fragment. (B) Electrostatic contour map for the R1 fragment. (C) Steric contour map for the R2 fragment. (D) Electrostatics contour map for the R2 fragment. Sterically favoured/unfavoured areas are shown in green/yellow contour, while the blue/red polyhedra depict the favourable sites for the positively/negatively charged groups.

Figure 5

Steric and electrostatics stdev* coefficient contour map for compound 30a by Topomer CoMFA analysis. (A) Steric contour map for the R1 fragment. (B) Electrostatic contour map for the R1 fragment. (C) Steric contour map for the R2 fragment. (D) Electrostatics contour map for the R2 fragment. Sterically favoured/unfavoured areas are shown in green/yellow contour, while the blue/red polyhedra depict the favourable site for the positively/negatively charged groups.

Figure 6

Steric and electrostatics stdev* coefficient contour map for compound 13 by Topomer CoMFA analysis. (A) Steric contour map for the R1 fragment. (B) Electrostatic contour map for the R1 fragment. (C) Steric contour map for the R2 fragment. (D) Electrostatics contour map for the R2 fragments. Sterically favoured/unfavoured areas are shown in green/yellow contour, while the blue/red polyhedra depict the favourable site for the positively/negatively charged groups.

Figure 7

Steric and electrostatics stdev* coefficient contour map for compound 5b by Topomer CoMFA analysis. (A) Steric contour map for the R1 fragment. (B) Electrostatic contour map for the R1 fragment. (C) Steric contour map for the R2 fragment. (D) Electrostatics contour map for the R2 fragment. Sterically favoured/unfavoured areas are shown in green/yellow contour, while the blue/red polyhedra depict the favourable site for the positively/negatively charged groups.

Figure 8

Scatter plot diagram for HQSAR analysis. The scatter plot displays the predicted versus actual activities of the training set (blue diamond) and the test set (red square) of compounds.

Figure 9

Positive and negative contribution of some fragments towards P-gp antagonism, obtained by HQSAR analysis. Grey = carbon atoms, where C2, C3 and Car indicates sp², sp³ and aromatic carbon. Red = oxygen atom, where O2 and O3 indicates sp² and sp³ oxygen, respectively. Blue = nitrogen atom, where N3, Nar indicates sp³ and aromatic nitrogen. Blue = ANY atom, and it may be hydrogen, carbon or oxygen.

Figure 10

Steric and electrostatics stdev* coefficient contour map for designed compound 28B by Topomer CoMFA analysis. (A) Steric contour map for the R1 fragment. (B) Electrostatic contour map for the R1 fragment. (C) Steric contour map for the R2 fragment. (D) Electrostatics contour map for the R2 fragment. Sterically favoured/unfavoured areas are shown in green/yellow contour, while the blue/red polyhedra depict the favourable site for the positively/negatively charged groups.