Synthesis, Biological Activity, and Molecular Dynamics Study of Novel Series of a Trimethoprim Analogs as Multi-Targeted Compounds: Dihydrofolate Reductase (DHFR) Inhibitors and DNA-Binding Agents

Abstract

Eighteen previously undescribed trimethoprim (TMP) analogs containing amide bonds (1-18) were synthesized and compared with TMP, methotrexate (MTX), and netropsin (NT). These compounds were designed as potential minor groove binding agents (MGBAs) and inhibitors of human dihydrofolate reductase (hDHFR). The all-new derivatives were obtained via solid phase synthesis using 4-nitrophenyl Wang resin. Data from the ethidium displacement test confirmed their DNA-binding capacity. Compounds 13-14 (49.89% and 43.85%) and 17-18 (41.68% and 42.99%) showed a higher binding affinity to pBR322 plasmid than NT. The possibility of binding in a minor groove as well as determination of association constants were performed using calf thymus DNA, T4 coliphage DNA, poly (dA-dT)², and poly (dG-dC)². With the exception of compounds 9 (IC50 = 56.05 µM) and 11 (IC50 = 55.32 µM), all of the compounds showed better inhibitory properties against hDHFR than standard, which confirms that the addition of the amide bond into the TMP structures increases affinity towards hDHFR. Derivatives 2, 6, 13, 14, and 16 were found to be the most potent hDHFR inhibitors. This molecular modelling study shows that they interact strongly with a catalytically important residue Glu-30.

Keywords: DHFR inhibitors; MGBAs; drug design; molecular dynamics; netropsin; trimethoprim.

Figure 1

Structure of methotrexate (MTX), trimethoprim (TMP), and netropsin (NT) as model compounds.

Figure 2

Structures (a) R=H (b) R=2,6-dichlorophenyl and (c) of selected TMP analogs directed at new targets.

Figure 3

Structures of previously synthesized TMP analogs (A–F).

Figure 4

Structures of 1–18 novel TMP derivatives.

Scheme 1

Synthesis of TMP analogs on the example of analogue 1. (a) Pyridine, dichloromethane (DCM), 18 h; (b) 1 M SnCl2, dimethylformamide (DMF), 18 h; (c) DCM, 4-dimethylaminopyridine (DMA)P,18 h; (d) TFA:DCM (50:50), 2 h.

Figure 5

(A) Binding modes of derivatives 2, 6 (green) and 13, 14, 16 (blue) at active site of human dihydrofolate reductase (hDHFR). Analogs with the same molecular scaffold overlap. (B) Figure on the right is a close-up of the active site with the main residues involved in interactions highlighted.

Figure 6

Superimposed crystal structures of hDHFR-inhibitor complexes from X-ray diffraction experiments. Polar hydrogen atoms have been added. MTX binding position (PDB: 1U72) [53] is in white, TMP (PDB: 2W3A) [54] is yellow, and orange is one of the propargyl-linked antifolate inhibitors (PDB: 4KD7) reported in work of Lamb et al. [55]. Our inhibitors 2 (green) and 14 (blue) are also shown. Interactions with Glu-30 and Phe-34 are highlighted.

Figure 7

Interactions between molecules (A) 2, (B) 6, (C) 13, (D) 14, (E) 16, and residues in the active site pocket of hDHFR, from molecular docking calculations. Conventional hydrogen bonds lengths are in (Å).

Figure 7

Interactions between molecules (A) 2, (B) 6, (C) 13, (D) 14, (E) 16, and residues in the active site pocket of hDHFR, from molecular docking calculations. Conventional hydrogen bonds lengths are in (Å).

Figure 8

Analysis of RMSD, SASA, RMSF, and Rg of unliganded enzyme (DHFR) and five TMP derivative complexes with protein during 20 ns MD simulations. (a) Root-mean-square deviation (RMSD) for Cα atoms. (b) Solvent accessible surface area (SASA). (c) Root Mean Square Fluctuation (RMSF) values for each residue averaged over the entire simulation. (d) Radius of gyration (Rg).

Figure 9

Number of H-bonds detected during 20 ns molecular dynamics simulation for (a) 2, (b) 6, (c) 13, (d) 14, and (e) 16. Cut-off distance for H-bonds was set to 3.2 Å and 30° angle. Green indicates interactions with Glu-30.