6-Arylpyrido[2,3-d]pyrimidines as novel ATP-competitive inhibitors of bacterial D-alanine:D-alanine ligase

Abstract

Background: ATP-dependent D-alanine:D-alanine ligase (Ddl) is a part of biochemical machinery involved in peptidoglycan biosynthesis, as it catalyzes the formation of the terminal D-ala-D-ala dipeptide of the peptidoglycan precursor UDPMurNAc-pentapeptide. Inhibition of Ddl prevents bacterial growth, which makes this enzyme an attractive and viable target in the urgent search of novel effective antimicrobial drugs. To address the problem of a relentless increase in resistance to known antimicrobial agents we focused our attention to discovery of novel ATP-competitive inhibitors of Ddl.

Methodology/principal findings: Encouraged by recent successful attempts to find selective ATP-competitive inhibitors of bacterial enzymes we designed, synthesized and evaluated a library of 6-arylpyrido[2,3-d]pyrimidine-based compounds as inhibitors of Escherichia coli DdlB. Inhibitor binding to the target enzyme was subsequently confirmed by surface plasmon resonance and studied with isothermal titration calorimetry. Since kinetic analysis indicated that 6-arylpyrido[2,3-d]pyrimidines compete with the enzyme substrate ATP, inhibitor binding to the ATP-binding site was additionally studied with docking. Some of these inhibitors were found to possess antibacterial activity against membrane-compromised and efflux pump-deficient strains of E. coli.

Conclusions/significance: We discovered new ATP-competitive inhibitors of DdlB, which may serve as a starting point for development of more potent inhibitors of DdlB that could include both, an ATP-competitive and D-Ala competitive moiety.

Figure 1. Three-dimensional structure of E. coli Ddl in complex with ADP (in grey sticks) and phosphoryl phosphonate-based inhibitor (PDB entry: 1IOV).

Figure 2. Reaction mechanism of Ddl.

Figure 3. Pyridopyrimidine-based inhibitor 33 of BC based on the protein kinase inhibitor pharmacophore.

Figure 4. Inhibitory activities of pyridopyrimidines toward DdlB.

Figure 5. Inhibitory activities of pyridopyrimidines toward DdlB (continued).

Figure 6. Inhibitory activities of pyridopyrimidines toward DdlB (continued).

Figure 7. Inhibitory activities of pyridopyrimidines toward DdlB (continued).

Figure 8. Reagents and conditions: (a) Na, EtOH, −5°C – room temp, 18

h (b) Ra-Ni, 98–100% HCOOH (c) NaH, EtO(CH₂)₂OH, reflux, 4 h. (d) NaH, R₂NCO/R₂NCS, DMF, room temp, 18 h. (e) R³NH₂, NH₂SO₃H, reflux, 42–72 h. (f) NaH, R₂NCO/R₂NCS, DMF, room temp, 18 h.

Figure 9. Reagents and conditions: (a) NaBH₄, MeOH, room temp, 2

h. (b) CBr₄, PPh₃, THF, room temp, 16 h. (c) NBS, Bz₂O₂, CCl₄, reflux, 4 h. (d) KCN, EtOH/H₂O  = 4/1, reflux, 4 h.

Figure 10. SPR analysis of compounds binding to DdlB.

Different concentrations of compounds were tested for the binding (left panels in all cases). The binding curves (right panels) were generated by ploting steady-state response levels, i.e. at the end of the association phase, vs concentration of the injected compound. The KDs were obtained from fitting the data to the steady-state affinity model and are reported in Table 1. For each compound 3–5 independent titrations were performed.

Figure 11. Calorimetric binding isotherms (▪) and the corresponding best fit model functions (–) for binding of a) 14 and b) 33 to DdlB at 37°C. r is the bound ligand to total protein concentration ratio.

Figure 12. Superposition of ADP (PDB entry: 1IOV, in yellow lines) and docked pose of inhibitor a) 14 (in grey sticks) and c) 33 (in grey sticks) in the E. coli DdlB active site (in green).

Schematic representation of interactions between b) 14 or d) 33 and E. coli DdlB active site residues as generated by PoseViewWeb .