Correlations of inhibitor kinetics for Pneumocystis jirovecii and human dihydrofolate reductase with structural data for human active site mutant enzyme complexes

Abstract

To understand the role of specific active site residues in conferring selective dihydrofolate reductase (DHFR) inhibition from pathogenic organisms such as Pneumocystis carinii (pc) or Pneumocystis jirovecii (pj), the causative agent in AIDS pneumonia, it is necessary to evaluate the role of these residues in the human enzyme. We report the first kinetic parameters for DHFR from pjDHFR and pcDHFR with methotrexate (MTX), trimethoprim (TMP), and its potent analogue, PY957. We also report the mutagenesis and kinetic analysis of active site mutant proteins at positions 35 and 64 of human (h) DHFR and the crystal structure determinations of hDHFR ternary complexes of NADPH and PY957 with the wild-type DHFR enzyme, the single mutant protein, Gln35Lys, and two double mutant proteins, Gln35Ser/Asn64Ser and Gln35Ser/Asn64Phe. These substitutions place into human DHFR amino acids found at those sites in the opportunistic pathogens pcDHFR (Q35K/N64F) and pjDHFR (Q35S/N64S). The K(i) inhibition constant for PY957 showed greatest potency of the compound for the N64F single mutant protein (5.2 nM), followed by wild-type pcDHFR (K(i) 22 nM) and then wild-type hDHFR enzyme (K(i) 230 nM). Structural data reveal significant conformational changes in the binding interactions of PY957 in the hDHFR Q35S/N64F mutant protein complex compared to the other hDHFR mutant protein complexes and the pcDHFR ternary complex. The conformation of PY957 in the wild-type DHFR is similar to that observed for the single mutant protein. These data support the hypothesis that the enhanced selectivity of PY957 for pcDHFR is in part due to the contributions at positions 37 and 69 (pcDHFR numbering). This insight will help in the design of more selective inhibitors that target these opportunistic pathogens.

Figure 1

Schematic representation of antifolates under study.

Figure 2

Fig. 2. (left) 2Fo-Fc difference electron density map (0.8σ, blue; 3σ, green) showing the fit of the inhibitor PY957 and NADPH in the active site of Q35S/N64F human DHFR ternary complex with NADPH and PY957. The map was phased using only the enzyme hDHFR as the search model. (right) Omit map from final refinement showing Fo-Fc density for PY957 (3σ, green). Fig. 2c. 2Fo-Fc difference electron density map (0.8σ, blue) showing the fit of the inhibitor PY957 in the active site of Q35S/N64F double mutant of human DHFR ternary complex with NADPH and PY957. Model of PY957 (green) in expected conformation interacting with Arg70 reveals negative (−3σ, red) density indicating that the side chain does not fit in this position. The correct model (yellow) indicates that water molecules interact with Arg70 while the side chain is folded such that it is between Phe31 and Phe64.

Figure 2

Fig. 2. (left) 2Fo-Fc difference electron density map (0.8σ, blue; 3σ, green) showing the fit of the inhibitor PY957 and NADPH in the active site of Q35S/N64F human DHFR ternary complex with NADPH and PY957. The map was phased using only the enzyme hDHFR as the search model. (right) Omit map from final refinement showing Fo-Fc density for PY957 (3σ, green). Fig. 2c. 2Fo-Fc difference electron density map (0.8σ, blue) showing the fit of the inhibitor PY957 in the active site of Q35S/N64F double mutant of human DHFR ternary complex with NADPH and PY957. Model of PY957 (green) in expected conformation interacting with Arg70 reveals negative (−3σ, red) density indicating that the side chain does not fit in this position. The correct model (yellow) indicates that water molecules interact with Arg70 while the side chain is folded such that it is between Phe31 and Phe64.

Fig. 3

Stereo view of the superposition of the hDHFR NADPH PY957 ternary complexes with: Q35K (violet), the Q35S/N64S double mutant (cyan), the Q35S/N64F double mutant (green), wild type hDHFR (pink), and pcDHFR wild type NADPH PY957 ternary complex (yellow) (16). The wild type structure has nearly the same conformation as the single mutant protein. Figure was drawn with PyMol (33).

Fig. 4

Stereo view of the interactions of PY957 side chain in the active site of the double mutant protein Q35S/N64F of hDHFR illustrating the hydrophobic contacts made between Phe31 and Phe64 of the mutant protein and the methylene carbons of the PY957 side chain. This conformation has the side chain flipped such that the carboxylate oxygen atoms no longer interact with Arg70 as observed in the other PY957 structures.