The Plasmodium falciparum malaria M1 alanyl aminopeptidase (PfA-M1): insights of catalytic mechanism and function from MD simulations

Abstract

Malaria caused by several species of Plasmodium is major parasitic disease of humans, causing 1-3 million deaths worldwide annually. The widespread resistance of the human parasite to current drug therapies is of major concern making the identification of new drug targets urgent. While the parasite grows and multiplies inside the host erythrocyte it degrades the host cell hemoglobin and utilizes the released amino acids to synthesize its own proteins. The P. falciparum malarial M1 alanyl-aminopeptidase (PfA-M1) is an enzyme involved in the terminal stages of hemoglobin digestion and the generation of an amino acid pool within the parasite. The enzyme has been validated as a potential drug target since inhibitors of the enzyme block parasite growth in vitro and in vivo. In order to gain further understanding of this enzyme, molecular dynamics simulations using data from a recent crystal structure of PfA-M1 were performed. The results elucidate the pentahedral coordination of the catalytic Zn in these metallo-proteases and provide new insights into the roles of this cation and important active site residues in ligand binding and in the hydrolysis of the peptide bond. Based on the data, we propose a two-step catalytic mechanism, in which the conformation of the active site is altered between the Michaelis complex and the transition state. In addition, the simulations identify global changes in the protein in which conformational transitions in the catalytic domain are transmitted at the opening of the N-terminal 8 Å-long channel and at the opening of the 30 Å-long C-terminal internal chamber that facilitates entry of peptides to the active site and exit of released amino acids. The possible implications of these global changes with regard to enzyme function are discussed.

Figure 1. Structure of PfA-M1 aminopeptidase N.

A. Ribbon diagram of PfA-M1 (PDB code 3EBH) coloured by domain: I (blue), II (green), III (yellow) and IV (red). Zinc ion depicted as purple sphere. Arrow indicates position of C-terminal channel opening. B. Catalytic domain II, with α-helices coloured red, loops green and β-strands yellow. Active site residues are shown in stick form with carbon cyan, oxygen red and nitrogen blue. Zinc ion depicted as purple sphere. The N-terminal lobe consisting of a 5-stranded β-sheet and α-helices 1–2 is to the right, the C-terminal lobe comprising α-helices 3–8 is to the left. The active site occurs in the cleft between the N- and C-terminal lobes. C. Ligand-bound active site. Active site of the starting structure of the ligand-bound complex used in the simulation. Ligand and active site residues are shown in stick form with carbon green (ligand) or yellow (PfA-M1), oxygen red and nitrogen blue. The nucleophilic water is labelled “W1” and the ligand N-terminal amino nitrogen “N”. Hydrogen bonds and metallo bonds are shown as red and purple dotted lines, respectively.

Figure 2. Approach profile of the nucleophilic water and Tyr-580 hydroxyl oxygens to the ligand carbonyl carbon.

Smoothed histogram plots of distances to the ligand carbonyl carbon measured in 0.1 Å bins every 50 ps over the 75 ns simulation of the ligand-bound complex. Solid line, distances to nucleophilic water oxygen; dashed line, distances to Tyr-580 hydroxyl oxygen.

Figure 3. Changes in interactions of active site residues in the ligand-bound complex.

Diagrams of the final frame (t = 75 ns) from the simulation of the ligand-bound complex. Ligand and active site residue sidechains are shown in stick form with carbon green (ligand) or light grey (PfA-M1), oxygen red, nitrogen blue and sulphur orange. Zn ion depicted as a purple sphere. Metallo bonds shown as purple dotted lines and hydrogen bonds as red dotted lines. A. Interaction of Glu-497 with Asn-468. B. Interaction of Glu-526 with His-496 and Arg-489. C. Network of interactions with the ligand N-terminal amino nitrogen and Tyr-575. GAMEN motif residues coloured yellow. D. Interaction of GAMEN backbone amide of Gly-460 and main-chain carbonyl oxygen of Ala-461 with the ligand. The Gly-460 amide-P1′ carbonyl oxygen hydrogen bond, present in the crystal structure, is lost due to conformational changes in the loop immediately N-terminal to Gly-460. GAMEN residues shown in stick form with carbon atoms coloured yellow.

Figure 4. Scheme for the catalytic mechanism of M1 aminopeptidases.

Figure 5. Conformational changes in the ligand-bound complex.

Structural alignments of frames at t = 0 and t = 75 ns from the simulation of the ligand-bound complex. Frames aligned using Cα coordinates of domain II. Ligand shown in stick form with carbon cyan, oxygen red and nitrogen blue. Zn ion depicted as a purple sphere. Secondary structural elements as indicated. A. Catalytic domain II (t = 0 green, t = 75 red). C-terminal of α-helix 1 is indicated by the arrow. B. Small N-terminal channel opening. Catalytic domain II (t = 0 green, t = 75 red). Domain IV (t = 0 tan, t = 75 mauve). Domain I, which would partially occlude the view of the opening to the active site, has been removed for clarity. C. Coupling of conformational changes in catalytic domain II (t = 0 green, t = 75 red) with changes in domains III (t = 0 yellow, t = 75 cyan) and IV (t = 0 tan, t = 75 mauve). Opening to C-terminal channel indicated by “C”.

Figure 6. Opening of the N-terminal small channel.

Surface representation (blue) of the starting structure (left) and the ligand-bound structure at t = 75 ns (right). Ligand is shown in stick form with carbon yellow, oxygen red and nitrogen blue. The N-terminal phenylalanine residue of the ligand is visible in the opened channel at t = 75 ns (right).

Figure 7. Propagation of conformational changes from the active site to domain III and α-helix 10.

Two roughly orthogonal views of PfA-M1 with secondary structural elements coloured in a spectrum pattern, blue to red, according to residue number. The internal chamber within the protein is shown in space filling representation and coloured violet. N-terminal channel indicated by an “N”. Left panel: Loop between β-strand 19 and α-helix 2 in domain II indicated by “L”. α-helices 11–19 are coloured orange through yellow and appear in the lower right quadrant. This region rotates in the direction of the oblique arrow. Domain III (lower left) moves in the direction of the small horizontal arrow. Right panel: α-helix 4 from domain II moves down and to the right, impinging on α-helix 10 at its N-terminus and also via α-helix 13.

Figure 8. Concerted motions of the α-helix 10 C-terminal arm toward the large C-terminal channel.

Two views of the overlayed maximum (orange) and minimum (slate) projection structures of PCA eigenvector 2 from the simulation of the ligand-bound complex. Opening to the C-terminal channel indicated by a “C”.