Targeting the fatty acid biosynthesis enzyme, beta-ketoacyl-acyl carrier protein synthase III (PfKASIII), in the identification of novel antimalarial agents

Abstract

The importance of fatty acids to the human malaria parasite, Plasmodium falciparum, and differences due to a type I fatty acid synthesis (FAS) pathway in the parasite, make it an attractive drug target. In the present study, we developed and a utilized a pharmacophore to select compounds for testing against PfKASIII, the initiating enzyme of FAS. This effort identified several PfKASIII inhibitors that grouped into various chemical classes of sulfides, sulfonamides, and sulfonyls. Approximately 60% of the submicromolar inhibitors of PfKASIII inhibited in vitro growth of the malaria parasite. These compounds inhibited both drug sensitive and resistant parasites and testing against a mammalian cell line revealed an encouraging in vitro therapeutic index for the most active compounds. Docking studies into the active site of PfKASIII suggest a potential binding mode that exploits amino acid residues at the mouth of the substrate tunnel.

Figure 1

Fatty acid initiation in P. falciparum. Malonyl-coenzyme A:ACP transacylase (MCAT) catalyzes the formation of malonyl-ACP from malonyl-coenzyme A. β-Ketoacyl–ACP synthase III (KASIII) catalyzes the condensation of malonyl-ACP and acetyl-CoA. The product of KASIII (β-ketoacetyl–ACP) is then elongated by the sequential action of four other FAS enzymes.

Figure 2

Refinement of the PfKASIII pharmacophore. (A) Initial pharmacophore mapped onto thiolactomycin. (B) Refined pharmacophore no longer maps to thiolactomycin.

Figure 3

Regression plot of observed vs predicted bioactivity of pharmacophore hypothesis Hypo1A.

Figure 4

Specificity of the refined PfKASIII pharmacophore. (A) Pharmacophore hypothesis Hypo1A (green, hydrogen bond acceptor; orange, ring aromatic; cyan, hydrophobic). (B) Compound 3a mapped onto Hypo1A showing the para methyl phenyl ring acting as the ring aromatic moiety, the hydroxyl group ortho to the sulfonyl group acting as the hydrogen bond acceptor while the naphthyl ring contributes as the hydrophobic moiety. (C) Compound 4a mapped onto Hypo1A showing the para methoxy phenyl ring acting as the ring aromatic moiety, the sulfur group acting as the hydrogen bond acceptor while the phenyl ring of the benzene sulfonic acid portion contributes as the hydrophobic moiety. (D) Compound 2b mapped onto Hypo1A showing the phenyl ring of the benzene sulfonic acid part acting as the ring aromatic moiety, no mapping on the hydrogen bond acceptor, and a poor mapping of the chlorine as the hydrophobic moiety. (E) Compound 3l mapped onto Hypo1A showing the naphthyl ring acting as the ring aromatic moiety, no mapping on the hydrogen bond acceptor, and a moderate mapping of the para methyl group of the toulamide portion as the hydrophobic moiety.

Figure 5

PfKASIII molecular model. (A) A ribbon representation of the PfKASIII homology model. The active site residues, cysteine 115, histidine 254, and asparagine 257, are shown in yellow sticks. The enzyme active site is buried near the core of the protein with a long substrate tunnel extending to the surface. This tunnel, largely, but not completely, lined with hydrophobic residues, binds coenzyme A, the acyl group donor for the reaction. The docked pose of compound 3a is shown in the active site. (B) The predicted interactions between 3a and PfKASIII. The position of the two ring systems in this pose is similar to the predicted poses for the majority of sulfones and sulfonamides: one ring system making van der Waals contacts with the active site triad and the other aligned up the active site tunnel rotated to make optimal contacts based on the particular composition of the second ring.