A potent series targeting the malarial cGMP-dependent protein kinase clears infection and blocks transmission

Abstract

To combat drug resistance, new chemical entities are urgently required for use in next generation anti-malarial combinations. We report here the results of a medicinal chemistry programme focused on an imidazopyridine series targeting the Plasmodium falciparum cyclic GMP-dependent protein kinase (PfPKG). The most potent compound (ML10) has an IC₅₀ of 160 pM in a PfPKG kinase assay and inhibits P. falciparum blood stage proliferation in vitro with an EC₅₀ of 2.1 nM. Oral dosing renders blood stage parasitaemia undetectable in vivo using a P. falciparum SCID mouse model. The series targets both merozoite egress and erythrocyte invasion, but crucially, also blocks transmission of mature P. falciparum gametocytes to Anopheles stephensi mosquitoes. A co-crystal structure of PvPKG bound to ML10, reveals intimate molecular contacts that explain the high levels of potency and selectivity we have measured. The properties of this series warrant consideration for further development to produce an antimalarial drug.Protein kinases are promising drug targets for treatment of malaria. Here, starting with a medicinal chemistry approach, Baker et al. generate an imidazopyridine that selectively targets Plasmodium falciparum PKG, inhibits blood stage parasite growth in vitro and in mice and blocks transmission to mosquitoes.

Fig. 1

In vivo efficacy of PKG inhibitors against rodent malaria parasites. a Groups of five female BALB/c mice were infected with 1x10⁷/ml P. berghei (ANKA) blood stage parasites in a Peters 4-day test and were given a twice daily dose (25 mg/kg) of one of three test compounds by oral gavage. Chloroquine was used as a positive control at a single daily oral dose of 10 mg/kg. b Groups of five female BALB/c mice were infected with 1×10⁷ P. chabaudi (AS) blood stage parasites and were given a single oral dose (50 mg/kg) of either ML1 or ML4 by oral gavage just prior to the predicted onset of schizogony. Chloroquine was used as a positive control at a single daily oral dose of 10 mg/kg. c Groups of five BALB/c mice were infected with 1×10⁷ P. chabaudi (AS) blood stage parasites and were given either a single or twice daily oral dose (50 mg/kg) of ML10 by oral gavage. The first dose was given to both groups of mice just prior to the predicted onset of schizogony and in one group this was followed 3 h later when schizogony was predicted to have been completed. The data are from single experiments each performed on a group of five mice. Error bars show the s.e.m

Fig. 2

Efficacy of ML10 against P. falciparum in the GSK PfalcHuMouse model and determination of in vitro killing dynamics. a Two mice were treated with vehicle and another two mice with either 50 or 100 mg/kg of ML10 to test proof of concept of efficacy in vivo. Parasitemia is shown over time in individual mice during the efficacy assay. The dotted horizontal line indicates 90% reduction in parasitaemia compared to vehicle-treated animals. Each symbol represents an individual mouse. b Microscopic and flow cytometric analysis of P. falciparum present in the peripheral blood of mice treated with vehicle or ML10. Samples taken after one (48 h) and two cycles (96 h) of exposure to the test compound were further analyzed. Flow cytometry dot plots from samples of peripheral blood show P. falciparum-infected human erythrocytes (blue rectangle). Images in the right-hand panels show Giemsa-stained blood stage parasites. Blood films from control untreated animals show normal staining and appearance. The parasites in ML10-treated animals show a relative enrichment in late schizonts at Day 5, whereas most cells remaining in peripheral blood at Day 7 are pyknotic (red circle). Scale bar, 3 µm. c The in vitro parasite reduction rate (PRR) assay was used to determine the onset of action and rate of killing as previously described. P. falciparum was exposed to ML10 at a concentration corresponding to 10× EC₅₀. The number of viable parasites at each time point was determined as described. Four independent serial dilutions were tested with each sample to correct for experimental variation; error bars show the standard deviation. Previous results reported on standard antimalarials tested at 10× EC₅₀ using the same conditions are shown for comparison

Fig. 3

Transmission-blocking activity of ML10. P. falciparum (NF54) gametocytes were incubated for 24 h with nine different concentrations of ML10 and were each fed to a separate cage of 30–40 Anopheles stephensi mosquitoes using a SMFA. Up to 20 surviving mosquitoes were dissected on day 7 post-feed and oocyst numbers assessed by microscopy. This experiment was performed twice and the figure shows the combined data for oocyst intensity in each mosquito as a function of the compound concentration for the replicate feeders. The left segment of the x-axis shows the oocyst intensities in the vehicle (0.1% DMSO) controls. The positive control was 10 μM dihydroartemisinin (DHA). Error bars show the s.e.m

Fig. 4

Side-by-side comparison with key features in the PvPKG/ML1 and PvPKG/ML10 co-structures. a Hydrogen bonds provide key interactions: At the hinge of the kinase domain, V614 forms hydrogen bonds with the amino-pyrimidine group of both ML1 (left) and ML10 (right). In addition, ML10 extends a sulfonamide moiety to form hydrogen bonds with D675 and F676, resulting in significantly greater inhibitory potency than ML1. b Gatekeeper confers specificity: Both ML1 (left) and ML10 (right) extend a fluorophenyl group to occupy a hydrophobic pocket adjacent to T611 (magenta). This explains in part the specificity of the inhibitors as a residue with a longer side-chain may potentially clash with this functional group. c Hydrophobic network enhances inhibitory potency: Plasmodium PKG also interacts with both ML1 (left) and ML10 using a network of hydrophobic residues (see Supplementary Fig. 13 for the same figure with amino-acid labels)