Pan-active imidazolopiperazine antimalarials target the Plasmodium falciparum intracellular secretory pathway

Abstract

A promising new compound class for treating human malaria is the imidazolopiperazines (IZP) class. IZP compounds KAF156 (Ganaplacide) and GNF179 are effective against Plasmodium symptomatic asexual blood-stage infections, and are able to prevent transmission and block infection in animal models. But despite the identification of resistance mechanisms in P. falciparum, the mode of action of IZPs remains unknown. To investigate, we here combine in vitro evolution and genome analysis in Saccharomyces cerevisiae with molecular, metabolomic, and chemogenomic methods in P. falciparum. Our findings reveal that IZP-resistant S. cerevisiae clones carry mutations in genes involved in Endoplasmic Reticulum (ER)-based lipid homeostasis and autophagy. In Plasmodium, IZPs inhibit protein trafficking, block the establishment of new permeation pathways, and cause ER expansion. Our data highlight a mechanism for blocking parasite development that is distinct from those of standard compounds used to treat malaria, and demonstrate the potential of IZPs for studying ER-dependent protein processing.

Fig. 1. GNF179-resistant yeast strains harbor mutations in endoplasmic reticulum(ER)-based lipid homeostasis and autophagy genes.

a Protein maps showing relevant mutations and PROSITE predicted protein domains, if applicable. Maps were generated using Illustrator of Biological Sequences (IBS) software package. Missense mutations are shown in yellow ovals, nonsense mutations in red pentagons, and frameshift mutations as purple arrows. b Protein–protein interaction (PPI) network was generated using the STRING database. Each node represents a S. cerevisiae protein and connecting lines delineate interactions. The PPI enrichment p-value (p = 1.38  × 10⁻¹⁴) indicates that the proteins show significantly more interactions among themselves than would be expected from a random subset of genes from the yeast genome.

Fig. 2. GNF179 localizes to the ER of early stage parasites.

a Chemical structure of canonical and NBD conjugated GNF179 and Coumarin-1 conjugated GNF179. b Dose response curves for GNF179 and Coumarin1 (left) or NBD (right) conjugated GNF179 in wild-type and KAF156-resistant clone (KAD452-R3, containing three mutations in pfcarl (M81I, L830V and S1076I). c Colocalization of Coumarin-1 conjugated GNF179 with ER-tracker red. d Colocalization of NBD conjugated GNF179 with ER-tracker red. Parasites are in mid-ring (6-hours post-infection) stage and were treated for 30 min with 2 µM GNF179-Coumarin1 and 100 nM GNF179-NBD. The blue signal represents the DAPI-stained parasite nucleus. Scale bars: 2 µm. Source data for b is provided as a Source Data file.

Fig. 3. Secretion reporter constructs demonstrate that GNF179 inhibits protein export of Plasmodium falciparum.

a Protein expression levels of PfEMP3-GFP reporter. This fusion includes the signal peptide and PEXEL motif of PfEMP3. By immunoblot, three protein products are seen with anti-GFP antibodies. The three indicated bands, in response to probing with GFP, as follows: 1. Full length protein (black arrow), 2. PEXEL cleaved protein (blue arrow), 3. GFP degradation product (magenta arrow). HSP70 is used as a loading control. b SERA5ss-GFP fusion reporter treated with GNF179. By immunoblot with anti-GFP antibodies we see two protein products for this construct: 1. Signal peptide cleaved (blue arrow) and 2. GFP degradation product (magenta arrow). HSP70 serves as a loading control. c ³⁵S incorporation of newly synthesized amino acids at different concentrations of KAF156, chloroquine (negative control for inhibition) and cycloheximide (positive control for inhibition). Counts were normalized to data obtained from no-drug controls (NDC) and presented as mean ± SD (n = 2 independent experiments). Source data for c is provided as a Source Data file.

Fig. 4. Confirmation of GNF179’s inhibition of protein export in P. falciparum blood stage parasites.

a Vector used for assays shown in b–d. b–d Imaging of GFP reporter to the plasma membrane under the indicated compound treatments for 16 h. Parasites were fixed and stained with Hoechst 33342 (blue), α-GFP (green), α-ERD2 (red) and α-PDI (cyan) antibodies. Scale bars: 2 µm unless otherwise indicated. Overlay: DIC images merged with fluorescent channels. Quantification of colocalization between KAHRP-GFP and ERD2 (e), and between KAHRP-GFP and PDI (f) are indicated by Pearson correlation coefficient (PCC); Each dot represents an individual parasitized red blood cell. Statistical significance was evaluated using an unpaired, two-tailed t-test where where * is p < 0.05, ** is p < 0.01, *** is p < 0.001, and **** is p < 0.0001. g Model for establishing new permeation pathways. If proteins are not exported to the red cell where they can form new permeation pathways, sorbitol will not be imported and cause lysis. h Percentage of rings (mean ± SEM, at least three independent experiments) from an asynchronous parasite population 24 h after sorbitol synchronization, treated as indicated by GNF179 (5 nM), Chloroquine (CQ, 500 nM) or brefeldin A (BFA, 5 µM). Statistical significance was determined using a paired, two-tailed t-test where * is p < 0.05, ** is p < 0.01, and *** is p < 0.001. Dd2 pfcarl experiments were conducted with the pfcarl evolved triple mutant (KAD452-R3) and edited Dd2-ACTStop mutant. For sensitive parasites (Dd2), 25 nM of GNF179 was used, while 1 µM was used on resistant parasites. Source data for e, f, and h are provided as a Source Data file.