Artemisinin-(Iso)quinoline Hybrids by C-H Activation and Click Chemistry: Combating Multidrug-Resistant Malaria

Abstract

A substantial challenge worldwide is emergent drug resistance in malaria parasites against approved drugs, such as chloroquine (CQ). To address these unsolved CQ resistance issues, only rare examples of artemisinin (ART)-based hybrids have been reported. Moreover, protein targets of such hybrids have not been identified yet, and the reason for the superior efficacy of these hybrids is still not known. Herein, we report the synthesis of novel ART-isoquinoline and ART-quinoline hybrids showing highly improved potencies against CQ-resistant and multidrug-resistant P. falciparum strains (EC₅₀ (Dd2) down to 1.0 nm; EC₅₀ (K1) down to 0.78 nm) compared to CQ (EC₅₀ (Dd2)=165.3 nm; EC₅₀ (K1)=302.8 nm) and strongly suppressing parasitemia in experimental malaria. These new compounds are easily accessible by step-economic C-H activation and copper(I)-catalyzed azide-alkyne cycloaddition (CuAAC) click reactions. Through chemical proteomics, putatively hybrid-binding protein targets of the ART-quinolines were successfully identified in addition to known targets of quinoline and artemisinin alone, suggesting that the hybrids act through multiple modes of action to overcome resistance.

Keywords: antimalarial agents; artemisinin; drug resistance; hybridization; proteomics.

Figure 1

Selected artemisinin derivatives and pharmaceutically relevant isoquinolines/quinolines. A) Artemisinin (naturally occurring), dihydroartemisinin, artesunate, and artemether (semisynthetic derivatives). B) Clinically applied drugs with a quinoline or isoquinoline core: paritaprevir, tafenoquine, chloroquine, and amodiaquine. C) First example of an artemisinin–quinine hybrid.9a

Figure 2

Novel hybrids 1–17 applied for activity examination against P. falciparum 3D7, Dd2, and K1 strains. Red, blue, and green indicate the parent pharmacophores of the molecules. The black moieties represent the linker groups of the molecules. The 3‐hydroxy‐desoxydihydroartemisinin unit of hybrid 3 is shown in orange.

Scheme 1

A) Synthesis of isoquinoline–artesunic acid hybrids 1–3. B) Synthesis of 7‐chloroquinoline–artesunic acid hybrids 4–7 and 7‐chloroquinoline–artemisinin hybrids 8–12. C) Synthesis of 7‐chloroquinoline–artesunic acid hybrids 13–16. D) Post‐modification of hybrid 14 to hybrid 17 in a novel domino process: in situ tertiary amide formation/intramolecular rearrangement reaction. Reagents and Conditions: A) (i) [Cp*Co(CO)I₂] (10 mol %), AgSbF₆ (20 mol %), NaOAc (20 mol %), 1,2‐dichloroethane, 120 °C, 1 h; (ii) 1) PdCl₂(PPh₃)₂ (1.0 mol %), trimethylsilylacetylene, triethylamine, 50 °C, 2 h, 2) K₂CO₃, methanol, 25 °C, 4 h; (iii) CuSO₄⋅5 H₂O, sodium ascorbate, CH₂Cl₂/H₂O, overnight. B) (i) CuSO₄⋅5 H₂O, sodium ascorbate, CH₂Cl₂/H₂O, rt, overnight. 23: n=1; 24: n=2; 25: n=3; 26: n=4; 28: n=1 (C‐10β); 29: n=1 (C‐10α); 30: n=2 (C‐10β); 31: n=3 (C‐10β); 32: n=4 (C‐10β). C) (i) DCC, DMAP, CH₂Cl₂, 0 °C→rt, overnight, N₂; (ii) EDCI, DMAP, CH₂Cl₂, 0 °C→rt, overnight, N₂; (iii) K₂CO₃, acetonitrile, 130 °C, overnight. D) In CH₂Cl₂, 0 °C→rt, 18 h, N₂.

Figure 3

Experimental design of the curative Thompson test (A), follow‐up of parasitemia (B), and animal survival (C) in P. berghei‐infected mice. Parasitemia and animal survival were measured using an n=6 group, unless indicated otherwise. Error bars indicate S.D. ***p<0.01 (two‐way ANOVA) versus vehicle group. ^# p<0.01, ^## p<0.005 (Log‐rank, Mantel‐Cox test) versus vehicle. dpi=days post‐infection. S.D.=standard deviation.

Figure 4

Experimental design (A), parasitemia (B), heme species (C), and summary of results (Table) from P. berghei‐infected mice within 24 h of oral treatment. In (B), parasitemia was determined by GFP signal using flow cytometry. In (C), heme species were quantified from the peripheral blood after mouse euthanasia. The results of two independent experiments with n=3 groups for each experiment are shown. Error bars are means and 95 % CI (95 % confidence interval). IC₅₀ and % reduction values are mean±S.D. In the table, β‐hematin formation was determined 24 h after drug incubation. [a] Values relative to the infected untreated group (vehicle). *p<0.05 (one‐way ANOVA and Newman‐Keuls multiple comparison test) for before versus after treatment. ^# p<0.05 in comparison to vehicle. CQ=Chloroquine, ARE=Artesunate. RI=Reference inhibitor (CQ at 310 μmol kg⁻¹).

Figure 5

Target Identification. A) The alkyne‐tagged hybrids were reacted with biotin azide to generate the affinity probes for interactive target enrichment. Affinity pull‐down coupled mass spectrometry identification was carried out to characterize the hybrids targets. B) Heat map representation of the relative abundance of individual targets of hybrids 16 and 17. The emPAI scores of individual proteins were used to generate the heat map with morpheus software. Complete datasets of the drug targets are shown in Table S5.

Figure 6

Gene ontology (GO) analysis of interactive proteins for A) hybrid 16 and B) hybrid 17.

Figure 7

Hybrid 16 (A) and hybrid 17 (B) targets are involved in multiple biological processes essential for parasite survival.