Artemisinin and its derivatives (collectively referred to as ARTs) rapidly reduce the parasite burden in Plasmodium falciparum infections, and antimalarial control is highly dependent on ART combination therapies (ACTs). Decreased sensitivity to ARTs is emerging, making it critically important to understand the mechanism of action of ARTs. Here we demonstrate that dihydroartemisinin (DHA), the clinically relevant ART, kills parasites via a two-pronged mechanism, causing protein damage, and compromising parasite proteasome function. The consequent accumulation of proteasome substrates, i.e., unfolded/damaged and polyubiquitinated proteins, activates the ER stress response and underpins DHA-mediated killing. Specific inhibitors of the proteasome cause a similar build-up of polyubiquitinated proteins, leading to parasite killing. Blocking protein synthesis with a translation inhibitor or inhibiting the ubiquitin-activating enzyme, E1, reduces the level of damaged, polyubiquitinated proteins, alleviates the stress response, and dramatically antagonizes DHA activity.

Highlights d K13 artemisinin resistance-conferring mutations lead to reduced K13 abundance d K13 is localized to cytostome-like structures at the parasite periphery d Reduced K13 abundance impairs hemoglobin catabolism and lessens artemisinin activation d Reduced artemisinin activation limits cell damage, permitting parasite survival

Fully synthetic endoperoxide antimalarials, namely, OZ277 (RBx11160; also known as arterolane) and OZ439 (artefenomel), have been approved for marketing or are currently in clinical development. We undertook an analysis of the kinetics of the in vitro responses of Plasmodium falciparum to the new ozonide antimalarials. For these studies we used a K13 mutant (artemisinin resistant) isolate from a region in Cambodia and a genetically matched (artemisinin sensitive) K13 revertant. We used a pulsed-exposure assay format to interrogate the time dependence of the response. Because the ozonides have physicochemical properties different from those of the artemisinins, assay optimization was required to ensure that the drugs were completely removed following the pulsed exposure. Like that of artemisinins, ozonide activity requires active hemoglobin degradation. Short pulses of the ozonides were less effective than short pulses of dihydroartemisinin; however, when early-ring-stage parasites were exposed to drugs for periods relevant to their in vivo exposure, the ozonide antimalarials were markedly more effective.

Current first-line artemisinin antimalarials are threatened by the emergence of resistant Plasmodium falciparum. Decreased sensitivity is evident in the initial (early ring) stage of intraerythrocytic development, meaning that it is crucial to understand the action of artemisinins at this stage. Here, we examined the roles of iron (Fe) ions and haem in artemisinin activation in early rings using Fe ion chelators and a specific haemoglobinase inhibitor (E64d). Quantitative modelling of the antagonism accounted for its complex dependence on the chemical features of the artemisinins and on the drug exposure time, and showed that almost all artemisinin activity in early rings (>80%) is due to haem-mediated activation. The surprising implication that haemoglobin uptake and digestion is active in early rings is supported by identification of active haemoglobinases (falcipains) at this stage. Genetic down-modulation of the expression of the two main cysteine protease haemoglobinases, falcipains 2 and 3, renders early ring stage parasites resistant to artemisinins. This confirms the important role of haemoglobin-degrading falcipains in artemisinin activation, and shows that changes in the rate of artemisinin activation could mediate high-level artemisinin resistance.

No abstract