Targeting Aurora Kinases as Essential Cell‐Cycle Regulators to Deliver Multi‐Stage Antimalarials Against Plasmodium Falciparum

Abstract

Kinases play critical roles in the development and adaptation of Plasmodium falciparum and present novel opportunities for chemotherapeutic intervention. Mitotic kinases that regulate the proliferation of the parasites by controlling nuclear division, segregation, and cytokinesis. We evaluated the potential of human Aurora kinase (Aur) inhibitors to prevent P. falciparum development by targeting members of the Aurora‐related kinase (Ark) family in this parasite. Several human AurB inhibitors exhibited multistage potency (< 250 nM) against all proliferative stages of parasite development, including asexual blood stages, liver schizonts, and male gametes. The most potent compounds, hesperadin, TAE684, and AT83, exhibited > 1000x selectivity towards the parasite. Importantly, we identified PfArk1 as the principal vulnerable Ark family member, with specific inhibition of PfArk1 as the primary target for hesperadin. Hesperadin's whole‐cell and protein activity validates it as a unique PfArk1 tool compound. Inhibition of PfArk1 results in the parasite's inability to complete mitotic processes, presenting with unsegregated, multi‐lobed nuclei caused by aberrant microtubule organization. This suggests PfArk1 is the main Aur mitotic kinase in proliferative stages of Plasmodium, characterized by bifunctional AurA and B activity. This paves the way for drug‐discovery campaigns based on hesperadin targeting PfArk1.

In this study, we repurposed human Aurora kinase‐specific inhibitors to identify potential antimalarial agents. Two inhibitors, hesperadin and TAE684, exhibited sub‐micromolar activity across multiple parasite stages, with hesperadin demonstrating significant potency and selectivity by specifically targeting PfArk1. Inhibition of PfArk1 resulted in anaphase‐like abnormalities, pointing to its role as the primary PfArk for cell division in malaria parasites.

Introduction

Plasmodia spp. parasites demonstrate pathogenic success due to their complex life cycle, alternating between non‐proliferative stages, where the cell cycle is quiescent, and stages characterized by rapid cell division events that lead to massive parasite population expansion^{[ 1} , ^{2 ]} Plasmodium falciparum, the causative agent of the most severe form of malaria,^{[ 3 ]} undergoes three unique and specialized cell division events. During hepatic and intra‐erythrocytic schizogony within the human host, a haploid parasite undergoes multiple rounds of closed asynchronous mitosis and karyokinesis, followed by a singular and synchronized cytokinesis event to produce a segmented schizonts.^{[ 4} , ^{5 ]} An equally unique cell division event occurs within the mosquito host, where male gametocytes (1n) undergo three rounds of rapid DNA replication (exflagellation) to generate eight flagellated male gametes (8n) in just ∼15 min.^{[ 6} , ^{7 ]} The parasite's ability to undergo rapid asexual replication is a key factor in its pathogenic success but requires extraordinary control. A detailed mechanistic understanding of the role players in regulating the parasite's atypical cell cycle in Plasmodium could lead to novel antimalarial therapeutic agents.

Cell‐cycle machinery and regulators, such as protein kinases (PKs), are crucial for accurate progression through various checkpoints in mammalian cells. Several conserved Ser/Thr mitotic PKs are considered primary regulators of the mitotic process, including the “Never In Mitosis” kinases (NIMA/Neks), Polo‐like kinases, and Aurora kinases (Aur). The Aur family is highly conserved amongst eukaryotes, with members identified, amongst others, in yeasts (Ipl1), humans (HsAurA, B, and C), Toxoplasma gondii (TgArk1–3),^{[ 8 ]} Trypanosoma brucei (TbAUK1).^{[ 9 ]} In P. falciparum, three aurora‐related kinases exist (PfArk1–3).^{[ 10} , ^{11 ]} Aur contributes to the assembly and disassembly of mitotic and meiotic centrosomes, regulating spindle‐pole structure and dynamics, chromosome segregation, and cellular fission during cytokinesis. Although Aur members are differentiated functionally depending on their localization, delocalization can cause moonlighting effects between HsAurA and B,^{[ 12 ]} although direct compensation for the loss of activity is not evident.^{[ 13 ]} During mitosis, HsAurA (“polar” Aur) localizes to the centrosome and spindle poles, and upon binding of microtubule‐associated protein TPX2, regulates centrosome maturation, separation, and microtubule spindle formation. HsAurB as “equatorial” Aur (with INCENP, survivin, and borealin), forms the chromosome passenger complex (CPC) as master controller of cell division, localized to centromeres, kinetochores, and the spindle midzone, allowing AurB to govern chromosome condensation, kinetochore attachment, sister chromatid segregation, and cytokinesis.

PfArks are implicated in critically regulating cell‐cycle progression of Plasmodium based on 1) the essentiality of all three PfArks to asexual proliferation (schizogony),^{[ 14 ]} 2) the unique expression patterns of the PfArks during cell‐cycle arrest and re‐entry,^{[ 15 ]} and 3) distinct, highly specific, and exclusive spatiotemporal associations during ABS schizogony.^{[ 16 ]} Although P. falciparum lacks a canonical centrosome, it possesses a microtubule‐organizing center (MTOC) characterized by a centriolar plaque (CP) embedded in the nuclear envelope, with inner CP (intranuclear body) and outer CP domains (cytoplasmic body).^{[ 17 ]} The CPs harbor several validated centrosomal proteins, including centrin and γ‐tubulin, and facilitate microtubule (MT) nucleation. PfArk1 (PF3D7_0605300) and PfArk2 (PF3D7_0309200) are associated with MTOCs during schizogony,^{[ 16 ]} with PfArk1 localizing to the outer CP domains of duplicated MTOCs in nuclei primed for division (similar to a “polar” Aur^{[ 10 ]}) while PfArk2 is additionally proposed to localize to kinetochores, akin to an “equatorial” Aur.^{[ 16 ]} PfArk3 (PF3D7_1356800) is found only in segmented nuclei associated with subpellicular microtubules (SPMTs) as cytosolic microtubules in merozoites, suggesting a role in cytokinesis.^{[ 16 ]}

The deregulation of HsAurs (especially HsAurA and HsAurB) has been linked to cancer and tumorigenesis, making them attractive targets for anticancer therapeutic strategies.^{[ 18 ]} Several inhibitors selectively target either HsAurA or HsAurB or have dual or pan‐reactive abilities. HsAurA inhibition leads to defects in mitotic spindle assembly and ultimately causes spindle checkpoint‐dependent mitotic arrest, cell cycle exit, and apoptosis.^{[ 19 ]} On the other hand, HsAurB inhibition causes abnormal chromosome alignment and overrides the mitotic spindle checkpoint, causing polyploidy, failure of cytokinesis, and endoreduplication.^{[ 20 ]}

Although several kinase families have been chemically and genetically validated as antimalarial targets (e.g., PfPI4K,^{[ 21 ]} PfPKG^{[ 22 ],} and PfCLK3^{[ 23 ]}), the unique requirement of Ark members for parasite‐proliferation processes has not been extensively explored to identify novel inhibitors specifically targeting this kinase family. Previous studies have shown that the HsAurB‐specific inhibitor hesperadin exhibits potent in vitro activity against P. falciparum, Trypanosoma brucei, and Leishmania donovani.^{[ 24} , ²⁵ , ^{26 ]} We also demonstrated that hesperadin treatment leads to mutations in PfArk1 that confer resistance.^{[ 27 ]} However, evidence of direct target engagement and inhibition of the PfArk members is lacking.

Here, we provide an in‐depth evaluation of PfArk inhibition and its influence on parasite survival. We systematically assessed HsAur inhibitors targeting all Aur classes to identify compounds that could be repurposed as antimalarials. We identify Ark inhibitors with multistage activity against proliferative stages of the parasite, including asexual blood stage (ABS) parasites, male gametes, and liver schizonts, correlating with the required function of Arks in cell proliferation events. Biochemical characterization showed the specific and sensitive inhibition of PfArk1 as the primary target of the inhibitors, with exquisite potency seen for hesperadin and NVP‐TAE684 (TAE684). Hesperadin is a bona fide inhibitor of only PfArk1, without targeting other kinases or exhibiting additional pleiotropic activity associated with the inhibition of hemozoin formation (the crystalline byproduct of hemoglobin digestion), as is the case with TAE684 and the other kinase inhibitors studied. A unique aspartate residue in the PfArk1 active site confers selectivity to hesperadin inhibition over the mammalian Aur and PfArk2, which have a lysine residue at the equivalent position. PfArk1 is shown to be the main Ark family member involved in mitotic processes, and interference with its activity results in aberrant nuclear division with no clear microtubule nucleation at the CPs in both ABS parasites and male gametes. To our knowledge, this is the first direct chemical evidence of an inhibitor specifically targeting any of the PfArk members. Notably, we demonstrate that inhibition of PfArk1 activity can be achieved at single‐digit nanomolar concentrations, with a large selectivity window for the parasite, and a favorable profile for potential drug candidates. These findings expand the current knowledge base regarding kinases as drug targets in malaria parasites and provide chemical validation of PfArk1 as a druggable target, with hesperadin presenting a starting point for further drug‐discovery initiatives.

Results and Discussion

Aur Inhibitors Demonstrate Effectiveness Against the Replicative Stages of P. falciparum Parasites

A set of commercially available compounds was chosen based on their specificity and potency to the HsAur members. The inhibitors were first assessed against various life cycle stages of Plasmodium parasites in vitro to determine their multistage antiplasmodium activity (Figures 1 and S1). Less than half of the HsAurA inhibitors targeted ABS proliferation of drug‐sensitive NF54 P. falciparum parasites with IC₅₀ values < 5 µM (Figures 1a and S1). By contrast, both HsAurB, most of the dual‐active inhibitors and five of the six pan‐active Aur inhibitors showed activity at this concentration. Six compounds (AAi I & Aki III [targeting HsAurA^{[ 28} , ^{29 ]}], hesperadin & AZD‐1152 [targeting HsAurB^{[ 30} , ^{31 ]}], AT83 & ZM‐39 [dual HsAurA&B^{[ 32} , ^{33 ]}]) were all potent at < 1 µM. All inhibitors that exhibit specificity towards HsAurB, even when developed as dual HsAurA&B, are generally more potent against P. falciparum ABS, with an IC₅₀ of 1.5 nM for hesperadin (Figure S2a), and <250 nM for AT83 and ZM‐39. The eight most active compounds retained activity against multidrug‐resistant PfDd2 and PfK1 strains with ≤ 2‐fold variance in IC₅₀ from PfNF54 (Figure S2b). TAE684 (Novartis’ first‐generation anaplastic lymphoma kinase (ALK)‐specific inhibitor) was also included in this assay as it was proposed to target a PfArk based on Kinobead competitive pulldown data^{[ 34 ]} (Figure S2c), with TAE684 also presenting potent activity against ABS parasites with an IC₅₀ of 280 ± 13 nM.

Figure 1.

Activity profile of Aur class anticancer inhibitors across the life cycle of Plasmodium falciparum parasites. a) Activity (IC₅₀) of inhibitors (grouped based on specificity against human Aur) against P.falciparum drug‐sensitive (PfNF54) asexual intra‐erythrocytic parasites as measured with SYBR Green I fluorescence as an indicator of viability and proliferation. Values obtained for the most active inhibitors are from three independent biological replicates, each performed in technical triplicate (n=3, mean ± S.E.), while the remainder is from two independent biological replicates, each performed in technical triplicate (n = 2, mean ± S.E.). b) Cytotoxicity (CC₅₀) of selected inhibitors against hepatocellular carcinoma (HepG2) and CHO cell lines. The Selectivity Index (SI = HepG2 CC₅₀/PfNF54 IC₅₀ and CHO CC₅₀/PfNF54 IC₅₀) is indicated below the graph (n = 3, mean ± S.E.). c) Single point activity profile of selected inhibitors at 5 µM against immature (iGc, II‐III) and late‐stage (LGc, IV/V) gametocytes determined by measuring luminescence of P. falciparum parasites expressing luciferase (n = 3, mean ± S.E.). d) Activity (IC₅₀) against male gamete formation (green) and female gametes (pink) for TAE684, AT83, and ZM‐39 (n = 3, mean ± S.E.). e) Structures of the seven most active compounds selected. ABS, asexual blood stages of the parasite; HsAur, human Aur.

With selectivity and mammalian cell toxicity often being concerns with kinase inhibitors, we evaluated the activity of the active compounds against hepatocellular carcinoma (HepG2) and Chinese Hamster Ovary (CHO) cell lines (Figure 1b). Several compounds exhibited cytotoxicity with selectivity indices (SI) < 10, including GSK‐916, AAi I, Aki III, and TAE684(CC_50s provided in S1)). The pan‐active inhibitor danusertib demonstrated the most pronounced cytotoxic effect against both lines. However, HsAurB inhibitors hesperadin and AZD‐1152 showed distinct selectivity in targeting P. falciparum compared to mammalian cell lines, indicating differentiation in action in the parasite, with SI>1000 (CC₅₀ of 4.0 ± 1.0 and 13.2 ± 4.9 µM against CHO and HepG2 cells) and >100‐fold (CHO/HepG2 CC₅₀ >15 µM), respectively. Similarly, the dual HsAurA&B inhibitors, AT83 (CC₅₀ of 4.5 ± 2.1 and CC₅₀ of 12.9 ± 4.5 µM, CHO & HepG2) and ZM‐39 (CHO/HepG2 CC₅₀ of > 15 µM), also showed SI>100‐fold.

We subsequently evaluated the ability of the compounds active on P. falciparum NF54 ABS parasites to target additional life‐cycle stages of P. falciparum parasites. All selected inhibitors displayed minimal (<50% inhibition at 5 µM) gametocytocidal activity, with only AAi I, GSK‐916, and TAE684 showing ∼70% inhibition of immature gametocyte viability (>80% stage II/III) whilst unable to kill mature gametocytes effectively (Figure 1c). This finding was not unexpected, as gametocytes are non‐proliferative cells, although all three PfArks are expressed during gametocytogenesis on a transcript and protein level.^{[ 16} , ^{35 ]} Such functional impairment of gametocytes has been described before in other gametocyte‐sterilizing compounds,^{[ 36 ]} with the effect evident during male gametogenesis. Subsequently, hesperadin (IC₅₀ of 10 nM^{[ 37 ]}), AT83 (786 ± 16 nM), and TAE684 (116 ± 17 nM) prevented male gamete exflagellation, whereas only TAE684 had any appreciable activity against female gametes (IC₅₀ of 2.3 ± 0.05 µM), with the rest inactive (>10 µM) (Figure 2d). This confirms the ability of these Aur inhibitors to target male gamete formation, which also requires DNA replication and cell division. Hesperadin and TAE684 further display a low nanomolar (IC₅₀ <200 nM) potency against P. berghei liver schizonts, confirming the preference of these compounds for proliferative forms of the parasite (Figure S2d). Taken together, these data reveal that a focused set of HsAur‐specific inhibitors demonstrates potent and selective antiplasmodium activity, preferentially targeting proliferative parasite stages related to cell cycle division, while exhibiting an acceptable margin of cytotoxicity. However, in addition to having potential as ABS active in TCP‐1 type strategies,^{[ 38 ]} the activity against gametes and liver stages raises the possibility of leveraging PfArk1 inhibition in transmission‐blocking strategies (with both TCP‐3 and TCP‐5 potential).^{[ 39 ]}

Figure 2.

Identification and validation of Plasmodium Aurora‐related kinase as the primary target. a) Activity (IC₅₀) against recombinant PfArk1 and PfArk3 proteins, using a three‐hybrid split‐luciferase competitive binding assay (KinaseSeeker). b) Effect of conditional knockdown (cKD) of PfArk1 and PfArk2 on parasite sensitivity relative to control conditions in the presence of high aTc. Representative dose‐response curves are presented for each cKD parasite line (n=3, mean ± S.E.) with an unpaired two‐tailed t‐test, *p<0.05; **p<0.01; ***p<0.001. c) Single‐point activity profile of selected inhibitors at 1 µM against recombinant PfNek1 (KinaseSeeker). d) Inhibitory activity against recombinant PfPKG, PfCLK3, and PvPI4Kβ. The mean IC₅₀ values ± SD were calculated using two independent experiments (n = 2), each with technical duplicates using the ADP‐Glo Kinase Assay. e) Stacked bar plots illustrating barcode populations on days 0 and 14 for no drug, drug control (halofuginone), and selected inhibitors.

PfArk1 was Identified and Validated as a Novel Plasmodium Kinase Target

Given that the selected inhibitors were developed and optimized to specifically target different HsAur mitotic kinases, we aimed to correlate the whole‐cell inhibition of the proliferative stages of P. falciparum to the biochemical evaluation of the inhibition of PfArk proteins (Figure 2). We first used the KinaseSeeker competitive binding assay to determine the inhibitory activity against PfArk1 and PfArk3.^{[ 40 ]} Among the selected inhibitors tested, AT83 and AAi I had a marginal effect (AT83: 30% inhibition of PfArk1, AAi I: 40% inhibition of PfArk3, Figure S3a). However, hesperadin and TAE684 exhibited potent activity against PfArk1, with IC₅₀ values of 2 ± 0.2 and 146 ± 20 nM, respectively (Figure 2a). Notably, both hesperadin and TAE684 had a significant preference towards PfArk1, with hesperadin ∼480‐fold more active against PfArk1 than PfArk3 (p = 0.0004 and p = 0.0001, respectively, n = 3, unpaired Student's t‐test). Moreover, the 2 nM activity of hesperidin on PfArk1 protein correlates with the in vitro activity against drug‐sensitive PfNF54 parasites at 1.5 nM.

We evaluated the involvement of PfArk2 inhibition by determining the loss of activity of the compounds against conditional knockdown (cKD) lines of P. falciparum for either PfArk1 or PfArk2.^{[ 41 ]} Most compounds did not show a change in activity against the cKD of either PfArk1 or PfArk2 (Figure 2b). However, cKD of PfArk1 resulted in an increased sensitivity to hesperadin and TAE684, as evidenced by a significant > 5‐fold decrease in the IC₅₀ values compared to the wild‐type control (p = 0.0002 and p = 0.00003, respectively, n = 3, unpaired Student's t‐test; Figures 2b and S3b), associated with decreased protein levels of PfArk1 under cKD conditions. Conversely, there was no change in the IC₅₀ value for these compounds in the PfArk2 cKD. This further supports PfArk1 as the primary protein target for hesperadin and TAE684 (Figure 2c).

The specificity towards PfArk1 was confirmed by evaluating the ability of the compounds to inhibit other currently relevant antimalarial Ser/Thr or lipid kinase drug targets. This included inhibition of PfNek1, as hesperadin resistance selections previously yielded mutations in this gene, suggesting an epistatic interaction between PfArk1 and PfNek1,^{[ 27 ]} and P. berghei Nek1 exhibits similar spatiotemporal associations with the outer CP domain as PfArk1.^{[ 42 ]} However, hesperadin did not inhibit PfNek1 activity, even at 1 µM, with only TAE684 showing a marginal 40% effect on this protein (Figure 2c). The inhibitors did not exhibit noteworthy activity against three other validated antimalarial kinase targets (PfPKG, PfCLK3, or PvPI4Kβ) (Figure 2d), except for AT83, which inhibited PfCLK3 with an IC₅₀ of 100 nM. However, it is important to note that the kinase assay was performed at a low ATP concentration (10 µM). According to the Cheng‐Prusoff equation, the PfCLK3 IC₅₀ will be ∼100‐fold higher in the presence of cellular ATP concentrations (∼3 mM in P. falciparum parasites). Thus, inhibition of PfCLK3 within the parasite is predicted to be weak and unlikely to contribute substantially to the observed antiplasmodial activity.

Additionally, hesperadin, TAE684, and AT83 were evaluated for their ability to inhibit the proliferation of a set of resistant P. falciparum parasites (including resistance mutants for PfPI4k and PfCLK3) using the antimalarial resistome barcode sequencing (AReBar) assay.^{[ 43 ]} All three compounds killed all of the resistant lines in the platform (Figures 2e, S3c, and S2), indicating no cross‐resistance with known antimalarial resistance mechanisms and a novel mode of action.

These data indicate that the specific inhibition of PfArk1 within the family of Arks in P. falciparum by hesperadin and TAE684 is the primary driver of ABS antiplasmodium activity. This correlates with the specific increase in abundance of this protein (over PfArk2) peaking at ∼28–34 hpi^{[ 44 ]} in preparation for its availability during schizogony (Figure S3d).

The validation of PfArk1 as a druggable target expands and diversifies the compendium of targetable mitotic kinases in Plasmodium beyond the current indication of PfNek3 inhibition with BI‐2536, a known potent human polo‐like kinase 1 inhibitor.^{[ 45 ]} This provides a clear starting point for drug repurposing and repositioning strategies against malaria. Hesperadin is also active against T. brucei and Leishmania major, with initial hit expansion indicating that analogues mirror hesperadin's activity and phenotype.^{[ 24 ]} This provides support for structure‐activity relationship (SAR) expansion studies and further development of hesperadin as an antiplasmodium chemotype. Initial data on the in vitro ADME properties of the frontrunner compounds (hesperadin, TAE684, and AT83) indicate metabolic liabilities for hesperadin, particularly in human liver microsomes, whereas AT83 displays a favorable half‐life of 395 min in microsomes, but with potential phase II liabilities as indicated by low stability in rat hepatocytes (Table 1). These data provide baseline information to guide design and optimization of a next generation of derivatives in hit‐2‐lead optimization campaigns to progress these compounds.

Table 1.

Summary of in vitro ADME properties of selected active compounds.

			HL microsome a)		Rat hepatocyte
	eLogD pH 7.4	Kin sol (µM) pH 7.4	CLint (µL min−1 mg−1)	t1/2 (min)	CLint (µL min−1/10e6 cells)	t1/2 (min)	Protein binding (%)b	tPSA	MW (Da)
AT83	2.2	22.2	3.5	395	12.6	55	16.5	111	381
TAE684	3.4	198.5	25.4	55	3.5	197	80	103	614
Hesperadin	2.8	2.5	76.1	18	14.2	49	37.3	98	517

^a) Human liver microsomes.

^b)Protein binding as measured in Albumax II media.

The Activity of the Additional HsAur Inhibitors is not Associated with Ark Inhibition

Since all the compounds selected for this study were based on their activity against HsAur, but we could only convincingly show that hesperadin and TAE684 target PfArks, we explored alternative mechanisms of activity for some of the most potent inhibitors (AT83, ZM‐39, Aki I, AAi III, and AZD‐1152). Several kinase inhibitors inhibit hemozoin formation due to structural similarities in the presence of multiple heteroaromatic rings, planar structures, and basic centers.^{[ 46 ]} Evaluation of the ability to block formation of synthetic β‐hematin (βH) in vitro in a cell‐free detergent‐mediated Nonidet P‐40 (NP‐40) assay^{[ 47 ]} indicated that AAi I, Aki III, AZD‐1152, and ZM‐39 (but not AT83) were potent inhibitors of βH formation (IC₅₀ < 20 µM), similar to the positive control chloroquine (CQ) (Figure 3a). This could implicate inhibition of hemozoin formation as a primary mode of action for these compounds. Interestingly, TAE684 displayed some effect against βH formation (IC₅₀ of 41.4 ± 2.3 µM, Figure 3a), whereas hesperadin was ∼3‐fold less active (IC₅₀ of 126.3 ± 42.0 µM). The inhibition of βH formation translated to inhibition of intracellular hemozoin formation for both ZM‐39 and TAE684,^{[ 48 ]} which caused a significant increase in free heme (p = 0.00000002 and p = 0.0003, respectively, n = 3, unpaired Student's t‐test), accompanied by a simultaneous decrease in hemozoin formation (p = 0.00001 and p = 0.00000003, respectively, n = 3, unpaired Student's t‐test) (Figure 3b). This effect was not observed in the hesperadin treatment, with no change in either heme or hemozoin levels, implying that hesperadin does not affect hemozoin formation in the parasite (Figure S4a).

Figure 3.

Phenotypic effect of potent HsAur inhibitors on intra‐erythrocytic development. a) Measuring the ability of the inhibitors to interfere with the formation of synthetic hemozoin, βH, in vitro in a cell‐free detergent‐mediated NP‐40 assay. Error bars represent ± SD for technical triplicates with an unpaired two‐tailed Student's t‐test. Exact p‐values provided in *p<0.05; **p<0.01, ***p<0.001, ****p<0.0001. b) Dose‐dependent changes in the heme Fe levels from intracellularly‐extracted fractions of hemozoin under ZM‐39 and TAE684 treatment. c) Phenotypic response of ABS PfNF54 parasites exposed to AT83 (3xIC₅₀, ∼600 nM), ZM‐39 (3xIC₅₀, ∼750 nM), TAE684 (3xIC₅₀, ∼900 nM) and hesperadin (IC₉₉, ∼3 µM). Parasite morphology was observed at 12 h intervals using thin blood smears and indicated enlarged food vacuoles (i, iv, and v), and small hemozoin crystals (ii and iii).

This was confirmed by a distinct phenotypic morphology, where a decreased hemozoin crystal size (Figure 3c–i,ii) and an enlarged food vacuole‐like structure (iii & iv) were observed for ZM‐39 treatment within the first 24 h post‐treatment (hpt). Importantly, the TAE684 treatment had a distinctly different phenotype, whereas hemozoin formation persisted for the first 24 h, but thereafter, a distinct vacuolar structure was present, and aberrant schizonts were formed. The morphological abnormalities associated with the food vacuole were not present in either hesperadin‐ or AT83‐treated parasites. For both situations, parasites normally progressed and entered schizogony; however, schizonts were abnormal, particularly following hesperadin treatment, and persisted for 60 hpt, indicating an arrested state and functional impairment in the completion of schizogony (Figures 3c and S4b). Co‐treatment of TAE684 or ZM‐39 with CQ, a known hemozoin formation inhibitor, showed an additive or indifferent effect, as evaluated by fixed‐ratio isobologram analysis, with ΣFIC₅₀ values of 1.3 and 1.4, respectively. By contrast, hesperadin was antagonistic to CQ (ΣFIC₅₀ 1.6) as well as to TAE684 (ΣFIC₅₀ 3.5) (Figure S5). Taken together, the data suggest that TAE684 exhibits polypharmacology, involving both the inhibition of hemozoin formation and PfArk1 inhibition, similar to other kinase inhibitors,^{[ 49 ]} but importantly, hesperadin has PfArk1 as its singular target.

In Silico Investigation of Inhibitor–Target Interactions in the ATP‐Binding Site of PfArk1

The specificity of hesperadin, an ATP‐competitive inhibitor, against PfArk1 was mechanistically investigated. PfArk1 shares ∼34% identity with mammalian AurA and B, with both the ATP‐binding signature and S/T PKs sites well‐conserved, including the catalytic lysine residue, Lys61 (Figures 4a and S5A). However, PfArk1 exhibits several critical changes in both the ATP binding site and the active site relative to both mammalian and protozoan Aur proteins, including alterations of two conserved Lys residues, one to Asp (at position 40) and Ala (at position 42), something only seen for the T. gondii Ark1, but not for PfArk2 (Figure S5b). Additionally, there are changes in the gatekeeper residue from Leu to a bulkier and more flexible Met at position 109 and an Ala to Cys change at position 112 within the hinge region (Figure 4a).

Figure 4.

In silico modeling predicts protein‐inhibitor interactions in the active site of PfArk‐1. a) Diagram of key characteristics of PfArk1 protein along with a protein sequence alignment of mammalian and Plasmodium falciparum Aur. b,c) Hesperadin binding pose in PfArk1 (green), overlayed with the binding pose of ATP (light blue), and the Xenopus laevis AurB crystal structure (2BFY, gray), with residue differences indicated in bold. The PfArk1 homology model was derived from the crystal structure of human Aur A (HsAurA) co‐crystallized with ATP (PDB 5DNR), with a root mean square deviation (RMSD) of 0.44Å and no stereochemical violations. Selected main chains and side chains of key residues conserved within the kinase domain are displayed (red and pink are for PfArk1, while gray is for XenAurB) with hydrogen bonds displayed as blue and salt bridge bonds as pink dashed lines.

In silico molecular docking studies revealed that hesperadin binds within the ATP‐binding pocket of PfArk1 but in a different pose to that observed for HsAurB and in the Xenopus laevis AurB co‐crystallized with hesperadin^{[ 50 ]} (Figure 4b,c). Hesperadin's indolinone moiety and sulfonamide group form hydrogen bonds, respectively, with the key conserved hinge region residues Glu110 and Tyr111. This causes the central phenyl to point into the active pocket of PfArk1, which would displace the α‐phosphate of ATP and prevent its interaction with the catalytic lysine (Lys61) (Figure 4c). Indeed, analogues lacking the bulky phenyl are not as active as hesperadin.^{[ 24 ]} Hesperadin is additionally stabilized in the ATP binding site pocket by an H‐bond between its piperidine ring and the unique Asp40 found in the Plasmodium enzyme (Figure 4c). PfArk1 is differentiated from PfArk2 in this position, with the latter containing a more conserved K–N modification compared to the HsAur, which does not accommodate stabilization of hesperadin to the same extent as in PfArk1. This data provides clarity on the selectivity of hesperadin for PfArk1 over both PfArk2 and mammalian AurA and B. Moreover, the unique Cys residue in the hinge region of PfArk1 is additionally particularly interesting from a drug development point of view as it undergoes unique interactions with the inhibitor, which could be exploited in the design of potential covalent inhibitors as highly potent and selective inhibitors, as has been shown for inhibition of PfCLK3 kinase.^{[ 51 ]}

Inhibition of PfArk1 Affects the Progression of Schizogony

We subsequently evaluated the effects of hesperadin and TAE684 against the proliferative stages of the parasites and their specificity for PfArk1. The rate at which hesperadin and TAE684 kill the parasite was evaluated by determining shifts in IC₅₀ over time.^{[ 52 ]} Hesperadin and TAE684 kill kinetics indicate that their effect is only evident with a significant IC₅₀ shift after 48 h, indicative of activity within one life cycle (Figure 5a). This profile resembles that seen for the PfPKG inhibitor ML10, which prevents parasite egress and invasion, but not for fast‐acting compounds such as CQ (Figure 5a). To further evaluate which developmental stage during asexual proliferation is affected by hesperadin and TAE684 treatment, we treated tightly synchronized parasites at 12 h intervals, correlating to ring, early, and late trophozoites and schizonts (Figure 5b). Hesperadin treatment had minimal impact on rings or trophozoites and did not limit the maturation from rings to trophozoites, whereas TAE684 treatment consistently affected trophozoite food vacuole formation (Figures 5b and 3c). The most pronounced effect for both compounds was associated with the completion of schizogony at ∼36–44 hpi, with no progression into the next cycle. Further evaluation of the morphologically aberrant schizonts revealed a significant reduction in the number of daughter merozoites formed for both hesperadin and TAE684 treatment (p < 0.000001, n = 30, unpaired Student's t‐test), an effect that was pronounced for the comparative treatment with TAE684 (Figure 5c). Hesperadin treatment did not prevent already formed schizonts from invading and forming rings in the next cycle, a process that was somewhat delayed by TAE684. Although this phenotype is reminiscent of the effect of PfPKG inhibition, which also prevents parasite egress and subsequent invasion,^{[ 53 ]} the lack of PfPKG inhibition for hesperadin and TAE684 and their indifferent effect when combined with ML10 (ΣFIC₅₀ of 1.1 and 1.3, respectively, Figure S6) support a PfPKG‐independent mechanism for these compounds. However, the data delineate a timeframe of action of hesperadin associated with the completion of late schizogony processes.

Figure 5.

Effect of PfArk1 inhibition on intra‐erythrocytic development. a) IC₅₀ speed of kill assay using unsynchronized PfNF54. CQ, pyrimethamine (PyR). b) Morphological evaluation (Giemsa‐stained thin smears) of P. falciparum asexual development following treatment with TAE684 (3xIC₅₀, ∼900 nM) and hesperadin (IC₉₉, ∼3 µM) (hpt – hours post‐treatment; hpi – hours post‐invasion). c) Nuclei count per schizont after treatment (n=30). Error bars represent the 95% confidence interval (CI) of the mean. d) Synchronized PfNF54 mature trophozoite and schizont populations treated with TAE684 and hesperadin for a 12 h period (solid line), washed off, and parasitemia measured at 12 h intervals using flow cytometry. e) Flow cytometric analysis of nuclear division following hesperadin and TAE684 treatment, sampled 12 hpt, and each subsequent 12 h until 60 hpt. Nuclei content was detected by consecutive staining with SYBR Green I (DNA fluorescence), detected in the FITC channel. Histograms overlaid for a representative sample of biological triplicates. All data are from three independent biological replicates, each performed in technical duplicates (n = 3, mean ± S.E.), significance was calculated using a two‐tailed Student's t‐test, *p<0.05, ***p<0.001, ****p<0.0001.

Flow cytometric quantification of this effect was performed after a 12 h drug treatment pulse on trophozoite and early schizont populations before drug washout (Figure 5d). Both hesperadin and TAE684‐treated trophozoites (∼30 hpi) were able to recover from a 12 h pulse and progress to schizonts; however, these parasites were unable to sufficiently establish reinvasion, resulting in a significant (p = 0.00015 and p = 0.00004, respectively, n = 3, unpaired Student's t‐test) decrease in parasitemia in the subsequent population. A similar significant effect was immediately evident with TAE684‐treated schizonts (p = 0.0002, n = 3, unpaired Student's t‐test) (∼40–42 hpi); however, no significant effect was observed for hesperadin‐treated schizonts (p = 0.139, n = 3, unpaired Student's t‐test), reminiscent of the phenotype observed in schizont‐treated samples (Figure 5b,d). Quantification of the nuclear content of treated parasites indicated entry into schizogony compared to untreated populations, but with the treatment primarily affecting parasites during schizogony, and the fraction of individual cells halted in the schizont stage containing ≥ 4n (DNA content) was higher in drug‐treated compared to untreated parasites (Figure 5e). Taken together, hesperadin as a PfArk1 inhibitor primarily affects parasite progression through schizogony (mid‐to‐late schizont development), where cells are unable to complete nuclear division and segregation successfully. However, it is ineffective against mature schizont stages, where these processes have already been completed.

PfArk1 is Critical to the Completion of Mitotic Processes

Fluorescence microscopy morphological evaluation of ABS parasites, where PfArk1 activity was inhibited by hesperadin or TAE684, revealed nuclear morphological abnormalities in schizonts. The nuclei of those treated with hesperadin appeared distinctly multi‐lobed, whereas TAE684‐treated schizonts had more elongated nuclei (Figure 6a), although packaging into daughter cell structures did occur, as evident by nuclear membrane detection (Figure S7a). The abnormal shape and decreased number observed during nuclear division were associated with abnormalities in MT structures in hesperadin‐ and TAE684‐treated schizonts. Compared to the well‐defined SPMTs seen in untreated schizonts, hesperadin treatment resulted in disorganized and interconnected MT structures. TAE684 decreased the appearance of MT structures entirely, with an overall decrease in nuclear content (Figure 6a).

Figure 6.

PfArk1 inhibition effect on microtubule and nuclear material morphology. Ring stage, synchronized, asexual intra‐erythrocytic PfNF54 parasites treated with hesperadin and TAE684 were harvested at ∼46 hpi. a) Representative images (maximum intensity projections) of the morphological abnormalities observed in nuclei (Hoechst, blue) & microtubules (anti‐⍺‐tubulin, pink), showing subpellicular microtubules (SPMTs, i), nuclei (ii). The images represent at least ten parasites per sample. Scale bars correspond to 2 µm. b) Representative images (maximum intensity projections) of expansion microscopy images of UT mature schizonts (top panel) compared to hesperadin‐treated (bottom panel). Red = tubulin, blue = DNA, and green = NHS‐ester. Scale bars correspond to 5 µm. c) Average microtubule (MT) branch lengths and number of MT branches per cell of UT compared to hesperadin‐treated, n = 4 schizonts, mean ± SD with an unpaired Welch's two‐tailed t‐test. d) Representative images of TAE684 and hesperadin‐treated Plasmodium falciparum male gametes labeled with anti‐⍺‐tubulin (green) and co‐stained with DAPI for nuclei (blue). Scale bars correspond to 5 µm.

Expansion microscopy (ExM) was subsequently used to provide more nuanced evaluation of hesperadin‐treated cells, which revealed the extent of abrogation of nuclear segmentation and packaging, as well as extensive MT defects (Figure 6b). Untreated parasites contained the expected well‐packaged and segregated daughter nuclei, with SPMTs clearly extending from the MTOCs associated with the CPs in proximity to well‐formed rhoptries (Figure 6b). No intranuclear MTs were evident as segregation was completed. By contrast, hesperadin treatment caused parasites to halt in schizogony at a point where the parasites contained multilobed nuclei, with the nuclear material not separated in most instances, and nuclei showing an increased nuclear volume. The MT organization was clearly affected by hesperadin treatment, with significantly extended MT structures (average lengths of 1.4 ± 0.7 µm, versus untreated parasites at 1.1 ± 0.4 µm, p = 0.04, n = 4, Figure 6c) spanning across and connecting different nuclear centers/CPs. Fewer of these MT structures were present per cell (29 versus 42 in hesperadin versus UT cells), and fewer could be associated with MTOC and the limited rhoptry pairs formed. This suggests that these microtubule structures are halted as interpolar MTs that are unable to retract during the interpolar to hemi‐spindle transition, as is typically required.^{[ 54} , ^{55 ]} This suggests that PfArk1 function is necessary for meta‐ and anaphase‐like transition processes during mitosis in the parasite. However, the potential for these aberrant MTs to be malformed, extended SPMTs cannot currently be excluded. Indeed, some of these aberrant MT structures closely localize to rhoptries connected to associated apical polar rings, as one would expect from SPMTs (Figure 6b).

We subsequently evaluated the broader involvement of PfArk1 in mitotic processes also present during male gametogenesis. Under normal conditions, when mature male gametocytes are successfully activated, exflagellation results in the formation of eight MT‐labeled flagella, each associated with segregated nuclei (Figure 6d) compared to non‐activated mature gametocytes (Figure S7b). Treatment with TAE684 and hesperadin was associated with some DNA replication proceeding normally, but the nuclear material remained highly compacted and lacked proper segregation. Strikingly, the MT structures in the drug‐treated cells are abnormal, thin, and wrapped around the nucleus, pointing to disorganized MT forming flagellae (Figure 6d). This suggests that DNA replication has mostly occurred, but the absence of PfArk1‐mediated signaling disrupted microtubule organization on the basal body in these stages. The extended, thin phenotype closely resembled that observed during late schizogony under PfArk1 inhibition (Figure 6b). These findings indicate that PfArk1 activity is required post‐DNA replication but before nuclear segregation and egress of male gametes, suggesting a mitotic block during gametogenesis.^{[ 37 ]} These findings suggest that the disruption of PfArk1 function during male gametogenesis does not affect DNA replication but rather may disrupt proper kinetochore attachment and spindle formation.

Our high‐resolution imaging provides the first evidence that disruption of PfArk1 function causes complete disorganization of MT structures required to coordinate and complete mitosis. The multilobed nuclei observed are reminiscent of the inhibition of AurB (equatorial Aur) in other organisms and apicomplexan parasites, which results in misaligned chromosomes, lagging chromatids, cytokinesis failure, and polyploidy.^{[ 26} , ³⁰ , ^{56 ]} Our observations support PfArk1 to be functionally similar to AurB in its association with kinetochores until metaphase, and the extended MT structures after loss of PfArk1 function imply a lack of translocation to the central spindle to coordinate cytokinesis. Although the disruption of AurB function can lead to disruption of microtubule dynamics and structure, these effects are also observed in the inhibition of AurA, which leads to unaligned chromosomes due to impaired centrosome separation and the formation of monopolar spindles, which could explain the spindle structures we observe in mature treated schizonts.^{[ 19} , ^{57 ]} Thus, based on these phenotypic observations, we suggest that PfArk1 fulfills the mitotic responsibilities of both AurA and AurB to control correct mitotic and interpolar spindle formation in ABS parasites and male gametocytes. Co‐localization with known markers of the centrosome (e.g., Centrins^{[ 58 ]}), kinetochores (e.g., NDC80^{[ 59 ]}), and chromosome passenger proteins (e.g., INCENP) will be needed to confirm if PfArk1 takes on a role of functional redundancy in P. falciparum for both AurA and B.

Conclusion

We demonstrate that PfArk1 is the most vulnerable target from the Ark protein family in P. falciparum, correlating with its essential requirement across multiple life cycle stages of the parasite.^{[ 10 ]} Moreover, PfArk1 inhibition abrogates mitotic‐associated proliferation processes, suggesting that PfArk1 is the primary Aur member governing mitotic‐related processes^{[ 14 ]} in ABS parasites, male gametes,^{[ 60 ]} and most likely also during hepatocyte schizogony. PfArk1 inhibition results in abnormal nuclear morphology, along with spindle structure defects that we previously described.^{[ 27 ]} With hesperadin as a tool compound that exclusively inhibits PfArk1, we could deduce the functional and mechanistic importance of this protein during mitosis. We propose that PfArk1 is the most important Aur regulating mitotic processes in the parasite and that hesperadin, as a potent and selective inhibitor of PfArk1 in P. falciparum, can be considered for future development as an antimalarial.

Supporting Information

The authors have cited additional references within the Supporting Information.^{[ 1} , ² , ³ , ⁴ , ⁵ , ⁶ , ⁷ , ⁸ , ⁹ , ¹⁰ , ¹¹ , ¹² , ¹³ , ¹⁴ , ¹⁵ , ¹⁶ , ¹⁷ , ¹⁸ , ¹⁹ , ²⁰ , ²¹ , ²² , ²³ , ²⁴ , ²⁵ , ²⁶ , ²⁷ , ²⁸ , ²⁹ , ³⁰ , ³¹ , ³² , ³³ , ³⁴ , ³⁵ , ^{36 ]}

Supplementary File with Supplementary Figures S1–S7 and Supplementary Methods.

Supplementary data files S1, S2.

Author Contributions

HL performed the work with MM, T Rabie, and JT (modeling and docking). HL and KM performed the hematin experiments. KM performed the hemozoin experiments under KJW, with NS performing recombinant protein work under LBC with KC. SGD, LCG, NB, MTF, MLdS, and JF contributed experimental data with interpretations and validations under JCN, MSL, MD, and EAW. LMB conceptualized the study and wrote the paper with HL. All co‐authors contributed to and approved the final version of the manuscript.

Conflict of Interests

The authors declare no conflict of interest.

Supporting information

Supporting Information

Supporting Information

Supporting Information

Langeveld H., Maepa K., Maree M., Thibaud J. L., Salomane N., Bridgwater R., Famodimu M. T., Godoy L. C., Pasaje C. F. A., Boonyalai N., de Souza M. L., Fong J., Rabie T., van der Watt M., Theart R. P., Ghidelli‐Disse S.,Niles J. C., Lee M. C. S., Winzeler E. A., Delves M. J., Chibale K., Wicht K. J., Coulson L. B., L.‐M. Birkholtz 0000‐0001‐5888‐2905, Angew. Chem. Int. Ed.. 2025, 64, e202518493. 10.1002/anie.202518493

Data Availability Statement

The data that support the findings of this study are available in the Supporting Information of this article.