https://orcid.org/0000-0001-5116-6973
Figures
Abstract
Malaria remains a global health threat exacerbated by emerging resistance to antimalarial therapies and insecticides, climate-driven outbreaks, and limited chemoprotective options. Here, we report the characterization of RUPB-61, the first orally bioavailable inhibitor of Plasmodium falciparum cGMP-dependent protein kinase (PfPKG). RUPB-61 prevents infection by P. falciparum and P. cynomolgi sporozoites, including the formation of hypnozoites by the latter. A single oral dose blocks liver infection by P. berghei sporozoites in vivo, demonstrating efficacy consistent with further development as a once-weekly prophylaxis based on pharmacokinetic modeling. The compound retains activity against field isolates resistant to chloroquine, mefloquine, cycloguanil, sulfadoxine and pyrimethamine, suggesting low likelihood of cross-resistance to existing antimalarials. Structural studies and free energy-based modeling guided-compound design prospectively validated the predictive accuracy of an in silico model of PfPKG interactions with this chemotype. While selectivity profiling identified off-target activity against human kinases, structural modeling provides a clear path for optimization. These results establish PfPKG inhibitors as promising candidates for chemoprevention and support further preclinical development of the RUPB-61 chemotype.
Author summary
Malaria remains a serious global health threat, made worse by growing resistance to existing drugs and the spread of insecticide-resistant mosquitoes. A critical gap in our prevention toolkit is the lack of safe, long-acting oral drugs that stop malaria parasites from infecting the liver - the first and obligatory step of every new infection. Here, we report the characterization of RUPB-61, a compound that blocks an enzyme that the malaria parasite needs to invade and survive in the liver. We show that a single oral dose completely prevents liver infection in mice, and that the compound remains active against parasites that are resistant to several currently used antimalarial drugs. Using X-ray crystallography and computer-based modeling, we determined how RUPB-61 binds to its target in atomic detail, providing a roadmap for designing improved versions. While the compound shows some unwanted interactions with human proteins, our structural data offer clear strategies to solve this challenge. Together, these findings provide a solid foundation for developing a new class of once-weekly oral malaria prevention drugs.
Citation: Dhiyebi H, Mbaye A, Thaniana A, Gilleran J, Eck T, Ashraf K, et al. (2026) An orally available PfPKG inhibitor blocks Plasmodium’s infection of the liver. PLoS Pathog 22(7): e1014322. https://doi.org/10.1371/journal.ppat.1014322
Editor: Leann Tilley, University of Melbourne, AUSTRALIA
Received: October 28, 2025; Accepted: May 30, 2026; Published: July 30, 2026
Copyright: © 2026 Dhiyebi et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All relevant data are within the manuscript and its Supporting Information files.
Funding: This research was funded by W81XWH2010386 from the Department of Defense to P.B. and J.Y.R. This project has been funded in whole or in part with Federal funds from the National Institute of Allergy and Infectious Diseases, National Institutes of Health, Department of Health and Human Services, under Contract No. 75N93022C00036 (S.L., A.D.). This research used resources the NYX beamline 19-ID, supported by the New York Structural Biology Center, at the National Synchrotron Light Source II, a U.S. Department of Energy (DOE) Office of Science User Facility operated for the DOE Office of Science by Brookhaven National Laboratory under Contract No. DE-SC0012704. The NYX detector instrumentation was supported by grant S10OD030394 through the Office of the Director of the National Institutes of Health (K.P.B.). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Introduction
Chemoprevention is one of two Target Product Profiles required for malaria elimination [1]. Target Candidate Profile 2 aims to address this need by developing drugs that prevent hepatic schizont and hypnozoite formation and are active against asexual stages. Such drugs will aid in reinvigorating the stalled progress in controlling malaria [2]. The first obligate step in malaria, referred to as the ‘pre-erythrocytic’ stage, is the infection by and development in the liver of Plasmodium sporozoites. Inhibition of sporozoite infection, through the use of vaccines or bed nets, significantly reduces disease severity and incidence [3,4]. Drugs that target pre-erythrocytic stages are an essential component of the anti-malarial effort because even if all symptomatic malaria cases could be treated, its global eradication requires the prevention of new infections [5]. Protection is needed for vulnerable populations, like children and pregnant women, as well as populations with reduced natural immunity due to decreased exposure to parasites, such as travelers from malaria-free zones visiting endemic areas and those living in newly malaria-free areas at risk of local outbreaks [5]. Infection can be prevented through vaccines, but current vaccines are only partially effective after 4–5 doses and are recommended only for pediatric use in Africa [6,7]. Drugs that block sporozoite infection can reduce hypnozoite formation by P. vivax, the most geographically widespread Plasmodium species, that is increasingly being detected in Africa and is the major cause of disease relapse [8,9]. Therapies that block its liver stage maturation and/or egress will prevent relapse caused by reactivated hypnozoites and reduce the number of asymptomatic individuals who serve as reservoirs of infection and transmission [10].
Current drugs targeting pre-erythrocytic stages – primaquine, tafenoquine and atovaquone – have several drawbacks. Primaquine and tafenoquine have a common chemotype, 8-aminoquinoline, which has serious side effects. It raises the risk of hemolytic anemia in individuals with genetic deficiencies in glucose-6-phosphate dehydrogenase [11], prevalent in malaria-endemic regions [12]. In addition, primaquine is a prodrug that undergoes CYP2D6-dependent metabolism for conversion into its active metabolite [13,14]. CYP2D6 genetic variants that are common in malaria-endemic regions are poor metabolizers of primaquine, limiting its efficacy in this population [13,15]. Meanwhile, atovaquone is expensive, and parasites acquire resistance easily [16]. These drawbacks create an acute need for drugs that target pre-erythrocytic stages with novel mechanisms of action.
PfPKG’s function is essential in pre-erythrocytic, asexual and sexual stages of the parasite [17–19]. It is chemically and biologically validated as a drug target during the parasite’s liver infection. Its inhibition impedes the motility of sporozoites and hence their invasion of hepatocytes [20]. The development and/or exit of mature liver stage parasites from the hepatocyte in the form of merosomes is also blocked when PKG function is lost [20,21]. Developmentally-arrested liver stages are targeted by the host immune response, and this response may potentially protect against subsequent infections [21]. PfPKG’s selective targeting is possible due to differences between its ATP-binding pocket and that of human PKG-I [22,23].
We previously reported a structure-activity relationship (SAR) study of a trisubstituted pyrrole (TSP, compound 1 [24]) with activity against PKG of related Apicomplexan parasites [25]. We used the SAR and crystallographic data to develop a free energy perturbation (FEP+) model for our series [25]. This work identified an early-lead compound, RUPB-61 (53 in [25]), with an IC50 of 13 nM against PfPKG and 50 nM against the drug-sensitive P. falciparum lab strain, 3D7. There is a 15-fold increase in its IC50 against mutant P. falciparum parasites carrying a Thr618Gln substitution of the ‘gatekeeper’ residue, demonstrating that its primary cellular target is PfPKG [25]. It has favorable in vitro pharmacokinetic (PK) properties - low preliminary cardiovascular risks as judged by hERG inhibition, high solubility and minimal first-pass metabolism, while also being more synthetically tractable due to the preferential formation of the cis isomer [25]. These data supported further investigation of the RUPB-61 chemotype. Here, we report its in vivo efficacy and activity against P. falciparum strains resistant to current anti-malarials. We solved the structures of P. vivax PKG (PvPKG) bound to the compound and its enantiomer. We used these data and the SAR of closely related analogs in the series to validate the FEP+ model for PfPKG interactions with this sub-chemotype. We tested RUPB-61’s binding affinity for the human kinome to assess its potential for cross inhibition of human kinases. We built in silico models for its interactions with these kinases to identify sites for future structural modifications aimed at mitigating these liabilities. These advances will enable the rational prospective design of molecules with improved potency against PfPKG and selectivity for the parasite.
Results
RUPB-61 prevents infection by P. falciparum and P. cynomolgi sporozoites
RUPB-61 is active against P. falciparum’s asexual erythrocytic stages and P. berghei sporozoites infecting HepG2 cells [25]. Here, we used infection of primary human hepatocytes (PHH) to test the activity of RUPB-61 against P. falciparum sporozoites. A phosphatidylinositol 4-kinase (PI(4)K) inhibitor, KDU-691 [26] was used as a positive control. RUPB-61 displays an IC50 of 79 nM in the P. falciparum sporozoite-PHH model with no cytotoxicity against PHH (Figs 1A, S1, Table 1). Next, we tested its effect on the hypnozoite-forming primate-infective species, P. cynomolgi. RUPB-61 prevented the formation of hepatic hypnozoites and schizonts with an IC50 of 190 nM and 170 nM, respectively (Fig 1B – 1C, Table 1). These data demonstrate that the scaffold holds promise for development as pre-exposure prophylaxis against human-infective sporozoites.
IC50 of RUPB-61 and TSP on infection of human or simian primary hepatocytes by P. falciparum and P. cynomolgi sporozoites. KDU-691 and tafenoquine were used as controls. Cellular cytotoxicity (CC50) was quantified in human and simian hepatocytes.
Infection was measured by quantifying the number of infected cells. Data shown are the mean of two biological replicates, each performed in technical duplicates. (B) Effect of RUPB-61 and KDU-691 on the number of P. cynomolgi hypnozoites formed in primary simian hepatocytes, normalized to vehicle control. Data shown are mean of three biological replicates, each performed in technical duplicates or triplicates. (C) Effect of RUPB-61 and KDU-691 on the number of P. cynomolgi schizonts in primary simian hepatocytes, normalized to vehicle control. Data shown are mean of three biological replicates, each performed in technical duplicates or triplicates.
RUPB-61 has a low propensity for cross-resistance
We previously reported the activity of RUPB-61 and related stereoisomers, RUPB-58, -59 and -60 (51, 50 and 52, respectively in [25]) against dihydroartemisinin (DHA)-resistant parasites that carry engineered mutations in PfKelch13 (CamC580Y) [25]. Their IC50 against the DHA-sensitive CamWT and DHA-resistant CamC580Y in ring-stage assays is similar [25]. Here, we assessed their cross-resistance against other clinically used antimalarial drugs by measuring their cellular activity against a panel of naturally occurring South American (7G8) and Southeast Asian (Dd2, K1 and TM90_C2B) P. falciparum field isolates. These strains collectively carry the most common alleles of pfdhfr, pfdhps, pfcrt and pfmdr1 that confer resistance to chloroquine, cycloguanil, sulfadoxine, atovaquone and pyrimethamine [27]. The change in the IC50 of RUPB-58–61 in the resistant parasites relative to 3D7 is below 5-fold. These results demonstrate that the compounds maintain their activity against multidrug resistant P. falciparum parasites (Table 2, S2 Fig).
IC50 of RUPB-58 - 61 against P. falciparum isolates resistant to current antimalarial drugs. Data shown are mean (standard error) of 3 biological replicates.
Replacement of the PfPKG ‘gatekeeper’ Thr618 with larger amino acids (e.g., Met, Phe, Tyr, Leu, Ile or Gln) are expected to lead to resistance to RUPB-61 and structurally related analogs since their bulkier sidechains decrease the volume of the binding pocket. Previous in vitro resistance studies using PfPKG inhibitors of different chemotypes either failed to select resistant mutants or found only low-level resistance from mutations in non-targets [28,29]. Here, we evaluated the risk of naturally occurring escape ‘gatekeeper’ mutants. We examined high-resolution genetic information from 20,892 parasite samples present in the MalariaGEN Pf7 database. These include isolates from major malaria-endemic regions of the world, laboratory clonal and mixed samples, and the results of experimental genetic crosses. Analysis revealed natural variations in the PfPKG protein – 493 synonymous changes (in 4823 samples) and 523 non-synonymous changes (in 1164 samples) (S1 Table). At the ‘gatekeeper’ position, two isolates carried synonymous changes (Thr618Thr), and one isolate carried a non-synonymous mutation (Thr618Ser). The number of samples with changes in the ‘gatekeeper’ position was not statistically significant compared to the total samples analyzed. The extreme rarity of non-synonymous mutations at the ‘gatekeeper’ position suggests that randomly occurring pre-existing mutations are unlikely to facilitate resistance to the RUPB-61 chemotype through this mechanism under natural conditions.
RUPB-61 is orally available
One of the requirements for a chemoprotective agent is oral administration at a frequency of at most once-weekly and ideally, once-monthly. Achieving these benchmarks requires compounds with extended oral half-lives. We tested RUPB-61’s ADME properties in mice after oral and intravenous dosing at 10 mg/kg. Plasma concentrations from both dosing regimens were identical from 2 - 24 h (Fig 2). The calculated oral bioavailability is 91%, and the oral half-life is approximately 9 h (Tables 3, S2). Allometric scaling translates this to approximately 2–4 days in humans [30]. The compound maintains exposure at >100 nM (~2x EC503D7) for 24 h with relative clearance around 50 mL/min/kg and large volume of distribution of 47.3 L/kg (Fig 2, Table 3, S2 Table).
Pharmacokinetic properties were determined after oral (PO) or intravenous (IV) administration of compound to mice. Data shown are average of 3 mice. Tmax: time to maximum distribution; Cmax: Maximum serum concentration; AUC: Area under the Curve; Cl: clearance; INF: infinite; MRT: Mean Residence Time; Vss: Volume of distribution.
Plasma concentration in mice was measured over a 24 h period after intravenous (IV) and oral (PO) dosing with 10 mg/kg (n = 3 per route).
A single oral dose of RUPB-61 is efficacious in a murine model of liver infection
The similarity in RUPB-61’s activity against P. falciparum and P. berghei sporozoites strongly supports the use of P. berghei as a model for testing compound activity in vivo. We tested the efficacy of RUPB-61 and the parent compound, TSP, against liver infection by P. berghei sporozoites expressing luciferase and GFP (PbLuc-GFP). In the first experiment, mice were administered three doses, intravenously (IV) or orally (PO), of compound (10 mg/kg) - two prior to sporozoite inoculation to block sporozoite invasion of hepatocytes and the third at 36 hpi to prevent egress of liver stages (Fig 3A). In vivo bioluminescent imaging (IVIS) of mice was utilized to detect and quantify relative parasitemia in the liver and flow cytometry was used to detect the appearance of blood stage parasites. Vehicle-treated mice were used as a negative control. For both treatments, liver parasitemia was detectable at 44 h post infection (hpi) in all vehicle-treated mice (10/10) (Fig 3B-3C, S3 Fig). Additionally, a majority of mice treated (IV) with TSP (3/5) displayed liver infection at 44 hpi. In contrast, none of the mice treated with RUPB-61 (10/10) showed liver parasitemia as late as 144 hpi (Fig 3B-3C, S3 Fig).
(A) Experimental design to test the efficacy of RUPB-61 against PbLuc-GFP sporozoite-initiated infection in mice. Mice were administered 3 doses of compound intravenously or orally. Liver parasitemia was quantified using bioluminescent imaging and blood stage parasitemia was detected using FACS. (B) Liver parasitemia at 44 hpi in infected mice administered compounds IV. (C) Liver parasitemia at 44 hpi in infected mice administered compounds orally. Parasitemia was measured in radiance (p/sec/cm2/sr). Dotted line represents background signal.
A decrease in liver stage parasitemia is expected to delay the appearance of blood stage parasites, as there are fewer hepatic merozoites to initiate erythrocytic infection. Therefore, we quantified the time taken to detect blood stage parasites after sporozoite inoculation (pre-patent period) in these mice. The pre-patent period in vehicle-treated animals was 3.6 days under IV administration and 4.6 days in oral treatment, while none of the RUPB-61-treated mice displayed blood stage parasites in the 3 week observation period (Table 4). In the group that received 3 intravenous doses of TSP, 2/5 mice did not show blood stage parasites for 3 weeks pi. This implies that the lack of detectable liver stage parasitemia at 44 hpi represents a complete block in sporozoite infection. In the remaining TSP-treated mice that had detectable liver parasitemia at 44 hpi, the pre-patent period trended longer than vehicle-treated mice of the same cohort (4.3 days versus 3.6 days) (Table 4), although the difference was not statistically significant.
Prepatent period of sporozoite infection was determined through detection of PbLuc-GFP asexual blood stages using flow cytometry. sd: standard deviation.
Finally, we tested the efficacy of single-dose RUPB-61 when administered intravenously or through oral gavage. Based on the in vivo pharmacokinetic data (Fig 2), the oral dose was provided at -3 hpi and the intravenous dose at -0.5 hpi to ensure high levels of compound in plasma at the time of sporozoite inoculation. Vehicle-treated mice (9/10) were positive for blood stage parasites by 4.8 days and 5.75 days, respectively whereas none of the RUPB-61-treated mice (10/10, PO and IV) developed detectable blood stage infection for the 3-week monitoring period (Table 4). We conclude that RUPB-61 is sufficient to prevent infection of the liver by P. berghei sporozoites at a single dose of 10 mg/kg delivered orally.
RUPB-61 has moderate selectivity for human kinases
A major issue affecting the further development of RUPB-61 is its human-kinome selectivity. To identify its significant off-target liabilities, we quantified its binding affinity for 400 unique human kinases (S4 Fig). The screen consists of binding competitions between the test compound and control ligands for each of the target human kinases. At 1 M RUPB-61 displaced >99% of the control ligands for CIT, ROCK1 and ROCK2, hence displaying moderate selectivity against the human kinome. We then determined the binding affinity (Kd) of RUPB-61 for the three kinases (Table 5). The Kd for CIT was the lowest, identifying it as the foremost off-target of RUPB-61.
Binding affinities (Kd, nM) of compounds for CIT, ROCK1 and ROCK2 were determined experimentally and predicted using a toxicology tool. Experimental ∆G was calculated using the relationships IC50 Kd and ∆G = -RTln(Kd) with 300K for temperature. ND: not determined.
Further development of RUPB-61 will require de-risking these kinases. In order to better understand their interaction with the RUPB-61 series of compounds, we determined the binding affinities of analogs with distinct structural modifications, RUPB-8, -18, -34 and -69 (Table 5), against ROCK1/2 and CIT. The compounds were selected because they carry significant structural modifications relative to RUPB-61 and maintain activity against PfPKG (Table 5). Compared to RUPB-61, the Kds of RUPB-8 (3-cyanopyridine), RUPB-18 (3-cyanopyrrole) and RUPB-34 (4-methoxyphenyl) are 40–700-fold higher for ROCK1 and ROCK2. The Kds of RUPB-18,-34 and -69 (dimethylbenzyl alcohol) for CIT were 5–125-fold higher compared to RUPB-61. In addition to experimental SAR, we utilized a predictive toxicology tool (Schrödinger Predictive Toxicology Models) for computational assessments of relative binding affinities of a given compound for human kinases. CIT was not available within the in silico kinase panel. However, since ROCK1 and CIT are both members of the AGC kinase family we postulated that ROCK1 modeling may also serve as a proxy for CIT and ROCK2. We compared experimental and model-predicted Kds (kcal/mol) of ROCK1 for RUPB-8, -18, -34, -61 and -69 (Table 5). The Pearson correlation coefficient squared (R2) for their measured and predicted ROCK1 ∆G is a 0.34. This contrasts with an R2 of 0.55 for our predicted IFD-MD PfPKG model (Table 6). The modest R2 for the ROCK1 model is most likely a reflection of the limited size of the data set. With additional data points and further refinement, the ROCK1 model should discriminate between weak, moderate and strong binders, leading to more accurate predictions.
The overall FEP statistics of the two models in terms of the squared Pearson correlation coefficient R2, between predicted and experimental ∆G and the root-mean-squared error (RMSE) of all pairwise ∆∆Gs. In parentheses are the 95% confidence interval RMSE values using bootstrapping.
Atomic structure of PvPKG bound to RUPB-61 and its enantiomer, RUPB-60
Further development of RUPB-61 will include improvement in its potency. Rational design of analogs will be facilitated by structural information. Towards this end, we solved the structure of RUPB-61 bound to P. vivax PKG (PvPKG) at 2.8 Å (PDB 9P74) (Fig 4, S3 Table). We compared the predicted RUPB-61-bound PfPKG structure from our prior work [25] with the experimental RUPB-61-bound PvPKG structure. These results showed that ligand poses are extremely similar. Structural superposition of the two kinase structures leads to a ligand heavy-atom RMSD of 0.82 Å. In both structures, the hinge interaction is formed using the pyridine nitrogen as an acceptor for a backbone NH (Val621 in PfPKG, Val614 in PvPKG) with aromatic CH interactions ortho to the pyridine nitrogen interacting with backbone carbonyls (Val621 and Glu619 in PfPKG, only Glu612 in PvPKG). The most significant difference between the two models lies in the conformation of the ‘P-loop’. In the predicted PfPKG structure, the P-loop dips down towards the ligand and in doing so allows Asp682 of the DFG motif (Asp-Phe-Gly motif) to hydrogen bond with the Thr551 side chain hydroxyl and the backbone NH in the P-loop. This Asp forms a hydrogen bond to the ligand pyrrole NH. In the experimental PvPKG structure, the P-loop is pointing up and away from the ATP pocket. The homologous Thr in PvPKG, Thr544 is unable to form an interaction with the DFG motif Asp, Asp675 in PvPKG, and evidently the hydrogen bond with the pyrrole is not formed. An additional difference is in the interaction with Glu625 in PfPKG. In the PvPKG structure, the equivalent Glu618 is forming a salt bridge to the ligand piperidine while in the PfPKG structure, Glu625 points out into solution.
Overlay of the predicted PfPKG structure with RUPB-61 [25] (white) and the experimental PvPKG structure (green). The two most significant differences between the two models are in the interactions of two carboxylates, E625 (E618 in PvPKG) which forms a salt-bridge to the ligand piperidine only in PvPKG and D682 (D675 in PvPKG) which only forms a charged hydrogen bind to the pyrrole NH in the PfPKG model.
RUPB-61 is chiral at the 3- and 4-positions of the piperidine ring. To understand the effect of chirality on its interactions with PfPKG, the structure of its diastereomer, RUPB-60, was also solved in PvPKG (PDB 9P75). RUPB-60 places a methyl substituent towards the solvent-exposed side of the ATP binding pocket (S5 Fig, S3 Table). In the RUPB-60 structure, the C-terminal loop is resolved with an additional three residues compared to the RUPB-61 structure. These additional three residues include Ala816 which hydrophobically packs against the methyl piperidine substituent. This Ala is conserved in PfPKG as Ala823, however, the additional hydrophobic packing does not appear to affect potency as RUPB-60 and RUPB-61 have similar IC50 against PfPKG variants [25].
To address whether the differences between the IFD-MD PfPKG model and experimental PvPKG model are material to modeling, we performed FEP using an identical congeneric series of compounds. We selected our previously reported compounds [25] as well as four additional compounds (one of which is represented as a pair of stereoisomers) which were prospectively designed using our prior IFD-MD model.
The statistics of the two models appear quite similar with a marginal improvement using the experimental PvPKG model (Table 6, S6 Fig). The RMSE values have overlapping 95% confidence intervals which is consistent with the two models appearing near equivalent in FEP. This would suggest that the differences between the two models may be due to thermal fluctuations, for example the solvent-exposed piperidine salt-bridge, as the crystal structure is solved within a 100K liquid nitrogen stream while the IFD-MD model is simulated at 300K. This also suggests that the PvPKG model and X-ray structures are valid for use as a proxy for PfPKG during molecule design.
Experimental confirmation of FEP+ model
The solved structures of PvPKG bound to either RUPB-61 or RUPB-60 match qualitatively and quantitatively (Table 6, S3 Table) to our previously reported FEP+ model of PfPKG [25]. For orthogonal validation of the FEP+ model, we designed molecules to test some key predictions (Table 7). RUPB-68, RUPB-69, and RUPB-72 explore displacement of a collection of high energy waters identified using WaterMap in our prior publication [25] and replacement of these interactions with an added hydroxyl or ester group compared to TSP. As a negative control, we tested the model’s accuracy at predicting weaker compounds by replacing the methyl groups in RUPB-69 with ethyl groups in RUPB-73. These groups are expected to clash in the binding cavity, resulting in weaker activity as confirmed by experimental measurements (Table 7). These data are included in the aggregate statistics shown in Table 6 with individual ligand statistics in S4 Table, but the predicted versus experimental affinity for these prospective ligands are shown separately in Table 7. Errors between prediction and experiment are around 1 kcal/mol which is consistent with published FEP+ performance using high quality crystal structures [31]. Both calculations (FEP+ in PvPKG and PfPKG) are run at 25 ns, 5x larger than the default, to help eliminate any random errors due to insufficient sampling or convergence. The congruence between these data, the co-crystal structures and predictions of the FEP+ model demonstrates the validity of the FEP+ model. This supports the use of the FEP+ model for future prospective design of more potent PfPKG inhibitors and for maintaining potency while eliminating undesirable human kinase binding.
IC50 was converted from concentration to kcal/mol using the relationships IC50 Kd and ∆G = -RTln(Kd) with 300K for the temperature.
The IFD-MD model can also be used to predict the binding affinity of the Thr618Ser mutation found in one isolate in MalariaGen database. We used the protein FEP approach, which assumes the ligand remains roughly unchanged while introducing the T618S mutation in silico to PfPKG. We measured the change in binding free energy (∆∆G) from this mutation. FEP modeling predicts that mutating T618 to S618 improves the binding affinity to the ligand slightly by 0.5 ± 0.43 kcal/mol, indicating that this mutation is not predicted to significantly affect binding. These calculations strongly suggest that the parasites carrying the Thr618Ser mutation in PfPKG will remain sensitive to RUPB-61.
Discussion
The global malaria landscape is under mounting threat due to resistance to artemisinin-based combination therapies and the spread of insecticide-resistant mosquito vectors. Climate-driven instability further compounds the risk, exemplified by the 5-fold surge in malaria cases in Pakistan during the 2022 floods. Even historically malaria-free regions, such as the United States, reported locally acquired infections in 2023 - the first in two decades - highlighting the urgent need for sustained innovation in prevention strategies.
A cornerstone of renewed malaria control efforts will be the development of drugs targeting pre-erythrocytic stages. These agents can complement artemisinin-based therapies in three-drug combinations or synergize with next-generation vaccines to provide robust chemoprotection, particularly for seasonal transmission or endgame elimination. Protection is needed for pregnant women, children, individuals traveling from malaria-free zones into endemic areas and for populations in newly malaria-free areas at risk of local malaria outbreaks, as these groups have reduced natural immunity to infection and disease [5]. However, there is a significant dearth of candidates that meet the stringent potency, pharmacokinetic, and safety criteria required for long-acting prophylactic agents, especially those intended for monthly oral or quarterly parenteral administration.
Our study identifies RUPB-61 as the first orally bioavailable inhibitor of PfPKG with demonstrated single-dose efficacy in a murine model. PfPKG is a compelling target for chemoprotection, given its dual essential functions in liver stage and blood stage egress and invasion. The lower parasite burden and lack of replication cycles in the liver further increase the vulnerability of the pre-erythrocytic stages to kinase inhibition. RUPB-61 exploits two temporally distinct windows of parasite susceptibility - during hepatocyte invasion and again during merosome formation [20,21] - positioning it as a versatile candidate for both prevention and treatment. The inhibitory effect of PfPKG inhibitors on P. cynomolgi sporozoites suggests that, in addition to preventing or mitigating disease, they could reduce the formation of P. vivax hypnozoites, which are the major cause of vivax-malaria relapse and the number of asymptomatic individuals who serve as reservoirs of P. vivax infection and transmission [10]. Studies in FRG huHEP mice will further define its efficacy against P. falciparum sporozoites and potential activity against P. vivax hypnozoite formation [32].
Importantly, PfPKG inhibitors fulfill the chemoprevention TPP’s requirement of a mechanism of action distinct from current antimalarials. Our cross-resistance studies demonstrate that RUPB-61 maintains potency in blood stage growth assays against parasites resistant to chloroquine, cycloguanil, sulfadoxine, pyrimethamine or dihydroartemisinin [25]. These findings suggest potential for co-formulation with existing frontline therapies without antagonism or cross-resistance, but this needs to be tested in combination drug assays.
Resistance to PfPKG inhibitors appears infrequent and biologically costly. Laboratory studies and data from gatekeeper mutants (e.g., T618Q) suggest that resistance-conferring mutations compromise ATP binding and enzymatic efficiency [33,34], reducing sporozoite yield and infectivity [20]. This supports the evolutionary robustness of the target and a low likelihood of transmission of strains arising from gatekeeper mutations. Future studies in P. falciparum mosquito transmission models will be crucial for validating this prediction. In addition, studies on lab-evolved resistance under drug pressure are needed to estimate the minimum inoculum of resistance and the susceptibility of the scaffold to the development of resistance in parasites.
The pharmacokinetic profile of RUPB-61 is also promising. Its similar IV and oral plasma concentrations - unusual for kinase inhibitors - result from high bioavailability and long oral half-life, which are key parameters for preclinical candidate advancement. The long half-life of the compound can result from intrinsic metabolic stability and/or depoting in tissues. RUPB-61’s volume of distribution suggests the latter. Since RUPB-61 is not highly lipophilic, its tissue depoting is unlikely to result from non-specific binding to membranes. We speculate that it may result from active uptake by transporters into tissues.
Structure-guided design has been instrumental in advancing the RUPB-61 chemotype. Prospective compounds synthesized using the IFD-MD model and FEP+ affinity predictions, prior to any retrospective fitting, validated the predictive power of the models. The ability to design novel, high-affinity ligands de novo represents a critical capability for advancing preclinical optimization and derisking development. One liability of the current scaffold is off-target activity against human kinases, notably CIT and ROCK1/2. However, our structure-based modeling provides a rational framework for selectivity optimization. Our results identify the fluorophenyl ring as an attractive target for substitutions aimed at improving selectivity. Importantly, these modifications can significantly enhance potency against PfPKG, for example the 10-fold improvement in the IC50 of RUPB-69 compared to RUPB-61. Future work will develop and apply experimental SAR across a larger number of analogs to drive iterative molecule design. These SAR data will also be used to build a more accurate IFD-MD model of ROCK1 (with the ROCK1 model serving as a proxy for the related ROCK2 and CIT). Greater accuracy should lead to predicted binding affinities that are within 1 kcal/mol of experimental determinations. With this refinement, the in silico models can be used to triage prospective molecules based on selectivity for the human kinome, hence assisting in rationalizing and prioritizing compound synthesis. Previous studies demonstrate that optimization of in silico models can be achieved using a ligand set of 9 – 10 compounds comprising strong, moderate and weak binders [35].
In summary, the RUPB-61 chemotype represents a promising lead for the development of a new class of orally active, long-acting, and resistance-resilient chemoprotective agents. Its dual-stage activity, favorable pharmacokinetics, and distinct mechanism of action position it well to fill a critical gap in the malaria elimination toolkit. With continued optimization and preclinical validation, PfPKG inhibitors may emerge as key components of integrated strategies for malaria prevention, treatment, and eventual eradication.
Methods
Ethics statement
All animal work in this project was reviewed and approved by the Institutional Animal Care and Use Committee of Rutgers New Jersey Medical School, following guidelines of the Animal Welfare Act, The Institute of Laboratory Animal Resources Guide for the Care and Use of Laboratory Animals, and Public Health Service Policy.
Parasite cultures and growth inhibition assays
Asexual blood stage P. falciparum strains, including 3D7 and drug-resistant field isolates, were cultured using standard conditions in RPMI 1640 with 0.5% Albumax as described previously [25]. EC₅₀ values were determined using a SYBR Green I fluorescence assay after 72 h of compound exposure. EC50 values were derived by nonlinear regression in GraphPad Prism version 10.
P. falciparum and P. cynomolgi sporozoite production and isolation
Liver stage antimalarial activity was evaluated using sporozoites from P. falciparum (NF54 strain) and P. cynomolgi bastianellii (B strain). P. falciparum-infected Anopheles stephensi mosquitoes were reared at WRAIR, while P. cynomolgi-infected Anopheles dirus mosquitoes were obtained from the Armed Forces Research Institute of Medical Sciences (AFRIMS), Thailand. Sporozoites were isolated by dissecting salivary glands under sterile conditions, followed by gentle mechanical disruption. Freshly isolated sporozoites were immediately used for hepatocyte infection.
Primary hepatocyte culture and sporozoite infection
Primary cryopreserved hepatocytes were selected based on host-parasite compatibility: primary simian hepatocytes (lot OQB) for P. cynomolgi and human hepatocytes (lot BGW) for P. falciparum (BioIVT Inc., Baltimore, MD, USA). Cells were thawed according to the manufacturer’s protocol and seeded into 384-well collagen-coated assay plates (Greiner Bio-One, Monroe, NC, USA) using InVitroGro CP plating medium (BioIVT) [36,37]. Seeding densities were optimized for uniform monolayer formation, and hepatocytes were allowed to adhere for 24–48 hours before infection. Sporozoites were added at a 1:1.1 hepatocyte-to-sporozoite ratio within 48 hours post-seeding. Automated robotic systems were used for media changes and compound additions to ensure assay consistency and minimize variability associated with manual pipetting.
Drug treatment regimens for P. falciparum and P. cynomolgi sporozoite assay
Test compounds (PfPKG inhibitors) and assay drug controls (KDU-691 and tafenoquine) were solubilized in DMSO or PBS and serially diluted into 8- or 12-point, 2- or 3-fold series, achieving final assay concentrations ranging from 20 µM to 0.019 µM. Compounds were introduced using a calibrated pin tool controlled by robotic arms (V&P Scientific), ensuring a 1:1000 dilution into assay wells. Media were replenished daily with fresh drug-containing HCM. Compounds were added 1-hour post-infection (p.i.) (day 0) and administered daily for 3 days. Cells were fixed on day 6 p.i. of P. falciparum culture and day 8 for P. cynomolgi infected culture, coinciding with liver stage schizont maturation.
Fixation and immunofluorescence staining for P. falciparum and P. cynomolgi assays
Cells were fixed using 4% paraformaldehyde for 20 minutes at room temperature, followed by three PBS washes, and stored at 4°C until further processing. Parasite detection was performed using species-specific primary antibodies, with P. cynomolgi parasites labeled using a mouse monoclonal anti-Plasmodium GAPDH antibody (1.6 μg/mL, European Malaria Reagent Repository) and P. falciparum parasites detected using a rabbit polyclonal anti-PfHSP70 antibody (1:70 dilution, LS Bio, USA). Primary antibodies were diluted in PBS containing 1% bovine serum albumin (BSA) and 0.03% Triton X-100 and incubated overnight at 4°C. Following primary antibody incubation, cells were washed and incubated overnight at 4°C with Alexa Fluor Plus 555-conjugated goat anti-mouse/anti-rabbit IgG secondary antibodies (2 μg/mL, Invitrogen, USA). Nuclei were counterstained using Hoechst 33342 (5–10 μg/mL) for one hour at room temperature before imaging. Each assay run included a control plate containing uninfected hepatocytes stained under identical conditions.
High-content imaging and analysis
High-content imaging was performed using the Operetta CLS system (Revvity) and image analysis was performed with Harmony software version 4.9 (Revvity, Waltham, MA, USA). Images were acquired at 10x or 20x magnification using DAPI and TRITC fluorescence channels. Following the established protocol [38,39], parasites were classified as hypnozoites or schizonts based on fluorescence intensity (mean and maximum), morphological features (cell roundness and size), and segmentation criteria. Hepatocyte nuclei were identified using Hoechst 33342 staining. Background signal was captured in uninfected, stained wells and subjected to the same parasite detection algorithm. Estimate erroneous parasite counts were then subtracted from infected well counts. Hepatocyte viability was determined by quantifying nuclear counts per well using Hoechst stain intensity thresholds. A total of 9 or 25 non-overlapping fields were captured per well, covering more than 95% of the total well surface area. An in-house built Python-based image analysis script identified parasite objects using predefined fluorescence intensity thresholds. Objects were validated based on size and shape criteria and confirmed via DAPI co-localization to eliminate false positives. Parasite size was quantified by calculating the average object area (µm²) of validated parasite structures per well. The classification pipeline was applied across all plates, accounting for debris and biological background estimated from uninfected stain controls. These values were subtracted from individual plate parasite counts to obtain normalized data. Dose-response inhibition and IC50 were calculated relative to in-plate negative controls (DMSO-treated wells). A four-parameter logistic regression model was fitted to the percentage survival or inhibition versus log concentration data using GraphPad Prism (Windows version 10.0.3, GraphPad Software Inc., USA) or a Python-based grid algorithm [40].
Efficacy studies
In vivo chemoprotective efficacy was evaluated using luciferase-expressing P. berghei ANKA sporozoites. Female Swiss-webster (Taconic) wild-type mice (6–8 weeks old) were administered RUPB-61 (10 mg/kg in PBS) orally or intravenously prior to intravenous injection of 0.5 - 1 x 104 sporozoites. Liver stage parasite burden was measured 42–48 h p.i. by bioluminescence imaging (IVIS Spectrum). Photon flux from the liver region was quantified using Living Image software. All animal procedures were approved by the Rutgers Institutional Animal Care and Use Committee (IACUC).
Bioinformatic analysis of P. falciparum genomes
This study included sequences of P. falciparum generated by the MalariaGen consortium and available on the European Nucleotide Archive (https://www.ebi.ac.uk/ena). Whole-genome sequencing data were downloaded using the ena-file-downloader.jar tool to fetch paired reads in FASTQ format via FTP (File Transfer Protocol). Sequencing reads were trimmed for quality control using Sickle [41] with the default parameters (length threshold of 20 and Phred quality score threshold of 20). Trimmed files were then aligned to the Pf3D7_01_v3 reference genome using bwa (v0.7.17-r1188). Samtools (v1.3.1) were used to generate BAM, sorted BAM, and index BAM files. Mutations were then identified from the aligned reads using Bcftools (v2.1) mpileup to generate genotype likelihoods at each genomic position with coverage and Bcftools call for variant calling. Variants were then annotated using SNPEff with Pf3D7_01_v3 as the reference genome. The resulting VCF files were compressed, indexed, and filtered by location (between 1490321 and 1494453 on chromosome 14). The VCF files were combined into one file for all samples. This final VCF file was downloaded and imported into R for further analysis. PfPKG sequence used is from positions 1490315 – 1494452 on chromosome 14.
Pharmacokinetic analysis
Mice were dosed with RUPB-61 (10 mg/kg, P.O. or I.V.) and blood collected over 24 h. Plasma concentrations were measured by LC-MS/MS. Non-compartmental PK parameters were calculated using Phoenix WinNonlin. Human half-life predictions were derived by allometric scaling for half-life (t1/2): t1/2,human = t1/2,mouse × (Whuman/Wmouse)0.25, where W = body weight. The exponent 0.25 is commonly used for time-based pharmacokinetic parameters. We assume an average mouse weight of 0.025 kg and an average human weight of 70 kg. With a mouse half-life of 9–12 hours, we compute the human half-life range to be 67.5 – 90 h.
Protein expression
P. vivax PKG was purified as previously described [42]. Briefly, Sf9 cells expressing the protein were flash frozen in liquid nitrogen. They were lysed hypotonically by thawing in 20 mM HEPES pH 7.0, 2 mM MgCl2, 0.5% Triton X-100 and EDTA-free protease inhibitor tablets (Roche). The whole cell lysate was flash frozen in liquid nitrogen. Thawed lysate was adjusted to 300 mM NaCl and 5% glycerol, and 1 mM TCEP, and insoluble material was removed by centrifugation at 10,000 RCF for 1 hour. The supernatant was treated with benzonase for 30 minutes, and debris pelleted at 230,000 rcf for 1 hour. Protein was purified by IMAC with an elution gradient of buffer A (20 mM HEPES pH 7.0, 300 mM NaCl, 1 mM TCEP, 5% glycerol) to buffer A with 500 mM Imidazole. Peak fractions were concentrated in a 10 kDa MWCO concentrator (Pall), then enriched by SEC on a 26/600 superdex 200 column (Cytiva) equilibrated in buffer A. Protein from peak fractions was concentrated to 10 mg/ml and flash frozen in liquid nitrogen.
Crystallization
Purified PvPKG, in 20 mM HEPES pH 7.0, 300 mM NaCl, 5% (v/v) glycerol, 1 mM TCEP was diluted to 10 mg/mL for crystallization screening. Protein-ligand complexes were prepared by adding 2 mM RUPB-60 and RUPB-61, from 100 mM DMSO stock solutions to aliquots of the protein and incubating on ice for 1 hour. All crystallization experiments were conducted using an NT8 drop-setting robot (Formulatrix Inc.) and UVXPO MRC (Molecular Dimensions) sitting drop vapor diffusion plates at 17 °C. 100 nL of protein and 100 nL crystallization solution were dispensed and equilibrated against 50 µL of the latter. Prismatic crystals were obtained overnight from the Morpheus [43] screen (Molecular Dimensions) condition B3 (20%(v/v) glycerol, 10% w/v PEG 4000, 100 mM Imidazole/MES, pH 6.5, 30 mM NaF, 30 mM NaBr and 30 mM NaI). Crystals were cryoprotected by layering the well solution onto the drops, harvesting directly and storing in liquid nitrogen. X-ray diffraction data were collected at the National Synchrotron Light Source II (NSLS-II) beamline 19-ID-I (NYX) using a Dectris Eiger2 9M XE pixel array detector. Data collection and refinement statistics are summarized in Supplementary Information.
Structure solution and refinement
Intensities were integrated using XDS [44,45] via Autoproc [46] and the Laue class analysis and data scaling were performed with Aimless [47]. Structure solution was conducted by molecular replacement with Phaser [48] using a previously determined structure (PDB Intensities were integrated using XDS [44,45] via Autoproc [46] and the Laue class analysis and data scaling were performed with Aimless [47]. Data from 6 crystals, obtained from the same crystallization condition, were scaled together to improve the data completeness and multiplicity due to the low symmetry P1 lattice. Structure solution was conducted by molecular replacement with Phaser [48] using a previously determined structure (PDB Intensities were integrated using XDS [44,45] via Autoproc [46] and the Laue class analysis and data scaling were performed with Aimless [47]. Data from 6 crystals, obtained from the same crystallization condition, were scaled together to improve the data completeness and multiplicity due to the low symmetry P1 lattice. Structure solution was conducted by molecular replacement with Phaser [48] using a previously determined PfPKG structure (PDB 4RZ7) as the search model. Structure refinement, manual model building and validation were conducted with Phenix [49], Coot [50] and Molprobity [51] respectively.
Accession codes
The coordinates and structure factors have been deposited to the Worldwide Protein Databank (wwPDB.org) with the accession codes 9P75 (RUPB-60) and 9P74 (RUPB-61).
Modeling
RUPB-61 was docked into prepared PDB structures of ROCK1 using the 25–2 release of Schrödinger’s IFD-MD [52]. The representations of ROCK1 were PDB IDs 2ETR, 3TVJ and 2ETR with the c-term loop deleted. IFD-MD utilizes a combination of induced-fit docking and molecular dynamics to predict the pose of RUPB-61 starting from the ROCK1 structures. Each predicted ligand-receptor complex produced by IFD-MD (15 poses in total for ROCK1) was then scored using Schrödinger’s FEP+ (Schrödinger Release 2025–2: FEP + , Schrödinger, LLC, New York, NY, 2025) using absolute binding free energy perturbation (AB-FEP) [53]. Due to finite simulation times, AB-FEP only accounts for a limited amount of the protein-reorganization free energy to transform the protein from the apo to holo state [53,54]. To accurately predict absolute binding affinity, an offset was added to each AB-FEP prediction to correct for this missing free energy cost. We pair this offset with each of the starting PDB structures and apply it as a post-hoc correction to the AB-FEP predicted ∆Gs. The final prediction is the most negative (favorable) AB-FEP ∆G after addition of the respective offset.
Kinome selectivity profiling
Off-target kinase inhibition was assessed using the DiscoverX scanMAX (Eurofins) panel at 1 µM compound concentration. Binding affinities of CIT, ROCK1 and ROCK2 were evaluated using the KdElect assay (Eurofins).
Synthetic strategy
The synthesis of late-stage intermediates 9 (Scheme 1) and 12 (Scheme 2) begins with a Sonogashira coupling to afford alkyne 4. Appel reaction with carbon tetrabromide to give 5 is followed by insertion of dimethyl sulfide, affording dimethylsulfonium salt 6. Condensation of aldehyde 1 with p-toluene sulfonamide affords Schiff base 2, that is then reacted with 6 to give cis-propargylic aziridine 7 under basic conditions. A platinum chloride catalyzed cycloisomerization in the presence of iodine affords the highly functionalized pyrrole 8. Subsequently, Suzuki coupling with 4-pyridylboronic acid gives selective coupling at the iodide of 8 to give late-stage intermediate 9. TBAF mediated tosyl deprotection of the pyrrole affords 10 that is then subjected to acidic deprotection conditions to give secondary amine 11 followed by reductive amination to arrive at late-stage intermediate 12.
Reagents and conditions: a) CuI, Pd(PPh3)2Cl2, TEA, 20°C, 4 h; b) CBr4, PPh3, DCM, 20°C, 1 h; c) Me2S, acetone, 20°C, 12 h; d) TsNH2, PhSO2Na, HCOOH/H2O, 20°C, 12 h; e) Cs2CO3, DCM, 20°C, 1.5 h; f) PtCl2, I2, MeCN/H2O, 80°C, 30 min; g) Boc2O, TEA, DCM, 20°C, 1 h; h) Pd(dppf)Cl2·CH2Cl2, 1,4-dioxane, 2M Na2CO3 (aq), μW, 90°C, 20 min.
a) 1M TBAF/THF, THF, rt, 16 h; b) 6M HCl/IPA, DCM, MeOH; c) formalin, AcOH, MeOH then NaCNBH3, 1 h.
Stille coupling of late-stage intermediate 12 and subsequent hydrolysis of the enone under acidic conditions gives 13. Borohydride reduction of 13 gives racemic secondary alcohol RUPB-68 (Scheme 3). Methyl addition to the ketone of 13 with methyl Grignard reagent affords tertiary alcohol RUPB-69.
a) tributyl(1-ethoxyvinyl)stannane, PdCl2(PPh3)2, 1,4-dioxane, μW, 130°C, 30 min; b) 1M HCl(aq), THF, water, rt, 1 h; c) NaBH4, MeOH, rt, 5 min; d) 3M MeMgBr in Et2O, Et2O, rt, 2 h, then NH4Cl(aq), rt, 30 min.
Palladium catalyzed carbonylation into the bromide of intermediate 9 affords ester 14. Using the previously described deprotection and methylation conditions, 14 is converted into RUPB-72. Diethyl insertion into the ester with ethylmagnesium bromide in the presence of titanium (IV) isopropoxide gives analog RUPB-73 (Scheme 4).
a) NaOAc, Mo(CO)₆, Pd(dppf)Cl2·CH2Cl2, MeOH, μW, 140°C, 1 h; b) 1M TBAF/THF, THF, rt, 1 h; c) TFA, DCM, 30 min; d) formalin, AcOH, MeOH, then NaCNBH3, 1 h; e) 3M EtMgBr in Et2O, Ti(OiPr)4, THF, rt, 30 min.
Experimental procedures for chemical synthesis and characterization
3-(3-Bromo-4-fluorophenyl)prop-2-yn-1-ol (4).
To a mixture of 2-bromo-1-fluoro-4-iodobenzene (3) (40.0 g, 0.13 mol, 1.0 equiv.) and trimethyl(prop-2-yn-1-yloxy)silane (23.9 g, 0.186 mol, 1.43 equiv.) in THF (500 mL) was added TEA (20.1 g, 0.199 mol, 1.53 equiv.), copper(I) iodide (2.53 g, 0.013 mol, 0.10 equiv.) and Pd(PPh3)2Cl2 (4.66 g, 6.64 mmol, 0.05 equiv.). Then the reaction mixture was stirred under N2 at 20°C for 4 hours. TLC (Petroleum ether: Ethyl acetate = 10:1) showed that the starting material was consumed, and a new spot was detected. The mixture was diluted with saturated aq. NH4Cl (500 mL) and extracted with ethyl acetate (500 mL × 3). The organic phase was concentrated under reduced pressure. The crude product was purified by silica gel chromatography (SiO2, Ethyl acetate/ Petroleum ether = 0–20%) to give 3-(3-bromo-4-fluorophenyl)prop-2-yn-1-ol (4) (23.0 g, 71.7% yield) as a colorless oil. 1H NMR (400 MHz, DMSO-d6) δ 7.78 (dd, J = 6.8, 2.0 Hz, 1H), 7.52 – 7.46 (m, 1H), 7.40 (t, J = 8.8 Hz, 1H), 5.37 (t, J = 6.0 Hz, 1H), 4.30 (d, J = 6.0 Hz, 2H).
2-Bromo-4-(3-bromoprop-1-yn-1-yl)-1-fluorobenzene (5).
To a solution of 3-(3-bromo-4-fluorophenyl)prop-2-yn-1-ol (4) (20.0 g, 87.3 mmol, 1.0 equiv.) in DCM (200 mL) was added CBr4 (34.7 g, 105 mmol, 1.2 equiv.). Then a solution of triphenylphosphine (27.5 g, 105 mmol, 1.2 equiv.) in DCM (200 mL) was added to the reaction mixture and stirred at 20°C for 1 hour. TLC (Petroleum ether: Ethyl acetate = 10:1) showed that the starting material was consumed, and a new spot was detected. The mixture was concentrated under reduced pressure. The crude product was diluted with n-hexanes (100 mL) and filtered. The filtrate was concentrated under reduced pressure to give 2-bromo-4-(3-bromoprop-1-yn-1-yl)-1-fluorobenzene (5) (24.0 g, 89.4% yield) as a colorless oil. 1H NMR (400 MHz, DMSO-d6) δ 7.84 (dd, J = 6.8, 2.0 Hz, 1H), 7.54 (ddd, J = 8.4, 4.8, 2.0 Hz, 1H), 7.42 (t, J = 8.8 Hz, 1H), 4.51 (s, 2H).
(3-(3-Bromo-4-fluorophenyl)prop-2-yn-1-yl)dimethylsulfonium bromide (6).
To a solution of 2-bromo-4-(3-bromoprop-1-yn-1-yl)-1-fluorobenzene (5) (24.0 g, 0.12 mol, 1 equiv.) in acetone (240 mL) was added Me2S (23.0 g, 0.37 mol, 3.09 equiv.). The reaction mixture was stirred at 20°C for 12 hours. The mixture was filtered, and the solid was collected and dried under vacuum to give (3-(3-bromo-4-fluorophenyl)prop-2-yn-1-yl)dimethylsulfonium bromide (6) (30.0 g, 90.0% yield) as a white solid. 1H NMR (400 MHz, DMSO-d6) δ 8.08 (dd, J = 6.8, 2.0 Hz, 1H), 7.68 (ddd, J = 8.4, 4.8, 2.0 Hz, 1H), 7.47 (t, J = 8.8 Hz, 1H), 4.70 (d, J = 13.6 Hz, 2H), 3.00 – 2.89 (m, 6H).
tert-Butyl 4-((tosylimino)methyl)piperidine-1-carboxylate (2).
To a mixture of tert-butyl 4-formylpiperidine-1-carboxylate (1) (55.0 g, 0.257 mol, 1 equiv.) and sodium p-toluenesulfinate (50.3 g, 0.28 mol, 1.09 equiv.) in FA/H2O (800 mL, 1: 1) was added 4-methylbenzenesulfonamide (44.0 g, 0.257 mol, 1 equiv.) The reaction mixture was stirred at 20°C for 12 hours. TLC (SiO2, Petroleum ether: Ethyl acetate = 1:1) showed that the starting material was consumed, and a new spot was detected. The mixture was filtered, and the filter cake was washed with water (500 mL × 2) and hexanes (500 mL × 2). The solid was dried under vacuum to give tert-butyl 4-((tosylimino)methyl)piperidine-1-carboxylate (2) (60.0 g, 0.154 mol, 60.2% yield, 100% purity by LCMS) as a white solid. ((MH-100)+) 267.10.
tert-Butyl 4-((2R,3S)-3-((3-bromo-4-fluorophenyl)ethynyl)-1-tosylaziridin-2-yl)piperidine-1-carboxylate (7).
To a solution of tert-butyl 4-((tosylimino)methyl)piperidine-1-carboxylate (2) (15.0 g, 40.7 mmol, 1 equiv.) in DCM (300 mL) was added (3-(3-bromo-4-fluorophenyl)prop-2-yn-1-yl)dimethylsulfonium bromide (6) (14.4 g, 40.7 mmol, 1 equiv.) and Cs2CO3 (13.3 g, 40.7 mmol, 1 equiv.). Then the reaction mixture was stirred at 20°C for 1.5 hours. LCMS showed that the starting material was consumed, and the desired product was detected as a major peak. The mixture was diluted with saturated aq. NH4Cl (200 mL) and extracted with DCM (200 mL × 2). The organic phase was concentrated under reduced pressure. The crude product was purified by silica gel chromatography (SiO2, Ethyl acetate/ Petroleum ether = 0–20%) to give tert-butyl 4-((2R,3S)-3-((3-bromo-4-fluorophenyl)ethynyl)-1-tosylaziridin-2-yl)piperidine-1-carboxylate (7) (20.0 g, 80.5% yield, 100% purity by LCMS) as a yellow solid. (MNa+) 599.05.
tert-Butyl 4-(5-(3-bromo-4-fluorophenyl)-4-iodo-1-tosyl-1H-pyrrol-2-yl)piperidine-1-carboxylate (8).
To a solution of tert-butyl 4-((2R,3S)-3-((3-bromo-4-fluorophenyl)ethynyl)-1-tosylaziridin-2-yl)piperidine-1-carboxylate (7) (20.0 g, 34.5 mmol, 1.0 equiv.) in MeCN/H2O (880 mL, v/v = 10: 1) was added I2 (17.5 g, 69.0 mmol, 1.5 equiv.) and PtCl2 (1.38 g, 5.17 mmol, 0.15 equiv.). The reaction mixture was heated to 80°C for 30 minutes. The mixture was diluted with 10% aq. Na2SO3 (800 mL) and extracted with ethyl acetate (800 mL × 2). The organic phase was concentrated under reduced pressure. The crude product was diluted with DCM (200 mL) followed by addition of Boc2O (8.28 g, 37.9 mmol) and TEA (6.97 g, 69.0 mmol) to the solution. Then the mixture was stirred at 20°C for 1 hour. LCMS showed that 30% starting material remaining with 40% desired product. The mixture was concentrated under reduced pressure. The crude product was purified by silica gel chromatography (SiO2, Ethyl acetate/ Petroleum ether = 0–15%) to give tert-butyl 4-(5-(3-bromo-4-fluorophenyl)-4-iodo-1-tosyl-1H-pyrrol-2-yl)piperidine-1-carboxylate (8) (4.00 g, 15.6% yield, 98.6% purity by HPLC) as a white solid. (MNa)+ 725.05. 1H NMR (400 MHz, DMSO-d6) δ 7.43 – 7.32 (m, 5H), 7.26 (dd, J = 6.4, 1.6 Hz, 1H), 7.22 (ddd, J = 8.4, 4.8, 2.0 Hz, 1H), 6.46 (s, 1H), 4.13 – 3.94 (m, 2H), 3.33 – 3.26 (m, 1H), 2.93 – 2.69 (m, 2H), 2.39 (s, 3H), 1.87 – 1.75 (m, 2H), 1.46 – 1.37 (m, 11H). 19F NMR (400 MHz, DMSO-d6) δ -108.03 (s, 1F).
tert-Butyl 4-(5-(3-bromo-4-fluorophenyl)-4-(pyridin-4-yl)-1-tosyl-1H-pyrrol-2-yl)piperidine-1-carboxylate (9).
A mixture of tert-butyl 4-(5-(3-bromo-4-fluorophenyl)-4-iodo-1-tosyl-1H-pyrrol-2-yl)piperidine-1-carboxylate (8) (700 mg, 1.0 mmol, 1 equiv.), pyridin-4-ylboronic acid (148 mg, 1.2 mmol, 1.2 equiv.) and Pd(dppf)Cl2·CH2Cl2 (82 mg, 0.10 mmol, 0.10 equiv.) was taken up in 1,4-dioxane (10 mL) and 2M Na2CO3 (aq) (5 mL) was heated in a microwave reactor for 20 minutes at 90°C. The reaction was determined to be complete by LCMS with a 73:9 ratio of mono-addition:double-addition products by LCMS at 254 nm. The crude reaction mixture was partitioned in EtOAc/water. The aqueous was extracted 3 x EtOAc. The combined organic was washed 1 x brine, dried over sodium sulfate, filtered and concentrated. The concentrate was purified by silica gel chromatography (0% → 5% MeOH/DCM) and product fractions repurified by silica gel chromatography (0% → 50% → 100% EtOAc/Hex). Product fractions were concentrated under reduced pressure and dried under vacuum to afford tert-butyl 4-(5-(3-bromo-4-fluorophenyl)-4-(pyridin-4-yl)-1-tosyl-1H-pyrrol-2-yl)piperidine-1-carboxylate (9) as a white solid (498 mg, 95% purity by LCMS). (MH+) 654.30, 656.30.
tert-Butyl 4-(5-(3-bromo-4-fluorophenyl)-4-(pyridin-4-yl)-1H-pyrrol-2-yl)piperidine-1-carboxylate (10).
To a solution of tert-butyl 4-(5-(3-bromo-4-fluorophenyl)-4-(pyridin-4-yl)-1-tosyl-1H-pyrrol-2-yl)piperidine-1-carboxylate (9) (389 mg, 0.59 mmol, 1 equiv.) in THF (20 mL) was added TBAF (1M in THF) (2.95 mL, 2.95 mmol, 5 equiv.). After stirring for 16 hours at room temperature, the reaction was determined complete by LCMS and was concentrated under reduced pressure. The residue was partitioned in EtOAc/water. The aqueous was separated and the organic was washed 1 x brine, dried over sodium sulfate, filtered and concentrated. The residue was purified by silica gel chromatography (0% → 2% MeOH/DCM). The combined product containing fractions were concentrated under reduced pressure. The solid was taken up in EtOAc, sonicated and filtered. The solid was washed with minimal EtOAc and dried under vacuum to yield tert-butyl 4-(5-(3-bromo-4-fluorophenyl)-4-(pyridin-4-yl)-1H-pyrrol-2-yl)piperidine-1-carboxylate (10) (258.8 mg, 0.52 mmol, 98% purity by LCMS) as a tan solid. (MH+) 500.20, 502.20.
4-(2-(3-Bromo-4-fluorophenyl)-5-(piperidin-4-yl)-1H-pyrrol-3-yl)pyridine (HCl salt) (11).
To a solution of tert-butyl 4-[5-(3-bromo-4-fluorophenyl)-4-(pyridin-4-yl)-1H-pyrrol-2-yl]piperidine-1-carboxylate (10) (258.8 mg, 0.52 mmol, 1 equiv.) in DCM (5 mL) and MeOH (5 mL) was added 6M HCl in IPA (1 mL) and stirred for 16 hours at rt. The reaction was determined to be complete by LCMS and concentrated under reduced pressure to give 4-(2-(3-bromo-4-fluorophenyl)-5-(piperidin-4-yl)-1H-pyrrol-3-yl)pyridine (HCl salt) (11) in quantitative yield (97% purity by LCMS at 254 nm). (MH+) 400.15, 402.20.
4-(2-(3-Bromo-4-fluorophenyl)-5-(1-methylpiperidin-4-yl)-1H-pyrrol-3-yl)pyridine (12).
To a solution of 4-[2-(3-bromo-4-fluorophenyl)-5-(piperidin-4-yl)-1H-pyrrol-3-yl]pyridine (11) (158 mg, 0.33 mmol, 1.0 equiv.) in MeOH (10 ml) was added AcOH (400 µL) and formalin (800 µL). After stirring for 1 hour at room temperature, an excess of sodium cyanoborohydride was added. Starting material and product coelute by LCMS. The reaction was determined to be complete when scan by scan analysis of the product peak showed only mass for methylated product with no visible signal for starting material mass present. After stirring for 30 minutes post addition of sodium cyanoborohydride, the reaction was concentrated under reduced pressure to remove the majority of the methanol. The residue was partitioned in EtOAc/water. The product was extracted into the acidic aqueous with 2 x water extractions of the EtOAc layer. The combined acidic aqueous was neutralized with NaHCO3 (sat, aq) and extracted 3 x DCM. The combined organic was dried over sodium sulfate, filtered and concentrated to afford 4-(2-(3-bromo-4-fluorophenyl)-5-(1-methylpiperidin-4-yl)-1H-pyrrol-3-yl)pyridine (12) as a yellow solid (134 mg, 99% purity by LCMS at 254 nm). (MH+) 414.10, 415.85.
1-(2-Fluoro-5-(5-(1-methylpiperidin-4-yl)-3-(pyridin-4-yl)-1H-pyrrol-2-yl)phenyl)ethan-1-one (13).
A mixture of 4-(2-(3-bromo-4-fluorophenyl)-5-(1-methylpiperidin-4-yl)-1H-pyrrol-3-yl)pyridine (12) (34.2 mg, 0.083 mmol, 1 equiv.), tributyl(1-ethoxyvinyl)stannane (74.5 mg, 0.21 mmol, 2.5 equiv.) and PdCl2(PPh3)2 (5.8 mg, 0.0083 mmol, 0.10 equiv.) was taken up in 1,4-dioxane (4 mL) and heated in a microwave reactor for 30 minutes at 130°C. The reaction was determined complete, as a mixture of the title compound and the enol intermediate. The mixture was diluted in EtOAc and filtered through celite to remove insoluble impurities. The filtrate was concentrated and taken up in a mixture of THF (4 mL) and water (4 mL). To this mixture was added 1M HCl(aq) (0.5 mL). After stirring for 1 hour at room temperature the reaction was determined complete by LCMS, with full conversion to the desired ketone product. The reaction was partitioned in DCM/water. The product partitioned into the acidic aqueous and the aqueous was washed 2 x DCM. The aqueous was basified with 1M NaOH(aq) and extracted 4 x DCM. The combined organic was dried over sodium sulfate, filtered and concentrated. The product was 98.5% purity after extraction but was purified by amino-d silica gel chromatography (0% → 2% MeOH/DCM) to remove any remaining impurities. The product fractions were combined and concentrated to afford 1-(2-fluoro-5-(5-(1-methylpiperidin-4-yl)-3-(pyridin-4-yl)-1H-pyrrol-2-yl)phenyl)ethan-1-one (13) (28.7 mg, 98% purity by LCMS at 254 nm) as a tan solid. (MH+) 378.30.
2-(2-Fluoro-5-(5-(1-methylpiperidin-4-yl)-3-(pyridin-4-yl)-1H-pyrrol-2-yl)phenyl)propan-2-ol (TFA salt) (RUPB-69).
To 1-(2-fluoro-5-(5-(1-methylpiperidin-4-yl)-3-(pyridin-4-yl)-1H-pyrrol-2-yl)phenyl)ethan-1-one (13) (15 mg, 0.040 mmol, 1 equiv.) in an oven-dried flask was added Et2O (3.5 mL). To this suspension was added MeMgBr (3M in Et2O) (60 µL, 0.20 mmol, 5 equiv.). After stirring for 1 hour at room temperature the reaction was roughly 45% complete by LCMS. An additional batch of MeMgBr (3M in Et2O) (60 µL, 0.20 mmol, 5 equiv.) was added. The reaction was roughly 67% complete by LCMS but was beginning to generate additional impurities. The reaction was quenched with NH4Cl (sat, aq) and was stirred for 30 min at rt. The mixture was partitioned in EtOAc/water and the aqueous was adjusted to pH = 2 with 1M HCl. The organic was extracted 2 x water. The combined aqueous was basified with 1M HCl and extracted 4 x DCM. The combined organic from the basic extraction was dried over sodium sulfate, filtered and concentrated. The concentrate was purified by prep HPLC (0.1% TFA(aq)/0.1% TFA(MeCN). The combined product fractions were concentrated and lyophilized to give 2-(2-fluoro-5-(5-(1-methylpiperidin-4-yl)-3-(pyridin-4-yl)-1H-pyrrol-2-yl)phenyl)propan-2-ol (TFA salt) (RUPB-69) (10.6 mg, 99% purity by LCMS) as a yellow/orange solid. (MH+) 394.30. 1H NMR (500 MHz, DMSO-d6) δ 11.89 – 11.72 (m, 1H), 10.37 – 10.17 (m, 1H), 8.63 – 8.51 (m, 2H), 7.73 – 7.65 (m, 3H), 7.34 (ddd, J = 8.3, 4.6, 2.4 Hz, 1H), 7.29 – 7.21 (m, 1H), 6.75 – 6.51 (m, 1H), 3.51 (d, J = 12.0 Hz, 2H), 3.17 – 3.03 (m, 2H), 2.91 – 2.73 (m, 4H), 2.28 – 2.17 (m, 2H), 1.87 (qd, J = 13.3, 3.7 Hz, 2H), 1.54 – 1.47 (m, 6H).
1-(2-Fluoro-5-(5-(1-methylpiperidin-4-yl)-3-(pyridin-4-yl)-1H-pyrrol-2-yl)phenyl)ethan-1-ol (TFA salt) (RUPB-68).
To a solution of 1-(2-fluoro-5-(5-(1-methylpiperidin-4-yl)-3-(pyridin-4-yl)-1H-pyrrol-2-yl)phenyl)ethan-1-one (13) (8 mg, 0.021 mmol, 1 equiv.) in MeOH (2.5 mL) was added NaBH4 (8 mg, 0.21 mmol, 10 equiv.). After stirring for 5 minutes at room temperature, the reaction was determined complete by LCMS. The reaction was concentrated under reduced pressure and the residue was purified by prep HPLC (0.1% TFA(aq)/0.1% TFA(MeCN). The combined product fractions were concentrated under reduced pressure and lyophilized to give 1-(2-fluoro-5-(5-(1-methylpiperidin-4-yl)-3-(pyridin-4-yl)-1H-pyrrol-2-yl)phenyl)ethan-1-ol (TFA salt) (RUPB-68) (8.2 mg, 99% purity by LCMS) as a yellow/orange solid. (MH+) 380.40. 1H NMR (500 MHz, DMSO-d6) δ 11.84 – 11.70 (m, 1H), 10.51 – 10.22 (m, 1H), 8.67 – 8.47 (m, 2H), 7.67 (d, J = 6.1 Hz, 2H), 7.56 (dd, J = 7.1, 2.4 Hz, 1H), 7.35 (ddd, J = 7.7, 4.9, 2.4 Hz, 1H), 7.26 (dt, J = 10.1, 7.0 Hz, 1H), 6.71 – 6.49 (m, 1H), 5.01 (q, J = 6.4 Hz, 1H), 3.50 (d, J = 12.2 Hz, 2H), 3.08 (q, J = 11.1 Hz, 2H), 2.90 – 2.74 (m, 4H), 2.24 (d, J = 13.2 Hz, 2H), 1.90 (tt, J = 14.6, 7.8 Hz, 2H), 1.34 (d, J = 6.3 Hz, 3H).
tert-Butyl 4-(5-(4-fluoro-3-(methoxycarbonyl)phenyl)-4-(pyridin-4-yl)-1-tosyl-1H-pyrrol-2-yl)piperidine-1-carboxylate (14).
A mixture of tert-butyl 4-(5-(3-bromo-4-fluorophenyl)-4-(pyridin-4-yl)-1-tosyl-1H-pyrrol-2-yl)piperidine-1-carboxylate (9) (156 mg, 0.24 mmol, 1 equiv.), sodium acetate (196 mg, 2.4 mmol, 10 equiv.), molybdenum hexacarbonyl (253 mg, 0.96 mmol, 4 equiv.) and Pd(dppf)Cl2·CH2Cl2 (39 mg, 0.048 mmol, 0.20 equiv.) was taken up in MeOH (6 mL) and heated in a microwave reactor for 1 hour at 130°C. The reaction was incomplete by LCMS and heated in the microwave reactor an additional 40 minutes at 140°C. The reaction was partitioned in DCM/water and filtered over celite to remove insoluble solids. The product was rinsed through the filter with DCM/water. The layers were separated and the aqueous was extracted 2 x DCM. The combined organic was dried over sodium sulfate, filtered and concentrated to afford a mixture of tert-butyl 4-(5-(4-fluoro-3-(methoxycarbonyl)phenyl)-4-(pyridin-4-yl)-1H-pyrrol-2-yl)piperidine-1-carboxylate (15) (24% area of LCMS at 254 nm, (MH+) 480.30) and tert-butyl 4-(5-(4-fluoro-3-(methoxycarbonyl)phenyl)-4-(pyridin-4-yl)-1-tosyl-1H-pyrrol-2-yl)piperidine-1-carboxylate (14) (42% area of LCMS at 254 nm, (MH+) 634.35) that was carried forward to the next reaction without further purification.
tert-Butyl 4-(5-(4-fluoro-3-(methoxycarbonyl)phenyl)-4-(pyridin-4-yl)-1H-pyrrol-2-yl)piperidine-1-carboxylate (15).
To the crude mixture of tert-butyl 4-(5-(4-fluoro-3-(methoxycarbonyl)phenyl)-4-(pyridin-4-yl)-1-tosyl-1H-pyrrol-2-yl)piperidine-1-carboxylate (14) and tert-butyl 4-(5-(4-fluoro-3-(methoxycarbonyl)phenyl)-4-(pyridin-4-yl)-1H-pyrrol-2-yl)piperidine-1-carboxylate (15) from the previous reaction in THF (10 mL) was added TBAF (1M in THF, 1 mL). The reaction was determined complete by LCMS after stirring for 1 hour at room temperature and was concentrated under reduced pressure to reduce the volume of THF. The concentrate was partitioned in EtOAc/water and the aqueous was extracted 3 x EtOAc. The combined organic was dried over sodium sulfate, filtered and concentrated. The concentrate was purified by silica gel chromatography (0% → 3% MeOH/DCM) and product fractions were combined, concentrated and dried under vacuum to afford tert-butyl 4-(5-(4-fluoro-3-(methoxycarbonyl)phenyl)-4-(pyridin-4-yl)-1H-pyrrol-2-yl)piperidine-1-carboxylate (15) as an orange solid (68.3 mg, 91% purity by LCMS at 254 nm). (MH+) 480.30.
Methyl 2-fluoro-5-(5-(piperidin-4-yl)-3-(pyridin-4-yl)-1H-pyrrol-2-yl)benzoate (TFA salt) (16).
To a solution of tert-butyl 4-(5-(4-fluoro-3-(methoxycarbonyl)phenyl)-4-(pyridin-4-yl)-1H-pyrrol-2-yl)piperidine-1-carboxylate (15) (68.3 mg, 0.14 mmol) in DCM (2 mL) was added TFA (0.5 mL). The reaction was determined complete by LCMS after stirring 30 minutes at room temperature. The crude reaction was concentrated to afford methyl 2-fluoro-5-(5-(piperidin-4-yl)-3-(pyridin-4-yl)-1H-pyrrol-2-yl)benzoate (TFA salt) (16) (quantitative yield, yellow/orange solid) that was used in the next reaction without further purification. (MH+) 380.15.
Methyl 2-fluoro-5-(5-(1-methylpiperidin-4-yl)-3-(pyridin-4-yl)-1H-pyrrol-2-yl)benzoate (RUPB-72).
To a solution of methyl 2-fluoro-5-(5-(piperidin-4-yl)-3-(pyridin-4-yl)-1H-pyrrol-2-yl)benzoate 2 x TFA (16) (85 mg, 0.14 mmol) in MeOH (4 mL) was added AcOH (200 µL) followed by formalin (400 µL). After stirring for 1 hour at room temperature, the reaction was quenched with the addition of excess sodium cyanoborohydride. The reaction was stirred for 30 minutes at room temperature and determined complete by LCMS. The crude reaction was concentrated under reduced pressure and the concentrate was partitioned in DCM and NaHCO3 (sat, aq). The aqueous was extracted 3 x DCM and the combined organic was dried over sodium sulfate, filtered, and concentrated. The product was purified by prep HPLC (0.1% FA(aq), MeCN(unmodified)). Product fractions were combined and concentrated to afford the title compound (64.6 mg yellow solid, formate salt, 80% purity by LCMS, (MH+) 394.15). The product was converted to the free base for solubility reasons related to the next reaction. The product was partitioned in DCM and NaHCO3 (sat, aq) and extracted 3 x DCM. The combined organic was dried over sodium sulfate, filtered and concentrated to give methyl 2-fluoro-5-(5-(1-methylpiperidin-4-yl)-3-(pyridin-4-yl)-1H-pyrrol-2-yl)benzoate (RUPB-72) as the free base (28.3 mg white solid, 80% purity by LCMS at 254 nm). (MH+) 394.15.
3-(2-Fluoro-5-(5-(1-methylpiperidin-4-yl)-3-(pyridin-4-yl)-1H-pyrrol-2-yl)phenyl)pentan-3-ol (TFA salt) (RUPB-73).
To a solution of methyl 2-fluoro-5-(5-(1-methylpiperidin-4-yl)-3-(pyridin-4-yl)-1H-pyrrol-2-yl)benzoate (RUPB-72) (20 mg, 0.051 mmol, 1 equiv.) in THF (4 mL) was added titanium(IV) isopropoxide (73 µL, 0.25 mmol, 5 equiv.) followed by ethylmagnesium bromide (3M in Et2O, 170 µL, 0.51 mmol, 10 equiv.). The reaction was stirred for 30 minutes at room temperature and quenched with water. The crude mixture was partitioned in EtOAc/water and filtered to remove solids. The product was rinsed through the filter with additional EtOAc/water. The organic was separated and the aqueous was extracted 2 x EtOAc. The combined organic was dried over sodium sulfate, filtered and concentrated. The concentrate was purified by prep HPLC (0.1% FA(aq)/MeCN(unmodified)). The crude reaction contained a mixture of unreacted starting material and the product of diethyl addition into the ester. Fractions containing pure products were combined separately and concentrated. Each of the products were then separately repurified by prep HPLC (0.1% TFA(aq)/0.1%TFA(MeCN). Clean fractions for each product were separately combined and concentrated to give the title compounds as TFA salts after concentration. Methyl 2-fluoro-5-(5-(1-methylpiperidin-4-yl)-3-(pyridin-4-yl)-1H-pyrrol-2-yl)benzoate (TFA salt) (RUPB-72) (5.8 mg sticky yellow/orange solid, 96.4% purity by HPLC at 254 nm. (MH+) 394.25. 3-(2-Fluoro-5-(5-(1-methylpiperidin-4-yl)-3-(pyridin-4-yl)-1H-pyrrol-2-yl)phenyl)pentan-3-ol (TFA salt) (RUPB-73) (9.6 mg sticky yellow/orange solid, 100% purity by HPLC at 254 nm). (MH+) 422.25. 1H NMR (500 MHz, DMSO-d6) δ 11.87 – 11.69 (m, 1H), 10.41 – 10.26 (m, 1H), 8.54 – 8.47 (m, 2H), 7.61 (d, J = 6.4 Hz, 2H), 7.54 (dd, J = 7.7, 2.5 Hz, 1H), 7.38 (ddd, J = 8.2, 4.6, 2.4 Hz, 1H), 7.27 (dt, J = 11.7, 7.6 Hz, 1H), 6.73 – 6.49 (m, 1H), 4.69 (s, 1H), 3.50 (d, J = 12.1 Hz, 2H), 3.08 (q, J = 11.6 Hz, 2H), 2.89 – 2.75 (m, 4H), 2.24 (d, J = 13.4 Hz, 2H), 1.92 (dddd, J = 22.5, 15.9, 10.9, 5.5 Hz, 4H), 1.70 (dq, J = 14.5, 7.3 Hz, 2H), 0.66 (t, J = 7.3 Hz, 6H).
RUPB-58, -59, -60 and -61 were synthesized as described previously [25].
RUPB-58 4-{5-[(3R,4R)-1,3-Dimethylpiperidin-4-yl]-2-(4-fluorophenyl)-1H-pyrrol-3-yl}pyridine Dihydrochloride
RUPB-59 4-{5-[(3R,4S)-1,3-Dimethylpiperidin-4-yl]-2-(4-fluorophenyl)-1H-pyrrol-3-yl}pyridine Dihydrochloride
RUPB-60 4-{5-[(3S,4R)-1,3-Dimethylpiperidin-4-yl]-2-(4-fluorophenyl)-1H-pyrrol-3-yl}pyridine Dihydrochloride
RUPB-61 4-{5-[(3S,4S)-1,3-Dimethylpiperidin-4-yl]-2-(4-fluorophenyl)-1H-pyrrol-3-yl}pyridine Dihydrochloride
Supporting information
S1 Fig. Activity and cytotoxicity of PfPKG inhibitors in primary hepatocyte models of P. falciparum and P. cynomolgi sporozoite infection.
(A) Dose-response curve of TSP against P. cynomolgi hepatic hypnozoites formed in simian hepatocytes. Tafenoquine was used as a positive control. Data shown are normalized to number of infected cells in vehicle control. Data shown are from a single trial performed in technical duplicates. (B) Dose-response curves from two independent trials of RUPB-61 against P. falciparum sporozoites in human hepatocytes. Data shown are normalized to the number of infected cells in vehicle control. KDU-691 was used as a positive control. (C) Effect of TSP on the survival of primary simian hepatocytes. Data shown are from a single biological replicate performed in technical duplicates. (D) Effect of RUPB-61 on the survival of primary human hepatocytes. Data shown are mean of two independent trials performed in technical duplicates. (E) Dose-response curves from three independent trials of RUPB-61 against P. cynomolgi sporozoites in primary simian hepatocytes. Data shown are normalized to the number of infected cells in vehicle control. KDU-691 was used as a positive control. (F) Effect of RUPB-61 on the survival of primary simian hepatocytes. Data shown are mean of three independent trials performed in technical duplicates or triplicates.
https://doi.org/10.1371/journal.ppat.1014322.s001
(PDF)
S2 Fig. Dose-response curves of RUPB-58 to -61 against multidrug resistant P. falciparum parasites.
Growth inhibition of 3D7, Dd2-B2, TM90_C2B, K1, 7G8 over 72h in asexual growth assays was measured using SybrGreen incorporation.
https://doi.org/10.1371/journal.ppat.1014322.s002
(PDF)
S3 Fig. Liver parasitemia after administration of PfPKG inhibitors.
Bioluminescent imaging of mice to quantify liver parasitemia in mice infected with PbLuc-GFP sporozoites. (A) Mice administered three intravenous doses of vehicle or compound. Images were obtained at 44 hpi and 144 hpi. (B) Mice administered three oral doses of vehicle or compound. Images were obtained at 44 hpi and 144 hpi.
https://doi.org/10.1371/journal.ppat.1014322.s003
(PDF)
S4 Fig. TREEspot Interaction map for RUPB-61.
Interaction of RUPB-61 at 1 µM with 400 unique human kinases was determined using a panel of in vitro competition binding assays. The effect of RUPB-61 on the binding of a target kinase to its test probe is reported as a percentage of enzyme binding in the presence of vehicle alone. Blue circles represent the major human off-targets, CIT, ROCK1 and ROCK2. TKL: Tyrosine Kinase-like; STE: Ste kinases; CK1: Casein Kinase 1; AGC: Protein Kinase A, G and C; CAMK: calmodulin kinase; CMGC: Cyclin-dependent kinases, Mitogen-activated protein kinases, Glycogen synthase kinases and CDC-like kinases.
https://doi.org/10.1371/journal.ppat.1014322.s004
(PDF)
S5 Fig. Experimental structure of RUPB-60 bound to PvPKG.
Compared to the structure with RUPB-61, the methyl substituent on the RUPB-60 (green) piperidine evidently leads to the resolution of Ala816 (grey) in the C-term tail (dotted line) (Ala823 in PfPKG).
https://doi.org/10.1371/journal.ppat.1014322.s005
(PDF)
S6 Fig. FEP+ correlation plot comparing prediction to experiment.
A congeneric set of 35 ligands was used in the (A) IFD-MD model of PfPKG (B) experimental PvPKG structure.
https://doi.org/10.1371/journal.ppat.1014322.s006
(PDF)
S1 Table. Natural variations in PfPKG open reading frame.
Bioinformatics analyses of PfPKG genomic sequence in MalariaGen isolates.
https://doi.org/10.1371/journal.ppat.1014322.s007
(XLSX)
S2 Table. In vivo ADMET properties of RUPB-61.
PK properties in mice after oral and intravenous dosing of 10 mg/kg compound. t1/2: half-life; Tmax: maximum time; Cmax: maximum concentration; AUC: area under the concentration-time curve; AUClast: AUC last observed concentration; AUCINF_obs: AUC extrapolated to infinity, based on the last observed concentration; CI_obs: Observed total plasma clearance.
https://doi.org/10.1371/journal.ppat.1014322.s008
(XLSX)
S3 Table. Crystallography data collection and refinement.
X-ray structures of RUPB-60 and RUPB-61 bound to P. vivax PKG were solved and deposited in the PDB database.
https://doi.org/10.1371/journal.ppat.1014322.s009
(XLSX)
S4 Table. Individual predicted ∆G from the IFD-MD PfPKG model for a congeneric set of 35 ligands versus their experimental ∆G.
*Experimental ∆G was inferred from IC50 values using the relationship IC50 ~ Kd and ∆G = -RTLn(Kd).
https://doi.org/10.1371/journal.ppat.1014322.s010
(XLSX)
Acknowledgments
This research was funded by W81XWH2010386 from the Department of Defense to P.B. and J.Y.R. This project has been funded in whole or in part with Federal funds from the National Institute of Allergy and Infectious Diseases, National Institutes of Health, Department of Health and Human Services, under Contract No. 75N93022C00036 (S.L., A.D.). This research used resources at the NYX beamline 19-ID, supported by the New York Structural Biology Center, at the National Synchrotron Light Source II, a U.S. Department of Energy (DOE) Office of Science User Facility operated for the DOE Office of Science by Brookhaven National Laboratory under Contract No. DE-SC0012704. The NYX detector instrumentation was supported by grant S10OD030394 through the Office of the Director of the National Institutes of Health (K.P.B.). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. The following reagents were obtained through BEI Resources, NIAID, NIH: Plasmodium falciparum, Strain K1, MRA-159, contributed by Dennis E. Kyle and strain 7G8, MRA-152, contributed by David Walliker. Strain TM90_C2B was a kind gift by Dr. Alison Roth, WRAIR. The pFastBac construct expressing P. vivax PKG was a kind gift from Dr. Halabelian and Dr. Aled Edwards at University of Toronto. We thank Alma Seitova for helpful discussion.
The material has been reviewed by the Walter Reed Army Institute of Research (WRAIR). There is no objection to its presentation and/or publication. The opinions or assertions contained herein are the private views of the authors and are not to be constructed as official or as reflecting the true views of the Department of the Army or the Department of Defense. Research was conducted under an IACUC-approved animal use protocol in an AAALAC International-accredited facility with a Public Health Service Animal Welfare Assurance and in compliance with the Animal Act and other federal statutes and regulations relating to laboratory animals. This research was supported in part by an appointment to the Department of Defense (DOD) Research Participation Program administered by the Oak Ridge Institute for Science and Education (ORISE) through an interagency agreement between the U.S. Department of Energy (DOE) and the DOD. ORISE is managed by ORAU under DOE contract number DE-SC0014664. All opinions expressed in this paper are the author’s and do not necessarily reflect the policies and views of DOD, DOE, or ORAU/ORISE.
References
- 1. El Gaaloul M, Tchouatieu AM, Kayentao K, Campo B, Buffet B, Ramachandruni H, et al. Chemoprevention of malaria with long-acting oral and injectable drugs: an updated target product profile. Malar J. 2024;23(1):315. pmid:39425110
- 2. Noor AM, Alonso PL. The message on malaria is clear: progress has stalled. Lancet. 2022;399(10337):1777. pmid:35461617
- 3. Alonso PL, Sacarlal J, Aponte JJ, Leach A, Macete E, Aide P, et al. Duration of protection with RTS,S/AS02A malaria vaccine in prevention of Plasmodium falciparum disease in Mozambican children: single-blind extended follow-up of a randomised controlled trial. Lancet. 2005;366(9502):2012–8. pmid:16338450
- 4. Gaudinski MR, Berkowitz NM, Idris AH, Coates EE, Holman LA, Mendoza F, et al. A monoclonal antibody for malaria prevention. N Engl J Med. 2021;385(9):803–14. pmid:34379916
- 5. Burrows JN, Duparc S, Gutteridge WE, Hooft van Huijsduijnen R, Kaszubska W, Macintyre F, et al. New developments in anti-malarial target candidate and product profiles. Malar J. 2017;16(1):26. pmid:28086874
- 6. Neafsey DE, Juraska M, Bedford T, Benkeser D, Valim C, Griggs A, et al. Genetic diversity and protective efficacy of the RTS,S/AS01 malaria vaccine. N Engl J Med. 2015;373(21):2025–37. pmid:26488565
- 7. Walk J, Reuling IJ, Behet MC, Meerstein-Kessel L, Graumans W, van Gemert G-J, et al. Modest heterologous protection after Plasmodium falciparum sporozoite immunization: a double-blind randomized controlled clinical trial. BMC Med. 2017;15(1):168. pmid:28903777
- 8. Kepple D, Hubbard A, Ali MM, Abargero BR, Lopez K, Pestana K, et al. Plasmodium vivax from duffy-negative and duffy-positive individuals share similar gene pools in East Africa. J Infect Dis. 2021;224(8):1422–31. pmid:33534886
- 9. Twohig KA, Pfeffer DA, Baird JK, Price RN, Zimmerman PA, Hay SI, et al. Growing evidence of Plasmodium vivax across malaria-endemic Africa. PLoS Negl Trop Dis. 2019;13(1):e0007140. pmid:30703083
- 10. Walldorf JA, Cohee LM, Coalson JE, Bauleni A, Nkanaunena K, Kapito-Tembo A, et al. School-age children are a reservoir of malaria infection in malawi. PLoS One. 2015;10(7):e0134061. pmid:26207758
- 11. Chu CS, Bancone G, Moore KA, Win HH, Thitipanawan N, Po C, et al. Haemolysis in G6PD heterozygous females treated with primaquine for plasmodium vivax malaria: a nested cohort in a trial of radical curative regimens. PLoS Med. 2017;14(2):e1002224. pmid:28170391
- 12. Ley B, Winasti Satyagraha A, Rahmat H, von Fricken ME, Douglas NM, Pfeffer DA, et al. Performance of the Access Bio/CareStart rapid diagnostic test for the detection of glucose-6-phosphate dehydrogenase deficiency: A systematic review and meta-analysis. PLoS Med. 2019;16(12):e1002992.
- 13. Bennett JW, Pybus BS, Yadava A, Tosh D, Sousa JC, McCarthy WF, et al. Primaquine failure and cytochrome P-450 2D6 in Plasmodium vivax malaria. N Engl J Med. 2013;369(14):1381–2. pmid:24088113
- 14. Pybus BS, Marcsisin SR, Jin X, Deye G, Sousa JC, Li Q, et al. The metabolism of primaquine to its active metabolite is dependent on CYP 2D6. Malar J. 2013;12:212. pmid:23782898
- 15. Dow G, Smith B. The blood schizonticidal activity of tafenoquine makes an essential contribution to its prophylactic efficacy in nonimmune subjects at the intended dose (200 mg). Malar J. 2017;16(1):209.
- 16. Vaidya AB, Mather MW. Atovaquone resistance in malaria parasites. Drug Resist Updat. 2000;3(5):283–7. pmid:11498396
- 17. Hopp CS, Bowyer PW, Baker DA. The role of cGMP signalling in regulating life cycle progression of Plasmodium. Microbes Infect. 2012;14(10):831–7. pmid:22613210
- 18. Alam MM, Solyakov L, Bottrill AR, Flueck C, Siddiqui FA, Singh S, et al. Phosphoproteomics reveals malaria parasite protein kinase G as a signalling hub regulating egress and invasion. Nat Commun. 2015;6:7285. pmid:26149123
- 19. Brochet M, Collins MO, Smith TK, Thompson E, Sebastian S, Volkmann K, et al. Phosphoinositide metabolism links cGMP-dependent protein kinase G to essential Ca²⁺ signals at key decision points in the life cycle of malaria parasites. PLoS Biol. 2014;12(3):e1001806. pmid:24594931
- 20. Govindasamy K, Jebiwott S, Jaijyan DK, Davidow A, Ojo KK, Van Voorhis WC, et al. Invasion of hepatocytes by Plasmodium sporozoites requires cGMP-dependent protein kinase and calcium dependent protein kinase 4. Mol Microbiol. 2016;102(2):349–63. pmid:27425827
- 21. Falae A, Combe A, Amaladoss A, Carvalho T, Menard R, Bhanot P. Role of Plasmodium berghei cGMP-dependent protein kinase in late liver stage development. J Biol Chem. 2010;285(5):3282–8. pmid:19940133
- 22. McRobert L, Taylor CJ, Deng W, Fivelman QL, Cummings RM, Polley SD, et al. Gametogenesis in malaria parasites is mediated by the cGMP-dependent protein kinase. PLoS Biol. 2008;6(6):e139. pmid:18532880
- 23. Rotella D, Siekierka J, Bhanot P. Plasmodium falciparum cGMP-dependent protein kinase - a novel chemotherapeutic target. Front Microbiol. 2020;11:610408. pmid:33613463
- 24. Gurnett AM, Liberator PA, Dulski PM, Salowe SP, Donald RGK, Anderson JW, et al. Purification and molecular characterization of cGMP-dependent protein kinase from Apicomplexan parasites. A novel chemotherapeutic target. J Biol Chem. 2002;277(18):15913–22. pmid:11834729
- 25. Gilleran JA, Ashraf K, Delvillar M, Eck T, Fondekar R, Miller EB, et al. Structure-activity relationship of a pyrrole based series of PfPKG Inhibitors as Anti-Malarials. J Med Chem. 2024;67(5):3467–503. pmid:38372781
- 26. Zeeman A-M, Lakshminarayana SB, van der Werff N, Klooster EJ, Voorberg-van der Wel A, Kondreddi RR, et al. PI4 kinase is a prophylactic but not radical curative target in plasmodium vivax-type malaria parasites. Antimicrob Agents Chemother. 2016;60(5):2858–63. pmid:26926645
- 27. Duffey M, Blasco B, Burrows JN, Wells TNC, Fidock DA, Leroy D. Assessing risks of Plasmodium falciparum resistance to select next-generation antimalarials. Trends Parasitol. 2021;37(8):709–21. pmid:34001441
- 28. Baker DA, Stewart LB, Large JM, Bowyer PW, Ansell KH, Jiménez-Díaz MB, et al. A potent series targeting the malarial cGMP-dependent protein kinase clears infection and blocks transmission. Nat Commun. 2017;8(1):430. pmid:28874661
- 29. Vanaerschot M, Murithi JM, Pasaje CFA, Ghidelli-Disse S, Dwomoh L, Bird M, et al. Inhibition of resistance-refractory P. falciparum kinase PKG delivers prophylactic, blood stage, and transmission-blocking antiplasmodial activity. Cell Chem Biol. 2020;27(7):806-816.e8. pmid:32359426
- 30. Lu Z, Kaspera R, Naritomi Y, Wang T. Dose finding in single dose studies by allometric scaling. Drug Discovery and Evaluation: Methods in Clinical Pharmacology. 2020.
- 31. Ross GA, Lu C, Scarabelli G, Albanese SK, Houang E, Abel R, et al. The maximal and current accuracy of rigorous protein-ligand binding free energy calculations. Commun Chem. 2023;6(1):222. pmid:37838760
- 32. Vaughan AM, Kappe SHI, Ploss A, Mikolajczak SA. Development of humanized mouse models to study human malaria parasite infection. Future Microbiol. 2012;7(5):657–65. pmid:22568719
- 33. Eck T, Laureano de Souza M, Delvillar M, Ashraf K, Yadav Bheemanaboina RR, Chakrasali R, et al. Characterization of competitive inhibitors of Plasmodium falciparum cGMP-dependent protein kinase. Chembiochem. 2022;23(7):e202100704. pmid:35044710
- 34. Fang H, Gomes AR, Klages N, Pino P, Maco B, Walker EM, et al. Epistasis studies reveal redundancy among calcium-dependent protein kinases in motility and invasion of malaria parasites. Nat Commun. 2018;9(1):4248. pmid:30315162
- 35. Miller EB, Hwang H, Shelley M, Placzek A, Rodrigues JPGLM, Suto RK, et al. Enabling structure-based drug discovery utilizing predicted models. Cell. 2024;187(3):521–5. pmid:38306979
- 36. Bheemanaboina RRY, de Souza ML, Gonzalez ML, Mahmood SU, Eck T, Kreiss T, et al. Discovery of imidazole-based inhibitors of Plasmodium falciparum cGMP-dependent protein kinase. ACS Med Chem Lett. 2021;12(12):1962–7. pmid:34917261
- 37. Pottenger AE, Roy D, Srinivasan S, Chavas TEJ, Vlaskin V, Ho D-K, et al. Liver-targeted polymeric prodrugs delivered subcutaneously improve tafenoquine therapeutic window for malaria radical cure. Sci Adv. 2024;10(16):eadk4492. pmid:38640243
- 38. Dodean RA, Li Y, Zhang X, Caridha D, Madejczyk MS, Jin X, et al. Development of next-generation antimalarial acridones with radical cure potential. J Med Chem. 2025;68(8):8817–40. pmid:40179277
- 39. Roth A, Maher SP, Conway AJ, Ubalee R, Chaumeau V, Andolina C, et al. A comprehensive model for assessment of liver stage therapies targeting Plasmodium vivax and Plasmodium falciparum. Nat Commun. 2018;9(1):1837. pmid:29743474
- 40. Wang Y, Jadhav A, Southal N, Huang R, Nguyen D-T. A grid algorithm for high throughput fitting of dose-response curve data. Curr Chem Genomics. 2010;4:57–66. pmid:21331310
- 41. Criscuolo A, Brisse S. AlienTrimmer: a tool to quickly and accurately trim off multiple short contaminant sequences from high-throughput sequencing reads. Genomics. 2013;102(5–6):500–6. pmid:23912058
- 42. El Bakkouri M, Kouidmi I, Wernimont AK, Amani M, Hutchinson A, Loppnau P, et al. Structures of the cGMP-dependent protein kinase in malaria parasites reveal a unique structural relay mechanism for activation. Proc Natl Acad Sci U S A. 2019;116(28):14164–73. pmid:31239348
- 43. Gorrec F. The MORPHEUS protein crystallization screen. J Appl Crystallogr. 2009;42(Pt 6):1035–42. pmid:22477774
- 44. Kabsch W. Automatic indexing of rotation diffraction patterns. Journal of Applied Crystallography. 1988;21(1):67–72.
- 45. Kabsch W. Xds. Acta Crystallogr D Biol Crystallogr. 2010;66(2):125–32. pmid:20124692
- 46. Vonrhein C, Flensburg C, Keller P, Sharff A, Smart O, Paciorek W, et al. Data processing and analysis with the autoPROC toolbox. Acta Crystallogr D Biol Crystallogr. 2011;67(Pt 4):293–302. pmid:21460447
- 47. Evans PR. An introduction to data reduction: space-group determination, scaling and intensity statistics. Acta Crystallogr D Biol Crystallogr. 2011;67(Pt 4):282–92. pmid:21460446
- 48. McCoy AJ, Grosse-Kunstleve RW, Adams PD, Winn MD, Storoni LC, Read RJ. Phaser crystallographic software. J Appl Crystallogr. 2007;40(Pt 4):658–74. pmid:19461840
- 49. Adams PD, Afonine PV, Bunkóczi G, Chen VB, Davis IW, Echols N, et al. PHENIX: a comprehensive Python-based system for macromolecular structure solution. Acta Crystallogr D Biol Crystallogr. 2010;66(Pt 2):213–21. pmid:20124702
- 50. Emsley P, Lohkamp B, Scott WG, Cowtan K. Features and development of Coot. Acta Crystallogr D Biol Crystallogr. 2010;66(Pt 4):486–501. pmid:20383002
- 51. Chen VB, Arendall WB 3rd, Headd JJ, Keedy DA, Immormino RM, Kapral GJ, et al. MolProbity: all-atom structure validation for macromolecular crystallography. Acta Crystallogr D Biol Crystallogr. 2010;66(Pt 1):12–21. pmid:20057044
- 52. Miller EB, Murphy RB, Sindhikara D, Borrelli KW, Grisewood MJ, Ranalli F, et al. Reliable and accurate solution to the induced fit docking problem for protein-ligand binding. J Chem Theory Comput. 2021;17(4):2630–9. pmid:33779166
- 53. Chen W, Cui D, Jerome SV, Michino M, Lenselink EB, Huggins DJ, et al. Enhancing hit discovery in virtual screening through absolute protein-ligand binding free-energy calculations. J Chem Inf Model. 2023;63(10):3171–85. pmid:37167486
- 54. Fajer M, Borrelli K, Abel R, Wang L. Quantitatively Accounting for Protein Reorganization in Computer-Aided Drug Design. J Chem Theory Comput. 2023;19(11):3080–90. pmid:37219932