Abstract
Cations of hydroxy-substituted 1,4-naphthoquinones were synthesized and evaluated as antiplasmodial agents against Plasmodium falciparum. The atovaquone analogues were found to be inactive as antagonists of parasite growth, which was attributed to ionization of the acidic hydroxyl moiety. Upon modification to an alkoxy substituent, the antiplasmodial activity was restored in the sub-100 nM range. Optimal inhibitors were found to possess IC50 values of 17.4–49.5 nM against heteroresistant P. falciparum W2.
Keywords: malaria; Plasmodium; 1,4-naphthoquinones; phosphonium cations
Malaria, particularly that caused by Plasmodium falciparum, remains one of the most important infectious diseases in the world.1,2 As malaria control is seriously limited by resistance to most available drugs,3−5 it is critical that new chemotherapeutics are developed to replace those now limited by drug resistance (e.g., chloroquine and antifolates) and toxicity (e.g., primaquine, amodiaquine, and mefloquine). Antimalarials that inhibit mitochondrial function such as atovaquone (ATV), which is available as a component of Malarone (ATV/proguanil), have particular clinical importance6−8 due to their ability to eradicate both liver and blood stages of the malarial parasite. Structural features common to ATV and related inhibitors are lipophilic substituents,9 which aid in permeability in the mitochondrial matrix membranes where the electron transport complexes are embedded. As a consequence, the physiochemical properties conferred by the hydrophobic groups have detrimental effects on the pharmacokinetic parameters, presenting a significant challenge in the development of mitochondrion-acting treatments of malaria.
Recently, we reported a new class of 1,4-naphthoquinone-based antiparasitic agents believed to be mitochondriotropic antagonists of electron transport.10 The inhibitory capacity of the compounds is thought to be conferred by a phosphonium group that facilitates passive transport of the lipophilic cations across plasma membranes into the energized plasmodial mitochondria. Electrostatic attraction11 is rationalized as the driving force for directed movement into the mitochondrion where the antagonists localize until the membrane potential collapses. This drug design strategy to enhance intracellular bioavailability is the subject of current investigations for improving the in vivo performance of antimalarials that target the Plasmodium mitochondrion.
The most potent of the identified antiplasmodial cations are 4- and 5-hydrocarbon analogues of vitamin K, 1a and 1b (Figure 1). The 1,4-naphthoquinone platform from which the inhibitors were constructed is found in many known antagonists of Plasmodium ubiquinone9,12 including ATV. The binding site of ATV and related naphthoquinone-based inhibitors is the cytochrome bc1 complex, which has been demonstrated in yeast to be mediated by the C-2 hydroxyl of the quinone ring.13 The significance of this alcohol as a H-donor in ATV and on antiplasmodial activity gave reason to examine 1,4-naphthoquinone cations containing a C-2 hydroxyl. Moreover, if the compounds possess potency similar to ATV against cultured P. falciparum, this result will provide evidence that the cations are acting on the bc1 complex. On this basis, we set forth to evaluate phosphonium cations of 2-hydroxy 1,4-naphthoquinone, which demonstrated unexpected antiplasmodial activities that led to the investigations of O-linked acyloxy and alkoxy derivatives of quinone 1.
Figure 1.

Antiplasmodial activities of naphthoquinone-based phosphonium cations against chloroquine-resistant P. falciparum W2.10
The synthetic plan to prepare hydroxyl analogues of naphthoquinone 1 involved first the installation of a n-bromoalkyl side chain to which the phosphonium substituent will subsequently be bound. The hydrocarbon linker was attached by the Kochi–Anderson radical decarboxylation procedure14,15 using 1,4-naphthoquinone 3 as the base material (Scheme 1). The alkylated product 4 was then oxidized to epoxide 5 and treated with sulfuric acid16 to give the 2-hydroxy naphthoquinone 6. Subsequent attempts to directly convert the naphthoquinone to phosphonium cations 9 were unsuccessful and resulted in the formation of complex mixtures of highly polar products. To overcome this problem, the hydroxyl was acetylated prior to the alkylation reaction with a tertiary phosphine (PR3). The phosphonium salts 8 were successfully generated at 100 °C from the acylated quinone 7 in 3:1 iPrOH:PhMe, a solvent combination that was found to minimize thermal degradation of the products during the sluggish reaction. In the final step, acid-catalyzed hydrolysis of the acetyl group provided the desired 2-hydroxy naphthoquinone product 9 (Table 1).
Scheme 1. Synthesis of 2-Hydroxy 1,4-Naphthoquinone Phosphonium Cation 9 (Table 1).
Table 1. Comparisons of IC50 Values for P. falciparum W2 Growth.

| compd | n | R | R1 | IC50 (nM) |
|---|---|---|---|---|
| 16a | 8 | Ph | Ac | 7123 |
| 17a | 1 | Ph | H | 6731 |
| 17b | 8 | Ph | H | 6642 |
| 16b | 2 | Ph | Ac | 5440 |
| 9a | 4 | Ph | H | 4075 |
| 9b | 9 | n-Bu | H | 3865 |
| 8a | 9 | Ph | Ac | 3592 |
| 9c | 9 | Bn | H | 3282 |
| 9d | 9 | Ph | H | 2175 |
| 8b | 4 | Ph | Ac | 1896 |
| chloroquine | 142.7 ± 0.8 | |||
| artemisinin | 18.1 ± 2.2 |
For comparison, a second series of 2-hydroxy 1,4-naphthoquinone cations were synthesized with a benzene ring linking the phosphonium hydrocarbon chain to the platform (Scheme 2). The analogues were prepared from phenyliodonium ylide 11(17) utilizing the BF3-mediated arylation procedure developed by Spyroudis and co-workers.18 The ylide was obtained in high yield from lawsone (10) and (diacetoxy)iodobenzene and then coupled with 4-(bromoalkoxy)-benzaldehydes 13 under reflux in CHCl3. Acetylation of the C-2 hydroxyl followed by alkylation of PPh3 and O-deprotection afforded the aryl-linked cations 17 (Table 1).
Scheme 2. Synthesis of 1,4-Naphthoquinone-Based Phosphonium Cations 16 and 17 (Table 1).
Determination of antiplasmodial activities was performed by assessing 50% inhibitory concentrations (IC50) against chloroquine-resistant P. falciparum (W2 strain) according to methods previously described.19 Surprisingly, each of the 2-hydroxy 1,4-naphthoquinone cations (9 and 17) possessed weak activity with IC50 values between 2175 and 6731 nM (Table 1). Similarly, the O-acetyl analogues (8 and 16) performed poorly as antiplasmodial agents in the medium–low micromolar range. On comparison with the uncharged controls 18 and 19 (IC50 339–392 nM), it was confirmed that the phosphonium moiety had an adverse effect on the IC50 values. With the unexpected results that the phosphonium modification to the 2-hydroxy 1,4-naphthoquinones did not enhance the activity as previously observed for cations 1, the possible rationale behind the reduced efficacy was investigated.
The vast majority of 2-hydroxy 1,4-naphthoquinones, including lawsone (pKa = 3.9820) and phthiocol (pKa = 5.0821), are weak acids that undergo ionization at physiological pH. Depending on the adjacent substituents, anion formation may result in delocalization of the negative charge over the C-2 and C-4 oxygen atoms, which can be ascertained as a bathochromic (red) shift in the UV–vis spectrum.22 In the case of the phosphonium cations 7 and 17, ionization was thought to have occurred in Plasmodium growth medium (pH 7.4), resulting in neutralization of the cationic charge of the inhibitors (Figure 2). The loss of charge needed to facilitate movement into the cell might therefore explain their lack of antiplasmodial activity. To probe if ionization occurred, UV–vis spectroscopy was employed to detect pH-dependent changes in electron delocalization levels of quinone cation 9a and its methyl analogue 1b.
Figure 2.

Ionization of naphthoquinone cations.
On comparison of the UV–vis absorbance spectra of 9a (IC50 = 4075 nM) in buffers ranging from pH 2 to 10, the ionization of the 2-hydroxy residue at neutral pH was confirmed (Figure 3a). Under acidic conditions (pH 2), the quinone gave rise to quinoid transition bands centered at 280 and 335 nm for benzenoid.22 When contrasted with the spectra obtained in neutral and basic pH buffers, the quinoid bands were repositioned at 270 nM and as a broad emission peak at 485 nm. Similar results, albeit of lesser intensity, were observed in 1-octanol, a medium that mimics the physiochemical properties of biological membranes.
Figure 3.

UV–vis absorption spectra of phosphonium cations 9a and 1b in 1-octanol and acidic–basic phosphate buffers.
The quinoid absorbance at the higher λ represents an increase in π electron delocalization.23 This was attributed to ionization of the hydroxy group and distribution of the resulting anionic charge into the quinone. The nearly identical emission spectra for quinone 9a at pH 7 and 10 was evidence that the deprotonated form exists in cells and the physiological environment. The cationic feature, which is believed to direct the molecules to the mitochondrion, would thereby be suppressed by the ionization, rendering the compounds inactive. As further confirmation of the pH-dependent change in the ionization state of hydroxy-substituted analogues, the UV–vis absorbance spectra of the C-2 methyl analogue 1b were compared, and little variability in pH 2–10 buffers was observed (Figure 3b).
On the basis of these results, it is concluded that in order for the naphthoquinone cations to function as growth antagonists, an acidic moiety cannot be present in the molecule. To test this hypothesis, the 2-methoxy analogue 21 of quinone cation 17a (IC50 = 6731 nM) was synthesized from 2-hydroxy naphthoquinone 14 and tested against P. falciparum W2 (Scheme 3). As expected, potent activity was restored (IC50 = 58.3 ± 14.3 nM) upon removal of the acidic hydroxy substituent on the quinone ring. With this finding, additional O-substituted quinone cations were synthesized for evaluation as antiplasmodial agents.
Scheme 3. Synthesis of 2-Methoxy 1,4-Naphthoquinone Phosphonium Cation 21.
For the preparation of 3-acyloxy- and 2-alkoxy-linked naphthoquinone cations, methyl-1,4-naphthoquinone (menadione, 22, Scheme 4) was used as the initial base material. Installation of the C-2 hydroxyl was achieved by acid-mediated cleavage of the corresponding menadione epoxide 23.16 Acylation of phthiocol (24) with 4-bromobutanoyl and 11-bromoundecanoyl chlorides followed by alkylation of PPh3 afforded the 3-acyloxy-linked cation derivatives 26a (n = 2) and 26b (n = 9).
Scheme 4. Synthesis of 3-Acyloxy 1,4-Naphthoquinone Cations 26 (Table 2).
Both phthiocol (24) and lawsone (10) were similarly employed as base materials for the synthesis of 2-alkoxy naphthoquinone cations 28 (Scheme 5). The n-bromoalkyl chain linkers were attached to the quinone platform under THF reflux with K2CO3, 18-crown-6, and tetrabutylammonium iodide (TBAI). Supplementing the reaction with two phase-transfer catalysts was found to enhance nucleophilicity of the C-2 oxygen and hinder formation of K+-naphthoquinone chelation complex.24 Alternatively, a n-bromopropyl chain was attached to phthiocol via a glycolic spacer prior to installation of a phosphonium group in quinone 30. In the final step, the naphthoquinone cations 28a–e (Table 2) and 31 were generated by the alkylation of PPh3 in 3:1 iPrOH:PhMe at 100 °C.
Scheme 5. Synthesis of 2-Alkoxy Naphthoquinone Phosphonium Cations 28a–e and 31 (Table 2).
Table 2. Comparisons of IC50 Values for P. falciparum W2 Growth.
| compd | n | R | IC50 (nM) |
|---|---|---|---|
| 26a | 2 | Me | 4604 |
| 26b | 9 | Me | 1377 |
| 31 | 1 | Me | 1230 |
| 28a | 8 | Me | 119.3 ± 11.3 |
| 28b | 4 | Me | 42.7 ± 1.4 |
| 28c | 1 | Me | 41.9 ± 4.0 |
| 28d | 1 | H | 40.0 ± 0.8 |
| 28e | 4 | H | 28.5 ± 3.0 |
| 36a | 4 | 4-ClBn | 49.5 ± 7.8 |
| 36b | 4 | Bn | 49.1 ± 6.3 |
| 36c | 4 | C6H11 | 47.9 ± 6.3 |
| 36d | 1 | Bn | 46.7 ± 4.4 |
| 36e | 1 | C6H11 | 42.3 ± 0.0 |
| ATV | 0.50 ± 0.2 |
For further comparison, additional 2-hydroxy-3-alkyl-naphthoquinones prepared from the algaecide Dichlon (32) were used to synthesize 2-alkoxy linked cation analogues 36 (Scheme 6). Derivatization of the C-3 position with cyclohexyl, benzyl, and 4-chlorobenzyl substituents was again achieved by radical decarboxylation.14,15 Following methanolysis of the remaining chlorine atom, the hydrocarbon linkers were attached to the C-2 hydroxyl, and the corresponding triphenylphosphonium bromides 36a–e (Table 2) were prepared as previously described.
Scheme 6. Synthesis of 1,4-Naphthoquinone-Based Phosphonium Cations 36a–e (Table 2).
Antiplasmodial testing revealed that the 2-alkoxy analogues 28 and 36 possessed significantly higher activity than their 3-acyloxy counterparts 26. The IC50 values of ester-linked derivatives including quinone 31 ranged from 1230 to 4604 nM, suggesting that the lipid chains with the triphenylphosphonium group bound were susceptible to intracellular hydrolysis. Conversely, the alkoxy-linked quinones (28b–e and 36a–e), which would be stable to esterase degradation possessed sub-50 nM IC50 values. Although few correlations between structure and activity could be ascertained for these analogues, compounds with shorter alkoxy linkers appeared to be slightly more effective antagonists of parasite growth. Little variability was also observed between analogues with different C-3 substituents with the nominal exception of the lawsone-derived cation 28e (IC50 = 28.5 ± 3.0 nM).
A structural feature lacking in each of the cations listed in Table 2 is a rigid substituent attached to the quinone ring. To compare the effect of restricted rotation of the C-3 residue on activity, an analogue bound with a p-anisyl group was synthesized (Scheme 7). Ylide 11 was once again employed as the base material to construct the 2-hydroxy naphthoquinone 38 via the BF3-mediated arylation procedure.18 Attachment of a n-propyl chain linker and installation of a triphenylphosphonium moiety afforded the p-anisyl quinone cation 40. Subsequent antiplasmodial testing revealed that the incorporation of rigidity at C-3 resulted in nearly a 2-fold increase in activity (IC50 = 17.4 ± 4.1 nM) as compared with analogues 28 and 36. An additional property of cation 40 that could be contributing to the enhanced efficacy is molecular planarity. With this finding, the correlation between planarity and antiparasitic activity of the cationic inhibitors will be a subject of future studies.
Scheme 7. Synthesis of 3-(4-Anisyl)-1,4-naphthoquinone Phosphonium Cation 40.
In summary, O-linked phosphonium cations of 1,4-napthoquinones were discovered to be effective growth inhibitors of cultured P. falciparum with IC50 values in the nanomolar range. The activity was greatest for quinones that did not possess an acidic functionality and were bound to the phosphonium moiety via a short hydrocarbon chain linker. Little variability in the IC50 values was observed for the 2-alkoxy cation analogues 28 and 36 with a H or sp3-hybridized carbon residue bound at the C-3 position. Alternatively, when a rigid aromatic group was attached to the C-3 position of the quinone, a definitive increase in inhibitory activity was observed.
Future investigations will continue to examine the effect of rigidity and planarity on activity. In addition, analogues possessing nitrogen-based cations will be evaluated as antiparasitic agents. Preliminary studies have revealed that the pyridinium (41) and imidazolium (42) derivatives of quinone 28c demonstrate weaker antiplasmodial activity than their phosphonium counterpart (Figure 4). Although the initial results have shown them to be several fold less active, cationic inhibitors containing a quaternary amine may possess better pharmaceutical properties and ultimately prove to have greater in vivo efficacy than the phosphonium analogues.
Figure 4.
Antiplasmodial activity comparisons of pyridinium and imidazolium analogues of triphenylphosphonium cation 28c against P. falciparum W2.
Supporting Information Available
Synthetic procedures and characterization data of reported compounds. This material is available free of charge via the Internet at http://pubs.acs.org.
Financial support was generously provided by the R. C. Wilson Pharmacy Fund from the College of Pharmacy at The University of Georgia to T.E.L., the National Institutes of Health to P.J.R., and the Doris Duke Charitable Foundation, with which P.J.R. is a Distinguished Clinical Scientist.
The authors declare no competing financial interest.
Funding Statement
National Institutes of Health, United States
Supplementary Material
References
- Greenwood B. M.; Bojang K.; Whitty C. J.; Targett G. A. Malaria. Lancet 2005, 365, 1487–98. [DOI] [PubMed] [Google Scholar]
- Snow R. W.; Guerra C. A.; Noor A. M.; Myint H. Y.; Hay S. I. The global distribution of clinical episodes of Plasmodium falciparum malaria. Nature 2005, 434, 214–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- A research agenda for malaria eradication: Drugs. PLoS Med. 2011, 8, e1000402. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Fidock D. A. Drug discovery: Priming the antimalarial pipeline. Nature 2010, 465, 297–298. [DOI] [PubMed] [Google Scholar]
- Greenwood B. Anti-malarial drugs and the prevention of malaria in the population of malaria endemic areas. Malar. J. 2010, 9Suppl. 3S2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Biagini G. A.; Fisher N.; Shone A. E.; Mubaraki M. A.; Srivastava A.; Hill A.; Antoine T.; Warman A. J.; Davies J.; Pidathala C.; Amewu R. K.; Leung S. C.; Sharma R.; Gibbons P.; Hong D. W.; Pacorel B.; Lawrenson A. S.; Charoensutthivarakul S.; Taylor L.; Berger O.; Mbekeani A.; Stocks P. A.; Nixon G. L.; Chadwick J.; Hemingway J.; Delves M. J.; Sinden R. E.; Zeeman A. M.; Kocken C. H.; Berry N. G.; O'Neill P. M.; Ward S. A. Generation of quinolone antimalarials targeting the Plasmodium falciparum mitochondrial respiratory chain for the treatment and prophylaxis of malaria. Proc. Natl. Acad. Sci. U.S.A. 2012, 109, 8298–8303. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Vaidya A. B. Mitochondrial and plastid functions as antimalarial drug targets. Curr. Drug Targets Infect. Disord. 2004, 4, 11–23. [DOI] [PubMed] [Google Scholar]
- Baggish A. L.; Hill D. R. Antiparasitic agent atovaquone. Antimicrob. Agents Chemother. 2002, 46, 1163–1173. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Rodrigues T.; Lopes F.; Moreira R. Inhibitors of the mitochondrial electron transport chain and de novo pyrimidine biosynthesis as antimalarials: The present status. Curr. Med. Chem. 2010, 17, 929–956. [DOI] [PubMed] [Google Scholar]
- Long T. E.; Lu X.; Galizzi M.; Docampo R.; Gut J.; Rosenthal P. J. Phosphonium lipocations as antiparasitic agents. Bioorg. Med. Chem. Lett. 2012, 22, 2976–2979. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Smith R. A.; Hartley R. C.; Cocheme H. M.; Murphy M. P. Mitochondrial pharmacology. Trends Pharmacol. Sci. 2012, 33, 341–352. [DOI] [PubMed] [Google Scholar]
- Porter T. H.; Folkers K. Antimetabolites of coenzyme Q. Their potential application as antimalarials. Angew. Chem., Int. Ed. Engl. 1974, 13, 559–569. [DOI] [PubMed] [Google Scholar]
- Kessl J. J.; Hill P.; Lange B. B.; Meshnick S. R.; Meunier B.; Trumpower B. L. Molecular basis for atovaquone resistance in Pneumocystis jirovecii modeled in the cytochrome bc(1) complex of Saccharomyces cerevisiae. J. Biol. Chem. 2004, 279, 2817–2824. [DOI] [PubMed] [Google Scholar]
- Anderson J. M.; Kochi J. K. Silver(I)-catalyzed oxidative decarboxylation of acids by peroxydisulfate - Role of silver(II). J. Am. Chem. Soc. 1970, 92, 1651–1659. [Google Scholar]
- Commandeur C.; Chalumeau C.; Dessolin J.; Laguerre M. Study of radical decarboxylation toward functionalization of naphthoquinones. Eur. J. Org. Chem. 2007, 3045–3052. [Google Scholar]
- Madej M. G.; Nasiri H. R.; Hilgendorff N. S.; Schwalbe H.; Unden G.; Lancaster C. R. D. Experimental evidence for proton motive force-dependent catalysis by the diheme-containing succinate: Menaquinone oxidoreductase from the Gram-positive bacterium Bacillus licheniformis. Biochemistry 2006, 45, 15049–15055. [DOI] [PubMed] [Google Scholar]
- Wu C. M.; Johnson R. K.; Mattern M. R.; Wong J. C.; Kingston D. G. I. Synthesis of furanonaphthoquinones with hydroxyamino side chains. J. Nat. Prod. 1999, 62, 963–968. [DOI] [PubMed] [Google Scholar]
- Glinis E.; Malamidou-Xenikaki E.; Skouros H.; Spyroudis S.; Tsanakopoulou M. Arylation of lawsone through BF3-mediated coupling of its phenyliodonium ylide with activated arenes and aromatic aldehydes. Tetrahedron 2010, 66, 5786–5792. [Google Scholar]
- Coteron J. M.; Catterick D.; Castro J.; Chaparro M. J.; Diaz B.; Fernandez E.; Ferrer S.; Gamo F. J.; Gordo M.; Gut J.; de las Heras L.; Legac J.; Marco M.; Miguel J.; Munoz V.; Porras E.; de la Rosa J. C.; Ruiz J. R.; Sandoval E.; Ventosa P.; Rosenthal P. J.; Fiandor J. M. Falcipain inhibitors: Optimization studies of the 2-pyrimidinecarbonitrile lead series. J. Med. Chem. 2010, 53, 6129–6152. [DOI] [PubMed] [Google Scholar]
- Zaugg H. E.; Rapala R. T.; Leffler M. T. Naphthoquinone antimalarials. 14. 2-Hydroxy-3-aryl-1,4-naphthoquinones. J. Am. Chem. Soc. 1948, 70, 3224–3228. [DOI] [PubMed] [Google Scholar]
- Ball E. B. Studies on oxidation-reduction. XXI. Phthiocol, the pigment of the human tubercle bacillus. J. Biol. Chem. 1934, 106, 515–524. [Google Scholar]
- Singh I.; Ogata R. T.; Moore R. E.; Chang C. W. J.; Scheuer P. J. Electronic spectra of substituted naphthoquinones. Tetrahedron 1968, 24, 6053–6073. [Google Scholar]
- Juster N. J. Color and chemical constitution. J. Chem. Educ. 1962, 39, 596–601. [Google Scholar]
- Lu X.; Althawari A.; Hansen E. N.; Long T. E.. Phase-transfer catalysts in the O-alkylation of 2-hydroxynapthoquinones. Synthesis, 2012, 44, in press. [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.









