Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2020 Oct 1.
Published in final edited form as: Inorganica Chim Acta. 2019 Jul 22;496:119029. doi: 10.1016/j.ica.2019.119029

Artesunate Activation by Heme in an Aqueous Medium

Laura Heller 1, Paul D Roepe 1,2,3,*, Angel C de Dios 1,3,*
PMCID: PMC7442224  NIHMSID: NIHMS1536117  PMID: 32831389

Abstract

The reaction between the antimalarial drug artesunate (ATS) and ferriprotoporphyrin_(IX) (FPIX) in the presence of glutathione (GSH) has been monitored by nuclear magnetic resonance (NMR) spectroscopy. By following the disappearance of resonances of protons near the endoperoxide group in ATS, the rate at which the drug is activated can be directly measured. In an aqueous medium, the rate of ATS activation is limited by the rate of reduction of the FPIX Fe(III) center by GSH. The reaction is observed to slow dramatically in the presence of other heme binding antimalarial drugs. These findings explain the long observed antagonism between artemisinin derivatives and quinoline-based drugs. This discovery suggests that combination therapy that involves artemisinin or any of its derivatives and a quinoline-based drug may be compromised.

Graphical Abstract

graphic file with name nihms-1536117-f0001.jpg

Monitoring artesunate protons by nuclear magnetic resonance spectroscopy provides a means for measuring the activation of this drug by reduced heme. The presence of quinoline-based antimalarial drugs such as chloroquine, quinine and amodiaquine are shown to inhibit this activation.

Introduction

Malaria remains a serious challenge to human health. Annually this disease affects about three billion people living in malaria-endemic regions as well as international travelers, governmental workers and deployed military personnel. Previous first - line drugs such as chloroquine (CQ) and sulfadoxine-pyrimethamine (SP) are no longer viable in many regions due to widespread drug resistance. These drugs have largely been replaced by artemisinin combination therapies (ACTs) composed of a derivative of artemisinin and one of several possible "partner" drugs (Figure 1).1 Unfortunately, recently, a harbinger of artemisinin resistance known as the "delayed clearance phenotype" (DCP) has been reported across Southeast Asia.2 Consequently, dihydroartemisinin (DHA) with piperaquine (PPQ; Figure 1) as a partner drug, followed by a single low-dose primaquine (PQ) treatment is now recommended in areas of Cambodia where both artemisinin resistance (ARTR) and PPQR is emerging.3,4 The molecular pharmacology of these drugs remains incompletely understood. Elucidating this pharmacology is essential for understanding pathways to resistance and for the development of next generation therapies.

Figure 1.

Figure 1.

Open in a new tab

Structures of antimalarial drugs used in this study and their common abbreviations. ATM, ATS and DHA are derivatives of the parent drug ART that are commonly used in the malaria clinic. LUM, PPQ, and AQ are commonly used “partner” drugs in artemisinin combination therapies (ACTs).

The ART 1,2,4 trioxane, specifically its bridging endoperoxide group, has been shown to be crucial for drug activity.5-9 However, the detailed molecular mechanism of endoperoxide activation within the malarial parasite, and how the activated drug then kills the parasite remain active areas of research. The trioxane group is highly reactive,10 and pioneering work of Meshnick et al11 suggested that antiparasitic reactivity requires activation by Fe(II). Two competing models have been proposed to describe this reaction. One envisions initial Fe(II) mediated cleavage of the peroxide bond.12-16 A second model involves ring opening driven by either protonation of the peroxide or action by a Lewis acid,17-20 thereby assigning only a minor role to Fe(II). Recent work suggests that ferri-protoporphyrin(IX), Fe(III)-PPIX, in its reduced form, Fe(II)PPIX, and not necessarily labile Fe(II), may account for the activation of ART - based drugs within malarial parasites. 21-23 Two studies have shown that >100 parasite proteins are potential ART drug alkylation targets,21,24 but overlap between these two proteomic data sets is limited, suggesting protein alkylation by activated ART - based drugs may be somewhat random.9,23 One recent study has found that activated DHA alkylates free FPIX within parasites23 and synthetic endoperoxide - based antimalarials have been shown to alkylate proteins.25 Currently it is unknown which ART - based drug adducts produced within the parasite are pharmacologically the most relevant. DHA - FPIX adducts are known to be toxic and some protein adducts might be predicted to activate stress response and /or cell death pathways, suggesting that both protein and FPIX adducts may be relevant.

The promiscuous targeting of ART and its activation by Fe(II)PPIX are somewhat consistent with current data on emerging ART resistance (ARTR), also known as the "delayed clearance phenotype" (DCP). Specifically, PfKelch13 amino acid substitutions are associated with DCP/ARTR primarily at the ring stage of parasite development,26-28 during which ART is not as efficiently activated due to lower availability of free Fe(II)PPIX relative to other parasite stages.22,23 It has been suggested that increasing ART treatment duration29 or the use of δ-aminolevulinic acid, a known heme precursor, may be effective in combating ARTR. 21

Fe(III)-PPIX is liberated during obligate parasite hemoglobin (Hb) catabolism that occurs during parasite red blood cell development, and is also a known target for other antimalarial drugs. For example, quinoline-based drugs such as chloroquine (CQ), quinine (QN) and amodiaquine (AQ) interact directly with Fe(III)-PPIX, and prevent its crystallization to inert hemozoin30-33. Using nuclear magnetic resonance (NMR) spectroscopy and the effects of the Fe(III)-PPIX paramagnetic Fe(III) center on the relaxation rates of drug protons, solution structures of noncovalent complexes formed between quinoline - based drugs and Fe(III)-PPIX have been determined.30,33 Furthermore, it is now known that an equilibrium exists between monomeric and dimeric Fe(III)-PPIX species. This equilibrium has been extensively studied as a function of pH, and in the presence of either CQ or QN.31 At low pH, the monomeric species is favored while at higher pH values, the antiferromagnetic dimer dominates. The transition in purely aqueous medium occurs at a pH lower than 5; only the antiferromagnetic species is then observed as precipitation begins to occur upon protonation of the heme propionic acid side chains. Abundant monomeric species is therefore observed only if a mixture of DMSO and H2O is used as solvent. In these previous studies, CQ was found to promote dimer formation while QN promotes the converse, highlighting that even closely related quinoline antimalarial drugs may have different molecular mechanisms of action.32 The state of Fe(III)-PPIX in solution is clearly consequential to how it interacts with quinoline antimalarial drugs,30-33 and is likely also important for interaction with ART - based drugs, which are activated by Fe(II)-PPIX.7,8,11,21,23 These issues are critical for elucidating the pharmacology of some ACTs, which combine ART - based and quinoline drugs.

ART - based drugs act rapidly against P. falciparum, and are eliminated quite rapidly, having a half-life in plasma of about 1 to 2 hours. Effective ACTs thus include drugs with longer half-lives such as LF, PPQ, and AQ (Figure 1).34,35 When Fe(II)-PPIX is released upon parasite Hb digestion, Fe(II) is presumably oxidized to Fe(III) due to the presence of molecular oxygen. For ART activation, Fe(II) is required. Therefore, if Fe(II)-PPIX is relevant for ART activation within malarial parasites, the drug must either encounter Fe(II)-PPIX as it is released during Hb catabolism or a reducing agent (presumably GSH) must convert Fe(III) to Fe(II). GSH is found at significant concentrations inside the parasite.36 Using GSH to reduce Fe(III)-PPIX, efficient DHA alkylation of heme has previously been demonstrated.23,37 Unfortunately, most in vitro studies involving Fe(II)-PPIX and ART - based drugs have been performed in organic solvents or mixed aqueous-organic medium that reduce Fe(II)-PPIX self-association relative to an aqueous environment. Rapid Fe(II)-PPIX and Fe(III)-PPIX self-association in aqueous environment can involve either cofacial π–π interactions or hydroxo-bridge bonds.38 The formation of Fe(II)/Fe(III) protoporhyrin IX complexes, on the other hand, has not been observed in DMSO,37 indicating that the behavior of FPIX, ART and GSH in an aqueous environment may be significantly different from what is observed in DMSO.39

Notwithstanding these complex studies using aqueous - DMSO solutions have been, and remain, useful.31 The higher viscosity relative to aqueous media dictates that Fe(III)-PPIX protons can be readily observed by NMR spectroscopy due to favorable rotational correlation times. In acidic DMSO, the FPIX monomer is dominant resulting in NMR signals that display substantial paramagnetic shifts. The four peaks observed with chemical shifts greater than 55 ppm from tetramethylsilane (TMS) arise from the protons of the four inequivalent methyl groups attached to the porphyrin ring in FPIX. In theory then, observing the reaction between ART and FPIX in the presence of GSH in acidic aqueous DMSO solution would provide concrete evidence of alkylation of the porphyrin ring. Also, the 55-70 ppm chemical shift region allows for non-invasive direct measurement of the rate of Fe(II)FPIX reduction by GSH. FPIX methyl peaks would disappear as Fe(III) is reduced to Fe(II), which would then correlate with appearance of oxidized glutathione (GSSG) peaks. As Fe(II) is oxidized back to Fe(III) upon reductive activation of the ART peroxide, methyl peaks would reappear. With subsequent alkylation of the FPIX porphyrin ring by ART, new methyl resonances would then be observed. In such studies, Fe(II)-PPIX protons would not be directly observable, however, drug protons can be monitored.

We have used these principles to examine the kinetics of ART - based drug / FPIX interactions and how they are influenced by well-known FPIX aggregation. In addition to elucidating important factors that affect how ART is activated by FPIX, these data provide insight on how ACT partner drug interactions with FPIX may influence ART - derived drug potency. We consider the following quinoline-based drugs; chloroquine, piperaquine, amodiaquine, lumefantrine, pyronaridine, quinine, primaquine and 9-epi-quinine.

C. Experimental Details

Ferriprotoporphyrin IX (Fe(III)-PPIX) chloride, glutathione (GSH), CQ diphosphate, PPQ tetraphosphate tetrahydrate, PQ bisphosphate, and lumefantrine (LUM) were from Sigma (St. Louis, MO), artemisinin (ART), artesunate (ATS), and artemether (ATM) were from Cayman (Ann Arbor, MI), and dihydroartemisinin (DHA) was from TCI America (Tokyo, Japan). QN monohydrochloride was from Acros Organics (Pittsburgh, PA), AQ dihydrochloride was from Fluka (Buchs, Switzerland), and PYR tetraphosphate (PY) was from Santa Cruz Biotech (Dallas, TX). Epiquinine (Epi-QN) was from eBioChem (Shanghai, China) and deuterated-DMSO(d6) and D2O were from Cambridge Isotope Labs (Andover, MA). Reaction mixtures were typically made to the following final concentrations: FPIX (0.375 mM), GSH (1.88 mM) and ATS (0.375 mM). To study the effect of partner antimalarial drugs on the reaction between ATS and FPIX, an equimolar amount (0.375 mM) of the partner drug was added to this mixture. The reaction mixtures were kept at 37°C and buffered with phosphate at pH 7.0. Reactions were monitored by 1H NMR spectroscopy using a Varian Unity INOVA 500 MHz NMR Spectrometer (Agilent).

Several of these reaction mixtures were also analyzed by mass spectrometry using FPIX chloride, GSH, ATS, and either QN or EPI (1:5:1:1 molar ratio) in H2O at pH 7.0. Solutions were diluted in 1:1 MeOH:H2O and the final mixtures were subjected to electrospray ionization mass spectrometry (ESI-MS), in the positive ion mode, using a Varian 500 ion trap mass spectrometer (Agilent). The electrospray ionization source was operated at 5 kV with N2 as the nebulizing gas. First-order ESI mass spectra were recorded in the mass range m/z 900–1400. The drug-heme solutions were introduced into the mass spectrometer by a syringe pump (Hamilton), employing a 1 mL syringe at a constant flow rate of 10 μL/min,

In the absence of GSH, unlike other antimalarial drugs, neither ART nor any of its derivatives is found to interact strongly with the Fe(III)-PPIX monomer, as observed in slightly acidic DMSO, or its μ-oxo dimer, as observed in water at neutral pH. The relaxation times of the protons in ART are not significantly perturbed by the presence of Fe(III)-PPIX. These drugs also appear to be stable in these solutions and no change in the NMR spectra is observed after a couple of days, except for the hydrolysis of the succinyl ester in ATS or the ether linkage in ATM. Aqueous solutions of ATS with GSH at the same concentration levels intended for the experiments described in this work have also been monitored to check for stability. And in forty eight hours, no reaction is observed between ATS with GSH. The deuterated solvents and all the reaction mixtures employed in this study have not been degassed since NMR relaxation time measurements do not indicate appreciable presence of molecular oxygen in any of the samples. The solvent employed in the final reaction mixture is 5%(by volume) d6-DMSO in D2O.

D. Results and Discussion

Reaction between FPIX and ATS was monitored by NMR spectroscopy in aqueous / DMSO solution. Most ATS peaks in the 1D proton NMR spectrum are multiplets except for those due to protons of the methyl group attached to C3 and the proton attached to C12 (Fig. 2). In the absence of GSH and FPIX, the 3-CH3 signal, being a singlet and arising from three 1H, is the most intense in the ATS 1D spectrum (Fig 3). This methyl is also proximal to the ATS endoperoxide (Fig. 2) such that the 3-CH3 signal disappears upon ATS reaction with Fe(II)- protoporphyrin IX (Figure 3; compare b vs a). The disappearance of the 3-CH3 signal is quite dramatic and is accompanied by the appearance of a resonance near 2.0 ppm (Fig. 3b), which can be assigned to the protons of an acetate ion or an acetyl group liberated during the reaction between ATS and Fe(II)-PPIX.13,14 Monitoring the rate of the loss of 3-CH3 signal provides a convenient method for measuring the rate of the reaction between Fe(II)-PPIX and ATS. In pure DMSO, where Fe(III)-PPIX is first incubated with GSH overnight, the reaction is quite fast such that it is complete before an NMR spectrum can be obtained. In water, on the other hand, where Fe(II)-PPIX is formed in situ, the reaction is much slower.

Figure 2.

Figure 2.

Open in a new tab

Artesunate (ATS) and the numbering scheme employed in this work.

Figure 3.

Figure 3.

Open in a new tab

The 3-CH3 resonance of ATS before (a) and after (b) reaction with Fe(II)FPIX heme formed in situ from Fe(III)FPIX and GSH. Peaks marked with “*” are tentatively assigned to acetate and acetyl proton.

We studied the Fe(II)FPIX - ATS reaction in aqueous medium in two different ways. The first initially incubated Fe(III)-PPIX with GSH to reduce all Fe(III) to Fe(II) before adding ATS. The second mixed aqueous solutions of Fe(III)-PPIX, GSH and ATS simultaneously, such that ATS reacted with Fe(III)-PPIX immediately upon its reduction to Fe(II). Overnight incubation of Fe(III)-PPIX with GSH showed a significant decrease in the rate of Fe(II)-PPIX activation of ATS (once ATS was added) as monitored by loss of the 3-CH3 resonance (Figure 4b), relative to the rate observed when Fe(III)-PPIX, GSH and ATS were mixed simultaneously (Fig. 4a). Using peak integrals to quantify loss of the 3-methyl signal, we compared these two reactions (Figure 5a). The difference in rate is easily explained by self-association of Fe(II)-FPIX, when formed in aqueous medium, as previously observed by Monti et al.38 That is, it is well known that Fe(II)- PPIX readily self-associates with Fe(III)-PPIX via oxo-bridges, which would significantly reduce ATS access to the Fe(II) center, thereby lowering ATS activation efficiency as observed (Fig. 4b).

Figure 4.

Figure 4.

Open in a new tab

(a) NMR resonance signal of 3-CH3 of ATS obtained in 30 minute intervals from an aqueous sample containing FPIX, GSH and ATS, (b) as in (a), but with overnight incubation of FPIX with GSH before adding ATS.

Figure 5.

Figure 5.

Open in a new tab

Conversion of ATS to C4 centered radical ATS vs time with (top, ×) and without (bottom, ◆) prior complete reduction of FPIX by GSH (see text).

As shown in Figure 5, the reaction in aqueous medium has not reached completion even after 3 hr. This is in stark contrast with the fast reaction observed in DMSO by Robert et al.37 In a nonpolar solvent that maintains FPIX in monomeric form, Fe(III)-PPIX alkylation by ATS is complete within a few minutes.40 It should also be noted that reactant concentrations in the current work are an order of magnitude lower than those used by Robert and colleagues.37 In D2O, where Fe(III)-PPIX protons are completely invisible, the reduction of Fe(III) in FPIX can be followed by monitoring oxidized GSH and/or reduced GSSG resonances, specifically those from β-cysteinyl protons. The appearance of GSSG peaks correlates with the disappearance of the 3-CH3 signal. We find that, in aqueous media, and at physiologically relevant concentrations, the reaction between ATS and FPIX strictly correlates with conversion of GSH to GSSG, suggesting that the reduction of Fe(III)-PPIX by GSH is rate limiting (Fig. 6).

Figure 6.

Figure 6.

Open in a new tab

Conversion of ATS to C4 centered radical ATS vs time (top, ◆) and oxidation of GSH into GSSG vs. time (bottom, ×)

When incubated alone with Fe(III)-PPIX, neither GSH nor ATS bind as strongly to FPIX Fe(II) as do quinoline antimalarial drugs such as CQ.30 NMR relaxation measurements of ATS or GSH protons in the presence of Fe(III)-PPIX do not show an appreciable decrease in longitudinal relaxation times (T1). The fact that NMR signals from ATS and GSH can still be observed in these samples suggests that the dissociation constant between either ATS or GSH with FPIX is > 1 mM. 1H attached to C10 or C12 in ATS demonstrate a T1 of about 0.8 s in D2O (not shown, see [31]). In the presence of Fe(III)-PPIX in D2O, the T1 for these protons is reduced to only 0.6 s. In DMSO, the reduction in T1 is stronger; to 0.3 s, indicating a slightly higher affinity for Fe(III)-PPIX monomer in this solvent. In contrast, proton resonances from CQ are barely observed in aqueous samples that contain 1 mM CQ and 100 μM Fe(III)-PPIX.31

As ATS reacts with Fe(II)-PPIX, new signals are observed near the methyl region of the proton NMR spectrum (Fig. 3b). These new signals (asterisks, Fig. 3b) are attributed to either an acetate or acetyl group (AcO) liberated during ATS alkylation of Fe(III)-PPIX.23 The growth of these signals does not occur at the same rate as the disappearance of the ATS 3-CH3 resonance due to activation of ATS (Fig. 3a). This is consistent with the formation of one or more alkylation intermediates as previously suggested41 and recently observed.23 The reaction scheme is shown in Figure 7. First, Fe(II)-PPIX breaks the ATS endoperoxide bridge, and due to the formation of an O-centered radical and proximity of the paramagnetic Fe(III), proton resonances from ATS are significantly perturbed in the NMR spectrum as an iron-oxo intermediate forms (Fig. 7, left). The O-centered radical then rearranges to form a primary carbon centered radical (Fig. 7, middle). This radical then alkylates the porphyrin ring of Fe(III)-PPIX (right) which gives rise to an acetyl peak in the proton NMR spectrum (see Fig. 3b). Since the appearance of free acetyl signal is delayed, the release of this acetyl group does not occur immediately after the alkylation of Fe(III)-PPIX (Fig. 7). This delay, however, may likewise suggest that alkylation of the porphyrin ring does not occur instantly. Intermediates prior to Fe(III)-PPIX alkylation are highly reactive as these contain either O- or C-centered radicals 23 If Fe(III)-PPIX alkylation does not occur instantly, these intermediate(s) might then facilitate alkylation of other targets such as parasite proteins24 after diffusion of the ATS radical. This covalent ATS mediated sequestration of Fe(III)-PPIX is distinctly different from noncovalent quinoline antimalarial drug sequestration.42

Figure 7.

Figure 7.

Open in a new tab

Proposed mechanism for alkylation of FPIX by ATS (see [23])

Quinoline antimalarial drugs such as CQ interact strongly with Fe(III)-PPIX in aqueous solutions.30 NMR studies in solution indicate that while interaction between quinoline drug and Fe(III)-PPIX are significant, the dominant complex formed is noncovalent and there is fast exchange between free and bound drug.30,33 Nevertheless, when quinoline drugs bind to Fe(III)- PPIX, access to the Fe(III) center is hindered.30,33 We therefore hypothesized that the reaction between Fe(II)-PPIX and ATS might be negatively impacted by the presence of a quinoline drug. Using NMR experiments similar to those described above, (e.g., simultaneous mixing of aqueous solutions of FPIX at pH 7.0, GSH, ATS, and equimolar heme-targeted antimalarial (HTA) drug), we observed results similar to those obtained with prior incubation of FPIX with GSH (Figure 8). Namely, the reaction between Fe(II)FPIX and ATS is significantly impeded by the presence of CQ and ACT partner drugs such as AQ, but not by 9-epi-quinine (EPI), which binds much more weakly to Fe(III)-PPIX, in an edge - to - face fashion.32 Figure 9 shows that even with a CQ:FPIX molar ratio of 1:4, the rate of disappearance of the 3-CH3 signal is still significantly slowed. With only one bound CQ molecule per four molecules of Fe(III)-PPIX,30 predicted ATS activation is only 20 percent compared to control (compare Figs. 9 vs 5 and 6). Similar results are obtained with QN and AQ where only 30 and 20 percent of ATS is activated after 4 hours with drug:FPIX ratio of 1:4 (not shown). Correspondingly, in the presence of quinoline drug, the increase in GSSG peaks, specifically those of β-cysteinyl proton resonances, are significantly delayed (not shown). The quinoline drug therefore prevents reduction of Fe(III)FPIX by GSH. Without Fe(II)-PPIX, ATS cannot be activated.

Figure 8.

Figure 8.

Open in a new tab

The amount of unreacted ATS as a function of time in the absence (●) vs presence of other FPIX - binding antimalarial drugs: CQ; ◇, PPQ; ◆, AQ; ◻, PY; ▴, LUM; +, QN; ▵, PQ; ∎, and 9-epi-quinine (EPI; ◯)

Figure 9.

Figure 9.

Open in a new tab

The amount of unreacted ATS as a function of time in the presence of CQ, at the following CQ:FPIX ratios: 1:1 (●), 1:2 (◯), 1:4 (◻), and 0:1 (∎).

Why equimolar quinolinal drug is not required for this effect illustrates the nature of Fe(III)-PPIX heme in aqueous medium. Access to the Fe(III) center would be limited even with a substoichiometric amount of the quinoline drug if heme already exists in an aggregated form as proposed.43 In such circumstances, only the ends of the FPIX aggregates would provide Fe(III) for GSH reduction. When a quinoline drug caps the end of these aggregates,30 thereby denying access to GSH, Fe(II) for activating ATS is less available. These aggregates have been suggested to be composed of Fe(III)-PPIX μ-oxo dimers or monomers interacting via π–π interactions.44 These aggregates are distinct from crystalline hemozoin (Hz) 45,46 Capping these aggregates presumably assists inhibition of Hz formation.47 Previous NMR studies30 of CQ binding to Fe(III)-PPIX in aqueous medium used large excess of CQ (≥ 10 fold) relative to Fe(III)-PPIX, which is sufficient to inhibit Fe(III)-PPIX aggregation. However, without such a large excess of CQ, Fe(III)-PPIX aggregates in aqueous medium even at neutral pH.44 In the presence of CQ and other quinolinal antimalarial drugs that are positively charged at neutral pH, aggregation can be enhanced by balancing the negative charge accumulating within the aggregate. This then explains why a substantial excess of drug is required to prevent Fe(III)-PPIX self-association.

Earlier results suggested that the diastereomer of quinine, 9-epi-quinine (EPI), is unique with regard to Fe(III)-PPIX interaction relative to other quinoline drugs. Specifically, while CQ and QN have been shown to bind to the face of the Fe(III)-PPIX porphyrin plane,32 ring-current shifts measured with a diamagnetic analog Zn(II)-PPIX suggested that EPI binds with a face-to-edge π- π interaction in which the quinoline ring is oriented along the side of the porphyrin.50 Edge binding would be predicted to stabilize but not cap Fe(III)-PPIX aggregates, and would not therefore prevent reduction of Fe(III) to Fe(II). In this work, we find that initially, the rate of ATS activation is limited by the amount of exposed heme. However, as ATS reacts and alkylates Fe(III)-PPIX, and frees a porphyrin unit from the aggregate, a new heme unit is now exposed. Thus, compared to the reaction when EPI is absent, the reaction is closer to being zero order as the concentration of heme available for reduction by GSH is nearly constant throughout the reaction. Correspondingly, (Fig. 8), the apparent initial rate of ATS activation is slower in the presence of EPI since with aggregation, the effective amount of Fe(III)-PPIX available for reduction of GSH is reduced.

Fe(III)-PPIX - ATS and Fe(III)-PPIX - ATM adducts were analysed by ESI - MS and showed m/z of 941, 854, respectively (Table 1), as previously reported.23 When ATM is reacted in the presence of EPI, ESI - MS reveals the presence of a single Fe(III)-PPIX molecule being alkylated twice (Table 1, Figs. 10 and 11). A clear peak at m/z = 1092 (Fig. 11, Table 1) can be assigned to a 1:2 FPIX:ATM adduct with both ATM moieties deacetylated (Fig. 10). In the absence of EPI, only the monoalkylated adduct, m/z = 854, is observed (Table 1). Double alkylation of FPIX in the presence of EPI suggests that an alkylated Fe(III)-PPIX with a regenerated Fe(III) center can be reduced and then activate another ATM molecule. This is more likely to happen if the FPIX aggregate is stable (as when EPI is binding edge - to - face to FPIX units in the aggregate) and a singly-alkylated heme at the end of an aggregate remains the only Fe(III) accessible to GSH.

Table 1.

Observed m/z peaks in electrospray ionization (ESI) mass spectrometry

Sample Observed m/z peaks
FPIX:GSH:ATM 854
FPIX:GSH:ATM:EPI 854, 1092
FPIX:GSH:ATS 941
FPIX:GSH:ATS:EPI 941, 1264
Open in a new tab

Figure 10.

Figure 10.

Open in a new tab

Proposed structures for FPIX - ATM adducts formed − (854) and + (1092) the presence of EPI (see text).

Figure 11.

Figure 11.

Open in a new tab

ESI - MS peaks observed for FPIX - ATM adduct formation in the presence of EPI (note peaks at 854 and 1092 corresponding to left, right Fig. 10)

Antagonism between a quinoline-based drug and artemisinin and its derivatives have long been observed in drug assays on both chloroquine sensitive and resistant strains of Plasmodium falciparum.51-54 Our current work provides a molecular explanation, quinoline-based drugs can prevent the reduction of Fe(III) in Fe(III)-PPIX. Without Fe(II)-FPIX, artemisinin or any of its derivatives cannot be activated.

Conclusions

In sum we find that Fe(II)-PPIX reacts with ATS to form Fe(III)-PPIX - ATS alkylated species and that the rate is limited by the reduction of Fe(III) to Fe(II) in Fe(III)-PPIX by GSH. This reaction is found to be dependent on the behavior of Fe(III)-PPIX in aqueous solution. Aggregation and capping of the aggregates by other antimalarial drugs can slow or prevent activation of ATS by Fe(III)-PPIX. However, in the special case of EPI, which binds very differently to Fe(III)-PPIX, aggregates are not capped but are instead stabilized by binding of EPI molecules on the sides of the aggregate. The Fe(III) center at the ends of these aggregates remains accessible for reduction by GSH.

Acknowledgement

This work was supported by National Institutes of Health Grant AI111962 to P.D.R. and the Department of Chemistry of Georgetown University.

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

References

  • (1).Tu Y The discovery of artemisinin (qinghaosu) and gifts from Chinese medicine. Nature Medicine 2011, 17, 1217–1220. [DOI] [PubMed] [Google Scholar]
  • (2).Ashley EA; Dhorda M; Fairhurst RM; Amaratunga C; Lim P; Suon S; Sreng S; Anderson JM; Mao S; Sam B; Sopha C; Chuor CM; Nguon C; Sovannaroth S; Pukrittayakamee S; Jittamala P; Chotivanich K; Chutasmit K; Suchatsoonthorn C; Runcharoen R; Hien TT; Thuy-Nhien NT; Thanh NV; Phu NH; Htut Y; Han K-T; Aye KH; Mokuolu OA; Olaosebikan RR; Folaranmi OO; Mayxay M; Khanthavong M; Hongvanthong B; Newton PN; Onyamboko MA; Fanello CI; Tshefu AK; Mishra N ; Valecha N; Phyo AP; Nosten F; Yi P; Tripura R; Borrmann S; Bashraheil M; Peshu J; Faiz MA; Ghose A; Hossain MA; Samad R; Rahman MR; Hasan MM; Islam A; Miotto O; Amato R; Maclnnis B; Stalker J; Kwiatkowski DP; Bozdech Z; Jeeyapant A; Cheah PY; Sakulthaew T; Chalk J; Intharabut B; Silamut K; Lee SJ; Vihokhern B; Kunasol C; Imwong M; Taming J; Taylor WJ; Yeung S; Woodrow CJ; Flegg JA; Das D; Smith J; Venkatesan M; Plowe CV; Stepniewska K; Guerin PJ; Dondorp AM; Day NP; White NJ Spread of artemisinin resistance in Plasmodium falciparum malaria. N. Engl. J. Med. 2014, 371, 411–423. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (3).Amaratunga C; Lim P; Suon S; Sreng S; Mao S; Sopha C; Sam B; Dek D; Try V; Amato R; Blessborn D; Song L; Tullo GS, Fay MP; Anderson JM; Tarning J; Fairhurst RM Dihydroartemisinin–piperaquine resistance in Plasmodium falciparum malaria in Cambodia: a multisite prospective cohort study. The Lancet Infectious Diseases 2016, 16, 357–365. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (4).Dondorp AM; Nosten F; Yi P; Das D; Phyo AP; Tarning J; Lwin KM; Ariey F Hanpithakpong W; Lee SJ; Ringwald P; Silamut K; Imwong M; Chotivanich K; Lim P; Herdman T; An SS; Yeung S; Singhasivanon P; Day NP; Lindegardh N; Socheat D; White NJ Artemisinin resistance in Plasmodium falciparum malaria. N. Engl. J. Med. 2009, 361, 455–467. Erratum in: N. Engl. J. Med. 361, 1714. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (5).O’Neill PM; Barton VE; Ward SA The molecular mechanism of action of artemisinin—the debate continues. Molecules 2010, 15, 1705–1721. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (6).Asawamahasakda W; Ittarat I; Pu Y; Ziffer H; Meshnick SR Reaction of antimalarial endoperoxides with parasite proteins. Antimicrob. Agents Chemother. 1994, 38, 1854–1858. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (7).Robert A; Benoit-Vical F; Claparols C; Meunier B The antimalarial drug artemisinin alkylates heme in infected mice. Proc. Natl. Acad. Sci. U. S. A. 2005, 103, 13676–13680. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (8).Meunier B; Robert A Heme as trigger and target for trioxane-containing antimalarial drugs. Acc. Chem. Res. 2010, 43, 1444–1451. [DOI] [PubMed] [Google Scholar]
  • (9).Hassett MR; Sternberg AR; Riegel BE; Thomas CJ; Roepe PD Heterologous expression, purification, and functional analysis of plasmodium falciparum phosphatidylinositol 3’-kinase. Biochemistry 2017, 56, 4335–4345. [DOI] [PubMed] [Google Scholar]
  • (10).Herwig JF; Soomro SA Chemical instability determines the biological action of artemisinins. Curr. Med. Chem. 2007, 14, 3243–3259. [DOI] [PubMed] [Google Scholar]
  • (11).Meshnick SR; Thomas A; Ran A; Xy CM; Pan HZ Artemisinin (qinghaosu): the role of intracellular hemin in its mechanism of antimalarial action. Mol. Biochem. Parasitol. 1991, 49, 181–189. [DOI] [PubMed] [Google Scholar]
  • (12).Posner GH; Oh CH Regiospecifically oxygen-18 labeled 1,2,4-trioxane: a simple chemical model system to probe the mechanism(s) for the antimalarial activity of artemisinin (qinghaosu). J. Am. Chem. Soc. 1992, 114, 8328–8329. [Google Scholar]
  • (13).Posner GH; Wang D; Cumming JN; Oh CH; French AN; Bodley AN; Shapiro TA Further evidence supporting the importance of and the restrictions on a carbon-centered radical for high antimalarial activity of 1,2,4-trioxanes like artemisinin. J. Med. Chem. 1995, 38, 2273–2275. [DOI] [PubMed] [Google Scholar]
  • (14).Posner GH; Oh CH; Wang DS; Gerena L; Milhous WK; Meshnick SR; Asawamahasadka W Mechanism-based design, synthesis, and in vitro antimalarial testing of new 4-methylated trioxanes structurally related to artemisinin: the importance of a carbon-centered radical for antimalarial activity. J. Med. Chem. 1994, 37, 1256–1258. [DOI] [PubMed] [Google Scholar]
  • (15).Jefford CW; Favarger F; Vicente MGH; Jacquier Y The decomposition of cis-fused cyclopenteno-l,2,4-trioxanes induced by ferrous salts and some oxophilic reagents. Helv. Chim. Acta 1995, 78, 452–458. [Google Scholar]
  • (16).Jefford CW; Vicente MGH; Jacquier Y; Favarger F; Mareda J; Millasson-Schmidt P; Brunner G; Burger U The deoxygenation and isomerization of artemisinin and artemether and their relevance to antimalarial action. Helv. Chim. Acta 1996, 79, 1475–1487. [Google Scholar]
  • (17).Haynes RK; Vonwiller SC The behaviour of qinghaosu (artemisinin) in the presence of heme iron(II) and (III). Tetrahedron Lett. 1996, 37, 253–256. [Google Scholar]
  • (18).Haynes RK; Vonwiller SC The behaviour of qinghaosu (artemisinin) in the presence of nonheme iron(II) and (III). Tetrahedron Lett. 1996, 37, 257–260. [Google Scholar]
  • (19).Haynes RK; Pai HH-O; Voerste A Ring opening of artemisinin (qinghaosu) and dihydroartemisinin and interception of the open hydroperoxides with Formation of N-oxides - a chemical model for antimalarial mode of action. Tetrahedron Lett. 1999, 40, 4715–4718. [Google Scholar]
  • (20).Haynes RK; Chan WC; Lung CM; Uhlemann AC; Eckstein U; Taramelli D; Parapini S; Monti D; Krishna S The Fe2+-mediated decomposition, PfATP6 binding, and antimalarial activities of artemisone and other arteminisins: the unlikehood of C-centered radicals as bioactive intermediates. ChemMedChem 2007, 2, 1480–1497. [DOI] [PubMed] [Google Scholar]
  • (21).Wang I; Zhang C-J; Chia WN; Loh CCY; Li Z; Lee YM; He Y; Yuan L-X; Lim TK; Liu M; Liew CX; Lee YQ; Zhang J; Lu N; Lim CT; Hua Z-C; Liu B; Shen H-M; Tan KSW; Lin Q Haem-activated promiscuous targeting of artemisinin in Plasmodium falciparum. Nat. Commun. 2015, 6, 10111. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (22).Heller LE; Roepe PD Quantification of free ferriprotoporphyrin IX heme and hemozoin for artemisinin sensitive vs delayed clearance phenotype Plasmodium falciparum malarial parasites. Biochemistry 2018. 57, 6927–6934. [DOI] [PubMed] [Google Scholar]
  • (23).Heller LE; Goggins E; Roepe PD Dihydroartemisinin - ferriprotoporphyrin IX adduct abundance in Plasmodium falciparum malarial parasites and relationship to emerging artemisinin resistance. Biochemistry 2018. 57, 6935–6945. [DOI] [PubMed] [Google Scholar]
  • (24).Ismail HM; Barton V; Phanchana M; Charoensutthivarakul S; Wong MHL; Hemingway J; Biagini GA; O’Neill PM; Ward SA Artemisinin activity-based probes identify multiple molecular targets within the asexual stage of the malaria parasites Plasmodium falciparum 3D7. Proc. Natl. Acad. Sci. U. S. A. 2016, 113, 2080–2085. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (25).Ismail HM; Barton VE; Panchana M; Charoensutthivarakul S; Biagini GA; Ward SA; O'Neill PM A Click Chemistry-Based Proteomic Approach Reveals that 1,2,4-Trioxolane and Artemisinin Antimalarials Share a Common Protein Alkylation Profile. Angew. Chem. 2016, 128, 6511–6515. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (26).Ariey F; Witkowski B; Amaratunga C; Beghain J; Langlois A-C; Khim N; Kim S; Duru V; Bouchier C; Ma L; Lim P; Leang R; Duong S; Sreng S; Suon S; Chuor CM; Bout DM; Mé nard S; Rogers WO; Genton B; Fandeur T; Miotto O; Ringwald P; Le Bras J; Berry A; Barale J-C; Fairhurst RM; Benoit-Vical F; Mercereau-Puijalon O; Me nard D A molecular marker of artemisinin-resistant Plasmodium falciparum malaria. Nature 2014, 505, 50–55. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (27).Witkowski B; Amaratunga C; Khim N; Sreng S; Chim P; Kim S; Lim P; Mao S; Sopha C; Sam B; Anderson JM; Duong S; Chuor CM; Taylor WRJ; Suon S; Mercereau-Puijalon O; Fairhurst RM; Menard D Novel phenotypic assays for the detection of artemisinin-resistant plasmodium falciparum malaria in cambodia: in-vitro and ex-vivo drug-response studies. Lancet Infect. Dis. 2013, 13, 1043–1049. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (28).Straimer J; Gnädig NF; Witkowski B; Amaratunga C; Duru V; Ramadani AP; Dacheux M; Khim N; Zhang L; Lam S; Gregory PD; Urnov FD; Mercereau-Puijalon O; Benoit-Vical F; Fairhurst RM; Ménard D; Fidock DA K13-propeller mutations confer artemisinin resistance in Plasmodium falciparum clinical isolates. Science 2015, 347, 428–431. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (29).Dogovski C; Xie SC; Burgio G; Bridgford J; Mok S; McCaw JM; Chotivanich K; Kenny S; Gnädig N; Straimer J; Bozdech Z; Fidock DA; Simpson JA; Dondorp AM; Foote S; Klonis N; Tilley L Targeting the Cell Stress Response of Plasmodium falciparum to Overcome Artemisinin Resistance. PLoS Biol. 2015, 13, el002132. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (30).Leed A; DuBay K; Ursos LMB; Sears D; de Dios AC; Roepe PD Solution structures of antimalarial drug–heme complexes. Biochemistry 2002, 41, 10245–10255 [DOI] [PubMed] [Google Scholar]
  • (31).Casabianca LB; An D; Natarajan JK; Alumasa JN; Roepe PD; Wolf C; de Dios AC Quinine and chloroquine differentially perturb heme monomer–dimer equilibrium. Inorg. Chem. 2008, 47, 6077–6081. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (32).Gorka AP; de Dios AC; Roepe PD Quinoline drug–heme interactions and implications for antimalarial cytostatic versus cytocidal activities. J. Med. Chem. 2013, 56, 5231–5246. [DOI] [PubMed] [Google Scholar]
  • (33).de Dios AC; Casabianca LB; Kosar A; Roepe PD Structure of the Amodiaquine-FPIX μ oxo dimer solution complex at atomic resolution. Inorg. Chem. 2004, 43, 8087–8084. [DOI] [PubMed] [Google Scholar]
  • (34).Eastman RT; Fidock DA Artemisinin-Based Combination Therapies: A Vital Tool in Efforts to Eliminate Malaria. Nature Reviews. Microbiology 2009, 864–874. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (35).Henrich PP; O'Brien C; Säenz FE; Cremers S; Kyle DE; Fidock DA Evidence for pyronaridine as a highly effective partner drug for treatment of artemisinin-resistant malaria in a rodent model. Antimicrob Agents Chemother. 2014, 58, 183–195. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (36).Atamna EL; Ginsburg H The malaria parasite supplies glutathione to its host cell--investigation of glutathione transport and metabolism in human erythrocytes infected with Plasmodium falciparum. Eur. J. Biochem. 1997, 250, 670–679. [DOI] [PubMed] [Google Scholar]
  • (37).Robert A; Coppel Y; Meunier B Alkylation of heme by the antimalarial drug artemisinin. Chem. Commun. 2002, 414–415. [DOI] [PubMed] [Google Scholar]
  • (38).Monti D; Vodopivec B; Basilico N; Olliaro P; Taramelli D A novel endogenous antimalarial: Fe(II)-protoporphyrin IXα (heme) inhibits hematin polymerization to β-hematin (malaria pigment) and kills malaria parasites. Biochemistry 1999, 38, 8858–8863. [DOI] [PubMed] [Google Scholar]
  • (39).Haynes RK Reply to comments on “highly antimalaria-active artemisinin derivatives: biological activity does not correlate with chemical reactivity”. Angew. Chem. IntT Ed. 2005, 44, 2064–2065. [DOI] [PubMed] [Google Scholar]
  • (40).Robert A; Coppel Y; Meunier B NMR characterization of covalent adducts obtained by alkylation of heme with the antimalarial drug artemisinin. Inorg. Chim. Acta, 2002, 339, 488496. [Google Scholar]
  • (41).Wu W-M; Chen Y-L; Zhai Z, Xiao S-H; Wu Y-L Study on the nechanism of action of artemether against schistosomes: The identification of cysteine adducts of both carbon-centered free radicals derived from artemether. Bioorg. Med. Chem. Lett. 2003, 13, 1645–1647. [DOI] [PubMed] [Google Scholar]
  • (42).Loup C; Leliévre J; Benoit-Vical F; Meunier B Trioxaquines and heme-artemisinin adducts inhibit the in vitro formation of hemozoin better than chloroquine. Antimicrob. Agents Chemother. 2007, 51, 3768–3770. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (43).Ursos LMB; DuBay KF; Roepe PD Antimalarial drugs influence the pH dependent solubility of heme via apparent nucleation phenomenon. Mol. Biochem. Parasitol. 2001, 112, 1117. [DOI] [PubMed] [Google Scholar]
  • (44).de Villiers KA; Kaschula CH; Egan TJ; Marques HM Speciation and structure of ferriprotoporphyrin IX in aqueous solution: spectroscopic and diffusion measurements demonstrate dimerization, but not μ-oxo dimer formation. J. Inorg. Biochem. 2007, 12, 101–117. [DOI] [PubMed] [Google Scholar]
  • (45).Pagola S; Stephenes PW; Bohle DS; Kosar A; Madsen SK The structure of malaria pigment B-haematin. Nature Lett. 2000, 404, 307–310. [DOI] [PubMed] [Google Scholar]
  • (46).Hempelmann E; Egan TJ Pigment biocrystallization in Plasmodium falciparum. Trends in Parasitol. 2002, 18, 11. [DOI] [PubMed] [Google Scholar]
  • (47).Ziegler J; Linck R; Wright DW Heme aggregation inhibitors: Antimalarial drugs targeting an essential biomineralization process. Curr. Med. Chem. 2001, 8, 171–189. [DOI] [PubMed] [Google Scholar]
  • (48).Gorka AP; Alumasa JN; Sherlach KS; Jacobs LM; Nickley KB; Brower JP; de Dios AC; Roepe PD Cytostatic versus cytocidal activities of chloroquine analogues and inhibition of hemozoin crystal growth. Antimicrob. Agents Chemother. 2013, 57, 356–364. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (49).Cabrera M; Paguio MF; Xie C; Roepe PD Reduced digestive vacuolar accumulation of chloroquine is not linked to resistance to chloroquine toxicity. Biochemistry 2009, 47, 11152–11154. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (50).Gorka AP; Sherlach KS; de Dios AC; Roepe PD Relative to quinine and quinidine, their 9-epimers exhibit decreased cytostatic activity and altered heme binding but similar cytocidal activity versus Plasmodium falciparum. Antimicrob. Agents Chemother. 2013, 57, 365–374. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (51).Chawira AN; Warhurst DC. The effect of artemisinin combined with standard antimalarials against chloroquine-sensitive and chloroquine-resistant strains of Plasmodium falciparum in vitro. J. Trop. Med. Hyg. 1987, 90:1–8. [PubMed] [Google Scholar]
  • (52).Davis TM; Hamzah EJ; Ilett KF; Karunajeewa HA; Reeder JC; Batty KT; Hackett S; Barrett PH In vitro interactions between piperaquine, dihydroartemisinin, and other conventional and novel antimalarial drugs. Antimicrob. Agents Chemother. 2006, 50:2883–2885. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (53).Fivelman QL; Walden JC; Smith PJ; Folb PI; Barnes KI The effect of artesunate combined with standard antimalarials against chloroquine-sensitive and chloroquine-resistant strains of Plasmodium falciparum in vitro. Trans. R. Soc. Trop. Med. Hyg. 1999, 93:429–432. [DOI] [PubMed] [Google Scholar]
  • (54).Fivelman QL; Adagu IS; Warhurst DC Effects of piperaquine, chloroquine, and amodiaquine on drug uptake and of these in combination with dihydroartemisinin against drug-sensitive and –resistant Plasmodium falciparum strains. Antimicrob. Agents Chemother, 2007, 51, 2265–2267. [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES