Safety assessment of peroxide antimalarials: clinical and chemical perspectives

Reactive Oxygen Species as the Brainbox in Malaria Treatment

Several measures are in place to combat the worldwide spread of malaria, especially in regions of high endemicity. In part, most common antimalarials, such as quinolines and artemisinin and its derivatives, deploy an ROS-mediated approach to kill malaria parasites. Although some antimalarials may share similar targets and mechanisms of action, varying levels of reactive oxygen species (ROS) generation may account for their varying pharmacological activities. Regardless of the numerous approaches employed currently and in development to treat malaria, concerningly, there has been increasing development of resistance by Plasmodium falciparum, which can be connected to the ability of the parasites to manage the oxidative stress from ROS produced under steady or treatment states. ROS generation has remained the mainstay in enforcing the antiparasitic activity of most conventional antimalarials. However, a combination of conventional drugs with ROS-generating ability and newer drugs that exploit vital metabolic pathways, such antioxidant machinery, could be the way forward in effective malaria control.

Safety assessment of MEFAS: an innovative hybrid salt of mefloquine and artesunate for malaria treatment

Malaria is a global public health problem that causes approximately 445 000 deaths annually worldwide, especially in underdeveloped countries. Because of the high prevalence and mortality of the disease, new and less toxic therapeutic agents need to be developed, such as MEFAS, a low-cost hybrid salt that consists of artesunate and mefloquine. However, the efficacy of MEFAS has been systematically demonstrated, its safety requires further investigation. This study investigated the acute toxicity of MEFAS and its precursors, artesunate, and mefloquine. A total of 42 female Swiss mice were divided into seven groups (n = 6/group) that were treated orally by gavage with vehicle (filtered water, negative control), MEFAS (50, 500, and 1000 mg/kg), and 1:1 concentrations of artesunate + mefloquine (50, 500, and 1000 mg/kg). Clinical signs of toxicity were observed for 14 d after treatment. On day 15, the animals were weighed, deeply anesthetized with isoflurane, and euthanized for subsequent collection of the liver, spleen, and kidneys. The relative organ weights were determined, followed by histopathological analysis. Artesunate + mefloquine produced toxic effects compared with the negative control group, reflected by changes in clinical signs, relative organ weights, and histopathological alterations. In MEFAS-treated animals, no changes were observed compared with the negative control group. These findings demonstrate that MEFAS is safer than artesunate + mefloquine after acute administration in mice.

Antiparasitic treatments in pregnant women: Update and recommendations

Parasitoses are a major cause of morbidity and mortality worldwide, especially in resource-poor countries where the prevalence of such infections is very high. Their consequences for pregnant women are a public health issue. It is very challenging to successfully control parasitic infections with the dedicated drugs, while protecting the fetus from the harmful effects of these medications. However, in both temperate and tropical regions, true antiparasitic innovations are rare, and the therapeutic armamentarium remains limited. Scientific data is incomplete as only a few clinical studies have included pregnant women so far. Therefore, physicians have to learn how to thoroughly handle the antiparasitic molecules available. They also need to know the embryo- and fetotoxic effects of each of them. Medical practices must be adapted to the trimester of pregnancy.

Novel series of 1,2,4-trioxane derivatives as antimalarial agents

Among three series of 1,2,4-trioxane derivatives, five compounds showed good in vitro antimalarial activity, three compounds of which exhibited better activity against P. falciparum resistant (RKL9) strain than the sensitive (3D7) one. Two best compounds were one from aryl series and the other from heteroaryl series with IC₅₀ values of 1.24 µM and 1.24 µM and 1.06 µM and 1.17 µM, against sensitive and resistant strains, respectively. Further, trioxane derivatives exhibited good binding affinity for the P. falciparum cysteine protease falcipain 2 receptor (PDB id: 3BPF) with well defined drug-like and pharmacokinetic properties based on Lipinski's rule of five with additional physicochemical and ADMET parameters. In view of having antimalarial potential, 1,2,4-trioxane derivative(s) reported herein may be useful as novel antimalarial lead(s) in the discovery and development of future antimalarial drug candidates as P. falciparum falcipain 2 inhibitors against resistant malaria.

Toxicokinetics and Hydrolysis of Artelinate and Artesunate in Malaria-Infected Rats

Comparative toxicokinetic (TK) and hydrolysis studies of intravenously administered two new antimalarial agents, artelinate (AL) and artesunate (AS), were performed in malaria-infected rats using three daily equimolar doses (96 μmoles/kg). The TK evaluation was related to select one drug for severe malaria treatment in U.S. Army. Drug concentration of AS with daily dose of 36.7 mg/kg was one-third less on day 3 than on day 1, which resembled its active metabolite, dihydroartemisinin (DHA), suggesting an autoinduction of hepatic drug-metabolizing enzymes for AS. The results were similar to other artemisinin drugs, but not for AL. TK parameters of AL were very comparable from day 1 to day 3 at same AS molecular dose at 40.6 mg/kg. AS is the prodrug of DHA with the DHA/AS ratio of 5.26 compared to the ratio of 0.01 for DHA/AL. Other TK parameters revealed that the total AUC1–3 days (84.4 μg · h ml−1) of AL was fivefold higher than that of AS (15.7 mu;g h ml−1 of AS plus DHA). The elimination half-life of AL (7.1 h) was much longer than that of AS (0.36 h) or DHA (0.72 h). The remarkable alteration of the TK shape of AL may be caused by poor conversion rates to DHA and an enterohepatic circulation, which is confirmed by the present TK and tissue distribution studies. Compared to AS, higher drug exposure levels and longer exposure time of AL in the rat blood may be the cause of its increased toxicity.

Endoperoxide antimalarials: development, structural diversity and pharmacodynamic aspects with reference to 1,2,4-trioxane-based structural scaffold

Malaria disease continues to be a major health problem worldwide due to the emergence of multidrug-resistant strains of Plasmodium falciparum. In recent days, artemisinin (ART)-based drugs and combination therapies remain the drugs of choice for resistant P. falciparum malaria. However, resistance to ART-based drugs has begun to appear in some parts of the world. Endoperoxide compounds (natural/semisynthetic/synthetic) representing a huge number of antimalarial agents possess a wide structural diversity with a desired antimalarial effectiveness against resistant P. falciparum malaria. The 1,2,4-trioxane ring system lacking the lactone ring that constitutes the most important endoperoxide structural scaffold is believed to be the key pharmacophoric moiety and is primarily responsible for the pharmacodynamic potential of endoperoxide-based antimalarials. Due to this reason, research into endoperoxide, particularly 1,2,4-trioxane-, 1,2,4-trioxolane- and 1,2,4,5-teraoxane-based scaffolds, has gained significant interest in recent years for developing antimalarial drugs against resistant malaria. In this paper, a comprehensive effort has been made to review the development of endoperoxide antimalarials from traditional antimalarial leads (natural/semisynthetic) and structural diversity of endoperoxide molecules derived from 1,2,4-trioxane-, 1,2,4-trioxolane- and 1,2,4,5-teraoxane-based structural scaffolds, including their chimeric (hybrid) molecules, which are newer and potent antimalarial agents.

General Pharmacology of Artesunate, a Commonly used Antimalarial Drug:Effects on Central Nervous, Cardiovascular, and Respiratory System

Artesunate, a semi-synthetic derivative of artemisinin, is used primarily as a treatment for malaria. Its effects on the central nervous system, general behavior, and cardiovascular, respiratory, and other organ systems were studied using mice, rats, guinea pigs, and dogs. Artesunate was administered orally to mice at doses of 125, 250, and 500 mg/kg and to rats and guinea pigs at 100, 200, and 400 mg/kg. In dogs, test drugs were administered orally in gelatin capsules at doses of 50, 100, and 150 mg/kg. Artesunate induced insignificant changes in general pharmacological studies, including general behavior, motor coordination, body temperature, analgesia, convulsion modulation, blood pressure, heart rate (HR) , and electrocardiogram (ECG) in dogs in vivo; respiration in guinea pigs; and gut motility or direct effects on isolated guinea pig ileum, contractile responses, and renal function. On the other hand, artesunate decreased the HR and coronary flow rate (CFR) in the rat in vitro; however, the extent of the changes was small and they were not confirmed in in vivo studies in the dog. Artesunate increased hexobarbital-induced sleeping time in a dose-related manner. Artesunate induced dose-related decreases in the volume of gastric secretions and the total acidity of gastric contents, and induced increases in pH at a dose of 400 mg/kg. However, all of these changes were observed at doses much greater than clinical therapeutic doses (2.4 mg/kg in humans, when used as an anti-malarial) . Thus, it can be concluded that artesunate is safe at clinical therapeutic doses.

Amplification activation loop between caspase-8 and -9 dominates artemisinin-induced apoptosis of ASTC-a-1 cells

Although caspases have been demonstrated to be involved in artemisinin (ARTE)-induced apoptosis, their exact functions are not well understood. The aim of this report is to explore the roles of caspase-8, -9 and -3 during ARTE-induced apoptosis in human lung adenocarcinoma (ASTC-a-1) cells. ARTE treatment induces a rapid generation of reactive oxygen species (ROS), and ROS-dependent apoptosis as well as the activation of caspase-8, -9 and -3 via time- and dose-dependent fashion. Of upmost importance, inhibition of caspase-8 or -9, but not caspase-3, almost completely blocks the ARTE-induced not only activation of the caspase-8, -9 and -3 but also apoptosis. In addition, the apoptotic process triggered by ARTE does not involve the Bid cleavage, tBid translocation, significant loss of mitochondrial membrane potential and cytochrome c release from mitochondria. Moreover, silencing Bax/Bak does not prevent the ATRE-induced cell death as well as the activation of caspase-8, -9 and -3. Collectively, our data firstly demonstrate that ARTE triggers a ROS-mediated positive feedback amplification activation loop between caspase-8 and -9 independent of mitochondria, which dominantly mediated the ARTE-induced apoptosis via a caspase-3-independent apoptotic pathway in ASTC-a-1 cells. Our findings imply a potential to develop new derivatives from artemisinin to effectively initiate the amplification activation loop of caspases.

Pharmacokinetic profiles of artesunate following multiple intravenous doses of 2, 4, and 8 mg/kg in healthy volunteers: phase 1b study

Background

Severe malaria results in over a million deaths every year, most of them in children aged less than five years and living in sub-Saharan Africa. Injectable artesunate (AS) was recommended as initial treatment for severe malaria by WHO in 2006. The Walter Reed Army Institute of Research (WRAIR) has been developing a novel good manufacturing practice (GMP) injection of AS, which was approved by the US FDA for investigational drug use and distribution by the CDC.

Methods

Tolerability and pharmacokinetics of current GMP intravenous AS, as an anti-malarial agent, were evaluated after ascending multiple doses of 2, 4, and 8 mg/kg daily for three days with 2-minute infusion in 24 healthy subjects (divided into three groups) in the Phase 1 clinical trial study.

Results

Results showed that there were no dose-dependent increases in any adverse events. Drug concentrations showed no accumulation and no decline of the drug during the three days of treatment. After intravenous injection, parent drug rapidly declined and was converted to dihydroartemisinin (DHA) with overall mean elimination half-lives ranging 0.15-0.23 hr for AS and 1.23-1.63 hr for DHA, but the peak concentration (C_max) of AS was much higher than that of DHA with a range of 3.08-3.78-folds. In addition, the AUC and C_max values of AS and DHA were increased proportionally to the AS climbing multiple doses.

Discussion

The safety of injectable AS, even at the highest dose of 8 mg/kg increases the probability of therapeutic success of the drug even in patients with large variability of parasitaemia.

Semisynthetic Artemisinin and Synthetic Peroxide Antimalarials

Since the discovery of the endoperoxide sesquiterpene lactone artemisinin, numerous second-generation semisynthetic artemisinins and synthetic peroxides have been prepared and tested for their antimalarial properties. Using a case-study approach, we describe the discovery of the investigational semisynthetic artemisinins artelinic acid (8) and artemisone (9), and the structurally diverse synthetic peroxides arteflene (10), fenozan B07 (11), arterolane (12), PA1103/SAR116242 (13), and RKA182 (14).

The pharmacokinetic evaluation of artemisinin drugs for the treatment of malaria in paediatric populations

INTRODUCTION

The use of artemisinin combination therapies to treat uncomplicated malaria is growing and, therefore, so is the number of children exposed to these agents. As a result, there is a huge drive to develop paediatric formulations. However, relatively limited data exist regarding the pharmacokinetic properties of these drugs in this vulnerable population.

AREAS COVERED

The article reviews the pharmacokinetic data for artemisinin drugs used for the treatment of malaria in paediatric populations. The authors discuss how developmental and environmental factors can produce significant variation in the pharmacokinetic properties of artemisinin drugs. The authors also discuss how this variation may lead to suboptimal therapeutic drug concentrations with implications on efficacy, safety and the development of parasite resistance to these drugs.

EXPERT OPINION

There is currently a lack of published studies on the pharmacokinetics of artemisinin drugs in children and this subject is complicated by several interdependent variables. Therefore, the construction of a systems-based model of this subject should be a priority area in order to identify gaps in current knowledge to ensure their continued effective and safe use.

The role of heme and the mitochondrion in the chemical and molecular mechanisms of mammalian cell death induced by the artemisinin antimalarials

The artemisinin compounds are the frontline drugs for the treatment of drug-resistant malaria. They are selectively cytotoxic to mammalian cancer cell lines and have been implicated as neurotoxic and embryotoxic in animal studies. The endoperoxide functional group is both the pharmacophore and toxicophore, but the proposed chemical mechanisms and targets of cytotoxicity remain unclear. In this study we have used cell models and quantitative drug metabolite analysis to define the role of the mitochondrion and cellular heme in the chemical and molecular mechanisms of cell death induced by artemisinin compounds. HeLa ρ(0) cells, which are devoid of a functioning electron transport chain, were used to demonstrate that actively respiring mitochondria play an essential role in endoperoxide-induced cytotoxicity (artesunate IC(50) values, 48 h: HeLa cells, 6 ± 3 μM; and HeLa ρ(0) cells, 34 ± 5 μM) via the generation of reactive oxygen species and the induction of mitochondrial dysfunction and apoptosis but do not have any role in the reductive activation of the endoperoxide to cytotoxic carbon-centered radicals. However, using chemical modulators of heme synthesis (succinylacetone and protoporphyrin IX) and cellular iron content (holotransferrin), we have demonstrated definitively that free or protein-bound heme is responsible for intracellular activation of the endoperoxide group and that this is the chemical basis of cytotoxicity (IC(50) value and biomarker of bioactivation levels, respectively: 10β-(p-fluorophenoxy)dihydroartemisinin alone, 0.36 ± 0.20 μM and 11 ± 5%; and with succinylacetone, >100 μM and 2 ± 5%).

Assessment of safety of the major antimalarial drugs

Antimalarial drugs remain the major intervention tool for the global malaria control efforts that save millions of lives. Nonetheless, emergence and spread of Plasmodium parasites resistant against chloroquine and other major antimalarial drugs has brought the urgency to develop a new generation of safe and effective drugs against malaria. In this article, the safety data for major antimalarial drugs is reviewed. Although an ample amount of clinical data regarding the safety and tolerability of several of these drugs in older children and adults is available, more critical safety and tolerability studies in pregnant women and young children is desirable. To offset the partial loss in efficacy due to drug resistance in malaria parasites acquired against specific drugs, treatment regimens often rely upon the combination of two or more drugs. However, combination therapy requires additional safety, toxicity and tolerability studies in all population groups where these drugs are administered. A uniform standard in assessing the safety and tolerability of antimalarial drugs will be useful in the formulation and implementation of malaria treatment policies that are based on the drug effectiveness, safety and tolerability.

Evidence for the involvement of carbon-centered radicals in the induction of apoptotic cell death by artemisinin compounds

Artemisinin and its derivatives are currently recommended as first-line antimalarials in regions where Plasmodium falciparum is resistant to traditional drugs. The cytotoxic activity of these endoperoxides toward rapidly dividing human carcinoma cells and cell lines has been reported, and it is hypothesized that activation of the endoperoxide bridge by an iron(II) species, to form C-centered radicals, is essential for cytotoxicity. The studies described here have utilized artemisinin derivatives, dihydroartemisinin, 10beta-(p-bromophenoxy)dihydroartemisinin, and 10beta-(p-fluorophenoxy)dihydroartemisinin, to determine the chemistry of endoperoxide bridge activation to reactive intermediates responsible for initiating cell death and to elucidate the molecular mechanism of cell death. These studies have demonstrated the selective cytotoxic activity of the endoperoxides toward leukemia cell lines (HL-60 and Jurkat) over quiescent peripheral blood mononuclear cells. Deoxy-10beta-(p-fluorophenoxy)dihydroartemisinin, which lacks the endoperoxide bridge, was 50- and 130-fold less active in HL-60 and Jurkat cells, respectively, confirming the importance of this functional group for cytotoxicity. We have shown that chemical activation is responsible for cytotoxicity by using liquid chromatography-mass spectrometry analysis to monitor endoperoxide activation by measurement of a stable rearrangement product of endoperoxide-derived radicals, which was formed in sensitive HL-60 cells but not in insensitive peripheral blood mononuclear cells. In HL-60 cells the endoperoxides induce caspase-dependent apoptotic cell death characterized by concentration- and time-dependent mitochondrial membrane depolarization, activation of caspases-3 and -7, sub-G(0)/G(1) DNA formation, and attenuation by benzyloxycarbonyl-VAD-fluoromethyl ketone, a caspase inhibitor. Overall, these results indicate that endoperoxide-induced cell death is a consequence of activation of the endoperoxide bridge to radical species, which triggers caspase-dependent apoptosis.

In vivo and mechanistic evidence of nuclear receptor CAR induction by artemisinin

Artemisinin (a sesquiterpene lactone endoperoxide) has become important in multi-drug treatment of malaria. There is evidence that artemisinin induces drug metabolism which could result in drug-drug interactions. The objective of this study was to characterize the inductive properties of artemisinin on drug-metabolizing cytochrome P450 (CYP450) enzymes. The possibility of artemisinin to induce CYP450 was studied in artemisinin-treated (orally for four days) and vehicle-treated rats using reverse transcriptase polymerase chain reaction (RT-PCR). The effect on enzymatic activities in mouse microsomes from multiple artemisinin administration (intraperitonally) to mice were also studied as well as the effect on the expression in mouse primary hepatocytes and HEK293 cells. Increased CYP2B1 mRNA levels in rats could be seen after artemisinin treatment as well as a weak but reproducible increase in the intensity of CYP1A2. Administration of artemisinin to mice up-regulated hepatic CYP2B10-dependent, and to a lesser extent, CYP2A5-dependent enzyme activities. In primary hepatocyte culture, artemisinin significantly increased the CYP2B10 mRNA levels whereas the CYP2A5 mRNA levels were increased to a lesser extent. No significant changes were seen in the levels of other CYP enzymes. Artemisinin was an activator of constitutive androstane receptor (CAR) but not pregnane X receptor (PXR) in HEK293 cells. The results demonstrate that the drug exerts its effects on drug metabolism via the CAR receptor that results in up-regulation of genes such as the Cyp2b. The weaker up-regulation of CYP2A5 might also be CAR-dependent or alternatively, a consequence of artemisinin toxicity. The results of this study are of importance when predicting potential drug-drug interactions in multi-drug therapies with artemisinin.

Hepatocellular bioactivation and cytotoxicity of the synthetic endoperoxide antimalarial arteflene

Arteflene is a synthetic endoperoxide antimalarial. Its peroxide bridge undergoes iron(II)-mediated reduction in vitro which yields a carbon-centered cyclohexyl radical and a mixture of cis- and trans-alpha,beta-unsaturated ketones (enones). The enones are biliary metabolites in rats and therefore surrogate markers of bioactivation. Arteflene is reported to be more cytotoxic to primary rat hepatocytes than some non-endoperoxide antimalarials. Hepatic metabolism of arteflene was investigated in recirculating isolated perfused rat livers, and the drug's metabolism and cytotoxicity were compared using hepatocytes from male rats. Both preparations metabolized [(14)C]arteflene to cis- and trans-[(14)C]enone, 8-hydroxyarteflene glucuronide and an unassigned isomeric glucuronide. During a 2 h liver perfusion, the cis- and trans-enones recovered in bile represented 8.1 +/- 3.4 and 11.3 +/- 4.6% (mean +/- S.D., N=6), respectively, of the [(14)C]arteflene (52 microM) added to the perfusate. After a 3 h incubation of [(14)C]arteflene (10 microM) with hepatocytes in suspension, the cis- and trans-enones comprised, respectively, 14.8 +/- 7.1 and 2.1 +/- 1.0% (N = 4) of the recovered radioactivity; the corresponding data for cultured hepatocytes being 18.6 +/- 6.9 and 3.3 +/- 2.2%. Arteflene was significantly (P < 0.05) toxic to isolated hepatocytes with reference to extramitochondrial reductase activity (tetrazolium reduction) but not enzyme leakage when the cells were exposed to drug concentrations > or =50 microM for 24 h. Cellular glutathione was depleted under these conditions. Therefore arteflene was acutely cytotoxic, though only at relatively high concentrations, when it was metabolized via a pathway which generates carbon-centered radicals.

Artemisinin derivatives: toxic for laboratory animals, safe for humans?

A discrepancy seems to prevail with regard to the toxicity and safety of the artemisinin family of antimalarials. While these compounds have been found to be virtually void of any serious side effects in humans, their neurotoxicity in animal models has raised concerns about their use. In this paper, we present selected examples of both pre-clinical and clinical studies dealing with adverse effects of artemisinin drugs. We suggest that the prolonged presence of artemisinins upon slow release from oil-based intramuscular formulations is the main cause of the observed toxicity in laboratory animals. In contrast, oral intake of these compounds, which is by far the most common formulation used for treatment of malaria patients, results in rapid clearance of these drugs and is thus unlikely to cause any toxicity in human subjects. Another plausible factor may be the relatively high doses of artemisinin compounds used in animal studies. In conclusion, the observation of the toxicity of artemisinin compounds in animals, but not in humans, is most likely due to different pharmacokinetic profiles after different routes of administrations.

Metabolism of fluorine-containing drugs

This article reviews current knowledge of the metabolism of drugs that contain fluorine. The strategic value of fluorine substitution in drug design is discussed in terms of chemical structure and basic concepts in drug metabolism and drug toxicity.

Synthesis, antimalarial activity, biomimetic iron(II) chemistry, and in vivo metabolism of novel, potent C-10-phenoxy derivatives of dihydroartemisinin

The combination of TMSOTf and AgClO(4) promotes the efficient C-10-phenoxylation of dihydroartemisinin (3) in good chemical yield and excellent stereoselectivity. All of the new phenoxy derivatives have potent in vitro antimalarial activity. On the basis of the excellent yield and stereoselectivity obtained for the p-trifluoromethyl derivative 7b, this compound and the parent phenyl-substituted derivative 5b were selected for in vivo biological evaluation against Plasmodium berghei in the mouse model and for metabolism studies in rats. Compound 7b demonstrated excellent in vivo antimalarial potency with an ED(50) of 2.12 mg/kg (cf. artemether = 6 mg/kg) versus P. berghei. Furthermore, from preliminary metabolism studies, this compound was not metabolized to dihydroartemisinin; suggesting it should have a longer half-life and potentially lower toxicity than the first-generation derivatives artemether and arteether. From biomimetic Fe(II)-catalyzed decomposition studies and ESR spectroscopy, the mechanism of action of these new lead antimalarials is proposed to involve the formation of both primary and secondary C-centered cytotoxic radicals which presumably react with vital parasite thiol-containing cellular macromolecules.

Antimalarial sulfide, sulfone, and sulfonamide trioxanes

A series of trioxanes featuring sulfide, sulfone, and sulfonamide substituents in diverse positions has been prepared. Structure-activity relationship (SAR) generalizations highlight two major factors controlling the antimalarial potency of these new chemical entities: (1) the proximity of the sulfur-containing substituent to the crucial peroxide bond and (2) the oxidation state of the sulfur-containing substituent. Generally, sulfones are more antimalarially potent than the corresponding sulfides.

Novel, potent, semisynthetic antimalarial carba analogues of the first-generation 1,2,4-trioxane artemether

Ten novel, second-generation, fluorinated ether and ester analogues of the potent first-generation analogues artemether (4a) and arteether (4b) have been designed and synthesized. All of the compounds demonstrate high antimalarial potency in vitro against the chloroquine-sensitive HB3 and -resistant K1 strains of Plasmodium falciparum. The most potent derivative 8 was 15 times more potent than artemisinin (2) against the HB3 strain of P. falciparum. In vivo, versus Plasmodium berghei in the mouse, selected derivatives were generally less potent than dihydroartemisinin with ED(50) values of between 5 and 8 mg/kg. On the basis of the products obtained from the in vitro biomimetic Fe(II)-mediated decomposition of 8, the radical mediator of biological activity of this series may be different from that of the parent drug, artemisinin (2).