Artemisone - A highly active antimalarial drug of the artemisinin class

Article abstract of DOI:10.1002/anie.200503071

Artemisinin - the next generation: Efficacies of artemisone against the malaria parasite are substantially greater than those of the current artemisinin "gold standard", artesunate. Also, in contrast to most current artemisinins it displays low lipophilic

Full text of DOI:10.1002/anie.200503071

Communications
Antimalarial Drugs
The artemisinins are the most potent and rapidly acting
antimalarial drugs. In response to the global threat of
multidrug-resistant malaria, they are now widely used in
combination therapies with drugs that have a longer half
life.[1] The parent, artemisinin (1), is converted by reduction
DOI: 10.1002/anie.200503071
Artemisone—A Highly Active Antimalarial Drug
of the Artemisinin Class**
Richard K. Haynes,* Burkhard Fugmann, Jörg Stetter,
Karl Rieckmann, Hans-Dietrich Heilmann,
Ho-Wai Chan, Man-Ki Cheung, Wai-Lun Lam,
Ho-Ning Wong, Simon L. Croft, Livia Vivas,
Lauren Rattray, Lindsay Stewart, Wallace Peters,
Brian L. Robinson, Michael D. Edstein,
Barbara Kotecka, Dennis E. Kyle,
Bernhard Beckermann, Michael Gerisch,
Martin Radtke, Gabriele Schmuck, Wolfram Steinke,
Ute Wollborn, Karl Schmeer, and Axel Römer
[*] Prof. Dr. R. K. Haynes, Dr. H.-W. Chan, Dr. M.-K. Cheung,
Dr. W.-L. Lam, Dr. H.-N. Wong
Department of Chemistry
Open Laboratory of Chemical Biology
Institute of Molecular Technology for Drug Discovery and Synthesis
The Hong Kong University of Science and Technology
Clear Water Bay, Kowloon, Hong Kong (P.R. China)
Fax: (+852)2358-1594
into dihydroartemisinin (DHA, 2)[2] and then into the hemi-
ester artesunate (ATS, 3; 65% yield from DHA).[3] Arte-
mether (4)[4] and arteether (5)[5] are obtained from DHA,
methanol, and an acid catalyst in about 60% yield after
fractional crystallization to remove the a-epimers. Therefore,
the most accessible derivative is ATS (3). Although artemi-
sinins display variable pharmacokinetics and low bioavaila-
bility,[6] such properties are countered by selective uptake into
parasitized erythrocytes.[7] Their induction of phase I
enzymes, primarily CYP3A4 and CYP2B6, leads to a time-
dependent decrease in efficacy.[8] Additionally, metabolism of
artesunate and the ethers to DHA (2) is facile.[8,9] The
artemisinins, especially DHA, are neurotoxic according to in
vivo assays with animal models[10] and in vitro studies with
neuronal cell cultures.[11–13] Although neurotoxicity is depen-
dent upon the route of administration in animal screens,[14] it
has not yet been definitively established to be a problem in
humans.[15] From a regulatory viewpoint, however, it is a
contentious issue, and the development of new artemisinin
therapies must be conducted with this in mind.
E-mail: haynes@ust.hk
Dr. B. Fugmann
Bayer Innovation GmbH (Germany)
Prof. Dr. J. Stetter
Bayer AG Central Research (Germany)
Prof. Dr. K. Rieckmann, Dr. M. D. Edstein, Dr. B. Kotecka,
Dr. D. E. Kyle[+]
Army Malaria Institute (AMI)
Enoggera, Queensland 4052 (Australia)
Prof. Dr. H.-D. Heilmann, Dr. B. Beckermann, Dr. M. Gerisch,
Dr. M. Radtke, Dr. G. Schmuck, Dr. W. Steinke
Bayer HealthCare (Germany)
Dr. S. L. Croft, Dr. L. Vivas, Dr. L. Rattray, Dr. L. Stewart
London School of Hygiene and Tropical Medicine (UK)
Prof. Dr. W. Peters, B. L. Robinson
Northwick Park Institute for Medical Research, Harrow (UK)
Dr. U. Wollborn
Bayer MaterialScience AG (Germany)
Dr. K. Schmeer
Despite an immense effort, no new artemisinin derivative
has reached development status. 10-Alkylaminoartemisinins
6 (R’ = H, alkyl; R’’ = alkyl; R’,R’’ = cycloalkyl) are an
artemisinin class whose preparation entails chemistry distinct
to that leading to other derivatives and which extends the
efficacy limit beyond those of the known artemisinins.[16,17]
For any such compound to be developed into a new drug, the
production cost must be comparable to that of ATS (3),[18]
unless a greater efficacy is obtained when a commensurate
increase in cost is permitted. As the interaction of artemisi-
nins with their likely target, the PfATP6 Ca2+ pump, is still
unclear,[19,20] a “rational design” approach is not possible and
emphasis must be placed on improving systemic behavior.
Whilst this depends upon a number of factors,[21] an enhance-
ment of aqueous solubility to a limiting value and attenuation
of lipophilicity as assessed through the logP parameter (P is
Bayer CropScience AG (Germany)
Dr. A. Römer
Analytik und Metabolismusforschung Service GmbH (Germany)
[+] Current affiliation:
Walter Reed Army Institute of Research (USA)
[**] Work at the HKUST was funded by Bayer AG Central Research,
Leverkusen, the Institute of Molecular Technology for Drug
Discovery and Synthesis through the HKSAR University Grants
Committee Areas of Excellence Fund, and the HKSAR Research
Grants Council through the Competitive Earmarked Research
Grants Scheme. The Medicines for Malaria Venture (MMV),
Geneva, also provided financial support of the work at Bayer, and
through Bayer at HKUST and other centers. Dr. W. Collins (Centers
for Disease Control, Atlanta) supplied P. falciparum FVO strain used
for the primate studies.
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
2082
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2006, 45, 2082 –2088

Angewandte
Chemie
the octanol–water partition coefficient) are benefi-
cial.[22,23] One report in the artemisinin area
addresses the lipophilicity issue, wherein enzymatic
hydroxylation of the lipophilic arteether (logP =
3.89–3.99) provides products with logP values of
2.6–2.7, considered to be in an optimal range for
systemic activity.[24] However, the corollary that
lipophilic artemisinins are neurotoxic[25,26] adds sub-
stantial impetus to the drive to attenuate lipophilic-
ity. As neurotoxicity is likely to be associated with a
specific target,[11,27] a linear correlation between
neurotoxicity and lipophilicity for a compound
series is unlikely.[25] However, it is clear that in
order to depress neurotoxicity, logP must be reduced
and, at the same time, phase I metabolism to DHA
must be blocked. As the alkylaminoartemisinins 6
will not undergo such metabolism, polar groups inert
to phase I and phase II metabolism[26] are best
incorporated into the 10-alkylamino substituent
itself. However, the use of hydroxy, carboxy, or
primary amino groups capable of forming phase II
conjugates is proscribed.
The amines 9–11[16,28,29] and 12–16[30] (40–60%)
were obtained from DHA a-trimethylsilyl ether 7
and trimethylsilyl bromide (TMSBr), followed by
treatment of the intermediate bromide 8 formed in
situ with the amine nucleophile (Scheme 1, route a).
Oxidation of the sulfide 12 also provided sulfoxide 13
or sulfone 14. A more straightforward route involved
treatment of DHA first with a mixture of NaBr and
TMSCl in toluene and then with the amine
Scheme 1. Preparation of 10-alkylamino derivatives 9–16; compound 13 is obtained by
oxidation of 12; compound 14 is obtained by either treatment of 8 with thiomorpholine-
S,S-dioxide or oxidation of 12, which itself is obtained from 8 and thiomorpholine.[16,29,30]
Compounds 17 and 18 were prepared according to published procedures.[28]
mCPBA=m-chloroperbenzoic acid; NMO=N-methylmorpholine-N-oxide;
TPAP=tetra-n-propylammonium perruthenate.
(Scheme 1, route b). Under these condition, DHA is con-
verted by the TMSBr formed in situ into the TMS ether 7, and
then into the bromide 8, whose presence was detected by
1H NMR spectroscopy.[28] For most compounds, products
from multigram- or multikilogram-scale reactions (com-
pounds 12 and 14) were isolated by crystallization of the
crude product mixtures. Thus, the candidate compounds are
relatively accessible and are produced as isomerically pure,
air-stable substances.
Assessment of physicochemical properties, neurotoxicity,
and efficacy against malaria was next carried out. For
comparative purposes, the lipophilic 4-fluorophenyl and 4-
chlorophenyl derivatives 17 and 18 were also studied.[16,28]
Measured and calculated[31] logP values (Table 1) reveal that
artemether 4 is relatively lipophilic. Because DHA (2) exists
as two epimers in solution, logP appears as two values, 2.35
and 2.73; the solubility of 2 could not be measured because of
decomposition. Compounds 9, 10, and 12, and especially 17
and 18, are lipophilic with high logP values and very low
aqueous solubility. Compound 14, which is highly crystalline
and easily purified, displays aqueous solubility that exceeds
the limiting requirements for a drug.[22] Compounds 13 and 15
are more soluble but are more costly to prepare than 14.
Although compound 16 has a low calculated logP value, it
also has a low aqueous solubility.
Neurotoxicity was assessed with the in vitro model
developed for screening compounds required for medicinal
and other purposes.[32] As artemisinins exert their neuro-
toxicity by acting selectively on brain stem cells,[11–13,27,33]
primary neuronal brain stem cell cultures from fetal rats,
with endpoints of cytotoxicity and neurofilament integrity,
were used. Intracellular ATP levels, which are a sensitive and
early marker of cytotoxicity,[27,33] were also measured
(Table 2). With a “no observable effect concentration”
(NOEC) on neurofilaments of less than 0.001 mgmLÀ1,
DHA displayed the greatest neurotoxicity. The neurotoxicity
of 9–11 contrasts with the nontoxicity of 14. NOEC and IC50
values for cytotoxicity with cortical neurons were generally
Table 1: Aqueous solubility and octanol–water partition coefficients.
1
3
4
9
10
11
12
13
14
15
16
17
18
Solubility [mgLÀ1 [a]
]
63
2.94
3.15
565
2.77
3.04
117
3.98
3.51
8
5.62
5.15
<2
4.78
4.07
28.4
3.05
3.26
<2
4.97
3.98
1251
2.63
2.03
89
2.49
2.08
294
3.36
2.29
10
[c]
2.30
<1
5.59
5.18
<1
6.15
5.58
logP[b]
logP (calcd)[d]
[a] Aqueous solubility at pH 7.2. [b] Octanol–water partition coefficient for the neutral compound determined at pH 7.4, except that for artesunate (3)
which was determined at pH 2 by HPLC. [c] Not determined. [d] Calculated log P value.[31]
Angew. Chem. Int. Ed. 2006, 45, 2082 –2088
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 2083

Communications
Table 2: Effects of the test compounds on viability and neurofilaments in rat primary neuronal brain
between in vivo and in vitro neurotoxicity
has already been noted with artemisinins 1–
4.[11,13,27,33]
stem cell cultures.[a]
Compound
Viability [mgmLÀ1 [b]
]
ATP [mgmLÀ1 [c]
]
Neurofilaments [mgmLÀ1 [d]
]
NOEC[e]
IC50
NOEC[e]
IC50
NOEC[e]
IC50
Antimalarial IC50 values from in vitro
screens of the 10-alkylamino compounds 9–
18 against Plasmodium falciparum chloro-
quine (CQ)-sensitive D6 or 3D7 and CQ-
resistant W2 or K1 strains generally were
substantially lower (to over 20-fold for
compounds 9 and 10) than those of ATS
(3). The compounds also display superior
activities in in vivo screens in the four-day
Petersꢀ model[36] against CQ-sensitive
P. berghei and CQ-resistant P. yoelii in
mice (Table 3). Compounds 9–12 were
exceedingly active but are excluded from
further consideration because of toxicity,
and unfavorable physicochemical proper-
ties. Overall, compound 14 (artemisone)
emerges as the development candidate.
1
2
3
4
9
10
11
12
14
15
17
18
1.0
0.1
0.1
1.0
0.1
0.01
0.1
1.0
>10
0.01
0.01
–
–
–
–
0.08
0.08
–
–
–
0.001
<0.001
0.1
0.01
10
0.05
0.01
5.0
5.0
>10
>10
>10
10
0.1
>10
–
0.1
3
0.04
>10
>10
>10
>10
4.5
<0.001
1.0
>10
1
<0.001
8.5
>10
0.001
0.001
>10
>10
>10
>10
>10
>10
>10
10
6.3
6.3
0.08
0.001
0.001
0.001
1.0
1.0
1
0.01
0.05
[a] Determined on fetal rat brain stem cells (E18–E19) cultured to generate a permanent neuronal
network during days 1–8; compounds in dimethyl sulfoxide (DMSO) were applied at 0.001, 0.01, 0.1, 1,
and 10 mgmLÀ1 for 7 days starting on day 9. [b] Cytotoxicity was measured by viability based on the
activity of neuron-specific enolases. [c] ATP (adenosine triphosphate) levels measured at 1 day after
initial administration.[27,33,35] [d] Neurotoxicity was assessed by the effect on the cytoskeleton on day 7.
[e] No observable effect concentration.
greater than 10 mgmLÀ1, even for
the highly neurotoxic 11. However,
DHA is also cytotoxic towards
cortical neurons (NOEC(DHA) =
Table 3: In vivo screens against CQ-sensitive P. berghei N strain and CQ-resistant P. yoelii NS strain.[a]
Compound
P. berghei
P. berghei
P. yoelii
P. yoelii
artesunate index[b]
sc
ED90 [mgkgÀ1
sc
]
artesunate index[b]
sc
ED90 [mgkgÀ1
]
sc
po
po
po
1 mgmLÀ1,
NOEC(artemisinin)
3
9
10
14
17
18
7.2
7.1
3.5
2.8
3.1
5.0
4.6
1.0
9.0
12
4.8
6.2
1.9
1.0
2.0
2.5
2.3
1.4
1.7
22.0
0.85
0.52
3.9
1.08
3.0
–
1.0
25.9
42.3
5.6
20.4
7.3
> 50 mgmLÀ1) and non-neuronal
cell lines (NOEC(liver HEP-G2,
epithelium liver IAR, kidney LLC-
PK1) = 0.1 mgmLÀ1, NOEC(arte-
misinin) > 10 mgmLÀ1).[27,32,33] In
general, the lipophilic compounds
are neurotoxic. Neurotoxicity of
the polar ATS (3) is likely due to
facile hydrolysis in situ to DHA.
As the relatively polar DHA and
artemisinin may undergo ring-
opening under aqueous conditions
to provide rather different struc-
tures,[7,34] the actual nature of the
neurotoxic agent is uncertain.
Compound 11 is anomalous in
0.8
0.6
1.5
1.16
3.8
3.0
2.0
5.0
–
–
3
11
12
4.6
0.18
0.51
9.3
1.3
1.9
1.0
25.6
9.0
1.0
7.15
4.9
42.0
1.25
0.61
–
1.84
2.0
1.0
33.6
81.0
3
13
12.0
9.0
–
–
1.0
1.3
–
–
50.0
10.0
–
–
1.0
5.0
[a] Peters’ four-day test[36] with mice treated daily subcutaneously (sc) or orally (po) from the day of
infection (day 0) through to day 3; results (ED90 values) are evaluated from parasite counts in peripheral
blood on day 4. [b] ED90(artesunate)/ED90(compound).
that it also elicits acute systemic toxicity in an in vivo
screen. The outcome of pilot tolerability studies in animal
models, which employ standard indicators of neurotoxic
effects,[35] reflects the results of the in vitro studies. Oral
administration via gavage in corn or sesame oil of ATS (3) at
10 and 50 mgkgÀ1 dayÀ1 for 14 days to female rats induces
uncoordinated gait and vocalization from days 7–11 at the
higher dose. Compounds 10 at the higher dose and 11 at the
lower dose are fatally toxic from days 2–4, generating prior
symptoms of acute systemic toxicity. At 50 mgkgÀ1 dayÀ1,
substantial body-weight loss, coupled with reduced motility,
uncoordinated gait, and piloerection are observed with
compounds 9, 12, 17, and 18. Compound 16 induces uncoor-
dinated gait from day 1, with a slight decrease in body weight.
In contrast, no effects are detectable with compounds 13–15;
the weight gains are no different to controls. The parallel
Even though it is not the most active candidate, its lack of
toxicity couples well with its accessibility and tractable
physicochemical properties. It was therefore subjected to
more-detailed screening in conjunction with the comparator
drug, ATS (3), the most accessible of the current artemisi-
nins.[1,9]
In comparative screens in vitro against a panel of multi-
drug-resistant isolates of P. falciparum, artemisone (14) is
significantly more active than ATS, CQ, and pyrimethamine,
irrespective of the resistance profile (Figure 1). Ex vivo
bioassay represents an effective method for assessing both
comparative bioavailability and efficacy of an orally admin-
istered drug.[38] Oral administration of a 30 mgkgÀ1 single
dose of artemisone (14) or ATS (3) to healthy Saimiri
monkeys and assay of plasma amples drawn from the
monkeys against P. falciparum K1 strikingly reveal that
2084 www.angewandte.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2006, 45, 2082 –2088

Angewandte
Chemie
tended to have later recrudescences
than monkeys in the ATS group.
Complete cure was achieved in
infected monkeys receiving
single oral dose of artemisone
(10 mgkgÀ1)
plus mefloquine
a
(MFQ, 5 mgkgÀ1), although recru-
descence occurred with a lower
MFQ dose (2.5 mgkgÀ1). Cure was
also achieved with oral artemisone
(10 mgkgÀ1 dayÀ1) plus amodia-
quine (AQ, 20 mgkgÀ1 dÀ1) given
for 3 days. Therefore, in vivo effi-
cacy is fully demonstrated. As the
problem of recrudescence is
common to all artemisinins, artemi-
sone will best be used in conjunc-
tion with a longer half-life drug
such as MFQ or AQ; this will
additionally protect each drug part-
ner against the emergence of resis-
tance.[1,9,18]
Figure 1. In vitro antimalarial activities for artemisone, artesunate, chloroquine (CQ), and pyrimeth-
amine (PYR) against both drug-sensitive (3D7) P. falciparum and to isolates that are variously
resistant (R) to CQ, PYR, and mefloquine (MFQ), assayed by administration of the compounds in
DMSO to parasites incubated with 3H-hypoxanthine. Inhibition of the uptake of hypoxanthine is
expressed as the concentration at 50% inhibition (IC50) in ngmLÀ1 [37]
3D7 (clone of NF54 isolate,
.
Netherlands); K1 (Thailand) CQ-R, PYR-R; VS1 (Vietnam) CQ-R, PYR-R; 7G8 (Brazil) CQ-R, PYR-R;
HB3-B2 (Honduras) PYR-R; FCB (Colombia) CQ-R; FCR3 (Gambia) CQ-R, PYR-R; Tm90 C2A
(Thailand) CQ-R, PYR-R, MFQ-R; W2 (Indochina) CQ-R, PYR-R; FCR-8 (Gambia); FCC2 (Hainan);
DD2 (clone from W2) CQ-R, PYR-R. Error bars represent the standard deviation about the arithmetic
mean.
To demonstrate the lack of
metabolism to DHA and to identify
metabolites formed in vitro and in
vivo, labeled artemisone 14* incor-
porating either 14C or 2H in the
metabolically
inert
15-methyl
artemisone has a greatly enhanced bioavailability, as reflected
in the greater and significantly more sustained activity in
plasma compared to ATS (Figure 2). Next, aotus monkeys
infected with P. falciparum FVO isolate, which is resistant to
CQ and antifolates, were treated with each of ATS and
artemisone at 10 mgkgÀ1 dayÀ1 for 3 days. The artemisone-
treated group cleared parasites within 24 h after commence-
ment of treatment, whereas parasites were still present 48 h
after treatment with ATS (Table 4). One monkey (no. 7,
Table 4) in the artemisone group was cured, and the others
group was required. The exocyclic methylidene was cleaved
from artemisitene (19)[39] to give the known[40] dicarbonyl
2
compound 16, treatment of which with H- or 14C-methyli-
dene triphenylphosphorane provided labeled artemisitene
19*, as shown in Scheme 2 for the deuteriated reagents. As
attempted reduction of the exocyclic methylidene with
conjugate hydride donors failed, the carbonyl group in 19*
was first reduced and the resulting allylic alcohol 21* was
reduced with either Et3SiH for the 14C-labeled compound or
Et3Si2H for the 2H-labeled compound to give glycal 22*. The
latter was converted into 12* by stereose-
lective addition of HCl and treatment of
the intermediate 10b-chloride with thio-
morpholine to give labeled 12*; oxidation
provided labeled artemisone 14*. This is
the first semisynthesis of an artemisinin
bearing a labeled group at a metabolically
inert region of the artemisinin skeleton, for
which total, albeit elegant, syntheses were
previously used.[41]
In vitro incubation of 14* with human
liver microsomes gave after 30 min metab-
olites M1
( ꢀ 34% of total radioactivity (TRA)), M2
( ꢀ 19% of TRA), M3 ( ꢀ 16% of TRA),
and smaller amounts of M4 and M5 (each
< 6% TRA) at approximately 77% con-
version of 14* (Scheme 2). Each metabolite
was isolated and identified unambiguously
by the use of 1Dand 2DNMR spectro-
scopic analyses coupled with high-resolu-
Figure 2. Ex vivo maximum inhibitory dilutions (MID) of plasma samples obtained from
healthy Saimiri monkeys at 1–4 and 6 h after a single oral dose of artemisone (14) or
artesunate (3; 10 mgkgÀ1 each) required to inhibit maturation of ꢁ90% P. falciparum K1
from rings to schizonts. Error bars represent the standard error of the mean of MID data
obtained for four monkeys with each drug.
Angew. Chem. Int. Ed. 2006, 45, 2082 –2088
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 2085

Communications
Table 4: In vivo activities of artesunate 3 and artemisone 14 against P. falciparum FVO in aotus
human liver microsomes and 14 recombi-
nant CYP isoforms together with selective
inhibitors indicated that only recombinant
CYP3A4 metabolizes artemisone in signifi-
cant amounts, with trolleandomycin, aza-
mulin, and ketoconazole being efficient
inhibitors. Treatment of human hepatocytes
with 4–10000 ngLÀ1 of artemisone had a
borderline inductive effect on CYP2B6 and
CYP3A4 at the highest concentration, with
no induction of CYP1A2 and CYP2C19.
Incubation of artemisone and standard
monkeys.[a]
Monkey
Initial parasitaemia
Compound
[10 mgkgÀ1 dayÀ1
days 0–2
Parasitaemia [”103 mLÀ1
on day n[b]
]
Time to
recrudescence
[days]
[”103 mLÀ1
]
]
day 1
day 2
day 3
1
2
3
4
5
6
7
145
515
340
490
10
3
3
3
14
14
14
14
2.1
15
40
0
0
0
0.2
0.6
0.4
0
0
0
0
0
0
0
0
0
0
9
20
18
20
20
29
–
150
85
0
0
[a] Two aotus monkeys were infected with blood containing aotus-adapted P. falciparum FVO isolate
resistant to CQ and antifolates. Blood was drawn from the infected monkeys and used to infect five
malaria-naïve monkeys. 6–9 Days after inoculation, ATS (3) and artemisone (14) were administered
orally for 3 days at 10 mgkgÀ1 dayÀ1 (total 30 mgkgÀ1). Thick blood films were examined microscopically
twice a week; these were considered to be negative when no parasites were detected after examination of
100 microscopic fields (ꢀ0.1 mL blood). [b] Parasitaemia on days after the initial treatment on day 0.
substrates
with
CYP1A2,
-2A6,
-2C9, -2C19, -2D6, -2E1, and -3A4 has no
effect on substrate metabolism. Therefore,
induction of CYP3A4 or CYP2B6 with
artemisone, in contrast to the current arte-
misinins, will likely not be a problem
in clinical use.[43]
All aspects of preclinical screen-
ing, including the results of pharma-
cokinetic studies and a detailed tox-
icological assessment, allow artemi-
sone to be carried forward into
clinical trials. We have shown that
artemisinins which are structurally
distinct to but significantly more
active than current artemisinins can
be prepared through embedding
metabolically inert polar groups
within the alkylamino substituents
attached directly to C10 of the arte-
misinin nucleus. The development
candidate chosen from within this
class, artemisone, has a different
metabolic profile to the current arte-
misinins and displays favorable phys-
icochemical properties and negligi-
ble neuro- and cytotoxicities. Nota-
bly, in key with the enhanced anti-
malarial activity of artemisone,
inhibition of the likely target, the
parasite Ca2+ pump PfATP6,[19] is
much greater for artemisone than
Scheme 2. Preparation of labeled artemisone 14* (C*H3 =14CH3 or C2H3) and CYP3A4-mediated formation of
metabolites M1–M5 (Ph3P 14CH2, Et3SiH for Clabel; Ph 3P C H2, Et3Si H for H label; yields quoted for
isolated deuteriated compounds). Reagents and conditions: a) NaIO4/RuCl3, C C4,l MeCN, 208C , 54%;
b) C2H3PPh3Br, n-C4H9Li, N2, THF, À788C; then 20, À78–208C , 20 h, 48%; c)iBu2AlH, CH2Cl2, À788C, 2 h,
>98%; d) Et3SiH or Et3Si2H, TFA, À788C, 4 h, 40%; e) HCl, CH2Cl2, 08C; then thiomorpholine, Et3N, 08C,
45 min; 50% to 12*; then mCPBA, K2CO3, Et2O, 08C , to14*, 72%. TFA=trifluoroacetic acid.
14
2
2
2
=
=
tion mass spectrometry. The identity of M1 was also verified
by independent preparation of unlabeled M1 from artemi-
sone and ozone, according to a method used to desaturate
tertiary amines.[42] Rat and dog microsomes produce M1 with
larger amounts of M2 and M3, and a large amount of M1 with
a small amount of M3, respectively. Incubation with human,
dog, and rat hepatocytes largely reflect the metabolism with
the microsomes. DHA could not be detected in any of the
systems. Plasma from rats administered orally with artemi-
sone contains M2 and M3 and minor amounts of M1, M4, and
M5. As the in vivo plasma profile of metabolites of rats
resembles the in vitro profiles from rat liver microsomes and
hepatocytes, the metabolism of artemisone in humans will
likely reflect formation of M1–M5 in human microsomes and
hepatocytes. Incubation of 14C-labeled artemisone with
artemisinin: Ki values for artemisinin and artemisone are
169 Æ 31 and 1.7 Æ 0.6 nm, respectively, and for the P. vivax
orthologue, the Ki values are 7.7 Æ 4.9 and 0.072 Æ 0.012 nm,
respectively.[20] The cost of artemisone, which is obtained in 2–
3 steps from artemisinin, is compatible with the requirement
for an antimalarial drug, and its enhanced activity over the
current artemisinins will permit the use of lower clinical dose
regimens, thereby augmenting the supply of artemisinin.
Ironically, the current supply shortfall[44] is largely due to the
demand for the fixed combination drug artemether–lumifan-
trine, of which that amount of drug containing half a gram of
artemether is required to cure a malaria patient.[45] In 2005,
the estimated global demand was for 120 million adult
treatment courses.[46] Thus, approximately 114 tons of arte-
misinin is required, given that the overall chemical yield of
2086 www.angewandte.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2006, 45, 2082 –2088

Angewandte
Chemie
[19] U. Eckstein-Ludwig, R. J. Webb, I. D. A. van Goethem, J. M.
artemether from artemisinin is 53%. The time to move on to
new, more-active artemisinins is at hand.
East, A. G. Lee, M. Kimura, P. M. OꢀNeill, P. G. Bray, S. A.
Ward, S. Krishna, Nature 2003, 424, 957 – 961.
[20] A.-C. Uhlemann, A. Cameron, U. Eckstein-Ludwig, G. Fisch-
barg, P. Iserovich, F. A. Zuniga, M. East, A. Lee, L. Brady, R. K.
Haynes, S. Krishna, Nat. Struct. Mol. Biol. 2005, 12, 628 – 629.
[21] G. L. Patrick, An Introduction to Medicinal Chemistry, 2nd ed.,
Oxford University Press, Oxford, 2001, chap. 10, pp. 223 – 224;
G. L. Patrick, An Introduction to Medicinal Chemistry, 2nd ed.,
Oxford University Press, Oxford, 2001, chap. 11, pp. 260 – 266.
[22] C. A. Lipinski, F. Lombardo, B. W. Dominy, P. J. Feeney, Adv.
Drug Delivery Rev. 2001, 46, 3 – 26.
Received: August 29, 2005
Revised: November 4, 2005
Published online: January 30, 2006
Keywords: antimalarial agents · artemisinins ·
.
medicinal chemistry · metabolism · neurotoxicity
[23] R. Manhold, R. F. Bekker, K. Dross, G. Bijloo, G. de Vries,
Quant. Struct.-Act. Relat. 1998, 17, 517 – 536.
[24] K. Ramu, J. K. Baker, J. Med. Chem. 1995, 38, 1911 – 1921.
[25] A. K. Bhattacharjee, J. M. Karle, Chem. Res. Toxicol. 1999, 12,
422 – 428.
[26] R. K. Haynes, Curr. Opin. Infect. Dis. 2001, 14, 719 – 772.
[27] G. Schmuck, E. Roehrdanz, R. K. Haynes, R. Kahl, Antimicrob.
Agents Chemother. 2002, 46, 821 – 827.
[28] R. K. Haynes, H. W. Chan, M.-K. Cheung, S.-K. Chung, W.-L.
Lam, H.-W. Tsang, A. Voerste, I. D. Williams, Eur. J. Org. Chem.
2003, 2098 – 2114.
[29] R. K. Haynes, W. L. Lam, H. W. Chan, M. K. Cheung, H. W.
Tsang, Synthesis and Antiparasitic Activity of Artemisinin
Derivatives (Endoperoxides), International patent WO
2000004024A1, 2000 [Chem. Abs. 2000, 132, 122783].
[30] R. K. Haynes, Antiparasitic Artemisinin Derivatives, Interna-
tional patent WO 03/076446A1, 2002 [Chem. Abs. 2003, 139,
261439].
[31] A. K. Ghose, G. M. Crippen, J. Chem. Inf. Comput. Sci. 1987, 27,
21 – 35.
[32] G. Schmuck, H. J. Ahr, G. Schlꢁter, In Vitro Mol. Toxicol. 2000,
13, 37 – 49.
[33] G. Schmuck, R. K. Haynes, Neurotoxic. Res. 2000, 2, 37 – 49.
[34] J. K. Baker, J. D. McChesney, H.-T. Chi, Pharm. Res. 1993, 10,
662 – 666; B. Yagen, Y. M. Pu, H. J. Yeh, H. Ziffer, J. Chem. Soc.
Perkin Trans. 1 1994, 843 – 846; D. S. Torok, H. Ziffer, Tetrahe-
dron Lett. 1995, 36, 829 – 832; R. K. Haynes, H.-O. Pai, A.
Voerste, Tetrahedron Lett. 1999, 40, 4715 – 4718.
[35] G. Schmuck, M. Temerowski, R. K. Haynes, B. Fugmann, Recent
Res. Dev. Antimicrob. Agents Chemother. 2003, 3, 35 – 47.
[36] W. Peters, B. L. Robinson in Handbook of Animal Models of
Infection (Eds.: O. Zak, M. Sande), Academic Press, London,
1999, Section VI, Parasitic Infection Models, chap. 92, pp. 756 –
771.
[37] R. E. Desjardins, C. J. Canfield, J. D. Haynes, J. D. Chulay,
Antimicrob. Agents Chemother. 1979, 16, 710 – 718.
[38] B. M. Kotecka, K. H. Rieckmann, Am. J. Trop. Med. Hyg. 1993,
49, 460 – 464.
[39] L. Zhang, W.-S. Zhou, Youji Huaxue 1988, 8, 329 – 330 [Chem.
Abs. 1989, 111, 78408]; L. Zhang, W.-S. Zhou, Huaxue Xuebao
1989, 47, 1117 – 1119 [Chem. Abs. 1990, 112, 217297]; S.
Paitayatat, B. Tarnchompoo, Y. Thebtaranonth, Y. Yuthavong,
J. Med. Chem. 1997, 40, 633 – 638; S. Ekthawatchai, J. Lertvor-
achon, P. Meepowpan, T. Thongpanchang, Y. Thebtaranonth, Y.
Yuthavong, Synth. Commun. 2003, 33, 1855 – 1860.
[1] P. L. Olliaro, W. R. Taylor, J. Postgrad. Med. 2004, 50, 40 – 44.
[2] J.-M. Liu, M.-Y. Ni, J.-F. Fan, Y.-Y. Tu, Z.-H. Wu, Y.-L. Wu, W.-S.
Zhou, Acta Chim. Sin. 1979, 37, 129 – 141.
[3] X. Liu, Yao Xue Tong Bao 1980, 15, 183; X. Liu, CN 85100781A,
1986 [Chem. Abstr. 1987, 107, 78111j]; Y. Li, P.-L. Yu, Y.-X.
Chen, R.-Y. Ji, Acta Chim. Sin. 1982, 40, 557 – 561.
[4] Y. Li, P.-L. Yu, Y.-X. Chen, L.-Q. Li, Y.-Z. Gai, D.-S. Wang, Y.-P.
Zheng, Acta Chim. Sin. 1981, 16, 429 – 439.
[5] A. Brossi, B. Venugopalan, L. Dominguez Gerpe, H. J. C. Yeh,
J. L. Flippen-Anderson, P. Buchs, X.-D. Luo, W. Milhous, W.
Peters, J. Med. Chem. 1988, 31, 645 – 650.
[6] Q.-G. Li, J. O. Peggins, L. L. Fleckenstein, K. Masonic, M. H.
Heiffer, T. G. Brewer, J. Pharm. Pharmacol. 1998, 50, 173 – 182;
V. Navaratnam, S. M. Mansor, N. W. Sit, J. Grace, Q.-G. Li, P.
Olliaro, Clin. Pharmacokinet. 2000, 39, 255 – 270.
[7] N. Vyas, B. A. Avery, M. A. Avery, C. M. Wyandt, Antimicrob.
Agents Chemother. 2002, 46, 105 – 109.
[8] J. M. Grace, A. J. Aguilar, K. M. Trotman, T. G. Brewer, Drug
Metab. Dispos. 1998, 26, 313 – 317; M. A. van Agtmael, V. Gupta,
T. H. van der Wösten, J.-P. B. Rutten, C. J. van Boxtel, Eur. J.
Clin. Pharmacol. 1999, 55, 405 – 410; T. Gordi, R. Xie, N. V.
Huong, D. X. Huong, M. O. Karlsson, M. Ashton, Br. J. Clin.
Pharmacol. 2005, 59, 189 – 198.
[9] C. J. Woodrow, R. K. Haynes, S. Krishna, Postgrad. Med. J. 2005,
81, 71 – 78.
[10] T. G. Brewer, S. J. Grate, J. O. Peggins, P. J. Weina, J. M. Petras,
B. S. Levine, M. H. Heiffer, B. G. Schuster, Am. J. Trop. Med.
Hyg. 1994, 51, 251 – 259; S. Kamchonwongpaisan, P. McKeever,
P. Hossler, H. Ziffer, S. R. Meshnick, Am. J. Trop. Med. Hyg.
1997, 56, 7 – 12; A. Nontprasert, M. Nosten-Bertrand, S.
Pukrittayakamee, S. Vanijanonta, B. J. Angus, N. J. White, Am.
J. Trop. Med. Hyg. 1998, 59, 519 – 522.
[11] D. L. Wesche, M. A. DeCoster, F. C. Tortella, T. G. Brewer,
Antimicrob. Agents Chemother. 1994, 38, 1813 – 1819.
[12] J. Fishwick, W. G. McLean, G. Edwards, S. A. Ward, Chem.-Biol.
Interact. 1995, 96, 263 – 271.
[13] W. G. McLean, S. A. Ward, Med. Trop. 1998, 58(Suppl. 3), 28 –
31.
[14] T. Gordi, E.-I. Lepist, Toxicol. Lett. 2004, 147, 99 – 107.
[15] R. Johann-Liang, R. Albrecht, Clin. Infect. Dis. 2003, 36, 1626 –
1627.
[16] R. K. Haynes, W.-Y. Ho, H.-W. Chan, B. Fugmann, J. Stetter,
S. L. Croft, L. Vivas, W. Peters, B. L. Robinson, Angew. Chem.
2004, 116, 1405 – 1409; Angew. Chem. Int. Ed. 2004, 43, 1381 –
1385; R. K. Haynes, Angew. Chem. 2005, 117, 2100 – 2101;
Angew. Chem. Int. Ed. 2005, 44, 2064 – 2065.
[40] N. Acton, J. M. Karle, R. E. Miller, J. Med. Chem. 1993, 36,
2552 – 2557.
[41] M. A. Avery, J. D. Bonk, S. Mehrotra, J. Labelled Compd.
Radiopharm. 1996, 38, 249 – 254; M. A. Avery, J. D. Bonk, J.
Bupp, J. Labelled Compd. Radiopharm. 1996, 38, 263 – 267.
[42] P. S. Bailey, D. A. Mitchard, A.-I. Y. Khashab, J. Org. Chem.
1968, 33, 2675 – 2680.
[43] HPLC fractions (solutions in DMSO) of the metabolites isolated
at Bayer Pharma (Germany) were submitted to preliminary
screens at the Australian Army Malaria Institute against
[17] R. K. Haynes, H.-W. Chan, W.-Y. Ho, C. K. F. Ko, L. Gerena,
D. E. Kyle, W. Peters, B. L. Robinson, ChemBioChem 2005, 6,
659 – 667.
[18] Malaria Control Department & Essential Drugs and Medicines
Policy Department, World Health Organization, Geneva,
Access to Antimalarial Medicines: Improving the Affordability
and Financing of Artemisinin-Based Combination Therapies,
WHO/CDS/MAL/2003.1095, p. 30.
Angew. Chem. Int. Ed. 2006, 45, 2082 –2088
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 2087

Communications
P. falciparum K1 in vitro. Metabolite M1 (IC50 = 1.7 ngmLÀ1
)
and M3 (IC50 = 2.3 ngmLÀ1) reveal similar activities to that of
artemisone (IC50 = 0.8 ngmLÀ1), but M2 is less active (IC50
25.7 ngmLÀ1).
=
[44] M. Enserink, Science 2005, 307, 33.
[45] M. van Vugt, S. Looareesuwan, P. Wilairatana, R. McGready, L.
Villegas, I. Gathmann, R. Mull, A. Brockman, N. J. White, F.
Nosten, Trans. R. Soc. Trop. Med. Hyg. 2000, 94, 545 – 548.
[46] from http://rbm.who.int/wmr2005/htm: Roll Back Malaria
World Malaria Report 2005: Section III Global Financing,
Commodities And Service Delivery; II. Commodities And
Service Delivery, Roll Back Malaria Partnership, World Health
Organization, Geneva.
2088 www.angewandte.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2006, 45, 2082 –2088