Artemisinin

Artemisinin (IPA: [arte’misinən]) is a drug Artemisinin

used to treat multi-drug resistant strains of
falciparum malaria. The compound (a
sesquiterpene lactone) is isolated from the
shrub Artemisia annua long used in
traditional Chinese medicine. Not all shrubs
of this species contain artemisinin.
Apparently it is only produced when the
plant is subjected to certain conditions. It
can be synthesized from arteminisic acid.[1] (3R,5aS,6R,8aS,9R,12S,12aR)-
octahydro-3,6,9-trimethyl-3,12-
Chemical name
epoxy-12H-pyrano[4,3-j]-
1,2-benzodioxepin-10(3H)-one
Chemical formula C15H22O5
Molecular mass 282.332 g/mol
Artemisinine
Synonyms
Qinghaosu
CAS number 63968-64-9
PubChem 68827
ATC code P01BE01
Density 1.24 ± 0.1 g/cm3
Melting point 152 - 157 °C
Boiling point decomp.
CC1CCC2C(C(=O)OC3
SMILES
C24C1CCC(O3)(OO4)C)C

History

Artemisia has been used by Chinese herbalists for more than a thousand years in the treatment
of many illnesses, such as skin diseases and malaria. In the 1960s a research program was set
up by the Chinese army to find an adequate treatment for malaria. In 1972, in the course of this
research, Tu Youyou discovered artemisinin in the leaves of Artemisia annua. The drug is
named qinghaosu in Chinese. It was one of many candidates then tested by Chinese scientists
from a list of nearly 200 traditional Chinese medicines for treating malaria. It was the only one
that was effective.

It remained largely unknown to the rest of the world for about ten years, until results were
published in a Chinese medical journal. The report was met with skepticism at first, because the

Page 1 of 3
Chinese had made unsubstantiated claims about having found treatments for malaria before. In
addition, the chemical structure of artemisinin, particularly the peroxide, appeared to be too
unstable to be a viable drug.

For many years, access to the purified drug and the plant it was extracted from were restricted
by the Chinese government. However, Artemisia annua is a common shrub and has been found
in many parts of the world, including along the Potomac River, in Washington, D.C.

Currently, artemisinin is widely used in China and Southeast Asia for treatment of malaria. It is
often used without taking precautions against conditions that might lead to resistance of the
malaria parasite to this drug, leading to concern that the effectiveness of artemisinin may be
reduced in the near future, as is the case with other classes of antimalarial drugs.

Because artemisinin itself has physical properties such as poor bioavailability that limit its
effectiveness, semi-synthetic derivatives of artemisinin, including artemether and artesunate,
have been developed. However, their activity is not long lasting, with significant decreases in
effectiveness after one to two hours. To counter this drawback, artemisinin is given alongside
lumefantrine to treat uncomplicated falciparum malaria. Lumefantrine has a half-life of about 3 to
6 days. Such a treatment is called ACT (artemisinin-based combination therapy); other
examples are artemether-lumefantrine, artesunate-mefloquine, artesunate-amodiaquine, and
artesunate-sulfadoxine/pyrimethamine. Recent trials have shown that ACT is more than 90%
effective, with a recovery of malaria after three days, especially for the chloroquine-resistant
Plasmodium falciparum.

The World Health Organization has recommended that a switch to ACT should be made in all
countries where the malaria parasite has developed resistance to chloroquine. Artemisinin and
its derivatives are now standard components of malaria treatment in China, Vietnam, and some
other countries in Asia and Africa, where they have proved to be safe and effective anti-malarial
drugs. They have minimal adverse side effects. Currently, artemisinin is not widely available in
the United States or Canada, but is easy to find in Africa and Asia. There have been some
concerns about the quality of some products on offer in Africa, but sticking to one of the
European (often Belgian) manufacturers could overcome this problem.

To counter the present shortage in leaves of Artemisia annua, researchers have been searching
for a way to develop artemisinin artificially in the laboratory. A recent paper in Nature presented
a geneticly engineered yeast that created a closely related compound which can be efficently
converted into Artemisinin. The compound called OZ-277 (also known as RBx11160),
developed by Jonathan Vennerstrom at the University of Nebraska, has proved to be even more
effective than the natural product in test-tube trials. A six month trial of the drug on human
subjects in Thailand was started in January 2005. There are also plans to have the plant grow in
other areas of the world (outside Vietnam and China).

Cancer Treatment

Artemisinin is under early research and testing for treatment of cancer. Artemisinin has a
peroxide lactone group in its structure. It is thought that when the peroxide comes into contact
with high iron concentrations (common in cancerous cells), the molecule becomes unstable and
releases reactive oxygen species. It has been shown to reduce angiogenesis and the
expression of vascular endothelial growth factor in some tissue cultures.

Page 2 of 3
 Mechanism of Action

The specific mechanism of action of artemisinin is not well understood, and there is ongoing
research directed at elucidating it. When the parasite that causes malaria infects a red blood
cell, it consumes hemoglobin and liberates free heme, an iron-porphyrin complex. The iron
reduces the peroxide bond in artemisinin generating high-valent iron-oxo species, resulting in a
cascade of reactions that produce reactive oxygen radicals which damage the parasite leading
to its death.[2]

Numerous studies have investigated the type of damage that these oxygen radicals may induce.
For example, Pandey et al. have observed inhibition of digestive vacuole cysteine protease
activity of malarial parasite by artemisinin.[3] These observations were further confirmed by ex
vivo experiments showing accumulation of hemoglobin in the parasites treated with artemisinin,
suggesting inhibition of hemoglobin degradation. They found artemisinin to be a potent inhibitor
of hemeozoin formation activity of malaria parasite.

A 2005 study investigating the mode of action of artemisinin using a yeast model demonstrated
that the drug acts on the electron transport chain, generates local reactive oxygen species, and
causes the depolarization of the mitochondrial membrane. [4]

Artemisinins have also been shown to inhibit PfATP6, a SERCA-type enzyme (calcium
transporter) and artemisinin has been shown to compete with thapsigargin for SERCA binding,
though artemesinin is much less toxic to mammalian cells. Resistance to artemisinin is
conferred by a single mutation in the calcium transporter (PfATP6). This mutation has been
studied in the laboratory but recently a study from French Guiana in field isolates of malaria
parasites has identified a different mutation in the calcium transporter (PfATP6) that is
associated with resistance to artemether, lending support to the idea that PfATP6 is the target
for artemisinins.[5]

References

1. ^ Acton, N. & Roth, R.J. On the conversion of dihydroartemisinic acid into artemisinin. J.
Org. Chem. 57, 3610-3614 (1992)
2. ^ Cumming, Jared N.; Ploypradith, Poonsakdi; Posner, Gary H. Antimalarial activity of
artemisinin (qinghaosu) and related trioxanes: mechanism(s) of action. Advances in
Pharmacology (San Diego) (1997), 37 253-297.
3. ^ Pandey et al
4. ^ Li et al., PLOS Genetics, September 2005, Volume 1, Issue 3
5. ^ A.-C. Uhlemann et al. Nature Struct. Mol. Biol. 12, 628-629;2005

Village Volunteers
5100 South Dawson Street, Suite 202 • Seattle, WA 98118 USA
206.577.0515 • info@villagevolunteers.org • www.villagevolunteers.org

Page 3 of 3
19.8 n&v AM 13/8/04 5:33 pm Page 838

news and views

Medicinal chemistry ring system (Fig. 1b). This was their first
breakthrough. Remarkably, not only were
A worthy adversary for malaria the resulting compounds stable, but when
tested against human strains of the malaria
Paul M. O’Neill parasite, they also had superior parasite-
killing properties even to artesunate and
A remarkable set of antimalarial drug candidates has been developed artemether.
by an international collaboration of scientists, using the age-old The first series of the compounds were
Chinese herbal medicine artemisinin as a template. poorly soluble in water, however, so the next
step was to increase the compounds’ water
solubility so that they could be more easily

N
early two billion people live in areas
where malaria is endemic, and the
incidence of this disease is increasing
dramatically, mainly because many malaria-
parasite strains have become resistant to the
available drugs1. The devastating spread of
parasite resistance has prompted the world-
wide search for new classes of effective anti-
malarial drugs. On page 900 of this issue,
Vennerstrom and colleagues2 describe the
development of a new class of synthetic drug
related to the natural antimalarial product
artemisinin.
Artemisinin has been used in traditional
Chinese herbal fever remedies for more
than 1,500 years3. It is extracted from sweet
wormwood (Artemisia annua), and is used
as the basis for synthesizing several modern
antimalarial drugs, such as artesunate and
artemether. These artemisinin derivatives
are increasingly important in the treatment
of drug-resistant malaria, as they are the
most potent antimalarials available, rapidly
killing all blood stages of the malaria parasite
Plasmodium falciparum. But the overall yield
of the artemisinin extraction process is poor,
so that its derivatives, though inexpensive by
the standards of developed countries, are
relatively expensive when put in the African
context. In addition, the drugs only act for a
short time, so the dosage regimen can lead to
patient non-compliance and subsequent
treatment failure. These shortcomings have
prompted medicinal chemists to try to make
entirely synthetic analogues of artemisinin
with improved antimalarial properties4.
Artemisinins contain a structural feature
known as an endoperoxide bridge (Fig. 1a)
that is key to their antimalarial activity, and
one of the major challenges in the pursuit of
synthetic analogues has been to introduce
this bridge into candidate drugs. Incor-
porating the endoperoxide ‘warhead’ is
now possible, but many of the candidate
analogues produced have significant draw-
backs, including poor antimalarial activity
create a new class of endoperoxide anti-
malarials that have superior properties to the
artemisinin derivatives. The team consisted
of chemists,parasitologists,pharmacokineti-
cists and toxicologists.
The first stage of their drug discovery
process involved a systematic examination of
a chemical group known as secondary
ozonides (or 1,2,4-trioxolanes; Fig. 1b). This
class of compounds is well known to organic
chemists as intermediates obtained by
exposing alkenes to ozone. Although they
have the requisite endoperoxide bridge,
these molecules tend to be highly unstable —
not a promising starting point, one might
think.And unsurprisingly,some of the initial
ozonide compounds tested turned out to be
very poor antimalarials.
In artemisinin, the sensitive peroxide
bridge is protected by bulky chemical rings,
and the team attempted to mimic this feature
by fusing a large, bulky group known as an
adamantane ring onto the standard ozonide
a
H
O
O
O
H
H
O
O
Artemisinin
b
R O O R
O O
R R
The starting point —
a secondary ozonide
c
O O
NH2
O
O
NH
absorbed from the gastrointestinal tract7.
Many potential drugs containing chemical
groups that increase water solubility were
examined, and this led to the discovery of
a candidate, OZ277 (Fig. 1c), that was
well absorbed when administered orally to
animals, and had outstanding antimalarial
properties both in vitro and in vivo. Unlike
the available artemisinin derivatives, OZ277
is structurally simple and its synthesis can be
scaled up in a way that is economically feasi-
ble.What’s more,its toxicological profiles are
satisfactory. Following the development of a
large-scale synthesis for OZ277, it has just
entered ‘first into man’studies.
Given the effectiveness of this new struc-
turally simple drug, the question arises: how
does it kill parasites? It seems probable that it
shares a common mechanism of action with
the artemisinins, where it is proposed that an
interaction of the endoperoxide bridge with
ferrous iron or haem from the red blood cell
Endoperoxide bridge
Adamantane ring
O O O O
O
Relatively unstable Highly potent
and inactive antimalarial
Endoperoxide bridge

and syntheses that are not stereoselective

or not amenable to being scaled up5,6. So,
although many research groups had focused
on the problem for some fifteen years, the
goal of a cheap, synthetic, endoperoxide-
based antimalarial had not been realized.
Vennerstrom et al.2 now provide the solu-
tions to many of these problems. They com-
piled a wish-list of ideal drug properties for
the target molecule and stuck strictly to it
through a process of drug-optimization to
OZ277
Figure 1 Artemisinin and its analogues. a, Chemical structure of artemisinin. The space-filled model
on the right shows how the sensitive endoperoxide bridge is shielded by the ring system. b, Drug
prototypes. Vennerstrom and colleagues2 fused an adamantane ring system onto the basic ozonide
ring to produce a new class of antimalarial drug with superior activity to artemisinin. c, The clinical
candidate, OZ277. The adamantane system protects the endoperoxide bridge, and the boxed chemical
group is key to the compound’s improved water solubility and pharmaceutical properties. Carbon
atoms, grey; hydrogen atoms, light blue; oxygen atoms, red; nitrogen atoms, dark blue.

838 NATURE | VOL 430 | 19 AUGUST 2004 | www.nature.com/nature

©2004 Nature Publishing Group
19.8 n&v AM 13/8/04 5:34 pm Page 839
in which the parasite lodges generates toxic,
carbon-centred free radicals8,9. Theoreti-
cally, these highly reactive molecules can
modify key parasite proteins, disabling
essential biological targets and killing the
parasite. Using a technique known as spin-
trapping, Vennerstrom and colleagues pro-
vide evidence that trioxolanes can indeed
generate a carbon-centred free radical in a
manner reminiscent of the artemisinins.
There are several proposed protein targets
for the noxious free radicals produced by
artemisinin, one of which is an enzyme
known as PfATP6 (ref. 10). Future studies
should reveal whether OZ277 and its chemi-
cal siblings target the same proteins as the
artemisinin derivatives.
The development of OZ277 is a flagship
project for the Medicines for Malaria Ven-
ture11,12. It is an excellent example of how a
well-managed partnership between acad-
emia and major pharmaceutical companies
can have a significant impact on antimalarial
product research and development.
Basing the drug development process on
a chemically unstable entity such as a sec-
ondary ozonide was a daring move. And
the research that enabled ozonides to be
redesigned, not only to increase the chemical
and metabolic stability, but also to provide
phenomenal antimalarial properties, is
impressive. The subsequent tailoring of the
‘ozonide’ molecule to enhance its availability
to the body was hugely successful — the new
synthetic analogues are more potent and
act for longer in vivo than artemether and
artesunate by some margin. As such, when
combined with a second antimalarial,
this new class could offer the best solution to
date for destroying drug-resistant malaria
parasites. ■ tal axis (where there is no magnetic induc-
Paul M. O’Neill is in the Departments of Chemistry
and of Pharmacology, University of Liverpool,
Liverpool L69 7ZD, UK.
e-mail: p.m.oneill01@liverpool.ac.uk
1. Winstanley, P. A., Ward, S. A. & Snow, R. W. Microbes Infect. 4,
157–164 (2002).
2. Vennerstrom, J. L. et al. Nature 430, 900–904 (2004).
3. Klayman, D. L. Science 228, 1049–1055 (1985).
4. Borstnik, K., Paik, I. H., Shapiro, T. A. & Posner, G. H.
Int. J. Parasitol. 32, 1661–1667 (2002).
5. O’Neill, P. M. & Posner, G. H. J. Medicin. Chem. 47, 2945–2964
(2004).
6. Vroman, J. A., Alvim-Gaston, M. & Avery, M. A. Curr. Pharm.
Design 5, 101–138 (1999).
7. Haynes, R. K. Curr. Opin. Infect. Dis. 14, 719–726 (2001).
8. Posner, G. H. & Oh, C. H. J. Am. Chem. Soc. 114, 8328–8329
(1992).
9. Wu, Y. K. Acc. Chem. Res. 35, 255–259 (2002).
10. Eckstein-Ludwig, U. et al. Nature 424, 957–961 (2003).
11. Vennerstrom, J. L., Dong, Y., Chollet, J. & Matile, H. US Patent
6,486,199 (2002).
12. www.mmv.org/pages/page_main.htm
Nature Outlook: Malaria
More about the challenges posed by malaria
appears in the collection of articles published
as a Nature Outlook supplement, beginning
on page 923 of this issue. The articles cover
scientific, social and political problems, with
the emphasis on Africa.
NATURE | VOL 430 | 19 AUGUST 2004 | www.nature.com/nature
Condensed-matter physics
Vortices weave a tangled web
David R. Nelson
In high-temperature superconductors, quantized vortex filaments can
be twisted up into a DNA-like double helix. An experiment is proposed
to test how easily these vortex lines cut through each other.
I
nside a superconductor, electrical currents
flow without resistance.Almost as remark-
able as this electron flow without dissipa-
tion are the quantized, thread-like vortices of
charge that swirl like miniature tornadoes
around lines of magnetic field. Last year,
Alexei Abrikosov shared the Nobel Prize in
Physics for his brilliant 1957 prediction
(made well before similar developments in
high-energy theory and astrophysics) that,
in a class of materials called ‘type II’
superconductors, a regular lattice of parallel
vortex filaments, aligned with an external
magnetic field, would form1. In high-
temperature copper-oxide superconductors,
discovered 30 years later, Abrikosov’s vortex
lattice actually ‘melts’ over an appreciable
range of magnetic field and temperature2,3.
Olson Reichhardt and Hastings4, writing in
Physical Review Letters, now propose a key
experiment that could unravel the physics of
these vortices as they become entangled in
the melted state.
A typical phase diagram for a high-tem-
perature superconductor is shown in Fig. 1,
overleaf, as a function of temperature and of
the magnetic field induced in the material5.
The Meissner phase of the diagram, in which
surface currents completely exclude an
applied magnetic field, follows the horizon-
tion in the superconductor) and terminates
at the temperature at which the material
ceases to be a superconductor — its transi-
tion temperature. According to Abrikosov’s
original theory, an applied magnetic field
would penetrate the material as a regular
lattice of quantized vortex filaments when
the magnetic induction is greater than zero
and below some critical field line (Fig. 1).
But thermal fluctuations melt this lattice
above the ‘coexistence region’ — where
liquid and solid phases coexist at slightly
different densities3.
What is this melted liquid of fluctuating,
randomly braided vortex filaments like? An
important property is its shear viscosity,
which determines the ease with which point-
like, linear or planar defects can pin the vor-
tices in place. These defects might be missing
atoms (for example, oxygen vacancies),
columnar damage tracks from heavy-ion
radiation, or more extended objects such as
grain boundaries or globules of a different
atomic phase. The pinning of vortex lines,
thereby immobilizing them relative to the
host material, is crucial in many high-field
©2004 Nature Publishing Group
news and views
applications. Indeed, if vortices are allowed
to move in response to an applied current
(which exerts a force on them), their motion
dissipates energy, and resistanceless flow in a
magnetic field is impossible. A vortex crystal
resists shear deformations, so that pinning
the Abrikosov flux lattice is relatively easy,
like tacking a carpet to a slippery floor. How-
ever, pinning the melted lattice in a vortex
liquid is much more difficult — this requires
a high viscosity, similar to the property that
allows us (temporarily, at least) to nail a
pancake of ‘silly putty’ to a wall. Silly putty is
composed of long, tangled polymer chains,
and it is tempting to think of a melted
liquid of Abrikosov vortices (Fig. 1) as a
related ‘directed polymer melt’ of entangled
vortex filaments. Crucial to the high viscos-
ity of real polymer melts, however, are large
energy barriers to the polymer chains cross-
ing each other.
Do vortices, which are singularities in the
underlying superconducting order, behave
as impenetrable lines, or do they cut freely
through one another? The answer depends
on field and temperature. The energy barri-
ers against flux cutting are very small near the
critical field line (Fig. 1), where quantized
vortices first form. However, when nearly
parallel vortex lines cross, the ‘quantum of
vorticity’ effectively doubles. The quantum
of vorticity is a universal combination of
Planck’s constant, the speed of light and the
electron charge, and determines the strength
of the swirling supercurrents around a
vortex. Well below the critical field line,
this doubling leads to a very large crossing
energy. Two entangled vortices can always
cut each other easily if the lines deform so as
to cross each other at a large angle, but this is
resisted by the line tension. Although there
is some experimental evidence for a large
viscosity in a vortex liquid, theoretical esti-
mates of the ratio of the melting temperature
and the crossing energy at that temperature
vary widely6,7.
Olson Reichhardt and Hastings4 propose
an elegant way to probe that all-important
vortex crossing energy, using a magnetic
force microscope (MFM).An MFM is similar
to an atomic force microscope, except that it
has a magnetized tip that pulls on the end of
an individual vortex line8. It should be possi-
ble to pin the entry points of two vortex
lines in place, and then use the MFM to
wind one vortex around another into a
double helix (Fig. 1). If the energy barriers to
839