A study of monoelectronic reduction of Artemisinin

Article abstract of DOI:10.1016/j.tetlet.2012.01.007

The reaction of Artemisinin with solvated electrons (e_solv ^-) generated by radiolytic methods led to the formation of acid 2 as the only detectable product. A possible mechanism leading to the formation of compound 2 is discussed.

Full text of DOI:10.1016/j.tetlet.2012.01.007

Tetrahedron Letters 53 (2012) 1296–1299
Contents lists available at SciVerse ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
A study of monoelectronic reduction of Artemisinin
⇑
Maria Luisa Navacchia, Massimo L. Capobianco, Mila D’Angelantonio, Giancarlo Marconi
CNR-Istituto per la Sintesi Organica e la Fotoreattività (ISOF), Via P. Gobetti 101, 40129 Bologna, Italy
a r t i c l e i n f o
a b s t r a c t
The reaction of Artemisinin with solvated electrons (eꢀsolv) generated by radiolytic methods led to the for-
mation of acid 2 as the only detectable product. A possible mechanism leading to the formation of com-
pound 2 is discussed.
Article history:
Received 17 November 2011
Revised 21 December 2011
Accepted 4 January 2012
Available online 10 January 2012
Ó 2012 Elsevier Ltd. All rights reserved.
Keywords:
Artemisinin
Malaria
c
-Radiolysis
Reduction
Alkoxyl radicals
Since its discovery in the late seventies of the last century,
Artemisinin (ART), a metabolic compound produced by the plant
Artemisia annua (Qinghao in Chinese), has imposed itself as the
most effective drug to fight malaria.1 Moreover, this molecule
and its direct derivatives have been found promising in the cure
of other serious diseases such as different kinds of solid cancer,2
schistosomiasis,3 and in the antiangiogenetic therapy.4 At the
moment, the cure for malaria suggested by WHO consists in a com-
bination of ART derivatives with some other drugs aimed at pro-
longing its therapeutic effect (the so called ACT therapy).5
Despite the number of studies devoted to clarify the action
mechanism of ART in vitro and in vivo, a univocal picture of its deg-
radation has not been yet obtained. ART (Fig. 1) is a sesquiterpenic
lactone that is conventionally divided into four rings. Among them,
the pharmaceutical active part is located in the ring A, a six mem-
bered trioxanic ring containing an endoperoxidic bond, and B, the
lactone ring, whereas the other two rings, that is the cyclohexane
and the oxacycloheptane seem to play only a minor role in its ac-
tion against the Plasmodium parasites. However not completely
understood, the mechanism of action of this drug seems to involve
free radicals generated during its catabolism. Specifically, induced
de novo production of reactive hydroxyl moieties and superoxides
within the malarial parasite has been reported to damage intracel-
lular processes and to cause death. One of the mechanisms pro-
posed involves a dissociative reduction of the peroxidic O–O
bond by a not yet identified reductive agent present in the blood
and, due the fact that the iron [Fe(II)] concentration is found to
be particularly high in the parasitic food vacuole, great attention
has been devoted to its involvement into the breaking of the per-
oxidic bond by direct electron transfer to the antibonding rr
*
LUMO localized on this part of the molecule.6 The two unstable dis-
tonic radicals oxyanions so generated, are claimed to undergo an
intramolecular rearrangement into C-centered radicals through
either a 1,5-hydrogen shift or a homolytic C–C bond cleavage. After
this stage, the destiny of the secondary or primary carbon radicals
produced by these intramolecular rearrangements is still not clear.
It was found that the ratio of products strongly depends on the
nature of the counterion (Clꢀ, SO42ꢀ) present in the solution which
contains iron as the main reducing agent.5 It was also found that
the presence of the carbonyl group, as well as the nature of substit-
uents on C-12 (see Scheme 2 for numbering) plays a decisive role
in orienting the deactivation mechanism of ART.7 An alternative
to the direct reduction is provided by the use of suitable photosen-
sitizers with energy close enough to that of the antibonding LUMO
of the endoperoxide, to allow a SET (Single Electron Transfer) lead-
ing to the formation of alkoxyl radicals.8 Beyond the experimental
studies, several theoretical investigations have been addressed to
O
D
A
C
O
O
H
H
B
O
O
1
⇑
Corresponding author. Tel.: +39 051 6399799; fax: +39 051 6399844.
E-mail address: marc@isof.cnr.it (G. Marconi).
Figure 1. ART.
0040-4039/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2012.01.007

M. L. Navacchia et al. / Tetrahedron Letters 53 (2012) 1296–1299
1297
Figure 2. Left: top, plot of the linear consumption of ART as a function of dose; bottom, radiation chemical yield (G) as a function of dose for the consumption of ART G
(ꢀART). Right: chromatogram profiles of irradiated mixtures (blue: dose = 0 Gy; red: dose = 2320 Gy; green: dose = 4780 Gy; pink: dose = 7080 Gy). HPLC analyses were
performed using a Zorbax C8 column (4.6 ꢂ 150 mm, 5
lm), linear gradient H2O/MeOH, flow rate 0.5 ml/min, detection at k 220 nm.
9
CH3 CHOH with itself, k = 1.5 ꢂ 10 Mꢀ1 sꢀ1, must be taken into
Å
the problem of identifying the most probable intermediates of the
ART reductions, most of them adopting only partial models of the
whole ART molecule.9,10 A review of these studies has also recently
appeared.11
account during the stationary irradiation.12
The absorbed radiation dose was determined with the Fricke
chemical dosimeter, by taking G (Fe3+) = 1.61 mol Jꢀ1 18
l . Samples
Our knowledge in radiolysis techniques prompted us to study
the reaction of solvated electrons (eꢀsolv) with ART by radiolytic
methods coupled with product studies.
were taken at different absorbed doses, analyzed by Reverse
Phase-High Performance Liquid Chromatography/Mass Spectrome-
try (RP-HPLC/MS) and characterized by 1H NMR after elimination
of the solvent under reduced pressure. Samples irradiated at dose
ca. 2320 and 7080 Gy were taken from repeated experiments and
used to improve the quantitative analysis. Quantitative analysis
for the consumption of ART was performed by RP-HPLC. The con-
centration values of ART taken from the different experiments re-
ported in Figure 2 as a function of dose were normalised at the
highest ART concentration used (3.8 mM).
The reaction of eꢀsolv with ART was previously studied by some of
us by pulse-radiolysis techniques in some detail.12 Experiments of
direct irradiation of a H2O/EtOH (1:1 v/v) solution of ART revealed
the initial formation of carbon centered radicals following the ra-
pid intramolecular rearrangement of the initially produced oxyan-
ion radicals in agreement with the reported data involving
chemical reductants such as Fe(II).13 However, in that case the pro-
posed mechanism was not supported by product analysis.
In the present communication we report preliminary results on
the gamma radiolysis of de-aerated solutions of ART (1) (ca.
3.5 mM, 15 mL) in H2O/EtOH (1:1 v/v) irradiated under station-
ary-state conditions, using a 60Co-Gammacell with a dose rate of
The G data, calculated at increasing absorbed doses, were ana-
lyzed as a function of dose and linearly extrapolated to zero dose
in order to obtain a G value unaffected by radiolytically induced
side reactions that can reduce the calculated radiolysis yields.17
In Figure 2 (left, top) is reported the linear consumption of ART
as a function of dose. In the same figure (left, bottom) G (ꢀART)
ꢁ6 Gy minꢀ1
.
The following equations describe the chemical processes occur-
ring during the irradiation:
was extrapolated to zero dose obtaining
0.27
mol Jꢀ1 which is a slightly higher value than that reported
in the literature for eꢀsolv 14 This finding indicates that only solvated
a value close to
l
.
þ
H2O=EtOH,eꢀsolv; CH3 CHOH; H3O =CH3CH2þOH2
ð1Þ
ð2Þ
ð3Þ
ð4Þ
Å
electrons are responsible for the reduction of ART and excludes the
ÅOH or ÅH þ EtOH ! CH3 CHOH þ H2O or H2
Å
Å
involvement of CH3 CHOH in the consumption of ART. In fact, the
ART þ eꢀsolv or CH3 CHOH ! radiolysis products
Å
Å
reaction of either eꢀsolv and CH3 CHOH with ART should have pro-
duced G = 0.54
CH3 CHOH yields.
l
mol Jꢀ1, that is the sum of solvated electrons and
Å
Å
CH3 CHOH þ CH3 CHOH ! products
14
Å
Radiolysis of neutral water/ethanol 1:1 v/v mixture leads to the
On this base Eq. 3 became:
+
species eꢀsolv (0.2), CH3 CHOH (0.34), H3O /CH3CH2+OH2 (0.2), Eq. 1,
Å
ART þ eꢀsolv ! radiolysis products
ð5Þ
14,15 where the values in parenthesis represent the radiation chem-
l . The dose absorbed by
mol Jꢀ1 16
The 1H NMR analysis performed on the reaction mixtures indi-
cates the presence of unreacted ART and only one reaction product
identified as the carboxylic acid 2. (Scheme 1). Compound 2 was
characterized after purification on RP column chromatography of
a large scale experiment reaction mixture by 1H NMR and MS
analyses.19
ical yields (G) expressed in
water/ethanol mixture was calculated by applying a correction fac-
tor of 0.7% to take into account the different electronic density of
ethanol and water.17 The use of water/ethanol mixtures for study-
ing the radiolysis of water-insoluble species of biological impor-
tance has been discussed in the past.14
The presence of EtOH as solvent (8.7 M) guarantees that
hydroxyl radicals and hydrogen atoms are scavenged efficiently
A quantitative analysis of the formation of acid 2 was performed
by 1H NMR of the reaction mixtures by comparing the integral of
H11 signal of 2 (d 2.85) with the integral of H11 signal of the unre-
acted ART 1 (d 3.10) and plotted in Figure 3.
Å
Å
(k
= 1.9 ꢂ 109 Mꢀ1 sꢀ1
,
and
k
= 1.7 ꢂ 107 Mꢀ1
ðEtOH þ OHÞ
ðEtOH þ OHÞ
sꢀ1), Eq. 2.
As reported in Eq. 4, the bimolecular reaction of
18

1298
M. L. Navacchia et al. / Tetrahedron Letters 53 (2012) 1296–1299
As reported in Figure 3 (left, top), the formation of acid 2 is lin-
ear with the dose. The analysis, in terms of G, of the data related to
the formation of the identified product (G(+Acid) = 0.097
mol Jꢀ1
shows that, comparing the G values at zero dose (Figs. 2 and 3 left,
bottom), ꢁ37% of reacted ART leads to the formation of this radiol-
ysis product.
O
l
)
e-
solv
1
O
H2O/EtOH
(ART)
OH
Acid 2 was previously reported in the literature as a minor
product obtained by the treatment of ART with agents such as
heme Iron(II) and (III).14
O
2
Scheme 1.
c-Radiolysis of ART.
Figure 3. Left: top, plot of the linear formation of acid 2 as a function of dose; bottom, radiation chemical yield (G) as a function of dose for the formation of acid 2 G (+acid).
Right: 1H NMR profiles of irradiated mixtures (blue: dose = 2320 Gy; red: dose = 4780 Gy; black: dose = 7080 Gy).
O
β
- scission
Path 1
O
2
3
HO
O
4
•O
9
5
6
H
O
1,6-HT
1
HO
e-solv
RO +H2
HO
O
O
(II)
7
O
O c
•O
1
10
H
H
8
O
13
12
O
11
O
O
b
a
•O
O
Path 2
β
- scission
HO
O
1,4-HT
(I)
- HO
X
H
O
O
e
O
O
d
OH
O
- H
O
O
R-O H
H
O
3
2
O
O
O
H
OR
R = H, CH3CH2
O
O
f
Scheme 2. Proposed mechanisms for the formation of carboxylic acid 2.

M. L. Navacchia et al. / Tetrahedron Letters 53 (2012) 1296–1299
1299
The formation of the detected product can be reasonably ex-
plained with the aid of the energy of the intermediates, quantum
mechanically calculated by Moles et al.10 Taking into account their
simplified model of ART, we propose the mechanism drawn in
Scheme 2. The initial distonic alkoxyl radical formed by the reduc-
tive cleavage of the peroxide bond is rapidly protonated to radical
a. As depicted in Scheme 2 radical a can follow, in principle, the
route (I) leading to the pharmacologically inactive deoxyartemisin
3 or the route (II), leading to the carboxylic acid 2. The product
analyses led us to exclude the route (I) since no deoxyartemisin
3 was found. Along the route (II), the first step of the decomposi-
tion of radical a starts with the scission of bond C3–O9 leading
to O9 radical b. For this process the calculated energy barrier is
around 14 kcal/mol, making this step particularly competitive with
respect to other possible rearrangement processes within this
route. From radical b to the final product f two different pathways
are possible both of them involving two steps. The first pathway
consists in breaking of the C7–C8 bond to give the secondary rad-
ical c, which in turn leads to radical e by intramolecular hydrogen
transfer between O1 and O9. The second one, requires an immedi-
ate hydrogen shift from O1 to O9 to give radical d, followed by
breaking of the C7–C8 bond leading to radical e. The accuracy of
the theoretical methods used to describe these reactions is proba-
bly not sufficient to choose between the two mechanisms. The
functional density B3LYP method in fact looks very favorable to
the former but does not allow to find a stationary point for the sec-
ond, while the HF/3–21G indicates a possible competition between
the two. The final step is the formation of the anhydride f through
the ÅH release from O9, analogously to what is supposed to happen
in the presence of Fe(II).6 The energies required to overcome the
transition states are 34.3, 24.7 kcal/mol for the processes b–d
and d–e, respectively, and 26.5, 37.2 kcal/mol for the routes b–c
and c–e, respectively, along the results of the HF/3–21G calcula-
tions.10 The step e–f to reach the final product requires 49 kcal/
mol. Beyond the reliable energies, the scheme proposed agrees
with the description of the spin densities calculated on the atoms
involved in the bond breakings and in the atom transfers, thus pro-
viding a further support to the mechanism proposed. It is worth
noting that product f could not be isolated nor detected due to
the reaction medium. It is well known in fact that anhydrides
hydrolyze in the presence of water leading to the corresponding
carboxylic acids and, as a consequence, it is not surprising that acid
2 is the only detected product.
In conclusion, we assessed that ART, under continuous c-radi-
olysis, reacts quantitatively with eꢀsolv leading to carboxylic acid 2
as the main and major reaction product. These preliminary results
make the highly debated and controversial degradation mecha-
nism of ART better understood supporting the route (II) over
the route (I).
References and notes
1. Li, G. Q.; Guo, X. B.; Arnold, K.; Jian, H. X.; Fu, L. C. Lancet 1984, 2, 1360–1361.
2. Du, J. H.; Zhang, H. D.; Ma, Z. J.; Ji, K. M. Cancer Chemother. Pharmacol. 2009, 65,
895–902.
3. Xiao, S. H. Acta Trop. 2005, 96, 153–167.
4. Chen, H. H.; Zhou, H. J.; Wu, G. D.; Lou, X. E. Pharmacology 2004, 71, 1–9.
5. World Health Organization. World Health Organization: Geneva, 2010.
6. O’Neill, P. M.; Posner, G. H. J. Med. Chem. 2004, 47, 2945–2964.
7. Wu, W. M.; Wu, Y. K.; Wu, Y. L.; Yao, Z. J.; Zhou, C. M.; Li, Y.; Shan, F. J. Am. Chem.
Soc. 1998, 120, 3316–3325.
8. Magri, D. C.; Donkers, R. L.; Workentin, M. S. J. Photochem. Photobiol., A 2001,
138, 29–34.
9. Gu, J. D.; Chen, K. X.; Jiang, H. L.; Leszczynski, J. J. Phys. Chem. A 1999, 103, 9364–
9369.
10. Moles, P.; Oliva, N.; Safont, V. S. J. Phys. Chem. A 2006, 110, 7144–7158.
11. Denisov, E. T.; Solodova, S. L.; Denisova, T. G. Russ. Chem. Rev. 2010, 79, 981–
1003.
12. Fuochi, P. G.; Marconi, G.; Mulazzani, Q. G. Int. J. Radiat. Biol. 2005, 81, 319–326.
13. Haynes, R. K.; Vonwiller, S. C. Tetrahedron Lett. 1996, 37, 253–256.
14. Jore, D.; Champion, B.; Kaouadji, N.; Jaygerin, J. P.; Ferradini, C. Radiat. Phys.
Chem. 1988, 32, 443–448.
15. Leu, A. D.; Jha, K. N.; Freeman, G. R. Can. J. Chem. 1983, 61, 1115–1119.
16. Buxton, G. V.; Mulazzani, Q. G. In Electron Transfer in Chemistry; Balzani, V., Ed.;
Wiley-VCH Verlag GmbH: Weinheim, 2001; pp 503–557.
17. Spinks, J. W. T.; Woods, R. J. Introduction to Radiation Chemistry, 3rd ed.; Wiley:
New York, 1990.
18. Buxton, G. V.; Greenstock, C. L.; Helman, W. P.; Ross, A. B. J. Phys. Chem. Ref.
Data 1988, 17, 513–886.
19. 1H NMR (400 MHz, CDCl3), d 11.90 (br s, COOH; disappeared on D2O shake),
2.85 (1H, m, H, collapsing to d upon irradiation at d 1.19, J = 5.6), 2.61 (1H, m,
H4, collapsing to br d upon irradiation at d 1.51), 2.54 (1H, m, H1), 2.54 (1H, m,
H1), 2.39 (1H, m, H6), 2.12 3H, s, CH3C), 1.90 (1H, m, H3, collapsing to br s upon
irradiation at d 1.51), 1.84–1.70 (2H, m, H2–H5), 1.51 (1H, m, H), 1.19 (3H, d,
J = 6.8, CH3, collapsing to s upon irradiation at d 2.85), 1.06 (3H, d, J = 6.0, CH3,
collapsing to s upon irradiation at d 1.51). Calculated mass (C14H22O4): 254.15
found ESI-MS+: 255 [M+H]+; 277 [M+Na]+; 531 [2 ꢂ M+Na]+; found ESI-MSꢀ:
253 [MꢀH]ꢀ.