Synthesis, antimalarial activity and cytotoxicity of 10-aminoethylether derivatives of artemisinin

Article abstract of DOI:10.1016/j.bmc.2012.06.014

In this study, a series of 11 10-aminoethylether derivatives of artemisinin were synthesised and their antimalarial activity against both the chloroquine sensitive (D10) and resistant (Dd2) strains of Plasmodium falciparum was determined. The compounds were prepared by introducing aliphatic, alicyclic and aromatic amine groups with linkers of various chain lengths through an ethyl ether bridge at C-10 of artemisinin using conventional and microwave assisted syntheses, and their structures were confirmed by NMR and HRMS. All derivatives proved to be active against both strains of the parasite. The highest overall activity was displayed by the short chain aromatic derivative 8 (IC₅₀ = 1.44 nM), containing only one nitrogen atom, while long chain polyamine derivatives were found to have the lowest activity against both strains. An interesting correlation between the IC₅₀, pK_a values and resistance index (RI) was found.

Full text of DOI:10.1016/j.bmc.2012.06.014

Bioorganic & Medicinal Chemistry 20 (2012) 4701–4709
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Synthesis, antimalarial activity and cytotoxicity of 10-aminoethylether
derivatives of artemisinin
Theunis T. Cloete a, J. Wilma Breytenbach b, Carmen de Kock c, Peter J. Smith c, Jaco C. Breytenbach a,
David D. N’Da a,
⇑
a Department of Pharmaceutical Chemistry, North-West University, Potchefstroom 2520, South Africa
b Statistical Consultation Services, North-West University, Potchefstroom 2520, South Africa
c Department of Pharmacology, University of Cape Town, Groote Schuur Hospital, Observatory 7925, South Africa
a r t i c l e i n f o
a b s t r a c t
Article history:
In this study, a series of 11 10-aminoethylether derivatives of artemisinin were synthesised and their
antimalarial activity against both the chloroquine sensitive (D10) and resistant (Dd2) strains of Plasmo-
dium falciparum was determined. The compounds were prepared by introducing aliphatic, alicyclic and
aromatic amine groups with linkers of various chain lengths through an ethyl ether bridge at C-10 of
artemisinin using conventional and microwave assisted syntheses, and their structures were confirmed
by NMR and HRMS. All derivatives proved to be active against both strains of the parasite. The highest
overall activity was displayed by the short chain aromatic derivative 8 (IC50 = 1.44 nM), containing only
one nitrogen atom, while long chain polyamine derivatives were found to have the lowest activity against
both strains. An interesting correlation between the IC50, pKa values and resistance index (RI) was found.
Ó 2012 Elsevier Ltd. All rights reserved.
Received 4 April 2012
Revised 29 May 2012
Accepted 5 June 2012
Available online 15 June 2012
Keywords:
Artemisinin
Plasmodium falciparum
Microwave
Malaria
1. Introduction
mostly used in artemisinin-based combination therapies (ACTs)
which entail combining a semi-synthetic artemisinin derivative
Malaria is a protozoan infection transmitted to humans by the
bite of an infected female anopheline mosquito. It is estimated that
malaria kills about 655,000 people each year, 91% of whom are liv-
ing in Africa and, most of them children under the age of 5 years.1
An increase in resistance against most of the drugs currently
used against malaria has led to the dramatic increase in the global
death toll. Chloroquine (CQ) and other synthetic quinoline
compounds has been the mainstay of malaria chemotherapy for
five decades, but resistance against these and the antifolate anti-
malarials have spread rapidly making these drugs obsolete in most
of the areas afflicted by this disease.2,3
with another drug of a different chemical class. These ACTs do
not only compensate for artemisinin’s poor pharmacokinetic prop-
erties, but also decrease the likelihood that resistance against this
class would emerge. Despite these efforts, resistance against
artesunate has already been reported at the Thai-Cambodian and
Thai-Myanmar borders, where significantly prolonged in vivo par-
asite clearance times have been observed.8,9 This is a stark remin-
der of the exceptional ability of the malaria parasite to acquire
resistance against a huge variety of chemical compounds, and a
realisation that the loss of this important drug class would have
dire consequences for millions of people around the world.
The artemisinin class of compounds (Fig. 1) are currently the ba-
sis of treatment preferred by the World Health Organization.4 Arte-
misinin, a trioxane endoperoxide compound, is one of the most
rapidly acting antimalarials with the broadest effective range but
is however poorly soluble in both oil and water. In an effort to
overcome this solubility problem, the first generation analogs of
artemisinin, viz. artemether, arteether and sodium artesunate,
were synthesised. Unfortunately these derivatives have relatively
short elimination half-lives creating an elevated risk of high recru-
descence rates.5–7
Polyamine compounds have been found to have implications in
a great number of various processes in the malaria parasite. The
natural occurring polyamines spermidine, putrescine and sperm-
ine have an important function in the regulation of growth and
differentiation in an array of cell types.10 These polyamine
compounds, existing as polycations in vivo, are taken up by the
malaria parasite through a polyamine transport system that recog-
nises specific point charges on these compounds and then actively
transports them into the malarial cells. The rapidly growing malar-
ia parasite needs a substantial amount of these polyamine com-
pounds and is reliant on exogenous sources of polyamines
together with the polyamine transport system to provide it with
the necessary quantities.11 Researchers have proposed that a moi-
ety that has the capability of being recognised by the polyamine
transport system could act as a vector when coupled to another
In an effort to prevent the artemisinin class of compounds from
suffering the same fate as the classic antimalarial drugs, they are
⇑
Corresponding author. Tel.: +27 18 299 2516; fax: +27 18 299 4243.
E-mail address: david.nda@nwu.ac.za (D.D. N’Da).
0968-0896/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.bmc.2012.06.014

4702
T. T. Cloete et al. / Bioorg. Med. Chem. 20 (2012) 4701–4709
(THF), 1-(3-aminopropyl)imidazole, N-phenylethylenediamine,
piperazine and toluene were purchased from Sigma–Aldrich
(Johannesburg, South Africa). Oxalic acid was purchased from
BDH laboratory reagents (Yorkshire, England). Furfurylamine was
purchased from Fluka (Johannesburg, South Africa). Isobutylamine
was purchased from Merck (Johannesburg, South Africa). All the
chemicals and reagents were of analytical grade. Chemicals were
used without further purification.
O
O
O
O
O
O
O
O
O
OR
Artemisinin
Dihydroartemisinin R = H
Artemether R = Me
Arteether R = Et
2.2. General procedures
The 1H and 13C NMR spectra were recorded on a Bruker
Avance™ III 600 spectrometer at a frequency of 600.17 and
150.913 MHz, respectively, in deuterated dimethyl sulfoxide
(DMSO-d6) or deuterated CDCl3. Chemical shifts are reported in
parts per million d (ppm) with the residual protons of the solvent
as reference. The splitting pattern abbreviations are as follows: s
(singlet), d (doublet), dd (doublet of doublet), ddd (doublet of dou-
blet of doublets), dt (doublet of triplets), dq (doublet of quartets), t
(triplet), td (triplet of doublets), tt (triplet of triplets), qd (quartet of
doublets), m (multiplet).
Artesunate R = COCH2CH2COONa
Figure 1. Artemisinin class of compounds.
drug molecule, increasing the concentration and the potency of
that given drug. Chadwick and co-workers11 have attached spermi-
dine and Boc-protected spermidine to artemisinin (ART) and have
observed disappointing activity in the former but potent activity in
the latter. They assigned the perceived disappointing activity to
either poor membrane permeability of the ART-spermidine
compounds, insufficient accumulation in the food vacuole of the
parasite, or failure of the compounds to be recognised by the poly-
amine transport system.
Mass spectra (MS) were recorded in positive mode on a Thermo
Electron LXQ™ ion trap mass spectrometer. The MS had an APCI
source set at 300 °C with Xcalibur 2.2 data acquisition and analysis
software and utilised direct infusion with a Harvard syringe pump
Calas et al.12 have proposed a second mechanism of action by
which their primary, secondary, tertiary and quaternary amino
compounds, with either one or two amino groups, had substantial
activity against the malaria parasite. These compounds are ana-
logues of choline, ethanolamine, serine and other polar head
compounds that are essential for phospholipid biosynthesis, a pro-
cess necessary for membrane synthesis in the malaria parasite. By
competing with these naturally occurring polar head compounds,
Calas et al.’ compounds were able to block the synthesis of phos-
pholipids, killing the parasite.12,13 In their study, the length of
the chain between the two amino groups, the degree of substitu-
tion and the number of lipophillic moieties around the amines
had a substantial influence on the activity.14
The third mechanism of action suggested to explain the activity
of amino compounds against malaria is ion trapping. The basis of
this mechanism relies on the concept that amino groups in certain
drugs, for example, chloroquine, act as weak bases. In the normal
in vivo environment (pH 7.4), these amines are neutral and unprot-
onated thus having the capability to cross lipophillic membranes
easily. However, when these unprotonated drug molecules reach
the acidic (pH 4.5–5.5) food vacuole of the malaria parasite, they
become protonated, making them less membrane permeable. This
results in their accumulation in the digestive vacuole and conse-
quently increases the potency of the given drug.15 Previous studies
designed to increase the concentration of artemisinin by ion trap-
ping resulted in a modest increase in antimalarial activity.16,17
The compounds described in this article are also amine deriva-
tives of artemisinin and with different lipophillic moieties, various
substituents and with varying chain lengths separating the amino
groups. Such compounds may serve as possible drugs in the fight
against malaria.
at a flow rate of 10
tained in 1 s, with a capillary voltage of 7 V while the corona dis-
charge was 10 A.
lL/min. A full scan from 100–1200 amu was ob-
l
Melting points (Mp) were determined with a BÜCHI melting
point B-545 instrument and were reported uncorrected. Thin layer
chromatography was performed using silica gel plates (60F254).
2.3. Biological testing
2.3.1. Antiplasmodial assay
The samples were tested in triplicate on one occasion against
the chloroquine sensitive (CQS) strain of Plasmodium falciparum
(D10), and on another occasion against the chloroquine resistant
(CQR) strain of Plasmodium falciparum (Dd2). Continuous in vitro
cultures of asexual erythrocyte stages of P. falciparum were
maintained using a modified method of Trager and Jensen.18
Quantitative assessment of antiplasmodial activity in vitro was
determined via the parasite lactate dehydrogenase assay using a
modified method.19 The test samples were prepared to a 20 mg/
ml stock solution in 100% DMSO and sonicated to enhance solubil-
ity. Samples were tested as a suspension if not completely dis-
solved. Stock solutions were stored at ꢁ20 °C. Further dilutions
were prepared on the day of the experiment. Sodium artesunate
(ArtNa), dihydroartemisinin (DHA) and chloroquine (CQ) were
used as the reference drugs in all experiments. A full dose-response
was performed for all compounds to determine the concentration
inhibiting 50% of parasite growth (IC50 value). Samples were tested
at a starting concentration of 100 ng/ml, which was then serially
diluted 2-fold in complete medium to give 10 concentrations; with
the lowest concentration being 0.2 ng/ml. The same dilution tech-
nique was used for all samples. The highest concentration of
solvent to which the parasites were exposed to had no measurable
effect on the parasite viability (data not shown). The IC50 values
were obtained using a nonlinear dose-response curve fitting anal-
ysis via Graph Pad Prism v.4.0 software.
2. Materials and methods
2.1. Materials
Dihydroartemisinin (mixture of 10-a and 10-b epimers) was
purchased from Changzhou Kaixuan Chemical Co (Chunjiang,
China). Boron trifluoride diethyl etherate (BF3ꢀEt2O), 2-bromoetha-
nol, N,N-dimethylformamide (DMF), morpholine, 4-(2-aminoethyl)
morpholine, piperidine, 1-(2-aminoethyl)piperidine, 2-(diisopro-
pylamino)ethylamine, N-benzylmethylamine, tetrahydrofuran
2.3.2. Cytotoxicity assay
The MTT-assay was used as a colorimetric assay for cellular
growth and survival, and compares well with other available
assays.20,21 The tetrazolium salt MTT was used to measure all
growth and chemosensitivity. The samples were tested in triplicate

T. T. Cloete et al. / Bioorg. Med. Chem. 20 (2012) 4701–4709
4703
on one occasion against the mammalian cell-line, Chinese hamster
ovarian (CHO). The same stock solutions used for antiplasmodial
testing were used for cytotoxicity testing. Dilutions were prepared
on the day of the experiment. Emetine was used as the reference
drug in all experiments. The initial concentration of emetine was
(ddd, J = 14.2, 7.5, 3.6 Hz, 1H, H-8b), 1.65–1.55 (m, 1H, H-7b),
1.52–1.37 (m, 5H, H-8a, H-5(b), H-14), 1.36–1.28 (m, 1H, H-6),
1.25–1.17 (m, 1H, H-5a), 0.89 (m, 7H, H-7(a
), H-15, H-16). 13C
NMR (151 MHz, CDCl3) d 104.10 (C-3), 102.02 (C-10), 88.12
(C-12), 81.02 (C-12a), 68.15 (C-a), 52.55 (C-5a), 44.33 (C-8a),
37.37 (C-6), 36.37 (C-4), 34.63 (C-7), 31.41 (C-b), 30.83 (C-9),
26.12 (C-14), 24.62 (C-5), 24.34(C-8), 20.31(C-15), 12.91(C-16).
HRMS m/z [M+H]+: 392.14 (Calcd for C21H35NO6: 391.130).
100
10-fold dilutions to give 6 concentrations, the lowest being
0.001 g/ml. The same dilution technique was applied to all test
lg/ml, which was serially diluted in complete medium with
l
samples. The highest concentration of solvent to which the cells
were exposed to had no measurable effect on the cell viability
(data not shown). The 50% inhibitory concentration (IC50) values
were obtained from full dose–response curves, using a non-linear
dose–response curve fitting analysis via Graph Pad Prism v.4.0
software.
2.5.2. General procedure for the coupling of a primary/
secondary amine to 1
The pure compound 1 was dissolved in DMF, THF or toluene
together with an excess of a given amine. The reaction mixture
was either heated in an oil bath (50–60 °C) or a microwave reactor
(in bursts of 60 watts and 40 °C for 4 min) until the reaction was
complete (Scheme 2). The reaction was stopped by firstly removing
the solvent in vacuo and then extracting it with DCM and water.
The DCM phase was then concentrated and the resulting oil was
purified by column chromatography eluting with various ratios
of MeOH, DCM, EtOAc and NH4OH.
After purification the product was converted to the correspond-
ing oxalate salt by stirring the obtained oil/powder with a solution
of oxalic acid dissolved in diethyl ether (1 g/100 ml). The resulting
precipitate was filtered to deliver the product as a white/off white
powder.
2.4. Statistical methods
2.4.1. Statistical analysis of antiplasmodial activity
One-way analyses of variances (ANOVA) were done to deter-
mine if significant differences existed between the means of the
test compounds and each of artesunate, DHA and CQ in general.
These were done for data on the D10 and Dd2 strains of Plasmo-
dium falciparum. Levene’s tests were performed to assure equality
of variances in each ANOVA’s case. In the case of inequality of
variances Welch tests were performed. Normal probability plots
on the residuals were done to assure that the data was fairly nor-
mally distributed.22 Dunnett’s tests were done to determine which
of the test compounds’ means differ statistically significantly from
the mean of the standard compounds. Tukey’s post hoc test was
done to determine which of the test compounds’ means differ sta-
tistically significantly from each other, omitting the reference
drugs from the analysis. These procedures were done using the
statistical data analysis software system.23 All tests were done at
a 0.05 significant level.
2.5.2.1. 1-(Morpholin-4-yl)-2-(10b-dihydroartemisinoxy)ethane
oxalate, 2.
Compound 1 (2.6 mmol, 1 g) was dissolved in
20 ml DMF and reacted with morpholine (10.2 mmol, 0.893 ml,
4 equiv) in an oil bath at 55 °C. Purification by column chromatog-
raphy (silica gel, MeOH/DCM 1:4) gave a white powder. The con-
version into the oxalate salt afforded 0.6 g (48%) of 2 as an off
white powder, mp 171.0 °C; C21H35NO6. 1H NMR (600 MHz, DMSO)
d 5.40 (s, 1H, H-12), 4.71 (d, J = 3.3 Hz, 1H, H-10), 3.93–3.85 (m, 1H,
H-a
H-ab), 3.07–2.80 (m, 6H, H-b, H-20), 2.40 (dt, J = 7.8, 5.7 Hz, 1H,
H-9), 2.17 (td, J = 14.1, 3.8 Hz, 1H, H-4 ), 2.03–1.93 (m, 1H, H-
4b), 1.86–1.76 (m, 1H, H-8 ), 1.71 (qd, J = 13.5, 3.1 Hz, 1H, H-5 ),
a
), 3.68 (d, J = 46.5 Hz, 4H, H-10), 3.59 (dt, J = 10.4, 5.4 Hz, 1H,
2.4.2. Statistical analysis of cytotoxicity
The statistical methods utilised to determine the cytotoxicity of
the test compounds against CHO were similar to the methods used
for antiplasmodial testing (see Section 2.4.1), using emetine as
reference drug.
a
a
a
1.66–1.59 (m, 1H, H-8b), 1.54 (dd, J = 13.0, 3.0 Hz, 1H, H-7b),
1.42–1.30 (m, 3H, H-5b, H-6, H-8a), 1.27 (s, 3H, H-14), 1.13 (td,
J = 11.4, 6.7 Hz, 1H, H-5a), 0.84 (ddd, J = 36.5, 21.4, 6.4 Hz, 7H,
2.5. Synthesis
H-7a
, H-15, H-16). 13C NMR (151 MHz, DMSO) d 103.44 (C-3),
100.91 (C-10), 87.07 (C-12), 80.55 (C-12a), 64.31 (C-10), 62.89
(C-a), 56.17 (C-b), 52.32 (C-20), 52.05 (C-5a), 43.66 (C-8a), 36.49
(C-6), 35.67 (C-4), 34.10 (C-7), 30.44 (C-9), 25.49 (C-14), 24.22
(C-5), 23.77 (C-8), 20.09 (C-15), 12.76 (C-16). HRMS m/z [M+H]+:
398.2545 (Calcd for C21H35NO6: 397.246438).
2.5.1. 2-Bromo-(10b-dihydroartemisinoxy) ethane, 1
The 1-bromo-2-(10b-dihydroartemisinoxy)ethane analogue 1
of DHA was synthesised by a method similar to that of Li and co-
workers24 (Scheme 1).
Mp 150.7 °C; C17H27BrO5. 1H NMR (600 MHz, CDCl3) d 5.46 5.46
(s, 1H, H-12), 4.82 (d, J = 3.4 Hz, 1H, H-10), 4.13 (dt, J = 10.8, 5.4 Hz,
2.5.2.2. 1-{[2-(Morpholin-4-yl)ethyl]amino}-2-(10-dihydroarte-
1H, H-a
a), 3.76 (dt, J = 10.9, 5.3 Hz, 1H, H-ab), 3.53 (t, 2H, H-b), 2.66
misinoxy)ethane oxalate, 3.
Compound 1 (1.3 mmol, 0.5 g)
(dt, J = 7.7, 5.7 Hz, 1H, H-9), 2.33 (tt, J = 20.2, 10.0 Hz, 1H, H-4
a
),
was dissolved in 3 ml DMF and reacted with 4-(2-aminoethyl)
morpholine (5.1 mmol, 0.670 ml, 4 equiv) in a microwave reactor
2.04–1.97 (m, 1H, H-4b), 1.91–1.80 (m, 2H, H-8(
a
), H-5
a
), 1.73
15
15
5
4
5
4
14
5a
14
5a
O2
O2
3
6
3
6
7
8
7
8
DCM
1
O
O1
BF3.Et2O
8a
8a
13O 12
1O1
13O
12a
10
12a
10
b
12
HO
+
0OC-r.t, 6-48h
9
Br
9
a
1O1
16
16
b
OH
DHA
O
Br
a
1
Scheme 1. Etheration of DHA using boron trifluoride etherate.

4704
T. T. Cloete et al. / Bioorg. Med. Chem. 20 (2012) 4701–4709
15
6
15
6
5
4
5
4
14
5a
14
5a
O2
O2
3
3
7
8
7
8
1
O1
O
8a
8a
13O 12
13O
12a
10
DMF or THF
or Toluene
12a
10
12
9
9
1O1
1O1
16
+
amine
16
50-60OC or
60 watts, 40OC
b
b
O
O
Br
R
a
a
2 - 12
______________________________
Compound
1
______________________________
Compound
R
R
1'
2'
_____________________________
______________________________
1'
2'
3'
8
2
N
N
O
1'
2'
1'
4'
5'
2'
3'
3'
4'
5'
2'
3'
HN
3
9
HN
N
4'
2'
2'
N
N
1'
6'
O
3'
4'
H
N
1'
3'
1'
4'
5'
2'
3'
HN
N
10
11
N
4
5
1'
3'
4'
2'
1'
2'
2'
2'
3'
HN
4'
5'
N
1'
NH
1'
3'
4'
4'
1'
3'
HN
12
2'
3'
2'
3'
6
7
HN
HN
N
4'
1'
1'
3'
O
4'
4'
4'
2'
3'
5'
Scheme 2. Synthesis of aminoethylether derivatives of artemisinin.
in bursts of 60 watts and 40 °C for 4 minutes at a time. Purification
by column chromatography (silica gel, MeOH) gave a white pow-
der. The conversion into the oxalate salt afforded 0.4 g (70 %) of
3 as an off white powder, mp 164.9 °C; C23H40N2O6. 1H NMR
(600 MHz, CDCl3) d 5.37 (s, 1H, H-12), 4.78 (d, J = 3.2 Hz, 1H, H-
20 ml DMF and reacted with piperidine (10.2 mmol, 1.01 ml,
4 equiv) in an oil bath at 55 °C. Purification by column chromatog-
raphy (silica gel, MeOH/DCM 1:4) gave a white powder. The con-
version into the oxalate salt afforded 0.968 g (76 %) of 4 as an off
white powder, mp 166.7 °C; C22H37NO5. 1H NMR (600 MHz, DMSO)
d 5.39 (s, 1H, H-12), 4.72 (d, J = 3.4 Hz, 1H, H-10), 4.00–3.91 (m, 1H,
10), 3.97 (dt, J = 10.7, 5.5 Hz, 1H, H-aa
), 3.70 (m, 4H, H-30), 3.48
(dt, J = 10.5, 5.3 Hz, 1H. H-ab), 2.80–2.73 (m, 2H, H-10), 2.69 (t,
J = 14.3, 8.3 Hz, 2H, H-b), 2.60 (s, 1H, H-9), 2.50–2.44 (m, 2H,
H-a
H-10), 2.45–2.36 (m, 1H, H-9), 2.17 (td, J = 14.1, 3.8 Hz, 1H, H-4
2.03–1.94 (m, 1H, H-4b), 1.85–1.76 (m, 1H, H-2’, H-8 ), 1.74–
1.58 (m, 6H, H-20, H-5 , H-8b0), 1.55–1.43 (m, 3H, H-30, H-7b),
1.35 (ddd, J = 18.2, 11.3, 5.0 Hz, 3H, H-5b, H-6, H-8a), 1.28 (s, 3H,
H-14), 1.19–1.02 (m, 1H, H-5a), 0.99–0.63 (m, 7H, H-7 , H-15, H-
a
), 3.69–3.58 (m, 1H,H-ab), 3.16 (dd, J = 46.2, 20.7 Hz, 5H, H-b,
a),
H-2’), 2.41 (s, 4H, H-40), 2.32 (dt, J = 33.2, 19.8 Hz, 1H, H-4
a
),
), 1.78–
), 1.59 (dd, J = 13.2, 3.1 Hz, 1H, H-7b),
1.51–1.36 (m, 5H, H-8a, H-5(b), H-14), 1.30 (dd, J = 12.4, 8.6 Hz,
1H, H-6), 1.25–1.14 (m, 1H, H-5a), 0.96–0.68 (m, 7H, H-7 , H-15,
a
2.03–1.90 (m, 1H, 4b), 1.83 (dt, J = 25.9, 13.3 Hz, 1H, H-8
1.66 (m, 2H, H-8(b), H-5
a
a
a
a
a
16). 13C NMR (151 MHz, DMSO) d 103.39 (C-3), 100.78 (C-10),
87.00 (C-12), 80.47 (C-12a), 62.53 (C-a), 55.41 (C-b), 52.75 (C-10),
52.03 (C-5a), 43.66 (C-8a), 36.48 (C-6), 35.82 (C-4), 34.15 (C-7),
30.28 (C-9), 25.49 (C-14), 24.22 (C-5), 23.91 (C-8), 22.63 (C-20),
21.52 (C-30), 20.14 (C-15), 12.69 (C-16). HRMS m/z [M+H]+:
396.2747 (Calcd for C22H37NO5: 395.267174).
H-16). 13C NMR (151 MHz, CDCl3) d 104.06 (C-3), 102.07 (C-10),
87.83 (C-12), 81.00 (C-12a), 67.86 (C-a), 66.99 (C-30), 58.35 (C-20),
53.72 (C-5a), 52.50 (C-40), 49.37 (C-10), 45.96 (C-b), 44.28 (C-8a),
37.41 (C-6), 36.37 (C-4), 34.60 (C-7), 30.84 (C-9), 26.14 (C-14),
24.64 (C-5), 24.52 (C-8), 20.30 (C-15), 13.08 (C-16). HRMS m/z
[M+H]+: 441.2958 (Calcd for C23H40N2O6: 440.288637).
2.5.2.4. 1-{[2-(Piperidin-1-yl)ethyl]amino}-2-(10-dihydroarte-
2.5.2.3.
1-(piperidin-1-yl)-2-(10-dihydroartemisinoxy)ethane
misinoxy)ethane oxalate, 5.
Compound 1 (1.3 mmol, 0.5 g)
oxalate, 4.
Compound 1 (2.6 mmol, 1 g) was dissolved in
was dissolved in 3 ml DMF and reacted with 1-(2-amino-

T. T. Cloete et al. / Bioorg. Med. Chem. 20 (2012) 4701–4709
4705
ethyl)piperidine (5.1 mmol, 0.7257 ml, 4 equiv) in a microwave
reactor in bursts of 60 watts and 40 °C for 4 minutes at a time. Puri-
fication by column chromatography (silica gel, MeOH/DCM/NH4
1:18:1) gave a white powder. The conversion into the oxalate salt
afforded 0162 g (23 %) of 5 as an off white powder, mp 160.4 °C;
2.5.2.7. 1-[Benzyl(methyl)amino]-2-(10-dihydroartemisinoxy)
ethane oxalate, 8. Compound 1 (2.6 mmol, 1 g) was dissolved
in 20 ml THF and reacted with N-benzylmethylamine (10.2 mmol,
1.3 ml, 4 equiv) in an oil bath at 55 °C. Purification by column chro-
matography (silica gel, EtOAc) gave a colourless oil. The conversion
into the oxalate salt afforded 0.577 g (43 %) of 8 as an off white
powder, mp 137.1 °C; C25H37NO5. 1H NMR (600 MHz, DMSO) d
7.50–7.32 (m, 5H, H-30,H-40, H-50), 5.37 (s, 1H, H-12), 4.73 (d,
J = 3.3 Hz, 1H, H-10), 4.09 (d, J = 29.8 Hz, 2H, H-10), 3.98–3.87 (m,
C
24H42N2O5. 1H NMR (600 MHz, DMSO) d 5.39 (s, 1H, H-12), 4.71
(d, J = 3.3 Hz, 1H, H-10), 3.90 (dt, J = 10.4, 5.1 Hz, 1H, H-aa), 3.61–
3.52 (m, 1H, H-ab), 3.27 (d, J = 25.0 Hz, 2H, H-10), 3.16 (d,
J = 5.8 Hz, 2H, H-b), 3.04 (s, 2H, H-20), 2.82 (d, J = 62.3 Hz, 4H, H-
30), 2.43–2.34 (m, 1H, H-9), 2.17 (td, J = 14.1, 3.8 Hz, 1H, H-4
a
),
), 1.67 (d,
), H-400), 1.57–1.50 (m, 1H, H-7b),
1.46 (s, 2H, H-50), 1.41–1.31 (m, 3H, H-8a, H-6), 1.28 (s, 3H, H-
14), 1.13 (td, J = 11.4, 6.7 Hz, 1H, H-5a), 0.94–0.76 (m, 7H, H-7( ),
1H, H-a
H-60), 2.40 (dt, J = 14.9, 5.7 Hz, 1H, H-9), 2.17 (td, J = 14.1, 3.8 Hz,
1H, H-4 ), 2.02–1.93 (m, 1H, H-4b), 1.84–1.76 (m, 1H, H-8 ),
1.74–1.57 (m, 2H, H-5 , H-8b), 1.52 (dd, J = 13.0, 2.8 Hz, 1H, H-
7b), 1.40–1.23 (m, 6H, H-5b, H-6, H-8a), 1.18–1.04 (m, 1H, H-5a),
0.95–0.76 (m, 8H, H-7
, H-15, H-16). 13C NMR (151 MHz, DMSO)
a), 3.67–3.58 (m, 1H, H-ab), 3.08 (s, 2H, H-b), 2.51 (s, 3H,
2.02–1.94 (m, 1H, H-4b), 1.84–1.76 (m, 1H, H-8
a
J = 24.9 Hz, 6H, H-8(b), H-5(
a
a
a
a
a
H-15, H-16). 13C NMR (151 MHz, DMSO) d 103.35 (C-3), 101.06
(C-10), 86.99 (C-12), 80.49 (C-12a), 63.50 (C-a), 53.00 (C-30),
52.73 (C-20), 51.98 C-5a), 46.43 (C-b), 43.74 (C-8a), 42.34 (C-10),
36.48 (C-6), 36.01 (C-4), 34.10 (C-7), 30.37 (C-9), 25.62 (C-14),
24.31 (C-40), 23.90 (C-5’), 23.48 (C-8), 22.17 (C-5), 20.13 (C-15),
12.60 (C-16). HRMS m/z [M+H]+: 439.3166 (Calcd for C24H42N2O5:
438.309373).
a
d 130.27 (C-20), 128.53 (C-30), 128.38 (C-40), 128.23 (C-50), 103.29
(C-3), 100.87 (C-10), 86.99 (C-12), 80.45 (C-12a), 63.34 (C-a),
59.75 (C-10), 55.02 (C-b), 51.88 (C-5a), 43.65 (C-8a), 40.48 (C-60),
36.52 (C-6), 35.78 (C-4), 33.99 (C-7), 30.20(C-9), 25.54 (C-14),
24.20 (C-5), 23.76 (C-8), 19.97 (C-15), 12.54 (C-16). HRMS m/z
[M+H]+: 432.2746 (Calcd for C25H37NO5: 431.267174).
2.5.2.8. 1-{[3-(1H-Imidazol-1-yl)propyl]amino}-2-(10-dihydro-
2.5.2.5. 1-({2-[Bis(propan-2-yl)amino]ethyl}amino)-2-(10-dihy-
artemisinoxy)ethane oxalate, 9.
Compound 1 (1.3 mmol,
droartemisinoxy)ethane oxalate, 6.
Compound 1 (2.6 mmol,
0.5 g) was dissolved in 3 ml DMF and reacted with 1-(3-aminopro-
pyl)imidazole (6.4 mmol, 0.766 ml, 5 equiv) in a microwave reactor
in bursts of 60 watts and 40 °C for 4 min at a time. Purification by
column chromatography (silica gel, MeOH/DCM/NH4 1: 13: 1) gave
a yellow oil. The conversion into the oxalate salt afforded 0.297 g
(44 %) of 9 as a off white powder, mp 148.2 °C; C23H37N3O5. 1H
NMR (600 MHz, DMSO) d 8.16 (d, J = 13.3 Hz, 1H, H-60), 7.37 (s,
1H, H-50), 7.16 (s, 1H, H-40), 5.38 (s, 1H, H-12), 4.69 (d, J = 3.4 Hz,
1H, H-10), 4.13 (t, J = 6.8 Hz, 2H, H-10), 3.89 (dt, J = 10.8, 5.3 Hz,
1 g) was dissolved in 15 ml DMF and reacted with 2-(diisopropyl-
amino) ethylamine (12.8 mmol, 2.2 ml, 5 equiv) in a microwave
reactor in bursts of 60 watts and 40 °C for 4 minutes at a time. Puri-
fication by column chromatography (silica gel, MeOH/DCM/NH4
1.5:13:0.5) gave a brown oil. The conversion into the oxalate salt
afforded 0.462 g (33 %) of 6 as a white powder, mp 139.4 °C;
C
25H46N2O5. 1H NMR (600 MHz, DMSO) d 5.38 (s, 1H, H-12), 4.71
(s, 1H, H-10), 3.88 (s, 1H, H-a
4H, H-b, H-40), 2.87 (d, J = 6.1 Hz, 2H, H-10), 2.67 (s, 2H, H-20),
2.36 - 2.45 (m, 1H, H-9), 2.17 (t, J = 13.5 Hz, 1H, H-4 ), 1.99 (d,
J = 13.8 Hz, 1H, H-4b), 1.79 (s, 1H, H-5 ), 1.74–1.58 (m, 2H, H-8
a), 3.54 (s, 1H, H-ab), 3.12–3.00 (m,
1H, H-a
2.93–2.81 (m, 2H, H-30), 2.41–2.32 (m, 1H, H-9), 2.22–2.03 (m,
3H,H-4
), H-20), 2.02–1.92 (m, 1H, H-4b), 1.82–1.74 (m, 1H, H-
), 1.69–1.54 (m, 2H, H-8b), H-5 ), 1.52 (dd, J = 12.9, 2.9 Hz, 1H,
H-7b), 1.43–1.25 (m, 6H, H-5b), H-6, H-8a, H-14), 1.15–1.00 (m,
1H, H-5a), 0.83 (ddd, J = 26.8, 18.5, 6.4 Hz, 8H, H-7 ), H-15, H-
a), 3.61–3.47 (m, 1H, H-ab), 3.19–3.05 (m, 2H, H-b),
a
a
a,
a
H-8b), 1.52 (d, J = 12.1 Hz, 1H, H-7b), 1.42–1.23 (m, 6H, H-5b,
8a
a
H-6, H-8a, H-14), 1.19–1.10 (m, 1H, H-5a), 1.02–0.80 (m, 21H, H-
7a
, H-15, H-16). 13C NMR (151 MHz, DMSO) d 103.44 (C-3),
a
101.03 (C-10), 86.90 (C-12), 80.34 (C-12a), 64.16 (C-a), 52.20
(C-5a), 48.54 (C-40), 47.31 (C-b), 46.78 (C-10), 43.70 (C-8a), 41.25
(C-20), 36.48 (C-6), 36.00 (C-4), 34.11 (C-7), 30.27 (C-9), 25.61 (C-
14), 24.22 (C-5), 23.91 (C-8), 20.26 (C-50), 20.10 (C-15), 12.62
(C-16). HRMS m/z [M+H]+: 455.3485 (Calcd for C25H46N2O5:
454.340673).
13
16). C NMR (151 MHz, DMSO) d 136.66 (C-60), 125.69 (C-50),
120.12 (C-40), 103.43 (C-3), 100.90 (C-10), 86.90 (C-12), 80.33
(C-12a), 63.34 (C-a), 52.05 (C-5a), 45.85 (C-b), 44.25 (C-30), 43.96
(C-10), 43.66 (C-8a), 36.33 (C-6), 36.01 (C-4), 34.10 (C-7), 30.27
(C-9), 26.90 (C-20), 25.62 (C-14), 24.21 (C-5), 23.76 (C-8), 20.12
(C-15), 12.62 (C-16). HRMS m/z [M+H]+: 436.2807 (Calcd for
C23H37N3O5: 435.273322).
2.5.2.6. 1-[(Furan-2-ylmethyl)amino]-2-(10-dihydroartemisin-
oxy)ethane oxalate, 7.
Compound 1 (2.6 mmol, 1 g) was dis-
2.5.2.9. 1-{[2-(Phenylamino)ethyl]amino}-2-(10-dihydroarte-
misinoxy)ethane oxalate, 10. Compound 1 (2.6 mmol, 1 g)
solved in 15 ml DMF and reacted with furfurylamine (10.2 mmol,
0.9454 ml, 4 equiv) in a microwave reactor in bursts of 60 watts
and 40 °C for 4 minutes at a time. Purification by column chroma-
tography (silica gel, EtOAc) gave a brown oil. The conversion into
the oxalate salt afforded 0.366 g (28 %) of 7 as a white powder,
mp 161.03 °C; C22H33NO6. 1H NMR (600 MHz, DMSO) d 7.75 (s,
1H, H-50), 6.57 (s, 1H, H-40), 6.52 (s, 1H, H-30), 5.40 (s, 1H, H-12),
was dissolved in 15 ml DMF and reacted with N-phenylethylenedi-
amine (12.8 mmol, 1.7 ml, 5 equiv) in a microwave reactor in
bursts of 60 watts and 40 °C for 4 min at a time. Purification by col-
umn chromatography (silica gel, MeOH) gave a brown oil. The con-
version into the oxalate salt afforded 0.970 g (74%) of 10 as an off
white powder, mp 151.63 °C; C25H38N2O5. 1H NMR (600 MHz,
DMSO) d 7.08 (t, J = 7.8 Hz, 2H, H-50), 6.65–6.51 (m, 3H, H-40,
H60), 5.39 (s, 1H, H-12), 4.71 (d, J = 3.4 Hz, 1H, H-10), 3.92 (dt,
4.71 (s, 1H, H-10), 4.20 (s, 2H, H-10), 3.89 (s, 1H, H-a
1H, H-ab), 3.08 (s, 2H, H-b), 2.42 (s, 1H, H-9), 2.18 (s, 1H, H-4
2.01 (s, 1H, H-4b), 1.80 (s, 1H, H-5 ), 1.66 (m, 2H, H-8 . H-8b),
1.54 (s, 1H, H-7b), 1.32 (m, 6H, H-6, H-8a, H-5b), 1.14 (s, 1H, H-
5a), 0.88 (m, 7H, H-7
, H-15, H-16). 13C NMR (151 MHz, DMSO)
a), 3.55 (s,
a
),
a
a
J = 10.7, 5.2 Hz, 1H, H-aa), 3.62–3.53 (m, 1H, H-ab), 3.34 (dd,
J = 16.1, 9.6 Hz, 2H, H-20), 3.23–3.15 (m, 2H, H-b), 3.13–3.05 (m,
2H, H-10), 2.43–2.34 (m, 1H, H-9), 2.17 (td, J = 14.1, 3.8 Hz, 1H,
a
d 146.84 (C-20), 143.96 (C-50), 111.41 (C-40), 110.87 (C-30), 103.43
(C-3), 101.06 (C-10), 86.89 (C-12), 80.34 (C-12a), 63.65 (C-a),
51.90 (C-5a), 45.75 (C-b), 43.83 (C-10), 42.83 (C-8a), 36.46 (C-6),
36.01 (C-4), 34.09 (C-7), 30.27 (C-9), 25.62 (C-14), 24.22 (C-5),
23.95 (C-8), 20.11 (C-15), 12.62 (C-16). HRMS m/z [M+H]+:
408.2375 (Calcd for C22H33NO6: 407.230788).
H-4
1.54 (m, 2H, H-8
1.39–1.19 (m, 6H, H-6, H-8a, H-5b), 1.19–1.04 (m, 1H, H-5a), 0.82
a
), 2.04–1.93 (m, 1H, H-4b), 1.84–1.74 (m, 1H, H-5
a), 1.72–
a, H-8b), 1.49 (dd, J = 13.0, 3.0 Hz, 1H, H-7b),
(ddd, J = 16.2, 13.6, 5.1 Hz, 7H, H-7a
, H-15, H-16). 13C NMR
(151 MHz, DMSO) d 147.97 (C-300), 128.89 (C-50), 116.32 (C-60),
112.34 (C-40), 103.43 (C-3), 101.03 (C-10), 86.91 (C-12), 80.35

4706
T. T. Cloete et al. / Bioorg. Med. Chem. 20 (2012) 4701–4709
(C-12a), 63.50 (C-a), 52.02 (C-5a), 46.21 (C-10), 46.02 (C-b), 43.66
(C-8a), 41.15 (C-20), 36.33 (C-6), 35.81 (C-4), 34.11 (C-7), 30.27
(C-9), 25.49 (C-14), 24.23 (C-5), 23.89 (C-8), 20.12 (C-15), 12.62
(C-16). HRMS m/z [M+H]+: 447.2858 (Calcd for C25H38N2O5:
446.278073).
(m, 1H, H-5a), 0.98–0.61 (m, 15H, H-30, 7 , H-15, H-16). 13C NMR
a
(151 MHz, DMSO) d 103.24 (C-3), 101.06 (C-10), 87.07 (C-12),
80.35 (C-12a), 63.36 (C-a), 54.32 (C-20), 51.75 (C-5a), 46.36 (C-b),
43.68 (C-8a), 36.47 (C-6), 36.00 (C-4), 34.09 (C-7), 30.11 (C-9),
25.62 (C-10), 25.36 (C-14), 24.37 (C-5), 23.93 (C-8), 20.12 (C-15),
19.94 (C-30), 12.63 (C-16). HRMS m/z [M+H]+: 384.2745 (Calcd for
2.5.2.10. 1-(Piperazin-1-yl) -2-(10-dihydroartemisinoxy)ethane
C21H37NO5: 383.267174).
oxalate, 11.
Compound 1 (2.6 mmol, 1 g) was dissolved in
3 ml toluene and reacted with piperazine (12.8 mmol, 1 g, 5 equiv)
in a microwave reactor in bursts of 60 watts and 40 °C for 4 min at
a time. Purification by column chromatography (silica gel, MeOH/
DCM/NH4 1:9:0.5) gave a brown oil. The conversion into the oxa-
late salt afforded 0.669 g (53%) of 11 as a white powder, mp
159.23 °C; C21H36N2O5. 1H NMR (600 MHz, DMSO) d 5.38 (s, 1H,
3. Results
3.1. Chemistry
Condensation of dihydroartemisinin with 2-bromoethanol pro-
duced intermediate 1 in good yield (Scheme 1).24 This intermediate
and all the other products were in the 10b form, as is confirmed by
a coupling constant of J = 3.4 Hz between H-10 and H-9, which is
consistent with a cis-equatorialaxial arrangement of these protons
in a chair pyranose ring25 and was previously determined by X-ray
analysis.26 A given amine was then reacted with 1 using either the
described method with conventional heating, or for the first time
with microwave radiation (Scheme 2) rendering products 2 to
12. Compounds 2 and 11 have already been synthesised by Li
and co-workers using the conventional heating method.24
H-12), 4.68 (d, J = 2.9 Hz, 1H, H-10), 3.78–3.70 (m, 1H, H-aa),
3.46 (dt, J = 10.5, 5.3 Hz, 1H, H-ab), 3.04 (s, 4H, H-20), 2.71–2.52
(m, 6H, H-b, H-10), 2.37 (d, J = 3.2 Hz, 1H, H-9), 2.17 (td, J = 14.1,
3.5 Hz, 1H, H-4
2H, H-8 , H-5 ), 1.65–1.57 (m, 1H, H-8b), 1.56–1.50 (m, 1H, H-
7b), 1.40–1.19 (m, 6H, H-5b, H-6, H-8a), 1.13 (dt, J = 17.9, 9.0 Hz,
1H, H-5a), 0.85 (dd, J = 35.6, 6.7 Hz, 7H, H-7
, H-15, H-16). 13C
a), 1.98 (d, J = 14.2 Hz, 1H,H-4a), 1.84–1.67 (m,
a
a
a
NMR (151 MHz, DMSO) d 103.36 (C-3), 100.79 (C-10), 86.89 (C-
12), 80.49 (C-12a), 64.92 (C-a), 56.90 (C-b), 52.10 (C-5a), 49.55
(C-10), 43.86 (C-20), 42.87 (C-8a), 36.82 (C-6), 36.07 (C-4), 34.16
(C-7), 30.48 (C-9), 25.68 (C-14), 24.34 (C-5), 23.97 (C-8), 20.22
(C-15), 12.81 (C-16). HRMS m/z [M+H]+: 397.2699 (Calcd for
3.2. In vitro antimalarial activity and cytotoxicity
C
21H36N2O5: 396.262423).
The antimalarial activity and cytotoxicity of the synthesised
compounds are given in Tables 1 and 2. They were all tested in
the 10b-isomer form. Comparing the antimalarial activity of the
compounds to that of artesunate, Dunnett0s test shows that the
mean IC50 values of compounds 2–7 and 9–12 differ statistically
significantly from the mean of artesunate since they have p-values
smaller than 0.05, implying that these compounds were less potent
than artesunate against the D10 strain. The IC50 value of compound
8, however, had a p-value greater than 0.05 indicating that there is
no statistically significant difference from the mean value of
artesunate implying that this compound was as potent against
the D10 strain as artesunate. Against the Dd2 strain, however,
derivatives 3, 5, 9, 10 and 12 were less potent whilst 2, 4, 6 - 8,
and 11 all displayed potency similar to that of artesunate (Table 1).
In comparison with DHA, all the derivatives were found to be
less potent against the D10 strain. Compounds 2, 7 and 8, however,
2.5.2.11. 1-[(2-Methylpropyl)amino]-2-(10-dihydroartemisin-
oxy)ethane oxalate, 12. Compound 1 (2.6 mmol, 1 g) was
dissolved in 15 ml DMF and reacted with isobutylamine
(10.2 mmol, 1.03 ml, 4 equiv) in a microwave reactor in bursts of
60 watts and 40 °C for 4 min at a time. Purification by column
chromatography (silica gel, MeOH) gave a colourless oil. The con-
version into the oxalate salt afforded 0.388 g (23 %) of 12 as a white
powder, mp 161.7 °C; C21H37NO5. 1H NMR (600 MHz, DMSO) d 5.40
(s, 1H, H-12), 4.71 (t, J = 5.9 Hz, 1H, H-10), 3.97–3.88 (m, 4H, H-aa),
3.57 (dd, J = 11.0, 5.6 Hz, 1H, H-ab), 3.20–3.05 (m, 2H, H-b), 2.88–
2.68 (m, 2H, H-10), 2.41 (dd, J = 7.1, 3.8 Hz, 1H, H-9), 2.17 (td,
J = 14.1, 3.6 Hz, 1H, 4
(m, 1H, 5 ), 1.72–1.57 (m, 2H, H-8a, H-8b), 1.53 (d, J = 10.3 Hz,
1H, H-7b), 1.44–1.17 (m, 6H, H-5b, H-6, H-8a, H-14), 1.19–1.09
a
), 2.06–1.84 (m, 2H, H-4b, H-20), 1.83–1.73
a
Table 1
Descriptive statistics of antimalarial IC50 values, results of ANOVAs and Dunnett’s tests of compounds 2–12, artesunate, DHA and chloroquine
Compound
D10
Dd2
RIb
na Mean IC50
(nM)
Std.
dev.
p-
p-Value: Dunnett
ArtNa DHA CQ
na Mean
IC50(nM)
Std.
dev.
p-
p-Value: Dunnett
Value:Welch
Value:Welch
ArtNa
0.920
DHA
CQ
0.000* 3.483
2
3
4
5
6
7
8
9
3
3
3
3
3
3
3
3
3
3
3
3
3
4
4.637
6.885
3.312
6.718
1.833
2.413
1.444
6.038
2.612
1.799
4.015
0.933
0.785
46.644
0.077
0.331
0.067
0.369
0.046
0.111
0.058
0.419
0.340
0.279
0.213
0.161
0.142
2.348
0.000* 0.000* 0.000*
0.000* 0.000* 0.000*
0.000* 0.000* 0.000*
0.000* 0.000* 0.000*
0.001* 0.000* 0.000*
0.000* 0.000* 0.000*
3
3
3
3
3
3
3
3
3
3
3
3
3
4
16.151
1.681
4.659
0.000
6.726
4.140
1.775
0.883
38.270
2.937
5.063
3.559
0.511
0.325
0.743
61.585
32.109
66.727
48.096
21.374
20.059
189.258
55.085
50.943
54.930
25.854
3.482
0.008* 0.000* 0.000* 8.945
0.995
0.043* 0.000* 9.695
0.002* 0.000* 0.000* 9.932
0.171
1.000
0.998
0.000* 0.000* 26.240
0.374
0.459
0.000* 8.858
0.114
0.023* 0.000*
0.000* 13.888
0.000* 0.000* 0.000*
0.000* 0.000* 0.000*
0.002* 0.000* 0.000*
0.000* 0.000* 0.000*
0.000* 0.000* 0.219
31.343
10
11
12
ArtNa
DHA
CQc
0.038* 0.000* 0.000* 21.086
0.095
0.000* 0.000* 28.318
0.039* 0.000* 0.000* 13.680
0.000*
0.000*
0.000*
0.000*
0.000*
27.719
4.433
3.618
168.740
12.656 0.000*
*
Statistically significant at 0.05 level. Note, values <0.05 indicates a statistically significant difference between the mean IC50 values with more than 95% certainty.
a
Replicates (n).
Resistance index (Dd2/D10).
b
c
CQ was tested as diphosphate salt.

T. T. Cloete et al. / Bioorg. Med. Chem. 20 (2012) 4701–4709
4707
Table 2
Descriptive statistics of cytotoxicity IC50 values, results of ANOVA and Dunnett’s test of compounds 7, 8, 10, 12 and emetine.
Compound
CHO
na
Mean IC50 (nM)
Std. dev.
4560.281
31708.460
14112.55
8151.421
10.298
p-value:Welch
p-value: Dunnett
Emetine
7
8
10
12
3
3
3
3
2
79672.800
140958.200
47471.900
72921.010
19.765
0.002*
0.000*
0.042*
0.004*
Emetine
0.000*
*
Statistically significant at 0.05 level. Note, values <0.05 indicates a statistically significant difference between the mean IC50 values with more than 95% certainty.
Replicates (n). Note that the values of compounds 2–6, 9 and 11 were above the maximum concentration used in this test and are therefore not included in the statistical
a
analysis.
had mean IC50 values against the Dd2 strain that were not statisti-
cally significantly different from that of DHA, implying that these
compounds were equipotent to DHA (Table 1).
Compared to chloroquine all derivatives were more potent
against both the D10 and Dd2 strains, with the exception of 9 that
was as potent as CQ against the Dd2 strain.
The structures of all synthesised compounds were confirmed by
HRMS, 1H and 13C NMR techniques. Due to the rigid structure and
asymmetric carbon units in the artemisinin nucleus, the methylene
protons in positions 4, 5, 7 and 8 are all non-equivalent to one an-
other and in each case appear as an AB pattern at d 2.17 (4
a), 2.06
(4b), 1.83 (5 ), 1.44 (5b), 1.53 (7 ), 0.98 (7b), 1.72 (8 and 8b),
a
a
a
The very low cytotoxicity of compounds 2–6, 9 and 11 is illus-
trated by their IC50 values being above the maximum concentra-
tion used in this test. Dunnett0s test showed that the cytotoxicity
of compounds 7, 8, 10 and 12 differed statistically significantly
from that of emetine indicating that these compounds were less
toxic than emetine (Table 2).
respectively. H-6, H-5a, H-8a and H-9, integrating for one proton
each, gave signals at d 1.44, 1.19, 1.17 and 2.41, respectively. The
above mentioned signals together with the characteristic signals
of H-12, H-10 and the methyl signals of H-14, H-15 and H-16, at d
5.40, 4.71, 1.44, 0.98 and 0.61, respectively, confirmed the presence
of the artemisinin moiety in the structures of all synthesised com-
pounds. The protons H-a of the ether linker are also non-equivalent,
appearing in a characteristic AB pattern at d 3.9 (aa) and 3.5 (ab),
4. Discussion
4.1. Chemistry
and serves as a marker that confirms the linker was still intact dur-
ing the second step of the synthetic process. The most characteristic
13C NMR signals of artemisinin are those of C-10, C-12,
C-3 and C-12a at d 100.90, 86.90, 103.43 and 80.33, respectively.
The signals of the methyl carbon atoms C-14, C-15 and C-16 can
also clearly be seen at d 25.62, 20.12 and 12.62, respectively. The
signal of the linker0s carbon atom C-a can usually be found at d
63.34.
The amino moieties were attached to DHA through an ethyl
ether linker. This ether bond is notoriously labile and is rapidly
metabolised by human liver microsomes.27 Artemether, an ether
derivative of artemisinin, is metabolised in vivo reproducing DHA
within two hours after intramuscular administration.28 The syn-
thesised compounds 2–12 were, however, tested in vitro which
preserves the ether bond and allows the compounds to act as
derivatives and not as prodrugs.
4.2. In vitro antimalarial activity and cytotoxicity
All the synthesised compounds were more potent than CQ
against the chloroquine sensitive (CQS) D10 strain. Against the
chloroquine resistant (CQR) Dd2 strain, they showed better activity
than CQ with only compound 9 being as active as CQ. Of all the
synthesised compounds, compound 8 (IC50 = 1.44 nM) had the
highest activity against the D10 strain with activity comparable
to that of artesunate. Against the Dd2 strain compounds 2, 4, 6–8
and 11 were as active as artesunate, whilst compounds 2, 7 and
8 had activity comparable to that of the even more potent DHA.
An interesting observation in this study was that compounds
with shorter chain lengths displayed higher activity than those
with longer chains connecting the DHA and amine moieties. An in-
crease in chain length from two to five methylene equivalent units
led to a reduction in activity as can be seen from the difference in
activity of 2 (IC50 = 4.64 nM) versus that of 3 (6.89 nM) and 4
Compounds 3, 5, 6, 7, 9, 10, 11 and 12 were synthesised using a
microwave reactor. This is to our knowledge the first time that
artemisinin compounds were subjected to microwave radiation.
Special precautions were taken to assure that the peroxide bond
remained intact during this procedure as this functional group is
reported to be susceptible to facile hydrolysis under basic and
acidic conditions29 and can also be expected to be sensitive to heat.
We found the peroxide bridge to be stable and were able to heat
the reaction to more than 90 °C without cleavage of the peroxide
bond. This perceived stability was confirmed by the literature
where the peroxide bond proved to be stable, even toward some
highly reactive organometallic reagents.30 Increasing either the
duration or the power of the radiation led to the formation of a
large number of by-products, most probably as result of the cleav-
age of the endoperoxide bridge.29 Microwave radiation decreased
the overall reaction time, but had no noticeable influence on the
yield of the reactions. This can be seen when looking at compound
11 that had a yield of 53% in this method, compared to the reported
60%.24 13C NMR spectrometry and high resolution MS confirmed
conclusively that the peroxide bridge was preserved during micro-
wave radiation in that no deviations from the expected values
were observed in the spectra of all prepared compounds.
(3.31 nM) versus
(16.15 nM) versus
5
3
(6.72 nM) for the D10 strain, and
(61.59 nM) and (32.11 nM) versus
2
5
4
(66.73 nM) for the Dd2 strain. For both strains Tukey0s test con-
firms that the mean IC50 values of 2 and 3, and that of 4 and 5 differ
statistically significantly from one another. While the difference in
activity can be specified with 95% certainty for both pairs against
the D10 strain, against the Dd2 strain the difference in activity
for latter pair can only be specified with 94% certainty, as can be
seen from the p-value of 0.057 (Tables 3 and 4). Calas et al.12 in
For stability and solubility reasons, the free base target com-
pounds were converted into oxalate salts.

4708
T. T. Cloete et al. / Bioorg. Med. Chem. 20 (2012) 4701–4709
Table 3
p-Values of Tukey’s test for compounds 2–12 on the D10 strain
Compound
Mean IC50 (nM)
2
3
4
5
6
7
8
9
10
11
12
4.637
6.885
3.312
6.718
1.833
2.413
1.444
6.038
2.612
1.799
4.015
2
3
4
5
6
7
8
9
4.637
6.885
3.312
6.718
1.833
2.413
1.444
6.038
2.612
1.799
4.015
0.000*
0.000*
0.000*
0.000*
0.999
0.000*
0.000*
0.000*
0.000*
0.000*
0.000*
0.000*
0.008*
0.000*
0.206
0.000*
0.000*
0.000*
0.000*
0.709
0.000*
0.015*
0.000*
0.082
0.000*
0.000*
0.068
0.000*
0.000*
0.000*
0.000*
1.000
0.153
0.802
0.000*
0.021*
0.143
0.000*
0.000*
0.000*
0.000*
0.000*
0.000*
0.000*
0.000*
0.000*
0.143
0.000*
0.065
0.000*
0.999
0.000*
0.000*
0.008*
0.000*
0.000*
0.068
0.000*
0.030*
0.995
0.000*
0.000*
0.000*
0.000*
0.000*
0.000*
0.000*
0.000*
0.000*
0.000*
0.015*
0.000*
0.000*
0.000*
0.000*
0.000*
0.000*
0.082
0.000*
0.000*
0.000*
0.206
0.709
0.000*
0.030*
1.000
0.000*
0.004*
0.004*
0.000*
0.995
0.153
0.000*
0.000*
0.000*
0.000*
0.000*
0.802
10
11
12
0.000*
0.000*
0.000*
0.000*
0.000*
0.000*
0.000*
0.065
0.021*
0.000*
0.000*
0.000*
*
Statistically significant at 0.05 level. Note, values <0.05 indicates a statistically significant difference between the mean IC50 values with more than 95% certainty.
their series of amino compounds found that those with longer
chain lengths displayed better activity than those with shorter
chains. This observation might suggest that our compounds per-
haps do not share the same mechanism of action, having too great
a difference in their pharmacophore.
possible reason for this phenomenon could be that the longer chain
polyamino compounds are too polar to cross the parasitic mem-
branes while their short chain counterparts were lipophillic en-
ough to reach the site of action. Interestingly, the difference in
activity of the long chain polyamino compounds 6, 9–11 against
the resistant and sensitive strains as displayed by their RI values,
is greater than for the short chain compounds. Whether this is
due to the specific method by which the parasite acquires resis-
tance, for example targeting polyamine compounds, still needs to
be elucidated.
Compound 2, with a short chain and one amino group, had the
lowest IC50 value of all the synthesised compounds against the Dd2
strain. This observation correlates well with the above findings
that short chain mono amino compounds retain a higher degree
of activity than the longer chain polyamino compounds. Com-
pound 8 with good activity against both the D10 and Dd2 strains
features a tertiary amino group and an aryl ring. Compound 9 with
a side chain containing a secondary amine and a terminal imidaz-
ole ring was the compound with the most nitrogen atoms and had
the lowest overall activity, especially against the Dd2 strain
(IC50 = 189.26 nM), and the highest RI value (RI = 31.34).
Against both strains, none of the compounds had activity higher
than that of DHA or artesunate. While the addition of amino
moieties to artemisinin had some favourable consequences, other
factors such as H bonding, molecular shape, polarity, flexibility
and charge/ionization also influence the accumulation of the drug
at the site of action thus increasing or decreasing the activity of the
drug. With DHA and artesunate having largely the same structure
as the synthesized compounds, it is not unexpected they might ex-
ert their antiplasmodial effect to roughly the same extend.32
Compounds 2, 4 and 11 forms a homologous series in that they
differ from one another in respect to the cyclic amino substituent,
with 2 having an oxygen atom in position 4, 4 a methylene group
and 11, a NH group. Tukey’s test confirms that the mean IC50 values
of 2, 4 and 11 differ statistically significantly from one another
against the D10 strain. The calculated pKa value of compound 2
is 6.75, followed by that of 4 at 8.81 and then that of 11 at
9.31.31 Their pKa values are thus in the order 11 > 4 > 2. On the ba-
sis of their antimalarial activity against the D10 strain, these com-
pounds can also be arranged in the order 11 > 4 > 2, from most to
least active. Similarly the order of the resistance index (RI) values
of these three compounds is 11 > 4 > 2, which is also the same or-
der as the pKa values showing that the compound with the lowest
pKa value has the lowest discrimination between the two strains. A
lower pKa value of the tertiary amine implies a weaker base and
this decreases the likelihood of the compound to be trapped in
the parasite’s acidic food vacuole via ion trapping. Whether there
is any correlation between the difference in activity against the
two strains, the pKa values of the amine moieties and the mode
by which the parasite acquires resistance warrants further
investigation.
Short chain compounds with fewer amino groups, that is, 6–8
and 11 had better activity than the long chain polyamines, that is
3, 5 and 9 against the D10 strain while the same observation was
made comparing 7 and 8 to 3, 5 and 9 against the Dd2 strain. A
Table 4
p-Values of Tukey’s test for compounds 2–12 on the Dd2 strain
Compound
Mean IC50 (nM)
2
3
4
5
6
7
8
9
10
11
12
16.151
61.585
32.109
66.727
48.096
21.374
20.059
189.258
55.085
50.943
54.930
2
3
4
5
6
7
8
9
16.151
61.585
32.109
66.727
48.096
21.374
20.059
189.258
55.085
50.943
54.930
0.005*
0.858
0.158
0.002*
1.000
0.098
0.944
0.857
0.717
1.000
0.017*
0.988
0.005*
0.256
1.000
0.013*
0.973
0.004*
0.205
1.000
0.000*
0.000*
0.000*
0.000*
0.000*
0.000*
0.000*
0.022*
1.000
0.448
0.978
1.000
0.069
0.052
0.000*
0.055
0.989
0.705
0.866
1.000
0.155
0.121
0.000*
1.000
0.023*
1.000
0.457
0.976
1.000
0.071
0.054
0.000*
1.000
1.000
0.001*
0.858
0.002*
0.098
1.000
1.000
0.000*
0.022*
0.055
0.023*
0.158
1.000
0.944
0.017*
0.013*
0.000**
1.000
0.988
1.000
0.057**
0.057**
0.857
0.988
0.973
0.000*
0.448
0.705
0.457
0.717
0.005*
0.004*
0.000*
0.978
0.866
0.976
0.256
0.205
0.000*
1.000
1.000
1.000
1.000
0.000*
0.069
0.155
0.071
0.000*
0.052
0.121
0.054
10
11
12
0.000*
0.000*
0.000*
1.000
1.000
1.000
*
Statistically significant at 0.05 level. Note, values <0.05 indicates a statistically significant difference between the mean IC50 values with more than 95% certainty.
Statistically significant at 0.06 level.
**

T. T. Cloete et al. / Bioorg. Med. Chem. 20 (2012) 4701–4709
4709
The conclusions drawn on the activity of compounds 2–12 are
References and notes
supported by statistical analysis of the data generated by the anti-
malarial test, but no attempts were made to draw any structure
activity relationships as the set of compounds (11 in total) does
not allow such an endeavour.
All of the synthesised compounds had a high degree of selectiv-
ity towards the parasite, with toxicity against CHO cells statisti-
cally significantly below that of the reference drug, emetine.
1. WHO. 2011. World Malaria Report 2011. [Web:] http://www.who.int/malaria/
world_malaria_report_2011/9789241564403_eng.pdf [Date used: 7 February
2012].
2. White, N. J. Malaria. In Manson’s tropical diseases; Cook, G. C., Manson, P., Zumla,
A. I., Eds.; W.B. Saunders Company: Philadelphia, 2008; pp 1201–1297.
3. Petersen, I.; Eastman, R.; Lanzer, M. FEBS Lett. 2011, 585, 1551.
4. WHO. 2010. Guidelines for the treatment of malaria. 2nd ed. http://
whqlibdoc.who.int/publications/2010/9789241547925_eng.pdf [Date used: 9
April 2012].
5. Ploypradith, P. Acta Tropica. 2004, 89, 329.
6. Krishna, S. Drug Resistance Updates 2004, 7, 233.
5. Conclusion
7. Duthaler, U.; Huwyler, J.; Rinaldi, L.; Cringoli, G.; Keiser, J. Vet. Parasitol. 2012,
186, 270.
In this study eleven 10-aminoethylether derivatives of artemis-
inin were successfully synthesised by attachment of various amino
moieties through an ether linkage to C-10 of DHA. The structures of
the products were confirmed by NMR and HRMS. The antimalarial
activity of the compounds against both the D10 and Dd2 strains of
Plasmodium falciparum were determined with CQ, sodium artesu-
nate and DHA as references. All compounds were found to be more
potent than CQ against either strain, with only compound 9 being
equally as active against the Dd2 strain. Derivative 8 was as potent
as artesunate against the D10 strain while compounds 2, 4, 6–8
and 11 had potency similar to that of artesunate against the Dd2
strain. No compound possessed higher or comparable activity to
DHA against the D10 strain. However, compounds 2, 7 and 8 were
found as potent as DHA against the Dd2 strain.
8. Dondorp, A. M.; Nosten, F.; Yi, P.; Das, D.; Phyo, A. P.; Tarning, J.; Lwin, K. M.;
Ariey, F.; Hanpithakong, W.; Lee, S. J.; Ringwald, P.; Silamut, K.; Imwong, M.;
Chotivanich, K.; Lim, P.; Herdman, T.; An, S. S.; Yeung, S.; Singhasivanon, P.;
Day, N. P. J.; Lindegardh, N.; Socheat, D.; White, N. J. N. Engl. J. Med. 2009, 361,
455.
9. Phyo, A. P.; Nkhoma, S.; Stepniewska, K.; Ashley, E. A.; Nair, S.; McGready, R.;
Moo, Ler C.; Al-Saai, S.; Dondorp, A. M.; Lwin, K. M.; Singhasivanon, P.; Day, N.
P. J.; White, N. J.; Anderson, T. J. C.; Nosten, F. The Lancet. 2012, in press.
10. Ramya, T. N. C.; Surolia, N.; Surolia, A. Biochem. Biophys. Res. Commun. 2006,
348, 584.
11. Chadwick, J.; Jones, M.; Mercer, A. E.; Stocks, P. A.; Ward, S. A.; Park, K. B.;
O’Neill, P. M. Bioorg. Med. Chem. 2010, 18, 2586.
12. Calas Le, M.; Cordina, G.; Bompart, J.; Ben Bari, M.; Jei, T.; Ancelin, M. L.; Vial, H.
J. Med. Chem. 1997, 40, 3557.
13. Delhaes, L.; Biot, C.; Berry, L.; Maciejewski, L. A.; Camus, D.; Brocard, J. S.; Dive,
D. Bioorg. Med. Chem. 2000, 8, 2739.
14. Calas Le, M.; Ancelin, L. M.; Codrina, G.; Portefaix, P.; Piquet, G.; Vidal-Sailhan,
V.; Vial, H. J. Med. Chem. 2000, 43, 505.
It was observed in a small group of three homologous com-
pounds that the order of the IC50 values against the D10 strain
was the same as the order of their pKa values as was the order of
their RI values.
15. Egan, T. J. Targets 2003, 2, 115.
16. Hindley, S.; Ward, S. A.; Storr, R. C.; Searle, N. L.; Bray, P. G.; Park, K.; Davies, J.;
O’Niel, P. M. J. Med. Chem. 2002, 45, 1052.
17. O’Niel, P. M.; Bishop, L. P.; Storr, R. C.; Hawley, S. R.; Maggs, J. L.; Ward, S. A.;
Park, K. J. Med. Chem. 1996, 39, 4511.
The short chain compounds were more active than the long
chain polyamine containing compounds irrespective of the strain.
In summary, this study delivered three compounds, viz. 2, 7 and
8, that were as potent as DHA against the chloroquine resistant
Dd2 stain, and one derivative, viz., 8, that was as potent as artesu-
nate against the D10 strain. Current efforts are underway to deter-
mine the in vivo IC50 values of these compounds and to elucidate
their pharmacokinetic properties. They stand as good drug candi-
dates to be further investigated in search for potent new antima-
larial drugs.
18. Trager, W.; Jensen, J. B. Science 1976, 193, 673.
19. Makler, M. T.; Ries, J. M.; Williams, J. A.; Bancroft, J. E.; Piper, R. C.; Gibbins, B.
L.; Hinrichs, D. J. Am. Soc. Trop. Med. Hyg. 1993, 48, 739.
20. Mosmann, T. J. Immunol. Methods. 1983, 65, 55.
21. Rubinstein, L. V.; Shoemaker, R. H.; Paull, K. D.; Simon, R. M.; Tosini, S.; Skehan,
P.; Scudiero, D. A.; Monks, A.; Boyd, M. R. J. Natl. Cancer Inst. 1990, 82(13), 1113.
22. Tabachnick, B. G.; Fidell, L. S. Using Multivariate Statistics, 4th ed.; Allyn &
Bacon: Boston, 2001. p. 966.
23. StatSoft, Inc. STATISTICA, 2011, version 10. http://www.statsoft.com.
24. Li, Y.; Zhu, Y.; Jiang, H.; Pan, J.; Wu, G.; Wu, J.; Shi, Y.; Yang, J.; Wu, B. J. Med.
Chem. 2000, 43, 1635.
25. Haynes, R. K.; Chan, H. W.; Cheung, M. K.; Lam, W. L.; Soo, M. K.; Tsang, H. W.;
Voerste, A.; Williams, I. D. Eur. J. Org. Chem. 2002, 1, 113.
26. Lombard, M. C.; Fernandes, M. A.; Breytenbach, J. C.; N’Da, D. D. Acta Cryst. E.
2010, 66, 2182.
Acknowledgements
27. Grace, J. M.; Aguilar, A. J.; Trotman, K. M.; Brewer, T. G. Drug Metab. Dispos.
1997, 26, 313.
We thank the National Research Foundation (NRF), the North-
West University (NWU) for financial support and the German Aca-
demic Exchange Service (DAAD) for a study bursary to T.T.C.
28. Karbwang, J.; Na-Bangchang, K.; Congpuong, K.; Molunto, P.; Thanavibul, A.
Eur. J. Clin. Pharmacol. 1997, 52, 307.
29. Torok, S. D.; Ziffer, H. Tetrahedron Lett. 1995, 6, 829.
30. O’Dowd, H.; Ploypradith, P.; Xie, S.; Shapiro, T. A.; Posner, G. H. Tetrahedron
1999, 55, 3625.
31. Advaced Chemistry Inc. ACD/ChemSketch, 2000, version 4.54. http://
www.acdlabs.com.
Supplementary data
32. Bigucci, F.; Kamsu-Kom, T.; Cholet, C.; Besnard, M.; Bonnet-Delpon, D.;
Ponchel, G. J. Pharm. Pharmacol. 2008, 60, 163.
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.bmc.2012.06.014.