New anti-malarial peroxides with in vivo potency derived from spongean metabolites

Article abstract of DOI:10.1016/j.bmcl.2003.08.073

The structure-activity relationship of the anti-malarial substance 3 having a 6-carbomethoxymethyl-3-methoxy-1,2-dioxane structure was studied. The ester portion of the peroxide 3, showing little in vivo efficacy in malaria-infected mice in spite of the potent in vitro activity, was hydrolyzed in serum to afford an inactive free acid 4. The amide analogues (8 and 9) robust to mouse serum were disclosed to exhibit in vivo anti-malarial potency.

Full text of DOI:10.1016/j.bmcl.2003.08.073

Bioorganic& Medicinal Chemistry Letters 13 (2003) 4081–4084
New Anti-Malarial Peroxides with In Vivo Potency Derived from
Spongean Metabolites
Nobutoshi Murakami,a Motoyuki Kawanishi,a Huq Mohammad Mostaqul,a
Jie Li,b Sawako Itagaki,b Toshihiro Horiib and Motomasa Kobayashia,*
aGraduate School of Pharmaceutical Sciences, Osaka University, 1-6 Yamada-oka, Suita, Osaka 565-0871, Japan
bResearch Institute for Microbial Diseases, Osaka University, 3-1 Yamada-oka, Suita, Osaka 565-0871, Japan
Received 25 July 2003; accepted 15 August 2003
Abstract—The structure–activity relationship of the anti-malarial substance 3 having a 6-carbomethoxymethyl-3-methoxy-
1,2-dioxane structure was studied. The ester portion of the peroxide 3, showing little in vivo efficacy in malaria-infected mice in spite
of the potent in vitro activity, was hydrolyzed in serum to afford an inactive free acid 4. The amide analogues (8 and 9) robust to
mouse serum were disclosed to exhibit in vivo anti-malarial potency.
# 2003 Elsevier Ltd. All rights reserved.
Today, malaria is one of the most deadly diseases on
earth and the leading cause of sickness and death in the
developing world. Annually, an estimated 300 million
cases of malaria occur throughout the world. Mortality
associated with the disease is estimated at over one mil-
lion per year.1 About 40% of the world population live
in malaria endemicocuntries. Due to the spreading
resistance to the conventional anti-malarials, there is an
urgent demand to search for new anti-malarial princi-
ples.2 In this context, we have been engaged in explora-
tion for new anti-malarial candidates originating in
natural resources.3,4
Previously, methyl esters (1 and 2) of spongean per-
oxides,5 peroxyplakoricaicds, were shown to display
potent in vivo anti-malarial activity. Furthermore, the
facile construction of their core skeleton was estab-
Figure 1.
lished, and the new, readily accessible anti-malarial
peroxide 3 with higher selectivity index than that of 1
and 2 was found.6 However, the promising candidate 3
was shown to exert little in vivo potency because of
lability in mouse serum. On the basis of the analysis for
the metabolized portion in 3, the acquirement of stabil-
ity in serum resulted in the discovery of new anti-
malarial peroxides 8 and 9 with in vivo potency. This
paper deals with the in vivo anti-malarial peroxides uti-
lizing 1 and 2 as scaffolds (Fig. 1).
When the peroxide 3 was examined by an in vivo system
using a mouse model, 3 showed little anti-malarial
potency. This undesired finding indicated that the ester
function in 3 may be metabolized in serum. Therefore,
stability of 3 in mouse serum was examined.7 Treat-
ment of 3 with fresh mouse serum afforded the corre-
sponding carboxylic acid 4 in 85% yield involving
complete disappearance of 3 within 1.5 h. The car-
boxylicacid 4 was almost stable after 24 h exposure to
mouse serum, while the anti-malarial activity (IC50
>1.2 mM) of 4 was significantly reduced. The malaria
parasite converts toxic-free heme to nontoxic hemozoin
by oxidative polymerization. In a mechanistic study of
*Corresponding author. Tel.: +81-6-6879-8215; fax: +81-6-6879-
8219; e-mail: kobayasi@phs.osaka-u.ac.jp
0960-894X/$ - see front matter # 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2003.08.073

4082
N. Murakami et al. / Bioorg. Med. Chem. Lett. 13 (2003) 4081–4084
.
Scheme 1. Reagents and conditions: (a) LiOH H2O, DMSO-d, quant; (b) esterase (from porcine liver), phosphate buffer (0.2 M, pH 7.4), 89%;
.
(c) pentafluorophenol, EDCI HCl, pyridine, 84%; (d) MeNH2 HCl, pyridine, 92% for 7; EtNH2 HCl, pyridine, 96% for 8; PrNH2, pyridine, 98%
.
.
.
for 9; Me2NH HCl, pyridine, 93% for 10; Et2NH, pyridine, 63% for 11.
the action of anti-malarial peroxides, the formation of
the heme-peroxide adducts was proposed to inhibit this
detoxication pathway.8 The formation of the heme-
peroxide adducts was shown to take place through the
production of a radical species in the food vacuole, an
acidic organelle characteristic of Plasmodium falci-
parum.9 Thus, the poor accumulation of the carboxylic
acid 4 into the food vacuole was presumed to be a
cause of the reduced anti-malarial activity. Taking the
stability against esterase in serum and their neutral
property into account, the amide analogues were
designed.
Preparation of the amide analogues was conducted as
illustrated in Scheme 1. Alkaline hydrolysis of 3 with
LiOH in DMSO conclusively provided an epoxy car-
boxylicacid 5 by concomitant cleavage of the dioxane
ring quantitatively. On the other hand, the treatment of 3
with commercially available esterase from porcine liver
in a phosphate buffer (pH 7.4) furnished 4 in 89% yield.
Conversion from the carboxylic acid 4 to amides analo-
gues was carried out via an activated pentafluorophenyl
ester.10 The carboxylic acid 4 was coupled with penta-
fluorophenol in the presence of N-ethyl-N0-3-dimethyl-
.
aminopropylcarbodiimide hydrochloride (EDCI HCl) in
pyridine to give an ester 6 in 84% yield. In the first
instance, the monoethyl and diethyl amides (8 and 11)11
were prepared in order to evaluate their stability in
.
mouse serum. Treatment of 6 with EtNH2 HCl or
Et2NH in pyridine afforded the corresponding amide in
96 and 63% yields, respectively. As a result of assess-
ment for stability in mouse serum, both analogues were
shown to be free from metabolism for 5 h (Fig. 2). This
encouraging biological property directed us to synthe-
size several amide analogues. In expectation of potency
in a mouse model, we further designed three analogues
(7, 9, 10) under the guidance of appropriate clogPs.12,13
ClogP is a parameter correlated to the permeability of
drugs, and is thus believed to be an important index in
predicting biological activity in animal models. In gen-
eral, clogP values larger than 4 tend to reduce in vivo
pharmacological efficacy significantly regardless of a
Figure 2. Stability of ester and amides in mouse serum.
Table 1. Anti-malarial activity of amide analogues
Compd
R1
R2
cLogP
IC50 (mM)
P. falciparum
Selectivity index
ED50 (mg/kg)
T/Ca
KB 3-1
P. berghei
7
8
9
10
H
H
H
Me
Et
Me
Et
Pr
Me
Et
2.06
2.59
3.12
2.45
3.15
2.17
0.52
0.54
0.31
0.31
0.49
0.12
5.2
14
12
4.0
13
43
10
25
39
13
27
360
n.d.b
11
9.3
13
20
>30
5.0
118
125
138
95
118
81
11
3
Artemisinin
145
aDose 3, 7–11: 10 mg/kg; artemisinin: 5 mg/kg. T/C is the quotient of the survival days of the treated animals (T) and those of the control animals
(C). T/C values of >120 are considered to be active.
bSome mice died by treatment with 30 mg/kg of 7, then ED50 could not be determined.

N. Murakami et al. / Bioorg. Med. Chem. Lett. 13 (2003) 4081–4084
4083
good in vitro biological score. The analogues (7, 9, 10)
were synthesized from 6 in the same fashion as the pre-
paration of 8 and 11 (Scheme 1) and all the analogues
were evaluated for anti-malarial activity in vitro.14 The
resulting biological outcome is summarized in Table 1.
Although all of the amide analogues reduced in vitro
inhibitory activity for proliferation of P. falciparum as
compared with 3, selectivity indices of more than 10
were observed in all compounds. Hence, anti-malarial
potency was tested by a conventional 4-day suppressive
test using mice infected by P. berghei in regard to all the
analogues.15 Among them, the monoethyl and mono-
propyl analogues (8 and 916) showed in vivo anti-
malarial potency ðT=C > 120Þ by intraperitoneal
administration.
amides, the ratio of stereoisomers was slightly changed. The
amides 8 and 11 were furnished as a mixture of syn and anti
isomers in a ratio of 8.7:1 and 11:1, respectively. The physico-
chemical data were determined as a mixture. In the H NMR
spectra, only the signals due to H-7 and OCH3 were definitely
1
separated between the two isomers. 8: colorless solid. IR umax
1
(KBr) cmꢂ1: 1650, 1550. H NMR (500 MHz, CDCl3) d: 0.88
(3H, t, J=7.4 Hz, ðCH2Þ4CH3), 1.13 (3H, t, J=7.3 Hz,
NCH2CH3), 1.25–1.35 (6H, m), 1.50–1.84 (4H, m), 1.86–1.90
(2H, m), 2.31 (1H, dd, J=15.9, 3.7 Hz, Ha-7), 2.35 (1H, dd,
J=15.9, 7.3 Hz, Hb-7, syn), 2.92 (1H, dd, J=14.7, 8.5 Hz, Hb-
7, anti), 3.20–3.33 (2H, m), 3.26 (3H, s, OCH3, syn), 3.28 (3H,
s, OCH3, anti), 4.38 (1H, m, H-6), 5.96 (1H, brs, NH). FAB-
MS m/z: 296 (M+Na)+. FAB-HR MS m/z: calcd for
C14H27NO4Na: 296.1838, found: 296.1837; 11: colorless oil.
IR umax (KBr) cmꢂ1: 1640. 1H NMR (500 MHz, CDCl3) d:
0.87 (3H, t, J=7.2 Hz, ðCH2Þ4CH3), 1.10 (3H, t, J=7.2,
NCH2CH3), 1.15 (3H, t, J=6.9 Hz, NCH2CH3), 1.25–1.45
(6H, m), 1.50–1.85 (6H, m), 2.27 (1H, dd, J=15.4, 5.7 Hz, Ha-
7), 2.54 (1H, dd, J=15.4, 5.9 Hz, Hb-7, syn), 2.94 (1H, dd-like,
J=ca. 14, 8 Hz, Hb-7, anti), 3.19–3.27 (2H, m, NCH2CH3),
3.24 (3H, s, OCH3, syn), 3.28 (3H, s, OCH3, anti), 3.31 (2H, q-
like, J=ca. 7 Hz, NCH2CH3), 4.54 (1H, m, H-6). FAB-MS
m/z: 324 (M+Na)+. FAB-HR MS m/z: calcd for
C16H31O4NNa: 324.2151, found: 324.2169.
In conclusion, we have developed new anti-malarial
peroxides 8 and 9 with in vivo potency by modifying the
ester portion of 3 to the amide function robust to mouse
serum. Further biological potency of 8 and 9 by use of a
complete-cure model with higher dosages is under
investigation.
12. Leo, A. J. Chem. Rev. 1993, 93, 1281. ClogP was calcu-
lated using the computer program (version 4.0, Bio Byte Cor-
poration, Claremont, CA 91711, USA).
Acknowledgements
13. Obata, Y.; Sato, H.; Li, C. J.; Takayama, K.; Higashiyama,
K.; Nagai, T.; Isowa, K. Int. J. Pharm. 2000, 198, 191.
This study was financially supported by Grants-in-Aid
for Scientific Research from the Ministry of Education,
Science, Sports, and Culture of Japan. The authors are
also grateful to the San-Ei Gen Foundation for Food
Chemical Research for financial support.
14. A strain of P. falciparum (FCR3, cycloguanil-resistant
from Gambia) was used in sensitivity test. After synchroniza-
tion by sorbitol treatment, 50 mL of a parasite culture at the
ring stage (0.55% parasitemia and 2% hematocrit) was added
to each well in 96-well microculture plates. The test samples
were dissolved in DMSO and diluted to the appropriate con-
centration using complete medium, then 50 mL of each sample
solution was inoculated. The final concentration of DMSO in
the culture is 1%. After incubation at 37 ꢀC for 48 h, the pro-
liferation of P. falciparum was assessed by Giemsa-stained
smear by observing 10,000 erythrocytes per one thin blood
film in triplicate. Quinine was used as a reference anti-malar-
ial. In this anti-malarial assay, quinine inhibited the prolifera-
tion of P. falciparum in a concentration-dependent manner
with IC50 of 40 ng/mL and IC90 of 90 ng/mL. Cytotoxic
potency was evaluated by the colorimetric MTT assay, in
which mitomycin C used as a positive control showed the IC50
of 0.1 mg/mL.
References and Notes
1. Sachs, J.; Malaney, P. Nature 2002, 415, 680.
2. Winstanley, P. A. Parasitol. Today 2000, 16, 146.
3. Murakami, N.; Kawanishi, M.; Itagaki, S.; Horii, T.;
Kobayashi, M. Tetrahedron Lett. 2001, 42, 7281.
4. Murakami, N.; Mostaqul, H. M.; Tamura, S.; Itagaki, S.;
Horii, T.; Kobayashi, M. Bioorg. Med. Chem. Lett. 2001, 11,
2445.
5. Kobayashi, M.; Kondo, K.; Kitagawa, I. Chem. Pharm.
Bull. 1993, 41, 1324.
6. Murakami, N.; Kawanishi, M.; Itagaki, S.; Horii, T.;
Kobayashi, M. Bioorg. Med. Chem. Lett. 2002, 12, 69.
7. Each sample (30 mL of 0.1 mg/mL solution) was treated
with fresh mouse serum (300 mL) and incubated at 37 ꢀC for 0,
30, 60, 180, 300 min, respectively. After extraction of the mix-
ture with EtOAc, each extract was concentrated under reduced
pressure. The residue was dissolved with 150 mL of n-hexane–
EtOAc(3:2), then an aliquot (25 mL) of the solution was ana-
lyzed by SiO2-phase HPLC (column: YMC-Pack SIL-06
15. In vivo anti-malarial activities of the compounds were
determined by the 4-day suppressive test using mice infected
with P. berghei (NK 65 strain). Five-week-old ddY female
mice obtained in sterile containers from Charles River Breed-
ing Laboratories Inc. (Yokohama, Japan) weighing 24–27 g
were used. They were housed under a natural day–night (12 h
each) cycle at 25 ꢀC. The mice were randomly assigned to
treated groups and housed in cages each containing five indi-
viduals. Parasites are collected by cardiac puncture from a
donor mouse harbouring about 15% parasitemia. The blood is
diluted with one-seventh volume of 3.2% trisodium citrate
solution, then a final concentration of the infected ery-
throcytes was adjusted to 5 ꢁ 106 by adding 0.9% NaCl solu-
tion. Initially, each mouse was inoculated intravenously in the
tail vein with 1 ꢁ 106 parasitized erythrocytes in 0.2 mL of the
infected suspension. Test compounds were prepared at doses of
1, 3, 10, and 30 mg/kg in dimethylsulfoxide and administrated
by 0.1 mL once a day from day 0 to day 3. The first adminis-
tration of the test compound started intraperitoneally 2 h after
parasite inoculation. Parasitemia levels were determined on day
4. To evaluate the anti-malarial activity of the compounds, we
4.6 mm i.d.
ꢁ
150 mm, mobile phase: 3; n-hexane/
EtOH=200:1, 8 and 11; n-hexane/EtOH=90:10, flow rate:
0.5 mL/min, detection: UV 220 nm) to determine the remain-
ing amounts of the test samples by the absolute calibration
method in triplicate.
8. Meshnick, S. R.; Jefford, C. W.; Posner, G. H.; Avery,
M. A.; Peters, W. Parasitol. Today 1996, 12, 79.
9. Akompong, T.; Kadekoppala, M.; Harrison, T.; Oksman,
A.; Goldberg, D. E.; Fujioka, H.; Samuel, B. U.; Sullivan, D.;
Haldar, K. J. Biol. Chem. 2002, 277, 28923.
10. Kisfaludy, L.; Schon, I. Synthesis 1983, 325.
11. In the transformation from 3 (syn/anti=8.3:1) into the

4084
N. Murakami et al. / Bioorg. Med. Chem. Lett. 13 (2003) 4081–4084
prepared tail blood smears and stained them with Giemsa
(E. Merck, Germany). Total 1 ꢁ 104 erythrocytes per one
thin blood film were examined under microscopy. On day
4, parasitemia of control mice were between 30 and 35%.
The suppression of parasitemia was calculated by the for-
mula: [(average of parasitemia for controlꢂaverage of
parasitemia for treated mice)/average parasitemia for
control]ꢁ100. Five infected and dimethylsulfoxide-dosed
mice were used as a control. The data are determined from
the five individuals in duplicate.
16. As for the amide 9, only the syn isomer was obtained.
9: colorless solid. IR umax (KBr) cmꢂ1: 1650, 1560. 1H
NMR (500 MHz, CDCl3) d: 0.88 (3H, t, J=6.7 Hz), 0.91
(3H, t, J=7.3 Hz) [ðCH2Þ4CH3 and NCH2CH3)], 1.27–1.45
(6H, m), 1.55–1.70 (6H, m), 1.81–1.91 (2H, m), 2.33 (1H,
dd, J=15.3, 2.4 Hz, Ha-7), 2.37 (1H, dd, J=15.3, 6.1 Hz,
Hb-7), 3.14–3.30 (2H, m), 3.25 (3H, s, OCH3), 4.38 (1H, m,
H-6), 6.03 (1H, brs, NH). FAB-MS m/z: 310 (M+Na)+.
FAB-HR MS m/z: calcd for C15H29NO4Na: 310.1995,
found: 310.2008.