New readily accessible peroxides with high anti-malarial potency

Article abstract of DOI:10.1016/S0960-894X(01)00673-4

In an explorative study for new anti-malarial substances using the methyl esters (1 and 2) of peroxyplakoric acids A₃ and B₃ as scaffolds, 6-carbomethoxymethyl-3-methoxy-3-pentyl-1,2-dioxane, which has been readily synthesized from 6

Full text of DOI:10.1016/S0960-894X(01)00673-4

Bioorganic & Medicinal Chemistry Letters 12 (2002) 69–72
New Readily Accessible Peroxides with High
Anti-Malarial Potency
a
a
b
Nobutoshi Murakami, Motoyuki Kawanishi, Sawako Itagaki,
b
a,
Toshihiro Horii and Motomasa Kobayashi *
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 13 August 2001; accepted 1 October 2001
Abstract—In an explorative study for new anti-malarial substances using the methyl esters (1 and 2) of peroxyplakoric acids A and
3
B as scaffolds, 6-carbomethoxymethyl-3-methoxy-3-pentyl-1,2-dioxane, which has been readily synthesized from 6-keto-a,b-
3
unsaturated ester, was found to exhibit potent anti-malarial activity with high selective toxicity. # 2001 Elsevier Science Ltd. All
rights reserved.
In the last two decades, malaria has regained its status
as a serious threat to the health and economic prosper-
ity of the human race. It is estimated that approximately
reduced as compared with those of 1 and 2 (Table 2).
Similarly, another peroxide relative to 1 and 2, chon-
drillin (4) isolated from a marine sponge of Xestos-
4
3
00 million clinical cases were observed and people in
pongia sp., was shown to display weaker anti-malarial
activity as well as low selective toxicity. These findings
allowed us to presume that the anti-malarial potency of
1 and 2 involving high selective toxicity would be rela-
ted to three functionalities: a 1,2-dioxane ring, a con-
jugated diene portion in a side chain, and a methyl
residue adjacent to a carbomethoxyl group (Fig. 1).
excess of 2.5 million die from this disease each year. Due
to resistance of the vector (Anopheles mosquito) to
insecticides and an ongoing spread of the drug-resistant
strains of Plasmodium falciparum against chloroquine
and other clinically used drugs, exploration of new
1
effective anti-malarials has been needed urgently.
In the previous paper, we reported two potent anti-
malarial spongean peroxides, the methyl esters (1 and 2)
As a preliminary investigation on participation of the
three functions, we designed accessible 6-carbo-
methoxymethyl-3-methoxy-1,2-dioxane analogues pos-
sessing linear side chains on C-3. On the basis of
cLogP, which is an effective parameter to predict both
permeability and in vivo biological activity of drugs,
two analogues were designed as shown in Figure 2.
Namely, one analogue 6 possesses nearly as large cLogP
as 1 and 2, the other 5 has cLogP that is considered to
belong to the range for exerting in vivo biological efficacy.
2
,3
of peroxyplakoric acids A and B . Furthermore, a
3
3
facile construction method for 6-carbomethoxymethyl-
3
5
,6
-methoxy-1,2-dioxane (3), the core structure of 1 and
3
2
, was developed. In our continued program exploring
for more potent anti-malarials by use of 1 and 2 as
scaffold substances, we have undertaken an accessible
synthesis of two congeners (5 and 6). This paper com-
municates the synthetic method of 5 and 6 and an
assessment of their anti-malarial potency.
The synthesis of the two analogues (5 and 6) was
conducted as illustrated in Scheme 1. Treatment of
When 6-carbomethoxymethyl-3-methoxy-3-methyl-1,2-
3
dioxane (3) (a mixture of diastereomers; 3a:3b=4.8:1)
was evaluated for inhibitory effect on proliferation of P.
falciparum, the anti-malarial effect of 3 was significantly
g-butyrolactone (7) with MeNHOMe and Me AlCl
2
7
afforded the corresponding Weinreb amide, which was
converted to a keto-alcohol (8) with pentylmagnesium
bromide in 78% yield for two steps. Swern oxidation of
8
followed by Wittig reaction with (carbethoxy-
methylene)triphenylphosphorane provided an a,b-unsa-
turated ester (10) in 90% yield for two steps. Next, the
*
8
Corresponding author. Tel.: +81-6-6879-8215; fax: +81-6-6879-
210; e-mail: kobayasi@phs.osaka-u.ac.jp
0
960-894X/02/$ - see front matter # 2001 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(01)00673-4

70
N. Murakami et al. / Bioorg. Med. Chem. Lett. 12 (2002) 69–72
(CF ) CHOH with CF CH OH resulted in increase of
the yield of 5. Reduction of the concentrations of 12,
3 2 3 2
Et NH, and reaction temperature stimulated the gen-
2
eration of 5. Eventually, the condition illustrated in
entry 6 afforded 5 (5a:5b=8.3:1) in moderate yield
(62%). By application of this condition, a nonyl con-
gener 6 (6a:6b=6:1) was also synthesized from 13 in a
similar yield as shown in entry 7.
First of all, we assessed biological potency of the ana-
logues without separating the diastereomeric mixtures.
Table 2 summarizes the growth inhibitory activity
against P. falciparum and the selective toxicity index
between parasite and cells of host-animals, human epi-
8
dermoid carcinoma KB 3-1. While the two designed
analogues (5 and 6) showed similar potent growth inhi-
bitory activities against P. falciparum comparable to the
methyl esters (1 and 2) of peroxyplakoric acids A and
3
B , the nonyl congener 6 showed about 10-fold stronger
3
Figure 1.
cytotoxic effect against KB 3-1 cells than 5. With respect
to the pentyl analogue 5, the reduction of cytotoxicity
resulted in a higher selective toxicity score (360) relative
to those of 1 and 2.
Next, the anti-malarial potency of 5 and 6 was eval-
uated after separation of the respective diastereomeric
isomers. In order to carry out unambiguous elucidation
of the relative stereostructures of both isomers, the
methyl congener 3, which showed a simple signal pat-
Figure 2.
1
unsaturated ester 10 was submitted to Sc(OTf) -medi-
3
ated peroxyhemiacetalization to afford 12 in 68%
yield. Contrary to our expectation, the intramolecular
tern in the H NMR spectrum, was analyzed. The cou-
3
pling constants and NOE correlations revealed the
methoxyl groups on C-3 to adopt axial orientation in
both isomers. In the NOESY spectrum of 3a, the fol-
lowing pairs of protons, Hax-4 and Hax-6, and 3-OCH3
and Hax-5, were correlated to each other. On the con-
trary, the NOESY spectrum of 3b showed the correla-
tions among Hax-4 and H-7b, Heq-5 and H-7a, Heq-4
3
Michael addition of 12 with Et NH in (CF ) CHOH
2
3 2
caused considerable reduction of the yield of the desired
,2-dioxane 5 (14%) unlike the preparation of 3.
1
Asignificant amount of formation of an epoxyketone
3
14 in comparison with the preparation for 3 led us to
presume that the larger steric hindrance of the pentyl
group in 5 would bring about a degradation of 5 in
and Hax-5, and Hax-4 and 3-CH . Therefore, the for-
3
mer 3a and the latter 3b were deduced to be the syn- and
9
1
the anti-dioxane, respectively. Comparison of the H
NMR data between 3a and 3b disclosed the conclusive
difference that the signals ascribable to H -7 in 3b were
(CF ) CHOH with high acidity (Table 1, entry 1). In
3 2
fact, treatment of 5 in (CF ) CHOH at room tempera-
3
2
2
ture gave a mixture of 5 and 10 in a ratio of 1:1. On the
other hand, compound 5 was completely stable in
CF CH OH. Thus, the reaction conditions from 12 to 5
resonated at lower field owing to deshielding effect from
oxygen atoms in the dioxane ring as compared with
those in 3a. By application of this physicochemical
property, syn- (5a and 6a) and anti-isomers (5b and 6b)
3
2
were investigated using CF CH OH as a reaction
3
2
1
0
solvent. As shown in entry 2, the replacement of
were completely distinguished after HPLC separation.
Scheme 1. Synthesis of the designed analogues (5 and 6). Reagents and conditions: (a) MeNHOMe, Me
steps 78% for 8 (n=4), 75% for 9 (n=8); (c) (COCl) , DMSO, Et N; (d) Ph , CH Cl , two steps 90% for 10 (n=4), 89% for 11
, MeOH, 68% for 12 (n=4), 67% for 13 (n=8); (f) Et NH, CF CH OH, 62% for 5 (n=4), 60% for 6
2 2 n 3
AlCl; (b) BrMg-R, [R=(CH ) CH ], two
2
3
3
P¼CHCOOCH
3
2
2
2.
O H
2
2
3
2
3
2
(
(
n=8); (e) H
n=8).
2
NCONH
, Sc(OTf)

N. Murakami et al. / Bioorg. Med. Chem. Lett. 12 (2002) 69–72
Table 1. Investigation on Michael addition of the peroxyhemiacetal
71
Entry
Temperature
Time
Concn of 12 (M)
Concn of base (M)
Solvent
Yield (%)a
1
2
5
10
14
1
2
3
4
5
6
rt
rt
0
rt
rt
0
8 h
8 h
2 day
8 h
8 h
0.16
0.16
0.16
0.016
0.016
0.016
0.016
0.01
0.01
0.01
0.001
0.005
0.005
0.005
(CF
CF
CF
CF
CF
CF
CF
3
3
3
3
3
3
3
)
2
CHOH
0
0
0
15
0
14
22
33
51
55
62
60
29
47
45
8
19
18
19
57
31
22
26
26
20
21
CH
CH
CH
CH
CH
CH
2
2
2
2
2
2
OH
OH
OH
OH
OH
OH
2 day
2 day
0
0
b
7
0
a
1
In all reactions, the peroxyhemiacetals (12 and 13) gave the three products exclusively. The yields were estimated by H NMR spectra after con-
centration of reaction mixtures.
The nonyl congener 13 was used in the reaction. The yields of the corresponding congeners were given.
b
As designated in Table 2, the syn-isomers exhibited
higher selective toxicity scores than the anti-ones in
either case. Nevertheless, this purification little affected
the anti-malarial potency of 5 and 6. Finally, participa-
tion of the absolute configurations in the biological
activity was examined in the case of the most potent
analogue 5a. The assessment of the biological property
after chiral column separation11 clarified that the asym-
metric centers of the peroxides are independent of their
anti-malarial potency.
logical activity slightly. Investigation of in vivo anti-
malarial effect of 5 is in progress in our laboratory.
Acknowledgements
The authors are grateful to the Kanae Foundation for
Life and Socio-medical Science for financial support.
This research is also financially supported by a Grant-
in-Aid for Scientific Research on Priority Areas from
the Ministry of Education, Science, Sports, and Culture
of Japan.
In summary, 6-carbomethoxymethyl-3-methoxy-3-pen-
tyl-1,2-dioxane (5), which was accessibly synthesized by
utilizing the spongean peroxides as a scaffold, was
shown to exhibit potent anti-malarial activity with
selective toxicity. Comparing the anti-malarial potency
among 1, 2, and 5, not only the methyl residues on the
asymmetric center but also the diene portions in the side
chains of 1 and 2 were shown to participate in the bio-
References and Notes
1
2
. Butler, D.; Maurice, J.; O’Brien, C. Nature 1997, 386, 535.
. Kobayashi, M.; Kondo, K.; Kitagawa, I. Chem. Pharm.
Bull. 1993, 41, 1324.
. Murakami, N.; Kawanishi, M.; Itagaki, S.; Horii, T.;
Kobayashi, M. Tetrahedron Lett. 2001, 42, 7281.
. Quinoa, E.; Kho, E.; Manes, L. V.; Crews, P. J. Org. Chem.
1986, 51, 4260.
3
Table 2. Anti-malarial activity and cytotoxicity of 1,2-dioxanes
4
Compd
IC50 (mM)
P. falciparum
Selective
toxicity
5
6
. Leo, A. J. Chem. Rev. 1993, 93, 1281.
. Obata, Y.; Sato, H.; Li, C. J.; Takayama, K.; Higashiyama,
KB 3-1
K.; Nagai, T.; Isowa, K. Int. J. Pharm. 2000, 198, 191. ClogP
was calculated using the computer program (version 4.0, Bio
Byte Corporation, Claremont, CA91711, USA).
7. Shimizu, T.; Osako, K.; Nakata, T. Tetrahedron Lett. 1997,
38, 2685.
1
2
3
4
5
0.15
0.12
>2.4
21
28
>0.49
0.63
43
53
23
4.4
4.4
2.2
77
140
230
>0.24
0.12
0.15
0.15
0.14
0.17
0.26
0.22
0.22
<2.6
360
350
150
31
26
8
350
370
5
5
6
6
6
a (syn)
b (anti)
8
. Astrain of P. falciparum (FCR3, cycloguanil-resistant from
Gambia) was used in sensitivity testing. After synchronization
by the sorbitol treatment, a 50 mL of parasite culture at 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 concentra-
tion using complete medium, then a 50 mL of each sample
a (syn)
b (anti)
(
(
+)-5a
ꢀ)-5a
82

72
N. Murakami et al. / Bioorg. Med. Chem. Lett. 12 (2002) 69–72
solution was inoculated. The final concentration of DMSO in
ꢁ
J=15.9, 5.5 Hz, Ha-7), 2.51 (1H, dd, J=15.9, 7.3 Hz, Hb-7),
3.26 (3H, s, 3-OCH ), 3.70 (3H, s, CO CH ), 4.49 (1H, m, H-
the culture is 1%. After incubation at 37 C for 48 h, the pro-
3
2
3
+
liferation of P. falciparum was assessed by Giemsa-stained
smear by observing 10,000 erythrocytes per one thin blood
film. Quinine was used as a reference anti-malarial. In this
anti-malarial assay, quinine inhibited the proliferation of P.
falciparum in a concentration-dependent manner with IC50 of
6). FAB-MS m/z 283 (M+Na) . FAB-HR MS m/z calcd for
Na: 283.1526, found: 283.1521. 5b: colorless oil. IR
13 24 5
C H O
ꢀ
1
1
umax (KBr) cm : 1745. H NMR (500 MHz, CDCl ) d: 0.90
3
(3H, t, J=7.3 Hz, CH ), 1.25–1.50 (8H, m), 1.50–1.80 (3H, m),
2.26 (1H, m), 2.62 (1H, dd, J=15.9, 6.1 Hz, Ha-7), 3.02 (1H,
3
40 ng/mL and IC90 of 90 ng/mL. Cytotoxic potency was eval-
uated by colorimetric MTT assay, in which mitomycin C used
dd, J=15.9, 7.9 Hz, Hb-7), 3.29 (3H, s, 3-OCH
CO CH ), 4.54 (1H, m, H-6). FAB-MS m/z 283 (M+Na) .
3
), 3.71 (3H, s,
+
2
3
as a positive control showed the IC50 of 0.1 mg/mL.
9
FAB-HR MS m/z calcd for C13
283.1521. 6a: colorless oil. IR umax (KBr) cm : 1745.
NMR (500 MHz, CDCl ) d 0.88 (3H, t, J=6.7 Hz, CH ), 1.20–
1.38 (15H, m), 1.58–1.70 (2H, m), 1.75–1.93 (3H, m), 2.38 (1H,
dd, J=15.9, 5.5 Hz, Ha-7), 2.51 (1H, dd, J=15.9, 7.3 Hz, Hb-
24 5
H O Na: 283.1526, found
. 3a: colorless oil. IR umax (KBr) cm : 1745. 1H NMR
ꢀ
1
ꢀ1
1
H
(
500 MHz, CDCl
J=15.3, 4.9, 4.3, 2.4 Hz, Heq-5), 1.67 (1H, ddd, J=16.5, 12.2,
.3 Hz, Hax-4), 1.75 (1H, dddd, J=15.3, 12.2, 11.0, 4.3 Hz,
3
) d 1.19 (3H, s, 3-CH
3
), 1.55 (1H, dddd,
3
3
4
Hax-5), 1.83 (1H, ddd, J=16.5, 4.9, 4.3 Hz, Heq-4), 2.32 (1H,
dd, J=15.9, 5.5 Hz, Ha-7), 2.44 (1H, dd, J=15.9, 7.9 Hz, Hb-
3 2 3
7), 3.26 (3H, s, 3-OCH ), 3.70 (3H, s, CO CH ), 4.49 (1H, m,
+
H-6). FAB-MS m/z 339 (M+Na) . FAB-HR MS m/z calcd
for C H O Na: 339.2156, found: 339.2147. 6b: colorless oil.
7
dddd, J=11.0, 7.9, 5.5, 2.4 Hz, H-6). FAB-MS m/z 227
), 3.24 (3H, s, 3-OCH ), 3.63 (3H, s, CO CH ), 4.43 (1H,
3 2 3
17 32
5
ꢀ
1
1
IR umax (KBr) cm : 1745. H NMR (500 MHz, CDCl
3
) d 0.88
+
(
M+Na) . FAB-HR MS m/z calcd for
C
H
9 16
O
5
Na:
(3H, t-like, J=ca. 6.5 Hz, CH ), 1.19–1.38 (14H, m), 1.38–1.49
3
2
27.2117, found: 227.2101; 3b: colorless oil. IR u
(KBr)
(2H, m), 1.50–1.80 (3H, m), 2.25 (1H, m), 2.62 (1H, dd,
J=15.9, 6.0 Hz, Ha-7), 3.02 (1H, dd, J=15.9, 7.9 Hz, Hb-7),
max
cm : 1745. 1H NMR (500 MHz, CDCl
ꢀ
1
) d 1.22 (3H, s, 3-
CH ), 1.37 (1H, m, Hax-5), 1.69 (1H, d, J=6.1 Hz, Hax-4),
3
3
3 2 3
3.29 (3H, s, 3-OCH ), 3.71 (3H, s, CO CH ), 4.54 (1H, m, H-
+
1
6
2
3
.70 (1H, dd, J=5.5, 1.8 Hz, Heq-4), 2.20 (1H, dddd, J=14.0,
.7, 6.1, 5.5 Hz, Heq-5), 2.54 (1H, dd, J=15.3, 6.1 Hz, Ha-7),
.93 (1H, dd, J=15.3, 7.9 Hz, Hb-7), 3.27 (3H, s, 3-OCH
.64 (3H, s, CO CH ), 4.46 (1H, dddd, J=7.9, 6.7, 6.1, 5.5 Hz,
6). FAB-MS m/z 339 (M+Na) . FAB-HR MS m/z calcd for
C H O Na: 339.2156, found: 339.2147.
1
7
32
5
3
),
11. Enatiomeric mixture of 5a was separated with Chir-
alcel OJ column (4.6 i.d.ꢂ250 mm, Daicel Chemicals Co.
Ltd.) using n-hexane/AcOEt (30:1) as an eluent to furnish
2
3
+
H-6). FAB-MS m/z 227 (M+Na) . FAB-HR MS m/z calcd
for C H O Na: 227.2117, found: 227.2101.
1
ꢁ
(+)-5a {[a]D +227 (c 0.1, CHCl )} and (ꢀ)-5a {[a]
9
16
5
3
D
0. 5a: colorless oil. IR umax (KBr) cm 1: 1745. 1H NMR
ꢀ
ꢀ222 (c 0.1, CHCl
ꢁ
3
)}. The absolute configurations of
(
(
500 MHz, CDCl ), 1.24–1.40
6H, m), 1.58–1.70 (3H, m), 1.75–1.94 (3H, m), 2.38 (1H, dd,
3
) d 0.89 (3H, t, J=7.3 Hz, CH
3
both have never elucidated because of their similar anti-
malarial potency.