Synthesis of 5′-methylenearisteromycin and its 2-fluoro derivative with potent antimalarial activity due to inhibition of the parasite S-adenosylhomocysteine hydrolase

Article abstract of DOI:10.1039/b418829b

5′-Methylenearisteromycin (5) and its 2-fluoro derivative 6, which were designed as antimalarial agents because of their AdoHcy hydrolase inhibition, were synthesized from D-ribose, using a stereoselective intramolecular radical cyclization as the key ste

Full text of DOI:10.1039/b418829b

View Article Online / Journal Homepage / Table of Contents for this issue
A R T I C L E
Synthesis of 5ꢀ-methylenearisteromycin and its 2-fluoro derivative
with potent antimalarial activity due to inhibition of the parasite
S-adenosylhomocysteine hydrolase1
Chieko Takagi,a Makoto Sukeda,a Hye-Sook Kim,b Yusuke Wataya,b Saori Yabe,c
Yukio Kitade,c Akira Matsudaa and Satoshi Shuto*a
a Graduate School of Pharmaceutical Sciences, Hokkaido University, Kita-12, Nishi-6, Kita-ku,
Sapporo, 060-0812, Japan. E-mail: shu@pharm.hokudai.ac.jp; Fax: +81-11-706-4980;
Tel: +81-11-706-3229
b Faculty of Pharmaceutical Sciences, Okayama University, Tsushima, Okayama, 700-8530,
Japan
c Faculty of Engineering, Gifu University, Yanagido 1-1, Gifu, 501-1193, Japan
Received 16th December 2004, Accepted 7th February 2005
First published as an Advance Article on the web 28th February 2005
5ꢀ-Methylenearisteromycin (5) and its 2-fluoro derivative 6, which were designed as antimalarial agents because of
their AdoHcy hydrolase inhibition, were synthesized from D-ribose, using a stereoselective intramolecular radical
cyclization as the key step to construct the carbocyclic structure. These compounds were evaluated as AdoHcy
hydrolase inhibitors with the recombinant human and malarial parasite enzymes. Although 5 and 6 were both potent
inhibitors of the malarial parasite AdoHcy hydrolase, the 2-fluoro derivative 6 proved to be superior due to its lower
inhibitory effect on the human enzyme. In addition, 6 was identified as a potent antimalarial agent using an in vitro
assay system with Plasmodium falciparum.
Introduction
The spreading resistance of Plasmodium falciparum (P. falci-
parum) to currently available drugs such as chloroquine, under-
scores the urgent need for the development of new, more effective
antimalarial agents.2 S-Adenosyl-L-homocysteine (AdoHcy) hy-
drolase, which is responsible for the hydrolysis of AdoHcy to
adenosine (Ado) and L-homocysteine (Hcy),3 has been recog-
nized as a new target for antimalarial agents.4 Since the parasite
has its own AdoHcy hydrolase,4d a drug, which inhibits this
hydrolase to increase the parasite AdoHcy level, would be highly
useful in the treatment of malaria. AdoHcy is a potent feedback
inhibitor of cellular transmethylation; consequently, inhibition
of AdoHcy hydrolase increases the levels of AdoHcy thereby
preventing transmethylation reactions using S-adenosyl-L-
methionine (AdoMet) as the methyl donor, e.g., mRNA methy-
lations, which are essential for the proliferation of the parasite.
Naturally occurring carbocyclic adenine nucleosides (Fig. 1)
such as aristeromycin (1) and neplanocin A (2) are known
to inhibit AdoHcy hydrolase.3 In recent years, we have been
engaged in the synthetic study of novel carbocyclic adenine
nucleosides in the hope of finding potent inhibitors to the
enzyme,5,6 and we have shown that these AdoHcy hydrolase
inhibitors actually exhibit antimalarial activity both in vitro and
in vivo.4e,6
For clinical use, such chemotherapeutic drugs should be selec-
tively toxic to eliminate the target pathogens without untoward
side effects. Unfortunately, the enzyme AdoHcy hydrolase is
essential not only for the proliferation of the parasite, but also
for the proliferation of mammalian cells.3 Accordingly, the most
desirable antimalarial drugs are those which inhibit the malarial
parasite AdoHcy hydrolase without affecting the enzyme of the
human cells.
Fig.
1 Carbocyclic adenine nucleosides as AdoHcy hydrolase
inhibitors.
falciparum) = 4.8), while its non-fluoro derivative 3 showed a
lower IC50 value for the human enzyme compared with that for
the parasite enzyme (human, IC50 = 1.1 lM; P. falciparum, IC50
= 3.1 lM; selectivity index = 0.35).6b
On the other hand, it has been recognized that adenine
nucleoside derivatives can be rapidly deaminated by adenosine
deaminase to a chemotherapeutically inactive inosine congener.
Introduction of a halogen atom at the 2-position of adenine
nucleosides allowed resistance to the adenosine deaminase.5c,7
From this metabolic viewpoint, the 2-fluoro-modification of
AdoHcy hydrolase inhibitors could bring about an improvement
of therapeutic potency.
A recent study suggested that introduction of a fluorine atom
at the 2-position of a carbocyclic adenine nucleoside derivative
might improve the selectivity index between human and malarial
parasite AdoHcy hydrolase inhibition.6 Thus, the IC50 values
of the dehydroxymethylated 2-fluoroaristeromycin derivative 4
against the human and the P. falciparum AdoHcy hydrolases
were 63 and 13 lM, respectively (selectivity index (human/P.
Prisbe and co-workers synthesized a series of 5ꢀ-substituted
aristeromycin derivatives in racemic forms, and found that ( )-
5ꢀ-methylenearisteromycin [( )-5] was an inhibitor of rabbit
erythrocyte AdoHcy hydrolase.8 We were interested in the
T h i s j o u r n a l i s
T h e R o y a l S o c i e t y o f C h e m i s t r y 2 0 0 5
O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 1 2 4 5 – 1 2 5 1
1 2 4 5
©

View Article Online
enantioselective synthesis of the eutomer (bioactive enantiomer)
5, the stereochemistry of which should be the same as that
of naturally occurring aristeromycin (1), in order to clarify
its inhibitory effect on both the human and the P. falciparum
AdoHcy hydrolases as well as to confirm its antimalarial
potency. The 2-fluoro derivative 6 might be superior to the
eutomer 5 as an antimalarial agent, particularly in the selectivity
index and the metabolic stability for the reasons mentioned
above.
In this report, we describe the enantioselective synthesis and
biological effect of 5ꢀ-methylenearisteromycin (5) and its 2-fluoro
derivative 6.
Results and discussion
Synthetic plan
In the synthesis of carbocyclic nucleosides, construction of the
carbocycles (particularly in an optically active form) is often
troublesome. We thought that the enantioselective construction
of the desired exomethylenecarbocyclic moiety would be accom-
plished using a radical reaction as the key step.
Fig. 2 Conceivable radical intermediates 1a and 1b derived from 7.
Gaudino and Wilcox synthesized the carbocyclic analogue
9 of D-ribose 5-phosphate, using a key radical reaction of the
heptyne derivative 7, prepared from D-ribose, to give a mixture
of the trans-product 8a and the cis-product 8b in a ratio of 6.4 :
1 in 63% yield (Scheme 1).9 The major product 8a preserves the
proper functional groups with the desired stereochemistry at the
2ꢀ, 3ꢀ, and 4ꢀ-positions for our target 5ꢀ-exomethylenecarbocyclic
nucleosides 5 and 6. Therefore, we decided to employ this
kind of radical reaction in this study. Based on the following
consideration, we expected that the stereoselectivity might be
improved. Stereoselectivity of the radical 5-exo-cyclizations has
been explained by a chair-like transition state model (Beckwith–
Houk model).10 Using this model, the radical reaction of 7
could be interpreted, as shown in Fig. 2, where the cyclization
could occur via the two chair-like transition states Ia and Ib.
Due to steric repulsion between the isopropylideneoxy and the
adjacent methoxymethyloxymethyl moieties in Ib, Ia seemed
to be more advantageous than Ib, and accordingly the trans-
cyclized product 8a might be formed as the major product.
We speculated that if the steric repulsion in the transition
state Ib was greater, the reaction might produce the desired
trans-product with higher selectivity. Thus, we designed the 5-
O-monomethoxytrityl (MMTr) -2,3-O-p-methoxybenzylidene–
protected substrate 10 (Fig. 3), the p-methoxyphenyl (PMP)
Fig. 3 Expected radical cyclization pathway selectively forming 11a.
group of which should be in the endo orientation. We expected
that the transition state IIa could be even more favored, because
the endo-oriented PMP group would exert steric repulsion in the
other transition state IIb, to give the desired trans-cyclization
product 11a selectively via IIa. From the trans-cyclization
product 11a, the target 5-methylenecarbocyclic nucleosides 5
and 6 would be synthesized via a stereo-inverted introduction of
adenine or 2-fluoroadenine at the allylic 1-position. In addition,
the MMTr and p-methoxybenzylidene groups were preferred
due to their easy and simultaneous removal under mild acidic
conditions, since the target compounds 5 and 6 might be
unstable because of the constrained exomethylene-carbocyclic
ring system.
Synthesis of the target compounds
The synthesis of the radical reaction substrate 10 is summa-
rized in Scheme 2. We found that the desired endo-2,3-O-p-
methoxybenzylidene-b-D-ribose (12) was obtained as the major
product (endo/exo = 12 : 1), when D-ribose was treated with
p-MeOPhCH(OMe)2–pyridinium p-toluenesulfonate (PPTS) in
DMF at 0 ◦C. The endo-stereochemistry of the benzylidene moi-
ety was determined by an NOE (2.2%) observed between the 3-
proton and the benzylidine-methyne proton. Since the endo/exo-
mixture could not be separated at this stage, after protection of
the 5-hydroxyl of 12 with MMTr group, nucleophilic addition of
acetylide to the resulting 13 was examined. When 13 was treated
with CH≡CMgBr in THF at −78 ◦C, the corresponding addi-
tion products 14 was obtained in 96% yield. At this point, the
minor exo-benzylidene isomer was successfully removed by silica
gel column chromatography. The Grignard reaction selectively
Scheme 1
1 2 4 6
O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 1 2 4 5 – 1 2 5 1

View Article Online
Scheme 2 Reagents: (a) p-MeOPhCH(OMe)2, PPTS, DMF, 60%; (b)
MMTrCl, py, 98%; (c) CH≡CMgBr, THF, 96%, (d) ClCO2Et, py,
CH2Cl2, 86%; (e) TCDI, CH2Cl2, 87%.
Fig. 4 NOE experimental results of (a) 11a; (b) 11b; and (c) 17.
radical reaction (entry 4), and MgBr2 did not affect the reaction
in terms of either yield or cis/trans-ratio (entry 5). However,
when the radical reaction with V-70L was performed in the
presence of Me3Al at rt, the stereoselectivity was significantly
improved, but an unknown product was formed and the yield
was not sufficient (entry 6). The stereoselectivity of the reaction
with Me3Al as the additive was decreased at higher temperature
(entry 7).
Although the stereoselectivity of the radical cyclization was
not very high, the procedure efficiently provided the key
intermediate 11a in six reaction steps from D-ribose. Thus, with
11a in hand, we next tried to synthesize the target carbocyclic
nucleosides 5 and 6 (Schemes 3 and 4).
After removal of the ethoxycarbonyl group of 11a, intro-
duction of a nucleobase to the resulting 16 was examined
under Mitsunobu reaction conditions (Scheme 3). When 16 and
6-chloropurine was treated with diisopropyl azodicarboxylate
(DIAD) and Ph3P in DMF at 0 ◦C and then at rt, the conden-
sation occurred regio- and stereoselectively to afford the desired
carbocyclic nucleosidic product 17 with b-stereochemistry in
79% yield. An NOE observed between the H-1ꢀ and H-4ꢀ (Fig. 4c)
showed the b-stereochemistry of 17, and the N9-regiochemistry
was confirmed by UV spectral data.13 After treatment of 17 with
NH3–MeOH to produce the corresponding adenine derivative
18, the protecting groups were simultaneously removed with
gave the desired S-product (R/S = 1 : 13),8 the stereochemistry
of which was determined after radical cyclization, as described
below. The hydroxyl at the propargyl position of 14 (R/S-
mixture) was selectively protected by an ethoxycarbonyl group to
give 15, which was obtained in a diastereomerically pure form
after silica gel column chromatography. Treatment of 15 with
N,Nꢀ-thiocarbonyldiimidazole (TCDI) in CH2Cl2 produced the
radical reaction substrate 10.
The radical reaction of 10 was investigated with Bu3SnH (1.1
eq.) as the reductant under various conditions, and the results
are summarized in Table 1. The reaction was first carried out
with AIBN as a radical initiator in benzene under reflux to give a
mixture of the desired trans-product 11a and the cis-product 11b
in 63% yield in a ratio of 3.0 : 1 (entry 1). The stereochemistry of
the products 11a and 11b was confirmed by NOE experiments,
shown in Fig. 4a and 4b. The reaction was next examined at
lower temperature using V-70L11 as an initiator. Although the
radical reaction did not occur at 0 ◦C (entry 2), the radical cy-
clization proceeded efficiently at rt (entry 3). However, the yield
and the ratio did not improve, compared with those at higher
temperature (entry 1). The effect of Lewis acids as an additive on
the reaction was next investigated, since Lewis acids sometimes
improve stereoselectivity in radical reactions.12 Methylaluminum
bis(2,4,6-tri-t-butylphenoxide) (MAT) completely inhibited the
Table 1 Radical cyclization of 10a
Entry
Initiator
Additive
Solvent
Temperature
Yield (%)
11a : 11b
1
2
3
4
5
6
7
AIBN
V-70L
V-70L
V-70L
V-70L
V-70L
AIBN
—
—
—
MAT
MgBr2
Me3Al
Me3Al
Benzene
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
Benzene
80 ◦
0 ◦
Rt
Rt
Rt
C
C
63
0
60
0
66
40
40
3.0 : 1
—
3.0 : 1
—
2.9 : 1
1.0 : 0b
1.6 : 1
Rt
80 ◦
C
a Reaction was carried out with Bu3SnH (1.1 eq.) and an initiator (1.0 eq.) in the absence or the presence of an additive (2.0 eq.). b An unknown
product was obtained in about 40% yield.
O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 1 2 4 5 – 1 2 5 1
1 2 4 7

View Article Online
Table 2 Inhibitory effect of 5ꢀ-exomethylenecarbocyclic nucleosides 5
and 6 on P. falciparum and human AdoHcy hydrolases
AdoHcy hydrolase, IC50/lM
Compound
P. falciparum
Human
Selectivity indexa
5
0.61
2.1
13
0.52
15.7
63
0.85
7.5
4.8
6
4b
a IC50 (human)/IC50 (P. falciparum). b Data were taken from ref. 6b
selectivity index of 0.85. Thus, chiral 5ꢀ-methylenearisteromycin
(5) proved to be a potent AdoHcy hydrolase inhibitor in this
evaluation system, as suggested by the previous results of the
racemic compound with the rabbit enzyme,8 even though the
selectivity index was low.
On the other hand, the 2-fluoro derivative 6 effectively
inhibited the parasite enzyme with an IC50 value of 2.1 lM
and only weakly inhibited the human enzyme (IC50 = 15.7 lM),
where the selectivity index improved to 7.5. These results showed
that introduction of a fluorine atom at the 2-position of 5
improved the efficacy, particularly, in regard to the selectivity
index, as we had expected.15
Scheme 3
Antimalarial effect
The antimalarial activity of 5 and 6 against P. falciparum (FCR-
3 strain) was evaluated in vitro, and the inhibitory effect of
these compounds on cell proliferation was also evaluated with
mouse FM3A cells in the growing phase.16 These results are
summarized in Table 3. While 5ꢀ-methylenearisteromycin (5) was
clearly cytotoxic to FM3A cells with an IC50 value of 0.31 lM,
it had only a weak antimalarial effect (IC50 = 16 lM). The 2-
fluoro derivative 6 showed significant antimalarial activity with
an IC50 value of 0.40 lM, and was demonstrated to exert lower
cytotoxicity against proliferation of FM3A cells (IC50 = 1.5 lM)
compared with 5. The selectivity indexes of 5 and 6 were 0.019
and 3.8, respectively.
Scheme 4
Therefore, 2-fluoro-5ꢀ-methylenearisteromycin (6) was shown
to be an effective antimalarial agent, and more potent than the
previously reported dehydroxymethylated 2-fluoroaristeromycin
derivative 4.6b Although the inhibitory effect of the 2-fluoro
compound 6 on the parasite AdoHcy hydrolase was weaker than
that of the non-fluoro derivative 5, its antimalarial activity was
superior to that of 5. A possible reason for the increased efficacy
of 6 might be that the compound is resistant to Ado deaminase,
keeping its concentration at a relatively higher level in the assay
system compared with the non-fluoro derivative 5.
aqueous AcOH to furnish the desired 5ꢀ-methylenearisteromycin
(5), as shown in Scheme 3.
The 2-fluoro derivative 6 was similarly synthesized from
16 (Scheme 4). When the Mitsunobu reaction of 16 was
performed with 2-fluoroadenine, instead of 6-chloropurine, the
desired carbocyclic nucleoside 19 was obtained along with
its derivative 20, which was formed by the condensation of
Ph3P at the N -position of 19. However, the N P bond was
found to be easily hydrolyzed under mild acidic conditions.
Thus, after the Mitsunobu reaction of 16 and 2-fluoroadenine,
the resulting mixture of 19 and 20, without purification, was
treated with aqueous AcOH to afford the target 2-fluoro-5ꢀ-
methylenearisteromycin (6) in high yield.
6
=
In summary, 2-fluoro-5ꢀ-methylenearisteromycin (6), synthe-
sized from D-ribose, with the key step being a stereoselective
intramolecular radical cyclization to construct the carbocyclic
structure, was identified as a potent antimalarial agent, which
selectively inhibits the malaria parasite AdoHcy hydrolase.
Therefore, compound 6 appears to be a promising drug can-
didate for the treatment of malaria parasite infections.
Inhibitory effect on human and P. falciparum AdoHcy hydrolases
In order to identify effective antimalarial agents, the activity of
which is due to the AdoHcy hydrolase inhibition, it is essential
to examine the inhibitory potency of the compounds with both
the malarial parasite and the human enzymes. Consequently, we
expressed the recombinant P. falciparum and also the human
AdoHcy hydrolases in E. coli and developed a method for the
evaluation of the inhibitors using these enzymes.14 Thus, the
inhibitory effect of the newly synthesized carbocyclic nucleosides
5 and 6 on the P. falciparum and human AdoHcy hydrolases were
evaluated by this method, and the results are shown in Table 2.
5ꢀ-methylenearisteromycin (5) significantly but non-selectively
inhibited both the parasite and the human AdoHcy hydrolases
with IC50 values of 0.61 and 0.52 lM, respectively, and with a
Table 3 In vitro antimalarial activity of 5ꢀ-exomethylenecarbocyclic
nucleosides 5 and 6
IC50/lM
Compound
P. falciparum
FM3A (growing)
Selectivity indexa
5
16
0.4
7.4
0.31
1.5
7.2
0.019
3.8
0.97
6
4b
a IC50 (FM3A)/IC50 (P. falciparum). b Data were taken from ref. 6b
1 2 4 8
O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 1 2 4 5 – 1 2 5 1

View Article Online
CDCl3, Me4Si) 7.42–721 (m, 14 H), 690–6.80 (m, 4 H), 5.75 (s,
1 H), 4.83 (m, 0.07 H), 4.75 (m, 0.93 H), 4.43 (dd, 0.07 H, J 3.2,
7.2 Hz), 4.35 (dd, 0.93 H, J 6.2, 8.5 Hz), 4.21 (m, 2 H), 4.04 (m,
1 H, H-6), 3.81 (s, 3 H), 3.78 (s, 3 H), 3.62 (d, 0.07 H, J 10.0),
3.51 (dd, 0.93 H, J 3.0, 10.0 Hz), 3.53 (dd, 0.07 H, J 6.0, 9.9
Hz), 3.32 (dd, 0.93 H, J 7.0, 9.8 Hz), 3.19 (d, 0.93 H, J 3.6 Hz),
2.92 (d, 0.07 H, J 4.9 Hz), 2.52 (d, 0.93 H, J 2.3 Hz), 3.10 (d,
0.07 H, J 2.3 Hz); m/z (FAB) 566.2309 (M+. C35H34O7 requires
566.2305).
Experimental
General methods
NMR spectra were recorded at 400 (1H) and 100 (13C) MHz,
and are reported in ppm downfield from Me4Si. The 1H NMR
assignments indicated were in agreement with COSY spectra.
Mass spectra were obtained by the fast atom bombardment
(FAB) method. Thin-layer chromatography was performed on
Merck coated plate 60F254. Silica gel chromatography was
performed with Merck silica gel 5715. Reactions were carried
out under an argon atmosphere.
(3S,4R,5R,6R)-3-Ethoxycarbonyloxy-6-hydroxy-7-
[(4-methoxyphenyl)diphenylmethoxy]-
4,5-(p-methoxybenzylidenedioxy)heptyne (15)
2,3-O-(p-Methoxybenzylidene)-b-D-ribofuranose (12)
A mixture of 14 (566 mg, 1.0 mmol), pyridine (243 lL,
3.0 mmol), and ClCO2Et (143 lL, 1.5 mmol) in CH2Cl2
(10 mL) was stirred at 0 C for 2 h. After addition of MeOH,
A mixture of D-ribose (750 mg, 5.0 mmol), p-MePhCH(OMe)2
(3.4 mL, 20 mmol), and PPTS (6.3 g, 25 mmol) in DMF
◦
◦
(50 mL) was stirred at 0 C for 24 h. After neutralization with
the resulting mixture was evaporated, and the residue was
partitioned between AcOEt and saturated aqueous NaHCO3.
The organic layer was washed with H2O and then with brine,
dried (Na2SO4), and evaporated. The residue was purified by
column chromatography (silica gel, 25% AcOEt in hexane) to
give 15 (549 mg, 86%) as a white foam; dH (400 MHz, CDCl3,
Me4Si) 7.46–7.20 (m, 14 H), 6.87–6.81 (m, 4 H), 5.81 (dd, 1 H,
J 2.3, 3.6 Hz), 5.76 (s, 1 H), 4.52 (dd, 1 H, J 3.8, 6.8 Hz), 4.28–
4.19 (m, 4 H), 3.81 (s, 3 H), 3.78 (s, 3 H), 3.47 (dd, 1 H, J 2.7,
9.7 Hz), 3.38 (dd, 1 H, J 5.3, 9.6 Hz), 2.59 (d, 1 H, J 5.3 Hz), 2.57
(d, 1 H, J 2.3 Hz), 1.31 (t, 3 H, J 7.1 Hz); m/z (FAB) 638.2524
(M+. C38H38O9 requires 638.2516).
NaHCO3, the resulting mixture was filtered through Celite, and
the filtrate was evaporated. The residue was purified by column
chromatography (silica gel, 30–50% AcOEt in hexane) to give
12 (800 mg, 60%) as a white solid. The endo/exo ratio was 12 : 1
based on the 1H NMR spectrum; dH (400 MHz, CDCl3, Me4Si)
for endo-isomer 7.43 (d, 2 H, J 8.7 Hz), 6.91 (d, 2 H, J 8.7 Hz),
5.74 (s, 1 H), 5.58 (d, 1 H, J 6.8 Hz), 4.92 (d, 1 H, J 6.2 Hz),
4.68 (d, 1 H, J 6.0 Hz), 4.60 (m, 1 H), 4.01 (d, 1 H, J 7.0 Hz),
3.85–3.76 (m, 5 H), 3.06 (dd, 1 H, J 2.8, 7.3 Hz); NOE irradiate
H-3/observed p-MeOPhCH (2.2%); for exo-isomer d 7.39 (d, 2
H, J 8.5 Hz), 6.91 (d, 2 H, J 8.7 Hz), 5.93 (s, 1 H, CH3OPhCH),
5.54 (d, 1 H, J 7.0 Hz), 5.02 (d, 1 H, J 5.6 Hz), 4.71 (d, 1 H,
J 5.7 Hz), 4.53 (m, 1 H), 4.11 (d, 1 H, J 7.0 Hz), 3.85–3.76
(m, 5 H), 2.93 (dd, 1 H, J 4.5, 5.7 Hz); m/z (FAB) 269.1014
(MH+. C13H17O6 requires 269.1025).
(3S,4R,5S,6R)-3-Ethoxycarbonyloxy-7-[(4-methoxyphenyl)-
diphenylmethoxy]-4,5-(p-methoxybenzylidenedioxy)-
6-(imidazolylthiocarbonyloxy)heptyne (10)
2,3-O-(p-Methoxybenzylidenedioxy)-
5-O-[(4-methoxyphenyl)diphenylmethyl]-D-ribofuranose (13)
A
mixture of 15 (128 mg, 0.20 mmol) and N,Nꢀ-
thiocarbonyldiimidazole (356 mg, 2.0 mmol) in CH2Cl2 (2 mL)
was stirred at rt for 2 d. The resulting mixture was partitioned
between AcOEt and H2O, and the organic layer was washed
with brine, dried (Na2SO4), and evaporated. The residue was
purified by column chromatography (silica gel, 15–30% AcOEt
in hexane) to give 10 (130 mg, 87%) as a white foam (Found C,
67.24; H, 5.35; N, 3.82. C42H40N2O9S requires C, 67.36; H, 5.38;
N, 3.74%); dH (400 MHz, CDCl3, Me4Si) 8.27 (m, 1 H), 7.58
(m, 1 H), 7.42–7.16 (m, 14 H), 7.05 (m, 1 H), 6.86 (m, 2 H), 6.69
(m, 2 H), 5.91 (s, 1 H), 5.82 (m, 1 H), 5.37 (dd, 1 H, J 2.1, 5.2
Hz), 5.01 (dd, 1 H, J 6.2, 8.3 Hz), 4.57 (dd, 1 H, J 6.2, 6.2 Hz),
4.08 (m, 1 H), 3.96 (m, 1 H), 3.83 (s, 3 H), 3.79 (m, 1 H) 3.75
(s, 3 H), 3.61 (dd, 1 H, J 3.6, 11.5 Hz), 2.44 (d, 1 H, J 2.3 Hz),
1.21 (t, 3 H, J 7.2 Hz); dX (100 MHz, CDCl3, Me4Si) 182.0, 160.5,
158.5, 153.5, 144.3, 143.9, 143.8, 136.8, 136.8, 134.8, 130.7,
130.3, 130.2, 128.3, 128.1, 128.1, 128.0, 127.9, 127.9, 127.7,
126.8, 126.8, 117.9, 117.8, 113.6, 113.5, 113.0, 103.7, 103.7, 86.5,
79.3, 77.5, 77.4, 46.5, 75.2, 65.7, 64.8, 60.5, 55.3, 55.2, 14.1; m/z
(FAB) 749 (MH+)
A mixture of 12 (537 mg 2.0 mmol), MMTrCl (695 mg,
2.25 mmol) in pyridine (17 mL) was stirred at rt for 12 h. After
addition of MeOH, the resulting mixture was evaporated, and
the residue was partitioned between AcOEt and aqueous HCl
(1 N). The organic layer was washed with H2O and then with
brine, dried (Na2SO4), and evaporated. The residue was purified
by column chromatography (silica gel, 25% AcOEt in hexane) to
give 13 (1.07 g, 98%, a/b = 4.5 : 1) as a white foam; dH (400 MHz,
CDCl3, Me4Si) for b-anomer (endo/exo = 7.8 : 1) 7.47–7.24 (m,
14 H), 6.94–6.84 (m, 4 H), 5.91 (s, 0.11 H), 5.77 (s, 0.89 H), 5.48
(d, 0.89 H, J 9.6 Hz), 5.44 (d, 0.11 H, J 9.2 Hz), 5.05 (d, 0.11
H, J 6.2 Hz), 4.93 (d, 0.89 H, J 6.2 Hz), 4.82 (d, 1 H, J 6.2 Hz),
4.46 (m, 0.11 H), 4.53 (m, 0.89 H), 4.14 (d, 1 H, J 9.6 Hz), 3.81
(s, 3 H), 3.80 (s, 3 H), 3.46 (dd, 1 H, J 3.4, 10.4 Hz), 3.37 (dd,
1 H, J 2.8, 7.3 Hz); for a-anomer (endo/exo = 12 : 1) 7.47–7.24
(m, 14 H), 6.94–6.84 (m, 4 H), 6.02 (s, 0.08 H), 5.88 (s, 0.92 H),
5.82 (dd, 1 H, J 4.5, 11.7 Hz), 4.86 (dd, 1 H, J 4.3, 6.6 Hz), 4.72
(d, 1 H, J 6.6 Hz), 4.40 (m, 1 H), 3.96 (d, 1 H, J 11.5 Hz), 3.81
(s, 3 H), 3.80 (s, 3 H), 3.48 (dd, 1 H, J 2.8, 9.6 Hz), 3.06 (dd, 1 H,
J 2.9, 9.8 Hz); m/z (FAB) 541.2245 (MH+. C33H33O7 541.2227
requires 541.2227).
(1S,2R,3R,4R)-1-Ethoxycarbonyloxy-2,3-(p-methoxybenzyl-
idenedioxy)-4-{[(4-methoxyphenyl)diphenylmethoxy]methyl}-
5-methylenecyclopentane (11a) and (1S,2R,3R,4S)-
1-ethoxycarbonyloxy-2,3-(p-methoxybenzylidenedioxy)-4-
{[(4-methoxyphenyl)diphenylmethoxy]methyl}-
5-methylenecyclopentane (11b)
(4S,5R,6R)-3,6-Dihydroxy-4,5-(p-methoxybenzylidenedioxy)-7-
[(4-methoxyphenyl)diphenylmethoxy]heptyne (14)
To a solution of 13 (13.5 g, 25 mmol) in THF (50 mL) was added
CH≡CMgBr (0.5 M in THF, 200 mL, 100 mmol) slowly over 2 h
at −78 ◦C, and the mixture was stirred at rt for 14 h. The resulting
mixture was partitioned between AcOEt and aqueous saturated
NH4Cl, and the organic layer was washed with H2O and then
with brine, dried (Na2SO4), and evaporated. The residue was
purified by column chromatography (silica gel, 30–35% AcOEt
in hexane) to give 14 (13.4 g, 96%) as a white foam. The R/S
ratio was 1 : 13 based on the 1H NMR spectrum; dH (400 MHz,
A mixture of10 (150 mg, 0.2 mmol), Bu3SnH (59 lL, 0.22 mmol),
AIBN (16 mg, 0.02 mmol) in benzene (10 mL) was heated under
reflux for 3 h. The resulting mixture was evaporated, and the
residue was purified by column chromatography (silica gel, 8–
10% AcOEt in hexane) to give 11a (59 mg, 47%) and 11b (20 mg,
16%) as white forms. Compound 11a; dH (400 MHz, CDCl3,
Me4Si) 7.44–7.19 (m, 14 H, Ph), 6.87–6.77 (m, 4 H, Ph), 5.71 (s,
=
1 H, CH3OPhCH), 5.57 (m, 1 H, H-1), 5.41 (s, 1 H, C CH), 5.27
O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 1 2 4 5 – 1 2 5 1
1 2 4 9

View Article Online
=
(s, 1 H, C CH), 4.93 (dd, 1 H, H-2, J 5.8, 5.8 Hz), 4.53 (d, 1 H,
After cooling to rt, the resulting mixture was evaporated, and
the residue was purified by column chromatography (silica gel,
2% MeOH in CHCl3) to give 18 (32 mg, 95%) as a white form;
UV (MeOH) kmax 261 nm; dH (400 MHz, CDCl3, Me4Si) 8.26 (s,
1 H), 7.76 (s, 1 H), 7.53–7.21 (m, 14 H), 6.95 (m, 2 H), 6.84 (m,
2 H), 5.96 (s, 1 H), 5.75 (br s, 2 H), 5.11 (m, 1 H), 5.15 (dd, 1 H,
J 5.8, 5.8 Hz), 5.04 (br s, 1 H), 4.85 (dd, 1 H, J 2.7, 6.5 Hz), 4.70
(br s, 1 H), 3.83 (s, 3H), 3.79 (s, 3 H), 3.48 (m, 2 H), 3.28 (m, 1 H);
dC (100 MHz, CDCl3, Me4Si) 160.6, 158.5, 155.4, 152.9, 152.9,
150.0, 147.8, 144.1, 144.1 139.6, 139.6, 135.2, 130.3, 128.4, 128.3,
128.2, 127.8, 126.9, 119.7, 113.8, 113.1, 112.5, 106.5, 86.7, 83.8,
83.0, 64.6, 63.6, 55.3, 55.2, 48.2; m/z (FAB) 668.2863 (MH+.
C40H38N5O5 requires 668.2873).
H-3, J 5.6 Hz), 4.21 (m, 2 H, OCH2CH3), 3.79 (s, 3 H, OCH3),
3.79 (s, 3 H, OCH3), 3.20 (m, 2 H, 4-CH2OMMTr), 2.63 (m, 1
H, H-4), 1.28 (t, 3 H, OCH2CH3, J 7.1 Hz); NOE irradiate
4-CH2OTr/observed H-1 (2.9%), H-3 (4.9%); dC (100 MHz,
CDCl3, Me4Si) 160.54, 158.51, 154.60, 146.92, 144.09, 143.94,
135.18, 130.26, 128.63, 128.57, 128.30, 128.27, 127.89, 127.80,
127.66, 126.91, 126.73, 113.56, 113.10, 112.98, 112.01, 105.21,
86.88, 82.57, 78.61, 65.47, 64.22, 55.29, 55.24, 48.72, 14.28; m/z
(FAB) 621.2563 (M+. C38H38O8 requires 621.2567). Compound
11b; dH (400 MHz, CDCl3, Me4Si) 7.48–7.17 (m, 14 H, Ph),
6.82–6.73 (m, 4 H, Ph), 5.72 (s, 1 H, CH3OPhCH), 5.15 (s, 1 H,
=
=
C CH), 5.12 (m, 1 H, H-1), 4.86 (m, 2 H, C CH, H-3), 4.81
(dd, 1 H, H-2, J 5.7, 5.8 Hz), 4.18 (m, 2 H, OCH2CH3), 3.80 (s, 3
H, OCH3), 3.75 (s, 3 H, OCH3), 3.63 (dd, 1 H, 4ꢀ-CH2OMMTr,
J 8.5, 8.5 Hz), 3.40 (dd, 1 H, 4-CH2OMMTr, J 5.9, 8.8 Hz), 2.63
(m, 1 H, H-4), 1.26 (t, 3 H, OCH2CH3, J 7.2 Hz); NOE irradiate
4-H/observed H-1 (9.5%); m/z (FAB) 621.2580 (M+. C38H38O8
requires 621.2567).
9-[(1R,2S,3R,4R)-2,3,-Dihydroxy-4-hydroxymethyl-5-methyl-
enecyclopentan-1-yl]adenine (5ꢀ-methylenearisteromycin, 5)
A solution of 18 (32 mg, 0.047 mmol) in aqueous AcOH (80%,
1 mL) was heated at 80 ◦C for 20 h. After cooling to rt, the
resulting mixture was evaporated, and the residue was purified
by column chromatography (silica gel, 5–20% MeOH in CHCl3)
to give 5 (13 mg, quant.) as a white form (Found C, 50.72; H,
5.47; N, 24.45. C12H16N5O3·0.4 H2O requires C, 50.66; H, 5.60;
N, 24.62%); UV (MeOH) kmax 260 nm; dH (400 MHz, DMSO-d6,
Me4Si) 8.16 (s, 1H, H-2), 8.08 (s, 1H, H-8), 7.22 (br s, 2H, NH2),
9-{(1R,2S,3R,4R)-2,3-(4-Methoxybenzylidenedioxy)-4-
{[(4-methoxyphenyl)diphenylmethoxy]methyl}-
5-methylenecyclopentan-1-yl}-6-chloropurine (17)
To a solution of 16 (1.10 g, 2.0 mmol), 6-chloropurine (927 mg,
6.0 mmol), and Ph3P (1.57 g, 6.0 mmol) in THF (20 mL)
was slowly added a solution of DIAD (1.26 mL, 6.0 mmol)
in THF (20 mL) at 0 ◦C, and the mixture was stirred at the same
temperature for 3 h and then at rt for 18 h. After evaporation of
the resulting mixture, to the residue was added AcOEt, and the
resulting suspension was filtered. The filtrate was evaporated,
and the residue was purified by column chromatography (silica
gel, 15–20% AcOEt in hexane) to give 17 (1.07 g, 79%) as a white
form; UV (MeOH) kmax 266 nm; dH (400 MHz, CDCl3, Me4Si)
8.64 (s, 1 H, H-2), 8.08 (s, 1 H, H-8), 7.52–7.22 (m, 14 H, Ph),
6.97 (m, 2 H, Ph), 6.84 (m, 2 H, Ph), 5.97 (s, 1 H, CH3OPhCH),
5.23 (dd, 1H, H-1ꢀ, J1 ,2
ꢀ
ꢀ
= 2.5, J1 ,C
ꢀ
= 2.5 Hz), 5.15 (d, 1H,
ꢀ
CH,1
=
CH
2ꢀ-OH, JOH,2
ꢀ
= 6.6 Hz), 5.04 (dd, 1H, C CH, JC=
= 2.4,
=
Jgem = 2.4 Hz), 4.99 (t, 1H, 4ꢀ-CH2OH, JOH,CH = 5.6 Hz), 4.85
(d, 1H, 3ꢀ-OH, JOH,3
ꢀ
= 3.0 Hz), 4.54 (m, 1H, H2-2ꢀ), 4.44(dd, 1H,
= 2.3, Jgem = 2.3 Hz), 4.01 (m, 1H, H-2ꢀ), 3.58
=
C CH, JC=
ꢀ
CH,4
(m, 2H, 4ꢀ-CH2OH), 2.62 (m, 1H, H-4ꢀ); dC (100 MHz, DMSO-
d6, Me4Si) 156.1, 152.1, 149.8, 148.2, 140.6, 120.0, 110.1, 74.3,
71.7, 63.4, 62.6, 51.8; m/z (FAB) 278.1261 (MH+. C12H16N5O3
requires 278.1253). Anal. (C12H16N5O3·0.4 H2O) C, H, N.
9-[(1R,2S,3R,4R)-2,3,-Dihydroxy-4-hydroxymethyl-5-
methylenecyclopentan-1-yl]-2-fluoroadenine
(2-fluoro-5ꢀ-methylenearisteromycin, 6)
5.58 (m, 1 H, H-1ꢀ), 5.15 (dd, 1 H, H-2ꢀ, J 5.7, 5.7 Hz), 5.09 (br s,
1 H, C CH), 4.88 (dd, 1 H, H-3 , J 2.3, 6.4 Hz,), 4.72 (br s, 1 H,
C CH), 3.84 (s, 3 H, OCH3), 3.80 (s, 3 H, OCH3), 3.52 (dd, 1 H,
ꢀ
=
=
4ꢀ-CH2OMMTr, J 5.5, 9.2 Hz), 3.45 (dd, 1 H, 4ꢀ-CH2OMMTr, J
5.9, 9.3 Hz), 3.31 (m, 1 H, H-4ꢀ); NOE irradiate H-3ꢀ/observed
H-6ꢀ (2.0%), irradiate H-1ꢀ/observed H-4ꢀ (2.1%); dC (100 MHz,
CDCl3, Me4Si) 160.7, 158.6, 151.9, 151.6, 151.1, 147.2, 144.4,
144.3, 144.0, 143.9, 135.0, 131.8, 130.3, 128.3, 128.1, 127.8,
127.0, 113.9, 113.3, 113.1, 106.6, 86.9, 83.8, 83.2, 65.4, 63.6, 55.4,
48.1, 21.8; m/z (FAB) 687.2372 (MH+. C40H36N4O5Cl requires
687.2374).
To a solution of 16 (110 mg, 0.20 mmol), 2-fluoroadenine (92 mg,
0.60 mmol), and Ph3P (157 mg, 0.60 mmol) in THF (2 mL)
was slowly added ◦a solution of DIAD (126 lL, 6.0 mmol) in
THF (2 mL) at 0 C, and the mixture was stirred at the same
temperature for 3 h and then at rt for 18 h. After evaporation of
the resulting mixture, to the residue was added AcOEt, and the
resulting suspension was filtered. The filtrate was evaporated,
and the residue was purified by column chromatography (silica
gel, 0.1–2% MeOH in CHCl3) to give a mixture of 19 and 20. A
solution of the mixture in aqueous AcOH (80%, 3 mL) was
heated at 80 ◦C for 20 h. After cooling to rt, the resulting
mixture was evaporated, and the residue was purified by column
chromatography (silica gel, 1–20% MeOH in CHCl3) to give 6
(53 mg, 89%) as a white form (Found C, 47.67; H, 4.76; N, 22.70.
C12H14FN5O3·0.5 H2O requires C, 47.37; H, 4.97; N, 23.02); UV
(MeOH) kmax 262 nm; dH (400 MHz, DMSO-d6, Me4Si) 8.08 (s,
1 H, H-8), 7.71 (br s, 2 H, NH2), 5.14 (br s, 1 H, 2ꢀ-OH), 5.08
(1S,2S,3R,4R)-1-hydroxy-2,3-(4-methoxybenzylidenedioxy)-
4-{[(4-methoxyphenyl)diphenylmethoxy]methyl}-5-
methylenecyclopentane (16)
A mixture of 11a (241 mg, 0.40 mmol) and NaOMe (28% in
MeOH, 154 lL, 0.80 mmol) in THF (1 mL) and MeOH (4 mL)
was stirred at rt for 5 h and then neutralized with Diaion
WK-20 resin (H+ form). After the resin was filtered off, the
filtrate was evaporated, and the residue was purified by column
chromatography (silica gel, 10–15% AcOEt in hexane) to give
16 (206 mg, 93%) as a white form; dH (400 MHz, CDCl3, Me4Si)
7.44–7.22 (m, 14 H), 6.90–6.80 (m, 4 H), 5.72 (s, 1 H), 5.43 (br s,
1 H), 5.24 (br s, 1 H), 4.75 (m, 1 H), 4.68 (dd, 1 H, J 6.0, 6.0
Hz), 4.54 (d, 1 H, J 6.0 Hz), 3.80 (s, 3 H), 3.80 (s, 3 H), 3.18 (d, 2
H, J 4.9 Hz), 2.88 (m, 1 H), 2.35 (d, 1 H, J 10.9 Hz); m/z (FAB)
551.2449 (MH+. C35H35O6 requires 551.2434).
ꢀ
ꢀ
=
(d, 1 H, H-1 ), 5.00 (br s, 1 H, C CH), 4.82 (br s, 2 H, 5 -OH,
ꢀ
ꢀ
=
3 -OH), 4.43 (br s, 1 H, C CH), 4.40 (m, 1 H, H-2 ), 3.95 (d, 1
H, H-3ꢀ, J3 ,2
ꢀ
ꢀ
= 4.0 Hz), 3.49 (m, 2 H, 4ꢀ-CH2OH), 2.55 (m, 1 H,
H-4ꢀ); dC (100 MHz, DMSO-d6, Me4Si) 158.5 (d, J 202.3 Hz),
157.4 (d, J 21.6 Hz), 151.0 (d, J 20.2 Hz), 147.6, 140.7, 117.2,
110.0, 74.0, 71.2, 63.0, 62.3, 51.5; m/z (FAB) 296.1160 (MH+.
C12H15N5O3F requires 296.1159).
Inhibitory effect on AdoHcy hydrolases
Assays were done according to previously reported methods.14
9-{(1R,2S,3R,4R)-2,3-(4-Methoxybenzylidenedioxy)-4-
{[(4-methoxyphenyl)diphenylmethoxy]methyl}-5-
methylenecyclopentan-1-yl}adenine (18)
Antimalarial effect and cytotoxicity
Assays were done according to previously reported methods.15
A solution of 17 (34 mg, 0.05 mmol) in methanolic ammonia
(saturated at 0 ◦C) was heated in a steel tube at 80 ◦C for 12 h.
1 2 5 0
O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 1 2 4 5 – 1 2 5 1

View Article Online
Shuto, Y. Saito, R. Snoeck, G. Andrei, J. Barzarini, E. De Clercq
and A. Matsuda, J. Med. Chem., 1996, 39, 3847–3852; (d) S. Shuto,
T. Obara, Y. Saito, K. Yamashita, M. Tanaka, T. Sasaki, G. Andrei,
R. Snoeck, J. Neyts, E. Padalko, J. Barzarini, E. De Clercq and A.
Matsuda, Chem. Pharm. Bull., 1997, 45, 1163–1168; (e) S. Shuto, S.
Niizuma and A. Matsuda, J. Org. Chem., 1998, 63, 4489–4493.
6 (a) M. Nakanishi, A. Iwata, C. Yatome and Y. Kitade, Tetrahedron,
2002, 58, 1271–1277; (b) Y. Kitade, H. Kojima, F. Zulfiqur, H.-S.
Kim and Y. Wataya, Bioorg. Med. Chem. Lett., 2003, 13, 3963–
3965.
7 (a) J. A. Montgomery, Cancer Res., 1982, 42, 3911–3917; (b) J. A.
Montgomery, in Nucleosides, Nucleotides, and Their Biological Appli-
cations, ed. J. L. Ridout, D. W. Henry and L. M. Beacher, Academic
Press; New York, 1983, pp. 19–46.
Acknowledgements
This investigation was supported by a Grant-in-Aid for Cre-
ative Scientific Research (13NP0401, A.M.) from the Japan
Society for the Promotion of Science and grant for scientific
research (16390031, K.H-S.) and Research on Priority Areas
(14021072W.Y., 16017266, K.H.-S.) from the Ministry of Edu-
cation, Culture, Sports, Science and Technology of Japan. We are
grateful to H. Matsumoto, A. Maeda, S. Oka, and N. Hazama
(Center for Instrumental Analysis, Hokkaido University) for
technical assistance with NMR, MS, and elemental analysis.
8 B. V. B. Madhavan, D. P. C. McGee, R. M. Rydzewski, R. Boehme,
J. C. Martin and E. Prisbe, J. Med. Chem., 1988, 31, 1798–1804.
9 J. J. Gaudino and C. S. Wilcox, J. Am. Chem. Soc., 1990, 112, 4374–
4380.
10 (a) A. L. J. Beckwith and C. H. Schiesser, Tetrahedron, 1985, 41, 3925–
3941; (b) D. P. Curran, N. A. Porter and B. Giese, Stereochemistry of
Radical Reactions, VCH, Weinheim, 1996.
References
1 This report constitutes part 231 of the series Nucleosides and
Nucleotides; for part 230, see: S. Hirano, S. Ichikawa and A.
Matsuda, Angew. Chem., Int. Ed., in press.
2 (a) W. Peters, Br. Med. Bull., 1982, 38, 187–192; (b) A. F. Cowman
and S. J. Foote, Int. J. Parasitol., 1990, 20, 503–513; (c) A. Slater,
Pharmacol. Ther., 1993, 57, 203–235; (d) D. F. Latel, F. Mangou and
J. Tribouley-Duret, Int. J. Parasitol., 1998, 28, 641–51.
11 Y. Kita, A. Sano, T. Yamaguchi, M. Oka, K. Gotanda and M.
Matsugi, Tetrahedron Lett., 1997, 38, 3549–3552.
3 (a) P. M. Ueland, Pharmacol. Rev., 1982, 34, 223–253; (b) M. S. Wolfe
and R. T. Borchardt, J. Med. Chem., 1991, 34, 1521–1530.
12 G. Bar and F. Parsons, Chem. Soc. Rev., 2003, 27, 251–263.
13 (a) A. Toyota, M. Katagiri and C. Kaneko, Synth. Commun., 1993,
23, 1295–1305; (b) C. Bonnal, C. Chavis and M. Lucas, J. Chem.
Soc., Perkin Trans. 1, 1994, 1401–1410.
14 M. Nakanishi, A. Iwata, C. Yamome and Y. Kitade, J. Biochem.,
2001, 129, 101–105.
15 2-Fluoroaristeromycin was also synthesized, which had a weak
inhibitory effect on P. falciparum and human AdoHcy hydrolases
(Y. Kitade, unpublished results). These results will be reported
elsewhere.
16 (a) H.-S. Kim, Y. Shibata, Y. Wataya, K. Tsuchiya, A. Masuyama
and M. Nojima, J. Med. Chem., 1999, 42, 2604–2609; (b) H.-S. Kim,
Y. Nagai, K. Ono, K. Begum, Y. Wataya, Y. Hamada, K. Tsuchiya,
A. Masuyama, M. Nojima and K. J. McCullough, J. Med. Chem.,
2001, 44, 2357–2361.
4 For examples see: (a) W. Trager, M. Robert-Gero and E. Lederer,
FEBS Lett., 1980, 85, 264–266; (b) J. M. Whaun, G. A. George,
N. D. Brown, R. K. Gordon and P. K. Chiang, J. Pharmacol. Exp.
Ther., 1986, 236, 277–283; (c) A. J. R. Bitonti, J. Baumann, E. T.
Jarvi, J. R. McCarthy and P. P. McCann, Biochem. Pharmcol., 1990,
40, 601–606; (d) K. A. Creedon, P. K. Rathod and T. E. Wellems,
J. Biol. Chem., 1994, 269, 16364–16370; (e) S. Shuto, N. Minakawa,
S. Niizuma, H.-S. Kim, Y. Wataya and A. Matsuda, J. Med. Chem.,
2002, 45, 748–751, and references therein.
5 (a) S. Shuto, T. Obara, M. Toriya, M. Hosoya, R. Snoeck, G. Andrei,
J. Balzarini and E. De Clercq, J. Med. Chem., 1992, 35, 324–331; (b) S.
Shuto, T. Obara, Y. Saito, G. Andrei, R. Snoeck, E. De Clercq and
A. Matsuda, J. Med. Chem., 1996, 39, 2392–2399; (c) T. Obara, S.
O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 1 2 4 5 – 1 2 5 1
1 2 5 1