Synthesis, conformation and antiviral activity of nucleoside analogues with the (2-hydroxy-1-phenylethoxy)methyl glycone - A family of nucleoside analogues related to d4T and aciclovir

Article abstract of DOI:10.1039/b510056a

A complete family of acyclic nucleoside analogues has been obtained by combining the (2-hydroxy-1-phenylethoxy)methyl glycone with the nucleoside bases adenine, guanine, cytosine, thymine and uracil. In each case both optical antipodes have been prepared

Full text of DOI:10.1039/b510056a

View Online / Journal Homepage / Table of Contents for this issue
P A P E R
Synthesis, conformation and antiviral activity of nucleoside
analogues with the (2-hydroxy-1-phenylethoxy)methyl glycone—a
family of nucleoside analogues related to d4T and aciclovir
David F. Ewing,*a Virginie Glacon,a Christophe Len*b and Grahame Mackenziea
¸
a Centre for Organic and Biological Chemistry, Department of Chemistry, University of Hull,
UK HU6 7RX. E-mail: d.f.ewing@hull.ac.uk
b
`
Synthese et Reactivite des Substances Naturelles, UMR 6514 CNRS, 40 avenue du Recteur
Pineau, 86022 Poitiers, France. E-mail: christophe.len@univ-poitiers.fr
´
´
Received (in Montpellier, France) 15th July 2005, Accepted 17th August 2005
First published as an Advance Article on the web 19th September 2005
A complete family of acyclic nucleoside analogues has been obtained by combining the (2-hydroxy-1-
phenylethoxy)methyl glycone with the nucleoside bases adenine, guanine, cytosine, thymine and uracil. In
each case both optical antipodes have been prepared in enantiomerically pure form from styrene via
asymmetric dihydroxylation. The conformation of the uracil and adenine derivatives has been compared by
comprehensive calculations using the PM3 method with the conformation of 20,30-didehydro-20,30-
dideoxyuridine (d4U) and the analogous 3-hydroxymethyl-1,3-dihydrobenzo[c]furan nucleoside (bfU). Some
antiviral activity has been observed.
crucial to the management of hepatitis B associated with organ
transplantation).18
Introduction
The realisation that nucleoside analogues have chemothera-
peutic potential as antiviral agents has led to prodigious
chemical effort directed to the synthesis of a very large number
of structural variations to both the heterocyclic base and the
sugar moiety of regular nucleosides. Although the discovery of
new anti-HIV chemotherapeutic agents has been of greatest
interest,1 compounds with activity against many other viral
infections have also been sought and found.2
Of particular importance has been the discovery of antiviral
activity in 20,30-didehydro-20,30-dideoxynucleosides (d4Ns, 1)
especially the thymidine derivative (d4T, 2),3 now a well known
agent (stavudine)4 employed in protocols for the clinical treat-
ment of AIDS through inhibition of reverse transcriptase. This
clinical advance has in turn generated intense effort to design
further structural variants of the d4N template and recent
examples include synthetic schemes to replace the ring oxygen
with carbon,5 to insert a 20- or 30-fluorine substituent,6 to move
the 40-substituent to the 30 position,7 to change the position of
the double bond,8 and to change the ring size.9 Although the
full range of standard nucleoside bases has not been employed
in all cases, several interesting examples of antiviral activity
have been found in this way. A parallel development in the
synthesis of nucleoside analogues with therapeutic potential
has been the replacement of the sugar ring of the nucleoside
with an acyclic group. This has followed from the highly
successful application of aciclovir (3) (and the valine ester
prodrug valaciclovir) to the treatment of herpes simplex virus
(HSV) infections and related infections produced by the herpes
family of viruses.10,11 Recent work includes the synthesis of
nucleoside bases with a variety of acyclic chains attached at the
glycone position.12–17 These novel acyclic nucleoside analogues
may incorporate an element of unsaturation in the side chains
as exemplified by the nucleoside family 4.13 The presence of an
acyclic glycone can lead to increased diversity of function (for
example as an oligonucleotide conjugate16 or a probe of
enzyme affinity17) and a widening of the activity range (ade-
fovir, 9-{2-[phosphonomethoxy]ethyl}adenine, has become
As part of a continuing effort to study the structure–activity
relationship of antiviral nucleoside analogues, we have re-
ported recently19–21 a number of routes to another variant of
the d4N template in which the glycone double bond is incor-
porated into a benzene ring, giving a derivative of the dihy-
drobenzo[c]furan system. The presence of the benzene ring in
this family of 3-hydroxymethyl-1,3-dihydrobenzo[c]furan nu-
cleosides (bfNs, 5) increases the lipophilicity relative to the
d4N family but there is also a significant increase in molecular
size and rigidity. A recent study22 of the anti-HIV activity of
the thymine member of the family of bfNs (5, B ¼ T) has found
that this compound and the analogous 50-triphosphate were
inactive as inhibitors of HIV-1 reverse transcriptase. This lack
of activity is likely to be due to the steric effect of the rigid
benzene ring, which reduces the ability of the molecule to
access the enzyme binding site. We now report the preparation
of another complete family of ten nucleosides based on the
(2-hydroxy-1-phenylethoxy)methyl glycone. This family of
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 a n d t h e
C e n t r e N a t i o n a l d e l a R e c h e r c h e S c i e n t i f i q u e 2 0 0 5
N e w J . C h e m . , 2 0 0 5 , 2 9 , 1 4 6 1 – 1 4 6 8
1461

View Online
9 by treatment with dimethoxymethane and P2O5 in chloro-
form. To promote better glycosidation the methoxy/acetyl
exchange was accomplished with boron trifluoride–diethyl
ether in acetic anhydride at 4 1C to give the corresponding
acetate 10 in 89% yield.
acyclic nucleosides 6 can be regarded as seco-derivatives of the
bfNs which implies that they will show the same increased
lipophilicity relative to the d4Ns but with a large increase in
molecular flexibility. More importantly, the nucleoside family
6 is obviously closely related to families 1 and 5 and provides
an interesting link between d4T 2 and aciclovir 3.
The benzoylated nucleosides 11–15 were obtained in 70–
98% yield by application of standard Vorbruggen chemistry
(condensation with the silylated purine or pyrimidine base in
anhydrous acetonitrile and toluene using dibenzo-18-crown-6-
ether in the presence of KI for phase transfer catalysis). The
purine bases reacted to give the N-9 isomers 11 and 12 with no
evidence for the formation of the analogous N-7 isomers.
Similarly the pyrimidine derivatives were obtained only as
the N-1 isomers 13–15. The benzoyl groups were removed
and the nucleoside analogues 16–20 were isolated in satisfac-
tory yields. The complete sequence was carried through start-
ing from both the (R)- and the (S)-diol (7). The retention of
chiral integrity was confirmed for the acetates (R)-10 and (S)-
10 by NMR spectroscopy using the europium chiral shift
reagent Eu(D-3-trifluoroacetylcamphorate)3 and for the nu-
cleoside analogues 11–15 by chiral chromatography on a
variety of columns.27 In most cases the enantiomeric purity
was 99% or better.
Although many variants of the (2-hydroxyethoxy)methyl
glycone present in 3 have been synthesised, there are few
reports of nucleosides incorporating the corresponding phenyl
substituted glycone. Pan et al.23 obtained some racemic pyri-
midine derivatives and this system was included in a German
patent24 but no chiral strategies were proposed. Some of our
preliminary studies have also been reported elsewhere.25
Results and discussion
Synthesis
The family of nucleosides 6 was obtained by the five step
sequence of reactions shown in Scheme 1. The starting phenyl-
ethanediols, (R)-7 and (S)-7, were obtained in 99% yield and
97% ee by asymmetric dihydroxylation of styrene using the
available reagents AD-mix b and AD-mix a, respectively.26
With the primary hydroxyl group protected as the benzoate
(8), the first attempt to derivatise the secondary hydroxyl group
involved conversion to a chloromethyl ether23 with parafor-
maldehyde and dry hydrogen chloride in methylene chloride.
This reaction proceeds in poor yield and was abandoned in
favour of the high yield conversion to the methoxymethyl ether
Molecular modelling
In order to investigate the shape and flexibility of the nucleo-
side analogues in series 6, especially when compared to both
the bfN series 5 and the d4N series 1, detailed calculations were
made at the semi-empirical level for each system. All calcula-
tions were carried out using the PM3 semi-empirical method in
the MOPAC 97 package as implemented in the Chem3D Ultra
(v. 5.0 or v. 6.0) program.28 The PM3 method has been widely
used for carbohydrate and nucleoside calculations.29 Indivi-
dual conformational minima were established by repeated
optimisation from at least six deviating conformations, ran-
domly selected, using the standard MOPAC convergence
criterion and convergence was finally confirmed by a calcula-
tion at 100-fold increase in SCF precision. Every combination
of rotatable bonds was investigated in this way using the
appropriate twofold or threefold rotational potential. The
values of the heat of formation for individual conformations
(DHf) are reproducible within ꢀ0.01 kJ molꢁ1
.
Although the thymine derivative d4T is the most active
compound in the d4N family, the present comparative analysis
was carried out on the analogous uracil, d4U (22) and the
corresponding uracil derivatives of the other two systems 20
and 21 since the presence of a methyl group in the pyrimidin-
2,4-dione ring has no effect on the relative energy (DHf) of
individual conformers corresponding to changes in the rest of
the molecule. Unfortunately the numbering is different in the
conformationally mobile part of these derivatives and, for
convenience, this numbering is shown in structures 16, 20–22
together with the labelling of the torsion angles. Where a
threefold rotational potential applies the conformation is
described by the conventional labels t, g1 or gꢁ for the variable
dihedral angles (although individual angles can deviate by 201
or more from the ideal value) and the overall molecular
conformation is described by a suitable combination of these
labels, one for each rotatable bond as defined below. For
Scheme 1 a: BzCl, pyridine; b: dimethoxymethane, P2O5, chloroform;
c: acetic anhydride, BF3 ꢂ Et2O; d: silylated base, dibenzo-18-crown-6-
ether, KI, acetonitrile, toluene; e: NH3 in MeOH.
1462
N e w J . C h e m . , 2 0 0 5 , 2 9 , 1 4 6 1 – 1 4 6 8

View Online
Table 1 Relative energies, populations and selected geometrical parameters for six conformers of 21
Relative DHf
Relative %
populationc
Separation of
N1 and C80 (A)
Separation of
H6 and H80 (A)
a
f2 (1)
b
f3 (1)
Conformer
(kJ molꢁ1
)
g1gꢁ
t g1
g1g1
t gꢁ
gꢁgꢁ
gꢁg1
a
75
172
65
ꢁ61
52
0.0
3.8
5.3
6.2
7.6
8.1
67
14
8
4.0
4.0
4.0
3.9
4.0
4.3
e
1.73
1.74
1.73
1.76
1.84d
1.77e
54
172
ꢁ75
ꢁ76
b
ꢁ63
ꢁ66
54
5
3
2
c
d
Angle O20C30C80O9. Angle C30C80O90H90. Based on all nine conformers. Separation of N1 and O90. Separation of H6 and H50.
compound 20 a more subtle labelling is required and that is
indicated below.
with a population greater than 1% are given in Table 2. The
average deviation from a plane for the ring atoms of the
dihydrofuran ring is less than 0.01 A. The uracil ring is planar,
and these two ring planes are approximately orthogonal, with
the line of intersection lying between C40 and C50. Although
there is a slight reordering of the populations of the minor
rotamers, the main species is again the g1gꢁ form, with a
virtually identical geometrical relationship between the ring
substituents as that observed in the bfU analogue 21. The very
strong similarity of the dominant conformation for the d4U
and bfU species is confirmed by the models in Fig. 1.
Conformational analysis of (1R,3S)-1-(3-hydroxymethyl-1,3-
dihydrobenzo[c]furan-1-yl)uracil (21)30
The uracil ring is expected to show a twofold rotational
potential with angle C2N1C10O20 (f1) around þ601 or
ꢁ1201. The anti orientation, f1 B ꢁ1201, was favoured by
16–17 kJ molꢁ1 for all conformations so the syn conformation
was not considered further. Angle f1 is ꢁ1191 in the lowest
energy conformation (the g1gꢁ in Table 1) and in the range
ꢁ1001 to ꢁ1211 in the other populated forms. There are nine
possible rotamers for 21 corresponding to rotation of the
CH2OH moiety and variations in angles O20C30C80O90 (f2)
and C30C80O90H90 (f3). A full geometry optimisation was
achieved in each case and details for the six significant con-
formers are given in Table 1. In all conformations the average
deviation from a plane for the nine ring atoms of the benzo[c]-
furan ring was in the range 0.013–0.028 A. This is a negligible
deviation, i.e. in every conformation the bicyclic dihydroben-
zo[c]furan was essentially planar. The uracil ring was also
planar, as expected, and these two ring planes are approxi-
mately orthogonal with the line of intersection lying between
C30 and C30a.
Of the six conformations with a population greater than 1%
(Table 1) the g1gꢁ form (labels correspond to f2 then f3)
clearly predominates, due in part to the weak H-bond which is
formed with the ring oxygen (H90ꢂ ꢂ ꢂO20 separation is 2.8 A).
The shape of the mobile part of the molecule varies very little
between the main conformers as shown (Table 1) by the
separation between N1 and C80, the two substituent atoms
attached to the dihydrofuran ring, and by the magnitude of the
closest non-bonded contact (usually H6ꢂ ꢂ ꢂH80).
Conformational analysis of 1-[(2-hydroxy-1-
phenylethoxy)methyl]uracil (20)
Although there are six rotatable bonds in compound 20, a full
analysis of the conformational space was undertaken, to
investigate not only the stable minima but also the overall
flexibility of this acyclic nucleoside analogue. Around two
thousand individual minimisations were undertaken including
some preliminary conformational calculations on model com-
pounds. The conformation around the N1–C10 bond linking
the uracil ring to the acyclic side chain in compound 20 is
described by a twofold rotational potential. The C10O20 bond
is approximately orthogonal to the uracil ring plane in both
orientations of the ring and hence the two conformations of the
dihedral angle C2N1C10O20 are labelled as o1 (f1 B þ901)
and oꢁ (f1 B ꢁ901). The position of the actual minimum
depends on the conformation of the rest of the molecule and
can deviate in some cases towards the position where one or
other of the H10 protons is eclipsed by the ring.
The conformation around the ether oxygen varied markedly
but, in the main, the ether linkages could each be treated as a
threefold potential with minima in the trans (t) and both
orthogonal positions (o1 and oꢁ). The gauche positions were
inaccessible except for g1 for f3. Thus, over the set of
conformations, angle N1C10O20C30 falls in three ranges, trans
(t) (f2 ¼ ꢁ1561 to ꢁ1671), orthogonal (o1) (f2 ¼ 841 to 1031)
and oꢁ (f2 ¼ ꢁ901 to ꢁ1101). Angle C10O20C30C40 falls in
ranges t (f3 ¼ ꢁ1621 to ꢁ1741), g1 (f3 ¼ 541 to 801), oꢁ(f3 ¼
ꢁ781 to ꢁ1051). In a few cases f2 and f3 are in an intermediate
range (1231 to 1351) labelled as anticlinal (a). The remaining
two dihedral angles in the side chain (f4 ¼ O20C30C40O50 and
f5 ¼ C30C40O50H50) are labelled as g1, gꢁ or t, although
Conformational analysis of (2R,5S)-1-(5-hydroxymethyl-2,5-
dihydrofuran-2-yl)uracil (22)31
Compound 22 is the member of the d4N family analogous to
the bfU species 21. The uracil ring again shows a twofold
rotational potential with the anti orientation strongly favoured
by about 18 kJ molꢁ1. The nine possible conformations for 22
were fully optimised for variations in angles O10C50C60O70
(f2) and C50C60O70H70 (f3) and details of the five conformers
Table 2 Relative energies, populations and selected geometrical parameters for five conformers of 22
Relative DHf
Relative %
populationc
Separation of
N1 and C60 (A)
Separation of
H6 and H60 (A)
a
b
Conformer
f2 (1)
f3 (1)
(kJ molꢁ1
)
g1gꢁ
t gꢁ
t g1
g1g1
gꢁg1
a
75
174
170
66
ꢁ60
ꢁ67
62
0.0
4.2
4.8
5.3
6.2
65
12
9
4.0
3.9
4.0
4.0
4.2
1.73
1.75
1.74
1.73
1.77
54
56
7
ꢁ77
b
5
c
Angle O10C50C60O70. Angle C50C60O70H70. Based on all nine conformers.
N e w J . C h e m . , 2 0 0 5 , 2 9 , 1 4 6 1 – 1 4 6 8
1463

View Online
Fig. 2 The oꢁo1t g1gꢁ (global minimum) conformation (a) and the
Fig. 1 The anti-g1gꢁ (global minimum) conformations of compound
21 and compound 22.
o1oꢁaꢁg1gꢁ conformation (b) of compound (S)-20.
values can deviate by ꢀ201 from the normal values of a
staggered chain.
The conformation of the phenyl group was generally close to
the position where the ring plane bisected the O20C30C40 angle
i.e. the C30–H30 bond was close to the benzene ring plane.32
(The orthogonal orientation for this bond was also explored).
In a few cases, for both the uracil and phenyl rings, a second
minimum of similar energy was found corresponding to ca. 201
rotation with respect to the primary conformation. These
additional forms are ignored since the overall molecular shape
is little affected.
There are 162 notional conformations for the combination
of dihedral angles detailed above for compound 20 but some of
these are impossibly crowded and are so unstable the mini-
misation reverts to another form. About 140 stable minima
were found and of these 18 had a relative DHf of 5 kJ molꢁ1 or
less, corresponding to a relative population of 1% or more.
Details of the ten conformations with the lowest energy are
given in Table 3. It is clear that compound 20 has great
flexibility and the global minimum is not as highly populated
as is the case for the rigid compounds 21 and 22.
Fig. 3 The oꢁo1t g1gꢁ (global minimum) conformation of compound
(S)-16.
shape over a large number of conformers confirms that com-
pound 20 is extremely flexible and able to adopt a range of
conformations of similar energy. Thus this family of nucleoside
analogues is of interest for potential antiviral activity.
At the global minimum compound 20 has the oꢁo1t g1gꢁ
conformation (labels in the order of the torsion angles f1 to f5)
which has the shape shown in Fig. 2(a). The two aromatic rings
lie in approximately parallel planes bridged by the ether
oxygen. The next lowest energy conformation (Table 3) has a
very similar overall shape and the third conformation is also
similar but with the uracil ring reversed. However this extended
linear shape is not the only available form since the fourth
entry in the table is a o1oꢁaꢁg1gꢁ conformation which has no
trans linkages and the shape shown in Fig. 2(b). In this case the
aromatic rings are facing and the reactive site is fully exposed
at the bottom of the V-shaped molecule. A range of molecular
shapes between these two extreme forms is found among the
lower population conformers. This diversity of molecular
Conformational analysis of 9-[(2-hydroxy-1-
phenylethoxy)methyl]adenine (16)
The adenine analogue 16 was also investigated over the
range of conformations which had a significant population in
the uracil derivative. Details of the five lowest energy forms
are given in Table 4. The global minimum conformation for the
adenine derivative (Fig. 3) is virtually identical to that for
the uracil derivative. Apart from a slight reordering of energies,
the five lowest energy forms are the same in both the uracil and
adenine derivatives (20 and 16, respectively). This suggests that
the whole class of acyclic nucleoside of type 6 is likely to show
the increased flexibility required for potential antiviral activity.
Table 3 Relative energies, populations and selected geometrical parameters for the ten most stable conformers of 20
Relative DHf
Relative %
populationh
Conformera
(kJ molꢁ1
)
b
f1 (1)
c
f2 (1)
d
f3 (1)
e
f4 (1)
f
f5 (1)
g
f6 (1)
oꢁo1t g1gꢁ
oꢁo1t t g1
o1o1t g1gꢁ
o1oꢁaꢁg1gꢁ
o1t g1g1gꢁ
oꢁo1t g1g1
o1oꢁgꢁg1g1
oꢁa1gꢁg1gꢁ
a1aꢁg1g1g1
o1oꢁgꢁgꢁgꢁ
a
ꢁ96
ꢁ97
96
97
95
ꢁ169
ꢁ168
ꢁ168
ꢁ134
57
74
163
73
ꢁ66
59
ꢁ4
19
0
20
10
9
1.7
1.8
2.7
2.7
2.9
3.1
3.3
3.8
3.9
94
ꢁ70
ꢁ66
ꢁ72
63
ꢁ11
ꢁ11
8
102
80
ꢁ93
ꢁ156
99
71
76
7
6
ꢁ94
96
ꢁ164
ꢁ89
ꢁ93
64
77
76
1
6
ꢁ95
142
62
ꢁ6
ꢁ8
1
6
ꢁ82
112
94
81
56
ꢁ66
59
5
ꢁ109
4
ꢁ98
ꢁ78
ꢁ40
ꢁ45
ꢁ5
4
The conformational labels are in the order of the dihedral angles f1 through to f5. The symbols o and a mean orthogonal and anticlinal,
b
c
d
e
f
g
respectively. Angle C2N1C10O20. Angle N1C10O20C30. Angle C10O20C30C40. Angle O20C30C40O50. Angle C30C40O50H50. Angle
h
H30C30C100C200. Based on all conformers with a population 41%.
1464
N e w J . C h e m . , 2 0 0 5 , 2 9 , 1 4 6 1 – 1 4 6 8

View Online
Table 4 Relative energies, populations and selected geometrical parameters for the five most stable conformers of 16
Relative DHf
Relative %
populationh
Conformera
(kJ molꢁ1
)
b
f1 (1)
c
f2 (1)
d
f3 (1)
e
f4 (1)
f
f5 (1)
g
f6 (1)
oꢁo1t g1gꢁ
o1o1t g1gꢁ
oꢁo1t t g1
a1oꢁaꢁg1gꢁ
a1aꢁg1g1gꢁ
a
ꢁ102
85
82
93
ꢁ175
ꢁ169
ꢁ167
ꢁ140
67
75
76
ꢁ66
ꢁ64
58
ꢁ5
ꢁ6
ꢁ6
ꢁ11
2
0
30
13
11
9
2.1
2.5
3.2
3.4
ꢁ102
113
118
92
ꢁ93
170
72
ꢁ66
ꢁ114
72
ꢁ62
8
The conformational labels are in the order of the dihedral angles f1 through to f5. The symbols o and a mean orthogonal and anticlinal,
b
c
d
e
f
g
respectively. Angle C4N1C10O20. Angle N1C10O20C30. Angle C10O20C30C40. Angle O20C30C40O50. Angle C30C40O50H50. Angle
H30C30C100C200. Based on all conformers with a population 41%.
h
(R)-14 against the Punta Toro virus, about five time less active
than ribavirin (48 mg mlꢁ1).
Biological results
The set of twenty compounds 11–20, comprising each enantio-
mer of the five different nucleosides with or without benzoyl
protection, was tested against a wide range of viruses using
three different cell lines.
In each of the cell lines employed for the above screens,
cytoxicity of the twenty nucleoside analogues was low, similar
to that determined for the reference antiviral compounds.
The antiherpes activity found for the adenine derivative
(S)-11 is the most interesting finding from the thirteen screens
reported above. The stereochemical selectivity favours the
configuration opposite to that in D-ribose, a consequence of
the conformational freedom conferred by the acyclic nature of
these compounds and established by the modelling studies
described above. Generally the compounds with a free hydro-
xyl group were inactive, probably indicating poor transport
through the cell wall.
Using HEL cell cultures antiviral activity was examined for
herpes simplex virus type 1 (HSV-1 KOS) and type 2 (HSV-2
G), an aciclovir resistant strain (HSV-1 ACVr), vaccinia virus
(VV) and vesicular stomatitis virus (VSV). Four established
antiviral compounds were used for comparison, namely 2-(2-
bromovinyl)deoxyuridine (BVDU), 1-b-D-ribofuranosyl-1,2,4-
triazole-3-carboxamide (ribavirin, RB), aciclovir 3 (ACV) and
9-[2-hydroxy-1-(hydroxymethyl)ethoxymethyl]guanine (ganci-
clovir, GCV). Activity was determined as the minimum
inhibitory concentration (mic), the concentration required to
reduce virus induced cytopathogenicity by 50%.
Conclusion
None of compounds 11–20 had activity approaching that of
ACV and GCV against HSV-1 KOS and HSV-2 G. However,
weak but significant antiherpes activity was observed for the
adenine derivative (S)-11 (mic 16 mg mlꢁ1) against HSV-1
KOS. It is notable that there is a clear dependence on substrate
stereochemistry because the activity of the other stereoisomer
(R)-11 is less (mic 480 mg mlꢁ1) and similar to that of most
other compounds in the set. The only other compound with
some activity was the cytosine derivative (R)-18 (mic 48 mg
mlꢁ1), where the stereochemical effect is reversed [(S)-18 has
mic 480 mg mlꢁ1].
Stereochemical discrimination is also evident in the activity
of the benzoylated adenine derivatives against HSV-1 ACVr,
VV and VSV although exact mic values were not determined
[(S)-11 mic 416 mg mlꢁ1 and (R)-11 480 mg mlꢁ1]. It is inter-
esting that the weak activity of (S)-11 against VSV is probably
significantly higher than that of any of the reference antivirals.
Using HeLa cell cultures the antiviral activities of 11–20
were determined for VSV, Coxsackie B4 virus (CB4V), and
respiratory syncytial virus (RSV). For these screens the refer-
ence antiviral compounds were BDVU, RB and (S)-9-(2,3-
dihydroxypropyl)adenine [(S)-DHPA].
None of compounds 11–20 had activity approaching that of
RB against VSV, CB4V and RSV but several of the benzoy-
lated compounds were slightly more active than BVDU or (S)-
DHPA. Furthermore this mild activity again showed a clear
dependence on the stereochemistry of the tested compound.
Thus the adenine derivative (S)-11 and the uracil derivative
(S)-15 showed mild activity against VSV, CB4V and RSV,
whereas the other enantiomers (R)-11 and (R)-15 were inactive
(mic 4400 mg mlꢁ1).
The modelling studies have shown that the conversion of the
nucleoside family 5 to nucleoside family 6 introduces extensive
flexibility which is likely to greatly increase the potential for
antiviral activity as observed in aciclovir and prodrugs.33 It is
also notable that some nucleosides with a hydroxylated acyclic
glycone have potent hepatitis B activity34 but the potential of
nucleoside family 6 extends into the wider antiviral area. This is
confirmed by finding mild activity against a wide range of
viruses.
Experimental
General procedures
NMR spectra were recorded with a JEOL Lambda 400 spec-
trometer using standard conditions with a data point resolu-
tion of ca. 0.1 Hz. 1H chemical shifts were measured relative to
Me4Si and 13C chemical shifts relative to CDCl3 (77.05 ppm),
(CD3)2SO (39.5 ppm) or CD3OD (49.0 ppm). All coupling
constants are given in hertz. Assignments of the 1H spectra
were made by detailed analysis using decoupling or correlation
techniques where appropriate. Column chromatography was
performed on silica gel (230–400 mesh; Prolabo) and TLC on
silica gel 60, F254 (Merck) with detection by UV absorbance or
phosphomolybdic acid. Chiral HPLC was carried out using
Chiralcel OD-H and Chiralpak AD columns [5 mm cellulose
tris(3,5-dimethylphenylcarbamate)], a Chiralpak AS column
[amylose tris{(S)-1-phenylethylcarbamate}] and a Chiralcel
OJ column [10 mm cellulose tris(methylbenzoate)] (all 250 ꢃ
4.6 mm stainless steel columns from Daicel Chemical Indus-
tries). Optical rotation values are given in 10ꢁ1 deg cm2 gꢁ1
.
Using Vero cell cultures the activity was also investigated
against parainfluenza-3 virus, reovirus-1, Sindbis virus, Cox-
sackie B4 virus and Punta Toro virus. All of compounds 11–20
showed mild activity against all of the above viruses [against
which BDVU and (S)-DHPA were inactive] but with no
evidence of stereochemical selectivity. Ribavirin also show mild
activity against all of the above viruses except Coxsackie B4. A
mic value of 240 mg mlꢁ1 was determined for both (S)-14 and
(R)- and (S)-Phenylethane-1,2-diol [(R)-7 and (S)-7]
AD-mix a or AD-mix b (50.0 g) in a mixture of tert-butanol
(180 mL) and water (180 mL) was stirred at room temperature
until both phases were clear. Styrene (3.72 g, 35.72 mmol)
was added to this mixture at ꢁ10 1C and the resulting slurry
was stirred vigorously at 0 1C for 1 h. Sodium sulfite (40 g) was
N e w J . C h e m . , 2 0 0 5 , 2 9 , 1 4 6 1 – 1 4 6 8
1465

View Online
added and the mixture stirred at 20 1C for 30 min, then diluted
with water (160 mL) and extracted with dichloromethane. This
extract was worked up and the crude product purified by
column chromatography (gradient of hexane–ethyl acetate,
7 : 3, then 1 : 1) to give the diol (R)-7 or (S)-7 as colourless
crystals, 4.9 g (99%), as reported previously.21
Preparation of nucleoside analogues 11–15
The unprotected nucleobase (A, C, G, T, U) (3.18 mmol) in
hexamethyldisilazane (13 mL) containing a catalytic amount of
NH4(SO4)2 was refluxed overnight then excess reagent cau-
tiously removed under reduced pressure. Dibenzo-18-crown-6-
ether (0.24 g, 0.67 mmol) and potassium iodide (0.22 g, 1.30
mmol) were added to a solution of the silylated nucleobase and
(R)-10 or (S)-10 (0.5 g, 1.59 mmol) in acetonitrile–toluene (1 : 1,
12 mL). The mixture was stirred overnight in the case of
pyrimidines and for 3 days in the case of purines under an
atmosphere of argon. The insoluble material was filtered off
and the filtrate evaporated under reduced pressure. The crude
product was chromatographed using CH2Cl2–CH3OH as spe-
cified below.
(R)- and (S)-(2-Hydroxy-2-phenyl)ethyl benzoate
[(R)-8 and (S)-8]
Benzoyl chloride (4.4 mL, 38.2 mmol) was added to (R)-7 or
(S)-7 (4.8 g, 37.7 mmol) in pyridine (70 mL) at 0 1C and the
mixture stirred for 24 h at room temperature. Ice–water was
added, the mixture stirred for 30 min and then extracted with
CH2Cl2. This extract was worked up and the crude product
purified by column chromatography (hexane–ethyl acetate,
7 : 3) to afford the protected diol (R)-8 or (S)-8 as colourless
crystals, 7.3 g (88%), mp 50 1C; Rf 0.3 (hexane–ethyl acetate,
7 : 3); dH (CDCl3) 4.40 (1H, dd, J 6.0, 8.6, H2a), 4.50 (1H, dd, J
2.5, 8.6, H2b), 5.07 (1H, dd, J 6.0, 2.5, H1), 7.37–7.42 (8H, m,
aromatic H), 8.03 (2H, m, aromatic H); dC (CDCl3) 69.8 (C2),
72.5 (C1), 126.2, 128.2, 128.4, 128.6, 129.7, 133.2, 139.8 (aro-
matic C), 166.7 (CO); (R)-8 had [a]2D2 ꢁ25.7 (c 0.58 in CHCl3);
(found: C, 74.06; H, 5.98; calc. for C15H14O3: C, 74.36; H,
5.82%); (S)-8 had [a]2D2 þ22.8 (c 0.19 in CHCl3); (found: C,
74.42; H, 5.83%).
(R)- and (S)-9-[(2-Benzoyloxy-1-phenylethoxy)methyl]adenine
[(R)-11 and (S)-11]
After chromatography with CH2Cl2–CH3OH (95 : 5) (R)-11 or
(S)-11 was recrystallised (71%) from ethanol, mp 155–157 1C;
Rf 0.3 (CH2Cl2–CH3OH, 95 : 5); dH (CDCl3) 4.34 (1H, dd, J
2.7, 8.8, H2a), 4.45 (1H, dd, J 6.3, 8.8, H2b), 4.97 (1H, dd, J
2.7, 6.3, H1), 5.55 (1H, d, J 8.4, CH2), 5.68 (1H, d, J 8.4, CH2),
6.22 (2H, s, NH2), 7.31–7.54 (8H, m, aromatic H), 7.79–7.81
(2H, m, aromatic H), 7.84 (1H, s, H2), 8.33 (1H, s, H8); dC
(CDCl3) 67.5 (C20), 71.3 (C10), 79.2 (CH2), 119.3 (C5), 127.0,
128.4, 128.5, 128.9, 129.5, 133.2, 136.6 (aromatic C), 140.5
(C8), 150.2 (C4), 153.5 (C2), 155.6 (C6), 165.3 (COPh); (R)-11
had [a]2D2 ꢁ67.1 (c 0.15 in CHCl3); (found: C, 62.55; H, 4.80; N,
17.71; calc. for C21H19N5O3: C, 64.77; H, 4.92; N, 17.98%);
(S)-11 had [a]2D2 þ65.5 (c 0.14 in CHCl3); (found: C, 64.51; H,
4.83; N, 17.69%).
(R)- and (S)-[2-(Methoxymethoxy)-2-phenyl]ethyl benzoate
[(R)-9 and (S)-9]
Phosphorus pentoxide (6.15 g, 43.34 mmol) was added, por-
tionwise with vigorous stirring, to a solution of (R)-8 or (S)-8
(7.0 g, 24.76 mmol) in anhydrous chloroform (100 mL) and
formaldehyde dimethyl acetal (5.46 mL, 61.91 mmol), the
temperature being maintained at 40–45 1C. The mixture was
then stirred at room temperature for 24 h. The supernatant was
diluted with CH2Cl2 (150 mL) and worked up and the crude
product purified by column chromatography (hexane–ethyl
acetate, 97 : 3) to afford (R)-9 or (S)-9 as an oil, 6.6 g (80%),
Rf 0.4 (hexane–ethyl acetate, 9 : 1); dH (CDCl3) 4.46 (1H, dd, J
2.9, 8.6, H2a), 4.53 (1H, dd, J 6.1, 8.6, H2b), 4.64 (1H, q, J 5.1,
7.8, CH2), 5.04 (1H, dd, J 2.9, 6.1, H1), 7.32–7.56 (8H, m,
aromatic H), 8.05 (2H, m, aromatic H); dC (CDCl3) 55.6
(CH3O), 68.2 (C2), 75.8 (C1), 94.4 (CH2), 127.2, 128.5, 128.7,
129.7, 130.1, 133.1, 138.0 (aromatic C), 166.4 (CO); (R)-9 had
[a]2D2 ꢁ100.7 (c 0.71 in CHCl3); (found: C, 71.16; H, 6.70; calc.
for C17H18O4: C, 71.31; H, 6.34%); (S)-9 had [a]2D2 þ103.9
(c 0.80 in CHCl3); (found: C, 71.34; H, 6.74%).
(R)- and (S)-9-[(2-Benzoyloxy-1-phenylethoxy)methyl]guanine
[(R)-12 and (S)-12]
After chromatography with CH2Cl2–CH3OH (90 : 10) (R)-12
or (S)-12 was recrystallised (70%) from dichloromethane, mp
238–240 1C; Rf 0.4 (CH2Cl2–CH3OH, 90 : 10); dH [(CD3)2SO]
4.27 (1H, dd, J 2.7, 8.8, H2a), 4.32 (1H, dd, J 5.9, 8.8, H2b),
5.03 (1H, q, J 2.7, 5.9, H1), 5.24 (1H, d, J 8.4, CH2), 5.48 (1H,
d, J 8.4, CH2), 6.48 (2H, s, NH2), 7.31–7.69 (5H, m, aromatic
H), 7.78 (1H, s, H8), 10.60 (1H, s, H1); dC [(CD3)2SO] 67.1
(C20), 70.7 (C10), 77.5 (CH2), 116.6 (C5), 127.0, 128.4, 128.5,
129.0, 129.3, 133.4, 137.4 (aromatic C), 137.6 (C8), 151.4 (C4),
153.8, (C2), 156.8 (C6), 165.3 (COPh); (R)-12 was too insoluble
for OR determination (found: C, 62.24; H, 4.56; N, 17.14; calc.
for C21H19N5O4: C, 62.22; H, 4.72; N, 17.27%); (S)-12 was too
insoluble for OR determination; (found: C, 62.29; H, 4.63; N,
17.12%).
(R)- and (S)-[2-(Acetoxymethoxy)-2-phenyl]ethyl benzoate
[(R)-10 and (S)-10]
Acetic anhydride (2.64 mL, 27.94 mmol) and boron trifluoride
–diethyl ether (1.05 mL, 8.38 mmol) were added to a solution
of (R)-9 or (S)-9 (4.0 g, 13.97 mmol) in anhydrous diethyl ether
(60 mL) at ꢁ10 1C. This mixture was stirred at 4 1C for 48 h
then poured into ice–water (35 mL), neutralised with saturated
NaHCO3 and extracted twice with diethyl ether (35 mL). The
extract was worked up and the crude product purified by
column chromatography (hexane–ethyl acetate, 95 : 05) to
afford (R)-10 or (S)-10 as an oil, 3.9 g (90%), Rf 0.5 (hexane–
ethyl acetate, 9 : 1); dH (CDCl3) 1.81 (3H, s, CH3), 4.44 (1H,
dd, J 5.9, 8.9, H2a), 4.48 (1H, dd, J 3.2, 8.9, H2b), 5.03 (1H,
dd, J 3.2, 5.9, H1), 5.14 (1H, d, J 4.8, CH2), 5.46 (1H, d, J 4.8,
CH2), 7.25–7.58 (8H, m, aromatic H), 8.04 (2H, m, aromatic
H); dC (CDCl3) 20.8 (CH3), 67.8 (C2), 79.5 (C1), 87.3 (CH2),
127.0, 128.5, 128.7, 128.8, 129.7, 130.0, 133.2, 137.6 (aromatic
C), 166.3 (COPh), 170.5 (COCH3); (R)-10 had [a]2D2 ꢁ63.3 (c
0.57 in CHCl3); (found: C, 69.08; H, 5.75; calc. for C18H18O5:
C, 68.78; H, 5.77%); (S)-10 had [a]2D2 þ64.2 (c 0.30 in CHCl3);
(found: C, 68.90; H, 5.86%).
(R)- and (S)-1-[(2-Benzoyloxy-1-phenylethoxy)methyl]cytosine
[(R)-13 and (S)-13]
After chromatography with CH2Cl2–CH3OH (95 : 5) (R)-13 or
(S)-13 was recrystallised (88%) from CH3OH, mp 183–185 1C;
Rf 0.5 (CH2Cl2–CH3OH, 90 : 10); dH [(CD3)2SO] 4.36 (1H, dd,
J 3.1, 8.8, H2a), 4.40 (1H, dd, J 5.7, 8.8, H2b), 5.06 (1H, dd, J
3.1, 5.7, H1), 5.19 (1H, dd, J 8.0, 12.8, CH2), 5.76 (1H, d, J 5.7,
H5), 7.34–7.90 (13H, m, aromatic H, H6, NH2), 8.39 (1H, s,
NH); dC [(CD3)2SO] 67.2 (C20), 76.1 (C10), 78.0 (CH2), 94.1
(C5), 126.9, 128.3, 128.5, 128.8, 129.2, 129.3, 133.5, 137.7
(aromatic C), 147.0 (C6), 152.0 (C2), 162.7 (C4), 165.4 (COPh);
(R)-13 had [a]2D2 ꢁ50.7 (c 0.90 in CHCl3); (found: C, 65.47; H,
5.61; N, 11.37; calc. for C20H21N3O4: C, 65.56; H, 5.78 N,
11.47%); (S)-13 had [a]D22 þ49.0 (c 0.13 in CHCl3); (found: C,
65.63; H, 5.58; N, 11.37%).
1466
N e w J . C h e m . , 2 0 0 5 , 2 9 , 1 4 6 1 – 1 4 6 8

View Online
(1H, s, H8), 10.60 (1H, s, H1); dC [(CH3)2SO] 65.6 (C20), 71.0
(R)- and (S)-1-[(2-Benzoyloxy-1-phenylethoxy)methyl]thymine
[(R)-14 and (S)-14]
(C10), 81.5 (CH2), 116.3 (C5), 126.8, 127.6, 128.1, 137.5 (aro-
matic C), 139.1, 151.3, 153.8, 156.7 (C8, C4, C2, C6); (R)-18
had [a]2D2 ꢁ100.3 [c 0.10 in (CH3)2SO]; m/z (CI) 302.1247
After chromatography with CH2Cl2–CH3OH (99 : 1) (R)-14 or
(S)-14 was recrystallised (98%) from CH3OH, mp 106–108 1C;
Rf 0.5 (CH2Cl2–CH3OH, 97 : 3); dH (CDCl3) 1.64 (3H, s, CH3),
4.32 (1H, dd, J 2.5, 9.0, H2a), 4.57 (1H, dd, J 6.7, 9.0, H2b),
4.96 (1H, dd, J 2.5, 6.7, H1), 5.04 (1H, d, J 8.2, CH2), 5.27 (1H,
d, J 8.2, CH2), 7.03 (1H, dd, J 0.84, 1.89, H6), 7.33–7.57 (8H,
m, aromatic H), 7.97–7.99 (2 H, m, aromatic H), 9.59 (1H, s,
NH); dC (CDCl3) 12.1 (CH3), 67.5 (C20), 75.1 (C10), 79.3
(CH2), 111.7 (C5), 127.2, 128.6, 128.9, 129.6, 129.8, 133.3,
136.9 (aromatic C), 139.0 (C6), 151.4 (C2), 164.2 (C4), 166.2
(COPh); (R)-14 had [a]2D2 ꢁ66.8 (c 0.14 in CHCl3); (found: C,
66.33; H, 5.33; N, 7.20; calc. for C21H20N2O5: C, 66.21; H,
5.29; N, 7.35%); (S)-14 had [a]2D2 þ65.8 (c 0.14 in CHCl3);
(found: C, 66.29; H, 5.30; N, 7.16%).
(M1 þ H, calc. for C14H16N5O3 302.1253); (S)-18 had [a]D22
100.2 [c 0.60 in (CH3)2SO]; m/z (CI) 302.1249.
þ
(R)- and (S)-1-[(2-Hydroxy-1-phenylethoxy)methyl]cytosine
[(R)-18 and (S)-18]
After chromatography with CH2Cl2–CH3OH (85 : 15) (R)-18
or (S)-18 was recrystallised (94%) from CH2Cl2, mp 101–
103 1C; Rf 0.3 (CH2Cl2–CH3OH, 85 : 15); dH (CD3OD) 3.51
(1H, dd, J 2.8, 9.1, H2a), 3.64 (1H, dd, J 6.1, 9.1, H2b), 4.61
(1H, dd, J 2.8, 6.1, H1), 5.19 (1H, q, J 7.9, 10.8, CH2), 5.73
(1H, d, J 5.5, H5), 7.25–7.33 (5H, m, aromatic H), 7.51 (1H, d,
J 5.5, H6); dC (CD3OD) 68.5 (C20), 79.4 (C10), 84.7 (CH2), 96.9
(C5), 128.9, 129.9, 130.3, 140.7 (aromatic C), 147.6 (C6), 159.4
(C4), 168.6 (C2); (R)-17 had [a]2D2 ꢁ97.2 (c 0.70 in CH3OH);
(found: C, 59.68; H, 6.15; N, 15.91; calc. for C13H17N3O2: C,
59.53; H, 6.53 N, 16.02%); (S)-17 had [a]2D2 þ100.8 (c 0.90 in
CH3OH); (found: C, 59.39; H, 6.39; N, 16.13%).
(R)- and (S)-1-[(2-Benzoyloxy-1-phenylethoxy)methyl]uracil
[(R)-15 and (S)-15]
After chromatography with CH2Cl2–CH3OH (98 : 2) (R)-15 or
(S)-15 was recrystallised (97%) from methanol, mp 106–
108 1C; Rf 0.4 (CH2Cl2–CH3OH, 97 : 3); dH (CDCl3) 4.34
(1H, dd, J 2.5, 8.9, H2a), 4.55 (1H, dd, J 6.5, 8.9, H2b), 4.96
(1H, dd, J 2.5, 6.5, H1), 5.10 (1H, d, J 8.2, CH2), 5.26 (1H, d, J
8.2, CH2), 5.54 (1H, dd, J 0.9, 5.9, H5), 7.24 (1H, d, J 5.9, H6),
7.33–7.58 (8H, m, aromatic H), 7.97–8.00 (2H, m, aromatic H),
9.88 (1H, s, NH); dC (CDCl3) 67.3 (C20), 75.3 (C10), 79.4
(CH2), 103.0 (C5), 127.0, 128.4, 128.7, 128.8, 129.5, 129.6,
133.2, 136.6 (aromatic C), 143.1 (C6), 151.1 (C2), 163.6 (C4),
166.1 (COPh); (R)-15 had [a]2D2 ꢁ69.8 (c 0.14 in CHCl3);
(found: C, 65.66; H, 5.05; N, 7.71; calc. for C20H18N2O5: C,
65.57; H, 4.95; N, 7.65%); (S)-15 had [a]2D2 þ67.8 (c 0.18 in
CHCl3); (found: C, 65.72; H, 4.93; N, 7.85%).
(R)- and (S)-1-[(2-Hydroxy-1-phenylethoxy)methyl]thymine
[(R)-19 and (S)-19]
After chromatography with CH2Cl2–CH3OH (98 : 2) (R)-19 or
(S)-19 was recrystallised (98%) from CH3OH, mp 132–134 1C;
Rf 0.2 (CH2Cl2–CH3OH, 97 : 3); dH (CD3OD) 1.72 (3H, d, J
0.9, CH3), 3.50 (1H, dd, J 2.7, 9.0, H2a), 3.64 (1H, dd, J 6.3,
9.0, H2b), 4.58 (1H, dd, J 2.7, 6.3, H1), 4.84 (1H, s, OH), 5.19
(2H, q, J 8.0, 9.9, CH2), 7.22–7.31 (6H, aromatic H, H6); dC
(CD3OD) 12.9 (CH3), 68.4 (C20), 78.3 (C10), 85.3 (CH2), 112.1
(C5), 128.8, 129.9, 130.2, 140.8 (aromatic C), 142.6 (C6), 153.7
(C2), 167.3 (C4); (R)-19 had [a]2D2 ꢁ79.8 (c 0.18 in CH3OH);
(found: C, 60.73; H, 5.81; N, 9.89; calc. for C14H16N2O4: C,
60.86; H, 5.84; N, 10.14%); (S)-19 had [a]2D2 þ79.0 (c 0.18 in
CH3OH); (found: C, 60.76; H, 5.80; N, 9.82%).
Deprotection of nucleoside analogues 11–15
Ammonia in methanol was added to a stirred solution of the
protected nucleoside in dioxane. The mixture was stirred over-
night at room temperature and then neutralised with acetic
acid. After evaporation under reduced pressure the crude
product was chromatographed using CH2Cl2–CH3OH as
specified below.
(R)- and (S)-1-[(2-Hydroxy-1-phenylethoxy)methyl]uracil
[(R)-20 and (S)-20]
After chromatography with CH2Cl2–CH3OH (98 : 2) (R)-20 or
(S)-20 was recrystallised (96%) from CH3OH, mp 132–134 1C;
Rf 0.2 (CH2Cl2–CH3OH, 95 : 5); dH (CD3OD) 3.51 (1H, dd, J
2.7, 9.0, H2a), 3.64 (1H, dd, J 6.3, 9.0, H2b), 4.60 (1H, dd, J
2.7, 6.3, H1), 4.87 (1H, s, OH), 5.20 (2H, s, CH2), 5.51 (1H, d, J
6.1, H5), 7.22–7.33 (4H, m, aromatic H, H6), 7.49–7.51 (2 H,
aromatic H); dC (CD3OD) 68.4 (C20), 78.4 (C10), 85.3 (CH2),
103.5 (C5), 128.8, 130.0, 130.3, 140.7 (aromatic C), 147.1 (C6),
153.5 (C2), 167.2 (C4); (R)-20 had [a]2D1 ꢁ90.1 (c 0.18 in
(R)- and (S)-9-[(2-Hydroxy-1-phenylethoxy)methyl]adenine
[(R)-16 and (S)-16]
After chromatography with CH2Cl2–CH3OH (90 : 10) (R)-16
or (S)-16 was recrystallised (95%) from ethanol, mp 172–
173 1C; Rf 0.5 (CH2Cl2–CH3OH, 90 : 10); dH (CD3OD) 3.52
(1H, dd, J 3.0, 8.8, H2a), 3.63 (1H, dd, J 6.1, 8.8, H2b), 4.55
(1H, s, OH), 4.69 (1H, dd, J 3.0, 6.1, H1), 5.70 (1H, q, J 8.2,
16.5, CH2), 7.24–7.30 (5H, m, aromatic H), 8.13 (1H, s, H2),
8.23 (1H, s, H8). dC (CD3OD) 68.3 (C20), 74.1 (C10), 84.9
(CH2), 120.6 (C5), 128.6, 129.8, 130.1, 140.5 (aromatic C),
143.5 (C8), 151.7 (C4), 154.8 (C2), 158.0 (C6); (R)-16 had [a]D22
ꢁ87.0 [c 0.10 in (CH3)2SO]; (found C, 58.80; H, 5.42; N, 24.42;
calc. for C14H15N5O2: C, 58.94; H, 5.30; N, 24.55%); (S)-16
had [a]2D2 þ89.1 [c 0.15 in (CD3)2SO]; (found: C, 58.74; H, 5.07;
N, 24.25%).
CH3OH); m/z (CI) 280.1300 (M1
þ
NH4, calc. for
C13H18N3O4 280.1297); (S)-20 had [a]2D2 þ88.5 (c 0.16 in
CH3OH); m/z (CI) 280.1299.
Acknowledgements
The authors are grateful for the laboratories of Professor E. De
Clercq for the screening results.
References
1
2
3
E. De Clercq, Biochim. Biophys. Acta, 2002, 1587, 258–275.
M. R. Keating, Mayo Clin. Proc., 1999, 74, 1266–1283.
J. Balzarini, G. J. Kang, M. Dalal, P. Herdewijn, E. De Clercq, S.
Broder and D. G. Johns, Mol. Pharmacol., 1987, 32, 162–167.
A. P. Lea and D. Faulds, Drugs, 1996, 51, 846–864.
(a) G. Y. Song, V. Paul, H. Choo, J. Morrey, R. W. Sidwell, R. F.
Schinazi and C. K. Chu, J. Med. Chem., 2001, 44, 3985–3993; (b)
see also the review by: L. Agrofoglio, E. Suhas, R. Condom,
S. R. Challand, R. A. Earl and R. Guedj, Tetrahedron, 1994, 50,
10611–10670.
(R)- and (S)-9-[(2-Hydroxy-1-phenylethoxy)methyl]guanine
[(R)-17 and (S)-17]
After chromatography with CH2Cl2–CH3OH (90 : 10) (R)-17
or (S)-17 was recrystallised (92%) from CH3OH, mp 4250 1C;
Rf 0.5 (CH2Cl2–CH3OH, 85 : 15); dH [(CD3)2SO] 3.40 (1H, m,
H2a), 3.48 (1H, m, H2b), 4.58 (1H, dd, J1,2a 3.2, J1,2b 5.5, H1),
4.88 (1H, t, OH), 5.28 (1H, d, J 8.2, CH2), 5.39 (1H, d, J 8.2,
CH2), 6.48 (2H, s, NH2), 7.21–7.32 (5H, m, aromatic H), 7.69
4
5
N e w J . C h e m . , 2 0 0 5 , 2 9 , 1 4 6 1 – 1 4 6 8
1467

View Online
6
7
(a) K. Lee, Y. Choi, G. Gumina, W. Zhou, R. F. Schinazi and C.
K. Chu, J. Med. Chem., 2002, 45, 1313–1320; (b) W. Zhou, G.
Gumina, Y. H. Chong, J. N. Wang, R. F. Schinazi and C. K. Chu,
J. Med. Chem., 2004, 47, 3399–3408.
(a) L. S. Jeong, H. O. Kim, H. R. Moon, J. H. Hong, S. J. Yoo, W.
J. Vhoi, M. W. Chun and C. K. Lee, J. Med. Chem., 2001, 44,
806–813; (b) see also: H. R. Moon, H. O. Kim and L. S. Jeong,
J. Chem. Soc., Perkin Trans. 1, 2002, 1800–1804.
24 A. Martens, E. Koch, B. Konig and H. Zilch, Ger. Offen. DE
¨
3906357, 1990.
25 (a) D. F. Ewing, V. Glac
¸
on, G. Mackenzie, D. Postel and C. Len,
Tetrahedron Lett., 2002, 43, 3503–3505; (b) D. F. Ewing, V.
Glacon, G. Mackenzie, D. Postel and C. Len, Collect. Symp.
Ser., 2002, 261–266.
¸
26 H. C. Kolb, M. S. VanNieuwenhze and K B. Sharpless, Chem.
Rev., 1994, 94, 2483–2547.
8
9
M. Ono, K. Nishimura, H. Tsubouchi, Y. Nagaoka and K.
Tomioka, J. Org. Chem., 2001, 66, 8199–8203 and references
therein.
H.-P. Guan, M. B. Ksebati, E. R. Kern and J. Zemlicka, J. Org.
Chem., 2000, 65, 5177–5184.
27 (a) E. Lipka-Belloli, V. Glac
C. Vaccher and J. P. Bonte, Chromatographia, 2002, 56, 567–572;
(b) E. Lipka-Belloli, V. Glacon, G. Mackenzie, D. Ewing, C. Len,
C. Vaccher and J. P. Bonte, J. Chromatogr., A, 2002, 972, 211–219;
(c) E. Lipka, V. Glacon, G. Mackenzie, D. Ewing, C. Len, D.
¸ on, G. Mackenzie, D. Ewing, C. Len,
¸
¸
10 E. De Clercq, Nucleosides, Nucleotides Nucleic Acids, 2000, 19,
1531–1541.
Postel, M. P. Vaccher, J. P. Bonte and C. Vaccher, Anal. Lett.,
2004, 37, 385–398; (d) E. Lipka, C. Daniel, M. P. Vaccher, V.
Glac¸ on, D. Ewing, G. Mackenzie, C. Len, J. P. Bonte and C.
11 (a) D. Ormrod, L. J. Scott and C. M. Perry, Drugs, 2000, 59, 839–
863; (b) D. Ormrod and K. Goa, Drugs, 2000, 59, 1317–1340.
12 D. S. Pedersen, T. Boesen, A. B. Eldrup, B. Kiaer, C. Madsen,
U. Henriksen and O. Dahl, J. Chem. Soc., Perkin Trans. 1, 2001,
1656–1661.
13 J. F. Du and G. Y. Wang, Nucleosides, Nucleotides Nucleic Acids,
2000, 19, 867–879.
14 K. Hirota, Y. Monguchi, H. Sajiki, C. Yatome, A. Hitaoka and Y.
Kitadi, Nucleosides Nucleotides, 1998, 17, 1333–1345.
15 C. M. Cheng, T. Shimo, K. Somekawa and M. Baba, Hetero-
cycles, 1998, 48, 491–498.
Vaccher, Electrophoresis, 2004, 25, 444–453.
28 Chem3D 5.0 Ultra, 1998, CambridgeSoft Corporation, Cambridge
MA, USA.
29 (a) U. Chiacchio, A. Corsaro, V. Pistara, A. Rescifina, D. Ian-
nazzo, A. Piperno, G. Romeo, R. Romeo and G. Grassi, Eur. J.
Org. Chem, 2002, 1206–1212; (b) S. Raic, M. Mintas, A. Dani-
lovski, M. Vinkovic, M. Pongracic, J. Plavec and D. VikicTopic, J.
Mol. Struct., 1997, 410, 31–33; (c) P. P. Mager, E. De Clercq, H.
Takashima, M. Ubasawa, K. Sekiya, M. Baba and H. Walther,
Eur. J. Med. Chem., 1996, 31, 701–712; (d) J. M. Molina and M.
R. Espinosa, J. Mol. Struct. (THEOCHEM), 1995, 332, 75–83;
(e) A. Schmidt and M. K. Kindermann, J. Org. Chem., 1997, 62,
3910–3918; (f) S. Obika, K. Morio, D. Nanbu and T. Imanishi,
Chem. Commun., 1997, 1643–1644.
16 H. Hakala, P. Ollikka, J. Degerholm and J. Hovinen, Tetrahedron,
2002, 58, 8771–8777.
17 Y. Kitade, M. Nakanishi and C. Yatome, Bioorg. Med. Chem.
Lett., 1999, 2737–2740.
18 H. M. Younger, A. J. Bathgate and P. C. Hayes, Aliment.
Pharmacol Ther., 2004, 20, 1211–1230.
30 The IUPAC name is 4-amino-1-[(1R,3S)-3-hydroxymethyl-1,3-
dihydro-2-benzofuran-1-yl]pyrimidin-2(1H)-one.
19 D. F. Ewing, N.-E. Fahmi, C. Len, G. Mackenzie, G. Ronco, P.
Villa and G. Shaw, Nucleosides Nucleotides, 1999, 18, 2613–2630.
20 D. F. Ewing, N.-E. Fahmi, C. Len, G. Mackenzie and A. Pranzo,
J. Chem. Soc., Perkin Trans. 1, 2000, 3561–3565.
21 D. F. Ewing, C. Len, G. Mackenzie, G. Ronco and P. Villa,
Tetrahedron: Asymmetry, 2000, 11, 4995–5002.
22 D. Egron, C. Perigaud, G. Gosselin, A.-M. Aubertin, A. Faraj, M.
´
´
Selouane, D. Postel and C. Len, Bioorg. Med. Chem. Lett., 2003,
13, 4473–4475.
31 The IUPAC name is 4-amino-1-[(2R,5S)-5-hydroxymethyl-2,5-
dihydrofuran-2-yl]pyrimidin-2(1H)-one; alternative names include
1-(2,3-dideoxy-b-D-glyceropent-2-enofuranosyl)uracil and 20,30-
didehydro-20,30-dideoxyuridine.
32 In isopropyl benzene the minimum energy conformation has the
a-CH bond in the ring plane: P. R. Richardson, M. A. Chapman,
D. C. Wilson, S. P. Bates and A. C. Jones, Phys. Chem. Chem.
Phys., 2002, 4, 4910–4915.
33 (a) H. Gao and A. K. Mitra, Synthesis, 2000, 329–351; (b) C. K.
Chu and S. J. Cutler, J. Heterocycl. Chem., 1986, 23, 289–319.
34 R. Kumar, M. Nath and D. L. J. Tyrrell, J. Med. Chem., 2002, 45,
2032–2040.
23 B.-C. Pan, Z.-H. Chen, G. Piras, G. E. Dutschman, E. C. Rowe,
Y.-C. Cheng and S. H. Chu, J. Heterocycl. Chem., 1994, 31,
177–185.
1468
N e w J . C h e m . , 2 0 0 5 , 2 9 , 1 4 6 1 – 1 4 6 8