Nucleosides and nucleotides. 184. Synthesis and conformational investigation of anti-fixed 3-deaza-3-halopurine ribonucleosides

Article abstract of DOI:10.1021/jo990638x

A versatile synthetic route to 3-deaza-3-haloinosines 6-8 and 46, - adenosines 15-17 and 64, and -guanosines 25, 26, and 52 is described. 3- Deaza-3-chloro-, -bromo-, and -iodopurine ribonucleosides can be synthesized by treating the 3-deazapurine ribonucleosides with N-halosuccinimides. For the synthesis of 3-deaza-3-fluoropurine ribonucleosides, 5-formylimidazole-4- carboxamide ribosides 34 and 35 prepared from 5-iodoimidazole-4-carboxamide derivatives 29 and 31 were used as the key intermediates. The reaction of 34 and 35 with lithium (trimethylsilyl)acetylide and sodium cyanide, respectively, followed by appropriate manipulations gave 3-deaza-3- fluoroinosine derivative 46 and 3-deaza-3-fluoroguanosine derivative 52. 3- Deaza-3-fluoroadenosine (64) was also synthesized by selective reductive deamination of 4,6-diamino-7-fluoroimidazo[4,5-c]pyridine derivative 51, followed by deprotection. A conformational analysis using ¹H NMR and NOE experiments showed that the introduction of halogens (chloro, bromo, and iodo) into the 3-position of 3-deazapurine ribonucleosides forced fixation of the glycosyl torsion angle in the anti region, but did not abnormally influence sugar puckering. On the other hand, 3-deaza-3-fluoropurine ribonucleosides should rotate freely around the glycosyl bond.

Full text of DOI:10.1021/jo990638x

7158
J . Org. Chem. 1999, 64, 7158-7172
Nu cleosid es a n d Nu cleotid es. 184. Syn th esis a n d Con for m a tion a l
In vestiga tion of An ti-F ixed 3-Dea za -3-h a lop u r in e
Ribon u cleosid es1,2
Noriaki Minakawa, Naoshi Kojima, and Akira Matsuda*
Graduate School of Pharmaceutical Sciences, Hokkaido University, Kita-12, Nishi-6,
Kita-ku, Sapporo 060-0812, J apan
Received April 16, 1999
A versatile synthetic route to 3-deaza-3-haloinosines 6-8 and 46, -adenosines 15-17 and 64, and
-guanosines 25, 26, and 52 is described. 3-Deaza-3-chloro-, -bromo-, and -iodopurine ribonucleosides
can be synthesized by treating the 3-deazapurine ribonucleosides with N-halosuccinimides. For
the synthesis of 3-deaza-3-fluoropurine ribonucleosides, 5-formylimidazole-4-carboxamide ribosides
34 and 35 prepared from 5-iodoimidazole-4-carboxamide derivatives 29 and 31 were used as the
key intermediates. The reaction of 34 and 35 with lithium (trimethylsilyl)acetylide and sodium
cyanide, respectively, followed by appropriate manipulations gave 3-deaza-3-fluoroinosine derivative
46 and 3-deaza-3-fluoroguanosine derivative 52. 3-Deaza-3-fluoroadenosine (64) was also synthesized
by selective reductive deamination of 4,6-diamino-7-fluoroimidazo[4,5-c]pyridine derivative 51,
1
followed by deprotection. A conformational analysis using H NMR and NOE experiments showed
that the introduction of halogens (chloro, bromo, and iodo) into the 3-position of 3-deazapurine
ribonucleosides forced fixation of the glycosyl torsion angle in the anti region, but did not abnormally
influence sugar puckering. On the other hand, 3-deaza-3-fluoropurine ribonucleosides should rotate
freely around the glycosyl bond.
In tr od u ction
Therefore, it would be interesting to devise a new way
to lock the conformation of nucleosides where the nucleo-
base is still held in a particular region, but where the
sugar ring is sufficiently flexible to adopt whatever
puckering the enzyme normally imposes on its sub-
strates. From this point of view, nucleosides that have a
bulky substituent on a nucleobase or a sugar portion may
be useful for fixing the nucleoside conformation. For
example, 8-bromoadenosine and 8-bromoguanosine have
been found to exist in the syn conformation in the solid
state as well as in solution due to steric repulsion
between the 8-bromo group and the sugar portion, and
hence, 8-bromoguanosine 5′-triphosphate is not a sub-
strate of Qâ replicase or Escherichia coli transcriptase,
which prefer an anti rotational isomer either during or
after the polymerization reaction.6 On the other hand,
8-bromoadenosine 5′-diphosphoribose bound to horse
liver alcohol dehydrogenase was found to assume an anti
conformation in its cocrystal, in contrast to the syn
conformation found in 8-bromoadenosine itself.7 As can
be seen from this result, the syn/anti conformation
around the glycosyl bond may be forced to change when
the nucleo(s/t)ide binds to certain enzymes or receptors,
and therefore, the position of an introduced substituent
would be important for such fixation. Iimori et al.8 have
attempted to restrict the conformation of sangivamycin
The conformation of a nucleoside is defined by several
parameters, i.e., sugar pucker modes, syn/anti conforma-
tion around the glycosyl bond, and orientation about the
C4′-C5′ bond.3 Concerning nucleoside-enzyme interac-
tions when acting as substrates or inhibitors, the syn/
anti conformation around the glycosyl bond is one of the
most important conformational aspects. In this regard,
several nucleosides with restricted conformation about
glycosyl torsion angles have been synthesized and evalu-
ated to understand the relationships between glycosyl
torsion angles and nucleoside-enzyme interactions. The
most straightforward fixation of glycosyl torsion angles
is found in cyclonucleosides that possess an extra cova-
lent linkage between the sugar and the nucleobase. These
cyclonucleosides cover a wide range of glycosyl torsion
angles, and the correlations between the direction of the
sign and magnitude of their CD spectra and the glycosyl
torsion angles have been well studied.4 However, such
fixation frequently results in a change in the sugar
conformation that is not normally seen in unconstrained
nucleosides. Like the syn/anti conformation around the
glycosyl bond, it has been revealed that the sugar pucker
is also an important conformational aspect in terms of
its correlation with nucleoside-enzyme interactions.5
* To whom correspondence and reprint requests should be ad-
dressed.Phone: +81-11-706-3228. Fax: +81-11-706-4980. E-mail:
matuda@pharm.hokudai.ac.jp.
(1) Part 183: Sugimoto, I.; Shuto, S.; Mori, S.; Shigeta, S.; Matsuda,
A. Bioorg. Med. Chem. Lett. 1999, 9, 385-388.
(2) Purine numbering has been used for all nucleosides throughout
the text except for the Experimental Section.
(3) Saenger, W. Principles of Nucleic Acid Structure; Springer-
Verlag: New York, 1984; pp 51-81.
(4) Yoshimura, Y.; Otter, B. A.; Ueda, T.; Matsuda, A. Chem. Pharm.
Bull. 1992, 40, 1761-1769 and references therein.
(5) For examples: (a) Parkanyi, L.; Kalman, A.; Czugler, M.; Walker,
R. T. Nucleic Acids Res. 1987, 15, 4111-4121. (b) Rodriguez, J . B.;
Marquez, V. E.; Mitsuya, H.; Barchi, J . J . J . Med. Chem. 1994, 37,
3389-3399. (c) Marquez, V. E.; Siddiqui, M. A.; Ezzitouni, A.; Russ,
P.; Wang, J .; Wagner, R. W.; Matteucci, M. D. J . Med. Chem. 1996,
39, 3739-3747.
(6) Tavale, S. S.; Sobell, H. M. J . Mol. Biol. 1970, 48, 109-123.
(7) Abdallah, M. A.; Biellmann, J . F.; Nordstrom, B.; Branden, C. I.
Eur. J . Biochem. 1975, 50, 475-481.
(8) Iimori, T.; Murai, Y.; Ohuchi, S.; Kodama, Y.; Ohtsuka, Y.; Oishi,
T. Tetrahedron Lett. 1991, 32, 7273-7276.
10.1021/jo990638x CCC: $18.00 © 1999 American Chemical Society
Published on Web 08/31/1999

3-Deaza-3-halopurine Ribonucleosides
J . Org. Chem., Vol. 64, No. 19, 1999 7159
Sch em e 1a
by introducing a methyl group on the sugar, but the steric
repulsion between the sugar and the nucleobase moiety
was insufficient. In addition, introduction of the methyl
group changed the conformation of the sugar moiety. On
the basis of these results, we envisioned the introduction
of a bulky substituent at the 3-position of 3-deazapurine
ribonucleosides to fix the conformation around the gly-
cosyl linkage in the anti region without greatly affecting
their sugar puckering. In our first approach, we previ-
ously reported the synthesis of 3-alkyl-3-deazainosines
that were successfully fixed in the anti conformation, and
the predominant sugar puckering was assigned to be C2′-
endo in solution, as in inosine and 3-deazainosine (3-
zI).9,10 However, the physicochemical properties such as
the pKa value of the nucleobase moiety would be ap-
preciably altered relative to those of 3-deazainosine itself
and its alkylated derivatives. In our previous paper,11 we
reported the synthesis of 3-deaza-3-haloinosines and that
such conformational fixation not only resulted in the
usual sugar puckering but also only minimally changed
the pKa values. In this paper, we describe in detail the
synthesis of 3-deaza-3-haloadenosines and -guanosines
in addition to -inosines and conformational studies of
a
Reagents: (a) (1) aqueous Me2NH in EtOH 80 °C, then
aqueous AcOH in EtOH, (2) Ac2O, pyridine; (b) N-halosuccinimide,
CH2Cl2 or DMF; (c) NH3/MeOH.
1
these compounds using H NMR and NOE experiments.
toward N-halosuccinimides. Although bromination and
iodination of 3-deazaadenosine derivative 913 achieved as
described above to give 11 and 12 in yields of 74 and 71%,
respectively, treatment of 9 with NCS in CH2Cl2 at -15
°C resulted in 10 in 16% yield.14 Thus, the exocyclic
amino group was protected by an acetyl group prior to
the chlorination. Treatment of 9 with an excess amount
of Ac2O in the presence of Et3N and (dimethylamino)-
pyridine (DMAP) in CH3CN, which gave a mixture of 13
and diacetylated derivative, followed by methanolic am-
monia gave 13 in 94% yield. When 13 was subjected to
chlorination, the desired product 14 was obtained in 75%
yield. Deprotection of the acetyl group was achieved by
treating 14 with methanolic ammonia in a steel container
at 50 °C (Scheme 2). For the synthesis of 3-deaza-3-
haloguanosines, all halogenations were carried out with
N-acetylated derivative 19 to give 20, 21, and 22 in good
yields. Removal of the acetyl groups can be done by
treatment with methanolic ammonia at room tempera-
ture to give 23 and 24; however, the 3-deaza-3-iodogua-
nosine derivative could not be obtained due to the
undesirable reduction of the iodo substituent under the
deprotection of the acetyl group (Scheme 3).15 All of the
halogenated 3-deazapurine ribonucleosides were con-
verted to free nucleosides 6-8, 15-17, 25, and 26 by
treatment with methanolic ammonia or tetrabutylam-
monium fluoride (TBAF).
Resu lts a n d Discu ssion
Ch lor in a tion , Br om in a tion , a n d Iod in a tion of
3-Dea za p u r in e Ribon u cleosid es. 3-Deazapurine ribo-
nucleosides have an imidazo[4,5-c]pyridine or imidazo-
[4,5-c]pyridinone skeleton at the nucleobase moiety that
could be expected to react with electrophilic reagents at
the 3-position. Chloro, bromo, and iodo substituents can
be easily introduced into the 3-position of 3-deazapurine
ribonucleosides by treatment with N-halosuccinimides.
When 3-deazainosine derivative 2, which was prepared
from 5-ethynyl-1-â-D-ribofuranosylimidazole-4-carbox-
amide (1)12 in 89% yield, was treated with N-chlorosuc-
cinimide (NCS) or N-bromosuccinimide (NBS) in CH2Cl2
at room temperature, 3-chloro and 3-bromo derivatives
3 and 4 were obtained in yields of 71 and 78%, respec-
tively. 3-Iodo derivative 5 was obtained by treatment of
2 with N-iodosuccinimide (NIS) in DMF in 75% yield
(Scheme 1). The position of halogenation was established
by comparing the 1H NMR spectra of 2 and 3 and by NOE
experiments. Thus, the vinylic protons of 2 at 7.36 (broad
doublet) and 6.65 ppm (doublet) that correspond to H-2
and H-3, respectively, disappeared after chlorination, and
a broad signal at 7.46 ppm due to H-2 was observed in
3. Furthermore, an NOE was observed at the proton
(12%) upon irradiation of the N1H proton. On the basis
of these results, it is likely that halogenation took place
at the expected 3-position.
Syn th esis of 3-Dea za -3-flu or op u r in e Ribon u cleo-
sid es. Since the need for organofluorine compounds has
increased not only for biological and pharmacological
application but also as organic materials, many fluori-
nating agents have been developed.16 We first examined
the direct fluorination of 2 with electrophilic fluorinating
reagents; however, all attempts with N-fluoropyridinium
In contrast to 3-deazainosine 2, 3-deazaadenosine and
3-deazaguanosine derivatives were much more reactive
(9) (a) Aoyagi, M.; Minakawa, N.; Matsuda, A. Tetrahedron Lett.
1993, 34, 103-106. (b) Yamagata, Y.; Kato, M.; Fujii, S.; Aoyagi, M.;
Minakawa, N.; Matsuda, A. Nucleosides Nucleotides 1994, 13, 1327-
1335.
(10) Coincident with our report, Acevedo et al. reported the synthesis
of 2′-deoxy-3-alkyl(aryl)-3-deazaguanosines. They indicated that
a
(13) Minakawa, N.; Matsuda, A. Tetrahedron 1993, 49, 557-570.
(14) Unlike the reaction with 2, the reaction mixture immediately
turned dark brown and the reaction gave a complex mixture. Although
purification and identification were unsuccessful, the byproducts
appeared to be polychlorinated 3-deazaadenosine derivatives.
(15) Although direct iodination of 18 with NIS in CH2Cl2 at -20 °C
gave the desired 3-deaza-3-iodoguanosine derivative, the product was
unstable and gave a complex mixture including 18 under usual storage
conditions.
bulky substituent such as a 1-naphthylethyl group forces the sugar
puckering into an unusual C3′-endo conformation, although the
nucleoside prefers the anti conformation; see: Acevedo, O. L.; Andrews,
R. S.; Cook, P. D. Nucleosides Nucleotides 1993, 12, 403-416.
(11) Minakawa, N.; Kojima, N.; Matsuda, A. Heterocycles 1996, 42,
149-154.
(12) Minakawa, N.; Takeda, T.; Sasaki, T.; Matsuda, A.; Ueda, T.
J . Med. Chem. 1991, 34, 778-786.

7160 J . Org. Chem., Vol. 64, No. 19, 1999
Minakawa et al.
Sch em e 2a
Sch em e 4a
a
Reagents: (a) TBDMSCl, imidazole, DMF; (b) isoamyl nitrite,
CH2I2, 100 °C; (c) (1) NH3/MeOH, (2) TBDMSCl, imidazole, DMF;
(d) methyl acrylate, Et3N, (PhCN)2PdCl2, CH3CN, 100 °C (e) (1)
O3, MeOH, -78 °C, then Me2S, or (2) OsO4, N-methylmorpholine
N-oxide, acetone-H2O, then NaIO4.
a
Reagents: (a) N-halosuccinimide, CH2Cl2 or DMF; (b) Ac2O,
Et3N, DMAP, CH3CN, then NH3/MeOH; (c) N-chlorosuccinimide,
CH2Cl2; (d) NH3/MeOH, 50 °C; (e) TBAF, THF.
Sch em e 3a
After silylation of a sugar hydroxyl group at the 5′-
position of 27, 28 was diazotized with isoamyl nitrite in
the presence of diiodomethane at 100 °C to give 29 in
69% yield. Although palladium-catalyzed formylation of
29 would be conceivably possible,18 treatment of 29 with
tributyltin hydride and carbon monoxide in the presence
of Pd(PPh3)4 resulted in recovery of the starting material
along with only a trace amount of 5-formyl derivative 34.
Therefore, 34 was synthesized by ozonolysis of 5(E)-(2-
carbomethoxyvinyl) derivative 32.11 Treatment of 29 with
methyl acrylate and Et3N in the presence of bis(benzoni-
trile)palladium dichloride in CH3CN gave 32 in 80%
yield. When 32 was ozonized in MeOH, followed by
reduction of the resulting ozonide with dimethyl sulfide,
5-formyl derivative 34 was obtained in 98% yield. In a
similar manner, 31 obtained from 3012 in two steps was
converted into 5(E)-(2-carbomethoxyvinyl) derivative 33.
Osmylation of 33, followed by in situ periodate oxidation,
gave a good result, compared with the ozonolysis of 33,
to give the desired 5-formyl derivative 35 in 85% yield.
Initially, we examined the synthesis of 3-deaza-3-
fluoroinosine (46) (Scheme 5). The addition reaction of
34 with vinylmagnesium bromide in THF gave 5-(1-
hydroxy-2-propenyl) derivative 36 in 93% yield. If the
resulting hydroxyl group can be replaced by a fluorine,
followed by conversion of the vinyl group into a formyl
group, the desired ring closure between the 5-carbox-
amide and the formyl groups to give a 3-deaza-3-fluoro-
hypoxanthine skeleton would be possible. When 36 was
treated with diethylaminosulfur trifluoride (DAST) in
CH2Cl2 at -15 °C, the desired 5-(1-fluoro-2-propenyl)
derivative 39 was obtained as a mixture of diastereomers
(about 3:2) in 37% yield along with 5(E)-(3-fluoro-1-
propenyl) derivative 41 in 31% yield. Formation of 41
may occur via an SN2′ reaction and/or generation of an
allylic cation during this reaction. The separated 39 was
then treated with potassium permanganate in aqueous
a
Reagents: (a) Ac2O, pyridine, then NH3/MeOH; (b) N-halo-
succinimide, CH2Cl2, or DMF; (c) NH3/MeOH; (d) TBAF, THF.
salts17 or F2 gas gave poor results with regard to recovery
or decomposition of the starting material. Therefore, we
planned to synthesize the desired compounds by prepar-
ing substrates with fluorinated substituents at the
5-position of imidazole nucleosides, followed by ring
closure. 5-Formylimidazole-4-carboxamide derivatives 34
and 35 were thought to be appropriate substrates for the
synthesis of the target compounds. The key intermediates
were synthesized as shown in Scheme 4.
(16) For review articles, see: (a) Purrington, S. T.; Kagan, B. S.;
Patrick, T. B. Chem. Rev. 1986, 86, 997-1018. (b) Wilkinson, J . A.
Chem. Rev. 1992, 92, 505-515. (c) Rozen, S. Chem. Rev. 1996, 96,
1717-1736. (d) Lal, G. S.; Pez, G. P.; Syvret, R. G. Chem. Rev. 1996,
96, 1737-1755.
(17) Commercially available N-fluoro-3,5-dichloropyridinium triflate,
N-fluoropyridinium tetrafluoroborate, and N-fluoro-2,6-dichloropyri-
dinium tetrafluoroborate were examined for direct fluorination.
(18) Baillargeon, V. P.; Stille, J . K. J . Am. Chem. Soc. 1986, 108,
452-461.

3-Deaza-3-halopurine Ribonucleosides
J . Org. Chem., Vol. 64, No. 19, 1999 7161
Sch em e 5a
a
Reagents: (a) vinylmagnesium bromide (or 1-bromo-2-methylpropene, t-BuLi, or trimethylsilylacetylene, n-BuLi), THF; (b) DAST,
CH2Cl2; (c) (1) KMnO4, 18-crown-6, THF, then NaIO4, or (2) H2, Lindlar, EtOH, then conditions (1); (d) aqueous NaHCO3, CHCl3, 80 °C;
(e) Ac2O, DMAP, pyridine, rt to 100 °C; (f) aqueous TFA.
THF in the presence of 18-crown-6, followed by sodium
periodate to give two new spots on TLC analysis. The
relative proportion of the more polar spot increased upon
lengthening the reaction time, and subsequent treatment
of the reaction mixture with aqueous sodium bicarbonate
gave the more polar spot predominantly. The structure
of the more polar spot was confirmed to be 44, but not
45, on the basis of its MS spectrum, in which a molecular
ion peak was detected at m/z 458 as MH+, and its UV
spectrum (λmax, 246 nm), together with the 1H NMR
spectrum, in which two dissociable protons were observed
at 8.36, 8.26 (in a ratio of 2:1 due to the NH), and 6.87
ppm (as a multiplet due to the OH). During the oxidation
reaction of 39, the less polar spot corresponding to 43
was detected by TLC analysis upon staining with 3,5-
dinitrophenylhydrazine. The intramolecular addition of
the carbamoyl group at the 4-position should be fast.
Surprisingly, the imidazo[4,5-c]-6,7-dihydropyridin-4(5H)-
one derivative 44 was found to be quite stable; treatment
of 44 with aqueous AcOH under heating did not give the
dehydrated product 45. Therefore, the hydroxyl group of
44 was acetylated using Ac2O and DMAP in pyridine, and
subsequent heating gave the desired 45 in 92% yield.
Deblocking of the protecting groups at the sugar moiety
in 45 was achieved with 75% trifluoroacetic acid (TFA)
to furnish 3-deaza-3-fluoroinosine (46) in quantitative
yield.
further stabilized by the terminal dimethyl group of 37
to give 42 predominantly. To improve the yield of 46, we
next tried using a propargyl alcohol system. Reaction of
34 with lithium (trimethylsilyl)acetylide in THF at -78
°C, followed by treatment with methanolic ammonia for
3 h at room temperature, gave a mixture of diastereomers
of 38, which was separable using a silica gel column.
Treatment of the mixture of diastereomers of 38 with
DAST in CH2Cl2 at -15 °C exclusively gave the desired
5-(1-fluoropropynyl) derivative 40 as a mixture of dia-
stereomers in 89% yield. Since the reaction of one of the
diastereomers 38 under the same conditions again gave
the mixture of diastereomers 40 (data not shown), the
fluorination reaction should proceed via a cationic inter-
mediate. A partial catalytic reduction of the acetylenic
moiety was performed using Lindlar catalyst to give 39
in 70% yield. After the hydrogenolysis, 44 was obtained
in 70% yield from 40 without isolation of 39.
The synthesis of 3-deaza-3-fluoroguanosine (52) can be
achieved by a strategy similar to that used for the
synthesis of 46. As shown in Scheme 6, if the hydroxyl
group of the cyanohydrin 47 can be converted into its
corresponding fluoride, ring closure from the carbamoyl
group at the 4-position would proceed smoothly. Since
the glycosyl linkage of 3-deazaguanosine derivatives is
expected to be unstable relative to that of 3-deazainosine
derivatives under acidic conditions,19 we used tris-TB-
DMS derivative 35 instead of 34 as a starting material.
Compound 35 was vigorously stirred with NaCN in a
mixture of EtOAc and aqueous 1 M sodium bicarbonate
solution, and the resulting cyanohydrin 47, without
purification due to the possibility of decomposition during
isolation, was then treated with DAST in CH2Cl2 at -15
°C to give the desired 5-cyanofluoromethyl derivative 48
as a mixture of diastereomers in 73% yield. The ring
closure of 48 to 3-deaza-3-fluoroguanine derivative 50
was achieved under heating at 60 °C in a mixture of
EtOH and aqueous 5% K2CO3 solution in 86% yield.
As expected, 34 was an important intermediate for the
synthesis of 46. However, fluorination of 36 with DAST
did not give a satisfying yield of 39. If the byproduct 41
was formed via an SN2′ mechanism, bulky substituents
at the terminus of the olefin would prevent undesired
fluorination. To examine this possibility, a carbanion
derived from 1-bromo-2-methylpropene was added to the
formyl group of 34. The resulting 37 was then treated
with DAST under the same conditions, but exclusively
gave the undesired 5(E)-(3-fluoro-3-methyl-1-butenyl)
derivative 42 in 80% yield (data not shown). On the basis
of these experiments, the fluorination of the allylic alcohol
by DAST should occur via the allylic cation, which was
(19) Minakawa, N.; Kojima, N.; Matsuda, A. Unpublished work.

7162 J . Org. Chem., Vol. 64, No. 19, 1999
Minakawa et al.
Sch em e 6a
Sch em e 7a
a
Reagents: (a) (1) Ac2O, DMAP, pyridine, then NH3/MeOH, or
(2) AcCl, pyridine, then NH3/MeOH; (b) isoamyl nitrite, THF, 60
°C; (c) NaOMe, MeOH; (d) TBAF, THF.
deamination. Acetylation of 51 with Ac2O in the presence
of DMAP in pyridine at room temperature gave a mixture
of mono- and diacetylated derivatives of 51 that was
successively treated with methanolic ammonia for a short
period of time at room temperature to give the monoacety-
lated derivative 54 in 81% yield from 51. To identify
which amino group was acetylated, the reductive deami-
nation of 54 was performed with isoamyl nitrite in THF22
to give 57 in 59% yield, which was then deblocked with
NaOMe in MeOH, followed by TBAF to furnish 59 in 60%
yield, but not the desired 3-deaza-3-fluoroadenosine (64)
(Scheme 7). The UV spectrum of 59 in H2O showed two
absorption maxima at 260 nm (ꢀ ) 6500) and 292 nm (ꢀ
) 3200), which were significantly different from those of
other 3-deaza-3-haloadenosines (see the Experimental
Section). Moreover, the 1H NMR spectrum of 59 sup-
ported its structure; i.e., the coupling constant between
H-6 and the fluorine was 1.5 Hz and was rather small
a
Reagents: (a) NaCN, aqueous NaHCO3-AcOEt; (b) DAST,
CH2Cl2; (c) POCl3, Et3N, CH2Cl2; (d) (1) aqueous Na2CO3-EtOH,
60 °C, or (2) NH3/EtOH, 120 °C; (e) (1) TBAF, THF, or (2) NH4F,
MeOH.
Deprotection of 50 with TBAF in THF furnished 3-deaza-
3-fluoroguanosine (52) in 70% yield as crystals. On the
other hand, the carbamoyl group in 48 was dehydrated
to a cyano group using POCl3 to give 49, which was
heated in methanolic ammonia at 120 °C in a steel
container to afford diamino derivative 51 in 79% yield.
Deprotection of 51 using TBAF proceeded smoothly, but
separation of the resulting 53 and the reagent used was
difficult. Therefore, 51 was deprotected by ammonium
fluoride in MeOH to give 53 in 74% yield.
For the synthesis of 3-deaza-3-fluoroadenosine deriva-
tives, conversion of the corresponding deazainosines as
a starting material is the usual approach. However,
deazapurines are more electron-rich than the corre-
sponding purines. Therefore, 6-chloro-3-deazapurine de-
rivatives were transformed to their 3-deazaadenosine
derivatives via treatment with hydrazine instead of
ammonia.20 Although the fluorine group at the 3-position
has an electron-withdrawing character, treatment of
6-chloro-3-deaza-3-fluoropurine derivative21 with am-
monia did not give the desired 3-deazaadenine derivative.
Even treatment with hydrazine gave a complex mixture
(data not shown). Therefore, we decided to convert the
2,6-diamino-3-deaza-3-fluoropurine derivative 51 into the
3-deaza-3-fluoroadenine derivative via selective reductive
relative to that of 3-deaza-3-fluoroinosine (46) (J H-2, F
)
6.1 Hz). On the other hand, treatment of 51 with AcCl
in pyridine gave the diacetamide derivative 55 in 92%
yield. Upon treatment of 55 with methanolic ammonia
at room temperature, none of the acetyl group was
removed. However, when 55 was heated at 80 °C with
methanolic ammonia in a steel container, 54 was ob-
tained in 90% yield. Thus, the structure of the diacety-
lated compound obtained when 51 was treated with acetic
anhydride should be 56. From these results, it was
revealed that selective acylation of the amino group at
2-position could be done by treatment of 51 with an
acylating reagent, followed by methanolic ammonia.
Although the acetyl group at the 6-position is less labile
than that of the 2-position, the acetyl group is stable
against methanolic ammonia at room temperature. Thus,
if an acyl-protecting group, which is deprotected by
methanolic ammonia at room temperature, was chosen
as that of the 2-position, selective protection of the amino
group at the 6-position could be done. We first selected a
trifluoroacetyl group to be introduced to the amino group
at the 2-position. Treatment of 51 with trifluoroacetic
(20) Rousseau, R. J .; Townsend, L. B.; Robins, R. K. Biochemistry
1966, 5, 756-760.
(21) Triacetylated 6-chloro-3-deaza-3-fluoropurine derivative was
obtained by treatment of triacetylated 3-deaza-3-fluoroinosine with
POCl3 and DMF in 94% yield.
(22) Nair, V.; Chamberlain, S. D. Synthesis 1984, 401-403.

3-Deaza-3-halopurine Ribonucleosides
J . Org. Chem., Vol. 64, No. 19, 1999 7163
Sch em e 8a
Ta ble 1. p Ka Va lu es of P u r in e a n d 3-Dea za p u r in e
An a logu es
a
b
compd
inosine
pKa1
pKa2
9.1
13.1
11.2
11.1
11.3
11.7
13.7
3.6
3-zI
46 (X ) F)
6 (X ) Cl)
7 (X ) Br)
8 (X ) I)
3-Me-3zI
adenosine
3-zA
7.0
64 (X ) F)
15 (X ) Cl)
16 (X ) Br)
17 (X ) I)
guanosine
3-zG
52 (X ) F)
25 (X ) Cl)
26 (X ) Br)
5.2
5.1
5.2
5.2
1.9
2.7
ND
2.0
2.1
9.2
12.3
10.4
10.5
10.5
a
Values represent pKa values of the N1 position for inosine and
adenosine derivatives and of the N7 position for guanosine deriva-
a
Reagents: (a) trifluoroacetic anhydride, Et3N, CH2Cl2, then
tives. Values represent pKa values of the N1 position. c Not
b
NH3/MeOH; (b) AcCl, pyridine; (c) NH3/MeOH; (d) isoamyl nitrite,
THF, 60 °C; (e) NH3/MeOH, 120 °C, then TBAF, THF.
determined.
N1H. Although fluorine is a very electronegative atom,
the fluorine atom also acts as an electron-donating
substituent due to participation of its antibonding 2p
orbital when it binds to unsaturated carbon.25 This would
explain why the pKa value of 46 was not proportional to
the electronegativity of its halogen atoms, due to both
its -I and +M effects. In contrast, the pKa value of
3-deaza-3-methylinosine (3-Me-3-zI) was 13.7, which is
much larger than that of 3-zI. The same tendency was
observed for 3-deazaadenosine (3-zA) and 3-deazagua-
nosine (3-zG) derivatives, and the pKa values of 3-deaza-
3-halopurine ribonucleosides were close to those of natu-
ral purine ribonucleosides.
anhydride in the presence of Et3N in CH2Cl2 followed by
treatment with methanolic ammonia afforded N2-trifluo-
roacetamide 60 in 91% yield. Subsequent acetylation of
60 with AcCl in pyridine gave N6-acetamido-N2-trifluo-
roacetamide derivative 61, which was treated again with
methanolic ammonia for 24 h at room temperature
without purification to give the desired 2-amino-N6-
acetamide derivative 62 in 80% yield from 51. The
reductive deamination of 62 gave 63 in 55% yield, which
was deprotected first with methanolic ammonia at 80 °C
and then with TBAF to give 3-deaza-3-fluoroadenosine
(64) in 57% yield from 63 (Scheme 8). The UV spectrum
of 64 showed an absorption maximum at 267 nm (ꢀ )
8000), which is close to those of 6-8, but different from
that of 59.
Acid -Ba se P r op er t ies of 3-Dea za -3-h a lop u r in e
Ribon u cleosid es. The pKa values of 3-deazapurine
ribonucleosides are appreciably different than those of
natural purine nucleosides,23 and thus, this should be
affected by nucleoside-enzyme interactions. As men-
tioned above, we envisioned introducing an electron-
withdrawing halogen atom as a bulky substituent at the
3-position not only to fix the conformation around the
glycosyl bond but also to minimize differences in the pKa
values between 3-deazapurine derivatives and natural
purine ribonucleosides. We first measured the pKa values
of the halogenated nucleosides, relative to those of
3-deaza and natural purine ribonucleosides.24 The results
are listed in Table 1. The pKa value of N1H of inosine
was 9.1, while that of 3-deazainosine (3-zI) was calculated
to be 13.1. The effects of the nitrogen atom at the
3-position of the purine ring influenced these differences.
Introduction of halogen atoms at the 3-position of 3-zI
increased the acidity of the N1H by about 2 pKa units.
Excluding fluorine, a higher electronegativity of the
halogen atom was associated with a higher acidity of the
Con for m a t ion a l St u d ies Usin g NMR a n d NOE
Exp er im en ts. Analyses with 1H NMR, 13C NMR, and
NOE spectroscopy have been used to study the dynamic
equilibrium of syn/anti around the glycosyl bond and
sugar pucker modes.26,27 Conformational studies of 3-deaza-
3-halopurine ribonucleosides were performed using NMR
and NOE experiments. In these investigations, an un-
1
expected phenomenon was observed in H NMR spectra.
The 1H NMR chemical shifts of the synthetic nucleosides
are listed together with those of natural inosine, adenos-
ine, and guanosine in Table 2. As can be seen, large
(25) It is known that the donor ability by resonance (+M effect) of
the halogens follows the series F > Cl > Br > I. For example, pKa
values of substituted phenols are 9.95 (phenol), 9.28 (m-F), 9.02 (m-
Cl), 9.11 (m-Br), and 9.17 (m-I), respectively. This trend is explained
in terms of electron-donating effect of fluorine; see: Modena, G.;
Scorrano, G. In The Chemistry of The Carbon-halogen Bond, Part 1;
Patai, S., Ed.; J ohn Wiley & Sons Ltd.: New York, 1973; Chapter 6,
pp 301-406.
(26) For examples of conformational studies on the syn/anti equi-
librium, see: (a) Davis, J . P.; Hart, P. A. Tetrahedron 1972, 28, 2883-
2891. (b) Stolarski, R.; Dudycz, L.; Shugar, D. Eur. J . Biochem. 1980,
108, 111-121. (c) Stolarski, R.; Hagberg, C.; Shugar, D. Eur. J .
Biochem. 1984, 138, 187-192. (d) Davies, D. B.; Rajani, P.; Sadikot,
H. J . Chem. Soc., Perkin Trans. 2 1985, 279-285. (e) Rosemeyer, H.;
Toth, G.; Golankiewicz, B.; Kazimierczuk, Z.; Bourgeois, W.; Kretschmer,
U.; Muth, H. P.; Seela, F. J . Org. Chem. 1990, 55, 5784-5790. (f) Cho,
B. P.; Evans, F. E. Biochem. Biophys. Res. Commun. 1991, 180, 273-
278.
(23) Seela, F.; Rosemeyer, H.; Fischer, S. Helv. Chim. Acta 1990,
73, 1602-1611.
(24) The pKa values of all nucleosides were measured by a slight
modification of the method reported by Shugar and Fox. A 0.2 M NaCl
solution was used for all pH ranges instead of buffers; see: Shugar,
D.; Fox, J . J . Biochim. Biophys. Acta 1952, 9, 199-218.
(27) For examples of conformational studies of the ribose conformer,
see: (a) Altona, C.; Sundaralingam, M. J . Am. Chem. Soc. 1973, 95,
2333-2344. (b) Davies, D. B.; Danyluk, S. S. Biochemistry 1974, 13,
4417-4434. (c) Yokoyama, S.; Yamaizumi, Z.; Nishimura, S.; Miyaza-
wa, T. Nucleic Acids Res. 1979, 6, 2611-2626.

7164 J . Org. Chem., Vol. 64, No. 19, 1999
Ta ble 2. 1H NMR Sp ectr a l Da ta a ,b
Minakawa et al.
c
d
d
compd
inosine
NH
H-8
H-2
NH2
H-1′(J 1′,2′
)
H-2′
H-3′(J 3′,4′
)
H-4′
H-5′
12.36
11.18
11.19
11.58
11.60
11.49
11.04
8.31
8.22
8.47
8.51
8.50
8.44
8.37
8.32
8.29
8.47
8.53
8.54
8.49
7.91
7.66
8.04
8.13
8.13
8.05
7.16
7.41
7.36
7.41
7.44
6.91
8.11
7.66
7.67
7.65
7.73
7.88
5.84 (5.9)
5.71 (6.6)
5.89 (5.7)
6.23 (4.9)
6.47 (5.0)
6.61 (5.5)
6.00 (5.4)
5.85 (6.6)
5.75 (6.2)
5.92 (5.7)
6.35 (5.1)
6.48 (5.5)
6.62 (5.5)
5.67 (5.9)
5.46 (6.1)
5.77 (5.5)
6.28 (4.9)
6.43 (4.9)
4.46
4.23
4.33
4.33
4.38
4.44
4.30
4.59
4.31
4.35
4.37
4.40
4.46
4.37
4.21
4.26
4.27
4.30
4.10 (3.9)
4.06 (3.3)
4.09 (3.4)
4.08 (4.4)
4.08 (3.8)
4.10 (3.3)
4.09 (3.4)
4.12 (3.3)
4.09 (3.3)
4.09 (3.4)
4.11 (4.4)
4.10 (4.4)
4.10 (4.4)
4.05 (3.3)
4.01 (3.9)
4.04 (3.9)
4.06 (4.4)
4.06 (4.4)
3.92
3.94
3.95
3.93
3.93
3.92
3.94
3.94
3.95
3.94
3.93
3.92
3.92
3.84
3.87
3.89
3.88
3.88
3.62, 3.52
3.62, 3.56
3.67, 3.57
3.68, 3.57
3.65, 3.54
3.59, 3.55
3.67, 3.57
3.64, 3.56
3.66, 3.59
3.65, 3.57
3.66, 3.56
3.64, 3.54
3.60, 3.55
3.58, 3.48
3.61, 3.51
3.60, 3.55
3.63, 3.53
3.61, 3.55
3-zI
46 (X ) F)
6 (X ) Cl)
7 (X ) Br)
8 (X ) I)
3-Me-3zI
adenosine
3-zA
64 (X ) F)
15 (X ) Cl)
16 (X ) Br)
17 (X ) I)
guanosine
3-zG
7.32
6.15
6.17
6.43
6.46
6.40
6.42
5.56
5.50
5.76
5.71
10.59
10.29
10.47
10.67
10.72
52 (X ) F)
25 (X ) Cl)
26 (X ) Br)
a
b
Chemical shifts in DMSO-d6. Assignment of 1H signals was based on two-dimensional NMR experiments. Full spectral data of
synthetic nucleosides are provided in the Supporting Information. c Values represent 6-NH2 for adenosine derivatives and 2-NH2 for
guanosine derivatives. Coupling constants (in parentheses).
d
Ta ble 3. 13C NMR Sp ectr a l Da ta a ,b
compd
inosine
C2
C3
C4
C5
C6
C8
C1′
C2′
C3′
C4′
C5′
145.8
129.5
115.1
128.6
131.1
136.6
152.3
137.7
126.0
140.1
142.7
148.8
153.6
142.4
135.4
143.9
144.6
148.1
138.3
129.3
134.3
135.2
137.2
149.0
139.4
126.6
133.7
134.7
137.1
151.2
147.6
132.0
137.6
138.5
124.4
131.8
132.3
132.6
133.0
133.3
119.3
126.8
128.5
127.9
128.2
128.5
116.7
123.0
122.3
124.1
124.5
156.5
157.9
156.3
156.8
156.9
157.1
156.1
152.0
149.4
151.6
152.0
152.5
156.7
156.4
154.2
155.2
155.4
138.7
139.2
139.5
139.6
139.7
139.8
139.8
140.2
140.2
140.1
140.2
140.3
135.5
136.6
137.0
137.0
137.2
87.4
88.7
89.4
88.4
87.6
86.1
87.9
88.7
89.4
88.4
87.6
86.1
86.3
88.1
89.0
88.0
87.4
74.0
74.5
75.2
75.3
75.0
74.4
73.4
73.9
74.8
75.0
74.8
74.2
73.6
73.8
74.7
74.9
74.7
70.2
70.1
69.8
69.4
69.4
69.7
70.6
70.1
69.9
69.6
69.6
69.8
70.3
70.1
69.9
69.5
69.5
85.6
85.4
85.5
85.0
85.0
85.1
85.8
85.6
85.4
85.0
84.9
84.9
85.1
85.2
85.2
84.7
84.7
61.2
61.1
60.8
60.4
60.5
60.8
61.6
61.2
61.0
60.7
60.7
60.9
61.4
61.2
61.0
60.6
60.7
3-zI
93.3
137.5
98.4
83.0
49.4
46 (X ) F)
6 (X ) Cl)
7 (X ) Br)
8 (X ) I)
adenosine
3-zA
64 (X ) F)
15 (X ) Cl)
16 (X ) Br)
17 (X ) I)
guanosine
3-zG
97.5
140.8
101.8
87.8
56.4
70.5
121.4
75.5
52 (X ) F)
25 (X ) Cl)
26 (X ) Br)
61.5
a
b
Chemical shifts in DMSO-d6. Assignment of 13C signals was based on DEPT and HSQC experiments. Full spectral data including
coupling constants of 46, 64, and 52 are provided in the Supporting Information.
differences in H-1′ chemical shifts were observed between
3-deazapurine ribonucleosides and 3-deaza-3-halopurine
ribonucleosides, especially the 3-chloro, bromo, and iodo
derivatives. For example, the H-1′ chemical shift of
3-deazainosine was characterized at 5.71 ppm, whereas
those of 6, 7, and 8 exhibited large downfield shifts (0.6-
0.9 ppm). The downfield shift was much less (0.18 ppm)
in 46. Although small downfield shifts (0.1-0.2 ppm)
were observed in H-2’s, other sugar protons exhibited
almost the same chemical shifts as 3-zI. Furthermore,
the same tendency was observed in 3-deazaadenosine and
3-deazaguanosine derivatives. Such anomalous downfield
shifts of proton signals were observed in steric compres-
sion molecules.28 As one interpretation of this phenom-
enon, intramolecular van der Waals effects, which cause
magnetic deshielding, with other proximate atoms in-
cluding hydrogen and halogen atoms can be considered.29
Furthermore, it is known that van der Waals effects
between a hydrogen atom and halogens such as chlorine
and bromine atoms show a larger magnetic deshielding
than those between two hydrogen atoms.29b The above
description is in good agreement with our 1H NMR
spectral data, since the downfield shifts in 3-Me-3zI, for
which the same effects is expected between the H-1′ and
one of the hydrogen atoms of methyl group, was smaller
(0.29 ppm) than those of 3-deaza-3-halopurine derivatives
such as 6-8. Although 3-deaza-3-halopurine ribonucleo-
sides are considered to be rather flexible molecules for
sugar pucker modes, which may also attribute to change
the proton chemical shifts, van der Waals effects would
be one of the explanations of the anomalous proton
deshielding. In the 13C NMR spectra (Table 3), small
upfield shifts of C-1′ due to van der Waals effects were
observed except for 3-deaza-3-fluoro derivatives.29a,30 The
(29) For examples, see: (a) Schaefer, T.; Reynolds, W. F.; Yonemono,
T. Can. J . Chem. 1963, 41, 2969-2976. (b) Yonemoto, T. Can. J . Chem.
1966, 44, 223-231. (c) De Coen, J . L.; Elefante, G.; Liquori, A. M.;
Damiani, A. Nature 1967, 216, 910-913. (d) Castellano, S.; Sun, C.;
Kostelnik, R. Tetrahedron Lett. 1967, 51, 5205-5209. (e) Bartle, K.
D.; Smith, J . A. S. Spectrochim. Acta 1967, 23A, 1689-1714. (f)
Rummens, F. H. A. NMR: Basic Princ. Prog. 1975, 10, 118pp. (g) Bohr,
J . E.; Hunt, K. L. C. J . Chem. Phys. 1987, 86, 5441-5448.
(30) Cheney, B. V.; Grant, D. M. J . Am. Chem. Soc. 1968, 90, 5386-
5390.
(28) For examples, see: (a) Slomp, G.; McGarvey, B. R. J . Am. Chem.
Soc. 1959, 81, 2200-2201. (b) Nagata, W.; Terasawa, T.; Tori, K. J .
Am. Chem. Soc. 1964, 86, 3746-3749. (c) Arnold, D. R.; Trecker, D.
J .; Whipple, E. B. J . Am. Chem. Soc. 1965, 87, 2596-2602. (d)
Winstein, S.; Carter, P.; Anet, F. A. L.; Bourn, A. J . R. J . Am. Chem.
Soc. 1965, 87, 5247-5249. (e) Cheney, B. V. J . Am. Chem. Soc. 1968,
90, 5386-5390.

3-Deaza-3-halopurine Ribonucleosides
J . Org. Chem., Vol. 64, No. 19, 1999 7165
Ta ble 4. Resu lts of NOE Exp er im en ts on 3-Dea za in osin e
Der iva tivesa ,b
65 66 inosine 3-zI 46
6
7
8
NOE of H-1′ (%) 7.8
1.1
9.4
3.0
5.1
4.7
0.9
5.2 2.4 1.0
3.3 4.8 7.8 10.1 10.8
0.8 0.8 1.5 1.4 1.8
0.6
0.6
H-2′
H-3′
0
0
H-2′ + H-3′ (%)
0
12.4
5.6
4.1 5.6 9.3 11.5 12.6
a
b
On irradiation of H-8 (H-2 for 66). Measured in DMSO-d6
F igu r e 1. Structures of conformationally fixed nucleosides.
(0.05 M, 400 MHz).
large differences in chemical shifts, especially at C-3,
between 3-deazapurine ribonucleosides and 3-deaza-3-
halopurine ribonucleosides were consistent with those
between benzene and monohalogenated benzenes.31 Thus,
downfield shifts in proportion to electronegativity were
observed in the 3-fluoro and 3-chloro derivatives. In
contrast, upfield shifts were observed due to heavy-atom
effects, which are greater than inductive effects, in the
3-bromo and 3-iodo derivatives.
The ribose conformation of the 3-deaza-3-halopurine
ribonucleosides is easily predicted from the coupling
constants of the ribose (J 1′,2′ and J 3′,4′) listed in Table 2.27
As is well-known, the sugar pucker modes of natural
purine ribonucleosides prefer a C2′-endo conformer.3 As
can be seen from the coupling constants, the C2′-endo
conformers of inosine and guanosine in DMSO-d6 were
calculated to have a population of 60%, while that of
adenosine was slightly higher (66%). Although the cou-
pling constants (J 1′,2′ and J 3′,4′) of the 3-deaza-3-halopurine
ribonucleosides show some scatter, all of the nucleosides
were found to exist preferentially as the C2′-endo in
solution, like natural purine ribonucleosides.
The syn/anti conformation around the glycosyl bond
was examined using NOE experiments. NOE spectros-
copy has been found to give a direct approach to qualita-
tive and semiquantitative information about the syn/anti
conformation. Thus, one can presume if irradiation of H-8
in purine ribonucleosides exhibits a strong NOE at H-1′
and small NOEs at H-2′ and H-3′, then the syn conformer
should be preferred. On the other hand, if irradiation of
H-8 exhibits a strong NOE at H-2′ and a smaller NOE
at H-1′, the anti conformer should dominate. Rosemeyer
et al.26e reported a semiquantitative estimation of the syn/
anti conformer populations of nucleosides using NOE
spectroscopy. In their method, N3,5′-anhydroisoguanosine
(65) and 2′-deoxywyosine (66) were used as syn- and anti-
fixed nucleosides, respectively, and NOE values at H-1′,
H-2′, and H-3′ upon irradiation of H-8 (H-2 for 66) were
used to established a calibration graph for the semiquan-
titative estimation of the syn and anti conformer popula-
tions of â-D-ribonucleosides. We used this method to
obtain a semiquantitative estimation of the syn and anti
populations of the 3-deaza-3-halopurine ribonucleosides.
Initially, 65 and 66 were prepared according to previous
reports (Figure 1).26e,32 The NOE measurements were
performed in the same manner as in Rosemeyer’s report,
and the NOE data are listed in Table 4. A maximal NOE
was observed at H-1′ upon irradiation of H-8 in syn-fixed
65 (7.8%), and simultaneously, no NOE was observed at
H-2′ and H-3′. On the other hand, anti-fixed 66 exhibited
F igu r e 2. Calibration graph for estimating the syn and anti
conformer populations of inosine and 3-deazainosine deriva-
tives.
large NOE values at H-2′ and H-3′ (12.4%) upon irradia-
tion of H-8, while a small NOE (1.1%) was observed at
H-1′. A calibration graph was set up with these data, and
the NOE data of the 3-deaza-3-halopurine ribonucleo-
sides, including those of inosine and 3-zI, were plotted
on the calibration graph (Figure 2). As shown in Figure
2, inosine and 3-zI slightly prefer the syn range (65%
syn). On the other hand, the introduction of halogens at
the 3-position resulted in a predominance for the anti
conformer with an increasing van der Waals radius of
the halogen substituents. Compound 8 exhibited the
same NOE tendency as anti-fixed 2′-deoxywyosine (66)
and was estimated to be a 100% anti substrate. The NOE
values at H-2′ and H-3′ of 7 and 6 were smaller than
those in the anti-fixed model, and were estimated to be
92% and 75% anti substrates, respectively. However, the
NOE values of these nucleosides at H-1′ were smaller
than that in the anti-fixed model, and thus, 7 and 6 are
also considered to be rather fixed in the anti conforma-
tional range due to steric repulsion of the bromo and
chloro substituents. 3-Deaza-3-fluoroinosine (46) was
estimated to be 45% anti and 30% syn based on the
calibration graph. One possible explanation for this
discrepancy is that this compound shows free rotation
around the glycosyl linkage and a predominant unusual
high-syn conformation with a reduction of NOE.26e These
NOE experiments also indicated that the sugar of 3-de-
azainosine derivatives prefers the C2′-endo conformation;
i.e., strong NOE’s were observed at H-2’s and smaller
ones were observed at H-3′.33 The same tendency was
(31) Silverstein, R. M.; Bassler, G. C.; Morrill, T. C. In Spectrometric
Identification of Organic Compounds, 4th Ed.; J ohn Wiley & Sons:
New York, 1981; pp 264-266.
(32) Golankiewicz, B.; Ostrowski, T.; Folkman, W. Nucleic Acids Res.
1990, 18, 4779-4782.

7166 J . Org. Chem., Vol. 64, No. 19, 1999
Minakawa et al.
1
foam: MS m/z 427, 429 (M+); H NMR (CDCl3) 13.06 (br s, 1
H), 8.27 (s, 1 H), 7.46 (s, 1 H), 6.75 (d, 1 H, J ) 4.4 Hz), 5.58
(dd, 1 H, J ) 3.9, 5.0 Hz), 5.43 (dd, 1 H, J ) 5.0, 5.5 Hz), 4.42
(m, 3 H), 2.20, 2.14, 2.12 (each s, each 3 H). Used immediately
in the next stage.
observed in 3-deazaadenosine and 3-deazaguanosine
derivatives (data not shown).
In conclusion, the synthesis of 3-deaza-3-halopurine
ribonucleosides was accomplished by direct halogenation
of the 3-deazapurine ribonucleosides or derivatization
from the 5-formylimidazole-4-carboxamide derivatives.
3-Deazapurine ribonucleosides with halogens (iodo, bromo,
and chloro) at the 3-position should be good model
compounds for studying nucleoside-enzyme interactions
with fixed glycosyl torsion angles in the anti region,
flexible sugar conformation, and pKa values at the
nucleobases close to those in natural ribonucleobases. On
the other hand, 3-deaza-3-fluoropurine ribonucleosides
should be better probes for such studies in models with
free rotation around the glycosyl bond than 3-deazapu-
rine ribonucleosides. The use of these analogues in
enzyme reactions should be the force of further study.
7-Br om o-1-(2,3,5-tr i-O-acetyl-â-D-r ibofu r an osyl)im idazo-
[4,5-c]p yr id in -4(5H)-on e (4). In a manner similar to that
above, the reaction of 2 (393 mg, 1 mmol) with N-bromosuc-
cinimide (NBS) in dry CH2Cl2 gave 4 (366 mg, 78%) as a bright
purple glass: MS m/z 471, 473 (M+); H NMR (CDCl3) 13.19
1
(br s, 1 H), 8.26 (s, 1 H), 7.51 (s, 1 H), 6.89 (d, 1 H, J ) 4.4
Hz), 5.59 (dd, 1 H, J ) 4.4, 5.0 Hz), 5.42 (dd, 1 H, J ) 5.0, 5.5
Hz), 4.47 (ddd, 1 H, J ) 5.5, 2.7, 2.2 Hz), 4.42 (s, 2 H), 2.19
2.13, 2.11 (each s, each 3 H). Used immediately in the next
stage.
7-Iod o-1-(2,3,5-tr i-O-a cetyl-â-D-r ibofu r a n osyl)im id a zo-
[4,5-c]p yr id in -4(5H)-on e (5). In a manner similar to that
above, the reaction of 2 (393 mg, 1 mmol) with N-iodosuccin-
imide (NIS) in dry DMF under shading gave 5 (391 mg, 75%)
as a yellow foam: FAB-MS m/z 520 (MH+); H NMR (CDCl3)
1
13.01 (br s, 1 H), 8.25 (s, 1 H), 7.66 (s, 1 H), 7.05 (d, 1 H, J )
5.0 Hz), 5.62 (t, 1 H, J ) 5.0 Hz), 5.44 (dd, 1 H, J ) 5.0, 5.5
Hz), 4.49 (ddd, 1 H, J ) 5.5, 2.7, 2.2 Hz), 4.41 (br s, 2 H), 2.20
2.14, 2.13 (each s, each 3 H). Used immediately in the next
stage.
7-Ch lor o-1-â-D-r ib ofu r a n osylim id a zo[4,5-c]p yr id in -
4(5H)-on e (3-Ch lor o-3-d ea za in osin e, 6). Compound 3 (300
mg, 0.70 mmol) was dissolved in methanolic ammonia (satu-
rated at 0 °C, 10 mL), and the mixture was kept overnight at
room temperature. The solvent was removed in vacuo, and the
residue was purified by a silica gel column (2.7 × 6 cm), eluted
with 5-30% EtOH in CHCl3, to give 6 (115 mg, 55%, crystal-
lized from MeOH): mp 171-173 °C dec; FAB-MS m/z 302, 304
(MH+); UV λmax (H2O) 261 nm (ꢀ 9700); UV λmax (0.5 N HCl)
274 nm (ꢀ 8300); UV λmax (0.5 N NaOH) 269 nm (ꢀ 10 200).
Anal. Calcd for C11H12ClN3O5‚1/3MeOH: C, 43.58; H, 4.30; N,
13.46. Found: C, 43.21; H, 4.19; N, 13.24.
7-Br om o-1-â-D-r ib ofu r a n osylim id a zo[4,5-c]p yr id in -
4(5H)-on e (3-Br om o-3-d ea za in osin e, 7). In the same man-
ner as described for 6, the reaction of 4 (294 mg, 0.62 mmol)
with methanolic ammonia (10 mL) gave 7 (150 mg, 70%,
crystallized from MeOH): mp 150-152 °C dec; FAB-MS m/z
346, 348 (MH+); UV λmax (H2O) 264 nm (ꢀ 9800); UV λmax (0.5
N HCl) 276 nm (ꢀ 8500); UV λmax (0.5 N NaOH) 269 nm (ꢀ
10 200). Anal. Calcd for C11H12BrN3O5‚3/4MeOH: C, 38.12; H,
4.08; N, 11.35. Found: C, 37.97; H, 4.16; N, 11.25.
7-Iod o-1-â-D-r ibofu r a n osylim id a zo[4,5-c]p yr id in -4(5H)-
on e (3-d ea za -3-iod oin osin e, 8). In the same manner as for
6, the reaction of 5 (391 mg, 0.66 mmol) with methanolic
ammonia (15 mL) gave 8 (239 mg, 92%, crystallized from
MeOH-H2O): mp 189-191 °C dec; FAB-MS m/z 394 (MH+);
UV λmax (H2O) 263 nm (ꢀ 10 600); UV λmax (0.5 N HCl) 280 nm
(ꢀ 8400); UV λmax (0.5 N NaOH) 269 nm (ꢀ 11 100). Anal. Calcd
for C11H12IN3O5‚3/5MeOH: C, 33.79; H, 3.52; N, 10.19. Found:
C, 33.65; H, 3.56; N, 9.95.
Exp er im en ta l Section
Gen er a l Meth od s. Physical data were measured as fol-
lows: Melting points are uncorrected. 1H and 13C NMR spectra
were recorded at 270, 400, or 500 MHz and 67.5, 100, or 125
MHz instruments in CDCl3 or DMSO-d6 as the solvent with
tetramethylsilane as an internal standard. Chemical shifts are
reported in parts per million (δ), and signals are expressed as
s (singlet), d (doublet), t (triplet), q (quartet), m (multiplet),
or br (broad). All exchangeable protons were detected by
addition of D2O. TLC was done on Merck Kieselgel F254
precoated plates. Silica gel used for column chromatography
was YMC gel 60A (70-230 mesh).
1-(2,3,5-Tr i-O-a cetyl-â-D-r ibofu r a n osyl)im id a zo[4,5-c]-
p yr id in -4(5H)-on e (2). An aqueous MeN2H solution (50%, 15
mL) was added to a suspension of 112 (777 mg, 2.91 mmol) in
EtOH (30 mL), and the mixture was heated at 80 °C in a steel
container. After 5 h, the reaction mixture was concentrated
in vacuo, and the residue was dissolved in EtOH-50% aqueous
AcOH (10 mL-10 mL), and the mixture was stirred overnight
at room temperature. The solvent was removed in vacuo and
then coevaporated several times with EtOH and benzene to
give crude 3-deazainosine. Acetic anhydride (1.1 mL, 11.6
mmol) was added to a suspension of 3-deazainosine in pyridine
(30 mL), and the reaction mixture was stirred at room
temperature overnight. Five milliliters of EtOH was added to
the mixture to quench the reaction. The mixture was concen-
trated in vacuo, and the residue was dissolved in CHCl3, which
was washed with saturated aqueous NaHCO3, followed by
saturated brine. The separated organic layer was dried (Na2-
SO4) and concentrated in vacuo and then coevaporated several
times with toluene. The residue was purified by a silica gel
column (3.2 × 9 cm), eluted with 0-12% EtOH in CHCl3, to
give 2 (1.02 g, 89% from 1, crystallized from EtOH): mp 200-
1
201 °C; MS m/z 393 (M+); H NMR (CDCl3) 12.69 (br s, 1 H,
NH), 8.04 (s, 1 H), 7.36 (br d, 1 H, J ) 7.1 Hz), 6.65 (d, 1 H,
J ) 7.1 Hz), 5.98 (d, 1 H, J ) 5.0 Hz), 5.49 (dd, 1 H, J ) 5.0,
5.5 Hz), 5.41 (dd, 1 H, J ) 5.5, 4.4 Hz), 4.49 (m, 1 H), 4.42 (br
s, 2 H), 2.17, 2.16, 2.11 (each s, each 3 H). Anal. Calcd for
4-Am in o-7-br om o-1-(2,3,5-tr i-O-ter t-bu tyld im eth ylsilyl-
â-D-r ibofu r a n osyl)im id a zo[4,5-c]p yr id in e (11). Compound
9 (122 mg, 0.20 mmol) was dissolved in dry CH2Cl2 (5 mL)
and cooled at -15 °C. To the mixture was added the suspen-
sion of NBS (53 mg, 0.30 mmol) in dry CH2Cl2 (2 mL), and
the resulting mixture was stirred for 30 min. After addition
of cyclohexene (20 µL), the mixture was concentrated in vacuo,
and the residue was dissolved in AcOEt, which was washed
with H2O, followed by saturated brine. The separated organic
layer was dried (Na2SO4) and concentrated in vacuo. The
residue was purified by a silica gel column (1.7 × 9 cm), eluted
with 25-50% AcOEt in hexane, to give 11 (102 mg, 74%) as a
C
17H19N3O8: C, 51.91; H, 4.87; N, 10.68. Found: C, 51.74; H,
4.77; N, 10.85.
7-Ch lor o-1-(2,3,5-tr i-O-acetyl-â-D-r ibofu r an osyl)im idazo-
[4,5-c]p yr id in -4(5H)-on e (3). A mixture of 2 (393 mg, 1.0
mmol) and N-chlorosuccinimide (NCS, 200 mg, 1.5 mmol) in
dry CH2Cl2 (10 mL) was stirred at room temperature. After
15 h, NCS (266 mg, 2.0 mmol) was further added to the
mixture and the mixture was stirred for an additional 24 h.
After addition of cyclohexene (0.2 mL) to decompose an excess
NCS, the mixture was concentrated in vacuo, and the residue
was purified by a silica gel column (2.7 × 13 cm), eluted with
0-6% EtOH in CHCl3, to give 3 (302 mg, 71%) as a bright
yellow
1
yellow solid: FAB-MS m/z 687, 689 (MH+); H NMR (CDCl3)
8.26 (s, 1 H), 7.90 (s, 1 H), 6.94 (d, 1 H, J ) 7.1 Hz), 5.16 (br
s, 2 H), 4.36 (dd, 1 H, J ) 7.1, 4.4 Hz), 4.18 (d, 1 H, J ) 4.4
Hz), 4.10 (dd, 1 H, J ) 2.7, 2.2 Hz), 3.89 (dd, 1 H, J ) 2.7,
11.5 Hz), 3.79 (dd, 1 H, J ) 2.2, 11.5 Hz), 0.98, 0.95, 0.73 (each
s, each 9 H), 0.18, 0.17, 0.12, 0.11, -0.13, -0.44 (each s, each
3 H). Used immediately in the next stage.
(33) Van de Ven, F. J . M.; Hilbers, C. W. Eur. J . Biochem. 1988,
178, 1-38.

3-Deaza-3-halopurine Ribonucleosides
J . Org. Chem., Vol. 64, No. 19, 1999 7167
4-Am in o-7-iod o-1-(2,3,5-tr i-O-ter t-bu tyld im eth ylsilyl-â-
D-r ibofu r a n osyl)im id a zo[4,5-c]p yr id in e (12). A mixture of
9 (608 mg, 1.0 mmol) and NIS (450 mg, 2.0 mmol) in dry DMF
(10 mL) was stirred for 11 h at room temperature under
shading. After addition of cyclohexene (0.1 mL), the mixture
was concentrated in vacuo, and the residue was dissolved in
AcOEt, which was washed with H2O, followed by saturated
brine. The separated organic layer was dried (Na2SO4) and
concentrated in vacuo. The residue was purified by a silica
gel column (3.2 × 12 cm), eluted with 25-50% AcOEt in
hexane, to give 12 (524 mg, 71%) as a yellow solid: MS m/z
734 (M+); 1H NMR (CDCl3) 8.23 (s, 1 H), 8.10 (s, 1 H), 7.10 (d,
1 H, J ) 7.7 Hz), 5.14 (br s, 2 H), 4.42 (dd, 1 H, J ) 7.7, 4.4
Hz), 4.17 (d, 1 H, J ) 4.4 Hz), 4.10 (dd, 1 H, J ) 2.8, 2.2 Hz),
3.88 (dd, 1 H, J ) 2.8, 11.5 Hz), 3.78 (dd, 1 H, J ) 2.2, 11.5
Hz), 0.98, 0.97, 0.72 (each s, each 9 H), 0.17, 0.16, 0.13, 0.11,
-0.13, -0.43 (each s, each 3 H). Used immediately in the next
stage.
m/z 685, 687 (MH+); 1H NMR (CDCl3) 8.55 (br s, 1 H), 8.43 (s,
1 H), 8.11 (s, 1 H), 6.75 (d, 1 H, J ) 7.1 Hz), 4.34 (dd, 1 H, J
) 7.1, 4.4 Hz), 4.19 (br d, 1 H), 4.12 (br s, 1 H), 3.93 (dd, 1 H,
J ) 2.8, 11.5 Hz), 3.80 (dd, 1 H, J ) 1.7, 11.5 Hz), 2.52 (s, 3
H), 0.99, 0.95, 0.73 (each s, each 9 H), 0.18, 0.17, 0.13, 0.11,
-0.11, -0.49 (each s, each 3 H). Used immediately in the next
stage.
4-Am in o-7-ch lor o-1-â-D-r ib ofu r a n osylim id a zo[4,5-c]-
p yr id in e (3-Ch lor o-3-d ea za a d en osin e, 15). A solution of
14 (430 mg, 0.63 mmol) in methanolic ammonia (saturated at
0 °C, 25 mL) was heated for 70 h at 50 °C in a steel container.
The solvent was removed in vacuo, and the residue was
purified by a short silica gel column (3.2 × 5 cm) to remove
acetamide, eluted with 33-50% AcOEt in hexane. The eluates
were concentrated to dryness in vacuo to give 10. A THF
solution of TBAF (1 M, 2.3 mL, 2.3 mmol) was added to a
solution of 10 in dry THF (10 mL) at 0 °C. The mixture was
stirred for 1 h at room temperature and concentrated in vacuo.
The residue was purified by a silica gel column (2.7 × 11 cm),
eluted with 5-30% EtOH in CHCl3, to give 15 (128 mg, 68%,
crystallized from EtOH-H2O): mp 225-228 °C dec; FAB-MS
m/z 301, 303 (MH+); UV λmax (H2O) 270 nm (ꢀ 11 400); UV λmax
(0.5 N HCl) 269 nm (ꢀ 12 400); UV λmax (0.5 N NaOH) 270 nm
(ꢀ 11 900). Anal. Calcd for C11H13ClN4O4‚H2O: C, 41.45; H,
4.74; N, 17.58. Found: C, 41.58; H, 4.73; N, 17.32.
6-Aceta m id o-1-(2,3,5-tr i-O-ter t-bu tyld im eth ylsilyl-â-D-
r ibofu r a n osyl)im id a zo[4,5-c]p yr id in -4(5H)-on e (19). A
solution of 1813 (842 mg, 1.35 mmol) in dry pyridine (10 mL)
containing Ac2O (0.64 mL, 6.75 mmol) was stirred overnight
at room temperature. To the mixture was added methanolic
ammonia (saturated at 0 °C, 10 mL), and the whole was kept
for 5 min at room temperature. The solvent was removed in
vacuo, and the residue was dissolved in CHCl3, which was
washed by H2O, followed by saturated brine. The separated
organic layer was dried (Na2SO4) and concentrated in vacuo.
The residue was purified by a silica gel column (3.2 × 12 cm),
eluted with 0-8% EtOH in CHCl3, to give 19 (811 mg, 90%)
as a pale yellow foam: MS m/z 666 (M+); 1H NMR (CDCl3)
11.73 (br s, 1 H), 10.63 (br s, 1 H), 8.14 (s, 1 H), 7.58 (s, 1 H),
5.77 (d, 1 H, J ) 5.5 Hz), 4.27 (m, 2 H), 4.14 (br s, 1 H), 3.99-
3.80 (m, 2 H), 2.48 (s, 3 H), 0.96, 0.95, 0.94 (each s, each 9 H),
0.15, 0.14, -0.05, -0.28 (each s, each 3 H), 0.11 (s, 6 H). Used
immediately in the next stage.
4-Am in o-7-b r om o-1-â-D-r ib ofu r a n osylim id a zo[4,5-c]-
p yr id in e (3-Br om o-3-d ea za a d en osin e, 16). A THF solution
of TBAF (1 M, 3.1 mL, 3.1 mmol) was added to a solution of
11 (540 mg, 0.79 mmol) in dry THF (10 mL) at 0 °C. The
mixture was stirred for 1 h at room temperature and concen-
trated in vacuo. The residue was purified by a silica gel column
(3.2 × 9 cm), eluted with 5-30% EtOH in CHCl3, to give 12
(250 mg, 92%, crystallized from EtOH-H2O): mp 224-227
°C dec; FAB-MS m/z 345, 347 (MH+); UV λmax (H2O) 272 nm
(ꢀ 11 700); UV λmax (0.5 N HCl) 270 nm (ꢀ 11 900); UV λmax
(0.5 N NaOH) 272 nm (ꢀ 12 900). Anal. Calcd for C11H13
-
BrN4O4: C, 38.28; H, 3.80; N, 16.23. Found: C, 38.31; H, 3.76;
N, 16.17.
4-Am in o-7-iod o-1-â-D-r ibofu r a n osylim id a zo[4,5-c]p yr i-
d in e (3-Dea za -3-iod oa d en osin e, 17). In the same manner
as for 16, the reaction of 12 (524 mg, 0.71 mmol) with TBAF
(1 M THF solution, 2.9 mL, 2.9 mmol) gave 17 (228 mg, 82%,
crystallized from MeOH-H2O): mp 213-215 °C dec; FAB-MS
m/z 393 (MH+); UV λmax (H2O) 273 nm (ꢀ 12 600); UV λmax (0.5
N HCl) 271 nm (ꢀ 12 100); UV λmax (0.5 N NaOH) 273 nm (ꢀ
13 200). Anal. Calcd for C11H13IN4O4: C, 33.69; H, 3.34; N,
14.29. Found: C, 33.97; H, 3.31; N, 14.24.
4-Aceta m id o-1-(2,3,5-tr i-O-ter t-bu tyld im eth ylsilyl-â-D-
r ibofu r a n osyl)im id a zo[4,5-c]p yr id in e (13). A mixture of
9 (500 mg, 0.82 mmol), Et3N (0.45 mL, 3.3 mmol), DMAP (10
mg), and Ac2O (0.3 mL, 3.3 mmol) in dry CH3CN (10 mL) was
stirred for 20 h at room temperature, and EtOH (2 mL) was
added to the mixture to decompose an excess of Ac2O. The
mixture was concentrated in vacuo, the residue (mono- and
diacetylated compounds) was dissolved in methanolic ammonia
(saturated at 0 °C, 10 mL), and the mixture was kept for 20
min at room temperature. The solvent was removed in vacuo,
and the residue was purified by a silica gel column (3.2 × 8
cm), eluted with 0-6% EtOH in CHCl3, to give 13 (501 mg,
94%) as a bright yellow foam: MS m/z 650 (M+); 1H NMR
(CDCl3) 8.44 (br s, 1 H), 8.15 (d, 1 H, J ) 6.0 Hz), 8.10 (s, 1
H), 7.42 (d, 1 H, J ) 6.0 Hz), 5.83 (d, 1 H, J ) 7.7 Hz), 4.38
(dd, 1 H, J ) 7.7, 5.0 Hz), 4.20 (d, 1 H, J ) 5.0 Hz), 4.15 (t, 1
H, J ) 2.2 Hz), 3.96 (dd, 1 H, J ) 2.2, 11.5 Hz), 3.84 (dd, 1 H,
J ) 2.2, 11.5 Hz), 2.52 (br s, 3 H), 0.99, 0.95, 0.73 (each s,
each 9 H), 0.20, 0.17, 0.13, 0.11, -0.14, -0.63 (each s, each 3
H). Used immediately in the next stage.
4-Acet a m id o-7-ch lor o-1-(2,3,5-t r i-O-ter t-b u t yld im et h -
ylsilyl-â-D-r ibofu r a n osyl)im id a zo[4,5-c]p yr id in e (14). A
mixture of 13 (314 mg, 0.48 mmol) and NCS (321 mg, 2.4
mmol) in dry CH2Cl2 (15 mL) was stirred at room temperature.
After 27 h, NCS (321 mg, 2.4 mmol) was further added to the
mixture, which was stirred for an additional 18 h. After
addition of cyclohexene (0.5 mL), the mixture was concentrated
in vacuo. Diethyl ether (10 mL) was added to the residue, and
an insoluble material was removed by filtration. The filtrate
was concentrated in vacuo, and the residue was dissolved in
AcOEt, which was washed with H2O, followed by saturated
brine. The separated organic layer was dried (Na2SO4) and
concentrated in vacuo. The residue was purified by a silica
gel column (2.7 × 13 cm), eluted with 25-50% AcOEt in
hexane, to give 14 (248 mg, 75%) as a white solid: FAB-MS
6-Acet a m id o-7-ch lor o-1-(2,3,5-t r i-O-ter t-b u t yld im et h -
ylsilyl-â-D-r ib ofu r a n osyl)im id a zo[4,5-c]p yr id in -4(5H )-
on e (20). In a manner similar to that for 3, the reaction of 19
(700 mg, 1.05 mmol) with NCS (280 mg, 2.1 mmol) in dry CH2-
Cl2 (20 mL) to give 20 (654 mg, 89%) as a pale yellow foam:
1
FAB-MS m/z 701, 703 (MH+); H NMR (CDCl3) 12.00 (br s, 1
H), 8.21 (s, 1 H), 8.04 (br s, 1 H), 6.61 (d, 1 H, J ) 7.1 Hz),
4.32 (dd, 1 H, J ) 7.1, 4.4 Hz), 4.16 (dd, 1 H, J ) 4.4, 3.8 Hz),
4.09 (ddd, 1 H, J ) 3.8, 2.7, 2.2 Hz), 3.89 (dd, 1 H, J ) 2.7,
11.5 Hz), 3.78 (dd, 1 H, J ) 2.2, 11.5 Hz), 2.29 (s, 3 H), 0.96,
0.93, 0.75 (each s, each 9 H), 0.16, 0.14, 0.11, 0.10, -0.09, -0.39
(each s, each 3 H). Used immediately in the next stage.
6-Acet a m id o-7-br om o-1-(2,3,5-t r i-O-ter t-b u t yld im et h -
ylsilyl-â-D-r ib ofu r a n osyl)im id a zo[4,5-c]p yr id in -4(5H )-
on e (21). In a manner similar to that for 4, the reaction of 19
(623 mg, 0.94 mmol) with NBS (183 mg, 1.03 mmol) in dry
CH2Cl2 at -20 °C gave 21 (548 mg, 79%) as a pale brown
1
foam: FAB-MS m/z 745, 747 (MH+); H NMR (CDCl3) 12.11
(br s, 1 H), 8.20 (s, 1 H), 8.00 (br s, 1 H), 6.81 (d, 1 H, J ) 7.1
Hz), 4.36 (dd, 1 H, J ) 7.1, 4.4 Hz), 4.15 (dd, 1 H, J ) 4.4, 3.8
Hz), 4.08 (br s, 1 H), 3.87 (dd, 1 H, J ) 2.8, 11.5 Hz), 3.77 (dd,
1 H, J ) 1.7, 11.5 Hz), 2.30 (s, 3 H), 0.96, 0.94, 0.74 (each s,
each 9 H), 0.16, 0.14, 0.12, 0.11, -0.10, -0.39 (each s, each 3
H). Used immediately in the next stage.
6-Aceta m id o-7-iod o-1-(2,3,5-tr i-O-ter t-bu tyld im eth ylsi-
lyl-â-D-r ibofu r an osyl)im idazo[4,5-c]pyr idin -4(5H)-on e (22).
In the same manner as for 5, the reaction of 19 (646 mg, 0.97
mmol) with NIS (327 mg, 1.46 mmol) gave 22 (548 mg, 79%)
1
as a yellow foam: FAB-MS m/z 793 (MH+); H NMR (CDCl3)
12.28 (br s, 1 H), 8.18 (s, 1 H), 8.13 (br s, 1 H), 7.04 (d, 1 H, J
) 7.7 Hz), 4.42 (dd, 1 H, J ) 7.7, 4.4 Hz), 4.15 (d, 1 H, J ) 4.4
Hz), 4.09 (br s, 1 H), 3.85 (dd, 1 H, J ) 2.7, 11.5 Hz), 3.76 (dd,

7168 J . Org. Chem., Vol. 64, No. 19, 1999
Minakawa et al.
1 H, J ) 2.2, 11.5 Hz), 2.30 (s, 3 H), 0.95 (s, 18 H), 0.73 (s, 9
H), 0.15, 0.14, 0.13, 0.11, -0.11, -0.37 (each s, each 3 H). Used
immediately in the next stage.
solution of isoamyl nitrite (5 mL) in CH2I2 (50 mL) was heated
at 100 °C, and a solution of 28 (3.51 g, 8.5 mmol) in CHCl3 (5
mL) was added dropwise to the preheated solution. After being
stirred for 15 min at 100 °C, the reaction mixture was absorbed
onto a silica gel column (5.0 × 20 cm), which was eluted with
0-4% EtOH in CHCl3 to give 29 (3.07 g, 69%, crystallized from
hexane-AcOEt): mp 159-160 °C; MS m/z 523 (M+); 1H NMR
(CDCl3) 8.01 (s, 1 H), 7.00 (br s, 1 H), 5.93 (d, 1 H, J ) 3.3
Hz), 5.36 (br s, 1 H), 4.82 (dd, 1 H, J ) 3.3, 5.9 Hz), 4.71 (dd,
1 H, J ) 5.9, 2.6 Hz), 4.35 (m, 1 H), 3.91 (dd, 1 H, J ) 1.9,
11.2 Hz), 3.80 (dd, 1 H, J ) 2.6, 11.2 Hz), 1.62, 1.36 (each s,
6-Am in o-7-ch lor o-1-(2,3,5-tr i-O-ter t-bu tyld im eth ylsilyl-
â-D-r ibofu r a n osyl)im id a zo[4,5-c]p yr id in -4(5H)-on e (23).
A solution of 20 (646 mg, 0.92 mmol) in methanolic ammonia
(saturated at 0 °C, 25 mL) was kept for 2 days at room
temperature. The solvent was removed in vacuo, and the
residue was purified by a silica gel column (3.2 × 12 cm), eluted
with 0-8% EtOH in CHCl3, to give 23 (464 mg, 76%) as a
white solid: FAB-MS m/z 659 (MH+); 1H NMR (CDCl3) 13.03
(br s, 1 H), 8.02 (s, 1 H), 6.68 (d, 1 H, J ) 7.1 Hz), 5.36 (br s,
2 H), 4.30 (dd, 1 H, J ) 7.1, 4.4 Hz), 4.16 (d, 1 H, J ) 4.4 Hz),
4.07 (br s, 1 H), 3.88 (dd, 1 H, J ) 2.7, 11.5 Hz), 3.78 (dd, 1 H,
J ) 2.2, 11.5 Hz), 0.96, 0.94, 0.77 (each s, each 9 H), 0.16,
0.14, 0.12, 0.10, -0.09, -0.36 (each s, each 3 H). Used
immediately in the next stage.
each 3 H), 0.88 (s, 9 H), 0.08 (s, 6 H). Anal. Calcd for C18H30
-
IN3O5Si: C, 41.30; H, 5.78; N, 8.02. Found: C, 41.56; H, 5.57;
N, 7.80.
5-Iod o-1-(2,3,5-t r i-O-ter t-b u t yld im et h ylsilyl-â-D-r ib o-
fu r a n osyl)im id a zole-4-ca r boxa m id e (31). A solution of 3012
(24.42 g, 49.3 mmol) in methanolic ammonia (saturated at 0
°C, 150 mL) was kept overnight at room temperature. The
solvent was removed in vacuo, and EtOH was added to the
residue. The suspension was heated in a water bath and then
cooled to room temperature. The precipitate was collected and
washed with EtOH. The resulting yellow solid was silylated
in the same manner as for 28 to give 31 (22.32 g, 64% as a
6-Am in o-7-br om o-1-(2,3,5-tr i-O-ter t-bu tyld im eth ylsilyl-
â-D-r ibofu r a n osyl)im id a zo[4,5-c]p yr id in -4(5H)-on e (24).
In the same manner as above, the reaction of 21 (548 mg, 0.74
mmol) with methanolic ammonia (saturated at 0 °C, 25 mL)
gave 24 (424 mg, 82%) as a brown foam: FAB-MS m/z 703,
1
705 (MH+); H NMR (CDCl3) 13.13 (br s, 1 H), 8.01 (s, 1 H),
1
6.88 (d, 1 H, J ) 7.1 Hz), 5.41 (br s, 2 H), 4.33 (dd, 1 H, J )
7.1, 4.4 Hz), 4.16 (d, 1 H, J ) 4.4 Hz), 4.07 (br s, 1 H), 3.86
(dd, 1 H, J ) 2.8, 11.5 Hz), 3.77 (br d, 1 H, J ) 11.5 Hz), 0.96,
0.95, 0.78 (each s, each 9 H), 0.15, 0.14, 0.12, 0.10, -0.10, -0.35
(each s, each 3 H). Used immediately in the next stage.
6-Am in o-7-ch lor o-1-â-D-r ib ofu r a n osylim id a zo[4,5-c]-
p yr id in -4(5H)-on e (3-Ch lor o-3-d ea za gu a n osin e, 25). A
THF solution of TBAF (1 M, 2.7 mL, 2.7 mmol) was added to
a solution of 23 (450 mg, 0.68 mmol) in dry THF (10 mL) at 0
°C. The mixture was stirred for 40 min at room temperature
and neutralized with AcOH. The solvent was removed in
vacuo, and a mixture of H2O-EtOH (10 mL-10 mL) was
added to the resulting white solid. The suspension was heated
in a water bath and then cooled to room temperature. The
precipitate was collected and washed with EtOH to give 25
(182 mg, 84%) as a white powder: FAB-MS m/z 317 (MH+);
UV λmax (H2O) 275 nm (ꢀ 10 000), 310 nm (ꢀ 8400); λmax (0.5 N
HCl) 293 nm (ꢀ 10 500), 320 (sh) nm (ꢀ 7400); λmax (0.5 N
NaOH) 287 nm (ꢀ 11 500). Anal. Calcd for C11H13ClN4O5‚
1/2EtOH: C, 42.42; H, 4.75; N, 16.49. Found: C, 42.11; H, 4.42;
N, 16.66.
yellow foam): MS m/z 711 (M+); H NMR (CDCl3) 8.10 (s, 1
H), 6.99 (br s, 1 H), 5.79 (d, 1 H, J ) 5.9 Hz), 5.31 (br s, 1 H),
4.29 (dd, 1 H, J ) 5.9, 3.6 Hz), 4.26 (m, 1 H), 4.18 (m, 1 H),
3.88 (dd, 1 H, J ) 2.6, 11.2 Hz), 3.75 (dd, 1 H, J ) 2.0, 11.2
Hz), 0.94, 0.91, 0.83 (each s, each 9 H), 0.12, 0.09, 0.08, -0.02,
-0.05, -0.20 (each s, each 3 H). Anal. Calcd for C27H54IN3O5-
Si3: C, 45.55; H, 7.65; N, 5.90. Found: C, 45.80; H, 7.46; N,
5.82.
5(E)-(2-Ca r b om et h oxyvin yl)-1-(5-O-ter t-b u t yld im et h -
ylsilyl-2,3-O-isopr opyliden e-â-D-r ibofu r an osyl)im idazole-
4-ca r boxa m id e (32). A mixture of 29 (1.57 g, 3.0 mmol),
bis(benzonitrile)palladium dichloride (86 mg, 0.23 mmol),
methyl acrylate (0.54 mL, 6.0 mmol), and Et3N (0.63 mL, 4.5
mmol) in dry CH3CN (10 mL) in a sealed glass tube was heated
for 14 h at 100 °C. The reaction mixture was filtered through
a Celite pad and washed with EtOH. The combined filtrate
and washings were concentrated in vacuo, and the residue was
purified by a silica gel column (3.6 × 15 cm), eluted with
hexane/AcOEt (2:1-1:2), to give 32 (1.15 g, 80% as a yellow
foam): EI-MS m/z 481 (M+); 1H NMR (CDCl3) 8.16, 6.93 (each
d, each 1 H, J ) 16.0 Hz), 7.93 (s, 1 H), 7.12, 5.40 (each br s,
each 1 H), 5.93 (d, 1 H, J ) 3.2 Hz), 4.79 (m, 2 H), 4.40 (m. 1
H), 3.87 (m, 2 H), 3.80 (s, 3 H), 1.64, 1.38 (each s, each 3 H),
0.89 (s, 9 H), 0.10, 0.01 (each s, each 3 H). Used immediately
in the next stage.
6-Am in o-7-b r om o-1-â-D-r ib ofu r a n osylim id a zo[4,5-c]-
p yr id in -4(5H)-on e (3-Br om o-3-d ea za gu a n osin e, 26). In
the same manner as for 25, the reaction of 24 (424 mg, 0.60
mmol) with TBAF (1 M THF solution, 2.4 mL, 2.4 mmol) gave
26 (136 mg, 63%) as a pale blue powder: FAB-MS m/z 361,
363 (MH+); UV λmax (H2O) 275 nm (ꢀ 9900), 311 nm (ꢀ 8400);
λmax (0.5 N HCl) 294 nm (ꢀ 10 100), 317 (sh) nm (ꢀ 7400); λmax
5(E)-(2-Ca r bom eth oxyvin yl)-1-(2,3,5-tr i-O-ter t-bu tyld i-
m e t h ylsilyl-â-D-r ib ofu r a n osyl)im id a zole -4-ca r b oxa m -
id e (33). In the same manner as described for 32, 31 (2.13 g,
3.0 mmol) was treated with methyl acrylate to give 33 (1.78
g, 89% as a yellow foam): EI-MS m/z 669 (M+); 1H NMR
(CDCl3) 8.12, 6.95 (each d, each 1 H, J ) 16.4 Hz), 7.95 (s, 1
H), 7.12, 5.37 (each br s, each 1 H), 5.89 (d, 1 H, J ) 7.1 Hz),
4.31 (dd, 1 H, J ) 7.1, 4.4 Hz), 4.19 (dd, 1 H, J ) 4.4, 1.2 Hz),
4.06 (ddd, 1 H, J ) 1.2, 3.1, 2.4 Hz), 3.86 (dd, 1 H, J ) 3.1,
11.4 Hz), 3.77 (dd, 1 H, J ) 2.4, 11.4 Hz), 3.78 (s, 3 H), 0.96,
0.94, 0.80 (each s, each 9 H), 0.14 (s, 6 H), 0.12, 0.11, -0.04,
-0.24 (each s, each 3 H). Used immediately in the next stage.
5-F or m yl-1-(5-O-ter t-b u t yld im et h ylsilyl-2,3-O-isop r o-
pyliden e-â-D-r ibofu r an osyl)im idazole-4-car boxam ide (34).
Ozonized oxygen was bubbled through a solution of 32 (4.53
g, 9.35 mmol) in MeOH (100 mL) at -78 °C. After 15 min, N2
gas was bubbled through the solution to remove excess
ozonized oxygen. Dimethyl sulfide (3.4 mL, 46.7 mmol) was
added to the solution at -78 °C, and then the reaction mixture
was allowed to warm to room temperature. The solvent was
removed in vacuo, and the residue was purified by a silica gel
column (6.0 × 10 cm), eluted with hexane/AcOEt (3:1-1:2), to
give 34 (3.91 g, 98%, crystallized from hexane-AcOEt): mp
(0.5 N NaOH) 285 nm (ꢀ 11 200). Anal. Calcd for C11H13
-
BrN4O5: C, 36.58; H, 3.63; N, 15.51. Found: C, 36.58; H, 3.72;
N, 15.13.
5-Am in o-1-(5-O-ter t-b u t yld im et h ylsilyl-2,3-O-isop r o-
pyliden e-â-D-r ibofu r an osyl)im idazole-4-car boxam ide (28).
A mixture of 27 (9.0 g, 30 mmol), imidazole (4.9 g, 72 mmol),
and tert-butyldimethylsilyl chloride (5.43 g, 36 mmol) in DMF
(120 mL) was stirred for 3.5 h at room temperature. The
reaction was quenched by addition of EtOH, and the mixture
was concentrated in vacuo. The residue was partitioned
between AcOEt and H2O, and the organic layer was washed
with H2O, followed by saturated brine. The separated organic
layer was dried (Na2SO4) and concentrated in vacuo. The
residue was purified by a silica gel column (6.0 × 8 cm), eluted
with hexane/AcOEt (1:1-0:1) to give 28 (12.5 g, quant as a
white foam): MS m/z 412 (M+); 1H NMR (CDCl3) 6.99 (s, 1
H), 6.46 (br s, 1 H), 5.56 (d, 1 H, J ) 3.3 Hz), 5.44 (br s, 2 H),
5.13 (br s, 1 H), 4.91 (m, 2 H), 4.16 (m, 1 H), 3.93 (dd, 1 H, J
) 2.0, 11.9 Hz), 3.82 (dd, 1 H, J ) 1.9, 11.9 Hz), 1.56, 1.34
(each s, each 3 H), 0.88 (s, 9 H), 0.09, 0.07 (each s, each 3 H).
Anal. Calcd for C18H32N4O5Si: C, 52.40; H, 7.82; N, 13.58.
Found: C, 52.12; H, 7.81; N, 13.25.
1
138-139 °C; EI-MS m/z 425 (M+); H NMR (CDCl3) 10.65 (s,
1 H), 8.21 (s, 1 H), 7.20, 5.58 (each br s, each 1 H), 6.57 (d, 1
H, J ) 1.7 Hz), 4.81 (dd, 1 H, J ) 1.7 Hz), 4.61 (m, 1 H), 4.38
(m. 1 H), 3.91 (m, 2 H), 1.63, 1.37 (each s, each 3 H), 0.90 (s,
5-Iod o-1-(5-O-ter t-bu tyld im eth ylsilyl-2,3-O-isop r op yl-
id en e-â-D-r ibofu r a n osyl)im id a zole-4-ca r boxa m id e (29). A

3-Deaza-3-halopurine Ribonucleosides
J . Org. Chem., Vol. 64, No. 19, 1999 7169
9 H), 0.12, 0.12 (each s, each 3 H). Anal. Calcd for C19H31N3O6-
Si: C, 53.63; H, 7.34; N, 9.87. Found: C, 53.60; H, 7.46; N,
9.93.
6.08 (d, 3/5 H, J ) 2.2 Hz), 5.51-5.32 (m, 1 H), 4.88-4.80 (m,
1 H), 4.71-4.61 (m, 1 H), 4.30 (dt, 2/5 H), 4.24 (dt, 3/5 H),
3.98 (dd, 3/5 H, J ) 2.2, 11.5 Hz), 3.95 (dd, 2/5 H, J ) 2.2,
11.5 Hz), 3.84 (dd, 3/5 H, J ) 2.2, 11.5 Hz), 3.83 (dd, 2/5 H, J
) 2.2, 11.5 Hz), 1.58, 1.35 (each s, each 3 H), 0.94 (s, 9 H),
0.14, 0.13 (each s, each 3 H).
5-F or m yl-1-(2,3,5-t r i-O-ter t-b u t yld im et h ylsilyl-â-D-r i-
bofu r a n osyl)im id a zole-4-ca r boxa m id e (35). A solution of
osmium tetraoxide in t-BuOH (0.02 M, 26.0 mL, 0.52 mmol)
was added dropwise to a solution of 33 (6.95 g, 10.4 mmol) in
acetone-H2O (9:1, 100 mL) containing N-methylmorpholine
N-oxide (2.43 g, 20.7 mmol), and the whole was stirred for 4
days at room temperature. An aqueous solution of sodium
periodate (9.10 g, 41.5 mmol in 25 mL of H2O) was added to
the mixture, which was stirred for an additional 23 h and then
filtered. The filtrate was reduced to a half volume in vacuo,
diluted with 1 M aqueous sodium thiosulfate (200 mL), and
extracted with AcOEt (3 × 70 mL). The combined organic
layers were washed with brine and dried (Na2SO4). The solvent
was removed in vacuo, and the residue was purified by a silica
gel column (5.0 × 12 cm), eluted with hexane/AcOEt (10:1-
3:1), to give 35 (5.39 g, 85% as a white foam): EI-MS m/z 613
(M+); 1H NMR (CDCl3) 10.63 (s, 1 H), 8.54 (s, 1 H), 7.25, 5.56
(each br s, each 1 H), 6.31 (d, 1 H, J ) 2.3 Hz), 4.22 (m, 1 H),
4.16 (m, 1 H), 4.11 (m, 1 H), 4.09 (dd, 1 H, J ) 2.2, 12.1 Hz),
3.83 (dd, 1 H, J ) 1.5, 11.8 Hz), 0.98, 0.91, 0.90 (each s, each
9 H), 0.18, 0.16, 0.11, 0.08, 0.06, -0.03 (each s, each 1H). Anal.
Calcd for C28H55N3O6Si3: C, 54.77; H, 9.03; N, 6.84. Found:
C, 54.59; H, 8.89; N, 6.66.
Physical data for compound 41: EI-MS m/z 455 (M+); UV
λ
max (MeOH) 271 nm; 1H NMR (CDCl3) 7.87 (s, 1 H), 7.16 (ddd,
1 H, J ) 16.5, 1.7, 4.4 Hz), 7.15, 5.31 (each br s, each 1 H),
6.74 (ddt, 1 H, J ) 16.5, 14.3, 5.5 Hz), 5.91 (d, 1 H, J ) 3.8
Hz), 5.08 (ddd, 1 H, J ) 1.7, 5.5, 47.3 Hz), 4.86 (dd, 1 H, J )
3.8, 6.0 Hz), 4.75 (dd, 1 H, J ) 6.0, 3.3 Hz), 4.37 (dt, 1 H, J )
3.3, 2.7 Hz), 3.92 (dd, 1 H, J ) 2.7, 11.5 Hz), 3.82 (dd, 1 H, J
) 2.7, 11.5 Hz), 1.62, 1.38 (each s, each 3 H), 0.91 (s, 9 H),
0.11, 0.10 (each s, each 3 H).
7-F lu or o-6-h yd r oxy-1-(5-O-ter t-bu tyld im eth ylsilyl-2,3-
O-isopr opyliden e-â-D-r ibofu r an osyl)-6,7-dih ydr oim idazo-
[4,5-c]p yr id in -4(5H)-on e (44). Meth od A. An aqueous so-
lution of KMnO4 (440 mg, 2.78 mmol, dissolved in 20 mL of
H2O) was added dropwise over 1 h to a solution of 39 (840
mg, 1.85 mmol) and 18-crown-6 (75 mg, 0.28 mmol) in dry THF
(30 mL) at 0 °C. After being stirred for 1 h, the reaction
mixture was filtered through a Celite pad, which was washed
with EtOH. The combined filtrate and washings were concen-
trated to dryness in vacuo, and the residue was partitioned
between CHCl3 and H2O. The aqueous layer was extracted
with CHCl3 three times. The combined organic layers were
concentrated in vacuo, and the residue was dissolved in CH2-
Cl2 (30 mL). To the solution was added an aqueous solution of
NaIO4 (1.19 g, 5.55 mmol, dissolved in 10 mL of H2O), and
the mixture was stirred for 17 h at room temperature. The
mixture was diluted with CHCl3 (40 mL), and the organic layer
was washed with H2O. The solvent was concentrated to half
of its original volume, and 5% aqueous Na2CO3 (20 mL) was
added to the solution, which was then heated for 1 h at 80 °C.
The reaction mixture was diluted with CHCl3, and the
separated organic layer was washed with H2O, followed by
brine. The organic layer was dried (Na2SO4) and concentrated
in vacuo, and the residue was purified by a silica gel column
(3.5 × 8 cm), eluted with 0-6% EtOH in CHCl3, to give 44
(624 mg, 74% as a white foam): FAB-MS m/z 458 (MH+); UV
5-(1-Hyd r oxy-2-p r op en yl)-1-(5-O-ter t-bu tyld im eth ylsi-
lyl-2,3-O-isop r op ylid en e-â-D-r ibofu r a n osyl)im id a zole-4-
ca r boxa m id e (36). A THF solution of vinylmagnesium bro-
mide (1 M, 27.5 mL, 27.5 mmol) was added dropwise over 30
min to a solution of 34 (3.9 g, 9.2 mmol) in dry THF (100 mL)
at -40 °C. After the mixture was stirred for 3.5 h at the same
temperature, the reaction was quenched by addition of 1 M
aqueous NH4Cl (55 mL) and then allowed to warm at room
temperature. The mixture was extracted with AcOEt (100 mL),
and the organic layer was washed with H2O, followed by brine.
The layer was dried (Na2SO4) and concentrated in vacuo. The
residue was purified by a silica gel column (4.8 × 12 cm), eluted
with hexane/AcOEt (2:1-1:3), to give 36 (3.85 g, 93%, crystal-
lized from hexane-AcOEt): mp 160-161 °C and 165-166 °C;
1
EI-MS m/z 453 (M+); H NMR (CDCl3) 7.74 (s, 1 H), 7.37 (d,
1
4/9 H, J ) 11.0 Hz), 7.02 (d, 5/9 H, J ) 11.0 Hz), 7.24, 5.54
(each br s, each 1 H), 6.17-6.00 (m, 1 H), 5.82 (d, 5/9 H, J )
3.3 Hz), 5.74 (d, 4/9 H, J ) 3.3 Hz), 5.41-5.24 (m, 1 H), 5.22-
5.12 (m, 2 H), 4.82 (dd, 4/9 H, J ) 6.0, 3.3 Hz), 4.80 (dd, 5/9 H,
J ) 6.0, 2.2 Hz), 4.75 (dd, 4/9 H, J ) 3.3, 6.0 Hz), 4.64 (dd, 5/9
H, J ) 3.3, 6.0 Hz), 4.42 (ddd, 5/9 H, J ) 2.2, 2.7, 2.8 Hz),
4.42 (ddd, 4/9 H, J ) 2.2, 2.7, 2.8 Hz), 3.92 (dd, 5/9 H, J ) 2.7,
11.5 Hz), 3.90 (dd, 4/9 H, J ) 2.7, 11.5 Hz), 3.80 (dd, 5/9 H, J
) 2.7, 11.5 Hz), 3.80 (dd, 4/9 H, J ) 2.7, 11.5 Hz), 1.59, 1.35
(each s, each 15/9 H), 1.59, 1.36 (each s, each 12/9 H), 0.87 (s,
36/9 H), 0.86 (s, 45/9 H), 0.08-0.07 (s × 4, 6 H). Anal. Calcd
for C21H35N3O6Si: C, 55.61; H, 7.78; N, 9.26. Found: C, 55.36;
H, 7.72; N, 9.22.
λmax (MeOH) 246 nm; H NMR (CDCl3) 8.38 (br d, 2/3 H, J )
2.9 Hz), 8.26 (br d, 1/3 H, J ) 3.0 Hz), 7.94 (d, 1/3 H), 7.88 (d,
2/3 H), 6.87 (m, 1 H), 5.87 (d, 1/3 H, J ) 2.7 Hz), 5.84 (d, 2/3
H, J ) 3.5 Hz), 5.80 (d, 2/3 H, J ) 50.7 Hz), 5.66 (d, 1/3 H, J
) 50.2 Hz), 5.51 (d, 2/3 H, J ) 1.2 Hz), 5.48 (d, 1/3 H, J ) 1.1
Hz), 4.89-4.78 (m, 2 H), 4.51 (m, 1/3 H,), 4.37 (m, 2/3 H), 3.93-
3.80 (m, 2 H), 1.61 and 1.37 (each s, each 6/3 H), 1.60, 1.38
(each s, each 3/3 H), 0.88 (s, 18/3 H), 0.82 (s, 9/3 H), 0.12, 0.03
(each s, each 3/3 H), 0.08, 0.07 (each s, each 6/3 H). Anal. Calcd
for
C
20H32FN3O6Si‚1/4EtOH: C, 52.49; H, 7.20; N, 8.96.
Found: C, 52.51; H, 7.16; N, 8.84.
Meth od B. A suspension of 40 (450 mg, 1.0 mmol) and
Lindlar catalyst (60 mg) in EtOH (30 mL) was shaken under
hydrogen (50 psi) on a Parr hydrogenation apparatus for 16 h
at room temperature. The reaction mixture was filtered
through a Celite pad, which was washed with EtOH. The
filtrate and washings were concentrated in vacuo to give 39.
Compound 39 was successively converted as above to give 44
(320 mg, 70% from 40).
5-(1-F lu or o-2-p r op en yl)-1-(5-O-ter t-bu tyld im eth ylsilyl-
2,3-O-isop r op ylid en e-â-D-r ibofu r a n osyl)im id a zole-4-ca r -
boxa m id e (39) a n d 5(E)-(3-flu or o-1-p r op en yl)-1-(5-O-ter t-
bu t yld im et h ylsilyl-2,3-O-isop r op ylid en e-â-D-r ib ofu r a n -
osyl)im id a zole-4-ca r boxa m id e (41). A solution of 36 (820
mg, 1.81 mmol) in dry CH2Cl2 was added dropwise over 40
min to a solution of DAST (0.5 mL, 3.6 mmol) in dry CH2Cl2
at -15 °C, and the whole was stirred for 1 h at that
temperature. To this mixture was added 1 M aqueous NaHCO3
(5 mL), and the whole was stirred for 20 min and then allowed
to warm at room temperature. The mixture was diluted with
AcOEt (100 mL), and the organic layer was washed with H2O,
followed by brine. The layer was dried (Na2SO4) and concen-
trated in vacuo, and the residue was purified by a silica gel
column (3.0 × 12 cm), eluted with hexane/AcOEt (2:1-2:3), to
give 39 (307 mg, 37% as a pale yellow foam) and 41 (255 mg,
31% as a yellow foam).
7-F lu or o-1-(5-O-ter t-b u t yld im et h ylsilyl-2,3-O-isop r o-
pylidene-â-D-ribofuranosyl)imidazo[4,5-c]pyridin-4(5H)-one
(45). A mixture of 44 (430 mg, 0.94 mmol), Ac2O (0.13 mL,
1.4 mmol), and DMAP (6 mg, 0.05 mmol) in dry pyridine (20
mL) was stirred for 1 h at room temperature. The reaction
was quenched by addition of EtOH (1 mL), and then the
mixture was heated for 30 min at 100 °C. The solvent was
removed in vacuo, and the residue was dissolved in CHCl3,
which was washed with H2O, followed by brine. The organic
layer was dried (Na2SO4) and concentrated in vacuo, and the
residue was purified by a silica gel column (3.0 × 9 cm), eluted
with 0-6% EtOH in CHCl3, to give 45 (380 mg, 92% as a white
foam): FAB-MS m/z 440 (MH+); UV λmax (MeOH) 259 nm; λmax
Physical data for compound 39: EI-MS m/z 455 (M+); UV
λ
max (MeOH) 240 nm; 1H NMR (CDCl3) 8.00 (s, 3/5 H), 7.93 (s,
2/5 H), 7.37-7.16 (m, 1 H, J ) 40.0 Hz), 7.10, 5.50 (each br s,
1
(H+) 260 nm; λmax (OH-) 268 nm; H NMR (CDCl3) 12.93 (br
each 1 H), 6.23-6.15 (m, 1 H), 6.12 (d, 2/5 H, J ) 2.2 Hz),
s, 1 H), 8.23 (s, 1 H), 7.29 (d, 1 H, J ) 5.2 Hz), 6.20 (d, 1 H, J

7170 J . Org. Chem., Vol. 64, No. 19, 1999
Minakawa et al.
) 3.0 Hz), 4.87 (dd, 1 H, J ) 6.0, 2.6 Hz), 4.81 (dd, 1 H, J )
3.0, 6.0 Hz), 4.42 (dt, 1 H, J ) 2.6, 3.0 Hz), 3.96 (dd, 1 H, J )
2.6, 11.4 Hz), 3.86 (dd, 1 H, J ) 3.0, 11.4 Hz), 1.63, 1.38 (each
s, each 3 H), 0.91 (s, 9 H), 0.12, 0.11 (each s, each 3 H). Anal.
Calcd for C20H30FN3O5Si: C, 54.65; H, 6.88; N, 9.56. Found:
C, 54.49; H, 6.78; N, 9.38.
FN3O5‚3/5MeOH: C, 45.76; H, 4.77; N, 13.80. Found: C, 45.47;
H, 4.65; N, 14.00.
5-Cya n oflu or om eth yl-1-(2,3,5-tr i-O-ter t-bu tyld im eth yl-
silyl-â-D-r ibofu r a n osyl)im id a zole-4-ca r boxa m id e (48). So-
dium cyanide (2.82 g, 57.6 mmol) was added to a solution of
35 (3.53 g, 5.86 mmol) in a mixture of AcOEt (150 mL) and 1
M aqueous NaHCO3 (150 mL), and the whole was vigorously
stirred for 20 h at room temperature. The separated organic
layer was washed with H2O, followed by brine. The organic
layer was dried (Na2SO4) and concentrated in vacuo to give
crude 47. A solution of 47 in dry CH2Cl2 was added dropwise
over 1 h to a solution of DAST (1.52 mL, 11.5 mmol) in dry
CH2Cl2 at -15 °C. The mixture was allowed to warm at room
temperature and stirred for 30 min. To this mixture was added
aqueous NaHCO3 (1 M, 40 mL), and the mixture was stirred
for 20 min. The mixture was extracted with CHCl3 (60 mL),
and the organic layer was washed with H2O, followed by brine.
The organic layer was dried (Na2SO4) and concentrated in
vacuo, and the residue was purified by a silica gel column (4.0
× 10 cm), eluted with hexane/AcOEt (9:1-3:1), to give 48 (2.69
g, 73% as a pale yellow foam): FAB-MS m/z 643 (MH+); UV
λmax (MeOH) 233 nm; 1H NMR (CDCl3) 8.20 (d, 3/5 H, J ) 1.8
Hz), 7.95 (d, 2/5 H, J ) 1.7 Hz), 7.74 (d, 2/5 H, J ) 42.9 Hz),
7.69 (d, 3/5 H, J ) 43.3 Hz), 7.13, 5.56 (each br s, each 1 H)
5.96 (d, 3/5 H, J ) 5.2 Hz), 5.91 (d, 2/5 H, J ) 6.6 Hz), 4.42
(dd, 1 H, J ) 6.6, 4.3 Hz), 4.30 (dd, 1 H, J ) 5.2, 4.1 Hz), 4.26-
4.22 (m, 1 H), 4.14-4.11 (m, 1 H), 4.10-3.78 (m, 2 H), 0.98,
0.93, 0.84 (each s, each 27/5 H), 0.96, 0.94, 0.83 (each s, each
18/5 H), 0.17, 0.17, 0.14, 0.14, 0.13, 0.12, 0.11, 0.10, -0.02,
5-(1-Hyd r oxyp r op yn yl)-1-(5-O-ter t-bu tyld im eth ylsilyl-
2,3-O-isop r op ylid en e-â-D-r ibofu r a n osyl)im id a zole-4-ca r -
boxa m id e (38). A hexane solution of BuLi (1.66 M, 15.0 mL,
24.8 mmol) was added at a rate that the temperature did not
exceed -65 °C to a solution of (trimethylsilyl)acetylene (3.6
mL, 25.4 mmol) in dry THF (40 mL). After the mixture was
stirred for 30 min, a solution of 34 (2.7 g, 6.35 mmol) in dry
THF (20 mL) was added dropwise over 1 h, and the whole was
stirred for an additional 1.5 h at -78 °C. The reaction was
quenched by addition of 1 M aqueous NH4Cl (40 mL) and then
allowed to warm at room temperature. The mixture was
extracted with AcOEt (150 mL), and the organic layer was
washed with H2O, followed by brine. The layer was dried (Na2-
SO4) and concentrated in vacuo. To the residue was added
methanolic ammonia (saturated at 0 °C, 100 mL), and the
mixture was kept for 3 h at room temperature. The solvent
was removed in vacuo, and the residue was purified by a silica
gel column (4.6 × 10 cm), eluted with hexane/AcOEt (3:1-2:
3), to give 38 (2.57 g, 90% as a white foam): EI-MS m/z 451
1
(M+); H NMR (CDCl3) 7.79 (d, 4/7 H, J ) 10.4 Hz), 7.71 (s,
4/7 H), 7.71 (s, 3/7 H), 7.28 (d, 3/7 H, J ) 10.7 Hz), 7.20, 5.62
(each br s, each 1 H), 5.90 (s, 3/7 H), 5.85 (d, 4/7 H, J ) 3.3
Hz), 5.76 (dd, 4/7 H, J ) 10.4, 2.2 Hz), 5.72 (dd, 3/7 H, J )
10.7, 2.2 Hz), 4.84-4.82 (m, 10/7 H), 4.74 (dd, 4/7 H, J ) 3.3,
6.0 Hz), 4.47 (m, 3/7 H), 4.42 (m, 4/7 H), 3.93 (dd, 4/7 H, J )
2.8, 11.5 Hz), 3.91 (dd, 3/7 H, J ) 2.4, 11.5 Hz), 3.82 (dd, 4/7
H, J ) 2.2, 11.5 Hz), 3.80 (dd, 3/7 H, J ) 2.8, 11.5 Hz), 2.54
(d, 4/7 H, J ) 2.2 Hz), 2.52 (d, 3/7 H, J ) 2.2 Hz), 1.61, 1.36
(each s, each 12/7 H), 1.61, 1.38 (each s, each 9/7 H), 0.88 (s,
36/7 H), 0.84 (s, 27/7 H), 0.09, 0.08 (each s, each 12/7 H), 0.07,
0.04 (each s, each 9/7 H). Used immediately in the next stage.
-0.03, -0.15, -0.16 (each s, total 18 H). Anal. Calcd for C29H55
-
FN4O5Si3: C, 54.17; H, 8.62; N, 8.71. Found: C, 54.15; H, 8.59;
N, 8.38.
6-Am in o-7-flu or o-1-(2,3,5-tr i-O-ter t-bu tyld im eth ylsilyl-
â-D-r ibofu r a n osyl)im id a zo[4,5-c]p yr id in -4(5H)-on e (50).
A solution of 48 (2.20 g, 3.43 mmol) in a mixture of EtOH (50
mL) and 5% aqueous Na2CO3 (50 mL) was heated for 30 min
at 60 °C. The solvent was removed, and the residue was
partitioned between CHCl3 and H2O. The aqueous layer was
extracted with CHCl3 three times. The combined organic layers
were dried (Na2SO4) and concentrated in vacuo. The residue
was purified by a silica gel column (4.0 × 12 cm), eluted with
0-10% EtOH in CHCl3, to give 50 (1.89 g, 86% as a yellow
foam): FAB-MS m/z 643 (MH+); UV λmax (MeOH) 273 nm; 1H
NMR (CDCl3) 11.15 (br s, 1 H), 8.00 (s, 1 H), 6.01 (d, 1 H, J )
6.6 Hz), 4.86 (br s, 2 H), 4.27 (dd, 1 H, J ) 6.6, 4.7 Hz), 4.19
(dd, 1 H, J ) 4.7, 1.6 Hz), 4.09 (ddd, 1 H, J ) 1.6, 2.8, 2.1 Hz),
3.91 (dd, 1 H, J ) 2.8, 11.5 Hz), 3.78 (dd, 1 H, J ) 2.1, 11.5
Hz), 0.96, 0.93, 0.80 (each s, each 9 H), 0.16, 0.15, 0.11, 0.10,
-0.06, -0.34 (each s, each 3 H). Anal. Calcd for C29H55FN4O5-
Si3: C, 54.17; H, 8.62; N, 8.71. Found: C, 54.09; H, 8.65; N,
8.59.
6-Am in o-7-flu or o-1-â-D-r ib ofu r a n osylim id a zo[4,5-c]-
p yr id in -4(5H)-on e (3-Dea za -3-flu or ogu a n osin e, 52). A
THF solution of TBAF (1 M, 11.0 mL, 11.0 mmol) was added
to a solution of 50 (1.77 g, 2.76 mmol) in THF (40 mL) at 0 °C.
The mixture was stirred for 1 h at room temperature and
neutralized with AcOH. The solvent was removed in vacuo,
and the residue was dissolved in H2O (100 mL), which was
washed with CHCl3 three times. The aqueous layer was
concentrated in vacuo, and the residue was coevaporated with
EtOH three times. The resulting solid was crystallized from
MeOH to give 52 (575 mg, 70%): mp >300 °C; FAB-MS m/z
301 (MH+); UV λmax (H2O) 274 nm (ꢀ 8300), 309 nm (ꢀ 7100);
5-(1-F lu or op r op yn yl)-1-(5-O-ter t-b u t yld im et h ylsilyl-
2,3-O-isop r op ylid en e-â-D-r ibofu r a n osyl)im id a zole-4-ca r -
boxa m id e (40). A solution of 38 (2.56 g, 2.84 mmol) in dry
CH2Cl2 was added dropwise over 1 h to a solution of DAST
(1.50 mL, 11.4 mmol) in dry CH2Cl2 at -15 °C. The mixture
was allowed to warm at room temperature and stirred for 30
min. To this mixture was added aqueous NaHCO3 (1 M, 40
mL), and the whole was stirred for 20 min. The mixture was
extracted with CHCl3 (100 mL), and the organic layer was
washed with H2O, followed by brine. The organic layer was
dried (Na2SO4) and concentrated in vacuo, and the residue was
purified by a silica gel column (4.0 × 10 cm), eluted with
hexane/AcOEt (3:1-1:1), to give 40 (2.3 g, 89%, crystallized
from hexane-AcOEt): mp 134-135 °C and 151-152 °C; FAB-
1
MS m/z 454 (MH+); H NMR (CDCl3) 8.07 (d, 3/5 H, J ) 0.9
Hz), 8.03 (d, 2/5 H, J ) 1.3 Hz), 7.53 (dd, 3/5 H, J ) 44.2, 2.3
Hz), 7.50 (dd, 2/5 H, J ) 44.2, 2.3 Hz), 7.12, 5.48 (each br s,
each 1 H), 6.45 (d, 3/5 H, J ) 2.8 Hz), 6.36 (d, 2/5 H, J ) 2.6
Hz), 4.94 (dd, 2/5 H, J ) 5.9, 3.8 Hz), 4.89-4.86 (m, 5/5 H),
4.66 (dd, 3/5 H, J ) 2.8, 5.7 Hz), 4.34 (m, 3/5 H), 4.29 (m, 2/5
H), 4.02 (dd, 3/5 H, J ) 2.1, 11.6 Hz), 3.97 (dd, 2/5 H, J ) 2.2,
11.6 Hz), 3.87 (dd, 3/5 H, J ) 2.5, 11.6 Hz), 3.86 (dd, 2/5 H, J
) 2.3, 11.6 Hz), 2.91 (d, 2/5 H, J ) 2.3 Hz), 2.90 (d, 3/5 H, J )
2.3 Hz), 1.61, 1.36 (each s, each 9/5 H), 1.60, 1.36 (each s, each
6/5 H), 0.95 (s, 27/5 H), 0.94 (s, 18/5 H), 0.14, 0.12 (each s,
each 3 H). Anal. Calcd for C21H32FN3O5Si: C, 55.61; H, 7.11;
N, 9.26. Found: C, 55.71; H, 7.13; N, 9.24.
λmax (0.5 N NaOH) 288 nm (ꢀ 9500). Anal. Calcd for C11H13
-
FN4O5‚1/4H2O: C, 43.35; H, 4.47; N, 18.38. Found: C, 43.63;
H, 4.32; N, 18.08.
7-F lu or o-1-â-D-r ib ofu r a n osylim id a zo[4,5-c]p yr id in -
4(5H)-on e (3-Dea za -3-flu or oin osin e, 46). An aqueous TFA
solution (75%, 10 mL) containing 45 (290 mg, 0.66 mmol) was
stirred for 1 h at room temperature. The solvent was removed
in vacuo and coevaporated with EtOH three times. The residue
was purified by a silica gel column (2.2 × 6 cm), eluted with
0-30% EtOH in CHCl3, to give 46 (196 mg, quant crystallized
from MeOH-H2O): mp >300 °C; FAB-MS m/z 286 (MH+); UV
λmax (H2O) 260 nm (ꢀ 9000); λmax (0.5 N HCl) 274 nm (ꢀ 8800);
5-Cya n oflu or om eth yl-1-(2,3,5-tr i-O-ter t-bu tyld im eth yl-
silyl-â-D-r ibofu r an osyl)im idazole-4-car bon itr ile (49). Phos-
phorus oxychloride (1.0 mL, 10.7 mmol) was added dropwise
to a mixture of 48 (2.3 g, 3.58 mmol) and Et3N (5.0 mL, 35.8
mmol) in dry CH2Cl2 at 0 °C, and the whole was stirred for 1
h at that temperature. The reaction mixture was poured into
ice-cold saturated aqueous NaHCO3 (80 mL) and extracted
with CHCl3. The organic layer was washed with H2O, followed
by brine. The organic layer was dried (Na2SO4) and concen-
λmax (0.5 N NaOH) 267 nm (ꢀ 9000). Anal. Calcd for C11H12
-

3-Deaza-3-halopurine Ribonucleosides
J . Org. Chem., Vol. 64, No. 19, 1999 7171
trated in vacuo, and the residue was purified by a silica gel
column (3.5 × 10 cm), eluted with hexane/AcOEt (20:1-6:1),
to give 49 (2.01 g, 90% as a pale yellow oil): FAB-MS m/z 625
more isoamyl nitrite (2.73 mL, 20.3 mmol) was added, and the
whole was stirred for an additional 4 h. The reaction mixture
was concentrated in vacuo, and the residue was purified by a
silica gel column (3.0 × 8 cm), eluted with 0-8% EtOH in
CHCl3, to give 57 (529 mg, 59% as a yellow foam): FAB-MS
m/z 669 (MH+); UV λmax (MeOH) 259 nm; λmax (H+) 270 nm;
1H NMR (CDCl3) 8.68 (d, 1 H, J ) 1.1 Hz), 8.48 (s, 1 H), 7.78
(br s, 1 H), 6.17 (d, 1 H, J ) 6.0 Hz), 4.33 (dd, 1 H, J ) 6.0, 4.7
Hz), 4.22 (dd, 1 H, J ) 4.7, 4.2 Hz), 4.14 (ddd, 1 H, J ) 4.2,
2.7, 2.2 Hz), 3.97 (dd, 1 H, J ) 2.7, 11.5 Hz), 3.82 (dd, 1 H, J
) 2.2, 11.5 Hz), 2.30 (s, 3 H), 0.98, 0.94, 0.75 (each s, each 9
H), 0.18, 0.17, 0.11, 0.11, -0.06, -0.37 (each s, each 3 H). Anal.
Calcd for C31H57FN4O5Si3: C, 55.65; H, 8.59; N, 8.37. Found:
C, 55.55; H, 8.43; N, 8.27.
1
(MH+); UV λmax (MeOH) 266 nm; H NMR (CDCl3) 7.95 (s, 1
H), 6.81 (d, 3/5 H, J ) 44.5 Hz), 6.63 (d, 2/5 H, J ) 44.5 Hz),
5.75 (d, 3/5 H, J ) 6.6 Hz), 5.73 (d, 2/5 H, J ) 7.2 Hz), 4.23-
4.08 (m, 3H), 3.97 (dd, 3/5 H, J ) 2.5, 11.9 Hz), 3.89 (dd, 2/5
H, J ) 2.7, 11.5 Hz), 3.86-3.79 (m, 1 H), 0.97, 0.96, 0.94, 0.94,
0.82, 0.82 (each s, total 27 H), 0.18, 0.18, 0.17, 0.17, 0.12, 0.11,
0.11, 0.10, -0.02, -0.04, -0.25, -0.32 (each s, total 18 H).
Used immediately in the next stage.
4,6-Dia m in o-7-flu or o-1-(2,3,5-tr i-O-ter t-bu tyld im eth yl-
silyl-â-D-r ibofu r a n osyl)im id a zo[4,5-c]p yr id in e (51). A so-
lution of 49 (1.93 g, 3.10 mmol) in ethanolic ammonia
(saturated at 0 °C, 50 mL) was heated for 20 h at 120 °C in a
steel container. The reaction mixture was concentrated in
vacuo, and the residue was purified by a silica gel column (3.5
× 8 cm), eluted with 0-4% EtOH in CHCl3, to give 51 (1.58 g,
79% as a yellow foam): FAB-MS m/z 642 (MH+); UV λmax
6-Am in o-7-flu or o-1-â-D-r ib ofu r a n osylim id a zo[4,5-c]-
p yr id in e (59). A MeOH solution of NaOMe (1 M, 2.8 mL, 2.8
mmol) was added to a solution of 57 (370 mg, 0.55 mmol) in
dry MeOH (20 mL). After being stirred for 24 h at room
temperature, more 1 M MaOMe (2.8 mL, 2.8 mmol) was added,
and the mixture was stirred for an additional 48 h at that
temperature. The reaction mixture was neutralized with 1 M
aqueous HCl, and the solution was concentrated in vacuo. The
residue was partitioned between CHCl3 and H2O, and the
aqueous layer was extracted three times with CHCl3. The
combined organic layers were dried (Na2SO4) and concentrated
in vacuo to give crude 58. A THF solution of TBAF (1 M, 1.75
mL, 1.75 mmol) was added to a solution of 58 in THF (15 mL)
at 0 °C, and the mixture was stirred for 90 min at room
temperature. The reaction mixture was concentrated in vacuo,
and the residue was purified on a silica gel column (2.5 × 10
cm), eluted with 0-25% EtOH in CHCl3, to give 59 (94 mg,
60%, crystallized from MeOH-H2O): mp 228-231 °C dec;
FAB-MS m/z 285 (MH+); UV λmax (H2O) 260 nm (ꢀ 6500), 292
nm (ꢀ 3200); λmax (0.5 N HCl) 265 nm (ꢀ 5900), 271 nm (ꢀ 5900),
319 nm (ꢀ 4700); λmax (0.5 N NaOH) 260 nm (ꢀ 6800), 293 nm
1
(MeOH) 283 nm; H NMR (CDCl3) 7.95 (s, 1 H), 6.02 (d, 1 H,
J ) 6.6 Hz), 4.81, 4.12 (each br s, each 2 H), 4.29 (dd, 1 H, J
) 6.6, 4.9 Hz), 4.19 (dd, 1 H, J ) 4.9, 4.4 Hz), 4.09 (ddd, 1 H,
J ) 4.4, 3.5, 2.5 Hz), 3.90 (dd, 1 H, J ) 3.5, 11.2 Hz), 3.78 (dd,
1 H, J ) 2.5, 11.2 Hz), 0.96, 0.94, 0.78 (each s, each 9 H), 0.15,
0.14, 0.11, 0.10, -0.07, -0.37 (each s, each 3 H). Anal. Calcd
for C29H56FN5O4Si3: C, 54.25; H, 8.79; N, 10.91. Found: C,
54.19; H, 8.50; N, 10.61.
4,6-Dia m in o-7-flu or o-1-â-D-r ibofu r a n osylim id a zo[4,5-
c]p yr id in e (53). A mixture of 51 (90 mg, 0.14 mmol) and
NH4F (103 mg, 2.8 mmol) in MeOH (5 mL) was heated
overnight under reflux. The solvent was removed in vacuo, and
the residue was purified by a silica gel column (3.0 × 4 cm),
eluted with 10-30% MeOH in CHCl3, to give 53 (31 mg, 74%,
crystallized from EtOH-H2O): mp 215-216 °C; FAB-MS m/z
300 (MH+); UV λmax (H2O) 282 nm (ꢀ 9300); λmax (0.5 N HCl)
276 nm (ꢀ 9300), 324 nm (ꢀ 8600); λmax (0.5 N NaOH) 280 nm
1
(ꢀ 3700); H NMR (DMSO-d6) 8.36 (s, 1 H), 8.23 (d, 1 H, J )
1
(ꢀ 9900); H NMR (DMSO-d6) 8.04 (s, 1 H), 5.81 (d, 1 H, J )
1.5 Hz), 5.88 (d, 1 H, J ) 5.8 Hz), 5.80 (br s, 2 H), 5.50 (d, 1 H,
J ) 6.1 Hz), 5.21 (d, 1 H, J ) 4.8 Hz), 5.06 (t, 1 H, J ) 5.3
Hz), 4.34 (ddd, 1 H, J ) 5.7, 6.1, 5.8 Hz), 4.08 (ddd, 1 H, J )
4.9, 4.8, 3.8 Hz), 3.93 (dt, 1 H, J ) 3.8, 4.1 Hz), 3.63 (ddd, 1 H,
J ) 4.1, 11.9, 5.3 Hz), 3.55 (ddd, 1 H, J ) 4.1, 11.9, 5.3 Hz).
13C NMR (DMSO-d6) 144.2 (J ) 9.2 Hz), 142.0, 136.7, 134.7
(J ) 5.5 Hz), 130.8 (J ) 250.0 Hz), 127.5 (J ) 5.6 Hz), 89.3 (J
) 3.6 Hz), 85.4, 74.5, 70.0, 61.2. Anal. Calcd for C11H13FN4O4:
C, 46.48; H, 4.61; N, 19.71. Found: C, 46.38; H, 4.78; N, 19.39.
4-Am in o-7-flu or o-6-t r iflu or oa cet a m id o-1-(2,3,5-t r i-O-
ter t-bu tyld im eth ylsilyl-â-D-r ibofu r a n osyl)im id a zo[4,5-c]-
p yr id in e (60). Trifluoroacetic anhydride (0.44 mL, 3.10 mmol)
was added to a solution of 51 (1.35 g, 2.10 mmol) in dry CH2-
Cl2 (35 mL) containing Et3N (0.43 mL, 3.10 mmol) in an ice
bath. After being stirred for 1.5 h at 0 °C, the reaction was
quenched by addition of EtOH (5 mL). The solvent was
removed in vacuo, and the residue was dissolved in AcOEt (70
mL), which was washed successively with H2O, saturated
aqueous NaHCO3, and saturated brine. The organic layer was
dried (Na2SO4) and concentrated in vacuo. Methanolic am-
monia (saturated at 0 °C, 30 mL) was added to the residue,
and the mixture was kept for 45 min at room temperature.
The solvent was concentrated in vacuo, and the residue was
purified on a silica gel column (3.5 × 6 cm), eluted with hexane/
AcOEt (3:1-3:2), to give 60 (1.50 g, 97% as a white foam):
FAB-MS m/z 738 (MH+); UV λmax (MeOH) 275 nm; λmax (H+)
6.0 Hz), 5.75, 5.19 (each br s, each 2 H), 5.45 (d, 1 H, J ) 6.0
Hz), 5.18 (d, 1 H, J ) 5.0 Hz), 5.02 (dd, 1 H, J ) 5.5, 4.9 Hz),
4.29 (dt, 1 H, J ) 6.0, 5.5 Hz), 4.05 (ddd, 1 H, J ) 5.5, 5.0, 3.8
Hz), 3.89 (ddd, 1 H, J ) 3.8, 4.1, 4.4 Hz), 3.61 (ddd, 1 H, J )
4.1, 12.1, 5.5 Hz), 3.55 (ddd, 1 H, J ) 4.4, 12.1, 4.9 Hz). 13C
NMR (DMSO-d6) 146.5, 141.8 (J ) 11.0 Hz), 138.0 (J ) 188.0
Hz), 128.8 (J ) 8.0 Hz), 126.3, 124.5, 89.2, 85.3, 74.6, 70.2,
61.4. Anal. Calcd for C11H14FN5O4: C, 44.15; H, 4.72; N, 23.40.
Found: C, 44.10; H, 4.67; N, 23.14.
6-Acet a m id o-4-a m in o-7-flu or o-1-(2,3,5-t r i-O-ter t-b u -
t yld im et h ylsilyl-â-D-r ib ofu r a n osyl)im id a zo[4,5-c]p yr i-
d in e (54). A mixture of 51 (1.08 g, 1.68 mmol), Ac2O (0.79
mL, 8.4 mmol), and DMAP (42 mg, 0.34 mmol) in dry pyridine
(40 mL) was stirred for 20 h at room temperature. The reaction
was quenched by addition of EtOH (5 mL), and the solvent
was removed in vacuo. The residue was dissolved in CHCl3,
which was washed with H2O, followed by brine. The organic
layer was dried (Na2SO4) and concentrated in vacuo, and
methanolic ammonia (saturated at 0 °C, 40 mL) was added to
the residue. After being kept for 20 min at room temperature,
the solvent was removed in vacuo, and the residue was
coevaporated several times with toluene. The residue was
purified by a silica gel column (3.5 × 10 cm), eluted with 0-8%
EtOH in CHCl3, to give 54 (929 mg, 81% as a pale yellow
foam): FAB-MS m/z 684 (MH+); UV λmax (MeOH) 279 nm; 1H
NMR (CDCl3) 8.23 (s, 1 H), 7.42 (br s, 1 H), 6.10 (d, 1 H, J )
6.0 Hz), 5.08 (br s, 2 H), 4.32 (dd, 1 H, J ) 6.0, 4.7 Hz), 4.21
(dd, 1 H, J ) 4.7, 4.4 Hz), 4.11 (ddd, 1 H, J ) 4.4, 3.0, 2.2 Hz),
3.94 (dd, 1 H, J ) 3.0, 11.5 Hz), 3.80 (dd, 1 H, J ) 2.2, 11.5
Hz), 2.24 (s, 3 H), 0.98, 0.94, 0.77 (each s, each 9 H), 0.17,
0.16, 0.11, 0.10, -0.06, -0.37 (each s, each 3 H). Anal. Calcd
for C31H58FN5O5Si3: C, 54.43; H, 8.55; N, 10.24. Found: C,
54.51; H, 8.36; N, 10.06.
1
271 nm; H NMR (CDCl3) 8.28 (s, 1 H), 8.09 (br s, 1 H), 6.09
(d, 1 H, J ) 5.5 Hz), 5.12 (br s, 2 H), 4.33 (dd, 1 H, J ) 5.5, 4.4
Hz), 4.21 (t, 1 H, J ) 4.4 Hz), 4.12 (ddd, 1 H, J ) 4.4, 2.5, 2.2
Hz), 3.94 (dd, 1 H, J ) 2.5, 11.0 Hz), 3.80 (dd, 1 H, J ) 2.2,
11.0 Hz), 0.97, 0.94, 0.77 (each s, each 9 H), 0.17, 0.15, 0.11,
0.11, -0.06, -0.37 (each s, each 3 H). Anal. Calcd for
C
31H55F4N5O5Si3: C, 50.45; H, 7.51; N, 9.49. Found: C, 50.39;
H, 7.29; N, 9.36.
6-Aceta m id o-7-flu or o-1-(2,3,5-tr i-O-ter t-bu tyld im eth yl-
silyl-â-D-r ibofu r an osyl)im idazo[4,5-c]pyr idin e (57). A THF
solution (50 mL) of 54 (920 mg, 1.35 mmol) was heated at 60
°C, and isoamyl nitrite (2.73 mL, 20.3 mmol) was added to
the mixture. After being stirred for 4 h at that temperature,
4-Acet a m id o-6-a m in o-7-flu or o-1-(2,3,5-t r i-O-ter t-b u -
t yld im et h ylsilyl-â-D-r ib ofu r a n osyl)im id a zo[4,5-c]p yr i-
d in e (62). Acetyl chloride (0.17 mL, 2.39 mmol) was added to
a solution of 60 (1.60 g, 2.17 mmol) in dry pyridine (40 mL),
and the whole was stirred for 50 min at room temperature.

7172 J . Org. Chem., Vol. 64, No. 19, 1999
Minakawa et al.
The reaction was quenched by addition of EtOH (5 mL), and
the solvent was removed in vacuo. The residue was dissolved
in AcOEt (70 mL), which was washed with H2O, followed by
saturated brine. The organic layer was dried (Na2SO4) and
concentrated in vacuo, and the residue was coevaporated with
toluene. Methanolic ammonia (saturated at 0 °C, 40 mL) was
added to the residue, and the mixture was kept for 24 h at
room temperature. The solvent was concentrated in vacuo, and
the residue was purified by a silica gel column (3.5 × 10 cm),
eluted with hexane/AcOEt (2:1-1:4), to give 62 (1.19 g, 80%,
crystallized from AcOEt-hexane): mp 174-175 °C; FAB-MS
m/z 684 (MH+); UV λmax (MeOH) 272, 313 nm; λmax (H+) 276,
H). Anal. Calcd for C31H57FN4O5Si3: C, 55.65; H, 8.59; N, 8.37.
Found: C, 55.39; H, 8.48; N, 8.26.
4-Am in o-7-flu or o-1-â-D-r ib ofu r a n osylim id a zo[4,5-c]-
p yr id in e (3-Dea za -3-flu or oa d en osin e, 64). A solution of 63
(400 mg, 0.60 mmol) in methanolic ammonia (saturated at 0
°C, 35 mL) was heated for 20 h at 120 °C in a steel container.
The reaction mixture was concentrated to dryness in vacuo to
give a crude deacetylated product. A THF solution of TBAF
(1 M, 2.0 mL, 2.0 mmol) was added to a solution of the
deacetylated product in THF (15 mL) at 0 °C, and the mixture
was stirred for 90 min at room temperature. The solvent was
removed in vacuo, and the residue was purified by a silica gel
column (2.5 × 10 cm), eluted with 0-25% EtOH in CHCl3, to
give 64 (97 mg, 57%, crystallized from MeOH-H2O): mp 217-
220 °C dec; FAB-MS m/z 285 (MH+); UV λmax (MeOH) 267 nm
(ꢀ 8000); λmax (0.5 N HCl) 265 nm (ꢀ 8600); λmax (0.5 N NaOH)
267 nm (ꢀ 8500). Anal. Calcd for C11H13FN4O4‚2/5MeOH: C,
46.09; H, 4.95; N, 18.86. Found: C, 45.74; H, 4.67; N, 18.84.
1
283, 338 nm; H NMR (CDCl3) 8.19 (br s, 1 H), 8.07 (s, 1 H),
6.04 (d, 1 H, J ) 6.6 Hz), 4.38 (br s, 2 H), 4.29 (dd, 1 H, J )
6.6, 5.5 Hz), 4.21 (dd, 1 H, J ) 5.5, 4.4 Hz), 4.11 (ddd, 1 H, J
) 4.4, 3.3, 2.2 Hz), 3.91 (dd, 1 H, J ) 3.3, 12.1 Hz), 3.79 (dd,
1 H, J ) 2.2, 12.1 Hz), 2.46 (br s, 3 H), 0.97, 0.94, 0.77 (each
s, each 9 H), 0.16, 0.15, 0.11, 0.11, -0.07, -0.40 (each s, each
3 H). Anal. Calcd for C31H58FN5O5Si3: C, 54.43; H, 8.55; N,
10.24. Found: C, 54.15; H, 8.52; N, 10.29.
Ack n ow led gm en t. This work was supported in part
by a Grant-in Aid for Encouragement of Young Scien-
tists from the Ministry of Education, Science, Sports,
and Culture of J apan.
4-Aceta m id o-7-flu or o-1-(2,3,5-tr i-O-ter t-bu tyld im eth yl-
silyl-â-D-r ibofu r a n osyl)im id a zo[4,5-c]p yr id in e (63). In the
same manner as described for 57, 62 was treated with isoamyl
nitrite in dry THF giving 63 (748 mg, 55% as a yellow foam):
FAB-MS m/z 669 (MH+); UV λmax (MeOH) 266 nm; λmax (H+)
1
Su p p or tin g In for m a tion Ava ila ble: Complete H NMR
1
289 nm; H NMR (CDCl3) 8.37 (s, 1 H), 8.37 (br s, 1 H), 8.05
spectral information for compounds 6-8, 15-17, 25, 26, 46,
52, and 64. Complete 13C NMR spectral information for
compounds 46, 52, and 64. This material is available free of
charge via the Internet at http://pubs.acs.org.
(d, 1 H, J ) 2.2 Hz), 6.13 (d, 1 H, J ) 5.5 Hz), 4.32 (t, 1 H, J
) 5.5 Hz), 4.22 (dd, 1 H, J ) 5.5, 4.4 Hz), 4.14 (ddd, 1 H, J )
4.4, 3.3, 2.2 Hz), 3.95 (dd, 1 H, J ) 3.3, 11.0 Hz), 3.81 (dd, 1
H, J ) 2.2, 11.0 Hz), 2.49 (s, 3 H), 0.98, 0.94, 0.76 (each s,
each 9 H), 0.18, 0.17, 0.12, 0.11, -0.07, -0.45 (each s, each 3
J O990638X