Nucleic acid related compounds. 127. Selective N-deacylation of N,O-peracylated nucleosides in superheated methanol

Article abstract of DOI:10.1021/jo051256w

Solutions of peracylated adenosine, cytidine, and related nucleoside derivatives undergo selective N-deacylation upon heating at elevated temperatures (oil bath ≥ 105 °C) in methanol. An increase in the bulk of the N-acyl group has little effect on the rate of N-deacylation but increases the N/O selectivity ratio. Extended heating is required for N-deacylation with arylcarboxylic acid derivatives. Contamination with acidic or basic reagent residues is avoided.

Full text of DOI:10.1021/jo051256w

SCHEME 1. N,O-Acylation and N-Deacylation of
Adenosine
Nucleic Acid Related Compounds. 127.
Selective N-Deacylation of N,O-Peracylated
Nucleosides in Superheated Methanol1
Ireneusz Nowak, Martin Conda-Sheridan, and
Morris J. Robins*
Department of Chemistry and Biochemistry,
Brigham Young University, Provo, Utah 84602-5700
morris_robins@byu.edu
Received June 18, 2005
Solutions of peracylated adenosine, cytidine, and related
nucleoside derivatives undergo selective N-deacylation upon
heating at elevated temperatures (oil bath g 105 °C) in
methanol. An increase in the bulk of the N-acyl group has
little effect on the rate of N-deacylation but increases the
N/O selectivity ratio. Extended heating is required for
N-deacylation with arylcarboxylic acid derivatives. Contami-
nation with acidic or basic reagent residues is avoided.
O-acetyl cleavage, upon heating a solution of 6-N-2′,3′,5′-
tri-O-tetraacetyladenosine in MeOH in a sealed vessel
at 105 °C. Holy reported that treatment of 4-N-2′,3′,5′-
tri-O-tetraacetylcytidine with 80% aqueous acetic acid at
reflux resulted in rearrangement of the acetyl group from
N4 to N3 of the cytosine ring. Such treatment of 6-N-
2′,3′,5′-tri-O-tetraacetyladenosine effected removal of the
N-acetyl group without its rearrangement to N1.8 Heat-
ing a methanolic solution of 4-N-2′,3′,5′-tri-O-tetraace-
tylcytidine at reflux for 40 h was required to solvolyze
the N-acetyl group.9 Cleavage of N-acyl groups with zinc
bromide and alcohols has been noted.10 We now report
the convenient and selective removal of N-acyl groups
from peracylated nucleosides with superheated methanol.
Adenosine was acylated with acetic, propionic, isobu-
tyric, and pivalic anhydrides in pyridine. Typically,
acetylation of adenosine at ambient temperature takes
several hours and produces a mixture of O-tri- and N,O-
tetraacetyl derivatives (∼3:1), which have similar chro-
matographic mobilities.11 We effected complete N-acyla-
tion by heating the reaction mixtures overnight at 50 or
80 °C, which gave both tetra-, 1, and pentaacylated, 2,
products (Scheme 1). (Because the pentaacylated com-
pounds 2 are converted quickly to their tetraacyl ana-
logues 1 in hot MeOH, we designate compounds 1 as
starting materials even when mixtures of both 1 and 2
are present.) Treatment of 1 (1-2 mmol) with super-
heated MeOH was examined (Table 1).
Selective protection and deprotection methodologies
are keystones of modern organic synthesis.2 Numerous
reports on manipulations of acylated nucleosides dem-
onstrate their removal under conditions that are orthogo-
nal to a variety of other protecting groups. Acyl groups
have been removed by both chemical3 and enzymatic4
procedures. Selective N-acylation of nucleobases has been
employed for temporary protection,5 especially during
glycosylation coupling reactions.6 However, N-acylation
is often encountered as an undesired side reaction during
protection of the sugar portion of nucleosides that have
an amino group on the nucleobase. Lowering the reaction
temperature can attenuate N-acylation with the less
reactive acid anhydrides but usually not with acid
chlorides.
Various target structures require reactions with the
free 6-amino group of adenosine. Current projects in our
laboratory involve diazotization of an amino group7 and
synthetic applications of adenosine N-oxides. We ob-
served rapid loss of the N-acetyl group, relative to
(1) Paper 126 is: Janeba, Z.; Balzarini, J.; Andrei, G.; Snoeck, R.;
De Clercq, E.; Robins, M. J. J. Med. Chem. 2005, 48, 4690-4696.
(2) Greene, T. W.; Wuts, P. G. M. Protective Groups in Organic
Synthesis; Wiley: New York, 1999.
(7) Liu, J.; Robins, M. J. Org. Lett. 2005, 7, 1149-1151.
(8) Holy, A. Collect. Czech. Chem. Commun. 1979, 44, 1819-1827.
(9) Beranek, J.; Pitha, J. Collect. Czech. Chem. Commun. 1964, 29,
625-633.
(3) (a) Asakura, J.-i. Nucleosides Nucleotides 1993, 12, 701-711 and
references therein. (b) Ishido, Y.; Sakairi, N.; Okazaki, K.; Nakazaki,
N. J. Chem. Soc., Perkin Trans. 1 1980, 563-573.
(10) Kierzek, R.; Ito, H.; Bhatt, R.; Itakura, K. Tetrahedron Lett.
1981, 22, 3761-3764.
(4) Singh, H. K.; Cote, G. L.; Sikorski, R. S. Tetrahedron Lett. 1993,
34, 5201-5204.
(11) Also indicated by the moderate yields and necessity for further
purifications reported by: (a) Reist, E. J.; Calkins, D. F.; Fisher, L.
V.; Goodman, L. J. Org. Chem. 1967, 33, 1600-1603. (b) Nair, V.;
Chamberlain, S. D. Synthesis 1984, 401-403.
(5) Igolen, J.; Morin, C. J. Org. Chem. 1980, 45, 4802-4804.
(6) Vorbru¨ggen, H.; Ruh-Pohlenz, C. Org. React. 2000, 55, 1-630.
10.1021/jo051256w CCC: $30.25 © 2005 American Chemical Society
Published on Web 07/26/2005
J. Org. Chem. 2005, 70, 7455-7458
7455

TABLE 1. N-Deacylation of 1/2 Mixtures in
Superheated Methanola
TABLE 2. N-Deacylation of 4/5 to Give 6 in Superheated
MeOHa
entryb
timec
R
R′
product
yieldd
entryb
timec
R
R′
R′′
Me
MePh
Ph
Me
Me
Y
Z
yieldd
1 (a)
2 (a)
1
12
1
Me
Me
3a
3a
3b
3b
3b
3c
3c
3c
3a
3a
3c
3c
3c
3d
3e
3e
3f
84 (92)
(45)
(87)
98
(80)
(56)
98
1 (a)
2 (b)
3 (c)
4 (d)
5 (e)
1.5
14
12
1.5
1.5
Ac
Tol
Bz
TBS
Ac
Ac
H
H
H
OTs
OTs
H
86 (93)
72
58
85
76 (90)
Me
Et
Me
Tol
Bz
Ace
Ac
H
3 (b)
Et
H
4 (b)
1.5
12
1
Et
Et
H
5 (b)
Et
Et
OAc
6 (c)
iPr
iPr
a Reactions were performed by the general procedure. b Starting
material in parentheses. c Hours. d Percent isolated or (% by 1H
NMR). e Product 6d (R′ ) H) had lost both the 6-N- and 3′-O-acetyl
groups.
7 (c)
3
iPr
iPr
8 (c)
12
1
iPr
iPr
(92)
(77)
70
9 (d)
10 (d)
11 (e)
12 (f)
13 (f)
14 (g)
15 (h)
16 (h)
17 (i)
Me
Me
iPr
t-Bu
t-Bu
t-Bu
Ph
3
3
93
1
iPr
7
SCHEME 2. N-Deacylation of Some Adenosine
Derivatives
12
14
14
14e
14
iPr
Ph
Ph
MePh
MePh
ClPh
88
Ph
84
MePh
MePh
ClPh
79
95
84
a Reactions were performed by the general procedure. b Starting
material in parentheses. c Hours. d Percent isolated or (% by 1H
NMR). e Temperature ) 120 °C.
As expected, the less sterically hindered N-(acetyl and
propionyl) groups underwent the most rapid methanoly-
sis (entries 1 and 3), followed by the N-(isobutyryl and
pivalyl) amides (entries 6 and 9). Prolonged heating must
be avoided, especially with N,O-acetates (entries 2 and
10), because the ester groups on the sugar moiety
undergo slow methanolysis. The more hindered N,O-
propionyl analogue 1b, and especially the N,O-isobutyryl
derivative 1c, underwent highly selective N-deacylation
in 1.5 and 3 h, respectively (entries 4 and 7). Hindered
rotation of the bulky isobutyryl groups was evidenced by
the appearance of twinned sets of signals (1:1) in the 1H
NMR spectrum of 1c. More steric interference would
occur with neighboring 2′- and 3′-O-pivalyl groups, and
complete O-pivalylation of adenosine was not observed.
The more bulky N-pivalyl group of 6-N-pivalyl-2′,3′,5′-
tri-O-isobutyryladenosine (1e) (prepared by pivalylation
of 3c) underwent selective methanolysis relative to the
less sterically hindered O-isobutyryl esters.
Loss of the N-pivalyl group was complete within 3 h
(entries 10 and 11). The lower yield of the desired product
from methanolysis of 1d (entry 10) results from partial
solvolysis of the O-acetyl groups on the sugar moiety.
Benzoylation of 3c with benzoyl chloride/pyridine gave
the 6-N,N-dibenzoyl derivative 2f. Methanolysis of 2f at
105 °C proceeded slowly and required 12 h for completion
(entries 12 and 13). The N-debenzoylated products 3d
and 3f were obtained in lower yields (84%) from the
perbenzoylated derivatives 2g and 2i (entries 14 and 17)
because of competing O-debenzoylation. Methanolysis of
the p-toluylated 2h gave 3e in 79% yield (entry 15). The
incomplete N-detoluylation was possibly due to the
inductive donating effect of the p-methyl group. Elevation
of the bath temperature to 120 °C resulted in deprotec-
tion of 2h to give 3e in 95% yield (entry 16).
relative to prior acylations that require extended incuba-
tion periods and proceed with varied N versus O selec-
tivities.
The preference for N versus O deprotection in super-
heated MeOH is reversed relative to that noted by
Khorana and co-workers for selective O-debenzoylation
of perbezoylated nucleosides with NaOH or NaOMe.12
Jones’ “transient protection” protocol also provides N-acyl
derivatives by complete trimethylsilylation followed by
acylation and selective removal of the O-silyl groups.13
Thus, the procedures of Khorana12 and Jones13 are
complementary to our deprotections in superheated
methanol.
We reasoned that these mild solvolysis conditions with
methanol should be compatible with 2′-deoxyadenosine
derivatives, which are highly susceptible to both acid-
catalyzed glycosyl-bond hydrolysis and base-promoted
eliminations.12 Indeed, the glycosyl bond in 4a (Scheme
2) was stable at 105 °C for 2 h, and essentially pure 6a
was obtained (Table 2). However, a lower yield of 6b
(72%) was obtained from 4b due to the loss of 6-N-
toluyladenine. A tosyl ester at O2′ was stable in super-
heated methanol (entries 3 and 4), but compound 6c
(58%) was obtained in a lower yield. An explanation
became evident when the 3′-O-benzoyl ester in 5c was
replaced with the more labile 3′-O-acetyl group in 5d
(entry 4). The N-deacetylation of 5d was accompanied
by complete loss of the 3′-O-acetyl group. Methanolysis
of the ester moiety at C3′ was facilitated by the electron-
withdrawing tosylate at C2′. It is noteworthy that the
5′-O-tert-butyldimethylsilyl group in 5d was stable under
these conditions. Such compatibility with silyl ether
protecting groups enhances the practicality of this depro-
tection method. The arabino epimer 5e also underwent
From a practical perspective, synthesis of the acetyl
3a, propionyl 3b, and isobutyryl 3c derivatives are one-
pot procedures. Isolation of intermediates 1a-c and
2a-c is not necessary; excess reagents are removed by
evaporation in vacuo. Methanolysis of the crude reaction
mixtures gives products that are clean or readily purified.
Short reaction times for N-deprotection are advantageous
(12) (a) Rammler, D. H.; Khorana, H. G. J. Am. Chem. Soc. 1962,
84, 3112-3122. (b) Schaller, H.; Weimann, G.; Lerch, B.; Khorana, H.
G. J. Am. Chem. Soc. 1963, 85, 3821-3827.
(13) Ti, G. S.; Gaffney, B. L.; Jones, R. A. J. Am. Chem. Soc. 1982,
104, 1316-1319.
7456 J. Org. Chem., Vol. 70, No. 18, 2005

SCHEME 3. Related Nucleosides Subjected to
MeOH/∆
SCHEME 4. Methanolysis of Acylated Cytidine
Derivatives
selective methanolysis to give a high yield of the N-
deprotected 6e (entry 5).
Tetraacetyl derivatives of formycin 7a, tubercidin 8a,
guanosine 9, and the pivalylated 2,6-diaminopurine
nucleoside 10a were subjected to our standard conditions
to further probe its applicability (Scheme 3). The N-acetyl
group was cleanly removed from the formycin derivative
7a within 3 h at 65 °C (refluxing MeOH) to give 2′,3′,5′-
tri-O-acetylformycin (7). It is likely that some 7a also was
formed during a previously reported acetylation of for-
mycin14 and that facile N-deacetylation occurred during
repeated addition and evaporation of ethanol. The tu-
bercidin derivative 8a was more stable and required 7 h
at 105 °C for complete N-deacetylation. This treatment
resulted in significant accompanying loss of sugar O-
acetyl groups. The 2-N-acetyl group of guanosine deriva-
tive 9 was not removed by prolonged heating (12 h) at
105 °C. It is noteworthy that the diaminopurine deriva-
tive 10a lost the more bulky 6-N-pivalyl group within 2
h; the 2-N-acetyl group remained unchanged after 3 h
at 105 °C.
Cytidine derivatives had reactivity patterns similar to
those of adenosine. The tetraacylcytidines 11a (R )
COMe), 11b (R ) COEt), and 11c (R ) COiPr) underwent
N-deprotection within 3 h at 105-120 °C to give 12a
(82% by NMR, 61% isolated), 12b (77% isolated), and 12c
(77% isolated). Heating the tetratoluyl derivative 11d
(R ) p-methylbenzoyl) for 14 h at 105 °C gave 12d (86%
isolated) plus 8% of the 4-methoxy-2-pyrimidinone15
derivative 13d (methanolysis at C4 to produce p-tolua-
mide) (Scheme 4). Concomitant loss of toluyl ester groups
from the sugar moiety was not observed. By comparison,
removal of the N-acetyl group from tetraacetylcytidine
had required 40 h at reflux (65 °C).9
the relative electron delocalization into the π-deficient
ring (i.e., away from the amino nitrogen) is provided by
pKa data. Amides derived by N6 acylation of adenosine
(pKa1 ) 3.616) undergo N-deacylation quite readily. Those
derived from the more basic cytidine (pKa1 ) 4.217) are
less reactive with MeOH at 105 °C, and N-deacetylation
of amide derivatives of tubercidin (pKa1 ) 4.717) is even
more difficult.
In conclusion, we have demonstrated that solutions of
peracylated adenosine and cytidine derivatives in super-
heated methanol undergo selective N-deacylation. Con-
tamination with acidic or basic residues is avoided with
this convenient neutral procedure. An increase in the
steric bulk of the N-acyl group has a limited effect on
the rate of N-deacylation, but it significantly increases
the N/O selectivity ratio (i.e., apparent retardation of the
rate of O-deacylation). Removal of N-acyl groups derived
from arylcarboxylic acids requires more prolonged heat-
ing but proceeds successfully. Judicious choices of tem-
perature and N- versus O-acyl groups allow virtually
complete selectivity for N-deacylation with superheated
methanol. In contrast to the reactivity of amides at C6
of purine nucleosides, 2-acetamidopurine derivatives
were completely resistant to methanolysis at 105 °C. This
allows selective removal of the sterically hindered pivalyl
group from N6 of a 2-acetamido-6-pivalamidopurine
compound. Adenosine and cytidine derivatives with ester
protection on the sugar moiety and a free amino group
on the base are important intermediates. Our methodol-
ogy makes such derivatives readily accessible.
A net shift of electron density from the exocyclic amino
nitrogen atom into the π-deficient heterocyclic ring occurs
with conjugated amidine systems in the pyrimidine rings
of nucleobases. The resulting enhanced electrophilicity
of the carbonyl carbon atom of amides formed with these
amino groups allows deprotection of such amides in the
presence of usually more stable esters. An estimate of
Experimental Section
General. 1H (500 MHz) and 13C (125 MHz) NMR spectra were
recorded with solutions in CDCl3 unless noted. 13C peaks with
identical chemical shifts for more than one carbon are specified,
and a shift range is indicated (ovlp) for overlapping peaks with
(16) Dawson, R. M. C.; Elliott, D. C.; Elliott, W. H.; Jones, K. M.
(14) Lindell, S. D.; Moloney, B. A.; Hewitt, B. D.; Earnshaw, C. G.;
Dudfield, P. J.; Dancer, J. E. Bioorg. Med. Chem. Lett. 1999, 9, 1985-
1990.
Data for Biochemical Research; Oxford University Press: New York;
1969.
(17) Ikehara, M.; Fukui, T.; Uesugi, S. J. Biochem. 1974, 76, 107-
(15) Robins, M. J.; Naik, S. R. Biochemistry 1971, 10, 3591-3597.
115.
J. Org. Chem, Vol. 70, No. 18, 2005 7457

multiple carbons. Electron impact (EI) mass spectra were
obtained at 70 eV. Mallinckrodt anhydrous methanol was used
for all “superheated MeOH” deprotection experiments. Other
chemicals and solvents were of reagent quality.
Acylation of Adenosine with Acid Anhydrides. Adenosine
(0.5 g, 1.87 mmol), pyridine (2 mL), and an excess of the acid
anhydride were heated overnight at 50 or 80 °C (oil-bath
temperature). Volatiles were evaporated in vacuo, and the
residue was chromatographed [EtOAc/hexanes (1:2) f EtOAc
and then EtOAc/MeOH (10:1)] to give quantitative yields of the
6-N,2′,3′,5′-tri-O-tetraacyladenosines, 1, and/or 6-di-N,N-2′,3′,5′-
tri-O-pentaacyladenosines, 2.
General Procedure for Methanolysis of the Peracylated
Adenosines. Compounds 1 and/or 2 (1-2 mmol) were dissolved
in MeOH (10 mL) in a 30 mL pressure vessel equipped with a
Teflon valve. The vessel was placed in an oil bath heated at 105
or 120 °C for the specified time. Volatiles were evaporated, and
the residues were dried in vacuo. Compounds 3b and 3c were
1H NMR pure (>98%). Compound 3a was purified by crystal-
lization, and 3d-f were purified by chromatography. Larger
scale methanolysis reactions required longer reaction times and/
or higher temperatures. For example, N-deacetylation of a
solution of 1a (15.0 g, 34.5 mmol) in MeOH (150 mL) in a 250
mL vessel required heating for 6 h at 110 °C.
7-N-2′,3′,5′-Tri-O-tetraacetylformycin (7a). Formycin (1.0
g, 3.75 mmol), Ac2O (2.0 mL, 2.16 g, 21.2 mmol), and pyridine
(2 mL) were stirred for 1 h at ambient temperature. Volatiles
were evaporated, and the residue was chromatographed (EtOAc)
to give 7a (1.62 g, quantitative): UV (MeOH) max 262, 295, 305
nm (ꢀ 6200, 6800, 6700); min 247, 278, 301 nm (ꢀ 5300, 4700,
6500). 1H NMR δ 2.08, 2.10, 2.14, 2.80 (4 × s, 4 × 3H), 4.29-
4.33 (m, 1H), 4.42-4.48 (m, 2H), 5.45 (d, J ) 5.4 Hz, 1H), 5.68-
5.70 (m, 1H), 5.99 (t, J ) 5.4 Hz, 1H), 6.62 (br s, 1H), 8.24 (br s,
1H), 8.44 (s, 1H); 13C NMR δ 20.45, 20.46, 20.6, 23.0, 63.5, 71.9,
72.9, 75.7, 79.8, 120.9, 146.2, 147.9, 151.6, 155.3, 169.5, 169.6,
170.5, 172.0; FAB-MS m/z 436 (M + H+); HRMS calcd for
C18H22N5O8 436.1468, found 436.1472.
(dd, J ) 3.9, 5.4 Hz, 1H), 5.74 (t, J ) 5.9 Hz, 1H), 6.45 (d, J )
3.9 Hz, 1H), 6.46 (d, J ) 5.9 Hz, 1H), 7.13 (d, J ) 3.9 Hz, 1H),
8.34 (s, 1H); 13C NMR δ 20.3, 20.5, 20.7, 63.4, 70.7, 72.8, 79.3,
85.0, 100.0, 103.6, 121.2, 150.9, 152.1, 157.0, 169.5, 160.7, 170.3;
FAB-MS m/z 415 (100%, M + Na+); HRMS calcd for C17H20N4O7-
Na 415.1230, found 415.1231. Method B. A solution of 8a (1.0
g, 2.30 mmol) in MeOH (20 mL) was heated at 105 °C for 7 h
(TLC). Compound 8 was the major product (55-60%; 1H NMR),
but other compounds were present that had lost O-acetyl groups
from the sugar moiety.
2-Acetamido-9-(2,3,5-tri-O-acetyl-â-D-ribofuranosyl)-6-
pivalamidopurine (10a). A solution of 2-acetamido-9-(2,3,5-
tri-O-acetyl-â-D-ribofuranosyl)-6-chloropurine20 (2.9 g, 6.2 mmol)
and NaN3 (2.0 g, 30.8 mmol) in DMF (60 mL) was stirred
overnight at ambient temperature. Volatiles were evaporated,
and the residue was chromatographed [hexanes/EtOAc (2:1) f
EtOAc) to give 2-acetamido-9-(2,3,5-tri-O-acetyl-â-D-ribofurano-
syl)-6-azidopurine (2.6 g, 88%). A solution of this material (2.0
g, 4.2 mol) in MeOH (40 mL) was stirred overnight under 1 atm
of H2 with Pd‚C (5%, 180 mg). The catalyst was filtered and
washed with MeOH, and volatiles were evaporated from the
combined filtrate. The residue was chromatographed [MeOH/
EtOAc (1:10)] to give 2-acetamido-9-(2,3,5-tri-O-acetyl-â-D-ribo-
furanosyl)adenine21 (10) (1.78 g, 94%) as a yellow foam. A
solution of this material (400 mg, 0.89 mmol), pyridine (2 mL),
and pivalic anhydride (1 mL) was stirred overnight at 80 °C.
Volatiles were evaporated in vacuo, and the residue was chro-
matographed [hexanes/EtOAc (2:1) f EtOAc f MeOH/EtOAc
(1:10)] to give 10a (390 mg, 83%): 1H NMR δ 1.39 (s, 9H), 2.09,
2.10, 2.16 (3 × s, 3 × 3H), 2.54 (br s, 3H), 4.37-4.51 (m, 3H),
5.74 (s, 1H), 5.90 (t, J ) 5.1 Hz, 1H), 6.12 (d, J ) 4.9 Hz, 1H),
8.04 (s, 1H), 8.48 (br s, 1H), 8.72 (br s, 1H); 13C NMR δ 20.0,
20.2, 20.4, 24.9, 26.9, 40.1, 62.8, 70.1, 72.7, 79.8, 86.2, 120.1,
140.7, 150.3, 151.9, 152.5, 169.1, 169.2, 170.0, 174.9; FAB-MS
m/z 557 (100%, M + Na+); HRMS calcd for C23H30N6O9Na
557.1972, found 557.1967.
4-N-2′,3′,5′-Tri-O-tetraacetylcytidine8,21 (11a). Cytidine
(150 mg, 0.62 mmol) was stirred with Ac2O (1 mL, 1.08 g, 10.6
mmol) and pyridine (3 mL) for 1 h at 80 °C. Volatiles were
evaporated in vacuo, and the residue was dried under vacuum
2′,3′,5′-Tri-O-acetylformycin18 (7). A solution of 7a (1.62 g,
3.72 mmol) in MeOH (40 mL) was refluxed for 3 h (TLC).
Volatiles were evaporated to dryness, and the residue was
chromatographed [EtOAc f MeOH/EtOAc (1:10)] to give 7 (1.24
g, 84%): UV (MeOH) max 229, 293 nm (ꢀ 6200, 9800); min 244
nm (ꢀ 3200). 1H NMR δ 2.05, 2.10, 2.11 (3 × s, 3 × 3H), 4.28
(dd, J ) 4.9, 12.2 Hz, 1H), 4.34-4.37 (m, 1H), 4.59 (dd, J ) 2.4,
11.7 Hz, 1H), 5.49 (d, J ) 5.9 Hz, 1H), 5.64 (t, J ) 5.4 Hz, 1H),
5.97 (t, J ) 5.6 Hz, 1H), 6.64 (br s, 3H), 8.27 (s, 1H); 13C NMR
δ 20.4 (2C), 20.7, 63.5, 71.7, 73.3, 75.7, 76.7, 79.7, 122.6 (br s),
139.5 (br s), 140.5 (br s), 150.8 (br s), 151.8, 170.06, 170.14, 171.6;
FAB-MS m/z 416 (100%, M + Na+); HRMS calcd for C16H19N5O7-
Na 416.1182, found 416.1178.
1
at 80 °C to give 11a (0.25 g, 100%): H NMR δ 2.09, 2.11, 2.16,
2.30 (4 × s, 4 × 3H), 4.39-4.44 (m, 3H), 5.33 (t, J ) 5.9 Hz,
1H), 5.45 (dd, J ) 4.4, 5.5 Hz, 1H), 6.09 (d, J ) 3.9 Hz, 1H),
7.51 (d, J ) 7.8 Hz, 1H), 7.93 (d, J ) 7.3 Hz, 1H), 10.26 (s, 1H);
13C NMR δ 20.24 (2C), 20.6, 24.6, 62.5, 69.4, 73.5, 79.6, 89.1,
97.2, 144.0, 154.6, 163.3, 169.2, 169.3, 170.0, 171.3; FAB-MS m/z
434 (100%, M + Na+); HRMS calcd for C17H21N3O9Na 434.1175,
found 434.1178.
2′,3′,5′-Tri-O-acetylcytidine9,21 (12a). A solution of 11a (250
mg, 0.62 mmol) in MeOH (6 mL) was heated at 105 °C for 3 h.
Volatiles were evaporated, and the residue [82% (1H NMR) of
12a plus other compounds with loss of O-acetyl groups from the
sugar moiety] was chromatographed [EtOAc f MeOH/EtOAc
(1:5)] to give 12a (61%): 1H NMR δ 2.09, 2.10, 2.13 (3 × s, 3 ×
3H), 4.29-4.41 (m, 3H), 5.40 (t, J ) 5.6 Hz, 1H), 5.46 (dd, J )
4.9, 5.9 Hz, 1H), 5.92 (d, J ) 4.4 Hz, 1H), 5.96 (d, J ) 7.8 Hz,
4-N-2′,3′,5′-Tri-O-tetraacetyltubercidin19 (8a). Tubercidin
(1.0 g, 3.76 mmol), Ac2O (2.0 mL, 2.16 g, 21.2 mmol), and
pyridine (3 mL) were stirred for 2 h at 50 °C. Volatiles were
evaporated, and the residue was chromatographed (EtOAc) to
give 8a (1.48 g, 91%): 1H NMR δ 2.06 (s, 3H), 2.16 (s, 6H), 2.35
(br s, 3H), 4.35-4.45 (m, 3H), 5.59 (t, J ) 4.9 Hz, 1H), 5.79 (t,
J ) 5.9 Hz, 1H), 6.55 (d, J ) 5.9 Hz, 1H), 7.03 (d, J ) 3.4 Hz,
1H), 7.34 (d, J ) 5.9 Hz, 1H), 8.58 (s, 1H), 10.30 (br s, 1H); 13C
NMR δ 20.2, 20.4, 20.6, 24.2, 63.2, 70.6, 72.8, 79.4, 85.1, 105.0,
109.1, 123.1, 150.3, 150.5, 153.0, 169.1 (br s), 169.2, 169.5, 170.1;
FAB-MS m/z 457 (100%, M + Na+); HRMS calcd for C19H22N4O8-
Na 457.1335, found 457.1341.
2′,3′,5′-Tri-O-acetyltubercidin (8): Method A. Tubercidin
(1.0 g, 3.76 mmol), Ac2O (2.0 mL, 2.16 g, 21.2 mmol), and
pyridine (3 mL) were stirred for 1 h at ambient temperature.
Volatiles were evaporated, and the residue was chromato-
graphed (EtOAc) to give 8 (1.31 g, 89%): 1H NMR δ 2.05, 2.147,
2.151 (3 × s, 3 × 3H), 4.32-4.41 (m, 3H), 5.38 (br s, 2H), 5.56
1H), 6.49 (br s, 1H), 7.40 (d, J ) 7.3 Hz, 1H), 8.14 (br s, 1H); 13
C
NMR δ 20.38, 20.40, 20.7, 62.9, 69.8, 73.3, 78.9, 90.1, 96.2, 140.9,
155.6, 166.2, 169.5, 169.6, 170.4; FAB-MS m/z 392 (100%, M +
Na+); HMRS calcd for C15H19N3O8Na 392.1070, found 392.1086.
Acknowledgment. We gratefully acknowledge NIH
Grant GM029332, pharmaceutical company gift funds
(M.J.R.), and Brigham Young University for support of
this research.
Supporting Information Available: Experimental pro-
cedures, spectral data, and 13C NMR spectra. This material is
available free of charge via the Internet at http://pubs.acs.org.
JO051256W
(18) Buchanan, J. G.; Stobie, A.; Wightman, R. H. Can. J. Chem.
1980, 58, 2624-2627.
(19) Wiley, P. F.; Johnson, J. H.; Hanze A. R. J. Antibiotics 1976,
29, 720-727. The site of N-acetylation (N4) and signal integration
(5.3-5.9 ppm range) were reported incorrectly for 8a.
(20) Kozai, S.; Yorikane, A.; Maruyama, T. Nucleosides Nucleotides
Nucleic Acids 2001, 20, 1523-1531.
(21) Saladino, R.; Mincione, E.; Crestini, C.; Mezzetti, M. Tetrahe-
dron 1996, 52, 6759-6780.
7458 J. Org. Chem., Vol. 70, No. 18, 2005