A new method for the synthesis of oligodeoxyribonucleotides containing 4-N-alkoxycarbonyldeoxycytidine derivatives and their hybridization properties

Article abstract of DOI:10.1021/jo010813l

Oligodeoxyribonucleotides incorporating 4-N-alkoxycarbonyldeoxycytidine derivatives were synthesized on polystyrene-type ArgoPore resins having a new benzyloxy(diisopropyl)silyl linker, by use of ZnBr₂ as the detritylating agent. The first 3′-terminal thymidine could be attached to the resin by successive in situ reactions of 5′-O-DMTr-thymidine with diisopropylsilanediyl ditriflate and an ArgoPore resin containing hydroxyl groups. The use of this new silanediyl-type linker allowed release of the DNA chain from the resin by treatment with TBAF under neutral conditions. The T_m experiments apparently showed that incorporation of 4-N-alkoxycarbonyldeoxycytidines into DNA strands resulted in higher hybridization affinity with the complementary DNA strands than that of 4-N-acyldeoxycytidines. In addition, comparable T_m studies using oligodeoxyribonucleotides incorporating acyl (RC(O)-) groups and alkoxyacyl (RO(CH₂)_nC(O)-) groups having the same chain length show that the latter tend to exhibit higher T_m values than the former. It turned out that 4-N-alkoxycarbonyldeoxycytidines can form base pairs not only with deoxyguanosine but also with deoxyadenosine. Based on the ab initio calculations of the hydrogen bond energies of the possible base pairs formed between 4-N-methoxycarbonyl-1-methylcytosine and 9-methyladenine and the NMR analysis of the base-pairs of ¹⁵N-labeled 4-N-alkoxycarbonyldeoxycytidines with deoxyadenosine derivatives, we conclude that the base pair involves two unique hydrogen bonds between the cytosyl 4-NH group and the adenyl N¹ atom and between the O atom of the ester group and the adenyl 6-NH group.

Full text of DOI:10.1021/jo010813l

476
J . Org. Chem. 2002, 67, 476-485
A New Meth od for th e Syn th esis of Oligod eoxyr ibon u cleotid es
Con ta in in g 4-N-Alk oxyca r bon yld eoxycytid in e Der iva tives a n d
Th eir Hybr id iza tion P r op er ties
Akio Kobori, Kenichi Miyata, Masatoshi Ushioda, Kohji Seio, and Mitsuo Sekine*
Department of Life Science, Tokyo Institute of Technology, Nagatsuta, Midoriku,
Yokohama 226-8501, J apan
msekine@bio.titech.ac.jp
Received August 9, 2001
Oligodeoxyribonucleotides incorporating 4-N-alkoxycarbonyldeoxycytidine derivatives were syn-
thesized on polystyrene-type ArgoPore resins having a new benzyloxy(diisopropyl)silyl linker, by
use of ZnBr2 as the detritylating agent. The first 3′-terminal thymidine could be attached to the
resin by successive in situ reactions of 5′-O-DMTr-thymidine with diisopropylsilanediyl ditriflate
and an ArgoPore resin containing hydroxyl groups. The use of this new silanediyl-type linker allowed
release of the DNA chain from the resin by treatment with TBAF under neutral conditions. The
Tm experiments apparently showed that incorporation of 4-N-alkoxycarbonyldeoxycytidines into
DNA strands resulted in higher hybridization affinity with the complementary DNA strands than
that of 4-N-acyldeoxycytidines. In addition, comparable Tm studies using oligodeoxyribonucleotides
incorporating acyl (RC(O)-) groups and alkoxyacyl (RO(CH2)nC(O)-) groups having the same chain
length show that the latter tend to exhibit higher Tm values than the former. It turned out that
4-N-alkoxycarbonyldeoxycytidines can form base pairs not only with deoxyguanosine but also with
deoxyadenosine. Based on the ab initio calculations of the hydrogen bond energies of the possible
base pairs formed between 4-N-methoxycarbonyl-1-methylcytosine and 9-methyladenine and the
NMR analysis of the base-pairs of 15N-labeled 4-N-alkoxycarbonyldeoxycytidines with deoxy-
adenosine derivatives, we conclude that the base pair involves two unique hydrogen bonds between
the cytosyl 4-NH group and the adenyl N1 atom and between the O atom of the ester group and
the adenyl 6-NH group.
In tr od u ction
group of deoxycytidine stabilizes DNA duplexes, and the
base recognition ability of this modified base has the
same tendency as deoxycytidine. A variety of functional
molecules such as biotin and fluorescein for DNA/RNA
labeling and metal chelate complexes for DNA/RNA
scission have long been introduced into the side chains
at the 5-position of the cytosine base via amide bonds.
Therefore, 4-N-acyldeoxycytidine derivatives would be
more straightforward carriers of these functional groups
that can be easily introduced into the 4-amino group by
simple acylation. However, with an increase of the length
of the side chain of the acyl group attached to the 4-N-
amino group of deoxycytidine, DNA duplexes destabilize
markedly because their increased hydrophobicity dis-
turbs the hydrogen-bond structure around the major
groove.12 This kind of destabilization effect was also
reported by use of N-alkylated deoxycytidines by Thong
et al.13 To stabilize DNA duplexes having a long side
chain via an amide bond without disturbance of the
hydration structure14 around the major groove, we in-
troduced more hydrophilic oxygen atoms into the alkyl
side chain of the N-acyl group. In this paper, we report
the duplex stability and base-recognition ability of oli-
godeoxyribonucleotides containing various 4-N-alkoxy-
N-Acylated nucleosides, such as 4-N-acetylcytidine
(Cac)1-7 and N-threonylcarbamoyladenosine (t6A),8,9 are
naturally occurring modified nucleosides found in tRNA
and rRNA. It is known that Cac exists at the first letter
of the anticodon of some tRNAs and selectively forms the
Watson-Crick-type base pair with guanosine.10 Recently,
we have reported the duplex stability of oligodeoxyribo-
nucleotides having 4-N-acyldeoxycytidine derivatives.11
Our results showed that acetylation of the 4-N-amino
(1) Ohashi, Z.; Harada, F.; Nishimura, S. Biochim Biophys. Acta
1972, 262, 209-214.
(2) Edmonds, C. G.; Crain, P. F.; Hashizume, T.; Gupta, R.; Astetter,
K. O.; McCloskey, J . A. J . Chem. Soc., Chem. Commun. 1987, 909-
910.
(3) Edmonds, C. G.; Crain, P. F.; Gupta, R.; Hashizume, T.; Hocart,
C. H.; Kowalak, J . A.; Pomerantz, S. C.; Stetter, K. O.; McCloskey, J .
A. J . Bacteriol. 1991, 173, 3138-3148.
(4) Bruenger, E.; Kowalak, J . A.; Kuchino, Y.; McCloskey, J . A.;
Mizushima, H.; Stetter, K. O.; Crain, P. F. FASEB J . 1993, 7, 196-
200.
(5) Limbach, P. A.; Crain, P. F.; McCloskey, J . A. Nucleic Acids Res.
1994, 22, 2183-2196.
(6) Grosjean, H.; Sprinzl, M.; Steinberg, S. Biochimie 1995, 77, 139-
141.
(7) Sprinzl, M.; Hartman, T.; Weber, J .; Blank, J .; Zeidler, R. Nucleic
Acids Res. Sequence Suppl. 1989, 17, r17-r67.
(8) Miler, J . P.; Hussain, Z.; Schweizer, M. P. Nucleic. Acids Res.
1976, 3, 1185-1201.
(9) Grosjean, H.; Soll, D. G.; Crothers, D. M. J . Mol. Biol. 1976, 103,
499-519.
(10) Stern, L.; Schulman, L. H. J . Biol. Chem. 1978, 253, 6132-
(12) Wada, T.; Kobori, A.; Kawahara, S.; Sekine, M. Submitted to
Eur. J . Org. Chem.
(13) Nguyen, H.; Auffray, P.; Asseline, U.; Dupret, D.; Thuong, N.
T. Nucleic Acids Res. 1997, 25, 3059-3067.
(14) Shui, X.; McFail-Isom, L.; Hu, G. G.; Williams, L. D. Biochem-
istry 1998, 37, 8341-8355.
6139.
(11) Wada, T.; Kobori, A.; Kawahara, S.; Sekine, M. Tetrahedron
Lett. 1998, 39, 6907-6910.
10.1021/jo010813l CCC: $22.00 © 2002 American Chemical Society
Published on Web 12/21/2001

Synthesis of Oligodeoxyribonucleotides
J . Org. Chem., Vol. 67, No. 2, 2002 477
Sch em e 1a
a
Key: (i) (A) excess HMDS, CH3CN, 60 °C, 30 min, (B) 1.5 equiv of chloroformate, 1.5 equiv of pyridine, CH2Cl2, 30 min, (C) concNH3,
30 min; (ii) 1.2 equiv of DMTrCl, pyridine, rt, 3 h; (iii) 1.8 equiv of diisopropylethylamine, 1.5 equiv of chloro(2-cyanoethoxy)(N,N-
diisopropylamino)phosphine, CH2Cl2, rt 1 h.
carbonyldeoxycytidine and 4-N-acyldeoxycytidine deriva-
tives throughout Tm and 1H NMR experiments, which
suggests that the 4-N-alkoxycarbonyldeoxycytidine de-
rivatives can form base pairs with not only deoxygua-
nosine but also deoxyadenosine.
allowed to react with the hydroxyl group of the ArgoPore
resin. After capping of unreacted hydroxyl functions with
Ac2O in the presence of DMAP, the DMTrT-loaded resin
(6: 51.6 µmol/g) was obtained. When this resin was
treated with 1 M TBAF-AcOH in THF for 1 h, 90% of
the nucleoside was released from the solid support, as
evidenced by the DMTr assay. Since partial cleavage of
the silanediyl linker during the treatment with 3.0%
Resu lts a n d Discu ssion
18
Syn th esis of P h osp h or a m id ite Un its. 4-N-Alkoxy-
carbonyl-5′-O-(4,4′-dimethoxytrityl)-2′-deoxycytidine-3′-
O-phosphoramidite derivatives 4a -c were synthesized
from 2′-deoxycytidine (1), as described in Scheme 1. These
alkoxycarbonyl groups were selectively introduced by the
use of the corresponding alkyl chloroformate derivatives
into the 4-N-amino function of deoxycytidine derivatives
by a transient protection method using hexamethyldisi-
lazane. Deprotection of the trimethylsilyl groups attached
to the hydroxyl functions by treatment with ammonia
gave 4-N-alkoxycarbonyldeoxycytidine derivatives 2a -c
as crystalline materials. 4-N-alkoxycarbonyl-5′-O-(4,4′-
dimethoxytrityl)deoxycytidine derivatives 3a -c were
conveniently synthesized in high yields by in situ treat-
ment of 2a -c with DMTrCl. Compounds 3a -c were
converted to compounds 4a -c by phosphitylation in the
usual way.15
Syn th esis of Nu cleosid e-Loa d ed Solid Su p p or ts.
Recently, Fraser et al.16 have reported a new diisopro-
pylsilanediyl linker that can release oligodeoxyribonucle-
otides from solid supports by treatment with TBAF under
neutral conditions. However, this linker proved to be
partially lost during the acid treatment prescribed for
removal of the DMTr group. In addition, Fraser’s method
required a four-step reaction for introduction of DMTrT
into the solid support. Therefore, we made efforts to
overcome these synthetic problems. To facilitate intro-
duction of DMTrT into a resin, we used a polystyrene-
type ArgoPore resin having hydroxyl functions, which
have recently been used in the field of peptide synthesis.17
DMTrT was linked with this resin in a more straight-
forward way, as shown in Scheme 2. Reaction of DMTrT
(5) with diisopropylsilanediyl ditriflate in the presence
of DBU gave a 3′-O-silylated species, which in turn was
trichloroacetic acid in CH2Cl2 was observed, we em-
ployed 1 M zinc bromide in CH2Cl2-i-PrOH (85/15,
v/v).12,19 This reagent did not cause undesired cleavage
of the linker.
Syn th esis, P u r ifica tion , a n d Ch a r a cter iza tion of
Oligod eoxyr ib on u cleot id es Con t a in in g Mod ified
Ba ses. Solid-phase synthesis of oligodeoxyribonucleotides
was carried out by use of the standard protocol. The usual
chain elongation followed by the TBAF treatment gave
oligodeoxyribonucleotides (7a -c, 8), as shown in Scheme
2. The product was purified by anion-exchange HPLC and
analyzed by reversed-phase HPLC. The isolated yields
of the 13mers containing one or five modified nucleosides
were 20-29%, as shown in Scheme 2. The structures of
the purified products were confirmed by MALDI-TOF
mass. The ratios of T and dC* (the dC* is a modified dC)
in these modified DNA oligomers were confirmed by
enzymatic digestion with snake venom phosphodiesterase
and calf intestinal alkaline phosphatase. As reference
materials, oligodeoxyribonucleotides d(T6XT6) containing
4-N-propionyldeoxycytidine (dCpr, 7d ), 4-N-hexanoylde-
oxycytidine (dChex, 7e), and 4-N-(4-methoxybutyryl)-
deoxycytidine (dCmeb, 7f) were also synthesized by the
method previously reported by us where a phthaloyl
linker11 was used for the polymer-supported synthesis.
Hybr idization P r oper ty of Oligodeoxyn u cleotides
Con ta in in g Mod ified Ba ses. The thermal stability of
DNA duplexes d(T6XT6)/d(A6GA6) containing a modified
nucleoside was investigated. The Tm values of DNA
duplexes in phosphate buffer (pH 7.0) containing 1.0 M
NaCl are listed in Table 1. Interestingly, the oligonucle-
otide [d(T6CmocT6)] (Table 1, entry 3) having a 4-N-
methoxycarbonyldeoxycytidine formed a stable duplex
with the complementary strand d(A6GA6) and had a
higher Tm value (+1.0 °C) than the unmodified d(T6CT6).
An oligodeoxyribonucleotide (8) [d(T3CmocTCmocCmocT2Cmoc
-
(15) Marugg, J . E.; Tromp, M.; Kuyl-Yeheskiely, E.; van der Marel,
G. A.; van Boom, J . H. Recl. Trav. Chim. Pays-Bas 1986, 105, 505-
506.
(16) Routledge, A.; Wallis, M.; Ross, K.; Fraser, W. Bioorg. Med.
Chem. Lett. 1995, 23, 2641-2647.
(17) Morphy, J . R.; Rankovic, Z.; Rees, D. C. Tetrahedron Lett. 1996,
37, 3209-3212.
(18) Agrawal, S. In Protocols for Oligonucleotides and Analogues;
Humana Press: Totowa, NJ , 1993.
(19) Kierzek, R.; Ito, H.; Bhat, R.; Itakura, K. Tetarahedron Lett.
1981, 22, 3761-3764.

478 J . Org. Chem., Vol. 67, No. 2, 2002
Kobori et al.
Sch em e 2
Ta ble 1. Meltin g Tem p er a tu r es a t 260 n m of a Na tu r a l
Du p lex a n d Th ose Con ta in in g a Mod ified CXa
of 4-N-acetyldeoxycytidines into oligodeoxyribonucle-
otides resulted in higher hybridization affinity with
complementary strands than the corresponding natural-
type oligonucleotides. However, the stabilization effect
was weakened by increasing the side chain length of the
acyl group.12 In this study, entries 4 and 6 of Table 1
showed the same tendency. These results suggested that
the hydrophobicity of the side chain of the acyl group was
partly responsible for the duplex destabilization, as
shown in Table 1, entries 4 and 6. In the case of the
duplexes of entries 4-7 (Table 1) which have longer side
chain lengths, they have lower Tm values by 1.2-2.8 °C
than the duplex of entry 1 (Table 1) because of their
increased hydrophobicity, but the Tm values of the
duplexes of entries 5-7 (Table 1) are higher than that of
the duplex of entry 4 (Table 1). The duplexes of entries
5-7 (Table 1) have commonly one or two oxygen atoms
in the side chain to keep the hydration structure in the
major groove that is essential for stabilization of the DNA
duplex structure. It is likely that this stabilization was
caused by increase of the hydrophilicity of the side chain
due to incorporation of the oxygen atom.
a
∆Tm1 is the difference in the Tm value between the duplex
having the modified bases and that having natural bases. ∆Tm2
is the difference in the Tm value between the duplex having an
N-acylated cytosine and that having an N-alkoxycarbonyl cytosine.
TCmocT)] (Table 1, entry 9), which has five 4-N-methoxy-
carbonyldeoxycytidines in the strand, had a much higher
Tm value (+9.0 °C) when hybridized with the comple-
mentary strand by comparison with the unmodified d(T3-
CTC2T2CTCT) (Table 1, entry 8). On the basis of these
results, it is apparently concluded that the dCmoc-dG base
pair stabilizes the DNA duplex more than the natural
dC-dG base pair to a degree of 1.8 °C/one modification.
Moreover, compared with an oligonucleotide (Table 1,
entry 2) containing a 4-N-propionyldeoxycytidine (dCpr)
with the same chain length as the methoxycarbonyl
group (Table 1, entry 3), it turned out that incorporation
of an oxygen atom to the neighboring R-position of the
carbonyl group apparently increased the duplex stability.
In the previous paper,11 we reported that incorporation
Ba se Recogn ition Ability of 4-N-Alk oxyca r bon -
yld eoxycytid in es. The Tm values of DNA duplexes
having a mismatched base pair in X-Y are summarized
in Table 2. The duplexes having amide-type, N-acylated
deoxycytidine derivatives (Table 2, entries 2-4) were
destabilized by the one-base mismatch to the same degree
as the unmodified duplex d(T6CT6)/d(A6GA6) (Table 2,
entry 1). These results are in agreement with the
previous result.11 The duplexes having urethane-type,
N-acylated deoxycytidine derivatives (Table 2, entries
5-7) were considerably more stabilized than the natural
duplex, when deoxyadenosine was located at the opposite
position of 4-N-alkoxycarbonyl deoxycytidine derivatives.

Synthesis of Oligodeoxyribonucleotides
J . Org. Chem., Vol. 67, No. 2, 2002 479
Ta ble 2. Meltin g Tem p er a tu r es a t 260 n m of a Na tu r a l
Du p lex a n d Th ose Con ta in in g a Mod ified CX
kinds of calculations. In the gas phase, the difference in
energy between 9a and 9b was very narrow at a higher
level calculation (B3LYP/6-31++G**). To calculate the
energies of 9a , 9b, and 9c in the solution phase, the effect
of a bulk solvent was simulated using the IPCM con-
tinuum reaction field23 with the isodensity contour set
to 0.0004 electrons per cubic bohr.24 Consequently, 9a
was confirmed to be the most stable form, while the
tautomer 9c was the most unfavorable structure in the
solution phase. These results strongly suggested that in
both gas and solution phases 4-N-methoxy-1-methylcy-
tosine exists in the form of 9a .
As shown in Table 2, all duplexes having a T-C*
mismatched base pair have Tm values between 27.3 and
29.7 °C. Acyl modification on the cytosine base did not
essentially affect the stability of these mismatched
duplexes regardless of the type of acyl groups. In the case
of the duplexes having a C-C* mismatched base pair,
there is a more varied range (18.5-23.3 °C) in the Tm
values observed. Such variation did not influence the
G-base recognition of dC* since the Tm values observed
were too low.
More significant effects were observed by replacement
of the dA-dC base pair with a dA-dC* base pair on the
duplex stability. Compared with a series of amide-type
N-acylated deoxycytidine derivatives, urethane-type de-
oxycytidine derivatives exhibited significant increases of
the Tm values, as shown in entries 5-7 of Table 2. In
the case of dCmoc having the MeOC(O) group (Table 2,
entry 5), the Tm value increased up to 35.9 °C, which is
6.3 °C higher than that of dC (Table 2, entry 1). The most
serious loss of the base recognition ability was observed
in the case of dCbuc having a BuOC(O) group (Table 2,
entry 6) . There is a difference of only 4.7 °C in the Tm
values between the duplexes having a dCbuc-dG matched
base pair and a dCbuc-dA mismatched base pair. The
approach of the Tm values to a similar level is interesting
with respect to studies of mutagenesis of DNA. It should
be clarified if incorporation of dCbuc into DNA can cause
DNA mutagenesis. Our results strongly suggested that
there might be a possibility that a series of urethane-
type N-acylated deoxycytidine derivatives could be used
for such studies.
Estim a tion of Hyd r ogen Bon d in g En er gies of
Mism a tch ed Ba se P a ir s by Use of a b In itio Ca lcu la -
tion . As possible factors to be discussed for duplex
stabilization due to the C*-A mismatched base pair, the
base-stacking mode, base-pairing mode, and hydrogen
bonding energy between the C*-A base pair should be
taken into consideration. To examine the most reasonable
mode of the mismatched base pair, the energy difference
between the structural isomers of 4-N-methoxycarbonyl-
1-methylcytosine and the hydrogen bond energies of the
possible base pairs formed between 4-N-methoxycarbo-
nyl-1-methylcytosine (9) and 9-methyladenine (10) were
estimated in the gas phase by using the Gaussian 98
package program.20 Figure 2 shows three possible iso-
meric structures (9a -c) of 4-N-methoxy-1-methylcy-
tosine. The structure 9b is a rotamer of 9a around O*-
C-N4-C4, while the structure 9c is a tautomer of 9a .
Table 3 shows their relative energies in gas (ꢀ ) 0) and
solution phases (ꢀ ) 40 under the same circumstance as
inside the DNA helix).21 The standard structure of 9a
having the same orientation of the carbonyl oxygen as
that of the crystal structure of 4-N-acetylcytidine22 was
found to be the most stable conformer throughout four
Figure 2 shows three possible base-pairing modes
11a -c between 4-N-methoxycarbonyl-1-methylcytosine
9 and 9-methyladenine 10, which can form two hydrogen
bonds without great deformation. In the case of hydrogen
bonding energy calculation, the Cmoc-A mismatched pair
including the unfavorable imino tautomeric form 9c was
the most stable, and hydrogen bonding energy was
estimated to be -8.28 kcal/mol, as shown in 11c of Figure
2. The origin of the stabilization effect of this base pair
was due to the markedly negatively charged 4-nitrogen
atom with the Mulliken charge of -0.91 e and the
positively charged (+0.56 e) proton of the NH group at
position 3. The Cmoc-A mismatched pair (11a ) has a
smaller hydrogen bonding energy (-6.34 kcal/mol) even
though it includes the most favorable amino tautomer.
The partial charge of the oxygen in the ester group was
-0.42 e and decreased by -0.07 e compared with the
partial charge in the monomer 9a in Figure 2 (data not
shown). This means that the oxygen atom in the ester
group of 9a can take part in hydrogen bonding with the
amino proton of A.26 The hydrogen bonding energy (-5.52
kcal/mol) of the Cmoc-A mismatched pair 11b was slightly
lower than that of the 11a type. The energy differences
of three types of mismatched base pairs were within 2.8
kcal/mol. Therefore, it was suggested that the rotation
of the amide bond and the tautomerism between the
amino and imino forms would be responsible for stabi-
lization of Cmoc-A mismatched base pair. To estimate the
energy of the mismatched base pair exactly, a higher level
calculation and solvation calculation using explicit water
molecules were required to reproduce the environment
of the DNA helix.
1H NMR Stu d ies of Ba se P a ir s F or m ed betw een
3′,5′-O-TBDMS-4-N -m e t h oxyca r b on yld e oxycyt i-
(20) Frisch, M. J .; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J . R.; Zakrzewski, V. G.; Montgomery, J . A.,
J r.; Stratmann, R. E.; Burant, J . C.; Dapprich, S.; Millam, J . M.;
Daniels, A. D.; Kudin, K. N.; Strain, M. C.; Farkas, O.; Tomasi, J .;
Barone, V.; Cossi, M.; Cammi, R.; Mennucci, B.; Pomelli, C.; Adamo,
C.; Clifford, S.; Ochterski, J .; Petersson, G. A.; Ayala, P. Y.; Cui, Q.;
Morokuma, K.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.;
Foresman, J . B.; Cioslowski, J .; Ortiz, J . V.; Stefanov, B. B.; Liu, G.;
Liashenko, A.; Piskorz, P.; Komaromi, I.; Gomperts, R.; Martin, R. L.;
Fox, D. J .; Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.;
Gonzalez, C.; Challacombe, M.; Gill, P. M. W.; J ohnson, B. G.; Chen,
W.; Wong, M. W.; Andres, J . L.; Head-Gordon, M.; Replogle, E. S.;
Pople, J . A. Gaussian 98, revision A.7; Gaussian, Inc.: Pittsburgh, PA,
1998.
(21) Florian, J .; Leszczynski, J . J . Am. Chem. Soc. 1996, 118, 3010-
3017.
(22) Parthasarathy, R.: Ginell, S. L.; De, N. C.; Chheda, G. B.
Biochem. Biophys. Res. Commun. 1978, 83, 657-663.
(23) Foresman, J . B.; Keith, T. A.; Weiberg, K. B.; Snoonian, J .;
Frisch, M. J . J . Phys. Chem. 1996, 100, 16098-16104.
(24) Wiberg, K. B.; Keith, T. A.; Frisch, M. J .; Murcko, M. J . Phys.
Chem. 1995, 99, 9072-9079.
(25) Nickson, A. J . Am. Chem. Soc. 1957, 79, 243-247.
(26) Saenger, W. In Principles of Nucleic Acid Structure; Springer-
Verlag: New York, 1984.

480 J . Org. Chem., Vol. 67, No. 2, 2002
Kobori et al.
F igu r e 1. Anion-exchange HPLC profile of 8 (A) and reversed-phase HPLC profile of the mixture obtained by enzymatic digestion
of 8 (B).
d in e a n d 3′,5′-O-TBDMS-d eoxya d en osin e. NMR mea-
surements were performed at 223 K in deuterated
chloroform to confirm if 4-N-methoxycarbonyl-3′,5′-O-bis-
(tert-butyldimethylsilyl)deoxycytidine (12) can form a
base pair with 3′,5′-O-bis(tert-butyldimethylsilyl)deoxy-
adenosine (14) via hydrogen bonds.
and the amide proton of 4-N-methoxydeoxycytidine plays
a major role in forming the mismatched base pair.
To determine the real base pair mode, we synthesized
15N-labeled 4-N-methoxycarbonyldeoxycytidine deriva-
tives 15 and 16 by use of a modification of the method
for the synthesis of the 15N-labeled ribonucleoside coun-
terparts reported by Ariza,27 as shown in Scheme 3.
Compounds 15 and 16 were obtained from 3′,5′-O-bis-
(tert-butyldimethylsilyl)-2′-deoxyuridine28 (17) in overall
yields of 52% and 44%, respectively. Figure 4A shows the
1H NMR spectrum of a 1:1 mixture of the deoxyadenosine
derivative 14 and the 4-15N labeled compound 15, and
Figure 4B shows that of a 1:1 mixture of 14 and 3-15N
labeled 16. Only in the case of Figure 4A, split signals
having a large coupling constant (90.7 Hz) derived from
1
Figure 3A shows the H NMR spectra of 12 at various
temperatures. The chemical shift of unexchangeable
protons of compound 12 did not change at any temper-
ature except for the 6H proton, which slightly changed
from 8.34 ppm at 313 K to 8.59 ppm at 223 K. This
change of 6H is also observed in the case of 3′,5′-O-bis-
(tert-butyldimethylsilyl)deoxycytidine (13) (data not
shown). The 5H vinyl proton at ca. 7.2 ppm did not move.
This implies that an intramolecular hydrogen bond exists
in the range between -50 and +40 °C. This chemical
shift of 7.2 ppm is remarkably moved to the downfield
from 5.6 ppm, which is usual for N-unprotected cytidine
derivatives.12,22,26 Therefore, this NMR study revealed
this compound 12 exists in the form of 9a , as suggested
by the ab initio calculation. Furthermore, the small
mobility of the chemical shift of the amide proton of 12
from 7.42 ppm at 313 K to 7.98 ppm at 223 K implied
that compound 12 did not form a self-assembled base pair
in CDCl3 even at low temperatures.
1
a covalently bonded 15N and H were clearly observed at
11.6 ppm. These results revealed that compound 12 forms
a base pair with compound 14 in the amino form in
CDCl3, as shown in Figure 4A. These results obtained
from the NMR experiments reflect those of Tm experi-
ments above-mentioned.
Con clu sion
Incorporation of dCac into oligodeoxyribonucleotides
resulted in stabilization of DNA duplexes, but, in reverse,
N-acylated deoxycytidine derivatives having a longer
hydrophobic side chain destabilized DNA duplexes. To
suppress this destabilization, hydrophilic oxygen atoms
capable of maintaining the hydration structure in the
major groove were introduced into the side chain of the
acyl group. Particularly, in the case of dCmoc, the modified
DNA duplex including five dCmoc monomer units at
separate positions exhibited a higher Tm value than the
natural DNA duplex. The most striking characteristic of
4-N-alkoxycarbonyldeoxycytidine derivatives is that they
can recognize not only guanosine but also adenosine.
NMR and X-ray crystal analysis of DNA duplexes having
a C-A mismatched base pair revealed that cytidine could
form a Wobble-type base pair with adenosine under acidic
conditions by protonation of the N1 position of adenine.29
In this study, it was concluded that 4-N-alkoxycarbon-
yldeoxycytidine derivatives can form base pairs with not
1
Figure 3B shows the H NMR spectra of 13, 14, and a
1:1 mixture of 13 and 14 at 223 K. From these spectra,
no significant change in the spectra was observed.
Therefore, it was concluded that an A-C base pair in
CDCl3 cannot be formed between 13 and 14.
1
Figure 3C shows the H NMR spectra of 12 and a 1:1
1
mixture of 12 and 14 at 223 K. Compared with the H
NMR spectrum of a 1:1 mixture of 13 and 14, the amide
proton of 12 in Figure 3C(b) was largely shifted (ca. 3.64
ppm) downfield in Figure 3C(a), when 14 was added to
12. Moreover, the two amino protons of 14 were detected
separately at 8.02 and 7.00 ppm in the presence of an
equivalent molar amount of 12 at low temperature in
Figure 3C(a). It was also found that the 5H proton of 12
did not shift appreciably when 14 was added to 12. This
result implies that the intramolecular hydrogen bonding
of the carbonyl oxygen of the methoxycarbonyl group with
the 5-vinyl proton remains intact upon formation of the
base pair. Therefore, the base pair mode depicted in the
form 11b in Figure 2 was ruled out.
(27) Ariza, X.; Vilarrasa, J . J . Org, Chem. 2000, 65, 2827-2829.
(28) Ogilvie, K. K. Can. J . Chem. 1973, 51, 3799-3807.
(29) Brown, T.; Kennard, O.; kneale G.; Rabinovich, D. Nature 1986,
320, 552-555. Patel, D.; Kozlowski, S. A.; Ikuta, S.; Itakura, K.
Biochemistry 1984, 23, 3218-3226.
These results suggested that 4-N-methoxycarbonylde-
oxycytidine derivatives can form a base pair with deoxy-
adenosine more easily than the deoxycytidine derivatives

Synthesis of Oligodeoxyribonucleotides
J . Org. Chem., Vol. 67, No. 2, 2002 481
F igu r e 2. Energy-optimized structures of the conformers 9a ,b and tautomer 9c of 4-N-methoxycarbonyl-1-methylcytosine by ab
initio MO calculation at the B3LYP/6-31++G** and energy-optimized structure of base pairs between 9a -c and 9-methyladenosine
(10) by ab initio MO calculation at the HF/6-31+G**.
associated with DNA mutagenesis.30 Although 4-N-
alkoxycarbonylcytidine derivatives have not been discov-
Ta ble 3. Rela tive Differ en ce in En er gy (k ca l/m ol) of th e
Ta u tom er s 9a -c a t th e Selected Ba sis Sets
ered from naturally occurring RNAs such as tRNAs and
rRNAs, there is a possibility that these modified nucleo-
sides would enable us to artificially change the codon
box when incorporated into the anticodon first letter
of tRNAs. Further studies are under way in this direc-
tion.
basis set
9a
9b
9c
HF/6-31G* (kcal/mol)
0
+3.76
+2.64
ꢀ ) 0
Hf/6-31++G** (kcal/mol)
0
0
+3.70
+4.77
+2.82
+6.00
ꢀ ) 0
1-PCM,HF/6-31++G**//HF/6-31++G**
(kcal/mol)
ꢀ ) 40
B3LYP/6-31++G** (kcal/mol)
ꢀ ) 0
0
+3.07
+3.01
Exp er im en ta l Section
Gen er a l Rem a r k s. 1H, 13C, and 31P NMR spectra were
recorded at 270, 68, and 109 MHz, respectively. The chemical
shifts were measured from tetramethylsilane or DSS for 1H
only deoxyguanosine but also deoxyadenosine under even
neutral conditions by using a geometry different from the
naturally occurring mismatched base pair of the Wobble
type. This inherent property of 4-N-alkoxycarbonyldeoxy-
cytidine derivatives would provide new insight in studies
(30) Singer, B.; Kusmierek, J . T. Annu. Rev. Genet. 1982, 52, 655-
693.

482 J . Org. Chem., Vol. 67, No. 2, 2002
Kobori et al.
F igu r e 3. Panel A: 1H NMR spectra of 12 in CDCl3 at various temperatures. The symbol “*” refers to the amide proton of 12.
Panel B: 1H NMR spectra at 223 K (CDCl3) of a mixture of 13 and 14 (a), compound 13 (b), and compound 14 (c). The symbols
“*” and “+” in (a) and (b) refer to the amino protons and the 5H proton, respectively, of 13. Panel C: 1H NMR spectra at 223 K
(CDCl3) of a 1:1 mixture of 12 and 14 (a) and compound 12 (b). The symbols “*” and “+” in (a) and (b) refer to the amide proton
and the 5H proton, respectively, of dCmoc 12.
Sch em e 3a
a
Key: (i) 1.2 equiv of diphenyl phosphate, 2 equiv of 1H-tetrazole, 2 equiv of TsCl, pyridine, rt, 2 d; (ii) 1.2 equiv of 15NH4Cl, 1.2 equiv
of KOH, 1.4 equiv of Et3N, CH3CN-H2O (3:1, v/v), rt, 24 h; (iii) 1.2 equiv of methyl chloroformate, 1.5 equiv of pyridine, CH2Cl2, 1 h; (iv)
4 equiv of (CF3CO)2O, 2 equiv of NH4NO3, CH2Cl2, 0 °C, 15 min; (v) 1.3 equiv of 15NH4Cl, 1.1 equiv of KOH, 1.3 equiv of Et3N, CH3CN-
H2O (3:1, v/v), rt, 6 d; (vi) 1.2 equiv of diphenyl phosphate, 2 equiv of 1H-tetrazole, 2 equiv of TsCl, pyridine, rt, 2 d, then excess NH3 aq,
rt, 2 h.
NMR spectra, CDCl3 (77 ppm), or DSS (0 ppm) for 13C NMR
spectra and 85% phosphoric acid (0 ppm) for 31P NMR spectra.
UV spectra were recorded on a U-2000 spectrometer. MALDI
TOF mass spectra were taken on a Voyager RP. Column
chromatography was performed with silica gel C-200 pur-
chased from Wako Co. Ltd., and a minipump for a goldfish
bowl was conveniently used to attain sufficient pressure for
rapid chromatographic separation. Anion-exchange HPLC was
done on a Gen-Pak FAX column (Waters, 4.6 × 100 mm) on
an LC module 1 with a Waters M-741 data module and a
Waters column heater with a 10-77% linear gradient of 25
mM phosphate, 1 M sodium chloride buffer (pH 6.0) in 25 mM
phosphate buffer (pH 6.0) at 50 °C at a flow rate of 1.0 mL/
min for 30 min. Pyridine was distilled two times from p-
toluenesulfonyl chloride and from calcium hydride and then
stored over molecular sieves 4A. Compounds 13, 14 and 17
were synthesized by the method reported by Ogilvie.28 Ele-
mental analyses were performed by the Microanalytical Labo-
ratory, Tokyo Institute of Technology at Nagatsuta.
P r ep a r a tion of Th ym id in e-Loa d ed Solid Su p p or t via
Sila n ed iyl Lin k er 8. Diisopropylsilanediyl ditriflate (180 µL,
0.6 mmol) and 1,8-diazabicyclo[5.4.0]-7-undecene (269 mg, 1.8
mmol) were dissolved in dry CH3CN (2 mL). 5′-O-(4,4′-
Dimethoxytrityl)thymidine (109 mg, 0.2 mmol) in dry CH3CN
(1 mL) was added dropwise to the solution at -40 °C under
argon atmosphere. After being stirred for 1 h at room tem-
perature, ArgoPore (500 mg, 0.4 mmol) was added and stirred
for an additional 1 h at room temperature. The solid support

Synthesis of Oligodeoxyribonucleotides
J . Org. Chem., Vol. 67, No. 2, 2002 483
a Beckman DU 650 spectrophotometer by increasing the
temperature at the rate of 1.0 °C/min.
2′-Deoxy-4-N-m eth oxycar bon yl-5′-O-(4,4′-dim eth oxytr i-
tyl)cytid in e (3a ). Gen er a l P r oced u r e. 2′-Deoxycytidine
hydrochloride (1) (2.64 g, 10 mmol) and hexamethyldisilazane
(1.1 mL, 50 mmol) was dissolved in CH3CN (100 mL), and the
mixture was stirred for 30 min at 60 °C. After evaporation,
the residue and pyridine (1.2 mL, 15 mmol) were redissolved
in CH2Cl2 (100 mL). Methyl chloroformate (1.2 mL, 15 mmol)
was added dropwise over 10 min to the solution at 0 °C. After
the mixture was stirred for 30 min, the organic layer was
washed three times with 5% NaHCO3 (aq), dried over MgSO4,
filtered, and evaporated under reduced pressure. Pyridine-
concentrated NH3 (1:1, v/v, 100 mL) was added to the residue.
After being stirred for another 30 min, the solution was
evaporated to dryness. The residue was diluted with CHCl3
(100 mL) and washed three times with 5% NaHCO3 (aq). The
organic layer was combined, dried over MgSO4, filtered, and
evaporated under reduced pressure. The residue was dried by
repeated coevaporation with dry pyridine and finally dissolved
in dry pyridine (100 mL). To the solution was added 4,4′-
dimethoxytrityl chloride (4.07 g, 12 mmol), and the mixture
was stirred at room temperature for 3 h. The mixture was
concentrated to half volume, diluted with CHCl3, and washed
three times with 5% NaHCO3 (aq), and the aqueous layer was
back-extracted with CHCl3. The organic layer and the wash-
ings were combined, dried over Na2SO4, filtered, and concen-
trated to dryness under reduced pressure. The residue was
chromatographed on a column of silica gel (100 g) with CHCl3
containing 1% pyridine, applying a gradient of methanol (1-
2%) to give 3a (5.41 g, 92%) as colorless foam: 1H NMR
(CDCl3) δ 2.22 (1H, m), 2.75 (1H, m), 3.34 (1H, br), 3.45 (2H,
dd, J ) 3.0, 10.7 Hz), 3.79 ÅA3.74 (9H, 2s), 4.15 (1H, m), 4.50
(1H, m), 6.27 (1H, dd, J ) 5.4, 5.6 Hz), 6.84 (4H, d, J ) 8.4
Hz), 6.99 (1H, d, J ) 7.4 Hz), 7.05-7.40 (9H, m), 7.83 (1H,
br), 8.22 (1H, d, J ) 7.4 Hz); 13C NMR (CDCl3) δ 41.7, 52.6,
54.8, 62.6, 70.4, 86.2, 86.3, 87.0, 94.7, 112.8, 126.6, 127.5, 127.7,
129.6, 129.6, 134.9, 135.1, 143.8, 158.1, 158.1, 144.0, 152.7,
154.9, 162.2. Anal. Calcd for C32H33N3O8‚1/2H2O: C, 64.42; H,
5.74; N, 7.04. Found: C, 64.19; H, 5.52; N, 6.89.
2′-Deoxy-4-N-bu toxyca r bon yl-5′-O-(4,4′-d im eth oxytr i-
tyl)cytid in e (3b). With use of the general procedure, com-
pound 3b was obtained from 2′-deoxycytidine hydrochloride
(1) (527 mg, 2 mmol) as a colorless foam (1.08 g, 86%): 1H
NMR (CDCl3) δ 0.94 (3H, t, J ) 7.3 Hz), 1.39 (2H, m), 1.63
(2H, m), 2.23 (1H, m), 2.72 (1H, m), 3.22 (1H, br), 3.45 (2H,
m), 3.80 (6H, s), 4.15-4.23 (3H, m), 5.00 (1H, m), 6.27 (1H,
dd, J ) 5.6, 5.6 Hz), 6.84 (4H, d, J ) 8.9 Hz), 7.00 (1H, d, J )
7.3 Hz), 7.23-7.41 (9H, m), 7.61 (1H, br), 8.22 (1H, d, J ) 7.3
Hz); 13C NMR (CDCl3) δ 13.7, 19.0, 30.6, 66.0, 55.2, 42.1, 62.7,
70.8, 87.2, 88.4, 86.7, 94.7, 113.1, 126.9, 127.8, 129.8, 129.9,
135.2, 135.4, 144.0, 144.1, 144.1, 152.3, 155.2, 158.4, 162.1.
Anal. Calcd for C35H39N3O8‚H2O: C, 64.90; H, 6.38; N, 6.49.
Found: C, 65.11; H, 6.02; N, 6.48.
2′-De oxy-4-N -m e t h oxye t h oxyca r b on yl-5′-O-(4,4′-d i-
m eth oxytr ityl)cytid in e (3c). With use of the general pro-
cedure, compound 3c was obtained from 2′-deoxycytidine
hydrochloride (1) (527 mg, 2 mmol) as a colorless foam (1.07
g, 85%): 1H NMR (CDCl3) δ 2.26 (1H, m), 2.51 (1H, br), 2.55
(1H, m), 3.40 (3H, s), 3.46 (2H, m), 3.63 (2H, t, J ) 4.6 Hz),
3.80 (6H, s), 4.11 (1H, m), 4.33 (2H, t, J ) 4.6 Hz), 4.49 (1H,
m), 6.25 (1H, dd, J ) 5.6, 5.9 Hz), 6.85 (4H, d, J ) 8.3 Hz),
6.99 (1H, d, J ) 7.3 Hz), 7.16-7.41 (9H, m), 8.23 (1H, d, J )
7.3 Hz); 13C NMR (CDCl3) δ 42.0, 55.1, 58.8, 62.7, 64.8, 70.1,
70.7, 86.4, 86.7, 87.2, 94.8, 113.1, 126.8, 127.7, 127.9, 129.8,
129.8, 135.2, 135.3, 144.0, 144.1, 152.2, 155.1, 162.1. Anal.
Calcd for C34H37N3O9‚1/2H2O: C, 63.74; H, 5.98; N, 6.56.
Found: C, 63.80; H, 5.97; N, 6.30.
F igu r e 4. 1H NMR spectra at 223 K (CDCl3) of a 1:1 mixture
of 4-N-methoxycarbonyl -3′,5′-bisTBDMS deoxycytidine (15)
and 14 (A) and a 1:1 mixture of 4-N-methoxycarbonyl -3′,5′-
O-bisTBDMS deoxycytidine (16) and 14 (B). The symbol “*”
refers to the amide proton of 15 and 16.
was filtered and washed with pyridine. The hydroxyl groups
were blocked by treatment with 0.1 M 4-(N,N-dimethyamino)-
pyridine and 10% (CH3CO)2O in pyridine. The resulting
support 6 was filtered, washed with pyridine, and dried. The
loading amount of thymidine was estimated by DMTr cation
assay to be 51.6 µmol/g.
Sta bility of DMT-T-Su p p or t tow a r d Tetr a bu tyla m m o-
n iu m F lu or id e. The DMT-T-ArgoPore gel support 6 (19.4 mg,
1.0 µmol) was suspended in a solution of tetrabutylammonium
fluoride hydrate (131 mg, 0.5 mmol) and acetic acid (30 mL,
0.5 mmol) in tetrahydrofuran (500 µL). After 1 h, the support
was filtered and washed with pyridine (1 mL × 3) and CH2-
Cl2 (1 mL × 3). The support was treated with 1% TFA in CH2-
Cl2, and the TFA solution was filtered. The filtrate was
concentrated to dryness, and the residue was dissolved in
perchloric acid-ethanol (3:2, v/v) to estimate the amount of
the released DMTr cation by measurement of the UV absor-
bance at 498 nm (ꢀ ) 71 700).
Typ ica l P r oced u r e for Solid -P h a se Syn th esis. Each
cycle of chain elongation consisted of detritylation (1 M zinc
bromide in CH2Cl2-i-PrOH (85/15, v/v), 1 mL, 40 min), washing
(pyridine (1 mL × 3), CH3CN (1 mL × 3)), coupling (0.1 M
amidite units, 0.4 M 1H-tetrazole in CH3CN (200 µL), 5 min),
and washing (CH3CN (1 mL × 3), pyridine (1 mL × 3)),
oxidation (0.1 M I2, THF-pyridine-H2O (10/10/1, v/v/v), 2
min), washing (pyridine (1 mL × 3), CH3CN (1 mL × 3), CH2-
Cl2 (1 mL × 3)). Generally, the average yield per cycle was
estimated to be 97-99% by the DMTr cation assay. After chain
elongation, the DMTr group was removed by treatment with
1 M zinc bromide in CH2Cl2-i-PrOH (85/15, v/v, 1 mL) for 40
min, and the resin was washed with pyridine(1 mL × 3) and
CH2Cl2(1 mL × 3). The cyanoethyl group was removed by
treatment with DBU-CH3CN (9/1, v/v, 500 mL) for 1 min, and
the resin was washed with CH3CN (1 mL × 3). The oligomer
was released from the ArgoPore resin by treatment with
tetrabutylammonium fluoride hydrate (131 mg, 0.5 mmol) and
acetic acid (30 mL, 0.5 mmol) in tetrahydrofuran (500 mL) for
1 h. The ArgoPore resin was removed by filtration and washed
with CH3CN(1 mL × 3). The filtrate was desalted using gel
filtration and purified by anion-exchange HPLC. The oligomers
isolated were analyzed by nuclease digestion and MALDI TOF
mass spectrometry.
Tm Mea su r em en t. An appropriate oligonucleotide and its
complementary DNA 13 mer d(A6GA6) were dissolved in a
buffer consisting of 1.0 M NaCl, 10 mM sodium phosphate,
and 0.1 mM EDTA adjusted to pH 7.0. The solution containing
oligonucleotides was kept at 60 °C for 10 min for complete
dissociation of the duplex to single strands, cooled at the rate
of -1.0 °C/min, and kept at 0 °C for 10 min. After that, the
melting temperatures (Tm) were determined at 260 nm using
2′-Deoxy-4-N-m eth oxycar bon yl-5′-O-(4,4′-dim eth oxytr i-
tyl)cytid in e 3′-(2-Cya n oeth yl N,N-d iisop r op ylp h osp h or -
a m id ite) (4a ). Compound 3a (580 mg, 1.0 mmol) was rendered
anhydrous by successive coevaporations with dry pyridine (1
mL × 3) and dissolved in dry CH2Cl2 (10 mL). To the mixture
were added diisopropylethylamine (313 µL, 1.8 mmol) and
chloro(2-cyanoethoxy)(N,N-diisopropylamino)phosphine (335

484 J . Org. Chem., Vol. 67, No. 2, 2002
Kobori et al.
µL, 1.5 mmol). The resulting mixture was stirred for 1 h and
extracted with 5% NaHCO3 and water. The usual workup
followed by silica gel column chromatography eluted with
hexanes-ethyl acetate containing 0.5% triethylamine gave 4a
(701 mg, 89%): 1H NMR (CDCl3) δ 1.05-1.29 (12H, m), 2.28
(1H, m), 2.43 (1H, t, J ) 6.4 Hz), 2.61 (1H, t, J ) 6.4 Hz), 2.74
(1H, m), 3.60-3.80 (15H, m), 4.19 (1H, m), 4.61(1H, m), 6.25
(1H, dd, J ) 5.6 Hz, J ) 4.9 Hz), 6.82-6.97 (5H, m), 7.22-
7.42 (9H, m), 8.19, 8.28 (1H, d, J ) 7.4 Hz); 13C NMR (CDCl3)
δ 20.1, 20.2, 20.3, 20.4, 21.5, 24.6, 24.6, 40.8, 41.1, 43.0, 43.1,
43.2, 43.3, 52.9, 55.2, 55.1, 58.0, 58.3, 61.8, 62.3, 71.3, 72.3,
85.4, 85.5, 85.5, 86.7, 94.5, 113.1, 117.2, 117.3, 126.9, 127.8,
127.9, 128.0, 128.0, 129.8, 129.9, 130.0, 135.0, 135.1, 135.2,
143.8, 144.0, 152.7, 154.7, 158.4, 162.0; 31P NMR (CDCl3) δ
149.4, 149.9. Anal. Calcd for C41H50N5O9P‚1/2H2O: C, 61.80;
H, 6.45; N, 8.79. Found: C, 61.76; H, 6.19; N, 8.72.
4-N-Bu toxyca r bon yl-2′-d eoxy-5′-O-(4,4′-d im eth oxytr i-
tyl)cytid in e 3′-(2-Cya n oeth yl N,N-d iisop r op ylp h osp h or -
a m id ite) (4b). With use of the general procedure, compound
(4b) was obtained from 4-N-butoxycarbonyl-5′-O-(4,4′-dimethoxy-
trityl)deoxycytidine (3b) (378 mg, 0.6 mmol) as colorless foam
(414 mg, 83%): 1H NMR (CDCl3) δ 0.95 (3H, t, J ) 7.26), 1.06-
1.44 (14H, m), 1.64 (2H, m), 2.36 (1H, m), 2.44, 2.61 (2H, t, J
) 6.3 Hz), 2.74 (1H, m), 3.38-3.82 (15H, m), 4.15-4.21 (3H,
m), 4.62 (1H, m), 6.45 (1H, m), 6.82-6.90 (4H, m), 6.93 (1H,
dd, J ) 7.3, 7.6 Hz), 7.23-7.42 (9H, m), 8.18, 8.28 (1H, d, J )
7.3, 7.6 Hz); 13C NMR (CDCl3) δ 13.5, 19.1, 20.1, 20.2, 20.3,
20.3, 24.4, 24.6, 24.6, 30.5, 41.5, 41.6, 43.0, 43.1, 43.3, 43.3,
55.3, 55.3, 55.3, 58.1, 58.3, 61.9, 62.3, 66.0, 71.7, 72.3, 85.5,
85.6, 85.7, 87.0, 94.5, 113.2, 117.2, 117.4, 127.4, 127.8, 127.9,
128.0, 128.0, 130.1, 130.2, 130.2, 135.4, 135.4, 144.1, 144.2,
152.3, 155.1, 158.6, 162.0; 31P NMR (CDCl3) δ 149.3, 149.9.
Anal. Calcd for C44H56N5O9P: C, 63.68; H, 6.80; N, 8.44.
Found: C, 63.81; H, 6.77; N, 8.21.
1-[3,5-O-Bis(ter t-bu tyld im eth ylsilyl)-2-d eoxy-â-D-r ibo-
fu r a n osyl]-4-(1H-tetr a zol-1-yl)p yr im id in -2(1H)-on e (18).
Compound 17 (500 mg, 1.1 mmol), 1H-tetrazole (154 mg, 2.2
mmol), diphenyl phosphate (330 mg, 1.3 mmol), and tosyl
chloride (419 mg, 2.2 mmol) were dissolved in pyridine (3 mL),
and the mixture was stirred at room temperature for 2 days.
Afterward, 1 mL of water was added, and the solution was
poured into aqueous Na2CO3 (saturated) and extracted with
CH2Cl2. The solvent was dried (Na2SO4) and removed by
coevaporation with toluene. The residue was purified by
column chromatography (hexanes-ethyl ether 60:40) to afford
18 (436 mg, 78%): 1H NMR (CDCl3) δ 0.07-0.16 (12H, m),
0.89-0.95 (18H, m), 2.17-2.26 (1H, m), 2.59-2.69 (1H, m),
3.82 (1H, dd, J ) 2.5 Hz, J ) 12.0 Hz), 4.01-4.06 (2H, m),
4.40 (1H, m), 6.25 (1H, dd, J ) 4.3 Hz, J ) 6.6 Hz), 7.12 (1H,
d, J ) 7.2 Hz), 8.87 (1H, d, J ) 7.2 Hz), 9.60 (1H, s); 13C NMR
(CDCl3) δ -5.4, -4.9, -4.5, 17.9, 18.3, 25.6, 25.9, 42.2, 61.5,
69.6, 87.8, 88.1, 94.1, 140.4, 148.1, 153.6, 156.8. MS m/z calcd
+
for C22H41N6O4Si2 509.2728, found 509.2738.
[4-15NH2]-3′,5′-O-Bis(ter t-bu tyld im eth ylsilyl)-2′-d eoxy-
cytid in e (19). 15NH4Cl (144 mg, 2.6 mmol) and KOH (85%,
174 mg, 2.6 mmol) were placed in a 30-ml round flask and
sealed with a septum. Water (4.4 mL), CH3CN (4.4 mL), Et3N
(430 mL, 3.1 mmol), and a solution of 1-[3,5-O-Bis(tert-
butyldimethylsilyl)-2-deoxy-â-D-ribofuranosyl]-4-(1H-tetrazol-
1-yl)pyrimidin-2(1H)-one (18) (1.1 g, 2.2 mmol) in CH3CN (8.8
mL) were then added sequentially with syringe equipment,
avoiding leakage of ammonia. After the mixture was stirred
for 24 h, the solvent was removed, and the residue was diluted
with CHCl3 and washed three times with 5% NaHCO3 (aq).
The organic layer was dried over Na2SO4, filtered, and
concentrated to dryness under reduced pressure. The residue
was chromatographed on a column of silica gel (20 g) with
CHCl3, applying a gradient of methanol (1-2%) to give 19 (964
mg, 80%) as colorless foam: 1H NMR (CDCl3) δ 0.05 (6H, s),
0.11 (6H, s), 0.88-0.93 (18H, m), 2.08-2.13 (1H, m), 2.38-
2.45 (1H, m), 3.74-3.94 (3H, m), 4.37 (1H, m), 5.62 (1H, d, J
) 7.4 Hz), 6.26 (1H, dd, J ) 5.7 Hz), 8.01 (1H, d, J ) 7.3 Hz);
13C NMR (CDCl3) δ -5.7, -5.6, -5.0, -4.7, 17.7, 18.1, 25.6,
25.7, 41.8, 61.8, 70.3, 85.3, 86.8, 94.7, 139.9, 155.7, 165.7 (d, J
2′-De oxy-4-N -m e t h oxye t h oxyca r b on yl-5′-O-(4,4′-d i-
m eth oxytr ityl)cytid in e 3′-(2-Cya n oeth yl N,N-d iisop r o-
p ylp h osp h or a m id ite) (4c). With use of the general proce-
dure, compound (4c) was obtained from 2′-deoxy-4-N-meth-
oxyethoxycarbonyl-5′-O-(4,4′-dimethoxytrityl) cytidine (3c) (190
mg, 0.3 mmol) as colorless foam (130 mg, 52%): 1H NMR
(CDCl3) δ 1.06-1.29 (12H, m), 2.29 (1H, m), 2.44, 2.62 (2H, t,
J ) 6.4 Hz), 2.74 (1H, m), 3.36-3.83 (11H, m), 3.81 (3H, s),
3.83 (3H, 1s), 4.20 (1H, m), 4.33 (2H, m), 4.61 (1H, m), 6.25
(1H, m), 6.84 (4H, m), 6.90, 6.93(1H, d, J ) 7.2, 7.6 Hz), 7.22-
7.41 (9H, m), 8.19, 8.29 (1H, d, J ) 7.2, 7.6 Hz); 13C NMR
(CDCl3) δ 20.1, 20.2, 20.2, 20.3, 24.5, 24.7, 24.7, 41.6, 41.7,
43.0, 43.2, 43.3, 43.3, 55.2, 55.2, 58.0, 58.2, 58.9, 61.9, 62.2,
64.6, 70.3, 71.6, 72.2, 85.5, 85.6, 85.6, 87.1, 94.6, 113.2, 117.3,
117.4, 127.3, 127.8, 127.9, 128.1, 128.1, 130.1, 130.2, 130.2,
135.3, 135.4, 135.4, 144.0, 144.2, 152.2, 155.0, 158.5, 162.0;
+
) 20.1 Hz); MS m/z calcd for C23H42N215NO4Si2 457.2684,
found 457.2685.
[4-15NH2]-3′,5′-O-Bis(ter t-bu tyld im eth ylsilyl)-2′-d eoxy-
4-N-m eth oxyca r bon ylcytid in e (15). Compound 15 was
obtained from [4-15NH2]-3′,5′-O-bis(tert-butyldimethylsilyl)-2′-
deoxycytidine (19) (183 mg, 0.4 mmol) as colorless foam (170
mg, 83%) with use of the same procedure of compound 12: 1H
NMR (CDCl3) δ 0.06 (6H, s), 0.12 (6H, s), 0.88 (9H, s), 0.93
(9H, s), 2.11-2.16 (1H, m), 2.50-2.55 (1H, m), 3.77-3.98 (6H,
m), 4.38-4.40 (1H, m), 6.24 (1H, dd, J ) 5.4 Hz, J ) 5.4 Hz),
7.17 (1H, d, J ) 7.3 Hz), 7.40 (1H, br), 8.38 (1H, d, J ) 7.3
Hz); 13C NMR (CDCl3) δ -5.6, -5.5, -5.0, -4.6, 17.8, 18.2,
25.6, 25.8, 42.1, 52.8, 61.8, 69.8, 86.5, 87.5, 94.2, 144.0, 152.9
(d, J ) 26.2 Hz), 154.5 (br), 162.1 (d, J ) 20.1 Hz); MS m/z
calcd for C23H44N215NO6Si2 515.2739, found 515.2751.
31P NMR (CDCl3) δ 149.4, 150.0. Anal. Calcd for C43H54
-
N5O9P: C, 62.08; H, 6.54; N, 8.42. Found: C, 62.15; H, 6.52;
N, 8.27.
3′,5′-O-Bis(ter t-bu tyld im eth ylsilyl)-2′-d eoxy-4-N-m eth -
oxyca r bon ylcytid in e (12). 3′,5′-O-Bis(tert-butyldimethylsi-
lyl)deoxycytidine28 (3.7 g, 8 mmol) and pyridine (1.0 mL, 12
mmol) were dissolved in CH2Cl2 (80 mL). Methyl chloroformate
(740 mL, 9.6 mmol) was added dropwise over 10 min to the
solution at 0 °C, and the mixture was stirred at room
temperature for 1 h. The mixture was concentrated to half
volume, diluted with CHCl3, and washed three times with 5%
NaHCO3 (aq). The aqueous layer was back-extracted with
CHCl3. The organic layer and the washings were combined
and dried over Na2SO4, filtered, and concentrated to dryness
under reduced pressure. The residue was chromatographed
on a column of silica gel (80 g) with CHCl3, applying a gradient
of methanol (1-2%) to give 12 (3.3 g, 80%) as colorless foam:
1H NMR (CDCl3) δ 0.11 (6H, s), 0.12 (6H, s), 0.88 (9H, s), 0.93
(9H, s), 2.10-2.18 (1H, m), 2.47-2.57 (1H, m), 3.76-3.98 (6H,
m), 4.35-4.41 (1H, m), 6.24 (1H, dd, J ) 5.4 Hz, J ) 5.4 Hz),
7.16 (1H, d, J ) 7.3 Hz), 7.46 (1H, br), 8.38 (1H, d, J ) 7.3
Hz); 13C NMR (CDCl3) δ -5.6, -5.5, -5.0, -4.6, 17.9, 18.3,
25.7, 25.9, 42.3, 53.0, 61.7, 70.0, 86.7, 87.7, 94.2, 144.5, 152.9,
154.9, 162.0; MS m/z calcd for C23H44N3O6Si2+ 514.2769, found
514.2781.
3′,5′-O-Bis(ter t-b u t yld im et h ylsilyl)-2′-d eoxy-3-n it r o-
u r id in e (20). Trifluoroacetic anhydride (2.82 mL, 20 mmol)
was added to a suspension of finely powdered NH4NO3 (800
mg, 10 mmol) in anhydrous CH2Cl2 (5 mL), at 0 °C. The
mixture was vigorously stirred at room temperature until the
solid was dissolved (ca. 1 h) and then cooled again. After
addition of 3′,5′-O-bis(tert-butyldimethylsilyl)-2′-deoxyuridine
(17) (2.28 g, 5.0 mmol), the mixture was stirred at 0 °C for 15
min, washed with cold phosphate buffer, dried, and evaporated
in vacuo. The residue was chromatographed on a column of
silica gel (40 g) with hexane, applying a gradient of ethyl
acetate (3-5%) to give 20 (2.1 g, 83%) as colorless syrup: 1H
NMR (CDCl3) δ 0.09-0.12 (12H, m), 0.90-0.93 (18H, m), 2.10-
2.17 (1H, m), 2.34-2.42 (1H, m), 3.76-3.98 (3H, m), 4.43 (1H,
m), 5.79 (1H, d, J ) 8.2 Hz), 6.25 (1H, dd, J ) 5.9 Hz), 8.01
(1H, d, J ) 8.2 Hz); 13C NMR (CDCl3) δ -5.5, -5.5, -4.8, -4.7,
18.0, 18.4, 25.7, 25.9, 42.0, 62.3, 71.1, 86.4, 88.2, 100.8, 139.7,
145.2, 155.1; MS m/z calcd for C21H40N3O7Si2+ 502.2405, found
502.2422.

Synthesis of Oligodeoxyribonucleotides
J . Org. Chem., Vol. 67, No. 2, 2002 485
[3-15N]-3′,5′-O-Bis(ter t-bu tyld im eth ylsilyl)-2′-d eoxyu r i-
d in e (21). 15NH4Cl (357 mg, 6.5 mmol) and KOH (85%, 309
mg, 5.5 mmol) were placed in a 10 mL round-bottom flask and
sealed with a septum. Water (12 mL), CH3CN (12 mL), Et3N
(909 mL, 6.5 mmol), and a solution of 3′,5′-O-bis(tert-butyldi-
methylsilyl)-2′-deoxy-3-nitrouridine (20) (2.5 g, 5.0 mmol) in
CH3CN (24 mL) were added sequentially, avoiding leakage of
ammonia. After the mixture was stirred for 6 days, the solvent
was removed, and the residue was diluted with CHCl3 and
washed three times with 5% NaHCO3 (aq). The organic layer
was dried over Na2SO4, filtered, and concentrated to dryness
under reduced pressure. The residue was chromatographed
on a column of silica gel (50 g) with CHCl3, applying a gradient
of methanol (1-2%) to give 21 (2.0 g, 87%) as a colorless
foam: 1H NMR (CDCl3) δ 0.08-0.11 (12H, m), 0.89-0.92 (18H,
m), 2.01-2.11 (1H, m), 2.28-2.37 (1H, m), 3.73-3.94 (3H, m),
4.41-4.43 (1H, m), 5.69 (1H, d, J ) 7.9 Hz), 6.29 (1H, dd, J )
5.9 Hz), 7.89 (1H, d, J ) 7.9 Hz), 8.43 (1H, d, J ) 91.0 Hz);
13C NMR (CDCl3) δ -5.6, -5.5, -4.9, -4.6, 17.9, 18.3, 25.7,
25.8, 41.7, 62.2, 70.9, 85.0, 87.5, 102.0, 139.8, 150.3 (br), 163.6
4.40 (1H, m), 5.63 (1H, d, J ) 7.4 Hz), 6.26 (1H, dd, J ) 5.7
Hz), 8.01 (1H, d, J ) 7.3 Hz); 13C NMR (CDCl3) δ -5.7, -5.7,
-5.0, -4.7, 17.8, 18.2, 25.6, 25.7, 41.8, 61.7, 70.3, 85.3, 86.9,
94.7, 140.0, 155.7 (d, J ) 7.8 Hz), 165.6 (d, J ) 6.1 Hz); MS
+
m/z calcd for C23H44N3O6Si2 457.2684, found 457.2685.
[3-15N]-3′,5′-O-Bis(ter t-b u t yld im et h ylsilyl)-2′-d eoxy-4-
N-m eth oxyca r bon ylcytid in e (16). Compound 16 was ob-
tained from [3-15N]-3′,5′-O-bis(tert-butyldimethylsilyl)-2′-de-
oxycytidine (22) (183 mg, 0.4 mmol) as a colorless foam (130
mg, 78%) by use of the same procedure as for compound 12:
1H NMR (CDCl3) δ 0.06 (6H, s), 0.12 (6H, s), 0.88 (9H, s), 0.93
(9H, s), 2.11-2.17 (1H, m), 2.50-2.55 (1H, m), 3.76-3.98 (6H,
m), 4.38-4.40 (1H, m), 6.24 (1H, dd, J ) 5.4 Hz, J ) 5.4 Hz),
7.16 (1H, d, J ) 7.3 Hz), 7.46 (1H, br), 8.38 (1H, d, J ) 7.3
Hz); 13C NMR (CDCl3) δ -5.5, -5.4, -5.0, -5.0, -4.6, 17.9,
18.3, 25.7, 25.9, 42.2, 52.9, 61.7, 69.9, 86.6, 87.6, 94.2, 144.1,
152.9, 154.7 (br), 162.0 (br); MS m/z calcd for C23H44N215NO6-
+
Si2 5115.2739, found 515.2757.
En zym a tic Tr ea tm en t of Oligod eoxyn u cleotid es (7a -c
a n d 8) w ith Sn a k e Ven om P h osp h od iester a se a n d Al-
k a lin e P h osp h a ta se. Oligodeoxynucleotides (7a -c and 8)
(0.5 A260) were each dissolved in 0.05 M Tris-HCl (pH 8.0, 200
mL) containing 0.01 M MgCl2, and 10 mL of SVPD (10 unit,
1.0 unit/mL) in glycerin-water (1:1, v/v) was added. The
resulting mixture was incubated at 37 °C for 4 h and heated
at 90 °C for 2 min. To the resulting mixture was added 20 µL
of alkaline phosphatase (20 unit, 1.0 unit/µL). After being
incubated for 3 h, the mixture was heated at 90 °C for 2 min.
The mixture was analyzed by reversed-phase HPLC.
(br); MS m/z calcd for C21H41N15NO5Si2 458.2524, found
+
458.2537.
[3-15N]-3′,5′-O-Bis(ter t-bu tyld im eth ylsilyl)-2′-d eoxycy-
tid in e (22). Compound 21 (500 mg, 1.1 mmol), tetrazole (154
mg, 2.2 mmol), diphenyl phosphate (330 mg, 1.32 mmol), and
tosyl chloride (419 mg, 2.2 mmol) were dissolved in pyridine
and stirred at room temperature for 2 days. Afterward, concd
NH3 (2 mL) was added to the mixture. After being stirred for
2 h, the mixture was concentrated to half volume, diluted with
CHCl3, and washed three times with 5% NaHCO3 (aq). The
aqueous layer was back-extracted with CHCl3. The organic
layer and the washings were combined and dried over Na2-
SO4, filtered, and concentrated to dryness under reduced
pressure. The residue was chromatographed on a column of
silica gel (80 g) with CHCl3, applying a gradient of methanol
(1-2%) to give 22 (436 mg, 87%) as colorless foam: 1H NMR
(CDCl3) δ 0.05 (6H, s), 0.10 (6H, s), 0.88-0.92 (18H, m), 2.04-
2.13 (1H, m), 2.38-2.47 (1H, m), 3.74-3.95 (3H, m), 4.33-
Ack n ow led gm en t. This work was supported by a
Grant from the “Research for the Future” Program of
the J apan Society for the Promotion of Science (J SPS-
RFTF97I00301) and a Grant-in-Aid for Scientific Re-
search from the Ministry of Education, Culture, Sports,
Science and Technology, J apan.
J O010813L