Synthesis and hybridization properties of oligodeoxynucleotides incorporating 2-N-carbamoylguanine derivatives as guanine analogs

Article abstract of DOI:10.1016/j.tetlet.2007.05.038

2-N-Carbamoyldeoxyguanosine and its derivatives were synthesized and incorporated into ODNs. The T_m analyses revealed higher selective base recognition ability of 2-N-carbamoylguanine than that of guanine. The new guanine analog must be useful

Full text of DOI:10.1016/j.tetlet.2007.05.038

Tetrahedron Letters 48 (2007) 5325–5329
Synthesis and hybridization properties of
oligodeoxynucleotides incorporating 2-N-carbamoylguanine
derivatives as guanine analogs
Takeshi Sasami,a,b Yoko Odawara,a,b Akihiro Ohkubo,a,b Mitsuo Sekinea,b, and
*
Kohji Seiob,c
aDepartment of Life Science, Tokyo Institute of Technology, 4259 Nagatsuta-cho, Midori-ku, Yokohama 226-8501, Japan
bCREST, JST (Japan Science and Technology Agency), 4259 Nagatsuta-cho, Midori-ku, Yokohama 226-8501, Japan
cFrontier Collaborative Research Center, Tokyo Institute of Technology, 4259 Nagatsuta-cho, Midori-ku, Yokohama 226-8501, Japan
Received 30 March 2007; revised 2 May 2007; accepted 10 May 2007
Available online 16 May 2007
Abstract—2-N-Carbamoyldeoxyguanosine and its derivatives were synthesized and incorporated into ODNs. The Tm analyses
revealed higher selective base recognition ability of 2-N-carbamoylguanine than that of guanine. The new guanine analog must
be useful for the development of functional oligodeoxynucleotides capable of precise base recognition.
Ó 2007 Published by Elsevier Ltd.
The highly selective hybridization of oligodeoxynucleo-
tides (ODNs) to the complementary ODN is the most
important property which can be applied to hybridiza-
tion-based technologies such as antisense DNA,1 DNA
chips,2 and PCR techniques.3 However, it is well known
that guanine can form not only the Watson–Crick G–C
base pair but also the rather stable wobble G–T, G–A
and G–G mismatch base pairs in canonical B-type
duplexes.4 Such mismatch formation could be the
reason for the inaccuracy of hybridization-based
technologies. Therefore, if the selective hybridization
could be achieved by use of modified nucleobases
instead of natural ones, ODN probes incorporating such
modified bases must improve the accuracy of DNA
microarray detection, and must be useful for the
above-mentioned techniques.
pair formed between c3G and C was less stable than the
canonical G–C pair, and c3G could not be used as the
modified base for the synthesis of the functional oligo-
nucleotides.5 Interestingly, it was also found by us that
modification of the 2-amino function of c3G gave a
new modified guanine base, 2-acetyl-3-deazaguanine
(a2c3G), which can form a Watson–Crick base pair as
stable as the canonical G–C pair. Moreover, the
sheared-type a2c3G–A mismatch and the wobble-type
a2c3G–U mismatch were significantly destabilized so
that the total base discrimination ability of a2c3G
became higher than that of the canonical guanine.7
Although these results indicated the usefulness of
a2c3G as an artificial base applicable to the functional
nucleic acids, simpler modified nucleosides which can
be prepared readily from canonical nucleosides must
be preferable because the synthesis of a2c3G nucleosides
requires multi-step chemical reactions.8–13
We have studied oligoribonucleotides having 3-deaza-
guanine (c3G) in place of the canonical guanine base.5
The studies of hybridization affinity of 20-O-methylated
RNA 10mers incorporating c3G for the target DNA or
RNA oligomers revealed that incorporation of c3G
could destabilize the sheared-type G–A mismatch6
because of the absence of the hydrogen-bond accepter
at the position 3. Unfortunately, the Watson–Crick base
In this Letter, we designed 2-N-carbamoylguanine deriv-
atives as new guanine analogs having high base discrim-
ination ability assuming that the acyl-type substitution
at the amino group can improve the base discrimination
ability of the guanine base as in the case of a2c3G. We
chose carbamoyl-type substituents because they can be
readily introduced to the amino group of guanine and
are stable toward the aqueous ammonia treatment,
which is necessary for removal of the protecting groups
*
Corresponding author. E-mail: msekine@bio.titech.ac.jp
0040-4039/$ - see front matter Ó 2007 Published by Elsevier Ltd.
doi:10.1016/j.tetlet.2007.05.038

5326
T. Sasami et al. / Tetrahedron Letters 48 (2007) 5325–5329
closed-type
O
open-type
O
OTMS
N
O
N
i), ii)
N
NH
NH2
H
O
H
N
N
N
N
N
² E
X
N
R
N
R
N
N
N
OPh
R
H
X
N
N
N
N
H
N
H
N
H
H
2a: R2 = H
2b: R2 = OCH3
H3C
H3C
3a: X = O, R1 = H, R2 = H
72% from 2a
N
R
O
N
3b: X = O, R1 = CH3, R2 = H
78% from 2a
iii)
NH
X
Figure 1. Structure of 9-methyl-cmG, X = O, R = H (1a), -mcmG,
X = O, R = CH3 (1b) and -mscmG X = S, R = CH3 (1c).
R1
N
R
N
N
N
H
3c: X = S, R1 = CH3, R2 = H
48% from 2a
H
3d: X = O, R1 = H, R2 = OCH3
55% from 2b
TBDMSO
O
in oligonucleotide synthesis.14 In addition, intramole-
cular hydrogen bond between the amino group of the
carbamoyl group and the nitrogen atom at position 3 of
guanine can stabilize the ‘open-type’ conformation which
can form a base pair (Fig. 1). It should be noted that with-
out such an intramolecular hydrogen bond, the ‘closed-
type’ conformation, which inhibits the formation of the
Watson–Crick base pair, must be predominant. A similar
conformational change by an intramolecular hydrogen
bond was also proposed in the case of a2c3G. In addition
to the simplest 2-N-carbamoyldeoxyguanosine (cmG), 2-
N-(N-methylcarbamoyl)deoxyguanosine (mcmG) and 2-
N-(N-methylthiocarbamoyl)deoxyguanosine (mscmG),
we also designed 2-N-carbamoyl-20-O-methylguanosine
(cmGm).
R =
TBDMSO
R2
Scheme 1. Reagents and conditions: (3a, 3b, 3d) (i) TMSCl (1.5 equiv),
pyridine, rt, 1 h; (ii) PhOC(=O)Cl (1.5 equiv), pyridine, rt, 4 h; (iii) For
3a 28% NH3aq (5.0 equiv), pyridine, rt, 2 h (from 2a, 3 steps, 68%).
For 3b 40% MeNH2/methanol (5.0 equiv), pyridine, rt, 2 h (from 2a, 3
steps, 78%). For 3d 28% NH3aq (5.0 equiv), pyridine, rt, 2 h (from 2b,
3 steps, 55%). (3c) (i) NaH (1.3 equiv), DMF, 70 °C, 1 h; (ii) methyl
thioisocyanate (3.0 equiv), DMF, 70 °C, 36 h, (from 2a, 3 steps, 48%).
ammonia (5.0 equiv) and methylamine (5.0 equiv) gave
compounds 3a and 3b, respectively. Compound 3c was
obtained by the base-selective reaction of the 30,50-O-
disilylated deoxyguanosine with methyl thioisocyanates
(3.0 equiv) in the presence of sodium hydride (1.3 equiv)
according to the procedure of Zimmerman for the
methyl carbamoyl modification of the guanine ring
(Scheme 1).18
First of all, we carried out theoretical studies on the
open-type and closed-type conformation of the cmG
and its derivatives. 9-Methyl-cmG (1a), 9-methyl-mcmG
(1b), and 9-methyl-mscmG (1c) were used as the models
of their nucleoside derivatives. The structures of the
open-type and closed-type conformers were optimized
at the HF/6-31G** level. The energies of the optimized
structures were evaluated with single-point calculations
at the MP2/6-31G** level.15,16
Nucleosides 3a–3c were converted to their phosphorami-
dite derivatives 7a–7c (Fig. 2 and Scheme 2) by use of the
standard phosphytilation procedure.19 To increase the
solubility of the deoxyribonucleosides, the carbonyl
group of each deoxyguanosine derivative was protected
by the diphenylcarbamoyl (dpc) group.20 On the other
hand, 20-O-methyl phosphoramidite 7d was synthesized
by removal of the TBDMS groups of 3d in 81% followed
by the successive treatments with DMTrCl (64%) and
phosphytilation (58%) under the conditions same as
those shown in the step (ii), (iii) and (iv) in Scheme 2.
In this case, the protection of the 6-O position was not
necessary because of the higher solubility of 7d than
7a–7c.
For both 9-methyl-cmG and 9-methyl-mcmG, the closed-
type conformers were found to be slightly more stable
than the open-type conformers by 1.56 and 1.31 kcal/
mol, respectively, than the corresponding open-type con-
formers. In contrast, the open-type conformer of mscmG
was more stable by 3.31 kcal/mol than the closed-type
one. Although these data suggested that the closed-type
conformers of 9-methyl-cmG and -mcmG were more sta-
ble than their open-type conformers, it should be noted
that Zimmerman and co-workers17,18 previously reported
that 2-N-[N-(n-butyl)carbamoyl]guanine could form sta-
ble supramolecular complexes having four hydrogen
bonds in open-type conformation at least at the nucleo-
side level. These theoretical results together with the
experimental data by Zimmerman et al. indicated that
all of cmG and mcmG and mscmG can form base pairs
in open-type conformation when incorporated in DNA
and RNA duplexes.
The solid-phase synthesis of ODNs and a 20-O-methyl-
RNA containing 2-N-carbamoylguanine derivatives
was carried out in a DNA synthesizer by use of the
standard phosphoramidite method. The ODNs were
released from the polymer supports and deprotected
by treatment with conc. NH3aq for 8 h at ambient
temperature. The 2-N-carbamoyl groups were stable
under such basic conditions. The products were purified
on a C18 cartridge column by the DMTr-ON purifica-
tion method and analyzed by use of anion-exchange
HPLC. The products were characterized by MALDI-
TOF mass.21
The synthesis of 2-N-carbamoylguanosine derivatives
and 2-N-thiocarbamoylguanosine derivatives 3a–3d is
outlined in Scheme 1. Acylation of the 30,50-O-disilyl-
ated derivatives 2a and 2b with phenyl chloroformate
(1.5 equiv) followed by treatment of the resulting 2-N-
phenoxycarbonylguanosine intermediate with aqueous
Next, the hybridization properties of these modified
ODNs, 50-d(CGGCXAGGAG)-30 where X is G, cmG,
mcmG or mscmG, were studied by measuring the Tm

T. Sasami et al. / Tetrahedron Letters 48 (2007) 5325–5329
5327
Odpc
N
O
N
N
N
N
X
NH
O
R1
N
N
N
H
N
N
NH2
DMTrO
DMTrO
O
O
O
O
H
H
7a: X = O, R1 = H
7b: X = O, R1 = CH3
7c: X = S, R1 = CH3
7d
O
P
O
P
OCH3
CN
CN
(
i-Pr)2N
(i-Pr)2N
Figure 2. Phosphoramidite units 7a–7c.
Odpc
Odpc
N
N
N
N
N
N
X
X
i)
ii)
3a-3c
TBDMSO
R2
R2
N
N
H
N
H
N
N
H
N
HO
O
O
H
4a: X = O, R2 = H
4b: X = O, R2 = CH3
4c: X = S, R2 = CH3
OTBDMS
OH 5a: X = O, R2 = H
5b: X = O, R2 = CH3
5c: X = S, R2 = CH3
Odpc
N
N
X
iii)
iv)
7a-7c
R
N
N
N
N
DMTrO
O
H
H
6a: X = O, R2 = H
6b: X = O, R2 = CH3
6c: X = S, R2 = CH3
OH
Scheme 2. Reagents and conditions: (i) DpcCl (1.3 equiv), (i-Pr)2NEt (2.0 equiv), pyridine, rt, 30 min, (4a: 91%, 4b: 94%, 4c: 81%); (ii) 3HFÆNEt3
(3.0 equiv), THF, rt, 10 h, (5a: quant, 5b: 81%, 5c: 80%); (iii) DMTrCl (1.2 equiv), pyridine, rt, 2 h, (6a: 76%, 6b: 87%, 6c: 78%); (iv) (a) (CEO)P[N(i-
Pr)2]2 (1.1 equiv), 1H-tetrazole (0.5 equiv), (i-Pr)2NH (0.5 equiv), CH2Cl2, rt, 2 h (7a: 75%); (b) Cl(i-Pr2N)P(OCE) (2.0 equiv), (i-Pr)2NEt (2.5 equiv),
CH2Cl2, rt, 2 h (7b: 68%); (c) Cl-(i-Pr2N)P(OCE) (1.3 equiv), (i-Pr)2NEt (1.6 equiv), CH2Cl2, rt, 2 h (7c: 47%).
values of the duplexes with the target ODN, 30-
d(GCCGYTCCTC)-50 where Y is C, A, G or T. As
shown in Table 1, the Watson–Crick base pairs formed
between the canonical G and C were as strong as that
between the cmG and C, Tm = 51.5 and 50.8 °C, respec-
tively. Interestingly, the cmG–T mismatch base pair was
more destabilized by than the G–T mismatch pair by
DDTm = ꢀ3.6 °C. Although the cmG–G mismatch base
pair was stabilized in comparison to the G–G mismatch
base pair (DDTm = 2.0 °C), the total selectivity of cmG
Table 1. Tm values (°C)a for 10mer duplexes containing 2-N-carbamoylguanine derivatives 50-d(CGGCXAGGAG)-30/30-d(GCCGYTCCTC)-50
X = G
Tm
(DTm)b
cmG
Tm
mcmG
Tm
mscmG
Tm
(DTm)b
(DTm)b
(DTm)b
{DDTm}c
{DDTm}c
{DDTm}c
Y = C
A
51.5 0.01
50.8 0.31
50.6 0.14
49.6 0.34
36.7 0.30
(ꢀ14.8 0.31)
35.5 0.84
(ꢀ15.3 0.54)
{ꢀ0.5}
36.0 0.53
(ꢀ14.6 0.41)
{+0.2}
35.8 0.78
(ꢀ13.8 0.43)
{+1.0}
G
T
34.8 0.47
(ꢀ16.7 0.46)
36.1 0.44
(ꢀ14.7 0.14)d
{+2.0}
37.3 0.28
(ꢀ13.3 0.20)d
{+3.4}
34.9 0.18
(ꢀ14.7 0.32)
{+2.0}
38.6 0.38
(ꢀ12.9 0.44)d
34.3 0.21
(ꢀ16.5 0.15)
{ꢀ3.6}
37.1 0.42
(ꢀ13.5 0.40)
{ꢀ0.6}
38.7 0.40
(ꢀ10.9 0.25)d
{+2.0}
a The averages and the standard deviations of three to four experiments were obtained under the following conditions: 2 lM duplexes, 10 mM sodium
phosphate buffer, 0.1 M NaCl, 0.1 mM EDTA, pH 7.0.
b The DTm is the difference of Tm of a X–C pair and the corresponding modified X–Y pair.
c The DDTm is the difference of DTm of a X–Y pair and the corresponding modified G–Y pair.
d The selectivity index defined as the DDTm of the smallest absolute value.

5328
T. Sasami et al. / Tetrahedron Letters 48 (2007) 5325–5329
(ꢀ14.7 °C) was higher than that of G (ꢀ12.9 °C), as
from deoxyguanosine. Therefore, unlike the previously
reported a2c3G, cmG and its derivatives are more
convenient for large-scale synthesis. The base pairing
profile of cmG and mcmG was characterized by the
stable Watson–Crick pairing with C, the destabilized
mismatch base pair with T, and the stabilized mismatch
base pair with G. This trend was also observed in 20-O-
methyl-RNA/DNA duplexes. The tendency was quite
similar to that of the previously reported a2c3G in the
20-O-methyl-RNA structures.7 Because a2c3G, cmG
and mcmG derivatives have the acyl-type substituents
on the 2-amino group, the base pairing profiles observed
in these studies might be common to 2-N-acylated guan-
ine derivatives. The mechanism of the destabilization of
the cmG–T and mcmG–T mismatch is not clear. One
plausible explanation is that cmG and mcmG are in
the ‘closed-type’ conformation in the cmG–T and
mcmG–T base pair, and the formation of wobble type
hydrogen bonds is blocked by the carbamoyl groups.
The hypothesis is supported by the fact that mscmG,
whose ‘open-type’ conformation is much more stable
than the ‘closed-type’ conformation, did not show such
base pair selectivity. It is unclear why the stability of the
cmG–G mismatches was increased. The G–G base pair
in the 50-CGA-30/30-GGT-50 sequence was reported to
exist in G(anti)-G(anti) conformation.22 Therefore, the
carbamoyl group might participate in the base pairing
in a unique manner. The structural studies together with
more detailed thermodynamic studies are under way to
clarify such a dynamic mechanism and the generality
in various sequences of the higher base recognition abil-
ity of the 2-N-acylated guanines. These results shown in
this Letter indicated the potential usefulness of oligode-
oxynucleotides incorporating these modified bases as
artificial hybridization probes capable of highly accurate
sequence recognition.
shown by the selectivity index.
The mcmG was similar to cmG in the base pairing
profile such as the base selectivity and the base pair
stabilities. In this case, the matched base pair (X = C)
was as stable as that of the canonical guanine (Tm =
50.6 °C), and the most stable mismatch pair was the
mcmG–G pair which was stabilized in comparison to
the G–G pair (DDTm = +3.4 °C). As the result, the
selectivity was proved to be 13.3 °C, which was between
those of cmG and G.
It should be noted for mcmG that each of the mis-
matches including mcmG was more stable than the cor-
responding mismatch pair incorporating cmG. These
results suggest the possibility that cmG could perform
more precise base recognition than mcmG. In contrast
to cmG and mcmG, the Watson–Crick base pairs of
mscmG–C were less stable than that of the canonical
G–C base pair (Tm = 49.6 °C). In terms of the selectiv-
ity, mscmG was the worst of all the four modified and
unmodified guanines as suggested by the selectivity
index of ꢀ10.9 °C.
The hybridization properties of the duplex of 20-O-
methyl-RNA incorporating cmGm and the target
ODN were analysed. As shown in Table 2, the base dis-
crimination of cmGm was qualitatively similar to but
quantitatively more significant than that in DNA shown
in Table 2. The cmGm–dC (Tm = 59.9 °C) base pair was
as strong as the canonical Gm–dC (Tm = 60.2 °C) and
the cmGm–T mismatch base pair was more destabilized
by DDTm = ꢀ4.3 °C than the Gm–T mismatch pair. The
cmGm–dG mismatch base pair was stabilized in compari-
son to the Gm–dG mismatch base pair (DDTm =
+1.8 °C), the selectivity of cmGm (14.0 °C) was higher
than that of Gm (9.7 °C), as shown by the selectivity
index.
Acknowledgments
In this study, we reported new oligonucleotides incorpo-
rating 2-N-carbamoylguanine base analogs such as
cmG, mcmG, mscmG, and cmGm. Deoxynucleosides
having these modified bases could be readily synthesized
This work was financially supported partly by a Grant-
in-aid from CREST, Japan Science and Technology
Agency. This study was also supported by Industrial
Technology Research Grant Program in ’05 from New
Energy and Industrial Technology Development Orga-
nization (NEDO) of Japan and in part by a grant of
the Genome Network Project from the Ministry of Edu-
cation, Culture, Sports, Science and Technology, Japan.
Table 2. Tm values (°C)a for 10mer duplexes containing 2-N-carb-
amoylguanine 50-(CmGmGmCmXAmGmGmAmGm)-30/30-d(GCCGYT-
a
CCTC)-50
X = Gm
Tm
cmGm
Tm
(DTm)
(DTm) {DDTm}
References and notes
Y = C
A
60.2 0.87
59.9 0.60
1. Agrawal, S. Biochim. Biophys. Acta 1999, 53.
2. Kurian, K. M.; Watson, C. J.; Wyllie, A. H. Pathol. 1999,
187, 267.
40.5 0.28
(ꢀ19.7 0.76)
38.4 0.25
(ꢀ21.5 0.70) {ꢀ1.8}
3. Freeman, W. M.; Walker, S. J.; Vrana, K. E. BioTechni-
ques 1999, 26, 112.
4. SantaLucia, J., Jr.; Allawi, H. T.; Seneviratne, P. A.
Biochemistry 1996, 35, 3555.
5. Seio, K.; Sasami, T.; Tawarada, R.; Sekine, M. Nucleic
Acids Res. 2006, 34, 4324.
6. SantaLucia, J., Jr.; Turner, D. H. Biochemistry 1993, 32,
12612.
G
T
41.7 1.25
(ꢀ18.5 0.26)
50.5 0.09
(ꢀ9.7 0.66)b
44.1 0.54
(ꢀ15.8 0.65) {+2.7}
45.9 0.19
(ꢀ14.0 0.32)b {ꢀ4.3}
a The Nm shows a 20-O-methyl-nucleoside residue.
b The selectivity index defined as the DDTm of the smallest absolute
value.

T. Sasami et al. / Tetrahedron Letters 48 (2007) 5325–5329
5329
7. Seio, K.; Sasami, T.; Ohkubo, A.; Ando, K.; Sekine, M.
J. Am. Chem. Soc. 2007, 129, 1026.
117.4, 117.5, 119.8, 126.9, 126.9, 127.8, 128.0, 128.0,
129.2, 129.9, 130.0, 130.0, 135.4, 135.5, 141.7, 144.4,
150.1, 153.0, 154.4, 154.5, 155.6, 155.6, 158.5; 31P
8. Robins, R. K.; Horner, J. K.; Greco, C. V.; Noell, C. Y.;
Beames, C. G., Jr. J. Org. Chem. 1963, 28, 3041.
9. Cook, P. D.; Rousseau, R. J.; Mian, A. M.; Dea, P.;
Meyer, R. B., Jr.; Robins, R. K. J. Am. Chem. Soc. 1976,
28, 1492.
NMR (CDCl3)
d 150.0, 150.3. MS m/z calcd for
C55H60N9NaO9P+: 1044.4149, found 1044.3683. Com-
1
pound 7c: H NMR (CDCl3) d 1.12–1.29 (14H, m), 2.46
(1H, m), 2.57–2.75 (3H, m), 2.25 (3H, m), 3.30–3.38 (2H,
m), 3.58–3.88 (4H, m), 3.75 (6H, m), 4.27–4.33 (1H, m),
4.67 (H, m), 6.32 (1H, m), 6.79 (4H, m), 7.16–7.43 (19H,
m), 8.10 (11H, m), 8.52 (1H, s), 10.59 (1H, m); 13C NMR
(CDCl3) d 20.3, 20.4, 20.6, 20.6, 24.7, 24.7, 24.8, 32.6,
39.8, 43.4, 43.4, 43.5, 55.3, 58.2, 58.3, 58.4, 58.4, 63.4,
63.6, 73.6, 74.1, 74.2, 84.8, 84.8, 86.1, 86.1, 86.3, 86.7,
113.3, 113.3, 117.5, 117.6, 120.2, 120.3, 127.1, 127.1,
128.0, 128.2, 128.2, 129.4, 130.1, 130.1, 130.2, 135.5,
135.6, 135.7, 141.7, 142.4, 144.5, 150.0, 151.8, 158.7,
158.7, 180.0; 31P NMR (CDCl3) d 150.0, 150.2. MS
m/z calcd for C55H60N9NaO8PS+: 1060.3921, found
1060.3889. Compound 7d: 1H NMR (CDCl3) d 1.02–
1.31 (14H, m), 2.36 (1H, m), 2.62–2.65 (3H, m), 3.30–3.38
(2H, m), 3.31–3.77 (4H, m), 3.71 (6H, m), 4.12–4.19 (1H,
m), 4.32 (1H, m), 4.56 (H, m), 6.00–6.02 (1H, m), 6.79–
6.83 (4H, m), 7.20–7.43 (9H, m), 7.86–7.82 (1H, m); 13C
NMR (CDCl3) d 9.1, 10.1, 23.4, 24.4, 25.4, 26.4, 43.2,
44.2, 45.0, 46.2, 47.3, 53.9, 55.0, 56.2, 57.3, 59.3, 83.0,
84.2, 87.0, 87.2, 112.9, 114.2, 114.2, 117.8, 120.3, 120.4,
126.7, 127.6, 127.6, 127.9, 128.0, 128.0, 128.9, 129.2,
129.2, 129.3, 129.9, 129.9, 131.0, 131.1, 131.2, 134.7,
135.7, 135.8, 135.9, 136.4, 144.7, 144.8, 144.8, 149.3,
157.0, 159.0; 31P NMR (CDCl3) d 151.3, 152.2. MS m/z
calcd for C42H51N8NaO9P+: 865.3414, found 865.3468.
20. Takamura, T.; Tsuchiya, M.; Urakami, K.; Koura, K.;
Sekine, M.; Shinozaki, K.; Miura, K.; Hata, T. J. Am.
Chem. Soc. 1984, 106, 4552.
10. Minakawa, N.; Matsuda, A. Tetrahedron 1993, 49, 557.
11. Tanaka, H.; Hirayama, M.; Matsuda, A.; Miyasaka, T.;
Ueda, T. Chem. Lett. 1985, 589.
12. Tanaka, H.; Hirayama, M.; Suzuki, M.; Miyasaka, T.;
Matsuda, A.; Ueda, T. Tetrahedron 1986, 42, 1971.
13. Seela, F.; Debelak, H.; Andrew, L.; Beigelman, L. Helv.
Chim. Acta 2003, 86, 2726.
14. Schulhof, J. C.; Molko, D.; Teoule, R. Nucleic Acids Res.
1987, 15, 397.
15. Kawahara, S.; Uchimaru, T.; Taira, K.; Sekine, M. J.
Phys. Chem. 2002, 106, 3207.
16. Kawahara, S.; Uchimaru, T. Eur. J. Org. Chem. 2003,
2577.
17. Park, T.; Zimmerman, S. C.; Nakashima, S. J. Am. Chem.
Soc. 2005, 127, 6520.
18. Park, T.; Todd, E. M.; Nakashima, S.; Zimmerman, S. C.
J. Am. Chem. Soc. 2005, 127, 18133.
19. Spectroscopic and mass analysis data of compound 7a–7d.
1
Compound 7a: H NMR (CDCl3) d 1.11–1.33 (14H, m),
2.46 (1H, m), 2.55–2.74 (3H, m), 3.32–3.40 (2H, m), 3.57–
3.87 (4H, m), 3.75 (6H, m), 4.26–4.32 (1H, m), 4.67–4.72
(1H, m), 5.36 (1H, s, br), 6.35 (1H, m), 6.79 (4H, m), 7.17–
7.50 (19H, m), 8.07 (1H, m), 8.53 (1H, s, br); 13C NMR
(CDCl3) d 20.6, 20.7, 20.8, 20.9, 24.9, 25.0, 25.0, 40.2, 43.6,
43.6, 43.7, 43.7, 55.6, 55.6, 58.5, 58.6, 58.6, 58.7, 63.7, 63.8,
73.9, 74.0, 74.4, 74.5, 84.7, 84.8, 86.2, 86.2, 86.4, 86.9,
113.6, 117.8, 117.9, 120.5, 127.3, 127.4, 128.3, 128.4, 128.5,
129.6, 130.3, 130.4, 130.4, 135.8, 135.8, 135.9, 135.9,
142.1, 144.7, 150.4, 153.1, 154.8, 154.9, 155.0, 156.2,
159.0; 31P NMR (CDCl3) d 149.9. MS m/z calcd for
C54H59N9NaO9P+: 1030.3993, found 1030.4001. Com-
21. Sequence, yield and MALDI-TOF mass analyses of
ODNs containing 2-N-carbamoylguanine derivatives, 50-
d(CGGC[cmG]AGGAG)-30; yield 23%, MALDI-TOF
mass [MꢀH]+ calcd for 3158.4, found 3158.6, 50-
d(CGGC[mcmG]AGGAG)-30; yield 26%, MALDI-TOF
mass [MꢀH]+ calcd for 3172.5, found 3172.6, 50-
d(CGGC[mscmG]AGGAG)-30; yield 7%, MALDI-TOF
mass [MꢀH]+ calcd for 3188.8, found 3188.6, 20-O-methyl
50-(CGGC[cmGm]AGGAG)-30; yield 27%, MALDI-TOF
mass [MꢀH]+ calcd for 3460.7, found 3460.1.
1
pound 7b: H NMR (CDCl3) d 1.13–1.27 (14H, m), 2.43
(1H, m), 2.58–2.76 (3H, m), 2.94 (3H, m), 3.35 (2H, m),
3.61–3.84 (4H, m), 3.73 (6H, m), 4.29–4.33 (1H, m), 4.71
(1H, m), 6.35 (1H, m), 6.80 (4H, m), 7.18–7.43 (19H, m),
7.63 (1H, m), 8.09 (1H, m), 8.61 (1H, s); 13C NMR
(CDCl3) d 20.1, 20.2, 20.3, 20.4, 24.5, 24.5, 24.6, 26.7, 39.5,
39.7, 43.2, 43.3, 55.1, 55.1, 58.1, 58.1, 58.2, 58.3, 63.3, 63.4,
73.4, 73.6, 74.0, 74.1, 84.4, 85.7, 85.8, 86.0, 86.5, 113.1,
22. Faibis, V.; Cognet, J. A. H.; Boulard, Y.; Sowers, L. C.;
Fazakerley, G. V. Biochemistry 1996, 35, 14452.