Synthesis of C4-Linked C0- and C2-Imidazole 2′-Deoxyribonucleoside Phosphoramidites and Imidazole Base-Pairing Effects on DNA

Article abstract of DOI:10.1055/s-0034-1378451

The synthesis of C4-linked imidazole C₀- and C₂-2′-deoxyribonucleoside phosphoramidites (dPAs), in which the final phosphitylations are greatly improved by 4,5-dicyanoimidazole-promoted conversion, is described. The respective dPAs a

Full text of DOI:10.1055/s-0034-1378451

PAPER
▌2815
paper
Synthesis of C4-Linked C0- and C2-Imidazole 2′-Deoxyribonucleoside
Phosphoramidites and Imidazole Base-Pairing Effects on DNA
C - and C -Imidazole 2′-Deoxyribonucleoside Phosphoramidites
a
a
a
b
c
0
2
Shinya Harusawa,* Hiroki Yoneyama, Yoshihide Usami, Daisuke Yamamoto, Zheng-yun Zhao
a
Laboratory of Pharmaceutical Organic Chemistry, Osaka University of Pharmaceutical Sciences, 4-20-1 Nasahara, Takatsuki, Osaka
569-1094, Japan
Fax +81(72)6901086; E-mail: harusawa@gly.oups.ac.jp
b
c
Biomedical Computation Center, Osaka Medical College, 2-7 Daigakucho, Takatsuki, Osaka 569-8686, Japan
Centre for Plant Science & Bio-polymer Research, Faculty of Engineering, Computing and Creative Industries, Edinburgh Napier University,
10 Colinton Road, Edinburgh, EH10 5DT, UK
Received: 20.05.2014; Accepted after revision: 17.06.2014
in the active sites of ribozymes.6c,8 Evidence from this ap-
proach indicated that the chemical mechanisms of VS and
hairpin ribozymes involve general acid–base catalysis via
the combination of the specific adenine (A) and guanine
Abstract: The synthesis of C4-linked imidazole C0- and C2-2′-de-
oxyribonucleoside phosphoramidites (dPAs), in which the final
phosphitylations are greatly improved by 4,5-dicyanoimidazole-
promoted conversion, is described. The respective dPAs are suc-
cessfully incorporated into the sequence of a 15-nt DNA, and the (G) nucleobases (e.g., A756 and G638 in the VS ribo-
zyme).8 Of particular interest is that the modified VS ribo-
zyme (G638C2Imz), which is obtained from a two-
abilities of one or two imidazoles to pair with different bases are in-
vestigated through thermal melting (Tm) experiments on the result-
ing DNA duplexes. Furthermore, computational models of the
carbon-elongated homologue [Imz-C2-PA (1c)], shows
imidazole-modified DNAs are found to be in good agreement with
the results of the thermal melting experiments.
significantly greater catalytic activity than G638C0Imz,
suggesting that the flexible C2-methylene spacer is a bet-
Key words: imidazole, phosphoramidite, C-deoxyribonucleoside,
ter structural mimic of the purine nucleobase.6c In addi-
base pairing, DNA
tion, starting from tetrazol(Tez)-C0- and -C2-PAs 2a and
2b, C5-linked C0- and C2-tetrazoles were incorporated
successfully into the VS ribozyme substrate to determine
Imidazole is present in many important biological mole-
more specifically which nucleobases of the ribozymes
cules,1 the most pervasive of which are histidine and his-
function as the acid or base.9 Ribose-(CH2)n-Imz or -Tez
tamine. The former plays a vital role in many proteins and
species can, in principle, be incorporated at any site of
enzymes,2 while the latter is an endogenous biogenic
RNA sequences through chemical synthesis.10 Functional
amine having a number of pathophysiological roles.3 Fur-
DNA oligonucleotides, in which a modified nucleobase
thermore, imidazole is present in many pharmaceuticals
bearing an alkylimidazole11 or an imidazole N-deoxyribo-
such as fungicides and antifungal, antiprotozoal and anti-
nucleoside5 was incorporated, have already been reported.
hypertensive medications.4
However, to the best of our knowledge, modified oligonu-
Imidazole, with a pKa of 7.1, is employed extensively as a
building block or as a Brønsted acid and base catalyst in
organic chemistry.1 However, imidazole-based com-
pounds have not fulfilled important roles as chemical
probes in nucleic acid chemistry, because to date there are
few imidazole-intercalating agents.5 From this viewpoint,
we previously reported the efficient synthesis of C4-
linked C0- to C3-imidazole ribonucleoside phosphoramid-
ites [Imz-Cn-PAs (1a–d)], as shown in Figure 1.6 During
the synthesis of these phosphoramidites (PAs), pivaloyl-
oxymethyl (POM)6a,b and cyanoethyl (CE)6c groups were
employed successfully as protecting groups for the imid-
azole τ-nitrogen and 2′-hydroxy functions, respectively.
Using Imz-Cn-PAs 1,6d we have developed a novel che-
mogenetic approach in the studies of the catalytic mecha-
nism of Varkud satellite (VS) and hairpin ribozymes,7,8
where conventional nucleobases were replaced by imidaz-
ole as a powerful tool to probe general acid–base catalysis
cleotides substituted with an imidazole C-deoxyribonu-
cleoside are not yet known.12,13 Therefore, if C4-linked
Imz-Cn-PAs based on 2′-deoxyribonucleosides could be
synthesised, as an extension of the diversification of on-
going heterocyclic phosphoramidite synthesis in our
group, they would greatly facilitate the chemogenetic ap-
proach to the investigation of various DNA functions.
From systematic studies, we herein report the practical
synthesis of Imz-C0-2′-deoxyribonucleoside phosphora-
midite (Imz-C0-dPA) (3a) and the elongated novel C2-ho-
mologue (Imz-C2-dPA) (3b) as a purine base mimic
(Figure 2, A). Phosphoramidites 3a and 3b were incorpo-
rated successfully into the centre of a 15-nt DNA se-
quence, and the abilities of the imidazole pairing with
different bases were investigated by measuring the melt-
ing temperature (Tm) of the formed duplexes (Figure 2, B).
Alternatively, two imidazole moieties were incorporated
into the same sequence (Figure 2, B). Furthermore,
Amber (Assisted model building with energy refinement)
models of their imidazole-incorporated DNA duplexes
were constructed to interpret the results of the Tm experi-
ments.
SYNTHESIS 2014, 46, 2815–2825
Advanced online publication: 31.07.2014
0
0
3
9
-
7
8
8
1
1
4
3
7
-
2
1
0
X
DOI: 10.1055/s-0034-1378451; Art ID: ss-2014-f0308-op
© Georg Thieme Verlag Stuttgart · New York

2816
S. Harusawa et al.
PAPER
O
yields and reaction times in brackets show the results of
the previous synthesis).6b Unfortunately, there was a seri-
ous limitation during the final step. The phosphitylation of
7 with 2-cyanoethyl N,N,N′,N′-tetraisopropylphosphor-
amidite (CETIP-PA) in the presence of diisopropylammo-
nium tetrazolide (DIPT)14a proceeded slowly (46 h) in 1,2-
dichloroethane (DCE) at 40 °C, and as a result, Imz-C0-
dPA (3a) was formed in 14–32% yields at most (Table 1,
entry 1). Moreover, the isolation of 3a from the reaction
mixture was troublesome owing to the presence of by-
products with similar polarities, resulting in low and vari-
able yields. Significant effort was made to improve the re-
action conditions, and additional experiments for the
phosphitylation step were carried out in the present study.
It was found that a shorter reaction time (22 h) provided
3a (48%) with recovery of 7 (23%) (Table 1, entry 2), but
a prolonged reaction (>22 h) resulted in generation of the
by-products. Further, the phosphitylation of 7 using the
alternative reagent, 2-cyanoethyl N,N-diisopropylchloro-
phosphoramidite, in the presence of N,N-diisopropyl-
ethylamine (DIPEA) proceeded smoothly at room
temperature in one hour, but column chromatographic pu-
rification of the crude reaction mixture was not easy ow-
ing to the hydrolysed excess of the phosphitylating
reagent, affording a moderate yield of 3a (63%) (Table 1,
entry 3). On the other hand, although diisopropylammoni-
um tetrazolide is used as a common reagent in the P(III)–
N activation strategy, 4,5-dicyanoimidazole (DCI) is an
alternative to tetrazole-based activators.14b It is less acidic
(pKa = 5.2), but more nucleophilic than tetrazole
(pKa = 4.8) and its derivatives. Thus, when the 3′-hydroxy
function of compound 7 was subjected to CETIP-PA (2
equiv) in the presence of 4,5-dicyanoimidazole (1.2
equiv) at 40 °C in 1,2-dichloroethane, the reaction pro-
ceeded cleanly to produce 3a (94%) in only one hour (Ta-
ble 1, entry 4). The crude product could be purified easily
by column chromatography on basic silica gel to yield 3a
as a white foam, in contrast to the more time-consuming
purification of the previous reaction6b using diisopro-
pylammonium tetrazolide (Table 1, entry 1). In addition,
although the previously reported debenzylation of 5 (re-
flux, 3 h) afforded diol 6 in moderate yield (<51%),6b the
present reaction was completed quantitatively in only 30
minutes under microwave (MW) activation (Scheme 1).9
Accordingly, the overall yield of Imz-C0-dPA (3a) in the
present synthetic study was 85% from deoxyribonucleo-
side 4, which is a 6- to 13-fold increase compared to the
aforementioned procedure, while the overall reaction time
was reduced 17-fold.
N
N
O
( POM )
( CE )
DMTO
(CH2)n
O
i-Pr2N
O
O
P
CN
O
NC
1a n = 0
1b n = 1
1c n = 2
1d n = 3
R1
N
N
N
N
DMTO
(CH2)n
O
i-Pr2N
O
O
P
R2
O
NC
2a R1 = MePOM, R2 = CE, n = 0
2b R1 = POM, R2 = TBDMS, n = 2
Figure 1 Structures of C4-linked imidazole- and C5-linked tetra-
zole-Cn-ribonucleoside phosphoramidites employed for solid-phase
RNA synthesis
O
A
N
N
O
DMTO
(CH2)n
O
i-Pr2N
O
P
O
NC
3a n = 0
3b n = 2
B
5'-T G G A C T C IMn C T C A A T G-3'
3'-A C C T G A G G A G T T A C-5'
X
and
5'-T G G A C IMn C T C IMn C A A T G-3'
3'-A C C T G G A G G T T A C-5'
X
X
Figure 2 (A) Structures of imidazole-Cn-2′-deoxyribonucleoside
phosphoramidites used for solid-phase DNA synthesis. (B) DNA du-
plexes containing one imidazole (IM) modification at the centre, and
two IM modifications separated by three bases. IMn = IM (n = 0,
from 3a) or IM2 (n = 2, from 3b; imidazole with a C2-linker). X = A,
G, C, T, IM, IM2 or Ab (Abasic)
Modified Synthesis of Imz-C0-dPA (3a)
Synthesis of Imz-C2-dPA (3b)
We previously reported the synthesis of Imz-C0-dPA
(3a)6b from 4(5)-(3,5-di-O-benzyl-2′-deoxy-β-D-ribofura-
nosyl)-1H-imidazole (4),13b involving the introduction of
a pivaloyloxymethyl group at the imN position of 4 (3 h,
90%), debenzylation of 5 to give 6 (3 h), dimethoxy (DM)
tritylation of diol 6 yielding 7 [overnight, 51% from 5 (2
steps)], and phosphitylation of the 3′-hydroxy compound
7 (46 h, 14–32%), as shown in Scheme 1 (the percentage
Starting from 3,5-di-O-benzyl-1-cyano-2-deoxy-β-D-ri-
bose (8),15 the selective synthesis of β-aldehyde 916 was
carried out in two steps: treatment with a base followed by
diisobutylaluminum hydride (DIBAL-H) reduction to
yield the aldehyde (Scheme 2).6c The Wittig reaction of al-
dehyde 9 with N-1-trityl-4-imidazolylmethylphosphoni-
um chloride (10)6c in toluene (–14 °C to r.t.) afforded the
E-alkene product 11 (40%) and the corresponding Z-iso-
Synthesis 2014, 46, 2815–2825
© Georg Thieme Verlag Stuttgart · New York

PAPER
C0- and C2-Imidazole 2′-Deoxyribonucleoside Phosphoramidites
2817
N
NH
N
N
POM
1) Pd(OH)2/C (20%)
cyclohexene, EtOH
MW, 100 °C, 0.5 h
N
N
POM
NaH
POMCl
BnO
BnO
HO
O
O
O
BnO
413b
BnO
5 [90%]6b
HO
6
N
N
POM
N
N
POM
DMTO
O
DMTO
(i-Pr2N)2POCH2CH2CN (2.0 equiv),
2) DMTCl, Et3N,
DMAP, Py, r.t., 2 h
O
DCI (1.2 equiv)
i-Pr2N
O
P
OH
DCE, 40 °C, 1 h
(Table 1, entry 4)
O
7 (quant from 5)
[51% from 5]6b
NC
3a (94%)
[14–32%]6b
Scheme 1 Modified synthesis of Imz-C0-dPA (3a) promoted by 4,5-dicyanoimidazole
mer 11 (16%), which were separated easily by column reoisomers at 147.9 ppm and 148.4 ppm (CDCl3). Mass
chromatography, while in tetrahydrofuran (–78 °C to r.t.) spectrometric analysis of 3b was performed using a ma-
there was little E/Z preference for the newly formed dou- trix system [triethanolamine–NaCl] on a fast atom bom-
ble bond [(E)-11 (26%), (Z)-11 (33%)]. Acid treatment of bardment mass spectrometer (FABMS) equipped with a
each isomer to remove the trityl group, followed by the in- double-focusing mass spectrometer, which was recently
troduction of a pivaloyloxymethyl group provided (E)-12 developed in our laboratories for various labile phosphor-
(54%) and the (Z)-isomer 12 (82%). Using (E)-12, we at- amidites.17 This technique successfully revealed the sodi-
tempted the synthesis of a constrained E-vinyl-C2-PA, but um ion adduct of 3b at m/z 851.4129, confirming the
debenzylation (Li naphthalenide, AlCl3–anisole, or BCl3) chemical formula as C46H61N4O8PNa.
whilst retaining the double bond of (E)-12 failed to afford
Base Pairing Experiments on Imidazole-Containing
Oligomers
the desired intermediate 14. Reductive debenzylation9c of
combined isomers (E/Z)-12, under microwave irradia-
For the assessment of the imidazole base-pairing abilities
with various bases, oligonucleotides containing imidazole
moieties were prepared via automated DNA synthesis
from the standard phosphoramidites (dA, dG, dC and T)
and imidazole deoxyribonucleoside phosphoramidites 3a
and 3b. Two experiments were designed. Firstly, imidaz-
ole moieties were introduced at the centre of a 15-nt du-
plex to replace a thymine–adenine (T/A) base pair [Figure
tion,9 involving simultaneous saturation of the double
bond followed by dimethoxytritylation afforded N-POM-
imidazole-C2-deoxyribonucleoside 13 (67%). Finally,
treatment of 13 with CETIP-PA in the presence of 4,5-di-
cyanoimidazole again proceeded smoothly to afford Imz-
C2-dPA (3b) (96%) as a white foam. The 31P NMR spec-
troscopic data of 3b indicated a pair of phosphorus diaste-
Table 1 Investigation of the Phosphitylation Conditions for the Synthesis of Compound 3a
N
N POM
DMTO
N
N
POM
phosphitylation
O
DMTO
O
i-Pr2N
O
P
OH
O
7
NC
3a
Entry
Conditions
Yield (%)
1
2
3
4
(i-Pr2N)2POCH2CH2CN (1.1 equiv), DIPTa (0.5 equiv), 40 °C, 46 h, DCE
(i-Pr2N)2POCH2CH2CN (1.1 equiv), DIPT (0.5 equiv), 40 °C, 22 h, DCE
i-Pr2N(Cl)POCH2CH2CN (2.0 equiv), DIPEA (2.5 equiv), r.t., 1 h, CH2Cl2
(i-Pr2N)2POCH2CH2CN (2.0 equiv), DCIb (1.2 equiv), 40 °C, 1 h, DCE
14–326b
48c
63
94
a DIPT = Diisopropylammonium tetrazolide.
b DCI = 4,5-Dicyano-1H-imidazole.
c Recovery of 7 (23%).
© Georg Thieme Verlag Stuttgart · New York
Synthesis 2014, 46, 2815–2825

2818
S. Harusawa et al.
PAPER
N
N
Tr
Cl
N
N
Tr
H
Ph3P
a (quant)
b (93%)
BnO
CN
BnO
BnO
O
10
c
O
O
d, e
O
BnO
BnO
BnO
(E)-11 (40%)
(Z)-11 (16%)
8
9
N
N POM
N
N POM
N
N POM
DMTO
f, g
DMTO
h
BnO
O
O
O
(E/Z)-12
i-Pr2N
O
OH
13 (67%)
BnO
P
(E)-12 (54%)
(Z)-12 (82%)
O
NC
3b (96%)
[(E)-12]
N
N POM
HO
O
HO
14
Scheme 2 Synthesis of Imz-C2-dPA 3b. Reagents and conditions: (a) NaOMe, MeOH, r.t., 15 h; (b) DIBAL-H, toluene, –78 °C, 0.5 h; (c) 10,
n-BuLi, toluene, –14 °C, then r.t., 2 h; (d) aq HCl, EtOH, reflux, 1 h; (e) NaH, POMCl, THF, r.t., 1 h; (f) 20% Pd(OH)2/C, cyclohexene, EtOH,
MW, 100 °C, 0.5 h; (g) DMTCl, Et3N, DMAP (cat.), r.t., 2 h; (h) CETIP-PA, DCI, DCE, 40 °C, 1 h.
2 (B) and Table 2].18 By varying the base directly opposite (1) In the first row of Table 3, it is seen that imidazole
the imidazole in the probe strand, the impact from the base (IM) pairs to a greater extent with purine bases (A or G)
pairing could be assessed, while the two arms of the du- than with pyrimidine bases (C and T), possibly through
plexes on each side with fully matched sequences could hydrogen bonding. The fact that the same low Tm value
hold the duplexes together. Next, two thymines separated (43.5 °C) was found for Abasic with Abasic (Ab, sugar
by three nucleobases were replaced by two imidazoles at moiety only) as for IM with IM (43.5 °C) indicates that
centrally symmetric positions of the probe strand (Figure there is little interaction between the IM moieties in the
2, B), and again, the opposite bases on the target strand context of the duplex. In addition, IM with any base has a
were changed simultaneously to the same base, in the higher Tm (above 43.5 °C) indicating that there are indeed
hope of obtaining further information. In both cases, the interactions between IM and the bases. However, all the
imidazole moieties only replaced the T bases in the probe modifications show lower Tm values than those of the ful-
strand. To assist the understanding of the interactions, we ly matched T/A (56 °C) duplex.
also prepared fully matched sequences with some modifi-
(2) In the case of the imidazole with a C2 linker (IM2),
cations at the centre. These oligonucleotides were synthe-
there was not much difference observed in its pairing with
sised by solid-phase chemistry based on the
different bases. IM2 paired with itself had a higher Tm of
phosphoramidite approach.19 Stepwise coupling yield in
48.5 °C. As we describe later, the modelling in the next
over 99% were obtained for the phosphoramidites (Imz-
section demonstrates some stacking interactions.
dPAs) 3a and 3b as monitored from trityl release. All the
(3) When two IMs were introduced, as in the third row of
prepared sequences are presented in Table 2. Details of
Table 3, the Tm values show some additional effects.
the deprotection, purification, and analysis can be found
Again, it was found that IM interacts more strongly with
in the Supporting Information. The thermal stability of
purine bases than pyrimidine bases, consistent with the
each duplex was determined by variable temperature
single IM substitution. However, two IM substitutions
UV/Vis spectroscopy by monitoring the absorbance
brought about more disruptions to duplex formation,
changes at 260 nm as a function of temperature. The melt-
hence lower Tm values were observed.
ing temperature of a fully matched duplex with a T/A base
pair in the centre was used as the control in each run of a
set of experiments. The results are presented in Table 3.
The following conclusions can be drawn:
(4) With two IMs having C2 linkers incorporated into one
strand, the corresponding duplexes showed pairing with
A, G, and T, and afford almost the same Tm values, while
pairing with cytosine (C) shows a higher value, however,
Synthesis 2014, 46, 2815–2825
© Georg Thieme Verlag Stuttgart · New York

PAPER
C0- and C2-Imidazole 2′-Deoxyribonucleoside Phosphoramidites
2819
Table 2 Sequences of the Synthesised DNA Oligonucleotides
Table 3 Thermal Melting (Tm; °C) Data of the Duplexes Formed by
Probes and Targetsa
Oligonucleotide
Control probe
Control or A target
IM probea
IM target
Sequence (5′ to 3′)
Target
T/A
A
G
C
T
Ab
Own
TGGACTCTCTCAATG
control
(IMn)
Probe
IM
CATTGAGAGAGTCCA
TGGACTC(IM)CTCAATG
CATTGAG(IM)GAGTCCA
TGGAC(IM)CTC(IM)CAATG
TGGACTC(IM2)CTCAATG
CATTGAG(IM2)GAGTCCA
TGGAC(IM2)CTC(IM2)CAATG
TGGACTC(Ab)CTCAATG
CATTGAG(Ab)GAGTCCA
CATTG(Ab)GAG(Ab)GTCCA
CATTGAGGGAGTCCA
CATTGAGCGAGTCCA
CATTGAGTGAGTCCA
CATTGGGAGGGTCCA
CATTGCGAGCGTCCA
CATTGTGAGTGTCCA
56
56
56
56
–
47
47
44
45
44
43.5
IM2
2IM
2IM2
Ab
46
47.5
30.5
34
47.5
26.5
37.5
–
47
44.5 48.5
36.5
32.5
–
27.5
33.5
–
25
–
–
–
2IM probe
IM2 probe
IM2 targeta
2IM2 probe
Ab probeb
Ab target
32
–
43.5
a Reaction conditions: 5.0 μM of each oligonucleotide in NaCl (100
mM), phosphate (10 mM), and pH 7.0 buffer. IM and IM2 refer to 3a
and 3b incorporated into sequences, and 2IM and 2IM2 refer to two
identical modifications in a sequence. Likewise, in the target, for ex-
ample, A stands for two As opposite the imidazole modifications in
the target sequence with respect to 2IM and 2IM2. Ab refers to Abasic
where there is no base, and only a sugar is present. Own refers to itself
(i.e., the same modification pairing with itself in the centre of a se-
quence).
2Ab target
G target
C target
structures of the imidazole-modified duplexes with that of
the fully-matched control duplex, as shown in Figure 3
(see the experimental section). The modelling indicated
predicted duplex structures: the T–A control duplex
(Tm = 56 °C), IM probe–A target (Tm = 47 °C), IM probe–IM
target (Tm = 43.5 °C), IM2 probe–A target (Tm = 46 °C),
and IM2 probe–IM2 target (Tm = 48.5 °C). In the control
15-nt DNA (Figure 3, A), two hydrogen bonds (3.39 Å
and 2.87 Å) between T and A were present in the eighth
base pair, and the aromatic rings of the eighth base pair
were parallel-stacked between the seventh and ninth base
pairs. Substitution of T with IM caused the breakdown of
these hydrogen bonds (Figure 3, B), but the stacking inter-
action appeared to be maintained without distortion of the
whole double-helix structure. In the case of the IM probe–
IM target (Figure 3, C), there was a localised breakdown
of the stacking interactions, and the imidazole rings of the
IMs were fixed by hydrophobic interactions with the sug-
ar moieties. In the IM2 probe–A target (Figure 3, D), an
irregular hydrogen bond (2.86 Å) between IM2 and G ad-
jacent to intrinsic A was formed, decreasing the regular
stacking interaction of the control duplex. Finally, in the
IM2 probe–IM2 target duplex (Figure 3, E), a hydrogen
bond (2.84 Å) between the imidazole rings of the IM2
units and the regular stacking interaction were recon-
structed in a similar way to those in the control duplex.
These predicted molecular structures of IM/IM2-modified
DNA duplexes are consistent with the decrease in the Tm
values in each case.
T target
2G target
2C target
2T target
a IM or IM2 represents the imidazole (3a) or the imidazole with a C2
linker (3b) incorporated into the DNA sequence via phosphoramidite
chemistry.
b Ab refers to Abasic, namely 2-deoxyribose without a base.
O
O
Ab
O
O
P
O
O
the reason for this is unclear. Perhaps, imidazole C2
linkers might be mimics of G. As it has been shown that
C2-linked imidazole is similar structurally to a purine
base, the N–H of imidazole would function similarly to G
and would donate a proton to form an imino linkage for
G/C pairs in the context of the DNA duplex. The fact that
in all situations the Tm values were higher for all bases
than with no bases, as in the case of Ab as a target, indi-
cates that the imidazoles do indeed interact with bases.
Computational Models of Imidazole (IM/IM2)-Con-
taining DNA
The theoretical values of the interaction energies of the
base pairs in these duplex structures are summarised in
Table 4. The value for the eighth pair in each structure
seemed to coincide with the above Tm values and the char-
acter of the molecular structure. From these results, we es-
timate the following order of stability: T–A > IM2–IM2 >
IM2–A = IM–A > IM–IM. The values for the three base
DNA duplex stability is a complicated issue, and might be
affected by electrostatic, base stacking, and hydrophobic
interactions, etc. Therefore, to understand the characteris-
tic decrease in Tm values of modified DNA caused by the
incorporation of IM or IM2, we compared four molecular
© Georg Thieme Verlag Stuttgart · New York
Synthesis 2014, 46, 2815–2825

2820
S. Harusawa et al.
PAPER
pairs from the seventh to the ninth showed the same ten- with the base ring moiety in the IM2–A and IM–A struc-
dency, but the difference in the average per pair was small tures as described above. We postulated that this relieved
compared with that of the eighth pair. In terms of the in- stability was also shown in other base pair partial struc-
teraction energy values for the five base pairs from the tures without hydrogen bonds, as is the case with a di-
sixth to the tenth, the difference became smaller, and the fluorotoluyl ribonucleoside in an RNA duplex structure.20
duplex stability was out of order, except for that of IM– Overall, the attack of water molecules on the localised un-
IM. Based on these findings, it was thought that the local- stable region might then spread along adjacent base pairs
ised lack of stability brought about by IM/IM2 incorpora- at a high temperature, similar to opening a zipper in both
tion might be relieved by the molecular structure of the directions.
surrounding base pairs, especially by stacking interactions
Figure 3 Predicted molecular structures of IM/IM2-incorporated DNA duplexes: (A) T–A control duplex; (B) IM probe–A target; (C) IM
probe–IM target; (D) IM2 probe–A target; (E) IM2 probe–IM2 target. Oxygen, nitrogen, and phosphorus atoms are shown in red, blue and
purple, respectively, and the carbon atoms of IM/IM2, the eighth nucleotide, G interacting with IM2(D) and other residues are shown in green,
yellow, light blue, and white, respectively. Some interatomic distances related to the eighth residue are shown with green numbers, and the
hydrogen bonds formed between atoms are shown as balls.
Synthesis 2014, 46, 2815–2825
© Georg Thieme Verlag Stuttgart · New York

PAPER
C0- and C2-Imidazole 2′-Deoxyribonucleoside Phosphoramidites
2821
Chromatorex NH-DM 1020 (Fuji Silysia Chemical Ltd.) was used
for basic (NH) silica gel chromatography. TLC was performed on
precoated TLC plates (Merck 60F254). 1H (400 MHz) and 13C NMR
(100 MHz) spectra were recorded on an Agilent 400-MR-DD2
spectrometer in CDCl3 with tetramethylsilane (TMS) as an internal
standard. 31P NMR spectra were recorded at 121 MHz on a Varian
Mercury-300 spectrometer and the chemical shifts were measured
relative to 85% H3PO4 as an external standard. High-resolution
mass spectra were obtained using a JEOL JMS-700 mass spectrom-
eter in positive-ion mode, with 3-nitrobenzyl alcohol (NBA) or
triethanolamine (TEOA)/NaCl as the matrix.
Table 4 Interaction Energies of Base Pairs in Imidazole-Substituted
DNA Duplexesa
Interaction energy (kcal/mol)
8th Pair
8th Pair only
7–9th Pairsb
6–10th Pairsb
DNA (T–A)
–13.604
–88.446
(–29.482)
–133.161
(–26.632)
IM–A
–3.871
–0.311
–3.040
–9.357
–80.99
(–26.999)
–138.303
(–27.661)
[4-(5-O-DMT-2-deoxy-D-β-ribofuranosyl)imidazolyl]methyl
2,2-Dimethylpropionate (7)
IM–IM
IM2–A
IM2–IM2
–67.253
(–22.418)
–125.354
(–25.071)
A mixture of compound 56b (149 mg, 0.31 mmol), 20% Pd(OH)2/C
(90 mg) and cyclohexene (0.95 mL, 9.30 mmol) in EtOH was ex-
posed to MW irradiation at 100 °C for 0.5 h. After filtration through
Celite, the filtrate was evaporated to afford crude diol 6 as an oil,
which was co-evaporated three times with pyridine (1 mL) and then
dissolved in anhydrous pyridine (4 mL). DMTCl (159 mg, 0.47
mmol), Et3N (64 μL, 0.47 mmol) and DMAP (4 mg, 0.03 mmol)
were added to the pyridine solution, and the mixture was stirred at
r.t. for 2 h. MeOH (1 mL) was added to the mixture which was then
evaporated. The residue was subjected to chromatography on an
NH-silica gel column (EtOAc) to afford 76b (190 mg, quant.) as an
amorphous product.
–79.488
(–26.496)
–136.365
(–27.273)
–81.696
(–27.232)
–129.744
(–25.949)
a The theoretical values of the interaction energies (kcal/mol) of the
8th pair, 7–9th pairs, and 6–10th pairs in the control DNA duplex as
well as the 8th-modified models (IM–A, IM–IM, IM2–A and IM2–
IM2) are listed. These values were calculated using the MOE package
under the same conditions of structural optimisation.
b Values in parentheses are the average values per pair.
[4-{5-O-DMT-2-deoxy-3-O-[2-cyanoethyl-(N,N-diisopropyl-
amino)phosphoramidyl]-β-D-ribofuranos-1-yl}imidazol-
yl]methyl 2,2-Dimethylpropionate (3a)
Compound 7 (60 mg, 0.10 mmol) was dissolved in anhydrous DCE
(3 mL), and DCI (14 mg, 0.12 mmol) and CETIP-PA (63 μL, 0.20
mmol) were added. The resulting mixture was stirred at 40 °C for 1
h and then evaporated. The residual oil was subjected to chromatog-
raphy on NH-silica gel (EtOAc–hexane, 40:60 v/v) to afford 3a6b
(75 mg, 94%) as a white foam.
In conclusion, a modified synthesis of C4-linked Imz-C0-
dPA (3a) was achieved via phosphitylation promoted by
4,5-dicyanoimidazole, significantly improving the overall
yield and reaction times. Further, novel Imz-C2-dPA (3b)
as a purine base mimic was synthesised efficiently. By us-
ing phosphoramidites 3a and 3b, C0- and C2-imidazoles
were incorporated, with over 99% stepwise coupling
yields into the sequence of a 15-nt DNA, and the imidaz-
ole pairing abilities with different bases were examined
through Tm experiments of the duplexes expressed as Tm
values. These preliminary results indicated that: (i) du-
Synthesis of 3a Using 2-Cyanoethyl N,N-Diisopropylchloro-
phosphoramidite (Table 1, Entry 3)
DIPEA (85 μL, 0.50 mmol) and 2-cyanoethyl N,N-diisopropylchlo-
rophosphoramidite (89 μL, 0.40 mmol) were added sequentially to
compound 7 (120 mg, 0.20 mmol) in anhydrous CH2Cl2 (10 mL).
After being stirred at r.t. for 1 h, the resulting mixture was diluted
plexes containing C0- or C2-imidazole showed lower Tm with CH2Cl2 (50 mL), and washed with sat. NaHCO3 solution (30
mL) followed by H2O (30 mL). The organic layer was dried over
MgSO4 and evaporated to afford a crude oil containing the hydro-
values than the fully matched T/A control in all cases; (ii)
a single imidazole paired with the purine bases (A and G)
lysed phosphitylation reagent, which was chromatographed twice
to a greater extent than with the pyrimidine bases (C and
T); and (iii) pairing of the C2-imidazole with nucleobases
did not show much difference, but the pairing of C2-imid-
azole with itself afforded higher Tm values. In addition,
the results of the Tm experiments were in good agreement
with those of the predicted models of imidazole-modified
DNA duplexes. We see no reason why the synthesis of 2′-
deoxyribonucleoside phosphoramidites 3a and 3b could
not be scaled up to meet the needs of DNA studies, and
therefore, 3a and 3b should be applicable for the insertion
of Cn-imidazoles into DNA oligonucleotides. Further
work on the applications of Cn-imidazole oligonucleotides
is underway and the results will be reported in due course.
over NH-silica gel (EtOAc–hexane, 25:75 v/v) to afford 3a6b (100
mg, 63%) as a white foam.
2-Deoxy-3,5-di-O-benzyl-β-D-ribofuranosylcarbaldehyde (9)
A mixture of NaOMe in MeOH (37.7 mL, 18.85 mmol, 0.5 M) and
compound 815 (4.06 g, 12.57 mmol) was stirred at r.t. for 15 h, and
then acidified with aq HCl at 0 °C and evaporated to afford a resi-
due, which was subsequently dissolved in EtOAc (100 mL). The or-
ganic layer was washed with H2O (100 mL), dried and evaporated.
The residue was purified by column chromatography on silica gel
(EtOAc–hexane, 30:70 v/v) to afford methyl (2-deoxy-3,5-di-O-
benzyl-β-D-ribofuranosyl)carboxylate (4.47 g, quant.) as an oil.
1H NMR (400 MHz, CDCl3): δ = 2.32–2.44 (m, 2 H), 3.54 (dd, J =
10.4, 4.4 Hz, 1 H), 3.56 (dd, J = 10.4, 4.4 Hz, 1 H), 3.70 (s, 3 H),
4.11 (quin, J = 2.8 Hz, 1 H), 4.38 (ddd, J = 4.4, 4.4, 2.8 Hz, 1 H),
4.45–4.57 (m, 4 H), 4.66 (dd, J = 8.0, 4.0 Hz, 1 H), 7.25–7.37 (m,
10 H).
Reactions with air- and moisture-sensitive compounds were carried
out under an Ar atmosphere. Microwave-assisted reactions were
performed using a Milestone MultiSYNTH multimodal reactor
with thermal control. Anhydrous solvents were purchased from
WAKO Chemical Co. Solvents were dried over Na2SO4 and re-
moved on a rotary evaporator under reduced pressure. Fuji Silysia
FL-60D silica gel was used for flash column chromatography.
13C NMR (100 MHz, CDCl3): δ = 35.9, 52.1, 70.2, 70.9, 73.4, 76.7,
79.5, 83.9, 127.5, 127.59, 127.62, 128.29, 128.34, 128.5, 137.8,
138.0, 173.4.
HRMS (FAB): m/z [M + H]+ calcd for C21H25O5: 357.1702; found:
357.1698.
© Georg Thieme Verlag Stuttgart · New York
Synthesis 2014, 46, 2815–2825

2822
S. Harusawa et al.
PAPER
{4-[(Z)-2-(2-Deoxy-3,5-di-O-benzyl-β-D-ribofuranos-1-yl)vi-
nyl]imidazolyl}methyl 2,2-Dimethylpropionate [(Z)-12]
Aq 2 M HCl (5 mL) was added to a solution of (Z)-11 (520 mg, 0.82
mmol) in EtOH (5 mL) and the mixture was heated at reflux tem-
perature for 1 h. The mixture was cooled to r.t. and EtOAc (100 mL)
was added. The resulting mixture was washed with sat. aq NaHCO3
solution (50 mL), dried and evaporated. The residue was purified by
column chromatography on silica gel (EtOAc) to afford 4-[(Z)-2-(2-
deoxy-3,5-di-O-benzyl-β-D-ribofuranos-1-yl)vinyl]-1H-imidazole
(290 mg, 91%) as an oil.
1H NMR (400 MHz, CDCl3): δ = 1.97 (ddd, J = 13.6, 8.0, 5.6 Hz, 1
H), 2.52 (ddd, J = 13.6, 6.8, 6.8 Hz, 1 H), 3.57 (dd, J = 10.4, 6.0 Hz,
1 H), 3.61 (dd, J = 10.4, 6.0 Hz, 1 H), 4.12 (ddd, J = 6.8, 5.6, 4.4 Hz,
1 H), 4.31 (ddd, J = 6.0, 4.4, 4.4 Hz, 1 H), 4.46–4.61 (m, 4 H), 4.92–
4.96 (br s, 1 H), 5.64 (dd, J = 11.6, 6.0 Hz, 1 H), 6.43 (dd, J = 11.6,
1.6 Hz, 1 H), 7.00 (s, 1 H), 7.26–7.36 (m, 11 H).
13C NMR (100 MHz, CDCl3): δ = 38.9, 70.3, 71.7, 73.4, 74.9, 80.4,
82.0, 121.3, 127.60, 127.63, 127.7, 127.9, 128.37, 128.44, 136.1,
137.7, 137.8.
DIBAL-H (4.0 mL, 4.0 mmol, 1 M solution in toluene) was added
dropwise over a period of 10 min to a solution of the carboxylate
(712 mg, 2.0 mmol) in toluene (10 mL) at –78 °C, and the mixture
was stirred for 30 min at the same temperature. The reaction was
quenched with MeOH (2 mL), and further stirred at r.t. Sat. NaH-
CO3 solution (4 mL) and EtOAc (20 mL) were added and the mix-
ture was stirred vigorously for 30 min. Anhydrous MgSO4 was
added to the resulting suspension, which was filtered through a
Celite pad. The filtrate was evaporated and the residue was purified
by column chromatography on silica gel (EtOAc–hexane, 1:3 v/v)
to afford 9 (604 mg, 93%) as an oil.
1H NMR (400 MHz, CDCl3): δ = 2.24–2.34 (m, 2 H), 3.45 (dd, J =
10.4, 5.2 Hz, 1 H), 3.52 (dd, J = 10.4, 4.4 Hz, 1 H), 4.09 (ddd, J =
4.0, 2.4, 1.6 Hz, 1 H), 4.36–4.56 (m, 6 H), 7.23–7.38 (m, 10 H), 9.75
(d, J = 1.6 Hz, 1 H).
13C NMR (100 MHz, CDCl3): δ = 34.5, 70.60, 70.63, 73.5, 79.5,
82.8, 84.3, 127.5, 127.69, 127.73, 127.74, 128.4, 137.4, 137.8,
204.7.
HRMS (FAB): m/z [M + H]+ calcd for C20H23O4: 327.1596; found:
327.1601.
HRMS (FAB): m/z [M + H]+ calcd for C24H27N2O3: 391.2021;
found: 391.2013.
4-[(Z)-2-(2-Deoxy-3,5-di-O-benzyl-β-D-ribofuranos-1-yl)vinyl]-
1-tritylimidazole [(Z)-11] and 4-[(E)-2-(2-Deoxy-3,5-di-O-ben-
zyl-β-D-ribofuranos-1-yl)vinyl]-1-tritylimidazole [(E)-11]
n-BuLi (1.1 mL, 1.78 mmol, 1.6 M solution in hexane) was added
dropwise over 10 min to a suspension of phosphonium salt 10 (1206
mg, 1.94 mmol) in THF (15 mL) at –78 °C, and the mixture was
stirred for 30 min at the same temperature. A solution of aldehyde
9 (530 mg, 1.62 mmol) in THF (5 mL) was added dropwise over 10
min. After stirring for 2 h, the mixture was allowed to reach to r.t.
and then quenched by the addition of H2O. The mixture was evapo-
rated and the residue was dissolved in EtOAc (100 mL). The organ-
ic layer was washed with H2O (100 mL), dried and evaporated. The
residue was purified by column chromatography on silica gel
(EtOAc–hexane, 20:80 to 30:70 v/v) to afford the first fraction, (Z)-
11 (340 mg, 33%), as an oil, and the second fraction, (E)-11 (270
mg, 26%), as an oil. When toluene was used as the solvent, the re-
action was performed at –14 °C, affording (E)-11 (40%) and (Z)-11
(16%).
Under stirring, NaH (60 mg, 1.50 mmol, 60% in mineral oil) was
added to THF (1 mL) to afford a suspension. A solution of the (Z)-
vinylimidazole (390 mg, 1.00 mmol) in THF (4 mL) was added to
the suspension and the resulting mixture was stirred at r.t. for 0.5 h.
Next, chloromethyl pivaloate (0.22 mL, 1.50 mmol) was added. Af-
ter 1 h, H2O was added and the mixture was dissolved in EtOAc
(100 mL). The organic layer was washed with H2O (50 mL), dried
and evaporated. The residue was purified by column chromatogra-
phy on silica gel (EtOAc–hexane, 40:60 v/v) to afford (Z)-12 (455
mg, 90%) as an oil.
1H NMR (400 MHz, CDCl3): δ = 1.16 (s, 9 H), 1.92 (ddd, J = 12.8,
7.6, 5.2 Hz, 1 H), 2.57 (ddd, J = 12.8, 6.4, 6.4 Hz, 1 H), 3.55 (dd,
J = 10.4, 4.8 Hz, 1 H), 3.59 (dd, J = 10.4, 4.8 Hz, 1 H), 4.19 (ddd,
J = 6.4, 5.2, 4.0 Hz, 1 H), 4.28 (ddd, J = 4.8, 4.8, 4.0 Hz, 1 H), 4.48–
4.62 (m, 4 H), 5.46–5.52 (m, 1 H), 5.72 (d, J = 10.8 Hz, 1 H), 5.75
(d, J = 10.8 Hz, 1 H), 5.83 (dd, J = 11.6, 8.0 Hz, 1 H), 6.30 (d, J =
11.6 Hz, 1 H), 7.06 (d, J = 1.2 Hz, 1 H), 7.24–7.36 (m, 10 H), 7.60
(d, J = 1.2 Hz, 1 H).
(Z)-11
13C NMR (100 MHz, CDCl3): δ = 26.8, 38.7, 67.6, 70.9, 71.5, 73.3,
75.4, 80.9, 82.3, 118.8, 121.6, 127.50, 127.54, 127.6, 128.30,
128.34, 132.0, 137.9, 138.2, 138.3, 139.9, 177.6.
HRMS (FAB): m/z [M + H]+ calcd for C30H37N2O5: 505.2703;
found: 505.2708.
1H NMR (400 MHz, CDCl3): δ = 1.88 (ddd, J = 12.8, 7.2, 5.6 Hz, 1
H), 2.50 (ddd, J = 12.8, 6.4, 6.4 Hz, 1 H), 3.45 (dd, J = 10.0, 5.2 Hz,
1 H), 3.50 (dd, J = 10.0, 5.2 Hz, 1 H), 4.11 (ddd, J = 6.4, 5.6, 4.0 Hz,
1 H), 4.25 (ddd, J = 5.2, 5.2, 4.0 Hz, 1 H), 4.45–4.54 (m, 4 H), 5.55
(ddd, J = 8.0, 7.2, 6.4 Hz, 1 H), 5.76 (dd, J = 11.6, 8.0 Hz, 1 H), 6.24
(d, J = 11.6 Hz, 1 H), 6.76 (d, J = 1.2 Hz, 1 H), 7.11–7.14 (m, 6 H),
7.23–7.33 (m, 19 H), 7.40 (d, J = 1.2 Hz, 1 H).
{4-[(E)-2-(2-Deoxy-3,5-di-O-benzyl-β-D-ribofuranos-1-yl)vi-
nyl]imidazolyl}methyl 2,2-Dimethylpropionate [(E)-12]
Aq 2 M HCl (5 mL) was added to a solution of (E)-11 (470 mg, 0.74
mmol) in EtOH (5 mL) and the mixture was heated at reflux tem-
perature for 1 h. The mixture was cooled to r.t. and EtOAc (100 mL)
was added. The resulting mixture was washed with sat. aq NaHCO3
solution (50 mL), dried and evaporated. The residue was purified by
column chromatography on silica gel (EtOAc) to afford 4-[(E)-2-
(2-deoxy-3,5-di-O-benzyl-β-D-ribofuranos-1-yl)vinyl]-1H-imidaz-
ole (234 mg, 81%) as an oil.
1H NMR (400 MHz, CDCl3): δ = 1.88 (ddd, J = 12.8, 7.2, 5.6 Hz, 1
H), 2.38 (ddd, J = 12.8, 6.4, 6.4 Hz, 1 H), 3.52 (dd, J = 10.4, 4.8 Hz,
1 H), 3.55 (dd, J = 10.4, 4.4 Hz, 1 H), 4.07–4.12 (m, 1 H), 4.17–4.24
(m, 1 H), 4.42–4.54 (m, 4 H), 4.60 (q, J = 7.2 Hz, 1 H), 6.22 (dd, J =
15.6, 7.2 Hz, 1 H), 6.43 (d, J = 15.6 Hz, 1 H), 6.88 (s, 1 H), 7.22–
7.34 (m, 10 H), 7.44 (s, 1 H).
13C NMR (100 MHz, CDCl3): δ = 38.6, 70.8, 71.4, 73.3, 75.2, 75.4,
80.9, 82.2, 121.3, 121.4, 127.4, 127.47, 127.54, 127.6, 128.0, 128.2,
128.3, 129.7, 131.3, 138.21, 138.23, 138.3, 138.8, 142.2.
HRMS (FAB): m/z [M + H]+ calcd for C43H41N2O3: 633.3118;
found: 633.3113.
(E)-11
1H NMR (400 MHz, CDCl3): δ = 1.92 (ddd, J = 12.8, 7.2, 5.6 Hz, 1
H), 2.40 (ddd, J = 12.8, 6.4, 6.4 Hz, 1 H), 3.55 (d, J = 4.0 Hz, 2 H),
4.12–4.21 (m, 2 H), 4.45–4.65 (m, 5 H), 6.37 (dd, J = 16.0, 6.8 Hz,
1 H), 6.45 (d, J = 16.0 Hz, 1 H), 6.76 (s, 1 H), 7.10–7.16 (m, 6 H),
7.25–7.35 (m, 19 H), 7.39 (s, 1 H).
13C NMR (100 MHz, CDCl3): δ = 38.7, 70.7, 71.6, 73.4, 75.3, 79.3,
80.5, 82.1, 119.4, 123.1, 127.5, 127.56, 127.60, 127.7, 127.9, 128.0,
128.31, 128.32, 128.6, 129.7, 138.1, 138.3, 138.6, 139.0, 142.2.
13C NMR (100 MHz, CDCl3): δ = 38.7, 70.8, 71.5, 73.4, 79.4, 80.4,
82.1, 121.4, 127.6, 127.9, 128.3, 135.7, 137.90, 137.92.
HRMS (FAB): m/z [M + H]+ calcd for C24H27N2O3: 391.2021;
HRMS (FAB): m/z [M + H]+ calcd for C43H41N2O3: 633.3118;
found: 633.3120.
found: 391.2020.
Synthesis 2014, 46, 2815–2825
© Georg Thieme Verlag Stuttgart · New York

PAPER
C0- and C2-Imidazole 2′-Deoxyribonucleoside Phosphoramidites
2823
Under stirring, NaH (11 mg, 0.28 mmol, 60% in mineral oil) was
added to THF (1 mL) to afford a suspension. A solution of the (E)-
vinylimidazole (110 mg, 0.28 mmol) in THF (4 mL) was added to
the suspension and the resulting mixture was stirred at r.t. for 0.5 h.
Next, chloromethyl pivaloate (0.06 mL, 0.42 mmol) was added. Af-
ter 1 h, H2O was added and the mixture was dissolved in EtOAc
(100 mL). The organic layer was washed with H2O (50 mL), dried
and evaporated. The residue was purified by column chromatogra-
phy on silica gel (EtOAc–hexane, 50:50 v/v) to afford (E)-12 (84
mg, 66%) as an oil.
1H NMR (400 MHz, CDCl3): δ = 1.16 (s, 9 H), 1.93 (ddd, J = 12.8,
7.2, 5.6 Hz, 1 H), 2.41 (ddd, J = 12.8, 6.4, 6.4 Hz, 1 H), 3.56 (d, J =
4.4 Hz, 1 H), 4.12–4.18 (m, 1 H), 4.22 (q, J = 4.4 Hz, 1 H), 4.46–
4.68 (m, 5 H), 5.78 (s, 2 H), 6.43 (dd, J = 16.0, 4.8 Hz, 1 H), 6.48
(d, J = 16.0 Hz, 1 H), 6.98 (d, J = 1.2 Hz, 1 H), 7.25–7.35 (m, 10 H),
7.60 (d, J = 1.2 Hz, 1 H).
0.20 mmol) were added. The resulting mixture was stirred at 40 °C
for 1 h and then evaporated. The residual oil was subjected to chro-
matography on NH-silica gel (EtOAc–hexane, 40:60 v/v) to afford
3b (80 mg, 96%) as two diastereomers (white foam).
1H NMR (400 MHz, CDCl3): δ = 0.99–1.06 (m, 2 H), 1.10–1.24 (m,
19 H), 1.72–1.85 (m, 1 H), 1.92–2.08 (m, 2 H), 2.32–2.44 (m, 2 H),
2.54–2.80 (m, 3 H), 3.06–3.13 (m, 1 H), 3.18–3.24 (m, 1 H), 3.48–
3.62 (m, 3 H), 3.70–3.87 (m, 1 H), 3.78 (s, 3 H), 3.79 (s, 3 H), 4.10–
4.22 (m, 2 H), 4.40–4.52 (m, 1 H), 5.69 (d, J = 10.8 Hz, 1 H), 5.73
(d, J = 10.8 Hz, 1 H), 6.79–6.83 (m, 5 H), 7.19–7.36 (m, 7 H), 7.43–
7.47 (m, 2 H), 7.57 (d, J = 1.2 Hz, 1 H).
13C NMR (100 MHz, CDCl3): δ (diastereomeric mixture) = 20.0,
20.1, 20.2, 20.3, 21.5, 24.3, 24.4, 24.46, 24.53, 24.87, 24.91, 26.7,
35.37, 35.41, 38.6, 39.66, 39.70, 39.74, 42.9, 42.99, 43.03, 43.1,
55.1, 57.5, 58.0, 58.2, 64.0, 64.2, 67.6, 75.0, 75.2, 75.3, 75.5, 76.7,
78.3, 83.6, 83.67, 83.72, 85.8, 112.9, 115.3, 117.46, 117.54, 126.56,
126.61, 127.7, 128.16, 128.22, 130.0, 130.1, 136.08, 136.13, 137.5,
143.2, 144.9, 158.3, 177.6.
13C NMR (100 MHz, CDCl3): δ = 26.8, 38.7, 38.8, 67.6, 70.7, 71.6,
73.4, 79.1, 80.5, 82.2, 116.8, 122.2, 127.5, 127.56, 127.62, 127.7,
128.3, 128.4, 129.6, 138.2, 138.26, 138.31, 140.6, 177.7.
HRMS (FAB): m/z [M + H]+ calcd for C30H37N2O5: 505.2703;
found: 505.2705.
31P NMR (121 MHz, CDCl3): δ = 147.9, 148.4.
HRMS (FAB: TEOA + NaCl):17 m/z [M + Na]+ calcd for
C46H61N4O8PNa: 851.4125; found: 851.4129.
{4-[2-(5-O-DMT-2-deoxy-β-D-ribofuranos-1-yl)ethyl]imidazol-
yl}methyl 2,2-Dimethylpropionate (13)
Oligonucleotide Synthesis and Purification
Oligonucleotides were synthesised with trityl-OFF on a 1.0 μM
scale on a controlled pore glass (CPG) solid support using a proce-
dure based on phosphoramidite chemistry.19 Commercially avail-
able DNA reagents and solvents were obtained from Links
Technologies, Applied Biosystems and Ruthburn Ltd., respectively.
The standard phosphoramidites (Bz-dA, ibu-dG, Ac-dC, and T) and
prepared imidazole phosphoramidites were dissolved in anhydrous
MeCN to a concentration of 0.1 M prior to the synthesis. Ethylthio-
1H-tetrazole (ETT; 0.25 M) in MeCN was used as an activator and
I2/H2O (0.02 M) was used as an oxidiser. A 1.0 μM DNA/RNA syn-
thetic cycle involving acid detritylation, coupling, capping and oxi-
dation steps was used. For IM modification, an extended coupling
time of 10 min was used, while for standard bases, a coupling time
of 25 s was used. The required sequences were cleaved from the sol-
id support by treatment with concd aq NH3 for approximately 1 h on
the synthesiser. The solutions were then heated in a sealed vial at
55 °C for 6 h for deprotection before being concentrated on a
Speedvac.
A mixture of (E/Z)-12 (270 mg, 0.53 mmol), 20% Pd(OH)2/C (160
mg) and cyclohexene (1.6 mL, 15.9 mmol) in EtOH was exposed to
MW irradiation at 100 °C for 0.5 h. After filtration through Celite,
the filtrate was evaporated to afford {4-[2-(2-deoxy-β-D-ribofura-
nos-1-yl)ethyl]imidazolyl}methyl 2,2-dimethylpropionate (195
mg, quant.) as an oil.
1H NMR (400 MHz, CD3OD): δ = 1.18 (s, 9 H), 1.63 (ddd, J = 12.8,
6.8, 6.0 Hz, 1 H), 1.83–2.03 (m, 2 H), 2.34 (ddd, J = 12.8, 6.4, 6.4
Hz, 1 H), 2.66–2.83 (m, 2 H), 3.47–3.63 (m, 2 H), 3.78 (q, J = 4.8
Hz, 1 H), 4.00–4.10 (m, 1 H), 4.20 (q, J = 6.0 Hz, 1 H), 6.04 (s, 2
H), 7.37 (s, 1 H), 8.67 (s, 1 H).
13C NMR (100 MHz, CD3OD): δ = 23.3, 27.1, 36.1, 39.7, 41.3,
63.3, 70.3, 73.4, 78.4, 86.8, 118.9, 138.1, 138.7, 178.7.
HRMS (FAB): m/z [M + H]+ calcd for C16H27N2O5: 327.1920;
found: 327.1920.
The prepared diol (195 mg, 0.53 mmol) was co-evaporated three
times with pyridine (1 mL) and dissolved in anhydrous pyridine (5
mL). DMTCl (271 mg, 0.80 mmol), Et3N (110 μL, 0.80 mmol) and
DMAP (6 mg, 0.05 mmol) were added to the solution, and the mix-
ture was stirred at r.t. for 2 h. MeOH (1 mL) was added and the mix-
ture was evaporated. The residue was subjected to chromatography
on an NH-silica gel column (EtOAc) to afford compound 13 (224
mg, 67%) as an amorphous product.
1H NMR (400 MHz, CDCl3): δ = 1.16 (s, 9 H), 1.68 (ddd, J = 12.8,
7.2, 6.4 Hz, 1 H), 1.89–2.03 (m, 2 H), 2.34 (ddd, J = 12.8, 6.4, 6.4
Hz, 1 H), 2.55–2.75 (m, 2 H), 3.09 (dd, J = 9.6, 6.0 Hz, 1 H), 3.27
(dd, J = 9.6, 4.8 Hz, 1 H), 3.78 (s, 6 H), 3.98 (ddd, J = 6.0, 4.8, 4.8
Hz, 1 H), 4.06–4.14 (m, 1 H), 4.28 (td, J = 6.4, 4.8 Hz, 1 H), 5.71
(d, J = 11.6 Hz, 1 H), 5.73 (d, J = 11.6 Hz, 1 H), 6.78 (d, J = 1.2 Hz,
1 H), 6.80–6.84 (m, 4 H), 7.17–7.22 (m, 1 H), 7.25–7.34 (m, 6 H),
7.41–7.44 (m, 2 H), 7.54 (d, J = 1.2 Hz, 1 H).
13C NMR (100 MHz, CDCl3): δ = 24.6, 26.8, 35.7, 38.7, 40.2, 55.2,
64.8, 67.6, 74.8, 77.9, 83.5, 86.2, 113.1, 113.2, 115.4, 126.7, 127.8,
128.1, 129.99, 130.01, 135.97, 136.04, 137.4, 143.2, 144.8, 158.4,
177.6.
HRMS (FAB): m/z [M + H]+ calcd for C37H45N2O7: 629.3227;
found: 629.3224.
The residues were dissolved in H2O (1.0 mL) and purified by re-
versed-phase HPLC on a C18 column (Phenomenex Clarity 5 μ Oli-
go-RP column, 150 × 10 mm), and eluted with 0.1 M TEAA (pH
7.0) and MeCN mixtures with a flow rate of 3.0 mL/min. The frac-
tions of the major peak, monitored at 260 nm, were collected and
evaporated. The residue was dissolved in H2O and desalted by pass-
ing through an NAP-10 column (GE Healthcare) following the
manufacturer’s instructions.
The purity of each oligonucleotide was assessed by analytical
HPLC and mass spectrometry. The analysis conditions and results
for key oligonucleotides are presented in the Supporting Informa-
tion.
UV/Vis spectra were recorded with a UV-1800 Shimadzu UV spec-
trophotometer. The concentrations of each oligonucleotide were de-
termined by measuring the absorbance at 260 nm. The molar
extinction coefficients of the oligonucleotides were obtained by
submitting the sequences on the IDT UV spectrum website
(http://biophysics.idtdna.com/UVSpectrum.html). The absorbance
of imidazole compared with those of the nucleobases at 260 nm was
negligible. The concentrations of these oligonucleotides were ob-
tained in the range 150–300 μM as solutions in H2O (1.5 mL).
Tm Measurements
[4-(2-{5-O-DMT-2-deoxy-3-O-[2-cyanoethyl-(N,N-diisopropyl-
amino)phosphoramidyl]-β-D-ribofuranos-1-yl}ethyl)imidazol-
yl]methyl 2,2-Dimethylpropionate (3b)
Compound 13 (63 mg, 0.10 mmol) was dissolved in anhydrous
DCE (3 mL) and DCI (14 mg, 0.12 mmol) and CETIP-PA (63 μL,
A 5.0 μM solution of each oligonucleotide strand in 1.0 mL of a pH
7.0 buffer containing phosphate buffer (10 mM), and NaCl (100
mM) was prepared. The solution was heated at 80 °C for 5 min and
then allowed to cool slowly to r.t. over a period of 4 h before being
© Georg Thieme Verlag Stuttgart · New York
Synthesis 2014, 46, 2815–2825

2824
S. Harusawa et al.
PAPER
placed in a UV spectrometer with temperature control for recording
the Tm measurements.
Jayaraman, K.; Revankar, G. R. Nucleic Acids Res. 1995, 23,
647. (c) Johnson, W. T.; Zhang, P.; Bergstrom, D. E. Nucleic
Acid Res. 1997, 25, 559.
Molecular Modelling of the DNA Duplex Model and IM/IM2
Incorporated Structures
(6) (a) Araki, L.; Harusawa, S.; Yamaguchi, M.; Yonezawa, S.;
Taniguchi, N.; Lilley, D. M. J.; Zhao, Z.; Kurihara, T.
Tetrahedron Lett. 2004, 45, 2657. (b) Araki, L.; Harusawa,
S.; Yamaguchi, M.; Yonezawa, S.; Taniguchi, N.; Lilley, D.
M. J.; Zhao, Z.; Kurihara, T. Tetrahedron 2005, 61, 11976.
(c) Araki, L.; Morita, K.; Yamaguchi, M.; Zhao, Z.; Wilson,
T. J.; Lilley, D. M. J.; Harusawa, S. J. Org. Chem. 2009, 74,
2350. (d) Araki, L.; Zhao, Z.; Lilley, D. M. J.; Harusawa, S.
Heterocycles 2010, 81, 1861. (e) Harusawa, S.; Fujii, K.;
Nishiura, M.; Araki, L.; Usami, Y.; Zhao, Z.; Lilley, D. M.
J. Heterocycles 2011, 83, 2041. (f) Araki, L.; Harusawa, S.
Yakugaku Zasshi 2010, 130, 1707.
(7) Lilley, D. M. J. The Hairpin and Varkud Satellite
Ribozymes, In Ribozymes and RNA Catalysis; Lilley, D. M.
J.; Eckstein, F., Eds.; Royal Society of Chemistry:
Cambridge, 2008, 66.
(8) (a) Zhao, Z.; McLeod, A.; Harusawa, S.; Araki, L.;
Yamaguchi, M.; Kurihara, T.; Lilley, D. M. J. J. Am. Chem.
Soc. 2005, 127, 5026. (b) Wilson, T. J.; Ouellet, J.; Zhao, Z.;
Harusawa, S.; Araki, L.; Kurihara, T.; Lilley, D. M. J. RNA
2006, 12, 980. (c) Lilley, D. M. J. Biol. Chem. 2007, 388,
699.
A theoretical molecular model for the DNA duplex of 5′-TG-
GACTCTCTCAATG-3′ (control probe) and 5′-CATTGAGA-
GAGTCCA-3′ (control target) was constructed as a B-DNA form using
the MOE (Molecular Operating Environment 2013.0801, Chemical
Computing Group Inc., Québec, CA; http://www.chemcomp.com/)
package for molecular structure analyses. The model was initially
optimised by energy minimisation using the Amber12 EHT force
field.21 The energy cut-off distance was set at 10 Å, and the dielec-
tric constant was distance-dependent, based on a value of 1.0 for the
DNA molecule and 80.0 for the exterior space. Subsequently, a 500-
ps molecular dynamics simulation was performed on the model at
300 K using a 0.002 ps time step, the same force field, and the NPA
(Nosé–Poincaré–Anderson) Hamiltonian equations.22 Before these
equilibrium iterations, 5-ps heating iterations were employed to ac-
count for stable equilibration. Finally, the most stable structure
during the latter 50-ps simulations was optimised by energy mini-
misation. Considering the molecular stability of the DNA molecule
at 300 K (26.85 °C), atoms of four nucleotides of both termini were
tethered in the space at 1.0 Å from those atomic positions in the ini-
tial B-DNA form during all the optimisation steps. Complex mole-
cules of IM probe–A, IM probe–IM, IM2 probe–A, and IM2 probe–
IM2 targets were modelled from the above B-DNA form, and these
molecular models were also optimised through the above methods.
All modelling operations, including calculations for the optimisa-
tion steps, were performed using MOE.
(9) (a) Harusawa, S.; Yoneyama, H.; Fujisue, D.; Nishimura,
M.; Fujitake, M.; Usami, Y.; Zhao, Z.; McPhee, S. A.;
Wilson, T. J.; Lilley, D. M. J. Tetrahedron Lett. 2012, 45,
5891. (b) Yoneyama, H.; Usami, Y.; Komeda, S.; Harusawa,
S. Synthesis 2013, 45, 1051. (c) The typical debenzylation of
3,5-di-O-benzyl-2-deoxy-β-D-riboses with Pd(OH)2/C–
cyclohexene in ethanol at reflux temperature proceeds
slowly, giving moderate yields of the products.
Acknowledgment
We thank Dr. M. Fujitake of Osaka University of Pharmaceutical
Sciences for MS measurements. This work was supported in part by
a Grant-in-Aid for Scientific Research [Grant No. 21590130 (S.H.)]
from JSPS, 2009–2013. We thank the EPSRC, Advantage West
Midlands, University of Birmingham and the analytical facility wit-
hin the School of Chemistry for support.
(10) Li, Z.; Zhou, H.; Wu, X.; Yao, H. Curr. Med. Chem. 2013,
20, 3641.
(11) (a) Sakthivel, K.; Barbas, C. F. III Angew. Chem. Int. Ed.
1998, 37, 2872. (b) Lee, S. E.; Sidorov, A.; Gourlain, T.;
Mignet, N.; Thorpe, S. J.; Brazier, J. A.; Dickman, M. J.;
Hornby, D. P.; Grasby, J. A.; Williams, D. M. Nucleic Acids
Res. 2001, 29, 1565. (c) Sawai, H.; Ozaki, A. N.; Satoh, F.;
Ohbayashi, T.; Masud, M. M.; Ozaki, H. Chem. Commun.
2001, 2604. (d) Perrin, D. M.; Garestier, T.; Hélène, C.
J. Am. Chem. Soc. 2001, 123, 1556.
Supporting Information for this article is available online
at
http://www.thieme-connect.com/products/ejournals/journal/
10.1055/s-00000084.SunpfgIpi
o
o
nr
i
(12) For recent reviews on C-nucelosides, see: (a) Boutureira, O.;
Matheu, M. I.; Díaz, Y.; Castillón, S. Synthesis of C-
Nucleosides, In Chemical Synthesis of Nucleoside
Analogues; Merino, P., Ed.; Wiley: Hoboken, 2013, 263.
(b) Wu, Q.; Simons, C. Synthesis 2004, 1533.
References
(1) Grimmett, M. R. Imidazoles, In Comprehensive
Heterocyclic Chemistry II; Vol. 3; Katritzky, A. R.; Rees, C.
W.; Scriven, E. F. V., Eds.; Pergamon: Oxford, 1996, 77.
(2) (a) Kopple, J. D.; Swendseid, M. E. J. Clin. Invest. 1975, 55,
881. (b) Fahey, R. C. Annu. Rev. Microbiol. 2001, 55, 333.
(3) (a) Marieb, E. N. In Human Anatomy & Physiology;
Benjamin-Cummings: San Francisco, 2001, 414, 796, 824
and 90. (b) Repka-Ramirez, M. S. Curr. Allergy Rep. 2003,
30, 227. (c) Hough, L. B. Mol. Pharmacol. 2001, 59, 415.
(4) (a) Madasu, S. B.; Vekariya, N. A.; Koteswaramma, C.;
Islam, A.; Sanasi, P. D.; Korupolu, R. B. Org. Process Res.
Dev. 2012, 16, 2025. (b) Walsh, A. J.; Davis, M. L.; Fraser,
W. Molecules 2006, 11, 486. (c) Mangas-Sánchez, J.; Busto,
E.; Gotor-Fernández, V.; Malpartida, F.; Gotor, V. J. Org.
Chem. 2011, 76, 2115. (d) Liang, J.; Lv, W.; Li, X.; An, K.;
Cushman, M.; Wang, H.; Xu, Y. Bioorg. Med. Chem. Lett.
2013, 23, 1387.
(c) Štambaský, J.; Hocek, M.; Kočovský, P. Chem. Rev.
2009, 109, 6729.
(13) For the synthesis of C4-linked imidazole 2′-deoxyribo-
nucleosides, see: (a) Bergstrom, D. E.; Zhang, P.; Zhou, J.
J. Chem. Soc., Perkin Trans. 1 1994, 3029. (b) Harusawa, S.;
Murai, Y.; Moriyama, H.; Imazu, T.; Ohishi, H.; Yoneda, R.;
Kurihara, T. J. Org. Chem. 1996, 61, 4405.
(14) (a) Beaucage, S. L. Oligodeoxyribonucleoside Synthesis, In
Protocols for Oligonucleotides and Analogs; Agrawal, S.,
Ed.; Humana Press: Totowa (NJ), 1993, 33. (b) Vargeese,
C.; Carter, J.; Yegge, J.; Krivjansky, S.; Settle, A.; Kropp,
E.; Peterson, K.; Pieken, W. Nucleic Acids Res. 1998, 26,
1046.
(15) Seitz, G.; Lachmann, J. Z. Naturforsch. 1999, 54b, 549.
(16) (a) The synthesis of a C2-α/β mixture (9αβ) of a
furancarbaldehyde was reported (see ref. 16b), but the
selective synthesis of 9β was first carried out in this paper.
(b) Krishna, P. R.; Reddy, V. V. R.; Srinivas, R. Tetrahedron
2007, 63, 9871.
(5) For oligodeoxyribonucleotides containing an imidazole N-
nucleoside, see: (a) Johannsen, S.; Megger, N.; Böhme, D.;
Siegel, R. K. O.; Müller, J. Nat. Chem. 2010, 2, 229.
(b) Durland, R. H.; Rao, T. S.; Bodepudi, V.; Seth, D. M.;
Synthesis 2014, 46, 2815–2825
© Georg Thieme Verlag Stuttgart · New York

PAPER
C0- and C2-Imidazole 2′-Deoxyribonucleoside Phosphoramidites
(18) Manchester, J.; Bassani, D. M.; Duprey, J.-L. H. A.;
2825
(17) (a) Fujitake, M.; Harusawa, S.; Araki, L.; Yamaguchi, M.;
Lilley, D. M. J.; Zhao, Z.; Kurihara, T. Tetrahedron 2005,
61, 4689. (b) Harusawa, S.; Fujitake, M.; Kurihara, T.; Zhao,
Z.; Lilley, D. M. J. Mass Determination of
Giordano, L.; Vyle, J. S.; Zhao, Z.-y.; Tucker, J. H. R. J. Am.
Chem. Soc. 2012, 134, 10791.
(19) Beaucage, S. L.; Caruthers, M. H. Tetrahedron Lett. 1981,
22, 1859.
(20) Xia, J.; Noronha, A.; Toudjarska, I.; Li, F.; Akinc, A.;
Braich, R.; Frank-Kamenetsky, M.; Rajeev, K. G.; Egli, M.;
Manoharan, M. ACS Chem. Biol. 2006, 1, 176.
(21) Salomon-Ferrer, R.; Case, D. A.; Walker, R. C. WIREs
Comput. Mol. Sci. 2013, 3, 198.
Phosphoramidites, In Current Protocols in Nucleic Acid
Chemistry; Beaucage, S. L.; Bergstrom, D. E.; Herdewijn,
P.; Matsuda, A., Eds.; John Wiley & Sons: New York, 2006,
10.11.1. (c) Fujitake, M.; Harusawa, S. Yakugaku Zasshi
2013, 133, 823.
(22) Sturgeon, J. B.; Laird, B. B. J. Chem. Phys. 2000, 112, 3474.
© Georg Thieme Verlag Stuttgart · New York
Synthesis 2014, 46, 2815–2825