The alkyl-connected 2-amino-6-vinylpurine (AVP) crosslinking agent for improved selectivity to the cytosine base in RNA

Article abstract of DOI:10.1016/j.bmc.2010.03.008

We have previously reported that the 2-amino-6-vinylpurine (AVP) nucleoside exhibits a highly efficient and selective crosslinking reaction toward cytosine and displayed an improved antisense inhibition in cultured cells. In this study, we further investigated the alkyl-connected AVP nucleoside analogs for more efficient crosslinking to the cytosine base (rC) of the target RNA. We synthesized three AVP analogs which connect the 2-amino-6-vinylpurine unit to the 2′-deoxyribose through a methylene, an ethylene, or a butylene linker. The ODN incorporating the AVP analog with the methylene or the butylene linker showed a slightly higher crosslinking to the target rC of RNA than the original AVP with no linker. In contrast, the AVP with the ethylene linker formed a selective and efficient crosslink to the rC of the target RNA.

Full text of DOI:10.1016/j.bmc.2010.03.008

Bioorganic & Medicinal Chemistry 18 (2010) 2894–2901
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
The alkyl-connected 2-amino-6-vinylpurine (AVP) crosslinking agent
for improved selectivity to the cytosine base in RNA
Yosuke Taniguchi a,c, Yusuke Kurose a, Takamasa Nishioka a, Fumi Nagatsugi b,c, Shigeki Sasaki a,c,
*
a Graduate School of Pharmaceutical Sciences, Kyushu University, 3-1-1 Maidashi, Higashi-ku, Fukuoka 812-8582, Japan
b Institute of Multidisciplinary Research for Advanced Materials, Tohoku University, 2-1-1 Katahira, Aoba-ku, Sendai-shi, Miyagi 98-8587, Japan
c CREST, Japan Science and Technology Agency, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
We have previously reported that the 2-amino-6-vinylpurine (AVP) nucleoside exhibits a highly efficient
and selective crosslinking reaction toward cytosine and displayed an improved antisense inhibition in cul-
tured cells. In this study, we further investigated the alkyl-connected AVP nucleoside analogs for more effi-
cient crosslinking to the cytosine base (rC) of the target RNA. We synthesized three AVP analogs which
connect the 2-amino-6-vinylpurine unit to the 20-deoxyribose through a methylene, an ethylene, or a butyl-
ene linker. The ODN incorporating the AVP analog with the methylene or the butylene linker showed a
slightly higher crosslinking to the target rC of RNA than the original AVP with no linker. In contrast, the
AVP with the ethylene linker formed a selective and efficient crosslink to the rC of the target RNA.
Ó 2010 Elsevier Ltd. All rights reserved.
Received 1 February 2010
Revised 3 March 2010
Accepted 4 March 2010
Available online 9 March 2010
Keywords:
Antisense oligonucleotide
Crosslink
RNA
Base selectivity
1. Introduction
RNA interactive oligonucleotides have significant potentials as
therapeutic agents for the treatment of a number of diseases, such
as cancers, diabetes, and infectious diseases.1–7 Their clinical appli-
cations have attracted more and more attention since the discov-
ery of the intrinsic and significant roles of non-coding functional
RNAs in the regulation of the gene expression in living cells. Among
the RNA binding molecules, the antisense oligonucleotides (ASO)
have the longest research history for therapeutic applications. A
variety of chemical modifications have been made on the natu-
ral-type ASO for their in vivo use, such as the phosphorothioate
backbone (PS-oligonucleotide), 20-O-methyl (20-OMe), 20-O-
methoxyethyl (20-MOE), and 20-O-40-C methylene locked nucleic
acid (LNA) modified nucleotides.8–11 These modifications are
beneficial for the tolerance to nucleases, thermal stability of the
complex with target RNAs, or penetration into cells. Another inter-
esting strategy of ASO is the use of chemically reactive nucleotide
analogs for the stabilization of the complex with the target RNA by
forming a covalent bond. Psolaren photo-crosslinking units have
been widely used in biological studies. We have developed the ori-
ginal crosslinking agent, the 2-amino-6-vinylpurine nucleoside
analog (AVP), for the selective crosslinking reaction with the
cytosine base (Scheme 1). Its stable precursor, the 2-amino-6-phe-
nylthioethylpurine nucleoside (AVP (SPh)), has exhibited an
Scheme 1. Predicted alkylation reaction of AVP (2-amino-6-alkyl-purine) and
cytosine in duplex DNA.
improved antisense inhibition through automatic activation within
the complex with the target RNA.12–15 This ‘induced alkylating sys-
tem’ has been further applied to intracellular antisense inhibition,
in which the ASO containing AVP (SMe) was conjugated with PEG
and formulated into nano-particles of the polyion complex mi-
celles with poly-L
-lysine.16 The AVP derivatives have other applica-
tions in the post-synthetic conjugation and in the DNA polymerase
reaction.17–19 However, it was realized in subsequent studies that
the reactivity of AVP depends on the target RNA sequence. It was
hypothesized that the local structure of the DNA/RNA heterodu-
plex slightly varies with the change in the sequence and affects
the closeness between the vinyl group of AVP and the amino group
of the cytosine base. Thus, the new AVP derivatives were designed
to have a spacer between the 2-amino-6-vinylpurine unit and 20-
deoxyribose by expecting improvement in the proximity effect to
the cytosine base (Fig. 1). The AVP analogs are called o-AVP with
no linker (original), me-AVP with the methylene linker, et-AVP
with the ethylene linker and bu-AVP with the butylene linker. In
our earlier study, et-AVP and bu-AVP were used for the selective
crosslinking to the adenine base of the TA pair, and the cytosine
* Corresponding author. Tel./fax: +81 92 642 6651.
E-mail address: sasaki@phar.kyushu-u.ac.jp (S. Sasaki).
0968-0896/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2010.03.008

Y. Taniguchi et al. / Bioorg. Med. Chem. 18 (2010) 2894–2901
2895
dite precursors 5 by the conventional method. The phosphorami-
dite precursors (5) were incorporated into ODNs using the DNA
automated synthesizer. The synthesized ODNs 6 were purified by
HPLC, the DMTr-protecting group was removed in an aqueous
AcOH solution, and their structures were identified by MALDI-
TOF MS measurements. These results are summarized in Table 1.
The purified ODNs were treated with magnesium monop-
eroxyphthalate (MMPP) followed by treatment with an aqueous
NaOH solution to generate the vinyl group of ODNs 7, which were
purified by HPLC again, and confirmed by a MALDI-TOF MS mea-
surement (data not shown).
2.2. Estimation of crosslinking reactions with oligonucleotides
containing AVP derivatives for DNA or RNA
The crosslinking reactions of 16 mer ODNs, o-ODN(8), me-ODN
(8), et-ODN(8), and bu-ODN(8) containing AVP were performed
using the FAM-labeled complementary DNA and RNA, and its pro-
gress was followed by the denatured gel mobility shift assay
including 7 M urea. The yield of the crosslinked products were
determined by quantification of the fluorescent bands based on
the bands corresponding to the non-reacted target DNA and RNA
Examples of the gel results of the o-ODN(8)/RNA and et-ODN(8)/
RNA heteroduplex are shown in Figure 2. The time-dependent
changes in the obtained alkylation yields are summarized in Figure
3. Previously, we reported that ODN having the original structure
of AVP, o-ODN(8), exhibited a highly efficient and selective cross-
linking reaction toward dC in the complimentary DNA (Fig. 3A).
However, almost no crosslinking ability by o-ODN(8) was observed
for any base in the target RNA (Figs. 2A and 3E). As the reactivity of
the vinyl group of AVP is highly dependent on the distance to the
amino group, these results indicate that the proximity effect be-
tween AVP and rC is lost in the DNA/RNA hetero duplex of this se-
quence. The ODN me-ODN(8) with AVP connected by the methyl
linker showed less reactivity and specificity in the crosslinking
reaction with the DNA targets, and displayed only a negligible
reactivity to the RNA targets (Fig. 3B and F). In a remarkable con-
trast, et-ODN(8) including AVP connected by the ethyl linker
exhibited a high crosslinking reactivity with a high selectivity to
rC in DNA/RNA hetero duplex (Figs. 2B and 3G), although a lower
reactivity and selectivity were observed to the DNA targets
(Fig. 3C). The ODN bu-ODN(8) did not show any significant cross-
linking reactivity either to the DNA or the RNA targets (Fig. 3D
and H). These results clearly indicated that the optimal length of
the spacer between the AVP unit and the 20-deoxyribose group is
the ethylene linker for the rC of DNA/RNA sequences used in this
study.
Figure 1. Structures of AVP analog monomers bearing alkyl linker between AVP
and 20f-deoxyribose.
base of the GC pair in the triplex DNA, respectively.20–22 In this pa-
per, we describe in detail the synthesis of the linker-containing
AVP analogs, and evaluation of the crosslinking properties of the
oligonucleotides incorporating them.
2. Result
2.1. Synthesis
2.1.1. Synthesis of the methylene-linked 2-amino-6-
vinylpurine derivative (me-AVP-SMe)
The synthesis of the new nucleoside analog, the methylene-
linked 2-amino-6-vinylpurine derivative 4, is shown in Scheme 2.
10-b-Cyano-20-deoxyribose 123 was converted to the tosylated sugar
unit 2 in 66% yield by a sequence of reactions including hydrolysis, 24
reduction of the resulting carboxylic acid using borate, and
subsequent treatment with tosylchloride. The displacement of
the tosylate group of 2 with 6-chloro-2-aminopurine afforded the
7 N-alkylated product as the major product in 68% after purifica-
tion by silica-gel column chromatography (7 N/9 N > 49:1). The
Pd(II)-catalyzed cross-coupling reaction with the 2,4,6-trivinylcy-
clotriboroxane pyridine complex25 gave the 2-amino-6-vinylpu-
rine (AVP) derivative, which afforded 3 by the reaction with
NaSMe. After removal of the toluoyl protecting groups, the corre-
sponding diol groups were transiently protected with TMS
groups,26 followed by protection of the 2-amino group with phen-
oxyacetyl chloride to produce 4 (me-AVP-SMe) in 68% yield from 3
in two steps. The synthesis of et-AVP-SMe and bu-AVP-SMe was
performed as previously reported.21,22 The abbreviation –SMe rep-
resents the fact that the vinyl group of AVP is protected by the
methylsulfide group.
As RNA generally forms a secondary structure under physio-
logical conditions, we tested the crosslinking ability of the al-
kyl-linked AVP structure to the 27 mer RNA target with a
hairpin-loop structure using o-ODN(9) and et-ODN(9), which
have the AVP unit at the complementary site toward the rC in
the loop position of the RNA target (Fig. 4A). As predicted from
the results of Figure 3E, o-ODN(9) did not show a high reactivity
to this loop RNA target (Fig. 4B). On the other hand, et-ODN(9)
2.1.2. Synthesis of the amidite precursor and the
oligonucleotide containing the AVP derivative
As shown in Scheme 3, the diol compounds 4 (o-, me-, et-, and
bu-AVP-SMe) were converted to the corresponding phosphorami-
Scheme 2. Reagents and conditions: (a) (1) concd HCl–dioxane; (2) BH3–THF, dry THF 80% (two steps); (b) p-TsCl, dry pyridine 82%; (c) 2-amino-6-chloropurine, t-BuOK, dry
DMSO 68%; (d) Pd(PPh3)4, vinylboronic anhydride, LiBr, K2CO3, H2O/dioxane; (e) MeSNa, CH3CN 97% (two steps); (f) NaOMe, dry MeOH; (g) TMSCl, dry pyridine, dry CH2Cl2;
(2) PhOAcCl; (3) MeOH 68% (two steps).

2896
Y. Taniguchi et al. / Bioorg. Med. Chem. 18 (2010) 2894–2901
Scheme 3. Reagents and conditions: (a) DMTrCl, pyridine; (b) 2-cyanoethyl-N,N-diisopropylchlorophosphoramidite, DIPEA, CH2Cl2; (c) (1) synthesis with an automated DNA
synthesizer; (2) 28% aqueous NH3; (3) HPLC purification; (4) 10% CH3COOH; (d) (1) 2 equiv MMPP; (2) aqueous NaOH.
Table 1
The sequences and MALDI-TOF MS results of oligonucleotides (6) containing reactive functional nucleoside derivatives in this study
ODN
Sequence
MALDI-TOF mass (m/z)
Calcd
Found
o-ODN(8)
me-ODN(8)
et-ODN(8)
bu-ODN(8)
o-ODN(9)
et-ODN(9)
ODN(10)
o-ODN(10)
et-ODN(10)
ODN(11)
50 d(CTT T(o-AVP-SMe) T TCT CCT TTC T) 30
50 d(CTT T(me-AVP-SMe) T TCT CCT TTC T) 30
50 d(CTT T(et-AVP-SMe) T TCT CCT TTC T) 30
50 d(CTT T(bu-AVP-SMe) T TCT CCT TTC T) 30
50 d(TAT ACT GAt CAA ATT (o-AVP-SMe) TA T)* 30
50 d(TAT ACT GAt CAA ATT (et-AVP-SMe) TA T)* 30
50 d(TTC GCC TCT TTG ATT AAC GCC) 30
4810.99
4825.02
4839.05
4867.11
6065.07
6093.10
6319.98
6378.00
6406.03
6611.13
6669.15
6691.18
4809.95
4824.39
4837.73
4871.34
6071.17
6098.34
6319.50
6381.08
6400.53
6610.10
6668.43
6696.54
50 d(TTC GCC TCT TT (o-AVP-SMe)ATT AAC GCC) 30
50 d(TTC GCC TCT TT (et-AVP-SMe)ATT AAC GCC) 30
50 d(ATG CCC ATA CTG TTG AGC AAT)* 30
o-ODN(11)
et-ODN(11)
50 d(ATG CCC ATA CT(o-AVP-SMe)TTG AGC AAT)* 30
50 d((ATG CCC ATA CT(et-AVP-SMe)TTG AGC AAT)* 30
*
These oligonucleotides have an amino linker at the 30 end.
displayed a selective crosslinking ability to rC, although the alkyl-
ating yield for the loop RNA is slightly lower than that with the
linear RNA (Fig. 4C).
To further check the sequence dependency, the 27 mer FAM-la-
beled RNAs (FAM-RNA15 and 16) were designed as the model
sequence of the luciferase mRNA. The time course of the crosslink-
ing yields between FAM-RNA15 with o-, et-ODN(10) is shown in
Figure 5A, and those between FAM-RNA16 and o-, et-ODN( 11)
are summarized in Figure 5B. Unexpected findings were that cross-
linked products were formed in the reaction of FAM-RNA15 having
rU at the target site with o-ODN(10) and et-ODN(10), although the
highest yields were obtained with et-ODN(10) toward rC at the tar-
get site of FAM-RNA15 (Fig. 5A). The structure determination of the
adduct with rU is needed for discussion of this unexpected cross-
linking, but an alkylation with the carbonyl group of rU might be
included like a new crosslinking agent for rU.27 On the other hand,
the results with the FAM-RNA16 were consistent with the
prediction of a low reactivity with o-ODN(11) and rC selectivity
with et-ODN(11) (Fig. 5B). These results have clearly suggested
that the sequence dependency of the original AVP analog has been
overcome with the ethylene-linked AVP analog, but nevertheless,
there is the still need to design a suitable RNA sequence for the
target site of crosslinking.
Figure 2. Gel shift assay to determine the crosslinking ability in the reaction buffer
containing 100 mm NaCl, 50 mM MES pH 50. (A) o-ODN(8)-AVP and FAM-labeled
16 mer target RNA. (B) et-ODN(8)-AVP and FAM-labeled 16 mer target RNA.

Y. Taniguchi et al. / Bioorg. Med. Chem. 18 (2010) 2894–2901
2897
Figure 3. Results of alkylating ability of 10
lM 16 mer oligonucleotides (o-ODN(8), me-ODN(8), et-ODN(8), bu-ODN(8)) having AVP derivatives for 1
l
M FAM-labeled target
DNA (A–D) and target RNA (E–H) (dC and rC black line, dT and rU red line, dA and rA green line, dG and rG blue line) in the reaction buffer containing 100 mM NaCl, 50 mM
MES pH 5.0.
Figure 4. (A) The sequence and the second structure of the target RNA. (B) Gel results and crosslinking yield of the sequence o-ODN(9) containing o-AVP to the target RNA. (C)
Gel results and crosslinking yield of the sequence et-ODN(9) containing et-AVP to the target RNA. o- and et-ODN(9): 50 d(TATACTGATCAAATTXTAT). Conditions: 10
lM
reactive ODNs, 1 lM FAM-labeled target RNA, 100 mM NaCl, 50 mM MES, pH 5 at 37 °C.

2898
Y. Taniguchi et al. / Bioorg. Med. Chem. 18 (2010) 2894–2901
Figure 5. Crosslinking ability for the model sequence of luciferase mRNA. (A) The results of 21 mer 20-deoxyoligonucleotides o-ODN(10) and et-ODN(10). (B) The results of 21
mer 20-deoxyoligonucleotides o-ODN(11) and et-ODN(11). Conditions: 10
pH 5 at 37 °C.
lM reactive ODNs, 1 lM FAM-labeled target RNA (Y = rC, rU, rA and rG), 100 mM NaCl, 50 mM MES,
2.3. The thermal stability and conformation of the DNA/RNA
duplexes formed with the ODN containing the AVP analog
estimated by measuring the melting temperature (Tm) of the du-
plex formed between the 21 mer ODN containing the AVP,
ODN(10 or 11) with the 27 mer RNA, RNA15, or RNA16 (Fig. 6A
and B), and the results are summarized in Table 2. The Tm value
of the RNA15/o-ODN(10) duplex was slightly lower than that of
the RNA15/et-ODN(10) duplex, and both were lower than that of
It may be assumed that the sequence dependency regarding the
reactivity and the selectivity may arise from the thermal stability
and the conformation of the duplexes. The thermal stability was
Figure 6. UV-melting curves (A and B) and CD spectra (C and D). A and C show ODN(10)/RNA15 heteroduplex in blue line, o-ODN(10)/RNA15 heteroduplex in black line and
et-ODN(10)/RNA15 heteroduplex in red line. B and D show ODN(11)/RNA16 heteroduplex in blue line, o-ODN(11)/RNA16 heteroduplex in black line and et-ODN(11)/RNA16
heteroduplex in red line. The spectra were recorded in 100 mM NaCl, 50 mM MES buffer, pH 5.0 at room temperature. The concentration of each strand was 1.5
sequences are described in Table 2.
lM. The RNA

Y. Taniguchi et al. / Bioorg. Med. Chem. 18 (2010) 2894–2901
2899
Table 2
ker with a different length to optimize the linker length for better
crosslinking efficiency.
Tm values of functional nucleosides in ODN with complementary RNAa
Among the three linkers, the methylene, ethylene, and butyl-
ene linker, the best improvement in the crosslinking reactivity
as well as the selectivity to rC was obtained with the ethylene lin-
ker analog, et-AVP. To the RNA targets where the original AVP did
not produce a significant crosslinked product, et-AVP produced
the crosslinked products in high yield with a high rC selectivity.
The sequence dependency of the reactivity of the original AVP
to rC in RNA has been overcome by et-AVP, although some limi-
tations still remain. The high reactivity of et-AVP toward rC in
RNA compared to the original AVP cannot be explained either
by the difference in the thermal stability or the difference in
the duplex conformation. Generally, the geometries of the GC
and AT base pairs are almost same in the DNA/DNA duplex and
the DNA/RNA heteroduplex. However, it is apparent that a slight
difference in the geometry of AVP to the rC complex significantly
affected the alkylating ability of the AVP analogs. The thermal
denaturing study and the CD results have shown that the DNA/
RNA heteroduplex is between the A- and B-type conformation
with only a slight distortion because of o-AVP-SMe and et-AVP-
SMe. It has been proposed by MD calculation that base pairs in
duplex RNA experience local opening events into the major
groove, while such events are not simulated in the DNA duplex.28
In the DNA/RNA heteroduplex in the conformation between A-
and B-type, the 2-amino-6-vinylpurine base may cause such a
conformational change to open the AVP-cytosine mismatch base
pair and push the cytosine base into the major groove (Fig. 7,
A1 vs. A2). In such a conformation, the amino group of the cyto-
sine base is no longer in close proximity with the vinyl group of
the original AVP without the linker. On the other hand, the ethyl-
ene-linked 2-amino-6-vinylpurine base (et-AVP) might reach the
cytosine base in a distorted conformation (Fig. 7, B1 vs. B2).
Although there is no experimental evidence, it may be speculated
that the ethylene linker makes the 2-amino-6-vinyl purine unit
flexible in order to adjust the movement of the cytosine base.
In conclusion, it has been clearly shown that the nucleoside ana-
log, et-AVP, having the 2-amino-6-vinylpurine unit through the
ethylene linker is an efficient and selective crosslinking agent to
the rC in the RNA target.
DNA
Target RNA*
Tm (°C)
ODN(10)
RNA15
RNA15
RNA15
RNA16
RNA16
RNA16
55.8
49.4
50.4
64.1
53.3
49.7
o-ODN(10)
et-ODN(10)
ODN(11)
o-ODN(11)
et-ODN(11)
* RNA15: 50 r(CGCUGGGCGUUAAUCAAAGAGGCGAAC) 30.
* RNA16: 50 r(ACGUGAAUUGUUCAACAGUAUGGGCAU) 30.
UV-melting profiles measured using 1.5
50 mM MES buffer, pH 5.0.
a
lM each of the strand in 100 mM NaCl,
the ODN(10)/RNA15 duplex (55.8 °C). In the case of the duplex
with RNA16, the Tm value with o-, et-ODN( 11) was more than
10 °C lower than that with the natural ODN(11) containing dG in-
stead of the AVP derivative. These results indicate that the duplex
is destabilized by the alkylating structure regardless of the pres-
ence or absence of the alkyl linker. The RNA16/et-ODN(11) duplex
produced a crosslinked product with a high rC selectivity, but gave
only low Tm values. Accordingly, it was shown that the thermal sta-
bility of the duplex is not the major contributor of the crosslinking
reactivity.
The circular dichroism spectra were measured to check the con-
formational change in the ODN/RNA heteroduplex. The CD spectra
of the duplex formed with RNA15 and ODN(10), o-ODN(10) and et-
ODN(10) are compared in Figure 6C, and those formed with RNA16
and ODN(11), o-ODN(11), and et-ODN(11) are shown in Figure 6D.
All of these CD spectra indicate that the heteroduplexes are in be-
tween the A- and B-conformation and are similar to each other,
suggesting that a large conformational change is not responsible
for the difference in the crosslinking reactivity.
3. Discussion
This study aimed to develop an efficient crosslinking agent
based on the 2-amino-6-vinylpurine (AVP) unit in order to cross-
link with the RNA molecules. The AVP analogs were designed to
connect the 2-amino-6-vinylpurine to the 20-deoxyribose via a lin-
Figure 7. Schematic representations of the 2-amino-6-vinylpurine and the cytosine base (A1 and A2) and the ethylene-linked 2-amino-6-vinylpurine and the cytosine base
(B1 and B2). In A1 and B1, the cytosine base is located at a natural conformation in DNA/RNA duplex (A1 and A2). A2 and B2 illustrate the cytosine base slightly flipped out to
the major groove. Only bases are shown for clarity.

2900
Y. Taniguchi et al. / Bioorg. Med. Chem. 18 (2010) 2894–2901
4. Materials and methods
4.1. General
and evaporated. The residue was purified by silica-gel column
chromatography (hexane/EtOAc = 1:2 to 1:1 to EtOAc only) to give
as
yellow caramel (330 mg, 0.57 mmol, 94%). 1H NMR
3
a
(400 MHz, CDCl3): d 7.90–7.83 (4H, m), 7.85 (1H, s), 7.25–7.20
(4H, m), 5.46 (1H, d, J = 6.1 Hz), 5.18 (2H, bs), 4.56–4.52 (2H, m),
4.45–4.36 (2H, m), 3.33–3.19 (1H, m), 3.02–3.00 (1H, m), 2.41 (3H,
s), 2.40 (3H, s), 2.38–2.29 (1H, m), 2.17 (3H, s), 2.13–1.96 (1H, m).
FTIR (neat): 1716, 1610 cmꢀ1. ESI-MS: 570.20 (M+H)+.
All reactions were performed under an argon atmosphere. The
1H NMR (400 and 500 MHz) spectra were recorded by a Varian
UNITY-400 and INOVA-500 spectrometer, respectively. The 31P
NMR (161 MHz) spectrum was recorded using 10% phosphoric acid
in D2O for the internal standard at 0 ppm. The IR spectra were ob-
tained using a Perkin Elmer FTIR-SpectrumOne. The high-resolu-
tion mass spectra were recorded by an Applied Biosystems
Mariner System 5299 spectrometer. The MALDI-TOF mass spectra
were recorded by a Bruker Daltonics Microflex instrument.
4.1.3. 9-[2-(20-Deoxy-b-
D-ribofuranosyl)methyl]-6-(2-
methylthioethyl)-2-phenoxyacetyl-aminopurine (4)
Sodium methoxide (153 mg, 2.7 mmol) was added to a solution
of 3 (155 mg, 0.27 mmol) in MeOH (7.0 ml), then the reaction mix-
ture was refluxed at 70 °C for 1 h. The reaction mixture was
quenched with a satd ammonium chloride solution and extracted
with EtOAc. The organic layer was continuously washed with
water and brine, dried over Na2SO4, and evaporated. The residue
was dissolved in pyridine/CH2Cl2 = 1:3 (6.4 ml) and cooled to 0 °C
in an ice bath. Trimethylsilyl chloride (0.34 ml, 2.69 mmol) was
added to this mixture. After stirring for 1.5 h at room temperature,
the mixture was cooled to 0 °C again. Phenoxyacetyl chloride
4.1.1. 2-Deoxy-1-tosyloxymethyl-3,5-O-ditoluoyl-b-D-
ribofuranose (2)
A solution of 1 (814 mg, 2.15 mmol), 1,4-dioxane (20 ml), and a
36% aqueous HCl solution (1.87 ml) was refluxed at 80 °C for 2 h.
The reaction mixture was quenched with a satd NaHCO3 solution
and extracted with CHCl3. The organic layer was dried over Na2SO4,
and evaporated. The residue was dissolved in THF and cooled to
0 °C. The BH3–THF complex (9.7 ml, 9.65 mmol) was slowly added
to this solution at 0 °C. After stirring for 1.5 h at room temperature,
the reaction mixture was quenched with a 10% aqueous HCl solu-
tion and extracted with CHCl3. The organic layer was dried over
Na2SO4, and then evaporated. The residue was purified by silica-
gel column chromatography (hexane/EtOAc = 1:1) to give a color-
less oil (680 mg, 1.72 mmol, 80% for two steps). p-Toluenesulfonyl
chloride (p-TsCl) (476 mg, 2.5 mmol) was added to a solution of the
colorless oil (600 mg, 1.56 mmol) in pyridine (6.0 ml) at 0 °C. After
stirring for 24 h, the reaction mixture was poured into ice water
and extracted with CHCl3. The organic layer was dried over Na2SO4,
and evaporated. The residue was purified by silica-gel column
chromatography (hexane/EtOAc = 1:1) to give 2 as a pale yellow
oil (693 mg, 1.29 mmol, 83%). 1H NMR (400 MHz, CDCl3): d 7.92–
7.87 (4H, m), 7.77–7.75 (2H, m), 7.28–7.22 (6H, m), 5.44 (1H, d,
J = 4.0 Hz), 4.48–4.30 (2H, m), 4.21–4.09 (2H, m), 2.41 (3H, s),
2.40 (3H, s), 2.35 (3H, s), 2.24–2.12 (2H, m). FTIR (neat): 1718,
1270 cmꢀ1. ESI-MS: 539.08 (M+H)+.
(37 ll, 0.39 mmol) was added to this mixture at 0 °C. The reaction
mixture was stirred for 3 h at the same temperature, and diluted
with CH3OH (1.6 ml). After stirring for 12 h at room temperature,
the reaction mixture was diluted with CHCl3. The organic layer
was washed with water, dried over Na2SO4, and evaporated. The
residue was purified by silica-gel column chromatography
(CHCl3/CH3OH = 10:1) to give 4 as a colorless caramel (86.4 mg,
0.18 mmol, 68% for two steps). 1H NMR (400 MHz, CDCl3): d 9.10
(1H, s), 8.09 (1H, s), 7.34–7.30 (2H, m), 7.05–7.00 (3H, m), 6.83
(1H, d, J = 6.3 Hz), 4.78–4.74 (2H, m), 4.54–4.50 (1H, m), 4.44–
4.39 (2H, m), 4.30–4.27 (1H, m), 3.85 (1H, d, J = 3.4 Hz), 3.64–
3.55 (2H, m), 3.43–3.39 (2H, m), 3.07–3.03 (2H, m), 2.16 (3H, s),
2.13–2.09 (1H, m), 1.96–1.92 (1H, m). FTIR (neat): 3379, 2918,
1691, 1600 cmꢀ1. ESI-MS: 474.24 (M+H)+.
4.1.4. 9-[2-(20-Deoxy-30-O-N,N-diisopropylcyanoethylphosphor-
amidyl-50-O-(4,40-dimethoxytrityl)b-
D-ribofuranosyl)methyl]-6-
(2-methylthioethyl)-2-phenoxyacetylaminopurine (5)
DMTrCl (92 mg, 0.26 mmol) was added to a solution of 4
(96 mg, 0.2 mmol) in pyridine containing molecular sieves 4 Å.
After stirring for 1 h, the reaction mixture was diluted with
CHCl3, and then washed with water and brine. The organic layer
was washed with water and brine, dried over Na2SO4, evapo-
rated, and then the residue was purified by silica-gel column
chromatography (CHCl3 containing 0.5% pyridine) to give the
DMTr-protected compound (96.1 mg, 0.12 mmol, 61%). Subse-
4.1.2. 2-Amino-9-[2-(20-deoxy-b-
D-ribofuranosyl)methyl]-6-(2-
methylthioethyl)purine (3)
2-Amino-6-chloropurine (432 mg, 2.55 mmol) was added to a
solution of potassium t-butoxide (301 mg, 2.55 mmol) in dimethyl-
sulfoxide (DMSO) (5.0 ml). After stirring for 1.5 h, a solution of 2
(686 mg, 1.27 mmol) in DMSO (5.0 ml) was added to the above mix-
ture, then the reaction mixture was refluxed for 3 h at 80 °C. The
reaction mixture was quenched with satd ammonium chloride solu-
tion and extracted with EtOAc. The organic layer was successively
washed with water and brine, dried over Na2SO4, and evaporated.
The residue was purified by silica-gel column chromatography (hex-
ane/EtOAc = 1:1) to give a colorless oil (357 mg, 0.67 mmol, 52%).
Tetrakis(triphenylphosphine)palladium(0) (70.5 mg, 0.061 mmol),
lithium bromide (64.9 mg, 0.73 mmol), potassium bicarbonate
(50.8 mg, 0.37 mmol), and 2,4,6-trivinylcyclotriboroxane pyridine
complex (212 mg, 0.85 mmol) were added to a solution of the color-
less oil (327 mg, 0.61 mmol) in water/dioxane = 1:3 (11.5 ml). The
reaction mixture was refluxed for 1 h at 120 °C, and quenched with
a satd ammonium chloride solution and extracted with EtOAc. The
organic layer was successively washed with water and brine, dried
over Na2SO4, and evaporated to give a brown caramel (506 mg). This
brown caramel solution in acetonitrile (5.0 ml) was treated with
sodium methanethiolate solution (0.43 ml, 0.92 mmol) for 15 min.
The reaction mixture was diluted with CHCl3. The organic layer
was continuously washed with water and brine, dried over Na2SO4,
quently, iP2NEt (67
of the DMTr-compound (50 mg, 0.06 mmol) and iPr2NP(-
Cl)OC2H4CN (43 l, 0.19 mmol) in CH2Cl2 at 0 °C. After stirring
ll, 0.39 mmol) was added to this solution
l
for 30 min, the reaction mixture was diluted with a satd NaH-
CO3 solution, and extracted with EtOAc. The organic layer was
dried over Na2SO4, evaporated, and then the residue was puri-
fied by flash silica-gel column chromatography (hexane/
EtOAc = 2:1) to give 5 as a white powder (39 mg, 0.04 mmol,
62%). 1H NMR (400 MHz, CDCl3): d 8.88 (1H, bs), 8.12 (1H, s),
7.43–7.28 (10H, m), 7.23–7.19 (2H, m), 7.06–7.01 (3H, m),
6.83 (4H, dd, J = 7.0, 1.5 Hz), 4.80–4.78 (2H, m), 4.59–4.39 (3H,
m), 4.19–4.11 (3H, m), 3.79 (3H, s), 3.78 (3H, s), 3.74–3.68
(1H, m), 3.66–3.57 (1H, m), 3.54–3.48 (2H, m), 3.46–3.39 (2H,
m), 3.16–3.15 (2H, m), 3.07 (2H, t, J = 7.63 Hz), 2.75 (1H, t,
J = 6.41 Hz), 2.40 (1H, t, J = 6.41 Hz), 2.18 (3H, s), 1.14 (6H, d,
J = 6.72 Hz), 1.10 (6H, d, J = 6.72 Hz). FTIR (neat): 2967, 2364,
1697, 1600 cmꢀ1. ESI-MS: 976.71 (M+H)+. 31P NMR (161 MHz,
CDCl3): d 148.7, 148.5.

Y. Taniguchi et al. / Bioorg. Med. Chem. 18 (2010) 2894–2901
2901
4.2. Oligonucleotides (6)
were converted to a molar ellipticity, which was performed by the
JASCO standard analysis software.
The 20-deoxy oligonucleotides (ODNs) containing the AVP deriv-
ative were synthesized using an automated DNA synthesizer (Ap-
Acknowledgments
plied Biosystems 394 DNA/RNA synthesizer) on a 1.0 lmol scale.
The synthesized ODNs were cleaved from CPG by 28% aqueous
ammonium hydroxide at room temperature and purified by HPLC
(conditions: column, Nacalai Tesque COSMOSIL 5C18AR-II
10 ꢁ 250 mm; flow rate, 3.0 ml/min; buffer A, 0.1 M TEAA; B,
CH3CN; B concentration 10–40%/20 min, 40–100%/30 min, linear
gradient). The DMTr-protecting groups of the purified ODNs were
removed by 10% acetic acid and the formed DMTrOH was extracted
with Et2O. The structure of the synthetic ODNs was confirmed by a
MALDI-TOF MS measurement. The MALDI-TOF MS results are sum-
marized in Table 1.
This work was supported by a Grant-in-Aid for Scientific Re-
search (S) from the Japan Society for Promotion of Science, and
CREST from the Japan Science and Technology Agency.
References and notes
1. Matsubara, H.; Takeuchi, T.; Nishikawa, E.; Yanagisawa, K.; Hayashita, Y.; Ebi,
H.; Yamada, H.; Suzuki, M.; Nagino, M.; Nimura, Y.; Osada, H.; Takahashi, T.
Oncogene 2006, 26, 6099.
2. Stenvang, J.; Lindow, M.; Kauppinen, K. Biochem. Soc. Trans. 2008, 36, 1197.
3. Horwich, M. D.; Zamore, P. D. Nat. Protoc. 2008, 3, 1537.
4. Fei, J. L.; Feiei, G.; Ming, L. Y.; Liu, Y. J. Drug Target 2008, 16, 688.
5. Yu, R. Z.; Lemonidis, K. M.; Graham, M. J.; Matson, J. E.; Crooke, R. M.; Tribble, D.
L.; Wedel, M. K.; Levin, A. A.; Geary, R. S. Biochem. Pharmacol. 2009, 77, 910–919.
6. Crooke, S. T. In Antisense Drug Technology Principles, Strategies, and Applications,
2nd ed.; CRC Press: Boca Raton, 2007.
4.3. The crosslinking reaction and the analysis by denatured
polyacrylamide gel
7. Aboul-Fadl, T. Expert Opin. Drug Discov. 2006, 1, 285.
8. Elmén1, J.; Lindow, M.; Schütz, S.; Lawrence, M.; Petri, A.; Obad, S.; Lindholm,
M.; Hedtjärn, M.; Hansen, H. F.; Berger, U.; Gullans, S.; Kearney, P.; Sarnow, P.;
Straarup, E. M.; Kauppinen, S. Nature 2008, 452, 896.
9. Elmén, J.; Lindow, M.; Silahtaroglu, A.; Bak, M.; Christensen, M.; Lind-Thomsen,
A.; Hedtjärn, M.; Hansen, J. B.; Hansen, H. F.; Straarup, E. M.; McCullagh, K.;
Kearney, P.; Kauppinen, S. Nucleic Acids Res. 2008, 36, 1153.
10. Tillman, L. G.; Geary, R. S.; Hardee, G. E. J. Pharm. Sci. 2008, 97, 225.
11. Davis, S.; Propp, S.; Freier, S. M.; Jones, L. E.; Serra, M. J.; Kinberger, G.; Bhat, B.;
Swayze, E. E.; Bennett, C. F.; Esau, C. Nucleic Acids Res. 2009, 37, 70.
12. Nagatsugi, F.; Uemura, K.; Nakashima, S.; Maeda, M.; Sasaki, S. Tetrahedron Lett.
1995, 36, 421.
13. Nagatsugi, F.; Uemura, K.; Nakashima, S.; Maeda, M.; Sasaki, S. Tetrahedron
1997, 53, 3035.
14. Nagatsugi, F.; Kawasaki, T.; Usui, D.; Maeda, M.; Sasaki, S. J. Am. Chem. Soc.
1999, 121, 6753.
15. Kawasaki, T.; Nagatsugi, F.; Ali, M. M.; Maeda, M.; Sugiyama, K.; Hori, K.;
Sasaki, S. J. Org. Chem. 2005, 70, 14.
16. Ali, M. M.; Oishi, M.; Nagatsugi, F.; Mori, K.; Nagasaki, Y.; Kataoka, K.; Sasaki, S.
Angew. Chem., Int. Ed. 2006, 45, 3136.
The methylthio group of ODN in the oligonucleotides 6 was
converted to the vinyl group 7 by MMPP (2.0 equiv) and aqueous
sodium hydroxide at room temperature. The RNase OUT (40 unit),
reaction buffer (final concd; 100 mM NaCl and 50 mM MES, pH 5)
and FAM-labeled target DNA or RNA (50 pmol) were added to the
above mixture at 0 °C. The reaction mixture was incubated at 37 °C,
and the reaction was stopped by mixing with a buffer (95% form-
amide, 20% EDTA, 0.05% bromophenol blue) at the appropriate
time. The reaction progress was analyzed by electrophoresis with
15% denatured polyacrylamide gel containing 7 M urea, and the
intensity of FAM-labeled bands were quantified by a luminescent
image analyzer LAS-3000 (Fujifilm).
4.4. Thermal denaturing study (Tm value)
17. Ali, M. M.; Nagatsugi, F.; Sasaki, S.; Nakahara, R.; Maeda, M. Nucleosides,
Nucleotides, Nucleic Acids 2006, 25, 159.
18. Nagatsugi, F.; Nakahara, R.; Inoue, K.; Sasaki, S. Arch. Pharm. Chem. Life Sci.
2008, 341, 562.
19. Ali, M. M.; Imoto, S.; Li, Y.; Sasaki, S.; Nagatsugi, F. Bioorg. Med. Chem. 2009, 17,
2859.
20. Nagatsugi, F.; Sasaki, S.; Miller, P. S.; Seidman, M. M. Nucleic Acids Res. 2003, 31,
e31.
21. Nagatsugi, F.; Matsuyama, Y.; Maeda, M.; Sasaki, S. Bioorg. Med. Chem. Lett.
2002, 12, 487.
22. Nagatsugi, F.; Usui, D.; Kawasaki, T.; Maeda, M.; Sasaki, S. Bioorg. Med. Chem.
Lett. 2001, 11, 343.
A mixture of the corresponding ODN 6 (1.5
lM) and 27 mer of
the complimentary RNA (1.5 M) in the buffer (100 mM NaCl and
l
50 mM MES, pH 5) was heated at 85 °C for 10 min then slowly
cooled. The melting curve was obtained by warming a sample at
the rate of 1.0 °C/min from 20 to 80 °C with monitoring at
260 nm by
a UV–vis spectrophotometer, DU 800 (Beckman
Coulter). The data were analyzed by the melt curve processing
program, MeltWin v. 3.5.
23. Grünefeld, P.; Richert, C. J. Org. Chem. 2004, 69, 7543.
24. Adamo, M. F. A.; Adlington, R. M.; Baldwin, J. E.; Day, A. L. Tetrahedron 2004, 60,
841.
25. Nagatsugi, F.; Ogata, Y.; Imoto, S.; Sasaki, S. Heterocycles 2007, 73, 493.
26. Fan, F.; Gaffney, B. L.; Jones, R. A. Org. Lett. 2004, 6, 2555.
27. Hattori, K.; Hirohama, T.; Imoto, S.; Kusano, S.; Nagatsugi, F. Chem. Commun.
2009, 6463.
4.5. Circular dichroism (CD) spectroscopy
The CD spectra were measured by a JASCO J-720W spectropho-
tometer at room temperature in a cylindrical quartz cell with a
path length of 0.1 cm. The samples were prepared in the same
manner as described in the thermal denaturing study. The Spectra
28. Pan, Y.; MacKerell, A. D., Jr. Nucleic Acids Res. 2003, 31, 7131.