Synthesis and structural characterization of diastereomeric isomers of RNA trimer adenylyl(3′-5′)adenylyl(3′-5′)adenosine

Article abstract of DOI:10.1016/j.tetasy.2005.07.026

We have synthesized the diastereomeric isomers of adenylyl(3′- 5′)adenylyl(3′-5′)adenosine (ApApA), and investigated their helical structures and hybridization properties with d-poly(U) by circular dichroism (CD) and UV melting experiments. The configurat

Full text of DOI:10.1016/j.tetasy.2005.07.026

Tetrahedron:
Asymmetry
Tetrahedron: Asymmetry 16 (2005) 2908–2917
Synthesis and structural characterization of diastereomeric
isomers of RNA trimer adenylyl(30-50)adenylyl(30-50)adenosine
Hidehito Urata,* Hisafumi Hara, Yoshihiro Hirata, Norihiko Ohmoto and Masao Akagi*
Osaka University of Pharmaceutical Sciences, 4-20-1 Nasahara, Takatsuki, Osaka 569-1094, Japan
Received 13 June 2005; accepted 21 July 2005
Available online 26 August 2005
Abstract—We have synthesized the diastereomeric isomers of adenylyl(30-50)adenylyl(30-50)adenosine (ApApA), and investigated
their helical structures and hybridization properties with D-poly(U) by circular dichroism (CD) and UV melting experiments.
The configuration of the 50-end residue of ApApA has little effect on the helical structure. The configuration of the 30-end residue
has little effect on the stability of the triplex, and the chiral modifications cause complicated effects on the helical structure of ApApA
and its triplex-forming ability with D-poly(U), depending on the site of the modification.
ꢀ 2005 Elsevier Ltd. All rights reserved.
1. Introduction
handed helical structure and forms a right-handed tri-
plex with D-poly(U), and the triplex is less stable than
that with natural D-(ApA).19 We found that heterochiral
ALpAD and ADpAL form the right- and left-handed
helical structure, respectively. ALpAD formed the triplex
with D-poly(U) while its thermal stability was quite sim-
ilar to the D-(ApA)ÆD-poly(U) triplex, whereas the tri-
plex of ADpAL with D-poly(U) showed a quite similar
stability to that of L-(ApA), whose stabilities were sig-
nificantly lower than those of D-(ApA) and ALpAD.17
This means that the configuration of the 30-end residue
of ApA primarily determines the helical sense of ApA
and its triplex-forming ability with poly(U). Here, we
report the synthesis of the eight diastereomeric isomers
of ApApA and the structural characterization, in partic-
ular, the effects of configuration of each residue on the
helical structure of ApApA.
Biological molecules such as sugars and amino acids are
chiral, and hence enantiomers exist. However, organ-
isms strictly discriminate between enantiomers, and usu-
ally synthesize and use only one of them. In particular,
organisms strictly discriminate both enantiomers of
ribose, the chiral component of nucleic acids. Nucleic
acids consist of only D-ribose as the sugar unit, and
the L-isomer has not yet been found in nature. It was
thus considered that the strict chiral selection during
the early stage of chemical evolution would have been
essential for the emergence of functional RNA mole-
cules. Since the chiral homogeneity is thought to be
essential for the formation of higher order structures
of biopolymers and their interaction with specific
ligands, it would be useful to understand the structures
and properties of chirally modified RNA molecules for
considering the establishing process and the significance
of the homochirality of RNA. Several groups have
reported the structure and properties of mirror image
DNAs1,2 and RNAs.3–8 However, there is little informa-
tion about the structure and properties of heterochiral
nucleic acids,9–16 especially heterochiral RNAs.17,18 In
this context, we have investigated the helical structure
of the diastereomeric isomers of adenylyl(30-50)adeno-
sine (ApA).17 Tsꢀo et al. reported the properties of
unnatural homochiral L-(ApA), which adopts a left-
2. Results and discussion
2.1. Synthesis of trimers
The diastereomeric isomers of ApApA were synthesized
by a phosphotriester method20 as shown in Schemes 1–5.
D- and L-N6-Benzoyladenosines were converted to
20,30-dibenzoates 3 and 10, as well as 20-O-tetrahydro-
furanyl-protected 30-nucleotides 7 and 14. Compounds
3 and 10 were allowed to condense with either 7 or 14
in the presence of MSNT [1-(2-mesitylenesulfonyl)-3-
nitro-1,2,4-triazolide] followed by deprotection of the
*
Corresponding authors. Tel.: +81 72 690 1089; fax: +81 72 690 1005
(H.U.); e-mail: urata@gly.oups.ac.jp
0957-4166/$ - see front matter ꢀ 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetasy.2005.07.026

H. Urata et al. / Tetrahedron: Asymmetry 16 (2005) 2908–2917
2909
BzA
BzA
BzA
HO
DMTO
HO
O
O
O
i
ii, iii
OH
OH
OH
OH
OBz OBz
1
2
3
iv
BzA
BzA
BzA
RO
DMTO
O
O
O
O
O
viii
v, vi, vii
TIPDS
O
O
O
OH
OH
O
O
P
O
O
O-
Et3NH
Cl
4
5 R = H-
6 R = DMT-
7
Scheme 1. Reagents and conditions: (i) DMT–Cl, pyridine; (ii) Bz2O, DMAP, CH2Cl2; (iii) 2% benzenesulfonic acid, MeOH/CH2Cl2; (iv) 1,3-
dichloro-1,1,3,3-tetraisopropyldisiloxane, pyridine; (v) 2,3-dihydrofuran, PPTS, CH2Cl2; (vi) 1 M TBAF/THF; (vii) DMT–Cl, pyridine;
(viii) o-chlorophenylphosphorditriazolide, pyridine, dioxane.
BzA
BzA
BzA
OH
ODMT
OH
O
O
O
i
ii, iii
OH
OH
OH
OH
OBz OBz
8
9
10
BzA
BzA
BzA
O
OH
ODMT
O
O
O
O
v, vi
vii
TIPDS
O
O
OH
O
O
OH
O
O
P
O
O-
Et3NH
Cl
11
12 R = H-
13 R = DMT-
14
Scheme 2. Reagents and conditions: (i) DMT–Cl, pyridine; (ii) Bz2O, DMAP, CH2Cl2; (iii) 2% benzenesulfonic acid, MeOH/CH2Cl2; (iv) 1,3-
dichloro-1,1,3,3-tetraisopropyldisiloxane, pyridine; (v) 2,3-dihydrofuran, PPTS, CH2Cl2; (vi) 1 M TBAF/THF; (vii) DMT–Cl, pyridine;
(viii) o-chlorophenylphosphorditriazolide, pyridine, dioxane.
DMT group to give 50-OH free dimers 15–18. Subse-
quently, these dimers were allowed to condense with
either 7 or 14 in the presence of MSNT to give fully pro-
tected trimers 19–26. After deprotection of these trimers
with concentrated ammonium hydroxide at 55 ꢁC for
8 h, followed by 80% acetic acid treatment, the purifica-
tion was conducted by using reversed phase HPLC. The
purities of the trimers were more than 95%.
and ALpAD. The trimers other than ALpALpAD,
ADpALpAL and L-(ApApA) could be completely hydro-
lyzed to adenosine and 50-AMP by these enzymes, and
their molar extinction coefficients were determined by
comparing absorbances before and after the hydrolysis
(Table 1).
We tried to characterize the isomers of ApApA by ana-
lyzing the digestion products obtained by the reaction
with these enzymes. However, SVPD, which is a 30-exo-
nuclease, showed the unexpected reactivity for trimers
that contain the L-adenosine residue at the 30-end. An
example is shown in Figure 1. ADpADpAL was expected
to be resistant to digestion by SVPD, however, the reac-
tion afforded degradation products in which both of the
phosphodiester bonds were cleaved (Fig. 1). Damha
et al. reported that SVPD cleaves L-oligomers very
2.2. Characterization of trimers
The reactivity of nuclease P1 and snake venom phos-
phodiesterase (SVPD) towards each isomer of diastereo-
meric ApAs has been well characterized.17 Nuclease P1
cleaves the phosphodiester bonds at the P–O(30) bond
of ADpAD and ADpAL, whereas SVPD cleaves the
phosphodiester bonds at the P–O(30) bond of ADpAD

2910
H. Urata et al. / Tetrahedron: Asymmetry 16 (2005) 2908–2917
BzA
7 or 14
HO
Cl
19 (DDD), 20 (LDD)
21 (DLD), 22 (LLD)
15 (DD)
16 (LD)
O
O
MSNT
O
O
O P
O
O
7 or 14
MSNT
BzA
O
7 or 14
MSNT
17 (DL)
18 (LL)
23 (DDL), 24 (LDL)
25 (DLL), 26 (LLL)
OBz OBz
15
i) 7, MSNT, pyridine
ii ) ZnBr2
7 or 14
MSNT
BzA
3
OH
O
i) 14, MSNT, pyridine
Scheme 5.
ii ) ZnBr2
O
Cl
O
O
O P
O
O
BzA
Table 1. Hypochromicity and molar extinction coefficients of the
trimers
O
Trimer
Hypochromicity (%)
e at 260nm (L/mol cm)
D-(ApApA)
ALpA0DpAD
ADpALpAD
ALpALpAD
ADpADpAL
ALpADpAL
ADpALpAL
L-(ApApA)
26.9
22.036,80
21.9
—
23.6
23.2
—
35,400
OBz OBz
16
36,800
36,300a
36,300
36,500
36,800a
35,400a
Scheme 3.
—
BzA
a Not determined. The values were assumed to be the same as the
corresponding enantiomers.
HO
O
O
O
Cl
O
O P O
O
BzA
O
OBz OBz
17
i) 7, MSNT, pyridine
ii) ZnBr2
BzA
10
OH
Cl
O
i) 14, MSNT, pyridine
ii) ZnBr2
O
O
O
O P
O
O
BzA
O
Figure 1. Reversed phase HPLC chromatogram of the reaction of
ADpADpAL with SVPD.
OBz OBz
18
find another enzyme that hydrolyzed the phosphodiester
bonds of ALpAD as well as SVPD does.
Scheme 4.
We thus changed the strategy to characterize the tri-
mers. Each trimer was hydrolyzed with aqueous potas-
sium hydroxide solution to give adenosine, 20- and 30-
AMPs, and the resulting mixture treated with calf intes-
tine alkaline phosphatase (CIAP) to give adenosine. The
enantiomers of adenosine were then separated by using
chiral HPLC to evaluate the molar ratios of D- and
L-adenosines (Fig. 2), and the trimers classified by the
slowly compared to the corresponding D-oligomers and
also endonucleolytically cleaves the heterochiral oligo-
mers that contain the L-nucleotide residue at the 30-
end, such as d(TTTTTCL).13 Our findings seem to be
similar to those reported by Damha et al. We tested sev-
eral enzymes (nuclease S1, micrococcal nuclease, RNase
T2) from commercial sources, however, we could not

H. Urata et al. / Tetrahedron: Asymmetry 16 (2005) 2908–2917
2911
Figure 2. HPLC chromatograms for the chiral resolution of D- and/or
L-adenosine obtained by the alkaline hydrolysis of the trimers followed
by treatment with APH.
D/L ratios of their constituent adenosine as shown in
Table 2, in which the expected degradation products
by nuclease P1 for each trimer are also shown. Although
some trimers afford the same degradation products as
each other by treatment with nuclease P1, for example,
both ADpADpAD and ADpADpAL afford 50-AMP and
adenosine, the classification of the trimers by the D/L
ratios made it possible to distinguish between them by
reversed phase HPLC. Figure 3 shows the reversed
phase HPLC chromatograms of the trimers after nucle-
ase P1 digestion. All the trimers gave the expected deg-
radation products and were successfully characterized
by a combination of the classification by the D/L ratios
and nuclease P1 treatment.
2.3. CD spectra of trimers
In order to understand the effects of chiral modification
on the helical structure of ApApA, CD experiments
Figure 3. Reversed phase HPLC chromatograms for the reactions of
the trimers with nuclease P1. D/L ratios are (a) 3:0or 0:3, (b) 2:1 and
(c) 1:2.
Table 2. Classification of the trimers by the D/L ratios of constituent
adenosine and expected products after nuclease P1 digestiona
D/L Ratio
Trimer
Expected products after
nuclease P1 digestion
were conducted with the results shown in Figure 4.
The natural trimer D-(ApApA) shows a typical conser-
vative CD spectrum by exciton interaction, with positive
and negative bands at 267 and 250nm, respectively. The
positive sign of the long-wavelength component of the
conservative bands indicates a right-handed helical twist
of the transition moment of the bases when the bases are
nearly perpendicular to the helix axis.21 The spectrum of
L-(ApApA) shows the opposite sign but the same inten-
sity as that of D-(ApApA). These results suggest that D-
(ApApA) has a right-handed helical sense and L-(ApA-
pA) is its mirror image, as expected. In the case of the
D:L = 3:0
D:L = 0 :3
D:L = 2:1
D-(ApApA)
L-(ApApA)
50-AMP + adenosine
L-(ApApA)
ALpADpAD
ADpALpAD
ADpADpAL
50-AMP + ALpAD
Adenosine + pALpAD
50-AMP + adenosine
D:L = 1:2
ALpALpAD
ALpADpAL
ADpALpAL
ALpALpAD
50-AMP + ALpAD
Adenosine + pALpAL
a The configuration of the monomers in the expected degradation
products are not indicated because of the inability of the reversed
phase HPLC in separating enantiomers.

2912
H. Urata et al. / Tetrahedron: Asymmetry 16 (2005) 2908–2917
of the other residues is essential for the formation of the
helical structure of ApApA.
2.4. Triplex stability of trimers with D-poly(U)
Figure 5 shows UV mixing curves of the ApApA iso-
mers with D-poly(U). All the cases showed an absorp-
tion minimum at A:U = 1:2, and it is thus very likely
that all the ApApA isomers form a triple helix with D-
poly(U), as seen in the case of the diastereomeric iso-
mers of ApA.17 Figure 6 shows the CD spectra of these
triplexes, which indicate that all the triplexes adopt a
right-handed helical structure. To evaluate the thermal
stability of each ApApA-poly(U) triplex, UV melting
experiments were conducted with the results shown
in Figure 7a. The triplexes other than ADpALpADÆ
2poly(U) and ADpALpALÆ2poly(U) show a cooperative
sigmoidal curve, suggesting a two-state transition from
the triplex to the single strand. However, ADpALpADÆ
2poly(U) and ADpALpALÆ2poly(U) reveal a non-
cooperative curve, which means a complex melting
process. Figure 7b shows the first derivative plots of
the melting curves of Figure 7a. ADpALpADÆ2poly(U)
exhibits two maxima (7 and 18 ꢁC) as well as ADpALpA-
LÆ2poly(U) (7 and 12 ꢁC). The presence of these two
maxima means that two kinds of complex formations
between the trimers and D-poly(U) exist. The maximum
at 7 ꢁC suggests a very weak complex formation between
the trimers and D-poly(U).
Figure 4. CD spectra of the trimers at 0 ꢁC. Samples contained 40 lM
trimer in 0.1 M NaCl, 10 mM sodium phosphate, pH 7.0.
heterochiral trimers, ALpADpAD and ADpALpAL also
exhibit spectra, which are symmetrical to each other.
Although these trimers show somewhat decreased CD
intensity when compared to D- and L-(ApApA), the
spectra are still considered to be conservative. Thus, it
is very likely that ALpADpAD has the right-handed heli-
cal sense, whereas ADpALpAL has the left-handed heli-
cal sense. Conversely, the other heterochiral ApApAs
exhibit the non-conservative CD spectra in extremely
weakened strength. It is thus unlikely that these trimers
form a definite helical structure. These results mean that
the trimers possessing the same chirality at the internal
and 30-end residues form the definite helical structure,
in other words, the chirality of the 50-end residue of
ApApA hardly affects the helical structure, whereas that
Table 3 shows the Tm values of the triplexes and their
relationship with the configuration of each residue of
the trimers. With respect to the configuration of the
internal residue, the trimers possessing the D-configura-
tion at the internal residue form the more stable triplexes
with D-poly(U) than those possessing L-configuration at
the same site. Among the trimers possessing the internal
D-configuration, those possessing D-configuration at the
50-end residue (DDD and DDL) form the more stable
Figure 5. UV mixing curves of the trimers with D-poly(U) in 10mM MgCl 2, 10mM Tris–HCl, pH 7.5 at 0 ꢁC. Total base concentration is 150 lM.

H. Urata et al. / Tetrahedron: Asymmetry 16 (2005) 2908–2917
2913
Figure 6. CD spectra of the triplexes of the trimers with D-poly(U) in 10mM MgCl 2, 10mM Tris–HCl, pH 7.5 at 0 ꢁC. Total base concentration is
90 lM.
Table 3. Relationship between the configuration of each residue of the
trimers and their triplex stability with D-poly(U)
Trimer
Chirality
Tm (ꢁC)
ADpADpAD
ADpADpAL
ALpADpAD
ALpADpAL
ALpALpAD
ALpALpAL
ADpALpAD
ADpALpAL
DDD
DDL
LDD
LDL
LLD
LLL
29.8
27.4
24.2
21.1
20.5
16.7
7.3
DLD
DLL
6.6
triplexes with D-poly(U) than those possessing L-config-
uration at the same site (LDD, LDL). However, among
the trimers possessing the internal L-configuration, those
possessing L-configuration at the 50-end residue (LLD,
LLL) form the more stable triplexes with D-poly(U)
than those possessing D-configuration at the same site
(DLD, DLL). Therefore, the same configuration at the
50-end and internal residues is somewhat more advanta-
geous for the triple helix formation of ApApA with D-
poly(U) than the different configuration at the same
sites.
In the case of the ApA diastereomeric isomers, D-(ApA)
and ALpAD were considered to have the right-handed
helical sense, whereas L-(ApA) and ADpAL were consid-
ered to have the left-handed helical sense from their
CD spectra. Furthermore, the triplexes of D-(ApA) and
ALpAD with D-poly(U) showed similar melting behaviors
(Tm value; 14.7 and 13.7 ꢁC, respectively), while their
stability was remarkably higher than that of the triplexes
of L-(ApA) and ADpAL with D-poly(U), which showed
a similar thermal stability (Tm value; 5.7 and 6.6 ꢁC,
respectively).17 It was thus thought that the configura-
tion of the 30-end residue of the dimers primarily deter-
mines the helical sense of ApA and its triplex stability
Figure 7. (a) UV melting profiles of the triplexes of the trimers with D-
poly(U). Base concentrations for trimer and D-poly(U) are 30and
60 lM, respectively, in 10mM MgCl 2, 10mM Tris–HCl, pH 7.5. (b)
First derivative plots of the UV melting curves of the triplexes of the
trimers with D-poly(U).

2914
H. Urata et al. / Tetrahedron: Asymmetry 16 (2005) 2908–2917
with D-poly(U). The present results indicate that the chi-
ral modifications cause complicated effects on the helical
structure of ApApA and its ability for the triplex forma-
tion with D-poly(U). Compared with homochiral D-
(ApApA) and L-(ApApA), the CD strengths of ALpAD-
pAD and ADpALpAL are significantly decreased,
although the spectra show the conservative Cotton band
as D-(ApApA) and L-(ApApA) (Fig. 4), respectively.
Thus, the chiral modification at the 50-end residue does
not induce such a dramatic structural alteration for
ApApA. However, the chiral modifications at the inter-
nal and 30-end residues induce dramatic distortions on
the helical structure. ADpADpAL, ADpALpAD and their
enantiomers show the CD spectra in extremely weak-
ened strength, which are no longer conservative, and
the trimers are not considered to form a definite helical
structure. Therefore, the configuration of the internal and
30-end residues is a crucial factor for maintaining the
helical structure of ApApA. From the viewpoint of the
triplex formation with D-poly(U), the configuration of
the internal and 50-end residues of ApApA plays impor-
tant roles for the stability, while the configuration of the
30-end residue had little effect on the triplex stability.
solution, followed by heating. For silica gel column
chromatography, either Wakogel FC-40 or C-400HG
(Wako Pure Chemical Industries) was used. The elution
was carried out by increasing the amounts of methanol
1
in chloroform unless otherwise noted. H NMR spectra
were measured on either a Varian Gemini 200 or Mer-
cury 300 spectrometer with tetramethylsilane as the
internal standard. Nuclease P1, snake venom phospho-
diesterase (SVPD) and calf intestine alkaline phospha-
tase (CIAP) were purchased from Yamasa Co. (Chiba,
Japan), Roche Diagnostics (Mannheim, Germany) and
Takara Bio Inc. (Kyoto, Japan), respectively. Reversed
phase HPLC was performed on a column of lBonda-
˚
sphere C18 5 lm 10 0 Acolumn (3.9 · 150mm, Waters),
and the elution was performed with a linear gradient of
acetonitrile (0–10%) in 50 mM potassium phosphate
(pH 4.0) on a Shimadzu LC-10A HPLC system. The
synthesis of L-ribose was reported previously.23 L-
Ribose was converted to 1-O-acetyl-2,3,5-O-tribenzoyl-L-
ribofuranose according to the literature procedure for
the corresponding D-isomer.24 N6-Benzoyl-L-adenosine
was synthesized by the trimethylsilyl trifluoromethane-
sulfonate-catalyzed coupling of silylated N6-benzoyl-
adenine with 1-O-acetyl-2,3,5-O-tribenzoyl-L-ribofur-
anose.25
In the template-directed oligomerization of activated
mononucleotides, oligomerization of the monomers pro-
ceeded in the 50!30 direction, and incorporation of an
L-nucleotide at the 30-end of oligomers prevents further
elongation by incorporation of either monomer.22 The
trimers without stable triplex-forming ability and the
stable triplex-forming trimers with an L-nucleotide at
the 30-end are thought to be inactive for further elonga-
tion. Furthermore, the stable triplex-forming trimers
with an L-nucleotide at the 30-end would compete with
the trimers proper for the elongation on the template.
These characteristics of heterochiral oligomers may lead
to an extremely reduced efficiency of template-directed
oligomerization of racemic mononulceotides.22
4.2. Synthesis of trimers
4.2.1. 50-O-(4,40-Dimethoxytrityl)-N6-benzoyl-D-adeno-
0
sine 2. A solution of 1 (2.97 g, 8.0mmol) and 4,4 -
dimethoxytritylchloride (3.25 g, 9.6 mmol) in dry pyri-
dine (40mL) was stirred at room temperature for 1 h.
After the addition of EtOH (4 mL), the solvent was
evaporated under reduced pressure. The residue was
extracted with chloroform and washed with satd NaH-
CO3. The organic layer was dried over Na2SO4 and con-
centrated. The residue was chromatographed on a silica
gel (100 g) to give 2 (4.74 g, 87.9%).
4.2.2. N6,20,30-O-Tribenzoyl-D-adenosine 3. A solution
of 2 (5.34 g, 7.92 mmol), benzoic anhydride (5.34 g,
23.6 mmol), triethylamine (5.21 mL, 23.6 mmol) and 4-
dimethylaminopyridine (575 mg, 4.72 mmol) in dry
dichloromethane (47 mL) was stirred at room tempera-
ture for 1 h. After the addition of EtOH (4 mL), the
solution was diluted with chloroform and washed with
satd NaHCO3 and 0.5 M KH2PO4 aqueous solutions.
The organic layer was dried over Na2SO4 and concen-
trated. The residue was dissolved in chloroform
(140mL) and the solution cooled with an ice bath and
treated with benzenesulfonic acid solution in MeOH
(6.67%, 60mL). After stirring at 0 ꢁC for 15 min, the
reaction was quenched by adding satd NaHCO3 aque-
ous solution. The solution was diluted with chloroform
and washed with satd NaHCO3 aqueous solution. The
organic layer was dried over Na2SO4 and concentrated.
The residue was chromatographed on a silica gel (100 g)
to give 3 (3.96 g, 86.3%).
3. Conclusion
In conclusion, the results reported herein indicate that
the homochirality of the internal and 30-end residues is
the primary factor for maintaining the helical structure
of ApApA, whereas the configuration of the 30-end res-
idue has little effect on the triplex formation. Thus, the
chiral modifications result in the different influences on
the helical structure of ApApA and on its triplex-form-
ing ability with D-poly(U). These findings would provide
useful information for the application of L-RNA to the
development of novel functional molecules and for also
considering the process of the chiral selection during the
chemical evolution of RNA.
4. Experimental
4.1. General
4.2.3. 30,50-O-(1,1,3,3-Tetraisopropyldisiloxanylidene)-
N6-benzoyl-D-adenosine 4. A solution of 1 (0.37 g, 1.0
mmol) and 1,1,3,3-tetraisopropyl-1,3-dichlorodisiloxane
(0.38 mL, 1.20 mmol) in dry pyridine (40 mL) was
TLC analyses were carried out on Merk silica gel 60F254
plates, which were visualized by UV illumination at
254 nm and spraying of diluted aqueous sulfuric acid

H. Urata et al. / Tetrahedron: Asymmetry 16 (2005) 2908–2917
2915
stirred at room temperature overnight. After the addi-
tion of H2O (2 mL), the solvent was evaporated under
reduced pressure. The residue was extracted with chloro-
form and washed with satd NaHCO3 aqueous solution.
The organic layer was dried over Na2SO4 and concen-
trated. The residue was chromatographed on a silica
gel (20g) to give 4 (0.61 g, 99.3%).
4.2.11. 50-O-(4,40-Dimethoxytrityl)-20-O- tetrahydrofuran-
yl-N6-benzoyl-L-adenosine 13. Yield: 0.733 g, 95.7%.
4.2.12. 50-O-(4,40-Dimethoxytrityl)-20-O-tetrahydrofuran-
yl-N6-benzoyl-L-adenosine
30-O-(2-chlorophenyl)phos-
phate 14. Yield: 1.24 g, quant.
4.2.13. 50-Free dimer 15. A mixture of 3 (0.107 g,
0.185 mmol) and 7 (0.23 g, 0.222 mmol) was coevapo-
rated with dry pyridine and dissolved in dry pyridine
(1.5 mL). After 1-(mesitylene-2-sulfonyl)-3-nitro-1,2,4-
triazole (MSNT, 0.132 g, 0.444 mmol) was added, the
solution was kept at room temperature for 40min.
The mixture was diluted with chloroform and washed
with 50mM TEAB. The organic layer was evaporated
under reduced pressure and the residue chromato-
graphed on a silica gel (10g) to give fully protected
dimer (0.233 g, 84.2%).
4.2.4. 20-O-Tetrahydrofuranyl-N6-benzoyl-D-adenosine
5. A solution of 4 (1.01 g, 1.65 mmol), 2,3-dihydro-
furan (0.25 mL, 3.36 mmol) and pyridinium p-toluene-
sulfonate (0.084 g, 0.34 mmol) in dry dichloromethane
(10mL) was stirred at room temperature overnight.
After addition of satd NaHCO3 aqueous solution, the
solution was diluted with chloroform. The mixture was
washed with satd NaHCO3 aqueous solution. The
organic layer was dried over Na2SO4 and concentrated.
The residue was treated with 1 M tetrabutylammonium
fluoride in THF solution (3.3 mL) for 10min. The
solvent was evaporated under reduced pressure and
the residue chromatographed on a silica gel (30g) to
give 5 (0.70 g, 95.6%).
The fully protected dimer (0.134 g, 0.09 mmol) was
coevaporated with dry pyridine and dry toluene, and
treated with 1 M ZnBr2 solution in dichloromethane–
i-PrOH (85:15) (5 mL, 5 mmol) at room temperature
for 30min. The reaction was quenched by adding 1 M
aqueous ammonium acetate and diluted with chloro-
form. The organic layer was separated and washed with
1 M aqueous ammonium acetate. The organic layer was
dried over Na2SO4 and evaporated under reduced pres-
sure. The residue was chromatographed on a silica gel
(10g) to give 15 (0.104 g, 97.2%).
4.2.5. 50-O-(4,40-Dimethoxytrityl)-20-O-tetrahydrofuran-
yl-N6-benzoyl-D-adenosine 6. A solution of 5 (0.21 g,
0.48 mmol) and 4,40-dimethoxytritylchloride (0.195 g,
0.58 mmol) in dry pyridine (3 mL) was stirred at room
temperature for 2 h. After the addition of EtOH
(0.5 mL), the solvent was evaporated under reduced
pressure. The residue was extracted with chloroform
and washed with satd NaHCO3 aqueous solution. The
organic layer was dried over Na2SO4 and was concen-
trated. The residue was chromatographed on a silica
gel (10g) to give 6 (0.33 g, 92.3%).
4.2.14. 50-Free dimer 16. Yield: 0.195 g, 90.4%.
4.2.15. 50-Free dimer 17. Yield: 0.113 g, 86.0%.
4.2.16. 50-Free dimer 18. Yield: 0.257 g, 90.5%.
4.2.6. 50-O-(4,40-Dimethoxytrityl)-20-O-tetrahydrofura-
nyl-N6-benzoyl-D-adenosine 30-O-(2-chlorophenyl)phos-
4.2.17. Fully protected trimer 19. A mixture of 15
(0.11 g, 0.088 mmol) and 7 (0.138 g, 0.133 mmol) was
coevaporated with dry pyridine and dissolved in dry pyr-
idine (1.5 mL). After MSNT (0.079 g, 0.266 mmol) was
added, the solution was kept at room temperature for
40min. The mixture was diluted with chloroform and
washed with 50mM TEAB. The organic layer was evap-
orated under reduced pressure and the residue chromato-
graphed on a silica gel (10g) to give 19 (0.151 g, 80.6%).
phate 7. Compound
6
(0.10 g, 0.13 mmol) was
coevaporated with dry pyridine three times, and treated
with 0.33 M 2-chlorophenylphosphoroditriazolide solu-
tion in dioxane (0.83 mL, 0.27 mmol). After 2 h,
50mM triethylammonium bicarbonate buffer (TEAB,
pH 7.5, 2 mL) was added, and the mixture extracted
with chloroform–pyridine (3:1) and washed with
50mM TEAB. The organic layer was evaporated under
reduced pressure and the residue was dissolved in small
amount of chloroform. The solution was added drop-
wise to 0.1% triethylamine in n-hexane (50mL) to give
7 (0.158 g, quant.).
4.2.18. Fully protected trimer 20. Yield: 0.205 g, 64.7%.
4.2.19. Fully protected trimer 21. Yield: 0.190 g, 77.8%.
4.2.20. Fully protected trimer 22. Yield: 0.176 g, 73.1%.
4.2.21. Fully protected trimer 23. Yield: 0.317 g, 77.0%.
4.2.22. Fully protected trimer 24. Yield: 0.259 g, 86.2%.
4.2.23. Fully protected trimer 25. Yield: 0.199 g, 83.8%.
4.2.24. Fully protected trimer 26. Yield: 0.233 g, 92.2%.
4.2.7. 50-O-(4,40-Dimethoxytrityl)-N6-benzoyl-L-adeno-
sine 9. Yield: 0.29 g, 85.0%.
4.2.8. N6,20,30-O-Tribenzoyl-L-adenosine 10. Yield:
0.23 g, 92.3%.
4.2.9. 30,50-O-(1,1,3,3-Tetraisopropyldisiloxanylidene)-
N6-benzoyl-L-adenosine 11. Yield: 0.84 g, 91.2%.
4.2.10. 20-O-Tetrahydrofuranyl-N6-benzoyl-L-adenosine
12. Yield: 0.455 g, 75.3%.
4.2.25. Deprotection and purification of trimers. The tri-
mer (10–20 lmol) was suspended in pyridine (2 mL) and

2916
H. Urata et al. / Tetrahedron: Asymmetry 16 (2005) 2908–2917
concentrated ammonium hydroxide (10mL). The reac-
tion vessel was sealed and heated at 55 ꢁC for 8 h. After
cooling, the solvents were evaporated under reduced
pressure and the residue treated with 80% aqueous ace-
tic acid for 2 h. The mixture was concentrated and
coevaporated with H2O several times, and the residue
dissolved in 20mM TEAB and washed with diethyl
ether. After the aqueous layer was lyophilized, the resi-
HCl (pH 7.5, 4 mL). The solution was placed in a
1 cm path-length quartz cell. Spectra were measured
on a JASCO J-820spectropolarimeter equipped with a
temperature control unit.
4.6. UV mixing curves
Each trimer and D-poly(U) was mixed with various
molar ratios of the bases at the total base concentration
of 150 lM. The samples were dissolved in 10mM MgCl 2
and 10mM Tris–HCl (pH 7.5, 3 mL), and the absor-
bance at 260nm of the solutions was measured at 0 ꢁC.
due was purified on
a C18 silica gel column
(7 · 150mm) with a linear gradient of acetonitrile (0–
15%) in 50mM triethylammonium acetate (TEAA,
pH 7.0). Fractions containing the pure timer were col-
lected and evaporated under reduced pressure, and the
residue was then lyophilized several times to remove
TEAA. The purity of the trimers tested by the above
HPLC system was more than 95%.
4.7. Measurements of melting curves
Each trimer and D-poly(U) was mixed at a total base
concentration of 90 lM with the molar ratio
A:U = 1:2, and dissolved in a buffer containing 10mM
MgCl2, 10mM Tris–HCl (pH 7.5, 4 mL). The solutions
containing triplexes were heated at 80 ꢁC and cooled
gradually to room temperature. Melting curves were
obtained at least twice by measuring the absorbance at
260nm every 0.2 ꢁC on a JASCO Ubest-55 spectropho-
tometer. The temperature was raised at a rate of 0.5 ꢁC/
min, and the Tm values were obtained by the first-deriv-
ative plots of the melting curves.
4.3. Enzymatic digestion
For nuclease P1 digestion, solutions of the trimers
(1 OD unit) in 50mM ammonium acetate (pH 5.0,
18 lL) were treated with nuclease P1 (1 mg/mL, 2 lL).
For SVPD digestion, solutions of the trimers (1 OD
unit) in 10mM MgCl 2 and 50mM Tris–HCl (pH 8.0,
18 lL) were treated with SVPD (0.5 mg/mL, 2 lL).
After the reactions were incubated at 37 ꢁC for 3 h,
30mM of EDTA (4 lL) was added. The mixtures were
filtered through Ultrafree-MC 5000NMWL (Millipore)
and analyzed by a reversed phase HPLC. Molar extinc-
tion coefficients of the trimers were determined by com-
parison of the absorbances of the trimer solutions before
and after the complete enzymatic digestion into adeno-
sine and 50-AMP. The coefficients of the trimers (AL-
pALpAD, ADpALpAL and ALpALpAL) that could not
be cleaved completely by any of nuclease P1, SVPD or
their combination were assumed as the same as those
of the enantiomers.
References
1. Urata, H.; Shinohara, K.; Ogura, E.; Ueda, Y.; Akagi, M.
J. Am. Chem. Soc. 1991, 113, 8174–8175.
2. Urata, H.; Ogura, E.; Shinohara, K.; Ueda, Y.; Akagi, M.
Nucleic Acids Res. 1992, 20, 3325–3332.
3. Ashley, G. W. J. Am. Chem. Soc. 1992, 114, 9731–9736.
4. Nolte, A.; Klußmann, S.; Bald, R.; Erdmann, V. A.;
Furste, J. P. Nat. Biotechnol. 1996, 14, 1116–1119.
¨
5. Pitsch, S. Helv. Chim. Acta 1997, 80, 2286–2314.
6. Garbesi, A.; Hamy, F.; Maffini, M.; Albrecht, G.; Klim-
kait, T. Nucleic Acids Res. 1998, 26, 2886–2890.
7. Garbesi, A.; Capobianco, M. L.; Colonna, F. P.; Maffini,
M.; Niccolai, D.; Tondelli, L. Nucleos. Nucleot. 1998, 17,
1275–1287.
8. Moyroud, E.; Biala, E.; Strazewski, P. Tetrahedron 2000,
56, 1475–1484.
9. Damha, M. J.; Giannaris, P. A.; Marfey, P.; Reid, L. S.
Tetrahedron Lett. 1991, 32, 2573–2576.
10. Urata, H.; Ueda, Y.; Suhara, H.; Nishioka, E.; Akagi, M.
J. Am. Chem. Soc. 1993, 115, 9852–9853.
11. Hashimoto, Y.; Iwanami, N.; Fujimori, S.; Shudo, K.
J. Am. Chem. Soc. 1993, 115, 9883–9887.
12. Blommers, M. J.; Tondelli, L.; Garbesi, A. Biochemistry
1994, 33, 7886–7896.
4.4. Degradation of trimers into adenosine and its optical
resolution
The trimers (1 OD unit) were treated with 0.1 M potas-
sium hydroxide (200 lL) at 80 ꢁC for 2.5 h. After the pH
was adjusted to about 9.0with 1 M HCl, a solution of
10mM MgCl 2, 500 mM Tris–HCl (pH 9.0, 20 lL) and
CIAP (0.5 units/lL, 2 lL) was added. After 3 h at
37 ꢁC, the mixtures were desalted with a Sep-Pak Plus
C18 cartridge (Waters), and analyzed by chiral HPLC.
The chiral HPLC analyses were performed on a column
of SUMICHIRAL OA-6000 (4.6 · 150mm, Sumika
Chemical Analysis Service, Japan) with an isocratic elu-
tion of 8% acetonitrile in 2 mM CuSO4.
13. Damha, M. J.; Giannaris, P. A.; Marfey, P. Biochemistry
1994, 33, 7877–7885.
14. Urata, H.; Akagi, M. Tetrahedron Lett. 1996, 37, 5551–
5554.
4.5. Measurements of CD spectra
15. Vichier-Guerre, S.; Morvan, F.; Fulcrand, G.; Rayner, B.
Tetrahedron Lett. 1997, 38, 93–96.
16. Vichier-Guerre, S.; Santamaria, F.; Rayner, B. Tetrahe-
dron Lett. 2000, 41, 2101–2104.
17. Urata, H.; Go, M.; Ohmoto, N.; Minoura, K.; Akagi, M.
Chem. Commun. 2002, 544–545.
18. Urata, H.; Shimizu, H.; Hiroaki, H.; Kohda, D.; Akagi,
M. Biochem. Biophys. Res. Commun. 2003, 309, 79–
83.
For measurements of CD spectra of the trimers, a solu-
tion (30 lM) of each trimer containing 0.1 M NaCl and
10mM sodium phosphate (pH 7.0) was placed in a 1 cm
path-length quartz cell. For the measurements of the
spectra of the ApApAÆD-poly(U) triplexes, each trimer
and poly(U) was mixed at a total base concentration
of 90 lM with the molar ratio A:U = 1:2, and dissolved
in a buffer containing 10mM MgCl 2 and 10mM Tris–

H. Urata et al. / Tetrahedron: Asymmetry 16 (2005) 2908–2917
2917
19. Tazawa, I.; Tazawa, S.; Stempel, L. M.; Tsꢀo, P. O. P.
Biochemistry 1970, 9, 3499–3514.
20. Ohtsuka, E.; Ohkubo, M.; Yamane, A.; Ikehara, M.
Chem. Pharm. Bull. 1983, 31, 1910–1916.
21. Johnson, W. C., Jr. Determination of the conformation
of nucleic acids by electronic CD. In Circular Dichroism
and the Conformational Analysis of Biomolecules; Fasman,
G. D., Ed.; Plenum Press: New York, 1996, pp 433–
468.
22. Joyce, G. F.; Visser, G. M.; van Boeckel, C. A. A.; van
Boom, J. H.; Orgel, L. E.; van Westrenen, J. Nature 1984,
310, 602–604.
23. Akagi, M.; Omae, D.; Tamura, Y.; Ueda, T.; Kumashiro,
T.; Urata, H. Chem. Pharm. Bull. 2002, 50, 866–868.
24. Recondo, E. F.; Rinderknecht, H. Helv. Chim. Acta 1959,
42, 1171–1173.
25. Vorbruggen, H.; Krolikiewicz, K. Angew. Chem., Int. Ed.
¨
1975, 14, 421–422.