Synthesis and silencing properties of siRNAs possessing lipophilic groups at their 3′-termini

Article abstract of DOI:10.1016/j.bmc.2008.07.010

Short-interfering RNAs (siRNAs) conjugated with lipophilic groups at their 3′-termini were synthesized. The properties of the synthesized siRNAs were examined in detail, and it was found that at low concentrations, their silencing abilities were dependent

Full text of DOI:10.1016/j.bmc.2008.07.010

Bioorganic & Medicinal Chemistry 16 (2008) 7698–7704
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Synthesis and silencing properties of siRNAs possessing lipophilic groups
at their 30-termini
a
e
a
a
e
Yoshihito Ueno a,b,d,f, , Koshi Kawada , Tomoharu Naito , Aya Shibata , Kayo Yoshikawa , Hye-Sook Kim ,
*
Yusuke Wataya e, Yukio Kitade a,b,c,d,
*
a Department of Biomolecular Science, Faculty of Engineering, Gifu University, Medical Information Sciences, Gifu University, 1-1 Yanagido, Gifu 501-1193, Japan
b Center for Emerging Infectious Diseases, Gifu University, Medical Information Sciences, Gifu University, 1-1 Yanagido, Gifu 501-1193, Japan
c Center for Advanced Drug Research, Gifu University, Medical Information Sciences, Gifu University, 1-1 Yanagido, Gifu 501-1193, Japan
d United Graduate School of Drug Discovery and Medical Information Sciences, Gifu University, 1-1 Yanagido, Gifu 501-1193, Japan
e Faculty of Pharmaceutical Sciences, Okayama University, Tsushima, Okayama 700-8530, Japan
f PRESTO, JST (Japan Science and Technology Agency), 4-1-8 Honcho Kawaguchi, Saitama 332-0012, Japan
a r t i c l e i n f o
a b s t r a c t
Short-interfering RNAs (siRNAs) conjugated with lipophilic groups at their 30-termini were synthesized.
The properties of the synthesized siRNAs were examined in detail, and it was found that at low concen-
trations, their silencing abilities were dependent on the positions of the modifications and the types of
organic molecules attached. Although the modification of siRNAs with palmitic acid or oleic acid at the
30-end slightly reduced their silencing activities, siRNAs had enough abilities to induce RNAi at 10 nM
concentrations. On the other hand, the modification of siRNAs with cholesterol at the 30-end of the pas-
senger strand was tolerated; however, the modification at the guide strand significantly reduces its
silencing activity. The siRNAs modified with the lipophilic groups did not possess ability to penetrate
the plasma membranes of HT-1080 cells without the transfection reagent. However, the results described
in this report will aid in designing novel siRNAs with cell membrane-permeable molecules.
Ó 2008 Elsevier Ltd. All rights reserved.
Article history:
Received 15 May 2008
Revised 2 July 2008
Accepted 3 July 2008
Available online 9 July 2008
Keywords:
RNAi
siRNA
Palmitic acid
Oleic acid
Cholesterol
1. Introduction
Since the discovery of RNA interference (RNAi) as a means to
silence expression of specific genes, small-interfering RNAs (siR-
NAs) have attracted significant attention as powerful tools for
targeting therapeutically important mRNAs and eliciting their
cleavage.1,2 siRNA has considerable potential as a new therapeu-
tic drug for intractable diseases because siRNAs can be logically
designed and synthesized if the sequences of disease-causing
genes are known. One of the critical problems in RNAi-based
therapeutic applications is the delivery of siRNA across plasma
membranes of cells in vivo. Thus far, a number of solutions
to this problem have been reported.3–9 Among them, the direct
chemical modification of siRNAs is an attractive proposition be-
cause various kinds of organic molecules, which are specifically
uptaken in specific tissues, can be specifically introduced into
siRNAs. Recently, Soutschek et al. have succeeded in reducing
plasma low-density lipoprotein (LDL) levels in mice by using
siRNAs that are modified with cholesterol at the 30-end of the
guide strand (antisense strand).4
Figure 1. Structures of siRNAs possessing lipophilic groups.
It is deduced that the ability of siRNAs to induce RNAi depends
on the positions of modifications and kinds of organic molecules
attached. However, the silencing activities of siRNAs modified with
* Corresponding authors. Tel.: +81 58 293 2639; fax: +81 58 230 1893.
E-mail addresses: uenoy@gifu-u.ac.jp (Y. Ueno), ykkitade@gifu-u.ac.jp (Y. Kitade).
0968-0896/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2008.07.010

Y. Ueno et al. / Bioorg. Med. Chem. 16 (2008) 7698–7704
7699
cell membrane-permeable molecules have not been investigated
systematically. In this paper, we report the synthesis and silencing
properties of siRNAs possessing lipophilic groups at their 30-ter-
mini (Fig. 1).
Argonaute2, a key component of RNA-induced silencing com-
plex (RISC), is responsible for mRNA cleavage in the RNAi path-
way.10 Argonaute2 is composed of PAZ, Mid, and PIWI domains.
The X-ray structural analysis of a co-crystal between a PIWI pro-
tein and an siRNA reveals that the 50-phosphate of a guide strand
(antisense strand) of siRNA is recognized by the PIWI domain
and that the 50-phosphate interacts with the carboxyl group of a
more flexible than hydroxyprolinol, as a linker to join RNAs and
lipophilic groups. Glycerol derivative 120 with 4,40-dimethoxytrityl
(DMTr) and tert-butyldimethylsilyl (TBDMS) groups at primary
hydroxyls was synthesized according to the reported method, con-
verted into its O-carbonylimidazolide, and reacted with 1,4-diam-
inobutane to afford an aminobutylcarbamoyl derivative 2 in 94%
yield (Scheme 1). Palmitic acid and oleic acid were introduced into
2 in the presence of 1-ethyl-3-(3-dimethylaminopropyl)carbodi-
imide hydrochloride (WSCI) to yield glycerol derivatives 3 and 4
with fatty acids in 69% and 86% yields, respectively. On the other
hand, cholesterol was converted into its O-carbonylimidazolide,
which was then reacted with 2 to afford a cholesterol-modified
glycerol 5 in 64% yield. After the desilylation of 3, 4, and 5 by tet-
rabutylammonium fluoride (TBAF), hydroxyl derivatives 6, 7, and 8
were succinated to give the corresponding succinates. They were
linked to a controlled pore glass (CPG) to produce solid supports
C-terminal residue of the domain through
a divalent metal
ion.11,12 It is observed that the modification of the 50-hydroxyl
groups of the guide strands of siRNAs by methoxy groups com-
pletely hampers the functions of siRNAs to induce RNAi in experi-
ments using Drosophila embryo lysate and HeLa cell extract.13,14
These results imply that the 50-terminus of a guide strand is not
suitable for modification by lipophilic groups.
9, 10, and 11 containing 6 (28
(88 mol/g), respectively.
lmol/g), 7 (83 lmol/g), and 8
l
On the other hand, X-ray structural analysis and a nuclear mag-
netic resonance (NMR) study have revealed that the 30-overhang
region of a guide strand of siRNA is recognized by the PAZ domain
and the 2-nucleotide (nt) 30-overhang is accommodated into a
binding pocket composed of hydrophobic amino acids in the do-
main.15–18 The length of the 30-overhang regions of siRNA influ-
ences the activities of siRNAs. It is reported that the 2-nt 30-
overhang is the most efficient in an experiment using 21-nt siRNA
in Drosophila embryo lysate; however, the multiple addition of 20-
deoxynucleotide to the 30-end of siRNAs is tolerated.19 On the basis
of this information, we introduced lipophilic groups at the 30-ends
of the passenger or guide strands of siRNAs.
All oligoribonucleotides (ONs) were synthesized using solid
supports 9, 10, and 11 by using a DNA/RNA synthesizer (Table 1).
Fully protected ONs (1.0 lmol each) linked to solid supports were
treated with concentrated NH4OH/EtOH (3:1, v/v) at room temper-
ature for 12 h and then with 1.0 M TBAF/THF at room temperature
for 12 h. Released ONs were purified by denaturing 20% polyacryl-
amide gel electrophoresis (20% PAGE) to afford deprotected ONs
25–44. These ONs were analyzed by matrix-assisted laser desorp-
tion/ionization time-of-flight mass spectrometry (MALDI-TOF/MS),
and observed molecular weights were in agreement with their
structures.
The ability of modified siRNAs to suppress gene expression was
studied by a dual-luciferase assay using a psiCHECK-2 vector,
which contained Renilla and firefly luciferase genes. Sequences of
siRNAs were designed to target Renilla luciferase. HeLa cells were
co-transfected with the vector and indicated amounts of siRNAs.
Signals of Renilla luciferase were normalized to that of firefly
2. Results and discussion
Soutschek et al. used 4-hydroxyprolinol as a linker for conjugat-
ing RNAs and lipophilic groups. We selected glycerol, which was
Scheme 1. Reagents and conditions: (a) (1) N,N0-carbonyldiimidazole, pyridine, rt, 1 h; (2) 1,4-diaminobutane, pyridine, rt, 3 h, 94%; (b) palmitic acid or oleic acid, WSCI,
CH2Cl2, rt, 24 h, 69% for 3 and 64% for 4; (c) cholesterol, N,N0-carbonyldiimidazole, pyridine, rt, 24 h, 64%; (d) TBAF, THF, rt, 80% for 6, 82% for 7, and 53% for 9; (e) (1) succinic
anhydride, DMAP, pyridine, rt, 24 h; (2) CPG, WSCI, DMF, rt, 72 h, 28 lmol/g for 9, 83 lmol/g for 10, and 88 lmol/g for 11.

7700
Y. Ueno et al. / Bioorg. Med. Chem. 16 (2008) 7698–7704
tached at the 30-termini of siRNAs (Fig. 2a and b). Although the
Table 1
Sequences of ONs and siRNAs used in this study
modifications of passenger strands (sense strands) by palmitic acid
and oleic acid reduced the silencing activities of siRNAs by a small
amount as compared to that of unmodified siRNA, siRNAs modified
with these fatty acids maintained their abilities to down-regulate
protein expression levels to 30% at 10 nM concentrations. The
silencing activity of siRNA modified with cholesterol at the 30-
end of the passenger strand was weaker than those of siRNAs mod-
ified with palmitic acid and oleic acid at each concentration. How-
ever, it down-regulated protein expression level to approximately
50% at 10 nM concentrations. The tendency of the silencing activi-
ties of siRNAs modified with palmitic acid and oleic acid at the 30-
ends of guide strands was similar to that of siRNAs modified at the
30-ends of guide strands. On the other hand, the modification of
siRNA with cholesterol at the 30-end of the guide strand signifi-
cantly reduced its silencing activity.
Argonaute2/eIF2C2 (hAgo2) has been identified as a key protein
with an endonuclease activity associated with RISC in the RNAi
pathway.21,22 In order to examine whether or not the observed
silencing activities could be attributed to RNAi, the activities of
modified siRNAs were studied after treating HeLa cells with
eIF2C2-targeting siRNAs. We considered that if the silencing activ-
ities of modified siRNAs resulted from RNAi, the expression levels
of luciferase proteins would recover by treating HeLa cells with
siRNAs targeting eIF2C2. Two kinds of siRNAs targeting eIF2C2—
one (siRNA20) targeted open reading frame (ORP) positions
1168–1188 and the other (siRNA21) targeted ORP positions
1897–1917—were used in this study. The siRNA19 is composed
of a random sequence. HeLa cells were transfected with the siR-
NA19, 20 or 21. After incubating them for 1 h, the cells were co-
transfected with a psiCHECK-2 vector and siRNAs modified with
lipophilic groups. After incubating for 24 h, the activities of Renilla
luciferase were measured. The siRNA19 composed of a random se-
quence did not largely change the expression level of Renilla lucif-
erase (Fig. 3) whereas the signals of Renilla luciferase were
recovered by treating them with siRNAs targeting eIF2C2 (Fig. 4).
These results indicated that the silencing activities of modified siR-
NAs were attributed to RNAi.
No. of siRNA
No. of ON
Sequence
siRNA12
ON25
ON26
50-GGCCUUUCACUACUCCUACtt-30
30-ttCCGGAAAGUGAUGAGGAUG-50
siRNA13
siRNA14
siRNA15
siRNA16
siRNA17
siRNA18
siRNA19
siRNA20
siRNA21
siRNA22
siRNA23
siRNA24
ON27
ON26
50-GGCCUUUCACUACUCCUACtt-30-Pal
30-ttCCGGAAAGUGAUGAGGAUG-50
ON28
ON26
50-GGCCUUUCACUACUCCUACtt-30-Ole
30-ttCCGGAAAGUGAUGAGGAUG-50
ON29
ON26
50-GGCCUUUCACUACUCCUACtt-30-Chl
30-ttCCGGAAAGUGAUGAGGAUG-50
ON25
ON30
50-GGCCUUUCACUACUCCUACtt-30
Pal-30-ttCCGGAAAGUGAUGAGGAUG-50
ON25
ON31
50-GGCCUUUCACUACUCCUACtt-30
Ole-30-ttCCGGAAAGUGAUGAGGAUG-50
ON25
ON32
50-GGCCUUUCACUACUCCUACtt-30
Chl-30-ttCCGGAAAGUGAUGAGGAUG-50
ON33
ON34
50-UUCUCCGAACGUGUGCACGUtt-30
30-ttAAGAGGCUUGCACAGUGCA-50
ON35
ON36
50-AGAUCCAUACGUCCGUGAAtt-30
30-ttUCUAGGUAUGCAGGCACUU-50
ON37
ON38
50-GCACCGGCAGGAGAUCAUAtt-30
30-gtCGUGGCCGUCCUCUAGUAU-50
ON39
ON40
50-GCUGUUCAAAACGAAGAUGtt-30-Pal
Pal-30-ttCGACAAGUUUUGCUUCUAC-50
ON41
ON42
50-GCUGUUCAAAACGAAGAUGtt-30-Ole
Ole-30-ttCGACAAGUUUUGCUUCUAC-50
ON43
ON44
50-GCUGUUCAAAACGAAGAUGtt-30-Chl
Chl-30-ttCGACAAGUUUUGCUUCUAC-50
The capital letters indicate ribonucleosides. Small italic letters represent 20-
deoxyribonucleosides.
luciferase. The silencing activities of siRNAs correlated with the
positions of the modifications and kinds of lipophilic groups at-
In order to examine whether or not the siRNAs modified with
the lipophilic groups can cross plasma membranes of cells without
a transfection reagent, next, we performed an experiment target-
ing an endogenous protein. We chose an mRNA of a human RNase
L protein, which is a constituent of the 2-5A system, as a target. It
Figure 3. Dual-luciferase assay. HeLa cells were transfected with siRNA19 (50 nM),
siRNA20 (50 nM), or siRNA21 (50 nM). After incubating for 1 h, the cells were co-
transfected with a psiCHECK-2 vector and siRNA12 (10 nM). After incubating for
24 h, the activities of Renilla luciferase were measured.
Figure 2. Dual-luciferase assay. (a) siRNAs modified at 30-ends of passenger (sense)
strands. (b) siRNAs modified at 30-ends of guide (antisense) strands.

Y. Ueno et al. / Bioorg. Med. Chem. 16 (2008) 7698–7704
7701
modified siRNAs had the abilities to down-regulate the expression
of the target protein. On the other hand, when the cells were trea-
ted with the abovementioned siRNAs without the transfection re-
agent, the expression levels of the human RNase L protein hardly
changed at both 24 and 48 h post-transfections. From these results,
it was revealed that the siRNAs modified with palmitic acid, oleic
acid, and cholesterol did not have abilities to penetrate the plasma
membranes of HT-1080 cells without the transfection reagent.
In conclusion, we have demonstrated the synthesis of siRNAs
possessing lipophilic groups at their 30-termini. It was found that
all the modified siRNAs had similar silencing abilities at high siRNA
concentrations but the silencing abilities of the modified siRNAs at
low siRNA concentrations were dependent on the kind and posi-
tions of the modifications. Although the modifications of siRNAs
with palmitic acid or oleic acid at their 30-end reduced their silenc-
ing activities by a small amount, siRNAs had enough abilities to in-
duce RNAi at their 10 nM concentrations. On the other hand, the
modification of siRNA with cholesterol at the 30-end of the passen-
ger strand was tolerated; however, the modification at the guide
strand significantly reduces its silencing activity. From these re-
sults, it is concluded that the 30-end of the passenger strand is bet-
ter than the guide strand for conjugating bulky molecules such as
fatty acids, particularly rigid molecules such as cholesterol. The
siRNAs modified with the lipophilic groups did not have abilities
to penetrate the plasma membranes of HT-1080 cells without the
transfection reagent. However, we believe that the results de-
scribed here will aid in designing novel siRNAs with cell-mem-
brane-permeable molecules.
Figure 4. Dual-luciferase assay. HeLa cells were transfected with siRNA20 (50 nM)
or siRNA21 (50 nM). After incubating for 1 h, the cells were co-transfected with a
psiCHECK-2 vector and siRNAs (10 nM) modified with lipophilic groups. After
incubating for 24 h, the activities of Renilla luciferase were measured.
3. Experimental
3.1. General remarks
NMR spectra were recorded at 400 MHz (1H) and at 100 MHz
(
13C), and are reported in parts per million downfield from tetra-
methylsilane. Coupling constants (J) are expressed in Hertz. Mass
spectra were obtained by fast atom bombardment (FAB). Thin-
layer chromatography was carried out on Merck-coated plates 60
Figure 5. (a) Western blot analysis for human RNase L and GAPDH. Lanes 5–7 and
12–14, without Lipofectamine 2000; lanes 8–10 and 15–17, with Lipofectamine
2000. Lane 1, marker; lane 2, 3.5 ng RNase L; lane 3, 1.75 ng RNase L; lanes 4 and 11,
non-treatment; lanes 5, 8, 12, and 15, siRNA 22; lanes 6, 9, 13, and 16, siRNA23;
lanes 5, 8, 12, and 15, siRNA22; lanes 7, 10, 14, and 17, siRNA24; lane 18,
cells + 1.75 ng RNase L. (b) Relative intensities of RNase L protein levels standard-
ized by GAPDH. LF indicates Lipofectamine 2000.
F254. Silica gel column chromatography was carried out on Wako
gel C-300. siRNAs directed against eIF2C2 (hAgo2) were purchased
from Qiagen Inc.
3.1.1.1-O-(4,40-Dimethoxytrityl)-2-O-[N-(4-aminobutyl)carbam-
oyl]-3-O-tert-butyldimethylsilylglycerol (2)
A mixture of 1-O-(4,40-dimethoxytrityl)-3-O-tert-butyldimeth-
ylsilylglycerol (1) (2.60 g, 5.11 mmol) and N,N0-carbonyldiimidaz-
ole (1.66 g, 10.2 mmol) in pyridine (17 mL) was stirred at room
temperature. After 1 h, 1,4-diaminobutane (2.63 g, 29.8 mmol)
was added to the mixture at 0 °C, and the whole was stirred at
room temperature. After 3 h, the mixture was diluted with CHCl3.
The organic layer was washed with H2O and brine, dried (Na2SO4),
and concentrated. The residue was purified by column chromatog-
raphy (SiO2, 10–25% MeOH in CHCl3) to give 2 (3.00 g, 4.82 mmol)
in 94% yield: 1H NMR (DMSO-d6) d 0.13 (s, 6H), 0.97 (s, 9H), 3.94 (s,
6H), 6.96–7.61 (m, 13H), all additional signals correspond to glyc-
erol and aminobutyl moieties; 13C NMR (DMSO-d6) d ꢁ5.6, 17.7,
25.6, 26.8, 27.8, 48.5, 55.0, 61.7, 62.3, 72.5, 85.2, 113.1, 126.6,
127.6, 127.7, 129.6, 135.5, 144.9, 149.3, 155.8, 158.0; HRMS
(FAB) calcd for C35H51N2O6Si (MH+) 623.3516, found 623.3508.
was reported that the absence of RNase L in vivo causes both a defi-
ciency in the antiviral activity of interferon and a major defect in
apoptosis23,24. We selected 23-nt of the open reading frame
(ORP) position 94–112, that starts with 50-AA-30 and ends with
50-TT-30, as a target sequence. Sequences of the siRNAs are listed
in Table 1. The siRNAs 22, 23, and 24 have palmitic acid, oleic acid,
and cholesterol at the 30-ends both of the passenger and guide
strands, respectively.
HT-1080 human fibrosarcoma (HT-1080) cells were transfected
with the siRNAs (200 nM concentrations) with and without the
transfection reagent, and the cell lysates were analyzed 24 and
48 h later by immunoblot analysis with RNase L-specific antibody.
Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) protein le-
vel served as a loading control. The results of the immunoblot anal-
ysis are represented in Figure 5a. Figure 5b shows the relative
intensities of RNase L protein levels standardized by GAPDH.
When the cells were transfected with the siRNAs 22, 23, and 24
by using the transfection reagent, the human RNase L protein
expression levels were reduced to ꢀ30% and ꢀ20% at 24 and
48 h post-transfections, respectively. Thus, it was found that the
3.1.2. 1-O-(4,40-Dimethoxytrityl)-2-O-[N-palmitoyl-N-(4-aminobutyl)-
carbamoyl]-3-O-tert-butyldimethylsilylglycerol (3)
A mixture of 2 (0.90 g, 1.44 mmol), palmitic acid (0.55 g,
2.16 mmol), and 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide
hydrochloride (WSCI) in CH2Cl2 (14 mL) was stirred at room

7702
Y. Ueno et al. / Bioorg. Med. Chem. 16 (2008) 7698–7704
temperature. After 24 h, the mixture was diluted with CHCl3. The
organic layer was washed with H2O and brine, dried (Na2SO4),
and concentrated. The residue was purified by column chromatog-
raphy (SiO2, 10–50% EtOAc in hexane) to give 3 (0.85 g, 0.99 mmol)
in 69% yield: 1H NMR (CDCl3) d 0.01 (s, 6H), 0.82 (s, 9H), 3.80 (s,
6H), 6.81–7.46 (m, 13H), all additional signals correspond to glyc-
erol, palmitoyl, and aminobutyl moieties; 13C NMR (CDCl3) d ꢁ5.4,
14.1, 18.1, 22.7, 25.7, 26.6, 27.5, 29.3, 29.4, 29.5, 29.6, 29.7, 31.9,
36.7, 39.0, 40.3, 55.1, 60.4, 62.1, 62.4, 74.0, 85.7, 113.0, 126.6,
127.7, 128.1, 130.0, 130.1, 136.0, 136.1, 144.9, 156.2, 158.3, 173.3.
After 12 h, the mixture was diluted with EtOAc. The organic
layer was washed with aqueous NaHCO3 (saturated) and brine,
dried (Na2SO4), and concentrated. The residue was purified by
column chromatography (SiO2, 10% EtOAc in hexane) to give 5
(0.39 g, 0.37 mmol) in 64% yield: 1H NMR (CDCl3) d 0.10 (s,
6H), 0.81 (s, 9H), 3.79 (s, 6H), 6.80–7.44 (m, 13H), all additional
signals correspond to glycerol, cholesterol, and aminobutyl moi-
eties; 13C NMR (CDCl3) d ꢁ5.4, ꢁ3.6, 11.8, 18.0, 18.7, 19.3, 21.0,
22.5, 22.8, 23.8, 24.3, 25.6, 25.8, 27.3, 28.0, 28.1, 28.2, 31.8,
31.9, 35.8, 36.2, 36.5, 37.0, 38.5, 39.5, 39.7, 40.6, 42.3, 50.0,
55.2, 56.1, 56.7, 62.1, 85.8, 113.0, 122.4, 126.6, 127.7, 128.1,
130.0, 136.1, 139.8, 144.9, 156.1, 158.3; Anal. Calcd for
3.1.3. 1-O-(4,40-Dimethoxytrityl)-2-O-[N-palmitoyl-N-(4-amino-
butyl)carbamoyl]glycerol (6)
C
63H94N2O8Siꢂ2H2O: C, 70.62; H, 9.22; N, 2.61. Found: C, 70.72;
A mixture of 3 (0.75 g, 0.87 mmol) and TBAF (1 M in THF,
2.00 mL, 2.00 mmol) in THF (4.4 mL) was stirred at room tempera-
ture. After 1 h, the mixture was diluted with EtOAc. The organic
layer was washed with H2O and brine, dried (Na2SO4), and concen-
trated. The residue was purified by column chromatography (SiO2,
EtOAc) to give 6 (0.52 g, 0.70 mmol) in 80% yield: 1H NMR (CDCl3) d
3.81 (m, 6H), 6.81–7.43 (m, 13H), all additional signals correspond
to glycerol, palmitoyl and aminobutyl moieties; 13C NMR (CDCl3) d
13.8, 14.1, 20.5, 22.7, 25.7, 26.7, 27.2, 29.3, 29.5, 29.6, 29.7, 31.9,
36.8, 39.0, 40.6, 52.1, 55.2, 62.9, 63.2, 74.6, 86.1, 113.1, 126.8,
127.8, 128.0, 130.0, 135.7, 144.6, 156.6, 158.4, 173.4; HRMS
(FAB) calcd for C45H67N2O7 (MH+) 747.4949, found 747.4955.
H, 9.45; N, 2.43.
3.1.7. 1-O-(4,40-Dimethoxytrityl)-2-O-[N-cholesteryloxycarbonyl-
N-(4-aminobutyl)carbamoyl]glycerol (8)
A mixture of 5 (0.34 g, 0.33 mmol) and TBAF (1 M in THF,
1.00 mL, 1.00 mmol) in THF (5 mL) was stirred at room tempera-
ture. After 4 h, the mixture was concentrated in vacuo. The residue
was purified by column chromatography (SiO2, 20–100% EtOAc in
hexane) to give 8 (0.61 g, 0.18 mmol) in 53% yield: 1H NMR (CDCl3)
d 3.73 (s, 6 H), 6.75–7.37 (m, 13H), all additional signals correspond
to glycerol, cholesterol, and aminobutyl moieties; 13C NMR (CDCl3)
d 11.8, 18.7, 19.3, 21.0, 22.5, 22.8, 23.8, 24.3, 27.1, 27.3, 28.0, 28.1,
28.2, 31.8, 31.9, 35.8, 36.1, 36.5, 36.9, 38.5, 39.5, 39.7, 40.7, 42.3,
49.9, 55.2, 56.1, 56.6, 62.9, 63.4, 74.5, 86.2, 113.1, 122.5, 126.8,
127.8, 128.0, 130.0, 130.1, 135.7, 135.7, 139.8, 144.6, 156.2,
158.5; Anal. Calcd for C57H80N2O8ꢂ1/3H2O: C, 73.83; H, 8.77; N,
3.02. Found: C, 73.87; H, 8.68; N, 2.92.
3.1.4. 1-O-(4,40-Dimethoxytrityl)-2-O-[N-oleoyl-N-(4-amino-
butyl)carbamoyl]-3-O-tert-butyldimethylsilylglycerol (4)
A
mixture of 2 (1.50 g, 2.41 mmol), oleic acid (1.15 mL,
3.62 mmol), and WSCI in CH2Cl2 (12 mL) was stirred at room tem-
perature. After 24 h, the mixture was diluted with CHCl3. The or-
ganic layer was washed with H2O and brine, dried (Na2SO4), and
concentrated. The residue was purified by column chromatography
(SiO2, 0–5% MeOH in CHCl3) to give 4 (1.83 g, 2.06 mmol) in 86%
yield: 1H NMR (CDCl3) d 0.04 (s, 6H), 0.85 (s, 9H), 3.82 (s, 6H),
6.84–7.49 (m, 13H), all additional signals correspond to glycerol,
oleoyl, and aminobutyl moieties; 13C NMR (CDCl3) d ꢁ5.4, 14.1,
18.1, 22.7, 25.7, 26.6, 27.1, 27.2, 27.5, 29.1, 29.2, 29.3, 29.5, 29.6,
29.7, 29.8, 31.9, 36.7, 39.0, 40.4, 50.6, 55.1, 60.4, 62.2, 62.4, 74.0,
85.8, 113.0, 126.6, 127.7, 128.1, 129.7, 130.0, 130.1, 136.0, 136.1,
144.9, 156.3, 158.4, 173.3; Anal. Calcd for C53H82N2O7SiꢂCHCl3: C,
64.43; H, 8.31; N, 2.78. Found: C, 64.32; H, 8.33; N, 2.75.
3.2. Solid support synthesis
A mixture of 6 (0.52 g, 0.70 mmol), succinic anhydride (0.21 g,
2.10 mmol), and DMAP (86 mg, 0.70 mmol) in pyridine (7 mL)
was stirred at room temperature. After 24 h, the solution was par-
titioned between CHCl3 and H2O, and the organic layer was washed
with H2O and brine. The separated organic phase was dried
(Na2SO4) and concentrated to give a succinate. Aminopropyl con-
trolled pore glass (0.81 g, 0.18 mmol) was added to a solution of
the succinate (0.55 g, 0.61 mmol) and WSCI (0.13 g, 0.70 mmol)
in DMF (18 mL), and the mixture was kept for 72 h at room tem-
perature. After the resin was washed with pyridine, a capping solu-
tion (20 mL, 0.1 M DMAP in pyridine/Ac2O = 9:1, v/v) was added
and the whole mixture was kept for 24 h at room temperature.
The resin was washed with MeOH and acetone, and dried in vacuo.
3.1.5. 1-O-(4,40-Dimethoxytrityl)-2-O-[N-oleoyl-N-(4-aminobu-
tyl)carbamoyl]glycerol (7)
A mixture of 4 (1.66 g, 1.88 mmol) and TBAF (1 M in THF,
3.76 mL, 3.76 mmol) in THF (18 mL) was stirred at room tempera-
ture. After 7 h, the mixture was diluted with EtOAc. The organic
layer was washed with H2O and brine, dried (Na2SO4), and concen-
trated. The residue was purified by column chromatography (SiO2,
0–10% MeOH in CHCl3) to give 7 (1.19 g, 1.54 mmol) in 82% yield:
1H NMR (CDCl3) d 3.78 (s, 6H), 6.81–7.43 (m, 13H), all additional
signals correspond to glycerol, oleoyl, and aminobutyl moieties;
13C NMR (CDCl3) d 14.1, 22.6, 25.7, 26.7, 27.1, 29.1, 29.2, 29.4,
29.6, 29.7, 31.8, 32.5, 36.7, 38.9, 40.5, 55.1, 62.8, 63.1, 74.5, 86.1,
113.1, 126.7, 127.8, 128.0, 129.7, 129.9, 135.7, 144.6, 156.6,
158.4, 173.3; Anal. Calcd for C47H68N2O7ꢂ10/3CHCl3: C, 51.63; H,
6.14; N, 2.93. Found: C, 51.79; H, 6.08; N, 2.48.
Amount of loaded compound 6 to solid support was 28 lmol/g
from calculation of released dimethoxytrityl cation by a solution
of 70% HClO4/EtOH (3:2, v/v). In a similar manner, solid supports
with 7 and 8 were obtained in 83 and 88 lmol/g loading amounts,
respectively.
3.3. RNA synthesis
Synthesis was carried out with a DNA/RNA synthesizer by phos-
phoramidite method. Deprotection of bases and phosphates was
performed in concentrated NH4OH/EtOH (3:1, v/v) at room
temperature for 12 h. 20-TBDMS groups were removed by 1.0 M
tetrabutylammonium fluoride (TBAF, Aldrich) in THF at room
temperature for 12 h. The reaction was quenched with 0.1 M TEAA
buffer (pH 7.0) and desalted on a Sep-Pak C18 cartridge. Deprotec-
ted ONs were purified by 20% PAGE containing 7 M urea to give the
highly purified ON27 (49), ON28 (24), ON29 (7), ON30 (41), ON31
(19), ON32 (11), ON39(13), ON40(8), ON41(9), ON42(9), ON43(11),
and ON44(6). The yields are indicated in parentheses as OD units at
3.1.6. 1-O-(4,40-Dimethoxytrityl)-2-O-[N-cholesteryloxy-
carbonyl-N-(4-aminobutyl)carbamoyl]-3-O-tert-butyldimethyl-
silylglycerol (5)
A mixture of cholesterol (0.22 g, 0.58 mmol) and N,N0-carbon-
yldiimidazole (0.10 g, 0.63 mmol) in pyridine (3 mL) was stirred
at room temperature. After 2 h, 2 (0.43 g, 0.70 mmol) was added
to the mixture, and the whole was stirred at room temperature.
260 nm starting from 1.0 lmol scale. Extinction coefficients of the

Y. Ueno et al. / Bioorg. Med. Chem. 16 (2008) 7698–7704
7703
ONs were calculated from those of mononucleotides and dinucleo-
tides according to the nearest-neighbor approximation method.25
ture medium containing 10% FBS. The contents of each dish were
transferred separately into 1.5 mL microfuge tubes and centri-
fuged at 2500 rpm for 5 min at 4 °C. The supernatant fluid was
discarded, and the cell pellet was resuspended in 2 volumes of
hypotonic buffer A (0.5% (v/v) Nonidet P-40, 20 mM Hepes, pH
7.5, 10 mM KOAc, 15 mM Mg(OAc)2, 1 mM dithiothreitol, 1 mM
3.4. MALDI-TOF/MS analyses of RNAs
Spectra were obtained with a time-of-flight mass spectrometer.
ON27: calculated mass, 7009.5; observed mass, 7003.4. ON28: cal-
culated mass, 7035.5; observed mass, 7025.8. ON29: calculated
mass, 7183.8; observed mass, 7177.2. ON30: calculated mass,
7318.8; observed mass, 7311.9. ON31: calculated mass, 7344.8;
observed mass, 7337.4. ON32: calculated mass, 7493.0; observed
mass, 7486.3. ON39: calculated mass, 7223.7; observed mass,
7220.4. ON40: calculated mass, 7074.5; observed mass, 7070.1.
ON41: calculated mass, 7249.8; observed mass, 7250.2. ON42: cal-
culated mass, 7100.6; observed mass, 7099.4. ON43: calculated
mass, 7398.0; observed mass, 7397.8. ON44: calculated mass,
6530.0; observed mass, 6538.4.
phenylmethylsulfonyl fluoride, 10
to swell in this buffer for 10 min before being broken either by
30 strokes in tight-fitting glass Dounce homogenizer for
lg/mL aprotinin) and allowed
a
2 min in ice. The homogenate was centrifuged at 10,000g for
10 min and the supernatant was pipetted off and stored at
ꢁ80 °C. Cell extracts were separated by 7.5% SDS–PAGE. The pro-
teins were electrophoretically transferred to PVDF membrane.
Primary antibodies, mouse monoclonal anti-human RNase L anti-
body, that were obtained from Taiho Pharmaceutical Co., Ltd,
were diluted 1:5600 in TTBS containing 5% BSA and placed on
a rocker platform for 16 h at 4 °C. The membranes were washed
once for 15 min and then three times for 5 min each in TTBST.
Secondary antibody conjugate (HRP conjugated anti-mouse IgG)
was diluted 1:100,000 in TTBS containing 5% BSA and placed
on a rocker platform for 16 h at 25 °C. The membranes were
washed once for 15 min and then three times for 5 min each
in TTBST. Immunoreactive bands were detected by enhanced
chemiluminescence (utilizing ECL plus chemiluminescence detec-
tion reagents from Amersham) and subsequent exposure to X-
ray film.
3.5. Dual-luciferase assay
HeLa cells were grown at 37 °C in a humidified atmosphere of
5% CO2 in air in minimum essential medium (MEM) (Invitrogen)
supplemented with 10% fetal bovine serum (FBS). Twenty-four
hours before transfection, HeLa cells (4 ꢃ 104/mL) were transferred
to 96-well plates (100 lL per well). They were transfected, using
TransFast (Promega), according to instructions for transfection of
adherent cell lines. Cells in each well were transfected with a solu-
tion (35
amounts of siRNAs, and 0.3
duced-Serum Medium (Invitrogen), and incubated at 37 °C. After
1 h, MEM (100 L) containing 10% FBS and antibiotics was added
to each well, and the whole was further incubated at 37 °C. After
24 h, cell extracts were prepared in Passive Lysis Buffer (Promega).
Activities of firefly and Renilla luciferases in cell lysates were deter-
mined with a dual-luciferase assay system (Promega) according to
the manufacturer’s protocol. The results were confirmed by at least
three independent transfection experiments with two cultures
each and are expressed as the average from four experiments as
means SD.
l
L) of 20 ng of psiCHECK-2 vector (Promega), the indicated
Acknowledgments
lg of TransFast in Opti-MEM I Re-
This work was supported by a Grant from PRESTO of the Japan
Science and Technology Agency (JST) and a Grant-in-Aid for Scien-
tific Research from the Japan Society for the Promotion of Science
(JSPS). We are grateful to Professor Y. Tomari (the University of
Tokyo) for helpful discussions. We are also grateful to Professor
Y. Hirata (Gifu University) and Professor K. Kiuchi (Gifu University)
for providing technical assistance in the dual-luciferase assay.
l
References
1. Fire, A.; Xu, S.; Montgomery, M. K.; Kostas, S. A.; Driver, S. E.; Mello, C. C. Nature
1998, 391, 806.
2. Bumcrot, D.; Manoharan, M.; Koteliansky, V.; Sah, D. W. Y. Nat. Chem. Biol.
2006, 2, 711.
3.6. Cell culture and transfection
HT-1080 human fibrosarcoma (HT-1080) cells were cultured in
RPMI 1640 supplemented with streptomycin (0.1 mg/mL), penicil-
lin (100 U/mL), and 10% heat-inactivated fetal bovine serum. Cells
were cultured at 37 °C with a 5% CO2 atmosphere. Cells were
seeded in 35-mm dishes at density of 120,000 cells per dish and al-
lowed to attach to the culture vessel for 24 h prior to transfection.
Transfection solutions were prepared according to the manufac-
turer’s protocols with an antibiotic-free medium. siRNA was di-
3. Song, E.; Lee, S.-K.; Wang, J.; Ince, N.; Ouyang, N.; Min, J.; Chen, J.; Shankar, P.;
Lieberman, J. Nat. Med. 2003, 3, 347.
4. Soutschek, J.; Akinc, A.; Bramlage, B.; Charisse, K.; Constien, R.; Donoghue, M.;
Elbashir, S.; Geick, A.; Hadwiger, P.; Harborth, J.; John, M.; Kesavan, V.; Lavin,
G.; Pandey, R. K.; Racie, T.; Rajeev, K. G.; Röhl, I.; Toudjarska, I.; Wang, G.;
Wuschko, S.; Bumcrot, D.; Koteliansky, V.; Limmer, S.; Manoharan, M.;
Vornlocher, H.-P. Nature 2004, 432, 173.
5. Wolfrum, C.; Shi, S.; Jayaprakash, K. N.; Jayaraman, M.; Wang, G.; Pandey, R. K.;
Rajeev, K. G.; Nakayama, T.; Charrise, K.; Ndungo, E. M.; Zimmermann, T.;
Koteliansky, V.; Manoharan, M.; Stoffel, M. Nat. Biotech. 2007, 25, 1149.
6. Minakuchi, Y.; Takeshita, F.; Kosaka, N.; Sasaki, H.; Yamamoto, Y.; Kouno, M.;
Honma, K.; Nagahara, S.; Hanai, K.; Sano, A.; Kato, T.; Terada, M.; Ochiya, T.
Nucleic Acids Res. 2004, 32, e109.
7. Song, E.; Zhu, P.; Lee, S.-K.; Chowdhury, D.; Kussman, S.; Dykxhoorn, D. M.;
Feng, Y.; Palliser, D.; Weiner, D. B.; Shankar, P.; Marasco, W. A.; Lieberman, J.
Nat. Biotech. 2005, 23, 709.
8. Morrissey, D. V.; Lockridge, J. A.; Shaw, L.; Blanchard, K.; Jensen, K.; Breen, W.;
Hartsough, K.; Machemer, L.; Radka, S.; Jadhav, V.; Vaish, N.; Zinnen, S.;
Vargeese, C.; Bowman, K.; Shaffer, C. S.; Jeffs, L. B.; Judge, A.; MacLachlan, I.;
Polisky, B. Nat. Biotech. 2005, 23, 1002.
luted in 250
micrograms of Lipofectamine 2000 (Invitrogen) was diluted in
250 l of RPMI 1640 without antibiotics and serum. After 5 min
lL of RPMI 1640 without antibiotics and serum. Five
l
of incubation, they were combined and incubated for 20 min at
room temperature to form the siRNA-lipid complexes. Five hun-
dred microliters of the transfection solution was dispensed per
dish. Transfections were conducted for 6 h, after which the solu-
tion was removed and replaced with growth medium containing
antibiotics. Cells were harvested for each analysis 24 or 48 h
post-transfection.
9. McNamara, J. O., II; Andrechek, E. R.; Wang, Y.; Viles, K. D.; Rempel, R. E.;
Gilboa, E.; Sullenger, B. A.; Giangrande, P. H. Nat. Biotech. 2006, 24, 1005.
10. Tolia, N. H.; Joshua-Tor, L. Nat. Chem. Biol. 2007, 3, 36.
11. Parker, J. S.; Roe, S. M.; Barford, D. Nature 2005, 434, 663.
12. Ma, J.-B.; Yuan, Y.-R.; Meister, G.; Pei, Y.; Tuschl, T.; Patel, D. J. Nature 2005, 434,
666.
3.7. Western blot analysis for human RNase L
13. Nykänen, A.; Haley, B.; Zamore, P. D. Cell 2001, 107, 309.
14. Schwarz, D. S.; Hutvágner, G.; Haley, B.; Zamore, P. D. Mol. Cell 2002, 10, 537.
15. Lingel, A.; Simon, B.; Izaurralde, E.; Sattler, M. Nature 2003, 426, 465.
16. Yan, K. S.; Yan, S.; Farooq, A.; Han, A.; Zeng, L.; Zhou, M.-M. Nature 2003, 426,
469.
HT-1080 cells were harvested by washing the cells once with
1ꢃ PBS, aspirated, and treated with 1 mL of a trypsin solution at
37 °C for 1 min. Trypsin was inactivated by the addition of a cul-

7704
Y. Ueno et al. / Bioorg. Med. Chem. 16 (2008) 7698–7704
17. Song, J.-J.; Liu, J.; Tolia, N. H.; Schneiderman, J.; Smith, S. K.; Martienssen, R. A.;
Hannon, G. J.; Joshua-Tor, L. Nat. Struct. Biol. 2003, 12, 1026.
18. Ma, J. B.; Te, K.; Patel, D. J. Nature 2004, 429, 318.
19. Elbashir, S. M.; Martinez, J.; Patkaniowska, A.; Lendeckel, W.; Tuschl, T. EMBO J.
2001, 20, 6877.
22. Martinez, J.; Patkaniowska, A.; Urlaub, H.; Lührmann, R.; Tuschl, T. Cell 2002,
110, 563.
23. Hassel, B. A.; Zhou, A.; Sotomayor, C.; Maran, A.; Silverman, R. H. EMBO J. 1993,
12, 3297.
24. Zhou, A.; Paranjape, J.; Brown, T. L.; Nie, H.; Naik, S.; Dong, B.; Chang, A.;
Trapp, B.; Fairchild, R.; Colmenares, C.; Silverman, R. H. EMBO J. 1997, 16,
6355.
20. De Napoli, L.; Di Fabio, G.; Messere, A.; Montesarchio, D.; Musumeci, D.;
Piccialli, G. Tetrahedron 1999, 55, 9899.
21. Mammond, S. M.; Boettcher, S.; Caudy, A. A.; Kobayashi, R.; Hannon, G. J.
Science 2001, 293, 1146.
25. Puglisi, J. D.; Tinoco, I., Jr.. In Methods in Enzymology; Dahlberg, J. E., Abelson, J.
N., Eds.; Academic Press, Inc.: San Diego, 1989; Vol. 180, pp 304–325.