Introduction of 2-O-benzyl abasic nucleosides to the 3′-overhang regions of siRNAs greatly improves nuclease resistance

Article abstract of DOI:10.1016/j.bmcl.2017.10.070

Chemically modified siRNAs containing 2-O-benzyl-1-deoxy-D-ribofuranose (R^H_OBn) in their 3′-overhang region were significantly more resistant towards serum nucleases than siRNAs possessing the natural nucleoside in this region. The knockdown efficacies and binding affinities of these modified siRNAs to the recombinant human Argonaute protein 2 (hAgo2) PAZ domain were comparable with that of siRNA with a thymidine dimer at the 3′-end.

Full text of DOI:10.1016/j.bmcl.2017.10.070

Bioorganic & Medicinal Chemistry Letters xxx (2017) xxx–xxx
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Introduction of 2-O-benzyl abasic nucleosides to the 30-overhang regions
of siRNAs greatly improves nuclease resistance
Yuki Nagaya a, Yoshiaki Kitamura b, Aya Shibata b, Masato Ikeda a,b,c, Yukihiro Akao a, Yukio Kitade b,d,
⇑
a United Graduate School of Drug Discovery and Medical Information Sciences, Gifu University, 1-1 Yanagido, Gifu 501-1193, Japan
b Department of Chemistry and Biomolecular Science, Faculty of Engineering, Gifu University, 1-1 Yanagido, Gifu 501-1193, Japan
c Gifu Center for Highly Advanced Integration of Nanosciences and Life Sciences (G-CHAIN), Gifu University, 1-1 Yanagido, Gifu 501-1193, Japan
d Department of Applied Chemistry, Faculty of Engineering, Aichi Institute of Technology, 1247 Yachigusa, Yakusa-cho, Toyota, Aichi 470-0392, Japan
a r t i c l e i n f o
a b s t r a c t
D
-ribofuranose (RHOBn) in their 30-overhang
Article history:
Chemically modified siRNAs containing 2-O-benzyl-1-deoxy-
Received 6 October 2017
Revised 26 October 2017
Accepted 27 October 2017
Available online xxxx
region were significantly more resistant towards serum nucleases than siRNAs possessing the natural nucle-
oside in this region. The knockdown efficacies and binding affinities of these modified siRNAs to the recom-
binant human Argonaute protein 2 (hAgo2) PAZ domain were comparable with that of siRNA with a
thymidine dimer at the 30-end.
Ó 2017 Elsevier Ltd. All rights reserved.
Keywords:
RNA
siRNA
RNAi
Abasic nucleoside analogues
Nuclease resistance
Small interfering RNAs (siRNAs) inhibit gene expression by RNA
interference (RNAi) and thus have great potential as nucleic acid
drugs.1 RNAi technology is a useful strategy in the fight against
cancers,2 viral infections,3 and other diseases.4 However, natural
RNA strands have many problems that complicate their therapeu-
tic application, such as rapid degradation in biological media, non-
specific gene silencing (off-target effects), and poor administration
using existing drug delivery systems. To overcome these problems,
artificially modified siRNAs have been developed extensively. An
siRNA is a short (18–26 nucleotides) double-stranded RNA (dsRNA)
containing a 2-nucleotide overhang at the 30-end of each strand.5
Once siRNAs are introduced into a cell by transfection, they are
incorporated into a RNA-induced silencing complex (RISC). Each
RISC contains a helicase that unwinds the siRNA helix. Upon
unwinding, one of the strands, known as an antisense strand (guide
strand), is retained in the RISC. This antisense RNA-RISC, called
mature RISC, binds to and degrades the complementary mRNA tar-
get through base-pairing interactions. Eukaryotic translation initi-
ation factor 2C2 (EIF2C2, Argonaute protein 2, Ago2), the core
component of RISC, is considered to be the major player in RNAi.
Ago2 has a conserved structure and includes PAZ, MID, and PIWI
domains. The PAZ domain specifically recognizes the antisense
strand of dsRNA through binding to the 30-overhang region.6,7
The binding site is a hydrophobic pocket composed of aromatic
amino acids.7–10
On the basis of these findings, it has been suggested that chem-
ical modification of the 30-overhang region is an effective tech-
nique for improving the functionality of siRNA for RNAi-based
therapy.11–25 We previously reported the design and synthesis of
various chemically modified functional RNAs bearing nucleic acid
mimics at their 30-end.11–15,17,23 As part of our ongoing studies,
we found that siRNAs containing 2-O-benzyl-1-deoxy-D-ribofura-
nose (RHOBn; 1) at the 30-ends showed high nuclease resistance and
a desirable knockdown effect (Fig. 1).
Synthesis: To synthesize the desired RNAs via the conventional
phosphoramidite method using a DNA/RNA synthesizer, we pre-
pared the phosphoramidite derivative of RHOBn (2) (Scheme 1). First,
1-deoxy-D-ribofuranose (RH, 3), which can be prepared from com-
mercially available 1-O-acetyl-2,3,5-tri-O-benzoyl-b-D-ribofuranose
via reductive cleavage of the anomeric position,15 was converted
to TIPDS-RH 4 using a disiloxane protection strategy. Subsequently,
benzylation with benzaldehyde, Et3SiH, and FeCl3 in CH3NO2,26 fol-
lowed by desilylation by treatment with Et3Nꢀ3HF, gave 1 in moder-
ate yield. Treatment of 1 with 4,40-dimethoxytrityl (DMTr) chloride
in pyridine gave the corresponding 5-DMTr derivative 5, which
was phosphorylated with 2-cyanoethyl N,N-diisopropylchlorophos-
phoramidite to produce 2 in 97% yield. Oligonucleotides (ONs) con-
taining RHOBn were synthesized using an automated nucleic acid
⇑
Corresponding author at: Department of Chemistry and Biomolecular Science,
Faculty of Engineering, Gifu University, 1-1 Yanagido, Gifu 501-1193, Japan.
E-mail address: ykkitade@gifu-u.ac.jp (Y. Kitade).
https://doi.org/10.1016/j.bmcl.2017.10.070
0960-894X/Ó 2017 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Nagaya Y., et al. Bioorg. Med. Chem. Lett. (2017), https://doi.org/10.1016/j.bmcl.2017.10.070

2
Y. Nagaya et al. / Bioorganic & Medicinal Chemistry Letters xxx (2017) xxx–xxx
Fig. 1. Conceptual diagram of this study.
Scheme 1. Reagents and conditions: (a) TIPDSCl2, imidazole, DMF, rt, 82%; (b) i)
PhCHO, Et3SiH, FeCl3, MeNO2, rt; ii) Et3Nꢀ3HF, THF, rt, 47%; (c) DMTrCl, pyridine, rt,
67%; (d) (i-Pr2N)P(Cl)O(CH2)2CN, i-Pr2NEt, THF, rt, 97%.
Fig. 2. Dual-luciferase assay.
to target the Renilla luciferase gene. HeLa cells were co-transfected
with the vector and the indicated amount of each siRNA. The signal
levels of Renilla luciferase were normalized to those of firefly luci-
ferase. Fig. 2 shows the silencing activities of the siRNAs. Renilla
luciferase was suppressed by each siRNA in a dose-dependent
manner. At 1.0 nM and 10 nM, the silencing activity of siRNA 7
with RHOBn dimers was almost equal to that of siRNA 6 with natural
thymidines.
Nuclease resistance: The susceptibilities of the ONs to snake
venom phosphodiesterase (SVPD), a highly active 30-exonuclease,
were examined. Thymidine-modified ON 14 and RHOBn-modified
ON 15, each labeled with fluorescein at their 50-end, were incubated
with SVPD and the reactions were analyzed by PAGE under denatur-
ing conditions. The half-life (t1/2) of ON 14 was <4 min, and that of
ON 15 carrying a RHOBn dimer was 19 min. ON 15 was at least 5 times
more resistant to the enzyme than thymidine-modified ON 14
(Fig. 3). Furthermore, the RHOBn-modified ON 15 was more resistant
to nucleolytic hydrolysis by SVPD than the corresponding modified
synthesizer with phosphoramidite derivative 2 (Table 1). The fully
protected ONs linked to a solid support were treated with NH4OH/
MeNH2 (40% in H2O), 1:1 (v/v) at 65 °C for 30 min and with Et3Nꢀ3HF
in DMSO at 65 °C for 2.5 h. The crude product can be precipitated by
adding 3 M NaOAc, followed by n-BuOH. The mixture was cooled at
ꢁ80 °C for 12 h and centrifuged at 4 °C, 12,500 rpm, for 30 min. After
removal of the supernatant, 70% EtOH was added to the pellet. The
resulting mixture was centrifuged at 4 °C, 12,500 rpm, for 15 min.
The supernatant was removed, and the washing step was repeated.
Deprotected ONs were purified by denaturing 20% polyacrylamide
gel electrophoresis (PAGE) to isolate the desired ONs bearing the RH-
modification. These ONs were analyzed by matrix-assisted laser
OBn
desorption/ionization time-of-flight mass spectrometry (MALDI-
TOF/MS), and the observed molecular weights were in good agree-
ment with their expected structures.27
Gene silencing of Renilla luciferase: The silencing activities of the
siRNAs were examined by a dual-reporter assay using the psi-
CHECK-2 vector in HeLa cells. This vector contains the Renilla and
firefly luciferase genes, and the siRNA sequences were designed
Table 1
Sequences of ONs and siRNAs used in this study.
No. of siRNA
No. of ON
Sequence
siRNA 6
ON 10
ON 11
ON 12
ON 13
ON 14
ON 15
50-GGCCUUUCACUACUCCUACtt-30
30-ttCCGGAAAGUGAUGAGGAUG-50
siRNA 7
50-GGCCUUUCACUACUCCUACROHBnROHBn-30
30-RHOBnRHOBnCCGGAAAGUGAUGAGGAUG-50
F-50-GUAGGAGUAGUGAAAGGCCtt-30
F-50-GUAGGAGUAGUGAAAGGCCdRHdRH-30
ꢁ
ꢁ
asiRNA 8
ON 16
ON 11
ON 16
ON 13
50-GGCCUUUCACUACUCCUAC-30
30-ttCCGGAAAGUGAUGAGGAUG-50
50-GGCCUUUCACUACUCCUAC-30
30-RHOBnRHOBnCCGGAAAGUGAUGAGGAUG-50
asiRNA 9
aCapital letters indicate ribonucleosides and small letters show 20-
deoxyribonucleosides.
bF denotes fluorescein.
Fig. 3. Nuclease resistance of ONs against SVPD.
Please cite this article in press as: Nagaya Y., et al. Bioorg. Med. Chem. Lett. (2017), https://doi.org/10.1016/j.bmcl.2017.10.070

Y. Nagaya et al. / Bioorganic & Medicinal Chemistry Letters xxx (2017) xxx–xxx
3
In conclusion, we demonstrated the synthesis of chemically
modified siRNAs bearing a 2-O-benzylated abasic nucleoside in
their 30-overhang region. It was found that introduction of the aba-
sic nucleoside at the 30-end of an siRNA greatly improves resistance
towards various nucleases. Furthermore, the knockdown efficacies
and binding affinities of these modified siRNAs to recombinant
hAgo2 PAZ domain were comparable with that of siRNA with a thy-
midine dimer. We believe that this modification method is a useful
technique for increasing the duration of silencing in vivo.
Acknowledgments
This work was supported by a Grant-in-Aid for Young Scientists
(B) (No. 16K18911 (Y.K.)) and a Grant-in-Aid for Scientific Research
(B) (No. 24390025 (Y.K.)) from the Japan Society for the Promotion
of Science (JSPS). We acknowledge the Division of Instrumental
Analysis and the Division of Genomics Research, Life Science
Research Center, Gifu University, for maintaining the instruments
used in this study.
Fig. 4. Serum stability of siRNAs.
A. Supplementary data
Supplementary data associated with this article can be found, in
the online version, at https://doi.org/10.1016/j.bmcl.2017.10.070.
References
1. For a recent review on synthetic RNAi-based drugs in phase 2-3 clinical trials,
see: Titze-de-Almeida R, David C, Titze-de-Almeida SS. Pharm Res.
2017;34:1339–1363.
2. For a recent review on RNAi-based strategies for cancer therapy, see: Kim HJ,
Kim A, Miyata K, Kataoka K. Adv Drug Deliv Rev. 2016;104:61–77.
3. For a recent review on RNAi-based strategies for therapy and prevention of
HIV-1/AIDS, see: Swamy MN, Wu H, Shankar P. Adv Drug Deliv Rev.
2016;103:174–186.
4. For a recent review on RNAi-based strategies for amelioration or prevention of
hypercholesterolemia, see: Fitzgerald K, White S, Borodovsky A, et al. New Engl J
Med. 2017;376:41–51.
5. Tomari Y, Matranga C, Haley B, Martinez N, Zamore PD. Science.
2004;306:1377–1380.
Fig. 5. Binding affinities of asiRNAs to the recombinant hAgo2 PAZ domain.
6. Takahashi T, Zenno S, Ishibashi O, Takizawa T, Saigo K, Ui-Tei K. Nucleic Acid Res.
2014;42:5256–5269.
7. Ma JB, Ye K, Patel DJ. Nature. 2004;429:318–322.
ON bearing the parent abasic ribo- and deoxyribonucleoside
(Fig. S1).15,23
8. Song JJ, Liu J, Tolia NH, et al. Nat Struct Biol. 2003;10:1026–1032.
9. Lingel A, Simon B, Izaurralde E, Sattler M. Nature. 2003;426:465–469.
10. Yan KS, Yan S, Farooq A, Han A, Zeng L, Zhou MM. Nature. 2003;426:469–474.
11. Ueno Y, Naito T, Kawada K, et al. Biochem Biophys Res Commun.
2005;330:1168–1175.
12. Ueno Y, Inoue T, Yoshida M, et al. Bioorg Med Chem Lett. 2008;18:5194–5196.
13. Ueno Y, Watanabe Y, Shibata A, et al. Bioorg Med Chem. 2009;17:1974–1981.
14. Somoza A, Terrazas M, Eritja R. Chem Commun. 2010;46:4270–4272.
15. Taniho K, Nakashima R, Mahmoud K, Kitamura Y, Kitade Y. Bioorg Med Chem
Lett. 2012;22:2518–2521.
Serum stability: Next, we tested the stability of chemically mod-
ified siRNA with RHOBn in 50% bovine serum. At various incubation
times, aliquots of each siRNA were analyzed by 20% native PAGE to
detect any degradation products. As shown in Fig. 4, siRNA 7 pos-
sessing RHOBn was more resistant to serum-derived nucleases than
unmodified siRNA 6 with natural thymidines. Unmodified siRNA 6
was largely degraded within 1 h, whereas less than 40% of modified
siRNA 7 was degraded after 3 h and approximately 15% remained
intact after 24 h incubation (Fig. S2).
16. Gaglione M, Potenza N, Di Fabio G, et al. ACS Med Chem Lett. 2013;4:75–78.
17. Luo X, Sugiura T, Nakashima R, Kitamura Y, Kitade Y. Bioorg Med Chem Lett.
2013;23:4157–4161.
Binding affinity with hAgo2 PAZ: We further investigated the
binding affinities of a recombinant hAgo2 PAZ domain protein to
the chemically modified siRNAs (Fig. 5).20 In order to simplify the
study, we used asymmetric siRNAs (asiRNAs) with an overhang
region at the 30-end of the antisense strand and a blunt end struc-
ture at the 30-end of the sense strand (ON 16). The asiRNA 8 con-
sists of ON 11 and ON 16, and the asiRNA 9 was duplex of ON 13
and ON 16. The addition of hAgo2 PAZ protein resulted in a slight
decrease in the absorbance at 312 nm. The binding affinity of
asiRNA 9 possessing RHOBn is almost the same as that of asiRNA 8 car-
rying a thymidine dimer (Fig. S3). This result suggests that asiRNA 9
with RHOBn can bind the hAgo2 PAZ domain in a very similar manner
to that of asiRNA 8 binding with natural thymidine dimers.
18. Kitamura Y, Masegi Y, Ogawa S, et al. Bioorg Med Chem. 2013;21:4494–4501.
19. Gaglione M, Mercurio ME, Potenza N, et al. Biomed Res Int. 2014;901617.
20. Inada N, Nakamoto K, Yokogawa T, Ueno Y. Eur J Med Chem. 2015;103:460–472.
21. Xu L, Wang X, He H, et al. Biochemistry. 2015;54:1268–1277.
22. Wang X, Zhang S, Dou Y, et al. PLoS Genet. 2015;11:e1005091.
23. Nagaya Y, Kitamura Y, Nakashima R, Shibata A, Ikeda M, Kitade Y. Nucleosides
Nucleotides Nucleic Acids. 2016;35:64–75.
24. Pendergraff HM, Debacker AJ, Watts JK. Nucleic Acid Ther. 2016;26:216–222.
25. Ma Y, Liu S, Wang Y, et al. Org Biomol Chem. 2017;15:5161–5170.
26. Iwanami K, Yano K, Oriyama T. Chem Lett. 2007;36:38–39.
27. MALDI-TOF/MS analyses of ONs. The following spectra were obtained by
MALDI-TOF/MS (negative mode). ON 12: calculated mass, 6462.9; observed
mass, 6461.8. ON 13: calculated mass, 6772.0; observed mass, 6772.8. ON 15:
calculated mass, 7310.1; observed mass, 7310.8.
Please cite this article in press as: Nagaya Y., et al. Bioorg. Med. Chem. Lett. (2017), https://doi.org/10.1016/j.bmcl.2017.10.070