A ‘catch and release’ strategy towards HPLC-free purification of synthetic oligonucleotides by a combination of the strain-promoted alkyne-azide cycloaddition and the photocleavage

Article abstract of DOI:10.1016/j.bmc.2017.09.014

A convenient strategy to purify oligonucleotides (ONs) synthesized by solid phase synthesis on an automatic DNA/RNA synthesizer was described. By attaching a photocleavable azide linker as the last phosphoramidite unit in the ON synthesis, only the desire

Full text of DOI:10.1016/j.bmc.2017.09.014

Accepted Manuscript
A ‘catch and release’ strategy towards HPLC-free purification of synthetic oli-
gonucleotides by a combination of the strain-promoted alkyne-azide cycload-
dition and the photocleavage
Yosuke Igata, Noriko Saito-Tarashima, Daiki Matsumoto, Kazuyuki Sagara,
Noriaki Minakawa
PII:
S0968-0896(17)31570-5
DOI:
http://dx.doi.org/10.1016/j.bmc.2017.09.014
BMC 13974
Reference:
Bioorganic & Medicinal Chemistry
To appear in:
Received Date:
Revised Date:
Accepted Date:
5 August 2017
7 September 2017
9 September 2017
Please cite this article as: Igata, Y., Saito-Tarashima, N., Matsumoto, D., Sagara, K., Minakawa, N., A ‘catch and
release’ strategy towards HPLC-free purification of synthetic oligonucleotides by a combination of the strain-
promoted alkyne-azide cycloaddition and the photocleavage, Bioorganic & Medicinal Chemistry (2017), doi: http://
dx.doi.org/10.1016/j.bmc.2017.09.014
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

A ‘catch and release’ strategy towards HPLC-free purification of synthetic
oligonucleotides by a combination of the strain-promoted alkyne-azide
cycloaddition and the photocleavage
Yosuke Igata, Noriko Saito-Tarashima, Daiki Matsumoto, Kazuyuki Sagara and Noriaki
Minakawa*
Graduate School of Pharmaceutical Science, Tokushima University, Shomachi 1-78-1, Tokushima
770-8505, Japan
Keywords: oligonucleotide synthesis, oligonucleotide purification, Staudinger ligation,
strain-promoted alkyne-azide cycloaddition, click chemistry on solid supports
*Author to whom correspondence and reprint requests should be addressed.
Phone and Fax: +81-88-633-7288.
E-mail: minakawa@tokushima-u.ac.jp
1

Abstract
A convenient strategy to purify oligonucleotides (ONs) synthesized by solid phase
synthesis on an automatic DNA/RNA synthesizer was described. By attaching a photocleavable
azide linker as the last phosphoramidite unit in the ON synthesis, only the desired full-length
sequence was ‘caught’ on a controlled pore glass (CPG) resin possessing an
aza-dimethoxycyclooctyne (DIBAC) derivative. Washing the resulting CPG resin to remove all
unbounded species, the subsequent photoirradiation allowed the pure ONs to be ‘released’ without
leaving any chemical modifications on native ON structure or chemical reagents from the solid
phase ON synthesis.
2

Introduction
Synthetic oligonucleotides (ONs) and their unnatural analogues are widely used, not only
as scientific research tools, but also as therapeutics agents. Thus, practical and scalable
preparation of pure synthetic ONs is required. Synthetic ONs are typically prepared on an
automatic DNA/RNA synthesizer using appropriately protected nucleoside phosphoramidite
building blocks. In the general phosphoramidite method, each synthetic cycle consists of
detritylation, coupling, capping, and oxidation steps.1,2 The first nucleoside, whose 5’-OH is
protected with a 4,4’-dimethoxytrityl (DMTr) group, is anchored to a controlled pore glass
(CPG) resin via its 3’-OH group through a cleavable linker under basic conditions. After
detritylation under acidic conditions, the next nucleoside phosphoramidite monomer can be
coupled with the 5’-OH of the CPG-anchored nucleoside. Normally, this coupling step is highly
efficient, but the generation of truncated failure sequences is unavoidable, especially in a long
ON synthesis. To aid purification of the resulting ON, the unreacted 5’-OH is capped with an
acetyl (Ac) group to protect it from reacting with the next phosphoramidite monomer. Then, the
phosphite(III) linkage is oxidized with iodine to give a phosphate(V) linkage. After repeating
these procedures to afford the intended length, the ONs are cleaved from the solid support, fully
deprotected under basic conditions, and purified using tedious techniques such as column
chromatography, gel electrophoresis, or reversed-phase HPLC (RP-HPLC).
There have been several reports concerning rapid purification methods for synthetic ONs
using biotinylation,3–5 polymerization,6,7 and perfluorination aproaches.8 Additionally, Agustin et
al. have recently reported a robust purification strategy using the inverse electron demand
Diels–Alder (IEDDA) reaction between trans-cyclooctene (TCO) and tetrazine.9 In their
approach, termed ‘click-to-release’, the nucleoside phosphoramidite derivative possessing a TCO
group on its nucleobase moiety is incorporated into ON as a final coupling reaction in the
standard phosphoramidite ON synthesis. The resulting ON having a TCO group at the 5’-end are
‘clicked’ to the tetrazine group on solid supports via the IEDDA reaction. Next, the solid
supports are washed to remove all unbounded species, and then suspended in water to allow the
‘release’ step to occur, yielding the purified ON. This approach based on IEDDA chemistry
allows for robust purification of synthetic ONs under more mild conditions than the previously
3

reported rapid synthetic ON purification methods; however, the resulting ON are inevitably
modified with an aminomethyl group on the 5’-end.
We have developed a convenient ON purification method termed the ‘catch and release’
method by combining a biotin–streptavidin interaction and a photocleavable group.5 In this
method, the biotinylated photocleavable group is introduced at the 2’-position of a uridine
phosphoramidite unit. This amidite unit is coupled to the 5’-end of the ON; and thus, the
resulting ON with a full-length sequence can be ‘caught’ on a streptavidin resin via the
high-affinity biotin–streptavidin interaction. After washing the streptavidin resin,
photoirradiation allows the purified ON to be ‘released’ without leaving any chemical
modifications on the native ON structure. Using a photocleavable group as the ‘release’ system
is thought to be useful for purifying ONs under mild conditions and without the addition of more
chemicals. Thus, an ON solution that is purified using this ‘catch and release’ method could be
directly applied to any type of biological usage. However, a ‘catch’ system based on the
biotin–streptavidin interaction still has room for improvement because preparation of the
phosphoramidite unit possessing a biotinylated photocleavable group is tedious and the
streptavidin resin is too expensive to market our ‘catch and release’ method as a convenient and
practical purification system. Herein, a suitable combination of chemistry and solid support was
explored to ‘catch’ synthetic ONs for a more convenient ‘catch and release’ purification method.
By using an azide-fictionalized photocleavable linker, we succeeding in rapidly purifying
synthetic ONs.
4

Results and Discussion
We first planned a Staudinger ligation between an azide and a triarylphosphine on a
polystyrene (PS) resin scaffold for the ‘catch’ reaction. Thus, a photocleavable azide linker,
termed the ‘catch and release’ tag, phosphoramidite unit 1 and a triarylphosphine derivative
anchored on PS resin 2 were designed (Figure 1).
Figure 1: Structure of the ‘catch and release’ tag and solid supports to ‘catch’ the ONs.
The synthesis of 1 is shown in Scheme 1. Starting with
1-(5-azidomethyl-2-nitrophenyl)ethanol (3),5,10 the azide group of 3 was converted into an amino
group with Ph3P to afford 4. Then, the resulting 4 was treated with 5-bromovaleric acid in the
presence of O-(benzotriazol-l-yl)-N,N,N’,N’-tetramethyluronium hexafluorophosphate (HBTU),
1-hydroxybenzotriazole (HOBt), and N,N-diisopropylethylamine (DIPEA) to give 5, which was
then converted into the corresponding phosphoramidite 1 without converting the bromo group
located on the end of the linker into an azide group. This is noteworthy because conversion of the
bromo to an azide group is problematic, since an azide group is easily reduced into an amino
group when treated with phosphite(III) reagents.11 As a proof-of-principle, the resulting
phosphoramidite derivative 1 was used to examine the ‘catch and release’ reaction on the
5

nucleoside monomer unit. Thus, phosphoramidite 1 was coupled with the 5’-OH of thymidine
derivative 612 with the aid of 5-benzylthio-1H-tetrazole (BTT) in CH3CN (Scheme 2), and the
subsequent oxidation of the phosphite(III) linkage using tert-butyl hydroperoxide (TBHP) gave
phosphotriester 7. Then, the bromo group located on the end of 7 was converted into an azide
group using sodium azide to afford a nucleotide unit possessing a ‘catch and release’ tag, 8.
Concurrently, the PS resin possessing triarylphosphine derivative 2 was also prepared (see the
Methods section for details).
Scheme 1: Synthesis of a the ‘catch and release’ tag phosphoramidite unit. Reagents and
conditions: (a) PPh3, THF–H2O, 83%; (b) 5-bromovaleric acid, HBTU, HOBt, DIPEA, DMF,
quant; (c) 2-cyanoethyl diisopropylchlorophosphoramidite, DIPEA, DMAP, CH2Cl2, 70%.
Scheme 2: Synthesis of a nucleoside monomer possessing the ‘catch and release’ tag
(photocleavable azide linker). Reagents and conditions: (a) 1, BTT, 4Å MS, CH3CN, 1.0 M
TBHP; (b) sodium azide, DMF, 78%.
Then, the ‘catch’ reaction of 8 with PS resin 2 via Staudinger ligation was examined. The
solution of nucleotide monomer 8 in CH3CN–H2O (4:1) was incubated with PS resin 2. Then, an
6

aliquot of the reaction mixture was analyzed using HPLC (Figure 2A). Most of the two peaks
observed around 10 min arising from the diastereomers of 8 were not consumed after 48 h
incubation. Because the ‘catch’ reaction between an azide group on the nucleoside monomer 8
and a triarylphosphine on the PS resin 2 was not efficient, we next planned to use a
strain-promoted alkyne-azide cycloaddition (SPAAC) instead. Thus far, SPAAC systems such as
oxanobornadienes,13 cyclooctynes,14 and dibenzo cyclooctynes,15,16 have been developed. From
these SPAAC systems, aza-dimethoxycyclooctyne (DIBAC)17 is one of the most synthetically
accessible alkyne analogues and its second-order rate constant (k2 = 0.31 M–1s–1) in SPAAC is
faster than that of Staudinger ligation (k2 = 0.0025 M–1s–1).18 In addition, the chemoselectivity of
the reaction between DIBAC and an azide group is reported to be superior than that achieved
between a triarylphosphine and an azide.19 A solution containing nucleotide monomer 8 in
CH3CN–H2O (4:1) was incubated with the DIBAC-anchored PS resin 9.20 Then, the ‘catch’
reaction was analyzed using HPLC. As can be seen in Figure 2B, two peaks arising from the
diastereomers of 8 observed at the retention time of 10 min almost disappeared after 12 h
incubation at room temperature, indicating that the catch reaction between an azide group and a
DIBAC anchored onto the PS resin preceded and 8 was completely immobilized on the resin.
After the resulting PS resin was washed with CH3CN–H2O (4:1) to remove all unbounded species,
the PS resin-containing suspension was irradiated with a xenon lamp (375 nm, 200 W) using a
Pyrex filter (>300 nm) (Figure 2C). In the HPLC profiles of an aliquot of the reaction solution,
only the peak indicating the thymidine 5’-monophosphate derivative was observed at around the
6 min retention time mark. With increased irradiation time, the peak intensity grew. Thus, our
‘catch and release’ approach conveniently afforded the purified nucleotide monomer.
7

A
8
48 h
24 h
0 h
0
5
10
R/T (min)
15
20
B
8
12 h
6 h
3 h
1 h
0 h
20
0
5
10
15
R/T (min)
C
O
N
NH
O
O
HO P O
OCE
O
AcO
30 min
20 min
10 min
5 min
1 min
0 min
system peak
0
5
10
R/T (min)
15
20
Figure 2: HPLC profiles of: (A) the ‘catch’ reaction between a nucleoside monomer 8 and
triarylphosphine derivative on PS resin 2, (B) the ‘catch’ reaction between nucleoside monomer
8 and DIBAC derivative on PS resin 9, and (C) the subsequent ‘release’ reaction to afford a
8

thymidine 5’-monophosphate derivative. The samples were analyzed using HPLC with a
-BONDASPHERE column (Waters, 150 × 3.9 mm), eluted with a linear gradient from 28% to
37% CH3CN in 0.1 N triethylammonium acetate (TEAA) buffer (pH 7.0) (1 mL/min for 20 min).
CE indicates a cyanoethyl group.
9

d(a)
DMTrO cagtgcactagagca
Step 2 Detritylation
Step 1
ODN synthesis
HO cagtgcactagagca
Coupled with 1
Step 3
O
NPE linker
O P O cagtgcactagagca
OCE
NO2
O
NH
NaN3
Step 4
Br
4
O
N3 NPE linker O P O cagtgcactagagca
OCE
‘Catch and release’ tag
Step 5 NH4OH
O
N3 NPE linker O P O cagtgcactagagca OH
O–
Step 6 ‘Catch’ reaction
‘Catch’ resin
Step 7 Wash
failuer sequences
reagents
O
N3 NPE linker O P O cagtgcactagagca OH
O–
Step 8 ‘Release’ reaction
N3 NPE linker
O
HO P O cagtgcactagagca OH
O–
Figure 3: Schematic representation for purifying synthetic ONs using our ‘catch and release’
strategy. CE indicates a cyanoethyl group.
With these successful results in hand, we expanded our purification approach to a
synthetic ON. A schematic representation of the experiment is shown in Figure 3. First, a 15-mer
ON sequence was prepared on an automatic DNA/RNA synthesizer (step 1). After deprotection
10

of the DMTr group at the 5’-end (step 2), the ‘catch and release’ tag, phosphoramidite derivative
1, was coupled with the resulting 5’-OH as the final phosphoramidite following standard
coupling, capping, and oxidation procedures on the CPG support (step 3). Then, the resulting
CPG support was treated with sodium azide for 12 h at room temperature to convert the bromo
group into an azide group (step 4). After the completion of solid phase synthesis, the resulting
ON was cleaved from the solid support and fully deprotected by the treatment of NH4OH (step 5).
The cleaved solid supports were filtered off and the filtrates were concentrated in vacuo.
At this stage, an aliquot of the sample was analyzed on RP-HPLC. As can be seen in the
bottom trace of Figure S1 (0 h), the full-length sequence possessing the ‘catch and release’ tag
(photocleavable azide linker) was observed at around retention time 18 min as two main peaks
arising from the existence of diastereomers at the photocleavable moiety. Multiple peaks that
indicate undesired sequences, failed syntheses, appear between 3 to 12 min. The crude solution
(containing the full-length sequence possessing a ‘catch and release’ tag, failure sequences, and
other impurities such as reagents) was then incubated with a DIBAC derivative anchored on PS
resin 9 in order to ‘catch’ the desired full-length ON (step 6). Figure S1 shows HPLC profiles of
the reaction mixture supernatant. The full-length sequence peaks were estimated to remain over
90% after 12 h incubation, indicating the SPAAC reaction of DIBAC on PS resin 9 against an
azide group on ON was quite slow compared with that against an azide group on a nucleotide
monomer (Figure S1 vs. Figure 2B). In this catch reaction, the DIBAC unit is considered to be
located in a hydrophobic environment because the PS resin is made of a styrene copolymer.
Conversely, the ON with the ‘catch and release’ tag is a hydrophilic molecule. The poor
accessibility between ON and PS resin arising from the hydrophilicity differences could lead to a
loss of reaction efficiency. Thus, we next examined the ‘catch’ reaction of ON using a DIBAC
derivative on the CPG resin 10 (step 6).21 Because CPG resin is used as a standard solid support
for ON solid phase synthesis, it is thought to be suitable for catching ONs. The solution
containing the full-length ON possessing a ‘catch and release’ tag, failure sequences, and other
impurities such as reagents was then incubated with a DIBAC derivative anchored on CPG resin
10. As shown in Figure 4A, only the peaks of the full-length sequences observed at around 18
min disappeared as the incubation times were elongated. After 120 min of incubation, unlike the
11

reaction on a PS resin, the full-length sequences possessing the ‘catch and release’ tag were
completely immobilized onto the CPG resin via the SPAAC reaction. Then, the resulting CPG
resin was washed 3 times with H2O, CH3CN, and acetone to remove all unbound species (this
process took 20 min) (step 7), suspended in 20 mM Tris-HCl buffer, and then irradiated with a
Xe lamp using a Pyrex filter (>300 nm) to execute the ‘release’ step (step 8). Figure 4B shows
the HPLC profiles of the reaction mixture supernatant after the indicated irradiation time. The
peak corresponding to the desired 15 mer ON possessing a 5’-phosphomonoester group was
observed at 6 min and its intensity increased as the irradiation times were elongated. The overall
efficiency of the ‘catch and release’ purification (overall recovery) was estimated to be over 85%
after 120 min irradiation, indicating our ‘catch and release’ strategy is very effective for the
purification of synthetic ONs (Figure S2). To examine a versatility of this ‘catch and release’
strategy, we further examined synthesis and purification of a longer ON (73 mer) in the similar
manner as described in Figure 3. The HPLC profiles of ‘catch’ and ‘release’ are shown in Figure
4C and 4D, respectively. As can be seen, it has been proved that this strategy can be applicable
even in a longer ON.
12

A
C
120 min
90 min
60 min
30 min
0 min
25
360 min
180 min
60 min
0 min
20
0
5
10
15
20
0
5
10
15
R/T (min)
R/T (min)
B
D
120 min
90 min
60 min
30 min
5 min
1 min
0 min
0 min
0
5
10
15
20
25
0
5
10
15
20
R/T (min)
R/T (min)
Figure 4: HPLC profiles of: (A) the ‘catch’ reaction between a 15 mer ON possessing the ‘catch
and release’ tag on its 5’-end and a DIBAC derivative on CPG resin 10, (B) the subsequent
‘release’ reaction to afford a purified synthetic ON, (C) the reaction with a 73 mer ON, and (D)
the subsequent ‘release’ reaction. Sequence of the 73 mer is
5’-aagtgcgttgctagtaccaactctcttgaagttggtactagcaacgcactatagtgagtcgtattaatttcgc-3’. The samples
were analyzed using HPLC with a -BONDASPHERE column (Waters, 150 × 3.9 mm), eluted
with a linear gradient from 14% to 22% CH3CN in 0.1 N TEAA buffer (pH 7.0) (1 mL/min for
25 min) for (A) and (B), and a linear gradient from 5% to 23% CH3CN in 0.1 N TEAA buffer
(pH 7.0) (1 mL/min for 25 min) at 50 °C for (C) and (D).
Conclusion
In summary, we have explored a suitable combination of click chemistry and solid support
to purify synthetic ONs using our ‘catch and release’ strategy. The phosphoramidite derivative of a
photocleavable bromo linker 1 was coupled with the 5’-end of a synthetic ON using an automatic
13

DNA/RNA synthesizer. After the conversion of the bromo group, located on the end of the linker
moiety, into an azide group, the resulting ON was detached from the solid support. Then, the
desired full-length ON possessing the photocleavable ‘catch and release’ azide linker was ‘caught’
on a DIBAC derivative affixed to a CPG resin via the SPAAC reaction, First and second
generation approaches using a triarylphosphine derivative on a PS resin via Staudinger ligation and
a DIBAC derivative on a PS resin via the SPAAC reaction, respectively, were not effective.
Following the ‘release’ by photoirradiation, pure synthetic ONs without any additional undesired
chemical modifications were obtained free from any reagents used in the solid phase ON synthesis.
We are currently expanding our ‘catch and release’ purification methods toward longer sequences
(> 100-mer) and RNA synthesis.
Experimental Section
General Methods.
Physical data were measured as follows; 1H, 13C, and 31P NMR spectra were recorded at 400,
or 500 MHz, 100, or 125, and 160 or 200 MHz instruments (Bruker FT-NMR AV400 or AV500),
respectively in CDCl3 or DMSO-d6 as the solvent with tetramethylsilane (for 1H and 13C NMR) or
phosphoric acid (31P NMR). Chemical shifts are reported in parts per million (), and signals are
expressed as s (singlet), d (doublet), t (triplet), q (quartet), m (multiplet), or br (broad). All
exchangeable protons were detected by addition of D2O. Mass spectra were measured on a SQD2
Waters), SYNAPT G2-Si HDMS (Waters), and microflex MALDI-TOF MS (MALDI, Bruker).
TLC was done on Merck Kieselgel F254 precoated plates. Silica gels used for column
chromatography were KANTO Chemical silica gel 60 and KANTO Chemical silica gel 60N
(neutral). Reagents for ON synthesis were purchased from Glen Research. PS-resin was purchased
from Sigma-Aldrich (0.5–1.0 mmlo/g, 100–200 mesh), while CPG was from Glen Research (500
Å, 108 mol/g).
Chemistry
1-{(5-Aminomethyl)-2-nitrophenyl}ethanol (4).5 To a solution of 35,10 (1.12 g, 5.04 mmol)
in THF (20 mL) containing Ph3P (1.59 g, 6.05 mmol) was added H2O (6 mL). After being stirred
for 4 h at room temperature, the solvent was removed in vacuo. The residue was purified by a
14

silica gel column, eluted with MeOH in CHCl3 (0%–20%), to give 4 (816 mg, 83%) as a yellow
solid: 1H NMR (CDCl3)  7.91 (1 H, d, J = 8.3 Hz), 7.79 (1 H, d, J = 1.8 Hz), 7.38 (1 H, dd, J = 1.8
and 8.3 Hz), 5.46 (1 H, q, J = 6.3 Hz), 3.99 (2 H, s), 1.57 (3 H, s).
1-[5-{N-(5-Bromopentanoyl)aminomethyl}-2-nitrophenyl]ethanol (5). To a solution of 4
(727 mg, 3.71 mmol) in DMF (20 mL) containing HBTU (2.12 g, 5.57 mmol), HOBt (54 mg, 0.4
mmol) and DIPEA (2 mL, 11.13 mmol) was added 5-bromovaleric acid (1.02 g, 5.57 mmol), and
the whole mixture was stirred for 10 min at room temperature. The reaction was quenched by
addition of ice, and the solvent was removed in vacuo. The residue was partitioned with AcOEt
and H2O, and the organic layer was washed with saturated aqueous NaHCO3, followed by brine,
dried (Na2SO4) and concentrated in vacuo. The residue was purified by a silica gel column, eluted
with hexane/AcOEt (1/2–0/1), to give 5 (1.21 g, quant) as a pale yellow oil: ESI-LRMS m/z 359
(MH+); ESI-HRMS calcd for C14H20BrN2O4 359.0606, found 359.0592; 1H NMR (CDCl3)  7.90 (1
H, d, J = 8.5 Hz), 7.73 (1 H, d, J = 1.7 Hz), 7.31 (1 H, dd, J = 1.7 and 8.5 Hz), 5.97 (1 H, br s),
5.44 (1 H, q, J = 6.2 Hz), 4.52 (2 H, d, J = 6.2 Hz), 3.42 (2 H, t, J = 6.2 Hz), 2.29 (2 H, t, J = 7.5
Hz), 1.96–1.80 (4 H, m), 1.55 (3 H, d, J = 6.2); 13C NMR (CDCl3,)


1-[5-{N-(5-Bromopentanoyl)aminomethyl}-2-nitrophenyl]-{1-O-(N,N-diisopropylamin
o-2-cyanoethoxyphosphino)}ethanol (1). To a solution of 5 (539 mg, 1.5 mmol) in CH2Cl2 (30
mL) containing 4-dimethylaminopyridine (DMAP) (9 mg, 0.075 mmol) and DIPEA (1.0 mL, 6.0
mmol) was added 2-cyanoethyl diisopropylchlorophosphoramidite (0.37 mL, 1.65 mmol) at 0 °C,
and the whole mixture was stirred for 10 min at room temperature. The reaction was quenched by
addition of ice, and the solvent was partitioned with CHCl3 and H2O, and the organic layer was
washed with saturated aqueous NaHCO3, followed by brine, dried (Na2SO4) and concentrated in
vacuo. The residue was purified by a silica gel column, eluted with hexane/AcOEt (1/2–1/4), to
give 1 (572 mg, 70%) as a yellow oil: ESI-LRMS m/z 559 (MH+); ESI-HRMS calcd for
C23H37BrN4O5P 559.1685, found 559.1671; 1H NMR (CDCl3)  7.89 (1 H, d, J = 8.2 Hz), 7.88 (1 H,
15

d, J = 8.5 Hz), 7.85 (1 H, d, J = 1.8 Hz), 7.76 (1 H, d, J = 1.7 Hz), 7.35 (1 H, dd, J = 1.8 and 8.5
Hz), 7.32 (1 H, dd, J = 1.7 and 8.2 Hz), 6.66 (1 H, br s), 6.27 (1 H, br s), 5.59–5.51 (2 H, m), 4.60
(1 H, dd, J = 6.3 and 15.1 Hz), 4.50 (2 H, d, J = 6.3 Hz), 4.40 (1 H, dd, J = 6.3 and 15.1 Hz),
3.99–3.83 (2 H, m), 3.77–3.61 (4 H, m), 3.52–3.45 (2 H, m) 3.42 (4 H, t, J = 6.3 Hz), 2.70–2.43 (4
H, m), 2.31–2.24 (4 H, m), 1.92–1.87 (4 H, m), 1.85–1.76 (4 H, m), 1.57 (3 H, d, J = 6.3 Hz), 1.56
(3 H, d, J = 6.3 Hz), 1.22 (6 H, d, J = 6.7 Hz), 1.17 (6 H, d, J = 6.7 Hz), 1.13 (6 H, d, J = 6.7 Hz),
0.90 (6 H, d, J = 6.7 Hz); 31P NMR (CDCl3)  147.14, 145.24.
1-(3-O-Acetyl-5-O-[1-{5-N-(5-bromopentanoyl)aminomethyl-2-nitrophenyl}ethyl](2-cy
anoethyl)phosphono-2-deoxy--D-ribofuranosyl)thymidine (7). To a solution of
3’-O-acetylthymidine 6 (88 mg, 0.31 mmol)12 in CH3CN (3.0 mL) containing powdery 4Å MS (10
mg) were added 1 (254 mg, 0.47 mmol) and 5-benzyl-1H-tetrazole (6.04 mL, 1.51 mmol). After
being stirred for 1 h at room temperature, a solution of 1.0 M TBHP/toluene (0.62 mL, 0.62 mmol)
was added to the reaction mixture, and the whole was stirred for 1.5 h at the same temparature. The
mixture was partitioned with AcOEt and H2O, and the organic layer was washed with saturated
aqueous NaHCO3, followed by brine, dried (Na2SO4) and concentrated in vacuo. The residue was
purified by a silica gel column, eluted with hexane/AcOEt (1/4–0/1) and then AcOEt/acetone
(4/1–2/3), to give 7 (161 mg, 68%) as a pale yellow oil: ESI-LRMS m/z 758 (MH+); ESI-HRMS
calcd for C29H38BrN5O12P 758.1438, found 758.1432; 1H NMR (CDCl3) see the spectrum in the
Supporting Information; 31P NMR (CDCl3)  –2.23, –2.28, –2.34, –2.52.
1-(3-O-Acetyl-5-O-[1-{5-N-(5-azidopentanoyl)aminomethyl-2-nitrophenyl}ethyl](2-cya
noethyl)phosphono-2-deoxy--D-ribofuranosyl)thymidine (8). To a solution of 7 (119 mg, 0.16
mmol) in DMF (1.6 mL) was added sodium azide (21 mg, 0.32 mmol), and the whole mixture was
stirred for 2 h at room temperature. The solvent was removed in vacuo, and the residue was
partitioned with AcOEt and H2O, and the organic layer was washed with saturated aqueous
NaHCO3, followed by brine, dried (Na2SO4) and concentrated in vacuo. The residue was purified
by a silica gel column, eluted with hexane/AcOEt (1/4–0/1), and then AcOEt/acetone (4/1), to give
8 (88 mg, 78%) as a pale yellow oil: ESI-LRMS m/z 721 (MH+); ESI-HRMS calcd for C29H38N8O2P
16

721.2347, found 721.2344; 1H NMR (CDCl3) see the spectrum in the Supporting Information; 31P
NMR (CDCl3)  –2.25, –2.31, –2.36, –2.51.
Preparation of the catch resin
PS resin possessing triarylphosphine unit (2). A suspension of PS resin (0.5–1.0 mmol/g,
300 mg) in DMF (5 mL) was kept for 10 min in a 20 mL of reservoir tube. Then,
1-methyl-2-diphenylphosphinoterephthalate (80 mg, 0.22 mmol)22,23, N,N’-diisopropylcarbodiimide
(DIC) (77 L, 0.5 mmol) and HOBt (68 mg, 0.5 mmol) were successively added to the above
suspension, and the whole mixture was gently shaken for 24 h at room temperature. After being
removed the solvent, the resin was washed with H2O, methanol, followed by acetone. The resin in
the reservoir tube was dried well by argon purge, and then, pyridine (5.4 mL) and acetic anhydride
(0.6 mL, 6.4 mmol) were added to the resin to cap the remaining amino groups. After being shaken
for 2 h, the resin was filtered, washed with H2O, methanol, followed by acetone and dried under
reduced pressure to give PS resin 2. The loading amount of cyclooctyne unit was estimated to be
0.57 mmol/g based on the weight of resin before and after the reaction.
PS resin possessing cyclooctyne unit (9). A suspension of PS resin (0.5–1.0 mmol/g, 600
mg) in DMF (6 mL) was kept for 10 min in a 20 mL of reservoir tube. Then,
dibenzocyclooctyne-acid (80 mg, 0.26 mmol)24, DIC (60 L, 0.4 mmol) and HOBt (54 mg, 0.4
mmol) were successively added to the above suspension, and the whole mixture was gently shaken
for 24 h at room temperature. After being removed the solvent, the resin was washed with H2O,
methanol, followed by acetone. The resin in the reservoir tube was dried well by argon purge, and
then, pyridine (5.4 mL) and acetic anhydride (0.6 mL, 6.4 mmol) were added to the resin to cap the
remaining amino groups. After being shaken for 2 h, the resin was filtered, washed with H2O,
methanol, followed by acetone and dried under reduced pressure to give PS resin 9. The loading
amount of cyclooctyne unit was estimated to be 0.25 mmol/g based on the comparison of UV
absorption (at 306 nm) of supernatant before and after the reaction.
CPG resin possessing cyclooctyne unit (10). In the similar manner as describe for 9, CPG
17

resin (108 mmol/g, 100 mg) in CH3CN (1 mL) was treated with dibenzocyclooctyne-acid (6.6 mg,
21.6 mol)24, HBTU (12.3 L, 32.4 mol) and HOBt (0.3 mg, 21.6 mol) for 24 h to give CPG
resin 10. The loading amount of cyclooctyne unit was estimated to be 0.72 mmol/g based on the
comparison of UV absorption (at 306 nm) of supernatant before and after the reaction.
Synthesis of ON.
ON containing ‘catch & release’ tag at the 5’-terminus was synthesized on a DNA
synthesizer (H-6 DNA/RNA synthesizer) by using 2 and commercially available
deoxyribonucleoside phosphoramidites at 0.4 mol scale following the standard phosphoramidite
chemistry. The phosphoramidite 2 and natural units were dissolved in dry CH3CN (0.1 M and
0.065 M, respectively). The coupling reactions for these phosphoramidites were carried out for 720
s (for 2) and 20s (for natural units). CPG supporting ON was treated with saturated NaN3 in DMF
(500 L) at room temperature for 12 h. The CPG was filtrated, washed successively with CH3CN
and acetone, and dried. ON was released from CPG by treatment with 28% aqueous ammonia for
18 h at room temperature, and the resulting solution was concentrated.
General protocol of ON purification using catch & release strategy.
A solution of ON containing ‘catch & release’ tag at the 5’-terminus (500 L, 100 nmol)
was added to an Eppendorf tube (2 mL) containing resin 10 (40 mg, 290 nmol). The suspension in
the tube was gently shaken at 55 °C, and then, an aliquot of the reaction solution (40 L each) was
analyzed by using HPLC with a C-18 reversed-phase column for completion of the catch reaction.
After the filtration, the resin was washed with CH3CN, followed by H2O. Then, the resin was
suspended in a 20 mM Tris-HCl buffer (pH 8.2, 1.2 mL), and photo irradiation was performed
using Moritex MUV-202U of a 3000 mW/cm2 Hg-Xe lamp using a Pyrex filter (>300 nm) at room
temperature. An aliquot of the supernatant of the reaction mixture (40 L each) was analyzed by
using HPLC with a C-18 reversed-phase column for completion of the release reaction.
Acknowledgement
This work was supported in part by the Grant-in-Aid for Scientific Research (B) (Grant
18

Number; 15H04656) and the Sasagawa Scientific Research Grant from The Japan Science Society.
The authors would like to thank Prof. Dr. A. Otaka and Dr. A. Shigenaga (The Institute of
Biomedical Sciences, Tokushima University Graduate School) for providing a technical assistance
of photoirradiation experiments and Mr. S. Kitaike (Center for Instrumental Analysis, Tokushima
University) for providing technical assistance of instrumental analysis. Y.I. and N.T. thank the
research program for the development of intelligent Tokushima artificial exosome (iTEX) from
Tokushima University.
References
1. L. J. McBride and M. H. Caruthers, An investigation of several deoxynucleoside
phosphoramidites useful for synthesizing deoxyoligonucleotides, Tetrahedron Lett., 1983,
24, 245–248.
2. S. Roy and M. Caruthers, Synthesis of DNA/RNA and Their Analogs via Phosphoramidite
and H-Phosphonate Chemistries, Molecules, 2013, 18, 14268–14284.
3. J. Olejnik, E. Krzymanska-Olejnik and K. J. Rothschild, Photocleavable biotin
phosphoramidite for 5’-end-labeling, affinity purification and phosphorylation of synthetic
oligonucleotides, Nucleic Acids Res., 1996, 24, 361–366.
4. S. Fang and D. E. Bergstrom, Reversible Biotinylation Phosphoramidite for
5’-End-Labeling, Phosphorylation, and Affinity Purification of Synthetic Oligonucleotides
Bioconjugate Chem., 2003, 14, 80–85.
5. K. Tomaya, M. Takahashi, N. Minakawa and A. Matsuda, Convenient RNA Synthesis
Using a Phosphoramidite Possessing a Biotinylated Photocleavable Group, Organic Lett.,
2010, 12, 3836–3839.
6. S. Fang and S. Fueangfung, Scalable Synthetic Oligodeoxynucleotide Purification with
Use of a Catching by Polymerization, Washing, and Releasing Approach, Organic Lett.,
2010, 12, 3720–3723.
7. Y. Yuan, S. Fueangfung, X. Lin, D. Pokharel and S. Fang, Synthetic 5’-phosphorylated
oligodeoxynucleotide purification through catching full-length sequences by
polymerization, RSC Advances, 2012, 2, 2803–2808.
19

8. W. H. Pearson, D. A. Berry, P. Stoy, K.-Y. Jung and A. D. Sercel, Fluorous Affinity
Purification of Oligonucleotides, J. Org. Chem., 2005, 70, 7114–7122.
9. E. Agustin, P. N. Asare Okai, I. Khan, M. R. Miller, R. Wang, J. Sheng and M. Royzen, A
fast click–slow release strategy towards the HPLC-free synthesis of RNA, Chem. Commun.,
2016, 52, 1405–1408.
10. J. Szychowski, A. Mahdavi, J. J. L. Hodas, J. D. Bagert, J. T. Ngo, P. Landgraf, D. C.
Dieterich, E. M. Schuman and D. A. Tirrell, Cleavable Biotin Probes for Labeling of
Biomolecules via Azide−Alkyne Cycloaddition, J. Am. Chem. Soc., 2010, 132,
18351–18360.
11. T. Wada, A. Mochizuki, S. Higashiya, H. Tsuruoka, S.-I. Kawahara, M. Ishikawa and M.
Sekine, Synthesis and properties of 2-azidodeoxyadenosine and its incorporation into
oligodeoxynucleotides, Tetrahedron Lett., 2001, 42, 9215–9219.
12. Abu T. Khan, S. Ghosh and Lokman H. Choudhury, A Simple and Useful Synthetic
Protocol for Selective Deprotection of tert-Butyldimethylsilyl (TBS) Ethers, European J.
Org. Chem., 2004, 2004, 2198–2204.
13. S. S. van Berkel, A. J. Dirks, M. F. Debets, F. L. van Delft, J. J. L. M. Cornelissen, R. J. M.
Nolte and F. P. J. T. Rutjes, Metal-Free Triazole Formation as a Tool for Bioconjugation,
ChemBioChem, 2007, 8, 1504–1508.
14. N. J. Agard, J. A. Prescher and C. R. Bertozzi, A Strain-Promoted [3+2] Azide−Alkyne
Cycloaddition for Covalent Modification of Biomolecules in Living Systems, J. Am. Chem.
Soc., 2004, 126, 15046–15047.
15. X. Ning, J. Guo, M. A. Wolfert and G.-J. Boons, Visualizing Metabolically Labeled
Glycoconjugates of Living Cells by Copper-Free and Fast Huisgen Cycloadditions, Angew.
Chem. Int. Ed., 2008, 47, 2253–2255.
16. N. E. Mbua, J. Guo, M. A. Wolfert, R. Steet and G.-J. Boons, Strain-Promoted
Alkyne–Azide Cycloadditions (SPAAC) Reveal New Features of Glycoconjugate
Biosynthesis, ChemBioChem, 2011, 12, 1912–1921.
17. M. F. Debets, S. S. van Berkel, S. Schoffelen, F. P. J. T. Rutjes, J. C. M. van Hest and F. L.
van Delft, Aza-dibenzocyclooctynes for fast and efficient enzyme PEGylation via
20

copper-free (3+2) cycloaddition, Chem. Commun., 2010, 46, 97–99.
18. C. P. Ramil and Q. Lin, Bioorthogonal chemistry: strategies and recent developments,
Chem. Commun., 2013, 49, 11007–11022.
19. F. Saito, H. Noda and J. W. Bode, Critical Evaluation and Rate Constants of
Chemoselective Ligation Reactions for Stoichiometric Conjugations in Water, ACS Chem.
Biol., 2015, 10, 1026–1033.
20. Preparation of the DIBAC anchored on PS resin 9 is described in section of the Method.
21. Preparation of the DIBAC anchored on CPG resin 10 is described in section of the
Method.
22. K. L. Kiick, E. Saxon, D. A. Tirrell and C. R. Bertozzi, Incorporation of azides into
recombinant proteins for chemoselective modification by the Staudinger ligation, Proc.
Natl. Acad. Sci. USA, 2002, 99, 19–24.
23. K. M. Gattás-Asfura and C. L. Stabler, Chemoselective Cross-Linking and
Functionalization of Alginate via Staudinger Ligation, Biomacromolecules, 2009, 10,
3122–3129.
24. C. Liu, T. Li and N. L. Rosi, Strain-Promoted “Click” Modification of a Mesoporous
Metal–Organic Framework, J. Am. Chem. Soc., 2012, 134, 18886–18888.
21

Solid support to ‘catch’ ONs
O
5’-end
of ON
O
O
O
3’
O
N
H
P
O–
‘Catch’
O
CPG
N
N3
4
N
H
O2N
‘Release’
‘Catch and release’ tag
photocleavable azide linker