Versatile solid supports for oligonucleotide synthesis that incorporate urea bridge

Article abstract of DOI:10.1080/15257770.2011.582851

The universal solid support, USIII, representing a new and improved version of commercial USII, as well as 2'-deoxynucleoside and 2'-deoxy-2'- fluoronucleoside bound supports, incorporating a labile phenoxyacetyl fragment, was synthesized by an aminomethyl polystyrene carbamoylation with corresponding azides in the presence of aqueous triethylammonium bicarbonate. All three solid phases incorporate a stable urea tether, thus bridging the polymer and functional linker. These new matrices proved to be potent solid phases for the synthesis of DNA, RNA, or modified oligonucleotides as well as randomized mixed 2'-ribo/2'-deoxy-2'-fluoro-RNA libraries and/or DNA libraries, randomized with trinucleotides (codons). Copyright Taylor and Francis Group, LLC.

Full text of DOI:10.1080/15257770.2011.582851

This article was downloaded by: [Washington University in St Louis]
On: 24 June 2013, At: 21:35
Publisher: Taylor & Francis
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered
office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK
Nucleosides, Nucleotides and Nucleic
Acids
Publication details, including instructions for authors and
subscription information:
http://www.tandfonline.com/loi/lncn20
Versatile Solid Supports for
Oligonucleotide Synthesis that
Incorporate Urea Bridge
Mr Andrey Yagodkin a , Mr Janne Weisel b & Professor Alex Azhayev a
c
a Faculty of Health Sciences, University of Eastern Finland, Kuopio,
Finland
b Biocenter Kuopio, University of Eastern Finland, Kuopio, Finland
c Metkinen Chemistry, Kuopio, Finland
To cite this article: Mr Andrey Yagodkin , Mr Janne Weisel & Professor Alex Azhayev (2011): Versatile
Solid Supports for Oligonucleotide Synthesis that Incorporate Urea Bridge, Nucleosides, Nucleotides
and Nucleic Acids, 30:7-8, 475-489
To link to this article: http://dx.doi.org/10.1080/15257770.2011.582851
PLEASE SCROLL DOWN FOR ARTICLE
Full terms and conditions of use: http://www.tandfonline.com/page/terms-and-conditions
This article may be used for research, teaching, and private study purposes. Any
substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing,
systematic supply, or distribution in any form to anyone is expressly forbidden.
The publisher does not give any warranty express or implied or make any representation
that the contents will be complete or accurate or up to date. The accuracy of any
instructions, formulae, and drug doses should be independently verified with primary
sources. The publisher shall not be liable for any loss, actions, claims, proceedings,
demand, or costs or damages whatsoever or howsoever caused arising directly or
indirectly in connection with or arising out of the use of this material.

Nucleosides, Nucleotides and Nucleic Acids, 30:475–489, 2011
Copyright ꢀC Taylor and Francis Group, LLC
ISSN: 1525-7770 print / 1532-2335 online
DOI: 10.1080/15257770.2011.582851
VERSATILE SOLID SUPPORTS FOR OLIGONUCLEOTIDE SYNTHESIS
THAT INCORPORATE UREA BRIDGE
Andrey Yagodkin,1 Janne Weisel,2 and Alex Azhayev1,3
1Faculty of Health Sciences, University of Eastern Finland, Kuopio, Finland
2Biocenter Kuopio, University of Eastern Finland, Kuopio, Finland
3Metkinen Chemistry, Kuopio, Finland
The universal solid support, USIII, representing a new and improved version of commercial
2
USII, as well as 2 ꢁ-deoxynucleoside and 2 ꢁ-deoxy-2 ꢁ-fluoronucleoside bound supports, incorporat-
ing a labile phenoxyacetyl fragment, was synthesized by an aminomethyl polystyrene carbamoylation
with corresponding azides in the presence of aqueous triethylammonium bicarbonate. All three solid
phases incorporate a stable urea tether, thus bridging the polymer and functional linker. These new
matrices proved to be potent solid phases for the synthesis of DNA, RNA, or modified oligonucleotides
as well as randomized mixed 2 ꢁ-ribo/2 ꢁ-deoxy-2 ꢁ-fluoro-RNA libraries and/or DNA libraries, ran-
domized with trinucleotides (codons).
Keywords Universal solid support; carbamoylation; urea bridge; randomized oligonu-
cleotide libraries; 2ꢁ-deoxy-2ꢁ-fluoronucleosides; trinucleotide (codons)
INTRODUCTION
Various advantages of universal supports for oligonucleotide synthesis
are well documented.[1] However, one feature of universal supports that is
rarely mentioned but appears to be very important in the case of pharma-
ceutical oligonucleotides, is due to the fact that a universal linker does not
contribute an atom to an active pharmaceutical ingredient and hence need
not be classified as a starting material. This feature is unlike conventional
nucleoside bound supports, where the nucleoside linked to a solid phase
becomes 3ꢁ-terminal unit in a target oligomer.
Received 16 March 2011; accepted 18 April 2011.
We thank Dr. James Callaway for reading the manuscript. This work was financially sup-
ported by the Technology Development Centre of Finland (www.tekes.fi), Metkinen Chemistry (www.
metkinenchemistry.com), and HyTest (www.hytest.com). Alex Azhayev gratefully acknowledges the re-
ceipt of Academy of Finland (www.aka.fi) one-year grant “Mobility in working life.”
Address correspondence to Prof. Alex Azhayev, Faculty of Health Sciences, University of Eastern
Finland, Yliopistoranta 1C, FIN-70211, Kuopio, Finland. E-mail: Alex.Azhayev@uef.fi
475

476
A. Yagodkin et al.
Commercial universal solid supports may be divided into two groups, and
this division depends on the timing of the dephosphorylation step that gen-
erates the 3ꢁ-hydroxyl of the target oligonucleotide.[1] In the first group, the
cleavage step of oligonucleotide synthesis initially leaves the intact tether
attached to the 3ꢁ-nucleoside through a phosphodiester linkage. The de-
phosphorylation step and the subsequent elimination of this tether occur
during the deprotection step and usually require strong basic conditions
that are similar to deprotection of the target oligonucleotide. In the second
group, the dephosphorylation step is effectively the cleavage step, and the
only oligonucleotides released into the solution already have a 3ꢁ-hydroxyl
group. Subsequent conventional deprotection leads to the fully deprotected
oligonucleotide.[2–4]
The universal solid support for oligonucleotide synthesis, USII
(Figure 1), first reported in 2001,[5,6] clearly belongs to the second
group.[1,5,6] The key feature of this solid matrix is related to the mech-
anism of cleavage for the preassembled oligonucleotide, which is based
on the reaction of cyclic phosphodiester formation, thus resulting in 3ꢁ-
dephosphorylation.[5]
Another important feature of the USII is the cleavage procedure, which
must be performed with the 4 M anhydrous ammonia in methanol for about
30 minutes at room temperature (RT), rather than the conventional lengthy
aqueous ammonia treatment. Moreover, it was noted that the subsequent
removal of protective groups could be alternatively performed with the
methanolic ammonia solution at RT, instead of using the more aggressive
aqueous ammonia.
Although the oligonucleotide base deprotection that is performed un-
der such a mild condition is not actually required for the simple synthesis
FIGURE 1 Structures of universal supports USII and USIII.

Versatile Solid Supports for Oligonucleotide Synthesis
477
of regular DNA fragments, it appears beneficial for the synthesis of RNA.
Indeed, treatment with ammonia in methanol at RT for both cleavage and
deprotection brings about higher yields of RNA oligonucleotides, due to a
much smaller amount of reaction by-products, that result from unwanted
chain scission, which takes place when a conventional aqueous ammonium
treatment is employed.
RESULTS AND DISCUSSION
Although the USII (Figure 1) remains a convenient solid matrix for
oligonucleotide synthesis that has several advantages, it also has one major
drawback. In particular, assembly of the USII support linker is performed
by consecutive (four step) attachment of the structural fragments onto the
solid matrix after a three-step synthesis of aminoglycol moiety.[5] Obviously,
this situation allows for a more straightforward procedure that provides a
support with similar desirable characteristics as the USII.
Recently, we have developed a procedure for the carbamoylation of
nucleophilic compounds by employing organic azides and aqueous triethy-
lammonium bicarbonate (TEAB) as in situ sources of isocyanates.[7,8] In this
publication,[7] we have demonstrated that the carbamoylation reaction may
be conveniently used for the modification of aminoalkylated solid surfaces
by attaching various organic molecules through the urea bridge.
This observation brought forward an idea to utilize this approach for
the preparation of a new version of the universal solid support by the direct
attachment of the universal linker to aminomethyl polystyrene. Thus, we
performed a simple three-step synthesis of azido linker 1 (Scheme 1), start-
ing from inexpensive ( )-3-chloro-1,2-propanediol to achieve about 45%
overall yield, followed by a one-step conjugation of compound 4 and the ma-
trix (see the EXPERIMENTAL section). Finally, the unreacted aminomethyl
groups of the aminomethyl polystyrene were capped with dichloroacetyl
imidazole to give the USIII. It is worth noting that dichloroacetyl imidazole
was chosen as a capping reagent in order to block unreacted amino groups
of the polymer, in addition to dichloroacetylation of the small amount of
hydroxyl groups of the universal linker. Such deacylated hydroxyl groups
may result from partial hydrolysis of dichloroacetyl ester functions during
the carbamoylation reaction in the presence of aqueous TEAB.
The universal support USIII was tested in automated oligonucleotide
synthesis by using various protocols and proved efficient for the synthesis
of various oligomer classes that include mixed 2ꢁ-ribo/2ꢁ-deoxy-2ꢁ-fluoro-
RNA[9] (Figure 2) and randomized 2ꢁ-ribo/2ꢁ-deoxy-2ꢁ-fluoro-RNA libraries
(Figure 3). Results of these syntheses appeared to be more than satisfactory in
terms of yield. High performance liquid chromatography (HPLC) traces of
reaction mixtures, obtained after cleavage and deprotection, are presented
in Figures 2 and 3 as an illustrative material.

478
A. Yagodkin et al.
SCHEME 1 Synthesis of universal support USIII.
The USIII also demonstrated good results in the solid phase synthe-
sis of randomized DNA libraries, by employing trimer (codon) phospho-
ramidites. After the assembly of such libraries, randomized sequences were
cleaved off the universal phase, deprotected, and purified as reported ear-
lier.[10] An illustration of a 15% denaturing polyacrylamide gel electrophore-
sis (PAGE) of two synthesized randomized libraries (Figure 4) is pre-
sented as an example. These libraries were of the following sequences: L1:
5ꢁ-CTGGAGTGGGTCTCA-(XXX)10-CGATTCACCATCTCCAG-3ꢁ (62 nt)
and L2: 5ꢁ-GCCTTGTATTACTGTGCAAAA-(XXX)10-GGGGGCAAGGGAC-
3ꢁ (65 nt), where (XXX)10 represents 10 consecutive randomizations with
the following mixture of 19 trinucleotides [TAC 30%, TCT 15%, GGT 15%
(Gly), CGT 2.5%, GCT 2.5%, GAC 2.5%, AAC 2.5%, CAG 2.5%, GAA 2.5%,
CAT 2.5%, ATC 2.5%, CTG 2.5%, AAA 2.5%, ATG 2.5%, TTC 2.5%, CCG
2.5%, ACT 2.5%, TGG 2.5%, and GTT 2.5%].
All of these results allowed us to conclude that this new solid support
(USIII) performs at least as well as the USII alternative. The synthetic changes

Versatile Solid Supports for Oligonucleotide Synthesis
479
FIGURE 2 “Trityl-on” RP HPLC trace of crude 27-mer 5ꢁ-DMT-aucguuuucucaucgguucuucgccuc-3ꢁ
(where u and c are 2ꢁ-deoxy-2ꢁ-fluororibonucleoside units, whereas a and g are ribonucleoside units),
obtained after cleavage (4 M NH3/MeOH, RT, 30 minutes) and deprotection (4 M NH3/MeOH, RT, 36
hours; evaporation to dryness; 1 M tetrabutylammonium fluoride/ACN 1:1, pH = 7.5, 40◦C, 12 hours).
The target compound elutes at about 26 minutes.
in the universal support preparation were made to facilitate the manufac-
ture and quality control of the universal support. It should be stressed that
because the mode of action of both supports is based on the same mecha-
nism, mediated by similar universal linkers, the structural differences have
no effect on the performance of two phases and therefore no impact on the
oligonucleotide product.
FIGURE 3 “Trityl-on” RP HPLC trace of crude 69-mer library 5ꢁ-DMT-caaacauguuuuuucguccga
(n30)ugggucgaaagaacaugu-3ꢁ (where u and c are 2ꢁ-deoxy-2ꢁ-fluororibonucleoside units, whereas a and
g are ribonucleoside units; n30—30 consecutive randomizations at equimolar with u, c, a, g), obtained
after the cleavage and deprotection (conditions are the same as described in Figure 2). The target library
elutes at about 18 minutes.

480
A. Yagodkin et al.
FIGURE 4 Photograph of 15% denaturing PAGE of randomized oligonucleotide libraries L1 and L2.
Lane 1, L1; lane 2, L2; lane 3, control 50-mer oligonucleotide; and XC, xylene cyanol marker.
Satisfactory results with the universal solid support USIII brought about
the idea of applying our carbamoylation procedure,[7,8] in order to prepare
a series of novel nucleoside bound supports for oligonucleotide synthesis.
Obviously, these new supports, incorporating the stable urea bridge, should
be relatively simple to prepare on any scale, and they should perform in
oligonucleotide synthesis at least as well as in commercial nucleoside bound
phases having a succinoyl diester bridge. On the other hand, the postsyn-
thetic cleavage of nucleic acid fragments from these new solid phases should
occur under conventional conditions (i.e., concentrated aqueous ammonia,
at RT), but preferably several times faster than in the case of nucleoside
bound solid supports.
Figure 5 shows structures of a conventional 2ꢁ-deoxynucleoside bound
matrix dNS-Su, the newer 2ꢁ-deoxynucleoside bound phase dNS-Ur, and 2ꢁ-
fluororibonucleoside bound phase fNS-Ur. Preparation of all three supports
FIGURE 5 Structures of dNS-Su, dNS-Ur, and fNS-Ur.

Versatile Solid Supports for Oligonucleotide Synthesis
481
appears to be a simple procedure. In the case with dNS-Su,[11] commercial
protected 2ꢁ-deoxynucleosides are succinylated with succinic anhydride, the
resulting 3ꢁ-succinyl nucleosides are then attached to the aminoalkyl solid
phase by the appropriate peptide type condensation, and, finally, the unre-
acted aminoalkyl groups of the starting solid matrix are capped by acetyla-
tion. In the cases with dNS-Ur and fNS-Ur, the corresponding protected nu-
cleoside is acylated by 4-azidomethylphenoxyacetic acid (synthesized from
inexpensive 4-hydroxymethylphenoxyacetic acid in three simple steps in
>80% yield, Scheme 2). 3ꢁ-acylated nucleoside is then purified by flash chro-
matography and finally loaded onto the aminomethyl polystyrene via the
urea linker[7,8] in excellent yield (>90%).
SCHEME 2 Synthesis of 3ꢁ-O-[2-(4-azidomethylphenoxy)acetyl]-5ꢁ-O-(4ꢁ,4ꢁꢁ-dimethoxytriphenylmethyl)
nucleosides 5--12.
Our experiments have demonstrated that our matrices dNS-Ur and fNS-
Ur perform equally well in the oligonucleotide assembly on the synthesizer as
their commercial counterparts dNS-Su. At the same time, oligonucleotides
assembled on dNS-Ur are completely cleaved with concentrated aqueous
ammonia at RT within 30 minutes, which is about four times faster than
from conventional dNS-Su supports.
The test syntheses for a number of oligodeoxyrobonucleotides on dNS-
Ur supports, as well as various mixed 2ꢁ-ribo/2ꢁ-deoxy-2ꢁ-fluoro-RNA and 2ꢁ-
fluoro-RNA (all nucleoside units are 2ꢁ-deoxy-2ꢁ-fluoronucleosides) on fNS-
Ur matrices, have demonstrated the applicability of both solid phases for

482
A. Yagodkin et al.
FIGURE 6 IE HPLC trace of crude 20-mer 5ꢁ-guucccuacucucgcguuaa-3ꢁ, where a, c, g, and u are 2ꢁ-
deoxy-2ꢁ-fluororibonucleoside units. The target compound elutes at about 15 minutes.
oligonucleotide synthesis. The ion exchange (IE) HPLC trace of crude 20-
mer 2ꢁ-fluoro-RNA is given in Figure 6 as a typical example.
In our opinion, the use of nucleoside bound supports, incorporating
a stable urea linker along with a labile phenoxyacetyl fragment, appears
to be potent solid phases for the synthesis of DNA, RNA, or modified
oligonucleotides. These phases may be superior to conventional nucleo-
side bound supports, which incorporate a succinate bridge, when milder
cleavage/deprotection conditions may bring about better yield and quality
of the target oligomer.
CONCLUSION
To our understanding, all three types of supports—USIII, dNS-Ur, and
dNS-Ur—may prove useful for DNA, RNA, and modified RNA synthesis in
principle. However, we believe that at least some of these may be optimal for
preparing certain classes of oligonucleotides, such as oligomers that contain
RNA units or certain combinatorial libraries. Owing to their intrinsic
chemical properties, all three solid phases allow a substantially milder post-
synthetic cleavage of oligonucleotides, when compared with conventional
solid supports. This feature seems especially important when various side
reactions, associated with monomers having vulnerable structures, become
a determinant factor in the successful preparation of a pure oligonucleotide
or oligonucleotide library. We are convinced that quality and sufficient
amounts of the desired target oligomers, or combinatorial oligolibraries,
may finally determine the success or failure of the final goal of achievement,

Versatile Solid Supports for Oligonucleotide Synthesis
483
for example, in the case of synthesis of a high-quality 2ꢁ-F-RNA library for
aptamer research.
EXPERIMENTAL
General
All reagents and solvents for organic synthesis and chromatography were
purchased from Sigma-Aldrich (Helsinki, Finland) and used without further
purification. Protected 2ꢁ-deoxyribonucleosides and 2ꢁ-deoxyribonucleoside
CE phosphoramidites were from Rasayan, Inc. (Carlsbad, CA, USA).
Macroporous aminomethyl polystyrene containing 0.034 mmol/g of amino
group loading, ribonucleoside TOM-CE phosphoramidites, and reagents
(solvents, activators, etc.) for oligonucleotide synthesis were purchased
from Glen Research Corporation (Sterling, VA, USA). Macroporous
aminomethyl polystyrene containing 0.12 mmol/g of amino group load-
ing, protected 2ꢁ-deoxy-2ꢁ-fluororibonucleosides, 2ꢁ-fluororibonucleoside CE
phosphoramidites, and trinucleotide (codon) phosphoramidites were from
Metkinen Chemistry (Kuopio, Finland).
Column flash chromatography was run using two C-601 pump modules, a
C-615 pump controller and borosilicate glass columns (Bu¨chi, Flawil, Switzer-
land). Silica gel 60 (40–63 µm) or Lichroprep RP-18 (40–63 µm), both from
Merck (Darmstadt, Germany), was used for column flash chromatography.
A gradient of methanol in dichloromethane was used in a former case and
a gradient of acetonitrile (ACN) in water was used in a latter case for purifi-
cation of compounds.
Reversed phase (RP) HPLC of mixed 2ꢁ-ribo/2ꢁ-deoxy-2ꢁ-fluoro-RNA was
performed by employing a Waters (Milford, MA, USA) Breeze gradient chro-
matograph on a xBridge C18 column (4.6 × 150 mm, 5 µm, Waters), us-
ing a gradient elution from 0% to 40% ACN in 0.1 M triethylammonium
acetate buffer, pH 7.5, over 30 minutes at 260 nm and flow rate 1 mL/
minute.
RP HPLC of randomized oligonucleotides was run on a Gilson (Villiers-
le-Bel, France) gradient chromatograph on a xBridge C18 column (4.6 ×
150 mm, 5 µm, Waters), using a gradient elution from 0% to 30% ACN in
0.1 M triethylammonium acetate buffer, pH 7.5, over 13 minutes, followed
by a gradient elution from 30% to 70% acetonitrile in the same buffer over
1 minute, and finally isocratic elution with 70% ACN in the same buffer over
16 minutes at 260 nm and flow rate 1 mL/minute.
IE HPLC was performed on a Waters (Milford, MA, USA) Breeze gradient
chromatograph on a Dionex DNA Pac PA-100 column (4 × 250 mm; buffer A:
0.1 M sodium acetate in 20% ACN, pH 8; buffer B: 0.1 M sodium acetate and
0.4 M sodium perchlorate in 20% ACN, pH 8), using a gradient elution from
5% to 45% buffer B over 20 minutes at 260 nm and flow rate 1 mL/minute.

484
A. Yagodkin et al.
Oligonucleotide syntheses were performed with an ASM-800 8-column
synthesizer (Biosset, Novosibirsk, Russia) on a 0.1 µM scale or with an ABI
(Applied Biosystems, Carlsbad, CA, USA) 392 2-column synthesizer on a 0.2
µM scale.
NMR spectra were collected with an Avance 500 spectrometer (Bruker,
Silberstetten, Germany) in D2O, CDCl3, or CD3CN. High-resolution mass
spectra (HRMS) were obtained with a QSTAR XL instrument (Applied
Biosystems, Carlsbad, CA, USA), using electrospray ionization (ESI). Ex-
act masses of compounds were calculated using the built-in calculator of
Analyst QS software.
( )-1-Azido-3-(4ꢀ,4ꢀꢀ-Dimethoxytriphenylmethoxy)-Propyl-2-
Dichloroacetate (1)
A precooled (+4◦C) solution of 4ꢁ,4ꢁꢁ-dimethoxytriphenylmethyl chlo-
ride (250 g, 0.738 mol) in dichloromethane (1 L) was added dropwise within
24 hours to a precooled (+4◦C) solution of ( )-3-chloro-1,2-propanediol
(94 g, 0.85 mol) in a mixture of dichloromethane and pyridine (4:1, 0.75 L).
The reaction mixture was washed with saturated sodium hydrogen carbon-
ate solution (3 × 500 mL) and evaporated to give oil. The oil was dissolved
in a solution of toluene and n-hexane (2:1, 1.5 L) and washed with water
(5 × 500 mL). The organic solution was evaporated to dryness and coevap-
orated with ethanol (2 × 500 mL) to give crude compound 2 as yellow
viscous oil. The oil was dissolved in ethanol (1.4 L), lithium azide (81 g,
1.65 mol) was added to the solution, and the resulting mixture was heated
at reflux with stirring overnight. The mixture was cooled to RT and evapo-
rated to dryness. The oil was dissolved in a solution of toluene and n-hexane
(2:1, 1.5 L) and washed with water (5 × 500 mL). The organic solution was
dried over sodium sulfate and evaporated to dryness to give crude 1-azido-
3-(4ꢁ,4ꢁꢁ-dimethoxytriphenylmethoxy)propan-2-ol as yellow viscous oil. The
oil was dissolved in dry dichloromethane (1.0 L). N,Nꢁ-carbonyldiimidazole
(CDI; 267 g, 1.65 mol) was added portionwise to a solution of dichloroacetic
acid (193.5 g, 1.5 mol) in dichloromethane (1.0 L) with stirring. After 30
minutes at RT, the resulting mixture was added to a solution of 1-azido-
3-(4ꢁ,4ꢁꢁ-dimethoxytriphenylmethoxy)propan-2-ol in dichloromethane (1.0
L) and was then stirred for an additional 3 hours at RT. The reaction
was quenched by addition of water (100 mL); the dichloromethane solu-
tion was extracted with saturated sodium hydrogen carbonate solution (3
× 500 mL) and evaporated to give oil. The oil was dissolved in a solution
of toluene and n-hexane (2:1, 1.5 L) and washed with saturated sodium
hydrogen carbonate solution (2 × 500 mL) and water (500 mL). The or-
ganic solution was evaporated to dryness and coevaporated with diisopropy-
lether (2 × 500 mL) to give crude compound 1 as yellow viscous oil. Two
successive crystallization from diisopropyl ether gave pure compound 1 as

Versatile Solid Supports for Oligonucleotide Synthesis
485
white colorless crystals, m.p. 83◦C–85◦C. The yield of azide 1 was 201.6 g
1
(44.7%). IR (KBr, cm−1): 2117 (N3); H NMR (CD3CN): δ 3.28–3.29 (m,
2H), 3.59–3.60 (m, 2H), 3.79 (s, 6H), 5.22 (m, 1H), 6.35 (s, 1H), 6.88–6.90
(m, 4H), 7.24–7.46 (m, 9H); 13C NMR (CD3CN): 51.10, 55.35, 62.38, 65.16,
71.95, 75.64, 86.63, 113.56, 127.36, 128.31, 128.35, 130.38, 130.39, 135.94,
135.95, 145.14, 159.21, 164.54; MS (ESI-MS, ACN in 50 mM NH4HCO3):
Calcd. for C26H25Cl2N3O5Na [M+Na]+ 552.1069; found 552.1071.
2-(4-Cloromethylphenoxy)Acetic Acid Methyl Ester (2)
Thionyl chloride (81.9 g, 50 mL, 688 mmol) was added to a suspension
of 4-hydroxymethylphenoxyacetic acid (36.5 g, 0.2 mol) in dichloromethane
(150 mL); the reaction mixture was heated at reflux for 5 hours, cooled down
to +4◦C, and diluted to a precooled (+5◦C) mixture of dichloromethane
and methanol (1:1, 100 mL). The resulting mixture was left for 24 hours at
RT and evaporated to dryness. The residue was dissolved in ether (250 mL),
washed with saturated sodium hydrogen carbonate solution (2 × 100 mL),
and evaporated to give yellow oil. The oily residue was distilled under re-
duced pressure (1 mmHg) and the fraction boiling at 108◦C–111◦C was
collected, which solidified spontaneously on standing to a crystalline mass,
1
m.p. 51◦C–54◦C. The yield of compound 2 was 38.2 g (89.3%). H NMR
(CDCl3): δ 3.81 (s, 3H), 4.56 (s, 2H), 4.64 (s, 2H), 6.88 (d, 2H, J = 8.7 Hz),
7.32 (d, 2H, J = 8.7 Hz); 13C NMR (CDCl3): 52.66, 54.65, 65.74, 115.36,
129.12, 130.17, 158.22, 169.60.
2-(4-Azidomethylphenoxy)Acetic Acid Methyl Ester (3)
Sodium azide (22.8 g, 0.351 mmol) was added to a solution of com-
pound 2 (33.7 g, 0.176 mol) in N,N -dimethylformamide (200 mL) and the
mixture was vigorously stirred for 36 hours at RT and finally evaporated to
dryness. Ether (300 mL) and water (300 mL) were added to the residue and
vigorously shaken. The ether layer was separated, dried over sodium sulfate,
and evaporated to give a crude azide 3 as yellow oil. The oily residue was
distilled under reduced pressure (12 mmHg) and the fraction boiling at
155◦C–158◦C was collected. The yield of pure azide 3 as a colorless oil was
1
31.0 g (79.7%). IR (νmax cm−1, KBr) 2115 (N3); H NMR (CDCl3): δ 3.81
(s, 3H), 4.27 (s, 2H), 4.65 (s, 2H), 6.91 (d, 2H, J = 8.6 Hz), 7.25 (d, 2H,
J = 8.6 Hz); 13C NMR (CDCl3): 46.38, 52.69, 65.73, 115.27, 130.55, 131.32,
158.25, 169.59.
Sodium 2-(4-Azidomethylphenoxy)Acetate (4)
Methyl ester 3 (30 g, 0.136 mol) was dissolved in methanol (85 mL).
A solution of sodium hydroxide (8.45 g, 0.211 mol) in water (35 mL) was
added to it. The mixture was stirred for 10 minutes; crystals of 7 were filtered
off, washed with ethanol (3 × 50 mL), and dried in vacuum to give 23.7 g
(76.1%) of salt 4 as colorless crystals, m.p. >250◦C. IR (νmax cm−1, KBr) 2115

486
A. Yagodkin et al.
1
(N3); H NMR (D2O): δ 4.23 (s, 2H), 4.38 (s, 2H), 6.86 (d, 2H, J = 8.7
Hz), 7.23 (d, 2H, J = 8.7 Hz); 13C NMR (D2O): 54.21, 67.22, 115.17, 128.77,
130.55, 158.16, 177.05; MS (ESI-MS, 80% ACN in 50 mM NH4HCO3): Calcd.
for C9H8N3O3 [M–Na]– 206.0565; found 206.0561.
3ꢀ-O-[2-(4-Azidomethylphenoxy)Acetyl]-5ꢀ-O-(4ꢀ,4ꢀꢀ-Dimethoxytriphenylmethyl)-
2ꢀ-Deoxynucleosides (5–8) and 3ꢀ-O-[2-(4-Azidomethylphenoxy)Acetyl]-5ꢀ-O-
(4ꢀ,4ꢀꢀ-Dimethoxytriphenylmethyl)-2ꢀ-Deoxy-2ꢀ-Fluoronucleosides (9--12)
Compound 4 (2.75 g, 12 mmol) was added to a solution of 5ꢁ-O-(4ꢁ,4ꢁꢁ-
dimethoxytriphenylmethyl)-2ꢁ-deoxy-2ꢁ-fluoronucleoside (10 mmol) in ACN
(60 mL) and pyridine (30 mL). A solution of 1-(2-mesitylenesulfonyl)-3-
nitro-1H-1,2,4-triazole (MSNT, 4.4 g, 15 mmol) in ACN (30 mL) was added
dropwise to the resulting suspension. After stirring the reaction mixture
for 2 hours at RT, solids were removed by filtration and the clear fil-
trate thus obtained was evaporated to dryness. The residue was dissolved
in dichloromethane (50 mL) and the organic solution was washed with
saturated aqueous sodium hydrogen carbonate, dried over sodium sulfate,
and evaporated to dryness. The desired protected 2ꢁ-deoxynucleosides 5–8
and 2ꢁ-deoxy-2ꢁ-fluoronucleosides 9–12 were then isolated using flash chro-
matography on silica gel to give pure compounds in 68%–74% yield as yellow
foams.
3ꢀ-O-[2-(4-Azidomethylphenoxy)Acetyl]-5ꢀ-O-(4ꢀ,4ꢀꢀ-Dimethoxytriphenylmethyl)-
N6-Benzoyl-2ꢀ-Deoxyadenosine (5)
1
IR (νmax cm−1, KBr) 2110 (N3); H NMR (CD3CN): δ 2.60–2.65 (m,
1H), 3.19–3.22 (m, 1H), 3.33–3.40 (m, 2H), 3.73 (ds, 6H), 3.73–3.75 (m,
1H), 4.29–4.31 (m, 1H), 4.32 (s, 2H), 4.77 (s, 2H), 5.63–5.65 (m, 1H),
6.42 (t, 1H, J = 7.0 Hz), 6.77–6.80 (m, 4H), 6.96–6.97 (m, 2H), 7.17–7.31
(m, 10H), 7.39–7.40 (m, 2H), 7.50–7.53 (m, 2H), 7.60–7.62 (m, 1H), 7.99
(d, 2H, J = 7.2 Hz), 8.23 (s, 1H), 8.55 (br.s, 1H), 9.48 (br.s, 1H); 13C
NMR (CD3CN): 36.60, 54.05, 55.32, 64.18, 65.42, 76.10, 84.40, 85.05, 86.67,
113.44, 115.21, 125.20, 127.27, 128.21, 128.37, 128.51, 129.06, 129.34, 130.33,
130.41, 132.96, 136.15, 136.17, 142.98, 145.36, 150.31, 152.24, 158.22, 159.06,
159.09, 165.63, 168.83; MS (ESI-MS, 80% ACN in 50 mM NH4HCO3): Calcd.
for C47H42FN8O8 [M+H]+ 847.3204; found 847.3125.
3ꢀ-O-[2-(4-Azidomethylphenoxy)Acetyl]-5ꢀ-O-(4ꢀ,4ꢀꢀ-Dimethoxytriphenylmethyl)-
N4-Benzoyl-2ꢀ-Deoxycytidine (6)
IR (νmax cm−1, KBr) 2115 (N3); 1H NMR (CD3CN): δ 2.36–2.42 (m, 1H),
2.66–2.70 (m, 1H), 3.38–3.44 (m, 2H), 3.76 (s, 3H), 3.77 (s, 3H), 4.27–4.28
(m, 1H), 4.32 (s, 2H), 4.72 (d, 1H, J Ha,Hb = 16.3 Hz), 4.76 (d, 1H, J Ha,Hb
= 16.5 Hz), 5.44–5.46 (m, 1H), 6.16 (t, 1H, J = 6.6 Hz), 6.87–6.96 (m,
6H), 7.20–7.34 (m, 11H), 7.43 (m, 2H), 7.51–7.54 (m, 2H), 7.62–7.65 (m,

Versatile Solid Supports for Oligonucleotide Synthesis
487
1H), 7.96 (d, 2H, J = 7.4 Hz), 8.10 (d, 1H, J = 7.5 Hz), 9.28 (s.br, 1H);
13C NMR (CD3CN): 38.90, 54.04, 55.34, 63.54, 65.38, 75.52, 84.65, 87.19,
87.52, 113.61, 115.19, 127.42, 128.38, 128.42, 128.49, 129.04, 129.33, 130.40,
130.43, 130.48, 133.28, 135.90, 136.00, 144.88, 145.07, 158.21, 159.21, 159.22,
163.17, 168.78; MS (ESI-MS, 80% ACN in 50 mM NH4HCO3): Calcd. for
C46H43N6O9 823.3092; found 823.3092.
3ꢀ-O-[2-(4-Azidomethylphenoxy)Acetyl]-5ꢀ-O-(4ꢀ,4ꢀꢀ-Dimethoxytriphenylmethyl)-
N2-Isobutyryl-2ꢀ-Deoxyguanine (7)
IR (νmax cm−1, KBr) 2110 (N3); 1H NMR (CD3CN): δ 1.16 (d, 6H, J = 6.8
Hz), 2.53–2.58 (m, 1H), 2.60–2.65 (m, 1H), 3.13–3.19 (m, 1H), 3.26–3.28
(m, 1H), 3.49–3.52 (m, 1H), 3.74 (ds, 6H), 4.27–4.29 (m, 1H), 4.31 (s, 2H),
4.73 (s, 2H), 5.48–5.50 (m, 1H), 6.23 (t, 1H, J = 7.0 Hz), 6.74–6.79 (m,
4H), 6.94 (d, 2H, J = 6.6 Hz), 7.19–7.38 (m, 12H), 7.83 (s, 1H); 13C NMR
(CD3CN): 18.19, 18.31, 35.65, 36.08, 53.59, 54.85, 54.88, 64.38, 64.92, 75.91,
84.47, 84.73, 86.09, 112.89, 112.91, 114.76, 115.00, 121.68, 126.84, 127.72,
128.00, 128.23, 128.94, 129.90, 130.01, 130.02, 135.63, 135.77, 138.18, 144.95,
147.76, 148.33, 155.30, 157.76, 158.60, 158.66, 168.46, 179.84; MS (ESI-MS,
80% ACN in 50 mM NH4HCO3): Calcd. for C44H44N8O9 847.3215; found
847.3181.
3ꢀ-O-[2-(4-Azidomethylphenoxy)Acetyl]-5ꢀ-O-(4ꢀ,4ꢀꢀ-
Dimethoxytriphenylmethyl)Thymidine (8)
IR (νmax cm−1, KBr) 2113 (N3); 1H NMR (CD3CN): δ 1.47 (d, 3H, J = 0.9
Hz), 2.40–2.50 (m, 2H), 3.34–3.42 (m, 2H), 3.75–3.79 (m, 7H), 4.13–4.14
(m, 1H), 4.31 (s, 2H), 4.72 (d, 1H, J Ha,Hb = 16.4 Hz), 4.75 (d, 1H, J Ha,Hb
=
=
ꢁ
ꢁ
ꢁ
ꢁ
16.4 Hz), 5.50–5.51 (m, 1H), 6.28 (dd, 1H, J H-1 ,H-2 a = 5.8 Hz, J H-1 , H-2 b
6.0 Hz), 6.88–6.95 (m, 6H), 7.21–7.34 (m, 11H), 7.45 (d, 2H, J = 7.4 Hz),
7.48 (d, 1H, J = 0.8 Hz), 9.25 (br.s, 1H); 13C NMR (CD3CN): 11.21, 36.93,
53.61, 54.95, 63.63, 64.92, 75.70, 83.42, 84.22, 86.71, 110.70, 113.19, 114.75,
127.06, 127.98, 128.03, 128.23, 128.91, 129.99, 130.06, 135.46, 135.50, 135.66,
144.78, 150.50, 157.78, 158.84, 163.62, 168.45; MS (ESI-MS, 80% ACN in 50
mM NH4HCO3): Calcd. for C39H38N5O9 [M–H]– 732.2670; found 732.2640.
3ꢀ-O-[2-(4-Azidomethylphenoxy)Acetyl]-5ꢀ-O-(4ꢀ,4ꢀꢀ-Dimethoxytriphenylmethyl)-
N6-Benzoyl-2ꢀ-Deoxy-2ꢀ-Fluoroadenosine (9)
IR (νmax cm−1, KBr) 2121 (N3); 1H NMR (CD3CN): δ 3.35–3.38 (m, 1H),
3.48–3.48 (m, 1H), 3.73 (ds, 6H), 4.32 (s, 2H), 4.38–4.41 (s, 1H), 4.79 (d, 1H,
J Ha,Hb = 16.8 Hz), 4.80 (d, 1H, J Ha,Hb = 16.8 Hz), 5.95–6.05 (m, 2H), 6.34
(dd, 1H, J H,F = 19.6 Hz, J H,H = 2.2 Hz), 6.77–6.80 (m, 4H), 6.93–6.94 (m,
2H), 7.18–7.28 (m, 9H), 7.37–7.39 (m, 2H), 7.52–7.55 (m, 2H), 7.62–7.65 (m,
1H), 8.00 (d, 2H, J = 7.4 Hz), 8.29 (s, 1H), 8.59 (s, 1H), 9.43 (s, 1H); 13C NMR
(CD3CN): 53.58, 54.89, 62.15, 64.64, 70.91, 71.03, 80.22, 86.26, 87.47, 87.74,

488
A. Yagodkin et al.
90.46, 91.97, 113.01, 114.75, 124.57, 125.27, 127.82, 127.91, 128.20, 128.67,
129.02, 129.92, 129.97, 130.00, 132.64, 133.76, 135.58, 135.67, 143.04, 144.83,
150.14, 151.53, 152.09, 157.65, 158.65, 165.28, 168.02; MS (ESI-MS, 80% ACN
in 50 mM NH4HCO3): Calcd. for C47H41FN8O8 [M+H]+ 865.3110; found
865. 3144.
3ꢀ-O-[2-(4-Azidomethylphenoxy)Acetyl]-5ꢀ-O-(4ꢀ,4ꢀꢀ-Dimethoxytriphenylmethyl)-
N4-Acetyl-2ꢀ-Deoxy-2ꢀ-Fluorcytidine (10)
1
IR (νmax cm−1, KBr) 2111 (N3); H NMR (CD3CN): δ 2.15 (s, 3H),
3.38–3.56 (m, 2H), 3.76 (s, 6H), 4.31 (s, 2H), 4.32–4.34 (m, 2H), 4.73 (d, 1H,
J Ha,Hb = 16.8 Hz), 4.77 (d, 1H, J Ha,Hb = 16.8 Hz), 5.34–5.51 (m, 2H), 5.93 (d,
1H, J H,F = 19.4 Hz), 6.86–6.92 (m, 6H), 7.15 (d, 1H, J = 7.5 Hz), 7.17–7.45
(m, 12H), 8.17 (d, 1H, J = 7.5 Hz), 9.04 (s, 1H); 13C NMR (CD3CN): 24.04,
53.57, 54.51, 61.19, 64.62, 69.68, 69.81, 79.91, 90.98, 91.27, 92.49, 92.92,
113.16, 114.75, 127.05, 127.99, 128.04, 128.24, 128.93, 129.99, 130.05, 135.50,
144.57, 145.55, 154.67, 157.62, 158.79, 163.05, 167.86, 170.72; MS (ESI-MS,
80% ACN in 50 mM NH4HCO3): Calcd. for C41H40FN6O9 779.2841; found
779.2817.
3ꢀ-O-[2-(4-Azidomethylphenoxy)Acetyl]-5ꢀ-O-(4ꢀ,4ꢀꢀ-Dimethoxytriphenylmethyl)-
N2-Benzoyl-2ꢀ-Deoxy-2ꢀ-Fluoroguanine (11)
1
IR (νmax cm−1, KBr) 2097 (N3); H NMR (CD3CN): δ 1.13 (d, 3H, J =
6.9 Hz), 1.15 (d, 3H, J = 6.9 Hz), 2.59 (m, 1H, J = 7.0 Hz), 3.40–3.43 (m,
1H), 3.47–3.50 (m, 1H), 3.74 (ds, 6H), 4.31 (s, 2H), 4.39–4.41 (m, 1H), 4.77
(d, 1H, J Ha,Hb = 17.0 Hz), 4.78 (d, 1H, J Ha,Hb = 17.0 Hz), 5.74–5.90 (m, 2H),
6.16 (dd, 1H, J H,F = 19.1 Hz, J H,H = 2.2 Hz), 6.75–6.94 (m, 6H), 7.17–7.39
(m, 13H), 7.88 (s, 1H), 9.24 (s, 1H), 12.00 (s, 1H); 13C NMR (CD3CN):
24.04, 53.57, 54.91, 61.19, 64.62, 69.67, 69.81, 79.91, 90.99, 91.27, 92.49,
95.91, 113.17, 114.75, 125.27, 127.04, 127.99, 128.04, 128.24, 128.94, 129.99,
130.05, 135.50, 144.57, 145.55, 154.67, 157.63, 158.79, 163.05, 167.86, 170.71;
MS (ESI-MS, 80% ACN in 50 mM NH4HCO3): Calcd. for C44H44FN8O9
[M+H]+ 847.3215; found 847.3181.
3ꢀ-O-[2-(4-Azidomethylphenoxy)Acetyl]-5ꢀ-O-(4ꢀ,4ꢀꢀ-Dimethoxytriphenylmethyl)-
2ꢀ-Deoxy-2ꢀ-Fluorouridine (12)
IR (νmax cm−1, KBr) 2105 (N3); 1H NMR (CD3CN): δ 3.34–3.37 (m, 1H),
3.45–3.48 (m, 1H), 3.77 (ds, 6H), 4.26 (m, 1H), 4.31 (s, 2H), 4.75 (d, 1H,
J Ha,Hb = 16.8 Hz), 4.77 (d, 1H, J Ha,Hb = 16.8 Hz), 5.37–5.51 (m, 3H), 5.85
(dd, 1H, J H,F = 19.9 Hz, J H,H = 1.8 Hz), 6.87–6.92 (m, 6H), 7.24–7.34 (m,
9.6H), 7.43–7.44 (m, 2H), 7.60 (d, 2H, J = 8.1 Hz); 13C NMR (CD3CN):
53.54, 54.92, 61.74, 64.61, 70.13, 70.26, 79.71, 86.61, 90.02, 90.30, 90.54,
92.05, 102.04, 113.15, 114.73, 127.01, 127.98, 129.02, 130.00, 130.05, 130.06,
135.48, 141.45, 144.74, 150.14, 157.62, 158.77, 162.95, 167.96; MS (ESI-MS,

Versatile Solid Supports for Oligonucleotide Synthesis
489
80% ACN in 50 mM NH4HCO3): Calcd. for C39H35FN5O9 [M–H]– 736.2419;
found 736.2402.
Universal Support USIII, 2ꢀ-Deoxynucleoside Bound Supports dNS-Ur, or
2ꢀ-Deoxy-2ꢀ-Fluoronucleoside Bound Supports fNS-Ur
A mixture of aminomethyl polystyrene (20 g, either containing 0.034
or 0.120 mmol/g of amino groups), dioxane (200 mL), freshly prepared 2
M TEAB buffer, pH 7.5–8.0 (4.8 mL), azide 1 or 5–8 or 9–12 (0.7 mmol),
and triphenylphosphine (3.2 g, 12 mmol) was quickly added to the reaction
mixture. The reaction flask was immediately tightly closed and vigorously
shaken for 48 hours. The resin was filtered, thoroughly washed with diox-
ane, and finally dried in vacuo. N,Nꢁ-CDI (7.3 g, 45 mmol) was added in
several portions to a solution of dichloroacetic acid (2.5 mL, 30 mmol) in
100 mL ACN. After evolution of carbon dioxide is complete, the mixture was
stirred for additional 30 minutes at RT, added to a dry resin, and the result-
ing mixture was shaken overnight at RT. The resin was filtered, thoroughly
washed with dioxane, and finally dried in vacuo. Solid supports for oligonu-
cleotide synthesis (USIII or dNS-Ur or fNS-Ur) contained 0.029–0.032 mmol
of dimethoxytrityl (DMT) groups per gram of resin.[11]
REFERENCES
1. Azhayev, A.; Antopolsky, M.; Tennila¨, T.; Mackie, H.; Randolph, J. A comparative study of commer-
cially available universal supports. Genet. Eng. News 2005, 25, 54–57.
2. Scott, S.; Hardy, P.; Sheppard, P.C.; McLean, M. A universal support for oligonucleotide synthesis, in
Innovations and Perspectives in Solid Phase Synthesis, 3rd International Symposium, ed. R. Epton, Mayflower
Worldwide Ltd., Birmingham, England, 1994, pp. 115–124.
3. Lyttle, M.H.; Hudson, D.; Cook, R.M. A new universal linker for solid phase DNA synthesis. Nucleic
Acids Res. 1996, 24, 2793–2798.
4. Azhayev, A. New universal solid support for oligonucleotide synthesis. Tetrahedron 1999, 55, 787–800.
5. Azhayev, A.; Antopolsky, M. Amide-group assisted 3ꢁ-deposphorylation of oligonucleotides synthe-
sized on universal A-supports. Tetrahedron 2001, 57, 4977–4986.
6. Azhayev, A.; Antopolsky, M. Universal supports for oligonucleotide synthesis. US Patent Serial No.
6770754 and European Patent No. 1404695.
7. Yagodkin, A.; Lo¨schcke, K.; Weisell, J.; Azhayev, A. Straightforward carbamoylation of nucleophilic
compounds employing organic azides, phosphines and aqueous trialkylammonium hydrogen car-
bonate. Tetrahedron 2010, 66, 2210–2221.
8. Yagodkin, A.; Azhayev, A. Carbamoylation of amines, thiophenols, mercaptanes and phenols employ-
ing organic azides. US Patent Application Serial No. 60/854,721 and International Patent Application
No. PCT/FI2007/050575.
9. Muchonen, P.; Tennila¨, T.; Azhayeva, E.; Parthasarathy, R.; Janckila, A.; Va¨a¨na¨nen, K.; Azhayev, A.;
Laitala-Leinonen, T. RNA interference tolerates 2ꢁ-fluoro modifications at the argonaute 2 cleavage
site. Chem Biodivers 2007, 4, 858–873.
10. Yagodkin, A.; Azhayev, A.; Roivainen, J.; Antopolsky, M.; Kayushin, A.; Korosteleva, M.; Miroshnikov,
A.; Randolph, J.; Mackie, H. Improved synthesis of trinucleotide phosphoramidites and genera-
tion of randomized oligonucleotide libraries. Nucleosides, Nucleotides Nucleic Acids 2007, 26, 473–
497.
11. Atkinson, T.; Smith, M. Phosphite-triester method, in Oligonucleotide Synthesis. A Practical Approach,
ed. M.J. Gait, IRL Press, Oxford, 1984, pp. 35–81.