Conjugation of nucleosides and oligonucleotides by [3+2] cycloaddition

Article abstract of DOI:10.1021/jo702023s

(Chemical Equation Presented) A procedure is presented for copper(I)-catalyzed [3+2] cycloaddition of nucleosides and nucleotides in near-quantitative yield. Azido-alkyne cycloaddition was applied for the preparation of a range of adenosine dimers and der

Full text of DOI:10.1021/jo702023s

hydrates lead to recognition by receptors on the surface of
designated cells.7 Consequently, a large number of antisense
conjugates have been prepared over the years, mostly involving
5′-conjugation due to the fact that 3′-modification is much less
straightforward. The alternative conjugation via a nucleobase
is more synthetically accessible but may interfere with Watson-
Crick base pairing and thus hybridization. Finally, a variety of
2′-substituted AONs has been described, in particular zwitte-
rionic derivatives, but no 2′-AON conjugates.
Conjugation of Nucleosides and Oligonucleotides
by [3+2] Cycloaddition
Anup M. Jawalekar,† Nico Meeuwenoord,‡
J. (Sjef) G. O. Cremers,† Herman S. Overkleeft,‡
Gijs A. van der Marel,‡ Floris P. J. T. Rutjes,† and
Floris L. van Delft*,†
Institute for Molecules and Materials, Radboud UniVersity
Nijmegen, ToernooiVeld 1, 6525 ED Nijmegen, The Netherlands,
and Leiden Institute of Chemistry, Bioorganic Chemistry,
Gorlaeus Laboratories, P.O. Box 9502,
Apart from antisense application, oligonucleotide conjugation
is also suitable for helical structure stabilization, e.g., with
intercalators or minor groove binders, or covalent attachment
of reporter groups, e.g., fluorescent, electrochemical, or spin
labels.8
2300 RA, Leiden, The Netherlands
f.Vandelft@science.ru.nl
We here wish to report the synthesis of 2′-azide or 2′-
acetylene modified adenosines as versatile building blocks for
application in the mild and efficient synthesis of a variety of
oligonucleotide hetero- and homoconjugates.
ReceiVed September 20, 2007
From the onset, a suitable technology for nucleotide conjuga-
tion appeared to be the Cu(I)-catalyzed Huisgen [3+2] cycload-
dition,9 a technique mild enough for bioconjugation of virus
capsid proteins,10 cell-surface labeling of E. coli,11 and protein
based profiling in living cells.12 On the basis of acetylene-
azide cycloaddition, triazole isosteres have been prepared of
A procedure is presented for copper(I)-catalyzed [3+2]
cycloaddition of nucleosides and nucleotides in near-
quantitative yield. Azido-alkyne cycloaddition was applied
for the preparation of a range of adenosine dimers and
derivatives with versatile functionality, as well as for the
smooth condensation of two oligonucleotide strands. The
described technology may find valuable application in the
synthesis of oligonucleotide dimers and conjugates.
(6) Shi, F.; Hoekstra, D. J. Controlled Release 2004, 97, 189-209.
(7) Manoharan, M. Antisense Nucleic Acid Drug DeV. 2002, 12, 103-
128.
(8) Zatepsin, T. S.; Stetsenko, D. A.; Gait, M. J.; Oretskaya, T. S. Bioconj.
Chem. 2005, 16, 471-489.
(9) (a) Tornøe, C. W.; Christensen, C.; Meldal, M. J. Org. Chem. 2002,
67, 3057-3064. (b) Rostovtsev, V. V.; Green, L. G.; Fokin, V. V.; Sharpless,
K. B. Angew. Chem., Int. Ed. 2002, 41, 2596-2599.
(10) Wang, Q.; Chan, T. R.; Hilgraf, R.; Fokin, V. V.; Sharpless, K. B.;
Finn, M. G. J. Am. Chem. Soc. 2003, 125, 3192-3193.
(11) Link, A. J.; Tirrell, D. A. J. Am. Chem. Soc. 2003, 125, 11164-
11165.
Synthetic oligonucleotides have proven to be highly valuable
as antisense structures for the selective silencing of specific
genes.1,2 However, clinical use of oligonucleotides is still limited
to a single example, i.e., fomivirsen (Vitravene),3 due to
drawbacks regarding potency, delivery, and duration of action.4
Second-generation involving 2′-O-modification, morpholino, or
locked nucleic acid (LNA)-based structures4 shows promise but
a remaining hurdle involves the poor cell internalization (∼1-
2%) of oligonucleotides.5 A promising solution to the inefficient
delivery of oligonucleotides lies in conjugation, e.g., with
cholesterol or polycationic groups for improvement of cellular
association.6 Alternatively, conjugates with peptides or carbo-
(12) Speers, A. E.; Adam, G. C.; Cravatt, B. F. J. Am. Chem. Soc. 2003,
125, 4686-4687.
(13) Bock, V. D.; Speijer, D.; Hiemstra, H.; van Maarseveen, J. H. Org.
Biomol. Chem. 2007, 5, 971-975 and references cited herein.
(14) Wilkinson, B. L.; Bornaghi, L. F.; Poulsen, S. A.; Houston, T. A.
Tetrahedron 2006, 62, 8115-8125.
(15) (a) Kuijpers, B. H. M.; Groothuys, S.; Keereweer, A. R.; Quaedflieg,
P. J. L. M.; Blaauw, R. H.; van Delft, F. L.; Rutjes, F. P. J. T. Org. Lett.
2004, 6, 3123-3126. (b) Dondoni, A.; Giovannini, P. P.; Massi, A. Org.
Lett. 2004, 6, 2929-2932. (c) Lin, H.; Walsh, C. T. J. Am. Chem. Soc.
2004, 126, 13998-14003.
(16) (a) Lazrek, H. B.; Engels, J. W.; Pfleiderer, W. Nucleosides
Nucleotides 1998, 17, 1851-1856. (b) Park, S. M.; Lee, Y. S.; Kim, B. H.
Chem. Commun. 2003, 2912-2913. (c) Gierlich, J. G.; Burley, G. A.;
Gramlich, P. M. E.; Hammond, D. M.; Carell, T. Org. Lett. 2006, 8, 3639-
3642. (d) Burley, G. A.; Gierlich, J.; Mofid, M. R.; Nir, H.; Tal, S.; Eichen,
Y.; Carell, T. J. Am. Chem. Soc. 2006, 128, 1398-1399. (e) Seela, F.;
Sirivolu, V. R. Chem. BiodiV. 2006, 3, 509-514. (f) Kosiova, I.; Kovackova,
S.; Kois, P. Tetrahedron 2007, 63, 312-320. (g) Seela, F.; Sirivolu, V. R.
HelV. Chim. Acta 2007, 90, 535-552.
(17) (a) Seo, T. S.; Li, Z.; Ruparel, H.; Ju, J. J. Org. Chem. 2003, 68,
609-612. (b) Lee, L. V.; Mitchell, M. L.; Huang, S.-J.; Fokin, V. V.;
Sharpless, K. B.; Wong, C.-H. J. Am. Chem. Soc. 2003, 125, 9588-9589.
(c) Weller, R. L.; Rajski, S. R. Org. Lett. 2005, 7, 2141-2144. (d) Devaraj,
N. K.; Miller, G. P.; Ebina, W.; Kakaradov, B.; Collman, J. P.; Kool, E.
T.; Chidsey, C. E. D. J. Am. Chem. Soc. 2005, 127, 8600-8601. (e)
Bouillon, C.; Meyer, A.; Vidal, S.; Jochum, A.; Chevolot, Y.; Cloarec, J.-
P.; Praly, J.-P.; Vasseur, J.-J.; Morvan, F. J. Org. Chem. 2006, 71, 4700-
4702. (f) Meyer, A.; Bouillon, C.; Vidal, S.; Vasseur, J.-J.; Morvan, F.
Tetrahedron Lett. 2006, 47, 8867-8871. (g) Kosiova, I.; Kovackova, S.;
Kois, P. Tetrahedron 2007, 63, 312-320.
† Radboud University Nijmegen.
‡ Leiden Insitute of Chemistry.
(1) (a) Zamecnik, P. C.; Stephenson, M. L. Proc. Natl. Acad. Sci. U.S.A.
1978, 75, 280-284. (b) Stephenson, M. L.; Zamecnik, P. C. Proc. Natl.
Acad. Sci. U.S.A. 1978, 75, 285-288.
(2) (a) Muthiah, M. Antisense Nucleic Acid Drug. DeV. 2002, 12, 103-
128. (b) Aboul-Fadl, T. Curr. Med. Chem. 2005, 12, 2193-2214.
(3) (a) Bennett, C. F. Biochem. Pharmacol. 1998, 55, 9-19. (b) Jarvis,
L. M. Chem. Eng. News 2006, 13-18.
(4) Corey, D. R. Nature Chem. Biol. 2007, 3, 8-11.
(5) (a) Gray, G. D.; Basu, S.; Wickstrom, E. Biochem. Pharmacol. 1997,
53, 1465-1476. (b) Stein, C. A.; Tonkinson, J. L.; Zhang. L. M.; Yakubov,
L.; Gervasoni, J.; Taub, R.; Rotenberg, S. A. Biochemistry 1993, 32, 4855-
4861. (c) Bennett, C. F.; Chiang, M. Y.; Chan, H.; Shoemaker, J. E.;
Mirabelli, C. K. Mol. Pharmacol. 1992, 41, 1023-1033.
10.1021/jo702023s CCC: $40.75 © 2008 American Chemical Society
Published on Web 12/04/2007
J. Org. Chem. 2008, 73, 287-290
287

TABLE 1. Copper(I)-Catalyzed [3+2] Cycloaddition of 2a with
Various Azides
SCHEME 1. Adenosine 2′-O-Alkylation and Processing to
Phosphoramidite Building Blocks 4a and 4b
biopolymers such as proteins,13 sugars,14 and combinations
thereof.15 Triazole derivatives of nucleotides have also been
reported, attached at the nucleobase16 or at the 5′-position.17
We recognized that, with a single exception,17f in all cases
acetylene modified nucleosides were applied, forming a limiting
factor in the design of the conjugation partner. Moreover, for
application in antisense conjugates, nucleobase modification will
in most cases be unfavorable for hybridization, whereas 5′-
conjugation is clearly of interest but only of marginal improve-
ment with respect to known protocols.
In contrast, substitution at 2′-OH, a known strategy to lock
ribose in a 3′-endo conformation, allows the introduction of
either an acetylene or an azide. Moreover, a 2′-modification
can be tolerated at any ribose of the oligonucleotide backbone,
including multiple modifications. Thus, we decided to prepare
2′-O-(3-azidopropyl) and 2′-O-(4-pentynyl) modified adenosine,
with reduced steric interaction in comparison to the shorter 2′-
O-(2-azidoethyl)18 and 2′-O-propargyl,19 and a potential positive
effect on hybridization.20 To avoid the common lengthy routes
to 2′-modified nucleobases, direct 4-pentynylation or 3-azi-
dopropylation of adenosine 2′-OH was investigated.21 Gratify-
ingly, it was found that treatment of a solution of adenosine
(1) in DMF with sodium hydride and an alkyl chloride at 5 °C,
followed by stirring for 3 days at 55 °C, led to a mixture of 2′-
and 3′-O-alkylated isomers (ratio ∼5:1), of which the desired
2′-O-isomer 2 could be readily fractionated by column chro-
matography and selective crystallization from ethanol.
water/tBuOH or water/DMSO at room temperature or at elevated
temperature led to poor conversion and significant green
coloring of the solution, an effect presumably caused by
complexation of copper(II) to the free adenine. A more
rewarding result was obtained by stirring a stoichiometric
mixture of 2a and benzyl azide in the presence of copper wire
in a 9:1 mixture of CH3CN and H2O at 35 °C, leading to the
formation of the desired triazole in excellent yield (Table 1,
entry 1). It was found that removal of the copper wire after 1
h avoided the undesired green coloring. Much to our satisfaction,
brief stirring in the presence of copper wire led to successful
[3+2] cycloaddition for a range of different azides, as sum-
marized in Table 1. For example, adenosine derivative 2a was
conjugated to a fluorescent coumarin label (entry 3), three
different azidonucleosides (entries 4-6), as well as a nitroxyl
spin label, leading to the desired triazole products in near-
quantitative yields in all cases.
Similar to the acetylene modified adenosine 2a, the 2′-O-
azide functionalized adenosine 2b (Table 2) underwent smooth
Huisgen cycloaddition with a diverse set of reaction partners,
including a propargylated coumarin (entry 3) and a fluorescence
quencher (entry 5). The latter compound may be of particular
interest for the application of FRET studies of antisense strand
labeled with fluorescent probes.
With the requisite adenosyl acetylene (2a) and azide (2b) at
hand, Cu(I) catalyzed [3+2] cycloaddition of 2a with benzyl
azide was explored. Initial attempts on Huisgen cycloaddition
catalyzed by Cu(OAc)2 and sodium ascorbate in a mixture of
(18) Grøtli, M.; Douglas, M.; Eritja, R.; Sproat, B. S. Tetrahedron 1998,
54, 5899-5914.
(19) Egli, M.; Minasov, G.; Tereshko, V.; Pallan, P. S.; Teplova, M.;
Inamati, G. B.; Lesnik, E. A.; Owens, S. R.; Ross, B. S.; Prakash, T. P.;
Manoharan, M. Biochemistry 2005, 44, 9045-9057.
(20) Cuenoud, R.; Casset, F.; Hu¨sken, N. F.; Wolf, R. M.; Almann, K.-
H.; Martin, P.; Moser, H. E. Angew. Chem., Int. Ed. 1998, 37, 1288-1291.
(21) (a) Yano, J.; Kan, L. S.; Ts’O, P. O. P. Biochim. Biophys. Acta
1980, 629, 178-183. (b) Wagner, E.; Oberhauser, B.; Halzner, A.; Brunar,
H.; Issakides, G.; Schaffner, G.; Cotten, M.; Knollmu¨ller, M.; Noe, C. R.
Nucl. Acids Res. 1991, 19, 5965-5971.
Having access to a range of functionalized adenosine
monomers, the remaining key question to be addressed con-
cerned the applicability of the building blocks 2a and 2b for
incorporation into an oligonucleotide strand. We were also
intrigued if such a high level of efficiency in cycloaddition
288 J. Org. Chem., Vol. 73, No. 1, 2008

TABLE 2. Copper(I)-Catalyzed [3+2] Cycloaddition of 2b with
Various Acetylenes
SCHEME 2. Solid-Phase Synthesis of 2′-O-Modified
Oligonucleotides 5 and 6 and [3+2] Cycloaddition
would be obtained in the conjugation of two oligonucleotide
strands. RNA chimeras find use, for example, in the selective
delivery of siRNA to prostate cancer cells,22 but so far such
chimeras have not been prepared via azide-acetylene condensa-
tion. Some Cu(I)-catalyzed cycloadditions of acetylene charged
nucleotides and azides are known, but require excess azide17a
or microwave conditions.17e,f Thus, to prepare the requisite
oligonucleotides, the 2′-modified nucleosides 2a and 2b were
converted into the corresponding 5′-O-DMT-3′-O-phosphora-
midite derivatives 4a and 4b with use of standard protocols. It
was noticed that, whereas the conversion of 3a into 4a proceeded
without incidence, the high instability of 4b necessitated rapid
synthesis (40 min), workup, purification on silica gel (5 min),
and concentration. At this stage, the suitability of phosphora-
midites 4a/b for either 3′- or 5′-incorporation into an oligo-
nucleotide (DNA) strand was investigated by automated solid-
phase synthesis on universal CPG (Scheme 2). Gratifyingly,
attachment of the acetylene-modified phosphoramidite 4a to the
resin, as well as repetitive coupling with standard 2′-deoxyri-
bonucleoside phosphoramidites, proceeded smoothly to give the
16-mer chain 3′-rA[(CH2)3(CtCH)d(GCGAGTATTGACCTA)-
5′ (5) in excellent yield and purity after cleavage from the resin.
The azido-containing phosphoramidite 4b, on the other hand,
failedtoincorporateintothenucleotide3′-(dTCATAACTGGATCGC)-
rA[(CH2)3N3]-5′, giving solely the truncated 15-mer. Apparently,
the azido function negatively influenced the coupling behavior
of the 3′-O-phosphoramidite, presumably due to iminophos-
phorane formation by Staudinger reaction, as reported for
2-azidoadenosine.23 Indeed, simply dissolving 4b in acetonitrile
led to spontaneous degradation, as judged by 31P NMR, with
near-complete disappearance of starting material after two
weeks. Consequently, an alternative approach was conceptual-
ized involving the introduction of azido functionality after
oligonucleotide assembly. To this end, an N-MMT protected
phosphoramidite was attached at the 5′-terminus of the 15-mer
oligonucleotide, followed by deprotection and azidoacetylation
to give, after base cleavage, compound 6.
Finally, [3+2] cycloaddition of fragments 5 and 6 was
attempted. First of all, a template-directed cycloaddition of the
two complementary sequences was attempted.24 It was found
that no reaction took place in the absence of copper, but
treatment of the two strands 5 and 6 in an CH3CN/H2O mixture
(1:9) with copper wire led to rapid disappearance of both
nucleotides and the formation of a single new compound as
judged by HPLC. The resulting product was unambiguously
identified as the triazole-linked conjugate 7 by MALDI-TOF
mass spectrometry.
In conclusion, a fast and practical procedure for the 2′-O-
alkylation of adenosine was developed for introduction of azido
or acetylene functionality, which could be converted into a range
of conjugates including other nucleosides, spin-labels, and
fluorescent probes. Conversion of the acetylene containing
compound 4a into the corresponding phosphoramidite proceeded
smoothly, but not for azide 4b, necessitating azide introduction
at a late stage. Conjugation of the oligonucleotides by 1,3-dipolar
cycloaddition readily occurred in the presence of copper wire,
(22) (a) McNamara, J. O.; Andrechek, E. R.; Wang, Y.; Viles, K. D.;
Rempel, R. E.; Gilboa, E.; Sullenger, B. A.; Giangrande, P. H. Nat.
Biotechnol. 2006, 24, 1005-1015. (b) Chu, T. C.; Twu, K. Y.; Ellington,
A. D.; Levy, M. Nucleic Acids Res. 2006, 34, e73.
(23) Wada, T.; Mochizuki, A.; Higashiya, S.; Tsuruoka, H.; Kawahar,
S.; Ishikawa, M.; Sekine, M. Tetrahedron Lett. 2001, 42, 9215-9219.
(24) (a) Poulin-Kerstien, A. T.; Dervan, P. B. J. Am. Chem. Soc. 2003,
125, 15811-15821. (b) Kumar, R.; El-Sagheer, A.; Tumpane, J.; Lincoln,
P.; Wilhelmsson, L. M.; Brown, T. J. Am. Chem. Soc. 2007, 129, 6859-
6864.
J. Org. Chem, Vol. 73, No. 1, 2008 289

General Procedure for Copper(I)-Catalyzed Azido-Alkyne
Cycloaddition (CuAAC). 2′-O-Pentynyladenosine 2a (67 mg, 0.2
mmol) and 1 equiv of an azido compound were dissolved in a
mixture of acetonitrile (2 mL) and water (0.2 mL). The reaction
mixture was stirred at 35 °C in the presence of a Cu-wired stirring
bar. After 1 h, the stirring bar was removed and stirring was
prolonged until all starting material was consumed. Concentration
in vacuo and purification by flash chromatography (CH2Cl2/MeOH,
95:5) furnished the triazole-functionalized nucleosides. The same
procedure was applied to azidopropyladenosine 2b by stirring with
an acetylene reaction partner.
Analytical Data for the CuAAC-Adduct of 2a and 5′-Azido-
5′-deoxyadenosine (entry 5 of Table 1). 1H NMR (300 MHz
DMSO) δ 8.37, 8.17, 8.13.(s,1H), 8.11 (s, 1H), 7.60 (s, 1H), 7.32
(s, 2H), 5.98 (d, J ) 6.0 Hz, 2H), 5.97 (d, J ) 5.1 Hz), 5.79-5.29
(br s, 4H), 4.66-4.65 (m , 3H), 4.46 (t, J ) 7.2 Hz, 1H), 4.31 (br
s, 1H), 4.21 (br s, 2H), 3.97 (d, J ) 3 Hz), 3.67-3.64 (m, 3H),
1.72-1.61 (m, 2H); 13C NMR (75 MHz DMSO) δ 156.1, 156.0,
152.7, 152.5, 149.2, 149.0, 146.2, 139.8, 122.6, 119.3, 119.2, 87.7,
86.3, 86.0, 82.3, 80.8, 72.6, 70.9, 68.9, 68.7, 61.5, 60.6, 51.2, 28.9,
21,23; HRMS calcd for C25H31N13O7 (M + Na+) 648.2367, found
648.2326.
thus paving the way for the synthesis of unique, spacer-linked
oligonucleotide chimeras.
Experimental Section
Synthesis of 2′-O-(1-Pentyn-5-yl)adenosine (2a). Adenosine (5
g, 18.72 mmol) was dissolved in hot anhydrous DMF (250 mL)
under an inert atmosphere. The solution was cooled to 5 °C, and
NaH (1 g, 25 mmol, 60% dispersion in mineral oil) was subse-
quently added, followed by the addition of TBAI (1.5 g, 4.06 mmol)
and 5-chloro-1-pentyne (2.5 mL, 23.84 mmol). The reaction mixture
was allowed to stir for 3 days at 55 °C. Evaporation of the solvent
under reduced pressure resulted in a suspension, which was
absorbed on silica gel. Flash chromatography (CH2Cl2/MeOH, 95:
5) gave a mixture of 2′-O- and 3′-O-alkynylated products. Selective
crystallization from anhydrous ethanol provided the title compound
1
1 (2.5 g, 40%) as colorless needles. H NMR (300 MHz, DMSO)
δ 8.36 (s, 1H), 8.12 (s, 1H), 7.33 (br s, 2H), 6.00 (d, J ) 6 Hz,
1H), 5.39 (m, 1H), 5.17 (d, J ) 5.4 Hz, 1H), 4.46 (m, 1H), 4.29
(m, 1H), 3.97 (m, 1H), 3.64-3.32 (m, 4H), 2.64 (m, 1H), 2.08 (m,
2H), 1.58 (m, 2H); 13C NMR (75 MHz, DMSO) δ 156.1, 152.5,
149.0, 139.7, 119.3, 86.3, 86.0, 83.9, 81.0, 71.1, 69.0, 68.3, 61.4,
28.1, 14.3; HRMS calcd for C15H19N5O4 (M + Na+) 356.1335,
found 356.1308.
Acknowledgment. The European Commission is kindly
acknowledged for financial support (FSG-V-RNA).
The same procedure with 3-azido-1-chloropropane instead of
1
5-chloro-1-pentyne provided compound 2b. H NMR (300 MHz,
DMSO) δ 8.36 (s, 1H), 8.12 (s, 1H), 7.32 (br s, 2H), 6.00 (d, J )
6 Hz, 1H), 5.38 (m, 1H), 5.20 (d, J ) 5.4 Hz, 1H), 4.46 (m, 1H),
4.29 (m, 1H), 3.97 (m, 1H), 3.64-3.43 (m, 4H), 3.27 (m, 2H),
1.67 (m, 2H); 13C NMR (75 MHz, DMSO) δ 156.1, 152.5, 149.0,
139.6, 119.3, 86.3, 86.0, 81.0, 68.0, 66.7, 61.4, 47.6, 28.5; HRMS
calcd for C13H18N8O4 (M + Na+) 351.1529, found 351.1525.
Supporting Information Available: General procedures,
synthesis, and characterizations of all new compounds. This material
is available free of charge via the Internet at http://pubs.acs.org.
JO702023S
290 J. Org. Chem., Vol. 73, No. 1, 2008