Synthesis of DNA conjugates by solid-phase fragment condensation via aldehyde-nucleophile coupling

Article abstract of DOI:10.1016/j.tetlet.2005.03.056

Oligodeoxyribonucleotides were synthesized that contain a novel nucleoside, 2′-O-(2,3-dihydroxypropyl)cytidine. Its 2′-diol group was blocked by an allyloxycarbonyl protecting group. Selective deprotection of diol group(s) of the support-immobilized blocked oligodeoxyribonucleotide by Pd(0) followed by periodate oxidation resulted in generation of the 2′-aldehyde group(s) on solid-phase. The modified oligonucleotides were used to prepare a number of conjugates with acridine, biotin and N-modified laminin peptides by oxime, hydrazone and hydrazine formation. The method may be applicable to the synthesis of oligonucleotide-peptide conjugates.

Full text of DOI:10.1016/j.tetlet.2005.03.056

Tetrahedron
Letters
Tetrahedron Letters 46 (2005) 3191–3195
Synthesis of DNA conjugates by solid-phase fragment
condensation via aldehyde–nucleophile coupling
Timofei S. Zatsepin,a Dmitry A. Stetsenko,b Michael J. Gaitb and Tatiana S. Oretskayaa,*
aDepartment of Chemistry, M. V. Lomonossov Moscow State University, Leninskie gory, Moscow 119992, Russia
bMRC Laboratory of Molecular Biology, Hills Road, Cambridge CB2 2QH, UK
Received 26 January 2005; accepted 10 March 2005
Available online 25 March 2005
Abstract—Oligodeoxyribonucleotides were synthesized that contain a novel nucleoside, 20-O-(2,3-dihydroxypropyl)cytidine. Its 20-
diol group was blocked by an allyloxycarbonyl protecting group. Selective deprotection of diol group(s) of the support-immobilized
blocked oligodeoxyribonucleotide by Pd(0) followed by periodate oxidation resulted in generation of the 20-aldehyde group(s) on
solid-phase. The modified oligonucleotides were used to prepare a number of conjugates with acridine, biotin and N-modified
laminin peptides by oxime, hydrazone and hydrazine formation. The method may be applicable to the synthesis of oligonucleo-
tide–peptide conjugates.
ꢀ 2005 Elsevier Ltd. All rights reserved.
Nowadays, oligonucleotide derivatives are widely used in
many fields of biotechnology and medicine for diagnos-
tics, for example, by PCR or hybridization assay, and
as potential therapeutic agents.1,2 However, synthesis
of modified oligonucleotides and their conjugates is still
evolving. Attachment of reporter groups enhances sensi-
tivity of detection of DNA sequences; coupling of pep-
tides, carbohydrates or cholesterol leads to increase in
cellular uptake. A variety of methods for conjugation
of oligonucleotides with target compounds have been
developed so far.3,4 Oligonucleotide conjugates may be
synthesized by use of phosphoramidite derivatives of tar-
get compounds, or by post-synthetic conjugation on
solid-phase, or in solution by reaction of oligonucleotides
with other molecules. Amongst the methods for covalent
joining of large biomolecules, aldehyde–nucleophile cou-
pling is gaining popularity in biomolecular chemistry5–8
and is particularly efficient in the case of difficult conjuga-
tion with multifunctional compounds, for example, pep-
tides. The advantages of this approach are in the high
yields of conjugates, chemoselectivity of the reaction
and use of smaller excess of the nucleophilic component.
Synthesis of oligonucleotide–peptide conjugates is well
attested.9–11 Total stepwise solid-phase synthesis is the
most direct method for the preparation of oligonucleo-
tide–peptide conjugates. Although it provides generally
good yields of the final product, difficulties in finding a
combination of compatible protecting groups for both
oligonucleotide and peptide moieties and losses due to
incomplete coupling at every elongation step limit the
nature and length of peptide sequences, especially those
suitable for cell delivery. To overcome this problem, so-
lid-phase coupling strategies have been developed for
synthesis of DNA–peptide12–14 and PNA–peptide15 con-
jugates. However, joining of protected peptide fragments
via acylation reactions proceeds slowly and in moderate
yields, due to their limited solubility in aprotic bipolar
solvents and adverse steric effects.
Here we report the preparation of blocked 20-O-(2,3-
dihydroxypropyl)cytidine phosphoramidite, its incorpo-
ration into oligonucleotides, and their subsequent use
for chemoselective conjugation. Use of allyloxycarbonyl
protecting groups ensures selective unmasking of the diol
group from the still protected and immobilized oligonu-
cleotide. Subsequent periodate oxidation on solid-phase
leads to the generation of an aliphatic aldehyde group in
the oligonucleotide. Then an efficient chemoselective
conjugation with a number of compounds, including
N-modified laminin peptides, was accomplished.
Keywords: Modified oligonucleotides; Solid-phase conjugation; Alde-
hyde; Oxime; Hydrazone.
The target phosphoramidite 3 was synthesized by a
route similar to that we published previously for the
20-O-(2,3-dihydroxypropyl)uridine17 (Scheme 1). Initial
*
Corresponding author. Tel.: +7 095 9395411; fax: +7 095 9393148;
e-mail: oretskaya@belozersky.msu.ru
0040-4039/$ - see front matter ꢀ 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2005.03.056

3192
T. S. Zatsepin et al. / Tetrahedron Letters 46 (2005) 3191–3195
NHBz
N
NHBz
N
NHBz
N
N
O
N
O
N
O
O
O
DMTrO
O
O
a, b
c, d, e
Si
O
Si
O
O
O
O
O
O
O
O
O
O
O
O
Si
O
O
O
Si
O
O
O
O
O
P
CN
N
O
1
2
3
Scheme 1. Preparation of the phosphoramidite 3. Reagents and conditions: (a) OsO4 (0.08 equiv), N-methylmorpholine N-oxide, THF–H2O (2:1,
v/v), 1 h, 92%; (b) AllocCl, Et3N, CH2Cl2, 4 ꢁC, 1 h, 82%; (c) TBAF, THF, 10 min, 91%; (d) DMTrCl, Py, 3 h, 87%; (e) (iPr2N)2PO(CH2)2CN,
diisopropylammonium tetrazolide, CH2Cl2, 1 h, 75%.16
derivative 1 was obtained as described by Sproat et al.18
from uridine. Oxidation of the 20-allyloxy group by
osmium tetroxide and N-methylmorpholine N-oxide
resulted in a mixture (60:40) of diastereomeric diols. The
portion of osmium tetroxide was increased in compari-
son to a previously published procedure, since cytosine
derivatives are hardly oxidized to the 5,6-diol.19 An
allyloxycarbonyl protecting group (Alloc) was chosen
to protect the 20-diol. Application of allyl and allyloxy-
carbonyl groups in oligonucleotide synthesis to protect
phosphates and heterocyclic bases was pioneered by
Hayakawa et al.20,21 Later Manoharan and co-work-
ers22 used Alloc for selective generation of an aliphatic
amino group in oligonucleotides on solid-phase. Very re-
cently, we have demonstrated successful unmasking of
a carboxylic acid group on solid support based on allyl/
Pd(0) chemistry.23,24 Previously, it was shown that side
modification of the 3-position of thymine or uracil by
allyl chloroformate may take place.25 The resultant deriva-
tive could not be deprotected by the Pd(0)/Ph3P system
under various conditions. This was the underlying
reason for selection of cytidine as the nucleoside. The
TIPDS protecting group was removed by fluoride
ion26 and the standard transformations27 of 2 were car-
ried out to provide the building block 328 suitable for
automated solid-phase synthesis. The modified oligo-
nucleotides were prepared by the phosphoramidite
approach. For the phosphoramidite 3, the coupling time
was increased to 10 min and the coupling was repeated
twice. Previously, tert-butyl hydroperoxide or 1-(S)-(+)-
(10-camphorsulfonyl)-oxaziridine had been used for
the P(III) oxidation in the case of allyl phosphorami-
dites or solid supports to prevent possible side-reactions
with the standard iodine oxidizer.20 In our hands, no
side-reactions were observed if 0.02 M I2 solution in
THF/Py/H2O (40:10:1) or 0.5 M 1-(S)-(+)-(10-cam-
phorsulfonyl)oxaziridine in MeCN were used as oxidants.
The sequences synthesized are shown in Table 1.
Deprotection of the 20-diol group of the modified oligo-
nucleotides was carried out under conditions similar to
those reported by Manoharan and co-workers22
(Scheme 2). However, in our case there was no need to
heat the reaction mixture. This may be explained by
the lower stability of the allyl carbonate linkage in com-
parison to allyl carbamate. The allyloxycarbonyl groups
were also deprotected during standard ammonia treat-
ment. Thus in effect, synthon 3 is a universal building
block for both solid-phase and homogeneous prepara-
tion of 20-aldehyde oligonucleotides, which is followed
by their conjugation. Preliminary deprotection of phos-
phate groups by DBU treatment had no influence on the
yields and time of the subsequent oxidation and conju-
gation, and therefore we avoided this additional step.
Generation of the 20-aldehyde group on solid-phase
was carried out by use of sodium periodate in a
DMSO—acetate buffer mixture.
Recently Lo¨nnberg and co-workers29,30 published a
method for solid-phase conjugation of various alde-
hydes to immobilized and protected oligonucleotides
bearing a free amino-oxy group. Here we adopted the
reverse way of coupling. Initial experiments on conjuga-
tion of the immobilized 20-aldehyde oligonucleotides with
a number of hydrazides, hydrazines and amino-oxy com-
pounds led to significant modification of heterocyclic
bases (Fig. 1c). Use of Ôultra fast deprotectionÕ 4-tert-
butylphenoxyacetyl phosphoramidites still afforded
about 10–15% of the base-modified oligonucleotide after
1 h treatment with 1000 M equiv of O-benzylhydroxyl-
amine followed by standard ammonolysis. However,
use of the appropriate salts, for example, hydrochloride,
trifluoroacetate or even acetate of the amino component
ensured conjugation without any traces of side-products
even in the case of standard phosphoramidites. Figure
1b demonstrates that if methoxyamine is used as its
hydrochloride, the reaction proceeds exclusively at the
20-aldehyde group. Addition of triethylamine or py-
ridine (1000 mol equiv) into the reaction mixture leads
to liberation of methoxyamine as its free base and for-
mation of a complex reaction mixture (Fig. 1c). The
oxime formed was stable under ammonia treatment
for several days at ambient temperature. Under these
conditions the 50-dimethoxytrityl group is partly depro-
tected during the conjugation step, so a DMTr OFF
mode of oligonucleotide assembly was used throughout.
The same procedure was also applied for the
Table 1. Sequences and MALDI-TOF data of the 20-modified
oligonucleotides
No.
Oligonucleotide
MALDI-TOF MS
calcd/found [M+H]+
I
CTCCCAGGCTCA
CTCCCAGGCUCA
3656.4/3658.1
3746.5/3646.9
II
C—20-O-(2,3-dihydroxypropyl)cytidine.

T. S. Zatsepin et al. / Tetrahedron Letters 46 (2005) 3191–3195
3193
protected DNA
CytBz
CytBz
O
DMTrO
O
O
O
a
O
O
O
O
O
O
O
O
P
O
O
O
P
O
O
O
O
O
CN
O
O
NC
N
O
protected DNA
b, c
CPG
CPG
3
4
DNA
O
protected DNA
DNA
CytBz
Cyt
Cyt
O
O
d or e
O
O
O
or
H
N
O
P
O
O
O
P
O
O
O
P
O
O
O
R2
N
R1
N
H
O
O
O
O
O
O
O
NC
DNA
protected DNA
DNA
CPG
CPG
6
7
5
Scheme 2. Preparation of the 20-aldehyde-containing oligonucleotides and their conjugates. Reagents and conditions: (a) automated oligonucleotide
synthesis; (b) Pd(Ph3P)4 (20 mg), Ph3P (30 mg), BunNH2–HCO2H (200 lL, 1 M in CH2Cl2), 1 h; (c) 0.05 M NaIO4, 0.2 M NaOAc, pH 5.0, 90% aq
DMSO, 3 h; (d) 1000 mol equiv R1ONH2–HCl or TFA in 100 lL DMSO, 3 h; (e) 1000 mol equiv R2NHNH2–AcOH or TFA in 100 lL DMSO, 3 h;
then NaBH3CN (10 mg), 1 h; (e) concd aq NH3, 25 ꢁC, 2 days. R1, R2-conjugated moiety (see Table 2).
Figure 1. RP-HPLC profiles of crude reaction mixtures (detection at 260 nm): (a) 20-aldehydo oligonucleotide I; (b) reaction of 20-aldehydo
oligonucleotide I with MeONH2 · HCl; (c) reaction of 20-aldehydo oligonucleotide I with MeONH2 free base; peak 1—20-aldehydo oligonucleotide
I, peak 2—20-oxime, peaks 3–5—side-products of base modification.
conjugation with 9-hydrazinoacridine. However, the
yield was only 3–10% (Fig. 2a). Addition of sodium
cyanoborohydride into the reaction mixture increased
the yield up to 70% (Fig. 2b). Thus the hydrazone link-
age is partly hydrolyzed during ammonia treatment
in the case of aromatic hydrazines. No product was
observed if biotin hydrazide was used without sodium
cyanoborohydride, whilst in situ hydrazone reduction
provided a good yield of the conjugate (Table 2). These
data do not correlate well with the results of Okamoto
et al.31 who described a solid-phase hydrazone formation.
3-Formylindole 20-deoxyriboside was incorporated into

3194
T. S. Zatsepin et al. / Tetrahedron Letters 46 (2005) 3191–3195
Figure 2. RP-HPLC profiles of crude conjugation mixtures (detection at 260 nm): (a) reaction of 20-aldehydo oligonucleotide
I with
9-hydrazinoacridine · HCl, peak 1—20-aldehydo oligonucleotide I, peak 2—hydrazone conjugate; (b) reaction of 20-aldehydo oligonucleotide I
with 9-hydrazinoacridine · HCl and NaBH3CN, peak 1—20-aldehydo oligonucleotide I, peak 3—hydrazino conjugate.
Table 2. Conversion yields and MALDI-TOF MS analysis of the conjugates of the 20-modified oligonucleotides
No.
Conjugated molecule
Oligonucleotide
I
II
8
MeONH2
Biotin hydrazide
90a (3654.4/3653.1)b
81a (3867.7/3868.1)b
10a (3818.6/3818.1)b
70a (3820.6/3822.0)b
87a (4543.4/4544.0)b
61a (4586.4/4587.9)b
84a (3750.5/3751.0)b
78a (4168.1/4167.9)b
3a (4055.9/4055.0)b
65a (4057.9/4059.1)b
71a (5508.4/5507.6)b
49a (5606.5/5608.9)b
9
10
11
12
13
9-Hydrazinoacridine (unreduced)
9-Hydrazinoacridine (reduced)
H2NOCH2CO-DPGYIGSR-NH2
H2NNHCO(CH2)2CO-DPGYIGSR-NH2
a Yield of the conjugate from integration of RP-HPLC trace.
b MALDI-TOF MS: calcd/found [M+H]+.
oligonucleotides followed by on-column reaction with
an ethanolic solution of a hydrazine for 48 h. The conju-
gate was then deprotected under standard ammonolytic
conditions with retention of the hydrazone linkage.
However, such a difference in stability may be ascribed
to the aromatic nature of the aldehyde group used by
Okamoto et al.31 N-Amino-oxy- and N-hydrazido-modi-
fied laminin peptide were successfully conjugated to the
20-aldehyde oligonucleotides I and II in the same way
as 9-hydrazinoacridine and biotin hydrazide (Table 2).
Stabilities of oxime and hydrazone conjugates corre-
sponded well with our previous results.32 It should be
emphasized that yields of conjugates were rather good,
even in the case of oligonucleotide II modified with
two reactive groups (Table 2), although one of the 20-
modified nucleotides was close to the solid support.
The conjugates were characterized by MALDI-TOF
MS. In all cases, the experimentally determined molecu-
lar masses were in good agreement with the calculated
values (Table 2).

T. S. Zatsepin et al. / Tetrahedron Letters 46 (2005) 3191–3195
3195
18. Sproat, B. S.; Iribarren, A.; Beijer, B.; Piels, U.; Lamond,
A. I. Nucleos. Nucleot. 1991, 10, 25–36.
In conclusion, we have described the application of
a new allyloxycarbonyl-protected 20-O-(2,3-dihydroxy-
propyl)cytidine synthon for efficient incorporation into
oligonucleotides via solid-phase phosphoramidite syn-
thesis, selective unblocking and quantitative oxidation
of its 20-diol group to the aldehyde, and successful con-
jugation of the modified oligonucleotides with a range of
nucleophilic molecules including peptides.
19. Burton, K.; Riley, W. T. Biochem. J. 1966, 98, 70–77.
20. Hayakawa, Y.; Kato, H.; Uchiyama, M.; Kajino, H.;
Noyori, R. J. Org. Chem. 1986, 51, 2400–2402.
21. Hayakawa, Y. Bull. Chem. Soc. Jpn. 2001, 74, 1547–1565.
22. Maier, M. A.; Barber-PeocÕh, I.; Manoharan, M. Tetra-
hedron Lett. 2002, 43, 7613–7616.
23. Kachalova, A. V.; Zubin, E. M.; Zatsepin, T. S.; Agapk-
ina, Yu. V.; Ivanova, Yu. M.; Stetsenko, D. A.; Gait, M.
J.; Oretskaya, T. S. In Innovation & Perspectives in Solid-
Phase Synthesis & Combinatorial Libraries; Epton, R.,
Ed.; Mayflower Worldwide: Kingswinford, 2004, pp 255–
260.
Acknowledgements
This work was supported by the Wellcome Trust CRIG
069419 and partly by the RFBR grant N 03-04-48957.
24. Kachalova, A.; Zubin, E.; Stetsenko, D.; Gait, M.;
Oretskaya, T. Org. Biomol. Chem. 2004, 2, 2793–2797.
25. Hayakawa, Y.; Hirose, M.; Noyori, R. J. Org. Chem.
1993, 58, 5551–5555.
References and notes
26. Matsuda, A.; Yasuoka, J.; Sasaki, T.; Ueda, T. J. Med.
Chem. 1991, 34, 999–1002.
1. Opalinska, J. B.; Gewirtz, A. M. Nat. Rev. Drug Discovery
2002, 1, 503–514.
2. Heller, M. J. Annu. Rev. Biomed. Eng. 2002, 4, 129–153.
3. Verkantesan, N.; Kim, S. J.; Kim, B. H. Curr. Med. Chem.
2003, 10, 1973–1991.
27. Caruthers, M. H.; Barone, A. D.; Beaucage, S. L.; Dodds,
D. R.; Fisher, E. F.; McBride, L. J.; Matteucci, M.;
Stabinsky, Z.; Tang, J.-Y. Methods Enzymol. 1987, 154,
287–313.
28. 1H NMR (500 MHz, CDCl3, d, ppm): 10.12 (s, 1H,
NHBz), 8.13 (d, 1H, H-6, J5,6 = 7.5 Hz), 7.64–7.52 (m, 6H,
Bz, H-5), 7.29–6.85 (m, 13H, DMTr), 6.23 (d, 1H, H-10,
4. Virta, P.; Katajisto, J.; Niittyma¨ki, T.; Lo¨nnberg, H.
Tetrahedron 2003, 59, 5137–5174.
0
0
5. Tam, J. P.; Xu, J.; Eom, K. D. Biopolymers 2001, 60, 194–
205.
J1 ,2 = 6.9 Hz), 5.95 (m, 2H, OCH2CH@), 5.32–5.25 (m,
4H, CH@CH2), 4.70 (m, 2H, CH2OC@O), 4.60 (m, 4H,
OCH2CH@), 4.53 (m, 1H, CHOC@O), 4.20 (m, 1H, H-
20), 4.14 (m, 1H, H-30), 4.01 (m, 1H, H-40), 3.73 (s, 6H,
MeO), 3.67 (m, 2H, POCH2), 3.41 (dd, 1H, H-50a,
6. Peri, F.; Nicotra, F. Chem. Commun. 2004, 623–627.
7. Forget, D.; Boturyn, D.; Defrancq, E.; Lhomme, J.;
Dumy, P. Chem. Eur. J. 2001, 7, 3976–3984.
8. Zatsepin, T. S.; Stetsenko, D. A.; Arzumanov, A. A.;
Romanova, E. A.; Gait, M. J.; Oretskaya, T. S. Biocon-
jugate Chem. 2002, 13, 822–830.
9. Stetsenko, D. A.; Arzumanov, A. A.; Korshun, V. A.;
Gait, M. J. Mol. Biol. (Russ.) 2000, 34, 998–1006.
10. Tung, C. H.; Stein, S. Bioconjugate Chem. 2000, 11, 605–
618.
11. Zubin, E. M.; Romanova, E. A.; Oretskaya, T. S. Russ.
Chem. Rev. 2002, 71, 239–264.
12. Peyrottes, S.; Mestre, B.; Burlina, F.; Gait, M. J. Tetra-
hedron 1998, 54, 12513–12522.
13. Zubin, E. M.; Romanova, E. A.; Volkov, E. M.; Tashlit-
sky, V. N.; Korshunova, G. A.; Shabarova, Z. A.;
Oretskaya, T. S. FEBS Lett. 1999, 456, 59–62.
14. Hwang, J. T.; Greenberg, M. M. J. Org. Chem. 2001, 66,
363–369.
15. Lewis, M. R.; Jia, F.; Gallazzi, F.; Wang, Y.; Zhang, J.;
Shenoy, N.; Lever, S. Z.; Hannink, M. Bioconjugate Chem.
2002, 13, 1176–1180.
J5 a,5 b = 11.8 Hz), 3.30 (dd, 1H, H-50b), 2.82 (m, 1H,
CH(CH3)2), 2.42 (t, 2H, CH2CN), 1.15 (s, 12H, CH3CH);
13C NMR (162 MHz, CDCl3, d, ppm): 167.60 (C@O, Bz),
162.59 (C-4), 158.67 (DMTr), 156.87 (C-2), 154.54, 154.08
(OC(O)O), 143.83 (DMTr), 142.41 (C-6), 134.58 (DMTr),
134.19 (C-4, Bz), 132.20 (CH@), 132.10, 128.97 (C-3, C-1,
Bz), 128.93, 128.03 (DMTr), 127.99 (C-2, Bz), 127.86,
127.11 (DMTr), 119.10 (CN), 118.79 (@CH2), 113.52
(DMTr), 96.55 (C-5), 89.79 (C-10), 87.11 (DMTr), 85.25
(C-20) 82.73 (C-40), 81.56 (C-30), 74.09 (C-200), 68.74 (C-
100), 68.35 (OCH2CH@), 66.25 (C-300), 66.01 (C-50), 58.18
(POCH2), 55.29 (MeO), 43.37 (PNCH), 24.66 (CH3CH),
20.21 (CH2CN); 31P NMR (202 MHz, CDCl3, d, ppm):
152.30, 151.33, mixture of diastereomers; MALDI-TOF
MS: [M+H]+ calcd 1093.1, found 1093.2.
0
0
29. Salo, H.; Virta, P.; Hakala, H.; Prakash, T. P.; Kawasaki,
A. M.; Manoharan, M.; Lo¨nnberg, H. Bioconjugate Chem.
1999, 10, 815–823.
30. Katajisto, J.; Virta, P.; Lo¨nnberg, H. Bioconjugate Chem.
2004, 15, 890–896.
16. Alloc–allyloxycarbonyl, Bz–benzoyl, TBAF–tetrabuty-
lammonium fluoride, DMTr–4,40-dimethoxytrityl, Py–
pyridine.
31. Okamoto, A.; Tainaka, K.; Saito, I. Tetrahedron Lett.
2002, 43, 4581–4583.
17. Kachalova, A. V.; Zatsepin, T. S.; Romanova, E. A.;
Stetsenko, D. A.; Gait, M. J.; Oretskaya, T. S. Nucleos.
Nucleot. Nucl. Acids 2000, 19, 1693–1707.
32. Zatsepin, T. S.; Ivanova, Y. M.; Stetsenko, D. A.; Gait,
M. J.; Oretskaya, T. S. Tetrahedron Lett. 2004, 45, 7327–
7330.