A novel versatile phosphoramidite building block for the synthesis of 5′- and 3′-hydrazide modified oligonucleotides

Article abstract of DOI:10.1081/NCN-200055723

We introduce a novel versatile phosphoramidite building block for the modification of oligonucleotides (ONs) with acyl hydrazides on the 5′- or 3′-terminus, or both. The reaction of these hydrazide functionalized ONs with 4-methoxyphenylaldehyde is demons

Full text of DOI:10.1081/NCN-200055723

Nucleosides, Nucleotides, and Nucleic Acids, 24 (3):211–226, (2005)
Copyright D 2005 Taylor & Francis, Inc.
ISSN: 1525-7770 print/ 1532-2335 online
DOI: 10.1081/NCN-200055723
A NOVEL VERSATILE PHOSPHORAMIDITE BUILDING BLOCK FOR
THE SYNTHESIS OF 5’- AND 3’-HYDRAZIDE
MODIFIED OLIGONUCLEOTIDES
5
Sergey I. Antsypovich
Department of Chemistry, Ruhr University, Bochum, Germany
5
Department of Chemistry, M. V. Lomonosov Moscow State University, Moscow, Russia
5
G¨unter von Kiedrowski
Department of Chemistry, Ruhr University, Bochum, Germany
5
We introduce a novel versatile phosphoramidite building block for the modification of
oligonucleotides (ONs) with acyl hydrazides on the 5’- or 3’-terminus, or both. The reaction of these
hydrazide functionalized ONs with 4-methoxyphenylaldehyde is demonstrated for solution derivatiza-
tion. Hydrazides are considered nowadays as promising reactants, which show enhanced reactivity at
neutral and slightly acidic conditions and higher stability of yielding products as compared to the
aliphatic amines, which are broadly used for ONs derivatization.
Our method to introduce hydrazides into ONs employs a phosphoramidite modifier designed to split,
during ammonia or lithium hydroxide treatment, into two hydrazides via b-elimination of a central bis-
2-carbonylethoxysulfone unit. It allows the creation of ONs derivatized with a hydrazide moiety at the
5’-, 3’- and both 5’- and 3’-termini, as well as two different hydrazide containing ONs at the same
time, viz. in one sequence on the same solid support. In latter case one can, for example, synthesize two
hydrazide containing ONs, where one is 5’-modified and second one is 3’-modified.
Keywords Hydrazide, Aldehyde, Modified Oligonucleotide, Phosphoramidite Modifier,
Automatic Oligonucleotide Synthesis
INTRODUCTION
The modification of ONs with nucleophilic functional groups is a routine
operation nowadays enabling numerous applications and assays in molecular
biology, biochemistry, and biotechnology.[1–4] One of them is amino group. In our
Received 7 September 2004, accepted 31 January 2005.
This work was supported by German Research Council (DFG) and the Federal Ministry of Education and
Research (BMBF, Germany). S.I. Antsypovich gratefully acknowledges a stipend and his fellowship from
German Academic Exchange Service (DAAD) and Alexander von Humboldt Foundation (AvH-Stiftung, Ger-
many), respectively.
Address correspondence to Sergey I. Antsypovich, Department of Chemistry, Ruhr University, Bochum,
Germany; Fax: +49 (234) 32-14355; E-mail: sergej.antsypowitsch@ruhr-uni-bochum.de
211
Order reprints of this article at www.copyright.rightslink.com

212
S. I. Antsypovich and G. von Kiedrowski
opinion, hydrazide modified ONs can be considered as a rather advantageous and
promising alternative to alkyl amino–modified ONs.
Briefly, it is well known that alkyl amino modified ONs react with aldehydes
to produce rather unstable Schiff’s bases, which require reduction to yield more
stable amine derivatives.[1,5–9] Moreover, for a large variety of applications the
applicability of aliphatic amino groups is limited due to the fact that their nu-
cleophilicity heavily depends on the pH. At low pH values amines are protonated,
and thus reactions with electophiles are suppressed. At high pH values amines
are reactive, but in many cases basic pHs are unacceptable due to other reasons.
For example, active esters, which are often used for conjugation with amines,
are rather unstable at basic conditions due to quick hydrolysis. Water-soluble
carbodiimides (CDIs), which are often used for conjugation of ONs with various
electrophilic ligands, are readily hydrolyzed under basic conditions too. More-
over, at basic pH, CDIs may be also attacked by exocyclic amino groups of
heterocyclic bases, thus producing undesirable side products.[10] Another
disadvantage is acyl migration, taking place on the CDI-activated intermediate.
This side process is common in CDI mediated couplings if carried out in aqueous
solution.[10]
On the contrary, the reactivity of hydrazides is similar to that of amines but due
to the lower pKa values (ꢀ2.5–5 for hydrazides versus 10–11 for amines) caused
by the a-effect of an NH group next to a NH2 group in the hydrazide
functionality,[11] hydrazides remain reactive at neutral and even at gently acidic
conditions. The main targets for hydrazides are aldehyde functions. Reactions of
hydrazides with aldehydes usually proceed rather fast, with high yield and with
almost no side products. The rate constant for the reaction with aromatic aldehydes
at pH 5–6 is in the order of 0.1–1 Mꢁ1 sꢁ1 according to our experiences.
Moreover, aldehydes being attacked by hydrazides form hydrazone structures
which are much more stable as compared to Schiff’s bases. Consequently,
aldehyde containing molecules can be coupled to hydrazide functionalized ones
without further reduction.[1,2,10,12,13] For example, when a 3’-terminal ribose is
oxidized by sodium periodate to give a bis-aldehyde derivative, the reaction with
hydrazides yields a stable dihydroxymorpholine linkage.[2,12] The hydrazone
obtainment in all cases does not require any additional activation and can be
readily performed in mild conditions suitable both for ONs and also double strand
nucleic acids survival.
The hydrazone conjugation chemistry enables two principle approaches
because the ONs can be derivatized both with hydrazide or with aldehyde groups.
The advantage of a hydrazide-modified ON is that no intra- or intermolecular
reaction can occur during the storage in a stock solution. This is not the case for an
aldehyde-modified ON, where an intra- or intermolecular attack of an exocyclic
amino group on aldehyde functionality may take place. Nevertheless, aldehyde-
modified ONs are also successfully employed; for example, an immobilization of
such ONs was reported on solid surfaces.[11,14]

Synthesis of 5’- and 3’-Hydrazide Modified Oligonucleotides
213
Due to the above features, hydrazone formation has found many applications
in protein and carbohydrate chemistry,[1,15] analytical biochemistry,[16–18] antisense
biotechnology[19] and biochip manufacturing.[2] One group of applications deals
with immobilization reactions. Examples include the immunochemical detection of
glycosphingolipids on thin-layer chromatograms,[16,18] the detection of glycopro-
teins,[16,17] and affinity chromatography.[20,21] Synthetic ONs modified at their 5’-
end with an aldehyde group can be easily attached to latex microspheres containing
hydrazide moieties.[11] The coupling can be carried out under mild conditions, the
reaction is usually rather fast, yielding a high binding capacity.[20] Another example
of the immobilization process consists of hydrazone formation, when a 5’-aldehyde
derivatized ON was immobilized on large-pore hydrazide-functionalized Sephacryl
beads and this construct was used for the affinity purification of plasmid DNA via
triple-helix formation.[14] Again, the strategy consists of the immobilization of ON
containing electrophilic group on the 5’-end onto a support containing hydrazide
group. High coupling yield and little leakage of ON material were reported.
The hydrazone chemistry has proved to be a promising way to produce
oligodeoxyribonucleotide (ODN) microchips.[2,22–26] A high and reproducible
immobilization yield was reported.[2] The polyacrylamide gel of chip microcells
was activated by introduction of hydrazide groups, and ODNs were synthesized
with a 3’-terminal 3-methyluridine unit that was then oxidized by sodium periodate
to produce dialdehyde groups on the 3’-terminus.[22,23] Immobilization on
hydrazide support showed high efficiency (82–93%) and low nonspecific binding
(ꢀ1%).[2,24] With this technology it was reported to be possible to achieve very high
immobilization capacity,[22] two orders of magnitude higher than an average two-
dimensional one of a glass surface.[23]
It should nevertheless be mentioned that the hydrazone linkage was reported
not to provide sufficient stability of ODN attachment onto microchips for the case
when multiple hybridizations are required.[25] In these cases reductive amination is
suspected to be advantageous. A number of reductants may be employed here:
sodium borohydride, sodium cyanoborohydride, pyridine-borane complex,
trimethylamine-borane complex, and others.[2] But even without reduction 10–15
(up to 50.[22]) cycles of hybridization could be performed in the gel microchip with
reliable data output.[25,26] This is nevertheless in contrast to an older report where
reduction of acyl hydrazones immobilized on latex beads was found to be
mandatory.[11] Here the attachment yield of 5’-aldehyde derivatized ONs into
hydrazide latex beads is roughly 50 times higher when the immobilization takes
place in the presence of sodium cyanoborohydride. There are a few other examples
for the immobilization of ONs on supports with hydrazone chemistry, where higher
yields were reported in the presence of sodium cyanoborohydride.[20] So it is
suspected, that in some cases it may be of advantage to reduce the hydrazone bond.
Nevertheless, despite the fact that a few researchers prefer to use two-step
conjugation procedures that include a reduction step, in principle the hydrazone
linkage is sufficiently stable under the conditions suitable for the most applications.

214
S. I. Antsypovich and G. von Kiedrowski
SCHEME 1 The formation of a stable hydrazone bond. R, R’ are the residues of molecules (e.g., peptides,
proteins, fluorophores, oligonucleotides, DNA) or solid surfaces labeled with hydrazine and aldehyde
moieties, respectively.
Of course, the stability of the hydrazone linkage depends on its structure. For
example TriLink Biotechnologies offers aromatic aldehyde and aromatic hydrazine
modifiers for coupling of biomolecules that are able to form a very stable hydrazone
bond (Scheme 1) and does not require any further reduction to be stable.
In our opinion, even considering the possible profit of a reduction step, the
hydrazone chemistry has a number of evident advantages, as it uses exceedingly
mild reagents and there are virtually no competing side reactions. The only side
process reported consisted of a reaction of the aldehyde group with the heterocyclic
bases of DNA,[10] but it proceeds slower than the reaction of the more nucleophilic
hydrazide moiety. Hydrazone adducts are usually formed with moderate yield
(>80%) and reaction times vary from several minutes to several hours. Thus the
hydrazone formation (sometimes followed by reduction) looks more advantageous
than the coupling of aldehydes and amino derivatives which requires reductive
amination in all cases.
Next to immobilization, hydrazone formation-based methods are widely used
for conjugating ONs and various reporter molecules,[4,10,12,19,27,28] including
markers, dyes, and peptides. Most often 5’-modified ONs are obtained through
post-synthetic introduction of the hydrazide function, and they were reported to be
stable in solution. Examples include methods for the direct attachment of ON
probes to enzyme reporter systems.[4] High conjugation yields were reported,
typically 80–85% with 3.5-fold excess of the aldehyde-modified enzyme. One more
promising application of hydrazone chemistry is the possibility to create carbazoyl
internucleotide linkage.[19]
There is a wealth of data about the reaction of hydrazides with bis-aldehyde
derivatives of nucleosides, nucleotides and ONs yielding dihydroxymorpholines as
a main product.[10,12] Hydrazide-based reagents can even be used for the
identification and quantitative determination of periodate-oxidized ribonucleosides
and RNA molecules.[12,28] The resulting derivatives are usually rather stable at
neutral and slightly acidic pH values but are hydrolyzed under basic conditions in
the presence of ammonia or primary amines.[12] In contrast to hydrazides, the
reactions with alkyl hydrazines, semicarbazide, or with thiosemicarbazide
sometimes result in the formation of complex mixtures including acyclic mono-
and bis-hydrazones and other products depending on the nature of initial reagents
and the reaction conditions.[12] The excess of hydrazines can even lead to
decomposition of adducts and modified nucleoside residues as well.[10,12,29]

Synthesis of 5’- and 3’-Hydrazide Modified Oligonucleotides
215
Thus the hydrazone conjugation chemistry in solution provides a simple and
practically advantageous route for the generation of numerous ON probes.
Essentially any aldehyde-modified reporter label can be tagged to hydrazide
ONs.[4] In the case when the modification occurs at the terminus of an ON, it does
not interfere at all with base-pairing interactions. For non-terminal modifications,
the synthesis of a nucleoside analogue containing a hydrazide derived imidazole
residue as a nucleobase was reported.[27] It can be incorporated into ONs both
chemically and enzymatically. Conjugation with aldehydes permits the synthesis of
a large variety of novel nucleobases in a single step. The coupling reaction of this
hydrazide moiety with several aromatic aldehydes was reported to proceed very fast
(in several minutes) under the conditions chosen (at pH 4) with high yield.[27]
In summary, hydrazide-functionalized ONs are undoubtedly very promising
compounds, which clearly will find more and more applications in the near future.
At the present time, most known methods propose to the inclusion of the hydrazide
moiety into ON post-synthetically.[3,4] Nevertheless, in our opinion automated
phosphoramidite-based synthesis can be considered as a most convenient method to
introduce hydrazide moieties into ONs. For this purpose appropriately functional-
ized phosphoramidites and/or solid supports must be used. Recently DMT and trityl
protected hydrazide phosphoramidites were described for the 5’-functionalization of
ONs.[1,30] Nevertheless according to our knowledge, methods to introduce
hydrazide moieties into 3’-end of ONs employing standard amidites in the process
of automated chemical synthesis are still unknown. This stimulated us to design a
versatile phosphoramidite building block enabling the synthesis of either 3’- or 5’-
hydrazide modified ONs or even both in one sequence on the same support.
RESULTS
Our hydrazide modifier 8 (Scheme 2) was synthesized from readily available
compounds 1, 3, and 4. The synthetic route is straightforward and proved to be
rather efficient and easily scalable.
First, the 6-hydroxyhexanoyl hydrazide (2) was obtained by treatment of 6-
caprolactone monomer (1) by hydrazine hydrate. Bis-2-hydroxyethylsulfone (3) was
derivatized with a double excess of 4-nitrophenylchloroformate (4) and activated
derivative 5 was treated by the hydroxyhydrazide 2 to yield the key adduct 6. The
latter was monomethoxytritylated to yield derivative 7 and then phosphitylated
with 2-cyanoethyl-N,N,N’,N’-tetraisopropyl phosphane to give the final phosphor-
amidite derivative 8.
6-Hydroxyhexanoylhydrazide (2) was obtained with quantitative yield. Among
a wide variety of derivatizing reagents, 4-nitrophenylchloroformate (4) was chosen
due to high reactivity and facile work-up conditions. No degradation of active
compound 5 was detected during washings or following flash chromatography.
Adduct 6 was synthesized in DMF as a solvent due to hydrophilic nature of both
reagents and a product; DMAP was added as a catalyst of acylation in the reaction

216
S. I. Antsypovich and G. von Kiedrowski
SCHEME 2 The synthesis of hydrazide modifier 8.
mixture. The MMT group was chosen for hydroxyl protection instead of DMT to
provide additional stability to the final derivative. Bannwarth’s reagent was
employed as a selective and mild phosphitylating reagent, thus avoiding tertiary
amines in the reaction mixture, which may provoke the decomposition of
derivatives 7 and 8 by b-elimination.[31,32]
Phosphoramidite 8 proved to be a straight-forward modifier in the synthesis of
hydrazide modified ONs. A standard protocol of ODN synthesis could be applied
except for the introduction of the modifier itself which required slightly longer
coupling times (4 min) and increased concentration of phosphoramidite solution

Synthesis of 5’- and 3’-Hydrazide Modified Oligonucleotides
217
TABLE 1 Sequences of Hydrazide Modified ODNs Obtained
No
Abbr.
Length
ODN sequence (5’ ! 3’)
TTT TTT TTT T X
1
AS1
10
2
3
4
5
AS2
AS3
AS4
AS5
13
15
15
15
X ATT AGA CGC CAT C
X ATT AGA CGC CAT CTA
X TGG TAG CCG CTA GAT
X ATT ATG GAC CGA TTT
X ATT ATG GAC CG
6
AS6
11
7
AS7
11
ATT ATG GAC CG X
8
9
ASK4
ASK5
ASK3
ASK32
AS32
ASD1 (ASK4 & AS6)
ASD2 (ASK5 & AS2)
ASD3 (AS7 & AS4)
15
15
15
15
ATC TAG CGG CTA CCA X
AAA TCG GTC CAT AAT X
TAG ATG GCG TCT AAT X
X TAG ATG GCG TCT AAT X
X ATT AGA CGC CAT CTA X
ATC TAG CGG CTA CCA X ATT ATG GAC CG
AAA TCG GTC CAT AAT X ATT AGA CGC CAT C
ATT ATG GAC CG X TGG TAG CCG CTA GAT
10
11
12
13*
14*
15*
15
15 & 11
15 & 13
11 & 15
X is a residue of hydrazide modifier.
*Combination sequences splitting into some of ODNs # 1–12 during post-synthetic work-up.
(0.12 M) to achieve a coupling yield of >90%. We observed a minor decrease of
coupling efficiency for the subsequent standard amidites. The latter was, however,
not considered as a problem requiring exhaustive optimization.
Fifteen modified ONs with different length and sequence were synthesized
employing phosphoramidite 8 (Table 1). The standard post-synthetic procedure
SCHEME 3 The formation of hydrazide containing aligonucleotides upon post-synthetic procedure. Ac: an acyl
protecting group; support: Primer Support (Pharmacia); Base, Base’: Ade, Gua, Thy, Cyt.

218
S. I. Antsypovich and G. von Kiedrowski
was applied to deprotect modified ODNs and cleave them from solid support
(Scheme 3). The structures of oligonucleotides obtained were confirmed by matrix
assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI TOF
MS) analysis.
We expected that a side reaction caused by ammonia nucleophilic attack at the
carbonyl group of the hydrazide moiety could probably take place. It may yield
side products containing an amide or an acid functionality instead of the hydrazide.
But actually only a minor amount of side products during deprotection with
aqueous ammonia was detected by MALDI TOF MS and HPLC. The formation of
side products depends on the time of ammonia treatment and the concentration of
ammonia solution as well as on the temperature of the process. As a rule of thumb
we used 5 h at 55°C for conc. ammonia and did not observe more than 5–10% of
side products. Further shortening of ammonia treatment did not result in a
significant improvement. We did not test a more labile protection scheme such as
possible using the commercially available PAC phosphoramidites to enable a
shortening of ammonia treatment under milder conditions. The latter, however,
should allow a decrease in the amount of side products down to trace levels. Also,
with such phosphoramidites it should be possible to use hydrazine-based
deprotection since the typical side reaction of standard amidites, hydrazination of
the N4-position of benzoyl cytidine, can be avoided here due to the nature of the
protection groups. It was reported, that the attack of hydrazine either at the amide
carbonyl group or at the C4 of the pyrimidine ring depends on the electronic
structure of the amide.[33,34]
As a further deprotection method we investigated the procedure originally
developed by Mirzabekov et al. for processing of ODNs synthesized on a biochip
surface.[35] It includes the treatment of support-bound ODNs by lithium hydroxide
FIGURE 1 MALDI TOF MS analysis of 5’-hydrazide modified ODN # 3 (AS3, 5’-X-ATT AGA CGC CAT
CTA-3’). X is a residue of hydrazide modifier. MW calculated for [M + Na]ꢁ: 4765.10; found: 4763.50. For
conditions see Experimental.

Synthesis of 5’- and 3’-Hydrazide Modified Oligonucleotides
219
in the presence of triethylamine and methanol. We considered this procedure to be
advantageous in our case as it circumvents ammonia treatment. We found,
however, that the outcome is rather similar to ammonia deprotection.
Most of the ODNs were synthesized in trityl-ON mode. ODNs containing mod-
ification on the 3’-end were purified by reverse-phase HPLC using the hydrophobic
properties of DMT group. ODNs containing the hydrazide modification at the
5’-end do not carry any hydrophobic protecting group. Nevertheless, it was found
that it is possible to purify these ODNs by reversed phase HPLC due to presence of
a residue of the hydrazide modifier on 5’-terminus which causes substantial shift of
retention time. All ONs were synthesized in 1.3 mmole scale, but the whole
procedure is easily scalable. Under the conditions chosen the overall yield of hy-
drazide ODNs was typically not less than 50%. Retention times of modified ODNs
proved to be different from those of unmodified ODNs of the same sequences,
FIGURE 2 MALDI TOF MS analyses of conjugates of 5’-modified ODNs with anisaldehyde. a. Conjugate of
ODN # 3 (AS3, 5’-X-ATT AGA CGC CAT CTA-3’) with anisaldehyde. MW calculated for [M]ꢁ: 4861.25; found:
4861.23. b. Conjugate of ODN #4 (AS4, 5’-X-TGG TAG CCG CTA GAT-3’) with anisaldehyde. MW calculated
for [M]ꢁ: 4933.30; found: 4934.92. X is a residue of hydrazide modifier. For conditions see Experimental.

220
S. I. Antsypovich and G. von Kiedrowski
both for DMT-containing ODNs and for ODNs without the DMT group. The
identities of synthetic ODNs containing the hydrazide functionality were con-
firmed by MALDI TOF MS (Figure 1). Analysis using negative-ion MALDI
TOF MS revealed experimental masses that are in agreement with theory.
The additional reliable evidence of the presence and chemical reactivity of the
hydrazide moiety in ODNs came from the successful conjugation with
anisaldehyde. The reaction was found to proceed rather fast and could be
monitored by reverse-phase HPLC. The conjugates of modified ODNs with
anisaldehyde were isolated by reverse-phase HPLC and structures were confirmed
by MALDI TOF MS analysis (Figure 2).
EXPERIMENTAL PART
General
Unless otherwise indicated, all reactions were performed under dry argon.
Reagents were obtained from Aldrich or Acros Organics Chemical Companies.
Column chromatography was accomplished using silica gel 60, 230–400 mesh
(Merck). ODN samples were evaporated using a Speed Vac Plus 110 A
concentrator (Savant).
1H-NMR spectra were recorded at 200.131 MHz with the Bruker DPX-200
spectrometer. 1H-NMR spectra are reported in units of d using tetramethyl silane as
1
an internal standard and coupling constants are reported in units of Hz. H-NMR
spectrum of the phosphoramidite 8 was recorded at 400.131 MHz and its 31P-NMR
spectrum was recorded at 161.992 MHz with the Bruker DRX-400 spectrometer.
ESI mass spectra were recorded with the Bruker Esquire LC 00153 spectrometer in
positive ionization mode. The MALDI TOF MS measurements were performed
with the Perseptive Voyager DE-RP and the Bruker Daltonics Autoflex instruments
in negative ionization mode. Samples were irradiated with a nitrogen laser operated
at 337 nm. Ions produced by laser desorption were energetically stabilized during a
delayed extraction period of 300 nanoseconds and then accelerated through the
linear TOF mass analyzer by 25 kV potential pulse or by 20kV in Bruker Autoflex.
As a matrix 3-hydroxypicolinic acid in a mixture of acetonitrile and water (1:1 v/v)
was used.
HPLC analyses and purifications were performed with Kontron HPLC system
in reverse-phase mode. The CC 250/4 Nucleodur 100-5 C18 ec column (Macherey-
Nagel) was used (size 4 ꢂ 250 mm, particle size 7 mm). Flow rate 1.0 ml/min at
room temperature was applied. 0.1 M Ammonium hydrogen carbonate buffer was
used as mobile phase A and 100% acetonitrile was used as mobile phase B. For
DMT group containing ODNs usually the gradient of 5% to 35% of mobile phase B
in 30 min (1% of mobile phase B per minute) was used; for ODNs without DMT
group usually the gradient of 5% to 20% of mobile phase B in 30 min (0.5% of mobile
phase B per minute) was used.

Synthesis of 5’- and 3’-Hydrazide Modified Oligonucleotides
221
The Synthesis of Phosphoramidite Building Block for the
Incorporation of Hydrazide Into ONs
6-Hydroxyhexanoylhydrazide (2). Two hundred millimoles of
aqueous hydrazine (35 wt. %) was slowly added to a solution of 21.2 ml (200
mmole) of 6-caprolactone monomer (1) in 200 ml of ethanol under vigorous
stirring. The reaction mixture was heated to reflux for 12 h and then cooled to room
temperature. The obtained colorless crystals were recrystallized from hot ethanol
and dried under vacuum to afford 28.4 g (97%) of 6-hydroxyhexanoylhydrazide (2).
1H-NMR (D2O): 1.31 (2H, m, CH2–(CH2)2–OH); 1.57 (4H, m, CH2–CH2–CH2–
CH2–OH); 2.21 (2H, t, CO–CH2, J=7.26 Hz); 3.58 (2H, t, CH2–OH, J=6.44 Hz).
ESI MS: C6H14N2O2, [M + Na]+ calculated: 169.18; found: 169.1.
Bis-[(4-nitrophenyl)oxycarbonyl-2-oxyethyl]sulfone
(5). Twenty-five grams (124 mmol) of 4-nitrophenylchloroformate (4) was
added to a solution of 8.63 g (56 mmol) of bis-2-hydroxyethylsulfone (3) in 300 ml
of anhydrous methylene chloride. Then 10.05 ml (125 mmol) of anhydrous
pyridine was slowly added dropwise under ice cooling and vigorous stirring. The
reaction mixture was vortexed for 1 h at room temperature. The precipitate was
filtered, washed twice with water, and dried under vacuum. A second fraction
was obtained from the organic phase. It was washed twice with water and dried
over anhydrous sodium sulfate. Organic phase and precipitate were combined
and concentrated in vacuo. The residue was purified by flash column
chromatography and eluted with a step gradient of 0–25% (v/v) of ethanol in
methylene chloride. Product containing fractions were concentrated and dried in
1
vacuo yielded 17.67 g (66.4%) of compound 5. H-NMR (DMSO-D6): 3.75 (4H, t,
CH2–SO2); 4.66 (4H, t, OCO–CH2, J=5.56 Hz); 7.55 (4H, d, OCO–C–CH); 8.31
(4H, d, NO2–C–CH, J=8.71 Hz). ESI MS: C18H16N2O12S, [M + NH4]+
calculated: 502.43; found: 503.1.
Bis-[(6-hydroxyhexanoyl)hydrazinocarbonyl-2-oxyethyl]sul-
fone (6). To a solution of 5.95 g (40.7 mmol) of 6-hydroxyhexanoylhydrazide
(2) in 350 ml of anhydrous DMF, 9.7 g (20 mmol) of bis-[(4-nitrophenyl)ox-
ycarbonyl-2-oxyethyl]sulfone (5) was added. Then, 1620 ml (20 mmol) of
anhydrous pyridine was slowly added dropwise under vigorous stirring followed
by the addition of 0.05 eq. of DMAP. The reaction mixture was vortexed during
20 h at room temperature. The DMF was evaporated in vacuo and 200 ml of
methylene chloride was added to reaction mixture, followed by 10 ml of acetic
acid and 200 ml of water. The product was extracted with water. The aqueous
layer was washed once with 50 ml of methylene chloride with 3 ml of acetic acid.
The solution was evaporated in vacuo and coevaporated twice with anhydrous
acetonitrile. The product was recrystallized from hot acetonitrile and dried in
1
vacuo to afford 6.5 g (65.2%) of compound 6. H-NMR (D2O): 1.38 (4H, m,

222
S. I. Antsypovich and G. von Kiedrowski
CH2–(CH2)2–OH); 1.58 (8H, m, CH2–CH2–CH2–CH2–OH); 2.31 (4H, t, CO–
CH2, J=7.07 Hz); 3.59 (4H, t, CH2–SO2, J=6.37 Hz); 3.68 (4H, bs, CH2–OH); 4.61
(4H, t, OCO–CH2). ESI MS: C18H34N4O10S, [M + Na]+ calculated: 521.54;
found: 521.2.
6-Monomethoxytrityloxy-6’-hydroxy-bis-(hexanoylhydrazino-
carbonyl-2-oxyethyl)sulfone (7). The solution of 2.93 g (9.5 mmol)
monomethoxytrityl chloride in 100 ml of anhydrous methylene chloride was
slowly added dropwise, while stirring vigorously, to a solution of 5 g (10 mmol) of
bis-[(6-hydroxyhexanoyl)hydrazinocarbonyl-2-oxyethyl]sulfone (6) in 120 ml of
anhydrous pyridine. The reaction mixture was vortexed during 3 h at room
temperature. The solvents were evaporated in vacuo and 100 ml of ethyl acetate
was added. The organic phase was washed once with 100 ml of brine, once with
100 ml of water, and dried over anhydrous sodium sulfate. After evaporation of
the solvent in vacuo the residue was purified by flash column chromatography,
eluting with a step gradient of 0–5% (v/v) of ethanol in methylene chloride
containing 0.5% (v/v) of pyridine. Product containing fractions were concentrated,
coevaporated with 20 ml of benzene for three times and dried in vacuo yielding
1
3.64 g (49.7%) of compound 7. H-NMR (DMSO-D6): 1.33 (4H, m, CH2–(CH2)2–
OH, CH2–(CH2)2–O–MMT); 1.50 (8H, m, CH2–CH2–CH2–CH2–OH, CH2–
CH2–CH2–CH2–O–MMT); 2.07 (4H, bs, NHCO–CH2); 2.92 (4H, t, CH2–SO2,
J=6.00 Hz); 3.54 (4H, bt, CH2–OH, CH2–O–MMT); 3.73 (3H, s, O–CH3); 4.34
(4H, t, OCO–CH2); 6.88 (2H, d, CH3O–C–CH); 7.23 (2H, d, CH3O–C–
CH=CH, J=8.53 Hz); 7.34 (10H, m, C6H5). ESI MS: C38H50N4O11S, [M + NH4]+
calculated: 788.93; found: 788.4.
6-Monomethoxytrityloxy-6’-[(2-cyanoethoxy)(diisopropyla-
mino)phosphanyloxy]-bis-(hexanoylhydrazinocarbonyl-2-
oxyethyl)sulfone (8). To a solution of 0.463 g (0.6 mmol) of 6-
monomethoxytrityloxy-6’-hydroxy-bis-(hexanoylhydrazinocarbonyl-2-oxyethyl)sul-
fone (7) and 0.334 g (1.95 mmol) of diisopropylamidotetrazolide in 15 ml of
anhydrous methylene chloride, a solution of 572 ml (1.8 mmole) of 2-cyanoethyl-
N,N,N’,N’-tetraisopropyl phosphane in 10 ml of anhydrous methylene chloride was
added dropwise via a syringe over a period of 5 min under vigorous stirring. The
reaction mixture was vortexed for 2 h at room temperature. Then mixture was
diluted with 125 ml of ethyl acetate and washed once with 150 ml of brine and once
with 150 ml of water. The organic phase was dried over anhydrous sodium sulfate,
evaporated in vacuo, and dried under vacuum, affording 0.54 g (92.7%) of
compound 8 as a light-yellow foam. As TLC analysis showed complete
consumption of initial compound 7, the resulting phosphoramidite 8 was used in
1
ODN synthesis without further purification. H-NMR (CDCl3): 1.18 (12H, bd,
((CH3)2CH)2N-, J=6.52 Hz); 1.24; 1.29; 1.40; 1.62 (12H, m, CH2–CH2–CH2–CH2–
O–P, CH2–CH2–CH2–CH2–O–MMT); 2.23 (4H, bt, NHCO–CH2, J=5.52 Hz);

Synthesis of 5’- and 3’-Hydrazide Modified Oligonucleotides
223
2.60 (2H, t, NC–CH2–CH2–O–, J=6.02 Hz); 3.06 (4H, bt, CH2–SO2, J=6.53 Hz);
3.30; 3.42; 3.58 (6H, m, ((CH3)2CH)2N–P, (CH2)4–CH2–O–P, (CH2)4–CH2–O–
MMT); 3.71 (2H, m, NC–CH2–CH2–O); 3.80 (3H, s, O–CH3); 4.54 (4H, bt, OCO–
CH2); 7.21; 7.27; 7.42 (14H, m, C6H4, C6H5). 31P-NMR (CDCl3): 147.49. ESI MS:
C47H67N6O12PS, [M + H]+ calculated: 972.12; found: 971.3.
The Synthesis of Hydrazide Containing ONs
ODNs were synthesized using solid-phase phosphoramidite chemistry on an
automated nucleic acids synthesizer Gene Assembler from Pharmacia Biotech in a
1.3 mmol scale. ‘‘Primer Supports’’ from Pharmacia Biotech (50–70 mm solid-phase
polystyrene bead matrix) were used as solid support. 4,5-Dicyanoimidazole in
acetonitrile was used as activator. The hydrazide containing phosphoramidite 8 was
applied as a 0.12 M solution in anhydrous acetonitrile. It was coupled at desired
positions in ODN sequence using 4 min coupling time. No other changes were
made in the standard protocol of phosphoramidite ODN synthesis.
Deprotection and Isolation of Hydrazide Containing ODNs
Two different deprotection and work-up strategies were applied.
Standard Protocol
The solid support bound ODN was placed in a 2-ml microtube with screw cap
(Sarstedt) and treated by 0.5 ml of 25% aqueous ammonia solution at 55°C. After
5 h the test tube was cooled to room temperature, ammonia solution was removed,
evaporated to dryness under reduced pressure, and the residue was redissolved into
0.5 ml of water.
Alternative Protocol
The solid support bound ODN was placed in a 2-ml test tube and 30 ml of
0.5 M solution of lithium hydroxide in water and then 300 ml of 3.5 M solution of
triethylamine in methanol were added. The reaction mixture was vortexed during
1 h at 75°C. 75 ml of glacial acetic acid was added at room temperature and the
solution was cooled at ꢁ 20°C for 5 min. After centrifugation the supernatant was
discarded, the precipitate was washed twice with 90% (v/v) acetonitrile in water,
twice with acetonitrile, and then redissolved in 0.5 ml of water.
The DMT group containing ODNs were analyzed and purified by reverse-
phase HPLC using the conditions described above. The DMT protecting group
was removed by treating the ODN with 80% acetic acid for 30 min at room
temperature. 500 ml of 50% (v/v) aqueous methanol was added to the acidic solution
which was then evaporated in vacuo. ODN was redissolved in 0.5 ml of water.
ODNs with free 5’-hydroxyl group were analyzed and purified by reversed phase
HPLC using the conditions described above.

224
S. I. Antsypovich and G. von Kiedrowski
SCHEME 4 Phosphoramidite modifier developed by Nanogen Recognomics.
Conjugation of Hydrazide Containing ODNs and Aldehyde
Five nanomoles of the hydrazide containing ODN was evaporated to dryness
and then dissolved in 10 ml of 0.1 M MES buffer (pH 5.0–5.3), 4 ml of a 1 M
solution of 4-methoxyphenylaldehyde in DMF was added. After 30 or 60 min at 30
to 37°C, ODN was desalted in NAP-10 Sephadex cartridge (Pharmacia Biotech).
An aqueous solution of ammonium acetate (1 mM) was used for the elution. The
ODN solution was concentrated under reduced pressure to a volume of 0.4 ml and
the reaction mixture was analyzed by reverse-phase HPLC in conditions described
above. The conjugates were isolated from the reaction mixtures under the same
conditions. Fractions containing conjugates were collected, evaporated to dryness
in vacuo, redissolved in 500 ml of water, and analyzed by HPLC and MALDI TOF
MS using conditions as described above.
CONCLUSIONS
Our method for the introduction of the hydrazide moiety into ONs was found to
be rather efficient. It is based on a versatile phosphoramidite building block
8 containing two hydrazide residues connected to b-eliminable unit, whose synthesis
(Scheme 2) is quick and straight-forward, starting from inexpensive commercially
available building blocks. Our phosphoramidite modifier can be employed in the
standard ON synthesis process and yield hydrazide containing ONs upon ammonia
or lithium hydroxide treatment. The hydrazide functions are liberated under the
mild conditions of a b-elimination yielding only low amount of side products. In
most cases the hydrazide modified ONs should nevertheless be purified using
reverse-phase HPLC. For the case of 3’-modified ONs it is further recommended to
keep the 5’-DMT group ‘‘on,’’ especially if one uses our method to prepare in a
single run two hydrazide-modified ONs of similar length. For the case of 5’-modified
ODNs, which are obtained without a lipophilic DMT group, it was nevertheless
quite easy to achieve the required level of ON purity using reverse-phase HPLC.
Recently, 5’-hydrazide modified ONs became commercially available from
Metabion company. Their synthesis is based on the modifier amidite 9 (Scheme 4)
developed by Nanogen Recognomics GmbH,[1] which formally allows to introduce

Synthesis of 5’- and 3’-Hydrazide Modified Oligonucleotides
225
the hydrazide moiety also at the 3’-end when employing reversed phosphoramidites
whose cost and coupling efficiency limit a wider applicability.
Our modifier has the advantage of being compatible with standard ON-
synthesis when 3’-modified ONs are required. Both 3’-and 5’-modified ONs are
easily accessible using our hydrazide modifier and may find particular applications
in the field of DNA nanotechnology.[36–38]
REFERENCES
1. Raddatz, S.; Mueller-Ibeler, J.; Kluge, J.; Wass, L.; Burdinski, G.; Havens, J.R.; Onofrey, T.J.; Wang, D.;
Schweitzer, M. Hydrazide oligonucleotides: new chemical modification for chip array attachment and
conjugation. Nucleic Acids Res. 2002, 30(21), 4793–4802.
2. Timofeev, E.N.; Kochetkova, S.V.; Mirzabekov, A.D.; Florentiev, V.L. Regioselective immobilization of short
oligonucleotides to acrylic copolymer gels. Nucleic Acids Res. 1996, 24(16), 3142–3148.
3. Grimm, G.N.; Boutorine, A.S.; Helene, C. Rapid routes of synthesis of oligonucleotide conjugates from non-
protected oligonucleotides and ligands possessing different nucleophilic or electrophilic functional groups.
Nucleosides, Nucleotides Nucleic acids 2000, 19(10–12), 1943–1965.
4. Ghosh, S.S.; Kao, P.M.; Kwoh, D.Y. Synthesis of 5’-oligonucleotide hydrazide derivatives and their use in
preparation of enzyme-nucleic acid hybridization probes. Anal. Biochem. 1989, 178(1), 43–51.
5. Podyminogin, M.A.; Lukhtanov, E.A.; Reed, M.W. Attachment of benzaldehyde-modified oligodeoxynu-
cleotide probes to semicarbazide-coated glass. Nucleic Acids Res. 2001, 29(24), 5090–5098.
6. Murata-Kamiya, N.; Kamiya, H. Methylglyoxal, an endogenous aldehyde, crosslinks DNA polymerase and
the substrate DANN. Nucleic Acids Res. 2001, 29(16), 3433–3438.
7. Timofeev, E.; Mirzabekov, A. Binding specificity and stability of duplexes formed by modified
oligonucleotides with a 4096-hexanucleotide microarray. Nucleic Acids Res. 2001, 29(12), 2626–2634.
8. Yasui, M.; Matsui, S.; Ihara, M.; Laxmi, Y.R.S.; Shibutani, S.; Matsuda, T. Translesional synthesis on a DNA
template containing N2-methyl-2’-deoxyguanosine catalyzed by the Klenow fragment of Escherichia coli DNA
polymerase I. Nucleic Acids Res. 2001, 29(9), 1994–2001.
9. Vaca, C.E.; Fang, J.L.; Schweda, E.K.H. Studies of the reaction of acetaldehyde with deoxynucleosides.
Chem. Biol. Interact. 1995, 98(1), 51–67.
10. Shabarova, Z.A.; Bogdanov, A.A. Advanced Organic Chemistry of Nucleic Acids. VCH Verlagsgesellschaft:
Weinheim, 1994.
11. Kremsky, J.N.; Wooters, J.L.; Dougherty, J.P.; Meyers, R.E.; Collins, M.; Brown, E.L. Immobilization of DNA
via oligonucleotides containing an aldehyde or carboxylic acid group at the 5’ terminus. Nucleic Acids Res.
1987, 15(7), 2891–2909.
12. Gritsenko, O.M.; Gromova, E.S. Dialdehyde-containing nucleic acid and their components: synthesis,
properties and affinity modification of proteins. Russ. Chem. Rev. 1999, 68(3), 241–251.
13. Turkova, J. Bioaffinity Chromatography. Elsevier: Amsterdam, 1993.
14. Schluep, T.; Cooney, C.L. Immobilization of oligonucleotides on a large pore support for plasmid purification
by triplex affinity interaction. Bioseparations 1999, 7(6), 317–326.
15. Hermanson, G.T. Bioconjugate Techniques. Academic Press: San Diego, CA, 1996.
16. Gershoni, J.M.; Bayer, E.A.; Wilchek, M. Blot analyses of glycoconjugates: enzyme-hydrazide—a novel
reagent for the detection of aldehydes. Anal. Biochem. 1985, 146(1), 59–63.
17. Heimgartner, U.; Kozulic, B.; Mosbach, K. Polyacrylic polyhydrazides as reagents for detection of
glycoproteins. Anal. Biochem. 1989, 181(1), 182–189.
18. Kniep, B.; Muhlradt, P.E. Immunochemical detection of glycosphingolipids on thin-layer chromatograms.
Anal. Biochem. 1990, 188(1), 5–8.
19. Magdalena, J.; Fernandez, S.; Ferrero, M.; Gotor, V. Synthesis of novel carbazoyl linked pyrimidine-
pyrimidine and pyrimidine-purine dinucleotide analogues. Tetrahedron Lett. 1999, 40(9), 1787–1790.
20. Wilchek, M.; Lamed, R. Immobilized nucleotides for affinity chromatography. In Methods in Enzymology,
XXXIV, Affinity Techniques, Enzyme Purification, Part B. Academic Press: London, 1974; 475–479.
21. Barker, R.; Trayer, I.P.; Hill, R.L. Nucleosides phosphates attached to agarose. In Methods in Enzymology,
XXXIV, Affinity Techniques, Enzyme Purification, Part B. Academic Press: London, 1974; 479–491.

226
S. I. Antsypovich and G. von Kiedrowski
22. Yershov, G.; Barsky, V.; Belgovskiy, A.; Kirillov, E.; Kreindlin, E.; Ivanov, I.; Parinov, S.; Guschin, D.;
Drobishev, A.; Dubiley, S.; Mirzabekov, A. DNA analysis and diagnostics on oligonucleotide microchips.
Proc. Natl. Acad. Sci. U. S. A. 1996, 93(10), 4913–4918.
23. Khrapko, K.R.; Lysov, Yu.P.; Khorlin, A.A.; Ivanov, I.B.; Yershov, G.M.; Vasilenko, S.K.; Florentiev, V.L.;
Mirzabekov, A.D. A method for DNA sequencing by hybridization with oligonucleotide matrix. DNA
Sequence 1991, 1(6), 375–388.
24. Ivanov, I.B.; Khrapko, K.R.; Chernov, B.K.; Khorlin, A.A.; Lysov, Yu.P.; Florentiev, V.L.; Mirzabekov, A.D.
Hybridization properties of gel-immobilized oligonucleotides. Nucleic Acids Res. Symp. Ser. 1991, 24, 189–
190.
25. Ivanov, I.B.; Yershov, G.M.; Barsky, V.E.; Belgovskiy, A.I.; Kirillov, E.V.; Kreindlin, E.Y.; Parinov, S.V.;
Mologina, N.V.; Mirzabekov, A.D. Diagnostics of genetic mutations on oligonucleotide microchips.
Molekulyarnaya Biologiya (Russ.) 1997, 31(1), 159–167.
26. Khrapko, K.R.; Lysov, Yu.P.; Khorlin, A.A.; Shick, V.V.; Florentiev, V.L.; Mirzabekov, A.D. An
oligonucleotide hybridization approach to DNA sequencing. FEBS Lett. 1989, 256(1–2), 118–122.
27. Strobel, H.; Dugue, L.; Marliere, P.; Pochet, S. Synthesis and recognition by DNA polymerases of a reactive
nucleoside, 1-(2-deoxy-b-D-erythro-pentofuranosyl)-imidazole-4-hydrazide. Nucleic Acids Res. 2002, 30(9),
1869–1878.
28. Randerath, K. 3-([3-125I]Iodo-4-hydroxyphenyl)propionyl carbohydrazide, a new radioiodination reagent for
ultrasensitive detection and determination of periodate-oxidized nucleoside derivatives and other carbonyl
compounds. Anal. Biochem. 1981, 115(2), 391–397.
29. Khym, J.X.; Cohn, W.E. Characterizations and some chemical reactions of periodate-oxidized nucleosides.
J. Am. Chem. Soc. 1960, 82(24), 6380–6386.
30. Kurz, M.; Lohse, P.; Wagner, R. Peptide Acceptor Ligation Methods. US Patent 6, 429, 300, August 6, 2002.
31. Braich, R.S.; Damha, M. Regiospecific solid-phase synthesis of branched oligonucleotides. Effect of vicinal
2’,5’- (or 2’,3’-) and 3’,5’-phosphodiester linkages on the formation of hairpin DNA. J. Bioconjugate Chem.
1997, 8(3), 370–377.
32. Peyrottes, S.; Mestre, B.; Burlina, F.; Gait, M.J. The synthesis of peptide-oligonucleotide conjugates by a
fragment coupling approach. Tetrahedron 1998, 54(41), 12513–12522.
33. Polushin, N.N.; Morocho, A.M.; Chen, B.C.; Cohen, J.S. On the rapid deprotection of synthetic
oligonucleotides and analogs. Nucleic Acids Res. 1994, 22(4), 639–645.
34. Polushin, N.N.; Pashkova, I.N.; Chakhmakhcheva, O.G.; Efimov, V.A. Use of hydrazine for rapid deblocking
of synthetic oligonucleotides. Bioorgan. Khimiya 1993, 19(3), 318–326.
35. Surzhikov, S.A.; Timofeev, E.N.; Chernov, B.K.; Golova, J.B.; Mirzabekov, A.D. Advanced method for
oligonucleotide deprotection. Nucleic Acids Res. 2000, 28(8), NAR Methods Online, e29, i–v.
36. Scheffler, M.; Dorenbeck, A.; Jordan, S.; Wustefeld, M.; von Kiedrowski, G. Self-assembly of
trisoligonucleotidyls: the case for nano-acetylene and nano-cyclobutadiene. Angew. Chem. Int. Ed. Engl.
1999, 38(22), 3311–3315.
37. Sievers, D.; von Kiedrowski, G. Self-replication of complementary nucleotide-based oligomers. Nature 1994,
369(6477), 221–224.
38. Luther, A.; Brandsch, R.; von Kiedrowski, G. Surface-promoted replication and exponential amplification of
DNA analogues. Nature 1998, 396(6708), 245–248.