Zwitterionic oligonucleotides: A study on binding properties of 2′-O-aminohexyl modifications

Article abstract of DOI:10.1081/NCN-200067400

2′-O-Aminohexyl side chains provide excellent conditions for zwitterionic interstrand and intrastrand interactions of oligonucleotides. 2′-O-Aminoalkylated phosphoramidites of adenosine and undine were synthesized and incorporated in increasing number int

Full text of DOI:10.1081/NCN-200067400

This article was downloaded by: [Eindhoven Technical University]
On: 15 November 2014, At: 19:49
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
Zwitterionic Oligonucleotides: A Study on Binding
Properties of 2′-O-Aminohexyl Modifications
Christian R. Noe a b , Johannes Winkler a , Ernst Urban a , Matthias Gilbert b , Georg
Haberhauer c & Helmut Brunar b
a Department für Medizinische/Pharmazeutische Chemie , Universität Wien,
Pharmaziezentrum , Wien, Austria
b Institut für Pharmazeutische Chemie , Johann Wolfgang Goethe-Universität, Biozentrum ,
Frankfurt, Germany
c Austrian Research Center , Seibersdorf, Austria
Published online: 31 Aug 2006.
To cite this article: Christian R. Noe , Johannes Winkler , Ernst Urban , Matthias Gilbert , Georg Haberhauer & Helmut
Brunar (2005) Zwitterionic Oligonucleotides: A Study on Binding Properties of 2′-O-Aminohexyl Modifications, Nucleosides,
Nucleotides and Nucleic Acids, 24:8, 1167-1185, DOI: 10.1081/NCN-200067400
To link to this article: http://dx.doi.org/10.1081/NCN-200067400
PLEASE SCROLL DOWN FOR ARTICLE
Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained
in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no
representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the
Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and
are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and
should be independently verified with primary sources of information. Taylor and Francis shall not be liable for
any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever
or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of
the Content.
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. Terms & Conditions of access and use can be found at http://
www.tandfonline.com/page/terms-and-conditions

Nucleosides, Nucleotides, and Nucleic Acids, 24 (8):1167–1185, (2005)
Copyright D Taylor & Francis, Inc.
ISSN: 1525-7770 print/ 1532-2335 online
DOI: 10.1081/NCN-200067400
ZWITTERIONIC OLIGONUCLEOTIDES: A STUDY ON BINDING
PROPERTIES OF 2’-O-AMINOHEXYL MODIFICATIONS
5
Christian R. Noe
Department fu¨ r Medizinische/Pharmazeutische Chemie, Universita¨t
Wien, Pharmaziezentrum, Wien, Austria and Institut fu¨ r Pharmazeutische Chemie, Johann
Wolfgang Goethe-Universita¨t, Biozentrum, Frankfurt, Germany
5
Johannes Winkler and Ernst Urban
Department fu¨ r Medizinische/Pharmazeutische
Chemie, Universita¨t Wien, Pharmaziezentrum, Wien, Austria
5
Matthias Gilbert
Institut fu¨ r Pharmazeutische Chemie, Johann Wolfgang Goethe-
Universita¨t, Biozentrum, Frankfurt, Germany
5
Georg Haberhauer
Austrian Research Center, Seibersdorf, Austria
5
Helmut Brunar
Institut fu¨ r Pharmazeutische Chemie, Johann Wolfgang Goethe-
Universita¨t, Biozentrum, Frankfurt, Germany
5
2’-O-Aminohexyl side chains provide excellent conditions for zwitterionic interstrand and intrastrand
interactions of oligonucleotides. 2’-O-Aminoalkylated phosphoramidites of adenosine and uridine were
synthesized and incorporated in increasing number into homo adenosine and homo uridine/thymidine
dodecamers, respectively. CD spectra of these dodecamers with complementary sense DNA exhibited a B-
DNA type structure. While duplex stability values of all tested oligonucleotides were lower than those of
the native oligonucleotides, they were significantly higher than those of 2’-O-heptyl modified
oligonucleotides. The destabilization amounted to 0.9, 1.5, and 2.7°C per modification for 2’-O-
aminohexyl adenosine, 2’-O-aminohexyl uridine, and 2’-O-heptyl adenosine substitutions. These findings
are pointing to a duplex stabilizing effect of the interaction of side chain amino groups with backbone
phosphoric acid.
INTRODUCTION
For about 20 years, gene therapy has been an important issue in the field of
therapy of diseases. Somatic gene therapy would bear the attribute of a perfect
therapy, if implemented in its theoretical perfection—a kind of molecular surgery.
Accepted 22 March 2005.
Address correspondence to Christian R. Noe, Department fu¨r Medizinische/Pharmazeutische Chemie,
Universit¨at Wien, Pharmaziezentrum, Althanstrasse 14, Wien 1090, Austria; Fax: +431-4277-9551; E-mail:
christian.noe@univie.ac.at
1167
Order reprints of this article at www.copyright.rightslink.com

1168
C. R. Noe et al.
Not surprisingly, it was considered to be the ultimate goal during the first decade of
therapeutic biotech research. Antisense drug therapy, the other gene therapeutic
approach, is basically a hybrid between somatic gene therapy and drug therapy.
On the one hand, silencing of genes by direct attack is the principle mode of action.
On the other hand, the compounds are applied in therapeutic doses in a manner
corresponding to that of regular drugs. While somatic gene therapy still may be
seen as a concept of the future, the antisense drug approach has reached the
market. Nevertheless, it was clear to experts in drug development from early phases
that the ‘‘rush for clinical trials’’ with a class of absolutely new drug compounds of
largely unknown pharmacokinetic and pharmacodynamic properties might easily
lead to failure and a backlash for this remarkable approach in drug therapy.[1]
Therefore, it is not surprising that the lack of success of the first drug fomivirsen and
the failing of most other projects in preclinical or clinical phases still shed doubt on
the ultimate success of this important therapeutic principle. The advent of siRNA
(short interfering RNA)-induced gene silencing has given new impetus to the whole
field and once more has opened a great chance for a class of drug compounds,
which, as nucleic acid drugs may act either on genes as antisense compounds or on
proteins as aptamers.
In terms of drug development, specific gene sequences are almost perfect drug
targets and their antisense sequences are excellent lead structures. However, in any
drug research project, optimizing the lead structure to achieve an optimum of
pharmacological parameters is an indispensable prerequisite for success. As a rule,
there is a significant difference between the structure of the lead compound and that
of the preclinical development candidate (PDC). An abundant variety of chemical
modifications of oligonucleotides has been reported.[2,3] It seems worthwhile to
reconsider the material available and to choose promising options, which might be
particularly suited as tools in nucleic acid drug lead optimization.
The present article summarizes some results of biophysical studies with a type
of particularly promising nucleic acid drug modifications, the so-called zwitterionic
oligonucleotides. A cationic substituent covalently linked to a nucleic acid, like an
amino group protonated under physiological conditions, may neutralize an anionic
phosphate group, thus decreasing the electrostatic repulsion between the two single
strands[4] and/or strengthening interstrand binding in duplexes. In addition, such
compounds may be expected to exhibit different protein-binding properties. The
term ‘‘zwitterionic’’ oligonucleotide has been coined at almost the same time
independently by two research groups: Hashimoto et al.[5,6] tethered aminohexyl
groups to carbon 5 of uridine and prepared the corresponding modified oligo-
nucleotides resulting in improved binding characteristics compared to oligonucleo-
tides bearing alkyl chains of the same length. Our own group made also use of
aminohexyl groups[4] but attached them to the 2’-O-position of the sugar moiety,
which had been found previously to be an excellent position for ligand attachment
by us and others.[7–9] The decision for a hexyl alkyl chain was, in our case, the
result of a Monte Carlo simulation study[10] on nucleic acid duplexes. According

Zwitterionic Oligonucleotides
1169
to the computed data, short alkyl chain up to four carbon atoms are too short
to contribute to ionic interstrand stabilization of a duplex, while chain lengths
from five to eight provide suitable conditions. A six-carbon tether gave optimal
steric stabilization.
Apart from double strand stabilization it was also the realistic hope for
improved cellular uptake that rendered the approach of zwitterionic oligonucleo-
tides with an aminoalkyl side chain particularly promising. It had been established
earlier that conjugation or complexation of antisense compounds to poly-L-
lysine[11–14] increases the efficacy of cellular uptake to a great extent. In fact, during
the last decade, our own group, together with the group of Kreuter and Zimmer,
focused on the design of aminoalkyl modified nanoparticles and liposomes as
oligonucleotide carriers.[15–18] Several promising ‘‘drug targeting’’ strategies
including other polycationic peptides like protamine[19] have been reported.
In the meantime, several publications have appeared on ‘‘zwitterionic
oligonucleotides’’ in general and on oligonucleotides bearing aminoalkyl groups
in particular. 2’-O-Aminopropyl[20,21] modified oligonucleotides were found to have
very favorable effects on both nuclease stability and hybridization properties. While
exhibiting melting temperatures that are 1.0°C per modification higher than
phosphorothioates,[22] their 3’-exonuclease resistance was 6–8 times better than that
of 2’-desoxyphosphodiesters.[20] Structure-based studies showed that the cationic
modification prevents the binding of a metal ion that is required for the enzyme to
efficiently catalyze the phosphoryl transfer reaction.[23] Other reported zwitterionic
2’-modifications include the 2’-O-dimethylaminopropyl and 2’-O-guanidiniu-
methyl[22] groups. Spermidine and other polyamines have been conjugated to
several positions of oligonucleotides.[5,24,25] However, only little data is published to
date referring to hybridization properties of these compounds. Hashimoto et al.
found that a duplex of his completely zwitterionic oligonucleotide with the natural
DNA complement is insensitive to the ionic strength of the solution in contrast to
wild-type DNA.[5] Two oligonucleotides containing four 2’-O-aminohexyl modified
thymidine moieties each showed an increase in melting point by 3.2 and 4.4°C,
respectively.[21]
Apart from the ‘‘ionic’’ function, aminoalkyl groups have been mainly reported
as linkers to attach dyes, reactive groups or other macromolecules to oligo-
nucleotides. 2’-O-Aminohexyl modified nucleotides were prepared and in-
corporated into oligonucleotides as a linker for attachment of cholesterol,[9]
fluorescein,[21] biotin,[26] anthraquinone,[27] and pyrene.[28]
The activation of RNAse H, a nuclease specifically cleaving duplexes consisting
of a DNA and an RNA strand, is deemed to be an important property for highly
efficient antisense oligonucleotides. Owing to their RNA-like character, 2’-O-
modified antisense oligonucleotides are generally unable to result in substrates for
RNAse H after hybridizing to the targeted RNA.[29,30] However, by creating so-
called gapmers, which incorporate a stretch of unmodified nucleotides flanked by
several 2’-O-modified nucleotides, this drawback can be overcome.[31]

1170
C. R. Noe et al.
In this article, we give a detailed description of the preparation of zwitterionic
oligonucleotides containing 2’-O-aminohexyl adenosine and uridine modifications
and a systematic study of the biophysical properties of these antisense oligonucleo-
tides. In contrast to the previously published characteristics of some 2’-O-aminoalkyl
modified oligonucleotides, we focused on one specific duplex sequence, poly(A)-
poly(U), and compared our results to those obtained with 2’-O-alkyl modifications of
similar lengths.
RESULTS AND DISCUSSION
The direct 2’-O-alkylation method was chosen to prepare the adenosine
nucleoside[32] (Scheme 1). First, the amino group of 6-hydroxyhexylamine (1) was
protected as trifluoracetamide using ethyl trifluoroacetate to give 2 (Scheme 2).
Substitution of the hydroxy group by iodide was achieved via the tosylate 3 and
subsequent reaction with potassium iodide giving 6-trifluoroacetamidohexane-1-
iodide (4). This alkylating agent was reacted with deprotonated adenosine to afford
2’-alkylated product 5 in a 30% yield. Additionally, 8% of 3’-alkylated product was
found after work-up as well as 2.5% of dialkylated adenosine. In the next step the
exocyclic amino group of adenosine was protected with a phenoxyacetyl group to
give 6 and the 5’-position was dimethoxytritylated resulting in 7.
While the synthesis of 2’-O-aminohexyl adenosine derivatives proceeded
straightforward in high analogy to that of 2’-alkylated ribonucleotides, amino-
alkylation of uridine was less easily achieved. ‘‘Soft’’ alkylating agents react in a SN2-
manner with nucleosides and the pKa of the imido nitrogen in position 3 is lower
than that of the hydroxyl group in 2’. Consequently, protection of the nucleobase at
position 3 is necessary for the synthesis of 2’-O-alkyl derivatives of uridine when
sodium hydride is used. Thus, we protected the imido group of 5’-DMT-uridine by
reaction of methoxyethoxymethylchloride and triethylamine.[33] Subsequent
alkylation with 6-phthalimidohexan-1-bromide gave rise to a mixture of 2’- and
3’-alkylated uridine derivatives with preferred alkylation of 2’ (2.3:1) and a total
yield of 30%. The two isomers could be separated by MPLC (medium pressure
liquid chromatography).
SCHEME 1 Synthesis of aminohexyl building block. a: CF3COOEt, CH2Cl2 (96%); b: p-TosCl, CH2Cl2, pyr
(85%); c: KI, acetone (85%).

Zwitterionic Oligonucleotides
1171
SCHEME 2 Synthesis of adenosine building block 7. d: NaH, 4 (30%); e: TMSCl, (PhOCH2CO)2O (36%);
f: DMTCl, pyr (73%).
With uridine an additional protecting step is required prior to 2’-O-alkylation. In
spite of the good selectivity in favor of 2’-alkylated product, we decided to check the
previously published procedure via a stannylene derivative (8)[13,33] to overcome
poor overall yields. Alkylation of the stannylene derivative 8 (Scheme 3), achieved
by reaction with dibutyltin oxide in nearly quantitative yield,[34] gave 9/10 in
better total yield of 44%, but with loss of selectivity for the 2’-position (55:45).
Separation of the two isomers by flash chromatography or MPLC did not succeed.
Dimethoxytritylation products 11 and 12, however, could be separated more
SCHEME 3 Synthesis of uridine building block 11. g: dibutyltin (IV) oxide (98%); h: 4; f: DMTCl, pyr (15% for
two steps).

1172
C. R. Noe et al.
easily by column chromatography. While the selectivity of a 2’-alkylation upon
activation with sodium hydride was higher, the overall yield of the stannylene
process was higher, adding up to 15% compared to 8% of the sodium hydride
method. Furthermore, trifluoracetic acid could not be used to protect the amino
group, when sodium hydride was used for deprotonation. Trifluoracetic acid is
superior to the phthalimido group, because it is cleaved readily after oligo-
nucleotide synthesis under the same conditions as the base protecting groups
(concentrated aqueous ammonia 55°C/16 h). Cleavage of the phthalimido groups
requires an additional step with 10% aqueous methylamine,[35] because the
deprotection with concentrated ammonia proceeds ineffectively.
The phosphoramidites of modified adenosine and uridine derivatives (13,14)
were prepared using the established method[36] in nearly quantitative yields. 2’-O-
Ethyl adenosine phosphoramidite was prepared according to the previously
published procedure.[32] All products were stored at À25°C in an argon atmo-
sphere and were stable for months. Oligonucleotide syntheses were performed in
an ABI 392 DNA synthesizer using the standard CE protocol. For coupling of the
2’-modified nucleotides, the coupling times were extended to 15 min. Still, average
coupling yields were poorer than for unmodified base building blocks and
amounted to only 92%. The steric hindrance of 3’ by the bulky 2’-O-substituent
might explain these somewhat lower yields. Nevertheless, yields were high enough
to provide the dodecamers after standard HPLC purification. We chose homomers
of adenosine and its counterstrand, thymidine homomers, as the model substances
because of the ease of analysis and purification. Circular dichroism spectra of
poly(A)-poly(T) are well defined and an influence of the modifications on the
secondary structure of the oligonucleotide duplex can easily be monitored.
One to six modified nucleotides were incorporated in oligonucleotides at the 5’-
end (15a,b--22a,b, Table 1) and characteristics were compared to the unmodified
DNA strands and to 2’-O-heptyl and 2’-O-ethyl modified adenosine dodecamers
(15c,d--22c,d). For the adenosine homomers, further oligonucleotides bearing 2’-
O-aminohexyl and 2’-O-ethyl modifications were prepared, one with unmodified
TABLE 1 Sequences of Synthesized Oligonucleotides
a, c, and d
b
15 5’-AAA AAA AAA AAA-3’
16 5’-AAA AAA AAA AAA-3’
17 5’-AAA AAA AAA AAA-3’
18 5’-AAA AAA AAA AAA-3’
19 5’-AAA AAA AAA AAA-3’
20 5’-AAA AAA AAA AAA-3’
a and c
5’-UUU UUU UUU UUU-3’
5’-UUU UUU UUU UUU-3’
5’-UUU UUU UUU UUU-3’
5’-UUU UUU UUU UUU-3’
5’-UUU UUU UUU UUU-3’
5’-UUU UUU UUU UUU-3’
21 5’-AAA AAA AAA AAA-3’
22 5’-AAA AAA AAA AAA-3’
Underlined nucleotides are modified (a, b: 2’-O-aminohexyl, c: 2’-O-ethyl, d: 2-O-
heptyl. (From Ref. [27].)

Zwitterionic Oligonucleotides
1173
TABLE 2 HPLC Retention Times of Oligonucleotides Containing 2’-O-Aminohexyl Modification, Using a
Linear Gradient 10–40% B in 0–30 min (A: 0.1 M TEAA, B: 0.1 M TEAA in 80% ACN)
2’-O-Aminohexyl
2’-O-Aminohexyl
2’-O-Ethyl
adenosine series
uridine series
adenosine series
15a
16a
17a
18a
19a
20a
9.08
15b
16b
17b
18b
19b
20b
17.35
18.23
18.08
17.75
17.36
17.10
15c
16c
17c
18c
19c
20c
10.52
13.40
15.04
16.50
17.79
18.69
10.70
11.40
12.11
12.59
14.47
triplets at both the 3’- and the 5’-end (21a,d) and one bearing modifications in an
alternating pattern (22a,d). Retention times in HPLC increased with increasing
number of 2’-O-aminohexyl adenosine nucleotides. However, in the uridine series,
maximum retention time was reached with two modified nucleotides and then
dropped with further increase of 2’-O-aminohexyl modifications (Table 2). MALDI
mass spectrometry of HPLC-purified products was performed to confirm the
structures. In contrast to reports by Griffey et al.,[20] but in accordance to the findings
of Behr et al.,[24] no side products due to transamidation of trifluoroacetyl amides
were observed in the capping step.
Circular dichroism spectra of the duplexes of these oligonucleotides with
complementary DNA strands were recorded. The resulting spectra indicated a B-
type DNA structure[37] for the duplexes of all compounds 15--20. It is known that
attachment of electron withdrawing moieties at the 2’ carbon shifts the conforma-
tional equilibrium toward the C3’-endo pucker, an RNA-like conformation. A shift
toward an A-DNA shape might also be expected along with the incorporation of an
increasing number of 2’-O-aminoalkyl modifications into oligodesoxynucleotides. A
transition from B-type to A-type DNA occurs in similar hybrids with increasing
FIGURE 1 CD spectra of modified oligonucleotides 15b--20b with complimentary DNA. 15b: – – – 16b:
— – – 17b: — – — – 18b: — — — 19b: – – 20b: —. All determinations were carried out in a solution of 0.10 M
Tris-HCl (pH 7.0) and 0.15 M NaCl.

1174
C. R. Noe et al.
number of RNA-like nucleosides.[38] However, no such occurrence could be
observed for the tested compounds. Even the products consisting of 50% RNA-like
nucleotides, the B-DNA structure was clearly favored (Figure 1). These findings are
characteristic of the poly(A)-poly(T) duplex.[39] With these, the number of base
pairs per helix turn is 10.5, compared to 10.1 for other B-DNA sequences.[40]
Obviously, the conformational rigidity of a homo-A-homo-T duplex is exception-
ally strong, because A-T-base pairs have a tendency to bind large amounts of water.
The major and minor grooves of B-type DNA provide more room for inclusion of
water than A-type DNA. B-type DNA is hence preferred. Similar shifts of DNA-
types can be induced in sequences of mixed base composition.[41] Even increase of
salt concentration does not induce conformational change. In poly(A)-poly(T)-
duplexes, a triple helix is formed instead.
Melting curves of these duplexes, except for that of 15a, all exhibit a slight
decrease of the melting temperature (Tm) with increasing number of modified
nucleotides (Table 3) compared to the native A12–T12 duplex. For both the
adenosine and the uridine series, the extent of decrease was less than that for 2’-O-
alkyl and ethylene glycol modified oligonucleotides with comparable alkyl length.[42]
For 20a, the decrease per modification was only 0.9°C, and for 20b, 1.5°C,
whereas homo-adenosine dodecamers with six 2’-O-butyl modifications showed a
destabilization of 11.9°C (2.0°C per modification) and with six 2’-O-heptyl chains
even 16.1°C (2.7°C per modification). Interestingly, the location of modified
nucleosides is crucial with regard to their hybridization affinity. When comparing
the melting temperatures of 20a,c, 21a,c, and 22a,c, all adenosine dodecamers
with six 2’-O-aminohexyl modifications, differences of five degrees are observed.
The highest melting temperature was found for 21a,c, the products with six
consecutive internal modifications. In comparison, when the modifications were
placed at the end of the adenosine dodecamer (20a,c), the melting temperatures
were 2 and 3°C lower for the aminohexyl and the ethyl series, respectively.
Modified nucleosides alternating with unmodified ones (22a,c) were the least
favored in that respect. The distribution of steric hindrance across the oligo-
nucleotide strand obviously eases thermal denaturation. It is evident that the amino
TABLE 3 Melting Temperatures of Duplexes of 2’-O-Modified Oligonucleotides
T12-A12
15a
16a
17a
18a
19a
20a
21a
22a
34.7
35.7
33.5
32.6
31.3
30.7
29.5
31.7
26.6
15b
16b
17b
18b
19b
20b
32.5
33.0
30.2
28.6
27.7
25.6
15c
16c
17c
18c
19c
20c
21c
22c
36.2
34.4
32.3
32.0
29.1
28.4
31.9
25.9
15d
16d
17d
31.8
29.5
26.4
19d
20d
23.0
18.6
(a: Aminohexyl modified adenosine, b: aminohexyl modified uridine, c: ethyl modified adenosine d: heptyl
modified adenosine[27]) with complimentary DNA.

Zwitterionic Oligonucleotides
1175
group can lead to duplex stabilization by reducing the net charge of the oligo-
nucleotide and compensate for the unfavorable effect of the steric bulk of the 2’-O-
alkyl substiment. The increase in melting point when compared to a modified
oligonucleotide with 2’-O-heptyl groups (same number of side chain atoms)
amounts to 10.9°C (1.8°C per modification). This increase of melting temperature is
attributed to the neutralizing effect of the protonated amino group.
Phosphorothioates, widely used in antisense experiments, are known to exhibit
a decrease in melting temperature of at least 0.5°C per modified nucleotide in
comparison to phosphodiester oligonucleotides.[2] A substitution of 2’-desoxyur-
idine for thymidine accounts for an average reduction of 0.5°C, shown for a
considerable number of different sequences.[2] Hence, duplex destabilization of the
2’-O-aminohexyl modification amounts to only about 0.5°C per modification for
both adenosine and uridine/thymidine dodecamers when compared to phosphor-
othioates, the current standard for antisense treatment. Griffey et al. found an
increase of 0.4°C for fully 2’-O-aminopropyl modified 15- and 17-mers when
compared to the corresponding phosphorothioate, but even higher melting
temperatures for 2’-O-propyl and 2’-O-methyl modified oligonucleotides of the same
sequence.[20] They used a lower salt concentration during their experiments,
possibly accounting for the differences in comparison to our results (Figure 2).
The nuclease resistance of the uridine series 15b--20b was tested using 5’-
exonuclease. The test substances were incubated with 200 mU 5’-exonuclease from
calf spleen for 30 min at 37°C. The extent of degradation was determined by HPLC
and can be seen in Figure 3. The unmodified dodecamer was nearly fully degraded
(90%) after 30 min, and 15b was partially (15%) degraded. Three modified
nucleotides at the 5’-end (17b) were sufficient to provide nearly full 5’-exonuclease
stability, with only 2% of oligonucleotide losing the 5’-terminal nucleotide. These
results are in accordance with the findings of Griffey et al. for aminopropyl
modified oligonucleotides.[20] In the molecule with a single terminal 2’-modification,
the aminohexyl groups seem to account for the higher nuclease resistance. After
incubation for 30 min, only 15% of the oligonucleotide showed a loss of the terminal
nucleotide, while a t1/2 of 20 min was determined for the aminopropyl modification.
FIGURE 2 Melting temperatures of 2’-O-aminohexyl modified adenosine (squares) and uridine series (triangles)
compared to 2’-O-heptylmodified adenosine series (rhombuses).

1176
C. R. Noe et al.
FIGURE 3 HPLC after incubation of oligonucleotides 15b, 17b, and control T12 with 5’-exonuclease (calf
spleen) for 30 min at 37°C.
In both experiments, the incorporation of three terminal modified nucleotides was
enough to achieve full resistance against nuclease degradation. However, these
results do not give a straightforward comparison on the effect of the length of the
alkyl linker on nuclease resistance because Griffey et al. used snake venom
phosphodiesterase and 3’-terminal modifications.
EXPERIMENTAL
Reagents were purchased from Merck or Aldrich in standard quality and used
without purification. Melting points were measured in a Bu¨chi melting point
apparatus and are uncorrected. Anhydrous solvents were obtained as follows:
tetrahydrofurane was refluxed on sodium and then distilled, and pyridine,
dichloromethane and triethylamine were distilled from calcium hydride. Purified
water was obtained from a Milli-Q-apparatus. NMR spectra were recorded on a
Bruker AC-200 MHz. Shifts are reported relative to the solvent peak (CHCl3 in
CDCl3: d 7.26 and 77.00, DMSO in d6-DMSO: d 2.50 and 39.50), and coupling
constants are in Hz. Thin-layer chromatography (TLC) was performed using silica
gel 60-F254 precoated aluminium plates by Merck. Column chromatography was

Zwitterionic Oligonucleotides
1177
performed with Merck silica gel 60. Elemental analyses were done at the Institut fu¨r
Organische Chemie der Johann Wolfgang Goethe Universit¨at Frankfurt/Main with
a Heraeus CHN Rapid.
N-(6-Hydroxyhexyl)trifluoroacetamide (2). To a solution of 6-
hydroxyhexyl-1-amine (1, 100 g, 0.853 mol) in dichloromethane (200 mL) ethyl
trifluoroacetate was added slowly. The mixture was stirred for 5 h and concentrated
to 100 mL in vacuo. Storage at room temperature resulted in precipitation of a
white solid. Collection by filtration, washing with CH2Cl2, and drying in vacuo
afforded 174.6 g (96%) pure product 2. m.p.: 49–50°C, 1H NMR (CDCl3): 7.16 (bs,
1H, H-N), 3.63 (t, J = 7.1, 2H, CH20H), 3.34 (q, J = 7.1, 2H, CH2NH), 1.60–1.40
(m, 8H, H-C(2), H-C(3), H-C(4), H-C(5)). 13C NMR (CDCl3): 157.34 (q, J = 36.5,
CO), 115.87 (q, J = 285.8, CF3), 62.50 (t, C(6)), 39.77 (t, C(1)), 32.29, 28.74, 26.26,
and 25.15 (4t, C(2), C(3), C(4), C(5)). ESI-MS: 214.0 (calc. 213.20).
N-Trifluoroacetyl-(6-aminohexyl)-1-tosylate (3). Dried 2 (50 g,
0.235 mol) was dissolved in dichloromethane and p-toluenesulfonyl chloride (49.57
g, 0.260 mol) was added to the stirred solution. Pyridine (34 g, 0.430 mol) was added
dropwise at 0°C and the resulting mixture was kept at 0°C for 1 h, then was allowed
to warm to room temperature. Stirring was continued for 21 h. The reaction was
quenched by pouring onto 0.5 M HCl/ice/H2O. The product was extracted with
CH2Cl2. The organic phase was dried over Na2SO4, and the solvent was removed in
vacuo. Recrystallization was afforded by stirring a suspension in 500 mL cyclo-
hexane at 4°C for 18 h. Filtration of crystals, washing with cyclohexane and drying in
1
vacuo gave product 3 (73.61 g, 85%). Fp: 33–36°C, H NMR (CDCl3): 7.75 (d,
J = 8.1, 2H, Ar-H), 7.33 (d, J = 8.1, 2H, Ar-H), 6.64 (bs, 1H, NH), 3.99 (t, J = 6.3,
2H, H-6), 3.29 (m, 2H, CH2NH), 2.43 (s, 3H, CH3), 1.68–1.19 (m, 8H, H-2, H-3, H-4,
H-5). 13C NMR (CDCl3): 157.34 (q, J = 36.5, CO), 144.84 (ArC), 132.96 (ArC),
129.85 (d, ArC), 127.78 (d, ArC), 115.87 (q, J = 285.75, CF3), 70.32 (t, C-6), 39.63 (t,
CH2NH), 28.61, 28.56, 25.83, and 24.83 (4t, C-2, C-3, C-4, C-5), 21.60 (q, CH3).
6-Trifluoroacetamidohexane-1-iodide (4). Tosylate 3 (73.6 g, 0.20
mol) was dissolved in 500 mL acetone and potassium iodide (35.97 g, 0.24 mol) was
added. The reaction mixture was heated to reflux for 5 h, cooled to room tem-
perature, and filtered. The filtrate was dried in vacuo and the residue was dissolved
in dichloromethane. The solution was washed with 0.1 M sodium thiosulfate solu-
tion until it was colorless, dried over Na2SO4 and the solvent was removed in
vacuo. Crystallization from dichloromethane/diethyl ether afforded 4 (55.0 g, 85%)
as colorless needles. Fp: 57–58°C, 1H NMR (CDCl3): 6.38 (bs, 1H, NH), 3.40–3.31
(m, 2H, CH2NH), 3.18 (t, J = 6.93, 2H, CH2I), 1.90–1.20 (m, 8H, H-2, H-3, H-4,
H-5). 13C NMR (CDCl3): 157.58 (q, J = 36.5, CO), 115.85 (q, J = 286.0, CF3) 39.80
(t, C-1), 33.10, 29.92, 28.78, 25.56 (4t, C-2, C-3, C-4, C-5). ESI-MS (negative mode):
322.1 (calc. 323.10).

1178
C. R. Noe et al.
Alkylation of Adenosine by Sodium Hydride Method
Adenosine (50 g, 0.187 mol) was dried over phosphorous pentoxide for 20 h,
partially dissolved in 400 mL DMF, and cooled to 0°C. Sodium hydride (9.0 g,
0.225 mol, 60% suspension in oil) was washed with n-hexane and added to the
suspension of adenosine. The reaction mixture was stirred for 30 min. Iodide 4
(60.5 g, 0.187 mol) was, added and the mixture was heated to 40°C for 5 h. The
reaction was quenched by adding 10 mL methanol and solvents were removed in
vacuo. Repeated flash chromatography (5–10% methanol in dichloromethane) gave
1
5 (25.9 g, 30%) as a yellow foam. H NMR (DMSO-d6): 8.84 (bs, 1H, NH), 8.12 (s,
1H, H-2), 8.00 (s, 1H, H-8), 6.96 (m, 2H, NH2), 6.27 (m, 1H, OH-5), 5.90 (d, J = 6,
1H, H-1’), 4.48 (m, 1H, H-2’), 4.36 (m, 1H, H-3’),4.12 (s, 1H, H-4’), 3.80–3.60 (m,
2H, H-5’a, H-5’b), 3.48 (m, 1H, OH-3’), 3.27 (m, 2H, CH2NH), 3.06 (m, 2H, OCH2),
1.30–1.10 (m, 8H, 4 CH2). 13C NMR (DMSO-d6): 156.17 (CO), 156.04 (C-6), 151.85
(d, C-2), 148.06 (C-4), 139.84 (d, C-8), 120.11 (C-5), 115.69 (q, J = 286.0, CF3), 87.89
(C-1’), 87.41 (C-4’), 80.65 (C-2’), 69.85 (t, OCH2), 69.50 (C-3’), 62.12 (t, C-5’), 39.51 (t,
CH2NH), 28.67, 28.02, 25.79, 24.68 (4t, CH2). C18H25N6O5F3: calc. C 46.75, H 5.45,
N 18.17, found C 46.75, H 5.56, N 18.05.
Alkylation of uridine by stannylene method (34). Dibutyltin
(IV) oxide is toxic! Protective gloves should be worn during
handling and dispersion of dust has to be prevented. Residual
reagent has to be collected in a moistured container. Uridine (25 g,
0.102 mol) and dibutyltin (IV) oxide (25.38 g, 0.102 mol) were suspended in 2000
mL dried methanol. After refluxing for 4 h, a clear solution was formed. After
cooling to room temperature, the solvent was removed in vacuo giving stannylene
derivative 8 (47.4 g, 98%) as colorless crystals, which were used for the next reaction
without further purification. m.p.: 232°C, 1H NMR (MeOH-d4): 7.93 (bs, 1H, H-6),
5.84 (bs, 1H, H-5), 5.71 (d, J = 7.6, 1H, H-1’), 4.40–3.70 (m, 5H, H-2’, H-3’, H-4’,
H-5’), 1.90–1.10 (m, 13H, CH2, OH-5’), 0.94 (t, J = 7.2, 6H, CH3). C17H28N2O6Sn:
calc. C 42.98, H 5.95, N 5.90; found C 43.14, H 6.08, N 5.79.
Stannylene derivative 8 (9.93 g, 20.9 mmol) was dissolved in 100 mL DMF and
iodide 4 (14.45 g, 44.7 mmol) was added under an atmosphere of argon. The mixture
was refluxed for 18 h. DMF was removed in vacuo and the residue was distributed
between ethyl acetate and water and extracted five more times with ethyl acetate.
Organic phases were dried over Na2SO4 and the solvent was removed in vacuo.
Purification with flash chromatography (5% methanol in dichloromethane) afforded a
mixture of 2’-(9) and 3’-alkylated (10) product (4.06 g, 44%, 9:10 = 55:45).
1
9: H NMR (DMSO-d6): 11.29 (bs, 1H, NH), 9.34 (bs, 1H, HNCF3), 7.92 (d,
J = 8.1, 1H, H-6), 5.82 (d, J = 4.5, 1H, H-1’), 5.62 (m, 1H, H-5), 5.10 (m, 1H, OH-5’),
5.00 (d, J = 5.9, 1H, OH-3’), 4.04 (m, 1H, H-3’), 3.95–3.20 (m, 6H, H-2’, H-4’, H-5’,
OCH2), 3.15 (t, J = 6.4, 2H, CH2NH), 1.60–1.20 (m, 8H, CH2). 13C NMR (DMSO-
d6): 163.09 (C-4), 156.07 (q, J = 35.7, COCF3), 150.47 (C-2), 140.33 (C-6), 115.93 (q,

Zwitterionic Oligonucleotides
1179
J = 286.6, CF3), 101.72 (C-5), 86.16 (C-1’), 85.04 (C-4’), 81.08 (C-2’), 69.63 (CH2N),
68.32 (C-3’), 60.45 (t, C-5’), 39.03 (t, CH2O), 29.15, 28.93, 25.94, 25.07 (4t, CH2).
1
10: H NMR (DMSO-d6): 11.29 (bs, 1H, NH), 9.34 (bs, 1H, HNCF3), 7.86 (d,
J = 8.1, 1H, H-6), 5.73 (d, J = 4.5, 1H, H-1’), 5.62 (m, 1H, H-5), 5.30 (d, J = 5.7, 1H,
OH-2’), 5.10 (m, 1H, OH-5’), 4.12 (m, 1H, H-3’), 3.95–3.20 (m, 6H, H-3’, H-4’, H-5’,
OCH2), 3.15 (t, J = 6.4, 2H, CH2NH), 1.60–1.20 (m, 8H, CH2). 13C NMR (DMSO-
d6): 163.09 (C-4),156.07 (q, J = 35.7, COCF3), 150.67 (C-2), 140.50 (C-6), 115.93 (q,
J = 286.6, CF3), 101.64 (C-5), 88.00 (C-1’), 82.76 (C-4’), 77.44 (C-3’), 72.59 (C-2’),
69.57 (CH2N), 60.75 (t, C-5’), 39.03 (t, CH2O), 29.15, 28.11, 25.89, 24.90 (4t, CH2).
N6-(2-Phenoxyacetyl)-2’-O-(6-trifluoroacetamido)hexyladeno-
sine (6). 2’-O-Alkylated derivative 5 (25.0 g, 54 mmol) was dissolved in 100 mL
dry pyridine. Chlorotrimethylsilane (24.7 g, 227 mmol) was added dropwise and
the mixture was stirred at room temperature for 2 h. Phenoxyacetic anhydride (21.7
g, 76 mmol) was added and stirring was continued for 5 h. The reaction was
quenched by pouring the mixture onto iced water/methanol (100 mL). After 30 min
sodium bicarbonate (40 g) was added and the solution was extracted with di-
chloromethane. Organic layers were combined, dried over Na2SO4, and the sol-
vent was removed in vacuo. Flash chromatography (3% i-PrOH in CH2Cl2) afforded
1
N6-protected adenosine derivative 6 (11.6 g, 36%) as a yellow foam. H NMR
(CDCl3): 9.72 (bs, NH), 8.78 (s, 1H, H-2), 8.21 (s, 1H, H-8), 7.40–7.00 (m, 5H, Ar-
H), 6.84 (m, 1H, OH-5’), 5.98 (d, J = 6.0, 1H, H-1’), 4.91 (s, 2H, CH2O), 4.77 (m,
1H, H-2’), 4.55 (m, 1H, H-3’), 4.40 (s, 1H, H-4’), 4.00–3.80 (m, 2H, H-5’a, H-5’b),
3.56 (m, 1H, OH-3’), 3.37 (m, 2H, CH2NH), 3.29 (m, 2H, OCH2), 1.50–1.20 (m, 8H,
CH2). 13C NMR (CDCl3): 167.03 (N–C = O), 157.45 (CO), 156.94 (ArC), 151.99
(C-2), 150.62 (d, C-6), 149.07 (C-4), 143.62 (d, C-8), 129.80 (d, ArC), 124.19 (C-
5),122.45 (d, ArC), 116.12 (q, J = 285, CF3), 114.86 (d, ArC), 89.51 (d, C-1’), 87.98
(d, C-4’), 81.07 (d, C-2’), 71.13 (t, OCH3), 70.62 (d, C-3’), 68.07 (t, CH2O), 63.05 (t,
C-5’), 39.59 (t, CH2NH), 29.16, 28.66, 26.07, and 25.24 (4t, CH2). C26H31N607F3:
calc. (0.2 H20) C 52.13 H 5.26 N 13.31, found C 52.22 H 5.34 N 13.26.
5’-O-Dimethoxytrityl-N6-(2-phenoxyacetyl)-2’-O-(6-trifluoroa-
cetamido)hexyladenosine (7). 6 (11.0 g, 18 mmol) was dissolved in
200 mL of a 50% mixture of pyridine and dichloromethane. 4-Dimethylaminopyr-
idine (0.11 g, 0.9 mmol) and 4,4’-dimethoxytriphenylmethyl chloride (7.5 g, 22
mmol) were added to the stirred solution under argon at room temperature. After
2 h the reaction was quenched by addition of CH2Cl2 and washed with 0.1 M
NaHCO3, dried over Na2SO4, and the solvent was removed in vacuo. Purification
by flash chromatography (3.5% i-PrOH in CH2CL2) gave fully protected adenosine
1
derivative 7 (12.2 g, 73%) as a yellow foam. H NMR (CDCl3): 9.45 (bs, 1H, NH),
8.73 (s, 1H, H-2), 8.25 (s, 1H, H-8), 7.40–6.80 (m, 18H, Ar-H), 6.16 (d, J = 7.8, 1H,
H-1’), 4.85 (m, 1H, H-2’), 4.84 (s, CH2O), 4.62 (m, 1H, H-3’), 4.43 (m, 1H, H-4’),
4.40–3.80 (m, 2H, H-5’a, H-5’b), 3.95 (s, 6H, OCH3), 3.60 (m, 1H, OH-3’), 3.40 (m,

1180
C. R. Noe et al.
2H, CH2NH), 3.31 (m, OCH2), 1.60–1.20 (m, 8H, 4 CH2). 13C NMR (CDCl3):
167.12 (N–C = O), 158.45 (ArC), 157.52 (CO), 157.03 (ArC), 152.06 (C-2),150.71
(d, C-6), 149.14 (C-4), 144.69 (ArC), 143.71 (d, C-8), 137.36, 130.89, 130.82, 130.62,
128.85, and 127.21 (ArC), 124.25 (C-5), 124.61 and 117.79 (ArC), 116.22 (q, J = 286,
CF3), 113.96 (ArC), 89.14 (d, C-1’), 85.63 (CPh3), 84.03 (d, C-4’), 81.97 (d, C-2’),
71.22 (t, OCH2), 70.68 (d, C-3’), 68.19 (t, CH2O), 63.19 (t, C-5’), 55.99 (q, OCH3),
39.64 (t, CH2NH), 29.18, 28.69, 26.10, 25.27 (4t, CH2). C47H49N6O9F3: calc. C 62.18
H 5.55 N 9.26, found C 62.20 H 5.79 N 8.97.
5’-O-Dimethoxytrityl-2’-O-(6-trifluoroacetamido)hexyluridine
(11). The mixture of 2’-and 3’-alkylated uridine 9/10 (4.0 g, 9.10 mmol) was
reacted as described for the adenosine derivative in 50 mL pyridine and with
DMTCl (4.62 g, 13.7 mmol). Workup and purification using MPLC (0–5% metha-
nol in CH2Cl2) gave dimethoxytritylated products 11 (3.37 g, 35.1%) and 12 (1.85 g,
274%).
1
11: H NMR (DMSO-d6): 11.34 (bs, 1H, NH-3), 9.35 (bs, 1H, NH), 7.71 (d,
J = 8.1, 1H, H-6), 7.73–7.22 (m, 9H, Ar-H), 6.88 (d, J = 8.8, 4H, Ar-H), 5.78 (d,
J = 3.7, 1H, H-1’), 5.28 (d, J = 8.1, H-5), 5.07 (d, J = 6.7, 1H, OH-3’), 4.15 (m, 1H,
H-3’), 3.94 (m, 1H, H-4’), 3.87 (m, 1H, H-2’), 3.73 (s, 6H, OCH3), 3.53 (m, 2H,
CH2O), 3.23 (m, 2H, H-5’a, H-5’b), 3.14 (m, 2H, CH2NH), 1.60–1.20 (m, 8H, 4
CH2). 13C NMR (DMSO-d6): 162.89 (q, C-4), 158.10 (q, ArC), 156.05 (q, J = 35.6,
CO), 150.21 (q, C-2), 144.56 (q, ArC), 140.12 (d, C-6), 135.31 and 135.05 (2q, ArC),
129.70, 127.81, 127.65, and 126.72 (4d, ArC), 115.98 (q, J = 279.0, CF3), 113.19 (d,
ArC), 101.41 (d, C-5), 87.10 (d, C-1’), 85.84 (q, ArC), 82.59 (d, C-4’), 80.81 (d, C-2’),
69.74 (t, OCH2), 68.45 (d, C-3’), 62.60 (C-5’), 54.98 (OCH3), 39.08 (CH2NH), 28.90,
28.09, 25.88, and 24.91 (4t, CH2). C38H42F3N3O9: calc. (0.52 H2O) C 60.76 H 5.78 N
5.59, found C 60.82 H 6.04 N 5.18.
12: 1H NMR (DMSO-d6): 11.33 (bs, 1H, 3-NH), 9.35 (bs, 1H, NHCF3), 7.74 (d,
J = 8.1, 1H, H-6), 7.74–7.20 (m, 9H, Ar-H), 6.88 (d, J = 8.7, 4H, Ar-H), 5.69 (d,
J = 3.7, 1H, H-1’), 5.38 (d, J = 5.7, 1H, OH-2’), 5.30 (d, J = 8.0, H-5), 4.25 (m, 1H, H-
2’), 3.91 (m, 1H, H-4’), 3.88 (m, 1H, H-3’), 3.73 (s, 6H, OCH3), 3.53 (m, 2H, CH2O),
3.23 (m, 2H, H-5’a, H-5’b), 3.14 (m, 2H, CH2NH), 1.60–1.20 (m, 8H, 4 CH2). 13C
NMR (DMSO-d6): 162.92 (q, C-4), 158.11 (q, ArC), 156.05 (q, J = 35.6, CO), 150.37
(q, C-2), 144.51 (q, ArC), 140.38 (d, C-6), 135.36 and 135.11 (2q, ArC), 129.81, 127.80,
127.63, and 126.73 (4d, ArC), 115.98 (q, J = 279.0, CF3), 113.18 (d, ArC), 101.33 (d,
C-5), 87.29 (d, C-1’), 85.89 (q, ArC), 80.38 (d, C-4’), 76.81 (d, C-3’), 71.98 (d, C-2’),
69.57 (t, OCH2), 62.41 (C-5’), 54.97 (OCH3), 39.08 (CH2NH), 29.05, 28.07, 25.92, and
25.00 (4t, CH2). C38H42F3N3O9: calc. C 61.53 H 5.71 N 5.66, found C 61.15 H 5.80 N
5.66.
5’-O-Dimethoxytrityl-N 6-(2-phenoxyacetyl)-2’-O-(6-trifluoro-
acetamido)hexyladenosine-3’-O-(2-cyanoethyl)-N,N-diisopropyl-
phosphoramidite (13). Protected adenosine derivative 11 (1.0 g, 1.1 mmol)

Zwitterionic Oligonucleotides
1181
was dissolved in a mixture of diisopropylethyl amine (0.42 g, 3.5 mmol) and 5 mL
dry CH2Cl2. 2-Cyanoethyl-N,N-diisopropylchloro phosphoramidite (0.4 g, 1.65
mmol) was added dropwise to the stirred solution under an argon atmosphere. After
stirring at room temperature for 5 h, the reaction was quenched by adding 0.2 mL
isopropanol and the solvents were removed in vacuo. Purification of the residue by
MPLC (75–0% petrol ether in ethyl acetate) afforded diastereomeric phosphor-
amidite 13 (1.0 g, 82%) as a colorless foam. 1H NMR (CDCl3): 9.42 (bs, NH), 8.70 (s,
1H, H-2), 8.21 (s, 1H, H-8), 7.40–6.80 (m, 18H, Ar-H), 6.12 (d, J = 5.0, 1H, H-1’), 4.83
(s, CH2O), 4.79 (m, 1H, H-2’), 4.58 (m, 1H, H-3’), 4.40 (m, 1H, H-4’), 3.80–3.30 (m,
16H, H-5’a, H-5’b, CH-CNE, CH(CH3)2, OCH3, CH2NH, OCH2), 2.65 (m, CH-
CNE), 2.65 (m, CH-CNE), 1.50–1.10 (m, 20H, CH3, CH2). 13C NMR (CDC13):
166.63 (N–C = O), 158.54 (ArC), 156.92 (C-6), 151.85 (C-2), 148.21 (C-4), 144.38 (C-
8), 135.56, 130.06, 129.83, 128.12, 127.83, and 122.46 (ArC), 117.63 (C-5), 114.90 and
113.11 (ArC), 86.66 (C-1’), 84.04 (C-4’), 80.50 (C-3’), 70.87 (O-CH2), 70.70 (C-2’),
67.95 (CHPh3), 63.01 (O-CH2), 62.71 (C-5’), 55.18 (OCH3), 43.16 and 43.06
(CH(CH3)3), 39.75 (CH2-NH), 29.20, 28.60, 26.23, and 25.36 (CH2), 24.59 and 24.53
(CH3), 20.35 and 20.29 (CH2-CN). 31P NMR (CDC13): 151.27, 150.79. ESI-MS:
1099.8. C56H64N8O10F3P: calc. C 59.11 H 5.91 N 9.85, found C 59.07 H 6.20 N 10.15.
5’-O-Dimethoxytrityl-2’-O-(6-trifluoroacetamido)hexyluri-
dine-3’-O-(2-cyanoethyl)-N,N-diisopropylphosphoramidite (14).
Synthesis of uridine phosphoramidite 14 after the method described above
1
starting with 1.0 g 11 gave diastereomeric 14 (1.25 g, 98%) as colorless foam. H
NMR (CDC13): 8.09 and 8.00 (d, J = 8.1, H-6), 7.44–7.22 (m, 9H, Ar-H), 6.95 (bs,
1H, NH-CF3), 6.88 (m, 4H, Ar-H), 5.97 (d, J = 2.8, H-1’), 5.22 and 5.19 (d, J = 8.1,
H-5), 4.65–3.40 (m, 15H, H-2’, H-3’, H-4’, H-5’a, H-5’b, OCH3, OCH2, CH(CH3)2),
3.39 (m, 2H, CH2NH), 2.76, 2.61, and 2.44 (3m, 4H, OCH2CH2CN), 1.80–1.00 (m,
20H, CH2, CH3). 13C NMR (CDC13): 163.09 (q, C-4), 158.79 (q, ArC), 157.25 (q,
J = 36.4, CO), 150.27 (q, C-2), 144.32, 144.18 (2q, ArC), 140.12 and 140.06 (2d, C-
6), 135.23, 135.14, 135.06, 134.95 (4q, ArC), 130.26, 128.28, 127.95, and 127.21 (4q,
ArC), 117.58 and 117.46 (CN), 115.98 (q, J = 279.0, CF3), 113.25 (d, ArC), 102.16
(d, C-5), 88.26 and 88.08 (d, C-1’), 87.18 and 87.02 (CPH3), 82.29 (d, C-4’), 81.61 (C-
2’), 70.86 and 70.57 (t, OCH2), 69.91 and 69.82 (d, J = 14.9, C-3’) 61.43 and 60.87
(C-5’), 58.23 (t, OCH2), 55.23 (q, OCH3), 45.40 and 45.28 (t, POCH2) 43.28 and
43.21 (d, J = 12.6, CH(CH3)2), 39.82 and 39.76 (t, NCH2), 29.33, 28.66, 28.60, 26.10,
25.47 and 25.36 (t, CH2), 24.69, 24.59, and 24.43 (q, CH(CH3)2), 20.24 (m, CH2CN).
31P NMR (CDC13): 150.85, 150.56. ESI-MS: 942.6.
Synthesis of Oligonucleotides
Oligonucleotides were prepared on an ABI 392B DNA Synthesizer in 1.0 mM
scale using standard synthesis protocols. 2’-O-Aminohexyl modified nucleotides
were coupled using extended coupling times of 5 min resulting in an average yield

1182
C. R. Noe et al.
of 92% as determined by trityl cation assay. Oligonucleotides were prepared in
DMT-off mode and cleaved from solid support using concentrated ammonia (1 h/
r.t.). Deprotection was afforded by heating the resulting solution to 55°C for 18 h.
Ammonia was removed in vacuo and the residue was redissolved in 0.5 mL water.
HPLC Analysis and Purification
Crude oligonucleotides were filtered on Schleicher & Schuell Spartan 13/20,
0.45 mm Braunrand H filters, concentrated on a speed vac, and redissolved in 300 mL
water. Concentrations were determined by measuring, UV-absorption at 260 nm.
Analytical HPLC of a 20-mL sample (0.2 mg/mL) was performed on an ET 250/8/4
Nucleosil1 100-5 C18 column (Macherey Nagel) with a flow rate of 1 mL/min and a
linear gradient of 10–40% B in 30 min (A: 0.1 M triethylammonium acetate in water,
B: 0.1 M triethylammonium acetate in 80% acetonitrile). Preparative HPLC was done
on a Aquapore Octyl Prep 20 Cartridge 250 Â 10 mm (Applied Biosystems) with a
flow rate of 4 mL/min and the same gradient as mentioned above. All
oligonucleotides were desalted using Sep-Pak-Cartridges1 (Waters, Milford, MA).
CD Spectroscopy
Concentrations of purified and desalted oligonucleotides were determined by
UV measurement at 260 nm. Molar extinction coefficients were calculated by
addition of nucleotides (184,800 cmÀ1MÀ1 for homo adenosine dodecamers,
105,600 cmÀ1MÀ1 for homo thymidine dodecamers). CD spectra were recorded on
a Jasco J-710 spectropolarimeter.
Oligonucleotides were diluted to a concentration of 9 mM in a solution of
0.15 M NaCl and 0.01 M Tris-HCl (pH 7.0) of a total volume of 300 mL.
Complementary strands were hybridized for 5 min at 80°C, then cooled to room
temperature to ensure duplex formation. Measurements were done in a quartz
cuvette of a path length of 1 mm. Wavelength range was set to 320–210 nm with a
scanning speed of 50 nm/min. For determination of melting temperature, the
duplex solution was heated from 0 to 80°C with a slope of 50°C per hour.
Nuclease Stability Test
To a solution of uridine dodecamer (2.71 nmol, 0.33 OD) in an Eppendorf
cap, 25 mL PDE-incubation buffer was added. The solution was diluted with
Milli-Q-Water to a total volume of 240 mL. After 10 min incubation at 37°C, 10 mL
(200 mU) of phosphodiesterase solution (calf spleen, Boehringer Mannheim, 4 U/
mL, diluted 1:200) was added. After mixing briefly, the resulting solution was
incubated at 37°C for 30 min. PDE-stop buffer (50 mL) was added and the mixture
was heated to 90°C for 10 min. Quantification of degradation was done using
HPLC as described above.

Zwitterionic Oligonucleotides
1183
CONCLUSION
A thorough study comparing 2’-O-alkyl– and 2’-O-aminohexyl–modified
oligonucleotides is presented. The modifications tested were dodecamers of
poly(A)-poly(T). Apart from studies on the secondary structure using circular
dichroism, melting points were primarily used as an indicator of the hybridization
affinity. 2’-O-Aminoalkyl–substituted oligonucleotides clearly offer more auspicious
characteristics in view of the therapeutic application compared to 2’-O-alkyl–
modified derivatives. The protonation of the negatively charged backbone by use
of the 2’-O-aminohexyl modification is a useful technique toward increasing the
stability of DNA-RNA duplexes. The destabilizing effect caused by the steric bulk
of the 2’-O-alkyl chain is greatly diminished by the stabilizing effect of the cationic
amino group. We have also shown that the incorporation of an increasing number
of zwitterionic nucleoside building blocks efficiently augments stability against
degrading 5’-exonucleases.
REFERENCES
1. Dean, N.M.; Bennett, C.F. Antisense oligonucleotide-based therapeutics for cancer. Oncogene 2003, 22(56),
9087–9096.
2. Freier, S.M.; Altmann, K.H. The ups and downs of nucleic acid duplex stability: structure-stability studies on
chemically modified DNA:RNA duplexes. Nucleic Acids Res. 1997, 25(22), 4429–4443.
3. Urban, E.; Noe, C.R. Structural modifications of antisense oligonucleotides. Farmaco 2003, 58(3), 243–258.
4. Noe, C.; Brunar, H. Preparation of Modified Oligonucleotides as Active Substances. Wo 9516696, June 22,
1995.
5. Hashimoto, H.; Nelson, M.G.; Switzer, C. Zwitterionic DNA. J. Am. Chem. Soc. 1993, 115(16), 7128–7134.
6. Hashimoto, H.; Nelson, M.G.; Switzer, C. Formation of chimeric duplexes between zwitterionic and natural
DNA. J. Org. Chem. 1993, 58(16), 4194–4195.
7. Cotten, M.; Oberhauser, B.; Brunar, H.; Holzner, A.; Issakides, G.; Noe, C.R.; Schaffner, G.; Wagner, E.;
Birnstiel, M.L. 2’-O-methyl. 2’-O-ethyl oligoribonucleotides and phosphorothioate oligodeoxyribonucleotides
as inhibitors of the in vitro U7 snRNP-dependent mRNA processing event. Nucleic Acids Res. 1991, 19(10),
2629–2635.
8. Brunar, H.; Holzner, A.; Issakides, G.; Knollmueller, M.; Noe, C.; Bimstiel, M.; Cotten, M.; Oberhauser, B.;
Wagner, E.; Schaffner, G. 2’-O-Alkyl-oligoribonucleotides, Their Synthesis and Use in Antisense
Oligonucleotides. Ger. Offen. De 41100851992.
9. Manoharan, M.; Tivel, K.L.; Andrade, L.K.; Cook, P.D. 2’-O- and 3’-O-pyrimidine amino-ether-containing
oligodeoxyribonucleotides: synthesis and conjugation chemistry. Tetrahedron Lett. 1995, 36(21), 3647–
3650.
10. Haberhauer, G. Ph.D. Thesis; University of Vienna: Austria, 1994.
11. Leonetti, J.P.; Degols, G.; Lebleu, B. Biological activity of oligonucleotide-poly(L-lysine) conjugates:
mechanism of cell uptake. Bioconjug. Chem. 1990, 1(2), 149–153.
12. Degols, G.; Leonetti, J.P.; Benkirane, M.; Devaux, C.; Lebleu, B. Poly(L-lysine)-conjugated oligonucleotides
promote sequence-specific inhibition of acute HIV-1 infection. Antisense Res. Dev. 1992, 2(4), 293–301.
13. Clarenc, J.P.; Degols, G.; Leonetti, J.P.; Milhaud, P.; Lebleu, B. Delivery of antisense oligonucleotides by
poly(L-lysine) conjugation and liposome encapsulation. Anti-Cancer Drugs Des. 1993, 8(1), 81–94.
14. Ginobbi, P.; Geiser, T.A.; Ombres, D.; Citro, G. Folic acid-polylysine carrier improves efficacy of
c-myc antisense oligodeoxynucleotides on human melanoma (M14) cells. Anticancer Res. 1997, 17(1A),
29–35.
15. Zimmer, A.; Zobel, H.P.; Werner, D.; Noe, C.R.; Kreuter, J. Cationic nanoparticles as enhancers for cellular
uptake of antisense oligonucleotides. Proc. Int. Symp. Control. Release Bioact. Mater. 1997, 679–680.
16. Zobel, H.P.; Kreuter, J.; Werner, D.; Noe, C.R.; Kumel, G.; Zimmer, A. Cationic polyhexylcyanoacrylate

1184
C. R. Noe et al.
nanoparticles as carriers for antisense oligonucleotides. Antisense Nucleic Acid Drug Dev. 1997, 7(5), 483–
493.
17. Zimmer, A.; Atmaca-Abdel Aziz, S.; Gilbert, M.; Werner, D.; Noe, C.R. Synthesis of cholesterol modified
cationic lipids for liposomal drug delivery of antisense oligonucleotides. Eur. J. Pharm. Biopharm. 1999,
47(2), 175–178.
18. Zobel, H.P.; Zimmer, A.; Atmaca-Abdel Aziz, S.; Gilbert, M.; Werner, D.; Noe, C.R.; Kreuter, J.; Stieneker, F.
Evaluation of aminoalkylmethacrylate nanoparticles as colloidal drug carrier systems. Part 1. Synthesis of
monomers, dependence of the physical properties on the polymerization methods. Eur. J. Pharm. Biopharm.
1999, 47(3), 203–213.
19. Junghans, M.; Kreuter, J.; Zimmer, A. Phosphodiester and phosphorothioate oligonucleotide condensation
and preparation of antisense nanoparticles. Biochim. Biophys. Acta 2001, 1544(1–2), 177–188.
20. Griffey, R.H.; Monia, B.P.; Cummins, L.L.; Freier, S.; Greig, M.J.; Guinosso, C.J.; Lesnik, E.; Manalili, S.M.;
Mohan, V.; Owens, S.; Ross, B.R.; Sasmor, H.; Wancewicz, E.; Weiler, K.; Wheeler, P.D.; Cook, P.D. 2’-O-
aminopropyl ribonucleotides: a zwitterionic modification that enhances the exonuclease resistance and
biological activity of antisense oligonucleotides. J. Med. Chem. 1996, 39(26), 5100–5109.
21. Manoharan, M.; Prakash, T.P.; Barber-Peoc’h, I.; Bhat, B.; Vasquez, G.; Ross, B.S.; Cook, P.D. N-(2-
cyanoethoxycarbonyloxy)succinimide: a new reagent for protection of amino groups in oligonucleotides.
J. Org. Chem. 1999, 64(17), 6468–6472.
22. Manoharan, M. Antisense Drug Technology; Crooke, S.T., Ed.; Marcel Dekker: New York, 2001; 391–470.
23. Teplova, M.; Wallace, S.T.; Tereshko, V.; Minasov, G.; Symons, A.M.; Cook, P.D.; Manoharan, M.; Egli, M.
Structural origins of the exonuclease resistance of a zwitterionic RNA. Proc. Natl. Acad. Sci. U. S. A. 1999,
96(25), 14240–14245.
24. Potier, P.; Abdennaji, A.; Behr, J.P. Synthesis and hybridization properties of oligonucleotides containing
polyamines at the C-2 position of purines: a pre-synthetic approach for the incorporation of spermine into
oligodeoxynucleotides containing 2-(4,9,13-triazatridecyl)-2’-deoxyguanosine. Chemistry 2000, 6(22), 4188–
4194.
25. Potier, P.F.; Behr, J.P. Recognition of DNA by strand invasion with oligonucleotide-spermine conjugates.
Nucleosides Nucleotides Nucleic Acids 2001, 20(4–7), 809–813.
26. Manoharan, M.; Inamati, G.; Tivel, K.L.; Wheeler, P.; Stecker, K.; Cook, P.D. 2’- and 3’-biotin conjugated
nucleoside building blocks: synthesis of biotinylated oligonucleotides. Nucleosides Nucleotides 1997, 16(7–
9), 1411–1413.
27. Whittemore, N.A.; Mullenix, A.N.; Inamati, G.B.; Manoharan, M.; Cook, P.D.; Tuinman, A.A.; Baker, D.C.;
Chambers, J.Q. Synthesis and electrochemistry of anthraquinone-oligodeoxynucleotide conjugates.
Bioconjug. Chem. 1999, 10(2), 261–270.
28. Manoharan, M.; Tivel, K.L.; Zhao, M.; Nafisi, K.; Netzel, T.L. Base-sequence dependence of emission
lifetimes for D141018-30-6NA oligomers and duplexes covalently labeled with pyrene: relative electron-
transfer quenching efficiencies of A, G, C, and T nucleosides toward pyrene. J. Phys. Chem. 1995, 99(48),
17461–17472.
29. Lamond, A.I.; Sproat, B.S. Antisense oligonucleotides made of 2’-O-alkylRNA: their properties and
applications in RNA biochemistry. FEBS Lett. 1993, 325(1–2), 123–127.
30. Manoharan, M. 2’-Carbohydrate modifications in antisense oligonucleotide therapy: importance of
conformation, configuration and conjugation. Biochim. Biophys. Acta 1999, 1489(1), 117–130.
31. Monia, B.P.; Lesnik, E.A.; Gonzalez, C.; Lima, W.F.; McGee, D.; Guinosso, C.J.; Kawasaki, A.M.; Cook,
P.D.; Freier, S.M. Evaluation of 2’-modified oligonucleotides containing 2’-deoxy gaps as antisense inhibitors
of gene expression. J. Biol. Chem. 1993, 268(19), 14514–14522.
32. Wagner, E.; Oberhauser, B.; Holzner, A.; Brunar, H.; Issakides, G.; Schaffner, G.; Cotten, M.; Knollmueller,
M.; Noe, C.R. A simple procedure for the preparation of protected 2’-O-methyl- or 2’-O-ethylribonucleoside 3’-
O-phosphoramidites. Nucleic Acids Res. 1991, 19(21), 5965–5971.
33. Corey, E.J.; Gras, J.L.; Ulrich, P. A new general method for protection of the hydroxyl function. Tetrahedron
Lett. 1976, 17(11), 809–812.
34. Wagner, D.; Verheyden, J.P.H.; Moffatt, J.G. Preparation and synthetic utility of some organo tin derivatives
of nucleosides. J. Org. Chem. 1974, 39(1), 24–30.
35. Hasan, S.K.; Wasson, F.I.; Wolfe, S. Five-membered rings. I. Reaction of 2-methyl-2-oxazoline with
phthalimidoacetyl chloride. Can. J. Chem. 1967, 45(17), 2000–2002.
36. Sinha, N.D.; Biernat, J.; Koester, H. b-Cyanoethyl N,N-dialkylamino/N-morpholinomonochlorophosphoami-
dites, new phosphitylating agents facilitating ease of deprotection and work-up of synthesized oligonucleotides.
Tetrahedron Lett. 1983, 24(52), 5843–5846.

Zwitterionic Oligonucleotides
1185
37. Cantor, C.R.; Schimmel, P.R. Biophysical Chemistry of Macromolecules, Pt. 3: The Behaviour of Biological
Macromolecules; W. H. Freeman: San Francisco, 1980.
38. Wang, A.H.; Fujii, S.; van Boom, J.H.; van der Marel, G.A.; van Boeckel, S.A.; Rich, A. Molecular structure
of r(GCG)d(TATACGC): a DNA-RNA hybrid helix joined to double helical DNA. Nature 1982,
299(5884), 601–604.
39. Zimmerman, S.B.; Pheiffer, B.H. A RNA:DNA hybrid that can adopt two conformations: an x-ray diffraction
study of poly(rA):poly(dT) in concentrated solution or in fibers. Proc. Natl. Acad. Sci. U. S. A. 1981, 78(1),
78–82.
40. Peck, L.J.; Wang, J.C. Sequence dependence of the helical repeat of DNA in solution. Nature 1981,
292(5821), 375–378.
41. Rhodes, D.; Klug, A. Sequence-dependent helical periodicity of DNA. Nature 1981, 292(5821), 378–380.
42. Werner, D.; Brunar, H.; Noe, C.R. Investigations on the influence of 2’-O-alkyl modifications on the base
pairing properties of oligonucleotides. Pharm. Acta Helv. 1998, 73(1), 3–10.