Fine-tuning furan toxicity: Fast and quantitative DNA interchain cross-link formation upon selective oxidation of a furan containing oligonucleotide

Article abstract of DOI:10.1039/b415092a

Oligonucleotides containing a furan modified internal nucleoside have been synthesized; selective in situ oxidation of the furan moiety to a reactive enal species in the presence of a complementary DNA strand gives rise to fast and efficient formation of an interstrand cross-link. The Royal Society of Chemistry 2005.

Full text of DOI:10.1039/b415092a

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Fine-tuning furan toxicity: fast and quantitative DNA interchain cross-
link formation upon selective oxidation of a furan containing
oligonucleotide{
Sami Halila, Trinidad Velasco, Pierre De Clercq and Annemieke Madder*
Received (in Cambridge, UK) 30th September 2004, Accepted 15th November 2004
First published as an Advance Article on the web 4th January 2005
DOI: 10.1039/b415092a
within the ON is possible and gives rise to a fast and quantitative
cross-link formation with complementary strands.
Oligonucleotides containing a furan modified internal nucleo-
side have been synthesized; selective in situ oxidation of the
furan moiety to a reactive enal species in the presence of a
complementary DNA strand gives rise to fast and efficient
formation of an interstrand cross-link.
Nucleosides equipped with a furan moiety have been prepared
previously.11,12 Instead of the introduction of the furan moiety on
the base we chose to link the furan via a 29-amino group on the
ribose and prepare both the ribo (1a) and arabino (1b) configured
stereoisomers. Depending on the stereochemistry at 29 and the
sugar pucker of the nucleoside within the ON—which in turn
depends on the target ON and the type of duplex formed—this in
principle allows for the selective orientation of the furan nucleus
towards minor or major groove. Synthesis of the two furan-
nucleoside building blocks was performed as shown in Scheme 1.
Nucleoside 3a was prepared starting from the known
1-(29-amino-29-deoxy-b-D-ribofuranosyl)uracil 2a13 by treatment
with pentafluorophenyl-3-(2-furyl)-propanoate 5 in the presence
of DBU. For the synthesis of 3b, 1-(29-amino-29-deoxy-b-D-
arabinofuranosyl)uracil hydrochloride14 was reacted with activated
ester 5. Further conversion into the required DMT-protected
phosphoramidites 1a and 1b was carried out using standard
methodology.
It has previously been shown that selective control of gene
expression is possible via artificially created nucleic acids that are
complementary to a given RNA (antisense approach) or DNA
(anti-gene approach) sequence. To overcome problems associated
with reversibility of hybridization, cross-linking strategies have
been proposed to ensure the required inhibition of gene
expression.1 Furthermore cross-linked nucleic acids have become
important in order to overcome the limitations associated with
using short synthetic oligonucleotides (ON) as models of high
molecular weight nucleic acids.2 Since interstrand cross-links
caused by environmental and chemotherapeutic agents represent
a serious impediment to cell survival, short cross-linked ON
duplexes can be of help in the study of natural repair processes,
possible causes of resistance to chemotherapeutics. A diverse array
of non-natural nucleosides have been developed for cross-linking
purposes with natural DNA/RNA targets.3–9 However most of
these methods require the introduction of modified bases and/or
extensive additional protection/deprotection schemes.
Scheme 2 summarizes the test reactions for oxidation of the
furan in nucleoside 3a. N-Bromosuccinimide (NBS) was found to
be the reagent of choice for oxidation under aqueous conditions.15
Upon treatment with 1 equivalent of NBS in pH 7.0 phosphate
buffer the corresponding furan oxidized enal derivative 6 (which
exists in equilibrium with its hydrated form) was formed next to
Inspired by the known toxicity of furan—which is believed to be
due to its in vivo oxidation into the reactive metabolite 2-butene-
1,4-dial10—we decided to explore the possibility of using a furan
moiety in an ON sequence as a masked reactive aldehyde (Fig. 1).
We here report on the preliminary but highly promising results of
these studies and show that selective in situ oxidation of the furan
Fig. 1 Schematic representation of the proposed cross-linking strategy.
{ Electronic supplementary information (ESI) available: full experimental
details for the preparation and characterization of nucleoside building
blocks. RPHPLC profiles and ESI-MS spectra of nucleoside oxidations,
oligonucleotides and cross-linked oligonucleotides. See http://www.rsc.org/
suppdata/cc/b4/b415092a/
Scheme 1 Reagents and conditions: (i) 5, DBU, DMF, 0 uC to room
temp., 2.5 h, (75%; 3a/3b); (ii) DMTCl, pyr, DMAP, room temp., 18 h,
(92%, 4a/96%, 4b); (iii) iPr2NP(Cl)OCH2CH2CN, iPr2NEt, DCM, room
temp., 1 h, (92%, 4a/97%, 4b).
*annemieke.madder@Ugent.be
936 | Chem. Commun., 2005, 936–938
This journal is ß The Royal Society of Chemistry 2005

View Article Online
Scheme 2 Oxidation of nucleoside 3a with NBS.
remaining 3a as shown by RPHPLC and ESI-MS. Subsequent
addition of benzylamine led to the conversion of 6 into an
unsaturated imine derivative 7. These preliminary experiments
showed the potential of the furan moiety to act as a masked
aldehyde group which is sufficiently reactive to allow for reaction
with potential nucleophiles in an oligonucleotide cross-linking
context.
Fig. 2 RPHPLC analysis of oxidation reaction mixtures: a) 8 before
NBS addition; b) 8 after 1 equiv. of NBS; c) 11 before NBS addition; d) 11
after 1 equiv. of NBS; e) 9:11 before NBS addition, t 5 t0; f) 9:11 after
1 equiv. of NBS, t 5 t0 + 15 min; g) 9:11 after 1 equiv. of NBS, t 5 t0 +
30 min; h) 9:11 after 4 equiv. of NBS, t 5 t0 + 60 min.
Both modified nucleosides were then successfully incorporated
into ONs 10 and 11 (Table 1). Replacement of the natural
ribonucleoside by a furan containing monomer in the 11-mers
resulted in a lowering of the duplex stability with 4 uC and 5 uC
respectively when compared to the unmodified DNA duplex 8:9.
A similar decrease in duplex stability was observed by Herdewijn
et al. when incorporating 29-acetamidonucleosides.16 The melting
temperatures remain sufficiently high to ensure duplex formation
at room temperature for the study of the cross-link formation.
We next set out to evaluate the potential of our proposed cross-
linking reaction. For this purpose the purified ONs (20 mM) were
treated with 1 equivalent of freshly prepared NBS solution in
phosphate buffer (pH 7.0, 10 mM, containing 10 mM NaCl) at
25 uC. In a first series of experiments it was shown that whereas
ON 8 (Fig. 2 a,b) was stable under these conditions, both ONs 10
and 11 gave a mixture of products (Fig. 2 c,d: example for 11).
The oxidation was then carried out with 10 and 11 in the
presence of the complementary DNA strand. Analysis of the
resulting reaction mixture under denaturing RPHPLC conditions
revealed after only 15 minutes the consumption of both strands
and the concurrent formation of a new slower running species.
Leaving the mixtures to stand for longer periods did not lead to
further conversion. Gradual addition of extra equivalents of NBS
and analysis of the reaction mixtures 15 minutes after each
addition revealed that a total amount of 4 equivalents is required
for the complete disappearance of the starting ONs (Fig. 2 e–h:
example for 9:11). Purification and analysis of the formed material
by ESI-MS revealed the complete conversion of the original ON
strands to the corresponding cross-linked duplex (9–10XL and
9–11XL).{ So in contrast to the complex mixture which is formed
when ONs 10 or 11 are oxidized as single strands, NBS treatment
of a duplex leads to clean cross-link formation. Further control
experiments were performed with 10 in the presence of the non-
complementary 59-d(GCCTGTCAG-C-G)-39. It was shown that
in this case no cross-linking took place and the furan-modified
ON 10 was converted into a mixture of products under the
reaction conditions while the non-complementary strand could be
recovered.
Table 1 Tm values for modified and reference duplexesa
Oligonucleotide duplexes
Tm/uC
8:
9:
59-d(GCCTGTCAG-U-G)-39
39-d(CGGACAGTC-A-C)-59
38.1
10:
9:
59-d(GCCTGTCAG-1a-G)-39
39-d(CGGACAGTC-A-C)-59
33.0
34.0
As for the site of cross-linking, further studies are necessary to
characterize the type and exact position of the covalent bond
between both strands. In view of the B-form of the DNA–DNA
duplex studied, the furan moiety is most likely located in the major
groove in both cases. Inspection of a molecular model reveals the
complementary A or its 59-neighbouring C as the most probable
ligation sites. Mass spectrometric data are consistent with a
dehydration reaction involving formation of an unsaturated imine
upon attack of the adenine or cytosine exocyclic amine on the
11:
9:
a
59-d(GCCTGTCAG- 1b-G)-39
39-d(CGGACAGTC-A-C)-59
UV-melting profiles were measured in 10 mM sodium phosphate
buffer (pH 7.0) containing 10 mM NaCl at a ramp rate of 0.5 uC
min21 at 260 nm using a Cary 3E UV/VIS spectrophotometer
(Varian) containing a Cary Thermoelectrical temperature-controlled
sample holder and 10 mm path-length cuvettes. The oligonucleotide
concentration was 4 mM for each strand.
This journal is ß The Royal Society of Chemistry 2005
Chem. Commun., 2005, 936–938 | 937

View Article Online
1 F. Nagatsugi, N. Tokuda, M. Maeda and S. Sasaki, Bioorg. Med.
Chem. Lett., 2001, 11, 2577 and references cited herein.
2 R. J. Cain and G. D. Glick, Nucleic Acids Res., 1997, 25, 836.
3 T. R. Webb and M. D. Matteucci, Nucleic Acids Res., 1986, 14, 7661.
4 J. C. Tabone, M. R. Stamm, H. B. Gamper and R. B. Meyer, Jr.,
Biochemistry, 1994, 33, 375.
5 L. M. Khalimskaya, A. S. Levina and V. F. Zarytova, Bioorg. Med.
Chem. Lett., 1994, 4, 1083.
6 R. S. Coleman and E. A. Kesicki, J. Org. Chem., 1995, 60, 6252.
7 R. S. Coleman and R. M. Pires, Nucleic Acids Res., 1997, 25, 4771.
8 S. Sasaki, Eur. J. Pharm. Sci., 2001, 13, 43.
9 K. Nakatani, T. Yoshida and I. Saito, J. Am. Chem. Soc., 2002, 124,
2118.
10 L.-J. Chen, S. S. Hecht and L. A. Peterson, Chem. Res. Toxicol., 1997,
10, 866.
11 D. Graham, A. Grondin, C. McHugh, L. Fruk and W. E. Smith,
Tetrahedron Lett., 2002, 43, 4785.
12 P. Wigerinck, L. Kerremans, P. Claes, R. Snoeck, P. Maudgal,
E. Declercq and P. Herdewijn, J. Med. Chem., 1993, 36, 538.
13 J. P. H. Verheyden, D. Wagner and J. G. Moffatt, J. Org. Chem., 1971,
36, 250.
aldehyde functionality. The exact position and nature of the
cross-link and a detailed analysis of sequence dependence and
importance of linker length are currently being investigated.
The results presented here clearly demonstrate the potential of
the in situ furan oxidation to trigger cross-linking within an ON
duplex as a more convenient and higher yielding alternative to
existing methods (e.g. acrolein modified ONs).17 Moreover the
described method offers an attractive way for the introduction of
aldehyde functionality into DNA for labeling purposes.
S. Halila and T. Velasco acknowledge the financial support
provided through the European Community’s Human Potential
Programme under contract HPRN-CT-2000-00014 [CombiCat].
Sami Halila, Trinidad Velasco, Pierre De Clercq and
Annemieke Madder*
Laboratory for Organic and Biomimetic Chemistry, Department of
Organic Chemistry, Krijgslaan 281, S4, 9000, Gent, Belgium.
E-mail: annemieke.madder@Ugent.be; Fax: +32-9-2644498;
Tel: +32-9-2644472
14 G. Walcher and W. Pfleiderer, Helv. Chim. Acta, 1996, 79, 1067.
15 Y. Kobayashi, M. Nakano, G. B. Kumar and K. Kishihara, J. Org.
Chem., 1998, 63, 7505.
16 C. Hendrix, B. Devreese, J. Rozenski, A. Van Aerschot, A. De Bruyn,
J. Van Beeumen and P. Herdewijn, Nucleic Acids Res., 1995, 23, 51.
17 I. D. Kozekov, L. V. Nechev, M. S. Mosely, C. M. Harris, C. J. Rizzo,
M. P. Stone and T. M. Harris, J. Am. Chem. Soc., 2003, 125, 50.
Notes and references
{ ESI-MS for 9–10XL: Mcalcd 5 6792.21, Mexp 5 6791.27; for 9–11XL:
Mcalcd 5 6792.21, Mexp 5 6792.23.
938 | Chem. Commun., 2005, 936–938
This journal is ß The Royal Society of Chemistry 2005