The design, synthesis and evaluation of hypoxia-activated pro-oligonucleotides

Article abstract of DOI:10.1039/b903331a

Hypoxia-activated pro-oligonucleotides were synthesized through the commercial phosphoramidite method, and could be readily cleaved to form normal oligos with good hypoxia selectivity in vitro under the effect of reductases, as well as in tumor cell extra

Full text of DOI:10.1039/b903331a

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
The design, synthesis and evaluation of hypoxia-activated
pro-oligonucleotidesw
Nan Zhang,ab Chunyan Tan,a Puqin Cai,a Peizhuo Zhang,c Yufen Zhaob
and Yuyang Jiang*ad
Received (in Cambridge, UK) 17th February 2009, Accepted 27th March 2009
First published as an Advance Article on the web 23rd April 2009
DOI: 10.1039/b903331a
Hypoxia-activated pro-oligonucleotides were synthesized
through the commercial phosphoramidite method, and could
be readily cleaved to form normal oligos with good hypoxia
selectivity in vitro under the effect of reductases, as well as in
tumor cell extract.
(2a) and 50-O-dimethoxytrityl thymidine-30-O-[(5-nitro-2-
thienyl)methyl N,N-diisopropylphosphoramidite] (2b) were
synthesized from 50-O-dimethoxytrityl protected thymidine
in two steps with overall yields at 68% and 85%, respectively.
Oligo 5, and pro-oligos 3 and 4, which contain different
numbers of modified monomer units in a T12-sequence,
were synthesized by the standard phosphoramidite protocol,
except that a prolonged coupling time was used to couple the
nitroheterocyle-modified monomer units. In contrast to base-
sensitive acylthioalkyl and acyloxyalkyl protecting groups,5–9
both 5-nitro-2-furylmethyl and 5-nitro-2-thienylmethyl groups
are quite stable under solid synthesis conditions. After being
removed from the solid support, the crude oligos were purified
by HPLC. The isolated yield was approximately 30%. The
sequences and the structures of the synthesized pro-oligos are
shown in Scheme 2.
Antisense oligonucleotides (oligos) are important tools for
regulating specific gene expression because of their simplified
design and synthesis, along with availability of the rapidly
expanding genomics. However, their broad applications in
biological research and gene therapy has been limited by their
low stability against nucleases and poor cellular uptake. The
pro-oligos approach has emerged as one of the potential
methods to overcome the above limitations by masking the
negative charges of the phosphate backbone with biodegradable
protection groups.1–5 However, special non-basic and non-
nucleophilic solid supports, and nucleobase protecting groups
have to be applied in the solid-phase synthesis, because such
protecting groups are sensitive to base and/or nucleophilic
treatment.5–9 More recently, the development of hypoxia-
activated prodrugs that preferentially ‘release’ a therapeutic
entity under hypoxic conditions has attracted more and
more attention.10–14 Being intrigued by the hypoxia-activated
modification, we designed pro-oligos containing nitroheterocycle-
modified phosphate internucleoside linkages in order to not
only make the synthesis of pro-oligos commercially applicable,
but also achieve better tumor selectivity. Herein, we report
our design, synthesis and bio-evaluation of hypoxia-activated
pro-oligos.
The mechanistic rationale for the proposed activation of
pro-oligos is shown in Scheme 3. Pro-oligos with nitrohetero-
cycle modifications can be bioreduced by reductase via a series
of one-electron reduction processes to form the hydroxylamino-
heterocycle- or aminoheterocycle-modified oligos, followed by
cleavage of the heterocycle groups and the release of the
desired T-sequence.14,15
The hydrolysis of pro-oligos by nitroreductase was monitored
by reversed-phase HPLC. 3a and 3b were incubated with
E. coli nitroreductase either in aerobic (air) or hypoxic (nitrogen)
conditions. As we expected, both pro-oligos were hydrolyzed
more readily in N2 than in air, as suggested by shorter lifetimes
of 3–4 h in N2 compared to those of 8–9 h in air (Fig. 1(a)
and Table S1w). The difference between the nitroheterocyclic
pro-oligos in N2 and in air is mainly due to the nitro anion
radical, which is the first intermediate during the multi-step
As shown in Scheme 1, the two thymidine phosphoramidite
monomer building blocks, 50-O-dimethoxytrityl thymidine-30-
O-[(5-nitro-2-furyl)methyl N,N-diisopropylphosphoramidite]
a Key Laboratory of Chemical Biology, Guangdong Province,
Graduate School at Shenzhen, Tsinghua University, Shenzhen,
518055, P.R. China. E-mail: jiangyy@sz.tsinghua.edu.cn
b Key Laboratory of Bioorganic Phosphorus Chemistry & Chemical
Biology, Ministry of Education, Department of Chemistry, Tsinghua
University, Beijing, 100084, P.R. China
c Shanghai GenePharma Co. Ltd, 1011 Halley Road, Z.-J. High Tech
Park, Shanghai, 201203, P.R. China
d School of Medicine, Tsinghua University, Beijing, 100084,
P.R. China
w Electronic supplementary information (ESI) available: Details on
experimental procedures of the synthesis of monomer building block
2a, 2b and oligonucleotides, hydrolysis with E. coli nitroreductase,
CEM cell extract preparation, hydrolysis in CEM cell extract, hydro-
lysis with snake venom phosphodiesterase, DNase I and in 10%
fetal bovine serum, and confocal microscopic analysis. See DOI:
10.1039/b903331a
Scheme 1 The synthesis of monomer building blocks with nitro-
heterocycle-modified phosphoramidites. Reagents and conditions:
(a) bis(N,N-diisopropylamino)chlorophosphine, N,N-diisopropylethyl-
amine, CH2Cl2, RT; (b) 5-nitro-2-hydroxymethylfuran or 5-nitro-2-
hydroxymethylthiophene, 4,5-dicyanoimidazole (DCI), CH2Cl2, RT.
ꢀc
This journal is The Royal Society of Chemistry 2009
3216 | Chem. Commun., 2009, 3216–3218

View Article Online
Scheme 2 The structures and sequences of synthesized pro-oligos 3
and 4, and oligo 5.
Fig. 2 A comparison of the stability of pro-oligos 4a (,) and 4b (’),
and oligo 5 (K) in the presence of SVPDE (a) and DNase I (b), as well
as in fetal bovine serum (c).
oligo 5 was rapidly hydrolyzed under the effect of both
SVPDE and DNase I with half-lives at 0.30 and 0.35 h,
respectively. Both pro-oligos 4a and 4b demonstrated significantly
improved nuclease resistance compared to oligo 5, with much
longer half-lives in the range of 8–10 h. The stability of these
three oligos in 10% fetal bovine serum was also studied. The
half-life of oligo 5 was only 0.2 h, whereas the half-lives of 4a
and 4b were 8.2 h and 8.5 h, respectively. As a comparison, the
reported half-lives of a pro-oligo with an S-acyl-2-thioethyl
group is 3 h in human serum and 19 h against SVPDE8 Our
system has shown a significantly enhanced nuclease resistance
in human serum. The stability enhancement of pro-oligos 4a
and 4b compared to non-modified oligo 5 strongly suggest that
pro-oligos with both nitrofuryl and nitrothienyl modifications
have good resistance against different nucleases, and thus have
potential applications in gene therapy.
Scheme 3 The mechanism of the hypoxia-activated release of the
normal oligo from pro-oligos.
A key challenge for the biomedical application of pro-oligos
is to improve the cellular uptake. In order to evaluate the
trans-membrane ability of our pro-oligos, fluorescein-labelled
oligo and pro-oligos with 4–6 protecting nitroheterocylcle
groups were synthesized (sequences are shown in Scheme 4).
After incubation with HeLa cells without any transfecting
reagent at 37 1C for 3 h, no fluorescence was detected for the
non-modified oligo under a confocal fluorescence microscope,
while as a distinctive comparison, all pro-oligos showed
fluorescence emission with different intensities, indicating
different degrees of cellular uptake (Fig. 3). The cellular
uptake was related to the lipophilicity of the oligos. As we
expected, the cellular uptake of pro-oligos increased with the
number of protecting groups. The most lipophilic pro-oligos,
9a and 9b, were more efficiently taken up by cells, whereas the
fluorescence signal for the most hydrophobic pro-oligos, 7a
and 7b, was remarkably reduced. The cellular uptake was
similar to that of the S-acyl-2-thioethyl protected pro-oligos.7,16
In summary, we have demonstrated for the first time the
design of a hypoxia-activated pro-oligo with T-sequence
Fig. 1 (a) The bio-degradation of pro-oligos 3a and 3b by E. coli
nitroreductases in N2 and in air: 3a in N2 (J), 3a in air (K), 3b in N2
(n), and 3b in air (.). (b) The bio-degradation of pro-oligos 4a and 4b
in CEM cell extract in N2 and in air: 4a in N2 (J), 4a in air (K), 4b in
N2 (n), and 4b in air (.).
process of reduction, undergoing oxidation back to the nitro
compound under the effect of oxygen.15
To reflect the intracellular degradation of pro-oligos, 4a and
4b were incubated in human leukemia CEM cell extract. As
shown in Fig. 1(b), pro-oligos 4a and 4b were hydrolyzed to
the model oligo, 5, with half-lives around 4 h under N2,
whereas the half-lives of 4a and 4b were more than 9 h
in air. The faster release of the pro-oligos in N2 than in air
by nitroreductases, both in vitro and in simulated in vivo
environment, suggests that such pro-oligo design could be
potentially used in the specific targeting of tumor cells which
are under the most hypoxic conditions, which in turn can
realize the tumor cell recognition and lower the side-effects of
antisense oligos to normal cells.
The stability of pro-oligos 4a and 4b against exonuclease
(snake venom phosphodiesterase, SVPDE) and endonuclease
(deoxyribonuclease I, DNase I) was evaluated and compared
to model oligo 5 (shown in Fig. 2 and ESI, Table S2w). Model
Scheme
4
The structures of the fluorescein-labelled oligo and
pro-oligos.
ꢀc
This journal is The Royal Society of Chemistry 2009
Chem. Commun., 2009, 3216–3218 | 3217

View Article Online
National Natural Science Foundation (20872077, 90813013)
for financial support.
Notes and references
1 N. Mignet, F. Morvan, B. Rayner and J. L. Imbach, Bioorg. Med.
Chem. Lett., 1997, 7, 851–854.
2 R. P. Iyer, D. Yu and S. Agrawal, Bioorg. Chem., 1995, 23, 1–21.
3 F. Morvan, J. J. Vasseur, E. Vives, B. Rayner and J. L. Imbach,
Pharm. Aspects Oligonucleotides, 2000, 79–97.
4 A. Hohlfeld and C. Meier, Nucleosides, Nucleotides Nucleic Acids,
2003, 22, 1123–1125.
5 P. Poijaervi, P. Heinonen, P. Virta and H. Loennberg, Bioconjugate
Chem., 2005, 16, 1564–1571.
6 T. Guerlavais, A. Meyer, J. L. Imbach and F. Morvan, Bioorg.
Med. Chem. Lett., 2001, 11, 2813–2816.
7 K. Alvarez, J. J. Vasseur and J. L. Imbach, Nucleosides, Nucleotides
Nucleic Acids, 1999, 18, 1435–1436.
8 G. Tosquellas, K. Alvarez, C. Dell’Aquila, F. Morvan,
J. J. Vasseur, J. L. Imbach and B. Rayner, Nucleic Acids Res.,
1998, 26, 2069–2074.
Fig. 3 Incubation of oligo 6 and pro-oligos 7–9 at 10 mM concentra-
tion with Hela cells at 37 1C for 3 h. The first row from left to right
shows oligo 6, and pro-oligos 7a, 8a and 9a; the second row from left
to right shows pro-oligos 7b, 8b and 9b.
9 J. C. Bologna, G. Tosquellas, F. Morvan, B. Rayner and
J. L. Imbach, Nucleosides, Nucleotides Nucleic Acids, 1999, 18,
1433–1434.
10 T. W. Failes and T. W. Hambley, Dalton Trans., 2006, 1895–1901.
11 K. O. Hicks, H. Myint, A. V. Patterson, F. B. Pruijn, B. G. Siim,
K. Patel and W. R. Wilson, Int. J. Radiat. Oncol., Biol., Phys.,
2007, 69, 560–571.
12 C. Schultz, Bioorg. Med. Chem., 2003, 11, 885–898.
13 A. V. Patterson, K. J. Williams, R. L. Cowen, M. Jaffar,
B. A. Telfer, M. Saunders, R. Airley, D. Honess, A. J. Van der
Kogel, C. R. Wolf and I. J. Stratford, Gene Ther., 2002, 97,
946–954.
14 R. F. Borch, J. Liu, J. P. Schmidt, J. T. Marakovits, C. Joswig,
J. J. Gipp and R. T. Mulcahy, J. Med. Chem., 2000, 43, 2258–2265.
15 M. Jaffar, K. J. Williams and I. J. Stratford, Adv. Drug Delivery
Rev., 2001, 53, 217–228.
synthesized using simple monomer preparation and the
commercial phosphoramidite route. These pro-oligos, which
contain mixed normal linkages and nitroheterocyle-protecting
phosphate internucleoside linkages, can be readily cleaved to
form normal oligos with good hypoxia-selectivity in vitro
under the effect of reductases, as well as in tumor cell extract
by cellular reductases. Moreover, the pro-oligos demonstrated
significantly improved nuclease resistance and cellular uptake.
Such favorable properties make these hypoxia-activated
pro-oligos good candidates for therapeutics to specifically
inhibit tumor related gene expression with low toxicity to
normal cells.
The authors would like to thank the Ministry of Science
and Technology of China (2007AA02Z160) and the Chinese
16 J. C. Bologna, E. Vives, J. L. Imbach and F. Morvan, Antisense
Nucleic Acid Drug Dev., 2002, 12, 33–41.
ꢀc
This journal is The Royal Society of Chemistry 2009
3218 | Chem. Commun., 2009, 3216–3218