Synthesis and characterization of DNA duplexes containing an N4C-ethyl-N4C interstrand cross-link

Article abstract of DOI:10.1021/ja003340t

Short DNA duplexes containing an N⁴C-ethyl-N⁴C interstrand cross-link, C-C, were synthesized on controlled pore glass supports. Duplexes having two, three, or four A/T base pairs on either side of the C-C cross-link and terminating with a C₄ overhang at their 5′-ends were prepared. The cross-link was introduced using a convertible nucleoside approach. Thus, an oligonucleotide terminating at its 5′-end with O⁴-triazoyl-2′-deoxyuridine was first prepared on the support. The triazole group of support-bound oligomer was displaced by the aminoethyl group of 5′-dimethoxytrityl-3′-O-tert-butyldimethylsilyl-N⁴- (2-aminoethyl)deoxycytidine to give the cross-link. The dimethoxytrityl group was removed, and the upper and lower strands of the duplex were extended from two 5′-hydroxyl groups of the cross-link using protected nucleoside 3′-phosphoramidites. The tert-butyldimethylsilyl group of the resulting partial duplex was then removed, and the chain was extended in the 3′-direction from the resulting 3′-hydroxyl of the cross-link using protected nucleoside 5′-phosphoramidites. The cross-linked duplexes were purified by HPLC and characterized by enzymatic digestion and MALDITOF mass spectrometry. Duplexes with three or four A/T base pairs on either side of the C-C cross-link gave sigmoidal shaped A₂₆₀ profiles when heated, a behavior consistent with cooperative denaturation of the A/T base pairs. Each cross-linked duplex could be ligated to an acceptor duplex using T4 DNA ligase, a result that suggests that the C-C cross-link does not interfere with the ligation reaction, even when it is located only two base pairs from the site of ligation. The ability to synthesize duplexes with a defined interstrand cross-link and to incorporate these duplexes into longer pieces of DNA should enable preparation of substrates that can be used for a variety of biophysical and biochemical experiments, including studies of DNA repair.

Full text of DOI:10.1021/ja003340t

VOLUME 123, NUMBER 15
APRIL 18, 2001
© Copyright 2001 by the
American Chemical Society
Synthesis and Characterization of DNA Duplexes Containing an
N4C-Ethyl-N4C Interstrand Cross-Link
David M. Noll,‡ Anne M. Noronha,† and Paul S. Miller*,†
Contribution from the Department of Biochemistry and Molecular Biology, School of Hygiene and
Public Health, Johns Hopkins UniVersity, 615 North Wolfe Street, and Department of Biophysics and
Biophysical Chemistry, School of Medicine, Johns Hopkins UniVersity, 725 North Wolfe Street,
Baltimore, Maryland 21205
ReceiVed September 11, 2000
Abstract: Short DNA duplexes containing an N4C-ethyl-N4C interstrand cross-link, C-C, were synthesized
on controlled pore glass supports. Duplexes having two, three, or four A/T base pairs on either side of the
C-C cross-link and terminating with a C4 overhang at their 5′-ends were prepared. The cross-link was introduced
using a convertible nucleoside approach. Thus, an oligonucleotide terminating at its 5′-end with O4-triazoyl-
2′-deoxyuridine was first prepared on the support. The triazole group of support-bound oligomer was displaced
by the aminoethyl group of 5′-dimethoxytrityl-3′-O-tert-butyldimethylsilyl-N4-(2-aminoethyl)deoxycytidine to
give the cross-link. The dimethoxytrityl group was removed, and the upper and lower strands of the duplex
were extended from two 5′-hydroxyl groups of the cross-link using protected nucleoside 3′-phosphoramidites.
The tert-butyldimethylsilyl group of the resulting partial duplex was then removed, and the chain was extended
in the 3′-direction from the resulting 3′-hydroxyl of the cross-link using protected nucleoside 5′-phosphoramidites.
The cross-linked duplexes were purified by HPLC and characterized by enzymatic digestion and MALDI-
TOF mass spectrometry. Duplexes with three or four A/T base pairs on either side of the C-C cross-link gave
sigmoidal shaped A260 profiles when heated, a behavior consistent with cooperative denaturation of the A/T
base pairs. Each cross-linked duplex could be ligated to an acceptor duplex using T4 DNA ligase, a result that
suggests that the C-C cross-link does not interfere with the ligation reaction, even when it is located only two
base pairs from the site of ligation. The ability to synthesize duplexes with a defined interstrand cross-link and
to incorporate these duplexes into longer pieces of DNA should enable preparation of substrates that can be
used for a variety of biophysical and biochemical experiments, including studies of DNA repair.
Introduction
form interstrand cross-links. Many of the bifunctional alkylators
are used clinically to treat cancer, and the DNA interstrand cross-
links formed are believed to be the primary lesion responsible
for their effectiveness as therapeutic agents.1-3 Interstrand cross-
A variety of bifunctional alkylating agents such as nitrogen
mustards, cis-platinum, cyclophosphamide, mitomycin C, and
photoreactive compounds such as psoralen react with DNA to
* Corresponding author. Phone: 410-955-3489. Fax: 410-955-2926.
E-mail: pmiller@jhsph.edu.
(1) Colvin, M. The Alkylating Agents; W. B. Saunders Co.: Philadelphia,
1982.
(2) Colvin, M. Alkylating Agents and Platinum Antitumor Compounds;
Lea & Febiger: Philadelphia and London, 1993.
(3) Pratt, W. B.; Ruddon, R. W.; Ensminger, W. D.; Maybaum, J.
CoValent DNA Binding Drugs; Oxford Press: New York, 1994.
† Department of Biochemistry and Molecular Biology, School of Hygiene
and Public Health.
‡ Department of Biophysics and Biophysical Chemistry, School of
Medicine.
10.1021/ja003340t CCC: $20.00 © 2001 American Chemical Society
Published on Web 03/23/2001

3406 J. Am. Chem. Soc., Vol. 123, No. 15, 2001
Noll et al.
links can be repaired by the cell through mechanisms that are
only now beginning to be characterized.4-14 Such repair would
be expected to reduce the therapeutic efficacy of these com-
pounds. Thus, a better understanding of repair processes could
potentially lead to the development of more effective therapeutic
agents.
Experimental Section
Materials. 5′-O-Dimethoxytrityldeoxyribonucleoside-3′-O-(â-cya-
noethyl-N,N-diisopropyl)phosphoramidites, 3′-O-dimethoxytritylthy-
midine-5′-O-(â-cyanoethyl-N,N-diisopropyl)phosphoramidite, and pro-
tected deoxyribonucleoside-controlled pore glass supports were purchased
from Glen Research, Inc., Sterling VA. Phosphoramidite solutions were
prepared using HPLC grade acetonitrile that was dried and stored over
calcium hydride. Reversed-phase HPLC was carried out using a Rainin
Microsorb-C-18 column (0.46 cm × 15 cm), and strong anion-exchange
(SAX) HPLC was carried out using a Rainin Dynamax II column (0.46
cm × 25 cm), both purchased from Varian Associates, Walnut Creek,
CA. The C-18 column was eluted with a 20 mL linear gradient of
acetonitrile in 50 mM sodium phosphate buffer (pH 5.8) at a flow rate
of 1.0 mL/min. The SAX column was eluted with 18 mL of a linear
gradient of ammonium sulfate in a buffer that contained 1 mM
ammonium acetate (pH 6.2) in 20% acetonitrile at a flow rate of 0.6
mL/min. The columns were monitored at 260 nm for analytical runs
and at 290 nm for preparative runs. Oligomers purified by HPLC were
desalted on C-18 SEP PAK cartridges (Waters Inc.). The SEP PAK
was pre-equilibrated by washing with 10 mL aliquots of acetonitrile;
50% aqueous acetonitrile and 2% acetonitrile in 50 mM sodium
phosphate, pH 5.8 (buffer A). The oligonucleotide solution was diluted
into buffer A to give a final acetonitrile concentration of 3%, and the
solution was applied to the cartridge. The cartridge was washed with
10 mL of water, and the oligomer was eluted with 3 mL of 50% aqueous
acetonitrile. Polyacrylamide gel electrophoresis was carried out on 20
cm × 20 cm × 0.75 cm gels containing 20% acrylamide and 7 M
urea. The running buffer was TBE, which contained 89 mM Tris, 89
mM boric acid, and 0.2 mM ethylenediaminetetraacetate buffered at
pH 8.0. The gel loading buffer contained 90% formamide, 0.05% xylene
cyanol, and 0.05% bromophenol blue.
A serious limitation to studying repair of DNA interstrand
cross-links is the lack of substrates of defined structure. Reaction
of DNA with bifunctional alkylating agents usually results in
the formation of many products, only a small percentage of
which is the desired interstrand cross-link. This problem can
be limited to some extent by carrying out reactions on short
oligonucleotide duplexes and isolating the interstrand cross-
linked duplex by gel or chromatography methods. This postsyn-
thetic approach has been used successfully to prepare, for
example, duplexes cross-linked with a nitrogen mustard,15,16
bifunctional pyrroles,17 mitomycin C,18,19 or N,N′-bis(2-chlo-
roethyl)nitrosourea.20 In some cases, these duplexes have been
ligated into plasmid DNAs which subsequently have been used
as substrates for repair studies.6,10,21
A more attractive synthetic strategy is a procedure in which
the cross-linked duplex is prepared directly on a solid support.
For example, Hopkins and co-workers22 have recently reported
solid support-mediated syntheses of nitrous acid cross-linked
duplexes. This direct synthetic approach has the potential
advantage of producing relatively large quantities of well-
defined compounds for use in both physical and biological
studies.
In this paper we describe the direct, support-mediated
syntheses of short DNA duplexes that contain a single N4C-
ethyl-N4C interstrand cross-link. These duplexes serve as
simple models for interstrand cross-linked DNA and are
potential substrates for DNA repair studies. In addition, we
describe some preliminary physical studies on these cross-linked
duplexes and their ligation into longer DNA duplexes.
Synthesis of 5′-O-Dimethoxytrityl-2′-deoxyuridine. A stirred solu-
tion of 2′-deoxyuridine (5 g, 22 mmol) in 110 mL of anhydrous pyridine
was treated with dimethoxytrityl chloride (8.91 g, 26.5 mmol) for 18
h at room temperature, after which an additional 500 mg of dimethox-
ytrityl chloride was added, and stirring was continued for 3 h. The
reaction solution was diluted with 25 mL of 95% ethanol and after 15
min of stirring was evaporated on a rotary evaporator. The resulting
syrup was dissolved in 300 mL of ethyl acetate, and the solution was
extracted with two 300 mL portions of 5% sodium bicarbonate followed
by 300 mL of saturated sodium chloride. The organic layer was dried
over anhydrous sodium sulfate. Following filtration, the solution was
evaporated, and the resulting foamy residue was dissolved in 100 mL
of ethyl acetate. This solution was added dropwise to 500 mL of hexane
with vigorous stirring. The resulting precipitate was collected by
centrifugation and dried under vacuum to give a solid, white powder
(10.9 g, 20.5 mmol, 93%). UV (methanol): λmax 204 nm, 231 nm, 262
nm. 1H NMR (DMSO-d6): δ 2.18 (m, 2H, H2′/2′′), 3.25 (m, 2H, H5′/
5′′), 3.78 (s, 6H, O-CH3), 3.87 (q, 1H, H3′), 4.23 (q, 1H, H4′), 5.33
(d, 1H, 3′-OH), 5.37 (d, 1H, H6), 6.14 (t, 1H, H1′), 6.8-7.4 (m, 13H,
Ph-H), 7.63 (d, 1H, H5), 11.33 (s, 0.3H, NH).
(4) Kumaresan, K. R.; Hang, B.; Lambert, M. W. J. Biol. Chem. 1995,
270, 30709-16.
(5) Calsou, P.; Sage, E.; Moustacchi, E.; Salles, B. Biochemistry 1996,
35, 14963-9.
(6) Berardini, M.; Mackay, W.; Loechler, E. L. Biochemistry 1997, 36,
3506-13.
(7) Bessho, T.; Mu, D.; Sancar, A. Mol. Cell. Biol. 1997, 17, 6822-30.
(8) Warren, A. J.; Maccubbin, A. E.; Hamilton, J. W. Cancer Res. 1998,
58, 453-61.
(9) Cullinane, C.; Bohr, V. A. Cancer Res. 1998, 58, 1400-4.
(10) Berardini, M.; Foster, P. L.; Loechler, E. L. J. Bacteriol. 1999, 181,
2878-82.
(11) Patrick, S. M.; Turchi, J. J. J. Biol. Chem. 1999, 274, 14972-8.
(12) Kumaresan, K. R.; Lambert, M. W. Carcinogenesis 2000, 21, 741-
51.
Synthesis of 5′-O-Dimethoxytrityl-3′-O-tert-butyldimethylsilyl-2′-
deoxyuridine. A stirred solution of 5′-dimethoxytrityl-2′-deoxyuridine
(10 g, 18.8 mmol) in 25 mL of anhydrous N,N-dimethylformamide
was treated with tert-butyldimethylsilyl chloride (6.2 g, 41 mmol) and
imidazole (5.6 g, 83 mmol) for 1 h at room temperature. A 10 mL
aliquot of methanol was added, and stirring was continued for 5 min.
The reaction solution was poured into 300 mL of ethyl acetate, and
the solution was extracted with two 300 mL portions of 5% sodium
bicarbonate followed by 300 mL of saturated sodium chloride. The
organic layer was evaporated and the foamy residue dissolved in 25
mL of ethyl acetate. This solution was added dropwise to 1.5 L of
vigorously stirred hexane. The hexane was decanted, and the precipitate
was washed with 200 mL of hexane. The precipitate was dissolved in
ethyl acetate and transferred to a round-bottom flask, and the solution
was evaporated to yield a tan foam (8.03 g, 12.5 mmol, 67%). 1H NMR
(DMSO-d6): δ 0.00 (s, 3H, Si-CH3), 0.01 (s, 3H, Si-CH3), 0.77 (s,
9H, -CH3), 2.20 (m, 2H, H2′/2′′), 3.17 (m, 2H, H5′/5′′), 3.73 (s, 6H,
(13) Mu, D.; Bessho, T.; Nechev, L. V.; Chen, D. J.; Harris, T. M.;
Hearst, J. E.; Sancar, A. Mol. Cell. Biol. 2000, 20, 2446-54.
(14) Kuraoka, I.; Kobertz, W. R.; Ariza, R. R.; Biggerstaff, M.;
Essignmann, J. M.; Wood, R. D. J. Biol. Chem. 2000, 275, 26632-6.
(15) Ojwang, J. O.; Grueneberg, D. A.; Loechler, E. L. Cancer Res. 1989,
49, 6529-37.
(16) Rink, S. M.; Solomon, M. S.; Taylor, M. J.; Rajur, S. B.;
McLaughlin, L. W.; Hopkins, P. B. J. Am. Chem. Soc. 1993, 115, 2551-7.
(17) Tsarouhtsis, D.; Kuchimanchi, S.; Decorte, B. L.; Harris, C. M.;
Harris, T. M. J. Am. Chem. Soc. 1995, 117, 11013-4.
(18) Rink, S. M.; Lipman, R.; Alley, S. C.; Hopkins, P. B.; Tomasz, M.
Chem. Res. Toxicol. 1996, 9, 382-9.
(19) Warren, A. J.; Hamilton, J. W. Chem. Res. Toxicol. 1996, 9, 1063-
71.
(20) Fischhaber, P. L.; Gall, A. S.; Duncan, J. A.; Hopkins, P. B. Cancer
Res. 1999, 59, 4363-8.
(21) Grueneberg, D. A.; Ojwang, J. O.; Benasutti, M.; Hartman, S.;
Loechler, E. L. Cancer Res. 1991, 51, 2268-72.
(22) Harwood, E. A.; Sigurdsson, S. T.; Edfeldt, N. B. F.; Reid, B. R.;
Hopkins, P. B. J. Am. Chem. Soc. 1999, 121, 5081-2.

N4C-Ethyl-N4C Interstrand Cross-Linked DNA Duplexes
J. Am. Chem. Soc., Vol. 123, No. 15, 2001 3407
O-CH3), 3.79 (q, 1H, H3′), 4.40 (q, 1H, H4′), 5.42 (d, 1H, H6), 6.12
(t, 1H, H1′), 6.88-7.38 (m, 14H, Ph-H), 7.70 (d, 1H, H5), 11.33 (s,
0.3H, NH).
°C for 16 h (5, n ) 3) or 48 h (5, n ) 4). The excess solution was
removed, the support (6) was transferred back to the synthesis column,
and the support was washed with two 10 mL aliquots of acetonitrile.
Chain extension was then continued in the 5′-direction after removal
of the 5′-dimethoxytrityl group from 6. After the final coupling step,
the synthesizer was programmed to acetylate the 5′-ends of the partial
cross-linked duplex to give 7. The support was then treated with 0.8
mL of anhydrous triethylamine for 32 h at room temperature. The
triethylamine was flushed from the column, and the support was washed
with two 10 mL aliquots of acetonitrile and then dried under vacuum.
The dry support was then treated with 1 mL of 1 M tetra-n-
butylammonium fluoride (TBAF) in tetrahydrofuran for 60 min at room
temperature. The TBAF solution was flushed from the support, and
the support was washed with 10 mL of 50% aqueous acetonitrile
followed by 10 mL of acetonitrile. Synthesis was then continued in
the 3′-direction using a 0.14 M solution of 3′-O-dimethoxytritylthy-
midine-5′-O-â-cyanoethyl-N,N-diisopropylphosphoramidite. The cou-
pling time was 180 s, and the 3′-terminal dimethoxytrityl group was
removed at the end of the synthesis to give support-bound oligomer 8.
Each support was treated with 400 µL of concentrated ammonium
hydroxide for 5 h at 65 °C. The supernatant was removed from the
support, and the support was washed with four 200 µL aliquots of 50%
aqueous acetonitrile. The combined supernatant and washings were
evaporated to dryness under vacuum at 37 °C. The cross-linked duplexes
were purified by HPLC. An aliquot (13 A260 units) of crude oligomer
1827 was first purified on the C-18 reversed-phase column using a
linear gradient of 2-50% acetonitrile. The product peak, which was
contaminated with shorter oligonucleotides, was desalted on a C-18
SEP PAK cartridge and then further purified on the SAX column using
a linear gradient of 0-0.5 M ammonium sulfate. The oligomer was
desalted on a C-18 SEP PAK cartridge to give 0.74 A260 unit of 1827.
Crude oligomer 1828 (45 A260 units) and 1829 (45 A260 units) were
each purified by SAX HPLC using a linear gradient of 0.001-0.8 M
ammonium sulfate. A total of 5.8 A260 units of 1828 and 4.5 A260 units
of 1829 were obtained after desalting on a SEP PAK cartridge.
Oligomer 1828 was further purified by C-18 reversed-phase HPLC
using a linear gradient of 2-20% acetonitrile. A total of 2.3 A260 units
of 1828 were obtained after desalting on a SEP PAK cartridge.
The cross-linked oligomers (0.02 A260 unit each) were phosphorylated
in 9.5 µL of solution that contained 70 mM Tris, pH 7.6, 10 mM
magnesium chloride, 5 mM dithiothreitol, and 63 µM γ-[32P]-ATP
(specific activity 17 µCi/mmol). The reactions were initiated by addition
of 0.5 µL (5 units) of T4 polynucleotide kinase and incubated at 37 °C
for 3 h. Each cross-linked oligomer migrated as a single band on a
20% polyacrylamide gel run under denaturing conditions as shown in
Figure 2.
The cross-linked oligomers were subjected to enzymatic digestion
with a combination of snake venom phosphodiesterase (SVPD) and
calf intestinal phosphatase (CIP). Oligomer 1827 (0.04 A260 unit),
dissolved in 16 µL of a solution containing 10 mM Tris (pH 8.1) and
2 mM magnesium chloride, was treated with 3 µL (6 ng) of SVPD
and 1 µL (10 units) of CIP for 16 h at 37 °C. Oligomers 1828 (0.1 A260
unit) and 1829 (0.1 A260 unit), each dissolved in 17 µL of enzyme buffer,
were treated with 2 µL of SVPD and 0.5 µL of CIP for 16 h at 37 °C.
A 10 µL aliquot of each reaction mixture was treated with an additional
2 µL of SVPD and 0.5 µL of CIP for 18 h at 37 °C. Each digest was
analyzed by C-18 reversed-phase HPLC using a linear gradient of
2-20% acetonitrile. The nucleoside ratios were calculated after dividing
the area of each peak by the appropriate extinction coefficient. The
ꢀ260 values used were as follow: dA, 14 100; dC, 7300; dT, 9000; dC-
dC, 16 240. The ꢀ260 values of dA, dC, dG, and dT were determined
experimentally in 50 mM sodium phosphate buffer, pH 5.8, that
contained 2% acetonitrile. The ꢀ260 value of dC-dC was calculated
using the extinction coefficient of N4-methyl-2′-deoxycytidine (ꢀ260
8210).23 The results are shown in Table 1. The molecular weights of
each cross-linked oligomer were determined by MALDI-TOF mass
spectrometry, and the results are shown in Table 1.
Synthesis of 5′-O-Dimethoxytrityl-3′-O-tert-butyldimethylsilyl-4-
(N-1-triazoyl)-2′-deoxyuridine (2). Phosphorous oxychloride (1.6 mL,
2.8 g, 17 mmol) was slowly added to a stirred suspension of 1,2,4-
triazole (5.8 g, 8.4 mmol) in 100 mL of anhydrous acetonitrile at 0 °C.
Anhydrous triethylamine (12 mL) was then added dropwise, and the
solution was stirred for 30 min. A solution of 5′-O-dimethoxytrityl-
3′-O-tert-butyldimethylsilyl-2′-deoxyuridine (1.4 g, 2.1 mmol) in 15
mL of anhydrous acetonitrile was then added dropwise, and stirring
was continued for 1.5 h. The reaction was quenched by addition of
150 mL of 5% sodium bicarbonate, and the aqueous solution was
extracted with ethyl acetate. The organic layer was washed with
saturated sodium chloride and dried over anhydrous sodium sulfate.
After filtration, the solvents were evaporated, and the product was
purified by silica gel flash column chromatography using ethyl acetate
as solvent. Evaporation of the solvents gave a white foam (1.0 g, 1.4
mmol, 68%). UV (95% ethanol): λmax 266 nm, 231 nm, λmin 255 nm.
1H NMR (DMSO-d6): δ 0.00 (s, 3H, Si-CH3), 0.06 (s, 3H, Si-CH3),
0.77 (s, 9H, -CH3), 2.37 (m, 2H, H2′/2′′), 3.35 (m, 1H, H5′/5′′), 3.73
(s, 6H, -OCH3), 3.93 (m, 1H, H4′), 4.40 (q, 1H, H3′), 6.11 (t, 1H,
H1′), 6.67 (d, 1H, H6), 6.8-7.3 (m, 14H, Ph-H), 8.39 (s, 0.5H,
triazole-H), 8.61 (d, 1H, H5), 9.45 (s, 0.5H, triazole-H).
Synthesis of 5′-O-Dimethoxytrityl-3′-O-tert-butyldimethylsilyl-N4-
(2-aminoethyl)-2′-deoxycytidine (3). A solution of 2 (1 g, 1.4 mmol)
in 15 mL of anhydrous pyridine was added dropwise with stirring to a
solution of ethylenediamine (3.6 g, 4.0 mL, 60 mmol) in 50 mL of
anhydrous pyridine. Stirring was continued for 1 h, and the solvents
were then evaporated. Residual pyridine was removed by coevaporation
with 95% ethanol. The resulting foamy residue was dissolved in 20%
methanol/1% triethylamine/chloroform (v/v) and the product purified
by silica gel flash chromatography to give a glassy solid (960 mg, 1.4
mmol, 100%). UV (95% ethanol): λmax 275 nm, 235 nm, λmin 258 nm.
1H NMR (DMSO-d6): δ -0.015 (s, 3H, Si-CH3), -0.073 (s, 3H, Si-
CH3), 0.77 (s, 9H, -CH3), 2.18 (m, 2H, H2′/2′′), 2.65 (t, 2H, -CH2-
CH2-NH2), 3.24 (t, 2H, -CH2-CH2-NH2), 3.42 (m, 2H, H5′), 3.74
(s, 6H, -OCH3), 3.79 (m, 1H, H3′), 4.08 (s, 0.9H, -NH2), 4.39 (q,
1H, H4′), 5.62 (d, 1H, H6), 6.15 (t, 1H, H1′), 6.98-7.38 (m, 14H,
Ph-H), 7.67 (d, 1H, H5).
Synthesis of 1,2-Bis-(N4-2′-deoxycytidylyl)ethane (1). A solution
of 5′-O-dimethoxytrityl-3′-O-tert-butyldimethylsilyl-4-(N-1-triazoyl)-
2′-deoxyuridine (2, 1.3 mg, 2 µmol) and 5′-O-dimethoxytrityl-3′-O-
tert-butyldimethylsilyl-N4-(2-aminoethyl)-2′-deoxycytidine (3, 3.2 mg,
5 µmol) in 50 µL of anhydrous pyridine was incubated at room
temperature for 16 h. Examination by silica gel TLC indicated that the
reaction was complete. The solvents were evaporated, and the residue
was treated with 200 µL of 0.1 N hydrochloric acid for 30 min at 65
°C. The solvents were evaporated, the residue was dissolved in 50%
aqueous acetonitrile, and the solution was examined by C-18 reversed-
phase HPLC using a 2-20% linear gradient of acetonitrile. Two
compounds were observed:N4-(2-aminoethyl)-2′-deoxycytidine [1H
NMR (D2O): δ 0.70 (m, 2H, H2′), 1.41 (t, 2H, N4-CH2-CH2-NH2),
1.95 (t, 2H, N4-CH2-CH2-NH2), 2.17 (m, 2H, H5′), 2.43 (q, 1H,
H4′), 2.82 (m, 1H, H3′) 4.26 (d, 1H, H5), 4.66 (t, 1H, H1′), 6.13 (d,
1H, H6)] and 1,2-bis-(N4-deoxycytidylyl)ethane (1) [UV (50% aqueous
1
acetonitrile): λmax 274 nm, λmin 240 nm. H NMR (D2O): 0.63 (m,
4H, H2′), 2.00 (s, 4H, N4-CH2-CH2-N4), 2.14 (m, 4H, H5′), 2.42
(q, 2H, H4′), 2.80 (m, 2H, H3′), 4.28 (d, 2H, H5), 4.62 (t, 2H, H1′),
6.05 (d, 2H, H6). The molar ratio of N4-(2-aminoethyl)-2′-deoxycytidine
to 1 was 1.47:1.
Syntheses of N4C-Ethyl-N4C Cross-Linked Duplexes. 5′-d-U-
(T)n∼CPG (5), where n ) 2, 3, or 4 and U is 4-triazoyl-2′-deoxyuridine,
was prepared on the DNA synthesizer. The concentration of the
protected nucleoside-3′-O-â-cyanoethyl-N,N-diisopropyl phosphora-
midites was 0.15 M, the coupling time was 120 s, and the synthesizer
was programmed to remove the 5′-terminal dimethoxytrityl group after
the last coupling step. The oligomer-derivatized support was transferred
to a 500 µL conical Reacti-Vial (Pierce Chemical Co.) and treated with
150 µL of a 0.1 M solution of nucleoside 3 in dry pyridine for 16 h at
room temperature (5, n ) 2) or 150 µL of 0.2 M of nucleoside 3 at 37
Thermal Denaturation Experiments. The cross-linked duplexes
(23) Handbook of Biochemistry and Molecular Biology, 3rd ed.; Nucleic
Acids, Vol. I; Fasman, G. D., Ed.; CRC Press: Cleveland, OH, 1975; p
143.

3408 J. Am. Chem. Soc., Vol. 123, No. 15, 2001
Noll et al.
Table 1. Nucleoside Ratios and Mass Spectral Data for
Cross-Linked Duplexes
Scheme 1
nucleoside ratio
mass
cross-linked duplex
expected
found
expected
5264
found
5269
1827
dA
2.00
4.00
2.00
1.00
1.87
3.91
2.15
1.00
dC
dT
dC-dC
1828
dA
6498
7733
6497
7735
1.00
1.33
1.00
0.33
0.92
1.32
1.00
0.33
dC
dT
dC-dC
1829
dA
1.00
1.00
1.00
0.25
1.07
1.11
1.00
0.28
dC
dT
dC-dC
Figure 1. Structure of the N4C-ethyl-N4C cross-link and sequences
of the C-C cross-linked duplexes.
duplex contains a central C-C cross-link flanked on either side
by two (1827), three (1828), or four (1829) A-T base pairs.
The C4 overhangs at the 5′-ends of each duplex provide sites
for ligation of the cross-linked duplexes into longer pieces of
DNA.
Examination of molecular models suggests that the ethyl
group that links the two opposing cytosines can be accom-
modated in the major groove with minimal distortion of the
B-form helix. As far as we know, this type of cross-link has
not been observed when DNA is treated with alkylating agents.
However, Romereo et al. recently reported isolation of an N3C-
N3C cross-link that results from reaction of mechlorethamine
with DNA that contains a C/C mismatch.25
or their constituent strands were dissolved in a buffer containing 50
mM Tris, pH 7.2, 100 mM sodium chloride, and 0.5 mM ethylenedi-
amine tetraacetate. The thermal denaturation experiments were carried
out using a Cary 3E UV-vis spectrophotometer fitted with a thermo-
stated sample holder and temperature controller as previously de-
scribed.24 Duplexes were heated from 0 to 70 °C at a rate of 0.5 °C/
min, and the absorbance at 260 nm was recorded as a function of the
temperature.
Ligation Reactions. Both 5′-hydroxyl groups of each cross-linked
duplex were end-labeled using γ-[32P]-ATP and T4 polynucleotide
kinase as described above. Following the reaction, the kinase was
inactivated by heating. A non-cross-linked control duplex, 1830, was
prepared by annealing d-[32P]-pCCCCAAAACTTTT and d-pC-
CCCAAAAGTTTT. An acceptor duplex, 1831, was prepared by
annealing equimolar amounts of d-CGTCCTCACTCCTGG and d-
pGGGGCCAGGAGTGAGGACG. Ligation reactions were carried out
with 0.1 µM cross-linked duplex or control duplex 1830 and 0.4 µM
acceptor duplex 1831 in a buffer that contained 50 mM Tris-HCl
(pH7.5), 10 mM magnesium chloride, 100 mM sodium chloride, 10
mM dithiothreitol, 1 mM ATP, and 25 µg/mL bovine serum albumin.
The reactions were initiated by the addition of 5 units of T4 DNA
ligase, and the reaction mixtures were incubated at 16 °C for 30 min.
The reactions were quenched by the addition of 90% formamide loading
buffer. Samples were electrophoresed on a denaturing 20% polyacry-
lamide gel and visualized by autoradiography as shown in Figure 5.
The N4C-ethyl-N4C (C-C) cross-link itself can be syn-
thesized readily in solution using a convertible nucleoside
approach,26-28 as shown in Scheme 1. Thus, in a series of test
reactions, 5′-O-dimethoxytrityl-3′-O-tert-butyldimethylsilyl-4-
(N-1-triazoyl)-2′-deoxyuridine (2) was converted to the protected
N4-(2-aminoethyl)deoxycytidine derivative (3) by reaction with
ethylenediamine at room temperature in pyridine. Subsequent
reaction in pyridine between 2 and an excess of aminoethyl
nucleoside 3 at room temperature results in formation of the
fully protected C-C cross-link, 4. Displacement of the triazole
group of 2 by the 2-amino group of 3 was complete after 16 h
at room temperature, as judged from silica gel TLC. Treatment
of 4 with 0.1 N hydrochloric acid removed the dimethoxytrityl
and tert-butyldimethylsilyl groups to give 1, which was
1
Results and Discussion
characterized by H NMR.
(25) Romero, R. M.; Mitas, M.; Haworth, I. S. Biochemistry 1999, 38,
Syntheses of Cross-Linked Duplexes. Oligodeoxyribonucle-
otide duplexes were synthesized that contain an N4C-ethyl-
N4C cross-link (1) whose structure is shown in Figure 1. The
sequences of these duplexes are also shown in Figure 1. Each
3641-8.
(26) Webb, T.; Matteucci, M. D. J. Am. Chem. Soc. 1986, 108, 2764-
5.
(27) MacMillan, A. M.; Verdine, G. L. J. Org. Chem. 1990, 55, 5931-
3.
(24) Kean, J. M.; Cushman, C. D.; Kang, H. M.; Leonard, T. E.; Miller,
P. S. Nucleic Acids Res. 1994, 22, 4497-503.
(28) Allerson, C. R.; Chen, S. L.; Verdine, G. L. J. Am. Chem. Soc.
1997, 119, 7423-33.

N4C-Ethyl-N4C Interstrand Cross-Linked DNA Duplexes
J. Am. Chem. Soc., Vol. 123, No. 15, 2001 3409
Scheme 2
with a solution of tetra-n-butylammonium fluoride (TBAF) in
tetrahydrofuran suggested that some degradation of the support-
bound oligomer had occurred. Similar problems were reported
during the syntheses of branched oligoribonucleotides29,30 and
during the syntheses of oligoribonucleotides carrying a 5′-O-
silyl protecting group.31 We reasoned that a possible cause of
this degradation could be attack of the 3′-oxygen anion,
produced during fluoride ion cleavage of the TBS group, on
adjacent phosphotriester linkages. To test this hypothesis, a 15-
mer, d-T15, was prepared. Treatment of support-bound 15-mer
with TBAF followed by cleavage of the oligomer from the
support with ammonium hydroxide gave the same HPLC profile
as material treated only with ammonium hydroxide. This result
suggests that fluoride ion alone does not cleave the cyanoeth-
ylphosphotriester linkages of the oligomer. The 5′-hydroxyl of
the support-bound 15-mer was converted to its tert-butyldim-
ethylsilyl derivative. Treatment of this TBS-d-T15 with TBAF
followed by ammonium hydroxide treatment gave rise to shorter
n - 1 and n - 2 oligomers, as judged from the strong anion-
exchange HPLC. In a separate experiment, the support-bound
TBS-d-T15 was first treated with anhydrous triethylamine.29,30
This treatment removes cyanoethyl phosphate protecting groups
but does not cleave the oligomer from the support.32 The
triethylamine-treated oligomer was then reacted with TBAF,
followed by ammonium hydroxide cleavage from the support.
Examination by HPLC showed the absence of the shorter
oligomers seen when TBS-d-T15 was treated with TBAF alone.
TBAF treatment did not appear to cleave significant amounts
of the oligomer from the support. Thus, 80-90% of the expected
OD was recovered from the TBAF-treated oligomer.
Support-bound partial duplex 7 was pretreated with triethy-
lamine followed by reaction with TBAF for 60 min at room
temperature. In initial experiments, approximately 70% of the
TBS groups were removed when the support was treated with
commercial TBAF solution. Virtually quantitative removal could
be achieved, however, if the TBAF solution was stored over
molecular sieves prior to use. This treatment most likely removes
traces of water that suppress the fluoride-mediated cleavage of
the silyl ether. Subsequent experiments showed that brief (10
min) exposure to dry TBAF was sufficient to remove the tert-
butyldimethylsilyl group, thus lessening the possibility of
cleaving the oligomer from the support.
The partial duplex was then extended in the 3′-direction using
3′-O-dimethoxytritylthymidine-5′-O-phosphoramidite to give 8.
The coupling reactions appeared to proceed in >90% yield, as
judged from the intensity of the color of the trityl cation released
after each synthetic cycle. The 3′-dimethoxytrityl group was
removed after the last coupling step, and the oligomer was
deprotected and cleaved from the support by treatment with
concentrated ammonium hydroxide solution. Following depro-
tection, each cross-linked duplex was purified by strong anion-
exchange HPLC alone or in combination with C-18 reversed-
phase HPLC.
The convertible nucleoside approach was combined with an
orthogonal synthesis scheme to produce the C-C cross-linked
duplexes as outlined in Scheme 2. Oligo-T terminated with a
4-triazoyl-2′-deoxyuridine (5) was prepared on a controlled pore
glass support using the DNA synthesizer. The support was
removed from the synthesizer and treated with a solution of
5′-O-dimethoxytrityl-3′-O-tert-butyldimethylsilyl-N4-(2-amino-
ethyl)-2′-deoxycytidine (3) in pyridine at 37 °C for periods up
to 48 h. To determine the extent of conversion to the cross-
linked oligomer, portions of the support were deprotected by
sequential treatment with concentrated ammonium hydroxide
and 80% aqueous acetic acid, and the products were analyzed
by reversed-phase HPLC. Unlike the result for the reaction in
solution, the percent conversion did not exceed 70%, even after
prolonged incubation. Failure to completely convert the 4-tria-
zoyl-dU to the C-C cross-link was not due to hydrolysis of
the 4-triazoyl-dU to deoxyuridine during the incubation. Thus,
digestion of the crude oligonucleotide with a combination of
snake venom phosphodiester and calf intestinal phosphatase did
not yield deoxyuridine. It appears that a portion of support-
bound oligomer 6 is inaccessible to reaction with the somewhat
bulky aminoethyl nucleoside 3. This may be due to steric
blocking caused by adjacent oligomer chains on the surface of
the support. Alternatively or additionally, some of the oligomer
chains may be sequestered in pores on the support and may
thus be unable to react with 3.
The 5′-terminal dimethoxytrityl group of 6 was removed, and
the upper and lower strands of the cross-link were extended in
the 5′-direction. Because of the symmetry of the duplex, the
upper and lower strands have the same sequence and can thus
be synthesized simultaneously. To maximize coupling yields,
the phosphoramidite concentration was increased from 0.1 to
0.15 M, and the coupling times were increased from 60 to 120
s. Under these conditions, coupling appeared to be almost
quantitative, as judged from the intensity of the color of the
dimethoxytrityl cation released after each coupling cycle. After
the final detritylation, the 5′-hydroxyl groups on the upper and
lower strands of the duplex were each capped with acetic
anhydride to give the 5′-acetylated, support-bound oligomer 7.
Selective removal of the tert-butyldimethylsilyl group from
the 3′-position of the cross-link allows extension of 3′-end of
the upper strand of the duplex. Braich and Damha have reported
previously using a similar strategy to extend one chain of a
branched RNA.29 In our hands, initial attempts to remove the
3′-O-tert-butyldimethylsilyl group of oligomer 7 by treatment
The synthetic scheme described above is similar to that used
by Harwood and co-workers to prepare duplexes that contain a
G-G nitrous acid cross-link.22 In our case, the C-C cross-link
is introduced using the convertible nucleoside approach on the
support-bound oligomer, whereas the G-G cross-link was
introduced in the form of the protected phosphoramidite. For
C-C cross-linked duplexes of the type described here, the
(30) Kierzek, R.; Kopp, D. W.; Edmonds, M.; Caruthers, M. H. Nucleic
Acids Res. 1986, 14, 4751-64.
(31) Scaringe, S. A.; Wincott, F. E.; Caruthers, M. H. J. Am. Chem.
Soc. 1998, 120, 11820-1.
(29) Braich, R. S.; Damha, M. J. Bioconjugate Chem. 1997, 8, 370-7.
(32) Hamma, T.; Miller, P. S. Biochemistry 1999, 38, 15333-42.

3410 J. Am. Chem. Soc., Vol. 123, No. 15, 2001
Noll et al.
Figure 3. Reversed-phase HPLC analysis of duplex 1829 after
digestion with snake venom phosphodiesterase and calf intestinal
phosphatase. The HPLC conditions are given in the Experimental
Section.
Figure 2. Polyacrylamide gel electrophoresis of cross-linked duplexes.
A linear 9-mer, d-CCCCAACTT (lane 1), and cross-linked duplexes
1827 (lane 2), 1828 (lane 3), and 1829 (lane 4) were each 5′-end labeled
with γ-[32P]-ATP and subjected to electrophoresis on a 20% polyacry-
lamide gel run under denaturing conditions. XC and BPB are xylene
cyanol and bromophenol blue dyes, respectively.
convertible nucleoside approach provides a versatile method for
introducing cross-links of varying chain length into the duplex
without the necessity of synthesizing individual protected cross-
link phosphoramidites. Such duplexes could provide interesting
substrates for physical studies and DNA repair studies.
Characterization of Cross-Linked Duplexes. The duplexes
were phosphorylated using γ-[32P]-ATP and polynucleotide
kinase and analyzed by electrophoresis on a denaturing 20%
polyacrylamide gel. As shown in Figure 2, each duplex migrated
as a single band on the gel in accordance with the size of the
duplex. Duplex 1829, which contains 26 nucleotides, migrated
with the xylene cyanol tracking dye. This mobility is similar to
that of a linear 28-mer, which comigrates with xylene cyanol
on this gel.33
The cross-linked duplexes could be hydrolyzed to their
component nucleosides and C-C cross-link by treatment with
a combination of snake venom phosphodiesterase, a 3′-exonu-
clease, and bacterial alkaline phosphatase. It appears that
hydrolysis of phosphodiester linkages in the vicinity of the cross-
link is slow. Thus, prolonged incubation, and in the cases of
duplexes 1828 and 1829 addition of a second aliquot of snake
venom phosphodiester, were required to effect complete deg-
radation. The digests were analyzed by C-18 reversed-phase
HPLC. A typical digest for oligomer 1829 is shown in Figure
3. The C-C cross-link produced in the digests had a retention
time identical to that of C-C synthesized in solution. As shown
in Table 1, the ratios of the component nucleosides and the
cross-link were consistent with the structure of each duplex.
In addition to enzymatic digestion, the duplexes were also
characterized by MALDI-TOF mass spectrometry. As indicated
in Table 1, the observed molecular weights were consistent with
the expected molecular weights for each cross-linked duplex.
Thermal Denaturation. The thermal stabilities of the cross-
linked duplexes were examined by monitoring their UV
absorbance at 260 nm as a function of temperature. The results
are shown in Figure 4. In addition to studying the cross-linked
duplexes, we also monitored absorbance vs temperature of d-C4-
(A)nC(T)n,which represent the component strands of the cross-
linked duplexes, and non-cross-linked duplexes formed by
d-C4(A)nC(T)n/d-C4(A)nG(T)n. The latter duplexes contain a
Figure 4. Thermal denaturation profiles of the cross-linked duplexes.
(A) d-C4A2CT2 (b); d-C4A2CT2/d-C4A2GT2 (0); and 1827 (O). (B)
d-C4A3CT3 (b); d-C4A3CT3/d-C4A3GT3 (0); and 1828 (O). (C) d-C4A4-
CT4 (b); d-C4A4CT4/d-C4A4GT4 (0); and 1829 (O). The experiments
were carried out in buffer that contained 50 mM Tris, pH 7.2, 100
mM sodium chloride, and 0.5 mM ethylenediamine tetraacetate. The
right axes show the A260 of the cross-linked duplexes, and the left axes
show the A260 of the control oligomers.
C-G base pair in the position occupied by the C-C cross-link
of the cross-linked duplexes.
Each of the d-C4(A)nC(T)n oligomers, where n ) 2, 3, or 4,
essentially showed a linear increase in A260 when heated over
the temperature range 0-70 °C. This behavior is consistent with
a lack of duplex formation. Duplex formation would require
(33) Maniatis, T.; Fritsch, E. F.; Sambrook, J. Molecular Cloning. A
Laboratory Manual; Cold Spring Harbor Laboratory: Cold Spring Harbor,
NY, 1982.

N4C-Ethyl-N4C Interstrand Cross-Linked DNA Duplexes
J. Am. Chem. Soc., Vol. 123, No. 15, 2001 3411
the presence of an unstable C/C mismatch in the center of the
duplex, and this most likely accounts for the failure of these
oligomers to form duplexes under these conditions.
The cross-linked duplexes showed different behavior. Duplex
1827 displayed a nonlinear increase in absorbance over the
temperature range 0-20 °C. This behavior suggests that minimal
base-pairing interaction takes place in this short cross-linked
duplex. In contrast, duplexes 1828 and 1829 each exhibited
distinct sigmoidal transitions with inflection points at 30 and
39 °C, respectively. These transition curves are qualitatively
similar to the melting curve observed for non-cross-linked
duplex d-C4(A)4C(T)4/d-C4(A)4G(T)4, whose Tm is 16 °C. Thus,
the shapes of the thermal transition curves of duplexes 1828
and 1829 suggest that the A/T base pairs of these cross-linked
duplexes denature in a cooperative manner. Because the C-C
cross-link is not hydrolyzed under these conditions, the strands
of these cross-linked duplexes cannot completely separate as
in the case of d-C4(A)4C(T)4/d-C4(A)4G(T)4. The presence of
the cross-link increases the thermal stability of the adjacent A/T
base pairs. Thus, the transition temperature of 1829 is 23 °C
higher than the Tm of the non-cross-linked duplex d-C4(A)4C-
(T)4/d-C4(A)4G(T)4. Similar cooperative thermal denaturation
and increases in stability have been observed in short DNA
duplexes that contain 4′-(hydroxymethyl)-4,5′,8-trimethylpso-
ralen34 or N2-G-mitomycin C-N2-G19 interstrand cross-links.
Ligation Experiments. For repair studies and other biophysi-
cal characterization studies, it will be necessary to be able to
ligate the cross-linked duplexes into longer duplex or plasmid
DNA. The ability of the C-C cross-link duplexes to serve as
substrates for DNA ligase was tested by following the ligation
of 5′-32P-labeled 1827, 1828, or 1829 to a 15-base pair acceptor
duplex, 1831, whose sequence is given in the Experimental
Section. Because the cross-linked duplexes are symmetrical,
successful ligation requires that each duplex undergo two
individual ligation events. Thus both 5′-CCCC overhangs of
the duplex are potential substrates for the ligation reaction.
The ligation reactions were carried out using a 4-fold molar
excess of the acceptor duplex, which is a 2-fold ligation excess
given that the cross-linked duplexes contain two ligation sites,
under salt conditions similar to those used in the thermal
denaturation studies. Each ligation reaction was incubated at
16 °C for 30 min prior to electrophoresis on a 20% denaturing
polyacrylamide gel.
Figure 5. DNA ligation experiments. Control duplex 1830 (lanes 1
and 2) and cross-linked duplexes 1827 (lanes 3 and 4), 1828 (lanes 5
and 6), and 1829 (lanes 7 and 8) were each incubated with acceptor
duplex 1831 in the absence (lanes 1, 3, 5, 7) or presence (lanes 2, 4, 6,
8) of T4 DNA ligase, and the reaction mixtures were subjected to
electrophoresis on a 20% polyacrylamide gel run under denaturing
conditions.
ing of one cross-link duplex and one acceptor duplex, followed
by formation of the fully ligated product consisting of one cross-
linked duplex and two acceptor duplexes. Thus, based on these
titration experiments, the bands shown in lanes 4, 6, and 8
correspond to duplexes that contain one cross-linked duplex
ligated to two acceptor duplexes.
Each of the cross-linked duplexes was completely converted
to its ligated product under the conditions of the reaction. This
result suggests that the C-C cross-link does not interfere with
the ligation reaction, even when the cross-link is located only
two base pairs from the site of the ligation, as is the case with
duplex 1827. It is interesting to note that, although cross-linked
duplex 1827 failed to exhibit any significant evidence of thermal
denaturation, its ability to support ligation appears to be
unhindered. While this behavior could be attributed to differ-
ences in buffer conditions between the two experiments, it seems
more likely that T4 DNA ligase, an enzyme capable of
catalyzing even blunt-ended ligations, stabilizes the interaction
between this short duplex and the acceptor.
Conclusions
As shown in Figure 5, lane 2, a control duplex, 1830, in which
the C-C cross-link of 1829 has been replaced with a G/C base
pair, is converted in the presence of T4 DNA ligase to two
slower migrating species. These species presumably correspond
to fully denatured 47-mer, the slowest moving band, and a
hairpin, which can form because the sequence of the 47-mer is
symmetrical. Under identical conditions, all three of the cross-
linked duplexes are capable of serving as substrates for the
ligation reaction, as shown by bands in lanes 4, 6, and 8. The
mobilities of these ligated duplexes, whose sizes differ by two
or four base pairs, are similar and, as expected, are less than
the mobilities of the single-stranded ligation products of the
non-cross-linked control duplex.
The results of our experiments show that the convertible
nucleoside approach can be used in conjunction with orthogonal
protecting groups to prepare interstrand cross-linked duplexes
directly on a DNA synthesizer. Although duplexes having an
ethyl cross-link between two opposing C residues were syn-
thesized here, the method should be applicable to syntheses of
other duplexes with opposed or staggered cross-links of varying
chain length. Such duplexes, which could be incorporated
enzymatically into longer pieces of DNA, should provide
valuable substrates for a variety of biophysical and biochemical
experiments.
In separate experiments (data not shown), we found that a
new product of intermediate gel mobility was formed when the
ligation reactions were carried out using a 1:1 molar ratio of
1829 to acceptor duplex. This product was “chased” into a
product of lower gel mobility when the ligation reaction was
continued after addition of excess acceptor duplex. This behavior
is consistent with first formation of half-ligated product consist-
Acknowledgment. This work was supported by a grant from
the National Cancer Institute (CA082785). D.M.N. was sup-
ported by grants from the Patterson Trust and the Alexander
and Margaret Stewart Trust. The authors thank Ms. Lori Mull
for assistance with the ligation experiments.
(34) Shi, Y.-b.; Hearst, J. E. Biochemistry 1986, 25, 5895-902.
JA003340T