Highly efficient sequence-specific DNA interstrand cross-linking by pyrrole/imidazole CPI conjugates

Article abstract of DOI:10.1021/ja028459b

We have developed a novel type of DNA interstrand cross-linking agent by synthesizing dimers of a pyrrole (Py)/imidazole (Im)-diamide-CPI conjugate, ImPyLDu86 (1), connected using seven different linkers. The tetramethylene linker compound, 7b, efficientl

Full text of DOI:10.1021/ja028459b

Published on Web 03/01/2003
Highly Efficient Sequence-Specific DNA Interstrand
Cross-Linking by Pyrrole/Imidazole CPI Conjugates
Toshikazu Bando,† Akihiko Narita,† Isao Saito,‡ and Hiroshi Sugiyama*,†
Contribution from the DiVision of Biofunctional Molecules, Institute of Biomaterials and
Bioengineering, Tokyo Medical and Dental UniVersity, 2-3-10 Surugadai, Kanda, Chiyoda,
Tokyo 101-0062, Japan, and Department of Synthetic Chemistry and Biological Chemistry,
Faculty of Engineering, Kyoto UniVersity, Yoshida, Sakyo, and Kyoto 606-8501, Japan, SORST,
Japan Science and Technology Corporation (JST)
Received September 8, 2002 ; E-mail: sugiyama.chem@tmd.ac.jp
Abstract: We have developed a novel type of DNA interstrand cross-linking agent by synthesizing dimers
of a pyrrole (Py)/imidazole (Im)-diamide-CPI conjugate, ImPyLDu86 (1), connected using seven different
linkers. The tetramethylene linker compound, 7b, efficiently produces DNA interstrand cross-links at the
nine-base-pair sequence, 5′-PyGGC(T/A)GCCPu-3′, only in the presence of a partner triamide, ImImPy.
For efficient cross-linking by 7b with ImImPy, one A‚T base pair between two recognition sites was required
to accommodate the linker region. Elimination of the A‚T base pair and insertion of an additional A‚T base
pair and substitution with a G‚C base pair significantly reduced the degree of cross-linking. The sequence
specificity of the interstrand cross-linking by 7b was also examined in the presence of various triamides.
The presence of ImImIm slightly reduced the formation of a cross-linked product compared to ImImPy.
The mismatch partners, ImPyPy and PyImPy, did not produce an interstrand cross-link product with 7b,
whereas ImPyPy and PyImPy induced efficient alkylation at their matching site with 7b. The interstrand
cross-linking abilities of 7b were further examined using denaturing polyacrylamide gel electrophoresis
with 5′-Texas Red-labeled 400- and 67-bp DNA fragments. The sequencing gel analysis of the 400-bp
DNA fragment with ImImPy demonstrated that 7b alkylates several sites on the top and bottom strands,
including one interstrand cross-linking match site, 5′-PyGGC(T/A)GCCPu-3′. To obtain direct evidence of
interstrand cross-linkages on longer DNA fragments, a simple method using biotin-labeled complementary
strands was developed, which produced a band corresponding to the interstrand cross-linked site on both
top and bottom strands. Densitometric analysis indicated that the contribution of the interstrand cross-link
in the observed alkylation bands was approximately 40%. This compound efficiently cross-linked both strands
at the target sequence. The present system consisted of a 1:2 complex of the alkylating agent and its
partner ImImPy and caused an interstrand cross-linking in a sequence-specific fashion according to the
base-pair recognition rule of Py-Im polyamides.
Introduction
a chlorambucil moiety at both the 5′ and the 3′ ends can alkylate
several sites on both DNA strands in a sequence-specific
DNA interstrand cross-linking agents effectively inhibit both
DNA replication and gene expression by preventing the melting
of the two strands, and therefore, they have considerable
potential in molecular biology and human medicine.1 Thus, the
mechanism of the action of interstrand cross-linking agents and
the design of new types of cross-linking agents has been
extensively investigated. Cross-linking agents, including anti-
tumor antibiotics, such as mitomycin C, carzinophilin A, and
synthetic antitumor agents, such as bizelesin and nitrogen
mustard derivatives, show intrinsic sequence-selectivity in the
formation of interstrand cross-links.2 However, interstrand cross-
linking agents that can target a predetermined sequence are quite
rare.3 By utilizing the sequence-specific recognition for triple
helix formation from the major groove of DNA, Kutyavin et
al. have demonstrated that a polypyrimidine 19-mer possessing
manner. However, the target sequence was limited to a poly-
purine sequence, and the efficiency of the interstrand cross-
linking was not high.4
Dickerson and colleagues have demonstrated using X-ray
crystallography that the natural product, netropsin, containing
two N-methylpyrrole (Py) groups forms a 1:1 complex in the
minor groove with an A‚T rich sequence.5 Dickerson and Lown
suggested that the conversion of Py to an imidazole (Im) in the
(2) (a) Norman, D.; Live, D.; Sastry, M.; Lipman, R.; Hingerty, B. E.; Tomasz,
M.; Broyde, S.; Patel, D. J. Biochemistry 1990, 29, 2861. (b) Fujiwara, T.;
Saito, I.; Sugiyama, H. Tetrahedron Lett. 1999, 40, 315. (c) Hodgkinson,
T. J.; Shipman, M. Tetrahedron 2001, 57, 4467. (d) Seaman, F. C.; Chu,
J.; Hurley, L. J. Am. Chem. Soc. 1996, 118, 5383. (e) Lawley, P. D.
BioEssays 1995, 17, 561. (f) Hopkins, P. B.; Millard, J. T.; Woo, J.;
Weidner, M. F.; Kirchner, J. J.; Sigurdsson, S. T.; Raucher, S. Tetrahedron
1991, 47, 2475.
(3) (a) Sigurdsson, S. T.; Rink, S. M.; Hopkins, P. B. J. Am. Chem. Soc. 1993,
115, 12633. (b) Zhou, Q.; Duan, W.; Simmons, D.; Shayo, Y.; Raymond,
M. A.; Dorr, R. T.; Hurley, L. H. J. Am. Chem. Soc. 2001, 123, 4865.
(4) Kutyavin, I. V.; Gamper, H. B.; Gall, A. A.; Meyer, R. B., Jr. J. Am. Chem.
Soc. 1993, 115, 9303.
† Tokyo Medical and Dental University.
‡ Kyoto University.
(1) Rajski, S. R.; Williams, R. M. Chem. ReV. 1998, 98, 2723.
9
10.1021/ja028459b CCC: $25.00 © 2003 American Chemical Society
J. AM. CHEM. SOC. 2003, 125, 3471-3485
3471

A R T I C L E S
Bando et al.
Scheme 1. Synthesis of 7ba
Figure 1. (a) The chemical structure of ImPyLDu86 (1) and a schematic
representation of the recognition of the 5′-PyGACPu-3′ sequence by the
homodimer of 1. The arrows indicate the site where alkylation takes place.
(b) A putative binding mode of the 1:2 complex of covalent dimer of 1,
7b, with ImImPy to 5′-PyGGCAGCCPu-3′ sequence.
A‚T binders should lead to G‚C specificity through the
introduction of a new hydrogen bond acceptor for the protruding
2-amino group of guanine.6 When these substitutions were made
in netropsin and distamycin A, rather complicated results were
obtained. Specific G‚C binding was achieved using the idea of
the 2:1 ligand binding motif to optimize contacts to the minor
groove.7 On the basis of these contributions to the concept of
minor groove sequence information readout, Dervan et al. have
designed and synthesized minor groove-binding Py-Im hairpin
polyamides that uniquely recognize each of the four Watson-
Crick base pairs.8 Sequence-specific DNA recognition in the
minor groove depends on the sequence of side-by-side pairings
oriented in the amino-carboxyl (N-C) direction with respect
to the 5′-3′ direction of the DNA helix. Antiparallel pairing of
Im opposite a Py group (Im/Py) recognizes a G‚C base pair,
whereas a Py/Py pair recognizes A‚T or T‚A base pairs.9
Hydroxypyrrole-Py pairing (Hp/Py) distinguishes the T‚A from
the A‚T base pairs.10 These Py-Im hairpin polyamides have a
i
a Conditions: i) Pd-C, NaBH4,MeOH then NO2ImCOCCl3, Pr2NEt,
CH2Cl2, 52%; ii) Pd-C, NaBH4, MeOH-AcOEt, then adipoyl chloride,
CH2Cl2, Pr2NEt, 36%; iii) DBU-H2O (1:1) then 1% aqueous HCl, 57%;
i
iv) CDI, DMF, 99%; v) segment A of DU-86, NaH, DMF, 61%.
binding affinity and sequence-specificity comparable to those
of transcription factors.11
Duocarmycin A (Duo) is a highly potent antitumor anti-
biotic,12 which binds to AT-rich sequences and selectively
alkylates the N3 site of adenine (A) at the 3′ end of three or
more consecutive A‚T base pairs in DNA.13 We discovered that
the addition of distamycin A (Dist) markedly modulates the
alkylation sites primarily at the G residues in GC-rich sequences
by forming a cooperative heterodimer between Duo and Dist.14
(10) (a) Kielkopf, C. L.; White, S.; Szewczyk, J. W.; Turner, J. M.; Baird, E.
E.; Dervan, P. B.; Rees, D. C. Science 1998, 282, 111. (b) White, S.;
Szewczyk, J. W.; Turner, J. M.; Baird, E. E.; Dervan, P. B. Nature 1998,
391, 468.
(11) (a) Trauger, J. W.; Baird, E. E.; Dervan, P. B. J. Am. Chem. Soc. 1998,
120, 3534. (b) Turner, J. M.; Swalley, S. E.; Baird, E. E.; Dervan, P. B. J.
Am. Chem. Soc. 1998, 120, 6219. (c) Bremer, R. E.; Baird, E. E.; Dervan,
P. B. Chem. Biol. 1998, 5, 119.
(5) Kopka, M. L.; Yoon, C.; Goodsell, D.; Pjura, P.; Dickerson, R. E. J. Mol.
Biol. 1985, 183, 553.
(6) (a) Kopka, M. L.; Yoon, C.; Goodsell, D.; Pjura, P.; Dickerson, R. E. Proc.
Natl. Acad. Sci. U.S.A., 1985, 82, 1376. (b) Lown, J. W.; Krowicki, K.;
Bhat, U. G.; Skorobogaty, A.; Ward, B.; Dabrowiak, J. C. Biochemistry
1986, 25, 7408.
(7) (a) Wade, W. S.; Mrksich, M.; Dervan, P. B. J. Am. Chem. Soc. 1992,
114, 8783. (b) Dwyer, T. J.; Geierstanger, B. H.; Bathini, Y.; Lown, J. W.;
Wemmer, D. E. J. Am. Chem. Soc. 1992, 114, 5911.
(8) (a) For recent review, see: Dervan, P. B. Bioorg. Med. Chem. 2001, 9,
2215. (b) Wemmer, D. E.; Dervan, P. B. Curr. Opin. Struct. Biol. 1997, 7,
355. (c) Trauger, J. W.; Baird, E. E.; Dervan, P. B. Nature 1996, 382, 559.
(d) Turner, J. M.; Baird, E. E.; Dervan, P. B. J. Am. Chem. Soc. 1997,
119, 7636. (e) Swalley, S. E.; Baird, E. E.; Dervan, P. B. J. Am. Chem.
Soc. 1997, 119, 6953. (f) Swalley, S. E.; Baird, E. E.; Dervan, P. B. Chem.
Eur. J. 1997, 3, 1600. (g) Trauger, J. W.; Baird, E. E.; Dervan, P. B. Angew.
Chem., Int. Ed. 1998, 37, 1421.
(9) (a) White, S.; Baird, E. E.; Dervan, P. B. J. Am. Chem. Soc. 1997, 119,
8756. (b) Dervan, P. B.; Burli, R. W. Curr. Opin. Chem. Biol. 1999, 3,
688. (c) Kielkopf, C. L.; Baird, E. E.; Dervan, P. B.; Rees, D. C. Nat.
Struct. Biol. 1998, 5, 104.
(12) (a) For recent review, see: Boger, D. L.; Boyce, C. W.; Garbaccio, R. M.;
Goldberg, J. Chem. ReV. 1997, 97, 787. (b) Takahashi, I.; Takahashi, K.;
Ichimura, M.; Morimoto, M.; Asano, K.; Kawamoto, I.; Tomita, F.; Nakano,
H. J. Antibiot. 1988, 41, 1915. (c) Ichimura, M.; Muroi, K.; Asano, K.;
Kawamoto, I.; Tomita, F.; Morimoto, M.; Nakano, H. J. Antibiot. 1988,
41, 1285. (d) Yasuzawa, T.; Iida, K.; Muroi, K.; Ichimura, M.; Takahashi,
K.; Sano, H.; Chem. Pharm. Bull. 1988, 36, 3728. (e) Asai, A.; Nagamura,
S.; Saito, H. J. Am. Chem. Soc. 1994, 116, 6, 4171. (f) Boger, D. L.; McKie,
J. A.; Nishi, T.; Ogiku, T. J. Am. Chem. Soc. 1997, 119, 311.
(13) (a) Sugiyama, H.; Hosoda, M.; Saito, I.; Asai, A.; Saito, H. Tetrahedron
Lett. 1990, 31, 7197. (b) Boger, D. L.; Ishizaki, T.; Zarrinmayeh, H. J.
Org. Chem. 1990, 55, 4499. (c) Boger, D. L.; Ishizaki, T.; Zarrinmayeh,
H.; Kitos, P. A.; Suntornwat, O. J. Am. Chem. Soc. 1990, 112, 2, 8961. (d)
Boger, D. L.; Ishizaki, T.; Zarrinmayeh, H. J. Am. Chem. Soc. 1991, 113,
3, 6645. (e) Boger, D. L. Acc. Chem. Res. 1995, 28, 8, 20. (f) Boger, D.
L.; Yun, W. Y.; Han, N. H. Bioorg. Med. Chem. 1995, 3, 1429. (g) Boger,
D. L.; Johnson, D. S. Angew. Chem., Int. Ed. Engl. 1996, 35, 1438. (h)
Schnell, J. R.; Ketchem, R. R.; Boger, D. L.; Chazin, W. J. J. Am. Chem.
Soc. 1999, 121, 5645.
9
3472 J. AM. CHEM. SOC. VOL. 125, NO. 12, 2003

DNA Interstrand Cross-Linking Agent Development
A R T I C L E S
Scheme 2 a
a Conditions: i) H2NCH2CH2CH2NMe2, 97%; ii) Pd-C, H2, MeOH-AcOEt then O2NImCOCCl3, iPr2NEt, CH2Cl2, 86%; iii) Pd-C, H2, MeOH-AcOEt
i
i
i
then O2NImCOCCl3, Pr2NEt, CH2Cl2, 44%; iv) Pd-C, NaBH4, MeOH then Ac2O, Pr2NEt, 47%; v) MeNH2, Pr2NEt, , 87%; vi) NaH, MeOH, 96%; vii)
i
NaOH, 79%; viii) DCC, HOBt; ix) Pd-C, H2, MeOH-AcOEt then Pr2NEt, DMF, 55%.
The NMR-refined structure of a Duo-Dist-d(CAGGTGGT)/
d(ACCACCTG) complex demonstrated that the three Py
residues of Dist recognize the complementary strand of the
reacting octamer, similar to the base-pair recognition rule of
Py-Im polyamides. In fact, the addition of Py-Im triamides to
the Duo reaction can effectively change the sequence-specificity
of Duo in a predictable manner.15 These results suggest that
Py-Im polyamides can be used as versatile recognition com-
ponents of sequence-specific DNA alkylating agents. Thus, we
synthesized hybrid molecules constructed from the A segment
of Duo and from Py-Im polyamides specifically alkylated at
predetermined sequences.16 The alkylating Py-Im hairpin poly-
amides selectively alkylated one of the matching sites within a
400-bp DNA fragment. Sequence-specific alkylation was ob-
served even at 50 nM concentrations of the alkylating agent.
Other groups also synthesized analogous CC-1065 and duocar-
mycin A related cyclopropapyrroloindole (CPI) or cyclopropa-
benzoindole (CBI) Py-Im conjugates and examined DNA
alkylating activities.17 Lown and colleagues have found that
insertion of a trans vinyl linker (L) between the Py and CPI
groups greatly enhanced the alkylating activity, as well as the
cytotoxicity.17a We also demonstrated that the Py-Im diamide-
CPI conjugates with the vinyl linker, ImPyLDu86 (1), and
alkylates double-stranded DNA at predetermined sequences
through the formation of a highly cooperative homodimer, as
shown in Figure 1a.18 We found that the efficiency of alkylation
by 1, which is the degree of DNA alkylation divided by that of
the agent, was 69%, thus confirming the unusually high
efficiency of dialkylation.
We extended our concept to synthesize the alkylating Py-Im
triamides, ImPyImLDu86 and ImImPyLDu86, which sequence-
specifically alkylate their target sequences in supercoiled plasmid
DNA.19 More recently, we have demonstrated that incorporation
of the vinyl linker pairing with Im dramatically improves the
reactivity of hairpin polyamide-CPI conjugates.20 These results
encouraged us to design sequence-specific interstrand cross-
linking agents,21 because these agents require a high efficiency
of alkylation on both sides, and the Py-Im polyamide CPI
(17) (a) Wang, Y.; Gupta, R.; Huang, L.; Luo, W.; Lown, J. W. Anticancer
Drug Des. 1996, 11, 15. (b) Jia, G.; Iida, H.; Lown, J. W. Heterocycl.
Commun. 1998, 4, 557. (c) Boger, D. L.; Han, N. Bioorg. Med. Chem.
1997, 233, 5. (d) Wurtz, N. R.; Dervan, P. B. Chem. Biol. 2000, 7, 154.
(e) Chang, A. Y.; Dervan, P. B. J. Am. Chem. Soc. 2000, 122, 4856.
(18) Tao, Z.-F.; Saito, I.; Sugiyama, H. J. Am. Chem. Soc. 2000, 122, 1602.
(19) Fujimoto, K.; Iida, H.; Kawakami, M.; Bando, T.; Tao, Z.-F.; Sugiyama,
H. Nucleic Acids Res. 2002, 30, 3748.
(14) (a) Yamamoto, K.; Sugiyama, H.; Kawanishi, S. Biochemistry 1993, 32,
1059. (b) Sugiyama, H.; Lian, C.; Isomura, M.; Saito, I.; Wang, A. H.-J.
Proc. Natl. Acad. Sci. U.S.A. 1996, 93, 14405.
(15) Fujiwara, T.; Tao, Z.-F.; Ozeki, Y.; Saito, I.; Wang, A. H.-J.; Lee, M.;
Sugiyama, H. J. Am. Chem. Soc. 1999, 121, 7706.
(16) (a) Tao, Z.-F.; Fujiwara, T.; Saito, I.; Sugiyama, H. Angew. Chem., Int.
Ed. 1999, 38, 650. (b) Tao, Z.-F.; Fujiwara, T.; Saito, I.; Sugiyama, H. J.
Am. Chem. Soc. 1999, 121, 4961.
(20) Bando, T.; Narita, A.; Saito, I.; Sugiyama, H. Chem. Eur. J. 2002, 8, 4781.
(21) Bando, T.; Iida, H.; Saito, I.; Sugiyama, H. J. Am. Chem. Soc. 2001, 123,
5158.
9
J. AM. CHEM. SOC. VOL. 125, NO. 12, 2003 3473

A R T I C L E S
Bando et al.
Figure 3. Denaturing polyacrylamide gel electrophoresis of TR18/ODN15
cross-linked by 7b with various concentrations of 7b with ImImPy: lane
1 ) 96 µM ImImPy and 48 µM 7b; lane 2 ) 48 µM ImImPy and 24 µM
7b; lane 3 ) 24 µM ImImPy and 12 µM 7b; lane 4 ) 12 µM ImImPy
and 6 µM 7b; lane 5 ) 6 µM ImImPy and 3 µM 7b; lanes 6-10 ) 24 µM
7b and 96, 48, 24, 12, and 6 µM ImImPy, respectively.
Im triamide, as shown in Figure 1b. To achieve efficient
interstrand cross-linking, we synthesized dimers of 1 possessing
various linkers (7a-g). Synthetic routes for compounds 7a-g
are summarized in Scheme 1. The dimeric unit 4a was
synthesized by the reduction of 3 with NaBH4, Pd-C, and H2
followed by coupling with glutaric dibenzotriazole ester, whereas
linkers 4b-g were prepared by coupling with the corresponding
linker dicarbonyl chlorides. The alkylating moiety, segment A
of DU-86 (CPI), was prepared via a five-step procedure
according to literature reports.23 After subsequent hydrolysis
with DBU, the carboxyl groups were activated by CDI to form
linkers 6a-g. The activated amides 6a-g were coupled with
CPI to give compounds 7a-g in a moderate yield. After
purification by high-performance liquid chromatography (HPLC),
the polyamide-CPI conjugates 7a-g were used in the DNA
alkylation experiments. Partner Py-Im triamides and tetraamides
with various N- and C-terminal groups 10-14 were synthesized
as shown in Scheme 2.
Reaction of 5′-[TR]-TTACAGTGGCTGCCAGCA-3′-
(TR18)/5′-TGCTGGCAGCCACTG-3′ (ODN15) with 7a-g
in the presence of ImImPy. The interstrand cross-linking
abilities of 7a-g were investigated using 5′-Texas Red-labeled-
TTACAGTGGCTGCCAGCA-3′ (TR18) and the complemen-
tary 5′-TGCTGGCAGCCACTG-3′ (ODN15). Alkylation was
carried out at 37 °C for 18 h, and the reaction was quenched
by the addition of calf thymus DNA. The reaction mixture was
Figure 2. (a) The denaturing polyacrylamide gel electrophoresis of 5′-
[TR]-TTACAGTGGCTGCCAGCA-3′ (TR18) and 5′-TGCTGGCAGC-
CACTG-3′ (ODN15) cross-linked by 7a-g in the presence and absence of
ImImPy: lane 1 ) DNA control; lane 2 ) 30 µM ImImPy; lane 3 ) 24
µM 7b; and lanes 4-10 ) 30 µM ImImPy and 24 µM with 7a-g,
respectively. (b) The denaturing polyacrylamide gel electrophoresis of TR18
cross-linked by 7b and ImImPy in the presence of different lengths of
complementary strand: lane 1 ) DNA control; lane 2 ) 7b (24 µM),
ImImPy (30 µM), and ODN15; lane 3 ) 7b (50 µM), ImImPy (100 µM),
and 5′-TTTTTATGCTGGCAGCCACTG-3′ (ODN21); lane 4 ) 7b (24
µM), ImImPy (30 µM), and ODN21. (c) Sequences of TR18, ODN15,
and ODN21. The alkylated bases and recognition sequences are represented
in red and blue, respectively.
conjugate with a vinyl linker fulfills this requirement. In this
study, we have examined in detail the sequence-specificity of
interstrand cross-linking by covalent dimers of 1 connected with
seven different linkers. The results indicate that sequence-
specificity of one of the dimers, 7b, can be controlled in a
predictive manner by changing the partner Py-Im triamide.
Using a new detection method, we have demonstrated that
sequence-specific interstrand cross-linking effectively occurs at
the target sequence within 67- and 400-bp DNA fragments.
Results and Discussion
Synthesis. To date, various types of CPI or CBI dimers,
including alternative points of attachment to CPI moiety, have
been synthesized.22 However, control of the sequence specificity
of the cross-linking agent is very difficult to achieve. We have
designed an interstrand cross-linking system by forming a 1:2
complex of an alkylating dimer component with a partner Py-
(22) (a) Kumar, R.; Lown, J. W. Org. Lett. 2002, 11, 1851. (b) Jia, G.; Iida, H.;
Lown, J. W. Heterocycl. Commun. 1998, 4, 557. (c) Reddy, B. S.; Sharma,
S. K.; Lown, J. W. Curr. Med. Chem. 2001, 8, 475. (d) Boger, D. L.;
Searcey, M.; Tse, W. C.; Jin, Q. Bioorg. Med. Chem. Lett. 2000, 10, 495.
(23) Nagamura, S.; Asai, A.; Kanda, Y.; Kobayashi, E.; Gomi, K.; Saito, H.
Chem. Pharm. Bull. 1996, 44, 1723.
9
3474 J. AM. CHEM. SOC. VOL. 125, NO. 12, 2003

DNA Interstrand Cross-Linking Agent Development
A R T I C L E S
Figure 4. Denaturing polyacrylamide gel electrophoresis of 5′-Texas Red-labeled oligonucleotides. Lanes 1-5 and lanes 6-10 used TR18/ODN15 and
5′-[TR]-TTATGCTGGCAGCCACTG-3′ (TR18L)/5′-CAGTGGCTGCCAGCA-3′ (ODN15L), respectively. Lanes 1 and 6 ) DNA controls; lanes 2 and 7
) reaction mixtures; lanes 3 and 8 ) reaction mixtures after exposure to heat and hot piperidine treatment; lanes 4 and 9 ) samples from lanes 3 and 8 after
exposure to alkaline phosphatase; lanes 5 and 10 ) 5′-[TR]-TTACAGTGGCTGCC-3′ and 5′-[TR]-TTATGCTGGCAGCC-3′, respectively.
concentrated at room temperature to avoid any depurination of
the alkylated bases, and the product was directly analyzed by
denaturing polyacrylamide gel electrophoresis using an auto-
mated DNA sequencer. As shown in Figure 2a (see lanes 4-6),
the compounds 7a-c generated slow-migrating bands in the
presence of ImImPy, while the other compounds, such as 7d,
e, and g, did not generate such slow migrating bands under the
same conditions (see lanes 7-10). When a longer complemen-
tary 5′-TTTTTATGCTGGCAGCCACTG-3′ (ODN21) was em-
ployed in the same reaction, a further shifted slower-migrating
band appeared (Figure 2b, lanes 3 and 4), indicating that these
slow-migrating bands are corresponding cross-linked products.
In the absence of ImImPy, compound 7b showed only a trace
of a slow-migrating band, indicating the indispensable role of
the partner in the formation of an interstrand cross-link (Figure
2a, lane 3). It is speculated that a trace of a slow-migrating
band may be due to alkylation products derived from the 1:1
monomeric binding of Py-Im polyamide.24 Interestingly, the
rigid linker compounds, 7d, e, and g did not produce any cross-
linked products, and only the flexible methylene linker com-
pounds, 7a-c caused cross-linking. The two faint bands
produced by 7f with ImImPy are assumed to be cross-linked
and monoalkylation products, respectively. Further identification
was not successful due to the low efficiency of the formations.
These results clearly indicate the importance of flexibility in
the linker region between two recognition moieties. These results
clearly demonstrate that the combination of the flexible meth-
ylene linker compounds, 7a-c and ImImPy efficiently produce
DNA interstrand cross-links.
To obtain the optimum conditions for efficient cross-linking,
the degree of interstrand cross-linking was monitored under
various concentrations of 7b and ImImPy, as shown in Figure
3. It was found that interstrand cross-linking largely depended
on the concentrations of 7b and ImImPy, and that 24 µM of
7b and 48 µM of ImImPy were required for the efficient cross-
linking of 3 µM of TR18/ODN15 (see lane 2). With 24 µM of
7b, more than 1 equiv of ImImPy was required for efficient
cross-linking (see lanes 6-8). Thus, the following alkylation
experiments were mostly performed under the standard condi-
tions of 24 µM of 7b and 30 µM of ImImPy.
Chemical Elucidation of Interstrand Cross-Linked Sites
in TR18/ODN15. To study the interstrand cross-linking bases
in TR18/ODN15, we analyzed the degradation products of
interstrand cross-links in detail, as shown in Figure 4. The
reaction mixture of TR18/ODN15 with 7b and ImImPy
containing a slow-migrating product (lane 2) was heated to 90
°C for 5 min and then subjected to a hot piperidine treatment
(90 °C, 20 min). Such a treatment without using heating at
neutral pH did not cause efficient strand cleavage at the adenine
alkylation sites. Using this treatment, the slow-migrating band
was converted to a faster migrating product (lane 3). The faster
migrating band was converted into a slightly slower migrating
band by dephosphorylation with alkaline phosphatase (lane 4),
the mobility of which was identical to that of the authentic
fragments of 5′-[TR]-TTACAGTGGCTGCC-3′ (lane 5). These
results clearly demonstrate that interstrand cross-linking selec-
tively occurred at the A15 position of TR18. Similar experiments
using labeled lower-strand 5′-[TR]-TTATGCTGGCAGCCACTG-
3′ (TR18L) and 5′-CAGTGGCTGCCAGCA-3′ (ODN15L)
(24) Urbach, A. R.; Dervan, P. B. Proc. Natl. Acad. Sci. U.S.A. 2001, 98, 4343.
9
J. AM. CHEM. SOC. VOL. 125, NO. 12, 2003 3475

A R T I C L E S
Bando et al.
Figure 5. Denaturing polyacrylamide gel electrophoresis of TR18 cross-
linked and monoalkylated by 7a with ImImPy in the presence of a different
complementary strand. (a) Lane 1 ) 7a (24 µM), ImImPy (24 µM), and
ODN15; lane 2 ) 7a (24 µM), ImImPy (24 µM), and 5′-TGCTGGCAGC-
CTCTG-3′ (ODN15T); lane 3 ) 7a (48 µM), ImImPy (48 µM), and
ODN15T; and lane 4 ) DNA control. (b) A schematic representation of
the monoalkylation of TR18/ODN15T by 7a with ImImPy.
Figure 6. Denaturing polyacrylamide gel electrophoresis of 5′-Texas Red-
labeled DNA fragments cross-linked by 7b and ImImPy. Lanes 1-4
employed TR18/ODN15, and lanes 5-8 employed 5′-[TR]-TTACAGCG-
GCTGCCGGCA-3′ (TR18G)/5′-TGCCGGCAGCCGCTG-3′ (ODN15G).
(a) Lanes 1 and 5 ) DNA control; lanes 2 and 6 ) 7b (24 µM) and ImImPy
(30 µM); lanes 3 and 7 ) 7b (12 µM) and ImImPy (30 µM); lanes 4 and
8 ) 7b (6 µM) and ImImPy (30 µM). (b) A schematic representation of
the interstrand cross-linking of TR18G/ODN15G by 7b with ImImPy.
produced an analogous slow-migrating band of the cross-link
product (lane 7). Heating and a hot piperidine treatment yielded
a faster migrating band (lane 8). This fragment was converted
by dephosphorylation to a slightly faster migrating band (lane
9), which comigrated with authentic 5′-[TR]-TTATGCTG-
GCAGCC-3′. These results clearly demonstrate that interstrand
cross-linking selectively occurred at the A15 position of TR18L.
The compounds with tri- and pentamethylene linker com-
pounds, 7a and 7c, gave fewer cross-linked products, indicating
that the tetramethylene linker is an optimal spacer for the
formation of interstrand cross-linking. The exact reason for the
efficient cross-linking by 7b is not known. However, flexibility
and an appropriate length of the linker would be important
factors for the cross-linking reaction. The band that appeared
slightly faster than the cross-linking product, observed in the
case of 7a with ImImPy, was assumed to be a monoalkylated
product. This is because this band was predominantly formed
when the same reaction using 7a was carried out with TR18/
5′-TGCTGGCAGCCTCTG-3′ (ODN15T) in which Target A,
in the lower strand, was substituted with Target T (Figure 5,
lanes 2 and 3).
reactions of 5′-[TR]-TTACAGTGGCTGCCGGCA-3′ (TR18G)
and 5′-TGCTGGCAGCCGCTG-3′ (ODN15G) with 7b in the
presence of 30 µM of ImImPy were examined. It was found
that three different concentrations of 7b cross-linked TR18G
and ODN15G in the presence of ImImPy with a similar
efficiency as that of TR18/ODN15 (Figure 6). These results
clearly indicate that the interstrand cross-link system of 7b with
ImImPy can alkylate either A or G at matching sites, i.e. a
5′-PyGGC(T/A)GCCPu-3′ sequence.
Base-Pair Preference of Linker Regions. As the γ-turn and
â-tail of hairpin polyamides25 and the linker region of tandem
hairpins recognize A‚T base pairs,26 we expected that one A‚T
base pair between two recognition moieties was required to
accommodate the linker region to enable efficient cross-linking.
In fact, deletion of the A‚T base pair of 5′-[TR]-TTACAGTG-
GCGCCAGCA-3′ (TR17) and 5′-TGCTGGCGCCACTG-3′
It has been demonstrated that alkylating Py-Im polyamide
can alkylate adenine and guanine,16,17a,18 whereas Duo mainly
alkylates adenine in AT-rich sequences.13 To test whether this
type of interstrand cross-link also occurs in G residues, the
(25) Swalley, S. E.; Baird, E. E.; Dervan, P. B. J. Am. Chem. Soc. 1999, 121,
1113.
(26) Herman, D. M.; Baird, E. E.; Dervan, P. B. Chem. Eur. J. 1999, 5, 975.
9
3476 J. AM. CHEM. SOC. VOL. 125, NO. 12, 2003

DNA Interstrand Cross-Linking Agent Development
A R T I C L E S
(ODN14), and the insertion of two A‚T base pairs of 5′-[TR]-
TTACAGTGGCTTGCCAGCA-3′ (TR19) and 5′-TGCTG-
GCAAGCCACTG-3′ (ODN16) significantly reduced the cross-
linking efficiency of 7a-c in the presence of ImImPy compared
with TR18/ODN15 (Figure 7a). Substitution of the A‚T linker
with a G‚C base pair, 5′-[TR]-TTACAGTGGCCGCCAGCA-
3′ (TR18CG) and 5′-TGCTGGCGGCCACTG-3′ (ODN15CG),
yielded several slow-migrating bands with some tailing present.
Some of these bands may be due to monoalkylation, which
includes newly produced half-match sites produced by the
incorporation of the C‚G base pair (Figure 7b). These results
clearly demonstrate that the linker region of the present cross-
linked system prefers single A‚T base pairs.
Optimal Partner Triamide for Interstrand Cross-Linkage
of TR18/ODN15 by 7b. We examined the interstrand cross-
linking by 7b in the presence of four different Py/Im triamides
at concentrations of 15 and 30 µM. The results of denaturing
polyacrylamide gel electrophoresis are shown in Figure 8. In
the case of ImImIm (lanes 3 and 7), the reduction of the
formation of a cross-linked product relative to ImImPy (lanes
2 and 6) was observed: 44-19% and 98-81% at 15 and 30
µM of triamide, respectively, indicating that the Im/Im pair binds
moderately to the G‚C base pairs to facilitate the interstrand
cross-link. These results indicate that the Py/Im pairing rule does
not strictly operate on the third residue of the triamides in the
present interstrand cross-linking system. A similar ambiguity
in base-pair recognition by Im/Im and Py/Py pairs of the
polyamides of the G‚C base pair has been previously reported.27
In clear contrast, the formation of the cross-linked product was
significantly reduced in the presence of ImPyPy or PyImPy
(lanes 4, 5, and 8). However, at 30 µM of PyImPy, a significant
DNA consumption was observed (lane 9), presumably due to
the monoalkylation at the half-matched site 5′-GCTG-3′. These
results indicate that the Py/Im pairing rule operates for the first
and second residues of the Py-Im triamides.
Optimal Terminal Group of Partner Triamide for Inter-
strand Cross-Link of TR18/ODN15 by 7b. We then examined
the interstrand cross-linking by 7b in the presence of a ImImPy
derivative with different terminal groups (10-14) at concentra-
tions of 15 and 30 µM. Substitution of the C-terminal diami-
nopropyl group with methylamine (11), methyl ester (12), or
carboxylic acid (13) decreased the efficacy of cross-linking
(Figure 9, lanes 1-3 and lanes 7-9). In addition, modification
to the N-terminal group of ImImPy (10 and 14) reduced the
activity (lanes 4, 5, 10, and 11). Densitometric analysis of the
gel electrophoresis indicated that the cross-linking efficiency
when compounds 11-14 were employed was less than 10% of
that of ImImPy at 15 µM. Although 30 µM of 11 yielded
several slow-migrating bands, the cross-linking efficiency by
compound 11 was apparently lower than that of ImImPy under
the same conditions. These results indicate that interstrand cross-
linking proceeds efficiently with ImImPy according to the base-
pair recognition rule of Py-Im polyamides.8 It is important to
note that 7b with the tetraamide, PyImImPy (14), did not yield
a cross-linked product (lanes 4 and 10). These results suggest
that steric hindrance between CPI and the N-terminal Py inhibits
an efficient cross-linking reaction.
Figure 7. (a) Denaturing polyacrylamide gel electrophoresis of 5′-Texas
Red-labeled DNA fragments treated by 7a-c in the presence of ImImPy.
Lanes 1-4 ) 5′-[TR]-TTACAGTGGCGCCAGCA-3′ (TR17)/5′-TGCTG-
GCGCCACTG-3′ (ODN14); lanes 5-8 ) TR18/ODN15; lanes 9-12 )
5′-[TR]-TTACAGTGGCTTGCCAGCA-3′ (ODN19)/5′-TGCTGGCAAGC-
CACTG-3′ (ODN16). Lane 1 ) DNA control; lanes 2-4 ) 30 µM ImImPy
and 24 µM 7a-c, respectively; lane 5 ) DNA control; lanes 6-8 ) 30
µM ImImPy and 24 µM 7a-c, respectively; lane 9 ) DNA control; lanes
10-12 ) 30 µM ImImPy and 24 µM 7a-c, respectively. (b) Denaturing
polyacrylamide gel electrophoresis of 5′-Texas Red-labeled DNA fragments
5′-[TR]-TTACAGTGGCCGCCAGCA-3′ (TR18CG)/5′-TGCTGGCGGC-
CACTG-3′ (ODN15CG) treated by 7b in the presence of ImImPy. Lane 1
) DNA control; lane 2 ) 15 µM ImImPy and 24 µM 7b; and lane 3 ) 30
µM ImImPy and 24 µM 7b. (c) Sequences of TR17, ODN14, TR19,
ODN16, TR18CG, and ODN15CG. The target bases and recognition
sequences are represented in red and blue, respectively. A schematic
representation of the monoalkylation of TR18CG/ODN15CG by 7b with
ImImPy.
(27) (a) Yang, X.-L.; Kaenzig, C.; Lee, M.; Wang, A. H.-J. Eur. J. Biochem.
1999, 263, 646. (b) Weyermann, P.; Dervan, P. B. J. Am. Chem. Soc. 2002,
124, 6872.
Control of Sequence Specificity by Changing the Py-Im
Triamide. We examined the sequence specificity of alkylation
9
J. AM. CHEM. SOC. VOL. 125, NO. 12, 2003 3477

A R T I C L E S
Bando et al.
Figure 9. Denaturing polyacrylamide gel electrophoresis of TR18/ODN15
cross-linked by 7b (24 µM) with the ImImPy derivatives, 10-14 (15 µM
or 30 µM). (a) Chemical structures of compounds 10-14. (b) Lane 1 ) 15
µM 11; lane 2 ) 15 µM 12; lane 3 ) 15 µM 13; lane 4 ) 15 µM 14; lane
5 ) 15 µM 10; lane 6 ) DNA control; lane 7 ) 30 µM 11; lane 8 ) 30
µM 12; lane 9 ) 30 µM 13; lane 10 ) 30 µM 14; and lane 11 ) 30 µM
10.
5′-TAGC(A/T)GCTA-3′ (sites 4 and 8), on both DNA strands
were observed in the presence of triamides (1 µM). However,
alkylations at 5′-TGACA-3′ (site 6) and 5′-TAGCTG-3′ (site
3) were also observed under these conditions. Alkylation at sites
3 and 6 can be explained by the binding of a 1:1 complex of
7b and triamide in which the alkylation part recognizes the base
sequence by a pair of Py/Im polyamides, the tetramethylene
linker region recognizes one or two AT base pairs, and the rest
of the molecule recognizes a base pair by the monomeric
recognition mode (Figure 10c).24 These results demonstrate that
sequence-specificity of DNA alkylation can be reasonably
controlled by changing the partner Py-Im triamide according
to the base recognition rule of Py-Im polyamides.
Sequence-Specific Interstrand Cross-Linking within a 400-
bp DNA Fragment. We evaluated the interstrand cross-linking
activity of 7b with ImImPy using a 5′-Texas Red-labeled 400-
bp DNA fragment possessing one matching site, 5′-TG-
GCAGCCG-3′. After DNA alkylation, the reaction mixture was
heated to 94 °C for 20 min, and thermally induced cleavage
fragments were analyzed using a sequencing gel. It was found
that a large excess of ImImPy (>25 µM) relative to 7b (100
nM) was necessary for efficient alkylation of the 400 bp DNA
fragment, suggesting the presence of a nonspecific monomeric
Figure 8. Denaturing polyacrylamide gel electrophoresis of TR18/ODN15
cross-linked by 7b (24 µM) with Py-Im triamide (15 µM or 30 µM). (a)
Chemical structures of Py-Im triamides. (b) Lane 1 ) DNA control; lane
2 ) 15 µM ImImPy; lane 3 ) 15 µM ImImIm; lane 4 ) 15 µM ImPyPy;
lane 5 ) 15 µM PyImPy; lane 6 ) 30 µM ImImPy; lane 7 ) 30 µM
ImImIm; lane 8 ) 30 µM ImPyPy; and lane 9 ) 30 µM PyImPy. (c) A
schematic representation of the interstrand cross-linking and monoalkylation
of TR18/ODN15 by 7b with triamide.
by changing the partner Py-Im triamides. We designed a 67-bp
DNA sequence including three matching sites for 7b with the
partner molecules ImPyPy, ImImPy, and PyImPy. If 7b and
each triamide were cooperatively bound to the minor groove
of DNA according to the base-pair recognition rule of Py-Im
polyamides, then sequence-specific interstrand cross-linking
should be observed for each matching site. We used two sets
of duplex 67-mers (TRF-67/R-67 and F-67/TRR-67), in which
the one of the strands was labeled using 5′-Texas Red, as shown
in Figure 10. In all three cases, sequence-selective alkylations
at their cross-linked matching sites, 5′-TGAC(A/T)GTCA-3′
(sites 1 and 5), 5′-TGGC(A/T)GCCA-3′ (sites 2 and 7), and
9
3478 J. AM. CHEM. SOC. VOL. 125, NO. 12, 2003

DNA Interstrand Cross-Linking Agent Development
A R T I C L E S
Figure 10. Sequence-specific alkylation of 5′-Texas Red-labeled: (a) TRF-67/R-67 and (b) F-67/TRR-67 DNA fragments by 7b with Py-Im triamide. Sites
of DNA alkylation were analyzed by thermal cleavage of the DNA strand at the alkylated sites using denaturing polyacrylamide gel electrophoresis. Lane
1 ) DNA control; lanes 2-5 ) 1 µM ImPyPy and 500, 250, 100, and 50 nM 7b, respectively; lanes 6-9 ) 1 µM ImImPy and 500, 250, 100, and 50 nM
7b, respectively; lanes 10-13 ) 1 µM PyImPy and 500, 250, 100, and 50 nM 7b, respectively. (c) A schematic representation of the putative interstrand
cross-linking and monoalkylation of a 67-bp DNA fragments by 7b with ImPyPy, ImImPy, and PyImPy. The arrows indicate the site of alkylation.
or dimeric binding of ImImPy (Figure 11a). Figures 11 (b) and
(c) clearly demonstrated that 7b with 30 µM of ImImPy
alkylates several sites (sites 1-11) on both DNA strands,
including the target sequence, 5′-PyGGC(A/T)GCCPu-3′ (sites
5 and 8). These results suggest that 7b with ImImPy alkylates
DNA in a 1:2 and a 1:1 binding mode, which is consistent with
the monoalkylations shown in Figure 10.
However, the sequencing analysis used employing top and
bottom strand individually labeled fragments could not distin-
guish between interstrand cross-linking and monoalkylation. To
determine the contribution of interstrand cross-links in the
observed alkylation at sites 5 and 8, a simple method using a
5′-biotin-labeled complementary strand was developed, as
outlined in Scheme 3. After treatment with 7b and ImImPy,
the DNA fragments were collected using streptavidin-bound
magnetic particles (MAGNOTEX-SA), and then the particles
were washed with alkaline solution. Under these conditions, only
cross-linked DNA fragments should be retained on the particles.
Ensuing heat and hot piperidine treatments should provide DNA
cleavage band only at the interstrand cross-linking site.
9
J. AM. CHEM. SOC. VOL. 125, NO. 12, 2003 3479

A R T I C L E S
Bando et al.
Figure 11. Sequence specific alkylation of 5′-Texas Red-labeled: (a) and (b) λDNA F 7315*-7714 (top), and (c) λDNA R 40789*-41188 (bottom) DNA
fragments with a 5′-biotin-labeled complementary strand by 7b and ImImPy. The sites of DNA alkylation were analyzed by denaturing polyacrylamide gel
electrophoresis. (a) Lane 1 ) DNA control; lanes 2-7 ) 0.5, 2, 5, 10, 25, and 30 µM ImImPy and 100 nM 7b; (b) and (c) lane 1 ) DNA control; lanes
2-6 ) 30 µM ImImPy and 25, 50, 100, 250, and 500 nM 7b. The asterisk indicates 5′-Texas Red modification. (d) A schematic representation of the
interstrand cross-linking and monoalkylation of a 400-bp DNA fragment by 7b with ImImPy. The arrows indicate the site of alkylation.
5′-Double-labeled DNA was alkylated with 7b under one-
hit conditions in the presence of ImImPy. After collection using
streptavidin-bound magnetic particles, the particles were washed
with alkaline solution. The particles were heated in a buffer
solution, and subjected to a hot piperidine treatment (90 °C, 20
min). The resulting DNA fragments were analyzed using a
sequencing gel. The results of the top and bottom strand of the
400-bp DNA are shown in Figure 12, a and b. The bands
corresponding to the two interstrand cross-link sites were
significantly retained relative to other monomeric sites. The
almost complete disappearance of the starting DNA also
confirmed the reliability of the present method. Densitometric
analysis demonstrated that the contribution of the interstrand
cross-links of the top and bottom strand were approximately
9
3480 J. AM. CHEM. SOC. VOL. 125, NO. 12, 2003

DNA Interstrand Cross-Linking Agent Development
A R T I C L E S
Figure 12. Interstrand cross-link of 5′-Texas Red-labeled: (a) λDNA F 7315*-7714 (top), (b) λDNA R 40789*-41188 (bottom), and (c) TRF-67 DNA
fragments with a 5′-biotin-labeled complementary strand by 7b with ImImPy. Sites of DNA alkylation were analyzed after heating and a hot piperidine
treatment by denaturing polyacrylamide gel electrophoresis both before, and after subjecting the solution to the detection method described in Scheme 3.
Lane 1 ) alkylation with 7b (a and b ) 100 nM, and c ) 500 nM) and ImImPy (a and b ) 25 µM, and c ) 1 µM); lane 2 ) alkylated DNA of lane 1
captured by streptavidin-bound magnetic particles, and then washed with alkaline solution. The asterisk indicates 5′-Texas Red modification. Schematic
representations of the interstrand cross-linking and monoalkylation by 7b with ImImPy are also included. The arrows indicate the site of alkylation.
Scheme 3
48% and 34%, respectively. The results of the 64-bp DNA
fragments are shown in Figure 12c, showing that the bands at
the target cross-linking sites were selectively retained. The
results demonstrated that 38% of the alkylation of the target
sequence of the top strand was due to interstrand cross-linking.
These results confirm that 7b with ImImPy efficiently cross-
links both strands at the target site.
These results suggest that the present system of a 1:2 complex
of 7b and ImImPy causes interstrand cross-linking in a
sequence-specific fashion according to the base-pair recognition
rule of Py-Im polyamides (Figure 13). To the best of our
knowledge, this is the first demonstration of such interstrand
cross-linking of a longer DNA fragment at a predetermined
sequence.
In conclusion, we have developed a novel DNA interstrand
cross-linking agent 7b that cross-links double strands only in
the presence of ImImPy in the nine-base-pair sequence, 5′-
PyGGC(T/A)GCCPu-3′. The present system will provide a
promising approach for the design of novel sequence-specific
DNA interstrand cross-linking agents. Targeting specific se-
9
J. AM. CHEM. SOC. VOL. 125, NO. 12, 2003 3481

A R T I C L E S
Bando et al.
Figure 13. (a) Energy-minimized structure of 7b-ImImPy2-d(CTGGCTGCCAC)/ d(GTGGCAGCCAG) interstrand cross-linked complex. DNA is drawn
in white. Py and Im residues of 7b and ImImPy are drawn in blue and red, respectively. (b) A DNA interstrand cross-linking mode of the 1:2 complex of
7b with ImImPy.
Proton NMR spectra were recorded in parts per million (ppm)
downfield, relative to tetramethylsilane. The following abbreviations
apply to spin multiplicity: s (singlet), d (doublet), t (triplet), q (quartet),
qu (quintet), m (multiplet), and br (broad). Electrospray ionization mass
spectra (ESMS) were produced on a PE SCIEX API 165 mass
spectrometer and fast atom bombardment mass spectra (FABMS) on a
JEOL-JMS-SX102A mass spectrometer. Polyacrylamide gel electro-
phoresis was performed on a HITACHI 5500-S DNA sequencer.
Fluorescent bands were analyzed using FRAGLYS2 (version 2.2,
HITACHI) and NIH image (version 1.61) on a Power Macintosh G3.
All deoxyoligonucleotides including 5′-Texas Red-labeled oligomers
and 5′-biotin-labeled oligomers were purchased from JbioS. Ex Taq
DNA polymerase and Suprec-02 purification cartridges were purchased
from Takara; the Thermo Sequenase core sequencing kit and loading
dye (dimethylformamide with fuchsin red) from Amersham; and 50%
Long Ranger gel solution from FMC Bioproducts. P1 nuclease and
calf intestine alkaline phosphatase (1000 units/mL) were purchased from
Roche Diagnostics. Magnetex-SA was purchased from JSR.
quences in the human genome by such a sequence-specific cross-
linking agent would constitute a powerful gene-regulating tool.29
Further studies on the applicability of this novel class of cross-
linking agents are currently in progress.
Experimental Section
Reagents and solvents were purchased from standard suppliers and
used without further purification. Abbreviations of some reagents:
i
DBU, 1,8-diazabicyclo[4.3.0]undec-7-ene; Pr2NEt, N,N-diisopropyl-
ethylamine; DMF, N,N-dimethylformamide; FDPP, pentafluorophenyl
diphenylphosphinate; CDI, 1,1′-carbonyldiimidazole. Reactions were
monitored by thin-layer chromatography (TLC) using 0.25-mm silica
gel 60 plates impregnated with a 254-nm fluorescent indicator (Merck).
Plates were visualized by UV illumination at 254 nm. NMR spectra
were recorded with a JEOL JNM-A 500 nuclear magnetic resonance
spectrometer, and tetramethylsilane was used as the internal standard.
(28) Kielkopf, C. L.; White, S.; Wzewczyk, J. W.; Turner, J. M.; Baird, E. E.;
Dervan, P. B.; Rees, D. C. Science 1998, 282, 111.
O2NPyLCO2Et (2). To a solution of HNO3 (1.5 mL, 35.7 mmol) in
Ac2O (3 mL) was slowly added 1-methyl-2-pyrrole carboxaldehyde
(3.0 g, 27.5 mmol) in Ac2O (5 mL) at -40 °C for 40 min. After the
reaction mixture was stirred for 2 h at -10 °C, hexane (100 mL) was
added. The precipitate was collected by filtration and washed with
hexane (10 mL × 2) to afford yellow crude amine (1.43 g, 34%), which
was used in the next step without further purification. To a solution of
(29) (a) Gottesfeld, J. M.; Neely, L.; Trauger, J. W.; Baird, E. E.; Dervan, P. B.
Nature 1997, 387, 202. (b) Henikoff, S.; Vermaak, D. Cell 2000, 103, 695.
(c) Janssen, S.; Cuvier, O.; Mu¨ller, M.; Laemmli, U. K. Mol. Cell. 2000,
6, 1013. (d) Ansari, A. Z.; Mapp, A. K.; Nguyen, D. H.; Dervan, P. B.;
Ptashne, M. Chem. Biol. 2001, 8, 583. (e) Ehley, J. A.; Melander, C.;
Herman, D.; Baird, E. E.; Ferguson, H. A.; Goodrich, J. A.; Dervan, P. B.;
Gottesfeld, J. M. Mol. Cell. Biol. 2002, 22, 1723. (f) Gottesfeld, J. M.;
Belitsky, J. M.; Melander, C.; Dervan, P. B.; Luger, K. J. Mol. Biol. 2002,
321, 249.
9
3482 J. AM. CHEM. SOC. VOL. 125, NO. 12, 2003

DNA Interstrand Cross-Linking Agent Development
A R T I C L E S
sodium hydride (372 mg, 9.29 mmol, 60% oil suspension) in THF (5
mL) was added dropwise triethylphosphonoacetate (1.93 mL, 9.75
mmol) at 0 °C. After 1-methyl-4-nitro-2-pyrrole carboxaldehyde (1.43
g, 9.29 mmol) was added in THF (10 mL) at 0 °C, the reaction mixture
was stirred for 2 h at room temperature. Following the evaporation of
the solvent H2O (10 mL) was added. The aqueous phase was extracted
with EtOAc (100 mL × 2). The combined organic layers were dried
(Na2SO4) and concentrated, and the residue was purified by column
chromatography (silica gel, 25-50% EtOAc in hexane, gradient elution)
tetra-CH2(CONHImPyLCOIm)2 (6b). To a solution of 5b (30.0
mg, 43.6 µmol) in DMF (1 mL) was added CDI (42 mg, 261 µmol).
The mixture was stirred overnight at room temperature. Evaporation
of the solvent gave a yellow residue, which was triturated by ethyl
ether (10 mL) to produce 6b (35.7 mg, 99%). 1H NMR (DMSO-d6) δ
1.59 (s, 4H), 2.34 (s, 4H), 3.78 (s, 6H), 3.96 (s, 6H), 7.11 (s, 2H), 7.16
(d, J ) 14.5 Hz, 2H), 7.32 (s, 2H), 7.45 (s, 2H), 7.49 (s, 2H), 7.87 (d,
J ) 14.5 Hz, 2H), 7.90 (s, 2H), 8.68 (s, 2H), 10.07 (br s, 2H), 10.23
(br s, 2H). ESMS m/e calcd for C38H41N14O6 [M+ + H] 789.1, found
789.3.
1
to afford linker 2 (1.69 g, 81%). H NMR (CDCl3) δ 1.34 (t, J ) 7.0
Hz, 3H), 3.77 (s, 3H), 4.26 (q, J ) 7.0 Hz, 2H), 6.31 (d, J ) 16.0 Hz,
1H), 7.11 (d, J ) 1.5 Hz, 1H), 7.48 (d, J ) 16.0 Hz, 1H), 7.56 (d, J
) 1.5 Hz, 1H). 13C NMR (CDCl3) δ 14.3, 35.4, 60.8, 106.1, 118.4,
125.3, 129.8, 130.1, 136.7, 166.5. FAB MS 225 [M+ + H].
tetra-CH2(CONHImPyLCODu86)2 (7b). To a solution of sodium
hydride (2.6 mg, 64.4 µmol, 60% oil suspension) in DMF (0.2 mL)
was added segment A of DU-8623 (12.5 mg, 48.3 µmol) in DMF (0.2
mL). After 6b (13.6 mg, 16.5 µmol) in DMF (0.2 mL) was added at 0
°C, the reaction mixture was stirred for 4 h at 0 °C. The reaction mixture
was quenched by addition of 0.01 M of sodium phosphate buffer (2
mL) at 0 °C. Evaporation of the solvent gave a yellow residue, which
was washed sequentially with H2O (10 mL), MeOH (10 mL), and ethyl
ether (10 mL) to produce 7b (12.2 mg, 61%). The purity of 7b was
more than 85% as determined by HPLC analysis. (Wakopak 5C18,
0.05 M ammonium formate containing 0-100% acetonitrile, linear
gradient, 20 min, at a flow rate of 1.0 mL/min; retention time of 7b,
O2NImPyLCO2Et (3). To a solution of 2 (500 mg, 2.23 mmol) in
MeOH (25 mL) was added 10% Pd-C (220 mg). After NaBH4 (153
mg, 4.04 mmol) in H2O (3 mL) was added dropwise at 0 °C, the reaction
mixture was stirred for 20 min at room temperature. The reaction
mixture was filtered through Celite and EtOAc (500 mL) was added.
The organic phase was washed with H2O (10 mL), dried (Na2SO4),
and concentrated in vacuo to afford 461 mg of brown crude amine,
which was used in the next step without further purification. To a
1
i
14.4 min). H NMR (DMSO-d6) δ 1.29 (m, 2H), 1.58 (s, 4H), 2.09
solution of crude amine in CH2Cl2 (10 mL) was added Pr2NEt (0.52
(m, 2H), 2.33 (s, 4H), 2.47 (s, 6H), 3.45 (m, 2H), 3.72 (s, 6H), 3.73 (s,
6H), 3.95 (s, 6H), 4.18 (m, 2H), 4.28 (m, 2H), 6.57 (d, J ) 14.5 Hz,
2H), 6.83 (br s, 2H), 6.99 (s, 2H), 7.41 (s, 2H), 7.44 (s, 2H), 7.58 (d,
J ) 14.5 Hz, 2H), 9.98 (s, 2H), 10.23 (s, 2H), 12.36 (br s, 2H). ESMS
m/e calcd for C60H61N14O12 [M+ + H] 1169.5, found 1169.6. A synthetic
procedure similar to that described for compound 7b was followed in
the preparation of 7a and 7b-i.
mL, 2.98 mmol). After O2NImCOCCl3 (550 mg, 2.02 mmol) was added
at 0 °C, the reaction mixture was stirred for 1 h at room temperature.
Evaporation of the solvent gave a brown residue, which was purified
by column chromatography (silica gel, 30-50% EtOAc in hexane,
1
gradient elution) to afford linker 3 (400 mg, 52% for two steps). H
NMR (CDCl3) δ 1.33 (t, J ) 7.0 Hz, 3H), 3.71 (s, 3H), 4.21 (s, 3H),
4.25 (q, J ) 7.0 Hz, 2H), 6.16 (d, J ) 16.0 Hz, 1H), 6.62 (d, J ) 1.5
Hz, 1H), 7.32 (d, J ) 1.5 Hz, 1H), 7.55 (d, J ) 16.0 Hz, 1H), 7.82 (s,
1H), 8.97 (br s, 1H). 13C NMR (CDCl3) δ 14.3, 34.4, 37.1, 60.3, 102.5,
114.2, 117.9, 122.1, 124.4, 127.6, 131.4, 137.2, 145.3, 154.4, 167.4.
FAB MS 348 [M+ + H].
tri-CH2(CONHImPyLCODu86)2 (7a): 1H NMR (DMSO-d6) δ 1.28
(m, 2H), 1.86 (m, 2H), 2.07 (m, 2H), 2.35 (s, 4H), 2.45 (s, 6H), 3.44
(m, 2H), 3.71 (s, 6H), 3.72 (s, 6H), 3.94 (s, 6H), 4.18 (m, 2H), 4.27
(m, 2H), 6.57 (d, J ) 15.0 Hz, 2H), 6.81 (br s, 2H), 6.97 (s, 2H), 7.40
(s, 2H), 7.44 (s, 2H), 7.56 (d, J ) 15.0 Hz, 2H), 9.94 (s, 2H), 10.21 (s,
2H), 12.34 (br s, 2H). ESI MS m/e calcd for C59H59N14O12 [M+ + H]
1155.4, found 1155.7.
tetra-CH2(CONHImPyLCO2Et)2 (4b). To a solution of 3 (250 mg,
0.72 mmol) in MeOH-EtOAc (1:1, 10 mL) was added 10% Pd-C
(120 mg). After NaBH4 (54.5 mg, 1.44 mmol) in H2O (0.5 mL) was
added dropwise at 0 °C, the reaction mixture was stirred for 20 min at
room temperature. The reaction mixture then subjected to flash column
chromatography (silica gel, EtOAc). The solvent was then removed
by evaporation under reduced pressure to give brown crude amine (141
mg), which was used in the next step without further purification. To
penta-CH2(CONHImPyLCODu86)2 (7c): 1H NMR (DMSO-d6) δ
1.27 (m, 4H), 1.56 (m, 4H), 2.07 (m, 2H), 2.30 (m, 4H), 2.46 (s, 6H),
3.55 (m, 2H), 3.70 (s, 6H), 3.71 (s, 6H), 3.93 (s, 6H), 4.18 (m, 2H),
4.27 (m, 2H), 6.56 (d, J ) 15.0 Hz, 2H), 6.81 (br s, 2H), 6.96 (s, 2H),
7.39 (s, 2H), 7.42 (s, 2H), 7.56 (d, J ) 15.0 Hz, 2H), 9.99 (s, 2H),
10.21 (s, 2H), 12.35 (br s, 2H). ESI MS m/e calcd for C61H63N14O12
[M+ + H] 1183.5, found 1183.6.
i
a solution of this crude amine in CH2Cl2 (2 mL) was added Pr2NEt
(0.25 mL, 1.33 mmol). After adipoyl chloride (32 µL, 0.22 mmol) was
added dropwise to the solution at 0 °C, the reaction mixture was stirred
for 14 h at room temperature. Evaporation of the solvent gave a yellow
residue, which was purified by column chromatography (silica gel,
0-3% MeOH in CHCl3, gradient elution) to afford 4b (96.2 mg, 36%
1
CO (NHImPyLCODu86)2 (7d): H NMR (DMSO-d6) δ 1.29 (m,
2H), 2.09 (m, 2H), 2.47 (s, 6H), 3.44 (m, 2H), 3.72 (s, 6H), 3.73 (s,
6H), 3.97 (s, 6H), 4.18 (m, 2H), 4.27 (m, 2H), 6.56 (d, J ) 14.5 Hz,
2H), 6.84 (br s, 2H), 7.01 (s, 2H), 7.26 (s, 2H), 7.42 (s, 2H), 7.57 (d,
J ) 14.5 Hz, 2H), 8.82 (br s, 2H), 10.16 (s, 2H), 12.35 (br s, 2H).
ESMS m/e calcd for C56H53N14O11 [M+ + H] 1084.4, found 1085.5.
trans-CHCH(CONHImPyLCODu86)2 (7e): 1H NMR (DMSO-d6)
δ 1.30 (m, 2H), 2.08 (m, 2H), 2.47 (s, 6H), 3.47 (m, 2H), 3.73 (s, 6H),
3.74 (s, 6H), 3.99 (s, 6H), 4.20 (m, 2H), 4.29 (m, 2H), 6.59 (d, J )
15.0 Hz, 2H), 6.79 (br s, 2H), 7.01 (s, 2H), 7.28 (s, 2H), 7.43 (s, 2H),
7.58 (d, J ) 15.0 Hz, 2H), 7.61 (s, 2H), 10.07 (br s, 2H), 10.89 (s,
2H), 12.36 (br s, 2H). ESMS m/e calcd for C58H55N14O12 [M+ + H]
1139.4, found 1139.5.
1
yield for two steps). H NMR (CDCl3) δ 1.32 (t, J ) 6.5 Hz, 6H),
1.81 (s, 4H), 2.42 (s, 4H), 3.67 (s, 6H), 4.04 (s, 6H), 4.24 (q, J ) 6.5
Hz, 4H), 6.10 (d, J ) 15.5 Hz, 2H), 6.51 (s, 2H), 7.34 (s, 2H), 7.41 (s,
2H), 7.53 (d, J ) 15.5 Hz, 2H), 8.04 (br s, 2H), 8.80 (br s, 2H). 13C
NMR (5% CD3OD in CDCl3) δ 14.1, 24.8, 34.1, 34.5, 34.6, 60.3, 102.3,
112.9, 114.3, 118.0, 122.8, 127.1, 131.8, 133.7, 135.8, 155.8, 168.0,
171.0. ESMS m/e calcd for C36H45N10O8 [M+ + H] 745.3, found 745.6.
tetra-CH2(CONHImPyLCO2H)2 (5b). To a solution of 4b (76.2
mg, 0.10 mmol) in H2O (0.6 mL) was added DBU (0.6 mL, 4.01 mmol).
The mixture was stirred for 15 h at room temperature. The solvent
was removed by evaporation under reduced pressure to give a brown
residue, which was triturated by Et2O (10 mL). The resulting DBU
salt was acidified with 1% aqueous HCl. The precipitate was collected
m-Ph(CONHImPyLCODu86)2 (7f): 1H NMR (DMSO-d6) δ 1.26
(m, 2H), 2.07 (m, 2H), 2.45 (s, 6H), 3.40 (m, 2H), 3.71 (s, 6H), 3.72
(s, 6H), 4.02 (s, 6H), 4.13 (m, 2H), 4.25 (m, 2H), 6.56 (d, J ) 15.0
Hz, 2H), 6.80 (br s, 2H), 6.98 (s, 2H), 7.41 (s, 2H), 7.56 (d, J ) 15.0
Hz, 2H), 7.65 (t, J ) 8.0 Hz, 1H), 7.67 (s, 2H), 8.16 (d, J ) 8.0 Hz,
2H), 8.59 (s, 1H), 10.01 (s, 2H), 10.82 (s, 2H), 12.33 (br s, 2H). ESMS
m/e calcd for C62H67N14O12 [M+ + H] 1189.4, found 1189.5.
1
by filtration and dried in vacuo to produce 5b (40.0 mg, 57%). H
NMR (DMSO-d6) δ 1.58 (s, 4H), 2.32 (s, 4H), 3.68 (s, 6H), 3.94 (s,
6H), 6.03 (d, J ) 15.5 Hz, 2H), 6.80 (s, 2H), 7.41 (s, 2H), 7.43 (s,
2H), 7.46 (d, J ) 15.5 Hz, 2H), 9.89 (br s, 2H), 10.24 (br s, 2H). ESI
MS m/e calcd for C32H37N10O8 [M+ + H] 689.3, found 689.5.
1
p-Ph(CONHImPyLCODu86)2 (7g): H NMR (DMSO-d6) δ 1.30
(m, 2H), 2.09 (m, 2H), 2.47 (s, 6H), 3.46 (m, 2H), 3.74 (s, 12H), 4.02
9
J. AM. CHEM. SOC. VOL. 125, NO. 12, 2003 3483

A R T I C L E S
Bando et al.
(s, 6H), 4.20 (m, 2H), 4.29 (m, 2H), 6.60 (d, J ) 15.0 Hz, 2H), 6.83
(br s, 2H), 7.01 (s, 2H), 7.43 (s, 2H), 7.58 (d, J ) 15.0 Hz, 2H), 7.68
(s, 2H), 8.12 (s, 4H), 10.05 (s, 2H), 10.96 (s, 2H), 12.36 (br s, 2H).
ESI MS m/e calcd for C62H67N14O12 [M+ + H] 1189.4, found 1189.5.
1H), 9.64 (s, 1H), 10.38 (s, 1H). ESMS m/e calcd for C22H32N11O4
[M+ + H] 514.3, found 514.4.
ImPyPy: 1H NMR (DMSO-d6) δ 1.60 (qu, J ) 7.0 Hz, 2H), 2.01
(s, 3H), 2.12 (s, 6H), 2.23 (t, J ) 7.0 Hz, 2H), 3.17 (dt, J ) 5.5, 7.0
Hz, 2H), 3.79 (s, 3H), 3.84 (s, 3H), 3.94 (s, 3H), 6.82 (d, J ) 1.5 Hz,
1H), 7.11 (d, J ) 1.5 Hz, 1H), 7.17 (d, J ) 1.5 Hz, 1H), 7.26 (d, J )
1.5 Hz, 1H), 7.41 (s, 1H), 8.04 (br t, J ) 5.5 Hz, 1H), 9.87 (s, 1H),
NO2PyCONH(CH2)3NMe2 (8). To N,N-dimethyl-1,3-propane-
diamine (1 mL, 8.33 mmol) was added O2NPyCOCCl3 (500 mg, 1.84
mmol) at 0 °C. The reaction mixture was stirred for 12 h at room
temperature. After the evaporation of the solvent, the residue was
9.94 (s, 1H), 10.22 (s, 1H). ESMS m/e calcd for C24H34N9O4 [M+
+
1
H] 512.3, found 512.2.
recrystallized from ethyl ether (3 mL) to afford 8 (460 mg, 97%) H
PyImPy: 1H NMR (DMSO-d6) δ 1.85 (m, 2H), 1.96 (s, 3H), 2.46
(m, 2H), 2.76 (s, 3H), 2.77 (s, 3H), 3.24 (m, 2H), 3.81 (s, 3H), 3.82 (s,
3H), 3.96 (s, 3H), 6.92 (s, 1H), 7.01 (s, 1H), 7.22 (s, 2H), 7.52 (s, 1H),
8.18 (br s, 1H), 9.81 (s, 1H), 9.93 (s, 1H), 10.18 (s, 1H). ESMS m/e
calcd for C24H34N9O4 [M+ + H] 512.3, found 512.3.
NMR (CDCl3) δ 1.75 (t, J ) 6.0 Hz, 2H), 2.34 (s, 6H), 2.54 (t, J )
6.0 Hz, 2H), 3.49 (q, J ) 6.0 Hz, 2H), 4.00 (s, 3H), 6.94 (s, 1H), 7.52
(s, 1H), 8.68 (br s, 1H). FAB MS 255 [M+ + H].
NO2ImPyCONH(CH2)3NMe2 (9). To a solution of 8 (460 mg, 1.81
mmol) in MeOH (3 mL) was added 10% Pd-C (120 mg), and the
reaction mixture was stirred for 2 h at room temperature under H2 gas.
The reaction mixture was filtered through Celite with MeOH (50 mL),
and concentrated in vacuo to afford 413 mg of yellow amine crude,
which was used in the next step without further purification. To a crude
AcNHImImPyCONHMe (11). To O2NPyCOCCl3 (1 g, 3.68 mmol)
in THF (10 mL) was added methylamine HCl salt (0.3 g, 4.44 mmol)
i
and Pr2NEt (2 mL, 11.48 mmol) at 0 °C. The reaction mixture was
stirred for 12 h at room temperature. After evaporation of the solvent,
the residue was recrystallized from ethyl ether (3 mL) to afford NO2-
i
solution in CH2Cl2 (8 mL) was added Pr2NEt (0.5 mL, 2.87 mmol).
1
PyCONHMe (586 mg, 87%, H NMR (DMSO-d6) δ 2.71 (d, J ) 5.0
After O2NImCOCCl3 (493 mg, 1.81 mmol) was added at 0 °C, the
reaction mixture was stirred for 15 h at room temperature. Evaporation
of the solvent gave a yellow residue, which was purified by successively
with water (30 mL) and diethyl ether (5 mL) and then dried in vacuo
Hz, 3H), 3.89 (s, 3H), 7.34 (d, J ) 2.0 Hz, 1H), 8.09 (d, J ) 2.0 Hz,
1H), 8.33 (br s, 1H)). A synthetic procedure similar to that described
for ImImPy was followed in the preparation of 11, with a yield of 9%
1
after six steps. H NMR (DMSO-d6) δ 2.03 (s, 3H), 2.68 (d, J ) 4.5
1
to produce 9 (587 mg, 86% for two steps). H NMR (DMSO-d6) δ
Hz, 3H), 3.80 (s, 3H), 3.97 (s, 3H), 3.99 (s, 3H), 6.90 (s, 1H), 7.22 (s,
1H), 7.50 (s, 1H), 7.55 (s, 1H), 7.93 (br s, 1H), 9.32 (s, 1H), 10.24 (s,
1H), 10.28 (s, 1H). ESMS m/e calcd for C19H24N9O4 [M+ + H] 442.2,
found 442.1.
1.61 (t, J ) 7.0 Hz, 2H), 2.14 (s, 6H), 2.24 (t, J ) 7.0 Hz, 2H), 3.19
(q, J ) 7.0 Hz, 2H), 3.81 (s, 3H), 4.04 (s, 3H), 6.97 (s, 1H), 7.27 (s,
1H), 8.14 (br s, 1H), 8.61 (s, 1H), 10.80 (br s, 1H). FAB MS 378 [M+
+ H].
AcNHImImPyCO2Me (12). To O2NPyCOCCl3 (1 g, 3.68 mmol)
in MeOH (20 mL) was added NaH (9 mg, 0.23 mmol) at 0 °C. The
reaction mixture was stirred for 1 h at room temperature. After adding
p-TsOH (45 mg, 0.24 mmol), the solution was concentrated. The
resulting white precipitate was collected by filtration, washed with
NO2ImImPyCONH(CH2)3NMe2 (10). To a solution of 9 (100 mg,
0.27 mmol) in MeOH (3 mL) was added 10% Pd-C (50 mg), and the
reaction mixture was stirred for 4 h at room temperature under H2 gas.
The reaction mixture was filtered through Celite with MeOH (50 mL),
and concentrated in vacuo to afford 65.3 mg of yellow amine crude,
which was used in the next step without further purification. To a crude
1
water, and dried to produce NO2PyCO2Me (651 mg, 96%, H NMR
(DMSO-d6) δ 3.79 (s, 3H), 3.92 (s, 3H), 7.31 (d, J ) 2.0 Hz, 1H),
8.27 (d, J ) 2.0 Hz, 1H)). A synthetic procedure similar to that
described for ImImPy was followed in the preparation of 12, with a
yield of 32% after six steps. 1H NMR (DMSO-d6) δ 2.04 (s, 3H), 3.73
(s, 3H), 3.84 (s, 3H), 3.97 (s, 3H), 3.99 (s, 3H), 7.01 (s, 1H), 7.50 (s,
1H), 7.52 (s, 1H), 7.56 (s, 1H), 9.30 (s, 1H), 10.27 (s, 1H), 10.40 (s,
1H). ESMS m/e calcd for C19H23N8O5 [M+ + H] 443.2, found 443.0.
AcNHImImPyCO2H (13). To a suspension of compound 12 (536
mg, 1.21 mmol) in 20 mL H2O was added NaOH (770 mg, 19.25
mmol). The solution was stirred for 48 h at room temperature, and the
aqueous solution was acidified to pH 2 at 0 °C. The resulting precipitate
was collected by filtration, washed with water, and dried to produce
compound 13 (408 mg, 79% yield) as a brown powder. 1H NMR
(DMSO-d6) δ 2.03 (s, 3H), 3.82 (s, 3H), 3.97 (s, 3H), 3.99 (s, 3H),
6.96 (d, J ) 2.0 Hz, 1H), 7.47 (d, J ) 2.0 Hz, 1H), 7.49 (s, 1H), 7.55
(s, 1H), 9.30 (s, 1H), 10.27 (s, 1H), 10.35 (s, 1H), 12.19 (br s, 1H).
ESMS m/e calcd for C18H21N8O5 [M+ + H] 429.2, found 429.1.
AcNHPyImImPyCONH(CH2)3NMe2 (14). To a solution of 9 (37.7
mg, 0.1 mmol) in MeOH-EtOAc (2:1, 1.5 mL) was added 10% Pd-C
(15 mg), and the reaction mixture was stirred for 18 h at room
temperature under H2 gas. The reaction mixture was filtered through
Celite with MeOH (20 mL), and concentrated in vacuo to produce 29.5
mg of yellow amine crude, which was used in the next step without
further purification. To a overnight-stirred solution of AcPyImCO2H
(30.5 mg, 0.1 mmol, 1H NMR (DMSO-d6) δ 1.95(s, 3H), 3.82 (s, 3H),
3.91 (s, 3H), 6.91 (s, 1H), 7.24 (s, 1H), 7.56 (s, 1H), 9.80 (s, 1H),
10.55 (s, 1H)), HOBt (15.3 mg, 0.1 mmol) and DCC (20.6 mg, 0.1
i
solution of CH2Cl2 (2 mL) was added Pr2NEt (0.1 mL, 0.57 mmol).
After O2NImCOCCl3 (72.2 mg, 0.27 mmol) was added at 0 °C, the
reaction mixture was stirred for 15 h at room temperature. Evaporation
of the solvent gave a yellow residue, which was purified by washing
in water (10 mL) and then diethyl ether (2 mL); the residue was then
1
dried in vacuo to produce 10 (58.2 mg, 44% for two steps). H NMR
(DMSO-d6) δ 1.61 (t, J ) 7.0 Hz, 2H), 2.15 (s, 6H), 2.25 (t, J ) 7.0
Hz, 2H), 3.19 (q, J ) 7.0 Hz, 2H), 3.80 (s, 3H), 4.01 (s, 3H), 4.06 (s,
3H), 6.91 (s, 1H), 7.23 (s, 1H), 7.58 (s, 1H), 8.13 (br s, 1H), 8.65 (s,
1H), 10.29 (s, 1H). FAB MS 501 [M+ + H].
ImImPy. To a solution of 10 (48.2 mg, 96.4 µmol) in MeOH-EtOAc
(2:1, 6 mL) was added 10% Pd-C (40 mg). After NaBH4 (8 mg, 0.21
mmol) in H2O (0.4 mL) was added dropwise at 0 °C, the reaction
mixture was stirred for 20 min at room temperature. The reaction
mixture was filtered through Celite and concentrated in vacuo to
produce 38 mg of yellow amine crude, which was used in the next
i
step without further purification. To a crude solution in Pr2NEt (0.1
mL, 0.57 mmol) was added Ac2O (0.1 mL, 1.05 mmol), and the reaction
mixture was stirred for 4 h at room temperature. Evaporation of the
solvent gave a yellow residue, which was purified by washing with
diethyl ether (2 mL) and dried in vacuo to produce ImImPy (23.2 mg,
1
47% for two steps). H NMR (DMSO-d6) δ 1.61 (m, 2H), 2.04 (s,
3H), 2.14 (s, 6H), 2.25 (m, 2H), 3.19 (m, 2H), 3.81 (s, 3H), 3.98 (s,
3H), 4.00 (s, 3H), 6.92 (s, 1H), 7.23 (s, 1H), 7.51 (s, 1H), 7.56 (s, 1H),
8.12 (br s, 1H), 9.34 (s, 1H), 10.27 (s, 1H), 10.30 (s, 1H). FAB MS
513 [M+ + H]. Synthetic procedures similar to those described for
compound ImImPy were followed in the preparation of ImImIm,
ImPyPy, and PyImPy.
ImImIm: 1H NMR (DMSO-d6) δ 1.63 (m, 2H), 2.04 (s, 3H), 2.14
(s, 6H), 2.25 (m, 2H), 3.22 (m, 2H), 3.96 (s, 3H), 3.97 (s, 3H), 4.01 (s,
3H), 7.50 (s, 1H), 7.52 (s, 1H), 7.65 (s, 1H), 8.34 (br s, 1H), 9.62 (s,
i
mmol) in DMF (0.6 mL) was added amine crude in Pr2NEt (0.1 mL,
0.57 mmol), and the reaction mixture was stirred for 3 h at room
temperature. Evaporation of the solvent gave a yellow residue, which
was purified by column chromatography (silica gel, 10-50% MeOH
in CHCl3, gradient elution) to produce 14 (34.8 mg, 55% for two steps).
9
3484 J. AM. CHEM. SOC. VOL. 125, NO. 12, 2003

DNA Interstrand Cross-Linking Agent Development
A R T I C L E S
1H NMR (DMSO-d6) δ 1.62 (m, 2H), 1.97 (s, 3H), 2.17 (s, 6H), 2.28
(m, 2H), 3.19 (m, 2H), 3.80 (s, 3H), 3.83 (s, 3H), 4.00 (s, 3H), 4.01 (s,
3H), 6.92 (s, 2H), 7.22 (s, 1H), 7.25 (s, 1H), 7.56 (s, 1H), 7.62 (s, 1H),
8.11 (br t, 1H), 9.38 (s, 1H), 9.83 (s, 1H), 10.23 (s, 1H), 10.36 (s, 1H).
ESMS m/e calcd for C29H39N12O5 [M+ + H] 635.3, found 635.2.
Denaturing Gel Electrophoresis of 7a-g Treated 5′-Texas Red-
Labeled Oligomers. Ten microliters of a reaction mixture containing
5′-Texas Red-labeled oligonucleotide with complementary strand (3
µM duplex concentration), the desired concentration of conjugates 7a-
g, ImImPy, and 10% DMF in 5 mM sodium cacodylate buffer (pH )
7.0) was incubated at 37 °C for 18 h. The reaction was quenched by
adding 1 µL of 1 mM calf thymus DNA, and then 110 µL of H2O was
added to this solution. A 1.5-µL aliquot was concentrated, and the
resulting residue was redissolved in 8 µL of loading dye (formamide
with New Fuchsin). A 2-µL aliquot was electrophoresed on a 15%
denaturing polyacrylamide gel using a Hitachi 5500-S DNA sequencer.
Chemical Elucidation of Interstrand Cross-Linked Sites in TR18/
ODN15. Ten microliters of a reaction mixture containing 5′-Texas Red-
labeled-TTACAGTGGCTGCCAGCA-3′ (TR18)/5′-TGCTGGCAGC-
CACTG-3′ (ODN15) or 5′-[TR]-TTATGCTGGCAGCCACTG-3′
(TR18L)/5′-CAGTGGCTGCCAGCA-3′ (ODN15L) (3 µM duplex
concentration), conjugate 7b (50 µM), ImImPy (100 µM), and 20%
DMF in 5 mM sodium cacodylate buffer (pH ) 7.0) was incubated at
37 °C for 15 h. A 1-µL aliquot was diluted to a volume of 15 µL with
H2O. A 1-µL aliquot of this solution was lyophilized, and the resulting
residue was redissolved in 8 µL of loading dye. A 2-µL aliquot of this
solution was subjected to the denaturing gel electrophoresis process as
described above. The rest of the reaction mixture was heated to 90 °C
for 5 min. One microliter of piperidine was added to this solution, and
the resulting solution was kept at 90 °C for 20 min. A 1-µL aliquot
was then subjected to gel electrophoresis. The solution was lyophilized
and then treated with calf intestine alkaline phosphatase (1 unit) in 11
µL of 5 mM sodium cacodylate buffer (pH ) 7.0) at 37 °C for 2 h. A
1-µL aliquot was subjected to the gel electrophoresis process as
described above.
Denaturing Gel Electrophoresis of Interstrand Cross-Link by
7b with Different Partner. Ten microliters of a reaction mixture
containing TR18/ODN15 (3 µM duplex concentration), conjugate 7b
(24 µM), Py-Im partner (15 or 30 µM), and 10% DMF in 5 mM sodium
cacodylate buffer (pH ) 7.0) was incubated at 37 °C for 18 h. The
reaction was quenched by adding 1 µL of 1 mM calf thymus DNA,
and then 110 µL of H2O was added to this solution. A 1.5-µL aliquot
was concentrated, and the resulting residue was redissolved in 8 µL of
loading dye. A 2-µL aliquot of this solution was subjected to the
denaturing gel electrophoresis process described above.
High-Resolution Gel Electrophoresis of 7b-Triamide-Treated 67-
Base Pair DNA Fragments. The top or bottom strand 5′-Texas Red-
labeled 67-base pair DNA fragments (TRF-67/R-67 or F-67/TRR-67,
10 nM duplex concentration) were alkylated using the desired
concentration of 7b with Py-Im triamides (1 µM) in 10 µL of 5 mM
sodium phosphate buffer (pH ) 7.0) containing 10% DMF maintained
at 23 °C for 21 h. The reaction was quenched by the addition of calf
thymus DNA (1 mM, 1 µL), and then heated for 5 min at 90 °C. The
DNA solution was concentrated under reduced pressure. The resulting
pellet was dissolved in 8 µL of loading dye, heated at 94 °C for 20
min, and then immediately cooled to 0 °C. A 2-µL aliquot of the
solution was electrophoresed on 15% denaturing polyacrylamide gel.
The (G + A) sequencing lanes were prepared by heating the solution
in 1 M piperidine at 90 °C for 10 min, according to published protocol.30
Preparation of 5′-Texas Red-5′-Biotin Double-Labeled 400-Base
Pair DNA Fragments. The 5′-Texas Red-labeled top strand and 5′-
biotin-labeled bottom strand 400-base pair DNA fragments (λDNA
F7315*-7714) were prepared by PCR using 5′-Texas Red-labeled-
AGAGGAGCTTGATGACACGG-3′ (λDNA forward 7315-7334), and
5′-biotin-labeled-ACTGCTCACCGGATGTAATGGTG-3′ (λDNA re-
verse 40789-40811) as primers, and purified by filtration using Suprec-
02. Similarly, λDNA R40789*-40811 were prepared by PCR using 5′-
Texas Red-labeled-ACTGCTCACCGGATGTAATGGTG-3′ (λDNA
reverse 40789-40811), and 5′-biotin-labeled-AGAGGAGCTTGAT-
GACACGG-3′ (λDNA forward 7315-7334) as primers. Their concen-
tration was determined from UV absorption data. The asterisk indicates
5′-Texas Red modification, and the nucleotide numbering sequence
begins with the replication site.
High-Resolution Gel Electrophoresis of 7b-ImImPy-Treated 400-
Base Pair DNA Fragments. The double-labeled 400-base pair DNA
fragments (9 nM duplex concentration) were alkylated by the desired
concentration of 7b and ImImPy in 10 µL of 5 mM sodium phosphate
buffer (pH ) 7.0) containing 10% DMF maintained at 23 °C for 18 h.
The reaction was quenched by adding calf thymus DNA (1 mM, 1
µL) and heating for 5 min at 90 °C. The DNA solution was concentrated
under reduced pressure. The resulting pellet was dissolved in 8 µL of
loading dye (formamide with Fuchsin red), heated at 94 °C for 20 min,
and then immediately cooled to 0 °C. A 2-µL aliquot of the resulting
solution was electrophoresed on a 6% denaturing polyacrylamide gel.
Identification of Interstrand Cross-Linking Sites in 400-Base Pair
DNA Fragments. The double-labeled 400-base pair DNA fragments
(9 nM duplex concentration) were alkylated by 7b (100 nM) with
ImImPy (25 µM) in 10 µL of 5 mM sodium phosphate buffer (pH )
7.0) containing 10% DMF maintained at 23 °C for 18 h. The reaction
was quenched by the addition of calf thymus DNA (1 mM, 1 µL).
MAGNOTEX-SA (0.2 mg) in a 2 × binding buffer (20 mM Tris-HCl
(pH ) 8.0), 2 mM EDTA, 2 M NaCl, 0.2% TritonX-100) (10 µL) was
added to the biotin-labeled DNA solution (10 µL). After mixing, the
tube was left at room temperature for 20 min, and the supernatant was
removed using a magnetic stand. A solution of 0.1 M aqueous NaOH-
NaCl (10 µL) was added to the reaction tube, and the supernatant was
removed. Ten microliters of H2O was added to reaction tube, and the
supernatant was removed while the tube remained on a magnetic stand.
Forty microliters of H2O was added to the reaction tube, and the solution
was stirred at 90 °C for 20 min. Piperidine (1 µL) was added to the
reaction tube, the solution was stirred at 90 °C for 20 min, and then
the supernatant was concentrated under reduced pressure. The resulting
pellet was dissolved in 8 µL of loading dye (formamide with Fuchsin
red), heated at 94 °C for 20 min, and then immediately cooled to 0 °C.
A 2-µL aliquot was electrophoresed on a 6% denaturing polyacrylamide
gel.
Identification of Interstrand Cross-Linking Sites in 67-Base Pair
DNA Fragments. The 5′-Texas Red-labeled top strand (TRF-67) and
5′-biotin-labeled bottom strand (R-67) 67-base pair DNA fragments
(10 nM) were alkylated by 7b (500 nM) with ImImPy (1 µM) in 20
µL of 5 mM sodium phosphate buffer (pH ) 7.0) containing 10% DMF
at 23 °C for 18 h. The reaction was quenched by the addition of calf
thymus DNA (1 mM, 1 µL). The reaction mixture was subjected a
detection method using a biotin-labeled complementary strand. The
results are shown in Figure 12c.
Molecular Modeling Studies. Minimizations were performed with
the Discover program (MSI, San Diego, CA) using CFF force-field
parameters. The starting structure was built on the basis of the crystal
structure of the (ImHpPyPy)2-d(CCAGTACTGG)2 complex26 and the
1H NMR structure of the Duo-Dist-d(CAGGTGGT)/d(ACCACCTG)
complex.13 The connecting parts were built using standard bond lengths
and angles. The CPI moiety of the assembled initial structure was energy
minimized using a distance-dependent dielectric constant of ꢀ ) 4r (r
stands for the distance between atoms i and j) and with convergence
criteria having an RMS gradient of less than 0.1 kcal/mol Å.
Acknowledgment. This study was partly supported by a
Grant-in-Aid for Priority Research from the Ministry of Educa-
tion, Science, Sports, and Culture, Japan.
(30) Maxam, A. M.; Gilbert, W. Proc. Natl. Acad. Sci. U.S.A. 1977, 74, 560.
JA028459B
9
J. AM. CHEM. SOC. VOL. 125, NO. 12, 2003 3485