Synthesis and biological studies of inducible DNA cross-linking agents

Article abstract of DOI:10.1002/anie.200700844

(Chemical Equation Presented) Checking the traps: When a phenyl selenide compound such as 1 was oxidized with sodium periodate or irradiated, an ortho-quinone methide intermediate formed and reacted with a DNA duplex by cross-linking. In the crystal structure of a derivative of 1 the biphenyl unit is twisted, and this might explain how 1 can interact effectively with the DNA backbone.

Full text of DOI:10.1002/anie.200700844

Communications
DOI: 10.1002/anie.200700844
DNA Cross-Linking
Synthesis and Biological Studies of Inducible DNA Cross-Linking
Agents**
Xiaocheng Weng, Lige Ren, Liwei Weng, Jing Huang, Shugao Zhu, Xiang Zhou,* and
Linhong Weng
Some antitumor drugs targeting DNA can induce DNA
interstrand cross-links (ISCs) or alkylation. Many bifunc-
tional alkylating agents (cross-linking agents), such as nitro-
gen mustard, cisplatin, and mitomycin C, can form covalent
bonds with DNA. This prevents the separation of DNA
double strands, disrupts DNA replication and transcription,
and finally leads to cell death.[1] One strategy to reduce the
toxicity of these alkylation agents for normal cells and
enhance potential medical application is to trigger the
prodrug in the tumor cells. Along these lines plenty of novel
inducible DNA cross-linking or alkylating agents have been
designed and studied because of their high intrinsic reactiv-
ities.[2] Quinone methides (QMs) have been found to be very
important intermediates in the process of DNA alkylation
and DNA cross-linkage.[3] They can be generated by many
methods, such as photo-irradiation,[4] oxidation,[5] and fluo-
ride-induced activation.[6] Our grouphas reported an efficient
DNA ISC agent based on a QM which triggers bisalkylation;
the biphenyl structure is suitable for effective interaction with
a DNA helix.[7] The biphenyl pharmacophore might be a good
starting point for the development of further modified and
decorated DNA recognition agents. Recently, Greenbergꢀs
laboratory developed a new method to effect efficient DNA
interstrand cross-linking by a nucleotide radical or methide
from a phenyl selenide precursor.[8,9] Meanwhile, selenium is
reported to be one of the most important elements used in
tumor therapy.[10] All of this information encouraged us to
design and investigate the new inducible reactivities of the
phenyl selenide compounds 1–3.
Compounds 1–3 were prepared from corresponding
commercially available phenol derivatives through successive
protection, bromination, substitution, and deprotection steps
(Scheme 1).[11] All new compounds have been fully charac-
terized by NMR spectroscopy and HRMS or elemental
analysis.[11]
[*] X.-C. Weng, L.-G. Ren, L.-W. Weng, J. Huang, S.-G. Zhu, Prof. X. Zhou
College of Chemistry and Molecular Sciences
Wuhan University
Hubei, Wuhan 430072 (P.R. China)
Fax: (+86)27-8733-6380
E-mail: xzhou@whu.edu.cn
Scheme 1. Synthesis of compound 2. a) CH3I, acetone, RT, 89%;
b) HBr, (CH2O)n, CH3COOH, 30%; c) (PhSe)2, NaBH4, DMF, RT, 62%;
d) BBr3, CH2Cl2, 08C, 47%.
Prof. L.-H. Weng
Department of Chemistry
Fudan University
Shanghai (P.R. China)
[**] X.Z. thanks the National Science Fund of China (no. 20672084), the
National Science Fund for Distinguished Young Scholars (no.
20425206), the Ministry of Education of China, the State Key
Laboratory of Natural and Biomimetic Drugs, and the State Key
Laboratory of Applied Organic Chemistry. We thank two referees for
their critical and helpful comments as well as Professor Dehua Pei,
Ohio State University, for insightful comments on this paper.
The DNA cross-linking abilities of compounds 1–3 were
studied using linearized plasmid DNA by denaturing alkaline
agarose gel electrophoresis (Figure 1, see Figure S1 in the
Supporting Information).[12] The duplex DNA was linearized
using the EcoRI restriction endonuclease digestion. DNA
cross-linking experiments were carried out in phosphate
buffer (pH 7.7).
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
8020
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2007, 46, 8020 –8023

Angewandte
Chemie
Figure 1. Concentration dependence of compounds 1–3 in DNA cross-
linking (NaIO4 oxidation). Lanes 1–8 with NaIO4 (5 mm): lane 1: 0.7 mg
pBR322 + 10 mm 1; lane 2: 0.7 mg pBR322 + 50 mm 1 (cross-linking
yield 50%); lane 3: 0.7 mg pBR322 + 100 mm 1 (84%); lane 4: 0.7 mg
pBR322 + 1 mM 2; lane 5: 0.7 mg pBR322 + 5 mm 2 (43%); lane 6:
0.7 mg pBR322 + 10 mm 2 (72%); lane 7: 0.7 mg pBR322 + 100 mm 3;
lane 8: 0.7 mg pBR322 + 1000 mm 3 (30%). Lanes 9–11 without
NaIO4: lane 9: 0.7 mg pBR322 + 100 mm 1; lane 10: 0.7 mg pBR322 +
10 mm 2; lane 11: 0.7 mg pBR322 + 1000 mm 3; lane 12: 0.7 mg
pBR322 + NaIO4; lane 13, 0.7 mg pBR322 (control). Mark lane: 1.5 mg
lambda DNA/Hindiii (molecular weight standard).
Figure 2. The duplex consisting of sequence OD1 and its comple-
ment, and the cross-linking of compounds 1 and 2 with OD1
(oxidation with 50 mm NaIO4). 1+OD1: 1 (10 mm) + OD1(4 mm) +
NaIO4; 2 + OD1: 2 (10 mm) + OD1(4 mm) + NaIO4; NM + OD1:
nitrogen mustard (10 mm) + OD1 (4 mm). A 39-mer oligonucleotide
served as a molecular weight marker. TAMRA= 6-carboxytetramethyl-
rhodamine, succinimidyl ester.
The DNA cross-linking abilities of the three compounds
in the presence of sodium periodate, an oxidant which does
not react with common proteins and nucleotides, were studied
(Figure 1).[13] Compound 2 produced nearly 80% DNA cross-
linking at a concentration as low as 10 mm and was found to be
about 10 times more efficient than compound 1 as a DNA
cross-linking agent. The biphenol compound 2 displays very
potent ISC properties; this is consistent with our previously
published result that structural conformation might be a key
factor for this kind of ISC.[7] However, compound 3, which is
synthesized from the unstable hydroquinone, produced only
slight DNA cross-linking even at high concentration
(Figure 1, lane 8). This may be explained by its instability,
especially upon oxidation by NaIO4. We propose that the
rapid and complete oxidation of the hydroquinone frame-
work prevented ISCs between DNA and compound 3.
Upon photochemical activation (50-W high-pressure
mercury lamp), with Rose bengal as the singlet oxygen
sensitizer,[9] these compounds also exhibit fine DNA cross-
linking abilities (see Figure S1 in the Supporting
Information). The phenomenon of interstrand cross-linking
was the most remarkable for compound 2, followed by
compound 1, and compound 3, which was consistent with the
results with NaIO4 as the oxidant.[11] These observations
suggest that the designed compound 2 could find potential use
as an auxiliary drug in photodynamic therapy.
To study the site specificity of the DNA cross-linking, we
chose the 5’-TAMRA-labeled 20-mer oligonucleotide OD1
and its complementary strand as substrates to study the
reaction on short oligonucleotide duplexes (Figure 2). After
annealing, the non-cross-linked duplex was mixed with the
phenyl selenide derivatives. Then NaIO4 solution was added
in excess to this mixture. The reaction mixture was analyzed
by 20% denaturing PAGE. As it is a well-known DNA ISC
reagent, nitrogen mustard served as a control; it induces
interstrand DNA cross-linking, which was evident from the
slowly migrating band in the gel.[14] The samples of oligonu-
cleotide with compounds 1 and 2 were also observed to form
this kind of band (Figure 2, see Figure S2 in the Supporting
Information).[11] Since the molecular weight of the 39-mer
oligonucleotide is similar to that of the cross-linked products,
it served as a molecular-weight marker in DPAGE. The
similar velocity of the bands arising from the 39-mer
oligonucleotide and the high-molecular-weight product pro-
duced by oligonucleotide and compound 1 or 2 indicates the
formation of cross-linked products (Figure 2, see Figure S3 in
the Supporting Information).[11]
The gel-purified cross-linked DNAs were isolated by
electrophoresis and treated by several methods to research
their alkylation and cross-linking sites. Under alkaline con-
ditions (1m piperidine, 908C), direct strand breaks at guanine,
cytosine, and adenine residues were observed, and these
suggested that the alkylation occurred at these three residues
(see Figures S4 and S5 in the Supporting Information).[11]
Owing to the high reactivity of the cross-linking agent, the
product mixture is too complex to determine the interstrand
cross-linked site. In a separate experiment the exact cross-
linking site was determined by hydroxyl radical cleavage with
each fluorescence-labeled strand. The decreased intensity of
fragments from C15 to C20 in OD1 and the absence of
cleavage product in OD2 defined that CAGACC from C15 to
C20 in OD1 as the cross-linked region (see Figures S6 and S7
in the Supporting Information).[11] An interesting phenom-
enon of the appearance of a strong band at C15 is currently
under investigation.
In keeping with the mechanism reported by Greenberg
et al.,[8,9] we propose that this reaction might proceed via an o-
quinone methide intermediate (see Scheme S1 in the
Supporting Information ).[11] When the phenyl selenide
compounds were oxidized in the presence of a large excess
of ethyl vinyl ether (EVE) as a trapping agent for o-QM, we
obtained the expected QM–EVE adducts (Scheme 2).[15]
Therefore, the ISC properties of phenyl selenide deriva-
tives might be affected by two factors. First, compound 1
reacts to form interstrand cross-linking in two steps, while
compound 2 probably generates a bis(quinone methide)
intermediate.[7] This conclusion could explain the obvious
difference between compounds 1 and 2 concerning cross-
linking properties.
Angew. Chem. Int. Ed. 2007, 46, 8020 –8023
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
8021

Communications
crystallized product was filtered, washed with water, and dissolved in
dichloromethane. The organic phase was washed with saturated
sodium bicarbonate solution several times, dried over anhydrous
Na2SO4, and concentrated to dryness. The residue was purified by
column chromatography (hexane/ethyl acetate, 10:1) to yield the
desired product as a white solid (0.92 g, 30% yield). 1H NMR
(CDCl3): d = 3.94 (s, 6H), 4.62 (s, 4H), 6.93 (d, J = 8.4 Hz, 2H), 7.45
(dd, J = 2.4, 8.4 Hz, 2H), 7.51 ppm (d, J = 2.40 Hz, 2H); 13C NMR
(CDCl3): d = 29.19, 56.02, 111.56, 126.65, 128.52, 129.44, 133.17,
156.92 ppm. FAB-MS [M+] 400.
2c: Under N2 atmosphere, diphenyl diselenide (0.622 g, 2 mmol)
and sodium borodydride (0.149 g, 4 mmol) were dissolved in anhy-
drous DMF (5 mL) at room temperature for 10 min. A solution of
compound 2b (0.400 g, 1 mmol) in DMF (10 mL) was added dropwise
by syringe and stirred overnight. The mixture was diluted with ethyl
acetate, washed with brine several times, dried with Na2SO4, and
concentrated to dryness. The product was purified by silica gel column
chromatography (chloroform/hexane, 2:3) to give the product as a
white solid (0.343 g, 62% yield). 1H NMR (CDCl3): d = 3.84 (s, 6H),
4.12 (s, 4H), 6.83 (d, J = 8.4 Hz, 2H), 6.92 (d, J = 2.1 Hz, 2H), 7.21–
7.25 (m, 8H), 7.48–7.52 ppm (m, 4H); 13C NMR (CDCl3): d = 27.31,
55.80, 110.98, 126.42, 127.51, 127.72, 128.74, 129.07, 130.93, 132.90,
134.52, 156.33 ppm; 77Se NMR (CDCl3): d = 374.44 ppm. Elemental
analysis calcd for C28H26O2Se2: C 60.88, H 4.74; found: C 60.50, H
4.60.
Scheme 2. Trapping reactions in the presence of EVE.
The crystal data and structure of compound 2c suggest
that the configuration might be another important factor for
the mode of action of 2 (Figure 3).[11] The dihedral angle
2: Under N2 atmosphere, BBr3 (neat, 0.26 mL, 2.88 mmol) was
added gradually to a solution of 2c (0.2 g, 0.36 mmol) in freshly
distilled CH2Cl2 (20 mL) at 08C. After the mixture had been stirred
for 2 h at room temperature, the solution was poured into a large
amount of ice water and extracted with CHCl3. The organic phase was
washed with brine, dried over anhydrous Na2SO4, and concentrated to
dryness. The residue was purified by column chromatography
(hexane/ethyl acetate, 3:1) to give the desired product as a white
solid (0.089 g, 46.7% yield). 1H NMR (CDCl3): d = 4.13 (s, 4H), 5.53
(s, 2H), 6.81–6.85 (m, 4H), 7.12 (dd, J = 2.4, 8.4 Hz, 2H), 7.23–7.27
(m, 6H), 7.46–7.49 ppm (m, 4H). 13C NMR (CDCl3): d = 28.08,
116.96, 124.67, 127.13, 128.04, 129.08, 129.30, 129.50, 133.54, 134.60,
153.25 ppm; 77Se NMR (CDCl3): d = 333.83 ppm. Elemental analysis
calcd for C26H22O2Se2: C 59.55, H 4.23; found: C 59.68, H 4.50.
Figure 3. X-ray crystal structure of 2c. ORTEP representation of 2c
with hydrogen atoms omitted for clarity.
Received: February 25, 2007
Revised: August 1, 2007
Published online: September 13, 2007
between the planes of the two phenyl rings of the biphenyl
core is around 298, and the two phenyl selenide groups are on
the same side like two arms; the structure of the deprotection
product, compound 2, should be analogous. This twist
structure probably makes the molecule suitable to interact
with the backbone of double-stranded DNA, and with the
appropriate configuration, the attached functional groups of 2
might bind DNA effectively.
In conclusion, we have synthesized a series of phenyl
selenide compounds and have shown their capability of
generating DNA cross-linking through the formation of o-
QM induced by periodate oxidation or light. We are currently
testing related compounds and testing their potency as DNA
cross-linking agents in biological and medicinal applications.
Keywords: antitumor agents · DNA alkylation ·
.
DNA cross-linking · quinone methide
[1] a) S. R. Rajski, R. M. Williams, Chem. Rev. 1998, 98, 2723 – 2796;
b) D. M. Noll, T. M. Mason, P. S. Miller, Chem. Rev. 2006, 106,
277 – 301.
[2] a) Y. Na, V. S. Li, Y. Nakanishi, K. F. Bastow, H. Kohn, J. Med.
Chem. 2001, 44, 3453 – 3462; b) K. Mitra, J. C. Marquis, S. M.
Hillier, P. T. Rye, R. Zayas, A. S. Lee, J. M. Essigmann, R. G.
Croy, J. Am. Chem. Soc. 2002, 124, 1862 – 1863; c) S. Park, C.
Anderson, R. Loeber, M. Seetharaman, R. Jones, N. Tretyakova,
J. Am. Chem. Soc. 2005, 127, 14355 – 14365.
[3] a) P. Wang, Y. Song, L. X. Zhang, H. P. He, X. Zhou, Curr. Med.
Chem. 2005, 12, 2893 – 2931; b) M. Freccero, C. Di Valentin, M.
Sarzi-Amadeꢀ, J. Am. Chem. Soc. 2003, 125, 3544 – 3553.
[4] a) E. Modica, R. Zanaletti, M. Freccero, M. Mella, J. Org. Chem.
2001, 66, 41 – 52; b) S. N. Richter, S. Maggi, S. C. Mels, M.
Palumbo, M. Freccero, J. Am. Chem. Soc. 2004, 126, 13973 –
13979; c) E. E. Weinert, R. Dondi, S. Colloredo, K. N. Franken-
field, C. H. Mitchell, M. Freccero, S. E. Rokita, J. Am. Chem.
Soc. 2006, 128, 11940 – 11947.
Experimental Section
Synthesis of compound 2: 2b: Paraformaldehyde (1.332 g, 44.4 mmol)
and 2a (1.90 g, 8.88 mmol) were suspended in 50 mL of acetic acid
and 5 mL of 40% HBr in 30 mL of acetic acid was added to the
mixture in one portion at 808C. The suspension cleared immediately.
The reaction was complete after 3 h at this temperature. The reaction
mixture was allowed to cool and was stored for 24 h at 48C. The
8022 www.angewandte.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2007, 46, 8020 –8023

Angewandte
Chemie
[5] a) J. Liu, H. Liu, R. B. van Breemen, G. R. J. Thatcher, J. L.
Bolton, Chem. Res. Toxicol. 2005, 18, 174 – 182; b) S. R. Angle,
W.-J. Yang, J. Org. Chem. 1992, 57, 1092 – 1097; c) D. C.
Thompson, K. Perera, R. London, Chem. Res. Toxicol. 1995, 8,
55 – 60.
[6] a) Q. P. Zeng, S. E. Rokita, J. Org. Chem. 1996, 61, 9080 – 9081;
b) P. Pande, J. Shearer, J. Yang, W. A. Greenberg, S. E. Rokita, J.
Am. Chem. Soc. 1999, 121, 6773 – 6779; c) W. F. Veldhuyzen, Y.-
F. Lam, S. E. Rokita, Chem. Res. Toxicol. 2001, 14, 1345 – 1351.
[7] P. Wang, R. P. Liu, X. J. Wu, H. J. Ma, X. P. Cao, P. Zhou, J. Y.
Zhang, X. C. Weng, X. L. Zhang, J. Qi, X. Zhou, L. H. Weng, J.
Am. Chem. Soc. 2003, 125, 1116 – 1117.
[8] a) I. S. Hong, M. M. Greenberg, Org. Lett. 2004, 6, 5011 – 5013;
b) I. S. Hong, M. M. Greenberg, J. Am. Chem. Soc. 2005, 127,
3692 – 3693; c) I. S. Hong, H. Ding, M. M. Greenberg, J. Am.
Chem. Soc. 2006, 128, 485 – 491; d) I. S. Hong, H. Ding, M. M.
Greenberg, J. Am. Chem. Soc. 2006, 128, 2230 – 2231.
[9] I. S. Hong, M. M. Greenberg, J. Am. Chem. Soc. 2005, 127,
10510 – 10511.
[10] L. Bjoerkhem-Bergman, U. Torndal, S. Eken, C. Nystroem, A.
Capitanio, E. H. Larsen, M. Bjoernstedt, L. C. Eriksson, Carci-
nogenesis 2004, 26, 125 – 131.
[11] See Supporting information.
[12] a) T. R. Cech, Biochemistry 1981, 20, 1431 – 1437; b) J. J. Tepe,
R. M. Williams, J. Am. Chem. Soc. 1999, 121, 2951 – 2955.
[13] a) L. Burdine, T. G. Gillette, H. J. Lin, T. Kodadek, J. Am. Chem.
Soc. 2004, 126, 11442 – 11443; b) B. Liu, L. Burdine, T. Kodadek,
J. Am. Chem. Soc. 2006, 128, 15228 – 15235.
[14] a) J. T. Millard, S. Raucher, P. B. Hopkins, J. Am. Chem. Soc.
1990, 112, 2459 – 2460; b) S. M. Rink, M. S. Solomon, M. J.
Taylor, S. B. Rajur, L. W. McLaughlin, P. B. Hopkins, J. Am.
Chem. Soc. 1993, 115, 2551 – 2557; c) S. M. Rink, P. B. Hopkins,
Biochemistry 1995, 34, 1439 – 1445.
[15] a) K. Nakatani, N. Higashida, I. Saito, Tetrahedron Lett. 1997, 38,
5005 – 5008; b) D. Brousmiche, P. Wan, Chem. Commun. 1998,
491 – 492.
Angew. Chem. Int. Ed. 2007, 46, 8020 –8023
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 8023