Efficient enhancement of DNA cleavage activity by introducing guanidinium groups into diiron(III) complex

Article abstract of DOI:10.1016/j.bmcl.2007.11.001

Inspired by the structures of natural nucleases, guanidinium groups were introduced into binuclear iron(III) systems. Compared with the corresponding analogue without guanidinium groups, the new diiron(III) system led to considerable rate enhancement on DNA cleavage. The cooperativity between metal ions and guanidine groups was evidenced by the fact that no significant cleavage was observed after incubating pBR322 plasmid DNA with non-metalated ligands or free Fe³⁺ ion. DNA binding experiments indicated that introduction of positively charged guanidinium groups can obtain more than one order of magnitude enhancement in the affinity of complex with DNA.

Full text of DOI:10.1016/j.bmcl.2007.11.001

Available online at www.sciencedirect.com
Bioorganic & Medicinal Chemistry Letters 18 (2008) 109–113
Efficient enhancement of DNA cleavage activity by
introducing guanidinium groups into diiron(III) complex
Xiaoqiang Chen,a Jingyun Wang,b,* Shiguo Sun,a Jiangli Fan,a Song Wu,b
Jianfeng Liu,b Saijian Ma,b Lizhu Zhanga and Xiaojun Penga,*
aState Key Laboratory of Fine Chemicals, Dalian University of Technology, Dalian 116012, PR China
bDepartment of Bioscience and Biotechnology, Dalian University of Technology, Dalian 116023, PR China
Received 3 August 2007; revised 19 October 2007; accepted 1 November 2007
Available online 5 November 2007
Abstract—Inspired by the structures of natural nucleases, guanidinium groups were introduced into binuclear iron(III) systems.
Compared with the corresponding analogue without guanidinium groups, the new diiron(III) system led to considerable rate
enhancement on DNA cleavage. The cooperativity between metal ions and guanidine groups was evidenced by the fact that no sig-
nificant cleavage was observed after incubating pBR322 plasmid DNA with non-metalated ligands or free Fe3+ ion. DNA binding
experiments indicated that introduction of positively charged guanidinium groups can obtain more than one order of magnitude
enhancement in the affinity of complex with DNA.
ꢀ 2007 Elsevier Ltd. All rights reserved.
Mimicking the activities of nucleases is currently an
attractive research area in molecular biology since artifi-
cial nucleases have potential applications as novel
restriction enzymes and anticancer therapeutic agents.1
In the past few years, various synthetic metallonucleases
have been reported as DNA cleavage agents.2 Further-
more, much effort has focused on di- or polynuclear me-
tal complexes, which are typically more reactive than the
corresponding mononuclear metal complexes.3 Never-
theless, their activities are still much lower than that of
the corresponding enzymes. In nature, the high reactiv-
ity of nucleases is attributed to the cooperation between
metal ions and several functional groups of the amino
acid side chains present in the active site. The positively
charged residues of lysine, arginine, and histidine are
thought to stabilize phosphorane-like transition states
by electrostatic interactions, hydrogen bonding, and/or
proton transfer.4 Inspired by the structures of metallo-
enzymes, several examples of synthetic systems have
been reported and exhibited the enhanced activity by
introducing ammonium/guanidinium group to mono-
metallic complex.5 However, no study upon introducing
these hydrogen bonding donors to dinuclear metal com-
plex as DNA cleavage agent was reported so far.
2,6-Bis[((2-hydroxybenzyl)(2-pyridylmethyl)amino)methyl]-
4-methylphenol (La), which provides a N4O3-donor
coordination sphere, can bind two Fe(III) or Zn(II) ions
to form corresponding dinuclear complexes.6 In this
study, we prepared a new dinucleating ligands Lb as
described in Scheme 1, whose structure features bisguan-
idinium-containing arms. In the previous literatures, the
complexes based on diiron(III) core have also showed
high activity in DNA hydrolysis.7 We expect that the
combination of diferric center and bisguanidinium
groups can further enhance DNA cleavage activity.
Beginning with 2-(4-hydroxy-benzyl)-isoindole-1,3-
dione, Lb was synthesized according to the reaction
NH2
NH
NH2
NH
HN
HN
OH
HO
OH
HO
OH OH
N
N
N
N
N
N
N
N
Lb
La
*
Corresponding authors. Tel.: +86 411 88993899; fax: +86 411
88993906 (X.P.); e-mail: pengxj@dlut.edu.cn
Scheme 1. Molecular structures of ligands used.
0960-894X/$ - see front matter ꢀ 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2007.11.001

110
X. Chen et al. / Bioorg. Med. Chem. Lett. 18 (2008) 109–113
OH
N
OH
OH O
NH2
HMT/TFA, 110 o
C
NaBH3CN
N
H
H
N
Route A:
N2, 20 h, HCl, 63%
CH3OH
Pht
Cl OH Cl
Pht
Pht
1
O
N
Pht =
O
Pht
Pht
OH
OH
CH2Cl2 HO
N
N
r.t., Et3N
N
N
6
OH
OH
O
OH Cl
OH OH
SOCl2/CH2Cl2
HMT/TFA, 110 o
C
NaBH3CN/ZnCl2
H
MeOH, N2, r..t.
12 h, HCl, 82%
N2, r.t., 96%
N2, 20 h, HCl, 63%
Pht
Pht
Pht
Pht
3
2
Route B:
1
N
OH
OH
O
OH O
NH2
NaBH4
93%
N
N
HMT/TFA
N
N
H
H
110 oC, 48 h, 42%
CH3OH
5
4
Pht
Pht
OH
OH
HO
CH2Cl2
N
N
N
r.t., Et3N, 73%
N
6
Scheme 2. Synthesis of intermediate 6.
Pht
NH2
Pht
NH2
N
OH
OH
1) N2H4, EtOH, r.t., 12 h
2) NaOH (aq.), r.t., 45 min, 78%
OH
OH
HO
HO
N
N
N
N
N
N
N
7
6
O
S
O
I
BOC
S
S
CH3I
CH3OH
O
O
N
N
O
H2N
NH2
H2N
NH2
O
H
NaOH(aq.)
8
NH2
NH
H2N
HN
O
N
O
HN
O
NH
O
O
NH
HN
O
N
N
NH
HN
CH2Cl2
56%
O
O
1) TFA, CH2Cl2, r.t., 6 h
2) resin, 94%
OH
OH
HO
OH
OH
HO
N
N
N
N
N
N
N
Lb
9
Scheme 3. Synthesis of Lb.

X. Chen et al. / Bioorg. Med. Chem. Lett. 18 (2008) 109–113
111
Figure 1. Left: agarose gel electrophoresis of pBR322 plasmid DNA treated with Fe–La (lanes 2–3) or Fe–Lb (lanes 4–6) for 1 h in Hepes buffer
(20 mM, pH 7.0) at 37 ꢁC. Key: Lane 1, control DNA; Lane 2, Fe–La 20 lM. Lane 3, Fe–La 50 lM. Lane 4, Fe–Lb 10 lM. Lane 5, Fe–Lb 20 lM.
Lane 6: Fe–Lb 50 lM. Middle: agarose gel electrophoresis of pBR322 plasmid DNA treated with ligands for 2 h in Hepes buffer (20 mM, pH 7.0) at
37 ꢁC. Key: Lane, 1 La 50 lM. Lane 2, Lb 50 lM. Lane 3, control DNA. Right: agarose gel electrophoresis of pBR322 plasmid DNA treated with
FeCl3 for 1 h in Hepes buffer (20 mM, pH 7.0) at 37 ꢁC. Key: Lane 1, control DNA. Lane 2, FeCl3 20 lM. Lane 3, FeCl3 40 lM. Lane 4, FeCl3
100 lM.
12
and then afforded the bisguanidinium ligand Lb by
passing through an anion-exchange column.
Stock solutions of complexes (Fe–La and Fe–Lb) using
in DNA cleavage experiment were prepared through
incubating ligands with 2-equiv FeCl36H2O in water.
Through HRMS spectra, the binuclear structures of
complexes were supported by the information that the
measured isotopic distribution is coincident with the cal-
culated one (Figure S1 and S2). Incubation of pBR322
plasmid DNA (0.02 lg/lL, 32 lM bp) with the stock
solution of complexes at pH 7.0 and 37 ꢁC for 1 h results
in a different extent of DNA cleavage depending on the
nature and concentration of the complexes (Fig. 1,
left)13. Complex Fe–La, lacking guanidinium subunit,
shows a much lesser cleavage activity: after incubation
of DNA in the presence of 50 lM of Fe–La, only about
66% of form I have been nicked. However, for Fe–Lb,
just 10 lM was required to convert completely from
the supercoiled form I to the nicked form II and eventu-
ally to the linearized form III. This result indicates a
much more pronounced DNA cleavage efficiency for
the introduction of bisguanidinium groups. The cooper-
ativity between metal ions and guanidine groups is evi-
denced by the fact that no significant cleavage was
observed after incubating pBR322 plasmid DNA with
50 lM non-metalated ligands for 2 h (Fig. 1, middle)
or with 20–100 lM FeCl3 for 1 h (Fig. 1, right). Further-
more, we carried out kinetic measurements of the DNA
cleavage reaction by Fe–Lb. As indicated in Fig. 2,
20 lM Fe–Lb led to complete cleavage of form I in a
very short time (no more than 1 min), then remains ac-
tive at a much slower rate. By the reason of too high ini-
Figure 2. Time-dependence of pBR322 plasmid DNA cleavage by
20 lM Fe–Lb in Hepes buffer (20 mM, pH 7.0) at 37 ꢁC. Key: Lane 1,
control. Lane 2, 1 min. Lane 3, 2 min. Lane 4, 4 min. Lane 5, 7 min.
Lane 6, 12 min.
sequences depicted in Schemes 2 and 3. To obtain the
precursor 6, we initially sought a similar route (Scheme
2, route A) used previously by Neves and co-workers in
the synthesis of La.6b Unexpectedly, the phthalimide
group (Pht) of monoformylated phenol was cleaved un-
der basic conditions due to the presence of pyridin-2-
ylmethanamine. Another route (Route B) was chosen
to keep phthalimide group from attack of primary
amine: first, functionalization in the ortho position of
phenol was achieved by formylation via the Duff reac-
tion,8 and monoformylated phenol was obtained
through controling the reaction condition; subsequently,
reduction of the monoformylated phenol by NaBH3CN
then gave hydroxymethyl phenol 2; after chlorination,
the resulting chloromethyl phenol 3 was finally incu-
bated with 2-hydroxy-5-methylisophthalaldehyde 5 in
the presence of triethylamine to yield the precursor 6.
Deprotection of 6 with hydrazine hydrate in ethanol
at room temperature produced 7. The synthesis of Lb
from 7 was preformed according to the similar proce-
dure described by Hassan and coworkers:9 after
guanylation of 7 using N,N0-bis(tert-butoxycarbonyl)-
S-methylisothiourea 8, the Boc-protected molecule 9
was deprotected in the presence of trifluoroacetic acid
Figure 3. Left: absorption spectra of Fe–Lb (2 · 10ꢀ4 M) in the presence of increasing amounts of CT DNA at room temperature. Right: the plot of
CDNA/(ea ꢀ ef) versus CDNA
.

112
X. Chen et al. / Bioorg. Med. Chem. Lett. 18 (2008) 109–113
tial activity, attempt to obtain the first-order rate con-
stant of DNA cleavage is unsuccessful.
In summary, guanidinium groups have been introduced
successfully to dinuclear iron(III) system. Compared
with the analogue (Fe–La), the new bimetallic system
led to considerable rate enhancement. The cooperative
action between the guanidinium groups and the two me-
tal centers is evidenced by the fact that there is no signif-
icant cleavage using non-metalated ligands or FeCl3.
Absorption titration experiments indicate that introduc-
ing of positively charged guanidinium groups to diiro-
n(III) complex can lead to more than one order of
magnitude enhancement in DNA binding abilities.
To exploring the interaction between complexes and
DNA, we carried out absorption titration measure-
ments14. With increasing concentration of CT DNA,
the changes of the intensity of the spectral band at
531 nm for Fe–La and 526 nm for Fe–Lb were moni-
tored. The binding constant was determined using the
following
equation:10
CDNA/(ea ꢀ ef) =
CDNA/
(eb ꢀ ef) + 1/ Kb(eb ꢀ ef), where ea, ef, and eb correspond
to Aobsd/[complex], the extinction coefficient for the free
iron complex, and the extinction coefficient for the iron
complex in the fully bound form, respectively. The intrin-
sic binding constants of complexes were obtained by the
Acknowledgments
This work was supported by the Ministry of Education
of China (No. 707016) and National Natural Science
Foundation of China No. 20706008 and 20725621.
ratio of the slope to intercept through the plot of CDNA
/
(ea ꢀ ef) versus CDNA: 1.9 · 104 Mꢀ1 for Fe–La and
3.0 · 105 Mꢀ1 for Fe–Lb (Fig. 3). Comparing with that
of Fe–La, more than one order of magnitude enhance-
ment in binding abilities of Fe–Lb should be attributed
to the more positive charges and hydrogen bonding ef-
fects of bisguanidinium groups. This result also indicated
that the introduction of guanidinium groups could
potentially accelate DNA cleavage by increasing the
DNA affinity of complex with DNA.
Supplementary data
Supplementary data associated with this article can be
found, in the online version, at doi:10.1016/j.bmcl.
2007.11.001.
A possible mechanism (Scheme 4) for the cleavage of
DNA by Fe–Lb was proposed. From the crystal struc-
ture data of complexes based on La, the distance be-
References and notes
1. (a) Hegg, E. L.; Burstyn, J. N. Coord. Chem. Rev. 1998,
173, 133; (b) Komiyama, M.; Sumaoka, J. Curr. Opin.
Chem. Biol. 1998, 2, 751; (c) Molenveld, P.; Engbersen, J.
F. J.; Reinhoudt, D. N. Chem. Soc. Rev. 2000, 29, 75; (d)
Niittyma¨ki, T.; Lo¨nnberg, H. Org. Biomol. Chem. 2006, 4,
15.
2. (a) Cowan, J. A. Chem. Rev. 1998, 98, 1067; (b) Morrow,
J. R.; Iranzo, O. Curr. Opin. Chem. Biol. 2004, 8, 192; (c)
Mancin, F.; Scrimin, P.; Tecilla, P.; Tonellato, U. Chem.
Commun. 2005, 2540.
3. (a) Chin, J. Curr. Opin. Chem. Biol. 1997, 1, 514; (b)
Williams, N. H.; Takasaki, B.; Wall, M.; Chin, J. Acc.
Chem. Res. 1999, 32, 485; (c) Rossi, P.; Felluga, F.; Tecilla,
P.; Formaggio, F.; Crisma, M.; Toniolo, C.; Scrimin, P.
J. Am. Chem. Soc. 1999, 121, 6948; (d) Iranzo, O.;
Kovalevsky, A. Y.; Morrow, J. R.; Richard, J. P. J. Am.
Chem. Soc. 2003, 125, 1988.
˚
tween Ca and Cb is calculated to be about 12 A
(Scheme 4, the top graphic).6b,c On the other hand, since
the distance between two consecutive phosphate groups
11
˚
in a B-DNA strand is about 7 A, we may conceive a
likely model of interaction between Fe–Lb and DNA:
the two guanidinium groups interact with the adjacent
phosphate groups, and such an arrangement forces the
central phosphate to interact with the two metal centers,
taking full advantage of their complementary roles for
the hydrolytic cleavage (Scheme 4, the bottom graphic).
N
N
M
M
4. Perreault, D. M.; Anslyn, E. V. Angew. Chem. Int. Ed.
Engl. 1997, 36, 432.
O
N
Cb
Ca
O
N
O
5. (a) Kovari, E.; Kramer, R. J. Am. Chem. Soc. 1996, 118,
12704; (b) Wall, M.; Linkletter, B.; Williams, D.; Lebuis,
A. M.; Hynes, R. C.; Chin, J. J. Am. Chem. Soc. 1999, 121,
4710; (c) Feng, G. Q.; Mareque-Rivas, J. C.; Williams, N.
H. Chem.Commun. 2006, 1845; (d) Feng, G. Q.; Mareque-
Rivas, J. C.; de Rosales, R. T. M.; Williams, N. H. J. Am.
Chem. Soc. 2005, 127, 13470; (e) Ait-Haddou, H.; Suma-
oka, J.; Wiskur, S. L.; Folmer-Andersen, J. F.; Anslyn, E.
V. Angew. Chem. Int. Ed. Engl. 2002, 41, 4014; (f) Livieri,
M.; Mancin, F.; Tonellatoa, U.; Chin, J. Chem. Commun.
2004, 2862; (g) Sheng, X.; Lu, X. M.; Zhang, J. J.; Chen,
Y. T.; Lu, G. Y.; Shao, Y.; Liu, F.; Xu, Q. J. Org. Chem.
2007, 72, 1799.
6. (a) Krebs, B.; Schepers, K.; Bremer, B.; Henkel, G.;
Althaus, E.; Miiller-Warmuth, W.; Griesar, K.; Haasel,
W. Inorg. Chem. 1994, 33, 1907; (b) Neves, A.; deBrito, M.
A.; Vencato, I.; Drago, V.; Griesar, K.; Haase, W. Inorg.
Chem. 1996, 35, 2360; (c) Adams, H.; Bradshaw, D.;
Fenton, D. E. Eur. J. Inorg. Chem. 2002, 535.
B
B
O
B
O
O
O
O
O
O
O
O
P
O
P
O
H
O-
P
O
O
O
O-
H
H
H
N
N
OH
Fe
NH
HN
O
Fe
O
NH2
NH2
N
N
N
N
O
Scheme 4. Top: stuctures of complexes based on La (M = Fe or Zn);
Bottom: proposed mechanism for cleavage of DNA by Fe–Lb.

X. Chen et al. / Bioorg. Med. Chem. Lett. 18 (2008) 109–113
113
7. (a) Liu, C. L.; Yu, S. W.; Li, D. F.; Liao, Z. R.; Sun, X. H.;
Xu, H. B. Inorg. Chem. 2002, 41, 913; (b) Schnaith, L. H.;
Hanson, R. S.; Jr, L. Q. Proc. Natl. Acad. Sci. USA 1994, 91,
569; (c) Neves, A.; Terenzi, H.; Horner, R.; Jr., A. H.;
Szpoganicz, B.; Sugai, J. Inorg. Chem. Commun. 2001, 4, 388.
8. (a) Ogata, Y.; Kawasaki, A.; Sugiura, F. Tetrahedron
1968, 24, 5001; (b) Lindoy, L. F.; Meehan, G. V.;
Svenstrup, N. Synthesis 1998, 7, 1029.
9. Ait-Haddou, H.; Wiskur, S. L.; Lynch, V. M.; Anslyn, E.
V. J. Am. Chem. Soc. 2001, 123, 11296.
10. Wolfe, A.; Schimer, G. H., Jr.; Meehan, T. Biochemistry
1987, 26, 6392.
13. pBR322 DNA cleavage experiments by binuclear metal
complexes: DNA cleavage experiments were performed
using pBR 322 (Takara) in 20 mM Hepes buffer, pH 7.0,
with incubating DNA (0.4 lg) at 37 ꢁC in the presence and
absence of metal complex for the definite time. The
cleavage reactions were stopped by addition of 3 lL of
loading buffer (0.25% bromophenol blue, 25% glycerol,
1 mM EDTA, 2% SDS). The electrophoresis of DNA
cleavage products was performed on 1% agarose gel. The
gels were run at 130 V for 45 min in 0.01 M, pH 7.0, TAE
buffer. The resolved bands were visualized by ethidium
bromide staining and quantified. A correction factor of
1.22 was utilized to account for the decrease in ability of
ethidium bromide to intercalate into form I DNA versus
forms II and III.
11. Sissi, C.; Rosso, P.; Felluga, F.; Formaggio, F.; Palumbo,
M.; Tecilla, P.; Tonilo, C.; Scrimin, P. J. Am. Chem. Soc.
2001, 123, 3169.
12. Ligand Lb: 1H NMR (CD3OD, 400 MHz, d ppm): 8.59 (d,
J = 6.0 Hz, 2H, Py–H), 8.07 (br, 2H, Py–H), 7.77 (d,
J = 6.0 Hz, 2H, Py–H), 7.60 (s, 2H, Ph–H), 7.36 (br, 2H,
Py–H), 7.35 (s, 2H, Ph–H), 7.15 (d, J = 8.0 Hz, 2H, Ph–
H), 6.72 (d, J = 8.0 Hz, 2H, Ph–H), 4.72 (s, 4H, –CH2N–),
4.62 (s, 4H, PyCH2–), 4.50 (s, 4H, –CH2N–), 4.27 (s, 4H,
–CH2N–), 2.64 (s, 3H, Ph–CH3); 13C (CDCl3, 100 MHz) d
158.49, 157.15, 154.55, 149.04, 147.16, 143.10, 136.66,
132.87, 132.80, 132.20, 129.25, 127.90, 127.03, 120.22,
117.82, 116.64, 56.44, 56.40,56.36, 45.23, 20.42; HRMS
calcd for C39H46N10O3, [M+H]+, 704.3911. Found:
704.3924.
14. Absorption titration of complexes binding to DNA: The
solutions of CT DNA gave a ratio of UV absorbance at
260 and 280 nm, A260/A280 > 1.8, indicating that the DNA
was sufficiently free of protein. The concentrated stock
solution of CT DNA (stored at 4 ꢁC and used not more
than 6 days) was prepared, and the concentration of DNA
in nucleotide phosphate was determined by UV absor-
bance at 260 nm. The molar absorption coefficient of CT
DNA was taken as 6600 Mꢀ1cmꢀ1. A solution (3 ml) of
complexes in water was incubated for 30 min at 25 ꢁC.
With addition of increasing amounts of DNA, the
influence of volume was negligible.