Geometry-Dependent Phosphodiester Hydrolysis Catalyzed by Binuclear Copper Complexes

Article abstract of DOI:10.1021/ic0340985

Two isomeric binuclear ligands PBTPA and MBTPA and their copper(II) complexes were prepared and examined for hydrolysis of a model phosphodiester substrate: bis(p-nitrophenyl) phosphate. A bell-shaped pH vs rate profile, which is in agreement with one mechanism proposed for bimetallonucleases/phosphatases, was observed for the binuclear complex of copper(II) and PBTPA. At pH 8.4, a maximum rate of 1.14 × 10^-6 s^-1-more than 10⁴-fold over uncatalyzed reactions-was achieved. However, the analogous complex of MBTPA did not show significant rate enhancement. The binuclear complex of copper(II) and PBTPA also showed 10-fold acceleration over mononuclear complex of copper(II) and tris(2-pyridylmethyl)amine (TPA) catalyzed reaction. A phage φX174 DNA assay showed that the complex of copper(II) and PBTPA promoted supercoiled phage φX174 DNA relaxation under both aerobic and anaerobic conditions, in contrast to the hydrolytic inactivity of the mononuclear complex of copper(II) and TPA.

Full text of DOI:10.1021/ic0340985

Inorg. Chem. 2003, 42, 7912−7920
Geometry-Dependent Phosphodiester Hydrolysis Catalyzed by Binuclear
Copper Complexes
Lei Zhu, Osvaldo dos Santos, Chi Wan Koo, Marc Rybstein, Louise Pape, and James W. Canary*
Department of Chemistry, New York UniVersity, New York, New York 10003
Received January 29, 2003
Two isomeric binuclear ligands PBTPA and MBTPA and their copper(II) complexes were prepared and examined
for hydrolysis of a model phosphodiester substrate: bis(p-nitrophenyl)phosphate. A bell-shaped pH vs rate profile,
which is in agreement with one mechanism proposed for bimetallonucleases/phosphatases, was observed for the
binuclear complex of copper(II) and PBTPA. At pH 8.4, a maximum rate of 1.14 × 10-6 s-1smore than 104-fold
over uncatalyzed reactionsswas achieved. However, the analogous complex of MBTPA did not show significant
rate enhancement. The binuclear complex of copper(II) and PBTPA also showed 10-fold acceleration over
mononuclear complex of copper(II) and tris(2-pyridylmethyl)amine (TPA) catalyzed reaction. A phage φX174 DNA
assay showed that the complex of copper(II) and PBTPA promoted supercoiled phage φX174 DNA relaxation
under both aerobic and anaerobic conditions, in contrast to the hydrolytic inactivity of the mononuclear complex of
copper(II) and TPA.
Introduction
hydrolysis of nucleic acid backbones or phosphodiester
model systems to various degrees.43-47 Among these, Cu-
The remarkable stability of the DNA phosphodiester
backbone is one of the essential requirements for the survival
and maintenance of life.1 The half-life for hydrolytic cleavage
of a typical DNA phosphodiester bond at 25 °C in neutral
water is estimated to be on the order of billions of years.2
On the other hand, the stability of the DNA phosphodiester
backbone provides a daunting task for scientists who wish
to engineer DNA structure, including specific cleavage and
ligation, to achieve desired functions.3 However, enzymes
such as polymerases, recombinases, topoisomerases, reverse
transcriptases, and others catalyze phosphoryl transfer reac-
tions efficiently under physiological conditions. Many such
enzymes include nuclease motifs with metal-ion-containing
active sites,4,5 which represent the structural basis for
biomimetic simulation.6
(7) For examples of hydrolyses of RNA/RNA model systems by metal
complexes, see the following and refs 8-18. Iranzo, O.; Kovalevsky,
A. Y.; Morrow, J. R.; Richard, J. P. J. Am. Chem. Soc. 2003, 125,
1988.
(8) Kawahara, S.-I.; Uchimaru, T. Eur. J. Inorg. Chem. 2001, 2437.
(9) Ait-Haddou, H.; Sumaoka, J.; Wiskur, S. L.; Folmer-Andersen, J. F.;
Anslyn, E. V. Angew. Chem., Int. Ed. 2002, 41, 4014.
(10) Worm, K.; Chu, F.; Matsumoto, K.; Best, M. D.; Lynch, V.; Anslyn,
E. V. Chem. Eur. J. 2003, 9, 741.
(11) Matsuda, S.; Ishikubo, A.; Kuzuya, A.; Yashiro, M.; Komiyama, M.
Angew. Chem., Int. Ed. 1998, 37, 3284.
(12) Rossi, P.; Felluga, F.; Tecilla, P.; Formaggio, F.; Crisma, M.; Toniolo,
C.; Scrimin, P. J. Am. Chem. Soc. 1999, 121, 6948.
(13) He, C.; Lippard, S. J. J. Am. Chem. Soc. 2000, 122, 184.
(14) Young, M. J.; Chin, J. J. Am. Chem. Soc. 1995, 117, 10577.
(15) Liu, S.; Hamilton, A. D. Bioorg. Med. Chem. Lett. 1997, 7, 1779.
(16) Liu, S.; Hamilton, A. D. Chem. Commun. 1999, 587.
(17) McCue, K. P.; Morrow, J. R. Inorg. Chem. 1999, 38, 6136.
(18) Fritsky, I. O.; Ott, R.; Kramer, R. Angew. Chem., Int. Ed. 2000, 39,
3255. For examples of hydrolyses of DNA/DNA model systems by
metal (except copper(II)) complexes, see refs 19-27.
An array of coordination compounds of transition metal
ions7-37 and lanthanide ions38-42 has been reported to catalyze
(19) Hettich, R.; Schneider, H.-J. J. Am. Chem. Soc. 1997, 119, 5638.
(20) Schnaith, L. M. T.; Hanson, R. S.; Que, L. J. Proc. Natl. Acad. Sci.
U.S.A. 1994, 91, 569.
* To whom the correspondence should be addressed. E-mail:
james.canary@nyu.edu. Fax: (212) 260 7905.
(1) Westheimer, F. H. Science 1987, 235, 1173.
(2) Williams, N. H.; Takasaki, B.; Wall, M.; Chin, J. Acc. Chem. Res.
1999, 32, 485.
(3) Hegg, E. L.; Burstyn, J. N. Coord. Chem. ReV. 1998, 173, 133.
(4) Stra¨ter, N.; Lipscomb, W. N.; Klabunde, T.; Krebs, B. Angew. Chem.,
Int. Ed. Engl. 1996, 35, 2024.
(21) Vichard, C.; Kaden, T. Inorg. Chim. Acta 2002, 337, 173.
(22) Gultneh, Y.; Khan, A. R.; Blaise, D.; Chaudhry, S.; Ahvazi, B.;
Marvey, B. B.; Butcher, R. J. J. Inorg. Biochem. 1999, 75, 7.
(23) Jurek, P.; Martell, A. E. Inorg. Chim. Acta 1999, 287, 47.
(24) Bencini, A.; Berni, E.; Bianchi, A.; Fedi, V.; Giorgi, C.; Paoletti, P.;
Valtancoli, B. Inorg. Chem. 1999, 38, 6323.
(25) Chapman, W. H.; Breslow, R. J. Am. Chem. Soc. 1995, 117, 5462.
(26) Leivers, M.; Breslow, R. Bioorg. Chem. 2001, 29, 345.
(27) Kondo, S.-I.; Shinbo, K.; Yamaguchi, T.; Yoshida, K.; Yano, Y. J.
Chem. Soc., Perkin Trans. 2 2001, 128.
(5) Wilcox, D. E. Chem. ReV. 1996, 96, 2435.
(6) Bashkin, J. K. Chem. ReV. 1998, 98, 937.
7912 Inorganic Chemistry, Vol. 42, No. 24, 2003
10.1021/ic0340985 CCC: $25.00 © 2003 American Chemical Society
Published on Web 11/06/2003

Phosphodiester Hydrolysis Catalyzed by Binuclear Cu Complexes
Chart 1 a
a
Mononuclear Cu(II) complex [Cu(II)(TPA)(H2O)]2+; binuclear ligands 2 and 3; bis(p-nitrophenyl)phosphate (BNPP).
(II) complexes have been studied extensively in the past
decade in promoting hydrolyses of DNA or model sub-
strates,3,28-37 because of Cu(II) ion’s similar coordination
chemistry and usually superior Lewis acidity to that of Zn-
(II), which has been observed as cofactor in many nu-
cleases.4,5
On the other hand, nucleic acid hydrolysis by many
metallonucleases/phosphatases is postulated to be facilitated
frequently by the cooperative action of two metal ions in
their active sites,4 with one metal ion activating the phos-
phodiester substrate, and the other increasing the pKa of the
bound water.5 Therefore, in a biomimetic two-metal-ion
system,2,48,49 one coordinated metal ion should be engineered
to bind a water ligand which ionizes to generate a hydroxide
nucleophile under physiological conditions. The other metal
ion would bind with phosphoryl oxygen atoms in the
substrate to neutralize the developing negative charges in
the transition state structures. The two coordination domains
should be connected by a suitable linker so that the intermetal
distance as well as the overall geometry of the binuclear
center can be optimized to generate high hydrolytic activity
toward given substrates. Therefore, binuclear, Cu(II)-contain-
ing complexes that feature appropriate geometries in regard
to the two metal binding sites would likely possess nuclease
activity in a biomimetic fashion34 (Chart 1).
The ligand tris(2-pyridylmethyl)amine (TPA, 1) is known
to form stable complexes with metal ions of small atomic
radii (∼0.8 Å) such as Zn(II) and Cu(II).50 In aqueous
solution, [Zn(1)OH2]2+ displays an acidic water molecule
with pKa 8.050 and does not aggregate under high-pH
conditions.51 A binuclear metal complex containing [Zn(II)-
(TPA)] motif has already been reported to be capable
of promoting hydrolysis of bis(p-nitrophenyl)phosphate
(BNPP).52 The pKa of [Cu(1)OH2]2+ is ∼7.4,50 which is
acidic enough to be deprotonated under near physiological
conditions to provide a metal ion bound hydroxide nucleo-
phile. In this Article, we first studied the coordination
chemistry between [Cu(1)]2+ and the model phosphodiester
compound BNPP in an effort to validate that [Cu(II)(TPA)]
domain binds a phosphodiester substrate in a fashion similar
to what is observed in metallonuclease-substrate complexes.
Second, binuclear ligands and complexes were prepared by
connecting two TPA motifs with proper spacers. Last,
structural and kinetics studies were performed with selected
binuclear coordination complexes.
(28) DNA/DNA model system hydrolyses by mononuclear copper(II)
complexes: the following and refs 29-33. Sreedhara, A.; Cowan, J.
A. Chem. Commun. 1998, 1737.
(29) Sreedhara, A.; Freed, J. D.; Cowan, J. A. J. Am. Chem. Soc. 2000,
122, 8814.
(30) Ren, R.; Yang, P.; Zheng, W.; Hua, Z. Inorg. Chem. 2000, 39, 5454.
(31) Hegg, E. L.; Mortimore, S. H.; Cheung, C. L.; Huyett, J. E.; Powell,
D. R.; Burstyn, J. N. Inorg. Chem. 1999, 38, 2961.
(32) Deck, K. M.; Tseng, T. A.; Burstyn, J. N. Inorg. Chem. 2002, 41,
669.
(33) Itoh, T.; Hisada, H.; Sumiya, T.; Hosono, M.; Usui, Y.; Fujii, Y. J.
Chem. Soc., Chem. Commun. 1997, 677.
(34) DNA/DNA model system hydrolyses by binuclear copper(II) com-
lpexes: the following and refs 35-37. Molenveld, P.; Engbersen, J.
F. J.; Kooijman, H.; Spek, A. L.; Reinhoudt, D. N. J. Am. Chem. Soc.
1998, 120, 6726.
(35) De Iuliis, G.; Lawrance, G. A.; Fieuw-Makaroff, S. Inorg. Chem.
Commun. 2000, 3, 307.
(36) Gajda, T.; Dupre, Y.; Torok, I.; Harmer, J.; Schweiger, A.; Sander,
J.; Kuppert, D.; Hegetschweiler, K. Inorg. Chem. 2001, 40, 4918.
(37) Rossi, L. M.; Neves, A.; Horner, R.; Terenzi, H.; Szpoganicz, B.; Sugai,
J. Inorg. Chim. Acta 2002, 337, 366.
(38) Wang, C.; Choudhary, S.; Vink, C. B.; Secord, E. A.; Morrow, J. R.
Chem. Commun. 2000, 2509.
(39) Epstein, D. M.; Chappell, L. L.; Khalili, H.; Supkowski, R. M.;
Horrocks, W. D., Jr.; Morrow, J. R. Inorg. Chem. 2000, 39, 2130.
(40) Welch, J. T.; Sirish, M.; Lindstrom, K. M.; Franklin, S. J. Inorg. Chem.
2001, 40, 1982.
(41) Branum, M. E.; Tipton, A. K.; Zhu, S.; Que, L. J. J. Am. Chem. Soc.
2001, 123, 1898.
(42) Kim, Y.; Franklin, S. J. Inorg. Chim. Acta 2002, 341, 107.
(43) Representative reviews on metal complex catalyzed nucleic acid/model
system hydrolyses: the following and refs 44-49. Franklin, S. J. Curr.
Opin. Chem. Biol. 2001, 5, 201.
(44) Ott, R.; Kramer, R. Appl. Microbiol. Biotechnol. 1999, 52, 761.
(45) Bashkin, J. K. Curr. Opin. Chem. Biol. 1999, 3, 752.
(46) Sreedhara, A.; Cowan, J. A. J. Biol. Inorg. Chem. 2001, 6, 337.
(47) Cowan, J. A. Curr. Opin. Chem. Biol. 2001, 5, 634.
(48) Chin, J. Curr. Opin. Chem. Biol. 1997, 1, 514.
(49) Molenveld, P.; Engbersen, J. F. J.; Reinhoudt, D. N. Chem. Soc. ReV.
2000, 29, 75.
(50) Anderegg, G.; Hubmann, E.; Podder, N. G.; Wenk, F. HelV. Chim.
Acta 1977, 60, 123.
(51) Chiu, Y.-H.; Canary, J. W. Inorg. Chem. 2003, 42, 5107-16.
(52) dos Santos, O.; Lajmi, A. R.; Canary, J. W. Tetrahedron Lett. 1997,
38, 4383.
Inorganic Chemistry, Vol. 42, No. 24, 2003 7913

Zhu et al.
should be handled in small quantity and with caution.)55 Zn(ClO4)2‚
6H2O (37 mg, 0.1 mmol) in MeOH (1.0 mL) and BPTPA (2, 33
mg, 0.05 mmol) in CH3CN (1.0 mL) were mixed together. The
mixture was heated until a homogeneous solution was formed. Most
of the solvent was then removed under vacuum so that the complex
(Zn2(2)(H2O)2(ClO4)4, 56 mg, 92%) was precipitated and collected
by filtration. Anal. Calcd for C42H42Cl4N8O18Zn2: C, 41.37; H, 3.47;
N, 9.19. Found: C, 41.39; H, 3.54; N, 8.85. Crystals suitable for
X-ray diffraction were prepared by diffusing various volatile
solvents into the CH3CN solution of the complex. Colorless
transparent crystals were obtained respectively with Et2O, THF,
EtOAc, and CH2Cl2 as diffusing solvents. The crystals prepared
from Et2O (7a) and THF (7b) diffusion were submitted for X-ray
structural determination.
[Cu2(2)(AN)2](ClO4)4 (8). (CAUTION! Perchlorate salts of metal
complexes with organic ligands are potentially explosive. They
should be handled in small quantity and with caution.)55 Cu(ClO4)2‚
6H2O (37 mg, 0.1 mmol) in MeOH (1.0, mL) and BPTPA (2, 33
mg, 0.05 mmol) in CH3CN (1.0 mL) were mixed together. A blue
precipitate appeared immediately. Most of the solvent was removed
under vacuum, and the bluish powdery complex (Cu2(2)(AN)2-
(ClO4)4, 60 mg, 95%) was collected by filtration. Anal. Calcd for
C46H44Cl4Cu2N10O16: C, 43.79; H, 3.51; N, 11.10. Found: C, 43.64;
H, 3.19; N, 10.93. Pleochroic (blue and violet) transparent crystals
in clusters (4-6 pieces/cluster) were obtained when EtOAc diffused
in the CH3CN solution of the complex.
Kinetic Measurements. Spectrophotometric grade DMSO was
used. All buffers were prepared with deionized H2O. The pH of
each solution was measured with a Mettler Toledo Inlab electrode
(Ag/AgCl reference) and an Accumet pH meter 10 at room
temperature at the end of each kinetic run. The free acids of HEPES,
EPPS, CHES, and CAPS (0.1 M with 0.2 M NaClO4 each) were
titrated to the desired pH with NaOH (1 M). All absorbance spectra
were recorded with a PerkinElmer Lambda 40 UV/vis spectrometer
fitted with a thermostated cuvette holder and sample transport
accessory. Samples were prepared in 1 cm path length glass
microcuvettes. Deionized water (160 µL), buffer (0.1 M, 500 µL),
DMSO (220 µL), the substrate BNPP (50 mM in DMSO, 40 µL),
binuclear ligand (5.0 mM in DMSO, 40 µL), and aqueous Cu(ClO4)2
(10.0 mM, 40 µL) solutions were measured into the cuvette
sequentially to constitute a 30% DMSO solution. After 15 min
equilibration time, the increase in concentration of p-nitrophenylate
(λ ) 400 nm) was measured every 100 s for 3 h. The initial slope
(<5% conversion) of a plot of the measured absorbance vs time
was determined (R2 (Cu(II)2(2)) > 0.999; R2 (Cu(II)2(3)) > 0.9)
The rate constants were calculated with the extinction coefficient
of p-nitrophenylate at 400 nm read from a pH-ꢀ correlation curve
for 30% DMSO solution at 55 °C. The experiment with mono-
nuclear catalyst Cu(II)(1) was performed in a similar way while
the mononuclear ligand (TPA, 1) concentration was 10.0 mM
instead of 5.0 mM.
Experimental Details
General. All reactions were performed in oven-dried glassware
under an atmosphere of argon gas. H NMR and 13C NMR were
1
recorded on a Varian-Gemini 200 MHz spectrometer. FAB mass
spectra were measured using the matrix NBA/CHCl3. ESI mass
spectra were measured with a PE SCIEX API electrospray LC/MS
spectrometer. Compounds TPA53 (1) and BrTPA54 (4) were
prepared by literature procedures.
[Cu(1)(BNPP)](ClO4). (CAUTION! Perchlorate salts of metal
complexes with organic ligands are potentially explosive. They
should be handled in small quantity and with caution.)55 Tris(2-
pyridylmethyl)amine (TPA, 1, 73 mg, 0.25 mmol), Cu(ClO4)2‚6H2O
(93 mg, 0.25 mmol), and bis(p-nitrophenyl)phosphate (BNPP, 85
mg, 0.25 mmol) were dissolved in MeOH (2.0 mL). To this meth-
anolic solution, NaOH (1.0 M) was added dropwise until the pH
reached 6. The solution was warmed to 50 °C. DMSO was added
in the heated solution dropwise with stirring until the precipitate
disappeared. The clear solution was allowed to stand at room
temperature overnight while blue crystals of [Cu(1)(BNPP)](ClO4)
were formed at the bottom of the container (124 mg, 65%). The
product was redissolved in CH3CN. The crystals suitable for X-ray
structure determination were grown by Et2O diffusion into its CH3-
CN solution over 48 h. MS (ESI): calcd (M+) 692.09, found 692.
PBTPA (2). To a solution of BrTPA (4, 369 mg, 1 mmol), 1,4-
benzene diboronic acid (5, 82 mg, 0.5 mmol), and Pd(PPh3)4 (173
mg, 0.15 mmol) in toluene (3.0 mL) and MeOH (2.0 mL) was added
aqueous Na2CO3 (3.0 mL, 2 M). The solution was stirred at reflux
for 16 h. After the reaction mixture was cooled, 5% HCl was added
dropwise until pH reached ∼1. The mixture was stirred for another
30 min. The crude product was diluted with H2O to 50 mL and
extracted with Et2O (2 × 50 mL). The aqueous layer was basified
carefully with 10% NaOH to pH > 10 and extracted with CH2Cl2
(3 × 50 mL). The combined organic fractions were collected and
dried over Na2SO4. Evaporation of the solvent under reduced
pressure gave the residue, which was chromatographed on alumina
gel with MeOH/CH2Cl2 (from 0/100 to 1.5/100) as eluant to give
1
the pure product (438 mg). The yield was 67%. H NMR (200
MHz, CDCl3): δ 8.54 (d, 4H, J ) 4.0 Hz), 8.13 (s, 4H), 7.8-7.5
(m, 14H), 7.13-7.17 (m, 4H), 3.97 (s, 12H). 13C NMR (50 MHz,
CDCl3): δ 159.9, 156.6, 149.6, 149.5, 140.2, 137.5, 136.8, 127.7,
123.4, 122.4, 121.9, 119.2, 60.8, 60.7. MS (FAB): calcd (M +
H+) 655.32, found 655.65. Anal. Calcd for C42H38N8: C, 77.04;
H, 5.85; N, 17.11. Found: C, 76.96; H, 5.93; N, 17.05.
MBTPA (3). MBTPA was prepared by Suzuki reaction between
4 and 1,3-benzene diboronic acid dipinacol ester (6) by the
procedure applied to 2 above. The crude product was chromato-
graphed on silica gel with MeOH/CH2Cl2 (from 0/100 to 2/100) in
the presence of 1% triethylamine to yield 278 mg of product. The
1
yield was 85%. H NMR (200 MHz, CDCl3): δ 8.63 (t, 1H, J )
1.5 Hz), 8.56-8.50 (m, 4H), 8.06 (dd, 2H, J ) 1.7, 7.8 Hz), 7.80-
7.50 (m, 15H), 7.13 (m, 4H), 3.97 (s, 12H). 13C NMR (50 MHz,
CDCl3): δ 160.1, 159.8, 157.0, 149.6, 140.4, 137.6, 136.9, 129.6,
127.9, 126.0, 123.4, 122.4, 121.7, 119.3 60.8. MS (MALDI-TOF):
calcd (M + H+) 655.32, found 655.9. Anal. Calcd for C42H40N8O
(3‚H2O): C, 74.98; H, 5.99; N, 16.65. Found: C, 74.64; H, 5.80;
N, 16.86.
Preparation of OX174 RFI DNA Solution. The EDTA in
φX174 RFI DNA purchased from New England Biolabs was
removed via ethanol precipitation followed by a 70% ethanol wash.
The pellet was suspended in 10 mM Tris-Cl buffer (pH 8.0, no
EDTA). The final concentration of the DNA solution was 0.1 µg/
µL.
Aerobic Reaction with Cu(II) Complexes. The reaction mixture
was prepared by adding Cu(ClO4)2 (2.0 µL, 50 mM), DMSO (2.0
µL), ligand 2 in DMSO (1.0 µL, 50 mM), the buffer (4.0 µL, 0.1
M CHES, 0.2 M NaClO4, pH 8.75), and φX174 DNA (1.0 µL, 0.1
g/L) to sterile 1.5 mL Eppendorf tubes sequentially; in the
experiments with no catalyst, with free Cu2+ or with mononuclear
[Zn2(2)(AN)2](ClO4)4 (7). (CAUTION! Perchlorate salts of metal
complexes with organic ligands are potentially explosive. They
(53) Toftlund, H.; Ishiguro, S. Inorg. Chem. 1989, 28, 2236.
(54) Chuang, C.-L.; dos Santos, O.; Xu, X.; Canary, J. W. Inorg. Chem.
1997, 36, 1967.
(55) Chem. Eng. News 1983, 61, 1 (Dec 5), 4.
7914 Inorganic Chemistry, Vol. 42, No. 24, 2003

Phosphodiester Hydrolysis Catalyzed by Binuclear Cu Complexes
Figure 1. Synthesis of binuclear ligands 2 and 3.
catalyst Cu(II)(1), the quantity of each added solution was adjusted
so that the percentage of DMSO remained at 30%.The total volume
was 10 µL. These mixtures were vortexed and centrifuged to
facilitate the complexation of ligands with Cu(II) ions before the
addition of φX174 DNA. The mixtures were then vortexed and
centrifuged again to initiate the reactions. Then the solutions were
incubated at 45 °C for 2 h. Immediately after incubation, 2 µL of
EDTA solution were added to all reaction mixtures, followed by 5
min of incubation at 37 °C to minimize precipitation. The reaction
mixtures were then quenched with 4 µL of FSM (purchased from
Sigma) before loading onto an agarose gel (purchased from Bio-
Rad). The DNA solutions were subjected to electrophoresis on a
1% agarose gel at either 22 mV in 1XTBE buffer solution
(purchased from United States Biochemical) overnight for the 20
cm long gel, or at 65 mV for 2 h in the same buffer for the midigel.
The gel was stained in 1 µg/mL ethidium bromide solution
(purchased from United States Biochemical) for 10 min and
destained for 2 h. The gel was photographed under UV light, and
bands of the scanned image were quantified by “Intelligent
Quantifier” software (RM Luton, Inc.).
intermetal distance in a bimetallic complex of 2 would be
∼4 Å, on the basis of the examination of CPK molecular
models, upon binding with a phosphodiester substrate, while
in 3 the distance would be shorter. The phenyl linkers also
provide some degree of rotational freedom so that they would
accommodate different substrates or transition state struc-
tures.56
Synthesis. The synthesis of 2 was completed through a
Suzuki reaction between BrTPA51,54 (4) and 1,4-benzene
diboronic acid (5) catalyzed by Pd(0) with 67% yield (Figure
1). For 3, 1,3-benzene diboronic acid was obtained57 first
and esterified with pinacol in CHCl3 to give 1,3-benzene
diboronic acid dipinacol ester (6), which was easily purified
by running through a silica plug. Compounds 4 and 6 were
then reacted in 2:1 ratio in the presence of Pd(0) catalyst to
give 3 in 85% yield.
The complex Zn2(2)(AN)2(ClO4)4 (7) was prepared by
mixing PBTPA (2) and Zn(ClO4)2 in 1:2 ratio in MeOH
and CH3CN. Crystals suitable for X-ray diffraction were
prepared by diffusing various volatile solvents into the CH3-
CN solution of the complex. Colorless transparent crystals
were obtained respectively with Et2O, THF, EtOAc, and CH2-
Cl2 as diffusing solvents. The crystals prepared from Et2O
and THF yielded two different structures (7a and 7b,
respectively) determined by X-ray diffraction. Complex Cu2-
(2)(AN)2(ClO4)4 (8) was prepared by the same method,
except that suitable crystals were only obtained when EtOAc
was diffused into the CH3CN solution of 8. The crystals were
formed as clusters (same phenomena were observed when
other solvents were used without exception). The pleochroic
crystals appeared either deep blue or pale violet as rotated
around their long axis.
Anaerobic Reaction with Cu(II) Complexes. Deoxygenated
water, buffer, and DNA solutions were prepared by either three or
four “freeze-pump-thaw” cycles. All solutions were stored under
a nitrogen atmosphere prior to use. All anaerobic stock solutions
were prepared in a nitrogen-filled inflatable polyethylene chamber
with built-in gloves. A sealed atomsbag was purged with N2 for 2
h prior to use. Reaction mixtures were prepared in the atomsbag
by the addition of appropriate volumes of stock solutions to the
reaction tubes. The procedure for mixing all reagents in the sterile
Eppendorf tubes was the same as described for the aerobic reaction.
Reactions between the complexes and DNA were initiated by quick
centrifugation, transfer to a nitrogen-filled vacuum desiccator, and
incubation in the sealed desiccator at 45 °C for 2 h. Procedures for
treatment of the mixtures after incubation are the same as those
described for the aerobic reaction.
X-ray Structures. All the crystal data and structural
refinements are summarized in Table 1. The X-ray structure
of [Cu(1)(BNPP)](ClO4) in Figure 258 shows that the Cu-
Results and Discussion
Both compounds PBTPA (2) and MBTPA (3) contain
two homogeneous N4 binding sites linked together by phenyl
linkers elaborated in such a fashion that both metal binding
sites are either para (2) or meta (3) to each other. The
(56) Breslow, R. Acc. Chem. Res. 1995, 28, 146.
(57) Nielsen, D. R.; McEwen, W. E. J. Am. Chem. Soc. 1957, 79, 3081.
(58) Farrugia, L. J. J. Appl. Crystallogr. 1997, 30, 565.
Inorganic Chemistry, Vol. 42, No. 24, 2003 7915

Zhu et al.
Table 1. Crystal Data and Structure Refinements for [Cu(1)(BNPP)](ClO4), [Zn2(2)(AN)2](ClO4)4 (7a, 7b), and [Cu2(2)(AN)2(ClO4)](ClO4)3 (8)
[Zn2(2)(AN)2](ClO4)4
(7a)
[Zn2(2)(AN)2](ClO4)4
(7b)
[Cu2(2)(AN)2(ClO4)](ClO4)3
(8)
[Cu(1)(BNPP)](ClO4)
empirical formula
fw
C30H26ClCuN6O12P
792.53
C56H63Cl4N13O17Zn2
1462.73
C52H56Cl4N13O16Zn2
1391.64
C46H44Cu2N10Cl4O16
1261.79
temp
173(2) K
173(2) K
173(2) K
173(2) K
wavelength
crystal system
space group
unit cell dimens
0.71073 Å
triclinic
P1h
0.71073 Å
monoclinic
Cc
0.71073 Å
monoclinic
P21/c
0.71073 Å
monoclinic
P21/c
a ) 14.652(2) Å
b ) 23.435(4) Å
c ) 15.434(3) Å
R ) 90°.
a ) 9.0338(7) Å
b ) 10.3878(8) Å
c ) 17.567(1) Å
R ) 90.969(1)°
â ) 100.544(1)°
γ ) 100.361(1)°
1592.1(2) Å3
2
a ) 21.411(3) Å
b ) 11.106(2) Å
c ) 29.054(4) Å
R ) 90°
a ) 11.7208(12) Å
b ) 18.3144(18) Å
c ) 14.1549(14) Å
R ) 90°
â ) 111.138(2)°
γ ) 90°
â ) 97.332(2)°
γ ) 90°
â ) 109.384(4)°.
γ ) 90°.
volume
Z
density (calcd)
absorption coeff
F(000)
6443.7(15) Å3
4
3013.6(5) Å3
2
4999.1(14) Å3
4
1.653 Mg/m3
0.897 mm-1
810
1.508 Mg/m3
0.988 mm-1
3016
1.534 Mg/m3
1.051 mm-1
1430
1.676 Mg/m3
1.148 mm-1
2576
crystal color, morphology
crystal size
θ range for data collection
index ranges
light blue, plate
0.38 × 0.32 × 0.16 mm3
1.18-27.53°
-11 e h e 11,
-13 e k e 13,
-22 e l e 22
19166
colorless, block
0.44 × 0.32 × 0.14 mm3
1.50-27.55°
-27 e h e 27,
-14 e k e 14,
-37 e l e 37
37432
colorless, block
0.30 × 0.25 × 0.13 mm3
1.75-25.04°
-13 e h e 13,
-21 e k e 21,
-16 e l e 16
22569
blue/violet (pleochroic), plate
0.25 × 0.20 × 0.10 mm3
1.47-25.17°
-17 e h e 16,
0 e k e 27,
0 e l e 18
56057
8946 [R(int) ) 0.0647]
7024
reflections collected
indep reflns
obsd reflns
7254 [R(int) ) 0.0180]
6392
14738 [R(int) ) 0.043]
12099
5322 [R(int) ) 0.037]
4387
completeness to θ ) 27.53° 98.7%
99.6%
multiscans
99.9%
multiscans
99.6%
multiscans
0.8939 and 0.7623
full-matrix least-
squares on F2
8946/0/707
absorption correction
max and min transmission
refinement meth
multiscans
1.000000 and 0.901804
full-matrix least-
squares on F2
7254/0/460
0.869 and 0.755
full-matrix least-
squares on F2
14738/2/706
1.026
0.864 and 0.764
full-matrix least-
squares on F2
5322/90/404
1.063
data/restraints/params
GOF on F2
1.019
1.057
final R indices [ I> 2σ(I)]
R indices (all data)
largest diff peak and hole
R1 ) 0.0295, wR2 ) 0.0805 R1 ) 0.0383, wR2 ) 0.0873 R1 ) 0.0481, wR2 ) 0.1232 R1 ) 0.0472, wR2 ) 0.1157
R1 ) 0.0352, wR2 ) 0.0831 R1 ) 0.0500, wR2 ) 0.0924 R1 ) 0.0604, wR2 ) 0.1345 R1 ) 0.0652, wR2 ) 0.1270
0.421 and -0.534 e Å-3
0.402 and -0.373 e Å-3
0.753 and -0.684 e Å-3
1.047 and -0.469 e Å-3
Table 2. Selected Bond Lengths (Å) and Angles (deg) of
[Cu(1)(BNPP)](ClO4)
Cu1-N1 2.0187(14)
Cu1-N2 2.0406(15) 83.28(6)
Cu1-N3 2.0258(15) 81.51(6)
Cu1-N4 2.1508(16) 80.31(6)
132.48(6)
109.45(6) 111.86(6)
Cu1-O1 1.9196(13) 174.27(6) 102.45(6) 94.40(6)
97.63(6)
N4
N1
N2
N3
Figure 3. ORTEP diagram (50% probability ellipsoids) of cationic portion
of [Zn2(2)(AN)2](ClO4)4 (7a). Hydrogen atoms are removed for clarity.
Figure 2. ORTEP diagram (50% probability ellipsoids) of the cationic
portion of [Cu(1)(BNPP)](ClO4). Hydrogen atoms are removed for clarity.
Cu(II) are almost identical to those of Zn(II) in the structure
of [Zn(1){OP(O)(C6H4NO2)O}Zn(1)](BPh4)2.59 These struc-
tures suggest that both [Cu(II)(TPA)] and [Zn(II)(TPA)]
would serve as valid phosphodiester binding motifs with
similar binding modes.
(II) center adopts a distorted trigonal bipyramidal geometry,
where one of the nonesterified oxygen atoms O1 of BNPP
occupies the apical position opposite to the tertiary nitrogen
atom N1. The four Cu-N bond lengths fall in the range
2.02-2.15 Å; the bond length of Cu-O1 is 1.92 Å (Table
2). The coordination chemistry and the geometric data of
(59) Adams, H.; Bailey, N. A.; Fenton, D. E.; He, Q.-Y. J. Chem. Soc.,
Dalton Trans. 1995, 697.
7916 Inorganic Chemistry, Vol. 42, No. 24, 2003

Phosphodiester Hydrolysis Catalyzed by Binuclear Cu Complexes
Figure 5. ORTEP diagram (50% probability ellipsoids) of [Cu2(2)(AN)2-
(ClO4)](ClO4)3 (8). Hydrogen atoms are removed for clarity.
Figure 4. ORTEP diagram (50% probability ellipsoids) of cationic portion
of [Zn2(2)(AN)2](ClO4)4 (7b). Hydrogen atoms are removed for clarity.
C26-C25 (47.923° in the structure) was varied. The shortest
intermetal distance that could be reached was 5.494 Å, which
is close to the values reported in some bimetallonucleases.61
The intermetal distance (Zn1-Zn1A) in structure 7b
(Figure 4) is 8.305 Å. The dihedral angle C28-C27-C26-
C25 (52.894° in the structure) was varied so that the shortest
distance between two metal centers was 5.480 Å, comparable
to that obtained in 7a. It was observed that the dihedral angles
between phenyl planes and the planes of the connecting
pyridyl groups in 7a and 7b differed by 5°. Thus in structure
7b, the plane of one pyridyl group in each [Zn(II)TPA]
moiety was almost parallel to the phenyl plane, whereas in
structure 7a this correlation was not observed. The subtle
difference from the two dihedral angles (which involved the
only rotational freedom in the complex) may very well have
contributed to the magnified difference in the crystal lattice
packing to result in two distinct solid structures.
Table 3. Selected Bond Lengths (Å) and Bond Angles (deg) for 7a,
7b, and 8
[Zn2(2)(AN)2]- [Zn2(2)(AN)2]- [Cu2(2)(AN)2(ClO4)]-
(ClO4)4
(ClO4)4
(ClO4)3
(7a)
(7b)
(8)
Bond Lengths
2.199(3)
M1-N1
M1-N2
M1-N3
M1-N4
2.166(3)
2.088(3)
2.051(3)
2.080(3)
2.026(3)
2.412(3)
1.992(3)
1.989(3)
2.036(3)
2.078(3)
2.085(3)
Bond Angles
80.98(13)
N1-M1-N2
N1-M1-N3
N1-M1-N4
N2-M1-N3
N2-M1-N4
N3-M1-N4
78.07(11)
79.84(11)
78.88(11)
120.98(11)
116.84(11)
111.31(10)
83.18(11)
82.14(12)
82.50(12)
85.16(11)
95.63(11)
164.42(13)
78.83(13)
79.49(12)
117.33(13)
114.89(13)
118.49(13)
Both 7a and 7b (Figures 3 and 4) structures displayed
distorted trigonal bipyramidal Zn(II) coordination geometry
with similar bond lengths and angles in the coordination
domains (Table 3). The ones grown from Et2O (7a) crystal-
lized in the space group Cc. Although there is evidence of
a center of symmetry in the structure, the space group is
almost certainly Cc and not C2/c. The main reason for this
deduction is that when the structure is solved in C2/c the
final R value is about 20%, some atoms display very poor
anisotropic thermal displacement parameters, and there are
areas of high residual difference electron density in nonsensi-
cal places. Another indicator is that the value of |E2 - 1| is
0.77, i.e., close to that expected for non-centrosymmetric
symmetry.
The crystals of 7a have both Et2O and CH3CN in the
lattice; however, the solvent molecules proved to be too
disordered to be modeled satisfactorily so their associated
electron density was removed from the structural refinement.
The ones grown from THF (7b) crystallized in the space
group P21/c. They have CH3CN in the lattice but no THF.
The solvent molecules were disordered but modeled satis-
factorily in the structural refinement.
In the Cu(II) complex [Cu2(2)(AN)2(ClO4)](ClO4)3 (8,
Figure 5), the Cu(II) ions, three nitrogen atoms from 2, and
a nitrogen atom from CH3CN form a square planar geometry,
with the Cu-N bond lengths ranging from 1.98 to 2.02 Å
(Table 3), in contrast to what was observed in the structure
of [Cu(1)(BNPP)](ClO4) and other documented Cu(II)-TPA
structures.62 The pyridyl nitrogen atom N2 and one oxygen
atom (O1) from a perchlorate associate weakly with Cu(II)
from the two apical positions. One perchlorate (Cl1) links
Cu1 and Cu2 on adjacent tetracations. These Cu‚‚‚O interac-
tions are very long with Cu1-O1 ) 2.621(4) Å and Cu2-
O2#1 ) 2.688(4) Å. These are at least 0.1 Å longer than
the mean value for Cu(II) 5,6 coordinate structures, as listed
in the International Tables for Crystallography, Vol. C, p
756. The nitrogen atoms from the p-di-2-pyridylbenzene
backbone do not bond strongly, if at all, to their respective
Cu atoms. Both are slightly longer than 2.4 Å and well
outside normal pyridyl bonding distances. These pyridyl rings
are also tilted 47.2° and 45.1° from the normal of the plane
formed by the Cu atoms and the remaining three N atoms,
respectively. The three perchlorate anions not involved in
linking the tetracations are involved in a number of important
weak C-H‚‚‚O type hydrogen bonds.
In structure 7a (Figure 3), the distance between two metal
ion centers (Zn1 and Zn2) is 8.350 Å. With the aid of a
commercial software package (MacroModel),60 the intermetal
distances were measured as the dihedral angle C36-C31-
The distance between two Cu(II) centers is 9.327 Å. By
varying the dihedral angle C42-C37-C6-C5 (33.453° in
(61) Kim, Y.; Eom, S. H.; Wang, J.; Lee, D.-S.; Suh, S. W.; Steitz, T. A.
Nature 1995, 376, 612.
(62) Allen, C. S.; Chuang, C.-L.; Cornebise, M.; Canary, J. W. Inorg. Chim.
Acta 1995, 239, 29.
(60) Mohamadi, F.; Richards, N. G. J.; Guida, W. C.; Liskamp, R.; Lipton,
M. C., C.; Chang, G.; Hendrickson, T.; Still, W. C. J. Comput. Chem.
1990, 11, 440.
Inorganic Chemistry, Vol. 42, No. 24, 2003 7917

Zhu et al.
ligands in species A (Figure 7).25 The solid curve is the
theoretical plot constructed from eq 163 employing the
second-order rate constants k1, k2, and k3 for species A, B,
and C and the appropriate pKa values (Figure 7). It is
presumed that A, B, and C are the only active catalytic
species of Cu(II)2(2) in solution.64 At low pH where species
A is dominant, there can be only Lewis acid activation of
the phosphodiester substrate, which is far from enough to
achieve significant enhancement,2 and little acceleration of
the hydrolysis is observed. Within pH range 8-9, the species
B is abundant in which one Cu(II)-bound-water is deproto-
nated to generate an active nucleophile. As in the two-metal-
ion mechanism proposed for a number of bimetallonucleases/
phosphatases,5 it is very likely that this species delivers a
nucleophilic hydroxide to the phosphate center while the
other metal ion is oriented to bind with the phosphoester
moiety to provide Lewis acid activation. When all the
geometric elements such as the intermetal distance are
adjusted to give an energetically favorable complex between
Cu(II)2(2) and phosphodiester transition state structure,
nucleophilic activation by one Cu(II) and Lewis acid
activation by the other would participate in the process in a
highly cooperative way to give more than the simple additive
result of both activations, and a facile hydrolysis reaction is
observed.2 At pH 8.38 where Cu(II)2(2) gives maximum
enhancement, the rate of catalyzed hydrolysis is more than
104-fold over uncatalyzed reaction, and ∼10-fold over the
reaction catalyzed by mononuclear catalyst Cu(II)(1).
Figure 6. Plot of the observed rates of hydrolyses of BNPP (2.0 mM)
catalyzed by Cu(II)2(2) (2, 0.20 mM, average of two runs), Cu(II)2(3) (9,
0.20 mM), and Cu(II)(1) (b, 0.40 mM) at various pH values (0.05 M buffer
solution containing 0.1 M NaClO4) in 30% DMSO solution (v/v). The
background rates in the absence of catalyst are subtracted. The solid line is
a theoretical fit by eq 1.
the structure), the shortest distance 6.480 Å could be reached.
However, since N6 and N2 do not bind, or bind very loosely,
to respective Cu atoms, this structure has more rotational
freedom than its Zn(II) counterpart. Therefore, even shorter
intermetal distances might be allowed in the solution
structures.
Hydrolysis of BNPP. In kinetic studies of BNPP hy-
drolysis probing the catalytic activity of the synthetic ligand-
metal complexes, a concentrated solution of substrate BNPP
(50 mM in DMSO), binuclear ligand (5.0 mM in DMSO),
and Cu(ClO4)2 (10.0 mM in H2O) were added sequentially
with H2O, DMSO, and buffer (0.1 M in H2O in the presence
of 0.2 M NaClO4 for keeping ionic strength) to give a 30%
DMSO solution. Final concentrations were 2.0 mM for
BNPP, 0.20 mM for binuclear ligand, 0.40 mM for Cu-
(ClO4)2, and 0.05 M for buffer so that binuclear ligand would
form 1:2 complex with Cu2+ in situ. Efficient metal binding
under these conditions was expected,50 and stable metal
complex formation was confirmed by the absence of
precipitation of metal hydroxide at high pH. The formation
of hydrolysis product p-nitrophenol/p-nitrophenylate was
monitored by absorbance at 400 nm. The observed first-order
rate constants (kobs) were calculated by the initial rate method.
In determination of pH vs rate correlation of hydrolysis
catalyzed by bimetallic catalysts, various buffer solutions
(HEPES, pH < 8.0; EPPS, 8.0 < pH < 8.9; CHES, 8.9 <
pH < 11.0; CAPS, pH > 11.0) were used to cover the pH
region of interest.25 A control experiment with mononuclear
ligand 1 was also performed, in which 1 was added at twice
the concentration of the binuclear ligands so that it would
form 1:1 complex with Cu2+.50 The pH vs rate profiles are
shown in Figure 6.
(k1aH2 + k2K5aH + k3K5K6)[Cu(II)2(2)]
kobs
)
(1)
2
K5K6 + K5aH + aH
When pH becomes >9, the second Cu(II)-bound water is
ionized, and the dominant species in solution is C. Species
C has two tightly bound hydroxide ions at Cu(II) centers,
which inhibit the coordination of the anionic phosphodiester
substrate to the catalyst.34,64 Although it is able to deliver
two nucleophilic hydroxides, it lacks the cooperativity
between two metal ion centers as in species B. Indeed, the
enhancement of hydrolysis of BNPP by C is about 4 times
less than the acceleration by B. (It is taken into account that
C delivers twice as much nucleophile as B.)
The complex Cu(II)2(3), however, did not significantly
enhance the hydrolysis of BNPP as shown in Figure 6.
Unlike the Cu(II)2(2) containing reaction mixture, which was
homogeneous at 55 °C throughout the reaction, Cu(II)2(3)
was difficult to dissolve under the same conditions, and
therefore formed a slightly heterogeneous mixture. Nonethe-
less the kinetic measurements were conducted. The observed
rate constants with R2 > 0.9 were plotted in Figure 6. The
k
obs values by Cu(II)2(3) float around 1 × 10-7 s-1 indepen-
dent of pH. CPK molecular models suggest that Cu(II)2(3)
In complex Cu(II)2(2) catalyzed BNPP hydrolysis, the pH
vs rate correlation in Figure 6 is an unsymmetrical bell-
shaped curve with a maximum observed first-order rate
constant (kobs) of 1.14 × 10-6 s-1 at pH ∼ 8.4, which is
most likely between the two pKas of the Cu(II)-bound water
(63) Chin, J.; Banaszczyk, M.; Jubian, V.; Zou, X. J. Am. Chem. Soc. 1989,
111, 186.
(64) Another potential explanation for the loss of activity at higher pH
could be reduction in the reaction order of the catalyst,3 although
related studies51 indicated that similar complexes do not aggregate in
aqueous solution.
7918 Inorganic Chemistry, Vol. 42, No. 24, 2003

Phosphodiester Hydrolysis Catalyzed by Binuclear Cu Complexes
Figure 7. Species A, B, and C are presumed active species in catalyzing hydrolysis of BNPP. The second-order rate constants for each species are
designated as k1, k2, and k3, respectively. The theoretical fit of kobs vs pH by eq 1 results in k1 ≈ 0; k2 ) (8.3 ( 0.7) × 10-3 M-1 s-1; k3 ) (3.8 ( 0.1) ×
10-3 M-1 s-1; K5 ) (1.8 ( 0.3) × 10-8 M; K6 ) (2.5 ( 0.9) × 10-9 M.
Figure 8. Electrophoretic separations of φX174 DNA (0.02 g/L) on 1% agarose gel followed cleavage with different catalysts (10.0 mM Cu(II) total
concentration) in pH 8.75 (0.10 M CHES, 0.20 M NaClO4) 30% DMSO solution (v/v) at 45 °C for 2 h. Catalyst X: Cu(II)2(2). Y: Cu(II)(1). 8a: experiment
under aerobic conditions. 8b: under anaerobic conditions.
has a shorter intermetal distance than Cu(II)2(2) does, which
might contribute to its ability to form species with atoms
bridged between two Cu(II) centers, thus hampering its
catalytic efficiency in BNPP hydrolysis.22,23 The other
probable explanation is that the two [Cu(II)TPA] domains
that are meta to each other in Cu(II)2(3) are unable to orient
the two Cu(II) centers into positions relative to the substrate
that are favorable for phosphoryl transfer reactions.23
(L ) 2, n ) 2; L ) 1, n ) 1) at pH 8.75 and 45 °C for 2 h
in 30% DMSO solution. The complete conversion of form
I (supercoiled) DNA to form II (nicked) was observed with
either 2 (5.0 mM) or 1 (10.0 mM) as the ligand at 10.0 mM
total Cu(II) concentration, whereas a significant amount of
form I remained when only free Cu2+ was present (Figure
8a). The same experiment conducted with Zn(II) as the
coordinating metal ion did not show catalytic advantage of
binuclear complex over either Zn(II)-TPA mononuclear
complex or free Zn(II) ion.
However, it is well-known that Cu(II) complexes promote
oxidative nucleic acid strand scission through various
pathways.65-67 In cleavage and ligation reactions of DNA
in nucleic acid biotechnology, they are preferred to be
reversible transformations rather than the irreversible process
that results from DNA oxidative cleavage.3 In order to probe
the hydrolytic activity of the Cu(II) complexes, an anaerobic
assay was conducted under otherwise identical conditions
(Figure 8b). The complete conversion of form I to form II
was again observed with Cu(II)2(2) as the catalyst. However,
complex Cu(II)(1) and free Cu2+ ions showed similarly
dismal activity in promoting hydrolytic supercoiled DNA
relaxation. The percentages of form I and form II products
obtained were semiquantified by densitometry, which showed
that 96% supercoiled φX174 DNA was hydrolyzed to the
The binuclear complex of Zn(II) and PBTPA was also
tested for BNPP hydrolysis activity. Under identical condi-
tions, the maximum first-order reaction rate 2.3 × 10-8 s-1
was observed at pH 8.2. It was about 100-fold slower than
what was observed with binuclear Cu(II) complex. The bell-
shaped pH vs rate profile characteristic of binuclear coop-
erativity was not observed. The difference between activities
of Zn(II) and Cu(II) complexes must in part originate from
the greater Lewis acidity of Cu(II). Additionally, the different
coordination geometries observed between Zn(II) and Cu-
(II) complexes in the X-ray structures may reflect different
solution structures between both complexes. The weakly
bound third pyridyl group in the Cu(II) complex structure
may provide additional rotational freedom, which the Zn-
(II) complex lacks, for binding the substrate in solution that
would result in an energetically favorable transition state
complex.
Hydrolysis of Phage OX174 DNA. A phage φX174 DNA
assay was performed to study the DNA cleavage activity of
Cu(II)2(2). Figure 8a shows the agarose gel electrophoresis
patterns for the cleavage of φX174 after treatment with CunL
(65) Pogozelski, W. K.; Tullius, T. D. Chem. ReV. 1998, 98, 1089.
(66) Melvin, M. S.; Tomlinson, J. T.; Saluta, G. R.; Kucera, G. L.;
Lindquist, N.; Manderville, R. A. J. Am. Chem. Soc. 2000, 122, 6333.
(67) Humphreys, K. J.; Karlin, K. D.; Rokita, S. E. J. Am. Chem. Soc.
2002, 124, 6009.
Inorganic Chemistry, Vol. 42, No. 24, 2003 7919

Zhu et al.
nicked form by binuclear Cu(II)2(2) complex, whereas only
61% hydrolysis was achieved by Cu(II)(1), which essentially
has the same activity as free Cu2+ ions. Therefore, the fact
that relaxation of supercoiled (form I) DNA occurred under
anaerobic conditions signifies that Cu(II)2(2) cleaves DNA
via a nonoxidative, hydrolytic pathway. Comparing Figure
8a with 8b, it is also noted that both Cu(II)2(2) and Cu(II)-
(1) are able to cleave form I via an oxidative pathway.
However, Cu(II)(1) did not enhance the hydrolysis over Cu2+
ion. The hydrolytic enhancement by participation of ligand
2 and the inactivity of complex Cu(II)(1) suggest that, in
this particular case, the activity of Cu(II)2(2) results from a
bimetallic cooperative mechanism, not merely the additive
result of two mononuclear metal binding units.
The two isomeric binuclear ligands 2 and 3 differ only
from the relative orientations of two metal binding sites with
respect to each other. Yet we observed drastic differences
in their catalytic efficiencies in hydrolyzing BNPP, a
phosphodiester model compound. Complex Cu(II)2(2) dis-
played greater activity toward catalyzing BNPP hydrolysis
than Cu(II)2(3) and mononuclear complex Cu(II)(1). At pH
8.4, the achieved rate enhancement by Cu(II)2(2) for BNPP
hydrolysis is >104. The fact that binuclear Cu(II)2(2) is a
much better catalyst than mononuclear Cu(II)(1) and its
observed bell-shaped pH vs rate profile are in agreement with
the two-metal-ion mechanism proposed for some bimetal-
lonucleases/phosphatases.5 The inactivity of Cu(II)2(3) com-
pared to Cu(II)2(2) underscores the subtlety in the biomimetic
design of effective catalysts. Furthermore, the φX174 RFI
DNA assay demonstrated that catalyst Cu(II)2(2) had much
greater activity toward hydrolyzing phage φX174 RFI DNA
than Cu(II)(1), a mononuclear Cu(II) complex. This study
shows that binuclear complex Cu(II)2(2) does not merely
double the activity of its mononuclear counterpart. It supplies
evidence for the two-metal-ion enzymatic mechanism by
achieving cooperativity between two metal ions in this
particular model system.
Conclusion
In summary, two isomeric homobinuclear ligands (2, 3)
designed to imitate the general active site structures of
bimetallonucleases/phosphatases were prepared. Published
metal-bound-water acidity data of [M(1)(H2O)]2+ (M ) Zn-
(II) or Cu(II)) complexes50 indicate that metal-bound hy-
droxide nucleophile would be provided in near neutral water;
the X-ray structure of [Cu(1)(BNPP)](ClO4) shows that [Cu-
(1)]2+ complex binds with the phosphodiester model com-
pound BNPP in a mode similar to what was observed in
numerous enzyme active sites. The X-ray structures of
coordination complexes [Zn2(2)(AN)2](ClO4)4 (7a, 7b) and
[Cu2(2)(AN)2(ClO4)](ClO4)3 (8) were examined. The Zn(II)
centers in 7a and 7b adopt a distorted trigonal bipyramidal
geometry, whereas each Cu(II) domain in 8 is square planar
with a perchlorate oxygen atom and a pyridyl nitrogen atom
loosely bound in apical positions. It was proved that the
intermetal distances between two metal centers in all the
structures could reach within 5.5 Å of each other, assuming
only two degrees of rotational freedom (the C-C bonds
between phenyl rings and the connecting pyridyl groups).
The distances could very well be even shorter if more
rotational freedom was allowed, which is apparent in the
structure of [Cu2(2)(AN)2(ClO4)](ClO4)3 (8).
Acknowledgment. We thank William W. Brennessel, Dr.
Neil R. Brooks, and Dr. Victor G. Young, Jr. in the X-Ray
Crystallographic Laboratory, Department of Chemistry,
University of Minnesota, for obtaining X-ray crystallographic
data. FAB mass spectra were obtained at the Michigan State
University Mass Spectrometry Facility. This work was
supported by the NSF (CHE-0316589). O.S. acknowledges
a NIH minority predoctoral fellowship.
Supporting Information Available: Derivation of eq 1 (PDF);
X-ray crystallographic files for [Cu(BNPP)(TPA)](ClO4), [Zn2-
(PBTPA)(AN)2](ClO4)4, and [Cu2(PBTPA)(AN)2(ClO4)](ClO4)3 in
CIF format. This material is available free of charge via the Internet
at http://pubs.acs.org.
IC0340985
7920 Inorganic Chemistry, Vol. 42, No. 24, 2003