Electronic structure and spectro-structural correlations of Fe IIIZnII biomimetics for purple acid phosphatases: Relevance to DNA cleavage and cytotoxic activity

Inorg. Chem. 2010, 49, 11421–11438 11421

DOI: 10.1021/ic101433t

Electronic Structure and Spectro-Structural Correlations of Fe^IIIZn^IIBiomimetics

for Purple Acid Phosphatases: Relevance to DNA Cleavage and Cytotoxic Activity

Rosely A. Peralta,^†Adailton J. Bortoluzzi,^†Bernardo de Souza,^†Rafael Jovito,^†Fernando R. Xavier,^†

Ricardo A. A. Couto,^†Annelise Casellato,^†Faruk Nome,^†Andrew Dick,^‡Lawrence. R. Gahan,^‡Gerhard Schenk,^‡

§

||

^

#

ꢀ

Graeme R. Hanson, Flavia C. S. de Paula, Elene C. Pereira-Maia, Sergio de P. Machado, Patricia C. Severino,

#

3

,†

,‡

ꢀ

Claus Pich, Tiago Bortolotto, Hernan Terenzi, Eduardo E. Castellano, Ademir Neves,* and Mark J. Riley*

†

´

ꢀ

Departamento de Quımica, Universidade Federal de Santa Catarina, Florianopolis, South Carolina 88040-900,

Brazil, ^‡School of Chemistry and Molecular Biosciences, The University of Queensland, St. Lucia, QLD 4072,

Australia, ^§Centre for Advanced Imaging, The University of Queensland, St. Lucia, QLD 4072, Australia,

||

´

Departamento de Quımica, Universidade Federal de Minas Gerais, Belo Horizonte, MG, 31270-901, Brazil,

^

´

^

´

Departamento de Quımica Inorganica, Instituto de Quımica, Universidade Federal do Rio de Janeiro,

#

´

Rio de Janeiro, RJ, 21941-909, Brazil, Departamento de Bioquımica, Universidade Federal de Santa Catarina,

3

ꢀ

´

~

´

Florianopolis, South Carolina 88040-900, Brazil, and Instituto de Fısica de Sao Carlos, Departamento de Fısica

ꢀ

~

e Informatica, Universidade de Sao Paulo, Sao Carlos, SP, 13566-590, Brazil

Received July 18, 2010

Purple acid phosphatases (PAPs) are a group of metallohydrolases that contain a dinuclear Fe^IIIM^IIcenter (M^II= Fe,

Mn, Zn) in the active site and are able to catalyze the hydrolysis of a variety of phosphoric acid esters. The dinuclear

complex [(H₂O)Fe^III( μ-OH)Zn^II(L-H)](ClO₄)₂(2) with the ligand 2-[N-bis(2-pyridylmethyl)aminomethyl]-4-methyl-

6-[N⁰-(2-pyridylmethyl)(2-hydroxybenzyl) aminomethyl]phenol (H₂L-H) has recently been prepared and is found to

closely mimic the coordination environment of the Fe^IIIZn^IIactive site found in red kidney bean PAP (Neves et al. J. Am.

Chem. Soc. 2007, 129, 7486). The biomimetic shows significant catalytic activity in hydrolytic reactions. By using a

variety of structural, spectroscopic, and computational techniques the electronic structure of the Fe^IIIcenter of this

biomimetic complex was determined. In the solid state the electronic ground state reflects the rhombically distorted

Fe^IIIN₂O₄octahedron with a dominant tetragonal compression aligned along the μ-OH-Fe-O_phenolatedirection.

To probe the role of the Fe-O_phenolatebond, the phenolate moiety was modified to contain electron-donating

or -withdrawing groups (-CH₃, -H, -Br, -NO₂) in the 5-position. The effects of the substituents on the electronic

properties of the biomimetic complexes were studied with a range of experimental and computational techniques. This

study establishes benchmarks against accurate crystallographic structural information using spectroscopic techniques

that are not restricted to single crystals. Kinetic studies on the hydrolysis reaction revealed that the phosphodiesterase

activity increases in the order -NO₂rBr rH rCH₃when 2,4-bis(dinitrophenyl)phosphate (2,4-bdnpp) was used as

substrate, and a linear free energy relationship is found when log(k_cat/k₀) is plotted against the Hammett parameter σ.

However, nuclease activity measurements in the cleavage of double stranded DNA showed that the complexes

containing the electron-withdrawing -NO₂and electron-donating -CH₃groups are the most active while the cytotoxic

activity of the biomimetics on leukemia and lung tumoral cells is highest for complexes with electron-donating groups.

Introduction

esters and anhydrides within the pH range 3-8.^1-4In the

structure of red kidney bean PAP (rkbPAP)^1,4,5the Fe^IIIion

is coordinated to a tyrosine (Y167), a histidine (H325) and an

aspartate (D135) while the Zn^IIion is coordinated by two

Purple acid phosphatases (PAPs) are a group of metallo-

hydrolases that contain a heterovalent dinuclear Fe^IIIM^II

center (M^II= Fe, Mn, Zn) in the active site, and which are

able to catalyze the hydrolysis of a variety of phosphoric acid

ꢀ

(3) Mitic, N.; Valizadeh, M.; Leung, E. W. W.; de Jersey, J.; Hamilton, S.;

*To whom correspondence should be addressed. E-mail: ademir@

qmc.ufsc.br (A.N.), m.riley@uq.edu.au (M.J.R.).

Hume, D. A.; Cassady, A. I.; Schenk, G. Arch. Biochem. Biophys. 2005, 439,

154–164.

€

ꢀ

(1) Klabunde, T.; Strater, N.; Frohlich, R.; Witzel, H.; Krebs, B. J. Mol.

(4) Schenk, G.; Elliott, T. W.; Leung, E. W. W.; Mitic, N.; Carrington,

Biol. 1996, 259, 737–748.

L. E.; Gahan, L. R.; Guddat, L. W. BMC Struct. Biol. 2008, 8, 6.

(5) Truong, N. T.; Naseri, J. I.; Vogel, A.; Rompel, A.; Krebs, B. Arch.

Biochem. Biophys. 2005, 440, 38–45.

ꢀ

(2) Mitic, N.; Smith, S. J.; Neves, A.; Guddat, L. W.; Gahan, L. R.;

Schenk, G. Chem. Rev. 2006, 106, 3338–3363.

r

2010 American Chemical Society

Published on Web 11/16/2010

pubs.acs.org/IC

11422 Inorganic Chemistry, Vol. 49, No. 24, 2010

Peralta et al.

Chart 1. General Structure of Complexes 1-4 (R = CH₃, H, Br, NO₂)

Figure 1. Representative active site of purple acid phosphatases (PAPs).

The residue labels refer to red kidney bean PAP (rkbPAP).

histidines (H286, H323) and an asparagine (N201). The

Fe^IIIZn^IIcenter is bridged by a monodentate carboxylate group

of an aspartate (D164) and by a μ-hydroxo group. While

crystallographically ambiguous, terminal hydroxo and aqua

ligands were also modeled to complete the 6-fold coordination

spheres of the Fe^IIIand Zn^IIions, respectively, resulting in the

[(OH)Fe^III( μ-OH)Zn^II(OH₂)] core¹as shown in Figure 1.

Because of the relatively low resolution of protein crystal

structures, in particular related to PAPs enzymes,^1,2,4model

complexes often play an important role in the structural

characterization of the active sites in metalloproteins. Recently,

we have established a general method for the preparation

of mixed-valence heterobinuclear Fe^IIIM^IIcomplexes using

the unsymmetrical donor ligand 2-[N-bis-(2-pyridylmethyl)-

aminomethyl]-4-methyl-6-[N⁰-(2-pyridylmethyl)(2-hydroxy-

benzyl) aminomethyl]phenol (H₂L-H). The unsymmetric ligand

L-H^2-has both hard (N₂O₂) and soft (N₃O) sites that bind the

Fe^IIIand M^IIions, respectively, with vacant coordinating sites

available for bridging ligands. The complexes Fe^IIIFe^IIL-H,

Fe^IIIZn^IIL-H, Fe^IIIMn^IIL-H, Fe^IIINi^IIL-H, Fe^IIICu^IIL-H, Fe^III-

Co^IIL-H, and Ga^IIIZn^IIL-H containing two bidentate acetate

bridging ions have previously been reported.^6-13These com-

plexes contain a terminal Fe^III-phenolate motif that results in the

characteristic purple color of the chromophore because of an

intense ligand to metal charge transfer (LMCT) transition in the

visible region of the spectrum. In the solid state most of these

complexes have two acetate bridges, in contrast to the enzymes

where a (hydr)oxide links the two metal ions.^1-4

However, under suitable conditions it has been possible to

prepare the [H₂OFe^III( μ-OH)Zn^IIL-H]^2þcomplexes without

the bridging acetates.¹⁴In these complexes the bridging

acetates are replaced by a μ-OH^-bridge and a terminally

bound water to the Fe^III, with the M^IIion becoming five

coordinate. Their structures can be described as a distorted

octahedron edge sharing with a distorted trigonal bipyramid.

Metalloproteins with tyrosine bound to iron, such as

lactoferrin¹⁵and PAPs,^2,15-17show intense colors because

of the tyrosinate-to-iron(III) LMCT bands.^17,18Their enzy-

matic activity is considered to be related to the Fe^III/Fe^II

potential,¹⁹which will also affect the energy of these LMCT

bands. It has been demonstrated in a series of high-spin Schiff

base iron(III) complexes that these LMCT bands shift to

higher energies, and that the Fe^III/Fe^IIredox potential

becomes more negative as the number of phenolate-contain-

ing donor sites increases.²⁰These relationships have been

pursued using a series of monomeric iron complexes with two

phenolate ligands,²¹and where the electronic properties of

the phenolate group itself is varied by substitutions in the

position para to the phenolate oxygen.

Herein we report the geometric and electronic structure

and the phosphodiesterase-like activity of the [H₂OFe^III-

( μ-OH)Zn^IIL-R]^2þ(R = -CH₃, -H, -Br, -NO₂) series

of complexes (1-4) in detail (Chart 1). Since Zn^IIis spectro-

scopically silent, this series of complexes enables the study of

the S = 5/2 high spin Fe^IIIcenter in the catalytic site and the

LMCT spectra associated with the Fe^III-phenolate chromo-

phore, without magnetic coupling effects in the ground state.

Specifically, structural (X-ray crystallography) and spectro-

scopic (electron paramagnetic resonance (EPR), magnetic

circular dichroism (MCD), and UV-vis) techniques, as

well as electrochemical, potentiometric titration, and kinetic

(6) Batista, S. C.; Neves, A.; Bortoluzzi, A. J.; Vencato, I.; Peralta, R. A.;

Szpoganicz, B.; Aires, V. V. E.; Terenzi, H.; Severino, P. C. Inorg. Chem.

Commun. 2003, 6, 1161–1165.

(7) Karsten, P.; Neves, A.; Bortoluzzi, A. J.; Lanznaster, M.; Drago, V.

Inorg. Chem. 2002, 41, 4624–4625.

(8) Lanznaster, M.; Neves, A.; Bortoluzzi, A. J.; Aires, V. V. E.;

Szpoganicz, B.; Terenzi, H.; Severino, P. C.; Fuller, J. M.; Drew, S. C.;

Gahan, L. R.; Hanson, G. R.; Riley, M. J.; Schenk, G. J. Biol. Inorg. Chem.

2005, 10, 319–332.

(9) Lanznaster, M.; Neves, A.; Bortoluzzi, A. J.; Szpoganicz, B.; Schwingel,

E. Inorg. Chem. 2002, 41, 5641–5643.

(10) Neves, A.; de Brito, M.; Vencato, I.; Drago, V.; Griesar, K.; Haase,

W. Inorg. Chem. 1996, 35, 2360–2368.

(15) Anderson, B. F.; Baker, H. M.; Dodson, E. J.; Norris, G. E.;

Rumball, S. V.; Waters, J. M.; Baker, E. N. Proc. Natl. Acad. Sci. U.S.A.

1987, 84, 1769–1773.

(16) Averill, B. A.; Davis, J. C.; Burman, S.; Zirino, T.; Sanders-Loehr, J.;

Loehr, T. M.; Sage, J. T.; Debrunner, P. G. J. Am. Chem. Soc. 1987, 109,

3760–3767.

(11) Schenk, G.; Peralta, R. A.; Batista, S. C.; Bortoluzzi, A. J.; Szpoganicz,

B.; Dick, A.; Herrald, P.; Hanson, G. R.; Szilagyi, R. K.; Riley, M. J.; Gahan,

L. R.; Neves, A. J. Biol. Inorg. Chem. 2008, 13, 139–155.

(17) Yang, Y. S.; McCormick, J. M.; Solomon, E. I. J. Am. Chem. Soc.

1997, 119, 11832–11842.

ꢀ

(12) Smith, S. J.; Casellato, A.; Hadler, K. S.; Mitic, N.; Riley, M. J.;

Bortoluzzi, A. J.; Szpoganicz, B.; Schenk, G.; Neves, A.; Gahan, L. R.

J. Biol. Inorg. Chem. 2007, 12, 1207–1220.

(13) Xavier, F. R.; Neves, A.; Casellato, A.; Peralta, R. A.; Bortoluzzi, A. J.;

Szpoganicz, B.; Severino, P. C.; Terenzi, H.; Tomkowicz, Z.; Ostrovsky, S.;

Haase, W.; Ozarowski, A.; Krzystek, J.; Telser, J.; Schenk, G.; Gahan, L. R.

Inorg. Chem. 2009, 48, 7905–7921.

(14) Neves, A.; Lanznaster, M.; Bortoluzzi, A. J.; Peralta, R. A.; Casellato,

A.; Castellano, E. E.; Herrald, P.; Riley, M. J.; Schenk, G. J. Am. Chem. Soc.

2007, 129, 7486–7487.

(18) Ainscough, E. W.; Brodie, A. M.; Plowman, J. E.; Brown, K. L.;

Addison, A. W.; Gainsford, A. R. Inorg. Chem. 1980, 19, 3655–3663.

(19) Bradley, F. C.; Lindstedt, S.; Lipscomb, J. D.; Que, L., Jr.; Roe,

A. L.; Rundgren, M. J. Biol. Chem. 1986, 261, 11693–11696.

(20) Ramesh, K.; Mukherjee, R. J. Chem. Soc., Dalton Trans. 1992, 1, 83–

89.

~

(21) Lanznaster, M.; Neves, A.; Bortoluzzi, A. J.; Assumpc-ao, A. M. C.;

Vencato, I.; Machado, S. P.; Drechsel, S. M. Inorg. Chem. 2006, 45, 1005–

1011.

Article

Inorganic Chemistry, Vol. 49, No. 24, 2010 11423

measurements, are used to show (i) how the structural and

electronic properties of the series of [H₂OFe^III( μ-OH)Zn^II-

L-R]^2þcomplexes are affected by the substituent R on the

phenolate ring, (ii) how these properties change upon dissol-

ving the solid complexes, (iii) how the hydrolase-like activity

(using a model substrate and DNA) and the cytotoxic activity

(IC50) of 1-4 are affected by the electron donor/acceptor

nature of the substituent on the terminally bound phenolate.

Importantly, it is demonstrated for complexes 1-4 that after

monodentate binding of the substrate to the Zn^IIcenter, the

terminal Fe^III-bound hydroxide rather than the bridging

hydroxide acts as the initiating nucleophile in the hydrolysis

of a diester phosphate. Such a mechanism is in agreement

with that proposed for the monoesterase activity of

rkbPAP,^22-25thus demonstrating that the model system does

not only mimic the two chemically distinct geometric environ-

ments around the Fe^IIIand Zn^IIcenters in the enzyme, but also

shares functional similarities.^1,14

2-[N-bis-(2-pyridylmethyl)aminomethyl]-4-methyl-6-[N⁰-(2-pyri-

dylmethyl)(2-hydroxy-5-methylbenzyl)aminomethyl]phenol; H₂L-

CH₃). The ligand N-(2-hydroxy-5-methylbenzyl)(2-pyridylmethyl)-

amine (hbpa-CH₃) was obtained as follows: 2.7 g (20 mmol) of

2-(aminomethyl)pyridine was added to a methanolic solution con-

taining 2.2 g (20 mmol) of 2-hydroxy-5-methylbenzaldehyde under

stirring. After 1 h, 0.76 g (20 mmol) of sodium borohydride was

slowly added. The pH was then adjusted to 6.0 using 2.0 mol L^-1

HCl, and the solvent evaporated under reduced pressure. A 100 mL

portion of CH₂Cl₂was added to the remaining solid, the solution

was washed with a sodium bicarbonate saturated aqueous solution

(5 ꢀ 40 mL), dried over anhydrous sodium sulfate, and the solvent

removed under vacuum. A 3.4 g portion (15 mmol) of hbpa-CH₃

was obtained as a pale yellow oil. Yield 75%. IR (KBr) in cm^-1

:

ν(N-H) 3445; ν (C-H_arand C-H_aliph) 3013-2918; ν (CdN and

CdC) 1594-1435; ν (C-O) 1253; ν (C-N) 1148; ν (C-H_ar) 769.

¹H NMR, δ_H(200 MHz; CDCl₃), inppm: 2.15 (s, 3H); 3.84 (s, 2H);

3.87 (s, 2H); 5.97 (s, 1 H, NH); 6.64-6.70 (m, 3H); 6.87-6.94 (dt,

1H); 7.14-7.17 (dd, 1H); 7.56-7.64 (dt, 1H); 8.49-8.52 (d, 1H).

The ligand H₂L-CH₃was synthesized by adding to a 60 mL of

CH₂Cl₂solution containing 8.77 g of 2-[N-bis(2-pyridylmethyl)-

aminomethyl]-4-methyl-6-chloromethylphenol hydrochloride -

Experimental Section

bpmamcf HCl⁷(21.7 mmol), 4.95 g of hbpa-CH₃(21.7 mmol),

3

Materials and Methods. All starting materials were purchased

from Aldrich, Acros, or Merck. ¹H NMR spectra were recorded

with a Bruker-FT 200 MHz instrument at room temperature, and

the chemical shifts referenced to the residual solvent (CDCl₃) signal.

IR spectra were recorded with a Perkin-Elmer-FTIR 2000 spectro-

photometer (KBr pellet). Tetrabutylammonium hexafluoropho-

sphate (TBAPF₆) was recrystallized from ethanol, 2-pyridinecar-

boxyaldehyde was distilled under reduced pressure, and 2,

4-dinitrophenol was recrystallized from chloroform. The sub-

strate bis-(2,4-dinitrophenyl)phosphate (2,4-bdnpp) was synthe-

sized, characterized, and purified as described in the literature.²⁶

Elemental analyses were made in a CHNS analyzer from Carlo

Erba model E-1110. Electronic spectra were recorded with a

Perkin-Elmer Lambda-19 spectrometer in quartz cuvettes of

1 cm path-length. All electrochemical experiments (recorded in a

potentiostat-galvanostat PAR 273) requiring an inert atmosphere

were performed in a conventional three-electrode cell filled with

ultrapure argon. Data were collected using a typical three-elec-

trode electrochemical cell. The auxiliary electrode was a platinum

sheet. The reference electrode consisted of Ag/Ag^þ, and the

working electrode was a glassy carbon previously polished in an

aluminaslurrytoa0.3μm finish. TBAPF₆or NaCl was used as the

supporting electrolyte. The redox pair Fc^þ/Fc (or sodium

hexacyanoferrate) was used as an internal standard.²⁷

and 3.5 g of triethylamine (25 mmol) previously dissolved in

50 mL of CH₂Cl₂. After reflux for 36 h, the mixture was washed

with an aqueous sodium bicarbonate solution (5 ꢀ 40 mL), dried

over anhydrous Na₂SO₄, and the solvent removed under vacuum.

The yellow solid remaining was dissolved in acetone and after some

hours under refrigeration the pale yellow solid obtained was

collected by filtration, washed with cold acetone, and dried in air.

Yield: 10.6 g, 90%; mp 86-88 ꢀC; IR (KBr) cm^-1: ν(C-H_Arand

C-H_aliph) 3055-2824; ν(CdC) 1594-1434; ν(O-H) 1372;

1

ν(C-O_phenol) 1252-1229; ν (C-H_Ar) 757. H NMR (200 MHz;

CDCl₃): δ 2.14 (s, 6H); 3.79-3.67 (m, 12H); 6.67 (s, 1H); 6.74 (s,

2H); 6.86-6.83 (m, 2H); 7.10-7.04 (t, 3H); 7.19 (s, 1H); 7.31-7.27

(m, 3H); 7.56-7.49 (dt, 3H); 8.50-8.47 (dt, 3H).

2-[N-bis-(2-pyridylmethyl)aminomethyl]-4-methyl-6-[N⁰-(2-pyri-

dylmethyl)(2-hydroxy-5-bromobenzyl)aminomethyl]phenol; H₂L-Br).

The ligand N-(2-hydroxy-5-bromobenzyl)(2-pyridylmethyl)amine

(hbpa-Br) was obtained as follows: 2.01 g (10 mmol) of 2-hydro-

xy-5-bromobenzaldehyde was dissolved in 50 mL of toluene and

was added dropwise to 1.49 mL of 2-(aminomethyl)pyridine

(10 mmol) while stirring. The flask was equipped with a Dean-

Stark trap and after 1 h of reflux, 0.37 g (10 mmol) of sodium

borohydride was slowly added. The pH was then adjusted to 6.0

using 2.0mol L^-1HCl. The solution was washed with a saturated

sodium bicarbonate (3 ꢀ 50 mL), dried over Na₂SO₄, filtered off,

and the solvent evaporated under reduced pressure. A 2.93 g

portion (10 mmol) of hbpa-Br was obtained as a pale yellow oil.

Yield: 99%; IR (KBr) in cm^-1: ν(N-H) 3435; ν (C-H_arand

C-H_aliph) 3062-2851; ν (CdN and CdC) 1592-1477; ν (C-O)

1267; δ (C-H_ar) 763; ¹H NMR, δ_H(200 MHz; CDCl₃), in ppm:

3.91 (s, 4H); 6.74-7.67 (m, 6H); 8.58 (d, 1H).

Preparation of Ligands. A general synthetic route for the pre-

paration of the ligands H₂L-R (R = -CH₃, -H, -Br, -NO₂) is

shown in Scheme S1 in the Supporting Information. The ligands

2-[N-bis-(2-pyridylmethyl)aminomethyl]-4-methyl-6-[N⁰-(2-pyri-

dylmethyl)(2-hydroxybenzyl)aminomethyl]phenol (H₂L-H),⁷

and 2-[N-bis-(2-pyridylmethyl)aminomethyl]-4-methyl-6-[N⁰-(2-

pyridylmethyl)(2-hydroxy-5-nitrobenzyl)aminomethyl]phenol

(H₂L-NO₂)²⁸were synthesized as previously reported.

The ligand H₂L-Br was synthesized by adding 1.60 g of hbpa-Br

(5.46 mmol) and 2.1 mL of triethylamine (15 mmol) to a 50 mL

CH₂Cl₂solution containing 2.21 g of bpmamcf HCl (5.46

(22) It should be noted that under certain circumstances (i.e., depending

on the metal ion composition and type of substrate used in assays) PAPs may

employ an alternative mechanism in which the substrate forms an initial μ-

1,3 phosphate complex prior to hydrolysis, thus placing the μ-(hydr)oxo

bridge in an ideal position to act as the reaction initiating nucleophile.^2,11,12

(23) Cox, R. S.; Schenk, G.; Mitic, N.; Gahan, L. R.; Hengge, A. C. J. Am.

Chem. Soc. 2007, 129, 9550–9551.

(24) Schenk, G.; Gahan, L. R.; Carrington, L. E.; Mitic, N.; Valizadeh,

M.; Hamilton, S. E.; de Jersey, J.; Guddat, L. W. Proc. Natl. Acad. Sci. U.S.

A. 2005, 102, 273–278.

3

mmol).⁷After reflux for 72 h, the solution was cooled, washed

with aqueous sodium bicarbonate solution, dried over anhydrous

sodium sulfate, and the solvent removed under vacuum. The oil

remaining was dried in a vacuum desiccator (0.1 mmHg) for 6 h,

and 2.39 g of a yellow oil was obtained. Yield: 70%; IR (KBr)

cm^-1: ν(C-H_Arand C-H_aliph) 3056-2820; ν(CdC and CdN)

1592-1480; δ(O-H) 1368; ν(C-O_phenol) 1262; ν (C-H_Ar) 758;

¹H NMR (200 MHz; CDCl₃): δ 2.19 (s, 3H); 3.86 (s, 8H); 3.92 (s,

2H); 3.97 (s, 2H); 6.73-7.21 (m, 7H); 7.33-7.42 (m, 4H);

7.60-7.64 (t, 3H); 8.57 (d, 3H).

(25) Smoukov, S. K.; Quaroni, L.; Wang, X.; Doan, P. E.; Hoffman,

B. M.; Que, J. L. J. Am. Chem. Soc. 2002, 124, 2595–2603.

(26) Bunton, C. A.; Farber, S. J. J. Org. Chem. 1969, 34, 767–772.

Preparation of Complexes. Caution! Perchlorate salts of metal

complexes are potentially explosive and therefore should be

prepared in small quantities.

ꢀ

(27) Gagne, R. R.; Farber, S. J. J. Org. Chem. 1980, 19, 2854–2855.

(28) Carvalho, M. L.; Santhiago, M.; Peralta, R. A.; Neves, A.; Micke,

G. A.; Vieira, I. C. Talanta 2008, 77, 394–399.

11424 Inorganic Chemistry, Vol. 49, No. 24, 2010

Peralta et al.

[Fe^IIIZn^II(L-CH₃)( μ-OH)(H₂O)](ClO₄)₂2.5H₂O (1). To a

60 mL methanol solution of H₂L-CH₃(0.28 g, 0.5 mmol) was

consisting of a Jobin/Yvon 750S monochromator, calcite polar-

izer, a Hinds PEM90 photoelastic modulator, an Oxford In-

struments SpectroMag 7T superconducting magnet with an

Hamamatsu S5 photomultiplier. The total and differential

circularly polarized light from a 250 W halogen lamp at refer-

ence frequencies of the modulator and a New Focus 7051 optical

chopper were detected using Stanford SR830 lockin amplifiers,

allowing the simultaneous measurement of absorbance and

MCD. Samples were dissolved in a glassing solution of

MeOH:EtOH (1:2) to concentrations of about 1.5 mM, and

the spectra were acquired using 1 mm path length quartz cells.

Potentiometric Titration. The potentiometric studies were

carried out with a Micronal B375 pH meter fitted with blueglass

and calomel reference electrodes calibrated to read -log [H^þ]

directly, designated as pH. Doubly distilled water in the pre-

sence of KMnO₄and reagent grade ethanol were used to prepare

the ethanol/water (70/30, v/v) solutions. The electrode was

calibrated using the data obtained from a potentiometric titra-

tion of a known volume of a standard 0.100 M HCl solution with

a standard 0.100 M KOH solution. The ionic strength of the

HCl solution was maintained at 0.100 M by addition of KCl.

The measurements were carried out in a thermostatted cell

containing a solution of the complex (0.05 mol/50 mL) with

ionic strength adjusted to 0.100 M by addition of KCl, at 25.00

( 0.05 ꢀC. The experiments were performed under an argon flow

to eliminate the presence of atmospheric CO₂. The samples were

titrated by addition of fixed volumes of a standard CO₂-free

KOH solution (0.100 M). An ethanol/water (70/30 v/v) solution

was chosen as the medium for the potentiometric titrations

because of the low solubility of complexes in water. The pK_w

of the mixture ethanol/water 70/30 v/v containing 0.100 M KCl

was 14.72(2).³¹The deprotonation constants were calculated

using the BEST7 program³²and species distribution were made

using SPE and SPEPLOT programs.³²

3

added with stirring 0.19 g (0.5 mmol) of Zn(ClO₄)₂6H₂O. To

3

this solution was added dropwise a methanol solution (60 mL)

of Fe(ClO₄)₃9H₂O (0.26 g, 0.5 mmol) and 3 mmol of NaOH in

3

6.5 mL of water. The stirring was maintained for approximately

15 min and then NaClO₄(0.12 g, 1 mmol) was added. The solu-

tion was filtered off and left to stand. The resulting blue product

was collected and washed with water and diethyl ether (Yield:

0.26 g, 55%). Single crystals suitable for X-ray analysis were

obtained after recrystallization from methanol. Anal. Calcd for

C₃₅H₄₃Cl₂N₅O_14.5FeZn: C: 43.88; H: 4.53; N: 7.31. Found: C:

43.71; H: 4.27; N: 7.34%. IR (KBr), cm^-1: ν(OH) 3426;

ν(C-H_Arand C-H_aliph) 3039-2861; ν(CdN and CdC),

1610-1437; ν(C-O) 1276; ν(Cl-O) 1097; δ(C-H_Ar) 767.

[Fe^IIIZn^II(L-H)( μ-OH)(H₂O)](ClO₄)₂H₂O (2). (2) was pre-

3

pared as described previously.¹⁴

[Fe^IIIZn^II(L-Br)( μ-OH)(H₂O)][ClO₄]₂H₂O (3). To a 60 mL

3

methanol solution of H₂L-Br (0.31 g, 0.5 mmol; 624.58 g mol^-1

was added with stirring Zn(ClO₄)₂6H₂O (0.19 g, 0.5 mmol) and

)

3

3 mmol of NaOH in 6.5 mL of water. To this solution was added

dropwise a methanol solution (60 mL) of Fe(ClO₄)₃9H₂O (0.26 g,

3

0.5 mmol). The solution was heated and stirred for 15 min.

NaClO₄, (0.12 g, 1 mmol) was added, and the solution left to sit

at room temperature. After about 15 min a purple precipitate

formed which was filtered and washed with water and cold ether

(0.30 g, 55% yield). After recrystallization from methanol, micro-

crystals were obtained. Anal. Calcd for FeZnC₃₄H₃₅BrN₅O₅-

(ClO₄)₂H₂O, C, 41.01; H, 3.75; N, 7.03%. Found: C, 40.95; H,

3

3.95; N, 7.23%. IR (KBr), cm^-1: ν(OH) 3395; ν(C-H_Arand

C-H_aliph) 3058-2880; ν(CdN and CdC) 1609-1437; ν(C-O)

1282; ν(Cl-O) 1101; δ(C-H_Ar).

[Fe^IIIZn^II(L-NO₂)( μ-OH)(H₂O)][ClO₄]₂H₂O (4). To a 60 mL

3

methanol solution of H₂L-NO₂(0.30 g, 0.5 mmol; 590.70 g mol^-1

was added with stirring and heating Zn(ClO₄)₂6H₂O (0.19 g,

)

3

Single-Crystal X-ray Structure Determinations. Crystallo-

graphic analyses of complexes 1 and 4 were carried out with

an Enraf-Nonius CAD4 diffractometer with graphite-mono-

3

0.5 mmol). To this solution was added dropwise a methanol

solution (60 mL) of Fe(ClO₄)₃9H₂O (0.26 g, 0.5 mmol) and 3 mmol

3

˚

chromated Mo-K_Rradiation (λ = 0.71069 A), at room tem-

of NaOH in 6.5 mL of water. The solution was heated and

stirred for 15 min. NaClO₄, (0.12 g, 1 mmol) was added, and the

solution left to sit at room temperature. After about 15 min a

red precipitate formed which was filtered and washed with

water and ether (0.29 g, 60% yield). Single crystals suitable for

X-ray structural analysis were obtained after recrystallization

perature. Cell parameters were determined from 25 carefully

centered reflections using a standard procedure.³³Intensities

were collected using the ω-2θ scan technique. All data were

corrected for Lorentz and polarization effects.³⁴An empirical

absorption correction based on the azimuthal scans of 7 appro-

priate reflections was also applied to the collected intensities

using the PLATON program.³⁵The structures were solved by

direct methods and refined by full-matrix least-squares methods

using the SHELXS97³⁶and SHELXL97³⁷programs, respec-

tively. H atoms attached to C atoms were placed at their

idealized positions, with C-H distances and U_eqvalues taken

from the default settings of the refinement program. Complex 1:

an irregular block cut off from a big dark red crystal was selected

and fixed at the end of a glass fiber for X-ray analysis. All non-

hydrogen atoms were refined with anisotropic displacement

parameters, except for oxygen atoms of the perchlorate coun-

terions and water solvate, which were refined isotropically. H

atoms of the water molecule of crystallization (O2W) could not

from methanol. Anal. Calcd for FeZnC₃₄H₃₅N₆O₆(ClO₄)₂

H₂O, C, 42.45; H, 3.89; N, 8.74%. Found: C, 41.90; H, 4.04;

3

N, 8.53%. IR (KBr), cm^-1: ν(OH) 3400; ν(C-H_Arand

C-H_aliph

) 3063-2864; ν(CdN and CdC) 1607-1440;

ν(N-O) 1292; ν(C-O) 1292; ν(Cl-O) 1091; δ(C-H_Ar) 770.

[Fe^IIIZn^II(L-H)( μ-OAc)₂][ClO₄]₂H₂O (5). (5) was prepared

3

as described previously.⁹

Electron Paramagnetic Resonance (EPR) Spectroscopy. EPR

spectra were collected with a Bruker Elexsys E500 continuous

wave spectrometer fitted with an X-band Super HighQ cavity. A

Bruker ER 035 M Gaussmeter and an EIP 548B microwave

frequency counter were used for calibration of the magnetic field

and microwave frequency. Low temperatures (2 K) were ob-

tained using an Oxford ESR910 flow through cryostat in

conjunction with an Oxford instruments ITC-4 temperature

controller. Bruker’s Xepr (v2.2b.38) software was used for

spectrometer tuning, signal averaging, and plotting. Computer simu-

lations of the collected data were performed using the XSophe-

Sophe-XeprView computer simulation software suite (v1.1.4).²⁹

Magnetic Circular Dichoism (MCD) Spectroscopy. Spectra

were acquired using a Lastek Pty. Ltd. developed instrument³⁰

(31) Schwingel, E. W.; Arend, K.; Zarling, J.; Neves, A.; Szpoganicz, B. J.

Braz. Chem. Soc. 1996, 7, 31–37.

(32) Martell, A. E.; Montekaitis, R. J. Determination and use of stability

constants, 2nd ed.; VCD: New York, 1992.

(33) CAD4 EXPRESS, ver 5.1/1.2; Enraf-Nonius: Delft, The Netherlands,

1994.

(34) Spek, A. L. HELENA: CAD-4 data reduction program; University of

Utrecht: The Netherlands, 1996.

(35) Spek, A. L. PLATON: molecular geometry and plotting program;

University of Utrecht: The Netherlands, 1997.

(29) Hanson, G.; Gates, K. E.; Noble, C. J.; Griffin, M.; Mitchell, A.;

Benson, S. J., Inorg. Biochem. 2004, 98, 903.

(30) Riley, M. J.; Krausz, E. R.; Stanco, A. J. Biol. Inorg. Chem. 2003, 96,

217.

(36) Sheldrick, G. M. SHELXS97: program for the solution of crystal

€

structures; University of Gottingen: Gottingen, Germany, 1990.

(37) Sheldrick, G. M. SHELXL97: program for the refinement of crystal

€

structures; University of Gottingen: Gottingen, Germany, 1997.

Article

Inorganic Chemistry, Vol. 49, No. 24, 2010 11425

be located. H atoms of the coordinated water and hydroxo

bridge were found from a Fourier difference map and treated

with a riding model. The oxygen atoms of the perchlorate

counterions were found to be disordered over two alternative

positions, and their refined site occupancies were 0.55(4) and

0.45(4). Complex 4: the crystallographic data were acquired

from a selected irregular block cut off from a large dark red

crystal taken from the crystalline sample. The crystals of this

complex showed low scattering power. From 6251 poorly

diffracted reflections, only 2572 reflections were observed with

I > 2σ. However, non-H atoms could be refined with aniso-

tropic displacement parameters, except for oxygen atoms of the

perchlorate counterions and water solvate. Hydrogen atoms of

the water molecules and hydroxo bridge were not found in the

Fourier map. Further relevant crystallographic data for com-

plexes 1 and 4 are summarized in Supporting Information,

Table S1.

k_H/k_D.¹⁴The reactions were monitored under a 125-fold excess of

substrate at 400 nm for complexes 1, 3, and 4.

Preparation of Plasmid DNA. Plasmid DNA pBSK II

(Stratagene) was obtained and purified according to standard

techniques.^40,41Briefly, Escherichia coli DH5alpha cells were

transformed with pBSK II. One colony was incubated overnight

in 5 mL of Lysogeny broth (LB) medium medium^42,43supple-

mented with 0.1 mg/mL of ampicillin at 37 ꢀC with aeration; and

then 1 mL was inoculated in 300 mL of LB medium supple-

mented with 0.1 mg/mL of ampicillin and incubated for 8 h at

37 ꢀC with aeration. From these cultures plasmid DNA was

purified using Qiagen Plasmid Maxi Prep Kit. DNA was

quantified by spectrophotometry using the ratio of the absor-

bance at 260 and 280 nm.⁴⁰The plasmid preparation was

analyzed by gel electrophoresis to ensure that it consisted of

approximately 90% intact supercoiled plasmid DNA.

DNA Cleavage Activity Assays. The ability of 1-4 to cleave

DNA was examined by following the conversion of supercoiled

plasmid DNA (F I) to open circular DNA (F II) or linear DNA

(F III) using agarose gel electrophoresis to separate the cleavage

products.^41,44

Theoretical Calculations. The geometry optimization of the

four cationic complexes 1-4 was performed using the B3LYP

hybrid density functional theory in combination with the 6-31G(d,

p) and LACVP* basis sets using Gaussian03.^38,39To ensure that

the geometry was minimized, a frequency calculation was per-

formed (B3LYP, LANL2DZ), finding only one weak, low energy

imaginary frequency corresponding to a minor rotatory vibration

of the methyl group.

In general, 400 ng of pBSK II (∼30 mmol L^-1bp) buffered

with PIPES (10 mmol L^-1, pH 6.5) were treated with 1-4 in

CH₃CN (25% reaction volume) for different time reactions at

50 ꢀC. Thereafter, each reaction was quenched adding 5 μL of a

loading buffer solution (0.25% bromophenol blue, 50% glycer-

ol and 250 mmol L^-1ethylenediamine tetraacetic acid (EDTA)

at pH 8.0) and then subjected to electrophoresis on a 0.8%

agarose gel containing 0.3 mg mL^-1of ethidium bromide in 0.5x

TBE buffer (44.5 mmol L^-1TRIS, 44.5 mmol L^-1boric acid and

1 mmol L^-1EDTA) at 90 V for approximately 1.5 h. The

resulting gels were visualized and digitized using a DigiDoc-It

gel documentation system (UVP, U.S.A.). The proportion of

plasmid DNA in each band was quantified using KODAK

Molecular Imaging Software 5.0 (Carestream Health, U.S.A.).

The portion of supercoiled DNA (F I) was corrected by a factor

of 1.47,⁴⁵since the ability of ethidium bromide to intercalate

into this DNA topoisomeric form is decreased relative to open

circular and linear DNA. To elucidate whether the mechanism

of DNA cleavage performed by complexes 1-4 is hydrolytic or

oxidative, different inhibitors of reactive oxygen species (ROS)

were added to the DNA treatments prior to the complexes.

These inhibitors were dimethyl sulfoxide (DMSO, 10%), KI

(500 mmol L^-1), superoxide dismutase (SOD, 15 Units), and

NaN₃(500 mmol L^-1). In addition, assays in the presence of the

DNA minor groove binder distamycin⁴⁶(50 μmol L^-1) and the

DNA major groove binder methyl green⁴⁷(50 μmol L^-1) were

also performed to clarify the DNA groove binding preference of

the complex-DNA interaction. The plasmid DNA was pre-

treated with distamycin and methyl green for 30 min and then

treated with the complex as described above. The kinetics of

plasmid DNA cleavage performed by 1-4 were evaluated

following the loss of supercoiled DNA fraction along the

Kinetic Assays for Complexes. Phosphodiesterase-like activity

was determined for complexes 1-4 by measuring hydrolysis of the

substrate bis-(2,4-dinitrophenyl)phosphate (2,4-bdnpp) in an UV-vis

Varian Cary 50 BIO spectrometer. The data were accompanied by

monitoring the release of 2,4-dinitrophenolate at 400 nm. The

rates were measured in real time at various pH values, and the

extinction coefficients of the product 2,4-dinitrophenolate were

measured at different pH’s under the same experimental condi-

tions of the kinetic measurements at different pH’s/(λ = 400 nm/ε,

M^-1cm^-1) (3.5/2125; 4.0/3410; 4.5/7180; 5.0/10080; 5.5/11400;

6.0/12,000; 6.5-10.0/12,100). Reactions were monitored to less

than 5% of conversion of substrate to product, and the data were

treated by the initial rate method. The effect of pH on the rate of

hydrolysis of 2,4-bdnpp between pH 3.6 and 10.25 (buffers - MES;

HEPES and CHES 0.05 mol L^-1) was investigated by using fixed

concentrations of substrate ([S]_final= 5 ꢀ 10^-3mol L^-1), and

complex ([C]_final= 4 ꢀ 10^-5mol L^-1for complexes 1 and 3 and

4 ꢀ 10^-5mol L^-1for complex 4), with excess of substrate at 25 ꢀC

(I = 0.05 mol L^-1with LiClO₄). Substrate dependence ([S]_final

=

5 ꢀ 10^-4-7 ꢀ 10^-3mol L^-1) of the catalytic rate was measured at

the optimum pH for each complex. The data were treated using the

Michaelis-Menten non-linear regression (V₀vs [bdnpp]). Isoto-

pic effects of deuterium on the hydrolysis of 2,4-bdnpp promoted

by complexes 1, 3, and 4 were investigated monitoring parallel

reactions, where the buffer solutions MES pH and pD 6.50 were

prepared in H₂O and D₂O, respectively to determine the relation

(38) Becke, A. D. J. Chem. Phys. 1993, 98, 5648–5652.

(39) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb,

M. A.; Cheeseman, J. R.; J. A. Montgomery, J.; Vreven, T.; Kudin, K. N.;

Burant, J. C.; Millam, J. M.; Iyengar, S. S.; Tomasi, J.; Barone, V.;

Mennucci, B.; Cossi, M.; Scalmani, G.; Rega, N.; Petersson, G. A.; Nakatsuji,

H.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.;

Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Klene, M.; Li, X.; Knox, J. E.;

Hratchian, H. P.; Cross, J. B.; Adamo, C.; Jaramillo, J.; Gomperts, R.;

Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.;

Ochterski, J. W.; Ayala, P. Y.; Morokuma, K.; Voth, G. A.; Salvador, P.;

Dannenberg, J. J.; Zakrzewski, V. G.; Dapprich, S.; Daniels, A. D.; Strain,

M. C.; Farkas, O.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.;

Foresman, J. B.; Ortiz, J. V.; Cui, Q.; Baboul, A. G.; Clifford, S.; Cioslowski,

J.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; ; Komaromi, I.;

Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.;

Nanayakkara, A.; ; Challacombe, M.; Gill, P. M. W.; Johnson, B.; Chen,

W.; Wong, M. W.; Gonzalez, C.; Pople, J. A. Gaussian 03, Revision C.02;

Gaussian Inc: Wallingford, CT, 2004.

(40) Ausubel, F. M.; Brent, R.; Kingston, R. E.; Moore, D. D.; Seidman,

J. G.; Smith, J. A.; Struhl, K. In Short protocols in molecular biology: A

compendium of methods from current protocols in molecular biology; Wiley:

New York, 1999.

(41) Scarpellini, M.; Neves, A.; Horner, R.; Bortoluzzi, A. J.; Szpoganicz,

B.; Zucco, C.; Silva, R. A. N.; Drago, V.; Mangrich, A.; Ortiz, W. A.; Passos,

W. A. C.; de Oliveira, M. C. B.; Terenzi, H. Inorg. Chem. 2003, 42, 8353–

8365.

(42) Bertani, G. J. Bacteriol. 2004, 186, 595–600.

(43) Bertani, G. J. Bacteriol. 1951, 62, 293–300.

(44) Neves, A.; Terenzi, H.; Horner, R.; Horn, A.; Szpoganicz, B.; Sugai,

J. Inorg. Chem. Commun. 2001, 388–439.

(45) Bernadou, J.; Pratviel, G.; Bennis, F.; Girardet, M.; Meunier, B.

Biochemistry 1989, 28, 7268–7275.

(46) Hiraku, Y.; Oikawa, S.; Kawanishi, S. Nucleic Acids Res. Suppl.

2002, 61, 95–96.

(47) Kim, S. K.; Norden, B. FEBS Lett. 1993, 315, 61–64.

11426 Inorganic Chemistry, Vol. 49, No. 24, 2010

Peralta et al.

Table 1. Comparison of Theoretical and Crystallographic Structural Data^a

-CH₃(1)

-H (2)

-Br (3)

-NO₂(4)

exp

calc

exp

calc

exp

Calc

Fe-Zn

Fe-O1

Fe-O2

Fe-O3

Fe-O4

Fe-N2

Fe-N4

Zn-O1

Zn-O2

Zn-N1

Zn-N3

Zn-N5

3.0862(18)

2.053(3)

1.920(7)

2.112(3)

1.893(3)

2.137(8)

2.186(3)

2.101(3)

1.969(6)

2.152(7)

2.073(8)

2.056(8)

3.136

2.077

2.002

2.206

1.891

2.153

2.230

2.066

2.008

2.216

2.128

2.143

3.041

2.028

1.934

2.054

1.923

2.136

2.178

2.059

1.966

2.141

2.067

2.051

3.135

2.063

1.998

2.204

1.887

2.145

2.227

2.073

2.009

2.212

2.142

2.126

3.135

2.064

1.998

2.204

1.896

2.149

2.230

2.074

2.010

2.213

2.126

2.141

3.067(2)

2.031(8)

1.932(8)

2.107(9)

1.910(8)

2.117(10)

2.189(9)

2.103(8)

1.981(9)

2.135(10)

2.087(11)

2.038(11)

3.130

2.033

1.990

2.198

1.900

2.137

2.225

2.093

2.014

2.205

2.122

2.137

Fe-O1-Zn

95.96(13)

105.0(3)

125.9(4)

-151.81

-45.07

128.5(4)

12.77

89.4(2)

124.4(2)

110.8(2)

77.9(2)

166.0(3)

101.4(3)

110.2(3)

80.8(3)

79.6(3)

120.6(3)

0.69

98.40

96.14

98.55

98.51

95.8(3)

103.2(4)

125.1(8)

-152.22

-47.49

134.5(7)

9.46

89.1(3)

125.0(4)

111.1(4)

78.4(3)

166.4(3)

102.3(4)

109.4(4)

80.3(4)

80.0(4)

119.6(4)

0.69

98.68

Fe-O2-Zn

102.87

131.92

-165.25

-31.38

131.40

22.35

102.47

123.43

-152.64

-48.36

134.03

6.70

102.89

131.51

-163.48

-34.28

131.66

21.80

102.91

131.41

-163.41

-33.71

131.73

21.46

102.82

129.75

-159.35

-39.58

132.28

19.96

Fe-O4-C41

Zn-Fe-O4-C41

Fe-O4-C41-C46

Fe-O1-C14

Fe-O1-C14-C13

O1-Zn-N1

O1-Zn-N3

O1-Zn-N5

O1-Zn-O2

O2-Zn-N1

O2-Zn-N3

O2-Zn-N5

N1-Zn-N3

N1-Zn-N5

N3-Zn-N5

τ = (R-β)/60

90.02

90.46

90.15

90.27

130.20

112.13

79.36

167.53

101.94

110.84

79.86

124.49

109.47

78.45

167.31

100.18

108.63

80.93

80.73

122.67

0.71

130.02

112.22

78.94

167.22

101.86

110.94

79.99

129.84

112.36

78.95

167.42

102.08

110.63

80.02

129.31

112.51

78.13

166.73

101.96

110.74

80.32

79.12

113.44

79.24

113.69

79.22

113.72

79.51

114.36

a

ꢀ

˚

Bond lengths are in A, and bond angles are in deg. The atomic numbering follows that given in Figure 2.

reaction time under pseudo-first-order conditions. The appar-

ent plasmid DNA cleavage rates (k_obs) were obtained from the

plot of ln[Supercoiled DNA (%)] versus time. The reaction

conditions were as the same as described above.

structures of the compounds 2 and 5 have been previously

published^9,14but for comparison are shown in Figure 2

together with the structures of 1 and 4. Because of low

intensity and quality of the diffraction data it was not

possible to completely refine the structure of 4. However,

the data are considered to be of sufficient quality to permit

structural details to be elucidated. Unfortunately, attempts

to obtain suitable crystals of complex 3 were unsuccessful.

We have used the electronic structure calculations to

supplement the limited crystallographic data.

Cell Lines and Cultures. K562 is a cell line of chronic myelo-

genous leukemia established from pleural effusion of a 53 year-old

female in terminal blast crisis, which was purchased from the Rio

de Janeiro Cell Bank (number CR083 of the RJCB collection).

The GLC4 cell line was derived from pleural effusion of a patient

with small cell lung carcinoma. The cell lines were cultured in

suspension inRPMI 1640 supplemented with10% fetal calf serum

in a humidified atmosphere with 5% CO₂at 37 ꢀC. Both cultures

were initiated at 10⁵cells/mL and grew exponentially to about

8 ꢀ 10⁵cells/mL in 3 days.

The molecular structures of 1 and 4 (Figure 2) show that

in the dinuclear [L-R(OH₂)Fe( μ-OH)Zn]^2þunit the Fe^III

ion is facially coordinated by the hard tridentate pendant

arm of the ligands L-CH₃^2-(1) or L-NO₂^2-(4) containing

the amine (N4) and pyridine nitrogen (N2) and the pheno-

late oxygen (O4) atoms, while Zn^IIis coordinated by the

soft side of the ligand through the amine (N1) and pyridine

(N3, N5) nitrogen atoms. The bridging phenolate oxygen

O1, the bridging hydroxo group (O2), and a terminal water

molecule (O3) complete the octahedral N₂O₄coordination

of Fe^III, while the distorted trigonal bipyramid of Zn^II(t =

0.69 for 1 and 4)⁴⁸is complemented by the bridging

phenolate (O1) and the hydroxo (O2) oxygen atoms.

Complexes 1 and 4 are isostructural with the previously

For cytotoxicity measurements, 1 ꢀ 10⁵cells/mL were incu-

bated for 72 h in the absence and presence of various concentra-

tions of tested compounds. Cell viability was checked by Trypan

Blue exclusion. Cell number was determined by Coulter counter

analysis. The sensitivity to the drug was evaluated by the con-

centration required to inhibit cell growth by 50%, the IC₅₀. The

mean IC₅₀(SD was determined in three independent experiments

each performed in duplicate.

Results and Discussion

Geometric and Electronic Structures of the Complexes.

(a). Crystallography. The H₂L-H ligand has been used to

generate a large number of heterovalent dinuclear com-

plexes with different metals ions together with auxiliary

ligands to complete their coordination spheres.^6-14Se-

lected crystallographic structural data are given in the

Supporting Information, Table S1 for complexes 1 and 4.

Bond lengths and angles are given in Table 1. The complete

reportedcomplex 2.¹⁴The Fe-μ-OH (1.920 A for 1, 1.934 A

˚

for 2, and 1.932 A for 4, respectively) and Zn-μ-OH (1.969 A

˚

for 1, 1.966 A for 2, and 1.981 A for 4, respectively) distances

˚

in 1, 2, and 4 are quite similar, which are somewhat shorter

(48) Addison, A. W.; Rao, T. N.; Reedijk, J.; van Rijn, J.; Verschoor,

G. C. J. Chem. Soc., Dalton Trans. 1984, 1349–1356.

Article

Inorganic Chemistry, Vol. 49, No. 24, 2010 11427

Figure 2. Crystal structures of (a) 1, (b) 2, (c) 4, and (d) 5.⁹

than the corresponding metal-μ-OH distances in rkbPAP,

the bridging phenolate oxygen (O1). In the μ-OAc com-

plex (5) the Fe^IIIion is coordinated by the tertiary amine,

pyridine and phenolate groups of the L-H^2-ligand (facial

arrangement) and the two bridging acetates to complete a

cis-N₂O₄coordination. The Zn^IIion of the μ-OAc com-

plex is also six coordinate, with a fac-N₃O₃coordination.

Figure 3 shows the structural changes that accompany the

change of the bridging acetates to a bridging hydroxide.

Figure 3a shows 5 viewed in the directions perpendicular

(upper) and parallel (lower) to a pseudo C₂axis of the

dimer that bisects the Fe-O1-Zn bridging angle, while

Figure 3b shows the analogous views of complex 2.

In complex 2 the Fe^IIIion retains the cis-N₂O₄coordi-

nation, with a bridging hydroxide and a terminally bound

water replacing the bridging acetates (Figure 3b). The

1

which have been modeled between 2.0 and 2.2 A. Further-

˚

II

more, the Fe^IIIZn distances (3.086 A for 1, 3.041 A for

˚

3 3 3

14

2 , and 3.067 A for 4) are shorter but comparable to the

˚

3.20 A found in rkbPAP. Importantly, the structures of 1, 2,

and 4 reveal the presence of a terminal Fe-bound water

˚

molecule (Fe-O3 = 2.112 A, 2.054 A, and 2.107 A,

respectively), which occupies a position equivalent to that

of one of the proposed nucleophiles in rkbPAP.^1,23

Although the distinct substituents (-CH₃, -H, -NO₂)

at the terminal Fe^III-bound phenolate do not cause any

significant effect in the arrangement of the donor atoms

around the metal centers and in the bond lengths and

angles within this series of complexes, the same is not true

regarding the physicochemical properties and reactivity

of the complexes, as will be discussed below.

It is also of interest to compare the structure of the

complex [H₂OFe^III( μ-OH)Zn^II(L-H)]^2þ(2)¹⁴with that

of [Fe^IIIZn^II( μ-OAc)₂(L-H)]^þ(5).⁹The complexes 2 and 5

are shown in Figure 3, the latter having diacetate bridges

in a bidentate mode and the former a hydroxo bridge,

with a terminal H₂O completing the coordination of the

Fe^IIIion. In both cases the Fe^IIIand Zn^IIions are bridged

by the phenolate oxygen of the L-H^2-ligand (O1 in

Figure 2), and the Fe^IIIion has the oxygen of the terminal

phenolate (O4; Figure 2) bound in the position trans to

Zn^IIion in 2, however, has a 5-fold N₃O₂coordination.

II

The Fe^III-Zn distances (3.040 A for 2 and 3.491 A for 5)

˚

bracket the value of 3.1 A measured in rkbPAP in the

presence of a sulfate ion that has been assumed to mimic

the initial substrate binding mode in the outer coordination

sphere,^4,5and 3.5 A in the presence of a metal ion bridging

˚

1

III

˚

phosphate. The Fe -O1 bond lengths are 2.028 A (2) and

III

˚

2.006 A (5), while the Fe -O4 bonds to the terminal

phenolate trans to O1 are considerably shorter (1.923 A

˚

(2) and 1.890 A (5)). As shown in Figure 3b complex 2 does

not have the pseudo-C₂symmetry seen in 5. The change can

11428 Inorganic Chemistry, Vol. 49, No. 24, 2010

Peralta et al.

Figure 3. Comparison of the structures of 5 ( μ-OAc)₂(a, top/bottom)

and 2 ( μ-OH) (b top/bottom) perpendicular and parallel to the pseudo C₂

axisofthe dimer thatbisects the Fe-O1-Febridgingangle. Inadditionto

the μ-phenoxo bridging ligand, 5 and 2 also contain ( μ-OAc)₂and μ-OH

bridging ligands, respectively. Hydrogen atoms have been deleted for

clarity. The arrows in (a) bottom indicates the rotation of the octahedron

on going from 5 to 2 (see text). Note the loss of the pseudo-C₂symmetry in

2 (b bottom) and the steric crowding of the aromatic groups.

Figure 4. (a) Orientation of the molecular axes and aromatic planes for

the Fe^IIIcenter in 2 and (b) the relevant transitions and the energy

expressions for the t_2gorbitals, having an energy ∼20,000 cm^-1. Energies

for the e_gorbitals is ∼30,000 cm^-1

.

be viewed as the Fe^IIIoctahedron rotating about the

R-phenolate-Fe-phenolate axis, such that the coordinate

position of one of the acetate ligands moves to the bridging

μ-OH position as indicated by the arrows in Figure 3. This

rotation causes the plane of the bridging phenolate to rotate

to be almost within the Fe-O1-Zn plane, while in complex

5 it forms a 45ꢀ angle with that plane (compare bottom

Figures 3a and 3b). This dramatic change also results in the

tertiary amine rotating toward this plane with a concomi-

tant shift of the attached pyridyl groups. The Zn coordina-

tion sphere undergoes a rotation in the opposite sense so

that the coordinate positions trans to both the tertiary

amines are directed to bind to the bridging OH^-in 2. The

Fe-O1-Zn angle decreases from 116.2ꢀ to 96.1ꢀ, and the

Fe-Zn distance decreases from 3.490 A to 3.040 A. The

vacant position of the Fe^IIIcoordination sphere is occupied

by a water molecule while that of the Zn^IIion is accom-

modated by a distortion toward a five coordinate trigonal

bipyramidal structure. The complex [Fe^IIIZn^II( μ-OH)-

(H₂O)L-H](ClO₄)₂is the first structural model for PAPs

with only one terminal phenolate, a μ-hydroxo, and a

molecule of water coordinated to the Fe^III.¹⁴

(b). Electronic Structure Calculations. The results for

the geometry optimization of the complexes 1-4 are given

in Table 1. A comparison between the available crystal-

lographic (1, 2, 4) and calculated structural data sets shows

good agreement. Small differences between the crystal data

and the theoretical values are observed, and these can be

assigned to the fact that the X-ray structures were measured

in a compacted crystalline form whereas the calculations

were performed in a vacuum, where crystal packing forces

are absent.

symmetry of Fe^IIIwill remove the otherwise degenerate xz,

yz orbitals. As discussed below, the aromatic planes of the

phenol and pyridine groups orient the anisotropic π bond-

ing of the ligating atoms. The energy of the d-orbitals will be

affected by interactions mainly with the out-of-plane

π-bonding orbitals of the aromatic groups, rather than

with the in-plane orbitals, which will be lower in energy.

In terms of the angular overlap model (AOM), it is the

antibonding interaction with the donor ligand orbitals that

raises the relative energy of the d-orbitals. From Figure 4a,

for the O1 (phenolate) ligand, the out-of-plane p orbital will

be approximately in the xz plane, the N2 (pyridine) p

orbital is approximately in the xz plane, and for the O4

(terminal phenolate) ligand the π bonding direction is

defined by the torsional angles given in Table 1. The

orientation of the O4 phenolate results in the out-of-plane

(p_op) and in-plane(p_ip) p orbitals being able to participate in

both σ- and π- bonding. From the location of the hydrogen

atoms of the OH-/H₂O ligands, the O2 (hydroxo bridge)

and O3 (water) atoms are expected to mainly undergo

π-bonding in the xy plane.

˚

The angle formed by the terminal phenolate and the Fe^III

is 123.43ꢀ, 125.9ꢀ, and 125.1ꢀ (Fe-O4-C41, Table 1) for

compounds 1, 2, and 4, respectively, while the torsional

angle of the phenolate plane with the Fe-O-C plane is

45.1ꢀ, 48.4ꢀ, and47.5ꢀ (Fe-O4-C41-C46; Table 1). These

valuescomparetothevalues136ꢀ and 39ꢀ for the analogous

angles in the tyrosine-Fe^IIIbond in the Fe^IIIZn^II

rkbPAP.¹⁶The close to 45ꢀ torsional angle results in the

p_opand p_iporbitals having mixed σ/π bonding character

and the π bonding interactions of O4 are approximately

equal with both xz and yz orbitals.

In six coordinate complexes, the ligand σ-bonding

interactions are with the e_g(x²-y², z²) orbitals only and

the π-bonding interactions affects the t_2g(xy, xz, yz)

orbitals only. Using the orientations discussed above,

and the approximation that the saturated N4 does not

undergo π-bonding, the relative energy expressions of the

t_2gorbitals in terms of the AOM π parameters are given in

Figure 4b.

Figure 4 shows a local coordinate system that we will use

to describe the geometric environment of the Fe^IIIion and

the effect on the energy of the d-orbitals. The z axis is

defined along the direction of the bridging and terminal

phenolate ions, whose short bond lengths impart a tetra-

gonally compressed geometry. The Fe^IIIZn^II( μ-OH)(O1)

core defines the yz plane, while the terminal water (O3) is

approximately along the x axis. The lower than tetragonal

Article

Inorganic Chemistry, Vol. 49, No. 24, 2010 11429

Figure 5. Illustrative absorption (top) and MCD (bottom) spectra of complexes 1 (a) and 2 (b) in a 1:1 methanol/ethanol at a temperature of 2 K and a

magnetic field of 7 T.

Table 2. Energy of HOMO, Orbital Parameters, Electronic Absorption Spectral Analysis, Square Voltammetric Data, and Zero Field Splittings for Complexes 1-4

E_1/2(mV vs NHE)

D cm^-1(E/D)

complex

σ_p

E_HOMOeV

λ

_maxnm (ε mol^-1cm^{-1 a}

)

pH 3.5

pH 6.5

EPR

MCD^b

1

2

3

4

-0.14

0

0.26

0.81

-10.278

-10.485

-10.300

-10.948

547(2195); 318(sh)

544(2800)

527 (1023)

57

60

63

70

-197

-180

-160

-100

-1.0(0.18)

1.8(0.06)

-0.7

0.4

500(4047); 353(20700)

4.0(0.065)

0.8

^aSolvent CH₃CN. ^b(E/D) values not fitted, taken from EPR.

In the low temperature absorption and MCD spectra of 1

and 2 (Figure 5) there are two groups of features at around

20,000 and 30,000 cm^-1that correspond to LMCT transi-

tions involving the charge transfer from a π orbital on the

terminal phenolate to the t_2gand e_gsets of the unfilled d

orbitals of Fe^III, respectively. These transitions are a char-

acteristic feature of PAPs and are responsible for these

enzymes’ purple color. The nature of the LMCT transitions

in PAPs¹⁶and in the model [Fe^IIIFe^II( μ-OAc)₂L-H]^þ

complex have been established by MCD¹⁷and resonance

Raman spectroscopy.⁴⁴This leads to four expected transi-

tions within the 20,000 cm^-1band and two transitions

within the 30,000 cm^-1band in the phenolate-Fe^IIILMCT

spectra as indicated in Figure 4b (only five of them are

Gaussian resolved in Figure 5). The electronic spectra of

complexes 1-4 in CH₃CN show the presence of two

absorption maxima in the region 300-600 nm and the

spectral data are summarized in Table 2. The lower and

higher energy bands correspond to the pπ f dπ^*and pπ f

dσ phenolate to Fe^IIILMCT transitions discussed above.

(c). Zero Field Splitting (ZFS) Parameters. The energies

of the d-orbitals in the Fe^IIId⁵system are also important for

determining the sign and magnitude of the zero-field splitting

(ZFS) parameters obtained from the EPR spectra. There are

examples of Fe^IIIwith N/O donor sets that result in both

high and low spin Fe^IIIdepending on the substituents and

steric effects.^49,50This implies that these types of ligands are

near the spin crossover region. While the present complexes

1-4 are all in the high spin configuration, there will be a

close-lying quartet ⁴T_1gexcited state, and spin-orbit mixing

is expected to generate relatively large ZFS values. The ZFS

values of the ⁶A₁ground state are determined by spin-orbit

induced mixture with the excited ⁴T_1gstate when it is split by

an orthorhombic low symmetry ligand field. From second

order perturbation theory one finds⁵¹

"

k_y²

ξ²2k²_z

4

D ¼

f T_1zÞ -

f T_1yÞ

6

10

Eð A₁

#

k_x²

4

-

f T_1xÞ

ð1Þ

6

Eð A₁

"

#

k_y²

ξ²

10

k_x²

4

E ¼

f T_1yÞ -

f T_1xÞ

6

Eð A₁

ꢁ

(49) Boca, R. Struct. Bonding (Berlin) 2006, 117, 1–264.

(50) Zang, Y.; Kim, J.; Dong, Y.; Wilkenson, E. C.; Appelman, E. H.;

Que, J. L. J. Am. Chem. Soc. 1997, 119, 4197–4205.

(51) Abragam, A.; Bleaney, B. Electron Paramagnetic Resonance of

Transition Ions; Dover: New York, 1986.

11430 Inorganic Chemistry, Vol. 49, No. 24, 2010

Peralta et al.

Figure 6. Experimental X-band EPR spectrum of complexes 1, 2, and 4 at the indicated temperatures in the solid state (microwave frequency 9.3499 GHz,

microwave power 2 mW, modulation amplitude 0.1 mT, and receiver level 74 dB). The simulated spectra were made using the parameters given in the text

(line width 400 G). The origin of the large signal at g ∼ 2 that appears in the 294 K spectrum of 1 is unknown.

where ξ is the spin-orbit coupling constant and k_iare the

orbital reduction factors, which are in general e1. We have

compared these expressions to those obtained from the full d⁵

matrix diagonalization and find that they are in remarkably

good agreement to within 5% over a range of parameters

appropriate to this system. The relative energies of the split

⁴T_1gcomponents are to a good approximation determined by

the relative energies of the t_2gd-orbitals. Assuming isotropic

orbital reduction factors, and taking into account that the

average energy of the ⁴T_1gcomponents is given by the strong

field expression 40B - Δ_o, where Δ_ois the overall effective

octahedral splitting, then the above expressions reduce to

through the effective octahedral splitting Δ_o(= 3σ - 4π) in

eq 2. Analogous expressions can be found for the specific

anisotropic π-bonding pattern in the model complexes 1-4

as described above. Using the expressions for the relative

energy of the t_2gorbitals given in Figure 4b, together with

eq 2 and typical values of Δ_o∼15,000 cm^-1,40B∼30,000 cm^-1,

ζ ∼ 460 cm^-1, and the values k ∼ 0.8 and a common

π ∼ 2000 cm^-1value.⁵²The later value is taken as represen-

tative of N/O πbonding.⁵³Restriction of |E/D< 1/3 requires

the zfx permutation of the axes in Figure 4, giving

D =-0.66 cm^-1and E=þ0.094 cm^-1. The surprising con-

clusion is that with this pattern of anisotropic π-bonding with

a single common π parameter, the principal axis of the ZFS

parameters is perpendicular (not collinear) with the geo-

metric axis of compression. That is, the principal axis of the

ZFS parameters is along the N2-Fe-O3, not the O1-

Fe-O4, direction. While it is not realistic to have a single

π-bonding parameter, the above clearly shows the sensitivity

of the ZFS parameters to the π- rather than the σ-bonding

character of the ligands.

ξ²k²

D ¼

½2EðxyÞ - EðxzÞ - EðyzÞꢁ

2

10ð40B - Δ_oÞ

ξ²k²

¼

½π_xþ π_y- 2π_zꢁ

ð2Þ

2

5ð40B - Δ_oÞ

EPR spectra for complexes 1, 2, and 4 are shown in

Figure 6, and the data were fitted to the following spin

Hamiltonian, where standard axes are assumed to restrict

the ZFS parameters to E/D e 1/3:

ξ²k²

E ¼

½EðxzÞ - EðyzÞꢁ

2

10ð40B - Δ_oÞ

ξ²k²

¼

½π_y- π_xꢁ

^

2

H_spin¼ βðg_xB_xS_xþ g_yB_yS_yþ g_zB_zS_zÞ

5ð40B - Δ_oÞ

2

^

2

^

2

^

þ DðS_z- 1=3SðS þ 1Þ þ EðS_x- S_yÞ

ð3Þ

in terms of the d-orbital energies E(xy), E(xz), E(yz). The π_x,

π_yand π_zparameters in the right-hand side expressions are

isotropic π-bonding parameters for ligands along the x, y,

and z axes. When evaluated for a tetragonal compression

(π_z> π_x= π_y) the expected negative value for D and a zero

value for E is obtained.

While the g-values would be expected to have rhombic

symmetry, in practice any anisotropy is very small⁵⁴as the

(52) Lever, A. B. P. Inorganic electronic spectroscopy, 2nd ed.; Elsevier:

New York, 1984.

(53) Figgis, B. N. Hitchman, M. A. Ligand field theory and its applications;

Wiley-VCH: New York, 2000.

(54) Solomon, E. I.; Brunold, T. C.; Davis, M. I.; Kemsley, J. N.; Lee, S.-

K.; Lehnert, N.; Neese, F.; Skulan, A. J.; Yang, Y.-S.; Zhou, J. Chem. Rev.

2000, 100, 235–349.

The above expressions demonstrate that it is the

π-bonding which is the most important factor in deter-

mining the signs of the ZFS parameters. The σ-bonding

contributes to the magnitude (but not the sign) of D and E

Article

Inorganic Chemistry, Vol. 49, No. 24, 2010 11431

Figure 7. Fits of the VTVH MCD curves measured at 20,000 cm^-1for the indicate temperatures. Panels (a)-(c) refer to samples 1, 2, 4, respectively. Solid

lines are fits using the parameters given in the text.

orbitally nondegenerate ⁶A_1gground state multiplet dif-

fers in spin multiplicity from all the excited ligand field

states. On the other hand, the large effective g-values

directly result from the action of the ZFS parameters

(Supporting Information, Figures S1, S2). For this reason

the ZFS parameters D and E are considered more useful

in probing spectro-structural correlations in d⁵complexes

using, for example, the simple perturbation expressions

relating the ZFS parameters to AOM bonding para-

meters given in eq 2. The effective g-values, g_eff, are a

sensitive function of E/D as shown in the rhombogram in

Supporting Information, Figure S2.

strong mixing with the excited state quartet, as do the

relatively large ZFS parameters. The inclusion of the higher

order terms in the spin Hamiltonian eq 4 improves the fits,

but the additional parametrization makes it difficult to find

a unique set.

The variable temperature variable field (VTVH) MCD

data are given in Figure 7. The simulation of the VTVH

MCD data was made using the formalism of Neese and

Solomon⁵⁶where, within a scaling factor K, the MCD

magnetization curves are calculated as:

Z

π

2π

X

Δε

¼ K

N_iðlÆS_xiæM_y^ef_z^fþ mÆS_yiæM_x^ef_z^f

However, when the first excited quartet states are low in

energy, the low-symmetry split components of the T_1g

E

4

0

i

6

þ nÆS_ziæM_x^ef_y^fÞ sin θ dθ dφ

ð5Þ

excited state will be mixed into the A_1gground state

multiplet to form “spin-admixed” states.⁴⁹A way of incor-

porating the effect of this admixture within the spin

Hamiltonian formalism is to use higher order terms (eq 4).⁵¹

where the summation is over the six S = 5/2 levels, and N_i

are the Boltzmann populations of the energy levels; ÆS_xiæ

is the expectation value of the spin S_xoperator in level i,

^

a

4

^

4

^

4

^

2

^

H_s⁴_pin

¼

fðS_xþ S_yþ S_zÞ - 7=4ð3S þ 3S þ 1Þg ð4Þ

and l, m, and n specify the direction cosines of the

magnetic field with respect to a fixed molecular axis

system with l = sin θ sin φ, m = sin θ cos φ, and n =

^

6

where the fine structure constant a is defined in terms of

the spin-orbit levels given in Supporting Information,

Figure S1.

eff

cos θ. The parameters M , M , M are the effective

xy

xz

yz

transition dipole moments that are treated as phenomen-

ological parameters. In principle, these moments can be

related to the intensities of particular transitions, but as

there are many contributions this must be done on a case

by case basis. In this particular case, the z polarized nature

Since the crystal structures show that the closest inter-

molecular Fe^III-Fe^IIIdistance between different dimers is

˚

large (7.29 A in 2), the dipole-dipole interaction, which is

proportional to 1/r³, will be significantly smaller than the

ZFS and will generally result in dipolar broadening of the

EPR resonances.⁵⁵A dipole-dipole broadened spectrum

was observed for 3 (results not shown). The low tempera-

ture EPR spectra have been simulated with isotropic

eff

of the transition (Figure 4) implies that M < M

=

xz

xy

eff

M , and we have set these values to 0.1, 1.0, 1.0,

yz

respectively. The integration in eq 5 is done numerically,

and the energy levels and wave functions are found by

solving the ground state spin Hamiltonian given in eq 3.

From the six energy levels and wave functions of the |S =

5/2, M_Sæ spin basis, the Boltzmann factors and spin

expectation values can then be obtained. The calculated

VTVH curves are fitted to the experimental curves using a

least-squares method by varying the spin Hamiltonian

parameters. These fitted curves are shown as the solid

g-values and the following ZFS parameters: D = -1.0 cm^-1

;

E/D = 0.18 (1); D = þ1.8 cm^-1; E/D = 0.06 (2); and D =

þ4.0 cm^-1; E/D = 0.065 (4). In all cases the fits were

improved by allowing the g-values to optimize to values

between 1.85-2.0. It is unusual for the g-values in a d⁵

system to depart appreciably from 2.0, and it would indicate

(55) A dipole-dipole broadened spectrum was observed for 3 (results not

shown).

(56) Neese, F.; Solomon, E. I. Inorg. Chem. 1999, 38, 1847.

11432 Inorganic Chemistry, Vol. 49, No. 24, 2010

Peralta et al.

lines in Figure 7 and are obtained using the parameters

given in Table 2. The trend is the same as that obtained

from the EPR spectra, but the values are significantly

smaller. This is not surprising, as the EPR experiments

were obtained with powdered solids, while the MCD data

were obtained from the samples that were frozen glass

solutions where conformational relaxation from the solid

state structure is possible.

(d). Electronic Structure and Hammett Parameters.

The Hammett parameter σ_pcan qualitatively describe the

electronic substituent effect of the structure on the reactivity

of a site X in the series p-XC₆H₄Y for different functional

groups Y. This effect is one of many such linear free-energy

relationships (LFERs) that have been developed,^57,58and

provides a simple measure of the electron withdrawing

(σ_p> 0) or donating (σ_p< 0) ability of the functional group

relative to Y = H. Values of σ_pfor complexes 1-4 are given

in Table 2, and they rank the electron withdrawing ability in

the order -CH₃< -H < -Br < -NO₂. One might expect

a longer Fe-O bond to the substituted phenolate (Fe-O4)

for the -NO₂substituted complex, but in the structural data

for these complexes given in Table 1, little experimental

variation in Fe-O4 is observed. However, as will be dis-

cussed below, although the Fe^III-O_phenolatebond length is

little affected by variation of the substituents, both the λ_maxin

the visible spectrum and the half-wave redox potential of the

Fe^III/Fe^IIprocess are strongly influenced by the electron

withdrawing or donating nature of the substituents at the

para position of the terminal Fe^III-bound phenolate. It is

interesting to note that the bond which is the most variable in

the series (Table 1) is that to the water (Fe-O3). This is the

only monodentate terminal ligand; the other coordinating

atoms are constrained by the conformation of the L-R^2-

ligand. At least for complexes 1 and 2 it is possible to observe

that the -CH₃substituent has an effect on the Fe^III-OH₂

Figure 8. Representation ofthe R-HOMO(a) andR-LUMO(b)orbitals

for complex 2.

charge transfer transition, which is also observed in the

spectrum of the free ligand. This shift was also found

previously for a similar nitro substituted phenolate

ligand.²¹

˚

bond distance since this distance increases to 2.112 A in

[Fe^IIIZn^II( μ-OH)(H₂O)L-CH₃]^2þwhen compared with the

Electrochemical Properties of the Complexes. The elec-

trochemical behavior of complexes 1-4 was studied by

cyclic voltammetry (CV) and square wave voltammetry (SW)

in an ethanol/water 70/30% v/v solution, and the SW data

obtained at pH values of 3.5 and 6.5 are listed in Table 2. For

all complexes the cyclic voltammograms show one reversible

redox process which can be attributed to the Fe^III/Fe^IIredox

couple. Typical square wave voltammograms at distinct pH

values for the [Fe^IIIZn^II( μ-OH)(H₂O)L-H]^2þcomplex are

shown in Figure 10.

isostructural complex containing the L-H^2-ligand (2.054 A).

˚

A graphical representation of the R-HOMO and β-LUMO

orbitals for complex 2 is shown in Figure 8. As can be seen,

both the π orbitals from the terminal phenolate group and the

d orbital from the Fe^IIIcontribute to this orbital. The d-orbital

has mainly d_z₂character as expected for a tetragonal compres-

sion along the O1-Fe-O4 axis. A linear correlation was

obtained (Figure 9a) by plotting the calculated energies of the

R-HOMO for the complexes 1-4 against the Hammett

parameter (σ_p) for the substituents -CH₃, -H, -Br,

and -NO₂, respectively.

Similarly, as shown in the plot in Figure 9b, the LMCT

process is shifted to a lower wavelength (higher energy) as

the electron-withdrawing effect of the substituent group

increases. Consequently, a hypsochromic shift is ob-

served for [Fe^IIIZn^II( μ-OH)(H₂O)L-NO₂]^2þ, whereas

the complex [Fe^IIIZn^II( μ-OH)(H₂O)L-CH₃]^2þshows a

bathochromic shift when compared to [Fe^IIIZn^II( μ-OH)-

(H₂O)L-H]^2þ. The band at 353 nm in the spectrum of

[Fe^IIIZn^II( μ-OH)(H₂O)L-NO₂]^2þis shifted with respect to

the other complexes because of an overlap between the

LMCT process and π-phenolate-to-NO₂intraligand

When square wave voltammograms for 1-4 are re-

corded at pH 3.5 a process at 60-70 mV versus NHE is

observed which becomes cathodically shifted when the

pH of the solution is raised to 6.5. We interpret this result

as being a consequence of deprotonation of the Fe^III

-

bound water molecule (generation of Fe^III-OH species)

which causes an increase of the electron density on the

Fe^IIIcenter, thus making it more difficult to reduce. On

the other hand, one can observe that within the series of

complexes 1-4, the half-wave (E_1/2) potentials become

more positive as the electron-withdrawing capacity of the

para substituent group on the terminal Fe^III-bound phe-

nolate increases (-CH₃< -H < -Br < -NO₂), a clear

indication that the electron spin density around the Fe^III

center is strongly dependent on the electronic nature of

the para-substituent (Figure 9c). While the donor effect of

the methyl group results in a greater stabilization of the

(57) March, J. Advanced Organic Chemistry: Reactions, Mechanisms and

Structure, 5th ed.; John Wiley & Sons: New York, 2001.

ꢀ

(58) Simon-Manso, Y. J. Phys. Chem. A 2005, 109, 2006.

Article

Inorganic Chemistry, Vol. 49, No. 24, 2010 11433

Figure 9. Linear correlation fromHammett parameter (σ_p) and (a) calculated energies of the R-HOMO for the complexes 1-4; (b) the lowest LMCT band

for complexes 1-4, in CH₃CN; (c) the half-wave potentials (E_1/2) for Fe^III/Fe^IIat pH 6.5 for complexes 1-4. Data obtained from square wave

voltammograms (pulse = 20 mV, frequency = 50 Hz) of [Fe^IIIZn^II(μ-OH)(H₂O)L-R]^2þcomplexes, in ethanol/water 70/30% v/v solution. [complex] =

5.0 ꢀ 10^-4mol L^-1, I = 0.1 mol L^-1NaCl; carbon working electrode, platinum wire auxiliary electrode, reference: Ag/AgCl (NaCl 3 mol L^-1); internal

standard: K₃[Fe(CN)₆)]; (d) the potentiometric pK_a2of complexes 1-4. (e) Hammett correlation for the hydrolysis of 2,4-bdnpp promoted by complexes

1-4. (f) Hammett relationship for IC₅₀for K562 and complexes 1-4. IC₅₀is the concentration required to inhibit 50% of cell growth. The mean IC₅₀( SD

was determined in three independent experiments each performed in duplicate.

3þ oxidation state of the Fe^IIIcenter, the opposite occurs

for [Fe^IIIZn^II( μ-OH)(H₂O)L-NO₂]^2þ, where the 2þ oxi-

dation state of the iron is more stabilized by the electron-

withdrawing effect of the nitro group. Interestingly, not

only does the λ_max(500-600 nm; Table 2) of the lowest

LMCT band correlate well with the Hammett parameter

(σ_p)⁵⁷but also the half-wave potentials of the Fe^III/Fe^II

couple (Figures 9b and 9c). This indicates that the ob-

served effects of the substituent groups on the spectral

and redox properties affect the HOMO energies to the

same extent (Supporting Information, Figure S3), thus

demonstrating that the theoretical calculations ade-

quately describe the electronic behavior of 1-4. Indeed,

the effects of the substituents on the electronic properties

of the biomimetics will be of crucial significance in

discussing the factors that determine the differences in

the catalytic activities of 1-4 in the hydrolysis of the

model diester 2,4-bdnpp and DNA cleavage (vide infra).

Potentiometric Titrations of the Complexes. Potentio-

metric titration studies of 1-4 in a water/ethanol (30:70)

solution^6,8,14showed the neutralization of 3 mol of KOH/

mol of complex in the pH range 2-10. The corresponding

11434 Inorganic Chemistry, Vol. 49, No. 24, 2010

Peralta et al.

Figure 11. Dependence of the initial reaction rate (V₀) on pH for the

hydrolysis of 2,4-bdnpp promoted by complexes 1, 3, and 4, respectively.

[complex] = 4.0 ꢀ 10^-5mol L^-1for complexes1 and 3 and 6.0 ꢀ 10^-5mol

Figure 10. Square wave voltammograms (pulse = 20 mV, frequency =

50Hz) of[Fe^IIIZn^II(μ-OH)(H₂O)L-H]^2þbyvaryingpHfrom3.5 to6.50 in

pH intervals of 0.25, in ethanol/water 70/30% v/v solution. [complex] =

5.0 ꢀ 10^-4mol L^-1, I = 0.1 mol L^-1NaCl; carbon working electrode,

platinum wire auxiliary electrode, reference: Ag/AgCl (NaCl 3 mol L^-1);

internal standard: K₃[Fe(CN)₆)].

L

^-1for complex 4; [2,4-bdnpp] = 5.0 ꢀ 10^-3mol L^-1; [buffers] = 0.05

mol L^-1; I = 0.05 mol L^-1(LiClO₄) in H₂O/CH₃CN (1:1) at 25 ꢀC. Data

for 2 has been given previously.¹⁴

expected within the series 1-4, the [Fe^IIIZn^II( μ-OH)-

(H₂O)L-CH₃]^2þcomplex shows the highest pK_a2value, while

the [Fe^IIIZn^II( μ-OH)(H₂O)L-NO₂]^2þderivative has the

lowest.⁵⁹

Table 3. Kinetic and Potentiometric pK_a’s for Complexes 1-4

kinetics

potentiometric

kinetic pH

complex optimum

Phosphodiesterase-Like Activity of the Complexes. It is

important to note that complexes 1-4 possess the follow-

ing features which are required for developing biomi-

metics of this type: (i) cis-oriented labile water molecules

to coordinate the substrate and provide a H₂O/OH^-

molecule as the nucleophile and (ii) a low pK_avalue for

one of the water molecules coordinated to the Fe^IIIcenter,

generating a good nucleophile under physiological pH

conditions. This last characteristic is present in all four

complexes and is strongly influenced by the para sub-

stituent group on the terminal Fe^III-bound phenolate as

can be observed in the pK_a’s obtained (Table 3).

pK₁

pK₂

pK_a1

pK_a2

pK_a3

1

2

3

4

6.5 ( 0.5 5.5 ( 0.1 8.7 ( 0.2 3.28

7.0 ( 0.5 5.1 ( 0.2 8.1 ( 0.2 3.01

6.5 ( 0.5 5.0 ( 0.2 8.0 ( 0.2 2.73

6.5 ( 0.5 5.0 ( 0.2 8.3 ( 0.1 2.55

4.88

4.66

4.64

4.25

8.43

8.25

7.80

8.25

pK_avalues are given in Table 3, and speciation plots are

shown in the Supporting Information, Figure S4. For all

of the complexes, the first dissociation constant (pK_a1)

can be attributed to the deprotonation of a μ-aqua species

in the equilibrium

The phosphodiesterase-like activity of complexes 1-4 was

evaluated in the hydrolysis reaction of the activated substrate

2,4-bdnpp²⁶and monitored spectrophotometrically at 400 nm,

and the extinction coefficients of the product 2,4-dinitro-

phenolate were measured at different pH’s under the same

experimental conditions of the kinetic measurements:

(pH/ε, M^-1cm^-1: 3.5/2125; 4.0/3410; 4.5/7180; 5.0/10080;

5.5/11400; 6.0/12000; 6.5-10.25/12100 for the 2,4-dnp

anion) at 25 ꢀC. The pH dependence of the catalytic activity

between pH 3.5 and pH 10.25 shows bell-shaped profiles

with optima at pH 6.5, 7.0, 6.5, and 6.5 for complexes 1-4,

respectively (Table 3, Figure 11). Overall, these values are

in the range found for the [M^IIIM^II( μ-OAc)₂(L-H)]

complexes^6,9,11,14under similar experimental conditions

and for PAPs.²For each complex, the reaction is strongly

influenced by the pH and the bell-shaped pH dependence of

the reactivity indicates that at least two protonation equili-

bria (pK₁= pK_a2and pK₂= pK_a3) are relevant to catalysis

(Table 3), in good agreement with the values of the pK_a’s

obtained from potentiometric titration. Indeed these data

suggest that deprotonation of the Fe^III-coordinated water

½ðOH₂ÞFeðμ -OH₂ÞZnðOH₂Þꢁ ðAÞ

h ½ðOH₂ÞFeðμ -OHÞZnðOH₂Þꢁ ðBÞ

ð6Þ

while pK_a2and pK_a3are attributed to the deprotonation

of terminally coordinated water molecules at the Fe^IIIsite

and the Zn^IIsite, respectively, according to the following

equilibria:

½ðOH₂ÞFeðμ -OHÞZnðOHÞꢁ ðBÞ

h ½ðOHÞFeðμ -OHÞZnðOH₂Þꢁ ðCÞ

h ½ðOHÞFeðμ -OHÞZnðOHÞꢁ ðDÞ

ð7Þ

Species B is the same as that determined by X-ray crystal-

lography but with an additional Zn-coordinated water

molecule. It is worth noting that the pK_a2values attributed

to deprotonation of the terminal Fe^III-bound water for

complexes 1-4 also have a linear dependence on the

Hammett parameters (Figure 9d). In fact, the dependence

of the pK_avalues on the Hammett parameters, as well as the

linear relationships discussed in the previous section, shows

the strong influence of the ligand on the electron spin

density on the Fe^IIIcenter as the electron donor/acceptor

nature of the ligand substituent is changed. Thus, as

(59) Potentiometric Titration of complexes 1-4 in water/CH₃CN (50:50)

results in a similar trend of the pK_avalues obtained in water/ethanol (30:70)

with pK_a’s in water/CH₃CN being ∼0.15 pH units higher.

Article

Inorganic Chemistry, Vol. 49, No. 24, 2010 11435

Table 4. Kinetic Parameters for the Hydrolysis of 2,4-bdnpp, Catalyzed by Complexes 1-4^a

complex

V_max(mol L^-1s^-1

)

K_M(mol L^-1

)

k_cat(s^-1

)

k_cat/K_M(mol^-1L s^-1

)

K_assoc(mol^-1L)

k_cat/k_nc

1

2

3

4

3.68 ( 0.15 ꢀ 10^-8

3.50 ( 0.12 ꢀ 10^-8

2.64 ( 0.09 ꢀ 10^-8

1.85 þ 0.10 ꢀ 10^-8

3.05 ( 0.10 ꢀ 10^-3

4.20 ( 0.09 ꢀ 10^-3

6.97 ( 0.11 ꢀ 10^-3

2.58 ( 0.07 ꢀ 10^-3

9.20 ꢀ 10^-4

9.13 ꢀ 10^-4

6.55 ꢀ 10^-4

4.63 ꢀ 10^-4

30 ꢀ 10^-2

20 ꢀ 10^-2

9.5 ꢀ 10^-2

18 ꢀ 10^-2

327

238

143

387

4.9 ꢀ 10³

4.8 ꢀ 10³

3.5 ꢀ 10³

2.5 ꢀ 10³

^aConditions: CH₃CN/H₂O 1:1; [complex] = 4.0 ꢀ 10^-5mol L^-1; [buffer] = 0.05 mol L^-1(MES or HEPES); I = 0.05 mol L^-1(LiClO₄); [2,4-bdnpp] =

1.0-8.0 ꢀ 10^-3mol L^-1at 25 ꢀC.

molecule generates the catalytic active species [(OH)Fe^III-

Scheme 1. Mechanism Proposed for the Hydrolysis of 2,4-bdnpp by

Complexes 1-4

( μ-OH)Zn^II(H₂O)] within the series of complexes 1-4,

although it does not necessarily discriminate whether the

initiating nucleophile is the terminal or the bridging Fe^III-

bound hydroxo group. The dependence of the initial rates, at

optimal pH for 1-4, on the concentration of the substrate

reveals saturation kinetics with Michaelis-Menten-like be-

havior. The resulting k_catand K_Mparameters were obtained

by fitting the data (V₀vs [bdnpp], Supporting Information,

Figure S5) using the non-linear regression of the Michae-

lis-Menten equation, and the values are listed in Table 4.

For the complexes 1-4, increasing catalytic activity is

measured (Table 4) in the following sequence of ligands:

L-CH₃(1) > L-H (2) > L-Br (3) > L-NO₂(4). The

complexes 1-2 and 3-4 show, respectively, ∼100 and

10 turnovers in 24 h at pH = 6.5 and 25 ꢀC. To evaluate

the monoesterase activity of 1-4, 1:1 stoichiometric

reactions between the complexes and the phosphate

turnover number (Table 4). It is important to emphasize that

diester 2,4-bdnpp were carried out. As observed

within the series of complexes investigated, the structural

previously,^14,41after 55 h at 50 ꢀC one 2,4-bdnp molecule is

unit [(OH)Fe^III( μ-OH)Zn^II(OH₂] of the catalyst is main-

released, indicating that only the diester can be hydrolyzed.

tained in solution as demonstrated by the potentiometric

Isotope effect studies for the hydrolytic cleavage of

titration experiments (vide supra) and thus the observation

2,4-bdnpp by 1-4 were carried out under identical con-

of different pK_a2’s (nucleophiles) and turnover rates must be

ditions using H₂O or D₂O to probe the possibility of a

intrinsically related to the change of the donor/acceptor

general base reaction mechanism. If the solvent deuterium

nature of the para-substituent on the terminal Fe^III-bound

isotope effect k_H/k_Dof identical reactions remains between

phenolate.

0.80 and 1.50, no proton transfer is present in the rate-limiting

Interestingly, a F value of -0.60 for the Hammett plot on

step of the reaction.⁶⁰Complexes 1-4 show k_H/k_Dvalues of

pK_a2(Figure 9d) arises from the strength of the interaction

1.03, 1.34,¹⁴1.12, and 0.93, respectively, providing further

between the aryl substituents and the acidic water bound to

evidence for an intramolecular nucleophilic attack in the

the Fe^IIIcenter. This interaction depends on the distance

hydrolysis of 2,4-bdnpp promoted by these complexes.

between the interacting sites and the identity of the interven-

As discussed in the previous sections, the para substituents

ing groups. In the simplest case, substituted phenols have no

on the terminal Fe^III-bound phenolate strongly influence the

PhOH 61

intervening groups and exhibit a F = -2.23 for pK_a

.

electronic properties of the Fe^IIIcenter demonstrated through

several linear relationships involving the dependence of these

electronic properties (Figure 9) on the Hammett parameters.

Particularly interesting is the fact that the pK_a2values

attributed to deprotonation of the terminally Fe^III-bound

water for complexes 1-4 have a linear dependence on the

Hammett constants, providing evidence that the Lewis acid-

ity of the Fe^IIIcenter is directly affected as the electron donor/

acceptor nature of the ligand substituent is changed.

In contrast, the polar substituent effect on the pK_a2of the

acidic water in 1-4 is transmitted through the aryl oxygen

atom and the ligand-bound Fe^IIIcausing a lower dependence

on the F value. This dependence relates to the product of

“attenuation factors” (f) for each intervening group in the

phenol, that is, F = F_phenolf_Of_Fe. Using f_O= 0.64 from the

literature,⁶¹a value of 0.42 is calculated for f_Fe.⁶¹

In addition, a LFER is found when log(k_cat/k₀) is

plotted against the Hammett parameter σ (Figure 9e),

and a least-squares treatment of the data gives a F value

of -0.40. Indeed, the F value of -0.40 for the Hammett

plot on k_catis also a function of the distance between the

aryl substituents and the nucleophilic hydroxide, but lies

0.2 units above the F value for pK_a2, which in relation to a

single proton is a consequence of the lower affinity of the

phosphoryl group by the metal coordinated hydroxo. The

The hydrolysis reactions of the phosphate diester

2,4-bdnpp catalyzed by the series of complexes 1-4 under

identical experimental conditions reveal that the turnover

rate (k_cat) decreases in the following sequence of substitu-

ents: -CH₃>-H>-Br > -NO₂. While complex 1 con-

taining the -CH₃substituent and the highest pK_a2shows

the highest catalytic reactivity, the [Fe^IIIZn^II( μ-OH)(H₂O)-

L-NO₂]^2þderivative has the lowest pK_a2and the lowest

(60) Deal, K. A.; Hengge, A. C.; Burstyn, J. N. J. Am. Chem. Soc. 1996,

118, 1713–1718.

(61) Williams, A. Free Energy Relationships in Organic and Bio-organic

Chemistry; Royal Society of Chemistry: Cambridge, 2003.

11436 Inorganic Chemistry, Vol. 49, No. 24, 2010

Peralta et al.

Figure 12. Supercoiled DNA cleavage by 1-4 (30 μmol L^-1) in PIPES buffer (pH 6.5, 25 mmol L^-1) in the absence (A) and in the presence of Distamycin

(50 μmol L^-1) (B) and in the presence of Methyl green (50 μmol L^-1) (C) for 16 h at 50 ꢀC.

Brønsted plot of log k_catversus the pK_a2of the Fe^III-OH₂

(Supporting Information, Figure S6) exhibits a good

linear correlation with a β_nuc=0.60. This value is sub-

stantially higher than those observed for the uncatalyzed

hydrolyses of phosphate diesters and triesters (β_nuc

values around 0.30),^62,63therefore indicating that the

hydroxide-catalyzed hydrolysis of the 2,4-bdnpp involves

a more nucleophilic contribution in the transition state.

In summary, given that complexes 1-4 are structurally

very similar and that the difference in reactivity results from

varying the strength of the nucleophile (Fe-OH) because of

the change of the para-substituent on the terminal Fe-bound

phenolate, as evidenced from the LFER/Hammett plot

(Figure 9e), a common reaction coordinate/transition state

and thus a common mechanism, as shown in Scheme 1, is

proposed for the hydrolysis reaction of the phosphate diester

2,4-bdnpp by these complexes.¹⁴In the first phase of the

mechanism [(OH)Fe^III( μ-OH)Zn^II(OH₂)] is the catalytically

active species and possesses a labile site that permits the

monodentate binding of the substrate to the Zn^IIcenter. This

hypothesis is strongly supported by EPR spectroscopic data

of 2, measured in the absence and presence of 2,4-bdnpp,

which indicate that the substrate does not interact with the

Fe^IIIcenter.¹⁴Next, an intramolecular nucleophilic attack by

the cis oriented Fe^III-bound hydroxide leads to the release of

2,4-dinitrophenolate with the monoester 2,4-dinitrophenyl-

phosphate remaining coordinated to the complex; and

finally, the μ-1,3-coordinated monoester intermediate under-

goes substitution by two water molecules and regenerates the

active site for the next catalytic cycle. This proposal, as

shown in Scheme 1, is in accordance with the X-ray structure

of a tetranuclear Fe₂Zn₂complex bridged by inorganic

phosphate that has been determined as the end product

resulting from the dimerization of catalyst 2 in a significantly

slower reaction.¹⁴Therefore, it has been concluded that the

μ-hydroxo is a significantly poorer nucleophile than the

terminally Fe^III-bound hydroxo group, and that hydrolysis

of the monoester to form inorganic phosphate and 2,4-

dinitrophenolate occurs exclusively because of the back-

ground reaction.

DNA Cleavage Catalyzed by 1-4. Data on DNA

cleavage by the complexes 1-4 indicated that all com-

plexes are able to moderately cleave the plasmid DNA

after 16 h at 50 ꢀC and low concentration (30 μmol L^-1),

converting the plasmid supercoiled DNA into open cir-

cular DNA (Figure 12A). All of ROS scavengers did not

inhibit the DNA cleavage performed by 1-4 clearly

suggesting that the DNA scission mechanism maybe

follows a hydrolytic rather than oxidative pathway

(Supporting Information, Figure S7). Then, the proposed

mechanism of 2,4-bdnpp hydrolysis can be also illustrate

to the DNA cleavage event.

Distamycin did not prevent the DNA cleavage by 1-4,

but on the contrary enhanced the DNA fragmentation

(Figure 12B). This effect may be attributed to the distamycin

binding at the DNA minor groove, modifying the DNA

structure in such a way that it enhances the binding affinity

of the complexes, as observed for other examples in the

literature.⁶⁵On the other hand, methyl green did not affect

the DNA cleavage promoted by 1 and 2, but slightly

decreases the activity of 3 and 4. Regarding these results it

is difficult to suggest an appropriate mechanism of complex-

DNA interaction. However, a DNA-groove binding mode

cannot be ruled out by these assays.

The DNA cleavage kinetic behavior was investigated

with concentration dependence on complexes 1-4. When

the obtained DNA cleavage rates (k_obs) were plotted

(64) Sreedhara, A.; Freed, J. D.; Cowan, J. A. J. Am. Chem. Soc. 2000,

122, 8814–8824.

(65) Ferrer, S.; Ballesteros, R.; Sambartolome, A.; Gonzalez, M.; Alzuet,

G.; Borras, J.; Liu, M. J. Inorg. Biochem. 2004, 98, 1436–1446. Thomas, A. M.;

Nethaji, M.; Mahadevan, S.; Chakravarty, A. R. J. Inorg. Biochem. 2003, 94,

171–178.

(62) Khan, S. A.; Kirby, A. J. J. Chem. Soc. 1970, 1172–1182.

(63) Kirby, A. J.; Younas, M. J. J. Chem. Soc. 1970, 1165–1172.

Article

Inorganic Chemistry, Vol. 49, No. 24, 2010 11437

Figure 13. Kinetics of supercoiled DNA cleavage by 1-4 at pH 6.5 under pseudo-Michaelis-Menten conditions. Conditions: [plasmid DNA] = 30 μmol

L

^-1; [complex] = 0-40 μmol L^-1; [PIPES buffer] = 25 mmol L^-1at 50 ꢀC. Samples were collected at 0, 60, 120, 240, 360, and 480 min.

Table 6. IC₅₀Values for Complexes 1, 2, 3, and 4

IC₅₀(μM ( s.d.)

Table 5. Kinetic DNA Degradation Parameters Obtained for the Complexes 1-4

kinetic parameters

1

2

3

4

k

_cat(h^-1

)

4.05 ꢀ 10^-22.09 ꢀ 10^-22.89 ꢀ 10^-23.63 ꢀ 10^-2

3.23 ꢀ 10^-621.4 ꢀ 10^-63.66 ꢀ 10^-63.73 ꢀ 10^-6

complex

GLC4

K562

K_M(mol L^-1

k_cat/K_M

)

1

2

3

4

6.3 ( 0.6

4.37 ( 0.4

5.73 ( 0.6

6.2 ( 0.5

21.9 ( 2.0

47.9 ( 4.0

1.39 ꢀ 10⁴

11.2 ꢀ 10⁵

0.98 ꢀ 10⁴

5.8 ꢀ 10⁵

0.79 ꢀ 10⁴

8.0 ꢀ 10⁵

0.97 ꢀ 10⁴

10.0 ꢀ 10⁵

(mol^-1L h^-1

)

a

E (k_cat/k_unc

)

18.2 ( 1.9

^aSpontaneous dsDNA hydrolysis (3.6 ꢀ 10^-8h^{-1 64}

)

complexes necessary to inhibit 50% of cell growth, IC₅₀, are

shown in Table 6. By comparing the activity of the com-

plexes, one observes that the effectiveness increases in the

order 4 < 3 < 2 < 1. The efficacy follows the trend of

increasing basicity of the active site. As demonstrated above,

an electron-withdrawing group such as -NO₂or -Br

decreases the basicity of the Fe^III-bound OH^-group while

a hydrogen or a methyl group, by an inductive effect,

increases the basicity of the hydroxo group, which acts as

the initiating nucleophile in the hydrolysis of phosphate

diester bonds (Scheme 1). Interestingly, a linear correlation

was found between the Hammett parameter σ_pand the IC₅₀

values for complexes 1-4 (Figure 9f). Nevertheless, on the

basis of the experimental results presented in this study, at

present we were unable to establish a clear relationship

between DNA in vitro cleavage and tumor cell death.

against the complexes’ concentration, the k_obsvalues

reach a saturation behavior, proper of a pseudo-Michae-

lis-Menten behavior (Figure 13). The kinetic parameters

obtained under these experimental conditions from these

plots are summarized in Table 5. Indeed, these results also

support the hydrolytic mechanism of DNA cleavage,

since metal complexes with oxidative DNA scission mech-

anism usually do not follow a saturation behavior. On the

basis of k_catvalues, the activity order of DNA cleavage

(1 = 4 > 3 > 2) does not follows the trend in reactivity

observed in the hydrolysis of the 2,4-bdnpp substrate

(1 > 2 > 3 > 4), which demonstrates that other factors

may contribute to complex-DNA interaction rather than

the presence of donating and electron withdrawing substit-

uents. Nevertheless, it should be noted that the K_Mvalues

for the complexes containing the bulky para-substituents

(R = 1, 3, 4) are about 6 times lower in relation to complex

1 (R = H) and thus suggest a higher hydrophobic interac-

tion between the substituted aromatic rings and the hydro-

phobic groups from other molecules such as the base pairs

of DNA.

Conclusion

Considerable success has been achieved in the synthesis of

model complexes that mimic structural and spectroscopic

aspects of corresponding enzymatic systems. However, in

comparison to their biological counterparts the synthetic

systems generally have catalytic efficiencies that are orders

of magnitude lower. In the active sites of enzymes, substrates

generally bind in a specific orientation relative to some

Cytotoxicity Activities of the Complexes. The in vitro

cytotoxic activities of complexes 1-4 were evaluated against

K562 and GLC4 tumor cell lines. The concentrations of

11438 Inorganic Chemistry, Vol. 49, No. 24, 2010

Peralta et al.

functional groups within the catalyst, and hydrogen bonds play

an important role in binding and orienting substrates opti-

mally. To a large extent the low catalytic efficiency in model

systems is associated with their inability to bind and orient

substrates in an optimal manner. In the present work we have

synthesized and studied the structural and spectroscopic prop-

erties of a series of phenolate coordinated iron(III) complexes

that resemble the catalytic center of PAPs. Considering that the

Fe^IIIplays an essential role in the reaction mechanism of PAPs

activating the hydrolysis-initiating nucleophile,²the interaction

between the phenolate ligand and the metal ion may play an

important role in the modulation of the activity. To probe the

role of the Fe^III-O_phenolatebond in the model systems (and, by

analogy, the Fe^III-tyrosinate moiety in PAPs) electron donating

and electron withdrawing substituents have been incorporated.

The effects of the substituents manifest themselves in a sys-

tematic alteration of the electronic properties, mapped through

the relationships between the Hammett parameter and various

ground and excited state parameters. The electron donating

substituents increase, and electron withdrawing substituents

decrease the hydrolytic activity of the biomimetic complex

toward a model phosphate substrate. The reverse, however

(with exception of [Fe^IIIZn^II( μ-OH)(H₂O)L-NO₂]^2þ), is true

for the effect on nuclease activity indicating that the mechanism

of hydrolysis for these two activities (i.e., phosphorolysis and

nucleolysis) is likely different. These observations suggest that the

goal of producing effective biomimetics will be attained through

a combination of structural and electronic enhancements in the

second coordination sphere which will effectively mimic the

substrate binding pockets and relevant hydrogen bonding inter-

actions in naturally occurring systems.⁶⁶The electronic effect of

the substituents seems also to impact the ability of complexes to

inhibit the growth of tumor cell lines. The cytotoxic activity

increases with the electron donating ability which raises the

basicity of the hydroxo group acting as a nucleophile. Further

studies aiming to establish a clear relationship between DNA in

vitro cleavage and tumor cell death by complexes 1-4 are

underway and will the subject of future reports.

Acknowledgment. The authors thank CNPq, CAPES-

ꢀ

PROCAD, FAPESC and INCT - Catalise (Brazil) and

the Australian Research Council through grants DP0986292,

DP0986613, DP0664039, and DP0558652 for financial support.

Supporting Information Available: Table S1 and Figures S1-S7

in PDF format and crystallographic data in CIF format. This

material is available free of charge via the Internet at http://pubs.

acs.org. Also, crystallographic data (without structure factors) for

the structure(s) reported in this paper have been deposited with the

Cambridge Crystallographic Data Centre as supplementary pub-

lication nos. CCDC-737630 and CCDC-737631. Copies of the data

can be obtained free of charge from the CCDC (12 Union Road,

Cambridge CB2 1EZ, U.K.; phone: (þ44) 1223-336-408; fax:

(þ44) 1223-336-003; e-mail: deposit@ccdc.cam.ac.uk; Web site

www.ccdc.cam.ac.uk).

(66) Piovezan, C.; Jovito, R.; Bortoluzzi, A. J.; Terenzi, H.; Fischer, F. L.;

Severino, P. C.; Pich, C. T.; Azzolini, G. G.; Peralta, R. A.; Rossi, L. M.;

Neves, A. Inorg. Chem. 2010, 49, 2580–2582.

Article Doi

Electronic structure and spectro-structural correlations of Fe IIIZnII biomimetics for purple acid phosphatases: Relevance to DNA cleavage and cytotoxic activity

DOI: 10.1021/ic101433t

Source and publish data:

Authors:

Article abstract of DOI:10.1021/ic101433t

Full text of DOI:10.1021/ic101433t

Products guided by the article

R&D Labs maybe for 1028059-11-1

Relevant to this article

Hot Product