Second-Sphere Effects in Dinuclear FeIIIZnII Hydrolase Biomimetics: Tuning Binding and Reactivity Properties

Article

pubs.acs.org/IC

Cite This: Inorg. Chem. XXXX, XXX, XXX −XXX

Second-Sphere Eﬀects in Dinuclear Fe^IIIZn^IIHydrolase Biomimetics:

Tuning Binding and Reactivity Properties

Tiago Pacheco Camargo,^†Ademir Neves,*^,†Rosely A. Peralta,^†Claudia Chaves,^†Elene C. P. Maia,^‡

́

Edgar H. Lizarazo-Jaimes,^‡Dawidson A. Gomes,^§Tiago Bortolotto,^⊥Douglas R. Norberto,^⊥

Hernan Terenzi,*^,⊥David L. Tierney,^∥and Gerhard Schenk^#

́

⊥

^†Departamento de Química, LABINC, and Departamento de Bioquímica, CEBIME, Universidade Federal de Santa Catarina,

́

Florianopolis, Santa Catarina 88040-900, Brazil

^‡Departamento de Química and ^§Departamento de Bioquímica, Universidade Federal de Minas Gerais, Belo Horizonte, Minas Gerais

31270-901, Brazil

^∥Department of Chemistry and Biochemistry, Miami University, Oxford, Ohio 45056, United States,

^#School of Chemistry and Molecular Biosciences, The University of Queensland, Brisbane 4072, Australia

S

* Supporting Information

ABSTRACT: Herein, we report the synthesis and character-

ization of two dinuclear Fe^IIIZn^IIcomplexes [Fe^IIIZn^IILP1] (1)

and [Fe^IIIZn^IILP2] (2), in which LP1 and LP2 are conjugated

systems containing one and two pyrene groups, respectively,

connected via the diamine −HN(CH₂)₄NH− spacer to the

well-known N₅O₂-donor H₂L ligand (H₂L = 2-bis{[(2-

pyridylmethyl)aminomethyl]-6-[(2-hydroxybenzyl)(2-

pyridylmethyl)]aminomethyl}-4-methylphenol). The complex

[Fe^IIIZn^IIL1] (3), in which H₂L was modiﬁed to H₂L1, with a

carbonyl group attached to the terminal phenol group, was

included in this study for comparison purposes.¹Both

complexes 1 and 2 were satisfactorily characterized in the

solid state and in solution. Extended X-ray absorption ﬁne structure data for 1 and 3 in an acetonitrile solution show that the

multiply bridged structure seen in the solid state of 3 is retained in solution. Potentiometric and UV−vis titration of 1 and 2 show

that electrostatic interaction between the protonated amino groups and coordinated water molecules signiﬁcantly decreases the

pK_aof the iron(III)-bound water compared to those of 3. On the other hand, catalytic activity studies using 1 and 2 in the

hydrolysis of the activated substrate bis(2,4-dinitrophenyl)phosphate (BDNPP) resulted in a signiﬁcant increase in the

association of the substrate (K_ass≅ 1/K_M) compared to that of 3 because of electrostatic and hydrophobic interactions between

BDNPP and the side-chain diaminopyrene of the ligands H₂LP1 and H₂LP2. In addition, the introduction of the pyrene motifs in

1 and 2 enhanced their activity toward DNA and as eﬀective antitumor drugs, although the biochemical mechanism of the latter

eﬀect is currently under investigation. These complexes represent interesting examples of how to promote an increase in the

activity of traditional artiﬁcial metal nucleases by introducing second-coordination-sphere eﬀects.

extent responsible for the discrepancy of the catalytic eﬃciency

between biomimetics and true enzymes.⁵As a consequence,

attempts to generate model complexes that may mimic

hydrogen-bonding and electrostatic interactions, as well as

hydrophobic eﬀects and van der Waals forces, have recently

gained momentum.⁵Speciﬁcally, among the strategies used to

increase the hydrolytic eﬃciency and selectivity of artiﬁcial

metal-ion-dependent hydrolytic systems, complexes modiﬁed

with noncoordinating functional groups such as amines,

guanidines, and hybrid mimics such as polyethylene imines,

dendrimers, and gold nanoparticles have been reported.⁵⁻⁸

Cyclodextrins and calix[n]arene-containing polypeptides, in

INTRODUCTION

■

Model complexes capable of cleaving phosphate ester bonds

have been attracting increasing interest in view of both their

importance at a fundamental level (e.g., as a biomimetic system

to elucidate the structure and mechanism of corresponding

enzymes) and their potential as new therapeutic drugs targeting

speciﬁc types of DNA or RNA (so-called “catalytic drugs”) and

antibiotics.²However, although synthetic model complexes are

capable of mimicking the structural and physicochemical

properties, and frequently mechanistic aspects, of the corre-

sponding active sites of their biological counterparts, the catalytic

eﬃciency of the metalloenzymes remains largely unrivaled.^3,4

Indeed, the absence of a well-deﬁned binding site that provides

relevant contact points for noncovalent interactions between the

model system and the reactants in model systems is to a large

Received: September 19, 2017

A

DOI: 10.1021/acs.inorgchem.7b02384

Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry

Article

which the biomimetic complex is covalently linked to the

supramolecular hosts, have also been described.^5,6

activity (k_cat) increased by around a factor of 10 in comparison to

that of europium(III) alone, the 15-fold increase of the

Michaelis−Menten constant, K_M, was unexpected, given that

intercalation of naphthylalkylamines into double-stranded DNA

(dsDNA) has been unequivocally established and thus the

aﬃnity of the catalyst for nucleic acid should be enhanced.¹³In a

zinc(II)-containing heptapeptide complex, the introduction of an

acridine moiety at the N-terminus of the peptide resulted in a

slight increase in the activity at low catalyst concentrations,

compensated for by a decrease at higher concentrations due to

the formation of dimers or larger aggregates of the peptide

induced by the presence of the acridine moiety in association

with the amphiphilic character of the helix.¹⁴More remarkable in

this context is a diiron^IIIcomplex bearing two acridine

intercalators, which shows a DNA hydrolytic cleavage rate of

6.6 × 10⁻³s⁻¹at 37 °C and pH 7.0. This value represents a 300-

fold rate acceleration compared to the same complex without the

acridine groups and a 14-fold acceleration relative to the complex

bearing only a single acridine group.¹⁵Another relevant example

was reported by Tonellato and co-workers.¹⁶These authors

studied the activity of a series of zinc(II) cis-1,3,5-triaminocy-

clohexane (Zn^II-tach)/anthraquinone intercalators linked with

alkyl spacers of varying length (C4−C8) in the cleavage of

plasmid DNA. Two important observations were made. First, the

complex containing the C8 alkyl spacer cleaves supercoiled DNA

15-fold more eﬃciently than the simple Zn^II-tach complex

without the anthraquinone moiety. Second, for the shorter

spacer (C4), no cleavage could be observed, indicating that the

advantage derived from increased DNA aﬃnity is canceled out by

the incorrect positioning of the reactive group. Hence, from the

examples presented above, it emerges that, apart from the

hydrolytic active site, the nature of the intercalator as well as the

type and size of the spacer between the catalytic center and the

intercalator plays a major role in the reactivity of such conjugated

systems.

The major functions of the second coordination sphere of

metallohydrolases are related to the activation of both the metal-

bound nucleophilic water (H₂O)/OH⁻and the substrate, as well

as stabilization of the transition state (TS).⁵Aliphatic amines are

particularly interesting in this case because they tend to be

protonated in solution (relevant pK_avalues are comparable to

that of lysine), and thus they may contribute to electrostatic

interactions. However, only a few examples of such biomimetics

have been described to date. Structural evidence for the

interaction between positively charged ammonium groups and

the substrate mimic phenylphosphate has been reported for a

dinuclear copper complex by Kovar

́

i and Kramer.⁹The X-ray

̈

structure of this complex indicated the presence of a hydrogen

bond between the tertiary ammonium moiety of the bipyridine

ligand and the metal-bridging phosphoryl group of the substrate

mimic. A dinuclear [Fe^III(μ-oxo)Fe^III] (Figure 1) core containing

Figure 1. Structure of the [(H₂L_x){Fe^III₂-O}(Cl)₄]²⁺complex

containing a cyclam group as an anchor.

Guided by the structure and physicochemical properties of

PAPs, our group has reported a great variety of mixed-valent

M^IIIM^IIcomplexes (M^III= Fe, Ga; M^II= Fe, Zn, Cu, Ni, Mn, Co,

Hg, Cd) with the unsymmetrical ligand H₂L = 2-bis{[(2-

pyridylmethyl)aminomethyl]-6-[(2-hydroxybenzyl)(2-

pyridylmethyl)]aminomethyl}-4-methylphenol as functional

and structural models for the active site of these enzymes.^1,4,17−24

Of particular relevance with regard to these M^IIIM^IIL

biomimetics is the fact that the dinuclear catalyst contains a

labile M^II-OH₂site for monodentate substrate binding and an

adjacent M^III-OH group as the initiating nucleophile that can be

generated at pH values close to 7.0.²¹Additionally, the

mechanism of phosphate diester bond cleavage in the DNA

model substrate bis(2,4-dinitrophenyl)phosphate (BDNPP) and

in DNA itself by these complexes is well established.^20,21Herein,

in an attempt to generate more eﬃcient and selective catalysts for

dsDNA cleavage, we describe a modiﬁcation of these model

systems by connecting a pyrene structural unit via the diamine

−HN(CH₂)₄NH− spacer to the well-characterized Fe^IIIZn^IIL

complex.¹In this complex, the ligand H₂L was modiﬁed to 2-

bis{[(2-pyridylmethyl)aminomethyl]-6-[N-(2-pyridylmethyl)-

aminomethyl]}-4-methyl-6-formylphenol (H₂L1), with a car-

bonyl group attached to the terminal phenol group, which is

ideally oriented for further chemical modiﬁcations. This H₂L1

ligand was used for the synthesis of the conjugated systems 2-

{[bis(pyridin-2-ylmethyl)amino]methyl}-6-{[[[2-hydroxy-5-

methyl-3-[[[4-[(pyren-1-ylmethyl)amino]butyl]amino]-

methyl]phenyl]methyl](pyridin-2-ylmethyl)amino]methyl}-4-

a protonated 1,4,8,11-tetraazacyclotetradecane (cyclam) plat-

form has been reported as a functional model for the

metallohydrolase purple acid phosphatase (PAP), catalyzing

the hydrolysis of the activated phosphomonoester 2,4-

dinitrophenyl phosphate (DNPP).¹⁰Secondary eﬀects exerted

by such a protonated cyclam platform have been proposed as

being responsible for the positioning and activation of the

monodentately iron(III)-bound substrate.¹¹

The design of small molecules able to cleave DNA under

physiological conditions is of great interest for the development

of novel therapeutic agents. Considerable eﬀort has been

invested in preparing metal complexes capable of binding and

breaking the DNA double helix.²However, there are only a few

examples of hydrolytically active metal complexes that have been

linked to DNA intercalating groups.¹²In 1987, Barton and co-

workers reported the ﬁrst example of an artiﬁcial DNA-

intercalating system with hydrolytic activity toward DNA.¹²

The system consisted of an intercalating ruthenium tris-

(phenanthroline) derivative attached to two bis(aminoethyl)-

amine (dien) moieties for complexing metal ions. The dinuclear

zinc(II) complex of this ligand cleaves ca. 40% of plasmid DNA in

5 h at 37 °C and pH 8.5, with a rate constant of ∼3 × 10⁻⁵s⁻¹.

This remarkably high rate for a zinc(II) complex may be due to

intercalation, although no direct evidence to support this

hypothesis has been reported. Schneider and co-workers¹³

reported an europium(III) complex containing two naphthalene

groups anchored to an azacrown ligand via C6 alkyl spacers as an

eﬃcient DNA cleaving agent. However, while the catalytic

B

DOI: 10.1021/acs.inorgchem.7b02384

Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry

Article

Chart 1. Proposed Structures for 1 and 2

electrochemical cell employed was of a standard three-electrode

conﬁguration: a glassy carbon (working), a platinum wire (counter),

and Ag/Ag⁺(reference). Potassium hexacyanoferrate(III) (E_1/2= 369

mV vs NHE) was used as the internal standard.²⁶Electrospray

ionization mass spectrometry (ESI-MS) of the complexes dissolved in

an ultrapure CH₃CN solution (500 nM) was performed using an

AmaZon X Ion Trap MS instrument (Bruker Daltonics) with an ion-

spray source using electrospray ionization in positive-ion mode. The

ion-source voltage was 4500 V. Nitrogen was used as the nebulizing (20

psi) and curtain (10 psi) gases. The samples were directly infused into

the mass spectrometer at a ﬂow rate of 180 μL/h. The scan range was

from m/z 100 to 1500. The simulated spectra were calculated using the

Mmass software.²⁷

Potentiometric studies of 1 were carried out in an CH₃CN/H₂O (1:1,

v/v) mixture with a Metrohm 848 Titrino Plus research pH meter ﬁtted

with blue-glass and Ag/AgCl reference electrodes, calibrated to read

−log [H⁺] directly, designated as the pH. Equilibrium measurements

were performed in a thermostated cell, purged with argon and

containing 50.00 mL of the CH₃CN/H₂O (1:1) solution and 0.03

mmol of the complex.²⁸The temperature was maintained at 25.00

0.05 °C, and the ionic strength of 0.100 M was maintained constant

through the addition of KCl. Computations of the results were carried

out with the BEST7 program, and species diagrams were obtained with

the SPE and SPEPLOT programs.²⁹

methylphenol (H₂LP1) and 2-{[[[3-[[[4-[bis(pyren-1-

ylmethyl)amino]butyl]amino]methyl]-2-hydroxy-5-

methylphenyl]methyl](pyridin-2-ylmethyl)amino]methyl}-6-

{[bis(pyridin-2-ylmethyl)amino]methyl}-4-methylphenol

(H₂LP2) and their corresponding Fe^IIIZn^IIcomplexes (Chart 1).

The diesterase activity (using the model substrate BDNPP and

DNA) and cytotoxic activity are signiﬁcantly increased because

of the presence of protonated amines and conjugated pyrene(s)

in the [Fe^IIIZn^IILP1] (1) and [Fe^IIIZn^IILP2] (2) catalysts. When

DNA was used as the substrate, we observed an important

increase in the cleavage activity in complexes containing pyrene

groups. Interestingly, using a footprinting technique, the contact

points of the three complexes on the DNA double-helix surface

were established, and then the ability of the complex containing

two pyrene moieties to protect a larger DNA surface was

demonstrated. Confocal ﬂuorescence microscopy reveals that

complex 1 accumulates in the lysosomes of mammalian cells, as

previously shown for other metal complexes, which, in addition

to their DNA interacting properties, suggests its potential as a

novel lead for antitumor drugs.

EXPERIMENTAL SECTION

■

Synthesis of Ligands and Complexes. The ligands H₂L and H₂L1

were obtained as described elsewhere.^1,21

Materials and Methods. All starting materials were purchased from

Aldrich, Acros, or Merck. The plasmid pBSK II (2961 bp, purchased

from Stratagene California, San Diego, CA) was used as a substrate for

DNA cleavage assays. The plasmid was transformed into DH5α

Escherichia coli competent cells, ampliﬁed as previously described,²⁵and

then puriﬁed from E. coli cells according to the manufacturer’s

instructions (Qiagen Plasmid Maxi Kit protocol). The self-comple-

mentary 49-mer oligonucleotide (5′-FAM-CCG ATT ATT TAA TCG

CCG GCC GCT TTT TGC GGC CGG CGA TTA AAT AAT C-3′,

where FAM is ﬂuorescein and the thymine nucleotides, shown in italics,

represent the predicted hairpin loop) used in DNase I footprinting

assays was purchased from IDT (Coralville, IA) and used without

further puriﬁcation. DNase I (from bovine pancreas, type IV, lyophilized

powder, ≥2000 Kunitz units/mg of protein) was purchased from Sigma-

Aldrich (St. Louis, MO), resuspended in 2.5 mM CaCl₂and 5 mM

Synthesis of (4-Aminobutyl)pyren-1-ylmethylamine (Pyr-Dab). In

a 250 mL round-bottom ﬂask, 1.25 g (14.0 mmol) of 1,4-diaminobutane

was added to 80 mL of methanol. In the next step, 0.46 g (2.0 mmol) of

1-pyrenecarboxaldehyde was added, and the reaction mixture was

allowed to react for 6 h at room temperature. The temperature was

lowered to 0 °C in an ice bath, 0.08 g (2.0 mmol) of sodium borohydride

was added, and the mixture was stirred for 3 h. The pH was adjusted to

5.0, and the solvent was removed under reduced pressure. The resulting

solid was dissolved in dichloromethane (DCM) and washed three times

with a saturated sodium bicarbonate solution and two times with H₂O,

resulting in a light-yellow foam (yield: 74%).

IR (KBr pellets, cm⁻¹): ν(N−H) 3367, 3290, ν(C−H_Ar) 3043, ν(C−

H_Aliph) 2934, 2810, ν(CC) 1594, ν(CH₂) 1443, δ(C−H_Ar) 843, 801,

δ(C−H_Ar) 708. ¹H NMR (CDCl₃): δ −1.53 (4 H, m), 2.73 (4 H, m),

4.46 (2 H, s), 5.76 (1 H, s), 8.05 (8 H, m), 8.35 (1 H, d).

1

MgCl₂at 0.03 units of DNase I/μL, and stored at −20 °C. H 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 PerkinElmer FTIR

Spectrum 100 spectrophotometer (KBr pellets). Elemental analysis was

performed on a Carlo Erba E-1110 analyzer. Electronic absorption

spectra in the 200−1200 nm range were recorded on a PerkinElmer

Lambda 750 spectrophotometer. Electrochemical measurements were

obtained using a BAS Epsilon potentiostat/galvanostat. Cyclic

voltammograms were obtained for the complexes in an acetonitrile

(CH₃CN)/H₂O (2:1) solution containing 0.1 M sodium tetraﬂuor-

oborate as the supporting electrolyte under an argon atmosphere. The

Synthesis of tert-Butyl-N-{4-[bis(pyren-1-ylmethyl)amino]butyl}-

carbamate (Pyr2-Dab-Boc). In a 125 mL round-bottom ﬂask

containing 0.56 g (3.0 mmol) of N-Boc-1,4-diaminebutane dissolved

in dry tetrahydrofuran (THF), 1.77 g (6.0 mmol) of 1-bromopyrene was

slowly added under an argon atmosphere, followed by 1 mL of

triethylamine. The mixture was stirred for 12 h under a saturated argon

atmosphere. The formation of triethylammonium bromide salt was

observed, which was removed by ﬁltration. The ﬁltrate was placed in a

rotary evaporator to give a yellowish solid. The solid was then solubilized

in DCM and washed with small portions of NaHCO₃(3 × 20 mL) and

brine (2 × 20 mL). The organic layer was dried with anhydrous Na₂SO₄,

C

DOI: 10.1021/acs.inorgchem.7b02384

Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry

Article

and the solvent was removed by rotary evaporation, resulting again in a

light-yellow foam (yield: 74%).

and left to stand. The resulting purple solid was collected and washed

with isopropyl alcohol and diethyl ether (yield: 0.36 g, 53%).

Anal. Calcd for FeZnC₆₂H₇₄N₇Cl₄O₁₅: C, 52.43; H, 5.25; N, 6.90.

Found: C, 52.58; H, 5.26; N, 6.91. IR (KBr pellets, cm⁻¹): ν(O−H)

3437, ν(C−H_Ar) 3037, ν(C−H_Aliph) 2922, 2861, ν(CC) 1607, ν(C

N) 1571, ν(CC) 1470, ν(Cl−O) 1090, δ(C−H_Ar) 843, δ(C−H_Ar)

758, δ(CC_Ar) 622.

IR (KBr pellets, cm⁻¹): ν(N−H) 3430, ν(C−H_Ar) 3043, ν(C−H_Aliph

)

2922−2794, ν(CO) 1697, ν(CC) 1505, ν(CH₂) 1362, δ(C−H_Ar)

843, δ(C−H_Ar) 708. ¹H NMR (CDCl₃): δ −1.39 (9 H, s), 1.46 (2 H, m),

1.68 (2 H, t), 2.61 (2 H, t), 2.86 (2 H, m), 4.31 (4 H, s), 8.05 (16 H, m),

8.34 (2 H, d).

Synthesis of N,N-Bis(pyren-1-ylmethyl)-1,4-butanediamine (Pyr2-

Dab). In a 125 mL round-bottom ﬂask, 1.24 g (2.0 mmol) of Pyr2-Dab-

Boc was dissolved in DCM. The system was cooled to 0 °C in an ice

bath, and 2.0 mL (ca. 1 mL/mmol) of triﬂuoroacetic acid was slowly

added (the solution acquired a greenish color). The mixture was stirred

at room temperature for another 3 h. The organic layer was washed (4 ×

20 mL) with an aqueous 2.0 M sodium hydroxide (NaOH) solution and

then dried with anhydrous Na₂SO₄. The solvent was then removed by

rotary evaporation (yield: 82%).

Synthesis of [Fe^IIIZn^II(LP2)(μ-OH)(H₂O)](ClO₄)₂(2). Complex 2 was

prepared similarly to complex 1 but using H₂LP2 instead of H₂LP1 as

the ligand and as solvents a mixture (1:1) of methanol and chloroform

(yield: 0.40 g, 56%).

Anal. Calcd for FeZnC₇₅H₇₉N₇Cl₅O₁₆: C, 55.15; H, 4.88; N, 6.00.

Found: C, 55.03; H, 4.91; N, 5.95. IR (KBr pellets, cm⁻¹): ν(O−H)

3443, ν(C−H_Ar) 3043, ν(C−H_Aliph) 2919, 2857, ν(CC) 1610, ν(C

C) 1474, ν(Cl−O) 1112, 1090, δ(C−H_Ar) 847, δ(C−H_Ar) 762, δ(C

C_Ar) 620.

IR (KBr pellets, cm⁻¹): ν(N−H) 3423, 3360, ν(C−H_Ar) 3035, ν(C−

H_Aliph) 2928, 2798, ν(CC) 1590, ν(CH₂) 1366, δ(C−H_Ar) 838, δ(C−

H_Ar) 708. ¹H NMR (CDCl₃): δ −1.03 (2 H, m), 1.12 (1 H, s), 1.39 (2 H,

m), 2.39 (2 H, t), 2.61 (2 H, t), 4.32 (4 H, s), 7.99 (16 H, m), 8.37 (2 H,

d).

Synthesis of [Fe^IIIZn^II(L1)(μ-OH)(H₂O)](ClO₄)₂(3). This complex was

prepared following procedures described in the literature. Anal. Calcd

for FeZnC₃₆H₄₀C_l2N₅O₁₄: C, 45.09; H, 4.20; N, 7.30. Found: C, 44.85;

H, 4.14; N, 7.25.

Crystallographic data for 3 were previously deposited.¹

Synthesis of H₂LP1. In a 125 mL round-bottom ﬂask, 1.32 g (2.20

mmol) of H₂L1 was dissolved in 40 mL of methanol/THF (1:1). In the

next step, 30 mL of a THF solution containing 0.66 g of Pyr-Dab (2.20

mmol) was added slowly. The mixture was stirred at room temperature

for 3 h. Subsequently, the mixture was cooled to 0 °C in an ice bath, and

small portions of NaBH₄were added directly (0.08 g, 2.20 mmol) under

constant stirring. The solution was stirred for another 3 h, during which

its yellow color gradually faded. A 1.0 M HCI solution was then added in

small portions to adjust the pH to 6, and the solvent was removed by

rotary evaporation. To the resultant oil was added 80 mL of DCM, and

the solution was washed with a saturated aqueous solution of NaHCO₃

(3 × 20 mL). The organic layer was dried with solid anhydrous Na₂SO₄,

and the solvent was removed under reduced pressure. As a result, 1.48 g

of a light-yellow foam was obtained (yield: 77%).

Reactivity Measurements. The phosphodiesterase-like activity

was determined for complexes 1 and 2 by monitoring hydrolysis of the

substrate BDNPP with a UV−vis Varian Cary 50 BIO spectrometer at

400 nm in an CH₃CN/H₂O (2:1, v/v) mixture due to the low solubility

of 2. The initial rates were measured in real time at various pH values

between 4.5 and 10.0 [the buﬀers used were 0.05 M 2-(N-

morpholino)ethanesulfonic acid (MES), 2-[4-(2-hydroxyethyl)-

piperazin-1-yl]ethanesulfonic acid (HEPES), and N-cyclohexyl-2-

aminoethanesulfonic acid (CHES)], and the relevant molar absorption

coeﬃcients of the reaction product 2,4-dinitrophenolate were

determined at each pH under the same experimental conditions as

those of the rate measurements.²⁰Reactions were monitored to less than

5% depletion of the substrate. The kinetic parameters k_catand K_Mwere

derived using the initial rate methods in a series of buﬀers ranging from

pH 4.5 to 10.0. The pH dependence of the rate (from 4 to 9) was

investigated using a substate concentration range of 0.5−6.5 mM and

complexes ([C]_final= 50.0 μM for complexes 1 and 2) at 25 °C (I = 0.05

M with LiClO₄). Isotopic eﬀects on the hydrolysis rate of BDNPP

promoted by complexes 1 and 2 (i.e., k_H/k_D) were investigated by

monitoring reactions in deuterated buﬀers.²¹

IR (KBr pellets, cm⁻¹): ν(O−H) 3430, ν(N−H) 3186, ν(C−H_Ar)

3043, 3007, ν(C−H_Aliph) 2922, 2810, ν(CC) 1586, ν(CN) 1568,

1

δ(O−H) 1365, δ(C−H_Ar) 843, δ(C−H_Ar) 768. H NMR (CDCl₃): δ

−1.52 (4 H, m), 2.16 (6 H, s), 2.63 (2 H, m), 2.75 (2 H, m), 3.66−3.80

(14 H, s), 4.43 (2 H, s), 6.76 (4 H, q), 7.09 (3 H, m), 7.34 (3 H, d), 7.55

(3 H, t), 8.03 (8 H, m), 8.29 (1 H, d), 8.50 (3 H, d).

Synthesis of H₂LP2. In a 125 mL round-bottom ﬂask, 1.17 g (2.0

mmol) of H₂L1 was added to 30 mL of a methanol/THF (1:1) mixture.

Next, 30 mL of a THF solution, containing 1.03 g of Pyr2-Dab (2.0

mmol), was slowly added. The reaction mixture was stirred at room

temperature for 2 h and then cooled to 0 °C in an ice bath. Small

portions of NaBH₄were then added directly (0.07 g, 2.0 mmol), and the

mixture was stirred for another 1 h. Subsequently, HCI was added in

small portions to adjust the pH to 6, and the solvent was removed by

rotary evaporation. To the resulting oil was added 60 mL of DCM, and

the solution was washed with a 1 M HCl solution (2 × 20 mL),

NaHCO₃(3 × 20 mL), and ﬁnally H₂O (1 × 20 mL). The organic layer

was dried with anhydrous Na₂SO₄, and the solvent was removed under

reduced pressure. As a result, 1.52 g of a light-yellow foam was obtained

(yield: 69%).

X-ray Crystallography and X-ray Absorption Fine Structure

(EXAFS) Analysis. The structure of Fe^IIIZn^IIL1 (3) has been previously

determined by X-ray crystallography.¹No crystals for 1 and 2 were

obtained. In order to compare the eﬀect of the modiﬁcations of ligand

L1 on the catalytic dinuclear Fe^IIIZn^IIcenter, Fe and Zn K-edge EXAFS

data were measured for 1 and 3 in ﬂuorescence mode in 2 mM CH₃CN

solutions. Data were collected at the National Synchrotron Light Source,

Beamline X3B. Data collection, reduction, and analysis followed

published procedures, using EXAFSPAK³⁰to process the data, which

were then ﬁtted in SIXPACK, using the calibrated phase and amplitude

functions obtained from FEFF 8.0.³¹Brieﬂy, for each data set, the

Fourier-ﬁltered ﬁrst shell was ﬁtted while allowing only the absorber−

scatterer distance and Debye−Waller factor to vary and stepping

through reasonable coordination numbers. The best-ﬁt coordination

number was then held ﬁxed in subsequent ﬁts that included outer-shell

scatterers.³²

Isothermal Titration Calorimetry (ITC). The binding interactions

between complexes 1 and 2 and calf-thymus DNA (CT-DNA) were

studied in a 30 mM HEPES buﬀer at pH 7.0 and 25 °C using a VP-ITC

titration calorimeter (MicroCal, Northampton, MA). All solutions were

thoroughly degassed before use by stirring under vacuum. The sample

cell was loaded with 1.43 mL of a 120 μM CT-DNA solution, and

titration was carried out using a 250 μL syringe ﬁlled with a 500 μM

complex solution, with stirring at 286 rpm. Injections were started after

baseline stability had been achieved. A titration experiment consisted of

40 consecutive 6 μL injections. Each injection lasted 12 s, with a 5 min

interval between injections. The values for the heat of dilution of the

complex were determined by injecting the complex solution into the

buﬀer alone, and the total observed values for the heat of binding were

IR (KBr pellets, cm⁻¹): ν(O−H) 3423, ν(C−H_Ar) 3041, 3004, ν(C−

H_Aliph) 2919, 2804, ν(CC) 1591, ν(CN) 1569, ν(CC) 1472,

1

δ(O−H) 1362, δ(C−H_Ar) 843, δ(C−H_Ar) 754, δ(CC_Ar) 708. H

NMR (CDCl₃): δ −1.32 (2 H, q), 1.68 (2 H, q), 2.17 (6 H, s), 2.22 (2 H,

m), 2.58 (2 H, m), 3.44 (2 H, s), 3.69−3.82 (12 H, s), 4.29 (4 H, s), 6.84

(4 H, q), 7.07 (3 H, m), 7.37 (3 H, d), 7.51 (3 H, q), 7.82−8.11 (16 H,

m), 8.35 (2 H, d), 8.53 (3 H, d).

Synthesis of [Fe^IIIZn^II(LP1)(μ-OH)(H₂O)](ClO₄)₂(1). In a 125 mL

round-bottom ﬂask, containing 60 mL of a THF/CH₃CN (1:1)

solution, the ligand H₂LP1 (0.48 g, 0.55 mmol) and Zn(ClO₄)₂·6H₂O

(0.20 g, 0.55 mmol) were added under stirring at room temperature. To

this mixture were added dropwise Fe(ClO₄)₃·9H₂O (0.28 g, 0.55 mmol)

in 60 mL of CH₃CN and 3 mmol of NaOH in 1.5 mL of H₂O. The

mixture was stirred for approximately 30 min, and sodium perchlorate

(NaClO₄; 0.13 g, 1.1 mmol) was then added. The solution was ﬁltered

D

DOI: 10.1021/acs.inorgchem.7b02384

Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry

Article

corrected accordingly. The heat released by the dilution of CT-DNA

was negligible. At least three titration experiments were performed for

each sample set to evaluate the reproducibility. The resulting data were

ﬁtted to a single set of identical site models using ORIGIN software

supplied with the instrument, and the apparent enthalpy change for the

binding, ΔH, the binding constant, K_B, and the binding stoichiometry, n,

were obtained. The relative Gibbs free energy change, ΔG, and entropy

change, ΔS, for the binding reaction were calculated from the

fundamental equations of thermodynamics^33,34

described above but with increases in the ionic strength of the reaction

medium obtained through the addition of sodium chloride (NaCl;

ranging from 50 to 500 mM). Also, diﬀerent reactive oxygen species

(ROS) scavengers,⁷including dimethyl sulfoxide (10%), KI (0.5 mM),

superoxide dismutase (20 units), and NaN₃(0.5 mM), were added to

the reaction mixture prior to the complex. Assays in the presence of the

DNA minor groove binder, netropsin⁴²(50 μM), or the DNA major

groove binder, methyl green (50 μM), were also performed to clarify the

DNA groove binding preference of the complexes. The plasmid DNA

was pretreated with netropsin or methyl green for 30 min and then

treated with the complexes as described above.

ΔG = −RT ln K_a

(1)

In addition, the kinetics of plasmid DNA cleavage performed by 1 and

2 was evaluated following the loss of the supercoiled DNA fraction over

time under pseudo-ﬁrst-order conditions.⁴³The apparent plasmid DNA

cleavage rates (k_obs) were obtained from the plot of ln [supercoiled DNA

(%)] versus time.

Cell Lines and Cell Culture. K562 cells, from a human chronic

myelogenous leukemia cell line (CR083; RJCB, Rio de Janeiro, Brazil),

were cultured in a RPMI 1640 (Sigma-Aldrich, St. Louis, MO) medium

ΔH − ΔG

ΔS =

(2)

T

where K_Bis the aﬃnity or binding constant (K_B= 1/K_D, the dissociation

constant), R is the ideal gas constant, and T is the temperature (in

Kelvin). In all experiments, the c value (c = K_B[CT-DNA])³³was in a

range that provides data suitable for the accurate determination of

binding constants (c ranged from 76 to 10).^35,36The reaction heat of the

interactions between the complex and DNA is composed of several

contributions, which is the reason why the enthalpy must be carefully

interpreted as an apparent or observed quantity. It includes not only the

eﬀects of the equilibrium between DNA and the complex but also

additional equilibria, including solvent eﬀects.³⁶⁻³⁸Nonetheless, it is

assumed that the quantity of heat absorbed or released is directly

proportional to the extent of binding.

supplemented with 10% fetal calf serum (CULTILAB, Sao Paulo,

̃

Brazil) at 37 °C in a humidiﬁed 5% carbon dioxide atmosphere.

SKHep-1 cells, from a human liver cancer cell line (ATCC, Manassas,

VA), were cultured at 37 °C in 5% carbon dioxide in Dulbecco’s

modiﬁed Eagle’s medium (Invitrogen, Carlsbad, CA) containing 10%

fetal bovine serum, 1 mM sodium pyruvate, 50 units/mL of penicillin,

and 50 g/mL streptomycin (Invitrogen, Carlsbad, CA).

Determination of the DNA Binding Constant from the

Electronic Spectrum. The intrinsic binding constant K_bwas

determined from spectral titration data using eq 3:

Cell Viability Assay. The cytotoxic eﬀect of the complexes was

studied by incubating K562 cells in the absence and presence of diﬀerent

complex concentrations for 24 h. The cells were then counted, and the

concentration required to inhibit 50% of cellular growth, IC₅₀, was

determined. The cell viability was checked by Trypan Blue exclusion.

The cell number was determined by Coulter counteranalysis.

C_DNA

ε_a− ε_f

C_DNA

ε_b− ε_f

1

=

+

K_b(ε_b− ε_f)

(3)

Assessment of Apoptosis. The quantiﬁcation of apoptotic cells

was performed by ﬂow cytometry using the Alexa Fluor 488 Annexin V/

Dead Cell Apoptosis Kit (Thermo Fisher Scientiﬁc, Waltham, MA).

K562 cells (1 × 10⁵M) were cultured for 24 h in the absence and

presence of a range of concentrations of complexes 1 and 3. The cells

were then washed twice with cold phosphate-buﬀered saline and stained

with Alexa Fluor 488 Annexin V and propidium iodide for 15 min at 37

°C in the dark, according to the manufacturer’s instructions. Cells were

subsequently analyzed with a Guava Easy Cyte 6-2L ﬂow cytometer

(Millipore, Burlington, MA). The Alexa Fluor 488 Annexin V and

propidium iodide ﬂuorescence data were collected using 525/30 and

583/26 emission ﬁlters, respectively. Prior to data collection, the

ﬂuorescence signals were compensated for using cells stained with each

one of the probes separately. All data were analyzed using FlowJo

Software (Tree Star, Ashland, OR). Cisplatin at 10 μM was used as a

positive control for apoptosis and late necrosis.

Fluorescence Microscopy. Nuclear Staining. SKHep-1 cells were

detached and incubated (1 × 10⁵cells/well) with 25 μM complex 1 and

0.5 μg/mL propidium iodide for 24 h. Representative cells were then

selected and photographed. Images were captured using an Olympus

IX70 microscope coupled with an Olympus UPlanSApo 60× (NA =

1.35) objective (Olympus, JAP), a Lumen 200PRO illumination source

(Prior Scientiﬁc Inc., Rockland, MA), and a cooled QIclick camera (Q

Imaging, Surrey, British Columbia, Canada). The emission and

excitation ﬁlters respectively used to collect the ﬂuorescence data for

the metal complexes were AT350/50 and AT455/50, and for propidium

iodide, they were ET555/25 and ET605/52. The ﬁlters and

components were controlled by ImagePro (Media Cybernetics,

Rockville, MD). Unstained cells were employed as a negative control,

and cells stained only with one ﬂuorochrome were employed to

compensate for the propidium iodide channel.

where C_DNAis the concentration of DNA in base pairs and the apparent

absorption coeﬃcients ε_a, ε_f, and ε_bcorrespond to A_obsd/[M], the

extinction coeﬃcient of the free compounds. In plots of C_DNA/(ε_a− ε_f)

versus C_DNA, K_bis given by the ratio of the slope to the intercept.

Absorption titration experiments were performed with ﬁxed concen-

trations of the complexes. Here the complexes were dissolved in a

solvent mixture of 25% CH₃CN and 75% HEPES buﬀer (5 mM HEPES

and 50 mM NaClO₄, pH 7.0) to avoid the precipitation of DNA. The

absorption titration experiments were performed in the absence of and

with increasing concentrations of CT-DNA. To obtain the absorption

spectra, an appropriate amount of CT-DNA was added to both the

compound solution and the reference solution to eliminate the

absorbance of CT-DNA itself.

Plasmid DNA Cleavage. The DNA cleavage activity of the

complexes was evaluated by analyzing conversion of the intact

supercoiled form of pBSK II DNA (F I) to open circular (F II) and

linear (F III) forms, which represent the plasmid forms containing

single- and double-stranded breaks, respectively.³⁹Typical reactions

were conducted in a ﬁnal volume of 20 μL using 330 ng of pBSK II DNA

(∼25 μM base pairs) in a 10 mM buﬀer (MES, HEPES, or CHES,

depending on the pH range).^7,40The cleavage reaction was initiated by

the addition of the complexes and incubation for up to 16 h at 37 °C.

Each reaction was quenched by adding 5 μL of a loading buﬀer solution

[50 mM Tris-HCl, pH 7.5, 0.01% bromophenol blue, 50% glycerol, and

250 mM ethylenediaminetetraacetic acid (EDTA)] and then subjected

to electrophoresis on a 1.0% agarose gel containing 0.3 μg/mL ethidium

bromide in 0.5 times TBE buﬀer (44.5 mM Tris, 44.5 mM boric acid,

and 1 mM EDTA at pH 8.0) at 90 V for 90 min. The resulting gels were

visualized and digitized using a DigiDoc-It gel documentation system

(UVP, Upland, CA). The proportion of plasmid DNA in each band was

quantiﬁed using Kodak Molecular Imaging Software 5.0 (Carestream

Health, Rochester, NY). The quantiﬁcation of supercoiled DNA (F I)

was corrected by a factor of 1.47 because the ability of ethidium bromide

to intercalate into this topoisomeric DNA form is lower compared with

that of open circular and linear DNA.⁴¹To determine the mechanism by

which the complexes perform DNA cleavage, external agents were

added to the reaction mixtures. In order to investigate the contribution

of electrostatic interactions in the reaction, assays were conducted as

Lysosome Staining. SKHep-1 cells were detached and incubated (1

× 10⁵cells/well) with 80 μM complex 1 and 50 nM Lysotracker Red

DND-99 for 5 h. Images were captured using a Zeiss 5 Live confocal

(Carl Zeiss, Jena, Germany) microscope with a 63× (NA = 1.4)

objective lens. The samples were excited at λ_ex= 405 nm and observed at

λ_em= 425 nm to detect complex 1 and at λ_ex= 543 nm and λ_em= 565−

615 nm to detect Lysotracker.

E

DOI: 10.1021/acs.inorgchem.7b02384

Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry

Article

Fe^IIIZn^IIL complex. This internuclear distance is in good

agreement with the EXAFS measurements of 1, which indicated

that the two metal ions are separated by ∼3.00 Å (Table 2). It is

RESULTS AND DISCUSSION

■

Synthesis and Characterization of the Ligands and

Complexes. Synthesis. The unsymmetrical ligand H₂L1,

containing a carbonyl moiety attached to the terminal phenol

group, was synthesized as described previously.¹The ligands

containing the intercalator pyrene (H₂LP1 and H₂LP2) were

obtained following the synthetic routes described in Schemes S1

Table 2. EXAFS-Derived Metrical Parameters (Å) for FeZnL1

a

(3) and FeZnLP1 (1)

sample

Fe−N/O

Zn−N/O

Zn−Fe

1

and S2, respectively, and were fully characterized through H

FeZnL1 (3)

6/1.95

6/1.97

5/2.07

5/2.11

3.00

NMR and ESI-MS (Figures S1 −S4). The conjugate mixed-

valence complexes FeZnLP1 (1) and FeZnLP2 (2) were

synthesized by employing the method described for FeZnL1

(3), but using a mixture of CH₃CN/THF (1:1) instead of

CH₃OH because of the decreased solubility of the ligands.

Solid-State Structure of 3. The molecular structure of 3

(Figure 2 and Table 1) shows that in the dinuclear [Fe^III(μ-

FeZnLP1 (1)

a

See Figure S5.

important to note that in addition to the terminal iron-bound

H₂O molecule [Fe···O3 = 2.082(6) Å], which has been proposed

as the nucleophile in the hydrolysis of diesters catalyzed by

[Fe^III(μ-OH)Zn^II(L)],²¹the structure of 3 contains a carbonyl

group (i.e., C···O5) located in the ortho position to the terminal

iron(III)-bound phenolate ligand. From the solid-state structure

of 3 (Figure 2), it is evident that this functional group is ideally

oriented for further chemical reactions. Consequently, ligand L1

was modiﬁed at this position to contain one (H₂LP1) or two

(H₂LP2) pyrene groups conjugated to the dinucleating ligand by

a 1,4-diaminobutane spacer (Chart 1).

Solution Structures of 1 and 3. The pyrene groups were

expected to be structurally mobile additions to ligand L1.

Consequently, it was no surprise that no crystals of the

corresponding FeZn complexes could be obtained. Thus, in

order to probe whether these ligand modiﬁcations aﬀect the

geometry of the catalytically relevant [Fe^III(μ-OH)Zn^II] unit, the

solution structures of 1 and 3 were studied by EXAFS, at both the

Zn and Fe K-edges (Table 2 and Figure 3; detailed information is

given in Figure S5).

Figure 2. ORTEP representation of complex 3.

Table 1. Main Bond Distances (Å) and Angles (deg) for the

Fe^IIIZn^IIL1 Complex (3)¹

Fe1−Zn1

Fe1−O2

Fe1−O4

Fe1−O1

Fe1−O3

Fe1−N2

3.0550(16)

1.912(6)

1.925(5)

2.032(5)

2.082(6)

2.132(7)

Fe1−N4

Zn1−O2

Zn1−N5

Zn1−N3

Zn1−O1

Zn1−N1

2.163(7)

1.978(6)

2.044(8)

2.062(8)

2.079(5)

2.149(6)

O2−Fe1−O4

O2−Fe1−O1

O4−Fe1−O1

O2−Fe1−O3

O4−Fe1−O3

101.2(3)

81.1(2)

177.7(3)

93.5(3)

89.2(2)

O1−Fe1−O3

O2−Fe1−N2

O4−Fe1−N2

O1−Fe1−N2

O3−Fe1−N2

90.4(2)

96.8(3)

86.2(3)

93.9(2)

169.4(3)

Figure 3. EXAFS Fourier transforms (solid lines) and corresponding

best ﬁts (for open symbols, see Table 2) for 2 mM CH₃CN solutions of

complexes 1 and 3 at the Fe and Zn K-edges.

OH)Zn^II(L1)] unit the coordination/arrangement of L1 and all

metric parameters around the metal centers are similar to those

observed in the isostructural complex containing ligand H₂L.²¹

Brieﬂy, in the dinuclear complex, the iron(III) center is facially

coordinated by the hard tridentate pendant arm of the L1 ligand

containing an amine (N4) and a pyridine (N2) nitrogen atoms

and a phenolate oxygen (O4) atom, while zinc(II) is coordinated

by the soft side of the ligand through an amine (N1) and two

pyridine (N3 and N5) nitrogen atoms. The bridging phenolate

oxygen (O1) atom, the bridging hydroxo group (O2), and a

terminal H₂O molecule (O3) complete the pseudooctahedral

N₂O₄coordination of iron(III), while the distorted trigonal

bipyramid of zinc(II) (τ = 0.67) is complemented by the bridging

phenolate (O1) and hydroxo (O2) oxygen atoms. The Fe···Zn

distance is 3.0550(16) Å in Fe^IIIZn^IIL1 and 3.040(1) Å in the

As can be seen in Figure 3, the iron and zinc EXAFS spectra of

the parent complex 3 and the asymmetric complex 1 are very

similar in a CH₃CN solution. Each Fourier transform is

dominated by the primary donors to the respective metals (R

+ α ∼ 1.6 Å) and one signiﬁcant long-range interaction.

Each data set was ﬁtted with a simple model of one shell of low-

Z atoms (N/O) and one long-distance metal−metal (Fe···Zn)

scattering interaction. The Zn−N/O distances (2.07 and 2.11 Å)

are fully consistent with a ﬁve-coordinate zinc(II) ion in a

nitrogen-rich environment and are in good agreement with the

average obtained by diﬀraction (2.06 Å). In contrast, the Fe···N/

O distances in a CH₃CN solution, while consistent with six-

F

DOI: 10.1021/acs.inorgchem.7b02384

Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry

Article

coordinate iron(III) in an oxygen-rich environment, are ∼0.08 Å

shorter than indicated by the crystal structure. This may reﬂect a

tendency for CH₃CN to facilitate dissociation of the terminal

hydroxyl, leaving a six-coordinate iron with one weakly bound

solvent. The iron X-ray absorption near-edge structure results

support an iron coordination number of 6, with integrated 1s →

3d areas of 9.3 × 10⁻²eV for both complexes (Figure S5).⁴⁴In

addition, and perhaps most importantly, the EXAFS data show

that the multiply bridged structure seen in the solid state is

retained in solution, with equivalent distances determined from

both the Zn and Fe K-edges, for the two complexes.

Solution Studies (ESI-MS, UV−vis, Electrochemistry,

and Potentiometric Titration). In order to establish the

catalytically active species in the hydrolysis of diester bonds, ESI-

MS, electronic spectroscopy, electrochemistry, and potentio-

metric titration studies were carried out in CH₃CN or CH₃CN/

H₂O solutions.

computational techniques.²⁰For these complexes, it has been

shown that withdrawing groups (−Br, −NO₂) tend to shift the

(pπ)phenolate → (dπ)Fe^IIILMCT band to higher-energy values,

while a bathochromic shift is observed for the donating −CH₃

group compared to −H. According to the λ_maxvalues of 1 and 2, a

positive charge on the side chain, due to the protonated amino

groups (see potentiometric titration), could shift this λ_maxvalue

to higher-energy values, which simulates the eﬀect of a

withdrawing group on the aromatic phenolate ring.

The electrochemical behavior of complexes 1 and 2 was

studied by cyclic voltammetry in an CH₃CN/H₂O (2:1, v/v)

solution (Table 3). For both complexes, the cyclic voltammo-

grams show one quasi-reversible redox process, which can be

attributed to the Fe^III/Fe^IIredox couple. Typical cyclic

voltammograms for the two complexes are shown in Figure S8.

The redox potentials of 1 (E^1/2= −189 mV) and 2 (E^1/2= −196

mV) vs NHE are comparable with those found for the [Fe^III(μ-

OH)Zn^II(L-R)] complexes (R = −CH₃, −H, −Br, −NO₂).^1,20

However, the inﬂuence of the positive charge on the amino

groups in the side chains of the modiﬁed ligands led to values that

are anodically shifted by around 50 mV compared to the E_1/2

value of complex 3. This observation is in full agreement with the

UV−vis spectra for the complexes (vide supra), which show that

the (pπ)phenolate → (dπ)Fe^IIILMCT band of 1 and 2 is shifted

to lower λ_max(higher energies) compared to that of 3.

ESI-MS analysis of complex 1 was performed in a pure

CH₃CN solution (500 nM). A group of peaks with a maximum

(100%) at m/z 517.1 with a 2+ charge was observed (Figure S6).

This signal can be assigned to the system [Fe^IIZn^II(μ-

OH)(LP1)] + OH₂+ Li⁺, suggesting a reduction in the iron(III)

center under the ESI-MS conditions. For the molecular ion of the

complex, it is reasonable to assume that two H₂O molecules are

terminally coordinated to the metal centers. Adducts formed

with some ions, such as Na⁺, K⁺, Li⁺, and NH₄, commonly occur

Equilibrium studies were carried out by the potentiometric

titration of complex 1 in an CH₃CN/H₂O (1:1) mixture.²⁸The

results indicate the neutralization of 4 mol of potassium

hydroxide per mole of complex in the pH range 3.00−12.00.

The corresponding pK_avalues are listed in Table 4, and species

+

in mass spectrometry experiments⁴⁵and possibly originate from

the glass containers used in the sample preparation or the sample

injection line. Complex 2 showed a group of peaks with a

maximum (100%) at m/z 649.2 with a 2+ charge. This peak is

assigned to the system [Fe^III(μ-OH)Zn^II(LP2)(OH₂)] + OH₂+

Na⁺, which corresponds to the molecular ion of the complex

having a H₂O molecule and a terminal hydroxo ion bound to the

metal centers, a hydration H₂O, and a sodium ion. Figure S7

shows the mass spectrum for complex 2.

Table 4. Values of the Protonation Constants for Complexes

a

1−3

pK_a1

pK_a2

pK_a3

pK_a4

1

2

3

4.21 0.03

4.27 0.08

5.07

6.85 0.02

9.09 0.02

11.39 0.04

The UV−vis absorption spectra of complexes 1−3 were

investigated in the range of 200−900 nm, using CH₃CN or H₂O/

CH₃CN (2:1, v/v) as the solvents. In pure CH₃CN solutions,

complexes 1 and 2 show strong absorption bands at λ_max, nm (ε,

M⁻¹cm⁻¹), of 512 (2100) and 510 (2300), respectively (Table

3), assigned to a ligand-to-metal charge-transfer (LMCT)

8.21

a

The data for 3 have been previously reported.

distribution graphs are shown in Figure S9. Because of low

solubility, complex 2 was studied by spectrophotometric titration

in an CH₃CN/H₂O (2:1) mixture, and only pK_a1could be

determined. Although the experimental conditions used for

complexes 1 and 2 were not strictly the same and thus we cannot

directly compare the pK_avalues, it seems reasonable to correlate

the pK_avalues among each other.

The ﬁrst protonation constant (pK_a1) may be attributed to a

H₂O molecule bound to iron(III), while the second constant

(pK_a2) is attributed to deprotonation of the H₂O molecule

bound to zinc(II). The third and fourth constants (pK_a3and

pK_a4) are assigned to the protonation/deprotonation equilibria

of two amines from the 1,4-diaminobutane group in the side

chain of complex 1, an assignment that is in agreement with the

values for free amines in solution.⁴⁷The ﬁrst and second pK_a

values of 1 are rather low for terminally coordinated H₂O

molecules and are approximately one pH unit lower than the

corresponding constants reported for complex 3 (Scheme 1 and

Figure S9). This observation may be interpreted in terms of

electrostatic interactions between the protonated amino groups

in the side chain of the complex and the iron(III)/zinc(II)-

coordinated H₂O molecules. A proposal for the protonation

equilibria observed for complex 1 is shown in Scheme 1. A similar

Table 3. Electronic Spectral and Electrochemical Data for

Complexes 1−3

λ_max, nm (ε, M⁻¹cm⁻¹

)

complex

CH₃CN

CH₃CN/H₂O (2:1)

E_1/2(mV vs NHE)

1

2

3

512 (2100)

510 (2300)

562 (2080)

506

491

−189

−196

−252

transition from the pπ orbitals of the phenolate to the dπ*

orbitals of the iron(III) ion. Similar transitions are also observed

in PAPs.⁴⁶For both complexes (as well as PAPs), a second

transition is observed as a shoulder at 330 nm, which is also

assigned to a LMCT transition involving the pπ orbitals of the

phenolate and the dσ* orbitals of the iron(III) center. Relative to

1 and 2, the lower-energy LMCT transition of 3 is shifted by

around 50 to ∼560 nm. The electronic properties of a series of

[Fe^III(μ-OH)Zn^II(L)] complexes, where the terminal phenolate

moiety was modiﬁed to contain electron-donating or -with-

drawing groups (i.e., −CH₃, −H, −Br, −NO₂), have been

studied in some detail with a variety of experimental and

G

DOI: 10.1021/acs.inorgchem.7b02384

Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry

Article

Scheme 1. Proposed Equilibrium Species for Complex 1 in a H₂O/CH₃CN (1:1) Solution

scheme is also likely for complex 2, although only pK_a1could be

determined through spectrophotometric titration because of the

low solubility of this species (Figure S10).

Reactivity Studies. The phosphodiesterase-like activity of

complexes 1 and 2 was evaluated in an CH₃CN/H₂O (2:1, v/v)

mixture using the activated substrate BDNPP under pseudo-ﬁrst-

order conditions ([S] ≫ [complex]). The reaction was

monitored spectrophotometrically at 400 nm, the absorption

maximum of the product 2,4-dinitrophenolate.²⁰The pH

dependence of the catalytic activity for both complexes was

obtained with the plot of k_cat/K_M(obtained at each pH value)

versus pH (Figure 5). At each pH value, the excess of the

substrate curve was obtained by using the initial rate method, and

it was possible to use the Michaelis−Menten equation V₀=

V_max{[S]/([S] + K_M)} to calculate the kinetic parameters k_catand

K_M. The kinetic pK_avalues were obtained from sigmoidal ﬁts of

the data for each curve at the corresponding adequate pH range.

At least two protonation equilibria are relevant to catalysis

(Figure 4 and Table 5). The optimal pH for 1 is ∼6.5, while 2

reaches its maximum activity at pH ∼8.0. While the pK_avalues

obtained via kinetic and potentiometric titration experiments for

3 are similar, the acid dissociation constants determined for 1 and

2 from kinetic assays, as far as available, are shifted to more

alkaline values compared to the values estimated from

potentiometric titrations (see Tables 4 and 5).

This observation may suggest that the presence of the BDNPP

substrate prevents interaction between the protonated amine

side chain of the ligand and the H₂O ligands coordinated to the

metal ions. This hypothesis was probed by recording the

ﬂuorescence spectra of 1 as a function of the BDNPP added

(Figure 5). The emission spectrum of 1 shows a strong band at

460 nm, which is typical for pyrene excimers, most probably due

to intermolecular interactions.⁴⁸In general, the broad emission

Figure 4. Dependence of the catalytic eﬃciency on the reaction rates at

the pH for hydrolysis of BDNPP by 1 ( ) and 2 (gray-shaded ).

●

▲

Table 5. Comparison of Values for the Potentiometric and

a

Kinetic pK_afor 1−3

potentiometric

kinetic

pK_a1

pK_a2

pK_a1

pK_a2

1

2

3

4.21 0.03

4.27 0.08

5.07

6.85 0.02

5.34 0.14

5.49 0.10

4.90

7.54 0.11

8.21

7.90

a

The data for 3 have been previously reported.

centered at 460−480 nm is extremely easy to recognize as the

result of the excimer of pyrene derivatives because monomer

H

DOI: 10.1021/acs.inorgchem.7b02384

Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry

Article

Figure 5. Quenching of the ﬂuorescence of complex 1 by successive additions of BDNPP.

ﬂuorescence takes place in the 380−420 nm wavelength range.

As can be seen in Figure 5, upon binding of the substrate, the

emission of the pyrene excimer decreases while the ﬂuorescence

of the monomer slightly increases, suggesting that the substrate

binds to the active site in the vicinity of the pyrene group, thus

aﬀecting the ﬂuorescence properties of this group but possibly

also the interactions of the protonated amino groups with the

metal-bound H₂O/OH⁻.

Similar to the assignments described in the potentiometric

titration analysis (see above), pK_a1= 5.34 obtained from the

kinetic data for 1 may be attributed to the deprotonation of the

terminal iron(III)-bound H₂O molecule and the formation of the

active species [(OH)Fe^III(μ-OH)Zn^II(OH₂)], while pK_a2is

assigned to deprotonation of the H₂O molecule bound to

zinc(II).²⁰The increase in k_catbetween pH 4.5 and 7.0 is due to

formation of the nucleophile, while the decrease between pH 7.5

and 8.5 is interpreted in terms of a reduced association

(characterized by K_ass) of the substrate [a deprotonated

zinc(II)-bound H₂O molecule may interfere with the eﬀective

substrate binding]. This interpretation is consistent with the

mechanism proposed for the related Fe^IIIZn^IIL complex.²¹The

pH dependence of 2 is more complex and resembles that of some

PAPs and related model complexes for which a minimum of

three relevant protonation equilibria have been identiﬁed. The

data were ﬁtted using a Boltzmann model, and a sigmoidal ﬁt of

the curves revealed the following pK_avalues: pK_a1= 5.32; pK_a2=

7.41; pK_a3= 8.48. The pK_a1and pK_a3values for 2 are likely to

correspond to the pK_a1and pK_a2values for 1 (Table 5). The

assignment of pK_a2for 2 remains obscure, but an attractive

suggestion invokes a role for the tertiary amine group of ligand

LP2 (Chart 1) in substrate binding. In its deprotonated (neutral)

form (i.e., above pK_a2), this residue may promote optimal

interactions between the two pyrene moieties of the ligand and

the phenol group of the substrate.

Figure 6. Dependence of the reaction rates on the concentration of

BDNPP for 1 and 2 at pH 6.5.

Table 6. Kinetic Parameters of Complexes 1−3 for BDNPP

a

Hydrolysis at pH 6.5

k_cat, ×10 ⁴s⁻¹

K_M, ×10 ⁴

M

K_ass, ×10 ⁻³M⁻¹k_cat/K_M, M⁻¹s⁻¹

1

2

3

2.10 0.14

2.90 0.29

3.80 0.10

9.0

6.60 0.34

5.10 0.18

5.5 0.11

1.5

0.32 0.34

0.56 0.62

0.67 0.61

0.25

1.9

b

c

1.8

35.0

0.27

a

b

The data for 3 have been previously reported.¹At pH 8.0.

c

Reference 1.

be seen from the data in Table 6 that, under these experimental

conditions, the substrate binding constant (K_ass) is only a little

lower than that obtained at pH 6.5, while k_catis increased,

suggesting that the maximum catalytic eﬃciency of 2 is attained

when the side-chain amine residue becomes deprotonated and

promotes optimal interactions with the substrate (vide supra).

The k_catvalues obtained for 1 and 2 at pH 6.5 are similar, but

the value for complex 3 is 3−4 times higher (Table 6). A possible

reason for this diﬀerence may originate from interaction of the

substrate BDNPP with the ligand side chain, thus increasing the

The eﬀect of the substrate concentration on the catalytic rate

at pH 6.5 reveals Michaelis−Menten-like behavior for 1 and 2.

Relevant kinetic parameters (Figure 6 and Table 6) were

obtained by ﬁtting the data to the Michaelis−Menten equation

by nonlinear regression.

The kinetic parameters for 2 were also determined at pH 8.0,

the optimum pH for this species in hydrolysis of BDNPP. As can

I

DOI: 10.1021/acs.inorgchem.7b02384

Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry

Article

K_assvalues for 1 and 2 by a factor of ∼6 compared to 3.

Nevertheless, despite the improved binding in comparison to 3

(i.e., increased K_ass= 1/K_Mvalues), the substrate may not be

adequately positioned for a nucleophilic attack by the Fe^III-OH

group in 1 and 2 (Figure 7). A comparison of the catalytic

of the activation parameters of the uncatalyzed reaction versus

that catalyzed by 3 reveals that the catalyst has only a minor

inﬂuence on ΔS^⧧of the process (similar TΔS^⧧values for the two

reactions indicate similar structural organizations of their TSs).

Therefore, the eﬀect of 3 on the process is basically enthalpic

(Table 7); it favors the reaction (k_cat) by bond making in the TS.

Table 7. Activation Parameters Obtained from Eyring Plots

ΔH^⧧(kJ mol⁻¹

)

TΔS^⧧(kJ mol⁻¹

)

ΔG^⧧(kJ mol⁻¹

)

a

H₂O

79.0

−34.4

113.8

1

2

3

48.6 0.5

48.7 0.3

58.5 0.1

−49.6 0.3

−45.8 0.2

−33.2 0.1

98.2 0.6

94.5 0.4

91.7 0.2

a

Reference 51.

In contrast, the inﬂuence of the protonated 1,4-diaminobutane

linker and the pyrene group(s) in complexes 1 and 2 is expected

to manifest itself via both enthalpic and entropic eﬀects. The

presence of an increased number of hydrogen bonds (secondary

eﬀects) is expected to enhance the stability of the TS compared

to 3 (Table 7).

However, the favorable contribution of enthalpy changes to

the reactivity of 1 and 2 are oﬀset by less favorable changes in

ΔS^⧧in comparison to 3. This may be a reﬂection of an enhanced

structural organization in the TS. A similar eﬀect has been

observed for reaction of the polyethylenimine-modiﬁed complex

FeCuL¹with the same substrate (BDNPP), in which the polymer

showed an even greater eﬀect on ΔS^⧧.⁷In summary, these results

demonstrate that adequate interactions between the ﬁrst and

second coordination spheres of the catalyst are necessary to

achieve higher catalytic activity. It is encouraging to note that the

presence of two pyrene groups in 2 reduces the negative impact

of ΔS^⧧, possibly by introducing a degree of asymmetry to the

complex/substrate in the TS.

Figure 7. Proposal for interactions between the substrate and complex

1.

eﬃciencies of 1 and 2 with that reported for a heterovalent

Fe^IIIFe^IIcomplex containing a derivative ligand of L1 with NH₂

substituents at the pyridyl groups for hydrogen bonding with

BDNPP reveals that these parameters are of the same order but

again with K_assfor 1 and 2 being 6 times higher for complexes

containing a lateral chain.⁴⁹

In order to determine whether the complexes also show

monoesterase activity, their capability of hydrolyzing DNPP, a

product of their diesterase activity, was tested. A stoichiometric

reaction (1:1 complex/substrate) was performed at 25 °C for 24

h. As observed for the FeZnL and 3 complexes, only the

diesterase activity was observed for both complexes 1 and 2.^20,21

1 and 2 showed respectively 10 and 13 turnovers in 24 h at pH

6.5 and 25 °C.

To investigate whether a general base reaction mechanism can

be attributed to 1 and 2, the eﬀect of the deuterium isotope on

the hydrolytic cleavage of BDNPP was studied under identical

conditions using either H₂O or D₂O. The values for k_H/k_Dfor

complexes 1 and 2 were respectively 1.24 and 1.12, indicating

that there is no proton transfer in the rate-limiting step of the

reaction (0.80 < k_H/k_D< 1.50), thus lending support to a

mechanism invoking an intramolecular nucleophilic attack by a

metal-bound hydroxide to initiate hydrolysis.⁵⁰

Activation Parameters. The inﬂuence of the temperature

on the hydrolysis reaction was investigated for complexes 1−3.

The experimental conditions were the same as those described

above for the pseudo-ﬁrst-order experiments, and the temper-

atures were varied from 25 to 50 °C (Figure S11). A comparison

On the basis of the combined structural and kinetic data, the

reaction mechanism shown in Scheme 2 is proposed for

hydrolysis of the phosphate diester BDNPP by 1.

First, the substrate binds monodentately to the zinc(II) ion of

the [(OH)Fe^III(μ-OH)Zn^II(OH₂)] unit by displacing its labile

H₂O ligand. Hydrogen-bonding and electrostatic interactions via

the protonated 1,4-diaminobutane linker as well as hydrophobic

interactions with the pyrene group(s) signiﬁcantly increase the

K_assocvalue of the substrate compared to complex 3. The next

step is characterized by the nucleophilic attack of the terminal

iron(III)-bound hydroxo group of the bound substrate and the

concomitant release of the 2,4-dinitrophenolate leaving group. In

this step, hydrogen-bonding formation and stabilization of the

negatively charged TS are likely because the ΔH^⧧values for 1 and

2 are decreased by ∼10 kJ mol⁻¹with respect to that of 3. Finally,

the μ-1,3-coordinated monoester product DNPP is displaced by

two H₂O molecules from the environment, and the active site is

regenerated. In conclusion, the main diﬀerence in the

mechanisms associated with 1−3 is related to the second-

coordination-sphere eﬀects, which lead to tighter substrate

binding and a higher degree of structural order in the TSs of 1

and 2) On the other hand, it is important to mention here that

the secondary eﬀects result in a decrease of the catalytic activity

(k_cat) of 1 and 2, most probably due to the strong electrostatic

interaction of BDNPP and the protonated amino group(s), and

thus the substrate may not be adequately positioned for a

nucleophilic attack by the Fe^III-OH group.

J

DOI: 10.1021/acs.inorgchem.7b02384

Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry

Article

Scheme 2. Proposed Mechanism for Hydrolysis of Phosphate Diester Promoted by Complex 1

DNA Binding Studies. The investigation of the interactions

of coordination compounds with DNA via absorption spectros-

copy generally results in hypochromism and also a bathochromic

shift in the charge-transfer bands.^48,52,53This is largely due to π-

stacking interactions between aromatic rings from the complex

and DNA base pairs. Here, binding of the potentially intercalative

complexes 1 and 2 to DNA was initially assessed through

titration, by following the changes in the absorbance intensity

and wavelength. The pyrene group has a well-known band at

around 350 nm related to a π−π* transition,⁴⁸which is

commonly used in the study of the pyrene group interaction

with DNA by spectrophotometric titration. Absorption spectra

for complexes 1 and 2 in the presence of CT-DNA are shown in

Figures S12 and S13. As the concentration of DNA was

increased, the π−π* transition bands of the pyrene group

exhibited hypochromism, indicating that the complexes bind to

CT-DNA through intercalation. On the basis of the decrease in

the absorbance values for complexes 1 and 2, their respective

binding aﬃnities (i.e., binding constants K_b) for CT-DNA could

be determined. The values obtained are reported in Table 8 and

Figures S12 and S13.

GCGCGCGCGCG-3′) were added in small portions to a

solution containing complex 1. The results show similar K_b

values for CT-DNA and the oligonucleotide pAT (AT-rich),

indicating a probable preference of pyrene-containing complexes

for A−T pair regions. For the pGC (GC-rich) oligonucleotide,

the K_bconstant values were around 25% of these values. Finally,

with the last oligonucleotide (p50AT), the constant decreased by

around 50%, as expected, because it has 50% of A−T pairs. The

binding interactions between the complexes and CT-DNA were

further examined by ITC (Table 9). These studies show positive

enthalpy terms, based on the endothermic proﬁle of the

reactions, with stoichiometries between 2 and 4 base pairs of

DNA per complex, that is, similar to those previously reported

for related systems.^54,55

The DNA binding constants of complexes with modiﬁed

ligands increase signiﬁcantly (from 1.42 × 10⁵to 1.08 × 10⁶M⁻¹)

compared to that of 3. Interestingly, the presence of two pyrene

groups (in 2) instead of only one (in 1) increases the aﬃnity by

around a factor of 2 (note that the binding constants determined

by ITC are in reasonable agreement with those obtained from the

spectrometric titrations).

An experiment using three diﬀerent oligonucleotides was

performed to identify a preference for A−T or C−G base pairs.

In a manner similar to that used for CT-DNA, three diﬀerent

oligonucleotides (pAT, 5′-CATATATATA-CCCC- TATATA-

TATG-3′; p50AT, 5′-CGCATATGGC-TTTT-GCCA-

TATGCG-3′; pGC, 5′-CGCGCGCGCGC-TTTT-

ITC measurements provide insight into the driving force for a

particular binding interaction.⁵⁶As shown in Figure 8, the

interaction between DNA and 3 is largely driven by favorable

entropy changes to overcome the large enthalpic penalty. In

contrast, the incorporation of pyrene groups in complexes 1 and

2 leads to nearly enthalpy-neutral interactions with DNA but

with considerably smaller contributions from entropy changes

compared to 3. These observations are in agreement with the TS

analysis (Table 9) and reinforce the interpretation that the

pyrene rings aﬀord an increased structural order of the catalyst/

substrate (i.e., DNA) complex.

Table 8. Binding Constants (K_b) for the Interaction of CT-

DNA, Oligonucleotides, and Complexes 1 and 2

K_b× 10⁻⁵K_b(pAT) × 10⁻⁵K_b(pGC) × 10⁻⁴

(M⁻¹(M⁻¹(M⁻¹

K_b(p50AT) ×

)

10⁻⁵(M⁻¹

1.3 0.4

)

Plasmid DNA Cleavage. Assays were performed to

determine the DNA cleavage eﬃciencies of 1 and 2 as a function

of the pH (Figure 9; the spontaneous fragmentation of DNA was

1

2

2.4 0.4

2.7 0.2 6.8 0.7

12.1 0.3

K

DOI: 10.1021/acs.inorgchem.7b02384

Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry

Article

a

Table 9. Thermodynamic Parameters K_B, ΔH, and n Directly Determined by ITC Using the Single Set of Identical Site Models

complex

n

K_B(M⁻¹

)

ΔH (kJ mol⁻¹

)

TΔS (kJ mol⁻¹

)

ΔG (kJ mol⁻¹

)

1

2

3

2.28 0.02

3.30 0.02

3.64 0.01

(6.3 0.2) × 10⁵

(1.1 0.1) × 10⁶

(1.4 0.2) × 10⁵

5.7 0.1

3.5 0.1

38.5 0.3

38.8 1.3

37.8 1.0

67.9 2.1

−33.1 0.2

−34.3 0.1

−29.4 0.1

a

The binding stoichiometry n is in terms of moles of DNA base pairs per mole of bound complex. The calculated binding free energy ΔG and

binding entropy ΔS for the reaction are also shown.

Figure 8. Enthalpic (ΔH) and entropic (−TΔS) contributions to the

Gibbs free energy (ΔG) of 1−3 binding to CT-DNA.

taken into account for the analysis, being ∼20% at pH 6.0−6.5

and negligible at other pH values). The pH proﬁle for the activity

of both complexes is bell-shaped with an optimum at pH ∼7.0

(Figure 9).

The DNA cleavage eﬃciencies of 1 and 2 were assessed in

terms of their concentrations (Figure 10). The two complexes

reveal similar patterns of concentration-dependent DNA

cleavage. In the range of 0.5−2.5 μM, 1 promotes only

conversion of the supercoiled DNA fraction (F I) to the open

circular form (F II), which represents a single-stranded scission

event. Only at concentrations above 2.5 μM does 1 convert F I to

F III, which represents a double-stranded cleavage event.

Complex 2, in contrast, is capable of double-stranded cleavage

even at 2.5 μM.

In order to assess the eﬀect of the pyrene groups in 1 and 2,

their DNA cleavage eﬃciencies were compared to that of 3. At

pH 7.0 and at a concentration of 5 μM, 3 promotes the cleavage

of only ∼30% of the plasmid DNA, which is signiﬁcantly lower

than that cleaved by 1 or 2 (∼90−100%). This observation

strongly suggests that the pyrene groups enhance DNA binding

and, consequently, also DNA cleavage

In aqueous solutions at pH 7.0, plasmid DNA is a large

polyanionic biopolymer. Under the same conditions, the amino

groups within the 1,4-diaminobutane linker in complexes 1 and 2

are expected to be positively charged. It is thus possible that

electrostatic interactions between the negatively charged DNA

and the positively charged linkers play an important role in

binding.

To probe this hypothesis, DNA cleavage reactions were

performed with increasing concentrations of NaCl (50−500

mM) to modulate the ionic strength of the reaction medium. The

increase in the ionic strength clearly reduces the activity toward

plasmid DNA for both complexes (Figure S14A,B). The same

result was obtained when assays were performed in the presence

of NaClO₄instead of NaCl (Figure S14C,D).

Because perchlorate ions are less likely to coordinate directly

to the metal ions than the chloride ion, the results in Figure S14

are in agreement with electrostatic interactions contributing to

Figure 9. pH proﬁle of plasmid DNA cleavage by 5 μM 1 (A) and 2 (B)

for 16 h at 37 °C in a 10 mM buﬀer (MES, pH 6.0−6.5; HEPES, pH

7.0−8.0; CHES, pH 9.0). Results are expressed as mean value

standard deviation obtained from three independent experiments.

the complex-mediated DNA cleavage process. However, these

results do not rule out other types of complex/DNA interactions,

such as groove binding. Thus, assays in the presence of the minor

groove binder netropsin (50 μM) or the major groove binder

methyl green (50 μM) were also performed (Figure S15). Both

groove binders diminish the complex-promoted DNA cleavage,

indicating that groove binding also contributes to the binding

interactions between the substrate and catalysts. However, no

groove binding preference or dependence was found because the

two groove binders inﬂuence the cleavage activity to the same

extent.

The mechanism of DNA strand scission was investigated by

assessing the eﬀect of ROS scavengers on the DNA cleavage

eﬃciencies of 1 and 2 (Figure S16). None of the scavengers had a

signiﬁcant eﬀect on the cleavage, suggesting that a hydrolytic

L

DOI: 10.1021/acs.inorgchem.7b02384

Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry

Article

(k_obs= 0.215 h⁻¹; t_1/2= ∼3.2 h). Compared to 3 (k_obs= 0.019 h⁻¹;

t

_1/2= ∼36.3 h), the pyrene-modiﬁed systems 1 and 2 exhibit 11-

and 19-fold higher activity, respectively. The combined results

for the reactivity and mechanism of DNA cleavage suggest that

the pyrene groups do not aﬀect the mechanism of hydrolysis but

instead enhance the binding of the catalysts. Similar results were

observed for a diiron(III) complex, with the symmetrical

dinucleating ligand 2,6-bis[(2-hydroxybenzyl)(2-

pyridylmethyl)aminomethyl]-4-methylphenol (H₃L4) contain-

ing two linked acridine groups¹⁵or a pyrene.⁵⁸

In an attempt to evaluate the binding preferences in terms of

nucleotide sequence or structural DNA motifs, DNase I

footprinting assays were performed with (1) − (3). It should

be noted that none of the complexes was able to cleave the

oligonucleotide substrate directly, and thus we were unable to

detect cleavage of the ﬂuorescent probe to establish the location

of the target site (data not shown). On the other hand, DNase I

protection assays were successful, as shown in Figure S17.

DNase I Footprinting. DNase I footprinting has been

routinely used as a reliable assay to determine the DNA binding

of several small organic compounds.⁵⁹Although few studies on

the DNase I footprinting of metal compounds have been

reported,⁶⁰⁻⁶²this methodology can be used to identify the DNA

binding sites of proteins and small molecules. In general, our

DNase I footprinting assays revealed that the three complexes

analyzed have distinct binding behaviors (Figure S17). Complex

3 binds preferentially at an AT-rich region covering four

nucleotides from T9 to A12 (A, TTAA). Two other weak

binding sites are located at C16−C17 (B) and C23−T24 (C).

Complex 1 showed additional binding sites including an

expanded binding site A up to T13 (TTAAT) and a fourth

binding site (D, T41−A42, and A44) that is A, complementary to

the A binding site, with an oﬀset between + and − strands of

three base pairs, which could be indicative of minor groove

contacts. Finally, the complex 2 binding site is similar to that of 1,

with the addition of an expanded binding site B including G15

and G18−C20 (E, GCCGGC). The proposed binding sites for

each complex are summarized in Figure 12. The presence of

Figure 10. Complex concentration dependence of plasmid DNA

cleavage by 1 (A) and 2 (B) for 16 h at 37 °C in 10 mM HEPES buﬀer

(pH 7.0).

rather than oxidative pathway is associated with the ability of the

complexes to promote DNA cleavage. The DNA cleavage

mechanism was also investigated by monitoring the catalytic

parameters of the reaction (Figure 11). Assays were conducted

under pseudo-Michaelis−Menten conditions as described

previously.⁵⁷Complexes 1 and 2 induce DNA precipitation at

concentrations higher than 10 μM (data not shown). Because of

this limitation, the concentration of the complexes used in the

experiments was 5 μM (Figure S16). Complex 2 is the most

eﬃcient system (k_obs= 0.355 h⁻¹; t_1/2= ∼1.9 h), followed by 1

Figure 12. Proposed DNA binding sites of 3, 1, and 2 within the 49-mer

oligonucleotide. Note that E refers to the expanded binding site B

including a G15 and G18−C20. The nucleotides in red refer to the

DNase I cleavage sites protected due to complex binding.

pyrene motifs in 1 and 2 apparently enhances the binding aﬃnity

of these compounds to DNA, as evidenced by the increase in the

apparent binding strength for sites A/D and B (empirically

evidenced by a strong inhibition of DNase I cleavage at these

sites), and also expands the range and number of binding sites

(Figure S17). Furthermore, the addition of pyrene increases the

structural size of the complexes, aﬀecting the number of

nucleotides that can interact with 1 and 2 inside each binding

site. Because pyrene behaves as a general intercalator, it can act as

Figure 11. Kinetics of plasmid DNA cleavage by 5 μM 1−3 monitored

over 8 h at 37 °C in a 10 mM HEPES buﬀer (pH 7.0). Measurements

were carried out in duplicate.

M

DOI: 10.1021/acs.inorgchem.7b02384

Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry

Article

a “chemical anchor”, allowing 1 and 2 to interact with additional

vicinal nucleotides.

Cell Cytotoxicity. Both complexes 1 and 3 are able to inhibit

cellular growth in a concentration-dependent manner, with the

IC₅₀values indicated in Table 10. Apoptosis, a form of

Confocal microscopic analysis of the nuclear epidermal growth

factor receptor (EGFR) with nuclei of ameloblast-like cells

(Figure 14) shows that complex 1 displays excellent colocaliza-

Table 10. IC₅₀for Complexes 1 and 3

IC₅₀

complex

72 h

24 h

1

3

2.76 0.30

8.80 0.90

6.52 0.60

18.76 1.70

a

IC₅₀is the concentration (μM) required to inhibit 50% of K562 cell

growth. The values are the mean of triplicate determinations.

programmed cell death, is an important determinant of the

response of tumor cells to chemotherapeutic agents. In apoptotic

cells, phosphatidylserine is translocated from the inner to the

outer leaﬂet of the plasma membrane. Annexin V can be used to

identify apoptotic cells because it binds to exposed phosphati-

dylserine on the outer leaﬂet. Propidium iodide, a nucleic acid

binding dye, is impermeable to live and apoptotic cells but stains

dead cells with red ﬂuorescence. The combination of the two

dyes allows apoptotic cells (annexin V positive and propidium

iodide negative), necrotic cells (double labeled with annexin V

and propidium iodide), and viable cells (unmarked) to be

diﬀerentiated.

At concentrations between 0.5 and 20 μM, complexes 1 and 3

induce cellular apoptosis in a concentration-dependent manner

(Figure S18). The apoptotic eﬀects of these two complexes are

more pronounced than that of cisplatin. Following treatment

with 10 μmol/L complex 1, approximately 60% of the cells

undergo apoptosis compared with 11% for complex 3 and only

5.3% for cisplatin. At the highest concentration tested (20 μM),

only ∼5% of the cells treated with complex 1 are viable, whereas

∼50% of the cells treated with complex 3 remain unmarked

(Figure S19). These results are in agreement with the IC₅₀values

determined in the cytotoxic assays. Furthermore, the necrotic

eﬀect of complex 3 and cisplatin is more pronounced than that of

complex 1, as shown by the percentage of cells stained with

propidium iodide.

Figure 14. Accumulation in lysosomes. Confocal microscopic analysis

showing colocalization of complex 1 with the lysotracker. Complex 1 is

pseudocolored in green and the lysotracker in red. Incubation time = 5 h,

and [1] = 80 μM. Original magniﬁcation: 63×.

tion with lysosomes. Lysosomes, which are cellular organelles

found in all animal cells, play key roles in cellular metabolism,

endocytosis, and the synthesis or assembly of hydrolases involved

in macromolecule digestion. Various proteases regulate

lysosomal traﬃcking and the intralysosomal pH.^49,63

Lysosome dysfunction has been implicated in the invasion,

metastasis, and maintenance of cancerous tissues.⁶⁴Lysosome

membrane permeabilization can induce cell death because the

release of lysosomal proteases in the cytosol causes the digestion

of vital proteins and the activation of additional hydrolases

including caspases.^65,66Partial lysosomal permeabilization occurs

in several models of apoptosis, a phenomenon known as the

“lysosomal pathway of apoptosis”. A large number of potential

anticancer drugs induce p53-independent apoptosis mediated by

partial lysosomal permeabilization, suggesting that these path-

ways are potential targets for anticancer drug development.⁶⁷

Platinum compounds with antitumor properties, such as a

ﬂuorophore-labeled cisplatin, were also localized in lysosomal

compartments.^68,69A series of luminescent platinum(II)

complexes containing pincer-type ligands with pyrazole moieties

also accumulates in the lysosomes of cancer cells, increasing the

lysosomal membrane permeability and inducing cell death.⁷⁰

Thus, the enhanced apoptotic eﬀect of 1, together with its

colocalization in lysosomes, indicates that this compound is an

excellent lead for novel anticancer drugs.

The strong ﬂuorescence of complex 1 allowed an assessment

of its intracellular location. Propidium iodide was used as a

nucleic acid binding dye and lysotracker as a lysosome dye. The

localization of 1 does not coincide with that of propidium iodide

(last column in Figure 13), which means that the compound does

not accumulate in the nucleus.

CONCLUSIONS

■

In summary, we have synthesized and characterized two new

dinuclear Fe^IIIZn^IIhydrolase biomimetics 1 and 2, in which the

carbonyl group in the ligand L1 was modiﬁed to contain one

(H₂LP1) or two (H₂LP2) pyrene groups conjugated to the

dinucleating ligand by a 1,4-diaminobutane spacer. The

diesterase activity (using the model substrate BDNPP and

DNA) and cytotoxic activity (estimated by the corresponding

IC₅₀values) of 1 and 2 are signiﬁcantly increased in comparison

to their parental counterpart, i.e., the [Fe^IIIZn^IIL1] (3) complex,

because of the presence of protonated amines and conjugated

pyrene group(s), which exert second-coordination-sphere eﬀects

on the catalytic eﬃciency of these species. Interestingly, confocal

ﬂuorescence microscopy analysis of the nuclear EGFR with

nuclei of ameloblast-like cells shows that complex 1 displays

excellent colocalization with lysosomes and does not seem to

Figure 13. Cellular localization of compound 1 observed by

ﬂuorescence microscopy. The complex is pseudocolored in blue and

the nuclear dye, propidium iodide, in red. In the last column (Merge),

one can see that the localization of the compound does not coincide with

that of the dye.

N

DOI: 10.1021/acs.inorgchem.7b02384

Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry

Article

́

(9) Kovari, E.; Kramer, R. Rapid Phosphodiester Hydrolysis by an

̈

accumulate in the nucleus of the cell. Further studies involving

variations in the side-chain length (ethyl, propyl, and hexyl), as

well as in the intercalating group are underway and will be the

subject of further reports.

Ammonium-Functionalized Copper(II) Complex. A Model for the

Cooperativity of Metal Ions and NH-Acidic Groups in Phosphoryl

Transfer Enzymes. J. Am. Chem. Soc. 1996, 118, 12704−12709.

(10) Comba, P.; Gahan, L. R.; Hanson, G. R.; Mereacre, V.; Noble, C.

J.; Powell, A. K.; Prisecaru, I.; Schenk, G.; Zajaczkowski-Fischer, M.

Monoesterase Activity of a Purple Acid Phosphatase Mimic with a

Cyclam Platform. Chem. - Eur. J. 2012, 18, 1700−1710.

ASSOCIATED CONTENT

* Supporting Information

The Supporting Information is available free of charge on the

ACS Publications website at DOI: 10.1021/acs.inorg-

chem.7b02384.

■

S

(11) Kendall, A. J.; Zakharov, L. N.; Gilbertson, J. D. Synthesis and

Stabilization of a Monomeric Iron(II) Hydroxo Complex via Intra-

molecular Hydrogen Bonding in the Secondary Coordination Sphere.

Inorg. Chem. 2010, 49, 8656−8658.

(12) Basile, L. A.; Barton, J. K. Design of a double-stranded DNA

cleaving agent with two polyamine metal-binding arms: Ru-

(DIP)₂Macron⁺. J. Am. Chem. Soc. 1987, 109, 7548−7550.

(13) Rammo, J.; Hettich, R.; Roigk, A.; Schneider, H.-J. Catalysis of

DNA cleavage by lanthanide complexes with nucleophilic or

intercalating ligands and their kinetic characterization. Chem. Commun.

1996, 1, 105−107.

(14) Sissi, C.; Rossi, P.; Felluga, F.; Formaggio, F.; Palumbo, M.;

Tecilla, P.; Toniolo, C.; Scrimin, P. Dinuclear Zn²⁺Complexes of

Synthetic Heptapeptides as Artificial Nucleases. J. Am. Chem. Soc. 2001,

123, 3169−3170.

(15) Chen, X.; Fan, J.; Peng, X.; Wang, J.; Sun, S.; Zhang, R.; Wu, T.;

Zhang, F.; Liu, J.; Wang, F.; Ma, S. Bisintercalator-containing dinuclear

iron(III) complex: An efficient artificial nuclease. Bioorg. Med. Chem.

Lett. 2009, 19, 4139−4142.

(16) Boseggia, E.; Gatos, M.; Lucatello, L.; Mancin, F.; Moro, S.;

Palumbo, M.; Sissi, C.; Tecilla, P.; Tonellato, U.; Zagotto, G. Toward

Efficient Zn(II)-Based Artificial Nucleases. J. Am. Chem. Soc. 2004, 126,

4543−4549.

(17) Gahan, L. R.; Smith, S. J.; Neves, A.; Schenk, G. Phosphate Ester

Hydrolysis: Metal Complexes As Purple Acid Phosphatase and

Phosphotriesterase Analogues. Eur. J. Inorg. Chem. 2009, 2009, 2745−

2758.

(18) Xavier, F. R.; Peralta, R. A.; Bortoluzzi, A. J.; Drago, V.; Castellano,

E. E.; Haase, W.; Tomkowicz, Z.; Neves, A. Bioinspired Fe^IIICd^IIand

Fe^IIIHg^IIcomplexes: Synthesis, characterization and promiscuous

catalytic activity evaluation. J. Inorg. Biochem. 2011, 105, 1740−1752.

(19) Schenk, G.; Peralta, R.; Batista, S.; Bortoluzzi, A.; Szpoganicz, B.;

Dick, A.; Herrald, P.; Hanson, G.; Szilagyi, R.; Riley, M.; Gahan, L.;

Neves, A. Probing the role of the divalent metal ion in uteroferrin using

metal ion replacement and a comparison to isostructural biomimetics.

JBIC, J. Biol. Inorg. Chem. 2007, 13, 139−155.

(20) Peralta, R. A.; Bortoluzzi, A. J.; de Souza, B.; Jovito, R.; Xavier, F.

R.; Couto, R. A. A.; Casellato, A.; Nome, F.; Dick, A.; Gahan, L. R.;

Schenk, G.; Hanson, G. R.; de Paula, F. C. S.; Pereira-Maia, E. C.; de P.

Machado, S.; Severino, P. C.; Pich, C.; Bortolotto, T.; Terenzi, H.;

Castellano, E. E.; Neves, A.; Riley, M. J. Electronic Structure and

Spectro-Structural Correlations of Fe^IIIZn^IIBiomimetics for Purple Acid

Phosphatases: Relevance to DNA Cleavage and Cytotoxic Activity.

Inorg. Chem. 2010, 49, 11421−11438.

Synthesis of the ligands H₂LP1 and H₂LP2 (Schemes S1

and S2), characterization of the ligands and their

complexes 1 and 2 (Figures S1−S10), kinetics (Figure

S11), DNA interaction of 1 and 2 (Figures S12−S17), and

cell interaction (Figures S18 and S19) (PDF)

AUTHOR INFORMATION

Corresponding Authors

*E-mail: ademir.neves@ufsc.br. Phone: +55 (48) 3721-3605.

*E-mail: hernan.terenzi@ufsc.br.

■

ORCID

Ademir Neves: 0000-0002-6713-798X

Rosely A. Peralta: 0000-0002-6417-2683

Author Contributions

The manuscript was written with contributions from all authors.

All authors have approved the ﬁnal version of the manuscript.

Notes

The authors declare no competing ﬁnancial interest.

ACKNOWLEDGMENTS

■

Financial support was received from CNPq, CAPES, FAPESC,

and INCT-catalise (Brazil). G.S. acknowledges the Australian

Research Council for the award of a Future Fellowship

(FT120100694).

REFERENCES

■

(1) 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. Heterodinuclear Fe^IIIZn^II-Bioinspired Complex Supported on

3-Aminopropyl Silica. Efficient Hydrolysis of Phosphate Diester Bonds.

Inorg. Chem. 2010, 49, 2580−2582.

(2) Desbouis, D.; Troitsky, I. P.; Belousoff, M. J.; Spiccia, L.; Graham,

B. Copper(II), zinc(II) and nickel(II) complexes as nuclease mimetics.

Coord. Chem. Rev. 2012, 256, 897−937.

(3) Mancin, F.; Scrimin, P.; Tecilla, P. Progress in artificial

metallonucleases. Chem. Commun. 2012, 48, 5545−5559.

́

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

Schenk, G. The Catalytic Mechanisms of Binuclear Metallohydrolases.

Chem. Rev. 2006, 106, 3338−3363.

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

Casellato, A.; Castellano, E. E.; Herrald, P.; Riley, M. J.; Schenk, G. An

Unprecedented Fe^III(μ-OH)Zn^IIComplex that Mimics the Structural

and Functional Properties of Purple Acid Phosphatases. J. Am. Chem.

Soc. 2007, 129, 7486−7487.

(5) Zhao, M.; Wang, H.-B.; Ji, L.-N.; Mao, Z.-W. Insights into

metalloenzyme microenvironments: biomimetic metal complexes with a

functional second coordination sphere. Chem. Soc. Rev. 2013, 42, 8360−

8375.

(22) Smith, S. J.; Peralta, R. A.; Jovito, R.; Horn, A.; Bortoluzzi, A. J.;

Noble, C. J.; Hanson, G. R.; Stranger, R.; Jayaratne, V.; Cavigliasso, G.;

Gahan, L. R.; Schenk, G.; Nascimento, O. R.; Cavalett, A.; Bortolotto,

T.; Razzera, G.; Terenzi, H.; Neves, A.; Riley, M. J. Spectroscopic and

Catalytic Characterization of a Functional Fe^IIIFe^IIBiomimetic for the

Active Site of Uteroferrin and Protein Cleavage. Inorg. Chem. 2012, 51,

2065−2078.

(23) 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. A new

heterobinuclear Fe^IIICu^IIcomplex with a single terminal Fe^III−

(6) Rebilly, J.-N.; Colasson, B.; Bistri, O.; Over, D.; Reinaud, O.

Biomimetic cavity-based metal complexes. Chem. Soc. Rev. 2015, 44,

467−489.

(7) de Souza, B.; Kreft, G. L.; Bortolotto, T.; Terenzi, H.; Bortoluzzi, A.

J.; Castellano, E. E.; Peralta, R. A.; Domingos, J. B.; Neves, A. Second-

Coordination-Sphere Effects Increase the Catalytic Efficiency of an

Extended Model for Fe^IIIM^IIPurple Acid Phosphatases. Inorg. Chem.

2013, 52, 3594−3596.

(8) de Souza, B.; Heying, R.; Bortoluzzi, A. J.; Domingos, J. B.; Neves,

A. The effect of chain size on the modeling of second sphere effects in

biomimetic complexes. J. Mol. Catal. A: Chem. 2015, 397, 76−84.

O

DOI: 10.1021/acs.inorgchem.7b02384

Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry

Article

O(phenolate) bond. Relevance to purple acid phosphatases and

nucleases. JBIC, J. Biol. Inorg. Chem. 2005, 10, 319−332.

́

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

tumoral: Cytotoxicity and DNA Cleavage. Inorg. Chem. 2011, 50, 6414−

6424.

(44) Roe, A. L.; Schneider, D. J.; Mayer, R. J.; Pyrz, J. W.; Widom, J.;

Que, L. X-ray absorption spectroscopy of iron-tyrosinate proteins. J. Am.

Chem. Soc. 1984, 106, 1676−1681.

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

reaction mechanism of the Ga(III)Zn(II) derivative of uteroferrin and

corresponding biomimetics. JBIC, J. Biol. Inorg. Chem. 2007, 12, 1207−

1220.

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

J. G.; Smith, J. A.; Struhl, K. Short Protocols in Molecular Biology: A

Compendium of Methods from Current Protocols in Molecular Biology, 4th

ed.; Wiley: New York, 1999.

(45) Kruve, A.; Kaupmees, K.; Liigand, J.; Oss, M.; Leito, I. Sodium

adduct formation efficiency in ESI source. J. Mass Spectrom. 2013, 48,

695−702.

́

(46) Schenk, G.; Mitic, N.; Hanson, G. R.; Comba, P. Purple acid

phosphatase: A journey into the function and mechanism of a colorful

enzyme. Coord. Chem. Rev. 2013, 257, 473−482.

(47) Cox, B. G. Acids and Bases: Solvent Eﬀects on Acid−base strength;

Oxford University Press: Oxford, U.K., 2013.

(26) Rock, P. A. The Standard Oxidation Potential of the

Ferrocyanide-Ferricyanide Electrode at 25° and the Entropy of

Ferrocyanide Ion. J. Phys. Chem. 1966, 70, 576−580.

(48) Winnik, F. M. Photophysics of preassociated pyrenes in aqueous

polymer solutions and in other organized media. Chem. Rev. 1993, 93,

587−614.

́

́ ̌

(27) Strohalm, M.; Kavan, D.; Novak, P.; Volny, M.; Havlícek, V.

mMass 3: A Cross-Platform Software Environment for Precise Analysis

of Mass Spectrometric Data. Anal. Chem. 2010, 82, 4648−4651.

(28) Barbosa, J.; Sanz-Nebot, V. Autoprotolysis constants and

standardization of the glass electrode in acetonitrile-water mixtures.

Effect of solvent composition. Anal. Chim. Acta 1991, 244, 183−191.

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

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

(49) Comba, P.; Gahan, L. R.; Mereacre, V.; Hanson, G. R.; Powell, A.

K.; Schenk, G.; Zajaczkowski-Fischer, M. Spectroscopic Character-

ization of the Active Fe^IIIFe^IIIand Fe^IIIFeII Forms of a Purple Acid

Phosphatase Model System. Inorg. Chem. 2012, 51, 12195−12209.

(50) Deal, K. A.; Hengge, A. C.; Burstyn, J. N. Characterization of

Transition States in Dichloro(1,4,7-triazacyclononane)copper(II)-

Catalyzed Activated Phosphate Diester Hydrolysis. J. Am. Chem. Soc.

1996, 118, 1713−1718.

(30) George, G. N. EXAFSPAK and EDG-FIT; Stanford Synchrotron

Radiation Laboratory: Stanford, CA, 2000.

(51) Bunton, C. A.; Farber, S. J. Hydrolysis of bis(2,4-dinitrophenyl)

phosphate. J. Org. Chem. 1969, 34, 767−772.

(31) Rehr, J. J.; Kas, J. J.; Vila, F. D.; Prange, M. P.; Jorissen, K.

Parameter-free calculations of X-ray spectra with FEFF9. Phys. Chem.

Chem. Phys. 2010, 12, 5503−5513.

(32) Tierney, D. L.; Schenk, G. X-Ray Absorption Spectroscopy of

Dinuclear Metallohydrolases. Biophys. J. 2014, 107, 1263−1272.

(33) Wiseman, T.; Williston, S.; Brandts, J. F.; Lin, L.-N. Rapid

measurement of binding constants and heats of binding using a new

titration calorimeter. Anal. Biochem. 1989, 179, 131−137.

(52) Sunita, M.; Anupama, B.; Ushaiah, B.; Gyana Kumari, C.

Synthesis, characterization, DNA binding and cleavage studies of mixed-

ligand copper (II) complexes. Arabian J. Chem. 2017, 10, S3367−S3374.

(53) Pages, B. J.; Ang, D. L.; Wright, E. P.; Aldrich-Wright, J. R. Metal

complex interactions with DNA. Dalton Trans. 2015, 44, 3505−3526.

(54) Haq, I.; Lincoln, P.; Suh, D.; Norden, B.; Chowdhry, B. Z.;

Chaires, J. B. Interaction of.DELTA.- and.LAMBDA.-[Ru-

(phen)₂DPPZ]²⁺with DNA: A Calorimetric and Equilibrium Binding

Study. J. Am. Chem. Soc. 1995, 117, 4788−4796.

(34) Holde, K. E. v.; Johnson, C.; Ho, P. S. Principles of Physical

Biochemistry, 2nd ed.; Prentice Hall: Upper Saddle River, NJ, 2006.

(35) Grossoehme, N.; Spuches, A.; Wilcox, D. Application of

isothermal titration calorimetry in bioinorganic chemistry. JBIC, J.

Biol. Inorg. Chem. 2010, 15, 1183−1191.

(55) Li, L.-Y.; Jia, H.-N.; Yu, H.-J.; Du, K.-J.; Lin, Q.-T.; Qiu, K.-Q.;

Chao, H.; Ji, L.-N. Synthesis, characterization, and DNA-binding studies

of ruthenium complexes [Ru(tpy)(ptn)]²⁺and Ru(dmtpy)(ptn)]²⁺. J.

Inorg. Biochem. 2012, 113, 31−39.

(56) Kuntz, I. D.; Chen, K.; Sharp, K. A.; Kollman, P. A. The maximal

affinity of ligands. Proc. Natl. Acad. Sci. U. S. A. 1999, 96, 9997−10002.

(57) Sreedhara, A.; Freed, J. D.; Cowan, J. A. Efficient Inorganic

Deoxyribonucleases. Greater than 50-Million-Fold Rate Enhancement

in Enzyme-Like DNA Cleavage. J. Am. Chem. Soc. 2000, 122, 8814−

8824.

́

(36) Pedroso, M. M.; Ely, F.; Mitic, N.; Carpenter, M. C.; Gahan, L. R.;

Wilcox, D. E.; Larrabee, J. L.; Ollis, D. L.; Schenk, G. Comparative

investigation of the reaction mechanisms of the organophosphate-

degrading phosphotriesterases from Agrobacterium radiobacter

(OpdA) and Pseudomonas diminuta (OPH). JBIC, J. Biol. Inorg.

Chem. 2014, 19, 1263−1275.

(37) Cooper, A.; Johnson, C. In Microscopy, Optical Spectroscopy, and

Macroscopic Techniques; Jones, C., Mulloy, B., Thomas, A., Eds.;

Humana Press, 1994; Vol. 22, Chapter 9, pp 109−124.

(38) Ladbury, J. E.; Chowdhry, B. Z. Sensing the heat: the application

of isothermal titration calorimetry to thermodynamic studies of

biomolecular interactions. Chem. Biol. 1996, 3, 791−801.

(58) Chen, X. Q.; Wang, J. Y.; Zhang, T. Y.; Zhang, L. Z.; Peng, X. J.

Synthesis and DNA cleavage activity of diiron(III) complex bearing

pyrene group. Chin. Chem. Lett. 2008, 19, 342−344.

(59) Bailly, C.; Denny, W. A.; Mellor, L. E.; Wakelin, L. P. G.; Waring,

M. J. Sequence specificity of the binding of 9-aminoacridine- and

amsacrine-4-carboxamides to DNA studied by DNase I footprinting.

Biochemistry 1992, 31, 3514−3524.

(39) Mancin, F.; Scrimin, P.; Tecilla, P.; Tonellato, U. Artificial

metallonucleases. Chem. Commun. 2005, 2540−2548.

(40) Muxel, A. A.; Neves, A.; Camargo, M. A.; Bortoluzzi, A. J.;

Szpoganicz, B.; Castellano, E. E.; Castilho, N.; Bortolotto, T.; Terenzi,

H. New La(III) Complex Immobilized on 3-Aminopropyl-Function-

alized Silica as an Efficient and Reusable Catalyst for Hydrolysis of

Phosphate Ester Bonds. Inorg. Chem. 2014, 53, 2943−2952.

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

Potassium monopersulfate and a water-soluble manganese porphyrin

complex, [Mn(TMPyP)](OAc)₅, as an efficient reagent for the oxidative

cleavage of DNA. Biochemistry 1989, 28, 7268−7275.

(42) Van Dyke, M. W.; Hertzberg, R. P.; Dervan, P. B. Map of

distamycin, netropsin, and actinomycin binding sites on heterogeneous

DNA: DNA cleavage-inhibition patterns with methidiumpropyl-

EDTA.Fe(II). Proc. Natl. Acad. Sci. U. S. A. 1982, 79, 5470−5474.

(43) Silva, P. P.; Guerra, W.; Silveira, J. N.; Ferreira, A. M. d. C.;

Bortolotto, T.; Fischer, F. L.; Terenzi, H.; Neves, A.; Pereira-Maia, E. C.

Two New Ternary Complexes of Copper(II) with Tetracycline or

Doxycycline and 1,10-Phenanthroline and Their Potential as Anti-

(60) Ruba, E.; Hart, J. R.; Barton, J. K. [Ru(bpy) (L)]Cl : Luminescent

̈

2

Metal Complexes That Bind DNA Base Mismatches. Inorg. Chem. 2004,

43, 4570−4578.

(61) Kikuta, E.; Aoki, S.; Kimura, E. New potent agents binding to a

poly(dT) sequence in double-stranded DNA: bis(Zn²⁺-cyclen) and

tris(Zn²⁺-cyclen) complexes. JBIC, J. Biol. Inorg. Chem. 2002, 7, 473−

482.

́ ́ ̌ ́ ́

(62) Chvalova, K.; Kasparkova, J.; Farrell, N.; Brabec, V.

Deoxyribonuclease I footprinting reveals different DNA binding

modes of bifunctional platinum complexes. FEBS J. 2006, 273, 3467−

3478.

(63) Yapici, N. B.; Bi, Y.; Li, P.; Chen, X.; Yan, X.; Mandalapu, S. R.;

Faucett, M.; Jockusch, S.; Ju, J.; Gibson, K. M.; Pavan, W. J.; Bi, L. Highly

Stable and Sensitive Fluorescent Probes (LysoProbes) for Lysosomal

Labeling and Tracking. Sci. Rep. 2015, 5, 8576.

(64) Koblinski, J. E.; Ahram, M.; Sloane, B. F. Unraveling the role of

proteases in cancer. Clin. Chim. Acta 2000, 291, 113−135.

P

DOI: 10.1021/acs.inorgchem.7b02384

Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry

Article

(65) Guicciardi, M. E.; Leist, M.; Gores, G. J. Lysosomes in cell death.

Oncogene 2004, 23, 2881−2890.

(66) Boya, P.; Kroemer, G. Lysosomal membrane permeabilization in

cell death. Oncogene 2008, 27, 6434−6451.

(67) Erdal, H.; Berndtsson, M.; Castro, J.; Brunk, U.; Shoshan, M. C.;

Linder, S. Induction of lysosomal membrane permeabilization by

compounds that activate p53-independent apoptosis. Proc. Natl. Acad.

Sci. U. S. A. 2005, 102, 192−197.

(68) Molenaar, C.; Teuben, J.-M.; Heetebrij, R. J.; Tanke, H. J.;

Reedijk, J. New insights in the cellular processing of platinum antitumor

compounds, using fluorophore-labeled platinum complexes and digital

fluorescence microscopy. JBIC, J. Biol. Inorg. Chem. 2000, 5, 655−665.

(69) Safaei, R.; Katano, K.; Larson, B. J.; Samimi, G.; Holzer, A. K.;

Naerdemann, W.; Tomioka, M.; Goodman, M.; Howell, S. B.

Intracellular Localization and Trafficking of Fluorescein-Labeled

Cisplatin in Human Ovarian Carcinoma Cells. Clin. Cancer Res. 2005,

11, 756−767.

(70) Tsai, J. L.-L.; Zou, T.; Liu, J.; Chen, T.; Chan, A. O.-Y.; Yang, C.;

Lok, C.-N.; Che, C.-M. Luminescent platinum(ii) complexes with self-

assembly and anti-cancer properties: hydrogel, pH dependent emission

color and sustained-release properties under physiological conditions.

Chem. Sci. 2015, 6, 3823−3830.

Q

DOI: 10.1021/acs.inorgchem.7b02384

Inorg. Chem. XXXX, XXX, XXX−XXX

Article Doi

Second-Sphere Effects in Dinuclear FeIIIZnII Hydrolase Biomimetics: Tuning Binding and Reactivity Properties

DOI: 10.1021/acs.inorgchem.7b02384

Source and publish data:

Authors:

Article abstract of DOI:10.1021/acs.inorgchem.7b02384

Full text of DOI:10.1021/acs.inorgchem.7b02384

Products guided by the article

R&D Labs maybe for 1217342-27-2

Relevant to this article

Hot Product