Three new dinuclear nickel(II) complexes with amine pendant-armed ligands: Characterization, DFT study, antibacterial and hydrolase-like activity

Article abstract of DOI:10.1016/j.ica.2020.119559

Herein, we report the synthesis and characterization of three new Ni(II) complexes, namely: [Ni₂(H₂LEt)(μ-OAc)₂(H₂O)]BPh₄?ClO₄ (1), (H₂LEt = 2-[(N-benzyl-N-2-pyridyl methylamine)]-4-methyl-6-[N-2(pyridylmethyl)aminomethyl)]-6((2-aminoethyl)amino)methyl phenol); [Ni₂(H₂LProp)(μ-OAc)₂(H₂O)](ClO₄)₂ (2), (H₂LProp = 2-[(N-benzyl-N-2-pyridylmethylamine)]-4-methyl-6-[N-(2-pyridylmethyl)aminomethyl]-6-((2-aminopropyl) amino)methylphenol); and [Ni₂(LBut)(μ-OAc)₂(H₂O)](HCl)₂ (3), (LBut = 2-[(N-benzyl-N-2-pyridylmethylamine)]-4-methyl-6-[N-(2-pyridylmethyl)aminomethyl]-6-((2-aminobuthyl) amino)methylphenol). All of them were characterized through spectroscopic techniques (elemental analysis, IR, UV–Vis spectroscopy), ESI-MS, electrochemistry and potentiometric titration. Density functional theory (DFT) was used to better understand the electronic and molecular structure of these complexes. The hydrolytic activity of complexes 1–3 towards the 2,4-BDNPP substrate was analyzed and complex 2 presented the highest catalytic efficiency (k_cat/K_M) of the three, possibly due to a greater interaction with the substrate. The complexes were also screened for their antibacterial activities using both Gram-positive and Gram-negative bacterial strains by minimum inhibitory concentration and minimum bactericidal concentration methods.

Full text of DOI:10.1016/j.ica.2020.119559

Journal Pre-proofs
Research paper
Three new dinuclear nickel(II) complexes with amine pendant-armed ligands:
characterization, DFT study, antibacterial and hydrolase-like activity
Cláudia C.V. Chaves, Giliandro Farias, Maíra D. Formagio, Ademir Neves,
Rosane M. Peralta, Jane M.G. Mikcha, Bernardo de Souza, Rosely A. Peralta
PII:
DOI:
Reference:
S0020-1693(20)30153-5
https://doi.org/10.1016/j.ica.2020.119559
ICA 119559
To appear in:
Inorganica Chimica Acta
Received Date:
Revised Date:
Accepted Date:
27 January 2020
25 February 2020
27 February 2020
Please cite this article as: C.C.V. Chaves, G. Farias, M.D. Formagio, A. Neves, R.M. Peralta, J.M.G. Mikcha, B.
de Souza, R.A. Peralta, Three new dinuclear nickel(II) complexes with amine pendant-armed ligands:
characterization, DFT study, antibacterial and hydrolase-like activity, Inorganica Chimica Acta (2020), doi:
https://doi.org/10.1016/j.ica.2020.119559
This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover
page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version
will undergo additional copyediting, typesetting and review before it is published in its final form, but we are
providing this version to give early visibility of the article. Please note that, during the production process, errors
may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
© 2020 Published by Elsevier B.V.

Three new dinuclear nickel(II) complexes with amine pendant-armed ligands:
characterization, DFT study, antibacterial and hydrolase-like activity
Cláudia C. V. Chavesa, Giliandro Fariasa, Maíra D. Formagioc, Ademir Nevesa, Rosane M.
Peralta,b, Jane M. G. Mikchac, Bernardo de Souzaa, Rosely A. Peralta*a
aDepartamento de Química, Universidade Federal de Santa Catarina, Florianópolis, Santa
Catarina, 88040-900, Brazil
bDepartamento de Bioquímica, Universidade Estadual de Maringá, Maringá, Paraná, 87015-900,
Brazil
cDepartamento de Análises Clínicas e Biomedicina, Universidade Estadual de Maringá, Maringá,
Paraná, 87020-900, Brasil
Corresponding author: Rosely A. Peralta
email:rosely.peralta@ufsc.br
KEYWORDS: Dinuclear Ni(II) complexes, hydrolase, second coordination sphere, antibacterial
activity
ABSTRACT: Herein, we report the synthesis and characterization of three new Ni(II) complexes,
namely: [Ni2(H2LEt)(μ-OAc)2(H2O)]BPh4‧ClO4 (1), (H2LEt = 2-[(N-benzyl-N-2-pyridyl
methylamine)]-4-methyl-6-[N-2(pyridylmethyl)aminomethyl)]-6((2-aminoethyl)amino)methyl
phenol); [Ni2(H2LProp)(μ-OAc)2(H2O)](ClO4)2 (2), (H2LProp
=
2-[(N-benzyl-N-2-
pyridylmethylamine)]-4-methyl-6-[N-(2-pyridylmethyl)aminomethyl]-6-((2-aminopropyl)
amino)methylphenol); and [Ni2(LBut)(μ-OAc)2(H2O)](HCl)2 (3), (LBut = 2-[(N-benzyl-N-2-
pyridylmethylamine)]-4-methyl-6-[N-(2-pyridylmethyl)aminomethyl]-6-((2-aminobuthyl)
amino)methylphenol). All of them were characterized through spectroscopic techniques
(elemental analysis, IR, UV-Vis spectroscopy), ESI-MS, electrochemistry and potentiometric
titration. Density functional theory (DFT) was used to better understand the electronic and
1

molecular structure of these complexes. The hydrolytic activity of complexes 1 – 3 towards the
2,4-BDNPP substrate was analyzed and complex 2 presented the highest catalytic efficiency
(kcat/KM) of the three, possibly due to a greater interaction with the substrate. The complexes were
also screened for their antibacterial activities using both Gram-positive and Gram-negative
bacterial strains by minimum inhibitory concentration and minimum bactericidal concentration
methods.
Abbreviations
2,4-DNPP
2,4-DNP
CHES
2,4-dinitrophenylphosphate
2,4-dinitrophenolate
N-cyclohexyl-2-aminoethanesulfonic acid
Deoxyribonucleic acid
DNA
H2BPPAMFF 2-[(N-benzyl-N-2-pyridylmethyl-amine)]-4-methyl-6-
[N-(2-pyridylmethyl)aminomethyl]-4-methyl-6-formylphenol
HEPES
MES
2-[4-(2-hydroxyethyl)-piperazin-1-yl]ethanesulfonic acid
2-(N-morpholino)ethanesulfonic acid
TD-DFT/TDA Time-dependent-density functional theory/Tamm-Dancoff approximation
TS
Transition state
1. Introduction
Hydrolases are well-known enzymes that are of considerable interest due to their ability to make
use of water molecules to promote the cleavage of chemical bonds. This feature plays a crucial
role in organisms, in which some enzymes are able to cleave the DNA in a hydrolytic fashion [1-
3]. The half-life of a phosphodiester bond in DNA has been estimated to be hundreds to thousands
of millions of years [4], which explains why phosphate esters are so resistant toward hydrolysis
under physiological conditions.
Phosphatases are enzymes that hydrolytically cleave phosphoester bonds in particular, playing an
important role in cellular regulation and signaling and they are among the most extensively studied
2

metalloenzymes [5,6]. Therefore, over the past few decades a series of model complexes able to
mimic their ability to cleave P-O bonds have been described in the literature [7-15]. Most of these
complexes target the properties and catalytic functions of phosphatases to develop possible
therapeutic drugs that could act as anticancer, antiviral and/or antibiotic agents. The design of some
of these compounds has, as a common feature, the presence of a dinuclear active site, in which
both metal centers play important roles in substrate hydrolysis. It has been postulated that one
metal ion acts as a Lewis acid, decreasing the pKa of the coordinated water molecule, while the
other activates the substrate [4,16-21]. The most commonly used metal ions include FeII/III, ZnII,
CuII, MnII and NiII [22-30].
Although NiII complexes are mostly used to mimic ureases in the literature, some of them
reportedly show significant phosphodiesterase activity [5,37-41], including that reported by
Greatti et al. The authors describe a NiIINiII complex with an unsymmetrical ligand containing an
imidazole moiety, able to catalyze the hydrolysis of the bis-(2,4-dinitrophenyl)phosphate (2,4-
BDNPP) substrate almost one million times faster than the noncatalyzed reaction [42]. A similar
study was reported by Piovezan et al, in which the NiIINiII complex had an unsymmetrical ligand
in its structure, with an aldehyde moiety that was used to attach the complex to silica surfaces, as
can be seen in Chart 1. The complex attached to silica was able to catalyze the hydrolysis of 2,4-
BDNPP over a hundred thousand times faster than the noncatalyzed reaction [43]. This is an
example of the role of the second coordination sphere in the catalysis. Therefore, since the so-
called first sphere of coordination has been previously studied by several research groups and for
a large number of metalloenzymes and their biomimetics [44-50], a new quest began in order to
understand the role of the amino acid residues surrounding the active site of the metalloenzymes.
These amino acid residues are called the second coordination sphere, and some authors have
reported the impact that changes in their structure have on the catalytic activity of the
metalloenzymes [51-59]. As for example, in the work reported by Hadler et al, on their studies
with glycerophosphodiesterase (GpdQ), an enzyme capable to hydrolyze P–O bond of tri-, di- and
monoesters, the replacement of the amino acid residue Asn80 by alanine leads to an increase in
reactivity, but at the cost of decreased in affinity with the substrate if compared with the original
enzyme [60]. The amino acid residues surrounding the active site of metalloenzymes can not only
help to bind and activate the substrate, but also enhance the nucleophilicity of the metal-bound
hydroxo groups, stabilize the charged transition state and assist the departure of the leaving group
3

through hydrogen-bonds or electrostatic interactions [5,54-56]. Some researchers have focused
their attention on the investigation of the effects of the second coordination sphere, employing
ammonium, guanidine or amido groups in the vicinity of the metal center of model complexes in
order to mimic the hydrogen-bonds and/or electrostatic interactions [54]. One example of this type
of study was reported by Silva and co-workers [61], who synthesized three FeIIIZnII complexes,
one of which presented an aldehyde moiety, while its derivative presented a diamine group. The
complex containing the diamine group presented a kcat value 3.2 times greater than the complex
containing the aldehyde moiety and had a kcat/KM value 7.0 times greater than its precursor. The
authors explained this through electrostatic interactions arising from the protonation of the primary
amine group, which could lead to a greater affinity to the substrate.
In addition to presenting higher catalytic activity in the hydrolysis of phosphodiester substrates in
comparison to other common metallic centers, such as ZnII, FeII/III and MnII, nickel coordination
compounds are also known to cleave the DNA of certain cancer cells [62-66]. Thus, the rational
design of new ligands presenting second coordination sphere interactions can bring extra insights
and novel approaches for cancer treatment. Nickel compounds can also penetrate the microbial
cells, inactivating their enzymes [67] and acting as antimicrobial compounds. Several authors [68-
75] already reported Ni(II) complexes presenting antimicrobial activity, which is dependent on the
chelate ring geometry and stability of the metal complexes. Therefore, the study of the
antimicrobial effect of nickel complexes on Gram-positive and Gram-negative bacteria is of great
interest, but the investigation of this subject remains insufficient.
In the aforementioned study, Piovezan and co-workers reported the synthesis of a dinuclear ligand
containing an aldehyde moiety (Chart 1), which can be functionalized with side chain groups [43].
This approach could evidently be applied to the study of the second coordination sphere effects,
and the insertion of amino groups in the side chain could, for example, lead to a greater interaction
with the substrate. With that in mind, modifications on the ligand H2BPPAMFF were performed
by inserting aliphatic diamines in the backbone of the ligand, providing the new H2LEt, H2LProp
and H2LBut ligands (Chart 2). Amine groups tend to be in their positively-charged ammonium
forms in solution around the pH range of hydrolytic catalytic activity, which can attract the anionic
substrate 2,4-BDNPP with electrostatic contributions [76].
Considering these results and given the lack of studies focused on Ni(II) complexes presenting
phosphodiesterase activity in the literature, herein we report the synthesis and characterization of
4

three new nickel compounds. Antibacterial studies were carried out with these compounds to
determine their action against bacterial strains of Escherichia coli, Salmonella typhi, Pseudomonas
aeruginosa, Aeromonas hydrophila, Listeria innocua and Staphylococcus aureus and their
hydrolase-like activity towards the model substrate 2,4-BDNPP was also investigated.
Chart 1 A) Ligand H2BPPAMFF and B) Complex A reported by Piovezan et al [43].
Chart 2 Ligands used in this study.
2. Experimental
2.1 Chemicals
The reagents, materials, gases and solvents of high purity grade used in the synthesis procedures
were purchased from commercial sources and used without further purification. The 3-
(chloromethyl)-2-hydroxy-5-methylbenzaldehyde [77], bis-(2,4-dinitrophenyl)phosphate [78] and
the
ligand
2-[(N-benzyl-N-2-pyridylmethylamine)]-4-methyl-6-[N-(2-
5

pyridylmethyl)aminomethyl]-4-methyl-6-formylphenol (H2BPPAMFF) [43] were synthesized by
previously described methods.
2.2 Physical measurements
Infrared spectra were obtained on a PerkinElmer Fourier transformed infrared-attenuated total
reflectance (FTIR-ATR) spectrophotometer (model Spectrum 100), using a ZnSe (45º) crystal, and
a triglycine sulfate (TGS) detector. The samples were analyzed directly by ATR, in the range of
4000-550 cm-1, and were corrected by a blank measurement. The elemental analysis of all
complexes and ligands was carried out with a CE Instruments C, H, N, S, O analyzer (model EA
1110 CHNS-O). The mass spectra for all ligands and complexes were obtained on an Amazon-ion
trap mass spectrometer via electrospray ionization (ESI-MS). High-resolution mass spectra were
obtained on a Bruker Daltonics spectrometer (model Q-micrOTOF Q-II) equipped with a KD
Scientific automatic syringe for sample injection. The ESI-MS analysis of both ligands and
complexes was carried out in an ultrapure acetonitrile solution at a concentration of 500 ppb at 180
μL min-1. The capillary temperature was maintained at 180 to 200 °C and the voltage at -400 to -
500 V. The simulated spectra were calculated using the Mmass software [79]. Nuclear magnetic
resonance spectra (1H NMR) were recorded with a Bruker FT 200 MHz instrument at room
temperature and the chemical shifts were recorded in ppm using tetramethylsilane (TMS) as the
internal reference (TMS, δ = 0.00 ppm) and deuterated chloroform as the solvent. Electronic
spectra were obtained on a Perkin-Elmer Lambda-750 spectrophotometer in the range of 250-800
nm in acetonitrile solution. The redox behavior of the complexes was investigated by cyclic and
square-wave voltammetry on a BAS potentiostat/galvanostat (model Epsilon), and the experiments
were performed in acetonitrile solution under argon atmosphere. In these experiments,
tetrabutylammonium hexafluorophosphate [NBu4](PF6) (0.1 mol L-1) was used as the supporting
electrolyte and measurements were taken in an electrolytic cell with three electrodes: glassy carbon
(working); platinum wire (counter); and Ag/Ag+ (reference). Ferrocene (E° = 0.4 V vs NHE) was
used as the internal standard [80]. The Ni determination was carried out by atomic absorption
spectrometry with a continuous source high resolution atomic absorption spectrometer, model
ContrAA 700 (Analytik Jena, Jena, Germany), coupled with a graphite furnace and flame
atomizers. The selected wavelength was 232.003 nm. A mixture of air (oxidant) and acetylene
(fuel) was used with a continuous flow of 55 L h-1. The burner height was fixed at 6 mm.
6

2.3 Potentiometric titrations
The potentiometric studies were carried out with a Methrom 848 Titrino Plus automatic titrator
fitted with a combination electrode (Ag/AgCl) calibrated to read -log[H+] directly, designated as
the pH. The experiment temperature was maintained at 25.00±0.05 °C and gas-free deionized
water was used to prepare the solutions. The ionic strength was kept constant at 0.1 mol L-1 through
the addition of KCl. Experiments were performed in 50 mL of a CH3CN/H2O 1:1 (pKw =15.40
[81]) solvent system, containing 0.025 mmol of the complex, purged with Ar, cleaned using two
bubbling towers with 0.10 mol L-1 KOH solutions and titrated with 0.10 mol L-1 standard CO2-
free KOH from pH 3.00 to 12.00 through the addition of 0.025 mL aliquots of base. The
experiments were run in duplicate for complex 1 and in triplicate for complexes 2 and 3. Data
analysis was performed using the BEST7 program [82] and the species diagrams were obtained
with the SPE [82] and SPEPLOT [82] programs.
2.4 Kinetic assays
The phosphodiesterase-like activity of the complexes was determined by monitoring the hydrolysis
of the substrate bis-(2,4-dinitrophenyl)phosphate (2,4-BDNPP) under substrate excess. The
experiments were carried out in triplicate using a UV-Vis Varian Cary 50 BIO spectrometer at 400
nm in an CH3CN/H2O (1:1, v/v) mixture. The initial rates were measured in real time at various
pH values between 5.00 and 9.50 (buffers used were 0.10 mol L-1 MES, pH 5.00-6.50, HEPES,
pH 7.00-8.50 and CHES, pH 9.00-9.50) with ionic strength I = 0.10 mol L-1, LiClO4. The molar
absorption coefficient of the product 2,4-dinitrophenolate was determined at each pH under the
same experimental conditions applied to obtain the rate measurements [83]. The reactions were
monitored to less than 5% of conversion. The pH dependence of the rate was investigated using
fixed concentrations of the substrate ([S]final = 5.33 × 10-3 mol L-1 for 1, [S]final = 3.33 × 10-4 mol
L-1 for 2 and [S]final = 6.66 × 10-4 mol L-1 for 3) and complexes ([C]final = 1.00 × 10-4 mol L-1 for 1,
[C]final = 3.33 × 10-6 mol L-1 for 2 and [C]final = 1.00 × 10-5 mol L-1 for 3) at 25.0 ± 0.5 °C. Substrate
dependence ([S]final = 2.00 × 10-4 mol L-1 – 4.40 × 10-3 mol L-1) was measured in the pH range of
5.00 to 9.50. The data were treated using the Michaelis-Menten equation by non-linear regression
[84]. In order to establish the number of molecules of substrate which are hydrolyzed per molecule
of complex, the reaction was monitored at 445 nm (ε = 3600 L mol-1 cm-1) [83], under a 50-fold
7

substrate excess ([S]final = 2×10-4 mol L-1 for 1 and 2 and [S]final = 3×10-4 mol L-1 for 3) relative to
the complex ([C]final = 4×10-6 mol L-1 for 1 and 2 and [C]final = 6×10-6 mol L-1 for 3), at pH 9.00
and 25 °C. Also, a stoichiometric reaction between complexes 1, 2 and 3 and the 2,4-BDNPP
substrate (1 × 10-5 mol L-1) was monitored (400 nm) at pH 9.00 and 25 °C. Isotopic effects
regarding the hydrolysis rate of 2,4-BDNPP promoted by the complexes were investigated by
monitoring the reactions in deuterated buffer. The effect of temperature on the reaction rates was
investigated in the range 20 – 40 °C, at pH 9.00, for all complexes under the same conditions as
those used to assess the substrate dependence. The studies were corrected for spontaneous
hydrolysis by direct difference between two identical parallel reactions.
2.5 Biological Studies
The antibacterial activity of the complexes was studied by agar selective and differential methods
at pH 7.2, using six strains of bacteria (Escherichia coli, Salmonella typhimurium, Pseudomonas
aeruginosa, Aeromonas hydrophila, Listeria innocua and Staphylococcus aureus) by the minimum
inhibitory concentration method. The different bacterial cultures were maintained in trypic soy
broth (TBS, Difco, Le Pont-de-Claix, France) supplemented with 20% glycerol at -20 °C. An
aliquot of the culture was seeded in brain heart infusion broth (BHI, Difco,Le Pont-de-Claix,
France) and incubated for 24 h at 35 °C with exception of S. aureus, which was incubated for 48
h.
2.5.1 Minimum inhibitory and bactericidal concentrations
The minimum inhibitory concentration (MIC) and minimum bactericidal concentration (MBC)
were determined using the 96-well microplate microdilution technique (TTP, Trasadingen,
Switzerland) according to the Clinical and Laboratory Standards Institute [85]. The compounds
were diluted in Mueller Hinton broth (MHB, Difco, Le Pont-de-Claix, France) in 2% acetonitrile
at the final concentration and added to the microplates in batches ranging from 0.97 to 1000 µg
mL-1.
The bacterial cultures cultivated in MHB were standardized to the McFarland scale 0.5 and diluted
to obtain 5 × 106 colony-forming units (CFU) mL-1 and 10 µL were inoculated in each microplate
well and incubated at 35 °C for 24 h. The MIC was defined as the lowest concentration that
inhibited bacterial growth by visual reading. From the wells in which the bacterial growth was not
8

observed, a 10 µL aliquot was inoculated in plates containing selective differential agar and
incubated at 35 °C for 24 h to determine the MBC. The MBC was determined as the lowest
concentration in which no bacterial growth was observed by visual inspection. The experiments
were performed in triplicate.
Culture mediums were added as controls without the addition of the antimicrobial and inoculum
(negative control) culture medium, with the addition of the bacterial inoculum (positive control)
and control of the acetonitrile solvent (2% v/v) with the addition of the inoculum.
2.6 Theoretical predictions
Geometry optimizations of complexes 1, 2 and 3 were carried out in a vacuum with the Orca 4.0.1
software package [86] at the density functional theory (DFT) level using the BP86 functional
[87,88]. The basis set chosen was Def2-TZVP for the nickel atom and Def2-SVP for the other
atoms [89-91]. These calculations also included Grimme’s dispersion correction (D3) with Becke-
Johnson damping (BJ) [92,93]. The vibrational frequencies for both complexes 1, 2 and 3 showed
no imaginary frequencies. The phosphate conjugates were also optimized, and the vibrational
frequencies showed only one small negative frequency due to rotation of the aromatic rings. In
order to simulate the absorption spectra, time-dependent DFT under the Tamm-Dancoff
approximation (TD-DFT/TDA) was employed to obtain the first 50 excitations, using the same
calculations protocol, differing only at the functional, which in this case was PBE0 [94,95], and
the basis set for coordinated atoms was the same used for the nickel atom, Def2-TZVP. The 3D
representations of the complexes were obtained using the Chemcraft program [96].
2.7 Synthesis of the ligands
Three new ligands containing side chain amino groups derived from the already known
H2BPAMFF ligand [43] were prepared.
Synthesis
of
2-[(N-benzyl-N-2-pyridylmethylamine)]-4-methyl-6-[N-
2(pyridylmethyl)aminomethyl)]-6-((2-aminoethyl)amino)methylphenol – H2LEt. In a 500 mL
round-bottom flask, 4.20 g (70 mmol; 60.10 g mol-1; 0.899 g mL-1) of 1,2-diaminoethane was
added to 150 mL of an ethanol/dichloromethane (2:1) solution, followed by 4.10 g (7 mmol; 586.72
g mol-1) of H2BPPAMFF in 50 mL of ethanol added dropwise. The reaction mixture was left to
9

react overnight at room temperature. The temperature was then lowered to 0°C in an ice bath and
3.18 g (84 mmol; 37.83 g mol-1) of solid sodium borohydride were added slowly, and the mixture
was stirred for 2 h. The pH was adjusted to 5.0 and the solvent was removed under reduced
pressure. The product was dissolved in 30 mL chloroform and washed first with a saturated sodium
bicarbonate solution (2 × 15 mL), then three times with a saturated sodium chloride solution (3 ×
15 mL). The organic layer was dried with anhydrous Na2SO4, then filtered off, and the solvent
removed under reduced pressure, resulting in a pale-pink foam (yield: 93%). Anal. calcd (found)
for C39.5H49ClN6O3: C, 68.63 (68.97); H, 7.14 (7.26); N, 12.16 (12.15). IR (ATR), in cm-1: ν (C-
Har and C-Haliph) 3066-2712; ν (C=N and C=C) 1589-1432; δ (O-Hphenol) 1365; ν (C – Ophenol) 1228;
ν (C – Har) 746-698. 1H NMR - δH (200 MHz; CDCl3), in ppm: 2.20 (s, 3HCH3); 2.21 (s, 3HCH3);
3.67 (s, 2HCH2); 2.60 – 2.66 (dt, 2HCH2); 2.75 – 2.80 (dt, 2HCH2); 3.66 (s, 2HCH2); 3.70 (s, 2HCH2);
3.74 (s, 2HCH2); 3.77 (s, 4HCH2); 3.82 (s, 2HCH2); 3.83 (s, 2HCH2); 6.81 – 6.90 (dd, 4Har); 7.11 –
7.21 (m, 3Har); 7.29 – 7.40 (m, 6Har); 7.56 – 7.68 (m, 2Har); 8.52 – 8.59 (dd, 2Hpy). ESI-MS: [H-
2LEt] + H+, m/z = 631.35.
Synthesis
of
2-[(N-benzyl-N-2-pyridylmethylamine)]-4-methyl-6-[N-(2-
pyridylmethyl)aminomethyl]-6-((2-aminopropyl)amino)methylphenol – H2LProp. In a 500
mL round-bottom flask, 5.20 g (70 mmol; 74.13 g mol-1; 0.886 g mL-1) of 1,3-diaminopropane
was added to 150 mL of an ethanol/dichloromethane (2:1) solution, followed by 4.10 g (7 mmol;
586.72 g mol-1) of H2BPPAMFF in 50 mL of ethanol added dropwise. The reaction mixture was
left to react overnight at room temperature. The temperature was lowered to 0°C in an ice bath
then 3.17 g (84 mmol; 37.83 g mol-1) of solid sodium borohydride were added, and the mixture
was stirred for 4 h. The pH was adjusted to 5.0 and the solvent was removed under reduced
pressure. The product was dissolved in 30 mL chloroform and washed first three times with a
saturated sodium bicarbonate solution (3 × 15 mL), and then two times with a saturated sodium
chloride solution (2 × 15 mL). The organic layer was dried with anhydrous Na2SO4, then filtered
off, and the solvent removed under reduced pressure, resulting in a white foam (yield: 98%). Anal.
calcd (found) for C41H52Cl2N6O3: C, 65.85 (65.62); H, 7.01 (6.97); N, 11.24 (11.15). IR (ATR), in
cm-1: ν (C-Har and C-Haliph) 3061-2712; ν (C=N and C=C) 1588-1433; δ (O-Hphenol) 1365; ν (C –
1
Ophenol) 1229; ν (C – Har) 748-697. H NMR - δH (200 MHz; CDCl3), in ppm: 1.61 – 1.67 (q,
2HCH2); 2.21 (s, 3HCH3); 2.69 – 2.71 (t, 4HCH2); 3.66 (s, 2HCH2); 3.70 (s, 2HCH2); 3.74 (s, 2HCH2);
10

3.77 (s, 4HCH2); 3.82 (s, 2HCH2); 3.83 (s, 2HCH2); 6.83 – 6.89 (dd, 4Har); 7.18 (m, 3Har); 7.29 – 7.39
(m, 6Har); 7.58 – 7.64 (m, 2Har); 8.57 (dd, 2Hpy). ESI-MS: [H2LProp] + H+, m/z = 645.46.
Synthesis
of
2-[(N-benzyl-N-2-pyridylmethylamine)]-4-methyl-6-N-(2-
pyridylmethyl)aminomethyl-6-((2-aminobuthyl)amino)methylphenol – H2LBut. In a 500 mL
round-bottom flask, 7.052 g (80 mmol; 88.15 g mol-1; 0.877 g mL-1) of 1,4-diaminobuthane was
added to 150 mL of an ethanol/dichloromethane (2:1) solution, followed by 4.69 g (8 mmol; 586.72
g mol-1) of H2BPPAMFF in 50 mL of ethanol added dropwise. The reaction mixture was left to
react overnight at room temperature. The temperature was lowered to 0 °C in an ice bath, 3.63 g
(96 mmol; 37.83 g mol-1) of solid sodium borohydride were added, and the mixture was stirred for
4 h. After this time, the pH was adjusted to 5.0 and the solvent was removed under reduced
pressure. The product was dissolved in 30 mL chloroform and washed first four times with a
saturated sodium bicarbonate solution (4 × 15 mL), and then three times with a saturated sodium
chloride solution (3 × 15 mL). The organic layer was dried with anhydrous Na2SO4, then filtered
off, and the solvent removed under reduced pressure, resulting in a light-yellow foam (yield: 98%).
Anal. calcd (found) for C42H56Cl2N6O4: C, 64.69 (64.15); H, 7.24 (7.20); N, 10.78 (10.73). IR
(ATR), in cm-1: ν (C-Har and C-Haliph) 3088-2711; ν (C=N and C=C) 1592-1432; δ (O-Hphenol
)
1367; ν (C – Ophenol) 1227; ν (C – Har) 748-696. NMR - δH (200 MHz; CDCl3), in ppm: 1.39 – 1.52
(m, 4HCH2); 2.21 – 2.22 (s, 6HCH3); 2.56 – 2.67 (m, 4HCH2); 3.66 – 3.83 (s, 14HCH2); 6.82 (s, 2Har);
6.90(s, 2Har); 7.11 – 7.21 (m, 3Har); 7.29 – 7.41 (m, 6Har); 7.55 – 7.68 (dt, 2Har); 8.52 – 8.58 (dd,
2Hpy). ESI-MS: [H2LBut] + H+, m/z = 659.45.
2.8 Synthesis of the Complexes
Complexes 1, 2 and 3 (Scheme 1) were synthesized using a procedure similar to that previously
described by Piovezan et al [43].
Synthesis of the Complex [Ni2(H2LEt)(μ-OAc)2(H2O)](ClO4)(BPh4) (1). Complex (1) was
synthesized in a (2:1) methanol/acetonitrile solution by mixing 0.37 g of Ni(ClO4)2‧6H2O (1 mmol;
365 g mol-1) and 0.32 g of the H2LEt ligand (0.5 mmol; 630.82 g mol-1) with stirring and mild
heating (55 °C). In the next step, 0.14 g of NaCH3COO‧3H2O (1 mmol; 136.08 g mol-1) and 0.17
g of NaBPh4 (0.5 mmol; 342.22 g mol-1) were added and a green solution was obtained. After
recrystallization in a (1:1) dichloromethane/hexane solution, a green solid was obtained (yield:
0.48 g, 68%). Anal. calcd (found) for Ni2C68H78BCl3N6O12: C, 58.09 (57.76); H, 5.59 (5.70); N,
11

5.98 (6.35). IR (ATR), in cm-1: ν (O – H) 3511; ν (C – Har and C – Halif) 3055 – 2912; ν (C=C)
1605 – 1445; νass (COO-) 1554; νs (COO-) 1427; ν (C – O) 1317; ν (Cl-O) 1090; δ (C-Har) 735 –
703. ESI-MS: [Ni2LEt(CH3COO)]+, m/z = 803.31. Atomic absorption, in mg mL-1: Ni,
2.97±0.0034.
Synthesis of the Complex [Ni2(H2LProp)(μ-OAc)2(H2O)](ClO4)2 (2). Complex (2) was
synthesized in a methanolic solution by mixing 0.37 g of Ni(ClO4)2‧6H2O (1 mmol; 365 g mol-1)
and 0.32 g of the H2LProp ligand (0.5 mmol; 644.82 g mol-1) with stirring and mild heating (45
°C). In the next step, 0.14 g of NaCH3COO‧3H2O (1 mmol; 136.08 g mol-1) and 0.17 g of NaBPh4
(0.5 mmol; 342.22 g mol-1) were added and a green solution was obtained. After recrystallization
in a (1:1) ethyl acetate/acetone solution, a green solid was obtained (yield: 0.36 g, 57%). Anal.
calcd (found) for Ni2C57H68B0.5Cl4N6Na0.5O15: C, 50.59 (50.68); H, 5.06 (5.08); N, 6.21 (6.20). IR
(ATR), in cm-1: ν (O – H) 3496; ν (C – Har e C – Haliph) 3058 – 2854; ν (C=C) 1606 – 1422; νass
(COO-) 1574; νs (COO-) 1437; ν (C – O) 1318; ν (Cl-O) 1081; δ (C-Har) 759 – 703. ESI-MS:
[NiLProp(CH3COO)]+, m/z = 877.37. Atomic absorption, in mg mL-1: Ni, 2.77±0.0171.
Synthesis of the Complex [Ni2(LBut)(μ-OAc)2(H2O)]‧(HCl)2 (3). Complex (3) was synthesized
in a (2:1) methanol/acetonitrile solution by mixing 0.25 g of Ni(OAc)2‧4H2O (1 mmol; 248.86 g
mol-1) and 0.33 g of the H2LBut ligand (0.5 mmol; 658.82 g mol-1) with stirring and mild heating
(45 °C). In the next step, 0.34 g of NaBPh4 (1 mmol; 342.22 g mol-1) was added to the mixture and
a green solution was obtained. The resulting green solid was collected and washed with distilled
water (yield: 0.58 g, 94%). Anal. calcd. (found) for Ni2C93H98B2Cl2N6Na2O7 :C, 66.98 (66.67); H,
5.92 (5.91); N, 5.04 (5.54). IR (ATR), in cm-1: ν (C – Har and C – Haliph) 3055 – 2859; ν (C=C)
1603 – 1412; νass (COO-) 1577; νs (COO-) 1439; ν (C – O) 1315; δ (C-Har) 757 – 701. ESI-MS:
[Ni2LBut(CH3COO)2(H2O)2(Na)2(Cl)]+, m/z = 1007.32. Atomic absorption, in mg mL-1: Ni,
3.36±0.0114.
Caution! Perchlorate salts of metal complexes are potentially explosive.
12

Scheme 1. Schemes showing the synthesis of complexes 1, 2 and 3.
3. Results and discussion
3.1 Synthesis of the ligands and complexes
13

The synthetic route to the unsymmetric ligands (H2LEt, H2LProp and H2LBut) is illustrated in
1
Figure S1. The three ligands were fully characterized by H NMR, IR and ESI-MS techniques
(Figures S26-S28). The reaction of the ligands with Ni(II) salts, as described above, afforded the
complexes 1, 2 and 3 as green solids.
3.2 Solid-state and DFT studies
3.2.1 Infrared (IR) spectroscopy
The IR spectra for 1, 2 and 3 (Figures S2, S3 and S4 and Table S1) present bands of νass (COO-)
and νs (COO-) with Δ = 127 cm-1, 137 cm-1 and 138 cm-1 for 1, 2 and 3 respectively, indicating the
coordination of a carboxylate group in a bridging mode [97]. Bands at 1090 cm-1 and 1081 cm-1
for 1 and 2, respectively, related to the stretching of the Cl-O bond of the counterion were present
in all spectra. The results obtained for complexes 1 – 3 agree with those reported for similar
complexes in the literature [42,43], as well as the frequencies obtained in the DFT studies (vide SI
- IR spectroscopy).
3.2.2 DFT modeling of the structures
Since no single crystals were obtained for X-ray analysis, DFT was used to predict likely structures
(and also to help to understand the experimental data) for complexes 1-3 and their phosphate
conjugates. The optimization of the structures for the three complexes was carried out only on the
predominant species at the optimum kinetic pH value. The calculations were performed with a
terminal hydroxo group bonded to the nickel center on the soft moiety of the ligand and a water
ligand in the nickel center on the hard moiety. In agreement with the potentiometric titration
studies, it was also assumed that the terminal amines of the ligands were protonated in their
phosphate conjugates. Also, the spin of the NiII ion was set equal to 1 as reported by Piovezan et
al [43]. As shown in Figure S5, all of the nickel centers exhibit a pseudo-octahedral geometry,
which is expected for these complexes, based on previous studies with similar systems [43]. As
can be seen in Table S2, all of the bonds and angles were in good agreement with those previously
reported for NiII(µ-OH)NiII complexes for which X-ray structures were obtained [43]. The
calculated vibrational frequencies and intensities were compared with the experimental infrared
spectra (Figures S6-S8), showing reasonable agreement. It is important to note that no perchlorate
14

anion was included in the theoretical predictions; hence, the lack of the intense band at around
1100 cm−1. Table S3 shows the assignments of the IR bands for complexes 1–3 (experimental and
calculated).
3.3 Solution studies
3.3.1 Mass spectrometry (MS)
The ESI-MS analysis of complexes 1, 2 and 3 was performed in a pure acetonitrile solution. The
mass spectrum of complex 1 showed a group of peaks with a maximum (100%) at m/z 460.15 with
a +2 charge, which can be assigned to the system [NiIINiII(μ-CH3COO)(HLEt)]+CH3OH+CH2Cl2
(Figure S9). Another group of peaks (intensity = 28%) at m/z 803.31 with a +1 charge was observed
and can be assigned to the [NiIINiII(μ-CH3COO)(LEt)] species (Figure S10). For complex 2 a
group of peaks with a maximum (100%) at m/z 817.32 and a +1 charge was observed and can be
assigned to [NiIINiII(μ-CH3COO)(LProp)] (Figure S11). The mass spectrum for complex 3 showed
a group of peaks with a maximum (100%) at m/z 1007.32 and a +1 charge, which can be assigned
to [NiIINiII(μ-CH3COO)2(HLBut)]+4H2O+Na++Cl- (Figure S12). Another group of peaks (29%) at
m/z 831.36 with a +1 charge can be seen, which can be assigned to [NiIINiII(μ-CH3COO)(LBut)]
species (Figure S13).
3.3.2 Electronic absorption spectroscopy and DFT studies
The electronic spectra of 1, 2 and 3 were investigated in the range of 250 – 800 nm in acetonitrile
solution and present bands at λmax, nm (ε, L mol-1 cm-1), of 609 (14), 616 (15) and 626 (23),
respectively (Figures S14-S16). These can be attributed to d-d transitions, showing a bathochromic
shift as the side chain amino group increases in size from two to four carbon atoms. The
bathochromic shift can also be observed when we compare the λmax values of complex
[Ni(HBPPAMFF)(µOAc)2(H2O)]BPh4 (A), reported by Piovezan and coworkers [43] – which
contains an aldehyde moiety in its structure – with those of complexes 1 – 3, as can be seen in
Table 1. A weak band is also observed at around 770 nm for all complexes and is attributed to a
spin-forbidden transition. The diffuse reflectance spectra of complexes 1 – 3 exhibit bands at
approximately 614 nm, 622 nm and 628 nm, respectively, as well as a weak band at around 770
nm (Figures S14-S16), indicating that the structures are maintained in solution. These values agree
with those for other complexes previously described in the literature [17,27,31,39,41,42,43,62,98].
15

The calculated frontier orbitals for complexes 1 – 3 are shown in Figure 1. It can be observed that
for all complexes the last alpha and beta occupied orbitals are spread over the substituted phenol
moiety of the ligand with a slight contribution of dπ orbitals of the Ni(II) center, and there is no
significant influence from the different lateral groups. The first unoccupied orbitals are located
over the pyridine coordinated to the same nickel center, also with a small contribution of dπ orbitals
from the Ni(II). Time-dependent DFT (TD-DFT) was used to simulate the absorption spectra and
help to understand the assignments from the experimental spectra (Figures S17-19). Based on these
calculations, the low energy bands (~550-900 nm) of 1-3 can be resolved in a few excitations, all
of which are related to d-d like transitions, where donor molecular orbitals have a small
contribution from the coordinated oxygen atoms and the acceptor molecular orbitals have a small
contribution from the phenol moieties. The transitions involving mostly the frontier orbitals
described above are located in the region between 550 and 400 nm. The remaining transitions at
the high energy end of the spectrum could be assigned as π-π*-like transitions with a small
contribution from CT transitions involving the Ni(II) center.
Figure 1. Calculated frontier orbitals of complexes 1, 2 and 3.
16

3.3.3 Electrochemical properties and molar conductivity
Cyclic and square-wave voltammograms were recorded in dry acetonitrile in the potential range
-2.0 to +2.0 V versus NHE, using [NBu4][(PF6)] as the supporting electrolyte and ferrocene as the
internal standard (E1/2 = 0.4 V vs NHE) [80]. In the cyclic voltammograms for complexes 1 – 3,
one irreversible wave was observed (Figures S20, S22 and S24) in the cathodic region, which can
be attributed to the one-electron transfer processes NiIINiII/NiIINiI while in the anodic region two
waves were detected for all complexes and are attributed to the oxidation of the ligands (Figures
S20, S22 and S24). However, the possibility of these processes being associated with the oxidation
of the Ni(II) centers cannot be excluded. The square-wave voltammograms for complexes 1 and 3
showed a quasi-reversible behavior (Figures S21 and S25), which can be attributed to the one-
electron transfer processes NiIINiII/NiIINiI and NiIINiII/NiIINiII, whereas for complex 2 only one
cathodic wave could be observed (Figure S23). From the values listed in Table 1 it is possible to
infer an anodic shift from complex A [43] to 3, which agrees with the bathochromic shift observed
on the UV spectra. This can be explained by interactions related to the protonated amino group
and the terminal water molecule bonded to the metal center (vide potentiometric data, section 3.4),
which causes the withdrawn of electron density from the metal center, thereby favoring its
reduction [17,61,99]. The measured molar conductivity of the complexes in acetonitrile solution
at 25 °C was 121 S cm2 mol-1 for 1, 140 S cm2 mol-1 for 2, which is characteristic of a 1:1
electrolyte. The value of 36 S cm2 mol-1 for 3 is characteristic for a neutral species in solution
[100]. In addition, the atomic absorption spectrometry analysis results for complexes 1 – 3 are
consistent with the presence of two nickel centers for each complex.
Table 1 Electronic spectra and electrochemical data for complexes 1 – 3, at 25ºC.
Complex
λmax, nm (ε, L mol-1 cm-1)
Epc, (V vs NHE)
Epa, (V vs NHE)a
CH3CN
CH3CN
A
1
604 (20)
-1.27 and -1.64
-0.960
609 (14)
-0.974
2
616 (15)
-0.850
3
626 (23)
-0.943
-0.916
B
621 (20)
-1.29 and -1.66
1.00 and 1.40
a[Complexes] = 1 × 10-3 mol L-1; B = [NiIINiII(L)(µ-CH3COO)2(OH2)]BPh4·H2O L= 2-[(4,7-diisopropyl-1,4,7-
triazonan-1-yl)methyl]-4-methyl-6[(pyridine-2-ylmethylamino)methyl]phenolate
[27].
Conditions
for
electrochemistry: glassy carbon (working); platinum wire (counter); and Ag/Ag+ (reference). Ferrocene (E° = 0.4 V
vs NHE) was used as the internal standard.
17

3.4 Potentiometric equilibrium studies
Potentiometric titrations of complexes 1 – 3 were carried out in CH3CN/H2O (1:1) solutions in
order to assess the presence of water molecules coordinated to the Ni(II) metal centers, since the
lability of the acetate bridges is facilitated by increasing the pH of the solution
[7,11,23,101,102,103], which generates aquo-complexes. The results indicate the neutralization of
7 mols of potassium hydroxide per mol of complex in the pH range of 3.00 – 12.00. The
corresponding pKa values are listed in Table 2 and the proposed assignments of the corresponding
equilibrium observed for complexes 1 – 3 are shown in Figure 2. The species distribution curves
as a function of pH for complexes 1 – 3 are presented in Figure 3.
For complexes 1 – 3, the first pKa can be assigned to the release and the deprotonation of one
acetate bridge bonded to the metal centers, leading to the [(H2O)2NiII(µ-OAc)NiII(OH2)] species.
The second pKa can be assigned to the release and the deprotonation of the second acetate bridge,
leading to the [(H2O)3NiIINiII(OH2)] species.
Table 2. Values of the protonation constants for complexes 1 – 3.
1
2
3
pKa1
pKa2
pKa3
pKa4
pKa5
pKa6
pKa7
5.09±0.02
5.07±0.17
5.27±0.15
Acetate
acetate
water
5.62±0.04
5.92±0.09
6.35±0.17
7.67±0.30
9.38±0.16
10.93±0.11
5.83±0.11
6.99±0.13
7.31±0.09
9.08±0.24
10.65±0.09
11.51±0.21
5.69±0.03
6.80±0.19
7.99±0.06
8.64±0.14
10.24±0.15
11.01±0.17
water
water
amine
amine
The deprotonation of a metal-bound water molecule leads to the µ-OH bridge, yielding the species
[(OH2)2NiII(µ-OH)NiII(OH2)]. The pKa4 and pKa5 can be attributed to the deprotonation of water
molecules bonded to the metal centers and can be tentatively attributed to the species
[(OH )(OH)NiII(µ-OH)NiII(OH )]
[(OH )(OH)NiII(µ-OH)NiII(OH)]. These attributions are in
⇌
2
2
2
18

agreement with those for previous Ni(II) complexes described in the literature [11,27,42,43,104]
and the lower values obtained for pKa4 and pKa5 may be interpreted in terms of electrostatic
interactions between the protonated amino groups in the side chain of the complexes and the water
molecules bound to the metal centers [51,61]. The last two deprotonation constants (pKa6 and pKa7)
can be assigned to the protonation/deprotonation of the amines in the side chain of complexes 1 –
3, which is in agreement with the values for free amines in solution [105].
Figure 2. Proposed assignment of observed protonation equilibria for complexes 1 – 3 in
CH3CN/H2O 1:1 v/v solution; I = 0.1 mol L-1 (KCl) at 25 ºC.
19

3.5 Reactivity studies
Kinetics experiments were carried out in order to determine the influence of hydrogen bonding on
the hydrolysis of phosphoesters using the activated substrate 2,4-BDNPP under pseudo-first-order
conditions ([S]>> [complex]). The hydrolysis activities were investigated in an CH3CN/H2O (1:1
v/v) mixture in the pH range 5.00 – 9.50 for complexes 1 and 3 and 5.00 – 9.00 for complex 2.
The reaction was monitored spectrophotometrically at 400 nm, corresponding to the maximum
absorption of the product 2,4-dinitrophenolate (2,4-DNP). The pH dependence of the catalytic
activity for the three complexes was obtained from the plot of kobs versus pH (Figure 3). It was not
possible to determine the kinetic pKa values for any of the three complexes but, as can be seen in
Figure 3, for all complexes the kobs is highly influenced by alkaline pH.
Comparing to the complex A[43], that presented an exponential dependence on the pH, it is
possible to observe a change in the kobs vs pH dependence profile for complexes 1 – 3 (dots in
Figure 3). The profile of kobs for complexes 1 – 3, with amino groups in the side chain, presents a
bell-type shape at pH values ranging from 5.00 to 8.50 for complexes 1 and 2, and 5.00 to 7.50 for
complex 3, and then a leap at pH 9.00. The non-negligible catalytic activity presented by
complexes 1 – 3 (Tables S5 – S7) in low-pH regions – when compared to other Ni(II) complexes
reported in the literature [17,39,42,43,47] – can probably be explained by interactions involving
the protonated amino groups in the side chain of the complexes and a water molecule coordinated
to the metal center. These interactions may lead to the lowering of the pKa of the metal-bound
water molecule, which generates the required species for the hydrolysis reaction to occur, i.e., with
a hydroxo molecule coordinated to the metal center.
20

Figure 3. The kobs vs pH plot for the hydrolysis reaction of the 2,4-BDNPP (represented as black
circles) promoted by complexes 1 (A), 2 (B) and 3 (C) in CH3CN/H2O (50% v/v) at 25º C.
[Complex] = 1.00 × 10-5 to 4.00 × 10-6 mol L-1; [2,4-BDNPP] = 1.07 × 10-3 to 3.73 × 10-3 mol L-1,
along with the species distributions shown as solid lines.
21

The dependence of the reaction rate on the substrate concentration was investigated at each pH
value, in the pH range of 5.00 – 9.50 for complexes 1 and 3 and 5.00 – 9.00 for complex 2. In these
experiments it was observed that, for all complexes and pH values, initially, the hydrolysis rate
increases linearly at low substrate concentrations and reaches saturation as the concentration
increases, which suggests the formation of a complex–substrate intermediate (Figure 4). The
Michaelis-Menten [106] model was employed to obtain the kinetic parameters. The kinetic data
for complexes 1 – 3 and similar compounds described in the literature, with and without second
coordination sphere, are listed in Table 3.
Figure 4. Dependence of the initial reaction rate (v0) on the 2,4-BDNPP concentration for the
hydrolysis reaction promoted by complexes 1 - 3. Conditions: [complex] = 4.00 × 10-6 mol L-1 for
complexes 1 and 2; [complex] = 5.18 × 10-6 for complex 3; [buffer] = 50.00 × 10-3 mol L-1 (CHES,
pH = 9.00); I = 50.00 × 10-3 mol L-1 (LiClO4) in CH3CN/H2O (50% v/v) at 25 ºC.
22

Table 3 Kinetic parameters of complexes 1 – 3 for 2,4-BDNPP hydrolysis and other complexes
for comparison.
Complex vmax × 107
KM × 103
kcat × 102
(s-1)
Kass
kcat/KM
f×10-4a pH
(mol L-1 s-1) (mol L-1)
(L mol-1) (L s-1 mol-1)
Ab
1
5.37
1.57
5.37
637.0
246.3
492.6
833.3
769.2
483.1
724.6
291
34.20
12.11±1.24
1.47±0.12
29.18±8.35
2.38±0.15
15.30±2.03
2.69±0.27
3.70
13.84 9.00
12.67 9.00
1.97±0.11
0.30±0.01
1.40±0.18
4.06±0.35 4.92±0.27
2.03±0.16 0.30±0.01
1.20±0.31 3.50±0.45
1
1.55
9.02
1.60
8.15
1.92
3.25
8.84
6.50
9.00
6.50
9.00
6.50
9.00
9.00
7.00
6.50
6.50
2
2
0.31±0.009 1.30±0.07 0.31±0.01
3
1.64±0.12
0.64±0.03
1.33
2.07±0.23 3.16±0.23
1.38±0.12 0.37±0.02
3
Bc
Cd
De
Ef
Fg
3.44
1.19
0.21
27.8
3.07
1.26
3.42
0.28
0.07
0.194
0.58
840.3
4762
36.0
28.82
0.01
0.34
0.02
2.11
1.00
325
0.63
a f = kcat/kunc (catalytic factor), where kunc = 3.88 × 10-7 s-1 at 25 ºC and pH = 9.00; kunc = 1.93 × 10-7 s-1 at 25 ºC and pH = 6.50
b
[78];
[Ni2(HBPPAMFF)(µOAc)2(H2O)]BPh4, H2BPPAMFF = 2-[(N-benzyl-N-2-pyridylmethylamine)]-4-methyl-6-[N-(2-
c
pyridylmethyl)aminomethyl)]-4-methyl-6-formylphenol [43];
diisopropyl-1,4,7-triazonan-1-yl)methyl]-4-methyl-6[(pyridine-2-ylmethylamino)methyl]phenolate
[NiIINiII(L)(µ-CH3COO)2(OH2)]BPh4·H2O,
L
[27];
=
2-[(4,7-
d
[Ni2(L1)(OAc)2(H2O)]ClO4·H2O,
L1
=
2-[N-bis(2-pyridylmethyl)aminomethyl]-4-methyl-6-[N-(2-
pyridylmethyl)aminomethyl]]phenol [42]; [Ni2(LClO)(µ-OAc)2](PF6)·3H2O, L = 2,6-bis[bis(2-pyridylmethyl)aminomethyl]-4-
e
f
clorophenol [107]; [FeZn(IPCPMP)(OAc)2(CH3OH)]PF6, IPCPMP = 2-(N-isopropyl-N-((2-pyridyl)methyl)aminomethyl)-6-(N-
(carboxylmethyl)-N-((2-pyridyl)methyl)aminomethyl)-4-methylphenol [108]; [FeZn(L1bpea-apur)(µ-OH)(H2O)]ClO4, L1bpea-
g
apur
=
2-((2-(9H-Purin-6-ylamino)ethylamino)methyl)-6-(((3-((bis(2-(pyridin-2-yl)ethyl)amino)methyl)-2-hydroxy-5-
methylbenzyl)(pyridin-2-ylmethyl)amino)methyl)-4-methyl-phenol [76].
The derivatization with pendant amino groups in the complexes 1 and 3 led the KM values higher
when compared to those obtained for complex A [43] (with the aldehyde ligand H2BPPAMFF),
C [42] (with a dinucleating ligand with three pyridines metal-donor groups attached) and D [107]
(with a very similar ligand reported in this study and presenting four pyridines units). Thus, there
is a decrease in the Kass, together with a decrease in kcat leading to an overall catalytic efficiency
lower than that of complexes A [43] and C [42]. A similar behavior was presented by complex
B, reported by Xavier et al, which presented a triazacyclononane macrocyclic ring unit in its
structure (Table 3). Also, the hydrolysis reaction with complex 1 was 126,700 times faster than
the uncatalyzed reaction (kunc = 3.88 × 10-7 s-1 at 25 ºC) [78], which is comparable to the value
found for complex A [43], despite the lower catalytic efficiency value presented for 1. This can
23

be explained by the fact that complex 1 presented the highest kcat value among the three
complexes. The hydrolysis reaction with complex 2 was 90,200 times faster than the uncatalyzed
reaction (kunc = 3.88 × 10-7 s-1 at 25 ºC) [78] and showed a decrease in both KM and kcat when
compared to complex A [43], presenting the highest Kass value of the three complexes with amino
groups, as well as the highest catalytic efficiency. Figure 5 illustrates the behavior of f (kcat/kunc)
for the substrate 2,4-BDNPP hydrolysis with respect to pH for the three complexes.
As can be seen from the Table 3, complexes 1 – 3 presents higher catalytic efficiency if compared
with analogous FeIIIZnII complexes in the same pH values. Complex F [76], reported by Pereira
et al, with an aminopurine group as side chain moiety of the ligand, present a meaningful second
coordination sphere effect regarding the hydrolysis of the 2,4-BDNPP substrate, while complex
E [108], reported by Jarenmark et al has no such effect. From the values listed in Table 3, it is
possible to observe a notable increase not only in the catalytic efficiency of the systems
containing Ni(II), but also in the kcat and in the catalytic acceleration factor f. The catalyzed
hydrolysis reaction for the complexes 1 – 3 was 1.50, 1.60 and 1.92 (for 1, 2 and 3 respectively)
times faster than the hydrolysis reaction promoted by complex F [76] at pH 6.50, indicating that
even at lower pH values (Table S5 – S7), the complexes studied in this work are more active
than other analogues complexes reported in the literature and presenting phosphoesterase activity
[61,76,108].
24

Figure 5 Plot of f (kcat/kunc) vs pH for the three complexes, using the substrate 2,4-BDNPP. pH
5.00, kunc = 1.77 × 10-7 s-1; pH 5.5, kunc = 1.22 × 10-7 s-1; pH 6.50, kunc = 1.93 × 10-7 s-1; pH 7.00,
kunc = 1.89 × 10-7 s-1; pH 7.50, kunc = 1.82 × 10-7 s-1; pH 9.00, kunc = 3.88 × 10-7 s-1 [78].
To improve our understanding of the catalytic process, the phosphate conjugates of complexes 1
– 3 were optimized and are shown in Figure 6. As can be seen, one minimum-energy structure for
all phosphate conjugates was found after a proton transfer between the charged amine and a
hydroxy group coordinated to the nickel center, in which complex 2 also shows a relevant
interaction between the amine and an oxygen atom present in the substrate. This additional
interaction with the substrate may explain the higher catalytic efficiency obtained with it.
25

Figure 6. Optimized ground state geometries for the phosphate conjugates of 1, 2 and 3 using
BP86/DEF2-TZVP (metals) or DEF2-SVP (all others). Some atoms are omitted for clarity.
In order to assess the possible hydrolysis of the monoester 2,4-DNPP, a stoichiometric reaction
between complexes 1 – 3 and the 2,4-BDNPP substrate was monitored. It was observed that 2
equivalents of 2,4-DNP are released over a period of 40 h, at 25 °C, which indicates the hydrolysis
of both the mono- and the diester. Based on this information, and assuming that the hydrolysis of
2,4-BDNPP leads to the formation of an intermediate (presumably the monoester coordinated to
the catalyst), the catalytic activity of the most efficient complex 2 towards the hydrolysis of the
monoester 2,4-DNPP was also investigated. The determination of the initial rates as a function of
the substrate concentration reveals saturation kinetics presenting Michaelis-Menten behavior
(Figure 7). It was possible to obtain values of kcat = 9.60 × 10-3 s-1 and Kass = 3833.34 mol L-1,
revealing that the constant for the association of the monoester with catalyst 2 is 4 times larger,
while the kcat is 3.6 times lower when compared to the parameters obtained with the corresponding
2,4-BDNPP diester. Also, complex 2 hydrolyzes 2,4-DNPP 35,400 times faster (f = 3.54 × 104 s-
1) than the uncatalyzed reaction (kunc = 2.71 × 10-7 s-1 at 25 ºC) [109]. As the 2,4-DNPP substrate
is a dianion, it could have coordinated with the complex in bridging mode so that the μ-OH could
act as the nucleophile, which may explain the lower kcat value presented by 2 in the hydrolysis
reaction of 2,4-DNPP.
26

Figure 7. Dependence of the initial reaction rate (v0) on the monoester 2,4-DNPP concentration
for the hydrolysis reaction promoted by complex 2. Conditions: [complex] = 4.00 × 10-6 mol L-1;
[buffer] = 50.00 × 10-3 mol L-1 (CHES, pH = 9.00); I = 50.00 × 10-3 mol L-1 (LiClO4) in
CH3CN/H2O (50% v/v) at 25 ºC.
In order to obtain the number of catalytic cycles for the hydrolysis of the 2,4-BDNPP, a hydrolysis
reaction promoted by complexes 1 – 3 (4 × 10-6 mol L-1) and the 2,4-BDNPP substrate (2 × 10-4
mol L-1) was monitored at 445 nm, at pH 9.00 and 25 °C, revealing that the complexes can be
considered as efficient catalysts and are able to hydrolyze 42 molecules of substrate in 24 h for
complex 1, 58 molecules in 12 h for complex 2 and 41 molecules in 8 h for complex 3.
To establish whether a general base catalysis or an intramolecular nucleophilic attack can be
attributed to complexes 1 – 3, the isotopic effect was evaluated in the hydrolysis of 2,4-BDNPP,
under identical conditions using either H2O or D2O. The values for kH/kD for complexes 1, 2 and 3
were, respectively, 5.22, 4.22 and 7.02, indicating that there is a proton transfer in the rate-limiting
step of the reaction (kH/kD>2 ) [110], suggesting a general base reaction mechanism.
The effect of the temperature on the hydrolysis reaction was investigated under the same
conditions described above for the substrate excess experiments at 20, 25, 30, 35 and 40 ºC. The
activation parameters Ea, ΔH≠, TΔS≠ and ΔG≠ for the hydrolysis catalyzed by complexes 1 – 3
27

were obtained through the Arrhenius and the Eyring equations [111] and the results are reported
in Table 4.
Table 4. Activation parameters for the hydrolysis of 2,4-BDNPP.
Complex
Ea
(kJ mol-1) (kJ mol-1)
ΔH≠
TΔS≠
ΔG≠
(kJ mol-1)a
(kJ mol-1)
1
2
39.44
68.86
36.93
66.34
103.66
80
-47.07
-17.23
17.43
-107
84.00
83.57
86.23
111
3
106.17
HO-b
aAt 298.15 K, bGeneral basic catalysis [112]
As observed, the data indicates that for complexes 1 and 2 an organization of the reactive species
occurs in the TS (ΔS≠ < 0), while for complex 3 a value of ΔS≠ > 0 suggests that the reactive
species becomes less structured in the TS. For all complexes ΔH≠ > 0, which reflects bond breaking
in the activated complex. For the calculated phosphate conjugates, there was an increase in the
distance of the hydrogen bond connecting the lateral amine and the coordinated water from 1.475
Å for 1 to 1.617 Å for 3, and as obtained from the Eyring plots, the enthalpic contribution to the
activation barrier also increases from 1 to 3. Both results suggest a mechanism where the water
transfers one proton back to the amine before the phosphate hydrolysis. The higher entropic effect
caused by the elongation of the carbon chain, that hence loses more degrees of freedom with
binding, can also be observed from 1 to 3, in agreement with the change in entropy from -47.07 kJ
mol-1 for 1 to 17.43 kJ mol-1 for 3.
3.5.1 Proposed reaction mechanism for 2,4-BDNPP hydrolysis by complexes 1 – 3
Based on the data detailed above, the reaction mechanism shown in Scheme 2 is proposed for the
hydrolysis of the 2,4-BDNPP substrate by complexes 1 – 3. First, the protonated amine transfers
a proton to the hydroxo molecule bound to the NiII center on the hard side of the complex, which
generates the [(H2O)(OH)NiII(µ-OH)NiII(H2O)] unit. The substrate then binds in a monodentate
fashion to the Ni(II) ion by displacing its labile H2O ligand. For complexes 1 and 3 a mechanism
28

is suggested in which the NiII-bound hydroxide activates an H-bonded water molecule to act as the
reaction-initiating nucleophile [76,113] (Figure 9), with concomitant release of the 2,4-DNP
leaving group. Since stoichiometric reactions between complexes 1 – 3 and 2,4-BDNPP released
2 equiv. of 2,4-DNP, it is proposed that after the release of the first 2,4-DNP unit an intramolecular
nucleophilic attack of the µ-OH would be responsible for the hydrolysis of the bound monoester.
The remaining phosphate is then displaced by excess water, regenerating the catalytic center. For
complex 2 a very similar mechanism is proposed, in which the sole difference is the interaction
between the amine group and the substrate, which increases the proximity of the substrate and the
complex, explaining its higher catalytic efficiency when compared to complexes 1 and 3.
29