Kinetico-mechanistic Study on the Oxidation of Biologically Active Iron

Nov 1, 2017 - Air oxidation of methanolic solutions of biologically active tridentate pyridyl thiosemicarbazone (TSC) complexes of the general formula...
0 downloads 0 Views 899KB Size
Article Cite This: Inorg. Chem. 2017, 56, 14284-14290

pubs.acs.org/IC

Kinetico-mechanistic Study on the Oxidation of Biologically Active Iron(II) Bis(thiosemicarbazone) Complexes by Air. Importance of NH···O2 Interactions As Established by Activation Volumes Paul V. Bernhardt,† Miguel A. Gonzálvez,‡ and Manuel Martínez*,‡ †

School of Chemistry and Molecular Biosciences, University of Queensland, Brisbane 4072, Australia Departament de Química Inorgànica i Orgànica, Secció de Química Inorgànica, Universitat de Barcelona, Martí i Franquès 1−11, E-08028 Barcelona, Spain

Downloaded via UNIV OF CALIFORNIA SANTA BARBARA on July 1, 2018 at 14:54:44 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.



S Supporting Information *

ABSTRACT: Air oxidation of methanolic solutions of biologically active tridentate pyridyl thiosemicarbazone (TSC) complexes of the general formula [FeII(TSC)2] has been studied at varying dioxygen concentrations, temperatures, and pressures. The data collected indicate that the activation entropy of the reaction increases linearly with the redox potential of the complexes in a more definite way than the activation enthalpy. However, a very distinct behavior is observed for the values of the activation volumes, which do not follow the expected entropy−volume parallel trend for all of the systems studied. The involvement of important interactions between the terminal NH groups of the coordinated TSC ligand and molecular dioxygen has been found to be significant by measurements carried out at varying hydrostatic pressures. Kinetic experiments run on analogous N-deuterated complexes confirm the importance of these noncovalent interactions, which are weaker for the less acidic ND groups. These interactions show the existence of an ordering/expansion process upon going from the reactants to the transition state, whenever an interaction between the polar terminal amino groups and dioxygen can be established.



INTRODUCTION Thiosemicarbazones (TSCs) are of particular interest in medicinal chemistry because of their anticancer, antibacterial, and antiviral activity.1 These compounds show antileukemic effects, which have been attributed to their inhibition of irondependent enzyme ribonucleotide reductase, which catalyzes the rate-limiting step of DNA synthesis where ribonucleotides are converted to the corresponding 2′-deoxyribonucleotide.2 In recent years, studies have suggested that this inhibition may be due to iron chelation by the TSC,3 thus disabling electron transfer in the enzyme, which normally cycles between its ferrous and ferric oxidation states.4 Although an essential element for all forms of life, high concentrations of iron within a cell may lead to the generation of cytotoxic reactive oxygen species (ROS), namely, superoxide and hydroxyl radicals, through Fenton chemistry.5 In this respect, TSCs have a potential dual role to play in being avid iron chelators (and being able to deactivate iron-dependent enzymes) and also forming redox-active complexes that are capable of generating ROS within a cell.6 TSC ligands are promising drugs from a pharmacological perspective because they can fight against tumor growth and metastasis and overcome drug resistance through multiple molecular targets.7 Furthermore, their characteristics as ligands allow fine-tuning of their structure, and the chemical properties © 2017 American Chemical Society

of their iron complexes can be altered accordingly, which, in turn, may modify their biological activity. The TSC ligands in Scheme 1 bind in their monoanionic form to iron with deprotonation of the NH group in the 2 position occurring at neutral pH. The redox potentials of the pyridyl TSC complexes shown in Scheme 1 typically fall in the range of ca. −50 to +200 mV versus NHE. Therefore, both the FeII- and FeIII(TSC) complexes are isolable, and oxidation of FeII is typically slow and easily measurable in solution. Because the redox activity of the Fe(TSC) complexes is central to their mode of action, here we present a kinetico-mechanistic study of the oxidation of [FeII(TSC)2] complexes from the ApT, DpT, and BpT series (Scheme 1) by dioxygen in methanol (MeOH) solution, with this being the solvent of choice becaue of limited solubility in water. Furthermore, evaluation of the thermal activation parameters provides information about the influence of the previously determined redox potentials in the kinetics of the process. This feature has already been found to have a rather unexpected behavior for the reaction of FeIII(TSC) complexes with oxyhemoglobin, where hydrogen bonding, rather than redox potential variations, was the dominant feature; 8,9 very recently, the unequivocal influence of Received: September 15, 2017 Published: November 1, 2017 14284

DOI: 10.1021/acs.inorgchem.7b02381 Inorg. Chem. 2017, 56, 14284−14290

Article

Inorganic Chemistry Scheme 1

interaction−activation strain models has also been analyzed in depth.10 In this respect, the dependence of the reaction rate constant on the pressure will also be used to obtain relevant information about the mechanism and the influence of NH···O2 interactions.11−13



RESULTS The oxidation reactions of the ferrous bis(thiosemicarbazone) complexes [FeII(TSC)2] of the HDpT analogues in Scheme 1 have been studied in aerated methanolic solutions at varying temperatures and dioxygen concentrations. In order to have reasonable absorbance changes, the concentration of the iron(II) complex had to be greater than ca. 4 × 10−5 M, which reduced de facto to half an order of magnitude the excess concentration range of dioxygen if pseudo-first-order conditions were to be maintained (4 × 10−4 M < [O2] < 2.1 × 10−3 M; MeOH air-saturation limit).14,15 As a consequence, even though the oxidation reaction is expected to occur via an outersphere mechanism with a rate law such as that indicated in eq 1,16,17 the observed trend of the observed pseudo-first-order rate constants (kobs) as a function of [O2] is linear (Figures 1 and S1). This indicates that KOS(O2) ≪ 1, so either KOS (the equilibrium constant for formation of the outer-sphere encounter complex) is inherently small or the limited values of [O2] achievable in MeOH are too low to approach the saturation behavior and the eventual plateau of kobs expected from eq 1. Previous studies conducted by us on the [FeIII(TSC)2]+ oxidation of oxyhemoglobin have revealed a similar behavior,8,9 producing an operative rate law of kobs = ketKOS[reactant] because the denominator of eq 1 is effectively unity. kobs =

ketK OS[O2 ] 1 + K OS[O2 ]

Figure 1. Plot of the [O2] dependence of the values of the observed pseudo-first-order rate constant of the single step observed upon oxidation of [FeII(Dp44mT)2] in MeOH solution at different temperatures.

From the slopes of these plots, the values of the second-order rate constants for the oxidation process (k = ketKOS) collected in Table 1 can be calculated. Furthermore, from the temperature dependence of these values, and using the Eyring equation, the values of the thermal activation parameters (ΔH⧧ and ΔS⧧) can also be determined (Table 1). Considering the very good linearity observed in Figures 1 and S1, the value of the second-order rate constant can be reasonably calculated as k = {kobs/[O2]}sat, thus allowing for a simplified experimental approach for some of the ApT, DpT, and BpT analogues studied. The values of the second-order rate constants and thermal activation parameters obtained for those systems are also included in Table 1 (Figure 2a). Using this simplified methodology, the value of the activation volumes indicated in Table 1 have also been determined (Figure 2b) from ln({kobs/ [O2]}sat) versus P plots.

(1) 14285

DOI: 10.1021/acs.inorgchem.7b02381 Inorg. Chem. 2017, 56, 14284−14290

Article

Inorganic Chemistry

Table 1. Summary of the Kinetic (298 K) and Thermal and Pressure Activation Parameters Determined for the Systems Indicated in a MeOH Solutiona 298

TSC

k = ketKOS/s−1 M−1

HDpT HDp4mT HDp44mT HApT HAp4mT HAp44mT HBpT HBp4mT HBp44mT a

0.81 0.64 0.69 9.6 7.3 2.8 1.6 0.93 1.1

ΔH⧧/kJ mol−1 b

E°(FeIII/II)/mV +165 +153 +166 +20 −3 +49 +120 +108 +119

55 55 57 39 37 43 50 49 51

± ± ± ± ± ± ± ± ±

1 5 4 2 1 1 2 1 2

ΔS⧧/J K−1 mol−1 b −64 −66 −59 −97 −106 −94 −75 −83 −75

± ± ± ± ± ± ± ± ±

284

ΔV⧧/cm3 mol−1 +3 ± 1 −6 ± 1 −13 ± 1

2 15 12 5 3 1 5 1 4

b b

−8 +8 −3 −9

± ± ± ±

1 1 1 1

Electrochemical data (vs NHE)18−20 are also included for further discussion. bToo fast for conventional measurement.

Figure 2. (a) Eyring plot of the temperature dependence of the second-order rate constant, {kobs/[O2]}sat, for oxidation of MeOH solution of [FeII(TSC)2] complexes of the ApT series. (b) ln k versus P plot for two of the DpT systems in MeOH solution at 21 °C.



DISCUSSION The complete reduction of dioxygen to water is a four-electron process, so the stoichiometric reaction would require the highly unlikely reaction where four iron(II) complexes react in succession with the same dioxygen molecule, i.e., 4[FeII(TSC)2] + O2 + 4H+ → 4[FeIII(TSC)2]+ + 2H2O. The more likely scenario in a protic solvent (MeOH in this case) is illustrated by eq 2(a)−(d) and as in other published studies of oxidation by dioxygen.21−23

determining. As seen in Figure 3, time monitoring of the UV− vis spectral changes shows the presence of various isosbestic

[Fe II(TSC)2 ] + O2 → [Fe III(TSC)2 ]+ + O2•− (a) [Fe II(TSC)2 ] + O2•− + 2H+ → [Fe III(TSC)2 ]+ + H 2O2

(b)

[Fe II(TSC)2 ] + H 2O2 + H+ → [Fe III(TSC)2 ]+ + •OH + H 2O

(c )

[Fe II(TSC)2 ] + •OH + H+ → [Fe III(TSC)2 ]+ + H 2O

(d )

2O2

•−

+

+ 2H → O2 + H 2O2

2H 2O2 → O2 + H 2O2

Figure 3. Time-resolved (total time 12000 s) changes in the UV−vis spectrum for the air oxidation of [FeII(Dp44mT)2] (4 × 10−5 M) in MeOH. [O2] = 2.15 × 10−3 M and T = 15 °C.

points, which indicate that the iron(II) complex converts into the iron(III) analogue without any decomposition or intermediate species. Furthermore, the fact that the initial spectrum agrees with that expected for the [FeII(TSC)2] complex, under the experimental conditions,18−20,24 also indicates that no reaction occurs before oxygen is introduced. The Specf it or ReactLab global spectral analysis software25,26 modeled the time-resolved data to a single-exponential process, which indicates that there is a single rate-determining reaction in the sequence shown in eq 2. Taking into account the fact that no initial fast oxidation is occurring on the solutions (see above), either eq 2(c) is very fast or it is irrelevant under the

(i ) (ii)

(2)

Equation 2(b) can be disregarded, given the fact that decomposition of superoxide [eq 2(i)] is much faster under the conditions used in this study (MeOH with traces of water); consequently, eq 2(ii) and thus eq 2(c) may become even more relevant than initially expected. Given the fact that the remaining radical reaction (d) in eq 2 is expected to be very fast, the processes indicated by (a) and (c) in eq 2 are rate 14286

DOI: 10.1021/acs.inorgchem.7b02381 Inorg. Chem. 2017, 56, 14284−14290

Article

Inorganic Chemistry

Figure 4. Plot of the thermal activation parameters, ΔH⧧ (a) and ΔS⧧ (b), versus E°(FeIII/II) in MeOH solution.

activation (ordering/contraction)28 does not exist for all of the systems studied here. Although, for systems with TSC ligands with −NMe2 terminal groups, the values of both ΔSP and ΔVP show the expected negative values arising from the contributions of both KOS and the electron-transfer step (generating a smaller iron(III) complex and a more electrostricting O2•− anion), in Figure 2b, it is clear that the negative activation volumes are only observed for the ligands with no H atoms in the N4 position. Figure 5 visualizes the trends observed. It

conditions used. Consequently, either hydrogen peroxide (H2O2) is a dead-end compound or reaction (ii) is dominant for the possible disappearance of H2O2. In this respect, experiments run at [H2O2] = [FeII]/2 = 1.5 × 10−5 M (