Radical Scavenging and Catalytic Activity of Metal−Phenolic

J. Phys. Chem. B 2005, 109, 24197-24202

24197

Radical Scavenging and Catalytic Activity of Metal-Phenolic Complexes Harbir S. Mahal,*,† Sudhir Kapoor,*,† Ashis K. Satpati,‡ and Tulsi Mukherjee† Radiation and Photochemistry DiVision, Analytical Chemistry DiVision, Bhabha Atomic Research Centre, Trombay, Mumbai 400085, India ReceiVed: September 1, 2005; In Final Form: October 7, 2005

A series of metal-ligand complexes were prepared by the reaction of various metal ions, namely, Cu(II), Mn(II), or Fe(II) with phenolic derivatives of [catechol, chlorogenic acid (CGA), n-propyl gallate (nPG), 3-hydoxy anthranilic acid, resveratrol, and rutin] and characterized by UV-vis spectroscopy. The metal/ ligand complexing ratio and complexation constants have been determined. The complexes were probed for their reactivity toward various free radicals (eaq-, CO2•-, and O2•-). Pulse radiolysis studies showed that the one-electron reduction of metal/phenol complexes by CO2•- radicals was metal-centered, and this was confirmed by the formation of an initial adduct with CO2•- radicals. Rate constants for the scavenging of superoxide anions with metal complexes ranged between 107-109 dm3 mol-1 s-1 and those for the reaction of eaq- with the metal complexes were in the range of (1-5) × 109 dm3 mol-1 s-1, depending on the pH of the solution. Cyclic and differential pulse voltammetric studies showed that the reduction potential of the complexes are found to range between -0.022 to 0.45 V vs normal hydrogen electrode.

Introduction It has been suggested that the superoxide free radical plays a key role in many deleterious biological processes and has been implicated in the damage caused to DNA, lipids, and enzymes.1 Inflammatory response is in part attributed to O2•- and •OH radicals. Under normal circumstances, the levels of superoxide anion (O2•-) produced by one-electron reduction of molecular oxygen through various pathways such as by normal cellular respiration,2 by inflammatory cells,3 during redox cycling of drugs, for example, quinone-semiquinone 4 (Q-SQ•-), and during arachidonic acid metabolism5 are kept under control by various endogenously present superoxide dismutase (SOD) enzymes6

SQ•- + O2 f Q + O2•-

(1)

In biological systems, a well-known mechanism of superoxide dismutation by Cu/Zn-SOD (metal-bound proteins) is that where the copper ions may be either oxidized or reduced7 according to the following reactions 2 and 3, with the net reaction being reaction 4

Cu(I) + O2•- + 2H+ f Cu(II) + H2O2

(2)

Cu(II) + O2•- f Cu(I) + O2

(3)

2 O2•- + 2H+ f H2O2 + O2

(4)

Thus, it is likely that a transition metal ion complex of Fe, Cu, or Mn with suitable ligands can act as an effective mimetic of SOD with therapeutic application8 if they have low molecular weight, high cell permeability, fair solubility in water, good * To whom correspondence should be addressed. E-mail: hsmahal@ barc.ernet.in (Dr. H. S. Mahal); E-mail: [email protected] (Dr. S. Kapoor). Fax: (+)-91-22-25505151. Tel: (+)-91-22-25590298. † Radiation and Photochemistry Division. ‡ Analytical Chemistry Division.

stability, and also if their reduction potential lies between the standard reduction potential of the oxygen/superoxide couple (-0.16 V vs NHE (normal hydrogen electrode) relative to 1 mol dm-3 O2) and that of the (O2•-/H2O2) couple at pH 7 is ∼0.94 V vs NHE.9 The reaction of superoxide ions with various low molecular weight complexes of Mn/iron with EDTA, HEDTA,10,11 tetrakis(4-N-methyl pyridyl) porphine,12 ferritin,13 cytochrome-c,14 desferrioxamine,15 Cu-histidine,16 macrocyclicpolyamine,17 and dipeptides18 has shown rather poor reaction rates with a rate constant of k ) 105-107 dm3 mol-1 s-1. A new class Mnporphyrin derivatives19 used as catalysts for scavenging superoxide radicals has been prepared. On the other hand, a few complexes of Cu-3,5-(diisopropylsalicylic acid)2 (Cu-DIPS),20 MnII-tetrakis(4-N-methylpyidyl)porphyrin (MnII-TMPyP4+),21 and Mn-β-octabromo-meso-tetrakis(N-methyl-pyridinium-4-yl)porphyrin (Mn-OBTMPyP4+)22 have shown a high reactivity with superoxide anions (k g 2 × 108 dm3 mol-1 s-1). It is suggested that the kinetic barrier to oxidation is the poor stability in neutral aqueous solutions due to the high pKa of H2O2. In case the products of the reaction become stabilized by complexation to a metal cation, oxidation by O2•- is conceivable. Flavonoids possess antioxidant and antimutagenic effects. It was previously established that the Cu-rutin complex has good ability to scavenge the superoxide radical anions.23 Catechols are biosynthesized and used as iron-sequestering agents by microorganisms. Catechols have low reduction potentials, and they can act as good ligands for forming chelates with other metal ions such as Cu and Mn. In literature, little information is available on transition metal complexes of iron, copper, and manganese with phenolic types of ligands, which behave as antioxidants and metal chelating properties24 and which may fulfill the role as synthetic low molecular weight SODs. In the present paper, we have tried to explore the mechanistic and kinetic information about the reactivity of primary and secondary species formed during the redox cycle mechanism. A pulse radiolysis technique was

10.1021/jp0549430 CCC: $30.25 © 2005 American Chemical Society Published on Web 11/24/2005

24198 J. Phys. Chem. B, Vol. 109, No. 50, 2005

Mahal et al.

eaq- + O2 f O2•-

employed to study the reactivity of free radicals; UV-vis absorption spectroscopy, XRD, and ESR techniques were used to monitor the complex formation.

k ) 2 × 1010 dm3 mol-1 s-1 (ref 28) H + O2 f HO2• y\z O2•-

H2O f •OH‚, eaq-, H•, H2O2, H2

(5)

eaq- + N2O + H2O f •OH + N2 + OH-

(6)

k ) 8.7 × 109 dm3 mol-1 s-1 (ref 28)

pKa ) 4.8

•

Experimental Section Materials. Ferrous ammonium sulfate, cupric chloride, cuprous chloride, sodium chloride, sodium formate, and manganese sulfate were from Sarabhai Merck (GR grade). Catechol, chlorogenic acid (CGA), n-propyl gallate (nPG), 3-hydroxy anthranilic acid (3-OHAA), resveratrol, curcumin, and rutin were obtained from Sigma and used as received. IOLAR grade N2 or N2O gas (purity > 99.9%) used for purging solutions was obtained from Indian Oxygen Limited. All solutions were prepared just before the experiments and kept in the dark to avoid photochemical reactions. Methods. The pulse radiolysis setup consists of an electron linear accelerator (Forward Industries, U.K.) capable of giving single pulses of 50 ns, 500 ns, or 2 µs of 7 MeV electrons. The pulse irradiates the sample contained in a 1 cm × 1 cm suprasil quartz cuvette kept at a distance of approximately 12 cm from the electron beam window, where the beam diameter is approximately 1 cm. An optical detection system comprised of a 450 W xenon arc lamp, lenses, mirrors, and a monochromator monitors the transient changes in absorbance of the solution following the electron pulse. The output from the photomultiplier tube (PMT) is fed through a DC offset circuit to the Y input of an L & T storage scope which can transfer 400 mega samples/s on each input channel at 250 ns/div time base range with a sensitivity of 5 mV/div and having a bandwidth of 100 MHz. Further details of the LINAC can be seen elsewhere.25 An aerated 10-2 mol dm-3 KSCN solution was used for dosimetry, and the (SCN)2•- radical was monitored at 475 nm. The absorbed dose per pulse was calculated26 assuming G[(SCN)2•-] ) 2.6 × 10-4 m2 J-1 at 475 nm. The dose employed in the present study, unless otherwise stated, was typically 13 Gy/ pulse. The rate constants for the reaction of several metal complexes (with differently substituted phenols) with eaq-, formate radical anions (CO2•-), and superoxide radical anions (O 2•-) have been determined by a pulse radiolysis technique by monitoring the disappearance of either eaq- or the superoxide radical anion or the formation of a product transient at an appropriate wavelength. The rates of reactions were determined by carrying out the experiments with at least four different solute concentrations. Bimolecular rate constants were derived from plots of the pseudo-first-order rates vs solute concentration. The rate constants reported are generally accurate to (15%. In the radiolysis of water, the primary species formed are OH radicals, solvated electrons, and H atoms (reaction 5), G(•OH‚) ) G(eaq-) ) 2.9 × 10-7 mol J-1, G(H•‚) ) 0.6 × 10-7 mol J-1.27 The solvated electrons can be converted into further •OH‚ radicals (reaction 6). In the presence of O2, they are converted into superoxide radical anions (O2•-) (reaction 7). The H• atoms are also scavenged by O2 and contribute to the superoxide yield at neutral pH (reaction 8). In the presence of formate ions, the ‚•OH radicals react with it (reaction 9) to give formate radicals. In the presence of O2, formate radicals convert O2 to superoxide radical anions (reaction 10)

(7)

(8)

k ) 2 × 1010 dm3 mol-1 s-1 (ref 29) OH + HCO2- f OH- + CO2•-‚

•

(9)

k ) 3.5 × 109 dm3 mol-1 s-1 CO2•-‚ + O2 f O2•-‚ + CO2

(10)

k ) 4.2 × 109 dm3 mol-1 s-1 (ref 30) •

OH/H• + (CH3)3COH f •CH2(CH3)2COH + H2O/H2 (11)

Cyclic Voltammetric Analysis. Reduction potentials of the ligands and the metal complexes with different ligands were determined by the cyclic voltammetric method. In some cases, because of the low solubility of the complexes, a differential pulse voltammetric technique was used to determine the redox potentials. All the voltammetric experiments were carried out using the Eco Chemie make Potentiostat/Galvanostat Autolab 100. Data acquisition and analysis were made by Autolab-GPES software. Electrochemical scanning was done taking the test solution (purged with high-purity nitrogen gas) in an electrochemical cell comprised of a glassy carbon working electrode, a saturated calomel electrode (SCE) as a reference electrode, and a platinum rod as the counter electrode. Preconditioning of the glassy carbon electrode was carried out prior to every measurement by polishing the surface of the electrode using very fine alumina powder and then rinsing it thoroughly before use. Aqueous solutions containing 10-3 mol dm-3 of the complex, 0.1 mol dm-3 KCl at 27 °C, were bubbled with pure N2 prior to measurement. All the experimental potentials values were then converted with respect to the NHE. The electrochemical system was calibrated by doing cyclic voltammetry and differential pulse voltammetry scans of Cd(II) and also of a standard ferricyanide/ferrocyanide couple to check the electrochemical setup daily. All the potential values were measured with respect to the SCE and converted to NHE by suitable corrections. Characterization. Absorption measurements were carried out on a Jasco-530 UV-vis spectrophotometer. The spectra were recorded at 27 °C using a 1 cm quartz cuvette. X-ray diffraction patterns were taken using a Phillips Analytical automated powder diffractometer employing Cu KR radiation. The ESR spectrum was recorded on a Bruker-ESP 300 X band EPR spectrometer operated at 100 kHz. Results and Discusions Stoichiometry of Complexes. In the present investigation, we have used iron, copper, and manganese salts in their divalent state. Mixing aqueous solutions of ligands such as CGA, nPG, 3-OHAA, resveratrol, and rutin with either Cu(II), Mn(II), and Fe(II) metal ions at near neutral pH resulted in the development of color, which is indicative of the formation of the complex. The UV-vis absorption spectra (Figure 1) show changes in the absorbance upon the addition of Cu ions to catechol solution, pH = 8. Apart form the 400 nm peak, the appearance of a broad band having λmax ∼ 615 nm observed for the Cu/catechol ratio

Metal-Phenolic Complexes

J. Phys. Chem. B, Vol. 109, No. 50, 2005 24199 5 and 6)

M + L ) ML K)

Figure 1. UV-vis absorption spectra of aqueous solution containing 2 × 10-3 mol dm-3 phosphate buffer, 10-3 mol dm-3 catechol, and (0, 30, 65, 100, 130, 160, 225, 310, 450, 590, 770, 950) × 10-6 mol dm-3 CuCl2, pH ) 8 at 28 °C.

[ML] [M][L]

(1) (2)

Where, K is the stability constant (expressed as a logarithm). M is the amount of metal ion, and L is the amount of the ligand such as phenol. The total concentration of the metal [M]0 can be expressed as [M]0 ) [M] + [ML] and [ML] ) K[M][L]. Hence, upon rearrangement, [M] ) [M]0/(1 + K[L]), which shows that the metal concentration [M] depends on the stability constant of the complex and the free concentration of the ligand [L] which is dependent on the corresponding pK and pH values. If [L]0 is the total ligand concentration, then one can write

[L] ) [L]0 - [ML] and [M] ) [M]0 - [ML] [ML] ([M]0 - [ML])([L]0 - [ML])

Figure 2. Job’s plots for Cu-catechol complex at 28 °C. Concentration of cupric chloride and catechol used was 10-3 mol dm-3 in 2 × 10-3 mol dm-3 phosphate buffer.

TABLE 1: Optical Properties of Metal Complexes in Aqueous Solutions complex

ratio

λmax nm

pH

FeII-nPG Mn-nPG Cu-catechol Cu-CGA Cu-rutin Fe-(3-OHAA) Cu-resveratrol Cu-curcumin

1:2 1:2 1:1 1:1 1:3 1:3 1:3 1:3

540 359 400 421 406 346 345 356

7.0 8.0 8.0 7.5 7.5 7.0 8.0 9.5

K, equilib const, dm3 mol-1 7.9 × 106 4 × 106 1.5 × 106 5 × 106 2 × 106 3.7 × 105 1 × 106

of 1:1 is due to the ligand-to-metal charge-transfer band. As a consequence of complex formation, Cu(II) is reduced to the Cu(I) state.31 Assuming that only one complex is formed between the ligand and metal ions, Job’s method of continuous variation was used to determine the composition of the complex. The maximum in Job’s plot corresponds to a condition where [molar fraction of metal] ) 1/(n + 1), where “n” represents the number of ligands in the complex. In the case of the Cu(II) complex with catechol, 1/(n + 1) ) 0.5 ) 1/2, which results in n ) 1. The data for other phenols and metal ions are summarized in Table 1. Figure 2 shows Job’s plot for the Cu-catechol complex. In the case of the phenolic compounds studied, it appears that the difference in the number of ligands originates from the availability of phenolic oxygens in n-propylgallate, catechol, or resrveratrol, and so forth, for the coordination bonds with the metal ions. It is important to mention here that the different complex ratio of ligands with metal ions is due to the difference in their structures as shown by Kawabata et al.32 Apparent Stability Constants of the Complexes. By keeping the concentration of the ligand constant and by varying the concentation of metal ion, it was possible to determine the stability constants of the complexes. Stability constants of a 1:1 M/L complex were determined from the following relation (eqs

(3)

since the concentration of the metal ion is very small and so is the concentration of the ligand, one can express eq 3 as

[M]0[L]0 - [L]0[ML] [ML]

(4)

[L]0 1 1 ) + [ML] K[L]0[M]0 [L]0[M]0

(5)

1 1 1 ) + [ML] K[L]0[M]0 [M]0

(6)

By plotting 1/ML vs 1/[M]0 in eq 7, we get slope 1/K[L]0. In general

1 1 1 ) + n [MLn] K[L ]0[M]0 [M]0

(7)

From this, an equilibrium constant K in dm3 mol-1 for the complex was estimated. The data for various phenols and metal ions are summarized in Table 1 and are in the range 3.7 × 105 to 7.9 × 106 dm3 mol-1, in fair agreement with the binding constant of the FeII-nPG complex (K ) 105 dm3 mol-1) reported by Reddan et al.33 The low binding constant values imply that these are weak complexes. As mentioned earlier, there exists a possibility of reduction of metal ions by phenols. Therefore, to confirm this, XRD and ESR experiments were carried out. It was observed that the diffraction pattern of Cu-nPG complexes was neither of Cu(II) nor of Cu(I). This could be due to the presence of both oxidation states of Cu in the complex. This was further confirmed by ESR experiments where it was seen clearly that the paramagnetic signal due to Cu(II) decreases drastically in the presence of phenols. This could be due to the formation of Cu(I) from Cu(II) in the presence of phenols. It is pertinent to mention here that the complex formation studies were carried out in an aerated solution, hence the possibility of a partial oxidation of phenols or metal ions cannot be ruled out completely. It was observed that the formation of the complex

24200 J. Phys. Chem. B, Vol. 109, No. 50, 2005

Figure 3. Transient absorption spectrum formed by the reaction of eaq- with the Cu-CGA complex in aqueous media at (1) 2 µs and (2) 15 µs after the pulse. Matrix: 2 × 10-3 mol dm-3 phosphate buffer, 2 × 10-4 mol dm-3 (1:1) Cu-CGA complex, 10-2 mol dm-3 tert-butyl alcohol, N2 saturated, pH 7. Dose: 13 Gy/pulse.

as well as the reduction of the metal ion are facilitated in the presence of O2. It was confirmed by forming complexes of Ag+ ions with phenols where after the reaction pure Ag metal particles were formed (results not shown). To confirm whether Cu(I) can also form a complex with phenols, Cu(I) complexes with phenols were prepared in aqueous solution containing 2 mol dm-3 NaCl. The absorption spectrum of the complex was roughly similar to that obtained on mixing Cu(II) and catechol. As the reduction potential of Cu+/Cu° has a very negative E° value (Cu+/Cu° ) -2.7 V vs NHE), it is unlikely that copper particles will be formed in the presence of phenols.34 The above observations clearly show that the presence of phenolic antioxidants may cause the reduction of metal ions. We therefore have tried to study the free radical reactions of metal-phenol complexes. Pulse Radiolysis Studies. Pulse radiolysis of an N2-bubbled aqueous solution (pH ) 7), containing 1.0 × 10-2 mol dm-3 tert-butyl alcohol, a scavenger for hydroxyl and •H atoms (reaction 11), shows an absorption maximum at ∼700 nm attributed to the formation of eaq-. The tert-butyl alcohol radical formed is quite innocuous and does not have any absorbance in the visible range. In the additional presence of (1-5) × 10-4 mol dm-3 of Cu-CGA, it was observed that the pseudo-firstorder decay rate of eaq- at 700 nm increased, with increasing solute concentration. There was also a concomitant increase in the build-up rate of the absorption band at 435 nm. By plotting the pseudo-first-order rate constants for the reaction of eaq- with Cu-CGA vs solute concentration [Cu-CGA], the bimolecular rate constant was estimated to be 5.1 × 109 dm3 mol-1 s-1 and this value correlated well with the formation rate at 435 nm. The above rate constant value is much lower than the reaction of eaq- with Cu2+ (k ) 3 × 1010 dm3 mol-1 s-1),35 confirming that Cu(II) is in the complexed form. Figure 3 shows a typical time-resolved absorption spectrum of the semi-reduced transient species of CGA formed at two different times in N2-bubbled solutions. The reactivity of eaq- with Cu-catechol and CunPG complexes is reported in Table 2. It should be noted that the present rate constant values are higher than the rate constant reported for the reaction of eaq- with Fe(III)-, Mn(III)-, Co(III)-, and Cu-transferrin complexes (k ) (0.61-2.3) × 108 dm3 mol-1 s-1).36 It is important to mention here that the observed spectra of metal complexes on reactions with eaq- were different from that observed with ligands. Reaction of CO2•- with Complexes. Figure 4 shows a typical transient absorption spectrum observed upon pulsing a N2Osaturated solution containing 0.1 mol dm3 formate (2 × 10-4

Mahal et al.

Figure 4. Transient absorption spectrum formed by the reaction of CO2•- radicals with the Cu-CGA complex in aqueous medium at 15 µs after the pulse. Matrix: 2 × 10-3 mol dm-3 phosphate buffer, 2 × 10-4 mol dm-3 (1:1) Cu-CGA complex, 10-1 mol dm-3 sodium formate, N2O saturated, pH 7.5. Dose: 13 Gy/pulse.

TABLE 2: Bimolecular Rate Constants of the Reaction of Metal-Phenol Complexes with Various Radicals radical

metal complex

eaq-. eaq-. eaq-. O2•O2•O2•O2•O2•O2•O2•O2•O2•CO2•CO2•CO2•-

Cu-CGA Cu-catechol Cu-nPG Fe-nPG Cu-resveretrol Cu-nPG Cu-rutin Cu-CGA Cu-catechol Cu-curcumin Fe(II)-3-OH-anthralinic acid Mn(II)-nPG Fe-CGA Cu-CGA Fe-nPG

ratio pH k (dm3 mol-1 s-1) 1:1 1:1 1:1 1:2 1:3 1:1 1:3 1:1 1:1 1:3 1:3 1:2 1:1 1:1 1:2

7 7 7 9 9 9 9 9 9 9 9 9 7.5 7.5 7.5

5.1 × 109 1.0 × 109 5.3 × 109 1.3 × 109 1.7 × 109 3.3 × 108 1.7 × 109 1.0 × 109 1.2 × 109 2 × 107 7.5 × 108 6.8 × 107 4.7 × 108

mol dm-3) of the Cu-CGA complex, pH ∼ 7.5. Since, in aqueous media, the monovalent copper ion (Cu+) shows an absorption maximum at ∼265 nm34 and 265 nm ) 400 dm3 mol-1 cm-1, the observed transient spectrum (λ ) 480 nm) could be due to the formation of the complex of CO2•- with the Cu(I)CGA complex. To confirm this, the above solution was exposed to repetitive electron pulses so as to convert most of the Cu2+ to Cu+, and the optical signals were recorded after application of the last pulse. The CO2•- radicals produced in the last pulse therefore had a greater probability to react with Cu+. Indeed, it was observed that with successive pulsing the concentration of Cu+ increased in the solution, hence the yield of absorption due to the complex formation also increased. Thus, it can be inferred that the absorption band around 480 nm is due to the complex formation of Cu+-CGA with the formate radical. The rate constant k for the reaction of formate radical anions with Cu-CGA was found to be 6.8 × 107 dm3 mol-1 s-1. Similar results were obtained with other complexes mentioned above. The bimolecular rate constants for the reaction of formate radicals with various complexes studied ranged between 0.68 × 108 and 7.5 × 108 dm3 mol-1 s-1 and are summarized in Table 2. The rate constants for the second-order reaction between CO2•- radicals and Cu complexes of histidine and histidyl peptides were e4.5 × 108 dm3 mol-1 s-1. The present values are in good agreement with the reported values. 37,38 These rate constants are, however, lower in comparison to the rate constants of eaq- with metal complexes obviously because CO2•- is a weaker reducing species. Reactions of Superoxide Radical Anions. As mentioned earlier, the formation of the reduced form of Cu(I) in solutions of Cu(II)-phenol complexes can result in the dismutaion of

Metal-Phenolic Complexes

J. Phys. Chem. B, Vol. 109, No. 50, 2005 24201 TABLE 3: Reduction Potentials of Phenolic Ligands and Their Complexes with Iron, Copper, or Manganese Ions as Determined by Cyclic Voltammetry

Figure 5. Transient absorption spectrum formed by the reaction of O2•- radicals with the Cu-catechol complex in aqueous medium at 20 µs after the pulse. Matrix: 2 × 10-3 mol dm-3 phosphate buffer, 8 × 10-6 mol dm-3 (1:1) Cu-catechol complex, 10-2 mol dm-3 sodium formate, O2 saturated, pH 9. Dose: 13 Gy/pulse.

superoxide radicals (reaction 2). As there exists a possibility of the presence of Cu(I) in Cu(II)-phenol complexes, it is therefore worthwhile to see the reactivity of the complexes toward O2•-. Figure 5 shows the transient absorption spectrum obtained having a λmax ) 310 nm when O2-bubbled solutions containing 8 × 10-6 mol dm-3 Cu-catechol complex and 1.0 × 10-2 mol dm-3 formate ions at pH 9 were pulsed. The tailing portion of this band toward the longer wavelengths signifies the interaction of superoxide radical anions with the complex. The spectrum of superoxide radical anions is well characterized with an absorption maximum at 245 nm and an 245 nm ) 2350 ( 120 dm3 mol-1 cm-1.39 The lifetime of the superoxide radical anions in aqueous solutions (pH ) 9) is k = 6 × 103 dm3 mol-1 s-1.39 Therefore, it can be inferred that the observed transient absorption spectrum is due the reaction of O2•- with the Cucatechol complex. By plotting the pseudo-first-order decay rates at 260 nm9 and varying the concentration of the Cu-catechol complex between 4 × 10-6 and 8 × 10-6 mol dm-3, the bimolecular rate constant for the reaction of O2•- with Cucatechol was found to be 1.0 × 109 dm3 mol-1 s-1. The observed rate constants are closer to those reported for the Cu(II) chelates with salicylate type of drugs.20 The high rate constant with superoxide radical shows that the complex mimics the enzyme for dismutaion of O2-. The reactivities of superoxide radicals with other complexes are summarized in Table 2. It was seen that the rate constant values were higher for phenolic antioxidants having two OH groups in the ortho position rather than in the meta position. However, the presence of a different oxidation state of Cu in the complex may have important implications in the biological reactions and therefore has to be studied in detail. A very generalized mechanism39 can be represented as follows through reactions 12-17

2O2•-‚ + 2H2O f H2O2 + O2 + 2OH-

(12)

LnM2+ + O2•-‚ a LnMO2+

(13)

LnMO2+ + O2•-‚ + 2H+ f LnM2+ + H2O2 + O2

(14)

LnMO2 + 2H+ f LnM3+ + H2O2

(15)

LnM3+ + O2•- a LnMO22+

(16)

LnMO22+ f LnM2+ + O2

(17)

where “n” stands for the number of ligand molecules “L” attached to the metal atom “M”. Redox Potential of Metal Complexes. The redox potential reflects the catalytic effects of metal ions present in the complex

complex

pH

E° (V) vs NHE

FeII-nPG n-PG Mn-nPG Cu-resveratrol resveratrol FeII-3-OH AA 3-OH anthranilic acid Cu-curcumin curcumin Cu-CGA CGA Cu-catechol catechol Cu-rutin rutin

7 7 7.0 10.5 10.5 7 7 9 9 7.5 7 7 7 7 7

0.450 0.42 0.038 0.3 0.5 & 0.74 0.33 0.58 -0.022 0.38 0.251 0.40 0.277 0.442 0.325 0.475

and the extent to which they alter the potential with respect to the ligand alone. Most of the ligands, namely, CGA, nPG, resveratrol, catechol, and so forth, are potent antioxidants, that is, electron donors. To investigate if their antioxidant action is affected by complexation with metal ions, cyclic voltammetry at pH 7, 9, or 10.5 was performed to determine the oxidation potentials of the ligands and their metal-complexed forms. In the case of Cu(I), which is considered to be a soft acid, the “d” shell has 10 electrons and prefers four coordination or trigonal three coordination geometries. It has affinity to bind with soft donors. The placement of such soft ligands in the coordination sphere also alters the Cu(II)/Cu(I) reduction potential. Complexation of metal ions such as Cu, Fe, and Mn with phenolic antioxidants results in the oxidation of phenols and reduction of metal ions. However, in the presence of oxygen, the reduced metal is reoxidized with the formation of superoxide. Thus, under aerated conditions such as the one used in our experiments, an equilibrium is likely to occur even in the complexed form, wherein the existence of both Cu(II) and Cu(I) ions is to be expected in the aqueous phase. It is therefore of significance to determine the reduction potentials of the metal chelates used to scavenge superoxide anions by the cyclic voltammetric method. Redox potentials of different systems were measured using cyclic voltammetry (CV) or differential pulse voltammetry, and the values are reported in Table 3. In the case of a reversible electrochemical system, E° was estimated from the cyclic voltammetric scans (at 50 mV s-1) by taking a mean of the cathodic and anodic peak potentials, that is, E° ) (Epc + Epa)/ 2. Whereas, in the case of irreversible systems, differential pulse voltammetry (DPV) was carried out with pulse amplitudes of 25 and 50 mV and the E° value was estimated from E° ) (Ep + Eh)/2, where Ep is the voltammetric peak potential and Eh is the applied pulse amplitude. Electrochemical scans were repeated three times, and the average E° value is reported. Compared to the reduction potentials of the ligands in aqueous solutions, the reduction potentials of the complexes with the metal ions are shifted to more cathodic potentials when present in the same environment thus indicating the greater ease with which they can act as reductants. The relatively small changes in the reduction potential values are due to weak interactions between the ligand donor atom and the metal ion. The observed E1/2 of the Cu(II) CGA complex is ∼0.25 V vs NHE. Whereas the redox reaction of O2 in neutral medium is as follows (reactions 18 and 19)

O2 + e ) O2•-

E° ) -0.16 V vs (NHE)

(18)

24202 J. Phys. Chem. B, Vol. 109, No. 50, 2005

O2•- + 2H+ + e ) H2O2

E° ) 0.94 V vs (NHE)

Mahal et al.

(19)

Therefore, by comparing the redox potential value of the Cu(II)-CGA complex with that of O2 in neutral medium, it is indicated that the Cu(II) complex could always be reduced but never be oxidized by O2•-. Thermodynamically feasible reactions could be O2•- giving up one electron to form O2 (step1) and Cu(II)-CGA is reduced to Cu(I)-CGA. Another feasible reaction is Cu(I)-CGA is oxidized to Cu(II)-CGA thereby reducing O2•- to H2O2 (step 2). In all, it shows SOD-like activity. Similar observations have also been made with other Cu complexes (Table 1). Therefore, the present study shows that the metal complexes formed with phenolic antioxidants show higher SOD-like activity and higher complexation stability. Conclusions Pulse radiolysis studies have shown that the metal complexes of Mn(II)-, Fe(II)-, or Cu(II)- with phenolic antioxidants are able to undergo electron-transfer reactions. In vitro studies have revealed that the rate constants for the scavenging of superoxide radical anions by the metal-phenol complexes are fairly comparable to those known for natural SODs at near physiological pHs. Hence, the likely mechanism is also thought to be similar to that observed in the case of natural SODs. The reduction potential of the complexes lends support to the prediction of the feasibility of the reaction with superoxide radicals. Acknowledgment. The authors thank Dr. M. C. Rath for giving some useful suggestions during the preparation of this manuscript. References and Notes (1) Cuzzocerea. S.; Riley, D. P.; Caputi, A. P.; Salvemini, D. Pharmacol. ReV. 2001, 53, 135. Chuaqui, C. A.; Petkau, A. Radiat. Phys. Chem. 1987, 30, 365. Murphy, M. P.; Echtay, K. S.; Blaikie, F. H.; AsinCayuela, J.; Cocheme´, H. M.; Green, K.; Buckingham, J. A.; Taylor, E. R.; Hurrell, F.; Hughes, G.; Miwa, S. Smith, R. A. J.; Brand, M. D. J. Biol. Chem. 2003, 278, 48534. (2) Rice-Evans, C. A. In Free radical damage and its control; RiceEvans, C. A., Burdon, R. H., Eds.; Elsevier Science: Amsterdam, The Netherlands, 1994. (3) Goldstein, S.; Czapski, G. Free Radical Res. Commun. 1991, 1213, 5. (4) Cohen, G.; Heikkila, R. E. J. Biol Chem. 1974, 249, 2447. (5) Czerniecki, B. J.; Witz, G. Carcinogenesis 1989, 10, 1769. (6) Fridovich, I. J. Biol. Chem. 1989, 264, 7761.

(7) Chaudie`re, J. Some Chemical and bio-chemical constraints of oxidative stress in living cells. In Free Radical Damage and its control; Rice-Evans, C. A., Burdon, R. H., Eds.; Elsevier Science: Amsterdam, The Netherlands, 1994; pp 25-66. (8) Riley, D. P. Chem. ReV. 1999, 99, 2573. (9) Goldstein, S.; Czapski, G.; Meyerstein, D. J. Am. Chem Soc. 1990, 112, 6489. (10) Ilan, Y. A.; Czapski, G. Biochim. Biophys. Acta 1977, 498, 386. (11) Butler, J.; Halliwell, B. Arch. Biochem. Biophys. 1982, 218, 174. (12) Pasternack, R. F.; Halliwell, B. J. Am. Chem. Soc. 1979, 101, 1026. (13) Buettner, G. R.; Saran, M.; Bors, W. Free Radical Res. Comm. 1986, 2, 369. (14) Fridovich, I. Cytochrome-c. In CRC Handbook of Methods for Oxygen Radical Research; CRC Press: Boca Raton, FL, 1985; pp 213215. (15) Goldstein S, Czapski G. Free Radical Res Commun. 1990, 11, 231. (16) Weinstein, J.; Bielski, B. H. J. J. Am. Chem. Soc. 1979, 102, 2, 4916. (17) Kimura, E.; Sakonaka, A.; Nakamoto, M. Biochim. Biophys. Acta. 1981, 678, 172. (18) Fengmei, L.; Ying, X.; Xixian, D.; Xuan, Z.; Dianhua, C. Radiat. Phys. Chem. 2002, 63, 675. (19) Batinic-Haberle, I.; Spasojevic, I.; Stevens, R. D.; Hambright, P.; Neta, P.; Okado-Matsumoto, A.; Fridovich, I. J. Chem. Soc., Dalton Trans. 2004, 1696. (20) Younes, M.; Lengfelder, E.; Zienau, S.; Weser, U. Biochem. Bophys. Res. Commun. 1978, 81, 576. (21) Wienraub, D.; Levy, P.; Faraggi, P. Int. J. Radiat. Biol. 1986, 50, 649. (22) Faulkner, K. M.; Liochev, S. I.; Fridovich, I. J. Biol. Chem. 1994, 260, 23471. (23) Moridani, M. Y.; Pourahmed, J.; Bui, H.; Siraki, K.; O’Brein, P. J. Free Radical Biol. Med. 2003, 34, 243. (24) Letan, A. J. Food Sci. 1966, 31, 395. (25) Mukherjee, T. In Atomic Molecular and Cluster Physics; Ahmad, S. A., Ed.; Narosa: New Delhi, India, 1997; p 299. (26) Buxton, G. V.; Stewart, C. R. J. Chem. Soc., Faraday Trans. 1995, 91, 279. (27) Spinks, J. W. T.; Woods, R. J. An Introduction to Radiation Chemistry; J. Wiley & Sons Inc.: New York, 1990. (28) Ross, A. B. Natl. Stand. Ref. Data Ser. U.S., Natl. Bur. Stand. 1975, No. 43. (29) Bielski, B. H. J. Photochem. Photobiol. 1978, 28, 645. (30) Ilan, Y. A.; Rabani, J. Int. J. Radiat. Phys. Chem. 1976, 8, 609. (31) Li, Y.; Seacat, A.; Kuppusamy, P.; Zweier, J. L.; Yager, J. D.; Trush, M. A. Mutat. Res. 2002, 518, 123. (32) Kawabata, T.; Schepkin, V.; Haramaki, N.; Phadke, R. S.; Packer, L. Biochem. Pharmacol. 1996, 51, 1569. (33) Reddan, J. R.; Giblin, F. J.; Sevilla, M.; Padgaonkar, V.; Dziedzic, D. C.; Leverenz, V. R. Misra, I. C.; Chang, J. S.; Pena, J. T. Exp. Eye Res. 2003, 76, 49. (34) Ershov, B. G.; Janata, E.; Michaelis, M.; Henglein, A. J. Phys. Chem. 1991, 95, 8996. (35) Buxton, G. V.; Greenstock, C. L.; Helman, W. P.; Ross, A. B. J. Phys. Chem. Ref. Data 1988, 17, 513. (36) Bielski, B. H. J.; Cabelli, D. E.; Arudi, R. L. Ross, A. B. J. Phys. Chem. Ref. Data 1985, 14, 1041. (37) Bettelheim, A.; Faraggi, M. Radiat. Res. 1977, 72, 71. (38) Ilan, Y.; Ilan, Y. A.; Czapski, G. Biochim. Biophys. Acta. 1978, 503, 399. (39) Bielski, B. H. J.; Cabelli, D. E. Int. J. Radiat. Biol. 1991, 59, 291.