Molecular Mechanism of Hydrogen Peroxide Conversion and

Toxicol. , 2001, 14 (10), pp 1353–1362 ... Publication Date (Web): September 11, 2001 ... JBIC Journal of Biological Inorganic Chemistry 2006 11 (4)...
0 downloads 0 Views 139KB Size
OCTOBER 2001 VOLUME 14, NUMBER 10 © Copyright 2001 by the American Chemical Society

Articles Molecular Mechanism of Hydrogen Peroxide Conversion and Activation by Cu(II)-Amikacin Complexes Małgorzata Jez˘ owska-Bojczuk,*,† Wojciech Les´niak,† Wojciech Bal,†,| Henryk Kozłowski,† Kazimierz Gatner,† Adam Jezierski,† Jarosław Sobczak,† Stefano Mangani,‡ and Wolfram Meyer-Klaucke§ Faculty of Chemistry, University of Wrocław, F. Joliot-Curie 14, 50-383 Wrocław, Poland, Institute of Biochemistry and Biophysics, Polish Academy of Sciences, Warsaw, Poland, Department of Chemistry, University of Siena, Via Aldo Moro, I-53100 Siena, Italy, and EMBL-Hamburg Outstation, Notkestrasse 85, D-22603 Hamburg, Germany Received February 27, 2001

The interactions between Cu(II)-amikacin complexes [Cu(II)-Ami] and hydrogen peroxide were studied by spectroscopy (EPR, UV-vis, CD, XAS) and cyclic voltammetry. A monomerdimer equilibrium was detected at complex concentrations above 5 mM (log Kdim ) 1.84 ( 0.03). The dimeric complex undergoes easy, although irreversible oxidation (ca. 0.5-0.6 V) to a Cu(III) species on platinum electrode. However, the monomeric complexes are able to catalyze hydrogen peroxide disproportionation reaction at pH 7.4 in a multistep process, mediated by hydroxyl radicals and involving both Cu(I)/Cu(II) and Cu(II)/Cu(III) redox pairs.

Introduction Novel or enhanced antibiotics are extremely needed to combat emerging antibiotic-resistant bacterial strains. Aminoglycosides are an important group of antibiotics, frequently used against bacterial infections, especially in severe sepsis. Unfortunately, widespread multi-drug resistance and a relatively high toxicity limit their usage. Amikacin (Ami)1 is a semisynthetic aminoglycoside antibiotic (a derivative of kanamycin A), active against * To whom correspondence should be addressed. Phone: +48-713757-281. Fax: +48-71-3282-348. E-mail: [email protected]. wroc.pl. † University of Wrocław. | Polish Academy of Sciences. ‡ University of Siena. § EMBL-Hamburg Outstation.

Gram-negative bacteria, with a broad spectrum of activity. It kills bacteria by inhibiting the translation step in microbial protein synthesis and subsequently damaging cytoplasmic membrane (1, 2). Its major disadvantage, common to aminoglycosides, is the oto- and nephrotoxicity (3-5). In consequence, amikacin has a narrow therapeutic window for blood plasma concentrations, reducing its use in therapy to life-threatening situations (6, 7). Aminosugars are specific and effective ligands for Cu(II) ions. The amino nitrogen serves as the metal anchor1 Abbreviations: Ami, amikacin; Cu(II)-Ami, cupric complex of amikacin; CT, charge transfer; CV, cyclic voltammetry; NDMA, N,Ndimethyl-p-nitrosoaniline; NBT, Nitro Blue Tetrazolium; TSP, sodium (3-trimethylsilyl)-2,2,3,3-tetradeuteriopropionate; XAS, X-ray absorption spectroscopy; XANES, X-ray absorption near edge structure; EXAFS, extended X-ray absorption fine structure; V0, initial reaction rate.

10.1021/tx010046l CCC: $20.00 © 2001 American Chemical Society Published on Web 09/11/2001

1354

Chem. Res. Toxicol., Vol. 14, No. 10, 2001

Figure 1. Molecule of Cu(II)-Ami at pH 7.4, based on a previous work (14). The CuHAmi and CuAmi complexes, differing by the presence of the H atom at the B5 oxygen are present at equimolar equilibrium at this pH.

ing donor, while a suitably located hydroxyl group completes the chelate ring (8-12). Aminoglycoside antibiotics contain multiple aminosugar and aminocyclitol rings, with the number of amino functions typically varying between 4 and 6. We undertook systematic studies of coordination abilities of aminoglycosides and related antibiotics toward Cu(II), including 1-deoxynojirimycin (13), amikacin (14), kanamycin B (15), geneticin (16), tobramycin (17), lincomycin (18) and kasugamycin (19). We found that aminoglycosides are strong Cu(II), chelators. We also found that such complexes may be very effective mediators of oxidative reactions at pH 7.4 in the presence of H2O2, exerting very efficient oxidation of 2′-deoxyguanosine (dG) and DNA double strand scission in vitro (1315, 18, 19). Using a combined potentiometric and spectroscopic methodology, we demonstrated that at pH 7.4 amikacin (Ami) forms two major complexes with Cu(II) ions, with CuHAmi and CuAmi stoichiometries (14). The pKa of formation of CuAmi is 7.57. In both these complexes Cu(II) is coordinated to the B ring donor atoms: the amino and amide nitrogens and the B5 oxygen. The difference between CuHAmi and CuAmi regards the protonation state of this oxygen. These complexes are presented in Figure 1. As the next step of our studies, in this paper we would like to present the elements of the mechanism of H2O2 activation at pH 7.4. The relevance of these studies is supported by recent findings indicating the involvement of free radicals and metal ion coordination in the mechanism of toxicity of aminoglycosides (20). For the sake of simplicity of description, the mixtures of CuHAmi and CuAmi are given a common name Cu(II)-Ami throughout the text. The detailed study of reactivity of Cu(II)-Ami toward dG, DNA and RNA is currently in progress in our laboratory.

Experimental Procedures Materials. Amikacin (Ami) was obtained from Fluka (Buchs, Switzerland). CuCl2, KNO3, K2SO4, ethanol, Nitro Blue Tetrazolium (NBT), N,N-dimethyl-p-nitrosoaniline (NDMA), H2O2, 5,5-dimethylpyrroline-N-oxide (DMPO), Chelex-100 resin, so-

Jez˘ owska-Bojczuk et al. dium and potassium phosphates, and all other simple chemicals were purchased from Sigma Chemical Co. (St. Louis, MO). 2,2,6,6-Tertmethylpiperidin-1-oxyl (TEMPO) was obtained from Aldrich (Milwaukee, WI). Electronic Absorption (UV-Vis). The electronic absorption spectra were recorded at 25° C on a Beckman DU-650 (Beckman, Palo Alto, CA) spectrophotometer over the spectral range of 200-900 nm, in 1 cm cuvettes. N,N-Dimethyl-p-nitrosoaniline (NDMA), a hydroxyl radical scavenger, and Nitro Blue Tetrazolium (NBT), a superoxide anion radical scavenger, were used in separate experiments as reporter molecules in tests of oxygen radical formation by Cu(II)-Ami complexes, at concentrations of 20 µM and 0.1 mM, respectively. These reaction mixtures also contained sodium phosphate buffer (5 mM), Ami and Cu(II) (each 50 µM), and H2O2 (0.5 mM). In preliminary studies, the course of reaction was monitored by periodically recording the whole spectra, to find out whether any absorbing reaction products would interfere with the detection of radicals. After eliminating this possibility, the reactions were monitored at the characteristic wavelength of the NDMA (440 nm). Each experiment was repeated five times. The NBT studies revealed no reactivity. CD Spectroscopy. The spectra were recorded at 25 °C on a JASCO J-715 spectropolarimeter (JASCO, Japan Spectroscopic Co., Hiroshima, Japan), over the range of 300-800 nm, using 1 cm cuvettes. Samples with 1:1 Cu(II)-to-Ami ratios and pH 7.4 were used, at complex concentration varying between 1 and 100 mM. The spectra are expressed in terms of ∆ ) l - r, where l and r are molar absorption coefficients for left and right circularly polarized light, respectively. EPR. The spectra of the Cu(II)-Ami complex at pH 7.4 were recorded at room temperature (r.t.) and at 77 K on a Bruker ESP 300E spectrometer (Karlsruhe, Germany) at the X-band frequency (9.3 GHz). Ethane-1,2-diol:water (1:2) was used as solvent in low-temperature experiments to obtain homogeneity of frozen samples. The concentration range of the samples was identical to that used in CD studies. H2O2 was added to some of the r.t. samples at 10-fold molar excess over the complex. In direct measurements of radical formation, EPR spectra were recorded at r.t., on an X-band spectrometer, model SE/X 2543 (Radiopan, Poznan˜, Poland) during 3 h after the mixing of reaction components. The instrumental conditions, typical for the quantitative EPR technique, were field setting of 334 mT, scan range of 10 mT, microwave power of 4 mW, and modulation amplitude of 0.1 mT. Concentrations of DMPO and TEMPO were 50 mM and 6 µM, respectively. The samples of reaction mixtures were otherwise prepared in the same way as in spectrophotometric measurements of kinetics of oxygen radical formation. The height of the low field line (h+1) of the TEMPO EPR signal was used as a measure of its relative intensity. NMR. Proton NMR spectra of 10 mM Ami samples in D2O, containing varied amounts of Cu(II), were recorded at 25 °C, on a Bruker AMX-300 spectrometer (Karlsruhe, Germany), at 300 MHz; TSP [sodium (3-trimethylsilyl)-2,2,3,3-tetradeuteriopropionate] was used as internal standard in the samples not containing Cu(II). In the presence of Cu(II), the signal of TSP broadened, indicting interaction. In these samples, tert-butyl alcohol was used as internal standard instead. All chemical shifts are expressed in δ (ppm), as relative to TSP. The pH* (pH reading of the electrode not corrected for isotope effects) of the samples was adjusted by adding small volumes of concentrated DCl or NaOD. Cyclic Voltammetry. CV measurements were done using an Autolab PG-12 potentiostat-galvanostat (Eco Chemie BV, Utrecht, Netherlands) with GPES 4.8 software, at voltage sweep rates between 10 and 100 mV/s. The working electrodes were a planar Pt electrode (2 mm diameter) and planar glassy carbon electrode (4 mm diameter). The auxiliary electrode was platinum wire and the reference electrode was Ag/AgCl (1 M KCl) with a Vycor tip. The supporting electrolyte was 100 mM K2SO4. Measurements were performed at 25 °C, under nitrogen. The concentrations of reagents in samples were varied between 1

Mechanism of H2O2 Activation by Cu(II) Complex of Amikacin and 60 mM and the Cu(II)-to-Ami ratio was kept constant at 1, where appropriate. The pH of the samples containing Ami was set to 7.4 with small amounts of concentrated NaOH or H2SO4. Additionally, some 1 mM samples were set to pH 7.0 or 8.1. The samples of Cu(II) alone were set to pH 6.5 to avoid precipitation of Cu(OH)2. All potentials are given relative to N. H. E. Dioxygen Generation Measurements. The generation of dioxygen in solutions containing combinations of Cu(II), Ami, and H2O2 was studied at 25 °C, for equimolar samples of Cu(II) and amikacin (concentrations from 50 µM to 50 mM) with varied concentrations of H2O2 (from 50 to 150 mM). Ethanol (12% v/v) was added to some samples. The reactions were carried out at constant volume in a thermostated glass-reactor equipped with a magnetic stirring bar and connected to an automatic gas absorption or desorption measuring apparatus controlled by a computer. All reactions were performed with nondegassed solutions. Degassing was shown to have no influence on H2O2 dismutation catalyzed by manganese- and iron-porphyrin complexes (21). Water solution of the Cu(II) complex with Ami, adjusted to pH 7.4, was added to the reactor at air atmosphere. Next, varied amounts of 30% aqueous H2O2 were introduced in one portion, immediately after which the stirring was started and the progress of dioxygen evolution was measured. The molar amounts of dioxygen evolved were calculated from the volume, atmospheric pressure, and temperature data using the ideal gas equation. XAS Measurements. XAS measurements were performed at the EMBL EXAFS beamline (c/o DESY, Hamburg). During the experiments the DORIS III storage ring was operating in dedicated mode at 4.5 GeV, with ring currents ranging from 90 to 150 mA. A Si(111) double crystal monochromator with an energy resolution of 1.6 eV at 8980 eV was used. The first monochromator crystal was detuned to 50% of its peak intensity in order to reject higher harmonics. The monochromator angle was converted to an absolute energy scale by using a calibration technique (22). The following samples were measured:

1. Cu(II)-Ami, 1:1 (solid) 2. Cu(II)-Ami, 1:1 (100 mM solution, pH 7.4) 3. Cu(II)-Ami, 1:1 (10 mM solution, pH 7.4) 4. Cu(II)-Ami, 1:1 (5 mM solution, pH 7.4) 5. Cu(II)-Ami-H2O2, 1:1:4 (solid) 6. Cu(II)-Ami-H2O2, 1:1:4 (100 mM solution, pH 7.4) 7. Cu(II)-Ami-H2O2, 1:1:6 (100 mM solution, pH 7.4) All spectra were collected on the frozen samples at 20 K. Solid-state samples were prepared from water solutions by ethanol precipitation, using a procedure similar to that by Sreedhara and Cowan (23). For samples 1, 2, and 5-7 the data were collected at the copper edge (∼8988.0 eV at the edge jump inflection point) by monitoring the sample absorption, while for the diluted samples 3 and 4 the fluorescence was detected with an energy discriminating 13 element Ge solid-state detector (Canberra). For samples 1, 2, and 5-7, three to five spectra were collected for each, from 8930 to 9810 eV with variable step widths. In the XANES and EXAFS regions, steps of 0.3 and 0.6-1.5 eV were used, respectively. For samples 3 and 4, the final spectrum resulted from averaging 20 scans for each sample. The scans spanned the same energy range of the absorption spectra with identical steps in the different regions of the spectrum. After inspection of each scan for edge consistency, the data were normalized by the edge jump and averaged. The EXAFS was then extracted by subtracting the slowly varying atomic background fitted with three cubic splines. Data reduction based on standard procedures was performed with the local set of

Chem. Res. Toxicol., Vol. 14, No. 10, 2001 1355 programs (24). The preedge peak (1s-3d electronic transition at ∼8980 eV) areas were calculated by subtracting an arctangent function from the normalized edge spectra and integrating over the range 8975-8985 eV. The analysis of the EXAFS data was performed with the set of programs EXCURV98 (25) by utilizing the rapid curved single and multiple scattering theories. The whole spectrum was analyzed by varying the atom types and the coordination numbers and iteratively refining the distance (R) and the Debye-Waller factor (2σ2) for each atomic shell. The quality of the fit obtained was assessed using the EXAFS Rfactor and an absolute index of goodness of fit , defined as follows: N

Rexafs )

∑1/σ(|χ

exp i

(k) - χitheory(k)|) × 100

i

N

∑w (χ

v2 ) 1/(Nind - p)(Nind/N)

i

exp

i

(k) - χith(k))2

i

where Nind is the number of independent data points and p is the number of parameters.

Results Spectroscopic Studies of Cu(II)-Ami System at pH 7.4. NMR spectra were recorded for a broad range of Ami:Cu(II) molar ratios, with Ami concentration kept constant at 10 mM. At Ami-to-Cu(II) ratios between 10 000 and 100 the increasing general broadening of the signals can be seen. At still lower ratios of 10 and 2, the spectra also exhibited a decrease of intensity, by 10 and 50%, respectively (measured by comparing integrals of Ami peaks to that of tert-butyl alcohol reference). However, a residual, very broad spectrum could still be detected for the 1:1 molar ratio. Its intensity was 4% of the initial value for Cu(II)-free amikacin. This observation indicated the formation of a minor diamagnetic species. This issue was studied further using EPR. Room-temperature X-band solution spectra of equimolar samples of Cu(II)-Ami, adjusted to pH 7.4, were measured for the range of concentrations spanning 2 orders of magnitude (1-100 mM). Parallel samples were prepared in 30% ethane-1,2-diol and measured at 77 K. The highest concentration obtained for the latter was 60 mM, due to solubility problems. No concentration effects on the spectral parameters were seen (giso ) 2.10, Aiso ) 65 G). The same experiment was repeated in the presence of 10-fold excess of H2O2 over Cu(II). Also, no influence of H2O2 on the spectral parameters was found. However, the intensity of the r.t. spectrum exhibited a significant deviation from a linear concentration dependence, both in the absence and presence of H2O2, while no deviation was seen for frozen solution spectra (Figure 2). This deviation was interpreted as an indication of formation of a diamagnetic dimer. A weakly coupled antiferromagnetic dimer could be excluded, because no additional EPR spectral lines could be detected in the area of g ) 4 (and in fact nowhere in the accessible spectral range). The formation constant for the apparent dimerization reaction (2 CuAmi ) Cu2Ami2) could be calculated by fitting the experimental curve, as shown on Figure 2 (log Kdim ) 1.84 ( 0.04). The dimer formation was also studied by CD spectroscopy. Increasing of the complex concentration resulted in the intensification of a CT band at 362 nm, while the d-d band of the complex remained unchanged (Figure 3).

1356

Chem. Res. Toxicol., Vol. 14, No. 10, 2001

Figure 2. Concentration dependence of the intensities of solution EPR spectra of Cu(II)-Ami at pH 7.4 (O). The curve calculated for a dimerization constant of 69 M-1 (solid) and the straight line, predicted for no dimer formation (dashed) are shown. The insert presents frozen solution experiments (9), for which a linear fit was found.

Figure 3. CD spectra of Cu(II)-Ami, 1:1 molar ratio, at pH 7.4 and various concentrations: 1 mM (a), 60 mM (b), 100 mM (c).

To gain deeper knowledge on the dimer structure, XAS measurements were performed on a series of Cu(II)-Ami samples, ranging from dilute solutions, through concentrated solutions to alcohol-precipitated solids. Figure 4 shows the edge region of four representatives of the seven samples studied. The comparison indicated that the average copper environment in all the seven samples was very similar to each other, with constant edge energy values, as judged by the values of the 1s-4p transition at 8986 eV and of the white line transition at 8994 eV. The edge shape is characteristic of Cu(II) complexed to nitrogen or oxygen donors in a centrosymmetrical coordination environment (26). The analysis of the 1s-3d preedge absorption at ∼8980 eV gave areas between 1 × 10-2 and 2 × 10-2 eV. These values fall in the characteristic range observed for Cu(II) complexes in slightly distorted six- or five-coordination geometry (26, 27). Furthermore, the preedge data show that the coordination of the metal ion did not change when hydrogen peroxide was added to the Cu(II)-Ami complex, either in solution or in the solid state. No differences were

Jez˘ owska-Bojczuk et al.

Figure 4. Comparison of the edge data for Cu(II)-Ami solid (a); Cu(II)-Ami solution (b); Cu(II)-Ami-H2O2, 1:1:4, solid (c); Cu(II)-Ami-H2O2, 1:1:4, solution (d).

observed in the preedge and edge regions between the solid and solution samples, independently of the addition of hydrogen peroxide. The EXAFS spectrum of sample 1 [Cu(II)-Ami, 1:1 solid] is characteristic of all light O/N type scatterers. The contribution to the total EXAFS of the first shell ligand can be evaluated with 5 O/N ligands at an average distance of 1.96 Å. This latter distance is perfectly consistent with a Cu(II) compound as indicated by the oxidation number of 2.15 obtained by applying the Bond Valence Sum method (28). The full spectrum is significantly better fitted by adding the contribution of one O/N atom at 2.33 Å, but the relatively large Debye-Waller factor obtained from the fitting (0.013 Å2) makes this contribution questionable. The sample 1 EXAFS spectrum and its Fourier transform are reported in Figure 5, panels a and b, respectively. Figure 5b shows the presence of outer shells of atoms, yielding peaks at about 3.4 and 3.9 Å from Cu(II). The former FT peak can be simulated by the contribution of the second shell carbon atoms (see Table 1). The outer peak is better reproduced by a copper atom at 3.86 Å. The spectrum obtained for sample 2 [Cu(II)-Ami, 1:1, 100 mM solution, pH 7.4] is noisier than the spectrum from the corresponding solid sample (Figure 6). While the edge region is identical for samples 1 and 2, slight differences can be noted in the EXAFS region that can be assigned to a different radial distribution function arising from changes occurring essentially in the outer coordination shells. The analysis of the spectrum indicates the same first coordination shell around Cu(II), consisting of 5 N/O ligands at an average distance of 1.96 Å. The contribution from atoms at about 3.3 Å, present in the solid sample, is maintained also in solution as is the contribution from a possible heavy backscatterer (Cu) at about 3.8 Å. At variance with sample 2, a new shell of atoms seems to be present between 2.8 and 3.0 Å. This may reflect a conformational change between solid and solution, but the complexity of the molecule prevents to propose any model. The Cu(II)-Ami complex in diluted solutions (samples 3 and 4, 10 and 5 mM) was also monitored to confirm the concentration dependence of dimerization. As shown in Figure 7 and Table 1, the FT of diluted solution spectra are identical (only the noise is obviously higher for the

Mechanism of H2O2 Activation by Cu(II) Complex of Amikacin

Figure 5. Comparison of experimental (continuous line) and theoretical (dashed line) EXAFS data (top) and of their Fourier transform (bottom) for solid Cu(II)-Ami (sample 1). The fitting parameters are reported in Table 1. Table 1. Results of Data Analysis of the Full EXAFS Spectrum for Six of the Seven Samplesa sample

shell

distance (Å)

DWF 2σ2 (Å2)

Rexafs (%)



1

5N/O 1N/O 2C 1Cu 5N/O 1N/O 4C 1Cu 5N/O 2N/O 2C 5N/O 1N/O 2C 1Cu 5N/O 1N/O 2C 1Cu

1.96(1) 2.33(2) 3.30(2) 3.85(2) 1.96(1) 2.79(2) 3.31(2) 3.84(2) 1.98(1) 2.81(2) 3.27(2) 1.96(1) 2.32(2) 3.31(2) 3.81(2) 1.97(1) 2.36(2) 3.30(2) 3.83(2)

0.008 0.013 0.006 0.020 0.011 0.004 0.016 0.006 0.009 0.010 0.009 0.010 0.019 0.008 0.022 0.010 0.022 0.007 0.025

17.5

0.82

30.3

4.1

25.1

1.4

14.0

0.4

33.8

4.2

2

3 5

6

a The results obtained for sample 4 are not reported since they are identical to those of sample 5. DWF is Debye-Waller Factor, Rexafs and  are indicators of the fit quality as reported in the Experimental Procedures.

more diluted sample) and do not show the outer shell peak at 3.8 Å, while maintaining the same distances for other shells. The EXAFS spectra of samples containing H2O2 (samples 5-7) exhibited no differences with sample 1, in agreement with the absence of effects in EPR.

Chem. Res. Toxicol., Vol. 14, No. 10, 2001 1357

Figure 6. Comparison of experimental (continuous line) and theoretical (dashed) EXAFS data (top) and of their Fourier transform (bottom) for the Cu(II)-Ami 100 mM solution (sample 2). The fitting parameters are reported in Table 1.

Cyclic Voltammetry. The cyclic voltammetry measurements were performed on platinum or glassy carbon electrodes. The cathodic and anodic ranges of potentials were scanned separately or jointly. The experiments were performed for the supporting electrolyte, Cu(II), Ami, and Cu(II)-Ami, for four Cu(II) and Ami concentrations [keeping Cu(II):Ami molar ratio at 1:1]: 1, 5, 30, and 60 mM. Ami alone was electrochemically inactive in the whole range of potentials applied and Cu(II) alone presented a typical reduction pattern. Expectedly, the oxidation of free Cu(II) to Cu(III) was not detected. Figures 8 and 9 present cyclic voltammograms obtained for the complex. The CV scans of 1 mM Cu(II)-Ami at pH 7.4 (Figure 8) showed an oxidation peak in the anodic range, attributable to a Cu(II) to Cu(III) transition, at +1.2 V. The Cu(II) f Cu(III) oxidation was hardly affected by altering the pH of the solution to 7.0, corresponding to the maximum of formation of CuHAmi, or to 8.1, where CuAmi was at maximum. The voltammograms a-c in Figure 8, obtained in the reduction/oxidation mode, exhibited a further broad oxidation peak, at ca. +0.9 V. It is likely due to reoxidation of products of irreversible complex reduction, because it was absent in oxidationonly voltammograms, like the one labeled d on Figure 8. In contrast, the reduction peak of Cu(II) to Cu(0), at +0.11 V at pH 7.4, was shifted by 50 mV toward the higher potentials at pH 7.0 and by 70 mV toward the lower potentials at pH 8.1. The potentials of all these reduction peaks had more negative values, compared to free Cu(II) aqua ion, but the reoxidation peak was less

1358

Chem. Res. Toxicol., Vol. 14, No. 10, 2001

Jez˘ owska-Bojczuk et al.

Figure 9. Concentration dependence of the oxidation of the Cu(II)-Ami complex at 1:1 molar ratio (0.1 M K2SO4, 25 °C, 25 mV/s on Pt electrode) at complex concentrations of 1 mM (a), 5 mM (b), 30 mM (c), and 60 mM (d), illustrating the increased susceptibility of the dimeric complex to oxidation.

Figure 7. Comparison of experimental (continuous line) and theoretical (dashed) EXAFS data (top) and of their Fourier transform (bottom) for the Cu(II)-Ami 10 mM solution (sample 3). The fitting parameters are reported in Table 1. Figure 10. Concentration dependences of dioxygen formation from H2O2, catalyzed by Cu(II)-Ami at 25 °C, pH 7.4. Inserts provide expansions of the time scale for the initial periods of reactions. (A) Dependence on H2O2 concentration: 50 mM (a), 75 mM (b), 100 mM (c), 150 mM (d); Cu(II)-Ami constant at 5 mM. (B) Dependence on Cu(II)-Ami concentration: 50 µM (a), 1 mM (b), 5 mM (c); H2O2 constant at 50 mM.

Figure 8. Cathodic/anodic CV scans of 1 mM Cu(II)-Ami complex, 1:1 molar ratio (0.1 M K2SO4, 25 °C, 25 mV/s on Pt electrode) at pH 7.1 (a), 7.4 (b), and 8.1 (c). The oxidation-only scan at pH 7.4 is also shown (d).

influenced, except for pH 8.1 where this peak disappeared. The increase of complex concentration (Figure 9) resulted in a gradual increase of a novel redox species, which is characterized by a very easy oxidation to Cu(III) (at +0.5 to +0.66 V) and the lack of a reduction process. The presence of this species coincided with the presence of the dimeric species, except for the 5 mM complex, for which no dimer formation was detected by spectroscopy. The positions of oxidation peaks for both the monomeric and the presumed dimeric species were

shifted in a similar fashion toward higher potentials upon the increase of complex concentration. The slenderness of the oxidation peak of the dimer suggests a two-electron electrode reaction. Dioxygen Generation. These experiments were prompted by an observation of oxygen bubbles forming upon the addition of H2O2 to concentrated Cu(II)-Ami solutions during the production of solid-state samples for XAS studies. The measurements were conducted using a gasometric apparatus, and the volumes of oxygen evolved were measured. Figure 10 presents the course of oxygen evolution reactions for various concentrations of components, grouped according to the variation in concentrations of Cu(II)-Ami (part A) and H2O2 (part B). The crucial initial period of dioxygen evolution is presented in the inserts. For longer times, the reactions slowed, which was accompanied by a decrease of pH, up to one unit for the highest H2O2 concentrations [the solutions were unbuffered, due to a limited solubility of Cu(II)-Ami in phosphate buffers]. Table 2 presents the O2 yields and turnover numbers (TON) for these reactions, proving the catalytic role of Cu(II)-Ami. Figure 11 presents the dependence of initial rates (V0) of O2

Mechanism of H2O2 Activation by Cu(II) Complex of Amikacin

Chem. Res. Toxicol., Vol. 14, No. 10, 2001 1359

Table 2. Influence of Concentrations of H2O2 and Cu(II)-Ami Complex on H2O2 Dismutation [H2O2] (mM)

[Cu(II)-Ami] (mM)

O2 yielda (%)

V0b (mmol/min)

TONc

50 75 100 150 50 50

5 5 5 5 1 0.05

100 96.6 82.2 72.4 72.4 32.8

0.21 0.27 0.34 0.42 0.061 0.000 58

9.75 14.5 16.0 21.7 36.2 320

a Yield of O is defined as the molar ratio (in %) of O formed 2 2 with respect to the theoretical value calculated from the initial amount of H2O2, knowing that 2 molecules of H2O2 are required in the dismutation reaction to generate one molecule of O2. b V0, initial rate of O2 evolution reaction. c TON, O2 yield (mol)/Cu(II)Ami catalyst (mol).

Figure 12. Decay of the band of NDMA at 25 °C, pH 7.4, in the presence of Cu(II)-Ami complexes and H2O2. The spectra were recorded every 10 min during 2 h. Concentrations: Cu(II) and Ami, 50 µM; H2O2 0.5 mM; NDMA, 20 µM; phosphate buffer, 5 mM.

Figure 11. Effect of H2O2 (A) and Cu(II)-Ami (B) concentrations on the initial rate of dioxygen formation at 25 °C, pH 7.4. Inserts present logarithmic plots, which indicate reaction orders in respect to individual components.

evolution reaction on H2O2 and Cu(II)-Ami concentrations. The reaction orders relative to Cu(II)-Ami and H2O2 concentrations were found as 1.2 ( 0.2 and 0.63 ( 0.04, respectively (inserts). Only the results for Cu(II)-Ami concentrations of up to 5 mM are presented. The O2 evolution reactions for higher concentrations of H2O2 and the catalyst were too rapid to be measured on our equipment. Therefore, the dependence presented in Figure 11A does not exhibit a typical saturation kinetics with excess of H2O2. However, the Lineweaver-Burk plot of the kinetic data (rate-1 vs [H2O2]-1, not shown) is linear. This confirms that the reaction follows the Michaelis-Menten kinetics. The Michaelis constant KM ) 154 ( 2 mM and kcat ) (2.8 ( 0.1) × 10-3 min-1 were determined from the slope and intercept of this plot. The addition of 12% ethanol resulted in reaction slow-down, by ca. 50%. Hydroxyl Radical Formation. The dioxygen evolution experiments demonstrated that the disproportionation of H2O2 was the main reaction catalyzed by Cu(II)Ami. Still, our preliminary studies of dG oxidation by Cu(II)-Ami (14) suggested that these complexes generate some radical oxygen species (ROS). To identify radical intermediates, we used two techniques, UV-vis and EPR. For an unambiguous identification of oxygen radicals in this system, DMPO spin trapping technique monitored by EPR was used (29, 30). The four line (1:2:2:1) spectrum, typical for •OH radicals was detected after the addition of DMPO (50 mM) to the reaction mixtures, containing Ami and Cu(II) (each 50 µM), and H2O2 (0.5

mM) in sodium phosphate buffer (5 mM). No traces •O2radicals were detected. The maximum of DMPO-•OH concentration was detected just after the mixing of the reagents. Then, the gradual decrease of the EPR signal was observed for several hours. For Cu(II) alone level of DMPO-•OH adducts was low, with a tendency to a very slow gradual increase. On the contrary, the presence of Ami in the sample resulted in a relatively high level of DMPO-•OH immediately after mixing. It suggests the maximal production of •OH at the beginning of reaction in the presence of Cu(II)-Ami complexes. In a separate experiment, ethanol was added at 1.7 M (12%) to the reaction mixture, resulting in the nearly complete quenching of DMPO-•OH formation. The kinetics of •OH formation was studied by EPR and UV-vis spectroscopies. In the former technique, the formation of the stable hydroxyl radical adduct of TEMPO was followed at 25 °C for pH 7.4. The TEMPO quenching reaction was monitored at 5 min intervals, due to technical limitations of an EPR spectrometer. Only second-order kinetics was seen. In the UV-vis experiments, the kinetic curves were obtained at 25 and 37 °C for Cu(II)-Ami and H2O2 concentrations similar to DMPO studies. The whole spectra were gathered initially, to find outs, whether the reaction products might interfere with the assay (Figure 12). After excluding this possibility, the kinetics was followed by monitoring the decrease in NDMA absorption at 440 nm (Figure 13). The experiments were performed for pH 7.4, as well as 7.1 and 8.1, in correspondence with CV studies. The analysis of the NDMA destruction curves indicated that initially (5-15 min, dependent on the reaction rate) the reporter molecule consumption followed the pseudofirst-order kinetics. Later, the concentrations of H2O2 decreased, resulting in reaction slow-down and higher order kinetics, as shown on logarithmic plots for pH 7.4. The values of 1st order rate constants obtained at 25 °C and 37 °C for pH 7.1, 7.4, and 8.1 are presented in Table 3.

Discussion Dimer Structure. The concentration dependence of the EPR signal intensity suggests the formation of a strongly antiferromagnetically coupled dimer. The CT

1360

Chem. Res. Toxicol., Vol. 14, No. 10, 2001

Figure 13. Experimental curves of kinetic measurements of destruction of NDMA by hydroxyl radicals generated from H2O2 in the presence of Cu(II)-Ami in sodium phosphate buffer, at pH 7.4 and 25 °C (A) or 37 °C (B). Inserts present evaluations of rate constants. Concentrations of reagents as in Figure 12. Table 3. First-Order Rates Constants [k (min -1)] for Quenching of •OH Radicals by NDMA temp/pH 7.1 7.4 8.1

25 °C (min-1) 10-3

(0.78 ( 0.3) × (2.30 ( 0.2) × 10-3 (5.33 ( 0.3) × 10-3

37 °C (min-1) (5.24 ( 0.4) × 10-3 (13.9 ( 0.5) × 10-3 (26.5 ( 0.4) × 10-3

band at 362 nm, seen in CD spectra, became stronger along with the loss of the EPR signal. CT bands resulting from Cu(II) coordination by hydroxy or alkoxy oxygens, bridging as well as nonbridging, are known to appear just in this spectral region (31, 32). Thus, the presence of this band, at low intensity, was found in conditions where only a monomeric form was present. The formation of the bridge resulted in the increase of intensity of this band. The backscatterer shell at ca.. 3.8 Å was observed in EXAFS only in the concentrated frozen solution samples and in the solids precipitated from such solutions with alcohol. This finding, and the good fit of the EXAFS data for the assumption of a Cu(II)-Cu(II) pair, confirm that the dimerization of the complex occurs at its high concentration. The Cu(II)-Cu(II) distance is consistent with an alkoxy-bridged dimer, for the bridging oxygen at ca.. 2.3 Å from Cu(II) (see Table 1). It then must be the B5 oxygen, which also participates in Cu(II) coordination in the parent monomeric complex at pH 7.4 (Figure 1). The analysis of the molecular model, using the Cu(II)-Cu(II) distance of 3.86 Å and the Cu(II)-O distance of 2.33 Å, provided by EXAFS measurements, yielded the Cu(II)-O-Cu(II) angle of 112°, thereby confirming the consistency of the data. The spectra also point against other kinds of bridges, e.g., with -O-Ounits, because those yield characteristic CT bands in the 500 nm region, not observed here (31-36). Redox Properties of the Monomeric and Dimeric Cu(II)-Ami Complexes. Cyclic voltammetry measurements demonstrated that the redox properties of the dimeric complex are distinctly shifted toward a Cu(II)/ Cu(III) redox pair, compared to the parent monomer, which is both reduction- and oxidation-prone. In an apparent discrepancy with the spectroscopic data, the dimer features could be seen already for 5 mM samples, perhaps due to interactions with electrode surface, which also seem responsible for a slight variation of the oxidation voltage of the dimer with concentration.

Jez˘ owska-Bojczuk et al.

The potential of reduction of the monomeric Cu(II)-Ami depends on pH, but the reduction current is more or less constant. Using stability constants obtained previously (14) one can calculate the relative abundances of CuHAmi and CuAmi at particular pH values: 7.0 (63.5% CuHAmi, 17% CuAmi); 7.4 (17% CuHAmi, 37% CuAmi); 8.1 (18% CuHAmi, 61% CuAmi). Taking these values into account, it can be stated that Cu(II) in both CuHAmi and CuAmi can be reduced, but this process is less favorable for CuAmi. In contrast, the oxidation potential is not dependent on pH, but the current is. The likely interpretation is that only CuAmi, which contains the deprotonated oxygen donor, can undergo oxidation to Cu(III), and the increase of current is simply due to the increase of concentration of the electroactive species. The amino/ amide donor set of Ami complexes, is similar to that present in Cu(II) complexes of dipeptides. Their redox properties were studied (37). It was found that almost all dipeptide complexes could be electrochemically reduced, but not oxidized to Cu(III), analogous to the CuHAmi complex. Thus, the oxidation of CuAmi to a Cu(III) species appears to be enabled by the presence of the deprotonated B5 oxygen in the coordination sphere. Its role is further supported by the particular ease of oxidation of the dimer, in which it is bridging the copper atoms. Mechanism of Reaction with H2O2. The kinetic characteristics of dioxygen production clearly demonstrate that the monomeric complex is an actual catalyst: the reaction order is one with respect to the complex concentration in the range of 50 µM to 5 mM, where the dimer is not formed. The fast H2O2 disproportionation process dominates: at high concentrations of reagents the conversion degree reached 100% within several minutes. The spin trapping experiments and oxidation of reporter molecules unequivocally showed, however, that the Cu(II)-Ami-H2O2 system generates hydroxyl radicals in the same conditions, while no superoxide radicals were detected by both methods. The synthetic catalysts for hydrogen peroxide disproportionation, although inspired by the catalases, do not always obey the reference kinetic law for catalase, for which all the stepwise reactions take place at the same catalytic center. Different orders in H2O2 or in catalyst concentrations and simple or more complicated rate laws have been reported (38, 39). The values of first-order rate constants for dioxygen generation and for hydroxyl radical quenching at 25 °C are of the same order of magnitude (see Tables 2 and 3). The latter value is lower by a factor of 5, but it is a reporter reaction, successive to the process of •OH formation. A 100% specificity of •OH capture by NDMA is not possible. Therefore, its rate is necessarily lower from the one for •OH formation, and we can assume that the •OH generation is an essential intermediate of the dioxygen generation process, accompanied by other, slower steps, obeying the Michaelis-Menten kinetics. The observed deactivation of the catalyst can be caused by hydroxyl radicals, which are known to initiate degradation of ligand by hydrogen-atom abstraction (40). Oxidation of Ami ligand coordinated to Cu(II) can lead to a decomposition of the coordination site. Cu(II)-Ami contains ca. equimolar amounts of CuHAmi and CuAmi complexes (ratio of ca. 5:3 for the range of complex concentrations used in H2O2 disproportionation studies). The CV results indicated that the CuHAmi

Mechanism of H2O2 Activation by Cu(II) Complex of Amikacin

complex can be reduced to Cu(I), while the CuAmi complex can be oxidized to Cu(III). Therefore, H2O2 activation by Cu(II)-Ami may proceed with the joint participation of Cu(I)/Cu(II) and Cu(II)/Cu(III) redox pairs. The rate of NDMA decomposition increased markedly with pH, along with the increase of abundance of CuAmi (17, 33, and 60% at pH 7.1, 7.4, and 8.1, respectively), providing additional evidence for the generation of •OH radicals by the Cu(II)/Cu(III) redox pair. The addition of ethanol resulted in complete scavenging of •OH radicals (29), but dioxygen formation slowed only ca. 2-fold. The plausible explanation is the formation of peroxo and hydroxo species bound to the copper amikacin complex, which would not be reactive toward the two scavengers used in our studies, NBT and DMPO. The hydroxo species would be in equilibrium with free, scavengable •OH radicals, while •O2- radicals would only be present in a copper-bound form. The presence of copper-bound, both Cu(I) and Cu(II), rather than free, oxygen radical species was proved in many instances, e.g., refs 41-44, and thus seems the likely explanation of the phenomena observed in our experiments. The probable mechanism of the H2O2 disproportionation, resulting from the above arguments is presented in the following reactions: •

Cu(II)-Ami + H2O2 ) Cu(I)-Ami- OOH + H

+

Cu(II)-Ami + H2O2 ) Cu(III)-Ami-•OH + OHCu(I)-Ami-•OOH + Cu(III)-Ami-•OH ) 2 Cu(II)-Ami + O2 + H2O 2 H2O2 ) O2 + 2 H2O (sum reaction) An alternative mechanism, proposed by Sigel et al. (45) and later by Lekchiri et al. (46) does not involve any diffusible radical species:

H2O2 T HOO- + H+ Cu(II)L + HOO- T Cu(II)LOOH Cu(II)LOOH + H2O2 T Cu(II)LOOH(H2O2) Cu(II)LOOH(H2O2) T Cu(II)L + O2 + H2O + OHThis kind of mechanism in fact requires a copper dimer, and is consistent with either Cu(I)/Cu(II) or Cu(II)/Cu(III) redox pair. It is not relevant for the reactivity observed in our experiments, but provides an interesting option in further studies of oxidative properties of Cu(II) complexes of aminoglycosides, if higher stability dimers would be detected, especially as it is likely to provide a much more efficient catalyst than the monomeric complex (the rates found in ref 46 are more than 10-fold higher from ours).

Conclusions The data presented above clearly indicate that Cu(II)Ami complexes are effective activators of hydrogen peroxide at pH 7.4. The multistep catalytic process of H2O2 disproportionation by monomeric Cu(II)-Ami com-

Chem. Res. Toxicol., Vol. 14, No. 10, 2001 1361

plexes was found. This process is mediated by hydroxyl radicals and involves both Cu(I)/Cu(II) and Cu(II)/Cu(III) redox pairs. The strongly coupled Cu(II)-Cu(II) dimer, formed at Ami and Cu concentrations higher than 10 mM, retains the Cu(II) coordination sphere of the monomer. The EXAFS experiments yielded the Cu-Cu distance of 3.86 Å, which is consistent with the Cu-O-Cu bridging by the B5 oxygen atom. This dimer can be oxidized electrochemically to a Cu(III) species at a very low potential of ca. +0.6 V. These properties were not relevant for the conditions in which the Cu(II)-Ami reactivity was studied, but could be of interest for the design of novel antibiotics.

Acknowledgment. This work was supported by the Polish State Committee for Scientific Research (KBN) Grant 3 T09A 06818. The authors also acknowledge the “European CommunitysAccess to Research Infrastructure Action of the Improving Human Potential Program to the EMBL Hamburg Outstation, contract number: HPRI-CT-1999-00017” for providing access to the XAS instrumentation in Hamburg.

References (1) Davis, B. D. (1987) Mechanism of bactericidal action of aminoglycosides. Microbiol. Rev. 51, 341-350. (2) Busse, H. J., Wostmann, C., and Bakker, E. P. (1992) The bactericidal action of streptomycin: membrane permeabilization caused by the insertion of mistranslated proteins into the cytoplasmic membrane of Escherichia coli and subsequent caging of the antibiotic inside the cells due to degradation of these proteins. J. Gen. Bacteriol. 138, 551-561. (3) Humes, H. D. (1999) Insights into ototoxicity. Analogies to nephrotoxicity. Ann. N. Y. Acad. Sci. 884, 15-18. (4) Schacht, J. (1993) Biochemical basis of aminoglycoside ototoxicity. Otolaryngol. Clin. North Amer. 26, 845-856. (5) Beaubien, A. R., Karpinski, K., and Ormbsy, E. (1995) Toxicodynamics and toxicokinetics of amikacin in the guinea pig cochlea. Hear. Res. 83, 62-79. (6) Follath, F., Wenk, M., and Vozeh, S. (1981) Plasma concentration monitoring of aminoglycosides. J. Antimicrob. Chemother. 8(A), 37-43. (7) Moore, R. D., Lerner, S. A., and Levine, D. P. (1992) Nephrotoxicity and ototoxicity of aztreonam versus aminoglycoside therapy in seriously ill nonneutropenic patients. J. Infect. Dis. 165, 683688. (8) Jez˘ owska-Bojczuk, M., Kozłowski, H., Trnka, T., and Cerny, M. (1994) Interaction of 1,6-anhydro-derivatives of aminosugars with copper(II) ions. Part II. Carbohydr. Res. 253, 19-28. (9) Jez˘ owska-Bojczuk, M., Kozłowski, H., Pettit, L. D., Micera, G., and Decock, P. (1995) Coordination ability of di-galactosamine, and di- and tri-galacturonic acids. Potentiometric and spectroscopic studies of Cu(II) complexes. J. Inorg. Biochem. 57, 1-10. (10) Jez˘ owska-Bojczuk, M., Lamotte, S., and Trnka, T. (1996) Potentiometric and spectroscopic study of copper(II) binding by 1,6anhydro-derivatives of aminosugars. J. Inorg. Biochem. 61, 213219. (11) Jez˘ owska-Bojczuk, M., Chruscin˜ska, E., Trnka, T., and Micera, G. (1996) Coordination ability of selected amino sugars with Cu(II) ions in solution. J. Inorg. Biochem. 63, 231-239. (12) Micera, G., and Kozłowski, H. (1995) Metal ion-aminosugar complexes. In Handbook of Metal-Ligand Interactions in Biological Fluids (Berthon, G. Ed.) Vol. 1, pp 707-716, Marcel Dekker: New York. (13) Jez˘ owska-Bojczuk, M., Bal, W., and Kasprzak, K. S. (1996) Cu(II) interactions with an experimental anti-HIV agent, 1-deoxynojirimycin and oxygen activation by resulting complexes. J. Inorg. Biochem. 64, 231-246. (14) Jez˘ owska-Bojczuk, M., and Bal, W. (1998) Coordination of copper(II) by amikacin. Complexation equilibria in solution and oxygen activation by resulting complexes. J. Chem. Soc., Dalton Trans. 153-159. (15) Jez˘ owska-Bojczuk, M., Bal, W., and Kozłowski, H. (1998) Kanamycin revisited: a combined potentiometric and spectroscopic

1362

(16) (17) (18)

(19) (20) (21)

(22) (23) (24) (25) (26)

(27) (28)

(29) (30) (31) (32)

Jez˘ owska-Bojczuk et al.

Chem. Res. Toxicol., Vol. 14, No. 10, 2001

study of copper(II) binding to kanamycin B. Inorg. Chim. Acta 275/276, 541-545. Jez˘ owska-Bojczuk, M., Karaczyn, A., and Bal, W. (1998) Binding of Cu(II) ions to geneticin, a gentamycin analog. J. Inorg. Biochem. 71, 129-134. Jez˘ owska-Bojczuk, M., Karaczyn, A., and Kozłowski, H. (1998) Copper(II) binding to tobramycin. Potentiometric and spectroscopic studies. Carbohydr. Res. 313, 265-269. Jez˘ owska-Bojczuk, M., Les´niak, W., Szczepanik, W., Gatner, K., Jezierski, A., Smoluch, M., and Bal, W. (2001) Copper(II) lincomycin: complexation pattern and oxidative activity. J. Inorg. Biochem. 84, 189-200. Jez˘ owska-Bojczuk, M., and Les´niak, W. (2001) Coordination mode and reactivity of copper(II) complexes with kasugamycin. J. Inorg. Biochem. 85, 99-105. Priuska, E. M., and Schacht, J. (1995) Formation of free radicals by gentamicin and iron and evidence for an iron/gentamicin complex. Biochem. Pharmacol. 50, 1749-1752. Belal, R., Momenteau, M., and Meunier, B. (1989) Influence of proximal ligand in the dismutation of hydrogen peroxide catalyzed by manganese and iron-porphyrin. J. Chem. Soc., Chem. Commun. 412-414. Pettifer, R. F., and Hermes, C. (1985) Absolute energy calibration of X-ray radiation from synchrotron sources. J. Appl. Crystallogr. 18, 404-412. Sreedhara, A., and Cowan, J. A. (1998) Efficient catalytic cleavage of DNA mediated by metalloaminoglycosides. Chem. Commun. 1737-1738. Nolting, H. F., and Hermes, C. (1993) EXPROG: EMBL EXAFS data analysis and evaluation program package for PC/AT, European Molecular Biology Laboratory, c/o DESY, Hamburg. Binsted, N., Strange, R. W., and Hasnain, S. S. (1992) Constrained and restrained refinement in EXAFS data analysis with curved wave theory. Biochemistry 31, 12117-12125. Kau, L. S., Spira-Solomon, D. J., Penner-Hahn, J. E., Hodgson, K. O., and Solomon, E. I. (1987) X-ray absorption edge determination of the oxidation state and coordination number of copper: application to the type 3 site in Rhus vernicifera laccase and its reaction with oxygen. J. Am. Chem. Soc. 109, 6433-6442. Wirt, M. D., Bubacco, L., Peisach, J. (1995) Investigation of solid and solution structure of N-substituted Cu(II) salicyldimines by X-ray absorption spectroscopy. Inorg. Chem. 34, 2377-2381. Liu, W., and Thorp, H. H. (1993) Bond valence sum analysis of metal-ligand bond length in metalloenzymes and model complexes. 2. Refined distances and other enzymes. Inorg. Chem. 32, 4102-4105. Halliwell, B., and Gutteridge, J. M. C. (1999) Free Radicals in Biology and Medicine, 3rd ed., Oxford University Press, Oxford. Jazen, E. G., and Blackburn, B. J. (1968) Detection and identification of short-lived free radicals by electron spin resonance trapping technique. J. Am. Chem. Soc. 90, 5909-5910. Solomon, E. I., Penfield, K. P., and Wilcox, D. E. (1983) Active sites in copper proteins. An electronic structure overview. Struct. Bonding 53, 1-57. Spira-Solomon, D. J., and Solomon, E. I. (1987) Chemical and spectroscopic studies of the coupled binuclear copper Site in type 2 depleted Rhus Laccase: Comparison to hemocyanins and tyrosinase. J. Am. Chem. Soc. 109, 6421-6432.

(33) Solomon, E. I., Tuczek, F., Root, D. E., and Brown, C. A. (1994) Spectroscopy of binuclear dioxygen complexes. Chem. Rev. 94, 827-856. (34) Kitajima, N., Fujisawa, K., Fujimoto, C., Moro-oka, Y., Hashimoto S, Kitagawa T, Toriumi K, Tatsumi K, and Nakamura A (1992) A new model for dioxygen binding in hemocyanin. Synthesis, characterization and molecular structure of the µ-η2:η2 peroxo dinuclear copper(II) complex, [Cu(HB(3,5-R2pz)3)]2(O2) (R ) i-Pr and Ph). J. Am. Chem. Soc. 114, 1277-1291. (35) Karlin, K. D., Haka, M. S., Cruse, R. W., Meyer, G. J., Farooq, A., Guitneh, Y., Hayes, J. C., and Zubieta, J. (1988) Dioxygencopper reactivity. Model for hemocyanin: reversible O2 and CO binding to structurally characterized dicopper(I) complexes containing hydrocarbon-linked bis[2-(2-pyridyl)ethyl]amine units. J. Am. Chem. Soc. 110, 1196-1207. (36) Karlin, K. D., Wei, N., Jung, B., Kaderli, S., Niklaus, P., and Zuberbu¨hler, A. D. (1993) Kinetics and thermodynamics of copperdioxygen adducts: oxygenation of mononuclear copper(I) complexes containing tripodal tetradentate ligands. J. Am. Chem. Soc. 115, 9506-9514. (37) Bonomo, R. P., Marchelli, R., and Tabbı´, G. (1995) Study of H2O2 interaction with copper(II) complexes with diamino-diamide type ligands, diastereoisomeric dipeptides, and tripeptides. J. Inorg. Biochem. 60, 205-218. (38) Sigel H. (1969) Catalase and peroxidase activity of Cu2+ complexes. Angew. Chem., Int. Ed. 8, 167-177. (39) Choua, S., Pacheco, P., Coquelet, C., and Bienvenu¨e, E. (1997) Catalase-like activity of a water-soluble complex of Ru(II). J. Inorg. Biochem. 65, 79-85, and references therein. (40) Francis, K. C., Cummins, D., and Oakes, J. (1985) Kinetic and structural investigations of [FeIII(edta)]-[edta ) ethylenediaminetetraacetate(4-)] catalysed decomposition of hydrogen peroxide. J. Chem. Soc., Dalton Trans. 493-501. (41) Oikawa, S., and Kawanishi, S. (1996) Site-specific DNA damage induced by NADH in the presence of copper(II): role of active oxygen species, Biochemistry 35, 4584-4590. (42) Yamamoto K., and Kawanishi S. (1989) Hydroxyl free radical is not the main active species in site-specific DNA damage induced by copper (II) ion and hydrogen peroxide. J. Biol. Chem. 264, 15435-15440. (43) Sobkowiak, A., Qui, A., Liu, X., Llobet, A., and Sawyer, D. T. (1993) Copper(I)/(t-BuOOH)-induced activation of dioxygen for the ketonization of methylenic carbons. J. Am. Chem. Soc. 115, 609614. (44) Masarwa, M., Cohen, H., Meyerstein, D., Hickman, D. L., Bakac, A., and Espenson, J. H. (1988) Reactions of low-valent transitionmetal with hydrogen peroxide. Are they “Fenton-like” or not? The case of Cu+aq and Cr2+aq. J. Am. Chem. Soc. 110, 4293-4297. (45) Sigel, H., Flierl, C., and Griesser, R. (1969) On the kinetics and mechanism of the decomposition of hydrogen peroxide, catalyzed by the Cu2+-2,2′-bipyridyl complex. J. Am. Chem. Soc. 91, 10611064. (46) Lekchiri, A., Brighli, M., Methenitis, C., Morcellet, J., and Morcellet, M. (1991) Complexation of a polyelectrolyte derived from glutamic acid with copper(II). Catalase-like activity of the complexes. J. Inorg. Biochem. 44, 229-238.

TX010046L