2944
Biomacromolecules 2010, 11, 2944–2948
Dissolution and Dissolved State of Cytochrome c in a Neat, Hydrophilic Ionic Liquid Malvika Bihari, Thomas P. Russell, and David A. Hoagland* Department of Polymer Science and Engineering, University of Massachusetts Amherst, Amherst, Massachusetts 01003 Received July 1, 2010; Revised Manuscript Received August 25, 2010
The dissolution and dissolved molecular state of cytochrome c were investigated in the room temperature ionic liquid ethylmethylimidazolium ethylsulfate, [EMIM][EtSO4], by viscometry, optical and vibrational spectroscopies, and peroxidase activity. In dilute mixtures, viscometry demonstrated true molecular dissolution of cytochrome c in the ionic liquid and uncovered a molecular size larger than that in aqueous buffer, suggesting altered solvation or slight denaturation. The protein’s heme unit absorbs light outside the spectral range masked by [EMIM], enabling conformational assessments by UV-visible and circular dichroism spectroscopies. Adding trends from fluorescence and Fourier transform infrared spectroscopy, unchanged secondary but perturbed tertiary structures were determined, consistent with the appreciable peroxidase activity measured. Different than in aqueous buffers, denaturation is not accompanied by aggregation. Results are relevant to the proposed application of ionic liquids as media for room temperature preservation of biomacromolecules.
Introduction New techniques are sought for the stabilization and storage of proteins and protein-based pharmaceuticals, as current approaches suffer from protein degradation (for lyophilized formulations) or high cost of temperature control (for frozen formulations).1,2 As outlined in several recent reports,3-8 ionic liquids (ILs) pose a new option, not just as media for protein preservation but also for biocatalysis. Several studies of proteins in ILs have been reported to date exploring behavior of proteins in these solvents.9-12 To evaluate prospects, more must be known about “rules” for IL-protein compatibility, protein conformation in ILs, and transfer of IL-dissolved proteins into aqueous environments. Here, the first two issues are addressed for a representative IL-protein pair that has properties conducive to in situ study of protein structure and activity. ILs are salts with melting points below 100 °C, and room temperature ionic liquids (RTILs) are a subset of ILs with melting points near or below room temperature. Attributes of ILs that attract the most interest are near-zero vapor pressure, high thermal stability, broad miscibility, and polarity and hydrophobicity that can easily be tuned. Because of negligible vapor pressure, ILs sometimes are viewed as “green” solvents, and because an immense number of ILs can be envisaged by switching of cation and anion, they are also viewed as “designer” solvents.13,14 Predicting the compatibility (i.e., molecular level mixing) of proteins and ILs is not yet possible. For neat ILs, i.e., those lacking measurable water or organic solvent, factors such as hydrogen-bonding, polarity (hydrophobicity, dielectric constant), liquid nanostructure, and solvent viscosity could all be important.6 Even with compatibility, Hofmeister effects could impact stability and activity, as already found for some IL-buffer mixtures.6 Strong coordination of an IL ion to protein can be an important factor favoring dissolution, but the same factor tends to disrupt the intramolecular interactions responsible for * To whom correspondence should be addressed. E-mail: hoagland@ mail.pse.umass.edu.
native structure. The combination of slight solubility and measurable enzyme activity for Candida Antarctica lipase B in the IL triethyl methyl ammonium methylsulfate demonstrates that appropriate interactions are possible.15 In one attempt, taskspecific ILs with low molar concentration of anions known to denature and inactivate enzymes, were synthesized that had the ability to dissolve high concentrations of biomolecules (sugars) while allowing lipase B to retain its function.16 Other novel studies involve synthesis of enzyme-compatible ILs,17,18 ILs with natural amino acid anions (AAILs)19 and dual functionality ILs incorporating pharmaceutically active cations composed of antibacterials, analgesics etc. in combination with aspirinate and salicylic acid anions.20,21Perhaps more promisingly, mixtures of the salt choline dihydrogen phosphate with water not only dissolve cytochrome c (cyt c), but kept at room temperature, sustain notable enzyme activity for over a year.22 The same IL denatures lysozyme.23 Consequently, an IL suited to one protein might be ill suited for another, and there is no guidance in predicting the outcome. Better molecule-scale information about IL-protein interactions, needed to redress this situation, has proven difficult to obtain. Common ILs display broad UV absorbances that mask protein spectral features between 280 and 350 nm, precluding assessment of structure through typical spectroscopic methods. As described here, the difficulty is avoided with cyt c, fortuitously discovered to dissolve in the hydrophilic IL ethylmethylimidazolium ethylsulfate, [EMIM][EtSO4], a nontoxic and hydrophilic RTIL available commercially in large volumes. Thermal transitions of this hygroscopic liquid are absent above -60 °C, and at room temperature, viscosity is moderate, ∼90 cP. Key to structural study is cyt c’s heme group, which outside the range of UV interference by IL, absorbs light in a manner sensitive to the nearby protein conformation. A wide variety of studies on cyt c in aqueous16 and nearly anhydrous ILs have been conducted involving liquid-liquid extraction in the presence of crown ether,24,25 covalent modification of protein by attaching poly(ethylene oxide) to induce solubility,26 and biocatalysis.27 Cyt c is a small globular protein (104 residues, M ) 12,384
10.1021/bm100735z 2010 American Chemical Society Published on Web 10/07/2010
Dissolution and Dissolved State of Cytochrome c
Figure 1. Schematic of dissolution behavior of Cyt c in IL.
Da) with an isoelectric point of 10.3. Buried in a crevice lined with hydrophobic amino acids, the heme group in horse heart cyt c is covalently attached on two of its in-plane corners through Cys14 and Cys17.28-30 Axial coordination of the low spin heme iron with His18 and Met80 strongly stabilizes the native tertiary structure. Figure 1 illustrates the structure of heme prosthetic group of native cyt c. The iron readily converts between its ferrous and ferric states, making cyt c an efficient biological electron transporter.31 In the current investigation, molecular solubility and size were characterized by dilute solution viscometry (since light absorbance precluded light scattering), and molecular structure was characterized by UV-vis, fluorescence, FTIR, and CD spectroscopies. In addition, differences in peroxidase activity between IL and buffer were determined.
Experimental Section Materials. Horse heart cyt c was purchased from Sigma-Aldrich and used without purification, and [EMIM][EtSO4] (purity >99%) was donated by Evonik-Goldsmidt, Inc. Depending on protein concentration c, heating under vacuum at 60 °C achieved apparent dissolution in IL, as reflected in visual homogeneity, after a period of 1-5 days. This preparation step also removed any residual water in IL to below the detection limit of optical refractometry, ∼0.3 wt %, a value also confirmed by Karl Fischer titration. (The refractive indices of IL and water are 1.479 and 1.333, respectively.) These IL-protein mixtures were stored under nitrogen between experiments. Aqueous protein solutions were prepared in 0.01 M phosphate buffer at pH ∼7.2. Viscometry. A semimicro Ubbelohde viscometer (Cannon Instruments) was chosen for measuring the intrinsic viscosity [η] of cyt c in both IL and buffer. The average Stokes radius Rs was extracted from the experimentally determined [η] via the standard relationship Rs3 ) 3M[η]/10πNA, where NA is Avogadro’s number and M is molecular weight. Solvent viscosities of IL and buffer were determined to be 93 and 0.96 cP, respectively, and for determing [η], c varied from 2 to 10 mg/mL in IL and 40 to 80 mg/mL in buffer; both ranges are dilute, as verified by linear plots of specific viscosity versus c. Viscometry was performed in a nitrogen-purged glovebag to avoid moisture uptake by the hygroscopic IL. UV-Vis Spectroscopy. A UV-vis spectrophotometer (PerkinElmer Lambda 25) recorded the absorption spectrum of cyt c solutions in standard 1-cm path-length cuvettes. The extinction coefficient measured at 409 nm, 1.11 × 105 M-1 cm-1, compares favorably to that reported in literature for the same buffer.32 Fluorescence Spectroscopy. For c equal to 0.05 mg/mL in buffer and 0.05 to 10 mg/mL in IL, fluorescence emission spectra were recorded (Perkin-Elmer LS50B) at a fixed excitation wavelength of 280 nm. FTIR Spectroscopy. Spectra were obtained using a FTIR spectrometer (Bruker Tensor 27) equipped with a LN-MCT detector and MIRacle, a single reflection horizontal ATR diamond crystal plate. A total of 100 background-subtracted scans at 2 cm-1 resolution were averaged at c equal to 2 mg/mL in buffer and 18 mg/mL in IL. Circular Dichroism (CD). At c equal to 0.2 mg/mL in both buffer and IL, CD spectra (Jasco 720) were measured across the Soret region, 410 to 450 nm, for samples in 1-cm path-length quartz cells.
Biomacromolecules, Vol. 11, No. 11, 2010
2945
Peroxidase Activity Assay. Assays were initiated by adding hydrogen peroxide (H2O2) to a solution mixture of cyt c and guaiacol. The latter was catalytically oxidized by cyt c to its tetramer (tetraguaiacol) or similar compound,33 which due to strong orange color, could be monitored through absorbance at 470 nm. Assays were performed in 1.5 mL quartz cuvettes by mixing 50 µL of 20 mM guaiacol, 50 µL of 50 mM H2O2, 450 µL of buffer, and 50 µL of 0.02-0.2 mg/mL protein solution. For buffer assay, the last solution was prepared in buffer, and for an IL assay, the last solution was prepared in IL. A control assay was performed as for a buffer assay except that 50 µL of the added buffer volume was replaced with 50 µL of IL. Thus, in the control, composition was identical to that of an IL assay but the protein was not introduced as a neat IL solution. The maximum of the first time derivative of the product formation curve was taken as the characteristic reaction rate.
Results and Discussion Protein Size. Dispersing cyt c in either IL or buffer endows a pronounced red color to the liquid due to the heme iron. In both cases, mixtures at c < 20 mg/mL are homogeneous and fully transparent, strong evidence for protein dissolution. A more rigorous test for dissolution is the demonstration that the dispersed objects have a molecular size commensurate with their molecular weight which is normally accomplished by dynamic light scattering. However, in the present case, due to protein coloration, attempts to perform dynamic light scattering under available instrument configurations failed, leading to pursuit of protein size determination through dilute solution capillary viscometry, which offers a slightly more difficult but more accurate size measurement. The extra difficulty arises from [EMIM][EtSO4] hygroscopicity, which necessitated measurements in nitrogen atmosphere, and high viscosity, which made membrane filtration, serial dilutions, and flow measurements time-consuming. One full day was required for each [η] determination. In IL, [η] was 17 ( 1 mL/g, and in buffer, [η] was 2.5 ( 0.6 mL/g, corresponding to Stokes radii of 3.2 ( 0.4 nm and 1.7 ( 0.1 nm, respectively. Because of their magnitudes, both sizes confirm the molecular level mixing of the protein and solvent, and the current measurement in buffer is consistent with literature measurements in similar buffers.34 Similar value of radius of gyration determined by small angle neutron scattering (SANS) for cyt c in 50 vol % aqueous [BMIM][Cl] has been reported,16 suggesting an expanded structure, perhaps reminiscent of the protein partially unfolded in guanidine hydrochloride, but apparently still largely retaining elements of secondary structure. When denaturants (urea or guanidine HCl) are added to aqueous buffer, [η] for cyt c increases,34,35 much as observed in IL, and the increase can be attributed to an expansion of threedimensional protein structure, i.e., a globule-to-coil transition driven by the loss of both the secondary and tertiary protein structure. By analogy, one might infer that cyt c in IL is also denatured, but other explanations for enlarged size cannot be discounted, for example, a thicker solvation layer. Unless strong association with added denaturant stabilizes molecular dispersion, protein denaturation in aqueous environments at finite c inevitably exposes hydrophobic residues that cause uncontrolled, disordered aggregation. In IL, the observed increase of cyt c size cannot be explained in terms of aggregation since the Stokes radius remains comparable to that in buffer and does not grow in time. All evidence points to molecularly disrupted, yet solution-stable cyt c, a motif possibly allowing protein refolding upon transfer from IL to buffer. ILs have been observed to suppress protein aggregation irrespective of their protein
2946
Biomacromolecules, Vol. 11, No. 11, 2010
stabilizing/destabilizing effect.10 Whether the molecules possess a defined molecular structure different than the one adopted in buffer or adopt statistical, random coils cannot be ascertained through viscometry. Spectroscopic Characterization. Due to its heme prosthetic group, cyt c displays characteristic absorption bands starting from the edge of the near UV and extending across the visible region. In comparison to other factors affecting UV-vis absorption, the difference in refractive index between buffer and IL has negligible impact.36,37 In its native conformation, cyt c has an intense Soret band maximum at ∼409 nm, a position reflecting low spin state iron, a form imposed by the heme’s axial ligands, His18 and Met80.38 Denatured cyt c, with one or both of the heme axial ligands replaced, and thus with iron in its high spin state, is characterized by a blue shift of the soret band to ∼394 nm.39-41 Q bands, weaker absorbance at around 530 nm, reflect heme’s π-π* transitions, and for reduced iron, there are two of these, at ∼520 nm and ∼550 nm, but for oxidized iron, there is just one, at ∼530 nm.31 A charge transfer band for native cyt c, arising from coordination of Met80 sulfur to heme iron, is observed at ∼695 nm, and this band disappears when sulfur-iron coordination is lost.28 Spectra across the relevant regions for (a) neat IL, (b) cyt c dissolved in buffer, and (c) cyt c dissolved in IL are shown in Figures 2a-c. Figure 2a establishes that UV spectroscopy of cyt c in IL is impossible due to IL interference at wavelengths below ∼360 nm. Figure 2b shows that cyt c in buffer exhibits all of the mentioned bands, as expected for native, oxidized protein, and Figure 2c reveals that cyt c in IL has a Soret band at ∼408 nm, a position similar to that in buffer. However, the Q-band region in IL possesses not just a broad peak at ∼530 nm but also a “shoulder” at ∼550 nm, the latter indicating molecules in their reduced state. Expanding the spectral region beyond 650 nm, the inset to Figure 2c reveals complete loss of the charge transfer band in IL, a feature indicative of perturbation or loss of heme’s axial sulfur-coordinated ligand. With the Soret band essentially as for native protein, any new IL ligand does not alter iron spin state. Based on similarity in ligand structures, we speculate that the imidazolium cation complexes with oxidized iron, displacing Met80, a scenario also proposed by investigators exploring the impact of exogenous nitrogeneous ligands such as imidazole on cyt c in buffer.42,43 Figures 3a and 3b display fluorescence emission spectra for cyt c solutions in buffer and IL, such fluorescence nominally expected from aromatic Trp59. Forster energy transfer between Trp59 and nearby heme, however, quenches all fluorescence of native cyt c,30,44 so the emission from the buffer solution at ∼300 nm represents the first Stokes band of water.45 Figure 3a shows no fluorescence from cyt c in IL at 10 mg/mL, but when c is lower, 0.05 mg/mL to 3 mg/mL, Figure 3b does reveal fluorescence, and in this range, c systematically alters emission intensity. These trends can be explained in terms of the quenching of IL fluorescence by protein (cyt c has a high absorbance at ∼409 nm reflected by the soret band in absorbance spectrum), as previously reported for cyt c in buffer upon addition of butyl methyl imidazolium chloride.46 Thus, no definitive conclusions about cyt c structure in IL can be drawn from fluorescence measurements. Far and near UV CD of cyt c in IL proved uninformative due to high IL absorbance, but there was no IL interference in the Soret region, 350 to 450 nm, which reveals integrity of native structure in the heme vicinity. Figure 4 shows such spectra for cyt c in buffer and IL. The former exhibits a strong positive band at ∼407 nm and a strong negative band at ∼416 nm. The
Bihari et al.
Figure 2. (a) UV-vis spectrum of neat IL, showing high absorbance up to ∼350 nm, (b) Soret and Q bands of cyt c in both 0.01 M, pH 7 phosphate buffer and IL, and (c) magnified Q-band with maximum at ∼530 nm, characteristic of native cyt c. The inset shows a charge transfer band at ∼695 nm for native cyt c, and this band absent in IL.
negative band results from interaction of Phe82 with heme on the Met80 side of the heme plane. Loss of Met80 ligand causes disturbance to the distance and/or orientation of Phe82 with respect to heme, the negative band consequently disappearing,29,47 as noted in IL. When buffer-dissolved cyt c is treated with denaturants such as guanidine HCl or urea, the same band likewise vanishes.48 Similar observations have been noted for cyt c in aqueous [BMIM][Cl] above 25 vol. %.16 The finding from CD confirms the disruption of coordination between Met80 and heme iron deduced through loss of the charge transfer band in the absorption spectrum. Figure 5 provides FTIR spectra of neat IL, cyt c in buffer, and cyt c in IL. In buffer, amide I and II peaks are at 1653 and 1548 cm-1, respectively, whereas in IL, these peaks are at 1651 and 1542 cm-1. Similarities in the two peak positions demon-
Dissolution and Dissolved State of Cytochrome c
Biomacromolecules, Vol. 11, No. 11, 2010
2947
Figure 5. FTIR spectra in amide I and II regions of neat IL, cyt c in 0.01 M, pH 7 phosphate buffer, and cyt c in IL.
Figure 3. (a) For excitation at 280 nm, fluorescence emission spectrum of cyt c in 0.01 M, pH 7 phosphate buffer and IL. (b) Emission from neat IL and IL with cyt c at c from 0.05 to 3 mg/mL. Figure 6. Cyt c peroxidase activity assay: comparison of activities in 0.01 M, pH 7 phosphate buffer and IL. In the control, IL was added to buffer prior to assay. Table 1. Comparison of cyt c Peroxidase Activities in Buffer, IL, and Control (IL Added after Protein Dissolution in Buffer) ∆A/∆t
Figure 4. Soret region CD spectrum of cyt c in 0.01 M, pH 7 phosphate buffer, and IL.
strate that the protein’s native secondary structure remains largely, but not completely, intact in IL. Unlike the preceding UV/vis and CD experiments, FTIR assesses the secondary structure across the protein, not just in the vicinity of heme. Peroxidase Activity. The preceding discussion indicates that dissolution of cyt c in IL causes changes to protein structure, particularly a perturbation or loss of Met80 as axial ligand to the heme and an overall expansion of molecular size. Cyt c acquires peroxidase activity with disturbance to heme ligand,33,49 In native cyt c, all six ferric coordinate bonds are occupied, while in a genuine heme peroxidase enzyme, such as horse radish peroxidase, there is a free coordination site. Additionally, in native cyt c the heme is deeply “buried” in the protein crevice, hence hindered access lowers peroxidase activity.33,50,51 The two impacts of IL on cyt c, molecular expansion and ligand disturbance, should both enhance peroxidase activity.
sample
cyt c conc ) 0.2 mg/mL
cyt c conc ) 0.02 mg/mL
cyt c-buffer cyt c-IL cyt c-buffer + IL (control)
5.9 × 10-4 18.4 2.4
4.6 × 10-4 15.5 2.7
Figure 6 compares peroxidase activities in buffer and IL, and Table 1 shows the maximum slopes ∆A/∆t characterizing the steady state phase. Cyt c in IL shows ∼3× higher peroxidase activity than in buffer, a result expected from the reported conformational characterizations. As a control, just prior to assay, sufficient IL was added to buffer-dissolved cyt c to create a mixture with the same composition and, here, peroxidase activity was lowest, an experiment and result demonstrating that increased activity in neat IL is generated by IL-induced conformational changes, not simply exposure to IL. From the low activity in the control, one might conclude that molecules in a mixture retain their native structure and hydration spheres, which in context of trends for neat IL, presages a complex, competitive interplay between solvation/coordination by water and IL. Protein behavior in IL-water mixtures will not be discussed here, but previous studies have also seen enhanced enzyme activity in such mixtures.4,52
Conclusions Because of its unique spectral features, cyt c provides an attractive model system for the study of proteins in ILs. Via a
2948
Biomacromolecules, Vol. 11, No. 11, 2010
combination of measurements performed in a neat hydrophilic IL that molecularly dissolves the protein, we conclude that, while secondary structure remains largely intact, tertiary structure changes significantly upon dissolution. Trends for cyt c in this IL resemble those for acid-denatured cyt c in buffer. In both cases, the iron-sulfur bond of the Met80 ligand of the heme group is disrupted. Dissolution may proceed by complexation of the freed heme coordination site with the IL’s imidazolium cation53 as illustrated in the schematic in Figure 1. Our results support belief that dissolution of enzyme in neat IL frequently requires a molecular interaction of sufficient strength to threaten the enzyme’s structural integrity. While dissolution in IL is accompanied by denaturation, reflected in an increase in overall molecular size, the denatured proteins in IL were never observed to aggregate or precipitate, a feature distinct from denaturation in buffer. Acknowledgment. Finanical support for this work was provided by the NSF-supported Materials Research Science and Engineering Center at UMass and the Department of Energy Office of Basic Energy Science. Evonik-Goldschmidt is acknowledged for donation of the IL.
References and Notes (1) Klibanov, A. M.; Schefiliti, J. A. Biotechnol. Lett. 2004, 26 (14), 1103– 1106. (2) Wang, W. Int. J. Pharm. 2000, 203 (1-2), 1–60. (3) Fujita, K.; Forsyth, M.; MacFarlane, D. R.; Reid, R. W.; Eiliott, G. D. Biotechnol. Bioeng. 2006, 94 (6), 1209–1213. (4) Sheldon, R. A.; Lau, R. M.; Sorgedrager, M. J.; van Rantwijk, F.; Seddon, K. R. Green Chem. 2002, 4 (2), 147–151. (5) Yang, Z.; Pan, W. B. Enzyme Microb. Technol. 2005, 37 (1), 19–28. (6) van Rantwijk, F.; Sheldon, R. A. Chem. ReV. 2007, 107 (6), 2757– 2785. (7) Zhao, H. J. Chem. Technol. Biotechnol. 2010, 85 (7), 891–907. (8) Byrne, N.; Angell, C. A. J. Mol. Biol. 2008, 378 (3), 707–714. (9) Byrne, N.; Wang, L. M.; Belieres, J. P.; Angell, C. A. Chem. Commun. 2007, (26), 2714–2716. (10) Constatinescu, D.; Herrmann, C.; Weingartner, H. Phys. Chem. Chem. Phys. 2010, 12 (8), 1756–1763. (11) Mann, J. P.; McCluskey, A.; Atkin, R. Green Chem. 2009, 11 (6), 785–792. (12) Byrne, N.; Belieres, J. P.; Angell, C. A. Aust. J. Chem. 2009, 62 (4), 328–333. (13) Rogers, R. D.; Seddon, K. R. Science 2003, 302 (5646), 792–793. (14) Ueki, T.; Watanabe, M. Macromolecules 2008, 41 (11), 3739–3749. (15) Lau, R. M.; Sorgedrager, M. J.; Carrea, G.; van Rantwijk, F.; Secundo, F.; Sheldon, R. A. Green Chem. 2004, 6 (9), 483–487. (16) Baker, G. A.; Heller, W. T. Chem. Eng. J. 2009, 147 (1), 6–12. (17) Zhao, H.; Baker, G. A.; Song, Z. Y.; Olubajo, O.; Crittle, T.; Peters, D. Green Chem. 2008, 10 (6), 696–705. (18) Zhao, H.; Jones, C. L.; Cowins, J. V. Green Chem. 2009, 11 (8), 1128– 1138. (19) Kagimoto, J.; Taguchi, S.; Fukumoto, K.; Ohno, H. J. Mol. Liq. 2010, 153 (2-3), 133–138. (20) Bica, K.; Rijksen, C.; Nieuwenhuyzen, M.; Rogers, R. D. Phys. Chem. Chem. Phys. 2010, 12 (8), 2011–2017. (21) Stoimenovski, J.; MacFarlane, D. R.; Bica, K.; Rogers, R. D. Pharm. Res. 2010, 27 (4), 521–526.
Bihari et al. (22) Fujita, K.; MacFarlane, D. R.; Forsyth, M.; Yoshizawa-Fujita, M.; Murata, K.; Nakamura, N.; Ohno, H. Biomacromolecules 2007, 8 (7), 2080–2086. (23) Vrikkis, R. M.; Fraser, K. J.; Fujita, K.; MacFarlane, D. R.; Elliott, G. D. J. Biomech. Eng.-Trans. ASME 2009, 131 (7), XXX. (24) Shimojo, K.; Kamiya, N.; Tani, F.; Naganawa, H.; Naruta, Y.; Goto, M. Anal. Chem. 2006, 78 (22), 7735–7742. (25) Shimojo, K.; Nakashima, K.; Kamiya, N.; Goto, M. Biomacromolecules 2006, 7 (1), 2–5. (26) Sakai, S.; Kodama, L.; Matsumura, H.; Nakamura, N.; Yohda, M.; Ohno, H. Polym. AdV. Technol. 2008, 19 (10), 1430–1435. (27) Laszlo, J. A.; Compton, D. L. J. Mol. Catal. B-Enzymatic 2002, 18 (1-3), 109–120. (28) Eaton, W. A.; Hochstrasser, R. M. J. Chem. Phys. 1967, 46 (7), 2533– 2539. (29) Santucci, R.; Fiorucci, L.; Sinibaldi, F.; Polizio, F.; Desideri, A.; Ascoli, F. Arch. Biochem. Biophys. 2000, 379 (2), 331–336. (30) Sinibaldi, F.; Mei, G.; Polticelli, F.; Piro, M. C.; Howes, B. D.; Smulevich, G.; Santucci, R.; Ascoli, F.; Fiorucci, L. Protein Sci. 2005, 14 (4), 1049–1058. (31) Ayato, Y.; Takatsu, A.; Kato, K.; Matsuda, N. IEICE Trans. Electron. 2008, E91C (12), 1899–1904. (32) Russell, B. S.; Melenkivitz, R.; Bren, K. L. Proc. Natl. Acad. Sci. U.S.A. 2000, 97 (15), 8312–8317. (33) Fujita, A.; Senzu, H.; Kunitake, T.; Hamachi, I. Chem. Lett. 1994, (7), 1219–1222. (34) Fisher, W. R.; Taniuchi, H.; Anfinsen, C. B. J. Biol. Chem. 1973, 248 (9), 3188–3195. (35) Varhac, R.; Antalik, M.; Bano, M. J. Biol. Inorg. Chem. 2004, 9 (1), 12–22. (36) Makarska-Bialokoz, M.; Pratviel, G.; Radzki, S. J. Mol. Struct. 2008, 875 (1-3), 468–477. (37) Renge, I. J. Phys. Chem. 1993, 97 (25), 6582–6589. (38) Tajima, H.; Ikeda, S.; Matsuda, M.; Hanasaki, N.; Oh, J. W.; Akiyama, H. Solid State Commun. 2003, 126 (10), 579–581. (39) Droghetti, E.; Smulevich, G. J. Biol. Inorg. Chem. 2005, 10 (6), 696– 703. (40) Moosavi-Movahedi, A. A.; Chamani, J.; Goto, Y.; Hakimelahi, G. H. J. Biochem. 2003, 133 (1), 93–102. (41) Babul, J.; Stellwag, E. Biopolymers 1971, 10 (11), 2359–2361. (42) Schejter, A.; Aviram, I. Biochemistry 1969, 8 (1), 149–153. (43) Liu, G. H.; Shao, W. P.; Zhu, S. M.; Tang, W. X. J. Inorg. Biochem. 1995, 60 (2), 123–131. (44) Tsong, T. Y. J. Biol. Chem. 1974, 249 (6), 1988–1990. (45) Loewenich, D.; Kleinermanns, K. Photochem. Photobiol. 2007, 83 (6), 1308–1312. (46) DiCarlo, C. M.; Compton, D. L.; Evans, K. O.; Laszlo, J. A. Bioelectrochemistry 2006, 68 (2), 134–143. (47) Santucci, R.; Ascoli, F. J. Inorg. Biochem. 1997, 68 (3), 211–214. (48) Gong, J.; Yao, P.; Duan, H. W.; Jiang, M.; Gu, S. H.; Chunyu, L. Biomacromolecules 2003, 4 (5), 1293–1300. (49) Vladimirov, Y. A.; Proskurnina, E. V.; Izmailov, D. Y.; Novikov, A. A.; Brusnichkin, A. V.; Osipov, A. N.; Kagan, V. E. BiochemistryMoscow 2006, 71 (9), 989–997. (50) Diederix, R. E. M.; Ubbink, M.; Canters, G. W. Eur. J. Biochem. 2001, 268 (15), 4207–4216. (51) Diederix, R. E. M.; Ubbink, M.; Canters, G. W. Biochemistry 2002, 41 (43), 13067–13077. (52) Kragl, U.; Eckstein, M.; Kaftzik, N. Curr. Opin. Biotechnol. 2002, 13 (6), 565–571. (53) Cheng, D. H.; Chen, X. W.; Shu, Y.; Wang, J. H. Talanta 2008, 75 (5), 1270–1278.
BM100735Z
