DOI: 10.1002/chem.201003485
A Heme–Peptide Metalloenzyme Mimetic with Natural Peroxidase-Like Activity Flavia Nastri,[a] Liliana Lista,[a] Paola Ringhieri,[a] Rosa Vitale,[a] Marina Faiella,[a] Concetta Andreozzi,[a] Paola Travascio,[a] Ornella Maglio,[a, b] Angela Lombardi,*[a] and Vincenzo Pavone*[a] Abstract: Mimicking enzymes with alternative molecules represents an important objective in synthetic biology, aimed to obtain new chemical entities for specific applications. This objective is hampered by the large size and complexity of enzymes. The manipulation of their structures often leads to a reduction of enzyme activity. Herein, we describe the spectroscopic and functional characterization of FeIII–mimochrome VI, a 3.5 kDa synthetic heme– protein model, which displays a peroxidase-like catalytic activity. By the use
of hydrogen peroxide, FeIII–mimochrome VI efficiently catalyzes the oxidation of several substrates, with a typical Michaelis–Menten mechanism and with several multiple turnovers. The catalytic efficiency of FeIII–mimochrome VI in the oxidation of 2,2’azino-di(3-ethyl-benzothiazoline-6-sulKeywords: bioinorganic chemistry · heme proteins · Michaelis–Menten kinetics · peroxidase activity · protein design
Introduction A considerable challenge in synthetic biology is the construction of artificial enzymes. Recently, excellent examples have been reported on the computational design of functional proteins.[1] The design of metalloenzymes has been recognized as being an even more intricate challenge,[2a,b] since both the requirements of protein structure and metal ion coordination should be fulfilled. Metal ions impart highly tunable redox and catalytic activities to biomolecules, thus rendering them capable of catalyzing important reactions, such as nitrogen fixation, methane hydroxylation, and CO oxidation or insertion, in environmentally benign conditions. However, metalloenzymes with asymmetrical coordination centers have rarely been mimicked,[2c] even through
[a] Prof. Dr. F. Nastri, Dr. L. Lista, Dr. P. Ringhieri, Dr. R. Vitale, Dr. M. Faiella, Dr. C. Andreozzi, Dr. P. Travascio, Dr. O. Maglio, Prof. Dr. A. Lombardi, Prof. Dr. V. Pavone Department of Chemistry, Complesso Universitario Monte S. Angelo University of Naples Federico II Via Cintia, 80126 Naples (Italy) Fax: (+ 39) 081-674090 E-mail:
[email protected] [email protected] [b] Dr. O. Maglio Permanent address: IBB, CNR, Via Mezzocannone 16 80134 Napoli (Italy) Supporting information for this article is available on the WWW under http://dx.doi.org/10.1002/chem.201003485.
4444
fonic acid (ABTS) and guaiacol (kcat/ Km = 4417 and 870 mm1 s1, respectively) is comparable to that of native horseradish peroxidase (HRP, kcat/Km = 5125 and 500 mm1 s1, respectively). FeIII–mimochrome VI also converts phenol to 4- and 2-nitrophenol in the presence of NO2 and H2O2 in high yields. These results demonstrate that small synthetic peptides can impart high enzyme activities to metal cofactors, and anticipate the possibility of constructing new biocatalysts tailored to specific functions.
the use of model organic compounds.[3] The successful synthesis and application of novel metalloenzyme models may have great potential for numerous areas of chemistry and biology, such as biocatalysis and biosensor technology, degradation of pollutants or biomass, and drug or food processing. Numerous synthetic biomimetics of metalloproteins have been prepared using designed organic ligands.[3] They have been fundamental in elucidating the structure and function of metalloproteins. However, significant questions remain, such as the mechanism by which the protein matrix finely tunes the properties of the metal center and effects catalytic processes. Peptide-based metalloenzyme models seem valuable candidates to accommodate a metal binding site with defined geometries that are required for function.[4] Their structures are smaller than native proteins, making them easier for practical applications. Simultaneously, they have sufficient size and chemical diversity to allow the construction of functional sites. We have applied this approach to the construction of heme–protein mimetics; previously we have developed a family of artificial heme proteins, named mimochromes, characterized by a bis-His coordination to the heme.[5] In this family of compounds, FeIII–deuterohemin is covalently linked to two peptide chains through two amide bonds between the heme-propionic groups and the e-amino groups of two Lys residues. X-ray analysis and NMR spectroscopy proved that, despite their small size, mimochromes adopt a well-defined secondary and tertiary structure in a sandwich
2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 4444 – 4453
FULL PAPER arrangement, with two helical peptides surrounding the heme on both faces.[5] The goal of the current work was to introduce peroxidase-like activity into mimochrome architecture and to mimic peroxidases. Among heme proteins, peroxidases catalyze a variety of oxidation reactions by using H2O2.[6] These enzymes are very appealing for laboratory or industrial oxidation reactions because they have high activity towards a wide range of substrates. However, their high production costs and instability during catalysis limit their potential applications. One solution to this problem would be to use simpler synthetic peroxidases with improved stability and efficiency. We report here the peroxidase-like activity of FeIII–mimochrome VI. This molecule is an evolution, through design and redesign, of previously described mimochromes. It embodies some of the key elements for functioning as a peroxidase-like catalyst.
Results Design and synthesis: FeIII–mimochrome VI was designed to create a proximal and a distal site environment around the heme. Figure 1 depicts the model structure, which is composed of a 14-residue peptide with a His residue at position 6, as axial ligand to the heme (proximal face) and a 10-residue peptide that is devoid of a heme-coordinating residue, therefore may create a cavity around the metal ion (distal face).
Figure 1. Schematic representation of the model structure of FeIII–mimochrome VI . The amino acid sequence is also reported.
The NMR-derived solution structure of CoIII–mimochrome IV[5d] and the model structure of CoIII–mimochrome II[5f] were used as starting point for the design. They are both examples of bis-His cobalt porphyrin–peptide conjugate; the first is a bis-tetradecapeptide–porphyrin conjugate and the second is a bis-nonapeptide–porphyrin conjugate. Both are pseudo-C2 symmetrical systems. We combined the sequences of mimochrome II and IV to construct an unsymmetrical five-coordinate mono-histidine model. The primary structure contains a histidine (His6) in the tetradecapeptide chain (derived from the sequence of mimochrome II), as a potential axial ligand to the metal (proximal face), and a serine (Ser6) in the decapeptide chain (derived from the sequence of mimochrome IV) to create a cavity near the metal site (distal face). Some substitutions from the sequences of mimochrome II and IV were introduced to further
Chem. Eur. J. 2011, 17, 4444 – 4453
stabilize the secondary and tertiary structure. Similarly to other mimochromes, both peptide chains are covalently linked to the porphyrin propionate through the Lys9 side chains. In terms of secondary structure, the tetradecapeptide was designed to adopt a short helical conformation (residues 1–9), a loop gaD (residues 10 and 11) and a short b-strand (residues 12–14) that folds back (through the loop) to interact with the helical part, the decapeptide also adopts a helical conformation (residues 1–8). The tertiary structure can be described as a sandwich, which is characteristic of the mimochrome class of molecules. The two peptide chains embrace the metalloporphyrin; the helical segments are antiparallel to each other and the helix axes are about parallel to the porphyrin plane. Stabilization of the tertiary structure was contributed by interchain ion pairs between the carboxylate side chains of glutamate residues (Glu2) on one helix and the guanidine groups of arginine (Arg10) on the other helix. The positively charged Arg10 and the negatively charged Glu2 at the C-terminal and N-terminal ends, respectively, with the opposite sign relative to the helix dipole, may also provide stabilization of the secondary structure. Finally, several glutamines (Gln3, 4, 8) and a serine (Ser7) were introduced in the solvent exposed positions to promote water solubility. FeIII–mimochrome VI was prepared similarly to other mimochromes (see the Supporting Information for details).[5a] The two peptides were synthesized by the solid-phase method using the 9-fluorenylmethoxycarbonyl (Fmoc) protection strategy, and they were coupled to deuteroporphyrin in solution to afford mimochrome VI. Mimochrome VI was then purified to homogeneity by RP-HPLC; MALDI-TOF mass spectrometry confirmed the expected molecular weight. Iron ion was inserted into mimochrome VI by following the acetate method,[7] which was slightly modified by our group.[5b] Spectroscopic analysis: CD spectroscopy was used to investigate the structural properties of the molecule. The CD spectrum of FeIII–mimochrome VI (phosphate buffer, 100 mm, pH 6.5) indicated a low helical content for the peptide chains (Figure 2 a).
Figure 2. Secondary structure of FeIII–mimochrome VI. a) CD spectra of FeIII–mimochrome VI (7.3 106 m) in phosphate buffer (10 mm, pH 6.5) at various TFE concentrations (v/v). b) Molar ellipticity at 222 nm as a function of TFE % (v/v).
2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4445
V. Pavone, A. Lombardi et al.
Addition of 2,2,2-trifluoroethanol (TFE) induced significant changes in the CD spectrum. The wavelength shift to 207 nm of the minimum at 202 nm, the increase of both the ellipticity at 222 nm, and the [q]ratio ([q]222/[q]min), indicated the presence of at least 50 % helical content at a high TFE concentration (Figure 2 b).[5d] To determine the coordination properties of the complex, a UV/Vis pH titration of FeIII–mimochrome VI was performed in the pH range 2.0–10.0 in aqueous solutions (Figure 3 a). Two well-resolved transitions, with midpoints at pH 3.4 and 7.4, were observed (Figure 3 b). At pH 2.0, the spectrum observed for FeIII–mimochrome VI is typical of a predominantly high spin ferric porphyrin (S = 5/2), with the Soret band at 389 nm, the Q bands at 495, 529 nm, and the CT (charge-transfer) band at 611 nm (Table 1). An increase in the pH from 2.0 to 5.5 causes the spectrum to remain essentially unchanged; the observed decrease in intensity of the Soret and CT bands, together with a slight red shift, are consistent with changes in the ferric ion axial coordination from H2O/H2O to His/H2O, in the high spin state.[8 ] Furthermore, the band near 350 nm (N band region) broadens and the relative intensity of this band to the Soret band changes from 0.17 to 0.34. Although the presence of the broad 350 nm band has been sometimes attributed to intermolecular heme–heme association,[9] experimental evidence suggests that this is not the case for FeIII–mimochrome VI. In fact, a linear correlation in aqueous neutral solutions, between absorbance and concentration in the range 106–105 m, indicates a negligible association phenomena. Sedimentation studies carried out on 1.6 105 m solutions also indicates that FeIII–mimochrome VI is monomeric in phosphate buffer (data not shown). Increasing the pH from 6.5 to 8.5 causes a further decrease of the Soret band intensity, with a concomitant increase of the N band intensity (ratio to the Soret band up to 0.71). In the visible region, the Q bands are observed as shoulders at 525 and 568 nm; the main bands at 485 and 600 nm can be attributed to charge-transfer bands of the high-spin iron hydroxide complex.[10] These spectroscopic features suggest that the optical transition at pH 7.4 corresponds to a ligand exchange from water to the hydroxide ion. The UV/Vis spectra of the complex are essentially unchanged between pH 8.5 and 10.0. Table 1. UV/Vis parameters of iron–mimochrome VI derivatives.[a] Compound x
Solvent
Soret
FeIII–mimochrome VI FeIII–mimochrome VI FeIII–mimochrome VI FeIII–mimochrome VI FeIII–mimochrome VI FeIII–mimochrome-VI Im[b] FeIII–mimochrome-VI NO FeII–mimochrome VI FeII–mimochrome-VI CO
water pH 2.0 buffer pH 5.5–6.5 buffer pH 8.5 buffer pH 6.5+50 % TFE buffer pH 7.0+50 % TFE buffer pH 7.0+50 % TFE buffer pH 7.0+50 % TFE buffer pH 7.0+50 % TFE buffer pH 7.0+50 % TFE
389 391 391 391 391 400 409 419 409
[a] sh = shoulder. [b] Im = imidazole.
4446
www.chemeurj.org
(110) (63) (39) (67) (65) (122) (122) (70) (78)
495 (4.7) 491 (4.2) 526 (sh) 490 (5.8) 490 (6.8) 525 (9.0) 524 (9.5) 544 (7.5) 530 (8.0)
Figure 3. Coordination properties of FeIII–mimochrome VI. a) UV/Vis spectra of FeIII–mimochrome VI (1.2 105 m) at various pH (2.0–10.0). The spectroscopic titration was performed using NaOH (at concentrations in the range 0.1–1 m), in a 1.0 cm path length cuvette. Arrows indicate changes of the Soret- and the 350 nm band from acidic to alkaline pH. b) Plot of the Soret absorbance at 387 nm against pH, showing a picture of the species (see arrows) involved in the pH-dependent equilibria.
Ligand binding: UV/Vis spectroscopy was also used to analyze the binding of small exogenous ligands to the heme. Figure 4 reports the UV/Vis spectra of ferric and ferrous mimochrome VI, together with the spectra of NO and CO adducts. Anaerobic addition of NO to III Fe –mimochrome VI resulted Visible region 1 1 in an optical spectrum typical l[nm] (e [mm cm ]) of a low-spin heme complex, 529 sh[a] 567 sh 611 (2.2) with a His/NO axial coordina530 sh 570 sh 614 (1.9) 565 sh 485 sh 594 (3.7) tion for the ferric ion 520 sh 602 (3.5) (Figure 4).[11] The binding of 520 sh 610 (4.7) NO was reversible, since the 565 sh original ferric spectrum was 558 (10) almost fully recovered by flush555 (7.0) ing the NO derivative with air for about 1 h. The dithionite-re-
2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 4444 – 4453
Heme–Peptide Metalloenzyme Mimetic
FULL PAPER
Figure 4. UV/Vis spectra of ferric and ferrous mimochrome VI (4.1 106 m) and analogues in (50 % TFE (v/v)) buffered solution at pH 7 (cell path length 1.0 cm); FeIII (·····); FeIII-NO (—);FeII (- -);FeII-CO (· - ·). Inset: visible region.
duced form, FeII–mimochrome VI, is characterized by a spectrum fully consistent with a high-spin five-coordinate FeII species, containing an axial histidine at the fifth coordination site (Figure 4).[12] Addition of carbon monoxide to a solution of FeII–mimochrome VI saturated with sodium dithionite, also results in a new UV/Vis spectrum typical of a heme-CO adduct.[13] Titration of FeIII–mimochrome VI with imidazole demonstrates the formation of a 1:1 complex with a dissociation constant of (5.26 0.02) 105 m ACHTUNGRE(Figure 5).[8a] The starting high-spin state is entirely converted to a low-spin state. UV/ Vis spectroscopy was also used to investigate the product of the reaction of FeIII–mimochrome VI and H2O2 in the absence of co-substrates. Addition of one equivalent of H2O2 to FeIII–mimochrome VI caused a 50 % decrease in the Soret band intensity without wavelength shifts, and flattened a/b bands (Figure 6). These spectral changes are consistent with the formation of a peroxidase compound I analogue.[14] FeIII–mimochrome VI compound I was stable in the reaction media for about 10 min, then it spontaneously decayed to the original ferric state, almost reversibly. Catalytic activity: The peroxidase-like activity of FeIII–mimochrome VI was assessed by the use of H2O2, 2,2’-azinodi(3-ethyl-benzothiazoline-6-sulfonic acid (ABTS),[15a] or guaiacol.[15b] Phenol nitration in the presence of NO2 was also studied. To optimize the experimental condition for catalytic activity, the kinetic parameters for the oxidation of ABTS by H2O2 were first measured as a function of TFE concentration and pH (Figure 7). The initial rate of ABTS oxidation is influenced by TFE, showing a maximum at 50 % TFE (v/v) in phosphate buffered solution, pH 6.5, (Figure 7 a). Next, the effect of pH on the activity was examined. The pH profile of the kcat for the ABTS+C formation is characterized by a bell-shaped curve, with a maximum activity at pH 6.5 (Figure 7 b).
Chem. Eur. J. 2011, 17, 4444 – 4453
Figure 5. Binding of imidazole to FeIII–mimochrome VI. UV/Vis spectra of FeIII–mimochrome VI at different imidazole concentration; arrows indicate changes of the Soret band upon imidazole addition. Upper inset: plot A– A0 against free ligand ([L], or unbound [Im0]) for titration of FeIII–mimochrome VI (5.4 106 m) with imidazole at 20 8C, in TFE (50 % (v/v)) phosphate buffered solution at pH 7.0. Absorbances are taken at 400 nm, and the data points correspond to [Im0] of 1.0 104 m, 2.0 104 m, 4.0 104 m, 1.2 103 m, 5.2 103 m, 10.8 103 m. Lower inset: plot of logACHTUNGRE(A– A0)/ACHTUNGRE(A1–A) against log of free ligand according to Equation (4) (see the experimental section); Hills coefficient, n = 1.03 0.03.
Figure 6. Reaction of FeIII–mimochrome VI with H2O2. UV/Vis spectra of FeIII–mimochrome VI (solid line), of FeIII–mimochrome VI compound I (dotted line) obtained by addition of 1 equivalent of H2O2 to the ferric complex (phosphate buffer pH 6.5, 50 % TFE (v/v)), and of FeIII–mimochrome VI (dashed line) regenerated after spontaneous decay of FeIII– mimochrome VI compound I. Inset: visible region.
Therefore, all the catalytic experiments were performed in 50 % TFE (v/v), pH 6.5. Kinetic parameters of FeIII–mimochrome VI were determined by varying H2O2 concentration using fixed concentrations of reducing substrate, and vice versa. Figure 8 reports the initial rates of ABTS and guaiacol oxidation at various substrate concentrations and fixed H2O2 concentration (50 mm H2O2 for ABTS oxidation,
2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4447
V. Pavone, A. Lombardi et al.
Figure 7. Kinetic parameters for the oxidation of ABTS by H2O2 as a function of TFE concentration and pH. a) Initial rates for ABTS oxidation at various TFE content. Reaction conditions were FeIII–mimochrome VI (2.0 107 m) in phosphate buffer (100 mm, pH 6.5), H2O2 (3.0 mm), ABTS (0.1 mm). b) kcat values for the oxidation of ABTS at various pHs. Solid line represents the data fitting according to Equation (8).
concentration at a fixed concentration of the reducing substrate (0.10 mm ABTS, 0.10 mm guaiacol). Data fitting, according to two-substrate Michaelis–Menten kinetics,[16] gave 3 2 2 O2 mm, K H = (8.7 0.3) mm, and kcat = K AH m = (9.2 0.4) 10 m (8.0 0.1) s1 (Table 2). In the experimental conditions for maximal activity (50 % TFE (v/v), pH 6.5), FeIII–mimochrome VI is capable of performing several turnovers (more than 4000) without bleaching (Figure 10 a), whereas it shows a linear reaction progress up to only 35 turnovers in the absence of TFE (Figure 10 b).
Figure 10. Progress curve for the oxidation of ABTS by FeIII–mimochrome VI (solid line) and FeIII–mimochrome VI mono-adduct (dashed line). a) phosphate buffer (100 mm, pH 6.5), TFE (50 %, (v/v)); b) phosphate buffer (pH 6.5, 100 mm). Reaction conditions; ABTS (0.10 mm), catalyst (2.0 108 m), and H2O2 (3.0 mm). Figure 8. Peroxidase-like activity of FeIII–mimochrome VI. a) Concentration-dependent initial rates for ABTS oxidation. Reaction conditions were FeIII–mimochrome VI (2.0 107 m), phosphate buffer (100 mm, pH 6.5), TFE (50 %, (v/v)), H2O2 (50 mm). b) Concentration-dependent initial rates for guaiacol oxidation. Reaction conditions were FeIII–mimochrome VI (2.0 107 m) in phosphate buffer (100 mm, pH 6.5), TFE (50 %, (v/v)), H2O2 (10 mm). The data was analyzed according to Equation (7) for a two-substrate Michaelis–Menten kinetic.
and 10 mm H2O2 for guaiacol oxidation). Data fitting, according to two-substrate Michaelis–Menten kinetics,[16] gave 2 2 2 O2 K AH mm, KH = (44 2) mm, and kcat = m = (8.4 0.2) 10 m 1 (371 14) s . Figure 9 reports the initial rates of ABTS (Figure 9 a) and guaiacol oxidation (Figure 9 b) versus H2O2
FeIII–mimochrome VI mono-adduct, characterized by the absence of the distal decapeptide chain, shows a linear reaction progress up to only 10 or 1500 turnovers in the absence or in the presence of 50 % TFE (v/v), respectively (Figure 10). Reaction of FeIII–mimochrome VI with phenol (1.0 mm) in the presence of NO2, and H2O2 resulted in the formation of both 4- and 2-nitrophenol (Scheme 1 and Figure 11).
Scheme 1. Reaction of phenol with NO2 and H2O2, in the presence of FeIII–mimochrome VI as a catalyst, to give both 4- and 2-nitrophenol.
Figure 9. Peroxidase-like activity of FeIII–mimochrome VI. a) Initial rate of ABTS oxidation at various H2O2 concentrations. Reaction conditions; FeIII–mimochrome VI (2.0 107 m) in phosphate buffer (100 mm, pH 6.5), TFE (50 %, (v/v)), and ABTS (0.10 mm). b) Initial rate of guaiacol oxidation at various H2O2 concentrations. Reaction conditions were FeIII–mimochrome VI (2.0 107 m) in phosphate buffer (100 mm, pH 6.5), TFE (50 %, (v/v)), guaiacol (0.10 mm). The data was analyzed according to Equation (7) for a two-substrate Michaelis–Menten kinetic.
4448
www.chemeurj.org
The NO2- and H2O2-concentration-dependent product yields were determined. Experiments were run at fixed concentration of phenol (1 mm) in the presence of FeIII–mimochrome VI (0.2 mm), and H2O2 (3 mm), at various NaNO2 concentrations (Figure 12 a), and at fixed concentration of phenol (1 mm) in the presence of FeIII–mimochrome VI (0.2 mm), and NaNO2 (20 mm), at various H2O2 concentrations (Figure 12 b). The yield of 4- and 2-nitrophenol at 40 min reaction time increased by increasing H2O2 and NO2 concentrations, and then it leveled off. The maximum yield of nitrophenols (4- and 2-) was about 15 % at NO2 and H2O2 concentration greater than 20 mm, and 1 mm, respectively.
2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 4444 – 4453
Heme–Peptide Metalloenzyme Mimetic
FULL PAPER
Table 2. Steady state kinetic parameters for H2O2 oxidation of ABTS and guaiacol catalyzed by FeIII–mimochrome VI, natural and mutated HRPs, and some heme–protein models.[a]
ABTS
Guaiacol
Enzyme
pH
2 O2 KH m [mm]
2 K AH m [mm]
kcat ACHTUNGRE[s1]
2 O2 kcat/K H m 1 1 ACHTUNGRE[mm s ]
2 kcat/K AH m ACHTUNGRE[mm1 s1]
FeIII–mimochrome VI MP8[b] MP11[b] HRP[c,d] HRP[c,e] * HRP[f,e] R38 L *HRP[e] R38 A hHRP[g,d] H42 A hHRP[h] FeIIIToCPP-13G10[i,j] FeIIIToCPP-14H7[i,j] FeIIIToCPP2[i,j] FeIII–mimochrome VI HRP[e] * HRP[e] R38L *HRP[e] R38 A hHRP[d] H42 A hHRP[h] Heme 2a 4[k]
6.5 7.0 7.0 4.6 7.0 7.0 7.0 4.6 4.6 4.6 5.0 5.0 6.5 7.0 7.0 7.0 9.0 6.0 7.4
44
8.4 102
371 2.6 103 13 103 4100 52.5 57.3 28.8 660 0.41 9.33 1.05 0.85 8.0 400 376 13.2 41 0.015 1.17
8.4 1.73 6.87
4417
11.5 103 13.6 103 8.2
16 9 42 8.7 10.7 102 8.3 102 3.5
0.8 5.1 3.8 10.8 0.40 6 103
9.2 103 0.8 0.9 2.3 0.6 3.8 103
9.0
4565 4213 3.51
0.58 0.12 0.02 0.91 3738 4530 3.8
5125 10.29 15.08 2.67 1650 68
870 500 417 5.7 68 4
0.13
[a] Parameters given refer to apparent values at fixed concentrations of the second substrate as indicated in the experimental section. kcat values correspond to variation of H2O2 at constant concentration of reducing substrate, but the values obtained with constant H2O2 and varying the reducing substrate were the same within the experimental error. [b] MP8 and MP11, the values of kcat/Km correspond to the apparent second order rate constant for ABTS formation (from ref. [14, 17] and this work), using the rate law dACHTUNGRE[ABTS+C]/dt = k2 [Enzyme] [H2O2]. [c] HRP = horseradish peroxidase isoenzyme C (glycosylated form). [d] Data from reference [18a]. [e] Data from reference [18b]. [f] *HRP = non-glycosylated recombinant HRP. [g] hHRP = polyhistidinetagged recombinant HRP. [h] Data from reference [18c]. [i] ToCPP = meso-a,a,a,b-tetrakis-orthocarboxyphenylporphyrin. [j] Data from reference [19]. [k] Data from reference [20].
Figure 11. HPLC profile obtained after 40 min reaction of phenol (1 mm) in the presence of FeIII–mimochrome VI (0.2 mm), NaNO2 (20 mm) and H2O2 (3 mm) in phosphate buffer (pH 6.5), and TFE (50 %, (v/v)).
Discussion FeIII–mimochrome VI encompasses basic features to work like a functional heme–protein mimetic including a proximal heme face with a His residue as axial ligand, and a distal face easily accessible to exogenous ligands. The spectroscopic characterization of FeIII–mimochrome VI demonstrated that in the pH range 5.5–6.5, the His/H2O species is predominant.[8, 21] The spectral parameters in this pH range are close to those observed for aquomet-deuteromyoglobin and deuterohemoglobin, in neutral solutions (Table 3). Microperoxi-
Chem. Eur. J. 2011, 17, 4444 – 4453
dases,[8, 22] chelated deuterohemin–histidine, and deuterohemin–peptide–histidine complexes[23] display similar spectral behaviors (Table 3). The first inflection point at pH 3.4, observed in the UV/Vis pH titration (Figure 3 b), represents the apparent pKa for the deprotonation of the His, with consequent coordination to the iron heme, and concomitant displacement of the proximal water molecule. This low His pKa is consistent with other studies on mimochrome analogues,[5a,c] and several porphyrin-peptide model systems (Table 4). The second inflection point at pH 7.4 corresponds to ligand exchange from water to the hydroxide ion.[10] Similar pKa values were exhibited by Aplysia myoglobin and Chironomus hemoglobin.[21] Binding of small exogenous ligands confirmed the presence of an easily accessible face of
Figure 12. Total concentrations of nitrophenols formed at various substrate concentrations. a) variable NaNO2 concentrations; phenol and H2O2 concentrations fixed at 1 mm and 3 mm, respectively. b) variable H2O2 concentrations; phenol and NO2concentrations fixed at 1 mm and 20 mm, respectively.
the heme (Figure 4 and Table 1). The UV/Vis spectral features, obtained upon NO addition, indicated a His/NO axial coordination for the ferric ion, similar to the NO adduct of HRP.[11] Furthermore, reaction of FeII–mimochrome VI with CO gave a complex with UV/Vis features very close to those of the CO adducts of HRP,[13a] DH-Mb and DHHb.[13b,c] Functional characterization demonstrates that FeIII–mimochrome VI behaves like an artificial peroxidase, because it catalyzes oxidation reactions by the activation of H2O2. The kinetic parameters were first optimized for the oxidation of
2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4449
V. Pavone, A. Lombardi et al.
Table 3. Electronic spectral data for some FeIII–porphyrin complexes and natural heme–proteins. Compound
Solvent Soret
AcMP-11[b] AcMP-11[b] AcMP-8[c] AcMP-8[c] DH[d] DH-His[d] DH-P11[e] DH-P11-Im[e] DH-Mb[f] DH-Hb[f] PSM-2 m[g]
water pH < 3 buffer (pH 7.5) water pH 1.4 buffer (pH 4–5) methanol/water methanol/water methanol methanol buffer (pH 7) buffer (pH 7) buffer (pH 7)+25 % PrOH
394 397 393.8 395 388 390 388 390 393 394 391
lmax [nm] b CT[a] 494 495 495 490 492 487 487 496 500 495
622 618.8 621.4 612 616 587 587 620 620 620
[a] CT = charge transfer band. [b] AcMP-11 = N-acetyl-microperoxidase– undecapeptide, from reference [22]. [c] AcMP-8 = N-acetyl-microperoxidase–octacapeptide, from reference [8]. [d] DH = deuterohemin, DHHis = chelated deuterohemin–histidine, from reference [23a]. [e] DHP11 = deuterohemin–undecapeptide complex, Im = imidazole, from reference [23b]. [f] DH-Mb = deuteromyoglobin, DH-Hb = deuterohemoglobin, from reference [24]. [g] PSM-2 m = peptide-sandwiched meso-heme– monopeptide analogue 2, from reference [25].
Table 4. The pKapp of histidine coordination to FeIII in various heme–peptide models. Complex
pKapp
[a]
MP-8 AcMP-8[b] MP-11[c] AcMP-11[d] MP9[e] FeIII–mimochrome I[f] FeIII–mimochrome IV[g] FeIII–mimochrome VI
3.5 3.1 3.4 3.4 2.9 2.5 3.9 3.4
[a] Microperoxidase–octapeptide, see reference [26]. [b] N-acetyl-MP-8, see reference [8a]. [c] Microperoxidase–undecapeptide, see reference [27]. [d] N-acetyl-MP-11, see reference [22]. [e] Microperoxidase– nonapeptide, see reference [28]. [f] Reference [5b]. [g] Reference [5c].
the ABTS as reducing substrate, and the maximal peroxidase activity was observed at pH 6.5, in 50 % TFE (v/v). The kcat bell-shaped pH dependence indicates that at least two ionizable groups are involved in the catalytic activity. Data fitting gave pKa1 = 5.37 and pKa2 = 7.61. This result indicates that the activity of FeIII–mimochrome VI is modulated by the deprotonation of an ionizable group having an apparent pKa of 5.37. The potential residues involved in the modulation of this effect are Asp or Glu residues in the distal side. The decrease in the kcat due to increasing the pH from 6.5 to 8.0 can be attributed to an increase in the molar fraction of the FeIII–mimochrome VI complex in the His/ OH iron coordinated form (Figure 2 b). The apparent pKa2 well agrees with the midpoint at pH 7.4 of the second optical transition observed in the UV/Vis pH titration. This pKa has been attributed to a ligand exchange of water with a hydroxide ion. This ligand exchange occurring at the distal site above pH 6.5, causes the complex to be less able to coordinate, and consequently to activate, the peroxide ion. The enhancement in the peroxidase activity induced by TFE, previously experimented by Mihara and co-workers,[20]
4450
www.chemeurj.org
suggests the possible structural role of the peptide chains in assisting the reaction. Preliminary NMR analysis of the diamagnetic CoIII–mimochrome VI derivative (phosphate buffered solution, pH 6.5, 50 % TFE (v/v)), shows the presence of several NOE contacts between the hydrophobic peptide chains and the heme. Therefore, it is reasonable to suppose that, similarly to previously developed mimochromes,[5] helix folding induced by TFE causes the two helical peptides to enwrap the heme faces, thus protecting the heme from solvent exposure. The structural basis for catalysis is further supported by the catalytic properties of FeIII–mimochrome VI mono-adduct, characterized by the absence of the distal decapeptide chain. This mono-adduct showed a kcat = (688 33) s1 in the presence of 50 % TFE (v/v), which is higher than the value displayed by the full FeIII–mimochrome VI molecule, in the same experimental conditions. This can be interpreted in terms of a faster access of the substrate and co-substrate to the iron site when the decapeptide chain is missing. At the same time, FeIII–mimochrome VI monoH2 O2 2 = (80 4) mm, adduct showed K AH m = (0.13 0.01) mm, K m which are about twice those observed for FeIII–mimochrome VI in the same experimental conditions. This is an indication that the decapeptide chain in FeIII–mimochrome VI assists the binding of the substrate and co-substrate. Most interestingly, the mono-adduct is more susceptible to oxidative degradation, as shown by the much lower number of turnovers that is able to perform, with respect to FeIII–mimochrome VI (see Figure 10), both in the presence or in the absence of 50 % TFE (v/v). This is a clear evidence of the protective effect to degradation exerted by the decapeptide chain. The kinetic parameters for H2O2 catalyzed oxidation of FeIII–mimochrome VI were compared with those of native and mutated peroxidases,[18] and some peptide- and proteinbased peroxidase models, such as microperoxidases[14, 17] and hemoabzymes[19] (see Table 2). In the oxidation of ABTS, none of the synthetic models or mutated proteins developed so far has kcat as high as that of HRP (kcat = 4100 s1, pH 4.6). However, FeIII–mimochrome VI approaches the HRP catalytic performance. It has a kcat value only 11-fold lower than that of HRP, in the experimental conditions for maximal activity for each enzyme, whereas the kcat value of FeIII–mimochrome VI is almost sevenfold higher than that of HRP at neutral pH. FeIII–mimochrome VI shows Km values in the mm range toward both the reducing substrates, ABTS and guaiacol. Notably, these values are much lower than those of HRPs, and are comparable to those of H42 A hHRP mutant. In addition, the apparent Km values for H2O2 activation are in the mm range and they are comparable to those of model compounds and mutated HRPs. The catalytic efficiency (kcat/Km) best describes the differences between FeIII–mimochrome VI and the other compounds reported in Table 2. In the activation of H2O2, FeIII–mimochrome VI is the most efficient catalyst known so far among synthetic models, although it is not as efficient as natural and recombinant HRPs. Instead, the catalytic efficiency (kcat/Km) for the reducing substrates approaches that of HRP for ABTS oxidation, and notably, it
2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 4444 – 4453
Heme–Peptide Metalloenzyme Mimetic
is about twofold higher than native HRP for guaiacol oxidation. FeIII-mimochrome VI is also an efficient catalyst in the nitration of phenols, by using NO2 in the presence of H2O2, as observed for several heme–protein and model compounds.[29, 30] The total yield of 2- and 4-nitrophenol in the presence of FeIII–mimochrome VI was higher than that in the presence of previously developed peroxidase mimetics,[30] and only fourfold lower than that of soybean peroxidase (SBP), one of the most efficient peroxidase in the enzymatic nitration of phenols.[29] The most important result for this activity is that FeIII–mimochrome VI is able to catalyze this reaction in the presence of high volume percent of organic co-solvent, such as TFE, whereas natural peroxidases are inactivated in similar experimental conditions. This finding could be very important for the application of this reaction in the nitration of compounds with limited aqueous solubility. Regarding the molecular mechanism of H2O2 activation by FeIII–mimochrome VI, spectroscopic analysis suggests that it is likely to proceed through the formation of compound I (Figure 6). This finding supports the hypothesis that the catalytic cycle of FeIII–mimochrome VI is similar to natural peroxidases.[6] Peroxidases (E) consume peroxide and oxidize a broad range of organic molecules (AH) to radicals (A*). The catalytic process occurs through several steps, which initially involves the reaction of hydrogen peroxide with heme, to give a two-electron oxidized protein and water. The oxidized protein is named compound I, and contains an oxoferryl center (FeIV=O), and a porphyrin or protein radical cation. Two sequential single-electron transfers from substrates reduce compound I to compound II, and then compound II to the ferric resting state.[6] E þ H2 O2 ! Compound I þ H2 O Compound I þ AH ! Compound II þ A* Compound II þ AH ! E þ A* þ H2 O Some substrates directly reduce the protein to the resting state without the formation of compound II, thus the reaction may proceed through a two-electron transfer, instead of two consecutive one-electron pathways. The activity of natural peroxidases is dictated by the nature of the proximal and distal heme environments.[31] The proximal His-axial ligand is hydrogen bonded to an aspartate residue and acts as an electron donor in stabilizing the high oxidation state of compound I (the push effect). The distal cavity of heme peroxidases is the site of the interaction with H2O2, and it is characterized by two constantly invariant amino acids, the distal histidine and the distal arginine; these two residues play a major role in the formation and stabilization of compound I (the pull effect). The catalytic activity of FeIII–mimochrome VI, which is comparable to those of complex proteins, demonstrated that its miniaturized structure holds essential elements for tuning the iron–porphyrin peroxidase activity.
Chem. Eur. J. 2011, 17, 4444 – 4453
FULL PAPER Conclusion All the spectroscopic and functional properties here described indicate that FeIII–mimochrome VI is an efficient heme–protein model, which accommodates a peroxidaselike active site. FeIII–mimochrome VI peptide framework, despite its small structure (a total of 24 amino acid residues), confers higher efficiency to the porphyrin cofactor than other peroxidase mimics. Three important outcomes deserve highlighting: 1) FeIII–mimochrome VI efficiently catalyzes the oxidation of different substrates, such as ABTS and guaiacol, by activating H2O2, and efficiently catalyzes the nitration of phenols; 2) FeIII–mimochrome VI displays a very high specific activity (104 mol mg1 s1 for ABTS oxidation), with respect to highly purified HRPs (91 mol mg1 s1 for ABTS oxidation at pH 4.6); these value highlights both its high catalytic efficiency and small molecular mass, compared with natural peroxidases that contain more than 300 amino acid residues; 3) FeIII–mimochrome VI exhibits multiple turnover kinetics; more than 4000 turnovers within 10 min were observed in the ABTS oxidation (Figure 10). All these features indicate that FeIII–mimochrome VI is an attractive, low-molecular-weight heme–enzyme, which will serve as an excellent scaffold for further design of hemebased biocatalysts.
Experimental Section General methods: TFE was supplied by Romil. MP-11 was from Aldrich. ABTS, guaiacol, and phenol were from Sigma, and they were used without further purification. H2O2 (30 %, v/v) was from Fluka; H2O2 concentrations were determined spectrophotometrically (lmax (e) = 240 nm (39.4 m1 cm1)). Design: Molecular design was performed on a Silicon Graphic Octane 2 workstation. The program package InsightII/Discover with Extensible Systematic Force Field (ESFF),[32] was used for energy minimization. Sample preparation for UV and CD analysis: Stock solutions of FeIII– mimochrome VI (1.0–3.0 104 m) in TFA (0.1 % (v/v)) were prepared and stored at 4 8C when not in use. Their concentrations were determined by flame atomic absorption spectrometry, on the basis of their metal content. Typically, 1.0–2.0 105 m solutions of metal-reconstituted mimochrome VI (corresponding to approx. 0.5–2 mg L1 of metal) in ultra pure metal-free water with TFA (0.1 % (v/v), pH 1.9) were directly aspired into an air-acetylene flame with no prior treatment. Concentrations were obtained by comparison with calibration curves. TFA/water mixtures (0.1 % (v/v), pH 1.9) used to dissolve FeIII–mimochrome VI were also checked for metal content, and found to be free of metal within the experimental error. FeIII–mimochrome VI stock solutions, analyzed for metal contents, were appropriately diluted and used for determining the extinction coefficients at the Soret band maximum wavelength. UV/Vis spectroscopy: UV/Vis spectra were recorded on a Cary Varian 5000 Spectrophotometer equipped with a multi-cell holder and temperature controller. Temperature within the cell was measured using the Peltier thermocouple. Quartz cuvettes with a pathlength of 1.0 cm were used for most measurements. Wavelength scans were performed at 25 8C (unless otherwise specified) from 200 to 800 nm, with a 300 nm min1 scan speed. Sample concentrations in the range 1.0 106–1.0 105 m were used for the determination of the extinction coefficients at the Soret maximum in acidic (0.1 % TFA in water) and neutral (10 mm, pH 7 phosphate buffer) solutions, respectively.
2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4451
V. Pavone, A. Lombardi et al.
NO gas was produced in situ by addition of degassed H2SO4 (2 m) to NaNO2 (3 g). (NO)-FeIII–mimochrome VI was then obtained by flushing the so-produced NO gas directly to a septum-sealed UV cell containing a degassed solution of FeIII–mimochrome VI in phosphate buffer (pH 7.0, 50 % TFE (v/v)). Reduction of the FeIII–mimochrome VI to the ferrous state was accomplished as follows: a solution of FeIII–mimochrome VI (4.1 106 m) in pH 7 phosphate buffer (10 mm) with 50 % TFE (v/v) was taken into a quartz cuvette, which had been thoroughly purged with nitrogen and fitted with a rubber septum. A degassed saturated solution of sodium dithionite (Na2S2O4) was subsequently added to the deoxygenated FeIII–mimochrome VI solution, followed by gentle shaking of the solution, and the spectrum was immediately recorded. The CO adduct was formed from the reduced sample by flushing with CO for 15 min. UV/Vis spectroscopy was also used to investigate the product of the reaction of FeIII–mimochrome VI and H2O2 in the absence of the co-substrates. Spectral changes upon addition of H2O2 (1 equiv) to FeIII–mimochrome VI (20 mm solution) in phosphate buffer (100 mm, pH 6.5) and TFE (50 % (v/v)) were recorded for 10 min. Imidazole titration: Titration of FeIII–mimochrome VI with imidazole was carried out in phosphate buffer (10 mm, pH 7) and TFE (50 % (v/v)) at room temperature. Imidazole was dissolved in phosphate buffer to final concentrations of 0.2 and 2.0 m, and the pH of the solution was adjusted with H3PO4. Aliquots were titrated into the cuvette containing the FeIII–mimochrome VI solution using a 10 mL Hamilton syringe and the solution thoroughly stirred (dilution was less than 1 %). Equilibrium was rapidly established, typically within 1 min of mixing time. No pH change was detected at the end of the titration experiment. The equilibrium constant for ligand–adduct formation by FeIII–mimochrome VI was determined by monitoring the absorbance changes at 400 nm (the Soret maximum for the imidazole adduct of FeIII–mimochrome VI) as a function of ligand concentration. The data was analyzed assuming the equilibria: M þ nL Ð MLn The equilibrium constant can be expressed by Equation (1):
K¼
½MLn ½M½Ln
ð1Þ
in which [M], [ML], and [L] represent the concentrations of unbound FeIII complex, FeIII-ligand adduct, and free ligand, respectively; the fraction of FeIII-ligand adduct formed can be expressed as in Equation (2): A A0 ½MLn ¼ A1 A0 ½MTOT
ð2Þ
in which A0, A and A1 are the absorbances at a given wavelength in the absence, presence, and at saturation concentrations of ligand, respectively; Equation (1) can then be rewritten as in Equation (3): K½Ln ¼
A A0 A1 A0
½Im0 ¼
½ImTOT 1þ
ð5Þ
½Hþ Ka1
The concentration of free ligand in solution was evaluated from Equation (6) ½L ¼ ½Im0 ½M0
A A0 A1 A0
ð6Þ
in which M0 is the total concentration of FeIII–mimochrome VI. Circular dichroism spectroscopy: CD experiments were performed on a Jasco J-715 circular dichroism spectropolarimeter at 20 8C (unless otherwise specified). Temperature control was achieved using a Peltier ( 0.5 8C accuracy). For all studies, cell path length (l) was 0.5 cm. Spectra in the 190–260 nm region were collected at 0.2 nm intervals with a 5 nm min1 scan speed, 2 nm bandwidth and a 16 s response. The concentration of FeIII–mimochrome VI was 7.3 106 m. Spectra are reported in terms of mean residue ellipticity, calculated by dividing the total molar ellipticity by the number of amino acids in the molecule. Catalytic studies on ABTS and guaiacol oxidation: The catalytic experiments were followed using a Varian Cary 50 spectrophotometer, by measuring the appearance of the products in the reaction medium. The substrates used in the catalytic assays were ABTS[15a] and guaiacol.[15b] The formation of ABTS+C cation radical was followed at 660 nm (lmax (e) = 660 nm (1.40 104 m1 cm1)). The formation of guaiacol oxidation product (cyclic tetraguaiacol) was followed at 470 nm, considering a e470 = 2.66 104 m1 cm1. Kinetic parameters of FeIII–mimochrome VI were determined by varying the H2O2 concentration by using fixed concentrations of the reducing substrates, and vice versa. In the experiments performed at various H2O2 concentrations (in the range 0.01–200 mm) the ABTS concentration was kept constant at 0.1 mm. In the experiments performed at various ABTS concentrations (in the range 0.005–0.1 mm) the H2O2 concentration was 50 mm. In all the experiments the FeIII–mimochrome VI concentration was 2.0 107 m, and the reaction volume was 600 mL. In the experiments with guaiacol as substrate, the guaiacol concentration was kept constant at 0.1 mm, while H2O2 concentrations were in the range 1–40 mm. In the experiments performed at various guaiacol concentrations, in the range 0.0025–0.07 mm, the H2O2 concentration was 10 mm. Kinetic studies of MP11 were carried out at H2O2 concentrations in the range 0.025–0.6 mm, while ABTS concentration was kept constant at 0.1 mm in individual set of experiments. A solution of MP11 (2.0 106 m), phosphate buffer (0.1 m), pH 7.0, was used in all the kinetic experiments. The Km and kcat were determined using a two-substrate Michaelis– Menten kinetic model.[16] The data were analyzed by using the following Equation (7): v¼
½E0 1 kcat
K
K
ð7Þ
þ kcatm½A þ kcatm½B A
B
ð3Þ
in which v is the initial rate, [E]0 is the enzyme concentration, [A] is the H2O2 concentration, [B] is the reducing agent concentration.
Ligand stoichiometry and the equilibrium constant were thus determined from Equation (4)
In experiments performed at different pHs, kcat values for the oxidation of ABTS were determined using a FeIII–mimochrome VI concentration of 2.0 107 m, a fixed concentration of H2O2 (3.0 mm), and ABTS concentration ranging from 0.0025 to 0.10 mm phosphate solution (0.1 m) and TFE (50 % (v/v)) at different pHs were used as the solvent.
log
A A0 ¼ log K þ n log½L A1 A0
ð4Þ
The data were analyzed according to Equation (8):[33]
In aqueous solutions imidazole undergoes the following dissociations: þ
kcat H0 ¼
ImH Ð Im0 Ð Im
with pKa1 = 7.02 and pKa2 = 14.44; pKa2, which can be neglected because the experiments were performed at neutral pH. Since only Im0 is capable of coordination to a metal center, its concentration can be calculated by Equation (5):
4452
www.chemeurj.org
k
1þ
cat ½Hþ Ka1
K
þ ½Ha2þ
ð8Þ
where kcatH’ is the pH independent turnover number, and Ka1 and Ka2 denotes the protolytic dissociation constants of ionizable residues. Phenol nitration: The standard incubation for phenol nitration was performed by using phenol (1 mm) in the presence of FeIII–mimochrome VI
2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 4444 – 4453
Heme–Peptide Metalloenzyme Mimetic
(0.2 mm), NaNO2 (20 mm), and H2O2 (3 mm) in phosphate buffer (pH 6.5) and TFE (50 % (v/v)). The reactions were performed at room temperature with an incubation time of 40 min. The reaction mixtures were analyzed by analytical HPLC on a Phenomenex Gemini C18 column (150 4.6 mm, 5 mm), eluted with a TFA A) H2O/0.1 % and TFA B) CH3CN/0.1 % linear gradient from 10 % to 90 % B over 20 min, at 1 mL min1 flow rate. The concentrations of the starting material and products were determined from calibration curves constructed using commercial samples and 4-cyanophenol as internal standard.
Acknowledgements This work was supported by the Italian Ministry of University and Scientific Research (PRIN 2007 KAWXCL). We thank Dr. Marco Trifuoggi for experimental assistance with metal content analysis.
[1] a) D. N. Bolon, C. A. Voigt, S. L. Mayo, Curr. Opin. Chem. Biol. 2002, 6, 125 – 129; b) R. L. Koder, P. L. Dutton, Dalton Trans. 2006, 3045 – 3051; c) D. N. Bolon, S. L. Mayo, Proc. Natl. Acad. Sci. USA 2001, 98, 14274 – 14279; d) J. Razkin, J. Lindgren, H. Nilsson, L. Baltzer, ChemBioChem 2008, 9, 1975 – 1984; e) L. Jiang, Science 2008, 319, 1387 – 1391; f) D. Rçthlisberger, Nature 2008, 453, 190 – 195; g) J. Kaplan, W. F. DeGrado, Proc. Natl. Acad. Sci. USA 2004, 101, 11566 – 11570; h) M. Faiella, C. Andreozzi, R. Torres Martin de Rosales, V. Pavone, O. Maglio, F. Nastri, W. F. DeGrado, A. Lombardi, Nat. Chem. Biol. 2009, 5, 882 – 884. [2] a) Y. Lu, N. Yeung, N. Sieracki, N. M. Marshall, Nature 2009, 460, 855 – 862; b) Y. Lu, Angew. Chem. 2006, 118, 5714 – 5728; Angew. Chem. Int. Ed. 2006, 45, 5588 – 5601; c) T. Ueno, Angew. Chem. 2010, 122, 3958 – 3959; Angew. Chem. Int. Ed. 2010, 49, 3868 – 3869. [3] a) D. Mansuy, C. R. Chim. 2007, 10, 392- 413; b) E. Y. Tshuva, S. J. Lippard, Chem. Rev. 2004, 104, 987 – 1012. [4] W. F. DeGrado, C. M. Summa, V. Pavone, F. Nastri, A. Lombardi, Ann. Rev. Biochem. 1999, 68, 779 – 819. [5] a) A. Lombardi, F. Nastri, V. Pavone, Chem. Rev. 2001, 101, 3165 – 3189; b) F. Nastri, A. Lombardi, G. Morelli, O. Maglio, G. D’Auria, C. Pedone, V. Pavone, Chem. Eur. J. 1997, 3, 340 – 349; c) A. Lombardi, F. Nastri, D. Marasco, O. Maglio, G. De Sanctis, F. Sinibaldi, R. Santucci, M. Coletta, V. Pavone, Chem. Eur. J. 2003, 9, 5643 – 5654; d) G. D’Auria, O. Maglio, F. Nastri, A. Lombardi, M. Mazzeo, G. Morelli, L. Paolillo, C. Pedone, V. Pavone, Chem. Eur. J. 1997, 3, 350 – 362; e) L. Di Costanzo, S. Geremia, L. Randaccio, F. Nastri, O. Maglio, A. Lombardi, V. Pavone, J. Biol. Inorg. Chem. 2004, 9, 1017 – 1027; f) A. Lombardi, F. Nastri, M. Sanseverino, O. Maglio, C. Pedone, V. Pavone, Inorg. Chim. Acta 1998, 275–276, 301 – 313. [6] H. B. Dunford, J. S. Stillman, Coord. Chem. Rev. 1976, 19, 187 – 251. [7] J. W. Buchler in The Porphyrins, Vol. 1: Structure and Synthesis (Ed.: D. Dolphin), Academic Press, New York, 1979, pp. 389 – 483. [8] a) O. Q. Munro, H. D. Marques, Inorg. Chem. 1996, 35, 3752 – 3767; b) H. D. Marques, Dalton Trans. 2007, 4371 – 4385. [9] D. W. Urry, J. W. Pettegrew, J. Am. Chem. Soc. 1967, 89, 5276 – 5283. [10] a) T. Yonetani, H. Anni, J. Biol. Chem. 1987, 262, 9547 – 9554; b) A. Feis, M. P. Marzocchi, M. Paoli, G. Smulevich, Biochemistry 1994, 33, 4577 – 4583. [11] T. Yonetani, H. Yamamoto, J. E. Erman, J. S. Leigh, G. H. Reed, J. Biol. Chem. 1972, 247, 2447 – 2455.
Chem. Eur. J. 2011, 17, 4444 – 4453
FULL PAPER [12] a) F. Adar in The Porphyrins, Vol. 3: Physical Chemistry, (Ed.: D. Dolphin), Academic Press, New York, 1978, Chapter 2; b) W. A. Eaton, J. Hofrichter, Methods Enzymol. 1981, 76, 175 – 261. [13] a) B. A. Wittenberg, E. Antonini, M. Brunori, R. W. Noble, J. B. Wittenberg, J. Wyman, Biochemistry 1967, 6, 1970 – 1974; b) A. Rossi-Fanelli, E. Antonini, Arch. Biochem. Biophys. 1957, 72, 243 – 246; c) E. Antonini, M. Brunori, A. Caputo, E. Chiancone, A. Rossi-Fanelli, J. Wyman, Biochim. Biophys. Acta Spec. Sect. Biophys. Subj. 1964, 79, 284 – 292. [14] P. A. Adams, R. D. Gold, J. Chem. Soc. Chem. Commun. 1990, 97 – 98. [15] a) R. E. Childs, W. G. Bardsley, Biochem. J. 1975, 145, 93 – 103; b) R. S. Koduri, M. Tien, J. Biol. Chem. 1995, 270, 22 254 – 22 258. [16] W. W. Cleland, Biochim. Biophys. Acta 1963, 67, 104 – 137. [17] P. A. Adams, J. Chem. Soc. Perkin Trans. 2 1990, 1407 – 1414. [18] a) M. I. Savenkova, J. M. Kuo, P. R. Ortiz de Montellano, Biochemistry 1998, 37, 10828 – 10836; b) J. N. Rodriguez-Lopez, A. T. Smith, R. N. F. Thorneley, J. Biol. Chem. 1996, 271, 4023 – 4030; c) M. I. Savenkova, S. L. Newmyer, P. R. Ortiz de Montellano, J. Biol. Chem. 1996, 271, 24 598 – 24 603; d) S. L. Newmyer, P. R. Ortiz de Montellano, J. Biol. Chem. 1996, 271, 14 891 – 14 896; e) J. N. RodriguezLopez, A. T. Smith, R. N. F. Thorneley, J. Biol. Inorg. Chem. 1996, 1, 136 – 142. [19] R. Ricoux, Q. Raffy, J. P. Mahy, C. R. Chim. 2007, 10, 684 – 702. [20] S. Sakamoto, S. Sakurai, A. Ueno, H. Mihara, Chem. Commun. 1997, 1221 – 1222. [21] M. Brunori, G. Amiconi, E. Antonini, J. Wyman, R. Zito, A. R. Fanelli, Biochem. Biophys. Acta Protein Struct. 1968, 154, 315 – 322. [22] A. D. Carraway, M. G. McCollum, J. Peterson, Inorg. Chem. 1996, 35, 6885 – 689. [23] a) L. Casella, E. Monzani, M. Gullotti, L. De Gioa, A. Strini, F. Chillemi, Inorg. Chem. 1996, 35, 439 – 444; b) L. Casella, M. Gullotti, L. De Gioa, F. Monzani, F. Chillemi, J. Chem. Soc. Dalton Trans. 1991, 2945 – 2953. [24] E. Antonini, M. Brunori, Hemoglobin and Myoglobin in their Reactions with Ligands, North-Holland Publishing, Amsterdam, 1971. [25] a) P. A. Arnold, D. R. Benson, D. J. Brink, M. P. Hendrich, G. S. Jas, M. L. Kennedy, D. T. Petasis, M. Wang, Inorg. Chem. 1997, 36, 5306 – 5315; b) D. R. Benson, B. R. Hart, X. Zhu, M. B. Doughty, J. Am. Chem. Soc. 1995, 117, 8502 – 8510. [26] D. A. Baldwin, H. M. Marques, J. M. Pratt, J. Inorg. Biochem. 1986, 27, 245 – 254. [27] M. T. Wilson, R. J. Ranson, P. Masiakowski, E. Czarnecka, M. Brunori, Eur. J. Biochem. 1977, 77, 193 – 199. [28] D. A. Baldwin, M. B. Mabuya, H. M. Marques, S. Afr. J. Chem. 1987, 40, 103. [29] C. L. Budde, A. Beyer, I. Z. Munir, J. S. Dordick, Y. L. Khmelnitsky, J. Mol. Catal. B 2001, 15, 55 – 64. [30] a) R. Ricoux, E. Girgenti, H. Sauriant-Dorizon, D. Blanchard, J. P. Mahy, J. Protein Chem. 2002, 21, 473 – 477; b) R. Ricoux, J. L. Boucher, D. Mansuy, Eur. J. Biochem. 2001, 268, 3783 – 3788. [31] J. H. Dawson, Science 1988, 240, 433 – 439. [32] S. Barlow, A. L. Rohl, S. Shi, C. M. Freeman, D. OHare, J. Am. Chem. Soc. 1996, 118, 7578 – 7592. [33] J. Wang, T. Araki, M. Matsuoka, T. Ogawa, Biochim. Biophys. Acta Protein Struct. Mol. Enzymol. 1999, 1435, 177 – 183.
2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Received: December 2, 2010 Published online: March 17, 2011
www.chemeurj.org
4453
