Cytochrome c covalently immobilized on mesoporous silicas as a peroxidase: Orientation Effect
Kun-Che Kao,a Chia-Hung Lee,a Tien-Sung Lin,*b and Chung-Yuan Mou*a
a
Department of Chemistry, National Taiwan University, Taipei, Taiwan 106
b
Department of Chemistry, Washington University, St. Louis, MO 63130 USA
*Corresponding authors: TSL: Fax: 314-935-4481; E-mail:
[email protected] CYM: Fax: +886-2-23660954; E-mail:
[email protected]
Abstract Cytochrome c (cyt c), a heme protein with positive electric charge and a global dimension of 2.5×2.5×3.7 nm, is immobilized by covalent bonding in the nanochannels and on the surface of IBN4 (pore size = 5.3 - 7.1 nm) mesoporous silicas. The composite material behaves as peroxidase in the oxidation of organic molecules in the presence of hydrogen peroxide. The surface of IBN4 was first modified with three different linkers, glutaric anhydride (GAC), glutaraldehyde (GAH) and succinimido 3-maleimido propanoate (SMP), to facilitate the binding (bioconjugation) with cyt c. Different linker exposes the catalytic active site (Fe-heme) in different environment which allows us to examine the orientation effect imposed by the binding linker. Molecular modeling further allows us to assess the orientation effect on catalytic activity arising from the distribution of electric charges of cyt c immobilized in different surface modified nanochannels. The accessibility of the Fe active center in immobilized cyt c is found in the following order: IBN4-N-SMP-cyt c > IBN4-N-GAH-cyt c > IBN4-N-GAC-cyt c which is correlated to the measured trend of the initial specific peroxidase-like activities of immobilized cyt c in three different modified surface towards oxidation of 4-aminoantipyrine. The surface modified nanochannels of mesoporous silica provide the confining spaces that could prevent cyt c from protein unfolding and orient the active site in a favorable location in the pores to facilitate its activity. However, there are much more structure decay after hydrothermal treatment and the activities diminish accordingly: the IBN4-N-GAH-cyt c sample lost most of its activity, the IBN4-N-SMP-cyt c lost its activity due to the less protection of the active center of Fe-heme, and the IBN4-N-GAC-cyt c retained good activity. These temperature effects are further confirmed in the UV-Vis spectra and EPR studies. Cyt c immobilized on functionalized IBN4 surface exists in high spin state as inferred from EPR and UV-Vis studies which differs from the primarily low
spin state of native cyt c. The high spin state arises from the replacement of Met-80 ligands of heme Fe (III) by water or silanol group on the silica surface, which could open up the heme groove for easy access of oxidants to iron center and facilitate the catalytic activity. Finally, we apply the covalently immobilized cyt c in the oxidation of a representative polycyclic aromatic hydrocarbon (PAH) – pyrene. The trend in activity can be understood from the design principle we learned in this work.
1. Introduction Peroxidases can catalyze degradation or transformation of polycyclic aromatic hydrocarbons and dyes.1-3 They are capable of treating various types of recalcitrant aromatic compounds in the presence of redox mediators in large scale. Our interest is to develop cytochrome c (cyt c) as an oxidation catalyst for oxidizing organic compounds including polycyclic aromatic hydrocarbons (PAHs). Cytochrome c is an interesting oxidation catalyst. Native cyt c, when bound on the outer membrane of mitochondria, functions as electron shuttle only.4 In this state, all the six coordination sites in Fe of cyt c are occupied so that no oxidants can approach Fe center. However, in recent years it is gradually realized that protein modification can confer additional peroxidase properties to cyt c.5 This opens up the possibility of employing cyt c as a peroxidase catalyst for oxidizing PAHs or other organic compounds in aqueous solution if we learn how to modify the structure of cyt c to enhance its peroxidase activity while still keeping it stabilized. Recently we reported cyt c, when immobilized on mesoporous silica, is a good biocatalyst (peroxidase-like) to decompose PAHs in the presence of hydrogen peroxide and organic hydroperoxides.6, 7 However, the degradation of a hydrophobic substrate like PAHs is usually limited by its low water solubility, diffusion rate and instability of the enzyme at extreme conditions. Many approaches have been reported to overcome these difficulties: utilization of nontoxic surfactants to enhance the degradation of PAHs,8 water-miscible organic solvents to solve the solubility problem,9 raising temperature to increase the solubility and diffusion rate. Most notably, Vazquez-Duhalt et al. used chemical modification on the heme protein surface with poly-(ethylene glycol).10 The modified cyt c has cage-like polymer protection and shows
good activity at high temperatures.11 Site-directed mutagenesis showed that catalytic activity was improved by ten-fold in Gly82:Thr102 variant in comparison with the wild-type.10, 12 It is found in mesoporous materials the confinement effect would often stabilize the enzyme if proper choice of pore size to fit the enzyme molecule is established. Mesoporous silica materials provide tunable and uniform pore system, functionalizable surfaces, and restricted nanospaces for enzyme immobilization.13 Mesoporous silica materials further possess some unique properties that are suitable for encapsulation of proteins with great utility, such as huge surface area (about 1000 m2/g), controllable geometric parameters and well-defined physical and chemical properties, good mechanical and chemical stability, and high affinity by proper design to the types and sizes of molecules that can penetrate them. These properties offer many options in internal decoration of the mesopore upon loading of an enzyme into the pores. The applications of mesoporous silicas further facilitate the separation and re-use in the catalytic processes.14 Excluded volume is one of the important factors determining the stability of the confined protein.15, 16 In our previous work of immobilization of cyt c in mesoporous silica materials, it was found cyt c adsorbed in MCM-41 and MCM-48 with just about the right pore size to fit cyt c snugly have the highest hydrothermal stability and overall catalytic activity compared to other mesoporous silica with larger or smaller pore size.6 We took the advantage of electrostatic interaction between cyt c and aluminosilicate to load the enzyme. The advantage of physical adsorption method is easy operation and control. However, its disadvantage is leakage, particularly at high ionic strength. On the other hand, covalent attachment of enzyme in the pore would be a better method to immobilize enzyme. Recently, Lee et al. have reviewed various methods of covalent attachment
of enzymes to mesoporous silica.13 There are many advantages of covalently attaching enzymes to mesoporous silica materials. First, the most obvious one is the avoidance of the enzyme leakage problem. Secondly, through careful choice of various linker groups, one has certain choices of the amino residue to be attached. This then opens up a new possibility, e.g. one can have some control over the orientation of the attachment by choosing the amino acid to be the anchor site. The orientation of enzyme in the constricted pore space is important because substrate’s access to the active center of the enzyme is often through narrow pocket along certain direction.17 Side chain specific chemical modification certainly can control to some extent the orientation of the enzyme.18, 19 Finally, selected covalent attachment of enzyme at certain local point can constitute a local perturbation of the protein structure which may be reflected in distortion of the environment near active center. This opens up possibility of modulation of catalytic activity. Cyt c is a particular interesting example in that local to global structural perturbation depends on linking site.20 Cyt c has multiple functions depending on the fine tuning of the local structure at the active site. It is the purpose of this work to report the effect of confining and orienting cyt c in the nanochannels of mesoporous silica materials as new materials for water treatment. We have three different ways of covalently attaching the cyt c molecule; each probes the perturbation of the structure on different sites. Thus, they will expose the catalytic active site in different orientations inside the cylindrical nanochannels. Specifically, we applied IBN4 mesoporous materials21 (pore size = 5.3 - 7.1 nm) to examine the stability and activity of immobilized cyt c toward electron donors in the presence of H2O2. We chose the IBN4 materials for its ultra-fine particle size and its relatively large pore size to accommodate cyt c molecules. The increased stability of the confined enzyme was investigated by UV-Vis spectra. The disruption of the
heme-Fe binding, crucial for the peroxidase activity, will be investigated by EPR technique.22 For testing the peroxidase activity, we first carry out the aqueous reaction between phenol and 4-aminoantipyrine to form quinoneimine.23 Then, we did molecular modeling of cyt c and calculated charge distribution of the various surface amino acids on cyt c from its tertiary structure. Thus, the spatial orientation of the cyt c molecule under the three different ways of attachment to the nanochannels can be pictured. This will help us understanding the accessibility of the active site. Combination of the activities of oxidation of 4-aminoantipyrine with spectroscopic and structural analysis a rich understanding of the structure-activity relationship in the peroxidase activity of cyt c immobilization in mesoporous silica materials was given. Finally, we investigated the catalytic activity of oxidation of pyrene in the presence of H2O2. Similar trends in catalytic activities in functionalization methods and hydrothermal treatment were observed. The correlations of activity and structural variations will provide design basis for catalytic materials to be applied in more practical fields, such as water treatment.
2. Experimental 2.1. Chemicals Cytochrome c from Saccharomyces cerevisia, horse heart cytochrome c, pluronic P123 (EO20PO70EO20), and 4-aminoantipyrine were obtained from Aldrich-Sigma Chemical Co. Tetraethyl orthosilicate (TEOS), 3-aminopropyltrimethoxy silane (APTMS), glutaric anhydride (GAC), glutaraldehyde (50% in H2O) (GAH), 1-Ethyl-3-(3-dimethylaminopropyl)carbodiimide (EDC), phenol, pyrene, hydrogen peroxide (35%) were obtained from Acros. Succinimido
3-maleimido propanoate (SMP) is from Fluka. FC-4, is from Yick-Vic Chemicals & Pharmaceuticals (HK) Ltd. All the chemicals were used as received without further purification. 2.2. Synthesis and functionalization of IBN4 IBN4 was synthesized by the method as reported by Han et al.21 Typically, 0.25 g of -
Pluronic P123 and 0.7 g of FC-4, (C3F7O(CFCF3CF2O)2CFCF3CONH(CH2)3N+(C2H5)2CH3I ), were dissolved in 40 ml of 0.02 M HCl solution and then 1.0 g of TEOS was added. After stirring at 30oC for 20 h, the solution was hydrothermal treated at 100oC for 1 day. The as-synthesized material was collected by centrifugation, dried at 60oC for 12 h, and then calcined at 560oC for 6 h to remove the surfactants. To functionalize the IBN4, APTMS was grafted into the silica surface by post-modification method (IBN4-N). Typically, 3.5 g of IBN4 was well-dispersed in 350 mL of toluene, and 1.75 g of APTMS was added to the mixture. After stirring at room temperature for 1 h and refluxing at 100oC for 9 h, the solids were centrifuged, washed by toluene, and dried in vacuum for 10 h. To further derive various linkers on the amino group, three reactions were performed: (1) the IBN4-N-GAC sample, 1 g of IBN4-N was dispersed in 30 mL of DMF and then 0.99 g of glutaric anhydride (GAC) was added. After stirring at room temperature for 1 h, the solution was heated to 60oC with stirring for 9 h. (2) the IBN4-N-SMP sample, 1 g of IBN4-N was dispersed in 45 mL of dichloromethane, 0.57 g of succinimido 3-maleimido propanoate (SMP) was added, and then stirring at room temperature for 10 h. (3) the IBN4-N-GAH sample, 1 g of IBN4-N was dispersed in 30 mL of sodium phosphate buffer (pH 7.4), 1.54 mL of glutaraldehyde (50%) (GAH) was added, and stirring at room temperature for 10 h. These three functionalized materials were collected by centrifugation, washed by acetone, and dried in vacuum for 10 h. 2.3. Immobilization of cyt c in different modified IBN4
40 μM of cyt c stock solution was prepared by dissolving 0.066 g of cyt c in 132 mL of 0.01 M NaH2PO4-Na2HPO4 buffer solution (pH 7.4). For IBN4-N-GAC, 0.2 g of sample was dispersed into 15 mL of 0.01 M NaH2PO4-Na2HPO4 buffer solution (pH 6.0) under stirring. 0.1 g of EDC was added into the mixture in order to activate carboxylic groups. After 10 min, 15 mL of cyt c stock solution was added under stirring for 4 h at ice-bath. For IBN4-N-SMP and IBN4-N-GAH, 0.2 g of samples were dispersed into 15 mL of 0.01 M NaH2PO4-Na2HPO4 buffer solution, respectively. After adding each 15 mL of cyt c stock solution, pH values were tune to 9.0 by adding 0.01 M NaOH and stirred for 4 h at ice-bath. After immobilization, the solid samples of IBN4-cyt c materials were collected by centrifugation, washed by buffer solution and dried in vacuum for 3 h. The loading amounts of cyt c were determined by monitoring the change of Soret band at 409 nm from UV-Vis spectra (ε = 100,000 M-1 cm-1, Soret band). The solid powders of the immobilized cyt c were stored in - 20oC. The overall surface functionalization and immobilization of cyt c is illustrated in Scheme 1. 2.4. Characterization TEM images were performed on a transmission electron microscopy Hitachi H-7100 operating at 100 kV. Dispersed IBN4 aqueous solution was deposited on a carbon coated Cu grids and dried under air atmosphere. The structures of various surface modified IBN4 samples were analyzed by powder X-ray diffraction (Scintag X1 diffractometer, CuKα radiation at λ = 0.154 nm). The surface area, pore size, and pore volume were determined by N2 adsorption-desorption isotherms obtained at 77 K on a Micrometric ASAP 2010 apparatus. The samples were degassed at 10-3 Torr at 100 °C for 16 h prior to the adsorption experiment. The pore size distribution curves were obtained from the analysis of the adsorption portion of the isotherms using the BJH (Barrett-Joyner-Halenda) method. Thermogravimetric analysis (TGA)
was performed on NETZSCH TG209 thermogravimetric analyzer. Samples were heated to 800oC at a heating rate of 10oC / min in air atmosphere. FT-IR spectra (transmission) were performed on a Nicolet 550 spectrometer. IBN4 and surface modified IBN4 samples were mixed with KBr and made into pellets for measurement. The diffuse reflectance and solution UV-Vis spectra were measured with a Hitachi U-3010 spectrophotometer. When dealing with solid samples, samples were loaded in a quartz cell and an integrating sphere was included to collect the reflected light. The spectra were collected in 300-700 nm wavelength range against a standard. EPR spectra were recorded on an X-band EPR spectrometer (Brüker EMX). A quartz tubing of 4 mm OD was used to measure the EPR spectra of solids. The spectrometer was equipped with a variable-temperature setup, which allowed us to perform low-temperature experiments (8 K) of solids. Typical spectrometer settings were: microwave frequency, 9.546 GHz; microwave power, 20 mW; center field, 2600 G; sweep width, 5000 G; time constant, 40.96 ms; modulation amplitude, 5 G. The low temperature EPR experiments were performed at Prof. SC Ke’s laboratory, National Dong-Hwa University, Taiwan. 2.5. Peroxidase activity assay of free and immobilized cyt c The peroxidase activity of cyt c was measured by a standard colorimetric method.23 It involves the coupling reaction between quinone and 4-aminoantipyrine to form the colorful quinoneimine. First, phenol is oxidized to quinone by the catalysis of H2O2 and cyt c. Second, the production of quinone can further react with 4-aminoantipyrine through nucleophilic addition. The reaction condition contained 1 mL of phenol (30 mM), 1 mL of 4-aminoantipyrine (7.5 mM), and 0.4 mL of hydrogen peroxide (21.625 mM). All reagents were dissolved in 0.01 M sodium phosphate buffer solution (pH 7.4). After incubation in the water bath for 3 min (30oC to 80oC), 1 mL of free cyt c solution (2 μM) or 1 mL of the immobilized-cyt c sample (1 mg / mL) was
added to start the reaction. The reaction was taken for 5 min. The absorption of the colored product was monitored at 507 nm (ε = 7100 M-1 cm-1) and it was proportional to the concentration of the product according to Beer-Lambert Law. One unit of activity was defined as the amount of product that will be produced by 1 μmol of cyt c per min under the reaction. All the activity measurements were performed three times. 2.6. Hydrothermal stability Hydrothermal stability was determined by treating 1 mg / mL of IBN4-cyt c samples under 0.01M sodium phosphate buffer at 80oC for 12 h. After hydrothermal treatment, immobilized cyt c was collected by centrifugation and dried in vacuum for 3 h. The diffuse reflectance UV-Vis absorption spectra of the immobilized cyt c were measured to determine the differences before and after hydrothermal treatment. In addition, the residual of the catalytic activity after hydrothermal treatment were measured by the same protocol in section 2.5. 2.7. Catalytic oxidation of pyrene It was reported that cyt c / H2O2 system can catalyze the oxidation of pyrene to form 1,8-pyrenodione.10 The PAHs assay was performed to determine the enzyme activity of immobilized cyt c. Typically, 5 mg of IBN4-cyt c solids were added to 5 ml of 20 μM pyrene solution (20% of CH3CN and 80% of 0.01 M NaH2PO4-Na2HPO4 buffer (pH 6.1)). After stirring the mixture for 5 min at 30oC, 25 μL of 200 mM H2O2 was added. The reaction was taken at 500 rpm for 5 min. After the reaction was finished, the solids were separated from the reaction mixture by centrifugation. The reaction extent was determined by the intensity decrease of the absorption peak at 335 nm in the solution phase. All the measurements were performed three times. Specific activity was determined by measuring the moles of substrate oxidized per mole of cyt c per min.
2.8. Regional surface charge calculation Regional surface charge of cyt c at pH 6.0 was calculated by the method previously reported by Lei et al.18 Cyt c crystal structure (Protein Data Bank ID: 1YCC) 24, 25 was defined in a cube and the six faces of the surface charge was calculated by a online software at EMBL WWW Gateway to Isoelectric Point Service.26 The most probable orientation of cyt c attached to the IBN4-N-GAC sample was determined by the most electrically positive face of the cyt c based on electrostatic interaction.
3. Results and discussion 3.1. Structural characterization of the mesoporous materials IBN4 and its functionalized form TEM images of the IBN4 materials are shown in Fig. 1. The well-ordered macrostructures of the materials was observed. The morphology showed short rods with a width near 80 nm and length between 200 to 500 nm. No aggregated form was observed so that they can well suspend in solution. XRD patterns and N2 adsorption-desorption isotherms of IBN4 and modified IBN4 samples are displayed in Fig. 2. It was reported that the mesostructure of IBN4 is 2-D hexagonal (p6m).21 The observed strong peak at 2θ = 0.9 corresponds to the (100) diffraction peak of hexagonal lattice symmetry. All other modified IBN4 has similar XRD diffraction patterns indicating intact structure of the surface functionalized samples. The thermogravimetric analysis (TGA) of IBN4 and surface modified IBN4 (shown in supporting materials) gave decomposition temperatures and weight loss. All the surface functionalized groups (GAC, SMP and GAH) were stable up to 300oC. The loadings were between 10 to 13% by weight. For example, for the IBN4-N-GAC sample, this is equivalent to 1.46 GAC / nm2 which is not far less than the
monolayer coverage. The FT-IR spectra of IBN4 and surface modified IBN4 are shown in Fig. 3. The topest curve (Fig. 3a) is for the bare IBN4 itself showing only the surface silanol group and the low frequency silica vibrations. The C-H stretches at 2875 and 2935 cm-1 were found in all surface functionalized samples (Fig. 3b - 3e). The amine-modified IBN4 sample displays a weak N-H bending mode (primary amine) near 1553 cm-1 (Fig. 3b). Further derivation of amino groups to GAC (Fig. 3c), SMP (Fig. 3d), and GAH (Fig. 3e) groups showed the different C=O stretch modes from carboxylic acid, amide, and aldehyde groups at near 1708 cm-1. In addition, the secondary amide of GAC and SMP samples showed relatively strong N-H bending bands at about 1553 cm-1 which attributed to a combination of C-N stretching band and an N-H bending band. Other bands at 1415 cm-1 attributed to the CH2 bends. 3.2. Immobilization of cyt c The nitrogen adsorption studies allow us to evaluate the pore size, pore volume and surface area of the IBN4 with various bioconjugates used in the experiments (see Table 1). We note that the pore size is reduced from 7.1 nm of the as-synthesized IBN4 to 5.3 nm in IBN4-N-GAH. Note that the bulk dimension of cyt c is (2.5 x 2.5 x 3.7 nm), thus the pore size of modified IBN4 samples still have more than enough room to accommodate cyt c molecules. In addition, in order to avoid the different diffusion rates due to pore size effect the three functionalized IBN4 samples have almost the same pore size (Table 1) so that the catalytic activities of the immobilized cyt c could be properly compared. However, the surface area is drastically decreased upon functionalization, reduced from 847 m2/g in the as-synthesized IBN4 to 361 m2/g in IBN4-N and finally to 249 m2/g in IBN4-N-GAC. Some of the nanopores may have been blocked after functionalization. The loadings of cyt c in various modified IBN4 are shown in Table 2. Immobilization efficiency is defined as the percentage of loaded enzyme relative to that
in the initial solution phase. Due to the isoelectric point of cyt c is about 10.0, it’s difficult to immobilized cyt c into the IBN4-N sample even at pH 9.0. The order of the three methods of immobilization efficiency is GAC > SMP > GAH. GAC is most efficient because of the electrostatic attraction between the negatively charged carboxylate and the positive charges of the cyt c molecule (amino groups from Lys residues). GAH is least efficient because the reversible reaction for the formation and hydrolysis of the imine bond. Maximum loading was established by decreasing the amount of mesoporous silica materials relative to the enzyme used and the results were summarized in Table 2. We note that we can achieve a very high loading of 8.32 μ mol/g ( 105 mg cyt c per gram of IBN4-N-GAC) at pH 6.0. However, the lower loading materials in the middle column were used in the catalysis and structure study in this work. The loading kinetics were also measured and presented in supporting materials. All three methods gave fast loading; finished within 30 min. To make sure the loadings were by covalent bonding instead of ionic interactions; the IBN4-cyt c samples were washed and sonicated in 1 M KCl for 5 min twice. For the IBN4-N-GAC-cyt c sample made with EDC added, no leached cyt c was detected in the solution. However, when EDC were not used in the linking process, 81% of cyt c was washed out by KCl solution. For the IBN4-N-SMP-cyt c sample, we loaded two different cyt c: the horse heart cyt c and yeast cyt c. The difference is that horse heart cyt c contains no cysteine group and could not be linked to SMP. We found 53% of horse heart cyt c was washed out while only 3.6% of yeast cyt c was washed out by 1M KCl. Thus, the covalent bonding nature of the enzyme loadings was ensured. 3.3. UV-Vis spectroscopic characterization The heme center of native cyt c is a low-spin Fe-(III) ion and the coordination of the axial ligands is methionine 80 (Met-80) and histidine 18 (His-18). The UV-Vis spectra of native cyt c
showed Soret band absorptions at 409 and 532 nm are attributed to the absorptions of the porphyrin chromophore. Q-band absorptions at 502 and 550 nm are from high spin and low spin heme, respectively. Fig. 4 shows the UV-Vis spectra of the free enzyme cyt c and enzyme-loaded IBN4 samples. We also show the spectra after the samples were hydrothermally treated for 12 h. Fig. 4a shows that the Soret band is rather sharp for native enzyme (at pH 7.4). The sharp small peak at 550 nm of Q band shows that native enzyme is in low spin state. After hydrothermally treated at 80oC for 12 h, the native cyt c completely denatured to give a featureless UV-Vis spectrum. The UV-Vis spectra of three confined cyt c samples all show some degree of changes in the Soret and Q band. There are some shifts in the peak positions of Soret band of the immobilized enzyme relative to the native enzyme (409 nm). The blue shifts (403 nm and 406 nm) can be attributed to the de-binding of the Met-80 to iron center.27 Notably, a peak at 620 nm due to the in-plane charge transfer band between the porphyrin and heme iron of the high spin6, 22 appears in the IBN4-N-GAC-cyt c and IBN4-N-SMP-cyt c samples. Moreover, the three immobilized samples show different hydrothermal stability as shown in the UV-Vis spectra. IBN4-N-GAC-cyt c is the most stable one while IBN4-N-GAH-cyt c seems denatured very much after 12 h of hydrothermal treatment. We can infer the different stability of the two samples from the higher hydrolysis reactivity of imine bond than the amide bond. Thus, the IBN4-N-GAH-cyt c sample may undergo hydrolysis and cyt c molecules are leaching from IBN4-N-GAH solids. 3.4. Low temperature EPR studies The covalent bonding of cyt c to the surface modified IBN4 materials could affect the spin state of heme group. In order to establish the spin state of heme Fe (III), we performed EPR experiments at 8 K. The EPR spectra are displayed in Fig. 5. As pointed out in our previous studies,6, 7 cyt c loaded in mesoporous silica exist mostly in high spin Fe (III). High spin Fe (III)
is formed when Met-80 is replaced by H2O or silanol group on the silica surface during cyt c and mesoporous silica ion-exchange process. We observed a signal at g = 6.01 in IBN4-N-GAC-cyt c (Fig. 5a) and in IBN4-N-SMP-cyt c (Fig. 5b), which is assigned to high spin Fe (III). Vazquez-Duhalt et al. chemically modified cyt c with poly-(ethylene glycol) to give high spin heme Fe (III) in six coordination configuration.11, 28 They also reported that such high spin species could improve both the activity and stability. Hirobe et al. immobilized cyt c with poly-γ-methyl- L-glutamate and found cyt c with high spin Fe (III) exhibits high stability in organic solvent.29 Deere et al. immobilized cyt c in MCM-41 and detected both low spin and high spin Fe (III) in their Raman studies.30 We considered the following signals: g = 6.01(high spin) and 2.00 (high spin and protein radical), g = 2.30 (low spin), and g = 4.26 (non-heme Fe (III)). High spin Fe (III) center is formed when Met-80 is detached from Fe (III) and replaced by H2O. The enzyme activity is enhanced when it is in high spin configuration which is attributed to a fully open heme groove and accessible active center. We did not observe any low spin state signal near g = 2.30 implying almost all the immobilized cyt c molecules are in high spin state. The appearance of the signal at g = 4.26 due to free Fe (III) ions may be a result from the abstraction of Fe from heme ring due to competitive binding with amino group on the solid surface. If the relative loading amount of cyt c in different surface modified samples are taken consideration (see Table 2), we note that the relative signal intensity at g = 6.01 of GAC-cyt c and SMP-cyt c are comparable and stronger than that of GAH-cyt c which shows almost zero intensity. The relative intensity strength of the high spin state is correlated closely to the catalytic activity to be discussed in the next section. The sharp strong signal observed at g = 2.00 in IBN4-N-GAH-cyt c (Fig. 5c) can be ascribed to protein radicals formed by transferring one electron to some amino acid (most likely
tryptophan) on cyt c.31 However, the origin and function of this protein radical are not clear. It needs further detailed study in the future. Nonetheless, the presence of free radicals in this material may react with the immobilized cyt c to destroy it and lower the catalytic activity of IBN4-N-GAH-cyt c. 3.5. Reactivity and stability To measure the peroxidase activity, we carried out the aqueous reaction between phenol and 4-aminoantipyrine to form quinoneimine. Fig. 6 gives peroxerdase activity of immobilized cyt c before and after hydrothermal treatment. For the intrinsic activities (before hydrothermal) of the immobilized enzyme, IBN4-N-SMP-cyt c gives the highest activity, the catalytic behavior is very much similar to the homogenous catalysis of free cyt c solution. IBN4-N-GAH-cyt c shows moderate activity while IBN4-N-GAC-cyt c gives the lowest intrinsic activity. However, after hydrothermal treatment the situation is reversed. IBN4-N-GAC-cyt c keeps about the same activity as before the treatment. IBN4-N-GAH-cyt c loses almost completely its catalytic activity. IBN4-N-SMP-cyt c also loses greatly its activity. The residual catalytic activities have the same trend as the hydrothermal stability as inferred from the study of UV-Vis spectra (Fig. 4). The reactive imine bond of the IBN4-N-GAH-cyt c sample may undergo a fast hydrolysis at high temperature in the hydrothermal process and therefore the lowest residual activity was observed. However, the better protective effects of the IBN4-N-GAC-cyt c (high Soret band intensity) lead to a higher residual activity of the IBN4-N-GAC-cyt c sample. In order to study the catalytic behavior of the immobilized cyt c, we also measured the peroxidase activity as a function of temperature (Fig. 7). Since the reaction is only taken five minutes, cyt c also showed the catalytic activity before it denatured at high temperature. IBN4-N-SMP-cyt c gives about the same catalytic curve as free cyt c throughout the whole range
of temperature from 30 to 80oC. IBN4-N-GAC-cyt c starts lowest at room temperature but gradually increases its activity as temperature is raised until finally at 80oC where it shows the highest activity. IBN4-N-GAH-cyt c on the other hand gives a decreasing activity as temperature is increased which should be attributed to the existence of protein radical observed in EPR spectrum (Fig. 5c) that can destroy cyt c rapidly at higher temperature. In, addition, we speculate that the catalytic center of cyt c in the IBN4-N-SMP sample has less steric hindrance and the EPR spectrum showed high spin state which the heme groove is fully open (Fig. 5b). Therefore, the active center is easily accessible to the substrate molecule so that raising temperature showed only slight contribution to the diffuse rate of substrate and a small increase in the specific activity. This behavior of the catalytic curve is similar to the fast diffusion of the reactant in the homogeneous catalysis. In contrast with IBN4-N-SMP-cyt c, IBN4-N-GAC-cyt c only shows high activity at high temperature even though the GAC-cyt c sample also showed a high spin state (Fig 5a). Thus, other factor may strongly affect the catalytic behavior of the immobilized cyt c. The reason for these trends will be further discussed in the next section of the structural analysis related to the orientation of cyt c in the nanochannel of mesoporous silica IBN4. 3.6. Spatial orientation effect In order to easily examine the relationship between the activity, structure and orientation of cyt c in the surface modified IBN4 materials, we express the crystal structure of cyt c (PDB ID: 1YCC)24, 25 in an imaginary cubic box as shown in Fig. 8. The electric charge distribution of cyt c in the cubic box at pH = 6.0 is tabulated in Table 3. Note that each plane surface of the cubic box carries different surface net charge corresponding to the aggregation of different charged amino acid residues in that region. Different linker therefore exposes the catalytic active sites (Fe-heme) in different environment which allows us to examine the orientation effect imposed
by the binding mode. When cyt c is immobilized in the IBN4-N-SMP sample, there is only one mercapto group from Cys-102 (Cys-14 and Cys-17 formed disulfide bond with heme group) would form C-S bond with surface functionalized SMP group. Since Cys-102 is on the C-terminal helix, it will be located just on the face of EFGH as shown in Fig. 8. Thus, if cyt c was immobilized in this material, the active site will be pointed away from the silica wall and fully opened to the substrate, i.e. less steric hindrance and the heme groove is fully open. On the other hand, when cyt c is immobilized in the IBN4-N-GAC sample, electrostatic interaction is the dominant factor to determine the orientation of cyt c. The positively charged face of cyt c would be attracted toward the negatively charged carboxylate in the IBN4-N-GAC sample. As shown in Table 3, the face of ABCD gives the most positive charge which is most likely to attach and form amide bond with surface functionalized GAC group. Thus, the Fe-porphyrin active site is in proximity to the silica wall and the entrance for the substrate-docking pocket may be partially blocked. For the IBN4-N-GAH sample, the glutaraldehyde group does not have strong electrostatic interaction with amine group, so the attachment site would be random. In Scheme 2, we display the orientation of the active site and specific binding of cyt c on various surface functionalized IBN4 samples. One can see clearly that the accessibility of the Fe active center in immobilized cyt c would be in the order of IBN4-N-SMP-cyt c > IBN4-N-GAH-cyt c > IBN4-N-GAH-cyt c which is the same order of the initial specific activities shown in Fig. 6. Therefore, the structural orientation analysis given here does correlate well with the catalytic activity. On the other hand, the peroxidase activity of immobilized cyt c after hydrothermal treatment is in different order. From UV-Vis spectra, one observed that GAC-linked cyt c possess the highest stability (Fig. 4b). Also in EPR spectra, IBN4-N-GAC-cyt c has a higher
ratio of high spin (g = 6.01) component to free Fe (g = 4.26) (Fig. 5a). Thus we expect IBN4-N-GAC-cyt c to give the highest residual activity after hydrothermal treatment. This correlated to the orientation of the cyt c which the active site is in proximity of the silica wall and thus well-protected the enzyme. In contrast, the lost of activity in IBN4-N-SMP-cyt c may be due to the less protection of the active center of Fe-heme after prolonged hydrothermal treatment. On the other hand, the IBN4-N-GAH-cyt c sample lost most of its activity after hydrothermal treatment because of extensive radical attack by the radicals (g = 2.00) (Fig. 5c) to destroy the protein structure. Finally, we should comment on the temperature trend in Fig. 7. These confined enzymes, except for GAH-cyt c, show rising activities against rising temperature and then decay after reaching the maximum. This kind of trend is typical of immobilized enzyme where the pore confinement protects the enzyme against denaturation. The behavior of IBN4-N-SMP-cyt c sample is almost the same as that of free enzyme. This may be due to the active center of IBN4-N-SMP-cyt c is least perturbed from the immobilization as it is hanged freely by the tail of Cys-102 group. However, cyt c in the IBN4-N-GAC sample would partially block and somehow protect the active site so the activity in high temperature could be raised and showed highest stability. 3.7. Oxidation of Pyrene Polycyclic aromatic hydrocarbons (PAHs) are often regarded as potential health hazardous materials arising from their toxicity, carcinogenicity, mutagenicity, resistance to biodegradation, and because they are ubiquitous.32 The transformation of PAHs to innocuous compounds is an important water treatment problem, and has been a subject of extensive research.33 They are recalcitrant towards simple oxidation treatment. However, these toxic PAHs have been
biodegraded by enzymes releasing microorganism.34 Here, we choose pyrene as a test case of PAH in using our IBN4-supported cyt c as the catalyst. As revealed in the preceding studies of design principles of immobilization for the peroxidase activities of supported cyt c, we applied the materials to the oxidation of pyrene as an example for PAH oxidation. Because we have shown that GAH-linked cyt c produces copious amount of protein radical which tends to degrade cyt c itself, we will not apply IBN4-N-GAH-cyt c in the case of pyrene. Fig. 9 shows the peroxerdase activity of immobilized cyt c towards pyrene before and after hydrothermal treatment. First, we notice that the initial reaction rates follow the order of IBN4-N-SMP-cyt c > IBN4-N-GAC-cyt c ~ free cyt c. Again, the sample IBN4-N-SMP-cyt c is most active because of the open active site in the nanochannels. We also found the catalyst IBN4-N-GAC-cyt c is most stable towards hydrothermal treatment because of the well protected active center. Surprisingly, IBN4-N-SMP-cyt c does not lost much in activity after hydrothermal treatment.
4. Conclusions We demonstrated that proper surface modification with three different linkers: GAC, SMP and GAH yields orientation effect that affects the peroxidase-like catalytic activity of immobilized cyt c in IBN4 mesoporous silicas. The measured relative peroxidase activities of these three immobilized cyt c are as follows: IBN4-N-SMP-cyt c > IBN4-N-GAH-cyt c > IBN-4-N-GAC-cyt c. Molecular modeling further provide insight into the relationship between peroxidase activity of immobilized cyt c and structural orientation due to charge distribution in the nanochannels: (1) in the GAC-cyt c sample, the active site is in the proximity of silica wall,
(2) in SMP-cyt c, the active site is pointed away from silica wall which would leave room for substrate to diffuse into the nanochannel and react with the active site of cyt c in the void space (middle) of channels, and (3) in GAH-cyt c, the cyt c has random orientation. In the hydrothermal treatment studies, there are much more structure decay after 12 h at 80oC treatment and the activities diminish accordingly: the IBN4-N-GAH-cyt c sample lost most of its activity, the IBN4-N-SMP-cyt c lost its activity due to the less protection of the active center of Fe-heme, and the IBN4-N-GAC-cyt c retained good activity. Thus, we show that we can chemically select and design a pore surface to achieve an optimal environment (site selection) to attain high reactivity and preserve the integrity of protein structures. The designed catalysts have been applied to the oxidation of pyrene. The activity trends do correlate well with our detailed study of orientation as structural integrity of confined cyt c in mesoporous silica.
Acknowledgements: We wish to thank Prof. SC Ke of National Dong-Hwa University for assisting the EPR measurements at low temperatures.
The research was supported by a grant
from the National Science Council of Taiwan through her Nanoscience program.
Scheme 1 Immobilization of cytochrome c in surface modified IBN4 mesoporous silica materials. O
O
O O
glutaric anhydride
O
O
N O
O O O O Si NH2 succinimido 3-maleimido propanoate O IBN4-N
O
O
O O-
N H IBN4-N-GAC
Cyt c-NH2
O O Si O
Cyt c-SH
O O Si O
O
N H
N H
Cyt c
O
O
N
O O Si O
O glutaraldehyde
O O Si O
O
N H
N
IBN4-N-SMP
O
O
O S
N H
N
Cyt c
O
O O O Si O
N
H
Cyt c-NH2
O O Si O
N
N
Cyt c
IBN4-N-GAH
Scheme 2 Orientation and specific binding of cyt c in (A) IBN4-N-GAC: active site (heme group) is in the proximity of silica wall, (B) IBN4-N-SMP: active site is pointed away from silica wall and (C) IBN4-N-GAH: orientation of active site is random.
Table 1 Pore size, pore volume and surface area of IBN4 and surface modified IBN4.
Table 2 Immobilization conditions of cyt c in surface modified IBN4 via covalent bonding.
Table 3 Net charges of the surface regions of cyt c calculated from the six faces of Fig. 8 at pH 6.0.
Fig. 1 TEM images of IBN4 materials.
Fig. 2 (A) Powder X-ray diffraction patterns and (B) N2 adsorption-desorption isotherms: (a) IBN4, (b) IBN4-N, (c) IBN4-N-GAC, (d) IBN4-N-SMP, and (e) IBN4-N-GAH.
(A)
(B)
Intensity
3
Volume adsorbed (cm /g)
(a)
(a)
(b)
(c) (d) (e)
(b) (c) (d) (e)
1
2
3
2 theta
4
0.0
0.2
0.4
0.6
P/P0
0.8
1.0
Fig. 3 FTIR spectra of (a) IBN4, (b) IBN4-N, (c) IBN4-N-GAC, (d) IBN4-N-SMP, and (e) IBN4-N-GAH.
Transmittance (% )
(a)
(b) (c) (d)
(e) 2875 1708 1710
2935
1415
1553
4000
3500
3000
2500
2000
1500 -1
Wavenumber (cm )
1000
500
Fig. 4 UV-Vis absorption spectra of (a) free cyt c solution and diffuse reflectance UV-Vis absorption spectra of (b) IBN4-N-GAC-cyt c, (c) IBN4-N-SMP-cyt c, and (d) IBN4-N-GAH-cyt c before (solid) and after (dot) hydrothermal at 80oC for 12 h. The solid state UV-Vis spectra of (b), (c) and (d) were collected using the IBN4-N-GAC, SMP, and GAH materials as reference, respectively. 0.8
0.35
(a)
409
(b)
403
0.7
0.30
0.6
Intensity (a.u.)
Intensity (a.u.)
0.25 0.5 0.4 0.3 0.2
0.20 0.15 532
0.10 523 550
0.1
620
0.05
0.0 300
400
500
600
0.00 300
700
Wavelength (nm)
400
500
600
Wavelength (nm) 0.25
0.30
(d)
(c) 406
406
0.20
0.20
Intensity (a.u.)
Intensity (a.u.)
0.25
0.15
0.10
0.15
0.10 550
532
0.05
0.05
0.00 300
700
622
400
500
600
Wavelength (nm)
700
0.00 300
400
500
600
Wavelength (nm)
700
Fig. 5 EPR spectra of immobilized cyt c at 8 K: (a) IBN4-N-GAC-cyt c, (b) IBN4-N-SMP-cyt c, and (c) IBN4-N-GAH-cyt c.
g=6.01
g=4.26
g=2.00
Intensity (a.u.)
(a)
(b)
(c)
500
1000
1500
2000
2500
3000
3500
4000
Gauss
Fig. 6 Relative peroxidase activity of immobilized cyt c on functionalized IBN4 surface for oxidation of 4-aminoantipyrine before and after hydrothermal at 80oC for 12 h.
Specific activity (mol/molmin)
14 12 10 8 86%
6 4
27%
2
11% 3%
0 free cyt c
IBN4-GAC-cyt c IBN4-SMP-cyt c IBN4-GAH-cyt c
Fig. 7 Peroxerdase activity of immobilized cyt c for oxidation of 4-aminoantipyrine as a function of temperature.
free cyt c
Specific activity (mol/molmin)
24
IBN4-GAC-cyt c IBN4-SMP-cyt c IBN4-GAH-cyt c
20 16 12 8 4 0 30
35
40
45
50
55
60
65
70
75
80
o
Temperature ( C)
Fig. 8 Crystal structure of cyt c (Protein data bank ID: 1YCC) in an imaginary cubic box. The arrow indicates the location of Cys-102.
Fig. 9 Relative peroxidase activity of immobilized cyt c on functionalized IBN4 surface for oxidation of pyrene before and after hydrothermal at 80oC for 12 h.
Specific activity (mol/molmin)
0.5
97%
0.4 69%
0.3
52%
0.2
0.1
0.0 free cyt c
IBN4-GAC-cyt c
IBN4-SMP-cyt c
References 1 2 3 4
5
6
7 8 9 10 11
12
13 14 15 16 17 18 19 20
21 22 23 24 25 26
27 28
29 30 31 32
33
Q. Husain, M. Husain, and Y. Kulshrestha, Critical Rev. Biotech., 2009, 29, 94. M. Chivukula, J. T. Spadaro and V. Renganathan, Biochemistry, 1995, 34, 7765. J. T. Spadaro and V. Renganathan, Archives of Biochemistry and Biophysics, 1994, 312, 301. M. Wikström, 'Biophysical and Structural Aspects of Bioenergetics', RSC Publishing, Cambridge, 2005. L. C. Godoy, C. Muńoz-Pinedo, L. Castro, S. Cardaci, C. M. Schonhoff, M. King, V. Tortora, M. Marin, J. F. Jiang, A. Kapralov, and R. Jemmerson, Proc. Natl. Acad. Sci. USA, 2009, 106, 2653. C.-H. Lee, J. Lang, C.-W. Yen, P.-C. Shih, T.-S. Lin, and C.-Y. Mou, J. Phys. Chem. B., 2005, 109, 12277. C.-H. Lee, C.-Y. Mou, S.-C. Ke, and T.-S. Lin, Mol. Phys., 2006, 104, 1635. A. Tiehm, Appl. Environ. Microbiol. , 1994, 60, 258. A. P. Borole, C. L. Cheng, and B. H. Davison, Biotechnol. Prog. , 2004, 20, 1251. R. Vazquez-Duhalt, J. Mol. Catal. B: Enzym., 1999, 7, 241. H. Garcia-Arellano, B. Valderrama, G. Saab-Rincon, and R. Vazquez-Duhalt, Bioconjugate Chem., 2002, 13, 1336. E. Torres, J. V. Sandoval, F. I. Rosell, A. G. Mauk, and R. Vazquez-Duhalt, Enzyme Microb. Tech., 1995, 17, 1014. C.-H. Lee, T.-S. Lin, and C.-Y. Mou, NanoToday, 2009, 4, 165. S. Hudson, J. Cooney, B. K. Hodnett and E. Magner, Chemistry of Materials, 2007, 19, 2049. R. Ravindra, S. Zhao, H. Gies, and R. Winter, J. Am. Chem. Soc., 2004, 126, 12224. J. Kim, J. W. Grate and P. Wang, Chemical Engineering Science, 2006, 61, 1017. T. E. Reeves, S. Paliwal, M. E. Wales, J. R. Wild, and A. L. Simonian, Langmuir, 2009, 25, 9615. C. Lei, T. A. Soares, Y. Shin, J. Liu, and E. J. Ackerman, Nanotechnology, 2008, 19, 125102. K. P. Fears, B. Sivaraman, G. L. Powell, Y. Wu, and R. A. Latour, Langmuir, 2009, 25, 9319. M. E. Aubin-Tam, W. Hwang, and K. Hamad-Schifferli, Proc. Natl. Acad. Sci. USA, 2009, 106, 4095. Y. Han and J. Y. Ying, Angew. Chem. Int. Ed., 2005, 44, 288. S. Oellerich, H. Wackerbarth, and P. Hildebrandt, J. Phys. Chem. B, 2002, 106, 6566. R. Radi, L. Thomson, H. Rubbo, and E. Prodanov, Arch. Biochem. Biophys. , 1991, 288, 112. G. V. Louie, W. L. B. Hutcheon, and G. D. Brayer, J. Mol. Biol., 1988, 199, 205. G. V. Louie and G. D. Brayer, J. Mol. Biol., 1990, 214, 527. http://www.embl-heidelberg.de/cgi/pi-wrapper.pl, in 'EMBL WWW Gateway to Isoelectric Point Service'. F. I. Rosell, J. C. Ferrer and A. G. Mauk, J. Am. Chem. Soc., 1998, 120, 11234. E. Busi, B. D. Howes, R. Pogni, R. Basosi, R. Tinoco, and R. Vazquez-Duhalt, J. Mol. Catal. B: Enzym., 2000, 9, 39. R. Akasaka, T. Mashino, and M. Hirobe, J. Chem. Soc., Perkin Trans. 1, 1994, 1, 1817. J. Deere, E. Magner, J. G. Wall, and B. K. Hodnett, J. Phys. Chem. B. , 2002, 106, 7340. J. Stubbe and W. A. van der Donk, Chemical Reviews, 1998, 98, 705-762. A. Luch, 'The Carcinogenic Effects of Polycyclic Aromatic Hydrocarbons', Imperial College Press, London, 2005. E. Torres, R. Tinoco, and R. Vazquez-Duhalt, Wat. Sci. Tech. , 1997, 36, 37.
34
N. P. Cheremisinoff, 'Biotechnology for Waste and Wastewater Treatment', Noyes Publication,1996.
