Adsorption of Myoglobin onto Porous Zirconium Phosphate and

Porous zirconium phosphate (P-ZrP) and zirconium benzenephosphonate (P-ZrBP) were prepared in the presence of an anionic surfactant acting as a templa...
4 downloads 0 Views 119KB Size
Download PDF
5064

Langmuir 2006, 22, 5064-5069

Adsorption of Myoglobin onto Porous Zirconium Phosphate and Zirconium Benzenephosphonate Obtained with Template Synthesis Francesca Bellezza, Antonio Cipiciani,* Umberto Costantino, and Fabio Marmottini Dipartimento di Chimica and Centro di Eccellenza Materiali InnoVatiVi Nanostrutturati (CEMIN), UniVersita` di Perugia, Via Elce di Sotto 8, 06123 Perugia, Italy ReceiVed NoVember 25, 2005. In Final Form: March 16, 2006 Porous zirconium phosphate (P-ZrP) and zirconium benzenephosphonate (P-ZrBP) were prepared in the presence of an anionic surfactant acting as a template. Poorly crystalline materials with a P/Zr molar ratio equal to 2 and having a relatively high surface area and micro/mesoporosity have been obtained. The interaction of myoglobin with the two types of surfaces, the hydrophobic P-ZrBP and the hydrophilic P-ZrP, was investigated, and the adsorption isotherms were determined at different pH and temperature values. A model was proposed for the mechanism of the interaction of the protein with the surface based on the shape of the adsorption isotherm and the physical-chemical properties of myoglobin. The pH has been found to be an important parameter for determining the maximum adsorption capacity of P-ZrBP and P-ZrP for myoglobin molecules because of the changes that occur in the type and net charge of the protein surface as the pH of the medium changes. Protein binding affinity and capacity increase when the temperature is increased. This phenomenon occurs because myoglobin varies its conformation at high temperature with an increase in the exposed hydrophobic region. This process causes a stronger hydrophobic interaction between the protein and the adsorbent and reduces the repulsion between the adsorbed molecules. Studies on the activities of the obtained biocomposites are in progress.

1. Introduction Enzymatic biotransformations have been extensively investigated for a wide range of important chemical processes as a result of their high selectivity and mild reaction conditions.1 However, low catalytic efficiency and stability of enzymes have been seen as deterrents for the development of large-scale operations that could compete with traditional chemical processes. Among the most investigated strategies for improving the functionality and performance of enzymes, immobilization offers easy catalyst recycling, feasible continuous operations, and simple product purifications.2 Recently, mesoporous silica materials were shown to be effective in bioimmobilization when compared with conventional materials because of their uniform pore size and large surface area as well as mechanical and chemical resistance.3-5 In recent years, surfactants have been used as powerful tools in the synthesis of nanostructured or mesostructured materials.6,7 The self-assembly of surfactants to create hydrophobic and hydrophilic structured domains has been exploited for molecular “imprinting” or “templating” to control pore size and shape.8 Since the discovery of surfactant-assisted synthesis of mesoporous silica,9 there has been great research interest in supramolecular templating of porous materials. The main areas of investigation have been the tailoring of pore diameter and size distribution as well as the possible application in catalysis. * Corresponding author. E-mail: [email protected]. Tel: 075/585540. Fax: 075/5855560. (1) Hult, K.; Berglund, P. Curr. Opin. Biotechnol. 2003, 14, 395. (2) Lalonde, J.; Margolin, A. Enzyme Catal. Org. Synth. 2002, 1, 163. (3) Deere, J.; Magner, E.; Wall, J. G.; Hodnett, B. K. J. Phys. Chem. B 2002, 106, 7340. (4) Goradia, D.; Cooney, J.; Hodnett B. K.; Magner, E. J. Mol. Catal. B: Enzym. 2005, 32, 231. (5) Wang, Y.; Caruso, F. Chem. Mater. 2005, 17, 953. (6) Schuth, F. Chem. Mater. 2001, 13, 3184. (7) Antonietti, M. Curr. Opin. Colloid Interface Sci. 2001, 6, 244. (8) Patarin, J.; Lebeau, B.; Zana, R. Curr. Opin. Colloid Interface Sci. 2002, 7, 107. (9) Kresge, C. T.; Leonowicz, M. E.; Roth, W. J.; Vartuli, J. C.; Beck, J. S. Nature 1992, 359, 710.

To create more suitable biocatalysts for enzymatic transformations, it is necessary to understand the factors that influence the immobilization behavior of protein within porous materials. Two factors may greatly influence the immobilization properties: the size of the mesopore and the surface characteristics. In general, proteins with a hydrodynamic diameter larger than the pore diameter adsorb on the external surface of the support, whereas small proteins interact with mesopores and show high loading. Enzyme leaching is one of the problems that has been reported in the use of porous materials as supports for enzyme immobilization. An approach to strengthen the interactions of the enzyme with the pore walls is the organic functionalization on the internal surfaces.10 The functional groups required for such a modification depend on the structure of the enzyme. An interesting alternative is to use materials that have suitable active groups anchored to the surface and to plan their synthesis with the aim of obtaining a larger surface area and greater porosity. In a series of papers, we showed that zirconium phosphates and phosphonates are suitable supports for the immobilization of proteins and enzymes because they join good chemical and mechanical properties with the possibility of functionalizing the layers with a large variety of organic groups.11-14 However, these phosphonates have a relatively small surface area and are not considered to be porous. A method for preparing materials with larger surface areas and variable mesoporosities could be of interest. Porous zirconium phosphates have been prepared by using a “surfactant-assisted” approach similar to that used to obtain porous silica.15-17 Furthermore, few examples of hybrid porous metal phosphonates have been reported.18,19 (10) Yiu, H. H. P.; Wright, P. A.; Botting, N. P. J. Mol. Catal. B: Enzym. 2001, 15, 81. (11) Bellezza, F.; Cipiciani, A.; Costantino, U.; Nicolis, S. Langmuir 2004, 20, 5019. (12) Bellezza, F.; Cipiciani, A.; Costantino, U.; Negozio, M. E. Langmuir 2002, 18, 8737. (13) Bellezza, F.; Cipiciani, A.; Costantino, U. J. Mol. Catal. B: Enzym. 2003, 26, 47. (14) Bellezza, F.; Cipiciani, A.; Quotadamo, M. A. Langmuir 2005, 21, 11099.

10.1021/la0531897 CCC: $33.50 © 2006 American Chemical Society Published on Web 04/19/2006

Myoglobin Adsorption onto P-ZrP

In this article, we report a simple method for preparing porous zirconium phosphate (P-ZrP) and benzenephosphonate (P-ZrBP) in the presence of anionic surfactant sodium bis(2-ethylhexyl)sulfosuccinate (AOT) and their use as supports for protein adsorption. These materials offer advantages of porous solids such as large surface areas combined with different specific functions on the surface that are suitable for interacting with biomolecules. As already discussed,12 P-ZrP has hydrophilic character because of the presence of OH groups, whereas P-ZrBP has hydrophobic character because of its phenyl groups on the surface. It seemed of interest to continue our studies on protein immobilization and, in particular, of myoglobin (Mb), focusing on the effects of surface area and porosity on protein loading. Furthermore, parameters such as protein concentration, temperature, and pH of the equilibrium adsorption isotherms have been analyzed in an attempt to investigate electrostatic and hydrophobic interactions of proteins at the support-equilibrating solution interface. Myoglobin is a globular protein with molecular mass of 17 800 Da, an isoelectric point (pI) of 7.0-7.2, and approximate dimensions of 25 × 35 × 45 Å3.20 It is therefore expected that the protein has access to the eventual mesopores of P-ZrP and P-ZrBP. The P-ZrP and P-ZrBP solids obtained in the presence of AOT were characterized in terms of X-ray diffraction patterns, chemical and thermogravimetric analyses, BET analysis of surface areas, and pore size distributions. N2 adsorption/desorption data of materials before and after protein adsorption are used to demonstrate the extent of pore filling with adsorbed biomolecules. 2. Experimental Section 2.1. General. Myoglobin (from horse heart, 90% pure, lot 0447006) was obtained from Sigma. ZrOCl2‚8 H2O was a Merck product, and H3PO4 (85%) was a Carlo Erba reagent. AOT (sodium bis(2-ethylhexyl)sulfosuccinate) and benzenephosphonic acid were obtained from Aldrich. 2.2. Synthesis of Porous Zirconium Phosphate (P-ZrP) and Benzenephosphonate (P-ZrBP). P-ZrP and P-ZrBP were synthesized according to the following procedure. AOT (0.2 g, 0.45 mmol) was mixed with 100 mL of deionized water and stirred until all of the surfactant was dissolved. Zirconyl chloride octahydrate (1.28 mmol) and the acid (phosphoric or benzenephosphonic, respectively) (5.12 mmol) were added to the solution and stirred for 16 h ([P] ) 0.05 M, [Zr] ) 0.0125 M, P/Zr ) 4). The white gel product was centrifuged (8000 rpm, 15 min), and the solid obtained was washed with water (3 × 100 mL) and ethanol (3 × 100 mL) to remove the surfactant. The solid was dried under vacuum at 60 °C. 2.3. Characterization of Porous Zirconium Phosphate (P-ZrP) and Benzenephosphonate (P-ZrBP). X-ray powder diffraction (XRPD) analyses were performed on a computerized Philips PW1710 diffractometer using Cu KR radiation, operating at 40 kV and 20 mA with a step scan of 1° min-1. N2 adsorption/desorption isotherms were measured at 77 K using a computer-controlled Micromeritics ASAP2010 volumetric adsorption analyzer. The surface-area (15) Jimenez-Jimenez, J.; Maireles-Torres, P.; Olivera-Pastor, P.; RodriguezCastellon, E.; Jimenez-Lopez, A.; Jones, D. J.; Roziere, J. AdV. Mater. 1998, 10, 812. (16) Sun, Y.; Afanasiev, P.; Vrinat, M.; Coudurier, G. J. Mater. Chem. 2000, 10, 2320. (17) Dong, A.; Ren, N.; Tang, Y.; Wang, Y.; Zhang, Y.; Hua, W.; Gao, Z. J. Am. Chem. Soc. 2003, 125, 4976. (18) Bhaumik, A.; Inagaki, S. J. Am. Chem. Soc. 2001, 123, 691. (19) Ren, N.; Tang, Y.; Wang, Y. J.; Hu, S. H.; Dong, A. G.; Hua, W. M.; Yue, Y. H.; Shen, J. Y. Chem. Lett. 2002, 1036. (20) Bos, M. A.; Shervani, Z.; Anusiem, A. C. I.; Giesbers, M.; Norde, W.; Kleijn, M. Colloids Surf., B 1994, 3, 91.

Langmuir, Vol. 22, No. 11, 2006 5065

Figure 1. Possible formation mechanism for P-ZrP and P-ZrBP around AOT micelles. measurement is based on the BET method,21 and the pore-size distribution is based on the BJH method.22 t-plot analysis is used to calculate the micropore volume and distribution size (with crystalline R-ZrP taken as a reference material). Thermogravimetric analyses (TGA) were performed with a Netzsch STA 449C thermobalance. Phosphate analysis was performed by ion chromatography using a Dionex 500 instrument with a Dionex AS4A column. The same instrument, with a Dionex 201 TP column, was used to determine the zirconium content following the method described in ref 23 (eluent, acetonitrile/water 55/45 v/v with acetic acid/sodium acetate at pH 4; flux, 1.0 mL/min; detection, spectrophotometric at 585 nm). 2.4. Myoglobin Adsorption. Samples of adsorbed Mb were prepared by mixing 30 mg of support with 3 mL of protein solution at different concentrations. In the case of P-ZrBP, the solid was first wetted with 0.1 mL of ethanol to facilitate contact between the hydrophobic support and the aqueous solution. The mixture was stirred from 0.25 to 16 h, and the solid was recovered by centrifuging at 8000 rpm for 15 min. The supernatant was assayed for protein content by UV spectroscopy (λ ) 410 nm,  ) 160 000 M-1 cm-1). The amount of adsorbed protein was calculated from the difference between the concentrations of the initial solution and the supernatant. The effect of pH on the adsorption process was examined using different buffer solutions (sodium acetate 0.2 M, pH 5.0; sodium dihydrogenphosphate 0.2 M, pH 6.0, 7.0, 8.0).

3. Results and Discussion 3.1. Synthesis and Characterization of Porous Zirconium Phosphate (P-ZrP) and Benzenephosphonate (P-ZrBP). The synthesis of P-ZrP and P-ZrBP in the presence of amphiphilic molecules has been realized with the AOT anionic surfactant acting as a “structure-directing species”. A possible formation mechanism of these porous materials is schematically illustrated in Figure 1. (21) Brunauer, S.; Hemmet, P. H.; Teller, E. J. Am. Chem. Soc. 1938, 60, 309. (22) Barret, E. P.; Joyner, L. G.; Halenda, P. P. J. Am. Chem. Soc. 1951, 730, 373. (23) Oszwaldowski, S.; Lipka, R.; Jarosz, M. Anal. Chim. Acta 1998, 361, 177.

5066 Langmuir, Vol. 22, No. 11, 2006

Bellezza et al.

Figure 3. N2 adsorption-desorption isotherms on P-ZrP (a) and Mb/P-ZrP (b). Figure 2. Thermogravimetric analyses of the ZrP precursor, obtained without ethanol extraction (a), and of P-ZrP (b) and P-ZrBP (c) obtained after ethanol extraction.

At the relatively low concentration of AOT used, the amphiphilic molecules are organized in spherical micelles with a negatively charged surface balanced by positive Na+ counterions. After the addition of ZrOCl2 to the micellar solution, the metal ion precursors interact with the negatively charged headgroups of micelles replacing, at least in part, the initial surfactant counterions and becoming the new counterions for the surfactant micelles. This counterion exchange could be the driving force for the modification of micelles into new surfactant microaggregates that are different in size and/or shape.8 The addition of phosphoric or benzenephosphonic acid involves the precipitation of zirconium phosphate (P-ZrP) and benzenephosphonate (P-ZrBP), respectively, whose lamellar framework surrounds the supramolecular array of self-assembled surfactant. A porous house of cards arrangement may be obtained after the removal of the template. The removal of template molecules from mesostructured solids is the crucial step in the whole process of surfactant templating because it can influence the porosity of the obtained solid.24 The most common methods of surfactant removal are calcination at relatively high temperature in the presence of oxygen and extraction with solvent, and the amounts removed can be evaluated using thermogravimetric analyses. When the surfactant is removed by calcination, a carbonaceous deposit is present on the material surface. Moreover, high calcination temperatures usually lead to decreased surface areas and, in the present case, to the loss of P-OH and P-C6H5 functional groups. To avoid this problem, we decided to use extraction with ethanol to remove the surfactant template (see experimental), and the efficiency of the procedure has been proved by TGA. Thermogravimetric analysis of the ZrP precursor obtained after precipitation and drying, avoiding extraction with ethanol, is reported in Figure 2, curve a, whereas curves b and c refer to TGA of P-ZrP and P-ZrBP samples obtained after extraction. The comparison between curves a and b reveals that the precursor contained about 30% w/w of template, which was almost completely removed after three extractions with ethanol. Curves b and c show the first step in weight loss that occurs at temperatures less than 200 °C, corresponding to the adsorbed water loss. It is possible to see that the amount of adsorbed water in the ZrP sample (3.25 mol/mol of ZrP) is more than that found in the ZrBP sample (1.3 mol/mol of ZrBP). This is probably due to the more hydrophilic nature of the hydrogenphosphate groups with respect to the benzenephosphonate groups. The second step (24) Kleitz, F.; Schmidt, W.; Schuth, F. Microporous Mesoporous Mater. 2001, 44-45, 95.

Figure 4. N2 adsorption-desorption isotherms on P-ZrBP (a) and Mb/P-ZrBP (b).

can be attributed to the condensation of the hydrogenphosphates or to the degradation of benzenephosphonate groups that occur in the temperature range between 200 and 700 °C. The XRD patterns indicate that the prepared samples were predominantly amorphous (data not shown), and the chemical analysis indicates that the molar ratios P/Zr were around 2.0 in the P-ZrP and P-ZrBP samples. Nitrogen adsorption-desorption isotherms of P-ZrP and P-ZrBP samples, obtained after drying at 60 °C, are shown in Figures 3a and 4a. The calculated BET specific surface areas were 83 and 270 m2/g for P-ZrP and P-ZrBP, respectively. When the same materials were dried at higher temperatures (200 °C, under vacuum), a small decrease in surface area was observed (i.e., for the P-ZrBP sample, the surface area decreased from 270 to 240 m2/g). By applying the t-plot theory, a negligible micropore volume was detected in the case of P-ZrP, obtained after drying at 60 °C, whereas P-ZrBP has an appreciable micropore volume of 0.094 cm3/g and an external surface area of 74 m2/g (the mean pore hydraulic radius is less than 5 Å). The presence of the hysteresis loop between the adsorption and desorption isotherms of the P-ZrP sample is related to the presence of the mesopores. A mesopore volume of 0.17 cm3/g has been calculated by applying BJH theory. The texture and pore characteristics of the materials dried at 200 °C were similar to that observed for the materials dried at 60 °C (data not shown). It may be noted that the BET surface area of P-ZrBP is much higher than that of P-ZrP. However, a large part of the calculated surface area, attributable to the P-ZrBP microporosity, is not accessible for protein adsorption. Therefore, the surface area accessible to the protein is similar for the two materials (74 and 83 m2/g, respectively). It is interesting that when the precipitation of zirconium phosphate and benzenephosphonate was carried out in the absence of any surfactant, materials with very low surface area (2-10 m2/g) and no significant porosity characteristics were obtained.12

Myoglobin Adsorption onto P-ZrP

Figure 5. Adsorption of Mb onto P-ZrP, expressed as the percent of maximum immobilization, at 20 °C and different pH values, as a function of time.

Figure 6. Adsorption of Mb onto P-ZrBP, expressed as the percent of maximum immobilization, at 20 °C and different pH values, as a function of time.

3.2. Myoglobin Adsorption. The adsorption of Mb as a function of time onto P-ZrP and P-ZrBP at 20 °C and different solution pH values ranging from 5.0 to 8.0 is shown in Figures 5 and 6, respectively. All of the experiments were conducted with an initial concentration of 90 mg of Mb/g of support. The adsorption rate and maximum uptake of Mb onto P-ZrP and P-ZrBP are affected by the pH value of the equilibrating solutions. For the hydrophilic P-ZrP support, the maximum amount of Mb adsorbed was reached at pH values of 5 and 6 within 2-4 h. At both of these pH values, the support had a strong affinity for the protein, and the adsorption curves were essentially the same. At pH 7.0, the rate of adsorption was slower; the maximum amount of Mb was adsorbed in 16 h and was only 20% of that recorded at pH 6. At pH 8.0, no adsorption of Mb was observed on P-ZrP. For the hydrophobic support P-ZrBP, the maximum amount of Mb adsorbed was also reached at pH values of 5.0 and 6.0 within 2-4 h. In the same time interval, the Mb uptake at pH 7.0 and 8.0 is considerably lower than the maximum uptake; however, 100% of immobilization is reached after 16 h at pH 7.0. The adsorption isotherms of Mb onto P-ZrP and P-ZrBP, at solution pH ranging from 5.0 to 8.0, are shown in Figures 7 and

Langmuir, Vol. 22, No. 11, 2006 5067

Figure 7. Adsorption isotherms of Mb onto P-ZrP at different pH values (t ) 2 h, T ) 20 °C).

Figure 8. Adsorption isotherms of Mb onto P-ZrBP at different pH values (t ) 2 h, T ) 20 °C).

8, respectively. All of the isotherms were fitted with the Langmuir model to evaluate the binding affinity and the maximum binding capacity of Mb onto the supports. The adsorption isotherms show a sharp initial uptake, indicating a high affinity between Mb and the adsorbent surface followed by a plateau. In these Langmuir-type profiles, the solid lines in the Figures represent the fit of the experimental data using the Langmuir equation (eq 1)25

Qe )

QmaxCe (1/aL) + Ce

(1)

where Ce is the concentration of Mb in solution at equilibrium (mg of Mb/mL), Qe is the amount of Mb adsorbed onto the support per unit weight (mg of Mb/g of support), Qmax is the maximum adsorption capacity of the support per unit weight (mg of Mb/g of support), and aL is the Langmuir constant (mL/ mg of Mb), which is correlated to the affinity between the solute and the adsorbent. High values of aL indicate a strong interaction between Mb and the support. For both supports, an effect of pH on the binding affinity and on the maximum adsorption capacity was observed (Table 1). (25) Al-Duri, B.; Yong, Y. P. J. Mol. Catal. B: Enzym. 1997, 3, 177.

5068 Langmuir, Vol. 22, No. 11, 2006

Bellezza et al.

Table 1. Adsorption Parameters for Mb Adsorbed onto P-ZrP and P-ZrBP with the Langmuir Model Qmax aL surface coverage support pH (mg of Mb/g of support) (mL/mg) (m2/g of support) P-ZrP

5.0 6.0 7.0 8.0

75 82 17 -

431 307 10 -

22-39 24-42 5-8 -

P-ZrBP 5.0 6.0 7.0 8.0

134 126 77 61

241 408 244 202

40-69 37-65 23-40 18-31

Table 2. Characterization of P-ZrP and P-ZrBP before and after Mb Adsorption material

BET surface area (m2/g)

mesopore volumea (cm3/g)

micropore volume (cm3/g)

P-ZrP Mb/P-ZrP P-ZrBP Mb/P-ZrBP

83 43 270 43

0.17 0.09 -

0.094 -

a

Pores with 2-30 nm width.

Because the approximate dimensions of Mb are 4.5 × 3.5 × 2.5 nm3, 1 mg of Mb, adsorbed as a closely packed monolayer, should cover 0.30-0.52 m2, depending on whether the molecules are adsorbed side-on or end-on. Considering the maximum amount of protein adsorbed (Qmax) onto P-ZrP and P-ZrBP at different pH values, we obtained the surface coverage shown in Table 1. These latter values do not exceed the calculated BET surface area for either compound (Table 2), confirming the monolayer adsorption model. The maximum adsorption capacity of the hydrophilic support P-ZrP (Figure 7) was obtained at pH 5.0 and 6.0, where the support has a strong affinity for the protein. At pH 7.0, the binding affinity and capacity were lower, whereas at pH 8.0 no Mb adsorption was observed. The isoelectric point (pI) of Mb is around 7.2,26 hence the protein is positively charged at pH 7.2. The number of positive charges present on the protein increases with decreasing pH from 7.0 to 5.0. The surface of ZrP, rich in P-OH groups, is negatively charged at a pH ranging from 5.0 to 8.0, and the protons have been exchanged by sodium ions of the buffered solution. These results suggest that the equilibrium adsorption capacity of the zirconium phosphate surface was due to the electrostatic interactions between the negatively charged adsorbent and the positively charged protein. The increase in monolayer adsorption capacity (Qmax) with the decrease in the solution pH value (from 8.0 to 6.0) could be attributed to a change in the net positive charge of the protein surface. The lack of adsorption capacity of Mb onto P-ZrP at a solution pH of 8.0, where both protein and adsorbent are negatively charged, is very likely assigned to the strong electrostatic repulsion between the protein and the support surface and the lateral repulsion between the protein molecules. The shapes of the adsorption isotherms of the hydrophobic support P-ZrBP (Figure 8) show a well-defined plateau at all of the pH values tested, indicating a high affinity between Mb and the support. The isotherms clearly show the effect of pH on protein uptake. The amount of adsorbed Mb increases with decreasing pH from 8.0 to 5.0. It should be noted that the (26) Antonini, E.; Brunori, M. Hemoglobin and Myoglobin in Their Reactions with Ligands; North-Holland Research Monographs, Frontiers in Biology; NorthHolland, Amsterdam, 1976.

adsorption of Mb onto P-ZrBP occurs not only at pH near pI, at which the sum of the charges on the protein surface is zero, but also at pH > pI, at which the protein surface is negatively charged, and at pH < pI, at which the protein surface is positively charged. These results suggest that the interactions between the nonpolar side chains of the amino acid residues present on the protein surface and the phenyl group of P-ZrBP overcame the repulsions between the negatively and positively charged protein surfaces, allowing the formation of the stable Mb/P-ZrBP complex at pH ranging from 5.0 to 8.0. At a pH near the isoelectric point, where the net charge of the protein is zero, the lateral interactions between the protein molecules are reduced, and the hydrophobic interactions with the support are dominant. Therefore, the maximum adsorption capacity of P-ZrBP for Mb molecules should be located in the region of the isoelectric point. However, the maximum amount of Mb adsorbed onto P-ZrBP was observed at pH 5.0 (Figure 8). This result can be explained by considering that at each pH value it is important to consider not only the net charge of the surface protein but also the distribution of hydrophobic and hydrophilic amino acid residues on the protein surfaces. When the hydrophobic P-ZrBP surface attracts apolar patches of Mb molecules, the accommodation of adsorbed proteins is probably determined by intermolecular repulsions between equally charged regions of molecules at pH ) pI that are stronger than those occurring at pH < pI. It is interesting that the maximum monolayer adsorption capacity of Mb onto P-ZrBP, obtained at pH 5.0, was twice the amount adsorbed onto P-ZrP at the same pH value. The variation in plateau value can be explained by considering the binding mechanism of the proteins onto the two supports. The binding mechanism of proteins onto hydrophobic supports is complex, and several subprocesses are involved.27 In solution, the structure of the water molecules or ions surrounding the protein and the hydrophobic surface of the support appear to play an important role in the hydrophobic interaction. From a thermodynamic point of view, the extent of protein adsorption is determined by the free energy of adsorption, which at any given temperature depends on the enthalpy and entropy of adsorption.28 Enthalpic driving forces include attractive or repulsive Coulombic forces and attractive van der Waals interaction between protein molecules and the interface. Entropic contributions to the free energy of adsorption come from hydrophobic dehydration and conformational changes in the protein. The contribution of hydrophobic dehydration comes from the removal of the water molecules surrounding the hydrophobic surfaces of the adsorbent and the hydrophobic parts of the protein. These specific water molecules are more ordered than those in the bulk phase. The adsorption of protein onto hydrophobic surfaces results in the displacement of water molecules. This process eventually leads to an increased entropy of water and hence of the entire system. The conformational changes associated with breaking internal noncovalent bonds in the protein core give rise to a large increase in entropy and in structural fluctuations of the adsorbed protein. Irrespective of the stability considerations, proteins in aqueous solution have a high tendency to adsorb onto apolar surfaces even under conditions of electrostatic repulsion.29 The greater surface coverage onto hydrophobic P-ZrBP compared to that onto P-ZrP is in line with this observation. The (27) Chen, W. Y.; Huang, H. M.; Lin, C. C.; Lin, F. Y.; Chan, Y. C. Langmuir 2003, 19, 9395. (28) Roth, C. M.; Lenhoff, A. M. Langmuir 1993, 9, 962. (29) Norde, W.; Anusiem, A. C. I. Colloids Surf. 1992, 66, 73.

Myoglobin Adsorption onto P-ZrP

Langmuir, Vol. 22, No. 11, 2006 5069

As far as the P-ZrP hydrophilic support is concerned, there were no significant effects due to the increase in temperature between 5 and 40 °C (Figure 10). However, a slight increase in binding capacity at 60 °C has been observed. This increase may still be attributed to the conformational changes in the protein at 60 °C. The increase in hydrophobicity reduces the lateral repulsion between the protein molecules and facilitates the promotion of protein aggregates30 with a consequent increase in the binding capacity. An important question that needed to be addressed is whether the protein is adsorbed into the pores of the support or merely on its external surface. To address this, the BET surface area of the support after Mb adsorption was calculated on the basis of N2 adsorption data.

Figure 9. Adsorption isotherms of Mb onto P-ZrBP at different temperatures (t ) 2 h, pH 5.0).

The N2 adsorption isotherms are shown in Figures 3b and 4b and are compared with those of the pristine supports. It may be seen that the isotherms have similar shapes. The BET surface area of samples decreased after Mb adsorption (Table 2). The mesopore volume of P-ZrP decreased from 0.17 to 0.09 cm3/g, whereas the micropore volume of P-ZrBP decreased from 0.094 cm3/g to a negligible volume. However, the isotherm shape did not vary appreciably, indicating that the pore typologies were not changed.

Conclusions The aim of the present work was to investigate the absorption of myoglobin onto hydrophilic zirconium phosphate and hydrophobic zirconium benzenephosphonate prepared with templatedirected synthesis to obtain protein supports with a high surface area and accessible porosity; it was also the aim of this work to determine the optimal pH and temperature of equilibrating solutions for the maximum uptake of myoglobin.

Figure 10. Adsorption isotherms of Mb onto P-ZrP at different temperatures (t ) 2 h, pH 5.0).

large energy gain due to the dehydration of the hydrophobic areas compensates for the repulsive lateral interactions between the adjacent charged adsorbed protein molecules and leads to a very compact monolayer. Figures 9 and 10 show the adsorption isotherms of Mb onto P-ZrBP and P-ZrP, respectively, obtained with equilibrium batch analysis at temperatures ranging from 5 to 60 °C and pH 5.0. The maximum binding capacities of Mb with the P-ZrBP hydrophobic support were almost unchanged at temperatures ranging from 5 to 40 °C where monolayer adsorption behavior is noted. The binding affinity between Mb and the support greatly increased at 60 °C as shown in Figure 9. This behavior could have been induced by changes in protein conformation. The increase in temperature might cause the breaking of noncovalent bonds (e.g., hydrogen bonds and ion pairs), allowing the exposure of the inner hydrophobic parts of the protein. This increase in the overall external hydrophobicity of Mb determines the increase in both binding affinity and capacity for the P-ZrBP support. In conclusion, temperature affects the binding capacity of Mb onto P-ZrBP, and a monolayer adsorption of protein occurring at lower temperatures (5-40 °C) and a multilayer adsorption occurring at a higher temperature (60 °C) are observed.

Concerning the effect of surface area, the comparison among adsorption data of Mb on ZrBP prepared with the classical method11 and on P-ZrBP indicates that an increase of about 8 times the external surface area produces an increase in Mb uptake from 35.6 to 128 mg/g of support as determined for some experimental conditions. The shape of the adsorption isotherms suggests that myoglobin has a relatively high affinity for both of the surfaces of the investigated supports; however, it has a higher affinity for the hydrophobic surface of P-ZrBP when compared to that of the hydrophilic P-ZrP. pH is an important parameter for determining the maximum adsorption capacity of P-ZrBP and P-ZrP for Mb molecules because of the modification of the type and net charge of the protein surface when the pH of the medium changes. The protein binding affinity and capacity increased when the temperature was increased. This phenomenon occurred because the myoglobin conformation changes at high temperature, increasing the exposed hydrophobic region of the protein surface. This process results in a stronger hydrophobic interaction between the protein and the support and a reduction of the repulsion between the adsorbed molecules. Work is in progress to investigate the activity of the biocomposites obtained with these new supports toward the redox reactions. LA0531897 (30) Nakanishi, K.; Sakiyama, T.; Imamura, K. J. Biosci. Bioeng. 2001, 91, 233.