- Email: [email protected]
Protein-containing hydrophobic coatings and films
Biomaterials 23 (2002) 441–448
Protein-containing hydrophobic coatings and films Scott J. Novicka, Jonathan S. Dordickb,* a
Department of Chemical and Biochemical Engineering, The University of Iowa, 125B Chemistry Building, Iowa City, IA 52242, USA b Department of Chemical Engineering, Rensselaer Polytechnic Institute, Troy, NY 12180, USA Received 5 July 2000; received in revised form 15 March 2001; accepted 23 March 2001
Abstract The incorporation of enzymes and other proteins into hydrophobic polymeric coatings and films has been investigated in this study with the goal of generating biologically active materials for biocatalysis, antifouling surfaces, and biorecognition. The protein– polymer composites are created using standard solution coating techniques with poly(methyl methacrylate), polystyrene, and poly(vinyl acetate) as polymers and a-chymotrypsin and trypsin as biocatalysts. The specific enzyme is first extracted into a nonpolar organic solvent using hydrophobic ion-pairing. The ion-paired enzyme is dried and redissolved into a solvent also miscible with the polymer. This solution is then poured over a surface and the solvent is allowed to evaporate to form the enzyme-containing coating, which can then be delaminated to form a film. Leaching of enzyme from and activity of the biocatalytic coatings and films were evaluated. The biocatalytic coatings showed no loss of activity over ca. one week. For the biocatalytic films, the leaching rate was initially high followed by a slow rate of enzyme loss. Activity was measurable for at least one month, with only ca. one-third of the initial activity lost in that time, while, being continuously incubated in a buffer solution. Activity was also exhibited on macromolecular (protein) substrates. The biocatalytic coatings could be reused over 100 times with only a modest loss of activity. Finally, coatings and films containing a lectin (Concanavalin A) were capable of selectively binding to glycoproteins, thereby extending the application of such films for use in bioseparations and biorecognition. r 2001 Elsevier Science Ltd. All rights reserved. Keywords: Enzyme–polymer composites; Biocatalytic coatings and films; Antifouling materials
1. Introduction Enzymes are used in a number of forms, including free in solution, immobilized to solid supports, and crosslinked. They have also been used in thin films and coatingsFoften for applications in biosensors [1,2]. For example, glucose oxidase was incorporated into polyethylene films, derivatized with acrylic acid using plasma induced graft copolymerization, to measure free glucose concentrations in solution [3]. Various oxidative enzymes, such as horseradish peroxidase, glucose oxidase, and lactate oxidase have been incorporated into polyacrylamide-based redox polymers and evaluated for detection of analytes (including glucose and lactate) by ‘‘wiring’’ the enzyme to the redox polymer [4]. Finally, glucose oxidase has been immobilized onto a cellulose acetate membrane and also evaluated as a glucose sensor [5]. In all cases, enzyme has been incorporated *Corresponding author. Fax: +1-518-276-2207. E-mail address: [email protected] (J.S. Dordick).
into rather hydrophilic polymers or attached to the surface of a hydrophobic material. This limits the nature of the polymeric materials that can be employed as a biocatalytic coating or film. The aim of the present work is to incorporate homogeneously an enzyme within a hydrophobic ‘‘plastic’’ polymer coating or film. Examples of these polymers include poly(methyl methacrylate), polystyrene, and poly(vinyl acetate), and hence are quite distinct from the materials used to date. This is accomplished by dissolving the polymer and enzyme into a solvent miscible to both then pouring the solution over a surface and letting the solvent evaporate leaving behind the biocatalytic coating. The coating can then be delaminated from the surface to form a biocatalytic film or membrane. The main challenge to this approach is the need to entrap homogeneously an enzyme within a hydrophobic polymer, therefore necessitating that both the polymer and enzyme be soluble in a single phase. Although native enzymes are essentially insoluble in organic
0142-9612/02/$ - see front matter r 2001 Elsevier Science Ltd. All rights reserved. PII: S 0 1 4 2 - 9 6 1 2 ( 0 1 ) 0 0 1 2 3 - 5
442
S.J. Novick, J.S. Dordick / Biomaterials 23 (2002) 441–448
solvents, there are a number of methods available to solubilize enzymes into organic solvents including covalent addition of polyethylene glycol moieties to the enzyme surface [6], formation of reversed micelles [7], coating the enzyme with a detergent [8], and ionpairing with surfactants [9–11], e.g., using anionic surfactant molecules that are bound electrostatically to the positively charged groups on the enzyme. We have chosen to use the latter technique, which yields highly active and stable enzyme solutions in nonaqueous media. Some advantages of this method include high catalytic activity [9] and solubility [12] of the ion-paired enzyme in organic solvents and the method is quick and facile. Once the protein is dissolved in a suitable organic solvent, the polymer can be added and protein containing coatings and films can be created. This study represents a new and simple method to incorporate homogeneously a variety of proteins, including enzymes and lectins, into hydrophobic, chemically, mechanically, and thermally stable coatings and films. Applications of enzyme-containing coatings and films may include synthetic applications, antifouling paints and coatings, and biocatalytically active films and membranesFthe later for use in simultaneous reaction and separation processes. The lectin-containing films and coatings may be employed as column packings for bioseparations or as well plate coatings for diagnostic use.
2. Materials and methods 2.1. Materials
containing 2 mm AOT was then added and the two phases were magnetically stirred for 3 min, then centrifuged at 4000 rpm for 10 min. After centrifugation, the organic phase was separated from the aqueous phase and most of the organic solvent was removed via rotary evaporation at 301C. The concentration of protein in the organic phase was determined by UV absorbance at 280 nm. 2.3. Formation of biocatalytic coatings and films The ion-paired enzyme (5 mg) was put into a separate vial and the remaining hexane was removed by purging with a stream of dry N2. Toluene (500 ml) was added and the vial was shaken to dissolve the ion-paired enzyme. The polymer solution (1 ml of a 10% w/v solution in toluene) was added and shaken well. This solution was then poured out into an aluminum weighing dish with a bottom diameter of 40 mm. The solvent was allowed to evaporate slowly (15 h) to prevent bubbles from forming. After the bulk of the solvent was removed the coatings were dried for an additional 2–3 h under vacuum to form the biocatalytic coating. To form biocatalytic films, the coatings were delaminated from the surface by pealing off the coatings with a forceps, (this was more easily done for the poly(methyl methacrylate) and polystyrene coatings). The thickness of the films was ca. 30–50 mm as measured by a digital caliper. To limit the gradual leaching of enzyme from the polymer, the coatings or films were treated with 4 ml of 50 mm glutaraldehyde for 2 h then rinsed briefly under a stream of DI water to wash off the residual glutaraldehyde.
a-Chymotrypsin (from bovine pancreas), trypsin (from porcine pancreas), Concanavalin A (from Jack Bean), soybean peroxidase, and casein as well as bis-tris propane buffer and the substrates N-succinly-l-ala-lala-l-pro-l-phe-p-nitroanilide, 4-methylumbelliferyl p-(N,N,N-trimethylammonium) cinnamate (MUTMAC), and 2,20 -azino-di-[3-ethyl-benzothiazoline-(6)sulphonic acid] (ABTS) were obtained from Sigma Chemical Co. (St. Louis, MO). The polymers, poly (methyl methacrylate) (Mw ca. 996 kDa), poly(styrene) (Mw ca. 280 kDa), and poly(vinyl acetate) (Mw ca. 167 kDa) were obtained from Aldrich (Milwaukee, WI). All other reagents and solvents were of the highest grade commercially available.
2.4. Leaching of protein from coatings and films
2.2. Enzyme solubilization into organic solvents
2.5. Active site titration of a-chymotrypsin coatings and films
Following the procedure by Wang et al. [13], the enzyme was dissolved in buffer (10 mm bis-tris propane, pH 7.8, 2 mm CaCl2) to give a concentration of 1 mg/ml. For Con A, the same buffer was used but at pH 7.0, containing 2 mm MnCl2. An equal volume of hexane
To measure the leaching of protein from the coatings or films, the materials were placed in 4 ml of buffer (same as that used for the extraction) to cover the surface entirely and shaken at 75 rpm. Periodically, aliquots of the supernatant were removed and the absorbance at 280 nm was measured. These aliquots were then returned to the supernatant following determination of the protein concentration. Known amounts of pure enzyme were used to construct a calibration curve to relate absorbance at 280 nm to the protein concentration. Fresh buffer was added to replace any that had evaporated to maintain the 4 ml volume.
The fluorometric active site titrant (MUTMAC) was used to titrate the active centers of the a-chymotrypsin containing materials [14]. This titrant will bind to the active site of the enzyme and in the process release a
S.J. Novick, J.S. Dordick / Biomaterials 23 (2002) 441–448
fluorophore into the supernatant. The concentration of this fluorophore will be equal to the active site concentration. The titration was performed in a 0.1 m Na-borate buffer, pH 7.5. Fresh coatings and films were prepared and treated with glutaraldehyde for 2 h. Discs of 0.7 and 1.3 cm were punched out and placed in 2 ml of buffer to which 100 ml of 0.83 mm MUTMAC solution (in water) was added. The mixture was vigorously shaken for 3 min after which the fluorescence of the supernatant was monitored over time (Ex: 360 nm; Em: 450 nm). The fluorescence data was linearly extrapolated back to the time of MUTMAC addition and the difference in fluorescence between the two samples was used to calculate the active site loading. Pure 4methylumbelliferyl was used as a standard. A two-film extrapolation was used to account for any MUTMAC that was hydrolyzed in the bottle or in solution prior to exposure to the enzyme. A one-film extrapolation would overestimate the active site loadings if prior hydrolysis was present. 2.6. Biocatalytic activity of coatings and films The chromogenic tetrapeptide substrate, N-succinyll-ala-l-ala-l-pro-l-phe-p-nitroanilide, was used to measure the activity of the a-chymotrypsin containing coatings and films. First, the incubation buffer was removed and the films were rinsed extensively with DI water. Next, 4 ml of fresh buffer was added and the coating or film was allowed to stir for 4 min at 75 rpm. The buffer was then removed and the activity of this ‘‘wash’’ solution was measured by adding substrate (0.2 mm) and monitoring the release of p-nitroaniline spectrophotometrically at 410 nm. The films were once again rinsed with DI water and 4 ml of 0.2 mm substrate was added while stirring at 75 rpm. A 1 ml aliquot was removed every 30 s for 2 min and the absorbance was measured at 410 nm then the aliquot was returned to the incubation mixture. The activity was determined by a linear fit of absorbance versus time. p-Nitroaniline was used as a standard. All measurements were performed at room temperature. The activity of the coatings or films was calculated by subtracting the activity of the wash from the measured activity of the coating or films. Finally, the coating or film was rinsed again in DI water and the original incubation buffer was returned to the incubation mixture. For the reusability experiments the activity of the wash was not measured. Rather, the substrate was added, the activity was measured as above, then the coating was rinsed with DI water and this procedure was repeated. The milk protein casein was also used to measure the activity of protease containing materials. The method used is similar to Rick [15], wherein casein was added to 100 mm borate buffer, pH 7.8, containing 2 mm CaCl2 added. The mixture was heated to 901C for about 20 min
443
to disperse the casein in the buffer, then cooled to 351C. To each coating or film, 5 ml of this mixture was added and shaken at 90 rpm and 351C. Every 20 min, 250 ml aliquots were removed and added to 750 ml of a 5% trichloroacetic acid solution. This caused precipitation of the large protein fragments, while the smaller ones remained soluble. After 30 min, these samples were centrifuged and the absorbance of the supernatant was measured at 280 nm. A linear fit of absorbance versus time was calculated to determine the activity of the biocatalytic coatings and films. All activity values reported were normalized to the total exposed surface area of the materialsF12.6 cm2 for the coatings and 25.1 cm2 for the films. 2.7. Lectin-affinity of protein-containing films and coatings The activity of soybean peroxidase bound to the Concanavalin A containing materials was measured by a method similar to Pu. tter and Becker [16]. Two stock solutions, one with 11 mg/ml ABTS and one with 10.2 ml/ml 30% H2O2 were prepared in 10 mm bis-tris propane, pH 7.0, containing 2 mm each of CaCl2 and MnCl2. The reaction solution consisted of 3.2 ml buffer, and 392 ml each of the ABTS and H2O2 solution. This solution was poured over the biocatalytic coating or film and the absorbance at 405 nm was monitored over a 3 min period while stirring at 75 rpm at room temperature. A linear fit of absorbance versus time was used to measure the coating/film activity. An extinction coefficient of 1.86 mm @1 min@1 was used to convert absorbance into concentration [16].
3. Results and discussion a-Chymotrypsin (CT) and trypsin (TR) were chosen as our model enzymes due to their well-defined activity and selectivity as well as their ease of extraction into organic solvents using hydrophobic ion pairing. Concanavalin A (Con A) was chosen as a model lectin for two reasons; its large molecular weight compared to the proteases (ca. 108,000 versus ca. 24,000 Da) allowed us to investigate the influence of protein size on leaching of the protein from the polymer, and because this protein has a well defined affinity to mannose-containing glycoproteins and is routinely used in affinity chromatography [17]. The polymers, poly(methyl methacrylate) (pMMA), poly(vinyl acetate) (pVAc), and polystyrene (pST), were chosen due to their chemical and mechanical stability as well as their overall general use in coatings and films as described in our previous studies [13,18]. The coatings and films were prepared using standard solution coating techniques as illustrated in Fig. 1.
444
S.J. Novick, J.S. Dordick / Biomaterials 23 (2002) 441–448
Because pVAc initially was strongly attached to the aluminum surface and was difficult to delaminate without tearing, it was only evaluated as a coating. On the other hand, pMMA and pST formed poor coatings as they easily delaminated (often spontaneously after drying under vacuum) and therefore were only evaluated as films. Because the protein is physically entrapped within the polymer it may be leached out upon use. For this reason, the coatings and films were treated with glutaraldehyde to crosslink the protein molecules within the polymer network. This effectively increased their size resulting in increased chain entanglement, decreasing their diffusion coefficients, and thereby reducing the rate and the extent of protein leaching. There was a significant decrease in the protein leaching rate after treatment with gluteraldehyde. Because gluteraldehyde is a small crosslinker and the protein loading is relatively low (5 w/w%), this may indicate the presence of a phase separation between the protein and the polymer. The protein may congregate in a ‘‘protein rich’’ phase where the protein molecules are much closer together and thus are able to be crosslinked with the gluteraldehyde. Evidence of this, e.g. using scanning electron microscopy, will be the subject of future work. The active site titrant, MUTMAC, has a molecular weight of 400 Da. This is comparable to the size of the low molecular weight (tetrapeptide) substrate used for activity. Thus, this will give a good approximation of not only the active but also the accessible (to low molecular weight substrates) entrapped enzyme molecules. Active site titration on the coatings and films showed on average 0.43 nmol of active and accessible CT per cm2. A simple calculation of accessible enzyme assuming a monolayer of enzyme on the surface (with () no void area and using a hydrodynamic radius of 30 A @3 2 gives 5.88 10 nmol per cm , or 72-fold lower. This
Fig. 1. Schematic for the synthesis of protein containing coatings, films, and membranes (circles represent entrapped enzyme).
demonstrates that the observed activity of these materials is mainly due to enzyme embedded within the polymeric coating or film rather than on the surface. 3.1. Activity and leaching of CT from pVAc coatings Fig. 2A shows the leaching of CT entrapped within pVAc coated onto an aluminum surface (40 mm diameter). As shown in this figure, there is a steady leaching of enzyme from the polymeric coating with ca. 20% of the initial CT loading lost after nearly one week. At longer times the coating began to delaminate from the surface, thus exposing the underside that was not treated with glutaraldehyde and significantly increased the rate of protein leaching (data not shown). The activity of the coating and the wash are shown in Fig. 3A for the hydrolysis of the tetrapeptide substrate. The actual coating activity (activity of the coating minus
Fig. 2. (A) Leaching of CT from pVAc/CT biocatalytic coatings. (B) Leaching of CT from pMMA/CT (K) and pST/CT (’) biocatalytic films.
S.J. Novick, J.S. Dordick / Biomaterials 23 (2002) 441–448
445
activity of the wash) remained relatively constant during the length of the experiment, 6.3 days. The average activity during this time was 8.08 [email protected] 10@4 mm product/(s cm2). The activity of the wash remained relatively steady and was on average 12.3% of the measured film activity. This relatively high activity of the wash (especially compared to the films below) may be due to slight delamination of the coating thus causing leaching of enzyme from the untreated underside. Note that the wash activity is not the activity of the entire enzyme that has leached out of the coating up to that time point. Instead it represents the activity of enzyme that has leached out of the coating during 4 min of incubation in fresh buffer at the given time point that the coating activity was measured. 3.2. Activity and leaching of CT from pMMA and pST films
Fig. 3. (A) Measured wash activity (K), and film minus wash activity (’) of pVAc/CT biocatalytic coating. (B) Measured wash activity (K), and film minus wash activity (’) of pMMA/CT biocatalytic film. (C) Measured wash activity (K), and film minus wash activity (’) of pST/ CT biocatalytic film.
Fig. 2B shows the leaching of CT from pMMA and pST films. The leaching characteristics for both polymers were nearly identical. Initially, for the first 2–3 days, the rate of enzyme loss was significant with about 25% of the initial enzyme leaching out. After this time the loss of enzyme was very gradual and equilibrated at ca. 70% enzyme retention. This result suggests that there are minimally two distinct enzyme phases within the biocatalytic film. Approximately 25–30% of the total enzyme exists in an ‘‘enzyme lean’’ phase, where the enzyme was not effectively crosslinked by the gluteraldehyde, and hence, susceptible to relatively rapid loss from the polymer matrix. Conversely, ca. 70% of the enzyme is retained in the polymer films, indicating that most of the enzyme is within an ‘‘enzyme rich’’ phase where it was more effectively crosslinked by the gluteraldehyde. In both regions, enzyme remains accessible to the substrate. The activity profiles of the film and wash for pMMA and pST are shown in Figs. 3B and C, respectively. The results with these two films were nearly identical; initially high activity that is gradually lost during the 30 days over which the experiment was conducted. The pST/CT material was initially 24% more activity than the pMMA/CT materials (as determined from the intercept of a linear fit of the data). However the pST material lost activity at a slightly higher rate, 1.2%/day, than the pMMA materials, 0.9%/day (also determined from a linear fit of the data). The decrease in the observed film activity could come from a number of sources. The incorporated enzyme could undergo inactivation or autolysis (by the free enzyme that has leached out into the supernatant or the enzyme within the polymer). Alternatively, inactivated enzyme that could leach out into the supernatant may become deposited onto the film surface and block substrate access to the enzyme incorporated into the film. The
446
S.J. Novick, J.S. Dordick / Biomaterials 23 (2002) 441–448
activity decrease is probably not due to the loss of enzyme since the leaching profile shows little loss of enzyme beyond the first 5 days. Unlike the pVAc/CT coating, the activity of the wash is essentially zero during the entire length of the experiment. Both the pMMA/CT and pST/CT initial film activities were higher than that of the pVAc/CT coating, 15.7% and 55.2% higher, respectively (compare Figs. 3A and B with C). The lower activity of the pVAc material could be due to differences in the polymer structures between the more hydrophilic pVAc and the more hydrophobic pMMA and pST. Specifically, the more hydrophilic pVAc may retard partitioning of the relatively hydrophobic tetrapeptide substrate.
3.4. Activity of the biocatalytic films and coating towards a large substrate
There are a number of applications in synthesis and sensing where reusability and long-term use of the biocatalytic coatings and films would be desirable. To that end, we tested the reusability of a pVAc/CT coating (containing 5% CT). Substrate was added, activity was measured, and the coating was then rinsed and this process was repeated. The biocatalytic coating showed high initial activity, which steadily decreased over the first 25 cycles and then maintained ca. 50% maximal activity for the next 75 consecutive cycles (100 cycles total). Fig. 4 shows the average of every five reuse cycles with the error bars representing the standard deviation. The initial loss of activity could be due to enzyme inactivation or from enzyme loss due to leaching. Nevertheless, this result demonstrates the long-term reuse of these materials.
From the previous results it is apparent that the biocatalytic coatings and films are capable of hydrolyzing low molecular weight substrates. We extended this to high molecular weight proteinaceous substrates by using the milk protein casein as substrate. Both CT and TR containing coatings and films were evaluated separately for protein hydrolysis activity as well as combined CT+TR containing films and coatings. Newly prepared materials were incubated in buffer for 2 days to allow loosely incorporated enzyme to be leached from the material. Thus the observed activity would be predominately from the film or coating and not from leached enzyme. All materials were prepared with 5% (w/w) initial protein loading. The casein hydrolysis activity of several films and coatings containing CT, TR, and combined CT+TR is depicted in Fig. 5. For CT containing materials, the pVAc coating showed the highest activity. This is contrary to what was observed for the hydrolysis of the tetrapeptide substrate. Active site titration of the CT containing material showed pVAc/CT to have the highest active site loading, 0.84 nmol active sites/cm2, followed by pST/CT and pMMA/CT with 0.35 and 0.092 nmol active sites/cm2, respectively. The activity towards casein, therefore, appears to correlate to the number of active and accessible sites. This is contrary to the results with the tetrapeptide. Once again, this may be explained by the relative partitioning of the substrate, in this case casein, into the polymers. Casein, being more hydrophilic than the tetrapeptide (the former is more water-soluble than the latter), is likely to partition into the pores of pVAc,
Fig. 4. Reusability of pVAc/CT biocatalytic coating.
Fig. 5. Activity of CT and TR biocatalytic coatings and films towards casein as a substrate.
3.3. Reusability of pVAc/CT coatings
S.J. Novick, J.S. Dordick / Biomaterials 23 (2002) 441–448
Fig. 6. Leaching of Con A from pVAc/Con A coating (K), and pMMA/Con A (’) and pST/Con A (~) films.
while partitioning is likely to be reduced relative to the tetrapeptide for the more hydrophobic pMMA and pST. The TR containing materials showed lower activity versus CT with the most active material being pST. Interestingly, the most active material overall was the coating containing both CT and TR. Because of the broader specificity afforded by the two enzymes (CT favors aromatic side chains [19] and TR favors positively charged residues [20]) this material will be able to hydrolyze casein into smaller fragments and at more sites than either protein individually. Because these two enzymes have very different amino acid specificity, it is not surprising that the activity of the pVAc/CT and pVAc/TR individually sum to that of the pVAc/CT+TR coating (Fig. 5). 3.5. Glycoprotein binding to Con A incorporated coatings and films The broader applicability of protein containing coatings and films was investigated via the incorporation of Con A, a lectin, which selectively binds to sugar moieties found on glycoproteins [17]. A pVAc coating and pMMA and pST films were prepared containing 5% (w/w) Con A. Leaching of the Con A from the coating and films is shown in Fig. 6. Initially, the loss of protein is low, particularly for pVAc, which loses less than 5% protein in the 5 day incubation. To determine whether the polymer-Con A composites are capable of binding to glycoproteins, these materials were incubated in the presence of 4 ml of 0.5 mg/ml soybean peroxidase (SBP), a glycoprotein with a high mannose content, for 20 h and then washed extensively
447
Fig. 7. Activity of SBP bound to coatings and films with and without Con A before (light gray) and after (dark gray) exposure to methyl-a-dmannopyranoside.
with DI water to remove any loosely bound or unbound SBP. The peroxidase activity of the coating and films were then measured using the colorimetric ABTS substrate. The materials were again washed and then incubated for 20 h with a large excess of methyl-a-dmannopyranoside (25 mm, 4 ml total), to compete with SBP for the glycosyl binding sites on Con A. The coating and films were washed again and the activity was measured. These results are shown in Fig. 7 along with controls, which included the same polymers with no Con A incorporated. As expected, neither the controls nor the Con A materials showed peroxidase activity prior to incubation with SBP (data not shown). After incubation with SBP, the controls did show some activity due to minor nonselective adsorption of SBP. Materials containing Con A, however, showed significantly higher activity, from 3.3-fold for pVAc/Con A to 4.5-fold for pST/Con A when compared to the controls. This demonstrates that Con A within a coating or film remains accessible and is able to bind glycoproteins. After the coating and films were incubated with methyla-d-mannopyranoside most of the activity was lost due to the replacement of SBP with the sugar. This again demonstrates that the incorporated Con A did indeed selectively bind SBP.
4. Conclusions The incorporation of enzymes into hydrophobic, mechanically and chemically stable coatings and films represents a new type of biocatalytically active material.
448
S.J. Novick, J.S. Dordick / Biomaterials 23 (2002) 441–448
The procedure for making these materials is facile and can be applied to a wide range of proteins and polymers. The incorporated enzymes remain, to a large extent, within the polymeric material and exhibit activity towards both small and large substrates for extended periods of time. Moreover, Con A incorporated coatings and films show the selective binding of glycoproteins. These materials lend themselves to applications ranging from biocatalytically active paints, coatings, films, and membranes as well as affinity materials for use in the synthetic, diagnostic and medical fields.
Acknowledgements This work was supported by grants from the Biotechnology Research and Development Corporation and the Office of Naval Research.
References [1] Wang B, Li B, Wang Z, Xu G, Wang Q, Dong S. Sol–gel thin-film immobilized soybean peroxidase biosensor for the amperometric determination of hydrogen peroxide in acid medium. Anal Chem 1999;71(10):1935–9. [2] Yao T, Takashima K. Amperometric biosensor with a composite membrane of sol–gel derived enzyme film and electrochemically generated poly(1,2-diaminobenzene) film. Biosensors Bioelectron 1998;13(1):67–73. [3] Wang CC, Hsiue GH. Immobilization of glucose oxidase on polyethylene film using a plasma induced graft copolymerization process. J Biomater Sci Polym Ed 1993;4(4):357–67. [4] de Lumley-Woodyear T, Rocca P, Lindsay J, Dror Y, Freeman A, Heller A. Polyacrylamide-based redox polymer for connecting redox centers of enzymes to electrodes. Anal Chem 1995;67(8):1332–8. [5] Wang S, Ji X, Yuan Z. Study of cellulose acetate membrane-based glucose biosensors. Clin J Biotechnol 1995;11(3):199–205.
[6] Delgado C, Francis GE, Fisher D. The uses and properties of PEG-linked proteins. Crit Rev Ther Drug Carrier Syst 1992;9(3,4):249–304. [7] Creagh AL, Prausnitx JM, Blanch HW. Structural and catalytic properties of enzymes in reverse micelles. Enzyme Microb Technol 1993;15:383–92. [8] Tsuzuki W, Sasaki T, Suzuki T. Effect of detergent attached to enzyme molecules on the activity of organic-solvent-soluble lipases. J Chem Soc Perkin Trans 1991;2(11):1851–4. [9] Paradkar VM, Dordick JS. Aqueous-like activity of a-chymotrypsin dissolved in nearly anhydrous organic solvents. J Amer Chem Soc 1994;116:5009–10. [10] Paradkar VM, Dordick JS. Mechanism of extraction of chymotrypsin into isooctane at very low concentrations of aerosol OT in the absence of reversed micelles. Biotechnol Bioeng 1994;43: 529–40. [11] Meyer JD, Kendrick BS, Matsuura JE, Ruth JA, Bryan PN, Manning MC. Generation of soluble and active subtilisin and a-chymotrypsin in organic solvents via hydrophobic ion pairing. Int J Pept Protein Res 1996;47(3):177–81. [12] Novick SJ, Dordick JS. Investigating the effects of polymer chemistry on the activity of biocatalytic plastic materials. Biotechnol Bioeng 2000;68:665–71. [13] Wang P, Sergeeva MV, Lim L, Dordick JS. Biocatalytic plastics as active and stable materials for biotransformation. Nat Biotechnol 1997;15:789–93. [14] Jameson GW, Roberts DV, Adams RW, Kyle WSA, Elmore DT. Determination of the operational molarity of solutions of bovine a-chymotrypsin, trypsin, thrombin and factor Xa by spectrofluorimetric titration. Biochem J 1973;131:107–17. [15] Rick W. Chymotrypsin. In: Bergmeyer HU, editor. Methods of enzymatic analysis, vol. 2. New York: Academic Press Inc., 1974. p. 1006–9. [16] Pu. tter J, Becker R. Peroxidases. In: Bergemeyer HN, editor. Methods of enzymatic analysis,, vol. 3. Deerfield Beach: Verlag Chemie, 1983. p. 286–93. [17] Liener IE, Sharon N, Goldstein IJ, editors. Lectins: properties functions and applications in biology and medicine. New York: Academic Press, 1986. [18] Dordick JS, Novick SJ, Sergeeva MV. Biocatalytic plastics. Chem Ind (London) 1998;1:17–20. [19] Berezin I, Martinek K. Specificity of a-chymotrypsin. FEBS Lett 1970;8:173. [20] Keil-Dlouh!a V, Zylber N, Tong N, Keil B. Cleavage of glucagon by a- and b-trypsin. FEBS Lett 1971;16:287.
Protein-containing hydrophobic coatings and films Scott J. Novicka, Jonathan S. Dordickb,* a
Department of Chemical and Biochemical Engineering, The University of Iowa, 125B Chemistry Building, Iowa City, IA 52242, USA b Department of Chemical Engineering, Rensselaer Polytechnic Institute, Troy, NY 12180, USA Received 5 July 2000; received in revised form 15 March 2001; accepted 23 March 2001
Abstract The incorporation of enzymes and other proteins into hydrophobic polymeric coatings and films has been investigated in this study with the goal of generating biologically active materials for biocatalysis, antifouling surfaces, and biorecognition. The protein– polymer composites are created using standard solution coating techniques with poly(methyl methacrylate), polystyrene, and poly(vinyl acetate) as polymers and a-chymotrypsin and trypsin as biocatalysts. The specific enzyme is first extracted into a nonpolar organic solvent using hydrophobic ion-pairing. The ion-paired enzyme is dried and redissolved into a solvent also miscible with the polymer. This solution is then poured over a surface and the solvent is allowed to evaporate to form the enzyme-containing coating, which can then be delaminated to form a film. Leaching of enzyme from and activity of the biocatalytic coatings and films were evaluated. The biocatalytic coatings showed no loss of activity over ca. one week. For the biocatalytic films, the leaching rate was initially high followed by a slow rate of enzyme loss. Activity was measurable for at least one month, with only ca. one-third of the initial activity lost in that time, while, being continuously incubated in a buffer solution. Activity was also exhibited on macromolecular (protein) substrates. The biocatalytic coatings could be reused over 100 times with only a modest loss of activity. Finally, coatings and films containing a lectin (Concanavalin A) were capable of selectively binding to glycoproteins, thereby extending the application of such films for use in bioseparations and biorecognition. r 2001 Elsevier Science Ltd. All rights reserved. Keywords: Enzyme–polymer composites; Biocatalytic coatings and films; Antifouling materials
1. Introduction Enzymes are used in a number of forms, including free in solution, immobilized to solid supports, and crosslinked. They have also been used in thin films and coatingsFoften for applications in biosensors [1,2]. For example, glucose oxidase was incorporated into polyethylene films, derivatized with acrylic acid using plasma induced graft copolymerization, to measure free glucose concentrations in solution [3]. Various oxidative enzymes, such as horseradish peroxidase, glucose oxidase, and lactate oxidase have been incorporated into polyacrylamide-based redox polymers and evaluated for detection of analytes (including glucose and lactate) by ‘‘wiring’’ the enzyme to the redox polymer [4]. Finally, glucose oxidase has been immobilized onto a cellulose acetate membrane and also evaluated as a glucose sensor [5]. In all cases, enzyme has been incorporated *Corresponding author. Fax: +1-518-276-2207. E-mail address: [email protected] (J.S. Dordick).
into rather hydrophilic polymers or attached to the surface of a hydrophobic material. This limits the nature of the polymeric materials that can be employed as a biocatalytic coating or film. The aim of the present work is to incorporate homogeneously an enzyme within a hydrophobic ‘‘plastic’’ polymer coating or film. Examples of these polymers include poly(methyl methacrylate), polystyrene, and poly(vinyl acetate), and hence are quite distinct from the materials used to date. This is accomplished by dissolving the polymer and enzyme into a solvent miscible to both then pouring the solution over a surface and letting the solvent evaporate leaving behind the biocatalytic coating. The coating can then be delaminated from the surface to form a biocatalytic film or membrane. The main challenge to this approach is the need to entrap homogeneously an enzyme within a hydrophobic polymer, therefore necessitating that both the polymer and enzyme be soluble in a single phase. Although native enzymes are essentially insoluble in organic
0142-9612/02/$ - see front matter r 2001 Elsevier Science Ltd. All rights reserved. PII: S 0 1 4 2 - 9 6 1 2 ( 0 1 ) 0 0 1 2 3 - 5
442
S.J. Novick, J.S. Dordick / Biomaterials 23 (2002) 441–448
solvents, there are a number of methods available to solubilize enzymes into organic solvents including covalent addition of polyethylene glycol moieties to the enzyme surface [6], formation of reversed micelles [7], coating the enzyme with a detergent [8], and ionpairing with surfactants [9–11], e.g., using anionic surfactant molecules that are bound electrostatically to the positively charged groups on the enzyme. We have chosen to use the latter technique, which yields highly active and stable enzyme solutions in nonaqueous media. Some advantages of this method include high catalytic activity [9] and solubility [12] of the ion-paired enzyme in organic solvents and the method is quick and facile. Once the protein is dissolved in a suitable organic solvent, the polymer can be added and protein containing coatings and films can be created. This study represents a new and simple method to incorporate homogeneously a variety of proteins, including enzymes and lectins, into hydrophobic, chemically, mechanically, and thermally stable coatings and films. Applications of enzyme-containing coatings and films may include synthetic applications, antifouling paints and coatings, and biocatalytically active films and membranesFthe later for use in simultaneous reaction and separation processes. The lectin-containing films and coatings may be employed as column packings for bioseparations or as well plate coatings for diagnostic use.
2. Materials and methods 2.1. Materials
containing 2 mm AOT was then added and the two phases were magnetically stirred for 3 min, then centrifuged at 4000 rpm for 10 min. After centrifugation, the organic phase was separated from the aqueous phase and most of the organic solvent was removed via rotary evaporation at 301C. The concentration of protein in the organic phase was determined by UV absorbance at 280 nm. 2.3. Formation of biocatalytic coatings and films The ion-paired enzyme (5 mg) was put into a separate vial and the remaining hexane was removed by purging with a stream of dry N2. Toluene (500 ml) was added and the vial was shaken to dissolve the ion-paired enzyme. The polymer solution (1 ml of a 10% w/v solution in toluene) was added and shaken well. This solution was then poured out into an aluminum weighing dish with a bottom diameter of 40 mm. The solvent was allowed to evaporate slowly (15 h) to prevent bubbles from forming. After the bulk of the solvent was removed the coatings were dried for an additional 2–3 h under vacuum to form the biocatalytic coating. To form biocatalytic films, the coatings were delaminated from the surface by pealing off the coatings with a forceps, (this was more easily done for the poly(methyl methacrylate) and polystyrene coatings). The thickness of the films was ca. 30–50 mm as measured by a digital caliper. To limit the gradual leaching of enzyme from the polymer, the coatings or films were treated with 4 ml of 50 mm glutaraldehyde for 2 h then rinsed briefly under a stream of DI water to wash off the residual glutaraldehyde.
a-Chymotrypsin (from bovine pancreas), trypsin (from porcine pancreas), Concanavalin A (from Jack Bean), soybean peroxidase, and casein as well as bis-tris propane buffer and the substrates N-succinly-l-ala-lala-l-pro-l-phe-p-nitroanilide, 4-methylumbelliferyl p-(N,N,N-trimethylammonium) cinnamate (MUTMAC), and 2,20 -azino-di-[3-ethyl-benzothiazoline-(6)sulphonic acid] (ABTS) were obtained from Sigma Chemical Co. (St. Louis, MO). The polymers, poly (methyl methacrylate) (Mw ca. 996 kDa), poly(styrene) (Mw ca. 280 kDa), and poly(vinyl acetate) (Mw ca. 167 kDa) were obtained from Aldrich (Milwaukee, WI). All other reagents and solvents were of the highest grade commercially available.
2.4. Leaching of protein from coatings and films
2.2. Enzyme solubilization into organic solvents
2.5. Active site titration of a-chymotrypsin coatings and films
Following the procedure by Wang et al. [13], the enzyme was dissolved in buffer (10 mm bis-tris propane, pH 7.8, 2 mm CaCl2) to give a concentration of 1 mg/ml. For Con A, the same buffer was used but at pH 7.0, containing 2 mm MnCl2. An equal volume of hexane
To measure the leaching of protein from the coatings or films, the materials were placed in 4 ml of buffer (same as that used for the extraction) to cover the surface entirely and shaken at 75 rpm. Periodically, aliquots of the supernatant were removed and the absorbance at 280 nm was measured. These aliquots were then returned to the supernatant following determination of the protein concentration. Known amounts of pure enzyme were used to construct a calibration curve to relate absorbance at 280 nm to the protein concentration. Fresh buffer was added to replace any that had evaporated to maintain the 4 ml volume.
The fluorometric active site titrant (MUTMAC) was used to titrate the active centers of the a-chymotrypsin containing materials [14]. This titrant will bind to the active site of the enzyme and in the process release a
S.J. Novick, J.S. Dordick / Biomaterials 23 (2002) 441–448
fluorophore into the supernatant. The concentration of this fluorophore will be equal to the active site concentration. The titration was performed in a 0.1 m Na-borate buffer, pH 7.5. Fresh coatings and films were prepared and treated with glutaraldehyde for 2 h. Discs of 0.7 and 1.3 cm were punched out and placed in 2 ml of buffer to which 100 ml of 0.83 mm MUTMAC solution (in water) was added. The mixture was vigorously shaken for 3 min after which the fluorescence of the supernatant was monitored over time (Ex: 360 nm; Em: 450 nm). The fluorescence data was linearly extrapolated back to the time of MUTMAC addition and the difference in fluorescence between the two samples was used to calculate the active site loading. Pure 4methylumbelliferyl was used as a standard. A two-film extrapolation was used to account for any MUTMAC that was hydrolyzed in the bottle or in solution prior to exposure to the enzyme. A one-film extrapolation would overestimate the active site loadings if prior hydrolysis was present. 2.6. Biocatalytic activity of coatings and films The chromogenic tetrapeptide substrate, N-succinyll-ala-l-ala-l-pro-l-phe-p-nitroanilide, was used to measure the activity of the a-chymotrypsin containing coatings and films. First, the incubation buffer was removed and the films were rinsed extensively with DI water. Next, 4 ml of fresh buffer was added and the coating or film was allowed to stir for 4 min at 75 rpm. The buffer was then removed and the activity of this ‘‘wash’’ solution was measured by adding substrate (0.2 mm) and monitoring the release of p-nitroaniline spectrophotometrically at 410 nm. The films were once again rinsed with DI water and 4 ml of 0.2 mm substrate was added while stirring at 75 rpm. A 1 ml aliquot was removed every 30 s for 2 min and the absorbance was measured at 410 nm then the aliquot was returned to the incubation mixture. The activity was determined by a linear fit of absorbance versus time. p-Nitroaniline was used as a standard. All measurements were performed at room temperature. The activity of the coatings or films was calculated by subtracting the activity of the wash from the measured activity of the coating or films. Finally, the coating or film was rinsed again in DI water and the original incubation buffer was returned to the incubation mixture. For the reusability experiments the activity of the wash was not measured. Rather, the substrate was added, the activity was measured as above, then the coating was rinsed with DI water and this procedure was repeated. The milk protein casein was also used to measure the activity of protease containing materials. The method used is similar to Rick [15], wherein casein was added to 100 mm borate buffer, pH 7.8, containing 2 mm CaCl2 added. The mixture was heated to 901C for about 20 min
443
to disperse the casein in the buffer, then cooled to 351C. To each coating or film, 5 ml of this mixture was added and shaken at 90 rpm and 351C. Every 20 min, 250 ml aliquots were removed and added to 750 ml of a 5% trichloroacetic acid solution. This caused precipitation of the large protein fragments, while the smaller ones remained soluble. After 30 min, these samples were centrifuged and the absorbance of the supernatant was measured at 280 nm. A linear fit of absorbance versus time was calculated to determine the activity of the biocatalytic coatings and films. All activity values reported were normalized to the total exposed surface area of the materialsF12.6 cm2 for the coatings and 25.1 cm2 for the films. 2.7. Lectin-affinity of protein-containing films and coatings The activity of soybean peroxidase bound to the Concanavalin A containing materials was measured by a method similar to Pu. tter and Becker [16]. Two stock solutions, one with 11 mg/ml ABTS and one with 10.2 ml/ml 30% H2O2 were prepared in 10 mm bis-tris propane, pH 7.0, containing 2 mm each of CaCl2 and MnCl2. The reaction solution consisted of 3.2 ml buffer, and 392 ml each of the ABTS and H2O2 solution. This solution was poured over the biocatalytic coating or film and the absorbance at 405 nm was monitored over a 3 min period while stirring at 75 rpm at room temperature. A linear fit of absorbance versus time was used to measure the coating/film activity. An extinction coefficient of 1.86 mm @1 min@1 was used to convert absorbance into concentration [16].
3. Results and discussion a-Chymotrypsin (CT) and trypsin (TR) were chosen as our model enzymes due to their well-defined activity and selectivity as well as their ease of extraction into organic solvents using hydrophobic ion pairing. Concanavalin A (Con A) was chosen as a model lectin for two reasons; its large molecular weight compared to the proteases (ca. 108,000 versus ca. 24,000 Da) allowed us to investigate the influence of protein size on leaching of the protein from the polymer, and because this protein has a well defined affinity to mannose-containing glycoproteins and is routinely used in affinity chromatography [17]. The polymers, poly(methyl methacrylate) (pMMA), poly(vinyl acetate) (pVAc), and polystyrene (pST), were chosen due to their chemical and mechanical stability as well as their overall general use in coatings and films as described in our previous studies [13,18]. The coatings and films were prepared using standard solution coating techniques as illustrated in Fig. 1.
444
S.J. Novick, J.S. Dordick / Biomaterials 23 (2002) 441–448
Because pVAc initially was strongly attached to the aluminum surface and was difficult to delaminate without tearing, it was only evaluated as a coating. On the other hand, pMMA and pST formed poor coatings as they easily delaminated (often spontaneously after drying under vacuum) and therefore were only evaluated as films. Because the protein is physically entrapped within the polymer it may be leached out upon use. For this reason, the coatings and films were treated with glutaraldehyde to crosslink the protein molecules within the polymer network. This effectively increased their size resulting in increased chain entanglement, decreasing their diffusion coefficients, and thereby reducing the rate and the extent of protein leaching. There was a significant decrease in the protein leaching rate after treatment with gluteraldehyde. Because gluteraldehyde is a small crosslinker and the protein loading is relatively low (5 w/w%), this may indicate the presence of a phase separation between the protein and the polymer. The protein may congregate in a ‘‘protein rich’’ phase where the protein molecules are much closer together and thus are able to be crosslinked with the gluteraldehyde. Evidence of this, e.g. using scanning electron microscopy, will be the subject of future work. The active site titrant, MUTMAC, has a molecular weight of 400 Da. This is comparable to the size of the low molecular weight (tetrapeptide) substrate used for activity. Thus, this will give a good approximation of not only the active but also the accessible (to low molecular weight substrates) entrapped enzyme molecules. Active site titration on the coatings and films showed on average 0.43 nmol of active and accessible CT per cm2. A simple calculation of accessible enzyme assuming a monolayer of enzyme on the surface (with () no void area and using a hydrodynamic radius of 30 A @3 2 gives 5.88 10 nmol per cm , or 72-fold lower. This
Fig. 1. Schematic for the synthesis of protein containing coatings, films, and membranes (circles represent entrapped enzyme).
demonstrates that the observed activity of these materials is mainly due to enzyme embedded within the polymeric coating or film rather than on the surface. 3.1. Activity and leaching of CT from pVAc coatings Fig. 2A shows the leaching of CT entrapped within pVAc coated onto an aluminum surface (40 mm diameter). As shown in this figure, there is a steady leaching of enzyme from the polymeric coating with ca. 20% of the initial CT loading lost after nearly one week. At longer times the coating began to delaminate from the surface, thus exposing the underside that was not treated with glutaraldehyde and significantly increased the rate of protein leaching (data not shown). The activity of the coating and the wash are shown in Fig. 3A for the hydrolysis of the tetrapeptide substrate. The actual coating activity (activity of the coating minus
Fig. 2. (A) Leaching of CT from pVAc/CT biocatalytic coatings. (B) Leaching of CT from pMMA/CT (K) and pST/CT (’) biocatalytic films.
S.J. Novick, J.S. Dordick / Biomaterials 23 (2002) 441–448
445
activity of the wash) remained relatively constant during the length of the experiment, 6.3 days. The average activity during this time was 8.08 [email protected] 10@4 mm product/(s cm2). The activity of the wash remained relatively steady and was on average 12.3% of the measured film activity. This relatively high activity of the wash (especially compared to the films below) may be due to slight delamination of the coating thus causing leaching of enzyme from the untreated underside. Note that the wash activity is not the activity of the entire enzyme that has leached out of the coating up to that time point. Instead it represents the activity of enzyme that has leached out of the coating during 4 min of incubation in fresh buffer at the given time point that the coating activity was measured. 3.2. Activity and leaching of CT from pMMA and pST films
Fig. 3. (A) Measured wash activity (K), and film minus wash activity (’) of pVAc/CT biocatalytic coating. (B) Measured wash activity (K), and film minus wash activity (’) of pMMA/CT biocatalytic film. (C) Measured wash activity (K), and film minus wash activity (’) of pST/ CT biocatalytic film.
Fig. 2B shows the leaching of CT from pMMA and pST films. The leaching characteristics for both polymers were nearly identical. Initially, for the first 2–3 days, the rate of enzyme loss was significant with about 25% of the initial enzyme leaching out. After this time the loss of enzyme was very gradual and equilibrated at ca. 70% enzyme retention. This result suggests that there are minimally two distinct enzyme phases within the biocatalytic film. Approximately 25–30% of the total enzyme exists in an ‘‘enzyme lean’’ phase, where the enzyme was not effectively crosslinked by the gluteraldehyde, and hence, susceptible to relatively rapid loss from the polymer matrix. Conversely, ca. 70% of the enzyme is retained in the polymer films, indicating that most of the enzyme is within an ‘‘enzyme rich’’ phase where it was more effectively crosslinked by the gluteraldehyde. In both regions, enzyme remains accessible to the substrate. The activity profiles of the film and wash for pMMA and pST are shown in Figs. 3B and C, respectively. The results with these two films were nearly identical; initially high activity that is gradually lost during the 30 days over which the experiment was conducted. The pST/CT material was initially 24% more activity than the pMMA/CT materials (as determined from the intercept of a linear fit of the data). However the pST material lost activity at a slightly higher rate, 1.2%/day, than the pMMA materials, 0.9%/day (also determined from a linear fit of the data). The decrease in the observed film activity could come from a number of sources. The incorporated enzyme could undergo inactivation or autolysis (by the free enzyme that has leached out into the supernatant or the enzyme within the polymer). Alternatively, inactivated enzyme that could leach out into the supernatant may become deposited onto the film surface and block substrate access to the enzyme incorporated into the film. The
446
S.J. Novick, J.S. Dordick / Biomaterials 23 (2002) 441–448
activity decrease is probably not due to the loss of enzyme since the leaching profile shows little loss of enzyme beyond the first 5 days. Unlike the pVAc/CT coating, the activity of the wash is essentially zero during the entire length of the experiment. Both the pMMA/CT and pST/CT initial film activities were higher than that of the pVAc/CT coating, 15.7% and 55.2% higher, respectively (compare Figs. 3A and B with C). The lower activity of the pVAc material could be due to differences in the polymer structures between the more hydrophilic pVAc and the more hydrophobic pMMA and pST. Specifically, the more hydrophilic pVAc may retard partitioning of the relatively hydrophobic tetrapeptide substrate.
3.4. Activity of the biocatalytic films and coating towards a large substrate
There are a number of applications in synthesis and sensing where reusability and long-term use of the biocatalytic coatings and films would be desirable. To that end, we tested the reusability of a pVAc/CT coating (containing 5% CT). Substrate was added, activity was measured, and the coating was then rinsed and this process was repeated. The biocatalytic coating showed high initial activity, which steadily decreased over the first 25 cycles and then maintained ca. 50% maximal activity for the next 75 consecutive cycles (100 cycles total). Fig. 4 shows the average of every five reuse cycles with the error bars representing the standard deviation. The initial loss of activity could be due to enzyme inactivation or from enzyme loss due to leaching. Nevertheless, this result demonstrates the long-term reuse of these materials.
From the previous results it is apparent that the biocatalytic coatings and films are capable of hydrolyzing low molecular weight substrates. We extended this to high molecular weight proteinaceous substrates by using the milk protein casein as substrate. Both CT and TR containing coatings and films were evaluated separately for protein hydrolysis activity as well as combined CT+TR containing films and coatings. Newly prepared materials were incubated in buffer for 2 days to allow loosely incorporated enzyme to be leached from the material. Thus the observed activity would be predominately from the film or coating and not from leached enzyme. All materials were prepared with 5% (w/w) initial protein loading. The casein hydrolysis activity of several films and coatings containing CT, TR, and combined CT+TR is depicted in Fig. 5. For CT containing materials, the pVAc coating showed the highest activity. This is contrary to what was observed for the hydrolysis of the tetrapeptide substrate. Active site titration of the CT containing material showed pVAc/CT to have the highest active site loading, 0.84 nmol active sites/cm2, followed by pST/CT and pMMA/CT with 0.35 and 0.092 nmol active sites/cm2, respectively. The activity towards casein, therefore, appears to correlate to the number of active and accessible sites. This is contrary to the results with the tetrapeptide. Once again, this may be explained by the relative partitioning of the substrate, in this case casein, into the polymers. Casein, being more hydrophilic than the tetrapeptide (the former is more water-soluble than the latter), is likely to partition into the pores of pVAc,
Fig. 4. Reusability of pVAc/CT biocatalytic coating.
Fig. 5. Activity of CT and TR biocatalytic coatings and films towards casein as a substrate.
3.3. Reusability of pVAc/CT coatings
S.J. Novick, J.S. Dordick / Biomaterials 23 (2002) 441–448
Fig. 6. Leaching of Con A from pVAc/Con A coating (K), and pMMA/Con A (’) and pST/Con A (~) films.
while partitioning is likely to be reduced relative to the tetrapeptide for the more hydrophobic pMMA and pST. The TR containing materials showed lower activity versus CT with the most active material being pST. Interestingly, the most active material overall was the coating containing both CT and TR. Because of the broader specificity afforded by the two enzymes (CT favors aromatic side chains [19] and TR favors positively charged residues [20]) this material will be able to hydrolyze casein into smaller fragments and at more sites than either protein individually. Because these two enzymes have very different amino acid specificity, it is not surprising that the activity of the pVAc/CT and pVAc/TR individually sum to that of the pVAc/CT+TR coating (Fig. 5). 3.5. Glycoprotein binding to Con A incorporated coatings and films The broader applicability of protein containing coatings and films was investigated via the incorporation of Con A, a lectin, which selectively binds to sugar moieties found on glycoproteins [17]. A pVAc coating and pMMA and pST films were prepared containing 5% (w/w) Con A. Leaching of the Con A from the coating and films is shown in Fig. 6. Initially, the loss of protein is low, particularly for pVAc, which loses less than 5% protein in the 5 day incubation. To determine whether the polymer-Con A composites are capable of binding to glycoproteins, these materials were incubated in the presence of 4 ml of 0.5 mg/ml soybean peroxidase (SBP), a glycoprotein with a high mannose content, for 20 h and then washed extensively
447
Fig. 7. Activity of SBP bound to coatings and films with and without Con A before (light gray) and after (dark gray) exposure to methyl-a-dmannopyranoside.
with DI water to remove any loosely bound or unbound SBP. The peroxidase activity of the coating and films were then measured using the colorimetric ABTS substrate. The materials were again washed and then incubated for 20 h with a large excess of methyl-a-dmannopyranoside (25 mm, 4 ml total), to compete with SBP for the glycosyl binding sites on Con A. The coating and films were washed again and the activity was measured. These results are shown in Fig. 7 along with controls, which included the same polymers with no Con A incorporated. As expected, neither the controls nor the Con A materials showed peroxidase activity prior to incubation with SBP (data not shown). After incubation with SBP, the controls did show some activity due to minor nonselective adsorption of SBP. Materials containing Con A, however, showed significantly higher activity, from 3.3-fold for pVAc/Con A to 4.5-fold for pST/Con A when compared to the controls. This demonstrates that Con A within a coating or film remains accessible and is able to bind glycoproteins. After the coating and films were incubated with methyla-d-mannopyranoside most of the activity was lost due to the replacement of SBP with the sugar. This again demonstrates that the incorporated Con A did indeed selectively bind SBP.
4. Conclusions The incorporation of enzymes into hydrophobic, mechanically and chemically stable coatings and films represents a new type of biocatalytically active material.
448
S.J. Novick, J.S. Dordick / Biomaterials 23 (2002) 441–448
The procedure for making these materials is facile and can be applied to a wide range of proteins and polymers. The incorporated enzymes remain, to a large extent, within the polymeric material and exhibit activity towards both small and large substrates for extended periods of time. Moreover, Con A incorporated coatings and films show the selective binding of glycoproteins. These materials lend themselves to applications ranging from biocatalytically active paints, coatings, films, and membranes as well as affinity materials for use in the synthetic, diagnostic and medical fields.
Acknowledgements This work was supported by grants from the Biotechnology Research and Development Corporation and the Office of Naval Research.
References [1] Wang B, Li B, Wang Z, Xu G, Wang Q, Dong S. Sol–gel thin-film immobilized soybean peroxidase biosensor for the amperometric determination of hydrogen peroxide in acid medium. Anal Chem 1999;71(10):1935–9. [2] Yao T, Takashima K. Amperometric biosensor with a composite membrane of sol–gel derived enzyme film and electrochemically generated poly(1,2-diaminobenzene) film. Biosensors Bioelectron 1998;13(1):67–73. [3] Wang CC, Hsiue GH. Immobilization of glucose oxidase on polyethylene film using a plasma induced graft copolymerization process. J Biomater Sci Polym Ed 1993;4(4):357–67. [4] de Lumley-Woodyear T, Rocca P, Lindsay J, Dror Y, Freeman A, Heller A. Polyacrylamide-based redox polymer for connecting redox centers of enzymes to electrodes. Anal Chem 1995;67(8):1332–8. [5] Wang S, Ji X, Yuan Z. Study of cellulose acetate membrane-based glucose biosensors. Clin J Biotechnol 1995;11(3):199–205.
[6] Delgado C, Francis GE, Fisher D. The uses and properties of PEG-linked proteins. Crit Rev Ther Drug Carrier Syst 1992;9(3,4):249–304. [7] Creagh AL, Prausnitx JM, Blanch HW. Structural and catalytic properties of enzymes in reverse micelles. Enzyme Microb Technol 1993;15:383–92. [8] Tsuzuki W, Sasaki T, Suzuki T. Effect of detergent attached to enzyme molecules on the activity of organic-solvent-soluble lipases. J Chem Soc Perkin Trans 1991;2(11):1851–4. [9] Paradkar VM, Dordick JS. Aqueous-like activity of a-chymotrypsin dissolved in nearly anhydrous organic solvents. J Amer Chem Soc 1994;116:5009–10. [10] Paradkar VM, Dordick JS. Mechanism of extraction of chymotrypsin into isooctane at very low concentrations of aerosol OT in the absence of reversed micelles. Biotechnol Bioeng 1994;43: 529–40. [11] Meyer JD, Kendrick BS, Matsuura JE, Ruth JA, Bryan PN, Manning MC. Generation of soluble and active subtilisin and a-chymotrypsin in organic solvents via hydrophobic ion pairing. Int J Pept Protein Res 1996;47(3):177–81. [12] Novick SJ, Dordick JS. Investigating the effects of polymer chemistry on the activity of biocatalytic plastic materials. Biotechnol Bioeng 2000;68:665–71. [13] Wang P, Sergeeva MV, Lim L, Dordick JS. Biocatalytic plastics as active and stable materials for biotransformation. Nat Biotechnol 1997;15:789–93. [14] Jameson GW, Roberts DV, Adams RW, Kyle WSA, Elmore DT. Determination of the operational molarity of solutions of bovine a-chymotrypsin, trypsin, thrombin and factor Xa by spectrofluorimetric titration. Biochem J 1973;131:107–17. [15] Rick W. Chymotrypsin. In: Bergmeyer HU, editor. Methods of enzymatic analysis, vol. 2. New York: Academic Press Inc., 1974. p. 1006–9. [16] Pu. tter J, Becker R. Peroxidases. In: Bergemeyer HN, editor. Methods of enzymatic analysis,, vol. 3. Deerfield Beach: Verlag Chemie, 1983. p. 286–93. [17] Liener IE, Sharon N, Goldstein IJ, editors. Lectins: properties functions and applications in biology and medicine. New York: Academic Press, 1986. [18] Dordick JS, Novick SJ, Sergeeva MV. Biocatalytic plastics. Chem Ind (London) 1998;1:17–20. [19] Berezin I, Martinek K. Specificity of a-chymotrypsin. FEBS Lett 1970;8:173. [20] Keil-Dlouh!a V, Zylber N, Tong N, Keil B. Cleavage of glucagon by a- and b-trypsin. FEBS Lett 1971;16:287.