Biomacromolecules 2005, 6, 35-38
35
Transglutaminase-Mediated Protein Immobilization to Casein Nanolayers Created on a Plastic Surface Noriho Kamiya,*,† Satoshi Doi,† Jo Tominaga,† Hirofumi Ichinose,‡ and Masahiro Goto*,†,‡ Department of Applied Chemistry, Graduate School of Engineering, Kyushu University, 6-10-1 Hakozaki, Fukuoka, Japan, and PRESTO, JST, 4-1-8 Honmachi, Kawaguchi, Saitama Received August 27, 2004; Revised Manuscript Received November 8, 2004
An enzymatic method for covalent and site-specific immobilization of recombinant proteins on a plastic surface was explored. Using Escherichia coli alkaline phosphatase (AP) with a specific peptide tag (MKHKGS) genetically incorporated at the N-terminus as a model (NK-AP), microbial transglutaminase (MTG)-mediated protein immobilization was demonstrated. To generate a reactive surface for MTG, a 96-well polystyrene microtiter plate was physically coated with casein, a good MTG substrate. Successful immobilization of recombinant AP to the nanolayer of casein on the surface of the microtiter plate was verified by the detection of enzymatic activity. Since little activity was observed when wild-type AP was used, immobilization of NK-AP was likely directed by the specific peptide tag. When polymeric casein prepared by MTG was used as a matrix on the plate, the loading capacity of AP was increased about 2-fold compared to when casein was used as the matrix. Transglutaminase-mediated site-specific posttranslational modification of proteins offers one way of generating a variety of protein-based solid formulations for biotechnological applications. 1. Introduction Protein-based devices have attracted much attention owing to their abilities such as strict molecular recognition and efficient signal amplification. In recent years, a number of studies have focused on controlled manipulation of protein molecules, because the handling of protein molecules without disruption of their function is a prerequisite for the construction of designed biomaterials. Protein immobilization has been a promising route for making the best use of protein functions. To avoid protein denaturation upon immobilization, site-specific immobilization of target proteins has been studied in the preparation of immobilized enzymes and is currently being studied in the development of protein microarrays.1,2 Most current approaches are directed to the combination of chemical and genetic manipulations to obtain functional protein formulations. Noncovalent and selective immobilization can be achieved by (i) conjugation of peptide tags to target proteins, such as the hexahistidine tag,3 an epitope tag,4 and an elastin-like polypeptide,5 (ii) immobilization of a capture protein like protein A for antibody immobilization6 or an antibody itself,7 and (iii) use of avidin-biotin interactions in which the intein-mediated sitespecific biotinylation at the C-terminus of target proteins8 or Sfp-catalyzed biotin labeling of a peptide carrier protein fused to target proteins9 ensures control of the molecular orientation of proteins.8,9 On the other hand, covalent and selective immobilization of proteins seems to be rather * To whom correspondence should be addressed. E-mail: noritcm@ mbox.nc.kyushu-u.ac.jp (N.K.);
[email protected] (M.G.). Tel: +81-92-642-3575. † Kyushu University. ‡ PRESTO.
difficult. Excellent approaches for that purpose have recently been demonstrated using an enzyme as the fusion protein of target proteins.10,11 Enzymes are capable of recognizing their substrates with high specificity and some substrates irreversibly form covalent bonds with reactive functional groups involved in the catalytic process. On the basis of the nature of enzymatic reaction, covalent and selective protein immobilization was demonstrated using a chemically derivatized surface with a designed substrate analogue, which is directed to the active site of enzymes fused to the proteins of interest.10,11 In this study, we employed an enzyme for covalent protein immobilization as well, but not as a tab for sticking. In a living organism, proteins are spatially organized at the molecular level according to their function, and posttranslational protein modification plays an important role in their organization. We have recently demonstrated that sequence-specific posttranslational modification catalyzed by microbial transglutaminase (MTG) is applicable to sitespecific cross-linking of recombinant proteins in solution.12-14 Here, we report on a feasible method for covalent and sitespecific immobilization of proteins using MTG. As a model protein to immobilize, E. coli alkaline phosphatase (AP) was selected and a peptide tag that is specifically recognized by MTG was genetically incorporated at the N-terminus. For a solid support, we chose a 96-well polystyrene microtiter plate that is routinely employed for biosensing applications (e.g., an enzyme-linked immunosorbent assay: ELISA). Consequently, the microplate format allows high throughput screening of the optimal reaction conditions for the MTGmediated protein immobilization. To generate a solid surface displaying the recognition sites of MTG, the wells were
10.1021/bm0494895 CCC: $30.25 © 2005 American Chemical Society Published on Web 12/04/2004
36
Biomacromolecules, Vol. 6, No. 1, 2005
physically coated with casein, a good substrate for MTG as well as a well-known surface-blocking reagent in ELISA. It was found that the casein-coated plastic plate effectively prevents nonspecific adsorption of proteins as anticipated, whereas selective immobilization of the recombinant AP onto the surface is permitted. The increase in the loading capacity of the recombinant AP by the polymerization of casein matrixes was also investigated. The present study provides us with an opportunity to carry out a controlled protein immobilization that is required to design a range of current solid protein formulations. 2. Materials and Methods 2.1. Materials. MTG was provided by Ajinomoto Co., Inc (Japan). Wild-type AP was purchased from Wako (Japan). The plasmid encoding AP gene was obtained as previously described.14 Bovine casein (sodium salt) was purchased from Sigma-Aldrich (USA). All other reagents were of commercially available analytical grade. 2.2. Preparation of a Specific Peptide-Fused Recombinant AP. A specific peptide consisting of six amino acids (MKHKGS, abbreviated as K-tag) was genetically attached to the N-terminus of AP. The resultant N-terminal K-tagged AP is abbreviated as NK-AP. The plasmid vector construction for the expression of NK-AP in E. coli was described elsewhere in detail.15 NK-AP was expressed in E. coli strain BL21(DE3) (Novagen) and was purified with a hexahistidine tag (His-tag) attached to the C-terminus of NK-AP. The purified protein was employed as NK-AP. 2.3. Casein Coating on a Polystyrene Plate. The aqueous solution of casein (5 mg/mL) was prepared with 50 mM Tris-HCl buffer (pH 7.5). The casein solution was added to each well of a 96-well polystyrene plate (100 µL) and incubated around 20 °C for 16 h. Prior to the immobilization of NK-AP with MTG, the wells were washed with PBST (phosphate buffer saline plus 0.05 vol % Tween20) three times, to wash out any residual soluble casein fraction. MTGcatalyzed polymerization of casein was conducted by incubating casein (5 mg/mL) and MTG in 50 mM Tris-HCl buffer for 2 h around 20 °C. The polymerization of casein with MTG was checked by SDS-PAGE and the resultant polymeric products were employed as polymeric casein. The apparent molecular weight of the polymeric casein used was estimated by the gel filtration analyses with Superdex200 and high-molecular-weight gel filtration calibration kit (Amersham Biosciences). The polymeric casein was administrated to the wells in the same manner as casein. 2.4. Immobilization of NK-AP on the Casein-Coated Plates. The immobilization of wild-type AP and NK-AP on the casein-coated plates was carried out using aqueous solutions of wild-type AP or NK-AP (0.05 mg/mL) and MTG (1 mg/mL) in 50 mM Tris-HCl buffer (pH 7.5). The enzymatic immobilization was conducted with NK-AP (2.5 µg/well) and MTG (50 µg/well) around 25 °C for 2 h. The total volume of the reaction mixture in each well was 100 µL. Physical adsorption of wild-type AP and NK-AP was also investigated in the absence of MTG but otherwise under the same conditions. After both enzymatic and physical
Communications
immobilization procedures, the wells were washed with PBST three times prior to the activity measurements. 2.5. Measurement of Catalytic Activity of Immobilized AP. The catalytic activity of immobilized AP was measured with p-nitrophenyl phosphate as the substrate at 25 °C. In 1 M Tris-HCl buffer (pH 8.0), hydrolysis of the substrate (1 mM) was followed by an increase in the absorbance at 410 nm (derived from p-nitrophenol) using a 96-well microtiter plate reader (PowerWave X, Bio-Tek Instruments Inc., USA). In a typical experiment, the p-nitrophenol formation was periodically checked over 2 h and the catalytic activity was evaluated by the change in the absorbance at 410 nm per hour (∆Abs/h). The activity measurements were done at least in triplicate. 2.6. Estimation of the Loading Capacity of NK-AP. The loading capacity of NK-AP onto the plastic surface was estimated on the basis of the change in the catalytic activity of immobilized AP upon varying the concentration of NK-AP applied to the wells. Experimental data obtained were fitted to the Langmuir adsorption isotherm θ ) K [NK-AP]/(1 + K [NK-AP]) where θ is the fraction of surface sites occupied by NK-AP upon the immobilization and K is the equilibrium constant of adsorption. Since θ equals q/qm (where q is the quantity of NK-AP immobilized and qm is the maximum surface coverage by NK-AP), analysis of the experimental data by the above equation yields the maximum loading quantity of NK-AP (qm) on a casein-coated surface. The data fitting was carried out using the nonlinear least-squares method with KaleidaGraph software (Synergy Software) on a personal computer. The surface coverage area of casein at the watertoluene interface was determined by an interfacial tensiometer (TVT2S, LAUDA, Germany). 3. Results and Discussion 3.1. Immobilization of NK-AP onto a Plastic Plate. To conduct covalent and site-specific protein immobilization, we focused on utilizing an intriguing biological event, the posttranslational protein cross-linking reaction. An enzyme involved in this reaction, transglutaminase (TG), catalyzes isopeptide bond formation between the side chains of Gln and Lys residues that fit the substrate specificity of this enzyme.16,17 One can imagine that if we could introduce TG recognition sites at desired positions in both the target proteins and the solid surfaces, they could be easily, covalently, and site-specifically cross-linked by TG. To demonstrate this idea, we separately prepared either a recombinant protein to immobilize or a specific solid surface to be immobilized. Microbial transglutaminase (MTG, 18) was employed as the catalyst for protein immobilization in this study. We selected AP as a model protein because it did not suffer from significant loss of native activity upon the incorporation of peptide tags at the N-terminus.4,14 In a preliminary experiment, a short peptide, MKHKGS, was found to work as a Lys-donor in the conjugation of NK-AP with casein in an aqueous solution.15 For a solid surface displaying reactive
Communications
Figure 1. Immobilization of AP on the casein-coated microtiter plate. Results obtained for a bare surface (lanes 1 and 2) and the solid surfaces coated with casein (lanes 3-7) or polymeric casein (lanes 8-10) are shown. The wells were treated with: wild-type AP (lanes 1 and 4); wild-type AP and MTG (lane 5); NK-AP (lanes 2, 6 and 9); NK-AP and MTG (lanes 7 and 10). The concentration of NK-AP applied to each well was 2.5 µg. Background levels of p-nitrophenol formation are shown in lanes 3 (with casein) and 8 (with polymeric casein).
Gln residues, we focused on casein, which is widely used as a surface-blocking reagent in ELISA to prevent nonspecific protein adsorption on microplates. It is known that casein adsorbs stably onto a polystyrene surface by van der Waals, hydrophobic, and hydrogen-bonding interactions,1 although the detailed mechanism has not yet been determined.19 We assumed that a polystyrene surface covered by physically adsorbed casein molecules can be regarded as a stable solid surface displaying reactive Gln residues for the immobilization of NK-AP. Hence, NK-AP immobilization on the casein nanolayer generated on the plastic surface was investigated. Figure 1 shows the detection of catalytic activity of wildtype AP or NK-AP on a bare and a casein-coated polystyrene plates after immobilization procedures under different reaction conditions. Nonspecific adsorption of both wild-type AP and NK-AP on a bare polystyrene plate was observed (lanes 1 and 2 in Figure 1). In contrast, as shown in lanes 4 and 6, the residual catalytic activities were negligible, suggesting that surface coating of the plate with casein effectively prevents physical adsorption of both proteins. Comparison of the results obtained with wild-type AP and NK-AP in the presence of MTG (lanes 5 and 7 in Figure 1) shows that a significantly higher amount of p-nitrophenol formation was observed for the latter, indicating covalent immobilization of NK-AP through the specific peptide at the N-terminus. The accessibility of NK-AP to casein physically adsorbed on the solid surface seems rather difficult due to steric hindrance; however, it was found that such a short peptide tag consisting of six amino acids could facilitate the immobilization of NK-AP. These results clearly indicate the potential utility of MTG in covalent and site-specific protein immobilization of recombinant proteins to which a specific peptide for MTG is genetically incorporated. 3.2. Increasing the Loading Capacity Using Polymeric Casein as a Matrix for Immobilization. In an attempt to increase the amount of immobilized NK-AP, we tried to create a 3D scaffold composed of casein. The polymeric casein was readily prepared by incubating casein with MTG
Biomacromolecules, Vol. 6, No. 1, 2005 37
Figure 2. Relationship between the NK-AP concentration used during the immobilization process and the AP activity derived from the immobilized NK-AP. Results obtained for a casein-coated surface (open circles) and a polymeric casein-coated surface (closed circles) are shown. The wells were treated with NK-AP and MTG for 2 h. The solid lines in the figure represent the results from fitting of the experimental data to the Langmuir adsorption isotherm (R ) 0.999 in both cases). See details in Materials and Methods.
in solution and was subsequently adsorbed onto the plate surface. The apparent molecular weight of the polymeric casein used was determined to be approximately 450 kDa according to the results of gel filtration analyses. The number of casein units in polymeric casein was thus estimated to be about 20, based on the apparent molecular weight. As shown in Figure 1 (lanes 7 and 10), the polymeric casein allows about 4-fold higher loading of NK-AP on the solid surface in comparison to casein itself. The results seem to be contradictory, because polymerization of casein by MTG could reduce the number of reactive Gln residues in the immobilization process. However, the higher loading of NK-AP with the coating of polymeric casein implies an increase in the total mass of casein on the plate surface, which results in an increase in the number of Gln residues accessible to MTG on the plate surface. 3.3. Characterization of Protein Immobilization on Casein-Coated Surface. Figure 2 depicts the relationship between the concentration of NK-AP used during the immobilization process and the activity detected after immobilization. The experimental data shows a saturation profile. For MTG, NK-AP is one of the substrates in a well and the product is NK-AP-casein conjugate on the plate surface. Given that the activity of immobilized NK-AP (i.e., y axis of Figure 2) is proportional to the amount of immobilized NK-AP, the saturation profile could be connected to the Langmuir adsorption isotherm (see the Supporting Information). Assuming that the catalytic activity of the immobilized NK-AP was comparable to that of NK-AP in solution [the activity-concentration profile of AP in solution showed a linear dependence over the range of AP activity shown in Figure 2 (data not shown)], the detected AP activity could be converted to the amount of NK-AP immobilized on the surface. Although the above assumption would underestimate the loading capacity, experimental data obtained showed a good fit to the Langmuir isotherm in each case (solid lines in Figure 2). The maximum loading quantity
38
Biomacromolecules, Vol. 6, No. 1, 2005
of NK-AP (corresponding to the qm values in the analyses) to casein- and polymeric casein-coated surfaces were estimated to be 35 and 65 ng/well, respectively. An approximately 2-fold increase in qm with polymeric casein may reflect the difference in the number of reactive Gln residues accessible to both MTG and NK-AP. For the casein-coated plate, the number of immobilized NK-AP molecules was roughly calculated to be one NK-AP molecule per 250 molecules of casein, based on the surface coverage area of casein at the water-toluene interface (ca. 2.5 × 105 m2/mol). In terms of the equilibrium constant (K), it was found that polymeric casein-coated surfaces showed a slightly larger value than casein-coated ones [the values for casein- and polymeric casein-coated surfaces were 0.73 × 10-2 and 1.3 × 10-2 (µg/well)-1, respectively], implying that the type of casein matrix affected the affinity for the target protein. These results suggest that suitable modulation of solid surfaces at the molecular level can increase the capacity for MTGmediated protein immobilization.
Communications
partly by the 21st Century COE Program, “Functional Innovation of Molecular Informatics” from the MEXT of Japan (to M.G.). Supporting Information Available. Connection of the saturation profile to the Langmuir adsorption isotherm. This material is available free of charge via the Internet at http:// pubs.acs.org. References and Notes (1) (2) (3) (4) (5) (6) (7) (8)
4. Conclusions
(9)
In this manuscript, we described a new methodology for the construction of solid protein-based formulations from the viewpoint of enzyme engineering. The experimental protocol is quite simple and can be further extended to a range of applications if both a target protein and the surface of supports are made available to transglutaminase. Optimization of the position of insertion of specific peptides into target proteins and derivatization of different types of solid supports suitable for MTG-mediated protein immobilization are now underway in our group.
(10)
Acknowledgment. We are grateful to Ajinomoto Co. Inc., for providing samples of MTG. The present work was supported mainly by a Grant-in-Aid for Scientific Research (No. 16760638) from the Ministry of Education, Culture, Science, Sports and Technology (MEXT) of Japan and Nagase Science and Technology Foundation (to N.K.) and
(11) (12) (13) (14) (15) (16) (17) (18) (19)
Wilson, D. S.; Nock, S. Curr. Opin. Chem. Biol. 2002, 6, 81-85. Kodadek, T. Chem. Biol. 2001, 8, 105-115. Zhang, J. K.; Cass, A. E. G. Anal. Biochem. 2001, 292, 307-310. Vishwanath, S. K.; Watson, C. R.; Huang, W.; Bachas, L. G.; Bhattacharyya, D. J. Chem. Technol. Biotechnol. 1997, 68, 294302. Nath, N.; Chilkoti, A. Anal. Chem. 2003, 75, 709-715. Johnson, C. P.; Jensen, I. E.; Prakasam, A.; Vijayendran, R.; Leckband, D. Bioconjugate Chem. 2003, 14, 974-978. Koepsel, R. R.; Russell, A. J. Biomacromolecules 2003, 4, 850855. Lesaicherre, M.-L.; Lue, R. Y. P.; Chen, G. Y. J.; Zhu, Q.; Yao, S. Q. J. Am. Chem. Soc. 2002, 124, 8768-8769. Yin, Y.; Liu, F.; Li, X.; Walsh, C. T. J. Am. Chem. Soc. 2004, 126, 7754-7755. Hodneland, C. D.; Lee, Y.-S.; Min, D.-H.; Mrksich, M. Proc. Natl. Acad. Sci. U.S.A. 2002, 99, 5048-5052. Kindermann, M.; George, N.; Johnsson, N.; Johnsson, K. J. Am. Chem. Soc. 2003, 125, 7810-7811. Kamiya, N.; Tanaka, T.; Suzuki, T.; Takazawa, T.; Takeda, S.; Watanabe, K.; Nagamune, T. Bioconjugate Chem. 2003, 14, 351357. Tanaka, T.; Kamiya, N.; Nagamune, T. Bioconjugate Chem. 2004, 15, 491-497. Takazawa, T.; Kamiya, N.; Ueda, H.; Nagamune, T. Biotechnol. Bioeng. 2004, 86, 399-404. Tominaga, J.; Kamiya, N.; Doi, S.; Ichinose, H.; Goto, M. Enzyme Microb. Technol. 2004, 35, 613-618. Lorand, L.; Graham, R. M. Nature ReV. Mol. Cell Biol. 2003, 4, 140-156. Griffin, M.; Casadio, R.; Bergamini, C. M. Biochem. J. 2002, 368, 377-396. Kashiwagi, A.; Yokoyama, K.; Ishikawa, K.; Ono, K.; Ejima, D.; Matsui, H.; Suzuki, E. J. Biol. Chem. 2002, 277, 44252-44260. Butler, J. E.; Solid phases in immunoassay. In Enzyme immunoassay; Academic Press: New York, 1996; Chapter 9, pp 205-225.
BM0494895
