High-Affinity Tags Fused to S-Layer Proteins Probed by Atomic Force

Institute of Biophysics, Johannes Kepler UniVersity of Linz, 4040 Linz, Austria, and Center for. NanoBiotechnology, UniVersity of Natural Resources an...
0 downloads 0 Views 440KB Size
Download PDF
1324

Langmuir 2008, 24, 1324-1329

High-Affinity Tags Fused to S-Layer Proteins Probed by Atomic Force Microscopy† Jilin Tang,‡ Andreas Ebner,‡ Nicola Ilk,§ Helga Lichtblau,§ Carina Huber,§ Rong Zhu,‡ Dietmar Pum,§ Micheal Leitner,‡ Vassili Pastushenko,‡ Hermann J. Gruber,‡ Uwe B. Sleytr,§ and Peter Hinterdorfer*,‡ Institute of Biophysics, Johannes Kepler UniVersity of Linz, 4040 Linz, Austria, and Center for NanoBiotechnology, UniVersity of Natural Resources and Applied Life Sciences, 1180 Vienna, Austria ReceiVed July 27, 2007. In Final Form: September 14, 2007

Two-dimensional, crystalline bacterial cell surface layers, termed S-layers, are one of the most commonly observed cell surface structures of prokaryotic organisms. In the present study, genetically modified S-layer protein SbpA of Bacillus sphaericus CCM 2177 carrying the short affinity peptide Strep-tag I or Strep-tag II at the C terminus was used to generate a 2D crystalline monomolecular protein lattice on a silicon surface. Because of the genetic modification, the 2D crystals were addressable via Strep-tag through streptavidin molecules. Atomic force microscopy (AFM) was used to investigate the topography of the single-molecules array and the functionality of the fused Strep-tags. In high-resolution imaging under near-physiological conditions, structural details such as protein alignment and spacing were resolved. By applying molecular recognition force microscopy, the Strep-tag moieties were proven to be fully functional and accessible. For this purpose, streptavidin molecules were tethered to AFM tips via ∼8-nm-long flexible polyethylene glycol (PEG) linkers. These functionalized tips showed specific interactions with 2D protein crystals containing either the Strep-tag I or Strep-tag II, with similar energetic and kinetic behavior in both cases.

Introduction The self-assembly of biomolecules is one of the most fascinating fields of nanobiotechnology. The preparation of nanostructures by self-assembly is often referred to as “bottomup” nanotechnology. In contrast to “top-down” techniques such as lithography, bottom-up can efficiently build blocks on the lower nanoscale. Nanotechnology offers exciting applications for the self-assembly of multifunctional biomolecules, as exemplified in the present study by fusing a crystallizable protein with widely used peptide tags. Crystalline bacterial cell surface layers (S-layers) are the most common surface structures of prokaryotic organisms.1-3 Isolated S-layer protein maintains its ability to reassemble and recrystallize into oblique, square, or hexagonal lattice symmetry on solid supports, such as silicon wafers, gold chips, and glass, as well as on Langmuir lipid films and liposomes.4,5 The modification of the S-layer protein with Strep-tag was designed in a way that did not interfere with the self-assembly and recrystallization properties of the S-layer protein moiety.6,7 Functional domains such as core streptavidin,8,9 †

Part of the Molecular and Surface Forces special issue. * Corresponding author. E-mail: [email protected]. ‡ Johannes Kepler University of Linz. § University of Natural Resources and Applied Life Sciences. (1) Sa´ra, M.; Sleytr, U. B. J. Bacteriol. 2000, 182, 859. (2) Sleytr, U. B. FEMS Microbiol. ReV. 1997, 20, 5. (3) Sleytr, U. B.; Beveridge, T. J. Trends Microbiol. 1999, 7, 253. (4) Sleytr, U. B.; Egelseer, E. M.; Ilk, N.; Pum, D.; Schuster, B. FEBS J. 2007, 274, 323. (5) Sleytr, U. B.; Huber, C.; Ilk, N.; Pum, D.; Schuster, B.; Egelseer, E. M. FEMS Microbiol. Lett. 2007, 267, 131. (6) Ilk, N.; Vo¨llenkle, C.; Egelseer, E. M.; Breitwieser, A.; Sleytr, U. B.; Sa´ra, M. Appl. EnViron. Microbiol. 2002, 68, 3251. (7) Huber, C.; Ilk, N.; Runzler, D.; Egelseer, E. M.; Weigert, S.; Sleytr, U. B.; Sa´ra, M. Mol. Microbiol. 2005, 55, 197. (8) Moll, D.; Huber, C.; Schlegel, B.; Pum, D.; Sleytr, U. B.; Sa´ra, M. Proc. Natl. Acad. Sci. U.S.A. 2002, 99, 14646. (9) Ebner, A.; Kienberger, F.; Huber, C.; Kamruzzahan, A. S.; Pastushenko, V. P.; Tang, J.; Kada, G.; Gruber, H. J.; Sleytr, U. B.; Sara, M.; Hinterdorfer, P. ChemBioChem 2006, 7, 588.

heavy-chain camel antibodies,10,11 IgG binding domains, and the major birch pollen allergen Bet v16 fused to the C-terminal end of S-layer protein SbpA of B. sphaericus CCM 2177 were used as binding sites for biologically active molecules. Functional S-layer fusion proteins recrystallized on solid supports were exploited as a new type of sensing layer in label-free detection systems such as SPR and QCMD,10,11 as microadsorbents in extracorporeal blood purification systems,12and for the development of novel types of antiallergic vaccines.6,13,14 S-layer protein SbpA of Bacillus sphaericus CCM 2177 with a molecular mass of 130 kDa is one of the most extensively studied S-layer proteins.6,7,10-12,15 It is generated from a 1268 amino acid-long protein precursor by the cleavage of a 30 amino acid-long signal peptide, whereupon it self-assembles into a square (p4) lattice6,16 To add functional sites, Strep-tags were fused to the S-layer by genetic engineering (Figure 1). Strep-tag I (AWRHPQFGG) and Strep-tag II (SNWSHPQFEK) are nine amino acid peptides that act as artificial ligands for streptavidin.17,18 SbpA was fused to a sequence encoding the Strep-tag I, forming the rSbpA-Strep-tag I protein. Another short affinity peptide, Strep-tag II, which can be fused not only to the C terminus (10) Pleschberger, M.; Neubauer, A.; Egelseer, E. M.; Weigert, S.; Lindner, B.; Sleytr, U. B.; Muyldermans, S.; Sara, M. Bioconjugate Chem. 2003, 14, 440. (11) Pleschberger, M.; Saerens, D.; Weigert, S.; Sleytr, U. B.; Muyldermans, S.; Sara, M.; Egelseer, E. M. Bioconjugate Chem. 2004, 15, 664. (12) Vo¨llenkle, C.; Weigert, S.; Ilk, N.; Egelseer, E.; Weber, V. F.; Loth Falkenhagen, D.; Sleytr, U. B.; Sa´ra, M. Appl. EnViron. Microbiol. 2004, 70, 1514. (13) Bohle, B.; Breitwieser, A.; Zwolfer, B.; Jahn-Schmid, B.; Sara, M.; Sleytr, U. B.; Ebner, C. J. Immunol. 2004, 172, 6642. (14) Breitwieser, A.; Egelseer, E. M.; Moll, D.; Ilk, N.; Hotzy, C.; Bohle, B.; Ebner, C.; Sleytr, U. B.; Sara, M. Protein Eng. 2002, 15, 243. (15) Gyo¨rvary, E. S.; Stein, O.; Pum, D.; Sleytr, U. B. J. Microsc. 2003, 212, 300. (16) Sleytr, U. B.; Messner, P.; Pum, D.; Sa´ra, M. Angew. Chem., Int. Ed. 1999, 38, 1034. (17) Voss, S.; Skerra, A. Protein Eng. 1997, 10, 975. (18) Schmidt, T. G.; Skerra, A. J. Chromatogr., A 1994, 676, 337.

10.1021/la702276k CCC: $40.75 © 2008 American Chemical Society Published on Web 11/15/2007

High-Affinity Tags Fused to S-Layer Proteins

Figure 1. Schematic drawing of rSbpA-Strep-tags with p4 symmetry. One unit cell is composed of four single S-layer proteins.

but also to the N terminus of the recombinant protein,17,19,20 was fused to SbpA, thereby constituting the protein rSbpA-Strep-tag II. Strep-tag I-modified S-layer proteins have been seen to bind streptavidin specifically.6 Atomic force microscopy (AFM) has been applied for imaging the surface of biological samples21,22 and also for measuring the binding force between ligands and receptors at the single-molecule level.23-25 Single-molecular interaction forces are typically studied in force spectroscopy experiments in which a ligandfunctionalized AFM tip is approaches the surface where the receptor is located. The ligand on the AFM tip can thereby recognize the receptor, resulting in the formation of a ligandreceptor complex. Subsequently, the tip is withdrawn from the surface, and the complex sustains with the increasing force until it dissociates when a critical force, the so-called unbinding force, is reached. In this study, S-layer proteins rSbpA-Strep-tag I and rSbpAStrep-tag II were recrystallized on silicon surfaces. The surface topography of S-layer crystals was investigated by AFM. The accessibility of Strep-tag I and Strep-tag II in the S-layer was proven, and the unbinding forces of streptavidin from the Streptag I and Strep-tag II proteins were studied. Experiment Cloning of the Genes Encoding rSbpA-Strep-tag I and rSbpAStrep-tag II. The gene encoding SbpA derivatives rSbpA31-1268/ Strep-tag I, carrying Strep-tag I at the C terminus, was cloned as described in Ilk et al.6 The gene encoding rSbpA-Strep-tag II, carrying Strep-tag II at the C-terminal end, was amplified by PCR from the chromosomal DNA of B. sphaericus CCM 2177 using oligonucleotide primer sbpA37 (forward primer) [5′-CGG AAT TCC ATG GCG CAA GTA AAC GAC TAT AAC AAA ATC-3′] and SbpAII/5 (reverse primer) [5′-GA CCG CTC GAG TTA GCA ACC ACC ACC TTT TTC GAA CTG CGG GTG AGA CCA ACC ACC TTC TGA ATA TGC AGT AGT TGC TGC-3′]. Primer SbpAII/5 introduced the sequence of Strep-tag II (italic) at the 3′end of the coding sequence of sbpA. For cloning, NcoI and XhoI restriction sites were introduced during PCR at the 5′and 3′ ends, respectively, and the gel-purified PCR fragment was inserted into the corresponding restrictions sites of plasmid pET28a, which was established in Escherichia coli TG1. (19) Schmidt, T. G. M.; Koepke, J.; Frank, R.; Skerra, A. J. Mol. Biol. 1996, 255, 753. (20) Korndo¨rfer, I. P.; Skerra, A. Protein Sci. 2002, 11, 883. (21) Kienberger, F.; Mueller, H.; Pastushenko, V.; Hinterdorfer, P. EMBO Rep. 2004, 5, 579. (22) Scheuring, S.; Ringler, P.; Borgnia, M.; Stahlberg, H.; Muller, D. J.; Agre, P.; Engel, A. EMBO J. 1999, 18, 4981. (23) Nevo, R.; Brumfeld, V.; Elbaum, M.; Hinterdorfer, P.; Reich, Z. Biophys. J. 2004, 87, 2630. (24) Moy, V. T.; Florin, E.-L.; Gaub, H. E. Science 1994, 266, 257. (25) Hinterdorfer, P.; Baumgartner, W.; Gruber, H. J.; Schilcher, K.; Schindler, H. Proc. Natl. Acad. Sci. U.S.A. 1996, 93, 3477.

Langmuir, Vol. 24, No. 4, 2008 1325 Expression of rSbpA-Strep-tag I and rSbpA-Strep-tag II. The heterologous expression of the genes encoding rSbpA-Strep-tag I and rSbpA-Strep-tag II in E. coli HMS(174)DE3 cultures was done as described in Ilk et al.6 Isolation and Purification of rSbpA-Strep-tag I and rSbpAStrep-tag II. The isolation of both rSbpA derivatives was performed as described by Jarosch et al.26 For purification, aliquots (120 mg) of the lyophilized recombinant S-layer protein forms were suspended in 5 mL of 2 M guanidine hydrochloride (GHCl) that also contained 50 mM Tris-HCl (pH 7.2, buffer A). After centrifugation at 16 000g for 5 min at 4 °C, the supernatants were filtered through a 0.45 µm RC membrane (Minisart RC 25), and the clear solutions were subjected to gel permeation chromatography (GPC) using a Superdex 200 column (Pharmacia) and equilibrated in buffer A for separation. Fractions containing S-layer protein were pooled and dialyzed against distilled water for 18 h at 4 °C, lyophilized, and stored at -20 °C. The purity of rSbpA-Strep-tag I and rSbpA-Strep-tag II was finally checked by SDS-PAGE. Recrystallization of S-Layer Fusion Proteins. One milligram of lyophilized protein (rSbpA-Strep-tag I and rSbpA-Strep-tag II) was dissolved in buffer A and dialyzed against Milli-Q water for 3 h. After centrifugation, the protein solution was diluted to 0.1 mg/mL with 0.1 mM CaCl2, 0.5 mM Tris buffer (pH 9). Silicon wafers (Si-Mat Silicon Materials, Germany) were cut into 8 × 8 mm2 pieces. The silicon pieces were cleaned with 7:1 H2SO4/H2O2 (Caution! piranha is extremely corrosiVe and potentially explosiVe.) for 30 min and washed with Milli-Q water three times. The cleaned silicon was incubated with S-layer protein solution (0.1 mg/mL) for 4 h, washed with Milli-Q water, and stored in Tris buffer (0.5 mM Tris, 150 mM NaCl, pH 7.5) at 4 °C. Functionalization of AFM Tips with Streptavidin. AFM tips (Si3N4, Veeco) for force microscopy were functionalized with streptavidin (Sigma) as previously described.27 In brief, after cleaning with ethanol and chloroform, AFM tips were incubated with 0.55 g/mL ethanolamine hydrochloride (Sigma) in DMSO overnight at room temperature, resulting in amino groups on the tip surface.28 The resulting amino groups were reacted with the NHS ester terminus of an ∼8-nm-long heterobifunctional PEG chain, and the benzaldehyde function on the free end of the PEG linker was used for the covalent binding of streptavidin via its lysine residue.27 For the coupling of streptavidin, tips with the functional aldehyde group were incubated with 200 µg/mL streptavidin in pH 7.5 PBS buffer (1.8 mM KH2PO4, 10.1 mM Na2HPO4, 2.7 mM KCl ,and 140 mM NaCl), together with 10 mM NaCNBH3 for 1 h at room temperature.27 The AFM tips were washed PBS (3×) and stored in PBS at 4 °C until use pH 7.5 (up to 3 days). AFM Measurements. AFM measurements were performed either on a PicoSPM or on a Pico Plus setup (Molecular Imaging, Tempe, AZ). AFM tips without a magnetic coating (Veeco, Santa Barbara, CA) with a nominal spring constant of 0.03-0.1 N/m were used for contact mode imaging. Magnetically coated AFM cantilevers with a nominal spring constant of 0.1 N/m (Molecular Imaging, Tempe, AZ) were used for MAC mode image. The measurement frequency was set to 20% below the resonance frequency. All measurements were carried out in the buffer containing 0.5 mM Tris, 150 mM NaCl (pH adjusted to 7.5 with HCl). The scan speed for imaging was 1 line/s at 512 data points per line. The spring constants of the cantilevers were determined in thermal-noise mode.29 Loading rates r were calculated from r ) Vkeff, with V being the pulling velocity and keff being the effective spring constant.30 (26) Jarosch, M.; Egelseer, E. M.; Huber, C.; Moll, D.; Mattanovich, D.; Sleytr, U. B.; Sara, M. Microbiology 2001, 147, 1353. (27) Ebner, A.; Wildling, L.; Kamruzzahan, A. S.; Rankl, C.; Wruss, J.; Hahn, C. D.; Holzl, M.; Zhu, R.; Kienberger, F.; Blaas, D.; Hinterdorfer, P.; Gruber, H. J. Bioconjugate Chem. 2007, 18, 1176. (28) Riener, C. K.; Stroh, C. M.; Ebner, A.; Klampfl, C.; Gall, A. A.; Romanin, C.; Lyubchenko, Y. L.; Hinterdorfer, P.; Gruber, H. J. Anal. Chim. Acta 2003, 479, 59. (29) Hutter, J. L.; Bechhoefer, J. ReV. Sci. Instrum. 1993, 64, 1868. (30) De Paris, R.; Strunz, T.; Oroszlan, K.; Gu¨therodt, H.-J.; Hegner, M. Single Mol. 2000, 1, 285.

1326 Langmuir, Vol. 24, No. 4, 2008

Tang et al.

Figure 2. Topographical images of S-layer protein rSbpA-Strep-tag II recrystallized on silicon. The cross-section profile along the white line shows the height of the S-layer crystal. The scale bars are (A) 300 and (B) 100 nm. The streptavidin-modified AFM cantilevers were blocked by the addition of 1 mM free biotin and subsequent incubation for about 1 h. Data Evaluation. For each data set shown in Figure 4C (solid and dashed lines) and D (solid and dashed lines), 500-1000 force distance cycles were recorded. The cycles, which show an unbinding event, were collected to build the histograms. To generate the probability density function of unbinding forces, the data were calculated according to their standard deviations. For each data point, the unbinding force and the corresponding standard deviation were used to construct a Gaussian distribution. The Gaussian distribution was summed to give the final probability density function.31 Image Processing. The signal-to-noise ratio in the images was enhanced by Fourier domain filtering using a digital filter mask. The shape of the filter holes was attenuated by a cos2 function, and the size was predefined as 0.2 of the geometric mean of the base vector lengths. The space group symmetry was determined from the Fourier coefficients. The image processing software was developed in house.

Results and Discussion AFM Imaging of rSbpA-Strep-tag II and rSbpA-Strep-tag I Proteins. The topographical images of the rSbpA-Strep-tag II protein layer recrystallized on silicon are shown in Figure 2. In the large scan area, the silicon surface was fully covered by an rSbpA-Strep-tag II protein layer with a thickness of 8 ( 2 nm (Figure 2A), indicating that rSbpA-Strep-tag II was recrystallized into a monolayer on the silicon surface.15 By enlarging smaller areas, more detailed information was observed in the topographical image (Figure 2B). The square lattice constants were determined to be a ) 13.93 ( 0.34 nm and b ) 13.84 ( 0.37 nm, which is typical for the SbpA S-layer protein.32,33 An overview topographical image of rSbpA-Strep-tag I protein recrystallized on silicon is shown in Figure 3A. The scan area was almost completely covered by crystals of rSbpA-Strep-tag I protein, with a thickness of 8 ( 2 nm indicating only a monolayer of protein on the silicon surface.15 The square lattice of rSbpA(31) Baumgartner, W.; Hinterdorfer, P.; Schindler, H. Ultramicroscopy 2000, 82, 85. (32) Martin-Molina, A.; Moreno-Flores, S.; Perez, E.; Pum, D.; Sleytr, U. B.; Toca-Herrera, J. L. Biophys. J. 2006, 90, 1821. (33) Toca-Herrera, J. L.; Moreno-Flores, S.; Freiedmann, J.; Pum, D.; Sleytr, U. B. Microsc. Res. Tech. 2004, 65, 226.

Strep-tag I was clearly visible. In the smaller scan size, the square lattice became clearly visible (Figure 3B). The square lattice parameters were a ) 13.48 ( 0.32 nm and b ) 13.95 ( 0.43 nm, in good agreement with the values of S-layer protein SbpA.32,33 The high-resolution topographical image of the rSbpA-Streptag I in Figure 3C was obtained by using contact mode. The recombinant rSbpA-Strep-tag I fusion protein also shows p4 lattice symmetry, although the lattice structure displayed in Figure 3C is slightly distorted (aspect ratio of the base vectors, 0.85; base angle, 84°). The four proteins constituting the morphological unit form a square, and the morphological units are interlinked. Figure 3C shows detailed information for one unit cell. The edge between two cell units is clearly visible. Digitized images were processed when several high-resolution spots were found beyond a 1/3.0 nm-1 threshold. The pronounced furrows running from left to right (slightly tilted from the horizontal) might be an effect caused by the scanning direction and are further enhanced by image processing (Figure 3C inset). The AFM results show that both rSbpA-Strep-tag I and rSbpAStrep-tag II could well be recrystallized on the silicon surface, forming a crystalline protein monolayer. The topography characteristics of rSbpA-Strep-tag I and rSbpA-Strep-tag II are very similar (Figures 2 and 3). Force Microscopy Studies of rSbpA-Strep-tag I and rSbpAStrep-tag II Proteins. Strep-tag I and Strep-tag II are artificial peptide ligands with high specificity for streptavidin and are regularly used as affinity handles for the purification and detection of recombinant fusion proteins.17 For the functionalization of S-layer protein SbpA of B. sphaericus CCM2177, Step-tag I or Strep-tag II was fused to the C-terminal end of the S-layer protein. In the present study, the accessibility of Strep-tag I or Streptag II for streptavidin in the S-layer lattice was investigated by single-molecule force microscopy. Streptavidin was bound to the AFM tips via a polyethylene glycol (PEG) linker using wellestablished surface chemistry (Figure 4A).27 The functionalized tip was used to detect the Strep-tag I and Strep-tag II on the surface of the S-layer lattice on silicon. Figure 4B shows a typical force distance cycle with a binding event between streptavidin and Strep-tag I. The tips modified with streptavidin approached

High-Affinity Tags Fused to S-Layer Proteins

Langmuir, Vol. 24, No. 4, 2008 1327

Figure 3. Topographical images of S-layer protien rSbpA-Strep-tag I recrystallized on silicon. The cross-section profile along the white line shows the height of the S-layer crystal. The inset of C shows a signal-to-noise ratio-enhanced enlarged view of the rSbpA-Strep-tag I layer. The scale bars are (A) 200, (B) 100, and (C) 100 nm.

the S-layer surface, allowing for association between streptavidin and Strep-tag I or between streptavidin and Strep-tag II. Then, an increasing force was applied to the specific bond by pulling on the complex until dissociation occurred when a critical force, the unbinding force fu, was reached. The subsequent addition of free biotin caused almost all binding events in the force distance cycles to disappear, thereby proving the specificity of the interaction between Strep-tag I and streptavidin (Figure 4B inset). The distribution of unbinding forces between Strep-tag I on the monolayer of rSbpA-Strep-tag I and streptavidin is shown in Figure 4C. The most probable unbinding force (the maximum of the distribution) between Strep-tag I and streptavidin was about 45 pN at a loading rate of 1.35 nN s-1, with a binding probability of about 23%. After blocking with free biotin, the binding probability dramatically decreased to about 2%. Thus, Strep-tag I in the protein crystal lattice is accessible for the specific binding of streptavidin. To minimize experimental errors, the accessibility of Strep-tag II in rSbpA-Strep-tag II was probed by the same streptavidin-modified tip as that are used for rSbpA-

Strep-tag I. The unbinding force between Strep-tag II and streptavidin was about 40 pN at a loading rate of 1.31 nN s-1, as shown in Figure 4D. The binding probability was about 13%. After streptavidin was blocked with free biotin, the binding probability decreased to 2.6%, again indicating the specificity of interaction. A second maximum at 68 and 70 pN can be observed in Figure 4C,D, respectively. The 2-fold unbinding force of the second peak was most likely the result of unbinding two Strep-tag-streptavidin interactions, probably by two streptavidin molecules on the apex of the AFM tip. It has been reported that the receptor-ligand unbinding forces fu depend not only on the molecules themselves but also on the loading rate r of the experiment. In the case of a single energy barrier, the unbinding forces fu rise linearly with the logarithm of the loading rate.34,35 The unbinding force fu is related to the loading rate r through eq 1 (34) Evans, E.; Ritchie, K. Biophys. J. 1997, 72, 1541. (35) Merkel, R.; Nassoy, P.; Leung, A.; Ritchie, K.; Evans, E. Nature 1999, 397, 50.

1328 Langmuir, Vol. 24, No. 4, 2008

Tang et al.

Figure 4. (A) Schematic representation of conjugating streptavidin to the AFM tip. The NHS-PEG-aldehyde cross linker was covalently bound to the ethanolamine-coated tip through the NHS-ester end. Streptavidin was conjugated to the aldehyde end. (B) Typical force-distance cycle for the specific interaction between Strep-tag I and streptavidin. The specific interaction is blocked using free biotin (inset). (C) Distribution of the unbinding force for the interaction of streptavidin and Strep-tag I on crystals of S-layer protein rSbpA-Strep-tag I in the absence (solid line) and presence (dashed line) of biotin. The dotted lines represent the bimodal Gaussian fit at the first and second peaks. The loading rate was 1.35 nN s-1. (D) Distribution of the unbinding force for the interaction of streptavidin and Strep-tag II in crystals of S-layer protein rSbpA-Strep-tag II in the absence (solid line) and presence (dashed line) of biotin. The dotted lines represent the bimodal Gaussian fit at the first and second peaks. The loading rate was 1.31 nN s-1. (E) Plot of the unbinding force between streptavidin and Strep-tag I on crystals of S-layer protein rSbpA-Strep-tag I against the logarithm of the loading rate (solid line) and plot of the unbinding force between streptavidin and Strep-tag II on crystals of S-layer protein rSbpA-Strep-tag II against the logarithm of the loading rate (dashed line).

High-Affinity Tags Fused to S-Layer Proteins

fu )

( )( ) kBT rxβ In xβ kBTkoff

Langmuir, Vol. 24, No. 4, 2008 1329

(1)

where xβ is the separation of the energy barrier from the equilibrium position, koff is the dissociation constant at zero force, and kBT is the thermal energy. fu and r are the most probable unbinding force and the loading rate, respectively. Although Strep-tag I and Strep-tag II showed an affinity for streptavidin, Strep-tag I had to be fused to the C teminus of the recombinant protein to maintain a free peptide carboxylate group, giving rise to a salt bridge with an Arg side chain of streptavidin. For Strep-tag II, the side chain of the penultimate Glu residue instead of the terminal peptide carboxylate group of the Streptag I gave rise to a salt bridge with the Arg side chain of streptavidin.17,19,20 Therefore, Strep-tag II can be fused not only to the C terminus but also to the N terminus of the recombinant protein. To compare the interactions of Strep-tag I and Strep-tag II with streptavidin, the most probable unbinding forces were plotted against the logarithm of the respective loading rate (Figure 4E). The unbinding forces fu of Strep-tag I-streptavidin (top) and Strep-tag II-streptavidin (bottom) increased linearly with the logarithm of the loading rate, which was shown in the semilogarithmic plot in Figure 4E. The unbinding forces fu of Strep-tag I-streptavidin (top) and Strep-tag II-streptavidin (bottom) varied in the range of 30-70 pN, depending on the applied loading rate (Figure 4E). Calculated from the curves in Figure 4E by using eq 1, the separation of the energy barrier from the equilibrium position xβ values and the dissociation constant at zero force koff values were xβ ) 0.286 nm and koff ) 3.15 s-1 for Strep-tag I-streptavidin and xβ ) 0.352 nm and koff ) 2.38 s-1 for Strep-tag II-streptavidin, respectively. These results indicate that the bond strength of Strep-tag I-streptavidin at loading rate of 0.5-5.5 nN s-1 is not significantly different from the bond strength of Strep-tag II-streptavidin, taking into account the experimental error. The unbinding forces of Strep-tag I-streptavidin and Strep-tag II-streptavidin were lower than for biotin-streptavidin in the same loading rate range.35 Biotin-

streptavidin also displayed a lower koff,36 consistent with the higher affinity to streptavidin compared to that to the Streptags.17,19,36

Conclusions We have successfully recrystallized the S-layer proteins, rSbpA-Strep-tag I and rSbpA-Strep-tag II, on the silicon surface. The square lattice clearly was observed on crystals rSbpA-Streptag I and rSbpA-Strep-tag II by high-resolution AFM. The interaction between the Strep-tags and streptavidin was studied at the single-molecule level. The Strep-tags on the surface of the recrystallized S-layer could be specially detected by streptavidinfunctionalized AFM tips. The unbinding force between Streptag I and streptavidin was 45 pN at a loading rate of 1.35 nN s-1. By comparison, the unbinding force between Strep-tag II and streptavidin was 40 pN at a loading rate of 1.31 nN s-1. The results demonstrated that the self-assembly of S-layer Strep-tag proteins gave crystalline nanostructures with exposed functional groups in defined positions and orientations on the nanometer scale. Because of the high reproducibility of crystallization and high affinity for streptavidin, rSbpA-Strep-tag I and rSbpAStrep-tag II are promising building blocks for nanobiotechnology. On top of the single-molecule force spectroscopy experiment, spatial force mapping and TREC, recently developed as a fast recognition imaging method, appear to be promising tools for localizing binding sites on S layers on the nanoscale. Acknowledgment. This work was supported by the FP6 ECSTREP project NASSAP 13523, U.S. Air Force Office of Scientific Research (AFOSR) project FA9550-07-0313, Austrian Science Fund (FWF) project P17170-B10, and the Human Science Frontier Program of the European Union (HFSP project RGP0053). We thank Christian Rankl for help in analyzing the data and Linda Wildling for the synthesis of cross linkers. LA702276K (36) Klumb, L. A.; Chu, V.; Stayton, P. S. Biochemistry 1998, 37, 7657.