Combination of Click Chemistry and Enzymatic Ligation for Stable and Efficient Protein Immobilization for Single-Molecule Force Spectroscopy

Abstract

Open AccessCCS ChemistryRESEARCH ARTICLE1 Feb 2022Combination of Click Chemistry and Enzymatic Ligation for Stable and Efficient Protein Immobilization for Single-Molecule Force Spectroscopy Shengchao Shi, Ziyi Wang, Yibing Deng, Fang Tian, Qingsong Wu and Peng Zheng Shengchao Shi State Key Laboratory of Coordination Chemistry, Chemistry and Biomedicine Innovation Center (ChemBIC), School of Chemistry and Chemical Engineering, Nanjing University, Nanjing 210023 , Ziyi Wang State Key Laboratory of Coordination Chemistry, Chemistry and Biomedicine Innovation Center (ChemBIC), School of Chemistry and Chemical Engineering, Nanjing University, Nanjing 210023 , Yibing Deng State Key Laboratory of Coordination Chemistry, Chemistry and Biomedicine Innovation Center (ChemBIC), School of Chemistry and Chemical Engineering, Nanjing University, Nanjing 210023 , Fang Tian State Key Laboratory of Coordination Chemistry, Chemistry and Biomedicine Innovation Center (ChemBIC), School of Chemistry and Chemical Engineering, Nanjing University, Nanjing 210023 , Qingsong Wu State Key Laboratory of Coordination Chemistry, Chemistry and Biomedicine Innovation Center (ChemBIC), School of Chemistry and Chemical Engineering, Nanjing University, Nanjing 210023 and Peng Zheng *Corresponding author: E-mail Address: [email protected] State Key Laboratory of Coordination Chemistry, Chemistry and Biomedicine Innovation Center (ChemBIC), School of Chemistry and Chemical Engineering, Nanjing University, Nanjing 210023 https://doi.org/10.31635/ccschem.021.202100779 SectionsSupplemental MaterialAboutAbstractPDF ToolsAdd to favoritesDownload CitationsTrack Citations ShareFacebookTwitterLinked InEmail Protein immobilization is an essential method for both basic and applied research for protein, and covalent, site-specific attachment is the most desirable strategy. Classic methods typically rely on a heterobifunctional cross-linker, such as N-hydroxysuccinimide (NHS)-linker-maleimide, or a similar two-step process. It utilizes the amino-reactive NHS and the thiol-reactive maleimide to conjugate protein to the solid support. However, NHS as a chemical is susceptible to hydrolysis during storage and handling, and maleimide reacts nonspecifically with all cysteines available in the protein, leading to an inconsistent result. To solve these problems, we have developed a method by combining a strain-promoted azide–alkyne cycloaddition (SPAAC) click reaction and an OaAEP1(C247A)-based enzymatic ligation. The method was demonstrated by the successful immobilization of enhanced green fluorescent protein (eGFP), which was visualized by fluorescent imaging. Moreover, the correct folding and stability of the immobilized protein were verified by atomic force microscopy-based single-molecule force spectroscopy (AFM-SMFS) measurement with a high success rate (89%). Finally, the strength of the 1,2,3-triazole linkage from the azide-dibenzocyclooctyne (DBCO)-based SPAAC reaction was quantified with an ultrahigh rupture force <1.7 nN. Thus, this stable, efficient, and site-specific immobilization method can be used for many challenging systems, especially SMFS studies. Download figure Download PowerPoint Introduction Protein immobilization is crucial for the characterization and application of many proteins,1,2 such as the study of the protein mechanism, detection, and single-molecule study.3–7 Among different methodologies, site-specific covalent bonding is most desirable, as it offers a reliable and robust protein attachment. This strategy typically relies on a heterobifunctional cross-linking reagent that can connect the solid support through the amino or carboxyl functional group on the surface and the protein through cysteines or lysines.1,8 Perhaps the most commonly used functional group for the glass surface is the amino-reactive reagent N-hydroxysuccinimide (NHS) ester. However, the NHS group is notorious for unintended hydrolysis. Thus, extra care must be paid during its repeated usages and storage to keep the chemical effective, and an inconsistent result will be obtained if handled inappropriately. Also, maleimide (Mal) reacts with all available cysteines/lysines of the protein of interest (POI) in an uncontrolled way. To overcome these issues, we combine a strain-promoted azide–alkyne cycloaddition (SPAAC) click reaction and OaAEP1-based enzymatic ligation for robust and site-specific protein immobilization. Click chemistry and enzymatic ligation have attracted increasing attention in chemical biology, surface chemistry, and protein engineering. Click chemistry is robust in water with a high yield, which has been used for protein labeling and immobilization.9–11 However, its direct reaction with protein typically requires the chemical modification or incorporation of nonnatural amino acids in the target protein, which limits its yield and application.12,13 Protein ligase is also used for similar purposes, such as sortase A and butelase 1.14–16 The bottlenecks are the low efficiency or poor accessibility of the enzymes.17 Recently, we developed a transpeptidase OaAEP1(C247A)-mediated enzymatic ligation for site-specific polyprotein construction and immobilization.18 Here, we combined the N3-dibenzocyclooctyne (DBCO)-based SPAAC click reaction and the enzymatic ligation to immobilize enhanced green fluorescent protein (eGFP), which were verified by both fluorescent imaging and atomic force microscopy-based single-molecule force spectroscopy (AFM-SMFS) unfolding experiments.18–20 The chemicals used for click chemistry are stable, leading to easier handling and robust results. Moreover, the strength of the 1,2,3-triazole linkage from the click reaction was quantified, showing an ultrahigh strength above 1700 pN, which provides a general method for measuring the strength of stable proteins and bonds from click chemistry. Experimental Methods All reagents were purchased from commercial suppliers and used accordingly. Imidazole-1-sulfonyl azide hydrochloride (ImSO2N3·HCl; Abydos Scientific, Nanjing, Jiangsu, China), DBCO-(polyethylene glycol)4-Mal (DBCO-PEG4-Mal; Biocone, Chengdu, Sichuan, China), and 4-(N-maleimidomethyl)cyclohexane-1-carboxylic acid 3-sulfo-N-hydroxysuccinimide ester, sodium salt (sulfo-SMCC; Thermo Fisher, Rockford, IL) were stored and used in the dark. It is noted that ImSO2N3·HCl is hygroscopic and reacts slowly with water to produce hydrazoic acid, which is sensitive and explosive. Care must be taken to prevent this from occurring. Microporous membrane filters (0.22 and 0.45 μm) were used for further purification (Jet Biofil, Guangzhou, Guangdong, China). The genes of eGFP, Coh, XDoc, GB1, I27, SdrG, and Fgβ were purchased from Genscript (Nanjing, Jiangsu, China). OaAEP1(C247A) is a cysteine 247 to alanine mutant of asparaginyl endoproteases 1 from Oldenlandia affinis, abbreviated as OaAEP1. I27 is the 27th immunoglobulin domain of human cardiac titin. GB1 is the B1 domain of Streptococcal Protein G.21 [Coh-XDoc] is the type III cohesion dockerin-X module domain complex from Ruminococcus flavefaciens.22 SdrG is an serine-aspartic acid (SD)-repeat protein G, a cell surface adhesin from Staphylococcus epidermidis, and Fgβ is the N terminus of the β chain of human fibrinogen.23 All proteins were overexpressed in Escherichia coli BL21(DE3) cells. The protein purification details can be found in the Supplementary Information. For the AFM-SMFS experiment, Nanowizard4 AFM is used. The cantilevers (MLCT-Bio-DC; Bruker, Camarillo, CA) were calibrated in the AFM buffer by the thermal noise method to measure the spring constant. All experiments were performed under a constant pulling velocity of 1000 nm/s. The data were first filtered using data processing software (JPK, Berlin, Germany). More details regarding AFM data selection and analysis can be found in the Supporting Information. Results and Discussion The principle of our immobilization method is shown in Figure 1. First, the SPAAC click reaction replaces the NHS reaction (Figures 1a and 1b). Starting from an amino-silanized coverslip, N3 is functionalized on the surface from the reaction between ImSO2N3·HCl and –NH2 (step 1).24 Then, a heterobifunctional DBCO-PEGn-Mal can be reacted and adds the Mal group (step 2). Then, instead of randomly reacting with all cysteines of POI, the Mal reacts with a heterobifunctional peptide including elastin-like polypeptide (ELP)n as a spacer, glycine-leucine-ELP-cysteine (GL-ELP-C) (step 3).25 It adds the GL tag, which can be recognized by several protein transpeptidases, such as OaAEP1 and Sortase A. These endopeptidases can function as protein ligases to connect the target protein with the corresponding peptide tag [asparagine-glycine-leucine (NGL) for OaAEP1, or leucine-proline-glutamic acid-threonine-glycine (LPETG) for Sortase A] to the surface (step 4).18,26 Figure 1 | (a) The classic two-step method uses an sulfo-NHS-linker-Mal (SMCC) to immobilize protein on the surface. (b) Our method: the surface is functionalized with N3 by reacting the shelf-stable ImSO2N3 with the amino group. Then, DBCO-PEGn-Mal reacts with –N3, followed by the addition of GL-ELP-C. Finally, the POI with an NGL tag was immobilized on the surface by reacting with the functionalized-GL catalyzed by OaAEP1. Download figure Download PowerPoint To allow easy confirmation, eGFP was used as the model protein, which was verified by a clear fluorescent pattern on the glass slides (Figures 2a and 2b and Supporting Information Figures S1–S4). The surface preparation protocol is provided in the Supporting Information. First, eGFP and OaAEP1 solution were added to the GL-functionalized glass slide and incubated by hand pipetting. Then, the slides were washed with high-salt buffer and deionized water to remove the nonspecific bound proteins and dried. Finally, it was visualized by fluorescent imaging in air, which showed a clear edge formed by the covalently attached eGFP (Figure 2a). In addition, using an automatic microarrays machine, a 3 × 3 protein array on the functionalized surface can be obtained in air (Figure 2b). Moreover, the fluorescence image of eGFP in the buffer showed no difference from that on the dry sample ( Supporting Information Figure S4a). Under the same conditions, the droplet without OaAEP1 showed a much weaker signal ( Supporting Information Figure S4b). Moreover, the transformation rate from amines to azide group on the glass was analyzed by further reacting azide-functionalized surfaces with SMCC to react with the unreacted amine in the last step as a control experiment. And the subsequent steps were the same as the classic SMCC method, in which only these unreacted amines were used for protein immobilization. After quantification with ImageJ (National Institutes of Health, Bethesda, MD) and normalization with the new method, it showed 5% fluorescent intensity per area of the new method, indicating a ∼95% transformation rate (see Figure 2c, Supporting Information, and Supporting Information Table S1). Figure 2 | Verification of immobilized eGFP by fluorescent imaging and AFM-SMFS measurement. (a) Immobilized eGFP on the glass as a droplet by hand pipetting shows a clear fluorescent signal, whose boundary was indicated by a dashed line from bright-field imaging ( Supporting Information Figure S3). (b) eGFP was added in a 3 × 3 microarray, showing an expected fluorescence. (c) The fluorescent intensity per area of the three methods, this new method, classic SMCC, and a control for determining the transformation rate of amines to azide, were quantified by ImageJ and normalized. The histograms show the intensity of eGFP with (green) and without (black) OaAEP1. (d) The schematic shows how AFM-SMFS measures eGFP. (e) Force-extension curves show the eGFP unfolding signals (curves 1 and 2, in green), while no signal was observed outside the droplet (curves 3 and 4). (f) Diagram shows the relationship between the F and ΔLc of eGFP (one-step in green, n = 238; two-step in purple, n = 30) and GB1 (in black). Download figure Download PowerPoint To confirm the folding and stability of the immobilized eGFP, we performed the AFM-SMFS unfolding experiment. AFM manipulates a single molecule27–31 or complexes mechanically,32–34 which can (un)fold the protein,35–38 break molecular interactions,39–41 and chemically bond.42–48 In our previous design, cohesin (Coh) was fused to the eGFP, which binds to the Xmod-dockerin (XDoc) site specifically and reversibly with a rupture force of ∼600 pN.2,49 Thus, we functionalized the AFM tip with GB1-XDoc using the same method for the AFM experiment. Here, GB1 is a marker protein with a contour length increment (ΔLc) of 18 nm.21 By pressing the tip on the glass, POI was captured through the [Coh-XDoc] interaction. Then, moving the tip up vertically at a constant velocity (1000 nm/s), the fused protein was stretched and unfolded (Figure 2d). Finally, the [Coh-XDoc] ruptured, and the tip moved to another place to repeat this measurement. By probing the droplet-covered area using AFM-SMFS, force-extension curves with eGFP unfolding peaks were observed. Fitting the curve with the worm-like chain model, besides the unfolding peak from GB1 with ΔLc of 18 nm, an additional peak with ΔLc of ∼80 nm was observed (Figure 2e, curve 1, in green). It agrees with the theoretical results of the full extension of eGFP (227 amino acids). The unfolding force is 96 ± 13 pN (average ± standard deviation, n = 238) with an ΔLc of 84 ± 3 nm (Figure 2f).50 In addition, a two-step unfolding scenario of eGFP was observed, similar to previous findings for GFP (curve 2). We also probed the area outside the eGFP immobilization as a control. As expected, no curve containing the eGFP unfolding signal was detected (curves 3 and 4). To further prove the immobilized protein was well-folded with proper stability and demonstrate the ability of our method to immobilize large proteins, we immobilized another protein domain, I27. I27 is the first protein domain characterized by AFM-SMFS with an unfolding force of ∼220 pN and a ΔLc of ∼28 nm (pulling speed: 1000 nm/s).35,51 We built a Coh-(GB1)2-(I27)8-NGL polyprotein, a large fused protein with more than 10 protein domains. We then performed the AFM experiment using the same GB1-XDoc-functionalized AFM tip for measurement (Figure 3a). Its force-extension curves showed a classic sawtooth-like pattern with 11 unfolding force peaks, containing 8 peaks from I27 (in red) and 3 peaks from GB1 (in blue), respectively (Figure 3b). The force was 228 ± 32 pN for I27 (n = 524) and 212 ± 37 pN for GB1 (n = 217) (Figures 3c and 3d). These values are comparable with that of the same polyprotein immobilized by the NHS(SMCC)-based method ( Supporting Information Figure S5). Statistically, 48 out of 54 (89%) experiments succeeded with a high single-molecule pick-up ratio (<1%). Figure 3 | AFM-SMFS results for Coh-(GB1)2-(I27)8 using a GB1-XDoc functionalized tip both immobilized by the new method. (a) The schematic shows how the polyprotein is unfolded. (b) Force-extension curves showed 11 unfolding peaks of the polyprotein as designed. The unfolding peaks from I27 were colored in red and marked by a red star, and peaks from GB1 were colored in blue and marked by a blue star. Force histograms show the unfolding force is 228 ± 32 pN for I27 (n = 524) (c), and 212 ± 37 pN for GB1 (n = 217) (d), respectively. Download figure Download PowerPoint Finally, the mechanical strength of the immobilization system, including the triazole complex from the N3-DBCO reaction, was quantified. To test the high-force limit, another protein–protein interaction pair [SdrG-Fgβ] with an ultrahigh rupture force over 2000 pN was used ( Supporting Information Figure S6),23 instead of the previous weaker [Coh-XDoc] pair. We built an Fgβ-(GB1)2-NGL and an SdrG-NGL and immobilized them in the AFM-SMFS system for measurement (Figure 4a). After two GB1 unfolding peaks (213 ± 34 pN, n = 318, Figures 4b and 4c), it showed a high detachment peak with an average force of 1717 ± 185 pN (n = 313) (Figures 4b and 4d, red star), indicating the 1,2,3-triazole complex in the system could withstand forces above ∼1700 pN. It is noted that this statistical value is one of the strongest “click bond” measured so far.52–54 Figure 4 | (a) Schematic shows the mechanical strength measurement of the new immobilization system, including the 1,2,3 triazole complex. (b) Force-extension curves showed two unfolding peaks from GB1 (blue star) followed by a high-force detachment (rupture) peak (red star). (c and d) Force histograms show the average unfolding force is 213 ± 34 pN for GB1 (c) and 1717 ± 185 pN for the detachment (d). Download figure Download PowerPoint Conclusion In this work, we demonstrated the combination of the SPAAC click chemistry and enzymatic ligation for stable and efficient protein immobilization. Although it requires two more steps compared with the NHS-based method, our method shows good results. Most chemicals used for the click reaction, such as ImSO2N3, are shelf-stable for weeks,24 leading to a robust and convenient experimental procedure and a better and more consistent result, as verified by AFM-SMFS. Furthermore, the 1,2,3-triazole linkage in our system was quantified, which can withstand nN-level force suitable for characterizing most stable systems. Finally, it provides a robust platform to investigate the mechanical strength of the click bond. Supporting Information Supporting Information is available and includes experimental methods and materials, supplementary notes, Table S1, and Figures S1–S6. Conflict of Interest There is no conflict of interest to report. Funding Information This work was financially supported by National Natural Science Foundation of China (grant nos. 21771103 and 21977047), by the Fundamental Research Funds for the Central Universities (no. 14380205), and by Natural Science Foundation of Jiangsu Province (nos. BK20200058 and BK20202004). Acknowledgments The authors wish to acknowledge Prof. Zijian Guo, Prof. Thomas Perkins, Prof. Michael Nash, and Dr. Lukas Milles for their inspiring discussions, Prof. Jie Li for his help with click chemistry, Prof. Ying Liu, Prof. Deju Ye, and Miss Yue Zhang for fluorescent imaging, and Dr. Jie Pang and Miss Furui Jin for protein microarrays preparation. References 1. Chen Y. X.; Triola G.; Waldmann H.Bioorthogonal Chemistry for Site-Specific Labeling and Surface Immobilization of Proteins.Acc. Chem. Res.2011, 44, 762–773. Google Scholar 2. Liu H. P.; Ta D. T.; Nash M. A.Mechanical Polyprotein Assembly Using SFP and Sortase-Mediated Domain Oligomerization for Single-Molecule Studies.Small Methods2018, 2, 1800039. Google Scholar 3. Walder R.; LeBlanc M. A.; Van Patten W. J.; Edwards D. T.; Greenberg J. A.; Adhikari A.; Okoniewski S. R.; Sullan R. M. A.; Rabuka D.; Sousa M. C.; Perkins T. T.Rapid Characterization of a Mechanically Labile Alpha-Helical Protein Enabled by Efficient Site-Specific Bioconjugation.J. Am. Chem. Soc.2017, 139, 9867–9875. Google Scholar 4. Popa I.; Berkovich R.; Alegre-Cebollada J.; Badilla C. L.; Rivas-Pardo J. A.; Taniguchi Y.; Kawakami M.; Fernandez J. M.Nanomechanics of HaloTag Tethers.J. Am. Chem. Soc.2013, 135, 12762–12771. Google Scholar 5. Li M. Y.; Cheng F.; Li H. Q.; Jin W. W.; Chen C.; He W.; Cheng G.; Wang Q.Site-Specific and Covalent Immobilization of His-Tagged Proteins via Surface Vinyl Sulfone-Imidazole Coupling.Langmuir2019, 35, 16466–16475. Google Scholar 6. Zakeri B.; Fierer J. O.; Celik E.; Chittock E. C.; Schwarz-Linek U.; Moy V. T.; Howarth M.Peptide Tag Forming a Rapid Covalent Bond to a Protein, through Engineering a Bacterial Adhesin.Proc. Natl. Acad. Sci. U. S. A.2012, 109, E690–E697. Google Scholar 7. Fu H.; Jiang Y.; Yang D.; Scheiflinger F.; Wong W. P.; Springer T. A.Flow-Induced Elongation of von Willebrand Factor Precedes Tension-Dependent Activation.Nat. Commun.2017, 8, 324. Google Scholar 8. Zimmermann J. L.; Nicolaus T.; Neuert G.; Blank K.Thiol-Based, Site-Specific and Covalent Immobilization of Biomolecules for Single-Molecule Experiments.Nat. Protoc.2010, 5, 975–985. Google Scholar 9. Jewett J. C.; Bertozzi C. R.Cu-Free Click Cycloaddition Reactions in Chemical Google Scholar J. M.; J.; D. R.; Click Chemistry for of Biomolecules in 8, Google Scholar M.; D.; M.; Methods for Force Google Scholar Liu C. C.; to the Google Scholar C. A.; D. Protein Immobilization Using Google Scholar J. J.; J.; of Sortase Site-Specific of to in Chem. Google Scholar Using and Google Scholar Tian F.; Li G.; Zheng B.; Liu Y.; Shi Deng Y.; Zheng of Sortase for Protein by Single-Molecule Force Spectroscopy and Google Scholar Chen I.; M.; Liu D. for the of Using Natl. Acad. Sci. U. S. Google Scholar Deng Y.; Wu T.; Wang M.; Shi G.; Li X.; H.; Wu B.; Zheng and Immobilization of Polyprotein at the Single-Molecule Google Scholar Yang R.; Wong Y. H.; T.; J. P.; J.; Wu a Efficient Protein Am. Chem. Soc.2017, 139, Google Scholar H.; T. J.; T.; D. J.; H. Protein Labeling by a Am. Chem. Google Scholar Y.; C.; Wang M.; Li Protein Chem. Google Scholar R. C.; E.; C.; H.; G.; E. A.; H. E.; Nash M. of in a Protein by and Single-Molecule Force Am. Chem. Google Scholar Milles F.; H. E.; R. of in a Google Scholar E. D.; R. and Google Scholar W.; M. A.; M. E.; Milles F.; Nash M. A.; H. for Single-Molecule Force Google Scholar Y.; Chen L.; Liu J.; Zhang H.; Chen Labeling for Am. Chem. Google Scholar M.; G.; D.; M.; W. H.; J. L.; H. E.; C.; Y. F.; D. Force 1, Google Scholar Zhang W.; Zhang of Google Scholar H.; Li Wang W.; Liu P.; Liu J.; Y.; Wu L.; Zhang W.; of an to 2, Google Scholar Fu L.; Wang H.; Li from Protein Single-Molecule to 1, Google Scholar M.; L.; of of by Chem. Google Scholar P.; Y. and of Using Force Google Scholar E.; A.; S. J.; P.; E.; A. E. M.; A.; Wang Y. J.; F.; C.; P.; of Proteins the Google Scholar S. J.; R. E. M.; L.; E.; D. of by Commun.2017, 8, Google Scholar M.; M.; F.; Fernandez J. M.; H. of by Google Scholar Li Q.; the of a Google Scholar A.; M.; of Protein Google Scholar R.; P.; P.; D. W.; D. in of Google Scholar M.; C.; Sullan R. M. A.; A. E.; R.; A.; Blank of in the Google Scholar Zhang Liu H.; Zhang H.; L.; the of by Single-Molecule Chem. Google Scholar F.; A.; L.; H.; and by Force Spectroscopy and Natl. Acad. Sci. U. S. Google Scholar M.; M.; M.; H.; H. a Covalent Google Scholar Y.; Li X.; Li H.; Zhang the Efficient 5, Google Scholar G.; Liu H.; Q.; Li X.; J.; J.; Zheng Force Spectroscopy that Proteins in Chem. Google Scholar W.; Wu X.; X.; Y.; H.; Shi Y.; Chen Y.; M.; Wang W.; through Google Scholar D.; Badilla C. L.; Fernandez J. M.; Alegre-Cebollada Reactions in a of Protein Google Scholar M. F.; A. L.; D.; M. Bond Chem. Google Scholar G.; X.; Liu H.; G.; Tian F.; Shi Yang Y.; Li G.; Zheng Force Spectroscopy that the Bond of the in a Google Scholar S. W.; Nash M. A.; D. B.; M.; Y.; E. A.; H. of the Natl. Acad. Sci. U. S. A.2012, 109, Google Scholar H.; M.; M.; F.; T.; J. P.; Engineering of for Single-Molecule Force 1, Google Scholar M.; A. F.; S. B.; E.; S. E.; J.; Fernandez J. and Chemical of a A Natl. Acad. Sci. U. S. Google Scholar A.; F.; B.; R.; Fang of an and by Force in the of Covalent Google Scholar D.; J.; M. U.; Bond by the a Chem. Google Scholar M. J.; G.; Blank of Cycloaddition Chem. Google Scholar Information Chemical force authors wish to acknowledge Prof. Zijian Guo, Prof. Thomas Perkins, Prof. Michael Nash, and Dr. Lukas Milles for their inspiring discussions, Prof. Jie Li for his help with click chemistry, Prof. Ying Liu, Prof. Deju Ye, and Miss Yue Zhang for fluorescent imaging, and Dr. Jie Pang and Miss Furui Jin for protein microarrays preparation.

Topics & Concepts