Real-Time Imaging of Single-Molecule Enzyme Cascade Using a

Figure 1. Schematic illustration of molecule-level enzyme cascade on 2D surface. ... We first demonstrated the dynamic tethering (allowing free motion...
8 downloads 0 Views 7MB Size
Download PDF
Article pubs.acs.org/JACS

Real-Time Imaging of Single-Molecule Enzyme Cascade Using a DNA Origami Raft Lele Sun,†,‡,§ Yanjing Gao,†,‡,§ Yan Xu,†,‡ Jie Chao,∥ Huajie Liu,† Lianhui Wang,∥ Di Li,*,† and Chunhai Fan† †

Division of Physical Biology & Bioimaging Center, Shanghai Synchrotron Radiation Facility, Key Laboratory of Interfacial Physics and Technology, Shanghai Institute of Applied Physics, Chinese Academy of Sciences, Shanghai 201800, China ‡ University of Chinese Academy of Sciences, Beijing 100049, China ∥ Key Laboratory for Organic Electronics and Information Displays (KLOEID), Institute of Advanced Materials (IAM), School of Materials Science and Engineering, Nanjing University of Posts and Telecommunications, Nanjing 210046, China S Supporting Information *

ABSTRACT: The dynamics of enzymes are directly associated with their functions in various biological processes. Nevertheless, the ability to image motions of single enzymes in a highly parallel fashion remains a challenge. Here, we develop a DNA origami raft-based platform for in-situ real-time imaging of enzyme cascade at the single-molecule level. The motions of enzymes are rationally controlled via different tethering modes on a two-dimensional (2D) supported lipid bilayer (SLB). We construct an enzyme cascade by anchoring catalase on cholesterol-labeled double-stranded (ds) DNA and glucose oxidase on cholesterol-labeled origami rafts. DNA functionalized with cholesterol can be readily incorporated in SLB via the cholesterol−lipid interaction. By using a total internal reflection fluorescence microscope (TIRFM), we record the moving trajectory of fluorophore-labeled single enzymes on the 2D surface: the downstream catalase diffuses freely in SLB, whereas the upstream glucose oxidase is relatively immobile. By analyzing the trajectories of individual enzymes, we find that the lateral motion of enzymes increases in a substrate concentrationdependent manner and that the enhanced diffusion of enzymes can be transmitted via the cascade reaction. We expect that this platform sheds new light on studying dynamic interactions of proteins and even cellular interactions.

■

INTRODUCTION The dynamics of proteins provides valuable information on their functions. The locations of different functional proteins in signal transduction are closely related to their physiological roles; in particular, the colocation of different proteins in a cellular environment provide valuable information on signal pathways.1−4 However, owing to the resolution limit of optical microscopy, it is difficult to trace the motion of individual enzymes in actions, thereby hampering real-time imaging of the dynamic interactions of multiple enzymes in physiological environments.5−7 Recent developments in total internal reflection fluorescence microscopy (TIRFM) enable one to image the locations, movements, and complex formations of biomolecules in living cells.8−11 In addition, the conformation changes of a single protein can be also monitored by fluorescence resonance energy transfer (FRET) between fluorophore-modified interested amino acid residues.12−15 However, the fluorescence intensity decreases exponentially as the fluorophores move farther away from the surface and disappears when the particle is more than 300 nm from the surface, which makes it difficult for TIRFM to trace 3D free diffusion of biomolecules in solution.10 Moreover, it is still challenging for TIRFM to obtain © 2017 American Chemical Society

an in-situ dynamic interaction process between two or more biomolecules because it is difficult to simultaneously track the motion of different fluorophores-labeled biomolecules with different laser channels. One possible solution is to convert the three-dimensional free diffusion of biomolecules to a lateral motion. For this purpose, a two-dimensional fluidic surface is required to immobilize biomolecules.16 Supported lipid bilayer (SLB) has been demonstrated to be a powerful platform for anchoring biomolecules17−20 or NPs,21,22 where the behavior of lateral motion of biomolecules on the confined fluidic interface is similar to its free diffusion in solution. For example, Knight and Falke et al. studied the lateral motion of proteins and protein domains on SLB that interact reversibly with membranes in response to intracellular signals.23−25 However, even on the fluidic interface it is still hard to obtain information on the dynamic interaction process between multiple biomolecules or to simultaneously monitor the moving trajectories of two or more biomolecules with different channels. Received: September 1, 2017 Published: November 13, 2017 17525

DOI: 10.1021/jacs.7b09323 J. Am. Chem. Soc. 2017, 139, 17525−17532

Article

Journal of the American Chemical Society

Figure 1. Schematic illustration of molecule-level enzyme cascade on 2D surface. (a) Schematic of the enzyme cascade on supported lipid bilayer (SLB). Downstream catalase modified with alexa647 and single-stranded DNA (ss-DNA) is dynamically tethered on SLB via DNA hybridization, while upstream glucose oxidase (GOx) modified with atto488 and another ss-DNA is statically tethered on SLB via a triangle DNA origami raft. Triangle DNA origami raft with 27 anchor strands statically immobilized on SLB is labeled with a Cy3. (b) AFM image of GOx assembled on triangle DNA origami. Triangle DNA origami holds 12 binding sites for GOx. (c) Representative procedure of in-situ tracing the motion of single catalase on 2D SLB. First, we use a 561 nm laser to excite Cy3 to ensure one immobilized origami in the imaging field. Second, switching to a 488 nm channel to excite atto488 assures the colocation of GOx and DNA origami. Third, switching to a 647 nm channel to excite alexa647, and the moving trajectories of catalase are recorded with the 647 nm channel.

Figure 2. Dynamic and static tethering of down- and upstream enzymes on SLB. (a) (Left) Schematic illustration of the dynamic tethering of alexa647-modified catalase on SLB via ds-DNA. (Right) TIRFM image of the catalase-tethered SLB upon exciting with a 647 nm laser. (b) FRAP measurements indicate the high mobility of catalase on SLB. Recovery curves are fitted with the single-exponential function f(t) = a + b × (1 − e−λt) for calculating the diffusion coefficients from FRAP experiments via the parameter λ. Inset images show the process of fluorescence recovery in the ROI (region of interest, in dash line) after initial photobleaching. (c) Moving trajectories of GOx on origami raft with different numbers of anchor strands. (d) Average diffusion coefficient of GOx on origami raft, as determined by single-particle tracking and MSD analysis from 100 particles. Insets show the colocation of Cy3 and atto488 spots between two fluorescent images within 3 s when 27 anchoring strands were used to immobilize DNA raft. α: 0.996 ± 0.107, 0.969 ± 0.110, 0.942 ± 0.109, and 0.873 ± 0.095 for 3, 9, 17, and 27 anchor strands, respectively. 17526

DOI: 10.1021/jacs.7b09323 J. Am. Chem. Soc. 2017, 139, 17525−17532

Article

Journal of the American Chemical Society

Figure 3. Enhanced motion of catalase in enzymatic reaction. (a) Schematic illustration of the substrate-dependent diffusion of downstream enzyme in cascade reaction. (b) Time-dependent MSD plots of dynamic tethered catalase on SLB in the absence of H2O2 obtained from 50 trajectories. (c) Average diffusion coefficient (D) of dynamic tethered catalase increased with increasing concentrations of H2O2, obtained from 300 trajectories. D: 3.99 ± 2.11, 5.36 ± 2.52, 6.61 ± 3.07, and 7.52 ± 2.95 μm2/s for 0, 0.001, 0.01, and 0.1 M H2O2, respectively. (d) Mean α also increased with increasing concentrations of H2O2. α: 1.01 ± 0.15, 1.03 ± 0.16, 1.06 ± 0.138, and 1.07 ± 0.13 for 0, 0.001, 0.01, and 0.1 M H2O2, respectively. (e) Average diffusion coefficient of catalase within the 5 μm range of immobilized GOx when cascade reactions were activated (3.91 ± 2.19 μm2/s) or not (4.56 ± 2.40 μm2/s) and catalase beyond the 5 μm range of immobilized GOx while the cascade reaction was activated (3.98 ± 2.20 μm2/s), obtained from 200 trajectories.

In the present study, we establish a cascade enzymatic reaction on a 2D fluidic surface as a proof-of-concept study of chemotaxis of enzymes in a molecule level (Figure 1a). The ability of cholesterol to insert into phospholipids membrane provides a powerful means to immobilize protein−DNA conjugates on SLB via cholesterol-functionalized DNA (cholDNA). The mobility of proteins on SLB could be rationally controlled by different tethering modes, i.e., dynamic tethering and static tethering. We introduce DNA origami as raft26,27 to “fix” enzymes on the highly fluidic SLB. The upstream enzyme (glucose oxidase, GOx) is statically tethered with origami raft (Figure 1b) and “fixed” on the fluidic surface, while the downstream enzyme (catalase) is dynamically tethered through ds-DNA and remains freely diffuse. The colocation spot of origami and GOx represents a coordinate origin, while the

moving trajectory of catalase can be monitored by another channel of TIRFM (Figure 1c). Our results suggested that the diffusion of downstream enzyme in enzymatic reactions is increased in a substrate-dependent manner; however, the enhanced diffusion in a substrate gradient does no lead to productive motion toward upstream enzyme, which is attributed to the balanced competition of translational and rotational motion of enzyme. The molecule-level observation of the moving trajectory of cascaded enzymes provides powerful insights of chemotaxis behaviors. The ability of rationally controlling the mobility of proteins and real-time monitoring their motion trajectories also opens up new possibilities for the study of massively parallel dynamic interactions of protein. 17527

DOI: 10.1021/jacs.7b09323 J. Am. Chem. Soc. 2017, 139, 17525−17532

Article

Journal of the American Chemical Society

■

not only as a barge to carry proteins but also as an “adjustable lipid raft” to rationally control the moving behavior of proteins on fluidic interfaces. Herein, we used 27 anchor stands to immobilize GOx on 2D SLB in further study of chemotaxis of enzymes and collected the moving trajectory of catalase within 3 s. Increased Diffusion of Downstream Enzyme in Cascade Reaction. Recent studies indicated that the diffusive mobility of enzymes increases in the enzymatic reaction in a substrate-dependent manner.33−36 The behind mechanism was demonstrated to be an acoustic wave generated by the heat released during catalysis, or a chemoacoustic effect.37 The ability of ds-DNA and DNA origami to control the mobility of downstream and upstream enzymes provides a powerful platform to study the diffusivity of enzymes in cascade reactions (Figure 3a). Catalase, which mediates the conversion of H2O2 to H2O and O2, has been a model to demonstrate the substrate-dependent enhanced diffusion.37 We first confirmed that the surface-immobilized GOx and catalase was still enzymatically active (Figures S10 and S11). We next interrogated the free lateral diffusion of catalase on the fluidic surface in the absence of H2O2 by single-particle tracking. The resulting MSD plots in Figure 3b reveal an almost linear increase over time from 50 tracked catalases in the absence of H2O2. The average exponent factor (α) of single catalase was obtained from eq 1 and calculated to be 1.01, suggesting that dynamically tethered catalase undergoes Brownian motion-like lateral diffusion on SLB. We next interrogated the diffusion coefficient of catalase in the presence of its substrate, H2O2, and found the diffusivity of catalase indeed follows a substrate-dependent manner (Figure 3c, Figure S12). Moreover, with the increase of H2O2, α increased from 1.01 ± 0.15 to 1.07 ± 0.13, suggesting an enhanced diffusion (Figure 3d). We also performed a control experiment to interrogate the diffusivity of catalase in the presence of both H2O2 and its inhibitor, NH2OH, and did not observe a significant change in the diffusion coefficient, which further confirmed the enhanced diffusivity is a result of enzymatic reaction (Figure S13). Of note, catalase has a very high kcat and converts H2O2 into H2O and O2. Although previous studies have excluded the possible contribution of bubble generation-induced enhanced diffusion of catalase,34 we also performed a control experiment with GOx-horseradish peroxidase (HRP) cascade, which does not generate an O2 bubble, to rule out the possibility of bubble-induced enhanced motion (Figure S14). This control experiment still suggested a substrate-dependent enhanced diffusion of HRP, which completely excluded the possible contribution of a bubble. The substrate-dependent enhanced diffusion suggested that the diffusion of downstream enzymes could be regulated by the cascade reaction. Thereby we further challenged the diffusivity of catalase in the cascade reaction. Previous studies have indicated that the distribution of H2O2 around GOx follows

RESULTS AND DISCUSSION Dynamic Tethering of Catalase on 2D SLB. We first demonstrated the dynamic tethering (allowing free motion) of proteins on SLB. 2D SLB was built on a solid surface by fusion of small unilamellar vesicles.17 The fluidity of SLB was controlled by the phase-transition temperature of the composed phospholipids (Figure S1). Fluorescence recovery after photobleaching (FRAP) experiments28 suggested the asbuilt 2D SLB possesses good fluidity and integrality (Figure S2). The ability of cholesterol to insert into phospholipids bilayers enables one to anchor cholesterol-functionalized DNA nanostructures and NPs on 2D SLB, which diffuses freely.18,26 Inspired by this work, we attached enzymes with cholesterolfunctionalized ss-DNA and immobilized enzymes on the 2D fluidic surface via DNA hybridization. Briefly, a cholesterolfunctionalized ss-DNA was used as anchoring strand to immobilize catalase on SLB through the hybridization between the anchoring strand and its complementary strand that was covalently modified on catalase (Figures S3−6). In order to obtain the real-time moving trajectory of catalase on SLB, catalase was also modified with a fluorophore, alexa647 (Figure 2a, Figures S4a and S5a, and Supporting movie). We also performed FRAP experiments to interrogate the lateral mobility of catalase on 2D SLB (Figure 2b, inset). Data fitting yielded a mobile fraction of about 93%, suggesting a majority of catalase could diffuse freely on 2D SLB, a behavior similar to the lateral motion of lipid-anchored membrane proteins on a cell membrane.7,29 Of note, we found that the specific tethering of catalase on SLB is highly dependent on Mg2+ (Figure S7); thereby, a buffer solution without Mg2+ is prerequisite to avoid nonspecific adsorption of proteins on SLB. Static Tethering of GOx on 2D SLB. Having confirmed that cholesterol-functionalized DNA could tether catalase on 2D SLB we then were challenged to control the mobility of enzymes on 2D SLB by increasing the number of anchoring DNA strands. To achieve this we introduced DNA origami30,31 with a different number of anchoring strands as a barge to immobilize atto488-modifed GOx and investigated the mobility of an origami barge on 2D SLB. The moving trajectory of GOx on an origami barge was monitored after switching to 488 nm channel and treated by image analysis software (Figure 2c). With the increasing number of anchoring strands, the 2D diffusion of GOx became more stationary. The diffusion coefficient (D) of GOx was extracted from the mean square displacement (MSD) of lateral motion of an individual origami barge in a short time scale (τ) (eq 1) MSD = 4Dτ α

(1)

where τ is the observed lag time and the exponent α carries rich information about the motion type.32 From the MSD analysis, we found that the increasing anchor strands indeed slowed the diffusivity of the origami barge (Figure 2d, Figure S8). More importantly, α drops from 0.996 ± 0.107 (3 anchoring strands) to 0.873 ± 0.095 (27 anchor stands), indicating the diffusion behavior changed from Brownian motion to subdiffusion in the trapped lipid bilayer. The stationary GOx on the DNA barge with 27 anchor strands was further confirmed by the colocation of Cy3 and atto488 spots (Figure 2d, inset). Of note, we designed 8 sites and 12 sites for hybridizing with Cy3-DNA and GOx-atto488-DNA, respectively. The distance between the farthest Cy3 and atto488 sites was theoretically calculated to be 96.5 nm, which is within the 200 nm resolution of TIRFM (Figure S9). These results suggested that DNA origami behaves

i=t /τ−1

∑

n(r , t ) =

i=0

−

⎡ 1 ⎢ exp ⎢⎣ [4πDH2O2(t − iτ )]3/2

⎤ r2 ⎥ 4DH2iO2(t − iτ ) ⎥⎦

(2)

where n(r,t) is the concentration of H2O2 at a distance (r, μm) from the initial position at the diffusion time (t, s) and DH2O2 is the diffusion coefficient of H2O2.38 The simulated H2O2 17528

DOI: 10.1021/jacs.7b09323 J. Am. Chem. Soc. 2017, 139, 17525−17532

Article

Journal of the American Chemical Society

Figure 4. Distribution analysis of catalase around immobilized GOx. (a) In situ imaging the distribution of alexa647-labeled catalase around immobilized GOx and subsequently localizing the coordinate of catalase. Scale bar is 200 nm in the zoomed figure. (b) TIRFM images showing the distribution of catalase around GOx upon different treatment. White dashed line represents a circular area with the spot of GOx as the center; white spots represent the location of catalase obtained from point spread function (PSF) mapping. These images are all for the same GOx location. (c) Mean standard deviation (σ) of 9 times parallel measurements in the X-axis and Y-axis direction changed along with time under different treatment. (d) Scatter distribution of the standard deviation (σ) in the X-axis direction and Y-axis direction corresponding to multiparallel measurements (n = 32) under different treatment. Coordinate origin is at the direction as indicated by the gray arrow.

Random Distribution of Catalase in Cascade Reaction. The enhanced diffusivity thus intrigues an interesting question, i.e., does the increased diffusion of enzymes lead to a directional migration toward higher substrates gradients or a chemotaxis behavior? Chemotaxis, the directed movement of organisms in response to a chemical stimulus, is often found in bacteria, cells, or multicellular organisms;39,40 however, it is obscure to think that enzymes might have evolved the capacity to seek substrates.41 Although very recent studies have suggested that an ensemble of enzyme molecules exhibits chemotaxis,34,35,42 this work did not provide insightful information by which this may occur. The ability of tracing the moving trajectories of enzymes in a fluidic surface provides a powerful tool to real-time locate the position of catalase and GOx. The position of GOx tightly immobilized on 2D SLB with an origami raft was thus defined as the original point in coordinate (x0, y0) (Figure 4a). The relative position of catalase (xi, yi) toward GOx was analyzed by

concentration gradient using eq 2 is shown in Figure S15 with the following parameters: diffusion coefficient (DH2O2) = 1000 μm2/s; kcat (GOx) = 300 s−1; and an integration time 3 s. The result indicated that the concentration gradient of H2O2 dominantly exists within 5 μm around GOx. Therefore, we analyzed the diffusion coefficient of catalase in the activated cascade reaction within and beyond the 5 μm diameter range with GOx as the center, within this range but the cascade reaction was not activated (Figure 3e). We, indeed, found an increase of the diffusion coefficient of catalase in cascade reaction within this range in the presence of glucose with p < 0.05, while beyond this range no obvious change in the diffusion coefficient was observed. These results further confirmed that the lateral motion of enzymes increases in a substrate concentration-dependent manner and indicated that the enhanced diffusion of enzymes could be transmitted by cascade reactions. 17529

DOI: 10.1021/jacs.7b09323 J. Am. Chem. Soc. 2017, 139, 17525−17532

Article

Journal of the American Chemical Society

Figure 5. Chemotactic behavior of catalase under confined rotary motion. (a) Model of the movement behavior of catalase around immobilized GOx. Gradual varied green circle around GOx represents H2O2 concentration gradient. (b) Schematic illustration of the rotation behavior of catalase tethered onto SLB via different methods. (c) Scatter distribution of the standard deviation (σ) in the X direction and Y direction corresponds to multiparallel measurements (n = 16) under different treatment. (d) Model of the movement behavior of catalase under confined rotary motion in a H2O2 concentration gradient field. Gradual varied green circle around GOx represents H2O2 concentration gradient. Red and green arrows represent the displacement toward and opposite to GOx, respectively.

standard deviation,43 σx in the x direction and σy in the y direction, with home-written software, where σx and σy are defined as σx =

σy =

n

κ=

∑ (Xi − X 0)2

1 N

∑ (Yi − Y0)2

i=1

(3)

N i=1

n

(5)

The calculated κ was 585 ± 8 in the absence of both substrate and inhibitor (glucose (−), NH2OH (−)), 586 ± 7 in the presence of substrate and absence of inhibitor (glucose (+), NH2OH (−)), and 585 ± 7 in the presence of both substrate and inhibitor (glucose (+), NH2OH (+)), respectively. The even value of κ suggests an unchanged relative position between up- and downstream enzymes in the cascade reaction, which does not support the biological chemotaxis explanation. This suggested the behind mechanism for the enhanced diffusion of enzymes in the catalytic reaction is more complicated.37 In classical chemotaxis of cells or bacteria, the motion is influenced by external signals through chemical interactions.39,44,45 The external signal could be either a positive stimuli (e.g., a gradient of nutrients) or a negative stimuli (e.g., poison). In the well-established E. coli chemotaxis model, E. coli sense, and process the chemotactic stimuli is performed by complexes that consist of several types of attractant-specific chemoreceptors in the cytoplasmic membrane and flagellar motors that preform swimming.46 The sensitive chemotaxis response is ensured by the allosteric interaction between chemosensory receptors and switch subunits of the flagellar

N

1 N

∑i = 1 Xσi 2 + Yσi 2

(4)

Obviously, the standard deviation (σ) represents the discrete degree of the distribution of catalase around immobilized GOx. However, upon interrogating with the substrate of GOx (glucose) and inhibitor of catalase (NH2OH) (Figure 4b), we found the distribution of catalase around immobilized GOx was stochastic (Figure 4c), and there were no apparent changes in σ along with time. Also, the moving trajectories of catalase around GOx were random after the addition of glucose (Figure S16). We further plotted the standard deviation in the X and Y directions (Figure 4d) and further introduced a chemotaxis factor “κ” (eq 5, Xσi is the standard deviation in the X direction, Yσi is the standard deviation in the Y direction, and n is the number of measurement times). A smaller κ would indicate a relatively concentrated distribution of catalase around GOx, or chemotaxis. 17530

DOI: 10.1021/jacs.7b09323 J. Am. Chem. Soc. 2017, 139, 17525−17532

Article

Journal of the American Chemical Society motors.47−49 To make temporal comparisons in chemotaxis, a prerequisite is the ability for cells or bacteria to sense their current environment and a temporal memory of the past. However, macroscopic chemotaxis of the enzyme ensembles under given conditions could not represent the underlying molecular mechanism.34,41 In the present work, we found that the lateral motion of downstream enzymes indeed increased in cascade reactions; however, the distribution of downstream enzyme is randomly around its upstream enzyme and no chemotaxis is observed. We argued that rotational diffusion of enzymes could not be ignored. Enzymes also undergo rotational diffusion in 3D solution environment (Figure S17); the directed chemotaxis, however, also requires the active center of enzymes always faces the substrate gradient or upstream enzymes. Therefore, we speculated that the chemotaxis behavior is a balanced competition of lateral and rotational motion (Figure 5a). Slowing down or eliminating the rotational motion might help to improve enzymatic reaction-induced directional motion. We thus designed a control experiment by rationally decreasing the rotational diffusion of catalase to explore the distribution of catalase around immobilized GOx. Catalase was tethered on SLB via biotin−streptavidin interaction instead of DNA hybridization (Figure 5b). Because of the labeling method (SI Experimental section), there were multiple biotin sites on the surface of catalase, which could recruit several streptavidin. The friction between streptavidin and SLB, together with the larger size of catalase−streptavidin complex, would result in a slower rotational diffusion of catalase compared with one chol-DNA. Of note, although this tether mode also decreased the translational diffusion, the diffusion coefficient of catalase still increased in the presence of H2O2 (Figure S18). We then examined the distribution of catalase around GOx and obtained a relatively smaller standard deviation (σ) in the X and Y directions after adding glucose (Figure 5c), as indicated by a smaller κ (from 537 ± 7 to 525 ± 6). This control experiment indicated that owing to the confined rotary motion, the orientation of the active center of catalase could be kept for a longer time, resulting in a rise to a fixed direction of acting force from catalytic reactions. To support this we further compared the translational and rotary motion rate of catalase on SLB. For catalase with a radius of 5 nm,50 we theoretically calculated the time scale of rotational diffusion to be 7.64 × 10−5 s and the translational rate to be 9.8 μm/s from the Einstein relation,37,38,51,52 which means that during spinning one circle catalase will translationally move 0.8 nm, which is much smaller than its own radius (see Supporting Information). These calculations suggested that the rotary of activated catalase was much faster than lateral motion, which further supported our conclusion that the chemotaxis behavior of enzymes is a balanced competition of lateral and rotational motion (Figure 5d).

easily expanded to study other dynamic protein interactions. For instance, the platform could be used to study posttranslational modifications (PTM) by colocation of different proteins in PTM pathways. We thus envisage that the method developed here opens brand new possibilities in studying protein interactions and inhibitor screening

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/jacs.7b09323. Experimental section, viability, quantitative analyses, all nucleic acid sequences, control experiments, fluorescence anisotropy measurement, and theoretical derivation (PDF) Real-time moving trajectory of catalase on SLB (AVI)

■

AUTHOR INFORMATION

Corresponding Author

*[email protected] ORCID

Lianhui Wang: 0000-0001-9030-9172 Di Li: 0000-0003-1674-0110 Chunhai Fan: 0000-0002-7171-7338 Author Contributions §

L.S. and Y.G. contributed equally to this work.

Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This work was supported by the National Basic Research Program of China (2013CB932803, 2013CB933800), National Key R&D Program of China (2016YFA0201200, 2016YFA0400900), NSFC (21675166, 21227804, 21390414, 21473236, 31371015, 21329501), and Key Research Program of Frontier Sciences, CAS (QYZDJ-SSW-SLH031). We thank Prof. Lehui Xiao and Prof. Wenhao Li for helpful discussions.

■

REFERENCES

(1) Goulian, M.; Simon, S. M. Biophys. J. 2000, 79, 2188−2198. (2) Golan, D. E.; Veatch, W. Proc. Natl. Acad. Sci. U. S. A. 1980, 77, 2537−2541. (3) Sbalzarini, I. F.; Koumoutsakos, P. J. Struct. Biol. 2005, 151, 182− 195. (4) Ke, G. L.; Liu, M. H.; Jiang, S. X.; Qi, X. D.; Yang, Y. R.; Wootten, S.; Zhang, F.; Zhu, Z.; Liu, Y.; Yang, C. J.; Yan, H. Angew. Chem., Int. Ed. 2016, 55, 7483−7486. (5) Serge, A.; Bertaux, N.; Rigneault, H.; Marguet, D. Nat. Methods 2008, 5, 687−694. (6) Thompson, M. A.; Casolari, J. M.; Badieirostami, M.; Brown, P. O.; Moerner, W. E. Proc. Natl. Acad. Sci. U. S. A. 2010, 107, 17864− 17871. (7) Weigel, A. V.; Simon, B.; Tamkun, M. M.; Krapf, D. Proc. Natl. Acad. Sci. U. S. A. 2011, 108, 6438−6443. (8) Orrit, M.; Ha, T.; Sandoghdar, V. Chem. Soc. Rev. 2014, 43, 973− 976. (9) Schneckenburger, H. Curr. Opin. Biotechnol. 2005, 16, 13−18. (10) Joo, C.; Balci, H.; Ishitsuka, Y.; Buranachai, C.; Ha, T. Annu. Rev. Biochem. 2008, 77, 51−76. (11) Liang, L.; Li, J.; Li, Q.; Huang, Q.; Shi, J. Y.; Yan, H.; Fan, C. H. Angew. Chem., Int. Ed. 2014, 53, 7745−7750. (12) Rasnik, I.; McKinney, S. A.; Ha, T. Acc. Chem. Res. 2005, 38, 542−548.

■

CONCLUSIONS In summary, we developed a method to control the moving behaviors of proteins on 2D fluidic surfaces by different tethering mode. The moving trajectory of fluorophore-labeled proteins on the 2D fluidic surfaces could be real-time monitored by TIRFM. The ability of recording the moving trajectory of multiple proteins provides new tools to study the dynamic interactions of proteins, which was exemplified in this work by studying substrate-enhanced diffusion of enzymes and chemotaxis in cascade reactions. However, this ability could be 17531

DOI: 10.1021/jacs.7b09323 J. Am. Chem. Soc. 2017, 139, 17525−17532

Article

Journal of the American Chemical Society (13) Kim, H.; Ha, T. Rep. Prog. Phys. 2013, 76, 016601. (14) Roy, R.; Hohng, S.; Ha, T. Nat. Methods 2008, 5, 507−516. (15) Gilburt, J. A. H.; Sarkar, H.; Sheldrake, P.; Blagg, J.; Ying, L.; Dodson, C. A. Angew. Chem., Int. Ed. 2017, 56, 11409−11414. (16) Lingwood, D.; Simons, K. Science 2010, 327, 46−50. (17) Block, S.; Zhdanov, V. P.; Hook, F. Nano Lett. 2016, 16, 4382− 4390. (18) Johnson-Buck, A.; Jiang, S. X.; Yan, H.; Walter, N. G. ACS Nano 2014, 8, 5641−5649. (19) Palla, K. S.; Hurlburt, T. J.; Buyanin, A. M.; Somorjai, G. A.; Francis, M. B. J. Am. Chem. Soc. 2017, 139, 1967−1974. (20) Avakyan, N.; Conway, J. W.; Sleiman, H. F. J. Am. Chem. Soc. 2017, 139, 12027−12034. (21) Hartman, K. L.; Kim, S.; Kim, K.; Nam, J. M. Nanoscale 2015, 7, 66−76. (22) Lee, Y. K.; Kim, S.; Oh, J. W.; Nam, J. M. J. Am. Chem. Soc. 2014, 136, 4081−4088. (23) Ziemba, B. P.; Li, J.; Landgraf, K. E.; Knight, J. D.; Voth, G. A.; Falke, J. J. Biochemistry 2014, 53, 1697−1713. (24) Ziemba, B. P.; Knight, J. D.; Falke, J. J. Biochemistry 2012, 51, 1638−1647. (25) Brandt, D. S.; Coffman, M. D.; Falke, J. J.; Knight, J. D. Biochemistry 2012, 51, 7654−7664. (26) Kocabey, S.; Kempter, S.; List, J.; Xing, Y. Z.; Bae, W.; Schiffels, D.; Shih, W. M.; Simmel, F. C.; Liedl, T. ACS Nano 2015, 9, 3530− 3539. (27) Czogalla, A.; Kauert, D. J.; Franquelim, H. G.; Uzunova, V.; Zhang, Y. X.; Seidel, R.; Schwille, P. Angew. Chem., Int. Ed. 2015, 54, 6501−6505. (28) Axelrod, D.; Koppel, D. E.; Schlessinger, J.; Elson, E.; Webb, W. W. Biophys. J. 1976, 16, 1055−1069. (29) You, M.; Lyu, Y.; Han, D.; Qiu, L.; Liu, Q.; Chen, T.; Sam Wu, C.; Peng, L.; Zhang, L.; Bao, G.; Tan, W. Nat. Nanotechnol. 2017, 12, 453−459. (30) Yao, G. B.; Li, J.; Chao, J.; Pei, H.; Liu, H. J.; Zhao, Y.; Shi, J. Y.; Huang, Q.; Wang, L. H.; Huang, W.; Fan, C. H. Angew. Chem., Int. Ed. 2015, 54, 2966−2969. (31) Chao, J.; Zhang, Y. N.; Zhu, D.; Liu, B.; Cui, C. J.; Su, S.; Fan, C. H.; Wang, L. H. Sci. China: Chem. 2016, 59, 730−734. (32) Qian, H.; Sheetz, M. P.; Elson, E. L. Biophys. J. 1991, 60, 910− 921. (33) Muddana, H. S.; Sengupta, S.; Mallouk, T. E.; Sen, A.; Butler, P. J. J. Am. Chem. Soc. 2010, 132, 2110−2111. (34) Sengupta, S.; Dey, K. K.; Muddana, H. S.; Tabouillot, T.; Ibele, M. E.; Butler, P. J.; Sen, A. J. Am. Chem. Soc. 2013, 135, 1406−1414. (35) Sengupta, S.; Patra, D.; Ortiz-Rivera, I.; Agrawal, A.; Shklyaev, S.; Dey, K. K.; Cordova-Figueroa, U.; Mallouk, T. E.; Sen, A. Nat. Chem. 2014, 6, 415−422. (36) Ma, X.; Hortelao, A. C.; Patino, T.; Sanchez, S. ACS Nano 2016, 10, 9111−9122. (37) Riedel, C.; Gabizon, R.; Wilson, C. A. M.; Hamadani, K.; Tsekouras, K.; Marqusee, S.; Presse, S.; Bustamante, C. Nature 2015, 517, 227−230. (38) Fu, J. L.; Liu, M. H.; Liu, Y.; Woodbury, N. W.; Yan, H. J. Am. Chem. Soc. 2012, 134, 5516−5519. (39) Ridley, A. J.; Schwartz, M. A.; Burridge, K.; Firtel, R. A.; Ginsberg, M. H.; Borisy, G.; Parsons, J. T.; Horwitz, A. R. Science 2003, 302, 1704−1709. (40) Csete, M. E.; Doyle, J. C. Science 2002, 295, 1664−1669. (41) Wand, A. J. Nature 2015, 517, 149−150. (42) Dey, K. K.; Sen, A. J. J. Am. Chem. Soc. 2017, 139, 7666−7676. (43) Ernst, D.; Kohler, J. Phys. Chem. Chem. Phys. 2013, 15, 845− 849. (44) Wadhams, G. H.; Armitage, J. P. Nat. Rev. Mol. Cell Biol. 2004, 5, 1024−1037. (45) Baggiolini, M. Nature 1998, 392, 565−568. (46) Meacci, G.; Tu, Y. Proc. Natl. Acad. Sci. U. S. A. 2009, 106, 3746−3751.

(47) Hazelbauer, G. L.; Falke, J. J.; Parkinson, J. S. Trends Biochem. Sci. 2008, 33, 9−19. (48) Kollmann, M.; Lovdok, L.; Bartholome, K.; Timmer, J.; Sourjik, V. Nature 2005, 438, 504−507. (49) Sourjik, V.; Berg, H. C. Nature 2004, 428, 437−441. (50) Mozaffar, S.; Ueda, M.; Kitatsuji, K.; Shimizu, S.; Osumi, M.; Tanaka, A. Eur. J. Biochem. 1986, 155, 527−531. (51) Golestanian, R.; Liverpool, T. B.; Ajdari, A. Phys. Rev. Lett. 2005, 94, 220801. (52) Saffman, P. G.; Delbruck, M. Proc. Natl. Acad. Sci. U. S. A. 1975, 72, 3111−3113.

17532

DOI: 10.1021/jacs.7b09323 J. Am. Chem. Soc. 2017, 139, 17525−17532