Molecular Calipers for Highly Precise and Accurate Measurements of

Feb 20, 2014 - *E-mail: [email protected]., *E-mail: [email protected]. ... and accurate length change measurements in protein mechanics ... We furthe...
1 downloads 0 Views 2MB Size
Download PDF
Article pubs.acs.org/Langmuir

Molecular Calipers for Highly Precise and Accurate Measurements of Single-Protein Mechanics Yanyan Wang,†,‡ Xiaodong Hu,*,† Tianjia Bu,‡ Chunguang Hu,† Xiaotang Hu,† and Hongbin Li*,†,‡ †

State Key Laboratory of Precision Measurements Technology and Instruments, School of Precision Instrument and Optoelectronics Engineering, Tianjin University, Tianjin, 300072 PR China ‡ Department of Chemistry, University of British Columbia, Vancouver, BC V6T 1Z1 Canada ABSTRACT: Single-molecule atomic force spectroscopy (AFM) has evolved into a powerful technique toward elucidating conformational changes in proteins when exposed to applied force. AFM technologies that are currently available allow for precise measurements of proteins length changes during conformational transitions. However, because of systematic errors in piezo calibration as well as errors originating from fitting experimental data using a worm-like chain model of polymer elasticity, high-precision measurements of length changes do not necessarily translate into highly accurate measurements of length changes, resulting in uncertainty in obtaining structural information about protein conformational changes. Actually achieving highly precise and accurate force spectroscopy measurements remains a challenge. Here, we report a protein caliper method that eliminates systematic errors that occur during single-protein force spectroscopy measurements, and thus achieves highly precise and accurate length change measurements in protein mechanics studies. To do this, a series of loop elongation variants of the small protein GB1, which differ by 2, 5, 10, 15, and 24 amino acid residues, were engineered. Differential measurements of amino acid residue length obtained from different AFM setups result in a precise measure of the length of a single amino acid residue, which varies within different AFM setups because of systematic error between individual AFM piezoelectric calibrations. The measured length of a single amino acid residue from a given AFM setup is then used as a caliper for the given setup to eliminate systematic error, leading to highly accurate and precise measurements of the number of amino acid residues that are involved in a conformation change of a polypeptide chain. We further developed a more precise, robust, and model-free method to determine the apparent size of single amino acid residues and conformational changes of proteins. This method improves the accuracy of single protein force spectroscopy measurements, providing an accurate means of measuring force-induced protein conformational changes.

■

INTRODUCTION Over the last two decades, the development and application of atomic force microscopy (AFM)-based single-molecule force spectroscopy has made it possible to examine the conformational dynamics of macromolecules with unprecedented detail at the single-molecule level.1−8 In particular, AFM-based singlemolecule force spectroscopy has evolved into a powerful tool to investigate protein folding−unfolding dynamics. By stretching a single protein molecule between a soft microfabricated cantilever and a solid substrate, the force−extension relationship of that particular protein can be directly measured. This applied force induces conformational changes in the protein, which are often accompanied by a length change between the two tethered amino acid residues. Thus, accurately measuring such length changes can provide useful structural information about proteins during their conformation transitions.9−11 Because of the compliance of a polypeptide chain, the apparent change in protein length depends on the force applied when the length change is observed;12,13 this is different from the change in contour length, which is constant and reflects the intrinsic length change in the protein. Developing a robust molecular caliper is thus critical to gaining precise and © 2014 American Chemical Society

accurate structural information about proteins under investigation using single-molecule force spectroscopy techniques. Current technology has made it possible to achieve highprecision measurements of length changes, which correspond to information about protein structural changes, in singlemolecule force spectroscopy studies.4−6,10 However, high precision measurements of length changes do not necessarily translate into highly accurate measurements of length changes or allow one to obtain highly accurate structural information about protein conformational changes. How to achieve highly precise and highly accurate force spectroscopy measurements remains a challenge. The length of a single amino acid residue in a fully extended polypeptide chain is a constant and can be accurately measured from 3D structures determined using X-ray diffraction or nuclear magnetic resonance spectroscopy.14 Thus, the length of a single amino acid residue has naturally been proposed as a molecular caliper for length measurements in protein Received: December 31, 2013 Revised: February 19, 2014 Published: February 20, 2014 2761

dx.doi.org/10.1021/la404978f | Langmuir 2014, 30, 2761−2767

Langmuir

Article

Figure 1. (A) Schematic showing the 3D structure of GB1. The second loop of GB1 is highlighted in green, and residues 39 and 40 are indicated in yellow. Loop elongation variants are constructed by inserting a specific number of residues between residues 39 and 40 in the second loop of GB1. (B) Representative force−extension curves of caliper proteins. Stretching polyproteins of GB1 loop insertion variants results in characteristic sawtoothlike force−extension curves in which the individual force peaks correspond to the mechanical unfolding of the individual mutant domains. Blue lines correspond to the WLC fits to experimental data. It is evident that the contour length increment increases as the number of inserted residues increases. (C) Superposition of force−extension curves of caliper proteins. It is clear that small differences in ΔLc between GB1 loop variants can be readily amplified by increasing the number of repeat domains in the polyprotein.

■

mechanics studies and has been used extensively in singlemolecule force spectroscopy studies to infer structural information about proteins.9,10,15,16 However, the contour length change upon conformational changes in proteins are usually measured by fitting experimental data to the worm-like chain (WLC) model of polymer elasticity.17 It remains untested whether the length of an individual amino acid measured by using the WLC fits is the same as that measured using X-ray diffraction and nuclear magnetic resonance spectroscopy. This effect coupled with the potential errors in piezo calibration (which is often done through optical interferometry18 or the use of calibration standards19−21) could account for the diverse values of the length of a single amino acid residue in the literature (ranging from 3.4 to 4.0 Å10,16,22−25). This discrepancy has led to uncertainty in the inferred structural information on proteins and made it difficult to compare data from different AFMs and different research groups. To circumvent this challenge, we report the development of caliper proteins that allows for a model-free way of defining the length of a single amino acid. Our method provides a timely solution for high-precision and high-accuracy measurements in single-protein mechanics studies.

MATERIALS AND METHODS

Protein Engineering. The model proteins chosen as potential calipers (GB1-L2, GB1-L5, GB1-L10, GB1-L15, and GB1-L24; for simplicity, we refer to them as GL2, GL5, GL10, GL15, and GL24) are loop-insertion variants based on the wild-type (wt) GB1 domain. To construct these caliper proteins, we inserted a nonpalindromic AvaI restriction site (CTC GGG) between codons for the 39th and 40th amino acid residues in the second loop of wt GB1. The insertion of the AvaI site was done via standard site-directed mutagenesis protocols. The resultant sequence encodes GL2, in which a two-residue linker LG was inserted into loop 2 of wt GB1 via the AvaI site. Genes encoding GL5 and GL24 were constructed as previously described.26 Genes encoding GL10 and GL15 were constructed in a similar way as for GL24. All of the caliper proteins have the following general sequence: MDTYKLILNG KTLKGETTTE AVDAATAEKV FKQYANDNGV InsertedSequence DGEWTYDDATKTFTVTE. Portions of the sequence noted in italics are from wt GB1; inserted sequences for each of the individual loop insertion variants were inserted between residue 39−40 of the wt GB1. The inserted sequences for different loop insertion variants are GL2: LG; GL5: GGGLG; GL10: GGGSGASGLG; GL15: GCGS(GAS)3CG; and GL24: LG(GSA)6GSLG Caliper polyproteins containing eight identical tandem repeats of the specific caliper protein were constructed using procedures similar to that used to engineer (GB1)8, which utilize the “sticky ends” of BamHI and BglII sites to insert protein domains. Polyprotein genes 2762

dx.doi.org/10.1021/la404978f | Langmuir 2014, 30, 2761−2767

Langmuir

Article

were subcloned into the expression vector pQE80L, which contains an N-terminal (His)6 tag to facilitate the purification of expressed proteins. The polyprotein was expressed in DH5α strain E. coli and purified using Ni-NTA affinity chromatography. Single-Molecule AFM. Single-molecule AFM measurements were performed on a custom-built atomic force microscope (AFM)27 (setup 1) and a commercial MFP-3D AFM (AsylumResearch) (setup 2) following published procedures.27 Setup 1 is equipped with a PicoCube actuator with capacitive sensors that have a resolution of 50 pm (Physik Instrumente). Setup 2 is equipped with a LVPD sensor that has a resolution of 260 pm (AsylumResearch). All cantilevers were calibrated in a phosphate-buffered saline (PBS) solution using the equipartition theorem. Briefly, 1 μL of the polyprotein solution (with a concentration of ∼1 mg/mL) was deposited onto a freshly cleaned glass coverslip with ∼50 μL of PBS; both solutions were thoroughly mixed. After ∼10 min of equilibration, single-molecule AFM experiments were performed at a pulling speed of 400 nm/s. All data analysis was accomplished using custom scripts written in Igor Pro 6.0.

unfolding force peaks to the WLC model of polymer elasticity allows for the measurement of the contour length increment upon domain unfolding. It is evident that ΔLc differences between any pair of GB1 variants (such as GB1 vs GL2) are small. When the first unfolding peaks within force−extension curves for two proteins are placed in register (Figure 1C), differences in ΔLc are not obvious within the first unfolding peaks. However, small differences in ΔLc between two variants are amplified by multiple domains, where unfolding force peaks become increasingly out of register, resulting in a sizable difference in ΔLc that can be easily measured.9 Figure 2A shows ΔLc histograms for the six GB1 variants measured from WLC fits to the experimental data. It is evident

■

RESULTS AND DISCUSSION Differential Measurements of the Molecular Caliper Size: A Protein Caliper Approach. Numerous attempts have been made to measure the length of a single amino acid residue.9,10,15 However, these attempts likely suffer from the negative impacts arising from protein structure compliance as well as potential systematic errors inherent in each of the methods used. To remove such uncertainty, we describe a protein caliper approach that utilizes a differential method to measure the length of a single amino acid residue precisely. The principle of this protein caliper approach is to use a series of loop insertion proteins that differ both by a known number of residues in length as well as contour length increment but demonstrate similar mechanical properties (such as protein structure compliance). Differences in the measured caliper protein contour length increment are linearly dependent upon the number of inserted residues, providing a precise measurement of the single amino acid residue length. To achieve this goal, the well-characterized GB1 domain was used as a model system. GB1 is a small α/β protein of 56 amino acid residues28 whose mechanical properties have been well characterized by single-molecule AFM.27 Our previous studies have shown that the second loop, which consists of five residues (residue 37−41) and connects the α helix to β strands 3, can accommodate the insertion of additional flexible residues without adversely impacting the overall structure of GB1.26,29,30 By inserting various numbers of flexible amino acid residues between residues 39 and 40, we engineered a series of loopinsertion variants (GL2, GL5, GL10, GL15, and GL24) in which 2, 5, 10, 15, and 24 residues, respectively, were inserted into the second loop of GB1 while the rest of GB1 sequence remains unchanged (Figure 1A). The structural integrity of these GB1 variants was confirmed by circular dichroism spectroscopy.26 The resolution of contour length increment measurement is ∼2 nm using the current AFM techniques.10,31 To improve the precision of the differential measurements, we constructed polyproteins consisting of eight identical tandem repeats of the given GB1 variant. The tandem modular construction amplifies the small length differences between different GB1 variants to improve the signal-to-noise ratio.9 Figure 1B shows the representative sawtooth-like appearance of the six polyproteins as they are unfolded under force. Each sawtooth peak corresponds to a mechanical unfolding event of a given domain within the polyprotein.13 Fitting consecutive

Figure 2. Contour length increment ΔLc histograms for caliper proteins as measured on a home-built AFM (A, setup 1) and an MFP3D AFM (B, setup 2).

that the average ΔLc for these caliper proteins is linearly proportional to the number of inserted residues in the loop insertion variants (Figure 3). A linear fit to the change in ΔLc with the number of inserted residues shows a ΔLc of 17.9 nm for wt GB1. The slope of this relationship, 0.37 nm/aa, provides a precise measurement of the length of a single amino acid within the unfolded polypeptide chain. This value is slightly larger than, but very similar to, the length of a single amino acid residue (0.36 nm/aa) expected for a fully extended protein chain.14,32 Use of Multiple Protein Calipers Provides an Accurate and Precise Structural Measure of Conformational 2763

dx.doi.org/10.1021/la404978f | Langmuir 2014, 30, 2761−2767

Langmuir

Article

experimental data can be effectively eliminated, allowing for precise and accurate structural information from different AFMs to be obtained and compared. To illustrate this possibility, we used GL15 as a case study. GL15 carries two cystein residues in the second loop that are capable of forming a disulfide bond under oxidizing conditions.29 Disulfide bond formation will result in the sequestration of approximately 11 residues and a decrease in the contour length increment corresponding to the loss of these 11 residues. Figure 4 shows representative force−extension

Figure 3. Use of multiple caliper proteins allows for the measurement of a single amino acid residue length. The ΔLc measured for each individual loop insertion variant is linearly proportional to the number of inserted residues. A linear regression results in single amino acid residue lengths of 0.37 ± 0.01 nm (for setup 1, R2 = 0.998) and 0.35 ± 0.01 nm (for setup 2, R2 = 0.999). The error bar indicates the standard error of the mean.

Changes. Using an engineered set of protein calipers combined with force spectroscopy experiments, we have determined the length of a single amino acid residue. The accuracy of this method ultimately depends on the accurate calibration of the piezo actuator. The measurement reported here is derived from our custom-made AFM equipped with a PicoCube actuator that comprises capacitive sensors with a resolution of 50 pm. Because of the accuracy of the piezo calibration and possible variations over time, different AFMs may offer different accuracy in their length measurements, even if these measurements are of high precision. To illustrate this point, we carried out the method described above utilizing the caliper protein set with a different AFM, the commercially available MFP3D equipped with an LVPD sensor, and a resolution of 260 pm. Data from this instrument is shown in Figures 2B and 3. A linear fit to the change in ΔLc with the increase in the number of inserted of residues results in a ΔLc of 17.90 nm for wt GB1 and a slope of 0.35 nm/aa (Figure 3). The length of a single amino acid residue measured from these different instruments differs by ∼5%. In some extreme cases, the measurement of a single amino acid residue length using AFMs that are equipped with an excellent position sensor can differ from the theoretical value by more than 10% (data not shown). These results clearly demonstrate the potential error in calibrating each individual AFM and the associated low accuracy of length measurements despite the high precision of individual experiments. It is also clear from Figure 3 that the difference in measured ΔLc between the two setups is proportional to the contour length increment. The larger the ΔLc, the larger these differences will be. This effect can be particularly problematic for comparing data obtained within different laboratories for proteins that show large contour length increments during unfolding and conformational changes. To improve the accuracy in obtaining structural information about protein conformational transitions, we propose that the single amino acid length measured by an individual AFM be used as a reference and that the molecular caliper set be used to “calibrate” the specific AFM. In doing so, systematic errors from the calibration in individual AFM and WLC fitting to the

Figure 4. Representative force−extension curves of GL15DB and GL15. The inset shows the sequence of the GL5 domain with the disulfide bond highlighted (blue line). Upon formation of the disulfide bond between two cysteine residues, 11 residues are sequestered (colored in green) and shielded from the stretching force, leading to a contour length increment reduction of GL15DB. Thin lines correspond to WLC fits to the experimental data.

curves of GL15 under oxidizing (GL15DB) and reducing conditions. As shown in Table 1, the ΔΔLc between the two redox states as measured by different AFM instruments differs by ∼10%. The student’s t test showed that the ΔΔLc values measured from the two AFM instruments are statistically different from each other at a significance level of 0.01. If we use the theoretical length of a single amino acid residue (0.36 nm) to calculate the number of sequestered residues that give rise to the measured ΔΔLc, results from the two AFMs differ by ∼1.2 aa. However, if we use the length measured from the specific AFM, then the numbers of residues measured are very similar (approximately ∼11 residues for each), which is in excellent agreement with the number of amino acid residues sequestered by the disulfide bond. This result clearly demonstrates that the use of caliper proteins can produce highly precise and accurate structural information about protein conformational changes, despite the potential low accuracy of the given AFM and piezo calibration. More Precise and Robust Method for Force Spectroscopy Experiments. Our results demonstrate that the described caliper protein approach reliably provides precise and accurate single protein force spectroscopy. However, determining the length of a single amino acid residue is based on WLC fits to the experimental data. To precisely fit each force spectra to the WLC, the low force regime of each sawtooth peak is of critical importance, as that is where the 2764

dx.doi.org/10.1021/la404978f | Langmuir 2014, 30, 2761−2767

Langmuir

Article

Table 1. Measured Length of a Single Amino Acid Residue Can Be Used to Obtain Structural Information about Protein Conformational Changesa setup 1 setup 2 a

ΔLc(GL15DB) (nm)

ΔLc(GL15) (nm)

ΔΔLc (nm)

Laa (nm)

number of sequestered aa

19.12 ± 0.08 (n = 148) 19.01 ± 0.06 (n = 103)

23.19 ± 0.09 (n = 172) 22.76 ± 0.11 (n = 213)

4.07 ± 0.13 3.65 ± 0.17

0.37 0.35

11.0 10.7

Data is presented as average ± standard error of the mean (n = number of events).

Figure 5. The interpeak spacing method provides precise measurements of length changes occurring during protein conformational changes. (A) Schematic illustration of the polyprotein interpeak spacing ΔLsi. (B) ΔLsi histogram measured from force−extension curves shown in part A. Seven unfolding events result in seven narrow peaks, where the intensity decreases from the first to the last. ΔLsi measures the interpeak spacing between unfolding events n and n + I and should be i*ΔLs1. (C) ΔLsi is linearly proportional to i. A linear fit provides a highly precise means of measuring the average interpeak spacing for the unfolding of a polyprotein. ΔLs1 histograms from caliper proteins measured on (D) setup 1 and (E) setup 2. It is evident that the distribution of ΔLs1 is very narrow.

WLC model is best fit.17 However, data within the low force range is usually not present or very limited due to the sawtooth appearance. In addition, standard WLC fits do not take into account the extensibility of each amino acid residue at higher forces.17,33,34 These factors limit the precision and accuracy of WLC fits to the experimental data. To circumvent these practical challenges, we propose a robust alternative method that is capable of achieving higher measurement precision. Instead of measuring the length of a single amino acid residue in the fully extended state (WLC fits), we can use the average length of a single amino acid at a given force range as a length caliper for force spectroscopy measurements. This method is model-free and offers a higher precision for length measurements. For a given sawtooth-like force extension curve, the spacing ΔLs between two unfolding force peaks (adjacent and nonadjacent) at the same force can be measured across the defined force range (Figure 5A). ΔLs1 corresponds to the length increment between any two adjacent unfolding force peaks, and thus measures the average length increment of the protein domain at the given force range. Similarly, ΔLsi

measures the length increment between the two nonadjacent unfolding force peaks n and n + i. Because all domains are identical within the given polyprotein, the interpeak spacing between consecutive force peaks should be identical.35 This property also leads to the interesting and useful observation that the interpeak spacing between nonadjacent peaks ΔLsi would be equal to i· ΔLs1. Figure 5B shows a representative interpeak spacing ΔLs histogram for the polyprotein (GB1)8. For seven unfolding events of GB1, the ΔLs histogram shows seven narrow peaks, with the intensity decreasing from the first peak to the last. The first peak represents the spacing ΔLs1 between two consecutive unfolding force peaks (peaks n and n + 1) and has the highest intensity. The second peak corresponds to ΔLs2 and measures the spacing between peaks n and n + 2. Similarly, the interpeak spacing ΔLsi is displayed as the ith peak in the histogram. Fitting each individual peak to a Gaussian distribution measures the average specific interpeak spacing ΔLsi. As expected, ΔLsi is linearly proportional to i. A linear regression of this relationship leads to a precise measurement of ΔLs1 (Figure 5C). 2765

dx.doi.org/10.1021/la404978f | Langmuir 2014, 30, 2761−2767

Langmuir

Article

Figure 5D shows the ΔLs1 histogram measured from different molecules. The interpeak spacing ΔLs1 shows a much narrower distribution than the ΔLc histograms (Figure 2), indicating that interpeak spacing ΔLs1 offers a higher precision measurement than those based on contour length increment. Comparing the measurements from the same set of data, we found that the standard deviation of interpeak spacing measurements is ∼3 times lower than that from contour length increment measurements. On the basis of this interpeak spacing measurement, we have used the caliper proteins to determine the average length Laa′ of a single amino acid precisely within an applied force range of 30−80 pN (Figure 5A, red lines). Figure 6 shows the interpeak

which is in excellent agreement with the number of amino acid residues sequestered by the disulfide bond. This result demonstrates that this method can be used as a robust means of obtaining accurate structural information about protein conformational changes. This method is model-free, can be automated, and has greatly improved the data analysis efficiency in single-protein mechanics studies.

■

CONCLUSIONS By combining single-molecule force spectroscopy with an engineered set of protein calipers, we have developed a differential measurement-based, highly precise, and highly accurate approach to extracting information about the number of amino acid residues that are involved in force-induced protein conformational changes. The engineered set of protein calipers comprises of a series of loop elongation variants that differ in length by a known number of amino acid residues. The use of such variants provides a robust method to measure the length of a single amino acid residue that serves as an inherent molecular length caliper for “calibrating” force spectrometers. This method allows for the elimination of the systematic errors associated with AFM calibrations and thus improves the accuracy of single-protein force spectroscopy measurements. We anticipate that the caliper proteins reported here can be used as a general standard to calibrate the length of a single amino acid residue in single-molecule AFM-based protein mechanics studies and thus help improve the comparability of force spectroscopy results among different laboratories.

■

Figure 6. Caliper proteins allow for the precise measurement of the average amino acid residue length in the range of applied force of 30− 80 pN. ΔLs1 measured for individual loop insertion variants is linearly proportional to the number of inserted residues. ΔLs1 measured from the two setups is linearly proportional to the number of inserted amino acid residues in loop 2. A linear regression has a slope of 0.280 ± 0.009 nm per amino acid (R2 = 0.998) for setup 1, with a slope of 0.270 ± 0.008 nm per amino acid (R2 = 0.997) for setup 2.

AUTHOR INFORMATION

Corresponding Authors

*E-mail: [email protected]. *E-mail: [email protected]. Notes

The authors declare no competing financial interest.

■

spacing ΔLs1 as a function of the inserted number of amino acid residues in the loop insertion variants. A linear regression of this relationship results in an average single amino acid length measurement (Laa′) of 0.28 nm/aa in the force range of 30−80 pN. As in contour length increment measurements, the average measured length Laa′ varies among different AFMs, reflecting systematic errors in AFM calibration. However, by using the measured Laa′ as a reference for a given AFM, an accurate measurement of the number of residues involved in conformational changes of proteins is possible. To demonstrate this possibility, unfolding data from GL15 and GL15DB were closely investigated. Figure 5 shows representative force−extension curves of GL15 under oxidizing and reducing conditions. Interpeak spacing analysis revealed that the change in interpeak spacing ΔΔLs upon reduction corresponds to the spacing change caused by 11.5 residues,

ACKNOWLEDGMENTS We thank Ms. Eileen Wang for her technical contributions at the beginning of this project. This work was supported by the Natural Sciences and Engineering Research Council of Canada, the Canada Foundation of Innovation, the Canada Research Chairs Program, and the National Natural Science Foundation of China. Y.W. and T.B. were supported by fellowships from the China Scholarship Council.

■

REFERENCES

(1) Borgia, A.; Williams, P. M.; Clarke, J. Single-molecule studies of protein folding. Annu. Rev. Biochem. 2008, 77, 101−125. (2) Brockwell, D. J. Force denaturation of proteins - an unfolding story. Curr. Nanosci. 2007, 3, 3−15. (3) Li, H. B. ‘Mechanical Engineering’ of Elastomeric Proteins: Toward Designing New Protein Building Blocks for Biomaterials. Adv. Funct Mater. 2008, 18, 2643−2657.

Table 2. Measured Average Length of a Single Amino Acid Residue within a Given Force Range Can Be Used to Obtain Structural Information about Proteins’ Conformational Changesa setup 1 setup 2 a

ΔLs1(GL15DB) (nm)

ΔLs1(GL15) (nm)

ΔΔLs1 (nm)

Laa′ (nm)

number of sequestered aa

16.08 ± 0.04 (n = 148) 15.53 ± 0.04 (n = 103)

19.26 ± 0.04 (n = 172) 18.63 ± 0.05 (n = 213)

3.18 ± 0.06 2.78 ± 0.08

0.28 0.27

11.4 11.5

Data is presented as average ± standard error of the mean (n = number of events). 2766

dx.doi.org/10.1021/la404978f | Langmuir 2014, 30, 2761−2767

Langmuir

Article

(26) Li, H. B.; Wang, H. C.; Cao, Y.; Sharma, D.; Wang, M. Configurational entropy modulates the mechanical stability of protein GB1. J. Mol. Biol. 2008, 379, 871−880. (27) Cao, Y.; Li, H. B. Polyprotein of GB1 is an ideal artificial elastomeric protein. Nat. Mater. 2007, 6, 109−114. (28) Gronenborn, A. M.; Filpula, D. R.; Essig, N. Z.; Achari, A.; Whitlow, M.; Wingfield, P. T.; Clore, G. M. A novel, highly stable fold of the immunoglobulin binding domain of streptococcal protein G. Science 1991, 253, 657−661. (29) Peng, Q.; Kong, N.; Wang, H. C.; Li, H. Designing redox potential-controlled protein switches based on mutually exclusive proteins. Protein Sci. 2012, 21, 1222−1230. (30) Peng, Q.; Li, H. Domain insertion effectively regulates the mechanical unfolding hierarchy of elastomeric proteins: toward engineering multifunctional elastomeric proteins. J. Am. Chem. Soc. 2009, 131, 14050−14056. (31) Zheng, P.; Chou, C. C.; Guo, Y.; Wang, Y.; Li, H. Single molecule force spectroscopy reveals the molecular mechanical anisotropy of the FeS4Metal center in rubredoxin. J. Am. Chem. Soc. 2013, 135, 17783−17792. (32) Eisenberg, D. The discovery of the α-helix and β-sheet, the principal structural features of proteins. Proc. Natl. Acad. Sci. U.S.A. 2003, 100, 11207−11210. (33) Bustamante, C.; Marko, J. F.; Siggia, E. D.; Smith, S. Entropic elasticity of lambda-phage DNA. Science 1994, 265, 1599−1600. (34) Rief, M.; Oesterhelt, F.; Heymann, B.; Gaub, H. E. Single molecule force spectroscopy on polysaccharides by atomic force microscopy. Science 1997, 275, 1295−1297. (35) This property has been used as an on-the-fly method to identify single-molecule stretching events of polyproteins in Professor Julio Fernandez’s group (http://fernandezlab.biology.columbia.edu/ downloads).

(4) Zoldak, G.; Rief, M. Force as a single molecule probe of multidimensional protein energy landscapes. Curr. Opin. Struct. Biol. 2013, 23, 48−57. (5) Marszalek, P. E.; Dufrene, Y. F. Stretching single polysaccharides and proteins using atomic force microscopy. Chem. Soc. Rev. 2012, 41, 3523−3534. (6) Puchner, E. M.; Gaub, H. E. Force and function: probing proteins with AFM-based force spectroscopy. Curr. Opin. Struct. Biol. 2009, 19, 605−614. (7) Best, R. B.; Brockwell, D. J.; Toca-Herrera, J. L.; Blake, A. W.; Smith, D. A.; Radford, S. E.; Clarke, J. Force mode atomic force microscopy as a tool for protein folding studies. Anal. Chim. Acta 2003, 479, 87−105. (8) Hoffmann, T.; Dougan, L. Single molecule force spectroscopy using polyproteins. Chem. Soc. Rev. 2012, 41, 4781−4796. (9) Carrion-Vazquez, M.; Marszalek, P. E.; Oberhauser, A. F.; Fernandez, J. M. Atomic force microscopy captures length phenotypes in single proteins. Proc. Natl. Acad. Sci. U.S.A. 1999, 96, 11288−11292. (10) Dietz, H.; Rief, M. Protein structure by mechanical triangulation. Proc. Natl. Acad. Sci. U.S.A. 2006, 103, 1244−1247. (11) Bertz, M.; Rief, M. Mechanical unfoldons as building blocks of maltose-binding protein. J. Mol. Biol. 2008, 378, 447−458. (12) Oberhauser, A. F.; Hansma, P. K.; Carrion-Vazquez, M.; Fernandez, J. M. Stepwise unfolding of titin under force-clamp atomic force microscopy. Proc. Natl. Acad. Sci. U.S.A. 2001, 98, 468−472. (13) Carrion-Vazquez, M.; Oberhauser, A. F.; Fowler, S. B.; Marszalek, P. E.; Broedel, S. E.; Clarke, J.; Fernandez, J. M. Mechanical and chemical unfolding of a single protein: A comparison. Proc. Natl. Acad. Sci. U.S.A. 1999, 96, 3694−3699. (14) Pauling, L.; Corey, R. B. The pleated sheet, a new layer configuration of polypeptide chains. Proc. Natl. Acad. Sci. U.S.A. 1951, 37, 251−256. (15) Takeda, S.; Ptak, A.; Nakamura, C.; Miyake, J.; Kageshima, M.; Jarvis, S. P.; Tokumoto, H. Measurement of the length of the alpha helical section of a peptide directly using atomic force microscopy. Chem. Pharm. Bull. 2001, 49, 1512−1516. (16) Yang, G.; Cecconi, C.; Baase, W. A.; Vetter, I. R.; Breyer, W. A.; Haack, J. A.; Matthews, B. W.; Dahlquist, F. W.; Bustamante, C. Solidstate synthesis and mechanical unfolding of polymers of T4 lysozyme. Proc. Natl. Acad. Sci. U.S.A. 2000, 97, 139−144. (17) Marko, J. F.; Siggia, E. D. Stretching DNA. Macromolecules 1995, 28, 8759−8770. (18) Jaschke, M.; Butt, H. J. Height calibration of optical-lever atomic-force microscopes by simple laser interferometry. Rev. Sci. Instrum. 1995, 66, 1258−1259. (19) Alliata, D.; Cecconi, C.; Nicolini, C. A simple method for preparing calibration standards for the three working axes of scanning probe microscope piezo scanners. Rev. Sci. Instrum. 1996, 67, 748− 751. (20) Brodowsky, H. M.; Boehnke, U. C.; Kremer, F. Wide range standard for scanning probe microscopy height calibration. Rev. Sci. Instrum. 1996, 67, 4198−4200. (21) Li, Y.; Lindsay, S. M. Polystyrene latex-particles as a size calibration for the atomic force microscope. Rev. Sci. Instrum. 1991, 62, 2630−2633. (22) Oesterhelt, F.; Oesterhelt, D.; Pfeiffer, M.; Engel, A.; Gaub, H. E.; Muller, D. J. Unfolding pathways of individual bacteriorhodopsins. Science 2000, 288, 143−146. (23) Erickson, H. P. Reversible unfolding of fibronectin type III and immunoglobulin domains provides the structural basis for stretch and elasticity of titin and fibronectin. Proc. Natl. Acad. Sci. U.S.A. 1994, 91, 10114−10118. (24) Trombitas, K.; Greaser, M.; Labeit, S.; Jin, J. P.; Kellermayer, M.; Helmes, M.; Granzier, H. Titin extensibility in situ: entropic elasticity of permanently folded and permanently unfolded molecular segments. J. Cell Biol. 1998, 140, 853−859. (25) Sarkar, A.; Caamano, S.; Fernandez, J. M. The elasticity of individual titin PEVK exons measured by single molecule atomic force microscopy. J. Biol. Chem. 2005, 280, 6261−6264. 2767

dx.doi.org/10.1021/la404978f | Langmuir 2014, 30, 2761−2767