HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Afrin, R. Articles by Ikai, A. Search for Related Content PubMed PubMed Citation Articles by Afrin, R. Articles by Ikai, A. Social Bookmarking                   What's this?

Pretransition and progressive softening of bovine carbonic anhydrase II as probed by single molecule atomic force microscopy

Laboratory of Biodynamics, Graduate School of Bioscience and Biotechnology, Tokyo Institute of Technology, Midori-ku, Yokohama, 226-8501, Japan

Reprint requests to: Atsushi Ikai, Laboratory of Biodynamics, Graduate School of Bioscience and Biotechnology, Tokyo Institute of Technology, 4259 Nagatsuka, Midori-ku, Yokohama, 226-8501, Japan; e-mail: aikai{at}bio.titech.ac.jp; fax: 81-45-924-5806.

(RECEIVED December 8, 2004; FINAL REVISION March 6, 2005; ACCEPTED March 11, 2005)

	Abstract

TOP Abstract Introduction Results Discussion Materials and methods References

 
To develop a simple method for probing the physical state of surface adsorbed proteins, we adopted the force curve mode of an atomic force microscope (AFM) to extract information on the mechanical properties of surface immobilized bovine carbonic anhydrase II under native conditions and in the course of guanidinium chloride–induced denaturation. A progressive increase in the population of individually softened molecules was probed under mildly to fully denaturing conditions. The use of the approach regime of force curves gave information regarding the height and rigidity of the molecule under compressive stress, whereas use of the retracting regime of the curves gave information about the tensile characteristics of the protein. The results showed that protein molecules at the beginning of the transition region possessed slightly more flattened and significantly more softened conformations compared with that of native molecules, but were still not fully denatured, in agreement with results based on solution studies. Thus the force curve mode of an AFM was shown to be sensitive enough to provide information concerning the different physical states of single molecules of globular proteins.

Keywords: bovine carbonic anhydrase II; atomic force microscopy (AFM); protein stretching and compression; Young’s modulus; progressive softening; pretransition softening; Tatara model

Article and publication are at http://www.proteinscience.org/cgi/doi/10.1110/ps.041282305.

	Introduction

TOP Abstract Introduction Results Discussion Materials and methods References

 
The atomic force microscope (AFM) is widely used to characterize biomacromolecules such as proteins, polysaccharides, and DNA at the single molecular level; providing a wealth of knowledge on their mechanical properties by the direct application of compressive or tensile force to individual molecules (Radmacher et al. 1994; Mitsui et al. 1996; Rief et al. 1997; Carrion-Vazquez et al. 1999; Clausen-Schaumann et al. 2000; Li et al. 2000; Wang et al. 2001; Alam et al. 2002; Hertadi et al. 2003; Möller et al. 2003). In most of the studies cited above, macromolecules were stretched by the application of tensile forces in opposing directions from two arbitrarily chosen or specifically defined points along their primary structures, providing information on the internal heterogeneity in their mechanical design. In just a limited number of cases has an AFM been used to compress protein molecules to extract information on the apparent Young’s modulus of individual molecules; for example, in the case of lysozyme, it was reported to be in the range of 0.5 ± 0.2 GPa (Radmacher et al. 1994). Although the actual measurement of Young’s modulus using either an AFM or a surface force apparatus has so far given values in a 0.2GPa to 2 GPa range (Morozov and Morozova 1981; Radmacher et al. 1994; Suda et. al. 1995 ), a theoretical assessment of protein rigidity has recently been put forward proposing 0.05–0.5GPa as a plausible range of the mechanical modulus of protein molecules (Vanselow 2002).

Development of a method to evaluate protein rigidity and, if possible, local variations of it is important because protein flexibility has been recognized as an essential requirement in the expression of enzyme activity, muscle contraction, ligand–receptor interactions, and many other protein functions (Cantor and Schimmel 1980). It has also been studied using different experimental techniques such as pressure-dependent NMR spectroscopy or acoustic measurement of the compressibility of proteins in solution (Gekko 2002; Akasaka 2003). Further, the physical state of surface adsorbed protein has been an important issue in the field of surface science (Karlsson et al. 2000; Ramsden 2002; Vermeer 2002) and biotechnology (Wahlgren and Arnebrant 1991; Elwing 1998; Tengvall et al. 1998; Drobny et al. 2003). Atomic force microscopy has the potential to make significant contributions to the study of protein flexibility by allowing researchers to press and/or pull single protein molecules in specified directions.

Bovine carbonic anhydrase II (BCA II) is a zinc metalloprotein with 259 amino acid residues and is known to undergo a transition from the native to a fully denatured state via a stable intermediate state(s) when the concentration of guanidinium chloride (GdmCl) is increased from 0 M to 6 M (Wong and Tanford 1973). Characterization of the pretransition intermediate states of the protein has been actively pursued for bovine as well as human CA by using a variety of optical, hydrodynamical, spin-labeling, and chromatography methods. It is now well accepted that the pretransition state of this protein under mildly denaturing conditions has properties of which many can be ascribed to the "molten globule" state of the protein (Uversky 1993; Borén et al. 1999; Carlsson and Johnsson 2000; Andersson et al. 2001). In the molten globule state, the hydrodynamics radius of the protein is similar to that of the native protein, but most of the experimental parameters indicate that the protein is devoid of the segmental interactions required for keeping its well-defined three-dimensional conformation, although the optical parameters strongly support that the secondary structure of the protein is formed almost to the same level as the native protein. The proposed nature of the molten globule-like intermediate state of BCA II presents itself as an ideal test case for application of AFM-based mechanical measurement at the single molecular level. We previously reported the mechanical response of the recombinantly modified protein with two cysteine residues outside of N and C termini under native and fully denatured conditions using the force curve mode of an AFM (Wang et al. 2001). In this report we investigate whether an AFMbased mechanical method can distinguish the pretransition intermediate state from either the native or fully denatured states.

	Results

TOP Abstract Introduction Results Discussion Materials and methods References

 
Results under native conditions
Figure 1 gives a schematic view of the force mode operation of the AFM with a BCA II molecule sandwiched between the tip and the substrate. The radius of the tip is shown as several times larger than that of the sample protein, which is close to the actual ratio of the two radii. The three circles in black dotted line indicate the positions of alternate crossings of the C- and N-terminal segments together forming a "knot structure."

View larger version (40K): [in this window] [in a new window]   Figure 1. A schematic view of force curve operation of the AFM with a bovine carbonic anhydrase molecule sandwiched between the AFM tip and the substrate through covalent cross-linkers. Three alternative crossings of the chain in the C-terminal region leading to the knot formation are indicated with circles in black dotted lines.

View larger version (13K): [in this window] [in a new window]   Figure 2. (A) An example of the raw output of the force curve mode of AFM operation. The tip starts from position 1 and touches the sample at position 2, from where the cantilever is pushed up to position 3. Thence, the tip reverses its movement up to position 4, and the cantilever is pulled downward to position 5 due to the presence of tensile sample. At position 5, a covalent bond yields to the tensile force, and the cantilever jumps back to its free position at 6. D, d, E, and I are, respectively, the distance between the tip and the substrate, cantilever deflection, sample extension, and the depth of compression. The total distance covered by the piezomotor under the sample stage is D = E+d. (B) The raw data above are converted to the F–E curve by plotting the tensile force, F = –k x d versus E, where k is the spring constant of the cantilever. The horizontal positions marked as H and E0, respectively, indicate the contact positions of the tip to the protein upon theirmutual approach and the point of their final separation upon retraction. They, therefore, respectively correspond to the approximate height of the sample and the contour length of the protein, provided that the chain is stretched normal to the substrate surface.

We first introduce in Figure 3A, the result of pulling experiments on BCA II in a 50 mM Tris-sulfate buffer (pH 7.5) at ~25°C. The inset figures give representative individual curves with initial extension of ~15 nm (1) and ~30 nm (2), respectively. A majority of the pulling curves showed an initial lowforce response (0.2–0.25 nN)for ~10– 30 nm, and then the force increased rapidly without any significant further extension of the sample (without protein samples, force curves were almost flat, showing little or no interactions). When the force reached the 1–2 nN range, the covalent cross-linking system was ruptured, and no tensile interactions were obtained beyond the rupture point. This behavior of BCAII in the early phase of forced stretching is similar to the previously reported observation of Wang et al. (2001), who attributed this effect to the presence of a "knot" structure in the C-terminal region of the protein (for BCA II, see Saito et al. 2004; for HCAII, see Eriksson et al. 1988; Håkansson et al. 1992). There appear to be two groups of force curves in Figure 3A with respect to the length of the chain extension achieved before the covalent bond was ruptured: one in ~15-nm range and the other in ~30-nm range. It is thus possible that the protein can be unfolded in two different ways with respect to the initial knot tightening processes. Our preliminary result of molecular dynamics simulations of chain extension of knotted BCA II confirmed the appearance of large force peaks at ~15 nm and ~30 nm in extension, which explains the presence of ~15 nm peak described above, but the disappearance of the same peak in ~30 nm extension remains to be explained.

View larger version (20K): [in this window] [in a new window]   Figure 3. (A) The F–E curves of BCA II under native conditions with 50 mM Tris-sulfate buffer (pH 7.5) and at ~25°C. The initial low level forces in the range of 0.1–0.2 nN are considered to represent frictional resistance in the C-terminal knotted region. Insets 1 and 2 contain representative individual curves with ~15 nm and ~30 nm initial extensions, respectively. (B) F–E curve obtained in the presence of the inhibitor. Inset is a collection of representative individual curves.

When the immobilized protein was incubated with the inhibitor, p-aminomethylbenzenesulfonamide, in the AFM liquid cell and used for force curve measurements, the F–E curve showed a slight change from those without the inhibitor, as shown in Figure 3B with an inset figure showing representative individual curves here and hereafter. The initial extension before the rupture of the covalent bond became longer by ~5–10 nm compared with those in Figure 3A and the depth of compression is shown to be only 2 nm compared with 3.5 ± 0.1 nm for the native protein, indicating an increased rigidity of the core structure accompanied by a slight loosening of surface residues of the protein as discussed later.

We next examine the compression part of the force curve. The part of the F–E curve in the contact region in Figure 4A is given as an overlap of 15 experimental curves obtained under native conditions. In this figure, the abscissa, I_H, is one half of the experimentally observed depth of compression, I, in accordance with the theoretical analysis of contact mechanics as explained below. H/2 is therefore one half of the contact height of the sample. Since the actual dimension of BCA II is ~6 x 4 x 4 nm (Saito et al. 2004), the value of H = 3.5 ± 0.1 nm was slightly less than the expected height of the protein. The result implied that the protein is not completely flattened under the force of 1–1.5 nN. The compression curve was reversible approximately up to 2 nm in the depth of compression.

View larger version (27K): [in this window] [in a new window]   Figure 4. (A) A collection of compressive curves obtained under the same conditions. The abscissa is the distance from the end of compression, and the ordinate is the compressive force. The pink, yellow, and cyan lines in the figure are best-fitting curves based on Equation 1 (Hertz model), Equation 2 (modified Hertz), and Equation 3' (Tatara model), respectively. (B) Four possible cases of protein compression. In each figure, the solid and dotted lines, respectively, represent relative positions of the tip, sample, and the substrate before and after compression. Symbols correspond to those defined in the text. Case 1 indicates a small deformation from both top and bottom side of a spherical sample; case 2, a small deformation from the top only (original Hertz model) of a mushroom shaped sample; case 3, a large deformation from the top and bottom (applied for the native protein) allowing lateral extension (indicated with horizontal arrows here and in case 4); and case 4, a large deformation from the top with lateral extension (applied for denatured protein).

(1)

where, and

When we apply the original Hertz model to obtain Young’s modulus of the protein molecule, we must remember that the model was derived for a semi-infinite elastic sphere, i.e., a large half sphere in a simple expression, assuming a small deformation at the contact site of the two spheres. In the analysis of compressive curves in Figure 4A, we considered four different possibilities, as shown schematically in the bottom part of Figure 4B. In cases 1 and 3, a protein molecule is compressed equally from the top and bottom with a small (1) or large (3) total deformation. The measured value of I in these cases must be halved to fit the experimental data to the theoretical curves because the Hertzian approach distance, I_H, used in theoretical models represents the depth of compression at a single contact site; in cases 2 and 4, a protein molecule is immobilized on the substrate with a large contact area compared with its contact site with the tip. In these cases, at least in the initial period of compression, the measured I can be taken as I_H and is applicable to denatured protein molecules on the substrate ("mushrooms" as later introduced).

Since Hertz model predicts 1.5th-order dependence of F on the Hertzian approach distance (I_H) (Johnson 1985), we first applied Equation 2 with a variable Young’s modulus, Y_app, where Y₀ and k are adjustable parameters, to reproduce the experimental curves for the initial estimation of the value of a:

(2)

where, Y_app = Y₀[1 + k exp (I_H)].

The resulting values of apparent Young’s modulus Y_app are plotted in Figure 5 as a function of I_H. The result indicates that for substantially denatured proteins (curves 4, 5, and 6), Y_app is almost constant and represents the Young’s modulus of the samples correctly, but for curves 1, 2, and possibly 3, we need more advanced analyses than the classical Hertz model.

View larger version (16K): [in this window] [in a new window]   Figure 5. Plot of apparent Young’s modulus of BCA II, Yapp, as a function of IH. The abscissa is the Hertzian approach distance, IH, defined as either IH = I/2 for curves 1 and 2, and IH = I for the rest of the curves, where I is the experimentally observed depth of indentation (see text for individual cases). Yapp was calculated applying Equation 2 to the experimental data. Numbers represent the following: 1, native; 2, complexed with an inhibitor; 3, in 1.0–1.5 M; 4, 2 M; 5, 3 M; and 6, 6 M GdmCl solutions.

(3)

where, .

Since a and a_c are related through Y₁, we can reduce Equation 3 to a single parameter equation of 3' below by assuming ₁ = 0.40, and use the latter equation for curve fitting to the experimental data.

(3')

The numerical coefficients of Equation 3' is only weakly dependent on the value of ₁ between 0.35 and 0.45. In the following analysis, we adopted ₁ = 0.40. The final value of Young’s modulus to be calculated from a then has an uncertainty range of ± 5% due to the (1 – ₁ ²) term for the range of ₁ = 0.35 to 0.45. Equation 3 is given by Tatara (1989) as an expansion formula of the implicit relation of I_H to F given below,

(4)

where, .

Since the Young’s modulus of silicon substrate is large, we can set a_c = 4Y₁R₁/(1+₁)(3–2₁) as already given. Tatara model was applied by Liu et al. (1998) for the analysis of compression curves of polyurethane spheres. Although both Hertz and Tatara models have been developed for materials with homogeneous and isotropic physical properties, we applied these models to nonhomogeneous and nonisotropic protein molecules to obtain a numerical measure of rigidity of BCA II during its denaturation process in GdmCl solutions.

In Figure 4A, experimental force curves are overlapped with the best-fitting curves to Equation 1 (pink), Equation 2 (yellow), and Equation 3' (cyan) in the middle part of the plot. The fitting curve based on Equation 3' reproduced the experimental data down to ~60%–70% of the observed depth of compression with a constant value of Y₁, which we call Y_exp, supporting the reliability of our estimate of Y_exp.

The Young’s modulus, Y_exp, obtained from the value of a in each fitting curve by assuming that the Poisson’s ratio = 0.40 and R = 2.3 nm(based on the estimated values of R₁ = 2.5 nm and R₂ = 30 nm), was 80 ± 10 MPa for Equation 1 in the initial 10% of the total compression depth, and 75 ± 10 MPa for both Equation 2 and Equation 3'. If we force Equation 1 to fit to 70% of the data starting from zero compression, Y_exp was increased to 150 ± 20 MPa, suggesting that we would get approximately twice as large Y_exp from a forced fitting of Equation 1 for a large deformation of protein molecules. Since, as noted above, Equation 3' could be fitted nicelyup to 60%– 70% of the total compression depth with a single parameter as shown in Figure 4A, we think it is the most reliable model and results cited in Table 1 are all based on Equation 3'. Deviation of the experimental points from Equation 3' is most likely due to nonconstant Young’s modulus of BCA II, but we need more precise data in higher compression regions, which will be left for future studies. In the following analysis of compression curves under different conditions, we used Equations 3' and 2 to obtain the value of a and calculated Y_exp, which covers 60%–70% of the data range in each case. The results are summarized in Table 1.

The Young’s modulus of BCA II obtained above is quite a low value compared with a previously reported value of 500 MPa for lysozyme adsorbed on mica (Radmacher et al. 1994). Structural differences of the two proteins and the experimental conditions (lysozyme was physically adsorbed instead of cross-linked to mica surface) may be responsible for the observed difference in the Young’s modulus (see Discussion).

When the enzyme inhibitor was added to the sample solution, the value of Y_exp became slightly lower than that of the native enzyme but the height was much reduced, indicating that the core of BCA II became more rigid than in the native state (Table 1).

The increased stability of the inhibitor (acetazolamide)-bound form of CA has been reported (Almstedt et al. 2004). The result of X-ray crystallography shows the distributionof the thermal B-factor in the native and inhibitor-bound form of human CA. In the native state, not surprisingly, the surface residues have larger B-factors compared with the internal residues, but in the inhibitor-bound form, the difference is more pronounced. For example, the results in Protein Data Bank (PDB) codes 1CAII and 1A42 [PDB] , respectively, for the native and inhibitor-bound forms show this tendency. If the B-factors can be taken as an indicator of relative rigidity, the rigidity of the surface residues is reduced, whereas that of internal residues enhanced. Mechanical rigidity of the inhibitor-bound form of BCA II, as cited in Table 1, shows a slightly reduced Young’s modulus in the initial stage of compression, but the total compression depth is only 60% of the native protein. We temporarily interpreted the result as indicating a softened surface residues and incompressibly rigidified internal core structure with the force of ~2nN. Compression of the protein molecule in both native and inhibitor-bound forms must be carried out, in the future, with an improved accuracy in a large force range.

Results in 1–1.5 M GdmCl
In 1–1.5 M GdmCl, BCA II is known to be in the "molten globule" state (Uversky 1993; Andersson et al. 2001; Bushmarina et al. 2001), and as expected, the pulling force curves under this condition showed that the protein yielded to much less force in the initial 30-nm extension and was stretched as a flexible chain thereafter up to almost the full contour length of the chain before the covalent cross-linking system was severed (Figure 6A). The result indicated BCA II lost the highly resistant -core structure observed under the native condition. Reference experiments using unmodified tips confirmed that the nonspecific adhesion force was always <0.05 nN. F–E curves revealed that the mechanical resistance of the protein against the tensile force was in the range of 0.2–0.3 nN up to 80 nm in extension, except in a few cases where there was still a remnant of force peaks in the 30-nm region. This is a clear indication of the loss of the substructures responsible for the strong mechanical resistance in the native protein, which was so prominent in Figure 3A.

View larger version (22K): [in this window] [in a new window]   Figure 6. (A) F–E curves under 1.5 M GdmCl showing various degrees of resistance against tensile force in the initial extension up to 40 nm, after which chains were extended as a flexible polymer. Inset is a collection of representative individual curves. (B) Collection of compressive curves with a fitting curve based on Equation 3' as a white line. The abscissa is the distance from the end of compression, and the ordinate is the compressive force.

Results in 2 and 3 M GdmCl
The pulling curves in 2 M GdmCl were similar to those obtained in 1.5 M GdmCl, in that they were characterized by initial plateau region of 0.2–0.3 nN followed by a temporal drop in the force value before the final steep rise toward the maximum extension of the chain (data not shown). In 3 M GdmCl, the tensile response of the protein was very similar to that of the fully denatured proteins in 6 M GdmCl in Figure 7A (data not shown).

View larger version (20K): [in this window] [in a new window]   Figure 7. (A) Collection of F–E curves in 6 M GdmCl fitted with the interpolation formula (Equation 5 in text) of the WLC model in red. Inset is a collection of representative individual curves. (B) Collection of compressing curves under the same conditions with a fitting curve based on Equation 3'' as a white line. The abscissa is the distance from the end of compression, and the ordinate is the compressive force.

(5)

The fitting curve in Figure 7A was obtained by adjusting P = 0.2 nm and E₀ = 90 nm. The result supports that the protein is fully denatured under this condition.

In Figure 7B, compression curves are overlapped with a fitting curve based on Equation 3'' given below, which has different coefficients from Equation 3' because of the differences in R₁ and R.

(3'')

The value of maximum compression, H (9 ± 1 nm), obtained from the above analysis is expected to be approximately equal to the height of the denatured protein on the solid surface because the data cluster at around x = 0 for a compressing force of 0.6–1.5 nN. Thus H corresponds to the maximum depth of compression and can be taken as the radius of denatured protein. Taking R₁ = 9 nm and R₂ = 30 nm, we obtain R = 6.9 nm. Y_exp of the denatured protein was in the range of 2 ± 0.5 MPa (I_H = I) (Fig. 5, curve 6). If the protein is assumed to be compressed from both top and bottom, Y_exp would be 5 ± 1 MPa.

It has been suggested that the height of randomly coiled polymer tethered on one end to a solid surface roughly corresponds to its radius. Since the unperturbed end-to-end distance of denatured BCA II is estimated to be in the range of 14–15 nm according to the equation given by Tanford (1968), the height probed by the AFM tip is close to one half of the end-to-end distance but significantly larger than the Stokes radius, R_S = 5.1– 5.2 nm as reported by Uversky (1993). The radius of gyration of randomly coiled polymer has been given as R_G = [3/2]R_S (Strobl 1997), which is equal to 7.7 nm in this case and close to the value of H. Therefore, we concluded that the cantilever we used in this work sensed the presence of denatured protein at a distance from the substrate close to either a half of the end-to-end distance, or the radius of gyration.

It should be added that the effect of the hard substrate on the compression experiment of small sample must be taken into consideration in a more precise work in the future (Akhremitchev and Walker 1999).

	Discussion

TOP Abstract Introduction Results Discussion Materials and methods References

 
Unfolding and refolding of globular proteins has been intensively studied to gain an understanding of the mechanisms of spontaneous creation of their three-dimensional conformations that have biological activity. BCA II has been regarded as one of the few examples providing clear evidence of the presence of intermediate states in the transition region, different from either the native or the fully denatured states. The intermediate state has been characterized by a variety of optical and hydrodynamic methods, but its mechanical rigidity has not been studied because such a property is only meaningful when measured at the single molecular level. The recent development of single molecule force spectroscopy has enabled us to measure the mechanical properties of individual molecules of proteins and other biological polymers. In this article, we probed the progressive softening of BCA II under mild to fully denaturing conditions by using an AFM.

Under native conditions, the pulling curve of BCA II was characterized by a short extension of 15–30 nm before the covalent system broke down. The inability of the chain to be extended beyond this level is most likely to be the result of a strong friction between the C-terminal and N-terminal chains in the knot region. From the compression curves, the molecule was shown to have a maximum compression depth of H ~ 3.5 ± 0.1 nm, which in this case is most likely to be less than the sample height, and by using Tatara equation, Y_exp was found to be in the range of 75 ± 10 MPa for ~60%of the compression range from the onset of compression. The large difference observed between the Young’s modulus of BCA II and that of lysozyme as measured by Radmacher et al. (1994) is probably due to, first, to the small size of the latter protein that is structurally reinforced by four disulfide bonds and, second, to the multipoint adhesion of lysozyme to the substrate, whereas BCA II in this work cross-linked to the substrate through a single cross-linker.

BCA II in the early transition region of 1.0–1.5 M GdmCl gave a similar or slightly lower value of H (~3 nm) than that of the native protein but had a much smaller starting value of Y_exp<20MPa, implying that the protein molecules were significantly flattened compared with the protein under native conditions but still not as softened as in 6 M GdmCl. A similar conclusion can be obtained from the analysis of pulling curves obtained under the same condition. Thus the AFM experiments suggested that the protein was in a different state from either the native or the denatured states, confirming the well-known fact of the non–two-state transition of denaturation of this protein induced by GdmCl. What is new in this work is the actual measurement of the F–E curves, together with the modulus of elasticity of the intermediate states of denaturation.

It is noteworthy that the presence of softened protein molecules was detected in AFM experiments even in the native condition, and that the relative population of the softened protein molecules increased as the concentration of GdmCl was increased. Such heterogeneity in the molecular population observed in pulling experiments can be pursued further in future study. As well, the extent of chain stretching at the final rupture point was short under native conditions for the majority of the molecules, but this length gradually increased as the concentration of the denaturant was increased from 1 to 3 M, where the majority of the protein molecules showed denatured type F–E curves. Both of these aspects in BCA II denaturation may have complex time dependencies and are left for future investigation.

Under fully denaturing conditions, each tethered molecule on a solid surface is expected to form a structure known in polymer physics as a "mushroom" (Strobl 1997; Jones and Richards 1999). The height of mushroom BCA II in 6 M GdmCl (~8–10 nm) was significantly greater than the Stokes radius of the same protein (~5 nm) under a similar denaturing condition. If we take the known relationship between R_S and R_G = (3/2)R_S known for an ideal Gaussian chain, we obtain R_G = 7.5 nm for an ideal Gaussian chain, which is closer to H than R_S is. Furthermore, the end-to-end distance of an unperturbed Gaussian chain of polypeptide having 260–290 amino acid residues can be estimated as ~14–15 nm according to the empirical equation given by Tanford (1968), which is also close to, but slightly less than, twice the experimental H obtained in 6 M GdmCl. Although the value of H is expected to depend on the force constant of the cantilever and the radius of the tip, such dependence was not studied in this work.

In summary, we have, for the first time, measured the mechanical softness of a globular protein under progressively denaturing conditions and proved that an AFM has the capability to follow the denaturation process at the single molecular level. It is especially interesting to have found that AFM successfully distinguished the intermediate states of denaturation of BCA II from either the native or a fully denatured form. Changes of physical properties of protein molecules either adsorbed or immobilized on various solid surfaces are attracting the attention of researchers concerned with proteins at interfaces. Our method will enlarge the repertoire of methods available to study such processes at the single molecular level.

	Materials and methods

TOP Abstract Introduction Results Discussion Materials and methods References

 
Protein
Recombinantly modified BCA II with two cysteine residues each at the N or the C terminus and a His₆-tag sequence outside of the N terminus containing extra 6 x His tag sequence was produced as described previously and purified by the metal chelating chromatography (Ni-NTA agarose) and affinity chromatography prepared with p-aminomethylbenzenesulfonamide followed by gel filtration chromatography (Alam et al. 2002). The total contour length between two cysteine residues is ~96 nm, assuming that the length of one amino acid residue is 0.37 nm. BCA II has recently been shown to have virtually the same three-dimensional structure as its human counterpart (Saito et al. 2004), and the modified protein used in this work had full enzymatic activity (Wang et al. 2001). The presence of extra His-tag residues interfered neither with the structure nor with the activity of the enzyme, and their effect was not explicitly considered in this work. We also confirmed by measuring the intrinsic fluorescence of the modified protein that it was denatured in the concentration range of 1–3 M of GdmCl as reported for wild-type protein (Wong and Tanford 1973; Bushmarina et al. 2001).

Chemicals
Succinimidyl 6-[3'(2-pyridyldithio)-propionamido]hexanoate (LC-SPDP) was purchased from Pierce and stored under highly dehydrated and low temperature conditions. The silanization reagent, 3-aminopropyltriethoxy silane (APTES), was purchased from Shin-Etsu Chemicals. GdmCl, p-aminomethylbenzenesulfonamide, and other chemicals were purchased from Sigma Chemical Co. All the reagents were used without further purification.

Silicon substrate and protein immobilization
Crystalline silicon wafers were purchased from Shin-Etsu Chemicals, cut into ~10 mm x 10 mm square pieces, and cleaned first in chloroform, then in ethanol, and finally in MilliQ water. Wafers were then treated in Piranha solution (H₂SO₄:H₂O₂ = 3 : 1) and washed with MilliQ water. They were further treated in an ozone cleaner for 2 min under O₂, then for 20 min UV irradiation, and finally for 6 min under N₂. Silanization with APTES was performed as described previously (Afrin et al. 2003). The silanized wafers were then derivatized with the hetero-bifunctional cross-linker LC-SPDP, which has the amino-reactive succinimidyl group and is expected to form a covalent cross-link with amino-silanized silicon surface. The other reactive group of the cross-linker, pyridyldithio group, is reactive toward -SH groups on the recombinant BCA II molecules, thus the cross-linker was used to covalently immobilize BCA II on the silanized silicon surface. The use of covalent cross-linkers to immobilize protein molecules to the substrate was an important step in this work because most of the AFM measurements were done in the presence of the denaturant GdmCl, which otherwise denatures and washes off noncovalently attached proteins on the substrate. The success of the immobilization of the protein on the silicon surface was confirmed by a TRIFT III TOF SIMS (time-of-flight secondary ion mass spectroscopy) apparatus, which demonstrated (m/z) peaks at 110, 120, and 130, corresponding to nitrogen containing fragments of C₆H₁₀N₂, C₈H₁₀N, and C₉H₈N, respectively (data not shown).

The flatness of the bare and APTES-modified silicon substrate was measured by imaging the surface roughness, which was found to be within ± 0.5 nm.

AFM experiments
NanoScope IIIa (Digital Instruments) was used at room temperature (~25°C) with the force curve mode. For force curve measurements, a silicon substrate with immobilized BCA II molecules was placed under 50 mM Tris-sulfate buffer (pH 7.5) containing various concentrations of GdmCl between 0 and 6 M. A modified tip with the cross-linker LC-SPDP was then brought in contact with the sample layer on the AFM. During their brief contact, covalent bonds were occasionally formed between the remaining -SH groups on immobilized BCA II and the dithiopyridyl groups on the tip. Covalent bond formation was verified after recording force curves that showed the presence of tensile material between the tip and the substrate and ended in a complete release of the cantilever with the final rupture force of ~1.0–2.5 nN (Afrin et al. 2003). To avoid simultaneous formation of more than two cross-links between the tip and the proteins, the probability of covalent bond formation against the total number of mutual approaches was kept to <20% by adjusting the protein concentration to <50 µg/mL in the immobilization step of BCA II. In compression experiments, an AFM tip was pushed onto the sample protein with the final force exceeding 1 nN to assure the maximum compression. When the tensile characteristics of the protein were probed in the retracting regime of the force curves, the initial compression force was kept as small as possible to guarantee reversible compression of the protein, i.e., in the range of 0.3–0.5 nN. In addition, since experiments were done in solutions having an ionic strength >100 mM, electrostatic contribution to the tip–sample interaction was considered negligibly small in the analysis of the force curves.

The silicon nitride NPS probes used in the AFM measurements were purchased from Digital Instruments (presently Veeco Instruments Japan). Triangle cantilevers with a nominal force constant of 0.06 N/m were used in this experiment. They were similarly functionalized with APTES and LC-SPDP as described above for silicon substrates. The spring constant of cantilevers used in the experiment was calibrated by the thermal vibration method (Hutter and Bechhoefer 1993). Collected force curves were analyzed using SPIP (Image Metrology), Microsoft Excel (Microsoft Corp.), and Igor Pro software (Wavemetrics).

	Acknowledgments

 
This work was supported by a grants-in-aid #99R16701 (Research for Future Program) and #15101004 (Basic Research [S]) from the Japan Society for Promotion of Science (JSPS) to A.I. R.A is a recipient of JSPS postdoctoral fellowship. We thank Alvac Phi Co. for conducting TOF SIMS measurements. We thank Professor Uno Carlsson of Linköping University (Sweden) for critically reading the manuscript.

	References

TOP Abstract Introduction Results Discussion Materials and methods References

 
Afrin, R., Arakawa, H., Osada, T., and Ikai, A. 2003. Extraction of membrane proteins from a living cell surface using the atomic force microscope and covalent crosslinkers. Cell Biochem. Biophys. 39: 101–117.[CrossRef][Medline]

Akasaka, K. 2003. Highly fluctuating protein structures revealed by variable-pressure nuclear magnetic resonance. Biochemistry 42: 10875–10885.[CrossRef][Medline]

Akhremitchev, B.B. and Walker, G.C. 1999. Finite sample thickness effects on elasticity determination using atomic force microscopy. Langmuir 15: 5630–5634.[CrossRef]

Alam, M.T, Yamada, T., Carlsson, U., and Ikai, A. 2002. The importance of being knotted: Effects of the C-terminal knot structure on enzymatic and mechanical properties of bovine carbonic anhydrase II. FEBS Lett. 519: 35–40.[CrossRef][Medline]

Almstedt, K., Lundqvist, M., Carlsson, J., Karlsson, M., Persson, B., Jonsson, B.H., Carlsson, U., and Hammarstrom, P. 2004. Unfolding a folding disease: Folding, misfolding and aggregation of the marble brain syndrome-associated mutant H107Y of human carbonic anhydrase II. J. Mol. Biol. 342: 619–633.[CrossRef][Medline]

Andersson, D., Hammarstrom, P., and Carlsson, U. 2001. Cofactor-induced refolding: refolding of molten globule carbonic anhydrase induced by Zn(II) and Co(II). Biochemistry 40: 2653–2661.[CrossRef][Medline]

Borén, K., Andersson, P., Larsson, M., and Carlsson, U. 1999. Characterization of a molten globule state of bovine carbonic anhydrase III: Loss of asymmetrical environment of the aromatic residues has a profound effect on both the near and far-UV CD spectrum. Biochim. Biophys. Acta 1430: 111–118.[CrossRef][Medline]

Bushmarina, N.A., Kuznetsova, I.M., Biktashev, A.G., Turoverov, K.K., and Uversky, V.N. 2001. Partially folded conformations in the folding pathway of bovine carbonic anhydrase II: A fluorescence spectroscopic analysis. Chembiochem. 2: 813–821.[CrossRef][Medline]

Bustamante, C, Marko, J.F., Siggia, E.D., and Smith, S. 1994. Entropic elasticity of lambda-phage DNA. Science 265: 1599–1600.[Free Full Text]

Cantor, C.R. and Schimmel, P.R. 1980. Structure of proteins. In Biophysical chemistry, pp. 41–154. W.H. Freeman and Company Springer-Verlag, New York, NY.

Carlsson, U. and Johnsson, B.H. 2000. Folding and stability of human carbonic anhydrase II. In The carbonic anhydrases: New horizons (eds. W.R. Chegwidden et al. ) pp. 241–259. Birkhauser Verlag, Basel, Switzerland.

Carrion-Vazquez, M., Oberhauser, A.F., Fowler, S.B., Marszalek, P.E., Broedel, S.E., Clarke, J., and Fernandez, J.M. 1999. Mechanical and chemical unfolding of a single protein: A comparison. Proc. Natl. Acad. Sci. 96: 3694–3699.[Abstract/Free Full Text]

Clausen-Schaumann, H., Rief, M., Tolksdorf, C., and Gaub, H.E. 2000. Mechanical stability of single DNA molecules. Biophys. J. 78: 1997–2007.[Abstract/Free Full Text]

Drobny, G.P., Long, J.R., Karlsson, T., Shaw, W., Popham, J., Oyler, N., Bower, P., Stringer, J., Gregory, D., Mehta, M., et al. 2003. Structural studies of biomaterials using double-quantum solid-state NMR spectroscopy. Annu. Rev. Phys. Chem. 54: 531–571.[CrossRef][Medline]

Elwing, H. 1998. Protein absorption and ellipsometry in biomaterial research. Biomaterials 19: 397–406.[CrossRef][Medline]

Eriksson, A.E., Jones, T.A., and Liljas, A. 1988. Refined structure of human carbonic anhydrase II at 2.0 Å resolution. Proteins 4: 274–282.[CrossRef][Medline]

Gekko, K. 2002. Compressibility gives new insight into protein dynamics and enzyme function. Biochim. Biophys. Acta 1595: 382–386.[CrossRef][Medline]

Håkansson, K., Carlsson, M., Svenson, L.A., and Lilijas, A. 1992. Structure of native and apo-carbonic anhydrase II and some of its anion-ligand complexes. J. Mol. Biol. 227: 1192–1204.[CrossRef][Medline]

Hertadi, R., Gruswitz, F., Silver, L., Koide, A., Koide, S., Arakawa, H., and Ikai, A. 2003. Unfolding mechanics of multiple Osp A substructures investigated with single molecule force spectroscopy. J. Mol. Biol. 333: 993–1002.[CrossRef][Medline]

Hutter, J.L. and Bechhoefer, J. 1993. Calibration of atomic-force microscope tips. Rev. Sci. Instrum. 64: 1868–1873.[CrossRef]

Johnson, J.K. 1985. Normal contact if elastic contacts: Hertz theory. In Contact mechanics, pp. 84–106. Cambridge University Press, Cambridge, UK.

Jones, R.A.L. and Richards, R.W. 1999. Tethered polymer chains in solutions and melts. In Polymers at surfaces and interfaces, pp. 244–292. Cambridge University Press, Cambridge, UK.

Karlsson, M., Mårtensson, L.-G., Jonsson, B.-H., and Carlsson, U. 2000. Adsorption of human carbonic anhydrase II variants to silica nanoparticles occurs stepwise: Binding is followed by successive conformational changes to a molten globule like state. Langmuir 16: 8470–8479.[CrossRef]

Li, H., Oberhauser, A.F., Fowler, S.B., Clarke, J., and Fernandez, J.M. 2000. Atomic force microscopy reveals the mechanical design of a modular protein. Proc. Natl. Acad. Sci. 97: 6527–6531.[Abstract/Free Full Text]

Liu, K.K, Williams, D.R., and Briscoe, B.J. 1998. The large deformation of a single micro-elastomeric sphere. J. Phys. D Appl. Phys. 31: 294–303.[CrossRef]

Mitsui, K., Hara, M., and Ikai, A. 1996. Mechanical unfolding of -2-macroglobulin with atomic force microscope. FEBS Lett. 385: 29–33.[CrossRef][Medline]

Möller, C., Fotiadis, D., Suda, K., Engel, A., Kessler, M., and Müller, D.J. 2003. Determining molecular forces that stabilize human aquaporin-1. J. Struct. Biol. 142: 369–378.[CrossRef][Medline]

Morozov, V.N. and Morozova, T.Y. 1981. Viscoelastic properties of protein crystals: Triclinic crystals of hen egg white lysozyme in different conditions. Biopolymers 20: 451–467.[CrossRef][Medline]

Radmacher, M., Fritz, M., Cleveland, J.P., Walters, D.A., and Hansma, P.K. 1994. Imaging adhesion forces and elasticity of lysozyme adsorbed on mica with the atomic force microscope. Langmuir 10: 3809–3814.[CrossRef]

Ramsden, J.R. 2002. Adsorption kinetics of proteins. In Encyclopedia of surface and colloid science (ed. A.T. Hubbard), pp. 240–261. Mercel Dekker Inc., New York, NY.

Rief, M., Oesterhelt, F., Heymann, B., and Gaub, H.E. 1997. Single molecule force spectroscopy on polysaccharides by atomic force microscopy. Science 275: 1295–1297.[Abstract/Free Full Text]

Saito, R., Sato, T., Ikai, A., and Tanaka, N. 2004. Structure of bovine carbonic anhydrase II at 1.95 Å resolution. Acta Crystallogr. D Biol. Crystallogr. 70: 792–795.

Smith, S.B., Cui, Y., and Bustamante, C. 1996. Overstretching B-DNA: The elastic response of individual double-stranded and single-stranded DNA molecules. Science 271: 795–799.[Abstract]

Strobl, G. 1997. Microscopic dynamical models. In The physics of polymers, 2d ed. p. 257–296. Springer-Verlag, New York, NY.

Suda, H., Sugimoto, M., Chiba, M., and Uemura, C. 1995. Direct measurement for elasticity of myosin head. Biochem. Biophys. Res. Commun. 211: 219–225.[CrossRef][Medline]

Tanford, C. 1968. Protein denaturation. In Advances in protein chemistry (eds. Anfinsen et al. ) Vol. 23, pp. 121–282. Academic Press, New York, NY.

Tatara, Y. 1989. Extensive theory of force-approach relations of elastic spheres in compression and impact. J. Eng. Mater. Tech. 111: 163–168.

———. 1991. On compression of rubber elastic sphere over a large range of displacements—Part 1: Theoretical study. J. Eng. Mater. Tech. 113: 285–291.

Tatara, Y., Shima, S., and Lucero, J.C. 1991. On compression of rubber elastic sphere over a large range of displacements—Part 2: Comparison of theory and experiment. 113: 292–295.

Tengvall, P., Lundstrom, I., and Liedberg, B. 1998. Protein adsorption studies on model organic surfaces: An ellipsometric and infrared spectroscopic approach. Biomaterials 19: 407–422.[CrossRef][Medline]

Uversky, V.N. 1993. Use of fast protein size exclusion liquid chromatography to study the unfolding of proteins which denature through the molten globule. Biochemistry 32: 13288–13298.[CrossRef][Medline]

Vanselow, D.G. 2002. Role of constraint in catalysis and high-affinity binding by proteins. Biophys. J. 82: 2293–2303.[Abstract/Free Full Text]

Vermeer, A.W.P. 2002. Conformation of adsorbed proteins. In Encyclopedia of surface and colloid science (ed. A.T. Hubbard), Vol. 1, pp. 1193–1212. Mercel Dekker Inc. New York, NY.

Wahlgren, M. and Arnebrant, T. 1991. Protein adsorption to solid surfaces. Trends Biotechnol. 9: 201–208.[CrossRef][Medline]

Wang, T., Arakawa, H., and Ikai, A. 2001. Force measurement and inhibitor binding assay of monomer and engineered dimer of bovine carbonic anhydrase B. Biochem. Biophys. Res. Commun. 285: 9–14.[CrossRef][Medline]

Wong, K-P. and Tanford, C. 1973. Denaturation of bovine carbonic anhydrase B by guanidine hydrochloride: A process involving separable sequential conformational transitions. J. Biol. Chem. 248: 8518–8523.[Abstract/Free Full Text]

CiteULike    Connotea    Del.icio.us    Digg    Reddit    Technorati    What's this?

This article has been cited by other articles:

				H. Li and C. Frieden Phenylalanine Side Chain Behavior of the Intestinal Fatty Acid-binding Protein: THE EFFECT OF UREA ON BACKBONE AND SIDE CHAIN STABILITY J. Biol. Chem., November 18, 2005; 280(46): 38556 - 38561. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager