Structure and dynamics of lysozyme in DMSO–water binary mixture: fluorescence correlation spectroscopy

Abstract

Proteins' structural stability and activity in aqueous–organic solvent mixtures is a fascinating topic of research in biochemistry. Here, the effect of dimethyl sulfoxide (DMSO) on the structure and conformational dynamics of lysozyme (labeled with alexa 488) is studied by fluorescence correlation spectroscopy (FCS). Circular dichroism and tryptophan emission indicate that DMSO molecules are accumulated around the tryptophan residues of lysozyme. According to the FCS data, the hydrodynamic radius (r_H) of the protein increases steeply from 18 Å at 0 mol% DMSO to 33 Å at 5 mol% DMSO. This suggests unfolding of the protein on addition of DMSO. On addition of DMSO beyond 5 mol%, size of lysozyme gradually decreases until 30 mol% of DMSO, and increases thereafter. The rate constants (k₊ and k₋) of fast folding and unfolding dynamics (contact between alexa 488 and amino containing residues e.g. tryptophan) show a chevron like plot on increasing DMSO concentration which indicates a two state pathway of folding dynamics.

---

1. Introduction

Water is universally believed to be the most favorable solvent for biological systems and biological activity. However, in recent years, non-aqueous solvents particularly, binary mixtures have been explored as biological solvents in many experiments, theory and simulations.^1–32 Among them the DMSO–water binary mixture has drawn major attention. Addition of DMSO to an aqueous solution of polymers and bio-molecules leads to interesting structural transitions. At low DMSO concentration, there is unfolding of protein which is followed by hydrophobic collapse (formation of a more compact structure because of interaction of hydrophobic residues with methyl groups) at higher concentration of DMSO.^10–12 The methyl groups of DMSO participate in hydrophobic aggregation while the S=O can accept hydrogen-bond. Interaction of DMSO and water leads to several anomalies such as viscosity maximum at an intermediate concentration of DMSO.^33,34 The actual experimental data indicates that viscosity maximum is at ∼35 mol% of DMSO.³⁴ This is ascribed to the formation of a 2 : 1 (water : DMSO) association complex.^33,34

In this work, we study structure and dynamics of a small protein, lysozyme, in DMSO–water mixtures. The 129 residues of lysozyme are arranged in 4 helices – A (5–15), B (25–36), C (88–101), and D (109–115) (Scheme 1A). Klibanov and co-workers showed that the refolding activity of lysozyme in DMSO–water mixture is maximum at ∼6 mol% DMSO (i.e. 20% v/v) in water.^1–3 Torreggiani et al. showed that thermally denatured lysozyme shows maximum enzyme activity at 10% (w/w) (2.5 mol%) of DMSO.¹⁷ Addition of DMSO to an aqueous solution of a protein reduces thermal stability of many proteins (e.g. NAD⁺ synthetase,⁴ RNAase,⁵ and lysozyme^6,7). NMR and Raman spectra indicate that addition of DMSO to an aqueous solution of lysozyme reduces the contribution of the hydrated turn, α-helix and the β-sheet.⁸ Bhattacharjya and Balaram, found that addition of 20% (v/v) (6 mol%) DMSO causes a broad structural transition of lysozyme to a partially unfolded state while at ∼20 mol% (i.e. 50% (v/v)) DMSO a partially folded state is formed with considerable modification of the tertiary structure.^28,29

Scheme 1 (A) Lysozyme (PDB ID-3ZEK). (B) Alexa Fluor 488 carboxylic acid, TFP ester.

A large scale computer simulation carried out by Bagchi and co-workers¹² reveal unfolding of Chicken Villin Headpiece (HP-36) as indicated by increase in size and root mean square displacement (RMSD) of C_α atoms from 0 to 10 mol% DMSO.¹² For the larger 129-residue protein, lysozyme, they simulated only the active site.¹⁰ RMSD of C_α atoms and SASA (solvent accessible surface area) of lysozyme undergoes a slight decrease up to 5 mol% DMSO, followed by a sharp increase from 5 to 15 mol% DMSO.¹⁰ Further, addition of DMSO up to 15 mol% causes an increase in loss of native like contact at the active site which is signature of partial unfolding.¹⁰

While the simulation¹⁰ considered only the effect of DMSO on the active site, other regions of the protein (lysozyme) may also be affected by DMSO. This may affect the size (r_H) and the conformational dynamics of lysozyme. In the present work, we have applied fluorescence correlation spectroscopy (FCS) to probe size (r_H) and the conformational dynamics of the protein lysozyme in DMSO–water mixture. FCS has recently been applied to study protein,^14,35,36 lipid,^37,38 vesicles³⁹ and biological cell.^40–42 For the FCS study, lysozyme (Scheme 1A) is covalently labeled by a fluorescent probe alexa fluor 488 TFP ester (Scheme 1B). For alexa 488 labeling, the major modification sites are Lys-33 (helix B) and Lys-97 (helix C).^43–46 The active site of lysozyme is the deep cleft between Glu-35 and Asp-52 (shown in orange, Scheme 1A). As shown in Scheme 1A, one of the locations of alexa probe (at Lys-33, shown in red cartoon, Scheme 1A) is close to the active site and the other (Lys-97, shown in red cartoon, Scheme 1A) is far from the active site.

2. Experimental section

2.1 Materials

Alexa fluor-488 carboxylic acid TFP ester and hen egg white lysozyme were purchased from Invitrogen and Fluka respectively. Both of them were used without further purification. Bio grade DMSO was used as the solvent. pH of DMSO–water mixtures was measured using Systronics Digital pH meter (model 335).

2.2 Methods

2.2.1 Circular dichroism (CD) spectra measurement. In 0% DMSO, the CD spectra were recorded on a JASCO J815 CD Spectrometer (model number J-815-1505) using ∼20 μM solution of unlabeled lysozyme in phosphate buffer with 0% and 0.005% Tween 20. In the presence of DMSO the CD spectra could not be recorded below 250 nm (helical region) because of strong absorption due to DMSO. In this case, the CD spectra of tryptophan region of lysozyme (315–270 nm) in mixtures of DMSO and phosphate buffer, was recorded using ∼60 μM unlabelled lysozyme.

2.2.2 Protein labeling and sample preparation for FCS experiment. For labeling of lysozyme by alexa 488 we followed the protocol described by Prieto and co-workers.^43,44 Briefly, required amount of the provided dye was added to a diluted solution of protein (2 mg mL⁻¹) in 0.1 M sodium bicarbonate buffer (pH ∼ 9.0) and the reaction mixture was stirred for 1 h at room temperature. The labeled protein was separated from free dye by column chromatography using PBS elution buffer (0.1 M potassium phosphate and 1.5 M NaCl) with 2 mM sodium azide (pH ∼ 7.2) (degree of labeling ∼ 70%). The first colored band was collected, which contains the labeled protein. For the FCS experiment, a diluted solution (∼1 nM) of probe labeled protein was prepared by dilution with the corresponding solvent mixtures (pH ∼ 7 phosphate buffer and DMSO).

In the FCS experiment, Tween-20 (0.005%, w/w) was used as a solvent additive to suppress attachment of the protein to the glass surface. The CD spectrum of the protein is found to be unchanged on addition of Tween 20 (ESI, Fig. S1†). This rule outs the possibility of structural change of the protein in the presence of Tween 20. FCS experiments were carried out at 20 °C, and laser power was kept at ∼50 μW. All experiments were repeated at least thrice.

2.2.3 FCS data measurement and analysis. For FCS studies, we have used very dilute solution (∼1 nM) of the protein. The number of molecules within the confocal volume is found to be ∼10. Our experimental setup for FCS and details regarding collection of FCS data are described in our earlier publication.⁴⁷

The autocorrelation function G(τ) of fluorescence intensity is defined as

where 〈F〉 is the average intensity and δF(τ) is the fluctuation in intensity at a delay time τ around the mean value, i.e., δF(τ) = 〈F〉 − F(τ).

We have fitted the FCS traces to two models. The quality of fit (residuals) to a free diffusion model and to a model involving single component diffusion and one component relaxation are given in ESI, Fig. S2.† The best fit is obtained for the model involving single component diffusion and one component relaxation. For this model, the autocorrelation function is given by,⁴⁸

where τ is the lag-time, N is the number of molecules in the observation volume, and τ_d is the diffusion time. F represents the amplitude of the relaxation time (τ_R) representing the fraction of molecules in the non-fluorescent state. ω denotes the structure parameter of the observation volume and is given by ω = ω_z/ω_xy, in which ω_z and ω_xy are the longitudinal and transverse radii of the observation volume, respectively. The known diffusion coefficient of Rhodamine 6G in water (426 μm² s⁻¹)⁴⁹ was used for calibrating the structure parameter (ω_xy ∼ 319 nm, ω ∼ 5). The confocal volume of our microscope setup is 0.9 fL.

In order to ascertain the contribution of blinking of the dye, the relaxation times were determined as a function of laser power (ESI, Fig. S3†). The lack of a systematic dependence of relaxation time on laser power conclusively rule outs the possibility of blinking.

The diffusion coefficient (D_t) is related to the diffusion time (τ_D) and transverse radius (ω_xy) as,

Assuming the spherical shape of lysozyme molecule, the hydrodynamic radius (r_H) can be obtained from diffusion coefficient (D_t) using the Stokes–Einstein relation,

The viscosity (η) of DMSO is 2.2 times higher than that of water and the refractive index of DMSO (1.479) is about 10% higher than that of water (1.33). For these systems, refractive index mismatch and viscosity effect have been corrected following the method discussed by Chattopadhyay et al.⁴⁸ and Sherman et al.⁵⁰ In order to account for the changes in refractive index and viscosity we used R6G as a diffusion standard (Table S1 and Fig. S4, ESI†) and the following equation.⁴⁸

where r_H and τ_D denote hydrodynamic radius and characteristic diffusion time, respectively.

3. Results

3.1 Circular dichroism (CD) spectra of lysozyme

Fig. 1 shows the effect of DMSO on the CD spectra of lysozyme in the tryptophan region. On addition of DMSO, there is a perceptible blue shift of the 287 nm band (ESI, Fig. S5†). This indicates decrease of local polarity presumably because of accumulation of DMSO around the tryptophan(s). As shown in Fig. 4, at pH ∼ 7 the ellipticity at 293 nm (due to the tryptophan residues) increases with increase in DMSO concentration up to 30 mol% DMSO. This suggests that the random coil structure around tryptophan region of lysozyme increases up to 30 mol% DMSO and decreases at higher DMSO concentration.

Fig. 1 Variation of ellipticity of CD spectra of lysozyme with mol% of DMSO.

The effect of addition of DMSO to the CD band at 208 nm (due to the helical structure) could not be monitored because of intense absorption of DMSO below 250 nm.

3.2 Blue shift of tryptophan emission on addition of DMSO

Fig. 2 shows the effect of DMSO on the tryptophan emission maximum of lysozyme in DMSO–water mixtures. It is readily seen that addition of DMSO causes a blue shift in the position of the emission maximum of the tryptophan residues from 339 nm in water (pH ∼ 7 phosphate buffers) to 332 nm in 5 mol% and to 327 nm in 30 mol% DMSO. This once again indicates that the methyl groups of DMSO molecules preferentially localize in the vicinity of the tryptophan residues causing a reduction of polarity of the environment around the tryptophan residues.

Fig. 2 Emission spectra of unlabelled lysozyme (λ_ex = 280 nm) in water as a function of mol% of DMSO.

3.3 FCS data: size of alexa 488 labeled lysozyme in DMSO–water mixture

Fig. 3 shows the normalized FCS traces of alexa-488 labeled hen egg white lysozyme in DMSO–water binary mixture at different mol% of DMSO (χ_DMSO). Table 1 summarizes the pH and viscosity of the mixtures, conformational relaxation time of lysozyme (τ_r) and hydrodynamic radius (r_H) of lysozyme in DMSO–phosphate buffer mixtures. Diffusion time of lysozyme (τ_D), amplitude of the relaxation time (F) and diffusion time of R6G (τ^R6G_D) are given in ESI, Table S1.†

Fig. 3 Normalized FCS traces of alexa-488 labeled lysozyme in 0 mol% (black), 5 mol% (red), 20 mol% (green) and 40 mol% (blue) in DMSO–water mixture.

Table 1 FCS data of lysozyme in DMSO–water mixture along with pH and viscosity

Mol% DMSO	pH^a	ηbulk (at 293 K) in cP^34	τr^b (μs)	rH^c (Å)	k+ (m s^−1)	k− (m s^−1)	Kfolding	ΔGunfolding (kJ mol^−1)
a ±0.1.b ±10 μs.c ±2 Å.
0	7.0	1.002	65	18	2.92	12.46	0.23	−3590.77
1.31	7.1	—	85	22	2.82	8.94	0.32	−2854.47
2.74	7.2	—	105	27	3.24	6.29	0.52	−1642.57
5	7.2	—	150	33	2.27	4.4	0.52	−1642.57
10	7.4	2.152	125	25	2.56	5.44	0.47	−1866.63
15	7.4	—	105	21	1.9	7.62	0.25	−3432.99
20	7.4	3.455	60	20	3.25	13.42	0.24	−3511.12
30	7.7	4.310	40	18	4	21	0.19	−4106.40
40	8.0	4.267	60	20	3.17	13.5	0.23	−3590.77

---

Fig. 4 describes the variation of r_H of lysozyme in DMSO–water mixtures. On addition of DMSO, size of lysozyme increase from 18 Å in 0% DMSO to 33 Å in 5 mol% DMSO followed by a decrease to 18 Å in 30 mol% DMSO. The increase in size at lower concentration of DMSO indicates unfolding of lysozyme.

Fig. 4 Lysozyme in DMSO–water mixture: variations of (A) hydrodynamic radius (r_H); (B) relaxation time (τ_R); and (C) amplitude of the non-fluorescent state (F).

3.4 Conformational dynamics of alexa 488 labeled lysozyme in DMSO–water mixture

In this section, we will discuss the effect of DMSO on the conformational dynamics of lysozyme. The relaxation time of conformational dynamics (τ_R) was obtained by fitting the FCS data to eqn (2). The time constant (τ_R) for lysozyme in DMSO–water mixture are listed in Table 1.

In the native state, lysozyme displays a relaxation time of 65 μs. This is similar to the 55 μs component detected by Majima and co-workers for Cyt C.⁵¹ Chattopadhyay and co-workers found a single relaxation component of 20–50 μs for Cyt C.^51–53 The variation of the relaxation component for alexa labeled lysozyme with mol% of DMSO is shown in Fig. 4 and is listed in Table 1. It is evident that the time component exhibits an initial rise from 65 μs at 0 mol% DMSO to 150 μs at ∼5 mol% DMSO and then decrease to 40 μs at 30 mol% DMSO (Fig. 4 and Table 1). F (contribution of non-fluorescent state) increases from 19% at 0 mol% DMSO to 34% at 5 mol% DMSO followed by a decrease to 16% at 30 mol% DMSO (Fig. 4).

4. Discussion

Before discussing the effect of DMSO on alexa labeled lysozyme, it may be emphasized that neither alexa-labeling nor slight change in pH due to addition of DMSO affects the structure of the protein. Melo et al.⁴⁴ has previously reported that labeling of lysozyme with alexa 488 keeps the CD spectra intact (Fig. S3 of ESI of ref. 42) and hence, does not affect the structure of the protein. Addition of DMSO causes slight increase in pH because of changes in pK_a of the buffer and the protein. The pH of DMSO–water mixtures increase from 7 at 0 mol% DMSO to 8 at 40 mol% DMSO (Table 1). Calorimetric studies indicate that the slight change in pH on addition of DMSO does not affect the thermodynamics of preferential solvation of lysozyme.⁶

Lysozyme contains six tryptophan residues (Trp-28, Trp-62, Trp-63, Trp-108, Trp-111 and Trp-123). Of these only two tryptophan residues (Trp-62 and 108) contribute to tryptophan emission.⁵⁴ Among the emitting tryptophan residues Trp-62 is near the active site of lysozyme while Trp-108 is not at the active site. The blue shift of tryptophan emission indicates that DMSO invades both the active site and other regions of the lysozyme. Note, clustering of the hydrophobic DMSO molecule at the active site of lysozyme at 5 mol% DMSO has also been detected in recent computer simulation and has been shown to give rise to a decrease in the solvent accessible surface area (SASA).¹⁰

In a small angle neutron scattering (SANS) study, Voets et al. found size (radius of gyration) of lysozyme remains unchanged up to ∼37 mol% (i.e. 70% (v/v)) DMSO.¹³ Voets et al. however, did not study any points between 0 mol% and 10 mol% (30% v/v) DMSO. We detected substantial fluctuation in size of lysozyme in this region (0–10 mol% DMSO).

Addition of 5 miol% DMSO causes a ∼1.7-fold increase in the size (r_H) of the protein (lysozyme) from 18 Å to 33 Å (Table 1). This indicates substantial amount of unfolding. According to Wilkins et al.⁵⁵ r_H of a protein is 4.75N^0.29 in the native state and 2.21N^0.57 in the fully denatured state. For lysozyme, N = 129. Thus, for lysozyme from Wilkins formula, r_H ≈ 19 Å, in the native state. For lysozyme in the native state (i.e. 0% DMSO), r_H is determined from the FCS data (Table 1 and Fig. 4), to be 18 Å. This is in very good agreement with the r_H obtained from the Wilkins formula, crystallography⁵⁶ and size exclusion chromatography.⁵⁷ On addition of DMSO up to 5 mol%, r_H of lysozyme increases steeply to 33 Å. According to Wilkins formula r_H of lysozyme in its fully denatured state should be 35 Å. Thus the protein is almost fully unfolded in 5 mol% DMSO. This is quite surprising because of the fact that lysozyme contain four di-sulfide bonds.¹⁹ Note, a recent MD simulation study¹⁰ shows that up to 15 mol% DMSO concentration there is a loss of native contact of lysozyme, which is a signature of unfolding. They also found that up to 15 mol% DMSO concentration, the methyl group of DMSO preferentially solvate the hydrophobic residue of enzyme.¹⁰ Using near UV-circular dichroism, Balaram and coworkers also indicated a broad structural transition of lysozyme to a partially unfolded state at 6 mol% (i.e. 20% v/v) DMSO concentration.^28,29 It seems that the favorable interaction of the hydrophobic residues (tryptophan) with DMSO is the driving force for the incorporation of the DMSO near the tryptophan residues. This causes unfolding of the protein and reduction in local polarity (i.e. blue shift of tryptophan emission).

Above 5 mol% DMSO, the size of lysozyme decreases and assumes native like size (r_H ∼ 18 Å) at 30 mol% DMSO. Thus the overall structure of lysozyme becomes smaller and hence, more compact above 5 mol% DMSO. The CD spectra also support this. However, presence of large number of DMSO at the active site (as evidenced by blue shift of CD and emission spectra) may impair the enzyme activity. On further addition of DMSO, r_H of lysozyme gradually increases again. Note, the variation of size of lysozyme on addition of DMSO is very different from that of the protein bovine serum albumin which exhibits a monotonic increase in hydrodynamic radius from 3.7 nm in 0% DMSO to 7 nm in 40 mol% DMSO.⁹ In another study, we have found that hydrodynamic radius of lysozyme shows non-monotonic and oscillating variation in ethanol–water binary mixture.⁵⁸

According to Bryngelson and Wolynes, protein folding affects two parameters – fraction of native contacts (η) and radius.^59,60 Using large scale MD simulation, Bagchi and co-workers¹⁰ showed that at the active site of lysozyme, there is an increase in the loss of native contact and total contact on addition of DMSO up to 15 mol%. In 15–20 mol% DMSO concentration, native contacts are regenerated and the overall structure suddenly collapsed due to the solvent effect.¹⁰ In this concentration region, the hydrophobic residues of lysozyme are sequestered in core structure due to “hydrophobic zipping”.¹⁰ It is obvious that addition of DMSO at low concentration disrupts native like contact which is indicated in increase in size up to 5 mol% DMSO. But at high concentration of DMSO size decreases because of collapse. The small size of lysozyme in 15–30 mol% DMSO concentration regions is due to collapse not refolding. This is manifested in differences in relaxation time and folding rate constants.

Finally, we discuss the effect of DMSO on the time constant for the fast folding dynamics of the protein chain from relaxation time (τ_R) given in eqn (2). The fluctuation in the fluorescence intensity (and hence FCS data) arises from quenching of the fluorescence of the probe (alexa 488) by amino groups of tryptophan and other residues.^47–53,61 As shown in Scheme 2, in the open conformation of the protein, the fluorescence probe (P) and the quenching groups (amino groups of tryptophan and other residues) are far apart and hence the open form is fluorescent. The closed form is non-fluorescent because of rapid electron transfer quenching. If k₊ and k₋ denote the rate constants of the formation of non-fluorescent state (open-to-closed inter-conversion) and fluorescent state (closed-to-open inter-conversion), respectively, in the case of two-states at equilibrium (open ↔ closed) one can write,⁶¹

where K_folding is the equilibrium constant and F is the average fraction of molecules in the non-fluorescent state. Evidently,

Scheme 2

From the equilibrium constant, we can find out the free energy of unfolding (closed to open transition) using the equation, ΔG_unfolding = −RT ln(1/K_folding).

We have previously showed that τ_R and F vary non-monotonically over the entire DMSO concentration. Interestingly we observed that (1/τ_R), the folding rate (inter-conversion rate, k₊) and unfolding rate constant (inter-conversion rate, k₋) show chevron like plot with increasing DMSO concentration (Fig. 5). This chevron plot indicates that transition between open-to-close state does not involve any intermediate (i.e. it undergo a two state pathway).

Fig. 5 Lysozyme in DMSO–water mixture: variations of inverse of relaxation time (1/τ_R), rate of formation of closed form (k₊), rate of formation of open form (k₋) and free energy of unfolding (ΔG_unfolding). Solid lines represent the B-splined line joining the data points.

We have also calculated the unfolding free energy (ΔG_unfolding). Contrary to our expectation, ΔG_unfolding shows a maximum around 5 mol% DMSO concentration. The maximum value of ΔG_unfolding indicates that unfolding is least favorable at 5 mol% DMSO. At 5 mol% DMSO the size of the protein is largest. At the large size of protein, already most of the native like contacts are ruptured.¹⁰ As a result the free energy difference of the closed and open form will be lower at 5 mol% DMSO.

5. Conclusion

This study demonstrates that the FCS probe the overall structure of the protein lysozyme and monitors the DMSO-induced change both at the active site and at other regions of the protein. Circular dichroism and emission spectra indicates that DMSO resides near the hydrophobic tryptophan residues (at the active site and away from it) presumably because of interaction of hydrophobic residues with methyl groups of DMSO. Large scale invasion of DMSO into lysozyme causes unfolding (change in size) of lysozyme resulting in a 1.7 fold increase in r_H from 18 Å at 0% to 33 Å at 5 mol% DMSO. Beyond 5 mol% DMSO the size of the protein decreases because of hydrophobic collapse as result of interaction of the methyl groups of DMSO with the hydrophobic residues. The folding rate (inter-conversion from open to closed i.e. contact of probe and amino group) and unfolding rate (inter-conversion from closed to open) constant show a chevron like plot as a function of DMSO concentration. This indicates a two state pathway of folding dynamics.

Acknowledgements

Thanks are due to Department of Science and Technology, India (Centre for Ultrafast Spectroscopy and Microscopy Project and J. C. Bose Fellowship) and the Council of Scientific and Industrial Research (CSIR) for generous research support. S.G., S.C. and R.C. thank CSIR for awarding fellowships.

References

1. A. M. Klibanov, Nature, 2001, 409, 241–246.
2. T. Knubovets, J. J. Osterhout, P. J. Connolly and A. M. Klibanov, Proc. Natl. Acad. Sci. U. S. A., 1999, 96, 1262–1267.
3. R. V. Rariy and A. M. Klibanov, Proc. Natl. Acad. Sci. U. S. A., 1997, 94, 13520–13523.
4. Z. W. Yang, S. W. Tendian, W. M. Carson, W. J. Brouillette, L. J. Delucas and C. G. Brouillette, Protein Sci., 2004, 13, 830.
5. A. L. Jacobson and C. L. Turner, Biochemistry, 1980, 19, 4534.
6. E. L. Kovrigin and S. A. Potekhin, Biochemistry, 1997, 36, 9195.
7. A. Giugliarelli, M. Paolantoni, A. Morresi and P. Sassi, J. Phys. Chem. B, 2012, 116, 13361–13367.
8. E. P. G. Areas, J. A. G. Areas, J. Hamburger, W. L. Peticolas and P. S. Santos, J. Colloid Interface Sci., 1996, 180, 578–589.
9. A. Pabbathi, S. Patra and A. Samanta, ChemPhysChem, 2013, 14, 2441–2449.
10. S. Roy, B. Jana and B. Bagchi, J. Chem. Phys., 2012, 136, 115103.
11. S. Banerjee, S. Roy and B. Bagchi, J. Phys. Chem. B, 2010, 114, 12875–12882.
12. S. Roy and B. Bagchi, J. Phys. Chem. B, 2013, 117, 4488–4502.
13. I. K. Voets, W. A. Cruz, C. Moitzi, P. Lindner, E. P. G. Areas and P. Schurtenberger, J. Phys. Chem. B, 2010, 114, 11875–11883.
14. H. P. Lu, Phys. Chem. Chem. Phys., 2011, 13, 6734–6749.
15. C. Gögelein, D. Wagner, F. Cardinaux, G. Nägele and S. U. Egelhaaf, J. Chem. Phys., 2012, 136, 015102.
16. M. Jackon and H. H. Mantsch, Biochim. Biophys. Acta, 1991, 1078, 231–235.
17. A. Torreggiani, M. Di Foggia, I. Manco, A. De Maio, S. A. Markarian and S. Bonora, J. Mol. Struct., 2008, 891, 115–122.
18. S. Roy and B. Bagchi, J. Chem. Phys., 2013, 139, 034308.
19. L. R. Nemzer, B. N. Flanders, J. D. Schmit, A. Chakrabarti and C. M. Sorensen, Soft Matter, 2013, 9, 2187–2196.
20. Y. L. Khmelnitskyl, A. B. Belova, A. V. Levashov and V. V. Mozhaev, FEBS Lett., 1991, 284, 267–269.
21. N. Byrne and C. A. Angell, J. Mol. Biol., 2008, 378, 707–714.
22. R. M. Parodi, E. Bianchi and A. Ciferri, J. Biol. Chem., 1973, 248, 4047–4051.
23. T. A. Page, N. D. Kraut, P. M. Page, G. A. Baker and F. V. Bright, J. Phys. Chem. B, 2009, 113, 12825–12830.
24. W. Dzwolak, J. Kalinowski, C. Johannessen, V. Babenko, G. Zhang and T. A. Keiderling, J. Phys. Chem. B, 2012, 116, 11863–11871.
25. V. Vagenende, M. G. S. Yap and B. L. Trout, Biochemistry, 2009, 48, 11084–11096.
26. S. Singh, S. K. Srivastava and D. K. Singh, RSC Adv., 2013, 3, 4381–4390.
27. I. T. S. Li and G. C. Walker, J. Am. Chem. Soc., 2010, 132, 6530–6540.
28. S. Bhattacharjya and P. Balaram, Proteins: Struct., Funct., Genet., 1997, 29, 492–507.
29. K. A. Satish, S. C. Shankaramma, S. Raghothama and P. Balaram, Biopolymers, 2001, 58, 465–476.
30. D. Hu, J. Yu, K. Wong, B. Bagchi, P. J. Rossky and P. F. Barbara, Nature, 2000, 405, 1030–1033.
31. V. Nikolakis, S. H. Mushrif, B. Herbert, K. S. Booksh and D. G. Vlachos, J. Phys. Chem. B, 2012, 116, 11274–11283.
32. M. Wang and K. Warncke, J. Am. Chem. Soc., 2008, 130, 4846–4858.
33. C. Nieto-Draghi, J. B. Ávalos and B. Rousseau, J. Chem. Phys., 2003, 119, 4782–4789.
34. S. A. Schichman and R. L. Amey, J. Phys. Chem., 1972, 76, 98–102 and the references there in.
35. A. Hoffmann, K. Neupane and M. T. Woodside, Phys. Chem. Chem. Phys., 2013, 15, 7934–7948.
36. R. Chowdhury, S. Chattoraj, S. Sen Mojumdar and K. Bhattacharyya, Phys. Chem. Chem. Phys., 2013, 15, 16286–16293.
37. T. Motegi, H. Nabika and K. Murakoshi, Phys. Chem. Chem. Phys., 2013, 15, 12895–12902.
38. S. Sen Mojumdar, S. Ghosh, T. Mondal and K. Bhattacharyya, Langmuir, 2012, 28, 10230–10237.
39. S. Ghosh, A. K. Mandal, A. K. Das, T. Mondal and K. Bhattacharyya, Phys. Chem. Chem. Phys., 2012, 14, 9749–9757.
40. D. K. Sasmal, S. Ghosh, A. K. Das and K. Bhattacharyya, Langmuir, 2013, 29, 2289–2298.
41. S. Ghosh, S. Chattoraj, T. Mondal and K. Bhattacharyya, Langmuir, 2013, 29, 7975–7982.
42. S. Ghosh, S. Chattoraj and K. Bhattacharyya, J. Phys. Chem. B, 2014 DOI:10.1021/jp412631d.
43. A. Coutinho, L. M. S. Loura, A. Fedorov and M. Prieto, Biophys. J., 2008, 95, 4726–4736.
44. A. M. Melo, M. Prieto and A. Coutinho, Biochim. Biophys. Acta, 2011, 1808, 2559–2568.
45. D. Suckau, M. Mak and M. Przybylski, Proc. Natl. Acad. Sci. U. S. A., 1992, 89, 5630–5634.
46. D. C. Philllips, Proc. Natl. Acad. Sci. U. S. A., 1967, 57, 484–495.
47. S. Sen Mojumdar, R. Chowdhury, S. Chattoraj and K. Bhattacharyya, J. Phys. Chem. B, 2012, 116, 12189–12198.
48. K. Chattopadhyay, S. Saffarian, E. L. Elson and C. Frieden, Biophys. J., 2005, 88, 1413–1422.
49. Z. Petrasek and P. Schwille, Biophys. J., 2008, 94, 1437–1448.
50. E. Sherman, A. Itkin, Y. Y. Kuttner, E. Rhoades, D. Amir, E. Haas and G. Haran, Biophys. J., 2008, 94, 4819–4827.
51. J. Choi, S. Kim, T. Tachikawa, M. Fujitsuka and T. Majima, Phys. Chem. Chem. Phys., 2011, 13, 5651–5658.
52. S. Haldar, S. Mitra and K. Chattopadhyay, J. Biol. Chem., 2010, 285, 25314–25323.
53. S. Haldar and K. Chattopadhyay, J. Biol. Chem., 2012, 287, 11546–11555.
54. T. Imoto, L. S. Forster, J. A. Rupley and F. Tanaka, Proc. Natl. Acad. Sci. U. S. A., 1971, 69, 1151–1155.
55. D. K. Wilkins, S. B. Grimshaw, V. Receveur, C. M. Dobson, J. A. Jones and L. J. Smith, Biochemistry, 1999, 38, 16424–16431.
56. P. Prinsen and T. Odijk, J. Chem. Phys., 2006, 125, 074903.
57. V. N. Uversky, Biochemistry, 1993, 32, 13288–13298.
58. S. Chattoraj, A. K. Mandal and K. Bhattacharyya, J. Chem. Phys., 2014 , in press.
59. J. D. Bryngelson and P. G. Wolynes, J. Phys. Chem., 1989, 93, 6902–6915.
60. J. D. Bryngelson, J. N. Onuchic, N. D. Socci and P. G. Wolynes, Proteins: Struct., Funct., Genet., 1995, 21, 167–195.
61. H. Neuweiler, M. Lollmann, S. Doose and M. Sauer, J. Mol. Biol., 2007, 365, 856–869.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c4ra00719k

---