Quantitative Evaluation of Myoglobin Unfolding in the Presence of

Dec 19, 2013 - Denaturation by guanidinium hydrochloride was monitored by loss of heme absorptivity and limited unfolding structural information was ...
2 downloads 0 Views 1MB Size
Download PDF
Article pubs.acs.org/JPCB

Quantitative Evaluation of Myoglobin Unfolding in the Presence of Guanidinium Hydrochloride and Ionic Liquids in Solution Olivia C. Fiebig,† Emily Mancini,† Gregory Caputo,†,‡ and Timothy D. Vaden*,† †

Department of Chemistry and Biochemistry and ‡School of Biomedical Sciences, Rowan University, 201 Mullica Hill Road, Glassboro, New Jersey 08028, United States S Supporting Information *

ABSTRACT: The use of ionic liquids in biochemical and biophysical applications has increased dramatically in recent years due to their interesting properties. We report results of a thermodynamic characterization of the chaotrope-induced denaturation of equine myoglobin in two different ionic liquid aqueous environments using a combined absorption/fluorescence spectroscopic approach. Denaturation by guanidinium hydrochloride was monitored by loss of heme absorptivity and limited unfolding structural information was obtained from Förster resonance energy transfer experiments. Results show that myoglobin unfolding is generally unchanged in the presence of ethylmethylimidazolium acetate (EMIAc) in aqueous solution up to 150 mM concentration but is facilitated by butylmethylimidazolium boron tetrafluoride (BMIBF4) in solution. The presence of 150 mM BMIBF4 alone does not induce unfolding but destabilizes the structure as observed by a decrease in threshold denaturant concentration for unfolding and an 80% decrease in the magnitude of ΔGunfolding from 44 kJ/ mol in the absence of BMIBF4 to 8 kJ/mol in the presence of 150 mM BMIBF4. Thus, the BMIBF4 significantly destabilizes the myoglobin structure while the EMIAc does not, likely due to differences in anion interaction capabilities. This is confirmed with control studies using NaAc and LiBF4 solutions. EMIAc may be chosen as cosolvent additive with minimal effects on protein structure while BMIBF4 may be used as a supplement in protein folding experiments, potentially allowing access to proteins which have been traditionally difficult to denature as well as designing ionic liquids to match protein characteristics. absorption/fluorescence studies,16 and molecular dynamics simulations.8,11 In aqueous solution, ILs become electrolytes, and in many systems, the IL molecular cation and anion have different effects on structure and stability. Hydrophobic cations like those in imidazolium-based ILs generally increase solution viscosity17 and decrease protein structure stability.10 The anion can potentially directly interact with proteins and peptides, especially if it is a Bronsted base (e.g., acetate or TFSI),10,13 but can also exhibit repulsive or negligible interactions.4,10,14 Several examples of the effects of various ionic liquids on protein stability have been reported for the well-characterized protein ribonuclease A and hen egg white lysozyme.18−21 However, the effects of ILs in solution on the inherent thermodynamic stabilities of proteins have in many cases not been quantified. Protein denaturation experiments have provided tremendous insights into the thermodynamics and kinetics of protein folding.21−28 The thermodynamic stability of a protein can be measured with denaturant-induced unfolding using, for example, guanidinium hydrochloride (GuHCl).29−31 Myoglo-

1. INTRODUCTION Scientific studies of room-temperature ionic liquids (ILs) have been continually increasing in depth and complexity over the past 15 years. Their unique properties such as negligible vapor pressures, low melting points, nonflammability, potential utility in green chemistry, good solvation of many organic and inorganic chemicals, and high ionic conductivities make them attractive options as solvents and cosolvent additives.1,2 ILs are used in electrochemistry and organic chemistry as both solvent and catalysts.1−3 While their utility to organic/inorganic chemistry and materials science is well-known, ILs have recently received significant attention as novel solution environments for biomolecules in various biomedical and biochemical applications.4 As cosolvent additives in aqueous solution, they have been used to inhibit (or promote) amyloid fibril formation,5,6 solubilize biomolecules7,8 and modulate enzyme activity9−11 for pharmaceutical applications. In most previously reported studies, IL applications require biomolecules to be structurally intact in the presence of IL molecular ions.12 It is known that high concentrations of some ILs in solution can induce protein unfolding, but other ILs can also be used with protein structures remaining intact.13 Recent studies have reported on the stability of protein structure in IL environments using circular dichroism spectroscopy,14,15 © 2013 American Chemical Society

Received: August 12, 2013 Revised: December 15, 2013 Published: December 19, 2013 406

dx.doi.org/10.1021/jp408061k | J. Phys. Chem. B 2014, 118, 406−412

The Journal of Physical Chemistry B

Article

bin, a heme-containing protein used in oxygen storage in muscles, has been the subject of intense examination with regard to protein structure, and its thermodynamic stability has been quantified by monitoring the change in heme absorption at 409 nm as a function of GuHCl-induced unfolding.32−35 The structural aspects of unfolding have also been studied with tryptophan-to-heme Förster resonance energy transfer (FRET) measurements.32,36 The myoglobin folding−unfolding phase transition has been investigated in solutions containing IL additives using the heme absorption, and GuHCl-assisted protein unfolding in imidazolium-based IL solutions has been studied for human serum albumin.12,15 However, in systems in which the protein retains its structure in IL solutions, the effect of the electrolytes on tertiary structure thermodynamic stability in these IL-containing mixed-solvent systems has not been quantified. In this work, we present the results of a GuHCl-induced unfolding experiment of myoglobin in aqueous solutions containing two well-characterized ILs as cosolvent additive electrolytes: ethylmethylimidazolium acetate (EMIAc)12,37 and butylmethylimidazolium boron tetrafluoride (BMIBF4).38,39 The imidazolium cation (EMI+ or BMI+) exhibits repulsive forces and likely destabilizes protein structures.10,18,20 The major difference between the two ILs is their anion: The acetate anion (Ac−) can interact with myoglobin via hydrogen bonds while the boron tetrafluoride anion (BF4−) most likely exhibits repulsive protein−IL forces, consistent with previous findings.10,18,20 By characterizing myoglobin unfolding in the presence of the two different ILs as cosolvent additives in aqueous solution, we can understand and quantify the different effects of the two IL molecular anions on protein unfolding and stability. The results show that moderate IL concentrations do not induce myoglobin unfolding and with EMIAc do not appear to affect unfolding process. However, BMIBF4 decreases the GuHCl concentrations at which protein unfolding occurs. From a standard thermochemical analysis we directly measure ΔGunfolding, the protein thermodynamic stability, in the environment of the IL solutions. This parameter is not strongly affected by EMIAc but dramatically decreases with increasing BMIBF4 concentration.

solutions. No decomposition attributable to HF generation was detected. Each prepared sample was measured with absorption and fluorescence spectroscopy. For the absorption measurements we used a PerkinElmer Lambda-35 spectrometer with a solution of 200 mM IL in DI water as blank. The absorbance values were normalized such that the 409 nm absorbance (corresponding to heme absorption) for the solution with no GuHCl (the most-folded sample) was unity to correct for slight differences in myoglobin concentration for different experiments. For the fluorescence measurements we used a PerkinElmer LS55 fluorometer. Samples were excited at 285 nm (corresponding to tryptophan excitation), and the fluorescence band from tryptophan was measured to infer tryptophan-to-heme FRET efficiency (related to the protein folding state). The fluorescence spectra were normalized such that the tryptophan fluorescence intensity at 355 nm for the 3.0 M GuHCl sample (the most-unfolded sample) was unity to correct for any constant myoglobin-IL quenching. Samples were incubated for at least 60 min prior to spectroscopic measurements.

3. RESULTS AND DISCUSSION 3.1. Absorbance Results. Myoglobin absorption measurements in the presence of EMIAc and BMIBF4 in solution (absence of GuHCl) are shown in Figure 1. This figure

2. EXPERIMENTAL SECTION Horse skeletal myoglobin was obtained from Sigma-Aldrich and used without further purification. EMIAc and BMIBF4 (SigmaAldrich, >98% purity) were also used without purification as the main impurity is water. LiBF4 and NaAc were obtained from Sigma-Aldrich. Solutions were prepared containing constant (0.2 mg/mL) myoglobin concentrations in 20 mM pH = 7 phosphate buffer. For each experiment, ∼12 solutions were prepared by pipetting from stock solutions with increasing GuHCl concentrations from 0 to 3.0 M, and for each experiment the solution contained a constant concentration of myoglobin and also of IL (either EMIAc or BMIBF4), prepared from a 500 mM stock solution of aqueous IL. The IL concentrations increased from 0 mM (first experiment) to 150 mM (final experiment). All samples were made immediately before measurements were taken. The BMIBF4 slowly hydrolyzes in water to form HF and BF3, and therefore myoglobin stored in IL solutions for more than a few days decomposes presumably due to the HF (data not shown). For this reason, experiments were performed with fresh BMIBF4

Figure 1. Absorption spectra of 0.2 mg/mL myoglobin in 20 mM pH = 7 phosphate buffer with increasing IL concentration. (A) Solutions with EMIAc from 40 to 400 mM show no significant differences except an increase at 275 nm, which arises from the IL. (B) Solutions with BMIBF4 from 10 to 400 mM show no significant differences except an increase at 275 nm, which arises from the IL. (C) With 1.0 M EMIAc, the myoglobin remains folded. (D) With 1.0 M BMIBF4 (top trace) the spectrum indicates that myoglobin is at least partially unfolded and also exhibits some interaction between the protein and the IL.

demonstrates that neither IL has any significant effect on the spectral properties of the folded protein over the range 50−400 mM. In fully folded myoglobin, the heme interacts with a histidine residue in a restricted pocket in the protein, leading to an intense Soret band. As the protein denatures, the histidine− heme interaction is lost and the absorbance of the Soret band decreases.40,41 In Figures 1A (EMIAc) and 1B (BMIBF4), the 407

dx.doi.org/10.1021/jp408061k | J. Phys. Chem. B 2014, 118, 406−412

The Journal of Physical Chemistry B

Article

spectral shape, intensity, and position (409 nm) of the heme Soret band are essentially unchanged, immediately indicating that the myoglobin remains in a predominantly folded conformation in solutions with up to 400 mM IL. To illustrate that myoglobin unfolding occurs with sufficiently high IL concentration, we extended the measurements in Figure 1B up to 1.0 M EMIAc and BMIBF4 solutions. Even up to 1.0 M, EMIAc does not affect the myoglobin spectrum (Figure 1C). Significant changes in the protein absorption spectrum are observed at the higher BMIBF4 concentration (Figure 1D). This change is evidenced by decreased absorbance at 409 nm and the appearance of a strong band at 510 nm that is only observed when BMIBF4 is present. For reference, spectra of folded and unfolded myoglobin (denatured with GuHCl) without IL can be found as Figure S1 in the Supporting Information. The 510 nm band likely arises from an interaction between the heme and the IL, although the nature of this interaction is not resolved. Figure 1D thus demonstrates that myoglobin unfolds in IL solutions above 1 M BMIBF4, and the unfolded protein structure likely interacts strongly with the IL. Figure 1 shows that moderate concentrations of ILs in solution do not induce myoglobin unfolding. To quantify the IL effects on the protein, we performed a series of GuHClinduced denaturing experiments in the presence of ILs. The absorbance intensity at 409 nm is proportional to the concentration of folded myoglobin molecules, and changes in 409 nm absorbance correlate to the myoglobin unfolding. GuHCl denatures myoglobin, and when the folded and unfolded structures are in equilibrium in the “transition region” of GuHCl concentrations, the unfolding equilibrium constant can be calculated from the normalized 409 nm absorbance intensity, A:41 K f→u =

Af − A A − Au

Figure 2. Myoglobin (0.2 mg/mL in pH = 7 phosphate buffer) unfolding in the presence of GuHCl and different EMIAc concentrations. (A) Absorbance at 409 nm as a function of increasing GuHCl molar concentration for three different EMIAc concentrations, showing the IL has a minimal effect on myoglobin unfolding. (B) Analysis of results showing the calculated values of ΔGunfolding (at 22 °C) (y-axis intercept for each fitted line) for each experiment.

solution without IL. With 150 mM IL, the sigmoidal curve is very slightly shifted to lower GuHCl concentration with a midpoint at ∼1.35 M GuHCl rather than 1.4 M. Based on Figure 2A, it appears that the EMIAc IL has a minimal effect on myoglobin unfolding. We used eqs 1 and 2 along with the data from Figure 2A to calculate ΔGf→u as a function of GuHCl concentration for the three experiments. The ΔGf→u values, calculated at the experimental laboratory temperature of 22 °C, are shown in Figure 2B. In the transition region (around the sigmoidal midpoint) the folded and unfolded myoglobin states are in equilibrium, and a plot of ΔGf→u versus [GuHCl] should be linear.41 These linear relationships can be extrapolated to 0 M GuHCl to derive the inherent thermodynamic stability of the myoglobin protein structure, i.e., ΔGunfolding in absence of denaturant. The ΔGunfolding values derived from linear fits of the data are included in Figure 2A. The R2 values for these linear fits are all between 0.87 and 0.95. The error bars are shown for a typical data set in Figure S2 of the Supporting Information. The value derived from the experiment in absence of EMIAc, 44 kJ/mol, is in very good agreement with the literature value of 45.7 kJ/mol for horse skeletal myoglobin under similar conditions.41 The ΔGf→u values for myoglobin in the presence of EMIAc appear to be very similar to the values in the absence of IL. They lie very close to each other in Figure 2B. The slopes of the linear fits of the IL data are slightly less than the linear slope in absence of IL, but the differences are not dramatic. Due to the differences in slope, the data in IL solutions extrapolate to lower values of ΔGunfolding. The ΔGunfolding values decrease slightly with increasing EMIAc concentration, from 44 kJ/mol for no IL to 29 kJ/mol with 100 mM IL and 24 kJ/mol with 150 mM IL. From R2 values in the linear fits (not shown) and fluctuations in absorbance values we can estimate the error in these ΔGunfolding values at around 20%; the difference between the two different IL concentrations is not significant. Thus, a visual inspection of Figure 2 combined with the values derived from the analysis appears to show that the EMIAc in solution may slightly decrease the stability of myoglobin, but the effect is minor. Moreover, the EMIAc effect on the myoglobin folded structure does not impart any significant enhancement of chaotrope-induced unfolding, indicating this effect is likely superficial. We next present the results for myoglobin unfolding in the presence of BMIBF4 in aqueous solution. The results, analogous to those presented in Figure 2, are summarized in

(1)

where Af is the absorbance of the fully folded (0 M GuHCl) solution and Au is the absorbance of the fully unfolded (3.0 M GuHCl) solution. From this equilibrium constant, ΔGf→u for the f → u phase transition can be readily calculated: ΔGf → u = −RT ln K f → u

(2)

By measuring ΔGf→u as a function of GuHCl concentration and extrapolating the data to zero GuHCl concentration, the free energy of unfolding and thermodynamic stability of myoglobin, ΔGunfolding, in the presence and absence of GuHCl has been measured. We performed this GuHCl-induced unfolding experiment for myoglobin in the presence of increasing IL solution concentrations. We first consider the results for myoglobin unfolding in the presence of EMIAc. The absorbance results are summarized in Figure 2. Values are normalized such that the absorbance is unity in absence of GuHCl. In the absence of IL, myoglobin begins to unfold with GuHCl concentrations above 1.0 M as signaled by a decrease in 409 nm absorbance. The unfolding follows a sigmoidal curve with midpoint at ∼1.4 M GuHCl and is completely unfolded with more than 2.0 M GuHCl. With increasing concentrations of EMIAc the curves in Figure 2A do not appear to change dramatically. With 100 mM IL, the absorbance values decrease slightly up to 1.0 M GuHCl, but significant decrease (signaling significant protein unfolding) does not occur until 1.4 M GuHCl, which is the same as the 408

dx.doi.org/10.1021/jp408061k | J. Phys. Chem. B 2014, 118, 406−412

The Journal of Physical Chemistry B

Article

group which allow for highly efficient FRET between the two groups (the R0 for the tryptophan−heme donor−acceptor pair is 29 Å; the absorbance and emission spectra are shown in Figure S4).42 The tryptophan−heme distance increases when myoglobin unfolds, resulting in decreased FRET efficiency. This can be monitored by measuring the tryptophan donor fluorescence at 355 nm, which is inversely proportional to the FRET efficiency. We performed FRET experiments (as described in section 2) on the same samples used for the absorbance measurements of GuHCl-induced myoglobin unfolding in IL environments. The FRET experiment results are shown for both IL results in Figure 4. Myoglobin was excited at 285 nm, corresponding

Figure 3. Figure 3A shows the absorbance at 409 nm as a function of increasing GuHCl molar concentration for three

Figure 3. Myoglobin (0.2 mg/mL in pH = 7 phosphate buffer) unfolding in the presence of GuHCl and different BMIBF 4 concentrations. (A) Absorbance at 409 nm as a function of increasing GuHCl molar concentration for three different BMIBF4 concentrations. (B) Analysis of results showing the calculated values of ΔGunfolding (at 22 °C) for each experiment.

different BMIBF4 concentrations. Values are normalized such that the absorbance is unity in absence of GuHCl. These results are clearly different from the results for EMIAc solutions (Figure 2). In the absence of IL (black squares) the myoglobin unfolds with GuHCl concentration above ∼1.0 M with a midpoint at ∼1.4 M (same data as Figure 2). However, with 100 mM BMIBF4 present (red circles) the myoglobin begins to unfold at lower GuHCl concentration, with GuHCl present above ∼0.75 M. With 150 mM BMIBF4 (blue triangles) significant protein unfolding begins to occur with only 0.25 M GuHCl. Thus, even though moderate IL concentrations do not cause myoglobin unfolding (Figure 1), they do lower the GuHCl-induced unfolding threshold. Control GuHCl-denaturation experiments performed in the presence of 200 mM NaCl, shown in Figure S3 of the Supporting Information, look very similar to the results in absence of IL (that is, the black squares in Figure 3) and indicate that the enhanced susceptibility to GuHCl denaturing and decrease in ΔGunfolding are not simply due to the ionic strength but rather to the special properties of the BMIBF4 IL electrolytes in solution. The data analysis for quantifying ΔGf→u and ΔGunfolding from the unfolding experiments in solutions of increasing BMIBF4 concentration is summarized in Figure 3B. As in Figure 2B, the R2 values for these linear fits are all between 0.87 and 0.95.The data in Figure 3B are clearly different from the data in Figure 2B and show that the BMIBF4 has a much more dramatic effect on myoglobin unfolding than the EMIAc. The ΔGunfolding values clearly decrease with increasing BMIBF4 concentration, from 44 kJ/mol for no IL to 20 kJ/mol with 100 mM IL and 8 kJ/mol with 150 mM IL. These numbers, which show a more than 75% decrease in ΔGunfolding for myoglobin in 150 mM BMIBF4, are more significant than the ΔGunfolding values derived for the EMIAc results in Figure 2B. It is therefore clear from Figure 3 that the thermodynamic stability of the myoglobin protein tertiary structure is decreased in the presence of the IL in solution and that increased IL concentration further decreases the myoglobin stability as reflected in the threshold concentrations of denaturant required to induce unfolding and in ΔGunfolding. 3.2. Fluorescence Results. We further investigated GuHCl-induced myoglobin unfolding in the presence of BMIBF4 IL in solution using Förster resonance energy transfer (FRET) experiments. In folded myoglobin, there are two tryptophan residues approximately 15 and 22 Å from the heme

Figure 4. Results of myoglobin FRET experiment on the same samples as those presented in Figures 2 and 3. Fluorescence intensities at 355 nm (normalized so that intensity is unity at [GuHCl] = 3.0 M) from tryptophan excitation at 285 nm are shown as a function of increasing GuHCl molar concentration for different IL concentrations. Increasing fluorescence intensity corresponds to decreased FRET efficiency due to myoglobin unfolding. (A) Myoglobin in solutions of increasing EMIAc concentration. (B) Myoglobin in solutions of increasing BMIBF4 concentration.

to tryptophan excitation. The full fluorescence spectra for 150 mM IL solutions are shown in Figure S5 of the Supporting Information. Unfolding results in an increase in tryptophan fluorescence intensity at ∼340−355 nm and decrease in heme fluorescence at 450 nm.40 The 355 nm fluorescence intensity in Figure 4 is shown as a function of increasing GuHCl concentration for the three different BMIBF4 concentrations. Figure 4A summarizes the results for the EMIAc solutions, and Figure 4B summarizes the results for the BMIBF4 solutions. The intensities are normalized so that the intensity is unity at 3.0 M GuHCl (completely unfolded myoglobin) to correct for any constant myoglobin−IL fluorescence quenching. In comparison to the samples lacking IL, those with 100 or 150 mM IL exhibit similar FRET efficiency in the absence of GuHCl (data at the left side of each figure), supporting the hypothesis that these amounts of IL do not cause any significant structural denaturation on their own. As the increase in tryptophan fluorescence signals protein unfolding due to decreased Trp-heme FRET, the interpretation of Figure 4 is analogous and consistent with Figures 2 and 3. The plots of normalized myoglobin fluorescence intensity vs GuHCl concentration are unchanged by the presence of 100 or 150 mM EMIAc (Figure 4A). This indicates (again) that EMIAc does not significantly affect the myoglobin unfolding behavior. However, the addition of BMIBF4 clearly changes the fluorescence data as shown in Figure 4B. With no IL present (black squares) the protein unfolds with GuHCl concentration above 1.0 M with a midpoint at ∼1.4 M, while in the presence of the BMIBF4 IL the myoglobin begins to unfold at lower GuHCl concentration. With 100 mM BMIBF4 the protein 409

dx.doi.org/10.1021/jp408061k | J. Phys. Chem. B 2014, 118, 406−412

The Journal of Physical Chemistry B

Article

begins to unfold above 0.75 M GuHCl with a midpoint of ∼1.2 M, and with 150 mM BMIBF4 the protein begins to unfold above 0.25 M GuHCl with a midpoint of ∼0.4 M. From Figure 4, denaturation experiments in the absence and presence of 100 or 150 mM IL aqueous solutions show similar patterns to those obtained using absorbance spectroscopy. Further inspection of the denaturation profiles obtained from our fluorescence and absorption experiments can be used to glean information related to the unfolding “pathway” as well as the potential presence of partially folded intermediate states. By comparing protein unfolding curves obtained by two different techniques, deviations between curve shape can often be attributed to structural intermediates in the unfolding pathway that respond differently to the different interrogation techniques (i.e., heme absorption vs Trp-heme FRET).43,44 We compare the unfolding curves from absorbance and fluorescence measurements in Figure 5. From the absorbance values from Figures 2A and 3A, we can compute the fraction of unfolded myoglobin, f u, from 0 (all folded) to 1 (all unfolded):40

fu =

Af − A Af − Au

environment and conformational state.47−50 CD spectroscopic results for myoglobin with BMIBF4 in the heme spectral region, around 409 nm, are included as Figure S6 in the Supporting Information as final evidence. The loss of 409 nm signal is consistent with the loss of protein structure and resultant change in environment around the heme. Figure S5 suggests that the structural unfolding pathway does not appear to be significantly altered by the BMIBF4 in solution. 3.3. Discussion. Taken together, the results presented in Figures 2−4 conclusively show that the presence of ILs as cosolvent additives to the aqueous environment of the myoglobin can have different effects on protein stability depending on the IL identity. More importantly, the free energy thermodynamic effects are quantified here using a GuHCl-induced protein unfolding experimental method. We report the ΔGunfolding values for myoglobin in solutions of increasing IL concentration. The experimental measurement of the thermodynamic stability provides quantitative insight into the effect of the IL on the protein stability. This thermodynamics result will be of interest to development and expanded use of ILs in protein science and biomedical applications. The results clearly highlight a significant difference between the two ILs in their effects on protein structure and stability as cosolvents in aqueous solution. At moderate concentrations, the EMIAc does not have any noticeable effect on the myoglobin unfolding and behaves essentially the same as NaCl (see Figure S2). In the pH-buffered solutions, the acetate remains deprotonated and likely interacts strongly with water molecules and remains solvated. Previous reports have shown protein unfolding in imidazolium acetate ILs but only after at 1.0 M concentration and addition of heat.12 Our results provide no evidence that acetate interacts with myoglobin to any extent (at the temperature and concentrations studied). The imidazolium cation likewise does not affect the myoglobin. In significant contrast, the BMIBF4 IL affects myoglobin unfolding by GuHCl. As stated in the Introduction, the BF4− anion would be predicted to destabilize proteins via repulsive interactions based on enzyme activity studies.10,51 Our results confirm this expectation and, furthermore, provide quantitative evidence of the effect of the IL on the protein thermodynamic stability. Moderate BMIBF4 concentrations do not result in myoglobin denaturing but instead lower the threshold concentration of GuHCl for inducing unfolding. The effects of HF generation via hydrolysis are not likely responsible for this decreased protein stability because the solutions were prepared fresh, although they would eventually cause the protein to decompose in BMIBF4 solution. Based on the GuHCl unfolding experiment, the presence of the IL lowers the thermodynamic stability of the myoglobin tertiary structure, by decreasing the unfolding free energy from 44 kJ/mol to less than 10 kJ/mol (a 75% decrease). Except for the thermodynamics, the results appear to suggest that the folded-tounfolded phase transition is largely unaffected by the IL (cf. Figure 4B and Figure S5). The main difference between the two ILs, BMIBF4 and EMIAc, is the anion; the imidazolium cation is essentially the same. It is therefore of interest to study myoglobin in solutions containing these anions with different cations. We performed the GuHCl-unfolding experiment in 150 mM LiBF4 solution to compare to the BMIBF4 solution results and in 150 mM NaAc solution to compare to the EMIAc solution results. The results are summarized in Figure 6. It is immediately clear that the

(3)

Figure 5. Plots of fluorescence data (fluorescence intensity at 355 nm, black squares) along with unfolded fractions computed from 409 nm absorbance data (red circles) for myoglobin in the presence of ILs in solution. (A) Data from the 150 mM EMIAc solutions. (B) Data from the 150 mM BMIBF4 solutions.

In Figure 5, the fluorescence intensities are plotted versus GuHCl concentration along with the unfolded fractions computed from the absorbance measurements for myoglobin in the presence of 150 mM IL in Figures 2A (EMIAc) and 3A (BMIBF4). The FRET data and unfolded fractions are consistent, and both show closely related sigmoidal curves indicating a single folding−unfolding transition without any significant intermediates.45 This figure provides further evidence that the GuHCl denaturation of myoglobin is unaffected by EMIAc but is facilitated by BMIBF4 due do decreased thermodynamic stability. It is notable that the two unfolding curves for normalized fluorescence and computed unfolding fraction are identical in shape when myoglobin is present in EMIAc IL solution (Figure 5A), but the two curves differ in the BMIBF4-supplemented samples (Figure 5B). This provides further evidence that the EMIAc does not affect the myoglobin unfolding in any significant way while the BMIBF4 does affect unfolding. Thus, Figure 5B provides some evidence for a specific myoglobin−BMIBF4 interaction that alters the unfolding process, although this deviation may also arise from more specific interactions between the IL and the fluorophores as seen in other protein−IL systems.46 Porphyrin groups are known to exhibit spectral changes in circular dichroism (CD) in response to changes in the local 410

dx.doi.org/10.1021/jp408061k | J. Phys. Chem. B 2014, 118, 406−412

The Journal of Physical Chemistry B

Article

denced in a dramatically decreased ΔGunfolding and threshold concentration required for GuHCl to induce denaturation. The similarity of the IL imidazolium cationic species indicates that the observed effects are resultant from the anionic species, Ac− vs BF4−. The wide variety of cation−anion pairs available as ionic liquids will facilitate further and continued investigation into specific IL properties and the resultant impacts on protein structure and stability and can lead to tuning of IL composition to impart desired effects to protein solutions, and the thermodynamic measurements presented here will benefit the design of future applications.

Figure 6. Myoglobin (0.2 mg/mL in pH = 7 phosphate buffer) unfolding in the presence of GuHCl and different electrolytes to compare to the IL solution systems. (A) Absorbance at 409 nm as a function of increasing GuHCl molar concentration for no additive, 150 mM EMIAc, and 150 mM NaAc. (B) Absorbance at 409 nm as a function of increasing GuHCl molar concentration for no additive, 150 mM BMIBF4, and 150 mM LiBF4.

■

ASSOCIATED CONTENT

S Supporting Information *

Myoglobin reference UV, fluorescence, and CD spectra, Trp/ heme excitation/emission spectra, representative data set with error bars, and unfolding with NaCl results. This material is available free of charge via the Internet at http://pubs.acs.org.

NaAc and LiBF4 results are different from each other. More importantly, the myoglobin unfolding in NaAc solution is exactly the same as the unfolding in EMIAc solution, and the unfolding in LiBF4 solution is exactly the same as the unfolding in BMIBF4 solution. This confirms that the destabilizing effects of the BMBF4 present as electrolytes in aqueous solution is due primarily to the anion (BF4−).20,51 Of additional interest is that the effect of the cation was insignificant which differs from the thermal-unfolding processes reported for Ribonuclase A and yeast alcohol dehydrogenase.20,51 This highlights the system-tosystem variability that can arise when changing the protein system being studied. We can speculate on the mechanism of protein destabilization as BF4− is a moderately polarizable anion capable of interacting with hydrophobic amino acid side chains. The anion may be able to “infiltrate” the inner hydrophobic core of myoglobin and partially disrupt the hydrogen bonds, providing tertiary structure stability. Some evidence of this is provided by the detection of a myoglobin−BMIBF4 interaction at high IL concentration in Figure 1. Furthermore, the supplementation of ILs in protein folding and denaturation experiments has a number of potentially significant benefits. From the perspective of ionic liquids, the variety of ILs available to allow for customizing and fine-tuning IL−protein interactions based on the cation−anion pair used in the IL. This includes the application of many well-studied ILs including cations and anions that have been shown to interact with proteins and modulate structure and activity.4−6,10,14,18−21 Additionally, the ability to enhance the efficacy of traditional chaotropes may allow access to denaturation experiments of previously difficult to unfold proteins, allow access to unique or previously inaccessible folding intermediates,21 or be applied in combination with other methods of denaturation such as thermal melts.18−21,27

■

AUTHOR INFORMATION

Corresponding Author

*E-mail [email protected], Ph +1-(856)-256-5457 (T.D.V.). Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS The authors thank Mr. Dylan Gary for performing the LiBF4 and NaAc control experiments and the College of Science and Mathematics at Rowan University and the American Chemical Society Project SEED program for financial support.

■

REFERENCES

(1) Hapiot, P.; Lagrost, C. Electrochemical Reactivity in RoomTemperature Ionic Liquids. Chem. Rev. 2008, 108, 2238−2264. (2) Zhang, Z.; Reddy, R. G. Thermal Stability of Ionic Liquids. In Epd Congress; Taylor, P. R., Ed.; TMS: Warrendale, PA, 2002. (3) Hallett, J. P.; Welton, T. Room-Temperature Ionic Liquids: Solvents for Synthesis and Catalysis. 2. Chem. Rev. 2011, 111, 3508− 3576. (4) Attri, P.; Venkatesu, P. Thermodynamic Characterization of the Biocompatible Ionic Liquid Effects on Protein Model Compounds and Their Functional Groups. Phys. Chem. Chem. Phys. 2011, 13, 6566− 6575. (5) Debeljuh, N.; Barrow, C. J.; Byrne, N. The Impact of Ionic Liquids on Amyloid Fibrillization of Abeta16-22: Tuning the Rate of Fibrillization Using a Reverse Hofmeister Strategy. Phys. Chem. Chem. Phys. 2011, 13, 16534−16536. (6) Kalhor, H. R.; Kamizi, M.; Akbari, J.; Heydari, A. Inhibition of Amyloid Formation by Ionic Liquids: Ionic Liquids Affecting Intermediate Oligomers. Biomacromolecules. 2009, 10, 2468−2475. (7) Gao, Y.-A.; Li, Z.-H.; Du, J.-M.; Han, B.-X.; Li, G.-Z.; Hou, W.-G.; Shen, D.; Zheng, L.-Q.; Zhang, G.-Y. Preparation and Characterization of Inclusion Complexes of B-Cyclodextrin with Ionic Liquid. Chem. Eur. J. 2006, 11, 5875−5880. (8) Pinkert, A.; Marsh, K. N.; Pang, S.; Staiger, M. P. Ionic Liquids and Their Interaction with Cellulose. Chem. Rev. 2009, 109, 6712− 6728. (9) Burney, P. R.; Pfaendtner, J. Structural and Dynamic Features of Candida Rugosa Lipase 1 in Water, Octane, Toluene, and Ionic Liquids BMIM-PF6 and BMIM-No3. J. Phys. Chem. B 2013, 117, 2662−2670. (10) Naushad, M.; Alothman, Z. A.; Khan, A. B.; Ali, M. Effect of Ionic Liquid on Activity, Stability, and Structure of Enzymes: A Review. Int. J. Biol. Macromol. 2012, 51, 555−560.

4. CONCLUSIONS The thermodynamic stability of proteins is key to their function in a cellular environment as well as in their applications in medical, pharmaceutical, and industrial settings. The results presented show that the supplementation of ionic liquids into protein solutions can affect the thermodynamic stability, and the effects are dependent on the molecular composition of the ionic liquid. Specifically, at moderate concentrations (50−150 mM) both the EMIAc and the BMIBF4 ionic liquids did not induce any significant myoglobin unfolding individually, but only the BMIBF4 significantly destabilized myoglobin evi411

dx.doi.org/10.1021/jp408061k | J. Phys. Chem. B 2014, 118, 406−412

The Journal of Physical Chemistry B

Article

(32) Ballew, R. M.; Sabelko, J.; Gruebele, M. Direct Observation of Fast Protein Folding: The Initial Collapse of Apomyoglobin. Proc. Natl. Acad. Sci. U. S. A. 1996, 93, 5759−5764. (33) Puett, D. The Equilibrium Unfolding Parameters of Horse and Sperm Whale Myoglobin. Effects of Guanidine Hydrochloride, Urea, and Acid. J. Biol. Chem. 1973, 248, 4623−4634. (34) Schechter, A. N.; Epstein, C. J. Spectral Studies on the Denaturation of Myoglobin. J. Mol. Biol. 1968, 35, 567−589. (35) Sykes, P. A.; Shiue, H. C.; Walker, J. R.; Bateman, R. C. Determination of Myoglobin Stability by Visible Spectroscopy. J. Chem. Educ. 1999, 76, 1283−1284. (36) Sanchez, K. M.; Schlamadinger, D. E.; Gable, J. E.; Kim, J. E. Forster Resonance Energy Transfer and Conformational Stability of Proteins: An Advanced Biophysical Module for Physical Chemistry Students. J. Chem. Educ. 2008, 85, 1253−1256. (37) Araujo, J. M. M.; Ferreira, R.; Marrucho, I. M.; Rebelo, L. P. N. Solvation of Nucleobases in 1,3-Dialkylimidazolium Acetate Ionic Liquid: NMR Spectroscopy Insights into the Dissolution Mechanism. J. Phys. Chem. B 2011, 115, 10739−10749. (38) Yu, L.; Clifford, J.; Pham, T. T.; Almaraz, E.; Perry, F. I.; Caputo, G. A.; Vaden, T. D. Conductivity, Spectroscopic, and Computational Investigation of H3O+ Solvation in Ionic Liquid BMIBF4. J. Phys. Chem. B 2013, 117, 7057−7064. (39) Yu, G.; Wen, L.; Zhao, D.; Asumana, C.; Chen, X. QSPR Study on the Viscosity of Bis(Trifluoromethylsulfonyl)Imide-Based Ionic Liquids. J. Mol. Liq. 2013, 184, 51−59. (40) Jones, C. M. An Introduction to Research in Protein Folding for Undergraduates. J. Chem. Educ. 1997, 74, 1306−1310. (41) Sykes, P. A.; Shiue, H.-C.; Walker, J. R.; Bateman, R. C., Jr. Determination of Myoglobin Stability by Visible Spectroscopy. J. Chem. Educ. 1999, 76, 1283−1284. (42) Wu, P.; Brand, L. Resonance Energy Transfer: Methods and Applications. Anal. Biochem. 1994, 218, 1−13. (43) Carra, J. H.; Privalov, P. L. Thermodynamics of Denaturation of Staphylococcal Nuclease Mutants: An Intermediate State in Protein Folding. FASEB J. 1996, 10, 67−74. (44) Guyett, P. J.; Gloss, L. M. Intermediates in Protein Folding. Compr. Biophys. 2012, 3, 43−71. (45) Mohan, P. M. K.; Hosur, R. V. Structure−Function−Folding Relationships and Native Energy Landscape of Dynein Light Chain Protein: Nuclear Magnetic Resonance Insights. J. Biosci. 2009, 34, 465−479. (46) Geng, F.; Zheng, L.; Yu, L.; Li, G.; Tung, C. Interaction of Bovine Serum Albumin and Long-Chain Imidazolium Ionic Liquid Measured by Fluorescence Spectra and Surface Tension. Process. Biochem. 2010, 45, 306−311. (47) Kajiyoshi, M.; Anan, F. K. Conformation of Biological Macromolecules. Circular Dichroism and Magnetic Circular Dichroism Studies of Metmyoglobin and Its Derivatives. J. Biochem. 1975, 78, 1087−1095. (48) Andersson, L. A.; Sono, M.; Dawson, J. H. Circular Dichroism Studies of Low-Spin Ferric Cytochrome P-450cam Ligand Complexes. Biochim. Biophys. Acta 1983, 748, 341−352. (49) Sono, M.; Dawson, J. H. Extensive Studies of the Heme Coordination Structure of Indoleamine 2,3-Dioxygenase and of Tryptophan Binding with Magnetic and Natural Circular Dichroism and Electron Paramagnetic Resonance Spectroscopy. Biochim. Biophys. Acta 1984, 789, 170−187. (50) Kawamura-Konishi, Y.; Kihara, H.; Suzuki, H. Reconstitution of Myoglobin from Apoprotein and Heme, Monitored by Stopped-Flow Absorption, Fluorescence and Circular Dichroism. Eur. J. Biochem. 1988, 170, 589−595. (51) Weibels, S.; Syguda, A.; Herrmann, C.; Weingartner, H. Steering the Enzymatic Activity of Proteins by Ionic Liquids. A Case Study of the Enzyme Kinetics of Yeast Alcohol Dehydrogenase. Phys. Chem. Chem. Phys. 2012, 14, 4635−4639.

(11) Takekiyo, T.; Yamazaki, K.; Yamaguchi, E.; Abe, H.; Yoshimura, Y. High Ionic Liquid Concentration-Induced Structural Change of Protein in Aqueous Solution: A Case Study of Lysozyme. J. Phys. Chem. B 2012, 116, 11092−11097. (12) Heller, W. T. Comparison of the Thermal Denaturing of Human Serum Albumin in the Presence of Guanidine Hydrochloride and 1-Butyl-3-Methylimidazolium Ionic Liquids. J. Phys. Chem. B 2013, 117, 2378−2383. (13) Sankaranarayanan, K.; Sathyaraj, G.; Nair, B. U.; Dhathathreyan, A. Reversible and Irreversible Conformational Transitions in Myoglobin: Role of Hydrated Amino Acid Ionic Liquid. J. Phys. Chem. B 2012, 116, 4175−4180. (14) Jaeger, V. W.; Pfaendtner, J. Structure, Dynamics, and Activity of Xylanase Solvated in Binary Mixtures of Ionic Liquid and Water. ACS Chem. Biol. 2013, 8, 1179−1186. (15) Shu, Y.; Liu, M.; Chen, S.; Chen, X.; Wang, J. New Insight into Molecular Interactions of Imidazolium Ionic Liquids with Bovine Serum Albumin. J. Phys. Chem. B 2011, 115, 12306−12314. (16) Mrestani-Klaus, C.; Richardt, A.; Wespe, C.; Stark, A.; Humpfer, E.; Bordusa, F. Structural Studies on Ionic Liquid/Water/Peptide Systems by HR-MAS NMR Spectroscopy. ChemPhysChem 2012, 13, 1836−1844. (17) Yu, G.; W, L.; Zhao, D.; Asumana, C.; Chen, X. QSPR Study on the Viscosity of Bis(Trifluoromethylsulfonyl)Imide-Based Ionic Liquids. J. Mol. Liq. 2013, 184, 51−59. (18) Weingartner, H.; Cabrele, C.; Herrmann, C. How Ionic Liquids Can Help to Stabilize Native Proteins. Phys. Chem. Chem. Phys. 2012, 14, 415−426. (19) Rodrigues, J. V.; Prosinecki, V.; Marrucho, I.; Rebelo, L. P.; Gomes, C. M. Protein Stability in an Ionic Liquid Milieu: On the Use of Differential Scanning Fluorometry. Phys. Chem. Chem. Phys. 2011, 13, 13614−13616. (20) Constantinescu, D.; Weingartner, H.; Herrmann, C. Protein Denaturation by Ionic Liquids and the Hofmeister Series: A Case Study of Aqueous Solutions of Ribonuclease A. Angew. Chem., Int. Ed. 2007, 46, 8887−8889. (21) Byrne, N.; Angell, C. A. Protein Unfolding, and the “Tuning in” of Reversible Intermediate States, in Protic Ionic Liquid Media. J. Mol. Biol. 2008, 378, 707−714. (22) Dill, K. A.; MacCallum, J. L. The Protein-Folding Problem, 50 Years On. Science 2012, 338, 1042−1046. (23) Dill, K. A.; Ozkan, S. B.; Weikl, T. R.; Chodera, J. D.; Voelz, V. A. The Protein Folding Problem: When Will It Be Solved? Curr. Opin. Struct. Biol. 2007, 17, 342−346. (24) Englander, S. W.; Mayne, L. Protein Folding Studied Using Hydrogen-Exchange Labeling and Two-Dimensional NMR. Annu. Rev. Biophys. Biomol. Struct. 1992, 21, 243−265. (25) Privalov, P. L. Intermediate States in Protein Folding. J. Mol. Biol. 1996, 258, 707−725. (26) Hill, R. B.; Raleigh, D. P.; Lombardi, A.; DeGrado, W. F. De Novo Design of Helical Bundles as Models for Understanding Protein Folding and Function. Acc. Chem. Res. 2000, 33, 745−754. (27) Byrne, N.; Wang, L. M.; Belieres, J. P.; Angell, C. A. Reversible Folding-Unfolding, Aggregation Protection, and Multi-Year Stabilization, in High Concentration Protein Solutions, Using Ionic Liquids. Chem. Commun. (Cambridge, U. K.) 2007, 2714−2716. (28) Constantinescu, D.; Weingartnet, H.; Herrmann, C. Protein Denaturation by Ionic Liquids and the Hofmeister Series: A Case Study of Aqueous Solutions of Ribonuclease A. Angew. Chem., Int. Ed. 2007, 46, 8887−8889. (29) England, J. L.; Haran, G. Role of Solvation Effects in Protein Denaturation: From Thermodynamics to Single Molecules and Back. Annu. Rev. Phys. Chem. 2011, 62, 257−277. (30) Englander, S. W. Protein Folding Intermediates and Pathways Studied by Hydrogen Exchange. Annu. Rev. Biophys. Biomol. Struct. 2000, 29, 213−238. (31) Maglio, O.; Nastri, F.; Pavone, V.; Lombardi, A.; DeGrado, W. F. Preorganization of Molecular Binding Sites in Designed Diiron Proteins. Proc. Natl. Acad. Sci. U. S. A. 2003, 100, 3772−3777. 412

dx.doi.org/10.1021/jp408061k | J. Phys. Chem. B 2014, 118, 406−412