Unravelling the Intricacy of the Crowded Environment through

Apr 7, 2017 - Behavior of bovine serum albumin in the presence of locust bean gum. Hongmei Zhang , Ying Wang , Yanqing Wang , Shixin Sun , Jian Cao...
0 downloads 0 Views 2MB Size
Article pubs.acs.org/JPCB

Unravelling the Intricacy of the Crowded Environment through Tryptophan Quenching in Lysozyme Priyanka Singh and Pramit K. Chowdhury* Department of Chemistry, Indian Institute of Technology Delhi, Hauz Khas, New Delhi 110016, India S Supporting Information *

ABSTRACT: Global conformational modulation of proteins in the presence of crowding agents is well-known. In this work, using the intrinsic tryptophan (Trp) residues of lysozyme, we have studied the effect of the crowding agents on the local conformation of this biomolecule. In presence of the macromolecular crowder Dextran 6, considerable quenching of tryptophan fluorescence was observed, which we have attributed to the enhanced proximity of the surrounding charged residues arising from local perturbation of protein structure. Accessibility of the Trp residues in the presence of crowders as a function of thermal and chemical denaturation has also been monitored using the traditional quenchers, acrylamide and iodide. Quenching in the presence of the crowding agents had to be modeled predominantly using the sphere-of-action model, with the sphere-of-volume “V” being postulated to be a signature of the cage-like environments that can exist in the solutions of such polymeric macromolecular crowding agents. Moreover, percolation of the quencher molecules through the entangled crowder systems was observed to be dependent on the micropolarity of the specific crowding agent being studied, with the neutral quencher acrylamide exhibiting maximum quenching in the presence of Dextran 6, while iodide being charged, exhibits higher quenching efficiency when Dextran 70 was the crowder. Additionally, control studies with the free amino acid tryptophan suggest that the variation in quenching so observed is not only due to the changes in the conformation of lysozyme and hence accessibility of the Trp residues but is also dictated by the underlying details and complexity of the crowder solutions to an appreciable extent.



INTRODUCTION Macromolecular crowding in past few years has gained tremendous importance owing to its myriad versatility in biophysical research.1−9 From the initial inception of crowding effects being attributed to the excluded volume phenomenon,10−12 recent studies have provided evidence of considerable soft interactions influencing the outcome in a crowded milieu.13−15 Several studies have reported the effects of macromolecular crowding agents on the global conformations of proteins and peptides.13,14,16−18 However, residue level or site-specific studies in the presence of crowding agents are rare to come across. Moreover, fluorescence spectroscopy has become a crucial and indispensable tool in biochemical research by virtue of its robustness, high sensitivity, and noninvasiveness. In this regard various intrinsic and extrinsic probes have been used to monitor the effect of external perturbations on biomacromolecules. Among the intrinsic protein fluorophores, tryptophan has been the most extensively studied one, used not only as a marker of protein denaturation but also as a sitespecific reporter for protein conformational changes and dynamics.19−23 In this study we have used the tryptophan fluorescence and quenching of the same to understand how the accessibility of these residues is modulated in the presence of © 2017 American Chemical Society

various macromolecular crowders under denaturing conditions. The protein we have chosen is the multitryptophan protein, hens egg white lysozyme (HEWL). HEWL is a globular protein containing 129 amino acids. The alpha-domain is composed of three alpha-helices and one C-terminal 310 helix, and the betadomain is composed of one antiparallel three-stranded betasheet and one loop. Besides, lysozyme contains four disulfide bonds.24,25 Among these disulfide bonds, two (Cys6-Cys127, Cys30-Cys115) are located in the alpha-domain, one (Cys64Cys80) in the beta-domain, and the remaining one (Cys76Cys94) acts as a linkage between the two domains.26−29 There are also six Trp residues within the globular protein. Among these Trp62, Trp63, and Trp108 are the most solventaccessible ones.28,29 Furthermore, HEWL is structurally homologous to human lysozyme (60% sequence homology), which has been found to be responsible for hereditary nonneuropathic systemic amyloidosis.30,31 Given the aforesaid features, hens egg white lysozyme therefore serves as a good Received: February 2, 2017 Revised: March 18, 2017 Published: April 7, 2017 4687

DOI: 10.1021/acs.jpcb.7b01055 J. Phys. Chem. B 2017, 121, 4687−4699

Article

The Journal of Physical Chemistry B

balance. Solutions of different concentrations of acrylamide and KI (freshly prepared) varying from 0.05 to 0.35 M in concentration, were prepared separately, in 50 mM phosphate buffer (pH 7). To ensure constant ionic strength, the total KI plus KCl concentration was held constant at 400 mM such that any effects on protein structure due to ionic strength variation are minimized. Thermal Unfolding of Native Hens Egg Lysozyme as a Function of Denaturants at pH 5. The protein samples (6 μM) were prepared in 50 mM acetate buffer that contained different concentrations of urea and GdmHCl in absence and presence of acrylamide and KI at 100 g/L and 200 g/L of crowder concentrations. The concentrations of urea and GdmHCl were verified by measuring the refractive index of the urea and GdmHCl stock solutions using a refractometer (Kruss optronics, Germany). In our study we have considered three different urea concentrations 3.5, 6, and 8 M and for the stronger denaturant GdmHCl the concentrations were 3.5 and 7 M. Instrumentation. Absorption measurements were performed in a double-beam Shimadzu UV−Vis spectrophotometer (UV-2450, Japan) using 1 cm path length cuvettes. Absorbance of the protein solutions was measured in the range of 200−600 nm and the molar extinction coefficient at 280 nm (ε280) used for HEWL was 37950 M−1cm−1. Fluorescence measurements were carried out on a FLS920 spectrofluorometer (Edinburgh Instruments, UK). The fluorescence spectra of protein samples at varying concentrations of crowders were measured using fluorescence quartz cuvettes having 1 cm path length. Prior to each experiment, the concentration of every sample was measured using the UV spectrophotometer. The intrinsic tryptophan fluorescence was monitored by exciting the protein at 295 nm (to avoid the excitation of tyrosine) and the emission was collected between 300 to 550 nm. The slit widths employed for the measurements were 0.5 and 4 nm for excitation and emission, respectively. For the experiments where thermal scans were not acquired, all the fluorescence spectra were recorded at 25 °C with an integration time of 0.4 s, the temperature being maintained by a Peltierbased controller. The samples were allowed sufficient equilibration time before acquiring their respective spectra. Excited-state lifetime measurements were performed using a time-correlated single photon counting (TCSPC) spectrometer (Edinburgh FLS920). For our experiments a LED having its central wavelength at 280 nm (EPLED 280) was used as the source for exciting the tryptophan (Trp) residues of HEWL. Emission was subsequently collected at 340 nm for the HEWL in buffer with and without the crowding agents through a single monochromator with a 10 nm bandpass and over a total time range (TAC) of 20 ns. Emission decays were fitted with appropriate instrument response functions (IRF) collected using a scattering solution (Ludox). The FWHM (full width at half-maximum) of the IRFs collected was typically in the range of 500 ps. The average tryptophan lifetime was calculated using the formula < τ> = Σiαiτi (αi = amplitude and τi = decay time of component i). Quenching Data Analyses. The quenching data were analyzed according to the models below: Stern−Volmer Model.

model system to investigate the effects of macromolecular crowders on its local conformational attributes. The most commonly used collisional fluorescence quenchers include acrylamide and iodide.32,33 While charged quenchers like iodide report predominantly on the surface exposed tryptophan residues, the neutral quencher acrylamide can access the often hydrophobic protein interior thereby reporting on the extent of accessibility of the buried residues. First, using the macromolecular crowders themselves as the quenchers, we observed a dramatic effect of the synthetic crowding agent Dextran 6 on native lysozyme as monitored using tryptophan quenching. Since the crowding agents used have little probability of accessing the buried and less-exposed Trp residues, we have hypothesized that the only way these crowders can quench the tryptophan fluorescence is by altering the local arrangement of the amino acids without affecting the global structure to a significant extent. Comparison with the quenching induced by small molecule crowders, like betaine, glucose, and sucrose, reveal that the quenching is strictly due to the macromolecular nature of the high molecular weight crowding agents used here. Moving on to the small molecule quenchers, interesting differences were observed between acrylamide and iodide based on the specific crowding agent being used. In presence of acrylamide as the quencher, quenching was the most for Dextran 6 while Dextran 70 was the crowder inducing the largest extent of quenching when iodide was the quencher. This phenomenon has been attributed to preferential percolation of the quencher molecule through the polymeric crowder matrix depending on the micropolarity of the latter. In addition, appreciable deviation from the commonly encountered linear Stern−Volmer quenching plots to those being described by the sphere-of-action model, the latter invoking an effective volume “V” which was also observed to be crowder dependent, reveals the complexity of the crowded milieu and hints at the cage-like arrangement that might exist under such conditions. As a consequence, the quenching phenomena were observed to be more under static control and less dictated by pure diffusion. We think that the results so obtained will be applicable in general for a crowded scenario, with the nature of the crowder influencing the manner in which quenching occurs along with the associated conformational changes of the protein being studied.



MATERIALS AND METHODS Hens egg white lysozyme (HEWL) was purchased from SigmaAldrich Chemical Co. (USA) and was used as received without further purification. Urea, GdmHCl (guanidinium hydrochloride), Dextran 6, Dextran 40, Dextran 70, Ficoll 70, and PEG 8000 were also purchased from Sigma-Aldrich. Sodium phosphate dibasic (Na2HPO4) anhydrous and mono basic (NaH2PO4) dihydrate, sodium acetate, acetic acid, acrylamide, and KI were purchased from Merck Specialties Private Limited (Mumbai, India). Thermal Unfolding of Native Hens Egg Lysozyme at pH 7. Stock solution of lysozyme was prepared in 50 mM of phosphate buffer (pH 7). The concentration of the protein samples used in steady state fluorescence measurements was kept at 6 μM and for the lifetime studies, the protein concentration was kept at 10 μM. For the quenching experiments a series of solutions with varying crowder concentrations (25 g/L to 200 g/L) were prepared in 50 mM phosphate buffer (pH 7) by weighing out appropriate amounts of the crowding agents using a Precisa analytical

F0 = 1 + KSV[Q ] F 4688

(1) DOI: 10.1021/acs.jpcb.7b01055 J. Phys. Chem. B 2017, 121, 4687−4699

Article

The Journal of Physical Chemistry B where F0 and F are respectively the fluorescence intensities in absence and presence of quenchers (here the macromolecular crowders) and KSV is the quenching constant. Linear fits were modeled according to eq 1. Collisional quenching of fluorescence was described by the following Stern−Volmer equation: F0 = 1 + kqτ0[Q ] = 1 + KD[Q ] F

(2)

where F0 and F are the fluorescence intensities in absence and presence of quencher, kq, is the bimolecular quenching constant; τ0 is the lifetime of the fluorophore in the absence of quencher, and Q is the concentration of quencher. The Stern−Volmer quenching constant is given by KD = kqτ0. If the quenching is known to be dynamic, the Stern−Volmer constant is represented by KD. Otherwise this constant is described as KSV. Sphere of Action Model. Quenching plots showing positive (upward) curvature were analyzed based on eq 3 which is a modified form of the Stern−Volmer equation and is known as the sphere-of-action model F0 = [(1 + KSV[Q ])exp(V [Q ])] F

Figure 1. Emission spectra of HEWL in the presence of (A) Dextran 6 and (B) PEG8000, respectively (λexc = 295 nm, λem = 340 nm). (C) Stern−Volmer plots of HEWL in the presence of different crowding agents. (D) Stern−Volmer plots in the presence of all the crowding agents except Dextran 6. The dotted lines are the fits to the respective quenching data according to eq 4b given in the text. In panels (A) and (B), lysozyme in the figure legend refers to the protein in buffer only, that is, in absence of any crowding agents.

(3)

with V being the volume of the sphere surrounding the fluorophore wherein for quenching to occur no diffusion needs to take place. For quenching induced by the crowding agents, eqs 1 and 3 were modified as follows:

F0 ′ [w] = 1 + KSV F

(4a)

F0 ′ [Mc])[Q ]exp(V [Q ]] = [1 + (KSV F

(4b)

bringing about changes in the microenvironment around the Trp residues. Analyses of the quenching curves thus obtained based on the Stern−Volmer equations (eqs 4a and 4b) also confirmed our observations, with the highest KSV’ observed in the presence of Dextran 6 (13 × 10−3 g−1L) followed by that of the higher molecular weight crowders of Ficoll 70 (3.2 × 10−3 g−1L), Dextran 70 (2.6 × 10−3 g−1L), and Dextran 40 (2.2 × 10−3 g−1L). The lowest KSV’ of 1.9 × 10−3 g−1L was observed in the presence of PEG 8000 (SI Table S1). The dramatic difference in the extent of Trp quenching between that observed for Dextran 6 and the other crowders can be rationalized as follows: (i) Dextran 6 having the least average molecular weight possesses the highest number of molecules (per g/L), thereby giving rise to more quenching and hence resulting in a higher KSV’ value, and (ii) the intrinsic response of the protein lysozyme and the conformational modulation therein, in the presence of the crowders. While our earlier report also showed a similar effect, with Dextran 6 being able to influence the native protein more than the other larger crowders,34 however the huge difference so observed in this study suggests that the outcome of the effect of the crowding agents is also influenced considerably by the type of biomolecule under investigation. The serum albumin proteins, human serum albumin (HSA), and bovine serum albumin (BSA), used in our earlier study, were both multidomain in nature, predominantly helical in secondary structure content, and possessed one and two tryptophans, respectively.34 On the contrary, lysozyme is much smaller in size and with six Trp residues, with the native form adopting mainly a helical conformation (∼30% of α helix; and ∼6% of β sheet). Thus, the smaller size of lysozyme and its similarity of molecular weight with that of Dextran 6, leads to enhanced excluded volume effect as exuded by the crowding agent. The heterogeneity of the conformational landscape in the presence of the crowding agents is further borne out by the different quenching profiles, that are, the linear (Stern−Volmer model)

where KSV ′ = KSV/Mc, with Mc being the average molecular weight of the crowder (6 kDa for Dextran 6, 40 kDa for Dextran 40, 70 kDa for Ficoll 70, and 8 kDa for PEG 8000) and w, the crowder concentration in g/l and V, volume of sphere in g−1L. KSV ′ can thus be considered to be the molar mass normalized form of the actual Stern−Volmer constant.



RESULTS Steady State Fluorescence. Crowder-Induced Quenching. A recent study has shown that macromolecular crowding can bring about appreciable quenching of the intrinsic tryptophan residues of proteins.34 In this regard, the tryptophan (Trp) fluorescence of lysozyme also underwent quenching to varying degrees in the presence of different macromolecular crowding agents, viz. Dextran (6, 40, and 70), Ficoll 70, and PEG 8000 (Figure 1). Dextran 6 exhibited the maximum quenching while PEG 8000 brings about the least reduction in Trp fluorescence in the native state for HEWL. Among the other crowders, Ficoll 70 induces the most quenching, followed by that of Dextran 70 and Dextran 40 (Figure 1D). To ascertain the fact that our observations are a result of the macromolecular nature of the polymer molecules, quenching experiments were also performed with the monomeric entities, like glucose (the monomer of dextran-based crowding agents), sucrose (the monomer of Ficoll 70), and a well-known osmolyte, betaine. As seen from the plots (Figure 1D), the latter exhibited the least Trp quenching thereby showing that the macromolecular crowders are more effective with respect to 4689

DOI: 10.1021/acs.jpcb.7b01055 J. Phys. Chem. B 2017, 121, 4687−4699

Article

The Journal of Physical Chemistry B response and the upward curvature (sphere-of-action model) induced by the crowders (Figure 1). It is unique that Dextran 6 is the only one exhibiting the sphere-of-action quenching while the others induced a linear response. This is an apt representation of the manner in which Dextran 6 affects the protein the most. The sphere-of-action quenching model arises from a group of quenchers that remain very close (i.e., adjacent) to the fluorophores and hence gives rise to instantaneous quenching without any diffusion needing to take place. Dextran 6 has been shown to exert its action primarily through the “excluded volume effect”.34 Thus, based on its higher number density it is quite obvious that it provides increasingly less conformational space to lysozyme as its concentration is increased. Hence at the higher concentrations where upward curvature comes into picture, the dynamics of the protein has been curtailed to an extent that the surrounding amino acid residues that are responsible for quenching are in very close proximity to the Trp residues, bringing about enhanced quenching without the need of diffusion, that is, large scale protein conformational motion need not take place. To probe further into the crowder induced quenching of the Trp fluorescence based on our assumption of static quenching arising from complex formation, we used the following equation: ⎡F − F⎤ log⎢ 0 = log K + n log[Q ] ⎣ F ⎥⎦

Figure 2. Stern−Volmer plots of (HEWL) with varying concentrations of acrylamide in the presence of different crowding agents at 25 °C (top panels) and 85 °C (bottom panels) (λexc = 295 nm, λem = 340 nm). Lysozyme in the figure legend signifies the quenching of the protein in absence of the crowders. The dotted lines are the fits to the respective quenching data according to eqs 1 and 3 as given in the text.

(5)

where K (M−1) is the binding constant and n is the number of binding sites that was determined by the slope of the fit to the plots (SI Figure S1). As shown (SI Table S2), Dextran 6 exhibits the highest binding constant K (27.11 M−1) among all the crowding agents with the n ∼ 2. Though this is a binding analyses, however, the fractional n values obtained that are quite less than 1 for almost all the other crowders used, imply that there is no direct association with the Trp residues but rather that the interaction is indirect, that is, mediated by the surrounding amino acid residues in response to crowder perturbation. Quenching by Traditional Fluorescence Quenchers. In the section above, the crowder induced quenching was mainly brought about by the changes in conformation in the immediate vicinity of the Trp residues. Since the crowders are quite massive in size, it is thus needless to say that there is negligible probability of these exhibiting direct interactions with the buried Trp residues. In other words, for such crowderinduced quenching, one is not measuring the accessibility of the intrinsic fluorophore but rather how the surrounding protein matrix responds or adjusts to the crowded environment. However, the extent of accessibility is also quite relevant since crowders are known to modulate protein conformations in a significant manner thereby affecting the path of approach to the intrinsic fluorophores. Hence quenching of the Trp residues using external quenchers provides important residue/region specific information about the impact of these macromolecules (crowders) on the protein conformation(s). In this study we have used two of the most commonly used fluorescence quenchers, acrylamide and potassium iodide (KI) (Figures 2 and 3) to provide insights into how the accessibility gets modulated in the crowded milieu. While acrylamide being neutral is considered to access both surface and buried residues (fluorophores) of proteins, however the halide ion, iodide (I−), due to its charged nature, provides

Figure 3. Stern−Volmer plots of (HEWL) with varying concentrations of iodide in the presence of different crowding agents at 25 °C (top panels) and 85 °C (bottom panels) (λexc = 295 nm, λem = 340 nm). Lysozyme in the figure legend signifies the quenching of the protein in absence of the crowders. The dotted lines are the fits to the respective quenching data according to eqs 1 and 3 as given in the text.

information mainly about the surface accessible residues since the interior of proteins can be quite hydrophobic.35 Since lysozyme has a distribution of buried and exposed Trp residues, using the two quenchers in tandem we hope to provide enough information about the local conformational changes of the protein as a function of several denaturing conditions, both thermal and chemical, in the presence of the macromolecular crowders. Thermal Denaturation. Native lysozyme has its tryptophan emission maximum at 340 nm. As the protein denatures as a function of temperature, there is not only a decrease in 4690

DOI: 10.1021/acs.jpcb.7b01055 J. Phys. Chem. B 2017, 121, 4687−4699

Article

The Journal of Physical Chemistry B fluorescence intensity but also a shift in the emission maximum to 346 nm. The largest shift occurs in the temperature range of 75−85 °C. A similar red shift was also observed in the presence of crowding agents. The reported Tm of the protein in buffer is ∼73 °C that coincides well with the temperature range where the largest shift in the Trp emission maximum was observed.36,37 Acrylamide. As shown in Figure 2, at 25 °C the maximum quenching was observed in the presence of 100 g/L of Dextran 6 with a KSV of 4.0 M−1 (SI Table S3). This is more than the KSV (2.6 M−1) observed for lysozyme in buffer only. Comparing the higher molecular weight crowders, KSV was observed to be more in the presence of 100 g/L of Ficoll 70 (3.1 M−1) as compared to that of Dextran 70 (1.8 M−1). On increasing the temperature from 25 to 85 °C a similar trend was observed with the highest quenching again in the presence of 100 g/L of Dextran 6 (KSV = 9.8 M−1; SI Table S6). Similar temperature dependent experiments were also carried out for lysozyme in the presence of 200 g/L of the crowders. In buffer only, that is, in absence of the crowders, increase of temperature leads to enhanced quenching beyond 65 °C (SI Figure S2), this corresponding to the aforesaid range within which the thermal unfolding transition takes place. This feature was also evident from the jump in KSV values beyond 65 °C (Figure 4).

sees an increase in the value of the quenching constant. It is to be noted here that for the profiles showing upward curvatures, the plotted KSV values only reflect on the linear part of the Trp emission response to quenching by acrylamide. For both the temperatures and crowder concentrations, the quenching and hence the overall accessibility of the Trp residues is the highest in the presence of Dextran 6, as obtained from the final F0/F values. Based on the fact that Dextran 6 itself induced such a large extent of quenching (Figure 1), this implies that the protein responded to and adjusted itself in the presence of this crowder in a manner such that the Trps remained more accessible not only than the other crowding agents but also when compared to that of the native protein. Moreover at 85 °C, the increase in the extent of quenching for 200 g/L Dextran 70 as compared to that of 100 g/L was not quite expected. Considering the fact that at higher crowder concentrations the protein would become more compact because of the associated increase in excluded volume, one would have thought that the Trps would be less exposed. Increase in the concentration of crowders is also accompanied by an increase in viscosity and since collisions have their origins in diffusion, be it dynamic or static in nature, increase in the extent of quenching, as seen here, was not a logical expectation, provided we assume that the quenching reactions here are strictly under diffusion control. KI. For selective quenching of the surface and hence more exposed Trp residues, the aforementioned experiments were also performed with I− (iodide) as the quencher. In absence of the crowders, unlike that observed for acrylamide, no sudden increase in the quenching constant at higher temperatures occurred here. In presence of the crowding agents a surprising modification in the quenching trend was observed. Dextran 6 no longer induced the maximum exposure; instead at 200 g/L for both the temperatures (25 and 85 °C), the highest degree of quenching was observed for Dextran 70, while at 100 g/L, almost all crowders induced very similar quenching. On increasing the concentration of the crowder(s) to 200 g/L, sphere of action quenching was observed in the presence of the dextran-based crowders while a linear Stern−Volmer plot was observed in the presence of Ficoll 70. Moreover, the KSV values for KI quenching showed no characteristic temperature dependence (Figure 4). This latter aspect confirms the fact that the halide ion preferentially quenches the surface Trp residues since the latter are not expected to track the cooperative unfolding process of lysozyme. Finally, closer analyses of the quenching plots (SI Figures S3−7) reveal that the acrylamide quenching profiles are dominated by the linear type while in case of KI the profiles are mostly described by the sphere-of-action model. This observation is quite in contrast to that in absence of the crowders wherein both the quenchers gave rise to linear quenching plots, thereby implying that the crowding agents are responsible for inducing such differences. Chemical Denaturation. In presence of chemical denaturants, all our experiments were carried out at pH 5 so as to unfold the protein which is otherwise very stable at the physiological pH.38,39 In our case we have chosen three urea concentrations namely 3.5, 6, and 8 M and two GdmHCl concentrations, 3.5 and 7 M. Native lysozyme shows an emission maximum at 340 nm which shifts to 341 and 342 nm for the urea concentrations of 6 and 8 M, respectively. In presence of the stronger denaturant GdmHCl, at 3.5 M, the λmax shifts to 344 nm while at 7 M, the fluorescence maximum undergoes a 10 nm shift, that is, from 340 to 350 nm, showing

Figure 4. KSV as a function of temperature for (HEWL) in the presence of acrylamide (top panels) and iodide (bottom panels) for 100 g/L and 200 g/L of different crowding agents.

Moreover, the observed quenching profiles in absence of the crowding agents were linear thereby implying a single type of quenching phenomenon (either dynamic or static). Analyses of the types of quenching profiles observed in the presence of the individual crowding agents (SI Figures S3−7 and SI Tables S3−6) show that except PEG8000 (sphere-of-action), for all the other crowders at 100 g/L the trends were linear. In presence of 200 g/L, both Dextran 6 and Ficoll 70 induced structural changes such that a sphere-of-action model for quenching needed to be invoked. Interestingly, both Dextran 6 and Ficoll 70 at 100 g/L induced higher KSV values and steeper jumps during the unfolding of the protein, signifying enhanced accessibility of the Trp residues under these conditions. For 200 g/L crowders, the KSV values are generally lower than that of 100 g/L, except that of Dextran 70 and PEG8000, where one 4691

DOI: 10.1021/acs.jpcb.7b01055 J. Phys. Chem. B 2017, 121, 4687−4699

Article

The Journal of Physical Chemistry B

Table 1. KSV (M−1) and V (M−1) for HEWL in the Presence of Dextran-based Crowding Agents for Acrylamide and Iodide at a Fixed Denaturant Concentration acrylamide

iodide

100 g/L steady state fluorescence 3.5 M urea

6 M urea

8 M urea

3.5 M GdmHCl

7 M GdmHCl

buffer Dextran Dextran Dextran buffer Dextran Dextran Dextran buffer Dextran Dextran Dextran buffer Dextran Dextran Dextran buffer Dextran Dextran Dextran

KSV 6 40 70 6 40 70 6 40 70 6 40 70 6 40 70

KSV = 1.58 2.32 1.71 KSV = 1.25 1.85 0.79 KSV = 1.89 1.88 1.11 KSV = 1.70 0.35 2.53 KSV = 3.27 2.78 6.23

200 g/L V

KSV

1.05

1.68 2.21 2.87

100 g/L V

1.66

2.83

1.37

1.31

1.59 1.87 2.13

1.35

1.41

1.72 1.41, V = 1.73 1.61 5.14 2.78, V = 1.49 1.25 2.99

1.66 1.67 1.41

1.53

0.70 1.22 6.54

3.09 2.43

3.45 3.49 1.93

1.43 1.01 3.22

that the tryptophans are mostly in a polar environment. The extent of quenching for both acrylamide and KI was the highest in the presence of 7 M GdmHCl, further proving that the Trp residues are the most exposed under such strong denaturing conditions (SI Figure S8). Moreover, for urea denaturation, all the observed quenching profiles were linear in trend while for GdmHCl, upward curvatures were noted. Having investigated the quenching profile under the denaturing conditions in absence of the crowders, our next step was to probe the effect of these quenchers in the presence of the dextran-based crowding agents on the Trp residues. 1. Urea (Figures 5, S9, and S10). Acrylamide. Linear quenching profiles were observed for Dextran 6 and Dextran 40 while Dextran 70 induced trends with upward curvatures that were fitted using the sphere of action model, throughout the urea concentrations used in this study. As shown in Table 1, on increasing the urea concentration from 3.5 to 8 M no such significant change in the KSV values were observed in the presence of Dextran 6 and Dextran 40. Dextran 70 however showed marked changes in KSV and the sphere of volume. The highest KSV of 2.87 M−1 was obtained in the presence of 200 g/ L of Dextran 70 at 3.5 M urea and was also accompanied by the highest V of 2.83 M−1, which therefore in combination gave rise to a high F0/F value of ∼4.0 (Figure 5). KI. In presence of KI as the quencher, similar sphere-ofaction quenching profiles were observed for Dextran 70 throughout. The value of “V” decreased as a function of the crowder concentration (Dextran 70) for a fixed urea value (Table 1). For example, “V” decreases from 3.72 M−1 to 2.94 M−1 at 3.5 M urea whereas at 6 M urea it decreases from 3.59 M−1 to 1.95 M−1 and at 8 M urea volume decreases from 7.33 M−1 to 1.95 M−1 on increasing the crowder concentration from 100 g/L to 200 g/L. Since “V” is a reflection of the volume and hence length scale over which the quenchers (iodide ions) are distributed, the data thus suggest that increase in the concentration of Dextran 70 is accompanied by a decrease in

KSV KSV = 0.98 0.81 0.30 KSV = 0.86 0.59 0.40 KSV = 1.15 1.21 0.23 KSV = 0.28 0.45 1.20 KSV = 2.23 0.63 1.11

200 g/L V

KSV

V

3.72

1.27 1.06 0.61

2.94

3.59

0.84 1.03 0.80

1.95

7.33

1.24 0.91 0.75

1.95

2.00

1.20

1.68

1.68 4.52 4.31

1.29 1.28 1.43

1.79, V = 1.58 5.69 3.22

2.97 2.93 0.59

4.24

Figure 5. Stern−Volmer plots of (HEWL) with varying concentrations of acrylamide and iodide in the presence of dextran-based crowding agents at 25 °C at 3.5 M urea (λexc = 295 nm, λem = 340 nm). The dotted lines are the fits to the respective quenching data according to eqs 1 and 3 as given in the text.

the immediate I− concentration around the Trp residues, probably arising from the crowder molecules displacing the latter from the immediate protein neighborhood. 2. GdmHCl (Figures 6 and S11). Acrylamide. Sphere of action profiles were observed in the presence of Dextran 6 and Dextran 40 for all the GdmHCl concentrations used whereas Dextran 70 mostly induced linear quenching profiles except at 200 g/L in 7 M GdmHCl in the presence of acrylamide. Both KSV and volume of sphere “V” changed for a fixed GdmHCl concentration as a function of the crowder concentrations as shown in Table 1. In presence of Dextran 6 at 3.5 M GdmHCl 4692

DOI: 10.1021/acs.jpcb.7b01055 J. Phys. Chem. B 2017, 121, 4687−4699

Article

The Journal of Physical Chemistry B KSV decreases from 1.70 M−1 to 0.70 M−1 whereas an increase was observed for Dextran 40 from 0.35 M−1 to 1.22 M−1 and for Dextran 70 from 2.53 M−1 to 6.54 M−1 on increasing the crowder concentration from 100 g/L to 200 g/L (Figure 6).

constants of 2.37 ns and 0.72 ns, giving rise to an average lifetime of 1.16 ns, which is in good agreement with the previous reports.28,29 It has been established through a series of mutations that Trp-108 and Trp-62 of lysozyme are the major contributors to the fluorescence of the native lysozyme and hence the lifetime is also a reflection of the same.40 It has also been reported that the fluorescence decay of Trp residues are influenced by neighboring side chains, such as amino groups, carboxylic groups, and disulfide linkages, through proton transfer and electron exchange interactions.29 Hence small and subtle changes in conformation of protein can bring about local changes that will affect the Trp fluorescence. In presence of the crowders only, the Trp lifetime is barely affected under denaturing and nondenaturing conditions (data not shown), that is, the decays were almost superimposable, thereby implying that the influence of dynamic quenching in the presence of these crowding agents is negligible. This observation can be explained based on the fact that the macromolecular crowding agents are quite massive in size and hence their penetration into the protein interior for accessing the intrinsic Trp residues is hindered, thus further confirming our hypothesis based on our steady-state measurements in Figure 1. Having established that crowder induced quenching is predominantly static in nature, in agreement with a previous study,34 it can thus be visualized that the quenching of the intrinsic Trp residues of the protein is due to weak ground-state complex formation between the Trp moieties and the surrounding amino acids as alluded to before. However, a significant contribution of dynamic quenching was observed when acrylamide and KI were used as quenchers. The decay profiles were faster in the presence of acrylamide as compared to KI in absence and presence of crowding agents at a fixed denaturant condition, as expected, because acrylamide can bring about quenching of both exposed and buried Trp residues. In presence of the quenchers, the longer component of the lifetime was the most affected with almost no change in the other component (SI Tables S8−S10). This suggests that the longer component arises from the contribution of those residues that get preferably quenched in this multitryptophan protein. The highest quenching constant was observed at 7 M GdmHCl in the presence of both the quenchers with KD values of 3.72 M−1 and 2.39 M−1 for acrylamide and iodide respectively, KD being the dynamic Stern−Volmer quenching constant (SI Table S7). In presence of acrylamide we observed mostly linear Stern−Volmer plots for both 100 g/L and 200 g/ L of the dextran-based crowding agents, except for 7 M GdmHCl, wherein for Dextran 70 the nature of graph had to be described by the sphere-of-action model. In case of iodideinduced quenching a nonlinear trend was however observed for Dextran 40. The dynamic quenching trends as obtained through the lifetime studies show only slight differences for the crowding agents under most of the denaturing conditions (SI Figures S12−14), signifying that the collision-type interactions are quite similar overall. This further implies that the static component has a major impact in the differences observed in the quenching profiles from the steady-state data. Quenching Studies with the Amino Acid Tryptophan. In order to provide further insights into the observed quenching data, we have studied the effect of acrylamide and KI on the emission of the free amino acid tryptophan, in the presence of the crowding agents at selected denaturant conditions, preferably where lysozyme would be more unfolded resulting

Figure 6. Stern−Volmer plots of (HEWL) with varying concentrations of acrylamide and iodide in the presence of dextran-based crowding agents at 25 °C at 3.5 M GdmHCl (λexc = 295 nm, λem = 344 nm). The dotted lines are the fits to the respective quenching data according to eqs 1 and 3 as given in the text.

However, at 7 M GdmHCl an increase in trend for KSV was followed for Dextran 40 from 2.78 M−1 to 3.49 M−1 while Dextran 70 showed a decrease in KSV from 6.23 M−1 to 1.93 M−1 on increasing the crowder concentration from 100 g/L to 200 g/L. On increasing the concentration of dextran-based crowders from 100 g/L to 200 g/L sphere-of-action quenching was observed with an increase in volume of sphere “V” from 1.61 M−1 to 3.09 M−1 in the presence of Dextran 6 and a decrease in “V” from 5.14 M−1 to 2.43 M−1 in the presence of Dextran 40 at 3.5 M GdmHCl. Moreover this decrease in trend for sphere of volume was followed in the presence of Dextran 40 at 7 M GdmHCl from 2.99 M−1 to 1.01 M−1. KI. An increase in KSV with increase in crowder concentration was observed in the presence of KI, except for Dextran 70 at 7 M GdmHCl where the decrease in KSV from 1.11 M−1 to 0.59 M−1 was compensated by a shift in the value of “V” to the higher side. At 7 M GdmHCl, the highest “V” of 5.69 M−1 was observed for 100 g/L of Dextran 40. To end this section, the overall quenching constants are much greater in 7 M GdmHCl confirming the fact that it is indeed the tryptophan exposure that dictates, to a large extent, the degree of quenching under these conditions. Time Resolved Fluorescence. To probe the contribution of dynamic quenching in the observed Stern−Volmer plots, we also performed time-resolved fluorescence studies for HEWL in the presence of the dextran-based crowders at a fixed denaturant concentration. For this study we have chosen 8 M urea and the GdmHCl concentrations of 3.5 and 7 M as the denaturing conditions wherein the exposure of the Trp residues are appreciably on the higher side. The lifetime of lysozyme was adequately described by a double exponential fit, with time 4693

DOI: 10.1021/acs.jpcb.7b01055 J. Phys. Chem. B 2017, 121, 4687−4699

Article

The Journal of Physical Chemistry B

Table 2. KSV (M −1) and V (M −1) Values of the Free Amino Acid Tryptophan in the Presence of Dextran-based Crowding Agents for Acrylamide and Iodide as a Function of Denaturant Concentration acrylamide

iodide

100 g/L steady state fluorescence Trp (buffer)

8 M urea

3.5 M GdmHCl

7 M GdmHCl

buffer Dextran Dextran Dextran buffer Dextran Dextran Dextran buffer Dextran Dextran Dextran buffer Dextran Dextran Dextran

KSV 6 40 70 6 40 70 6 40 70 6 40 70

KSV = 54.5 10.6 14.9 KSV = 17.1 9.0 9.1 KSV = 19.5 16.2 12.8 KSV = 12.9 9.0 7.3

200 g/L V

KSV

100 g/L V

11.8, V = 2.2 1.7 1.3 12.1, V = 1.9 2.1 1.6 12.4, V = 1.2 3.9 1.3 10.5, V = 1.4 0.5 2.3

26.7 9.4 16.5

2.2 0.9

36.4 3.4 14.2

2.5 0.7

1.1 9.4 15.7

5.1 1.2 0.9

8.3 8.5 4.7

1.5 2.1

KSV

200 g/L V

KSV= 10.7 5.8 4.2 1.4 7.0 1.5 KSV = 7.7 4.3 3.6 7.0 KSV = 3.8, V = 1.8 3.1 1.9 0.9 5.4 4.7 KSV = 3.3, V = 1.1 1.2 3.6 2.7 2.4 5.3

KSV

V

2.9 2.1 4.9

2.7

2.0 2.5 3.4

1.7

3.2 0.7 3.8 1.2 1.8 2.3

6.0 1.8

3.4

fact that the increase in viscosity of the polymer solutions (with increase in polymer concentration) did not seem to have any effect on the KSV value also implies that the observed quenching is not strictly dynamic in the presence of the crowding agents. This proposition was further confirmed when a dramatic increase in the value of KSV in the presence of Dextran 6 was obtained, with the quenching constant being as high as 54.5 M−1. While the linearity in profile confirms a single type of quenching, the enhanced quenching we presume arises from a cage like arrangement of the Dextran 6 molecules around tryptophan. This has the effect of increasing the local concentration and hence the “activity” of acrylamide, thereby increasing either the frequency of collisions with the fluorophore (for the dynamic part) or giving rise to enhanced complex formation (if quenching is static in nature). The same is amplified for Dextran 6 because of its higher packing density (arising from its low molecular weight) that allows little room for even the small acrylamide molecules to venture far out from tryptophan. In this regard the decrease in KSV to almost half at 200 g/L of Dextran 6 is not a result of viscosity but rather points to the possibility of increased polymer entanglement of Dextran 6 at this concentration thereby restricting the access of acrylamide molecules through to the tryptophan residues. KI. In absence of any denaturants, iodide quenching of tryptophan in buffer is the most followed by that in the presence of Dextran 70. A similar situation prevails for almost all the denaturing conditions too (Table 2 and SI Figures S15− 17) with Dextran 70 in some cases inducing more quenching or comparable to that observed in absence of any crowders. Unlike that in case of acrylamide, on average the quenching by iodide is quite reduced in the presence of the crowders and also in the presence of urea and GdmHCl. The difference between acrylamide and iodide arises mostly from the charged nature of the latter quencher, since the size of both are quite similar (diameter of iodide: ∼ 410 pm; diameter of acryamide: ∼ 430 pm, a linear structure was assumed for acrylamide). In buffer only, both acrylamide and iodide have comparable KSV values (Table 2), thus further confirming the size similarity. However, for the neutral quencher (acrylamide), the profile had to be analyzed using the sphere-of-action model suggesting the

in the Trp residues on average being more exposed and hence experiencing a more polar microenvironment (Table 2). Acrylamide. In buffer only, the Stern−Volmer plot showed an upward curvature (Figure 7) with the obtained KSV and “V”

Figure 7. Stern−Volmer plots for the free amino acid tryptophan with varying concentrations of acrylamide and KI at 25 °C for dextranbased crowding agents of different molecular weights as mentioned in the figure legend. The dotted lines are the fits to the respective quenching data according to eqs 1 and 3 as given in the text (λexc = 295 nm, λem = 350 nm).

being very similar to that reported in a recent study.41 This suggests that in addition to the assumed dynamic collisionbased component, a part of the quenching arises from quenchers proximal to the fluorophore, an aspect that we did not expect to occur in the dilute buffer conditions. The nature of quenching was the same in the presence of 100 g/L and 200 g/L Dextran 40 and Dextran 70 as the crowding agents, with Dextran 70 showing slightly higher KSV values. Moreover the 4694

DOI: 10.1021/acs.jpcb.7b01055 J. Phys. Chem. B 2017, 121, 4687−4699

Article

The Journal of Physical Chemistry B

Cys94 and Cys6-Cys127.38,40,44,45 Thus, taking into account the fact that Trp-108 and Trp-62 are the major contributors to the observed fluorescence in the native state, thus logically it can be expected that extensive pressure from the surrounding Dextran 6 molecules leads to increased compression on the protein around these two tryptophan residues thereby leading to the observed quenching. The presence of several charged residues (the negatively charged carboxylic group of Glu35 is close to Trp 108 and the guanidylic group of Arg 61 is close to Trp 62) in close proximity to the aforementioned Trp moieties (Figure 8) are suitably poised to bring about the aforementioned

presence of higher local concentration of the quencher around tryptophan giving rise to the static contribution. In presence of Dextran 6, the dramatic drop in the extent of quenching in buffer as signified by ∼10-fold decrease in the quenching constant for KI as compared to that of acrylamide is quite intriguing. This difference cannot be attributed to the lack of interstitial voids for Dextran 6 (presumed to exhibit high packing density), since as discussed above, the quenchers are of similar size. Thus, the data suggest that movement of the charged species is specifically being hindered in the presence of not only Dextran 6 but also the other dextran based crowders (Table 2). Quenching studies involving charge separated states have been shown to be polarity dependent with nonpolar solvents/environments adversely affecting the rate of quenching.42,43 Our data thus point to the fact that the intricate polymer network of the crowding agents presents a predominantly hydrophobic environment which the negatively charged iodide ion finds difficult to percolate. Evidence of the same is also obtained from pyrene based micropolarity studies of these crowded systems (unpublished results) wherein Dextran 70 exhibited a higher micropolarity than that of Dextran 6. In support of this argument, Dextran 70 induced the highest quenching among the crowders for KI.



DISCUSSION Quenching analyses of intrinsic tryptophan (Trp) residues in proteins have been extensively used to study the nature of folding and unfolding reactions that a protein undergoes when subjected to suitable perturbing conditions. While most of such reports have focused on the general accessibility of the Trp residues to the various quenchers, however a few studies have shown that the quenching mechanism is not as simple as one might assume. For example acrylamide quenching has been shown to be not only viscosity independent in many cases but can also provide deeper insights into the conformational dispositions of proteins, an aspect that has received much lesser attention than it deserves. Moreover, with the realization that the crowded cellular interior has significant impact on several biological phenomena, and in particular on conformations of proteins, thus analyses of tryptophan quenching should provide much needed insights into how the local protein structure, that is, in the immediate neighborhood of the Trp residues, is modulated in the crowded milieu. Crowder induced quenching of the Trp fluorescence of lysozyme was an ample reflection of the manner in which these macromolecules/polymers can influence the local environment without causing much change in the global structure as observed from the CD studies (SI Table S11). The dramatically enhanced quenching induced by Dextran 6 which we have proposed to be arising from the higher number density of the crowder molecules also suggests that under these conditions the protein conformational subspace is severely restricted such that the surrounding amino acids responsible for quenching are in very close proximity leading to increased complexation with the Trp moieties. Based on the crystal structure (PDB: 2LYZ) of HEWL, three Trp residues, namely; Trp62, Trp63, and Trp108 are located in the substrate-binding site, two (Trp28 and Trp111) are located in the core hydrophobic region and the remaining Trp residue, Trp123 is located apart from all other residues. Trp62 and Trp108 are the ones contributing to the Trp fluorescence the most since it has been reported that the fluorescence of other Trp residues (Trp63, Trp111, and Trp123) is quenched by the nearby disulfide linkages of Cys76-

Figure 8. 3D representation of hen egg white lysozyme (HEWL) (2LYZ, Protein Data Bank) generated using PyMOL software. (A) Trp62, Trp63, Trp108, Trp 111, Trp 28, and Trp 123 are the six tryptophan residues that have been highlighted in (red). (B) Along with the Trp residues, the residues that might be responsible for quenching the Trp fluorescence because of their proximity has also shown as follows: Residues highlighted as in close proximity to Trp 62 are Arg 61 and Arg 73 (cyan), Asp 101 (yellow) close to Trp 63, Glu 35 (orange) and Met 105 (marine blue) close to Trp 108, and Lys 116 (magenta) close to Trp 111.

quenching. Moreover, recent papers have suggested that even the monomeric building blocks of the macromolecular crowders and some polysaccharides as small molecule crowders can have appreciable effects on the protein conformation.13 However, our results with glucose, sucrose, and betaine confirm that the effects are predominantly due to the macromolecular nature of the crowding agents, with sucrose being a possible exception, in agreement with a recent study.46 Furthermore, though Dextran 6 and PEG8000 have similar molecular weights and hence similar number density, however the stark difference in their effects on the Trp fluorescence and hence the protein structure is a direct consequence of the dissimilar internal architecture of these crowders, again in consonance with previous reports.47,48 Since there are six Trp residues in lysozyme, the idea behind the usage of multiple quenchers, one charged (KI) and the other neutral (acrylamide), was to be able to selectively investigate the buried and the exposed tryptophans, thereby gaining further information about the local conformation modulation of the protein in the presence of the crowding agents. To aid this analysis, we also carried out the quenching studies for both thermally and chemically denatured lysozyme, keeping in mind the fact that differences might exist in the nature of denatured states so encountered. Additionally, to reduce the complexity of the study, for chemical denaturation, we compared the outcome for only the dextran-based crowding 4695

DOI: 10.1021/acs.jpcb.7b01055 J. Phys. Chem. B 2017, 121, 4687−4699

Article

The Journal of Physical Chemistry B

presence of cage-like structures arising from polymer chain entanglement of these crowding agents, which we propose is also partially responsible for the high KSV values obtained in quite a few cases, as summarized in Table 2. “V”, the active volume, takes into account contributions from static quenching and transient effects arising from quenchers in the proximity of the chromophore at the time of excitation.41 One of the main differences among the quenchers is the magnitude of the static component, V, which on average is much more in case of KI as compared to acrylamide (Table 1). For example the largest “V” value of 7.33 M−1 at 8 M urea was observed in the presence of KI for 100 g/L of Dextran 70. In presence of GdmHCl, the “V” values obtained in the presence of KI were consistently on the higher side for both the protein and the free amino acid. Since iodide is a charged quencher, its presence in the vicinity of the protein will be dictated by the positively charged side chains and/or patches on the protein surface. This factor is augmented in the presence of the guanidinium cation that favors closer association with the oppositely charged iodide ion, thereby giving rise to a higher “V”. Based on eq 5, we have also carried out binding analyses of all the quenching curves for HEWL showing upward curvatures in the presence of the chemical denaturants. All these plots are linear in profile therefore further confirming that the nature of quenching is predominantly due to complex formation (SI Figures S18−23). Moreover, higher binding constants are invariably accompanied by higher average number of binding sites, in particular, for KI in the presence of Dextran 70 in GdmHCl as the denaturant (SI Figures S22 and S23). Closer analyses of the quenching curves in the presence of chemical denaturants also bring forth an additional interesting feature of the quenching process. The extent of quenching brought about by acrylamide is always more in the presence of the crowders than in their absence while just the reverse is observed for KI, with Dextran 70 being the exception in the latter case. This suggests that in the presence of the crowding agents the exposed Trp residues as probed by iodide become more buried, while for acrylamide, the buried ones become more exposed, leading to increased quenching for the neutral molecule. However, the quenching results are also mirrored to a certain extent by the changes brought about for free tryptophan only (Table 2) under identical conditions, indicating that the observed results for lysozyme are not solely due to changes in the protein conformation as assumed, but are partly a characteristic of the crowded milieu and their differential interactions with the neutral and charged quenchers. We propose that the increase in acrylamide quenching as compared to KI is an outcome of the nature of entanglement of the macromolecular crowding agents, with the preference for the neutral quencher over the charged one implying the not-sopolar nature of the polymer (crowder) mesh. This is further augmented by the observed predominance of the sphere of action models that needed to be invoked, thereby reflecting on the complexity of such reactions taking place in the crowded milieu. Moreover, as mentioned before, the micropolarity of Dextran 70 being more than that of the other two dextranbased crowders, the iodide ions thus find themselves in a more compatible environment and hence are more effective as quenchers. Assuming the quenching mechanism to be diffusion controlled, the bimolecular rate constant kq (= kD) at a fixed temperature, is inversely proportional to the viscosity based on the following sets of equations:

agents, that is, Dextran 6, Dextran 40, and Dextran 70. The latter should not be having too much of an effect on our analyses since the these crowding agents typically encompass all the characters and complexities that a crowded environment can possess, starting from being predominantly the excluded volume type, as in Dextran 6, to exhibiting appreciable soft interactions, as in Dextran 70.48,49 In case of acrylamide, plot of the quenching constants (KSV) as a function of temperature in the presence of the crowders show distinct jumps in values around the melting temperature while the same is absent when KI was used as the quencher. These data thus almost conclusively prove that the iodide ions are indeed probing the surface exposed Trp residues since these are not expected to be sensitive to the cooperative nature of the denaturation of lysozyme. Moreover, the large increase in KSV for lysozyme only, from 2.6 M−1 at 25 °C to 5.1 M−1 at 85 °C in the presence of acrylamide, as compared to the small enhancement from 1.7 M−1 to 2.3 M−1 under similar conditions for KI, is evidence of the fact the though the previously buried tryptophans do become accessible, but not to the extent that the iodide can easily quench their fluorescence. Even in the presence of the crowding agents the aforementioned trend is maintained indicating that the overall disposition of the surface exposed Trp residues vary only slightly at the lower and higher temperatures for the present solvent compositions. Additionally a few other interesting points also emerge from the thermal studies. In presence of acrylamide, Dextran 6 is the crowder that induces maximum quenching. Considering the fact that Dextran 6 itself also brought about dramatic quenching of the Trp residues which we hypothesized above as resulting from increased protein compaction, this result for acrylamide was thus quite surprising. In other words the observations imply that in the presence of this crowding agent, the protein structure is affected in a manner such that specific channels or gates might open up thereby facilitating the approach of the quencher to the Trp residue(s). Evidence of such gated mechanism has been suggested in a few previous reports wherein such a gate, when open, allows transient exposure of the fluorophore to the quencher molecule in the protein matrix.50−52 For quencher molecules which cannot penetrate the protein matrix, quenching of a buried fluorophore of a protein requires the residue to become occasionally exposed to the solvent (and therefore to the quencher molecules as well). With KI as the quencher, however, Dextran 70 at 200 g/L induces the maximum effect, an aspect we feel is partially dictated by the nature of the crowder rather than only the protein conformation itself and has been further addressed below. Another striking aspect in the quenching profiles is the frequency at which plots become nonlinear, that is, where the sphere-of-action needs to be invoked. For example, in case of quenching during thermal denaturation (SI Tables S3−S6), KI was more prone to give rise to profiles having upward curvatures as compared to that of acrylamide, while during chemical denaturation experiments, this trend was reversed, particularly in case of GdmHCl as the denaturant (Table 1). For further insights in this regard we also performed quenching studies of the free amino acid tryptophan in selected denaturant (urea and GdmHCl) concentrations. Even in buffer only, that is, in absence of the crowders but in the presence of the denaturants, the quenching induced for acrylamide was far from linear. Existence of the sphere-of-action volumes in case of tryptophan-only quenching by crowders points toward the 4696

DOI: 10.1021/acs.jpcb.7b01055 J. Phys. Chem. B 2017, 121, 4687−4699

Article

The Journal of Physical Chemistry B kD =

4πRND 1000

(6)

D=

kBT 6πRη

(7)

medium in general: (a) a weak viscosity dependence suggesting a predominantly non-diffusion limited regime, (b) considerable presence of the sphere of action volume, “V”, suggesting preferential distribution of quenchers which increases in the presence of crowding agents, (c) higher concentration of crowding agents leading to increase in the local concentration of quencher thereby bringing about a change in the mechanism by which quenching occurs, and (d) the extent of quenching is dictated to an appreciable extent by the nature of the quencher and its ability to percolate through the crowded medium, the latter being dependent on the micropolarity of the polymer mesh.

where kD is Smoluchowski’s diffusion controlled rate constant, R is the sum of molecular radii of the fluorophore and the quencher, and D is the sum of the diffusion coefficients of the quencher and fluorophore, N is the Avogadro’s number, kB is Boltzmann’s constant, T is the absolute temperature, R is the radius of the species, and η is the solvent viscosity. Eq 6 is the Smoluchowski expression for calculating the diffusion controlled rate constant while eq 7 is the extensively used Stokes−Einstein equation for determination of diffusion coefficients.41,53,54 Assuming all other factors to remain unchanged with measurements being carried out at a fixed temperature, one can write the following expression:

kq1 kq2

=



CONCLUSION In this study we have focused on the quenching phenomenon of a multitryptophan protein, HEWL, in the presence of several macromolecular crowding agents. The type of quenching that was observed was of primarily two types: (i) quenching brought about by the macromolecular crowder itself, albeit indirectly, through conformational modulations and (ii) quenching of the intrinsic tryptophan residues using small molecule quenchers. Quenching induced by the crowders themselves was strictly static in nature, with the extent of complex formation being dictated by the surrounding amino acids in proximity of the Trp residues and the crowder in question. Dextran 6 was observed to induce the maximum quenching signifying extensive local perturbation even for the native state of the protein without any appreciable change in the secondary structure, an aspect that we have attributed to the high number density of this small molecular weight crowding agent. Surprisingly, PEG8000, having similar average molecular weight to that of Dextran 6 was not that effective at all thereby bringing to light the intrinsic differences in the nature of entanglement of these crowders and their subsequent effects on the conformations of biomolecules. Quenching by acrylamide and KI provided further insights into the complexity of the crowded milieu. In presence of Dextran 70, iodide induced the maximum quenching, a feature that we have attributed to the greater ability of the charged quencher to percolate through the crowder mesh based on the underlying micropolarity of the polymer solution. This combined with the fact that in the presence of the crowding agents, the quenching brought about by acrylamide and KI were dominated by the sphere-of-action type quenching model, wherein a large fraction of the quencher molecules do not need to diffuse to exhibit their quenching effect, reveals the presence of a high local concentration of these quencher molecules in the immediate proximity of the protein, more specifically, the tryptophan residues. The latter is a direct effect of the increase in activity of the quenching solutes in the presence of the crowding agents, confirmation of which is also obtained from the quenching studies on the free amino acid (tryptophan).

η2 η1

(8)

For the different dextran-based crowding agents used here the bulk viscosity varied (in absence of denaturant) as follows: 2.65 cP and 4.78 cP for 100 g/L and 200 g/L of Dextran 6, 4.22 cP and 12.46 cP for 100 g/L and 200 g/L of Dextran 40, and 8.16 cP and 36.15 cP for 100 g/L and 200 g/L of Dextran 70. The ratio of the viscosities for increasing crowder concenη trations from 100 g/L to 200 g/L, that is 200 , increases as 1.80 η100

for Dextran 6, 2.95 for Dextran 40, and 4.43 for Dextran 70. As shown (SI Table S7) there is hardly any correlation between the

kq100 kq200

and ratios, suggesting that the observed quenching is not

strictly diffusion controlled. For convenience we have assumed η the ratio of the viscosity that is 200 to be the same in the η100

presence of the denaturants too. These have further been confirmed in the quenching of the free amino acid, where the viscosity of the solution is not the sole determinant of the observed KSV values as can also be understood from the quenching data in the presence of the denaturants. For 8 M urea and 7 M GdmHCl wherein the viscosity increases, the quenching of the Trp in buffer only (that is in absence of crowders) remains almost the same. Additionally in 8 M urea for Dextran 6, the quenching constant in 200 g/L of the crowder is just twice that in 100 g/L, thus further pointing to a quenching mechanism that is independent of viscosity. Quenching of tryptophan residues in proteins involving different quenchers has been addressed in quite a few studies.55−57 The observed quenching can arise from several mechanisms, namely, (i) a viscosity dependent diffusioncontrolled process, (ii) reaction under the rapid diffusion limit (RDL) in which the quenching rate is typically independent of viscosity, and (iii) penetration of the quenchers through the protein matrix. Presence of macromolecular crowding agents introduces additional complexities with regards to not only affecting the nature of diffusion because of variation in the intrinsic viscosity arising from the entangled nature of the polymers, but also due to the excluded volume effect, with the latter known to increase the activity of the quencher leading to enhanced collisions and hence quenching. Thus, from this study we have the following important observations that we think will be applicable for any crowded



ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.jpcb.7b01055. Tables showing (i) Ksv, Ksv’, and “V” values of HEWL for different crowders in the presence of quenchers at a fixed temperature; (ii) KD, kq, and Ksv/KD values of HEWL in the presence of Dextran-based crowders for varying concentrations of acrylamide and iodide at a fixed 4697

DOI: 10.1021/acs.jpcb.7b01055 J. Phys. Chem. B 2017, 121, 4687−4699

Article

The Journal of Physical Chemistry B



denaturant concentration; (iii) fluorescence lifetimes of HEWL in the absence and presence of Dextran-based crowders for varying concentrations of acrylamide and iodide at a fixed denaturant concentration; (iv) binding plots to confirm the static nature of quenching; and (v) percentage change in ellipticity of HEWL for a crowder concentration of 200 g/L in the presence of varying concentrations of denaturants; and F0/F and τ0/τ plots for HEWL and F0/F plots for the free amino acid Trp in presence and absence of dextran-based crowding agents (PDF)

(13) Benton, L. A.; Smith, A. E.; Young, G. B.; Pielak, G. J. Unexpected Effects of Macromolecular Crowding on Protein Stability. Biochemistry 2012, 51, 9773−9775. (14) Senske, M.; Tork, L.; Born, B.; Havenith, M.; Herrmann, C.; Ebbinghaus, S. Protein Stabilization by Macromolecular Crowding Through Enthalpy Rather than Entropy. J. Am. Chem. Soc. 2014, 136, 9036−9041. (15) Sapir, L.; Harries, D. Origin of Enthalpic Depletion Forces. J. Phys. Chem. Lett. 2014, 5, 1061−1065. (16) Dhar, A.; Samiotakis, A.; Ebbinghaus, S.; Nienhaus, L.; Homouz, D.; Gruebele, M.; Cheung, M. S. Structure, Function, and Folding of Phosphoglycerate Kinase Are Strongly Perturbed by Macromolecular Crowding. Proc. Natl. Acad. Sci. U. S. A. 2010, 107, 17586−17591. (17) Cheung, M. S.; Klimov, D.; Thirumalai, D. Molecular Crowding Enhances Native State Stability and Refolding Rates of Globular Proteins. Proc. Natl. Acad. Sci. U. S. A. 2005, 102, 4753−4758. (18) Homouz, D.; Perham, M.; Samiotakis, A.; Cheung, M. S.; Wittung-Stafshede, P. Crowded, Cell-like Environment Induces Shape Changes in Aspherical Protein. Proc. Natl. Acad. Sci. U. S. A. 2008, 105, 11754−11759. (19) Royer, C. A. Probing Protein Folding and Conformational Transitions with Fluorescence. Chem. Rev. 2006, 106, 1769−1784. (20) Vallée-Bélisle, A.; Michnick, S. W. Visualizing Transient Proteinfolding Intermediates by Tryptophan-scanning Mutagenesis. Nat. Struct. Mol. Biol. 2012, 19, 731−736. (21) Munishkina, L.; Fink, A. L. Fluorescence as a Method to Reveal Structures and Membrane-interactions of Amyloidogenic Proteins. Biochim. Biophys. Acta, Biomembr. 2007, 1768, 1862−1885. (22) Vivian, T. J.; Callis, P. R. Mechanisms of Tryptophan Fluorescence Shifts in Proteins. Biophys. J. 2001, 80, 2093−2109. (23) Prigozhin, M. B.; Chao, S.-H.; Sukenik, S.; Pogorelov, T. V.; Gruebele, M. Mapping Fast Protein Folding with Multiple-site Fluorescent Probes. Proc. Natl. Acad. Sci. U. S. A. 2015, 112, 7966− 7971. (24) Tokunaga, Y.; Sakakibara, Y.; Kamada, Y.; Watanabe, K.; Sugimoto, Y. Analysis of Core Region from Egg White Lysozyme Forming Amyloid Fibrils. Int. J. Biol. Sci. 2013, 9, 219−227. (25) Laurents, D. V.; Baldwin, R. L. Characterization of the Unfolding Pathway of Hen Egg White Lysozyme. Biochemistry 1997, 36, 1496−1504. (26) Matagne, A.; Dobson, C. M. The Folding Process of Hen Lysozyme: A Perspective from the ‘New View’. Cell. Mol. Life Sci. 1998, 54, 363−371. (27) Blake, C. C. F.; Koenig, D. F.; Mair, G. A.; North, A. C.; Phillips, D. C.; Sarma, V. R. Structure of Hen Egg-White Lysozyme: A ThreeDimensional Fourier Synthesis at 2 Angstrom Resolution. Nature 1965, 206, 757−761. (28) Yamashita, S.; Nishimoto, E.; Yamasaki, N. The Steady State and Time-resolved Fluorescence Studies on the Lysozyme-Ligand Interaction. Biosci., Biotechnol., Biochem. 1995, 59, 1255−1261. (29) Nishimoto, E.; Yamashita, S.; Yamasaki, N.; Imoto, T. Resolution and Characterization of Tryptophyl Fluorescence of Hen Egg-White Lysozyme by Quenching- and Time-Resolved Spectroscopy. Biosci., Biotechnol., Biochem. 1999, 63, 329−336. (30) Booth, D. R.; Sunde, M.; Bellotti, V.; Robinson, C. V.; Hutchinson, W. L.; Fraser, P. E.; Hawkins, P. N.; Dobson, C. M.; Radford, S. E.; Blake, C. C. F.; Pepys, M. B. Instability, Unfolding and Aggregation of Human Lysozyme Variants underlying Amyloid Fibrillogenesis. Nature 1997, 385, 787−793. (31) Pepys, M. B.; Hawkins, P. N.; Booth, D. R.; Vigushin, D. M.; Tennent, G. A.; Soutar, A. K.; Totty, N.; Nguyen, O.; Blake, C. C.; Terry, C. J.; et al. Human Lysozyme Gene Mutations cause Hereditary Systemic Amyloidosis. Nature 1993, 362, 553−557. (32) Phillips, S. R.; Wilson, L. J.; Borkman, R. F. Acrylamide and Iodide Fluorescence Quenching as a Structural Probe of Tryptophan Microenvironment in Bovine Lens Crystallins. Curr. Eye Res. 1986, 5, 611−619. (33) Eftink, M. R.; Ghiron, C. A. Fluorescence Quenching Stuides with Proteins. Anal. Biochem. 1981, 114, 199−227.

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]; Tel: +911126591521. ORCID

Pramit K. Chowdhury: 0000-0002-9593-2577 Notes

The authors declare no competing financial interest.



ACKNOWLEDGMENTS Priyanka Singh thanks Indian Institute of Technology (IIT) Delhi for her institute fellowship. Pramit K Chowdhury thanks the Department of Science and technology (DST), New Delhi, India, for financial support under the Fast Track Scheme for Young Scientists (SR/FT/CS-007/2010) and IIT Delhi for startup funding.



REFERENCES

(1) Ellis, R. J. Macromolecular Crowding: Obvious but Underappreciated. Trends Biochem. Sci. 2001, 26, 597−604. (2) Ellis, R. J. Macromolecular Crowding: An Important but Neglected Aspect of the Intracellular Environment. Curr. Opin. Struct. Biol. 2001, 11, 114−119. (3) Zhou, H. X.; Rivas, G.; Minton, A. P. Macromolecular Crowding and Confinement: Biochemical, Biophysical, and Potential Physiological Consequences. Annu. Rev. Biophys. 2008, 37, 375−397. (4) Rivas, G.; Minton, A. P. Macromolecular Crowding In Vitro, In Vivo, and In Between. Trends Biochem. Sci. 2016, 41, 970−981. (5) Golkaram, M.; Hellander, S.; Drawert, B.; Petzold, L. R. Macromolecular Crowding Regulates the Gene Expression Profile by Limiting Diffusion. PLoS Comput. Biol. 2016, 12, e1005122. (6) Vazquez, A.; Oltvai, Z. N. Macromolecular Crowding Explains Overflow Metabolism in Cells. Sci. Rep. 2016, 6, 31007. (7) Nakano, S. I.; Sugimoto, N. Model Studies of the Effects of Intracellular Crowding on Nucleic Acid Interactions. Mol. BioSyst. 2017, 13, 32−41. (8) Ando, T.; Yu, I.; Feig, M.; Sugita, Y. Thermodynamics of Macromolecular Association in Heterogeneous Crowding Environments: Theoretical and Simulation Studies with a Simplified Model. J. Phys. Chem. B 2016, 120, 11856−11865. (9) Wirth, A. J.; Platkov, M.; Gruebele, M. Temporal Variation of a Protein Folding Energy Landscape in the Cell. J. Am. Chem. Soc. 2013, 135, 19215−19221. (10) Sharp, K. Analysis of the Size Dependence of Macromolecular Crowding Shows That Smaller Is Better. Proc. Natl. Acad. Sci. U. S. A. 2015, 112, 7990−7995. (11) Minton, A. P. Models for Excluded Volume Interaction between an Unfolded Protein and Rigid Macromolecular Cosolutes: Macromolecular Crowding and Protein Stability Revisited. Biophys. J. 2005, 88, 971−985. (12) White, D. A.; Buell, A. K.; Knowles, T. P. J.; Welland, M. E.; Dobson, C. M. Protein Aggregation in Crowded Environments. J. Am. Chem. Soc. 2010, 132, 5170−5175. 4698

DOI: 10.1021/acs.jpcb.7b01055 J. Phys. Chem. B 2017, 121, 4687−4699

Article

The Journal of Physical Chemistry B

(53) Strambini, G. B.; Gonnelli, M. Protein Phosphorescence Quenching: Distinction between Quencher Penetration and External Quenching Mechanisms. J. Phys. Chem. B 2010, 114, 9691−9697. (54) Arık, M.; Ç elebi, N.; Onganer, Y. Fluorescence Quenching of Fluorescein with Molecular Qxygen in Solution. J. Photochem. Photobiol., A 2005, 170, 105−111. (55) Chen, Y.; Barkley, M. D. Toward Understanding Tryptophan Fluorescence in Proteins. Biochemistry 1998, 37, 9976−9982. (56) Calhoun, D. B.; Vanderkooi, J. M.; Holtom, G. R.; Englander, S. W. Protein Fluorescence Quenching by Small Molecules: Protein Penetration Versus Solvent Exposure. Proteins: Struct., Funct., Genet. 1986, 1, 109−115. (57) Gałęcki, K.; Kowalska-Baron, A. Room Temperature Fluorescence and Phosphorescence Study on the Interactions of Iodide ions with Single Tryptophan Containing Serum Albumins. Spectrochim. Acta, Part A 2016, 169, 16−24.

(34) Singh, P.; Chowdhury, P. K. Crowding-Induced Quenching of Intrinsic Tryptophans of Serum Albumins: A Residue-Level Investigation of Different Conformations. J. Phys. Chem. Lett. 2013, 4, 2610−2617. (35) Geddes, C. D. Optical Halide Sensing Using Fluorescence Quenching: Theory, Simulations and ApplicationsA Review. Meas. Sci. Technol. 2001, 12, R53−R88. (36) Venkataramani, S.; Truntzer, J.; Coleman, D. R. Thermal Stability of High Concentration Lysozyme Across Varying pH: A Fourier Transform Infrared Study. J. Pharm. BioAllied Sci. 2013, 5, 148−153. (37) Knubovets, T.; Osterhout, J. J.; Connolly, P. J.; Klibanov, A. M. Structure, Thermostability, and Conformational Flexibility of Hen Egg-White Lysozyme Dissolved in Glycerol. Proc. Natl. Acad. Sci. U. S. A. 1999, 96, 1262−1267. (38) Kumari, M.; Dohare, N.; Maurya, N.; Dohare, R.; Patel, R. Effect of 1-Methyl-3-Octyleimmidazolium Chloride on the Stability and Activity of Lysozyme: A Spectroscopic and Molecular Dynamics Studies. J. Biomol. Struct. Dyn. 2016, 1−15. (39) Roy, A. S.; Ghosh, P. Characterization of the Binding of Flavanone Hesperetin with Chicken Egg Lysozyme using Spectroscopic Techniques: Effect of pH on the Binding. J. Inclusion Phenom. Macrocyclic Chem. 2016, 84, 21−34. (40) Formoso, C.; Forster, L. S. Tryptophan Fluorescence Lifetimes in Lysozyme. J. Biol. Chem. 1975, 250 (10), 3738−3745. (41) Strambini, G. B.; Gonnelli, M. Fluorescence Quenching of Buried Trp Residues by Acrylamide Does Not Require Penetration of the Protein Fold. J. Phys. Chem. B 2010, 114, 1089−1093. (42) Muino, P. L.; Callis, P. R. Solvent Effects on the Fluorescence Quenching of Tryptophan by Amides via Electron Transfer. Experimental and Computational Studies. J. Phys. Chem. B 2009, 113, 2572−2577. (43) Jensen, R. L.; Arnbjerg, J.; Ogilby, P. R. Reaction of Singlet Oxygen with Tryptophan in Proteins: A Pronounced Effect of the Local Environment on the Reaction Rate. J. Am. Chem. Soc. 2012, 134, 9820−9826. (44) Kuznetsova, I. M.; Biktashev, A. G.; Bushmarina, N. A.; Uversky, V. N.; Turoverov, K. K. Understanding the Contribution of Individual Tryptophan Residues to Intrinsic Lysozyme Fluorescence. ChemBioChem 2000, 7, 411−420. (45) Lehrer, S. S. Solute Perturbation of Protein Fluorescence. The Quenching of the Tryptophyl Fluorescence of Model Compounds and of Lysozyme by Iodide Ion. Biochemistry 1971, 10 (17), 3254−3263. (46) Beg, I.; Minton, A. P.; Hassan, I.; Islam, A.; Ahmad, F. Thermal Stabilization of Proteins by Mono- and Oligosaccharides: Measurement and Analysis in the Context of an Excluded Volume Model. Biochemistry 2015, 54, 3594−3603. (47) Kuznetsova, I. M.; Turoverov, K. K.; Uversky, V. N. What Macromolecular Crowding Can Do to a Protein. Int. J. Mol. Sci. 2014, 15, 23090−23140. (48) Biswas, S.; Chowdhury, P. K. Correlated and Anticorrelated Domain Movement of Human Serum Albumin: A Peek into the Complexity of the Crowded Milieu. J. Phys. Chem. B 2016, 120, 4897− 4911. (49) Biswas, S.; Chowdhury, P. K. Unusual Domain Movement in a Multidomain Protein in the Presence of Macromolecular Crowders. Phys. Chem. Chem. Phys. 2015, 17, 19820−19833. (50) Strambini, G. B.; Gonnelli, M. Acrylamide Quenching of Trp Phosphorescence in Liver Alcohol Dehydrogenase: Evidence of Gated Quencher Penetration. Biochemistry 2009, 48, 7482−7491. (51) Strambini, G. B.; Gonnelli, M. Amplitude Spectrum of Structural Fluctuations in Proteins from the Internal Diffusion of Solutes of Increasing Molecular Size: A Trp Phosphorescence Quenching Study. Biochemistry 2011, 50, 970−980. (52) Somogyi, B.; Norman, J. A.; Punyiczki, M.; Rosenberg, A. Viscosity Dependence of Acrylamide Quenching of Ribonuclease T1 Fluorescence: The Gating Mechanism. Biochim. Biophys. Acta, Protein Struct. Mol. Enzymol. 1992, 1119, 81−89. 4699

DOI: 10.1021/acs.jpcb.7b01055 J. Phys. Chem. B 2017, 121, 4687−4699