Estimating conformation content of a protein using citrate-stabilized Au nanoparticles

Abstract

Herein we report the use of the optical properties of citrate-stabilized gold nanoparticles (Au NPs) for estimation of native or denatured conformation content in a mixture of a protein in solution. The UV-vis extinction spectrum of citrate-stabilized Au NPs is known to broaden differently in the presence of native and denatured states of α-amylase, bovine serum albumin (BSA) or amyloglucosidase (AMG). On the other hand, herein we show that when a mixture of native and denatured protein was present in the medium, the broadening of the spectrum differed for different fractional content of the conformations. Also, the total area under the extinction spectrum varied linearly with the change in the mole fraction content of a state and for a constant total protein concentration. Transmission electron microscopy (TEM) measurements revealed different levels of agglomeration for different fractional contents of the native or denatured state of a protein. In addition, time-dependent denaturation of a protein could be followed using the present method. The rate constants calculated for denaturation indicated a possible fast change in conformation of a protein before complete thermal denaturation. The observations have been explained based on the changes in extinction coefficient (thereby oscillator strength) upon interaction of citrate-stabilized NPs with proteins being in different states and levels of agglomeration.

---

Proteins, being ubiquitous in biological systems, have drawn considerable research attention in terms of understanding their structures and functions. This has special relevance to diseases caused by misfolding or unfolding and consequent agglomeration of proteins. Furthermore, the use of proteins in catalyses makes the understanding of structure–function relationships crucial to the chemical and biochemical industries. Fortunately, availability of a large number of crystal structures has made structure–function correlation easier.¹ However; a knowledge of solution structure of the protein holds the key to the understanding of the function of the molecules in solution. It would be worthwhile if one could follow the evolution of conformation content of a protein—as it degrades, folds or unfolds in solution—in addition to understanding the folding and unfolding processes. It is interesting to observe that although there are several methods, based on nuclear magnetic resonance,^2,3 infrared spectroscopy,^4–6 differential scanning calorimetry,^7,8 near-UV circular dichroism,^3,4,9,10 fluorescence^4,11 and surface plasmon resonance^12–16—among others—that are currently available for following the conformational changes in a protein, the estimation of conformation content in solution has somehow eluded appropriate attention.

On the other hand, there is a growing body of literature on the use of nanoscale materials for bioassays and diagnostics. For example, metal nanoparticles (NPs) especially those of gold (Au) and quantum dots have been of great use in the assay of protein^17–21 and DNA^22–26 and for observing denaturation of protein.^12–16 The general approach to probing biomolecules such as proteins and DNA using NPs has been to functionalize the NPs with specific groups that would attach specifically to target molecules, although use of unmodified Au NPs for probing DNA sequences is known.²⁴ The approach provides an easy way of identification and estimation of biomolecules using optical properties of the NPs. Interestingly, specificity of interactions between the probe and biomolecules has been deemed of such importance that non-specific interactions are considered as non-informative or of little significance. However, biomolecules such as proteins provide an opportunity to study the interaction between ‘non-functionalized’ citrate-stabilized Au NPs and the protein with some degree of specificity, as the (protein) molecules consist of both hydrophobic and hydrophilic groups in abundance. In addition, the conformational state of a protein could provide the required specificity in the interaction with citrate-stabilized Au NPs. In this regard, we have recently shown that the optical properties of citrate-stabilized Au NPs could be used to probe the concentration of protein with the distinction of its native and denatured states.²⁷ In other words; interactions between native and denatured states of a protein with citrate-stabilized Au NPs are of sufficient difference to be able to distinguish them in aqueous solution. Therein we had observed that interaction of proteins with citrate-stabilized Au NPs can lead to significant changes in the extinction spectrum of citrate-stabilized Au NPs, which is dependent on both the concentration of the protein and its conformation. For example, the change in the area under the UV-vis extinction spectrum of citrate-stabilized Au NPs as a result of addition of a protein solution varied linearly with the concentration of the protein. Moreover, the extent of change depended on the conformation of the protein, thus providing a way of distinguishing the conformations.

Herein we report the use of citrate-stabilized Au NPs in the estimation of native and denatured fractions in a mixture of a protein. We have used α-amylase, bovine serum albumin (BSA) and amyloglucosidase (AMG) as the model proteins for the present study. For all the proteins, the changes in the area under the absorption spectrum due to citrate-stabilized Au NPs varied linearly with the mole fraction of the native (and thus denatured) conformation. However, slopes of the graphs not only were different in magnitude but also in sign, indicating specificity of interactions not only based on conformational states but also on the nature of the protein. In addition, time-dependence of denaturation of a protein at various temperatures could be followed in terms of fraction of native (or denatured) conformations. The rate constants for denaturation indicated faster conformational changes in comparison to known rate of denaturation. The results have been explained by Mie scattering theory based on the changes in the dielectric constant of the stabilizer as a result of interaction of citrate-stabilized Au NPs with protein. In addition, significant change in the spectrum was associated with agglomeration of NPs in the presence of proteins. The observations have been exploited to develop a method for quick and easy assay of the purity of a protein in terms of its conformation content. This may be useful at least in routine laboratory analyses of known proteins using a UV-vis spectrophotometer. Our investigations also suggest that in a mixture of native and denatured proteins there may not be any associative interaction between the native protein-stabilized Au NPs and the denatured protein-stabilized Au NPs.

It is known that UV-vis extinction spectrum of citrate-stabilized Au NPs broadens in the presence of proteins such as α- amylase, BSA and AMG, when proteins are either in the native or denatured form.²⁷ Also, the area under the extinction curve changes linearly with the change in concentration of the protein (being either in native or denatured form) up until certain concentration of the protein. This limiting concentration varies from protein to protein and also between the native and denatured conformations of a particular protein. It is important to mention here that the total concentration of protein that has been used here was decided in a way such that it falls within linear region of total area under the UV-vis curve versus concentration of protein plot. Thus when 0.054 μg mL⁻¹ of native α-amylase was added to the citrate-stabilized Au NP solution the UV-vis spectrum broadened significantly as shown in Fig. 1A. Also, when the same amount of denatured protein was added to the Au NP solution, broadening could be observed. However, broadening was much less when denatured protein was added in comparison to the addition of same amount of native protein. Now, when a mixture of native and denatured protein was added by systematically varying their ratio, intermediate broadening was observed. In brief, systematic broadening of Au NP UV-vis extinction spectrum was observed in the presence of various ratios of native and denatured α-amylase when the total concentration of protein was kept constant. Similar broadening of spectrum was observed in the presence of 1.64 μg mL⁻¹ native or denatured form of BSA (Fig. 1B). However, here broadening was more in the presence of denatured protein than the native conformation at the same concentration. Intermediate broadening of the spectrum was observed in the presence of a mixture of native and denatured form, as is clear from Fig. 1B. When the experiments were carried out with 0.125 μg mL⁻¹ native or denatured form of AMG the results were similar to those of BSA. The details of the results (involving AMG) are available in the ESI (Fig. S1†). Essentially, broadening of the UV-vis extinction spectrum of citrate stabilized Au NPs could be observed in the presence of a mixture of native and denatured forms of a protein under study, where the variation of the ratio of the concentrations of the conformations led to discernable difference in the broadening of the spectrum. This formed the basis for finding the ratio of native and denatured conformations of a protein present in aqueous solution. Further, it is important to mention here that when similar experiments were carried out with thioglycolic acid (TGA) stabilized Au NPs, no UV-vis spectral broadening was observed in the presence of α-amylase (refer to the ESI; Fig. S2†), indicating specificity of interaction of protein with citrate-stabilized Au NPs.

Fig. 1 UV-vis spectra of citrate-stabilized Au NPs before and after addition of protein solution with increasing mole fraction of the native form of (A) α-amylase and (B) BSA. The arrows in the Figures show the increasing mole fraction of the native form.

That the change in the UV-vis extinction spectrum of citrate-stabilized Au NPs upon addition of native and denatured proteins was due to interaction of the stabilized NPs with the protein, was further substantiated by photoluminescence studies. It has been established that citrate-stabilized Au NPs quench efficiently the fluorescence from fluorophores, especially proteins.²⁸ In this regard, the decrease in fluorescence emission at ca. 365 nm owing to tryptophan (trp) residue in the protein acts as a marker of interaction between the citrate-stabilized Au NPs and the protein. In the present set of experiments when native, denatured or a mixture of the two forms of protein was incubated with citrate-stabilized Au NPs in buffer solutions, the fluorescence emission at ca. 365 nm disappeared completely. The excitation wavelength was kept at 280 nm. The results for α-amylase and BSA are shown in Fig. 2, while those of AMG are in the ESI (Fig. S3†). As is clear from the Figures, in all cases the disappearance of the emission peak (at the total protein concentrations that were used in the UV-vis extinction studies and at the Au NP concentration that was used in each measurement) supported that indeed the proteins were attached to the citrate-stabilized Au NPs present in the medium. Also, it is important to note here that complete disappearance of the peak at 365 nm due to a mixture of native and denatured forms (0.5 mole fractions) suggests that both forms of protein were completely attached to the NPs. Further, the fluorescence spectrum of native α-amylase in the presence of TGA-stabilized Au NPs was recorded to check whether the protein interacts with such Au NPs. The result is shown in the ESI, Fig. S4,† which demonstrated quenching of the fluorescence emission of α-amylase in presence of TGA-stabilized Au NPs. The quenching of fluorescence suggests that there is interaction between the protein molecules and the TGA-stabilized NPs. In addition, as mentioned earlier the UV-vis spectrum of TGA-stabilized NPs exhibited no discernible broadening in the presence of the protein (α-amylase). These mean that interaction of protein with citrate-stabilized Au NPs is different from that with TGA-stabilized Au NPs. It is plausible that citrate, being electrostatically attached to the Au NPs, could be amenable to partial (if not complete) replacement by the protein molecules, leading to agglomeration of the NPs. On the other hand TGA, being covalently bound to Au NPs through the S–Au bond, could not be replaced, even though the protein molecules interact with the TGA-stabilized Au NPs thus showing no signs of agglomeration of the NPs. Further, a recent study by Brewer et al.²⁹ suggests that BSA molecules bind to the citrate stabilized Au NPs predominantly by an electrostatic mechanism, which involves salt-bridges, for example, of the carboxylate-ammonium type, between the citrate and the lysine on the protein surface and displacement of citrate by the protein molecule could not be supported by their studies.

Fig. 2 Fluorescence spectra of various compositions of (A) α-amylase and (B) BSA, in the presence and absence of citrate-stabilized Au NPs.

In order to further probe the interaction between the proteins and citrate-stabilized Au NPs, TEM measurements were performed with samples prepared using α-amylase. We have earlier observed that the native form of α-amylase—beyond a certain concentration—leads to substantial aggregation of citrate-stabilized Au NPs.²⁷ On the other hand, the denatured form of the protein does not lead to agglomeration of the NPs. Herein, TEM measurements were performed with samples evaporated on grids consisting of citrate-stabilized Au NPs only, citrate-stabilized Au NPs in the presence of native α-amylase, denatured α-amylase and a mixture of a 0.5 mole fraction of each native and denatured protein. The images are shown in Fig. 3A–D. As is evident from the Figures, the NPs in the presence of native protein exhibited maximum agglomeration (Fig. 3B), whereas those in the presence of denatured α-amylase had the minimum agglomeration of NPs (Fig. 3D). On the other hand, in the presence of 50% of either of native and denatured proteins (Fig. 3C), the level of agglomeration was roughly in between the two. In other words, in the presence of a 0.5 mole fraction content in the mixture there was agglomeration of the NPs; however, the NPs were not substantially agglomerated like those in the presence of only native form at the same total protein concentration. There was substantial presence of unagglomerated NPs. What is interesting here is that the level of agglomeration was dependent on the conformation of the proteins, which possibly led to a differential broadening of UV-vis extinction spectrum of citrate-stabilized Au NPs. Essentially, the extent of agglomeration varied with the mole fraction of the native (or denatured) protein and this helped make the UV-vis spectral broadening a quantitative measure of the conformation content of the protein in a mixture.

Fig. 3 Transmission electron microscopic images of (A) citrate-stabilized Au NPs, (B) citrate-stabilized Au NPs in the presence of native protein (α-amylase) only, (C) citrate-stabilized Au NPs in the presence of 50% native and 50% denatured protein mixture and (D) citrate-stabilized Au NPs in the presence of 100% denatured protein. The total concentration of protein in all the samples was the same.

The extinction of light by a colloidal Au NP solution with N particles per unit volume could be expressed as,³⁰

Here, I₀ and I are the intensities of incident and transmitted light, l is the path length and Q_ext is the extinction coefficient of a single NP. Using Mie scattering theory of small spherical metal particles, the extinction coefficient can be written as,³⁰Here, R is the radius of the NP, λ is the wavelength of the light, ε_m is the dielectric function of the medium. On the other hand, ε′ and ε′′ are the real and imaginary parts of frequency dependent dielectric function (ε) of the NP (ε = ε′ + iε′′). In case of the particle being coated by a layer of organic or inorganic moiety having a different dielectric constant, the extinction coefficient is then written as,^14,30Here, ε₁ and ε₂ are the complex dielectric functions of the particle core and the surface coating, whereas g is the volume fraction occupied by the surface coating. Therefore, the extinction of light would also be varying systematically provided the stabilizer constituting the surface coating of the NP is changed consistently. This could occur in two ways. Firstly, addition of protein to citrate-stabilized Au NPs could result into gradual changes in the constitution of the stabilizing layer of the NPs. At higher concentrations of the protein there could possibly be agglomeration of the protein–NP composite, thereby further changing the dielectric environment of the NP (and hence the extinction of light). Secondly, if the stabilizing layer consists of proteins in denatured or native conformation then there would be significant difference in the extinction coefficient even at the same total concentration of the protein. It has been established that there is a linear relationship between the concentration of the protein and the extinction of light due to citrate-stabilized Au NPs (in the presence of protein in either of native and denatured states).²⁷ Thus in the linear regime, the changes in the extinction spectrum would be additive and would linearly depend on the mole fraction of the native or the denatured form when a mixture of the two forms is present. In other words, area under the extinction spectrum (related to oscillator strength) would vary linearly with the mole fraction of the native form (or denatured form) at a fixed total protein concentration. This would be the case provided there is no associative interaction between the proteins’ (native and denatured conformations) stabilizing citrate-stabilized Au NPs, that would have otherwise given rise to non-linear changes. Thus in the presence of protein with two different forms, citrate-stabilized Au NPs would have extinction of light due to contributions from only citrate-stabilized Au NPs, native form of protein attached to citrate-stabilized Au NPs and denatured form of protein attached to citrate-stabilized Au NPs. Then the total area under the extinction spectrum, at fixed total Au NP and protein concentrations, could be written as follows.

∫A(λ)dλ = x_NP∫A_NP(λ)dλ + x_N∫A_N(λ)dλ + x_D∫A_D(λ)dλ (4)

Here, A(λ) is the wavelength-dependent extinction of light by citrate-stabilized Au NPs in the presence of protein, A_NP(λ) is the wavelength-dependent extinction of light by citrate-stabilized Au NPs, A_N(λ) is the wavelength-dependent extinction of light by native protein attached to citrate-stabilized Au NPs and A_D(λ) is the wavelength-dependent extinction of light by completely denatured protein attached to citrate-stabilized Au NPs. Also, x_N and x_D are the mole fractions of native protein and denatured protein, respectively. It is important to mention here that x_N + x_D = 1. Here x_NP is an empirical dimensionless parameter associated with the contribution of citrate-stabilized Au NPs in the overall extinction of light (in the presence of protein). In the absence of protein x_NP = 1 and when all the NPs are stabilized by protein then x_NP = 0. One can rewrite the equation as

∫A(λ)dλ = x_NP∫A_NP(λ)dλ + x_N∫A_N(λ)dλ + (1 − x_N)∫A_D(λ)dλ (5)

or ∫A(λ)dλ = x_NP∫A_NP(λ)dλ + ∫A_D(λ)dλ + x_N{∫A_N(λ)dλ − ∫A_D(λ)dλ} (6)

There are variations of size of NPs and its distributions in different sets of preparations which will be reflected in the extinction spectra. Thus the total area under the curve would vary from sample to sample. If the variation is sufficiently small then the associated changes could be accommodated by dividing both sides of the eqn (6) by ∫A_NP(λ)dλ. The resultant equation could be expressed as below.

The ratio of the areas in the left hand side of eqn (7) will henceforth be called the normalized area. Thus, if there is no associative interaction between the native and denatured protein attached to citrate-stabilized Au NPs then the normalized area under the extinction curve ought to vary linearly with the mole fraction of the native (or denatured) content in the protein mixture. This is because at the highest concentration of either of native and denatured proteins (used here) both ∫A_N(λ)dλ and ∫A_D(λ)dλ would have constant values, in addition to ∫A_NP(λ)dλ being constant. In order to probe this, the normalized area of each of the graphs obtained after addition of different mole fractions of each protein was calculated and plotted against the mole fraction content of the native form. The results corresponding to the normalized areas under the extinction graphs of citrate-stabilized Au NPs obtained after addition of different mole fractions of α-amylase are shown in Fig. 4A. The extinction curves used here correspond to those in Fig. 1A. Also shown in Fig. 4B is the same normalized area plots calculated from the extinction graphs, obtained in the presence of different mole fractions of BSA, corresponding to those in Fig. 1B. It is interesting to note from both Figures that the normalized area changed linearly with the mole fraction of native (and hence denatured) form of the proteins. Similar results were obtained with AMG, the details of which are available in the ESI (Fig. S5†). The linearity of the relationship between the normalized area and mole fraction content of the protein helps establish the present method of estimation of the concentration of native and denatured proteins in a mixture of the two as a sound one. It also indicates that in a mixture of the native and denatured forms of a protein there is possibly no associative interaction between the two forms—at least for the proteins under investigation.

Fig. 4 Ratio of the area under the UV-vis spectrum of citrate-stabilized Au NPs in presence of protein to that of citrate-stabilized Au NPs only for different composition of native:denatured protein for (A) α-amylase (the enzyme being denatured at 70 °C). (B) BSA (the enzyme being denatured at 70 °C). The data shown are the mean of three sets.

However, there is an interesting and significant difference between the two—i.e. the slope of the graph in the presence of α-amylase is positive whereas that in the presence of BSA is negative. The different values of the slopes (0.131 for α-amylase and −0.137 for BSA) along with the sign demonstrate that interaction between citrate-stabilized Au NPs and protein is specific to the protein and sensitive to its conformation. The difference in the slopes could be attributed to the difference in the extent of broadening by the native and denatured states of a protein. For the case of α-amylase the broadening is greater in the presence of the native form than that in the presence of denatured form at the same concentration of the protein. Hence the slope would be positive. On the other hand, the broadening is greater in the presence of denatured form of BSA in comparison to that in the presence of native form. Hence the slope is negative.

A desirable consequence of the present method of estimation of the fractional content of the native (or denatured) form of a protein would be to be able to follow time-dependent denaturation of a protein at a particular temperature. This was pursued by first keeping several samples of the native protein solution at a designated temperature, followed by withdrawal of the samples at certain intervals of time and then quickly cooling to room temperature. A portion of the sample was then added to the citrate-stabilized Au NP solution and UV-vis spectrum of the mixture was then recorded. The results of time-dependent changes in normalized area under the Au NP extinction graphs, after addition of α-amylase to citrate-stabilized Au NPs that were kept at 60 °C, 65 °C and 70 °C respectively, are shown in Fig. 5A–C. As is clear from the Figures, the normalized areas for all the samples decreased with time and the major decrease in the normalized area occurred for samples kept for 4 min at all three temperatures. It is known from the experiments reported above that the denatured form of α-amylase broadens the extinction curve of Au NP less than that by the native form. Thus the decrease in the area under the extinction curve could be considered to be associated with the denaturation of the protein α-amylase. Hence, denaturation—if at all—was nearly complete by 4 min, when the protein was heated at the above temperatures. However, if one looks closely at the Figures one would observe that the area under the graph at saturation (after 12 min of heating) for protein treated at 60 °C was higher than those treated at 65 °C and 70 °C. For example, the normalized area for the sample treated at 60 °C was 1.01 at 12 min of heating, whereas that was 0.99 and 0.98 for the samples treated at 65 °C and 70 °C respectively. The results indicate that heating α-amylase at 60 °C for 12 min may not lead to complete denaturation of the protein. In order to have a clearer picture, the time-dependences of the normalized area under the graphs were converted into a mole fraction of the native protein content and then plotted against time. It is important to mention here that the mole fractions were calculated by using the slope in the graph in Fig. 4A, in conjunction with eqn (7) and results in Fig. 5A–C. The results for the α-amylase treated at the above three temperatures are shown in Fig. 5D–F. It is interesting to observe that the protein treated at 60 °C did not denature completely, even after 12 min; with a value of mole fraction equivalent to 0.1 it appeared to have remained in the native form. On the other hand, when the protein was heated to 65 °C or 70 °C, complete denaturation occurred in about 4 min. Since the measurements were made at room temperature after being cooled from 60 °C, it is plausible that the results indicate that heating the protein at 60 °C and then cooling to room temperature may lead to partial regeneration of the original structure. On the other hand, heating α-amylase at 65 °C or higher temperatures did not lead to reversion to the original native structure of the protein. Similar observations have been made by others with respect to denaturation of BSA probed using circular dichroism spectroscopy and the primary conclusion has been that structural changes are partially reversible for BSA upon heating to 65 °C.⁹ It could also be possible that the protein (α-amylase) when denatured at 60 °C, followed by cooling to room temperature, reached a conformation different from complete denaturation occurring at 70 °C and thus resulting in different area of the extinction spectrum. However, the results reported herein clearly indicate the possibility of a simple yet powerful method of quantitative estimation of structural changes that occur in the protein upon heating. The results with respect to heating of BSA at 50 °C, 60 °C and 70 °C indicated that the protein was completely denatured at 70 °C. On the other hand, partial denaturation occurred with samples heated at 50 °C and 60 °C, with the extent of denaturation higher for the sample heated at 60 °C than the same being heated at 50 °C. The details of the results are included in the ESI (Fig. S6†). The results corresponding to AMG were similar to those of BSA and are included in the ESI (Fig. S7†).

Fig. 5 (A–C) Ratio of the area of UV-vis spectrum of solution of α-amylase (0.05 mL of 0.054 μg mL⁻¹) in 3 mL citrate-stabilized Au NPs solution to that of only citrate-stabilized Au NPs plotted against the time of denaturation. (D–F) Time-dependent changes in the mole fraction of native protein of α-amylase denatured at different temperatures. The legend shows the temperature at which the protein solution was thermally heated for denaturation. The mole fractions in D–F, corresponding to area ratio of A–C, were calculated from the area of the Au NP peaks based on data in Fig. 4A. The data shown are the mean of three sets.

Further, in order to obtain rate constants for denaturation of the proteins the mole fraction versus time plots were fitted to single exponentials (the data fits are shown in the ESI, Fig. S8–10†). The rate constants for α-amylase were found to be 60.90 ± 2.79 h⁻¹, 62.40 ± 1.98 h⁻¹ and 82.74 ± 9.66 h⁻¹ at heating temperatures of 60 °C, 65 °C and 70 °C respectively. The results indicate that the rate of denaturation increased with temperature. On the other hand, the values of rate constants at all temperatures were an order of magnitude higher than those reported in the literature.³¹ For example, denaturation of α-amylase (from Bacillus licheniformis) studied at 60 °C, 65 °C and 70 °C has been reported to have two rate constants with the values of the first rate constants (higher than the second ones) to be 6.2 h⁻¹ 6.3 h⁻¹ and 15.7 h⁻¹ respectively. Similar results were obtained for studies with BSA and AMG. The results are shown in Table 1. The apparent higher values of the rate constants observed in the present experiments could be rationalized as follows. The literature studies primarily report the actual denaturation of protein as followed by activity studies, although there are some reports where experiments were performed spectroscopically in the presence of a dye molecule. On the other hand, the present studies may indicate initial changes in conformations of the protein resulting in changes in the UV-vis spectrum of citrate-stabilized Au NPs, which is very sensitive to the conformations of the stabilizing protein.¹⁴ In other words, initial rapid changes in conformation of a protein at an elevated temperature, which does not necessarily lead to complete loss of activity of the protein, may also contribute to the changes in the extinction, in addition to actual denaturation. Thus the increase in apparent rate constants for denaturation of the protein observed here. In short, the present method not only captures the denaturation of a protein but also possibly change in conformation which ultimately leads to thermal denaturation of a protein. Further, in order to account for the rate constants one must consider that source of protein and pH of the medium also contribute to difference in the observed rate constants of denaturation of a protein at elevated temperatures.

Table 1 Rate constants for temperature-dependent denaturation of α-amylase, BSA and AMG and their comparison to literature values. All the three proteins in the present study and α-amylase in reported literature³¹ were denatured at pH = 7.0; however BSA in the reported literature³² was denatured at pH = 5.5 and AMG at pH = 4.5. ³³

Name of the protein	Temp./°C	Rate constant, k/h^−1
		Present study value	Literature value
		k 1	k 1	k 2
α-Amylase	60	60.90	6.23	0.482
	65	62.40	6.37	0.349
	70	82.74	15.66	0.666
BSA	50	10.80	—	—
	60	20.94	0.123	—
	70	32.70	1.957	—
AMG	70	20.64	0.871	—
	80	42.78	—	—
	90	63.72	—	—

---

In brief, the method reported herein takes advantage of the changes in the UV-vis extinction spectrum of citrate-stabilized Au NPs in order to estimate the fractional content of native (or denatured) proteins in aqueous solution. The fundamental observation is that the area under the extinction curve of citrate-stabilized Au NPs broadened in the presence of the protein and the extent of broadening depended linearly on the fractional content of native or denatured form of a protein thus providing a simple method of estimation. The origin of the changes has been ascribed to the differential agglomeration of citrate-stabilized Au NPs in the presence of protein with a fractional difference in conformation. Further, the linear dependence of the area under the extinction graphs on the fractional content of the protein (in a particular conformational state) supported that there was no associative interactions between a native protein attached to citrate-stabilized-Au-NPs and a denatured protein attached to citrate-stabilized-Au-NPs. In addition, the method provided a new way of probing time-dependent denaturation of a protein at various temperatures, with the knowledge of the fractional content of the conformation. This simple yet quantitative measure of the purity of a protein is expected to find use in routine laboratory analyses at least for tests of purity, in comparison to classical tests such as those involving activity tests of a protein. Although we have reported the results of studies with three proteins, it would be worthwhile to establish the generality of the procedure by extending the studies to involve a large number of proteins. In addition, the specificity of the interactions between the citrate-stabilized Au NPs and different proteins with specific conformations may provoke studies leading to a newer understanding of the interaction between citrate-stabilized Au NPs and proteins.

Experimental

Preparation of citrate-stabilized Au NPs

1.0 mL of 1.7262 × 10⁻² M HAuCl₄ (Sigma- Aldrich Chemical Co.) was added to 40.0 mL of MilliQ grade water and then heated to boiling. Then, 1.0 mL of 0.5 M trisodium citrate 2-hydrate (E. Merck India Limited, Mumbai) solution was added to the boiling solution (all at once) and refluxed for another 30 min to ensure complete reduction of HAuCl₄. The solution turned deep red, indicating the formation of Au NPs.

Preparation of native/denatured protein solution

1.0 mg mL⁻¹ solution of each protein, namely α-amylase (from hog pancreas, Fluka), BSA (Merck specialties private limited, Mumbai) and AMG (from Aspergillus niger, Fluka) was prepared by dissolving the respective solid proteins in phosphate buffer (of pH 7.0). However since α-amylase was only sparingly soluble, the solution obtained was further stirred for 15 min followed by centrifugation at 5000 rpm. The supernatant obtained was the 1.0 mg mL⁻¹ solution. This was further diluted 10× to obtain 0.1 mg mL⁻¹ protein solution. For denaturation, a 3.0 mL portion of the 0.1 mg mL⁻¹ of enzyme/protein solution, prepared as discussed above, was kept in a water bath at 70 °C for α-amylase and BSA and at 80 °C for AMG for 15 min. The solution was brought to room temperature and the volume was made up for the loss due to evaporation. The actual protein content in the as-prepared 1.0 mg mL⁻¹ solution was calculated based on the standard Bradford test, which was reported in earlier studies²⁷ from the laboratory and the obtained values were used in the present study. The concentration of each protein expressed throughout the manuscript is thus the actual protein content.

Preparation of TGA-stabilized Au NPs

600.0 μL of 1.7262 × 10⁻² M HAuCl₄ (Sigma- Aldrich Chemical Co.) was added to 50.0 mL of MilliQ grade water and was kept in ice-water bath with constant magnetic stirring. To this solution 1.2 mL of 20.0 mM freshly prepared NaBH₄ (Merck specialties private limited, Mumbai) was added drop wise and with continued stirring for 2 h, after which the solution turned red, indicating formation of Au NPs. To the above solution 4.0 mL of 14.0 mM NaOH containing 120.0 μL of thioglycolic acid (Fluka) was added and stirred for another 1 h, which led to the formation of purple precipitates. The solution along with the precipitate thus formed was centrifuged at 10 000 rpm for 20 min and washed twice with ethanol (followed by centrifugation) to remove excess TGA. The pellet was then redispersed in 8.0 mL of phosphate buffer (pH = 7.0) for further experiments.

UV-vis extinction spectra of citrate-stabilized Au NPs in the presence of proteins

The citrate stabilized Au NPs solution as prepared was diluted (in phosphate buffer of pH 7.0) 2× so that the final absorbance of the solution was ∼1.0 and the final pH was 7.0. 3.0 mL of the diluted Au NP solution was taken in a plastic disposable cuvette and its UV-vis spectrum was recorded (using a Hitachi U-2900 spectrophotometer). 50.0 μL of 0.1 mg mL⁻¹ native protein solution was added to the Au NP, mixed well and left for 5 min followed by another wavelength scan. For preparing different mole fractions of the protein, definite amounts of native and denatured protein solutions were mixed together. 50.0 μL of the mixture thus obtained was added to 3.0 mL of Au NP solution as described before and the UV-vis spectrum was recorded before and after addition of the protein. It may be mentioned here that in order to avoid sticking of the protein stabilized Au NPs to the walls of the cuvette, a new cuvette was used for recording of each spectrum. The ratio of the peak area of UV-vis spectrum of Au NP plus protein to that of Au NP only was calculated and plotted against the respective mole fraction of the protein.

Calculation of the area under the UV-vis spectrum

This was performed using the software associated with the data acquisition of the spectrophotometer. The area was calculated by selecting two wavelengths as end points in the absorption spectrum. The area under the graph (in between the selected points) was taken as the area value. For all the proteins, the extreme wavelengths were set at 445 and 650 nm. A typical view of such area is shown in the ESI (Fig. S11†).

Temperature dependent denaturation study of proteins

Different test tubes with 1.0 mL solution of native α-amylase were taken and heated in a water bath at 60 °C. The test tubes were taken out at various time intervals and cooled quickly in a room temperature water bath, to have protein solutions denatured to different extents. 50.0 μL of each of this solution (as well as that of native solution) was added to 3.0 mL of citrate-stabilized Au NPs incubated for 5 min and the resultant UV-vis spectrum was recorded. The ratio of the peak area for citrate-stabilized Au NPs with protein to that of Au NP only were calculated by following the same method as already described before in the Experimental section. These area ratios were plotted against the time of denaturation. Untreated protein solution corresponds to 0 min of denaturation. This was done at 65 °C and 70 °C also. Similar experiments were performed for BSA (at 50 °C, 60 °C and 70 °C) and AMG (at 70 °C, 80 °C and 90 °C).

Calculation of mole fraction of native protein in the solutions denatured for different time intervals (at a particular temperature)

To calculate the exact mole fraction corresponding to each value of area/area of Au NP denoted in Fig. 5 for α-amylase, first of all the observed area/area of Au NP values were normalized (by multiplying by a common factor) so that the value at time equal to 0 min is same as that corresponding to mole fraction equal to 1 in Fig. 4(A). The normalized values thus obtained (say y′) were used in the following equation to calculate the corresponding mole fractions: y′ = mx′ + c, where m is the slope and c the intercept of the linear fit in Fig. 4(A). The x′ values are the required mole fraction values and were plotted against time in min (of denaturation) as shown in Fig. 5. Similarly calculations were done for BSA and AMG.

Sample preparation for TEM analysis

Solutions containing citrate-stabilized Au NPs and those of Au NPs in the presence of 50.0 μL each of native, denatured and mixture of 50 : 50 native–denatured were drop cast onto Cu grids (5 min after protein addition and UV-vis analyses) and then left for air drying overnight. These grids were further analyzed by a Jeol 2100 TEM machine (operated at a maximum voltage of 200 kV).

Fluorescence studies of proteins

50.0 μL of 0.1 mg mL⁻¹ native α-amylase was added to 3.0 mL of phosphate buffer, and the emission spectrum was recorded after setting the excitation wavelength at 280 nm (using a Perkin Elmer LS 55 Fluorescence Spectrometer). This was compared with that of 50.0 μL of 0.1 mg mL⁻¹ native α-amylase in the presence of citrate-stabilized Au NPs (of dilution as used in regular experiments) after 5 min incubation. Similarly, fluorescence emission spectra for same amount of denatured α-amylase and a mixture of native and denatured α-amylase (50 : 50) were recorded in presence and absence of citrate-stabilized Au NPs. Also, fluorescence spectra of native, denatured and mixed BSA and AMG were recorded in a similar fashion.

Acknowledgements

We thank the Department of Science and Technology (SR/S5/NM-01/2005, 2/2/2005-S.F.), the Department of Biotechnology and the Council of Scientific and Industrial Research (01(2172)/07/EMR-II), India for funds. J. D. acknowledges the fellowship from CSIR (09/731(67)/2008-EMR-I).

References

1. http://www.rcsb.org/ .
2. M. Lundqvist, I. Sethson and B. H. Jonsson, Langmuir, 2004, 20(24), 10639.
3. A. M. Tsai, J. H. van Zanten and M. J. Betenbaugh, Biotechnol. Bioeng., 1998, 59(3), 273.
4. X. Wu and G. Narsimhan, Langmuir, 2008, 24(9), 4989.
5. A. Filosa, Y. Wang, A. A. Ismail and A. M. English, Biochemistry, 2001, 40(28), 8256.
6. G. Anderle and R. Mendelsohn, Biophys. J., 1987, 52, 69.
7. J. M. Sanchez-Ruiz, J. L. Lopez-Lacomba, M. Cortijo and P. L. Mateo, Biochemistry, 1988, 27(5), 1648.
8. J. H. Teichroeb, J. A. Forresta and L. W. Jones, Eur. Phys. J. E, 2008, 26, 411.
9. K. Takeda, A. Wada, K. Yamamoto, Y. Moriyama and K. Aoki, J. Protein Chem., 1989, 8(5), 653.
10. M. S. Celej, G. G. Montich and G. D. Fidelio, Protein Sci., 2003, 12, 1496.
11. B. E. Cohen, A. Pralle, X. Yao, G. Swaminath, C. S. Gandhi, Y. N. Jan, B. K. Kobilka, E. Y. Isacoff and L. Y. Jan, Proc. Natl. Acad. Sci. U. S. A., 2005, 102(4), 965.
12. T. Mannen, S. Yamaguchi, J. Honda, S. Sugimoto, A. Kitayama and T. Nagamune, Anal. Biochem., 2001, 293, 185.
13. S. Chah, C. V. Kumar, M. R. Hammond and R. N. Zare, Anal. Chem., 2004, 76(7), 2112.
14. S. Chah, M. R. Hammond and R. N. Zare, Chem. Biol., 2005, 12, 323.
15. J. H. Teichroeb, J. A. Forresta, V. Ngai and L. W. Jones, Eur. Phys. J. E, 2006, 21, 19.
16. F. Wang, J. Wang, X. Liu and S. Dong, Talanta, 2008, 77, 628.
17. Y. M. Chen, C. J. Yu and W. L. Tseng, Langmuir, 2008, 24(7), 3654.
18. D. Casanova, D. Giaume, M. Moreau, J. L. Martin, T. Gacoin, J. P. Boilot and A. Alexandrou, J. Am. Chem. Soc., 2007, 129(42), 12592.
19. L. R. Hirsch, J. B. Jackson, A. Lee, N. J. Halas and J. L. West, Anal. Chem., 2003, 75(10), 2377.
20. C. P. Soman and T. D. Giorgio, Langmuir, 2008, 24(8), 4399.
21. K. A. Mahmoud, S. Hrapovic and J. H. T. Luong, ACS Nano, 2008, 2(5), 1051.
22. R. Elghanian, J. J. Storhoff, R. C. Mucic, R. L. Letsinger and C. A. Mirkin, Science, 1997, 277, 1078.
23. J. Wang, G. Liu and A. Merkoci, J. Am. Chem. Soc., 2003, 125, 3214.
24. H. Li and L. Rothberg, Proc. Natl. Acad. Sci. U. S. A., 2004, 101(39), 14036.
25. S. J. Park, T. A. Taton and C. A. Mirkin, Science, 2002, 295, 1503.
26. J. J. Storhoff, R. Elghanian, R. C. Mucic, C. A. Mirkin and R. L. Letsinger, J. Am. Chem. Soc., 1998, 120, 1959.
27. J. Deka, A. Paul and A. Chattopadhyay, J. Phys. Chem. C, 2009, 113(17), 6936.
28. M. Iosin, F. Toderas, P. L. Baldeck and S. Astilean, J. Mol. Struct., 2009, 924–926, 196.
29. S. H. Brewer, W. R. Glomm, M. C. Johnson, M. K. Knag and S. Franzen, Langmuir, 2005, 21(20), 9303.
30. P. Mulvaney, Langmuir, 1996, 12(3), 788.
31. M. Violet and J. C. Meunier, Biochem. J, 1989, 263, 665.
32. R. Lavecchia and A. Zuorro, Chem. Lett., 2010, 39, 38.
33. G. M. Zanin and F. F. De Moraes, Appl. Biochem. Biotechnol., 1998, 70–72, 383.

---

Footnote

† Electronic supplementary information (ESI) available: Additional UV-vis and fluorescence spectra and graphs based on UV-vis studies. See DOI: 10.1039/c0nr00154f

---