Effects of Arginine and Other Solution Additives on the Self

ACS2GO © 2018. ← → → ←. loading. To add this web app to the home screen open the browser option menu and tap on Add to homescreen...
0 downloads 0 Views 948KB Size
Download PDF
ARTICLE pubs.acs.org/Langmuir

Effects of Arginine and Other Solution Additives on the Self-Association of Different Surfactants: An Investigation at Single-Molecule Resolution Shubhasis Haldar and Krishnananda Chattopadhyay* Structural Biology and Bioinformatics Division, Indian Institute of Chemical Biology, Council of Scientific and Industrial Research, 4 Raja S. C. Mullick Road, Kolkata 700032, India

bS Supporting Information ABSTRACT: Fluorescence correlation spectroscopy is used to monitor the self-association of SDS and DTAB monomers at single-molecule resolution. Tetramethylrhodamine-5-maleimide (TMR) has been chosen as a probe because rhodamine dyes have been shown to bind surfactant micelles. Correlation functions obtained by FCS experiments have been fit using conventional discrete diffusional component analysis as well as the more recent maximum entropy method (MEM). Hydrodynamic radii calculated from the diffusion time values increase with surfactant concentration as the monomers self-associate. Effects of several solution additives on the self-association property of the surfactants have been studied. Urea and glycerol inhibit self-association, and arginine shows a dual nature. With SDS, arginine favors self-association, and with DTAB, it inhibits micelle formation. We propose surfactant self-association to be a “supersimplified” model of protein aggregation.

’ INTRODUCTION The self-association of biological molecules has been studied extensively because of their relevance to a number of biological questions. For example, the self-association of proteins may lead to the formation of amyloids, which has been implicated in several human diseases. Additionally, the self-association of therapeutic proteins could lead to immunogeneicity and other undesirable complications in their manufacturing and formulation development. The self-association of surfactants is considered to be important because they offer a suitable model system for biological membranes.15 Encapsulation by surfactants has been used as an efficient method of drug delivery.6 The selfassociation of surfactants leads to the formation of micelles and other aggregated forms that may play crucial roles in their interaction with proteins or other biomolecules.711 We have been studying protein self-association using fluorescence correlation spectroscopy (FCS) and other biophysical methods. Fluorescence correlation spectroscopy (FCS) is an important single-molecule technique in the study of the diffusional and conformational properties of labeled molecules.1220 In an FCS experiment, the sample is kept in a small observation volume at thermodynamic equilibrium. Fluorescence fluctuations, which occur because of the molecular diffusion in and out of the observation volume, are analyzed by measuring the correlation functions providing a measure of the hydrodynamic radius (rH). FCS can be used in combination with other fluorescence techniques to study r 2011 American Chemical Society

chemical kinetics and dynamics on the microsecond timescale.13,21,22 Using FCS, we have shown recently that arginine, an amino acid, can be used to inhibit protein self-association.23 We have shown further by measuring the microsecond conformational dynamics using FCS that arginine inhibits the formation of misfolded intermediate states in the unfolding transition of proteins.18 Our data and other results available in the literature suggest that the ability of arginine to suppress protein self-association may arise from its interaction with the side chains of the native and nativelike states of proteins.2427 This property would be in contrast with other protein stabilizers and osmolites that interact only with the unfolded state of proteins without affecting their native states.18 In this article, we have carried out FCS measurements to study the effect of different stabilizers, including arginine, on the selfassociation of two surfactants with the same chain length and opposite charge. We show that the use of a high concentration of urea or glycerol affects the process of self-association presumably by affecting the interaction of water with the surfactant molecules. We observe that self-association is favorable in the presence of salt. The behavior of arginine, however, depends on the nature of the surfactants. Arginine favors the self-association of SDS, a negatively Received: February 14, 2011 Revised: March 28, 2011 Published: April 13, 2011 5842

dx.doi.org/10.1021/la200580z | Langmuir 2011, 27, 5842–5849

Langmuir

ARTICLE

charged surfactant. However, with DTAB, a surfactant of identical chain length but opposite charge, arginine inhibits micellization.

’ MATERIALS AND METHODS Sample and Reagents. Teramethylrhodamine-5-maleimide (TMR) was obtained from Molecular Probes (Eugene, OR). SDS, DTAB, urea, and NaCl were purchased from Sigma-Aldrich in the highest available grade. FCS Experiments with TMR. FCS experiments were carried out using a commercial Confocor 3 LSM system (Carl Zeiss, Evotec, Jena, Germany), with a 40 water-immersion objective. Approximately 350 μL of each sample (containing 50 nM TMR) was placed into a Nunc chamber (NalgeNunc) in the presence of different concentrations of SDS. All of the experiments were carried out at room temperature. The samples were excited with an argon laser at 514 nm. The fluorescence signal was separated from the excited line using a main dichroic filter and collected using a pair of Avalanche photodiodes (APDs). The counts detected by the APDs were used by the correlator to calculate the correlation function. FCS experiments were carried out in 20 mM sodium phosphate buffer at pH 7.4. To correct for the refractive index and viscosity of the urea, glycerol, sucrose, and arginine solutions, necessary correction measures were taken using a microscope correction collar and the height as described previously.14 Briefly, the objective correction collar settings and the height are adjusted to obtain an optimized focal volume that maximizes the counts per particle and minimizes the observed diffusion time (τD) and number of particles (N). (See below for the definitions of τD and N.) The value of τD obtained for a particular solution condition was further normalized with respect to the value of τD observed in water to exclude the contributions of the refractive index and viscosity completely. Analysis of the Correlation Functions. For a system of one diffusing species without any conformational events, the diffusion time (τD) of a fluorophore and the number of particles in the observation volume (N) can be calculated by fitting the correlation function (G(τ)) to eq 1 22 0 10 10:5 GðτÞ ¼ 1 þ

C 1 n B 1 C 1 CB B C ai B N i ¼ 1 @1 þ τ A@1 þ SP2 τ A τDi τ Di

∑

ð1Þ

where SP is the structure parameter, which is the depth to diameter ratio of the Gaussian observation volume. For a single-component system, n = 1, which implicates the presence of only one kind of species with a diffusion time of τD. For a two-component system, n = 2, leading to the presence of two different kinds of particles (one particle with a diffusion time of τD1 and an amplitude of a1 and the second particle with a diffusion time of τD2 and an amplitude of a2, for example). The diffusion coefficient (Di) of the molecules can be calculated from τDi using eq 2 τ Di ¼

ω2 4Di

ð2Þ

where ω is the beam radius of the observation volume, which can be obtained by measuring the diffusion time of a dye with a known diffusion coefficient (D). The hydrodynamic radius (rH) of a labeled molecule can be calculated from D using the StokesEinstein equation (eq 3)21 D¼

kT 6πηrH

ð3Þ

where k is the Boltzmann constant, T is the temperature, and η corresponds to the viscosity of the solution. Data Analysis Using the Maximum Entropy Method (MEM). The correlation function data was further analyzed by the maximum entropy

method (MEM) recently applied to FCS using the MEMFCS algorithm.28 In this method, the multicomponent correlation function can be represented by n, the number of noninteracting fluorescent species, each of which can have a diffusion time of 0.001 to 500 ms. MEMFCS minimizes the parameter χ2 and also maximizes the entropic quantity S = Σipi ln pi (where pi = Ri∑j Rj) to obtain an optimized fit.23,28

’ RESULTS Two common surfactants, namely, sodium dodecyl sulfonate (SDS) and dodecyl trimethyl ammonium bromide (DTAB), are chosen for the present study. We chose SDS because its selfassociation properties have been extensively studied. As a model ionic surfactant, SDS is also widely used to study the folding of different proteins.29,30 SDS has been shown to stabilize partially folded intermediate states whose characterization by NMR and other spectroscopic techniques provides useful information about the thermodynamics and kinetics of protein folding.31 SDS has been shown to refold proteins under strongly denaturing conditions and even in the presence of 10 M urea.31 Although the exact mechanism of SDS-induced refolding is still not known, the surfactants have been found to interact with proteins using ionic and hydrophobic interactions. Although significant refolding of the secondary structure of cytochrome c has been observed with SDS, a charged surfactant, the extent of refolding is much less with lauryl maltoside, a neutral surfactant with an identical chain length.31 We monitor the diffusion of a fluorescent dye molecule, tetramethylrhodamine-5-maleimide (TMR), in the presence of different concentrations of surfactants. It has been assumed that the interaction of TMR with the surfactant monomer can be differentiated from its interaction with surfactant self-associated molecules by the hydrodynamic radii (rH) measurements. Rhodamine fluorophores have been shown to bind SDS.32 Fluorescence Correlation Spectroscopy Measurements with TMR in the Presence of Different Concentration of SDS. Figure 1a shows representative correlation functions ob-

tained by FCS experiments in the absence and presence of different concentration of SDS at pH 7.4. The correlation functions have been fit to eq 1 using a single-component system to obtain the values of the diffusion time (τD). Using an equation containing two diffusing components (the first component with a diffusion time of τD1 and an amplitude of a1 and the second component with a diffusion time of τD2 and an amplitude of 1-a1) does not improve the fit. The goodness of the fit in each case has been established by the randomness of the residual distributions. As an additional test, the correlation function data have been analyzed further using MEM (Figure 1b). Because MEM is a model-independent method, it provides a bias-free solution to complement the conventional method of analyzing the correlation function data. MEM analyses also show the presence of only one diffusion component in the presence of different concentration of SDS (Figure 1b). Moreover, the use of one diffusion component to fit the correlation functions obtained for the surfactant micellization have been reported by others.43,46 The absence of the contribution of any free label in the correlation function could be due to the use of a significantly smaller concentration of the label (50 nM TMR) in the FCS experiments compared to the concentration of the surfactants used (in the millimolar range). The values of τD are used to calculate the diffusion coefficients (D) and hydrodynamic radii (rH) of the surfactant molecules. In the absence of SDS, the value of τD has been observed to be 31 μs when fit using the conventional method. τD increases to 53 and 5843

dx.doi.org/10.1021/la200580z |Langmuir 2011, 27, 5842–5849

Langmuir

ARTICLE

Figure 2. Urea dependence of the self-association of SDS at pH 7.4: (a) MEM profiles obtained with 0, 3, and 10 mM SDS in the absence (bottom panel) and presence of 3 M (middle panel) and 5 M urea (top panel). (b) SDS concentration dependence of rH in the absence and presence of 3 and 5 M urea. Experiments were carried out in the presence of 20 mM phosphate buffer at pH 7.4.

Figure 1. (a) Normalized correlation functions obtained with TMR in the absence and presence of 3 and 10 mM SDS at pH 7.4. With increasing SDS concentration, the correlation plots shift to the right, suggesting an increase in the diffusion time (τD). The data were fit to eq 1 using a model containing one diffusing component. (b) Profiles of the MEM analyses of the correlation function data in the absence and presence of 3 and 10 mM SDS. (c) Dependence of rH and counts per particle on the SDS concentration. The left axis represents the rH data, and the right axis shows the data for the counts per particle. FCS experiments are carried out at room temperature using 20 mM sodium phosphate buffer at pH 7.4.

110 μs in the presence of 3 and 10 mM SDS, respectively. In the absence of SDS, the MEM fit of the data shows the maximum at 25 μs, which is similar to the value of τD (31 μs) obtained by conventional analysis. The presence of 3 and 10 mM SDS shifts the maxima to higher values (43 and 93 μs, respectively; Figure 1b). Figure 1c shows the variation of rH and the counts per particle with SDS concentration. Both observables follow sigmoidal profiles. The values of rH increase with SDS concentration as self-association increases the size of the molecules. Because the

self-association of monomers leading to large particles decreases the number of monomers (and hence the number of particles) for a particular count rate, the counts per particle increase with self-association as observed in the present data (Figure 1c). Interestingly, the increase in rH begins at 2.5 mM SDS, which is less than the first critical micellar concentration (cmc1) of SDS under this solution condition (8 mM).33 There could be several reasons behind this early onset of the increase in rH with SDS concentration. First, there may be adduct formation between TMR and negatively charged SDS monomers occurring below the cmc as reported earlier.33 Second, the surfactant monomers may self-associate to form small surfactant aggregates at less than the reported value of the cmc as shown by Zettl. et al.34 To obtain further insight into this, we have carried out FCS experiments with DTAB using TMR. Interestingly, the increase in rH with DTAB concentration starts at 12 mM (see below), which is similar to the reported cmc of DTAB.35 The present FCS experiments hence indicate a concentration of self-association (Csa) where Csa is similar to the cmc of DTAB but is significantly less than the cmc of SDS. Effect of Urea Concentration on SDS Self-Association. Figure 2a shows MEM profiles of TMR at 0, 3, and 10 mM SDS in the absence and presence of different concentrations of urea at pH 7.4. In the absence of urea, MEM profiles are well separated, with maxima at 25, 43, and 93 μs for 0, 3, and 10 mM SDS, respectively (Figure 2a, bottom panel). In the presence of 3 M urea, their separation decreases significantly and profiles 5844

dx.doi.org/10.1021/la200580z |Langmuir 2011, 27, 5842–5849

Langmuir

ARTICLE

Figure 4. Effect of urea on the self-association of SDS at pH 2 and 11 as monitored by rH. The presence of urea inhibits the formation of aggregates at both pH values.

Figure 3. pH dependence of the self-association of SDS: (a) MEM profiles of 0, 3, and 10 mM SDS at pH 7.4 (bottom panel), pH 2 (middle panel), and pH 11 (top panel). (b) SDS concentration dependence of rH at pH 7.4, 2, and 11. At pH 2, self-association begins at low surfactant concentration, but at pH 7.4 and 11, aggregation begins at high SDS concentration.

overlap with each other (Figure 2a, middle panel). As a result, the maxima of MEM profiles do not show any major SDS concentration variations (27, 27, and 49 μs in the presence of 0, 3, and 10 mM, respectively). In the presence of 5 M urea, their separation decreases even further, resulting in significant overlaps between MEM profiles (Figure 2a, top panel). As discussed earlier, the values of τD (obtained from conventional or MEM analysis) have been used to determine rH. Figure 2b shows the variation of rH with SDS concentration in the presence of different concentrations of urea. In the absence of urea (filled squares), rH increases drastically as the self-association of the surfactant monomers begins. The presence of urea (open squares for 3 M and filled circles for 5 M urea) hinders selfassociation, resulting in an increase in the concentration at which self-association begins (Figure 2b). Effect of pH on SDS Self-Association. Figure 3a shows MEM profiles of 0, 3, and 10 mM SDS at pH 2 and 11. MEM profiles at pH 7.4 are shown as a comparison. MEM profiles are similar at pH 7.4 and 11. At low pH, however, the profile for 3 mM SDS is significantly shifted. Also, at pH 2, MEM profiles for 3 and 10 mM SDS are superimposable. Figure 3b shows the variation of rH with SDS concentration at pH 2 and 11; the data at pH 7.4 is shown as a comparison. Compared to that at pH 7.4, self-association at pH 2 occurs at a very low SDS concentration. The transition of self-association is

Figure 5. Effect of solution additives on MEM profiles in the absence and presence of 3 and 10 mM SDS. Experiments were performed in 20 mM phosphate buffer at pH 7.4. The names of the additives are shown.

considerably sharper at pH 2 with a high degree of cooperativity. The concentration of self-association is slightly higher at pH 11 than at pH 7.5. We have studied the effect of urea on the self-association of SDS at pH 2 and 11. Figure 4 shows the variation of rH with SDS concentration at pH 2 and 11 in the presence of different urea concentrations. As observed at pH 7.4, the use of urea at pH 2 and 11 increases the concentration of SDS needed to initiate selfassociation. Effect of Solution Additives on SDS Self-Association. We have carried out FCS experiments with TMR in the presence of different solution additives (e.g., glycerol, sucrose, and arginine) to study their effects on the self-association of SDS. MEM profiles with different additive solutions in the absence and presence of 3 and 10 mM SDS are shown in Figure 5. MEM profiles with sucrose show less dispersion between 0 and 3 M SDS. With glycerol, MEM profiles do not show any dispersion and the data at 0, 3, and 10 mM SDS overlap. With arginine, MEM profiles show a large shift between 0 and 3 mM SDS. No change in MEM profiles is observed between 3 and 10 mM SDS. To gain better insight into the effects of different solution additives, rH values are plotted with SDS concentrations in the 5845

dx.doi.org/10.1021/la200580z |Langmuir 2011, 27, 5842–5849

Langmuir

ARTICLE

Figure 7. The midpoint of self-association calculated from the dependence of hydrodynamic radii with SDS concentration data is plotted with the concentration of each additive. The midpoints are calculated using sigmoidal fits to the data. The top axis represents the concentration values for urea and glycerol and the bottom axis corresponds to that for the rest of additives.

Figure 8. FCS experiments on the self-association of DTAB. The values of rH are plotted with the concentration of DTAB in the presence of different solution additives. Figure 6. Variation of rH with SDS concentration in the presence of different additives: (a) the effect of sucrose; (b) the effect of glycerol; and (c) the effect of 500 mM arginine and salt. The concentration of each additive is mentioned in its respective plot.

presence of different concentrations of each additive (Figure 6). The data at Figure 6a suggest that self-association occurs at higher SDS concentrations in the presence of sucrose. With glycerol, the process of self-association is slow and gradual (Figure 6b), which is supported by the MEM profile data in Figure 5. The presence of glycerol leads to the complete absence of dispersion in the MEM profiles at different concentrations of SDS (Figure 5). The addition of 500 mM arginine, in contrast, results in an earlier onset of the formation of self-associated species (Figure 6c), which is also shown in the MEM profiles (Figure 5). The midpoint of the transition is calculated for each of the solution additives using a sigmoidal fit and plotted with the molar concentration of the additive (Figure 7). The lines drawn through

the data in Figure 7 are generated using third-order polynomial fits. These lines should be used for representation purposes only and to understand the trend. It is evident from Figure 7 that, in the case of glycerol and urea, the midpoint of the self-association increases with the concentration of the additives, indicating that they inhibit the self-association of SDS. In the case of NaCl and arginine, however, the trend is reversed and these two additives facilitate the self-association of SDS. According to the data presented here, the propensity of SDS self-association follows the order arginine ≈ NaCl > pH 2 > phosphate buffer, pH 7.4 ≈ sucrose > pH 11 > urea > glycerol. Effect of Solution Additives on the Self-Association of DTAB. FCS experiments with TMR in the presence of DTAB, a positive surfactant with the same chain length as SDS, yield a concentration of self-association (Csa) of 12 mM, which is similar to the reported cmc.35 We have carried out FCS experiments in the presence of different solution additives in order to understand the effects of these additives on the self-association of DTAB. Figure 8 shows rH values of different concentrations of 5846

dx.doi.org/10.1021/la200580z |Langmuir 2011, 27, 5842–5849

Langmuir DTAB in the presence of 5 M urea, 500 mM sucrose, 20% glycerol, and 500 mM arginine. Representative MEM profiles of different concentrations of SDS in the presence of additives are shown in Figure 1S (Supporting Information). As observed before in the case of SDS, the presence of urea, sucrose, and glycerol increases the Csa of DTAB molecules. Similarly, NaCl decreases the Csa, suggesting that the presence of salt favors selfassociation. Interestingly, the effect of arginine has been found to be exactly opposite to its behavior toward SDS self-association. Although arginine favors the self-association of SDS, the presence of it inhibits that of DTAB (Figure 8).

’ DISCUSSION The self-association of surfactants has been studied in great detail using multiple techniques including conductivity, light scattering, and rheology measurements.3638 Steady-state and time-resolved fluorescence studies using suitable fluorophores have also been performed to explore the solution properties of surfactants and their association behaviors.3942 Although the use of single-molecule techniques including FCS is becoming popular in the study of different aspects of nucleic acid and protein chemistry, their use in the field of surfactants has been relatively uncommon.34,43,44 FCS data are generally analyzed by using models of a number of discrete diffusing components. Because the individual components and their amplitudes can influence each other, particularly in the case of multicomponent systems, this method can be quite confusing and potentially inadequate for analyzing heterogeneous systems. We have used MEM, which has recently been developed by Sengupta et al. and used extensively to study protein aggregation and other systems.28,45 MEM provides a bias-free method of fitting FCS data using the continuous distribution of a large number of diffusing components. The FCS study presented in this article shows an increase in rH that begins at around 2.5 mM SDS. This concentration is significantly less than the reported cmc value of SDS in water (8 mM). This apparent discrepancy may arise because FCS is sensitive enough to detect the formation of relatively small particles in solution such as the formation of adducts between SDS and TMR. Assuming rH of TMR (molecular weight 481) to be approximately 7.5 Å, the calculated rH of a TMR-SDS (TMRbound SDS monomer) molecule can be calculated to be 8.9 Å, which is remarkably similar to the value of rH observed (8.8 Å) in the present study at 1 mM SDS. Likewise, the formation of small complexes of three and six molecules of TMR-bound SDS would yield rH values of approximately 13 and 16 Å, and identical values are observed at 3 and 3.5 mM SDS. Short complexes in such a size range would be difficult to detect by techniques traditionally used to measure the cmc values of a surfactant molecule. Experiments are underway in our laboratory to detect these small adducts using high-resolution transmission electron microscopy. The ability of FCS to detect and monitor these early events makes it an important technique. It would be very exciting if the singlemolecule resolution of FCS could be used to monitor the lag phase of protein self-association. Although the presence of conformationally altered intermediates would add to the complexity in the initial stage of the protein aggregation pathway, FCS has been shown to monitor subtle changes in the conformational orientation of a number of proteins.13,14,18,23 Although FCS offers high sensitivity and single-molecule resolution for studying surfactant micellization, there are several

ARTICLE

important issues that need careful consideration. Because micelles are dynamic, it is expected that the diffusion coefficient measured by FCS may depend on the relative binding affinity of the used fluorophore (here TMR) to the surfactant micelles compared to that of the surfactant monomer. Because this relative binding affinity would almost certainly change under different solution conditions (e.g., at low pH or in the presence of glycerol or urea), the value of Csa obtained by FCS may be influenced significantly. This effect can be overcome using a fluorophore that is covalently attached to a monomer molecule. To determine if the binding affinity between TMR and the surfactants has any significant influence on the measured Csa values, we have performed separate dynamic light scattering measurements with SDS at pH 7.4 in the presence of several solution additives. The values of the count rate at different SDS concentrations (representative plots are shown in Figure 2S, Supporting Information) are used to calculate the cmc of SDS. Although the cmc values determined by the count rate data have been found to be different from Csa, their relative trend has been found to be similar to that observed by the FCS data. Second, MEM assumes the diffusing components to be noninteracting, but this assumption may not hold in the case of highly interacting micellar systems. It is important to analyze the FCS data obtained with a heterogeneous and dynamic system using different methods and models to rule out any possible fitting artifacts. Finally, care should be taken to exclude the contributions of the refractive index and viscosity, which not only could lead to an artificial increase in the diffusion time but the refractive index mismatch could potentially lead to significant focal volume aberrations. A comparison between the protein and surfactant selfassociation behavior could be interesting. Protein self-association has been studied in detail because of its relevance in several human diseases and also because of its critical role in the formulation development of biotechnologically derived drugs or “biologics”.4750 Research in our laboratory emphasizes the role of the stabilizers or excipients such as arginine to influence the conformational properties of the native, nativelike, and unfolded states of a protein.18,23 The self-association of a surfactant may be considered to be a “supersimplified” model of the self-association of a protein excluding the contribution of the native or nativelike states. The self-association of a protein is an equilibrium composed of at least three interacting units: the native state, the unfolded states, and the self-associated states. The presence of native or un-folded-like intermediate states with altered conformational properties also adds to the complexity. The exclusion of the side chains in a surfactant system (and conformational heterogeneity) excludes the presence of the native state, resulting in a two-state equilibrium between the unfolded state and the self-associated state. Although research in protein folding had traditionally focused on the native state, there is growing recent interest to study the unfolded state of a protein.5157 Another significant difference between these two kinds of self-association comes from the degree of reversibility. Surfactant self-association has been found to be reversible in the presence of all of the additives studied. (Figure 3S in the Supporting Information shows one representative example in phosphate buffer at pH 7.4.) The reversibility of protein self-association, however, depends on the nature of the proteins, the solution conditions, and the nature of the additives present in the solution (unpublished data). The self-association of an ionic surfactant (such as SDS and DTAB used in this study) can be governed by two opposing 5847

dx.doi.org/10.1021/la200580z |Langmuir 2011, 27, 5842–5849

Langmuir physicochemical interactions. The hydrophobicity of the surfactants would favor their association, but the headgroup charges would repel each other. The use of low pH is expected to neutralize the negative charge of the SDS molecules, leading to an overall enhancement of the favorable hydrophobic interaction that is consistent with the present results at pH 2. An apparent similarity occurs in the case of proteins because many proteins have been shown to exhibit an enhanced aggregation propensity at low pH because of the enhanced hydrophobic interactions. A high pH, such as pH 11 used in this study, should have no effect on SDS self-association because the overall charge and hydrophobicity of SDS molecules remain identical to these values at pH 7.4, which is also observed in this study. The observed effect of salt in increasing the affinity of the self-association of SDS and DTAB can be explained by the screening of the charges of the surfactant headgroups, resulting in more favorable hydrophobic interactions. The effect of urea, glycerol, and other small additive molecules on the self-association of ionic and nonionic surfactants has been studied.58 The addition of urea to increase the cmc of surfactants has been reported earlier, and it has been proposed that urea weakens the hydrophobic interactions in aqueous solution.59 Two different mechanisms have been regularly used in the literature to explain the effect of urea on surfactant self-association. The indirect mechanism considers urea to be a “water breaker”, and the direct mechanism suggests the replacement of water molecules from the hydration shell as induced by urea.59,60 A third mechanism has also been proposed that suggests that urea increases the water polarity, increasing the solvation of the polar groups.59,64 It has been shown recently that the use of a direct or indirect mechanism to understand urea action is inadequate in general terms; instead the actual mechanism depends on a particular surfactant system, depending on how the solvation occurs.59 For a protein system, it is generally believed that urea interacts with the side chains and backbone of a protein more tightly than water or another hydrophobic molecule, leading to its ability to dissolve the core hydrophobic region.61 Urea has been commonly used to break protein aggregates and to purify protein molecules from inclusion bodies. Cosolvents such as glycerols have been shown to have complex effects on self-association that cannot be explained by simple salting-in or salting-out effects.58 Glycerol is known for its effect to stabilize the native state of a globular protein. It is expected to act like an Osmolite, which interacts with the unfolded state of a protein and excludes itself from the surface of the protein. The use of glycerol and other small organic molecules, however, is shown to decrease the cohesive forces in water, which results in an increase in the solubility of surfactants.62 The end result would be an increase in the concentration at which self-association begins as observed in the case of glycerol used in this study. The use of ethanol, isopropanol, and glycerol as additives has been shown to disfavor micellization and decreases in the aggregation number.62 We have chosen arginine for this study for several reasons. Although arginine has been widely used as an efficient agent to inhibit protein aggregation,63 little or no information is available on its effect on the self-association of a surfactant. Although the structure of arginine is similar to that of guanidine, a molecule commonly used to denature proteins, arginine is commonly used as a protein stabilizer and as an inhibitor of protein self-association. It has been suggested that arginine may have dual effects: a proteinstabilizing osmolyte-like effect and a destabilizing denaturant-like

ARTICLE

effect.23,63 Its ability to interact with the protein side chains, particularly with the aromatic amino acids, has been proposed to be responsible for its behavior in inhibiting protein selfassociation.63 The present results suggest that in the case of a “supersimplified” model that does not have the presence of sidechain contributions, the effect of arginine depends solely on the charge of the surfactant systems. Arginine favors the self-association of negatively charged SDS, but it play an inhibitory role in DTAB self-association. DTAB is positively charged, and hence it may offer repulsive interactions with the positive guanidine group present in arginine.

’ CONCLUSIONS In this article, we have carried out a detailed study of the selfassociation of two surfactant systems, namely, SDS and DTAB using FCS. These two surfactants are chosen because of their same chain length and opposite charge. The data have been analyzed using conventional methods and MEM. MEM provides a model-free method of analyseis of any heterogeneous multicomponent system. Using FCS, we could follow the initial adduct formation between SDS and TMR that is not present in DTAB. The effects of different solution additives have also been investigated. Urea and glycerol are shown to inhibit surfactant selfassociation. Although arginine favors the self-association of SDS, it inhibits the self-association of DTAB. ’ ASSOCIATED CONTENT

bS

Supporting Information. Additional information consisting of the MEM profiles of the FCS experiments with DTAB and separate dynamic light scattering data and reversibility measurements. This material is available free of charge via the Internet at http://pubs.acs.org.

’ ACKNOWLEDGMENT This work has been funded by CSIR network project grant nwp 005. We thank Prof. Sudipta Maiti of the Tata Institute of Fundamental Research for providing us with the MEM analysis software used extensively to analyze our data. We thank Prof. Siddhartha Roy of the Indian Institute of Chemical Biology for his help and support. ’ REFERENCES (1) Helenius, A.; McCaslin, D. R.; Fries, E.; Tanford, C. Methods Enzymol. 1979, 56, 734–749. (2) Helenius, A.; Simons, K. Biochim. Biophys. Acta 1975, 415, 29–79. (3) Hjelmeland, L. M. Proc. Natl. Acad.Sci. U.S.A. 1980, 77, 6368–6370. (4) Lichtenberg, D.; Robson, R. J.; Dennis, E. A. Biochim. Biophys. Acta 1983, 737, 285–304. (5) Silvius, J. R. Annu. Rev. Biophys. Biomol. Struct. 1992, 21, 323–348. (6) Narang, A. S.; Delmarre, D.; Gao, D. Int. J. Pharm. 2007, 345, 9–25. (7) Chakraborty, T.; Chakraborty, I.; Moulik, S. P.; Ghosh, S. Langmuir 2009, 25, 3062–3074. (8) Lu, R. C.; Cao, A. N.; Lai, L. H.; Xiao, J. X. Colloids Surf., B 2008, 64, 98–103. (9) Wu, D.; Xu, G.; Sun, Y.; Zhang, H.; Mao, H.; Feng, Y. Biomacromolecules 2007, 8, 708–712. 5848

dx.doi.org/10.1021/la200580z |Langmuir 2011, 27, 5842–5849

Langmuir (10) De, S.; Girigoswami, A.; Das, S. J. Colloid Interface Sci. 2005, 285, 562–573. (11) Wintergerst, E.; Manz-Keinke, H.; Plattner, H.; Schlepper-Schafer, J. Eur. J. Cell Biol. 1989, 50, 291–298. (12) Berland, K. M. Methods Mol. Biol. 2004, 261, 383–398. (13) Chattopadhyay, K.; Saffarian, S.; Elson, E. L.; Frieden, C. Proc. Natl. Acad. Sci. U.S.A. 2002, 99, 14171–14176. (14) Chattopadhyay, K.; Saffarian, S.; Elson, E. L.; Frieden, C. Biophys. J. 2005, 88, 1413–1422. (15) Chen, H.; Rhoades, E.; Butler, J. S.; Loh, S. N.; Webb, W. W. Proc. Natl. Acad. Sci. U.S.A. 2007, 104, 10459–10464. (16) Digman, M. A.; Brown, C. M.; Sengupta, P.; Wiseman, P. W.; Horwitz, A. R.; Gratton, E. Biophys. J. 2005, 89, 1317–1327. (17) Elson, E. L. J. Biomed. Opt. 2004, 9, 857–864. (18) Haldar, S.; Mitra, S.; Chattopadhyay, K. J. Biol. Chem. 2010, 285, 25314–25323. (19) Schwille, P.; Haupts, U.; Maiti, S.; Webb, W. W. Biophys. J. 1999, 77, 2251–2265. (20) Schwille, P. Cell Biochem. Biophys. 2001, 34, 383–408. (21) Chattopadhyay, K.; Elson, E. L.; Frieden, C. Proc. Natl. Acad. Sci. U.S.A. 2005, 102, 2385–2389. (22) Haupts, U.; Maiti, S.; Schwille, P.; Webb, W. W. Proc. Natl. Acad. Sci. U.S.A. 1998, 95, 13573–13578. (23) Ghosh, R.; Sharma, S.; Chattopadhyay, K. Biochemistry 2009, 48, 1135–1143. (24) Hamada, H.; Arakawa, T.; Shiraki, K. Curr. Pharm. Biotechnol. 2009, 10, 400–407. (25) Arakawa, T.; Tsumoto, K.; Nagase, K.; Ejima, D. Protein Expression Purif. 2007, 54, 110–116. (26) Tsumoto, K.; Ejima, D.; Kita, Y.; Arakawa, T. Protein Pept. Lett. 2005, 12, 613–619. (27) Arakawa, T.; Tsumoto, K. Biochem. Biophys. Res. Commun. 2003, 304, 148–152. (28) Sengupta, P.; Garai, K.; Balaji, J.; Periasamy, N.; Maiti, S. Biophys. J. 2003, 84, 1977–1984. (29) Ferreon, A. C.; Deniz, A. A. Biochemistry 2007, 46, 4499–4509. (30) Martinez, E.; Jimenez, M. A.; Segui-Real, B.; Vandekerckhove, J.; Sandoval, I. V. J. Mol. Biol. 1997, 267, 1124–1138. (31) Chattopadhyay, K.; Mazumdar, S. Biochemistry 2003, 42, 14606–14613. (32) Haglund, B. O.; Sundelof, L.-O.; Upadrashta, S. M.; Wurster, D. E. J. Chem. Educ. 1996, 73, 889–893. (33) Romani, A. P.; Machado, A. E.; Hioka, N.; Severino, D.; Baptista, M. S.; Codognoto, L.; Rodrigues, M. R.; de Oliveira, H. P. J. Fluoresc. 2009, 19, 327–332. (34) Zettl, H.; Portnoy, Y.; Gottlieb, M.; Krausch, G. J. Phys. Chem. B 2005, 109, 13397–13401. (35) Pawel, W.; Maria, P. J. Colloid Interface Sci. 2005, 286, 387–391. (36) Rodriguez-Pulido, A.; Casado, A.; Munoz-Ubeda, M. F.; Junquera, E.; Aicart, E. Langmuir 2010, 26, 9378–9385. (37) Lundberg, D.; Shi, L.; Menger, F. M. Langmuir 2008, 24, 4530–4536.  lvarez, J. G.; Perez-Lopez, J. H.; (38) Fernandez, V. V. A.; Tepale, N.; A MacIas, E. R.; Bautista, F.; Pignon, F.; Rharbi, Y.; Gamez-Corrales, R.; Manero, O.; Puig, J. E.; Soltero, J. F. A. J. Colloid Interface Sci. 2009, 336, 842–849. (39) Dey, S.; Sasmal, D. K.; Das, D. K.; Bhattacharyya, K. ChemPhysChem 2008, 9, 2848–2855. (40) Unruh, J. R.; Kuczera, K.; Johnson, C. K. J. Phys. Chem. B 2009, 113, 14381–14392. (41) Mahata, A.; Sarkar, D.; Bose, D.; Ghosh, D.; Das, P.; Chattopadhyay, N. J. Colloid Interface Sci. 2009, 335, 234–241. (42) Chaudhuri, A.; Haldar, S.; Chattopadhyay, A. Biochem. Biophys. Res. Commun. 2009, 390, 728–732. (43) Yu, L.; Tan, M.; Ho, B.; Ding, J. L.; Wohland, T. Anal. Chim. Acta 2006, 556, 216–225. (44) Hink, M. A.; Hoek, A. V.; Visser, A. J. W. G. Langmuir 1999, 15, 992–997.

ARTICLE

(45) Sengupta, P.; Garai, K.; Sahoo, B.; Shi, Y.; Callaway, D. J.; Maiti, S. Biochemistry 2003, 42, 10506–10513. (46) Luschtinetz, F.; Dosche, C. J. Colloid Interface Sci. 2009, 338, 312–315. (47) Dobson, C. M. Trends Biochem. Sci. 1999, 24, 329–332. (48) 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.; Pepys, M. B. Nature 1997, 385, 787–793. (49) Andya, J. D.; Hsu, C. C.; Shire, S. J. AAPS Pharm. Sci. 2003, 5, 1–11. (50) Andya, J. D.; Maa, Y. F.; Costantino, H. R.; Nguyen, P. A.; Dasovich, N.; Sweeney, T. D.; Hsu, C. C.; Shire, S. J. Pharm. Res. 1999, 16, 350–358. (51) Hong, J.; Gierasch, L. M. J. Am. Chem. Soc. 2010, 132, 10445–10452. (52) Voelz, V. A.; Singh, V. R.; Wedemeyer, W. J.; Lapidus, L. J.; Pande, V. S. J. Am. Chem. Soc. 2010, 132, 4702–4709. (53) Mallam, A. L.; Rogers, J. M.; Jackson, S. E. Proc. Natl. Acad. Sci. U.S.A. 2010, 107, 8189–8194. (54) Ferreon, A. C.; Moran, C. R.; Gambin, Y.; Deniz, A. A. Methods Enzymol. 2010, 472, 179–204. (55) Uversky, V. N. J. Biomed. Biotechnol. 2010, 568068, 1–14. (56) Sahara, N.; Maeda, S.; Takashima, A. Curr. Alzheimer Res. 2008, 5, 591–598. (57) Sosnick, T. R.; Shtilerman, M. D.; Mayne, L.; Englander, S. W. Proc. Natl. Acad. Sci. U.S.A. 1997, 94, 8545–8550. (58) Patel, H.; Raval, G.; Nazari, M.; Heerklotz, H. Biophys. Chem. 2010, 150, 119–128. (59) Hierrezuelo, J. M.; Molina-Bolivar, J. A.; Carnero Ruiz, C. J. Phys. Chem. B 2009, 113, 7178–7187. (60) Raghuraman, H.; Pradhan, S. K.; Chattopadhyay, A. J. Phys. Chem. B 2004, 108, 2489–2496. (61) Zangi, R.; Zhou, R.; Berne, B. J. J. Am. Chem. Soc. 2009, 131, 1535–1541. (62) Sarkar, B.; Lam, S.; Alexandridis, P. Langmuir 2010, 26, 10532–10540. (63) Arakawa, T.; Ejima, D.; Tsumoto, K.; Obeyama, N.; Tanaka, Y.; Kita, Y.; Timasheff, S. N. Biophys. Chem. 2007, 127, 1–8. (64) Dias, L. G.; Florenzano, F. H.; Reed, W. F.; Baptista, M. S.; Souza, S. M. B.; Alvarez, E. B.; Chaimovich, H.; Cuccovia, I. M.; Amaral, C. L. C.; Brasil, C. R.; Romsted, L. S.; Politi., M. J. Langmuir 2002, 18, 319–324.

5849

dx.doi.org/10.1021/la200580z |Langmuir 2011, 27, 5842–5849