Specific Cation Effects on Hemoglobin Aggregation below and at

20 Nov 2013 - ACS Biomaterials Science & Engineering 2016 2 (5), 741-751. Abstract | Full ... Drew F. Parsons , Timothy T. Duignan , Andrea Salis. Int...
0 downloads 0 Views 2MB Size
Article pubs.acs.org/Langmuir

Specific Cation Effects on Hemoglobin Aggregation below and at Physiological Salt Concentration Luca Medda,† Cristina Carucci,† Drew F. Parsons,‡ Barry W. Ninham,‡ Maura Monduzzi,† and Andrea Salis*,†,‡ †

Department of Chemical and Geological Sciences, University of Cagliari-CSGI and CNBS, Cittadella Universitaria, S.S. 554 bivio Sestu, 09042 Monserrato (CA), Italy ‡ Department of Applied Mathematics, The Australian National University, Acton 0200 ACT, Australia S Supporting Information *

ABSTRACT: Turbidity titrations are used to study the ion specific aggregation of hemoglobin (Hb) below and physiological salt concentration in the pH range 4.5−9.5. At a salt concentration 50 mM cations promote Hb aggregation according to the order Rb+ > K+ ∼ Na+ > Cs+ > Li+. The cation series changes if concentration is increased, becoming K+ > Rb+ > Na+ > Li+ > Cs+ at 150 mM. We interpret the puzzling series by assuming that the kosmotropic Li+ will bind to kosmotropic carboxylates groupsaccording to the law of matching water affinities (LMWA)whereas the chaotropic Cs+ will bind to uncharged protein patches due to its high polarizability. In fact, both mechanisms can be rationalized by invoking previously neglected ionic nonelectrostatic forces. This explains both adsorption to uncharged patches and the LMWA as a consequence of the simultaneous action of electrostatic and dispersion forces. The same interpretation applies to anions (with chaotropic anions binding to chaotropic amine groups). The implications extend beyond hemoglobin to other, still unexplained, ion specific effects in biological systems.

1. INTRODUCTION Ion specific or “Hofmeister”1 phenomena are effects that are not accommodated by classical theories of electrolytes.2−4 The terminology puts the cart before the horse because ion specific effects are the rule rather than the exception. A plethora of old and recent works have shown the ubiquity of these phenomena in chemistry and biology. They include viscosities,5 bubble coalescence,6 polymer solubility,7−9 pH of buffer solutions,10 oxide surfaces,11,12 enzyme activities,13,14 protein cloud points,15,16 protein surface charges,17 electrophoretic mobilities,18−20 electrochemistry of redox enzymes,21 and many other examples which are impossible to list exhaustively. Among the many systems in which ion specific effects have been found to be important, protein systems, the subject of Hofmeister’s1 pioneering (and other early)22,23 experiments, remain of major interest. Naturally enough, because in living organisms proteins are called upon for a multitude of disparate functions. For example, blood proteins are essential for the transport of lipids, hormones, gases, and vitamins. They are also involved in the immune response, in blood clotting, in inflammatory processes, and in the regulation of various activities of the organism (i.e., enzymes). Protein water systems contain strong (e.g., Na+, Ca2+, K+, Cl−, etc.) and weak (e.g., phosphate, carbonate) electrolytes. In order to function, the chemical nature and physiological concentration of ions have to be strictly prescribed. Each ion has its own biochemical importance, being called on to play a role as part of a team in © 2013 American Chemical Society

enzyme activities, protein conformation, osmotic pressure, or pH regulation. On the one hand, then, there is a full awareness of the importance of specific ions in regulating biochemical pathways, but on the other hand, physical chemistry has still not given to biochemists the theoretical tools to explain them. Thus, e.g., classical intuition based on Debye−Huckel theory or its extensions suggests there should be no difference in behavior among ions of the same valence. Any deviation from limiting law (infinite dilution) behavior is just assigned to the mean activity coefficients. In fact, these cannot be predicted beyond 10−3−10−2 M without arbitrary parameters. This is at least 1 order of magnitude lower than the physiological electrolyte concentration. The situation is worse still for mixed electrolytes that occur in real biology. But fortunately, after many decades with theory frozen on the question of specificity, some progress is on the way.3,4,24,25 Protein systems are certainly not the easiest system to deal with due to the complexity and inhomogeneity of protein surfaces. Nevertheless, they constitute a benchmark test of any theory which is aimed to explain ion specificity. It is now widely recognized that ions adsorb specifically at aqueous/protein interface, thus affecting its physicochemical properties, like folding. But which driving forces are responsible for which specific ion binding and which surface sites are involved are still Received: September 17, 2013 Published: November 20, 2013 15350

dx.doi.org/10.1021/la404249n | Langmuir 2013, 29, 15350−15358

Langmuir

Article

Figure 1. (a) Variation of pH as a function of time during a turbidimetric pH titration experiment. (b) Turbidimetric pH titration: relative transmittance, T/T0, versus pH (red continuous line). Derivative of transmittance, d(T/T0)/dpH, versus pH (dashed blue line). [NaCl] = 50 mM; temperature = 25 °C.

open questions. There are at the present time two, only apparently, different approaches which are being used to explain ion specific effects. The former is an empirical rule known as Collins’ “law of matching water affinities” (LMWA),26 and the latter is a theory which considers specific ion−ion and ion−surface interactions due to the cooperative action of electrostatic and nonelectrostatic (NES) forces.27 The most recent developments of these approaches can be found elsewhere.2,25,28,29 But whatever the mechanism which brings to specific ion binding, the chemical nature of the protein binding sites has been partially understood for anions. It has been proposed that besides the positive chargesdue to the protonation of lysine, arginine, and histidine amino acids anions can bind also to amide backbones.30 Binding strength has been found to increase with increasing anion polarizability (SCN− > I− > NO3− > Br− > Cl−). This has led to an important advance in understanding why, for example, the anionic Hofmeister series reverses by changing pH (below or above IEP) or salt concentration.15,19,31,32 Although there is a vast literature on specific anion effects, this has not occurred for cations which give, in general, a weaker or even no effect.18 The result is that neither the type of interaction nor the specific binding site for cations binding has been identified with any good degree of certainty. Here we investigated the cation (co-ion) specific aggregation of native hemoglobin (Hb) protein as a function of pH in the presence of background salts in the range of concentration 50− 150 mM. Hemoglobin is likely the most studied protein due to its biological importance for the life of vertebrates.33 Hb aggregation was studied through turbidimetric pH titrations according to an experimental protocol very recently proposed by Yan et al.34 We searched for a set of experimental conditions which permit to observe a marked specific cation effect at salt concentrations which are comparable to those in physiological systems. We preliminarily show that Hb aggregates very easily just changing solution conditions like pH, ionic strength, or salt anion. This last effect is very strong already at the lowest investigated salt concentration (50 mM). We found that the obtained specific cation series cannot be explained by considering that cation adsorption is driven by a matching in water affinity or by an increase of polarizability only. A more subtle mechanism is likely at work. The insights may be useful

in understanding the puzzling ion−protein and protein− protein interactions that occur in living systems.

2. EXPERIMENTAL SECTION 2.1. Chemicals. Human hemoglobin was purchased from SigmaAldrich (H7379) and used without further purification. Sodium chloride (>98%), sodium bromide (99%), sodium nitrate (99%), sodium iodide (≥99.5%), lithium chloride (99%), potassium chloride (≥99%), rubidium chloride (≥99%), and cesium chloride (≥98%) were from Sigma-Aldrich (Milan, Italy). Standard buffers at pH 1, 4, 6, 9, and 10 were purchased from Hanna instruments (Szeged, Hungary). Sodium hydroxide and hydrochloric acid standard solutions (0.1 M) were from Fluka (Milan, Italy). 2.2. Sample Preparation. All salts were dried overnight at 110 °C, cooled at room temperature in a desiccator, and dissolved in purified water (conductivity ≤0.054 μS cm−1), prepared by means of a Millipore water purification system (Millipore, UK). Hemoglobin dispersions (1 mg/mL) in 50 mM salt (NaCl, NaBr, NaNO3, NaI) solutions were prepared to verify the expected effect of anions, whereas the effect of cations (LiCl, NaCl, KCl, RbCl, CsCl) was studied at three different concentrations, that is, 50, 100, and 150 mM. 2.3. Instrumentation. Turbidimetric pH titrations and kinetic measurements (see Supporting Information) were carried out by monitoring simultaneously the transmittance and the pH variations of hemoglobin suspensions due to the addition of a volume of a standard titrant solution. A thermostatic cell connected to a thermostatic bath (Huber, polystat cc1) was used to keep the sample temperature at 25 °C. The addition of the titrant solution and the measure of pH were done by means of an automatic titrator, Titrando 836 from Metrohm (Herizau, Switzerland) interfaced to a PC with software Tiamo 1.3, equipped with a glass electrode which was calibrated by a 5-point calibration by means of standard buffers. We should also take into account that absolute pH measurements are affected by salt type and concentration because ion adsorption at the glass electrode surface affects the electrochemical potential and, hence, the pH calculated through the conventional Nernst equation.10 But this effect becomes relevant at salt concentrations higher than those explored here. Hence, although we are aware that the interpretation of pH measurements is still an open question, we consider that effect to be negligible with respect to the ion specific effects on turbidimetric titrations. This might not be true at higher salt concentrations. The transmittance of the protein suspension was measured by using a UV−vis spectrophotometer Varian Cary 50 equipped with an optic fiber probe (path length = 1 cm). Transmittance (T%) was acquired at 780 nm, a wavelength at which there is no optical absorption, and the variation of T% is only due to the light scattering caused by protein aggregates. A T0 value, which was taken as T = 100%, was recorded for each experiment at the initial titration point (pH = 4.5). 15351

dx.doi.org/10.1021/la404249n | Langmuir 2013, 29, 15350−15358

Langmuir

Article

Figure 2. Effect of NaCl concentration on turbidimetric pH titration of hemoglobin dispersions: (a) T/T0 versus pH; (b) derivative plot. Temperature = 25 °C. 2.4. Turbidimetric pH Titrations. In a typical titration experiment a freshly prepared hemoglobin dispersion was first pretitrated at pH 4.5 with the standard HCl solution (0.1 M). The dispersion was then titrated with a standard solution of NaOH (0.1 M) up to pH = 9.5. In the pH range (4.5−9.5) here investigated, hemoglobin would not be subjected to denaturation. Indeed, according to Tanford,35 the range of reversibility of the hemoglobin titration curve (that is, the range of pH at which the protein keeps its native form) is between pH 4.4 and 11.5. The titrant was added at a constant rate of 5 μL/s (a volume of 5 μL is indeed the minimum value which the automatic titrator is able to add with accuracy) to obtain an average rate of pH change of 0.3 pH unit/min. Each curve presented below is the average of at least five different measurements. Before proceeding with cation measurements, we first confirmed that the procedure generates the expected anion effects. At higher concentration (i.e., [NaI] = 0.1 M, [NaNO3] = 0.15 M) we found such a strong specific anion effect on Hb aggregation that the dispersion was not optically clear at pH 4.5, thereby preventing an effective comparison between the effects of different anions at concentrations higher than 50 mM. The same confoundingly strong aggregation was given by NaSCN at 50 mM, so that the titration curve relative to such a chaotropic anion could not be included in the plot showing the specific effect of anions on turbidimetric titration of Hb (Figure 3). These observations confirm the sensitivity of hemoglobin to aggregation conditions, which prevents a detailed study of specific anion effects at higher salt concentrations but, on the other hand, means that this system is ideal for deeply investigating the specific effect of cations, which is usually weaker than the anion effect.

of T/T0 close to that at the beginning of the titration. The effect of pH on protein aggregation can be explained to first approximation by invoking the DLVO theory of colloid science36−38 based on the balance of attractive and repulsive interactions. Attractive interactions are primarily van der Waals forces, while repulsive interactions are due to adsorption of ions. For proteins, repulsive forces are pH-dependent since the electric charge on the protein surface, responsible for electrostatic adsorption of ions, is produced by the proton dissociation/association with weak acidic (R−COOH) and basic (R−NH2 or similar nitrogen-based) groups carried by surface amino acids. The repulsive force is then minimal at the isoelectric point (IEP, zero net charge) and maximal at pH ≫ IEP or pH ≪ IEP. The curve in Figure 1b does not have a symmetric shape, probably because the aggregation and disaggregation phenomena follow different kinetic mechanisms.34 But here we focus the attention mainly on the left side of the curve where the aggregation occurs. The relationship between the turbidimetric pH titrations (continuous red curve in Figure 1b) and aggregation kinetics (detailed in the Supporting Information) can be found through the first derivative of the transmittance, d(T/T0)/dpH, plotted versus pH (dashed blue curve) in Figure 1b. This derivative is equivalent to the kinetic aggregation rate shown in Figure S1c because of the linear variation of pH with titration time (Figure 1a). In Figure 1b, we find the aggregation rate varies with pH and the maximal aggregation rate is obtained at pH = 6.60, a value which is comparable with that obtained by the kinetic experiments (pH = 6.88, Figure S1c). This result is in agreement with that reported by Yan et al. for β-lactoglobulin.34 The two methods are in principle equivalent, but the turbidimetric titration is more convenient from an experimental point of view because the pH dependence of the initial aggregation rate of protein molecules can be obtained in a single experiment rather than through several experiments, each of them at a different pH value. Moreover, from what is reported in Figure 1b (red continuous curve), we can identify two characteristic points. The first is the pH of the inflection point (pHinflection), which corresponds to the minimum in the derivative plot (blue dashed curve). This point is the pH at which the maximal initial aggregation rate for hemoglobin molecules occurs. The other characteristic point (pHminimum) is the pH at which the minimum value of T/T0 is

3. RESULTS 3.1. Turbidimetric pH Titrations. Turbidimetric pH titrations can be used to study the effect of pH on the aggregation/disaggregation of protein molecules. This method has been recently shown by Yan et al.34 to be equivalent, but more convenient by the experimental point of view, to the kinetic measurements shown in the Supporting Information. The titrant (NaOH) addition was carried out at a constant rate in order to achieve variations of about 0.3 pH units/min. Figure 1a shows that pH varies almost linearly with time in a wide pH range which includes that where aggregation and part of disaggregation phenomena occur. Figure 1b shows the turbidimetric pH titration as a function of pH (continuous red curve). The hemoglobin suspension is optically clear below pH 5 and then becomes cloudy after about pH 6 due to the aggregation of protein molecules. Transmittance reaches a minimum at about pH 7.4 (pHminimum), and then it increases, due to the redissolution of protein aggregates, reaching a value 15352

dx.doi.org/10.1021/la404249n | Langmuir 2013, 29, 15350−15358

Langmuir

Article

Figure 3. Specific anion effects on turbidimetric pH titration of hemoglobin dispersions: (a) T/T0 versus pH; (b) derivative plot. [NaX] = 50 mM; temperature = 25 °C.

then we observe a less marked variation along the investigated pH range for cations than in the case of anions. This agrees with the general notion that cations display a less important effect than anions, particularly in the case of proteins.18 Nevertheless a quite marked effect occurs. If we consider the trends at pH ∼ pHminimum, cations promote hemoglobin aggregation in the order Rb+ > K+ ∼ Na+ > Cs+ > Li+. Similar experiments carried out at 100 and 150 mM gave the following series: K+ ∼ Rb+ > Na+ > Cs+ > Li+ (100 mM) and K+ > Rb+ > Na+ > Li+ > Cs+ (150 mM). As for the case of anions, also in this case the different curves cross each other at about pH 9 (about pH 8 for salt concentration 150 mM). The order of cations in the series changes with increasing salt concentration. This effect has already been observed for anions in other protein systems.15 But the most surprising result is that at the lower concentration (50 mM) the cations with the biggest difference in size (that is Li+ and Cs+) behave similarly in promoting hemoglobin aggregation.41−43

obtained (this corresponds to an inflection point in the derivative plot). At pHminimum the aggregation and disaggregation phenomena occur with the same rate. The two points will be useful in the following parts of this work 3.2. Ionic Strength and Specific Anion Effects. Besides pH, also ionic strength and the specific anion affect hemoglobin aggregation. These two effects were preliminarily studied in order to find the optimal experimental conditions to investigate specific cation effects. Figure 2a shows that an increase of ionic strength produces a decrease of T/T0. All curves have a pHminimum close to the literature value of isoelectric point of hemoglobin (IEP ∼ 7.1 ± 0.3).39 These two points are evidently correlated, since at the IEP the repulsive force is minimal and protein molecules experience the maximal attractive interaction which is responsible for the maximal aggregation (minimum of T/T0). We also observe that the increase of salt concentration results in a small shift of pHminimum toward lower pH values. A shift of IEP with increasing NaCl concentration was recently experimentally found and theoretically explained in a recent work dealing with bovine serum albumin, by the occurrence of higher dispersion interaction of chloride anion with respect to sodium cation.40 The increase of ionic strength has also the effect to shift toward lower pH values the starting point of aggregation (Figure 2a) and the pHinflection (6.55 at 50 mM; 6.45 at 100 mM; and 6.1 at 150 mM) which, we recall, is directly related to the maximal aggregation rate of Hb (Figure 2b). We confirmed a strong anion specific effect with NaCl, NaBr, NaNO3, and NaI at 50 mM (Figure 3a,b). Both pHminimum and pHinflection as well as the extent of aggregation given by the value T/T0 at different pH values are anion specific and follow an inverse Hofmeister series: I− > NO3− > Br− > Cl−. This behavior corresponds to anion binding to the surface (or surface sites) which decreases the net positive surface charge of protein molecules and hence decreases the repulsion among them. A crossing point with reversal of the anion Hofmeister series is found at pH 9, above which a direct Hofmeister series is obtained.31 3.3. Specific Cation Effects. Our chief measurements were turbidimetric pH titrations with different 50 mM chloride salts, carried out to investigate cation specific effects. At the starting pH 4.5 ( NO3− > Br− > Cl−. But this is also the order of decreasing polarizability (if nitrate is not considered) and of partial molar volumes, so that in this case it is not easy to understand which parameter plays the most important role. We may argue that all are at work and, more importantly, for anions they operate in the same direction. Then, the same correlation procedure tested for of anions has been applied to cations. First of all, cation specific differences in the pHminimum, distinct from the case of anions, are here hardly observed. Correlations between pHinflection and the difference between ΔHhydration of acetate48 (which is taken as a reference anionic group due to its similarity with carboxylates occurring at protein surfaces) minus ΔHhydration of the different cations as well as with polarizability and partial molar volumes, at different salt concentrations, are shown in Figure 6. According to LMWA, the strength of interaction of cations with negatively charged carboxylates (classified as kosmotropes) would decrease along the series Li+ > Na+ > K+ > Rb+ > Cs+. But the order would be the opposite if polarizability is considered a dominant factor. In fact, the obtained trends agree neither with LMWA (Hb aggregation would have to decrease in the order Cs+ > Rb+ > K+ > Na+ > Li+) nor with polarizability order (Hb aggregation would have to decrease in the order Li+ > Na+ > K+ > Rb+ > Cs+). Also, the correlation plot of pHinflection versus partial molar volume does not follow a monotonic trend. We argue that differently from the case of anions, the correlation measures for cations are here operating in opposite directions and what is

Table 1. Literature Values of Enthalpies of Hydration (ΔHhydration),46 Static Polarizabilities (α0),3 and Partial Molar Volumes (Vi)47 ion −

Cl Br− NO3− I− Li+ Na+ K+ Rb+ Cs+

ΔHhydration (kJ mol−1)

α0 (Å3)

Vi (cm3 mol−1)

−381 −347 −314 −305 −519 −409 −322 −293 −264

4.220 6.028 4.008 8.967 0.028 0.131 0.795 1.348 2.354

23.3 30.2 34.5 41.7 −5.2 −5.7 4.5 9.5 16.9

literature values of these parameters.3,46,47 We first test the correlation procedure for the anion specific effects observed at 50 mM. The insight obtained will be useful in the following part dedicated to specific cation effects in the concentration range 50−150 mM. Figure 5a shows a plot of both pHmin and pHinflection as a function of the difference between ΔHhydration of the different anions minus ΔHhydration of ammonium cation (which is taken as a reference cation due to its similarity to the amine groups occurring at protein surfaces). Similarly, the same two parameters have been correlated with anion static polarizability and anion partial molar volumes in Figures 6b and 6c, respectively. We see enough good correlation of pHminimum and

Figure 5. Correlation of experimental pHmin (empty circles) and pHinflection (filled circles) for the anions with (a) the hydration enthalpy (ΔHhydration) difference between the anion and ammonium ion (−307 kJ mol−1), (b) anion static polarizability (α0), and (c) partial molar volumes (Vi). 15355

dx.doi.org/10.1021/la404249n | Langmuir 2013, 29, 15350−15358

Langmuir

Article

Figure 6. Correlation of experimental pHinflection for the cations with (a) the hydration enthalpy (ΔHhydration) difference between acetate ion (−425 kJ mol−1) and the cation; (b) cation static polarizability (α0); and (c) partial molar volumes (Vi). The dashed curves do not have any physical meaning and are only a guide for the eye.

neutral sites of Hb, making its surface more positive and so contrasting protein aggregation. Differently from the case of anions, specific cation effects demonstrate that LMWA gives only a partial explanation of what is going on. According to the LMWA, we can exclude that chaotropic cations bind to kosmotropic carboxylates. Instead, chaotropic cations would adsorb at uncharged sites as predicted by NES forces theory. Other experimental techniques have to be used to identify the chemical nature of these surface binding sites.30

observed is the result of their combination. A similar behavior has in fact been observed in several protein/enzyme systems.41−43 4.3. A Possible Explanation of Cation Specificity. Since protein surfaces are in general highly inhomogeneous, at almost any pH, both positively charged and negatively charged sites coexist with uncharged patches. In order to rationalize our results, let us consider that Hb aggregation is affected by specific ion binding to specific sites at its surface. That is, protein surfaces present in general different binding sites for specific ions. In our experimental conditions (a titration from acidic toward basic pH values), anion binding promotes Hb aggregation through a decrease of its net positive charge. Highly polarizable (chaotropic) iodide will adsorb at a higher extent than less polarizable chloride anions. In such case the adsorbing sites are (according with LMWA and NES forces theory) the chaotropic positively charged groups of Hb. Moreover, according to NES forces theory, Rember et al. have shown that also neutral patches can act as anion adsorbing sites.30 Hence, anion correlation plots (Figure 5) are consistent with both ion/charged-site binding (the only mechanism considered by LMWA) and ion/neutral-site binding (the additional mechanism due to the action of NES forces). The specific effect of cations seems to be less easy to deal with. We may think that cations form an ion pair with kosmotropic carboxylates in an order which would follow the LMWA, that is, Li+ > Na+ > K+ > Rb+ > Cs+. The effect of kosmotropic Li+ is to decrease the number of negative charges and hence to increase the positive net charge of Hb. This would result in a higher repulsion and hence would negatively affect Hb aggregation, which was indeed observed. But the experimental turbidity trends show that Cs+ behaves similarly to Li+. According to LMWA, the chaotropic Cs+ cation cannot form an ion pair with carboxylates; therefore, the only way to explain the experimental results is to consider that it adsorbs to the uncharged patches of Hb driven by its high polarizability. Cs+ adsorption will hence increase the positive net charge of the protein surface, strengthen the repulsion among Hb molecules, and thus disfavor protein aggregation. That is, Li+ and Cs+ produce a similar effect on aggregation but through two very different mechanisms. An increase of concentration results in a more complicated trend likely because the previously observed phenomenon of “series reversal” takes place.15 As the charged sites become saturated, even the other chaotropic cations (i.e., Rb+ and K+) are able to bind to the

5. CONCLUSIONS In this work the cation specific effects on the aggregation of hemoglobin at low and physiological salt concentration in the pH range 4.5−9.5 were investigated. Because of the zwitterionic nature of proteins, both cations and anions coming from salt dissociation play a simultaneous role in affecting Hb aggregation. This is likely a general mechanism since in physiological (pH) conditions most proteins have both positively and negatively charged (as well as uncharged) residues which act as binding sites of both anions and cations. Anions at fixed concentration (50 mM) strongly affect Hb aggregation according to an expected15,19 inverse Hofmeister series: I− > NO3− > Br− > Cl−. Cations showed a less marked effect on aggregation than anions since they are co-ions and have lower polarizability values. Nevertheless, cations affect Hb solubility. At pH < IEP and salt concentration 50 mM, they promote Hb aggregation according to the series Rb+ > K+ > Na+ > Cs+ > Li+. Even more challenging trends are obtained at 100 and 150 mM. These trends cannot be explained by considering the law of matching water affinities (LMWA) as the only mechanism at work. The results obtained here can be explained with the occurrence of different binding sites and mechanisms depending on the nature of the ion. It has been proposed that anions bind at both charged and uncharged amide backbone sites.30 Chaotropic anions would strongly bind to both types of sites whereas kosmotropic anions would bind poorly.30 Chaotropic cations likewise bind well to uncharged backbone sites but bind poorly to kosmotropic carboxylate sites, and vice versa for kosmotropic cations. Hence, binding to both kind of sites is cooperative for anions (acting in the same direction) but antagonistic for cations. Anions and cations at the protein interface do not show “symmetric” behavior because protein positively and negatively charged groups are chaotropic and kosmotropic, respectively. Hence, the simultaneous action of 15356

dx.doi.org/10.1021/la404249n | Langmuir 2013, 29, 15350−15358

Langmuir

Article

(14) Varhac, R.; Tomásková, N.; Fabián, M.; Sedlák, E. Kinetics of cyanide binding as a probe of local stability/flexibility of cytochrome c. Biophys. Chem. 2009, 144, 21−26. (15) Zhang, Y.; Cremer, P. S. The inverse and direct Hofmeister series for lysozyme. Proc. Natl. Acad. Sci. U. S. A. 2009, 106, 15249− 15253. (16) Lo Nostro, P.; Peruzzi, N.; Severi, M.; Ninham, B. W.; Baglioni, P. Asymmetric partitioning of anions in lysozyme dispersions. J. Am. Chem. Soc. 2010, 132, 6571−6577. (17) Medda, L.; Barse, B.; Cugia, F.; Bostrom, M.; Parsons, D. F.; Ninham, B. W.; Monduzzi, M.; Salis, A. Hofmeister challenges: Ion binding and charge of the BSA protein as explicit examples. Langmuir 2012, 28, 16355−16363. (18) Gokarn, Y. R.; Fesinmeyer, R. M.; Saluja, A.; Razinkov, V.; Chase, S. F.; Laue, T. M.; Brems, D. N. Effective charge measurements reveal selective and preferential accumulation of anions, but not cations, at the protein surface in dilute salt solutions. Protein Sci. 2011, 20, 580−587. (19) Salis, A.; Cugia, F.; Parsons, D. F.; Ninham, B. W.; Monduzzi, M. Hofmeister series reversal for lysozyme by change in pH and salt concentration: insights from electrophoretic mobility measurements. Phys. Chem. Chem. Phys. 2012, 14, 4343−4346. (20) Cugia, F.; Monduzzi, M.; Ninham, B. W.; Salis, A. Interplay of ion specificity, pH and buffers: insights from electrophoretic mobility and pH measurements of lysozyme solutions. RSC Adv. 2013, 3, 5882−5888. (21) Medda, L.; Salis, A.; Magner, E. Specific ion effects on the electrochemical properties of cytochrome c. Phys. Chem. Chem. Phys. 2012, 14, 2875−2883. (22) Robertson, B. T. J. Biol. Chem. 1911, 9, 303. (23) Green, A. A. Studies in the physical chemistry of the proteins: X. The solubility of hemoglobing in solutions of chlorides and sulfates of varying concentration. J. Biol. Chem. 1932, 95, 47−66. (24) Schwierz, N.; Horinek, D.; Netz, R. R. Anionic and cationic Hofmeister effects on hydrophobic and hydrophilic surfaces. Langmuir 2013, 29, 2602−2614. (25) Parsons, D. F.; Bostrom, M.; Lo Nostro, P.; Ninham, B. W. Hofmeister effects: interplay of hydration, nonelectrostatic potentials, and ion size. Phys. Chem. Chem. Phys. 2011, 13, 12352−12367. (26) Collins, K. D. Charge density-dependent strength of hydration and biological structure. Biophys. J. 1997, 72, 65−76. (27) Ninham, B. W.; Yaminsky, V. Ion binding and ion specificity: the Hofmeister effect and Osanger and Lifshits theories. Langmuir 1997, 13, 2097−2108. (28) Collins, K. D. Why continuum electrostatics theories cannot explain biological structure, polyelectrolytes or ionic strength effects in ion-protein interactions. Biophys. Chem. 2012, 167, 43−59. (29) Kunz, W. Specific ion effects in colloidal and biological systems. Curr. Opin. Colloid Interface Sci. 2010, 15, 34−39. (30) Rembert, K. B.; Paterova, J.; Heyda, J.; Hilty, C.; Jungwirth, P.; Cremer, P. S. Molecular mechanisms of ion-specific effects on proteins. J. Am. Chem. Soc. 2012, 134, 10039−10046. (31) Boström, M.; Parsons, D. F.; Salis, A.; Ninham, B. W.; Monduzzi, M. Possible origin of the inverse and direct Hofmeister series for lysozyme at low and high salt concentrations. Langmuir 2011, 27, 9504−9511. (32) Schwierz, N.; Horinek, D.; Netz, R. R. Reversed anionic Hofmeister series: The interplay of surface charge and surface polarity. Langmuir 2010, 26, 7370−7379. (33) Baldwin, J. M. Structure and function of haemoglobin. Prog. Biophys. Mol. Biol. 1976, 29, 225−320. (34) Yan, Y.; Seeman, D.; Zheng, B.; Kizilay, E.; Xu, Y.; Dubin, P. L. pH-dependent aggregation and disaggregation of native β-lactoglobulin in low salt. Langmuir 2013, 29, 4584−4593. (35) Tanford, C. The interpretation of hydrogen ion titration curves of proteins. In Advances in Protein Chemistry; Anfinsen, C. B., Bailey, K., Anson, M. L., Edsall, J. T., Eds.; Academic Press: New York, 1963; Vol. 17, pp 69−165.

the two competing mechanisms allows two so different ions (Li+ and Cs+) to yield similar macroscopic effects (protein aggregation). Our future goal will be to prove that this mechanism can, in fact, be modeled by means of the tools of the ion dispersion forces theory. Since the salt concentration investigated here is close to the physiological electrolyte concentration, we expect that a mechanism similar to what is hypothesized here might play a role in affecting the behavior of proteins in living organisms.



ASSOCIATED CONTENT

S Supporting Information *

Description of the aggregation kinetics method and the results. This material is available free of charge via the Internet at http://pubs.acs.org.



AUTHOR INFORMATION

Corresponding Author

*Tel +61 2 6125 9184; Fax +61 2 6125 0732; e-mail andrea. [email protected] (A.S.). Notes

The authors declare no competing financial interest.

■ ■

ACKNOWLEDGMENTS MIUR, PRIN grant 2010BJ23MN-002, is thanked for financial support. A.S. thanks the ARC Discovery project. REFERENCES

(1) Hofmeister, F. Arch. Exp. Pathol. Pharmakol. 1888, 24, 247−260. (2) Lo Nostro, P.; Ninham, B. W. Hofmeister Phenomena: An Update on Ion Specificity in Biology. Chem. Rev. 2012, 112, 2286− 2322. (3) Ninham, B. W.; Lo Nostro, P. Molecular Forces and Self Assembly In Colloid, Nano Sciences and Biology; Cambridge University Press: Cambridge, 2010. (4) Kunz, W. Specific Ion Effects; World Scientific Publishing: Singapore, 2010. (5) Jones, G.; Dole, M. The viscosity of aqueous solutions of strong electrolytes with special reference to barium chloride. J. Am. Chem. Soc. 1929, 51, 2950−2964. (6) Craig, V. S. J. Bubble coalescence and specific-ion effects. Curr. Opin. Colloid Interface Sci. 2004, 9, 178−184. (7) Zhang, Y.; Furyk, S.; Bergbreiter, D. E.; Cremer, P. S. Specific ion effects on the water solubility of macromolecules: PNIPAM and the Hofmeister series. J. Am. Chem. Soc. 2005, 127, 14505−14510. (8) Deyerle, B. A.; Zhang, Y. Effects of Hofmeister anions on the aggregation behavior of PEO-PPO-PEO triblock copolymers. Langmuir 2011, 27, 9203−9210. (9) Lutter, J. C.; Wu, T.-y.; Zhang, Y. Hydration of cations: A key to understanding of specific cation effects on aggregation behaviors of PEO-PPO-PEO triblock copolymers. J. Phys. Chem. B 2013, 117, 10132−10141. (10) Salis, A.; Pinna, M. C.; Bilanicova, D.; Monduzzi, M.; Lo Nostro, P.; Ninham, B. W. Specific anion effects on glass electrode pH measurements of buffer solutions: bulk and surface phenomena. J. Phys. Chem. B 2006, 110, 2949−2956. (11) Lützenkirchen, J. Specific ion effects at two single-crystal planes of sapphire. Langmuir 2013, 29, 7726−7734. (12) Morag, J.; Dishon, M.; Sivan, U. The governing role of surface hydration in ion specific adsorption to silica: An AFM-Based account of the Hofmeister universality and its reversal. Langmuir 2013, 29, 6317−6322. (13) Toth, K.; Sedlak, E.; Sprinzl, M.; Zoldak, G. Flexibility and enzyme activity of NADH oxidase from Thermus thermophilus in the presence of monovalent cations of Hofmeister series. Biochim. Biophys. Acta, Proteins Proteomics 2008, 1784, 789−795. 15357

dx.doi.org/10.1021/la404249n | Langmuir 2013, 29, 15350−15358

Langmuir

Article

(36) Wagner, M.; Reiche, K.; Blume, A.; Garidel, P. The electrokinetic potential of therapeutic proteins and its modulation: Impact on protein stability. Colloids Surf., A 2012, 415, 421−430. (37) Verwey, E. J. W.; Overbeek, J. T. G. Trans. Faraday Soc. 1946, 42B, 117. (38) Derjaguin, B. On the repulsive forces between charged colloid particles and on the theory of slow coagulation and stability of lyophobe sols. Trans. Faraday Soc. 1940, 35, 203−215. (39) Adair, G. S.; Adair, M. E. The determination of the isoelectric and isoionic points of haemoglobin from measurements of membrane potentials. Biochem. J. 1934, 28, 1230−1258. (40) Salis, A.; Boström, M.; Medda, L.; Cugia, F.; Barse, B.; Parsons, D. F.; Ninham, B. W.; Monduzzi, M. Measurements and theoretical interpretation of points of zero charge/potential of BSA protein. Langmuir 2011, 27, 11597−11604. (41) Katz, A. M. Effects of alkali metal ions on the Mg2+-activated ATPase activity of reconstituted actomyosin. Biochim. Biophys. Acta, Bioenerg. 1968, 162, 79−85. (42) Lindenmayer, G. E.; Schwartz, A. Conformational changes induced in Na+, K+-ATPase by ouabain through a K+-sensitive reaction: Kinetic and spectroscopic studies. Arch. Biochem. Biophys. 1970, 140, 371−378. (43) Poillon, W. N.; Bertles, J. F. Deoxygenated sickle hemoglobin. Effects of lyotropic salts on its solubility. J. Biol. Chem. 1979, 254, 3462−7. (44) Collins, K. Biophys. Chem. 2006, 119, 271−281. (45) Collins, K. D. Ions from the Hofmeister series and osmolytes: effects on proteins in solution and in the crystallization process. Methods 2004, 34, 300−311. (46) Smith, D. W. Ionic hydration enthalpies. J. Chem. Educ. 1977, 54, 540. (47) Millero, F. J. Partial molal volume of ions in various solvents. J. Phys. Chem. 1969, 73, 2417−2420. (48) Kherb, J.; Flores, S. C.; Cremer, P. S. Role of carboxylate side chains in the cation Hofmeister series. J. Phys. Chem. B 2012, 116, 7389−7397.

15358

dx.doi.org/10.1021/la404249n | Langmuir 2013, 29, 15350−15358