|Home | About | Journals | Submit | Contact Us | Français|
The hydrophobic spin probe 2,2,6,6-tetramethyl-piperidin-1-oxyl-4-yl octadecanoate (TEMPO-stearate) was used to study the hydration of the polar shell of Triton® X-100 micelles as functions of the concentration of the electrolytes KCl, NaCl and LiCl, and temperature. It was shown that the hydration of the polar shell of the Triton® X-100 micelle decreases with both increasing electrolyte concentration and increasing temperature. The effect of Li+ on the hydration of the polar shell was found to be smaller than those of Na+ and K+, which have almost identical behavior. The effective water concentration decreases from 18.3 M to 15.8 M for LiCl and from 18.3 M to 13.9 M For NaCl and KCl when the concentration of the electrolyte in the solution increases from 0 to 2.5 M. The dehydration of the polar shell was correlated to the average value of the cation hydration number calculated from literature data; the greater the cation hydration number, the greater the dehydration for the same increase in electrolyte concentration. Also, it was shown that the cloud point is strongly correlated to the hydration of the polar shell.
((p-(1,1,3,3-tetramethylbutyl)phenyl poly(ethylene glycol)), Triton® X-100, is a nonionic surfactant that is extensively used to solubilize and purify transmembrane proteins.1, 2 Triton® X-100 mixed micelle systems have also been successfully employed with numerous lipid–dependent enzymes in the study of surface dilution kinetics. 3, 4, 5 Due to its importance in biochemistry applications, the physicochemical properties of pure and mixed Triton® X-100 micellar solutions have been studied with a wide range of techniques such as nuclear magnetic resonance (NMR) 6, 7, 8 quasi-elastic light scattering (QELS) 9, 10, 7 light scattering (LS) 11 time resolved fluorescence quenching (TRFQ), 7 rotating disk electrode voltammetry 12, small-angle X-ray scattering, 13 and viscometry. 14
Recently, the growth and hydration of Triton® X-100 micelles was studied as a function of the concentration of added electrolyte. 10, 12, 11 It was found that micelles become more compact with increasing electrolyte concentration, while at the same time there is an increase in the aggregation number of the micelle accompanied with an even greater increase in the amount of entrapped water at the periphery of the micelle. 12, 11 It was suggested that the observed decrease in the partial specific volume with increasing electrolyte concentration is very likely caused by dehydration of the hydrophilic section of the micelle. 11 In other words, the hydrophilic polar shell of the micelle becomes dehydrated with the addition of electrolyte while the number of water molecules per surfactant increases.
The spin probe electron paramagnetic resonance (EPR) method has been used to study the surface hydration of sodium dodecyl sulfate (SDS) micelles as a function of electrolyte concentration. 15, 16 The method relies on the sensitivity of the 14N hyperfine coupling constant of a hydrophobic spin label probe to the average fraction of the volume occupied by water in the region of the nitroxide moiety that is located in the polar shell (hydrophilic part) of the micelle. The effective water concentration or hydration index 17 in the polar shell can be obtained through a calibration curve of the spin probe measured in a series of solutions of known hydration, 15, 18 which then in turn may give the number of water molecules per monomer if the geometry of the micelle is well known. 15, 19, 20 Also, the relative heights of the resonance lines give estimates of the rotational mobility of the spin probe yielding insights into the dynamic nature of the probe’s environment. 21, 18
Since the substrate for studying the kinetics of lipid–dependent enzymes is usually solubilized by Triton® X-100 micelles, and the enzyme activity strongly depends on the physicochemical properties of the polar shell, 3, 4 such as surface hydration, it is important to gain a better understanding of the hydration state of the polar shell of Triton® X-100 micelles. 22, 23 The hydrophobic spin probe 2,2,6,6-tetramethyl-piperidin-1-oxyl-4-yl octadecanoate (TEMPO-stearate) has been used recently to study the interfacial properties of a variety of phosphatidylcholine vesicles 18 and of vesicles composed of the negatively charged dimyristoyl-phosphatidylglycerol (DMPG) and the zwitterionic dimyristoyl-phosphatidylcholine (DMPC) mixed with lysomyristoylphosphatidylcholine (LMPC). 19 TEMPO-stearate is very sparingly soluble in water, so it is certainly located in the Triton® X-100 micelles. Although, Triton® X-100 micelle has a relatively large hydrophilic corona and a small (true) hydrophobic core, it is very likely that the hydrocarbon tail of the spin probe most of the time resides in the hydrophobic core, while the nitroxide moiety resides in the polar shell of the micelle. Therefore, TEMPO-stearate is an ideal spin probe for studying the interfacial properties of Triton® X-100 micelles, because its location in the micelle is very likely the same as the location of the phospholipid molecule, so the nitroxide moiety senses the same water concentration as the polar head of the phospholipid molecule.
The purpose of the present work is to study the hydration properties of the polar shell of Triton® X-100 micelles using the spin probe EPR method, especially the effect of electrolyte, such as KCl, NaCl and LiCl, on the hydration of the polar shell of Triton® X-100 micelles.
The Triton® X-100 surfactant was obtained from Sigma (Saint Louis, MI).The spin label Tempo 2,2,6,6-tetramethylpiperiden-1oxyl-4-yl octadecanoate (TEMPO-stearate) was obtained from Molecular Probes, Inc. (Eugene, OR). Monovalent alkali salts - NaCl, KCl, LiCl were purchased from Sigma Aldrich (Saint Louis, MI). The distilled water was purchased from Arrowhead (Arrowhead, CA). All chemicals were used as purchased.
The appropriate amount of TEMPO-stearate ethanol solution was added to a vial and then dried with a stream of N2. Thereafter, a 25 mM water solution of Triton® X-100 was added to the vial to produce a molar ratio [Triton® X-100]/[spin probe] of 250, that is all the samples had 0.1 mM of TEMPO-stearate in the presence of 25 mM Triton® X-100. Each salt was added to this solution to achieve the desired concentration of electrolyte.
EPR spectra were measured on a Bruker ESP 300 E spectrometer equipped with a Bruker variable temperature unit (Model B-VT-2000). Spin labeled Triton® X-100 samples were prepared a day before the EPR measurements and stirred over night. On the day of measurement, the samples, not degassed, were drawn into glass capillaries and sealed with a gas torch. The glass capillary was then placed into the Dewar within the EPR cavity. The temperature of the sample was controlled with a Bruker temperature unit and was stable within ± 0.2 °C; the accuracy of the temperature was estimated to be within ± 1 °C. Five EPR spectra were acquired, one after the other, employing a sweep time of 84 s; time constant, 20 ms; microwave power, 5 mW; sweep width, 50.2 G; and modulation amplitude, 1.0 G. The magnetic sweep width was measured with Bruker’s NMR Gaussmeter operating in the 1-mG resolution mode.
The EPR spectra were transferred from the spectrometer to a PC where they were analyzed by the computer program Lowfit. The Lowfit program is based on the fact that the ESR lineshape, which is inhomogenously broadened due to unresolved hyperfine interaction, can be excellently described by a Lorentzian-Gaussian sum function. 24 One of the parameters that was obtained by the Lowfit analysis was the Lorentzian linewidth which was then used to calculate the rotational correlation times as described previously in reference 18. Since the TEMPO-stearate spin probe undergoes anisotropic Brownian rotational motion which is axially symmetric about an axis parallel to the hydrocarbon chain, one needs two independent rotational correlation times. Thus, one of the rotational correlation times τ‖ characterizes the rotation of the spin probe about the stearate acyl chain symmetry axis, while the second rotational correlation time τ characterizes the rotation perpendicular to it. Since the rotational correlation times τ‖ and τ characterize the rotation of the probe, which is sensitive to its surroundings, these quantities can give us useful information about the probe’s environment.
The Lowfit program also gives precise values of the EPR line positions, which are used in calculation of the nitrogen hyperfine coupling spacing A+. A representative EPR spectrum of 0.1 mM TEMPO-stearate in 25 mM Triton® X-100 micelles, its fit and the definition of A+ are given in the Supporting Information. The value of A+ is sensitive to the amount of water in the probe’s surroundings, 15, 25, 19 and therefore it is a good measure of effective water concentration in the vicinity of the probe. The method for estimating the effective water concentration in the polar shell of Triton® X-100 micelles is based on a calibration curve measured for TEMPO-stearate in a series of mixtures of ethanol-water and ethanol-1,4-dioxane covering the water fraction range in the solution from 0 to 80%. The hydration index of each mixture, 17 defined as the ratio of the molar concentration of OH dipoles in the solution mixture to that of pure water can be easily evaluated. 25 Since the hydration index is a linear function of molar concentration of water, the effective water concentration is given by the following calibration equation: 18
where 55.345 M is the molar concentration of water.
The effective water concentration [H2O] sensed by TEMPO-stearate hyperfine and corresponding coupling spacing A+ in the polar shell of the Triton® X-100 micelles mixed with different concentrations of NaCl as a function of temperature are shown in Figure 1. As can be noted, the effective water concentration and the hyperfine coupling spacing decrease with increasing temperature, as well as with increasing electrolyte concentration. The experimental points can be fitted very well to the exponential expression [H2O] = [H2O]0 exp(−εH2O /RT), where [H2O]0 is the effective water concentration at 1/T = 0, −εH2O is the activation energy for surface dehydration, and R = 8.315 J/mol•K is the gas constant, the solid lines in Figure 1. The graphs of [H2O] and A+ versus temperature with different electrolyte concentration, not shown here, for LiCl and KCl are very similar to Figure 1. Since the plots for all three electrolytes are the same in their general features, there is no need to display the other two plots. Thus, to save space we use the exponential function to present our experimental data in Table 1. The values of the correlation coefficients Rc also attest for the goodness of the fits.
The rotational correlation times τ‖ and τ characterize the axial rotation of the spin probe, which is affected by the physicochemical properties of the probe’s surroundings. Therefore, those relaxation times can give us additional information about the water structure and dynamics in the polar shell of Triton® X-100 micelles. Figure 2 shows the parallel rotational correlation time τ‖, characterizing the rotation about the probe hydrocarbon chain, for Triton® X-100 micelles as function of temperature for different concentrations of NaCl. As expected, the rotational correlation time τ‖ decreases with increasing temperature, while the addition of NaCl increases the value of τ‖.
Next, we present the perpendicular rotational correlation time τ, characterizing the rotation of the nitroxide perpendicular to the hydrocarbon chain, as a function of temperature with different concentration of NaCl, Figure 3. Again, the rotational correlation time τ decreases with increasing temperature and becomes longer with the addition of electrolyte. The value of τ is greater than the value of τ‖ for a given temperature at all temperatures, which is common for axial motion. Again, we do not show plots of experimental data for the rotational correlation times for KCl and LiCl. Instead we present the experimental data using the fitting parameters A and b of exponential fits (A e−bT) to the data, where the parameters of the parallel rotational correlation time τ‖ are given in Tables 2, while the parameters of the perpendicular rotational correlation time τ are given in Tables 3. As can be seen from Figure 2 and Figure 3, and the correlation coefficients in Table 2 and Table 3, the experimental data can be fitted extremely well to an exponential function. As far as we know, the fitting function used for the τ‖ and τ data does not have any physical meaning; it is just a good fit to the experimental data.
Experimental methods such as NMR 8 quasi-elastic light scattering (QELS) 9, 10 light scattering (LS) 11 rotating disk electrode voltammetry 12, small-angle X-ray scattering, 13, and viscometry. 14 measure the total number of water molecules associated with Triton® X-100 micelles. By contrast, the spin probe EPR method measures the effective water concentration sensed by the nitroxide moiety in its surroundings 18, 19, 25, 21. Therefore, in order to clearly understand the results of this work, it is necessary to compare the effective water concentration, in other words, the effective number of water molecules measured by the spin probe EPR method to those hydration values measured by other methods. Since the spin probe EPR method measures water in the immediate proximity of the probe, while some hydrodynamic methods measure water in a more extended region, we have to be very careful when comparing the same quantities measured by those methods. That is, data gathered by different methods very often sense different regions and might be painting somewhat different pictures.
The effective water concentration [H2O] can be easily found from the measured hyperfine splitting A+, equation (1), which can be then converted into the number of water molecules per surfactant sensed by the spin probe according to the following equation:
where VW = 30 Å3 is the volume of a water molecule and VOE = 637 Å3 is the volume of the ethylene oxide group.26 Although the location of the spin probe is well known, that is the hydrocarbon tail is in the hydrocarbon core of the Triton® X-100 micelle and the nitroxide moiety is in the polar shell of the micelle, the range of motion of the nitroxide is not, which introduces an uncertainty when the effective water concentration is converted into number of water molecules. This uncertainty was discussed in more detail in Reference 19. Here we assume that the nitroxide moiety samples the volume that is equal to the volume of the ethylene oxide (EO) group plus the volume of the water molecules surrounding the OE group, the denominator in Equation (2). We also assume that the addition of salt does not noticeably change the relative position of the spin probe and Triton® X-100 molecules. The effective numbers of water molecules calculated from the effective water concentration presented in Figure 4 (full squares) for pure Triton® X-100 micelles at different temperatures are given in Table 4. Hydration numbers for pure Triton® X-100 micelles have been reported to be in the range from 20 to 50 water molecules. 12 Taking into account that other methods measure both the water in the polar shell and the water hydrodynamically associated with the micelles, while our method measures only the water molecules in the polar shell, it is not surprising that our hydration numbers are slightly smaller, and as such are in very good agreement with previously reported numbers. Note that we avoid identifying the water sensed by nitroxide to what other authors call hydrogen bound water. 12, 11 Since the nitroxide moiety is very likely located in the polar shell the water sensed by nitroxide is slightly less than the hydrogen bound water and as long as the micelle does not undergo drastic structural changes, it is a good measure of that water.
The water concentration in the polar shell of globular structures such as micelles can be calculated from the geometrical parameters of the micelle and the surfactant, and the aggregation number 15, 21, 19 by
, where VS is the volume of the polar shell of the micelle and the VH = Nagg VOE is the volume of all the surfactant heads in the polar shell. VS is given by, where r0 is the effective radius of the micelle and Vhg is the volume of the hydrophobic group of a Triton® X-100 molecule. In other to further validate our method we use the radius of the Triton® X-100 micelle from Table II in Reference 14, and the aggregation number (143 molecules) and the volumes of the hydrophobic (363 Å3) and hydrophilic (637 Å3) parts of Triton® X-100 reported by Robson and Dennis 26 to calculate the effective water concentration in the polar shell of the Triton® X-100 micelle. The calculated [H2O] (full circles) together with measured [H2O] (full squares) are presented in Figure 4. As can be seen, the agreement is very good, and the temperature dependence of [H2O] is the same, which means that the spin probe EPR method can successfully be used to study the effective water concentration in Triton® X-100 micelles as a function of both electrolyte concentration and temperature.
The effective water concentration in the polar shell of the Triton® X-100 micelle decreases with increasing temperature, Figure 1 and Figure 4, as well as with increasing electrolyte concentration, Figure 1. In other words, the polar shell gets dehydrated with an increase in both temperature and electrolyte concentration. The good exponential fits indicate that the dehydration process is an activated process and the fact that −εH2O is positive means the process is spontaneous, in other words the water molecules transferring from the polar shell to the aqueous micellar surroundings do not encounter a potential barrier, but do fall into a potential well. Also, it can be clearly seen that the activation energy is directly proportional to the concentration of electrolyte, Table 1. Several authors have observed a decrease in intrinsic viscosity with increasing electrolyte concentration, which indicates a decrease in specific micellar volume. The decrease in specific micellar volume was then attributed to collapse of the EO chains due to dehydration. 12, 11 Micellar surface dehydration was also inferred from a decrease of area per surfactant with increasing electrolyte concentration. 11 Obviously, our observation of micellar surface dehydration is in good agreement with those previously reported.
We present a plot of the hyperfine splitting and effective water concentration of Triton® X-100 micelles as a function of electrolyte concentration at 35 °C for NaCl, KCl and LiCl, Figure 5. The plots at other temperatures are similar. From Figure 5, it is obvious that dehydration is more pronounced in the presence of NaCl and KCl than LiCl. The effective water concentration decreases from 18.3 M to 15.8 M when the concentration of LiCl in the solution increases to 2.5 M, while for the same mole increase of NaCl and KCl the value of [H2O] decreases to 13.9 M. In terms of water molecules, calculated by Equation (2), the presence of 2.5 M LiCl in the solution reduces the number of water molecules per EO group from 10 to 8, whereas the presence of the same concentration of the other two electrolytes reduces the number of water molecules from 10 to 6.8.
To see how the presence of an electrolyte affects the motion of nitroxide, that is the parallel and perpendicular rotational times, we present τ, Figure 6, and τ‖, Figure 7, as a function of electrolyte concentration at 35 °C for all three electrolytes. As expected, the values of both τ and τ‖ increase with increasing electrolyte concentration indicating the dehydration of the hydrophilic shell of the micelle. Again, the effect of LiCl is quite different than that of the other two electrolytes, but this time the effect of LiCl is greater than the effect of NaCl and KCl. The fact that the values of τ and τ‖ are longer in the presence of LiCl indicates that the structure of water molecules in the polar shell of the micelle is slightly more rigid which might be explained by possible forming of complexes between Li+ and Triton® X-100 molecules. 27 Na+ and K+ do not form complexes with polyoxyethylated surfactants. 27
One of the macroscopic properties of a nonionic surfactant is its cloud point, which is the temperature where the surfactant solution starts to separate into two phases making the solution look cloudy. 12, 11 Bolivar et al 11 and Charlton and Doherty 12 showed that the cloud point was reduced with increasing electrolyte concentration due to salting out behavior. The presence of cations in the aqueous solution decreases the water concentration in the polar shell, so the number of hydrogen bonds formed with the ether groups decrease, which in turn reduces the barrier for micelle-micelle interactions. The final result of salting-out is that the micelle-micelle interaction becomes more likely and the cloud point decreases. Since the EPR method measures the effective water concentration in the polar shell, we present a plot of cloud point versus effective water concentration (lower axis) and hyperfine coupling constant (upper axis) for all three electrolytes in Figure 8. The data for the cloud points of Triton® X-100 in the presence of NaCl and LiCl are taken from Figure 4, Reference 11, and in the presence of KCl from Figure 5, Reference 12. As can be seen, there is an almost perfect linear relation between cloud point and [H2O] for all three electrolytes. Notice that the addition of LiCl lowers the cloud point much less than the addition of the other two electrolytes. Charlton and Doherty 12 suggested that the effect of Na+ and K+ on cloud points should be identical. Thus, it is not at all surprising that the data for NaCl and KCl in Figure 8 overlap.
Although a large number of published reports have been devoted to the most accurate determination of the hydration numbers of ions by a variety of methods, 28 the reported numbers for any given ion have a wide range of values, and are dependent on the methods of measurement. For these reasons, Israelachvili stated in his book 29: “The hydration number is more of a qualitative indicator of the degree to which ions bind water rather than an exact value.” Table 1 in Reference 28 contains the hydration numbers of Li+ measured by diffraction methods in the range 4–6, or in terms of average value and standard deviation, 4.3 ± 0.7; the hydration numbers of Na+ measured by diffraction methods are in the range 4–8, or 5.6 ± 1.7; the hydration numbers of K+ measured by diffraction methods are in the range 4–8 and one measurement of 2.2 and 3.2, or after averaging 5.0 ± 1.6. Statistically, the average value of a quantity is a better representation of the quantity than any individual measurement. Therefore, it is very likely that the dehydration of the micellar surface is related to the hydration (coordination) number of the cations, because the average value of the hydration number of Li+ is obviously lower than the other two, which are statistically very close. This implies that even in the case of Li+ the hydration of the polar shell of the micelle in part depends on the cation hydration, and not just on the formation of complexes between Li+ and Triton® X-100. In summary, the statement by Molina-Bolivar et al. 11 that the hydration of the polar shell of the micelle decreases with increasing electrolyte concentration can now include Li+.
The experimental EPR data presented here have clearly shown that the hydration of the polar shell of the Triton® X-100 micelle depends strongly on the concentration and the nature of the electrolyte added to the micellar solution. An increase in the concentration of electrolyte increases the dehydration of the hydrophilic part of the Triton® X-100 molecules. The dehydration of the polar shell is much smaller in the case of Li+ than in the case of Na+ and K+, while the dehydration due to Na+ and K+ is identical. The presence of 2.5 M LiCl in the Triton® X-100 solution reduces the number of water molecules per EO group from 10 to 8, whereas the presence of the same concentration of the other two electrolytes reduces their number from 10 to 6.8. This result is reasonably well correlated to the average value of the hydration of cations calculated from literature data. Also, it was shown that the polar shell hydration is strongly correlated to the cloud point for all three electrolytes, as expected.
The authors gratefully acknowledge support from NIH Grant S06 GM48680 (BLB and MP).
Supporting Information Available: Fitting of an EPR spectrum. This material is available free of charge via the Internet at http://pubs.acs.org.