14816
J. Phys. Chem. B 2009, 113, 14816–14823
Vibrational Sum Frequency Generation Spectroscopic Investigation of the Interaction of Thiocyanate Ions with Zwitterionic Phospholipid Monolayers at the Air-Water Interface P. Viswanath,†,‡ A. Aroti,§ H. Motschmann,*,†,| and E. Leontidis*,§ Max Planck Institute of Colloids and Interfaces, Am Mu¨hlenberg 1, D14424 Golm/Potsdam, Germany, and Department of Chemistry, UniVersity of Cyprus, Nicosia 1678, Cyprus ReceiVed: July 8, 2009; ReVised Manuscript ReceiVed: September 24, 2009
Thiocyanate (SCN-) is a highly chaotropic anion of considerable biological significance, which interacts quite strongly with lipid interfaces. In most cases it is not exactly known if this interaction involves direct binding to lipid groups, or some type of indirect association or partitioning. Since thiocyanate is a linear ion, with a considerable dipole moment and nonspherical polarizability tensor, one should also consider its capability to adopt different or preferential orientations at lipid interfaces. In the present work, the interaction of thiocyanate anions with zwitterionic phospholipid monolayers in the liquid expanded (LE) phase is examined using surface pressure-area per molecule (π-AL) isotherms and vibrational sum frequency generation (VSFG) spectroscopy. Both dipalmitoyl phosphatidylcholine (DPPC) and dimyristoyl phosphatidylethanolamine (DMPE) lipids, which form stable monolayers, have been used in this investigation, since their headgroups may be expected to interact with the electrolyte solution in different ways. The π-ΑL isotherms of both lipids indicate a strong expansion of the monolayers when in contact with SCN- solutions. From the C-H stretch region of the VSFG spectra it can be deduced that the presence of the anion perturbs the conformation of the lipid chains significantly. The interfacial water structure is also perturbed in a complex way. Two distinct thiocyanate populations are detected in the CN stretch spectral region, proving that SCN- associates with zwitterionic phospholipids. Although this is a preliminary investigation of this complex system and more work is necessary to clarify certain points made in the discussion, a potential identification of the two SCN- populations and a molecular-level explanation for the observed effects of the SCN- on the VSFG spectra of the lipids is provided. 1. Introduction The specific effects of electrolytes on biological and biologically relevant systems have always attracted strong scientific interest.1-4 The lyotropic or Hofmeister series of ions, which orders ions of the same charge according to their ability for salting-out proteins from aqueous solutions, is still not entirely understood.1-6 Research on specific salt effects has recently intensified, largely because of several new experimental tools available for probing the behavior of water and ions at interfaces.7 We have recently demonstrated in a detailed and systematic way that Langmuir monolayers of zwitterionic lipids at the surface of electrolyte solutions are useful model systems to examine specific anion effects,8-11 following a long and fruitful tradition of using these systems to simulate interactions at membrane surfaces.12,13 In our previous work we concentrated on anionic effects on zwitterionic lipid monolayers, using a series of sodium salts, and discovered that, while the wellorganized, crystal-like liquid condensed (LC) phase is hardly affected by even high electrolyte concentrations, the salts affect mostly the disordered liquid expanded (LE) phase of the lipid monolayer.10 The surface pressure-molecular area (π-AL) * To whom correspondence should be addressed. E-mail:
[email protected] (H.M.);
[email protected] (E.L.). † Max Planck Institute of Colloids and Interfaces. ‡ Present address: Centre for Liquid Crystal Research, Jalahalli, Bangalore, 560013 India. § University of Cyprus. | Present address: Institute of Physical and Theoretical Chemistry, University of Regensburg, D-3040 Regensburg, Germany.
isotherms of dipalmitoyl phosphatidylcholine (DPPC) monolayers are shifted upward over electrolyte solutions compared to the DPPC isotherm over pure water. The surface pressure increment in the LE phase was subsequently modeled using a range of electrostatic models,8,9 which provide anion-lipid association parameters. Being chemical potential differences, these parameters were assumed to reflect the major changes that anions encounter upon moving from bulk water to the lipid monolayer headgroup region. They were found to correlate very well with a complex function of ionic radius, which describes both the free energy loss of ion-water interactions and the gain from a lower energy cavity at the monolayer.9 Thiocyanate (SCN-) was found to deviate considerably from this correlation, showing an affinity toward the DPPC monolayer that is much higher than what is expected on the basis of its size alone. This finding is in line with older work, which reported that SCNstrongly affects DPPC bilayers leading to lipid interdigitation.14 The strong interaction of SCN- with lipid bilayers was also verified in recent work by some of the present authors, but the nature of this interaction was not clarified.15,16 Other chaotropic anions also affect lipid monolayers and bilayers quite strongly, a situation reviewed by some of us in a recent publication.15 However, SCN- appears to be a rather special case; in addition, it offers certain important advantages for interfacial spectroscopic work (see below); hence, it will be the focus of the present investigation. Thiocyanate is an important chaotropic anion in the Hofmeister series.1-3,8,9 In contrast to other monovalent anions that are widely studied, it does not have a spherically symmetric charge distribution, possessing both a considerable dipole moment and
10.1021/jp906455k CCC: $40.75 2009 American Chemical Society Published on Web 10/13/2009
Interaction of SCN- with Phospholipid Monolayers an asymmetric polarizability.17 Given the strong chaotropic character of SCN-, and its poorly understood behavior at lipid interfaces, we have decided to further investigate the interaction between SCN- and DPPC monolayers at the molecular level, with the goal being to answer the following questions: (a) Is SCN- really concentrated at the lipid-water interface, as the π-ΑL isotherms suggest? (b) Does SCN- have a preferred orientation at the lipid-water interface, and would that orientation create specific interactions, enhancing the ion’s affinity toward the lipid monolayer? (c) Does SCN- have an effect on the conformation and/or the hydration of the lipids, and especially on the headgroup area? The latter question is particularly significant, since there is only scarce knowledge about the effects of chaotropic anions on the molecular conformations of organized lipid structures. Questions b and c above can only be addressed by surfacesensitiVe methods that provide information at the molecular scale. Surface-sensitive spectroscopies, such as second harmonic generation (SHG) or vibrational sum frequency generation (VSFG), are particularly relevant for such investigations.7,18-22 In this work we have used VSFG, which is a more specific probe and provides some distinct advantages for the present system. VSFG can in principle examine the conformation and hydration of the lipids, by measuring changes in the alkyl (2800-3000 cm-1) and carbonyl (1700-1800 cm-1) spectral regions, while it has been used to probe even the phosphate band (1050-1150 cm-1).20,21 In addition, the CN stretch frequency of SCN- itself (2000-2150 cm-1) is easily observed and provides a good opportunity to study the orientation and local interactions of the ion22-26 and to potentially distinguish between populations of ions in different “solvent” environments.27 Finally, the effect of the ion on the surface water structure can be examined by looking at the interfacial water features (3200-3800 cm-1),7,22,23,28-32 while selective deuteration of lipids and/or solvent can provide further insights.20,21 Among systems pertinent to the present study, VSFG has been applied until now to the free water surface,7,18,31,33-36 to the surfaces of electrolyte solutions,7,28,30-32,39-41 and to surfactant, macromolecule, and lipid Langmuir monolayers.18-21,25,29,40-44 Many ground-breaking VSFG studies examined the effects of electrolytes on the structure of the air-water (A-W) interface,7,28,30-32,39-41 andstogether with other methods7,45-53sthey were instrumental in establishing the presence of large chaotropic anions close to the surface, as suggested by recent computer simulations.54-59 There have been rather few applications of VSFG to the study of ionic effects on Langmuir monolayers. Knock and Bain have studied the effects of halide ions on monolayers of water-soluble cationic surfactants and concluded that the effect of the ions on the conformation of the surfactant chains is rather small.41 Watry et al. observed a considerable decrease of the intensity of the interfacial water peaks at phospholipid monolayers in the presence of NaCl.29 They attributed this finding to the decrease of the interfacial electric field through screening by the salt. More recently, in an investigation that is close in spirit to the present work, the group of Cremer studied the effects of a series of sodium salts on poly(N-isopropyl acrylamide) (PNIPAM) Langmuir monolayers.44 These authors observed large effects of the anions on the surface water structure and used the oscillator strength at 3200 cm-1 as a measure of anion adsorption at the monolayer. The Cremer group has also examined the effect of sodium salts on octadecylamine monolayers spread over D2O.43 By looking at the ratio of the intensities of the CH3-SS and CH2-SS peaks, they concluded that chaotropic anions increase the disorder of
J. Phys. Chem. B, Vol. 113, No. 44, 2009 14817 the monolayer to a considerable extent, in agreement with the Hofmeister series. This they attributed to anion penetration within the monolayers, a particularly important suggestion for the present work. The actual CN stretching mode of SCN- has been examined by Duffy et al. in a study of SCN- interaction with monolayers of tetradecyltrimethylammonium adsorbed on an octadecylthiol-gold surface.25 They concluded that SCNbinds preferentially, compared to bromide, to these monolayers and that SCN- ions do have a preferential orientation, approaching the interface with the sulfur atom. The CN stretching mode was also studied by some of the present authors in a recent study of the A-W interface of a KSCN solution.22,23 On the basis of the ratio of the observed VSF intensities of the SCNstretch under different polarizations, an averaged orientation of about 45° with respect to the surface normal was inferred, in contrast to a SHG investigation by Petersen et al.,45 who assumed a parallel orientation of the ion to the surface, on the basis of molecular dynamics simulation results. It was also speculated that the SCN- ions close to the interface are solvated so that the N-atom points toward the bulk of the solution. From the previous discussion it appears that VSFG can provide valuable insights regarding the interaction of SCN- with uncharged phospholipid monolayers. The present investigation is to our knowledge the first attempt to apply this quite powerful surface spectroscopic method to this important problem. Given the relative insensitivity of the LC phases of zwitterionic phospholipid monolayers to the presence of electrolytes, we have opted to examine the effect of NaSCN mostly on the LE phases. 2. Experimental Section 2.1. Materials. The phospholipids DPPC (1,2-dipalmitoylsn-glycerophosphocholine) and DMPE (1,2-dimyristoyl-snglycerophosphoethanolamine) were obtained from Avanti Polar Lipids (for the isotherm work) or from Sigma-Aldrich (for the VSFG work) and used without further purification. The use of phospholipids from two different suppliers resulted from the fact that isotherms were measured in Cyprus and VSFG spectra in Germany, the two laboratories having bought the chemicals from different suppliers. The issue of impurities in DPPC and DMPE is not critical for insoluble monolayer work, as it is for soluble monolayers. Our recent experience with lipids from Avanti, Aldrich, and Anatrace is that they are very pure and give practically identical isotherms over pure water at room temperature. Chloroform (p.a. grade, Merck) was used as a solvent to prepare 1 mM solutions of DPPC or DMPE. The solutions were kept at -20 °C for a maximum of 1 week. Sodium thiocyanate was purchased from Merck with purity >98%. Salt solutions were prepared using ultrapure water (specific resistance of 18.2 MΩ cm) produced by a Millipore or a Sartorius reverse osmosis unit. 2.2. Pressure-Area Isotherms of Langmuir Monolayers. Isotherm measurements were carried out with a KSV 3000 Langmuir trough (KSV Instruments) equipped with a Wilhelmy plate for the determination of the surface pressure with an accuracy of (0.01 mN m-1. The effective trough surface area was 795 cm2, and the subphase volume was 1.2 L. All experiments were performed at 22.0 ( 0.1 °C. The temperature of the subphase was maintained constant with a Julabo recirculating thermostat. DPPC (DMPE) monolayers were obtained by spreading 90-100 µL of a 1 mM chloroform solution of DPPC (DMPE) on a painstakingly cleaned water surface. After 15 min of evaporation time for the spreading solvent, the π-AL isotherms were registered while compressing the monolayers at a constant speed of 10 mm/min. Different
14818
J. Phys. Chem. B, Vol. 113, No. 44, 2009
Viswanath et al.
solvent evaporation times (10-30 min) and different compression speeds (2-10 mm/min) were used as well and were found to have no systematic effect on the isotherms. The DPPC (DMPE) isotherms were measured as many times as necessary to obtain an accurate average isotherm. For salt experiments, in each particular day, the DPPC (DMPE) isotherm over pure water was also measured as a basis for comparison and in order to check cleanliness conditions. 2.3. VSFG Experiments. The chloroform solutions of the lipids were deposited drop by drop over the subphase (Millipore water, >18.2 MΩ cm) of a Riegle-Kirsten GmBH Langmuir trough using a microsyringe. After waiting for about 15 min (allowing the solvent to evaporate), the monolayer was compressed to the desired surface pressure and equilibrated for about 10 min. We have employed vibrational sum frequency generation spectroscopy to explore the aqueous-lipid interfaces at different NaSCN concentrations. The infrared (IR) and visible (532 nm) beams overlapped both spatially and temporally at the interface (effective beam diameter ∼ 0.5 mm). When the vibrational frequency of a molecule is in resonance with the frequency of the incident IR beam, a coherent sum frequency response is generated at the interface, provided that the vibration under study is both IR- and Raman-active and that the oscillator (or collection of oscillators) does not possess a center of symmetry. The sum frequency intensity is normalized with respect to the intensities of the incident IR and visible beams. The ssp and ppp polarizations were used for the present study. The normalized sum frequency intensity is fitted to a standard equation of the following form:
Isfg ∝ |χnr + Σ Aq /(ωq - ωir + Γq)| 2 q
Figure 1. Surface pressure-area per molecule isotherms of (a) DPPC and (b) DMPE Langmuir monolayers over pure water and solutions of various concentrations of sodium thiocyanate at 295 K.
TABLE 1: Molecular Areas of DPPC Monolayers at 5, 7, and 30 mN m-1 over Water and 0.1 and 0.5 M Aqueous Solutions of NaSCN
(1)
Isfg refers to the normalized sum frequency intensity with respect to the incident IR and visible beams. χnr is the nonresonant part of the second-order susceptibility, which, in general, is a small complex quantity (here assumed to be real). Aq, ωq, and Γq are the vibrational amplitude, the resonant frequency, and the damping constant of the qth mode and ωir refers to the tunable frequency of the incident IR beam. The details of the fitting procedure are provided in the Supporting Information. The experimental setup used for VSFG is similar to the one described elsewhere.23 Briefly, the laser is used in the scanning mode. For the CN stretch of SCN- (1950-2200 cm-1) for both DPPC and DMPE monolayers the SFG signal was recorded by averaging 200 shots/point for every 2 or 3 cm-1. For the CO stretch (1650-1800 cm-1) of DPPC the SFG signal was recorded by acquiring 200 shots/point in steps of 2 and 5 cm-1. In the case of the CO stretch of the DMPE monolayer (scanning from 1650 to 1800 cm-1 again) the signal was recorded by acquiring about 75 shots/data in steps of 3 cm-1. For the water region, for both DMPE and DPPC, scanning was done from 2800 to 3800 cm-1 in steps of 3 cm-1 acquiring 75 shots/(data point). 3. Results and Discussion 3.1. Surface Pressure-Area per Molecule Isotherms. Parts a and b of Figure 1 show the π-ΑL isotherms of DPPC and DMPE for various concentrations of NaSCN in the aqueous subphase. We focus on the range of surface pressures from 0 to 10 mN/m and molecular areas from 60 to 120 Å2, which are pertinent for the present investigation. As will be seen below, VSFG measurements were carried out in monolayers at “0”, 5, and 7 mN m-1. A few measurements at 30 mN m-1 (where the
surface pressure/ (mN m-1)
area per molecule for monolayer over pure H2O/Å2
area per molecule for monolayer over 0.1 M NaSCN/Å2
area per molecule for monolayer over 0.5 M NaSCN/Å2
5 7 30
80 71 47
86 81 48.7
97.5 92 51
monolayer is in the LC phase) were also carried out for comparison. In Figure 1a it is found that at 7 mN m-1 the monolayer of DPPC over pure water is already in the phase coexistence region of the LE and LC phases. In contrast, even modest NaSCN concentrations displace the DPPC isotherm toward higher molecular areas (for a fixed pressure). As a result, at 7 mN m-1 the DPPC monolayer is in a pure LE phase in the presence of NaSCN, as ascertained by Brewster angle microscopy images (not shown). From Figure 1a we also note that when we compare properties at a specific surface pressure for water and a NaSCN solution, the comparison is really made at a different area per molecule of the lipid monolayer. Similar conclusions can be drawn from Figure 1b for DMPE monolayers in the presence of NaSCN. The molecular areas of DPPC and DMPE monolayers in the presence and absence of NaSCN for 5, 7, and 30 mN m-1 are reported in Tables 1 and 2. The DMPE monolayer is always more condensed than the DPPC monolayer at the same surface pressure and NaSCN concentration in the subphase. This is mostly due to the different headgroup structure. It is known that the carbonyl and phosphate groups present in DMPE and DPPC act as hydrogen bond acceptors, while the amine group of DMPE acts as a hydrogen bond donor. The ethanolamine headgroups associate with each other through hydrogen bonds between the ammonium and the carbonyl and phosphate groups.30,60 The hydration of the two monolayers is thus rather different.
Interaction of SCN- with Phospholipid Monolayers
J. Phys. Chem. B, Vol. 113, No. 44, 2009 14819
TABLE 2: Molecular Areas of DMPE Monolayers at 5, 7, and 30 mN m-1 over Water and 0.1 and 0.5 M Aqueous Solutions of NaSCN surface pressure/ (mN m-1)
area per molecule for monolayer over pure H2O/Å2
area per molecule for monolayer over 0.1 M NaSCN/Å2
area per molecule for monolayer over 0.5 M NaSCN/Å2
5 7 30
70 66 42
76 71.5 43.5
83 78.5 45
3.2. VSFG Results. DPPC Monolayers. To avoid confusion, we will first discuss the results obtained with the DPPC monolayers and then move to DMPE and make the appropriate comparisons. Figure 2 shows the VSFG spectrum of DPPC monolayers over pure water and on 0.5 M NaSCN solutions at
Figure 2. VSFG spectra of DPPC monolayers over pure water and on 0.5 M NaSCN at surface pressures of 5 and 7 mN m-1 under ssp polarization. The alkyl (a) and interfacial water (b) for 5 and (c) for 7 mN m-1 regions of the spectra are shown. The points and the lines represent the experimental data and the fit, respectively.
TABLE 3: AR Values (Amplitude Ratios of the CH3-SS vs the CH2-SS Peaks) for DPPC and DMPE Monolayers in the Presence and in the Absence of NaSCN at Various Surface Pressures system
surface pressure/(mN m-1)
AR values
DPPC-H2O DPPC-H2O DPPC-0.1 M NaSCN DPPC-0.5 M NaSCN DPPC-0.5 M NaSCN DMPE-H2O DMPE-H2O DMPE-0.1 M NaSCN DMPE-0.1 M NaSCN DMPE-0.1 M NaSCN DMPE-0.5 M NaSCN DMPE-0.5 M NaSCN
5 7 30 5 7 5 7 3 7 30 5 7
0.92 2.92 1.48 1.63 1.04 0.92 1.59 2.14 2.22 9.93 1.25 0.96
surface pressures of 5 and 7 mN m-1 under ssp polarization in the ranges of 2800-3000 (for the alkyl stretch, Figure 2a) and 3000-3800 cm-1 (for the water stretch, Figure 2b for 5 mN m-1 and Figure 2c for 7 mN m-1). For the alkyl region, we have used four peaks to fit the spectra under ssp polarization, viz., 2845 (CH2-SS), 2878 (CH3-SS), 2927 (CH2-AS), and 2945 cm-1 (CH3-FR).20,21,30,41-45 Furthermore, the amplitude ratio (AR) of the CH3-SS to the CH2-SS vibrations has been used as a measure to quantify the ordering of the alkyl chains23,24,29,30,41-44 in the presence and in the absence of electrolyte, a higher ratio implying a more ordered monolayer. The presence of salt decreases the C-H intensities considerably at both surface pressures. We attribute this to the increased overall monolayer disorder, which stems from the monolayer expansion at constant pressure in the presence of NaSCN (see Figure 1a and Table 1). Interestingly, while the AR value decreases in the presence of salt at 7 mN m-1 (expected behavior), the opposite appears to happen at 5 mN m-1, despite the strong monolayer expansion (see Table 3 for a collection of AR values for the DPPC and DMPE systems). We speculate that this might reflect a clustering of lipid molecules under sodium binding, as has been predicted by molecular simulation methods61-64 and revealed by AFM measurements in supported lipid bilayers,65,66 and by fluorescence spectroscopy in bilayers.62 This point will be further discussed in connection with the behavior of the interfacial water peaks and the carbonyl stretch. The interfacial water region of the spectra in Figure 2b,c have been fitted using three peaks, at 3250, 3400, and 3730 cm-1. The last of those is attributed to free or “dangling” (nonhydrogen-bonded) OH groups at the surface, and it is very weak under a lipid monolayer. The amplitudes of the other two peaks are reduced in the presence of the salt. These two peaks have originally been assigned to ice-like (at 3250 cm-1) and liquidlike (at 3400 cm-1) water molecules,34 but this assignment has recently been challenged by theory and simulation, which point out that these broad peaks probably arise from collective vibrations with strong intermolecular components.67,68 These interfacial water peaks have been the subject of numerous investigations and have been examined in detail at the surfaces of aqueous electrolyte solutions, proving that the ions affect the interfacial water structure.7,31-33,35,37-40 In our case, the intensity of the “bonded” water peaks at 5 mN m-1 (Figure 2b) decreases considerably in the presence of NaSCN. This could either mean a decreased water polarization due to a reduction of the local electric field or a decreased number of molecules contributing to the signal. Both interpretations could be valid here. Na+ and SCN- ions adsorb at the lipid monolayer. They
14820
J. Phys. Chem. B, Vol. 113, No. 44, 2009
Viswanath et al.
Figure 4. VSFG spectra of the DPPC monolayer over 0.5 M NaSCN solutions probing the CN stretch (1950-2250 cm-1) at the surface pressures of “0”, 5, and 7 mN m-1 under ssp polarization. The points and the lines represent the experimental data and the fit, respectively.
Figure 3. VSFG spectra of DPPC monolayers over (a) pure water and (b) a 0.5 M NaSCN solution probing the carbonyl stretch (1650-1800 cm-1) at surface pressures of “0”, 5, and 7 mN m-1 under ssp polarization. The points and the lines represent the experimental data and the fit, respectively.
might change the local potentials felt by the interfacial water molecules, promoting a more centrosymmetric orientation, or they might just push water molecules away from the interface by “crowding” it. The intensities of the bonded water at 7 mN m-1 are not reduced significantly by the presence of NaSCN, even though the previous mechanisms that would lead to reduction should still be operational. We speculate that the expansion of the monolayer and some lipid clustering may produce a countereffect. Figure 3 contains the carbonyl stretch peaks for DPPC monolayers obtained under ssp polarization over water (Figure 3a) and 0.5 NaSCN (Figure 3b). No peaks can be recorded under ppp polarization. The intensities of the peaks are severely reduced in the presence of the salt. The CdO stretch region of phospholipid monolayers has not attracted much attention in the literature, in contrast to the CdO stretch of carboxylic acid monolayers.69,70 The intensities in Figure 3a are roughly independent of the surface pressure (or the molecular area) over pure H2O. Intensities are much lower in the presence of NaSCN, probably because of the monolayer expansion, but the fact that the observed signal decreases with surface pressure in Figure 3b must be attributed to changes in conformation upon ion adsorption. Recent computer simulations claim that sodium binds to the carbonyl groups of DPPC bilayers.61-64 Could this be the reason for the decrease of the signal? The VSFG spectra of the CN stretch of the SCN- ions under DPPC monolayers under ssp polarization are shown in Figure 4. Two major CN peaks, at 2065 cm-1 and ca. 2130 cm-1, are observed under both ssp and ppp polarization (see Figure S1 in Supporting Information for the latter). The intensities of these
Figure 5. VSFG spectra of DPPC monolayers over 0.1 M NaSCN solution probing the CN stretch (1950-2250 cm-1) at the surface pressures of 5 and 30 mN m-1 under ssp polarization. The points and the line represent the data and the fit, respectively.
peaks are weakly dependent on surface pressure in the range of 0-7 mN m-1. In our previous work, which examined the SCNVSFG spectrum at the surface of KSCN solutions, we detected only one peak at 2065 cm-1. Figure 4 therefore implies the existence of two distinct populations of SCN- ions under the DPPC monolayers. It is tempting to identify the peak at 2065 cm-1 as deriving from molecules that have a similar orientation to that found at the free water surface. Further clarification about the nature of these two peaks is given in Figure 5, where we plot the SCN- spectrum for DPPC over 0.1 M NaSCN at 5 and 30 mN m-1. In Figure 5 we observe that as the surface pressure increases, the intensity at 2065 cm-1 decreases, while that at 2130 cm-1 increases considerably. Since at 30 mN m-1 the DPPC monolayer over 0.1 M NaSCN issto a very large extentsin the LC phase, we conclude that this peak corresponds to SCN- ions located under an organized region of lipid headgroups. Why is then this peak present even at very low surface pressures? We speculate that this is again due to lipid clustering, which creates small clusters with locally organized headgroups. SCN- peaks at frequencies higher than 2100 cm-1 were reported by Tadjeddine and Guyot-Sionnest25 and Bain and co-workers,26,27 who used VSFG to study the adsorption of the ion at noble metal electrode surfaces. SCN- adsorbs at these interfaces with the sulfur atom pointing toward the electrodes, and the CN stretch frequency increases almost linearly with the electrode potential (higher for more positive potentials). It is possible that the population of SCN- ions with frequencies at
Interaction of SCN- with Phospholipid Monolayers
J. Phys. Chem. B, Vol. 113, No. 44, 2009 14821
Figure 6. SCN- adsorption under a lipid monolayer gives rise to two distinct populations that provide two peaks in the VSFG spectra.
2135 cm-1 experience a positive electric potential being closer to the choline headgroup of the lipids. The picture that we are envisioning to explain the SCNspectra is presented in Figure 6, where we speculate on the interfacial structure at low and high surface pressures. According to this figure, sodium ions induce or enhance lipid clustering at low surface pressures. At high surface pressures most SCNions are lying under the organized monolayer. It would be interesting to measure the VSFG spectrum at even higher surface pressures, to check if the peak at 2065 cm-1 would gradually disappear, but this was not done here. 3.3. VSFG Results. DMPE Monolayers. Figure 7 contains the alkyl and interfacial water regions of the VSFG spectrum for DMPE monolayers under ssp polarization in the absence and in the presence of 0.5 M NaSCN. The conclusions drawn from this figure are similar to those for the DPPC system. The alkyl intensities are strongly reduced in the presence of NaSCN as a result of monolayer expansion, but the AR values (Table 3) at 5 mN m-1 increase in the presence of salt, implying some type of local chain ordering. This is also observed for 0.l M NaSCN up to 7 mN m-1. The interfacial water peaks are strongly reduced in the presence of salt at both 5 and 7 mN m-1 (Figure 7b,c), and the effect appears to be stronger than for DPPC, although the areas per molecule of the DMPE monolayer are smaller for the same surface pressure. The carbonyl region of the spectrum is not qualitatively similar to that for DPPC and is presented in Figure 8. There is no signal decrease in the presence of the ionssin fact there is a rather interesting increase of the signal at very low surface pressures. This may be attributed to the fact that the DMPE monolayer exhibits considerable lipid clustering through H-bonding even at low surface pressures: hence, the presence of the ions barely affects the carbonyl conformations. Finally, the two different SCNpopulations are also observed at this monolayer as well (Figure 9 and Figure S2 in Supporting Information). The most significant difference here is that the low-frequency peak appears at 2050 cm-1 compared to the value of 2065 cm-1 at the DPPC monolayer. Despite this difference, we believe that the lowfrequency peaks in both monolayers reflect similar environments, not widely different from that at the free water surface. Interestingly, the -NH stretch (occurring at ∼3085 cm-1), which has also been reported earlier,44,71 is detected in the case of DMPE on pure water but disappears in the presence of 0.5 M NaSCN (see Figure 7b,c), suggesting a possible interaction of the positively charged amine group of DMPE with SCN-. This may explain the frequency shift for the population of SCNions with the lower CN frequency. 4. Conclusions The VSFG measurements have shown that SCN- is present at the lipid-water interface of DPPC and DMPE Langmuir
Figure 7. VSFG spectra of DMPE monolayers over pure water and on 0.5 M NaSCN solutions probing the alkyl (a) and amine and interfacial water (b) for 5 and (c) for 7 mN m-1 under ssp polarization. The points and the lines represent the experimental data and the fit, respectively.
monolayers. Two well-resolved peaks in the CN stretching spectrum reveal the existence of two distinct populations of anions at the interface. By examining the evolution of these populations as the surface pressure increases, we were able to identify the peak at 2135 cm-1 as derived from ions that are probably lying “below” the lipid head groups. The other population, which gives a signal centered at 2065 cm-1 for DPPC and 2050 cm-1 for DMPE, is thought to consist of interfacial anions in a hydration environment similar to that of the free air-water surface, although the lower frequency in the case of DMPE also suggests a more direct interaction of the ions with the amine group of DMPE, which is also supported
14822
J. Phys. Chem. B, Vol. 113, No. 44, 2009
Viswanath et al. AR despite the increase in molecular area at 5 mN m-1 and also as a considerable change in the carbonyl stretch intensities. We speculate that clustering may also create some more space at the interface for water molecules. We have thus demonstrated that VSFG is a powerful tool to understand the complex interactions of ions with soft lipid interfaces. Despite the large number of data collected, we still view this as a preliminary investigation, since many interesting questions are left open, and some of the discussion is certainly speculative. In particular, a more complete investigation will allow a good estimate of the average SCN- orientations of the two populations that we have detected. The nature of the two populations can be further ascertained by working at more extreme surface pressures. The selective deuteration of the chains or of the choline headgroup of the lipid or using heavy water (D2O) as the subphase could provide useful information, especially under additional polarization measurements. We hope however that the present work will provide impetus for further studies of anion interactions with lipids, both experimental and computational. The relevant recent computer simulations61-64 address mostly the lipid bilayer geometry and do not focus on complex but important chaotropic anions such as ClO4- or SCN-.
Figure 8. VSFG spectra of DMPE monolayers over (a) pure water and (b) a 0.5 M NaSCN solution probing the carbonyl stretch (1650-1800 cm-1) at the surface pressures of “0”, 5, and 7 mN m-1 under ssp polarization. The points and the lines represent the experimental data and the fit, respectively.
Acknowledgment. A.A. and E.L. would like to thank the University of Cyprus for a generous 3 year internal research grant (2005-2007) that enabled them to carry out this work. P.V. is grateful to the French-German network program on “Complex Fluids: From 2D to 3D” for a postdoctoral research fellowship. We thank Prof. H. Mo¨hwald for enlightening discussions, constant support, and encouragement. Supporting Information Available: Description of the fitting procedure that we used to analyze the VSFG spectra and two figures containing the ppp polarization spectra for the thiocyanate band for the DPPC and the DMPE monolayer. This information is available free of charge via the Internet at http:// pubs.acs.org. References and Notes
Figure 9. VSFG spectra of DMPE monolayers over 0.5 M NaSCN solutions probing the CN stretching region (1950-2250 cm-1) at the surface pressures of “0”, 5, and 7 mN m-1 under ssp polarization. The points and the lines represent the experimental data and the fit, respectively.
by the disappearance of the NH intensity in the presence of NaSCN. This is in fact a distinct difference between the monolayers of the two lipids. It was found that the electrolyte affects both the interfacial water structure and the lipid chain ordering. In general the water peaks are considerably reduced in intensity, and we believe that this is mostly due to the “condensation” of ions at the surface. Chain ordering also decreases in most cases, because of the monolayer expansion upon salt adsorption, which leads to higher areas per lipid molecule. However, we have obtained some evidence for increased local chain order at low salt concentration, especially for the DPPC interface. This manifests itself as an increased
(1) Collins, K. D.; Washabaugh, M. W. Q. ReV. Biophys. 1985, 18, 323. (2) Cacace, M. G.; Landau, E. M.; Ramsden, J. J. Q. ReV. Biophys. 1997, 30, 241. (3) Baldwin, R. L. Biophys. J. 1996, 71, 2056. (4) Collins, K. D. Biophys. Chem. 2006, 119, 271. (5) Ninham, B. W. AdV. Colloid Interface Sci. 1999, 83, 1. (6) Kunz, W.; Lo Nostro, P.; Ninham, B. W. Curr. Opin. Colloid Interface Sci. 2004, 9, 1. (7) Thematic issue. Chem. ReV. 2006, 106, 1137. (8) Aroti, A.; Leontidis, E.; Maltseva, E.; Brezesinski, G. J. Phys. Chem. B 2004, 108, 15238. (9) Aroti, A.; Leontidis, E.; Dubois, M.; Zemb, T.; Brezesinski, G. Colloids Surf., A 2007, 303, 144. (10) Leontidis, E.; Aroti, A.; Belloni, L. J. Phys. Chem. B 2009, 113, 1447. (11) Leontidis, E.; Aroti, A. J. Phys. Chem. B 2009, 113, 1460. (12) Kaganer, V. M.; Mo¨hwald, H.; Dutta, P. ReV. Mod. Phys. 1999, 71, 779. (13) Brezesinski, G.; Mo¨hwald, H. AdV. Colloid Interface Sci. 2003, 100-102, 563. (14) Cunningham, B. A.; Lis, L. J. Biochim. Biophys. Acta 1986, 861, 237. (15) Aroti, A.; Leontidis, E.; Dubois, M.; Zemb, T. Biophys. J. 2007, 93, 1580. (16) Leontidis, E.; Aroti, A.; Belloni, L.; Dubois, M.; Zemb, T. Biophys. J. 2007, 93, 1591. (17) Serr, A.; Netz, R. R. Int. J. Quantum Chem. 2006, 106, 2960. (18) Richmond, G. L. Chem. ReV. 2002, 102, 2693. (19) Eisenthal, K. B. Chem. ReV. 1996, 96, 1343. (20) Ma, G.; Allen, H. C. Langmuir 2006, 22, 5341. (21) Ma, G.; Allen, H. C. Langmuir 2007, 23, 589. (22) Viswanath, P.; Motschmann, H. J. Phys. Chem. C 2007, 111, 4484.
Interaction of SCN- with Phospholipid Monolayers (23) Viswanath, P.; Motschmann, H. J. Phys. Chem. C 2008, 112, 2099. (24) Tadjeddine, A.; Guyot-Sionnest, P. Electrochim. Acta 1991, 36, 1849. (25) Ong, T. H.; Davies, P. B.; Bain, C. D. J. Phys. Chem. 1993, 97, 12047. (26) Duffy, D. C.; Ward, R. N.; Davies, P. B.; Bain, C. D. J. Am. Chem. Soc. 1994, 116, 1125. (27) Schultz, P. W.; Leroi, G. E.; Popov, A. I. J. Am. Chem. Soc. 1996, 118, 10617. (28) Gragson, D. E.; Richmond, G. L. J. Phys. Chem. B 1998, 102, 3847. (29) Raymond, E. A.; Richmond, G. L. J. Phys. Chem. B 2004, 108, 5051. (30) Watry, M. R.; Tarbuck, T. L.; Richmond, G. L. J. Phys. Chem. B 2003, 107, 512. (31) Shultz, M. J.; Schnitzer, C.; Simonelli, D. Int. ReV. Phys. Chem. 2000, 19, 123. (32) Gopalakrishnan, S.; Liu, D.; Allen, H. C.; Kuo, M.; Shultz, M. J. Chem. ReV. 2006, 106, 1155. (33) Liu, D.; Ma, G.; Levering, L. M.; Allen, H. C. J. Phys. Chem. B 2004, 108, 2252. (34) Du, Q.; Superfine, R.; Freysz, E.; Shen, Y. R. Phys. ReV. Lett. 1993, 70, 2313. (35) Shultz, M. J.; Baldelli, S.; Schnitzer, C.; Simonelli, D. J. Phys. Chem. B 2002, 106, 531. (36) Allen, H. C.; Raymond, E. A.; Richmond, G. L. J. Phys. Chem. A 2001, 105, 1649. (37) Schnitzer, C.; Baldelli, S.; Campbell, D. J.; Shultz, M. J. J. Phys. Chem. A 1999, 103, 6383. (38) Schnitzer, C.; Baldelli, S.; Shultz, M. J. J. Phys. Chem. B 2000, 104, 585. (39) Mucha, M.; Frigato, T.; Levering, L. M.; Allen, H. C.; Tobias, D. J.; Dang, L. X.; Jungwirth, P. J. Phys. Chem. 2005, 109, 7617. (40) Gopalakrishnan, S.; Jungwirth, P.; Tobias, D. J.; Allen, H. C. J. Phys. Chem. 2005, 109, 8861. (41) Guyot-Sionnest, P.; Hunt, J. H.; Shen, Y. R. Phys. ReV. Lett. 1987, 59, 1597. (42) Knock, M. M.; Bain, C. D. Langmuir 2000, 16, 2857. (43) Bonn, M.; Roke, S.; Berg, O.; Juurlink, L. B. F.; Stamouli, A.; Mu¨ller, M. J. Phys. Chem. B 2004, 108, 19083. (44) Gurau, M. C.; Lim, S.-M.; Castellana, E. T.; Albertorio, F.; Kataoka, S.; Cremer, P. S. J. Am. Chem. Soc. 2004, 126, 10522. (45) Petersen, P. B.; Saykally, R. J.; Mucha, M.; Jungwirth, P. J. Phys. Chem. B 2005, 109, 10915.
J. Phys. Chem. B, Vol. 113, No. 44, 2009 14823 (46) Chen, X.; Yang, T.; Kataoka, S.; Cremer, P. S. J. Am. Chem. Soc. 2007, 129, 12272. (47) Cheng, J.; Hoffmann, M. R.; Colussi, A. J. J. Phys. Chem. B 2008, 112, 7157. (48) Petersen, P. B.; Saykally, R. J. Chem. Phys. Lett. 2004, 397, 51. (49) Petersen, P. B.; Johnson, J. C.; Knutsen, K. P.; Saykally, R. J. Chem. Phys. Lett. 2004, 397, 46. (50) Petersen, P. B.; Saykally, R. J. J. Phys. Chem. B 2005, 109, 7976. (51) Winter, B.; Weber, R.; Schmidt, P. M.; Hertel, I. V.; Faubel, M.; Vrbka, L.; Jungwirth, P. J. Phys. Chem. B 2004, 108, 14558. (52) Bo¨hm, R.; Morgner, H.; Oberbrodhage, J.; Wulf, M. Surf. Sci. 1994, 317, 407. (53) Ghosal, S.; Hemminger, J. C.; Bluhm, H.; Mun, B. S.; Hebenstreit, E. L. D.; Ketteler, G.; Ogletree, D. F.; Requejo, F. G.; Salmeron, M. Science 2005, 307, 563. (54) Perera, L.; Berkowitz, M. L. J. Chem. Phys. 1991, 95, 1954. (55) Stuart, S. J.; Berne, B. J. J. Phys. Chem. 1996, 100, 11934. (56) Jungwirth, P.; Tobias, D. J. J. Phys. Chem. B 2001, 105, 10468. (57) Jungwirth, P.; Tobias, D. J. J. Phys. Chem. B 2002, 106, 6361. (58) Dang, L. X.; Chang, T.-M. J. Phys. Chem. B 2002, 106, 235. (59) Archontis, G.; Leontidis, E.; Andreou, G. J. Phys. Chem. B 2005, 109, 17957. (60) Yeagle, P. L. Acc. Chem. Res. 1978, 11, 321. (61) Pandit, S. A.; Bostick, D.; Berkowitz, M. L. Biophys. J. 2003, 84, 3743. (62) Böckmann, R. A.; Hac, A.; Heimburg, T.; Grubmu¨ller, H. Biophys. J. 2003, 85, 1647. (63) Gurtovenko, A. A.; Vattulainen, I. J. Phys. Chem. B 2008, 112, 1953. (64) Cordomi, A.; Edholm, O.; Perez, J. J. J. Phys. Chem. B 2008, 112, 1397. (65) Garcia-Manyes, S.; Oncins, G.; Sanz, F. Biophys. J. 2005, 89, 1812. (66) Oncins, G.; Garcia-Manyes, S.; Sanz, F. Langmuir 2005, 21, 7373. (67) Auer, B. M.; Skinner, J. L. J. Phys. Chem. B 2009, 113, 4125. (68) Noah-Vanhoucke, J.; Smith, J. D.; Geissler, P. L. J. Phys. Chem. B 2009, 113, 4065. (69) Miranda, P. B.; Du, Q.; Shen, Y. R. Chem. Phys. Lett. 1998, 286, 1. (70) Tang, C. Y.; Allen, H. C. J. Phys. Chem. A 2009, 113, 7383. (71) Gragson, D. E.; McCarty, B. M.; Richmond, G. L. J. Am. Chem. Soc. 1997, 119, 6144.
JP906455K
