5256
Langmuir 2006, 22, 5256-5260
Effect of Sodium Dodecyl Sulfate at Different Hydration Conditions on Dioleoyl Phosphatidylcholine Bilayers Studied by Grazing Incidence X-ray Diffraction J. Pereira-Lachataignerais,*,† R. Pons,† H. Amenitsch,‡ M. Rappolt,‡ B. Sartori,‡ and O. Lo´pez† Departamento de Tecnologı´a de TensioactiVos, Instituto de InVestigaciones Quı´micas y Ambientales de Barcelona (IIQAB), Consejo Superior de InVestigaciones Cientı´ficas (CSIC), Calle Jordi Girona 18-26, 08034 Barcelona, Spain, and Institute of Biophysics and X-ray Structure Research, Austrian Academy of Sciences, Schmiedlstrasse 6, 8042 Graz, Austria ReceiVed NoVember 28, 2005. In Final Form: April 6, 2006 The effect of the surfactant content and hydration conditions in the phases of dioleoyl phosphatidylcoline (DOPC)/ sodium dodecyl sulfate (SDS) mixtures was studied. To this end, surface X-ray diffraction experiments have been performed on bilayers of the mixtures deposited on hydrophobic silicon wafers by dip coating. To investigate the effect of relative humidity (RH) on bilayer organization, a humidity chamber with dry-wet air control was used, and RH values were fixed between 1 and 65%. Our results showed, in addition to the lamellar phase, a rhombohedral phase in mixtures at low hydration conditions (RH < 30%). The d spacing between lamellae increased with the RH and SDS content. This fact could be associated with a swelling effect that is probably due to the localization of water molecules between the polar headgroups of the DOPC and SDS forming the bilayers. The electron-density profiles calculated by Fourier reconstruction of the lamellar stacking for the different samples also confirmed this fact. In addition, the increase in d spacing could be related to the increase in the hydrophilic character of the mixture when the SDS content increases. The rhombohedral phase was more clearly observed in mixtures with high SDS content. Thus, the stalk structure of the rhombohedral phase could be facilitated because of the SDS contribution to inverse structures.
Introduction Lipids form thermodynamically stable bilayers that are the universal basis for membrane structure. For this reason, there is a great deal of interest the study of physical properties of phospholipid bilayers. Supported bilayers allow the study of membrane constituents in a controlled environment.1-3 Thus, experimental approaches have produced an abundance of structural data on lipid bilayers using X-ray scattering.4,5 Phospholipids usually bind considerable amounts of water at their headgroups, often establishing a liquid-crystalline lamellar arrangement known as the LR phase. This phase consists on an alternating stack of planar, undisrupted bilayers and intervening layers of solvent. However, lipid hydration is known to affect not only the molecular interactions in the headgroup region but also the organization and motions of the acyl chains comprising the hydrocarbon core (HC).6 One important characteristic of lipid bilayers is the phase transition during membrane fusion. The formation of solvent-filled intralamellar defects in lamellar phases may be regarded as being related to the formation of the so-called intermediate phases. The first intermediate state of two fusing lipid bilayers is the merging of two opposite monolayers, * Corresponding author. E-mail:
[email protected]. Tel: (34-93) 400 61 00 ext 301. Fax: (34-93) 204 59 04. † Instituto de Investigaciones Quı´micas y Ambientales de Barcelona (IIQAB). ‡ Austrian Academy of Sciences. (1) Plant, A. L. Langmuir 1993, 9, 2764-2767. (2) Sackmann, E. Science 1996, 271, 43-48. (3) Groves, J. T.; Boxer, S. G. Acc. Chem. Res. 2002, 35, 149-157. (4) Wiener, M. C.; White, S. H. Biophys. J. 1992, 61, 428-433. (5) Hristova, K.; White, S. H. Biophys. J. 1998, 74, 2419-2433. (6) Ge, M.; Freed, J. H. Biophys. J. 1998, 74, 910.
resulting in an hourglass-shaped structure called a stalk (Figure 1). This structure is involved in the 3D rhombohedral phase, which was first reported by Luzzati and exists only in a very narrow region of the phase diagram.7 In a recent publication, Yang et al.8 showed the observation of a rhombohedral phase in a pure DOPC system at reduced hydration (RH < 45%). In previous papers, we investigated the solubilization of liposomes by surfactants from a structural viewpoint.9,10 Different intermediate lipid-surfactant phases in the solubilization process were studied by time-resolved small-angle X-ray scattering (SAXS) using a stopped flow cell and synchrotron radiation.11 This technique has been demonstrated to be a very good method for studying the phase behavior of other lipid-surfactant systems and for monitoring very fast biological processes.12,13 In this work, we study systems formed by pure DOPC and lipidsurfactant (DOPC/SDS) mixtures. To study the lipid-surfactant interaction in more detail, a highly organized multilayer system was prepared. Thus, aligned multilayers of lipid formed on silicon wafers with different SDS contents were studied by grazing incidence X-ray diffraction. We have analyzed the phase behavior at several relative humidity values. At low-humidity conditions, (7) Luzzati, V.; Gulik-Krzywicki; T., Tardieu, A. Nature 1968, 218, 10311034. (8) Yang, L.; Huang, H. W. Biophys. J. 2003, 84, 1808-1817. (9) Lo´pez, O.; de la Maza, A.; Coderch, L.; Lo´pez-Iglesias; C.; Wehrli; E.; Parra, J. L FEBS Lett. 1998, 426, 314-318. (10) Lo´pez, O.; Co´cera, M.; Wehrli, E.; Parra, J. L.; de la Maza, A. Arch. Biochem. Biophys. 1999, 367, 153-160. (11) Co´cera, M.; Lo´pez, O.; Pons, R.; Amenitsch, H.; de la Maza, A. Langmuir 2004, 20, 3074-3079. (12) Laggner, P.; Kriechbaum, M. Chem. Phys. Lipids 1991, 57, 121-145. (13) Woenckhaus, J.; Ko¨hling, R.; Thiyagarajan, P.; Littrell, K. C.; Seifert, S.; Royer, C. A.; Winter, R. Biophys. J. 2001, 80, 1518-1523.
10.1021/la053207k CCC: $33.50 © 2006 American Chemical Society Published on Web 05/10/2006
Surface X-ray Difraction on Lipid Bilayers
Langmuir, Vol. 22, No. 12, 2006 5257
Figure 1. Model of the stalk structure (A). Example of stalk structure for DOPC and for DOPC and SDS (B).
a 3D rhombohedral phase was detected. The behavior of the system with increasing SDS content was also studied.
The electron-density profiles were derived by standard Fourier reconstruction h)m
Materials and Methods Lipid Bilayer Preparation. Solutions of pure DOPC and DOPC/ SDS mixtures were prepared in 2:1 chloroform/methanol. Sample deposition was carried out on silicon wafers that were previously cleaned as described by Amenitsch et al.14 Lipid and lipid/surfactant multilayers were deposited by dipping the silicon wafer into the solutions and pulling them out at a constant speed of 1.5 mm/s. This process was controlled through LabView software and using equipment made by the SAXS beamline laboratory personal at the ELETTRA synchrotron light source. After the dip-coating procedure, the wafers were kept under vacuum for 24 h at room temperature in order to remove any residual solvent contamination. The samples were measured at three degrees of hydration: 1, 30, and 65% relative humidity (RH). This procedure was made possible by using a humidity chamber with dry-wet air control where the RH was measured using a Keithley 195A digital multimeter electrode. The samples were kept at constant RH for at least 30 min before making measurements. Surface X-ray Diffraction Measurements. All of the measurements were made at the SAXS beamline of the ELETTRA synchrotron light source (Trieste, Italy). X-ray diffraction measurements were carried out using a model FR591 rotating anode Mar research X-ray generator at an energy of 55 kV and 4.5 kW at room temperature (T ) 26°C). The sample-detector distance was fixed at 200 mm, and the exposure time was between 100 and 900 s. Silver behenate was used to calibrate the angular scale of the measured intensity. The samples were measured with a scanning angle of 6° from the alignment angle. The X-ray diffraction patterns were recorded by reflection on an area detector; the incident angle was 0.5° relative to the sample. X-ray Data Recorder, Analysis, and Electron-Density Reconstruction. The 2D images were analyzed using FIT2D software (available at http://www.esrf.fr/computing/scientific/FIT2D/). The intensities along the main reflection were integrated. The settings used correct for the geometrical transformation from a flat detector to the desired to 2θ plot. The q scale was obtained from silver behenate calibration (d spacing ) 58.38 Å). All diffraction peaks were fitted to a Gaussian,and from these fits we obtained values for both the d spacing and the amplitude. From those d spacing values, disregarding the values corresponding to the first reflection, we have estimated an error of 0.06 nm for a confidence of 95%. (14) Amenitsch, H.; Rappolt, M.; Teixeira, C. V.; Majerowicz, M.; Laggner, P. Langmuir 2004, 20, 4621-4628.
∆F(x) )
( ) h
∑ (F cos 2πxd h
h)1
where h is the order of reflection, Fh is the form factor of the hth Bragg peak, and d is the lamellar repeat distance. Form factors are given by the square root of the product of h and the integrated areas of the respective Lorentz and background-corrected Bragg peaks. The phase information for each diffraction order is either +1 or -1 for a centrosymmetric electron-density profile such as that for a lipid bilayer. To determine the correct phasing, we calculated all possible combinations of phases. Half of these combinations are symmetrical with the other half. The eight remaining possibilities (those that have a lower electronic density at the center of the bilayer than at the edge) were examined. This complete information can be found as Supporting Information for pure DOPC in Figure S1. According to this, we found that the best phasing choice for the LR phase is (- - + -). This assignation is based on (1) the similarity in the hydrocarbon chain region with previous studies related to the LR phase of dioleoyl phosphatidyl choline5 and (2) the agreement with structural parameters of the DOPC bilayer reported by Wiener and White.4 The other phase combinations lead to improper structural features such as a hydrocarbon core that is too large, a missing methyl trough, and a headgroup size that is too small. In addition, other authors also reported this combination of phases for phospholipid bilayer structure.15 The same phase combination was used for the DOPC/SDS mixed samples because of the same analysis. (The example for DOPC/SDS 127/55.2 mM can be found as Supporting Information, Figure S2.) Different methods of SAXS evaluation are available in the literature that can be applied to oriented lamellar systems. These analysis approaches allow one to obtain refined structural parameters from the SAXS curves such as the degree of orientation and the type of lamellar stacking,15,16,20 membrane elasticity,17 lattice distortion, and preferred orientation.18 In this work in which the lamellar stacking of rhombohedral structure was studied, Fourier reconstruction was used. (15) Winter, I.; Pabst, G.; Rappolt, M.; Lohner, K. Chem. Phys. Lipids 2001, 112, 137-150. (16) Ruland, W.; Smarsly, B. J. Appl. Crystallogr. 2004, 37, 575-584. (17) Pabst, G.; Rappolt, M.; Amenitsch, H.; Laggner, P. Phys ReV. E 2000, 62, 4000-4009. (18) Smarsly, B.; Gibaud, A.; Ruland, W.; Sturmayr, D.; Brinker, C. J. Langmuir 2005, 21, 3858-3866. (19) Yang, L.; Ding, L.; Huang, H. W. Biochemistry 2003, 42, 6631-6635. (20) Rappolt, M.; Amenitsch, H.; Strancar, J.; Teixeira, C. V.; Kriechbaum, M.; Pabst, G.; Majerowicz, M.; Laggner, P. AdV. Colloid Interface Sci. 2004, 111, 63-77.
5258 Langmuir, Vol. 22, No. 12, 2006
Pereira-Lachataignerais et al.
Figure 2. Unit cell parameter a as a function of the mole fraction of SDS.
Figure 3. Diffraction pattern of the rhombohedral phase of DOPC/ SDS (127/55.2) at RH ) 1%. BP (1) and BP (2) correspond to the first two orders of Bragg peaks.
Results and Discusion Phase Behavior as a Function of RH and SDS Content. The X-ray reflection patterns were obtained at three degrees of hydration (1, 30, and 65%). A lamellar phase (LR) was observed in all of the samples at RH ) 65% and was characterized by a single series of Bragg peaks. A 3D rhombohedral phase (R) was identified in the diffraction patterns obtained at RH ) 1 and 30% for all of the samples. The characteristic diffraction pattern of the LR and R lattices can be observed in Figures 3 and 4. The R phase was characterized on DOPC by Yang et al., who described a diffraction pattern that agrees with our results. This group studied the phase behavior of pure DOPC in the range of RH from 35 to 90% and at temperatures from 25 to 65 °C. Under these conditions, they reported the existence of lamellar and rhombohedral phases being the last present only at low humidity. Our result corroborates that the R phase does not exist at high hydration.19,20 The R phase involves the presence of stalk structure in the sample (depicted in Figure 1A), which is associated with an intermediate state during membrane fusion and pore formation.8 According to our results, the stalk motifs could be stabilized at low and controlled humidity conditions. To compare the behavior of the R phase for different samples, the specific cell parameter of the rhombohedral lattice was calculated from the diffraction pattern following the method described by Rappolt et al.20 This parameter a is associated with the distance between pores of the stalk structure. Figure 2 shows the variation of this parameter as a function of SDS mole fraction at RH ) 1 and 30%. It can be observed that there is a notable increase in parameter a from pure DOPC to DOPC/SDS samples at RH ) 1%, whereas the cell parameter a has an almost constant
Figure 4. Diffraction patterns of the rhombohedral phase (RH ) 1 and 30%) and lamellar phase (RH ) 65%) for pure DOPC (A, B, and C) 127 mM, DOPC/SDS (D, E, and F) 127/0.69, and DOPC/ SDS (G, H, and I) 127/55.2.
value at RH ) 30%. An increase in unit cell parameter a is detected from RH ) 1 to 30% for pure DOPC, which can be attributed to the water hydration of the headgroup. Rappolt et al. reported the dimensions of the unit cell of the R phase in DOPC, where the parameter a corresponds approximately to 67 Å at RH ) 35%,20 which is close to our value a ) 64.2 Å for DOPC at RH ) 30%. In the presence of SDS, the opposite occurssthe cell parameter a is smaller for 30% hydration than for 1%. At RH ) 30%, SDS does not seem to alter the critical packing parameter in favor of the headgroup. However, at RH ) 1%, the SDS seems to increase the unit cell parameter a, which could mean that the effective headgroup size is changing dramatically. That is, the interaction is mainly with the headgroup and has only a minor influence on the hydrocarbon tail region. However, this increase is not linear, as shown in Figure 2, and it seems that there is a limiting SDS content that holes of the R phase could admit. Figure 1B shows the possible stalk structures proposed for DOPC and that for DOPC/SDS molecules. Figure 4 shows the diffraction patterns of pure DOPC and DOPC/SDS: 127/0.69 and 127/55.2 samples at RH ) 1, 30, and 65%. It can be observed that the R pattern is stronger at RH ) 1% (Figure 4A, D, and G) than at RH ) 30%, which was expected because this phase is stabilized at low hydration conditions.8 In Figure 4, we can also observe that the diffraction peaks of the R phase become sharper with increasing SDS content. This is an unexpected result taking into account the hydrophilic character
Surface X-ray Difraction on Lipid Bilayers
Langmuir, Vol. 22, No. 12, 2006 5259 Table 1. d-Spacing Values at Different RH and SDS Contenta molar ratios (mM)
RH ) 1% d(Å)
d(Å)
127/0 127/0.69 127/6.9 127/27.6 127/41.4 127/55.2
40.5 41.5 40.9 41.0 41.7 41.5
43.7 43.7 43.6 43.5 44.4 44.1
RH ) 30% dB(Å) dW(Å) 34.5 35.6 34.9 33.5 35.4 36.0
9.2 8.1 8.7 10.0 9.0 8.1
d(Å) 47.8 47.9 47.5 47.5 48.5 47.9
RH ) 65% dB(Å) dW(Å) 36.4 38.1 36.4 35.8 37.5 38.0
11.4 9.8 11.1 11.7 11.0 9.9
a d, lamellar repeat distance; dB, thickness of the bilayer region; and dW, thickness of the interlamellar water layer.
Figure 5. X-ray diffractograms of DOPC 127 mM (A) and DOPC/ SDS 127/55.2 mM (B). From bottom to top, samples are measured at RH ) 1, 30, and 65%, respectively.
of the SDS molecule. The increase in SDS content would increase the hydrophilic character of the bilayer. Thus, it would be reasonable to assume that an increase in the hydrophilic character of the bilayer could have the same effect as an increase in the RH (i.e., a decrease in the R phase); nevertheless, this effect was not detected. We should consider that at low hydration the surfactant dissociation is low and, therefore, acts as a low-polarity molecule. Analyzing the stalk structure (Figure 1), we can observe the contribution of an inverse structure.21 The incorporation of SDS molecules in the DOPC bilayers could be related to the formation of this inverse structure. Given that SDS is able to form inverse micelles at very low hydration conditions, the formation of the inverse structure of stalk motifs could be stabilized and potentiated with the increase in SDS content, only at low hydration. Huang et al. reported that the incorporation of weakly polar fatty acids would reduce the effective hydration of the polar headgroup, expanding the hydrocarbon chain region and causing the formation of inverted phases.22 d-Spacing Dependency on RH and SDS Content. Figure 5 A shows the X-ray diffraction patterns of DOPC 127 mM and 5 B DOPC/SDS 127/55.2 at the three RH values studied. The Bragg peaks observed in the diffraction patterns allow us to calculate the lamellar repeat distance (d spacing), which is a parameter that is related to the thickness of the bilayer (both in the lamellar and rhombohedral phases). These values allow us to evaluate the effect of SDS on DOPC multilayers for all samples. The d-spacing values at different RH and SDS content are shown in Table 1. The results show the increase in d spacing with the RH and SDS content increases. The d-spacing variation was especially notable with the RH and with a linear increase for all samples. The values in Table 1 show that the evolution of d spacing with the increase in SDS content is similar, regardless of the system RH. However, there is a slight minimum in the d spacing as a function of SDS content. In previous work, our group found that the addition of low SDS content to liposomes first led to a progressive decrease in the size of surfactant-lipid (21) Verkleij A. J.; Mombers C.; Gerritsen W. J.; Leunissen-Bijvelt, L.; Cullis, P. R. Biochim. Biophys. Acta 1979, 555, 358-361. (22) Huang, Z.; Epand, R. M. Chem. Phys. Lipids 1997, 86, 161-169.
Figure 6. Electron-density profile with the definition of the bilayer parameters t: d, the lamellar repeat distance; dB, bilayer region thickness; and dW, interlamellar water layer thickness.
vesicles because of the formation of contracted surfactant-lipid vesicles.9 This vesicle contraction was attributed to an electrostatic effect because of the incorporation of charged surfactant monomers into the vesicle bilayer. Increasing the SDS content led to a progressive rise in the mixed vesicle size. The slight minima obtained in our results could be compared with those previously published.9 Thus, in our study, the electrostatic effect could be affecting the d spacing of DOPC/SDS. It can be observed that this minimum is more emphasized at the highest RH used. The increase in hydration could be favoring the electrostatic effect because of the presence of ionic surfactant. Thus, considering the surfactant-lipid vesicles to be fully hydrated systems, the DOPC and DOPC/SDS samples at RH ) 65% could have similar behavior. An increase in the repeat distance as a function of RH from 1 to 65% is observed. We associate this increase in d spacing with a swelling effect that is evidenced when the water content in the bilayer increases.23 However, the d-spacing increase for the samples seems to be independent of SDS content; see the values in Table 1. The null or near-zero effect of SDS on this behavior of DOPC/SDS samples is remarkable. Analysis of Electron-Density Profiles. To obtain more information about the DOPC and DOPC/SDS systems, we calculated the electron-density profiles from the diffraction patterns. Figure 6 shows a profile of pure DOPC at RH ) 65%. The high-electron-density region, corresponding to the main maxima, reflects the presence of the phospholipid headgroups whereas the low-density region reflects the hydrocarbon chains.24 In some of the profiles, two secondary maxima in the main electron-density maximum are observed (Figures 6 and 7). This corresponds to the headgroup separation (dW) because of the incorporation of a water layer (swelling effect).16 The lamellar repeat distance (i.e., d spacing) is the sum of the bilayer region (23) Fragneto, G.; Charitat, T.; Bellet-Amalric, E.; Cubitt R.; Graner, F. Langmuir 2003, 19, 7695-7702. (24) Bouwstra, J. A.; Gooris, G. S.; Bras, W.; Talsma, H. Chem. Phys. Lipids 1993, 64, 89-98.
5260 Langmuir, Vol. 22, No. 12, 2006
Pereira-Lachataignerais et al.
Figure 7. Electron-density profiles calculated for pure DOPC (A) and DOPC/SDS (127/0.69) (B) and (127/55.2) (C) at the three relative humidities studied.
thickness (dB) and the interlamellar water layer thickness (dW) when the swelling effect is observed (Figure 6). Table 1 shows the bilayer parameter values (d, dB, and dW) of the samples analyzed, which allow us to discuss the evolution of the system with RH. The polar headgroup separation (dW) as a function of the increase in the RH was detected for all DOPC/SDS ratios. It can be observed that this separation is higher in a pure DOPC sample at RH ) 30 and 65% than in DOPC/SDS samples. In DOPC/SDS samples, the electron density in the interbilayer region is slightly higher than in pure DOPC. In this case, the sodium counterion from sulfate in SDS is present in the water layer, thus increasing the electron density in the headgroup separation region. The profiles shown in Figure 7 correspond to pure DOPC, DOPC/SDS (127/0.69 mol/mol), and DOPC/SDS (127/55.2 mol/ mol), measured at RH 1, 30, and 65%. At very low RH, only two main maxima are observed, which suggests that dW ) 0 (i.e., close proximity of polar headgroups in the bilayer). However, the electron-density profiles obtained at RH ) 30 and 65% show two secondary maxima in each main maximum, indicating an increase in the separation between headgroups. At RH ) 65%, the lipid polar headgroup separation is maximized, dW ≈ 11Å, (Figure 7, Table 1), as was expected. According to results from the literature, the electron-density trough in the center of the profiles indicates the localization of the low-density methyl groups even at high RH values.4,15 In the present work, the electron-density profiles at the highest RH studied show those methyl group contributions. However, this low-density region in the center of the profiles is not detected in the samples represented in Figure 7: (A) RH ) 1%, (B) RH ) 1-30%, and (C) RH ) 1-30%. In the pure DOPC sample, the interdigitation effect appears only at RH ) 1%. This would be expected, assuming that the multilayers are more compacted at low hydration. However, in the DOPC/SDS samples analyzed,
the profiles show this behavior also at RH ) 30%. The slight decrease in the bilayer thickness could be related to the surfactant charges involved in the bilayer formation. Then, the electrostatic effect could also favor the interdigitation of the hydrocarbon tails in the mixed samples at RH ) 30%. Thus, the contraction of the chains due to the presence of the SDS molecule would not affect either the pure DOPC sample or the samples at the highest RH.
Conclusions
The 3D rhombohedral phase was detected in DOPC multilayers at relative humidities of 1 and 30% at room temperature. This phase was also observed in DOPC/SDS mixed samples and was even powered by increasing SDS content. The fact that SDS potentiated this phase was not expected because of the hydrophilic character of this molecule. At low humidity, an inverse structure with SDS molecules participating in the high-curvature regions could stabilize the stalk motifs of the rhombohedral phase. The lamellar arrangement, characteristic of both LR and R phases, was detected in all samples, allowing us to obtain the d spacing parameter, which increases with RH and SDS content. The increase in headgroup separation due to the increase in RH was observed in the electron-density profiles. Acknowledgment. We thank Professor Peter Laggner for access to all the facilities of the Institute of Biophysics and X-ray Structure Research, Austrian Academy of Sciences at Synchrotron ELETTRA. This work was supported by funds from CICYT (MAT2004-1188-C02-02), and J.P.-L. acknowledges financial support from grant FP-2001-1568. Supporting Information Available: Phasing combination analysis for the DOPC bilayer at RH ) 65% and DOPC/SDS 127/0.69 mM at RH ) 65%. This material is available free of charge via the Internet at http://pubs.acs.org. LA053207K
