Lipid Compositions Modulate Fluidity and Stability of Bilayers

Article pubs.acs.org/Langmuir

Lipid Compositions Modulate Fluidity and Stability of Bilayers: Characterization by Surface Pressure and Sum Frequency Generation Spectroscopy Wei Liu,† Zhuguang Wang,† Li Fu,† Roger M. Leblanc,‡ and Elsa C. Y. Yan*,† †

Department of Chemistry, Yale University, 225 Prospect Street, New Haven, Connecticut 06520, United States Department of Chemistry, University of Miami, 1301 Memorial Drive, Cox Science Center, Coral Gables, Florida 33146, United States

‡

S Supporting Information *

ABSTRACT: Cell membranes are crucial to many biological processes. Because of their complexity, however, lipid bilayers are often used as model systems. Lipid structures influence the physical properties of bilayers, but their interplay, especially in multiple-component lipid bilayers, has not been fully explored. Here, we used the Langmuir−Blodgett method to make mono- and bilayers of 1,2-dihexadecanoyl-sn-glycero-3-phospho-(1′-rac-glycerol) (DPPG), 1-palmitoyl-2-oleoyl-sn-glycero-3-phospho-(1′-rac-glycerol) (POPG), and 1-hexadecanoyl-2-(9Z-octadecenoyl)-sn-glycero-3-phospho-L-serine (POPS) as well as their 1:1 binary mixtures. We studied the fluidity, stability, and rigidity of these structures using sum frequency generation (SFG) spectroscopy combined with analyses of surface pressure−area isotherms, compression modulus, and stability. Our results show that single-component bilayers, both saturated and unsaturated, may not be ideal membrane mimics because of their low fluidity and/or stability. However, the binary saturated and unsaturated DPPG/POPG and DPPG/POPS systems show not only high stability and fluidity but also high resistance to changes in surface pressure, especially in the range of 25−35 mN/m, the range typical of cell membranes. Because the ratio of saturated to unsaturated lipids is highly regulated in cells, our results underline the possibility of modulating biological properties using lipid compositions. Also, our use of flat optical windows as solid substrates in SFG experiments should make the SFG method more compatible with other techniques, enabling more comprehensive future surface characterizations of bilayers.

1. INTRODUCTION Cell membranes are a crucial component of all living organisms.1 They not only create the boundary between cells and their environments, but they also host components crucial for various biological processes, such as homeostasis maintenance, intercellular communication, and transportation.2−7 Lipids, proteins, sterols, and carbohydrates, typical components of a cell membrane, all play important biological roles.8 Although membrane-related processes are well-studied, investigation of individual components is difficult because of the complexity of cell membranes.9−11 Hence, it is important to model simplified cell membranes while still maintaining active biological functions. Therefore, establishing methods that enable quantitative control of physical properties of lipid bilayer models for mimicking cell membranes is important. Various methods have been developed to mimic cell membranes, including black lipid membranes, solid-supported planar lipid bilayers, self-assembled monolayers, liposome fusion, polymer-cushioned lipid bilayers, nanomaterial-supported lipid bilayers, and microfluidic platforms.12−19 Most of these methods start with Langmuir monolayer deposition or liposome fusion on solid substrates.12,20 A model for membrane © 2013 American Chemical Society

bilayers should fulfill several criteria, including simplicity of setup, reliability and reproducibility, functionality comparable to biological membranes, and compatibility with various characterization techniques.12,20,21 Among those methods, the solid-supported planar lipid bilayer is applied most commonly because it is easy to set up and compatible with a wide range of optical methods.12 In this study, we introduced a setup to use the Langmuir−Blodgett method to prepare supported bilayers on flat optical windows. The lipid bilayer is prepared by combining two lipid monolayers, one deposited onto the optical windows and the other prepared at the air/water interface using the Langmuir trough. We characterized the surface properties of the single- and binary-component monolayers at the air/water interface by evaluating the surface pressure isotherm, compression modulus, and stability. Subsequently, we characterized the bilayers and the corresponding monolayers using surface-specific vibrational sum frequency generation (SFG) spectroscopy. Received: September 24, 2013 Revised: November 18, 2013 Published: November 18, 2013 15022

dx.doi.org/10.1021/la4036453 | Langmuir 2013, 29, 15022−15031

Langmuir

Article

approaches for controlling the physical properties and stability of lipid bilayers by varying the lipid structure and mixture compositions.

The SFG spectroscopic method has been used to explore protein folding, lipid behaviors, molecular assembly, and dynamics of biological interactions at the air/water interface, solid silica surfaces, polymer surfaces, and supported lipid bilayers.22−27 Recently, chiral SFG spectroscopy has been exploited to characterize chiral biopolymers, such as proteins, peptides, and DNA.28−33 Lipid−water interfaces have been studied extensively using SFG, including kinetic studies of lipid flip-flop in bilayers,34 observations of interactions between proteins and biomembrane mimics,35−38 characterization of various lipid monolayers at the air−water interface,39−42 and investigation of polymer-cushioned lipid bilayers.43 Although most of those works focus on single-component lipids, mixed lipid bilayers have been investigated recently. Conboy and coworkers reported the flip-flop property of lipid bilayers with mixed lipids and cholesterol.38,44 Davies and co-workers also reported SFG spectroscopic studies on the hybrid bilayer with lipids in various compositions.45,46 How to modulate physical properties and stability by tuning the bilayer composition and structure, though, remains largely unexplored. To approach this problem, we first made the precursor monolayers using various binary lipid compositions at the air/ water interface and characterized the surface properties of the monolayers. Then, we correlated the lipid compositions to surface properties of the monolayers prior to bilayer formation. We evaluated the stability and robustness of the bilayers formed on flat optical windows using SFG vibrational spectroscopy. Because the cell membrane components are diverse, ranging from saturated and unsaturated lipids to glycolipid and proteinassociated lipids,47,48 it is important to investigate the interplay of bilayer properties with lipid structures and compositions. We used three types of lipids in single and binary compositions, namely, 1,2-dihexadecanoyl-sn-glycero-3-phospho-(1′-rac-glycerol) (DPPG), 1-palmitoyl-2-oleoyl-sn-glycero-3-phospho-(1′rac-glycerol) (POPG), and 1-hexadecanoyl-2-(9Z-octadecenoyl)-sn-glycero-3-phospho-L-serine (POPS). DPPG and POPG share a common hydrophilic glycerol headgroup, while POPG and POPS share the same hydrophobic carbon chains containing one unsaturated double bond (Figure 1). Our results show that binary saturated and unsaturated DPPG/ POPG and DPPG/POPS systems provide desirable properties for mimicking cell membranes. They show not only high stability and fluidity but also high resistance to changes in surface pressure, especially in the range of 25−35 mN/m, the typical range for a cell membrane. Our studies provide

2. MATERIALS AND METHODS 2.1. Materials and Reagents. DPPG (99%), POPS (99%), and POPG (99%) were purchased from Avanti Polar Lipids, Inc. (Alabaster, AL) and used without further purification. Chloroform (spectroscopy purpose, J.T. Baker Chemical Co., Central Valley, PA) was used to dissolve lipids. A phosphate buffer (10 mM, pH 7.4) was prepared using sodium phosphate dibasic heptahydrate (Na2HPO4· 7H2O, Sigma) and sodium phosphate monobasic anhydrous (NaH2PO4, Fisher Scientific). Water (resistivity of 18.2 MΩ cm) for all experiments was supplied by the Milli-Q plus system. 2.2. Surface Chemistry Analysis of the Lipid Monolayer. The surface pressure−molecular area (π−A) isotherm was obtained using a KSV mini Langmuir trough (KN2002, KSV Instruments, Ltd., Finland). Two symmetric barriers on the Langmuir trough were controlled by KSV Nima software, and a Wihelmy plate recorded the surface pressure during compression. The chloroform solutions of lipids were added and spread onto the buffer subphase using a Hamilton microsyringe. Chloroform was allowed to evaporate for 10 min before lipid monolayers were compressed. The stability of lipid monolayers with different compositions was assessed by the following experiment. First, the monolayer was compressed and maintained at a constant surface pressure. If the monolayer is not stable, molecules at the surface may desorb into the subphase. Because the mean molecular area is calculated with the assumption that all molecules stay at the surface, desorption at a constant surface pressure leads to a decrease of the mean molecular area. Thus, a decrease in the mean molecular area at a constant surface pressure reveals instability of the lipid monolayers. In this experiment, we also monitored the fluctuation of the surface pressure, which reveals the rigidity of the lipid monolayers. Each experiment was repeated at least 3 times. Finally, the surface pressure isotherms of lipid monolayers with different compositions were used to compute the compression modulus CS−1 49,50

CS−1 = − A(dπ /dA)

(1,)

where A is the mean molecular area per molecule at the given surface pressure π. Its value corresponds to the packing and rigidity of the monolayer at different surface pressures. A high value means that a monolayer is rigid and well-ordered. To prepare solid-supported lipid bilayers, we used the Langmuir− Blodgett method to deposit a lipid monolayer onto thin optical windows. Two slides of the windows were stacked and attached to the KSV dipper (Figure 2A). Before dipping, the lipid monolayer was compressed to ∼34 mN/m to mimic the surface pressure of a biomembrane.51 Then, the slides were pulled up vertically at a constant rate of 5 mm/min, while the surface pressure was maintained at ∼34 mN/m. The efficiency of the deposition for single and binary lipid monolayers on the slides ranged from 91 to 95% (see Table S1 of the Supporting Information). Once the deposition was finished, two slides covered with a monolayer of lipid were separated for further preparation of supported bilayers (Figure 2A). Meanwhile, the Teflon container was placed in the dipping well of the Langmuir trough, and a new lipid monolayer at the air−water interface was prepared (Figure 2B). The container is equipped with a small handle that can be connected to the dipper and lifted at a constant rate (3 mm/min). Once the monolayer reached ∼34 mN/m,51 the container was lifted to transfer the ∼34 mN/m lipid monolayer in the Langmuir trough to the Teflon container (Figure 2). The lipid monolayer-coated optical window was transferred onto the Teflon container with the deposited lipid monolayer facing down (Figure 2B). The Teflon container has a small hole where a syringe can be inserted to control the liquid level by pulling out or pushing in the subphase buffer. Increasing and decreasing the liquid level can lead to formation and separation of the lipid bilayer, respectively. Unlike previous lipid bilayer setups using

Figure 1. Chemical structures of selected phospholipids: (A) DPPG, (B) POPG, and (C) POPS. 15023


Langmuir

Article

Figure 2. (A) Deposition of a lipid monolayer onto two optical windows. (B) Construction and manipulation of lipid bilayers using a syringe to modify liquid levels. prisms as solid substrates in SFG experiments, the flat optical window can be adjusted to horizontal positions, providing flexibility in optical alignment and making it straightforward to examine bilayer homogeneity. 2.3. SFG Experiments. The detailed setup of the broad-bandwidth SFG spectrometer was described previously by Ma et al.40 It contained a 6 W regenerative amplifier seeded by a 120 fs, 1.9 W Ti/sapphire oscillator (Mai Tai, Spectra-Physics) and was pumped by two Nd/YLF pump lasers (16 W, Empower, Spectra-Physics). Half of the amplifier output (3 W) was pumped by an optical parametric amplifier (OPA, TOPAS, Spectra-Physics) to generate a 120 fs pulsed infrared (IR) beam in the range of 3800−900 cm−1. The other half (3 W) of the amplifier (800 nm output) entered a pulse shaper to yield ∼2 ps pulses to a narrow bandwidth of ∼7 cm−1. The pulse shaper consisted of a grating, a planoconvex cylindrical lens, and a homemade slit. The reflected SFG signal was filtered, focused onto the slit of the monochromator (SP-2558, Princeton Instruments), and detected by a charge-coupled device (CCD, Spec-10:400BR/LN, Princeton Instruments). The 800 nm beam has an incident angle of 56°, and the IR beam has an incident angle of 69°. We used ssp polarization (spolarized SFG, s-polarized visible, and p-polarized IR) to obtain the C−H stretch spectra of lipids at 2800−3000 cm−1 with an acquisition time of 10 s and used eq 2, to analyze the spectra

ISFG ∝

(2) χNR

+

∑ q

Figure 3. Surface pressure−molecular area isotherm of singlecomponent lipid monolayers.

2

Aq ωIR − ωq + i Γq

difference is likely due to the difference in the headgroup, where DPPG and POPG have non-charged glycerol headgroups and POPS has a charged serine. The isotherms of POPS and POPG have a similar collapsing area and pressure, while POPS has a smaller surface pressure than POPG over the whole range, suggesting that POPS is likely unstable at the air− water interface. These observations suggest that the saturated DPPG lipid monolayer is more rigid and ordered than POPS and POPG. Figure 4 presents the π−A isotherms of binary-component lipid monolayers (green) with the corresponding singlecomponent isotherms (red and blue). Figure 4A shows the isotherm for the DPPG/POPG mixture, which has a significantly larger rising area (140 Å2) than both DPPG and POPG (∼120 Å2). In comparison to the single-component isotherms, the isotherm of the binary DPPG/POPG monolayer has a higher surface pressure over the range from ∼45 to ∼140 Å2. Figure 4B presents the isotherm for the monolayer of the DPPG/POPS mixture. It shows that the rising area is slightly larger than that of the DPPG monolayer isotherm. In comparison to the single-component isotherm, the mixture has a higher surface pressure than the single-component systems in the range of 60−120 Å2. Finally, Figure 4C shows the isotherm for the monolayer of the POPG/POPS mixture, which has a significantly larger rising area of 140 Å2 (120 Å2 for POPG and 100 Å2 for POPS). The binary lipid monolayer isotherm has a higher surface pressure than the single-

(2,)

χ(2) NR

is the non-resonant secondwhere ISFG is the intensity of SFG, order susceptibility, ωIR is the input IR frequency, and Aq, ωq, and Γq are the amplitude, resonant frequency, and damping factor of the qth vibration mode, respectively.

3. RESULTS 3.1. Surface Characterizations of Single-Component Lipid Monolayers. Figure 3 shows the surface pressure−area isotherms of single-component lipid monolayers made of DPPG, POPS, and POPG. The DPPG monolayer isotherm shows a clear phase transition from the liquid phase to the solid phase at ∼15 mN/m. The rising area is ∼120 Å2, and the limiting area is ∼56 Å2. The collapsing point is at ∼55 mN/m, in agreement with previous reports.52,53 The POPS monolayer isotherm does not show a clear border between the solid and liquid phases. The rising area is ∼100 Å2, and the limiting area is ∼70 Å2, while the collapsing point is at ∼45 mN/m. The POPG monolayer isotherm also lacks a clear border between the solid and liquid phases. The rising area of POPG is ∼120 Å2, and the collapsing point is at ∼42 mN/m. The isotherms show that, in comparison to POPG, DPPG has a higher collapsing pressure and lower pressure over the range from ∼45 to ∼120 Å2. The DPPG and POPG monolayers have the same rising area, while POPS has a much smaller rising area. This 15024


Langmuir

Article

Figure 4. Surface pressure−molecular area isotherms of binary-component lipid monolayers (green) compared to those of the corresponding singlecomponent lipid monolayers (red and blue): (A) DPPG/POPG, (B) DPPG/POPS, and (C) POPS/POPG.

Figure 5. Compression modulus versus surface pressure for binary-component lipid monolayers (green) and the corresponding single-component lipid monolayer (red and blue): (A) DPPG/POPG, (B) DPPG/POPS, and (C) POPS/POPG.

Figure 6. Stability analyses of single-component and 1:1 binary-component lipid monolayers: (A) DPPG, (B) POPG, (C) POPS, (D) DPPG/ POPG, (E) DPPG/POPS, and (F) POPG/POPS. The surface pressure (blue) and mean molecular area (red) are monitored. All experiments were performed for 1 h or until the distance of the two barriers of the Langmuir trough reached its minimum.

component isotherms over the range of 45−120 Å2. In general, all binary mixtures show smoother isotherms and higher surface pressure in the liquid phase than the corresponding singlecomponent monolayers, suggesting that the binary mixtures are more loosely packed and/or contain self-assembled lipid rafts. They also have larger rising areas, indicating less ordered gasphase packing. Figure 5 shows the compression modulus against surface pressure for each binary monolayer (green) together with the

corresponding single-component monolayers (red and blue). As discussed earlier, the compression modulus indicates molecular packing, rigidity, and sensitivity to environmental change at various surface pressures. Figure 5A shows the compression modulus of the DPPG/POPG binary monolayer together with that of the POPG and DPPG monolayers. The DPPG monolayer has a significantly higher compression modulus than both the POPG single-component and DPPG/ POPG binary-component lipid monolayers at surface pressures 15025


Langmuir

Article

Figure 7. SFG spectra and spectral fitting for the single- and binary-component lipid monolayers deposited on flat optical windows: (A) DPPG, (B) POPG, (C) POPS, (D) DPPG/POPG, (E) DPPG/POPS, and (F) POPG/POPS.

the barriers controlled by the software. The surface pressure (blue) and mean molecular area (red) were monitored (Figure 6). Because the mean molecular area was calculated assuming that all lipid molecules remain at the interface, a decrease in the mean molecular area indicates desorption of lipid molecules from the interfaces and, thus, instability of the monolayers. Panels A−C of Figure 6 show the time-dependent mean molecular area for single-component lipid monolayers. After initial compression of the DPPG monolayer, the mean molecular area of DPPG stays constant for more than 1 h (Figure 6A), implying monolayer stability. The surface pressure for the DPPG monolayer shows a relatively large fluctuation over time, however, indicating a large response to the adjustment of the surface pressure. This response suggests that the DPPG monolayer is relatively rigid. In contrast, the mean molecular area of both POPG and POPS monolayers decreases over time, suggesting relatively rapid surface desorption. The POPS monolayer is the least stable, and the surface area reaches the minimum value of the instrument in only 1200 s. The fluctuation in the surface pressure for the POPG and POPS monolayers is smaller than that for the DPPG monolayer, indicating that these monolayers have higher fluidity. We also monitored the surface pressure and mean molecular area for the binary lipid monolayers (panels D−F of Figure 6). Intriguingly, the results reveal that the binary lipid monolayers have higher stability or fluidity than the corresponding single lipid monolayers. For example, the binary DPPG/POPG (Figure 6D) and DPPG/POPS (Figure 6E) monolayers show a small decrease in the mean molecular area over time, indicating that the binary monolayers are as stable as the DPPG monolayer (Figure 6A) but more stable than the corresponding

over 20 mN/m. The maximum value of the DPPG compression modulus is 100 mN/m at a surface pressure of ∼45 mN/m, in agreement with a previous report by Wydro et al.54 This result confirms that the DPPG monolayer has ordered packing and carries high rigidity. From 25 to 38 mN/m, the DPPG/POPG binary monolayer has the lowest modulus, suggesting high tolerance to changes in the surface pressure. Figure 5B shows the compression modulus of the DPPG/POPS binary monolayer. Similarly, the binary DPPG/POPS monolayer has a lower modulus over the range of ∼25−40 mN/m than the single-component DPPG and POPS monolayers. Figure 5C shows the compression modulus of the POPS/POPG binary monolayer and the corresponding single-component monolayers. Both POPS and POPG monolayers have very similar moduli, likely because both lipids share the same carbon chains. However, these two monolayers have lower moduli than the saturated DPPG monolayer, confirming that the unsaturated lipid monolayers are loosely packed and less rigid than the saturated DPPG lipid monolayer. The modulus of the binary POPS/POPG monolayer, however, is even lower than that of the single-component POPS and POPG monolayers in the range of 25−38 mN/m. Without an exception, all moduli of binary monolayers are lower than those of the corresponding two single-component monolayers in the range of 25−35 mN/ m, indicating higher stability and resistance to environmental pressure change. This range of surface pressure overlaps with the surface pressure of a native cell membrane (25−35 mN/ m),51 suggesting possible evolutionary advantages for cells with membranes containing multiple lipid components. We then examine the stability and fluidity of the single and binary lipid monolayers. The monolayers were first compressed to 34 mN/m and then maintained at this surface pressure by 15026


Langmuir

Article

Figure 8. ssp SFG spectra of the single-component lipid systems. A lipid bilayer formation/separation cycle: before formation (green), bilayer (blue), and after separation (red) for (A) DPPG, (B) POPG, and (C) POPS. The SFG spectra of the first three formation/separation cycles for the (D) DPPG, (E) POPG, and (F) POPS monolayers.

∼2937 cm−1, and CH3 asymmetric stretch (AS) at ∼2966 cm−1 (fitting parameters are reported in the Supporting Information).55,56 For the unsaturated POPG and POPS lipids, those three peaks are in similar positions (panels B and C of Figure 7). However, an additional peak at ∼2852 cm−1 of the CH2 SS is observed because of the double bonds in the unsaturated lipid that disrupt the packing of the carbon chain, resulting in detectable CH2 arrangement.24,41,42,57 In addition, CH2 SS and CH3 SS peaks in the POPS spectrum are different from those in the POPG spectrum. Because POPG and POPS have identical hydrophobic chains, the difference can only be explained by the different properties of the hydrophilic heads (glycerol in POPG and serine in POPS), which may lead to different orientations and packing of the monolayer and, thus, affect the SFG signal. It is to be noted that the C−H stretching mode of the double bond in POPG and POPS was not observed, which is likely due to the low surface density of the C−H groups in lipids and/ or the disordering of the C−H groups, leading to cancellation of SFG signals. Because the focus of the SFG studies is to examine the robustness of the lipid bilayers, the other spectral regions, such as the O−H and CO stretching modes, were not explored. Nonetheless, the CO stretch of DPPG was observed in our previous studies.58 For all binary lipid monolayers, all four peaks are observable, confirming the disorder because of the presence of double bonds in the unsaturated POPG and POPS lipids (panels D−F of Figure 7). Then, we performed the following experiments to characterize the single and binary lipid bilayers as well as their corresponding monolayers on flat optical windows using SFG. As shown in Figure 2B, we first obtained the SFG spectra of the monolayer of the lipid deposited on the flat optical windows (Figure 7). Then, we used a syringe to raise the water level,

single POPG (Figure 6B) and POPS (Figure 6C) lipid monolayers. Fluctuation was observed to different extents in the surface pressure curves for different bilayers. By calculating the ratio of standard deviation to the average surface pressure after reaching 34 mN/m, we obtained 1.01, 0.32, and 0.31% for the single lipid DPPG (Figure 6A), POPG (Figure 6B), and POPC (Figure 6C) bilayers, respectively, and 0.15, 0.24, and 0.21% for the binary DPPG/POPG (Figure 6D), DPPG/POPS (Figure 6E), and POPG/POPS (Figure 6F) bilayers, respectively. Because a rigid bilayer responds slowly to the modulation of surface pressure, a large fluctuation indicates high rigidity. Therefore, the results suggest that only the binary lipid bilayer with both saturated and unsaturated components is both highly fluidic and highly stable. More surprising is the stability enhanced by mixing two unsaturated POPG and POPS lipids. The binary POPG/POPS lipid monolayer shows a decrease in the mean surface area much smaller than that of the single POPS lipid monolayer, revealing that the presence of POPG can slow desorption and enhance the stability of POPS at the interface. The binary POPG/POPS lipid monolayer (Figure 6F) shows a similar magnitude of fluctuation as the POPG (Figure 6B) and POPS (Figure 6C) single-component monolayers, suggesting that fluidity is maintained in the binary lipid monolayer but with improved stability. 3.2. SFG Spectroscopic Study of the Lipid Monolayer/ Bilayer. Using SFG, we first characterized the C−H stretch vibration structures of the single and binary lipid monolayers deposited on flat optical windows using the setup shown in Figure 2B. Figure 7 shows the SFG spectra with peak assignments. The saturated DPPG lipid shows three peaks in the C−H region of SFG spectra (Figure 7A): CH3 symmetric stretch (SS) at ∼2877 cm−1, CH3 Fermi resonance (FR) at 15027


Langmuir

Article

Figure 9. ssp SFG spectra of the 1:1 binary lipid systems. A lipid bilayer formation/separation cycle: before formation (green), bilayer (blue), and after separation (red) for (A) DPPG/POPG, (B) DPPG/POPS, and (C) POPS/POPG. The SFG spectra of the first three formation/separation cycles for the (D) DPPG/POPG, (E) DPPG/POPS, and (F) POPG/POPS binary monolayers.

such that the monolayer at the air/water interface and the monolayer on the optical windows made contact, forming a supported bilayer (Figure 2B). Subsequently, we obtained the SFG spectra of the lipid bilayer. Then, we decreased the water level to separate the layers and obtained a SFG spectrum of the solid-supported lipid monolayer on the optical window. We carried out this cycle multiple times on a single optical window. We repeated this set of experiments 3 times with new monolayers of lipid. At each step in the experiments, we checked the homogeneity of the lipid structures by taking spectra at a minimum of three locations on the windows. Figure 8A presents the spectra for the DPPG singlecomponent system. Before the liquid level is raised (green), the solid-supported DPPG monolayer shows strong SFG signals. To confirm the homogeneity of the monolayer, we took spectra at various spots on the optical windows (see Figure S1 of the Supporting Information). The highly similar spectra suggest that the DPPG monolayer has high homogeneity. After the liquid level is raised to form the lipid bilayer (blue), all SFG signals vanish because of the symmetric orientation of the lipid bilayer, as shown previously by Davies et al. and Ye et al.59−61 Upon decreasing the liquid level again, the SFG vibrational peaks reappear at the same positions with similar intensity (red), indicating the separation of the bilayer and reformation of the monolayer. The formation/separation cycle can be performed up to 10 times with similar results (first three cycles are shown in Figure 8D), indicating that the DPPG bilayer is exceptionally stabile and robust. Because high fluidity is expected to cause quick lipid exchanges between two layers as well as desorption of lipid from the solid substrate, our observation implies, in agreement with previous reports, that the DPPG bilayer undergoes slow lipid exchange.38,44

Panels B and C of Figure 8 show the results for the POPG and POPS single-component systems, respectively. For the solid-supported monolayer (green), POPG and POPS show similar spectra. When the liquid level is raised to form a bilayer (blue), both the POPG and POPS systems show greatly reduced peaks. For the monolayers (red), which form after the liquid level decreases, the C−H stretch peaks reappear at the same positions, with significantly reduced intensity. The differences in SFG spectra at different positions suggest that the homogeneity of the unsaturated system is lower than that of the DPPG system (see Figure S1 of the Supporting Information). Both of these bilayers are less robust than the DPPG bilayer; after only two to three cycles of formation/ separation of the bilayers, the SFG signal vanished (panels E and F of Figure 8). Figure 9 presents the SFG spectra of the binary DPPG/ POPG, DPPG/POPS, and POPG/POPS 1:1 lipid systems. Before the liquid level is raised (green), both binary DPPG/ POPG (Figure 9A) and DPPG/POPS (Figure 9B) monolayers show strong SFG signals. Upon checking at least three different locations on the optical windows to examine homogeneity (see Figure S2 of the Supporting Information), we observed slight variation in the SFG spectra at different points, suggesting that the binary lipid system is less homogeneous than the singlecomponent system. This difference could be due to the disordered packing of the lipid monolayer and/or formation of local islets. After the buffer level is raised to form bilayers (blue), the signals almost vanish. The residual signals could be due to inhomogeneity, which can lead to local asymmetry. After the separation of the lipid bilayers (red), the SFG signals reappear with slightly lower intensity, suggesting lipid desorption or increased disordering of the bilayer. The lipid 15028


Langmuir

Article

While the results for the surface characterization of the single-component lipid monolayers are intuitive, the results for the binary lipid monolayers reveal some intriguing properties. While the binary unsaturated POPG/POPS lipid monolayers show surface properties similar to those of the singlecomponent POPG and POPS lipid monolayers (Figures 5 and 6), the binary saturated/unsaturated DPPG/POPG and DPPG/POPS monolayers have some desirable properties for mimicking cell membranes. They show improved stability compared to the single-component unsaturated POPG and POPS monolayers and higher fluidity than the singlecomponent saturated DPPG monolayer, conclusions supported by the observations that the binary monolayers show low fluctuations in surface pressure and small changes in mean molecular area in the stability analyses (Figure 6). Moreover, the binary DPPG/POPG and DPPG/POPS monolayers have relatively low compression moduli compared to singlecomponent monolayers, indicating higher tolerance to environmental change. As Figure 10 shows, it is likely that saturated lipids contribute to high rigidity and stability, while unsaturated lipids contribute to high fluidity and low compression moduli in the binary saturated/unsaturated lipid monolayers. In loosely packed lipid bilayers, interdigitation between two monolayers is possible, as a previous study shows.62 However, our experiments do not facilitate identification of such processes. These results generally agree with the SFG studies. First, the DPPG monolayer on the optical window shows strong C−H stretch peaks preserved after several bilayer formation/ separation cycles (Figure 8D). These cycles place considerable mechanical stress on the lipid structures. The fact that the DPPG bilayer can undergo at least 10 cycles with preserved bilayer structures shows that the DPPG bilayer is very robust. For the POPG and POPS bilayers, the C−H signals decrease rapidly after a single cycle, implying high fluidity but low stability. Moreover, the binary saturated/unsaturated DPPG/ POPG and DPPG/POPS bilayers show diminishing C−H signals over one to two cycles, suggesting that saturated/ unsaturated mixed lipid bilayers have improved or similar stability compared to the single unsaturated lipid bilayers but higher fluidity than the saturated DPPG bilayer (Figure 10). Our results also show that changes in surface properties that results from mixing two lipids are indicated in the compression modulus (Figure 5). The binary lipid monolayers always have lower moduli than the corresponding single lipid monolayers. The effect is particularly obvious in the surface pressure range of 25−35 mN/m of a typical cell membrane. Because lowering the compression modulus can minimize environmental impacts on the cell membrane, it could be important in stabilizing the cell membrane. Because each type of cell has a characteristic lipid composition, one may speculate based on our results that the lipid compositions of various types of cell membrane may have been selected through evolution to enhance stability and optimize fluidity for biological functions. The combined results of surface characterization and SFG studies support the premise that the stability and fluidity of lipid bilayers may be adjusted by varying the lipid compositions. Further characterization of the surface properties of lipid monolayers and correlation of these properties to the stability and fluidity of lipid bilayers will allow for the use of the Langmuir−Blodgett method to optimize the physical properties of supported bilayers and eventually to address problems related to cell membranes in both fundamental and applied research.

bilayer formation/separation cycle is performed multiple times. The SFG spectra for the first three cycles are shown in panels D and E of Figure 9 for the DPPG/POPG and DPPG/POPS systems, respectively. The binary saturated/unsaturated lipid systems are more stable than the corresponding single unsaturated lipid bilayers, and several cycles of bilayer formation/separation can be performed with appreciable SFG signals. However, the SFG signals diminish along with the cycles, in agreement with the argument that the binary saturated/unsaturated lipid bilayers remain fluid. The results for the POPS/POPG binary lipid system are shown in panels C and F of Figure 9. Because this system involves a mixture of unsaturated lipids, there are several notable differences when compared to the unsaturated/ saturated binary lipid systems. First, the SFG signals of the POPS/POPG binary lipid system are significantly lower after a single bilayer formation/separation cycle. Second, the cycle can only be performed 1 time before the SFG signals vanish. Both observations imply that the unsaturated/unsaturated binary lipid system is as fluid as the single unsaturated POPG and POPS bilayers.

4. DISCUSSION We characterized the monolayers and bilayers made of singlecomponent DPPG, POPG, and POPS lipids and their binary 1:1 mixtures by analyzing the surface pressure−area isotherms, compression modulus, and stability together with results from SFG spectra. The results of single-component lipid monolayers show that the monolayers formed by the saturated DPPG lipid are more stable and rigid than those formed by the unsaturated POPG and POPS. This conclusion agrees with our observations that the DPPG monolayer has a higher collapsing pressure (Figure 3) and larger modulus value (Figure 5), as well as larger fluctuation in surface pressure and smaller changes in molecular area in the stability measurements (Figure 6). These results are expected because the saturated DPPG carbon chains are highly ordered (Figure 10). On the other hand, the POPG

Figure 10. Packing of lipids in single- and binary-component lipid bilayers.

and POPS monolayers show similar but slightly different stability and fluidity values. They have similar compression moduli, suggesting that the structure of the headgroup has little impact on the rigidity (Figure 5). Both have small fluctuations in the surface pressure in the stability measurements (Figure 6), implying high fluidity. Moreover, both show significant decreases in molecular area (Figure 6), indicating lower stability than the saturated DPPG monolayer. The POPS monolayer is most unstable, and its mean molecular area falls to the minimum instrumental limit within 1200 s. 15029


Langmuir

Article

(3) Alberts, B.; Johnson, A.; Lewis, J.; Raff, M.; Roberts, K.; Walter, P. Molecular Biology of the Cell, 5th ed.; Garland Science: New York, 2007. (4) Brown, M. S.; Goldstein, J. L. The SREBP pathway: Regulation of cholesterol metabolism by proteolysis of a membrane-bound transcription factor. Cell 1997, 89 (3), 331−340. (5) Edidin, M. TimelineLipids on the frontier: A century of cellmembrane bilayers. Nat. Rev. Mol. Cell Biol. 2003, 4 (5), 414−418. (6) Brown, D. A.; London, E. Functions of lipid rafts in biological membranes. Annu. Rev. Cell Dev. Biol. 1998, 14, 111−136. (7) Cullis, P. R.; Dekruijff, B. Lipid polymorphism and the functional roles of lipids in biological membranes. Biochim. Biophys. Acta 1979, 559 (4), 399−420. (8) Shipley, G. Recent X-ray diffraction studies of biological membranes and membrane components. Biol. Membr. 1973, 2, 1−89. (9) Saffman, P.; Delbrück, M. Brownian motion in biological membranes. Proc. Natl. Acad. Sci. U. S. A. 1975, 72 (8), 3111−3113. (10) Korn, E. D. Structure of biological membranes. Science 1966, 153 (3743), 1491−1498. (11) Brown, D.; London, E. Structure and origin of ordered lipid domains in biological membranes. J. Membr. Biol. 1998, 164 (2), 103− 114. (12) Castellana, E. T.; Cremer, P. S. Solid supported lipid bilayers: From biophysical studies to sensor design. Surf. Sci. Rep. 2006, 61 (10), 429−444. (13) Mao, H.; Yang, T.; Cremer, P. S. Design and characterization of immobilized enzymes in microfluidic systems. Anal. Chem. 2002, 74 (2), 379−385. (14) Spinke, J.; Yang, J.; Wolf, H.; Liley, M.; Ringsdorf, H.; Knoll, W. Polymer-supported bilayer on a solid substrate. Biophys. J. 1992, 63 (6), 1667. (15) Yang, T.; Baryshnikova, O. K.; Mao, H.; Holden, M. A.; Cremer, P. S. Investigations of bivalent antibody binding on fluid-supported phospholipid membranes: The effect of hapten density. J. Am. Chem. Soc. 2003, 125 (16), 4779−4784. (16) Berquand, A.; Mazeran, P.-E.; Pantigny, J.; Proux-Delrouyre, V.; Laval, J.-M.; Bourdillon, C. Two-step formation of streptavidinsupported lipid bilayers by PEG-triggered vesicle fusion. Fluorescence and atomic force microscopy characterization. Langmuir 2003, 19 (5), 1700−1707. (17) Proux-Delrouyre, V.; Laval, J.-M.; Bourdillon, C. Formation of streptavidin-supported lipid bilayers on porous anodic alumina: Electrochemical monitoring of triggered vesicle fusion. J. Am. Chem. Soc. 2001, 123 (37), 9176. (18) McConnell, H.; Watts, T.; Weis, R.; Brian, A. Supported planar membranes in studies of cell−cell recognition in the immune system. Biochim. Biophys. Acta 1986, 864 (1), 95−106. (19) Cremer, P. S.; Boxer, S. G. Formation and spreading of lipid bilayers on planar glass supports. J. Phys. Chem. B 1999, 103 (13), 2554−2559. (20) Nagle, J. F.; Tristram-Nagle, S. Structure of lipid bilayers. Biochim. Biophys. Acta 2000, 1469 (3), 159−195. (21) Tamm, L. K.; McConnell, H. M. Supported phospholipid bilayers. Biophys. J. 1985, 47 (1), 105−113. (22) Eisenthal, K. Liquid interfaces probed by second-harmonic and sum-frequency spectroscopy. Chem. Rev. 1996, 96 (4), 1343−1360. (23) Chen, X.; Yang, T.; Kataoka, S.; Cremer, P. S. Specific ion effects on interfacial water structure near macromolecules. J. Am. Chem. Soc. 2007, 129 (40), 12272−12279. (24) Richmond, G. Molecular bonding and interactions at aqueous surfaces as probed by vibrational sum frequency spectroscopy. Chem. Rev. 2002, 102 (8), 2693−2724. (25) Shen, Y.-R. The Principles of Nonlinear Optics; WileyInterscience: New York, 1984; p 575. (26) Stiopkin, I. V.; Weeraman, C.; Pieniazek, P. A.; Shalhout, F. Y.; Skinner, J. L.; Benderskii, A. V. Hydrogen bonding at the water surface revealed by isotopic dilution spectroscopy. Nature 2011, 474 (7350), 192−195.

This work also introduces a method for preparing lipid bilayers on flat optical windows (Figure 2). Unlike previous setups that use prisms as solid substrates in the SFG studies, the position of the flat optical window can be adjusted horizontally while maintaining the optical alignment, enabling examination of the bilayers for homogeneity. The flat optical windows are more affordable than prisms and make SFG experiments compatible with in situ characterizations of bilayers using other optical methods, such as ultraviolet−visible (UV−vis) spectroscopy, fluorescence spectroscopy, Brewster angle microscopy, and IR reflection−absorption spectroscopy.

5. CONCLUSION We investigated the physical properties of lipid monolayers and bilayers by combining surface chemistry techniques with SFG spectroscopy using the DPPG, POPG, and POPS lipids as well as their binary 1:1 mixtures. In general, the saturated DPPG lipid bilayer has the highest stability and packing order. However, this high packing order makes it very sensitive to pressure fluctuation and low in fluidity. In contrast, the unsaturated lipid bilayers have lower stability and disordered packing, resulting in high fluidity, low sensitivity to pressure fluctuation, and low stability. For the binary lipid systems, mixtures of saturated and unsaturated lipids, such as DPPG/ POPG and DPPG/POPS, have both high stability and fluidity. The advantage of mixing saturated/unsaturated lipids is particularly evident when the surface pressure ranges from 25 to 35 mN/m, which is the surface pressure typical of a cell membrane. Our results provide insight into the molecular evolution of lipid compositions in cell membranes and highlight the possibility of modulating bilayer properties to better mimic cell membranes. Our setup for preparing lipid bilayers on flat optical windows, as well as the data obtained regarding the properties of lipid bilayers with various components, may shed light on surface-specific research as well as membrane-related biological studies that use lipid bilayers as cell membrane mimics.

■

ASSOCIATED CONTENT

S Supporting Information *

Parameters of lipid monolayer deposition (Table S1), SFG spectra of single-component lipid systems (Figure S1), SFG spectra of binary-component lipid systems (Figure S2), and peak assignments and fitting parameters of the SFG spectra (Table S2). This material is available free of charge via the Internet at http://pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Notes

The authors declare no competing financial interest.

■ ■

ACKNOWLEDGMENTS This work was supported by the National Science Foundation Grant CHE 1213362 to Elsa C. Y. Yan. REFERENCES

(1) Steck, T. L. Organization of proteins in human red blood-cell membraneReview. J. Cell Biol. 1974, 62 (1), 1−19. (2) Joscelyne, S. M.; Tragardh, G. Membrane emulsificationA literature review. J. Membr. Sci. 2000, 169 (1), 107−117. 15030


Langmuir

Article

(27) Shen, Y. Surface properties probed by second-harmonic and sum-frequency generation. Nature 1989, 337, 519−525. (28) Fu, L.; Xiao, D.; Wang, Z.; Batista, V. S.; Yan, E. C. Chiral sum frequency generation for in situ probing proton exchange in antiparallel β-sheets at interfaces. J. Am. Chem. Soc. 2013, 135 (9), 3592−3598. (29) Fu, L.; Wang, Z.; Yan, E. C. Chiral vibrational structures of proteins at interfaces probed by sum frequency generation spectroscopy. Int. J. Mol. Sci. 2011, 12 (12), 9404−9425. (30) Fu, L.; Liu, J.; Yan, E. C. Chiral sum frequency generation spectroscopy for characterizing protein secondary structures at interfaces. J. Am. Chem. Soc. 2011, 133 (21), 8094−8097. (31) Stokes, G. Y.; Gibbs-Davis, J. M.; Boman, F. C.; Stepp, B. R.; Condie, A. G.; Nguyen, S. T.; Geiger, F. M. Making “sense” of DNA. J. Am. Chem. Soc. 2007, 129 (24), 7492−7493. (32) Wang, J.; Chen, X.; Clarke, M. L.; Chen, Z. Detection of chiral sum frequency generation vibrational spectra of proteins and peptides at interfaces in situ. Proc. Natl. Acad. Sci. U. S. A. 2005, 102 (14), 4978−4983. (33) Rocha-Mendoza, I.; Yankelevich, D. R.; Wang, M.; Reiser, K. M.; Frank, C. W.; Knoesen, A. Sum frequency vibrational spectroscopy: The molecular origins of the optical second-order nonlinearity of collagen. Biophys. J. 2007, 93 (12), 4433−4444. (34) Liu, J.; Conboy, J. C. Direct measurement of the transbilayer movement of phospholipids by sum-frequency vibrational spectroscopy. J. Am. Chem. Soc. 2004, 126 (27), 8376−8377. (35) Chen, X.; Chen, Z. SFG studies on interactions between antimicrobial peptides and supported lipid bilayers. Biochim. Biophys. Acta 2006, 1758 (9), 1257−1273. (36) Ye, S.; Nguyen, K. T.; Clair, S. V. L; Chen, Z. In situ molecular level studies on membrane related peptides and proteins in real time using sum frequency generation vibrational spectroscopy. J. Struct. Biol. 2009, 168 (1), 61−77. (37) Smith, K. A.; Conboy, J. C. Using micropatterned lipid bilayer arrays to measure the effect of membrane composition on merocyanine 540 binding. Biochim. Biophys. Acta 2011, 1808 (6), 1611−1617. (38) Liu, J.; Conboy, J. C. Phase behavior of planar supported lipid membranes composed of cholesterol and 1, 2-distearoyl-sn-glycerol-3phosphocholine examined by sum-frequency vibrational spectroscopy. Vib. Spectrosc. 2009, 50 (1), 106−115. (39) Pittler, J.; Bu, W.; Vaknin, D.; Travesset, A.; McGillivray, D.; Lösche, M. Charge inversion at minute electrolyte concentrations. Phys. Rev. Lett. 2006, 97 (4), 046102. (40) Ma, G.; Liu, J.; Fu, L.; Yan, E. C. Probing water and biomolecules at the air−water interface with a broad bandwidth vibrational sum frequency generation spectrometer from 3800 to 900 cm−1. Appl. Spectrosc. 2009, 63 (5), 528−537. (41) Ma, G.; Allen, H. C. DPPC Langmuir monolayer at the air− water interface: Probing the tail and head groups by vibrational sum frequency generation spectroscopy. Langmuir 2006, 22 (12), 5341− 5349. (42) Chen, X.; Hua, W.; Huang, Z.; Allen, H. C. Interfacial water structure associated with phospholipid membranes studied by phasesensitive vibrational sum frequency generation spectroscopy. J. Am. Chem. Soc. 2010, 132 (32), 11336−11342. (43) Wang, T.; Li, D.; Lu, X.; Khmaladze, A.; Han, X.; Ye, S.; Yang, P.; Xue, G.; He, N.; Chen, Z. Single lipid bilayers constructed on polymer cushion studied by sum frequency generation vibrational spectroscopy. J. Phys. Chem. C 2011, 115 (15), 7613−7620. (44) Anglin, T. C.; Conboy, J. C. Kinetics and thermodynamics of flip-flop in binary phospholipid membranes measured by sumfrequency vibrational spectroscopy. Biochemistry 2009, 48 (43), 10220−10234. (45) Kett, P.; Casford, M.; Davies, P. Structure of mixed phosphatidylethanolamine and cholesterol monolayers in a supported hybrid bilayer membrane studied by sum frequency generation vibrational spectroscopy. J. Phys. Chem. B 2011, 115 (20), 6465−6473.

(46) Kett, P. J.; Casford, M. T.; Davies, P. B. Sum frequency generation (SFG) vibrational spectroscopy of planar phosphatidylethanolamine hybrid bilayer membranes under water. Langmuir 2010, 26 (12), 9710−9719. (47) Zs-Nagy, I. The role of membrane structure and function in cellular aging: A review. Mech. Ageing Dev. 1979, 9 (3−4), 237. (48) Abercrombie, M.; Ambrose, E. The surface properties of cancer cells: A review. Cancer Res. 1962, 22 (5, part 1), 525−548. (49) Davies, J. T.; Rideal, E. K. Interfacial Phenomena; Academic Press: New York, 1961. (50) Harkins, W. D.; Debye, P. The Physical Chemistry of Surface Films; Reinhold Publishing Corporation: New York, 1952. (51) Marsh, D. Lateral pressure in membranes. Biochim. Biophys. Acta 1996, 1286 (3), 183. (52) Vollhardt, D.; Fainerman, V.; Siegel, S. Thermodynamic and textural characterization of DPPG phospholipid monolayers. J. Phys. Chem. B 2000, 104 (17), 4115−4121. (53) Panda, A. K.; Vasilev, K.; Orgeig, S.; Prestidge, C. A. Thermodynamic and structural studies of mixed monolayers: Mutual mixing of DPPC and DPPG with DoTAP at the air−water interface. Mater. Sci. Eng., C 2010, 30 (4), 542−548. (54) Wydro, P.; Knapczyk, S.; Lapczynska, M. Variations in the condensing effect of cholesterol on saturated versus unsaturated phosphatidylcholines at low and high sterol concentration. Langmuir 2011, 27 (9), 5433−5444. (55) Conboy, J. C.; Messmer, M. C.; Richmond, G. L. Investigation of surfactant conformation and order at the liquid−liquid interface by total internal reflection sum-frequency vibrational spectroscopy. J. Phys. Chem. 1996, 100 (18), 7617−7622. (56) Liljeblad, J. F.; Bulone, V.; Rutland, M. W.; Johnson, C. M. Supported phospholipid monolayers. The molecular structure investigated by vibrational sum frequency spectroscopy. J. Phys. Chem. C 2011, 115 (21), 10617−10629. (57) Hall, S. A.; Jena, K. C.; Trudeau, T. G.; Hore, D. K. Structure of leucine adsorbed on polystyrene from nonlinear vibrational spectroscopy measurements, molecular dynamics simulations, and electronic structure calculations. J. Phys. Chem. C 2011, 115 (22), 11216−11225. (58) Fu, L.; Ma, G.; Yan, E. C. Y. In situ misfolding of human islet amyloid polypeptide at interfaces probed by vibrational sum frequency generation. J. Am. Chem. Soc. 2010, 132 (15), 5405−5412. (59) Holman, J.; Davies, P. B.; Nishida, T.; Ye, S.; Neivandt, D. J. Sum frequency generation from Langmuir−Blodgett multilayer films on metal and dielectric substrates. J. Phys. Chem. B 2005, 109 (40), 18723−18732. (60) Nishida, T.; Johnson, C. M.; Holman, J.; Osawa, M.; Davies, P. B.; Ye, S. Optical sum-frequency emission from Langmuir−Blodgett films of variable thickness: Effects of the substrate and polar orientation of fatty acids in the films. Phys. Rev. Lett. 2006, 96 (7), 077402. (61) Tong, Y.; Zhao, Y.; Li, N.; Osawa, M.; Davies, P. B.; Ye, S. Interference effects in the sum frequency generation spectra of thin organic films. I. Theoretical modeling and simulation. J. Chem. Phys. 2010, 133 (3), 034704. (62) Kranenburg, M.; Vlaar, M.; Smit, B. Simulating induced interdigitation in membranes. Biophys. J. 2004, 87 (3), 1596−1605.

15031


Lipid Compositions Modulate Fluidity and Stability of Bilayers

Recommend Documents