Miscibility Behavior and Nanostructure of Monolayers of the Main

Nov 16, 2011 - We used atomic force microscopy to study the structural properties of ... (10)Although biological membranes are bilayers, there is litt...
0 downloads 0 Views 3MB Size
ARTICLE pubs.acs.org/Langmuir

Miscibility Behavior and Nanostructure of Monolayers of the Main Phospholipids of Escherichia coli Inner Membrane  Laura Picas,† Carme Suarez-Germa, M. Teresa Montero, Oscar Domenech, and Jordi Hernandez-Borrell* Department of Physical-Chemistry, Faculty of Pharmacy and IN2UB, University of Barcelona, E-08028 Barcelona, Spain

bS Supporting Information ABSTRACT: We report a thermodynamic study of the effect of calcium on the mixing properties at the airwater interface of two phospholipids that mimic the inner membrane of Escherichia coli: 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphoethanolamine and 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphoglycerol. In this study, pure POPE and POPG monolayers and three mixed monolayers, χPOPE = 0.25, 0.5, and 0.75, were analyzed. We show that for χPOPE = 0.75, the values of the Gibbs energy of mixing were negative, which implies attractive interactions. We used atomic force microscopy to study the structural properties of LangmuirBlodgett monolayers that were transferred onto mica substrate at lateral surface pressures of 25 and 30 mN m1. The topographic images of pure POPE and POPG monolayers exhibited two domains of differing size and morphology, showing a step height difference within the range expected for liquid-condensed and liquid-expanded phases. The images captured for χPOPE = 0.25 were featureless, and for χPOPE = 0.5 small microdomains were observed. The composition that mimics quantitatively the proportions found in the inner membrane of E. coli, χPOPE = 0.75, showed large liquid condensed domains in the liquid expanded phase. The extension of each domain was quantitatively analyzed. Because calcium is used in the formation of supported bilayers of negatively charged phospholipids, the possible influence of the nanostructure of the apical on the distal monolayer is discussed.

’ INTRODUCTION The study of the physicochemical properties of phospholipids that constitute biomembranes is critical to our understanding of many physiological and pharmacological events. These include signal transduction, substrate transport, toxicity, or development of drug resistance mediated by transmembrane proteins. A large body of evidence establishes that nonhomogeneous lateral distribution of lipids is common in natural membranes.1 It is also believed that peptides and proteins interact preferentially with definite domains2 and induce domain formation.3 As a result of differences in their physicochemical properties, it has been demonstrated by using model membranes such as monolayers,4 giant unilamellar vesicles (GUVs),5 and supported lipid bilayers (SLBs)6 that mixtures of phospholipids undergo phase separation and become laterally structured in domains. Indeed, SLBs have been widely adopted as a suitable cell membrane model for imaging separated domains by atomic force microscopy (AFM).7 Lateral segregation, however, may have several origins. Thus, as they are mesomorphic, phospholipids can exist in several physical states. In excess of water, phospholipids form fully solvated lipid bilayers, which undergo the gel (Lβ) to liquid-crystalline (Lα) phase transition at a temperature (Tm) that is characteristic for each species and mixture.8 For phospholipidcholesterol mixtures, a distinction between liquid-ordered (lo) and liquiddisordered (ld) phases has also been described.9 Besides, below their corresponding critical temperature, phospholipids monolayers can be compressed and may undergo several phase transitions. In this case, there are four main monolayer phases: gaseous (G), r 2011 American Chemical Society

liquid-expanded (LE), liquid-condensed (LC), and solid (S) phases.10Although biological membranes are bilayers, there is little doubt about the importance of local lateral pressure and phospholipid phase changes of the monolayers in different mechanisms involved in the modulation of peripheral membrane proteins11 and transmembrane protein function.12,13 Hence, the investigation of the phase and mixing properties of phospholipid monolayers may help us to understand some of the properties of bilayers in biomembranes.14 Gram-negative bacteria such as Escherichia coli present two distinct membranes, outer and inner, which limit the cell boundaries and are separated by a network of peptidoglycan. While the outer membrane contains large numbers of pore-like proteins through which bulk transport may occur, the inner or cytoplasmic membrane contains a number of specific transport systems such as lactose permease (LacY) or the dicarboxylic acid transport system. Beyond its high content in proteins, the inner lipid bilayer is composed of three main phospholipids: phosphatidylethanolamine (PE, zwitterionic, 74% of the total molar phospholipid content), phosphatidylglycerol (PG, bearing a negative charge, 19%), and cardiolipin (CL, bearing two negative charges, 3%).15 Regarding lateral separation, there is considerable evidence of phospholipid domains in the membrane of E. coli.16 Specifically, it seems that PG and PE segregate laterally Received: September 28, 2011 Revised: November 11, 2011 Published: November 16, 2011 701

dx.doi.org/10.1021/la203795t | Langmuir 2012, 28, 701–706

Langmuir

ARTICLE

experiment. The compression barrier speed was 5 cm2 min1. Every πA isotherm was repeated at least three times, and the isotherms showed satisfactory reproducibility. Briefly, the thermodynamic analysis was performed according to the classical formalism. The excess Gibbs energy (GE), which can be calculated from:

into domains, both in vivo17 and in SLBs.18 Importantly, we have recently demonstrated19 the preferential insertion in fluid phases that occurs when LacY is reconstituted in SLBs formed from proteoliposomes of POPE:POPG. The aim of this Article is 2-fold. On the one hand, we aim to investigate the mixing properties at the airwater interface of two phospholipids mimicking the E. coli inner membrane: 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphoethanolamine (POPE) and 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphoglycerol (POPG). On the other hand, we aim to obtain further insight into the characterization of two-dimensional domains in Langmuir Blodgett films (LBs). Regarding the E. coli inner membrane model, domains have been characterized and confirmed in SLBs of POPE:POPG obtained either by extension of liposomes or by successive extraction of two LBs.18-20 However, there is evidence that the solid support and extension conditions have a strong influence on the features of the SLBs observed.21 Normally, to enhance liposome rupture and spreading, or for the successive deposition of two LBs onto a solid substrate, buffers with divalent cations, mostly Ca2+, with concentrations ranging from 2 to 20 mM, are used.22,23 Although a washing step with a buffer free of divalent cations may be performed before observation by AFM, little is known about the distribution of the lipids between the two leaflets of supported bilayers. Besides, this is a very important point that may have a strong influence on the distribution of protein membranes when SLBs are formed from reconstituted proteoliposomes. Therefore, the study of the nanostructure of monolayers of POPE and POPG may provide information not only for their biological significance regarding the interplay between the leaflets in a bilayer, but also on how the nanostructure of the apical monolayer (head groups facing the mica surface) may influence the nanostructure of the distal monolayer (head groups facing the buffer) and in the final orientation of proteins reconstituted in supported systems.24

GE ¼

Z

π 0

½A12  χ1 A1  χ2 A2  dπ

ð1Þ

where A12 is the area per molecule of the mixed monolayer and χ1, A1 and χ2, A2 are the molar fractions and the area per molecule of the pure components 1 and 2, respectively. This expression can be treated as an indication of the interaction and stability of the mixed monolayer at a constant surface pressure π and at constant temperature. The inverse of the isothermal compressibility or elastic modulus of area compressibility (Cs1) was calculated using:   ∂π ð2Þ Cs 1 ¼ ð  AÞ ∂A T, n The derivative of the experimental data was computed by fitting a straight line to a window of area width of 0.2 nm2 molec1 around any given surface pressure value, so that experimental noise was filtered out.

2.3. LangmuirBlodgett Extraction and AFM Imaging and Processing. LB films for AFM observations were transferred onto

freshly cleaved mica at specific surface pressure (25 and 30 mN m1), by lifting the substrate at a constant rate of 0.1 cm2 min1. The transfer ratios were evaluated and found to be near unity, indicating that the mica was almost fully covered with the monolayer. AFM experiments were performed on a Multimode microscope controlled by Nanoscope V electronics (Bruker AXS Corporation, Santa Barbara, CA) 24.0 ( 0.2 °C . Images were acquired in air and in tapping-mode (TM-AFM) operation with beam-shaped silicon oxide tapping tips (37th series B cantilever, MikroMasch, Portland, OR) with a nominal spring constant of 0.3 N m1. All images were recorded at minimum vertical force, maximizing the amplitude set point value and maintaining the vibration amplitude as low as possible. Image processing was performed through commercial NanoScope Analysis Software (Bruker AXS Corporation, Santa Barbara, CA) and Gwyddion software, a modular free program for SPM data analysis (gwyddion.net). Area occupancy (%) of LELC domains on AFM topography images was obtained using particle analysis on Image J software.

2. MATERIALS AND METHODS 1-Palmitoyl-2-oleoyl-sn-glycero-3-phosphoethanolamine (POPE) and 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphoglycerol (POPG), specified as 99% pure, were purchased from Avanti Polar Lipids (Alabaster, AL). For monolayer experiments, the subphase buffer used was Hepes 20 mM, NaCl 150 mM, pH 7.4, CaCl2 20 mM prepared in Ultrapure water (Milli Q reverse osmosis system, 18.2 MΩ cm resistivity). Chloroform and methanol, HPLC grade, were purchased from SIGMA (St. Louis, MO). 2.1. Monolayer Conditions. The monolayers were prepared in a 312 DMC LangmuirBlodgett trough manufactured by NIMA Technology Ltd. (Coventry, England). The trough (total area: 137 cm2) was placed on a vibration-isolated table (Newport, Irvine, CA) and enclosed in an environmental chamber. The resolution of surface pressure measurement was (0.1 mN m1. In all experiments, the temperature was maintained at 24.0 ( 0.2 °C via an external circulating water bath. Before each experiment, the trough was washed with chloroform/methanol and rinsed thoroughly with purified water. The cleanliness of the trough and subphase was ensured before each run by cycling the full range of the trough area and aspirating the airwater surface, while at the minimal surface area, to zero surface pressure. 2.2. Analysis of Isotherms. The experiments and the analysis of the isotherms were carried out as described elsewhere.25 The lipid was dissolved in chloroformmethanol (2:1, v/v) to a final concentration of 1 mg mL1. The corresponding aliquot of lipid was spread onto the surface of the subphase solution with a Hamilton microsyringe. A 15 min period was required to allow the solvent to evaporate before each

3. RESULTS AND DISCUSSION Surface pressurearea (πA) isotherms of pure POPE and POPG and mixed monolayers at 24.0 ( 0.2 °C are shown in Figure 1. The features for the pure phospholipid monolayers are in agreement with isotherms published elsewhere.25,26 In both pure monolayers, a shift to lower molecular areas was observed by comparison to the same compression isotherms performed in the absence of Ca2+.19 This observation is particularly consistent with what has been reported for anionic phospholipids in the presence of calcium where the shift to lower molecular areas is interpreted as a result of the binding capability of Ca2+.27,28 The pure POPE isotherm features the characteristic plateau from ∼0.43 to ∼0.34 nm2 molec1 and collapses at 50.3 mN m1.29 In turn, the POPG isotherm shows a monotonous increase in surface pressure, until it collapses at ∼47.6 mN m1.30 The isotherms for the binary POPE:POPG systems are similar to that of pure POPG, and the molecular areas are within the range of pure monolayers. All monolayers reached maximal values of Cs1 between 25 and 35 mN m1, whereas only the POPE pure monolayer 702

dx.doi.org/10.1021/la203795t |Langmuir 2012, 28, 701–706

Langmuir

ARTICLE

Table 1. Excess Gibbs Energy as a Function of Composition for Mixed POPE:POPG Monolayers at Different Surface Pressures GE (J mol1) χPOPE 1

π (mN m )

Figure 1. Surface pressure (π) as a function of the area per molecule for pure POPE (9), POPE:POPG (3:1) (b), POPE:POPG (1:1) (2), POPE:POPG (1:3) (0), and POPG (O). Inset: Compressibility modulus (Cs1) as a function of area per molecule for the same compositions.

0.25

0.50

0.75

5

2 ( 20

0 ( 20

860 ( 20

10

5 ( 20

2 ( 20

1600 ( 20

15

7 ( 20

3 ( 20

2300 ( 20

20

9 ( 20

5 ( 20

2900 ( 20

25 30

10 ( 30 11 ( 30

6 ( 30 7 ( 30

3400 ( 30 3900 ( 30

35

12 ( 30

9 ( 30

4400 ( 30

40

12 ( 50

10 ( 40

4800 ( 50

45

13 ( 50

12 ( 50

5200 ( 50

and used throughout this work. Actually, this is the same concentration used in previous works18,20 where the effect of Ca2+ on bilayers of POPE:POPG was studied in detail. To gain insight into the specific nanostructure of POPE: POPG mixtures, POPE:POPG monolayers at different χPOPE were spread onto the buffer and transferred onto a mica substrate at different surface pressures, specifically, 25 and 30 mN m1, both considered biologically relevant.13 According to the Cs1 values at these pressures, the monolayer shows an LE phase in both cases. As mentioned above, the characterization of the phase state based on Cs1 is merely phenomenological. Indeed, the pressures at which the LBs are transferred correspond to a region in which both LC and LE phases coexist in the monolayers. Nevertheless, the extension of each domain depends on the specific surface pressure and the conditions in which the LBs were extracted. Attempts to analyze the LELC distribution in monolayers have been based on interfacial techniques such as epifluorescence,37 Brewster angle microscopy,38 and confocal microscopy.39 Among the microscopic techniques, however, AFM can provide better resolution at the nanostructure level. This was demonstrated in a study in which confocal microscopy was used in conjunction with AFM to characterize LCLE domains in L-α-dipalmitoylphosphatidylcholine.36 The aim of the present study was to use AFM to examine the nanostructure of LBs in the POPE:POPG system. Under the prevalent conditions, this enables us to explore the nanostructure of apical monolayers when blistering bilayers of POPE:POPG20 and, particularly, to get a quantitative approach to LC and LE coexistence. For the sake of clarity, the mixing effect on the surface properties of POPE and POPG monolayers at different χPOPE will be rather evaluated at 25 mN m1, whereas the effect of increasing surface pressure in LCLE distribution will be further discussed. As observed from the topographic images in Figures 2a,e and 3a,e, at any of the surface pressure regimes studied, both pure POPE and POPG monolayers showed two domains with different size and morphology. It is worth mentioning that these features, as well as the reach nanostructure of the mixed monolayers (see below), are only observable at 20 mM Ca2+ (see Figure 2, Supporting Information). The measured step heights between both domains in pure POPE and POPG are provided in Table 2 and fall within the range of differences expected between LC and LE phases.40,41 In both cases, LC domains protrude a

showed a decrease of Cs1, before it collapsed at π ≈ 40 mN m1 (Figure 1). For all pure and mixed monolayers, the Cs1 values remain far below 100 mN m1, which is considered to be the lowest value for the LC phase.31 This behavior is qualitatively similar to what has been observed in the absence of Ca2+19 and confirms that under the present conditions the monolayer remains in LE phases. Obviously, this does not imply, however, that the nanostructure of the LBs would be the same. To explore the stability and possible phase separation, the values of GE for the binary mixtures at different surface pressures were calculated from eq 1. For POPE molar fraction (χPOPE) of 0.25 and 0.50, values of GE show small positive deviation from ideality in the whole range of surface pressures investigated (Table 1). These values are almost negligible and compatible with an ideal mixing behavior, which is characterized by similar interactions between phospholipids in the χPOPE = 0.25 and 0.50. Remarkably, for χPOPE = 0.75, GE values were significant and negative, which implies attractive interactions. Besides, as can be seen in Table 1, the higher is the surface pressure, the more negative the GE values became. This is an indication of monolayer stability. The value at 30 mN m1 is much more negative than in the absence of Ca2+19 and points to the direct implication of divalent cations in enhancing monolayer stability. The attractive forces observed at χPOPE = 0.75 may arise from the high packing array of PE and PG molecules through hydrogen bonds.32,33 In relative terms, this means that the contribution of the repulsive interactions to the net GE values becomes relevant only at higher proportions of POPG. This behavior is similar to that observed in monolayers of diacyl saturated PG and PE species34 spread on water. It is well-known that the addition of divalent cations causes the zeta potential to become less negative35 and promotes the adsorption of liposomes onto mica.36 In our study, to obtain firm adsorption onto mica acceptable transfer ratios, a convenient amount of Ca2+ was previously established. Because POPG and mica are both negatively charged, only phospholipid patches were obtained (see Figure 1, Supporting Information). To get a complete coverage of the substrate, 20 mM of Ca2+ was required 703

dx.doi.org/10.1021/la203795t |Langmuir 2012, 28, 701–706

Langmuir

ARTICLE

Figure 2. AFM topographic images in air of POPE:POPG monolayers at different POPE molar fractions (χPOPE): 0.0, 0.25, 0.50, 0.75, and 1.00 (a, b, c, d, and e, respectively). Samples were extracted at a constant π of 25 mN m1. Image size bare stands for 5 μm and color scale for 12 nm. Profile analysis along the white dashed line is displayed above the corresponding image, showing height differences of protruding LC domains.

Figure 3. AFM topographic images in air of POPE:POPG monolayers at different POPE molar fractions (χPOPE): 0.0, 0.25, 0.50, 0.75, and 1.00 (a, b, c, d, and e, respectively). Samples were extracted at a constant π of 30 mN m1. Image size bar stands for 5 μm and color scale for 12 nm. Profile analysis along the white dashed line is displayed above the corresponding image, showing height differences of protruding LC domains.

Table 2. Average Step Heights of Protruding LC Domains in POPE:POPG Monolayers at Different χPOPE Values and for the Two Surface Pressures Studied, 25 and 30 mN m1

Table 3. Percentage of LC Phase Domains Occupancy in POPE:POPG Monolayers at Different χPOPE Values and for the Two Surface Pressures Studied, 25 and 30 mN m1 χPOPE

χPOPE 0.0

0.25

0.50

0.75

1.0

0.0

Step Height (nm)

0.25

0.5

0.75

1.0

LC Phase Occupancy (%)

25 mN m1 1.15 ( 0.05 0.27 ( 0.04 1.05 ( 0.10 0.64 ( 0.08 0.80 ( 0.09 30 mN m1 1.20 ( 0.07 0.38 ( 0.04 1.11 ( 0.13 1.01 ( 0.10 0.98 ( 0.06

mean height difference of ∼1 nm, the area occupied being 1.8% and 31.9% of the total surface at 25 mN m1 for pure POPE and POPG, respectively (see Table 3). The size of the LC domains in the POPG monolayers may be attributed to the clustering effect of the binding of Ca2+ to negatively charged phospholipids. As discussed for other acidic phospholipids,25,26 Ca2+ typically binds two PG head groups, which reduces the electrostatic repulsion between them and promotes aggregation. The scenario changes slightly when considering different POPE:POPG mixtures. While the topographic images obtained at χPOPE = 0.25 (Figures 2b and 3b) were featureless, at χPOPE = 0.50 small microdomains were clearly distinguished at both surface pressures studied (Figures 2c and 3c). These observations are consistent with the prediction of the near-ideal mixing behavior of both components given by the thermodynamic analysis performed above. Thus, at χPOPE = 0.25, most of the individual POPE molecules are surrounded by POPG molecules, which hinders the Ca2+ binding. Consequently, the LC domains cannot be seen. As the proportion of POPE is raised to χPOPE = 0.50,

25 mN m1

1.8

36.5

60.5

31.9

30 mN m1

15.9

40.6

70.1

35.4

POPE appears to reach its limit of solubility in POPG, and a second phase appears. Conversely, for χPOPE = 0.75, LC and LE phases are clearly distinguishable (Figures 2d and 3d). The attractive forces predicted by the thermodynamic analysis may be responsible for the lateral segregation observed. Besides, the increase in POPE proportion results in an increase of the LC phase protrusion domain from the small, isolated LC domains at χPOPE = 0.5 to larger, more continuous LC microdomains at χPOPE = 0.75. This is illustrated in Table 3, which shows that the area occupancy for the LC domains increased from 31.9% to 60.5% over the LE phase. Obviously, the composition of the LC domains is different in each case. Thus, while pure LC domains of POPG are observed in images a in Figures 2 and 3, the LC domains observed at χPOPE = 0.75 (image c in the same figures) represent a POPE enriched phase. As pointed out in a theoretical study,31 hydrogen bonds between PE and PG molecules may be responsible for this behavior. We also studied the effect of the compression on the nanostructure and domain distribution of POPE:POPG monolayers. 704

dx.doi.org/10.1021/la203795t |Langmuir 2012, 28, 701–706

Langmuir

ARTICLE

Figure 4. Area occupancy (%) of LELC phase on POPE:POPG monolayers at different POPE molar fractions (χPOPE = 0.0, 0.25, 0.50, 0.75, and 1.00) and for the two surface pressures studied, 25 and 30 mN m1. LE phase is displayed in gray, LC in black, and the absence of phase coexistence in white.

Specifically, we examined the effect of an increase of 5 mN m1 on the LELC distribution. Compression normally increases the LC domains (compare Figures 2 and 3). However, while an increase of 5 mN m1 led to the formation of larger globular domains for pure POPG, the shape and size of the domains did not change for pure POPE. Strikingly, at χPOPE = 0.25, there was no evidence of LC growth, and at χPOPE = 0.50 the shape and size of the domains were similar to those of pure POPE. For χPOPE = 0.75 at 25 mN m1, the LC domains formed extensive regions. Figure 2d shows two large LC domains separated by a rim formed by the LE phase. At 30 mN m1, the large domains coalesced and surrounded small, irregular shaped LE domains (Figure 3d). LC step height obtained from profile analysis of topographical images of POPE:POPG monolayers showed a slight increase upon increasing the surface pressure. This effect is a consequence of the arrangement of the phospholipids tails in a more ordered conformation.42 To complement these observations, we have listed the area occupied by each domain for each χPOPE and surface pressure studied (Table 3 and Figure 4). Despite differences in composition and morphology, the most remarkable increases occurred for pure POPG (14.1%) and for χPOPE = 0.75 (9.6%) monolayers. Elsewhere, we have reported two observations on the POPE: POPG system that are, in principle, unrelated. These are (i) LacY inserts preferentially in the temperature-dependent Lα phase,43 and (ii) the nanomechanical properties (breakthrough and adhesion forces) determined in the Lα phase of SLBs are ∼2.5-fold lower than those of Lα phases obtained by blistering two monolayers through the LB technique.20 In the present study, we now provide evidence of the predominance of the LC phase in the apical leaflet of the χPOPE = 0.75 monolayer. This observation indicates that the use of 20 mM Ca2+ in the formation of SLBs from vesicles with negatively charged lipids leads to asymmetrical bilayers. This may also influence the final nanostructure of the bilayers obtained by successive extraction of two LBs.20 This is consistent with observations reported in other systems, in which an asymmetrical distribution in the supported bilayer results in the exposure of larger areas of the LE phase in the distal leaflet of the bilayer.44 Therefore, considering that the viscosity and diffusion coefficients in the LE phase are similar to those measured in the Lα phase,39 we conclude that the insertion of LacY and other transmembrane proteins40,45 depends on the influence between both leaflets of the bilayers. Beyond this, when LacY is reconstituted in preformed supported lipid bilayers,43

and knowing the preference of the protein for the fluid phases,19 it could be worthy to investigate if the nanostructure of the apical layer may influence its distribution and self-segregation. Works are undertaken in our laboratory to elucidate this problem.

’ ASSOCIATED CONTENT

bS

Supporting Information. AFM topographic image of a LB of pure POPG extracted in the absence of Ca2+ and POPE: POPG mixed monolayers extracted in the presence of 10 mM Ca2+. This material is available free of charge via the Internet at http://pubs.acs.org.

’ AUTHOR INFORMATION Corresponding Author

*Phone: (+34) 934035986. Fax: (+34) 934035987. E-mail: [email protected]. Present Addresses †

INSERM U1006, Institut Curie, 25 rue d’Ulm, Paris, France.

’ ACKNOWLEDGMENT C.S.-G. is recipient of a FPI fellowship from the Ministerio de Ciencia e Innovacion of Spain. This work has been supported by grant CTQ-2008-03922/BQU from Ministerio de Ciencia e Innovacion of Spain. J.H.-B. acknowledges the support of COST Action AFM4NanoMedBio (TD1002). ’ REFERENCES (1) Jacobson, K.; Mouritsen, O. G.; Anderson, R. G. W. Nat. Cell Biol. 2007, 9:1, 7–14. (2) Nyholm, T. K.; Ozdirekcan, S.; Killian, J. A. Biochemistry 2007, 46, 1457–1465. (3) Poveda, J. A.; Fernandez, A. M.; Encinar, J. A.; Gonzalez-Ros, J. M. Biochim. Biophys. Acta 2008, 1778, 1583–1590. (4) Rilfors, L.; Lindblom, G. Colloids Surf., B 2002, 26, 112–124. (5) Bagatolli, L. Biochim. Biophys. Acta 2006, 1758, 1541–1556. (6) Domenech, O.; Redondo, L.; Picas, L.; Morros, M. T.; HernandezBorrell, J. J. Mol. Recognit. 2007, 20, 546–553. (7) Singh, S.; Keller, D. J. Biophys. J. 1991, 60, 1401–1410. (8) Houslay, M. D.; Stanley, K. K. Dynamics of Biological Membranes; Wiley & Sons: New York, 1982; pp 116. (9) Ipsen, J. H.; Karlstron, G.; Mouritsen, O. G.; Wennerstron, H.; Zuchermam, M. J. Biochim. Biophys. Acta 1987, 905, 167–172. 705

dx.doi.org/10.1021/la203795t |Langmuir 2012, 28, 701–706

Langmuir

ARTICLE

(10) Petty, M. C. Langmuir-Blodgett Films: An Introduction; Cambridge University Press: New York, 1996. (11) Calvez, P.; Demers, E.; Boisselier, E.; Salese, Ch. Langmuir 2011, 27, 1373–1379. (12) Cantor, R. S. J. Phys. Chem. B 1997, 101, 1723–1725. (13) Marsh, D. Biochim. Biophys. Acta 1996, 1285, 183–223. (14) Dorfler, H. D.; Koth, C.; Retting, W. Langmuir 1995, 11, 4803–4810. (15) Dowhan, W. Annu. Rev. Biochem. 1997, 66, 199–232. (16) Mileskowskaya, E.; Dowhan, M. J. Bacteriol. 2000, 182, 1172–1175. (17) Vanounou, S.; Parola, A. H.; Fishov, I. Mol. Microbiol. 2003, 49, 1067–1079. (18) Picas, L.; Montero, M. T.; Morros, A.; Caba~nas, M. E.; Seantier, B.; Milhiet, P. E.; Hernandez-Borrell, J. J. Phys. Chem. B 2009, 112, 4646–4655. (19) Suarez-Germa, C.; Montero, M. T.; Ignes-Mullol, J.; Hernandez J. Phys. Chem. B 2011, 115, 12778–12784. Borrell, J.; Domenech, O. (20) Picas, L.; Suarez-Germa, C.; Montero, M. T.; HernandezBorrell, J. J. Phys. Chem. B 2010, 114, 3543–3549. (21) Richter, R. P.; Mauray, N.; Brisson, A. R. Langmuir 2005, 21, 299–304. (22) Richter, R.; Mukhopadhay, A.; Brisson, A. Biophys. J. 2003, 85, 3035–3047. (23) Jass, J.; Tj€arnhage, T.; Puu, G. Biophys. J. 2000, 79, 3153–3163. (24) Berquand, A.; Levy, D.; Gubellini, F.; Le Grimellec, C.; Milhiet, P. E. Ultramicroscopy 2007, 107, 928–933.  Sanz, F.; Montero, M. T.; Hernandez-Borrell, J. (25) Domenech, O.; Biochim. Biophys. Acta 2006, 1758, 213–221. (26) Takamoto, D. Y.; Lipp, M. M.; Von Nahmen, A.; Lee, K. Y. C.; Warning, A. J.; Zasadzinski, J. A. Biophys. J. 2001, 81, 153–169. (27) Sovago, M.; Wurpel, G. W. H.; Smits, M.; M€uller, M.; Bonn, M. J. Am. Chem. Soc. 2007, 129, 11079–11084. (28) Park, C. K.; Schmitt, F. J.; Evert, L.; Schwarts, D. K.; Israelachvili, J. N.; Knobler, C. M. Langmuir 1999, 15, 201–206.  Ignes-Mullol, J.; Montero, M. T.; Hernandez(29) Domenech, O.; Borrell, J. J. Phys. Chem. B 2007, 111, 10946–10951. (30) Sanchez-Martín, M. J.; Haro, I.; Alsina, M. A.; Busquets, M. A.; Pujol, M. J. Phys. Chem. B 2010, 114, 448–456. (31) Davies, J. T.; Rideal, E. K. Interfacial Phenomena; Academic Press: New York, 1963; p 263.  Torrent-Burgues, J.; Merino, S.; Sanz, F.; (32) Domenech, O.; Montero, M. T.; Hernandez-Borrell, J. Colloids Surf., B 2005, 41, 233–238. (33) Murzyn, K.; Rog, T.; Pasenkiewicz-Gierula, M. Biophys. J. 2005, 88, 1091–1103. (34) Wydro, P.; Witkowska, K. Colloids Surf., B 2009, 72, 32–39. (35) McLaughlin, S.; Mulrine, N.; Gresalfi, T.; Vaio, G.; McLughlin, A. J. Gen. Physiol. 1981, 77, 445–473. (36) Garcia-Manyes, S.; Oncins, G.; Sanz, F. Biophys. J. 2005, 89, 1812–1826. (37) Nag, K.; Boland, C.; Rich, N.; Keough, K. M. W. Biochim. Biophys. Acta 1991, 1068, 157–160. (38) Honig, D.; Mobius, D. J. Phys. Chem. 1991, 95, 4590–4592. (39) Veatch, S. L.; Keller, S. L. Biophys. J. 2005, 85, 3074–3083. (40) Hollars, C. W.; Dunn, R. C. Biophys. J. 1998, 75, 342–353. (41) Saulnier, P.; Foussard, F.; Boury, F.; Proust, J. E. J. Colloid Interface Sci. 1989, 218, 40–46. (42) M€ohwald, H. Phospholipid Monolayers. In Handbook of Biological Physics; Lipowsky, R., Sackmann, E, Eds.; Elsevier Science: New York, 1995; Chapter 4. (43) Picas, L.; Carretero-Genevrier, A.; Montero, M. T.; VazquezIbar, J. L.; Seantier, B.; Milhiet, P. E.; Hernandez-Borrell, J. Biochim. Biophys. Acta 2010, 1798, 1014–1019. (44) Rosseti, F. F.; Textor, M.; Reviakine, I. Langmuir 2006, 22, 3476–347. (45) Seeger, H. K.; Bortolotti, C. A.; Alessandrini, A.; Facci, P. J. Phys. Chem. B 2009, 113, 16654–16659.

706

dx.doi.org/10.1021/la203795t |Langmuir 2012, 28, 701–706