ARTICLE pubs.acs.org/Langmuir
Structure and Phase Behavior of Archaeal Lipid Monolayers Christoph Jeworrek,† Florian Evers,‡,|| Mirko Erlkamp,† Sebastian Grobelny,† Metin Tolan,‡ Parkson Lee-Gau Chong,§ and Roland Winter*,† †
Physical Chemistry I, Faculty of Chemistry, TU Dortmund University, Dortmund, Otto-Hahn-Strasse 6, D-44221 Dortmund, Germany Faculty of Physics and DELTA, TU Dortmund University, Maria-Goeppert-Mayer-Strasse 2, D-44221 Dortmund, Germany § Department of Biochemistry, Temple University School of Medicine, Philadelphia, Pennsylvania 19140, United States ‡
bS Supporting Information ABSTRACT: We report X-ray reflectivity (XRR) and grazing incidence X-ray diffraction (GIXD) measurements of archaeal bipolar tetraether lipid monolayers at the airwater interface. Specifically, Langmuir films made of the polar lipid fraction E (PLFE) isolated from the thermoacidophilic archaeon Sulfolobus acidocaldarius grown at three different temperatures, i.e., 68, 76, and 81 °C, were examined. The dependence of the structure and packing properties of PLFE monolayers on surface pressure were analyzed in a temperature range between 10 and 50 °C at different pH values. Additionally, the interaction of PLFE monolayers (using lipids derived from cells grown at 76 °C) with the ion channel peptide gramicidin was investigated as a function of surface pressure. A total monolayer thickness of approximately 30 Å was found for all monolayers, hinting at a U-shaped conformation of the molecules with both head groups in contact with the interface. The monolayer thickness increased with rising film pressure and decreased with increasing temperature. At 10 and 20 °C, large, highly crystalline domains were observed by GIXD, whereas at higher temperatures no distinct crystallinity could be observed. For lipids derived from cells grown at higher temperatures, a slightly more rigid structure in the lipid dibiphytanyl chains was observed. A change in the pH of the subphase had an influence only on the structure of the lipid head groups. The addition of gramicidin to an PLFE monolayer led to a more disordered state as observed by XRR. In GIXD measurements, no major changes in lateral organization could be observed, except for a decrease of the size of crystalline domains, indicating that gramicidin resides mainly in the disordered areas of the monolayer and causes local membrane perturbation, only.
’ INTRODUCTION About 90% of the lipid components in the plasma membrane of the thermoacidophilic archaeon Sulfolobus acidocaldarius are tetraether lipids,1 among which the polar lipid fraction E (PLFE) is one of the main constituents.2 PLFE contains a mixture of bipolar tetraether lipids with either a glycerol dialkylcalditol tetraether (GDNT or calditoglycerocaldarchaeol; ∼90% of total PLFE) or a glycerol dialkylglycerol tetraether (GDGT, or caldarchaeol; ∼10% of total PLFE) skeleton.24 Both GDGT and GDNT are bisubstituted in the polar head group regions and are thus designated as bipolar tetraether lipids. The nonpolar regions of these lipids consist of a pair of 40-carbon biphytanyl chains, each of which contains up to four cyclopentane rings. The number of the cyclopentane rings in each biphytanyl chain increases with increasing growth temperature.5 The structure of the backbone of bipolar tetraether lipids is shown in Figure 1.6,20 In PLFE liposomes, lipids span the entire lamellar structure, forming a monomolecular spanning membrane,7 in contrast to the bilayer structure formed by monopolar diester (or diether) phospholipids in mammalian cells.810 Since PLFE is one of the r 2011 American Chemical Society
major polar lipid components in the plasma membrane of S. acidocaldarius, PLFE liposomes have been used as a model system for studying thermoacidophilic archaeal membranes. PLFE liposomes exhibit high thermal stability and unusually low solute permeability when compared to monopolar diester or diether liposomes (reviewed by Chong4). The thermal stability with respect to leakage of dye originally trapped inside PLFE liposomes has been attributed to the negative charges on the membrane surface and to the tight and rigid lipid packing.1113 The low proton permeability in PLFE liposomes has been ascribed to the network of hydrogen bonds between the sugar residues of PLFE exposed at the outer face of the membrane12 and to the tight and rigid lipid packing.13,14 Both the dye leakage and proton permeation experiments suggest that membrane packing, in either the hydrocarbon or the polar head group regions or both, is a central issue in understanding PLFE lipid membranes. Received: May 31, 2011 Revised: July 23, 2011 Published: September 12, 2011 13113
dx.doi.org/10.1021/la202027s | Langmuir 2011, 27, 13113–13121
Langmuir
ARTICLE
Figure 1. Bipolar tetraether lipids: molecular structures of (a) GDGT (or caldarchaeol) and (b) GDNT (or calditolglycerocaldarchaeol). GDG(N)T-0 and GDG(N)T-4 contain zero and four cyclopentane rings, respectively. The number of cyclopentane rings in each biphytanyl chain can vary from zero to four for the polar lipid fraction E (PLFE) derived from S. acidocaldarius. The different head groups of GDNT and GDGT are presented at the bottom.6,20
To study membrane packing in PLFE liposomes, lateral and rotational diffusions of membrane probes have been examined. The lateral mobility of 1-palmitoyl-2-(10-pyrenyl)-decanoyl)-snglycero-3-phosphatidylcholine (PyrPC) in PLFE liposomes was found to be highly restricted and only became appreciable at temperatures > 48 °C.15 This indicates a significant structural change near 48 °C in the PLFE hydrocarbon core. These studies also suggest that the hydrocarbon region of PLFE liposomes is rigid and tightly packed below ∼48 °C. Above ∼48 °C, the hydrocarbon core of PLFE membranes begins to gain appreciable membrane fluidity, which would be required for the functionality of archaeal membranes. The polar head group region of PLFE membranes, on the other hand, may still be rigid and tightly packed through the hydrogen-bond network16,17 at elevated temperatures (>48 °C) to maintain a large proton gradient (pH 23 outside and pH 6.5 inside the cell) across the membrane at the growth temperature. This proposition4 explains why low proton permeability and appreciable membrane fluidity can occur at the same time in thermoacidophiles at high growth temperatures.4 This point is supported by a spin-label study,4 which showed that at high temperatures (∼85 °C) the nonitol (more precisely, calditol) head group of tetraether lipids from the thermoacidophilic archaeon S. solfataricus was relatively immobile, whereas the hydrocarbon region possessed some degree of mobility.18 The structural properties and phase behavior of PLFE lipid membranes have been explored to some extent. In previous studies, some structural properties of PLFE liposomal membranes in bulk solution were revealed by using SAXS,6 and the thermal phase transitions and volumetric properties were studied by calorimetric methods.19,20 However, little is known about the structural composition and the packing properties of PLFE monolayer films spread at the airwater interface. To shed light on the structural properties of PLFE membranes under such
conditions, surface X-ray scattering techniques provide an appropriate methodology.2125 Here, we present a comprehensive study on the influence of cell growth temperature, subphase temperature, lateral film pressure, and subphase pH value on the structure and ordering of PLFE monolayer films at the airwater interface. Additionally, the interaction of PLFE monolayers (derived from cells grown at 76 °C) with the ion channel peptide gramicidin D was investigated at various temperatures with a subphase pH 6.5, mimicking the conditions inside thermoacidophilic archaeon cells. The results lead to a better understanding of the thermal stability and the packing properties of PLFE lipid membranes. Such knowledge is also needed for potential applications of tetraether lipids. Owing to their high stability, archaeal lipids are widely discussed for technological applications such as coatings on solid surfaces, crystallization of membrane-bound proteins, immunoassays, and delivery vehicles for vaccines or drugs.4
’ MATERIALS AND METHODS Materials. PLFE lipids were extracted from S. acidocaldarius, grown at 68, 76, and 81 °C, as previously described.2,11 Chloroform, methanol, hydrochloric acid, and gramicidin from Bacillus aneurinolyticus were purchased from Sigma (Taufkirchen, Germany) and used without further purification. Aqueous solutions were prepared using water filtered through a Milli-Q purification system, yielding a specific electrical resistivity of >18 MΩ 3 cm. The pH of the aqueous subphase was adjusted to the desired pH. PLFE was dissolved in a mixture of chloroform, methanol, and water (15:3.8:1 vol) and spread on the aqueous subphase. For samples containing gramicidin, gramicidin (2 wt %, 2.5 mol % in PLFE monolayers) was dissolved into the lipid solution prior to spreading at the airwater interface. GramicidinPLFE interactions were studied at a subphase pH value of 6.5. 13114
dx.doi.org/10.1021/la202027s |Langmuir 2011, 27, 13113–13121
Langmuir
ARTICLE
Figure 2. (left) X-ray reflectivity data (symbols) obtained from PLFE lipid monolayers (cells grown at 68 °C) at the airwater interface, presented together with the best fits (solid lines) normalized to the Fresnel reflectivity, RF. For clarity, the reflectivity curves are shifted along the ordinate with increasing film pressure. (right) Normalized electron density profiles as retrieved from the fits on the left-hand side. PLFE monolayers were spread on subphases at different subphase temperatures of (a) 10 and (b) 50 °C, and analyzed as a function of lateral film pressure.
Methods. Archaeal lipid stock solutions were spread at the air water interface on a Langmuir trough (Riegler & Kirstein, Potsdam, Germany). Lipid films were examined at various lateral film pressures, Π, between 10 and 40 mN/m and different subphase temperatures, ranging from 10 to 50 °C. Further details of the experimental procedures and methods are presented in the Supporting Information. The X-ray scattering experiments were conducted at the liquid surface diffractometer of beamline ID10B at the synchrotron light source ESRF (European Synchrotron Radiation Facility, Grenoble, France).26 A monochromatic X-ray wavelength, λ, of 1.558 or 1.523 Å (corresponding to photon energies, E, of 7.95 or 8.13 keV) was selected by a diamond (111) crystal. Reflectivity measurements were performed in an angular range 0.4αc < αi < 30αc with the critical angle of total reflection of the airwater interface, αc (αc = 0.15° at the given wavelength). A typical X-ray reflectivity scan took about 20 min. In a typical grazing incidence X-ray diffraction (GIXD) scan, the detector with a horizontal resolution of ΔQxy = 0.0051 Å1 is moved in 200 steps in an angular range of 19° < 2θ < 29° (corresponding to 1.36 Å1 < Qxy < 2.07 Å1), counting 15 s per step. For these experiments, the Langmuir trough was mounted on the diffractometer. We have used a similar experimental setup in previous studies on lipid films and lipidpeptide interactions.24,25,27 The analysis of surface X-ray scattering data21,22,2833 and the experimental parameters are outlined in detail in the Supporting Information. X-ray reflectivity (XRR) data are plotted as R/RF versus Qz, with the reflectivity, R, the Fresnel reflectivity, RF, and the vertical wave vector transfer, Qz. In a GIXD experiment, the momentum transfer has horizontal and vertical components, Qxy and Qz.21,30
’ RESULTS Vertical Structure of PLFE Monolayers as a Function of Lateral Surface Pressure. In the following, we report X-ray
scattering data on PLFE monolayers spread on subphases of different temperatures and pH values as a function of lateral film pressure. Figure 2 presents X-ray reflectivity data and inferred electron density profiles of PLFE monolayers, at varying lateral film pressures and different subphase temperatures. In this experiment, PLFE lipids were isolated from cells grown at 68 °C. A full list of fitting parameters and additional X-ray reflectivity data (for PLFE lipids derived from cells grown at 76 and 81 °C) are presented in the Supporting Information (Table S1, Figures S1 and S2). All X-ray reflectivity data of pure PLFE monolayers could be adequately described by a two-layer model, accounting for a PLFE lipid head group and a lipid hydrocarbon chain region. The accuracy of the fitting parameters is less than or equal to (0.005 e/Å3 for the electron density, (0.2 Å for the layer thickness, and (0.3 Å for the interfacial roughness (for more details see the Supporting Information). Thus, from the analysis of the XRR measurements, variations in lateral surface pressure and subphase temperature/pH can be related to changes in the vertical structure of PLFE monolayers. For convenience, we will use Tgrow to refer to the cell growth temperature and Tsub for the subphase temperature. For PLFE (Tgrow at 68 °C) measured at Tsub = 10 °C (Figure 2), an increase in film pressure results in a successively rising total monolayer thickness (from 30.2 Å at Π = 10 mN/m to 32.8 Å at Π = 30 mN/m), which can be ascribed to an increase in thickness of the head group region (from 5.6 to 9.2 Å). The head group and hydrocarbon chain thicknesses have been determined from changes in the slope of the electron density profile F(z) in the ranges between about 10 and 10 Å, and 10 and 30 Å, respectively. The electron density of the head group region decreases slightly and, concomitantly, the electron density 13115
dx.doi.org/10.1021/la202027s |Langmuir 2011, 27, 13113–13121
Langmuir
ARTICLE
Figure 3. Influence of subphase temperature on the PLFE monolayer structure at different surface pressures (right, 10 mN/m; left, 30 mN/m) for PLFE isolated from cells grown at 68 (upper) and 76 °C (lower).
of the lipid chains increases at higher film pressures. At Tsub = 50 °C (Figure 2), changes of the monolayer thickness with increasing film pressure are not as pronounced as at low Tsub temperatures. Increasing film pressure from Π = 10 mN/m to Π = 20 mN/m leads to an enhanced electron density of the head group and an increase in the thickness of the hydrocarbon region. From Π = 20 mN/m to Π = 30 mN/m, the electron density profile (EDP) does not vary markedly. PLFE monolayers composed of PLFE with Tgrow at 76 and 81 °C (Figures S1 and S2 in the Supporting Information) show a different response to increased film pressure at low and elevated subphase temperatures. In these cases, variations of the total monolayer thickness are more pronounced at a subphase temperature of 50 °C than at 10 °C (Table S1 in the Supporting Information). The strongest effect of Tsub on the total monolayer thickness was observed at a PLFE monolayer with Tgrow at 81 °C, where the overall thickness changes between 10 and 30 mN/m by only 0.8 Å at Tsub = 10 °C and by 3.4 Å at Tsub = 50 °C. For both PLFEs with Tgrow values at 76 and 81 °C, the head group and the hydrocarbon chain region thicknesses increase with increasing lateral film pressure. Influence of Subphase Temperature on the Vertical Structure of PLFE Monolayers. Figure 3 highlights the effect of varying the subphase temperature (Tsub) on the electron density profile of PLFE monolayers at different film pressures. Again, lipid monolayers composed of PLFE obtained from different Tgrow's show slightly different responses to variations of Tsub. However, two general trends become visible: the total monolayer thickness increases with rising film pressure and decreases with rising Tsub. (Table S1 in the Supporting Information). For monolayers composed of PLFE derived from Tgrow = 68 °C, changing Tsub from 10 to 50 °C at a low film pressure of Π = 10 mN/m mainly alters the organization of the film’s head group; in
particular, the thickness of the hydrocarbon chain region decreases by 2 Å, while the thickness of the head group region increases by 3 Å. At an elevated surface pressure of Π = 30 mN/ m, the decrease of total monolayer thickness (2 Å) with rising Tsub is pronounced. At this high lateral film pressure, the electron density and the thickness of the head group as well as those of the chain region are diminished. For lipid monolayers composed of PLFE derived from Tgrow at 76 °C, both at a low film pressure of Π = 10 mN/m and at a high film pressure of Π = 30 mN/m, increasing Tsub from 10 °C over 40 °C to 50 °C successively decreases the electron density of both the head group and the chain region, while the thickness changes are not that striking. For example, at Π = 30 mN/m, the total monolayer thickness drops from 31.4 Å over 31.2 Å to 31.0 Å with increasing Tsub (Table S1 in the Supporting Information). Effect of Cell Growth Temperature (Tgrow) on the Vertical Structure of PLFE Monolayers. Figure S3 in the Supporting Information summarizes the influence of cell growth temperature (Tgrow) on the electron density profile of PLFE monolayers under different environmental conditions. In all cases, monolayers composed of PLFE derived from cells grown at 76 and 81 °C show rather similar structures. However, films composed of PLFE derived from cells grown at 68 °C exhibit an enlarged total monolayer thickness (by ∼1 Å) at low Tsub and high film pressure Π as well as at elevated Tsub and low Π. Lateral Structure of PLFE Monolayers As Revealed by Grazing Incidence Diffraction. A typical GIXD map, I(Qxy, Qz), as obtained from PLFE monolayer (Tgrow = 81 °C) at the airwater interface with a lateral film pressure of Π = 30 mN/m and a subphase temperature of Tsub = 10 °C is shown in Figure 4. Two distinct peaks are observed, indicating the existence of highly ordered domains in the monolayer with a distorted hexagonal packing. Please note that that the highly ordered domains occupy 13116
dx.doi.org/10.1021/la202027s |Langmuir 2011, 27, 13113–13121
Langmuir
Figure 4. Two-dimensional contour plots of the intensity I(Qxy,Qz) along the horizontal (Qxy) and vertical (Qz) scattering directions as obtained from the PLFE monolayer (cell growth temperature 81 °C) at a lateral film pressure of Π = 30 mN/m and a subphase temperature of 10 °C.
Figure 5. GIXD pattern I(Qxy) obtained by integrating along Qz for an archaeal lipid monolayer (cell growth temperature 81 °C) at a lateral film pressure of Π = 30 mN/m and a subphase temperature of 10 °C (square symbols). Peaks were fitted by Gaussian functions (solid lines). The peaks of the distorted hexagonal packing can be denoted by their indices;from left to right;{10, 01}, {11}. (inset) Typical Bragg rod intensity profile I(Qz) obtained by integrating along the Qxy region of the Bragg peak. The absence of a peak at Qz 6¼ 0 indicates little or no molecular tilt.
only a fraction of the interfacial area and therefore all parameters derived from GIXD represent only these crystalline domains in the monolayer and not the overall monolayer structure. In all GIXD measurements, two peaks were found, suggesting an identical structure of the crystalline unit cell for all samples studied. All fit parameters are presented in Table S2 in the Supporting Information in detail. In Figure 5, the GIXD pattern I(Qxy) obtained by integrating along Qz for the PLFE monolayer (Tgrow = 81 °C) at a lateral film pressure of Π = 30 mN/m is presented (square symbols). The peaks of the distorted hexagonal packing can be denoted by their indices;from left to right;{10, 01} and {11}. All peaks were fitted by Gaussian functions in order to locate the position of the peak maxima as well as to determine their full widths at halfmaximum (fwhm). From the positions of the peak maxima, the
ARTICLE
repeat distances of the lattice (d-spacings) were calculated. The two lattice constants were calculated from the maximum positions of the peaks to a = b ≈ 4.58 Å and c ≈ 5.08 Å. The angle between the two lattice vectors, γ, is approximately 112.6°, which is slightly smaller than the expected value of 120° for an ideal hexagonal packing. The occupied area per hydrocarbon chain in the crystalline domains, A, was directly calculated from these lattice parameters to 19.4 Å2/chain. The lateral crystalline domain lengths, Lxy, could be calculated using the Scherrer equation from the full widths at half-maximum of the Bragg peaks, corrected for instrumental resolution.21 For the {10, 01} peak at about Qxy = 1.49 Å1 and the {11} peak at about Qxy = 1.65 Å1, Lxy values on the order of 2200 and 1860 Å were obtained, respectively. These values represent huge areas of high crystallinity on the monolayer at Π = 30 mN/m and Tsub = 10 °C. In the inset of Figure 5, a typical Bragg rod intensity profile I(Qz) obtained by integrating along the Qxy region of the {10, 01} Bragg peak is depicted. The maximum at Qz = 0 is the so-called VineyardYoneda peak,34 which arises from the interference between X-rays diffracted up into the Bragg rod and rays diffracted down and then reflected up by the interface. The absence of a peak at Qz 6¼ 0 indicates negligible or no molecular tilt of the hydrocarbon chains relative to the surface normal. The vertical lengths of the crystalline domains, Lz, was obtained from the full width at half-maximum of the Bragg rods. For the two rods corresponding to both Bragg peaks, heights of the vertical domain of 30.3 and 30.9 Å were found, respectively. PLFE lipids derived from different growth temperatures (68, 76, and 81 °C) were spread at the airwater interface at Tsub = 10 °C and compressed to a lateral film pressure of 30 mN/ m. According to the GIXD data, no significant differences in the lattice parameters and dimensions of crystalline domains were found between the different growth conditions (for details see Table S2 in the Supporting Information). GIXD signals were only obtained at Tsub of 10 and 20 °C. No crystalline domains could be found at higher temperatures, probably due to the higher thermal fluctuations and temperature-induced disorder of the lipid hydrocarbon region. The lattice parameters of PLFE monolayer with Π = of 30 mN/m at Tsub = 20 °C show only insignificant variations to the parameters observed at 10 °C. However, the dimension of the crystalline domains decreases with increasing Tsub temperature by about 20% in the lateral direction (Lxy) and by approximately 15% from 30 to 26 Å in the vertical direction (Lz). Effect of Subphase pH Value on PLFE Monolayer Structure. The lipid monolayers have been studied at three different pH values: 2.2, 5.0, and 6.5 (Table S1 in the Supporting Information). A pH value of 2.2 mimics the extracelluar environment of thermoacidophilic archaeon cells, and a pH value of 6.5 the pH value inside archaeon cells. At the higher pH values (especially at pH 6.5), PLFE monolayers exhibit a different behavior in response to increasing lateral film pressure in the XRR experiments: while, at a pH value of 2.2, increasing film pressure Π leads to a continuously rising monolayer thickness, at pH 5 and 6.5, the largest thickness can be found at low film pressures. At intermediate Π values, the monolayer thickness decreases and rises slightly at high values of Π again. These changes can be related to a growing thickness of the chains (∼4 Å) and a decreasing thickness of the head groups (∼4 Å) upon increasing Π. In the GIXD experiments, no significant influence of the subphase pH on the unit cell parameters can be found. The 13117
dx.doi.org/10.1021/la202027s |Langmuir 2011, 27, 13113–13121
Langmuir
ARTICLE
Figure 6. Interaction of PLFE monolayers with gramicidin D. (left) X-ray reflectivity data (symbols) obtained from archaeal lipid monolayers at the airwater interface, presented together with the best fits (solid lines) normalized to the Fresnel reflectivity, RF. For clarity, the reflectivity curves are shifted along the ordinate with increasing film pressure. (right) Normalized electron density profiles as retrieved from the fits on the left-hand side. PLFE monolayers with and without gramicidin D were prepared at different film pressures and a subphase temperature of 20 °C.
lateral crystalline domain length decreases by about 10% with increasing pH value from 2.2 to 6.5, only. With increasing lateral film pressure, the PLFE monolayer (Tgrow of 76 °C) on a subphase with pH 6.5 at 20 °C becomes more ordered. The lateral crystalline domain lengths increase between 10 and 30 mN/m from approximately 760 to 1620 Å. The height of the crystalline domains increases slightly by about 5%. Interestingly, the area per hydrocarbon chain also increases slightly with increasing lateral film pressure, from 19.1 Å2/chain at 10 mN/m to 19.3 Å2/chain at 30 mN/m. At the same time, the angle of the unit cell, γ, increases from 111.5 to 112.9°. Interaction between Archaeal Lipids and the Model Ion Channel Peptide Gramicidin. Figure 6 shows exemplarily X-ray reflectivity data and electron density profiles of PLFE monolayers (Tgrow = 76 °C) spread on a subphase with pH 6.5 and a temperature of 20 °C as well as data of the interaction of PLFE with gramicidin D on the same subphase (see also Table S1 in the Supporting Information). Remarkably, when both gramicidin and PLFE are present, a less ordered, more heterogeneous interfacial structure appears at low film pressures. Such structures were adequately described by a one-layer model (cf. cyan curve in Figure 6) because of the absence of pronounced Kiessig oscillations, which are characteristic of well-defined lipid films. In the presence of gramicidin D, the total monolayer thickness rises with increasing film pressure. Comparing the film structures in the presence and absence of gramicidin at high film pressures reveals that the addition of gramicidin leads to an increase of the electron density of the chain region, while the electron density of the head group is slightly decreased (Figure 6). The addition of 2 wt % (2.5% mol) gramicidin into PLFE monolayer (Tgrow at 76 °C) at the airwater interface (pH 6.5) at 20 °C and a lateral film pressure of 30 mN/m leads to no significant changes in the unit cell parameters and vertical dimensions of crystalline domains observed by GIXD. This indicates that the interaction of the gramicidin with the lipid occurs mainly in the disordered areas of the monolayer. The lateral length of these highly ordered domains decreases to about 50% compared to samples without 2.5 mol % gramicidin at 30 mN/m. PLFE monolayer (Tgrow of 76 °C) containing 2 wt % gramicidin has been investigated at 20 °C for four different film pressures (10, 20, 30, and 40 mN/m). Only minor changes in the unit cell parameters could be observed, as for example a slight increase of the area per lipid chain by 1% with increasing film
pressure. There is a strong change in the dimensions of the crystalline domains with growing lateral film pressure. The lateral domain size decreases linearly by approximately 40% between 10 and 40 mN/m to about 920 Å.
’ DISCUSSION Lipid Conformation at the AirWater Interface. For all PLFE lipid films, an overall monolayer thickness of around 30 Å has been found in the XRR measurements. This indicates that the lipids adopt a U-shaped conformation at the interface with both lipid head groups in contact with the subphase rather than an upright standing conformation with only a single head group in contact with the aqueous subphase. This finding is corroborated by the fact that the XRR curves of the PLFE monolayers can be accurately described applying a two-layer model accounting for a single head group region in contact with the subphase and a single hydrocarbon chain region protruding into the air. Interestingly, the vertical length of crystalline domains found by GIXD describing the thickness of the crystalline part of the monolayer is slightly larger than 30 Å at elevated lateral film pressures, indicating a tight packing of the whole lipid film including the hydrocarbon chain and head group. Effect of Lateral Film Pressure and Cell Growth Temperature (Tgrow) on the Structure of PLFE Monolayers. At pH 2.2, an increase of the total monolayer thickness of all monolayers is observed with increasing film pressure at all subphase temperatures. For lipids isolated from cells grown at 68 °C, this increase originates;at a low subphase temperature of 10 °C;mainly from the lipid head group region. In the GIXD measurements, the size of crystalline domains in lateral and vertical directions increases significantly with increasing film pressure. The increase of thickness in the lipid head group region might indicate a conformational change of the lipid head group with increasing film pressure. With increasing lateral pressure, the head groups probably adopt a conformation that is more compact and the head groups are standing upright relative to the interface. This would explain the increase in the head group region observed by XRR. No more detailed structural information can be obtained, since in XRR the electron density profile measured is laterally averaged. At high film pressures, the lipid hydrocarbon chains become more tightly packed, as indicated by the huge crystalline domain lengths on the order of 16002200 Å found by GIXD. This tighter packing results in an increase in the electron density 13118
dx.doi.org/10.1021/la202027s |Langmuir 2011, 27, 13113–13121
Langmuir of the chain region concomitant with an increase in head group thickness and a decrease in head group electron density, as observed in the XRR measurements. The fact that no significant molecular tilt angle of the lipids could be found by the GIXD scans indicates a similar cross-sectional area of the lipid chain and head group region at all film pressures, which is possible by filmpressure-dependent changes in head group orientation. At a high subphase temperature of 50 °C, the changes in chain packing are much less pronounced and the majority of the changes occur at low film pressures between 10 and 20 mN/m. The PLFE lipids isolated from cells grown at 76 and 81 °C show a different behavior. Here, the effects of lateral film pressure are stronger at high temperatures (50 °C) than at cold conditions (10 °C). This can be explained by the difference in the hydrocarbon chain region of the lipids. With increasing growth temperature, the number of cyclopentane rings increases and with it the rigidity of the lipids chain region increases, too. For PLFE lipids isolated from higher Tgrow, the thermal energy is not high enough to disrupt the monolayer at low Tsub temperatures. On the other hand, the chains of the lipids isolated from cells grown at 68 °C are more flexible. Therefore, effects of film pressure are stronger for PLFE lipids derived from low growth temperatures compared to more rigid PLFE lipids isolated from higher growth temperatures. At high Tsub values, even at elevated lateral film pressures the chains stay rather disordered, and therefore, the effect of film pressure is much smaller than at low Tsub values for lipids isolated from cells grown at 68 °C. The strongest increase in total monolayer thickness with increasing lateral film pressure has been found in a monolayer built by archaeal lipids from cells grown at 81 °C on a subphase with a temperature of 50 °C. The hydrocarbon chains of this lipid have the highest rigidity. At a low film pressure, the monolayer is very thin, especially its head group region. The available area per head group region is therefore large. With increasing film pressure, the chains straighten, resulting in a larger monolayer thickness. At the same time, the available area for the lipid head groups decreases, forcing the head groups also into an elongated conformation as observed in the XRR measurements, where the head group thickness increases from 4.5 Å at 10 mN/m to 7.7 Å at 30 mN/m. Compared to lipids grown at 76 and 81 °C, the lipids obtained from the lowest growth temperature (68 °C) exhibit an enlarged total monolayer thickness at low temperature and high film pressure as well as high temperature and low film pressure. These effects can be explained by the lower number of cyclopentane rings and hence higher flexibility of the hydrocarbon chain region of this lipid which allows for a better packing at high lateral film pressures, leading to a more elongated conformation and therefore to a larger total monolayer thickness. This can be shown comparing the head group thicknesses of lipids isolated from Tgrow of 76 and 81 °C measured at Tsub = 10 °C and 30 mN/m of 6.3 and 7.4 Å, respectively, compared to a value of 9.2 Å for the head group of the lipid obtained from Tgrow = 68 °C. The increased overall thickness at high subphase temperature and low film pressure in a monolayer of the lipid isolated from cells grown at 68 °C can be explained by the low rigidity of the hydrocarbon chains of this lipid. The lipids can be tightly packed already at low lateral film pressures forcing the lipid head groups into an extended conformation, which increases the overall thickness of the monolayer. The head group thickness at 50 °C and 10 mN/m for lipids isolated from cells grown at 68 °C is 8.6 Å, compared to 5.0 and 4.5 Å for lipids derived from Tgrow = 76 and 81 °C, respectively.
ARTICLE
Interestingly, there is no effect of the number of cyclopentane rings on the parameters of the unit cell in GIXD measurements, indicating a nearly identical crystalline packing for all the archaeal lipids. This indicates that at high film pressures the packing is governed by the lipid head group region, which is identical in all lipids in this study. Additionally, the dimensions of crystalline areas were similar for all lipids. The main difference between lipids isolated from cells grown at different temperatures as observed by XRR originates therefore probably mainly from the less crystalline, more disordered areas of the monolayer. Influence of Subphase Temperature on the Structure of PLFE Monolayers. With increasing temperature, the monolayer thickness of all lipids at pH 2.2 decreases. The most pronounced behavior can be found in a monolayer of PLFE lipids isolated from cells grown at 68 °C. Here, the thicknesses of the head group and the hydrocarbon region decrease with increasing temperature at both low and high lateral film pressures. The lipids isolated from higher temperature exhibit a smaller change in overall monolayer thickness, but a decrease in the electron density of both layers. In the more flexible lipids isolated from cells grown at low temperature, the increasing subphase temperature increases the conformational disorder leading to a shorter overall monolayer thickness, but the packing density remains similar at all temperatures. The more rigid lipids isolated from cells grown at higher temperatures cannot change their conformation as strongly as the lipids obtained at low growth temperatures. Here, the increase in Tsub leads to stronger fluctuations in the membrane leading to a less dense packing, but a barely changed monolayer thickness. In both cases, this increase in conformational disorder is the reason no signals could be obtained from highly crystalline domains in the GIXD experiments at high subphase temperature. The results thus reveal a phase transition from a coexistence region of a crystalline phase and a disordered phase at low temperature to a completely disordered phase at high temperature, similar to what has been observed previously in the bulk phase.19,20 As the conformational properties in the bulk phase (multilamellar vesicles) are very different from those of PLFE at the waterair interface (U-shaped conformations), no direct structural correlations can be discussed. In the GIXD measurements, no significant difference in the unit cell parameters can be found between measurements with subphase temperatures of 10 and 20 °C at 30 mN/m. The size of crystalline domains decreases by 20% in the lateral direction and by 15% in the vertical direction between 10 and 20 °C. This illustrates the strong degree of decreasing overall order in the monolayer with increasing temperature, where much less and only smaller highly ordered domains can be found. Effect of Subphase pH Value on PLFE Monolayer Structure. In the XRR experiments, a distinct effect of the subphase pH can be found. At pH 2.2, the overall monolayer thickness increases with increasing lateral film pressure. At pH 5 and 6.5, the largest monolayer thickness is found at the lowest film pressure. Here, the lipid head group region is more than twice as large as at pH 2.2 and decreases with increasing film pressure. Simultaneously, the thickness of the lipid chain region increases as already observed at low pH. Thus, the major effect of pH is on the head group region. Again, a different conformation might be adopted at the elevated pH, as indicated by the larger head group thickness and significantly lower electron density of this layer compared to lower pH values. The parameters of the crystalline unit cell obtained by GIXD do not change with changing subphase 13119
dx.doi.org/10.1021/la202027s |Langmuir 2011, 27, 13113–13121
Langmuir
conformation of the chain. With no rotational degrees of freedom; as in a solid-ordered lipid phase;a much tighter packing is possible, resulting in a significantly smaller area per hydrocarbon chain as observed here by GIXD. In addition, in these monolayer systems, the PLFE chains adopt an U-shaped conformation extending into the air, which might allow for a more efficient packing. Please note that the lipids outside these highly ordered crystalline domains detected by GIXD must have a significantly larger area per hydrocarbon chain. This is supported by the much larger overall areas per molecule at the airwater interface reported in the literature.40
’ CONCLUSIONS To conclude, we found a total monolayer thickness of ∼30 Å for PLFE membranes, hinting at a U-shaped conformation of the molecules at the water/air interface. Compared to diester phospholipid monolayers, films built of PLFE lipids show a smaller response to temperature and lateral film pressure. In general, increasing temperature increases the disorder in the monolayer and increasing film pressure decreases it. Not only the two hydrocarbon chains of a single archaeal lipid, but also the chains of neighboring lipid molecules, adopt an extremely tight packing, in particular at high film pressures. For lipids derived from cells grown at higher temperatures, a more rigid structure in the lipid dibiphytanyl chains was observed. A change in the pH of the subphase had an influence only on the structure of the lipid head groups. At low temperatures (10 and 20 °C), the PLFE monolayer film exhibits highly crystalline domains with a distorted hexagonal packing coexisting with extended disordered regions of the monolayer. At high temperature (50 °C), the monolayer shows an overall disordered structure with no evidence of crystalline domains. The number of cyclopentane rings, which increases with increasing growth temperature, does not change the parameters of the domain’s unit cell, indicating that at high film pressures the packing is governed by the lipid head group region. The addition of the model ion channel gramicidin led to a minor increase in conformational disorder of the PLFE monolayer. No major changes in lateral organization could be observed at low temperatures except for a decrease of the size of crystalline domains, indicating that gramicidin resides mainly in the disordered areas of the monolayer and causes local membrane perturbations, only. ’ ASSOCIATED CONTENT
bS
Supporting Information. Experimental methods; fitting parameters of X-ray reflectivity curves and calculated parameters of grazing incidence diffraction measurements as a function of film pressure Π; X-ray reflectivity data obtained from PLFE lipid monolayers at the airwater interface and normalized electron density profiles; effect of PLFE growth temperature on the PLFE monolayer structure at different subphase temperatures and surface pressures. This material is available free of charge via the Internet at http://pubs.acs.org.
’ AUTHOR INFORMATION Corresponding Author
*E-mail:
[email protected]. Present Addresses
)
pH, indicating a nearly identical packing for all pH values. The crystalline domain length in the lateral direction decreases with increasing pH, indicating a lower order in the monolayer at high pH. This could be explained by an increase of charge repulsion of the phosphate groups at the higher pH value. Interaction between Archaeal Lipids and the Model Ion Channel Peptide Gramicidin. According to the XRR data, the addition of gramicidin leads to a more heterogeneous, disordered state at low film pressures. At higher film pressures, the presence of gramicidin leads to a slightly higher overall thickness of the monolayer compared to a measurement without gramicidin. The main difference between measurements of pure archaeal lipids and a system containing gramicidin is an increasing electron density of the lipid chain region and decreasing head group region with increasing film pressure in the presence of gramicidin. This indicates the incorporation of gramicidin into the hydrophobic region of the monolayer. In GIXD, no major changes in the unit cell parameters can be found. The lateral length of crystalline domains decreases to 50% after the addition of gramicidin at 30 mN/m. Also, for samples containing gramicidin, a decreasing lateral domain size with increasing film pressure is observed. This indicates the incorporation of gramicidin in the disordered areas of the monolayer. With increasing film pressure, gramicidin incorporation increases the disorder leading to smaller lateral crystalline domain lengths. Comparison to Other Lipid Systems. Compared to diester phospholipid monolayers, the films built by PLFE lipids have a higher electron density of the head group region (about 1.4(Fhead/Fwater) for PLFE lipids compared to ∼1.25(Fhead/Fwater) for DPPC at 30 mN/m), but at the same time this layer is significantly thinner for PLFE lipids (the typical head group thickness of DPPC is on the order of 11 Å compared to ∼69 Å found for PLFE in this study).27 The overall thickness of a diester phospholipid monolayer is nearly identical with the thickness found for the archaeal lipid film, which is only possible for PLFE lipids acquiring an U-shape conformation, which would be in agreement with earlier suggestions.4,3741 The area per hydrocarbon chain of PLFE lipids found by GIXD, representing only highly ordered domains in the monolayer, is surprisingly smaller than the one of diester phospholipids at the same lateral film pressure, although the structure of the chains is more complex in archaeal lipids. Compared to the value of ∼19.3 Å2/chain found in this study, the area per lipid chain of diester phospholipids at comparable lateral film pressures is considerably larger. For example, for DPPC and DPPG values of ∼23.3 and ∼22.6 Å2/chain, respectively, have been observed.35,36 Not only the two hydrocarbon chains of a single PLFE lipid but also the chains of neighboring lipid molecules adopt an extremely tight packing. The area per hydrocarbon chain of a glycolipid LPS Re, where one lipid has six chains, is, with a value of ∼19.1 Å2/chain, in the same range as the one found for archaeal lipids.25 This value is again significantly smaller than that of diester phospholipids, where two chains are attached to one lipid head group, but the chains are highly flexible. This flexibility of the chains seems to be an important factor controlling the size of the area occupied by each chain. In glycolipids, the restriction of flexibility originates from the large head group, where six chains are attached close to each other. Here, in PLFE lipids, this restriction of flexibility in the crystalline domains is imposed by the cyclopentane rings in the chain, by the two opposing head groups, both attached to the same pair of chains, and by the strong kink in the chain due to the U-shaped
ARTICLE
Condensed Matter Physics Laboratory, Heinrich-Heine-Universit€at D€usseldorf, Universit€atstrasse 1, D-40225 D€usseldorf, Germany.
13120
dx.doi.org/10.1021/la202027s |Langmuir 2011, 27, 13113–13121
Langmuir
’ ACKNOWLEDGMENT Financial support from the DFG and the NSF (DMR0706410 and -1105277) is gratefully acknowledged. We thank Oleg Konovalov and Alexei Vorobiev for their help setting up the X-ray scattering experiment and the ESRF for providing synchrotron radiation. ’ REFERENCES (1) Langworthy, T. A.; Pond, J. L. In Thermophiles: General, Molecular, and Applied Microbiology; Brock, T. D., Ed.; John Wiley & Sons: New York, 1986; p 107. (2) Lo, S.-L.; Chang, E. L. Biochem. Biophys. Res. Commun. 1990, 167, 238–243. (3) Sugai, A.; Sakuma, R.; Fukuda, I.; Kurosawa, N.; Itoh, Y. H.; Kon, K.; Ando, S.; Itoh, T. Lipids 1995, 30, 339–344. (4) Chong, P. L.-G. Chem. Phys. Lipids 2010, 163, 253–265. (5) De Rosa, M.; Esposito, E.; Gambacorta, A.; Nicholaus, B.; Bu'Lock, J. D. Phytochemistry 1980, 19, 827–831. (6) Chong, P. L.-G.; Zein, M.; Khan, T. K.; Winter, R. J. Phys. Chem. B 2003, 107, 8694–8700. (7) Elferink, M. G. L.; de Wit, J. G.; Demel, R.; Driessen, A. J. M.; Konings, W. N. J. Biol. Chem. 1992, 267, 1375–1381. (8) Winter, R. Biochim. Biophys. Acta 2002, 1595, 160–184. (9) Winter, R.; Czeslik, C. Z. Kristallogr. 2000, 215, 454–474. (10) Winter, R.; Jeworrek, C. Soft Matter 2009, 5, 3157–3173. (11) Chang, E. L. Biochem. Biophys. Res. Commun. 1994, 202, 673–679. (12) Elferink, M. G. L.; de Wit, J. G.; Driessen, A. J. M.; Konings, W. N. Biochim. Biophys. Acta 1994, 1193, 247–254. (13) Komatsu, H.; Chong, P. L.-G. Biochemistry 1998, 37, 107–115. (14) Baba, T.; Minamikawa, H.; Masakatsu, H.; Handa, T. Biophys. J. 2001, 81, 3377–3386. (15) Kao, Y. L.; Chang, E. L.; Chong, P. L.-G. Biochem. Biophys. Res. Commun. 1992, 188, 1241–1246. (16) Elferink, M. G. L.; de Wit, J. G.; Driessen, A. J. M.; Konings, W. N. Eur. J. Biochem. 1993, 214, 917–925. (17) Vilalta, I.; Gliozzi, A.; Prats, M. Eur. J. Biochem. 1996, 240, 181–185. (18) Bruno, S.; Cannistraro, S.; Gliozzi, A.; De Rosa, M.; Gambacorta, A. Eur. Biophys. J. 1985, 13, 67–76. (19) Chong, P. L.-G.; Ravindra, R.; Khurana, M.; English, V.; Winter, R. Biophys. J. 2005, 89, 1841–1849. (20) Chong, P.L.-G.; Sulc, M.; Winter, R. Biophys. J. 2010, 99, 3319–3326. (21) Jensen, T. R.; Kjaer, K. In Novel Methods to Study Interfacial Layers; M€obius, D., Miller, R., Eds.; Elsevier Science: Amsterdam, 2001; Vol. 11, p 205. (22) X-ray and Neutron Reflectivity: Principles and Applications; Daillant, J., Gibaud, A., Eds.; Lecture Notes in Physics; Springer: Berlin, 2009. (23) Jacquemain, D.; Wolf, S. G.; Leveiller, F.; Deutsch, M.; Kjaer, K.; Als-Nielsen, J.; Lahav, M.; Leiserowitz, L. Angew. Chem., Int. Ed. 1992, 31, 130–152. (24) Evers, F.; Jeworrek, C.; Tiemeyer, S.; Weise, K.; Sellin, D.; Paulus, M.; Struth, B.; Tolan, M.; Winter, R. J. Am. Chem. Soc. 2009, 131, 9516–9521. (25) Jeworrek, C.; Evers, F.; H€owe, J.; Brandenburg, K.; Tolan, M.; Winter, R. Biophys. J. 2011, 100, 2169–2177. (26) Smilgies, D.-M.; Boudet, N.; Struth, B.; Konovalov, O. J. Synchrotron Radiat. 2005, 12, 329–339. (27) Paulus, M.; Lietz, D.; Sternemann, C.; Shokuie, K.; Evers, F.; Tolan, M.; Czeslik, C.; Winter, R. J. Synchrotron Radiat. 2008, 15, 600–605. (28) Tolan, M. X-Ray Scattering from Soft-Matter Thin Films; Materials Science and Basic Research; Springer Tracts in Modern Physics; Springer: Berlin, 1999. (29) Russell, T. P. Mater. Sci. Rep. 1990, 5, 171–271.
ARTICLE
(30) Als-Nielsen, J.; Jacquemain, D.; Kjaer, K.; Leveiller, F.; Lahav, M.; Leiserowitz, L. Phys. Rep. 1994, 246, 251–313. (31) Evers, F.; Shokuie, K.; Paulus, M.; Sternemann, C.; Czeslik, C.; Tolan, M. Langmuir 2008, 24, 10216–10221. (32) Parratt, L. G. Phys. Rev. 1954, 95, 359–369. (33) H€usecken, A. K.; Evers, F.; Czeslik, C.; Tolan, M. Langmuir 2010, 26, 13429–13435. (34) Vineyard, G. H. Phys. Rev. B 1982, 26, 4146–4159. (35) Aroti, A.; Leontidis, E.; Maltseva, E.; Brezesinski, G. J. Phys. Chem. B 2004, 108, 15238–15245. (36) Neville, F.; Ishitsuka, Y.; Hodges, C. S.; Konovalov, O.; Waring, A. J.; Lehrer, R.; Lee, K. Y. C.; Gidalevitz, D. Soft Matter 2008, 4, 1665–1674. (37) Bakowsky, U.; Rothe, U.; Antonopoulos, E.; Martini, T.; Henkel, L.; Freisleben, H. Chem. Phys. Lipids 2000, 105, 31–42. (38) Heathcock, C. H.; Finkelstein, B. L.; Aoki, T.; Poulter, C. D. Science 1985, 229, 862–864. (39) de Rosa, M. Thin Solid Films 1996, 284285, 13–17. (40) Rolandi, R.; Schindler, H.; de Rosa, M.; Gambacorta, A. Eur. Biophys. J. 1986, 14, 19–27. (41) Gulik, A.; Luzzati, V.; de Rosa, M.; Gambacorta, A. J. Mol. Biol. 1985, 182, 131–149.
13121
dx.doi.org/10.1021/la202027s |Langmuir 2011, 27, 13113–13121
