3D-Membrane Stacks on Supported Membranes Composed of Diatom

Article pubs.acs.org/Langmuir

3D-Membrane Stacks on Supported Membranes Composed of Diatom Lipids Induced by Long-Chain Polyamines Oliver Grab̈ ,† Maryna Abacilar,‡ Fabian Daus,‡ Armin Geyer,‡ and Claudia Steinem*,† †

Institute of Organic and Biomolecular Chemistry, University of Göttingen, Tammannstr. 2, 37077 Göttingen, Germany Faculty of Chemistry, Philipps-University Marburg, Hans-Meerwein-Str. 4, 35032 Marburg, Germany

‡

S Supporting Information *

ABSTRACT: Long-chain polyamines (LCPAs) are intimately involved in the biomineralization process of diatoms taking place in silica deposition vesicles being acidic compartments surrounded by a lipid bilayer. Here, we addressed the question whether and how LCPAs interact with lipid membranes composed of glycerophospholipids and glyceroglycolipids mimicking the membranes of diatoms and higher plants. Solid supported lipid bilayers and monolayers containing the three major components that are unique in diatoms and higher plants, i.e., monogalactosyldiacylglycerol (MGDG), digalactosyldiacylglycerol (DGDG), and sulfoquinovosyldiacylglycerol (SQDG), were prepared by spreading small unilamellar vesicles. The integrity of the membranes was investigated by fluorescence microscopy and atomic force microscopy showing continuous flat bilayers and monolayers with small protrusions on top of the membrane. The addition of a synthetic polyamine composed of 13 amine groups separated by a propyl spacer (C3N13) results in flat but three-dimensional membrane stacks within minutes. The membrane stacks are connected with the underlying membrane as verified by fluorescence recovery after photobleaching experiments. Membrane stack formation was found to be independent of the lipid composition; i.e., neither glyceroglycolipids nor negatively charged lipids were required. However, the formation process was strongly dependent on the chain length of the polyamine. Whereas short polyamines such as the naturally occurring spermidine, spermine, and the synthetic polyamines C3N4 and C3N5 do not induce stack formation, those containing seven and more amine groups (C3N7, C3N13, and C3N18) do form membrane stacks. The observed stack formation might have implications for the stability and expansion of the silica deposition vesicle during valve and girdle band formation in diatoms.

■

INTRODUCTION Lipid membranes are of pivotal importance for every prokaryotic and eukaryotic organism, as they generate defined reaction compartments and can interact with a number of biomacromolecules. To study the interaction of biomacromolecules under well-controlled conditions in vitro, model membrane systems are highly desirable as their lipid composition is tunable.1 However, while lipid compositions, mimicking the plasma membranes of animal and bacterial cells, have intensively been used, only a few studies using lipid compositions commonly found in plants have been published. The three major components that are unique in plants are glyceroglycolipids, i.e., monogalactosyldiacylglycerol (MGDG), digalactosyldiacylglycerol (DGDG), and sulfoquinovosyldiacylglycerol (SQDG). Some studies report on the use of glyceroglycolipid-containing vesicles,2−7 black lipid membranes,8 and lipid monolayers at the air−water interface and deposited on a solid support by the Langmuir−Blodgett technique.9−11 However, to the best of our knowledge, supported lipid bilayers (SLBs) composed of mixtures of all three of these glyceroglycolipids have not been produced. © XXXX American Chemical Society

Although SLBs lack a large internal compartment as compared to vesicles, they have the advantage of being planar and longterm stable.12 Thus, they are accessible by a number of surface sensitive methods such as fluorescence microscopy and atomic force microscopy, which allow monitoring their integrity and biophysical properties as well as their lateral organization.13 They are also versatile tools to monitor biomacromolecule− lipid interactions in a quantitative manner.14 In this study, we made use of a lipid composition determined for the diatom Cyclotella meneghiniana15 to prepare SLBs from vesicle spreading allowing us to investigate the interaction of longchain polyamines (LCPAs) with these membranes. Diatoms are unicellular algae protected by a cell wall made of amorphous silica.16 Parts forming this cell wall are produced in highly specific, acidic, membrane enclosed organelles, the silica deposition vesicles (SDVs).17−21 The current model for biomineralization includes those biomolecules that have been Received: July 13, 2016 Revised: September 2, 2016

A

DOI: 10.1021/acs.langmuir.6b02575 Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir

Glimmer “V5”, Plano, Wetzlar, DE) were fixed onto glass slides with UV curing glue. Before vesicle addition the upper mica layers were freshly cleaved with adhesive film, and the areas around the mica slices were covered with 0.5 mm thick Teflon foil. SUV suspensions with a concentration of 0.5 mg mL−1 (500 μL) were added and incubated for 30 min at room temperature. The samples were rinsed five times with acetate buffer (5 × 500 μL, 50 mM KCl, 40 mM NaOAc/HOAc, pH 5.5) and stored in buffer. For those experiments, in which additional vesicles were added after SLB formation, they were rinsed more often (11 × 500 μL). Preparation of Supported Lipid Monolayers. Supported lipid monolayers were prepared on hydrophobic functionalized glass surfaces. Cover slides (D 263 M Schott glass, ibidi GmbH, Munich, DE) were cut into quadratic pieces (6 × 6 mm2) and fixed with UV curing glue onto glass slides. After rinsing with detergent solution (Hellmanex, Hellma Analytics, Müllheim, DE) and ultrapure water (2×), for 15 min each, the slides were dried under vacuum at 90 °C for 2 h. Afterward, the cover slides were covered with hexamethyldisilazane (20 μL), and the liquid was allowed to evaporate overnight at room temperature (Supporting Information). Further sample preparation and vesicle spreading was as described above for SLBs on mica surfaces. Long-Chain Polyamines (LCPAs). LCPAs are termed CXNY, where Y is the number of amine groups and X the number of carbon atoms of the alkyl spacer. C3N13 was synthesized by solid phase synthesis as described previously.50 C3N4, C3N5, C3N7, and C3N18 were also synthesized by solid phase synthesis according to a previously published protocol.31 LCPA stock solutions were prepared in ultrapure water and added to the solid supported membranes for at least 30 min at room temperature. Concentrations of the LCPAs are given with respect to the number of nitrogen atoms in the molecule (cN = 30 μM, Supporting Information) for better comparison. Stock solutions of FITC−C3N13 were prepared in acetate buffer and added directly to the solid supported membranes (cN = 30 μM). Samples were incubated for at least 30 min at room temperature. If SUVs were added in addition to LCPAs, SUVs (0.2 mg mL−1) were mixed with LCPAs and added to the solid supported membranes to final concentrations of 4 μg mL−1 (SUVs) and 30 μM (polyamines). All measurements were performed after 2 h of incubation at room temperature. Synthesis of FITC−C3N13. The C3N13 precursor was synthesized by solid phase peptide synthesis as described previously.50 Before cleavage from the resin the precursor resin was swollen in dimethylformamide (DMF, 2.0 mL mmol−1) for 30 min at ambient temperature under a nitrogen atmosphere. After filtration the resin was treated with N,N-diisopropylethylamine (4.0 equiv) and fluorescein isothiocyanate (FITC, 1.0 equiv) in DMF (2.0 mL mmol−1) for 24 h at ambient temperature.51 Finally, the resin was washed several times with DMF, methanol, and dichloromethane before it was dried under vacuum. FITC−C3N13 was cleaved from the resin with 95% aqueous trifluoroacetic acid (TFA).52 After shaking at ambient temperature for 3 h the resin was filtered and washed with TFA. The combined filtrates were concentrated under reduced pressure and precipitated from cold, dry diethyl ether (30.0 mL). The precipitate was washed three times with diethyl ether and lyophilized from water. The nonspecifically labeled FITC−C3N13 was obtained as a yellow voluminous solid. 1H NMR (500 MHz, DMSO-d6): δ [ppm] = 10.27−10.09 (bs, 2H, OH), 9.22−8.43 (m, 14H), 7.97−7.91 (m, 1H, CHarom), 7.90−7.81 (m, 1H, CHarom), 7.77−7.71 (m, 1H, CHarom), 7.24−7.15 (m, 1H, CHarom), 6.72−6.65 (m, 1H, CHarom), 6.60−6.48 (m, 2H, CHarom), 3.69−3.23 (m, 32H, CH2), 3.23−2.71 (m, 20H, CH2), 2.16−1.76 (m, 12H, CH2), 1.73−1.59 (m, 4H, CH2), 1.52−1.39 (m, 4H, CH2). 13C NMR (125 MHz, DMSO-d6): δ [ppm] = 133.5, 128.7, 123.3, 121.0, 112.5, 102.2, 59.6, 51.8, 48.8, 43.8, 29.0, 22.3, 19.8. Confocal Laser Scanning Microscopy and Fluorescence Recovery after Photobleaching (FRAP). Confocal fluorescence microscopy and FRAP experiments were performed on a FV1200 laser scanning microscope (Olympus, Tokyo, JP) with either a 20× (NA = 0.5, UMPLFLN 20XW, Olympus, Tokyo, JP) or 60× water immersion objective (NA = 1, LUMPLFLN 60XW, Olympus, Tokyo, JP). A 488

found to be associated with biosilica and which can be divided into a soluble and an insoluble fraction. The soluble fraction, i.e., organic molecules that become soluble after dissolving the biosilica with a mildly acidic solution of ammonium fluoride, contains LCPAs,22 silaffins,23 and silacidins.24 The insoluble fraction comprises organic material with well-defined nano- and micropatterns25,26 probably interacting with the soluble parts and serving as a template for silica formation.27,28 While the aggregation of most of these biomolecules has been studied in solution,26,29−31 the influence of the lipid bilayer surrounding the SDV (silicalemma) and its interaction with such biomacromolecules has not been addressed. In this study, we have focused on soluble LCPAs to elucidate their interaction with lipid membranes resembling the composition of those of diatoms. LCPAs are linear molecules with a number of amine groups bridged by alkyl linkers. Analysis of LCPAs from various diatom species revealed species-specific mixtures of LCPAs with different chain lengths, thus suggesting a direct relationship between amine structure and the patterning of the diatom’s cell wall. The molecular mass of these LCPAs ranges from about 600 to 1500 Da.22,32 While LCPAs are rather specific for diatoms, shorter polyamines, e.g., spermidine and spermine, are omnipresent in all pro- and eukaryotic cells and are involved in various biological processes.33−38 Polyamines buffer the pH, ionic strength, and osmotic pressure in cells39,40 and also interact with nucleic acids41−43 and plasma membranes.44 Besides binding to lipid bilayers,45 they can also mediate lipid− lipid interactions like enhancing vesicle aggregation and fusion.46−49 Here, we used chemically well-defined LCPAs to address the question how their interaction influences SLBs composed of diatom specific lipids.

■

EXPERIMENTAL SECTION

Materials. Monogalactosyldiacylglycerol (MGDG), digalactosyldiacylglycerol (DGDG), 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphoglycerol (POPG), and 1,2-dioleoyl-sn-glycero-3-phosphocholine (DOPC) were purchased from Avanti Polar Lipids (Alabaster, AL). Sulfoquinovosyldiacylglycerol (SQDG) was obtained from Larodan (Solna, SE). Sulforhodamine 101−1,2-dihexadecanoyl-sn-glycero-3phosphoethanolamine (Texas Red−DHPE), 2-(4,4-difluoro-5-methyl-4-bora-3a,4a-diaza-s-indacene-3-dodecanoyl)-1-hexadecanoyl-snglycero-3-phosphocholine (Bodipy-C12-HPC), and fluorescein isothiocyanate (FITC) were purchased from ThermoFisher Scientific (Waltham, MA). Spermine and spermidine were obtained from SigmaAldrich (St. Louis, MO), and hexamethyldisilazane was from Merck Chemicals (Darmstadt, DE). UV glue was from Norland optical adhesive (Norland Products, Cranbury, NJ). All other chemicals were standard analytical grade products. Aqueous solutions were prepared in ultrapure water (Milli-Q Gradient A 10, Millipore, Eschborn, DE). Preparation of Small Unilamellar Vesicles (SUVs). Lipid stock solutions (1−10 mg mL−1) were prepared in chloroform/methanol (5:1) and mixed in a glass test tube yielding the desired molar lipid ratios and masses needed for each experiment. Solvent was removed under a nitrogen stream at 30 °C. Residual solvent was removed for 3.5 h under vacuum at 30 °C providing lipid films at the bottom of the glass test tubes. The lipid films were stored at 4 °C. SUV suspensions were freshly prepared for each experiment by rehydration of the lipid film in citrate buffer (50 mM KCl, 20 mM trisodium citrate/HCl, 0.1 mM EDTA, 0.1 mM NaN3, pH 4.8). The lipid films were incubated for 30 min at room temperature followed by vortexing (3 × 30 s) with 5 min breaks. To form unilamellar vesicles, the suspensions were sonicated for 30 min (Sonopuls HD2070, resonator cup, Bandelin, Berlin, DE). Preparation of Supported Lipid Bilayers (SLBs). All SLBs were prepared on mica surfaces. Quadratic mica slices (6 × 6 mm2; B


Article

Langmuir nm laser was used to excite FITC-C3N13 (λex/em = 490/525 nm) and Bodipy-C12-HPC (λex/em = 500/510 nm) and a 561 nm laser for Texas Red−DHPE (λex/em = 595/615 nm). Diffusion coefficients were calculated from the FRAP data using Hankel transformations.53 Atomic Force Microscopy (AFM). All measurements were performed in buffer solution using intermittent contact mode on a MFP-3D AFM (Asylum Research, Santa Barbara, CA) and MSNL-10 cantilevers (resonant frequency (air) 90−160 kHz, spring constant 0.3−1.4 N m−1, Bruker, Camarillo, US). The AFM was combined with an inverse epifluorescence microscope (IX51, Olympus, Tokyo, JP), a 60× air objective (NA = 0.7, LUCPLFLN 60×, Olympus, Tokyo, JP), and a CCD camera (Infinity 2, Lumenera, Ottawa, CA) to directly overlay AFM and fluorescence micrographs.

spots are distributed all over the substrate, which remain on the surface even after extensive rinsing. Topographic images measured by AFM show a homogeneous featureless membrane surface. An AFM image showing a defect is presented in Figure 1B to demonstrate the contrast between bilayer and support and which allowed us to extract a characteristic bilayer thickness of about 4 nm from the height profile. FRAP experiments (Figure 1D) provide solid evidence for the formation of continuous lipid bilayers on the mica surface with full recovery and a lateral mobility of the lipids with a diffusion coefficient of 0.3 ± 0.1 μm2 s−1. Also, the fluorescence intensity of the bright spots fully recovers. This observation suggests that these spots are protrusion-like membrane structures connected to the underlying lipid bilayer. If SUVs composed of MGDG/DGDG/SQDG/POPG/ DOPC/Texas Red−DHPE (34:20:15:5:25:1) are spread on a hydrophobically silanized glass substrate (Supporting Information), lipid monolayers instead of lipid bilayers are generated.56 The result of spreading SUVs on such a functionalized glass substrate is shown in Figure 1C. Similar to the results obtained for bilayer formation, a homogeneous fluorescence with numerous brighter spots is observed. The formation of a continuous lipid monolayer with laterally mobile lipids was proven by FRAP experiments (Figure 1E) showing complete recovery of the bleached membrane area with a lipid diffusion coefficient of 0.3 ± 0.1 μm2 s−1. Interaction of the Long-Chain Polyamine C3N13 with SLBs. After the establishment of lipid monolayers and bilayers composed of MGDG/DGDG/SQDG/POPG/DOPC (35:20:15:5:25), we investigated the influence of LCPAs on the morphology of the planar membranes. We started out with a LCPA that is chemically readily accessible and is long enough to mimic the LCPAs found in diatoms.22 The LCPA C3N13 (Figure 2B) was synthesized as described previously.50 Fluorescence micrographs demonstrate that the addition of C3N13 leads to the formation of micrometer-sized domains with a brighter fluorescence intensity within a few minutes (Figure 2A). These domains show uniform fluorescence and grow with a kinetics that follows an exponential decay function starting from a central point. To be able to localize C3N13 on the membrane surface, the LCPA was fluorescently labeled with the fluorescent dye FITC (Figure 2B). Figure 2C shows a correlation between the green fluorescence intensity of FITC− C3N13 and the red fluorescence intensity of the Texas Red− DHPE-doped lipid bilayer. A clear co-localization between the membrane domains formed upon LCPA addition and an increased green fluorescence of accumulated C3N13 was observed. Control measurements without fluorescent labels in the SLB ruled out any cross-talk between the dyes. Further insight into the structure of the LCPA-induced domains was gathered by comparison of topographic AFM images and the distribution of the lipid-bound dye derived from fluorescence intensities measured by fluorescence microscopy. Figure 3A shows a topography map of two domains formed after addition of C3N13 to a SLB on mica. Even though the soft membrane structures were difficult to image by AFM, height differences between the surrounding membrane and the domains of about 8 and 15 nm were measured. While height differences caused by phase separation within a lipid bilayer are normally in the range of only 0.5−1 nm,57 we explain our findings with the formation of larger 3D membrane structures. The corresponding fluorescence micrograph (Figure 3B) showed a good correlation between the higher domains

■

RESULTS AND DISCUSSION SLBs and Lipid Monolayers Resembling the Lipid Composition of a Diatom. In the first set of experiments, planar lipid mono- and bilayers on solid supports were established with a lipid composition resembling the one found in the diatom Cyclotella meneghiniana (Table 1).15 The Table 1. Lipid Composition of the Diatom Cyclotella meneghiniana15 and Used in This Study: Monogalactosyldiacylglycerol (MGDG), Digalactosyldiacylglycerol (DGDG), Sulfoquinovosyldiacylglycerol (SQDG), Phosphatidylglycerol (PG), and Phosphatidylcholine (PC)a

a

R refers to different predominantly (poly)unsaturated alkyl chains of the fatty acids.

success of continuous lipid mono- or bilayer formation on hydrophobic and hydrophilic solid supports obtained by spreading of small unilamellar vesicles (SUVs) was monitored by fluorescence microscopy and atomic force microscopy (AFM) as well as by fluorescence recovery after photobleaching (FRAP) experiments. Figure 1A shows a fluorescence micrograph of a supported lipid bilayer (SLB) composed of MGDG/ DGDG/SQDG/POPG/DOPC/Texas Red−DHPE (34.5:20:15:5:25:0.5) obtained after spreading of SUVs on hydrophilic mica. The DOPC content was increased compared to the native lipid mixture15 to stabilize the lamellar phase containing otherwise large amounts of non-bilayer-forming MGDG.54,55 A homogeneous red fluorescence is observed indicating full coverage of the surface with membrane. In addition, brighter C


Article

Langmuir

Figure 1. (A) Fluorescence micrograph of a SLB formed by spreading of small unilamellar vesicles (SUVs) on a hydrophilic mica surface. Scale bar: 50 μm. (B) Atomic force topography image of a SLB with a large defect revealing a bilayer thickness of about 4 nm. Scale bar: 2 μm. (C) Fluorescence micrograph of a supported lipid monolayer on hydrophobically functionalized glass obtained by spreading of SUVs. Scale bar: 50 μm. Results of FRAP experiments on (D) SLBs and (E) lipid monolayers. Top: a series of fluorescence micrographs before (0 s), right after bleaching (6 s, 11 s), and at the end of the experiment (240 s, 214 s) are shown. Scale bars: 10 μm. Bottom: time-resolved relative fluorescence intensities F/F0 within the regions of interests (white circles) showing full recovery of the initial fluorescence intensity F0 of the membrane. These results confirm the formation of a continuous, fluid lipid mono- and bilayer with full lateral lipid mobility.

recorded. After bleaching, full recovery of the stack’s initial fluorescence intensity was observed, demonstrating a connection between the stacked and underlying membrane. FRAP experiments were carried out on both observed stacks: those with doubled and tripled fluorescence intensities compared to the underlying lipid bilayer. Full fluorescence recovery was observed in both cases within a few minutes. Because of the fast exchange of lipid-bound dye molecules, it is likely that several connections exist between the bilayers. To further understand the process of stack formation, we replaced the lipid bilayers supported on mica with lipid monolayers on a hydrophobically functionalized glass substrate (Supporting Information). With this approach, we addressed the question whether a lipid bilayer or only a monolayer is required to form LCPA-induced bilayer stacks. Figure 4 shows a fluorescence micrograph of a lipid monolayer on a functionalized glass substrate upon addition of C3N13. Membrane stacks are formed in the same manner as observed on SLBs. This observation rules out a mechanism of membrane stack formation, where a full lipid bilayer is folded into two layers as it was observed previously at the border of spread giant unilamellar vesicles.59 Instead, only one leaflet of the underlying membrane is involved in LCPA-induced stack formation. As the size of the membrane stacks is in the micrometer regime, the question arises, where the lipid material comes from

observed in the topographic AFM images and domains of brighter fluorescence intensity. Analysis of the fluorescence intensities of the two domains and the surrounding membrane revealed a roughly doubled respectively tripled fluorescence intensity of the domains compared to the planar membrane. From a statistical analysis of all observed membrane stacks (n = 1204), we found that 80% of the stacks are double bilayers (F/ F0 = 2.0 ± 0.2), while 20% are triple bilayers (F/F0 = 3.1 ± 0.4). Thus, the observed height differences of about 8 and 15 nm are supposed to correspond to one and two stacked lipid bilayers58 on top of the planar membrane, respectively. As the lipid bilayer itself is only 4 nm in height, we assume that the LCPAs form an aqueous layer of another 4 nm in between the stacks, theoretically resulting in a thickness of either 8 nm (one stack) or 16 nm (two stacks) (see Figure 8). Taken these observations together, it can be proposed that not membrane domains but membrane stacks (one or two additional lipid bilayers stacked on the membrane) are formed upon addition of the LCPA. FRAP experiments were employed to analyze whether there is a connection between the underlying and the stacked lipid bilayers, which would allow exchange of lipids and lipid-bound dye molecules between the stacks. Figure 3C shows a series of fluorescence micrographs during a FRAP experiment. The fluorescence of a whole membrane stack is bleached by a short laser pulse, and the time-resolved fluorescence recovery is D


Article

Langmuir

Figure 2. (A) Time series of Texas Red−DHPE fluorescence images showing the growth of bright fluorescent domains after addition of the LCPA C3N13. The domains grow following an exponential decay function starting from a central point. Scale bar: 20 μm. (B) Chemical structures of C3N13 and FITC−C3N13. (C) Fluorescence micrograph of FITC−C3N13 (green) and Texas Red−DHPE (red) after the addition of the LCPA showing a co-localization of a brighter FITC−C3N13 fluorescence and the membrane domain doped with Texas Red−DHPE. Scale bar: 10 μm.

Figure 3. (A) Top: AFM topography image of LCPA induced membrane stacks and (B) Texas Red−DHPE fluorescence image of the same membrane stacks. Scale bars: 10 μm. Bottom: corresponding height and intensity profile along the white lines depicted in the top images. (C) FRAP experiment on a membrane stack. Top: a whole membrane stack is bleached as visualized by a series of fluorescence micrographs before (0 s), right after (4 s), and about 10 min after bleaching (644 s). Scale bar: 10 μm. Bottom: relative fluorescence intensity F/F0 within the region of interest (white circle) is plotted as a function of time. The fluorescence intensity of the membrane stack F normalized to the fluorescence intensity of the surrounding bilayer F0 fully recovers proving a connection and exchange of lipid material between the stack and its underlying membrane.

to form these stacks. In principle, two lipid membrane pools can be envisioned: (i) an internal lipid pool where lipids are recruited from the underlying membrane. If this were the case, the formation of defects or a reduction of the protrusions observed on the initial lipid bilayer in the vicinity of the membrane stacks would be expected; (ii) an external lipid pool.60 If this were the case, the addition of lipid vesicles together with the LCPA would result in an enhanced formation of membrane stacks. To prove the two possible scenarios, we inspected the membranes during and after stack formation by means of fluorescence microscopy but did not observe any changes of the membrane surrounding the bilayer stacks. Furthermore, we analyzed the change of the bright spots on the membrane surface by fluorescence time lapse images during stack growth. No significant decrease in intensity or number of the protrusions was observed. To elucidate whether an external lipid pool provides the lipid material required to form stacks, we added Texas Red−DHPE-labeled SUVs to SLBs doped with Bodipy-C12-HPC. Figure 5 shows the fluorescence micrographs after addition of SUVs and C3N13. While the Bodipy-C12-HPC fluorescence image (green, left image) shows membrane stacks, the Texas Red−DHPE fluorescence image (red, right) clearly demonstrates that lipid material stemming from the SUVs were incorporated into the formed membrane stacks and the SLB. The connection between the underlying and the stacked membranes allowed an exchange of lipids and enabled mixing

of both dyes after membrane stack formation. This observation supports the notion of an external lipid pool required for membrane stack formation. The exact mechanism by which the lipid material gets inserted into the SLB leading to stack formation remains elusive. However, it is well conceivable that fusion of SUVs with the SLB inserts lipid material into the SLB. For shorter polyamines such as spermine or spermidine, enhancing effects in vesicle fusion have been reported.44,46−48 The SLBs under investigation contain large amounts of glyceroglycolipids, especially the lamellar phase forming DGDG and the non-bilayer-forming MGDG.61,62 To understand how the lipid composition influences bilayer stack formation, we varied the lipid composition and added C3N13. Figure 6A−D shows fluorescence micrographs after addition of C3N13 to differently composed lipid bilayers on mica. In Figure 6A−C fluorescence micrographs of SLBs lacking in each case one of the three glyceroglycolipids MGDG, DGDG, or SQDG are E


Article

Langmuir

Figure 4. Fluorescence micrograph of a lipid monolayer composed of MGDG/DGDG/SQDG/POPG/DOPC/Texas Red−DHPE (34:20:15:5:25:1) on hydrophobically functionalized glass after addition of the LCPA C3N13. Scale bar: 20 μm. Figure 6. Fluorescence micrographs of SLBs composed of (A) DGDG/SQDG/POPG/DOPC/TexasRed-DHPE (20:15:5:59.5:0.5), (B) MGDG/SQDG/POPG/DOPC/Texas Red−DHPE (34.5:15:5:45:0.5), (C) MGDG/DGDG/POPG/DOPC/TexasRedDHPE (34.5:20:20:25:0.5), and (D) POPG/DOPC/TexasRedDHPE (20:79.5:0.5) after the addition of C3N13. Scale bars: 20 μm.

C3N18 membrane stack formation was observed. These results suggest that reducing the charge repulsion of two opposing membranes by adsorption of amine groups is required for stack formation. It has been shown that the overall charge of the polyamine increases with polyamine chain length thus enhancing membrane interactions.63 However, our results further show that this is not sufficient. A minimum number of amine groups (Y > 5) in a chain is necessary to induce membrane stack formation. We speculate that a certain length is required to be able to bridge the two membranes forming a LCPA cushion of several nanometers between the two membranes (Figure 8). If the polyamines are too small, bridging would not be possible, thus making stack formation energetically unfavorable. In contrast to these observations, polymeric poly(ethylene imine)s with molar masses of about 78 000 g mol−1 lead to defects within lipid bilayers.64 In summary, we propose that membrane stacks are formed upon an intimate interaction between the headgroup region of the lipid membrane, presumably the phosphate groups of the glycerophospholipids, and the LCPAs (Figure 8). Between both membranes at least one connection is formed by fusion of the two neighboring membrane leaflets, allowing exchange of lipids.

Figure 5. Fluorescence micrographs of a lipid bilayer composed of MGDG/DGDG/SQDG/POPG/DOPC/Bodipy-C12-HPC (34.5:20:15:5:25:0.5) (left image, green, λex = 488 nm) after addition of Texas Red−DHPE containing SUVs composed of MGDG/ DGDG/SQDG/POPG/DOPC/Texas Red−DHPE (34.5:20:15:5:25:0.5) (right image, red, λex = 561 nm) and C3N13. Scale bar: 20 μm.

shown. All SLBs form LCPA-induced membrane stacks. SLB composed only of POPG and DOPC (Figure 6D) lacking any glyceroglycolipid or even only composed of DOPC (data not shown) also show membrane stack formation after C3N13 addition. These findings demonstrate that the stack formation process is not limited to membranes containing glyceroglycolipids but might be mediated by the direct interaction of the LCPAs with the phosphate groups of the glycerophospholipids.31 As the lipid composition appears to be not a decisive parameter for membrane stack formation, we further addressed the question, what the molecular requirements are concerning the LCPAs. We investigated the influence of the polyamine chain length employing fluorescence microscopy (Figure 7) using the short naturally occurring polyamines spermidine (Figure 7A) and spermine (Figure 7B) as well as four synthetic polyamines termed C3N4 (Figure 7C), C3N5 (Figure 7D), C3N7 (Figure 7E), and C3N18 (Figure 7F). Upon addition of these polyamines together with 4 μg mL−1 SUVs to the SLBs on mica, only for the two longest polyamines C3N7 and

■

SUMMARY AND CONCLUSIONS The process of silica formation in diatoms takes place within silica deposition vesicles being acidic organelles surrounded by a lipid bilayer. Within these silica deposition vesicles, positively charged long-chain polyamines (LCPAs) are one of the prominent biomacromolecules. Here, we have shown that LCPAs are capable of interacting with lipid membranes harboring negatively charged phosphate groups attached to glycerophospholipids. However, they do not only bind to these membranes but induce three-dimensional and characteristic flat F


Article

Langmuir

Figure 7. Fluorescence micrographs of SLBs composed of MGDG/DGDG/SQDG/POPG/DOPC/TexasRed-DHPE (34.5:20:15:5:25:0.5) after the addition of different polyamines. The influence of polyamine chain length was investigated by adding (A) spermidine, (B) spermine, (C) C3N4, (D) C3N5, (E) C3N7, and (F) C3N18 to the SLBs. Scale bars: 50 μm.

Figure 8. Model of membrane stack formation induced by the LCPA C3N13. The LCPA (blue) connects the bilayer stacks (red) via bridging the lipid head groups. At least one connection between the stacked and underlying membrane is formed by interaction of the two neighboring membrane leaflets. The structure of the connection is however not known and might depend on the lipid composition.

membrane stacks. For stack formation to occur, a minimum number of amine groups within the chain is required that is presumably needed to bridge the two opposing membranes forming a LCPA cushion of several nanometers. It is known that polyamines are capable of stabilizing lipid membranes as a result of electrostatic interactions. Together with the finding that LCPAs induce and stabilize three-dimensional membrane stacks, in which additional lipid material is stored, it can be hypothesized that the lipid−LCPA interaction helps stabilizing and expanding the SDV during valve and girdle band formation in diatoms.

■

Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS Generous support by the DFG (Research Unit FOR 2038, Nanopatterned Organic Matrices in Biological Silica Mineralization, SP-6, SP-7) is gratefully acknowledged. We thank Dr. I. Mey for fruitful discussions and V. Reusche and J. Gerber-Nolte for experimental support.

■

ASSOCIATED CONTENT

(1) Richter, R. P.; Bérat, R.; Brisson, A. R. Formation of solidsupported lipid bilayers: An integrated view. Langmuir 2006, 22, 3497−3505. (2) Webb, M. S.; Green, B. R. Permeability properties of large unilamellar vesicles of thylakoid lipids. Biochim. Biophys. Acta, Biomembr. 1989, 984, 41−49. (3) Webb, M. S.; Green, B. R. Effects of neutral and anionic lipids on digalactosyldiacylglycerol vesicle aggregation. Biochim. Biophys. Acta, Biomembr. 1990, 1030, 231−237. (4) Kraayenhof, R.; Sterk, G. J.; Wong Fong Sang, H. W.; Krab, K.; Epand, R. M. Monovalent cations differentially affect membrane surface properties and membrane curvature, as revealed by fluorescent probes and dynamic light scattering. Biochim. Biophys. Acta, Biomembr. 1996, 1282, 293−302. (5) Yang, C.; Boggasch, S.; Haase, W.; Paulsen, H. Thermal stability of trimeric light-harvesting chlorophyll a/b complex (LHCIIb) in

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.langmuir.6b02575. AFM images and contact angle measurements of glass substrates and HMDS-coated glass substrates, adsorption isotherm (reflectometric interference spectroscopy measurements) of C3N13 onto membrane-covered Si/SiO2 surfaces (PDF)

■

REFERENCES

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected] (C.S.). G


Article

Langmuir

(27) Kröger, N.; Poulsen, N. Diatoms  From cell wall biogenesis to nanotechnology. Annu. Rev. Genet. 2008, 42, 83−107. (28) Sumper, M. A phase separation model for the nanopatterning of diatom biosilica. Science 2002, 295, 2430−2433. (29) Kröger, N.; Lorenz, S.; Brunner, E.; Sumper, M. Self-Assembly of Highly Phosphorylated Silaffins and Their Function in Biosilica Morphogenesis. Science 2002, 298, 584−586. (30) Sumper, M.; Brunner, E. Learning from diatoms: Nature’s tools for the production of nanostructured silica. Adv. Funct. Mater. 2006, 16, 17−26. (31) Bernecker, A.; Wieneke, R.; Riedel, R.; Seibt, M.; Geyer, A.; Steinem, C. Tailored synthetic polyamines for controlled biomimetic silica formation. J. Am. Chem. Soc. 2010, 132, 1023−1031. (32) Sumper, M.; Kröger, N. Silica formation in diatoms: the function of long-chain polyamines and silaffins. J. Mater. Chem. 2004, 14, 2059−2065. (33) Edreva, A. Polyamines in Plants. Bulg. J. Plant Physiol. 1996, 22, 73−101. (34) Smith, T. A historical perspective on research in plant polyamine biology. Biochemistry and Physiology of Polyamines in Plants, pp 1−22. (35) Cohen, S. S. A Guide to the Polyamines; Oxford University Press: New York, 1998. (36) Tabor, C. W.; Tabor, H. Polyamines. Annu. Rev. Biochem. 1984, 53, 749−790. (37) Hamana, K.; Tanaka, T.; Hosoya, R.; Niitsu, M.; Itoh, T. Cellular polyamines of the acidophilic, thermophilic and thermoacidophilic archaebacteria, Acidilobus, Ferroplasma, Pyrobaculum, Pyrococcus, Staphylothermus, Thermococcus, Thermodiscus and Vulcanisaeta. J. Gen. Appl. Microbiol. 2003, 49, 287−293. (38) Wallace, H. M. The physiological role of the polyamines. Eur. J. Clin. Invest. 2000, 30, 1−3. (39) Flores, H. E. Changes in polyamine metabolism in response to abiotic stress. Biochemistry and Physiology of Polyamines in Plants, pp 213−227. (40) Tiburcio, A. F.; Besford, R. T.; Capell, T.; Borrell, A.; Testillano, P. S.; Risueño, M. C. Mechanisms of polyamine action during senescence responses induced by osmotic stress. J. Exp. Bot. 1994, 45, 1789−1800. (41) Chroboczek, J. Interaction of spermidine with viral RNA and its influence on protein synthesis. Plant Mol. Biol. 1985, 4, 23−30. (42) Wilson, R. W.; Bloomfield, V. A. Counterion-induced condensation of deoxyribonucleic acid. A light-scattering study. Biochemistry 1979, 18, 2192−2196. (43) Raspaud, E.; Olvera de la Cruz, M.; Sikorav, J.-L.; Livolant, F. Precipitation of DNA by polyamines: A polyelectrolyte behavior. Biophys. J. 1998, 74, 381−393. (44) Schuber, F. Influence of polyamines on membrane functions. Biochem. J. 1989, 260, 1−10. (45) Chung, L.; Kaloyanides, G.; McDaniel, R.; McLaughlin, A.; McLaughlin, S. Interaction of gentamicin and spermine with bilayer membranes containing negatively charged phospholipids. Biochemistry 1985, 24, 442−452. (46) Meers, P.; Hong, K.; Bentz, J.; Papahadjopoulos, D. Spermine as a modulator of membrane fusion: Interactions with acidic phospholipids. Biochemistry 1986, 25, 3109−3118. (47) Hong, K.; Schuber, F.; Papahadjopoulos, D. Polyamines Biological modulators of membrane fusion. Biochim. Biophys. Acta, Biomembr. 1983, 732, 469−472. (48) Schuber, F.; Hong, K.; Duzgunes, N.; Papahadjopoulos, D. Polyamines as modulators of membrane fusion: aggregation and fusion of liposomes. Biochemistry 1983, 22, 6134−6140. (49) Ohki, S.; Duax, J. Effects of cations and polyamines on the aggregation and fusion of phosphatidylserine membranes. Biochim. Biophys. Acta, Biomembr. 1986, 861, 177−186. (50) Abacilar, M.; Daus, F.; Geyer, A. Chemoselective silicification of synthetic peptides and polyamines. Beilstein J. Nanotechnol. 2015, 6, 103−110.

liposomes of thylakoid lipids. Biochim. Biophys. Acta, Bioenerg. 2006, 1757, 1642−1648. (6) Hincha, D. K. Effects of α-tocopherol (vitamin E) on the stability and lipid dynamics of model membranes mimicking the lipid composition of plant chloroplast membranes. FEBS Lett. 2008, 582, 3687−3692. (7) Zhou, F.; Liu, S.; Hu, Z.; Kuang, T.; Paulsen, H.; Yang, C. Effect of monogalactosyldiacylglycerol on the interaction between photosystem II core complex and its antenna complexes in liposomes of thylakoid lipids. Photosynth. Res. 2009, 99, 185−193. (8) Fuks, B.; Homblé, F. Permeability and electrical properties of planar lipid membranes from thylakoid lipids. Biophys. J. 1994, 66, 1404−1414. (9) Kerth, A.; Brehmer, T.; Meister, A.; Hanner, P.; Jakob, M.; Klösgen, R. B.; Blume, A. Interaction of a Tat substrate and a Tat signal peptide with thylakoid lipids at the air-water interface. ChemBioChem 2012, 13, 231−239. (10) Hoyo, J.; Guaus, E.; Torrent-Burgués, J. Monogalactosyldiacylglycerol and digalactosyldiacylglycerol role, physical states, applications and biomimetic monolayer films. Eur. Phys. J. E: Soft Matter Biol. Phys. 2016, 39, 39. (11) Schneider, J.; Dufrêne, Y. F.; Barger, W. R.; Lee, G. U. Atomic force microscope image contrast mechanisms on supported lipid bilayers. Biophys. J. 2000, 79, 1107−1118. (12) Janshoff, A.; Steinem, C. Transport across artificial membranes− an analytical perspective. Anal. Bioanal. Chem. 2006, 385, 433−451. (13) Johnston, L. J. Nanoscale imaging of domains in supported lipid membranes. Langmuir 2007, 23, 5886−5895. (14) Richter, R. P.; Him, J. L. K.; Brisson, A. Supported lipid membranes. Mater. Today 2003, 6, 32−37. (15) Vieler, A.; Wilhelm, C.; Goss, R.; Süß, R.; Schiller, J. The lipid composition of the unicellular green alga Chlamydomonas reinhardtii and the diatom Cyclotella meneghiniana investigated by MALDI-TOF MS and TLC. Chem. Phys. Lipids 2007, 150, 143−155. (16) Round, F. E.; Crawford, R. M.; Mann, D. G. The Diatoms: Biology & Morphology of the Genera; Cambridge University Press: Cambridge, NY, 1990. (17) Drum, R. W.; Pankratz, H. S. Post mitotic fine structure of Gomphonema parvulum. J. Ultrastruct. Res. 1964, 10, 217−223. (18) Coombs, J.; Halicki, P. J.; Holm-Hansen, O.; Volcani, B. Studies on the biochemistry and fine structure of silica shell formation in diatoms. Exp. Cell Res. 1967, 47, 315−328. (19) Schmid, A.-M. M.; Schulz, D. Wall morphogenesis in diatoms: Deposition of silica by cytoplasmic vesicles. Protoplasma 1979, 100, 267−288. (20) Li, C.-W.; Volcani, B. E. Aspects of silicification in wall morphogenesis of diatoms. Philos. Trans. R. Soc., B 1984, 304, 519− 528. (21) Vrieling, E. G.; Gieskes, W. W. C.; Beelen, T. P. M. Silicon deposition in diatoms: Control by the pH inside the silicon deposition vesicle. J. Phycol. 1999, 35, 548−559. (22) Kröger, N.; Deutzmann, R.; Bergsdorf, C.; Sumper, M. Speciesspecific polyamines from diatoms control silica morphology Speciesspecific p silica morphology. Proc. Natl. Acad. Sci. U. S. A. 2000, 97, 14133−14138. (23) Kröger, N.; Deutzmann, R.; Sumper, M. Polycationic peptides from diatom biosilica that direct silica nanosphere formation. Science 1999, 286, 1129−1132. (24) Wenzl, S.; Hett, R.; Richthammer, P.; Sumper, M. Silacidins: Highly acidic phosphopeptides from diatom shells assist in silica precipitation in vitro. Angew. Chem. 2008, 120, 1753−1756. (25) Scheffel, A.; Poulsen, N.; Shian, S.; Kröger, N. Nanopatterned protein microrings from a diatom that direct silica morphogenesis. Proc. Natl. Acad. Sci. U. S. A. 2011, 108, 3175−3180. (26) Kotzsch, A.; Pawolski, D.; Milentyev, A.; Shevchenko, A.; Scheffel, A.; Poulsen, N.; Shevchenko, A.; Kröger, N. Biochemical composition and assembly of biosilica-associated insoluble organic matrices from the diatom Thalassiosira pseudonana. J. Biol. Chem. 2016, 291, 4982−4997. H


Article

Langmuir (51) Gasparini, G.; Sargsyan, G.; Bang, E.-K.; Sakai, N.; Matile, S. Ring tension applied to thiol-mediated cellular uptake. Angew. Chem., Int. Ed. 2015, 54, 7328−7331. (52) Jullian, M.; Hernandez, A.; Maurras, A.; Puget, K.; Amblard, M.; Martinez, J.; Subra, G. N-terminus FITC labeling of peptides on solid support: The truth behind the spacer. Tetrahedron Lett. 2009, 50, 260−263. (53) Jönsson, P.; Jonsson, M. P.; Tegenfeldt, J. O.; Höök, F. A method improving the accuracy of fluorescence recovery after photobleaching analysis. Biophys. J. 2008, 95, 5334−5348. (54) Chupin, V.; van’t Hof, R.; de Kruijff, B. The transit sequence of a chloroplast precursor protein reorients the lipids in monogalactosyl diglyceride containing bilayers. FEBS Lett. 1994, 350, 104−108. (55) Castro, V.; Dvinskikh, S. V.; Widmalm, G.; Sandström, D.; Maliniak, A. NMR studies of membranes composed of glycolipids and phospholipids. Biochim. Biophys. Acta, Biomembr. 2007, 1768, 2432− 2437. (56) Keller, C. A.; Kasemo, B. Surface specific kinetics of lipid vesicle adsorption measured with a quartz crystal microbalance. Biophys. J. 1998, 75, 1397−1402. (57) Rinia, H. A.; Snel, M. M.; van der Eerden; Jan, P. J. M.; de Kruijff, B. Visualizing detergent resistant domains in model membranes with atomic force microscopy. FEBS Lett. 2001, 501, 92−96. (58) Kučerka, N.; Nieh, M.-P.; Katsaras, J. Fluid phase lipid areas and bilayer thicknesses of commonly used phosphatidylcholines as a function of temperature. Biochim. Biophys. Acta, Biomembr. 2011, 1808, 2761−2771. (59) Kocun, M.; Lazzara, T. D.; Steinem, C.; Janshoff, A. Preparation of solvent-free, pore-spanning lipid bilayers: Modeling the low tension of plasma membranes. Langmuir 2011, 27, 7672−7680. (60) Adams, P. G.; Lamoureux, L.; Swingle, K. L.; Mukundan, H.; Montaño, G. A. Lipopolysaccharide-induced dynamic lipid membrane reorganization: Tubules, perforations, and stacks. Biophys. J. 2014, 106, 2395−2407. (61) Sprague, S. G.; Staehelin, L. Effects of reconstitution method on the structural organization of isolated chloroplast membrane lipids. Biochim. Biophys. Acta, Biomembr. 1984, 777, 306−322. (62) Webb, M. S.; Green, B. R. Biochemical and biophysical properties of thylakoid acyl lipids. Biochim. Biophys. Acta, Bioenerg. 1991, 1060, 133−158. (63) Finger, S.; Schwieger, C.; Arouri, A.; Kerth, A.; Blume, A. Interaction of linear polyamines with negatively charged phospholipids: The effect of polyamine charge distance. Biol. Chem. 2014, 395, 769−778. (64) Hong, S.; Leroueil, P. R.; Janus, E. K.; Peters, J. L.; Kober, M.M.; Islam, M. T.; Orr, B. G.; Baker, J. R.; Banaszak Holl, M. M.; Mark, M. Interaction of polycationic polymers with supported lipid bilayers and cells: Nanoscale hole formation and enhanced membrane permeability. Bioconjugate Chem. 2006, 17, 728−734.

I


3D-Membrane Stacks on Supported Membranes Composed of Diatom

Recommend Documents