Induced Rupture of Vesicles Adsorbed on Glass by Pore Formation at

Subscriber access provided by TULANE UNIVERSITY

Article

Induced Rupture of Vesicles Adsorbed on Glass by Pore Formation at the Surface–Bilayer Interface Chiho Kataoka-Hamai, and Tomohiko Yamazaki Langmuir, Just Accepted Manuscript • DOI: 10.1021/la5042822 • Publication Date (Web): 09 Jan 2015 Downloaded from http://pubs.acs.org on January 20, 2015

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Langmuir is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

Induced Rupture of Vesicles Adsorbed on Glass by Pore Formation at the Surface–Bilayer Interface

Chiho Kataoka-Hamai†,* and Tomohiko Yamazaki‡

†

International Center for Materials Nanoarchitectonics, National Institute for Materials Science,

1-1 Namiki, Tsukuba, Ibaraki 305-0044, Japan ‡

International Center for Materials Nanoarchitectonics, National Institute for Materials Science,

1-2-1 Sengen, Tsukuba, Ibaraki 305-0047, Japan

Corresponding Author *Tel: +81-29-860-4548. E-mail: [email protected]

ACS Paragon Plus Environment

1

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ABSTRACT

Page 2 of 29

Supported lipid bilayers (SLBs) are often formed by spontaneous vesicle rupture

and fusion on a solid surface. A well-characterized rupture mechanism for isolated vesicles is pore nucleation and expansion in the solution-exposed nonadsorbed area. In contrast, pore formation in the adsorbed bilayer region has not been investigated to date. In this work, we studied the detailed mechanisms of asymmetric rupture of giant unilamellar vesicles (GUVs) adsorbed on glass using fluorescence microscopy. Asymmetric rupture is the pathway where a rupture pore forms in a GUV near the edge of the glass–bilayer interface with high curvature and then expansion of the pore yields a planar bilayer patch. We show that asymmetric rupture occasionally resulted in SLB patches bearing a defect pore. The defect formation probability depended on lipid composition, salt concentration, and pH. Approximately 40% of negatively charged GUVs under physiological conditions formed pore-containing SLB patches, while negatively charged GUVs at low salt concentration or pH 4.0 and positively charged GUVs exhibited a low probability of defect inclusion. The edge of the defect pore was either in contact with (on-edge) or away from (off-edge) the edge of the planar bilayer. On-edge pores were predominantly formed over off-edge defects. Pores initially formed in the glass-adsorbed region before rupture, most frequently in close contact with the edge of the adsorbed region. When a pore formed near the edge of the adsorbed area or when the edge of a pore reached that of the adsorbed area by pore expansion, asymmetric rupture was induced from the defect site. These induced rupture mechanisms yielded SLB patches with an on-edge pore. In contrast, off-edge pores were produced when defect pore generation and subsequent vesicle rupture were uncoupled. The current results demonstrate that pore formation in the surface-adsorbed region of GUVs is not a negligible event.


2

Page 3 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

INTRODUCTION Giant unilamellar vesicles (GUVs) adsorb and deform on a solid surface and then rupture to form supported lipid bilayer (SLB) patches.1 The bilayer patches formed from GUVs have recently been used to seal nanopores and micropores in solid supports.2, 3, 4 The pore-spanning SLBs can serve as electrical barriers with gigaohm resistance3 and permeability barriers for small molecules and macromolecules.5 The transport properties of a membrane protein and viral membrane fusion have also been demonstrated using SLB patches covering microcavities.4 To foster the development of such biotechnological applications, it is of great importance to better understand GUV rupture processes. In particular, the inclusion of bilayer pores is undesirable for pore-spanning SLB systems because these defect pores allow ions and molecules to pass across the bilayers. Nonetheless, it remains unclear if GUVs maintain their continuous lipid packing without forming defect pores during transformation into a planar bilayer patch. If defects form, the underlying principles of this process and the conditions to minimize the probability of defect formation should also be examined. To address these questions, here we investigated the mechanisms of GUV rupture in detail with a particular focus on bilayer defects. We used a flat glass surface as a solid support because we anticipated that such simple surface morphology would help us to elucidate the basic membrane dynamics influenced by bilayer–surface interactions. In addition to the biotechnological point of view, vesicle adsorption and rupture are important events from a fundamental perspective. This is partly because there are analogies between adsorption-induced bilayer deformation and membrane-involved steps in many biological processes, e.g., vesicle traffic. Vesicle adsorption has long been of interest.6, 7, 8, 9 Furthermore, fusion of small unilamellar vesicles (SUVs) or large unilamellar vesicles (LUVs) with a diameter


3

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 29

of ∼50–200 nm has been widely used to form SLBs.10, 11 The application range of this vesicle fusion technique may be extended by better understanding of adsorption behavior. Therefore, GUV rupture has previously been studied to clarify the formation mechanisms of SLBs by vesicle fusion.1 The present study shows that for negatively charged GUVs under physiological conditions, approximately 40% of their rupture events are induced by defect pores that form in the bilayeradsorbed region prior to rupture. Interestingly, previous studies have shown that bilayer pores are formed in the free bilayer area and the adsorbed region remains unaffected. A previous study on transient pores reported that vesicles are stretched by adsorption and then form a pore in the solution-exposed area to relax tension.12 In addition, a report on cell lysis interpreted adsorptioninduced lysis of red blood cells as the leakage of cell content through a pore formed in the nonadsorbed membrane area.13 A theoretical study modeled a rupture pathway where pore formation was initiated in the solution-exposed region.14 In previous studies, the mechanism of GUV rupture has been examined using negatively charged vesicles with a low salt concentration.1 (Note that in the current study, such conditions were found to produce very few defect-containing SLB patches.) After adsorption on glass, isolated vesicles deform through a favorable interaction with the surface and then undergo asymmetric or symmetric rupture to form SLB patches. In the asymmetric pathway, a rupture pore begins to open in deformed vesicles near the edge of the glass–bilayer interface. The pore then grows in the nonadsorbed area, creating an almost heart-shaped bilayer patch (asymmetric patch) with inside-up orientation. In contrast, the symmetric pathway involves the nucleation of a rupture pore near the apex of an adsorbed vesicle, yielding an almost circular bilayer patch (symmetric patch). The asymmetric rupture pathway is dominant over the symmetric one. For


4

Page 5 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

example, the percentage of asymmetric rupture was 100% and 92% for GUVs containing 0.5 and 3 mole % of a fluorescent lipid, respectively.1 Facile pore opening near the edge of the bilayeradsorbed area is expected to stem from the large stress on the lipid packing in this highly curved region.15 In this work, we investigate defect pore formation that occurs during the dominant asymmetric rupture mechanism. Spontaneous rupture of GUVs on glass is observed by fluorescence microscopy. We first examine the mechanisms of defect pore formation and the dependence of defect formation probability on salt concentration, pH, and lipid composition. We then discuss the reasons why pore formation occurs in the adsorbed area.

MATERIALS AND METHODS Materials. 1,2-Dioleoyl-sn-glycero-3-phosphocholine (DOPC), 1,2-dioleoyl-sn-glycero-3phospho-L-serine (sodium salt) (DOPS), and 1,2-dioleoyl-3-trimethylammonium propane chloride salt (DOTAP) were purchased from Avanti Polar Lipids (Alabaster, AL, USA). Texas Red 1,2-dihexadecanoyl-sn-glycero-3-phosphoethanolamine triethylammonium salt (TR-DHPE) was purchased from Life Technologies (Carlsbad, CA, USA). 7X detergent (No. 76-670-93) was purchased from MP Biomedicals (Santa Ana, CA, USA). Glass coverslips (22×22 mm, No. 1) were purchased from Matsunami Glass Ind. (Osaka, Japan). All chemicals were used as received. Histidine-tagged green fluorescent protein (GFP) was purified from BL21 (DE3) Escherichia coli cells using a Ni-NTA column (Qiagen, Hilden, Germany). The protein concentration was determined by the bicinchoninic acid (BCA) assay. Preparation of GUVs. GUVs were prepared as follows.16 DOPC, a charged lipid (DOPS or DOTAP), and fluorescent probe (TR-DHPE) were mixed in chloroform to give the desired


5

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 29

composition. Negatively charged DOPS or positively charged DOTAP was added to all lipid preparations to increase the yield of unilamellar vesicles.16 Bulk chloroform was removed under a stream of nitrogen, and any remaining solvent was evaporated under vacuum. Sucrose solution was added to the dried lipid film to yield a total lipid concentration of 2 mM. GUVs were then diluted 1000-fold in buffer before adsorption onto glass. The sucrose concentration was varied to prevent osmotic shock of the GUVs upon dilution. For instance, when the dilution buffer contained 100 mM NaCl and 10 mM NaH2PO4-Na2HPO4 (sodium phosphate) at pH 7.2 (0.225 osmolar), GUVs were prepared in 225 mM sucrose. The dilution buffers contained either sodium phosphate at pH 7.2, HEPES at pH 7.2, or CH3COOH-CH3COONa (sodium acetate) at pH 4.0 at a concentration of 10 mM to maintain constant pH. When the effect of calcium ions on defect formation at pH 7.2 was examined, HEPES was used instead of sodium phosphate to avoid precipitation of calcium phosphate. Lipid compositions of GUVs are expressed as molar ratios. Glass Cleaning. Glass coverslips were cleaned before use as described previously.17 Briefly, coverslips were soaked for 20 min in a solution of 7X detergent that was heated until clear, and then rinsed with water. After drying under a stream of nitrogen, the coverslips were baked at 440 °C for 5 h and then cooled to room temperature. The root-mean-square surface roughness of the glass after cleaning was determined to be 0.34 ± 0.018 nm (N = 3) by atomic force microscopy (AFM) (Supporting Information). Fluorescence Microscopy. A fluorescence microscope (Eclipse Ti-E, Nikon, Tokyo, Japan) equipped with an electron multiplying charge-coupled device camera (iXonEM + 897, Andor Technology, Belfast, Northern Ireland, UK) was used to image GUVs and bilayer patches adsorbed on glass. Samples were observed under mercury-arc lamp illumination at 150× magnification by combining a 100× oil-immersion objective with a 1.5× Optovar. The filter sets


6

Page 7 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

used to excite TR-DHPE and GFP were FF01-561/12:Di01-R561:FF01-609/54 and FF01482/18:Di01-R488:FF01-525/45 (Semrock, Rochester, NY, USA), respectively. The pixel size of the camera was 16×16 µm. Zeta Potential Measurements. LUVs were prepared as follows. Lipids were mixed in chloroform to give the desired compositions. Bulk chloroform was removed under a stream of nitrogen, and any remaining solvent was evaporated under vacuum. Water was added to the dried lipids to yield a total lipid concentration of 4 or 10 mM. The samples were then subjected to ten freeze-thaw cycles (liquid nitrogen/room temperature), followed by extrusion through a polycarbonate membrane filter with 100-nm pores using a mini-extruder (Avanti Polar Lipids, Alabaster, AL, USA) 11 times. Lipid compositions in LUVs are expressed as molar ratios. The lipid sample was diluted in an appropriate solution to yield a lipid concentration of 0.4 mM. The zeta potential for LUVs was then determined using a light scattering analyzer (ELSZ1000Z, Otsuka Electronics, Osaka, Japan). Experimental Values. Values are quoted as the mean ± standard error (SE).

RESULTS Defect Pores in Planar Bilayer Patches. To examine whether defects are formed in planar bilayer patches after asymmetric rupture, DOPC/DOPS/TR-DHPE (88:10:2) GUVs (2 µM lipid) were deposited on glass in a buffer solution of 100 mM NaCl and 10 mM sodium phosphate at pH 7.2. The sample surface was observed 30 min after deposition using fluorescence microscopy. SLB patches produced by asymmetric rupture of isolated unilamellar vesicles are characterized by almost heart-shaped bilayer patches, and therefore can be distinguished from those resulting from other rupture


7

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 29

mechanisms including symmetric rupture of isolated GUVs and induced rupture by vesicle– vesicle and vesicle–bilayer edge interactions.1 Thus, almost heart-shaped bilayer patches were focused on to examine defect generation during asymmetric rupture events. Many bilayer patches produced by asymmetric rupture contained a defect pore that showed very low fluorescence intensity (Figure 1A, B). SLB patches had a single defect pore, if any, and none with multiple defects were observed. Defect pores were located close to (Figure 1A) or away from (Figure 1B) the planar bilayer edge, which are hereafter called on-edge and off-edge pores, respectively. A point on the on-edge pore rim (Figure 1A, red arrow) was located at the bilayer boundary. The very weak fluorescence signals in the defect areas indicate that lipids were almost removed from these regions. This situation was further validated by protein adsorption experiments (Figure 1C–F). DOPC/DOPS/TR-DHPE (88:10:2) GUVs (2 µM lipid) were adsorbed on glass for 20 min. The sample surface was washed with buffer (100 mM NaCl and 10 mM sodium phosphate at pH 7.2), followed by incubation with 0.2 mg/mL GFP for 30 min. After rinsing with buffer to remove unbound proteins, the fluorescence distributions of TRDHPE and GFP were examined. SLB areas labeled with Texas Red (Figure 1C, E) showed much lower fluorescence intensities from GFP than that of the surrounding glass surface (Figure 1D, F), suggesting that protein adsorption occurred on the glass surface and was prevented on the planar bilayers. Because the defect pores showed GFP fluorescence intensities similar to those on the glass surface, it seemed likely that planar bilayer structures were not present in the defect pores.


8

Page 9 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

Figure 1. Representative defect-containing SLB patches formed by asymmetric rupture of DOPC/DOPS/TR-DHPE (88:10:2) GUVs in buffer solution (100 mM NaCl and 10 mM sodium phosphate at pH 7.2) at 150× magnification. (A, B) GUVs were adsorbed to glass and visualized using the fluorescence emission of Texas Red. The scale bar in A is 5 µm; the scale is the same for B. (A) On-edge defect pore. The pore and planar bilayer patch have a common boundary point (red arrow). (B) Off-edge defect pore. This pore is located away from the edge of the planar bilayer. (C–F) Texas Red (C, E) and GFP (D, F) fluorescence patterns were observed after SLB patches were incubated with GFP (0.2 mg/mL). The scale bar in C is 5 µm; the scale is the same for D–F.


9

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 29

Mechanisms of Defect Pore Formation. To elucidate the mechanisms of defect pore formation, we observed vesicle rupture with a time resolution of 2 s after deposition of DOPC/DOPS/TR-DHPE (88:10:2) GUVs on glass in buffer solution consisting of 400 mM NaCl and 10 mM sodium phosphate at pH 7.2 (Figure 2). Of 60 GUVs, 33 (55%) ruptured to form a planar bilayer patch with no defect (Figure 2A). The data recording speed was much slower than the time scale of rupture pore expansion (∼10–20 ms).1 Therefore, the formation of defect-free bilayer patches (Figure 2A) was identified by a sequence of two images consisting of an adsorbed vesicle (0 s) and resultant planar bilayer patch (2 s). Previous studies have shown that such successive fluorescence data can be attributed to rupture processes.1,

18

Because the focal plane was on glass, adsorbed vesicles created a circular

fluorescence pattern that arose from the glass-adsorbed area.1,

18

All rupture events yielding

defect-free SLBs were expected to occur through the asymmetric mechanism because of the formation of almost heart-shaped bilayer patches and the relative locations of the adsorbed vesicles and final bilayer patches, as reported previously.1 We observed no symmetric pathway, consistent with the previous observation that asymmetric rupture was predominant.1 Of 60 GUVs, 27 (45%) produced SLB patches with a defect pore. Defect formation pathways were grouped into four classes (Figure 2B–E). The first (Figure 2B) was observed for five rupture events, where a defect pore was found for planar bilayer patches (2 s), but not for the adsorbed vesicles before rupture (0 s). The defects were on-edge pores for all five rupture events. These data indicate that on-edge defect formation and vesicle rupture occurred faster than the time resolution of 2 s, and thus do not reveal detailed mechanisms of defect formation. However, the remaining 22 rupture events obviously showed the initial stage of defect formation and the


10

Page 11 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

subsequent stages of defect growth and vesicle rupture (Figure 2C–E). Therefore, it seems likely that in most cases, defect pores nucleated and expanded on the second time scale. The second fluorescence pattern, which is illustrated in Figure 2C, was most frequently observed for defect-bearing SLB formation (14 events). A defect pore was generated near the edge of the adsorbed area (2 s). The vesicle then ruptured from the defect site through the asymmetric rupture mechanism to yield a planar bilayer patch with an on-edge pore (12 s). This rupture mode is schematically illustrated in Figure 3A. The third fluorescence pattern (Figure 2D) was observed for seven events, where a defect pore was formed away from the edge of the glass-adsorbed region (2 s). The pore grew until it came in contact with the edge of the adsorbed area (510 s). The vesicle then ruptured at the common edge point of the defect pore and adsorbed region through the asymmetric pathway, resulting in planar bilayer patches with an on-edge defect (512 s). This rupture process is schematically depicted in Figure 3B. The forth fluorescence pattern (Figure 2E) was observed for only a single rupture event, where the vesicle (0 s) ruptured at a different site from the defect formed near the center of the adsorbed area, creating a planar bilayer patch with an off-edge pore (2 s). The dark region observed in the adsorbed vesicle (0 s) existed for >228 s, which was much longer than the lifetime of a rupture pore. Thus, this region with low fluorescence was attributed to a bilayer pore formed in the adsorbed area rather than a rupture pore opened at the apex of the nonadsorbed area. These data indicate that in this case, the vesicle ruptured with no direct involvement of the defect pore. This uncoupled mechanism is schematically depicted in Figure 3C. As the area of each bilayer defect increased (e.g., Figure 2D, 2–510 s), no fluorescent debris was observed in the vesicle lumen. This suggests that the lipids that had been previously attached to the defect regions did not move into the solution phase, but were rather integrated into


11

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 29

peripheral bilayer areas. The pore expansion process appeared to alter the global vesicle structure. In representative data (Figure 2D), the glass-adsorbed area at 0 and 510 s (i.e., the area with intense fluorescence emission) was ∼7.5 and ∼6.0 µm2, respectively. This implies that the lipids corresponding to a bilayer area of ∼1.5 µm2 were repartitioned into the nonadsorbed region during defect pore expansion.


12

Page 13 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

Figure 2. Spontaneous rupture of DOPC/DOPS/TR-DHPE (88:10:2) GUVs on glass in 400 mM NaCl and 10 mM sodium phosphate at pH 7.2 at 150× magnification. Scale bar in A, 5 µm; the scale is the same for all images. (A) An adsorbed vesicle (0 s) ruptured to yield a continuous SLB patch with no defect (2 s). (B) An adsorbed vesicle without a defect pore (0 s). An on-edge defect then appeared in the resulting SLB patch after rupture (2 s). (C) An adsorbed vesicle without a defect (0 s). A defect pore then began to form near the edge of the adsorbed area (2 s). After pore enlargement (4–10 s), the vesicle ruptured from the defect site to yield a SLB patch with an on-edge pore (12 s). (D) An initial vesicle without a defect pore (0 s). A pore then formed away from the edge of the adsorbed area (2 s). The edge of the defect pore reached the edge of the adsorbed area by pore expansion (510 s). Vesicle rupture then occurred at this contact point to yield a planar bilayer patch with an on-edge pore (512 s). (E) A defect pore was observed in the center of the adsorbed area (0 s). The vesicle ruptured at a different point than the defect to create a SLB patch with an off-edge pore (2 s).


13

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 29

Figure 3. Schematic illustrations of defect formation mechanisms. Two viewpoints are shown: from above the vesicle (upper) and a cross section along the red line (lower). (A) A defect pore begins to form close to the edge of the adsorbed region, and then induces rupture pore formation at the defect site. The defect-induced rupture results in SLB patches containing an on-edge defect pore. (B) A defect pore forms away from the edge of the adsorbed area. The edge of defect pore contacts the boundary of the adsorbed region by pore expansion, which then promotes asymmetric rupture at the defect position. The final SLB patches have an on-edge defect pore. (C) Defect pore formation and asymmetric rupture occur in an uncoupled manner, yielding SLB patches with an off-edge pore.


14

Page 15 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

Dependence of Defect Formation Probability on Salt Concentration, pH, and Lipid Composition. To examine the properties that affect the frequency of defect formation, GUVs (2 µM lipid) containing DOPS or DOTAP were adsorbed for 20–90 min on glass in NaCl-containing buffer at pH 7.2 or 4.0. The sample surface was observed with fluorescence microscopy to determine the proportion of defect-bearing SLB patches (Figure 4). Thirty or fifty bilayer patches formed by asymmetric rupture were analyzed for each sample surface. The defect ratios were calculated so that the sum of the percentage of on-edge pore-containing SLBs (circles with solid lines), offedge pore-containing SLBs (squares with dotted lines), and defect-free SLBs (not shown) was equal to 100. An average of measurements taken for three samples has been reported. The probability of off-edge pore formation (squares) was ≤6%, and did not show any meaningful dependence on the experimental conditions tested. In contrast, the probability of onedge pore formation (circles) varied markedly with salt concentration, pH, and lipid composition. When DOPC/DOPS/TR-DHPE (88:10:2) GUVs were adsorbed on glass in phosphate buffer at pH 7.2 (black circles), the percentage of on-edge pores increased sharply from 2% to 40% as the NaCl concentration increased from 30 to 100 mM. Almost stable values were then obtained in the salt concentration range of 100–400 mM. A similar trend was observed for the same lipid composition in HEPES buffer at pH 7.2 (red circles). Addition of 2 mM CaCl2 to HEPES buffer slightly increased the ratio of on-edge defects (blue circles). Changing the pH to 4.0 using acetate buffer decreased the ratio of on-edge defects (orange circles). Nonetheless, the defect ratio increased with salt level in a similar manner to that at pH 7.2. The increase of defect ratio was smaller as more salt was added. These data show that increasing the concentration of


15

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 29

monovalent and divalent salt ions promoted the formation of on-edge defect pores. In addition, replacement

of

DOPS

with

DOTAP

caused

large

changes

in

defect

formation.

DOPC/DOTAP/TR-DHPE (88:10:2) GUVs in phosphate buffer at pH 7.2 (green circles) yielded a much lower percentage of on-edge pores than DOPS-containing GUVs under the same solution conditions (black circles). The percentage of on-edge pores for DOTAP-containing vesicles increased with NaCl concentration, as observed for DOPS-containing GUVs. We found that the probability of defect inclusion was independent of the area of the planar bilayer patches (1.9–186 µm2) (data not shown). Thus, it seems that defect formation depended directly on salt concentration, pH, and lipid composition, but not vesicle size. Salt content, pH, and lipid ratio all affect the electrostatic properties of the bilayer and glass surface. This suggests that electrostatic interactions determined the probability of defect inclusion in the SLB patches. To further understand the electrostatic interactions in these systems, the zeta potential of LUVs was determined (Figure 5). DOPC/DOPS/TR-DHPE (88:10:2) vesicles in sodium phosphate buffer at pH 7.2 (black) showed negative zeta potential because of the fully deprotonated carboxyl groups in DOPS (pKa ∼3.6)19 and negatively charged TR-DHPE. The negative potential observed in HEPES buffer at pH 7.2 (red) moved to less negative potential by addition of CaCl2 (blue) because calcium ions bind strongly to the DOPS molecules.20 The same lipid mixtures at pH 4.0 (orange) were less negatively charged than at pH 7.2 because of protonation of the DOPS carboxyl group.19 In contrast DOPC/DOTAP/TR-DHPE (88:10:2) vesicles were positively charged at low salt concentration because DOTAP is cationic. The zeta potential varied greatly with the salt level in the range of ∼30–100 mM because of effective screening by counterions. Meanwhile, changes in salt content above ∼100 mM had little effect on zeta potential. This trend was similar to those observed for the proportion of on-edge


16

Page 17 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

pores for DOPC/DOPS/TR-DGPE GUVs (Figure 4, black, red, blue, and orange circles): the onedge defect ratio exhibited large changes in the NaCl concentration range of ∼30–100 mM, and moderate or almost no changes at higher salt level (>100 mM). Thus, at least for the DOPScontaining lipid mixtures, the electrostatic properties of bilayers appear to influence defect formation. The glass surface is negatively charged at pH 7.2, and its charge density is decreased at pH 4.0 because of protonation of dissociated silanol groups (–SiO−). Thus, GUVs containing DOPS experience electrostatic repulsion from the glass surface at pH 7.2 and 4.0, whereas those containing DOTAP are electrostatically attracted to the glass surface. Electrostatic potential near the glass surface is considered to change markedly with salt concentration for low salt levels and show no obvious dependence for high salt levels, in a similar manner to the zeta potential of LUVs. The extent of charge screening for a solid surface can be quantified by the Debye length, which is the distance where the electrostatic potential decreases to 1/e of that at the surface. Debye lengths for monovalent salt concentrations of 30, 60, and 100 mM were calculated to be 1.8, 1.3, and 0.97 nm, respectively. Comparing these values with the typical SLB–surface separation (1–2 nm),21, 22, 23 increasing the salt level in the low-salt regime by a few tens of millimolar is expected to substantially influence the electrostatic bilayer–surface interaction. Meanwhile, further increasing the salt content in the high-salt regime is not expected to greatly alter the electrostatic interaction between the bilayers and glass. This reasoning is also consistent with the salt-dependent defect ratio for the DOPS-containing bilayers (see Figure 4). Therefore, it seems likely that adsorption strength affects on-edge pore formation, at least for the DOPC/DOPS/TR-DHPE systems.


17

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 29

Figure 4. Dependence of the percentages of SLB patches containing on-edge (circles with solid lines) and off-edge defect pores (squares with dotted lines) on NaCl concentration. SLB patches were formed through asymmetric rupture of isolated vesicles. DOPC/DOPS/TR-DHPE (88:10:2) GUVs were adsorbed on glass in 10 mM sodium phosphate at pH 7.2 (black), 10 mM HEPES at pH 7.2 (red), 2 mM CaCl2 and 10 mM HEPES at pH 7.2 (blue), and 10 mM sodium acetate at pH 4.0 (orange) containing different concentrations of NaCl. DOPC/DOTAP/TRDHPE (88:10:2) GUVs were adsorbed on glass in 10 mM sodium phosphate at pH 7.2 containing different concentrations of NaCl (green).


18

Page 19 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

Figure 5. Dependence of zeta potential determined for LUVs (diameter, ∼100 nm) on NaCl concentration. DOPC/DOPS/TR-DHPE (88:10:2) LUVs in 10 mM sodium phosphate at pH 7.2 (black), 10 mM HEPES at pH 7.2 (red), 2 mM CaCl2 and 10 mM HEPES at pH 7.2 (blue), and 10 mM sodium acetate at pH 4.0 (orange) containing different concentrations of NaCl. DOPC/DOTAP/TR-DHPE (88:10:2) LUVs in 10 mM sodium phosphate at pH 7.2 containing different concentrations of NaCl (green). Values (± SE) are the average of three to four determinations.

DISCUSSION The major conclusions from this study are: 1) GUV rupture is induced by defect pore formation at the glass–bilayer interface; and 2) the probability of defect-induced rupture occurring is affected by salt concentration, pH, and lipid composition. We now rationalize the second conclusion by considering energetics. Defect pores preferentially formed in close vicinity to the edge of the adsorbed area (Figure 2C, 3A), occurring twice as often as pore formation at a site distant from the boundary of the


19

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 29

adsorbed region (Figure 2D, 3B). These results can be explained by considering bilayer curvature. DOPC and DOPS with near-zero curvatures are ill-suited for curved monolayers with markedly higher negative and positive curvatures than their own.24, 25 A pore is therefore readily nucleated at the most highly curved region to relax the stress imposed by the curvature. We expect that a small primitive pore formed in this curved region subsequently grows toward either the adsorbed or nonadsorbed region. When a pore expands to the adsorbed area, it develops into a defect pore. When a pore expands to the free bilayer region, it becomes a rupture pore. Because the highly curved region exists close to the glass surface, the curvature stress is relaxed only by pore expansion in the nonadsorbed area. In contrast, pore growth in the adsorbed region requires the displacement of adsorbed lipids and therefore, at first glance, is disfavored energetically because of the adsorption energy barrier. In a weak adsorption regime where vesicles are not deformed much, pore formation at the glass–bilayer interface may occur because of low curvature energy and low adsorption energy. However, this hypothesis is inconsistent with the results obtained for DOPS-containing GUVs (Figure 4). As the salt concentration was increased, the adsorption energy was expected to increase because of the screening of electrostatic repulsion between the negatively charged glass surface and anionic vesicles. Thus, the above hypothesis predicts that the proportion of defect-bearing SLB patches should decrease with increasing salt level. However, we obtained the opposite result; the proportion of defectbearing bilayer patches increased at higher salt concentration. Therefore, other energetic factors in addition to the curvature and adsorption energies must be considered. We propose that the difference in lateral tension between the adsorbed and nonadsorbed regions may account for the observed results, as discussed below.


20

Page 21 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

Free vesicles with a pore and adsorbed vesicles with or without a pore have been previously expressed using energetic contributions from curvature, adsorption, line tension, and lateral tension.8, 26, 27, 28, 29 The curvature is the energy required to bend the unstressed bilayer, and is correlated with molecular curvature, as explained above.30, 31 The adsorption energy increases in proportion to the adsorbed area, assuming a homogeneous bilayer–surface interaction over the surface.8 Line tension arises from the molecular arrangement at the pore edge deviating from the unperturbed bilayer structure.29,

32

Lipids at the edge are probably oriented with their polar

headgroups pointing toward the solution to shield their hydrocarbon chains from the aqueous environment.33 Lateral tension results from a change in bilayer area.28 It is believed that the attractive interaction from the surface stretches bilayers to induce lateral tension.12, 34 A local difference in lateral tension between the adsorbed and nonadsorbed regions has not been considered in previous literature.9, 12 Now consider a vesicle with a spherical cap shape where the highly curved region lies a very limited distance above the glass surface. We assume that the curvature stress is released by the formation of a pinhole in this curved area and is negligible for the subsequent hole enlargement. In this rough estimate, the energies required to expand a circular pore of radius r from the pinhole in the curved area to the nonadsorbed (Eb) and adsorbed (Es) regions are different as follows: − = 2π − + π − − ,

(1)

where γb and γs are the line tension along the pore edge in the bulk solution and on the surface, respectively, Σb and Σs are the lateral tension of the nonadsorbed and adsorbed regions, respectively, and A is the adsorption energy per unit area. The difference in line tension can then be ignored (i.e., γb ≈ γs) because the pore edge in the adsorbed region does not contact with the


21

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 22 of 29

glass surface and therefore is expected to exist in a similar environment to a pore edge in the nonadsorbed region. Thus, we obtain the following relation: − = π − − .

(2)

When the lateral tension is uniform over the vesicle surface (Σb = Σs), the following relation holds: − = −π .

(3)

Equation 3 indicates that strong adsorption prevents pore formation in the glass-adsorbed region. However, this argument does not agree with the results obtained for DOPS-containing GUVs, as discussed above. The data in Figure 4 show that as the salt concentration (i.e., A) increased, pore formation in the adsorbed region occurred more readily. We therefore postulate that the lateral tension was locally different; that is, Σb ≠ Σs. It is intuitively conceivable that the glass surface affects the tension of the adsorbed region (Σs) more than that of the nonadsorbed region (Σb). Equation 2 may explain the results in Figure 4 in the following manner. For negatively charged GUVs, the contribution from the difference in tension (Σb − Σs) became larger with increasing salt content because the attractive force from the surface became stronger. For positively charged GUVs, the contribution from A was large compared with that from local tension because of strong electrostatic attraction from the glass surface. The on-edge defect ratio for DOPScontaining GUVs (Figure 4), the zeta potential for LUVs with the same lipid composition (Figure 5), and the glass surface potential changed considerably with salt concentration in the few tens of millimolar range, but did not change markedly in the few hundred millimolar range. Thus, the electrostatic interaction between the glass surface and GUVs is likely to affect the probability of defect generation, consistent with the notion that the balance between (Σb − Σs) and A influences the formation of defect pores.


22

Page 23 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

More precise understanding of this phenomenon, especially of how the lateral tension varies locally with adsorption strength, requires further investigation. In addition, it is necessary to take into account unequal lipid partitioning in the two leaflets of bilayers. Asymmetric interleaflet distribution of DOPS or DOTAP sometimes occurs in planar bilayers formed by adsorption of SUVs or LUVs on a charged surface, which is explained by an electrostatic- or divalent ionmediated effect.10, 35, 36, 37 Asymmetric partitioning of TR-DHPE depending on pH has also been found for GUVs adsorbed on glass, which is rationalized by the balance between electrostatic repulsion and hydrogen bonding.18 Consideration of such asymmetry in lipid partitioning will help us to better understand the role of electrostatic interactions in pore formation behavior. The present results revealed that pores can form in both the adsorbed and nonadsorbed bilayers in GUVs. It should be emphasized that a substantial percentage (~40%) of negatively charged vesicles under physiological conditions ruptured through defect-induced mechanisms. Biological membranes are often negatively charged. Thus, one may need to consider pore formation in adsorbed regions when studying biological membranes and their model materials.

CONCLUSIONS We have shown that asymmetric rupture of GUVs is induced by defect pores formed in the adsorbed bilayer region. Pores formed in the glass-adsorbed area before rupture. When a pore formed near the edge of the adsorbed area or when the pore edge came in contact with the edge of the adsorbed region by expansion, asymmetric rupture was triggered from the defect site. These induced rupture mechanisms yielded SLB patches with an on-edge pore. Vesicles seldom ruptured at a different site from the defect pore; this rare event yielded SLB patches with an offedge defect pore. The frequency of defect-induced rupture increased with salt concentration, and


23

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 24 of 29

was lower for positively charged GUVs compared with negatively charged ones. We proposed that local differences in lateral tension may play an important role in pore formation in the adsorbed area of GUVs.

ASSOCIATED CONTENT Supporting Information. AFM data for the clean glass surface. This material is available free of charge via the Internet at http://pubs.acs.org.

ACKNOWLEDGMENTS This work was supported by the Japan Science and Technology Agency (KAKENHI 24710127) and the National Institute for Materials Science Molecule & Material Synthesis Platform (the Nanotechnology Platform Project operated by the Ministry of Education, Culture, Sports, Science and Technology, Japan).

REFERENCES 1.

Hamai, C.; Cremer, P. S.; Musser, S. M. Single Giant Vesicle Rupture Events Reveal

Multiple Mechanisms of Glass-Supported Bilayer Formation. Biophys. J. 2007, 92, 1988−1999. 2.

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.


24

Page 25 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

3.

Kresák, S.; Hianik, T.; Naumann, R. L. C. Giga-Seal Solvent-Free Bilayer Lipid

Membranes: From Single Nanopores to Nanopore Arrays. Soft Matter 2009, 5, 4021−4032. 4.

Kusters, I.; van Oijen, A. M.; Driessen, A. J. M. Membrane-on-a-Chip: Microstructured

Silicon/Silicon-Dioxide Chips for High-Throughput Screening of Membrane Transport and Viral Membrane Fusion. ACS Nano 2014, 8, 3380−3392. 5.

Lazzara, T. D.; Carnarius, C.; Kocun, M.; Janshoff, A.; Steinem, C. Separating Attoliter-

Sized Compartments Using Fluid Pore-Spanning Lipid Bilayers. ACS Nano 2011, 5, 6935−6944. 6.

Bernard, A.-L.; Guedeau-Boudeville, M.-A.; Jullien, L.; di Meglio, J.-M. Strong

Adhesion of Giant Vesicles on Surfaces: Dynamics and Permeability. Langmuir 2000, 16, 6809− 6820. 7.

Lipowsky, R.; Seifert, U. Adhesion of Membranes: A Theoretical Perspective. Langmuir

1991, 7, 1867−1873. 8.

Seifert, U. Configurations of Fluid Membranes and Vesicles. Adv. Phys. 1997, 46, 13−

137. 9.

Lipowsky, R.; Seifert, U. Adhesion of Vesicles and Membranes. Mol. Cryst. Liq. Cryst.

1991, 202, 17−25. 10.

Richter, R. P.; Bérat, R.; Brisson, A. R. Formation of Solid-Supported Lipid Bilayers: An

Integrated View. Langmuir 2006, 22, 3497−3505. 11.

Castellana, E. T.; Cremer, P. S. Solid Supported Lipid Bilayers: From Biophysical

Studies to Sensor Design. Surf. Sci. Rep. 2006, 61, 429−444. 12.

Sandre, O.; Moreaux, L.; Brochard-Wyart, F. Dynamics of Transient Pores in Stretched

Vesicles. Proc. Natl. Acad. Sci. U.S.A. 1999, 96, 10591−10596.


25

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

13.

Page 26 of 29

Hategan, A.; Law, R.; Kahn, S.; Discher, D. E. Adhesively-Tensed Cell Membranes:

Lysis Kinetics and Atomic Force Microscopy Probing. Biophys. J. 2003, 85, 2746−2759. 14.

Dimitrievski, K.; Kasemo, B. Simulations of Lipid Vesicle Adsorption for Different

Lipid Mixtures. Langmuir 2008, 24, 4077−4091. 15.

Fuhrmans, M.; Müller, M. Mechanisms of Vesicle Spreading on Surfaces: Coarse-

Grained Simulations. Langmuir 2013, 29, 4335−4349. 16.

Akashi, K.; Miyata, H.; Itoh, H.; Kinosita, K. Preparation of Giant Liposomes in

Physiological Conditions and Their Characterization Under an Optical Microscope. Biophys. J. 1996, 71, 3242−3250. 17.

Hamai, C.; Yang, T. L.; Kataoka, S.; Cremer, P. S.; Musser, S. M. Effect of Average

Phospholipid Curvature on Supported Bilayer Formation on Glass by Vesicle Fusion. Biophys. J. 2006, 90, 1241−1248. 18.

Kataoka-Hamai, C.; Higuchi, M. Packing Density Changes of Supported Lipid Bilayers

Observed by Fluorescence Microscopy and Quartz Crystal Microbalance-Dissipation. J. Phys. Chem. B 2014, 118, 10934−10944. 19.

Tsui, F. C.; Ojcius, D. M.; Hubbell, W. L. The Intrinsic pKa Values for

Phosphatidylserine and Phosphatidylethanolamine in Phosphatidylcholine Host Bilayers. Biophys. J. 1986, 49, 459−468. 20.

Sinn, C. G.; Antonietti, M.; Dimova, R. Binding of Calcium to Phosphatidylcholine-

Phosphatidylserine Membranes. Colloids Surf., A 2006, 282, 410−419. 21.

Bayerl, T. M.; Bloom, M. Physical Properties of Single Phospholipid Bilayers Adsorbed

to Micro Glass Beads. A New Vesicular Model System Studied by 2H-Nuclear Magnetic Resonance. Biophys. J. 1990, 58, 357−362.


26

Page 27 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

22.

Fromherz, P.; Kiessling, V.; Kottig, K.; Zeck, G. Membrane Transistor with Giant Lipid

Vesicle Touching a Silicon Chip. Appl. Phys. A 1999, 69, 571−576. 23.

Kiessling, V.; Tamm, L. K. Measuring Distances in Supported Bilayers by Fluorescence

Interference-Contrast Microscopy: Polymer Supports and SNARE Proteins. Biophys. J. 2003, 84, 408−418. 24.

Fuller, N.; Benatti, C. R.; Rand, R. P. Curvature and Bending Constants for

Phosphatidylserine-Containing Membranes. Biophys. J. 2003, 85, 1667−1674. 25.

Szule, J. A.; Fuller, N. L.; Rand, R. P. The Effects of Acyl Chain Length and Saturation

of Diacylglycerols and Phosphatidylcholines on Membrane Monolayer Curvature. Biophys. J. 2002, 83, 977−984. 26.

Taupin, C.; Dvolaitzky, M.; Sauterey, C. Osmotic Pressure Induced Pores in

Phospholipid Vesicles. Biochemistry 1975, 14, 4771−4775. 27.

Zhdanov, V. P.; Kasemo, B. Comments on Rupture of Absorbed Vesicles. Langmuir

2001, 17, 3518−3521. 28.

Seifert, U.; Lipowsky, R. Adhesion of Vesicles. Phys. Rev. A 1990, 42, 4768−4771.

29.

Karatekin, E.; Sandre, O.; Guitouni, H.; Borghi, N.; Puech, P. H.; Brochard-Wyart, F.

Cascades of Transient Pores in Giant Vesicles: Line Tension and Transport. Biophys. J. 2003, 84, 1734−1749. 30.

Gruner, S. M. Intrinsic Curvature Hypothesis for Biomembrane Lipid Composition: A

Role for Nonbilayer Lipids. Proc. Natl. Acad. Sci. U.S.A. 1985, 82, 3665−3669. 31.

Zimmerberg, J.; Kozlov, M. M. How Proteins Produce Cellular Membrane Curvature.

Nat. Rev. Mol. Cell Biol. 2006, 7, 9−19.


27

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

32.

Page 28 of 29

May, S. A Molecular Model for the Line Tension of Lipid Membranes. Eur. Phys. J. E

2000, 3, 37−44. 33.

Tieleman, D. P.; Leontiadou, H.; Mark, A. E.; Marrink, S. J. Simulation of Pore

Formation in Lipid Bilayers by Mechanical Stress and Electric Fields. J. Am. Chem. Soc. 2003, 125, 6382−6383. 34.

Brochard-Wyart, F.; de Gennes, P. G.; Sandre, O. Transient Pores in Stretched Vesicles:

Role of Leak-Out. Phys. A 2000, 278, 32−51. 35.

Richter, R. P.; Him, J. L. K.; Tessier, B.; Tessier, C.; Brisson, A. R. On the Kinetics of

Adsorption and Two-Dimensional Self-Assembly of Annexin A5 on Supported Lipid Bilayers. Biophys. J. 2005, 89, 3372−3385. 36.

Rossetti, F. F.; Textor, M.; Reviakine, I. Asymmetric Distribution of Phosphatidyl Serine

in Supported Phospholipid Bilayers on Titanium Dioxide. Langmuir 2006, 22, 3467−3473. 37.

Solon, J.; Streicher, P.; Richter, R.; Brochard-Wyart, F.; Bassereau, P. Vesicles Surfing

on a Lipid Bilayer: Self-Induced Haptotactic Motion. Proc. Natl. Acad. Sci. U.S.A. 2006, 103, 12382−12387.


28

Page 29 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

TOC Graphic-


29

Induced Rupture of Vesicles Adsorbed on Glass by Pore Formation at

Recommend Documents