Spatio-Temporal Kinetics of Supported Lipid Bilayer Formation on

Subscriber access provided by Queen Mary, University of London

Biophysical Chemistry, Biomolecules, and Biomaterials; Surfactants and Membranes

Spatio-Temporal Kinetics of Supported Lipid Bilayer Formation on Glass via Vesicle Adsorption and Rupture Mokhtar Mapar, Silver Jõemetsa, Hudson P. Pace, Vladimir P. Zhdanov, Björn Agnarsson, and Fredrik Höök J. Phys. Chem. Lett., Just Accepted Manuscript • DOI: 10.1021/acs.jpclett.8b02092 • Publication Date (Web): 23 Aug 2018 Downloaded from http://pubs.acs.org on August 23, 2018

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 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 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.

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 20 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

The Journal of Physical Chemistry Letters

Spatio-temporal Kinetics of Supported Lipid Bilayer

Formation

on

Glass

via

Vesicle

Adsorption and Rupture

Mokhtar Mapara, Silver Jõemetsaa, Hudson Pacea, Vladimir P. Zhdanova,b, Björn Agnarssona and Fredrik Hööka,* a

Division of Biological Physics, Department of Physics, Chalmers University of Technology,

Göteborg, Sweden b

Boreskov Institute of Catalysis, Russian Academy of Sciences, Novosibirsk, Russia

Corresponding Author Fredrik Höök: [email protected]

1 ACS Paragon Plus Environment

The Journal of Physical Chemistry Letters 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 Supported lipid bilayers (SLBs) represent one of the most popular mimics of the cell membrane. Herein, we have used total internal reflection fluorescence microscopy for indepth characterization of the vesicle-mediated SLB formation mechanism on a common silica-rich substrate, borosilicate glass. Fluorescently labelling a subset of vesicles allowed us to monitor the adsorption of individual labeled vesicles, resolve the onset of SLB formation from small seeds of SLB patches, and track their growth via SLB-edge-induced autocatalytic rupture of adsorbed vesicles. This made it possible to perform the first quantitative measurement of the SLB front velocity, which is shown to increase up to one order of magnitude with time. This effect can be classified as dramatic, because in many other physical, chemical, or biological kinetic processes the front velocity is either constant or decreasing with time. The observation was successfully described with a theoretical model and Monte Carlo simulations implying rapid local diffusion of lipids upon vesicle rupture.

TOC GRAPHICS:


Page 2 of 20

Page 3 of 20 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


Since the pioneering work by McConnell et al.1, supported lipid bilayers (SLBs) have become one of the most widely used mimics of cellular membranes, prompting many investigations into characterizing their formation.2–4 Due to its planar geometry and preserved fluidity over macroscopic areas, SLBs offer unique opportunities to apply surface-sensitive instrumentation for studies of membrane proteins5, membrane phase separation6, membrane interaction with viruses7 and nanoparticles8, as well as many other membrane-related processes. SLBs have also been exploited in biotechnology applications, related e.g. to sensors and drug-screening.3,4,6 In all such studies the quality of SLBs is a critical factor which can affect their utility and therefore, significant efforts have been invested to gain an in-depth understanding of the SLB formation process. Among the different protocols to form SLBs (Ref. 9 and references therein), the most popular one proposed originally by Kalb et al.10 is based on vesicle adsorption and rupture. Numerous studies were motivated not only by potential applications of this method, but also by the desire to clarify the corresponding non-trivial mechanistic details behind it. Already in the early works employing quartz crystal microbalance with dissipation monitoring (QCMD), surface plasmon resonance (SPR)11, fluorescence microscopy12, and atomic force microscopy13, it was shown that the initial vesicle adsorption step is typically mass-transport limited and that often, e.g. in the generic case of the SiO2-based supports, SLB formation is not initiated until a critical vesicle coverage has been reached. The analysis of the QCM-D kinetics11,14 and related Monte Carlo simulations15 implied that the vesicle rupture can (i) be fully spontaneous, i.e. take place at the level of single vesicles, (ii) occur via vesicle fusion, and / or (iii) be induced by already formed SLB patches at the patch boundaries, and propagate via wave-like spreading, and the conclusion was that channel (iii) dominates. Later, QCM-D and localized SPR techniques were widely used to identify the effect of various governing parameters (e.g. vesicle size, lipid composition, ionic strength, osmotic 3 ACS Paragon Plus Environment


pressure, pH, and temperature) on the kinetics of the SLB formation process.16–18 In the context of mechanistic details, an important piece of information was recently added by Kukura et al.19 Using interferometric scattering microscopy with single-vesicle resolution and high image acquisition rate, they investigated the SLB formation process on borosilicate glass and provided direct evidence for the previously suggested concepts including (i) onset of bilayer patch formation in local nanoscopic regions with high vesicle coverage, (ii) dependence of the onset of SLB formation on vesicle dimension, as well as (iii) wave-like growth of SLB patches during completion of the SLB formation process. In contrast, the fluorescence microscopy inspection of SLB formation20,21 suggests growth of irregular SLB patches already at low vesicle coverage. In this work, we address this question by quantifying the spatio-temporal kinetics of SLB formation at the single-vesicle level on borosilicate glass employing total internal reflection fluorescence (TIRF) microscopy. In comparison to how fluorescence microscopy was previously employed to study SLB formation,20,21 labelling of a small fraction (1%) of the vesicles allowed us to both discern individual adsorbed labeled vesicles and increase the optical contrast between planar bilayer patches and their surrounding regions.22 In this way, we were able to track and temporally resolve the surface density of adsorbed vesicles, observe the onset of SLB formation, and investigate the growth kinetics of the SLB patches. We particularly focus on the SLB patch growth kinetics in the context of two theoretically possible limits regarding the incorporation of lipids into the SLB patch after the rupture of surface-adsorbed vesicles at the border of a SLB patch: (i) local relaxation of released lipids, characterized by the SLB growth being restricted to the immediate vicinity of each ruptured vesicle, and (ii) global relaxation of released lipids, characterized by rapid coalescence of all the lipids into the SLB patch resulting in a perfectly circular patch. As detailed below, this


Page 4 of 20

Page 5 of 20 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


combined experimental and theoretical approach allowed us to obtain the first quantitative characterization of SLB patch growth kinetics during SLB formation. All experiments were performed under stagnant condition in PDMS wells formed by the conformal contact of a PDMS slab on a clean borosilicate glass. Each PDMS well (diameter ~2 mm) was filled with 9 µL of buffer followed by addition and mixing of 1 µL of 1:100 mixture of labeled and unlabeled lipid vesicles (ranging from 70-90 nm in size according to nanoparticle tracking analysis (NTA)) at a lipid concentration of 1 mg/mL (see Supporting Information). Typical TIRF micrographs of the SLB formation process (figure 1a) show four different phases, including initial binding of vesicles (t ≈ 250 s), appearance of the initial expanding SLB patches (t ≈ 600 s), accelerated growth of multiple SLB patches (t ≈ 700 s), and coalescence of adjacent bilayer patches (t ≈ 800 s) prior to formation of a macroscopic continuous SLB (not shown). With only 1% of the vesicles being fluorescently labeled, adsorption of each individual labeled vesicle could be readily resolved up to and beyond the initiation of SLB patch formation. The size of the vesicles is smaller than the wavelength, but rupture events can nevertheless be detected, because the size scale characterizing the decay of the evanescent field is small and the rupture is accompanied by the change of the dye-related light intensity. Additionally, mixing of lipids from labeled and unlabeled vesicles ruptured during SLB patch formation, made it possible to simultaneously visualize and temporally resolve the appearance of weakly fluorescent SLB patches with submicrometer resolution.



Figure 1. Spatio-temporal kinetics of the SLB formation: (a) Four micrographs illustrating different phases of the SLB formation process where 1% of the vesicles in the solution are fluorescently labeled. The rupture of vesicles is schematically shown on top. The first observed rupture (red contour) is denoted with an orange asterisk. (b) Number of detected intact tracer vesicles (red circles) and surface density of tracer vesicles (black squares) plotted versus time, for a 100 × 100 µm2 area. The open grey circles along the red curve indicate the moments corresponding to the micrographs. The inset illustrates how for 3 instances the recorded intensity raises and drops as vesicles are adsorbed on the surface and later rupture and fuse with an SLB.(c) Estimated SLB area normalized to the observation area plotted versus time. The inset magnifies the initial stage of the process showing sparse rupture of vesicles early in the adsorption stage.

Figure 1b shows how the number of unruptured vesicles on the surface increases until a critical surface density is reached, at which the rupture of vesicles cascades into complete SLB formation (red curve). The surface density of vesicles is seen to increase by about 8% from the appearance of the first growing SLB patches until the SLB completely covers the surface (black curve). The inset in figure 1b shows representative curves exhibiting how the recorded intensities of the selected areas first increase and then drop to a common intensity value, which is attributed to adsorption of labeled vesicles followed by SLB-patch induced rupture and fusion with the patch. 6 ACS Paragon Plus Environment

Page 6 of 20

Page 7 of 20 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


Under the reasonable assumption that fluorescently labeled vesicles constitute 1% of the adsorbed vesicles, the relative areal growth of the SLB (figure 1c) was estimated (see Supporting Information) based on the number of rupturing fluorescent vesicles in each frame, the surface density of adsorbed vesicles, the surface area of substrate available for SLB formation, and the 1/100 scaling factor. The analysis reveals sparse appearance of SLB patches early in the adsorption process (figure 1c, inset) with a negligible contribution to the total SLB area, which instead grows at a considerably higher rate in the later stages of the process. Assuming that a lipid occupies the same surface area in a vesicle and a SLB, a direct translation of the surface density of adsorbed vesicles (figure 1b, black curve) suggests that the total lipid mass prior to the appearance of patch growth is ~1.3 times that needed for a complete SLB on the substrate. Note, though, that lipid coverage is likely somewhat lower because (i) the lipid content of a typical vesicle is estimated based on the hydrodynamic radius rather than the real radius23 and (ii) there might be a slight preference towards adsorption of smaller vesicles24 since the initial adsorption is mass-transport limited11. Taken together, our observations are thus in excellent agreement with previously proposed models suggesting that SLB formation is initiated via local vesicle rupture into small SLB patches followed by subsequent growth via edge-catalyzed vesicle rupture11,12,19,25. To quantify the individual SLB patch evolution and different patch-growth scenarios, the boundary of each patch was determined for each analyzed frame and subsequently used to calculate the average front velocity (see Supporting Information). Figure 2 illustrates the commonly observed scenarios including (a) single patch formation and subsequent growth, (b) which typically merges with other patches into a larger patch, and (c) occasionally leads to a large SLB patch advancing across the entire field of view. In the corresponding snapshots, the local SLB front velocity, v(t), is represented by the color of the contour indicating the front location. Also shown is the average front velocity evolution during the 7 ACS Paragon Plus Environment


patch formation and expansion (each data point represents the average velocity for the corresponding frame). In cases (a) and (b), as the patches grew larger, the average front velocity increased. In case (c), there is slight rise and fall in the propagation rate which is tentatively attributed to local variations in the adsorbed vesicle density or existence of dark patches. During the observation interval, there usually was a 5-10-fold increase in the bilayer front velocity while the surface density of adsorbed vesicles increased less than ~8% (figure 1b, black curve, 600 < t < 770 s). This modest increase in the vesicle density cannot account for the significant increase in the bilayer front velocity observed in a short interval, strongly indicating that the SLB-front propagation is mainly governed by the already adsorbed vesicles. This minor increase in the surface density of vesicles is also not expected to have any significant influence on the stress or deformation of the adsorbed vesicles, suggesting that this contribution to the vesicle rupture rate near the patch edges is negligible. Our observation thus confirms that the SLB-front propagation is dominated by the autocatalytic edge-induced rupture of already adsorbed vesicles.


Page 8 of 20

Page 9 of 20 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


Figure 2. Representative patterns observed during the SLB-patch growth and the corresponding dependence of the average front velocity on time: (a) Single patch formation and expansion (snapshot interval 60 s). (b) Small patches merging into a bigger patch (snapshot interval 24 sec). (c) Propagating SLB front (snapshot interval 30 sec). Scale bars are 20 µm. The color of the lines showing the front of the SLB patch, indicates the local front velocity. For the surface density curve corresponding to (a) and (b), see figure 1b. The fluctuations in the average growth rate (including plateaus) are related to the stochastic nature of the growth, and that their fine structure is influenced by the frame interval.

To clarify the mechanism of SLB front propagation, we used two generic coarse-grained models with identical schemes of vesicles adsorption and SLB-edge-induced rupture, but alternative schemes of lipid relaxation after vesicle rupture. The corresponding analysis was performed employing Monte Carlo (MC) simulations based on the standard Gillespie algorithm, taking into account that the front propagation is experimentally observed to occur at relatively high vesicle surface density and that the adsorption of additional vesicles to the surface during the vesicle rupture cascade is negligible. The initial arrangement of adsorbed vesicles was generated using the conventional random-sequential-adsorption algorithm. After 9 ACS Paragon Plus Environment


attachment to the substrate, the vesicle-substrate interaction was assumed to induce physically reasonable cap-like vesicle deformation in which the radius of the vesiclesubstrate contact area is chosen to be equal to the vesicle radius, R, in the non-deformed state. The density of the vesicles was selected so that πR2c = 0.45, which is sufficient for the front propagation. The SLB formation was initiated by the rupture of an adsorbed vesicle, forming a small SLB patch with the area of 4πR2. This patch was further assumed to induce rupture of adjacent vesicles that are in contact with it, at a rate constant k, thereby resulting in front propagation. The simulations were performed on a 500R × 500R area. The first scheme of lipid relaxation after vesicle rupture implies that this process is local. In particular, the lipids belonging to a ruptured vesicle are assumed to form a circular SLB patch within radius 2R around the vesicle center coinciding with that of the vesicle center irrespective of the arrangement of the other vesicles or earlier formed SLB patches. Whether such a patch contacts other vesicles depends on their center-to-center distance. For the assumed hemispherical shape of adsorbed vesicles, an SLB patch formed after a vesicle rupture contacts and induces rupture of a neighboring vesicle if their center-to-center distance is within or equal to 3R. In this scheme, the rearrangement of lipids in newly-formed patches is assumed to be negligible, implying that overlap with the earlier formed patches can be interpreted as either a double bilayer or material loss. Consequently the areal growth of the SLB region due to rupture of a vesicle is always smaller than its maximum value, 4 . The radius of the whole SLB patch, , was further identified with the SLB front coordinate and determined by relating it with the number of ruptured vesicles, = or = ⁄

( ⁄ )

where c is the already defined initial surface concentration of vesicles.

The second scheme of lipid relaxation implies that this process is global. In particular, we instead assume that upon SLB-induced rupture of an adsorbed vesicle, the released lipids


Page 10 of 20

Page 11 of 20 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


rapidly diffuse along the substrate and associate with the boundary of a perfectly circular SLB patch with radius, (), resulting in an area increment by 4 . The two models described above from the perspective of MC simulations were treated also analytically by using reasonable simplifications. In particular, we exploit that the average time of a rupture for a vesicle contacting a SLB is ∆ = 1⁄ . For the local-relaxation model, the increment of corresponding to this time increment is comparable with R, i.e., we have

( + ∆) = + where ≅ 2 is a dimensionless factor (which depends on the vesicle coverage and vesicle shape). For a well-developed front (at ≫ ∆) this relation yields

≅

(∆) () ∆

= , or () = .

(1)

In the case of global relaxation, we take into account that the patch of radius () is formed from rupturing of vesicles located within < " ( − ∆). The number of vesicles used to form this patch is ( − ∆) , and accordingly its area is () = ( − ∆)4 , or

() = $ ( − ∆), where $ = (4 )⁄ . Using the latter equation, we have % ⁄% ≅ & () − ( − ∆)'⁄∆ = (1 − 1⁄$ ) (), or ≅ (0)exp&(1 − 1⁄$ )',

(2)

where rf(0) is the patch radius at the onset of patch formation. The MC simulations of the two scenarios are illustrated in figure 3a. Except for a short initial period, the front propagation for the local-relaxation model is linear at ≥ 8 and exponential for the global-relaxation model at ≥ 3 which is in agreement with the analytical models represented by Eqs. 1 and 2. Due to restrictions in lipid diffusivity, which can be understood as regions with overlapping SLBs or loss of lipid material, the localrelaxation model (Eq. 1) predicts, as one could expect, a significantly slower SLB growth



than that predicted by the global-relaxation model (Eq. 2) where all lipids from ruptured vesicles merge with the SLB patch to form a circular SLB patch.

Figure 3. Radius of the SLB patch as a function of dimensionless time (kt): (a) Monte Carlo simulations for two hypothetical SLB-formation mechanisms with global (blue) and local (red) relaxation of lipids after vesicle rupture. (b) Experimental data for six rather circular patches, including a fit to the global-relaxation model. The numbers noted for the different curves indicate the lipid content of the adsorbed vesicles relative to that of an SLB patch covering a similar area (or, in other word, the lipid content ratio which equals $ ) at the time of patch formation.

Manifested by increasing front velocities (figure 2), the experimental observations are in agreement with the global-relaxation model. To compare the measured growth kinetics with the theoretical model (figure 3b), least square fits of Eq. 2 were made to the growth kinetics of six representative single circular patches of the type shown in figure 2a for an SLB formation using POPC vesicles with a nominal hydrodynamic radius of 35 nm, measured with NTA. The local surface density of vesicles around each patch was estimated from the TIRF micrographs. The obtained value for the rate constant of vesicle rupture, k, was between 0.08 and 0.5 s-1, with a slight tendency towards lower values for higher local coverages. Similar observations were made with POPC vesicles of similar size containing 1 mol% PEGylated lipids (see Supporting Information). Although the rupture time constant (k) varied by a factor of up to 5 from case to case, the correlation between high local vesicle coverage and high growth rate was not appreciable 12 ACS Paragon Plus Environment

Page 12 of 20

Page 13 of 20 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


(figure S2), suggesting the reason for the observed differences is not related to variations in local vesicle coverage, but rather to heterogeneities in: (i) size distribution of vesicles within the suspensions and their lamellarity (the presence of a small fraction of multilamellar vesicles cannot be excluded), which are both factors that are likely to influence the vesicle rupture process, but not possible to accurately control using conventional vesicle preparation methods,27,28 and (ii) surface properties (e.g., local variations in surface roughness, defect density and charge), including how these properties are influenced by cleaning protocols and handling of the surface prior to measurements. This conclusion is further supported by the fact that although the SLB formation process was indeed dominated by the scenarios presented above (figure 2a-c), a number of less frequent mechanistic features were occasionally observed during the process which highlights the complexity of the SLB formation process. One such observation concerns the patches formed early in the vesicle adsorption process (figure 4a). These patches do not grow (the first patch) or have negligible or decaying growth rate (the second patch). However, the later formed patches in the same area (e.g. the third patch) experience a sustainable growth. The three patches in figure 4a have a lipid content ratio (the lipid content of the adsorbed vesicles relative to that of an SLB patch covering a similar area, i.e. $ ) of about 0.9, 1 and 1.15 respectively at the time of formation, suggesting that the local concentration of vesicles adjacent to the patch has to reach a certain minimum threshold to sustain the growth process. This is indeed in accordance with the global-relaxation model in which bilayer growth halts if β < 1, i.e. when the lipid content of the adsorbed vesicles per surface area is below that of an SLB (Eq. 2). Hence, when the local surface density of adsorbed vesicles surrounding a spontaneously formed patch is not sufficient to support accelerating SLB growth, the patch will simply cover as large surface area as possible before halting and become stagnant. Such a stagnant patch can, however, start growing again if more lipid material is incorporated into it for



example by the spontaneous rupture of vesicles at its boundary or catalyzed rupture of a vesicles later adsorbed infinitely close to its edge or if it experiences a small expansion or rearrangement over time thereby allowing it to come in contact with surrounding vesicles (or even upon merger with a small “invisible” patch). Further, upon coalescing of such a stagnant or slowly expanding patch with a SLB front that follows the more common growth process, there is typically an instantaneous increase in the front propagation velocity of the stagnant (or slowly expanding) patch (figure 4b). Such a sudden increase in the front velocity suggests that the local conditions around the SLB patch, such as the local surface density of adsorbed vesicles, may not be the only factor defining the rate of SLB front propagation of an already formed patch, including the fairly large spread in k values. It is beyond the scope of this work to inspect these curious, yet most likely very complex events within the SLB formation process in further detail. In future work, further investigation of the heterogeneity could be addressed by employing combined label-free and label-based imaging approaches with higher acquisition rates in addition to including experimental optimization with respect to reduced vesicle and surface heterogeneity.

Figure 4. (a) Patches formed early on during the vesicle adsorption process commonly have no or negligible growth rate. (b) When a stagnant bilayer patch is reached by a growing bilayer patch or


Page 14 of 20

Page 15 of 20 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


front, it typically starts growing rapidly in all directions. The color of the lines showing the front indicates the local front velocity.

To conclude, we have presented the first quantitative study of the kinetics of the SLBpatch growth during vesicle adsorption, and our key finding is that the front velocity may increase considerably, although the added material to the surface due to continuous adsorption of vesicles is negligible in this regime. This effect can be considered to be a novel observation, because in most other physical, chemical, or biological kinetic processes the front velocity is either constant, e.g., during the transition from a kinetically unstable or metastable state to a stable state (Fisher waves), or tends to decrease with time, e.g., during growth of aggregates (via Ostwald ripening) or domains. Further, we demonstrate that the main SLB formation mechanism on a silica-based substrate begins with the emergence of small, local SLB patches in conjunction with a minimum threshold surface density of adsorbed vesicles, and that the following accelerated growth could be reproduced using a global-relaxation model. In this model, the rate limiting step is characterized by a characteristic rupture time constant, which was found to display a fairly broad variation. This heterogeneity merits attention as it likely influences the quality of the resulting SLB, and should therefore be kept in mind in future efforts undertaken to understand the formation of SLBs with more complex lipid compositions and/or particularly, when forming SLBs from natural cell membrane vesicles.22,29



ASSOCIATED CONTENT

Supplementary Information. The Supplementary information contains the materials and methods section. AUTHOR INFORMATION

Author Contributions M.P. performed the TIRFM experiments, M.P., S.J. and B.A. analyzed the data, V.P.Z. developed the theory. H.P. and F.H. conceived the project. M.P. and S.J. wrote the article together with the other co-authors. Funding Sources The research leading to these results received funding from the Swedish Research Council under grant 2014-5557 and the Swedish Foundation for Strategic Research under grant no. RMA11-0104.

Notes The authors have no competing financial interests.


Page 16 of 20

Page 17 of 20 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


REFERENCES 1.

McConnell, H. M., Watts, T. H., Weis, R. M. & Brian, A. A. Supported Planar

Membranes in Studies of Cell-Cell Recognition in the Immune System. Biochim. Biophys. Acta - Rev. Biomembr. 864, 95–106 (1986). 2.

Cho, N.-J., Frank, C. W., Kasemo, B. & Höök, F. Quartz Crystal Microbalance With

Dissipation Monitoring of Supported Lipid Bilayers on Various Substrates. Nat. Protoc. 5, 1096–1106 (2010). 3.

Hardy, G. J., Nayak, R. & Zauscher, S. Model Cell Membranes: Techniques to Form

Complex Biomimetic Supported Lipid Bilayers via Vesicle Fusion. Curr. Opin. Colloid Interface Sci. 18, 448–458 (2013). 4.

van Weerd, J., Karperien, M. & Jonkheijm, P. Supported Lipid Bilayers for the

Generation of Dynamic Cell-Material Interfaces. Adv. Healthc. Mater. 4, 2743–2779 (2015). 5.

Roder, F. et al. Reconstitution of Membrane Proteins into Polymer-Supported

Membranes for Probing Diffusion and Interactions by Single Molecule Techniques. Anal. Chem. 83, 6792–6799 (2011). 6.

Alessandrini, A. & Facci, P. Phase Transitions in Supported Lipid Bilayers Studied by

AFM. Soft Matter 10, 7145–7164 (2014). 7.

Rydell, G. E., Dahlin, A. B., Hook, F. & Larson, G. QCM-D Studies of Human

Norovirus Vlps Binding to Glycosphingolipids in Supported Lipid Bilayers Reveal StrainSpecific Characteristics. Glycobiology 19, 1176–1184 (2009). 8.

Zhao, F. et al. TiO2 Nanoparticle Interactions with Supported Lipid Membranes – an

Example of Removal Of Membrane Patches. RSC Adv. 6, 91102–91110 (2016).



9.

Tabaei, S. R., Choi, J.-H., Haw Zan, G., Zhdanov, V. P. & Cho, N.-J. Solvent-

Assisted Lipid Bilayer Formation on Silicon Dioxide and Gold. Langmuir 30, 10363–10373 (2014). 10.

Kalb, E., Frey, S. & Tamm, L. K. Formation of Supported Planar Bilayers by Fusion

of Vesicles to Supported Phospholipid Monolayers. Biochim. Biophys. Acta - Biomembr.

1103, 307–316 (1992). 11.

Keller, C. A., Glasmästar, K., Zhdanov, V. P. & Kasemo, B. Formation of Supported

Membranes from Vesicles. Phys. Rev. Lett. 84, 5443–5446 (2000). 12.

Johnson, J. M., Ha, T., Chu, S. & Boxer, S. G. Early Steps of Supported Bilayer

Formation Probed by Single Vesicle Fluorescence Assays. Biophys. J. 83, 3371–3379 (2002). 13.

Schönherr, H., Johnson, J. M., Lenz, P., Frank, C. W. & Boxer, S. G. Vesicle

Adsorption and Lipid Bilayer Formation on Glass Studied by Atomic Force Microscopy. Langmuir 20, 11600–11606 (2004). 14.

Reimhult, E.; Larsson, C.; Kasemo, B.; Hook, F., Simultaneous Surface Plasmon

Resonance and Quartz Crystal Microbalance With Dissipation Monitoring Measurements of Biomolecular Adsorption Events Involving Structural Transformations and Variations in Coupled Water. Analytical Chemistry 76, 7211-7220 (2004). 15.

Zhdanov, V. P., Keller, C. A., Glasmästar, K. & Kasemo, B. Simulation of adsorption

kinetics of lipid vesicles. J. Chem. Phys. 112, 900–909 (2000). 16.

Richter, R. P., Bérat, R. & Brisson, A. R. Formation of Solid-Supported Lipid

Bilayers: An Integrated View. Langmuir 22, 3497–3505 (2006). 17.

Jackman, J. A., Choi, J., Zhdanov, V. P. & Cho, N. Influence of Osmotic Pressure on

Adhesion of Lipid Vesicles to Solid Supports. Langmuir 29, 11375–11384 (2013). 18 ACS Paragon Plus Environment

Page 18 of 20

Page 19 of 20 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


18.

Oh, E. et al. Contribution of Temperature to Deformation of Adsorbed Vesicles

Studied by Nanoplasmonic Biosensing. Langmuir 31, 771–781 (2015). 19.

Andrecka, J., Spillane, K. M., Ortega-Arroyo, J. & Kukura, P. Direct Observation and

Control of Supported Lipid Bilayer Formation with Interferometric Scattering Microscopy. ACS Nano 7, 10662–10670 (2013). 20.

Weirich, K. L., Israelachvili, J. N. & Fygenson, D. K. Bilayer Edges Catalyze

Supported Lipid Bilayer Formation. Biophys. J. 98, 85–92 (2010). 21.

Plunkett, P., Camley, B. A., Weirich, K. L., Israelachvili, J. & Atzberger, P. J.

Simulation of Edge Facilitated Adsorption And Critical Concentration Induced Rupture of Vesicles at a Surface. Soft Matter 9, 8420 (2013). 22.

Pace, H. et al. Preserved Transmembrane Protein Mobility in Polymer-Supported

Lipid Bilayers Derived from Cell Membranes. Anal. Chem. 87, 9194–9203 (2015). 23.

Lundgren, A. et al. Nonspecific Colloidal-Type Interaction Explains Size-Dependent

Specific Binding of Membrane-Targeted Nanoparticles. ACS Nano 10, 9974–9982 (2016). 24.

Olsson, T., Zhdanov, V. P. & Höök, F. Total Internal Reflection Fluorescence

Microscopy For Determination of Size of Individual Immobilized Vesicles: Theory and Experiment. J. Appl. Phys. 118, 64702 (2015). 25.

Muresan, A. S. & Lee, K. Y. C. Shape Evolution of Lipid Bilayer Patches Adsorbed

on Mica: an Atomic Force Microscopy Study. J. Phys. Chem. B 105, 852–855 (2001). 26.

Richter, R. P. & Brisson, A. R. Following the Formation of Supported Lipid Bilayers

on Mica: A Study Combining AFM, QCM-D, and Ellipsometry. Biophys. J. 88, 3422–3433 (2005).



27.

Larsen, J., Hatzakis, N. S. & Stamou, D. Observation of Inhomogeneity in the Lipid

Composition of Individual Nanoscale Liposomes. J. Am. Chem. Soc. 133, 10685–10687 (2011). 28.

Tabaei, S. R. & Cho, N. J. Lamellar Sheet Exfoliation of Single Lipid Vesicles by a

Membrane-Active Peptide. Chem. Commun. 51, 10272–10275 (2015). 29.

Richards, M. J. et al. Membrane Protein Mobility and Orientation Preserved in

Supported Bilayers Created Directly from Cell Plasma Membrane Blebs. Langmuir 32, 2963–2974 (2016).


Page 20 of 20

Spatio-Temporal Kinetics of Supported Lipid Bilayer Formation on

Recommend Documents