Preparation and Characterization of Solid-Supported Lipid Bilayers

She then received a Miller Research Fellowship and did her postdoctoral research at UC Berkeley and Lawrence Berkeley National Laboratory. In 1994, sh...
1 downloads 0 Views 2MB Size
Subscriber access provided by Kaohsiung Medical University

Invited Instructional Article

Preparation and Characterization of Solid Supported Lipid Bilayers Formed by Langmuir-Blodgett Deposition – a Tutorial James Kurniawan, Joao Francisco Ventrici de Souza, Amanda T. Dang, Gang-yu Liu, and Tonya L. Kuhl Langmuir, Just Accepted Manuscript • DOI: 10.1021/acs.langmuir.8b03504 • Publication Date (Web): 22 Nov 2018 Downloaded from http://pubs.acs.org on November 24, 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 55 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

Preparation and Characterization of Solid Supported Lipid Bilayers Formed by Langmuir-Blodgett Deposition – a Tutorial James Kurniawan1, João Francisco Ventrici de Souza2, Amanda T. Dang3, Gang-yu Liu2 and Tonya L. Kuhl1,3* 1Department

of Chemical Engineering, 2Department of Chemistry, and 3Department of Materials Science and Engineering, University of California, Davis, CA 95616, USA

ABSTRACT: The structure, phase behavior and properties of cellular membranes are derived from their composition, which includes phospholipids, sphingolipids, sterols, and proteins with various levels of glycosylation. Due to the intricate nature of cellular membranes, a plethora of in vitro studies have been carried out with model membrane systems that capture particular properties such as fluidity, permeability, or protein binding, but vastly simplify the membrane composition in order to focus in detail on a specialized property or function. Supported lipid bilayers (SLB) are widely used as archetypes for cellular membranes and this instructional review primarily focuses on the preparation and characterization of SLB systems formed by Langmuir deposition methods. Typical characterization methods, which take advantage of the planar orientation of SLBs, are illustrated and references that go into more depth are included. The instructional review is written so that non-experts can quickly gain an in-depth knowledge regarding the preparation and characterization of SLBs. In addition, this work goes beyond traditional instructional reviews to provide expert readers with new results that cover a wider range of SLB systems than those previously reported in the literature. The quality of an SLB is frequently not well described and details, such as topological defects, can influence the results and conclusions of an individual study. This paper quantifies and compares the quality of SLBs fabricated from a variety of gel and fluid compositions, in correlation with preparation

ACS Paragon Plus Environment

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 2 of 55

techniques and parameters, to generate general rules of thumb to guide construction of designed SLB systems. Introduction Cells, their organelles, and other functional volumes are compartmentalized by membranes.1 Cellular membranes themselves contain hundreds of different constituent molecules, including lipids, sterols, and proteins. These moieties interact laterally to create so-called lipid rafts where the majority of cellular signaling and transport are hypothesized to take place.2, 3, 4 The complexity of cellular membranes, due in large part to the enormous variety of chemical species present and their active function, makes their study challenging. In addition, cellular membranes are constantly in flux and dynamically respond to their local environment. Together, these properties make it daunting to tease out the temporal and spatial variation and even more difficult to correlate structure-function relationships. To make headway, various biomimetic or model membrane platforms have been developed. Though vastly simplified, model membranes can mimic essential physical and chemical properties of biological membranes such as membrane elasticity, fluidity, phase behavior, and can provide an appropriate environment for studying protein function.5 For example, membranes containing mixtures of saturated lipid with high melting point, unsaturated lipid with low melting point, and sterols can form a variety of coexisting phases and partitioning of different molecules into these phases has been of particular interest.6 In an earlier review, Castellana and Cremer7 described a number of lipid bilayer platforms that have been used as model membrane systems. These platforms allow for the study of a variety of processes. For example, freestanding black lipid membranes are used to study transport across the membrane, while phase behavior and membrane fluidity are frequently studied using giant unilamellar vesicles.8 Some of the model membrane platforms provide overlapping information,

2 ACS Paragon Plus Environment

Page 3 of 55 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

but the specific type of platform dictates the characterization techniques that can be used. Powerful surface sensitive techniques for measuring membrane interactions, adhesion and high-resolution scans of membrane topography can easily be performed using supported lipid bilayers (SLBs).9 Freestanding membranes such as unilamellar or multilamellar lipid vesicles enable studies free from the influence of the underlying inorganic support and are useful systems for micropipette mechanical measurements, small angle scattering, and various imaging microscopies, but are incompatible with most high resolution, nanoscopic characterization techniques. SLBs immobilized on clean, smooth, largely hydrophilic surfaces have emerged as a powerful platform to study biomimetic membranes, including nano to microscopic phase separation, lipid rafts and the influence or function of small molecules such as polypeptides and membrane proteins on the nanoscale. The underlying support stabilizes the membrane and such systems make excellent platforms for sensing applications like heterogeneous analytical assays for environmental monitoring, drug discovery and drug testing.10 However, defects in the SLB could influence the outcomes of individual studies. This instructional review focuses on solid-supported bilayer systems primarily formed by Langmuir-Blodgett (LB) deposition methods and is organized into three main sections. In the first, methods to prepare SLBs are explained in detail while the second part of the review covers commonly used characterization techniques of supported membrane systems. The last section of the paper characterizes a variety of SLB systems from the nano to macroscale to elucidate how lipid composition, phase state, deposition conditions, and substrate influences the resulting SLB. LB deposition conditions to create high quality, low defect SLBs are emphasized and validated by characterizing a plethora of SLBs deposited on various substrates to reveal the impact of surface roughness, lipid phase state, and deposition parameters.

3 ACS Paragon Plus Environment

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 55

Primary Methods in Preparation of Solid Supported Bilayers There are two main methods used to create solid supported bilayers: Langmuir trough deposition and vesicle fusion. These methods can be used separately or in conjunction to create symmetric or asymmetric membranes of controlled composition. Recently, a third method based on solvent spreading and/or spin coating has emerged as a rapid means to fabricate solid supported membranes and is briefly described. Descriptions of each deposition technique are provided below. Substrates and Langmuir-Blodgett (LB) Technique. The LB technique, which dates back to 1917, was pioneered by Irving Langmuir11 for compressing and depositing fatty acid monolayers from the air-water interface onto substrates. Katherine Blodgett12 then advanced the method to multilayers with repeated dipping of a substrate through the air-water interface. The LB technique has subsequently been used to measure thermodynamic properties of lipids and other surface active, water insoluble monolayer films. The trough, usually made with a low surface energy material such as Teflon, is filled with water or any other subphase such as physiological buffer solution. Widely used substrates include freshly cleaved mica (0001),13, 14 quartz,15, 16 borosilicate glass (microscope slides),15,

17

silicon wafers, and thin films of metal18,

19

or silicon dioxide

(SiO2).20 The quality of the transferred monolayer is greatly enhanced by using ultra-clean, hydrophilic substrates with low surface roughness which yield more well-packed, uniform SLBs. Most high resolution studies use mica or high quality oxidized silicon wafers because of their low root mean square roughness, 0.2 Å and 2-3 Å, respectively. Recently, a KOH treatment method that greatly reduces the roughness of glass coverslips has been described.21 To ensure substrate cleanliness, mica should be freshly cleaved right before use. We typically prepare our quartz and silicon wafer substrates through methodical coarse to refined cleaning steps: Sonication in acetone, sonication in isopropanol and then rinsing in copious MilliQ water to remove contaminants. We

4 ACS Paragon Plus Environment

Page 5 of 55 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

then sonicate the substrate in Hellmanex (Hellma Analytics), an alkaline soap for optical components, and again rinse with copious MilliQ water. The substrate is then dried using clean nitrogen gas. The clean, dry substrate is subsequently treated for at least 10 minutes with high intensity UV-ozone and used immediately. UV pen lights can also be used if the exposure time is extended to 30 minutes. After coarse solvent cleaning, another method to remove organic residue and hydroxylate the silica surfaces by increasing silanol groups is “piranha” treatment with a mixture of concentrated sulfuric acid and 30 % hydrogen peroxide. A common ratio is three or four parts acid to one part peroxide. However, care should be given as longer than 5 minute exposures can increase the surface roughness.22 Subsequently, the substrate is rinsed with copious MilliQ water and dried with clean nitrogen gas. The clean, hydrophilic substrates are immersed in the subphase before spreading lipids on the air-water interface in the trough. The lipid spreading solution is prepared in a solvent, such as chloroform, hexane, or a mixture of chloroform and methanol. The concentration of the spreading solution is typically between 0.1 to 1 mg/mL. The lipid must fully dissolve in the solvent, and solvent solubility in the trough subphase should be minimal. The choice of solvent and concentration is determined by the solubility of the lipids in the chosen solvent. For example, phosphatidylcholine (PC) lipids fully dissolve in chloroform at room temperature, but phosphatidylethanolamine (PE) lipids require a mixture of 9:1 chloroform:methanol for complete dissolution. The spreading solution is dispensed carefully, droplet by droplet, onto the air-water interface to create a thin layer of lipids whose hydrocarbon acyl chains anchor the molecules to the interface. For accuracy, an arbitrary solution concentration is typically chosen to achieve a deposited volume of 50 to 100 μL, depending on the surface area of the LB trough and desired compression of the film. Once the solvent has evaporated, a barrier compresses the lipid monolayer

5 ACS Paragon Plus Environment

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 55

to create a highly compressed two-dimensional film on the air-water interface at a pre-determined surface pressure. After reaching the target pressure and allowing the compressed film to equilibrate, the solid hydrophilic substrate is vertically drawn out of the water to deposit the inner monolayer. A diagram for bilayer deposition by LB/LB technique is shown in Figure 1A. Throughout the deposition process, the surface pressure is usually kept constant by compressing the remaining film as the transfer occurs. Once the substrate has cleared the water level, it is resubmerged to deposit the outer monolayer of the SLB. LB/LB deposition can yield a bilayer featuring a single component lipid or mixtures of lipids with few topological defects. The composition of the bilayer leaflets may be identical or different for the inner and outer monolayer (symmetric vs. asymmetric/hybrid bilayer). Some important parameters must be considered in order to achieve a high-quality solid supported membrane through the LB deposition technique. During deposition, the phase of the lipids greatly affects the resulting monolayer transfer onto the substrate. Lipid phase is dictated by lipid type, subphase temperature, and the film’s surface pressure (as measured by a Wilhelmy plate). Optimum deposition pressures are usually assessed through the plot of the surface pressure as a function of the area per lipid molecule (П-A isotherm) for the lipid or mixture being deposited. In general, high changes in surface pressure with area,

| |, correspond to better transfers to the 𝑑П 𝑑𝐴

substrate. LB transfer requires strong cohesion of the lipids which is greater at high pressures and rapid changes in

| |. Low values of | |, as found in regions of high fluidity or phase coexistence 𝑑П

𝑑П

𝑑𝐴

𝑑𝐴

where the pressure only changes modestly with area, correspond to weaker lipid cohesion and result in variable area per molecule and poor transfer efficacy. At very low surface pressures, little material is transferred from the air-water interface to the substrate.23

6 ACS Paragon Plus Environment

Page 7 of 55 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

∆𝐴𝑡𝑟𝑜𝑢𝑔ℎ

The transfer ratio, 𝑇𝑅 = 𝐴𝑠𝑢𝑏𝑠𝑡𝑟𝑎𝑡𝑒, is used to quantify the quality of the monolayer transfer from the air-water interface to the substrate. A value of TR = 1 is indicative that the surface pressure and, therefore, area per molecule were maintained from the air-water interface to the substrate.24, 25 Dipping speed must also be considered in order to achieve an efficient transfer of lipid from the air-water interface to the substrate. A relatively high transfer ratio cannot be obtained if the substrate moves too quickly through the lipid monolayer. In addition, a fast dipping speed sometimes causes delamination of the inner monolayer rather than deposition of the outer layer. Typically, a dipping speed between 0.1 to 5 mm/min is chosen in order to obtain a high transfer ratio (TR ~ 1.0). Details for parameters required to obtain an SLB with as few topological defects as possible are discussed in section 4, and Girard-Egrot and Blum provide an in-depth discussion of the LB technique.26 Langmuir-Schaefer (LS) Technique. Vincent Schaefer, in collaboration with Irving Langmuir, first deposited a monolayer of urease on metal and glass plates with a method similar to the LB technique.27 The LS technique, sometimes called horizontal deposition, deposits a lipid monolayer by stamping the substrate with a parallel orientation to the air-water interface through a compressed lipid monolayer (Figure 1B). Typically, the LS technique is used to deposit the outer leaflet of the membrane to create symmetric or asymmetric SLB with the inner monolayer deposited by LB. The LS technique should be used to create the SLB in cases where the physiochemical interaction of the inner layer to the substrate is insufficient to ensure stability. In other words, LS is particularly useful in situations where the inner monolayer can delaminate from the substrate during vertical LB deposition.28 Lipid phase state is also important for the LS technique. Gerelli et al. determined that fluid leaflets mixed during the LS deposition presumably due to mechanical shock.29 To maintain leaflet 7 ACS Paragon Plus Environment

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

asymmetry, which is frequently used to study lipid flip-flop rates,29,

Page 8 of 55

30, 31, 32

lipids should be

deposited in the gel phase. Dipping speed is not as crucial for achieving a quality bilayer using the LS method. However, the orientation (that is, how level the substrate surface is relative to the water surface) is critical for successful LS deposition. When the substrate is at an angle, film material can be pushed away, leading to a less well-packed transfer. A detailed comparison of SLBs fabricated with LB/LB technique versus LB/LS technique is provided in section 4. Vesicle Fusion (VF) Technique. Many SLB studies use the vesicle fusion method to prepare bilayers on substrates because of its ease and simplicity. To create a SLB, small unilamellar vesicles are incubated on a clean hydrophilic surface to create a symmetric bilayer (Figure 1D). Vesicles can also be incubated with an LB deposited monolayer of a different composition to make asymmetric membranes (Figure 1C).33 Vesicle fusion relies on the instability of vesicles interacting with the support and attractive interactions of the vesicles with the support (including already fused membrane regions) to yield spontaneous SLB formation.17,

34

Osmotic stress,

addition of salts and divalent ions, and temperature cycling can also be done to aid SLB formation, SLB coverage of the substrate, and break any absorbed, intact vesicles.35, 36, 37, 38, 39 However, the quality of SLBs deposited by VF or LB/VF is often more variable compared to LB/LB or LB/LS deposited bilayers (Figure 1E).

36, 37, 38, 39, 40VF

is less reproducible and may exhibit incomplete

surface coverage with holes or defects as well as adsorbed, non-fused vesicles..39, 40, 41, 42 In LB or LS techniques, the monolayer surface pressure is precisely controlled, which generally leads to fewer topological defects or variations in the SLB by comparison. Regarding high-resolution membrane structure, parity across techniques is possible. Watkins et al.24 established that 1,2dipalmitoyl-sn-glycero-3-phosphocholine (DPPC) membranes deposited by vesicle fusion mimic depositions by LB/LS technique at surface pressures of 38±3 mN/m.

8 ACS Paragon Plus Environment

Page 9 of 55 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

The first step in performing vesicle fusion is preparation of the vesicle solution. A small amount of concentrated lipid in organic solvent is prepared in a glass vial or similar container. The solvent is then evaporated by a stream of nitrogen while rotating the vial to leave a thin lipid layer on the walls of the container. The amount of lipid solution needed depends on the required volume of the vesicle solution. In general, the concentration of the vesicle suspension used for SLB formation is between 0.2 to 1 mg/ml. After drying with nitrogen, the vial is placed in a vacuum for at least 2 hours to fully remove all solvent before hydrating the lipids to create the vesicle suspension. Applying a vacuum is essential, as any solvent present upon addition of water will disturb vesicle formation. The dried lipids are then hydrated with water or buffer solution and the suspension is vortexed to obtain multilamellar vesicles of heterogeneous sizes. Freeze-thaw cycles can be used to break the multilamellar structures into unilamellar vesicles. Small unilamellar vesicles (SUVs, diameter ~100 nm) are preferable for vesicle fusion because they more readily absorb/rupture on the substrate to form an SLB. To make small, uniform SUVs, the vortexed vesicles are probe-tip sonicated (typically 1 – 2 minutes) or extruded through a polycarbonate membrane multiple times with the solution temperature above the lipid or lipid mixture melting point. Extrusion is the preferred method, as probe-tip sonication can release contaminant titanium particles into the vesicle solution, though these can be removed by centrifugation or an extra filtration step. The substrate is then incubated with the freshly prepared vesicle suspension for at least 20 minutes based on surface plasmon resonance and quartz crystal microbalance studies of the time required to reach saturation.43 Excess vesicles can be washed away after the incubation. Vesicle fusion usually results in a bilayer with many topological defects, adhered vesicles, and suboptimal surface coverage.

39, 40, 41

However, there are a number of ways to increase the

quality of vesicle fusion SLBs.17, 40, 44 Typical preparation methods include incubating a room

9 ACS Paragon Plus Environment

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 55

temperature substrate with a warm SUV solution ( in general the vesicle solution is always above the melting point of the lipids), freeze-thaw cycling, and rinsing with water at an elevated temperature or applying an osmotic shock by rinsing with a solution of a different salt concentration to obtain higher bilayer surface coverage. In particular, the incubation must be performed with a fresh SUV solution because small unilamellar vesicles are unstable and will readily fuse to the clean hydrophilic substrate to make symmetric a SLB or to a monolayer to create an asymmetric SLB.45, 46 Freeze-thaw cycling will rupture adsorbed vesicles through formation of ice crystals, so they can deform and cover the substrate.47 Recently, fusogenic peptides have been incorporated to improve vesicle fusion.48, 49 After incubation, excess and physisorbed vesicles are removed by exchanging the incubation solution with a vesicle-free buffer solution or water. If the rinsing solution has a different salt concentration than the vesicle solution, the concentration gradient creates osmotic flow that swells or shrinks and ruptures any excess vesicles stuck to the membrane or substrate and thus results in a more uniform supported membrane. This washing step is usually repeated a few times to ensure removal of excess vesicles. Care must be taken to ensure that the SLB is not exposed to air, which will disrupt the bilayer. After the SLB is deposited on the substrate, thermal cycling/annealing can be done to increase the surface coverage after incubation.38 However, the heating and cooling cycle must be done cautiously to prevent excessive loss of lipids into the subphase (Figure 2A-C). For comparison, the bottom panel (Figure 2D-F) shows higher resolution AFM topographs of SLBs of the same composition 2:2:1 DPPC:DOPC:cholesterol formed by vesicle fusion on mica. Spin Coating, Spreading Techniques, and Rapid Solvent Exchange. Spin coating and spreading are quick and easy techniques for deposition of solid supported lipid membranes. In spin coating, a lipid solution with concentration ranging between 0.25 and 5 mM in a volatile solvent that wets

10 ACS Paragon Plus Environment

Page 11 of 55 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

the substrate is deposited on a clean substrate. After spreading, the substrate is rapidly accelerated to a certain rotation speed (e.g. 2000 rpm for 2 min) to quickly remove the solvent leaving a thin dry lipid film. In the spreading technique, one microliter of lipid solution with a similar concentration range is deposited on a clean substrate and the solvent is allowed to evaporate.50, 51 Once a thin, dry lipid layer is formed on top of the substrate, the sample can then be partially or fully hydrated to create a stack of lipid membranes. Mennicke and Salditt52 described parameters for the spin coating process in detail, such as lipid solution concentration and rotation speed, in order to deposit multiple bilayers (up to 22 layers). Excess floating bilayers can be subsequently removed by a fluid jet to leave a single supported bilayer in the treated region.53 The fluid jet is typically water or buffer sprayed onto the substrate from a syringe or MilliQ water dispenser. While this technique can be used for solid supported membrane preparation, it is more suited for creating multiple bilayers on a substrate. This is due to the numerous topological defects and low stability of multilayers in bulk water as characterized by X-ray scattering and other methods.53, 54, 55

Rapid solvent exchange or solvent-assisted lipid bilayer (SALB) formation utilizes the phase behavior of lipid, water, isopropanol mixtures to drive the formation of the SLB.56 Although other miscible alcohols can be used (e.g. methanol, ethanol, n-propanol), the most uniform SLBs were formed using isopropanol.57 A range of lipid and isopropanol concentrations have been demonstrated to work, ranging from 0.5 mg/ml lipid in pure isopropanol followed by water/buffer exchange to 2 mg/ml lipid in a 50:50 volume solution of water:isopropanol followed by water/buffer exchange. The main requirement is during the incubation period, the lipids in the isopropanol-water mixture should be below the micelle-to-vesicle transition point. Upon increasing the relative water content, the critical micelle concentration of the lipid is lowered and

11 ACS Paragon Plus Environment

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 55

the lipid – substrate interaction drives the formation of the SLB. Similar lateral diffusivities were obtained for vesicle fusion and SALB formed SLBs, however care must be taken during water washing steps to ensure elimination of residual alcohol. A distinct advantage of the solvent exchange method is that no specialized equipment is required. In addition, the formation of SLBs with high cholesterol content can be obtained, which is especially challenging with VF methods.58 Characterization of Solid Supported Bilayers A solid supported membrane created by any of the methods described above can be characterized by a variety of techniques. Characterization techniques provide information on SLB properties, and the results from one technique can corroborate data collected by another method. Commonly used characterization techniques are described below. Fluorescence Microscopy (FM). Fluorescence microscopy is a basic but powerful technique that relies on doping in a small amount of fluorescent lipids (usually 0.25 to 1 mole%) to characterize solid supported membrane systems. White light is directed through a bandpass (excitation) filter that only allows light with a specific wavelength range to pass through and is directed to the sample using a dichroic mirror. To reduce noise and fluorescent contributions from off-target molecules in the sample, the wavelength range is selected to match the absorption wavelength of the fluorescent dopant. Illumination excites the dye, resulting in the release of light with a longer wavelength that passes back through the dichroic mirror and an emission filter into the detector. FM enables the lateral organization, including phase separation and domains of the SLB to be studied at the macro scale, since a given fluorescence dye will often partition preferentially into one phase more than the other. FM also permits the analysis of membrane fluidity through fluorescence recovery after photobleaching (FRAP). FRAP enables determination of the diffusion coefficient of the fluorescently labeled lipid, and the mobile and immobile fraction 12 ACS Paragon Plus Environment

Page 13 of 55 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

of the SLB. High quality SLBs have uniform diffusivities and low immobile fractions. Loren et al. and Rayan et al. have published recent reviews on FRAP.59, 60 In a typical measurement, a small region of the membrane is photo-bleached by exposing the region to a high intensity light source. The rate of fluorescence recovery in the bleached spot is then used to calculate the lipid diffusion coefficient while the recovered intensity compared to the initial intensity is used to determine the fraction of the SLB that is mobile. Figure 3 shows an exemplar FRAP measurement and fluorescence recovery time curve from an SLB system. The diffusivity is greatly impacted by the phase state of the membrane. Values typically range from on the order of 5𝜇𝑚2 𝑠 for fluid phase supported membranes, to 0.1𝜇𝑚2 𝑠 for liquid-ordered (Lo), phase supported membranes, to 10 ―3 𝜇𝑚2 𝑠 for gel phase supported membranes.15, 61 There are additional configurations of fluorescence microscopy that are used to characterize SLBs. For example, Fӧrster resonance energy transfer (FRET) can be used to observe microdomains in a lipid raft62 and the interaction between proteins in the membrane.63, 64 Total internal reflection fluorescence (TIRF) microscopy and two-photon excitation fluorescence microscopy (P2FM) can be utilized to study the directional alignment of the lipids in a gel phase domain.65,

66

Confocal fluorescence correlation microscopy, an

improvement upon traditional FM, increases the resolution of lateral diffusivity measurements and can produce a three-dimensional structural scan of the solid supported membrane if there are features protruding out of the membrane plane.67 The fluorescent dyes used in FM have several key characteristics, such as head group or acyl chain labeling, partitioning preference into more ordered or disordered phases, their excitation/emission spectrum, quantum yield, and lifetime, which must be carefully considered against the demands of the experiment. For example, FRET requires two fluorescence dyes such that the absorption spectrum of one dye overlaps the excitation spectrum of the other dye. FRAP

13 ACS Paragon Plus Environment

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 55

necessitates a dye with robust quantum yield and lifetime that is still easy to photobleach. Berezin and Achilefu qualified various fluorescence dyes in detail, including their spectra and lifetime.68 Online tools from fluorescent dye vendors also aid in determining the best dye for each specific use.69 One caution is that the presence of fluorescent probes can shift the miscibility and phase transitions in some systems.70, 71, 72 Atomic Force Microscopy. Atomic force microscopy(AFM) scans a sharp tip over samples and provides a powerful means to characterize solid supported bilayer systems due to its highresolution, label-free nature and versatility to work in various environments, including ambient, liquid and culture media.73, 74 Local membrane structure and properties have been clearly revealed by AFM including topography, domain size, layer thickness, friction, adhesion, and visco-elastic properties.38, 41, 75, 76, 77, 78 SLBs prepared with various deposition techniques described previously can be imaged using AFM under defined medium. Acquiring AFM images of monolayers in ambient or nitrogen environment is relatively straightforward. For imaging in water or liquid media, cautions are necessary when designing and constructing sample holders to minimize sample perturbation and maintain the SLB’s structural integrity.79 Two raster scan imaging modes are primarily used to characterize the structure and morphology of SLBs: tapping and contact mode.38, 80 Contact mode with soft cantilevers typically yields high-resolution images for gel phases, while tapping mode is necessary for liquid phases or layers with hydrophilic termini, which tend to have high adhesion to Si and Si3N4 AFM tips. In contact mode, the tip apex is in firm contact with the surface throughout the scan, while maintaining a set load.81 In this case, topographic and lateral force images are acquired simultaneously for both trace and retrace scan directions. The topographic images reveal the surface contour, while the lateral force images show local functionality, such as hydrophilic versus

14 ACS Paragon Plus Environment

Page 15 of 55 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

hydrophobic domains. In tapping mode, the cantilever tip is modulated at the resonance frequency of the cantilever, whose amplitude would decay as the tip approaches the surface. The control electronics maintains the tip-surface separation based on a set damping, e.g. 80% of initial amplitude. Typically, topographic and phase images are acquired simultaneously. The former reveals feature heights, while the latter is related to the tip-surface interaction, which again depends on the local functionality and mechanics. In addition to imaging, force versus distance measurements can be acquired to reveal tip surface interactions at chosen locations. This spectroscopic measurement is also known as a forcedeformation profile or curve, from which the repulsive and adhesive force between the tip and film can be extracted.82 With designed functionalization of the AFM tip, the adhesive force may be correlated to the hydrophilicity of the surface feature, DNA hybridization force, or specific interactions such as ligand-receptor binding.83, 84, 85 For AFM imaging of SLB systems presented in this work, tapping mode was frequently used at a speed of 20 m/s, 80% of initial amplitude (45 nm). Contact mode imaging was done at the same speed, but under a load of 6.2 nN. In both cases, the tip was an AC240 (Olympus, Japan) with spring constant of 1.7 N/m. Figure 4 shows an example AFM topograph of an SLB acquired using tapping mode. The SLB consisted of an inner monolayer of 1,2-dipalmitoyl-sn-glycero-3phosphoethanolamine (DPPE), and outer layer of 3:7 DPPE: 1,2-dioleoyl-sn-glycero-3phosphocholine (DOPC) lipids. At room temperature, DPPE (Tm = 63 °C) and DOPC (Tm = -20 °C) phase separate. The AFM topography image clearly shows phase separation in the SLB system. The gel phase DPPE domain in Figure 4 is about 1.5 nm taller than the surrounding DOPC fluid phase. The lowest imaging force should be used to avoid compression of the SLB.

15 ACS Paragon Plus Environment

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 55

Surface Force Apparatus (SFA). Various direct force-measuring techniques have been developed which enable measurement of the interaction between membranes of various geometries.83,

86

Using a variety of (interchangeable) force-measuring springs, the SFA can directly measure the full force profile between two SLBs in liquid at the angstrom (0.1 nm) level with a resolution of 10-8 N. The solution in the SFA can be exchanged in-situ to study SLB interactions under different conditions.14, 20, 87, 88 SFA characterization is very powerful because it is possible to unambiguously determine the interaction force profile, thickness and refractive index of the SLBs, as well as observe structural rearrangements when the SLBs are in contact.13, 88 Figure 5 shows example interaction force-distance profiles between liquid-ordered SLBs, which reveal nanoscopic structural rearrangement of the SLBs. A positive force indicates repulsion and a negative value indicates adhesion between the SLBs. The measured profile is the sum of any interactions present such as electrostatic, steric, van der Waals, depletion, hydration, and hydrophobic forces. By careful experimental design, the contribution of the various interactions can be separated and quantified. The substrate used in SFA is usually back-silvered, molecularly smooth mica that is glued onto a cylindrical disk (radius of ~1.5 cm). Once the mica is glued onto the disk, the desired SLB can be deposited, typically using the more controlled LB technique. After membrane deposition, the surfaces are transferred under water into the SFA and the solution is saturated with lipid of the same composition as the membrane to minimize desorption of lipids from the surfaces. The two surfaces are positioned in a cross-cylindrical configuration that is locally equivalent to a sphere near a flat surface or two spheres close together. White light is passed through the opposing surfaces and the emerging beam is focused onto the slit of a grating spectrometer. The silver layer on each disk partially transmits light directed normally through the surfaces, which constructively

16 ACS Paragon Plus Environment

Page 17 of 55 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

interferes and produces fringes of equal chromatic order (FECO). The distance between the two surfaces can be adjusted using a motor connected to the lower surface or piezo connected to the upper surface. The separation between the surfaces is measured by monitoring the wavelength of the FECO. The lower surface is supported on a double cantilever spring with a known spring constant (typically around 1 ― 2 × 105 𝑚𝑁/𝑚 for SLBs). Both repulsive and attractive forces can be measured and the force profile can be obtained over a large distance regime under quasi-static conditions. Once the force F as a function of distance D is measured for the two surfaces (of radius R), the adhesion or interfacial energy E per unit area can be calculated using the Derjaguin approximation: E = F/2R. Thus, for R ~ 1 cm, and given the measuring sensitivity in F of about 10-8 N, the sensitivity in measuring adhesion and interfacial energies is approximately10-2 mJ/m2 (erg/cm2). In addition to measuring interaction forces between SLBs,14, 20, 87, 88, 89, 90 the SFA has been used to probe the interaction between SLBs and substrates,91 hemifusion of SLBs,89, membrane mediated receptor-ligand interactions,93,

94

92

the refractive index and thickness of

supported monolayers and membranes,13, 14 and the interactions of adsorbed vesicle layers.95, 96 X-ray and Neutron Reflectivity. X-ray and neutron reflectivity are powerful surface-sensitive characterization techniques which provide information such as the thickness, density profile, and roughness down to the atomic scale of a substrate supported thin-film. Reflectivity, R, is defined as the intensity ratio of X-rays or neutrons (hereafter referred to as “particles”) elastically and specularly scattered from the surface relative to the incident particle beam. The reflectivity is measured as a function of the wave-vector transfer 𝑞𝑧 =

4𝜋𝑠𝑖𝑛𝜃 𝜆

perpendicular to the interface, where

𝜃 is the angle of the beam to the sample and 𝜆 is the wavelength of the particle beam. When

measured this way, the reflectivity curve contains information regarding the average scattering length density of the sample normal to the interface and can be used to determine the concentration 17 ACS Paragon Plus Environment

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 55

of atomic species at a particular depth in the film. Detailed descriptions of reflectivity measurements of SLBs can be found in the literature.97, 98, 99, 100, 101, 102, 103, 104, 105, 106 An example X-ray reflectivity profile from a DPPC SLB LB/LS deposited on quartz is shown in Figure 6A.16 The visible fringes in the reflectivity profile arise from interference between particles reflected from the membrane-solution interface and membrane-substrate interface. The amplitude of the fringes is due to the scattering length density (SLD) contrast between the substrate, lipid 2𝜋

headgroups, and acyl chains of the bilayer, while the fringe spacing Δ𝑞𝑧 ≈ 𝑑𝑙𝑎𝑦𝑒𝑟 is related to the thickness of the film (𝑑𝑙𝑎𝑦𝑒𝑟). From the measured reflectivity profile, the SLD, thicknesses, and

roughness of the various layers can be determined by modeling the expected SLD profile and iterating to minimize the difference between the measured reflectivity profile and that obtained from the modeled SLD profile. However, as the majority of reflectivity measurements only provide intensity information, the structural information of interest is indirectly contained within the reflectivity data. The transformation of the data from inverse space to real space, in the absence of phase information, is mildly ill-posed and multiple solutions can be obtained.107, 108 Limiting the possible solutions through constraints based on the known chemical identities of the layers, expected thickness, and other information is extremely helpful. The corresponding scattering length density profile of the DPPC bilayer fitted to the measured reflectivity profile is shown in Figure 6B. There are a number of key differences between neutron and X-ray scattering. Neutrons can penetrate large sample volumes and do not damage the sample as the beam is scattered by atomic nuclei. Selective contrast can be achieved by deuteration of lipids/proteins and solvent contrast variation.33 On the other hand, X-ray sources are much more brilliant. Because the beam is scattered by electrons in the sample, significant energy is deposited in the sample which can

18 ACS Paragon Plus Environment

Page 19 of 55 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

quickly degrade organic samples.109 Although contrast is limited to the electron density of the film, the high intensity of X-rays allows much more rapid and higher resolution measurements as shown in Figure 6A. The intensity also enables diffraction measurement from ordered monolayers and SLBs, yielding molecular details such as unit cell parameters, tilt angle, coupling between leaflets, and lateral coherence lengths from the off-specular diffraction signals (grazing incidence X-ray diffraction or GIXD).16, 24, 104 SLBs for neutron and X-ray reflectivity are prepared on smooth, ultra-polished singlecrystal substrates (e.g. quartz, oxidized silicon, sapphire) using LB-LS or vesicle fusion. The sample is mounted in a holder and the neutron/X-ray beam is reflected off the substrate through the SLB, which sits in a thin water layer. The scan area for neutron reflectivity is approximately 10-100 cm2; a significantly smaller area of about 0.1 cm2 can be used for X-ray reflectivity. The scattered signal and the quality of the data is dependent on multiple factors, such as interfacial roughness (quality of the substrate and SLB), contrast of the various layers, and the level of incoherent scattering from the bulk liquid. For the purposes of obtaining basic structural information about an SLB with high resolution, higher fluxes of the X-ray source coupled with faster measurement times make X-ray reflectivity generally superior to neutron reflectivity. On the other hand, neutron reflectivity can take advantage of deuteration to enable specialized measurements such as the flip-flop of lipids between the inner and outer leaflets of a bilayer.29 Details about neutron scattering characterization methods, including a discussion of advantages, limiting factors, recent works of SLB characterization, and extension of the use of neutron scattering on various platforms can be found in several recent publications.98, 110, 111 Quartz Crystal Microbalance (QCM). QCM can be used to track the quantity of absorbed mass on a solid with time. The quartz crystal oscillates with a frequency dependent on its mass. When films

19 ACS Paragon Plus Environment

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 55

are deposited/adsorb on the crystal, the frequency of the oscillation decreases and the change in the oscillation frequency can be correlated to a change in mass. The thickness of the deposited/absorbed layer can be calculated using an equation developed by Sauerbrey.112 In the case of SLBs, QCM measurements have been used to study the process of vesicle fusion as a function of time and are becoming a standard technique for optimizing the conditions for SLB formation.34, 113 In-depth studies of the mechanics of vesicle fusion with various lipid mixtures have been performed to determine parameters that affect vesicle adsorption kinetics upon various substrates.114, 115, 116 QCM can also be coupled with other techniques such as AFM and surface plasmon resonance (SPR) to visualize the various stages of SLB formation via vesicle fusion.117, 118

In addition to obtaining information on the mechanism of vesicle fusion, QCM can be used to

probe the interaction between various lipids with different types of substrates,119 and the interaction between deposited SLBs and proteins120 or nanoparticles.121 An in-depth discussion of the QCM technique, including its diverse uses extending beyond SLB systems, can be found in works by Cooper and coworkers.112, 122 Impact of Deposition Conditions on the Quality of Supported Lipid Bilayers In most cases, a clean, stable, and well-packed membrane provides a good platform for studies involving SLBs. Membrane topological defects, which span the outer monolayer or in some cases penetrate the inner monolayer, are typically present when SLBs are deposited with any commonly used preparation technique (hereafter referred to as membrane defects). In most cases, these defects are nanoscopic and not visible through optical microscopy, such as fluorescence or confocal microscopy.23, 92, 123, 124 These nanoscopic features can be resolved by high-resolution AFM imaging and can play an important role in altering the structure, properties, and interactions of the SLB. In this final section qualitative and quantitative comparisons of SLBs deposited under 20 ACS Paragon Plus Environment

Page 21 of 55 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

different conditions are provided and discussed to optimize procedures and conditions towards constructing continuous SLBs with closely packed molecules and few defects. As vesicle fusion SLBs are more variable, LB deposition methods are emphasized. Vesicle fusion can yield high quality SLBs, but the interplay between vesicle-substrate interaction, vesicle size, composition, substrate properties, and especially solution conditions, such as ionic strength and temperature all must be carefully considered and optimized to yield high quality SLBs. Moreover, complementary characterization techniques must be used to fully characterize the SLB. Namely, small defects and adsorbed/unfused vesicles can be difficult to detect. A number of papers have addressed these aspects.113, 125, 126 Further, many of the important experimental deposition parameters described for LB deposition in the next section also impact vesicle fusion SLB quality such as substrate roughness and lipid phase state, while other conditions such as surface pressure during deposition, packing properties of the inner monolayer leaflet, and other preparation conditions are controllable and explored via LB deposited films. These parameters are probed quantitatively by transfer ratio measurements conducted with LB deposition technique and high resolution AFM topographs to reveal the heterogeneity and coverage of the resulting SLB. The characterization of the SLBs moves from the macro/micro to nanoscale in the following subsections. Quantification of SLB quality by transfer ratio experiments and selection of deposition pressure. One of the most important Langmuir deposition parameters that dictates the quality of the resulting SLB is the surface pressure during deposition. The surface pressure-area (П-A) isotherm depends on the lipid mixture and temperature, and can be used to select the desired lipid packing (area per molecule) of the deposited monolayer. Typically, deposition using LB or LS technique is 𝑑П

conducted where the slope of the П-A isotherm curve is at its steepest (𝑑𝐴, largest change in the surface pressure with small shift in area per molecule), and the monolayer is still stable (showing 21 ACS Paragon Plus Environment

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 55

little loss in area with time prior to deposition).26 The surface pressure is held constant during the deposition by decreasing the area of the monolayer film as material is transferred from the airwater interface to the substrate. Because the change in pressure with area is small, monolayers cannot be transferred well at low surface pressure or in coexisting regions. For example, Bassereau and Pincet investigated the impact of deposition pressure in SLBs composed of inner leaflets of DMPE and outer leaflets of DOPC formed by LB/LB deposition on mica. When the outer DOPC monolayer was transferred at low surface pressures (𝜋 ≤ 25 𝑚𝑁/𝑚), holes were found in the SLBs. The size and surface coverage of the holes increased as the DOPC surface pressure decreased. Bassereau and Pincet were able to correlate the formation of holes with desorption of the inner DMPE monolayer from the substrate back to the air-water interface during transfer of the outer, poorly packed DOPC monolayer. The desorption of DMPE was further enhanced at slow dipping speeds. Similar results were found with symmetric SLBs composed of pure DOPC and DMPE, demonstrating the need for high surface pressure and high lateral cohesion of the lipids during LB deposition. Subsequently, Benz and coworkers investigated the morphology and interactions of SLBs as a function of deposition conditions of the outer monolayer.92 Their work clearly demonstrated that defects were present in the outer leaflet and that the density and size of the defects increased with decreasing deposition pressure. At ultra-low outer monolayer deposition pressures (4 mN/m), membrane spanning holes were observed due to rearrangement of the inner leaflet. As described in section 4.2, membrane spanning holes in SLBs are minimized by using well packed inner and outer leaflets deposited at high surface pressures, consistent with the requirement of strong lateral cohesion between lipids in the deposited monolayers for formation of high quality SLBs. Gel phase inner leaflets such as DPPE or DPPC also minimize the formation of membrane holes.

22 ACS Paragon Plus Environment

Page 23 of 55 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

Once the pressure for the deposition is selected, the quality of the deposited inner and outer monolayers can be quantified at the macroscale based on a transfer ratio measurement. The transfer ratio is defined as the ratio of the area of lipids removed from the air-water interface to the substrate area coated during the deposition (as described in section 2.1). In all cases here, transfers were done with the surface pressure held constant by computer control during the deposition. Deposition speeds were typically 1 mm/min. A systematic study of deposition speed on SLB transfer quality remains to be done. Table 1 summarizes the transfer ratios of various lipid mixtures deposited on different substrates with the LB technique. The impact of specific conditions on TR are highlighted below. Substrate roughness. To quantify the effect of substrate roughness on the deposition quality, a comparison of mica, silicon wafers, and microscope slides (borosilicate glass) was carried out. In general, the transfer of lipid monolayers onto glass slides or coverships resulted in higher transfer ratios than onto mica or silicon wafers. This is because of differences in substrate roughness. As measured by contact mode AFM, glass slides have a significantly larger surface roughness (root mean square roughness, RMS = 8-10 Å) compared to that of silicon wafers (3-4 Å) or mica (0.2 Å), creating a larger effective surface area on the glass. The transfer ratio of pure, fluid and transition phase lipids increased from about 95% surface coverage on mica to 100% or greater on glass. Similar observations were seen for mixed lipid systems containing cholesterol, where the transfer ratio increased from less than 90% to about 100%. The only cases where the transfer onto mica was greater were for gel phase DPPE and DPPC. This is likely due to the greater stiffness of the gel phase and lateral cohesion of the lipid monolayer, which prevents good conformity and physisorption onto rougher glass slides. Conversely, mica and silicon’s ultra-smooth surfaces are especially well suited for transfer of gel phase monolayers. When the monolayer contains

23 ACS Paragon Plus Environment

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 55

cholesterol, or exists at fluid or transition phases, it can conform to the roughness of the glass substrate and result in a higher transfer ratio. However, as shown through high resolution AFM topography scans in section 4.2, the SLB is of lower quality on rougher substrates. We also examined the transfer ratio of various lipid mixtures onto DPPE and DPPC monolayers on mica, glass slides, oxidized silicon, and quartz. However, the inner monolayer always delaminated when attempting to LB deposit the outer leaflet onto SiO2 surfaces, reinforcing the need for the LS method with these substrates.28 Unfortunately, measuring TR with LS is very challenging due to disruption of the remaining film during transfer. As a result, outer layer transfer ratios were only done using mica substrates. Lipid phase state – impact of deposition pressure. The lipid monolayer phase state and substrate roughness influenced the transfer ratio of the inner leaflet as described above. Here, we compare how the phase state of the inner leaflet effected the transfer ratio of the outer leaflet. At room temperature, the transfer ratio of fluid onto fluid (DTPC on DTPC) and transition phase onto transition phase (DMPC on DMPC) were below 80%. However, the transfer of gel onto gel (DPPC on DPPC), gel onto fluid (DPPC on DTPC), and fluid onto gel (DTPC on DPPC) were significantly higher. Interleaflet mixing and desorption of some of the deposited lipid in the inner monolayer disrupts the outer layer transfer in more fluid or transition phase depositions.23, 29 When at least one of the layers is in the gel phase, interleaflet mixing is diminished and higher transfer ratios are obtained. In general, gel phase lipids provided a good base for LB deposition of outer monolayers of various phases. As detailed in section 4.2, gel phase base monolayers have very few defects resulting in more complete SLB formation. We also found the outer monolayer transfer was slightly enhanced when the inner leaflet was dried by comparing transfer ratios onto freshly deposited DPPE inner monolayers and monolayers that were dried overnight (relative humidity