Electrochemically Triggered Release of Reagent to ... - ACS Publications

Jun 14, 2017 - selectively to the proximal leaflet of a supported lipid bilayer is described. Selective delivery is achieved by creating a spanning li...
0 downloads 0 Views 3MB Size
Download PDF
Article pubs.acs.org/Langmuir

Electrochemically Triggered Release of Reagent to the Proximal Leaflet of a Microcavity Supported Lipid Bilayer H. Basit, S. Maher, R. J. Forster, and T. E. Keyes* School of Chemical Sciences, National Centre for Sensors Research, Dublin City University, Dublin 9, Ireland S Supporting Information *

ABSTRACT: A novel and versatile approach to electrichemically triggering the release of a reagent, β-cyclodextrin (β-CD), selectively to the proximal leaflet of a supported lipid bilayer is described. Selective delivery is achieved by creating a spanning lipid bilayer across a microcavity array and exploiting the irreversible redox disassembly of the host−guest complex formed between thiolated ferrocene (Fc) and β-cyclodextrin (β-CD) in the presence of chloride. Self-assembled monolayers of the ferrocene−alkanethiols were formed regioselectively on the interior surface of highly ordered 2.8 μm cavities while the exterior top surface of the array was blocked with a monolayer of mercaptoethanol. The Fc monolayers were complexed with β-CD or β-CD-conjugated to streptavidin (β-CD-SA). Phospholipid bilayers were then assembled across the array via combined Langmuir−Blodgett/vesicle fusion leading to a spanning bilayer suspended across the aqueous filled microcavities. Upon application of a positive potential, ferrocene is oxidized to ferrocinium cation, disrupting the inclusion complex and leading to the release of the β-CD into the microcavity solution where it diffuses to the lower leaflet of the suspended bilayer. Disassembly of the supramolecular complex within the cavities and binding of the β-CD-SA to a biotinylated bilayer was followed by voltammetry and impedance spectroscopy where it caused a large increase in membrane resistance. For unmodified β-CD, the extraction of cholesterol from a cholesterol containing bilayer was evident in a decrease in the bilayer resistance. For the first time, this direct approach to targeted delivery of a reagent to the proximal layer of a lipid bilayer offers the potential to build models of bidirectional signaling (inside-out vs outside-in) in cell membrane model systems.

■

1.5 nm thick.18 The available volume is not dramatically improved on cushioned or tethered bilayers, and modification of the surface to release reagent may affect the functioning of the bilayer cushion or tether. In liposomes, because of inaccessibility of the interior of the structure, there is no simple way of accomplishing the release of reagent to the interior leaflet of the bilayer especially in a triggered way that allows the dose to be controlled. Whereas some black lipid membranes (BLMs) on open apertures can be addressed from both sides,19−21 there are a number of drawbacks to BLMS as lipid models, including poor control over lipid layer thickness, solvent residues, stability, and they lack the compositional versatility of SLBs.22 Also, the time taken between adding a reagent into the system and its arrival at the membrane can be long and hard to precisely control. An alternative model membrane approach which is being increasingly validated is to span lipid bilayers across enclosed nano- to micrometer pores within arrays.19,23,24 We recently described how such an approach can be used to support lipid bilayers spanned across aqueous filled cavity arrays of

INTRODUCTION Model lipid membranes are important tools for the study of a wide range of membrane-associated biological processes from cell−membrane interactions to lipid diffusion dynamics, protein−ligand binding, and ion permeation.1 The unit functions of biological membranes can be studied in exquisite detail using in vitro models without the complexity of the cellular environment.2−6 However, a major challenge in the area remains the ability to selectively and independently address each leaflet of a lipid bilayer membrane which is extremely important in real cell membranes as biochemical processes such as signaling occur bilaterally at both the cytoplasmic and extracellular sides of the cell membrane.7 The inside-out and outside-in signaling of integrin is an important example of such bilateral signal transduction in biology.8 In the most commonly applied lipid membrane models such as supported lipid membranes and related cushioned9−13 and tethered lipid bilayers,14−17 or in liposomes, mimicking bidirectional biochemical signaling through reagent release to both lipid leaflets is extremely challenging. In SLBs and related models there is insufficient volume between the bilayer and lipid to accommodate reagent; for example, in a glass supported SLB the aqueous interface intervening between the solid substrate and the proximal leaflet of the lipid bilayer is between 0.5 and © XXXX American Chemical Society

Received: March 29, 2017 Revised: June 7, 2017 Published: June 14, 2017 A

DOI: 10.1021/acs.langmuir.7b01069 Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir micrometer-scale dimensions, giving free-standing lipid bilayers with lateral fluidity similar to lipid vesicles.22 We demonstrated Langmuir−Blodgett deposition combined with vesicle fusion provides a versatile approach to spanning lipid bilayers across aqueous filled microcavity arrays prepared from PDMS or from gold.23,25 The use of an aqueous filled truncated spherical cavity as a support for lipid bilayers leads to lipid bilayers structures that have stability approaching that of SLBs but can be used to support membrane proteins or ionophores in an environment in which the fluidity of both lipid and protein components are conserved and similar to those of free-standing liposomes. This is enabled by the aqueous reservoir which is deep relative to the bilayer thickness and effectively decouples the lipids and other membrane components from the substrate. Furthermore, because it is possible to fill the microwell independently of the exterior solution, establishing gradients, (e.g., ionic, pH, etc.) across the membrane is facile.23 The aqueous filled cavities also offer the possibility of addressing both sides of the suspended bilayer by providing a deep well into which reagent can be incorporated. Furthermore, in the case of gold substrates, we have described how such arrays can be selectively modified and used as electrodes.23,26 Combined, these elements offer the prospect of electrochemically inducing release of reagent from the cavity surface into the aqueous volume to allow for spatiotemporally controlled release of reagent to the proximal leaflet of the lipid bilayer. We have previously reported on the electrochemically induced release of fibrinogen from the interior of a microcavity array (in the absence of bilayer) via reductive desorption.23 This mode of release though was impractical for biological applications because the released protein was denatured, most likely due to its adsorption directly on the gold substrate prior to its release and possibly also because of the high reducing potentials needed to induce desorption. The approach we report in the present work is illustrated in Scheme 1. To avoid direct interaction of fragile biomolecules with the substrate surface, we exploit the electrochemically reversible 1:1 host−guest complex of ferrocene or bis(cyclopentadienyl)iron(II) (Fc) with β-cyclodextrin (β-CD) as an intermediary.26−28 A self-assembled monolayer of ferrocene− ferrocenethiol was selectively formed on the interior interface of the cavities, onto which β-CD was allowed to complex. On oxidation of the ferrocene to ferrocinium cation, the host− guest complex is irreversibly disassembled in the presence of chloride due to the charge formed, disrupting the hydrophobic interactions between β-CD and Fc.29−31 In forming the Fc monolayers selectively at the interior of the cavity surface beneath a suspended lipid bilayer, on electrochemical dissociation of the host−guest association, β-CD is released into the solution in contact with the proximal leaflet of the bilayer. As illustrated in Scheme 1, we demonstrate electro-release using streptavidin modified β-CD to a DOPC bilayer containing DOPE labeled biotin. This release of reagent from the electrode was triggered by voltammetry, and the attachment to the bilayer was followed by monitoring membrane resistance changes via electrochemical impedance spectroscopy. We believe this facile approach to releasing reagent to a supported bilayer will be of use across a range of biophysical studies particularly given the demonstrated value of the cavity array approach in supporting membrane proteins.

Scheme 1. Schematic Illustrating the Ferrocenethiol−βCyclodextrin Modified Gold Microcavity Array with Supported Lipid Bilayer Reconstituted with Biotin-Labeled DOPEa

a The cavity arrays are selectively modified at the top surface with mercaptoethanol (ME) and comodified with ME and Fc at the cavity interior using sphere templating. The Fc forms an inclusion complex with streptavidin (blue) conjugated β-CD. Bottom: after application of potential of +0.4 V vs Ag/AgCl, the ferrocene is oxidized, leading to irreversible release of the streptavidin modified β-CD from the now cationic ferrocene monolayer permitting it to diffuse to and bind at the biotin (red) modified bilayer interface. The DOPE-biotin is present at both bilayer leaflets but is shown here only at the inner leaflet for clarity.

■

EXPERIMENTAL SECTION

Materials. Silicon wafers coated with a 100 nm layer of gold on a 50 Å layer of titanium were obtained from AMS Biotechnology Inc. The 2.94 μm diameter polystyrene spheres were obtained from Bangs Laboratories Inc. 1,2-Dioleoyl-sn-glycero-3-phosphocholine (DOPC), 1,2-dioleoyl-sn-glycero-3-phosphoethanolamine-N-Biotinyl (DOPEBiotin), 1,2-dioleoyl-sn-glycero-3-phospho-L-serine (DOPS), and cholesterol were all obtained from Avanti Polar Lipids Inc. Gold electrodeposition solution (TG-25RTU) was obtained from Technic Inc. All other reagents were obtained from Sigma-Aldrich and used as obtained. Gold Microcavity Array Preparation. Gold microcavity arrays were prepared using a previously described method.25,32,33 Silicon wafers coated with a 100 nm layer of gold on a 50 Å layer of titanium were cut into 2 cm × 1 cm pieces and cleaned with acetone, ethanol, and finally water and dried with nitrogen. A drop of 100 μL of a 0.5% (w/v) solution of 2.88 μm diameter polystyrene spheres was placed onto one end of the wafers and allowed to dry at room temperature overnight. Gold was then electrochemically deposited onto the wafers using a three-electrode cell with a silver/silver chloride reference electrode, a platinum counter electrode, and the gold-coated silicon wafer as the working electrode. The electrolyte used was a commercially available gold solution. The working electrode was held at a potential of −0.95 V (vs Ag/AgCl) using a CH660A potentiostat (CH Instruments) until a charge of 0.7 C had passed. The array was then removed from the electrolyte, rinsed with water, and dried under a gentle stream of nitrogen. Coupling of Streptavidin to Succinyl β-Cyclodextrin. Streptavidin was coupled to succinyl β-cyclodextrin via EDC coupling. 100 μL of a 0.62 mg solution of succinyl β-cyclodextrin dissolved in 20 mM HEPES buffer at pH 8.5 was added dropwise to 0.32 mg of EDC dissolved in 10 μL of HEPES pH 8.5. The reaction was stirred for 30 min, after which a 1 mg/mL solution of streptavidin was added dropwise to the reaction solution. The coupling reaction was allowed to stir overnight, resulting in streptavidin coupled to β-cyclodextrin (βCD-SA). The CD-SA was recovered by Sephedex column and B

DOI: 10.1021/acs.langmuir.7b01069 Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir

Figure 1. (A) Cyclic voltammetry of a gold microcavity array modified across the entire surface with mercaptoethanol monolayer (black). Cyclic voltammetry of the same electrode after its incubation with 1 mM ferrocene−hexanethiol for 24 h (red). (B) Cyclic voltammetry of microcavity array treated with mercaptoethanol on the top surface and ferrocene−hexanethiol inside the cavities. The measurements were performed in Tris NaCl buffer at pH 7.4 using a standard three-electrode setup. The potential was scanned from 0 to 0.7 V (vs Ag/AgCl reference) at 100 mV/s. coupling confirmed by UV and fluorescence spectroscopy (Supporting Information Figure S2). Selective Surface Modification of the Microcavity Array. Prior to removal of the templating spheres, the cavity arrays were placed in a 1 mM ethanol solution of 2-mercaptoethanol overnight to selectively modify the top surface of the cavities with mercaptoethanol. The arrays were then placed in THF and sonicated for 30 min to remove the polystyrene spheres. The arrays were then rinsed in water and placed overnight in an ethanol solution containing 1 mM solution containing 80 mol % 2-mercaptoethanol and 20 mol % ferrocene− hexanethiol to selectively modify the interior surface of the cavities. Following incubation, the arrays were rinsed with ethanol and dried under a stream of nitrogen. The cavities where then incubated in an aqueous solution containing1 mg/mL solution of β-CD for 30 min to allow the Fc/β-CD complex to form. Vesicle Preparation. Lipids from a chloroform solution, at an initial concentration of 50 mg/mL, were dried with nitrogen and then further dried under vacuum for 30 min. The lipids were then resuspended in 1 mL of Tris buffer containing 20 mM tris(hydroxymethyl)aminomethane and 150 mM NaCl at pH 7.4. The vesicles were then extruded through a 100 nm polycarbonate membrane 11 times to give unilamellar vesicles of 100 nm in diameter. Unless otherwise specified, vesicles were composed of DOPC:DOPE-Biotin (9:1 mol %). Supported Lipid Bilayer Preparation. Supported lipid bilayers where deposited across the SAM modified and aqueous filled gold microcavity arrays using a combination of the Langmuir−Blodgett (LB) and vesicle fusion techniques as described previously.23 Lipids from a 50 mg/mL chloroform solution were deposited onto the air− water interface of a NIMA 102D Langmuir−Blodgett trough. The chloroform was allowed to evaporate, after which the lipids were compressed to a surface pressure of 32 mN/m. This surface pressure ensures the formation of a compact monolayer of lipids on the air water interface The array was then immersed in the LB trough at high speed to ensure no lipid transfer at the hydrophilic interface and then slowly withdrawn at a rate of 5 mm/s while maintaining a surface pressure of 32 mN/m to deposit the lower leaflet of the bilayer. The array was then removed from the trough and a 1 mL solution of the preformed unilamellar vesicles was kept in contact with the cavity surface for about 10 min to complete the formation of a bilayer. Any remaining vesicles were washed away with buffer, and the bilayer coated arrays were kept in contact with buffer at all times. Electrochemical Impedance Spectroscopy. Electrochemical impedance spectroscopy was performed using a CH660A potentiostat (CH Instruments). A standard three-electrode setup was employed which consisted of an Ag/AgCl (1 M KCl) reference electrode, a platinum counter electrode, and the microcavity array used as the working electrode. The electrochemical impedance was measured over a frequency range of 100 000 to 0.01 Hz with an ac modulation amplitude of 0.01 V at a potential bias of 0 V (vs Ag/AgCl). All

measurements were carried out in a glass cell (approximate volume of 15 mL) in Tris NaCl buffer at pH 7.4.

■

RESULTS AND DISCUSSION Formation and Functionalization Gold Microcavity Arrays. Gold microcavity array electrodes were formed by microsphere lithography as described previously.25,33,32 Briefly, silicon wafers coated with 100 nm layer of gold were cut into 2 cm × 1 cm pieces, cleaned with acetone, ethanol, and finally water, and then dried with nitrogen. 150 μL of a 0.5% (w/v) solution of 2.88 μm diameter polystyrene (PS) spheres was placed onto one end of the wafers and allowed to dry for 24 h at room temperature. Gold was then electrochemically deposited through the templating spheres and onto the wafers by holding the electrode at a potential of −0.95 V (vs Ag/ AgCl) until a charge of 0.7 C had passed at the surface. SEM images of the resulting gold microcavity arrays indicate that under these conditions 2.88 μm diameter pores are prepared, with deposition occurring to a depth of 50% of the sphere diameter (see Figure S1). The gold microcavity arrays were selectively functionalized, using a modification of a stepwise sphere masking protocol reported previously by us.26 This was achieved by submerging the substrates in a 1 mM solution of mercaptoethanol (ME) prior to the removal of the polystyrene (PS) templating spheres. As described previously, the PS spheres prevent the ME SAM forming at the pore interior, limiting the SAM to the interstitial planar regions at the top surface of the array between pores.26,34 The ME deposition step was necessary, as described previously, to form a stable bilayer at the aqueous filled pores. The PS spheres were then removed by sonicating the cavities in THF for 30 min. As described previously, this step does not remove the SAM from the top surface.32 The substrate was then placed into an ethanol solution containing both ferrocene−hexanethiol (Fc) and ME (20:80 mol %) for 24 h. This resulted in a mixed monolayer of these reagents formed selectively at the cavity interior as the ME blocks the top surface. To confirm that the selective modification of the top surface of the electrode with mercaptoethanol (ME) was effective at preventing Fc-thiol from adsorbing at this surface, the cyclic voltammetry of gold array electrodes, modified with ME across the entire substrate, was compared with arrays selectively ME decorated at the top surface after each had been incubated for 24 h with Fc. The CVs were conducted in contact with Tris NaCl at a scan rate of 100 mV/s. C

DOI: 10.1021/acs.langmuir.7b01069 Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir

Figure 2. (A) Voltammetry of release of β-CD bound to Fc from the microcavity array showing the first and second scans of an ME top surface modified and Fc/ME modified interior after treatment with β-CD. (B) Overlay of current normalized voltammograms of ME/Fc and ME/Fc/β-CD modified electrodes. The measurements were performed in Tris NaCl buffer at pH 7.4 using a standard three-electrode setup. The scan was run from 0 to 0.7 V (vs Ag/AgCl reference) at 100 mV/s. Two consecutive scans were run on the same electrode; the first scan is shown above in black and the second in red.

Figure 1A shows the voltammetry of a 1 cm2 gold microcavity electrode modified across its entire surface with an ME monolayer (black) and the voltammogram of the same electrode after its incubation in Fc-thiol for 24 h (red). The electrodes are washed prior to electrochemistry to ensure that no unbound Fc remains. Significantly, the latter (red) shows a weak and irreversible oxidation peak at approximately 300 mV. This feature is attributed to the oxidation of a small quantity of Fc that adsorbs at the blocked electrode after its premodification with ME which occurs due to thiol exchange which has been reported before.35 However, it is evident from the results shown here that this is a slow process which results in minimal Fc absorption at the electrode with a preformed ME SAM. This result indicates that Fc adsorption at the top surface of the array will be minimal compared to Fc adsorption at unmodified bare cavity. The irreversible behavior of the Fc oxidation in the presence of chloride contrasts with the electrochemically reversible (vide supra) behavior observed in perchlorate electrolyte and arises due to the presence of chloride ion as described below. Figure 1A shows the voltammetry for the same dimensioned electrodes after incubation in Fc for 24 h following prior modification of only the top surface of the array (i.e., where the cavity surfaces are unmodified by ME). The voltammetric peak associated with Fc (Figure 1B, red) is far better resolved in this voltamogram, and the area under the voltammetric peak is approximately 4 times greater than observed for cavities where the entire surface had been modified with ME (Figure 1A). The presence of a weak Fc peak for the entirely ME modified surface indicates that although not 100% efficient, ME is highly effective at blocking the surface of gold preventing the Fc thiol from forming a uniform monolayer on the electrode surface. On the basis of this data, we can conclude that it is possible to selectively modify arrays where ME blocks the top surface so as to isolate the majority of the electrochemically active Fc monolayer within the cavity interiors. Ferrocene was used as an electrochemically addressable switch to trigger reagent release in this study since its oxidation reaction occurs at a relatively low potential in aqueous media, thus avoiding any potential damage to the lipid bilayer. The use of a mixed Fc/ME monolayer at the cavity interior surface is key to controlling the disassembly of the host−guest complex. Previous work has shown that a mixed monolayer system is needed to ensure efficient oxidation of the Fc metal center.36,37 It was reported that for ferrocene−hexanethiol SAMs the Fc

concentration in the deposition solution should be maintained below 20% of the overall monolayer concentration so as to inhibit lateral Coulombic interaction between neighboring oxidized and reduced Fc pairs during oxidation as such interactions force the Fc oxidation to higher potential.36 Correspondingly, here, to reduce such lateral interactions the Fc-thiol monolayer is diluted with mercaptoethanol ME (20:80 mol %). The success of this approach is reflected in Figure 1B which shows the cyclic voltammetry of the gold microcavity arrays modified with an ME monolayer at the top surface and with a mixed ME/Fc monolayer coating the interior walls of the cavities. An anodic voltammetric peak is observed at approximately 350 mV and is attributed to the oxidation of ferrocene localized at the cavity interior. The associated reduction process, is observed at 200 mV vs Ag/AgCl, but the integrated area of this peak is 85% lower than that of the associated oxidation peak. This indicates that the ferrocene redox reaction is chemically irreversible under the conditions used in this study. Furthermore, the peak-to-peak separation at 120 mV indicates irreversible, non-Nernstian behavior which was further confirmed by running voltammograms at varying scan rates which shows that the peak potential shifts with scan rate (see Figure S3). On a subsequent scan, the relative area of the oxidation peak diminishes further with no subsequent changes observed beyond these first three scans. (see Figure 1B). The irreversible voltammetric behavior of Fc has been reported previously and is attributed to the formation of an ion pair between Fc+ and Cl−, which prevents ferrocinium rereduction.38,39 To confirm this mechanism, we carried out a CV of the selectively modified Fc pore array in contact with sodium perchlorate electrolyte. Correspondingly, the voltammetry of the Fc center is fully reversible in the absence of chloride (Figure S4). Irreversible oxidation is essential in the context of the present application since it allows the potential to be scanned and the functionalized CD released into solution without the need to maintain the potential at a positive value to block reduction and consequent reassembly of the CD ferrocene monolayer complex. Thus, all measurements in this work were performed in Tris NaCl buffer. As previously described, β-cyclodextrin (β-CD) spontaneously forms a stable inclusion complex with ferrocene due to the hydrophobic interaction between the ferrocene and β-CD interior.28,40,41 Upon oxidation, ferrocene forms ferrocenium cation which due to its charge and increased hydrophilicity cannot form an inclusion complex with β-CD. We thus D

DOI: 10.1021/acs.langmuir.7b01069 Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir

Figure 3. (A) Nyquist plot of mercaptoethanol/ferrocene treated microcavity array before (black) and after incubation in 100 μL of β-CD (Red). (B) Nyquist plot of mercaptoethanol/ferrocene treated microcavity after incubation in 100 μL of β-CD (black) and after electrochemical release of β-CD (red). Measurements were performed at 0 V using a frequency range of 100 000 to 0.01 Hz in Tris NaCl buffer at pH 7.4 using a standard three-electrode setup consisting of a platinum counter, Ag/AgCl reference, and the microcavity array as the working electrode.

the cavities and its release from the interior of the cavities to the cavity solution. Figure 3A shows EIS data obtained from cavity arrays modified with ME (top surface) and Fc/ME mixed monolayer (cavity interior), both before (red circles) and after (black boxes) complexation of β-CD. It is evident from the Nyquist plots in Figure 3 that there is a significant increase in impedance upon introduction of β-CD, which is consistent with the association of the β-CD with the ferrocene monolayer at the cavity array. The EI spectra were fit to an equivalent circuit model (ECM) shown in Scheme 2, which was described

assembled the Fc-β-CD complex at the ferrocene mixed monolayer at the cavity walls and exploited the irreversible oxidation of Fc in the presence of chloride to induce disassembly of the Fc-β-CD layer under applied potential by oxidation of Fc.29 Representative cyclic voltammograms of a gold cavity array, which is selectively modified with ferrocene in the cavity interior followed by complexation with β-CD, are shown in Figure 2A. The initial scan in Figure 2A shows the ferrocene oxidation process as a peak at approximately 370 mV (black curve). The potential at which Fc oxidizes is shifted positive by 35 mV upon inclusion of β-CD compared to that observed for a ferrocene monolayer alone. This shift agrees with previous reports and confirms that the inclusion complex has formed.42 The associated reductive process is observed at 150 mV vs Ag/AgCl and is similar to the CV recorded for the Fc monolayer in the absence ofβ-CD. As before, the integrated peak area for this peak is 75% lower than the associated oxidation process due to the quasi-reversible oxidation of ferrocene in the presence of chloride. Upon a second scan (red curve), the Fc oxidation peak potential shifts from 370 to about 300 mV, and the associated reductive process is almost absent. The shift in the oxidation potential to lower values indicates that the β-CD does not reassociate with the Fc once it has been released upon oxidation. To confirm this, in Figure 2B we show the voltammograms recorded for an electrode selectively modified with Fc-thiol in the cavity interiors (black) overlaid with that for an electrode selectively modified with Fc-thiol complexed with β-CD in the cavity interiors (red). These results indicate the Fc-β-CD complex is irreversibly disrupted when the ferrocene is oxidized in the presence of chloride ion, and therefore the β-CD is released from the SAM under these conditions. To confirm the stability of the cavity Fc/β-CD interfacial assembly (Figure S5), CVs of Fc-treated cavities before and after β-CD association were performed in contact with blank aqueous sodium perchlorate. As mentioned previously in this electrolyte, the oxidation of the Fc is fully reversible. A characteristic shift in the peak potentials of approximately 30 mV is observed when the Fc/β-CD complex has formed.42 Further scans were performed at hourly intervals over an 8 h window and showed no changes to peak potentials or peak current, indicating that under these conditions the interfacial complex is stable. Electrochemical impedance spectroscopy (EIS) was used to follow both the binding of β-CD to ferrocene in the interior of

Scheme 2. Equivalent Circuit Model Used To Fit Electrochemical Impedance Spectroscopy Dataa

a

Rsol = electrolyte solution resistance, Rm = resistance of deposited layer on the electrode surface, Cm = capacitance of deposited layer, Rarray = the resistance of the cavity arrays, and Cdl = double-layer capacitance.

previously by us for the microcavity supported lipid bilayers.25 The ECM consists of the electrolyte solution resistance (Rsol) in series with the resistance and capacitance of the deposited layer on the electrode surface in parallel with each other (Rm, Cm). The circuit also contains a component for the resistance of the cavity arrays, Rarray, in parallel to the double-layer capacitance, Cdl. Constant phase elements (CPE) are used instead of pure capacitors to account for surface defects on both electrode surface, which may cause it to stray from purely capacitive behavior. The impedance of a CPE is given by ZCPE = Q−1(jω)−α, where Q is the magnitude of the capacitance of the CPE, ω is the angular frequency, and α is a real number between 1 and 0 (the closer α gets to 1 the more ideal the capacitive behavior of the CPE). A good agreement between model and experimental data is seen by the overlap of fitted EIS data (solid lines) on the experimental data (squares or circles). In blank electrolye, consistent with the voltammetric data, the EIS signal was stable in the absence of appied potential, but E

DOI: 10.1021/acs.langmuir.7b01069 Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir

a streptavidin conjugated CD was assembled at the Fc monolayer with a biotinylated bilayer suspended across the cavity array. While Fc/β-CD is electrochemically dissociable, we exploited the high affinity interaction between streptavidin and biotin whose association constant at a biotinylated bilayer is approximately 107 to follow β-CD release and binding at the bilayer by electrochemical impedance. Succinyl-β-cyclodextrin was coupled to streptavidin via EDC coupling to form β-CD-SA (S1). The gold microcavity electrodes were prepared as before, with ME selectively assembled at the top surface and a mixed monolayer of ME/ Fc at the cavity interiors. The electrodes were then incubated in β-CD-SA at 1 mg/mL for 30 min at room temperature to allow β-CD-Fc assembly, after which point the electrodes were rinsed with Tris HCl buffer to eliminate any unbound β-CD. A bilayer composed of DOPC/DOPE-biotin (90:10 mol %) was then suspended over the aqueous buffer filled microcavities using a combination of Langmuir−Blodgett and fusion of vesicles as described previously.23 The EIS signal from the array was stable in the absence of applied potential over the experimental window, indicating that although the association constants of the β-CDSA-Fc and β-CD-SA-biotin are expected to differ by a factor of approximately 100, the spontaneous dissociation of β-CDSA-Fc leading to an equilbirium in favor of β-CD-SA-biotin at the cavity supported bilayer system is slow relative to the experimental time scale, possibly due to the spatial isolation of the biotinylated lipid at the cavity aperture. The β-CD-SA was released as before, by scanning the potential of the electrode from 0 to 0.7 V at a scan rate of 100 mV for three cycles. The concentration of SA released was estimated from the release CV and found to be 0.55 nM (see Supporting Information). Impedance measurements carried out on the Me/FC-β-CDSA bearing cavity arrays, both before and after conversion of the Fc to Fc+, are shown in Figure 4. Figure 4A shows representative Nyquist plots obtained before and after release of β-CD-SA to a DOPC bilayer containing 10 mol % DOPEbiotin. Upon fitting the data to the equivalent circuit described above (Table 2), it is clear that the resistance of the deposited membrane increases upon release of β-CD-SA. This increase suggests that the β-CD-SA released from electrode is binding to the bilayer. The increase in resistance is accompanied by a slight decrease in film capacitance. To confirm the binding of the released β-CD-SA to the bilayer, we carried out a control experiment wherein a DOPC bilayer was suspended across the

as shown in Figure 3, the impedance of the ME/Fc-β-CD modified electrode changes significantly following application of a sufficiently positive potential to oxidize the ferrocene and dissociate the β-CD:Fc complex. Figure 3 shows the representative Nyquist plot following release of β-CD from the electrode surface, after scanning from 0 to 0.7 V at a scan rate of 100 mV/s. A pronounced decrease in impedance is observed upon application of potential (red circles) compared to the impedance recorded before application of potential, consistent with β-CD removal from the surface once Fc is oxidized. This is confirmed by the resistance and capacitance values obtained by fitting the data to the ECM described above. As seen in Table 1, the resistance of the membrane is observed to Table 1. EIS Values for the Release of β-CD from the Electrode Surfacea sample

Rm (MΩ cm2)

Cm (μF/cm2)

α

ME/Fc modified cavities β-CD-SA modified after release of β-CD-SA

4.94 ± 0.26 5.58 ± 0.87 3.73 ± 0.49

1.21 ± 0.04 1.13 ± 0.25 1.35 ± 0.34

0.93 ± 0.01 0.92 ± 0.01 0.93 ± 0.01

a

Values shown are the average value obtained from measuring three separate electrodes.

increase slightly upon addition of β-CD to a ME/Fc, whereas the capacitance shows a decrease, consistent with the addition of a layer of β-CD. However, a pronounced decrease in resistance is observed of about 1.8 MΩ cm2 , and a corresponding increase in capacitance (by about 0.22 μF/ cm2) upon release of β-CD from the Fc- β-CD complex bound to the electrode. Intriguingly, the values of resistance and capacitance upon release of β-CD do not return to those measured before the assembly of β-CD, i.e., the values of the ME/Fc modified cavities. This we attribute to the quasi-irreversible nature of the ferrocene oxidation arising from formation of the charged ferrocinium cation leading to a surface charge with lower resistance and higher capacitance than the ME/Fc surface bearing neutral ferrocene. Electrochemical Release of β-CD-SA to a Biotinylated Lipid Bilayer and Detection of Binding to the Bilayer. To investigate if the strategy of electrochemically releasing reagent (β-CD) from the surface into solution can be exploited to release reagent to the lower leaflet of a suspended lipid bilayer,

Figure 4. Nyquist plot of microcavity array with a DOPC; DOPE-biotin (90; 10 mol %) bilayer before (red) and after (black) the electrochemical release of β-CD-SA. Measurements were performed at 0 V using a frequency range of 100 000 to 0.01 Hz in Tris NaCl buffer at pH 7.4 using a standard three-electrode setup consisting of a platinum counter an Ag/AgCl reference and the microcavity array as the working electrode. F

DOI: 10.1021/acs.langmuir.7b01069 Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir Table 2. EIS Data for the Release of β-CD from the Electrode Surface to a DOPC:DOPE-Biotin and to a DOPC Bilayera sample

Rm (MΩ cm2)

Cm (μF/cm2)

α

DOPC; DOPE-biotin bilayer after release of β-CD-SA DOPC bilayer after release of β-CD-SA

9.42 ± 0.26

1.84 ± 0.17

0.92 ± 0.005

11.90 ± 0.78 6.57 ± 1.34 3.88 ± 0.24

1.74 ± 0.10 1.36 ± 0.06 1.64 ± 0.06

0.92 ± 0.004 0.93 ± 0.001 0.91 ± 0.003

of LB and vesicle fusion as described previously on the cavity arrays bearing ME/Fc in the interior of the cavities. The β-CD was electrochemically released by oxidizing the Fc in the Me/ Fc monolayer at the gold microcavity interface, and the EIS was measured before and after the voltammetric release of the β-CD from the cavity interior surfaces; the spectra were fit to the equivalent circuit described in Scheme 2. Figure 5 shows representative time-dependent impedance data for the β-CD release.

a

Values shown are the average value obtained from measuring three separate electrodes.

array without any DOPE-Biotin (Figure 4B). It is clear from the data and from the fit using the model (Table 2) that in the absence of biotin β-CD-SA release consistently induces a decrease in the average impedance across all three replicate experiments/substrates. The contrasting decrease in the resistance is attributed to the loss of material (the CD) from the interior of the cavity surface and its dissolution in the cavity buffer solution without its associated binding to the lipid membrane. Although β-CD is known in some cases to bind to and extract lipids from lipid bilayers,43 Ohvo and Slothe have shown that a solution of 1.4 mM β-CD did not extract any DPPC lipids.44 β-CD may nonetheless contribute to the observed reduction in resistance observed as it may induce changes the packing of the bilayer even in the absence of biotin. Such β-CD−lipid interactions are likely prevented in the presence of the biotinylated layer due to the very high affinity SA has for biotin, which essentially traps the β-CD at these sites. Electrochemical Release of β-Cyclodextrin to Cholesterol Containing Bilayers. The results obtained in the preceding section indicate that the approach of releasing material from gold microcavities to a suspended lipid bilayer using ferrocene/β-cyclodextrin host guest complex is promising for spatially selective modification of separate bilayer leaflets. Furthermore, in addition to introducing reagent to a bilayer, we explored the application of the β-CD release mechanism to use it as a means of altering lipid bilayer composition selectively from the lower leaflet. To examine this possibility, we explored the impact of the electrochemical release of β-CD from the pores of cavity arrays suspended with a lipid bilayer containing cholesterol. β-CD is well-known to bind to cholesterol with a binding constant45 of 1.7 × 104 M−1 and is used to remove cholesterol from the plasma membrane of some cells.46 In the case of the latter, the degree of depletion is dependent on the derivative of β-CD used, concentration, temperature, incubation time, and cell type. Methyl-β-cyclodextrins (M-β-CD) have been shown to be the most efficient at removing cholesterol. At concentrations of 5−10 mM they have been shown to deplete 80−90% of the total cholesterol from cellular membranes within approximately 2 h.47 M-β-CD has also been shown to form holes in lipid bilayers composed of DOPC and sphingomyelin (SM) which is thought to be due to the removal of lipids from the bilayer.48 Although lipid affinity is typically considerably lower for β-CD,49 it was shown that βCD has a stronger affinity for cholesterol than DPPC and SM, whereas α-CD has a stronger affinity for DPPC than SM while γ-CD has equal affinity for all three. A lipid bilayer composed of DOPC:DOPS:cholesterol (80:10:10 mol) in both leaflets was formed using a combination

Figure 5. Nyquist plot of β-CD released to bilayer containing cholesterol suspended over gold microcavities. Impedance was measured before and after release of β-CD to the cavities and after release of β-CD from the cavities. All measurements were performed in Tris NaCl buffer using a platinum counter electrode, Ag/AgCl reference, and the gold cavities as the working electrode.

Notably, the average (n = 3) bilayer resistance was determined as 16.65 ± 1.38 MΩ cm2 for the DOPC:DOPS:cholesterol (80:10:10 mol) bilayers, which is nearly 3 times the value for DOPC alone. This results demonstrates the sensitivity of the approach to this analytical bilayer structure. Cholesterol has a condensing effect on the lipid bilayer as it induces elongation of lipid chains, the area occupied per lipid decreases, and the membrane thickens.50 Consequently, the film is considerably more capacitive than a bilayer made from DOPC alone with an average membrane capacitance of 5.87 ± 0.42 μF/cm2.51 Such effects have been noted previously in SLBs.52 The β-CD was released, as before, by scanning the CV from 0 to +0.70 V at a scan rate of 100 mV/s for three consecutive scans. The EIS of the bilayer was then measured 15, 30, and 60 min after the release (Figure 5). As seen from the fitted EIS data in Table 3, the impedance of the cholesterol containing film decreases significantly by approximately 60% from 16.65 ± 1.38 to 6.19 ± 0.92 MΩ cm2 over the first 15 min. A further change of almost 10% from 6.19 ± 0.92 to 5.71 ± 0.97 MΩ cm2 was observed over the remaining hour. This can be attributed to the release of β-CD from the electrode surface and may also be due to some removal of cholesterol from the lipid bilayer. This behavior contrasts with an analogous control experiment where β-CD was released to a DOPC-only bilayer for which EIS was measured over the same time scale; the data are shown in Figure 6. As with the cholesterol containing bilayer, the EIS data show an initial decrease in the membrane resistance of approximately 55% from 6.92 ± 0.47 to 2.96 ± 0.09 MΩ cm2 in the first 15 min after release (see Table 3), which can be attributed to the release of β-CD from the cavity G

DOI: 10.1021/acs.langmuir.7b01069 Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir Table 3. EIS Data for the Release of β-Cyclodextrin from the Electrode Surface to a DOPC:DOPE-Biotin and to a DOPC Bilayera sample

Rm (MΩ cm2)

Cm (μF/cm2)

α

DOPC; DOPS; cholesterol bilayer 15 min after release 30 min after release 1 h after release DOPC bilayer 15 min after release 30 min after release 1 h after release

16.65 ± 1.38

2.87 ± 0.42

0.94 ± 0.03

6.19 6.57 5.71 6.92 2.96 3.76 3.79

± ± ± ± ± ± ±

0.92 1.59 0.97 0.47 0.09 0.10 0.05

2.33 1.94 1.91 1.44 1.82 1.65 1.64

± ± ± ± ± ± ±

1.16 1.09 1.04 0.23 0.25 0.24 0.26

0.91 0.92 0.92 0.96 0.94 0.95 0.95

± ± ± ± ± ± ±

The ability of the host−guest complex to release material to a lipid bilayer was demonstrated by functionalizing the β-CD with streptavidin (β-CD-SA) and voltametrically releasing this species to a bilayer containing biotin. Electrochemical impedance spectroscopy allowed facile monitoring of the process and showed a clear increase in the membrane resistance after the release of β-CD-SA to the bilayer which is attributed to the binding of SD to the biotin. In a second demonstration EIS was used to study the release of β-CD to a bilayer containing 10% mol/mol cholesterol. β-CD release caused an irreversible decrease in the resistance which is attributed to the removal of cholesterol from the bilayer. The advantages of this approach are that it can be performed in a buffer solution at sufficiently low potential that it does not affect the bilayer integrity. Furthermore, as release of reagent is indirect, mediated through a ferrocene oxidation, it does not require direct adsorption of the reagent onto the electrode which can be potentially damaging in the case, for example, of protein release. The drawback of the systems presented here are that they are based on release of noncovalently self-assembled structures and rely on equilibria between associated and dissociated components. Future work may focus on covalent or electrostatically assembled components for clean switching of release. Nonetheless, the examples presented here indicate that spatiotemporal release of reagent to the proximal leaflet of a bilayer is possible and offers the opportunity to address both sides of a lipid bilayer independently. As the distal or external leaflet is readily addressable through the external contacting solution, this method in principle permits bidirectional signaling to lipid or protein reconstituted into the bilayer and opens the possibility for mimicking such bidirectional signaling which is important in cell function.

0.04 0.03 0.03 0.01 0.01 0.01 0.01

a

Values shown are the average values obtained from measuring three separate electrodes.

■

Figure 6. Nyquist plot of β-CD released to a DOPC bilayer over gold microcavities. Impedance was measured before and after release of βCD to the cavities and after release of β-CD from the cavities. All measurements were performed in Tris NaCl buffer using a platinum counter electrode, Ag/AgCl reference, and the gold cavities as the working electrode.

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.langmuir.7b01069.

■

surface. However, the resistance did not decrease further beyond this time but slightly increased to 3.79 ± 0.05 MΩ cm2 which may be due to some re-formation of the Fc-β-CD complex on the interior of the cavities.

Figures S1−S6 (PDF)

AUTHOR INFORMATION

Corresponding Author

■

*E-mail [email protected] (T.E.K.).

CONCLUSIONS This work establishes a simple and novel method for the controlled release of material to the proximal leaflet of a supported lipid bilayer. The method relies on the use of microcavity supported lipid bilayers which provides for stable bilayers with an addressable interface in contact with an aqueous volume below the bilayer. The release mechanism is based on an established supramolecular interaction where an interfacial ferrocene/β-cyclodextrin host−guest complex is dissociated on oxidation of the ferrocene guest which forms a self-assembled monolayer selectively at the interior of the gold cavities. The formation and disruption of this supramolecular complex at the interior of the cavity was followed by voltammetry and electrochemical impedance spectroscopy in the absence of a bilayer which confirmed the irreversible nature of the ferrocene/β-cyclodextrin dissociation in the presence of Cl− ions that prevent reassembly of the host−guest complex.

ORCID

T. E. Keyes: 0000-0002-4604-5533 Present Address

H.B.: Chemistry Research Laboratory, University of Oxford, 12 Mansfield Road, OX1 3TA, Oxford, United Kingdom. Author Contributions

H.B. and S.M. contributed equally. Funding

This material is based upon work supported by the Science Foundation Ireland under Grant No. s 10/CE/B1821, and 14/ IA/2488 and the National Biophotonics and Imaging Platform, Ireland, funded by the Irish Programme for Research in Third Level Institutions, Cycle 4, Ireland’s EU Structural Funds Programmes 2007−2013. Notes

The authors declare no competing financial interest. H

DOI: 10.1021/acs.langmuir.7b01069 Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir

■

(20) Schmitt, E. K.; Vrouenraets, M.; Steinem, C. Channel Activity of OmpF Monitored in Nano-BLMs. Biophys. J. 2006, 91 (6), 2163− 2171. (21) Grime, J. M.; Edwards, M. A.; Rudd, N. C.; Unwin, P. R. Quantitative Visualization of Passive Transport across Bilayer Lipid Membranes. Proc. Natl. Acad. Sci. U. S. A. 2008, 105 (38), 14277− 14282. (22) Hladky, S. B.; Gruen, D. W. R. Thickness Fluctuations in Black Lipid Membranes. Biophys. J. 1982, 38, 251−258. (23) Basit, H.; Gaul, V.; Maher, S.; Forster, R. J.; Keyes, T. E. Aqueous-Filled Polymer Microcavity Arrays: Versatile and Stable Lipid Bilayer Platforms Offering High Lateral Mobility to Incorporated Membrane Proteins. Analyst 2015, 140 (9), 3012−3018. (24) Khan, M. S.; Dosoky, N. S.; Berdiev, B. K.; Williams, J. D. Electrochemical Impedance Spectroscopy for Black Lipid Membranes Fused with Channel Protein Supported on Solid-State Nanopore. Eur. Biophys. J. 2016, 45, 843−852. (25) Maher, S.; Basit, H.; Forster, R. J.; Keyes, T. E. Micron Dimensioned Cavity Array Supported Lipid Bilayers for the Electrochemical Investigation of Ionophore Activity. Bioelectrochemistry 2016, 112, 16−23. (26) Mallon, C. T.; Jose, B.; Forster, R. J.; Keyes, T. E. Protein Nanopatterning and Release from Gold Nano-Cavity Arrays. Chem. Commun. 2010, 46 (1), 106−108. (27) Harada, A.; Takahashi, S. Preparation and Properties of Cyclodextrin−ferrocene Inclusion Complexes. J. Chem. Soc., Chem. Commun. 1984, 645−646. (28) Wu, J.-S.; Toda, K.; Tanaka, A.; Sanemasa, I. Association Constants of Ferrocene with Cyclodextrins in Aqueous Medium Determined by Solubility Measurements of Ferrocene. Bull. Chem. Soc. Jpn. 1998, 71 (7), 1615−1618. (29) Frasconi, M.; D’Annibale, A.; Favero, G.; Mazzei, F.; Santucci, R.; Ferri, T. Ferrocenyl Alkanethiols−Thio β-Cyclodextrin Mixed SelfAssembled Monolayers: Evidence of Ferrocene Electron Shuttling Through the β-Cyclodextrin Cavity. Langmuir 2009, 25 (22), 12937− 12944. (30) Chmurski, K.; Temeriusz, A.; Bilewicz, R. β-Cyclodextrin-Based Ferrocene-Imprinted Gold Electrodes. Anal. Chem. 2003, 75 (21), 5687−5691. (31) Rojas, M. T.; Koeniger, R.; Stoddart, J. F.; Kaifer, A. E. Supported Monolayers Containing Preformed Binding Sites. Synthesis and Interfacial Binding Properties of a Thiolated.beta.-Cyclodextrin Derivative. J. Am. Chem. Soc. 1995, 117 (1), 336−343. (32) Jose, B.; Mallon, C. T.; Forster, R. J.; Blackledge, C.; Keyes, T. E. Lipid Bilayer Assembly at a Gold Nanocavity Array. Chem. Commun. 2011, 47 (46), 12530. (33) Bartlett, P. N.; Baumberg, J. J.; Birkin, P. R.; Ghanem, M. A.; Netti, M. C. Highly Ordered Macroporous Gold and Platinum Films Formed by Electrochemical Deposition through Templates Assembled from Submicron Diameter Monodisperse Polystyrene Spheres. Chem. Mater. 2002, 14 (5), 2199−2208. (34) Adamson, K.; Spain, E.; Prendergast, U.; Moran, N.; Forster, R. J.; Keyes, T. E. Peptide-Mediated Platelet Capture at Gold Micropore Arrays. ACS Appl. Mater. Interfaces 2016, 8 (47), 32189−32201. (35) Hostetler, M. J.; Templeton, A. C.; Murray, R. W. Dynamics of Place-Exchange Reactions on Monolayer-Protected Gold Cluster Molecules. Langmuir 1999, 15 (11), 3782−3789. (36) Auletta, T.; van Veggel, F. C. J. M.; Reinhoudt, D. N. SelfAssembled Monolayers on Gold of Ferrocene-Terminated Thiols and Hydroxyalkanethiols. Langmuir 2002, 18 (4), 1288−1293. (37) Chidsey, C. E. D.; Bertozzi, C. R.; Putvinski, T. M.; Mujsce, A. M. Coadsorption of Ferrocene-Terminated and Unsubstituted Alkanethiols on Gold: Electroactive Self-Assembled Monolayers. J. Am. Chem. Soc. 1990, 112 (11), 4301−4306. (38) Uosaki, K.; Sato, Y.; Kita, H. Electrochemical Characteristics of a Gold Electrode Modified with a Self-Assembled Monolayer of Ferrocenylalkanethiols. Langmuir 1991, 7 (7), 1510−1514. (39) Mertins, O.; Buriez, O.; Labbé, E.; Fang, P.-P.; Hillard, E.; Vessières, A.; Jaouen, G.; Tian, Z.-Q.; Amatore, C. Further Insights

REFERENCES

(1) Groves, J. T.; Dustin, M. L. Supported Planar Bilayers in Studies on Immune Cell Adhesion and Communication. J. Immunol. Methods 2003, 278 (1−2), 19−32. (2) Benda, A.; Beneš, M.; Marecek, V.; Lhotskỳ, A.; Hermens, W. T.; Hof, M. How to Determine Diffusion Coefficients in Planar Phospholipid Systems by Confocal Fluorescence Correlation Spectroscopy. Langmuir 2003, 19 (10), 4120−4126. (3) Seu, K. J.; Pandey, A. P.; Haque, F.; Proctor, E. A.; Ribbe, A. E.; Hovis, J. S. Effect of Surface Treatment on Diffusion and Domain Formation in Supported Lipid Bilayers. Biophys. J. 2007, 92 (7), 2445−2450. (4) Filippov, A.; Orädd, G.; Lindblom, G. Sphingomyelin Structure Influences the Lateral Diffusion and Raft Formation in Lipid Bilayers. Biophys. J. 2006, 90 (6), 2086−2092. (5) Meadows, K. E.; Nadappuram, B. P.; Unwin, P. R. A New Approach for the Fabrication of Microscale Lipid Bilayers at Glass Pipets: Application to Quantitative Passive Permeation Visualization. Soft Matter 2014, 10 (42), 8433−8441. (6) Gritsch, S.; Nollert, P.; Jahnig, F.; Sackmann, E. Impedance Spectroscopy of Porin and Gramicidin Pores Reconstituted into Supported Lipid Bilayers on Indium-Tin-Oxide Electrodes. Langmuir 1998, 14, 3118−3125. (7) Grecco, H. E.; Schmick, M.; Bastiaens, P. I. H. Signaling from the Living Plasma Membrane. Cell 2011, 144 (6), 897−909. (8) Hu, P.; Luo, B.-H. Integrin Bi-Directional Signaling across the Plasma Membrane. J. Cell. Physiol. 2013, 228 (2), 306−312. (9) Karatekin, E.; Rothman, J. E. Fusion of Single Proteoliposomes with Planar, Cushioned Bilayers in Microfluidic Flow Cells. Nat. Protoc. 2012, 7 (5), 903−920. (10) Kaufmann, S.; Ilg, K.; Mashaghi, A.; Textor, M.; Priem, B.; Aebi, M.; Reimhult, E. Supported Lipopolysaccharide Bilayers. Langmuir 2012, 28 (33), 12199−12208. (11) Wong, J. Y.; Majewski, J.; Seitz, M.; Park, C. K.; Israelachvili, J. N.; Smith, G. S. Polymer-Cushioned Bilayers. I. A Structural Study of Various Preparation Methods Using Neutron Reflectometry. Biophys. J. 1999, 77 (3), 1445−1457. (12) Baumgart, T.; Offenhäusser, A. Polysaccharide-Supported Planar Bilayer Lipid Model Membranes †. Langmuir 2003, 19 (5), 1730− 1737. (13) Diaz, A. J.; Albertorio, F.; Daniel, S.; Cremer, P. S. Double Cushions Preserve Transmembrane Protein Mobility in Supported Bilayer Systems. Langmuir 2008, 24 (13), 6820−6826. (14) Atanasov, V.; Knorr, N.; Duran, R. S.; Ingebrandt, S.; Offenhäusser, A.; Knoll, W.; Köper, I. Membrane on a Chip: A Functional Tethered Lipid Bilayer Membrane on Silicon Oxide Surfaces. Biophys. J. 2005, 89 (3), 1780−1788. (15) Gupta, G.; Staggs, K.; Mohite, A. D.; Baldwin, J. K.; Iyer, S.; Mukundan, R.; Misra, A.; Antoniou, A.; Dattelbaum, A. M. Fluid and Resistive Tethered Lipid Membranes on Nanoporous Substrates. J. Phys. Chem. B 2015, 119 (40), 12868−12876. (16) Basit, H.; Van der Heyden, A.; Gondran, C.; Nysten, B.; Dumy, P.; Labbé, P. Tethered Bilayer Lipid Membranes on Mixed SelfAssembled Monolayers of a Novel Anchoring Thiol: Impact of the Anchoring Thiol Density on Bilayer Formation. Langmuir 2011, 27 (23), 14317−14328. (17) Becucci, L.; D’Amico, M.; Daniele, S.; Olivotto, M.; Pozzi, A.; Guidelli, R. A Metal-Supported Biomimetic Micromembrane Allowing the Recording of Single-Channel Activity and of Impedance Spectra of Membrane Proteins. Bioelectrochemistry 2010, 78 (2), 176−180. (18) Kim, J.; Kim, G.; Cremer, P. S. Investigations of Water Structure at the Solid/Liquid Interface in the Presence of Supported Lipid Bilayers by Vibrational Sum Frequency Spectroscopy. Langmuir 2001, 17 (23), 7255−7260. (19) Steinem, C.; Janshoff, A.; Galla, H.-J.; Sieber, M. Impedance Analysis of Ion Transport through Gramicidin Channels Incorporated in Solid Supported Lipid Bilayers. Bioelectrochem. Bioenerg. 1997, 42, 213−220. I

DOI: 10.1021/acs.langmuir.7b01069 Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir into Hydrophobic Interactions between Ferrocenyl-Tamoxifen Drugs and Non-Polar Molecular Architectures at Electrode Surfaces. J. Electroanal. Chem. 2009, 635 (1), 13−19. (40) Bertrand, G. L.; Faulkner, J. R.; Han, S. M.; Armstrong, D. W. Substituent Effects on the Binding of Phenols to Cyclodextrins in Aqueous Solution. J. Phys. Chem. 1989, 93 (18), 6863−6867. (41) Matsue, T.; Evans, D. H.; Osa, T.; Kobayashi, N. ElectronTransfer Reactions Associated with Host-Guest Complexation. Oxidation of Ferrocenecarboxylic Acid in the Presence of.beta.Cyclodextrin. J. Am. Chem. Soc. 1985, 107 (12), 3411−3417. (42) Huangxian, J.; Zong, D.; Leech, D. Electrochemical Determination of Electroinactive Guests of β-Cyclodextrin at a Self-Assembled Monolayer Interface. Sci. China, Ser. B: Chem. 2002, 45 (1), 46−53. (43) Jose, B.; Steffen, R.; Neugebauer, U.; Sheridan, E.; Marthi, R.; Forster, R. J.; Keyes, T. E. Emission Enhancement within Gold Spherical Nanocavity Arrays. Phys. Chem. Chem. Phys. 2009, 11 (46), 10923. (44) Ohvo, H.; Slotte, J. P. Cyclodextrin-Mediated Removal of Sterols from Monolayers Effects of Sterol Structure and Phospholipids on Desorption. Biochemistry 1996, 35, 8018−8024. (45) Breslow, R.; Zhang, B. Cholesterol Recognition and Binding by Cyclodextrin Dimers. J. Am. Chem. Soc. 1996, 118 (35), 8495−8496. (46) Cellular Cholesterol Efflux Mediated by Cyclodextrins. (47) Levitan, I.; Christian, A. E.; Tulenko, T. N.; Rothblat, G. H. Membrane Cholesterol Content Modulates Activation of VolumeRegulated Anion Current in Bovine Endothelial Cells. J. Gen. Physiol. 2000, 115 (4), 405−416. (48) Giocondi, M.-C.; Milhiet, P. E.; Dosset, P.; Le Grimellec, C. Use of Cyclodextrin for AFM Monitoring of Model Raft Formation. Biophys. J. 2004, 86, 861−869. (49) Irie, T.; Fukunaga, K.; Pitha, J. Hydroxypropylcyclodextrins in Parenteral Use. I: Lipid Dissolution and Effects on Lipid Transfers in Vitro. J. Pharm. Sci. 1992, 81 (6), 521−523. (50) Hung, W.-C.; Lee, M.-T.; Chen, F.-Y.; Huang, H. W. The Condensing Effect of Cholesterol in Lipid Bilayers. Biophys. J. 2007, 92 (11), 3960−3967. (51) Nezil, F. A.; Bloom, M. Combined Influence of Cholesterol and Synthetic Amphiphillic Peptides upon Bilayer Thickness in Model Membranes. Biophys. J. 1992, 61, 1176−1183. (52) Wiegand, G.; Arribas-Layton, N.; Hillebrandt, H.; Sackmann, E.; Wagner, P. Electrical Properties of Supported Lipid Bilayer Membranes. J. Phys. Chem. B 2002, 106 (16), 4245−4254. (53) Nguyen, T. T.; Sly, K. L.; Conboy, J. C. Comparison of the Energetics of Avidin, Streptavidin, NeutrAvidin, and Anti-Biotin Antibody Binding to Biotinylated Lipid Bilayer Examined by Second-Harmonic Generation. Anal. Chem. 2012, 84 (1), 201−208.

J

DOI: 10.1021/acs.langmuir.7b01069 Langmuir XXXX, XXX, XXX−XXX