Robust Photoelectric Biomolecular Switch at a Microcavity-Supported Lipid Bilayer

Biomolecular devices based on photo-responsive proteins have been widely proposed for medical, electrical, and energy storage and production applications. Also, bacteriorhodopsin (bR) has been extensively applied in such prospective devices as a robust photo addressable proton pump. As it is a membrane protein, in principle, it should function most efficiently when reconstituted into a fully fluid lipid bilayer, but in many model membranes, lateral fluidity of the membrane and protein is sacrificed for electrochemical addressability because of the need for an electroactive surface. Here, we reported a biomolecular photoactive device based on light-activated proton pump, bR, reconstituted into highly fluidic microcavity-supported lipid bilayers (MSLBs) on functionalized gold and polydimethylsiloxane cavity array substrates. The integrity of reconstituted bR at the MSLBs along with the lipid bilayer formation was evaluated by fluorescence lifetime correlation spectroscopy, yielding a protein lateral diffusion coefficient that was dependent on the bR concentration and consistent with the Saffman–Delbrück model. The photoelectrical properties of bR-MSLBs were evaluated from the photocurrent signal generated by bR under continuous and transient light illumination. The optimal conditions for a self-sustaining photoelectrical switch were determined in terms of protein concentration, pH, and light switch frequency of activation. Overall, a significant increase in the transient current was observed for lipid bilayers containing approximately 0.3 mol % bR with a measured photo-current of 250 nA/cm2. These results demonstrate that the platforms provide an appropriate lipid environment to support the proton pump, enabling its efficient operation. The bR-reconstituted MSLB model serves both as a platform to study the protein in a highly addressable biomimetic environment and as a demonstration of reconstitution of seven-helix receptors into MSLBs, opening the prospect of reconstitution of related membrane proteins including G-protein-coupled receptors on these versatile biomimetic substrates.


Labelling of bacteriorhodopsin for Fluorescence Lifetime (Cross)Correlation Spectroscopy -FLCCS/FLCS
The incorporation of bacteriorhodopsin (bR) to proteoliposomes and to MSLBs was conducted using fluorescence studies, such FLCCS and FLCS. To do so, bR was marked ATTO-532 using NHS-ester coupling to an amino residue to covalently bound the ATTO dye to the protein. The labelling procedure used was provided by ATTO-TECH and followed as it is. The labelling efficiency was evaluated by UV-Vis spectroscopy and the degree of labelling (DOL) was measured after dialysis of unreacted ATTO532 dye from the labelled protein. Figure S7 shows the UV-Vis absorption spectra of unlabelled bR (black line) and labelled bR-ATTO532 (green line). The DOL was calculated by comparing the absorbance of protein versus labelled protein using equation S1, indicating a DOL of approximately 60%. = 532 . ,280 ( ,280 -532 ) . 532 eq. (S1) Here the extinction coefficient of ATTO532 at 532nm and bR at 280nm were ε ATTO532 = 1.15x10 5 M -1 cm -1 ( 1 ) and ε bR280 ≈ 63000 M -1 cm -1 ( 2 ), respectively. S4 Figure S1. UV-Vis spectra of bR after NHS-ester coupling of bR with ATTO532. The grey insertion represents the aromatic amino acids (a.a.a) absorption region. 3 The purple insertion illustrates the bR-ATTO532 conjugate absorption with maximum absorbance 530-550 nm.

Determination of optimal Titron-X100 concentration for protein incorporation
The optimal concentration of Titron-X100 for the protein incorporation using liposomes comprised of DOPC was determined by the change on the optical density of the liposomes solution, as previously proposed (4). Additionally, the change on the liposomes size was monitored by Dynamic Light Scattering (DLS) (Malvern, Zetasizer). The objective of this method is to create spaces into the liposome with the detergent and fill the gaps with bR. The UV-Vis ( Figure S5, open blue circles) shows that the optical density (OD) of the solution against 540 nm initially increases to around 2 mM of Triton X-100. Then OD linearly decreases until approximately 10.5 mM of Triton X-100. At this point, the liposomes are partially to completely dissolved by the detergent in solution. 4 The size of liposomes monitored against detergent concentration indicates that the liposomes swell due to the insertion of detergent to the lipid bilayer until disruption at 10 -11 mM. ( Figure S5, black circles) Therefore, to avoid partial disruption of the liposomes during bR incorporation, the concentration of Triton X-100 used in this work was 5 mM. Schematic S1. Schematic illustration of proteoliposomes preparation comprised of DOPC and containing bR.

Electroplating of Gold
In order to obtain a semi-hemispherical gold microcavity array, gold was electrodeposit to pre-casted monolayer of polystyrene microspheres using a cyanide free electroplating solution (Technic Inc.) The gold solution was purged with nitrogen for at least 30 min before reaction. Gold was electrodeposited at -0.6 V (vs. Ag/AgCl) using a 3-electrod set up with the gold array as working electrode, Ag/AgCl (KCl 1M) as reference electrode and a platinum wire as counter-electrode. The gold deposition was controlled by the evolution of current over time ( Figure S3a). The electrodeposition was stopped at the lower current, which represents

Characterization of PDMS and Gold µcavity arrays
The lipid bilayers comprised of DOPC/bR were spanned across PDMS and Gold arrays prepared using polystyrene (PS) spheres templating. In order to obtain a highly closed packed array, the Gold arrays were prepared by gravity-assisted lithography of PS, followed by gold electroplating. The dimensions of microcavities were analyzed with FESEM as illustrated in     Figure S7. c) Actual image obtained from the focal spot using a CCD camera.
Membrane impermeable probe pyranine was introduced in the pores of the array before assembly of the bilayer presonicating the microcavity arrays in 1 M pyranine solution. After, we prepared a lipid bilayer using the method here, by LB followed by proteoliposomes fusion.

S14
The substrate was imaged using a confocal microscope (Leica TCS SP8, Ex: 405 nm, detection window 440 -470nm). As shown in figure S9, we observed that after the bilayer preparation, the dye remain encapsulated within the microcavities, which indicate the bilayer is both suspended across the pore aperture and forms an effective impermeable seal, this persists in this way for several hours up to days Figure S9. Membrane Leakage test using membrane impermeable probe pyranine. The pryanine is introduced by sonication in solution into the microcavities by prior to monolayer transfer and proteoliposomes fusion. We observed that once the bilayer is in place the pyranine was retained within the cavity with no loss of emission intensity over the window of our experiments confirming the bilayer was spanning and its integrity maintained during our experimental windows.