Cholesterol-Induced Nanoscale Variations in the Thickness of Phospholipid Membranes

Graphene-induced energy transfer (GIET) is a recently developed fluorescence-spectroscopic technique that achieves subnanometric optical localization of fluorophores along the optical axis of a microscope. GIET is based on the near-field energy transfer from an optically excited fluorescent molecule to a single sheet of graphene. It has been successfully used for estimating interleaflet distances of single lipid bilayers and for investigating the membrane organization of living mitochondria. In this study, we use GIET to measure the cholesterol-induced subtle changes of membrane thickness at the nanoscale. We quantify membrane thickness variations in supported lipid bilayers (SLBs) as a function of lipid composition and increasing cholesterol content. Our findings demonstrate that GIET is an extremely sensitive tool for investigating nanometric structural changes in biomembranes.


Substrate preparation
Coverslips coated with single sheets of graphene were prepared by a transfer method based on the manufacturer's instruction. Briefly, the purchased monolayer graphene (Easy Transfer Monolayer, Graphenea) was sandwiched between a polymer film and a thin sacrificial film and cut into ∼1×1 cm 2 pieces, which were submersed into water for removing the polymer film. The floating graphene sheet with the sacrificial film was captured with a plasma-cleaned coverslip and dried at room temperature for 30 min, followed by heating at 150 • C for 1 h. Dried graphene-coated coverslips were stored under vacuum for at least 24 h to prevent detachment of the graphene from the coverslip. Finally, the sacrificial layer was removed by placing the coverslip into hot acetone (50 • C) for 1 h and into isopropyl alcohol for another 1 h. Coverslips were dried in a stream of N 2 and stored in a desiccator. In a next step, graphene-coated coverslips were coated with a SiO 2 spacer of 10 nm thickness by chemical vapor deposition using an electron beam source (Univex 350, Leybold) under high vacuum conditions (10 −5 mbar). Deposition was done at a slow rate of 1 Ås −1 to ensure maximal homogeneity of the deposited quartz layer.
Layer thickness was continuously monitored during deposition with an oscillating quartz unit.

Sample preparation
Small unilamellar vesicles (SUV) of DLPC and DOPC with and without cholesterol were prepared by extrusion. Briefly, a 60 µl droplet of chloroform solution containing unlabeled lipids (10 mg/ml DLPC or DOPC), cholesterol (15, 30 and 44 mol%), and 1 µl of 0.01 mg/ml DLPE-Atto655 (for DLPC SLB) or DPPE-Atto655 (for DOPC) was dried in vacuum for 1.5 h at 30 • C to evaporate the chloroform. The obtained lipid film was re-suspended with 500 µl of Tris buffer (20 mM Tris-Cl, 100 mM NaCl, 10 mM CaCl 2 , pH 7.4) in an ultra-sonic bath for 5 min, followed by stirring (Thermomixer Comfort, Eppendorf) at 30 • C for 1 h. The solution was then extruded for 15 cycles through a polycarbonate filter (Whatman) with 50 nm pore diameter. The resulting vesicle solutions were used within 3 days and stored at 4 • C before use. Giant unilamellar vesicles (GUV) of DPhPC with and without cholesterol were prepared by electro-formation as described before. Briefly, 100 µl of a chloroform solution containing DPhPC lipid (10 mg/ml) and cholesterol (15, 30 and 44 mol%) and 2 µl of 0.01 mg/ml DMPE-Atto655 was filled into a custom-built chamber, followed by evaporation for 3 h under vacuum at 30 • C. The chamber was re-filled with 500 µl of 300 mM sucrose solution, after which an alternating electric current of 15 Hz frequency and a peak-to-peak voltage of 1.6 V was applied for 3 h, followed by a lower frequency voltage of 8 Hz for another 30 min. Formed GUVs were collected by rinsing the electrode surface with 500 µl of a Tris-Cl buffer solution. Next, DLPC and DOPC SLBs were formed via vesicle fusion. Before placing a SUV solution onto the GIET substrate, the substrate's surface was activated with a plasma cleaner (Harrick Plasma, New York, United States) at low intensity for 30 s. After that, a droplet of SUV solution was deposited on the substrate and incubated for 3 h to ensure the formation of a uniform bilayer with minimal defects. This was followed by washing with copious buffer. For forming a homogeneous SLB with high cholesterol content, samples were incubated at room temperature for 6 h before further measurements. DPhPC SLBs were formed by putting a droplet of diluted GUV solution (10 times dilution from the stock solution) onto the GIET substrate and incubating for 10 min, followed by washing with buffer solution.

Determination of dye orientation
To determine the orientation of fluorophore atto655 in the membrane, we prepared two kinds of GUVs: 1) Pure DOPC and DPPE-atto655; 2) 30 mol% Chol in DOPC and DPPE-atto655. Then the GUVs are imaged under polarized excitation light by a epi-fluorescence microscopy. As shown in Figure S1 (3), both for the two GUVs, the observed intensity distribtuion is consistent with a dye orientation parallel to the bilayer surface.

Working principle of GIET
The inset of Figure 1 Figure 1

(B) in main text). By solving
Maxwell's equations, one find the emission power of the dipole emitter S(θ, z 0 ) as a function of its distance z 0 from the substrate surface and of its relative orientation defined by the angle θ between its dipole axis and the normal to the substrate surface. This emission power is inversely proportional to the radiative transition rate from the electronic excited state to the ground state. Taking into account also the non-radiative transition rate, the observable excited-state fluorescence lifetime (τ f ) is then found as where τ 0 is the free-space fluorescence lifetime in absence of the graphene layer, ϕ repre-