Unroofing site-specific α-synuclein–lipid interactions at the plasma membrane

Significance α-Synuclein is a neuronal protein with an ill-defined biological function that is central to Parkinson’s disease etiology. While considered to be involved in exocytosis, how α-synuclein facilitates synaptic vesicle fusion and release remains an open question. To address this, we investigated α-synuclein–lipid interactions at the plasma membrane through the technique of cellular unroofing, which uncovers an intact basal membrane. We conclusively show that α-synuclein is recruited to exocytic sites, preferring liquid-ordered lipid domains. Importantly, heterogeneous populations of α-synuclein conformers are revealed by measurements of fluorescence lifetime distributions, which are not adequately described by current models of α-synuclein structures. Membrane-bound α-synuclein is conformationally dynamic, exquisitely sensitive to lipid/protein composition, enabling the protein to carry out its function.

Protein Labeling. Stock solutions of the DyLight-488 maleimide (Thermo Fisher 46602) and IANBD Amide (N,N'-Dimethyl-N-(Iodoacetyl)-N'-(7-Nitrobenz-2-Oxa-1,3-Diazol-4-yl) Ethylenediamine, Thermo Fisher D2004) were prepared in dimethylformamide (DMF). To eliminate intermolecular disulfide bonds, 20 mM DTT was added to AcV26C-, AcV40C-, and AcY136C-α-syn at RT for 30 min prior to labeling. After the DTT treatment, the reduced protein was buffer exchanged into the reaction buffer (phosphate-buffered saline (PBS) and 1 mM EDTA at pH 7.4) using a PD-10 desalting column (GE Healthcare). A 3-molar excess of Dylight-488 maleimide or IANBD was then added to the protein solution ([α-syn] = 50 µM), and the reaction mixture was incubated for 2 h at RT in the dark. Afterwards, excess dye was removed by using a PD-10 desalting column. Concentration of labeled protein and degree of labeling were calculated using the following formulas: where the extinction coefficients were 5,970 M −1 cm −1 for AcV26C and AcV40C and 4470 M −1 cm −1 for AcY136C. Typical degree of labeling was 100% and molecular weight of the labeled protein was confirmed by LC-ESI-MS (NHLBI Biochemistry Core). V26CNBD and V40CNBD was measured to be 14,797.7 Da. Y136CDy488 and Y136CNBD were measured to be 15,219.7 and 14,765.23 Da, respectively.
Cell Culture. SK-MEL-28 (ATCC HTB-72) cells were maintained in phenol-free minimum essential medium (MEM, Thermo Fisher 41061307) supplemented with 10% fetal bovine serum (FBS, ATCC 30-2020) and 2% penicillin/streptomycin at 37 °C in 5% CO2. SK-MEL-28 cells at 70-80% confluence in T-75 flasks were trypsinized with the addition of 0.25% Trypsin-EDTA (Thermo Fisher 25200056) for 2 min. Resuspended cells (1 mL) and fresh media were added to circular #1.5 glass coverslips coated with poly-D-lysine (Neuvitro H-25-1.5-PDL) in 6-well plates and allowed to grow for 24 h. Fluorescent Labeling of Unroofed Cells. The lipid stains were prepared by diluting Vybrant DiD (Thermo Scientific V22889) and cholera toxin subunit-B (CT-B, Alexa-555 conjugate, Invitrogen C34776) into stabilization buffer (30 mM HEPES, 70 mM KCl, 5 mM MgCl2, 3 mM EGTA, pH 7.4) to a final concentration of 2.5 and 1 ng/mL, respectively. Unroofed cells were allowed to incubate with each stain for 5 min at RT in the dark. After staining, the coverslips were rinsed with stabilization buffer three times to remove any excess dye. For confocal fluorescence microscopy experiments, appropriate amounts of dye-labeled α-syn were first diluted into stabilization buffer to a final concentration of 1 µM and added to the stained unroofed cells. These samples were imaged immediately in the presence of α-syn in solution without fixation.

Confocal Fluorescence Microscopy.
Samples were imaged using a UPLSAPO 100×/1.35 NA silicone oil objective (Olympus, Tokyo, Japan) on an OlympusIX73 inverted microscope fitted with a Thorlabs Confocal Laser Scanner (CLS-SL) fiber coupled to a multichannel CMLS-E laser source. CT-B and Alexa Fluor 532 were excited using a 532 nm laser line and emission was collected using a 582 ± 75 nm bandpass filter. DiD and Alexa Fluor 642 were excited using a 642 nm laser line and emission was collected using a 660 nm long pass filter. NBD-labeled α-syn and Y136CDy488 α-syn were excited using a 488 nm laser line and emission was collected using a 512 ± 25 bandpass filter. A 50-μm pinhole was used and the scale was 0.1 μm/pixel. Images were analyzed with Fiji. Briefly, a Gaussian blur was applied to the raw image and the resulting image was used to subtract background signal. Next, the Otsu method was used to determine a threshold for binary images. Colocalized pixels were determined using Boolean Algebra ("intersection" function) to identify the pixels in which α-syn and the corresponding protein were both detected. Colocalization was analyzed by calculating a Pearson correlation coefficient (PCC) using the ImageJ plugin for colocalization (Coloc_2) for all acquired images. To verify the PCC values generated from Coloc2, object-based analysis was also used (ImageJ, JACoP). Multiple unroofed cells were imaged (n ≥ 50) and trends were verified with at least two biological replicates and two independent treatments.
Fluorescence recovery after photobleaching (FRAP) experiments were acquired on a Zeiss 780 confocal microscope (NHLBI Light Microscopy Core) using a 63 oil immersion objective (NA 1.4). Fluorescence intensity recovery was monitored for approximately 120 s. CT-B fluorescence was excited by a 561-nm laser line and fluorescence was collected using a 605 ± 50 bandpass filter. DiD fluorescence was excited by a 633 nm laser line and fluorescence was collected using a 633 S4 nm long pass filter. Photobleaching was performed using 100% laser power for each laser line for 50 iterations.

Phospholipid Micelle and Vesicle Preparation.
Micelle stocks (100 mM) were produced by hydrating 16:0 LPC (Avanti, 855675P) in PBS buffer. Small unilamellar vesicles (SUVs) were prepared following previously established protocols (1). SUVs composed of 1,2-dioleoyl-snglycero-3-phosphocholine (DOPC; Avanti, 850375) and ganglioside GM1 (Avanti, 860065P) at a molar ratio of 4:1 were prepared by mixing of the different lipids in chloroform stock solutions. Lipids were dried under flowing nitrogen gas and to ensure removal of organic solvent the lipids were allowed to dry overnight at 50 °C in a vacuum oven. Lipid films were rehydrated in buffer (PBS, pH 7.4) and resuspended using bath sonication for 10 min (Branson 2510 Ultrasonic Cleaner). To produce SUVs, the resuspended lipids were probe-tip sonicated for 30 min in a water bath (Branson 450, output 6, 50% duty cycle). Vesicles were then centrifuged (17,000× , 20 min) and syringe filtered through a 0.22 µm membrane (Millipore). Vesicle size (r ~ 40 nm) was determined by dynamic light scattering using a Wyatt Synapro NanoStar (NHLBI Biophysics Core). Micelles and SUVs were used immediately after preparation.
Fluorescence Spectroscopy. NBD-labeled protein (1 µM) was added to varying concentrations of 16:0 LPC and DOPC/GM1 vesicles. Emission was excited at 480 nm and monitored from 490 to 700 nm at 25 °C on a Fluorolog FL-3 instrument (Horiba Scientific) using 1-nm slit widths and 0.5 s integration times. All spectra were averages of 3 accumulations and their buffer/lipid backgrounds have been subtracted. Circular Dichroism Spectroscopy. CD measurements were carried out on a JASCO J-715 instrument (NHLBI Biophysics Core) in a 1-mm quartz cuvette. Spectra were collected from 200-260 nm with the following settings: continuous mode, 1 nm steps, 100 nm/min, 1 nm bandwidth, 1 s integration time, and three accumulations at 20 °C. Buffer/lipid background subtractions were applied to all spectra and the mean residue ellipticity (MRE) was calculated using the following equation (4): where MRW represents the mean residue weight, represents the ellipticity measured by the instrument, and represent the pathlength (cm) of the cuvette used and protein concentration (g/mL), respectively.
[ ] = 10  1 Data were fit to an exponential decay function: ( ) = ∑ exp ( − τ ⁄ ) using the SymPhoTime 64 software (PicoQuant) where and τ are the pre-exponential and fluorescence lifetime, respectively. Uncertainty of the fit parameter is indicated. Instrument response ≤ 340 ps.   1 Data were fit to an exponential decay function: ( ) = ∑ exp ( − τ ⁄ ) using the SymPhoTime 64 software (PicoQuant) where and τ are the pre-exponential and fluorescence lifetime, respectively. Uncertainty of the fit parameter is indicated. Instrument response ≤ 340 ps.