Extended Data Fig. 1: LC3-positive membranes associate with and deform p62 droplets.
From: Wetting regulates autophagy of phase-separated compartments and the cytosol
a, p62 droplet contours were obtained by live-cell imaging of RPE1 cells expressing mCherry–p62 and analysed. Binned droplet circularities are plotted as a histogram (n = 421 droplets assessed from three independent experiments). The frequency peak centred at a circularity of 0.90 (Lorentz fit ± s.d., a circularity of 1 indicates a perfect circle). b, Non-circularity correlates with increasing droplet size. Filled data points represent n > 5 droplets and the open data point n = 1 droplet. Data are from a. c, Four representative droplets were observed continuously for >10 min at 12 frames/min and circularities determined. Circularity is plotted for each single droplet individually over time. Frames in which another droplet moved into the field of view were excluded from analysis. d, Mean circularities of single droplets shown in c; n = data points/droplet (black n = 145, red n = 143, blue n = 136, green n = 139). The mean of the four droplets was 0.91 (s.d. = 0.01). e, f, Multiple small autophagosomes form at the surface of large p62-containing droplets. Representative confocal images of a live cell in Eagle’s balanced salt solution (e) and a fixed specimen prepared for ultrastructural analysis of the structure highlighted in f. All images are of RPE1 cells expressing GFP–LC3 and mCherry–p62. Magnified areas in f show single confocal z-sections of autophagosomal membranes (ii and iii); corresponding tomographic sections are shown in Fig. 1f, g. g, Electron tomography reconstruction of autophagosomal membranes i, ii and iii without (left) or with (right) the p62 droplet. The mean radius of the autophagosomal structures i, ii)and iii was 0.23% ± 0.09% ( ± s.d.) of the droplet radius. h, Schematic of droplet deformation by autophagosomal membranes during piecemeal sequestration, as observed in Fig. 1d, e. i, Immunolabelling of a 200-nm section of RPE1 cells expressing mCherry–p62 with an anti-p62-antibody and 10 nm gold particles. Left, transmission electron microscopy image of the section; right, surface tomography slice. The broken pink line delineates the droplet boundary. j, Left, enlarged surface tomography slice of the region indicated in i (right panel). Right, enlarged tomography slice through the section centre of the same region. Five droplets of the same preparation were examined with similar results (i, j). k, FRAP of proteins localizing to p62 droplets. LC3, p62 and p62(ΔLIR) all localized to droplets. Regions of radius a = 1.7 μm were bleached and the recovery of fluorescence recorded. The normalized intensity change over time was fitted to an exponential curve f(x) = A × (1 – e−x/τ) to obtain the apparent recovery time τ and estimate apparent diffusion coefficients (Dapp = a2/τ): DLC3 = 8.44 ± 0.50 μm2/min, Dp62 = 0.54 ± 0.01 μm2/min, and Dp62(ΔLIR) = 0.50 ± 0.01 μm2/min. The curves show mean recovery ± s.d. from 74 ROIs examined over 5 independent experiments (p62), 68 ROIs examined over 3 independent experiments (p62(ΔLIR)) and 62 ROIs examined over 5 independent experiments (LC3). l, Left, FRAP of a droplet-bound autophagosome in RPE1 cells expressing mCherry–p62 and GFP–LC3 (bleach spot and autophagosome as indicated). The green arrow indicates the position of the droplet-attached autophagosome. Intensities of single-channel microscopy images are inverted. Right, line profiles as indicated. Scale bars, 5 μm (e, f), 1 μm (l) or 0.5 μm (i, j). Representative images of n independent experiments (n = 2 for f, l; n = 7 for e).
