Skip to main content
. 2020 May 4;9:e55038. doi: 10.7554/eLife.55038

Figure 3. Chol and GSLs induce surface release of caveolae via an EHD2-dependent mechanism.

(A) Representative images of maximum projected confocal z-stacks of Cav1-mCh HeLa cells. Untreated cells or cells treated with LacCer-Bodipy liposomes for 1 h, fixed and immunostained for endogenous EHD2. High-magnification images (dotted square) show localization of EHD2 to Cav1-mCh (see scatterplot for quantification). n ≥ 60, two independent experiments, mean ± SEM. ***, p≤0.001 vs. control. (B) Experimental protocols analogous to (A), with exception of endogenous cavin1 immunostaining. n ≥ 60, mean ± SEM. (C) Confocal FRAP of Cav1-mCh HeLa cells treated with either EHD2 siRNA or Bodipy-LacCer liposomes. A ROI was photobleached and recovery of mCherry FI monitored over 5 min. mCherry FI was normalized to background and reference. n ≥ 10, mean ± SEM. (D) Representative time-lapse series showing control Cav1-mCh HeLa cells and cells treated with either EHD2 siRNA or Bodipy-LacCer liposomes. The photobleached area is outlined with white circles. mCherry FI is intensity‐coded using LUT. (E) Effects of lipids on track duration of Cav1-mCh structures were analyzed following siRNA-mediated depletion of EHD2. n ≥ 8, two independent experiments, mean ± SEM. (F) Quantification of track duration of untreated Cav1-mCh HeLa cells or cells transiently expressing EHD2-BFP with or without incubation with liposomes. Changes in track duration are relative to EHD2-BFP control (indicated by dotted line). n ≥ 8, two independent experiments, mean + SEM. ***, p≤0.001 vs. control cells. (G) Representative live cell confocal image of EHD2-647 microinjected into Cav1-mCh HeLa cells. (H) Quantification of track duration of Cav1-mCh cells treated with Bodipy-LacCer and following microinjection of EHD2-647. n = 8, mean + SEM. All scale bars, 10 μm.

Figure 3—source data 1. Excel file containing source data pertaining to Figure 3A–C,E,F and H.
elife-55038-fig3-data1.xlsx (110.2KB, xlsx)

Figure 3.

Figure 3—figure supplement 1. Cavin1 localization to Cav1-mCh before and after liposome treatment.

Figure 3—figure supplement 1.

(A) Representative TIRF micrographs of cavin1-GFP transfected Dox-induced Cav1-mCh HeLa cells. The same cells were imaged before and after incubation with fusogenic liposomes as depicted (final total lipid concentration 7 nmol/ml). Scale bar, 10 μm. (B) Bar graph shows fold change of the percentage of colocalizing spots before and after liposome treatment relative to untreated cells. n ≥ 10, at least two independent experiments, mean ± SEM.
Figure 3—figure supplement 1—source data 1. Excel file containing source data pertaining to Figure 3—figure supplement 1B.
Figure 3—figure supplement 2. EHD2 depletion in Cav1-mCh HeLa cells.

Figure 3—figure supplement 2.

(A) Representative immunoblots of Cav1-mCh HeLa cells treated with Ctrl siRNA or siRNA against EHD2. Clathrin HC served as loading control. (B) Effect of lipids on track duration of Cav1-mCh structures analyzed following control siRNA-treatment. n ≥ 8, two independent experiments, mean ± SEM. (C) Quantification of track mean speed of Cav1-mCh structures from TIRF movies treated with control siRNA or siRNA against EHD2. Fold changes are relative to untreated cells (Cav1-mCh). n ≥ 8, two independent experiments, ***, p≤0.001 vs. untreated.
Figure 3—figure supplement 2—source data 1. Excel file containing source data pertaining to Figure 3—figure supplement 2B and C.
Figure 3—figure supplement 3. Stabilization of caveolae to the PM by EHD2-I157Q cannot be reversed by addition of Bodipy-labeled LacCer or Chol.

Figure 3—figure supplement 3.

(A) Representative time-lapse images of Cav1-mCh positive for EHD2-BFP (yellow arrows) or lacking EHD2-BFP (white arrows) in double Flp-In EHD2-BFP Cav1-mCh HeLa cells. Dotted box shows higher magnification region. Numbering corresponds to number of frames. Scale bar, 10 μm; inset scale bars, 2 μm. (B) Differences in track duration of Cav1-mCh structures positive for EHD2-BFP or lacking EHD2-BFP in double Flp-In EHD2-BFP Cav1-mCh HeLa cells. Percentage of Cav1-mCh structures positive or lacking EHD2-BFP are indicated. n = 8, mean ± SEM. (C) Representative immunoblots of double Flp-In EHD2-BFP Cav1-mCh HeLa cells induced with 1 ng/ml Dox. (D) FRAP curves of mCh-tagged EHD2 wt or EHD2 I157Q expressing HeLa cells. A ROI was photobleached and recovery of mCherry fluorescence intensity (mCherry FI) was monitored. Intensities were normalized to background and reference. n = 8, mean ± SEM. (E) Cav1-mCh HeLa cells transiently expressing EHD2-I157Q-BFP were incubated with Bodipy-LacCer or Bodipy-Chol liposomes and track duration was analyzed. n ≥ 8, two independent experiments, mean + SEM. Imaris software was used to analyze data. Changes in track duration are relative to EHD2-I157Q-BFP control (indicated by dotted line).
Figure 3—figure supplement 3—source data 1. Excel file containing source data pertaining to Figure 3—figure supplement 3B,D and E.
Figure 3—figure supplement 4. Microinjection of EHD2-647 into Cav1-mCh HeLa cells.

Figure 3—figure supplement 4.

(A) Representative live cell TIRF images of Cav1-mCh HeLa cells untreated or treated with Bodipy-LacCer and with or without microinjection of EHD2-647. (B) Quantification of the colocalization of microinjected EHD2-647 to Cav1-mCh in control cells and cells treated with Bodipy-LacCer liposomes prior to injection. n ≥ 5, mean ± SEM . Scale bar, 10 μm. Imaris software was used to analyze data.
Figure 3—figure supplement 4—source data 1. Excel file containing source data pertaining to Figure 3—figure supplement 4B.