JCS024588 Supplementary Material

Files in this Data Supplement:

• Supplemental Figure S1 (Adobe PDF) -

  Fig. S1. Alignment of caveolin sequences used to generate phylogenetic trees. Residues 54-158 from human Cav1 (HsCav1) were aligned with the corresponding residues of the indicated species. The residue numbers for HsCav1 are used throughout the text. The other numbering systems (denoted FMB and alignment) were used during data analysis. Latin species names are indicated with the genus abbreviated by a single letter. Residues highlighted in grey are conserved in &γτ;90% of all sequences and residues collared in red are only conserved in caveolin sequences that can form caveolae. Sequences are named as described for Fig 1. Gaps in alignment are noted as −. Mutations or deletions of Cav3 found in human muscle diseases are denoted by either black asterisks or delta symbols, respectively. Triple amino-acid deletions of Cav3 are denoted by a line. Cav1 mutations involved in human disease are denoted by red asterisks. palmitoylated cysteine residues are denoted by +.

• Supplemental Figure S2 (Adobe PDF) -

  Fig. S2. Tubular structures formed in cells expressing flotillin-1 and flotillin-2. Cav1-null MEFs were transiently transfected with flotillin-1-GFP and flotillin-2-RFP. Cells were FAC sorted, fixed and examined by electron microscopy. No caveolae-like structures (as determined by morphology analysis) were observed in transfected cells, but an increase in tubular structures was observed.

• Supplemental Figure S3 (Adobe PDF) -

  Fig. S3. CeCav-a does not behave in a manner similar to HsCav1 in mammalian cells. (A-C) Microinjection of Cav1-null MEFs was used to cause the overexpression of either HsCav1-HA (A,B), CeCav-a-flag (A,C). Cholera toxin binding (CTB) subunit was bound to the surface of cells, then incubated at 37°C for 40 minutes and fixed and processed for immunofluorescence confocal microscopy using antibodies against GM130 and either the HA epitope or the FLAG epitope. In control Cav1-null MEFs, CTB accumulated in a GM130-positive compartment, judged as the Golgi complex (A). In cells overexpressing HsCav1-HA, CTB accumulation in the Golgi complex was reduced (asterisks), as compared with neighbouring control cells. Overexpression of CeCav-a-FLAG did not affect CTB accumulation in the Golgi complex. The fluorescence intensity of CTB accumulation in the Golgi complex was measured in injected and non-injected cells (graphs B and C). There was a significant reduction in the accumulation of CTB in the Golgi complex in cells over expressing HsCav1-HA, although no significant difference in cells injected with CeCav-a-Flag as compared to controls. (**P<0.001). (D-E) Cav1-null MEFs expressing HsCav1-HA (D,F) or CeCav-a-HA (E) were fixed, incubated with anti-HA antibodies, and processed for TIRF microscopy. HsCav1-HA appeared in defined punctate structures within the plasma membrane (see inset). CeCav-a-HA had a more diffuse irregular distribution within the plasma membrane than HsCav1-HA Scale bars: 10 µm; 5µm (inset). The frequency at which defined values of fluorescent intensity of the HsCav1-HA punctate occurred were plotted (F). HsCav1-HA puncta can be classified into groups according to their fluorescent intensity. The fluorescent intensities of the group increase in a defined quantal manner. A circular mask of fixed size was used to select 135 individual puncta, and the average intensity of these regions was recorded. The frequency of an intensity was calculated by creating pools (0.01 to 100, 100.01 to 200.01 etc..) from 0 to 4000. HsCav1-HA clearly exhibited punctate fluorescent structures of defined quantal units that correspond to either single caveola or double caveolar structures, unlike previous reports we were unable to identify multiple caveolae structures. This maybe due to the use of both primary and secondary antibodies to detect Cav1-HA compared with analysing Cav1-GFP (Pelkmans and Zerial, 2005) where the fluorescent signal may be a more linear representation of the amount of Cav1.

• Supplemental Figure S4 (Adobe PDF) -

  Fig. S4. CeCav-a does not behave in a manner similar to HsCav1 in mammalian cells − continued analysis. (A) Confluent BHK cells transfected with HsCav1-HA and CeCav-a-Flag, were allowed to recover for 5 hours after the cell monolayer was wounded. Cells were fixed, permeablised and probed with anti-FLAG (red) and anti-HA (green) antibodies before being processed for immunofluorescent microscopy. The wound in the cell monolayer was analysed for polarised migrating cells, an arrow shows direction of cell migration and the cell has been outline in red. (B) Cav1-null MEFs transiently expressing HsCav1-YFP or CeCav-a-Flag, were incubated with oleic acid for 15 hours and then processed for immunofluorescence microscopy. Cells were labelled with the neutral lipid dye, Oil Red. All cells examined had accumulated Oil Red-positive lipid droplets. Oleic acid treatment caused a redistribution of HsCav1-YFP to lipid droplets but did not affect CeCav-a localisation (see insets). (C) Cav1-null MEFs co-expressing Rab5^Q79L and either HsCav1 or CeCav-a. CeCav-a, but not HsCav1, accumulates in Rab5^Q79L-enlarged early endosomes. Scale bars: 10 µm (A), 20µm (B,C), 10 µm (insets).

• Supplemental Figure S5 (Adobe PDF) -

  Fig. S5. Analysis of mutant caveolin proteins using a combination of light- and electron-microscopy assays. (A) Cav1-null MEFs expressing truncation mutants of CeCav-a and hybrid proteins between HsCav1 and CeCav-a were either fixed and directly processed, or incubated with anti-HA antibodies, before being processed for confocal microscopy. HCav-HA^49-132, HCav-HA^1-132 and HCav-HA^CeCav1-110 all had a distribution similar to Cav1 in WT MEFs and formed caveolae. HCav-HA^102-132, HCav-HA^80-132 and HCav-cherry^60-132 were localised to the plasma membrane but failed to form caveolae as determined by EM assay. CeCav-a-HA^Δ1-86 was localised to intracellular vesicular. HCav^1-101 was restricted to a perinuclear compartment. The results of the EM caveolae biogenesis assay are indicated in the table below. (B) Cav1-null MEFs transiently expressing Cav-HA trunctation mutants were fixed and labelled with anti-HA antibodies and examined by confocal microscopy. The transiently expressed deletion mutants of CfCav1-HA^Δ157-178, CfCav1-HA^Δ147-178 and MmCav3 ^{Δ 1-16} were localised to the PM and formed caveolae. All other constructs had a restricted perinuclear localisation resembling the Golgi complex. The presence or absence of caveolae according to the results of the EM caveola biogenesis assay are also indicate. (C) Cav1-null MEFs expressing CfCav1-HA^CYS-, HsCav1-HA^S80A, HsCav1-HA^S80E, were fixed and incubated with anti-HA antibodies, before being processed for confocal microscopy or processed for EM caveolae biogenesis assay. CfCav1-HA^CYS- and HsCav1-HA^S80A had a distribution similar to Cav1 in WT MEFs and formed caveolae. Cav1-HA^S80E was restricted to a perinuclear compartment. The results of the EM caveolae biogenesis assay are indicated in the table below. −, experiment was not performed. Scale bars: 10 µm; inset, 5µm.

• Supplemental Figure S6 (Adobe PDF) -

  Fig. S6. Additional analysis of CeCav-a and the interaction of HCav constructs and Cav2. (A) Electron micrograph of Cav1-null MEFs expressing HCav^1-101. Injected cells were identified by electron-dense nucleus (N+) as compared to non-injected cells (N-). No CTB-HRP-positive surface-connected flask- shaped or vesicular structures 50 nm in diameter were observed. Scale bar 5µm. (B) Cav1-null MEFs expressing HsCav1-HA, HCav^1-101-HA, and HCav-cherry^60-132 were fixed and, where appropriate, incubated with anti-HA antibodies and processed for TIRF microscopy. HsCav1-HA appeared in defined punctate structures within the PM, but both HCav^1-101 and HCav^60-132 had more irregular evenly distribution PM localisation. Only in a subset of cells was HCav^1-101 localised to the PM. Scale bars 10 µm. (C) Cav1-null MEFs were transiently transfected with HsCav1-HA, HCav1-HA^102-132, HCav1-HA^80-132 and HCav1-cherry^60-132. Localisation of Cav2 within these cells was investigated by immunofluorescence microscopy. Cells were probed with antibodies against Cav2 and HA. In untransfected Cav1-null MEFs Cav2 was retained in a perinuclear compartment. Cav1-HA and the hybrid proteins examined caused a redistribution of Cav2 from the Golgi complex even through they cannot form caveolae. Scale bars: 10 µm.

• Supplemental Table S1 -

  Adobe PDF

• Supplemental Table S2 -

  Adobe PDF