FAPP2, cilium formation, and compartmentalization of the apical membrane in polarized Madin-Darby canine kidney (MDCK) cells

Abstract

We have analyzed the role of the phosphatidylinositol-4-phosphate adaptor protein-2 (FAPP2), a component of the apical transport machinery, in cilium formation in polarized Madin-Darby canine kidney (MDCK) cells. We show that ciliogenesis is defective in FAPP2 knockdown cells. Furthermore, by using fluorescence recovery after photobleaching studies of domain connectivity and the generalized polarization spectra of Laurdan, we demonstrate that FAPP2 depletion impairs the formation of condensed apical membrane domains. Laurdan staining also revealed that the ciliary membrane has a highly condensed bilayer domain at its base that could function as a fence to separate the ciliary membrane from the surrounding apical membrane. These results indicate that the compartmentalization of the apical membrane in MDCK cells into the ciliary membrane and the surrounding membrane depends on the balance of raft and nonraft domains.

Fig. 1.

FAPP2 is required for ciliogenesis. Primary cilia on filter-grown MDCK cells and on Matrigel cysts. MDCK cells were grown at confluence for 4 or 7 days on Transwell filters (A–C) or Matrigel (D–F), respectively, and then processed for IF. Cilia were labeled with an antibody to acetylated tubulin to reveal the ciliary axoneme (A, B, D, and E, red) and with DAPI to stain the nucleus (D and E, blue). (C and F) Quantification of the effect of FAPP2 depletion on cilia formation on filter-grown MDCK cells and cysts, respectively. Data are means ± SE of three separate experiments. A and D and black columns in C and F are controls. B and E and white columns in C and F are KD cells. (Scale bars, 10 μm.)

Fig. 2.

FAPP2 depletion causes an accumulation of subapical vesicles. (A–F) Distribution of apical (A and B, podocalyxin, red), subapical (C and D, galectin-3), and basolateral (A and B, green, E-cadherin; E and F, gp58) markers in control (A, C, and E) and FAPP2-KD cells (B, D, and F). (G and H) Electron micrographs of control (G) and KD cells (H). H also shows the zoom in of the microtubule-organizing center region. The cells were grown at confluence for 1 day on Transwell filters and then processed for indirect immunofluorescence or for EM. (Scale bars, 10 μm.)

Fig. 3.

FAPP2 depletion induces morphological changes in the apical lumen of cysts. (A) The transepithelial resistance (TER, graph) was determined in filter-grown cells as described in Materials and Methods. The cells were seeded on polycarbonate filters for 4 days. The black column corresponds to control, and the white column corresponds to KD cells. Data are means ± SE of three separate experiments. (B–E) KD and control cells were suspended in Matrigel and cultured in the matrix for 4–7 days. The cysts were fixed and processed. Confocal sections passing through the middle of each cyst are shown. Whereas >60% of the KD cells (C) had formed cysts with a collapsed apical lumen, as visualized by actin staining with TRITC-phalloidin (red) and nuclear staining with DAPI (blue), only 20% of control cells (B) displayed a collapsed lumen. Apical (podocalyxin, red) and basolateral (gp58, green) distribution in control (D) and FAPP2-KD (E) cells is shown. (Scale bars, 10 μm.)

Fig. 4.

FAPP2 depletion changes the physical state of the apical plasma membrane. (A and B) Forssman glycolipid distribution in filter-grown control (A) and KD (B) cells. (Scale bars, 10 μm.) (C–F) Control (C and E) and FAPP-2 KD (D and F) cells were plated onto coverslips at a density of 100 × 10³ per 2 cm², grown to confluent monolayers for 24 h, Laurdan labeled, and stained for apical marker protein podocalyxin (C and D) or basolateral marker gp58 (E and F). The distributions of ≈12 GP images from a single experiment taken at a focal depth corresponding to either apical or basolateral membranes were normalized and fitted to two Gaussian populations (black line through data). The fit yielded in a fluid population (P_f, dark gray) and a more ordered population (P_o, light gray). The mean GP of these populations as well as the coverage of the population (corresponding to the area under the curves) is given in each subpanel. Erfs for C–F are 0.0031, 0.0025, 0.0049, and 0.0018, respectively. (G) Table showing the diffusion coefficient, D, and the fraction that recovered in FRAP experiments on control and FAPP2-KD cells. Values are presented as averages ± 1 SD for the nonraft marker protein (EGFR-TMD-GFP in blue) and the raft marker protein (LAT-TMD-GFP in red). A circular spot of 4-μm² area was bleached on the apical membrane, and the fluorescence recovery due to diffusion was recorded at room temperature. The raw experimental data were corrected for bleaching during the recording, normalized, and fitted to a 2D diffusion model (12, 13). The number of FRAP curves n used for the theoretical analysis is given by n = xy, where x is the number of independent experiments, and y is the average number of cells examined per experiment.

Fig. 5.

The ciliary membrane has a condensed lipid domain at its base. (A, D, and J) Primary cilia revealed by using an antibody to acetylated tubulin. (B) FGL distribution. (C) FGL labels both apical membrane and cilia. (E and G) YFP-GL-GPI distribution. (F) YFP-GL-GPI is excluded from the center of the apical membrane that corresponds to the site of outgrowth of the primary cilium (red). (H) Galec tin-3 distribution. (I) Galectin-3 is confined to a subdomain of the apical membrane surrounding the site of outgrowth of the primary cilium. (J–L) Filter-grown MDCK cells were fixed and simultaneously imaged for the Laurdan intensity and acetylated tubulin. For Laurdan intensity, images were converted to GP images (K) as described in Materials and Methods. (Scale bars, 10 μm.) (M) GP values at apical membrane domain (▴, n = 39), base (▾, n = 28), and tip (♦, n = 28) of cilia were determined for individual images. Means (indicated by horizontal lines) and SDs are indicated in the graph.