|
|
|
|
|
|
|
|
|||||
| Home Help Feedback Subscriptions Archive Search Table of Contents | |||||
| ||||||||||||||||||||
Files in this Data Supplement:
Fig. S1. Characterization of the preapical patch and the compact structure. (A) Localization of F-actin and gp135 in solid two-cell aggregates with gp135 at the basal cell surface, with gp135-positive compact structure, with a PAP, or with a lumen. (B-C) Localization of gp58 and E-cadherin (B) or CFTR (C) in solid two-cell aggregates without PAP, with a PAP, and with a lumen. Middle optical sections of fixed and immunostained aggregates are shown as well as some corresponding DIC images. Asterisk indicates lumen. (D) Partial colocalization of F-actin with the gp135-positive compact structure. Magnified view of the central region (outlined with a dashed line in the top row) is shown in the bottom row. Note groups of gp135-positive vesicles (black filled arrowheads) and actin structures (white arrowheads) that do not overlap and an unresolved compartment, where the signals overlap (asterisk). (E) Deconvolved confocal section showing the partial colocalization of prominin-1 with gp135 in the PAP of a two-cell aggregate. Magnified view of the central region, outlined with a dashed line, is shown in the bottom row. Scale bars: 5 µm. (F) Magnification of electron-dense junctional material of zone I in Fig. 2C. Scale bar: 200 nm.
Fig. S2. Gp135-positive membranes enter the plane of the contact membrane during PAP formation (aggregate of Fig. 2A). Deconvolved optical sections of an aggregate expressing mRFP-gp135 and E-cadherin-GFP at various Z-levels (0, 2, 4, 6, 8, and 10 µm indicated in top row) are shown at -30 minutes (A) and 0 minutes (B). White arrowheads highlight the gap in the E-cadherin-positive surface. Scale bars: 5 µm.
Fig. S3. Analysis of PAP formation. (A) Localization of E-cadherin-positive vesicles during PAP formation (aggregate of Fig. 2A). XY-projections of complete Z-stacks at 0, 150, and 390 minutes are shown. E-cadherin-positive vesicles are indicated by open arrowheads. Scale bar, 5 µm. (B) Time-sequence of deconvolved middle optical sections of a two-cell aggregate expressing mRFP-gp135 (top row) and pp-YFP (middle row). Relative time in minutes is shown in the top row. Scale bar: 5 µm.
Fig. S4. Statistical analysis of morphometric data. Volume of individual cells (A) and total aggregate volume (B) normalized by the number of cells in fixed aggregates. The 1 to 99 percentile distribution (vertical line) was derived from individual data points (gray circles). The population mean is reported as a horizontal black line enclosed in a rectangular box whose length represents the s.e.m. P-values were obtained by Mann-Whitney test. For these data sets 14 MDCK aggregates of 2 to 25 cells with lumen and 10 aggregates without lumen were analyzed in a total of four different experiments. (C) The averaged normalized basal surface (Sb) stays constant between 60 minutes before and after lumen appearance. Individual data points (± s.e.m.) were normalized to the values at time 0 (only cells facing the lumen were measured in seven-cell aggregates; n=18). P-value was calculated by linear regression analysis shown in (H). (D-H) Linear regression models of data shown in C and in Fig. 5A-D. Linear regression was used to model the statistical relationship of (D) normalized cell volume, (E) normalized sum of apical and contact surface, (F) normalized aggregate volume, (G) aggregate sphericity over time, and (H) basal cell surface. Both, individual data points (black squares) and the fitted linear curve (red line) are shown.
Fig. S5. Reduction of myosin light chain phosphorylation correlates with increased PAP and lumen formation. (A) PAP and lumen formation in two-cell aggregates grown for 24 hours in floating collagen gel supplemented with 0.3% DMSO, ML-7, blebbistatin, Y-27632 or in gel without additives (medium). The averaged frequencies of randomly picked two-cell aggregates with basal gp135 localization, with a PAP, or with a lumen (n=50; three experiments) are shown as columns with error bars (± s.e.m.). The values that are significantly different from the control (P<0.001) are marked with asterisks. (B) Longer culture time reduced the amount of aggregates with basal gp135 and correspondingly increased PAP and lumen containing aggregates. Quantification of 2-cell aggregates grown for 40 hours in floating collagen gel which were during last 24 hours treated with 3 µM Y-27632, 30 µM Y-27632, or not (medium). The averaged frequencies of randomly picked two-cell aggregates with basal gp135 localization, with a PAP, or with a lumen in three to five experiments (n=50) are shown as columns with error bars (± s.e.m.). The values that are significantly different from the control (P<0.001) are marked with asterisks. (C) Quantification of the pMLC and ppMLC levels shown in Fig. 6C. The band intensities were normalized by the input signal and are shown in percents of the control values (Y-27632) or in percents of DMSO values (ML-7) with error bars representing the s.e.m. (three experiments).
Fig. S6. (A) Localization of gp135 and TJ component ZO-1 in two-cell MDCK aggregates with a PAP, with or without TJ, and in aggregates with a lumen. Middle optical sections of deconvolved confocal Z-stacks are shown. Circular TJ appear in the optical section as two bright dots (arrowheads). (B) Effect of mSnail expression on E-cadherin, desmoplakin, and occludin-1. Deconvolved widefield images of plastic grown MDCK are shown. In the double staining E-cadherin/desmoplakin the recoding set up per antigen was relative to its maximal intensity within the three conditions; analogical the occludin representation. Scale bar, 20 µm. (C) Effect of mSnail expression on aggregate morphology. Middle-slices of confocal Z-stacks showing two-cell aggregates with PAP or lumen (red, gp135; blue, nuclei). Scale bars, 5 µm.
Movie 1. 4D-microscopy of a MDCK two-cell aggregate during PAP formation shown in Fig. 2. MDCK cells expressing E-cadherin-GFP (green) and RFP-gp135 (red) were grown in collagen and were imaged 24 hours after plating over 8 hours. One Z-stack was recorded every 30 minutes. In the 17 consecutive time points optical sections of the same Z-position are presented (displayed at 8 frames per second). Scale bar: 5 µm.
Movie 2. 4D-microscopy of a MDCK three-cell aggregate during lumen initiation shown in Fig. 3. MDCK cells expressing pp-YFP (green) and mRFP-gp135 (red) were cultured in collagen and were imaged 24 hours after plating. One Z-stack was recorded every 30 minutes. In the eight consecutive time points optical sections of the same Z-position are presented (displayed at 1 frame per second). Scale bar: 5 µm.
Movie 3. PAP position determines lumen appearance. Time series images of an MDCK two-cell aggregate during lumen initiation. MDCK cells expressing E-cadherin-GFP (green) and RFP-gp135 (red) were grown in collagen and were imaged 24 hours after plating. One Z-stack was recorded every 30 minutes (one middle optical section demonstrating the largest diameter of the lumen is shown; displayed at 4 frames per second). Scale bar: 5 µm.
| ||||||||||||||||||||
