|
|
|
|
|
|
|
|
|||||
| Home Help Feedback Subscriptions Archive Search Table of Contents | |||||
| ||||||||||||||||||||
Files in this Data Supplement:
Fig. S1. Specificity of anti-Cdc42 immunostaining. (A) MTLn3 cells were stimulated with EGF for 180 seconds, fixed and immunostained with anti-Cdc42 antibody preabsorbed with GST- or GST-Cdc42. (B) MTLn3 cells were transfected with control luciferase siRNA (upper panels) or with Cdc42 siRNA (lower panels). The cells were starved for 3 hours, stimulated without or with EGF for 3 minutes, fixed and immunostained with anti-Cdc42. Bar, 10 μm. (C). MTLn3 cells were stimulated for 60 seconds, fixed and stained with Rhodamine-Phalloidin (shown in red) and anti-Cdc42 (shown in green); separate and merged images are shown. Inset shows increased magnification. Bar, 10 μm.
Fig. S2. Expression of the eGFP-mero-ISO-IAA Cdc42 biosensor in MTLn3 cells. (A) Wild type MTLn3 cells microinjected with the Cdc42 biosensor were stimulated with EGF for 3 minutes. Cells were imaged in the GFP and I-SO channels. Images were pooled according to mean GFP fluorescence in the total cell area. The same cells were scored for mean Cdc42 activation ratio in the total cell area. The ratio was then plotted as a function of mean GFP fluorescence. Since the Cdc42 biosensor was microinjected, the cellular concentration of the biosensor (detected through measuring the mean GFP fluorescence/cell area) was variable. Correlation analysis of the GFP fluorescence to the mean Cdc42 activation ratio showed a decrease in ratio as the GFP level increased (Fig S1C). Therefore, in all the experiments we only analyzed cells that contained relatively low levels of the Cdc42 biosensor (GFP<700). (B) MTLn3 cells were starved for 3 hours and microinjected with the Cdc42 biosensor. After a 1 hour recovery, the cells were then stimulated with 5 nM EGF at 37°C, and fixed at 3 minutes. The Cdc42 activation ratio (I-SO:eGFP) increased from 1 in the inactive state to 3 in the GTP-loaded active Cdc42 state. Activation was localized within 2 μm of the protruding cell edge. Bar, 10 μm.
Fig. S3. Visualization of adhesive structures in Cdc42-knockdown cells. (A) MTLn3 cells were transfected with control (luciferase) or Cdc42 siRNA, stimulated with EGF and stained with anti-paxillin antibody. (B) Frequency histogram of the distance of adhesion clusters from the leading edge in control and Cdc42 siRNA-transfected cells (n=45 cells, ten measurements per cell per treatment). The mean distance (± s.e.m.) of the adhesion sites from the cell edge for control versus Cdc42 siRNA-treated cells is indicated. (C) Live imaging of control luciferase-siRNA-transfected or Cdc42-siRNA-transfected MTLn3 cells transfected with GFP-vinculin and protruding in serum. Bars, 10 μm.
Fig. S4. Cdc42 knockdown inhibits EGF-stimulated protrusion but not EGF receptor signaling or internalization. MTLn3 cells were transfected with control siRNA (luciferase) or with a pool of four Cdc42 siRNA oligos. (A) Representative western blot showing siRNA knockdown of Cdc42 (upper panel) as compared to Rac (lower panel). (B) MTLn3 cells were transfected with control luciferase or Cdc42 siRNA and starved for 3 hours. Cells were then imaged via time lapse during a 10-minute EGF stimulation with images collected every 20 seconds. The graph shows quantification of the surface area of EGF-stimulated luciferase control cells (open circles) and Cdc42 siRNA-treated cells (closed circles) normalized to unstimulated cells. Data are the mean ± s.e.m. from at least 30 cells. (C) MTLn3 cells were transfected with control (luciferase) or Cdc42 siRNA. After 48 hours, cells were stimulated with EGF for 1 minute, lysed and blotted for phosphorylated Erk1/2 (Thr202/ThrY204) and for Cdc42 expression (Cdc42). The graph shows quantification of the bands by densitometry, and is presented as fold increase as compared with control unstimulated cells. The data are the mean ± s.e.m. from threeexperiments. (D) Control luciferase−siRNA-treated or Cdc42-siRNA-treated cells were incubated with 125I-EGF for 10 minuts. At 2-minute-intervals, cells were rapidly shifted to 0°C, rinsed, and acid washed. The internalization-rate constant ke is defined as the slope of the plot of internalized ligand against the integral of surface-bound ligand. (E.) MTLn3 cells were transfected with control luciferase or Cdc42 siRNA and starved for 3 hours. Cells were then imaged via time lapse during a 10-minute insulin stimulation (1 μM) with images collected every 20 seconds. Maximal protrusion at 10 minutes for insulin-stimulated luciferase and Cdc42 siRNA-treated cells was measured with NIH Image, and was normalized to unstimulated cells. Data are the mean ± s.e.m. from at least 20 cells.
Fig. S5. Cdc42 knockdown inhibits Rac and PI3K. (A) MTLn3 cells were starved for 3 hours, stimulated with EGF for 3 minutes, fixed and immunostained with antibodies against Rac and Cdc42. The merged images show the co-localization of Rac and Cdc42 to the protruding edge of the cells in response to EGF. (B) Control and Cdc42 KnDn MTLn3 cells were stimulated with EGF for 1 minute or unstimulated and then assayed for Rac activation as in Fig. 1. (C) Control and Cdc42 KnDn MTLn3 cells were stimulated with EGF for 1 and 3 minutes or unstimulated, lysed and immunoblotted with antibodies against pS473Akt, total Akt, Cdc42 and actin. The bands were quantitated by densitometry and the data presented as p-Akt normalized to total Akt (see bar graph below). Data is the mean ± s.e.m. (n=3). (D) MTLn3 cells transfected with Luciferase or Rac siRNA were starved, stimulated with EGF for 1 and 3 minutes, fixed and immunostained with anti-WAVE2 antibody. Bars, 10 μm.
Fig. S6. Cdc42 and Ras regulate protrusion in EGF-stimulated cells through independent pathways. (A) MTLn3 cells transfected with control or Cdc442 siRNA were stimulated with EGF for 1 and 3 minutes or not (0 min), and then assayed for Ras activation using a GST-Ras-binding-domain pulldown. GTP-Ras. Total Ras, and Cdc42 were detected through western blot. The bands were quantified by densitometry and the bar graph shows data as fold increase in Ras activation as compared with control unstimulated cells. The data are the mean ± s.e.m. (n=3). (B) Control and K-Ras knockdown MTLn3 cells were stimulated with EGF for 1 and 3 minutes or not (0 min), and then assayed for Cdc42 activation as in Fig. 1. K-Ras knockdown was validated by quantitative RT-PCR, and was approximately 80% effective (data not shown). The bands were then quantified by densitometry and the bar graph shows data as fold increase in Cdc42 activation compared with control unstimulated cells. The data are the mean ± s.e.m. (n=3).
Movie 1. Cdc42 inhibition leads to a decrease in cell motility. Phase-contrast time-lapse images of MTLn3 cells transfected with control siRNA, in 5% in serum. Total time was 2 hours. Increments are 4 minutes per frame.
Movie S2. Cdc42 inhibition leads to a decrease in cell motility. Phase-contrast time-lapse images of MTLn3 cells transfected with Cdc42 siRNA, in 5% serum. Total time was 2 hours. Increments are 4 minutes per frame.
| ||||||||||||||||||||
