|
|
|
|
|
|
|
|
|||||
| Home Help Feedback Subscriptions Archive Search Table of Contents | |||||
| ||||||||||||||||||||
Files in this Data Supplement:
Fig. S1. Calibration of the imaging system. (A) Recombinant EGFP was diluted at decreasing concentrations in saline solution. The micelles were prepared by dispersing the solution in mineral oil. Imaging was performed on the confocal microscope in carefully controlled conditions to ensure applicability of the calibration to the live cell experiments. The three panels show micelles at three different concentrations; quantification has been performed in the central part of the micelles to avoid spurious effects because of microlensing. Images are shown in false colours to enhance differences in brightness. Bar, 10 μm. (B) Calibration data for six different concentrations. Each point is the average of at least ten measures (30 at the three lower concentrations). These measures are relative to the FV 300 microscope. Each microscope used in the study required a specific calibration curve.
Fig. S2. ERK2-GFP localizes in the cytoplasm or nucleus according to its activation state. Starved fibroblasts were fixed in control conditions (A) or after 15 minutes in FGF4 (B). To demonstrate the nuclear exclusion or localization of pERK (red) and ERK-GFP (green) we stained the nuclei with the DNA stain Hoechst 33342 (blue; 10 minutes in 1 μg/ml solution). Cells were imaged by combining one and two photon excitation: ERK-GFP and pERK fluorescence was excited at visible wavelengths (488 and 543 nm, respectively), whereas Hoechst was excited at 870 nm. Calibration bar, 10 μm. (C) Immunoblotting against GFP on NIH-3T3 fibroblasts displays a band at the correct weight (69 kDa, ERK2 is 42 kDa and GFP is 27 kDa; green arrowhead) in cells transfected with the ERK2-GFP vector. Cells transfected with GFP only show a band at 27 kDa (black arrowhead). Non-transfected cells (wild type) do not show any signal. Control indicates cells transfected with an out-of-frame ERK fusion protein. (D) Immunoblotting demonstrates that the pERK antibody recognized both the WT ERK1/2 doublet (red arrowhead) and the fusion protein (green) after stimulation with 10% serum. Experimental conditions: N. Growth, cells kept in normal medium (10% serum); Starv, cells kept for 24 hours in 1% serum; Serum, cells stimulated for 15 minutes with 20% serum after starvation. (E) Stimulation for 15 minutes with FGF (80 ng/ml in all experiments) caused an increase in the pERK signal (red) and nuclear localization of the fusion protein (green). Note that the excitation used to image the pERK signal is eight times brighter in the upper panel. Bar, 20 μm. (F) Distribution of the cumulative probability of the concentration index measured in cells transfected with ERK2-GFP and imaged after starvation (red curve) or after stimulation with 10% serum (P≤0.001, Kolmogorov-Smirnov test). The cumulative probability is integral to the probability distribution and it is very effective in displaying the differences between two populations, regardless of their normality. (G) The N-terminal and C-terminal fusion proteins have similar subcellular distribution in starved cells and after stimulation with 10% serum. Bar, 10 μm.
Fig. S3. Demonstration of the nuclear accumulation of ERK2-GFP in two additional cell types. (A) NIH L1 preadipocytes were starved for 24 hours in 1% serum before stimulation with 10% serum. Bar, 10 μm. NIH-L1 preadipocytes were cultured according to ATCC protocols. (B) Primary fibroblasts were starved 24 hours prior to the experiment and stimulation with FGF4. In both panels time 0 indicates the beginning of stimulation. Primary fibroblasts were dissociated from the tail of an adult mouse and transfected at passage 3. Both cell types were transfected and imaged as described for NIH-3T3.
Fig. S4. Regulation of ERK2 activation. (A) ERK2-GFP distribution in the presence of the MEK inhibitor U0126 (20 μM, stock diluted in DMSO) in response to stimulation and after the washout of the blocker in the presence of serum or FGF. (B) Time course of the concentration index averaged on nine cells stimulated with either 10% serum or FGF (80 ng/ml). Treatment with control vehicle only (saline + 0.1% DMSO) did not have any effect on either the cell viability or on ERK-GFP localization (not shown). Bar, 10 μm. (C) Inhibition of phosphatase caused ERK1/2 phosphorylation and translocation. Starved cells have been treated with sodium orthovanadate for 40 minutes before fixation and immunofluorescence with the phosphospecific antibody. Representative images of pERK distribution in starved cells and during treatment with phosphatase inhibitor. Bar, 10 m. (D) Cumulative probability of the translocation coefficient in the two conditions measured after immunofluorescence with the pERK antibody (P≤0.001, Kolmogorov-Smirnov test).
Fig. S5. Estimate of ERK2 phosphorylation after stimulation. (A) Description of the experimental scheme. In the stimulated cells there is a mix of phosphorylated (green) and non-phosphorylated ERK1 and ERK2 (red) representing the total amount of ERK. To evaluate the ratio between phosphorylated and non-phosphorylated ERK upon serum stimulation, we performed a two-step immunoprecipitation experiment in the NIH-3T3 cell line. In the first step the lysate of stimulated cells (15 minutes, serum 10%) was incubated with a pERK antibody conjugated with sepharose. After incubation the lysate was centrifuged and the pellet containing pERK was collected. In the second step the pERK-immunodepleted supernatant (containing only ERK) was incubated with sepharose conjugated to antibodies against ERK. After incubation the lysate was centrifuged and the pellet containing ERK was collected. The obtained samples were run on 10% SDS-PAGE and subjected to western blotting. (B) Anti-ERK western blot. The left lane shows the pERK immunoprecipitation, the right lane shows the pERK-immunodepleted fraction: this lane shows that the supernatant contained a small but significant residual fraction (11%±3 of total, n=3) of ERK2 that was not precipitated by the p-specific antibody. The band at approximately 50 kDa corresponds to the heavy chain of immunoglobulins. (C) To verify that this band did not contain any pERK, we stripped the membrane and reprobed it with the anti-pERK antibody. (D) Test of linearity of gel quantification. A calibration standard was prepared by loading a gel with an increasing amount of total protein. The gel was probed with the pERK antibody and acquired together with the immunoprecipitation gels. (E) Gels were quantified by measuring the lane transmission T averaged in a fixed area centred on the band, and linearity was checked by plotting the transmission of the calibration bands against the total protein content. The plot shows that, in the range of band density obtained in our experiments (T≥0.4), transmission was linearly proportional to protein content (R2=0.96). All experiments were performed in triplicate, and the gels were quantified with Image J.
Fig. S6. (A) Phosphorylation-dependent regulation of ERK2 shuttling in NIH-L1 preadipocytes and the primary fibroblast ERK pathway was activated with 10% serum (L1) or FGF (fibroblasts) and imaging was completed within 45 minutes from stimulus onset. In both cell types the turnover of ERK2-GFP was strongly accelerated by activation (**P≤0.001; *P≤0.02). (B) For weakly expressing cells, the rate of ERK2 exchange is independent of the concentration of ERK2-GFP. The panel shows typical recordings from starved cells: fluorescence was converted in average nuclear concentration of ERK2-GFP according to the calibration (supplementary material Fig. S1). In a logarithmic plot exponentials of similar τ are parallel. (C) The time constant of recovery is independent of the average concentration of ERK2-GFP.
Fig. S7. ERK2-GFP is extruded from the nucleus by the operation of CRM1-mediated export. (A) Imaging of a 3T3 cell after starvation in 1% serum and during treatment with the CRM1 blocker leptomycin B (5 ng/ml in ethanol). (B) Time course of the translocation index averaged on seven cells. Leptomycin caused a gradual accumulation of ERK2-GFP beginning after approximately 20 minutes from treatment onset. Treatment with vehicle only (0.07% methanol in saline) had no effect (data not shown). (C) Combined effect of leptomycin, FGF and U0126 measured in 12 cells. Starved cells were pretreated with leptomycin, then FGF4 was added and finally U0126, as indicated by the bars. The red curve is an exponential fit used to evaluate the rate of loss of nuclear ERK2-GFP (τ=4.47±0.20 minutes, R2=0.98) during inhibition of the ERK pathway and CRM1 export.
Movie 1. Translocation of ERK2-GFP in the nucleus after stimulation with FGF in a transfected 3T3 fibroblast.
Movie 2. A fibroblast transfected with ERK2-GFP was stimulated initially with FGF. When translocation reached a peak, the upstream inhibitor U0126 was added to the bath, causing the immediate loss of nuclear fluorescence.
| ||||||||||||||||||||
