JCS02707 Supplementary Material

Files in this Data Supplement:

• Supplemental Figure 1 -

  Fig. S1. Validation of the microsopical setup as a semiquantitative approach for measuring ErbB2 internalization and cleavage. (A) Diagram showing the dissociation rate of Sc08 and GaM-488 from fixed SK-BR-3 cells. Cells were trypsinized lightly and fixed in 2% PFA, washed by centrifugation at 16,000 g for 1 minute, blocked in 5% NGS in PBS, incubated for 1 hour with Sc08 (1:100) in 5% NGS, washed three times in PBS, incubated with GaM-488 for 30 minutes, and washed three times in PBS. Right after the last wash the cells were resuspended in PBS and the fluorescence intensity measured continually for almost 1 hour in a FACSAria (BD-biosciences). Red dots are individual cells while the black points are average fluorescence for 5 minute intervals. Error bars represent the standard deviation of each 5 minute interval. (B) Diagram showing the amount of ErbB2_surface in SK-BR-3 cells after geldanamycin-stimulation. Geldanamycin-induced decrease in ErbB2_surface labeling seen over time in the confocal microscope was similar to that obtained when analyzing ErbB2 surface levels using FACS as in A, suggesting that the microscopical setup was indeed quantitative. Cells were incubated with 3 µM Geldanamycin for the indicated time period, fixed, and labeled for ErbB2_surface. (C) One potential pitfall when using r² values as measures for internalization and cleavage is that high amounts of internalization (or cleavage) could result in a high r² value, because bleed-through could cause a high correlation between the ErbB2_surface  and ErbB2_total  signals of internalized ErbB2 as well – although the ErbB2_surface / ErbB2_total  ratio would be lower. In order to estimate the impact from this, we simulated the r²_i values with various amounts of ErbB2 internalized. The graph presents the Pearson’s correlation coefficients (r², y-axis) in relation to the amount of colocalizing fluorescence (x-axis, assumed to be similar to the fraction of the ErbB2 population present on the cell surface). Three different scenarios were simulated: simple colocalization (squares), a high level of ‘shine through’ from other focal planes (diamonds), and both ‘shine through’ and a high level of bleedthrough from the total labeling to the surface labeling (triangles). It was evident that a non-linear relationship existed between the amount of ErbB2 internalized and the drop in r²_i, thus internalization of 20% of the ErbB2 would yield an r²_i value of ~0.5. If a 10% bleed-through of the ErbB2_surface into the ErbB2_total signal (which is a lot more than we observe) as well as 25% non-focal ‘shine through’ from other focal planes were included in the simulation, an increase in r²_i could be observed when almost all ErbB2 was internalized. However, in worst case this required more than 95% of ErbB2 to be internalized before it would affect the r²_i: a situation that in practice never occurred in the cells observed. Each datapoint is based on 1000 simulations each containing 1000 virtual pixels. (D) Another potential pitfall of the method is that the degradation of ErbB2 in response to Geldanamycin would reduce the fluorescence at later time points resulting in a worse signal-to-noise ratio. Reduced signal-to-noise ratio would itself cause decreased r² values independently of internalization or cleavage. In order to address the extent of this problem, we analyzed the relation between ErbB2_total average signal intensity and r²_i in an unstimulated cell population. Untreated SK-BR-3 cells were labeled for ErbB2_surface and ErbB2_total as described in Figure 2A, and the r²_i values between ErbB2_surface and ErbB2_total were obtained for each cell. Although the r²_i values dropped at average fluorescence intensities below 250 greyscale units, the r²_i values were almost unaffected at medium-low to high average intensity levels. Accordingly, cells with average intensities below 300 greyscale units were excluded from all analysis and presentation.

• Supplemental Figure 2 -

  Fig. S2. Confirmation in T47D cells of the effect of geldanamycin on internalization of ErbB2. Confocal images of T47D cells stained for ErbB2 as described in Fig. 1A. (A) Control T47D cells. (B) The cells are stimulated with 3 µM geldanamycin for 2 hours at 37°C. Note that geldanamycin also stimulates internalization of ErbB2 in T47D cells. (C) Treatment with 200 nM bafilomycin for 3 hours at 37°C. (D) Pretreatment of cells with 200nM bafilomycin for 1 hour before 3µM geldanamycin is added and incubated for another 2 hours at 37°C. Note that there is an increase in the intracellular content of ErbB2 when the lysosomes are inhibited with bafilomycin. (E) The cells are treated with 10 µM lactacystin for 3 hours at 37°C. (F) Pretreatment of cells with 10 µM lactacystin for 1 hour before 3 µM geldanamycin is added and incubated for another 2 hours at 37°C. Note that lactacystin inhibits internalization of ErbB2 after geldanamycin treatment. (G) The cells are treated with 10 µM monensin for 2 hours at 37°C. (H) Cells are treated with both 10 µM monensin and 3 µM geldanamycin for 2 hours at 37°C. It is seen that ErbB2 is internalized only after geldanamycin treatment. Bars, 20 µm.