Figure 1.
PP1 loss impairs electrotaxis in HeLa cells.
A. Treatment of parental HeLa Tet-Off (HTO) cells with siRNA strongly depletes PP1 levels 48 h post transfection. Endogenous PP1 levels were visualized with PP1 antibodies that recognize all isoforms. B. Plot diagrams show that loss of PP1 impairs the ability of cells to migrate towards the cathode. Each line represents the migration trajectory of a single cell. The starting point for each cell migration track is at the origin. Cell tracks with end positions to the right appear in red (“C”, cathode) and those to the left appear in black (“A”, anode). EF-untreated cells were assayed as controls. Control siRNA cells migrate strongly towards the cathode; PP1 siRNA treated cells are unable to migrate in response to a DC EF. Scales show distance migrated in µm. C. PP1 depletion strongly reduces distance migrated, speed, and directedness in response to physiological DC EF. Error bars are S.E.M. p values for significant differences in distance, speed and directedness are shown. D. Localization of endogenous PP1 and distribution of filamentous-actin in control and PP1 depleted cells treated with DC EF. Endogenous PP1 levels were visualized with PP1 antibodies that recognize all isoforms (green) and polymerised actin was detected using rhodamine phalloidin (red). The nuclei have been stained with DAPI (blue). Arrows mark cells with a strong decrease in PP1 levels which correlate with defects in the formation of actin rich protrusions. Representative images are shown. Scale bar is 50 µm. E. Numbers of cells with filopodia were quantified by counting 100 cells. Error bars are S.E.M. p values for significant differences are shown. Images show a detail of cell protrusions in control siRNA and PP1 siRNA cells. Arrows mark numerous filopodia in control cells and outline areas with a major lack of filopodia at the cell edges in PP1 siRNA cells.
Figure 2.
Loss of the PP1 interactor NIPP1 impairs the electrotactic response of HeLa cells.
A. Treatment of parental HeLa Tet-Off (HTO) cells with siRNA strongly depletes NIPP1 levels 48 h post transfection. Cell lysates were analysed by SDS/PAGE and immunoblotting. Bands corresponding to all PP1 isoforms were detected and GAPDH was used as loading control. B. Plot diagrams show that loss of NIPP1 impairs the ability of cells to migrate towards the cathode. Control siRNA cells migrate strongly towards the cathode; NIPP1 siRNA treated cells show a much reduced cathodal response. Scales show distance migrated in µm. Scales are different between diagrams in order to include the tracks of every cell assayed. C. NIPP1 depletion strongly reduces distance migrated, speed, and directedness in response to physiological DC EF. Data are from at least three experiments. Error bars are S.E.M. p values for significant differences in distance, speed and directedness are shown. D. Localization of endogenous NIPP1 and distribution of filamentous-actin in control and NIPP1 depleted cells treated with DC EF. Endogenous NIPP1 levels were recognized with a rabbit anti-NIPP1 antibody (green) and polymerised actin was detected using rhodamine phalloidin (red). Nuclei are stained with DAPI (blue). NIPP1 localizes to the nucleus in EF-treated and untreated cells and its levels are depleted by siRNA. Scale bar is 50 µm. E. Numbers of cells with filopodia were quantified by counting 100 cells. Error bars are S.E.M. p values for significant differences are shown. Images show a detail of cell protrusions in control siRNA and NIPP1si RNA cells. Arrows mark numerous filopodia in control cells and outline areas with a major lack of filopodia at the cell edges in NIPP1 siRNA cells.
Figure 3.
Loss of the PP1 interactor NIPP1 impairs the electrotactic response of PC-3-M cells.
A. Treatment of PC-3-M cells with IPTG induces NIPP1 depletion. Cell lysates were analysed by SDS/PAGE and immunoblotting. Bands corresponding to the PP1 isoforms were detected and GAPDH was used as loading control. B. Plot diagrams show that loss of NIPP1 impairs the ability of PC-3-M cells to migrate anodally. Migration trajectories were tracked for three hours. The starting point for each cell migration track is at the origin. Cell tracks with end positions to the right appear in red and those to the left appear in black. Cathode is marked as “C” and anode as “A” when a DC EF is applied to cells. Control scrambled PC-3-M cells migrate strongly anodally (negative directedness value); cells expressing shRNA targeting NIPP1 show a much reduced anodal response. Scales show distance migrated in µm. Scales are different between diagrams in order to include the tracks of every cell assayed. C. NIPP1 depletion strongly reduced distance migrated and directedness in response to physiological DC EF. Data are from at least three experiments. Error bars are S.E.M. p values for significant differences in distance, speed and directedness are shown.
Figure 4.
NIPP1 expression in HTO cells and control of EF-induced directional migration via its binding to PP1.
A. Cartoon of endogenous NIPP1 and the different FLAG-tagged NIPP1 variants expressed after doxycyclin removal in HeLa Tet-Off (HTO) cell lines. All three NIPP1 variants have a forkhead associated domain (FHA). The consensus PP1-binding sequence, RVTF in W.T-NIPP1 has been mutated to RATA in the FLAG-mNIPP1 variant. The C-terminal auto-inhibitory (ID) domain is not included in the FLAG tagged ΔC-NIPP1 protein, resulting in the expression of a constitutively active PP1/NIPP1 holoenzyme. B. Expression of NIPP1 variants confirmed by Western blotting after removal of doxycyclin. Cell lysates were analysed by SDS/PAGE and immunoblotting. Bands corresponding to the PP1 isoforms were detected and GAPDH was used as loading control. C. NIPP1 expression and localization in the HTO cells was confirmed by ICC in EF-treated and untreated HTO cells. Anti-FLAG antibody and rhodamine phalloidin have been used to detect the FLAG-tagged NIPP1 variants (green) and F-actin (red). The nuclei were stained with DAPI. Overexpressed NIPP1 localizes to the nucleus in EF-treated and untreated cells. Scale bar is 50 µm. Plot diagrams show that an EF of physiological strength (200 mV/mm) induced distinct migratory responses in the HTO cells expressing different NIPP1 variants. EF-untreated HTO cells are shown as controls. Migration trajectories were tracked for three hours in the absence and presence of EF. Each cell’s position at 0 h is positioned at the origin (0, 0). Cells whose end position is to the right are coloured red and those to the left appear in black. Cathode is marked as “C” and anode is marked as “A” when DC EF is applied to cells. Scales show distance migrated in µm. Note that scales are different among diagrams in order to include the tracks of every cell assayed.
Figure 5.
Centrosome polarization in the HTO cells mirrors directional migration in EF.
A. A DC EF polarizes centrosomes to the cathode in parental HTO cells as seen by counting the cells in 5 regions, top (t), Right (cathode in EF-treated cells), bottom (b), left (anode in EF-treated cells) and centre of the nucleus (marked as a white dot). B. Parental cells and mNIPP1 cells position their centrosomes cathodally in an EF, whereas overexpression of W.T-NIPP1 disrupts cathodal centrosomal polarisation and overexpression of ΔC-NIPP1 shifts cathodal polarisation of centrosomes to anodal. 100 cells were counted in each case and results are expressed as percentages.
Table 1.
List of genes from the Cdc42 pathway that are significantly upregulated by the overexpression of W.T-NIPP1 (WT) or ΔC-NIPP1 (ΔC), but not by mNIPP1 (m), in the HTO cells and compared to parental HTO cells.
Figure 6.
Effect of pharmacological inhibition of Cdc42-GTPase on the HTO cells.
A. Effect of ML141 on Cdc42 GTPase activity in unstimulated cells cultured in complete medium and in EF-stimulated HTO cells overexpressing the FLAG-NIPP1 protein variants. Levels of Cdc42-GTP determined by G-LISA in parental, W.T-NIPP1, ΔC-NIPP1 and mRATA cells in the absence or presence of DC EF and in cells pre-treated with 10 µM of ML141 before electrical stimulation. p values parental to W.T-NIPP1 and parental to ΔC-NIPP1 in complete medium were 0.1 and 0.01, respectively; p values comparing samples in the absence and presence of ML141 were in all cases <0.01. B. Cdc42 inhibition rescues cathodal polarisation and this correlates with centrosome positioning. Directedness values for the migration of EF-treated cells incubated with ML141. Cdc42 inhibition rescues the positive cell directedness decreased by W.T-NIPP1 overexpression. The strongly negative directedness value displayed by ΔC-NIPP1 cells becomes closer to 0 when cells are pretreated with Cdc42 inhibitor. For simplification directedness values in the absence of EF of the parental, W.T-NIPP1, ΔC-NIPP1, and mNIPP1 with and without ML141 have not been included in the diagram. These were, without ML141, −0.07±0.04; 0.05±0.09; −0.08±0.05 and −0.01±0.04, respectively; with ML141 were −0.07±0.04; 0.09±0.05; −0.07±0.05 and −0.01±0.04, respectively. In the absence of EF values were in all cases very close to 0 and differences between the four lines were not statistically significant in any of the cases. Data was quantified from at least three experiments. Error bars are S.E.M. p values for significant differences in directedness are shown. Polarisation index of centrosomes calculated as explained in materials and methods. Polarisation index of W.T-NIPP1 and ΔC-NIPP1 cells becomes similar to the polarisation index of parental cells when cells are treated with the Cdc42 inhibitor ML141.
Figure 7.
Cartoon showing the basic organization of the cervical epithelium and a mechanistic model to explain how PP1/NIPP1 may contribute to invasiveness of tumour cells.
Cervical and vaginal epithelia have lumen potentials of about −25 to −50 mV [65], [66]. Such a lumen potential would correspond to a transepithelial voltage gradients of 1.7 V/cm (170 mV/mm). In these electrophysiological conditions cervical epithelial cells would migrate towards the lumen as they turn over the epithelial lining layer (green arrow). Upregulation of NIPP1 and its recruitment to PP1 would reverse migration into the lumen, encouraging invasion of the surrounding tissue (red arrow).