A novel role for atypical MAPK kinase ERK3 in regulating breast cancer cell morphology and migration

Abstract

ERK3 is an atypical Mitogen-activated protein kinase (MAPK6). Despite the fact that the Erk3 gene was originally identified in 1991, its function is still unknown. MK5 (MAP kinase- activated protein kinase 5) also called PRAK is the only known substrate for ERK3. Recently, it was found that group I p21 protein activated kinases (PAKs) are critical effectors of ERK3. PAKs link Rho family of GTPases to actin cytoskeletal dynamics and are known to be involved in the regulation of cell adhesion and migration. In this study we demonstrate that ERK3 protein levels are elevated as MDA-MB-231 breast cancer cells adhere to collagen I which is concomitant with changes in cellular morphology where cells become less well spread following nascent adhesion formation. During this early cellular adhesion event we observe that the cells retain protrusive activity while reducing overall cellular area. Interestingly exogenous expression of ERK3 delivers a comparable reduction in cell spread area, while depletion of ERK3 expression increases cell spread area. Importantly, we have detected a novel specific endogenous ERK3 localization at the cell periphery. Furthermore we find that ERK3 overexpressing cells exhibit a rounded morphology and increased cell migration speed. Surprisingly, exogenous expression of a kinase inactive mutant of ERK3 phenocopies ERK3 overexpression, suggesting a novel kinase independent function for ERK3. Taken together our data suggest that as cells initiate adhesion to matrix increasing levels of ERK3 at the cell periphery are required to orchestrate cell morphology changes which can then drive migratory behavior.

Keywords: ERK3; actin filaments; cell adhesion; cell motility; cell protrusion; cell-cell adhesion; mitogen-activated protein kinase 6.

Figure 1.

MDA-MB-231 cells show a significant decrease of relative spread area after 8 hours of seeding. (A) MDA-MB-231 cells were seeded onto collagen I coverslips for the following time course 2, 4, 6, 8 hours and were fixed and stained with TRITC-phalloidin to show F-actin and Dapi. Images were taken by confocal microscopy. (B) Cell spread area, perimeter and elongation ratio were calculated using ImageJ (NIH) software. The results shown are mean ± s.e.m of over 30 cells from each population in each of three separate experiments. Statistical significance was analyzed using the student test, *P ≤ 0 .05 and **P ≤ 0 .005. Scale bar: 10 μm.

Figure 2.

Cell protrusion activity increases as MDA-MB-231 cells adhere to collagen. (A) Representative images of movie stills (t=0 and t=192 mins) from a movie of MDA-MB-231 imaged following plating on collagen I. Phase contrast to reveal cell outline detection (see materials and methods) and red/green pseudocolour to reveal areas of protrusion (green) and retraction (red). (B) Quantification of mean protrusion per unit perimeter over time. N=2 movies (20 cells). Images were quantified using in house Mathematica^TM software.

Figure 3.

ERK3 increasing level is coincided with the significant decrease of relative spread area. (A) MDA-MB-231 cells were seeded onto collagen I coated six-well plate and harvested at the following time course 2, 4, 6, 8 hours and probe for endogenous ERK3. The figure shown is a representative of three separate experiments. (B) Relative intensity was calculated for the time course blot 2, 4, 6, 8 hours and analyzed using student test. The figure shown is a representative of three separate experiments. (C) Overexpressed ERK3 localizes mainly in the nucleus and at the plasma membrane of the cell. MDA-MB-231 cells were transfected with MYC-ERK3 for 24 hours, fixed and stained with TRITC-phalloidin for F-actin (red) and MYC tag as required (green). (D) MDA-MB-231 cells were transfected with MYC-ERK3 or (E) MYC-ERK3S189A for 24 hours. The transfected cells were then seeded onto collagen I plates for the following time course 2, 4, 6, 8 hours and were fixed and stained with TRITC-phalloidin to show F-actin and Dapi. For ERK3 detection (green), ERK3 monoclonal antibody was used followed by Alexa Flour 488 anti-mouse. Confocal images were taken. Scale bar: 10 μm.

Figure 4.

Over expression of ERK3 in different breast cell lines induces a reduction in spread area and elongation ratio. (A) MDA-MB-231 cells were transfected with GFP control vector or GFP-ERK3 for 24 hours, the cells were than fixed and stained with TRITC-phalloidin for F-actin and Dapi. (B) MDA-MB-231 Cells were transfected with MYC-ERK3 for 24 hours, the cells were than fixed and stained with TRITC-phalloidin for F-actin, Dapi and MYC tag as required. (C) MCF-7 cells were transfected with MYC-ERK3 for 24 hours, the cells were than fixed and stained with TRITC-phalloidin for F-actin, Dapi and MYC tag as required. All relative spread area and elongation ratio were calculated using ImageJ (NIH) software. The results shown are mean ± s.e.m of over 30 cells from each population in each of three separated experiments. Statistical significance values were calculated using Student's t-test, *P ≤ 0.05, **P ≤ 0 .005 and ***P ≤ 0.0005.

Figure 5.

ERK3 has induced the MDA-MB-231 cells mobility and the actin cytoskeleton rearrangement. (A) MDA-MB-231 cells were transfected with GFP control vector or GFP-ERK3 for 24 hours. Cell images were collected using a Sensicam (PCO Cook) CCD camera, taking a frame every 5 minutes for 16 hours from each of the six wells using AQM acquisition software. Subsequently, cells were tracked using AQM tracker. Over 10 cells were tracked over six separate films from three separate experiments for each experimental condition. Mathematical analysis was then carried out using Mathematica 6.0™ workbooks (ANOVA). Mean track speeds for each condition were compared using the Student's T-test, **P ≤ 0 .005. (B) MDA-MB-231 cells were transfected with MYC-ERK3 for 24 hours, the cells were than fixed and stained with TRITC-phalloidin for F-actin, Dapi and MYC tag as required. Images of F-Actin were taken using Time-lapse microscopy. An increase of ERK3 level has an effect in F-Actin organization.

Figure 6.

ERK3 transfected cells tend to dissociate from neighboring cells. (A) MCF-7 cells were transfected with MYC-ERK3 for 24 hours, the cells were then fixed and stained with TRITC-phalloidin for F-actin, Dapi and MYC tag as required. ERK3 transfected cells lose their contact with other cells. (B) The percentage of MCF-7 cells that had been detected to be dissociated. 41.7% of cells were totally isolated, 46.6% cells were partially separated and 11.6% were within the colony. Scale bar: 10 μm.

Figure 7.

The depletion of ERK3 protein induces an increasing in spread area and elongation ratio. (A) ERK3 was knocked down in MDA-MB-231 cells. For control LUC protein was knocked down (see material and methods) (B) MDA-MB-231 cells were seeded onto collagen I coverslips for the following time course 2, 4, 6, 8 hours and were fixed and stained with TRITC-phalloidin to show F-actin and Dapi. Cells were imaged by Time-lapse microscopy. (C) Cell spread area, perimeter and elongation ratio were calculated using ImageJ (NIH) software. The results shown are mean ± s.e.m of over 30 cells from each population in three separate experiments. Statistical significance was analyzed using the student test, *P ≤ 0 .05 and **P ≤ 0 .005. Scale bar: 10 μm. (D) LUCKD, ERK3KD and ERK3KD cells transfected to express Flag-zfERK3 (Flag-zfERK3-ERK3KD) were seeded onto collagen I coverslips for 8 hours, fixed and stained with TRITC-phalloidin to show F-actin, Dapi and Flag tag as required. (E) Cell spread area and elongation ratio of LUCKD, ERK3KD and Flag-zfERK3-ERK3KD were calculated using ImageJ (NIH) software. The results shown are mean ± s.e.m of over 30 cells from each population in three separate experiments. Statistical significance was analyzed using the student test, *P ≤ 0 .05 and ***P ≤ 0 .0005. Scale bar: 10μm. (F) LUCKD and ERK3KD cells were seeded on collagen I wells and cell images collected for 16 hours using AQM acquisition software. Cell track plots for LUCKD and ERK3KD cells with all tracked plotted from 0,0 are illustrated (G) Individual cells were tracked and the mean migration speed and persistence of direction was calculated using in house Mathematica^TM software.

Figure 8.

ERK3 kinase activity has no effect on MDA-MB-231 cell morphology alteration. MDA-MB-231 cells were transfected with either MYC-ERK3 or a Kinase dead mutant MYC-ERK3 (MYC-ERK3D171A). After 24 hours, the cells were fixed and stained with TRITC-phalloidin for F-actin, Dapi and MYC tag as required. Cell spread area and elongation ratio were calculated using ImageJ (NIH) software. The results shown are mean ± s.e.m of over 30 cells from each population in each of three separate experiments. Statistical significance was analyzed using the student test, *P ≤ 0 .05 and **P ≤ 0 .005 and ***P ≤ 0 .0005.

Figure 9.

ERK3 activity during nascent adhesion. As cells adhere to collagen I, Rac is activated, activated Rac binds to PAK1 and induces phosphorylation of ERK3 at serine189 – phosphorylation leads to peripheral localization of ERK3 (green). Phosphorylation also leads to interaction with MK5 although this may not be part of the ERK3 morphological response. Localization of ERK3 at the cell periphery is required for the morphological changes that occur during nascent adhesion.