Murine missing in metastasis (MIM) mediates cell polarity and regulates the motility response to growth factors

Abstract

Background: Missing in metastasis (MIM) is a member of the inverse BAR-domain protein family, and in vitro studies have implied MIM plays a role in deforming membrane curvature into filopodia-like protrusions and cell dynamics. Yet, the physiological role of the endogenous MIM in mammalian cells remains undefined.

Principal findings: We have examined mouse embryonic fibroblasts (MEFs) derived from mice in which the MIM locus was targeted by a gene trapping vector. MIM(-/-) MEFs showed a less polarized architecture characterized by smooth edges and fewer cell protrusions as compared to wild type cells, although the formation of filopodia-like microprotrusions appeared to be normal. Immunofluorescent staining further revealed that MIM(-/-) cells were partially impaired in the assembly of stress fibers and focal adhesions but were enriched with transverse actin filaments at the periphery. Poor assembly of stress fibers was apparently correlated with attenuation of the activity of Rho GTPases and partially relieved upon overexpressing of Myc-RhoA(Q63L), a constitutively activated RhoA mutant. MIM(-/-) cells were also spread less effectively than wild type cells during attachment to dishes and substratum. Upon treatment with PDGF MIM(-/-) cells developed more prominent dorsal ruffles along with increased Rac1 activity. Compared to wild type cells, MIM(-/-) cells had a slower motility in the presence of a low percentage of serum-containing medium but migrated normally upon adding growth factors such as 10% serum, PDGF or EGF. MIM(-/-) cells were also partially impaired in the internalization of transferrin, fluorescent dyes, foreign DNAs and PDGF receptor alpha. On the other hand, the level of tyrosine phosphorylation of PDGF receptors was more elevated in MIM depleted cells than wild type cells upon PDGF treatment.

Conclusions: Our data suggests that endogenous MIM protein regulates globally the cell architecture and endocytosis that ultimately influence a variety of cellular behaviors, including cell polarity, motility, receptor signaling and membrane ruffling.

Figure 1. Preparation of MIM knockout mice.

(A) schematic illustration of the genomic structure of the murine MIM gene. The trapping vector, which carries a gal-neo (geo) fusion gene, targeted at the position corresponding to the nucleotide 57,067 in the third intron. (B) Genotyping of MIM knockout mice by PCR with the primers flanking the trapping vector. The top band (625 bp) indicates the mutant allele, and the lower band (400 bp) indicates the wild-type allele. (C) Western blot analysis of MEFs using MIM antibody, which detected a MIM characteristic duplex of p115 and p100 in the lysate of the MIM^+/+ cells. The same membrane was re-blotted with cortactin antibody for the loading control (the lower panel). (D) Immunostaining of a MEF cell using MIM antibody. The arrow head indicates a representative lamellipodia staining. The nuclear staining was likely non-specific and observed with both MIM^−/− and MIM^+/+ cells. (E) Expression of several cortical proteins in MIM^+/+ and MIM^−/− MEFs.

Figure 2. MIM^−/− cells display a distinct morphology and altered actin cytoskeleton reorganization.

(A) MIM^+/+ (WT) MEFs (a, c, e, and g) and MIM^−/− (KO) cells (b, d, f and h) were grown in 10% serum-containing medium on fibronectin covered slips, inspected by phase-contrast view (a and b) or by confocal microscopy following co-staining with cortactin antibody (c and d) and phalloidin (e and f). The images c and e were merged as g, and the images d and f were merged as h. Scale bars: a and b, 50 µm; c–h, 10 µm. (B) Quantification of the percentage of cells bearing prominent body projections as revealed under phase-contrast microscopy. (C) Quantification of the cell surface perimeter measured by MetaMorph program. (D) Quantification of the percentage of cells with prominent cortactin-enriched lamellipodia. (E) Activation of Rho is elevated in MIM^−/− cells. Cells were arrested in 0.2% serum-containing medium and stimulated with 30 ng/ml PDGF for 10 min. The level of GTP-Rho was measured by pull-down assay as described in the Materials and Methods. As a control, cells grown in 10% serum were also analyzed in parallel. Quantification of Rho activation is presented on the right based on three experiments (mean ± SEM). All the p values (t-test) refer to the difference between MIM^+/+ and MIM^−/− cells.

Figure 3. MIM^−/− MEFs were impaired in spreading during cell attachment.

(A) Cells were trypsinized, plated on fibronectin-coated coverslips in a serum-containing medium, fixed at 30 min and 2 h and inspected by phase-contrast microscopy. An enlarged area in each 30-min image was shown in inset. (B) The number of cells with apparent cytoplasm extensions was countered at different times after plating. The data represents mean ± SEM (n = 3). The p value (Anova test) refers to the difference between MIM^−/− and MIM^+/+ cells during the course of attachment. (C) Cells at 2 h after plating were fixed, stained with phalloidin and scanned with confocal microscope with each optical section of 0.30 µm in thickness. The images corresponding to different sections were compiled by Zeiss LSM Image browser.

Figure 4. PDGF induces prominent dorsal ruffles in MIM^−/− MEFs.

(A) MIM^+/+ (a and b) and MIM^−/− (c and d) cells were arrested by incubating for 24 h in 0.2% serum-containing medium and then treated with PDGF for 10 min (b and d) followed by staining with phalloidin. The stained cells were inspected by epifluorescent microscopy. Scale bar: 50 µm. (B) Quantification of dorsal ruffles. Cells were treated with PDGF for the times as indicated. The number of cells showing large ruffling areas was counted. The data shown are the mean ± SEM based on four independent experiments. In each experiment 70 cells were analyzed. The p value was calculated by Anova test, referring to the difference between MIM^−/− and MIM^+/+ MEFs during the response to PDGF. (C) PDGF treated cells were analyzed for Rac1 activation by pull-down assay followed by Western blot using anti-Rac1 antibody. The Rac1 activation was quantified based on three independent experiments. (D) Quiescent MIM^−/− cells grown on fibronectin-coated coverslips were treated with and without Rac1 inhibitor NSC23766 at the concentration of 50 µM for 48 h. The treated cells were then stimulated with PDGF for 10 min, and the Rac1 activation was measured as above. (E) NSC23766 treated cells were also treated with PDGF, and the formation of dorsal ruffles (DR) was quantified based on three independent experiments.

Figure 5. MIM deficiency affects cell motility.

(A) Cells were grown in 10% serum-containing medium until confluence, then wounded by rubber policeman and incubated in medium containing either 10% or 0.2% of serum. The images showing the same wounded areas were taken at the beginning of incubation (0) and 22 h later. (B) Wound-healing rates were measured as described in the Materials and Methods. (C) Cells were arrested in 0.2% serum-containing medium and placed on the top chamber of Transwell plates in which the bottom chamber was filled with 0.2% serum medium or medium containing PDGF (30 ng/ml). The plates were incubated for 4 h, stained and photographed. (D) Quantification of the motility of cells in Transwell plates containing 0.2% serum, 30 ng/ml PDGF, 30 ng/ml EGF and 10% serum, respectively. All the presented values are the mean ± SEM of three independent experiments.

Figure 6. MIM plays a positive role in internalization of extracellular particles.

(A) MEFs were grown in medium supplemented with either 0.2% or 10% serum and incubated with biotin-labeled transferrin (Bio-Tfn) for 5 min. Uptake of Bio-Tfn was measured as described in the Materials and Methods. (B) Cells were treated with Alexa Fluor 647 and incubated for the times as indicated. Internalization of the fluorescent dye was measured by flow cytometry. (C) NIH3T3 cells infected by retroviruses encoding GFP-MIM or GFP only were incubated with Bio-Tfn for 5 min. The internalized Bio-Tfn was measured as described in (A). All the error bars represent the mean ± SEM of three independent experiments. P values (t test) refer to the difference between the samples as indicated. (D) Quiescent MEFs were stimulated with PDGF (50 ng/ml) for the times as indicated. The cell surface (s) and total (t) PDGFRα proteins were analyzed as described in the Materials and Methods. The data represents two-independent experiments. (E) Quiescent MEFs were stimulated with PDGF for 5 min. The total cell lysates were subjected to Western blot using anti-phosphotyrosine antibody (4G10).