Loss of Tpm4.1 leads to disruption of cell-cell adhesions and invasive behavior in breast epithelial cells via increased Rac1 signaling

Abstract

Here we report the identification and characterization of a novel high molecular weight isoform of tropomyosin, Tpm4.1, expressed from the human TPM4 gene. Tpm4.1 expression is down-regulated in a subset of breast cancer cells compared with untransformed MCF10A breast epithelial cells and in highly metastatic breast cancer cell lines derived from poorly metastatic MDA-MD-231 cells. In addition, patients with invasive ductal breast carcinoma show decreased TPM4 expression compared with patients with ductal breast carcinoma in situ, and low TPM4 expression is associated with poor prognosis. Loss of Tpm4.1 using siRNA in MCF10A cells increases cell migration in wound-healing and Boyden chamber assays and invasion out of spheroids as well as disruption of cell-cell adhesions. Down-regulation of Tpm4.1 in MDA-MB-231 cells leads to disruption of actin organization and increased cell invasion and dissemination from spheroids into collagen gels. The down-regulation of Tpm4.1 induces Rac1-mediated alteration of myosin IIB localization, and pharmacologic inhibition of Rac1 or down-regulation of myosin IIB using siRNA inhibits the invasive phenotypes in MCF10A cells. Thus Tpm4.1 plays an important role in blocking invasive behaviors through Rac1-myosin IIB signaling and our findings suggest that decreased expression of Tpm4.1 might play a crucial role during tumor progression.

Keywords: TPM4; cell adhesions; invasion; migration; tropomyosin.

Figure 1. Expression of Tpm4.1 in breast cancer cells

(A) Detection of tropomyosin isoforms with indicated antibodies in an untransformed breast epithelial cell line, MCF10A, and breast cancer cell lines, MDA-MB-468, BT20, BT474, SKBR3, MCF7, T47D, and MDA-MB-231 by immunoblot. HeLa was used for a standard for the expression of Tpm4.1. (B) mRNA expression of Tpm4.1 in a normal breast cell line and breast cancer cell lines was measured by RT-PCR. (C) Schematic diagram of the exon organizations of Tpm4.1 and Tpm4.2 expressed from the TPM4 gene [5]. (D) Oncomine box plot of TPM4 expression levels in DCIS (n = 10) and IDC (n = 1,556) [38]. *p < 0.05. (E) Kaplan-Meier curve of relapse free survival of breast cancer patients with a compilation of all breast cancer patients independent of subcategories (n = 3,554). Data was obtained from the Kaplan-Meier plotter breast cancer database [39]. (F) Endogenous expressions of Tpm4.1 and Tpm4.2 in poorly metastatic parental MDA-MB-231 and its derivative highly metastatic cell lines, LM2, BoM2, and BrM2.

Figure 2. Loss of Tpm4 isoforms affects cell migration and invasion

(A) Migratory ability of MCF10A cells with siRNAs was analyzed by wound healing assay. Left panel: Representative images of wound healing assay. Scale bars, 500 μm. Right upper panel: The graph was from three independent experiments. *p < 0.05, **p < 0.005. Right lower panel: Representative images of immunoblot. (B–C) Upper panel: Analysis of migration (B) and invasion (C) of MCF10A cells with the indicated siRNAs in Boyden chamber assay with uncoated and Matrigel-coated membrane respectively. Each graph was from three independent experiments. *p < 0.05, **p < 0.005. Lower panel: Representative images of immunoblot.

Figure 3. Loss of Tpm4.1 impairs cell-cell adhesions

(A) Expressions of tropomyosin isoforms and cell-cell junction components were analyzed by immunoblot after siRNA transfection. (B) Time-lapse images of MCF10A cells with the indicated siRNAs in every 30 minutes. Arrows, dividing cells. Scale bars, 20 μm. (C) MCF10A cells with the indicated siRNAs were stained with the E-cadherin antibody and phalloidin to detect F-actin. Scale bars, 20 μm. (D) Rescue of cell-cell junctions by cDNA transfection following siRNA transfection. Immunoblot analysis was done after siRNA and cDNA plasmids transfection (e, endogenous tropomyosin isoforms). (E) MCF10A cells with the indicated siRNAs and cDNA plasmids were stained with the E-cadherin antibody and phalloidin to detect F-actin. Arrowheads indicate E-cadherin and actin bundles along cell-cell adhesions among the GFP-Tpm4.1-transfected cells. Scale bars, 20 μm.

Figure 4. Loss of Tpm4.1 decreases stress fibers and focal adhesions in MDA-MB-231

(A) MDA-MB-231 cell lysates were analyzed by immunoblot after siRNA transfection to examine silencing efficiency and phosphorylation of paxillin. (B) MDA-MB-231 cells with the indicated siRNAs were stained with the vinculin antibody and phalloidin. Scale bars, 20 μm. (C) Immunoblot analysis after siRNA and cDNA plasmids transfection (e, endogenous tropomyosin isoforms). (D) MDA-MB-231 cells with the indicated siRNAs and cDNA plasmids were stained with the vinculin antibody and phalloidin. (E) The cDNA plasmid-transfected cells were randomly captured and the sizes of the cells were measured using ImageJ software. The cell sizes were averaged as division of the measured area by the number of nucleus. n, nucleus. The graph was from three independent experiments. *p < 0.05, **p < 0.005

Figure 5. Localization of tropomyosin in MCF10A cells

(A) MCF10A cells in confluent condition were stained with the indicated tropomyosin antibody and phalloidin. Scale bars, 20 μm. (B) MCF10A cells were stained with δ/9d and E-cadherin antibodies and phalloidin. Scale bars, 20 μm. (C) After transfection of GFP and GFP-Tpm4.1, MCF10A cells were stained with E-cadherin antibody and phalloidin. Scale bars, 20 μm.

Figure 6. Tpm4.1 regulates cell spheroid aggregation and invasion

(A) Representative images of MCF10A spheroids with the indicated siRNA embedded in collagen I. Scale bars, 100 μm. (B) MDA-MB-231 cells with the indicated siRNA were grown in ultra-low attachment plate with Matrigel to form spheroids. Scale bars, 500 μm. (C) Spheroids of MDA-MB-231 cells were embedded in collagen I. Scale bars, 500 μm.

Figure 7. Depletion of Tpm4.1 induces Rac1 activation and it is responsible for increase in cell migration

(A) The level of GTP-bound Rac1 was measured by G-LISA Rac1 activity assay in MCF10A cells after siRNA transfection. The graph was from three independent experiments. *p < 0.05, **p < 0.005 (B) Left panel: Immunoblot analysis of Rac1 activity assay samples. Right upper panel: The intensity of p-MLC band was divided by that of total MLC band. The graph was from the samples of three independent experiments used in the analysis. *p < 0.05, **p < 0.005. (C) siRNA-transfected MCF10A cells were treated with 50 μM NSC23766 and migratory ability was analyzed by wound healing assay. The graph was from three independent experiments. *p < 0.05, **p < 0.005 (D) Immunoblot analysis of wound healing assay samples.

Figure 8. The Rac1 activation induces redistribution of myosin IIB and it induces the invasive phenotypes of Tpm4.1-silenced cells

(A) Immunofluorescence localization of myosin IIA and myosin IIB in MCF10A cells with the indicated siRNAs. Scale bars, 20 μm. (B) Immunoblot analysis after siRNA transfection. (C) Migratory ability of MCF10A cells with the indicated siRNAs was analyzed by wound healing assay. The graph was from three independent experiments. *p < 0.05, **p < 0.005 (D) MCF10A cells with the indicated siRNAs were stained with the E-cadherin antibody and phalloidin. Scale bars, 20 μm. (E) Immunofluorescence localization of myosin IIB with treatment of 50 μM NSC23766 for 1 hr in MCF10A cells with the indicated siRNAs.