Active Fraction from Embryo Fish Extracts Induces Reversion of the Malignant Invasive Phenotype in Breast Cancer through Down-regulation of TCTP and Modulation of E-cadherin/β-catenin Pathway

Abstract

Some yet unidentified factors released by both oocyte and embryonic microenvironments demonstrated to be non-permissive for tumor development and display the remarkable ability to foster cell/tissue reprogramming, thus ultimately reversing the malignant phenotype. In the present study we observed how molecular factors extracted from Zebrafish embryos during specific developmental phases (20 somites) significantly antagonize proliferation of breast cancer cells, while reversing a number of prominent aspects of malignancy. Embryo extracts reduce cell proliferation, enhance apoptosis, and dramatically inhibit both invasiveness and migrating capabilities of cancer cells. Counteracting the invasive phenotype is a relevant issue in controlling tumor spreading and metastasis. Moreover, such effect is not limited to cancerous cells as embryo extracts were also effective in inhibiting migration and invasiveness displayed by normal breast cells undergoing epithelial-mesenchymal transition upon TGF-β1 stimulation. The reversion program involves the modulation of E-cadherin/β-catenin pathway, cytoskeleton remodeling with dramatic reduction in vinculin, as well as downregulation of TCTP and the concomitant increase in p53 levels. Our findings highlight that-contrary to the prevailing current "dogma", which posits that neoplastic cells are irreversibly "committed"-the malignant phenotype can ultimately be "reversed", at least partially, in response to environmental morphogenetic influences.

Keywords: E-cadherin/β-catenin; Embryo fish extract; TCTP; Tumor Reversion; cytoskeleton; p53.

Figure 1

Effect of Zebrafish embryo F6 fraction on apoptosis of MDA-MB-231 (a) and MCF-7 (b) cells after 72 h of treatment with F6 at 0.3 and 3 μg/mL. Histograms showing the percentage of apoptotic cells; each column represents the mean value ± SD of four independent experiments. * p < 0.05 versus ctrl by ANOVA followed by Bonferroni post-test.

Figure 2

Effect of 5FU, 5FU+F6, and F6 on proliferation of MDA-MB-231 (a) and MCF-7 (b) cells. Cell proliferation was determined after 24 h of treatment by cell count assays performed by a particle count and size analyzer. Values, expressed as fold increase of control value considered as 1, are means of three independent experiments performed in triplicate, with SD represented by vertical bars. * p < 0.05 versus ctrl by ANOVA followed by Bonferroni post-test.

Figure 3

Effect of 5FU, 5FU+F6, and F6 on invasion in MDA-MB-231 cells. Transwell invasion assay (a,b) and urokinase plasminogen activator (uPA) levels (c) was performed in MDA-MB-231 cells untreated (ctrl) and treated with 5FU, 5FU+F6, and F6 for 24h. Values, expressed as fold increase of control value considered as 1, are means of three independent experiments performed in duplicate, with SD represented by vertical bars. * p < 0.05; *** p < 0.001 versus ctrl; ^# p < 0.05 versus 5FU by ANOVA followed by Bonferroni post-test. Images were obtained by optical microscopy, with 100× magnification.

Figure 4

Effect of 5FU, 5FU+F6, and F6 on migration in MDA-MB-231 cells. Transwell migration assay (a,b) was performed in MDA-MB-231 cells untreated (ctrl) and treated with 5FU, 5FU+F6, and F6 for 24 h. Values, expressed as fold increase of control value considered as 1, are means of three independent experiments performed in duplicate, with SD represented by vertical bars. * p < 0.05; ** p < 0.01; *** p < 0.001 versus ctrl; ^# p < 0.05 versus 5FU by ANOVA followed by Bonferroni post-test. Images were obtained by optical microscopy, with 100× magnification.

Figure 5

Effect of 5FU, 5FU+F6, and F6 on invasion (a) and migration (b) in MCF-10A cells. Transwell assays were performed in MCF-10A cells untreated (ctrl), and pre-treated with TGF-β1 for 5 days. TGF-β1 stimulated MCF-10A cells were then treated with 5FU, 5FU+F6, and F6 for 24 h. Values, expressed as fold increase of control value considered as 1, are means of three independent experiments performed in duplicate, with SD represented by vertical bars. * p < 0.05; ** p < 0.01; *** p < 0.001 versus ctrl; ^# p < 0.05; ^## p < 0.01; ^### p < 0.001 versus TGF-β1; ^@ p < 0.05; ^@@ p < 0.01 versus 5FU by ANOVA followed by Bonferroni post-test. Images were obtained by optical microscopy, with 100× magnification.

Figure 6

Distribution pattern of vinculin and F-actin in wound-healing assay performed on MDA cells cultured in control condition or exposed to 5FU, F6, and 5FU+F6. Confocal microscopy analysis of F-actin staining with rhodamine-phalloidin (red signal, (a) column) merged with anti-vinculin immunofluorescence (FITC/green signal, b column) on MDA cells subjected to wound healing assay, and cultured with or without 5FU, F6, 5FU+F6. The white arrows in the images of the left column indicate the direction of cellular movement toward the gap. In column (c) we reported higher magnification of the merging pictures shown in column (b).

Figure 7

Effect of 5FU, 5FU+F6, and F6 on expression of vinculin (a) and Rock1 (b) in MDA-MB-231 cells. Columns represent densitometric quantification of optical density (OD) of specific protein signal normalized with the OD values of GAPDH served as a loading control and they are expressed as fold increase of control value considered as 1. Each column represents the mean value ± SD of three independent experiments. * p < 0.05 versus ctrl by ANOVA followed by Bonferroni post-test. Representative western blot analysis relating to vinculin and Rock1expression in MDA-MB-231 cells untreated (ctrl) and treated with 5FU, 5FU+F6, and F6 for 24 h. GAPDH was used as loading control.

Figure 8

Effect of 5FU, 5FU+F6, and F6 on expression of E-cadherin (a), β-catenin (b), and E-cadherin/ β-catenin ratio (c) in MDA-MB-231 cells. Columns represent densitometric quantification of optical density (OD) of specific protein signal normalized with the OD values of GAPDH served as a loading control and they are expressed as fold increase of control value considered as 1. Each column represents the mean value ± SD of three independent experiments. * p < 0.05 versus ctrl; # p < 0.05 versus 5FU by ANOVA followed by Bonferroni post-test. Representative western blot analysis relating to E-cadherin and β-catenin expression in MDA-MB-231 cells untreated (ctrl) and treated with 5FU, 5FU+F6, and F6 for 24 h. GAPDH was used as loading control. (d) Confocal microscopy analysis of beta-catenin immunostaining (FITC/green signal) on MDA cells cultured in control condition, or with the following treatments: 5FU, F6, 5FU+F6. Distribution of β-catenin increase behind the cell membrane mostly in F6-treated cells.

Figure 9

Effect of 5FU, 5FU+F6, and F6 on expression of TCTP (a) and acetyl-p53 (b) in MDA-MB-231 cells. Columns represent densitometric quantification of optical density (OD) of specific protein signal normalized with the OD values of GAPDH served as a loading control and they are expressed as fold increase of control value considered as 1. Each column represents the mean value ± SD of three independent experiments. * p < 0.05; ** p < 0.01 versus ctrl; # p < 0.05 versus 5FU by ANOVA followed by Bonferroni post-test. Representative western blot analysis relating to TCTP and acetyl-p53 expression in MDA-MB-231 cells untreated (ctrl) and treated with 5FU, 5FU+F6, and F6 for 24 h. GAPDH was used as loading control.