Interaction between circulating galectin-3 and cancer-associated MUC1 enhances tumour cell homotypic aggregation and prevents anoikis

Abstract

Background: Formation of tumour cell aggregation/emboli prolongs the survival of circulating tumour cells in the circulation, enhances their physical trapping in the micro-vasculature and thus increases metastatic spread of the cancer cells to remote sites.

Results: It shows here that the presence of the galactoside-binding galectin-3, whose concentration is markedly increased in the blood circulation of cancer patients, increases cancer cell homotypic aggregation under anchorage-independent conditions by interaction with the oncofetal Thomsen-Friedenreich carbohydrate (Galbeta1,3GalNAcalpha-, TF) antigen on the cancer-associated transmembrane mucin protein MUC1. The galectin-3-MUC1 interaction induces MUC1 cell surface polarization and exposure of the cell surface adhesion molecules including E-cadherin. The enhanced cancer cell homotypic aggregation by galectin-MUC1 interaction increases the survival of the tumour cells under anchorage-independent conditions by allowing them to avoid initiation of anoikis (suspension-induced apoptosis).

Conclusion: These results suggest that the interaction between free circulating galectin-3 and cancer-associated MUC1 promotes embolus formation and survival of disseminating tumour cells in the circulation. This provides new information into our understanding of the molecular mechanisms of cancer cell haematogenous dissemination and suggests that targeting the interaction of circulating galectin-3 with MUC1 in the circulation may represent an effective therapeutic approach for preventing metastasis.

Figure 1

MUC1 expression prevents and MUC1-galectin-3 interaction promotes homotypic aggregation of human colon cancer cells. A: Cell surface staining with B27.29 anti-MUC1 antibody followed by analysis with flow cytometry shows higher cell surface MUC1 expression in HT29-5F7 than in HT29 cells. B: Representative flow cytometry plots from the aggregation assessment of human colon cancer HT29-5F7 and HT29 cells in the presence or absence of 1 μg/ml recombinant galectin-3. The top right (blue) in the bivariate correlation plot are the cell population containing both DiO- and Dil-labelled cells that are defined in this study as cell aggregates. C: The more strongly MUC1-expressing HT29-5F7 cells show less spontaneous cell aggregation than the parental HT29 cells. Data are expressed as mean ± SEM of triplicate determinations from three independent experiments. D: galectin-3 treatment induces a dose-dependent increase of HT29-5F7 but not HT29 cell aggregation. Data are expressed as mean ± SEM of triplicate determinations from four independent experiments. **p < 0.01, ***p < 0.001.

Figure 2

MUC1 expression prevents and MUC1-galectin-3 interaction promotes homotypic aggregation of MUC1 positively but not negatively transfected human breast epithelial cells. A: Representative flow cytometry plots from the aggregation assessment of MUC1 positive-transfectants (HCA1.7+) and negative-revertants (HCA1.7-) of HBL-100 human breast epithelial cells in the presence or absence or 1 μg/ml recombinant galectin-3. B: HCA1.7+ cells show less spontaneous cell-cell aggregation than HCA1.7- cells. Data are expressed as mean ± SEM of triplicate determinations from four independent experiments. C: Galectin-3 (1 μg/ml) increases HCA1.7+ but not HCA1.7- cell aggregation. Data are expressed as mean ± SEM of triplicate determinations from four independent experiments. D: Galectin-3 induces dose-dependent aggregation of HCA1.7+ cells. Data are expressed as mean ± SEM of triplicate determinations from three independent experiments. E: The presence of lactose (10 μM) blocks the increase of HCA1.7+ cell aggregation induced by 1 μg/ml galectin-3. Data are expressed as mean ± SEM of triplicate determinations from two independent experiments. *p < 0.05, **p < 0.01, ***p < 0.001.

Figure 3

Effects of TF expression on galectin-3-mediated cell aggregation. A: O-Glycanase treatment of HCA1.7+ cells reduces TF-expression on MUC1. B: Reduction of TF-expression by O-glycanase treatment abolishes the effect of galectin-3 (1 μg/ml) on HCA1.7+ cell aggregation. Data are expressed as mean ± SEM of triplicate determinations from two independent experiments. *p < 0.05.

Figure 4

Galectin-3 and B27.29 anti-MUC1 mAb both induce increase of cell aggregation. A: B27.29 mAb induces dose-dependent increase of HCA1.7+ cell aggregation. HCA1.7+ cell aggregation was determined after pre-incubation of the cells with or without various concentrations of B27.29 Mab, 1 μg/ml control mouse immunoglublin or CT-2 anti-MUC1 antibody. Data are expressed as mean ± SEM of triplicate determinations from three independent experiments. B: B27.29 mAb increases HCA1.7+ cell aggregation regardless of the presence or absence of recombinant galectin-3. HCA1.7+ cell aggregation was determined after pre-incubation of the cells with or without 1 μg/ml recombinant galectin-3 in the presence or absence of 1 μg/ml B27.29, BSA or control immunoglublin. Data are expressed as mean ± SEM of triplicate determinations from three independent experiments. C: Galectin-3 or B27.29 mAb at 1 μg/ml fails to induce HCA1.7+ cell aggregation at 4°C or to paraformaldehyde-prefixed cells. Data are expressed as mean ± SEM of triplicate determinations from four independent experiments. *p < 0.05,**p < 0.01, ***p < 0.001. D: MUC1 localization in cell aggregates. Separate aliquots of HCA1.7+ cells pre-labelled with DiO (green) or DiI (red) were mixed in the presence of 1 μg/ml galectin-3 or B27.29 mAb and incubated for 1 hr at 37°C before fixation and subsequent analysis of MUC1 localization by immunohistochemistry. Representative images of the MUC1 localization in cell aggregates are shown. F: localization of the cell aggregates-associated MUC1 (green) after treatment of the cells with 1 μg/ml galectin-3 for 48 hr under suspension (red: cell nucleuses). Bar = 10 μm.

Figure 5

Involvement of cell surface E-cadherin in galectin-3-induced cell aggregation. A: The presence of anti-E-cadherin antibody reduces spontaneous aggregation of HT29 but has little effect on aggregation of HT29-5F7 cells. Data are expressed as mean ± SEM of triplicate determinations from three independent experiments. B: Anti-E-cadherin antibody at 20 μg/ml prevents HT29-5F7 cell aggregation-induced by (1 μg/ml) galectin-3. Data are expressed as mean ± SEM of triplicate determinations from three independent experiments. C: E-cadherin localization in single cells and cell aggregates. HT29-5F7 cells released by NECDS and labelled separately with DiO and DiI were mixed in the presence or absence of galectin-3 for 1 hr at 37°C. The cells were fixed, probed with mAb B27.29 anti-MUC1 or anti-E-cadherin and fluorescent-labelled secondary antibody and analysed by fluorescent microscopy. E-cadherin shows localization/accumulation at the cell-cell contact points (arrowed) in the cell aggregates. D: Western blot showing that transfection of HT29-5F7 cells with siRNA for E-cadherin suppresses E-cadherin expression. E: E-cadherin immunohistochemistry shows marked reduction of the cell surface E-cadherin (green) in HT29-5F7 cells after treatment with E-cadherin siRNA for 72 hr (red: cell nucleus). F: SiRNA-mediated knock-down of E-cadherin expression prevents galectin-3-mediated HT29-5F7 cell aggregation. Data are expressed as mean ± SEM of triplicate determinations from three independent experiments. G: flow cytometry analysis of HT29 and HT29-5F7 cells stained with the MAB1838 anti-E-cadherin antibody show similar E-cadherin cell surface expressions. Open histogram: E-cadherin; shaded histogram: immunoglobulin isotype control.

Figure 6

Galectin-3-induced cell aggregation enhances cell survival by avoiding initiation of anoikis. A: Galectin-3-induced cell aggregation is associated with increased survival of the cells under anchorage-independent conditions. HT29-5F7 cells treated with or without 1 μg/ml galectin-3 or BSA for 1 hr followed by culture of the cells in suspension for 3 days at 37°C. After separation of the cells by cell strainers, the viability of the strained (single) cells and cell aggregates was assessed. The data are presented as mean ± SEM of triplicate determinations from two independent experiments. B: The presence of 20 μg/ml anti-E-cadherin antibody inhibited galectin-3-mediated increase of survival of HT29-5F7 cell aggregates. The data are presented as mean ± SEM of triplicate determinations from two independent experiments. C and D: Galectin-3 treatment has no effect on anoikis of the single cells (C) but significant reduction of anoikis of the cell aggregates (D). Anoikis was assessed by Annexin-V cell surface binding after treatment of the cells with galectin-3 and subsequent separation of the single cells and cell aggregates. The data are presented as mean ± SEM of triplicate determinations from two independent experiments. E. Representative flow cytometry plots from the anoikis assessments of HT29-5F7 cells in the presence of 1 μg/ml galectin-3 or BSA for 48 hr at 37°C. Annexin-V positive and PI negative (early apoptotic, at the bottom right in the bivariate correlation plot) and Annexin-V positive and PI positive (late apoptotic, at the top right in the correlation plot) cells are considered as apoptotic cells.