Breast cancer cell migration is regulated through junctional adhesion molecule-A-mediated activation of Rap1 GTPase

Abstract

Introduction: The adhesion protein junctional adhesion molecule-A (JAM-A) regulates epithelial cell morphology and migration, and its over-expression has recently been linked with increased risk of metastasis in breast cancer patients. As cell migration is an early requirement for tumor metastasis, we sought to identify the JAM-A signalling events regulating migration in breast cancer cells.

Methods: MCF7 breast cancer cells (which express high endogenous levels of JAM-A) and primary cultures from breast cancer patients were used for this study. JAM-A was knocked down in MCF7 cells using siRNA to determine the consequences for cell adhesion, cell migration and the protein expression of various integrin subunits. As we had previously demonstrated a link between the expression of JAM-A and β1-integrin, we examined activation of the β1-integrin regulator Rap1 GTPase in response to JAM-A knockdown or functional antagonism. To test whether JAM-A, Rap1 and β1-integrin lie in a linear pathway, we tested functional inhibitors of all three proteins separately or together in migration assays. Finally we performed immunoprecipitations in MCF7 cells and primary breast cells to determine the binding partners connecting JAM-A to Rap1 activation.

Results: JAM-A knockdown in MCF7 breast cancer cells reduced adhesion to, and migration through, the β1-integrin substrate fibronectin. This was accompanied by reduced protein expression of β1-integrin and its binding partners αV- and α5-integrin. Rap1 activity was reduced in response to JAM-A knockdown or inhibition, and pharmacological inhibition of Rap1 reduced MCF7 cell migration. No additive anti-migratory effect was observed in response to simultaneous inhibition of JAM-A, Rap1 and β1-integrin, suggesting that they lie in a linear migratory pathway. Finally, in an attempt to elucidate the binding partners putatively linking JAM-A to Rap1 activation, we have demonstrated the formation of a complex between JAM-A, AF-6 and the Rap1 activator PDZ-GEF2 in MCF7 cells and in primary cultures from breast cancer patients.

Conclusions: Our findings provide compelling evidence of a novel role for JAM-A in driving breast cancer cell migration via activation of Rap1 GTPase and β1-integrin. We speculate that JAM-A over-expression in some breast cancer patients may represent a novel therapeutic target to reduce the likelihood of metastasis.

Figure 1

JAM-A knockdown reduces breast cancer cell adhesion and migration. (a) Transfection of MCF7 breast cancer cells with short interfering RNA (siRNA) targeting JAM-A reduces JAM-A protein expression. (b) Adhesion of MCF7 cells transfected with JAM-A siRNA to a fibronectin substrate was reduced relative to that induced under control conditions. (c) Individual cell migration of MCF7 cells transfected with JAM-A siRNA or control siRNA MCF7 cells across fibronectin-coated, 8-μm pore filters was reduced relative to control conditions. Images show DAPI (4'-6-diamidino-2-phenylindole)-stained migrated cells on transwell filters. Error bars refer to standard deviation of pooled triplicate experiments. JAM-A, junctional adhesion molecule-A.

Figure 2

JAM-A regulates β1-integrin protein expression and Rap1 GTPase activity in breast cancer cells. (a) Western blot analysis of a panel of alpha- and beta-subunit integrins was conducted on MCF7 cell lines transfected with control short interfering RNA (siRNA) or JAM-A-siRNA. Assessment of actin expression was performed to control for protein loading. (b) Densitometry analysis of triplicate Western blot experiments showing relative protein expression of alpha- and beta-subunit integrins. Error bars refer to standard deviation of triplicate experiments. (c) β1-integrin Western blot analysis of equal total protein lysates (input) and JAM-A immunoprecipitates (IP:JAM-A) from MCF7 cells transfected with control siRNA or JAM-A-siRNA. (d) Western blot analysis of JAM-A, total Rap1, and active Rap1 protein expression in MCF7 cells transfected with control or JAM-A siRNA and in MCF7 cells treated with isotype control antibody or JAM-A inhibitory antibody. (e) Densitometric analysis of triplicate JAM experiments showing the ratio of active to total Rap1 after JAM-A knockdown or antagonism. Error bars refer to standard deviation of triplicate experiments. JAM-A, junctional adhesion molecule-A.

Figure 3

Rap1 or β1-integrin protein antagonism reduces breast cancer cell migration. (a) Western blot analysis of active Rap1 protein expression in MCF7 cells following treatment with the Rap1 pharmacological inhibitor GGTI-298 (10 μM). Migration of MCF7 cells in scratch-wound assays after pretreatment for 2 hours with Rap1 inhibitor (b), MAb 13 anti-β1-integrin inhibitory antibody (c), J104 anti-JAM-A inhibitory antibody (d), or a combination of all three antagonists (e). Error bars refer to standard deviation and represent triplicate values in a representative experiment. (f) Graphic representation of migration differences between control treatments and inhibitor/antibody treatments alone and in combination. Error bars refer to standard deviation in pooled triplicate experiments. JAM-A, junctional adhesion molecule-A.

Figure 4

JAM-A co-associates with AF-6 and PDZ-GEF2 in breast cancer cells. Immunoblot (IB) analysis for JAM-A, Rap1, AF-6, and PDZ-GEF2 in equivalent concentrations of total protein lysates (input) and JAM-A immunoprecipitates (IP:JAM-A) from mock-transfected MCF7 cells and MCF7 cells transfected with control short interfering RNA (siRNA) or JAM-A-siRNA. JAM-A, junctional adhesion molecule-A.

Figure 5

JAM-A signalling is increased in breast cancer primary tumor cells. (a) Immunoblot analysis for JAM-A, AF-6, PDZ-GEF2, Rap1, and β1-integrin protein expression conducted on input lysates of equivalent protein concentrations (left) and JAM-A immunoprecipitates (right) isolated from matched tumor (T) and non-tumor (N) primary breast cultures. (b) Comparison of tumor-to-non-tumor ratios from pooled samples of input protein and JAM-A immunoprecipitates for JAM-A, AF-6, and PDZ-GEF2 individually. JAM-A, junctional adhesion molecule-A.

Figure 6

Model of JAM-A signalling in human breast cancer cells. Our working model hypothesizes that, in normal breast cells, a baseline level of JAM-A signalling via AF-6 and PDZ-GEF2 leads to a low level of β1-integrin-mediated cell migration required for crucial normal physiological processes such as wound healing (a). However, in breast cancer cells, overexpression of JAM-A leads to its increased association with PDZ-GEF2 protein and this in turn hyperactivates the GTPase Rap1. We suggest that this culminates in increased β1-integrin-mediated cancer cell migration and leads to increased risk of invasion and metastasis (b). ECM, extracellular matrix; JAM-A, junctional adhesion molecule-A.