VE-PTP and VE-cadherin ectodomains interact to facilitate regulation of phosphorylation and cell contacts

Abstract

VE-cadherin is the essential adhesion molecule in endothelial adherens junctions, and the regulation of protein tyrosine phosphorylation is thought to be important for the control of adherens junction integrity. We show here that VE-PTP (vascular endothelial protein tyrosine phosphatase), an endothelial receptor-type phosphatase, co-precipitates with VE-cadherin, but not with beta-catenin, from cell lysates of transfected COS-7 cells and of endothelial cells. Co-precipitation of VE-cadherin and VE-PTP required the most membrane-proximal extracellular domains of each protein. Expression of VE-PTP in triple-transfected COS-7 cells and in CHO cells reversed the tyrosine phosphorylation of VE-cadherin elicited by vascular endothelial growth factor receptor 2 (VEGFR-2). Expression of VE-PTP under an inducible promotor in CHO cells transfected with VE-cadherin and VEGFR-2 increased the VE-cadherin-mediated barrier integrity of a cellular monolayer. Surprisingly, a catalytically inactive mutant form of VE-PTP had the same effect on VE-cadherin phosphorylation and cell layer permeability. Thus, VE-PTP is a transmembrane binding partner of VE-cadherin that associates through an extracellular domain and reduces the tyrosine phosphorylation of VE-cadherin and cell layer permeability independently of its enzymatic activity.

Fig. 1. Association of VE-cadherin and Flag-VE-PTP in COS-7 cells. (A) Schematic representation of the putative domain structure of VE-PTP and truncated VE-PTP where all extracellular domains, except the most membrane-proximal domain, are replaced by a Flag-tag (Flag-VE-PTP). TM, transmembrane domain. (B) COS-7 cells were transiently transfected either with VE-cadherin alone, with VE-cadherin and Flag-VE-PTP, with JAM alone, with JAM and Flag-VE-PTP, with β-catenin alone or with β-catenin and Flag-VE-PTP. Cell lysates were immunoprecipitated with anti-Flag antibodies, and precipitated proteins were transferred to filters and analysed by immunoblotting with anti-Flag antibodies (top). The same filters were re-probed with anti-VE-cadherin or anti-JAM or anti-β-catenin (middle). Similarly, total cell lysates were immunoblotted for these proteins (bottom). The arrowhead marks the heavy chain of IgG. Molecular mass markers (in kDa) are shown on the left.

Fig. 2. Deduced amino acid sequence of VE-PTP. The predicted signal peptide and the transmembrane region are indicated by bold underlining. The conserved catalytic domain that is homologous to other members of the PTP family is defined by thin underlining. The beginning of each predicted FNIII motif is marked by an arrow.

Fig. 3. VE-cadherin does not require its cytoplasmic tail for association with full-length VE-PTP. (A) Schematic representation of the domain structure of full-length VE-cadherin and truncated VE-cadherin lacking its complete cytoplasmic domain (VE-cad EC). (B) COS cells were transfected either with full-length VE-PTP and full-length VE-cadherin (VE-PTP/VE-cadherin), with VE-cadherin alone (VE-cadherin), with full-length VE-PTP and truncated VE-cadherin lacking its complete cytoplasmic domain (VE-PTP/VE-cad EC) or with the truncated version of VE-cadherin alone (VE-cad EC). Cell lysates were subjected to immunoprecipitations with anti-VE-PTP antibodies, and precipitated proteins were immunoblotted either with anti-VE-cadherin antibodies (upper) or with anti-VE-PTP antibodies (middle). Note that full-length and truncated VE-cadherin were co-precipitated with VE-PTP. To control transfection efficiency, total cell lysates were immunoblotted for VE-cadherin. The asterisk marks the degradation product of VE-cadherin. Molecular mass markers (in kDa) are shown on the left.

Fig. 4. The 17th FNIII repeat of VE-PTP is sufficient for binding to VE-cadherin. (A) Schematic representation of the domain structure of full-length VE-PTP (dark grey), full-length JAM (light grey) and three chimeric fusion proteins containing an N-terminal Flag-tag and various parts of VE-PTP and JAM: the cytoplasmic tail of VE-PTP and the extracellular and transmembrane domain of JAM (JAM–VE1), the cytoplasmic and transmembrane domain of VE-PTP fused to the extracellular domain of JAM (JAM–VE2) or the extracellular 17th FNIII repeat of VE-PTP and the cytoplasmic and transmembrane domain of JAM (VE–JAM). (B) COS cells were transfected with truncated VE-cadherin lacking its cytoplasmic tail (VE-cad EC, lanes 1–4) and, in addition, co-transfected with one of the three different VE-PTP– JAM fusion proteins (lanes 1–3). Proteins immunoprecipitated with anti-Flag antibodies were immunoblotted first with anti-VE-cadherin antibodies (top), and then the same filters were re-probed with anti-Flag antibodies (middle). Total cell lysates were immunoblotted with anti-VE-cadherin antibodies (bottom). Note that truncated VE-cadherin was only co-precipitated with the VE-PTP–JAM fusion protein that included the 17th FNIII repeat of VE-PTP (lane 3). The asterisk marks the IgG heavy chain detected by the secondary antibody. Molecular mass markers (in kDa) are shown on the left.

Fig. 5. The fifth cadherin domain of VE-cadherin is sufficient for binding to VE-PTP. (A) Schematic representation of the domain structure of truncated VE-cadherin lacking the cytoplasmic part (VE-cad EC) and four truncation mutants with the same cytoplasmic truncation and lacking, in addition, the first (VE2–5), the first two (VE3–5), the first three (VE4–5) or the first four (VE5) extracellular cadherin (EC) domains. (B) COS cells were either transfected with one of the five different truncated VE-cadherin constructs alone or, in addition, co-transfected with the VE–JAM fusion protein containing the Flag-tagged 17th FNIII domain of VE-PTP and the transmembrane and cytoplasmic domains of JAM. Cell lysates were immunoprecipitated with anti-Flag antibodies, precipitated proteins were transfered to a filter and the same filter was first reacted with anti-VE-cadherin antibodies and then with anti-Flag antibodies. Molecular mass markers (in kDa) are shown on the left.

Fig. 6. Endogenous VE-cadherin in mouse endothelioma cells interacts with native VE-PTP. Lysates of bEnd3 mouse endothelioma cells were immunoprecipitated with polyclonal antibodies against VE-PTP (anti-VE-PTP) or control rabbit IgG (rabbit IgG). Precipitated proteins were immunoblotted either with anti-VE-cadherin antibodies (top) or with anti-VE-PTP antibodies (bottom). Molecular mass markers (in kDa) are shown on the left.

Fig. 7. In vivo dephosphorylation of VE-cadherin by Flag-VE-PTP. COS cells were transfected with VE-cadherin, VEGFR-2 (Flk-1) and Flag-VE-PTP (lane 1); with VE-cadherin, VEGFR-2 (Flk-1) and the dominant negative Flag-VE-PTP R/A mutant (lane 2); with VE-cadherin and VEGFR-2 (Flk-1) (lane 3); or with VE-cadherin alone (lane 4). Cell lysates were immunoprecipitated with either anti-VE-cadherin, anti-VEGFR-2 (Flk-1) or anti-Flag antibodies, and immunoprecipitates were analysed by immunoblotting. Note that VE-cadherin was only phosphorylated in the presence of Flk-1 and that this phosphorylation was reversed by recombinant VE-PTP, but not by its point-mutated version. Molecular mass markers (in kDa) are shown on the left.

Fig. 8. FACS analysis of CHO cells stably co-transfected with VE-cadherin, VEGFR-2 and inducible Flag-VE-PTP or Flag-VE-PTP R/A. The triple-transfected CHO-F12 cells were either stimulated with mifepristone for 0 h (A), 4 h (B) or 24 h (C) or left unstimulated (D and E). Cells were analysed by flow cytometry with antibodies against the Flag-tag, recognizing tagged VE-PTP (A–C), VE-cadherin (D) or VEGFR-2 (E). CHO-RA7 cells (F) were analysed with anti-Flag antibodies after 24 h stimulation with mifepristone (bold line) or left unstimulated (grey line). The thin lines indicate signals obtained with mock-transfected CHO cells.

Fig. 9. VE-PTP and VE-PTP R/A reduce paracellular permeability in VE-cadherin/Flk-1/VE-PTP triple-transfected CHO cells and inhibit tyrosine phosphorylation of VE-cadherin. Paracellular permeability for FITC-dextran was analysed with monolayers of various stably transfected CHO cells. (A) VE-cadherin-transfected CHO cells (dark grey) had a 52 ± 8% reduced permeability compared with mock-transfected CHO (light grey.). (B) No significant difference in permeability was observed if CHO-C3 cells double transfected with VE-cadherin and inducible VE-PTP were either stimulated with mifepristone (dark grey) or mock stimulated (light grey). (C) Analysis of triple-transfected CHO-F12, transfected with VE-cadherin, VEGFR-2 and inducible Flag-VE-PTP. Stimulation with mifepristone for 10 h resulted in a 38 ± 12% reduction of cell permeability (dark grey) as compared with mock-stimulated cells (light grey). All three experiments are representative of nine similar experiments. (D) Analysis of triple-transfected CHO-RA7, transfected with VE-cadherin, VEGFR-2 and inducible Flag-VE-PTP R/A. Stimulation with mifepristone for 10 h resulted in a 38 ± 13% reduction of cell permeability (dark grey) as compared with mock-stimulated cells (light grey). This experiment is representative of five similar experiments. CHO-F12 cells (E) and CHO-RA7 cells (F) were analysed by immunoprecipitations with anti-VE-cadherin antibodies followed by immunoblotting with an antibody against phospho-tyrosine (anti-PY) (top). The same filter was re-probed with an antibody against VE-cadherin, demonstrating equal loading of VE-cadherin (middle). Immunoprecipitations with anti-Flag antibodies, followed by immunoblotting with the same antibody, revealed strong induction of truncated VE-PTP on mifepristone treatment (bottom, only shown for CHO-F12 cells).