RCAN1.4 regulates VEGFR-2 internalisation, cell polarity and migration in human microvascular endothelial cells

Abstract

Regulator of calcineurin 1 (RCAN1) is an endogenous inhibitor of the calcineurin pathway in cells. It is expressed as two isoforms in vertebrates: RCAN1.1 is constitutively expressed in most tissues, whereas transcription of RCAN1.4 is induced by several stimuli that activate the calcineurin-NFAT pathway. RCAN1.4 is highly upregulated in response to VEGF in human endothelial cells in contrast to RCAN1.1 and is essential for efficient endothelial cell migration and tubular morphogenesis. Here, we show that RCAN1.4 has a role in the regulation of agonist-stimulated VEGFR-2 internalisation and establishment of endothelial cell polarity. siRNA-mediated gene silencing revealed that RCAN1 plays a vital role in regulating VEGF-mediated cytoskeletal reorganisation and directed cell migration and sprouting angiogenesis. Adenoviral-mediated overexpression of RCAN1.4 resulted in increased endothelial cell migration. Antisense-mediated morpholino silencing of the zebrafish RCAN1.4 orthologue revealed a disrupted vascular development further confirming a role for the RCAN1.4 isoform in regulating vascular endothelial cell physiology. Our data suggest that RCAN1.4 plays a novel role in regulating endothelial cell migration by establishing endothelial cell polarity in response to VEGF.

Keywords: Angiogenesis; Endothelial; Migration; Polarisation; RCAN1; VEGFR-2.

Fig. 1

RCAN1 regulates VEGFR-2 levels in HDMECs. a Endothelial cells were left untransfected or transfected with non-silencing siRNA (N.S. siRNA) or RCAN1 siRNA. Cells were left unstimulated or stimulated with VEGF165 (50 ng/ml) for a range of time (5, 10, 30, 60, 180 min). Cells were lysed and immunoblotted with antibodies to VEGFR-1, phospho-VEGFR-2 (Y1175), VEGFR-1, phospho-AKT (S473), phospho-ERK1/2 (T202/Y204), GAPDH and RCAN1. Results are from one experiment representative of three separate experiments. b Quantification of levels of VEGFR-2, phospho-VEGFR-2, and RCAN1.1 and RCAN1.4. Levels are calculated relative to % of basal untransfected (mean ± s.e.m. n = 3 independent experiments). *P < 0.05, ** P < 0.01, ***P < 0.001 (unpaired student’s t test comparing RCAN1 siRNA and N.S. siRNA)

Fig. 2

RCAN1 regulates VEGFR-2 internalisation. a HDMECs were left untransfected or transfected with either non-silencing siRNA (N.S. siRNA) or RCAN1 siRNA. Cells were stimulated with VEGF (50 ng/ml) for 10, 30 and 60 min. Cells were fixed in 2% PFA and left unpermeabilised. Cells were incubated with goat anti-VEGFR-2 antibody (recognising the N-terminal, extracellular domain) followed by incubation with donkey anti-goat Alexa488 antibody. Nuclei were stained with Hoechst 33342. Scale bar represents 20 μm. b HDMECs were left untransfected or transfected with non-silencing siRNA (N.S.siRNA) or RCAN1 siRNA. Cells were surface labelled with cleavable biotin and stimulated or not with VEGF (50 ng/ml) for 10 min. Biotin remaining on the surface was cleaved with MesNa and biotinylated proteins immunoprecipitated with streptavidin-agarose. Internalised VEGFR-2 and transferrin receptor (TfR) were detected by Western blotting. Efficiency of biotin cleavage was analysed in untransfected cells without MesNa. The level of VEGFR-2, TfR and GAPDH was analysed in total cell lysate. c Quantification of internalised VEGFR-2 (mean ± S.D. from three individual experiments). *P < 0.05, **P < 0.01, (unpaired student’s t test)

Fig. 3

RCAN1 regulates VEGF-mediated cytoskeletal reorganisation and cell polarity. a HDMECs were grown on gelatin-coated glass coverslips. Cells were left untransfected or transfected with either non-silencing siRNA (N.S. siRNA) or RCAN1 siRNA. A horizontal wound was introduced on the glass coverslip. Cells were stimulated or not with either VEGF (50 ng/ml) or HGF (50 ng/ml) for 3 h. Cells were fixed in 2% PFA, permeabilised and stained for RCAN1. Cytoskeletal reorganisation was visualised by staining F-actin with phalloidin-Alexa Fluor 588. Nuclei were stained with Hoechst 33342. Scale bar represents 20 μm. b HDMECs were grown on gelatin-coated glass coverslips. Cells were left untransfected or transfected with either control non-silencing siRNA (N.S. siRNA) or RCAN1 siRNA. A horizontal wound was introduced on the glass coverslip. Cells were stimulated or not VEGF (50 ng/ml) for 3, 6 and 12 h. Cells were fixed in 2% PFA, permeabilised and stained for golgi (GM130) and centrosomes (pericentrin). Cell polarisation towards the wound was assessed by drawing a circle centred over the nucleus with a 120° segment facing the wound. Cells were polarised if the golgi and centrosome were located within this 120° segment. Scale bar represents 20 μm. c Quantification of cell polarisation. The percentage of polarised cells was calculated by counting 100 cells per treatment. Results are plotted as % of cells polarised at the leading edge (mean ± s.e.m. n = 3 independent experiments). *P < 0.05, **P < 0.01, ***P < 0.001

Fig. 4

RCAN1 regulates VEGF-mediated directional cell migration. HDMECs were plated in gelatin-coated 24-well plates and transfected with either non-silencing siRNA (N.S. siRNA) or RCAN1 siRNA prior to incubation in media containing 1% FCS for 24 h. A horizontal scratch wound was introduced. Cell were stimulated or not with either VEGF₁₆₅ (50 ng/ml) or HGF (50 ng/ml). Low-magnification (4× objective) phase contrast time-lapse series were obtained (images collected every 15 min for 18 h). a Images of cells following 18 h stimulation. b Migration trajectories of representative cells (n = 20). Tracks of cells migrating in a representative field of imaging. c Comparison of velocity (displacement/time), straightness (displacement/total path length) and forward migration index (FMI) (mean ± S.D., n = 3). *P < 0.05, ***P < 0.001 (unpaired student’s t test)

Fig. 5

RCAN1.4 isoform stimulates endothelial cell migration. HDMECs were plated on gelatin-coated 24-well plates and incubated with adenovirus (Ad-Control) or adenovirus encoding RCAN1.1 isoform (Ad-RCAN1.1) or RCAN1.4 isoform (Ad-RCAN1.4) for 24 h. a Cells were lysed and western blotted for levels of RCAN1.1 (39 kDa) and RCAN1.4 (29 kDa) and GAPDH. VEGF stimulation (50 ng/ml for 1 h) of untransfected cells were analysed for levels of RCAN1.4 as a control. b Cells were incubated in media containing 1% FCS for 24 h and a horizontal scratch wound was introduced. Cells were incubated with media (Basal) or stimulated with either VEGF (50 ng/ml) or HGF (50 ng/ml) for 18 h. Cells were fixed in 2% PFA. c Quantification of cell migration. Results are plotted as cell migration, % of basal untreated response (mean ± S.D., n = 3). ***P < 0.001 (unpaired student’s t test)

Fig. 6

RCAN1.4 stimulated cell migration requires VEGFR-2. a HDMECs were plated on gelatin-coated plates and transfected with either non-silencing siRNA (N.S. siRNA) or KDR siRNA prior to incubation with adenovirus (Ad-Control) or adenovirus encoding RCAN1.4 (Ad-RCAN1.4). Cells were lysed and immunoblotted with antibodies to VEGFR-2, GAPDH and RCAN1. b Cells were incubated with siRNA and adenovirus as above before a horizontal wound was introduced to the well. Cells were fixed in 2% PFA. c Results are plotted as cell migration, % of basal untreated response (mean ± S.D., n = 3). ***P < 0.001 (unpaired student’s t test)

Fig. 7

RCAN1 and VEGFR-2 interaction at the leading edge of migrating cells. a HDMECs were plated on gelatin-coated glass coverslips. A horizontal wound was introduced on the glass coverslip. Cells were stimulated or not with VEGF (50 ng/ml) for 1, 2, 5, 10 and 60 min. Cells were fixed in 2% PFA, permeabilised and stained for interaction between RCAN1 and VEGFR-2 using proximity ligation reaction (PLA) with antibodies to RCAN1 and VEGFR-2. Scale bar represents 20 μm. b Quantification of RCAN1/VEGFR-2 proximity ligations. Cells were analysed at the leading edge and also in the main body with the number of proximity ligations counted for 100 cells (mean ± s.e.m., n = 3 independent experiments). **P < 0.01, ***P < 0.001

Fig. 8

RCAN1.4 binds to VEGFR-2. a HDMECs were plated on gelatin-coated dishes and incubated with adenovirus (Ad-Control) or adenovirus encoding RCAN1.1 (Ad-RCAN1.1) or RCAN1.4 (Ad-RCAN1.4) isoforms. Cells were stimulated with VEGF (50 ng/ml) for 0, 2, 10 and 30 min. Cells were lysed and immunoprecipitation (IP) was performed using antibodies to RCAN1 or control rabbit IgG. Proteins were separated by SDS-PAGE and immunoblotted with antibodies to phospho-VEGFR-2 (Y1175), VEGFR2, calcineurin-A, GAPDH and RCAN1. Total cell lysate (TCL) was also analysed by Western blotting. b Quantification of levels of VEGFR-2 and phospho-VEGFR-2 in TCL. Levels are calculated relative to % of basal Ad-Control (mean ± s.e.m. n = 3 independent experiments). **P < 0.01, ***P < 0.001 (unpaired student’s t test)

Fig. 9

RCAN1.4 regulates endothelial cell sprouting and vascular formation in zebrafish. a In vitro sprouting angiogenesis assay. HDMECs were transfected with control non-targeting siRNA (N.S. siRNA) or RCAN1 siRNA prior to incubation with cell tracker dye green or orange. Cells were mixed 1:1 and incubated with carboxymethycellulose in ULA 96-well plates for 18 h prior to embedding in collagen matrix. Cells were stimulated with VEGF (50 ng/ml) for 18 h and fixed in 2% PFA prior to fluorescence imaging. b Quantification of endothelial tip and stalk cell formation. Sprouting tubes from VEGF treated spheroids were quantified (mean ± S.D., n = 3). ***P < 0.001 (unpaired Student’s t test). c Transgenic fli1-GFP zebrafish embryos were injected with mismatch or RCAN1a-4 morpholino oligonucleotides and imaged at 48 hpf. Scale bar represents 500 μm. d Analyses of total vessel length and junction number, n = 14 for mismatch, n = 15 for RCAN morpholino. Asterisk indicates P < 0.01 (unpaired student’s t test)

Fig. 10

Proposed mechanism for the role of RCAN1.4 in regulating VEGF action on endothelial cells. VEGF binding to VEGFR-2 leads to autophosphorylation on a number of c-terminal tyrosine residues resulting in the activation of PLC-γ and cleavage of PtdIns(4,5)P₂ (PIP₂) generating diacylglycerol (DAG) and Ins(1,4,5)P₃ (IP₃). Following VEGFR-2 phosphorylation, RCAN1.4 transiently interacts with VEGFR-2 to regulate internalisation at the leading edge. The resulting Ins(1,4,5)P₃-mediated release of Ca²⁺ from the endoplasmic reticulum (ER) results in activation of calcineurin/PP2B and dephosphorylation of the nuclear factor activated T-cells (NFAT) transcription factor which translocates to the nucleus and increases transcription of RCAN1.4 mRNA. The resulting increase in RCAN1.4 protein acts on calcineurin to decrease NFAT-mediated gene transcription. Increased RCAN1.4 protein also acts independently of calcineurin to regulate VEGFR-2-mediated cytoskeletal reorganisation, filopodia and lamellipodia formation resulting in endothelial cell polarisation and directed migration