PRL-2 phosphatase is required for vascular morphogenesis and angiogenic signaling

Abstract

Protein tyrosine phosphatases are essential modulators of angiogenesis and have been identified as novel therapeutic targets in cancer and anti-angiogenesis. The roles of atypical Phosphatase of Regenerative Liver (PRL) phosphatases in this context remain poorly understood. Here, we investigate the biological function of PRL phosphatases in developmental angiogenesis in the postnatal mouse retina and in cell culture. We show that endothelial cells in the retina express PRL-2 encoded by the Ptp4a2 gene, and that inducible endothelial and global Ptp4a2 mutant mice exhibit defective retinal vascular outgrowth, arteriovenous differentiation, and sprouting angiogenesis. Mechanistically, PTP4A2 deletion limits angiogenesis by inhibiting endothelial cell migration and the VEGF-A, DLL-4/NOTCH-1 signaling pathway. This study reveals the importance of PRL-2 as a modulator of vascular development.

Fig. 1. Postnatal endothelial Ptp4a2 deletion impairs vascular development.

a Generation strategy for Ptp4a2^fl/fliEKO mice and gene deletion through intragastric tamoxifen injection in postnatal mice. b Mice weight in P6 Ptp4a2^fl/fl and Ptp4a2^fl/fliEKO (Mann–Whitney U test: ns nonsignificant p value > 0.05). c qPCR analysis of mouse lung endothelial cell isolated from P6 Ptp4a2^fl/fl and Ptp4a2^fl/fliEKO (two-way ANOVA: ****p value < 0.0001). Each individual data point represents a mouse. d Retina whole-mount staining with isolectin B4 (IB4), and quantification of e vascular outgrowth and f density in P6 Ptp4a2^fl/fl and Ptp4a2^fl/fliEKO mice. Each individual data point represents a mouse (average of two retinas) (Mann–Whitney U test: **p value < 0.01, ****p value < 0.001). g Retina whole-mount staining with isolectin B4 and high magnification of vascular front. h quantification of the number of sprouts in P6 Ptp4a2^fl/fl and Ptp4a2^fl/fliEKO mice. Each individual data point represents a mouse i high magnification of retina whole mounts stained with IB4/ERG1/2/3 and j filopodia quantification per sprout (Mann–Whitney U test: *p value < 0.05, **p value < 0.01). Each n represents individual mouse. Scale bars: d 250 μm, g 100 μm, i 25 μm. Error bars represent mean ± s.e.m.

Fig. 2. Ptp4a2 postnatal endothelial deletion reduced endothelial cell number and increased vascular regression.

a Retina whole-mount staining with isolectin B4 and ERG1/2/3 and b quantification of the number of ERG1/2/3-positive endothelial cells in P6 Ptp4a2^fl/fl and Ptp4a2^fl/fliEKO mice. c Retina whole-mount staining with isolectin B4 and KI67 and ERG1/2/3, and d quantification of proliferative endothelial cells in P6 Ptp4a2^fl/fl and Ptp4a2^fl/fliEKO mice. e Retina whole-mount staining with isolectin B4 and collagen IV and f IB4⁻/COLIV⁺ sleeves quantification in P6 Ptp4a2^fl/fl and Ptp4a2^fl/fliEKO mice (Mann–Whitney U test: *p value < 0.05, **p value < 0.01, ns nonsignificant p value > 0.05). Each individual data point represents a mouse. Scale bars: a, c 100 μm, e 15 μm. Error bars represent mean ± s.e.m.

Fig. 3. Vascular defects in global Ptp4a2 knockout mice.

a Western blot of protein isolated from the Ptp4a2^+/− (het), ^+/+ (wt), or ^−/− (ko) mouse retina. b Retina whole-mount staining with isolectin B4 and c vascular outgrowth quantification in P6 Ptp4a2^+/+ and Ptp4a2^−/− mice. Each individual data point represents a mouse (average of two retinas). d Retina whole-mount staining with isolectin B4, and quantification of e vascular density and f branch points in P6 Ptp4a2^+/+ and Ptp4a2^−/− mice. Each individual data point represents a mouse (average of two retinas). g Retina whole-mount staining with isolectin B4 and high magnification of the vascular front and h quantification of the number of sprouts in P6 Ptp4a2^+/+ and Ptp4a2^−/− mice. Each n represents an individual mouse. i Retina whole-mount staining with isolectin B4 and quantification for j vascular outgrowth and k deeper plexus vascularization in P9 Ptp4a2^+/+ and Ptp4a2^−/− mice (Mann–Whitney U test: **p value < 0.01, ***p value < 0.001 ****p value < 0.0001). Each individual data point represents a mouse (average of two retinas). Scale bars: b 500 μm, d, g, i 250 μm. Error bars represent mean ± s.e.m.

Fig. 4. PRL-2 is essential for arterial–venous patterning.

a Retina whole-mount staining with isolectin B4 to quantify number of arteries and veins in P6 Ptp4a2^fl/fl and Ptp4a2^fl/fliEKO mice. b Retina whole-mount staining with isolectin B4 and a-SMA to quantify number of arteries and veins in P6 Ptp4a2^+/+ and Ptp4a2^−/− mice, and quantification of main vein and artery in c, d Ptp4a2^fl/fl and Ptp4a2 ^fl/fliEKO mice, or e, f Ptp4a2^+/+ and Ptp4a2^−/− mice. g Retina whole-mount staining with isolectin B4 to see arteries in P21 Ptp4a2^+/+ and Ptp4a2^−/− mice, (h) and quantification (Mann–Whitney U test: **p value < 0.01, ****p value < 0.0001, ns nonsignificant p value > 0.05). Each individual data point represents a mouse. Scale bars: a, b 250 and 125 μm, g 500 μm. Error bars represent mean ± s.e.m.

Fig. 5. Effect of PRL-2 silencing on endothelial cell migration and sprouting in vitro.

a qPCR analysis in HUVEC cells after CTRL or PTP4A2 siRNA treatment (at least n = 6 independent experiments; two-way ANOVA: *p value < 0.05). b Western blot in HUVEC cells, and c PRL-1 and PRL-2 protein quantification of western blot shown in b after CTRL or PTP4A2 siRNA treatment (n = 10 independent experiments; two-way ANOVA: ***p value < 0.001, ****p value < 0.0001). d Scratch wound assay performed on HUVEC monolayer in the presence of VEGF-A (50 ng/ml) after CTRL or PTP4A2 siRNA treatment. Images at 0 and 16 h after scratch. e Analysis and quantification of the wound closure in HUVEC cells after CTRL or PTP4A2 siRNA treatment in the presence (50 and 100 ng/ml), or absence of VEGF-A (PBS). (Each individual data point represents a biological replicat from n = 3 independent experiments, two-way ANOVA: ****p value < 0.0001, ns nonsignificant p value > 0.05). f, g EdU incorporation assay (8 h) on HUVEC cells after CTRL or PTP4A2 siRNA treatment (Mann–Whitney U test: ns nonsignificant p value > 0.05). Each individual data point represents a biological replicat from n = 3 independent experiments. h HUVEC sprouting assay embedded in 3D fibrinogen gel and stimulated with VEGF-A (100 ng/ml) after CTRL or PTP4A2 siRNA treatment. Quantification of i protrusion length, j number of cells per protrusion, k number of protrusions per beads. Results were expressed as average/beads. Each individual data point represent a beads; n = 3 (Mann–Whitney U test: ***p value < 0.001, ****p value < 0.0001). Scale bars: d 250 μm, f 50 μm, h 100 and 50 μm. Error bars represent mean ± s.e.m.

Fig. 6. Impact of PRL-2 silencing on VEGF signaling in vascular endothelial cells.

a qPCR analysis of VEGFA in HUVEC cells after CTRL or PTP4A2 siRNA treatment (Mann–Whitney U test: *p value < 0.05; n = 5). b Effects of VEGF-A treatment on VEGFR-2 phosphorylation (p-Y1175, p-Y951) in siRNA CTRL and PTP4A2 knockdown cells. c, d Quantification of phosphorylation of VEGFR-2 Y1175 and Y951 normalized to total VEGFR-2 and compared to PBS treated control (n = 3 independent experiments; two-way ANOVA: ***p value < 0.001, ns nonsignificant p value > 0.05). e Effects of VEGF-A treatment on ERK1/2 and AKT phosphorylation in HUVECs, with CTRL siRNA and with siRNA targeting PTP4A2. f, g Quantification of p-ERK normalized to total ERK and pAKT normalized to total AKT (n = 3 independent experiments; two-way ANOVA: **p value < 0.01, ***p value < 0.001, ns nonsignificant p value > 0.05). Error bars represent mean ± s.e.m.

Fig. 7. PRL-2 regulates NOTCH-1 cleavage.

a qPCR analysis in HUVEC cells after CTRL or PTP4A2 siRNA treatment (n = 4 independent experiments; two-way ANOVA: ***p value < 0.001, ****p value < 0.0001). b Western blot analysis on HUVEC cells and quantification after CTRL or PTP4A2 siRNA treatment. c Quantifications of blots shown in b (n = 3 independent experiments; two-way ANOVA: **p value < 0.01, ***p value < 0.001, ns nonsignificant p value > 0.05). d, e Relative mRNA level of HEY2 and HES1, respectively, assessed by qPCR analysis in HUVEC cells treated with CTRL, PTP4A2, or NOTCH1 siRNA and plated overnight on DLL-4-coated plate (n = 3 independent experiments; two-way ANOVA: *p value < 0.05 **p value < 0.01, ***p value < 0.001, ****p value < 0.0001, ns nonsignificant p value > 0.05). f Ptp4a2 qPCR analysis in HUVEC cells after CTRL, PTP4A2, or NOTCH1 siRNA treatment (n = 4 independent experiments; two-way ANOVA: ****p value < 0.0001, ns nonsignificant p value > 0.05). g Schematic representation for PRL-2’s role in the vasculature is depicted in the figure. This schematic was created with images adapted from Servier Medical Art licensed under a Creative Commons Attribution 3.0. Error bars represent mean ± s.e.m.