NRP1 regulates VEGFA-mediated permeability in an organotypic manner

(A) Schematic illustration of experimental design to induce recombination in Nrp1 iECKO mice.

(B) Western blot quantification of lung lysates from tamoxifen-treated control and Nrp1 iECKO mice (n=4).

(C) Representative images showing leakage of 2000 kDa FITC-dextran (Pseudo-colour) in response to intradermal VEGFA injection in the ear dermis of control and Nrp1 iECKO mice.

(D) Leakage sites per vessel length in response to intradermal VEGFA stimulation in the ear skin of control and Nrp1 iECKO mice. n=6 mice, two or more acquisitions/mouse.

(E) Lag period between intradermal VEGFA injection and initiation of leakage in the ear skin of control and Nrp1 iECKO mice. n≥3 mice, two or more acquisitions/mouse, three or more sites/acquisition.

(F) Quantification of 2000 kDa dextran extravasation over time in the ear skin of control and Nrp1 iECKO mice following intradermal VEGFA stimulation. Black lines represent lines of best fit for the slope between leakage initiation and leakage termination. n≥3 mice, two or more acquisitions/mouse, three or more sites/acquisition.

(G-H) Leakage of fixable 2000 kDa FITC dextran in trachea (G) and back skin (H) after systemic administration of VEGFA in control and Nrp1 iECKO mice. Left, representative images. Right, quantification of tracer leakage area / vessel area (n ≥ 8 mice, 2 or more fields of view/mouse).

(I-J) Leakage of 70 kDa (I) and 2000 kDa (J) dextran extracted from perfused tissues after systemic administration of VEGFA in kidney, skeletal muscle and heart in control and Nrp1 iECKO mice (n ≥ 8 mice).

Error bars; mean ± SEM. Statistical significance: Two-tailed unpaired student’s t-test and linear regression with ANCOVA. Scale bar: 100 μm

Global loss of NRP1 reduces VEGFA-mediated vascular permeability

(A) Schematic illustration of experimental design to induce recombination in Nrp1 iKO mice.

(B) Western blot quantification of lung lysates from tamoxifen-treated control and Nrp1 iKO mice (n ≥ 2 mice).

(C) Representative images showing leakage of 2000 kDa FITC dextran (Pseudo-colour) in response to intradermal VEGFA injection in the ear dermis of control and Nrp1 iKO mice.

(D) Leakage sites per vessel length in response to intradermal VEGFA stimulation in the ear skin of control and Nrp1 iKO mice. n=3 mice, two or more acquisitions/mouse.

(E-F) Leakage of fixable 2000 kDa FITC dextran in trachea (E) and back skin (F) after systemic administration of VEGFA in control and Nrp1 iKO mice. Left, representative images. Right, quantification of tracer leakage area / vessel area (n ≥ 5 mice, 3 or more fields of view/mouse).

Error bars; mean ± SEM. Statistical significance: Two-tailed unpaired student’s t-test. Scale bar: 100μm

Heterogeneous expression of perivascular NRP1

(A) Schematic illusatration of mouse strain and representation of NRP1 expression in endothelial, and perivascular cells using Pdgfrβ-GFP mice. White dashed line encompasses vascular area, space between white and yellow dashed lines represents perivascular area. Note perivascular, NRP1 expressing cells in between yellow and white lines.

(B-D) Images, left, and quantification, right, showing vascular and perivascular NRP1 expression in arterioles (A), capillaries (B) and venules (C) of ear skin and back skin and its relative expression.

Error bars; mean ± SEM. Statistical significance: Two-tailed unpaired student’s t-test. Scale bar: 50μm

NRP1 distribution modifies VEGFA-induced vascular permeability

(A) Schematic illustration of breeding scheme to obtain homozygous Vegfr2 Y949F/Y949F; Nrp1 iECKO mice and experimental design to induce recombination.

(B) Western blot quantification of lung lysates from tamoxifen treated Vegfr2 Y949F/Y949F and Vegfr2Y949F/Y949F; Nrp1 iECKO mice (n=3 mice).

(C) Representative images showing leakage in response to intradermal VEGFA injection in the ear dermis of Vegfr2 Y949F/Y949F and Vegfr2 Y949F/Y949F; Nrp1 iECKO mice.

(D) Leakage sites per vessel length in response to intradermal VEGFA stimulation in the ear skin of in control, Nrp1 iECKO, Vegfr2 Y949F/Y949F and Vegfr2 Y949F/Y949F; Nrp1 iECKO mice, normalized to control (n ≥ 6 mice, two or more acquisitions/mouse).

(E-F) Leakage of 2000 kDa dextran in trachea (E) and back skin (F) of control, Nrp1 iECKO, Vegfr2 Y949F/Y949Fand Vegfr2 Y949F/Y949F; Nrp1 iECKO mice after systemic administration of VEGFA. Left, representative images. Right, quantification of tracer leakage area/vessel area (n ≥ 3 mice, 2 or more fields of view/mouse).

Error bars; mean ± SEM. Statistical significance: Two tailed unpaired student’s t-test and Two-way ANOVA. Scale bar: 100μm

Perivascular NRP1 modifies VEGFA-induced signalling

(A-B) Phosphorylation of VE-Cadherin (VEC) Y685 in response to intradermal PBS or VEGFA injections in the ear dermis of control and Nrp1 iECKO (A) or Nrp1 iKO (B) mice. Left, representative images.

Right, quantification of phosphorylated VEC Y685 area per total VEC area normalized to PBS control in the ear dermis of control and Nrp1 iECKO or Nrp1 iKO mice. n≥3 mice, two or more fields of view/mouse.

(C-D) Phosphorylation of VE-Cadherin Y685 in response to intradermal PBS or VEGFA injections in the back skin of control and Nrp1 iECKO (C) or Nrp1 iKO (D) mice. Left, representative images. Right, quantification of phosphorylated VEC Y685 area per total VEC area normalized to PBS control in the back skin of control and Nrp1 iECKO or Nrp1 iKO mice. n≥3 mice, two or more fields of view/mouse.

Error bars; mean ± SEM. Statistical significance: Two-way ANOVA. Scale bar: 50μm

Schematic model of NRP1’s spatial effects on VEGFA-induced vascular permeability

With global loss of NRP1, VEGFA/VEGFR2 signalling regulating vascular leakage is suppressed and thus leakage is reduced (left). In the presence of EC NRP1, a VEGFA/VEGFR2/NRP1 complex is formed in cis, which has a rapid and transient effect on VEGFR2 downstream signalling (middle). Loss of EC NRP1 but the presence of perivascular NRP1 leads to formation of a stable VEGFA/VEGFR2/NRP1 complex in trans and enhanced VEGFR2 signalling and increased vascular permeability.

NRP1 expression in the ear dermis and back skin of adult mice

(A-B) Whole mount immunostainings showing NRP1 expression in ear dermis (A) and back skin (B) of control, Nrp1 iECKO and Nrp1 iKO mice. Scale bar: 50μm

Organotypic assessment of VEGFA-induced leakage

(A,B) Leakage of 10 kDa (A) and 70 kDa (B) dextran after systemic administration of only dextran (control) and dextran plus VEGFA in liver, kidney, skeletal muscle and heart of wild-type C57Bl/6 mice (n ≥ 3 mice).

(C-E) Leakage of 2000 kDa dextran after systemic administration of only dextran (control) and dextran plus VEGFA in ear dermis (C), back skin (D) and trachea (E) of wild-type C57Bl/6 mice (n ≥ 2 mice).

Error bars; mean ± SEM. Statistical significance used: Two-tailed unpaired student’s t-test

Loss of EC NRP1 reduces VEGFA mediated vascular permeability in the back skin

Evans blue leakage with intradermal administration of VEGFA in the back skin of Nrp1 iECKO mice. Left, representative images. Right, quantification of Evans blue extravasation shown as VEGFA induced leakage fold over PBS (n ≥ 3 mice).

Error bars; mean ± SEM. Statistical significance used: Two-tailed unpaired student’s t-tes

Global inactivation of NRP1 reduces VEGFA mediated vascular permeability

(A) VEGFA-induced leakage in the ear skin of C57Bl/6 mice treated intradermally with isotype control or NRP1 blocking antibody. Left, representative images. Right, quantification of leakage sites per 100 µm of vessel length (n=3 mice, 3 acquisitions/mouse).

(B) Leakage of 2000 kDa dextran in trachea of C57Bl/6 mice treated systemically with isotype control or NRP1 blocking antibody. Left, representative images. Right, quantification of tracer area/vessel area (n=3 mice).

(C) PBS and VEGFA-induced leakage in the back skin of C57Bl/6 mice treated intradermally with isotype control or NRP1 blocking antibody. Left, representative images. Right, quantification of Evans blue leakage (n=3 mice).

Error bars; mean ± SEM. Statistical significance used: Two-tailed unpaired student’s t-test. Scale bar: 100μm

Perivascular NRP1 modifies PLCγ phosphorylation

(A-B) Phosphorylation of PLCγ in response to intradermal PBS or VEGFA injections in the ear dermis of control and Nrp1 iECKO (A) or Nrp1 iKO (B) mice. Left, representative images. Right, quantification of phosphorylated PLCγ area per total VEC area normalized to PBS control in the ear dermis of control and Nrp1 iECKO or Nrp1 iKO mice. n≥3 mice, two or more fields of view/mouse.

(C-D) Phosphorylation of PLCγ in response to intradermal PBS or VEGFA injections in the back skin of control and Nrp1 iECKO (C) or Nrp1 iKO (D) mice. Left, representative images. Right, quantification of phosphorylated PLCγ area per total VEC area normalized to PBS control in the back skin of control and Nrp1 iECKO or Nrp1 iKO mice. n≥3 mice, two or more fields of view/mouse.

Error bars; mean ± SEM. Statistical significance: Two-way ANOVA. Scale bar: 50μm