Dynamic alterations in decoy VEGF receptor-1 stability regulate angiogenesis

Abstract

Blood vessel expansion is driven by sprouting angiogenesis of endothelial cells, and is essential for development, wound healing and disease. Membrane-localized vascular endothelial growth factor receptor-1 (mVEGFR1) is an endothelial cell-intrinsic decoy receptor that negatively modulates blood vessel morphogenesis. Here we show that dynamic regulation of mVEGFR1 stability and turnover in blood vessels impacts angiogenesis. mVEGFR1 is highly stable and constitutively internalizes from the plasma membrane. Post-translational palmitoylation of mVEGFR1 is a binary stabilization switch, and ligand engagement leads to depalmitoylation and lysosomal degradation. Trafficking of palmitoylation enzymes via Rab27a regulates mVEGFR1 stability, as reduced levels of Rab27a impaired palmitoylation of mVEGFR1, decreased its stability, and elevated blood vessel sprouting and in vivo angiogenesis. These findings identify a regulatory axis affecting blood vessel morphogenesis that highlights exquisite post-translational regulation of mVEGFR1 in its role as a molecular rheostat.

Figure 1. mVEGFR1 is stable and constitutively internalizes in endothelial cells.

(a) Immunoblot of HUVEC treated as indicated for 8 h; CHX, cycloheximide; three replicates. (b) Immunoblot of HUVEC treated as indicated; three replicates. (c) Immunoblot of HUVEC or HBMVEC with indicated treatments; two replicates. (d,e) (d) Immunofluorescence for VEGFR1 or VEGFR2 in HUVEC angiogenic sprouts with indicated treatments, scale bar: 100 μm. (e) Quantification of fluorescence via integrated density. (VEGFR1 n=18 sprouts per condition; VEGFR2, n=25 sprouts per condition); three replicates. (f) VEGFR1 immunofluorescence of HUVEC with indicated treatments; three replicates, scale bar: 20 μm. (g) Immunoblot of HUVEC at 48 h post-knockdown (KD) with indicated treatments; two replicates. (h,i) (h) VEGFR1 immunofluorescence of HUVEC 24 h post KD with indicated treatments, scale bar: 20 μm. (i) Quantification of fluorescence via integrated density. (No. of cells: ntRNA/H₂O, n=22; ntRNA/CHX, n=23; siRab11a/H₂O, n=19; siRab11a/CHX, n=14); three replicates. Statistics: Shown are means +95% CI. One-way ANOVA with pairwise comparison and post-hoc Tukey's range test. *P≤0.05; **P≤0.01; NS, not significant.

Figure 2. Palmitoylation regulates mVEGFR1 stability.

(a,b) (a) VEGFR1 immunofluorescence of HUVEC with indicated treatments. (b) Quantification of fluorescence via integrated density. (No. of cells: EtOH/H₂O, n=20; EtOH/CHX, n=12; 2-BH/H₂O n=24; 2-BH/CHX n=18); CHX, cycloheximide; two replicates, scale bar: 25 μm. (c) Immunoblot of HUVEC with indicated treatments; four replicates. (d,e) (d) VEGFR1 immunofluorescence of d5 HUVEC angiogenic sprouts with indicated treatments; 2-BH, 2-bromohexadecadnoic acid, scale bar: 25 μm. (e) Quantification of fluorescence via integrated density. (No. of sprouts: EtOH/H₂O, n=20; EtOH/CHX, n=12; 2-BH/H₂O, n=24; 2-BH/CHX, n=18); two replicates. (f) Quantification of VEGFR1 fluorescence in HUVEC with indicated treatments via integrated density. (n=50 cells per condition); four replicates. (g) Immunoblot of HUVEC stimulated with 50 ng ml⁻¹ PLGF or PBS for indicated times; four replicates. (h) Quantification of VEGFR1 fluorescence of HUVEC with indicated treatments and 50 ng ml⁻¹ PLGF or VEGF-A via integrated density. Pal-B, palmostatin-B. (No. of cells: DMSO/PBS, n=51; DMSO/PLGF, n=37; DMSO/VEGF-A, n=45; Pal-B/PBS, n=50; Pal-B/PLGF n=48; Pal-B/VEGF-A, n=51); three replicates. (i) Immunoblot of HUVEC with indicated treatments and 50 ng ml⁻¹ PLGF or VEGF-A for 1 h. Values are mVEGFR1 normalized to respective PBS controls; three replicates. (j,k) (j) VEGFR1 immunofluorescence of d5 HUVEC angiogenic sprouts with indicated treatments and 75 ng ml⁻¹ PLGF or PBS for 4 h, scale bar: 25 μm. (k) Quantification of fluorescence via integrated density. (No. of sprouts: DMSO/PBS n=13; DMSO/PLGF, n=17; Pal-B/PBS n=12; Pal-B/PLGF n=16); two replicates. Statistics: Shown are means+95% CI. One-way ANOVA with pairwise comparison and post-hoc Tukey's range test. *P≤0.05; **P≤0.01; NS, not significant.

Figure 3. Palmitoylation regulates VEGFR1 trafficking.

(a,b) (a) Immunofluorescence of VEGFR1 and PECAM or (b) VEGFR1 and Rab11a in HUVEC with indicated treatments; 2-BH, 2-bromohexadecadnoic acid, scale bars: 20 and 10 μm for a,b, respectively. (c,e) Mander's Correlation Coefficient quantification of overlap of VEGFR1 and PECAM (c) or Rab11a (e) in HUVEC treated as indicated. (No. of cells: VEGFR1/PECAM: EtOH, n=14; 2-BH, n=12; VEGFR1/Rab11: EtOH, n=23; 2-BH, n=27); three replicates. (d) Immunoblot of cell surface or intracellular VEGFR1 from HUVEC with indicated treatments. PKM1/2, cytoplasmic marker; H-Ras, specificity control. Statistics: Shown are means+95% CI. Student's t-test with post-hoc Tukey's range test. **P≤0.01.

Figure 4. VEGFR1 is palmitoylated in endothelial cells to regulate turnover.

(a) Workflow for acyl-resin assisted capture (Acyl-RAC). PA, palmitic acid; MMTS, methyl-methanethiosulfonate. (b) Acyl-RAC and immunoblot of HUVEC with indicated treatments. 2-BH, 2-bromohexadecadnoic acid; three replicates. (c) Acyl-RAC and immunoblot of VEGFR1 from HUVEC with indicated treatments for 45 min. CHQ, chloroquine; three replicates. (d) Diagram illustrating palmitic acid addition and removal. DHHC, protein acetyl-transferase; APT, acyl-protein thioesterase. (e,f) (e) VEGFR1 immunofluorescence of HUVEC with indicated treatments for 24 h, scale bar: 25 μm. (f) Quantification of fluorescence via integrated density. Shown are means+95% CI. (No. of cells: ntRNA, n=21; siDHHC3, n=22); four replicates. (g) Immunoblot of mVEGFR1 from HUVEC with indicated treatments; three replicates. (h) Acyl-RAC and immunoblot of HUVEC with indicated treatments for 24 h; three replicates. (i) Immunoblot of VEGFR1 and HA-tagged DHHC3 in HUVEC with indicated treatments for 24 h; three replicates. (j,k) Immunoblot of HUVEC with indicated treatments at 48 h post knockdown and treated with 50 ng ml⁻¹ VEGF-A (j) or PLGF (k); three replicates. (l–n) (l) HUVEC angiogenic sprouts with indicated treatments, scale bar: 100 μm. (m,n) Quantification of angiogenic parameters. (sprouts per bead no. of beads: ntRNA, n=25; siDHHC3, n=27; sprout length, no. of sprouts: ntRNA, n=288; siDHHC3, n=365); two replicates. Statistics: Students t-test and post-hoc Tukey's range test. **P≤0.01.

Figure 5. Rab27a regulates DHHC3 localization and palmitoylation of VEGFR1.

(a,b) DHHC3 and TGN46 (Golgi) immunofluorescence (a) and Mander's Correlation Coefficient quantification (b) of overlap in HUVEC with indicated treatments; three replicates, scale bar: 20 μm. (c) Subcellular fractionation and immunoblot of HUVEC with indicated treatments. Pyruvate kinase1/2 (PKM1/2), soluble marker; vascular-endothelial cadherin (VE-CAD), membrane and cytoskeletal marker; cis-Golgi Marker 130 (GM130), membrane marker; two replicates. (d,e) (d) VEGFR1 immunofluorescence of d4 HUVEC angiogenic sprouts with indicated siRNAs, scale bar: 25 μm. (e) Quantification of fluorescence via integrated density. (No. of sprouts: ntRNA, n=7; siRab27a, n=9; siRab3a, n=5); three replicates. (f) Immunoblot of HUVEC with indicated treatments. Values are relative mVEGFR1 levels. Doublet in VEGFR2 lane is likely due to glycosylation; four replicates. (g,h) (g) VEGFR1 immunofluorescence of HUVEC with indicated treatments, scale bar: 15 μm. (h) Quantification of fluorescence via integrated density. (No. of cells: ntRNA, H₂O, n=28; CHQ, n=22; siRab27a, H₂O, n=13; CHQ, n=15); two replicates. (i) Immunoblot of HUVEC with indicated treatments. Values are relative mVEGFR1 levels; 3 replicates. (j) Acyl-RAC and immunoblot of HUVEC with indicated siRNAs; three replicates. Statistics: Shown are means +95% CI. One-way ANOVA with pairwise comparison and post-hoc Tukey's range test (e,h); student's t-test (b). **P≤0.01; ***P≤0.001; NS, not significant.

Figure 6. Rab27a regulates vascular sprouting.

(a–c) (a) Aortic rings with indicated siRNAs. Small boxes, indicated stainings, scale bars: 500 and 25 μm for large and small boxes, respectively. (b,c) Quantification of angiogenic parameters after indicated treatments. (No. of rings, ntRNA=14; siRab27a=12); three replicates. (d–f) (d) HUVEC angiogenic sprouts with indicated treatments. Small boxes, indicated stainings, scale bars 50 and 25 μm for large and small boxes, respectively. (e,f) Quantification of angiogenic parameters. (sprouts per bead: ntRNA=12; siRab27a=14; sprout length: ntRNA=30; siRab27a=30); five replicates. (g,h) (g) Morphological analysis of HUVEC derived-sprouts with indicated treatments, scale bars: 25 and 10 μm, for large and small boxes, respectively. (h) Quantification of filopodia per μm of vessels. Arrows, filopodia. (ntRNA=10; siRab27a=14); five replicates. Statistics: Shown are means +95% CI. (b,c,e,f,h) Student's t-test. **P≤0.01; NS, not significant.

Figure 7. VEGFR1 is epistatic to Rab27a.

(a,b) (a) HUVEC angiogenic sprouts and (b) quantification of sprouting after indicated treatments, scale bar: 200 μm. (sprouts per bead, No. of beads: ntRNA=18; siRab27a=15; siVEGFR1=16); three replicates. (c) Immunoblot of HUVEC with indicated treatments; three replicates. (d–g) (d) HUVEC angiogenic sprouts with indicated treatments, scale bar: 200 μm. FC, control; R1-FC, recombinant human VEGFR1. (e) Quantification of sprouts per bead. (f) pERK immunofluorescence of HUVEC angiogenic sprouts with indicated treatments, scale bar: 20 μm. (g) pERK quantification via integrated density. (pERK fluorescence, No. of sprouts: ntRNA/FC=8; ntRNA/R1-FC=9; siRab27a/FC=10; siRab27a/R1-FC=7; sprouts per bead, No. of beads: ntRNA/FC=14; ntRNA/R1-FC=14; siRab27a/FC=12; siRab27a/R1-FC=15); three replicates. (h,i) (h) pVEGFR2 (Y1175) Immunofluorescence of HUVEC angiogenic sprouts with indicated treatments, scale bar: 10 μm. (i) pVEGFR2 quantification via integrated density. (No. of sprouts: ntRNA=9; siRab27a=9); three replicates. Statistics: Shown are means +95% CI. One-way ANOVA and pairwise comparison with post-hoc Tukey's range test. *P≤0.05; **P≤0.01.

Figure 8. Ashen retinal vessels have reduced FLT-1 and excess filopodia.

(a–c) (a) FLT-1 immunofluorescence of retinas from P5 mice with indicated genotypes, scale bars: 50 μm. (b,c) Quantification of FLT-1 in indicated areas via integrated density. Scale bar, 25 μm. (No. of retinas: +/+, n=4; ash/+, n=6; ash/ash, n=4). (d–g) (d) Immunofluorescence of retinas. Quantification of (e) filopodia in the plexus (f) sprouting, and (g) filopodia at the vascular front from P5 mice with indicated genotypes. Red box, magnified at right; arrows, filopodia. Scale bars: 25 μm. (No. of retinas from two independent litters: +/+, n=6, ash/+ n=5; ash/ash, n=7). (h,i) Immunoblot for mFLT-1 in retinal lysates from indicated genotypes at P5 (h) and P8 (i); four replicates per stage. (j) Acyl-RAC and immunoblot for mFLT-1 and Rab27a from retinal lysates at indicated stages; three replicates. (k) Immunoblot of retinal lysates as indicated; three replicates. Statistics: Shown are means +95% CI. One-way ANOVA and pairwise comparison with post-hoc Tukey's range test. *P≤0.05; **P≤0.01; ***P≤0.001; NS, not significant.

Figure 9. Model for regulation of mVEGFR1 stability in endothelial cells.

mVEGFR1 is normally (left) highly stable and slowly recycled to and from the surface via Rab11a, requiring Rab27a and DHHC3 for palmitoylation; reduced Rab27a levels (right) destabilize mVEGFR1 via reduced palmitoylation, leading to increased signalling through VEGFR2.