FLVCR1a expression is enhanced in angiogenic endothelial cells (aECs). A qRT-PCR analysis showing Flvcr1a mRNA levels in wild-type mouse embryos at different developmental stages (i.e., embryonic day (E) 9.5, 11.5, E13.5, E15.5). n = 4. B qRT-PCR analysis showing Flvcr1a transcript levels in embryonic ECs (CD31+) and non-endothelial fractions (CD31−) collected from mouse embryos at E9.5 and E13.5. n = 3. C qRT-PCR analysis showing Flvcr1a mRNA levels in embryonic angiogenic ECs (E9.5) and tumor-associated ECs (TECs) from Lewis Lung Carcinoma Cells (LLC)-xenografts. D qRT-PCR analysis showing Flvcr1a transcript levels in mouse ECs isolated from adult lung and LLC tumors. E, F Representative Western Blot analysis (E) and quantification (F) of FLVCR1a-myc-tagged protein expression in murine ECs from adult lung and LLC tumor. GFLVCR1a expression levels in human qECs (capillary ECs) and aECs (immature/stalk and tip cells) isolated from lung adenocarcinoma. Data were taken from public available database. aECs angiogenic ECs, NEC/TEC Normal/Tumor ECs, RQ Relative Quantification. Data are representative of at least 3 independent experiments and are expressed as mean ± SEM. *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001. For statistical analyses, ordinary one-way ANOVA test with Tukey’s multiple comparisons (A, B), parametric unpaired t test (C, D, F), and Mann–Whitney test (G) were used

Flvcr1a targeting impairs developmental angiogenesis in mouse and zebrafish embryos. A(i) Phenotype of E13.5 control (Flvcr1a^fl/fl) and Flvcr1a endothelial-specific constitutive knockout (Flvcr1a^fl/fl;Tie2-Cre) mouse embryos. (ii) Whole-mount staining with CD31/Pecam-1 antibody showing the embryonic vasculature in E11.5 embryos. Enlarged pictures of forelimbs are shown. White dashed line refers to the transverse section displayed in panels iii-iv. (iii-iv) Transverse sections of E11.5 forelimb from Flvcr1a^fl/fl and Flvcr1a^fl/fl;Tie2-Cre embryos stained with CD31/Pecam-1 antibody. B Quantification of the average vessel size in Flvcr1a^fl/fl and Flvcr1a^fl/fl;Tie2-Cre limbs transverse sections shown in (iii). C, D Photographs of transgenic (Tg) kdrl:GFP zebrafish embryos injected with CTRL, MO-ATG, or MOl1EX2 morpholinos alone or with Flvcr1a cRNA. Images were collected at 48 hpf. Magnification of intersegmental vessels in controls, morphants and rescued embryos, are shown. ****p < 0.0001. For statistical analyses, parametric unpaired t test was used. A.U. arbitrary units, MO morpholino, hpf hours post fertilization

Retinal angiogenesis is compromised in endothelial Flvcr1a-null mice. A Schematic representation of tamoxifen (TAM) treatment and retina dissection in Flvcr1a;Cdh5-Cre^ERT2 and Flvcr1a control mice. Briefly, pups were injected with TAM into the stomach for 3 consecutive days (P1–P3) to induce Flvcr1a gene deletion in the developing retinal vasculature. Retinas were isolated at P8 to analyze angiogenesis. B Representative PCR products from DNA analysis on Flvcr1a^fl/fl and Flvcr1a^fl/fl;Cdh5-Cre^ERT2 pups after tamoxifen treatment. Deletion of the first exon of Flvcr1a gene mediated by Cre recombinase gave rise to a band referred to as ‘‘null allele.’’ Specific primers were used to amplify Cre recombinase (270 bp), Flvcr1a floxed (280 bp), and Flvcr1a null (320 bp) alleles. C Whole-mount CD31 staining (green) of P8 retinal primary vascular plexus from Flvcr1a^fl/fl and Flvcr1a^fl/fl;Cdh5-Cre^ERT2 mice. Scale bar: 200 µm. D–I Quantification of vascular parameters of Flvcr1a^fl/fl and Flvcr1a^fl/fl;Cdh5-Cre^ERT2 P8 retinas, i.e., D radial expansion, E vascular area, F average nearest neighbor distance (NND), G vessel distribution variability, H vascular total length, I number of branches. J CD31 (gray)/pHH3 (red) double staining of Flvcr1a^fl/fl and Flvcr1a^fl/fl;Cdh5-Cre^ERT2 P8 retinas, and K quantification of pHH3⁺ ECs. Scale bar: 100 µm. L Higher magnification of the angiogenic front in Flvcr1a^fl/fl and Flvcr1a^fl/fl;Cdh5-Cre^ERT2 P8 retinas stained with CD31 antibody to highlight the vasculature. Tip cells are white circled. Scale bar: 100 µm. M Tip cells and N filopodia quantification is shown. O Whole-mount staining of Flvcr1a^fl/fl and Flvcr1a^fl/fl;Cdh5-Cre^ERT2 P8 retinas with CD31 (green) and NG2 (red) antibodies to highlight ECs and pericyte, respectively. Scale bar: 200 µm. P Quantification of pericyte coverage as % ratio of NG2 area and CD31 area. Data are representative of at least 3 independent experiments and are expressed as mean ± SEM. *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001; For statistical analyses ,an unpaired Student’s t test was used. TAM tamoxifen, P post-natal day, bp base pairs, A.U. arbitrary unit

Endothelial Flvcr1a is dispensable in adult healthy tissues. A Representative histological sections of lung and liver tissue from Flvcr1a^fl/fl and Flvcr1a^fl/fl;Cdh5-Cre^ERT2 mice. (i, ii) Tissue sections stained with hematoxylin and eosin (HE) to look at the cellular morphology. Scale bar: 200 µm. (iii) Confocal images of liver sections from Flvcr1a^fl/fl and Flvcr1a^fl/fl;Cdh5-Cre^ERT2 stained with CD31 antibody (green) and DAPI (blu) to look at vascular network organization. Scale bar: 200 µm. B Liver vascular area quantification expressed as % of liver area positive to CD31. C–F Gadolinium-based DCE-MRI analysis to evaluate (C) the total vascular volume (Vp) and D–F vessel permeability in Flvcr1a^fl/fl and Flvcr1a^fl/fl;Cdh5-Cre^ERT2 livers. HE, hematoxylin/eosin DCE-MRI, Dynamic Contrast-Enhanced Magnetic Resonance Imaging. Data are representative of at least 3 independent experiments and are expressed as mean ± SEM. **p < 0.01; ***p < 0.001; ns not significant. For statistical analyses, an unpaired Student’s t test was used

Endothelial Flvcr1a deficiency impairs tumor angiogenesis. A Schematic representation of tamoxifen (TAM) treatment in Flvcr1a^fl/fl;Cdh5-Cre^ERT2 and Flvcr1a^fl/fl control mice. B Confocal images of CD31 (green) and DAPI (blu)-stained tumor sections from Flvcr1a^fl/fl and Flvcr1a^fl/fl;Cdh5-Cre^ERT2 to highlight the tumor vasculature. C Tumor vascular area quantification as % CD31 of tumor area. Scale bar: 200 µm. D Tumor vascularization expressed as number of vessels/mm². E Gadolinium-based DCE-MRI analysis to measure tumor vascular volume (Vp) in Flvcr1a^fl/fl and Flvcr1a^fl/fl;Cdh5-Cre^ERT2 animals. E–G Quantification of (E) average vessel size, F vessel area distribution and G average vessel perimeter in Flvcr1a^fl/fl and Flvcr1a^fl/fl;Cdh5-Cre^ERT2 tumors. H Representative images showing the ImageJ “skeletonize tool” used to infer network descriptive parameters (e.g., vessel perimeter, Feret’s diameter) on tumor sections. I–J Histograms showing quantification of I average vessel perimeter and J vessel tortuosity (expressed as perimeter/Feret’s diameter) in Flvcr1a^fl/fl and Flvcr1a^fl/fl;Cdh5-Cre^ERT2 tumor sections. K High magnification of CD31 (green) and DAPI (blu)-stained Flvcr1a^fl/fl and Flvcr1a^fl/fl;Cdh5-Cre^ERT2 tumor vessels. ECs nuclei are comprised in yellow dashed lines. Scale bar: 20 µm. L Quantification of average number EC nuclei per vessel perimeter in Flvcr1a^fl/fl and Flvcr1a^fl/fl;Cdh5-Cre^ERT2 tumors. M Confocal images of CD31 (green) and NG2 (red) stained thick tumor Sects. (20 µm) from Flvcr1a^fl/fl and Flvcr1a^fl/fl;Cdh5-Cre^ERT2 mice. Scale bar: 50 µm. N Quantification of pericyte coverage as % NG2 area/CD31 area. LLCs Lewis Lung Carcinoma cells, TAM tamoxifen. Data are representative of at least 3 independent experiments and are expressed as mean ± SEM. *p < 0.05; ***p < 0.001; ****p < 0.0001. For statistical analyses, an unpaired Student’s t test was used

Endothelial Flvcr1a targeting increases tumor vessel permeability and hypoxia. A–D DCE-MRI analysis to evaluate tumor vessel permeability in LLC xenografts developed in Flvcr1a^fl/fl;Cdh5-Cre^ERT2 and Flvcr1a^fl/fl mice. The B enhancement curve, C K^trans constant, and D k^ep constant are shown. E Immunofluorescence analysis on Flvcr1a^fl/fl and Flvcr1a^fl/fl;Cdh5-Cre^ERT2 tumor sections stained with hypoxia marker pimonidazole (red) and DAPI (blu). Scale bar: 200 µm.; ECs endothelial cells, PIMO pimonidazole, DCE-MRI Dynamic Contrast-Enhanced Magnetic Resonance Imaging. Data are representative of at least 3 independent experiments and are expressed as mean ± SEM. **p < 0.01; For statistical analyses, an unpaired Student’s t test was used

FLVCR1a modulates the angiogenic potential of human tumor-derived ECs in vitro. A qRT-PCR analysis showing FLVCR1a mRNA levels in normal (Hmec) and tumor-associated (Btec) human ECs. B Western Blot analysis showing FLVCR1a protein expression in Btec and Hmec. The protein quantification is shown in C. Proliferation of control Btec (scr), FLVCR1a-silenced Btec (flv), and Hmec at multiple time points. E in vitro tubulogenesis assay performed with control (scr) and FLVCR1a-silenced (flv) Btec. F–H Quantification of F number of nodes, number of G master junctions, and H total length of the networks obtained with Btec-scr and Btec-flv. I Transendothelial resistance (TEER) measurement performed on control (scr) and FLVCR1a-silenced (flv) Btec. Data are representative of at least 3 independent experiments and are expressed as mean ± SEM. *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001. For statistical analyses, an unpaired Student’s t test was used

Endothelial FLVCR1a is required for proper angiogenesis. A Schematic representation of endothelial FLVCR1a expression during angiogenic vascular growth. FLVCR1a is low in the quiescent vasculature and strongly increases in aECs that have to proliferate and arrange in a new vascular network. Once the de novo vasculature is fully formed and mature, FLVCR1a levels decrease. B Schematic representation of vascular defects upon endothelial Flvcr1a targeting. Given the importance of FLVCR1a in aECs, FLVCR1a deficiency does not affect fully formed vascular networks. Conversely, endothelial-specific FLVCR1a targeting severely compromises neo-angiogenesis by reducing vascular expansion, ECs proliferation, and pericyte recruitment while increasing vessel permeability