Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2018 Apr 3;9(1):1296.
doi: 10.1038/s41467-018-03692-0.

Heterogeneity in VEGFR3 Levels Drives Lymphatic Vessel Hyperplasia Through Cell-Autonomous and Non-Cell-Autonomous Mechanisms

Affiliations
Free PMC article

Heterogeneity in VEGFR3 Levels Drives Lymphatic Vessel Hyperplasia Through Cell-Autonomous and Non-Cell-Autonomous Mechanisms

Yan Zhang et al. Nat Commun. .
Free PMC article

Abstract

Incomplete delivery to the target cells is an obstacle for successful gene therapy approaches. Here we show unexpected effects of incomplete targeting, by demonstrating how heterogeneous inhibition of a growth promoting signaling pathway promotes tissue hyperplasia. We studied the function of the lymphangiogenic VEGFR3 receptor during embryonic and post-natal development. Inducible genetic deletion of Vegfr3 in lymphatic endothelial cells (LECs) leads to selection of non-targeted VEGFR3+ cells at vessel tips, indicating an indispensable cell-autonomous function in migrating tip cells. Although Vegfr3 deletion results in lymphatic hypoplasia in mouse embryos, incomplete deletion during post-natal development instead causes excessive lymphangiogenesis. Analysis of mosaically targeted endothelium shows that VEGFR3- LECs non-cell-autonomously drive abnormal vessel anastomosis and hyperplasia by inducing proliferation of non-targeted VEGFR3+ LECs through cell-contact-dependent reduction of Notch signaling. Heterogeneity in VEGFR3 levels thus drives vessel hyperplasia, which has implications for the understanding of mechanisms of developmental and pathological tissue growth.

Conflict of interest statement

The authors declare no competing interests.

Figures

Fig. 1
Fig. 1
VEGFR3 is indispensable for tip cell function during embryonic dermal lymphatic vessel sprouting. a Schematic of the genetic constructs and 4-OHT administration schedule (Cre induction, red (6 × 1 mg); estimated effective period (24 h), light red). The timing of dermal lymphatic vessel formation in the dorsal skin is indicated. b Whole-mount immunofluorescence of E17.5 Vegfr3flox;R26-mTmG;Prox1-CreERT2 skin. Boxed areas are magnified and single channel images for VEGFR3 staining are shown. Note efficient depletion of VEGFR3 in the Cre-targeted (GFP+) LECs and the presence of non-targeted cells at the tips of hypoplastic vessel sprouts (arrows) in the mutant (flox/flox) skin. cg Quantification of dermal lymphatic vessel parameters in E17.5 Vegfr3flox;R26-mTmG;Prox1-CreERT2 embryos. Bars represent mean (n = 4–10 embryos, as indicated) ± s.e.m. h, i Whole-mount analysis of E16.5 skin after mosaic deletion of one (flox/+) or two (flox/flox) Vegfr3 alleles by a single 4-OHT (1 mg) administration at E13.5. j Quantification of Cre-targeted (GFP+) LECs at the dorsal midline (area depicted in i). Bars represent mean (n = 5–6 embryos, as indicated) ± s.e.m. * P < 0.05, *** P < 0.001. Two-tailed unpaired Student’s t test (cg, j). Scale bars: 200 µm (b, i). ns: not significant
Fig. 2
Fig. 2
Early post-natal deletion of Vegfr3 results in dermal lymphatic vessel hyperplasia during a critical post-natal period of 2 weeks. a Time course analysis of the effect of post-natal Vegfr3 deletion on dermal lymphatic development. Whole-mount immunofluorescence of ear skin of indicated stages and Tam/4-OHT treatment regimes. b Quantification of Cre-targeted (GFP+, Vegfr3 deleted) LECs in the distal vessel tips in P21 mice treated with Tamoxifen at P2 and P3. Bars represent mean (n = 6 fl/+ and n = 4 fl/fl mice) ± s.e.m. c Quantification of vessel branch points in ear skin analyzed in Fig. 2a after indicated Tam treatment regimes. Tamoxifen administration schedule (Cre induction, red (n × 150 µg); estimated effective period (72 h), light red) (left) is shown. Bars represent mean (n = 3–5 mice, as indicated) ± s.e.m. d Whole-mount immunofluorescence of ear skin of 5 weeks old mice treated with Tamoxifen at P2, P4 and P6. e, f Quantification of vessel branch points and area in ear skin analyzed in Fig. 2d. Bars represent mean (n = 3–6 mice, as indicated) ± s.e.m. *P < 0.05, ***P < 0.001, ****P < 0.0001. Two-tailed unpaired Student’s t test (b, c, e, f). Scale bars: 200 µm (a, d). ns: not significant
Fig. 3
Fig. 3
Lymphatic vascular hyperplasia in Vegfr3-deleted ears is driven by VEGF-C signaling. a Quantification of blunt-ended vessels (lymphatic capillaries) in 5 weeks old mice treated with Tamoxifen at P2, P4, and P6. Central region excludes 560 μm area (tip) from the edge of the ear. Bars represent mean (n = 3–6 mice, as indicated) ± s.e.m. b Whole-mount immunofluorescence of Vegfr3flox/flox;R26-mTmG;Prox1-CreERT2 ear skin showing vessel interconnections (arrows) formed of non-targeted (GFP) LYVE1+ (left) and VEGFR3+ (right) LECs. c Whole-mount immunofluorescence of ear skin of 3 weeks old Vegfr3flox/flox;R26-mTmG;Prox1-CreERT2 and Cre-negative littermate mice treated with Tamoxifen at P2, P4, and P6, and inhibitors of VEGF-C signaling (the VEGF-C-trap VEGFR3-Ig, the VEGFR3 inhibitor MAZ51, or the VEGFR2 blocking antibody DC101). Quantification of (d) vessel area and (e) branch points. Bars represent mean (n = 3 (untreated groups) or n = 4 (treated groups) mice) ± s.e.m. fh Visualization by whole-mount immunofluorescence (arrows in f) and quantification of blunt-ended vessels in the (g) entire and (h) central region of the ear in Vegfr3flox/flox;R26-mTmG;Prox1-CreERT2 mice treated with the VEGFR3 kinase inhibitor MAZ51. In g, h, bars represent mean (n = 3 (untreated) or n = 4 (MAZ51, DC101) mice) ± s.e.m. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001. Two-tailed unpaired Student’s t test (a, d, e, h). Scale bars: 200 µm (b, c, f). ns: not significant
Fig. 4
Fig. 4
Excessive proliferation of non-targeted VEGFR3+ LECs contributes to lymphatic vessel hyperplasia in Vegfr3-deleted ears. a, b Fluorescence-activated cell sorting (FACS) evaluation of Tomato and GFP expression in Vegfr3flox/+ and Vegfr3flox/flox;R26-mTmG;Prox1-CreERT2 LECs. Tamoxifen (150 µg) was administered daily from P2 until harvest at P11–P12. Representative dot plots of GFP and Tomato expression, and histogram of Tomato expression in GFP+ LECs are displayed with Cre littermate as a control. In b, graph of combined results from two litters, showing the average distribution of cells expressing Tomato only (red), GFP and Tomato (yellow), and GFP only (green). Dots show % of GFP+ LECs in individual mice, horizontal line represents mean (n = 3 mice) ± s.e.m. c Vegfr3 exon1 deletion (left) and mRNA expression (right) in sorted Tomato+GFP (red), Tomato+GFP+ (yellow), and TomatoGFP+ (green) LEC populations from P11–P14 Vegfr3flox/flox mutants (R3 fl/fl). Tamoxifen (150 µg) was administered at P2, P4, and P6. LECs from Cre mice (Ctrl) were used as controls with the average ΔCT cycle threshold as the reference value for relative quantification. Horizontal line represents mean (n = 3–6 mice). d, e 5-ethynyl-2'-deoxyuridine (EdU) incorporation evaluated by FACS analysis in LECs gated on Tomato and GFP expression, shown in Fig. 3c. Representative histograms of EdU incorporation evaluated 24 h after injection and graph of combined results from two litters are shown in e. Horizontal line represents mean (n = 3–5 mice). f Proposed model for the mechanism underlying post-natal Vegfr3 loss-induced lymphatic vessel hyperplasia in the ear. Incomplete deletion of Vegfr3 promotes selection of non-targeted VEGFR3+ cells to the tips of vessel sprouts and their excessive proliferation and vascular anastomosis. Proliferation of non-targeted VEGFR3+ cells is sustained by continuous induction of new VEGFR3- neighbors through repeated Tamoxifen administrations during a critical post-natal period of 2 weeks. **P < 0.01, ***P < 0.001. Two-tailed unpaired Student’s t test (e). ns: not significant
Fig. 5
Fig. 5
Vegfr3-deleted LECs induce proliferation of neighboring VEGFR3+ cells in a cell-contact-dependent manner. a VEGF-C concentration measured by ELISA in the ear skin at indicated stages. Bars represent mean (n = 2–8 mice, as indicated) ± s.d. b Experimental set up for co-cultures of WT and VEGFR3 (KO) primary LECs from Vegfr3flox/flox;R26-mTmG;Prox1-CreERT2 mice (left) and immunofluorescence showing VEGFR3 depletion in Cre targeted (GFP+) LECs (right, arrows). c Left: proliferating WT cells were frequently associated with KO cells. Right: quantification of the proportion of EdU+ WT cells that were in direct contact with KO cells (mean (n = 6 biological replicates) ± s.e.m). d Assessment of cell proliferation in co-cultures of primary human LECs treated with control (siCTRL) or VEGFR3 (siVEGFR3) siRNA. Cell tracker-labeled siCTRL cells (purple) were mixed with unlabeled siVEGFR3 cells in 40:60%, or they were cultured alone. e Quantification of proliferating EdU+ cells in co-cultures of siCTRL and siVEGFR3 LECs mixed in different ratios. Data are normalized to the proliferation rate (0.8 ± 0.2%) in siCTRL alone and represent mean (n = 5 biological replicates) ± s.e.m. f Predicted and observed proportions of siCTRL LECs (all vs. EdU+) that are in direct contact with siVEGFR3 LECs in 80:20% siCTRL:siVEGFR3 co-cultures (mean (n = 5 biological replicates) ± s.e.m). Predicted proportion was calculated based on random distribution and n = 5 (5.2 ± 0.2, quantified from n = 172 LECs stained for VE-cadherin) neighbors. g Quantification of proliferating EdU+ cells in siCTRL LECs cultured alone (−), or co-cultured without direct cell–cell contacts with siVEGFR3 cells seeded on Transwell inserts (+). Data are normalized to the proliferation rate in siCTRL alone and represent mean (n = 6 biological replicates) ± s.e.m. *P < 0.05, **P < 0.01, ***P < 0.001. Two-tailed unpaired Student’s t test (a, e, g) and Fisher’s exact test (f). Scale bars: 50 µm (b, c), 200 µm (d). ns: not significant
Fig. 6
Fig. 6
DLL4 downregulation and decrease in Notch signaling upon loss of VEGFR3. a qRT-PCR analysis of Notch pathway genes in control (siCTRL) and VEGFR3 siRNA (siVEGFR3)-treated human LECs. Bars represent mean relative expression (n = 3 biological replicates with n = 2 technical replicates each) ± s.d. b qRT-PCR analysis of Dll4 and Notch1 expression in FACS-sorted LECs from P13-P14 Cre (Ctrl) and Vegfr3flox/flox;R26-mTmG;Prox1-CreERT2 (R3 fl/fl) mice. Tamoxifen (150 µg) was administered at P2, P4, and P6. Bars represent mean relative expression (n = 5 Ctrl and n = 4 R3 fl/fl mice) ± s.d. c Whole-mount immunofluorescence of ear skin of 3 weeks old Vegfr3flox/flox;R26-mTmG;Prox1-CreERT2 and Cre-negative littermate mice treated with Tamoxifen at P2, P4, and P6. Note downregulation of DLL4 protein in targeted GFP+ LECs (arrows) compared to non-targeted GFP LECs (arrowhead) in the mutant (flox/flox) ear. d Hey1, (e) Efnb2 and (f) Ccnb1 expression analyzed by qRT-PCR in FACS-sorted LECs from P13-P14 Cre (Ctrl) and Vegfr3flox/flox;R26-mTmG;Prox1-CreERT2 (R3 fl/fl) mice. Results from non-targeted (Tomato+GFP; red), and targeted (TomatoGFP+; green) LECs are displayed separately. Horizontal lines represent mean relative expression (n = 5 Ctrl and n = 4 R3 fl/fl mice with n = 2 technical replicates for each). *P < 0.05, **P < 0.01, ***P < 0.001. Two-tailed unpaired Student’s t test (a, b, df). Scale bars: 50 µm (c). ns: not significant
Fig. 7
Fig. 7
Global and mosaic loss of DLL4 promote LEC proliferation through inhibition of Notch signaling in neighboring cells. a, b DAPT induced effect on the proliferation of control (siCTRL) or VEGFR3 (siVEGFR3) siRNA-treated LECs. a Tile scan immunofluorescence images and (b) quantification of proliferating EdU+ LECs after 3 h of incorporation (mean (n = 3 biological replicates) ± s.e.m). c Western blot analysis of cell lysates from LECs treated with siCTRL or NOTCH1 siRNA (siNOTCH1) and (d) quantification of proliferating EdU+ LECs cultured with or without VEGF-C (data are normalized to the proliferation rate in siCTRL alone and represent mean (n = 3 biological replicates) ± s.e.m). e Western blot analysis of cell lysates from LECs treated with siCTRL or DLL4 siRNA (siDLL4) and (f) quantification of proliferating EdU+ cells in co-cultures of siCTRL and siDLL4 LECs mixed in different ratios (data are normalized to the proliferation rate in siCTRL alone and represent mean (n = 6 biological replicates) ± s.e.m). g Model of cell-autonomous and non-cell-autonomous effects of VEGFR3 expression defining LEC responses during sprouting lymphangiogenesis. VEGF-C-VEGFR3 signaling positively regulates DLL4 expression leading to activation of Notch and cell cycle arrest in neighboring LECs. VEGFR3 deficiency (green cell) consequently leads to downregulation of DLL4 and inhibition of Notch in neighboring LECs, but only VEGFR3 expressing (red) cells respond by proliferation and sprouting. When VEGFR3 cells are in excess, VEGFR3+ cells interact predominantly with LECs with low DLL4 and respond by proliferation (dashed line) *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001. Two-tailed unpaired Student’s t test (b, d, f) or one-way ANOVA with Tukey’s post hoc test (d; comparison between VEGF-C untreated and treated groups). Scale bars: 200 µm (a). ns: not significant

Similar articles

See all similar articles

Cited by 4 articles

References

    1. Potente M, Mäkinen T. Vascular heterogeneity and specialization in development and disease. Nat. Rev. Mol. Cell Biol. 2017;18:477–494. doi: 10.1038/nrm.2017.36. - DOI - PubMed
    1. Augustin HG, Koh GY. Organotypic vasculature: from descriptive heterogeneity to functional pathophysiology. Science. 2017;357:eaa12379. doi: 10.1126/science.aal2379. - DOI - PubMed
    1. Aspelund A, Robciuc MR, Karaman S, Makinen T, Alitalo K. Lymphatic system in cardiovascular medicine. Circ. Res. 2016;118:515–530. doi: 10.1161/CIRCRESAHA.115.306544. - DOI - PubMed
    1. Geudens I, Gerhardt H. Coordinating cell behaviour during blood vessel formation. Development. 2011;138:4569–4583. doi: 10.1242/dev.062323. - DOI - PubMed
    1. Zarkada G, Heinolainen K, Makinen T, Kubota Y, Alitalo K. VEGFR3 does not sustain retinal angiogenesis without VEGFR2. Proc. Natl Acad. Sci. USA. 2015;112:761–766. doi: 10.1073/pnas.1423278112. - DOI - PMC - PubMed

Publication types

Substances

LinkOut - more resources

Feedback