The axonal attractant Netrin-1 is an angiogenic factor

Abstract

Blood vessels and nerves often follow parallel trajectories, suggesting that distal targets use common cues that induce vascularization and innervation. Netrins are secreted by the floor plate and attract commissural axons toward the midline of the neural tube. Here, we show that Netrin-1 is also a potent vascular mitogen. Netrin-1 stimulates proliferation, induces migration, and promotes adhesion of endothelial cells and vascular smooth muscle cells with a specific activity comparable to vascular endothelial growth factor and platelet-derived growth factor. Our evidence indicates that the netrin receptor, Neogenin, mediates netrin signaling in vascular smooth muscle cells, but suggests that an unidentified receptor mediates the proangiogenic effects of Netrin-1 on endothelial cells. Netrin-1 also stimulates angiogenesis in vivo and augments the response to vascular endothelial growth factor. Thus, we demonstrate that Netrin-1 is a secreted neural guidance cue with the unique ability to attract both blood vessels and axons, and suggest that other cues may also function as vascular endothelial growth factors.

Fig. 1.

Netrin-1 stimulates proliferation of vascular endothelial cells and VSMCs. The mitogenic potential of Netrin-1, compared to known vascular growth factors, was assessed by measuring the increase in cell number of cultured endothelial cells (A and B) and VSMCs (C and D) after application of these factors to the culture medium. The data are presented as fold increase in cell number relative to control cells that were treated with BSA in serum-free media. Each panel represents the results of at least three independent experiments, each performed in triplicate. (A) After 48 h of treatment, Netrin-1 (50 ng/ml, 0.7 nM) and VEGF (10 ng/ml, 0.6 nM) stimulated proliferation to a similar extent in both primary HMVEC and HAEC. (B) HMVEC exhibit a dose-dependent response to Netrin-1. Treatment of HMVEC with increasing concentrations of Netrin-1 reveals a peak mitogenic activity comparable to that of VEGF at an intermediate concentration of 50 ng/ml. (C and D) Netrin-1 is a mitogen for VSMC exhibiting a potency similar to, or exceeding, that of PDGF. (C) After 48 h of treatment with either Netrin-1 (50 ng/ml, 0.7 nM) or PDGF (30 ng/ml, 2.5 nM), a 2-fold increase in the number of primary rat or human VSMC was observed compared to BSA-treated (control) cells. (D) VSMC exhibits a dose-dependent response to Netrin-1. Cellular proliferation is enhanced with increasing concentrations of Netrin-1. Asterisks indicate treatments that yield a statistically significant increase (P < 0.05, Student's t test) in proliferation. The results are expressed as mean ± SD.

Fig. 2.

Netrin-1 induces migration of endothelial cells and VSMCs. A modified Boyden chamber assay was used to assess the ability of Netrin-1 to stimulate endothelial cells (A-C) and VSMCs (D-F) to migrate across a porous filter, compared to the known chemoattractants VEGF and PDGF, respectively. The data are presented as the fold increase in migration observed with the test factor relative to the control cultures treated with BSA. The figures represent the results of at least three independent experiments, each performed in triplicate. (A-C) Netrin-1 is chemotactic for endothelial cells with a comparable specific activity to VEGF. (A) Netrin-1 at 50 ng/ml (0.7 nM) and VEGF at 10 ng/ml (0.6 nM) each induced similar degrees of HAEC and HMVEC migration in a Boyden chamber. (B) HMVEC cell migration is enhanced with increasing concentrations of Netrin-1. (C) Netrin-1 induces a directional migration, as suggested by the loss of migration when equamolar concentrations are present in both the upper and lower chambers. (D-F) Netrin-1 is chemotactic for VSMC with similar, or greater, specific activity than PDGF. (D) Netrin-1 at 50 ng/ml (0.7 nM) and PDGF at 30 ng/ml (2.5 nM) each induced migration to a comparable extent in either rat or human VSMC. (E) Dose-response curves for Netrin-1-induced VSMC migration identified the peak activity around 50 ng/ml. (F) Netrin-1 induces a directional migration, because an equamolar concentrations of Netrin-1 in each chamber reduces VSMC migration. The asterisk indicates treatments that yield a statistically significant increase (P < 0.05, Student's t test) in migration. The results are expressed as the mean ± SD.

Fig. 3.

Netrin-1 promotes adhesion of VSMCs, but not endothelial cells. Quantitative analysis of the adhesion of VSMCs and endothelial cells (EC) to Netrin-1 is shown. The wells of cell culture dishes were coated with test matrix elements or Netrin-1. EC and VSMC were seeded on the treated wells for 30 min, followed by PBS wash. Adhesion was quantified by comparing the number of cells that adhered to a test coating relative to BSA-coated wells. The images represent the results of at least three independent experiments each performed in triplicate. (A) Primary rat and human VSMC adhered to Netrin-1 and fibronectin (FN), whereas HAEC and HMVEC did not adhere to the Netrin-1, but bound tightly to FN. (B) Representative fields of hematoxylin and eosin-stained wells of human EC and VSMC from adhesion assays. (C) Adhesion of VSMC to Netrin-1 is comparable to FN, Laminin-1, Collagen I, and Collagen IV. The asterisk indicates treatments that yield a statistically significant increase (P < 0.05, Student's t test) in adhesion. The results are expressed as the mean ± SD.

Fig. 4.

Neogenin mediates netrin signaling in VSMCs. The expression and functional contribution of the previously reported Netrin-1 receptors, DCC and Neogenin, to Netrin-1-induced migration and adhesion were assessed in endothelial cells and VSMC. (A) Neogenin was detected in VSMC, and to a much lesser degree in HMVEC and HUVEC, via RT-PCR analysis. DCC receptor expression was not detected in either endothelial cells or VSMC. Human brain RNA served as a positive control for detection of both receptors. (B) Antibodies that recognize DCC, Neogenin, the endothelial cell marker Robo4, and smooth muscle cell actin (SMA) were used in Western blot analysis of primary endothelial cell and VSMC lines. Cell lines known to express the designated gene products were used as controls. Neogenin protein, but not DCC, was detected in VSMCs. Neogenin and DCC were not detected in either endothelial cell type. Antibodies that recognize SMA and the endothelial-specific receptor Robo4 confirmed the identity of VSMC and endothelial cells, respectively. (C) Neogenin mediates VSMC migration to Netrin-1. Neogenin-blocking antibodies inhibited the migration of VSMC to Netrin-1. Conversely, this antibody did not block endothelial cell migration or PDGF-mediated VSMC migration. (D) Neogenin mediates VSMC adhesion to Netrin-1. Neogenin blocking antibodies inhibited netrin-mediated VSMC adhesion and had no effect on fibronectin-mediated VSMC adhesion. The asterisk indicates treatments that give a statistically significant decrease (P < 0.05, Student's t test) in migration or adhesion. The results are expressed as the mean ± SD.

Fig. 5.

Netrin-1 induces angiogenesis in vivo CAM assay. Netrin-1 induces vascular sprouting in a chick CAM assay. Sponges soaked with 1 μg of Netrin-1, VEGF, FGF, and BSA were placed on chick CAMs. The number of vessels sprouting into the sponges after 72 h was counted and quantified as the fold increase over BSA-treated sponges (control). For each test factor, a total of 24 CAM assays were performed. Quantitative assessment and visual representations of each treatment are shown. The asterisks indicate treatments that yield a statistically significant increase (P < 0.05, Student's t test) in angiogenesis.

Fig. 6.

Netrin-1 is synergistic with VEGF in both the in vivo corneal micropocket assay and the in vitro migration assay. (A and B) Netrin-1 induced angiogenesis in a murine corneal micropocket assay. Hydron pellets containing equimolar amounts of either Netrin-1 (200 ng) or VEGF (50 ng) stimulated comparable levels of blood vessel growth. When combined, the two factors generated an accentuated response that was greater than the sum of the responses to each factor individually. In a single experiment, each test factor was placed on six corneas. Each experiment was repeated a minimum of three times. Quantitative assessment and visual representations are shown. (C) Treatment of endothelial cells with Netrin-1 and VEGF together synergistically enhances migration. Either 10 ng/ml Netrin-1 or 2.5 ng/ml VEGF alone showed a <2-fold increase over BSA control. However, when 10 ng/ml Netrin-1 and 2.5 ng/ml VEGF were combined, the response was greater than the sum of each factor treated separately. The asterisks indicate treatments that yield a statistically significant increase (P < 0.05, Student's t test) in angiogenesis. The double asterisks indicate treatment that yields a statistically significant increase (P < 0.001) over individual Netrin-1- or VEGF-treated samples.