Endothelial and Smooth Muscle-derived Neuropilin-like Protein Regulates Platelet-derived Growth Factor Signaling in Human Vascular Smooth Muscle Cells by Modulating Receptor Ubiquitination*

Endothelial and smooth muscle cell-derived neuropilin-like protein (ESDN) is up-regulated in the neointima of remodeling arteries and modulates vascular smooth muscle cell (VSMC) growth. Platelet-derived growth factor (PDGF) is the prototypic growth factor for VSMCs and plays a key role in vascular remodeling. Here, we sought to further define ESDN function in primary human VSMCs. ESDN down-regulation by RNA interference significantly enhanced PDGF-induced VSMC DNA synthesis and migration. This was associated with increased ERK1/2, Src, and PDGF receptor (PDGFR)β phosphorylation, without altering total PDGFRβ expression levels. In binding assays, ESDN down-regulation significantly increased 125I-PDGF maximum binding (Bmax) to PDGF receptors on VSMCs without altering the binding constant (Kd), raising the possibility that ESDN regulates PDGFR processing. ESDN down-regulation significantly reduced ligand-induced PDGFRβ ubiquitination. This was associated with a significant reduction in the expression level of c-Cbl, an E3 ubiquitin ligase that ubiquitinylates PDGFRβ. Thus, ESDN modulates PDGF signaling in VSMCs via regulation of PDGFR surface levels. The ESDN effect is mediated, at least in part, through effects on PDGFRβ ubiquitination. ESDN may serve as a target for regulating PDGFRβ signaling in VSMCs.

Vascular injury initiates a cascade of events that ultimately leads to vascular remodeling and often intimal hyperplasia. Vascular smooth muscle cell (VSMC) 2 proliferation and migration are key cellular events in this process. Platelet-derived growth factor (PDGF)-BB is released by platelets, endothelial cells, VSMCs, and inflammatory cells at the sites of vascular injury and is a particularly potent regulator of VSMC proliferation and migration (1). PDGF binding to PDGF receptor (PDGFR)␤ in VSMCs leads to receptor dimerization, autophosphorylation, and activation of downstream signaling pathways, including MAPK. The ligand-bound receptor is internalized through the endocytotic pathway and may either recycle to the membrane or undergo ubiquitination and lysosomal degradation (2). A number of endogenous stimulatory and inhibitory regulators, including the E3 ubiquitin ligase, c-Cbl (3), tightly regulate the mitogenic stimulus by modulating the duration and intensity of the signal.
We have identified endothelial and smooth muscle cell-derived neuropilin-like protein (ESDN, also called CLCP1 or DCBLD2) as a marker and regulator of cell proliferation in vascular remodeling (4). ESDN is a transmembrane protein with a domain structure similar to neuropilins (5,6). ESDN can be induced by PDGF-BB and serum and is highly expressed in the neointima of injured rat (5), mouse (4), and human (4) arteries. ESDN expression parallels cell proliferation in the vessel wall in vivo (4). Furthermore, ESDN is up-regulated in proliferating VSMCs, and ESDN overexpression inhibits VSMC growth (4). Here, we expand the scope of our previous studies to demonstrate that ESDN regulates PDGF-induced VSMC migration and inhibits PDGF signaling in VSMCs. We further establish that this effect is mediated, at least in part, through changes in the surface expression of PDGF receptors. Finally, our study indicates that ESDN mediates PDGFR␤ ubiquitination by regulating c-Cbl gene expression.
Cells-Primary human VSMCs were isolated as described (4) from human aortas, obtained under protocols approved by the Yale University Human Investigation Committee, and grown in M199 with 20% fetal bovine serum. Only VSMCs within the first four passages were utilized. Each experiment was repeated with VSMCs from at least three different donors.
Quantitative Reverse Transcription-PCR (RT-PCR)-Relative transcript expression levels of various genes were assessed by real time RT-PCR. Total RNA extraction, cDNA synthesis, and PCR conditions were performed as described previously (4). PCR primers used in these studies were synthesized by Yale University Keck Facility, and their sequences are listed in supplemental Table 1. For quantitative RT-PCR analysis, each template was tested in triplicate. The abundance of each gene was determined relative to GAPDH.
[ 3 H]Thymidine Incorporation Assay-DNA synthesis was quantitated by measuring [ 3 H]thymidine incorporation. VSMCs transduced with ESDN or green fluorescent protein retrovirus (4) and VSMCs transfected with either ESDN or control siRNA for 48 h seeded on 24-well culture plates at a density of 1 ϫ 10 4 cells/well were serum-starved overnight and treated with PDGF (25 ng/ml) or control buffer in eight replicate wells for 24 h. During the last 6 h, the cells were incubated with 0.5 l/well of [ 3 H]thymidine (0.0185 MBq; GE Healthcare). After rinsing, cells were fixed in ice-cold methanol, and DNA was precipitated by 5% trichloroacetic acid and recovered with NaOH (0.3 N) at room temperature. Aliquots were assayed for [ 3 H]thymidine incorporation by liquid scintillation counting (Beckman Coulter, Inc. Fullerton, CA). Counts were normalized to the control sample and expressed as relative [ 3 H]thymidine incorporation.
Migration Assay-Forty-eight h after siRNA transfection, confluent VSMC monolayer cultures were placed in serum-free medium. Twenty-four h later, cells were scratched with a scraper to create a wound. After wounding, the medium was changed to serum-free medium with or without PDGF-BB (25 ng/ml). Random fields of view were photographed with phase-contrast microscope (Nikon Instruments, Inc., Melville, NY) at baseline and after 24 h. The distance between the leading edge of migrating cells and the wound edge was measured and averaged on 10 random photos for each experimental sample using NIH ImageJ software. The cells that migrated out from wound edge were counted from 10 random photos as well.
Immunoblotting-Forty-eight h following transfection with siRNAs, VSMCs were serum-starved for 24 h, followed by stimulation with PDGF (25 ng/ml) or control buffer for the indicated time. Cells were washed twice with chilled phosphatebuffered saline (PBS) and lysed in ice-cold lysis buffer (20 mM Tris, pH 7.4, 150 mM NaCl, 1% Triton X-100, and complete proteinase inhibitor (Roche Diagnostics) and phosphatase inhibitor (Sigma) cocktails). Cell homogenates were centrifuged (13,000 ϫ g, 15 min, 4°C), supernatants were collected, and protein concentrations were determined (DC protein assay reagent, Bio-Rad). Equal amounts of protein were resolved by SDS-PAGE and transferred electrophoretically to an Immun-Blot polyvinylidene difluoride membrane (Bio-Rad). The membranes were probed with primary antibodies, followed by horseradish peroxidase-conjugated secondary antibodies (Zymed Laboratories Inc., Carlsbad, CA) and developed using a chemiluminescence detection system (PerkinElmer Life Sciences). GAPDH was used as loading control. Films were scanned and quantitated using Kodak 1D 3.5 software.
Immunoprecipitation-Cell lysates (containing 500 g of protein) were incubated with 20 l of protein A-agarose beads for 1 h at 4°C, after which the beads were removed by centrifugation to clear the lysate. Precleared lysates were incubated with the respective antibodies for 1 h at 4°C and subsequently incubated with 15 l of protein A-agarose beads ans/or protein G-agarose beads for 1 h at 4°C, followed by washing with lysis buffer (20 mM Tris, pH 7.4, 150 mM NaCl, 1% Triton X-100, and complete proteinase inhibitor; Roche Applied Science). Immunocomplexes were resuspended in 15 l of SDS-PAGE sample buffer.
PDGF Binding Assay-VSMCs transfected with ESDN or control siRNA for 48 h were plated in complete growth medium in 96-well plates at 3 ϫ 10 3 cells/well. Twenty-four h after seeding, cells were rinsed with PBS and incubated in M199 with 0.5% fetal bovine serum containing increasing concentrations of 125 I-PDGF-BB in quadruplicates at 4°C. After 2 h, the cells were washed with ice-cold buffer and solubilized in RIPA buffer. Radioactivity in an aliquot of solubilized cells was measured by a gamma well counter (Cobra, Auto-Gamma, Packard Instrument Company). Cell numbers in parallel wells were determined after gentle trypsinization and used to normalize the results. Nonspecific binding was determined using 200-fold excess unlabeled PDGF-BB. Specific binding was calculated by subtracting nonspecific binding from total binding. The equilibrium binding constant (K d ) and maximum specific binding (B max ) were determined by non-linear regression (one-site specific binding) using GraphPad Prism software (La Jolla, CA).
Statistical Analysis-Data are expressed as means Ϯ S.E. of the mean. Statistical evaluation of the data were performed using Student's t test (for parametric values), ratio t test (for paired non parametric values), or two-way analysis of variance with Bonferroni post hoc test (for multiple groups). Statistic analysis of B max and K d from non-linear regression curve was done using a two-tailed, unpaired t test. p Ͻ 0.05 was considered to indicate a significant difference.

ESDN Impairs PDGF-induced VSMC Proliferation-VSMC
proliferation and migration, key cellular events in neointimal hyperplasia, are potently induced by PDGF. Therefore, we examined the effect of ESDN on these PDGF-induced cellular events. We have previously reported that ESDN overexpression inhibits VSMC growth, whereas ESDN knockdown shows an opposite effect (4). To establish whether the observed effect of ESDN on VSMC growth is, at least in part, through regulation of cell proliferation, we assessed the effect of ESDN knockdown on VSMC DNA synthesis. Three ESDN siRNA duplexes targeting distinct regions of the ESDN mRNA (769, 1081, 2245; the numbers refer to the oligonucleotide position within ESDN mRNA) were tested for their ability to reduce ESDN expression in cultured VSMCs. An siRNA (2245) targeting the mRNA sequence, GGAAUUGUUGGUACACUUCAUCAAA, was found to exert the greatest down-regulation of ESDN expres-sion as evaluated by quantitative RT-PCR (to ϳ40%, Fig. 1A and also supplemental Fig. 1) without altering the stress responsive gene OAS1 (supplemental Fig. 1) in both control and PDGF-treated cells (supplemental Fig. 2). The inhibitory effect of this siRNA on the VSMC ESDN protein level was confirmed by Western blotting (Fig. 1B). Therefore, we used this siRNA for the following experiments unless indicated otherwise. ESDN knockdown did not affect 3 (Fig. 1C). To confirm the observation from siRNA down-regulation that ESDN inhibits PDGF-induced VSMC proliferation, we transduced primary human VSMCs with a retroviral vector encoding ESDN cDNA (4). Transduction efficiency assessed by flow cytometry using a control green fluorescent protein-expressing vector was estimated to be around ϳ50%. ESDN overexpression modestly attenuated PDGF-induced 3 [H] incorporation (data not shown). Collectively, these results demonstrate the inhibitory role of ESDN on PDGF-induced VSMC proliferation.
ESDN Attenuates PDGF-induced VSMC Migration-Next, we addressed the effect of ESDN down-regulation on VSMC migration using a scratch wound assay. In the absence of growth factor, there was minimal cell migration over a period of 24 h (data not shown). In the presence of PDGF (25 ng/ml), siRNA-mediated ESDN inhibition significantly enhanced PDGF-induced VSMC migration as compared with cells transfected with the control siRNA (migration distance: 650 Ϯ 20 and 920 Ϯ 50 m; and migrating cells per microscopic field: 33 Ϯ 1 and 42 Ϯ 1, respectively, for the control and ESDN siRNAs, p Ͻ 0.0001) (Fig. 2, A-E). Interestingly, PDGF-induced VSMC migration was proportional to the degree of ESDN reduction (supplemental Fig. 3). These results indicate that similar to its effects on VSMC proliferation, ESDN regulates PDGF-induced VSMC migration.
ESDN Regulates PDGF Signaling-PDGF-induced VSMC differentiation, proliferation, and migration is dependent on MAPK activation and the resultant Elk-1 phosphorylation and transcriptional activation of many growth factor-inducible genes (for review, see Ref. 7). To elucidate the mechanism of the ESDN effect on PDGF-induced VSMC proliferation and migration, we examined the effect of ESDN down-regulation on MAPK ERK1/2 phosphorylation by Western blotting. Transfection of VSMCs with ESDN siRNA markedly enhanced PDGF-induced ERK1/2 phosphorylation as compared with the control siRNA ( Fig. 3A and supplemental Fig. 4). In addition to ERK1/2 signaling, Src-mediated signaling cascade also plays a role in VSMC migration and mitogenic responses (8 -10). Therefore, we assessed the effect of ESDN down-regulation on Src phosphorylation and demonstrated that ESDN inhibition by RNA interference enhances PDGF-induced Src phosphorylation (Fig. 3B). Together, these data indicate that ESDN modulates activation of several PDGF-related signaling pathways.
PDGFR␤ is the major receptor for PDGF-BB in VSMCs, and its phosphorylation mediates PDGF-induced ERK1/2 and Src phosphorylation (11). As such, PDGFR␤ stands out as a candidate for the observed effect of ESDN in regulating PDGF signaling in VSMCs. To explore a potential effect of ESDN on PDGFR␤, we assessed the effect of ESDN down-regulation on PDGFR␤ expression and phosphorylation in VSMCs. ESDN down-regulation did not affect PDGFR␤ mRNA level (data not shown), and its effect on its protein expression was minimal (Fig. 3C). However, it markedly enhanced PDGF-induced PDGFR␤ phosphorylation in VSMCs (Fig. 3C).

ESDN Alters the Number of PDGF Binding Sites on VSMCs-
The observed effect of ESDN down-regulation on PDGFR␤ phosphorylation may be explained by a potential effect on VSMC surface receptor numbers or affinity for the ligand. Therefore, we evaluated the effect of ESDN on PDGF binding to its receptors on VSMCs using 125 I-labeled PDGF-BB. VSMCs transfected with ESDN siRNA displayed significantly higher numbers of PDGF-BB binding sites as compared with control  cells (B max : 773 Ϯ 9 and 492 Ϯ 5 fmol/10 6 cells, respectively, p Ͻ 0.0001), without altering PDGF affinity for its receptors (K d 849 Ϯ 61 and 656 Ϯ 47 pM, respectively, p Ͼ 0.05) (Fig. 4). These results indicate that ESDN affects the number of PDGF receptors on the VSMC cell surface.
ESDN Affects Ligand-induced Ubiquitination of PDGFR␤-Polyubiquitination of PDGFR␤ is a necessary step for receptor degradation in response to PDGF stimulation and plays a negative regulatory role on PDGFR␤ signaling (12,13). This raises the possibility that PDGFR␤ ubiquitination plays a role in the observed ESDN effect on PDGFR␤ signaling. Therefore, we compared the levels of ubiquitinated PDGFR␤ after PDGF-BB stimulation in VSMCs transfected with ESDN or control siRNA. Ubiquitinated PDGFR␤ was detected by immunoblotting of immunoprecipitated PDGFR␤ with an anti-ubiquitin antibody. As expected, PDGF-stimulation led to PDGFR␤ ubiquitination and concomitant reduction in PDGFR␤ levels (Fig. 5A). Furthermore, the ratio of ubiquitinated to total PDGFR␤ was significantly lower in ESDN down-regulated, as compared with control cells (Fig. 5B). As such, PDGFR␤ ubiquitination, a negative regulator of PDGFR␤ signaling, is reduced by ESDN down-regulation, resulting in increased PDGFR␤ signaling.
ESDN Regulates c-Cbl Expression-c-Cbl is a cytoplasmic ubiquitin ligase that has been implicated as a negative regulator of PDGFR␤ signaling (3). To address the mechanism of the ESDN effect on PDGFR␤ ubiquitination, we investigated whether ESDN down-regulation affects VSMC c-Cbl protein level by Western blotting. In VSMCs transfected with ESDN siRNA c-Cbl was significantly reduced in conjunction with the reduction in ESDN protein level (Fig. 6A).
Finally, we sought to determine whether ESDN modulates c-Cbl protein level by regulating its gene expression. Transfecting primary human VSMCs with ESDN siRNA significantly reduced c-Cbl mRNA level in parallel with the expected reduction in ESDN mRNA level (to 61% Ϯ 0.02% of the level observed in cells transfected with control siRNA) (Fig. 6B). ESDN siRNA transfection did not affect the expression of a stress responsive gene, OAS1, a cytoskeletal protein, ␤-actin, and a growth factor receptor, vascular endothelial growth factor receptor-1, in  VSMCs were transfected with control or ESDN siRNAs for 2 days, followed by 24 h serum starvation, and then incubated with or without PDGF (25 ng/ml) for 0 -60 min. Cell lysates (containing 0.5 mg of protein) were immunoprecipitated with an anti-PDGFR␤ antibody, followed by immunoblotting with an anti-ubiquitin (Ub) antibody (upper panel) or an anti-PDGFR␤ antibody (lower panel). The quantification represents means Ϯ S.E. of three independent experiments. Level of ubiquitin (ubiquitylated PDGFR␤) was normalized to total PDGFR␤ immunoprecipitated. *, p Ͻ 0.05. VSMCs (supplemental Fig. 1), indicating that the observed inhibitory effect on c-Cbl mRNA level is very specific.

DISCUSSION
ESDN was initially cloned from human coronary arterial (5) and highly metastatic lung cancer (6) cells. We previously demonstrated that in animal models of immune or mechanical injury-induced vascular remodeling, ESDN expression was temporally and spatially associated with cell proliferation, thus providing a potential regulatory mechanism for the response to injury (4). In vitro, ESDN overexpression suppressed and its down-regulation enhanced VSMC growth. Here, we extend the scope of those findings to demonstrate that ESDN down-regulation potentiates PDGF-induced VSMC DNA synthesis and migration, indicating that ESDN functions as a suppressor of PDGF signaling in VSMCs. Despite the relatively modest changes in ESDN mRNA and protein levels achieved by RNA interference in primary human VSMCs, the modulatory effects of ESDN on PDGF-induced VSMC migration and growth were consistently detectable. It is expected that more pronounced changes in the ESDN level (expected e.g. in the ESDN knockout mouse) would lead to more striking effects on VSMC biology. These observations led us to further explore a potential regulatory effect of ESDN on PDGF signaling in VSMCs.
PDGF was identified more than 30 years ago as a serum growth factor for VSMCs and fibroblasts. The PDGF family consists of homo-or heterodimers of four distinct polypeptide chains, namely PDGF-A, -B, -C, or -D, which interact with two distinct receptors (PDGFR␣ and ␤). PDGF-BB, produced by endothelial cells, megakaryocytes, and neurons, is the prototypic growth factors for VSMCs. It is unique in its ability to induce profound suppression of VSMC differentiation marker genes through PDGFR␤ signaling (1). The role of PDGF in triggering VSMC proliferation and migration is supported by in vitro as well as in vivo evidence. As such, PDGF-BB antibodies can inhibit VSMC proliferation (14) and migration (15) after carotid injury in the rat. Existing data suggest that these effects are mediated by PDGFR␤, but not PDGFR␣, signaling as the latter cannot be activated by PDGF-BB (7). PDGF binding to PDGFR␤ leads to receptor dimerization, autophosphorylation, and recruitment of Src homology 2 (SH2) domain-containing signaling molecules (16). This in turn leads to MAPK (ERK1/2, p38 MAPK, and stress-activated protein kinase/c-Jun N-terminal kinase) activation (17). The resultant ERK1/2-induced Elk-1 phosphorylation is a key event in PDGF-induced VSMC differentiation, proliferation, and migration (1,18). As expected from its observed effects on VSMC proliferation and migration, ESDN down-regulation potentiated PDGF-induced ERK1/2, Src, and PDGFR␤ phosphorylation, demonstrating that ESDN regulation of VSMC biology is, at least in part, linked to its effect on PDGFR␤ signaling.
Upon binding to PDGF, PDGFR␤ is phosphorylated, ubiquitinylated, and internalized and undergoes lysosomal degradation or recycling to the cell membrane (19,20). The E3 ubiquitin ligase, c-Cbl, is activated after PDGF stimulation and associates with phosphorylated PDGFR␤ (21,22). This in turn leads to PDGFR␤ ubiquitination, internalization via the clathrin-mediated pathway, and lysosomal degradation negatively regulating PDGFR␤ signaling and cell proliferation (21). In addition to ubiquitin ligase, SHP-2 tyrosine phosphatase (23) and T-cell protein tyrosine phosphatase (24) bind to, dephosphorylate, and negatively regulate PDGFR␤ signaling. ESDN down-regulation increased the number of PDGF binding sites without significant changes in total receptor level, raising the possibility that it may alter receptor ubiquitination. It is known that Cbl-mediated enhanced lysosomal degradation limits the biological half-life of only the small but important active pool of receptor tyrosine kinases. This is thought to be an attenuation mechanism, regulating the global impact of receptor tyrosine kinase activation within a cell. It is not surprising that ubiqitination of the active PDGFR␤ pool does not significantly affect total receptor levels. Indeed, we observed a clear decrease in PDGF-induced PDGFR␤ ubiquitination in ESDN down-regulated cells. This was associated with a reduction in c-Cbl protein and mRNA levels, providing a potential mechanism for the significantly reduced PDGF-induced PDGFR␤ ubiquitination in ESDN down-regulated VSMCs. Of note, c-Cbl overexpression enhances PDGFR␤ ubiquitination and subsequent degradation in NIH3T3 fibroblasts (21).
The data presented here support our speculation that ESDN up-regulation participates in controlling VSMC proliferation and migration in injury-induced vascular remodeling, potentially by reducing the response to growth factors in a negative feedback loop (4). Other regulators of VSMC proliferation and migration and neointima formation have been described. Similar to ESDN, fibulin-5 is minimally expressed in normal arteries, is up-regulated in conjunction with neointima formation in remodeling arteries, and negatively regulates VSMC proliferation and migration (25). Integrins, such as ␣v␤3 (26) and ␣7␤1 (27,28), are induced and modulate VSMC proliferation and migration in vascular remodeling at multiple levels, including potentiation of ligand-dependent and -independent growth factor receptor signaling (29) through matrix engagement. Although many studies have identified c-Cbl and other Cblfamily proteins as key regulators of receptor tyrosine kinase and other signaling pathways (for review, see (30)), to our knowledge, this is the first report of regulating receptor signaling through regulation of c-Cbl expression. Although diverse functions of c-Cbl have been widely studied, little is known about its gene regulation and further studies are needed to elucidate how ESDN regulates c-Cbl gene expression.
In conclusion, our data establish a novel regulatory role for ESDN in PDGF-induced signaling in VSMCs. The ESDN effect is, at least in part, mediated by c-Cbl which regulates PDGFR degradation and recycling. All of the experiments reported here used early passage primary human VSMCs. Similar ESDN functions have been observed in two non-primary cell lines: ESDN overexpression in 293T cells suppressed bromodeoxyuridine uptake (6), and its overexpression in gastric cancer cells inhibited colony formation in both anchorage-dependent and -independent cultures as well as cell invasion through the collagen matrix (31). ESDN expression has also been linked to acquisition of metastatic capability in lung cancer in vivo (6). None of these reports established the mechanisms of ESDN function. The observed effect of ESDN in regulating PDGFR␤ signaling raises the possibility that ESDN modulates signaling of other receptor tyrosine kinases through similar mechanisms. Regulation of VSMC biology by ESDN raises the possibility that ESDN may serve as a therapeutic target for pathological states, such as post-angioplasty restenosis and graft arteriosclerosis where VSMC proliferation and migration play key roles in the pathogenesis. Demonstrating a similar inhibitory effect of ESDN on PDGFR signaling in other cell types and other signaling pathways would expand the potential clinical applications of ESDN to other proliferative disorders.