Notch1 activates angiogenic regulator Netrin4 in endothelial cells

Abstract Netrin4 (NTN4) is a chemotropic factor that regulates angiogenesis. We found that endothelial expression of the activated, intracellular domain of Notch1 (NICD1), significantly up‐regulated NTN4 mRNA as well as intracellular NTN4 protein in both transgenic mice and cultured human umbilical vein endothelial cells (HUVECs). Notch1 activation also increased NTN4 secretion from HUVECs. We subsequently demonstrated that NICD1 bound to CSL (CBF1, Suppressor of Hairless, Lag‐1), a core component of Notch transcription complex, at the −53 element of the human NTN4 gene promoter. Loss of the −53 element compromised NICD1‐induced NTN4 expression. Our results suggest a conserved role for Notch signalling in transcriptional regulation of endothelial NTN4.


| INTRODUC TI ON
Netrin4 (NTN4) is a secreted protein that is structurally related to laminins and plays a role in chemotaxis. The Netrin family of proteins contributes either attractive or repulsive guidance cues during neurite outgrowth and neuronal migration. 1 In addition, NTN4 regulates angiogenesis. Ntn4 expression is localized to areas of embryonic angiogenesis in zebrafish, 2 and is functional postnatally during ischaemia-induced angiogenesis in the rat retina. 3 Both in vivo and in vitro experiments have demonstrated that stimulation with soluble NTN4 protein markedly promotes angiogenesis by inducing proliferation, migration and tube formation of vascular or lymphatic endothelial cells. 2,4,5 This is mediated at least in part through activation of the ERK, AKT, JNK and FAK signalling pathways. 2,5 In mouse models of ischaemia, treatment with Ntn4 protein increases vascular density in ischaemic areas and is beneficial to post-ischaemic reperfusion. 4,6 Despite these pro-angiogenic findings, some studies suggest a contrary role in angiogenesis. Netrin4 suppresses VEGF-stimulated endothelial cell migration and tube formation by binding to the transmembrane receptor Neogenin (Neo1). 7 Elevated NTN4 also decreases pancreatic and colorectal cancer growth by inhibiting tumour angiogenesis. 8,9 In the corneal mouse model, Ntn4 inhibits suture-mediated neovascularization. 10 Whether NTN4 acts as a pro-or anti-angiogenic factor may be context dependent. Here, we report that NTN4 is a target of Notch signalling, a conserved pathway which is important in both embryonic and disease-related angiogenesis.
Notch signalling is highly conserved between species and is critical for differentiation, proliferation and fate determi- Our study demonstrates a role for Notch signalling in transcriptional regulation of NTN4.

| Animals
The generation of the endothelial-specific NICD1 Tg mice has been previously described. 14,15 All animal experiments were approved by the Ethics Committee of Shandong Provincial Qianfoshan Hospital.

| Western blot, Coomassie blue staining and antibodies
Protein samples were prepared in RIPA-0.1% SDS lysis buffer

| Chromatin immunoprecipitation assay
A Chromatin immunoprecipitation (ChIP) assay kit was used to precipitate the target chromatin fragments (EMD Millipore). CSL (Cell Signaling Technology, Danvers, MA) or Notch1 (Abcam, Cambridge, MA) antibody was used for immunoprecipitation with non-binding isotype control IgG for comparison (Cell Signaling Technology). The immunoprecipitated fragments were amplified with a PCR master mix (Vazyme, Nanjing, China) and analysed on 1% agrose gel.

| Dual-luciferase reporter assay
Human NTN4 promoter fragments were amplified from human gDNA through polymerase chain reaction (Fermentas, Waltham, MA). The fragments were subcloned into the pGL3-basic plasmid through Xho I/Hind III digestion. Primer sequences are disclosed in supporting materials (Table S1) (Table S1).

| Statistical analysis
All quantitative results were calculated from at least three independent experiments. Values represented means ± standard error.
Student's t test was utilized for statistical analysis.

| Activation of Notch signalling increases NTN4 expression
Both Notch signalling and Netrins regulate cell guidance and angiogenesis. We speculated that there may be a relationship between Notch signalling and the Netrin family protein, NTN4. To test this hypothesis, we first examined the expression of Ntn4 in the yolk sac of NICD1 Tg mice (NICD1 Tg), in which NICD1 expression was limited to the zone where its expression was dictated by the endothelial-specific Tie2 promoter ( Figure 1A). Quantitative RT-PCR (qRT-PCR) and Western blot results demonstrated that mRNA ( Figure 1B) and protein levels ( Figure 1C) of Ntn4 in the yolk sac were both markedly elevated in NICD1 Tg mice. To confirm this regulatory effect of Notch on Ntn4 expression, we activated Notch signalling by infecting HUVECs with lentiviral particles comprising the coding sequence of the Notch1 intracellular domain (NICD1) ( Figure S1). We used qRT-PCR to F I G U R E 1 Notch1 activation increase NTN4 gene expression. A, Strategy to generate endothelial-specific NICD1 transgenic mice. In ZEG-NICD1 Tg mice, the transcription of NICD1 is initially repressed by a LoxP flanked β-geo/STOP signal. To activate expression, the ZEG-NICD1 Tg mice have been crossed to Tie2-Cre Tg mice, in which Cre transcription is controlled by an endothelial-specific Tie2 promoter. In double Tg offspring, the β-geo/STOP signal is removed, leading to endothelial Notch1 intracellular domain (NICD1) overexpression. (B,C) The yolk sac was isolated from NICD1-expressing Tg mice (NICD1 Tg) and control littermates (labelled as WT). Ntn4 mRNA (B) and protein levels (C) were detected by quantitative RT-PCR and Western blot analysis respectively (β-actin as loading control). (D-F) Human umbilical vein endothelial cells (HUVECs) were transduced with NICD1-expressing lentiviral particles or control virus and puromycin selected to ensure expression. D, mRNA level of HES1 and NTN4 detected by qRT-PCR (data normalized to β-actin). E, Level of cellular NICD1, Netrin4 (NTN4) and HES1 protein in transduced cells. F, Secreted NTN4 protein in culture media detected by Western blot analysis. Coomassie blue staining of the gel was used as loading control. *P < 0.05, **P < 0.01 demonstrate that HES1, a canonical downstream target of Notch, was markedly up-regulated by lentiviral NICD1 (Figure 1D). Concurrently, NTN4 mRNA expression was significantly elevated, suggesting it is also a downstream target of the Notch pathway ( Figure 1D). We confirmed through Western blotting that protein expression was elevated for HES1 and NTN4 in NICD1-overexpressing endothe-

lial cells. Quantitation of protein elevation in lentivirally transduced
HUVECs showed an average of 2.86-and 2.52-fold change for HES1 and NTN4, respectively, compared to control cell lysates ( Figure 1E; Figure S2). Because NTN4 is a secreted protein, we asked whether NICD1-overexpression increases the abundance of secreted NTN4 in the extracellular environment. To confirm this, HUVEC culture media were collected and subjected to Western blot analysis. Netrin4 in NICD1-overexpressing HUVEC culture was significantly elevated ( Figure 1F; Figure S3), indicating that activated Notch1 enhances secretion of NTN4. To examine the regulatory effect of Notch signalling on NTN4 expression under physiological condition, HUVECs were cultured in Delta-like ligand 4 (DLL4)-coated dishes. In these cells, Notch signalling was activated, marked by a significant increase of HES1 expression ( Figure S4A,B) and NICD1 production ( Figure S4B).
Moreover, dibenzazepine (DBZ), a specific inhibitor of γ-secretase and Notch signalling, dramatically reduced the expression of HES1 ( Figure S5A,B) and the production of NICD1 ( Figure S5B) in HUVECs.
Consequently, transcription of the NTN4 gene ( Figure S5A), cellular ( Figure S5B) and secreted NTN4 protein level ( Figure S5C) was markedly decreased in DBZ-treated HUVECs. These results imply that NTN4 lies downstream of Notch signalling. We next asked whether NTN4 involves in Notch signalling-regulated angiogenic activities.
The expression of NTN4 in NICD1-overexpressing HUVECs was down-regulated by NTN4 siRNAs ( Figure S6A). As shown in Figure   S6B and C, NTN4 knockdown moderately rescues Notch1 activation-

| The −53 TGGGAA promoter element is essential for Notch-induced NTN4 transcription
We examined the NTN4 promoter for sequence that might mediate transactivation downstream of Notch. We found that the F I G U R E 2 A −53 bp TGGGAA element in the NTN4 promoter is essential for Notch transactivation. A, Human NTN4 promoter fragments inserted into reporter plasmids for this study. (B,C) Reporter plasmids were transfected into human umbilical vein endothelial cells (HUVECs). Basal activity of corresponding fragments was monitored by dual-luciferase reporter (DLR) assay. D, The Fragment 3 wildtype and mutant plasmids were cotransfected into HUVECs expressing Notch1 intracellular domain (NICD1). The transcriptional activity was monitored by DLR assay. E, The binding of CSL to the −53 bp element of the NTN4 promoter was detected by ChIP assay. CSL antibody was used to pulldown CSL. IgG was used as negative control for ChIP. F, The binding of NICD1 protein to the −53 bp element of the NTN4 promoter was monitored by ChIP assay. The quantitative analysis of band intensity was normalized to the input. *P < 0.05; **P < 0.01; ns, no significance; M, marker. FL, full length CSL-preferred binding element, TGG/TGAA, occurred at three unique sites in the NTN4 promoter ( Figure 2A) and sought to determine which, if any, functioned to regulate basal activation of the reporter. We constructed a series of DLR assay plasmids that contained at least one of these predicted binding sites (Figure 2A; Figure S7). The DLR assay in HUVECs showed that ablation of the sequence from −233 bp to +388 bp, which contained a predicted CSL site at −53 bp, had a drastic effect on the transcription activation of the NTN4 promoter plasmid ( Figure 2B). Ablation of other fragments failed to alter the reporter activity. To validate that it was the −53 bp CSL-binding site that played an important role in activating the NTN4 reporter plasmid, we created a mutant that contained five nucleotide substitutions at the −53 binding site (TGGGAA to GTGTCC) (Figure 2A). Deletion of the CSL-binding site lead to a more than 60% decrease in basal transcriptional activity ( Figure 2C).
Next we evaluated the role of Notch signalling in transcriptional activation. As shown in Figure 2D, co-expression of NICD1 up-regulated transactivation through the CSL-binding element at −53 bp (Fragment 3, but not Fragment 3 with TGGGAA to GTGTCC) ( Figure 2D). To confirm that NTN4 is a direct Notch target, we studied the binding activity of CSL and NICD1 protein on the predicted −53 bp site by ChIP assay. Either a CSL ( Figure 2E) or Notch1 ( Figure 2F) antibody was employed for immunoprecipitation. Subsequent PCR amplification of immunoprecipitated chromatin was accomplished with primers that flanked the −53 binding site. Both CSL and Notch1 antibodies enabled identification of a 90 bp band, the size predicted ( Figure 2E,F). Moreover, NICD1 overexpression could markedly increase the band intensity, implying that exogenous NICD1 protein was recruited to the −53 bp binding site ( Figure 2F). Collectively, these results suggest that NTN4 is a direct target of a Notch1 transactivating complex.
Direct binding occurs at the −53 bp, TGGGAA site of NTN4 promoter. This finding provides the first evidence connecting these angiogenic regulatory signals. Notch signalling may work, at least in part, to pattern the vasculature through NTN4.

ACK N OWLED G EM ENTS
This study is supported by grants from the National Natural Science

CO N FLI C T O F I NTE R E S T
All the authors declared no conflict of interest.

AUTH O R S' CO NTR I B UTI O N S
JL designed the research study; CL, JL contributed essential reagents or tools; QL WS, JL, YW, TA performed the experiments and analysed the data; QL CL, TA, JL wrote the paper. All the authors read and approved the manuscript.