Skip main navigation

Endothelial TIE1 Restricts Angiogenic Sprouting to Coordinate Vein Assembly in Synergy With Its Homologue TIE2

Originally publishedhttps://doi.org/10.1161/ATVBAHA.122.318860Arteriosclerosis, Thrombosis, and Vascular Biology. 2023;43:e323–e338

Abstract

BACKGROUND:

Vascular growth followed by vessel specification is crucial for the establishment of a hierarchical blood vascular network. We have shown that TIE2 is required for vein development while little is known about its homologue TIE1 (tyrosine kinase with immunoglobulin-like and EGF [epithelial growth factor]-like domains 1) in this process.

METHODS:

We analyzed functions of TIE1 as well as its synergy with TIE2 in the regulation of vein formation by employing genetic mouse models targeting Tie1, Tek, and Nr2f2, together with in vitro cultured endothelial cells to decipher the underlying mechanism.

RESULTS:

Cardinal vein growth appeared normal in TIE1-deficient mice, whereas TIE2 deficiency altered the identity of cardinal vein endothelial cells with the aberrant expression of DLL4 (delta-like canonical Notch ligand 4). Interestingly, the growth of cutaneous veins, which was initiated at approximately embryonic day 13.5, was retarded in mice lack of TIE1. TIE1 deficiency disrupted the venous integrity, displaying increased sprouting angiogenesis and vascular bleeding. Abnormal venous sprouts with defective arteriovenous alignment were also observed in the mesenteries of Tie1-deleted mice. Mechanistically, TIE1 deficiency resulted in the decreased expression of venous regulators including TIE2 and COUP-TFII (chicken ovalbumin upstream promoter transcription factor, encoded by Nr2f2, nuclear receptor subfamily 2 group F member 2) while angiogenic regulators were upregulated. The alteration of TIE2 level by TIE1 insufficiency was further confirmed by the siRNA-mediated knockdown of Tie1 in cultured endothelial cells. Interestingly, TIE2 insufficiency also reduced the expression of TIE1. Combining the endothelial deletion of Tie1 with 1 null allele of Tek resulted in a progressive increase of vein-associated angiogenesis leading to the formation of vascular tufts in retinas, whereas the loss of Tie1 alone produced a relatively mild venous defect. Furthermore, the induced deletion of endothelial Nr2f2 decreased both TIE1 and TIE2.

CONCLUSIONS:

Findings from this study imply that TIE1 and TIE2, together with COUP-TFII, act in a synergistic manner to restrict sprouting angiogenesis during the development of venous system.

Highlights

  • TIE1 (tyrosine kinase with immunoglobulin-like and EGF [epithelial growth factor]-like domains 1) was dispensable for cardinal vein development while TIE2 deficiency altered the identity of cardinal vein endothelial cells with the aberrant expression of DLL4 (delta-like canonical Notch ligand 4).

  • Loss of TIE1 disrupted the vein formation in skin and mesentery during embryogenesis as well as in the retinas after the endothelial Tie1 deletion.

  • TIE1 deficiency led to the decreased expression of venous regulators including TIE2 and COUP-TFII (chicken ovalbumin upstream promoter transcription factor, encoded by Nr2f2, nuclear receptor subfamily 2 group F member 2) while angiogenic regulators were upregulated, suggesting that TIE1 participates in the regulation of vein assembly by the restriction of angiogenesis.

  • Combining the endothelial deletion of Tie1 with 1 null allele of Tek resulted in a more severe retinal vein defects than that of Tie1 deletion alone, suggesting that TIE1 and TIE2 act in a synergistic manner during vein development.

  • COUP-TFII was reduced after the induced deletion of TIE1 or TIE2 while the endothelial deletion of COUP-TFII also led to the decrease of both TIE1 and TIE2, suggesting a positive feedback loop among TIE receptors and COUP-TFII for the coordinated regulation of venogenesis.

Specification of vascular endothelial cell (EC) identity, together with the parallel formation of arteries and veins, are crucial events for the construction of vascular network in development. While we have gained a better understanding about the molecular players for arteriogenesis, mechanisms underlying vein development are less well characterized. COUP-TFII (chicken ovalbumin upstream promoter transcription factor, encoded by Nr2f2, nuclear receptor subfamily 2 group F member 2), a transcription factor expressed in venous ECs, has been shown to regulate venous identity via the inhibition of NOTCH-mediated signals.1 AKT (RAC-alpha serine/threonine-protein kinase, also known as protein kinase B, PKB) activation was shown to inhibit RAF1 (RAF proto-oncogene serine/threonine-protein kinase)-ERK1/2 (extracellular signal-regulated kinase 1/2) signaling in ECs to favor venous specification.2 Recently, the upstream regulators have also been identified. The ANGPT (angiopoietin) receptor TIE2 was demonstrated to play a crucial role in the venous specification and maintenance via its downstream PI3K (phosphoinositide 3-kinase)/AKT-mediated stabilization of COUP-TFII.3 Mutations with angiopoietin receptor TIE2 led to venous malformation.4 Consistently, the specific deletion of cardiomyocyte Angpt1 was shown to disrupt coronary vein formation in the developing heart.5 Combined insufficiency of TIE2 ligands ANGPT1 and ANGPT2 in mice disrupted the formation of sclera venous sinus (Schlemm’s canal), a type of vessel with also lymphatic characteristics.6,7

TIE1 (tyrosine kinase with immunoglobulin-like and EGF [epithelial growth factor]-like domains 1) is a member of the receptor tyrosine kinase family with a high degree of homology with TIE2. Genetic studies have revealed that angiopoietins and TIE receptors are differentially required during blood vascular and lymphatic development. Attenuation of ANGPT1 and TIE2-mediated signals disrupted vein formation in addition to other vascular phenotypes.3,5,8–10Tie1 and Angpt2 null mice displayed abnormal lymphatic formation and also blood vascular defects.11–18 In spite of the important roles of TIE1 in vascular development, the underlying mechanism of its functions remains incompletely understood. On the one hand, previous studies showed that TIE1 could heterodimerize with TIE2 and exert an inhibitory role in TIE2 signaling.19–21 TIE1 expression was also suggested to negatively regulate TIE2 presentation at the cell surface in sprouting endothelial tip cells.17 On the other hand, ANGPT1 could induce TIE1 phosphorylation when coexpressed with TIE2 in cultured cells.22 TIE1 and TIE2 heteromeric complexes in endothelial cell-cell junctions were shown to be required for TIE2 activation.16,23,24 Interestingly, ANGPT1 and ANGPT2 delivered via adenoviral vectors induced capillary vessel enlargement in trachea, and this vascular phenotype was not observed in mice with Tie1 deletion.16

By employing genetic mouse models targeting Tie1, Tek (encoding TIE2), or both, we have explored functions of TIE1 and its relationship with TIE2 in the process of venogenesis. We found in this study that vein formation was retarded in TIE1-deficient mice, which showed an increase of vein-associated angiogenic sprouting during embryogenesis as well as in the postnatal retinal vascular development. Loss of TIE1 led to the decrease of TIE2 on the levels of mRNA and protein. Combining the endothelial deletion of Tie1 plus 1 null allele of Tek led to the formation of retinal vein-associated vascular tufts similar to that observed in the endothelial Tek-deleted mice. We further showed that the endothelial deletion of Nr2f2 resulted in the decrease of TIE1 and TIE2, suggesting that COUP-TFII regulates the expression of TIE receptors. Findings from this study imply that TIE1 acts in synergy with TIE2 to restrict angiogenesis for venous assembly during the establishment of a hierarchical vascular network.

MATERIALS AND METHODS

The authors declare that all supporting data are available within the article (and its Supplemental Material).

Mouse Models

All animal experiments were performed in accordance with the institutional guidelines of Soochow University Animal Center. All the mice used in this study were housed in a SPF (specific pathogen free) animal facility with a 12/12 hours dark/light cycle and were free to food and water access. Normal mouse diet (Suzhou Shuangshi Experimental Animal Feed Technology, Co, Ltd) and cage bedding (Suzhou Baitai Laboratory Equipment, Co, Ltd) were used. Two genetically modified mouse models targeting Tie1 gene were employed in this study. One mouse line targets TIE1 intracellular kinase domain (ICD), with exon 15 and exon 16 floxed, Tie1ICDFlox/Flox.12 The other line is a knockout first mouse model established from EUCOMM embryonic stem cells (EPD0735-3B07) targeting Tie1 gene (Tie1tm1a/tm1a), in which targeting cassette is recombined downstream of exon 7 (with exon 8 floxed). Tek knockout mouse model was generated as previously reported3 and was crossbred with Tie1ΔICD/ΔICD mouse model to obtain Tie1ICD and Tek double knockout mice (Tie1ΔICD/ΔICD;Tek-/-). To generate mice with endothelial cell–specific gene deletion mouse models, we employed the Cdh5-CreERT2 mouse line.25 In all the phenotype analysis, wildtype or heterozygous littermates were used as controls. Mice were bred in SPF experimental animal facilities. For the genotyping of Tie1tm1a knockout allele, the primers used were forward primer (5’- GCATGAAACTTCGCAAGCCA -3’) and reverse primer (5’- CTCTGCTGTGGTCCTGTCTG -3’), to amplify a 326 bp fragment for the wild-type allele and a 387 bp for the knockout allele. For the genotyping of Tie1ICD and Tek mouse lines, the primers used were as previously described.12Nr2f2Flox/Flox mice was generated and kindly provided by Dr Tsai’s laboratory.26 The genetic background of Tie1tm1a/tm1a is on C57BL/6 N, and the other lines are on C57BL/6J or C57BL/6J/SV129.

Induced Gene Deletion

Induction of gene deletion was performed as previously described by tamoxifen treatment.3,27 Briefly, new-born pups were treated by 3 or 4 daily intragastric injections of tamoxifen from postnatal day 1 (Sigma-Aldrich, T5648-5G; 60 μg daily). The genotypes of the Tie1/Tek knockout and control mice are as follows: Tie1ICDFlox/−;Cdh5-CreERT2 labeled as Tie1ICDiECKO and Tie1ICDFlox/−;Cdh5-CreERT2;Tek+/− labeled as Tie1ICDiECKO;Tek+/−, and the corresponding littermate control mice (labeled as control) are Tie1ICDFlox/+;Cdh5-CreERT2 and Tie1ICDFlox/+;Cdh5-CreERT2;Tek+/+. The genotypes of the Nr2f2 knockout and control mice are as follows: Nr2f2Flox/Flox;Cdh5-CreERT2 labeled as Nr2f2iECKO and Nr2f2Flox/Flox used as control. Tissues were collected for analysis at postnatal day 7-21. The retina dissection was according to the protocol by Pitulescu et al.28 Retinal vascularization index was quantified as the ratio of vascularized area to total retinal area as previously published.3

RNA Sequencing Analysis

Skin tissues were harvested from Tie1 knockout and control mice (E17.5) and kept frozen in liquid nitrogen. Total RNA was extracted from the tissues using Trizol (Invitrogen) according to the manufacturer’s instruction. RNA was qualified and quantified using a Nano Drop and Agilent 2100 bioanalyzer (Thermo Fisher Scientific). Oligo(dT)-attached magnetic beads were used to purify mRNA. Purified mRNA was fragmented, converted into cDNA and amplified by PCR, and finally made into RNAseq libraries. Single-end 50 bases reads were generated with a target depth of 20 million reads on BGIseq500 platform (BGI-Shenzhen, China). The sequencing data were filtered with SOAPnuke (v1.5.2), and clean reads were obtained and stored in FASTQ format. The clean reads were mapped to the reference genome using HISAT2 (v2.0.4). Bowtie2 (v2.2.5) was applied to align the clean reads to the reference coding gene set. Gene expression level was calculated by RSEM (v1.2.12), and differential expression analysis was performed using the DESeq2 (v1.4.5) with Q value ≤0.05.29–33 Gene enrichment analysis was performed using GSEA R package clusterProfiler (v3.14.3).34–36 The raw and processed data of RNA-seq analysis have been deposited in GEO (accession number: GSE229118).

RT-PCR

Total RNA from the lung and skin tissues of Tie1ΔICD/ΔICD and control embryos (E15.5) were extracted by TRIzol following the manufacturer’s protocol. cDNA was synthesized using a reverse transcriptional reaction kit (RevertAid First Strand cDNA Synthesis Kit, Thermo Scientific). Real-time quantitative PCR was performed using a SYBR Premix Ex Taq kit (Takara RR420A) in the Applied Biosystems 7500 Real-Time PCR System. Primers of qRT-PCR are as follows: Gapdh: 5’-GGTGAAGGTCGGTGTGAACG-3’, 5’-CTCGCTCCTGGAAGATGGTG-3’; Tie1: 5’-GCTGTGGTAGGTTCCGTCTC-3’, 5’-AAGGTCCCTGAGCTGAACTG-3’; Tek: 5’-GATTTTGGATTGTCCCGAGGTCAAG-3’, 5’-CACCAATATCTGGGCAAATGATGG-3’; Aplnr: 5’-CAGTCTGAATGCGACTACGC-3’, 5’-CCATGACAGGCACAGCTAGA-3’; Efnb2: 5’-TGTTGGGGACTTTTGATGGT-3’, 5’-GTCCACTTTGGGGCAAATAA-3’; Notch1: 5’-TGTTGTGCTCCTGAAGAACG-3’, 5’-TCCATGTGATCCGTGATGTC-3’; Dll4: 5’-TGCCTGGGAAGTATCCTCAC-3’, 5’-GTGGCAATCACACACTCGTT-3’. The transcripts of venous and arterial markers were normalized against Gapdh, and the relative expression level of every gene in the Tie1ΔICD/ΔICD mice was normalized against that of littermate control mice.

Cell Culture and siRNA Transfection

Human umbilical vein endothelial cells (HUVECs) were cultured in endothelial cell medium (ScienCell Research Laboratories #1001). To knock down Tie1 expression in HUVECs, cells were transfected with siRNA targeting human Tie1 (s14141 or s14142; Invitrogen) using Lipofectamine RNAiMax (Invitrogen), and siRNA negative control duplexes (12935300; Invitrogen) were used as a control.

Western Blotting Analysis

Lung and skin tissues of embryos were harvested, snap-frozen in liquid nitrogen and stored at -80 °C freezer. Tissues or cells were lysed in NP-40 lysis buffer (Beyotime P0013F) supplemented with protease inhibitor cocktail (complete Mini, Roche 04693124001), phosphatase inhibitor cocktail (PhosSTOP, Roche 04906837001), 10 mmol/L NaF and 1 mmol/L PMSF. Protein concentration was determined using the BCA protein assay kit (PIERCE), and equal amounts of protein were used for analysis. Briefly, after the protein transfer from gels to PVDF membranes (IPVH00010, Millipore) and the antibody incubations, images were acquired by the chemiluminescent detection method (NEL105001EA, PerkinElmer) using X-ray film (XBT, 6535876, Carestream). The protein markers were manually marked on the films overlapped with the PVDF membranes with the prestained protein ladder (26616, ThermoFisher Scientific). For the detection of beta-actin as the loading controls in this study, the entire blot was washed and reprobed to visualize it as described above. The following antibodies were used in this study, including goat polyclonal anti-TIE2 (R&D Systems AF762), goat polyclonal anti-TIE1 (R&D Systems AF619), rabbit polyclonal anti-AKT (Cell Signaling Technology No. 9272), rabbit monoclonal anti-Phospho-AKT (Ser473, Cell Signaling Technology No. 4060), mouse monoclonal anti-COUP-TFII (R&D Systems No. PP-H7147-00), mouse monoclonal to beta-Actin (Santa Cruz sc-47778), Goat anti-Mouse IgG, HRP Conjugated (Fcmacs Biotech FMS-MS01), Bovine Anti-Goat IgG, HRP Conjugated (Jackson ImmunoResearch Labs 805-035-180), Goat anti-Rabbit IgG, HRP Conjugated (R&D Systems HAF008).

Immunostaining

For the embryo studies, female mice were mated in the late afternoon and vaginal plugs were checked in the morning of the following day. The embryonic stages are estimated considering midday of the day on which the vaginal plug is present as embryonic day 0.5 (E0.5). Yolk sacs or tail tips from embryos were collected for genotyping. For immunostaining, tissues were harvested and processed as previously described.37 Briefly, the tissues were fixed in 4% paraformaldehyde, blocked with 3% (w/v) skim milk in PBS-TX (0.3% Triton X-100), and incubated with primary antibodies overnight at 4 °C. The antibodies used were: rat-anti-mouse PECAM1 (platelet endothelial cell adhesion molecule 1; BD 553370), 660-mouse-anti-mouse αSMA (alpha smooth muscle actin; eBioscience 50-9760), Cy3-mouse-anti-mouse aSMA (Sigma C6198), mouse-anti-mouse αSMA (Sigma A2547), goat-anti-human TIE1 (R&D AF619), goat-anti-mouse TIE2 (R&D AF762), goat-anti-mouse DLL4 (delta-like canonical Notch ligand 4; R&D AF1389), goat-anti-mouse EphB4 (R&D AF446), rat-anti-mouse endomucin (eBioscience 14-5851), rabbit-anti-all NG2 (Chemicon AB5320). Appropriate Alexa 488, Alexa 594 (Invitrogen) conjugated secondary antibodies were used. All fluorescently labeled samples were mounted and analyzed with a confocal microscope (Olympus Flueview 1000), or Leica MZ16F fluorescent dissection microscope. For comparison, the parameters were kept consistent for all the confocal microscopic imaging in this study.

Statistical Analysis

For the statistical analysis of multiple comparisons, the 1-way ANOVA was performed if data passed the D’Agostino-Pearson normality test, or the nonparametric Kruskal-Wallis test was used instead (GraphPad Prism 7). For the statistical analysis for 1-way ANOVA and Kruskal-Wallis test, Dunn’s multiple comparisons test was performed as post hoc test. For the 2-group comparison, the unpaired t test was performed with Welch’s correction if data passed the D’Agostino-Pearson normality test, or the unpaired nonparametric Mann-Whitney U test was applied using GraphPad Prism 7. Data are expressed as mean±SD. All statistical tests were 2-sided.

RESULTS

Differential Requirement of TIE1 and TIE2 in Cardinal Vein Specification

To investigate if TIE receptors are required for the specification of cardinal veins, we analyzed the Tie1 and Tek mutant mice during early embryogenesis. As shown by immunofluorescent staining of the pan-endothelial marker PECAM1, plus TIE1 or TIE2 in Figure 1A (embryonic day 8.75, E8.75), both TIE receptors were expressed in the dorsal aortas (DA) and anterior cardinal veins. To further characterize their functions in the cardinal vein formation, we deleted Tie1, Tek or both as previously reported.3,12 We found that the cardinal veins as well as dorsal aortas (7 to 14-somite stages, E8.75) appeared normal in Tie1 deleted mice (Figure 1B), with DLL4 expression in the dorsal aortas but not in cardinal veins. However, TIE2 deficiency led to the upregulation of DLL4 in anterior and posterior cardinal veins (Figure 1C; Figure S1). This is consistent with the previous observation that TIE2 is required for the specification of venous identity.3 Furthermore, the increased DLL4 expression was also detected in the cardinal veins of double Tie1/Tek knockout mice (Tie1ΔICD/ΔICD;Tek−/−), which did not seem to have additional abnormality with cardinal veins as shown in Figure 1D. Aberrant expression of DLL4 was also detected in head regions of Tek and Tie1/Tek double mutant mice (asterisks in Figure 1E and Figure S1). This is consistent with the lack of vein formation in the head regions at later stages (E10.5) observed in Tek mutant mice.3 As shown in Figure 1F, the DLL4 immunostaining signals are first quantified in anterior cardinal veins, which were then normalized against its level in dorsal aortas (DAs) of Tie1ΔICD/ΔICD, Tek−/−, Tie1ΔICD/ΔICD;Tek−/− and control mice. For all the images used for the quantification, the parameters were kept consistent for the confocal microscopic imaging.

Figure 1.

Figure 1. Differential requirement of TIE1 (tyrosine kinase with immunoglobulin-like and EGF [epithelial growth factor]-like domains 1) and TIE2 in cardinal vein specification at early embryogenesis. A, Analysis of cardinal veins and aortas (E8.75) by the whole-mount immunostaining for PECAM1 (platelet endothelial cell adhesion molecule 1; green) and TIE1 or TIE2 (red). B through E, Tie1 deletion did not affect cardinal vein formation (B), while Tek deletion altered the endothelial cell identity of cardinal veins (C–E; E8.5-9.0, 7 to 14-somite stages). Expression of arterial marker DLL4 (delta-like canonical Notch ligand 4; red, green for PECAM1) was detected in cardinal veins. Vein formation was disrupted in Tek null (Tek−/−, C) or Tie1/Tek double knockout mice (Tie1∆ICD/∆ICD;Tek−/−, D, E). F, Quantification of DLL4 immunostaining signals in anterior cardinal vein (ACV) normalized against its level in dorsal aorta (DA) at E8.75 (Tie1ΔICD/ΔICD: 0.11±0.08, n=5; Control: 0.18±0.18, n=5; P=0.6905. Tek−/−: 0.49±0.10, n=4; control: 0.14±0.08, n=6; P=0.0095. Tie1ΔICD/ΔICD;Tek−/−: 0.60±0.17, n=5; control: 0.14±0.09, n=4; P=0.0159). Asterisk in E points to DLL4 negative vessels, which are absent in Tie1/Tek double knockout mice. Arrows point to anterior cardinal veins and arrowheads to dorsal aortas. Scale bar: 200 µm in A, B, and D (far left); and the rest scale bars in A–E, 50 µm.

Disruption of Cutaneous Vein Formation and Alignment With Arteries After Tie1 Deletion

TIE2 is known to be more abundant in veins than arteries during early embryogenesis.3 Also, TIE1 was detected mainly in veins in head (Figure 2A) and somite regions at E10.5 (Figure 2B). Although TIE2 deficiency disrupted the vein formation in head and somite regions as observed at E9.5,3 lack of TIE1 did not have an obvious effect on vein formation (Figure 2A and 2B, E10.5), or the recruitment of vascular mural cells in head regions (Figure S2A). However, in addition to an increase of nonvascularized regions in the dorsal skin (Figure S2B), the vein formation was disrupted in the skin of Tie1 mutant mice in comparison with the heterozygous and wildtype controls (Figure 2C; Figure S2C). Small arteries but no veins were detected in regions close to the midline of the dorsal skin in the Tie1ΔICD/ΔICD mice (Figure 2D). Blood vessels in these regions also showed more angiogenic sprouting and less pericyte coverage in the Tie1 mutants in comparison with the controls (Figure S2D). There was also retarded lymphatic growth toward the middle line of the back skin (Figure S3). Interestingly, the veins were detected at the upper part of the dorsal cutaneous vascular network close to the axilla region (E15.5, Figure 2D and D’), but they were not properly aligned with the arteries.

Figure 2.

Figure 2. Defective vein formation after Tie1 deletion. A and B, Visualization of blood vessels in head (A) and somite regions (B, E10.5) of Tie1∆ICD/∆ICD and control mice by whole-mount immunostaining for PECAM1 (platelet endothelial cell adhesion molecule 1; red) and TIE1 (tyrosine kinase with immunoglobulin-like and EGF [epidermal growth factor]-like domains 1; green). Note that TIE1 is mainly expressed in veins (arrows) but with less expression in arteries in head and somite regions (arrowheads) at early stages of embryogenesis. C, Analysis of blood vessels in the skin of Tie1 deleted and control mice (E15.5) by the immunostaining for PECAM1 (green) and αSMA (alpha smooth muscle actin; red). Note that in comparison with the well-aligned veins and arteries in the skin of control mice, there were no veins but arteries detected in regions close to the midline of the dorsal skin in Tie1 mutant mice. Subcutaneous edema was indicated by asterisk in Tie1 mutants (C). D and E, Underdeveloped veins were accompanied by the increased staining of DLL4 in the skin of Tie1∆ICD/∆ICD (E15.5) and the wild-type littermates were used as the control (PECAM1, green; αSMA, red; DLL4, white). Note that veins (arrows in D and D’, DLL4 negative) were detected in the upper part of the dorsal cutaneous vascular network but not properly aligned with arteries in Tie1 mutant mice. The red dashed lines indicate the middle line of back skin. Arrows point to veins and arrowheads to arteries. Scale bar: 200 µm in A, C, D’, and 100 µm in B, D.

Retardation of Venous Assembly Upon TIE1 Deficiency

As previously reported,12 the Tie1∆ICD/∆ICD mutant mouse model was originally designed to investigate functions of TIE1 intracellular kinase domain in vascular development. However, because of the low level of expression of the truncated form of TIE1 lacking the intracellular domain (TIE1∆ICD), it is almost equivalent to a Tie1 complete knockout mouse line. The lymphatic phenotype of Tie1∆ICD/∆ICD mutants was similar to that observed in Tie1 knockout mice.38 To verify the role of TIE1 in the regulation of vein formation, we generated a new mouse line targeting Tie1 gene (Tie1tm1a), which was a knockout first allele with Tie1 exon 8 flanked by loxP sites (Figure S4A). Tie1 deletion was confirmed by the PCR genotyping and Western blotting analysis (Figure S4B and S4C). Sequential analysis of cutaneous vascular development at different stages of embryogenesis revealed that vein formation was initiated at approximately E13.5 close to the axilla region of dorsal skin, proceeding towards the midline (Figure 3A). Veins aligned with the arteries were observed from E14.0 onwards in the control mice (E14.0-15.5; Figure 3B, 3C, and 3E). However, the development of veins was seriously retarded in Tie1tm1a/tm1a mutant mice, being detected only at the upper part of dorsal cutaneous vascular network. Furthermore, DLL4 staining throughout the cutaneous vascular network was increased in the Tie1 deleted mice in comparison with their littermate controls (Figure 3C and 3E). As shown in Figure 3D, the DLL4 immunostaining signals are first quantified in nonarterial vessels, which were then normalized against its level in arteries in the skin of Tie1tm1a/tm1a and control mice at E14.5. For comparison, the parameters were kept consistent for the confocal microscopic imaging. Furthermore, the veins were not aligned properly with arteries compared with those in the littermate control mice (Figure 3F).

Figure 3.

Figure 3. Requirement of TIE1 (tyrosine kinase with immunoglobulin-like and EGF [epithelial growth factor]-like domains 1for the formation of cutaneous veins in alignment with arteries. A–E, Sequential analysis of vein formation in the skin of Tie1tm1a/tm1a mice and littermate controls between E13.5-E15.5 (A, E13.5; B, E14.0; C, E14.5; E, E15.5) by whole-mount immunostaining for PECAM1 (platelet endothelial cell adhesion molecule 1) or Endomucin (green) and DLL4 (delta-like canonical Notch ligand 4; red). Note that arteries (DLL4 positive, arrowheads) were readily detected while the process of vein formation (PECAM1 positive but negative for DLL4) started at approximately E13.5 (A, arrows). Veins positive for Endomucin but negative for DLL4 were detected in the skin of Tie1 mutant mice at the upper part of the vascular network near the axilla region of dorsal skin (B, C, E) but failed to grow along arteries (C, box 1´ and 2´), as compared with those of the control mice (C, box 1 and 2). Note that there was an obvious increase of DLL4 signals in Tie1 null mice as shown by the immunostaining. Quantification of the DLL4 immunostaining signals in nonarterial vessels normalized against its level in arteries of dorsal skin at E14.5 (D, Tie1tm1a/tm1a: 0.33±0.06, n=4; Control: 0.21±0.04, n=8; P=0.0081). F, Retardation and misalignment of veins with arteries in the skin of Tie1∆ICD/∆ICD mice (E18.5). Consistent with the observation at E15.5, veins were detected at the upper part of the dorsal cutaneous vascular network in the skin of Tie1 mutant mice (E18.5) but did not align properly with arteries as those in the littermate controls. Arrows point to veins and arrowheads point to arteries. Scale bar: 100 µm in A through F.

Abnormal Sprouting Angiogenesis After the Loss of TIE1

Consistent with the increased DLL4 expression in blood vessels of Tie1 mutant mice, we found that lack of TIE1 was associated with abnormal vascular sprouting that impaired the venous integrity in the head regions (Figure 4A). The delayed formation of veins was also observed in the mesentery of Tie1tm1a/tm1a mice. The vascular structures positive for endomucin, expressed by venous but not arterial endothelial cells, were not properly formed in the mutant mice compared with the littermate controls (Figure 4B, E13.5). In contrast to the mature mesentery veins aligned with arteries in the control mice (E17.5), mesentery veins of Tie1 mutant mice still underwent active sprouting (Figure S4D). This was further confirmed by the immunostaining for VE-cadherin showing that mesentery veins of Tie1tm1a/tm1a mutants (E17.5) displayed abnormal sprouting and that venous endothelial cell junctions appeared sparse while there were no obvious defects observed in arteries (Figure 4C). Similar venous abnormalities were observed in Tie1ΔICD/ΔICD mice as shown in Figure S5A and S5B. The vein-associated sprouting was also accompanied by the aberrant expression of DLL4, an arterial endothelial cell marker with also weak expression by lymphatic ECs (Figure S5C).

Figure 4.

Figure 4. Disruption of venous integrity accompanied by angiogenic sprouting upon Tie1 deletion. A, Disorganized endothelial junctions were observed in the vessel wall of veins in head regions as shown by PECAM1 (platelet endothelial cell adhesion molecule 1) staining (A, E10.5, Tie1tm1a/tm1a; TIE1 [tyrosine kinase with immunoglobulin-like and EGF (epithelial growth factor)-like domains 1], green; PECAM1, red). B, The mesentery vein formation was retarded in Tie1tm1a/tm1a mice compared with that of the littermate controls (E13.5). Arrows point to veins (Endomucin, green) and arrowheads to arteries (DLL4 [delta-like canonical Notch ligand 4], red). C, Analysis of veins and arteries of mesentery (E17.5) as demonstrated by the staining for VE-Cadherin (green). The EC junctions appeared to have an increased distance in the vein walls of Tie1 mutant mice compared with the controls. Veins were not properly aligned with arteries in the mesentery of Tie1tm1a/tm1a mice and arrows point to vein-associated angiogenic sprouts in Tie1 mutant mice (C). Consistent with previous findings, the formation of collecting lymphatics were disrupted in Tie1 mutant mice (PROX1, white). D, Presence of red blood cells in the mesentery lymphatic vessels of Tie1tm1a/tm1a mice (E17.5, yellow arrows). White arrows point to veins and arrowheads to arteries. Scale bar: 50 µm in A–D.

Consistent with the lymphatic defects observed in Tie1ΔICD/ΔICD mutants,12 collecting vessel formation was also disrupted in Tie1tm1a/tm1a mice compared with the control at E17.5 (Figure 4C). The loss of vascular integrity was confirmed by the presence of red blood cells detected in the mesenteric lymphatic vessels of Tie1tm1a/tm1a mice (Figure 4D; Figure S4E).

Alteration of Venous and Angiogenic Gene Expression After Tie1 Deletion

RNA sequencing analysis of skin tissues from TIE1 deficient mice (E17.5, Tie1tm1a/tm1a) revealed the altered expression of vascular genes involved in the regulation of venogenesis, angiogenesis as well as hypoxia (Figure 5A through 5C). The hallmark gene sets, including the hallmarks for vein, artery and angiogenesis used in the GSEA (Gene Set Enrichment Analysis, Table S1), are mainly defined according to the transcriptome atlas of murine endothelial cells by Kalucka et al.39 Briefly, the top 50 marker genes of venous, angiogenic or arterial endothelial cells identified in ten tissues by Kalucka et al were pooled and the genes expressed in 3 or more tissues were included in the hallmark gene sets of vein, artery and angiogenesis. The hallmark gene sets of hypoxia and other biological processes are based on the molecular signatures database (h.all.v7.4.symbols).34,35 As shown in Figure 5B, the venous genes were significantly downregulated whereas arterial genes were not significantly altered in the skin of Tie1 mutant mice (E17.5). Interestingly, Tie1 deletion led also to a significant upregulation of genes involved in angiogenesis and hypoxia (Figure 5B). Heatmaps of the subset of the differentially expressed vein or angiogenesis related genes were shown in Figure 5C. It is worth pointing out that Tie1 deletion led to a decrease of venous regulators at the transcript level such as Tek and Aplnr (Figure 5C). This was confirmed by the quantitative PCR analysis of skin and lung tissues from Tie1ΔICD/ΔICD mice (Figure 5D; Table 1 and Table 2). Consistent with the immunofluorescent staining, Dll4 transcripts were upregulated while there were no obvious changes with the expression of arterial markers such as Efnb2 and Notch1 (Figure 5D; Table 1 and Table 2).

Figure 5.

Figure 5. Alteration of venous and angiogenic gene expression after TIE1 (tyrosine kinase with immunoglobulin-like and EGF [epithelial growth factor]-like domains 1) deficiency. A through C, RNAseq analysis of the skin tissues of Tie1 null (Tie1tm1a/tm1a) and control mice. The Volcano plot of the differentially regulated genes was shown in A. GSEA (gene set enrichment analysis) analysis revealed that there was a significant decrease in the expression of venous genes but an increase in angiogenesis and hypoxia genes after Tie1 deletion. In contrast, there was no significant difference in the expression of artery genes (B). The subset of the upregulated angiogenesis genes and downregulated venous genes (P<0.05) was shown as heatmaps in C. D, Quantitative expression analysis of venous genes including Aplnr and Tek and arterial genes including Efnb2, Notch1, and Dll4 in skin and lung tissues of Tie1 mutant (Tie1∆ICD/∆ICD) and control mice (E15.5; Tables 1 and 2). E and F, Western blotting analysis and quantification of COUP-TFII and TIE2 protein normalized against beta-actin in lungs (COUP-TFII/beta-actin: Tie1ΔICD/ΔICD: 0.55±0.13, n=6; Control: 1.00±0.11, n=6; P=0.0022. TIE2/beta-actin:Tie1ΔICD/ΔICD: 0.62±0.17, n=6; Control: 1.00±0.17, n=6; P=0.0087), and skin of Tie1 mutant and control mice (COUP-TFII/beta-actin: Tie1ΔICD/ΔICD: 0.40±0.13, n=6; Control: 1.00±0.13, n=6; P=0.0022. TIE2/beta-actin:Tie1ΔICD/ΔICD: 0.50±0.24, n=6; Control: 1.00±0.18, n=6; P=0.0087). Quantification of AKT (RAC-alpha serine/threonine-protein kinase; Ser473) phosphorylation in lung is as follows (pAKT/tAKT [phosphorylated AKT/total AKT], Tie1ΔICD/ΔICD: 0.70±0.21, n=6; Control: 1.00±0.02, n=6; P=0.0043).

Table 1. The mRNA (messenger RNA) Expression Level of Venous and Arterial Endothelial Cell Markers in Skin Tissues of Tie1 Knockout and Control Mice

Skin transcript levels of venous and arterial markers
Venous and arterial markersmRNA expression level (skin)n (Tie1ΔICD/ΔICD)n (Control)P value
ControlTie1ΔICD/ΔICD
Tie11.00±0.060910<0.0001
Tek1.00±0.160.59±0.05910<0.0001
Aplnr1.00±0.100.27±0.10910<0.0001
Efnb21.00±0.130.95±0.159100.454
Notch11.00±0.140.98±0.179100.733
Dll41.00±0.072.00±0.36910<0.0001

Table 2. The mRNA (messenger RNA) Expression Level of Venous and Arterial Endothelial Cell Markers in Lung Tissues of Tie1 Knockout and Control Mice Lung Transcript Levels of Venous and Arterial Markers

Lung transcript levels of venous and arterial markers
Venous and arterial markersmRNA expression level (lung)n (Tie1ΔICD/ΔICD)n (Control)P value
ControlTie1ΔICD/ΔICD
Tie11.00±0.16099<0.0001
Tek1.00±0.100.67±0.199<0.0001
Aplnr1.00±0.160.53±0.1599<0.0001
Efnb21.00±0.141.04±0.19990.618
Notch11.00±0.141.07±0.25990.458
Dll41.00±0.131.31±0.28990.00684

Consistent with the downregulation of Tek expression, Tie1 deletion led to a decrease of TIE2 as well as COUP-TFII proteins as demonstrated in the skin and lung tissues of Tie1ΔICD/ΔICD mutant mice (Figures 5E and 5F). There was also a significant decrease of AKT phosphorylation in the lung of Tie1 mutants (Ser473, Figure 5F), which may account for the decrease of COUP-TFII as demonstrated in Tek-deficient mice.3 Furthermore, decrease of TIE2 and COUP-TFII proteins was further verified in cultured human umbilical vein endothelial cells (HUVECs) when TIE1 was reduced by the siRNA-mediated knockdown (Figure 6A through 6C). This suggests that TIE1 may function in vein development, at least partly, by regulating the expression of TIE2. Interestingly, the endothelial deletion of Tek also led to the decrease in TIE1 (Figure S6A through S6C).

Figure 6.

Figure 6. Synergy of TIE1 (tyrosine kinase with immunoglobulin-like and EGF [epithelial growth factor]-like domains 1) and TIE2 in retinal venous development. A through C, Western blotting analysis of COUP-TFII (chicken ovalbumin upstream promoter transcription factor, encoded by Nr2f2, nuclear receptor subfamily 2 group F member 2) protein in HUVECs after siRNA-mediated Tie1 knockdown. Two small-interfering (siRNAs) targeting Tie1 were used in this study, including S14141 and S14142. Quantification of TIE1, TIE2, and COUP-TFII protein level was as follows. For S14141: COUP-TFII was decreased to 0.61±0.27 vs control (1.00±0.24, n=15, P=0.0003) and TIE2 to 0.72±0.26 vs control (1.00±0.18, n=15, P=0.0018), when TIE1 expression was reduced to 0.15±0.07 vs control (1.00±0.14, n=15, P<0.0001; normalized by beta-actin; values from 5 independent experiments). For S14142: COUP-TFII was decreased to 0.54±0.20 vs control (1.00±0.23, n=9, P=0.0004) and TIE2 to 0.72±0.24 vs control (1.00±0.11, n=9, P=0.0092), when TIE1 expression was reduced to 0.20±0.06 vs control (1.00±0.15, n=9, P<0.0001; normalized by beta-actin; values from 3 independent experiments). D, Tamoxifen intragastric (i.g.) administration and the analysis scheme. E and F, Analysis of blood vessels in the retinas of Tie1ICDiECKO, Tie1ICDiECKO;Tek+/− and control mice between P7 and P21. Arrows point to hemangioma-like vascular tufts (E). Quantification of the vascularization index (F; ratio of vascularized area to total retina area normalized against the littermate controls at P7; Control: 1.00±0.05, n=10; Tie1ICDiECKO: 0.80±0.07, n=6; Tie1ICDiECKO; Tek+/−: 0.56±0.05, n=6; 1-way ANOVA Kruskal-Wallis test, P<0.0001; followed by Dunn multiple comparisons test: Tie1ICDiECKO vs Control, adjusted P=0.0502; Tie1ICDiECKO;Tek+/− vs Control, adjusted P<0.0001; Tie1ICDiECKO vs Tie1ICDiECKO;Tek+/−, adjusted P=0.3255). G, The deletion efficiency of TIE1 was examined by the Western blotting analysis of lung from Tie1ICDiECKO;Tek+/−, Tie1ICDiECKO and control mice at P7. H and I, Visualization of retinal blood vessels of Tie1ICDiECKO;Tek+/− mice by immunostaining for PECAM1 (platelet endothelial cell adhesion molecule 1) at P9, P11, P15 and P21. Arrows point to veins and arrowheads to arteries (H). Analysis of the 3 layers of retinal blood vessels in Tie1ICDiECKO, Tie1ICDiECKO;Tek+/− and control mice at P21 (I). Note that the vein-associated vascular defects were more severe in Tie1ICDiECKO;Tek+/− mice compared with the Tie1 deletion alone. Scale bar: 200 µm in H; 50 µm in I.

Synergy of TIE1 and TIE2 in Retinal Vein Morphogenesis

To further confirm the synergistic effect of TIE1 and TIE2 on vascular development, particularly vein formation, we generated mice with the endothelial cell–specific deletion of Tie1 (Tie1ICDiECKO: Tie1ICDFlox/−/Cdh5-CreERT2),12,25 and also the compound knockout mice targeting Tie1 plus 1 null allele of Tek (Tie1ICDiECKO;Tek+/−). The scheme for Tie1 deletion by the intragastric administration of tamoxifen is shown in Figure 6D. Deletion of Tie1 alone (Tie1ICDiECKO, P1-4) led to a significant decrease in the retinal vascularization at the postnatal day 7 (P7; Figure 6E and 6F). Interestingly, lack of TIE1 plus 1 null allele of Tek (Tie1ICDiECKO;Tek+/−) resulted in a more severe decrease in the retinal blood vessel growth (Figure 6E and 6F). Quantification of the retinal vascularization index was performed as previously published,3 and shown in Figure 6F. Consistently, mice heterozygous for Tie1/Tek double deletion displayed a decrease of retinal vascularization at the postnatal day 5, while there was no obvious effect observed on the retinal vascular formation with the deletion of 1 Tek allele at the same stage (Figure S7A and S7B).

Similar to the vascular defects observed in the retinas of Tek deleted mice,3 there was also a progressive increase of vein-associated angiogenesis leading to the formation of hemangioma-like vascular tufts in Tie1ICDiECKO mice and this became more severe in Tie1ICDiECKO;Tek+/− mice (from P9 to P21; Figure 6E, 6H, and 6I; Figure S7C; Figure 8A through 8D). Tie1 deletion was demonstrated by the immunostaining of retinas (Figure 8B) and Western blotting analysis of lung tissues from Tie1ICDiECKO, Tie1ICDiECKO;Tek+/− and control mice (Figure 6G). The retinal veins and arteries were identified by the immunostaining for EphB4 and DLL4 (Figure 8C and 8D). Vascular growth towards the deep layers of retinas was diminished in Tie1ICDiECKO and Tie1ICDiECKO;Tek+/− mice compared with the littermate controls (Figure 6I). Notably, the retinal venous defects were more severe in Tie1ICDiECKO;Tek+/− mice than the mutant mice with Tie1 deletion alone (Tie1ICDiECKO; Figure 6E, 6H, and 6I; Figure S7C), suggesting that TIE1 and TIE2 function in a synergistic manner in the regulation of retinal vascularization, particularly in retinal vein development.

Furthermore, the induced deletion of endothelial Nr2f2 resulted in the decrease of both TIE1 and TIE2 in lungs of Nr2f2iECKO compared with that of control mice (Figure 7A through 7C). Consistent with the observation in Tie1 (Tie1ICDiECKO) or Tie1/Tek double mutant mice (Tie1ICDiECKO;Tek+/−), lack of the endothelial COUP-TFII suppressed the retinal vascularization with abnormal sprouting angiogenesis (Figure 7D and 7E). It is worth noting that all the Nr2f2iECKO mice died in 2 weeks when the genetic knockout was induced at the neonatal stage from P1-3, indicating the successful deletion of endothelial Nr2f2. Shown in Figure 7F is a schematic illustration of TIE1 and TIE2, together with COUP-TFII, in the regulation of vein development.

Figure 7.

Figure 7. Decrease of TIE1 (tyrosine kinase with immunoglobulin-like and EGF [epithelial growth factor]-like domains 1) and TIE2 after the endothelial deletion of Nr2f2. A, Tamoxifen intragastric (i.g.) administration and the analysis scheme. B and C, Western blotting analysis and quantification of TIE1 and TIE2 protein normalized against beta-actin in lungs of Nr2f2iECKO and control mice at P7 (TIE1/beta-actin: Control: 1.00±0.21, n=10; Nr2f2iECKO: 0.71±0.15, n=10, P=0.0029. TIE2/beta-actin: Control: 1.00±0.08, n=10, Nr2f2iECKO: 0.66±0.10, n=10, P<0.0001). D and E, Analysis of blood vessels in the retinas of Nr2f2iECKO and control mice at P7 (D) and quantification of the vascularization index (E; Control: 1.00±0.10, n=5; Nr2f2iECKO: 0.68±0.06, n=5, P=0.0079). Arrows point to angiogenic sprouts in Nr2f2iECKO mice. F, Schematic illustration of the role of TIE1 in the establishment of the venous system. Lack of TIE1 leads to the abnormal angiogenesis associated with veins as well as in other regions with abnormal angiogenic sprouts within the vascular network of Tie1 mutant mice. In contrast to the crucial role of TIE2 in the regulation of venous fate, evidence from this study suggests that TIE1 participates in the process of vein assembly by the restriction of angiogenic sprouting. At the molecular level, COUP-TFII (chicken ovalbumin upstream promoter transcription factor, encoded by Nr2f2, nuclear receptor subfamily 2 group F member 2) is reduced after the induced deletion of TIE1 or TIE2 while the endothelial deletion of COUP-TFII also leads to the decrease of both TIE1 and TIE2. This suggests that TIE receptors and COUP-TFII coordinately regulate vein development by a positive feedback loop. Scale bar: 50 µm in D.

DISCUSSION

Functional mechanisms of endothelial TIE receptors have attracted intensive research over the last 3 decades since their discovery. The goal of this study is to investigate roles of TIE1 and its relationship with TIE2 in the establishment of veins. We show evidence in this study that TIE1 plays a crucial role in restricting angiogenesis for the assembly of veins and acts in a synergistic manner with TIE2 in this process.

TIE1 and TIE2 are differentially required during the blood vascular and lymphatic network formation.9,15,40 Recent studies have shown that TIE1 plays important roles in the regulation of blood vascular growth,16–18 in addition to its requirement in the formation of collecting lymphatics.12,15,38,41 TIE2 is required for the formation and maturation of blood vascular network, especially in vein development.3,9 We show in this study that the formation of cardinal veins is independent of TIE1. Cardinal veins are formed by a process of vasculogenesis, involving de novo formation of vessels from angioblasts.42 This suggests that TIE1 is not essential for the differentiation of venous ECs. Consistent with the previous observation,3 TIE2 is critical in the specification of cardinal venous ECs as evidenced by the aberrant expression of DLL4 upon the loss of TIE2. Interestingly, in contrast to the complete lack of cutaneous veins in Tek mutant mice,3 lack of TIE1 disrupted the initial formation of veins in skin. Interestingly, cutaneous veins were detected at later stages of embryonic development but displaying the arteriovenous misalignment. Therefore, TIE1 deficiency delayed the formation of veins, suggesting the differential requirement of TIE1 and TIE2 during the vein development.

In spite of the distinct functions, we have found that TIE1 and TIE2 act in a synergistic manner in the vein development. Induced deletion of Tek in neonatal mice led to the formation of angioma-like vascular tufts resulting from uncontrolled vein-associated angiogenesis in retinas.3 Loss of TIE1 resulted in a similar but relatively mild increase of retinal vein-associated angiogenesis in comparison with that of Tek mutants.3 Although there was no obvious vascular abnormality with mice heterozygous for Tek deletion in retinas, combining the induced endothelial knockout of Tie1 with 1 null allele of Tek produced a more severe vein-associated vascular phenotype in retinas than that of Tie1 deletion alone, with a comparable formation of the vascular tufts as those of Tek mutant mice. This points to a synergistic function of TIE1 and TIE2 in the restriction of sprouting angiogenesis during the vein formation. Synergistic roles of TIE1 and TIE2 in blood vascular formation were also demonstrated in other studies. Insufficient TIE1 decreased the ANGPT1-induced TIE2 activation and suppressed the enlargement of tracheal vessels by the treatment with angiopoietins.16 Delivery of soluble TIE2 by the adeno-associated virus (AAV) vectors plus the genetic deletion of Tie1 produced an additive effect on the inhibition of tumor angiogenesis.18 Furthermore, both TIE1 and TIE2 could regulate the expression of COUP-TFII, a key factor for the venous endothelial cell fate. Consistent with the findings in Tek mutant mice,3 we observed that COUP-TFII expression decreased in mice by the genetic deletion of Tie1 or in cultured endothelial cells by siRNA-mediated knockdown of Tie1. TIE1 deficiency also disrupted the arterial-venous alignment in tissues including skin and mesentery. This resembles the abnormal arteriovenous alignment of the Aplnr (also known as APJ) or Tek-deficient mice.3,43 A significant reduction of Aplnr expression was detected in the Tie1 deficient mice, which was also observed in Tek mutants as previously published.3 As the TIE1 deficiency reduced the expression of Tek, it is likely that TIE1 exerts its function, at least partly, via the regulation of TIE2 in the venous development.

How do TIE receptors exert their synergistic functions during the vascular development? TIE2 activation suppresses the Forkhead box protein O1 (FOXO1)-mediated transcriptional regulation via AKT pathway and inhibition of TIE2 activates FOXO1, leading to the increase of ANGPT2 expression.23,44 Induced deletion of Tie1 has also been shown to increase FOXO1 nuclear localization and transcriptional activation.16 In this study, we found that Tie1 deletion led to a decrease in the expression of venous endothelial related genes in embryonic skin such as Tek, Aplnr, Emcn, and an increase of angiogenic regulators including Angpt2, Vegfa and Dll4. This may account for the vein-associated increase of angiogenesis during embryogenesis and also in the retinas after the postnatal deletion of Tie1, which became more severe in the double mutants combining the endothelial Tie1 deletion with 1 null allele of Tek.

Furthermore, loss of TIE1 led to the disruption of vascular integrity as demonstrated by the vascular bleeding. Platelets play a central role in primary hemostasis during physiological and pathological conditions.45 It is worth noting that TIE1 deficiency leads to the decrease of P-selectin expression, which may interfere with the endothelial-platelet interaction.46 In addition, the decrease of VWF expression upon TIE1 deficiency also points to a role of TIE1 in the regulation of the coagulation system for vascular integrity. Besides the above discussed, endothelial adherens junctions are among the key regulatory components for the establishment and homeostasis of endothelial integrity.47 TIE1 and TIE2 are expressed by endothelial tip and/or stalk cells.17 In cultured endothelial cells, TIE1 and TIE2 are translocated to endothelial cell contacts upon ANGPT1 treatment,48 suggesting that TIE receptors are important components of the intercellular adhesions in addition to transducing signals for the assembly of vascular wall.

In summary, findings from this study indicate that TIE1 is required for the establishment of venous system. At the molecular level, COUP-TFII is reduced after the induced deletion of TIE1 or TIE2 while the endothelial deletion of COUP-TFII also leads to the decrease of both TIE1 and TIE2. This suggests a positive feedback loop among TIE receptors and COUP-TFII for the coordinated regulation of venogenesis.

ARTICLE INFORMATION

Acknowledgments

The authors thank the staff in Animal facility of Soochow University for technical assistance.

Supplemental Material

Figures S1–S8

Table S1

Major Resources Table

Nonstandard Abbreviations and Acronyms

COUP-TFII

chicken ovalbumin upstream promoter transcription factor, encoded by Nr2f2, nuclear receptor subfamily 2 group F member 2

EC

endothelial cell

ICD

intracellular domain

TEK

TEK receptor tyrosine kinase

TIE1

tyrosine kinase with immunoglobulin-like and EGF-like domains 1

Disclosures None.

Footnotes

For Sources of Funding and Disclosures, see page e337.

*X. Cao, T. Li, and B. Xu contributed equally.

Supplemental Material is available at https://www.ahajournals.org/doi/suppl/10.1161/ATVBAHA.122.318860.

Correspondence to: Dr Yulong He, Cyrus Tang Hematology Center, Collaborative Innovation Center of Hematology, National Clinical Research Center for Hematologic Diseases, State Key Laboratory of Radiation Medicine and Protection, Cam-Su Genomic Resources Center, Suzhou Medical College of Soochow University, 199 Ren-Ai Road, Suzhou 215123, China. Email

REFERENCES

  • 1. You LR, Lin FJ, Lee CT, DeMayo FJ, Tsai MJ, Tsai SY. Suppression of Notch signalling by the COUP-TFII transcription factor regulates vein identity.Nature. 2005; 435:98–104. doi: 10.1038/nature03511CrossrefMedlineGoogle Scholar
  • 2. Ren B, Deng Y, Mukhopadhyay A, Lanahan AA, Zhuang ZW, Moodie KL, Mulligan-Kehoe MJ, Byzova TV, Peterson RT, Simons M. ERK1/2-Akt1 crosstalk regulates arteriogenesis in mice and zebrafish.J Clin Invest. 2010; 120:1217–1228. doi: 10.1172/JCI39837CrossrefMedlineGoogle Scholar
  • 3. Chu M, Li T, Shen B, Cao X, Zhong H, Zhang L, Zhou F, Ma W, Jiang H, Xie P, et al. Angiopoietin receptor Tie2 is required for vein specification and maintenance via regulating COUP-TFII.eLife. 2016; 5:e21032. doi: 10.7554/eLife.21032CrossrefMedlineGoogle Scholar
  • 4. Natynki M, Kangas J, Miinalainen I, Sormunen R, Pietila R, Soblet J, Boon LM, Vikkula M, Limaye N, Eklund L. Common and specific effects of TIE2 mutations causing venous malformations.Hum Mol Genet. 2015; 24:6374–6389. doi: 10.1093/hmg/ddv349CrossrefMedlineGoogle Scholar
  • 5. Arita Y, Nakaoka Y, Matsunaga T, Kidoya H, Yamamizu K, Arima Y, Kataoka-Hashimoto T, Ikeoka K, Yasui T, Masaki T, et al. Myocardium-derived angiopoietin-1 is essential for coronary vein formation in the developing heart.Nat Commun. 2014; 5:4552. doi: 10.1038/ncomms5552CrossrefMedlineGoogle Scholar
  • 6. Thomson BR, Heinen S, Jeansson M, Ghosh AK, Fatima A, Sung H-K, Onay T, Chen H, Yamaguchi S, Economides AN, et al. A lymphatic defect causes ocular hypertension and glaucoma in mice.J Clin Invest. 2014; 124:4320–4324. doi: 10.1172/JCI77162CrossrefMedlineGoogle Scholar
  • 7. Kim J, Park DY, Bae H, Park DY, Kim D, Lee C-K, Song S, Chung T-Y, Lim DH, Kubota Y, et al. Impaired angiopoietin/Tie2 signaling compromises Schlemm’s canal integrity and induces glaucoma.J Clin Invest. 2017; 127:3877–3896. doi: 10.1172/JCI94668CrossrefMedlineGoogle Scholar
  • 8. Kim J, Park JR, Choi J, Park I, Hwang Y, Bae H, Kim Y, Choi WJ, Yang JM, Han S, et al. Tie2 activation promotes choriocapillary regeneration for alleviating neovascular age-related macular degeneration.Sci Adv. 2019; 5:eaau6732. doi: 10.1126/sciadv.aau6732CrossrefMedlineGoogle Scholar
  • 9. Augustin HG, Koh GY, Thurston G, Alitalo K. Control of vascular morphogenesis and homeostasis through the angiopoietin-Tie system.Nat Rev Mol Cell Biol. 2009; 10:165–177. doi: 10.1038/nrm2639CrossrefMedlineGoogle Scholar
  • 10. Kim KH, Nakaoka Y, Augustin HG, Koh GY. Myocardial angiopoietin-1 controls atrial chamber morphogenesis by spatiotemporal degradation of cardiac jelly.Cell Rep. 2018; 23:2455–2466. doi: 10.1016/j.celrep.2018.04.080CrossrefMedlineGoogle Scholar
  • 11. Qu X, Tompkins K, Batts LE, Puri M, Baldwin S. Abnormal embryonic lymphatic vessel development in Tie1 hypomorphic mice.Development. 2010; 137:1285–1295. doi: 10.1242/dev.043380CrossrefMedlineGoogle Scholar
  • 12. Shen B, Shang Z, Wang B, Zhang L, Zhou F, Li T, Chu M, Jiang H, Wang Y, Qiao T, et al. Genetic dissection of tie pathway in mouse lymphatic maturation and valve development.Arterioscler Thromb Vasc Biol. 2014; 34:1221–1230. doi: 10.1161/ATVBAHA.113.302923LinkGoogle Scholar
  • 13. D’Amico G, Korhonen EA, Waltari M, Saharinen P, Laakkonen P, Alitalo K. Loss of endothelial Tie1 receptor impairs lymphatic vessel development-brief report.Arterioscler Thromb Vasc Biol. 2010; 30:207–209. doi: 10.1161/ATVBAHA.109.196618LinkGoogle Scholar
  • 14. Gale NW, Thurston G, Hackett SF, Renard R, Wang Q, McClain J, Martin C, Witte C, Witte MH, Jackson D, et al. Angiopoietin-2 is required for postnatal angiogenesis and lymphatic patterning, and only the latter role is rescued by angiopoietin-1.Dev Cell. 2002; 3:411–423. doi: 10.1016/s1534-5807(02)00217-4CrossrefMedlineGoogle Scholar
  • 15. He Y. Angiopoietins and TIE Receptors in Lymphangiogenesis and Tumor Metastasis. In: Marmé D, ed. Tumor Angiogenesis. Springer Nature Switzerland AG; 2019:135–156.CrossrefGoogle Scholar
  • 16. Korhonen EA, Lampinen A, Giri H, Anisimov A, Kim M, Allen B, Fang S, D'Amico G, Sipilä TJ, Lohela M, et al. Tie1 controls angiopoietin function in vascular remodeling and inflammation.J Clin Invest. 2016; 126:3495–3510. doi: 10.1172/JCI84923CrossrefMedlineGoogle Scholar
  • 17. Savant S, La Porta S, Budnik A, Busch K, Hu J, Tisch N, Korn C, Valls AF, Benest AV, Terhardt D, et al. The orphan receptor Tie1 controls angiogenesis and vascular remodeling by differentially regulating Tie2 in tip and stalk cells.Cell Rep. 2015; 12:1761–1773. doi: 10.1016/j.celrep.2015.08.024CrossrefMedlineGoogle Scholar
  • 18. D’Amico G, Korhonen EA, Anisimov A, Zarkada G, Holopainen T, Hägerling R, Kiefer F, Eklund L, Sormunen R, Elamaa H, et al. Tie1 deletion inhibits tumor growth and improves angiopoietin antagonist therapy.J Clin Invest. 2014; 124:824–834. doi: 10.1172/JCI68897CrossrefMedlineGoogle Scholar
  • 19. Seegar TC, Eller B, Tzvetkova-Robev D, Kolev MV, Henderson SC, Nikolov DB, Barton WA. Tie1-Tie2 interactions mediate functional differences between angiopoietin ligands.Mol Cell. 2010; 37:643–655. doi: 10.1016/j.molcel.2010.02.007CrossrefMedlineGoogle Scholar
  • 20. Yuan HT, Venkatesha S, Chan B, Deutsch U, Mammoto T, Sukhatme VP, Woolf AS, Karumanchi SA. Activation of the orphan endothelial receptor Tie1 modifies Tie2-mediated intracellular signaling and cell survival.FASEB J. 2007; 21:3171–3183. doi: 10.1096/fj.07-8487comCrossrefMedlineGoogle Scholar
  • 21. Marron MB, Singh H, Tahir TA, Kavumkal J, Kim HZ, Koh GY, Brindle NP. Regulated proteolytic processing of Tie1 modulates ligand responsiveness of the receptor-tyrosine kinase Tie2.J Biol Chem. 2007; 282:30509–30517. doi: 10.1074/jbc.M702535200CrossrefMedlineGoogle Scholar
  • 22. Saharinen P, Kerkela K, Ekman N, Marron M, Brindle N, Lee GM, Augustin H, Koh GY, Alitalo K. Multiple angiopoietin recombinant proteins activate the Tie1 receptor tyrosine kinase and promote its interaction with Tie2.J Cell Biol. 2005; 169:239–243. doi: 10.1083/jcb.200411105CrossrefMedlineGoogle Scholar
  • 23. Kim M, Allen B, Korhonen EA, Nitschké M, Yang HW, Baluk P, Saharinen P, Alitalo K, Daly C, Thurston G, et al. Opposing actions of angiopoietin-2 on Tie2 signaling and FOXO1 activation.J Clin Invest. 2016; 126:3511–3525. doi: 10.1172/JCI84871CrossrefMedlineGoogle Scholar
  • 24. Leppanen VM, Saharinen P, Alitalo K. Structural basis of Tie2 activation and Tie2/Tie1 heterodimerization.Proc Natl Acad Sci U S A. 2017; 114:4376–4381. doi: 10.1073/pnas.1616166114CrossrefMedlineGoogle Scholar
  • 25. Okabe K, Kobayashi S, Yamada T, Kurihara T, Tai-Nagara I, Miyamoto T, Mukouyama Y-suke, Sato TN, Suda T, Ema M, et al. Neurons limit angiogenesis by titrating VEGF in retina.Cell. 2014; 159:584–596. doi: 10.1016/j.cell.2014.09.025CrossrefMedlineGoogle Scholar
  • 26. Takamoto N, You LR, Moses K, Chiang C, Zimmer WE, Schwartz RJ, DeMayo FJ, Tsai MJ, Tsai SY. COUP-TFII is essential for radial and anteroposterior patterning of the stomach.Development. 2005; 132:2179–2189. doi: 10.1242/dev.01808CrossrefMedlineGoogle Scholar
  • 27. Cao X, Xu B, Li X, Li T, He Y. A genetically engineered mouse model of venous anomaly and retinal angioma-like vascular malformation.Bio Protoc. 2021; 11:e4117. doi: 10.21769/BioProtoc.4117CrossrefMedlineGoogle Scholar
  • 28. Pitulescu ME, Schmidt I, Benedito R, Adams RH. Inducible gene targeting in the neonatal vasculature and analysis of retinal angiogenesis in mice.Nat Protoc. 2010; 5:1518–1534. doi: 10.1038/nprot.2010.113CrossrefMedlineGoogle Scholar
  • 29. Kim D, Langmead B, Salzberg SL. HISAT: a fast spliced aligner with low memory requirements.Nat Methods. 2015; 12:357–360. doi: 10.1038/nmeth.3317CrossrefMedlineGoogle Scholar
  • 30. Love MI, Huber W, Anders S. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2.Genome Biol. 2014; 15:550. doi: 10.1186/s13059-014-0550-8CrossrefMedlineGoogle Scholar
  • 31. Langmead B, Salzberg SL. Fast gapped-read alignment with Bowtie 2.Nat Methods. 2012; 9:357–359. doi: 10.1038/nmeth.1923CrossrefMedlineGoogle Scholar
  • 32. Li B, Dewey CN. RSEM: accurate transcript quantification from RNA-Seq data with or without a reference genome.BMC Bioinf. 2011; 12:323. doi: 10.1186/1471-2105-12-323CrossrefMedlineGoogle Scholar
  • 33. Li R, Li Y, Kristiansen K, Wang J. SOAP: short oligonucleotide alignment program.Bioinformatics. 2008; 24:713–714. doi: 10.1093/bioinformatics/btn025CrossrefMedlineGoogle Scholar
  • 34. Subramanian A, Tamayo P, Mootha VK, Mukherjee S, Ebert BL, Gillette MA, Paulovich A, Pomeroy SL, Golub TR, Lander ES, et al. Gene set enrichment analysis: a knowledge-based approach for interpreting genome-wide expression profiles.Proc Natl Acad Sci U S A. 2005; 102:15545–15550. doi: 10.1073/pnas.0506580102CrossrefMedlineGoogle Scholar
  • 35. Liberzon A, Birger C, Thorvaldsdottir H, Ghandi M, Mesirov JP, Tamayo P. The Molecular Signatures Database (MSigDB) hallmark gene set collection.Cell Syst. 2015; 1:417–425. doi: 10.1016/j.cels.2015.12.004CrossrefMedlineGoogle Scholar
  • 36. Morgan M FS, Gentleman R. GSEABase: Gene set enrichment data structures and methods. Bioconductor R package version 1.56.0. 2021. https://doi.org/10.18129/B9.bioc.GSEABase.Google Scholar
  • 37. Zhang L, Zhou F, Han W, Shen B, Luo J, Shibuya M, He Y. VEGFR-3 ligand-binding and kinase activity are required for lymphangiogenesis but not for angiogenesis.Cell Res. 2010; 20:1319–1331. doi: 10.1038/cr.2010.116CrossrefMedlineGoogle Scholar
  • 38. Qu X, Zhou B, Scott Baldwin H. Tie1 is required for lymphatic valve and collecting vessel development.Dev Biol. 2015; 399:117–128. doi: 10.1016/j.ydbio.2014.12.021CrossrefMedlineGoogle Scholar
  • 39. Kalucka J, de Rooij L, Goveia J, Rohlenova K, Dumas SJ, Meta E, Conchinha NV, Taverna F, Teuwen L-A, Veys K, et al. Single-cell transcriptome atlas of murine endothelial cells.Cell. 2020; 180:764–779 e720. doi: 10.1016/j.cell.2020.01.015CrossrefMedlineGoogle Scholar
  • 40. Saharinen P, Eklund L, Alitalo K. Therapeutic targeting of the angiopoietin-TIE pathway.Nat Rev Drug Discovery. 2017; 16:635–661. doi: 10.1038/nrd.2016.278CrossrefMedlineGoogle Scholar
  • 41. Korhonen EA, Murtomaki A, Jha SK, Anisimov A, Pink A, Zhang Y, Stritt S, Liaqat I, Stanczuk L, Alderfer L, et al. Lymphangiogenesis requires Ang2/Tie/PI3K signaling for VEGFR3 cell-surface expression.J Clin Invest. 2022; 132:e155478. doi: 10.1172/JCI155478CrossrefMedlineGoogle Scholar
  • 42. Corada M, Morini MF, Dejana E. Signaling pathways in the specification of arteries and veins.Arterioscler Thromb Vasc Biol. 2014; 34:2372–2377. doi: 10.1161/ATVBAHA.114.303218LinkGoogle Scholar
  • 43. Kidoya H, Naito H, Muramatsu F, Yamakawa D, Jia W, Ikawa M, Sonobe T, Tsuchimochi H, Shirai M, Adams RH, et al. APJ regulates parallel alignment of arteries and veins in the skin.Dev Cell. 2015; 33:247–259. doi: 10.1016/j.devcel.2015.02.024CrossrefMedlineGoogle Scholar
  • 44. Fukuhara S, Sako K, Minami T, Noda K, Kim HZ, Kodama T, Shibuya M, Takakura N, Koh GY, Mochizuki N. Differential function of Tie2 at cell-cell contacts and cell-substratum contacts regulated by angiopoietin-1.Nat Cell Biol. 2008; 10:513–526. doi: 10.1038/ncb1714CrossrefMedlineGoogle Scholar
  • 45. Ho-Tin-Noe B, Boulaftali Y, Camerer E. Platelets and vascular integrity: how platelets prevent bleeding in inflammation.Blood. 2018; 131:277–288. doi: 10.1182/blood-2017-06-742676CrossrefMedlineGoogle Scholar
  • 46. Frenette PS, Denis CV, Weiss L, Jurk K, Subbarao S, Kehrel B, Hartwig JH, Vestweber D, Wagner DD. P-Selectin glycoprotein ligand 1 (PSGL-1) is expressed on platelets and can mediate platelet-endothelial interactions in vivo.J Exp Med. 2000; 191:1413–1422. doi: 10.1084/jem.191.8.1413CrossrefMedlineGoogle Scholar
  • 47. Dejana E, Tournier-Lasserve E, Weinstein BM. The control of vascular integrity by endothelial cell junctions: molecular basis and pathological implications.Dev Cell. 2009; 16:209–221. doi: 10.1016/j.devcel.2009.01.004CrossrefMedlineGoogle Scholar
  • 48. Saharinen P, Eklund L, Miettinen J, Wirkkala R, Anisimov A, Winderlich M, Nottebaum A, Vestweber D, Deutsch U, Koh GY, et al. Angiopoietins assemble distinct Tie2 signalling complexes in endothelial cell-cell and cell-matrix contacts.Nat Cell Biol. 2008; 10:527–537. doi: 10.1038/ncb1715CrossrefMedlineGoogle Scholar