Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2022 Sep 1.
Published in final edited form as: Dev Biol. 2021 May 18;477:70–84. doi: 10.1016/j.ydbio.2021.05.008

ALTERATIONS IN THE SPATIOTEMPORAL EXPRESSION OF THE CHEMOKINE RECEPTOR CXCR4 IN ENDOTHELIAL CELLS CAUSE FAILURE OF HIERARCHICAL VASCULAR BRANCHING

Wenling Li 1, Chengyu Liu 2, Nathan Burns 1, Jeffery Hayashi 1, Atsufumi Yoshida 1, Aparna Sajja 1, Sara González-Hernández 1, Ji-Liang Gao 3, Philip M Murphy 3, Yoshiaki Kubota 4, Yong-Rui Zou 5, Takashi Nagasawa 6, Yoh-suke Mukouyama 1,*
PMCID: PMC8277738  NIHMSID: NIHMS1709040  PMID: 34015362

Abstract

The C-X-C chemokine receptor CXCR4 and its ligand CXCL12 play an important role in organ-specific vascular branching morphogenesis. CXCR4 is preferentially expressed by arterial endothelial cells, and local secretion of CXCL12 determines the organotypic pattern of CXCR4+ arterial branching. Previous loss-of-function studies clearly demonstrated that CXCL12-CXCR4 signaling is necessary for proper arterial branching in the developing organs such as the skin and heart. To further understand the role of CXCL12-CXCR4 signaling in organ-specific vascular development, we generated a mouse model carrying the Cre recombinase-inducible Cxcr4 transgene. Endothelial cell- specific Cxcr4 gain-of-function embryos exhibited defective vascular remodeling and formation of a hierarchical vascular branching network in the developing skin and heart. Ectopic expression of CXCR4 in venous endothelial cells, but not in lymphatic endothelial cells, caused blood-filled, enlarged lymphatic vascular phenotypes, accompanied by edema. These data suggest that CXCR4 expression is tightly regulated in endothelial cells for appropriate vascular development in an organ-specific manner.

Keywords: CXCR4, CXCL12, gain-of-function, vascular development, lymphatic vessel development, coronary development

Graphical Abstract

graphic file with name nihms-1709040-f0001.jpg

1. Introduction

The formation of the vascular system is crucial for all aspects of tissue growth and physiology during vertebrate embryonic development. Blood vessel formation consists of two distinct processes: vasculogenesis, the process which forms the honeycomb-shaped primitive vascular plexus from angioblasts, and angiogenesis, the process which forms a hierarchically branched vascular network from pre-existing vessels with the onset of the heart beat and establishment of circulation (reviewed in (Adams and Alitalo, 2007; Herbert and Stainier, 2011; Naito et al., 2020). During the phase of angiogenesis in mice, the established vessels from the primitive capillary plexus remodel to expand and increase in diameter, resulting in a highly organized, branched vascular network, and the arteries and veins are formed by arterial-venous specification and differentiation (reviewed in (Adams and Alitalo, 2007; Herbert and Stainier, 2011; Naito et al., 2020) Subsequently, lymphatic endothelial cells (LECs) begin specification and separation from a subset of venous endothelial cells to form the whole lymphatic vascular network (reviewed in (Adams and Alitalo, 2007; Oliver and Srinivasan, 2010; Petrova and Koh, 2018). Blood and lymphatic vessels form vital networks in every organ and play major roles in tissue development and physiological functions. Therefore, developmental programs of blood and lymphatic vessel formation and their branching morphogenesis must adapt to the organ- and tissue-specific angiogenic signals (Augustin and Koh, 2017; Petrova and Koh, 2018).

Recent studies have identified the G protein-coupled receptor (GPCR) superfamily, which mediates developmental and pathological angiogenesis (reviewed in (De Francesco et al., 2017; Richard et al., 2001). Chemokine receptors are members of the GPCR superfamily and are closely involved in cell migration at the cellular level, and play a critical role in diverse physiological and pathological processes at the tissue level. CXCR4, the receptor of CXCL12 (also known as SDF1) is a homeostatic receptor that is preferentially expressed on specialized endothelial cells, termed tip cells, at the vascular front of outgrowing capillaries (Bussmann et al., 2011; del Toro et al., 2010; Strasser et al., 2010), and arterial endothelial cells during vascular development (Li et al., 2013; Tachibana et al., 1998). CXCL12-CXCR4 signaling plays a prominent role in vascular development, including organ- and tissue-specific blood and lymphatic vessel formation (Cavallero et al., 2015; Cha et al., 2012a, b; Harrison et al., 2015; Ivins et al., 2015; Li et al., 2013; Siekmann et al., 2009). Loss-of-function experiments with conventional and endothelial cell-specific Cxcr4 knockouts demonstrated defective arterial development in the developing gut (Ara et al., 2005; Tachibana et al., 1998), kidney (Takabatake et al., 2009), heart (Cavallero et al., 2015; Harrison et al., 2015; Ivins et al., 2015) and skin (Li et al., 2013), strongly suggesting that CXCR4 is necessary for arteriogenesis during vascular development. Yet, conditional gain-of-function studies on vascular development in mice have not been carried out.

In the developing skin vasculature, the model which provides a highly stereotypic and recognizable vascular branching (Mukouyama et al., 2002), CXCR4 expression was detectable in a subset of capillary endothelial cells at embryonic (E) day 13.5, the stage when the capillary network encompasses the skin (Li et al., 2013). By E15.5, vascular remodeling forms a branched, hierarchical vascular structure, and remodeled arteries align with CXCL12-expressing nerves. At this time, CXCR4 expression was restricted to the nerve-associated arteries (Li et al., 2013). Moreover, Cxcl12 and Cxcr4 mutants fail to form arterial branching and nerve-vessel alignment. Given that CXCL12 exposure did not lead to arterial differentiation while VEGF-A did, nerve-derived CXCL12 appears to function as a long-range chemokine guidance cue to recruit CXCR4-expressing endothelial cells, and nerve-derived VEGF-A promotes arterial differentiation (Li et al., 2013). Consistent with this observation, previous studies on Cxcr4a mutants in the regenerating zebrafish fin model demonstrated that CXCL12-CXCR4 signaling is important for arterial branching morphogenesis but not for arterial differentiation of endothelial cells (Xu et al., 2014).

To better understand the roles of CXCR4 in vascular development, we generated a conditional Cxcr4 gain-of-function mouse model that expresses Cxcr4 in all endothelial cells during vascular development. Our focus was on the roles of endothelial CXCR4 in the processes of vascular remodeling of the pre-existing capillary network to form a hierarchically branched vascular network. We found that endothelial cell-specific Cxcr4 gain-of-function results in abnormal vascular development such as absence of or severe defects in arterial branching and reduction in venous branching in the developing skin vasculature, as well as severe defects in coronary development. The endothelial cell- specific Cxcr4 gain-of-function mutants also show enlarged lymphatic vasculature, edema and hemorrhage-like phenotypes in the skin. However, lymphatic endothelial cell-specific Cxcr4 gain-of-function mutants have no such lymphatic abnormalities, suggesting that the lymphatic abnormalities observed in the endothelial cell-specific Cxcr4 gain-of-function mutants may indeed be secondary to the venous phenotypes. Taken together, these data suggest that there may be a window of appropriate CXCR4 expression in endothelial cell types required for proper development of the embryonic vasculature.

2. Materials and methods

2.1. Mice

All animal procedures were approved by the National Heart, Lung, and Blood Institute (NHLBI) Animal Care and Use Committee in accordance with NIH research guidelines for the care and use of laboratory animals. The following mice were used in this study: C57BL/6J mice (The Jackson Laboratory), CD-1 mice (Charles River Laboratory), Rosa26-loxP-STOP-loxP-Cxcr4-hGFP mice (ROSA-LSL-Cxcr4-hGFP, generated by the Mukouyama lab and the NHLBI Transgenic Core Facility), Cad5-BAC-CreERT2 mice (Okabe et al., 2014), Cxcl12GFP/+ knock-in mice (Ara et al., 2003), Cxcr4WHM (Balabanian et al., 2012), Prox1-BAC-Cre mice (generated by Gene Expression Nervous System Atlas (GENSAT) Project at Rockefeller University and were commercially available from MMRRC (#036644-UCD). The Cre-mediated excision was induced by administrating 3mg tamoxifen (T-5648, Sigma-Aldrich) by intraperitoneal injection (I.P.) at embryonic days (E) 10.5, 11.5 and 12.5 and embryos were harvested at E13.5 and E15.5 for analysis.

2.2. Generation of a conditional Cxcr4 gain-of-function mouse model

Mouse Cxcr4 tagged with green fluorescence protein (Cxcr4-hGFP) was obtained from MCSV-mCxcr4-hGFP vector generated by Dr. Yong-Rui Zou at the Feinstein Institute of Medicine Research (Nie et al., 2008). This Cxcr4-hGFP fragment was inserted into the downstream of CAG promoter and loxP-STOP-loxP (LSL) cassette (CAG-LSL-Cxcr4-hGFP) of TARGATT 6.1 vector (Applied StemCell Inc., #AST3050) between the PstI and Notl restriction sites. The CAG-LSL-Cxcr4-hGFP was knocked into the mouse Rosa26 locus using the TARGATT™ knock-in technology (Chen-Tsai, 2019, 2020; Tasic et al., 2011) to generate the site-specific Cxcr4 transgenic mouse line (ROSA-LSL-Cxcr4-hGFP). Briefly, the CAG-LSL-Cxcr4-hGFP DNA construct (3 μg/ml) and ΦC31 integrase mRNA (3 μg/ml) were co-microinjected into the pronuclei of zygotes collected from heterozygous Rosa26-FVB mice (obtained from Applied Stem Cell Inc.), which contained three tandem copies of the attP docking sites at the Rosa26 locus. Injected zygotes were cultured at 37°C with 6% CO2, and those embryos that reached 2-cell stage of development were implanted into the oviducts of pseudopregnant foster mothers (CD-1 mice from Charles River Laboratory). Mice born to the foster mothers were genotyped by genomic PCR analyses of tail tip DNA samples using the TARGATT™ Genotyping Kit (Applied StemCell Inc. # AST 2006) with the following three project-specific primer sets:1) Cxcr4-F: 5’-GTTGCCATGGAACCGATCAG-3’ and Cxcr4-R: 5’-TCCGTCATGCTCCTTAGCTT-3’, 2) ROSA-F: 5’-CCCAAAGTCGCTCTGAGTTGTTAT-3’ and CAG-R:5’-CATATATGGGCTATGAAACTAATGACCCCGT-3’ and 3) flox-F: 5’- AGAGCCTCTGCTAACCATGT-3’ and flox-R: CCTTTTCAGCCAGCAGTTTC-3’. The PCR condition included 34 cycles of 45 seconds at 94°C, 30 seconds at 60°C, 1 minutes at 72°C, followed by 7 minutes of terminal elongation at 72°C.

2.3. Cell culture

Human embryonic kidney 293T (ATCC CRL-3216) and MSS31 mouse endothelial cells (Yanai et al., 1991) were used in culture study. The pCAG-Cre vector was obtained from Addgene (plasmid #13775). To examine the Cre-mediated recombination efficiency and Cxcr4 transgene expression, the pCAG-LSL-Cxcr4-hGFP and pCAG-Cre vectors were transfected into 293T or MSS31 cells using X-tremeGENE kit (Roche, #06365779001) following by the manufacturer’s protocol. After 72 hours co-transfection, the cells were fixed by 4% paraformaldehyde at room temperature (RT) for 10min followed by immunostaining with rabbit anti-GFP antibody (ThermoFisher Scientific, 1:200).

2.4. Flow cytometry analysis

Transfected MSS31 cells were harvested and flow cytometry was carried out following a modified procedure from that previously reported (Mukouyama et al., 2002). Cell viability was assessed using 7-aminoactinomycin D (Thermo Fisher Scientific, A1310). Flow cytometry analysis was performed with a FACS Aria II SORP instrument (BD Bioscience).

2.5. Chemotactic migration assay

Chemotactic migration assay was performed with a 48-well modified Boyden chamber (NeuroProbe) using a polycarbonate membrane with an 8 μm pore size. Membrane was coated with 10 ng/ml fibronectin (Biomedical Technologies) overnight at 4°C. pCAG-Cre vector, pCAG-LSL-Cxcr4-hGFP vector or both transfected into serum free-293T cells were loaded in the upper chamber wells, and CXCL12 (300 ng/ml, R&D systems, #350- NS-010) recombinant was added to the lower chamber. After 5 hours of incubation at 37°C, the chamber was disassembled, and cells were removed mechanically by Q-tip from the upper side of the filter membrane. The membrane was fixed in 100% methanol, and stained with hematoxylin solution. The number of cells that migrated to the bottom side were counted, and statistical significance was assessed using the Student’s t-test.

2.6. Whole-mount immunohistochemistry

Staining was performed essentially as described previously (Li and Mukouyama, 2011; Nam et al., 2013). In brief, embryonic forelimb skin and heart were collected and fixed by 4% paraformaldehyde at 4°C overnight and after washed by PBS the next day, the samples were incubated with primary antibodies in PBS-Triton X-100 solution with 10% heat inactivated goat serum buffer at 4°C overnight. The primary antibodies used were rabbit anti-GFP antibody (ThermoFisher Scientific, 1:200) to detect CXCR4-GFP or CXCL12-GFP; rabbit anti-CXCR4 antibody (Biotrend, 1:10) to detect endogenous or ectopic CXCR4, rat anti-PECAM-1 antibody (clone MEC13.3, BD Pharmingen, 1:300) or Armenian hamster anti-CD31 (Chemicon, 1:200) to detect endothelial cells; rabbit anti- Cx40 (Alpha Diagnostic Intl. Inc., 1:200) and goat anti-neuropilin1 (Nrp1, R&D systems, 1:150) antibodies as arterial endothelial cell markers; rat anti-endomucin antibody (EMCN, Santa Cruz Biotechnology, 1:500) as venous endothelial cell marker; Cy3- conjugated anti-αSMA antibody (mouse monoclonal antibody,clone1A4, Sigma, 1:500) to detect smooth muscle cells; rabbit anti-LYVE-1 (Abcam, 1:200) or rat anti-LYVE-1 (MBL, 1:200) and rabbit anti-Prox1 (Milipore/Sigma, 1:200) antibodies to detect lymphatic endothelial cells; rat anti-Ter119-FITC antibody (ThermoFisher Scientific, 1:250) to detect erythrocytes; mouse anti-neuron specific class III β-tubulin antibody (Biolegend, Tuj1, 1:500) or anti-neurofilament (NF, Abcam, 1:1000) to detect peripheral axons, rat anti-pHH3 (Sigma,1:500) to detect proliferative cells and rabbit anti-Cleaved Caspase 3 (Cell Signaling, 1:200) to dected cell apoptosis. For immunofluorescence detection, either Cy3-, Alexa-488-, Alexa-568-, Alexa-647-, or Cy5-conjugated secondary antibodies (Jackson ImmunoResearch or ThermoFisher Scientific, 1:250, 1 hour at room temperature) were used. All stained samples were mounted with ProLong Gold Antifade Mounting solution (Thermo Fisher Scientific). All confocal microscopy was carried out on a Leica TCS SP5 confocal (Leica). The average mean fluorescence (pixel/area) and length were quantified using Fiji software. The branching points and vessel diameter were analyzed with AngioTool (Zudaire et al., 2011). Number of embryos is indicated as “n” in figure legends. Statistical significance of samples was assessed using Student’s t-test.

2.7. Section immunohistochemistry

Staining was performed essentially as described previously (Li et al., 2013). Embryos were fixed with 4% paraformaldehyde/ PBS at 4°C overnight, sunk in 30% sucrose/ PBS at 4°C and then embedded in OCT compound. Embryos were cryosectioned at 12μm thickness and collected on pre-cleaned slides (Matsunami, Japan). Staining was performed using anti-PECAM-1 antibody to detect endothelial cells; anti-endomucin (EMCN) to detect venous endothelial cells, anti-GFP to detect CXCL12-GFP in Cxcl12GFP knock-in samples, anti-LYVE-1 and anti-Prox1 antibodies to detect lymphatic endothelial cells and rat anti-Ter119-FITC antibody to detect erythrocytes. For immunofluorescent detection, either Alexa- 488-, Alexa-568-, Alexa-647 or Cy3- conjugated secondary antibodies (Jackson Immuno Research or Thermo Fisher Scientific, 1:250) were used. All confocal microscopy was carried out on a Leica TCS SP5 confocal (Leica).

3. Results

3.1. Generation of an endothelial cell-specific Cxcr4 gain-of-function mouse model

We first generated a conditional Cxcr4 gain-of-function mouse harboring loxP-STOP- loxP-Cxcr4-hGFP in the Rosa26 locus (ROSA-LSL-Cxcr4-hGFP) using the ΦC31 recombinase-mediated TARGATT technology (Fig. 1A, (Chen-Tsai, 2019). The correct knockin event was verified by the genotyping PCR (Fig. 1B). Prior to the generation of ROSA-LSL-Cxcr4-hGFP mice, we transfected pCAG-LSL-Cxcr4-hGFP and pCAG-Cre vectors into human embryonic kidney 293T cells and MSS31 mouse endothelial cells, and examined whether the STOP sequence is excised by Cre recombinase and Cxcr4- hGFP transgene is expressed in these cells (Supplemental Fig. S1). CXCR4-GFP was detectable in 293T cells by immunostaining with anti-GFP antibody (Supplemental Fig. S1A–F) and MSS31 cells by flow cytometry (Supplemental Fig. S1G–I), when both pCAG-LSL-Cxcr4-hGFP and pCAG-Cre vectors were transfected in these cells. Moreover, we confirmed the responsiveness of the cells expressing Cxcr4-hGFP transgene to CXCL12, using Boyden chamber for a chemotactic migration assay (Supplemental Fig. S1J). The forced expression of Cxcr4-hGFP transgene led to enhance cell migration in response to CXCL12 (Supplemental Fig. S1K). We then generated the endothelial cell-specific Cxcr4 gain-of-function mice using the endothelial cell-specific Cad5-BAC-CreERT2 driver mice to express Cxcr4-hGFP transgene in all endothelial cells (Fig. 1C). Because previous studies revealed that the primitive capillary network is established in the mouse forelimb by E11.5 (Mukouyama et al., 2002) and some CXCR4+ endothelial cells emerge in the skin capillary endothelial cells by E13.5 (Li et al., 2013), we decided to induce Cxcr4-Hgfp transgene in Cad5-BAC- CreERT2;ROSA-LSL-Cxcr4-hGFP embryos with tamoxifen administrations at E10.5, 11.5 and 12.5 (Fig. 1C). In this setting, ectopic CXCR4 expression was detectable in PECAM-1+ endothelial cells of the skin vasculature at E15.5: CXCR4-GFP expression was detected by whole-mount immunostaining with anti-GFP antibody (Fig. 1D versus F, E versus G; Fig. S2E versus G, F versus H, arrows) and CXCR4 expression was detected by whole-mount immunostaining with anti-CXCR4 antibody (Fig. S2A versus C, B versus D, arrows). Moreover, CXCR4-GFP expression was also detectable in PECAM-1+/endomucin (EMCN)+ venous endothelial cells and capillaries as well as PECAM-1weak/EMCN− lymphatic endothelial cells (Supplemental Movie). Combined, we successfully generated an endothelial cell-specific Cxcr4 gain-of-function mouse model to study the roles of CXCR4 on vascular development.

Fig. 1. Generation of a conditional Cxcr4 gain-of-function mouse model.

Fig. 1.

Open in a new tab

(A) Targeting strategy for knock-in of CAG-LSL-Cxcr4-hGFP into the Rosa26 locus. The hGFP-tagged Cxcr4 transgene (Cxcr4-hGFP) is preceded by a loxP-flanked transcriptional stop (loxP-STOP-loxP, LSL) cassette. In cells that express Cre recombinase, the STOP sequence is excised and Cxcr4-hGFP transgene is expressed. Dashed arrows indicate genotyping primers to identify the floxed locus. (B) Confirmation of the floxed locus (ROSA-LSL-Cxcr4-hGFP) using three sets of genotyping primers indicated in (A): the LSL cassette (~2200bp), the CAG promoter in the knockin locus (~650bp) and Cxcr4 transgene (~200bp). (C) Generation of the conditional Cxcr4 gain- of-function embryos from the timed mating of ROSA-LSL-Cxcr4-hGFP mice. Black arrows indicate time points of intraperitoneal (IP) administration of 3mg tamoxifen into pregnant mice, and red arrows indicate time points for harvest and analysis of embryos. (D-G) Whole-mount immunofluorescence analysis of limb skin from E15.5 control littermates (ROSA-LSL-Cxcr4-hGFP, D and E) and Cad5-BAC-CreERT2;ROSA-LSL- Cxcr4-hGFP embryos (F and G) with antibodies to the pan-endothelial cell marker PECAM-1 (D and F, red) and anti-GFP antibody (D and F, green; E and G, white). Activation of CreERT2 was achieved by the tamoxifen injection as indicated in (C) and embryos were harvested at E15.5. Arrows indicate representative CXCR4-hGFP- expressing endothelial cells in the embryonic skin vasculature of Cad5-BAC- CreERT2;ROSA-LSL-Cxcr4-hGFP embryos (F and G). Scale bars represent 100 μm.

3.2. Abnormal morphological appearance in endothelial cell-specific Cxcr4 gain- of-function embryos

We analyzed double transgenic embryos and control littermates at E13.5, E15.5 and E17.5 by crossing heterozygous ROSA-LSL-Cxcr4-hGFP mice with Cad5-BAC-CreERT2 (Fig. 1C). Double transgenic mutant embryos (ROSA-LSL-Cxcr4-hGFP flox/wt; Cad5- BAC-CreERT2) were viable with expected ratios (25%) at E13.5 and E15.5 (χ2 test p>0.05, Fig. 2G), but all double transgenic mutant embryos exhibited a visible hemorrhage-like phenotype at E13.5 and E15.5 (Fig. 2A versus B, C versus D, asterisks) and severe edema at E15.5 and E17.5 (Fig. 2C versus D, E versus F, arrows). Indeed, fewer than the expected number of double transgenic embryos were observed at E17.5 (χ2 test p<0.05, Fig. 2G), suggesting that endothelial cell-specific Cxcr4 gain- of-function resulted in embryonic lethality at mid- to late-gestation. Thus, we focused on studying vascular development in the endothelial cell-specific Cxcr4 gain-of-function embryos at E13.5~E15.5.

Fig. 2. Gross appearance of endothelial cell-specific Cxcr4 gain-of-function embryos.

Fig. 2.

Open in a new tab

(A-F) Morphology of control littermates (A, C, E) and Cad5-BAC-CreERT2;ROSA-LSL- Cxcr4-hGFP embryos (B, D, F) collected from the same litter and photographed at the same magnification for each time point (E13.5, E15.5, E17.5) are shown. Note that hemorrhage-like phenotype (B, D, asterisks) and edema (D, F, arrows) were found in Cad5-BAC-CreERT2;ROSA-LSL-Cxcr4-hGFP embryos. Scale bars represent 100 μm. (G) Genotype data of embryos from the timed mating of Cad5-BAC-CreERT2 x ROSA- LSL-Cxcr4-hGFP is shown. The number of control littermates and Cad5-BAC- CreERT2;ROSA-LSL-Cxcr4-hGFP double transgenic embryos at each time point are indicated inside each bar. Chi squared p values at each time points are indicated above each bar. p<0.05 in χ2 test is considered statistically significant difference from the 1:4 ratio expected and marked with asterisk (*). With the timed mating of Cad5-BAC- CreERT2 hemizygous mice with ROSA-LSL-Cxcr4-hGFP heterozygous mice, Cad5-BAC- CreERT2;ROSA-LSL-Cxcr4-hGFP embryos were viable with expected ratios (25%) at E13.5 and E15.5 but not at E17.5.

3.3. Abnormal vascular branching pattern and defective nerve-vessel alignment in the skin of endothelial cell-specific Cxcr4 gain-of-function embryos

We next investigated whether the endothelial cell-specific Cxcr4 gain-of-function disrupts proper vascular development in the skin at E13.5~E15.5. As shown previously (Li et al., 2013), the expression of CXCL12 was predominantly observed in the nerves, particularly in non-myelinated, migrating Schwann cells associated with neurofilament (NF)+ nerves (Fig. 3A and B, arrowheads), while there are a small fraction of CXCL12- expressing cells that are not associated with the nerves (Fig. 3A and B, open arrowheads). At E13.5, a honeycomb-shaped capillary network was observed in the skin of Cad5-BAC-CreERT2;ROSA-LSL-Cxcr4-hGFP embryos and control littermates (Fig. 3C versus D), and there is no significant difference in the vascular branching complexity between them (Fig. 3G). In contrast, at E15.5, the stage when the primitive capillary network has already remodeled to form a branched, hierarchical vascular network in the skin, Cad5-BAC-CreERT2;ROSA-LSL-Cxcr4-hGFP embryos displayed a decreased vascular branching complexity (Fig. 3E versus F, 3G), albeit forming small- and middle- diameter remodeled blood vessels (Fig. 3E versus F, arrows) and large-diameter blood vessels (Fig. 3E versus F, open arrows) as seen in control littermates.

Fig. 3. Abnormal vascular network in the skin vasculature of endothelial cell- specific Cxcr4 gain-of-function embryos.

Fig. 3.

Open in a new tab

(A-B) Whole-mount immunohistochemical analysis of forelimb skin from E15.5 Cxcl12GFP knock-in embryos with antibody to PECAM-1 (A, red), GFP (CXCL12-GFP, A and B, green) and neurofilament (NF, A and B, blue) are shown. Arrows indicate representative PECAM-1+ blood vessels align with NF+ nerves. Arrowheads indicate representative CXCL12-GFP+ Schwann cells associated with NF+ nerves. Open arrowheads indicate representative CXCL12-GFP+ cells not associated with NF+ nerves. Scale bars represent 100 μm. (C-F) Whole-mount immunohistochemical analysis of forelimb skin from E13.5 and E15.5 control littermates and Cad5-BAC-CreERT2;ROSA- LSL-Cxcr4-hGFP embryos with antibody to PECAM-1 is shown. Arrows indicate representative small- and middle-diameter remodeled blood vessels, open arrows indicate a representative large-diameter blood vessel, and open arrowheads indicate representative lymphatic vessels with a relatively weaker expression of PECAM-1 in the skin vasculature (E and F). Scale bars represent 100 μm. (G) Quantification of PECAM-1+ blood vessel branching points (junctions per mm2) by AngioTool at each embryonic stage (E13.5, n=3; E15.5, n=5 per genotype). Bars represent mean ± SEM. Asterisks indicate statistically significant reduction in vascular branching points (p<0.05) according to Student’s t-test.

The vascular branching abnormality affects co-patterning of blood vessel and nerve branching in Cad5-BAC-CreERT2;ROSA-LSL-Cxcr4-hGFP embryos. At 13.5, no association between nerves and blood vessels was yet evident in both Cad5-BAC- CreERT2;ROSA-LSL-Cxcr4-hGFP embryos and control littermates (Supplemental Fig. S3A versus B). At E15.5, the stage when small- and middle-diameter remodeled blood vessels associate with nerves in the skin of control littermates, nerve-vessel alignment was greatly disrupted in Cad5-BAC-CreERT2;ROSA-LSL-Cxcr4-hGFP embryos (Supplemental Fig. S3C versus D, arrows and arrowheads), leading to a ~62% reduction in nerve-vessel alignment (Supplemental Fig. S3E). Combined, the endothelial cell-specific Cxcr4 gain-of-function leads to an abnormal vascular branching network and perturbs nerve-vessel alignment in the developing skin vasculature.

3.4. Defective arterial and venous development in the skin vasculature of endothelial cell-specific Cxcr4 gain-of-function embryos

We next examined arteriogenesis in the skin vasculature of the endothelial cell-specific Cxcr4 gain-of-function embryos at E13.5 and E15.5. To assess this, we first analyzed expression of the arterial endothelial cell marker connexin 40 (CX40), which is restricted to middle-diameter remodeled arteries (Li et al., 2013; Mukouyama et al., 2005; Mukouyama et al., 2002). As such, we did not detect CX40 expression in the primitive capillary network in the skin of control littermates at E13.5 (Fig. S4A, C and E). Likewise, we did not detect any precocious CX40 expression in the skin of Cad5-BAC- CreERT2;ROSA-LSL-Cxcr4-hGFP embryos at E13.5 (Fig. S4B, D and F). To our surprise, at E15.5, CX40+ arterial branching was almost undetectable in Cad5-BAC- CreERT2;ROSA-LSL-Cxcr4-hGFP embryos (Fig. 4A and C versus B and D, arrows; 4E). This arterial defect was confirmed by the near absence of neuropilin 1 (Nrp1), which is another arterial endothelial cell marker and expressed at an earlier stage of arterial differentiation (Li et al., 2013; Mukouyama et al., 2005; Mukouyama et al., 2002) (Supplemental Fig. S5A and C versus B and D, arrows). Combined, these data suggest that the endothelial cell-specific Cxcr4 gain-of-function abolishes arteriogenesis in the developing skin vasculature.

Fig. 4. Defective arteriogenesis in the skin vasculature of endothelial cell-specific Cxcr4 gain-of-function embryos.

Fig. 4.

Open in a new tab

(A-D) Whole-mount double immunofluorescence labeling of forelimb skin with antibodies to the arterial marker Connexin 40 (CX40, green, arrows) together with PECAM-1 (red) in control littermates (A and C) and Cad5-BAC-CreERT2;ROSA-LSL- Cxcr4-hGFP embryos (B and D) at E15.5 is shown. The CX40+ remodeled arteries were nearly abolished in the skin vasculature of Cad5-BAC-CreERT2;ROSA-LSL-Cxcr4-hGFP embryos (B and D). Scale bars represent 100 μm. (E) Quantification of CX40 expression in PECAM-1+ blood vessels (n=4 per genotype). Bars represent mean ± SEM. Asterisk indicates statistically significant difference in the Cx40 expression (P<0.05) according to Student’s t-test.

The formation of remodeled veins occurs along with remodeled arteries in the developing skin vasculature (Kidoya et al., 2015; Li and Mukouyama, 2013). Using the venous and capillary endothelial cell marker endomucin (EMCN), we investigated venous development in the developing skin vasculature. At E13.5, all capillary endothelial cells are positive for EMCN in Cad5-BAC-CreERT2;ROSA-LSL-Cxcr4-hGFP embryos and control littermates (Fig. S4). At E15.5, in control littermates, EMCN+/PECAM-1+ small- and middle-diameter remodeled veins were being formed along EMCN- or weak/PECAM-1+ small- and middle-diameter remodeled arteries (Fig. 5A and B; arrows and arrowheads indicate arteries and veins, respectively). Such arteriovenous association was not detectable in Cad5-BAC-CreERT2;ROSA-LSL-Cxcr4- hGFP embryos (Fig. 5C and D; arrowheads and open arrowheads indicate veins and lymphatic vessels, respectively). In addition, Cad5-BAC-CreERT2;ROSA-LSL-Cxcr4- hGFP embryos displayed a decreased EMCN+ vascular branching complexity (Fig. 5E and F versus G, F versus H; 5I), albeit forming small- and middle-diameter remodeled EMCN+ veins (Fig. 5E and G, arrowheads) and large-diameter veins (Fig. 5E and G, open arrows) as seen in control littermates. These data suggest that the endothelial cell-specific Cxcr4 gain-of-fuRnction leads to abnormal venous branching network in the developing skin vasculature.

Fig. 5. Venous and capillary phenotypes in the skin vasculature of endothelial cell-specific Cxcr4 gain-of-function embryos.

Fig. 5.

Open in a new tab

(A-H) Whole-mount double immunofluorescence labeling of forelimb skin with antibodies to venous and capillaries markers endomucin (EMCN, green) together with PECAM-1 (red) in control littermates (A, B, E, and F) and Cad5-BAC-CreERT2;ROSA- LSL-Cxcr4-hGFP embryos (C, D, G, and H) at E15.5 is shown. The boxed region in A, C, E, G is magnified in B, D, F, H, respectively. Arrowheads indicate representative EMCN+ (PECAM-1+) small-and middle-diameter veins, open arrow indicates a representative EMCN+ large-diameter vein and arrow indicates representative EMCNweak (PECAM-1+) remodeled artery. Open arrowheads indicate EMCN− (PECAM- 1weak) lymphatic vessels. Scale bars represent 100 μm. (I) Quantification of EMCN+ venous and capillary branching points (junctions per mm2) by AngioTool (n=4 per genotype). Bars represent mean ± SEM. Asterisks indicate statistically significant reduction in vascular branching points (p<0.001) according to Student’s t-test.

Recruitment and association of vascular smooth muscle cells in remodeled blood vessels are important steps to form a branched, hierarchical vascular network (Li et al., 2013; Mukouyama et al., 2002). Indeed, no αSMA+ vascular smooth muscle coverage was observed in the primitive capillary network of both control littermates and Cad5- BAC-CreERT2;ROSA-LSL-Cxcr4-hGFP embryos at E13.5 (Fig. S4A versus B, G versus H). At E15.5, in control littermates, αSMA+ vascular smooth muscle coverage was observed in small- and middle-diameter arteries as well as large-diameter remodeled arteries and veins (Fig. 6A, C, and E; arrows and arrowheads indicate arteries and veins, respectively). As expected, arterial branching, accompanied with αSMA+ vascular smooth muscle coverage, was not detectable in the skin vasculature of Cad5-BAC- CreERT2;ROSA-LSL-Cxcr4-hGFP embryos (Fig. 6A versus B, C versus D, E versus F, arrows; 6G; Fig. S6 A versus B, C versus D, E versus F, G versus H, arrows). Interestingly, the ectopic expression of CXCR4 in endothelial cells caused an increase in αSMA+ vascular smooth muscle coverage in large-diameter veins and precocious vascular smooth muscle coverage in middle-diameter veins (Fig. 6A versus B, C versus D, E versus F; arrowheads and open arrows indicate middle- and large-diameter remodeled veins, respectively; 6G; Fig. S6A versus B, E versus F, open arrows and arrowheads). These data suggest that the endothelial cell-specific Cxcr4 gain-of- function leads to abnormal vascular smooth muscle coverage in remodeled blood vessels in the developing skin vasculature. Indeed, these findings are consistent with the recent genetic studies in zebrafish showing that the endothelial CXCL12-CXCR4- PDGFb axis regulates vascular smooth muscle cell recruitment (Stratman et al., 2020).

Fig. 6. Abnormal vascular smooth muscle coverage in small-, middle-, and large- diameter remodeled vessels in the skin vasculature of endothelial cell-specific Cxcr4 gain-of-function embryos.

Fig. 6.

Open in a new tab

(A-F) Whole-mount double immunofluorescence labeling of limb skin with antibodies to the vascular smooth muscle cell marker αSMA (A and B, green; E and F, white), together with PECAM-1 (A and B, red; C and D, white) in control littermates (A, C, and E) and Cad5-BAC-CreERT2;ROSA-LSL-Cxcr4-hGFP embryos (B, D, and F) at E15.5 is shown. Arrows indicate representative αSMA+ vascular smooth muscle cells associated with remodeled arteries, arrowheads and open arrows indicate representative αSMA+ vascular smooth muscle cells coverage associated with middle- and large-diameter remodeled veins, respectively. Pseudo colors in C and D represent remodeled arteries (red) and veins (blue). Scale bars represent 100 μm. (G) Quantification of αSMA+ vascular smooth muscle cell coverage in remodeled arteries and veins (n=4 per genotype). Bars represent mean ± SEM. Asterisk indicates statistically significant difference (*P<0.05 and ** P<0.001) according to Student’s t-test.

3.5. Blood-filled, enlarged lymphatic vascular phenotypes are not primary to Cxcr4 gain-of-function in lymphatic endothelial cells

The observation of the hemorrhage-like phenotype and severe edema in the skin of Cad5-BAC-CreERT2;ROSA-LSL-Cxcr4-hGFP embryos prompted us to investigate whether the endothelial cell-specific Cxcr4 gain-of-function disrupts lymphatic vessel development. Our whole-mount limb skin immunostaining with the erythrocyte marker Ter119 and lymphatic endothelial cell marker LYVE-1 revealed that in control littermates, Ter119+ erythrocytes were mainly detectable in PECAM-1+ (LYVE-1−) blood vessels but not in LYVE-1+ (PECAM-1weak) lymphatic vessels (whole-mount images with maximum intensity projection in Fig. 7A and C; whole-mount images with 3D rendering in Fig. S7A, C, E and I, open arrowheads). In contrast, the skin of Cad5-BAC-CreERT2;ROSA- LSL-Cxcr4-hGFP embryos displayed blood-filled, enlarged lymphatic vessels with Ter119+ erythrocyte aggregates (Fig. 7B and D; Fig. S7B, D, F and J, open arrowheads). Compared to control littermates, a significant reduction of LYVE-1+ (PECAM-1weak) lymphatic vascular branching complexity, accompanied by an increase in vessel diameter, was observed in the skin of Cad5-BAC-CreERT2;ROSA-LSL-Cxcr4-hGFP embryos (Fig. 7A, C and E versus B, D, and F, 7G and H). We next performed whole- mount limb skin immunostaining with the proliferation marker Phosphohistone H3 (pHH3) and the lymphatic endothelial cell marker Prox1 together with PECAM-1 to define lymphatic endothelial cell proliferation with the nuclear staining by these two markers. Although there was a siginicant increase in the total number of Prox1+ (PECAM-1weak) lymphatic endothelial cells in the skin of Cad5-BAC-CreERT2;ROSA- LSL-Cxcr4-hGFP embryos compared to control littermates (Fig. S8A versus B, C versus D; S8F), we did not find any significant differences in Prox1+ (PECAM-1weak) lymphatic endothelial cell proliferation between them (Fig. S8A versus B, C versus D, arrowheads; S8E). We also performed whole-mount limb skin immunostaining with the apoptosis marker Cleaved Caspase 3 together with LYVE-1 and PECAM-1 to define apoptotic lymphatic endothelial cells. We did not find any changes in LYVE-1+ (PECAM-1weak) lymphatic endothelial cell death (Fig. S8G versus H, I versus J, K versus L, arrowheads; S8M), although there was a significant increase in PECAM-1+ (LYVE-1−) blood endothelial cell death in Cad5-BAC-CreERT2;ROSA-LSL-Cxcr4-hGFP embryos (Fig. S8K versus L, arrows; S8N). These data suggest that the lymphatic phenotypes in the endothelial cell-specific Cxcr4 gain-of-function embryos may not be caused by an abnormal proliferation or cell death of lymphatic endothelial cells in the skins at E15.5.

Fig. 7. Blood-filled, enlarged lymphatic vascular phenotypes are not primary to Cxcr4 gain-of-function in lymphatic endothelial cells.

Fig. 7.

Open in a new tab

(A-F) Whole-mount triple immunofluorescence labeling of forelimb skin with antibodies to the erythrocyte marker Ter119 (A-D, green) and lymphatic endothelial cell marker LYVE-1 (A-D, red; E and F, white) together with PECAM-1(blue) in control littermates (A, C, and E) and Cad5-BAC-CreERT2;ROSA-LSL-Cxcr4-hGFP embryos (B, D, and F) at E15.5 is shown. Open arrows indicate representative Ter119+ erythrocytes and open arrowheads indicate representative LYVE-1+ (Pecam-1weak) lymphatic vessels. LYVE-1+ round-shape cells are tissue-localized macrophages. Note that Ter119+ erythrocytes (open arrows) were detected in PECAM-1+ (LYVE-1−) blood vessels in the control littermate skin vasculature (A and C), while the Cad5-BAC-CreERT2;ROSA-LSL-Cxcr4- hGFP skin shows the accumulation of Ter119+ erythrocytes (open arrows) in enlarged lymphatic vessels. Scale bars represent 100 μm. (G and H) Quantification of LYVE-1+ lymphatic vessels branching points (G, junctions per mm2) and diameter (H) by AngioTool in each genotype (n=6 per genotype). Bars represent mean ± SEM. Asterisks indicate statistically significant reduction in vascular branching points (p<0.001) according to Student’s t-test. (I-N) Whole-mount double immunofluorescence labeling of forelimb skin with antibodies to LYVE-1 (I and J, green; K and L, white; open arrowheads) and PECAM-1 (I and J, red: M and N, white; arrows) in control littermates (I, K, and M) and lymphatic endothelial cell-specific Cxcr4 gain-of-function (Prox1- Cre;ROSA-LSL-Cxcr4-hGFP) embryos (J, L, and N) at E15.5 was performed. Note that no significant abnormality in the branching and size of PECAM-1+ (LYVE-1−) blood vasculature and LYVE-1+ (PECAM-1weak) lymphatic vasculature was detected in the skin of Prox1-Cre;ROSA-LSL-Cxcr4-hGFP embryos. Scale bars represent 100 μm. (O-P) Quantification of LYVE-1+ lymphatic vessels branching points (O, junctions per mm2) and diameter (P) by AngioTool in each genotype (n=5 per genotype). Bars represent mean ± SEM.

Next we examined whether the endothelial cell-specific Cxcr4 gain-of-function affects the formation of lymph sac, from which most of the lymphatic vasculature is derived (Yang and Oliver, 2014). Interestingly, we found blood-filled, enlarged lymph sacs in the trunk section of Cad5-BAC-CreERT2;ROSA-LSL-Cxcr4-hGFP embryos at E13.5, the stage when lymphatic endothelial cells migrate away from the jugular veins to form intact lymph sacs (Fig. 8A versus B, C versus D). Ectopic expression of the venous endothelial cell marker EMCN, together with the expression of the lymphatic markers Proxl and LYVE-1, albeit with a relatively lesser level, was observed in the lymph sac of Cad5-BAC-CreERT2;ROSA-LSL-Cxcr4-hGFP embryos (Fig. 8E versus F and F’, I versus J and J’, arrowheads). These data suggest a potential failure of lymph sacs to separate from jugular veins in the endothelial cell-specific Cxcr4 gain-of-function embryos.

Fig. 8. Blood-filled lymph sac in endothelial cell-specific Cxcr4 gain-of-function embryos.

Fig. 8.

Open in a new tab

(A-D) Triple immunofluorescence confocal microscopy of E13.5 trunk sections of control littermates (A and C) and Cad5-BAC-CreERT2;ROSA-LSL-Cxcr4-hGFP embryos (B and D) with antibodies for LYVE-1 (green), Ter119 (red), and PECAM-1 (A and B, blue; C and D, white). Blood-filled, enlarged lymph sac was observed in Cad5-BAC- CreERT2;ROSA-LSL-Cxcr4-hGFP embryos. (E-H) Triple immunofluorescence confocal microscopy with antibodies to LYVE-1 (E and F, green), EMCN (E and F, red) and PECAM-1 (G and H, white). The boxed region in F is magnified in F’. Arrowheads indicate representative LYVE-1+/EMCN+ lymphatic endothelial cells in the lymph sac of Cad5-BAC-CreERT2;ROSA-LSL-Cxcr4-hGFP embryos. Pseudo colors in C, D, G and H represent lymph sac (LS, green) and jugular vein (JV, red). Scale bars represent 100 μm. (I-L) Triple immunofluorescence confocal microscopy of E13.5 trunk sections of control littermates (I and K) and Cad5-BAC-CreERT2;ROSA-LSL-Cxcr4-hGFP embryos (J and L) with antibodies for Prox1 (green), EMCN (red), and PECAM-1 (K and L, white). Arrowheads indicate representative Prox1+/EMCN+ endothelial cells in lymph sac of Cad5-BAC-CreERT2;ROSA-LSL-Cxcr4-hGFP embryos (J’). The rpresentative images are shown in each genotype (n=3 per genotype). The boxed region in J is magnified in J’. Scale bars represent 100 μm. LS: lymph sac, JV: jugular vein, CA: carotid artery, SCG: superior cervical ganglion, NT: notochord.

We were interested in determining whether the overexpression of CXCR4 in lymphatic endothelial cells could cause the blood-filled, enlarged lymphatic vascular phenotypes. To address this, we generated a lymphatic endothelial cell-specific Cxcr4 gain-of- function mouse using a lymphatic endothelial cell-specific Prox1-Cre driver mouse to express Cxcr4-hGFP transgene in lymphatic endothelial cells (Supplemental Fig. S9). No visible abnormality in the gross morphology and embryonic size was observed in Prox-1-Cre;ROSA-LSL-Cxcr4-hGFP embryos (Supplemental Fig. S9A versus B). No significant hemorrhage-like phenotype along with edema was detected (Supplemental Fig. S9A versus B). In this setting, CXCR4-GFP expression was detectable in LYVE-1 + lymphatic endothelial cells of the skin vasculature at E15.5 (Supplemental Fig. S9C versus D, E versus F, G versus H, open arrows and open arrowheads). Further whole- mount limb skin immunostaining revealed no significant difference in lymphatic and blood vascular branching morphogenesis and vessel diameter between Prox-1- Cre;ROSA-LSL-Cxcr4-hGFP and control littermates (Fig. 7I versus J, K versus L, M versus N, arrows and open arrowheads; 7O and P). These data suggest that the blood- filled, enlarged lymphatic vascular phenotypes are not primary to Cxcr4 gain-of-function in lymphatic endothelial cells.

3.6. Defective coronary vascular development in endothelial cell-specific Cxcr4 gain-of-function embryos

Previous studies demonstrated that CXCR4 expression is restricted to endothelial cells of coronary vasculature in the myocardium of the ventricles and endothelial CXCR4 is required for coronary development, especially coronary artery formation in response to CXCL12 (Cavallero et al., 2015; Ivins et al., 2015). Indeed, CXCL12-GFP expression was detectable in the ventricular epicardium and myocardium (Supplemental Fig. S10A). We also found CXCL12-GFP expressing cells in close proximity to PECAM-1+ coronary vasculature (Supplemental Fig. S10B, arrows and arrowheads). We next investigated whether the endothelial cell-specific Cxcr4 gain-of-function disrupts coronary development. Our whole-mount heart ventricle immunostaining visualized a branched, hierarchical venous network in the subepicardial layer of the dorsal and ventral ventricular wall and arterial branching in the myocardial layer of the dorsal and ventral ventricular wall (Nam et al., 2013). At E15.5, in control littermates, the subepicardium of the dorsal and ventral ventricular surface displayed a highly branched EMCN+/PECAM- 1+ venous network including three large-diameter venous branches in the dorsal surface (Fig. 9A and E, arrowheads) and two large-diameter venous branches near the aorta in the ventral surface (Fig. 9I and M, arrowheads). The myocardium of the dorsal and ventral ventricular wall displayed branched EMCN−/PECAM-1+ capillaries and large- diameter arterial branches (Fig. 9C and G, arrows) including proximal coronary arteries from the aorta (Fig. 9K and O, arrows). In Cad5-BAC-CreERT2;ROSA-LSL-Cxcr4-hGFP embryos, the coronary vasculature was disorganized relative to the stereotype pattern found in control littermates: remodeled large-diameter venous and arterial branches were almost undetectable (Fig. 9A versus B, E versus F, coronary veins in the dorsal subepicardium; I versus J, M versus N, coronary veins in the ventral subepicardium; C versus D, G versus H, coronary arteries in the dorsal myocardium; K versus L, O versus P, coronary arteries in the ventral myocardium). Additionally, we analyzed cardiac lymphatic vascular formation by whole-mount heart ventricle immunostaining. At E15.5, a highly branched LYVE-1+ (PECAM-1weak) lympthatic network was developed in control littermates (Fig. S10C, D, G and H, arrowheads). In contrast, abnormal LYVE-1 + (PECAM-1weak) lymphatic vessel branching was observed in the subepicardium of the dorsal ventricular surface of Cad5-BAC-CreERT2;ROSA-LSL-Cxcr4-hGFP heart. Combined, these data suggest the endothelial cell-specific Cxcr4 gain-of-function leads to failure of coronary vascular remodeling for the formation of a hierarchical coronary vascular network and abnormal lymphatic vasculature in the ventricular wall of the developing heart.

Fig. 9. Defective coronary vascular development in the heart of endothelial cell- specific Cxcr4 gain-of-function embryos.

Fig. 9.

Open in a new tab

(A-H) Whole-mount immunofluorescence confocal microscopy of E15.5 heart was performed with antibodies to PECAM-1 (green or white) and EMCN (red) to visualize coronary veins in the dorsal subepicardium and arteries in the dorsal myocardium of the ventricular wall in control littermates (A and E, subepicardium; C and G, myocardium) or Cad5-BAC-CreERT2;ROSA-LSL-Cxcr4-hGFP embryos (B and F, subepicardium; D and H, myocardium). EMCN+/PECAM-1+ large-diameter venous branches were observed in the dorsal subepicardium of control littermates (A and E, arrowheads) but not in Cad5- BAC-CreERT2;ROSA-LSL-Cxcr4-h GFP embryos (B and F). EMCN−/PECAM-1+ remodeled arterial branches were observed in the dorsal myocardium of the ventricular wall in control littermates (C and G, arrows) but were barely observed in Cad5-BAC- CreERT2;ROSA-LSL-Cxcr4-hGFP embryos (D and H, arrows). (I-P) Whole-mount immunofluorescence confocal microscopy of E15.5 heart to visualize coronary veins and arteries in the ventral subepicardium and myocardium of the ventricular wall in control littermates (I and M, subepicardium; K and O, myocardium) or Cad5-BAC- CreERT2;ROSA-LSL-Cxcr4-hGFP embryos (J and N, subepicardium; L and P, myocardium). EMCN+/PECAM-1+ venous branches and EMCN−/PECAM-1+ large- diameter arterial branches were observed in the ventral subepicardium and myocardium of the ventricular wall in control littermates (I and M, arrowheads; K and O, arrows) but were barely observed in Cad5-BAC-CreERT2;ROSA-LSL-Cxcr4-hGFP embryos (J, N, L, P). The rpresentative images are shown in each genotype (n=3 per genotype). LV, left ventricle; RV, right ventricle; DA, dorsal aorta. Scale bars represent 200 μm.

4. Discussion

In this study, we examined vascular development in the developing skin and coronary vasculature of the endothelial cell-specific Cxcr4 gain-of-function mutants. Cxcr4 transgene was induced in endothelial cells during primitive capillary network formation and we focused our analysis on the process of vascular remodeling of the pre-existing capillary network to form a hierarchically branched vascular network with appropriate arterial and venous branching and vascular smooth muscle coverage. The endothelial cell-specific Cxcr4 gain-of-function resulted in abnormal vascular development such as the absence of or severe defects in arterial branching and reductions in venous branching in the developing skin, as well as severe defects in coronary development. Given a dependence on CXCL12-CXCR4 signaling for nerve-vessel alignment and arterial branching in the skin vasculature (Li et al., 2013), the endothelial cell-specific Cxcr4 gain-of-function perturbed nerve-vessel alignment and abolished arteriogenesis. Blood-filled, enlarged lymphatic malformations were caused by a potential failure of lymph sacs to separate from jugular veins in the endothelial cell-specific Cxcr4 gain-of- function embryos. Taken together, these vascular phenotypes in the endothelial cell- specific Cxcr4 gain-of-function suggest that appropriate CXCR4 expression in the appropriate endothelial cell type is required for tissue-specific vascular network formation

Previous studies demonstrated that loss-of-function with conventional and endothelial cell-specific Cxcr4 knockouts perturbed nerve-vessel alignment and abolished arteriogenesis in the developing skin (Li et al., 2013). Given that CXCL12 is preferentially expressed by sensory nerves and Cxcl12 mutants exhibit an almost identical phenotype in cardiovascular development, CXCL12-CXCR4 signaling is necessary for sensory nerve-mediated arterial branching in the skin vasculature. A subset of endothelial cells in the primitive capillary network express CXCR4 and nerve- derived CXCL12 selectively recruits CXCR4+ endothelial cells to form remodeled blood vessels aligned with nerves and undergo arteriogenesis (Li et al., 2013). Interestingly, the forced expression of CXCR4 in all endothelial cells of the primitive capillary network results in a similar vascular phenotype in the developing skin. Despite the significant importance of CXCL12-CXCR4 signaling in vascular development, the similar vascular phenotypes observed in both Cxcr4 loss- and gain-of-function in endothelial cells suggest that there may be a window of appropriate CXCL12-CXCR4 signaling for the sensory nerve-mediated arterial branching.

Why do the Cxcr4 gain-of-function mutants fail to develop proper nerve-vessel alignment and arteriogenesis in the skin? In wild-type embryos, CXCR4+ endothelial cells in the primitive capillary network can be pre-specified for association with nerves. During the vascular remodeling of the primitive capillary network to form the hierarchical vascular network, nerve-derived CXCL12 recruits CXCR4+ endothelial cells to form vessels to align with the nerves. Then, nerve-derived VEGF-A promotes arterial differentiation (Li et al., 2013). Indeed, CXCL12 exposure did not lead to arterial differentiation while VEGF-A did (Li et al., 2013). Vascular smooth muscle cells associate with branched, remodeled blood vessels only after such vessels have become aligned with nerves and begun to express arterial markers (Li et al., 2013; Mukouyama et al., 2005; Mukouyama et al., 2002). Therefore, arteriogenesis – arterial differentiation of endothelial cells and association of vascular smooth muscle cells – is initiated in branched, remodeled blood vessels only after they associate with nerves to form nerve-vessel alignment. One explanation for nerve-vessel alignment defects found in the endothelial cell-specific Cxcr4 gain-of-function mutants is as a result of competition for limiting amounts of nerve-derived CXCL12 in the developing skin vasculature: ectopic expression of CXCR4 may inhibit nerve-mediated endothelial cell migration touring to nerves by CXCL12 sequestration. Failure of nerve-vessel alignment results in defective arteriogenesis. Another explanation is that the increased levels of CXCR4, a member of the G protein-coupled receptors family, may enhance G-protein- dependent intracellular signaling such as the mitogen-activated protein kinase, JNK, p38, and phosphoinositide 3-kinase pathways, resulting in dysregulation of CXCL12- mediated directional migration and VEGF-A-mediated arterial differentiation. Given that different endothelial cell cycle states affect arteriovenous specification (Fang et al., 2017; Luo et al., 2021), the enhanced G-protein-dependent intracellular signaling might change endothelial cell cycle states, resulting in the inhibition of arterial differentiation. However, no vascular phenotype was detectable in constitutive active mutations of Cxcr4 in the immunodeficiency disease, Warts, Hypogammaglobulinemia, recurrent Infections and Myelokathexis (WHIM) syndrome (Hernandez et al., 2003). WHIM mutations of Cxcr4 (Cxcr4WHIM) increase the signaling because they disrupt the receptor’s negative regulatory elements in the carboxy-terminus (reviewed in (McDermott and Murphy, 2019). We did not find any significant change in arterial differentiation (no precocious arterial marker expression at E13.5 and proper arterial branching at E15.5), vascular smooth muscle coverage (proper vascular smooth muscle coverage in arteries at E15.5), as well as nerve-vessel alignment in the skin vasculature of Cxcr4WHIM embryos (Fig. S11). These data suggest that the increased level of CXCR4 in arterial endothelial cells is unlikely to lead to abnormal vascular phenotypes in the skin.

The endothelial cell- and lymphatic endothelial cell-specific Cxcr4 gain-of-function experiments demonstrated that blood-filled, enlarged lymphatic vascular phenotypes are not primary to the Cxcr4 gain-of-function in lymphatic endothelial cells: Ectopic CXCR4 expression in jugular veins but not lymph sac and lymphatic vasculature may cause a failure of lymph sac to separate from jugular vein. A series of seminal studies clearly demonstrated that Prox1+ lymphatic endothelial cell precursors, stimulated by VEGF-C in the surrounding mesenchyme, then delaminate from the anterior cardinal veins and migrate along a specific dorsolateral pathway to form the primary lymph sacs, which lie in close proximity to the cardinal veins (Yang and Oliver, 2014). One potential explanation for the failure in lymph sac separation from jugular vein in the endothelial cell-specific Cxcr4 gain-of-function mutants is as a result of aberrant delamination and migration of Prox1+ lymphatic endothelial cell precursors: ectopic expression of CXCR4 in venous endothelial cells of the jugular vein may affect VEGF-C-mediated delamination and directional migration, accompanied with attenuated expression of the lymohatic marker Prox-1 and LYVE-1 and aberrant expression of the venous marker endomucin (EMCN). The failure in lymph sac separation from jugular vein causes blood-filled lymph sac. Indeed, mutations in two genes, PKD1 (polycystin-1) and PKD2 (polycystin-2), co-receptor for WNT proteins coupled to Ca signaling (Kim et al., 2016), cause an intrinsic defect in directional migration of lymphatic endothelial cells, resulting in blood-filled lymph sac formation (Outeda et al., 2014). Endothelial cell specific Rac1 deficiency impairs budding and directional migration of lymphatic endothelial cells, resulting in abnormal lymphatic-blood vessel separation and blood- filled lymph sac formation (D’Amico et al., 2009).

Interestingly, CXCR7, a scavenger or decoy receptor for CXCL12 and adrenomedullin, is expressed by both jugular vein and lymph sac and loss of Cxcr7 results in similar lymphatic phenotypes in the skin and lymph sac (Klein et al., 2014). The lymphatic phenotypes in Cxcr7 loss-of-function was found to be caused by excessive proliferative signaling from adrenomedullin, a non-chemokine mitogenic peptide which is critical for lymphatic vascular development (Klein et al., 2014). Given that CXCR7 heterodimerizes with CXCR4 to regulate CXCL12 availability (Quinn et al., 2018), ectopic CXCR4 expression may induce the CXCR4/CXCR7 heterodimer interacting with CXCL12 and reduce CXCR7-adrenomedullin interaction in jugular veins, resulting in the above- mentioned lymphatic abnormalities. On the other hand, genetic studies of Cxcl12 and Cxcr4 loss-of-function and Cxcl12 gain-of-function experiments in zebrafish demonstrated that CXCL12-CXCR4 signaling controls lymphatic endothelial cell migration to form trunk lymphatic vessels (Cha et al., 2012a). Accordingly, dynamic changes in the expression of CXCL12 and CXCR4 appear to regulate the stepwise growth of trunk lymphatic vessels. However, we did not observe any significant lymphatic vessel abnormality in the lymphatic endothelial cell-specific Cxcr4 gain-of- function mutants. Although the previous studies in zebrafish did not carry out a comparable experiment such as lymphatic endothelial cell-specific Cxcr4 gain-of- function designed for similar purposes, the increased level of CXCR4 in lymphatic endothelial cells may not be sufficient to promote lymphatic vessel development.

Murine coronary vessels develop in two distinct anatomical components: coronary veins develop on the ventricular surface (subepicardial layer), while the coronary arteries develop in the myocardial layer. Sinus venosus-derived endothelial cells migrate caudally over the dorsal side of the heart ventricle and penetrate the ventricules from the outside towards the inside, giving rise to both coronary arterial and venous endothelial cells, whereas endocardium-derived endothelial cells sprout throughout the interventricular septum onto the ventral surface and into the lateral walls towards the apex (reviewed in Lupu et al., 2020). Proper connection of coronary plexus to coronary sinus or aorta marks the onset of vascular remodeling to form the hierarchical coronary network. The venous plexus form the connections with the coronary sinus, driven by semaphorin 3D-ErbB2/Neuropilin-1 signaling, allowing for blood circulation through the coronary vasculature and triggering the vascular remodeling to form the hierarchical coronary venous network (Aghajanian et al., 2016). In coronary artery development, CXCL12-CXCR4 signaling (Cavallero et al., 2015; Harrison et al., 2015; Ivins et al., 2015) and VEGF-C/VEGFR3 signaling (Chen et al., 2014) are important for polarization and directional migration of plexus endothelial cell during anastomosis of coronary plexus to the aorta. During the vascular remodeling of coronary arterial plexus, CXCL12-CXCR4 signaling controls endothelial cell migration against blood flow into remodeling arteries (Chang et al., 2017). As CXCR4 is preferentially expressed by endothelial cells of the coronary vasculature in the ventricular myocardium, which are fated to form capillaries and arteries, and CXCL12 is mainly expressed by the epicardium and myocardium, Cxcl12 and Cxcr4 loss-of-function in mice exhibits normal assembly of higher order of coronary vein structures in the ventricular subepicardium but failure of vascular remodeling of the coronary plexus to form large-diameter coronary arteries in the ventricular myocardium (Cavallero et al., 2015; Harrison et al., 2015; Ivins et al., 2015). Interestingly, the endothelial cell-specific Cxcr4 gain-of-function embryos fail to undergo the vascular remodeling of coronary plexus to form remodeled large-diameter coronary arteries and veins. The observation that both Cxcr4 loss-of- function and gain-of-function mutants form coronary plexus suggest that CXCL12- CXCR4 signaling is not indispensable for appropriate migration of coronary endothelial progenitor populations which arise from the sinus venosus and endocardium, and migrate into the subepicardium and myocardium to contribute to coronary plexus (reviewed in (Lupu et al., 2020). One explanation for the failure of hierarchical coronary network formation in the endothelial cell-specific Cxcr4 gain-of-function mutants is that the increased level of CXCR4 may cause abnormal polarization and migration of plexus endothelial cells through the enhanced G-protein-dependent intracellular signaling during the coronary plexus anastomoses with the aorta and sinus and subsequently the vascular remodeling of the plexus into the hierarchical vascular network. At the mechanistic level, whether the endothelial cell-specific Cxcr4 gain-of-function affects semaphorin 3D-ErbB2/Neuropilin-1 signaling in coronary vein development and VEGF- C/VEGFR3 signaling in coronary artery development remains to be investigated.

The data presented here demonstrate that the endothelial cell-specific Cxcr4 gain-of- function causes abnormal vascular development in the skin and heart. Given that endothelial cell-specific Cxcr4 loss- and gain-of-function exhibits almost identical vascular phenotypes, CXCR4 expression could be tightly regulated in endothelial cells for appropriate vascular development. Our studies also demonstrate the potential uses of the conditional Cxcr4 gain-of-function in understanding cell type-specific CXCR4 functions at different developmental stages and pathological conditions such as ischemic heart disease (Das et al., 2019) in the adult, as well as abnormal origins of the coronary arteries in infants and children.

Supplementary Material

1
Download video file (30MB, mp4)
2
3
Download video file (30MB, mp4)

Highlights:

  • Endothelial cell-specific Cxcr4 gain-of-function embryos exhibit abnormal vascular development

  • Endothelial cell-specific Cxcr4 gain-of-function embryos exhibit abnormal coronary development

  • Ectopic Cxcr4 expression in blood endothelial cells causes lymphatic vascular phenotypes

Acknowledgements

ROSA-LSL-Cxcr4-hGFP mice were generated in the NHLBI Transgenic Core Facility. Cxcr4WHIM mice were generated by the Bachelerie laboratory (INSERM). Thanks to P. Dagur, and P. J. McCoy for FACS assistance, C. Combs for assistance with image processing, M. Lindhurst (NHGRI) for χ2 test assistance and J. Hawkins and the staff of NIH Bldg50 animal facility for assistance with mouse breeding and care; K. Gill for laboratory management and technical support and V. Sam for administrative assistance. Thanks also to R. Adelstein, L. Leatherbury, W. Kowalski for editorial advice and discussion, and members of Laboratory of Stem Cell and Neuro-Vascular Biology for technical help and thoughtful discussion. None of the authors has any financial or other conflicts of interest. This work was supported by the Intramural Research Program of the National Heart, Lung, and Blood Institute, National Institutes of Health (HL005702-15 and HL006116-10).

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

References

  1. Adams RH, Alitalo K, 2007. Molecular regulation of angiogenesis and lymphangiogenesis. Nat Rev Mol Cell Biol 8, 464–478. [DOI] [PubMed] [Google Scholar]
  2. Aghajanian H, Cho YK, Manderfield LJ, Herling MR, Gupta M, Ho VC, Li L, Degenhardt K, Aharonov A, Tzahor E, Epstein JA, 2016. Coronary vasculature patterning requires a novel endothelial ErbB2 holoreceptor. Nat Commun 7, 12038. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Ara T, Tokoyoda K, Okamoto R, Koni PA, Nagasawa T, 2005. The role of CXCL12 in the organ-specific process of artery formation. Blood 105, 3155–3161. [DOI] [PubMed] [Google Scholar]
  4. Ara T, Tokoyoda K, Sugiyama T, Egawa T, Kawabata K, Nagasawa T, 2003. Long-term hematopoietic stem cells require stromal cell-derived factor-1 for colonizing bone marrow during ontogeny. Immunity 19, 257–267. [DOI] [PubMed] [Google Scholar]
  5. Augustin HG, Koh GY, 2017. Organotypic vasculature: From descriptive heterogeneity to functional pathophysiology. Science 357. [DOI] [PubMed] [Google Scholar]
  6. Balabanian K, Brotin E, Biajoux V, Bouchet-Delbos L, Lainey E, Fenneteau O, Bonnet D, Fiette L, Emilie D, Bachelerie F, 2012. Proper desensitization of CXCR4 is required for lymphocyte development and peripheral compartmentalization in mice. Blood 119, 5722–5730. [DOI] [PubMed] [Google Scholar]
  7. Bussmann J, Wolfe SA, Siekmann AF, 2011. Arterial-venous network formation during brain vascularization involves hemodynamic regulation of chemokine signaling. Development 138, 1717–1726. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Cavallero S, Shen H, Yi C, Lien CL, Kumar SR, Sucov HM, 2015. CXCL12 Signaling Is Essential for Maturation of the Ventricular Coronary Endothelial Plexus and Establishment of Functional Coronary Circulation. Dev Cell 33, 469–477. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Cha YR, Fujita M, Butler M, Isogai S, Kochhan E, Siekmann AF, Weinstein BM, 2012a. Chemokine signaling directs trunk lymphatic network formation along the preexisting blood vasculature. Dev Cell 22, 824–836. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Cha YR, Fujita M, Butler M, Isogai S, Kochhan E, Siekmann AF, Weinstein BM, 2012b. Chemokine signaling directs trunk lymphatic network formation along the preexisting blood vasculature. Dev. Cell 22, 824–836. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Chang AH, Raftrey BC, D’Amato G, Surya VN, Poduri A, Chen HI, Goldstone AB, Woo J, Fuller GG, Dunn AR, Red-Horse K, 2017. DACH1 stimulates shear stress-guided endothelial cell migration and coronary artery growth through the CXCL12-CXCR4 signaling axis. Genes Dev 31, 1308–1324. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Chen HI, Poduri A, Numi H, Kivela R, Saharinen P, McKay AS, Raftrey B, Churko J, Tian X, Zhou B, Wu JC, Alitalo K, Red-Horse K, 2014. VEGF-C and aortic cardiomyocytes guide coronary artery stem development. J Clin Invest 124, 4899–4914. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Chen-Tsai RY, 2019. Using TARGATT Technology to Generate Site-Specific Transgenic Mice. Methods Mol Biol 1874, 71–86. [DOI] [PubMed] [Google Scholar]
  14. Chen-Tsai RY, 2020. Integrase-Mediated Targeted Transgenics Through Pronuclear Microinjection. Methods Mol Biol 2066, 35–46. [DOI] [PubMed] [Google Scholar]
  15. D’Amico G, Jones DT, Nye E, Sapienza K, Ramjuan AR, Reynolds LE, Robinson SD, Kostourou V, Martinez D, Aubyn D, Grose R, Thomas GJ, Spencer-Dene B, Zicha D, Davies D, Tybulewicz V, Hodivala-Dilke KM, 2009. Regulation of lymphatic-blood vessel separation by endothelial Rac1. Development 136, 4043–4053. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Das S, Goldstone AB, Wang H, Farry J, D’Amato G, Paulsen MJ, Eskandari A, Hironaka CE, Phansalkar R, Sharma B, Rhee S, Shamskhou EA, Agalliu D, de Jesus Perez V, Woo YJ, Red-Horse K, 2019. A Unique Collateral Artery Development Program Promotes Neonatal Heart Regeneration. Cell 176, 1128–1142 e1118. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. De Francesco EM, Sotgia F, Clarke RB, Lisanti MP, Maggiolini M, 2017. G Protein-Coupled Receptors at the Crossroad between Physiologic and Pathologic Angiogenesis: Old Paradigms and Emerging Concepts. Int J Mol Sci 18. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. del Toro R, Prahst C, Mathivet T, Siegfried G, Kaminker JS, Larrivee B, Breant C, Duarte A, Takakura N, Fukamizu A, Penninger J, Eichmann A, 2010. Identification and functional analysis of endothelial tip cell-enriched genes. Blood 116, 4025–4033. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Fang JS, Coon BG, Gillis N, Chen Z, Qiu J, Chittenden TW, Burt JM, Schwartz MA, Hirschi KK, 2017. Shear-induced Notch-Cx37-p27 axis arrests endothelial cell cycle to enable arterial specification. Nat Commun 8, 2149. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Harrison MR, Bussmann J, Huang Y, Zhao L, Osorio A, Burns CG, Burns CE, Sucov HM, Siekmann AF, Lien CL, 2015. Chemokine-guided angiogenesis directs coronary vasculature formation in zebrafish. Dev Cell 33, 442–454. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Herbert SP, Stainier DY, 2011. Molecular control of endothelial cell behaviour during blood vessel morphogenesis. Nat Rev Mol Cell Biol 12, 551–564. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Hernandez PA, Gorlin RJ, Lukens JN, Taniuchi S, Bohinjec J, Francois F, Klotman ME, Diaz GA, 2003. Mutations in the chemokine receptor gene CXCR4 are associated with WHIM syndrome, a combined immunodeficiency disease. Nat Genet 34, 70–74. [DOI] [PubMed] [Google Scholar]
  23. Ivins S, Chappell J, Vernay B, Suntharalingham J, Martineau A, Mohun TJ, Scambler PJ, 2015. The CXCL12/CXCR4 Axis Plays a Critical Role in Coronary Artery Development. Dev Cell 33, 455–468. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Kidoya H, Naito H, Muramatsu F, Yamakawa D, Jia W, Ikawa M, Sonobe T, Tsuchimochi H, Shirai M, Adams RH, Fukamizu A, Takakura N, 2015. APJ Regulates Parallel Alignment of Arteries and Veins in the Skin. Dev Cell 33, 247–259. [DOI] [PubMed] [Google Scholar]
  25. Kim S, Nie H, Nesin V, Tran U, Outeda P, Bai CX, Keeling J, Maskey D, Watnick T, Wessely O, Tsiokas L, 2016. The polycystin complex mediates Wnt/Ca(2+) signalling. Nat Cell Biol 18, 752–764. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Klein KR, Karpinich NO, Espenschied ST, Willcockson HH, Dunworth WP, Hoopes SL, Kushner EJ, Bautch VL, Caron KM, 2014. Decoy receptor CXCR7 modulates adrenomedullin-mediated cardiac and lymphatic vascular development. Dev Cell 30, 528–540. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Li W, Kohara H, Uchida Y, James JM, Soneji K, Cronshaw DG, Zou YR, Nagasawa T, Mukouyama YS, 2013. Peripheral nerve-derived CXCL12 and VEGF-A regulate the patterning of arterial vessel branching in developing limb skin. Dev Cell 24, 359–371. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Li W, Mukouyama YS, 2011. Whole-mount immunohistochemical analysis for embryonic limb skin vasculature: a model system to study vascular branching morphogenesis in embryo. J Vis Exp. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Li W, Mukouyama YS, 2013. Tissue-specific venous expression of the EPH family receptor EphB1 in the skin vasculature. Dev Dyn 242, 976–988. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Luo W, Garcia-Gonzalez I, Fernandez-Chacon M, Casquero-Garcia V, Sanchez- Munoz MS, Muhleder S, Garcia-Ortega L, Andrade J, Potente M, Benedito R, 2021. Arterialization requires the timely suppression of cell growth. Nature 589, 437–441. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Lupu IE, De Val S, Smart N, 2020. Coronary vessel formation in development and disease: mechanisms and insights for therapy. Nat Rev Cardiol 17, 790–806. [DOI] [PubMed] [Google Scholar]
  32. McDermott DH, Murphy PM, 2019. WHIM syndrome: Immunopathogenesis, treatment and cure strategies. Immunol Rev 287, 91–102. [DOI] [PubMed] [Google Scholar]
  33. Mukouyama YS, Gerber HP, Ferrara N, Gu C, Anderson DJ, 2005. Peripheral nerve-derived VEGF promotes arterial differentiation via neuropilin 1-mediated positive feedback. Development 132, 941–952. [DOI] [PubMed] [Google Scholar]
  34. Mukouyama YS, Shin D, Britsch S, Taniguchi M, Anderson DJ, 2002. Sensory nerves determine the pattern of arterial differentiation and blood vessel branching in the skin. Cell 109, 693–705. [DOI] [PubMed] [Google Scholar]
  35. Naito H, Iba T, Takakura N, 2020. Mechanisms of new blood-vessel formation and proliferative heterogeneity of endothelial cells. Int Immunol 32, 295–305. [DOI] [PubMed] [Google Scholar]
  36. Nam J, Onitsuka I, Hatch J, Uchida Y, Ray S, Huang S, Li W, Zang H, Ruiz- Lozano P, Mukouyama YS, 2013. Coronary veins determine the pattern of sympathetic innervation in the developing heart. Development 140, 1475–1485. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Nie Y, Han YC, Zou YR, 2008. CXCR4 is required for the quiescence of primitive hematopoietic cells. J Exp Med 205, 777–783. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Okabe K, Kobayashi S, Yamada T, Kurihara T, Tai-Nagara I, Miyamoto T, Mukouyama YS, Sato TN, Suda T, Ema M, Kubota Y, 2014. Neurons limit angiogenesis by titrating VEGF in retina. Cell 159, 584–596. [DOI] [PubMed] [Google Scholar]
  39. Oliver G, Srinivasan RS, 2010. Endothelial cell plasticity: how to become and remain a lymphatic endothelial cell. Development 137, 363–372. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Outeda P, Huso DL, Fisher SA, Halushka MK, Kim H, Qian F, Germino GG, Watnick T, 2014. Polycystin signaling is required for directed endothelial cell migration and lymphatic development. Cell Rep 7, 634–644. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Petrova TV, Koh GY, 2018. Organ-specific lymphatic vasculature: From development to pathophysiology. J Exp Med 215, 35–49. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Quinn KE, Mackie DI, Caron KM, 2018. Emerging roles of atypical chemokine receptor 3 (ACKR3) in normal development and physiology. Cytokine 109, 17–23. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Richard DE, Vouret-Craviari V, Pouyssegur J, 2001. Angiogenesis and G-protein- coupled receptors: signals that bridge the gap. Oncogene 20, 1556–1562. [DOI] [PubMed] [Google Scholar]
  44. Siekmann AF, Standley C, Fogarty KE, Wolfe SA, Lawson ND, 2009. Chemokine signaling guides regional patterning of the first embryonic artery. Genes Dev. 23, 2272–2277. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Strasser GA, Kaminker JS, Tessier-Lavigne M, 2010. Microarray analysis of retinal endothelial tip cells identifies CXCR4 as a mediator of tip cell morphology and branching. Blood 115, 5102–5110. [DOI] [PubMed] [Google Scholar]
  46. Stratman AN, Burns MC, Farrelly OM, Davis AE, Li W, Pham VN, Castranova D, Yano JJ, Goddard LM, Nguyen O, Galanternik MV, Bolan TJ, Kahn ML, Mukouyama YS, Weinstein BM, 2020. Chemokine mediated signalling within arteries promotes vascular smooth muscle cell recruitment. Commun Biol 3, 734. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Tachibana K, Hirota S, Iizasa H, Yoshida H, Kawabata K, Kataoka Y, Kitamura Y, Matsushima K, Yoshida N, Nishikawa S, Kishimoto T, Nagasawa T, 1998. The chemokine receptor CXCR4 is essential for vascularization of the gastrointestinal tract. Nature 393, 591–594. [DOI] [PubMed] [Google Scholar]
  48. Takabatake Y, Sugiyama T, Kohara H, Matsusaka T, Kurihara H, Koni PA, Nagasawa Y, Hamano T, Matsui I, Kawada N, Imai E, Nagasawa T, Rakugi H, Isaka Y, 2009. The CXCL12 (SDF-1)/CXCR4 axis is essential for the development of renal vasculature. J Am Soc Nephrol 20, 1714–1723. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Tasic B, Hippenmeyer S, Wang C, Gamboa M, Zong H, Chen-Tsai Y, Luo L, 2011. Site-specific integrase-mediated transgenesis in mice via pronuclear injection.Proc Natl Acad Sci U S A 108, 7902–7907. [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Xu C, Hasan SS, Schmidt I, Rocha SF, Pitulescu ME, Bussmann J, Meyen D, Raz E, Adams RH, Siekmann AF, 2014. Arteries are formed by vein-derived endothelial tip cells. Nat Commun 5, 5758. [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Yanai N, Satoh T, Obinata M, 1991. Endothelial cells create a hematopoietic inductive microenvironment preferential to erythropoiesis in the mouse spleen. Cell Struct Funct 16, 87–93. [DOI] [PubMed] [Google Scholar]
  52. Yang Y, Oliver G, 2014. Development of the mammalian lymphatic vasculature. J Clin Invest 124, 888–897. [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Zudaire E, Gambardella L, Kurcz C, Vermeren S, 2011. A computational tool for quantitative analysis of vascular networks. PLoS One 6, e27385. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

1
Download video file (30MB, mp4)
2
NIHMS1709040-supplement-2.docx (81.4MB, docx)
3
Download video file (30MB, mp4)

RESOURCES