Introduction
Desmosomes provide strong adhesion to maintain tissue function and organ architecture. Organs that frequently experience mechanical stress, such as the skin and heart, particularly express abundant desmosomes to provide plasma membrane attachment sites for adjacent cells [
1]. Desmosomes are adhesive intercellular junctions comprising two cadherin proteins, desmogleins (Dsg) and desmocollins [
2]. Human genome encodes four desmogleins (Dsg1–4) which are single-pass transmembrane proteins with five extracellularly tandem conserved cadherin domains (EC1-EC5) and an intracellular domain that bind to intermediate filaments via adaptor proteins, desmoplakin and plakoglobin [
1]. Intercellular junctions of cadherin binding sites are composed of EC1 domains revealed by electron tomography studies of native desmosomes [
3,
4]. The specificity of adhesion had been confirmed by function-blocking peptides derived from EC1 domain [
5]. Differentially proteolytic cleavage fragments containing EC domains had been determined in human cancer lines [
6]. Clinically, shedding of Dsg2 extracellular domains are detected in patients with ulcerative colitis [
7]. Mutations of Dsg2 are detected in patients with arrhythmogenic right ventricular cardiomyopathy (ARVC) [
8], and expression of Dsg2 is increased in several epithelial-derived malignancies including basal-cell carcinomas, squamous cell carcinomas, and metastatic prostate cancer [
9‐
11]. These studies show the importance of Dsg2 homeostasis for the regulation of signaling in cell proliferation, migration, and epithelial-mesenchymal transition (EMT).
The therapeutic potential of endothelial progenitor cells (EPCs) has gained great interest since the observations that a significant number decrease of circulating EPCs was detected in patients with severe conditions, such as diabetes and repeated hospitalization for heart attacks [
12]. EPCs isolated from peripheral bloods consistently produce two distant subtypes which had been named as early EPCs and endothelial colony-forming cells (ECFCs), also called late EPCs for their late appearance in culture. Early EPCs, which produce paracrine factors, have limited culturing passages, and ECFCs, which directly incorporate into vasculature, have a strong growth capacity. Intramuscular injection of human ECFCs rescues blood perfusion of hindlimb ischemic mice [
13] that provides rationale for clinical trials using ECFC infusion as ischemic cardiovascular disease therapy [
14].
Previously, we had identified the antagonist role of Dsg2 on cancer metastasis [
15]. Polyclonal Dsg2 antibody and the immunogenic epitope derived from EC2 domain suppress EMT and invasion of human melanoma, breast cancer, and prostate cancer cells, consistent with the observation that Dsg2 exhibits a non-adhesive function for cell migration and morphogenesis [
1,
5,
6]. Here, we use Dsg2 antibody and its immunogeic peptide KC21 to test their effects on the control of vessel overgrowth in vivo and to screen the candidates involved in Dsg2-mediated ECFC angiogenesis.
Methods
Isolation, Characterization, and Culture of Human ECFCs
Ethical approval (No. 15MMHIS112) was granted by the Mackay Memorial Hospital Institutional Review Board, Taipei, Taiwan. Informed consent was obtained from healthy donors before the collection of peripheral blood (80 mL). The peripheral blood mononuclear cells (PBMCs) obtained from healthy donors were fractionated from other blood components by centrifugation. EPCs were isolated using CD34 MicroBead kit and MACS cell separation system (Miltenyi Biotec). In this study, PBMCs were cultured for 28 days to get ECFCs (late ECFCs) as described [
16]. ECFCs were defined as CD34
+KDR
+AC133
+CD31
+ as described [
13]. ECFCs were cultured in MV2 complete medium (PromoCell, Germany) with hEGF (5 ng/ml), hVEGF (0.5 ng/ml), hFGF-B (10 ng/ml), IGF-1(20 ng/ml), ascorbic acid (1 μg/ml), hydrocortisone (0.2 μg/ml), and 20% fetal bovine serum. 1 × 10
4 cells/cm
2 were seeded on 1% gelatin-coated dish (BD Biosciences) and maintained in the 37 °C incubator under a humidified 95% air and 5% CO
2 atmosphere.
Cell Viability and Proliferation Analysis
Cell viability was measured using the cell counting kit-8 (CCK-8) (Sigma-Aldrich) to reflect the dehydrogenase activity of living cells. ECFCs were seeded onto 96-well plates and treated with Dsg2-derived peptides (100, 200, and 400 μM). Twenty-four hours later, CCK-8 solutions were added to each well for 4 h, and the medium was harvested for the measurement of absorbance at 450 nm using a microplate reader. For cell proliferation assay, ECFCs were treated with Dsg2-derived peptide (100, 200, and 400 μM) for 4 h and then fixed. Cells labeled with 5-bromo-2′-deoxyuridine (BrdU) were subsequently identified with a primary antibody against BrdU and visualized with a secondary antibody conjugated with horseradish peroxidase using tetramethylbenzidine as a substrate.
Immunohistochemistry
Immunostaining of cell cultures was described previously [
13]. Antibodies used and dilution ratio were rabbit antiDsg2 (1:100, [EPR6767(B)], GeneTex), mouse antiplakoglobin (1: 200, clone 15/γ-Catenin, BD Biosciences), and mouse antiPECAM1 (1: 100, clone JC70A, Dako). Rat monoclonal antiendomucin (1: 100, clone V.7C7.1, abcam). Cells were fixed with −20 °C methanol for 10 min. Phalloidin (1: 1000, PHDG1, cytoskeleton) staining was performed at the step of secondary antibody incubation for 2 h at room temperature. Fluorescent images were acquired by confocal microscope (TCS SP5, Leica).
Zymography Assay
ECFCs were seeded at 80% confluence on 60-mm dishes in MV2 complete medium. Next day, cells were incubated in MV2 medium with 2% FBS with various concentrations of KC21 peptides for 24 h. Fifty μg of the conditioned medium was analyzed by 10% zymogram gel containing 0.1% gelatin [
17]. After electrophoresis, the gels were washed and incubated at 37 °C for 24 h. The gels were stained with Coomassie Blue to visualize proteinase activity. The digested area appeared clear over a blue background, indicating the location of matrix metalloproteinase 2 and 9 (MMP2 and 9) activity. MMP3 activity was determined by casein zymography [
18].
PAI-1 Activity Assay
ECFCs were treated with KC21 or scramble peptides for 24 h. Conditioned media were harvested for PAI-1 activity determination. PAI activity assay kit (Chemicon) utilized a chromogenic substrate cleaved by active uPA and detected by its optical density at 405 nm. Addition of PAI-1 in conditioned media blocked the cleavage of substrate by uPA. The relative PAI-1 activity was obtained by plotting with the standard curve of 10 units of uPA inhibited by a series dilution of PAI-1 incubated at 37 °C for 2 h as the assay instructions described.
Western Blot
ECFCs were lysed with SB-20 buffer (0.2 g/mL SDS, 10 mM EDTA, 100 mM Tris-HCl, pH 6.8), and protein concentrations were determined by modified Lowry’s method. Aliquots of cell lysates were loaded into 10% SDS-polyacrylamide gels, electrophoresed, and transblotted onto polyvinylidene fluoride membranes (Millipore). The blots were blocked with 10% bovine serum albumin for 1 h and probed with indicated primary antibody for two hours. The blots were further incubated with alkaline phosphatase-conjugated secondary antibodies for 1 hour at a room temperature. Immunoreactivity was visualized using CDP-star system (Roche) according to the manufacturer’s instruction. Primary antibody for Dsg2 (Santa Cruz), p38, p-p38, Akt, p-Akt, ERK, p-ERK, MMP9, and PAI-1 (Cell signaling) were diluted with PBS in 1 to 1000.
Wound Healing Assay
ECFCs were grown on twenty-four-well plates to reach confluence. Cell-free gap was generated using SPLScar™ Block (0.5 mm width, #201905, SPL Life Sciences, Korea) and photographed by optical microscopy (Leica, Germany) at × 40 magnification as basal line. After 4-h culture, cells were fixed and imagined to measure new growth areas using Image-J software (NIH). The ratio of the new migration area was calculated relative to the initial wound area and normalized to that for PBS-treated cells as described [
19].
Growth factor-reduced Matrigel (BD Biosciences) was thawed at 4 °C before use. Twenty-four-well plates were coated with Matrigel (200 μL/well) and polymerized for 30 min at 37 °C. ECFCs resuspended with various concentrations of KC21 peptides, scramble peptides, and antiDsg2 antibody (10 ng/ml or bev (0.25 μg/mL)) were seeded on Matrigel-coated wells at a density of 5 × 104 cells in MV2 medium containing 2% FBS for 24 h at 37 °C in a 5% CO2 humidified incubator. Each sample was tested in triplicate on the same plate, and wells were photographed with a Leica microscope with camera (× 40 magnification). Five fields were randomly chosen in each well to measure tube length and junction number manually using Image-Pro Plus 6.0 (Rockville, MD). Total tube length and junction number per field were calculated.
Animal Experiments
All animal experiments were approved by the Institutional Animal Care and Use Committee of the Mackay Memorial Hospital. C57BL/6 mice were kept and bred in accordance with the institutional ethical committee guidance (approval number: MMH-A-S-105-67).
Matrigel Plug Assay
To assess the antiangiogenic effects of KC21 peptides in vivo, growth factor-reduced liquid Matrigel (0.5 mL) containing heparin (60 U/mL), VEGF (10 ng/mL, with the exception of control), and KC21 peptides or scramble peptides were subcutaneously injected into the mice near the abdominal midline. Seven days after injection, mice were euthanized and Matrigel plugs were surgically removed. For macroscopic analysis of angiogenesis, hemoglobin content in Matrigel was measured with Drabkin’s reagent kit 525 (Sigma-Aldrich).
Oxygen-Induced Retinopathy Assay
Retinal neovascularization was induced by the use of a well-established murine model of oxygen-induced retinopathy [
20]. Neonatal mouse (C57BL/6) pups at postnatal day 7 (P7) with their nursing mothers were maintained for 5 days in 75% oxygen and then returned to room air (relative hypoxia) to produce retinal neovascularization at P12. PBS, scramble, KC21 peptides (25 μg), or Bev (10 µg) were then administered by intravitreal injection into mouse eyes at P12. The animals were sacrificed and the mouse eyes were enucleated at P17. Mouse eye cups were fixed in 4% paraformaldehyde for 2 h. The retinas were carefully separated from eye cups and then incubated with fluorescein-labeled isolectin-B4 (Life technologies) at 4 °C overnight. Samples were mounted with Vectashield medium (Vector Laboratories), and the isolectin labeling was examined by using the × 20 objective of a Leica TCS SP5 confocal microscope. Fluorescence volume measurements were recorded by creating image stacks of optical slices within lesions with QWIN software.
Discussion
Our study showed that the designed KC21 peptides have the ability to inhibit human ECFC migration in vitro, to reduce VEGF-induced capillary growth and to suppress oxygen-induced retinal neovascularization in vivo. At the cellular level, KC21 peptides specifically attenuate VEGF-induced activation of p38 MAPK, but not the signaling targets of Akt and ERK. Also, KC21 peptides regulate MMP9 and PAI-1 to stabilize the extracellular matrix of ECFCs.
The functions of Dsg2 on angiogenesis have gained attention recently. Depletion of Dsg2 by siRNA impaired tube-like structure formation in MVECs [
31]. These findings suggested that Dsg2 is a molecular target in regulating angiogenesis. Consistently, Dsg2 antibody profoundly inhibits tube-like structure formation (Fig.
S2). The effects of Dsg2 antibody and KC21 on ECFC angiogenesis inhibition are comparable to the application of Dsg2 antibody and EC2 domain peptides on disturbing intercellular barriers in human colon carcinoma (Caco) enterocytes [
32]. Our previous study also showed that Dsg2 antibody and the immunogenic epitope derived from EC2 domain suppress EMT and metastasis of human cancer cell lines [
15]. As EC2 domain is not involved in Dsg2 homotypic interaction and Dsg2 is ubiquitously distributed in ECFCs, the effects of KC21 and Dsg2 antibody on suppressing ECFC angiogenesis and migration support the observation that Dsg2 exhibits a non-adhesive function to regulate cell migration and tissue morphogenesis [
1,
5,
6]. Our results suggest that KC21 peptides may attenuate ECFC migration through inhibiting VEGF-mediated p38 activation (Fig.
4), consistent with the finding that p38 MAPK mediates VEGF-induced migration in HUVECs [
33].
Proteolysis of extracellular matrix is an initiation step for the recruitment of endothelial progenitor cells to establish new capillaries. Matrix metalloproteinases (MMPs) are extracellular endopeptidases selectively degrading components of the extracellular matrix. Shedding Dsg2 ectodomains by MMPs had been detected in the inflamed intestinal mucosa of mice with colitis and patients with ulcerative colitis [
7]. In this study, KC21 peptides specifically inhibit the activity of MMP9 but not MMP2 and MMP3 (Fig.
5c, d), consistent with the results that ectodomains of Dsg2 are substrates of MMP 9 but not of the other MMPs [
7]. As MMP9 activity is regulated by plasmin and inhibited by PAI-1, increase of PAI-1 cellular level and activity by KC21 peptides may further inhibit MMP9 activity (Fig.
5e).
The therapeutic functions of mature endothelial cells on ischemic diseases have been tested in hindlimb ischemic animal model [
34]. Both KC21 and bev strongly suppress angiogenesis in mature endothelial cells including human aortic endothelial cells (HAECs) and human umbilical vein endothelial cells (HUVECs) (Fig.
S6). However, to our surprise, KC21 profoundly inhibits ECFC tube formation (Fig.
6a), while bev has no such effect. As both mature ECs and ECFCs contribute to the progression of pathogenic angiogenesis [
35‐
37], KC21 might be a more potent peptide drug than bev on ocular neovascularization control.
To our knowledge, this study is the first report showing evidences for Dsg2-derived peptides to suppress retinal neovascularization in the mouse oxygen-induced-retinopathy (OIR) model. Neovascularization causes ocular vessel leakiness, edema, retinal detachment, and even blindness. VEGF is a major hypoxia-induced angiogenic factor and is found to be increased in the vitreous and retina to exacerbate retinopathy. Clinically, therapeutic agents against VEGF, such as bev and aflibercept, have been widely used to control the overgrowth of retinal blood vessels for vision rescue. However, more than 30% of the patients do not respond to these therapies and adverse events were also reported [
38]. The finding in this study that the inhibitory effects of KC21 peptides on neovascularization in the mouse OIR model is comparable to that of Bev (Fig.
6b) provides a new potential target for developing alternative or combined therapeutic options for retinal vascular diseases.
OpenAccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.