Transgenic Overexpression of Vascular Endothelial Growth Factor-B Isoforms by Endothelial Cells Potentiates Postnatal Vessel Growth In Vivo
http://www.100md.com
循环研究杂志 2005年第9期
The Queensland Institute of Medical Research (A.W.M., S.A.G., M.M.C., I.D.T., D.B., C.P., A.Z., N.K.H., G.F.K.), Herston, Brisbane, Queensland
AMRAD Corp Ltd (A.N., P.S.), Richmond, Victoria, Australia.
Abstract
Vascular endothelial growth factors (VEGFs) play significant roles in endothelial growth, survival, and function, and their potential use as therapeutic agents to promote the revascularization of ischemic tissues in being avidly explored. VEGF-A has received most attention, as it is a potent stimulator of vascular growth. Results in clinical trials of VEGF-A as a therapeutic agent have fallen short of high expectations because of serious edematous side effects caused by its activity in promoting vascular permeability. VEGF-B, a related factor, binds some of the VEGF-A receptors but not to VEGF receptor 2, which is implicated in the vascular permeability promoting activity of VEGF-A. Despite little in vitro evidence to date for the ability of Vegf-B to directly promote angiogenesis, recent data indicate that it may promote postnatal vascular growth in mice, suggesting that it may have potential therapeutic application. We have specifically studied the effects of VEGF-B on vascular growth in vivo and on angiogenesis in vitro by analyzing transgenic mice in which individual isoforms (VEGFB167Tg and VEGFB186Tg) of VEGF-B are overexpressed in endothelial cells. VEGFB167Tg and VEGFB186Tg mice displayed enhanced vascular growth in the Matrigel assay in vivo and during cutaneous wound healing. In the aortic explant assay, explants from VEGFB167Tg and VEGFB186Tg mice displayed elevated vascular growth, suggesting a direct effect of VEGF-B isoforms in potentiating angiogenesis. These data support the use of VEGF-B as a therapeutic agent to promote vascular growth, in part, by potentiating angiogenesis. Furthermore, the lack of vascular permeability activity associated with either transgenic overexpression of the VEGF-B gene in endothelial cells or application of VEGF-B protein to the skin of mice in the Miles assay indicates that use of VEGF-B as a therapy should not be associated with edematous side effects.
Key Words: vascular endothelial growth factor-B growth factor angiogenesis endothelium
Introduction
Promoting the revascularization of ischemic tissues by local or systemic administration of growth factors presents an attractive potential therapy for the treatment of cardiovascular disease. Numerous growth factors have been shown to stimulate angiogenesis in vitro and blood vessel growth in vivo1 but have failed to meet relevant end points in clinical trials and/or have been associated with adverse side effects. For example, vascular endothelial growth factor (VEGF)-A, which is a potent angiogenic factor and has been considered a prime candidate for promoting revascularization of ischemic tissues, causes tissue edema and hypotension because of its potent vascular permeability activity.2 As a result, there is a need to identify factors that can promote vessel growth in the absence of other activities to provide more selective treatments.
Signaling through VEGF receptor (VEGFR)-1 has emerged as a promising therapeutic strategy for use in promoting tissue revascularization.3 VEGF-A, VEGF-B, and placental growth factor (PlGF) bind to VEGFR-1.4 These factors promote vascular growth in vivo and are proposed as potential therapeutic agents for the treatment of ischemia.3 The suitability of these factors as therapeutic agents will not depend only on their ability to promote vessel growth but also on their side effect profile. VEGF-A, VEGF-B, and PlGF also bind to neuropilin-1 (in an isoform-specific manner)5 and VEGF-A additionally binds to VEGFR-2.6 Gene knockout studies in mice indicate that, despite the overlapping receptor binding profiles, each of these factors may have distinct roles in vascular signaling.7eC10 VEGF-A directly stimulates endothelial cell proliferation and migration.11 It is also a potent inducer of vascular permeability12; hence, it is also known as vascular permeability factor. In contrast, PlGF does not directly promote endothelial cell proliferation or migration and is only a weak stimulator of vascular permeability in the Miles assay.8,13 Characterization of PlGF gene knockout mice has revealed that PlGF is required for VEGF-AeCinduced endothelial cell proliferation and vascular permeability.8 Furthermore, transgenic overexpression of PlGF under the control of the K14 promoter leads to both vascular growth and edema in the skin of mice, which, in part, may be explained by the induction of expression of VEGF-A.14 This indicates that PlGF signaling is closely linked to that of VEGF-A, suggesting that, similar to VEGF-A, the use PlGF as a therapeutic agent to promote tissue revascularization may be associated with an edematous side effect.
VEGF-B promotes vascular growth in adult mice when administered either in the form of recombinant protein in Matrigel plugs or when expressed in situ by adenoviral vector at sites of experimentally induced tissue ischemia.15 This activity of VEGF-B was shown to be mediated by VEGFR-1, a receptor tyrosine kinase expressed by endothelial cells16 and a variety of other cell types,17eC19 and which is induced in ischemic tissue via a hypoxia-inducible factor-1aeCdependant mechanism.20,21 Signaling through VEGFR-1 has also been implicated in postnatal vessel growth in other model systems.22,23 There is no evidence to suggest that VEGF-B is either required for or can potentiate the activity of VEGF-A, and there have been no reports that VEGF-B can promote vascular permeability. On the contrary, we have previously reported that the ability of VEGF-A to induce vascular leakage in the Miles assay is not impaired in Vegfb gene knockout mice.10 Collectively, these data indicate that VEGF-B may be able to promote vascular growth without induction of vascular leakage and, as a therapeutic agent, may have a better side-effect profile than that predicted for VEGF-A and PlGF. However, not all reports support such therapeutic potential of using VEGF-B, as other studies found VEGF-B to be ineffectual at stimulating vessel growth in muscle and periadventitia when delivered by adenoviral vector.24,25 This may reflect tissue-specific differences in effectiveness of VEGF-B on vascular signaling and/or variations in the localized environment, such as the presence of other growth factors. These conflicting results from studies investigating the ability of VEGF-B to promote vessels growth are confounded by a lack of in vitro evidence supporting a direct angiogenic activity of VEGF-B.
To clarify the function of VEGF-B isoforms in aspects of postnatal vessel growth, we have generated transgenic mice in which VEGF-B167 and VEGF-B186 expression is driven by the endothelial celleCspecific Tie2 promoter/enhancer. Analysis of these mice demonstrates a role for both isoforms of VEGF-B in the potentiation, rather than the initiation, of postnatal vessel growth in vivo and of angiogenesis in vitro. Further, the absence of edema in the skin of transgenic animals in combination with an inability of recombinant VEGF-B to induce vascular permeability in the Miles assay indicates that VEGF-B does not induce vascular leakage.
Materials and Methods
Recombinant VEGF-B Expression and Refolding
Recombinant VEGF-B isoforms were produced essentially as previously reported.26 The coding regions of the cDNA corresponding to amino acid residues 10 to 108 and 1 to 167 of both human and mouse VEGF-B were cloned into the pET-15b vector (EMD Biosciences, San Diego, Calif) and transformed into Escherichia coli strain BL21-CodonPlus-(DE3)-RP (Stratagene, La Jolla, Calif). Recombinant protein expression was induced mid-log phase in Superbroth (30% wt/vol tryptone, 15% wt/vol yeast extract, 0.5% wt/vol NaCl) with 10 nmol/L isopropyl eCD-thiogalactoside. Inclusion bodies were isolated and the material purified under reducing denaturing conditions (6 mol/L guanidine hydrochloride, 0.1 mol/L NaH2PO4, 10 mmol/L -mercaptoethanol) by nickel metal affinity chromatography (MAC). The purified material was reduced and refolded by dialysis in redox solution as previously described.26 The refolded material was acidified by dialysis in 0.1 mol/L acetic acid and purified by reverse-phase high-performance liquid chromatography. The N-terminal His6 tag used for nickel MAC was subsequently removed by cleavage with Genenase I (New England Biolabs, Beverly, Mass).
Human VEGFR-1 Chimeric Receptor CelleCBased Assay
Development of the human VEGFR-1 celleCbased assay has been previously reported.27 A cDNA encoding a chimeric receptor consisting of the extracellular domain of human VEGFR-1 fused to the transmembrane and intracellular domains of mouse erythropoietin receptor was generated by introducing a BglII site into the VEGFR-1 cDNA before the sequence encoding the transmembrane domain, followed by ligation of a BglIIeCNotI fragment consisting of the transmembrane and intracellular domains of erythropoietin receptor.28 The cDNA was subcloned into the pEF-BOS expression vector29 and transfected by electroporation into the interleukin (IL)-3eCdependent pre-B cell line, BaF3, together with pcDNA3.1(+) zeo vector (Invitrogen, Carlsbad, Calif) encoding Zeocin resistance. The transfected cells were selected in complete media (DMEM [Invitrogen]; 5% FCS, 50 U/mL penicillin, 50 e/mL streptomycin, 2 mmol/L GlutaMAX [Invitrogen]) supplemented with 250 e/mL Zeocin (Invitrogen), and a Zeocin-resistant line able to proliferate in response to VEGF-A or VEGF-B in the absence of IL-3 was subsequently isolated and cloned by limit dilution. hVEGFR1/EpoR/BaF3 clones were routinely cultured in complete media supplemented with 5 ng/mL VEGF-B167.
For the assay of VEGFR-1 ligand activity, selected ligands were diluted into 96-well flat bottom tissue culture plates (TPP, Trasadingen, Switzerland). hVEGFR1/EpoR/BaF3 cells, washed 5 times in PBS, were plated at 6x105/well to give a final assay volume of 100 mL. For ligand neutralization assays, a fixed concentration of ligand was preincubated with antibody for 1 hour before the addition of hVEGFR1/EpoR/BaF3 cells. Cells were incubated for 72 hours at 37°C, 10% CO2, in a humidified container and proliferation assessed using an methanethiosulfonate (MTS) dye reduction assay (Promega, Madison, Wis) with the colorimetric change measured at a wavelength of 450 nm (Wallac VICTOR microplate reader, PerkinElmer, Boston, Mass).
Generation of Transgenic Mice
All mice in this study were treated in accordance with the Australian National Health and Medical Research Council guidelines for the care and use of experimental animals. Transgenic mice (FVB/N-TgN[Tie2-VEGFB167]1Gfk [henceforth referred to as VEGFB167Tg] and FVB/N-TgN[Tie2-VEGFB186]1Gfk [henceforth referred to as VEGFB186Tg]) were generated by standard techniques in FVB/N strain mice using the expression constructs shown in Figure 1a. The constructs were designed to result in expression of each isoform of VEGF-B (GenBank Nos. U43368 and U43369) under the control of an altered Tie2 enhancer/promoter cassette.30 This enhancer/promoter combination has been reported to drive constitutive endothelial celleCspecific transgene expression.30 Transgenic animals were genotyped by PCR amplification of tail-tip DNA using primers (F1 5'-ctg tct ccc agc ctg atg cc-3' and R1 5'-ctg ggt tga gct cta agc cc-3') specific for the human VEGFB cDNA sequence. The lines of transgenic mice reported here were chosen for the studies after analysis by reverse transcription RT-PCR and immunohistochemistry to confirm that the transgene was being appropriately expressed (see below).
The number of transgene copies incorporated in each transgenic mouse line was estimated from genomic DNA preparations by semiquantitative PCR (Figure 1b). The primers used to amplify the transgenes were the F1 and R1 primers (as above) multiplexed with primers that specifically amplify a region from the endogenous mouse Vegfb gene.7 Briefly, PCR products from increasing cycles of amplification were electrophoresed through an agarose/Tris-acetate-EDTA gel stained with Vistra-green (Amersham Pharmacia Biotech, Uppsala, Sweden). PCR products were visualized by fluorescence imaging on an Amersham Pharmacia Biotech Typhoon 8600 laser scanner (Amersham Pharmacia Biotech). ImageQuant software was used for image analysis and quantification of the PCR products. The gene dosage was estimated by measuring band intensity of the PCR products at their threshold detection levels relative to amplification cycle in comparison to the endogenous diploid Vegfb gene sequence.
Transgene expression levels were compared in RNA isolated from lung tissue for the VEGFB167Tg and VEGFB186Tg mice by semiquantitative RT-PCR and referenced to the level of Hprt expression, as previously described31 (Figure 1c), for transgene copy number estimation. The primers used for RT-PCR amplification of the transgenes were F2 5'-cgc cgc act cct gca gct-3' and R2 5'-gct ggc agg tag cgc gag-3'.
Expression and localization of human VEGF-B protein in VEGFB167Tg and VEGFB186Tg mice was determined by immunofluorescence (IF) on cryosections of normal heart (Figure 1d) and by immunohistochemistry (IHC) on paraffin sections of healing cutaneous wounds (Figure 1e). To specifically detect transgene expressed VEGF-B protein in cryosections of heart using the antieCVEGF-B monoclonal antibody (mAb) (clone 2H10),26 which detects both human and mouse VEGF-B; transgenic mice were crossed with VegfbeC/eC mice7 to generate VEGFB167Tg;VegfbeC/eC and VEGFB167Tg;VegfbeC/eC genotype animals. Similar sections from wild-type (C57BL/6J strain) and VegfbeC/eC mice (backcrossed for 12 generations onto the C57BL/6J background) were used as controls. Sections prepared from cutaneous wounds were chosen because Tie-2 expression has been shown to be upregulated in newly formed vessels within the granulation tissue and existing vessels at wound edges.32 For IHC analysis using antieChuman VEGF-B polyclonal Ab (N-19, Santa Cruz Biotechnology, Santa Cruz, Calif), full-depth cutaneous wound samples including surrounding tissue (at 5 days postwounding) were fixed by immersion in 4% paraformaldehyde (PFA) in PBS overnight at 4°C and paraffin embedded.
Matrigel Assay of Neovessel Growth in Vivo
A modified in vivo assay of neovessel growth described by Passaniti et al33 was used. Briefly, 0.5 mL of ice-cold Matrigel matrix (BD Biosciences, Bedford, Mass) infused with 300 ng/mL of recombinant human basic fibroblast growth factor (FGF)34 (Invitrogen) was injected SC near the abdominal midline of VEGFB167Tg, VEGFB186Tg, and wild-type littermate mice (8- to 10-week-old females). Soon after injection, the Matrigel solidified to form a semisolid plug. The mice were then maintained under normal conditions for 6 days, when the experiment was terminated and the angiogenic response studied. The mice were euthanized by CO2 asphyxiation and the Matrigel plugs with the supporting skin and peritoneal lining intact were dissected and frozen in OCT compound (Tissue-Tek, Sakura Finetek, Torrance, Calif). Vessels were identified by CD31 IHC on cryosections and neovessel growth quantified by measuring the density of vessels at the Matrigel/skin interface as described below. As macrophages also express CD31, single cells positive for CD31 were counted only if they surrounded a visible vessel lumen (ie, capillaries).
Assessment of Neovessel Growth in Healing Wounds
VEGFB167Tg, VEGFB186Tg, and wild-type littermate mice (8- to 10-week-old males and females) were anesthetized by IP injection of anesthetic (0.8 mg/kg fentanyl citrate, 25 mg/kg fluanisone, 12.5 mg/kg midazolam hydrochloride) and 2 full-depth cutaneous wounds were made on the backs using a 6-mm Keyes dermal punch. Wounds were left uncovered, and after 5 or 10 days, the mice were euthanized by CO2 asphyxiation for wound analysis. Wounds were dissected with the surrounding skin intact and fixed by immersion in 4% PFA in PBS overnight at 4°C. Wounds were bisected at the widest point and embedded in paraffin. Vessels were identified on paraffin sections by CD31 IHC and vessel density determined by counting CD31-positive structures and single positive cells surrounding a visible vessel lumen (capillaries) per microscopic field within the granulation tissue.
Aortic Explant Assay
A modified version of the mouse aortic ring assay (Masson et al35) was used to assess angiogenesis in aortic explants from VEGFB167Tg and VEGFB186Tg mice. Essentially, the descending aorta from VEGFB167Tg, VEGFB186Tg, and wild-type littermate mice (C57BL/6JxFVB/N strain, F1 progeny) were dissected and placed into ice-cold DMEM. Aortas were then placed in a Petri dish containing DMEM and the periaortic fibroadipose tissue removed under a stereo dissection microscope using fine-tipped forceps and microdissection scissors. The aortas were then longitudinally bisected and cut into 1-mm squares. After washing in fresh DMEM, individual squares of aorta tissue were embedded in a drop (25 e蘈) of rat tail collagen solution35 that was placed centrally in the bottom of a well of a 24-well flat bottom tissue culture plate (BD Labware, Franklin Lakes, NJ). The collagen solution was polymerized by placing the tissue culture plate in a humid incubator for 10 minutes at 37°C. Assay medium35 (300 mL of MCDB131 [Invitrogen] supplemented with 2.5% serum from VegfbeC/eC mice,7 25 mmol/L NaHCO3, 1% glutamine, and penicillin/streptomycin) was then added to each well.
In some experiments, antieCVEGF-B mAb (clone 2H10) or an isotype control mAb (IgG2a, clone C44) was added to the assay medium to a final concentration of 50 e/mL. In other experiments, aortic explants from VEGFB167Tg, VEGFB186Tg, and wild-type littermate mice were cocultured in the same collagen gel with aortic explants from VegfbeC/eC mice (C57BL/6J strain). Cultures were incubated in a humid incubator at 37°C/5% CO2 for 6 to 8 days for optimal endothelial vesseleClike growth. For analysis, collagen gels containing the aortic explants were fixed in situ with 2% PFA in PBS (20 minutes at room temperature [RT]) and endothelial vesseleClike outgrowths identified by immunofluorescence using fluorescein isothiocyanate (FITC)-conjugated BS-1 lectin (10 e/mL in 10 mmol/L HEPES-buffered saline, pH 7.5; Sigma-Aldrich) or by inverted light microscopy. Angiogenesis was assessed by counting the number of vessel-like sprouts and the number of vessel-like branches per aortic explant.
Endothelial Progenitor Cell Assay
The levels of circulating endothelial progenitors (EPCs) in VEGFB167Tg, VEGFB186Tg, and wild-type littermate mice were determined from 500 e蘈 of heparinized blood samples obtained by heart puncture as previously described.36
Miles Assay of Vascular Permeability
The vascular permeability at sites of intradermal injections of recombinant human VEGF-B167 (0.1 to 100 ng in 20 e蘈 of 0.01% BSA/PBS) were compared with a single dose of recombinant human VEGF-A165 (10 ng/site; R&D Systems, Minneapolis, Minn) using an adaptation of the Miles assay37 in FVB/N strain mice, as previously described.10
Antibodies and Immunohistochemistry
Endothelial cell proliferation in adult tissues was assessed in paraffin sections using MCM-7 IHC.38 Briefly, paraffin sections were dewaxed in xylene and hydrated through a graded series of alcohol. Antigen epitope was heat retrieved in 10 mmol/L citrate buffer, pH 6.0, and the sections permeabilized by incubation in 0.1% Triton X-100/PBS for 10 minutes at RT. Endogenous peroxidase activity and nonspecific protein-binding sites were sequentially blocked by incubation with 3% hydrogen peroxide/1% NaN3 in PBS for 30 minutes and 1% BSA/PBS for 20 minutes at RT. Sections were then incubated with 1 mg/mL of biotinylated MCM7 Ab2 (Clone 47DC141; NeoMarkers, Fremont, Calif) at RT for 90 minutes followed by incubation with streptavidin-conjugated horseradish peroxidase (1/600 dilution; DAKO Cytomation, Glostrup, Denmark) at RT for 30 minutes. Peroxidase activity was visualized using diaminobenzidine (DAB) and the slides lightly counter stained with hematoxylin. Endothelial cells and macrophages in tissue sections were identified by CD31 and F4/80 IHC. CD31 and F4/80 IHC were performed on paraffin sections of adult tissue and wound-healing samples and on cryosections of Matrigel samples. For paraffin-embedded sections, samples were dewaxed in xylene and hydrated through a graded series of alcohol. Antigen epitopes were retrieved by incubating the sections with 1 e/mL proteinase K in PBS at RT for 10 minutes or with 0.1% trypsin in PBS for 15 minutes at 37°C for CD31 and F4/80, respectively. Endogenous peroxidase activity was blocked by incubation with 3% hydrogen peroxide/1% NaN3 in PBS for 30 minutes at RT. To block nonspecific protein binding and tertiary antibody binding sites, the slides were sequentially incubated in 4% skim milk powder in PBS and 10% normal goat serum in PBS at RT for 15 and 30 minutes, respectively. Samples were then incubated overnight at 4°C with anti-mouse CD31 (1/800 dilution, clone MEC 13.3, rat IgG2a; Pharmingen, San Jose, Calif) or anti-mouse macrophage F4/80 (1/50 dilution of culture supernatant clone AC-1, rat IgG2a). Primary antibodies were detected by sequential incubation with polyclonal rabbit anti-rat IgG (1/750 dilution; Sigma-Aldrich, St Louis, Mo) and DAKO EnVision+ anti-rabbit peroxidase system (DAKO Cytomation) at RT for 30 minutes. Peroxidase activity was visualized using DAB and the slides lightly counter stained with hematoxylin. Cryosections (7 e) were air dried overnight, fixed with ice cold acetone for 5 minutes at RT, and air dried again. After rehydration in PBS, IHC for CD31 and F4/80 were performed essentially as described above for paraffin sections without the antigen retrieval.
VEGF-B protein expression was detected in heart tissue by IF. Cryosections were air dried overnight and fixed by immersion in ice-cold 50% acetone/50% methanol for 10 minutes. Endogenous biotin was blocked using DAKO Biotin Blocking System (DAKO Cytomation). Nonspecific protein-binding sites were blocked using 1% cold-water fish gelatin (Sigma-Aldrich) in PBS for 30 minutes at RT. Sections were then incubated with biotinylated antieCVEGF-B mAb (clone 2H10, 1 e/mL) in PBS overnight at 4°C and bound antibody detected by incubation with Rhodamine streptavidin conjugate (1/500 dilution in PBS) for 30 minutes at RT. IHC detection of human VEGF-B protein was confirmed on paraffin sections of cutaneous wounds, as previously described,39 using anti-human VEGF-B polyclonal peptide Ab (10 e/mL, N-19, Santa Cruz Biotechnology) with heat retrieval of antigen epitopes using citrate buffer, pH 6.0.
VEGFR-1 expression was detected by IHC on paraffin-tissue sections from VEGFB167Tg, VEGFB186Tg, and wild-type littermate mice. Following blocking of endogenous peroxidase activity and nonspecific protein-binding sites by sequential incubation with 3%H202/1% NaN3 in PBS and 4% skim milk power in PBS, dewaxed sections were incubated for 2 hours at RT with antieCVEGFR-1 rabbit polyclonal peptide Ab (4 e/mL,C-17), Santa Cruz Biotechnology). Sections were then washed in PBS and the bound antibody detected by incubation with DAKO EnVision+ anti-rabbit peroxidase system (DAKO Cytomation) at RT for 30 minutes. Peroxidase activity was visualized using DAB and the slides lightly counter stained with hematoxylin. As a negative control for all IF and IHC, primary antibody was omitted.
Results
Expression and Biological Activity of Human and Mouse VEGF-B Isoforms
Recombinant human (rh) and mouse VEGF-B10 to 108 and VEGF-B167 were expressed as N-terminal His6-tagged proteins in Escherichia coli. The monomeric proteins were purified by MAC and refolded by redox dialysis. The dimeric protein was separated from the monomeric material by reverse-phase high-performance liquid chromatography. SDS-PAGE analysis of the human and mouse isoforms is shown in Figure 2a.
Human and mouse VEGF-B isoforms were assessed for biological activity in the hVEGFR1/EpoR/BaF3 celleCbased assay. Both human and mouse VEGF-B10 to 108 and VEGF-B167 were shown to be active (Figure 2b), with human and mouse VEGF-B10 to 108 showing activity similar to that of human VEGF-A165 (EC50=0.1 nmol/L). In contrast, full-length human and mouse VEGF-B167 had 10- and 30-fold less activity, with EC50=1 and 3 nmol/L respectively.
Characterization of a Species Cross-Reactive VEGF-BeCSpecific Neutralizing mAB
The generation and preliminary characterization of the human VEGF-B neutralizing mouse mAb 2H10 has been described previously.26 To determine whether mAbs raised against human VEGF-B were cross-reactive with mouse VEGF-B and, thus, potentially of value in the analysis of VEGF-B activity in mouse models, the binding of 2H10 to the human and mouse VEGF-B was assessed by Western blot analysis (Figure 2c). Results presented in Figure 2c demonstrate that mAb 2H10 was able to bind to the human and mouse VEGF-B10 to 108 and VEGF-B167 isoforms. This interaction was confirmed using a variety of other approaches including ELISA and Biacore analysis (results not shown). 2H10 did not bind to reduced human and mouse VEGF-B, indicating that the 2H10 epitope is structural.
In addition to binding to both human and mouse VEGF-B, mAb 2H10 was shown to be a potent antagonist of both of these proteins. Results presented in Figure 2d demonstrate that mAb 2H10, but not the isotype control C44 antibody, inhibited human and mouse VEGF-BeCmediated proliferation of hVEGFR1/EpoR/BaF3 cells. The mAb appeared to be slightly more potent against human VEGF-B (IC50=1 nmol/L) than against mouse VEGF-B (IC50=3 nmol/L).
Evaluation of Transgene Expression in VEGFB167Tg and VEGFB186Tg Mice
A small number of viable individual transgenic founder animals with incorporation of the VEGFB transgenes were identified by PCR amplification of genomic DNA from tail-tip biopsies. More frequent transgene positive individuals were identified when mid-gestation embryonic stages were genotyped (data not shown), suggesting that high levels of expression of the transgene might be detrimental to development. Similarly, attempts to generate viable transgenic mice with ubiquitous overexpression of either VEGFB167 or VEGFB186 under the control of the HmgCoA gene promoter were unsuccessful, despite the identification of transgene positive mid-gestation embryos (data not shown). Of the viable transgenic mouse lines successfully derived, only 1 line each of VEGFB167Tg and VEGFB186Tg mice had detectable levels of transgene mRNA, as assessed by RT-PCR analysis of lung samples. We estimated by PCR amplification of the VEGFB transgenes in comparison with diploid endogenous Vegfb gene that approximately 16 copies of the VEGFB167 and 2 copies of the VEGFB186 transgenes were incorporated in the genome of these mouse lines (Figure 1b). Transgene expression levels, as estimated by RT-PCR from total RNA derived from lung tissue, showed that expression was approximately 12.7% and 1.5% of the Hprt expression level in the VEGFB167Tg and VEGFB186Tg mice, respectively (Figure 1c).
VEGFB167Tg and VEGFB186Tg mice are viable, fully fertile and appeared healthy. Aged-matched mice VEGFB167Tg and VEGFB186Tg mice (6- to 8-weeks old) displayed similar vessel densities in normal tissues to wild-type littermates, as assessed by platelet-endothelial celleCadhesion molecule (PECAM)-1 IHC on section of various organs (ie, liver, heart, and kidney; data not shown). Endothelial cell proliferation in vessels of the heart, liver, and kidney was also comparable between VEGFB167Tg, VEGFB186Tg, and their wild-type littermates, as determined by MCM-7 IHC (data not shown). IF staining for VEGF-B protein on cryosections of heart from VEGFB167Tg;VegfbeC/eC and VEGFB186Tg;VegfbeC/eC mice using antieCVEGF-B mAb (2H10) confirmed transgene protein expression. Cryosections of heart from wild-type (C57BL6J strain) and VegfbeC/eC mice were used as positive and negative controls, respectively. Positive staining was observed in wild-type heart sections, and no staining was observed in VegfbeC/eC heart sections, verifying specific VEGF-B protein staining (Figure 1d). Broad vascular staining was observed in the hearts of both VEGFB167Tg;VegfbeC/eC and VEGFB186Tg;VegfbeC/eC, consistent with Tie2 promoter/enhancereCdriven transgene expression in endothelial cells (Figure 1d). IHC using anti-human VEGF-B (rabbit polyclonal, N19; Santa Cruz Biotechnology) on paraffin sections of various tissues including heart, kidney, and liver and on 5-day-old, full-depth cutaneous wounds from both VEGFB167Tg and VEGFB186Tg mice was also performed as previously described39 but was less sensitive in detecting the transgene expression. Staining was only observed in the endothelium in blood vessels at wound edges (Figure 1e) and occasional newly formed vessels in the granulation tissue of VEGFB167Tg and VEGFB186Tg wounds. This indicates that transgene expression was increased in the vascular endothelium of vessels at wound edges and in neovessels during wound healing, which is consistent with previous reports showing upregulation of Tie-2 expression in this region in healing cutaneous wounds in mice.32 No staining of the vascular endothelium was observed in similar wound sections from wild-type mice (Figure 1e).
To determine whether the VEGFB transgenes altered expression of VEGFR-1, expression of the receptor was compared by IHC on paraffin sections of normal tissues from VEGFB167Tg and VEGFB186Tg mice with similar sections from wild-type littermates. In all genotypes, positive staining for VEGFR-1 was observed in endothelial cells in a restricted number of vessel types, predominantly small to medium veins and arteries. VEGFR-1 expression was not detected in capillary endothelium in any of the organs tested, including the heart, 1 of the sites where VEGFB transgene expression was detected (see Figure 1d). The intensity of endothelial staining and the types of vessels showing positive staining of endothelial cells were similar between the genotypes indicating that VEGFR-1 expression was not altered by VEGFB transgene expression (data not shown).
Assessment of Neovascular Growth in VEGFB167Tg and VEGFB186Tg Mice in the Matrigel Plug Assay and During Cutaneous Wound Healing
Vessel density at the Matrigel/skin interface of Matrigel plugs infused with 300 ng/mL bFGF was measured after 6 days in VEGFB167Tg, VEGFB186Tg, and wild-type littermate mice. Histological examination of H&E-stained cryosections of these plugs revealed an encapsulation of the Matrigel plug by granulation tissue and significant cellular invasion into the Matrigel matrix, largely consisting of macrophages as identified by F4/80 IHC, which was most concentrated at the periphery of the plug. Cellular invasion into the Matrigel matrix was associated with substantial angiogenesis as evidenced by in growth of PECAM-1eCpositive vessel structures into the Matrigel matrix from the surrounding granulation tissue (Figure 3a). Vessel density at the Matrigel/skin interface was significantly increased in Matrigel plugs from VEGFB167Tg (75% increase, P=0.0054 versus wild-type littermates) and VEGFB186Tg mice (52% increase, P=0.0068 versus wild-type littermates) when compared with vessel density in the same region of Matrigel plugs from wild-type littermates (Figure 3b). Levels of macrophages surrounding and invading the Matrigel plugs appeared similar in all genotypes (data not shown).
Apart from reproductive tissues, wound healing is 1 of the few nonpathological processes where neovessel growth is observed in the adult. During wound healing, neovessel growth is associated with the proliferative phase of wound healing and may facilitate the formation and, in later stages, the resolution of the granulation tissue. Based on previously published studies showing the kinetics of Tie-2 expression in wounds in mice,32 we compared angiogenic responses in granulation tissue at 5 and 10 days following full-depth cutaneous wounding among VEGFB167Tg, VEGFB186Tg, and wild-type littermates by measuring vessel density. The vessel density in the granulation tissue was significantly increased in both VEGFB167Tg and VEGFB186Tg compared with wild-type littermates at both 5 and 10 days postwounding (Figure 3c). The level of increase in vessel density at 10 days postwounding between the transgenic animals and the wild-type littermates was comparable to that observed between the same transgenic mice and wild-type littermates in the Matrigel plug assay (66% and 41% increase over wild-type vessel density for VEGFB167Tg and VEGFB186Tg, respectively; Figure 3c). The granulation tissue at 5 and 10 days postwound consisted primarily of macrophages and vascular tissue, as assessed by F4/80 (Figure 3d) and PECAM-1 (Figure 3e) IHC, respectively, and the levels of F4/80 positive cells in the granulation tissue appeared similar among genotypes when wounds from the various mouse genotypes were compared at the same time point (data not shown).
Decreased EPCs in the Circulation of VEGFB167Tg Mice
The levels of EPCs in the blood were significantly reduced in VEGFB167Tg mice (P=0.004) when compared with levels of these cells in the blood of wild-type littermates (Figure 4). The levels of EPCs in the blood of the VEGFB186Tg mice were not significantly different from the levels of these cells in the blood of wild-type littermates.
VEGFB167Tg and VEGFB186Tg Mice Display Increased Endothelial VesseleCLike Sprouting in the Aortic Explant Assay
The dynamics of endothelial vesseleClike outgrowth (angiogenesis) from cultured aortic explants was similar to that previously described.35 Vessel-like endothelial outgrowth was preceded by the migration of fibroblast-like cells from aortic explants into the surrounding collagen matrix, which was visible within the first 24 hours of culture (data not shown). Vessel-like endothelial outgrowths (identified by fluorescence microscopy after staining with FITC-conjugated BS-1 lectin; Figure 5a) from the aortic explants were first observed after 4 days in culture and maximal vessel growth was observed 4 days later, as assessed by both the number of vessels sprouts and the complexity of vessel branching (as measured as a ratio of the number of branches to the number of vessel sprouts). VEGFB167Tg and VEGFB186Tg explants displayed significantly greater numbers of vessel sprouts when compared with explants from wild-type littermates (Figure 5b). Branching complexity was comparable among the different mouse genotypes (wild-type littermates=1.44±0.14, VEGFB167Tg=1.56±0.07, and VEGFB186Tg=1.70±0.09 branches/vessel like sprout). The addition of antieCVEGF-B mAb (clone 2H10) to the culture medium significantly reduced the number of vessel sprouts in VEGFB167Tg and VEGFB186Tg aortic explant cultures when compared with their respective isotype mAb (clone C44)eCtreated control explant cultures (Figure 5c). In the presence of antieCVEGF-B antibody, the number of vessel sprouts in VEGFB167Tg and VEGFB186Tg aortic explants cultures were equivalent to that observed in wild-type littermate aortic explant cultures treated with isotype control mAb. In contrast, addition of antieCVEGF-B mAb was without effect on the level of vessel sprouts in wild-type littermate explants when compared with similar explants treated with the isotype control mAb. The complexity of vessel branching was similar among all genotypes regardless of the mAb treatment. In further studies, VegfbeC/eC aortic explants cocultured with VEGFB167Tg and VEGFB186Tg aortic explants displayed increased number endothelial vesseleClike sprouts in comparison to VegfbeC/eC explants that were cocultured with explants from wild-type littermates (Figure 5d). Consistent with the lack of effect of antieCVEGF-B antibody treatment on endothelial-like vessel sprouting in wild-type aortic explants, VegfbeC/eC aortic explants cocultured with wild-type aortic explants displayed similar levels of endothelial-like vessel sprouting to VegfbeC/eC aortic explants cocultured with VegfbeC/eC aortic explants. These data suggest that endogenous VEGF-B was without effect in this assay, which may reflect the low expression of endogenous Vegf-B normally displayed by this tissue.40
VEGF-B Does Not Induce Vascular Permeability
Intradermal injection with 10 ng of rhVEGF-A165 induced a significant increase in Evans blue dye leakage in comparison with control vehicle (0.01% BSA/PBS) (Figure 6). In contrast, intradermal injection of rhVEGF-B167 failed to induce vascular permeability, when tested at concentrations ranging from 0.1 to 100.0 ng/site, over levels observed with control vehicle (ie, 0 ng/mL).
Discussion
VEGF-B has been reported to promote postnatal angiogenesis in vivo.15 Although postnatal blood vessel growth is commonly assumed to be solely governed by angiogenesis, recent data indicate that it also occurs through the recruitment of bone marroweCderived EPCs in a process akin to embryonic vasculogenesis.41 The VEGF-B receptor VEGFR-1 has recently been implicated in mobilization of EPC from the bone marrow,36 suggesting that VEGF-B may be able to promote this process. This raises the question of whether VEGF-B is truly angiogenic, as suggested in previous in vivo studies, or whether VEGF-B may have promoted vessel growth by vasculogenesis through mobilization/recruitment of EPCs. The true mode of action of VEGF-B in promoting blood vessel growth (angiogenesis, vasculogenesis, or a combination of both) needs to be determined, so that its full therapeutic potential can be realized.
Transgenic overexpression of either of the human isoforms of VEGF-B by resting-state endothelial cells had no detectable effect on endothelial proliferation or vessel density in normal tissues, including the liver, heart, and kidney, suggesting that expression of transgene VEGF-B alone at basal levels is not sufficient to stimulate blood vessel growth. Although it may be argued that this might reflect subthreshold levels of expression of the VEGFB transgenes at resting states, this result is consistent with the minimal effect of deletion of the VEGF-B gene on vessel densities in organs normally associated with VEGF-B expression, including the heart and central nervous system.7,42,43 Further, adenoviral vectoreCexpressed VEGF-B was also without effect on vessel growth in muscle and periadventitia.24,25 One possible explanation for lack of effect of VEGF-B on stimulation of angiogenesis in normal tissue may involve insufficient levels of VEGFR-1 expression by endothelial cells.22
In contrast, the data presented here show that both transgene expressed VEGF-B isoforms were found to potentiate neovessel growth in the Matrigel plug assay, where blood vessel in growth was initiated by the action of bFGF. This was evidenced by a significant elevation in vessel density at the Matrigel-skin interface in Matrigel plug samples from VEGFB167Tg and VEGFB186Tg mice when compared with similar samples from wild-type littermates. Neovessel growth was also increased in cutaneous wounds within the granulation tissue from VEGFB167Tg and VEGFB186Tg mice when compared with those from wild-type littermates. Further, because the level of inflammatory cells or more specifically, the level of macrophages in the granulation tissue appeared similar between wounds in VEGFB167Tg and VEGFB186Tg mice compared with those from wild-type littermates, we believe that our results reflect a specific action of VEGF-B in the potentiation of vascular growth. We suggest that previous studies also reflect potentiation, rather than initiation, of vessel growth by VEGF-B.15 For example, expression of VEGF-B by adenoviral vector was found to enhance collateral artery growth at site of vessel ligation, where other growth factors including VEGF-A have been shown to be unregulated.44 VEGF-B also promoted vessel growth in Matrigel matrix, which is a basement membrane matrix extracted from EngelbretheCHolmeCSwarm mouse tumors rich in numerous growth factors and matrix components that supplies an environment favorable to the promotion of vessel growth.33 This may involve upregulation of VEGFR-1 expression by endothelial cells, such as is observed in healing wounds,8 enabling transduction of angiogenic signals.22 Collectively, these data demonstrate that both isoforms of VEGF-B potentiate physiological neovessel growth rather than initiate it.
The levels of EPCs were decreased in the blood of VEGFB167Tg mice when compared with wild-type littermates, implicating VEGF-B in aspects of EPC biology. At this stage we are not able to definitively demonstrate the cause of the reduced numbers of circulating EPCs in the VEGFB167Tg mice but suggest that this may not reflect negative regulation of EPC mobilization from the bone marrow. Vascular growth in the granulation tissue of healing wounds involves the recruitment of EPCs.45 We found that neovessel growth was equally increased in the Matrigel and wound-healing models in the VEGFB167Tg and VEGFB186Tg. We propose that this observation is inconsistent with a negative effect of VEGF-B on EPC mobilization and suggest that the reduced levels of EPCs in the blood of the VEGFB167Tg mice may be attributable to increased tissue recruitment or endothelial adherence of EPCs to the vascular endothelium in these animals. Unlike VEGF-B186, which is freely secreted, VEGF-B167 remains cell associated, where it potentially may exhibit an adhesion molecule-like action,46 and when expressed by endothelial cells as in the VEGFB167Tg mice may result in increased adherence of EPCs to the vasculature, thus removing them from the circulation. Tie-2 expression is upregulated in neovessels in healing wounds,32 and, accordingly, VEGFB167Tg mice and VEGFB186Tg mice displayed increased transgene expression in neovessel endothelium (compared with resting-state skin). This is associated with a significant increase in vessel density in granulation tissue of VEGFB167Tg and VEGFB186Tg wounds in comparison to wild-type littermate wounds. During wound healing, localized transgene expression may be sufficient to promote preferential recruitment and/or differentiation of EPCs from the circulation to the site of neovessel growth. In our wound healing and Matrigel studies, as in previous studies,15 the ability of VEGF-B to promote vascular growth in vivo may, therefore, represent effects on angiogenesis and/or vasculogenesis.
To determine the direct effects of VEGF-B on angiogenesis without the contribution of circulating EPCs to the vascular growth, we specifically selected the aortic explant assay in preference to other in vitro endothelial celleCbased angiogenesis assays because it more closely resembles the physiological angiogenesis process. Vessel growth in this system using rat aortic explants has been shown to directly involve angiogenesis,47 although the contribution of tissue resident EPCs in vessel growth in this assay has not been explored and should not be excluded. Similar to the results obtained in both the Matrigel assay and wound-healing studies, aortic explants from VEGFB167Tg and VEGFB186Tg mice exhibited increased vessel-like endothelial outgrowth in comparison to wild-type littermate aortic explants. Increased vessel growth in both VEGFB167Tg mice and VEGFB186Tg mice was caused by greater vessel-like endothelial sprouting rather than increased vessel branching. This result is in contrast to a previous study in a similar system using embryonic quail heart explants reporting a role for VEGF-B in vessel branching rather than vessel sprouting.48 In that study, the role of VEGF-B was determined by the addition of an antieCVEGF-B polyclonal antibody (C19; Santa Cruz Biotechnology). However, these data should be interpreted with caution because the ability of this antibody to neutralize VEGF-B signaling through its receptor has not, to our knowledge, been specifically assessed. Furthermore, this antibody was designed against a peptide region mapping to the carboxyl terminus of VEGF-B and, therefore, can only bind to a single isoform of VEGF-B,46 which we have determined to be VEGF-B167 by immunoblotting of recombinant VEGF-B isoforms (A.M., M.C., G.K., unpublished observation, 2003). In our studies, the action of both VEGF-B transgenes on potentiation of vessel growth was completely blocked in the presence of antieCVEGF-B mAb (clone 2H10), which is shown to neutralize Vegf-B signaling through VEGFR-1, confirming a direct effect of both VEGF-B isoforms in the potentiation of angiogenesis rather than inherent differences in the endothelial responsiveness of aortic explants between wild-type littermates and transgenic animals. Although the contribution of tissue-resident EPCs to vascular growth in this assay remains undetermined, we conclude that the differences in vessel growth between transgenic and wild-type explants cannot be solely attributable to increased levels of tissue-resident EPCs in the transgenic explants because VEGF-B activity was required during the assay to potentiate vessel growth. Support for a direct action of both VEGF-B transgenes on potentiation of angiogenesis comes from the evidence of a diffusible factor in the potentiation of angiogenesis from VEGFB167Tg and VEGFB186Tg aortic explants to VegfbeC/eC explants in a coculture system. From the early in vitro characterization of the VEGF-B isoforms,46 we had predicted that only the VEGF-B186 should be able to signal to the remote VegfbeC/eC explants because VEGF-B167 has been shown to be retained by the secreting cell. It is possible that the VEGF-B transgene effect involves a second messenger, but this is outside the scope of the present study.
The vascular permeability activity of VEGF-A presents a problem for the use of VEGF-A as a therapeutic agent for promoting tissue revascularization because it also promotes edema. This is illustrated by studies where overexpression of VEGF-A in the skin of mice induced vascular leakage that was evident by disruption of skin architecture.49 In contrast, there was no evidence of edema in the skin of VEGFB167Tg and VEGFB186Tg mice by histological examination. Injections of recombinant VEGF-B167 protein failed to stimulate vascular leakage in the Miles assay when used at concentration up to 10-fold excess of a concentration of VEGF-A, which potently induced vascular leakage in the same animal (wild-type mice, FVB/N strain). This indicates that, unlike VEGF-A, VEGF-B does not promote vascular leakage and indicates that VEGF-B may provide a more favorable therapy for promoting/potentiating tissue revascularization, as it should not have edematous side effects.
In conclusion, our data demonstrate that both isoforms of VEGF-B potentiate, rather than initiate, postnatal vessel growth in vivo. We also present in vitro evidence to show that VEGF-B can potentiate angiogenesis. Because VEGF-B does not appear to promote vascular leakage, its use as a therapeutic agent to promote vascular growth should not be associated with edematous side effects, unlike related factors VEGF-A and PlGF. These data suggest that VEGF-B may provide an effective therapy to promote/enhance vascular growth in diseases associated with impaired revascularization and ischemia, such as ischemic heart disease.
Acknowledgments
This work was supported by a research grant from AMRAD Corporation Ltd, Richmond, Victoria, Australia.
The Queensland Institute of Medical Research and AMRAD Corporation Ltd have a formal agreement whereby both parties share revenue from commercialization of VEGF-B as a potential therapeutic agent or target.
References
Cao Y, Hong A, Schulten H, Post MJ. Update on therapeutic neovascularization. Cardiovasc Res. 2005; 65: 639eC648.
Rissanen TT, Rutanen J, Yla-Herttuala S. Gene transfer for therapeutic vascular growth in myocardial and peripheral ischemia. Adv Genet. 2004; 52: 117eC164.
Eriksson U, Alitalo K. VEGF receptor 1 stimulates stem-cell recruitment and new hope for angiogenesis therapies. Nat Med. 2002; 8: 775eC777.
Tammela T, Enholm B, Alitalo K, Paavonen K. The biology of vascular endothelial growth factors. Cardiovasc Res. 2005; 65: 550eC563.
Neufeld G, Kessler O, Herzog Y. The interaction of neuropilin-1 and neuropilin-2 with tyrosine-kinase receptors for VEGF. Adv Exp Med Biol. 2002; 515: 81eC90.
Terman BI, Dougher-Vermazen M, Carrion ME, Dimitrov D, Armellino DC, Gospodarowicz D, Bohlen P. Identification of the KDR tyrosine kinase as a receptor for vascular endothelial cell growth factor. Biochem Biophys Res Commun. 1992; 187: 1579eC1586.
Bellomo D, Headrick JP, Silins GU, Paterson CA, Thomas PS, Gartside M, Mould A, Cahill MM, Tonks ID, Grimmond SM, Townson S, Wells C, Little M, Cummings MC, Hayward NK, Kay GF. Mice lacking the vascular endothelial growth factor-B gene (Vegfb) have smaller hearts, dysfunctional coronary vasculature, and impaired recovery from cardiac ischemia. Circ Res. 2000; 86: e29eCe35.
Carmeliet P, Moons L, Luttun A, Vincenti V, Compernolle V, De Mol M, Wu Y, Bono F, Devy L, Beck H, Scholz D, Acker T, DiPalma T, Dewerchin M, Noel A, Stalmans I, Barra A, Blacher S, Vandendriessche T, Ponten A, Eriksson U, Plate KH, Foidart JM, Schaper W, Charnock-Jones DS, Hicklin DJ, Herbert JM, Collen D, Persico MG. Synergism between vascular endothelial growth factor and placental growth factor contributes to angiogenesis and plasma extravasation in pathological conditions. Nat Med. 2001; 7: 575eC583.
Luttun A, Brusselmans K, Fukao H, Tjwa M, Ueshima S, Herbert JM, Matsuo O, Collen D, Carmeliet P, Moons L. Loss of placental growth factor protects mice against vascular permeability in pathological conditions. Biochem Biophys Res Commun. 2002; 295: 428eC434.
Mould AW, Tonks ID, Cahill MM, Pettit AR, Thomas R, Hayward NK, Kay GF. Vegfb gene knockout mice display reduced pathology and synovial angiogenesis in both antigen-induced and collagen-induced models of arthritis. Arthritis Rheum. 2003; 48: 2660eC2669.
Yoshida A, Anand-Apte B, Zetter BR. Differential endothelial migration and proliferation to basic fibroblast growth factor and vascular endothelial growth factor. Growth Factors. 1996; 13: 57eC64.
Six I, Kureishi Y, Luo Z, Walsh K. Akt signaling mediates VEGF/VPF vascular permeability in vivo. FEBS Lett. 2002; 532: 67eC69.
Sawano A, Takahashi T, Yamaguchi S, Aonuma M, Shibuya M. Flt-1 but not KDR/Flk-1 tyrosine kinase is a receptor for placenta growth factor, which is related to vascular endothelial growth factor. Cell Growth Differ. 1996; 7: 213eC221.
Oura H, Bertoncini J, Velasco P, Brown LF, Carmeliet P, Detmar M. A critical role of placental growth factor in the induction of inflammation and edema formation. Blood. 2003; 101: 560eC567.
Silvestre JS, Tamarat R, Ebrahimian TG, Le-Roux A, Clergue M, Emmanuel F, Duriez M, Schwartz B, Branellec D, Levy BI. Vascular endothelial growth factor-B promotes in vivo angiogenesis. Circ Res. 2003; 93: 114eC123.
Fong GH, Zhang L, Bryce DM, Peng J. Increased hemangioblast commitment, not vascular disorganization, is the primary defect in flt-1 knock-out mice. Development. 1999; 126: 3015eC3025.
Lalla RV, Boisoneau DS, Spiro JD, Kreutzer DL. Expression of vascular endothelial growth factor receptors on tumor cells in head and neck squamous cell carcinoma. Arch Otolaryngol Head Neck Surg. 2003; 129: 882eC888.
Casella I, Feccia T, Chelucci C, Samoggia P, Castelli G, Guerriero R, Parolini I, Petrucci E, Pelosi E, Morsilli O, Gabbianelli M, Testa U, Peschle C. Autocrine-paracrine VEGF loops potentiate the maturation of megakaryocytic precursors through Flt1 receptor. Blood. 2003; 101: 1316eC1323.
Sawano A, Iwai S, Sakurai Y, Ito M, Shitara K, Nakahata T, Shibuya M. Flt-1, vascular endothelial growth factor receptor 1, is a novel cell surface marker for the lineage of monocyte-macrophages in humans. Blood. 2001; 97: 785eC791.
Marti HJ, Bernaudin M, Bellail A, Schoch H, Euler M, Petit E, Risau W. Hypoxia-induced vascular endothelial growth factor expression precedes neovascularization after cerebral ischemia. Am J Pathol. 2000; 156: 965eC976.
Tomanek RJ, Lund DD, Yue X. Hypoxic induction of myocardial vascularization during development. Adv Exp Med Biol. 2003; 543: 139eC149.
Hiratsuka S, Maru Y, Okada A, Seiki M, Noda T, Shibuya M. Involvement of Flt-1 tyrosine kinase (vascular endothelial growth factor receptor-1) in pathological angiogenesis. Cancer Res. 2001; 61: 1207eC1213.
Luttun A, Tjwa M, Carmeliet P. Placental growth factor (PlGF) and its receptor Flt-1 (VEGFR-1): novel therapeutic targets for angiogenic disorders. Ann N Y Acad Sci. 2002; 979: 80eC93.
Rissanen TT, Markkanen JE, Gruchala M, Heikura T, Puranen A, Kettunen MI, Kholova I, Kauppinen RA, Achen MG, Stacker SA, Alitalo K, Yla-Herttuala S. VEGF-D is the strongest angiogenic and lymphangiogenic effector among VEGFs delivered into skeletal muscle via adenoviruses. Circ Res. 2003; 92: 1098eC1106.
Bhardwaj S, Roy H, Gruchala M, Viita H, Kholova I, Kokina I, Achen MG, Stacker SA, Hedman M, Alitalo K, Yla-Herttuala S. Angiogenic responses of vascular endothelial growth factors in periadventitial tissue. Hum Gene Ther. 2003; 14: 1451eC1462.
Scotney P, MacKenzie A, Maccarone P, Fabri L, Scrofani S, Gooley P, Nash A. Human vascular endothelial growth factor B: characterization of recombinant isoforms and generation of neutralizing monoclonal antibodies. Clin Exp Pharmacol Physiol. 2002; 29: 1024eC1029.
Makinen T, Jussila L, Veikkola T, Karpanen T, Kettunen MI, Pulkkanen KJ, Kauppinen R, Jackson DG, Kubo H, Nishikawa S, Yla-Herttuala S, Alitalo K. Inhibition of lymphangiogenesis with resulting lymphedema in transgenic mice expressing soluble VEGF receptor-3. Nat Med. 2001; 7: 199eC205.
Achen MG, Roufail S, Domagala T, Catimel B, Nice EC, Geleick DM, Murphy R, Scott AM, Caesar C, Makinen T, Alitalo K, Stacker SA. Monoclonal antibodies to vascular endothelial growth factor-D block its interactions with both VEGF receptor-2 and VEGF receptor-3. Eur J Biochem. 2000; 267: 2505eC2515.
Mizushima S, Nagata S. pEF-BOS, a powerful mammalian expression vector. Nucleic Acids Res. 1990; 18: 5322.
Schlaeger TM, Bartunkova S, Lawitts JA, Teichmann G, Risau W, Deutsch U, Sato TN. Uniform vascular-endothelial-cell-specific gene expression in both embryonic and adult transgenic mice. Proc Natl Acad Sci U S A. 1997; 94: 3058eC3063.
Brown AL, Kay GF. Bex1, a gene with increased expression in parthenogenetic embryos, is a member of a novel gene family on the mouse X chromosome. Hum Mol Genet. 1999; 8: 611eC619.
Wong AL, Haroon ZA, Werner S, Dewhirst MW, Greenberg CS, Peters KG. Tie2 expression and phosphorylation in angiogenic and quiescent adult tissues. Circ Res. 1997; 81: 567eC574.
Passaniti A, Taylor RM, Pili R, Guo Y, Long PV, Haney JA, Pauly RR, Grant DS, Martin GR. A simple, quantitative method for assessing angiogenesis and antiangiogenic agents using reconstituted basement membrane, heparin, and fibroblast growth factor. Lab Invest. 1992; 67: 519eC528.
Gatto C, Rieppi M, Borsotti P, Innocenti S, Ceruti R, Drudis T, Scanziani E, Casazza AM, Taraboletti G, Giavazzi R. BAY 12eC9566, a novel inhibitor of matrix metalloproteinases with antiangiogenic activity. Clin Cancer Res. 1999; 5: 3603eC3607.
Masson VV, Devy L, Grignet-Debrus C, Bernt S, Bajou K, Blacher S, Roland G, Chang Y, Fong T, Carmeliet P, Foidart JM, Noel A. Mouse aortic ring assay: a new approach of the molecular genetics of angiogenesis. Biol Proced Online. 2002; 4: 24eC31.
Heil M, Mitnacht-Krauss R, Issbrucker K, van den Heuvel J, Dehio C, Schaper W, Clauss M, Weich HA. An engineered heparin-binding form of VEGF-E (hbVEGF-E). Biological effects in vitro and mobilizatiion of precursor cells. Angiogenesis. 2003; 6: 201eC211.
Wise LM, Veikkola T, Mercer AA, Savory LJ, Fleming SB, Caesar C, Vitali A, Makinen T, Alitalo K, Stacker SA. Vascular endothelial growth factor (VEGF)-like protein from orf virus NZ2 binds to VEGFR2 and neuropilin-1. Proc Natl Acad Sci U S A. 1999; 96: 3071eC3076.
Padmanabhan V, Callas P, Philips G, Trainer TD, Beatty BG. DNA replication regulation protein Mcm7 as a marker of proliferation in prostate cancer. J Clin Pathol. 2004; 57: 1057eC1062.
Gunningham SP, Currie MJ, Han C, Robinson BA, Scott PA, Harris AL, Fox SB. VEGF-B expression in human primary breast cancers is associated with lymph node metastasis but not angiogenesis. J Pathol. 2001; 193: 325eC332.
Aase K, Lymboussaki A, Kaipainen A, Olofsson B, Alitalo K, Eriksson U. Localization of VEGF-B in the mouse embryo suggests a paracrine role of the growth factor in the developing vasculature. Dev Dyn. 1999; 215: 12eC25.
Masuda H, Asahara T. Post-natal endothelial progenitor cells for neovascularization in tissue regeneration. Cardiovasc Res. 2003; 58: 390eC398.
Aase K, von Euler G, Li X, Ponten A, Thoren P, Cao R, Cao Y, Olofsson B, Gebre-Medhin S, Pekny M, Alitalo K, Betsholtz C, Eriksson U. Vascular endothelial growth factor-B-deficient mice display an atrial conduction defect. Circulation. 2001; 104: 358eC364.
Sun Y, Jin K, Childs JT, Xie L, Mao XO, Greenberg DA. Increased severity of cerebral ischemic injury in vascular endothelial growth factor-b-deficient mice. J Cereb Blood Flow Metab. 2004; 24: 1146eC1152.
Prior BM, Lloyd PG, Ren J, Li H, Yang HT, Laughlin MH, Terjung RL. Time course of changes in collateral blood flow and isolated vessel size and gene expression after femoral artery occlusion in rats. Am J Physiol Heart Circ Physiol. 2004; 287: H2434eCH2447.
Keswani SG, Katz AB, Lim FY, Zoltick P, Radu A, Alaee D, Herlyn M, Crombleholme TM. Adenoviral mediated gene transfer of PDGF-B enhances wound healing in type I and type II diabetic wounds. Wound Repair Regen. 2004; 12: 497eC504.
Olofsson B, Pajusola K, von Euler G, Chilov D, Alitalo K, Eriksson U. Genomic organization of the mouse and human genes for vascular endothelial growth factor B (VEGF-B) and characterization of a second splice isoform. J Biol Chem. 1996; 271: 19310eC19317.
Kruger EA, Duray PH, Tsokos MG, Venzon DJ, Libutti SK, Dixon SC, Rudek MA, Pluda J, Allegra C, Figg WD. Endostatin inhibits microvessel formation in the ex vivo rat aortic ring angiogenesis assay. Biochem Biophys Res Commun. 2000; 268: 183eC191.
Tomanek RJ, Holifield JS, Reiter RS, Sandra A, Lin JJ. Role of VEGF family members and receptors in coronary vessel formation. Dev Dyn. 2002; 225: 233eC240.
Larcher F, Murillas R, Bolontrade M, Conti CJ, Jorcano JL. VEGF/VPF overexpression in skin of transgenic mice induces angiogenesis, vascular hyperpermeability and accelerated tumor development. Oncogene. 1998; 17: 303eC311.(Arne W. Mould, Sonia A. G)
AMRAD Corp Ltd (A.N., P.S.), Richmond, Victoria, Australia.
Abstract
Vascular endothelial growth factors (VEGFs) play significant roles in endothelial growth, survival, and function, and their potential use as therapeutic agents to promote the revascularization of ischemic tissues in being avidly explored. VEGF-A has received most attention, as it is a potent stimulator of vascular growth. Results in clinical trials of VEGF-A as a therapeutic agent have fallen short of high expectations because of serious edematous side effects caused by its activity in promoting vascular permeability. VEGF-B, a related factor, binds some of the VEGF-A receptors but not to VEGF receptor 2, which is implicated in the vascular permeability promoting activity of VEGF-A. Despite little in vitro evidence to date for the ability of Vegf-B to directly promote angiogenesis, recent data indicate that it may promote postnatal vascular growth in mice, suggesting that it may have potential therapeutic application. We have specifically studied the effects of VEGF-B on vascular growth in vivo and on angiogenesis in vitro by analyzing transgenic mice in which individual isoforms (VEGFB167Tg and VEGFB186Tg) of VEGF-B are overexpressed in endothelial cells. VEGFB167Tg and VEGFB186Tg mice displayed enhanced vascular growth in the Matrigel assay in vivo and during cutaneous wound healing. In the aortic explant assay, explants from VEGFB167Tg and VEGFB186Tg mice displayed elevated vascular growth, suggesting a direct effect of VEGF-B isoforms in potentiating angiogenesis. These data support the use of VEGF-B as a therapeutic agent to promote vascular growth, in part, by potentiating angiogenesis. Furthermore, the lack of vascular permeability activity associated with either transgenic overexpression of the VEGF-B gene in endothelial cells or application of VEGF-B protein to the skin of mice in the Miles assay indicates that use of VEGF-B as a therapy should not be associated with edematous side effects.
Key Words: vascular endothelial growth factor-B growth factor angiogenesis endothelium
Introduction
Promoting the revascularization of ischemic tissues by local or systemic administration of growth factors presents an attractive potential therapy for the treatment of cardiovascular disease. Numerous growth factors have been shown to stimulate angiogenesis in vitro and blood vessel growth in vivo1 but have failed to meet relevant end points in clinical trials and/or have been associated with adverse side effects. For example, vascular endothelial growth factor (VEGF)-A, which is a potent angiogenic factor and has been considered a prime candidate for promoting revascularization of ischemic tissues, causes tissue edema and hypotension because of its potent vascular permeability activity.2 As a result, there is a need to identify factors that can promote vessel growth in the absence of other activities to provide more selective treatments.
Signaling through VEGF receptor (VEGFR)-1 has emerged as a promising therapeutic strategy for use in promoting tissue revascularization.3 VEGF-A, VEGF-B, and placental growth factor (PlGF) bind to VEGFR-1.4 These factors promote vascular growth in vivo and are proposed as potential therapeutic agents for the treatment of ischemia.3 The suitability of these factors as therapeutic agents will not depend only on their ability to promote vessel growth but also on their side effect profile. VEGF-A, VEGF-B, and PlGF also bind to neuropilin-1 (in an isoform-specific manner)5 and VEGF-A additionally binds to VEGFR-2.6 Gene knockout studies in mice indicate that, despite the overlapping receptor binding profiles, each of these factors may have distinct roles in vascular signaling.7eC10 VEGF-A directly stimulates endothelial cell proliferation and migration.11 It is also a potent inducer of vascular permeability12; hence, it is also known as vascular permeability factor. In contrast, PlGF does not directly promote endothelial cell proliferation or migration and is only a weak stimulator of vascular permeability in the Miles assay.8,13 Characterization of PlGF gene knockout mice has revealed that PlGF is required for VEGF-AeCinduced endothelial cell proliferation and vascular permeability.8 Furthermore, transgenic overexpression of PlGF under the control of the K14 promoter leads to both vascular growth and edema in the skin of mice, which, in part, may be explained by the induction of expression of VEGF-A.14 This indicates that PlGF signaling is closely linked to that of VEGF-A, suggesting that, similar to VEGF-A, the use PlGF as a therapeutic agent to promote tissue revascularization may be associated with an edematous side effect.
VEGF-B promotes vascular growth in adult mice when administered either in the form of recombinant protein in Matrigel plugs or when expressed in situ by adenoviral vector at sites of experimentally induced tissue ischemia.15 This activity of VEGF-B was shown to be mediated by VEGFR-1, a receptor tyrosine kinase expressed by endothelial cells16 and a variety of other cell types,17eC19 and which is induced in ischemic tissue via a hypoxia-inducible factor-1aeCdependant mechanism.20,21 Signaling through VEGFR-1 has also been implicated in postnatal vessel growth in other model systems.22,23 There is no evidence to suggest that VEGF-B is either required for or can potentiate the activity of VEGF-A, and there have been no reports that VEGF-B can promote vascular permeability. On the contrary, we have previously reported that the ability of VEGF-A to induce vascular leakage in the Miles assay is not impaired in Vegfb gene knockout mice.10 Collectively, these data indicate that VEGF-B may be able to promote vascular growth without induction of vascular leakage and, as a therapeutic agent, may have a better side-effect profile than that predicted for VEGF-A and PlGF. However, not all reports support such therapeutic potential of using VEGF-B, as other studies found VEGF-B to be ineffectual at stimulating vessel growth in muscle and periadventitia when delivered by adenoviral vector.24,25 This may reflect tissue-specific differences in effectiveness of VEGF-B on vascular signaling and/or variations in the localized environment, such as the presence of other growth factors. These conflicting results from studies investigating the ability of VEGF-B to promote vessels growth are confounded by a lack of in vitro evidence supporting a direct angiogenic activity of VEGF-B.
To clarify the function of VEGF-B isoforms in aspects of postnatal vessel growth, we have generated transgenic mice in which VEGF-B167 and VEGF-B186 expression is driven by the endothelial celleCspecific Tie2 promoter/enhancer. Analysis of these mice demonstrates a role for both isoforms of VEGF-B in the potentiation, rather than the initiation, of postnatal vessel growth in vivo and of angiogenesis in vitro. Further, the absence of edema in the skin of transgenic animals in combination with an inability of recombinant VEGF-B to induce vascular permeability in the Miles assay indicates that VEGF-B does not induce vascular leakage.
Materials and Methods
Recombinant VEGF-B Expression and Refolding
Recombinant VEGF-B isoforms were produced essentially as previously reported.26 The coding regions of the cDNA corresponding to amino acid residues 10 to 108 and 1 to 167 of both human and mouse VEGF-B were cloned into the pET-15b vector (EMD Biosciences, San Diego, Calif) and transformed into Escherichia coli strain BL21-CodonPlus-(DE3)-RP (Stratagene, La Jolla, Calif). Recombinant protein expression was induced mid-log phase in Superbroth (30% wt/vol tryptone, 15% wt/vol yeast extract, 0.5% wt/vol NaCl) with 10 nmol/L isopropyl eCD-thiogalactoside. Inclusion bodies were isolated and the material purified under reducing denaturing conditions (6 mol/L guanidine hydrochloride, 0.1 mol/L NaH2PO4, 10 mmol/L -mercaptoethanol) by nickel metal affinity chromatography (MAC). The purified material was reduced and refolded by dialysis in redox solution as previously described.26 The refolded material was acidified by dialysis in 0.1 mol/L acetic acid and purified by reverse-phase high-performance liquid chromatography. The N-terminal His6 tag used for nickel MAC was subsequently removed by cleavage with Genenase I (New England Biolabs, Beverly, Mass).
Human VEGFR-1 Chimeric Receptor CelleCBased Assay
Development of the human VEGFR-1 celleCbased assay has been previously reported.27 A cDNA encoding a chimeric receptor consisting of the extracellular domain of human VEGFR-1 fused to the transmembrane and intracellular domains of mouse erythropoietin receptor was generated by introducing a BglII site into the VEGFR-1 cDNA before the sequence encoding the transmembrane domain, followed by ligation of a BglIIeCNotI fragment consisting of the transmembrane and intracellular domains of erythropoietin receptor.28 The cDNA was subcloned into the pEF-BOS expression vector29 and transfected by electroporation into the interleukin (IL)-3eCdependent pre-B cell line, BaF3, together with pcDNA3.1(+) zeo vector (Invitrogen, Carlsbad, Calif) encoding Zeocin resistance. The transfected cells were selected in complete media (DMEM [Invitrogen]; 5% FCS, 50 U/mL penicillin, 50 e/mL streptomycin, 2 mmol/L GlutaMAX [Invitrogen]) supplemented with 250 e/mL Zeocin (Invitrogen), and a Zeocin-resistant line able to proliferate in response to VEGF-A or VEGF-B in the absence of IL-3 was subsequently isolated and cloned by limit dilution. hVEGFR1/EpoR/BaF3 clones were routinely cultured in complete media supplemented with 5 ng/mL VEGF-B167.
For the assay of VEGFR-1 ligand activity, selected ligands were diluted into 96-well flat bottom tissue culture plates (TPP, Trasadingen, Switzerland). hVEGFR1/EpoR/BaF3 cells, washed 5 times in PBS, were plated at 6x105/well to give a final assay volume of 100 mL. For ligand neutralization assays, a fixed concentration of ligand was preincubated with antibody for 1 hour before the addition of hVEGFR1/EpoR/BaF3 cells. Cells were incubated for 72 hours at 37°C, 10% CO2, in a humidified container and proliferation assessed using an methanethiosulfonate (MTS) dye reduction assay (Promega, Madison, Wis) with the colorimetric change measured at a wavelength of 450 nm (Wallac VICTOR microplate reader, PerkinElmer, Boston, Mass).
Generation of Transgenic Mice
All mice in this study were treated in accordance with the Australian National Health and Medical Research Council guidelines for the care and use of experimental animals. Transgenic mice (FVB/N-TgN[Tie2-VEGFB167]1Gfk [henceforth referred to as VEGFB167Tg] and FVB/N-TgN[Tie2-VEGFB186]1Gfk [henceforth referred to as VEGFB186Tg]) were generated by standard techniques in FVB/N strain mice using the expression constructs shown in Figure 1a. The constructs were designed to result in expression of each isoform of VEGF-B (GenBank Nos. U43368 and U43369) under the control of an altered Tie2 enhancer/promoter cassette.30 This enhancer/promoter combination has been reported to drive constitutive endothelial celleCspecific transgene expression.30 Transgenic animals were genotyped by PCR amplification of tail-tip DNA using primers (F1 5'-ctg tct ccc agc ctg atg cc-3' and R1 5'-ctg ggt tga gct cta agc cc-3') specific for the human VEGFB cDNA sequence. The lines of transgenic mice reported here were chosen for the studies after analysis by reverse transcription RT-PCR and immunohistochemistry to confirm that the transgene was being appropriately expressed (see below).
The number of transgene copies incorporated in each transgenic mouse line was estimated from genomic DNA preparations by semiquantitative PCR (Figure 1b). The primers used to amplify the transgenes were the F1 and R1 primers (as above) multiplexed with primers that specifically amplify a region from the endogenous mouse Vegfb gene.7 Briefly, PCR products from increasing cycles of amplification were electrophoresed through an agarose/Tris-acetate-EDTA gel stained with Vistra-green (Amersham Pharmacia Biotech, Uppsala, Sweden). PCR products were visualized by fluorescence imaging on an Amersham Pharmacia Biotech Typhoon 8600 laser scanner (Amersham Pharmacia Biotech). ImageQuant software was used for image analysis and quantification of the PCR products. The gene dosage was estimated by measuring band intensity of the PCR products at their threshold detection levels relative to amplification cycle in comparison to the endogenous diploid Vegfb gene sequence.
Transgene expression levels were compared in RNA isolated from lung tissue for the VEGFB167Tg and VEGFB186Tg mice by semiquantitative RT-PCR and referenced to the level of Hprt expression, as previously described31 (Figure 1c), for transgene copy number estimation. The primers used for RT-PCR amplification of the transgenes were F2 5'-cgc cgc act cct gca gct-3' and R2 5'-gct ggc agg tag cgc gag-3'.
Expression and localization of human VEGF-B protein in VEGFB167Tg and VEGFB186Tg mice was determined by immunofluorescence (IF) on cryosections of normal heart (Figure 1d) and by immunohistochemistry (IHC) on paraffin sections of healing cutaneous wounds (Figure 1e). To specifically detect transgene expressed VEGF-B protein in cryosections of heart using the antieCVEGF-B monoclonal antibody (mAb) (clone 2H10),26 which detects both human and mouse VEGF-B; transgenic mice were crossed with VegfbeC/eC mice7 to generate VEGFB167Tg;VegfbeC/eC and VEGFB167Tg;VegfbeC/eC genotype animals. Similar sections from wild-type (C57BL/6J strain) and VegfbeC/eC mice (backcrossed for 12 generations onto the C57BL/6J background) were used as controls. Sections prepared from cutaneous wounds were chosen because Tie-2 expression has been shown to be upregulated in newly formed vessels within the granulation tissue and existing vessels at wound edges.32 For IHC analysis using antieChuman VEGF-B polyclonal Ab (N-19, Santa Cruz Biotechnology, Santa Cruz, Calif), full-depth cutaneous wound samples including surrounding tissue (at 5 days postwounding) were fixed by immersion in 4% paraformaldehyde (PFA) in PBS overnight at 4°C and paraffin embedded.
Matrigel Assay of Neovessel Growth in Vivo
A modified in vivo assay of neovessel growth described by Passaniti et al33 was used. Briefly, 0.5 mL of ice-cold Matrigel matrix (BD Biosciences, Bedford, Mass) infused with 300 ng/mL of recombinant human basic fibroblast growth factor (FGF)34 (Invitrogen) was injected SC near the abdominal midline of VEGFB167Tg, VEGFB186Tg, and wild-type littermate mice (8- to 10-week-old females). Soon after injection, the Matrigel solidified to form a semisolid plug. The mice were then maintained under normal conditions for 6 days, when the experiment was terminated and the angiogenic response studied. The mice were euthanized by CO2 asphyxiation and the Matrigel plugs with the supporting skin and peritoneal lining intact were dissected and frozen in OCT compound (Tissue-Tek, Sakura Finetek, Torrance, Calif). Vessels were identified by CD31 IHC on cryosections and neovessel growth quantified by measuring the density of vessels at the Matrigel/skin interface as described below. As macrophages also express CD31, single cells positive for CD31 were counted only if they surrounded a visible vessel lumen (ie, capillaries).
Assessment of Neovessel Growth in Healing Wounds
VEGFB167Tg, VEGFB186Tg, and wild-type littermate mice (8- to 10-week-old males and females) were anesthetized by IP injection of anesthetic (0.8 mg/kg fentanyl citrate, 25 mg/kg fluanisone, 12.5 mg/kg midazolam hydrochloride) and 2 full-depth cutaneous wounds were made on the backs using a 6-mm Keyes dermal punch. Wounds were left uncovered, and after 5 or 10 days, the mice were euthanized by CO2 asphyxiation for wound analysis. Wounds were dissected with the surrounding skin intact and fixed by immersion in 4% PFA in PBS overnight at 4°C. Wounds were bisected at the widest point and embedded in paraffin. Vessels were identified on paraffin sections by CD31 IHC and vessel density determined by counting CD31-positive structures and single positive cells surrounding a visible vessel lumen (capillaries) per microscopic field within the granulation tissue.
Aortic Explant Assay
A modified version of the mouse aortic ring assay (Masson et al35) was used to assess angiogenesis in aortic explants from VEGFB167Tg and VEGFB186Tg mice. Essentially, the descending aorta from VEGFB167Tg, VEGFB186Tg, and wild-type littermate mice (C57BL/6JxFVB/N strain, F1 progeny) were dissected and placed into ice-cold DMEM. Aortas were then placed in a Petri dish containing DMEM and the periaortic fibroadipose tissue removed under a stereo dissection microscope using fine-tipped forceps and microdissection scissors. The aortas were then longitudinally bisected and cut into 1-mm squares. After washing in fresh DMEM, individual squares of aorta tissue were embedded in a drop (25 e蘈) of rat tail collagen solution35 that was placed centrally in the bottom of a well of a 24-well flat bottom tissue culture plate (BD Labware, Franklin Lakes, NJ). The collagen solution was polymerized by placing the tissue culture plate in a humid incubator for 10 minutes at 37°C. Assay medium35 (300 mL of MCDB131 [Invitrogen] supplemented with 2.5% serum from VegfbeC/eC mice,7 25 mmol/L NaHCO3, 1% glutamine, and penicillin/streptomycin) was then added to each well.
In some experiments, antieCVEGF-B mAb (clone 2H10) or an isotype control mAb (IgG2a, clone C44) was added to the assay medium to a final concentration of 50 e/mL. In other experiments, aortic explants from VEGFB167Tg, VEGFB186Tg, and wild-type littermate mice were cocultured in the same collagen gel with aortic explants from VegfbeC/eC mice (C57BL/6J strain). Cultures were incubated in a humid incubator at 37°C/5% CO2 for 6 to 8 days for optimal endothelial vesseleClike growth. For analysis, collagen gels containing the aortic explants were fixed in situ with 2% PFA in PBS (20 minutes at room temperature [RT]) and endothelial vesseleClike outgrowths identified by immunofluorescence using fluorescein isothiocyanate (FITC)-conjugated BS-1 lectin (10 e/mL in 10 mmol/L HEPES-buffered saline, pH 7.5; Sigma-Aldrich) or by inverted light microscopy. Angiogenesis was assessed by counting the number of vessel-like sprouts and the number of vessel-like branches per aortic explant.
Endothelial Progenitor Cell Assay
The levels of circulating endothelial progenitors (EPCs) in VEGFB167Tg, VEGFB186Tg, and wild-type littermate mice were determined from 500 e蘈 of heparinized blood samples obtained by heart puncture as previously described.36
Miles Assay of Vascular Permeability
The vascular permeability at sites of intradermal injections of recombinant human VEGF-B167 (0.1 to 100 ng in 20 e蘈 of 0.01% BSA/PBS) were compared with a single dose of recombinant human VEGF-A165 (10 ng/site; R&D Systems, Minneapolis, Minn) using an adaptation of the Miles assay37 in FVB/N strain mice, as previously described.10
Antibodies and Immunohistochemistry
Endothelial cell proliferation in adult tissues was assessed in paraffin sections using MCM-7 IHC.38 Briefly, paraffin sections were dewaxed in xylene and hydrated through a graded series of alcohol. Antigen epitope was heat retrieved in 10 mmol/L citrate buffer, pH 6.0, and the sections permeabilized by incubation in 0.1% Triton X-100/PBS for 10 minutes at RT. Endogenous peroxidase activity and nonspecific protein-binding sites were sequentially blocked by incubation with 3% hydrogen peroxide/1% NaN3 in PBS for 30 minutes and 1% BSA/PBS for 20 minutes at RT. Sections were then incubated with 1 mg/mL of biotinylated MCM7 Ab2 (Clone 47DC141; NeoMarkers, Fremont, Calif) at RT for 90 minutes followed by incubation with streptavidin-conjugated horseradish peroxidase (1/600 dilution; DAKO Cytomation, Glostrup, Denmark) at RT for 30 minutes. Peroxidase activity was visualized using diaminobenzidine (DAB) and the slides lightly counter stained with hematoxylin. Endothelial cells and macrophages in tissue sections were identified by CD31 and F4/80 IHC. CD31 and F4/80 IHC were performed on paraffin sections of adult tissue and wound-healing samples and on cryosections of Matrigel samples. For paraffin-embedded sections, samples were dewaxed in xylene and hydrated through a graded series of alcohol. Antigen epitopes were retrieved by incubating the sections with 1 e/mL proteinase K in PBS at RT for 10 minutes or with 0.1% trypsin in PBS for 15 minutes at 37°C for CD31 and F4/80, respectively. Endogenous peroxidase activity was blocked by incubation with 3% hydrogen peroxide/1% NaN3 in PBS for 30 minutes at RT. To block nonspecific protein binding and tertiary antibody binding sites, the slides were sequentially incubated in 4% skim milk powder in PBS and 10% normal goat serum in PBS at RT for 15 and 30 minutes, respectively. Samples were then incubated overnight at 4°C with anti-mouse CD31 (1/800 dilution, clone MEC 13.3, rat IgG2a; Pharmingen, San Jose, Calif) or anti-mouse macrophage F4/80 (1/50 dilution of culture supernatant clone AC-1, rat IgG2a). Primary antibodies were detected by sequential incubation with polyclonal rabbit anti-rat IgG (1/750 dilution; Sigma-Aldrich, St Louis, Mo) and DAKO EnVision+ anti-rabbit peroxidase system (DAKO Cytomation) at RT for 30 minutes. Peroxidase activity was visualized using DAB and the slides lightly counter stained with hematoxylin. Cryosections (7 e) were air dried overnight, fixed with ice cold acetone for 5 minutes at RT, and air dried again. After rehydration in PBS, IHC for CD31 and F4/80 were performed essentially as described above for paraffin sections without the antigen retrieval.
VEGF-B protein expression was detected in heart tissue by IF. Cryosections were air dried overnight and fixed by immersion in ice-cold 50% acetone/50% methanol for 10 minutes. Endogenous biotin was blocked using DAKO Biotin Blocking System (DAKO Cytomation). Nonspecific protein-binding sites were blocked using 1% cold-water fish gelatin (Sigma-Aldrich) in PBS for 30 minutes at RT. Sections were then incubated with biotinylated antieCVEGF-B mAb (clone 2H10, 1 e/mL) in PBS overnight at 4°C and bound antibody detected by incubation with Rhodamine streptavidin conjugate (1/500 dilution in PBS) for 30 minutes at RT. IHC detection of human VEGF-B protein was confirmed on paraffin sections of cutaneous wounds, as previously described,39 using anti-human VEGF-B polyclonal peptide Ab (10 e/mL, N-19, Santa Cruz Biotechnology) with heat retrieval of antigen epitopes using citrate buffer, pH 6.0.
VEGFR-1 expression was detected by IHC on paraffin-tissue sections from VEGFB167Tg, VEGFB186Tg, and wild-type littermate mice. Following blocking of endogenous peroxidase activity and nonspecific protein-binding sites by sequential incubation with 3%H202/1% NaN3 in PBS and 4% skim milk power in PBS, dewaxed sections were incubated for 2 hours at RT with antieCVEGFR-1 rabbit polyclonal peptide Ab (4 e/mL,C-17), Santa Cruz Biotechnology). Sections were then washed in PBS and the bound antibody detected by incubation with DAKO EnVision+ anti-rabbit peroxidase system (DAKO Cytomation) at RT for 30 minutes. Peroxidase activity was visualized using DAB and the slides lightly counter stained with hematoxylin. As a negative control for all IF and IHC, primary antibody was omitted.
Results
Expression and Biological Activity of Human and Mouse VEGF-B Isoforms
Recombinant human (rh) and mouse VEGF-B10 to 108 and VEGF-B167 were expressed as N-terminal His6-tagged proteins in Escherichia coli. The monomeric proteins were purified by MAC and refolded by redox dialysis. The dimeric protein was separated from the monomeric material by reverse-phase high-performance liquid chromatography. SDS-PAGE analysis of the human and mouse isoforms is shown in Figure 2a.
Human and mouse VEGF-B isoforms were assessed for biological activity in the hVEGFR1/EpoR/BaF3 celleCbased assay. Both human and mouse VEGF-B10 to 108 and VEGF-B167 were shown to be active (Figure 2b), with human and mouse VEGF-B10 to 108 showing activity similar to that of human VEGF-A165 (EC50=0.1 nmol/L). In contrast, full-length human and mouse VEGF-B167 had 10- and 30-fold less activity, with EC50=1 and 3 nmol/L respectively.
Characterization of a Species Cross-Reactive VEGF-BeCSpecific Neutralizing mAB
The generation and preliminary characterization of the human VEGF-B neutralizing mouse mAb 2H10 has been described previously.26 To determine whether mAbs raised against human VEGF-B were cross-reactive with mouse VEGF-B and, thus, potentially of value in the analysis of VEGF-B activity in mouse models, the binding of 2H10 to the human and mouse VEGF-B was assessed by Western blot analysis (Figure 2c). Results presented in Figure 2c demonstrate that mAb 2H10 was able to bind to the human and mouse VEGF-B10 to 108 and VEGF-B167 isoforms. This interaction was confirmed using a variety of other approaches including ELISA and Biacore analysis (results not shown). 2H10 did not bind to reduced human and mouse VEGF-B, indicating that the 2H10 epitope is structural.
In addition to binding to both human and mouse VEGF-B, mAb 2H10 was shown to be a potent antagonist of both of these proteins. Results presented in Figure 2d demonstrate that mAb 2H10, but not the isotype control C44 antibody, inhibited human and mouse VEGF-BeCmediated proliferation of hVEGFR1/EpoR/BaF3 cells. The mAb appeared to be slightly more potent against human VEGF-B (IC50=1 nmol/L) than against mouse VEGF-B (IC50=3 nmol/L).
Evaluation of Transgene Expression in VEGFB167Tg and VEGFB186Tg Mice
A small number of viable individual transgenic founder animals with incorporation of the VEGFB transgenes were identified by PCR amplification of genomic DNA from tail-tip biopsies. More frequent transgene positive individuals were identified when mid-gestation embryonic stages were genotyped (data not shown), suggesting that high levels of expression of the transgene might be detrimental to development. Similarly, attempts to generate viable transgenic mice with ubiquitous overexpression of either VEGFB167 or VEGFB186 under the control of the HmgCoA gene promoter were unsuccessful, despite the identification of transgene positive mid-gestation embryos (data not shown). Of the viable transgenic mouse lines successfully derived, only 1 line each of VEGFB167Tg and VEGFB186Tg mice had detectable levels of transgene mRNA, as assessed by RT-PCR analysis of lung samples. We estimated by PCR amplification of the VEGFB transgenes in comparison with diploid endogenous Vegfb gene that approximately 16 copies of the VEGFB167 and 2 copies of the VEGFB186 transgenes were incorporated in the genome of these mouse lines (Figure 1b). Transgene expression levels, as estimated by RT-PCR from total RNA derived from lung tissue, showed that expression was approximately 12.7% and 1.5% of the Hprt expression level in the VEGFB167Tg and VEGFB186Tg mice, respectively (Figure 1c).
VEGFB167Tg and VEGFB186Tg mice are viable, fully fertile and appeared healthy. Aged-matched mice VEGFB167Tg and VEGFB186Tg mice (6- to 8-weeks old) displayed similar vessel densities in normal tissues to wild-type littermates, as assessed by platelet-endothelial celleCadhesion molecule (PECAM)-1 IHC on section of various organs (ie, liver, heart, and kidney; data not shown). Endothelial cell proliferation in vessels of the heart, liver, and kidney was also comparable between VEGFB167Tg, VEGFB186Tg, and their wild-type littermates, as determined by MCM-7 IHC (data not shown). IF staining for VEGF-B protein on cryosections of heart from VEGFB167Tg;VegfbeC/eC and VEGFB186Tg;VegfbeC/eC mice using antieCVEGF-B mAb (2H10) confirmed transgene protein expression. Cryosections of heart from wild-type (C57BL6J strain) and VegfbeC/eC mice were used as positive and negative controls, respectively. Positive staining was observed in wild-type heart sections, and no staining was observed in VegfbeC/eC heart sections, verifying specific VEGF-B protein staining (Figure 1d). Broad vascular staining was observed in the hearts of both VEGFB167Tg;VegfbeC/eC and VEGFB186Tg;VegfbeC/eC, consistent with Tie2 promoter/enhancereCdriven transgene expression in endothelial cells (Figure 1d). IHC using anti-human VEGF-B (rabbit polyclonal, N19; Santa Cruz Biotechnology) on paraffin sections of various tissues including heart, kidney, and liver and on 5-day-old, full-depth cutaneous wounds from both VEGFB167Tg and VEGFB186Tg mice was also performed as previously described39 but was less sensitive in detecting the transgene expression. Staining was only observed in the endothelium in blood vessels at wound edges (Figure 1e) and occasional newly formed vessels in the granulation tissue of VEGFB167Tg and VEGFB186Tg wounds. This indicates that transgene expression was increased in the vascular endothelium of vessels at wound edges and in neovessels during wound healing, which is consistent with previous reports showing upregulation of Tie-2 expression in this region in healing cutaneous wounds in mice.32 No staining of the vascular endothelium was observed in similar wound sections from wild-type mice (Figure 1e).
To determine whether the VEGFB transgenes altered expression of VEGFR-1, expression of the receptor was compared by IHC on paraffin sections of normal tissues from VEGFB167Tg and VEGFB186Tg mice with similar sections from wild-type littermates. In all genotypes, positive staining for VEGFR-1 was observed in endothelial cells in a restricted number of vessel types, predominantly small to medium veins and arteries. VEGFR-1 expression was not detected in capillary endothelium in any of the organs tested, including the heart, 1 of the sites where VEGFB transgene expression was detected (see Figure 1d). The intensity of endothelial staining and the types of vessels showing positive staining of endothelial cells were similar between the genotypes indicating that VEGFR-1 expression was not altered by VEGFB transgene expression (data not shown).
Assessment of Neovascular Growth in VEGFB167Tg and VEGFB186Tg Mice in the Matrigel Plug Assay and During Cutaneous Wound Healing
Vessel density at the Matrigel/skin interface of Matrigel plugs infused with 300 ng/mL bFGF was measured after 6 days in VEGFB167Tg, VEGFB186Tg, and wild-type littermate mice. Histological examination of H&E-stained cryosections of these plugs revealed an encapsulation of the Matrigel plug by granulation tissue and significant cellular invasion into the Matrigel matrix, largely consisting of macrophages as identified by F4/80 IHC, which was most concentrated at the periphery of the plug. Cellular invasion into the Matrigel matrix was associated with substantial angiogenesis as evidenced by in growth of PECAM-1eCpositive vessel structures into the Matrigel matrix from the surrounding granulation tissue (Figure 3a). Vessel density at the Matrigel/skin interface was significantly increased in Matrigel plugs from VEGFB167Tg (75% increase, P=0.0054 versus wild-type littermates) and VEGFB186Tg mice (52% increase, P=0.0068 versus wild-type littermates) when compared with vessel density in the same region of Matrigel plugs from wild-type littermates (Figure 3b). Levels of macrophages surrounding and invading the Matrigel plugs appeared similar in all genotypes (data not shown).
Apart from reproductive tissues, wound healing is 1 of the few nonpathological processes where neovessel growth is observed in the adult. During wound healing, neovessel growth is associated with the proliferative phase of wound healing and may facilitate the formation and, in later stages, the resolution of the granulation tissue. Based on previously published studies showing the kinetics of Tie-2 expression in wounds in mice,32 we compared angiogenic responses in granulation tissue at 5 and 10 days following full-depth cutaneous wounding among VEGFB167Tg, VEGFB186Tg, and wild-type littermates by measuring vessel density. The vessel density in the granulation tissue was significantly increased in both VEGFB167Tg and VEGFB186Tg compared with wild-type littermates at both 5 and 10 days postwounding (Figure 3c). The level of increase in vessel density at 10 days postwounding between the transgenic animals and the wild-type littermates was comparable to that observed between the same transgenic mice and wild-type littermates in the Matrigel plug assay (66% and 41% increase over wild-type vessel density for VEGFB167Tg and VEGFB186Tg, respectively; Figure 3c). The granulation tissue at 5 and 10 days postwound consisted primarily of macrophages and vascular tissue, as assessed by F4/80 (Figure 3d) and PECAM-1 (Figure 3e) IHC, respectively, and the levels of F4/80 positive cells in the granulation tissue appeared similar among genotypes when wounds from the various mouse genotypes were compared at the same time point (data not shown).
Decreased EPCs in the Circulation of VEGFB167Tg Mice
The levels of EPCs in the blood were significantly reduced in VEGFB167Tg mice (P=0.004) when compared with levels of these cells in the blood of wild-type littermates (Figure 4). The levels of EPCs in the blood of the VEGFB186Tg mice were not significantly different from the levels of these cells in the blood of wild-type littermates.
VEGFB167Tg and VEGFB186Tg Mice Display Increased Endothelial VesseleCLike Sprouting in the Aortic Explant Assay
The dynamics of endothelial vesseleClike outgrowth (angiogenesis) from cultured aortic explants was similar to that previously described.35 Vessel-like endothelial outgrowth was preceded by the migration of fibroblast-like cells from aortic explants into the surrounding collagen matrix, which was visible within the first 24 hours of culture (data not shown). Vessel-like endothelial outgrowths (identified by fluorescence microscopy after staining with FITC-conjugated BS-1 lectin; Figure 5a) from the aortic explants were first observed after 4 days in culture and maximal vessel growth was observed 4 days later, as assessed by both the number of vessels sprouts and the complexity of vessel branching (as measured as a ratio of the number of branches to the number of vessel sprouts). VEGFB167Tg and VEGFB186Tg explants displayed significantly greater numbers of vessel sprouts when compared with explants from wild-type littermates (Figure 5b). Branching complexity was comparable among the different mouse genotypes (wild-type littermates=1.44±0.14, VEGFB167Tg=1.56±0.07, and VEGFB186Tg=1.70±0.09 branches/vessel like sprout). The addition of antieCVEGF-B mAb (clone 2H10) to the culture medium significantly reduced the number of vessel sprouts in VEGFB167Tg and VEGFB186Tg aortic explant cultures when compared with their respective isotype mAb (clone C44)eCtreated control explant cultures (Figure 5c). In the presence of antieCVEGF-B antibody, the number of vessel sprouts in VEGFB167Tg and VEGFB186Tg aortic explants cultures were equivalent to that observed in wild-type littermate aortic explant cultures treated with isotype control mAb. In contrast, addition of antieCVEGF-B mAb was without effect on the level of vessel sprouts in wild-type littermate explants when compared with similar explants treated with the isotype control mAb. The complexity of vessel branching was similar among all genotypes regardless of the mAb treatment. In further studies, VegfbeC/eC aortic explants cocultured with VEGFB167Tg and VEGFB186Tg aortic explants displayed increased number endothelial vesseleClike sprouts in comparison to VegfbeC/eC explants that were cocultured with explants from wild-type littermates (Figure 5d). Consistent with the lack of effect of antieCVEGF-B antibody treatment on endothelial-like vessel sprouting in wild-type aortic explants, VegfbeC/eC aortic explants cocultured with wild-type aortic explants displayed similar levels of endothelial-like vessel sprouting to VegfbeC/eC aortic explants cocultured with VegfbeC/eC aortic explants. These data suggest that endogenous VEGF-B was without effect in this assay, which may reflect the low expression of endogenous Vegf-B normally displayed by this tissue.40
VEGF-B Does Not Induce Vascular Permeability
Intradermal injection with 10 ng of rhVEGF-A165 induced a significant increase in Evans blue dye leakage in comparison with control vehicle (0.01% BSA/PBS) (Figure 6). In contrast, intradermal injection of rhVEGF-B167 failed to induce vascular permeability, when tested at concentrations ranging from 0.1 to 100.0 ng/site, over levels observed with control vehicle (ie, 0 ng/mL).
Discussion
VEGF-B has been reported to promote postnatal angiogenesis in vivo.15 Although postnatal blood vessel growth is commonly assumed to be solely governed by angiogenesis, recent data indicate that it also occurs through the recruitment of bone marroweCderived EPCs in a process akin to embryonic vasculogenesis.41 The VEGF-B receptor VEGFR-1 has recently been implicated in mobilization of EPC from the bone marrow,36 suggesting that VEGF-B may be able to promote this process. This raises the question of whether VEGF-B is truly angiogenic, as suggested in previous in vivo studies, or whether VEGF-B may have promoted vessel growth by vasculogenesis through mobilization/recruitment of EPCs. The true mode of action of VEGF-B in promoting blood vessel growth (angiogenesis, vasculogenesis, or a combination of both) needs to be determined, so that its full therapeutic potential can be realized.
Transgenic overexpression of either of the human isoforms of VEGF-B by resting-state endothelial cells had no detectable effect on endothelial proliferation or vessel density in normal tissues, including the liver, heart, and kidney, suggesting that expression of transgene VEGF-B alone at basal levels is not sufficient to stimulate blood vessel growth. Although it may be argued that this might reflect subthreshold levels of expression of the VEGFB transgenes at resting states, this result is consistent with the minimal effect of deletion of the VEGF-B gene on vessel densities in organs normally associated with VEGF-B expression, including the heart and central nervous system.7,42,43 Further, adenoviral vectoreCexpressed VEGF-B was also without effect on vessel growth in muscle and periadventitia.24,25 One possible explanation for lack of effect of VEGF-B on stimulation of angiogenesis in normal tissue may involve insufficient levels of VEGFR-1 expression by endothelial cells.22
In contrast, the data presented here show that both transgene expressed VEGF-B isoforms were found to potentiate neovessel growth in the Matrigel plug assay, where blood vessel in growth was initiated by the action of bFGF. This was evidenced by a significant elevation in vessel density at the Matrigel-skin interface in Matrigel plug samples from VEGFB167Tg and VEGFB186Tg mice when compared with similar samples from wild-type littermates. Neovessel growth was also increased in cutaneous wounds within the granulation tissue from VEGFB167Tg and VEGFB186Tg mice when compared with those from wild-type littermates. Further, because the level of inflammatory cells or more specifically, the level of macrophages in the granulation tissue appeared similar between wounds in VEGFB167Tg and VEGFB186Tg mice compared with those from wild-type littermates, we believe that our results reflect a specific action of VEGF-B in the potentiation of vascular growth. We suggest that previous studies also reflect potentiation, rather than initiation, of vessel growth by VEGF-B.15 For example, expression of VEGF-B by adenoviral vector was found to enhance collateral artery growth at site of vessel ligation, where other growth factors including VEGF-A have been shown to be unregulated.44 VEGF-B also promoted vessel growth in Matrigel matrix, which is a basement membrane matrix extracted from EngelbretheCHolmeCSwarm mouse tumors rich in numerous growth factors and matrix components that supplies an environment favorable to the promotion of vessel growth.33 This may involve upregulation of VEGFR-1 expression by endothelial cells, such as is observed in healing wounds,8 enabling transduction of angiogenic signals.22 Collectively, these data demonstrate that both isoforms of VEGF-B potentiate physiological neovessel growth rather than initiate it.
The levels of EPCs were decreased in the blood of VEGFB167Tg mice when compared with wild-type littermates, implicating VEGF-B in aspects of EPC biology. At this stage we are not able to definitively demonstrate the cause of the reduced numbers of circulating EPCs in the VEGFB167Tg mice but suggest that this may not reflect negative regulation of EPC mobilization from the bone marrow. Vascular growth in the granulation tissue of healing wounds involves the recruitment of EPCs.45 We found that neovessel growth was equally increased in the Matrigel and wound-healing models in the VEGFB167Tg and VEGFB186Tg. We propose that this observation is inconsistent with a negative effect of VEGF-B on EPC mobilization and suggest that the reduced levels of EPCs in the blood of the VEGFB167Tg mice may be attributable to increased tissue recruitment or endothelial adherence of EPCs to the vascular endothelium in these animals. Unlike VEGF-B186, which is freely secreted, VEGF-B167 remains cell associated, where it potentially may exhibit an adhesion molecule-like action,46 and when expressed by endothelial cells as in the VEGFB167Tg mice may result in increased adherence of EPCs to the vasculature, thus removing them from the circulation. Tie-2 expression is upregulated in neovessels in healing wounds,32 and, accordingly, VEGFB167Tg mice and VEGFB186Tg mice displayed increased transgene expression in neovessel endothelium (compared with resting-state skin). This is associated with a significant increase in vessel density in granulation tissue of VEGFB167Tg and VEGFB186Tg wounds in comparison to wild-type littermate wounds. During wound healing, localized transgene expression may be sufficient to promote preferential recruitment and/or differentiation of EPCs from the circulation to the site of neovessel growth. In our wound healing and Matrigel studies, as in previous studies,15 the ability of VEGF-B to promote vascular growth in vivo may, therefore, represent effects on angiogenesis and/or vasculogenesis.
To determine the direct effects of VEGF-B on angiogenesis without the contribution of circulating EPCs to the vascular growth, we specifically selected the aortic explant assay in preference to other in vitro endothelial celleCbased angiogenesis assays because it more closely resembles the physiological angiogenesis process. Vessel growth in this system using rat aortic explants has been shown to directly involve angiogenesis,47 although the contribution of tissue resident EPCs in vessel growth in this assay has not been explored and should not be excluded. Similar to the results obtained in both the Matrigel assay and wound-healing studies, aortic explants from VEGFB167Tg and VEGFB186Tg mice exhibited increased vessel-like endothelial outgrowth in comparison to wild-type littermate aortic explants. Increased vessel growth in both VEGFB167Tg mice and VEGFB186Tg mice was caused by greater vessel-like endothelial sprouting rather than increased vessel branching. This result is in contrast to a previous study in a similar system using embryonic quail heart explants reporting a role for VEGF-B in vessel branching rather than vessel sprouting.48 In that study, the role of VEGF-B was determined by the addition of an antieCVEGF-B polyclonal antibody (C19; Santa Cruz Biotechnology). However, these data should be interpreted with caution because the ability of this antibody to neutralize VEGF-B signaling through its receptor has not, to our knowledge, been specifically assessed. Furthermore, this antibody was designed against a peptide region mapping to the carboxyl terminus of VEGF-B and, therefore, can only bind to a single isoform of VEGF-B,46 which we have determined to be VEGF-B167 by immunoblotting of recombinant VEGF-B isoforms (A.M., M.C., G.K., unpublished observation, 2003). In our studies, the action of both VEGF-B transgenes on potentiation of vessel growth was completely blocked in the presence of antieCVEGF-B mAb (clone 2H10), which is shown to neutralize Vegf-B signaling through VEGFR-1, confirming a direct effect of both VEGF-B isoforms in the potentiation of angiogenesis rather than inherent differences in the endothelial responsiveness of aortic explants between wild-type littermates and transgenic animals. Although the contribution of tissue-resident EPCs to vascular growth in this assay remains undetermined, we conclude that the differences in vessel growth between transgenic and wild-type explants cannot be solely attributable to increased levels of tissue-resident EPCs in the transgenic explants because VEGF-B activity was required during the assay to potentiate vessel growth. Support for a direct action of both VEGF-B transgenes on potentiation of angiogenesis comes from the evidence of a diffusible factor in the potentiation of angiogenesis from VEGFB167Tg and VEGFB186Tg aortic explants to VegfbeC/eC explants in a coculture system. From the early in vitro characterization of the VEGF-B isoforms,46 we had predicted that only the VEGF-B186 should be able to signal to the remote VegfbeC/eC explants because VEGF-B167 has been shown to be retained by the secreting cell. It is possible that the VEGF-B transgene effect involves a second messenger, but this is outside the scope of the present study.
The vascular permeability activity of VEGF-A presents a problem for the use of VEGF-A as a therapeutic agent for promoting tissue revascularization because it also promotes edema. This is illustrated by studies where overexpression of VEGF-A in the skin of mice induced vascular leakage that was evident by disruption of skin architecture.49 In contrast, there was no evidence of edema in the skin of VEGFB167Tg and VEGFB186Tg mice by histological examination. Injections of recombinant VEGF-B167 protein failed to stimulate vascular leakage in the Miles assay when used at concentration up to 10-fold excess of a concentration of VEGF-A, which potently induced vascular leakage in the same animal (wild-type mice, FVB/N strain). This indicates that, unlike VEGF-A, VEGF-B does not promote vascular leakage and indicates that VEGF-B may provide a more favorable therapy for promoting/potentiating tissue revascularization, as it should not have edematous side effects.
In conclusion, our data demonstrate that both isoforms of VEGF-B potentiate, rather than initiate, postnatal vessel growth in vivo. We also present in vitro evidence to show that VEGF-B can potentiate angiogenesis. Because VEGF-B does not appear to promote vascular leakage, its use as a therapeutic agent to promote vascular growth should not be associated with edematous side effects, unlike related factors VEGF-A and PlGF. These data suggest that VEGF-B may provide an effective therapy to promote/enhance vascular growth in diseases associated with impaired revascularization and ischemia, such as ischemic heart disease.
Acknowledgments
This work was supported by a research grant from AMRAD Corporation Ltd, Richmond, Victoria, Australia.
The Queensland Institute of Medical Research and AMRAD Corporation Ltd have a formal agreement whereby both parties share revenue from commercialization of VEGF-B as a potential therapeutic agent or target.
References
Cao Y, Hong A, Schulten H, Post MJ. Update on therapeutic neovascularization. Cardiovasc Res. 2005; 65: 639eC648.
Rissanen TT, Rutanen J, Yla-Herttuala S. Gene transfer for therapeutic vascular growth in myocardial and peripheral ischemia. Adv Genet. 2004; 52: 117eC164.
Eriksson U, Alitalo K. VEGF receptor 1 stimulates stem-cell recruitment and new hope for angiogenesis therapies. Nat Med. 2002; 8: 775eC777.
Tammela T, Enholm B, Alitalo K, Paavonen K. The biology of vascular endothelial growth factors. Cardiovasc Res. 2005; 65: 550eC563.
Neufeld G, Kessler O, Herzog Y. The interaction of neuropilin-1 and neuropilin-2 with tyrosine-kinase receptors for VEGF. Adv Exp Med Biol. 2002; 515: 81eC90.
Terman BI, Dougher-Vermazen M, Carrion ME, Dimitrov D, Armellino DC, Gospodarowicz D, Bohlen P. Identification of the KDR tyrosine kinase as a receptor for vascular endothelial cell growth factor. Biochem Biophys Res Commun. 1992; 187: 1579eC1586.
Bellomo D, Headrick JP, Silins GU, Paterson CA, Thomas PS, Gartside M, Mould A, Cahill MM, Tonks ID, Grimmond SM, Townson S, Wells C, Little M, Cummings MC, Hayward NK, Kay GF. Mice lacking the vascular endothelial growth factor-B gene (Vegfb) have smaller hearts, dysfunctional coronary vasculature, and impaired recovery from cardiac ischemia. Circ Res. 2000; 86: e29eCe35.
Carmeliet P, Moons L, Luttun A, Vincenti V, Compernolle V, De Mol M, Wu Y, Bono F, Devy L, Beck H, Scholz D, Acker T, DiPalma T, Dewerchin M, Noel A, Stalmans I, Barra A, Blacher S, Vandendriessche T, Ponten A, Eriksson U, Plate KH, Foidart JM, Schaper W, Charnock-Jones DS, Hicklin DJ, Herbert JM, Collen D, Persico MG. Synergism between vascular endothelial growth factor and placental growth factor contributes to angiogenesis and plasma extravasation in pathological conditions. Nat Med. 2001; 7: 575eC583.
Luttun A, Brusselmans K, Fukao H, Tjwa M, Ueshima S, Herbert JM, Matsuo O, Collen D, Carmeliet P, Moons L. Loss of placental growth factor protects mice against vascular permeability in pathological conditions. Biochem Biophys Res Commun. 2002; 295: 428eC434.
Mould AW, Tonks ID, Cahill MM, Pettit AR, Thomas R, Hayward NK, Kay GF. Vegfb gene knockout mice display reduced pathology and synovial angiogenesis in both antigen-induced and collagen-induced models of arthritis. Arthritis Rheum. 2003; 48: 2660eC2669.
Yoshida A, Anand-Apte B, Zetter BR. Differential endothelial migration and proliferation to basic fibroblast growth factor and vascular endothelial growth factor. Growth Factors. 1996; 13: 57eC64.
Six I, Kureishi Y, Luo Z, Walsh K. Akt signaling mediates VEGF/VPF vascular permeability in vivo. FEBS Lett. 2002; 532: 67eC69.
Sawano A, Takahashi T, Yamaguchi S, Aonuma M, Shibuya M. Flt-1 but not KDR/Flk-1 tyrosine kinase is a receptor for placenta growth factor, which is related to vascular endothelial growth factor. Cell Growth Differ. 1996; 7: 213eC221.
Oura H, Bertoncini J, Velasco P, Brown LF, Carmeliet P, Detmar M. A critical role of placental growth factor in the induction of inflammation and edema formation. Blood. 2003; 101: 560eC567.
Silvestre JS, Tamarat R, Ebrahimian TG, Le-Roux A, Clergue M, Emmanuel F, Duriez M, Schwartz B, Branellec D, Levy BI. Vascular endothelial growth factor-B promotes in vivo angiogenesis. Circ Res. 2003; 93: 114eC123.
Fong GH, Zhang L, Bryce DM, Peng J. Increased hemangioblast commitment, not vascular disorganization, is the primary defect in flt-1 knock-out mice. Development. 1999; 126: 3015eC3025.
Lalla RV, Boisoneau DS, Spiro JD, Kreutzer DL. Expression of vascular endothelial growth factor receptors on tumor cells in head and neck squamous cell carcinoma. Arch Otolaryngol Head Neck Surg. 2003; 129: 882eC888.
Casella I, Feccia T, Chelucci C, Samoggia P, Castelli G, Guerriero R, Parolini I, Petrucci E, Pelosi E, Morsilli O, Gabbianelli M, Testa U, Peschle C. Autocrine-paracrine VEGF loops potentiate the maturation of megakaryocytic precursors through Flt1 receptor. Blood. 2003; 101: 1316eC1323.
Sawano A, Iwai S, Sakurai Y, Ito M, Shitara K, Nakahata T, Shibuya M. Flt-1, vascular endothelial growth factor receptor 1, is a novel cell surface marker for the lineage of monocyte-macrophages in humans. Blood. 2001; 97: 785eC791.
Marti HJ, Bernaudin M, Bellail A, Schoch H, Euler M, Petit E, Risau W. Hypoxia-induced vascular endothelial growth factor expression precedes neovascularization after cerebral ischemia. Am J Pathol. 2000; 156: 965eC976.
Tomanek RJ, Lund DD, Yue X. Hypoxic induction of myocardial vascularization during development. Adv Exp Med Biol. 2003; 543: 139eC149.
Hiratsuka S, Maru Y, Okada A, Seiki M, Noda T, Shibuya M. Involvement of Flt-1 tyrosine kinase (vascular endothelial growth factor receptor-1) in pathological angiogenesis. Cancer Res. 2001; 61: 1207eC1213.
Luttun A, Tjwa M, Carmeliet P. Placental growth factor (PlGF) and its receptor Flt-1 (VEGFR-1): novel therapeutic targets for angiogenic disorders. Ann N Y Acad Sci. 2002; 979: 80eC93.
Rissanen TT, Markkanen JE, Gruchala M, Heikura T, Puranen A, Kettunen MI, Kholova I, Kauppinen RA, Achen MG, Stacker SA, Alitalo K, Yla-Herttuala S. VEGF-D is the strongest angiogenic and lymphangiogenic effector among VEGFs delivered into skeletal muscle via adenoviruses. Circ Res. 2003; 92: 1098eC1106.
Bhardwaj S, Roy H, Gruchala M, Viita H, Kholova I, Kokina I, Achen MG, Stacker SA, Hedman M, Alitalo K, Yla-Herttuala S. Angiogenic responses of vascular endothelial growth factors in periadventitial tissue. Hum Gene Ther. 2003; 14: 1451eC1462.
Scotney P, MacKenzie A, Maccarone P, Fabri L, Scrofani S, Gooley P, Nash A. Human vascular endothelial growth factor B: characterization of recombinant isoforms and generation of neutralizing monoclonal antibodies. Clin Exp Pharmacol Physiol. 2002; 29: 1024eC1029.
Makinen T, Jussila L, Veikkola T, Karpanen T, Kettunen MI, Pulkkanen KJ, Kauppinen R, Jackson DG, Kubo H, Nishikawa S, Yla-Herttuala S, Alitalo K. Inhibition of lymphangiogenesis with resulting lymphedema in transgenic mice expressing soluble VEGF receptor-3. Nat Med. 2001; 7: 199eC205.
Achen MG, Roufail S, Domagala T, Catimel B, Nice EC, Geleick DM, Murphy R, Scott AM, Caesar C, Makinen T, Alitalo K, Stacker SA. Monoclonal antibodies to vascular endothelial growth factor-D block its interactions with both VEGF receptor-2 and VEGF receptor-3. Eur J Biochem. 2000; 267: 2505eC2515.
Mizushima S, Nagata S. pEF-BOS, a powerful mammalian expression vector. Nucleic Acids Res. 1990; 18: 5322.
Schlaeger TM, Bartunkova S, Lawitts JA, Teichmann G, Risau W, Deutsch U, Sato TN. Uniform vascular-endothelial-cell-specific gene expression in both embryonic and adult transgenic mice. Proc Natl Acad Sci U S A. 1997; 94: 3058eC3063.
Brown AL, Kay GF. Bex1, a gene with increased expression in parthenogenetic embryos, is a member of a novel gene family on the mouse X chromosome. Hum Mol Genet. 1999; 8: 611eC619.
Wong AL, Haroon ZA, Werner S, Dewhirst MW, Greenberg CS, Peters KG. Tie2 expression and phosphorylation in angiogenic and quiescent adult tissues. Circ Res. 1997; 81: 567eC574.
Passaniti A, Taylor RM, Pili R, Guo Y, Long PV, Haney JA, Pauly RR, Grant DS, Martin GR. A simple, quantitative method for assessing angiogenesis and antiangiogenic agents using reconstituted basement membrane, heparin, and fibroblast growth factor. Lab Invest. 1992; 67: 519eC528.
Gatto C, Rieppi M, Borsotti P, Innocenti S, Ceruti R, Drudis T, Scanziani E, Casazza AM, Taraboletti G, Giavazzi R. BAY 12eC9566, a novel inhibitor of matrix metalloproteinases with antiangiogenic activity. Clin Cancer Res. 1999; 5: 3603eC3607.
Masson VV, Devy L, Grignet-Debrus C, Bernt S, Bajou K, Blacher S, Roland G, Chang Y, Fong T, Carmeliet P, Foidart JM, Noel A. Mouse aortic ring assay: a new approach of the molecular genetics of angiogenesis. Biol Proced Online. 2002; 4: 24eC31.
Heil M, Mitnacht-Krauss R, Issbrucker K, van den Heuvel J, Dehio C, Schaper W, Clauss M, Weich HA. An engineered heparin-binding form of VEGF-E (hbVEGF-E). Biological effects in vitro and mobilizatiion of precursor cells. Angiogenesis. 2003; 6: 201eC211.
Wise LM, Veikkola T, Mercer AA, Savory LJ, Fleming SB, Caesar C, Vitali A, Makinen T, Alitalo K, Stacker SA. Vascular endothelial growth factor (VEGF)-like protein from orf virus NZ2 binds to VEGFR2 and neuropilin-1. Proc Natl Acad Sci U S A. 1999; 96: 3071eC3076.
Padmanabhan V, Callas P, Philips G, Trainer TD, Beatty BG. DNA replication regulation protein Mcm7 as a marker of proliferation in prostate cancer. J Clin Pathol. 2004; 57: 1057eC1062.
Gunningham SP, Currie MJ, Han C, Robinson BA, Scott PA, Harris AL, Fox SB. VEGF-B expression in human primary breast cancers is associated with lymph node metastasis but not angiogenesis. J Pathol. 2001; 193: 325eC332.
Aase K, Lymboussaki A, Kaipainen A, Olofsson B, Alitalo K, Eriksson U. Localization of VEGF-B in the mouse embryo suggests a paracrine role of the growth factor in the developing vasculature. Dev Dyn. 1999; 215: 12eC25.
Masuda H, Asahara T. Post-natal endothelial progenitor cells for neovascularization in tissue regeneration. Cardiovasc Res. 2003; 58: 390eC398.
Aase K, von Euler G, Li X, Ponten A, Thoren P, Cao R, Cao Y, Olofsson B, Gebre-Medhin S, Pekny M, Alitalo K, Betsholtz C, Eriksson U. Vascular endothelial growth factor-B-deficient mice display an atrial conduction defect. Circulation. 2001; 104: 358eC364.
Sun Y, Jin K, Childs JT, Xie L, Mao XO, Greenberg DA. Increased severity of cerebral ischemic injury in vascular endothelial growth factor-b-deficient mice. J Cereb Blood Flow Metab. 2004; 24: 1146eC1152.
Prior BM, Lloyd PG, Ren J, Li H, Yang HT, Laughlin MH, Terjung RL. Time course of changes in collateral blood flow and isolated vessel size and gene expression after femoral artery occlusion in rats. Am J Physiol Heart Circ Physiol. 2004; 287: H2434eCH2447.
Keswani SG, Katz AB, Lim FY, Zoltick P, Radu A, Alaee D, Herlyn M, Crombleholme TM. Adenoviral mediated gene transfer of PDGF-B enhances wound healing in type I and type II diabetic wounds. Wound Repair Regen. 2004; 12: 497eC504.
Olofsson B, Pajusola K, von Euler G, Chilov D, Alitalo K, Eriksson U. Genomic organization of the mouse and human genes for vascular endothelial growth factor B (VEGF-B) and characterization of a second splice isoform. J Biol Chem. 1996; 271: 19310eC19317.
Kruger EA, Duray PH, Tsokos MG, Venzon DJ, Libutti SK, Dixon SC, Rudek MA, Pluda J, Allegra C, Figg WD. Endostatin inhibits microvessel formation in the ex vivo rat aortic ring angiogenesis assay. Biochem Biophys Res Commun. 2000; 268: 183eC191.
Tomanek RJ, Holifield JS, Reiter RS, Sandra A, Lin JJ. Role of VEGF family members and receptors in coronary vessel formation. Dev Dyn. 2002; 225: 233eC240.
Larcher F, Murillas R, Bolontrade M, Conti CJ, Jorcano JL. VEGF/VPF overexpression in skin of transgenic mice induces angiogenesis, vascular hyperpermeability and accelerated tumor development. Oncogene. 1998; 17: 303eC311.(Arne W. Mould, Sonia A. G)