| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
,1
* Department of Oncology and Pathology, Cancer Centre Karolinska, and
Department of Clinical Neuroscience, Section of Ophthalmology and Vision, Karolinska Institute, St. Erik’s Eye Hospital, Stockholm, Sweden;
Vascular Biology Laboratory, Cancer Research UK, London, UK; and
BioInvent International AB, Sölvegatan 41, Lund, Sweden
2Correspondence: Department of Oncology and Pathology, CCK, Cancer Centre Karolinska, Karolinska Institute, SE 17176, Stockholm, Sweden. E-mail: lars.holmgren{at}ki.se
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: angiostatin • neoplasia • vascular/ocular disease • migration
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. Vascular entothelial growth factor (VEGF) -A, which signals via the tyrosine receptor kinases VEGF-R1 and -R2, plays a pivotal role in the regulation of physiological and pathological angiogenesis (2)
. Indeed, recent seminal studies have shown that inhibitors of VEGF may prolong life in cancer patients and improve visual acuity in patients suffering from neovascular age-related macular degeneration (3
, 4)
. However, compensatory pathways may be activated when VEGF signaling/activation is inhibited in cancer model systems (5
, 6)
. The upregulation of other VEGF-independent signaling pathways appears not to be restricted to tumors, as other proangiogenic stimuli may affect pathogenesis in proliferative diabetic retinopathy as well as age-related macular degeneration (7
, 8)
. These findings suggest that therapies that target more than one pathway or endothelial functions downstream of these pathways could improve the efficacy of antiangiogenic therapies.
Endogenous inhibitors of angiogenesis include various antiangiogenic peptides, hormone metabolites, and apoptosis modulators (reviewed in (9
, 10)
. Some of the identified endogenous angiogenesis inhibitors, such as thrombospondin, endostatin, angiostatin, and others, target the endothelial cells directly and render them unresponsive to a variety of mitogenic and migratory signals (11
, 12)
. Angiostatin, (a fragment of plasminogen containing kringle domains 1–4 or 1–3) has been shown specifically to inhibit endothelial cell migration and effectively inhibits vascularization in models of tumor growth, choroidal neovascularization, and oxygen-induced retinopathy in mice (13
, 14)
. However, low half-life in circulation (3 h) and high production costs hinder clinical use of angiostatin (15)
.
In a strategy to identify the angiostatin receptor by yeast-two hybrid system, we discovered a novel protein named angiomotin (Amot) (16)
. Amot is a membrane-associated protein that mediates antimigratory effects of angiostatin in vitro. It consists of two protein isoforms with apparently distinct roles, the shorter isoform, p80, confers a hypermigratory and invasive phenotype in transfected cells (17)
, whereas the p130 isoform localizes to tight junctions and regulates cell shape (18
, 19)
. We have shown recently that Amot plays an essential role in vascular development in vivo. Of the Amot knockout mice, 75% die between embryonic day 11 (E11) and E11.5 and exhibit severe vascular insufficiency in the intersomitic region as well as dilated vessels in the brain (20)
. Furthermore, knockdown of Amot in zebrafish reduced the number of filopodia of endothelial tip cells and severely impaired the migration of intersegmental vessels (20)
.
The role of Amot in cell migration and the expression in endothelial cells during blood vessel formation indicates that Amot is a potential target for antiangiogenic therapy. Indeed, targeting Amot using DNA vaccination efficiently inhibited angiogenesis and tumor growth in vivo. DNA vaccination blocked angiogenesis in the matrigel plug assay in vivo and prevented growth of transplanted tumors for up to 150 days (21)
.
These findings indicate that angiomotin plays an essential role during blood vessel formation. For this reason, we generated a human monoclonal antibody, B06, which binds to angiomotin on the cell surface and inhibits endothelial migration. We further show that this antibody decreases the number of filopodia of tip cells during retinal angiogenesis, resulting in inhibition of vascular migration and neovascularization in vivo. The efficacy of the angiomotin intervention in three different model systems argues that antibodies generated against Amot may be important tools for treating angiogenesis-dependent diseases.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. In brief, human p80 Amot was expressed in eukaryotic cells and binding assays were performed with intact cells or cell lysate or in a purified form, in 3 consecutive rounds of selection. Selected scFvs were screened for specific Angiomotin binding in an automated system with an ELISA format and luminescence as the readout. The scFvs identified as being specific for Angiomotin were also screened for cross-reactivity to murine Amot with an ELISA format with luminescence as the readout. The recombinant scFv contained a C-terminal His tag that allowed for IMAC purification using Ni-NTA sepharose (Qiagen, Valencia, CA, USA). The purity of the preparations exceeded 95%, as determined from SDS-PAGE. Fab fragments and full-length IgG antibodies were produced through cloning into modified pcDNA3 vectors, followed by transient transfections into HEK293 cells with Lipofectamin (Invitrogen, Carlsbad, CA, USA). Fab fragments and full-length IgG antibodies were purified on a MabSelect protein A column (Amersham Biosciences, Uppsala, Sweden). The purity of the preparations exceeded 98%, as determined from SDS-PAGE. The binding specificity of the scFv was tested using luminescence-based ELISA, where dilutions of the scFv antibody fragments were incubated in test plate wells coated with purified Angiomotin. The Fab fragments utilized for PEGylation contained a C-terminal cystein, which allowed for a single site-specific PEGylation using a 20 kDa PEG-maleimid compound (Nectar Therapeutics, Huntsville, AL, USA). 2-Mercaptoethylamine HCL (Fluka, Buchs, Switzerland), 5 mM, pH 7.0, at 37°C for 90 min, was used for selective reduction of the C-terminal cysteins of the Fab fragments. The molar ratio (PEG:Fab) in the PEGylation reaction was 5:1. The PEGylation reaction was conducted at 25°C, pH 7, under nitrogen for 2 h. MabSelect protein A column chromatography was used to purify the Fab fragments from unreacted PEG-maleimid. PEGylated Fab could be separated from non-PEGylated Fab using size exclusion chromatography (GE Healthcare, Uppsala, Sweden).
Cell culture
Spontaneously immortalized mouse aortic endothelial (MAE) cells (23)
transfected with angiomotin or vector (16)
were maintained in Dulbecco’s modified Eagle’s medium (DMEM, Sigma, Stockholm, Sweden) containing 10% fetal bovine serum (FBS, Life Technologies, Inc., Stockholm, Sweden), 1% penicillin and 1% glutamine. The TUBO cell line, kindly provided by Dr. Guido Forni (Department of Clinical and Biological Sciences, University of Turin, Orbassano, Italy), was derived from a spontaneous mammary tumor which arose in a BALB NeuT transgenic mouse expressing a transforming rat neu oncogene (24)
. Cells were cultured in Iscoves Modified Dubecco’s Medium (IMDM, Sigma, Sweden) with 10% FBS. hTERT+-immortalized bovine capillary endothelial (BCE) cells (25)
were grown in DMEM (Sigma, Sweden) with 10% fetal bovine serum, 2 ng/mL FGF-2 (Peprotech, London, UK), 1% glutamine, and 1% penicillin/streptomycin.
FACS analysis
For evaluation of cell surface binding by flow-cytometry, MAE cells were incubated for 1.5 h with individual scFv clones at a concentration of 10 µg/ml in PBS (Invitrogen) containing 0.5% w/v bovine serum albumin (DPBS-B), followed by incubation with anti-flag-biotin (Sigma, Sweden) and streptavidin-Alexa 647 Fluor (Molecular Probes, Eugene, OR, USA). Dead cells were excluded by SYTOX Green Nucleic Acid Stain, (Molecular Probes). All incubations were performed on ice.
Tissue cross-reactivity
Antibodies were tested at concentrations of 5 and 20 µg/ml toward acetone-fixed frozen sections (8 µm) of normal human tissues (placenta from 3 individuals; liver, kidney, heart, pancreas, lymph node, and cerebrum from two individuals). The human IgGs were preincubated in tubes with a biotinylated monovalent goat Fab anti-human IgG fragment (Jackson ImmunoResearch Laboratories Inc., West Grove, PA, USA; Cat. No. 109-067-003). Staining was performed using the avidin-biotin complex (ABC) method. Slides were evaluated under a light microscope (Labophot-2, Nikon Instruments Europe B.V., Badhoevedorp, Netherlands), and photos were taken with a Leica DMR microscope (Leica Microsystems, Wetzlar, Germany). Monoclonal mouse anti-human CD34 class II, clone QBEnd (DakoCytomation, Copenhagen, Denmark; Code No. M 7165) was used as a positive control antibody while an n-CoDeR-derived IgG1 directed toward FITC was used as negative control.
Migration assay
Migration assays were performed in a modified Boyden chamber using a 48-well chemotaxis chamber (Neuroprobe Inc., Gaithersburg, MD, USA) as described earlier (26)
. Briefly, 8 µm nucleopore polyvinylpirrolidine-free polycarbonate filters were coated with 100 µg/ml of collagen type 1 (Cohesion, Palo Alto, CA, USA) overnight. hTERT+-BCE cells were starved in 0.2% FCS-DMEM for 16 h. The cells were trypsinized and resuspended in DMEM containing 0.1% bovine serum albumin (BSA), and 30,000 cells were added with or without B06 scFv or control scFv to each well of the upper chamber. FGF-2 (PeproTech EC Ltd., Rocky Hill, NJ, USA) at 30 ng/ml or VEGF at 50 ng/ml (PeproTech EC Ltd., USA) were used as chemoattractants in the lower chambers. The chemotaxis chambers were incubated for 3–5 h at 37°C with 10% CO2 to allow cells to migrate through the collagen-coated polycarbonate filter. Nonmigrating cells on the upper surface of the filter were removed and the filter was stained with Giemsa Stain (VWR International Ltd, West Chester, PA). The total number of migrated cells per field was counted at x20; each sample was tested in quadruplicates in at least three independent experiments.
Matrigel assay in vitro
Liquid matrigel (150 µl; Becton Dickinson Biosciences, Bedford, MA, USA) was added to each well of an 8-well chamber slide (BD FalconTM) and incubated at 37°C for 30 min to allow the gel to polymerize. Mouse aortic endothelial cells (MAE; 1.5x105) maintained in serum-free DMEM with 0.5% BSA were pretreated for 16 h with 5 µg/ml of CT17 scFv control or B06 scFv (BioInvent International, AB, Lund, Sweden) and seeded on a layer of polymerized matrigel as described previously (16)
. After 24 h the changes in cell morphology were examined with a phase-contrast microscope.
Matrigel plug assay in vivo
Matrigel plug assays were performed as described previously, with some modifications (27)
. Balb/C mice were anesthetized with Isofluran (Forene®, Abbot, Sweden) and injected with matrigel mixture subcutaneously in the lower quadrant of the abdomen. In the FGF-2-induced angiogenesis model, every matrigel plug contained 0.5 ml of matrigel (Becton Dickinson Biosciences), 200 ng/ml FGF-2 (PeproTech EC Ltd., London, UK), 250 µg of control antibodies in one group, and 250 µg of B06 scFv in another. Two control groups of mice were injected either with matrigel containing only FGF-2 or matrigel alone. Matrigel plugs were excised 7 days after implantation, photographed, and processed for histological studies. In the tumor-induced angiogenesis model, every matrigel plug contained 75,000 TUBO cells and 500 µg of single-chain antibodies in 0.5 ml of liquid matrigel. On day 7 after gel implantation, matrigel plugs have been removed and prepared for immunohistochemical examination. For the systemic treatment with Amot antibody we implanted into every mouse 0.5 ml of matrigel containing 150,000 TUBO cells. Animals were treated every second day with 500 µg/injection B06 Fab-PEG antibody during 2 wk. Rat anti-mouse CD31 mAb (PECAM-1) (BD PharmingenTM) and anti-rat-FITC-conjugated (Jackson ImmunoResearch) antibodies were used to visualize vascularization of matrigel plugs. To analyze the microvessel density, three images from every matrigel plug have been taken with a Zeiss Axioplan 2 fluorescence microscope (Carl Zeiss, Oberkochen, Germany). The vessels were counted under the microscope at x20.
Retinal angiogenesis assay
Intraocular injections were performed as described (28)
. Briefly, pups (P4) were anesthetized by isofluran inhalation. Injections (0.5 µl of 
5 µg/µl B06 scFv or CT17 scFv in PBS) were performed using 10 µl gastight Hamilton syringes equipped with 34-gauge needles attached to a micromanipulator. Three litters of C57Bl6/J mice were treated, 10 pups per group. The uninjected eyes served as additional control. After 24 h pups were euthanized, eyes were collected and fixed in 4% paraformaldehyde, and retinas were dissected and treated as described (28)
. For immunohistochemistry, endothelial cells and microglial cells were visualized with biotinylated isolectin B4 (Bandeiraea simplicifolia; Sigma, Sweden), followed by streptavidin Alexa Fluor 488 (Molecular Probes, Invitrogen). Retinas were flatmounted and analyzed using a Zeiss SV11 fluorescence stereomicroscope equipped with an Axiocam HRc. The distance from optic nerve to vascular front was measured using Axiovision 4.5 software (Zeiss). Images for filopodia analysis were taken on a Zeiss LSM 510 confocal microscope using 40 x 1.2 NA lens (settings: pinhole 1 airy unit, 1024x1024 pixel). A total of 10 z-images was collected at 0.4 µm intervals and presented as projection. Projection-images were converted to grayscale and inverted to facilitate filopodia visibility. The outline of the vessels at the migration front was measured and filopodia were counted using ImageJ 1.36b (public domain software, National Institutes of Health, Bethesda, Maryland, USA). Data were analyzed by unpaired two-tailed t test and graphed using Prism 4 software (GraphPad, San Diego, CA, USA).
Laser-induced choroidal neovascularization (CNV) in Mice
CNV was generated by krypton laser-induced rupture of Bruch’s membrane, as described previously (29)
. Briefly, three krypton laser photocoagulation burns (50 µm spot size, 0.1 s duration, 120 mW power) were induced in each eye of C57BL/6J mouse by using a handheld contact lens (647 nm, Spectra-Physics 265 Exciter, Lasertek, Helsinki, Finland). Mice received IP injections with 400 µg B06 Fab PEG or CT17 every second day, with the first injection given one day before laser treatment. Eyes were enucleated 10 days after laser treatment and fixed in 4% paraformaldehyde for 30 min, the cornea and lens were removed and the entire retina was carefully dissected from the eyecup. The RPE-choroid-sclera eyecups were rinsed in PBS, permeabilized in 0.5% Triton X-100, and blocked with 3% goat serum in PBS/Triton X-100. The eyecups were incubated with biotinylated isolectin B4 (1:100 dilution, lectin from Griffonia simplicifolia, Sigma, St. Louis, MO, USA) and anti-CD31 (BD Biosciences Pharmingen, San Jose, CA, USA) overnight at 4°C, followed by an incubation with Texas Red Streptavidin (1:100 dilution, Vector Laboratories, Burlingame, CA, USA) and Alexa 488 goat antirat (1:100, Molecular Probes). Radial cuts were made from the edge of the eyecup to the equator, and the eyecup was flattened and mounted with antifade medium (Vectashield Mounting Medium, Vector Laboratories) with the sclera facing down and the choroid facing up. Flat-mounted retinas were analyzed by fluorescence microscopy using a Carl Zeiss Axioskop 2 microscope equipped with a digital camera software (Axiocam HRm). Lesions were measured manually in a masked fashion, and data from each lesion were treated as a single statistical point. The outline of isolectin B4 staining was used to estimate the total plaque area, and the vascularization was estimated from the PECAM-1 staining by quantifying the number of PECAM-1 positive pixels per plaque.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. Three consecutive selections generated over 93 unique phage antibody clones. The individual phage clones were converted to scFv format and expressed in E. coli. Binding of purified scFv to both human and mouse Amot was assessed by ELISA as described in Materials and Methods (Fig. 1
A, B). Cross-reactivity to mouse Amot was of importance, as future screenings included mouse angiogenesis model systems. Another criterion was the ability of Amot-specific scFv to bind extracellular epitopes of angiomotin. FACS analysis showed cell surface binding of the B06 single-chain Fv (B06 scFv) to p80 Amot transfected cells (Fig. 1C
). Binding to irrelevant antigens may result in immunopathological effects and increase the toxicity profile of the antibody. Cross-reactivity was examined to a limited panel of human normal tissues (Supplemental Data 1). Cytotrophoblasts of the placenta, cells that previously have been shown to express high levels of Amot (16)
, were used as positive controls (data not shown). Binding to angiogenic vessels was verified by immunofluorescent stainings of vessels induced by FGF-2 in the in vivo matrigel plug assay (Fig. 2
).
|
|
Angiomotin-binding B06 single chains inhibit migration and tube formation in vitro
For functional analysis, C-terminal His-tagged scFv was produced in HEK 293 cells as described in Materials and Methods. We used transfected immortalized mouse aortic endothelial (MAE) cells to screen for inhibitory migratory activity of angiomotin-binding single chains. The MAE cells lack detectable endogenous expression of angiomotin, whereas transfection of p80 Amot promotes migration and renders the cells responsive to angiostatin treatment (16)
. This set-up has a major advantage in that unspecific inhibition of motility can be excluded by parallel testing of the MAE-vector control cell line. We screened over 83 scFvs in the Boyden chamber assay, where 40% (37/83) of the single chains showed marked inhibition of FGF-2- and VEGF-induced MAE Amot cell migration (Fig. 3
A, B). The B06 scFv was chosen as a prime candidate due to its low cross-reactivity to normal human tissues and the relatively high potency in inhibiting both FGF-2 and VEGF-induced MAE Amot migration (B06 IC50=100 pg/ml, angiostatin IC50=500 ng/ml) (16)
. In addition, B06 scFv inhibited migration of bovine capillary endothelial cells that express endogenous angiomotin (Fig. 3C
).
|
We also assessed the ability of B06 scFv to inhibit tube formation in the matrigel assay in vitro. MAE-p80 cells were plated on matrigel in the presence of either B06 scFv or the control single chain (CT17 scFv). In this assay, cells adhere to the matrix and spontaneously migrate to form an interconnecting tubular network. B06 scFv-treated cells adhered to matrix and formed aggregates but did not migrate out to form tubes similar to what we previously have observed with angiostatin treatment (16)
. No detectable inhibition of tube formation could be found in the angiomotin-negative control cells (Fig. 4
).
|
Inhibition of angiogenesis in vivo
To assess antiangiogenic efficacy of B06 scFv in vivo, we used an assay in which matrigel plugs implanted in mice are invaded by host endothelial cells to form a capillary network. Matrigel is a solubilized basement membrane preparation extracted from Engelbreth-Holm-Swarm (EHS) murine sarcoma, which is liquid at 4°C but solidifies at 37°C and allows slow release of growth factors that stimulate ingrowth of vessels. We have previously shown that the vessels inside the plug express high levels of angiomotin, whereas the vessels in surrounding tissues are negative (21)
. We injected each mouse with 0.5 ml of matrigel containing 200 ng/ml of FGF-2 in combination with 500 µg/ml of CT17 scFv or B06 scFv. The animals were euthanized seven days after matrigel implantation and analyzed by PECAM-1 staining (Fig. 5
A). Analysis of the vascular density showed an almost complete inhibition of angiogenesis B06 scFv-containing plugs as compared to the CT17 scFv control plugs (Fig. 5B
).
|
In our experience, VEGF has a modest stimulatory effect on angiogenesis in the matrigel plug assay, even when applied at high concentrations (data not shown). To circumvent this problem we induced angiogenesis by embedding TUBO breast cancer cells into the matrigel. TUBO is a cell line generated from a spontaneous mammary gland carcinoma from a Balb-neuT mouse and induces tumors very similar to the alveolar type human lobular mammary carcinomas (30)
. It has been shown to induce angiogenesis via VEGF-dependent pathways (31)
. TUBO tumors embedded into matrigel were treated with locally delivered single-chain Amot antibody. The PECAM-1 staining of matrigel plug cryosections with local treatment (Fig. 6
A, B) showed a 70% inhibition of vascular infiltration in B06 scFv-treated plugs as compared to controls (Fig. 6C
).
|
Systemic clearance of scFv antibodies is fast and occurs mainly through renal excretion, since the molecular weight is below the filtration barrier in the kidney glomeruli. PEGylation is one of the best validated strategies to extend the serum half-life of therapeutic agents and to decrease their immunogenicity (32)
. To assess efficacy of systemic B06 treatment, mice were treated with intraperitoneal injections every second day using pegylated B06 or CT17 Fab antibodies. Inhibition of TUBO-induced angiogenesis could be observed by visual observation (Fig. 6G, H
) as well as by PECAM-1 immunostaining of cryosections (Fig. 6D, E
). Systemic treatment with B06 Fab-PEG antibody resulted in significant reduction of vessel density in the plugs (Fig. 6F
). These data show that local or systemic administration of the B06 antibody efficiently inhibited tumor angiogenesis.
Inhibition of neonatal retinal angiogenesis
The vascularization of the mouse retina has been used as a model system to study physiological angiogenesis due to the relative ease of visualization of the vessels. The mouse retinal vasculature develops a network originating from the optic nerve, which spreads to vascularize the retina within 8 days (28)
. The vessel migration is orchestrated by matrix-bound VEGF, which is detected by the filopodia of tip cells at the leading edge of the migrating vessels (33)
. We have previously shown that angiomotin is expressed in blood vessels during the vascularization of the postnatal mouse retina (18)
. To assess the effect of B06 scFv on the vessel expansion and migration, we treated 4-day-old pups with intraocular injections of either B06 scFv or CT17 scFv. Animals were euthanized 24 h later, and the retinas were processed for whole mount immunofluorescent staining with isolectin B4. Expansion of the retinal network was estimated by measuring the distance from the optical nerve to the leading edge (Fig. 7
A). No effect on the vascular expansion was observed in CT17 scFv-treated eyes as compared to untreated eyes, demonstrating that the single chains or the intraocular injections per se, do not affect retinal angiogenesis. Treatment with B06 scFv significantly inhibited the migration of vessels during the 24 h period (Fig. 7C
). However, treatment with B06 scFv resulted in a significant reduction in the number of filopodia extensions in the tip cells (Fig. 7B, D
). Analysis of the vessel morphology revealed no differences in the number of vessel branch points (Fig. 7E
).
|
Inhibition of choroidal neovascularization
Age-related macular degeneration is a major cause of severe visual loss in elderly patients. Retinal damage is often associated with subretinal plaque formation as a consequence of CNV. We studied the effect of angiomotin intervention in a mouse model where CNV is induced by disruption of the Bruch’s membrane/retinal pigment epithelium complex by laser treatment. Mice were treated systemically with intraperitoneal injections every second day using B06 or CT17 Fab-PEG. Mice were euthanized on day 10, and eyes were processed for whole mount staining by using both isolectin B4 and PECAM-1. Vascularization was estimated by quantifying the number of PECAM-1 positive pixels per plaque. Systemic treatment with B06 Fab-PEG resulted in 73% reduction of plaque vessels as compared to the animals treated with the CT17 Fab-PEG control (Fig. 8
A, B).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The formation of blood vessels is a complex process that involves endothelial proliferation, migration, and lumen formation. Our strategy has been to target endothelial migration by generation of antibodies against Amot. Extensive evidence indicates that Amot is involved in the regulation of endothelial cell migration. First, overexpression of p80 Amot in mouse aortic endothelial cells stimulates migration and sensitizes cells to respond to chemotactic factors (16)
. Furthermore, Amot expression promotes tumor invasion in vitro and in vivo (17)
. Conversely, expression of a dominant-negative mutant of Amot in the endothelial lineage during mouse embryogenesis inhibited endothelial migration and vascularization of the mouse neuroectoderm during embryogenesis (34)
. Finally, Amot-deficient endothelial cells also exhibit a defective response to chemotactic factors (20)
.
In this paper, we provide in vitro and in vivo evidence that the B06 Amot antibody inhibits endothelial cell migration. In vitro, we used the Boyden chamber assay to screen and verify antimigratory activity. The B06 antibody inhibited migration of both Amot-transfected cells and endothelial cells expressing endogenous Amot. However, the mechanism behind the antimigratory effect is not clear from the in vitro experiments. In vivo, the vascular network of the retina of neonatal mice is formed by the concerted migration of specialized endothelial tip cells. These cells are polarized migratory cells extending multiple filopodia in the direction of migration. Filopodial function is required in cell migration and many morphogenetic events, including gastrulation, axonal path finding, epithelial cell adhesion, wound healing etc. (35)
. During angiogenesis, VEGF appears to play an essential role of vessel migration by activating VEGF-R2 receptors located on tip cell filopodia (28)
. Our data show that intraocular injections inhibited the motility of postnatal retinal vessels. Treatment did not affect the number of branch points of established vessels or cause any other obvious visible effects. However, the number of filopodia elaborated by the tip cells was significantly lower. Interestingly, genetic knockdown of Amot in zebrafish inhibits migration of intersegmental vessels as well as down-regulates the number of filopodia of the tip cells (20)
. These findings may be explained by a potential role of Amot in the cytoskeleton rearrangements required for the projection of filopodia, or AMOT may affect cell polarity. A role of angiomotin in cell polarity was recently provided by Wells et al., (36)
who showed that angiomotin positively controls Cdc42 activity via its association with the Cdc42 GAP protein Rich. Taken together, these data suggest that angiomotin may act as a scaffold for GTPase-regulating proteins and locally control the GTPase activity. It is tempting to speculate that Amot interference by treatment with either angiostatin or Amot antibodies inhibits polarization of endothelial cells and therefore renders them insensitive to chemotactic cues.
It has been suggested that angiogenesis is controlled by an "angiogenic balance" of systemic or local levels of stimulators and inhibitors. It is now well established that the positive signals from the VEGF pathway represent an important target for antitumor therapy as well as age-related macular degeneration and rheumatoid arthritis. The therapeutic potential of only a limited number of endogenous inhibitors (angiostatin, endostatin, thrombospondin) has been tested in the clinic. The overall conclusion is that more knowledge is needed regarding their mechanisms of action to optimize their efficacy. We initially identified Amot as a receptor for angiostatin, and the data presented in this paper reveal that Amot plays an essential role in both physiological and pathological angiogenesis. The data further support our hypothesis that Amot is a positive regulator of blood vessel formation and that angiostatin may acts as an antagonist of Amot-mediated migration. Our findings indicate that the antimigratory effect is an important component of the antiangiogenic activities of angiostatin. This does not rule out, however, that angiostatin may inhibit proliferation and induce apoptosis via other receptors, such as ATP synthase (37)
. Human therapeutic antibodies provide several important advantages, including decreased dosage, increased binding specificity and improved half-life in circulation. In contrast to angiostatin, laboratory-engineered therapeutic antibodies have considerably better half-life in circulation and are thus more suitable for long-term treatments.
Selective targeting of angiomotin by antibodies may be a promising approach to inhibit pathological blood vessel formation. One possible advantage of targeting angiomotin is that it appears not to be dependent on an individual signaling pathway. The observed inhibition of angiogenesis in cancer as well as in CNV indicates that it may have a wide applicability in the treatment of angiogenesis-dependent diseases.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
Received for publication August 6, 2007. Accepted for publication September 20, 2007.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  E. Roudier, N. Chapados, S. Decary, C. Gineste, C. Le Bel, J.-M. Lavoie, R. Bergeron, and O. Birot Angiomotin p80/p130 ratio: a new indicator of exercise-induced angiogenic activity in skeletal muscles from obese and non-obese rats? J. Physiol., August 15, 2009; 587(16): 4105 - 4119. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Zheng, S. Vertuani, S. Nystrom, S. Audebert, I. Meijer, T. Tegnebratt, J.-P. Borg, P. Uhlen, A. Majumdar, and L. Holmgren Angiomotin-Like Protein 1 Controls Endothelial Polarity and Junction Stability During Sprouting Angiogenesis Circ. Res., July 31, 2009; 105(3): 260 - 270. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Ernkvist, N. L. Persson, S. Audebert, P. Lecine, I. Sinha, M. Liu, M. Schlueter, A. Horowitz, K. Aase, T. Weide, et al. The Amot/Patj/Syx signaling complex spatially controls RhoA GTPase activity in migrating endothelial cells Blood, January 1, 2009; 113(1): 244 - 253. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
