|
|
||||||||
Cardiovascular Research Division, Health Care Research Center, Procter and Gamble Pharmaceuticals, Mason, Ohio 45040
Submitted 6 January 2004 ; accepted in final form 24 February 2004
| ABSTRACT |
|---|
|
|
|---|
inhibitor; protein; coronary artery; Tie-2; vascular endothelial growth factor receptor
27 million people and 16% of the population over age 55, indicates that PVD represents a clearly unmet medical need (50). Presently, approaches to treat occlusive cardiovascular disorders include interventional or surgical procedures to reestablish the circulation to the occluded artery (34). However, some patients with occluded arteries are asymptomatic due to the development or enlargement of collateral vessels that maintain blood flow distal to the site of occlusion (38). Nevertheless, in most patients with atherosclerosis, the collateral circulation remains insufficient for complete abrogation of symptoms or protection from ischemic tissue damage, prompting the search for novel therapeutic approaches to enhance collateral development.
The process of blood vessel development is orchestrated by angiogenic polypeptide growth factors that act via a family of conserved endothelial receptor tyrosine kinases (RTKs; Refs. 6, 12). For example, vascular endothelial growth factor receptor-2 (VEGFR-2) and Tie-2, the receptors for VEGF and the angiopoietins, respectively, are expressed principally in endothelial cells, and disruption of their function results in embryonic lethality due to distinct vascular defects. Although these receptors exhibit unique biological activity, their signaling is initiated by a common mechanism (20, 39). Growth factor binding results in receptor dimerization/oligomerization, activation of the intracellular kinase domain, and autophosphorylation of the receptor on specific tyrosine residues. Receptor phosphorylation serves a dual purpose, first to further increase tyrosine kinase activity, and second, to create binding sites for the recruitment of signaling molecules with phosphotyrosine binding motifs. These interactions lead to amplification of downstream effectors including activation of other protein kinases (e.g., mitogen-activated protein kinases, Akt), which converge to promote endothelial cell migration, proliferation, and survival, all of which are important events in blood vessel assembly.
Demonstrating the therapeutic potential of activating endothelial RTKs for the treatment of atherosclerotic cardiovascular disease, delivery of polypeptide growth factors in animal models, including fibroblast growth factor (FGF), VEGF, and angiopoietin-1 (Ang-1) results in improved collateral blood flow (4, 26, 42, 44, 49, 56). However, because polypeptide growth factors are expensive to produce and difficult to administer, their use as therapeutic agents is currently limited. In addition, like the development of the embryonic vasculature, optimal collateral development in vivo may require the coordinated activation of multiple RTKs and necessitate the delivery of multiple growth factors (4, 15, 37). Thus there has been growing interest in developing alternative approaches to activating endothelial RTKs for treatment of occlusive atherosclerotic cardiovascular disease.
An emerging alternative approach to enhancing RTK activation is to block the action of protein tyrosine phosphatases (PTPs) that counteract RTK activity (33, 57). Indeed, the tyrosine phosphatase PTP-1B negatively regulates insulin signaling by dephosphorylating the insulin RTK, and mice lacking PTP-1B exhibit enhanced insulin-mediated insulin receptor autophosphorylation, which results in enhanced insulin sensitivity (7, 25). As a proof of concept that PTP-1B inhibition could improve insulin signaling, vanadium compounds such as the potent but nonselective tyrosine phosphatase inhibitors sodium orthovanadate (NaVO4) and bis(maltolato)oxovanadium IV (BMOV) have been shown to augment insulin receptor activation and insulin action in diabetic rats (36, 47).
On the basis of these studies, we hypothesized that as with the insulin receptor, the actions of RTKs that promote angiogenesis and collateral development could be limited by PTP activity. Although most investigators use pan-tyrosine phosphatase inhibitors in cell lysis buffers to maintain RTK phosphorylation status, the concept of inhibition of PTPs to increase RTK signaling for promoting collateral blood flow is presently unsupported. To test this hypothesis, the organovanadium phosphatase inhibitor BMOV was used to demonstrate that inhibition of endothelial PTP activity enhanced the activation of endothelial RTKs, leading to augmentation of cellular responses important for angiogenesis and enhanced angiogenesis in the aortic ring explant assay. Consistent with these studies, BMOV also augmented collateral blood flow in a rat model of PVD. These data support the hypothesis that PTPs regulate endothelial RTK activation and provide the first demonstration of PTP inhibition as a novel approach for the treatment of occlusive atherosclerotic cardiovascular disease.
| MATERIALS AND METHODS |
|---|
|
|
|---|
Cell culture and receptor activation assays. Human umbilical vein endothelial cells (HUVECs; Clonetics) were grown to 90% confluence on 100-mm dishes. Cells were preincubated with BMOV at various concentrations as described in the text and then stimulated with either recombinant Ang-1 or recombinant VEGF165 (165-amino acid isoform of VEGF; R&D Systems) for 7 or 5 min, respectively. After stimulation, the cells were lysed in Triton lysis buffer (TLB: 20 mmol/l Tris·HCl, pH 8.0, 137 mmol/l NaCl, 10% glycerol, 1% Triton X-100, 2 mmol/l EDTA, 1 mM sodium orthovanadate, 1 mmol/l sodium fluoride, 1 mmol/l PMSF, 1 µg/ml leupeptin, and 1 µg/ml pepstatin) at 4°C. Cell lysates stimulated with Ang-1 were immunoprecipitated using 2 µg of anti-Tie-2 antibody (19) and 25 µl of protein A/G Plus. Lysates stimulated with VEGF165 were immunoprecipitated with 25 µl of wheat germ agglutinin (lectin from Triticum vulgaris). After overnight incubation, the complexes were centrifuged, washed once in TLB, and eluted by boiling in 30 µl of 1x sample buffer (50 mmol/l Tris·HCl, pH 6.8, 10% glycerol, 2% SDS, 0.1 mol/l DTT, and 0.1% bromphenol blue). Each sample (20 µl) was loaded onto a 6% SDS-PAGE gel, transferred to PVDF, and phosphotyrosine Western blotting was performed using anti-phosphotyrosine antibody PY-99. After exposure, the blots were stripped and reprobed with the anti-Tie-2 receptor Ab33 or anti-VEGFR-2 (51). The resulting films were scanned and quantitated using Quantity One software.
Cell survival assay. HUVECs were plated in 96-well tissue culture plates (20,000 cells/well) and serum starved in DMEM with 0.2% BSA. After 2 h, cells were treated with DMEM/0.2% BSA and growth factor and/or BMOV for 72 h. Changes in luminescence were measured using a Victor V plate reader.
Rat aortic ring assay of angiogenesis.
The isolated rat aortic ring model of angiogenesis was performed based on previously described studies (31). Male Sprague-Dawley rats (150175 g; Charles River) were anesthetized with Nembutal (50 mg/kg body wt ip) injection. The thoracic aorta was excised and transferred to endothelial basal media (EBM; Clonetics) that contained 1% antibiotics-antimycotic. Aortas were cross-sectioned at
1-mm intervals, and the resulting tissues were placed in a 24-well tissue-culture plate. Vessels were overlaid with 450 µl of collagen solution that consisted of 8 volumes of 3 mg/ml rat-tail collagen (Sigma) in 0.1% acetic acid, 1 volume of 10x medium 199 (phenol red free) supplemented with 100 mg/l L-glutamine, and 1 volume of 10x collagen buffer (0.05 N NaOH, 200 mmol/l HEPES, 260 mmol/l NaHCO3). The collagen solution was allowed to solidify for 1 h and was covered with 1 ml of EBM with or without BMOV. Images (TIFF) were taken using a SPOT camera connected to a Nikon inverted microscope, and vessel growth was quantitated using automated image-analysis software.
In vivo Tie-2 receptor phosphorylation.
Fasted rats (body wt, 250300 g) were infused with either saline or BMOV (1.73 mg/100 g of body wt) for 5 min, which was followed immediately by a 2-min infusion with either saline or Ang-1 (5 µg/100 g of body wt) via a carotid artery catheter (a polypropylene glycol-coated cannula was inserted 2.5 cm into the left carotid such that compounds were delivered into the aortic arch/descending aorta). Eight minutes after the dosing, the animals were euthanized and lung tissue was harvested using a 6-mm disposable punch biopsy. Samples were flash-frozen in liquid nitrogen and stored at 80°C until assayed. Frozen tissue (250 mg) was homogenized in RIPA buffer [50 mM Tris, pH 7.5, 150 mM NaCl, 1 mM EDTA, 1% Nonidet P-40, 0.25% SDS, 1 mM sodium orthovanadate, 1 mM NaF, 10 nM okadaic acid, and 1 complete protease inhibitor tablet (Roche)]. Homogenates were centrifuged at
21,000 g for 30 min at 4°C. Supernatants were recovered, and protein concentrations were determined by bicinchoninic acid assay (Pierce). Extracted protein (1 mg) was precleared with 25 µl of protein A/G-Plus agarose beads (Santa Cruz Biotechnology) for 1 h at 4°C. The Tie-2 receptor was immunoprecipitated as described above.
Femoral artery ligation and blood flow measurements. Male Sprague-Dawley rats (body wt, 325350 g; Taconic) were used in this study. The basic design of this model consisted of an initial sterile surgery at which time both femoral arteries were ligated (distal to the inguinal ligament and proximal to the femoral circumflex), and in the case of the VEGF groups, an osmotic pump was implanted for continuous delivery of the VEGF (15 µg·kg1·day1) through the jugular vein to allow systemic perfusion of the peptide (53). Dosing for the other treatment groups consisted of subcutaneous injections every 4 days beginning immediately after the surgery and continuing for up to 4 wk with either BMOV (dose, 0.8 mg/100 g of tissue; injection volume, 3.6 ml/kg) or vehicle (injection volume, 3.6 ml/kg). At the end of the treatment period, animals were instrumented (catheters were inserted into a carotid and caudal artery) for blood flow measurements using 0.5 ml of 15-µm fluorescent microspheres (Molecular Probes FluoroSpheres; orange and yellow-green; 5 x 105 microspheres) and subjected to exercise challenges (25 m/min at a 7° incline) on the treadmill to assess peak collateral flow. Tissues were digested with 4 N KOH overnight and processed (vacuum-filter extracted). Fluorescence in the supernatant was measured using a luminescence spectrometer (LS50B; PerkinElmer) at wavelengths of 352 and 506 nm for orange and yellow-green microspheres, respectively. Raw data were converted to blood flow (in ml·min1·100 g of tissue1) by the equation {[tissue fluorescence/tissue weight (in g)]/[reference blood fluorescence/blood withdraw rate (in ml/min)]} x 100 g. "Acute" animals were ligated on the same day as the exercise challenge. Animals in the 4-wk treatment groups for BMOV and vehicle were given injections on days 0, 4, 8, 12, 16, 20, and 24; in the 2-wk treatments, injections were given on days 0, 4, 8, and 12. VEGF165 was delivered continuously for either 2 or 4 wk. Flow values for left- and right-leg tissues were averaged together to create one value for each animal as long as even distribution was exhibited between the kidneys.
Statistical analysis. Values represent means ± SE. Statistical comparisons were made with Student's t-test and one- or two-way ANOVA (Tukey) where indicated with significance imparted at P values <0.05.
| RESULTS |
|---|
|
|
|---|
|
|
|
|
At 2 wk of treatment, blood flow in saline-injected animals was greater than that of acutely ligated animals (Fig. 5A), which indicates activation of innate mechanisms to restore collateral blood flow in these rats. VEGF165- or BMOV-treated animals exhibited similar increases in blood flow and flow-reserve recovery expressed as a percentage of maximal blood flow in sham-operated animals (177 ± 22 ml·min1·100 g of tissue1) to the calf muscle compared with either saline controls or acutely ligated animals (Fig. 5A). The blood flow reported value for sham-operated animals was similar to previous data (54). At 4 wk postsurgery, collateral blood flow in control and VEGF165-treated rats reached a plateau, as values were similar to those obtained at 2 wk (Fig. 5B). However, collateral blood flow in animals treated with BMOV for 4 wk was significantly improved compared with all 2-wk treatments and, surprisingly, exhibited a greater improvement in blood flow than VEGF165-treated animals for 4 wk. Furthermore, BMOV-treated rats recovered 55% of the maximal blood flow than sham-operated rats compared with 45 and 36% in VEGF and saline-treated rats, respectively. The augmented blood flow in the treatment groups occurred independently of calf weight normalized to body weight or alteration in heart rate (data not shown). Interestingly, improved blood flow in BMOV-treated rats occurred despite a modest decrease in peripheral blood pressure under resting (vehicle: 143 ± 3 mmHg, n = 6; VEGF: 135 ± 5 mmHg, n = 5; not significant vs. vehicle; BMOV: 130 ± 2 mmHg, n = 6; P < 0.05 vs. vehicle) or exercise conditions (vehicle: 148 ± 4 mmHg, n = 6; VEGF: 148 ± 6 mmHg, n = 5; not significant vs. vehicle; BMOV: 130 ± 4 mmHg, n = 6; P < 0.05 vs. vehicle or VEGF). Taken together with the findings in the aortic ring model, these results suggest that PTP inhibition can enhance the innate mechanisms of collateral vessel growth and development.
|
| DISCUSSION |
|---|
|
|
|---|
Mechanisms of blood flow enhancement by BMOV. Activation of endothelial RTKs such as VEGFR-2 and Tie-2 has been implicated in blood vessel development and growth (6, 12), and their respective ligands have been shown to augment collateral development in animal models of occlusive cardiovascular disease (15, 26, 42, 44, 49, 56). Supporting a role for endothelial phosphatases as negative regulators of endothelial RTK activation, BMOV enhanced activation of VEGFR-2 and Tie-2 in cultured endothelial cells. Additionally supporting this contention, BMOV also enhanced VEGF- and Ang-1-mediated endothelial cell survival, thereby demonstrating that augmentation of receptor activation translates into an enhanced biological response. Moreover, BMOV increased angiogenesis in the rat aortic ring, which indicates that phosphatase inhibition can enhance blood vessel growth and remodeling in a complex multicellular system.
Consistent with the in vitro studies, BMOV augmented Tie-2 activation in vivo and enhanced maximal collateral blood flow in a rat model of PVD. Surprisingly, after 4 wk of treatment, BMOV increased collateral blood flow beyond that observed with VEGF. The increase in blood flow in the treated groups was not due to differential regulation of vascular tone, because measurements were made during strenuous exercise to ensure maximal vasodilation and blood flow. The superior effect of BMOV on collateral blood flow over VEGF alone may reflect enhanced activation of Tie-2 or other angiogenic receptors including FGF, platelet-derived growth factor, and the insulin receptor. Indeed, previous studies in arterial insufficiency models indicated that VEGF and angiopoietin improved collateral blood flow beyond the effect of either growth factor alone, which suggests that optimal collateral development may require the activation of multiple angiogenic or vascular remodeling pathways (5, 41). Whichever pathways are affected, the results presented herein strongly suggest that PTP inhibition led to enhanced activation of endothelial RTKs and is the molecular mechanism underlying the beneficial effects of BMOV on collateral blood flow.
Although BMOV increased growth factor-induced RTK phosphorylation, endothelial cell survival, and angiogenic responses in vitro, PTP inhibition was also efficacious in the absence of exogenously added growth factors in vivo, which suggests that PTP inhibition may provide a clinical benefit by augmenting innate signaling cues for vascular remodeling in pathological states. A plausible stimulus for this effect may be mediated through amplification of inherent signaling mechanisms provided by changes in shear stress (16, 23, 48). Indeed, shear stress has been shown to be an important contributor in vascular remodeling as a source for ligand-independent activation of several RTKs including VEGFR-2, Tie-2, and the insulin receptor (21, 27). Furthermore, PTP inhibition and shear stress have been demonstrated to activate endothelial nitric oxide synthase in a calcium-independent manner (10), which would presumably promote vascular remodeling and collateral blood flow.
Although BMOV increased collateral blood flow in the hindlimb, we cannot at present distinguish whether this effect was due to angiogenesis (development of new collateral channels) or enlargement of existing collateral vessels (sometimes referred to as arteriogenesis). Previous studies have demonstrated that remodeling of preexisting collateral vessels occurs in this model, which suggests that collateral enlargement is the major contributor to blood flow improvements (18, 53). In addition, stimulation of angiogenesis (increased capillary density) in the rabbit hindlimb using adenoviral expression of VEGF165 did not improve hindlimb blood flow (17), which suggests enlargement of preexisting collaterals, not angiogenesis, is the most prominent factor to collateral blood flow. However, BMOV did enhance angiogenesis in the rat aortic ring model, thereby indicating that neovascularization cannot be excluded. In the present study, using fluorescent microspheres to measure collateral blood flow precluded measurements of anatomical changes in the hindlimb vasculature in response to phosphatase inhibition, but future studies will address this question.
PTPs implicated in regulation of endothelial signaling. PTPs play important roles in a number of biological systems including the nervous system, immune system, and hematopoietic system (28, 30). In addition, some PTPs, such as SHPTP-2, positively regulate growth factor receptor signaling (28, 59) and may perhaps provide an alternative explanation for inhibition of vessel growth in the rat aortic ring model at higher concentrations of BMOV. On the basis of these findings, it is likely that a selective PTP inhibitor will have advantages both in safety and efficacy. Thus it is important to identify the target PTPs that are responsible for the negative regulation of endothelial RTK signaling.
Although our study demonstrates the utility of PTP inhibition for the treatment of vascular disease, we cannot at present distinguish which PTPs may be involved in the beneficial effects of BMOV. However, a number of PTPs have recently been implicated in the regulation of endothelial RTK activation and are likely candidates. For example, we have shown that adenoviral-mediated overexpression of HCPTP-A, a PTP that is identified by its association with VEGFR-2, attenuated VEGFR-2 activation, endothelial cell proliferation and migration, and inhibited angiogenesis in the rat aortic ring assay (19). In another study, small-molecule inhibitors of PTP-1B were found to enhance VEGFR-2 activation in endothelial cells and to augment angiogenesis in a Matrigel plug assay (43). These findings are particularly relevant to the present study, because BMOV was recently shown to inhibit several of these PTPs including HCPTP-A, HPTP-
, PTP-1B, and SHP-2 (36). Future studies will aim at deciphering the specific PTP(s) involved in modulating vascular remodeling.
Intriguingly, just as the VEGFR family and the Tie family of RTKs are expressed predominantly in endothelial cells and play important roles in vascular development, several receptor type PTPs (rPTPs) are also expressed in endothelial cells, but their roles in vascular development have not been fully clarified (2). HPTP-
and HPTP-
, two closely related rPTPs, are both expressed predominantly in endothelial cells and their expression increases with cell density, a finding that suggests a role for these receptors in growth control (3, 11). Consistent with this possibility, HPTP-
was recently shown to interact with Tie-2 and negatively regulate its phosphorylation when cotransfected in COS-1 cells (9). Another receptor PTP, PTP-
, is also expressed predominantly in endothelial cells, and retroviral overexpression in HUVECs negatively regulated survival and proliferation (45). Although these data suggest a role for receptor PTPs in regulation of endothelial signaling, clearly much work is needed to understand their functions during blood vessel assembly and remodeling.
Limitations to current approach. Although this study demonstrates the therapeutic potential of PTP inhibition, our data suggest that its usefulness may be limited. BMOV promoted angiogenesis in the aortic ring at low concentrations; however, vessel formation was inhibited with higher doses. Moreover, the positive effects of BMOV on endothelial cell survival were reversed at higher concentrations (>100 µmol/l; data not shown), which indicate that selective PTP inhibition is likely necessary to provide clinical benefit. In addition, those promoting proangiogenic therapy must be mindful of the balance between enhancing collateral remodeling and potentially augmenting pathological angiogenesis. Thus patients receiving any proangiogenic therapy should be screened for conditions such as diabetic retinopathy or cancer that could be accelerated by enhancing angiogenesis.
Potential advantages of PTP inhibition for proangiogenesis and therapeutic implications. It is well accepted that patients with a robust network of preexisting collateral blood vessels fair better in response to occlusive artery disease (38). Furthermore, the risk of death from cardiovascular events in patients with PVD are similar to patients with coronary artery or cerebrovascular disease (14), which emphasizes the importance of developing novel means of establishing collateral vessels in high-risk patients.
To this end, data presented in this study suggest that inhibiting PTPs to enhance RTK activation and collateral remodeling may offer advantages over current approaches using growth factors. First, PTP inhibition is amenable to the development of a small molecule (57) and does not depend on the delivery of peptides in vivo, which are more difficult to produce and deliver and can be degraded by endogenous peptidases (8). Identification of a selective small molecule seems plausible based on the limited number of PTPs that have been identified as regulators of endothelial RTKs and the recent data that indicate success in generating specific small-molecule inhibitors of PTP-1B (58). Second, inhibition of PTPs could circumvent the need for administering multiple growth factors, providing that a single phosphatase can be identified that regulates multiple receptors. The ability of BMOV to enhance blood flow in vivo in the absence of exogenously added growth factors suggests that PTP inhibition may not require a combinatory approach, although this possibility requires further investigation. Finally, decreasing PTP activity could allow for more control of the vascular remodeling process because PTP inhibitors would amplify the body's physiological response to ischemia. This may be critical to successful proangiogenic therapy, because administration of large amounts of exogenous VEGF can lead to untoward side effects including vascular leakage, edema, inflammation, and even angioma formation (29, 40, 46). Likewise, deciphering the precise roles of phosphatase(s) involved in the improvement of collateral blood flow observed in this study is critical in separating the beneficial and adverse consequences of PTP inhibition. In any case, our findings indicate that PTPs regulate endothelial RTK activation and validate PTP inhibition as a novel approach for proangiogenic therapy of occlusive cardiovascular disease.
| ACKNOWLEDGMENTS |
|---|
| FOOTNOTES |
|---|
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
| REFERENCES |
|---|
|
|
|---|
, a tyrosine phosphatase expressed in endothelium, negatively regulates endothelial cell proliferation. Am J Physiol Heart Circ Physiol 281: H396H403, 2001.This article has been cited by other articles:
![]() |
A. N. Carr, B. W. Howard, H. T. Yang, E. Eby-Wilkens, P. Loos, A. Varbanov, A. Qu, J. P. DeMuth, M. G. Davis, A. Proia, et al. Efficacy of systemic administration of SDF-1 in a model of vascular insufficiency: Support for an endothelium-dependent mechanism Cardiovasc Res, March 1, 2006; 69(4): 925 - 935. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. Duerschmied, L. Olson, M. Olschewski, A. Rossknecht, G. Freund, C. Bode, and C. Hehrlein Contrast ultrasound perfusion imaging of lower extremities in peripheral arterial disease: a novel diagnostic method Eur. Heart J., February 1, 2006; 27(3): 310 - 315. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. Kappert, K. G. Peters, F. D. Bohmer, and A. Ostman Tyrosine phosphatases in vessel wall signaling Cardiovasc Res, February 15, 2005; 65(3): 587 - 598. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH |