Recent studies have demonstrated that lymphocyte-derived microparticles (LMPs) impair endothelial cell function. However, no data currently exist regarding the contribution of LMPs in the regulation of angiogenesis. In the present study, we investigated the effects of LMPs on angiogenesis in vivo and in vitro and demonstrated that LMPs strongly suppressed aortic ring microvessel sprouting and in vivo corneal neovascularization. In vitro, LMPs considerably diminished human umbilical vein endothelial cell survival and proliferation in a concentration-dependent manner. Mechanistically, the antioxidants U-74389G and U-83836E were partially protective against the antiproliferative effects of LMPs, whereas the NADPH oxidase (NOX) inhibitors apocynin and diphenyleneiodonium significantly abrogated these effects. Moreover, LMPs increased not only the expression of the NOX subunits gp91phox, p22phox, and p47phox, but also the production of ROS and NOX-derived superoxide (O2−). Importantly, LMPs caused a pronounced augmentation in the protein expression of the CD36 antiangiogenic receptor while significantly downregulating the protein levels of VEGF receptor type 2 and its downstream signaling mediator, phosphorylated ERK1/2. In summary, LMPs potently suppress neovascularization in vivo and in vitro by augmenting ROS generation via NOX and interfering with the VEGF signaling pathway.
- NADPH oxidase
- vascular endothelial growth factor receptor type 2
microparticles (MPs) are small membrane vesicles (14, 24) released upon activation or during apoptosis from various cell types including lymphocytes, platelets, and endothelial cells (23, 31, 38). MPs have been implicated in the pathogenesis of cardiovascular and inflammatory diseases that are associated with vascular damage and impaired angiogenesis. Of relevance, lymphocyte-derived microparticles (LMPs) have been detected at elevated levels in atherosclerotic plaques (32) and in patients with myocardial ischemia or preeclampsia (31, 49). Recent observations have further demonstrated that MPs released from apoptotic lymphocytes or from plasma of diabetic patients induce endothelial dysfunction by modulating nitric oxide (NO) pathways (46).
Angiogenesis is involved in physiological events such as embryonic development and wound healing, as well as in pathological conditions such as tumor growth, diabetic retinopathy, and chronic inflammation (8, 18). This tightly regulated and complex process involves endothelial cell survival, proliferation, migration, differentiation, and tube formation (17). It is widely accepted that angiogenesis is determined by a relative balance between pro- and antiangiogenic factors (22). Vascular endothelial growth factor (VEGF) is one of the most potent angiogenic factors known and exerts its mitogenic effects primarily through the VEGF receptor type 2 (VEGFR2), which is almost exclusively expressed on endothelial cells. Moreover, VEGFR2 possesses intrinsic tyrosine kinase activity and therefore transducers signals leading to stimulation of mitogen-activated protein kinases (MAPK) (44). Nonetheless, angiogenesis is also determined by the presence of angiostatic molecules. CD36 is a potent antiangiogenic surface receptor that is expressed by microvascular endothelial cells and binds to numerous ligands, including thrombospondin (TSP)-1, an endogenous inhibitor of angiogenesis (15). Interestingly, a previous study demonstrated that activation of CD36 by TSP-1 downmodulated VEGFR2 expression and p38 MAPK phosphorylation (39). Then again, increased CD36 expression has been associated with pro-oxidative conditions such as atherosclerosis, inflammation, and ischemia (12, 15, 40).
Reactive oxygen species (ROS) are involved in the development and progression of various cardiovascular diseases, and oxidative stress is considered the central mechanism (10). Furthermore, oxidative stress is thought to contribute to angiogenesis by mediating endothelial cell proliferation and migration (30, 43). The major source of ROS in endothelial cells is NADPH oxidase (NOX); increasing NOX-driven ROS stimulates VEGF expression and enhances VEGFR2 autophosphorylation (28, 47). In this context, LMPs could be one of the key factors linking oxidative stress and angiogenesis.
Previously published studies have documented that microparticles released from platelets (PMPs) induce angiogenesis and stimulate postischemic revascularization, whereas endothelial cell-derived microparticles (EMPs) suppress angiogenesis by altering the redox balance (25, 35). Nevertheless, the involvement of LMPs in regulating angiogenesis has yet to be established. Based on the present study, we report for the first time that LMPs significantly inhibited blood vessel formation in the ex vivo aortic ring angiogenesis assay and in vivo corneal neovascularization (CNV) model. Moreover, the current findings suggest that LMPs strongly diminish VEGF-induced endothelial cell proliferation and migration by enhancing ROS production primarily from NOX, with accompanied increases in CD36 expression and suppression of VEGFR2 signaling.
MATERIALS AND METHODS
Compounds and reagents.
Actinomycin D, 3-(4,5-dimethyl-2-thiazolyl)-2,5-diphenyl tetrazolium bromide (MTT), N,N-dimethyl-9,9′-biacridinium dinitrate (lucigenin), and angiotensin were obtained from Sigma Aldrich. β-Actin was obtained from Novus Biologicals. Flk-1 (VEGFR2) rabbit polyclonal antibody and horseradish peroxidase-linked anti-rabbit IgG, antibodies against gp91phox, p22phox (FL-195), p47phox, ERK1/2, phospho-ERK1/2, TSP-1, and rabbit polyclonal CD36 antibody were obtained from Santa Cruz Biotechnology (Santa Cruz, CA). U-83836E and U-74389G were obtained from Biomol (Plymouth Meeting, PA); [3H]thymidine from Amersham (Mississauga, ON, Canada); and human recombinant (hr)VEGF, apocynin, and diphenyleneiodonium (DPI) from Calbiochem (La Jolla, CA). Mitomycin C was obtained from Fluka Biochemika, and annexin-V-Cy5 from BD Pharmagen (San Diego, CA). The Vybrant apoptosis assay kit, propidium iodide (PI), and fluorescent microbeads (1 μm) were obtained from Molecular Probes (Eugene, OR). NADPH was obtained from Roche Diagnostics (Laval, QC, Canada).
CEM T cells were purchased from ATCC (Manassas, VA) and cultured with X-VIVO medium (Cambrex, Walkersville, MD). Human umbilical vein endothelial cells (HUVEC) were purchased from Cambrex (Walkersville, MD) and cultured as recommended. The immortalized human microvascular endothelial cell line-1 (HMEC-1) was kindly supplied by Dr. F. J. Candal (Centers for Disease Control and Prevention, Atlanta, GA). HMEC-1 were grown in endothelial basal medium (Cambrex) supplemented with 10% fetal bovine serum (GIBCO, Gaithersburg, MD), 100 μg/ml streptomycin, 100 U/ml penicillin, 10 ng/ml epidermal growth factor (Becton Dickinson, Oakville, ON, Canada), and 1 μg/ml hydrocortisone (Sigma).
LMPs were generated as described previously (33). Briefly, CEM T cells were treated with 0.5 μg/ml actinomycin D for 24 h, and a supernatant was obtained by centrifugation at 750 g for 15 min and then at 1,500 g for 5 min to remove cells and large debris. MPs from the supernatant were washed after three centrifugation steps (50 min at 12,000 g) and recovered in saline or basic cell culture media. Washing medium from the last supernatant was used as control. LMPs were characterized with annexin V staining by fluorescence-activated cell sorting (FACS) analysis and gated using 1.0-μm beads in which 97% of MPs (≤1 μm) were annexin-V-Cy5 positive (41). The concentrations of LMPs were determined using the Bio-Rad protein assay. Using the same protocol, we also generated LMPs from hyperoxia (95% O2, 36 h)- or hypoxia (5% O2, 36 h)-exposed CEM T cells.
Six-week-old male C57BL/6 mice purchased from Charles River (St-Constant, QC, Canada) were used following a protocol approved by the Sainte-Justine Research Center Animal Care Committee.
Aortic ring angiogenesis assay.
The aortic ring assay was performed as described previously (37). In brief, 1-mm thoracic aortas were embedded in three-dimensional growth factor-reduced Matrigel (BD Biosciences) and cultured in EGM-2 medium at 37°C. The culture medium was changed on day 3, and the aortic rings were treated with saline or 30 μg/ml LMPs until day 7. Aortic rings were photographed on days 5 and 7 with a Nikon eclipse TE300 inverted microscope. The angiogenic response was determined by measuring the area of neovessel formation using Image-Pro Plus software.
Murine model of corneal neovascularization.
Angiogenesis was investigated in vivo using a murine model of CNV as described previously (37). Briefly, each mouse was anesthetized with isoflurane (Abbott Canada), and topical proparacaine (Alcon Canada) and 2 μl of 0.15 M NaOH were applied to the central cornea. The corneal and limbal epithelia were removed by scraping with a scalpel. Gentamicin sulfate ophthalmic solution (Sabex, Quebec, Canada) was instilled immediately following epithelial denudation. Buprenorphine (0.05 mg/kg; Schering-Plough) was administered postoperatively for analgesia. Twenty-four hours after corneal injury, mice were randomly divided into two groups that received either saline or 50 μg/ml LMPs. Treatments were administered topically three times daily for 7 days, after which corneas were harvested, flat mounted, and immunostained with FITC-conjugated anti-CD31. Images were captured with a Nikon DXM 1200 digital camera using Nikon ACT 1 version 2.62 software. The CNV was quantified in a masked fashion by using Adobe Photoshop 7.0 image analysis software. The total corneal surface area was outlined by using the innermost vessel of the limbal arcade as the border, and the ratio [(neovascularized area/total cornea area) × 100] was used to provide a measure of the percentage of vascularized cornea.
Cell viability assay.
Cells at ∼60% confluence were incubated for 24 h with vehicle or the indicated concentrations of LMPs. Cell viability was estimated by mitochondria-dependent reduction of MTT. Essentially, MTT [0.5 mg/ml in PBS (pH 7.4)] was added to the culture medium and incubated at 37°C for 3 h, the medium was aspirated, the formazan product was solubilized with acidified isopropanol, and the optical density was read at 545 nm with reference wavelength at 690 nm.
[3H]thymidine incorporation assay.
HUVEC (4 × 104) were plated and serum starved for 24 h. After synchronization, cells were cultured in complete medium with vehicle or 10 μg/ml LMPs for an additional 24 h. Thereafter, 1 μCi/ml [3H]thymidine was added to each well and incubated for 24 h. [3H]thymidine DNA incorporation was assayed by scintillation counting.
HUVEC were treated with or without 10 μg/ml LMPs for 8, 18, and 24 h and then treated with reagents from the Vybrant apoptosis assay kit (Molecular Probes, Invitrogen), followed by flow cytometry analysis according to the manufacturer's protocol. The rate of apoptosis or necrosis was expressed as the percentage of apoptotic cells relative to the total number of cells per condition.
Measurement of ROS generation and the NADPH oxidase assay.
Induction of ROS was measured using the fluoroprobe 2′,7′-dichlorohydofluorescein diacetate (DCFDA; Molecular Probes). Endothelial cells were cultured in 24-well plates and treated with LMPs and/or apocynin at indicated concentrations for 3 h or with angiotensin (ANG II; 100 nM) for 45 min as a positive control. Cells were stained with DCFDA (10 μM) for another 30 min. After staining, the extracellular dye was washed twice with 10 mM HEPES buffer (pH 7.4), and the fluorescence was measured at an excitation wavelength of 485 nm and an emission wavelength of 535 nm, using a multiwell fluorescent plate reader (Wallac 1420 VICTOR multilabel counter).
NOX activity was measured using the lucigenin-enhanced chemiluminescence method as described previously (20) Briefly, HUVEC were treated with 10 μg/ml LMPs for different time periods, washed in ice-cold PBS, harvested, and homogenized via sonication (1 s) (Brandson Sonifier 150) in lysis buffer (20 mM KH2PO4, pH 7.0, 1 mM EGTA, 10 mM complete protease inhibitor cocktail). Homogenates were centrifuged at 800 g at 4°C for 10 min to remove the unbroken cells and debris, and aliquots were used immediately. To initiate the assay, 100-μl aliquots of the homogenates were added to 900 μl of 50 mM phosphate buffer, pH 7.0, containing 1 mM EGTA, 150 mM sucrose, 5 μM lucigenin, and 100 μM NADPH. Photon emission in terms of relative light units was measured in a luminometer every 2 min for 30 min. There was no measurable activity in the absence of NADPH. Superoxide anion production was expressed as relative chemiluminescence (light) units per microgram of protein. Protein content was determined using the Bio-Rad protein assay.
Western blot analysis.
Cells were plated at a density of 1 × 106 cells per 100-mm plate and incubated with 7.5, 10, and 15 μg/ml LMPs for 24 h. Soluble proteins were extracted using cell lysis buffer [10 mM Tris·HCl, 1.5 mM MgCl2, 1 mM DTT, 1 μM pepstatin, 0.75 mM EDTA, 1% (vol/vol) SDS, and 10 mM protease inhibitor cocktail (pH 7.5; Roche)]. After centrifugation, the supernatant was collected and total protein concentration was determined (Bio-Rad assay). Protein (25 μg) was fractionated using SDS-PAGE. The resolved proteins were transferred onto a polyvinylidene difluoride membrane on a semidry electrophoretic transfer cell (Bio-Rad) for Western blot analysis. Membranes were blocked and then incubated overnight at 4°C with an anti-VEGFR2 polyclonal antibody (1:500 dilution), anti-gp91phox antibody (1:100), anti-p22phox antibody (1:200), anti- p47phox antibody (1:200), phospho-ERK1/2 antibody (1:200), ERK1/2 antibody (1:200), TSP-1 antibody (1:400), and anti-CD36 polyclonal antibody (1:400). After washing, membranes were incubated with a horseradish peroxidase-linked anti-rabbit IgG (1:5,000) for 1 h at room temperature. β-Actin was used as a loading control (1:10,000). Proteins were visualized using the ECL Western blotting detection system (Perkin Elmer).
Cell migration assay.
Two cell migration assays were used to facilitate our analysis. Cell migration was first determined using a coverslip border migration assay. Briefly, 0.5 × 106 HUVEC were seeded onto 12-mm coverslips in a 24-well plate. Cells were serum starved for 4 h, and proliferation was inhibited by adding 10 μg/ml mitomycin C for 30 min. Next, coverslips were carefully removed, washed with fresh medium, and transferred into a 12-well plate containing 10 ng/ml VEGF in the presence or absence of 10 μg/ml LMPs. Images were captured between 48 and 72 h by using an Axiovert 200M inverted microscope (Zeiss). At 72 h, the coverslips were removed and the proportion of migrated cells was quantified by MTT assay.
The Boyden chamber migration assay was also used. A 96-well chemotaxis chamber with a 5-μm polycarbonate filter was purchased from Corning (Corning, NY). The filter was placed over a bottom chamber containing 10 ng/ml hrVEGF; 10,000 HUVEC were seeded to each well in the upper chamber. For testing the effects of LMPs and apocynin on the cell migration, HUVEC were incubated with LMPs and/or apocynin in the upper chambers. The assembled chemotaxis chamber was incubated for 24 h at 37°C with 5% CO2 to allow cells to migrate through the filter. Nonmigrated cells on the upper surface of the filter were removed by scraping with a wiper tool (Neuro Probe, Gaithersburg, MD) and a cotton swab, and the filter was stained with Coomassie blue. The total number of migrated cells per well were counted; the assays were performed in quadruplicate.
All experiments were repeated at least three times, and values are presented as means ± SE. Data were analyzed using one-way ANOVA followed by post hoc Bonferroni tests for comparison among means. Statistical significance was set at P < 0.05.
LMPs suppress aortic ring angiogenesis and in vivo corneal neovascularization.
The first objective was to determine whether LMPs affect vessel development. For this purpose, we utilized the aortic ring angiogenesis assay and a pathophysiologically relevant CNV model that is largely driven by VEGF. Incubation of aortic rings with saline or 30 μg/ml LMPs for 48 and 96 h significantly reduced neovessel formation by 50% (2.2 ± 0.2 vs. 1.1 ± 0.1 mm2; P < 0.05) and 58% (7.7 ± 0.3 vs. 3.2 ± 0.5 mm2; P < 0.001) (Fig. 1, A and B), respectively. Having established that LMPs inhibit ex vivo angiogenesis, we analyzed its significance in vivo by treating mice subjected to CNV with saline or 50 μg/ml LMPs three times daily for 7 days. Compared with saline treatment, LMPs caused a 23% reduction in CNV (80.0 ± 3.6 vs. 61.6 ± 2.3%; P < 0.001; Fig. 1, C and D).
LMPs inhibit human endothelial cell survival and proliferation.
Cell survival and proliferation are critical steps during angiogenesis. To determine the effect of LMPs on vascular cell survival, we exposed HUVEC and HMEC-1 to different concentrations of LMPs and assessed their viability by MTT assay. LMPs significantly diminished cell viability in both cell types in a concentration-dependent manner (Fig. 2, A and B). To determine whether the effect of LMPs on cell proliferation is stimulus dependent, we generated LMPs from hyperoxia or hypoxia exposure. LMPs produced under both hyperoxic and hypoxic conditions potently suppressed HUVEC proliferation (45.8 ± 1.4 and 50.8 ± 2.3%, respectively; P < 0.001 vs. control) to a degree comparable to actinomycin D-derived LMPs (49.0 ± 0.%8; P < 0.001 vs. control). This indicates that the effects of LMPs are not stimulus dependent.
The observed reduction in cell survival could be caused by decreased cell proliferation or increased apoptosis or necrosis. [3H]thymidine DNA incorporation was applied, and LMPs (10 μg/ml) reduced HUVEC proliferation by 60% (P < 0.001; Fig. 2C). To ascertain whether LMPs were inducing apoptosis or necrosis, we doubled-labeled both LMP-treated and control HUVEC with FITC-conjugated annexin-V and PI; however, induction of apoptosis or necrosis was not observed under any test conditions (P > 0.05) (Fig. 2D).
Antioxidants partially block the antiproliferative effects of LMPs.
Previous studies have shown that EMPs increase superoxide production (9) and lead to impairment of angiogenic pattern (35). Moreover, NOX, a major source of superoxide free radicals, is highly expressed by endothelial cells (16). We therefore postulated that LMPs exert their antiangiogenic properties via oxidative stress mechanisms. To address this hypothesis, we utilized two well-known lipid peroxidation inhibitors, namely, U-83836E and U-74389G (3, 29, 45), and tested their ability to attenuate the antiproliferative effects of LMPs. U-83836E and U-74389G at 5 and 10 μM concentrations, respectively, led to a partial but statistically significant increase in cell proliferation compared with LMP treatment alone (P < 0.05; Fig. 3A). In addition, pretreatment of HUVEC with two specific NOX inhibitors, apocynin (1.5 mM) and DPI (5 μM), significantly abrogated the LMP-induced antiproliferative effects (P < 0.05 and P < 0.001, respectively; Fig. 3, B and C).
LMPs increase ROS and NOX activity.
Having demonstrated the important role of oxidative stress in LMP-mediated activities, we were next interested in investigating the effects of LMPs on ROS generation. The latter was determined by measurement of intracellular ROS levels using dichlorofluorescein (DCF) fluorescence following a 3-h pretreatment with 10 μg/ml LMPs. As shown in Fig. 4A, compared with control, LMPs significantly increased ROS production as indicated by a rise in the DCF signal (P < 0.05). Moreover, LMP-induced ROS generation was significantly attenuated by pretreatment with apocynin (1.5 mM; P < 0.05).
Because the superoxide-generating NADPH oxidase has been described to largely contribute to ROS formation in endothelial cells (27), we next investigated the effect of LMPs on superoxide generation from this enzyme. Superoxide anion production was measured in LMP-treated HUVEC as lucigenin-enhanced chemiluminescence using NADPH as the substrate. As indicated in Fig. 4B, LMPs increased the rate of superoxide formation after 1 h of incubation and reached a peak after 8 h (P < 0.001).
LMPs induce protein levels of p22phox, p47phox, gp91phox, and the CD36 scavenger receptor.
Given the ability of LMPs to induce NOX activity, we next investigated their effect on the expression of p22phox, p47phox, and gp91phox, which are critical subunits of NADPH oxidase (48). Indeed, LMPs strongly upregulated p22phox, p47phox, and gp91phox protein expression in a concentration-dependent fashion (P < 0.05; Fig. 5, A–F). Although LMPs demonstrated very low-level expression of p22phox, p47phox and gp91phox were undetected.
The CD36 scavenger receptor and its endogenous ligand TSP-1 are potent inhibitors of in vitro and in vivo angiogenesis (37, 39) whose expression is potentiated in pro-oxidative environments (12, 13, 40) as well as by NOX activation (4). In this context, HMEC-1 treated with 10 and 15 μg/ml LMPs dose-dependently augmented CD36 protein levels by 1.9- and 2.3-fold, respectively (Fig. 5, G and H), whereas expression of TSP-1 was not significantly changed. Moreover, TSP-1 was not detected in LMPs per se, which is in agreement with the published results from the proteomic analysis of LMPs (36).
LMPs mediated antimigratory effects are reversed by NOX inhibitors.
Because cell migration plays a pivotal role in angiogenesis, we sought to elucidate the effect of LMPs on VEGF-induced cell migration. HUVEC were plated onto coverslips and exposed to 10 ng/ml VEGF with or without LMPs. Cell migration was substantially decreased by 58% after 72 h of LMP treatment (P < 0.001; Fig. 6, A and B).
Cell migration was also evaluated using the modified Boyden chamber assay. LMPs strongly inhibited VEGF-induced cell migration by 40% (P < 0.001; Fig. 6C), and apocynin (1.5 mM) was able to partially rescue LMP-mediated antimigratory effects (P < 0.01 vs. LMP; Fig. 6C).
LMPs reduce VEGFR2 protein and phospho-ERK levels.
Having observed that LMPs induced CD36 expression, we surmised that LMPs were further suppressing angiogenesis by antagonizing the VEGF signaling pathway. This hypothesis is corroborated by evidence that activation of CD36 leads to suppression of VEGF-induced VEGFR2 phosphorylation (39). Accordingly, HUVEC proliferation was assessed following preincubation with 1.5 μg/ml anti-VEGFR2 antibody in the presence or absence of LMPs (10 μg/ml). As expected, the anti-VEGFR2 antibody alone strongly decreased cell proliferation (P < 0.01); however, cotreatment with the anti-VEGFR2 antibody and LMPs did not result in a synergistic reduction of cell proliferation (P > 0.05 compared with anti-VEGFR2 antibody treatment alone; Fig. 7A). Consistent with these data, Western blot analysis of HUVEC treated with 7.5 and 15 μg/ml LMPs showed a dose-dependent downregulation of VEGFR2 protein expression by 50 and 65%, respectively, compared with control (P < 0.01; Fig. 7, B and C). Phospho-ERK1/2 levels were also significantly inhibited by 35% (P < 0.05; Fig. 7, D and E).
MPs are known to contribute to the pathogenesis of cardiovascular diseases, including inflammation and vascular dysfunction. Another important action of MPs in the vascular system is their ability to modulate angiogenesis (34). Nevertheless, despite the escalation in MP research, very little is known regarding the role of LMPs in regulating angiogenesis (19, 34). Presently, we report that LMPs inhibited angiogenesis in vivo and in vitro by suppressing vascular cell survival, proliferation, and migration. Of particular interest, our data demonstrate that LMPs induced ROS production via NOX activation, whereas antioxidants and NOX inhibitors attenuated the antiangiogenic effects of LMPs. Furthermore, through CD36 induction and VEGFR2 and phospho-ERK1/2 downregulation, we provide evidence that LMPs interfered with the VEGF signaling pathway. Together, these findings strongly support a role for LMPs in regulating angiogenesis during pathological conditions.
MPs are released from the plasma membrane during cell activation by apoptosis, shear stress, or agonists. In our study, microparticles were obtained by apoptosis from T lymphocytes treated with actinomycin D. Moreover, the characteristics of MPs seem to depend on the mechanism of stimulation and the activation status of the cell from which they originate (25, 35). This is clearly highlighted by the reported effects of MPs on angiogenesis. For example, although we have shown that LMPs possess antiangiogenic properties, others have shown that MPs from endothelial cells inhibit, whereas platelet-derived MPs promote, angiogenesis (7, 25, 35). In our study, the antiangiogenic effects of LMPs seem to occur as a result of decreased cell proliferation and not increased cell apoptosis or necrosis (Fig. 2). This is in agreement with observations by Martin et al. (33), who showed that pathophysiological levels of LMPs failed to induce endothelial cell apoptosis. Nonetheless, the pathophysiological relevance of our study is limited given that there is presently no standardized method for measuring MPs, and one cannot rely on the numbers provided by other studies, since they highly depend on the method of measurement and the sensitivity of detection (1, 6, 26).
It has been documented that oxidative stress is one of the central mechanisms responsible for endothelial cell dysfunction (16, 21). The major sources of ROS in endothelial cells are endothelial nitric oxide synthase (eNOS) and NOX. In line with this, there is a general consensus that NO inhibits both vascular smooth muscle and endothelial cell proliferation (11, 50, 51). In our study, nitrite levels were unchanged by LMP treatment, and eNOS blockers did not prevent the antiproliferative effects of LMPs (data not shown). Conversely, LMPs increased ROS levels and NOX activity (Fig. 4) as well as upregulating expression of the gp91phox, p22phox, and p47phox NOX subunits (Fig. 5). Consistent with this, inhibition of NOX partly abrogated the inhibitory effects of LMPs on both cell proliferation and migration (Figs. 3, B and C, and 6C). Collectively, these results support the suggestion that the NOX pathway, and not the eNOS-NO signaling pathway, is involved in generating ROS that mediate the angiostatic effects of LMPs.
One of the detrimental consequences of oxidative stress is peroxidation of membrane lipids. Lipid peroxidation induces site-specific changes in the organization of the phospholipid bilayer, thus leading to cellular dysfunction. The lipid peroxidation inhibitors U-74389G and U-83836E are lipophilic steroid compounds that intercalate into biological membranes, thus enhancing their stability in the event of oxidative stress (5, 42). In this study, both compounds partially attenuated the antiproliferative effects of LMPs (Fig. 3A), thus suggesting that the angiostatic activities of LMPs also involve increased lipid peroxidation.
Ample studies have demonstrated that oxidative stress stimulates CD36 expression and that antioxidants attenuate its expression and function (12, 13, 40). It was therefore intriguing to observe that LMPs upregulated CD36 expression, which is consistent with the pro-oxidant actions of LMPs (Figs. 3, 4, and 6). However, we presume that the LMP-mediated upregulation of CD36 is TSP-1 independent, since LMPs had no significant effect on TSP-1 expression. Moreover, because CD36 is a well-established antiangiogenic receptor (15, 37), it is tempting to speculate that the generation of ROS by LMPs occurs upstream of the induction of CD36 with subsequent suppression of the VEGF/VEGFR2 signaling pathway, as has been proposed by us and others (37, 39).
It is well known that VEGF plays a pivotal role in developmental and pathological angiogenesis. VEGF stimulates angiogenesis through VEGFR2 (KDR/Flk-1), which is expressed mainly on endothelial cells (2, 44). In the present study, several lines of evidence supported our hypothesis that LMPs antagonized the VEGF/VEGFR2 pathway. First, we demonstrated that LMPs potently inhibited VEGF-induced inflammatory CNV (Fig. 1, C and D). Second, VEGF-induced endothelial cell migration was dramatically reduced by LMPs (Fig. 6). Third, inhibition of VEGFR2 activity had no synergistic effect on the antiproliferative effects of LMPs, suggesting that both VEGFR2 and LMPs signal via the same pathway (Fig. 7A). Finally, we showed that LMPs significantly downregulated VEGFR2 and phosphorylated ERK1/2 expression (the main downstream effector of the VEGF signaling pathway) (Fig. 7) while increasing CD36 protein levels (Fig. 5, G and H), a known negative regulator of this pathway (37, 39).
In conclusion, we have provided evidence for the first time that MPs from T cells inhibit angiogenesis in vivo and in vitro. We have demonstrated that LMPs impair vascular cell survival, proliferation, and migration. The present data also suggest that LMPs regulate angiogenesis by acting through the NADPH oxidase and VEGFR2 pathways. Given the pivotal role of the VEGF/VEGFR2 signaling pathway in angiogenesis, understanding the mechanisms of how LMPs interrupt VEGFR2 signaling could provide attractive therapeutic strategies aimed at reducing the deleterious effects of MPs on the vascular system.
Perspectives and Significance
Having long been considered as cellular debris, MPs constitute reliable markers of vascular damage. Released into biological fluids, MPs are involved in the modulation of key functions including immunity, inflammation, vascular remodeling, and angiogenesis. LMPs can be considered a hallmark of stress-injured or dying lymphocytic cells and may be recognized in the future as a marker of lymphocytic dysfunction. Our data demonstrate that LMPs have considerable impact on angiogenesis in vitro and in vivo. In view of this, LMPs might be important contributors to the pathogenesis of diseases that are accompanied by impaired angiogenesis and could thus influence vascular function (microvascular angiogenesis and vasopermeability) of ischemic tissue, alerting the body for special attention and the need for emergency repair procedures. Pharmacological modulation of circulating LMP concentrations could become a major future therapeutic target.
This work was supported by grants from the Hospital for Sick Children Foundation and Fight for Sight Foundation, a Canadian National Institute for the Blind E. A. Baker Research Grant, and Canadian Institutes of Health Research Grant MOP 85050. C. Yang, B. Mwaikambo, and S. Seshadri are recipients of awards from Centre Hospitalier Universitaire Ste-Justine Research Center, the Foundation Fighting Blindness-Canada, and Vision Research Network Fonds de la Recherche en Santé du Québec (FRSQ), respectively. P. Hardy is the recipient of a scholarship from FRSQ.
We thank Carmen Gagnon for invaluable technical skills.
↵* C. Yang and B. R. Mwaikambo contributed equally to this work.
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.
- Copyright © 2008 the American Physiological Society