Galectin-3 is an important mediator of VEGF- and bFGF-mediated angiogenic response

Abstract

Recent studies have shown that a carbohydrate-binding protein, galectin-3, is a novel pro-angiogenic molecule. The mechanism by which galectin-3 promotes angiogenesis remains unknown. We demonstrate here that galectin-3 is a mediator of vascular endothelial growth factor (VEGF)- and basic fibroblast growth factor (bFGF)-mediated angiogenic response. Angiogenesis assays revealed that galectin-3 inhibitors, beta-lactose and dominant-negative galectin-3, reduce VEGF- and bFGF-mediated angiogenesis in vitro and that VEGF- and bFGF-mediated angiogenic response is reduced in galectin-3 knockdown cells and Gal3(-/-) animals. Integrin alphavbeta3 was identified as the major galectin-3-binding protein and anti-alphav, -beta3, and -alphavbeta3 integrin function-blocking antibodies significantly inhibited the galectin-3-induced angiogenesis. Furthermore, galectin-3 promoted the clustering of integrin alphavbeta3 and activated focal adhesion kinase. Knockdown of GnTV, an enzyme that synthesizes high-affinity glycan ligands for galectin-3, substantially reduced: (a) complex N-glycans on alphavbeta3 integrins and (b) VEGF- and bFGF-mediated angiogenesis. Collectively, these data suggest that galectin-3 modulates VEGF- and bFGF-mediated angiogenesis by binding via its carbohydrate recognition domain, to the GnTV synthesized N-glycans of integrin alphavbeta3, and subsequently activating the signaling pathways that promote the growth of new blood vessels. These findings have broad implications for developing novel, carbohydrate-based therapeutic agents for inhibition of angiogenesis.

Figure 1.

Galectin-3 promotes angiogenesis in vivo in a dose-dependent manner. (A) Angiogenesis in vivo was evaluated using the mouse corneal micropocket assay. Sustained-release polymer pellets containing various doses of galectin-3 (20–160 ng/pellet) were implanted in the corneas of Gal3^+/+ animals. (i) 5 d after surgery, the vessel area was calculated as described in the text. Data are expressed as mean ± SEM (n = 4 or more). *, P < 0.01 compared with control pellets (Con) without any protein. (ii) Representative fluorescence images of corneas are shown in A (i). White asterisks indicate pellets. (B–D) Full-length galectin-3 but not truncated galectin-3 (Gal3C) promotes angiogenesis in vitro and in vivo. (B) Modified Boyden Chamber Assay. The cells that migrated from the upper chamber to the lower chamber during the period of 6 h in response to the protein gradients created by full-length galectin-3 and/or Gal3C were counted in three nonoverlapping fields under a microscope (320×). Data are expressed as mean ± SEM (n = 3/group). *, P < 0.01 compared with cells incubated with media alone (Con). (C) Capillary tubule formation assay. HUVECs premixed with galectin-3 and/or Gal3C were plated on solidified matrigel (4.5 mg/ml) and after 6-h incubation, the extent of capillary tubule formation was evaluated by counting the number of branch points. Data are expressed as mean ± SEM (n = 3 or more/group). *, P < 0.01 compared with cells incubated with media alone (Con). (D) Angiogenesis in vivo was evaluated using the mouse corneal micropocket assay using pellets containing full-length (Gal3) and truncated galectin-3 (Gal3C; 80 ng/pellet). Data are expressed as mean ± SEM (n = 4 or more). *, P < 0.01 compared with control pellets (Con). Representative fluorescence images of corneas are shown in D (ii). White asterisks indicate pellets. Bars, 200 µm. Data are representative of three or more independent experiments.

Figure 2.

Galectin-3 inhibitors block VEGF- and bFGF-mediated angiogenesis in vitro. (A) Modified Boyden chamber assay. The effect of galectin-3 inhibitors on VEGF- and bFGF-induced cell migration was evaluated. The cells that migrated from the upper chamber to the lower chamber were counted as described in Fig. 1. (B) Capillary tubule formation assay. HUVEC suspensions premixed with cytokines were plated on solidified matrigel (4.5 mg/ml), and, after 6-h incubation, the extent of capillary tubule formation was evaluated by counting the number of branch points. Representative images from each group are shown in B (ii). Images of cells incubated in media alone (not depicted) were indistinguishable from those incubated in media containing VEGF or bFGF in the presence of β-lactose (Lac). The assays were performed in the absence (Control) and presence of truncated galectin-3 (Gal3C, a dominant-negative inhibitor of galectin-3), full-length galectin-3 (Gal3), 0.1 M β-lactose (Lac, a competing saccharide), and 0.1 M sucrose (Suc, a noncompeting control saccharide). Data are expressed as mean ± SEM (n = 3 or more/group), *, P < 0.01; **, P < 0.05 compared with respective controls incubated without galectin-3 inhibitors (Control) or with a noncompeting saccharide (Suc). Bar, 100 µm. Data are representative of three or more independent experiments.

Figure 3.

Galectin-3 knockdown reduces VEGF- and bFGF-mediated angiogenesis in vitro. (A and B) HUVECs were transfected with siRNA duplexes directed against galectin-3 (Gal3 siRNA) or control nontargeting siRNA duplexes (Control siRNA). The knockdown of galectin-3 was evaluated by RT-qPCR (A) and Western blot analysis (Bi: a representative image, Bii: quantification of protein knockdown by ImageJ). (C and D) VEGF and bFGF induced cell migration and capillary tubule formation is reduced in galectin-3–knockdown cells. The ability of 25 ng/ml VEGF (open bars) and 10 ng/ml bFGF (solid bars) to promote chemotaxis (C) and capillary tubule formation (D) in galectin-3 siRNA (Gal3 siRNA) transfected and nontargeting siRNA-transfected (control siRNA) HUVECs, as well as galectin-3 siRNA transfected HUVECs in the presence of 10 µg/ml exogenous galectin-3 (Rescue) was compared. A value of 1.0 was assigned to cell migration and capillary tubule formation values of control cells. The values of all other cells were calculated as a fold change with respect to control cells. Data are expressed as mean ± SEM (n = 3/group). *, P < 0.01 and **, P < 0.05 as compared with nontargeting siRNA transfected HUVECs). Representative images of the cells that migrated from the upper chamber to the lower chamber and capillary tubule formation in response to VEGF and bFGF stimulation are shown in C (ii) and D (ii), respectively. Bars: (C, ii) 100 µm; (D; ii) 200 µm. Data are representative of two or more independent experiments.

Figure 4.

VEGF- and bFGF-induced angiogenesis is reduced in Gal3^−/− mice. Angiogenesis in vivo was evaluated using the mouse corneal micropocket assay as described in the text using VEGF and bFGF pellets. 5 d after surgery, the animals were perfused with FITC-BS1, and the extent of angiogenesis was evaluated by examining the flat mounts of corneas by fluorescence microscopy. Blood vessel area was calculated using ImageJ. (A) Vessel area of neovascularization expressed in pixel² × 10⁴. Data are expressed as mean ± SEM (n = >4/group). *, P < 0.05 compared with Gal3^+/+ control animals. (B) Representative fluorescence images of corneas. White asterisks indicate pellets. Bar, 200 µm. Data are representative of two independent experiments.

Figure 5.

Integrin αvβ3 serves as a galectin-3 counter-receptor. (A) HUVEC lysates were applied to a galectin-3 affinity chromatography column and bound proteins were sequentially eluted with a noncompeting saccharide, 0.1 M sucrose (Suc), and a competing saccharide, 0.1 β-lactose (Lac). Proteins eluted from the affinity column were identified by Western blot analysis. Both αv and β3 integrins were detected in the lactose eluate, but not in the sucrose eluate. (B and C) Integrin-blocking antibodies directed against integrins αv, β3, and αvβ3 inhibit galectin-3–induced cell migration (B) and capillary tubule formation (C). The ability of 10 µg/ml galectin-3 to promote chemotaxis (Boyden chamber assay) and capillary tubule formation in the presence and absence of 10 µg/ml αv, β3, and αvβ3 integrin-blocking antibodies or control mouse IgG was compared. Data are expressed as mean fold change over control cells incubated in media alone ± SEM (n = 3/group). **, P < 0.05; *, P < 0.01 as compared with cell migration and capillary tubule formation induced by galectin-3. (D) Galectin-3 promotes clustering of integrin αvβ3. HUVECs were serum starved overnight and treated with 10 µg/ml galectin-3 for 10 min. At the end of the incubation period, the distribution of integrin αvβ3 was evaluated by confocal microscopy after immunostaining with an anti–human integrin αvβ3 monoclonal antibody. Treatment with galectin-3 resulted in a marked redistribution of integrins into a punctate pattern, indicative of integrin clustering. Overlapping Normanski images are shown on the bottom (NM). Bar, 50 µm. (E) Galectin-3 mediates FAK activation. HUVECs were serum starved for 2 h and treated with 10 µg/ml of galectin-3 for various time periods (0, 0.5, 1, 2, 5, and 10 min). HUVEC lysates were then subjected to immunoblot analysis for pFAK (Y397; top) and FAK (bottom). Galectin-3 treatment promoted a time-dependent activation of FAK that started at 1 min and peaked at 5 min. Total FAK levels remained unchanged at all time periods. Data are representative of three or more independent experiments.

Figure 6.

Disruption of GnTV reduces binding of galectin-3 to cell surface glycoprotein receptors with a concomitant reduction in galectin-3 and growth factor-mediated angiogenesis in vitro and in vivo. (A) HUVECs were transfected with lentiviral particles expressing shRNA directed against GnTV (shGnTV) or nontargeting shRNA control (shControl). The knockdown of endogenous GnTV was evaluated at the mRNA level by RT-qPCR. The value for HUVECs transfected with GnTV and control shRNA is expressed as the change in the expression level with respect to untransfected cells which served as a calibrator. *, P < 0.01 compared with control cells. (B) Expression of GnTV product, core β 1,6-branched saccharides, in HUVECs transfected with lentiviral particles was evaluated by staining with rhodamine-conjugated l-PHA, which specifically reacts with core β 1,6-branched saccharides. Bar, 100 µm. (C) The cell surface proteins of HUVECs transfected with lentiviral particles expressing shRNA directed against GnTV (shGnTV) or nontargeting shRNA control (shControl) were biotinylated and cross-linked in a single step, the biotinylated proteins were isolated by streptavidin pull-down, electrophoresed in reducing conditions and then probed with anti–galectin-3 mAb. Regardless of whether the Western blot analysis was performed using the streptavidin pull-down proteins (PD:Strep) or whole-cell lysates (WCL) of the labeled cells, reduced level of galecin-3 was detected in the GnTV knockdown cells compared with the cells transfected with control, nontargeting lentiviral particles. (D) GnTV modifies integrins αv and β3. RIPA buffer cell lysates (WCL) of untransfected HUVECs (H) and HUVECs transfected with control (C) and GnTV (M) targeting lentiviral particles were incubated with either l-PHA-agarose or ConA-agarose (1 h at 4°C). Bound proteins were eluted from the agarose beads by boiling in SDS-PAGE sample buffer, separated on SDS-PAGE, and immunostained to detect integrins αv and β3. Note that GnTV knockdown cell lysates show markedly reduced binding of integrins αv and β3 to l-PHA indicating that integrins αv and β3 are modified by GnTV. (E and F) The ability of galectin-3 (10 µg/ml), VEGF (25 ng/ml), and bFGF (10 ng/ml) to promote cell migration (E, Boyden chamber assay) and capillary tubule formation (F) is substantially reduced in GnTV knockdown cells. Data are expressed as mean fold change over control cells incubated in media alone ± SEM (n = 3/group), *, P < 0.01; **, P < 0.05 compared with untransfected controls. (G) VEGF- and bFGF-induced angiogenesis is reduced in GnTV^−/− mice. Angiogenesis in vivo was evaluated by the mouse corneal micropocket assay using VEGF and bFGF pellets. On day 5 after surgery, the animals were perfused with FITC-BS1, and the extent of angiogenesis was evaluated by examining the flat mounts of corneas by fluorescence microscopy. Blood vessel area was calculated using ImageJ. (i) Vessel area of neovascularization expressed in pixel² x 10⁴. Data are expressed as mean ± SEM (n = at least 4/group). *, P < 0.05 compared with GnTV^+/+ control animals. (ii) Representative fluorescence images of corneas. White asterisks indicate pellets. Bar, 100 µm. Control pellets, which did not contain any protein, did not promote angiogenesis [not depicted; images identical to that shown in Fig. 1 A (ii)]. Data are representative of two independent experiments each repeated in triplicate.