Coactivation of Tie2 and Wnt signaling using an antibody-R-spondin fusion potentiates therapeutic angiogenesis and vessel stabilization in hindlimb ischemia

Abstract

Therapeutic angiogenesis by intentional formation of blood vessels is essential for treating various ischemic diseases, including limb ischemia. Because Wnt/β-catenin and angiopoietin-1/Tie2 signaling play important roles in endothelial survival and vascular stability, coactivation of these signaling pathways can potentially achieve therapeutic angiogenesis. In this study, we developed a bifunctional antibody fusion, consisting of a Tie2-agonistic antibody and the Furin domains of R-spondin 3 (RSPO3), to simultaneously activate Tie2 and Wnt/β-catenin signaling. We identified a Tie2-agonistic antibody T11 that cross-reacted with the extracellular domain of human and mouse Tie2, and evaluated its ability to increase endothelial cell survival and tube formation. We generated a bifunctional T11-RF12 by fusing T11 with the Furin-1 and -2 domains of RSPO3. T11-RF12 could bind not only to Tie2, but also to LGR5 and ZNRF3, which are counterparts of the Furin-1 and -2 domains. T11-RF12 significantly increased Wnt/β-catenin signaling, as well as the formation of capillary-like endothelial tubes, regardless of the presence of Wnt ligands. Coactivation of Tie2 and Wnt/β-catenin signaling by T11-RF12 increased the blood flow, and thereby reduced foot necrosis in a mouse hindlimb ischemia model. In particular, T11-RF12 induced therapeutic angiogenesis by promoting vessel stabilization through pericyte coverage and retaining endothelial expression of Frizzled 10 and active β-catenin. These results indicate that the agonistic synergism of Tie2 and Wnt/β-catenin signaling achieved using T11-RF12 is a novel therapeutic option with potential for treating limb ischemia and other ischemic diseases.

Keywords: Bifunctional antibody; R-spondin; Tie2; Wnt; hindlimb ischemia; therapeutic angiogenesis; vessel stabilization.

Figure 1.

Identification of a Tie2-agonistic antibody acting as an ang-1. (a) Cross-species binding of T11 IgG analyzed using biolayer interferometry. Recombinant hTie2-ecd or mTie2-ecd (100 nM) was immobilized on an AR2G biosensor and allowed to bind to T11 IgG (60 nM). Binding kinetics was analyzed using global fitting analysis of the binding curves. (b) Activation of Tie2 and cascade signaling by T11. Serum-deprived HUVECs were treated with various concentrations of T11 IgG as indicated, and cell lysates were immunoprecipitated using an anti-Tie2 antibody. Phosphorylation of Tie2 and AKT was evaluated using western blotting of immuno-precipitated samples and whole cell lysate (WCL), respectively. (c) Effect of T11 on the survival of endothelial cells. HUVECs were treated with various concentrations of T11 (0.5, 1, 5, 50, or 200 nM) or control IgG (200 nM) for 40 h. Cell viability was assessed using crystal violet staining. (d) Formation of capillary-like tubes of HUVECs induced by T11. HUVECs were seeded on matrigel-coated plates and incubated in the presence of control IgG (ctrl IgG; 100 nM), T11 IgG (100 nM), or recombinant ang-1 (400 ng/mL). After 18 h, tube-like structure in each well was photographed (×40). (e) Endothelial sprouting activity of T11. Mouse aortas were seeded on dishes precoated with matrigel and polymerized for 30 min. Ctrl IgG (100 nM), T11 IgG (100 nM), or ang-1 (400 ng/mL) was added twice on day 0 and 2, and aortic endothelial cell sprouting was imaged on day 4 (n = 6). NS: not significant; **p < 0.01.

Figure 2.

Generation of T11-based bifunctional antibodies fused with furin-1 and −2 (Fu1/2) domains of RSPO3. (a) Effect of T11 and ang-1 on Wnt/β-catenin signaling. 293T-hTie2 cells transfected with STF and Renilla luciferase reporter plasmids were plated and treated with control IgG (10 μg/mL), T11 IgG (10 μg/mL), ang-1 (300 ng/mL), recombinant Wnt3a (300 ng/mL) and/or recombinant RSPO3 (20 ng/mL) for 24 h. Relative luciferase values represent mean of triplicate measurements. (b) A schematic of bifunctional antibody–RSPO3 fusions. T11 was utilized as a backbone IgG, and the Fu1/2 domains of RSPO3 were appended on the C-terminus of HC or LC of T11, to generate T11–RF12 or T11–RF12-LC. (c Analysis of purified antibody fusions. T11–RF12 or T11–RF12-LC was purified and analyzed using SDS-PAGE under non-reducing and reducing conditions. (d) and (e) binding activity of antibody fusions. Binding of T11–RF12 and T11–RF12-LC to counterpart molecules was assessed using indirect ELISA with recombinant Tie2-ecd (D), LGR5, or ZNRF3 (E). One-hundred nanomolar of control IgG, T11, T11–RF12, or T11–RF12-LC was allowed to bind to each protein. Values represent mean ± SD of duplicate measurements.

Figure 3.

Enhanced activation of Wnt/β-catenin signaling and endothelial tube formation by T11–RF12. (a) and (b) activation of Wnt/β-catenin signaling by antibody fusions. 293T-hTie2 cells transfected with STF and Renilla luciferase reporter plasmids were plated and treated with 100 nM of control IgG, T11, T11–RF12, T11–RF12-LC, or recombinant RSPO3 (20 ng/mL) without (A) or with (B) recombinant Wnt3a (300 ng/mL). Relative luciferase values represent mean ± SD (n = 4). (c) and (d) induction of endothelial tube formation by T11–RF12. HUVECs were seeded on matrigel-coated plates and incubated in the presence of control IgG (ctrl IgG), T11 (100 nM), or T11–RF12 (100 nM), or RSPO3 (20 ng/mL) without (a) or with (b) recombinant Wnt3a (300 ng/mL). After 18 h, tube formation in each well was photographed (×40), and relative tube lengths or branch points were quantified by analyzing the images. Values represent mean ± SD of triplicate measurements. *p < 0.05; **p < 0.01; ***p < 0.001.

Figure 4.

T11–RF12 treatment improves blood perfusion and prevents foot necrosis in a mouse hindlimb ischemia model. (a) Study schedule. Mice were subjected to peripheral artery surgery involving arterial and venous ligation at day 0. Antibodies were injected on day 1, 3, 7, 14, and 21 post surgery, and laser Doppler perfusion imaging (LDPI) of mouse hindlimbs was performed on day 1, 7, 14, 21, and 28. Cohorts of mice were sacrificed on day 28, and their gastrocnemius muscle was extracted for immunohistochemical analysis. (b) Representative LDPI photographs of mouse hindlimbs. (c) Quantitative analysis of the blood perfusion recovery. The LDPI ratio was calculated as the ratio of ischemic to nonischemic hindlimb blood perfusion. Values indicate mean ± SD (n = 5 per group). *p < 0.05; **p < 0.01; ***p < 0.001. (d) Evaluation of foot necrosis. Representative foot images were photographed on day 28. The foot necrosis scores were statistically analyzed for each antibody-treated group. Results are presented as mean ± sd (n = 5). *p < 0.05; **p < 0.01.

Figure 5.

T11 and T11–RF12 induce therapeutic angiogenesis and vessel stabilization by activating Tie2 phosphorylation. (a) Immunofluorescence analysis for assessing Tie2 activation. Tissue sections were stained to detect phosphorylated Tie2 (P-Tie2) and an endothelial cell marker (CD31). Representative images for each group are shown. Scale bar = 300 μm. (b) and (c) quantification of endothelial P-Tie2 activity. Fluorescence intensities of P-Tie2⁺ (B) or CD31⁺ and P-Tie2⁺ areas (C) in acquired images were quantified using the ImageJ software. Relative values were calculated with respect to the value of the control IgG group (B). NS: not significant; **p < 0.01. (d) Immunofluorescence analysis for assessment of enhancement in angiogenesis and vascular stabilization. Tissue sections were stained with CD31 and αSMA (a marker for pericytes). Representative images for each group are shown. Scale bar = 200 μm. (e) Quantification of angiogenesis. Fluorescent intensities of CD31⁺ endothelial cells were measured using the ImageJ software, and relative values were compared between the groups. NS: not significant; **p < 0.01. (f) Evaluation of vessel-stabilization activity. CD31⁺ areas surrounded by αSMA⁺ areas were quantified to assess stabilized vessels with pericyte coverage. **p < 0.01; ***p < 0.001.

Figure 6.

T11–RF12 maintains the expression of Fzd10 in endothelial cells and increases the levels of active β-catenin. (a) Effect of T11–RF12 on Fzd10 expression. CD31 and Fzd10 were costained to assess Fzd10 expression in endothelial cells. Representative images for each group are shown. Scale bar = 300 μm. (b) Quantification of endothelial Fzd10 expression. Fluorescence intensity of Fzd10⁺ was quantified using the ImageJ software and number of CD31⁺ and Fzd10⁺ vessels was counted from each image. *p < 0.05; **p < 0.01; ***p < 0.001. (c) Activation of β-catenin by T11–RF12. CD31 and active form of β-catenin were costained to determine Wnt/β-catenin signaling activation in endothelial cells. Representative images for each group were shown. Scale bar = 300 μm. (d) Quantification of active β-catenin in endothelial cells. Fluorescence intensity of active β-catenin was quantified using the ImageJ software and number of CD31⁺ and β-catenin⁺ vessels was counted from each image. *p < 0.05; **p < 0.01; ***p < 0.001.