Eph-B4 mediates vein graft adaptation by regulation of endothelial nitric oxide synthase

Abstract

Objective: Vein graft adaptation is characterized by loss of expression of the tyrosine kinase receptor Eph-B4, the embryonic determinant of venous identity, without increased expression of its ligand ephrin-B2, the embryonic determinant of arterial identity. Endothelial nitric oxide synthase (eNOS) is an important mediator of vessel remodeling. We hypothesized that the mechanism of action of Eph-B4 during vein graft adaptation might be through regulation of downstream eNOS activity.

Methods: Mouse lung endothelial cells were stimulated with ephrin-B2/Fc, without and with preclustering, without and with the eNOS inhibitor N_ω-nitro-l-arginine methyl ester hydrochloride or the Eph-B4 inhibitor NVP-BHG712, and assessed by Western blot and immunofluorescence for eNOS and Eph-B4 phosphorylation. Nitric oxide (NO) production was assessed using an NO-specific chemiluminescence analyzer. Cell migration was assessed using a Transwell assay. Human and mouse vein graft specimens were examined for eNOS activity by Western blot, and vessel remodeling was assessed in vein grafts in wild-type or eNOS knockout mice.

Results: Ephrin-B2/Fc stimulated both Eph-B4 and eNOS phosphorylation in a bimodal temporal distribution (n = 4; P < .05), with preclustered ephrin-B2/Fc causing prolonged peak Eph-B4 and eNOS phosphorylation as well as altered subcellular localization (n = 4; P < .05). Ephrin-B2/Fc increased NO release (n = 3; P < .01) as well as increased endothelial cell migration (n = 6; P < .05) in an eNOS-dependent fashion. Both human and mouse vein grafts showed increased eNOS phosphorylation compared with normal veins (n = 3; P < .05). Vein grafts from eNOS knockout mice showed less dilation and less wall thickening compared with wild-type vein grafts (n = 7; P < .05).

Conclusions: eNOS is a mediator of vein graft adaptation to the arterial environment. Eph-B4 stimulates eNOS phosphorylation in vitro and may mediate vein graft adaptation by regulation of eNOS activity in vivo.

Figure 1

Ephrin-B2/Fc stimulates Eph-B4 and eNOS phosphorylation. A–B. Bar graphs show time course of Eph-B4 phosphorylation (A) and eNOS phosphorylation (B) following Ephrin-B2/Fc (2 μg/ml). IP, immunoprecipitation; IB: immunoblot. A, n=4; *, P=.0041; **, P=.0097; ***, P=.0029. B, n=5; *, P=.0161; **, P=.0366. C. Bar graphs show eNOS phosphorylation in response to EphrinB2/Fc (2 μg/ml, 15 min), without or with NVP-BHG712 (1 μmol, 1 hour). n=3,*, P=.0313; **, P=.0018. D–E. Bar graph show time course of Eph-B4 phosphorylation (D) and eNOS phosphorylation (E) following preclustered Ephrin-B2/Fc (2 μg/ml, Ephrin-B2/Fc: Anti-Fc ratio, 1:5) stimulation. D, n=3; *, P=.0496; **, P=.0146; ***, P=.0032. E, n=4; *, p=.0484. (F) Schematic of the Eph-B4-eNOS pathway in vitro. In isolated endothelial cells, treatment of Eph-B4 with Ephrin-B2/Fc stimulates Eph-B4 and eNOS phosphorylation; these effects are enhanced (red arrows) in the presence of the clustered multimeric Ephrin-B2/Fc.

Figure 2

Colocalization of Eph-B4 and peNOS in EC. Representative immunofluorescence showing Eph-B4 (green), peNOS (red) and merge (yellow) in EC following control (top row), Ephrin-B2/Fc (2 μg/ml, middle row), or clustered Ephrin-B2/Fc (2 μg/ml, Ephrin-B2/Fc: Anti-Fc ratio, 1:5, bottom row) stimulation for 5 min (left columns), 15 min (middle columns) or 6 hours (right columns). Blue color is DAPI. Yellow arrows show merge of Eph-B4 and peNOS, yellow arrowheads show Eph-B4 in the cytoplasm. Scale bar, 20 μm.

Figure 3

Eph-B4 functions in endothelial cells are mediated by eNOS. A. Bar graph showing mean NO release in response to Ephrin-B2/Fc (2 μg/ml, 60 min), without or with L-NAME pretreatment (1 mM, 30 min). n=3; *, P=.0018; **, P=.0013; ***, P<.0001. B. Bar graph showing mean NO release in response to Ephrin-B2/Fc (2 μg/ml, 30 min) in WT or eNOS KO EC. n=4; *, P=.0076; **, P<.0001; ***, P<.0001. C. Bar graph showing EC migration in response to Ephrin-B2/Fc (2 μg/ml, 8 hr) or 10% fetal bovine serum (FBS), without or with L-NAME pretreatment (1 mmol, 30 min). n=6; *, P=0.0376; **, P<.0001. D. Bar graph showing EC migration in response to Ephrin-B2/Fc (2 μg/ml, 8 hr) or 10% FBS in WT or eNOS KO EC. n=5; *, p<.0001; **, p=.0217; ***, p<.0001; ****, p<.0001.

Figure 4

Increased eNOS phosphorylation in human and mouse vein grafts. A. Western blot showing eNOS phosphorylation in human saphenous veins and vein grafts. Bar graph shows the ratio of densitometry values of phosphorylated to total eNOS in veins and vein grafts. n=3; *, P=.0138. B. Western blot showing eNOS phosphorylation in mouse inferior vena cava (IVC) and vein grafts, 3 weeks after implantation. Bar graph shows the ratio of densitometry values of phosphorylated to total eNOS in mouse IVC and vein grafts. n=2; *, P=.0198. C. Representative immunofluorescence showing eNOS phosphorylation in human saphenous vein and vein graft (first row), and mouse IVC and vein graft (second row). Yellow arrows show peNOS positive cells. Scale bar: 25 μm.

Figure 5

eNOS mediates venous remodeling during mouse vein graft adaptation. A. Line graph showing time course of venous wall thickness of wild type (WT) and eNOS KO mice, measured using Doppler ultrasound. Solid line represents WT mice, dashed line represents eNOS KO mice. n=5; *, P<.0001. B. Representative photomicrographs showing vein grafts in WT mice (left column) and eNOS KO mice (right column), stained with H&E. Black arrowheads show the venous wall thickness. Scale bar: low power 20μm, high power 10μm. C-E. Bar graphs showing (C) mean wall thickness, (D) total vein area, and (E) lumen area in WT and eNOS KO mice. C, n=7; *, P=.0462. D, n=7; *, P=.028. E, n=7; P=.3273.

Figure 6

Increased Eph-B4 expression in eNOS knockout vein grafts. A. Representative immunofluorescence showing eNOS (red, left column), Eph-B4 (green, second column), p-Tyr (red, third column) and merge of Eph-B4 and p-Tyr (yellow, fourth column) in WT (top row) and eNOS KO (bottom row) vein grafts. n=3; scale bar, 100μm. B. Bar graph showing mean densitometry of eNOS, Eph-B4 and merge of Eph-B4 and p-Tyr. n=3; *, P=.0017; **, P=.0011; ***, P=.0002. C. Schematic of the Eph-B4-eNOS pathway during vein graft adaptation in vivo. During vein graft adaptation, Eph-B4 expression is decreased, allowing eNOS activity to promote adaptive venous remodeling. eNOS negatively regulates Eph-B4 activity in a feedback loop; when eNOS is not present, venous remodeling is reduced and Eph-B4 activity is enhanced.