Short-term preoperative protein restriction attenuates vein graft disease via induction of cystathionine γ-lyase

Abstract

Aims: Therapies to prevent vein graft disease, a major problem in cardiovascular and lower extremity bypass surgeries, are currently lacking. Short-term preoperative protein restriction holds promise as an effective preconditioning method against surgical stress in rodent models, but whether it can improve vein graft patency after bypass surgery is undetermined. Here, we hypothesized that short-term protein restriction would limit vein graft disease via up-regulation of cystathionine γ-lyase and increased endogenous production of the cytoprotective gaseous signalling molecule hydrogen sulfide.

Methods and results: Low-density lipoprotein receptor knockout mice were preconditioned for 1 week on a high-fat high-cholesterol (HFHC) diet with or without protein prior to left common carotid interposition vein graft surgery with caval veins from donor mice on corresponding diets. Both groups were returned to a complete HFHC diet post-operatively, and vein grafts analysed 4 or 28 days later. A novel global transgenic cystathionine γ-lyase overexpressing mouse model was also employed to study effects of genetic overexpression on graft patency. Protein restriction decreased vein graft intimal/media+adventitia area and thickness ratios and intimal smooth muscle cell infiltration 28 days post-operatively, and neutrophil transmigration 4 days post-operatively. Protein restriction increased cystathionine γ-lyase protein expression in aortic and caval vein endothelial cells (ECs) and frequency of lung EC producing hydrogen sulfide. The cystathionine γ-lyase inhibitor propargylglycine abrogated protein restriction-mediated protection from graft failure and the increase in hydrogen sulfide-producing ECs, while cystathionine γ-lyase transgenic mice displayed increased hydrogen sulfide production capacity and were protected from vein graft disease independent of diet.

Conclusion: One week of protein restriction attenuates vein graft disease via increased cystathionine γ-lyase expression and hydrogen sulfide production, and decreased early inflammation. Dietary or pharmacological interventions to increase cystathionine γ-lyase or hydrogen sulfide may thus serve as new and practical strategies to improve vein graft durability.

Keywords: Cardiovascular surgery; Diet and nutrition; Vascular disease.

Published on behalf of the European Society of Cardiology. All rights reserved. © The Author(s) 2019. For permissions, please email: journals.permissions@oup.com.

Figures

Figure 1

Short-term protein restriction attenuates vein graft disease. (A) Schematic representation of the development of intimal hyperplasia 4 weeks after implantation of a vein graft (inferior vein cava, IVC) from a separate donor mouse. (B) Schematic representation of diets (left) and dietary alterations (right) during the 3 week run-in, 1 week preconditioning, and 4 week post-operative periods. All mice were subjected to a 3-week run-in period on a HFHC diet. (C) Representative images of Masson trichome-stained vein grafts 28 days after surgery. Boundary between intimal (I) and medial + adventitial (M+A) layers is traced in yellow. Scale bars = 0.5 mm or 200 µm as indicated. (D–G) Vein grafts assessments 28 days after surgery in HFHC vs. PR-HFHC preconditioned mice as indicated; n = 15/group. (D) I/M+A area ratios (2.1 ± 0.2 vs. 0.9 ± 0.1, P < 0.0001, Student’s t-test). (E) I/M+A thickness ratios (2.7 ± 0.2 vs. 1.2 ± 0.1, P < 0.0001, Student’s t-test). (F) Intimal area (0.422 mm2 ± 0.03 vs. 0.320 mm2 ± 0.029, P = 0.0294, Mann–Whitney test) and M+A area (0.22 mm2 ± 0.021 vs. 0.367 mm2 ± 0.021, P < 0.001, Student’s t-test). (G) Intimal thickness (0.156 mm ± 0.009 vs. 0.131 mm ± 0.015) and M+A thickness (0.064 mm ± 0.005 vs. 0.112 mm ± 0.006, P < 0.0001, Student’s t-test). All data expressed as mean ± SEM; *P < 0.05, ****P < 0.0001.

Figure 2

Preoperative protein restriction limits smooth muscle cell migration and inhibits leucocyte transmigration. (A–I) Assessment of intimal hyperplasia and neutrophil infiltration in vein grafts 28 days (A–D) or 4 days (E–I) after surgery in LDLr−\− mice preconditioned on the indicated diets; n = 5–10/group. (A) Representative images of SMC-α (green), Ki-67 (red), and DAPI (blue) stained vein grafts. Included are lumen, intima and media + adventitia as histological landmarks and the intima/M+A border is illustrated by yellow lining. (B) SMC-α positive cells in intimal layer (12.7±2.1 vs. 6.9±1.6; P = 0.0446, Student’s t-test) or M+A layer expressed as a percentage of area occupied. (C) Number of proliferating SMC-α and Ki-67 double positive cells in the intimal layer per mm2 (Student’s t-test). (D) Percentage of intimal and M+A area occupied by collagen. (E) Representative images of SMC-α (green), Ki-67 (red), and DAPI (blue) stained vein grafts from LDLr−\− mice preconditioned as indicated and analysed on post-operative Day 4. Included are lumen, intima and M+A as histological landmarks and the Intima/M+A border is illustrated by a yellow line. (F) SMC-α positive cells in intimal or M+A layer expressed as a percentage of area occupied. (G) Number of proliferating SMC-α and Ki-67 double positive cells in the indicated layer per mm2. (H) Representative images of neutrophil anti-elastase (brown)-stained vein grafts from LDLr−\− mice preconditioned as indicated and analysed on post-operative Day 4, with the intima/M+A border depicted by a yellow line. (I) Quantitation of neutrophil transmigration in intimal layer (805 ± 160 vs. 322 ± 131 cells/mm2, P = 0.0184, Mann–Whitney test), perivascular adipose tissue (PVAT) layer (423±73 vs. 220±50 cells/mm2, P = 0.0077, Student’s t-test), and total vein graft (447 ± 45 vs. 233 ± 61 cells/mm2  P = 0.0146, Student’s t-test). Scale bars = 100 µm as indicated. All data expressed as mean ± SEM; *P < 0.05, **P < 0.01.

Figure 3

Contribution of donor and recipient response to protein restriction in attenuation of vein graft disease. (A–E) Vein grafts assessments 28 days after surgery in PR-HFHC preconditioned donor only or recipient only vs. neither (HFHC control) or both (PR-HFHC control) from Figure 1. (A) Intimal(I)/Media+Adventitia (M+A) area ratios from donors, recipients or both preconditioned on the indicted diet. HFHC (both) and PR-HFHC (both) data are from Figure 1B; PR-HFHC (donor only; 2.1 ± 0.2); PR-HFHC (recipient-only; 1.3 ± 0.2); one-way ANOVA with Dunnett’s multiple comparisons test vs. HFHC (both). (B and C) Intimal (B) and M+A (C) areas from the indicated diet groups; Kruskal–Wallis test with Dunnett’s multiple comparisons test vs. HFHC (both). (D and E) Percent of intimal (D) or M+A (E) area occupied by collagen; one-way ANOVA. (F) Vein graft lumen diameter proximal, mid, or distal to the heart 4 weeks after graft implantation as determined by in vivo duplex ultrasound; two-way ANOVA with Turkey’s multiple comparisons test. All data expressed as mean ± SEM; **P < 0.01, ****P < 0.0001.

Figure 4

Protein restriction up-regulates cystathionine γ-lyase and H2S levels in endothelial cells. (A and B) Representative images of inferior vena cava (IVC) immunofluorescence staining of CBS (A) or CGL (B). (C–H) Western blots (C, E, G) and quantitation (D, F, H) of CGL and CBS from whole thoracic aorta (C and D), aortic endothelial cells (ECs) (E and F), or IVC ECs (G and H) isolated from LDLr−\− mice after 1 week on the indicated diet; Ponceau stained membranes were used as loading controls and CGL−/− ECs were used as a control for CGL antibody specificity (E). (I–M) LDLr−\− mice in the indicated treatment groups (HFHC, PR-HFHC, or PR-HFHC + PAG) were harvested after one week. (I and J) Whole thoracic aorta gene expression of CGL (I) and ATF4 (J); one-way ANOVA with Dunnet’s multiple comparisons test vs. HFHC control group. (K–M) Flow cytometric analysis of single cell isolates from lung after staining with CD31 and P3 (fluorescent H2S probe). (K) Representative dot plots from the indicated groups with CD31/P3 double positive cells in the box. (L and M) Fold change, relative to HFHC group, in mean fluorescent intensity of P3 (L) and frequency (M) of CD31/P3 double positive cells; HFHC and PR-HFHC, n = 10/group; PR-HFHC+PAG, n = 4/group; Kurskal–Wallis test with Dunnett’s multiple comparisons test. All data expressed as mean ± SEM; *P < 0.05, **P < 0.01.

Figure 5

Cystathionine γ-lyase is required for protein restriction mediated attenuation of intimal hyperplasia. (A–D) Vein grafts assessments 28 days after surgery in PR-HFHC preconditioned LDLr−/− mice injected with either vehicle or PAG during the one week preconditioning period prior to vein grafting; n = 5–6/group. (A) Representative images of Masson trichome-stained vein grafts; boundary between intimal (I) and medial + adventitial (M+A) layers is traced in yellow. Scale bars = 0.5 mm and 200 µm as indicated. (B and C) I/M+A area (B, 0.7 ± 0.2 vs. 1.6 ± 0.2, P = 0.014, Student’s t-test) and thickness (C, 0.9 ± 0.2 vs. 2.0 ± 0.2, P = 0.014, Student’s t-test) ratios. (D and E) Intimal and M+A area (D) and thickness (E); Student’s t-test between ± PAG treatment groups within layer. All data expressed as mean ± SEM; *P < 0.05.

Figure 6

CGL overexpression protects against vein graft disease but does not increase basal endothelial cell H2S levels. (A–D) Western blot of CGL (A and B) and hydrogen sulfide production capacity (C and D) in homogenates of kidney (A, C) and liver (B, D) from hemizygous CGL transgenic (CGLtg) and WT littermate mice as indicated. (E–N) Vein grafts assessments in WT vs. CGLtg mice (n = 5–8/group) 28 days (E–L) or 4 days (M–N) after surgery. (E) Representative images of Masson’s trichome-stained vein grafts; boundary between intimal (I) and medial + adventitial (M+A) layers is traced in yellow. Scale bars = 0.5 mm and 200 µm as indicated. (F) I/M+A area ratios (1.5 ± 0.2 vs. 0.8 ± 0.1, P = 0.0066, Student’s t-test). (G) I/M+A thickness ratios (2.0 ± 0.4 vs. 1.1 ± 0.2, P = 0.0159, Student’s t-test). (H and I) Intimal and M+A area (H) and intimal and M + A thickness (I) (0.088 ± 0.007 vs. 0.114 ± 0.017, P = 0.0295, Mann–Whitney test). (J) Representative images of vein grafts stained with SMC-α (brown); scale bar = 100µm. (K) SMC-α positive cells in intimal layer (61 vs. 38, P = 0.032, Student’s t-test) or M+A layer expressed as a percentage of area occupied. (L) Vein graft lumen diameter proximal, mid or distal to the heart as indicated (two-way ANOVA with Turkey’s multiple comparisons test). (M) Representative images of grafts stained with anti-neutrophil-elastase on post-operative Day 4; scale bars = 50 µm. (N) Quantitation of neutrophil transmigration in intimal layer (Student’s t-test), adventitial + perivascular adipose tissue (PVAT) layer and total vein graft (Mann–Whitney test). (O–Q) Endogenous H2S in lung endothelial cells of WT or CGLtg littermates (n = 5/group). (O) Representative dot plots with CD31/P3 double positive cells indicated within the box. (P, Q) Fold change, relative to WT group, in mean fluorescent intensity of P3 (P) and frequency (Q) of CD31/P3 double positive cells; WT and CGLtg  n = 8–10/group; Mann–Whitney test. All data expressed as mean ± SEM; *P < 0.05, **P < 0.01.

Figure 7

Model for attenuation of vein graft disease by short-term protein restriction via up-regulation of cystathionine γ-lyase.