Increased shear stress with upregulation of VEGF-A and its receptors and MMP-2, MMP-9, and TIMP-1 in venous stenosis of hemodialysis grafts

Abstract

Venous injury and subsequent venous stenosis formation are responsible for hemodialysis graft failure. Our hypothesis is that these pathological changes are in part related to changes in wall shear stress (WSS) that results in the activation of matrix regulatory proteins causing subsequent venous stenosis formation. In the present study, we examined the serial changes in WSS, blood flow, and luminal vessel area that occur subsequent to the placement of a hemodialysis graft in a porcine model of chronic renal insufficiency. We then determined the corresponding histological, morphometric, and kinetic changes of several matrix regulatory proteins including VEGF-A, its receptors, matrix metalloproteinase (MMP)-2, MMP-9, tissue inhibitor of matrix metalloproteinase (TIMP)-1, and TIMP-2. WSS was estimated by obtaining blood flow and luminal vessel area by performing phase-contrast MRI with magnetic resonance angiography in 21 animals at 1 day after graft placement and prior to death on day 3 (n = 7), day 7 (n = 7), and day 14 (n = 7). At all time points, the mean WSS at the vein-to-graft anastomosis was significantly higher than that at the control vein (P < 0.05). WSS had a bimodal distribution with peaks on days 1 and 7 followed by a significant reduction in WSS by day 14 (P < 0.05 compared with day 7) and a decrease in luminal vessel area compared with control vessels. By day 3, there was a significant increase in VEGF-A and pro-MMP-9 followed by, on day 7, increased pro-MMP-2, active MMP-2, and VEGF receptor (VEGFR)-2 (P < 0.05) and, by day 14, increased VEGFR-1 and TIMP-1 (P < 0.05) at the vein-to-graft anastomosis compared with control vessels. Over time, the neointima thickened and was composed primarily of alpha-smooth muscle actin-positive cells with increased cellular proliferation. Our data suggest that hemodialysis graft placement leads to early increases in WSS, VEGF-A, and pro-MMP-9 followed by subsequent increases in pro-MMP-2, active MMP-2, VEGFR-1, VEGFR-2, and TIMP-1, which may contribute to the development of venous stenosis.

Fig. 1

Placement of the polytetrafluoroethylene (PTFE) hemodialysis graft and representative MRI and phase-contrast magnetic resonance angiography (PC MRA) of the vein-to-graft anastomosis. A: placement of the PTFE hemodialysis graft. CA, control carotid artery; CV, control jugular vein; VS, venous stenosis, IA, inflow artery. B: MRI with PC MRA performed on day 14 that shows high grade VS formation. RSA, right subclavian artery; LSA, left subclavian artery. C: PC MRA showing the direction of blood vessels in the opposite direction as depicted by black in the VS and white in the IA. D: magnitude of the blood flow in the same vessels. E: a hematoma surrounding the PTFE graft. T, trachea.

Fig. 2

Morphological changes in the venous wall and the development of neointimal hyperplasia. Animals were killed at the three various time points, and representative sections from a day 14 animal are shown. The explanted vein-to-graft anastomosis and contralateral jugular vein were stained using hematoxyln and eosin (H&E), Verhoeff's van Gieson (VVG) stain, α-smooth muscle (SM) actin, PCNA, and Masson's trichrome. The results show that neointimal hyperplasia had developed by day 14 compared with the control vein. An analysis of the expression of α-SM actin at the vein-to-graft anastomosis and normal vein was performed. The control external jugular vein showed well-organized layers of SM cells expressing α-SM actin appearing red-brown. On day 14 after the graft placement, the neointima and media at the venous anastomosis stained more intensely than the control for α-SM actin. Cells positive for PCNA are brown. There was an increased number of cells staining brown in the neointima of the venous anastomosis by day 14 compared with the control vein. The control external jugular vein showed the most extracellular matrix (appearing blue by trichrome staining) in the media and adventitia. By day 14, more cells (appearing pink) accumulated in the media and neointima accompanied by prominent extracellular matrix deposition in the media and inner neointima.

Fig. 3

Thickness of the nointima, media, and adventitia at the vein-to-graft anastomosis (VS) and control veins (CV) over time. Specimens from explanted vein-to-graft anastomosis and control vessels were obtained from pigs killed on days 3, 7, and 14. The thickness of the neointima, media, and adventitia from the vein-to-graft anastomosis and control vessels were individually quantified using computer-assisted planimetry on VVG-stained slides and was pooled for the vein-to-graft anastomosis and control veins on days 3, 7, and 14. The average thickness of the intima from the vein-to-graft anastomosis increased over time compared with control vessels. By days 7 and 14, it was significantly increased compared with control vessels (P < 0.05). The mean thickness of the media and adventitial of the vein-to-graft anastomosis was higher than that of the control vein at all time points. By days 3 and 7, the average thickness of the media was significantly higher than the control veins (P < 0.05). At all time points, the average thickness of the adventitia of the vein-to-graft anastomosis was significantly higher than the control vein (P < 0.05). Data are means ± SD. *P < 0.05.

Fig. 4

MRI measurements of the vein-to-graft anastomosis on days 1, 3, 7, and 14. The wall shear stress (WSS) was calculated from the MRI data. At all time points, the mean WSS of the VS was significantly higher (P < 0.05) than the control vein. By day 7, the WSS was significantly higher than on days 3 and 14 (P < 0.05). Data are means ± SD. *P < 0.05

Fig. 5

A: graph representing the appropriate protein band of VEGF-A from Western blot analysis. B: graph representing the appropriate band for β-actin from Western blot analysis for protein loading. C: semiquantitative analysis for VEGF-A performed on days 3, 7, and 14. The normalized density of VEGF-A was significantly higher at the vein-to-graft anastomosis compared with the control vein by day 3. *Significantly higher value (P < 0.05). Data are means ± SD. D: immunohistochemistry for VEGF-A for localization of expression on day 3. By immunohistochemistry, brown staining cells are positive for VEGF-A. Representative sections from a day 3 animal are shown. The left and middle left columns are from the vein-to-graft anastomosis, and middle right and right columns are from the control vein. The left and middle right columns are ×10 magnification, and the middle left and right columns are ×40 magnification. There were more cells staining brown located in the intima and media at the vein-to-graft anastomosis compared with controls

Fig. 6

A: graph representing the appropriate protein band of VEGF receptor (VEGFR)-2 from Western blot analysis. B: graph representing the appropriate band for β-actin from Western blot analysis for protein loading. C: semiquantitative analysis performed on days 3, 7, and 14. The normalized density of VEGFR-2 was significantly higher on day 7 compared with day 3. *Significantly higher value (P < 0.05). Data are means ± SD

Fig. 7

A: graph representing the appropriate protein band of VEGFR-1 from Western blot analysis. B: graph representing the appropriate band for β-actin from Western blot analysis for protein loading. C: semiquantitative analysis performed on days 3, 7, and 14. The normalized density of VEGFR-1 was significantly higher on day 14 compared with the control vein. *Significantly higher value (P < 0.05). Data are means ± SD. D: immunohistochemistry for VEGFR-1 for localization of expression on day 14. Representative sections from a day 14 animal are shown. The left and middle left columns are from the vein-to-graft anastomosis, and the middle right and right columns are from the control vein. The left and middle right columns are ×10 magnification, and the middle left and right columns are ×40 magnification. By immunohistochemistry, brown staining cells are positive for VEGF-A and VEGFR-1. There were more cells staining brown located in the intima and media at the vein-to-graft anastomosis compared with controls.

Fig. 8

A: graph representing the appropriate protein bands of pro-matrix metalloproteinase (MMP)-2 and MMP-2 by zymography. STD, standard. B: semiquantitative analysis performed on days 3, 7, and 14. The normalized density of pro-MMP-2 and active MMP-2 at the vein-to-graft anastomosis was significantly higher than the control veins by days 7 and 14. *Significantly higher value (P < 0.05). Data are means ± SD. C: graph representing the appropriate protein band of pro-MMP-9 by zymography. D: semiquantitative analysis performed on days 3, 7, and 14. The normalized density of pro-MMP-9 was significantly higher (*) at all time points compared with the control vein. E: graph representing the appropriate protein band of tissue inhibitor of matrix metalloproteinase (TIMP)-1 by Western blot analysis. F: graph representing the appropriate band for β-actin from Western blot analysis for protein loading. G: semiquantitative analysis performed on days 3, 7, and 14. The normalized density of TIMP-1 was significantly higher on day 14 compared with the control vein. *Significantly higher value (P < 0.05).

Fig. 9

Schematic depicting potential pathways from the present study