CorMatrix Wrapped Around the Adventitia of the Arteriovenous Fistula Outflow Vein Attenuates Venous Neointimal Hyperplasia

Venous neointimal hyperplasia (VNH) at the outflow vein of hemodialysis AVF is a major factor contributing to failure. CorMatrix is an extracellular matrix that has been used in cardiovascular procedures primarily as scaffolding during surgery. In the present study, we sought to determine whether CorMatrix wrapped around the outflow vein of arteriovenous fistula (AVF) at the time of creation could reduce VNH. In mice, the carotid artery to the ipsilateral jugular vein was connected to create an AVF, and CorMatrix scaffold was wrapped around the outflow vein compared to control mice that received no scaffolding. Immunohistochemistry, Western blot, and qRT-PCR were performed on the outflow vein at 7 and 21 days after AVF creation. In outflow veins treated with CorMatrix, there was an increase in the mean lumen vessel area with a decrease in the ratio of neointima area/media + adventitia area (P < 0.05). Furthermore, there was a significant increase in apoptosis, with a reduction in cell density and proliferation in the outflow veins treated with CorMatrix compared to controls (P < 0.05). Immunohistochemical analysis revealed a significant reduction in fibroblasts, myofibroblasts, macrophages, and leukocytes with a reduction in Tnf-α gene expression (P < 0.05). In conclusion, outflow veins treated with CorMatrix have reduced VNH.

Outflow veins treated with CorMatrix have a significant decrease in the average venous neointimal hyperplasia with an increase in average lumen vessel area. Vascular remodeling of the outflow vein was determined at 7 and 21 days after AVF placement by performing histomorphometric analysis using Verhoeff-Van Gieson (VVG) staining as previously described by our laboratory 14,15 . With VVG staining, the different layers of the vessel wall could be identified including the neointima, media + adventitia, and the vascular lumen as shown ( Fig. 2A). By day 21, the average lumen vessel area was significantly increased in the outflow vein from Group S compared to Group C (41, 541.1 ± 8, 919 μm 2 vs. 5, 079.7 ± 832.5 μm 2 , respectively; average increase: 817%, P < 0.0001). The average ratio of neointimal area/media + adventitia area was significantly reduced in the outflow veins from Group S compared to Group C at day 21 (1.35 ± 0.18 vs. 3.0 ± 0.76, respectively; average decrease: 55%, P < 0.05, Fig. 2B). Hematoxylin and Eosin (H&E) staining was used to assess cell density (images not shown). The results showed the average cell density in the neointima of the outflow veins from Group S was significantly lower than Group C at day 21 (7, 990.7 ± 1, 027.3/μm 2 vs. 43, 041.7 ± 8, 377/μm 2 , respectively; average decrease: 81%, P < 0.001, Fig. 2C). Next, we examined the changes in collagen 1 and 3 in the outflow veins using Picrosirius red staining. Qualitatively, this demonstrated a reduction in the intensity of Sirius red staining in Group S compared to Group C. In aggregate, this indicates that there is a decrease in constrictive remodeling of the outflow veins removed from Group S compared to Group C ( Supplementary Fig. S1).
Outflow veins treated with CorMatrix have a significant increase in TUNEL staining with reduced cellular proliferation. We speculated that the decrease in cell density could be due to either an increase in apoptosis 16 or a decrease in cell proliferation. Apoptosis was evaluated using Terminal deoxynucleotidyl transferase (TdT) dUTP Nick-End labeling (TUNEL) staining. The mean TUNEL (+) cells in Group S was significantly increased by day 7 (18.8 ± 1.79 vs. 1.95 ± 0.46; average increase: 962%, P < 0.001, Fig. 3A upper row) and remained increased at day 21 (26.1 ± 2.8 vs. 5.6 ± 0.5; average increase 516%, P < 0.001) when compared to Group C (Fig. 3B, Supplementary Fig. S2). To confirm these results, we measured cleaved caspase-3 levels in the outflow veins using Western blot analysis at day 21 (Fig. 4). Outflow veins removed from Group S had higher caspase-3 levels compared to Group C. Cell proliferation was evaluated using Ki-67 staining (Fig. 3A, bottom row). There was a significant reduction in Ki-67 staining in Group S compared to Group C by day 7 (26.3 ± 1.7 vs. 43 ± 3.9; average decrease: 38.9%, P < 0.01, Fig. 3C). However, there was no significant difference by day 21. Together, these results suggest that outflow veins treated with CorMatrix wrap had decreased cellular proliferation at day 7 with increased apoptosis at 7 and 21 days after AVF placement.
By day 21, there was a significant decrease in the average number of α-SMA (+) cells in the outflow veins removed from Group S compared to Group C (9.0 ± 1.9 vs. 20.0 ± 2.1; average decrease: 54.9%, P < 0.01, Fig. 5). Taken together, these results suggest that outflow vein treated with CorMatrix wrap have a significant decrease in fibroblasts and smooth muscle cells. Western blot analysis of α-SMA and FSP-1 confirmed these results (Fig. 4).

Outflow veins treated with CorMatrix have a significant decrease in HIF-1α expression.
Hypoxia has been implicated as one of the factors contributing to VNH 12,19 . To examine this, we performed HIF-1α staining. At day 7, we observed that the average staining density for HIF-1α was significantly reduced in outflow veins removed from Group S compared to Group C (10.8 ± 2.3 vs. 37.0 ± 2.6; average decrease: 70.6%, The average lumen vessel area in Group S vessels is significantly increased compared to Group C at day 21 (P < 0.0001). (C) The ratio of neointima/media + adventitia (M + Ad) is decreased in Group S compared to Group C at day 21 (P < 0.05). (D) Cell density in the neointima is decreased in Group S compared to Group C at day 21 (P < 0.0001). Each bar represents mean ± SEM (n = 4-6). * P < 0.05; ## P < 0.0001.
Outflow veins treated with CorMatrix wrap have a significant reduction in macrophage and leukocyte infiltration. Previous studies in clinical specimens and in pre-clinical AVF models have demonstrated that there is increased CD68 (+) cells present 12,19 . In order to assess whether CorMatrix wrap had an effect on macrophage infiltration, we stained tissue sections for CD68. By day 7, a significant decrease in the average CD68 (+) cells in outflow veins treated with Group S compared to Group C was observed (5.8 ± 2.2 vs.14.2 ± 2.3; average decrease: 58.6%, P < 0.05, Fig. 7). By day 21, there was a no significant difference between both groups. These results were confirmed using Western blot analysis for CD68 levels in outflow veins treated with Group S compared to Group C (Fig. 4).
Leukocyte infiltration in the vessel walls 20 was assessed by CD45 staining. CD45 staining was significantly decreased in outflow veins from Group S compared to Group C at day 7 (18.6 ± 4.2 vs. 43.8 ± 7.2; average decrease: 57.5%, P < 0.05, Fig. 7) and remained significantly decreased by day 21 (33.0 ± 7.1 vs. 59.1 ± 0.9; average Dark brown nuclei are positive for TUNEL (top row) and brown nuclei are positive for Ki-67 (bottom row) as indicated by arrows. All images were taken at 100× magnification and the scale bar is 20 μm. (B) Quantitative analysis of TUNEL staining for apoptosis. Mean TUNEL staining was significantly increased in Group S compared to Group C at day 7 (P < 0.0001) and day 21 (P < 0.0001). (C) Quantitative analysis of Ki-67 staining for cell proliferation was performed. The average Ki-67 density was significantly reduced in Group S compared to Group C at day 7 (P < 0.01). Each bar represents mean ± SEM (n = 6). # P < 0.01; ## P < 0.0001.

Outflow veins treated with CorMatrix Wrap have a significant decrease in Tnf-α gene expression.
In order to identify a mechanism for the present observations, we performed gene expression of Tumor necrosis factor alpha (Tnf-α) using qRT-PCR in the outflow veins. The average gene expression of Tnf-α gene was significantly decreased in outflow veins treated with Group S compared to Group C by day 7 ( Fig. 8; P < 0.01) demonstrating a reduction in inflammatory cytokine expression.

Discussion
Our results from the present study identify an important role for using CorMatrix wrap around outflow veins of AVF in reducing VNH in an experimental mouse model. The benefits of CorMatrix wrap include an increase in lumen vessel area of the outflow vein with positive vascular remodeling of the AVF with a reduction in collagen deposition. This was accompanied by an increase in apoptosis with a reduction in cellular proliferation. There is a reduction in fibroblasts, myofibroblasts, macrophages, and leukocytes accompanied with a decrease in gene expression of Tnf-α in outflow veins treated with CorMatrix ( Fig. 9), Recently, CorMatrix scaffold material, a decellularized extracellular matrix, has been used for cardiovascular applications [6][7][8][9][10][11] . Study performed by Kohler, et al. examined the effect of reducing the radius of the vessel wall acutely and thus diminishing wall stress using rigid polytetrafluoroethylene (PTFE) material and its effect on VNH. This study found that the vessel wall area and smooth muscle cell density were significantly reduced in tight-PTFE wrapped venous segments 21 . This also resulted in a decrease in the outflow vein dilation of AVF. In the present study, CorMatrix scaffold was wrapped loosely around the outflow vein of AVF reduces VNH accompanied by an increase in lumen area and results in positive outward vascular remodel.
The involvement of adventitial fibroblasts in the pathogenesis of hemodialysis vascular access dysfunction has been well described 17,18 . In the present study, a significant reduction in fibroblasts and myofibroblasts was observed in the outflow vein treated with CorMatrix compared to the control group. The results were confirmed by Western blot for α-SMA and FSP-1. We speculate that CorMatrix wrap provides a natural infrastructure for fibroblasts to migrate to and reside in, thereby reducing fibroblast migration toward the intimal region of the vessel wall.
Studies from clinical samples and pre-clinical models show that an increase in cell proliferation and a decrease in apoptosis are critical pathologic changes observed in failed hemodialysis vascular access [22][23][24] . This occurs when arterial blood flows into the vein, which leads to an increase in blood flow rate and interstitial pressure, which this is hypothesized to cause an increase in apoptosis 25 . Taken together, this results in an imbalance between cell apoptosis and proliferation in VNH. In the present study, we observed a significant reduction in cellular proliferation in vessels treated with CorMatrix wrap compared to controls. One of the reasons for using the scaffold wrap is that it exerts an anti-proliferative effect in the vessel wall, which may be due to a decrease in fibroblast or myofibroblast differentitation. Moreover, in addition to reduced cellular proliferation, there is a substantial increase in TUNEL staining and cleaved caspase-3 levels in scaffold-wrapped vessels indicating that CorMatrix wrap increases apoptosis in the outflow vein. Collectively, these results suggest that CorMatrix wrap modulates cell turnover in the vessel wall and thus reduces VNH.
Increased HIF-1α expression has been observed in the outflow vein of AVF in animal models and clinical specimens removed from patients of hemodialysis vascular access failure 12,19 . Hypoxic injury to the vascular wall is one of the initial factors contributing to VNH and arterial bypass graft failure [26][27][28] . Consistent with this, we found that HIF-1α staining was markedly decreased in the outflow veins treated with CorMatrix. This may be due to a decrease in cell density that may reduce oxygen demand in the vessel wall, resulting in decreased HIF-1α expression. Decreased HIF-1α expression may reduce the hypoxia-induced inflammatory response in the outflow veins treated with CorMatrix scaffold. One possible explanation for hypoxic injury is the disruption of vasa vasorum that supplies blood to the vessel wall during AVF creation 14 . In addition the wound healing process is a trigger for hypoxia signaling. Hypoxia signaling is known to modulate cell migration, proliferation, and differentiation resulting in VNH formation and vascular stenosis. In support of this notion, hypoxia accelerates the conversion of fibroblast to myofibroblasts (α-SMA-positive cells) and promotes proliferation and invasion of myofibroblasts 14,15 .
A decrease in inflammatory cells such as leukocytes and macrophages has been associated with improved fistula function in experimental animal models [29][30][31] . Anti-inflammatory molecules such as simvastatin have been shown to increase lumen vessel area in AVF outflow veins 15 . Consistent with these observations, we found a substantial decrease in macrophages and leukocytes in the vascular walls of outflow veins wrapped with CorMatrix, which was associated with reduced VNH.
To explore the molecular mechanism of CorMatrix wrap reducing VNH, gene expression of Tnf-α was analyzed by qRT-PCR. This demonstrated a significant decrease in the average gene expression of Tnf-α in outflow veins treated with CorMatrix wrap compared to control. Studies have shown that TNF-α is produced by different cell types including fibroblasts, smooth muscle cells and endothelial cells in response to hypoxia, sheer stress and inflammation 32 . Furthermore, fibroblasts tend to acquire pro-inflammatory and fibrogenic phenotype during The average α-SMA density is significantly reduced in Group S compared to Group C at day 21 (P < 0.01). (C) Quantitative analysis for FSP-1 staining. The average FSP-1 density is significantly reduced in Group S compared to Group C at day 7 (P < 0.01) and day 21 (P < 0.0001). Each bar represents mean ± SEM (n = 6). # P < 0.01; ## P < 0.0001.
SCIenTIfIC RepoRTS | 7: 14298 | DOI:10.1038/s41598-017-14696-z hypoxia 33 . In the present study, vessels wrapped with CorMatrix wrap have shown a decrease in fibroblast content. The decrease in inflammatory fibroblasts may explain the low Tnf-α gene levels. Collectively, our results suggest the effect of CorMatrix wrap in reducing VNH may be due to the fact that the CorMatrix wrap acts as a trap for fibroblast migration and also attenuates the proliferation of fibroblast and myofibroblasts through Tnf-α pathway.

Limitations.
A murine model of normal kidney function was employed due to the high mortality in immunodeficient mice with chronic kidney disease, and thus the effects of chronic kidney disease could not be evaluated. Further studies should be performed to corroborate these findings using immunocompetent animals. Moreover, a larger animal model should be employed for the study.
In conclusion, this study demonstrates that utilizing CorMatrix wrap around the outflow vein in a murine model improves fistula function and reduces VNH. This is accompanied by a reduction of several cells, including fibroblasts, myofibroblasts, macrophages, and leukocytes. The present study provides a potential therapeutic approach to reducing VNH formation associated with AVF.

Creation of the Arteriovenous Fistula with outflow veins treated with CorMatrix wrap. B6.
Cg-Foxn1 nu /J mice (n = 46) underwent a surgical procedure for placement of an AVF as described previously 14,15,34 . The mice were anesthetized with ketamine hydrochloride (0.1-0.2 mg/g) and xylazine (0.02 mg/g) by intraperitoneal injection. The skin overlying the common carotid artery and ipsilateral external jugular vein Quantitative analysis for CD45 staining in the outflow vein is demonstrated. There is a significant reduction in CD45 staining in Group S compared to Group C at day 7 (P < 0.05) and day 21 (P < 0.05). Each bar represents mean ± SEM (n = 6 per group). * P < 0.05. Figure 8. Tnf-α gene expression at day 7 in AVF outflow veins. Tnf-α gene expression using qRT-PCR was assessed and it was significantly reduced in Group S compared to Group C (P < 0.01). Each bar represents mean ± SEM (n = 6 per group). # P < 0.01. was incised and these vessels identified and separated. All branches of the jugular vein were ligated. The common carotid artery was clamped proximally; the distal side was tied immediately below the carotid bifurcation and transected. The distal end was pulled through, inverted over, and tied to the cuff (Laboratory PolyE Polyethylene Non-Sterile Tubing, Harvard Apparatus, Holliston, MA) of the AVF. The jugular vein was ligated distally and transected. The transected end was then placed over a cuff and tied with an 8-0 Ethilon suture to form an end-to-end anastomosis between the common carotid artery and jugular vein. A piece of CorMatrix scaffold (1 × 4 mm) was wrapped around the outflow vein and sutured using 8-0 nylon to secure the scaffold to the outflow vein as shown in Fig. 1. After the surgical procedure, all animals were allowed to recover under heat lamps to maintain body temperature. Post recovery, the animals were transferred to their housing facility.
At the end of the experiment, 7 and 21 days after AVF placement, the mice were sacrificed while under anesthesia. Animals were perfused with 4% buffered formalin through the heart as described previously 15 . The outflow veins were excised under the microscope, and fixed in 4% formalin for immunohistochemistry (n = 6 per group, for both time points). Outflow veins and contralateral jugular vein were harvested and flash frozen in liquid nitrogen for Western blotting (n = 3 per group, for day 7 only). Outflow veins and contralateral jugular veins were harvested and placed in RNA later for PCR analysis (n = 6 per group, for day 7 only).

Verhoeff-Van Gieson Staining.
To measure the formation of neointima and different layers of the vessel, VVG staining (Newcomer Supply, Middleton, WI) was performed following the manufacturer's protocols.

Picrosirius Red staining.
To measure collagen deposition, tissue sections of AVF outflow veins at day 7 and day 21 were stained for Picrosirius red as described previously 15 .
TUNEL Staining. The extent of apoptotic cell death in the outflow veins was measured using TUNEL staining (Trevigen, Gaithersburg, MD) following the manufacturer's protocols. Tissue sections stained with TUNEL but without terminal deoxynucleotidyl transferase enzyme treatment served as negative controls. Western Blot. The outflow veins were washed in PBS and homogenized in RIPA buffer containing inhibitors for proteases and phosphatases. Each group had 3 samples that were pooled together for Western blot analysis due to low amount of protein per sample. Tissue lysates were clarified by centrifugation at 12,000 × g for 30 minutes at 4 °C. Protein content was determined using the Bio-Rad DC Protein Assay Kit (Bio-Rad Laboratories, Hercules, CA). Lysates (30-μg) were resolved on 4% to 20% gradient SDS-PAGE and then transferred to PVDF membrane following standard protocols. The membranes were probed for α-SMA, CD68, FSP-1, caspase-3, and cleaved caspase-3 (activated caspase-3).
Real-time Polymerase Chain Reaction Analysis. The outflow vein and contralateral jugular vein were harvested and kept in RNA stabilization solution (Qiagen, Gaithersburg, MD). Samples were homogenized and the total RNA was isolated using the RNeasy Mini Kit (Qiagen). Gene expression was examined using real-time PCR analysis. Primer sequences for TNF-α included: Forward, GCTCTTCTGTCTACTGAACTTCG, Reverse, GATGAGAGGGAGGCCATTTG. PCR conditions included 95 °C for 5 min, 95 °C 15 sec, 55 °C 30 sec, 45 cycles using iTaq Universal SYBR mix in a Bio-Rad C1000 thermal cycler with CFX96 real-time system. The fold change in the gene expression was calculated following ΔΔ CT method after normalizing with 18S RNA and contralateral jugular vein.
Morphometric and Image Analysis. Image analysis was done as described previously 15 . Specifically, images of the entire cross section (5× magnification) were captured using a Carl Zeiss microscope and analyzed using Zen 2.3 digital imaging software (Zeiss, US). Immunostaining for Ki-67 (brown), α-SMA (brown), FSP-1 (brown), CD68 (brown), TUNEL positive (dark brown), and HIF-1α (brown) were selected, in turn, by selecting the appropriate red-green-blue color intensity range and then counted. The color intensity was kept the same for all sections for a particular antibody staining. Cell density as well as the neointima, media + adventitia area, and lumen vessel area were measured by manually tracing the vessel wall using the Zen 2.3 software. In the present study, the area of the neointima was identified by the internal elastic lamina, which is poorly defined in veins. In addition, the collagen deposition in the outflow vein of the AVF can be used to identify the neointima, as there is excessive collagen deposition in the neointima when compared to media and adventitia. An independent technician blinded to the experiment completed all image analysis. Slides from animals with poor tissue quality were excluded from analysis. Statistical Analysis. Data are expressed as mean ± SEM. The Student t test was conducted with post hoc Bonferroni correction. Significant differences between two groups were indicated by *P < 0.05, # P < 0.01, ## P < 0.0001. JMP Pro software (SAS Institute, Inc., Cary, NC) was used for statistical analysis.