Background: Proteoglycan 4 (PRG4; lubricin) is a member of two gene co-expression network modules associated with human vein graft failure. However, little is known about PRG4 and the vascular system. Therefore, we have investigated the effects of recombinant human PRG4 (rhPRG4) on cell migration and proliferation in human veins. Methods: Effects of rhPRG4 on cell migration, proliferation, and neointima formation were determined in human venous tissue and cultured venous smooth muscle cells (SMCs), adven-titial cells, and endothelial cells. Expression of PRG4 by cultured human saphenous veins, failed vein grafts, and varicose veins was determined by immunostaining or Western blotting. Results: Limited expression of PRG4 in fresh saphenous veins was dramatically increased around medial SMCs after culture ex vivo. rhPRG4 inhibited the migration of cultured SMCs, adventitial cells, and endothelial cells, as well as the proliferation of endothelial cells. rhPRG4 also inhibited the migration of SMCs and adventitial cells from tissue explants, but there was no effect on cell proliferation or neointima formation in ex vivo whole veins. Finally, PRG4 was largely absent in two examples of venous pathology, that is, failed human vein grafts and varicose veins. Conclusions: Although rhPRG4 can inhibit the migration of venous SMCs, endothelial cells, and adventitial cells, and the proliferation of endothelial cells, PRG4 was only increased around medial SMCs in veins after ex vivo culture. PRG4 was not observed around medial SMCs in failed human vein grafts and varicose veins, suggesting the possibility that a failure of PRG4 upregulation may promote these pathologies. (C) 2020 Elsevier Inc. All rights reserved.

Objectives: Approximately 30% of human vein grafts fail within the first year of implantation secondary to neointimal hyperplasia and adverse remodeling. This process has been difficult to study in animal models because of lack of a similar pathology. Therefore, we created a novel xenograft model where human saphenous vein was implanted into the infrarenal aorta of Rowett nude rats. Methods: Weekly ultrasound was performed before pressure perfusion fixation on postoperative day 28. Paraffin-embedded cross-sections were analyzed for human vs rat cells (using human-specific antibodies to mitochondria and Ku80) and for changes in wall thickening and extracellular matrix (ECM; Movat’s stain and antibodies to versican and aggrecan). Results: Successful human vein grafts (n=4) were ~150% larger than the native rat aorta as documented by photomicrograph (Figure 1A) and ultrasound (Figure 1B top) and showed appropriate arterial Doppler waveforms (Figure 1B bottom). Particularly robust neointimal, but also medial and adventitial, hyperplasia was observed in Movat’s stained cross-sections of the vein (Figure 1C). The neointimal layer was primarily ECM with relatively few cells. Surprisingly, the neointima stained poorly for versican, but was strongly positive for aggrecan. While neither human-specific antibody stained all cells, human cells were noted throughout the vein wall in similar numbers seen in human saphenous vein controls. Conclusions: This novel human vein xenograft model will be suitable for testing therapeutic strategies for decreasing neointimal hyperplasia leading to human vein graft failure. The novel observation of aggrecan in the neointima suggests new functions for aggrecan in the vascular setting.

Objectives: One third of infrainguinal vein bypasses may fail within the first 1.5 years. Pro- and anti-inflammatory mechanisms are thought to be involved in these graft stenoses and occlusions. In previous studies, low levels of anti-phosphorylcholine IgM (anti-PC IgM, an innate anti-inflammatory IgM) have been associated with increased cardiovascular events. In this study, the peri-operative dynamics of anti-PC IgM levels were established during leg bypass surgery, and associations assessed between anti-PC IgM levels and primary graft patency. Design and methods: This was a prospective, observational cohort study of infrainguinal autogenous vein bypass for peripheral arterial occlusive disease involving four university affiliated hospitals. Plasma cytokine and anti-PC IgM levels were measured pre- and post-operatively. The outcome of interest was loss of primary graft patency because of occlusion or intervention for graft stenosis. Results: One hundred and forty-two consecutive patients were enrolled: mean age 66 (46-91); 91% white race and male; 72.5% critical limb ischaemia (Fontaine III or IV). Median pre-operative anti-PC IgM levels were 49 units/mL (IQR 32.3-107.7, mean 89.8 + 101 sd). During follow up of an average of 1.8 years (1 month-7.4 years), 50 (35.2%) grafts lost primary patency. Pre-operative levels of interleukin 6 or C-reactive protein did not predict graft failure. Patients with pre-operative anti-PC IgM values in the lowest quartile had a twofold increased risk of graft failure (multivariable Cox proportional hazard, p = .03, HR 2.11, 95% CI 1.09-4.07), even after accounting for the other significant factors of conduit diameter, distal anastomosis, smoking, and the severity of leg ischaemia. Conclusions: Low levels of anti-PC IgM are associated with vein bypass graft failure. This biological mediator may be a useful marker to identify patients at higher risk, and offers the potential for novel, directed therapies for vascular inflammation and its consequences.

Changes in extracellular matrix proteins may contribute significantly to the adaptation of vein grafts to the arterial circulation. We examined the production and distribution of versican and hyaluronan in intact human vein rings cultured ex vivo, veins perfused ex vivo, and cultured venous adventitial and smooth muscle cells. Immunohistochemistry revealed higher levels of versican in the intima/media compared to the adventitia, and no differences in hyaluronan. In the vasa vasorum, versican and hyaluronan associated with CD34+ progenitor cells. Culturing the vein rings for 14 days revealed increased versican immunostaining of 30-40% in all layers, with no changes in hyaluronan. Changes in versican accumulation appear to result from increased synthesis in the intima/media and decreased degradation in the adventitia as versican transcripts were increased in the intima/media, but unchanged in the adventitia, and versikine (the ADAMTS-mediated cleavage product of versican) was increased in the intima/media, but decreased in the adventitia. In perfused human veins, versican was specifically increased in the intima/media in the presence of venous pressure, but not with arterial pressure. Unexpectedly, cultured adventitial cells express and accumulate more versican and hyaluronan than smooth muscle cells. These data demonstrate a differential regulation of versican and hyaluronan in human venous adventitia vs. intima/media and suggest distinct functions for these extracellular matrix macromolecules in these venous wall compartments during the adaptive response of vein grafts to the arterial circulation.

Objective: When an autogenous vein is harvested and used for arterial bypass, it suffers physical and biologic injuries that may set in motion the cellular processes that lead to wall thickening, fibrosis, stenosis, and ultimately graft failure. Whereas the injurious effects of surgical preparation of the vein conduit have been extensively studied, little is known about the influence of the clinical environment of the donor leg from which the vein is obtained. Methods: We studied the cellular responses of fresh saphenous vein samples obtained before implantation in 46 patients undergoing elective lower extremity bypass surgery. Using an ex vivo model of response to injury, we quantified the outgrowth of cells from explants of the adventitial and medial layers of the vein. We correlated this cellular outgrowth with the clinical characteristics of the patients, including the Wound, Ischemia, and foot Infection classification of the donor leg for ischemia, wounds, and infection as well as smoking and diabetes. Results: Cellular outgrowth was significantly faster and more robust from the adventitial layer than from the medial layer. The factors of leg ischemia (P<.001), smoking (P=.042), and leg infection (P=.045) were associated with impaired overall outgrowth from the adventitial tissue on multivariable analysis. Only ischemia (P=.046) was associated with impaired outgrowth of smooth muscle cells (SMCs) from the medial tissue. Co-culture of adventitial cells and SMCs propagated from vein explants revealed that adventitial cells significantly inhibited the growth of SMCs, whereas SMCs promoted the growth of adventitial cells. The AA genotype of the -838C>A p27 polymorphism (previously associated with superior graft patency) enhanced these effects, whereas the factor of smoking attenuated adventitial cell inhibition of SMC growth. Comparing gene expression, the cells cultured from the media overexpress Kyoto Encyclopedia of Genes and Genomes pathways associated with inflammation and infection, whereas those from the adventitia overexpress gene families associated with development and stem/progenitor cell maintenance. Conclusions: The adverse clinical environment of the leg may influence the biologic behavior of the cells in the vein wall, especially the adventitial cells. Chronic ischemia was the most significant factor that retards adventitial cell outgrowth. The cells arising from the vein adventitia may be key players in determining a healthy adaptive or a pathologic response to the injuries associated with vein grafting. (J Vasc Surg 2018;68:165S-76S.) Clinical Relevance: We prospectively studied saphenous veins from 46 infrainguinal bypasses and found that ex vivo cellular outgrowth was significantly impaired by the deleterious clinical factors of leg ischemia, smoking, and infection-especially for the adventitial layer of the vein. The adventitial cells may play an important role in the healthy remodeling of a successful vein graft.

Introduction: A p27 Kip1 single nucleotide polymorphism (SNP) is associated with vein graft failure. The protective genotype (AA) is associated with slower growth of human saphenous vein adventitial cells, but not of smooth muscle cells (SMCs). We investigated the influence of patient clinical characteristics on the migration of adventitial cells and SMCs from vein tissue and the interaction between adventitial cells and SMCs on cell growth. Methods and Results: Tissue explants of the adventitia and intima/media were prepared from samples of vein used for leg bypass, and the number of cells emerging from explants was counted over time (15 replicates/vein, 43 veins). Correlating migration with clinical characteristics of the patient, the degree of leg ischemia was an important factor. Migration of adventitial cells was slower with more severe ischemia (AAI<0.6; N=12) compared to milder ischemia (AAI≥0.6; N= 31): 37.0 ± 4.9 vs 53.8 ± 7.7 cells/explant day 7; mean ± SEM, P<.03). Migration of SMCs from the intima/media explants showed no influence of ischemia (19.1 ± 3.0 vs 24.1 ± 6.1; P=.12). The p27 SNP genotype, diabetes, smoking, and wounds/infection bore no correlation with migration. To test cell interactions, adventitial cells and SMCs from 11 veins were seeded in triplicate on opposite sides of 0.4 um Transwell filters that prevent cell migration (6 AA and 5 CC p27 SNP genotypes), such that the same or different cells were opposed. Cells were counted for a Day4/Day1 proliferation ratio. Compared to AA SMCs, AA adventitial cells significantly inhibited AA SMC growth (10.9 ± 2.5% inhibition; P<.01). In contrast, CC adventitial cells did not significantly inhibit CC SMC growth (4.1 ± 3.5% inhibition, P=.20). Conclusions: Migration of adventitial cells, but not SMCs, from vein tissue is inhibited by increased ischemia. This suggests that adventitial cells may be beneficial, since more severe ischemia has been associated with poorer graft patency. Supporting this, adventitial cells with the protective AA genotype (but not the CC genotype) inhibit the growth of SMCs. These data suggest that adventitial cells and their in situ precursors may inhibit intimal hyperplasia and graft failure by inhibiting SMC proliferation in a p27 SNP genotype-dependent manner.

Clowes et al., Circulation Research 58(6): 839-845, 1986. Currier et al., JACC 17(6): 118B-125B, 1991. Schmid et al., Seminars in Thrombosis and Hemostasis 19 Suppl. 1: 155–159, 1993. European Pat. Off.. WIPO. Edelman et al., Proc. Natl. Acad. Sci. U.S.A. 87:3773-3777, 1990. Rubin et al., The Lancet 1 (8599): 1353-1356, 1988. Ferns et al., Science 25: 1129-1132, 1991. Linder et al., J. Clin. Invest. 85:2004-2008, 1990. Jawien et al., J. Clin. Invest. 89: 507-511, 1992. Linder et al., J. Clin. Invest. 90: 2044-2049, 1992. Popma et al., Circulation 84(3): 1426-1436, 1991. Ferrell et al., Circulation 85(4): 1630-1631, 1992. Reilly et al., J. Cell. Phys. 136: 23-32, 1988. Guyton et al., Circulation Research 46(5): 625-634, 1980. Cavari et al., Cell Biology International 17(8): 781-786, 1993. Kimura et al., Japan. J. Pharmacol. 59: 51-56, 1992. Castellot, Jr. et al., J. Cell. Biol. 109(6): 3147-3155, 1989. Bioworld Today 5(4): 1, 3; 1994. Buchwald et al., Circulation 86(2): 531-537, 1992. Harris et al. TEBTECH 11:42-44(1993). Mullins Arterioscler Thromb 14:1047-1055 (1994). DeFeudis DN&P 5(1) 49-51 (1992). Kelly et al. J. Biol Chem. 266 8987-8992(1991). Kawahara et al. Biochem Biophys Res 147:839-845 (1987). Commun

Objective: Venous valves are prone to injury, thrombosis and fibrosis. We compared the behavior and gene expression of smooth muscle cells (SMCs) in the valve sinus vs non-valve sites to elucidate biological differences associated with vein valves. Methods: SMC migration was measured using 2.5 mm 2 explants of the intima/media of valve sinus segments (without valve leaflets) vs. non-valve segments of human saphenous veins. Proliferation and death of SMCs was determined by staining for Ki67 and TUNEL, respectively. Proliferation and migration of passaged valve vs non-valve SMCs was determined by cell counts and using microchemotaxis chambers. Global gene expression in valve vs non-valve intima/media was determined by RNA-Seq. Results: Valve SMCs demonstrated greater proliferation within tissue explants compared to non-valve SMCs (19.3±5.4% vs. 6.8±2.0% Ki67 positive nuclei at 4 days, respectively; mean ± SEM, 5 veins; P Conclusions: Valve, compared to non-valve, SMCs have greater rates of migration and proliferation, which may in part explain the propensity for pathological lesion formation in valves. While FGF2 mediates these effects in cultured SMCs, the mediators of these stimulatory effects in valve wall tissue remain unidentified. Here, the newly identified differentially expressed genes may play a role.

Objective: Markers containing dyes such as crystal violet (CAS 548-62-9) are routinely used on the adventitia of vein bypass grafts to avoid twisting during placement. Because little is known about how these dyes affect vein graft healing and function, we determined the effect of crystal violet on cell migration and proliferation, which are responses to injury after grafting.Methods: Fresh human saphenous veins were obtained as residual specimens from leg bypass surgeries. Portions of the vein that had been surgically marked with crystal violet were analyzed separately from those that had no dye marking. In the laboratory, they were split into easily dissected inner and outer layers after removal of endothelium. This cleavage plane was within the circular muscle layer of the media. Cell migration from explants was measured daily as either (1) percentage of migration-positive explants, which exclusively measures migration, or (2) number of cells on the plastic surrounding each explant, which measures migration plus proliferation. Cell proliferation and apoptosis (Ki67 and TUNEL staining, respectively) were determined in dye-marked and unmarked areas of cultured vein rings. The dose-dependent effects of crystal violet were measured for cell migration from explants as well as for proliferation, migration, and death of cultured outer layer cells. Dye was extracted from explants with ethanol and quantified by spectrophotometry.Results: There was significantly less cell migration from visibly blue compared with unstained outer layer explants by both methods. There was no significant difference in migration from inner layer explants adjacent to blue-stained or unstained sections of vein because dye did not penetrate to the inner layer. Ki67 staining of vein in organ culture, which is a measure of proliferation, progressively increased up to 6 days in nonblue outer layer and was abolished in the blue outer layer. Evidence of apoptosis (TUNEL staining) was present throughout the wall and not different in blue-stained and unstained vein wall segments. Blue outer layer explants had 65.9 +/- 8.0 ng dye/explant compared with 2.1 +/- 1.3 for nonblue outer layer explants. Dye applied in vitro to either outer or inner layer explants dose dependently inhibited migration (IC50 similar to 10 ng/explant). The IC(50)s of crystal violet for outer layer cell proliferation and migration were 0.1 and 1.2 mg/mL, whereas the EC50 for death was between 1 and 10 mg/mL.Conclusions: Crystal violet inhibits venous cell migration and proliferation, indicating that alternative methods should be considered for marking vein grafts.

Objective: Approximately 30% of autogenous vein grafts develop luminal narrowing and fail because of intimal hyperplasia or negative remodeling. We previously found that vein graft cells from patients who later develop stenosis proliferate more in vitro in response to growth factors than cells from patients who maintain patent grafts. To discover novel determinants of vein graft outcome, we have analyzed gene expression profiles of these cells using a systems biology approach to cluster the genes into modules by their coexpression patterns and to correlate the results with growth data from our prior study and with new studies of migration and matrix remodeling.Methods: RNA from 4-hour serum-or platelet-derived growth factor (PDGF)-BB-stimulated human saphenous vein cells obtained from the outer vein wall (20 cell lines) was used for microarray analysis of gene expression, followed by weighted gene coexpression network analysis. Cell migration in microchemotaxis chambers in response to PDGF-BB and cell-mediated collagen gel contraction in response to serum were also determined. Gene function was determined using short-interfering RNA to inhibit gene expression before subjecting cells to growth or collagen gel contraction assays. These cells were derived from samples of the vein grafts obtained at surgery, and the long-term fate of these bypass grafts was known.Results: Neither migration nor cell-mediated collagen gel contraction showed a correlation with graft outcome. Although 1188 and 1340 genes were differentially expressed in response to treatment with serum and PDGF, respectively, no single gene was differentially expressed in cells isolated from patients whose grafts stenosed compared with those that remained patent. Network analysis revealed four unique groups of genes, which we term modules, associated with PDGF responses, and 20 unique modules associated with serum responses. The "yellow" and "skyblue" modules, from PDGF and serum analyses, respectively, correlated with later graft stenosis (P=.005 and P=.02, respectively). In response to PDGF, yellow was also associated with increased cell growth. For serum, skyblue was also associated with inhibition of collagen gel contraction. The hub genes for yellow and skyblue (ie, the gene most connected to other genes in the module), scavenger receptor class A member 5 (SCARA5) and suprabasin (SBSN), respectively, were tested for effects on proliferation and collagen contraction. Knockdown of SCARA5 increased proliferation by 29.9% +/- 7.8% (P<.01), whereas knockdown of SBSN had no effect. Knockdown of SBSN increased collagen gel contraction by 24.2% +/- 8.6% (P<.05), whereas knockdown of SCARA5 had no effect.Conclusions: Using weighted gene coexpression network analysis of cultured vein graft cell gene expression, we have discovered two small gene modules, which comprise 42 genes, that are associated with vein graft failure. Further experiments are needed to delineate the venous cells that express these genes in vivo and the roles these genes play in vein graft healing, starting with the module hub genes SCARA5 and SBSN, which have been shown to have modest effects on cell proliferation or collagen gel contraction.

Objectives: p27 kip1 , a gene affecting human response to arterial injury, affects collateralization after hindlimb ischemia (femoral artery ligation), possibly by regulating matrix metalloproteinase 2 (MMP2). We hypothesized that MMP2 mRNA expression would increase significantly more in ko mice after hindlimb ischemia and that MMP inhibition would affect p27 -/- (ko) collateralization less than p27 +/+ (wt). Methods: Ko and wt mice had their femoral and collateral arteries harvested seven days after hindlimb ischemia for RNA isolation and qRT-PCR. Ko and wt mice were also fed chow without or with doxycycline (a broad spectrum MMP inhibitor) and subjected to hindlimb ischemia. The mice were followed with weekly footpad laser Doppler perfusion imaging for 28 days, and then sacrificed for microCT scanning. Vascular smooth muscle cells (VSMC) isolated from either wt or ko aortae were used in Boyden chamber migration assays without and with an MMP2 specific inhibitor. Results: MMP2 mRNA expression increased significantly more in ko collaterals than wt collaterals (312% vs. 50%, respectively). However, blood flow recovery in ko and wt mice treated with doxycycline decreased to a similar extent (27±2.5% and 33±1% decrease respectively), although ko mice still revascularized better than wt mice ( p < 0.01). Gracilis collateral diameters increased less after doxycycline treatment in both groups compared to mice fed regular chow (78.5±12 and 80±7 vs 98±18 and 106±17 μm, respectively). The bridge collateral diameters increased less in ko mice fed doxycycline (66±32 vs 158±18.3 μm), and no bridge collaterals were detected in wt mice fed doxycycline. Wt mice fed regular chow had rare bridge collaterals. Ko VSMC migration was less affected by MMP2 inhibition than wt (29% less migration vs 48% less migration, p <0.01). Conclusions: MMP2 mRNA expression increased significantly more in p27 ko collaterals after hindlimb ischemia. The non-specific MMP inhibitor doxycycline inhibited revascularization after hindlimb ischemia, independent of p27 expression. p27 ko mice still revascularized better than wt mice after doxycycline treatment, suggesting that p27 has additional effects other than regulating MMP expression in the collateralization response.

Introduction: In spite of great advances in endovascular therapies, leg bypass with autogenous vein is still frequently required for peripheral arterial disease (PAD). However, primary graft failure may occur in up to 30% within the first two years, due to vein graft stenosis and/or thrombosis. A deranged inflammatory response to injury is suspected as the underlying pathobiology, but there are no known specific therapies, nor biologic markers for vein graft failure. Humans have natural antibodies that target a common inflammatory component of damaged cell membranes, phosphorylcholine (PC), which is exposed during cell injury. We performed an observational study of patients undergoing leg bypass, to see if their plasma levels of this anti-inflammatory IgM antibody correlated with the long term outcomes of their bypass grafts. Methods: A prospective, observational study of patients undergoing infrainguinal bypass with autogenous vein (excluding those on hemodialysis, and operations for aneurysmal disease). The primary outcome was the loss of primary patency (loss of graft flow, or surgical or endovascular intervention for a critical vein graft stenosis). Secondary outcome was the composite of myocardial infarction, stroke, and vascular death. Patients were followed regularly at clinically prescribed intervals to assess for graft and cardiovascular events. Pre- and postoperative plasma samples were assayed for levels of anti-PC IgM with the CVDefine kit from Athera Biotechnologies (Solna, Sweden). Results: 141 patients were followed for an average of 1.7 years (1 month to 7.4 years). Their mean age was 66 (46–91), 91% were white males. Indications for surgery were claudication (Fontaine II-39, 27.7%), rest pain (Fontaine III-30, 21.3%), and critical ischemia with ulceration (Fontaine IV-72, 51.1%). During follow-up, 44 (31%) of grafts lost primary patency due to thrombosis (11) or intervention for critical graft stenoses (33). Kaplan–Meier survival analysis revealed that primary graft patency was significantly worse in the patients in the lowest quartile of preoperative anti-PC IgM levels, compared with those in the top three quartiles (log rank p = .027). Combining primary graft and secondary cardiovascular endpoints, K–M survival showed a similar pattern of worse outcomes in the lowest quartile of anti-PC IgM levels (p = .06). Clinical factors known to affect graft patency were equally distributed within the two categories of IgM levels (including indication for surgery, intraoperative technical difficulties, length or diameter of graft). Multivariate Cox proportional hazards analysis revealed that patients with IgM values in the lowest quartile had a 2-fold increased risk of graft failure (95% confidence interval: 1.05–3.69), even after accounting for these other factors. Conclusion: Low levels of the naturally occurring, anti-PC IgM are associated with vein bypass graft failure, and higher levels may be protective. These effects may be mediated by the anti-inflammatory actions of the antibody, which are thought to quench the inflammatory effects of the exposed PC in damaged vascular cells. This novel biological mediator may be a useful marker to identify patients at higher risk of graft failure, and offers the potential for novel, directed therapies for vascular inflammation and its consequences. Disclosure of Interest: M. Sobel Research/Education Support from: Department of Veterans Affairs Research and Development Grant, M. Yagi: None Declared, G. Tang: None Declared, T. Kohler: None Declared, E. Wijelath: None Declared, R. Kenagy: None Declared, K. Moreno: None Declared.

Introduction: p27 Kip1 (p27) is a cell-cycle inhibitor whose -838C>A single nucleotide polymorphism (SNP) accounts for ~40% of the risk of peripheral vein graft failure. However, whether this SNP is functional has not been definitely established. Methods: To determine functionality, we investigated paired adventitial cells and smooth muscle cells (SMC) derived from fresh human saphenous veins (N≥7 lines per group). After growth arrest in 2% serum followed by stimulation with 10 ng/ml PDGF-BB, we measured p27 mRNA, p27 protein, and cell proliferation using qRT-PCR, Western blotting, and cell counts, respectively. Results: The SNP genotype was associated with 72 hour adventitial cell growth, but not SMC growth. Adventitial AA cells grew 33% slower than those with the CC genotype (Figure 1A, B, p=.004). We expected AA adventitial cells to produce more p27 mRNA, but paradoxically, p27 mRNA was lower in both cell types of the AA genotype (Figure 1C, D, P<.01). However, levels of p27 protein (a single ~27 kD band in both cell types) were ~2 fold higher in AA adventitial cells and SMCs compared to the CC cells at both 0 and 72 hours (Figure 1E, 72 hours; P<.001 for genotype; 0 h data not shown). Adventitial cells and SMCs made comparable amounts of p27 protein. Conclusion: The p27 Kip1 -838C>A SNP regulates levels of p27 in both venous adventitial cells and SMCs. There is an effect of genotype on the growth of adventitial cells, but not on SMCs. AA cells produce ~2 fold higher p27 protein than CC cells, despite AA cells having ~2 fold lower levels of p27 mRNA. These data demonstrate that the p27 SNP is functional, that different cell types within the vein wall respond uniquely to this genetic polymorphism, and suggest an important role in graft failure for the precursor of the cultured adventitial cell.

Introduction: About 30% of vein grafts fail because of intimal hyperplasia or negative remodeling. We have reported that vein graft cells from patients that develop stenosis proliferate more than cells from patients that maintain patent grafts. We have now analyzed gene expression of the same cell lines using a systems biology approach to cluster genes into modules based on their co-expression patterns and to correlate the results with graft outcome, growth data from our prior study, and with new studies of migration and matrix remodeling. Methods: RNA from 4 hour serum- or PDGF-BB-stimulated cells (13 non-stenotic and 7 stenotic cell lines) was used for microarray analysis of gene expression followed by weighted gene co-expression network analysis. Cell migration in microchemotaxis chambers in response to PDGF-BB and cell-mediated collagen gel contraction in response to serum were also determined. Gene function in growth or collagen gel contraction was determined using siRNA to inhibit gene expression. Results: Neither migration nor collagen gel contraction were correlated with graft outcome. While 1,188 and 1,340 genes were differentially expressed in response to serum and PDGF, respectively, graft outcome was not correlated with expression of any single gene. Network analysis revealed one module each from the separate analysis of the PDGF and serum data sets, which were called “Yellow” and “Skyblue” respectively, that were correlated with later graft stenosis (P=.005 and .02, respectively). Yellow was also associated with increased cell growth, and Skyblue was also associated with inhibition of collagen gel contraction. The hub genes for Yellow and Skyblue (i.e. the gene most correlated with other genes in the module), SCARA5 and SBSN, respectively, were tested for effects on proliferation and collagen contraction. SCARA5, but not SBSN, inhibited proliferation, and SBSN, but not SCARA5, inhibited collagen gel contraction. Conclusion: Using weighted gene co-expression network analysis of cultured vein graft cell gene expression, we have discovered a small number of genes of interest in vein graft failure. Further experiments are needed to discriminate the roles these genes play in vein graft healing starting with the module hub genes SCARA5 and SBSN.

Introduction: Lesions causing coronary and peripheral vein graft failure often occur at valves. To test the hypothesis that intrinsic differences in cells of the valve wall contribute to graft failure we measured cell migration from tissue explants and migration and proliferation of cultured cells from valve and non-valve sections of human saphenous vein (HSV). Methods: After removal of endothelium, HSVs were dissected into an outer wall with adventitia and some media and an inner wall with intima and some media. Cell migration from explants and migration (microchemotaxis chambers) and proliferation of cultured cells at passage 6 were measured. Results: Cells migrated faster from valve compared to non-valve tissue of the inner wall (Figure A; P<0.01, N=24 veins), but not the outer wall (not shown). PDGF-BB- or serum-mediated migration of valve inner wall smooth muscle cells (VSMC) was greater than with non-valve SMCs (NVSMC; P<0.05, N=6 pairs, 6.5 ± 1.2 vs. 4.2 ± 0.8 fold of control for PDGF-BB and 4.7 ± 0.6 vs. 3.4 ± 0.4 for serum). VSMC also proliferated faster than NVSMC after stimulation with PDGF-BB plus 2% serum (P<0.01, N=6 pairs, 3.3 ± 0.4 vs. 2.2 ± 0.2 fold of control). Finally, a blocking antibody to FGF2 had no effect on PDGF-mediated migration or proliferation of NVSMC (migration: IgG 2.2 ± 0.3 vs anti-FGF2 2.0 ± 0.4, N=6; proliferation: IgG 2.6 ± 0.1 vs anti-FGF2 2.5 ± 0.2, N=7), but largely blocked the enhanced migration and proliferation of VSMC (migration: Figure B; P<0.01, N=6; proliferation: control IgG 3.4 ± 0.3 vs. Anti-FGF2 Ab 2.7 ± 0.3 fold; P<0.01, N=7). Conclusion: The increased migration and proliferation of VSMC compared to NVSMC is largely mediated by FGF2.

Introduction: p27Kip1 is a cell-cycle inhibitor whose -838C>A single nucleotide polymorphism (SNP) accounts for ~40% of the risk of peripheral vein graft failure. Methods: To determine SNP function, we tested growth of cells derived from explants of the outer and inner wall of veins (cleavage through medial circular muscle yields outer wall with adventitia and some media and inner wall with intima and some media). Results: Growth of cells at passage 6 was measured over 3 days in response to 10 ng/ml PDGF-BB + 2% serum. Outer wall cells with the AA p27 SNP genotype grow slower than those with the CC genotype, but inner wall cells do not show a genotype-dependent response (Figure). Gene expression of quiescent outer and inner wall cells (5pairs) measured by microarray show that 2876 genes are differentially expressed (≥2 fold difference with P≤.05). While there is no difference in expression of traditional smooth muscle markers (smooth muscle α-actin, smooth muscle myosin heavy chain, smoothelin, SM22α, and calponin1), expression of the pericyte markers, NG2 and PDGFRβ, is significantly higher in outer wall cells (4.2 and 2.2 fold over inner wall cells; P<.02). We also found CD31 immunostaining of the endothelial cells of the vasa vasorum was restricted to the outer wall. Conclusion: These results support the functionality of the p27Kip1-838C>A SNP in cellular proliferation, but only in outer wall cells. Pericytes of the vasa vasorum or other progenitors in the outer wall may be involved, but further studies must address the identity of precursors of the outer wall cells that are associated with graft failure.

Markers containing dyes such as crystal violet (CAS 548-62-9) are routinely used on the adventitia of vein bypass grafts to avoid twisting during placement. Since little is known about how these dyes affect vein graft healing and function, we determined the effect of crystal violet on cell migration, a primary response to injury after grafting. Human saphenous veins were split into adventitia and media after removal of endothelium. Cell migration from 2.5 square mm explants in DMEM with 20% fetal bovine serum was measured daily for 8 days as either 1) % migration positive explants (≥ 1 cell/explant), which only measures migration, or 2) the number of cells/explant, which measures a combination of migration and proliferation. Dye was extracted from explants with ethanol and quantified by spectrophotometry. Cell migration by both methods was significantly less from blue, compared to non-blue, adventitial explants (P<.05, mean±SEM, N=11 veins with 15-30 explants/condition; Figures A and B). No medial explants were visibly blue, and there was no significant difference in migration of cells from medial explants of blue and non-blue sections of vein (data not shown). Blue adventitial explants had 65.9±8.0 ng dye/explant compared to 2.1±1.3 for non-blue explants (mean±SEM, N=7-11 veins). Dye applied in vitro to either adventitial or medial explants dose-dependently inhibited migration (IC50=21.5 ng/explant). Conclusion: Crystal violet is toxic and inhibits venous adventitial cell migration; alternative methods should be considered for marking vein grafts.

Objective: Factors responsible for the variability in outcomes after lower extremity vein bypass grafting (LEVBG) are poorly understood. Recent evidence has suggested that a single nucleotide polymorphism (SNP) in the promoter region of the p27(Kip1) gene, a cell-cycle regulator, is associated with coronary in-stent restenosis. We hypothesized an association with vein graft patency.Methods: This was a retrospective genetic association study nested within a prospective cohort of 204 patients from three referral centers undergoing LEVBG for claudication or critical ischemia. The main outcome measure was primary vein graft patency.Results: All patients were followed up for a minimum of 1 year with duplex graft surveillance (median follow-up, 893 days; interquartile range, 539-1315). Genomic DNA was isolated and SNP analysis for the p27(Kip1)-838C>A variants was performed. Allele frequencies were correlated with graft outcome using survival analysis and Cox proportional hazards modeling. The p27(Kip1)-838C>A allele frequencies observed were CA, 53%; CC, 30%; and AA, 17%, satisfying Hardy-Weinberg equilibrium. Race (P = .025) and history of coronary artery disease (P = .027) were different across the genotypes; all other baseline variables were similar. Primary graft patency was greater among patients with the -838AA genotype (75% AA vs 55% CA/CC at 3 years; P = .029). In a Cox proportional hazards model including age, sex, race, diabetes, critical limb ischemia, redo (vs primary) bypass, vein type, and baseline C-reactive protein level, the p27(Kip1)-838AA genotype was significantly associated with higher graft patency (hazard ratio for failure, 0.4; 95% confidence interval, 0.17-0.93). Genotype was also associated with early (0-1 month) changes in graft lumen diameter by ultrasound imaging.Conclusions: These data suggest that the p27(Kip1)-838C>A SNP is associated with LEVBG patency and, together with previous reports, underscore a central role for p27(Kip1) in the generic response to vascular injury. (J Vasc Surg 2013;57:1179-85.)