Published online before print August 14, 2006, doi: 10.1161/CIRCULATIONAHA.106.635292
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
(Circulation. 2006;114:820-829.)
© 2006 American Heart Association, Inc.
Molecular Cardiology |
2(VIII) Collagen Substrata Enhance Endothelial Cell Retention Under Acute Shear Stress Flow via an 2ß1 Integrin–Dependent Mechanism
An In Vitro and In Vivo Study
From UK Centre for Tissue Engineering (N.J.T., M.O.M., C.M.K., R.A.B., M.G.W., A.E.C.), Universities of Manchester (N.J.T., M.O.M., C.M.K., M.G.W., A.E.C.) and Liverpool (R.A.B.); Department of Vascular Surgery (N.J.T., M.O.M., M.G.W.), Manchester Royal Infirmary; Wellcome Trust Centre for Cell Matrix Research (C.M.K., C.A.S., M.J.H., A.E.C.), University of Manchester; and Faculty of Medicine and Human Sciences (A.E.C.), University of Manchester; Manchester, United Kingdom.
Correspondence to Dr Ann Canfield, Michael Smith Building, Oxford Rd, University of Manchester, Manchester, M13 9PT, United Kingdom. E-mail ann.canfield{at}manchester.ac.uk
Received October 14, 2005; de novo received April 21, 2006; revision received June 8, 2006; accepted June 22, 2006.
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionConclusionsReferences |
|---|
Background— Essential to tissue-engineered vascular grafts is the formation of a functional endothelial monolayer capable of resisting the forces of blood flow. This study targeted
2(VIII) collagen, a major component of the subendothelial matrix, and examined the ability of and mechanisms by which endothelial cells attach to this collagen under static and dynamic conditions both in vitro and in vivo.
Methods and Results— Attachment of human endothelial cells to recombinant 2(VIII) collagen was assessed in vitro under static and shear conditions of up to 100 dyne/cm2. 2(VIII) collagen supported endothelial cell attachment in a dose-dependent manner, with an 18-fold higher affinity for endothelial cells compared with fibronectin. Cell attachment was significantly inhibited by function-blocking anti-2 (56%) and -ß1 (98%) integrin antibodies but was not RGD dependent. Under flow, endothelial cells were retained at significantly higher levels on 2(VIII) collagen (53% and 51%) than either fibronectin (23% and 16%) or glass substrata (7% and 1%) at shear rates of 30 and 60 dyne/cm2, respectively. In vivo studies, using endothelialized polyurethane grafts, demonstrated significantly higher cell retention rates to 2(VIII) collagen-coated than to fibronectin-coated prostheses in the midgraft area (P<0.05) after 24 hours’ implantation in the caprine carotid artery.
Conclusions— These studies demonstrate that 2(VIII) collagen has the potential to improve both initial cell attachment and retention of endothelial cells on vascular grafts in vivo, which opens new avenues of research into the development of single-stage endothelialized prostheses and the next generation of tissue-engineered vascular grafts.
Key Words: collagen • endothelium • cell adhesion molecules • integrins • vascular grafts
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionsReferences |
|---|
Development of a successful tissue-engineered vascular graft requires the formation of a functional endothelial monolayer that is capable of resisting the forces of blood flow. A major limitation of current grafts is the low number of endothelial cells (ECs), and the proportion of the endothelialized surface, remaining after exposure to fluid shear stresses representative of blood flow. In vitro, exposure to chronic shear stress most accurately replicates the conditions in vivo and can stimulate EC adhesion to the substratum.1 However, of particular importance from a tissue-engineering perspective is the requirement for ECs to resist the sudden, acute period of high flow rate associated with the resumption of blood flow. It has previously been demonstrated in vitro that up to 70% of ECs coating a vascular graft can be lost after the initiation of flow.2,3 Therefore, improving the resistance of ECs to acute shear stress potentially has significant benefits in promoting the long-term patency and success of tissue-engineered vascular grafts.
Clinical Perspective p 829
Cell adhesion to biomaterial surfaces can be enhanced either by modifying surface properties or by altering surface architecture through different manufacturing techniques.4,5 However, one of the most widely used and simplest techniques is surface coating with extracellular matrix (ECM) components.6–9 Type VIII collagen is a short-chain, network-forming collagen originally identified as a product of ECs10,11 and is thought to play a key structural role in the vasculature, where it is a major component of the subendothelial matrix and the tunica media.12 Type VIII collagen comprises 2 distinct polypeptide chains designated 1(VIII) and 2(VIII). These chains can exist as both heterotrimers and homotrimers, forming a nonfibrillar collagen with a short triple-helical domain flanked by noncollagenous C-terminal NC1 and N-terminal NC2 domains.11,13 In vitro studies have suggested an involvement of type VIII collagen in the process of EC differentiation and organization and in angiogenesis,14,15 although this role has been questioned.16
Interestingly, type VIII collagen may be antithrombogenic, because compared with collagen types I and III, it only weakly supports platelet adhesion,17 which suggests that this collagen has potential as a nonthrombogenic substrate to support EC attachment to vascular prostheses. However, its use as an attachment factor for ECs either in vitro or in vivo has not been investigated. The present study has investigated the ability of 2(VIII) collagen homotrimers to act as an attachment factor for ECs and has examined the retention of ECs to this substratum under defined acute shear stress in vitro, and to 2(VIII) collagen-coated prostheses in vivo.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionsReferences |
|---|
Cell Culture
Human ECs were obtained from saphenous vein and cultured as described previously.18 Caprine aortic ECs were sourced essentially as described by Jaffe et al19 and cultured as for human ECs. Cells were characterized as endothelial based on their positive immunoreaction to antibodies against CD31 and von Willebrand factor and were used between passages 4 and 6.
Transfected 293-EBNA cells were maintained in serum-free Ex-Cell HEK 293 medium (JRH Biosciences, Lenexa, Ks) supplemented with 2 mmol/L L-glutamine, 50 µg/mL ascorbic acid-2-phosphate (Acros Chemicals, Geel, Belgium), 0.5 µg/mL puromycin (Invitrogen, Paisley, United Kingdom), and 500 µg/mL Geneticin (Calbiochem, Lutterworth, United Kingdom). All cells were maintained in a humidified incubator at 37°C in an atmosphere of 5% CO2.
Production of Recombinant 2(VIII) Collagen
Recombinant 2(VIII) collagen was produced with a mammalian episomal expression system in 293-EBNA cells and characterized as detailed in the online Data Supplement.
Cell-Attachment Assays
Cell-attachment assays were performed in 96-well plates, in triplicate, with a dye-staining method as described previously.20 Wells were coated with recombinant 2(VIII) collagen or empty vector samples (0 to 0.5 µg/mL), with human fibronectin (0 to 10 µg/mL, Sigma-Aldrich, Gillingham, United Kingdom) used as a positive control. In all experiments, ECs were plated at 5x104 cells/well for 40 minutes.
To identify integrins involved in EC adhesion, cells were preincubated with 10 µg/mL function-blocking or non–function-blocking antibodies for 20 minutes. Antibodies against the following integrin subunits were used: mouse anti-human 2 (clones JA218 and 10A4),21 mouse anti-human 3 (clones P1B5, Chemicon, Temecula, Calif, and IA3, R&D Systems, Abingdon, United Kingdom), rat anti-human 5 (MAb16 and MAb11),22,23 rat anti-human ß1 (MAb13),22 and mouse anti-human ß1 (K20, Dako Cytomation, Ely, United Kingdom). To confirm the results of the integrin inhibition assay, cells were preincubated with cyclic arginine-glycine-aspartic acid (RGD)-blocking peptides [K(aha)CGRGDSP] for 20 minutes before plating. For RGD studies, poly-L-lysine (500 µg/mL, Sigma)–coated wells were used to identify nonspecific interactions. For both of these studies, cell attachment was assessed on 0.3 µg/mL 2(VIII) collagen as described above, with Pronectin F Plus (10 µg/mL, Sanyo Chemical Industries, Kyoto, Japan), containing only the RGD domain of fibronectin, used as a positive control.
Cell-Spreading Assay
Cell spreading was assessed on 2(VIII) collagen (0.3 µg/mL), fibronectin (10 µg/mL), and glass substrata, with the surfaces prepared as for the cell-attachment assay. ECs were plated at 5x105 cells/cm2 for 1 hour. Cells were classified as being spread if the nucleus and cytoplasmic extensions were distinguishable. For each substratum, 10 images (920x480 µm) were captured, and the percentage of spread cells per image was calculated.
EC Retention Under Physiological Shear Stress
A parallel-plate flow system (30x10x0.127 mm, GlycoTech, Rockville, Md) was used to assess EC retention to glass, 2(VIII) collagen (0.3 µg/mL), and fibronectin (10 µg/mL) substrata under shear stress. After a 1-hour incubation with ECs (5x105 cells/cm2), the flow system was run for 5 minutes at flow rates calculated to produce shear stresses of 0, 30, 60, and 100 dyne/cm2. Calculation of Reynolds number confirmed laminar flow at all flow rates.
For each sample, 10 fields of view were captured. Cell counts were performed with the UTHSCSA ImageTool program (developed at the University of Texas Health Science Center at San Antonio, Tex) and the cells/cm2 calculated.
Preparation of EC-Seeded Polyurethane Vascular Grafts
Polyurethane grafts (5 cmx4.5 mm) were prepared by electrospinning as detailed in the online Data Supplement. Grafts were precoated with either 2(VIII) collagen (0.3 µg/mL) or fibronectin (10 µg/mL) for 1 hour with constant rotation (0.1 rpm). Uncoated prostheses were included as controls. ECs, prelabeled with 5 µmol/L green fluorescent tracker dye (Invitrogen, Paisley, United Kingdom), were seeded onto the grafts at 1.5x105 cells/cm2 in 1 mL of Hanks’ medium 199. The grafts were placed within a rotation device (Endostrabilisator, Beigler, Mauerbach, Austria) for 4 hours at 0.1 rpm.
Analysis of EC Retention In Vivo
All procedures were performed with the authorization of the Home Office and complied with the Animals (Scientific Procedures) Act of 1986. Five female adult goats (weight 65 to 95 kg) were used. The animals were allowed to acclimatize for at least 1 week before undergoing surgery and were fed normal laboratory chow for the duration of the experiment.
All goats were premedicated with 150 mg of aspirin for 48 hours before surgery and given nothing by mouth overnight. Induction of anesthesia was achieved with inhaled 2% isoflurane, nitrous oxide (2 L/min), and oxygen (2 L/min). No paralyzing agents were used, and the anesthesia used allowed spontaneous respiration. After intravenous injection of midazolam, titrated to abolish laryngeal spasm, a size 9 endotracheal tube was sited down to the midtrachea and secured. Heart rate and oxygen saturation were monitored constantly.
After aseptic preparation of the incision site and intravenous injection of 1 g of flucloxacillin, a midline pretracheal incision was made. A substernomastoid plane was developed to reveal both common carotid arteries. After exposure of the artery, 2000 U of heparin (Leo, Princes Risborough, United Kingdom) was administered intravenously, with 2 minutes allowed for systemic distribution. Each artery was transected transversely between clamps, and an endothelialized graft [4 2(VIII) collagen, 3 fibronectin, and 3 uncoated] was inserted end-to-end with continuous 6.0 Prolene suture (Ethicon Limited, Edinburgh, United Kingdom). Analysis of 5-mm portions from the proximal end of every graft confirmed the presence of a confluent EC monolayer before implantation. After clamp release and securing of hemostasis, the wound was closed in a routine fashion.
The animals were allowed to recover and given postoperative analgesia as required. After 24 hours, each animal was humanely killed with intravenous sodium phenobarbital, after which each graft was explanted and fixed in 2.5% (vol/vol) glutaraldehyde in PBS.
Grafts were divided transversely into 1.5-cm sections to allow the proximal, mid, and distal areas to be analyzed separately. These sections were cut longitudinally into 4.5-mm-wide strips for microscopy. Each strip was viewed at x10 magnification with a Leica SP2 AOBS confocal microscope (Leica, Wetzlar, Germany). For each strip, 3 random fields (1.5x1.5 mm) were imaged, and fluorescently labeled cells were counted with the UTHSCA ImageTool program.
Statistical Analysis
All image analysis was performed blinded to the substratum. In vitro data are presented as mean±SD. In vivo data are presented as mean±SEM. ANOVA, Dunnett, and Student t tests were performed with SPSS software (SPSS Inc, Chicago, Ill) to determine statistical significance.
The authors had full access to the data and take full responsibility for its integrity. All authors have read and agree to the manuscript as written.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionsReferences |
|---|
Expression of Recombinant
2(VIII) Collagen HomotrimersPlasmids containing cDNA encoding the
2(VIII) collagen chain were transfected into 293-EBNA cells for recombinant expression. Cells transfected with the pCEP-His vector alone were generated as controls. After nickel affinity chromatography was performed, fractions were analyzed for type VIII collagen by SDS-PAGE and Western blotting. Immunoreactive bands were identified in fractions 7 to 18 corresponding to 2(VIII) collagen (Figure 1A). Fractions 8 to 14, with the highest concentration of 2(VIII) collagen, and the corresponding fractions from empty vector controls were pooled and used for all subsequent experiments. Total 2(VIII) collagen yield was calculated at &75 ng/mL culture medium.
|
2(VIII) collagen samples were subjected to pepsin digestion at 4°C and 37°C to investigate helical stability. At both temperatures, pepsin digestion resulted in the production of 8 distinct bands, ranging in size from 250 to 40 kDa, that were not present in undigested samples (Figure 1B; compare 3 and 4 with 1 and 2). Collagenase digestion of the pepsin-digested samples confirmed that all the bands except for 1 at &40 kDa (Figure 1B, track 6) were collagenous, which suggests that the 2(VIII) chains did assemble to form a helical collagen molecule that was stable at physiological temperatures.
2(VIII) Collagen Supports EC Attachment
Cell-attachment assays demonstrated that both 2(VIII) collagen and fibronectin supported EC attachment in a dose-dependent manner (Figure 2A). 2(VIII) collagen significantly increased EC attachment compared with both baseline levels and fibronectin at concentrations of 
0.005 µg/mL (P<0.05), whereas a significant increase above baseline attachment was only observed at concentrations 0.1 µg/mL for fibronectin (P<0.05). Half-maximal attachment was achieved with a concentration of 0.43 µg/mL for fibronectin but only 0.024 µg/mL for 2(VIII) collagen. Cell attachment to the empty vector controls was not significantly different from untreated, BSA-blocked controls (data not shown).
|
P<0.05). n=20.
2(VIII) Collagen Enhances Retention of ECs Under Shear Stress Flow
Having demonstrated that human ECs attached to 2(VIII) collagen under static conditions, the retention of ECs on this substratum after exposure to acute shear stress was evaluated. Cell counts performed on slides that had not been exposed to flow (0 dyne/cm2) showed that there were significant differences in initial cell attachment to the 3 substrata. Cell attachment to untreated glass slides was low, with 1.6x105±7.4x104 cells/cm2 attached compared with 4.2x105±6.5x104 cells/cm2 for fibronectin and 4.8x105±5.3x104 cells/cm2 for 2(VIII) collagen (P<0.0001). In addition, cell spreading was lowest on glass substrata, with only 13±10% of the attached cells identified as spread (Figure 3A). Statistical analysis demonstrated this to be significantly lower than for either fibronectin or 2(VIII) collagen (P<0.001 for both). There was no significant difference in the percentage of spread cells between fibronectin and 2(VIII) collagen substrata, with 46±14% and 41±5% of cells spread, respectively (Figures 3B and 3C). Therefore, subsequent statistical analysis of cell retention was performed on data normalized to the initial cell attachment numbers to give values of cell retention as a percentage of initial attachment.
|
Under flow, cell retention on glass substratum was low, with only 7±8% of cells remaining after exposure to 30 dyne/cm2 flow. At 60 dyne/cm2, this had decreased to 1±2%, with 0.4±1% of cells remaining after exposure to 100 dyne/cm2 flow (Figure 2B). In contrast, 23±16% of ECs remained attached to the fibronectin substratum after exposure to 30 dyne/cm2 flow, decreasing to 16±11% at 60 dyne/cm2. After exposure to 100 dyne/cm2 flow, cell numbers decreased slightly to 14±10% (Figure 2B). On the 2(VIII) collagen substratum, 53±26% of cells were retained after exposure to 30 dyne/cm2 flow, decreasing slightly to 51±27% after exposure to 60 dyne/cm2 flow. However, only 9±3% of the total cells were retained after 100 dyne/cm2 flow (Figure 2B).
ANOVA and Dunnett’s t tests demonstrated that a significant loss of cells occurred on all substrata after initial exposure to flow at 30 dyne/cm2 (P<0.001). However, cell retention on the 2(VIII) collagen and fibronectin substrata was significantly higher at all flow rates than on the glass substratum [P<0.001 for 2(VIII) collagen, P<0.05 for fibronectin]. EC retention to 2(VIII) collagen was significantly higher than to fibronectin at shear stresses of 30 and 60 dyne/cm2 (P<0.001 for both).
EC Attachment to 2(VIII) Collagen Is Mediated by Integrin 2ß1
On 2(VIII) collagen substrata, function-blocking anti-2 (JA218) and anti-ß1 (MAb13) antibodies significantly inhibited EC attachment by 56±6% and 98±0.4%, respectively (P<0.001 for both), with non–function-blocking antibodies (10A4 and K20) producing no significant difference in cell attachment compared with controls (Figures 4A and 4B). Incubation with both function-blocking anti-3 (P1B5) and anti-5 (MAb16) antibodies resulted in a small but statistically significant reduction in attachment; however, in both cases, this was not significantly different to the non–function-blocking controls.
|
In control experiments with Pronectin F Plus, EC attachment was significantly inhibited by the anti-ß1 (98±0.3% reduction, P<0.001) and anti-5 (92±3% reduction, P<0.001) function-blocking antibodies (Figures 4B and 4D). In both cases, no significant difference in attachment was seen with the non–function-blocking antibody compared with untreated controls. Furthermore, no significant difference in cell attachment was seen with either function-blocking or non–function-blocking anti-2 or anti-3 antibodies (Figures 4A and 4C).
Addition of cyclic RGD blocking peptides did not inhibit EC attachment to 2(VIII) collagen substratum (Figure 5); however, a significant reduction in EC attachment to the Pronectin F Plus control (P<0.0001) was observed. This inhibition was dose dependent, with the half-maximal response achieved at 29 µg/mL. Addition of RGD peptides did not significantly inhibit EC attachment to poly-L-lysine substratum.
|
2(VIII) Collagen Enhances EC Retention to Polyurethane Prostheses In Vivo
Image analysis of control prostheses, which had not been implanted, confirmed the formation of a confluent EC monolayer along the entire length of every graft (Figures 6A through 6C). There was no significant difference in overall cell density between the 3 surface treatments, with average cell counts of 2343±20.0, 2335±54.7, and 2303±23.11 cells/field for 2(VIII) collagen, fibronectin, and uncoated grafts, respectively. In addition, no significant difference was observed when proximal, middle, and distal portions within each graft type were compared (Figure 6D), which demonstrates that the seeding process produced a reproducible, uniform EC coating.
|
All grafts, except 1, were implanted without complication. One graft, an 2(VIII) collagen–coated prosthesis, thrombosed above the distal clamp, which necessitated graft exploration and thrombectomy to reestablish blood flow. Because this graft had not received the same handling as the other grafts, it was excluded from analysis. This procedure did not affect flow through the contralateral graft.
On explant, all grafts showed formation of white thrombus at the distal end corresponding to a 2-mm section of uncoated, unseeded polyurethane caused by the luer-lock connector used in the seeding process; however, this did not occlude blood flow. In all of the uncoated prostheses, this white thrombus also resulted in the formation of red thrombus that extended 1 to 2 cm along the prosthesis, occluding the graft. Red thrombus was additionally observed in 1 2(VIII) collagen–coated and 1 fibronectin-coated graft, both of which had been implanted in the same animal.
2(VIII) collagen–coated and fibronectin-coated grafts demonstrated similar levels of cell retention, with 50.0±10.5% and 41.7±1.6% of ECs remaining after implantation, respectively. EC retention to 2(VIII) collagen and fibronectin coatings was significantly higher than to uncoated prostheses (P<0.001 for both), which exhibited an average retention of only 8.6%±4.8% cells. When the proximal, middle, and distal portions were considered separately (Figure 7A), midgraft endothelialization, a crucial factor for graft success, was significantly greater on 2(VIII) collagen–coated than on fibronectin-coated grafts (1438.3±179.32 compared with 997.9±112.0 cells/field, respectively; P<0.05). No significant difference was observed in cell retention between 2(VIII) collagen– and fibronectin-coated grafts in either the proximal (1173.2±143.4 and 978.7±159.2 cells/field, respectively) or distal (911.0±115.8 and 901.1±125.6 cells/field, respectively) areas. In all cases, EC retention to 2(VIII) collagen and fibronectin coatings was significantly higher than to uncoated prostheses. Interestingly, although &50% of cells had been lost from the 2(VIII) collagen–coated grafts, these prostheses still retained a confluent monolayer of cells over the majority of the surface, with the cells spreading to fill in areas of cell loss (Figure 7B). This also occurred to some extent on fibronectin-coated prostheses, but the monolayer was only 80% to 90% confluent (Figure 7C).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionsReferences |
|---|
This study conclusively proves that
2(VIII) collagen significantly enhances EC adhesion, demonstrating an 18-fold increase in the affinity of ECs for 2(VIII) collagen compared with fibronectin. In addition, under dynamic conditions both in vitro and in vivo, 2(VIII) collagen significantly enhances EC retention compared with fibronectin. Furthermore, this study demonstrates for the first time that this attachment is mediated by the integrin 2ß1.
Acute shear stress is a major problem for the development of tissue-engineered vascular grafts. Normal blood flow in peripheral arteries produces a mean shear stress of &15 dyne/cm2. However, the resumption of blood flow to a vessel causes a sudden, short exposure of the EC monolayer to high flow rates and therefore high shear stresses. The present study has demonstrated that in vitro, 50% of the seeded cells were still present on the 2(VIII) collagen substratum after flow at 60 dyne/cm2, a major increase compared with other substrata. Furthermore, analysis of the in vivo data demonstrated similar levels of EC retention to 2(VIII) collagen–coated polyurethane grafts compared with the in vitro flow studies, which suggests good synergy between these in vitro and in vivo models of acute shear stress.
The in vitro model aimed to assess retention of newly adhered cells. Thus, although a confluent monolayer was not achieved, this assay gives an accurate measure of cell retention purely due to attachment strength and not influenced by cell spreading. Indeed, there was no significant difference in EC spreading between 2(VIII) collagen and fibronectin, which indicates increased adhesion strength was the key factor for increased EC retention under flow. The adult caprine carotid model was used in the present study because it simulates the hemodynamic forces encountered in the human superficial femoral and popliteal arteries.24 An added advantage is that both carotid arteries are easily accessible, providing an intraexperiment control and allowing all combinations of graft and position to be included, which minimizes experimental variability.
The use of polyurethane as a vascular graft has seen a resurgence due to the development of novel polyurethane compounds with stable hemodynamic and degradation properties. However, controversy still exists over the high thrombogenicity of the luminal surface. Although considered less thrombogenic than either Dacron or expanded polytetrafluoroethylene (ePTFE),25 many studies have shown high levels of thrombosis with polyurethane both in vitro and in human trials.26–28 Indeed, in the present study, all uncoated, endothelialized prostheses occluded through the development of red thrombus. In addition, all implanted grafts showed formation of white thrombus on the uncoated, unseeded portion of the graft located at the distal anastomosis, which is indicative of high blood flow rates.
The present in vivo data showed &50% of cells were retained on both 2(VIII) collagen–coated and fibronectin-coated grafts, which is comparable to other recently published studies using treated ePTFE or polyurethane.29–31 In contrast, only 8% of cells were retained on uncoated grafts. In addition, using fluorescently labeled ECs, we were able to evaluate retention of ECs in the proximal, mid, and distal positions by immunofluorescence, which eliminated the possibility of migrating host ECs skewing our data. Whereas other studies have used immunofluorescence to evaluate EC coverage,32 to the best of our knowledge, the present study is unique in using this method to evaluate EC retention after in vivo implantation. The present study demonstrated significantly enhanced EC retention to 2(VIII) collagen–coated grafts in the midgraft area compared with other substrata. This is of particular importance for graft success, because in humans, only a small area around the anastomoses naturally endothelializes via host ECs,33 which leaves the midgraft area devoid of an intact EC monolayer, increasing the risk of thrombosis.
The 2(VIII) collagen produced in the present study was thermally stable, resistant to SDS, and formed homotrimers of the predicted size.34 Pepsin digests demonstrated that the recombinant 2(VIII) collagen was stable under physiological conditions, with a banding pattern that was similar to that seen in pepsin-digested type VIII collagen from smooth muscle cells.34 Collagenase digestion demonstrated that all these bands were collagenous except for a band at &40 kDa. It has been shown previously that the NC1 domain of 1(VIII) collagen is collagenase resistant and can form an aggregate of &50 kDa.35 Therefore, it is possible that this 40-kDa band is an aggregate of the NC1 domains of the 2(VIII) collagen.
Cell attachment to ECM ligands is mediated mainly by integrins. Of the 24 known integrin combinations, 4 function primarily as collagen receptors, namely, 1ß1, 2ß1, 10ß1, and 11ß1.36 ECs express at least 13 different integrins,37 with human saphenous vein cells, used in the present study, known to express combinations of the 2, 3, 5, v, and ß1 integrins.38 The present study demonstrated that antibodies directed against both the 2 and ß1 integrin significantly inhibited EC attachment to 2(VIII) collagen. The level of inhibition by the anti-2 function-blocking antibody was consistent with its allosteric mode of cell adhesion. It has recently been demonstrated that the 2ß1 integrin is responsible for attachment of chondrocytes to type X collagen, which shares considerable homology with type VIII collagen.39 Furthermore, analysis of smooth muscle cell migration on type VIII collagen also demonstrated the involvement of the 2ß1 integrin in this process.40
The use of RGD blocking peptides showed that although the 2(VIII) collagen contained 3 RGD domains within its triple-helix region, ECs did not attach to the 2(VIII) collagen in an RGD-dependent manner. This suggests that the RGD-binding domains are buried within the collagen triple helix and unavailable for cell attachment, or their conformation is inappropriate for binding the integrin. These data are consistent with the identification of 2ß1 as mediator of EC attachment to 2(VIII) collagen. Indeed, analysis of the amino acid sequence has identified a GLOGER motif within the 2(VIII) chain, previously shown to be a recognition motif for 2ß1 integrin.41
|
Conclusions |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionsReferences |
|---|
Historically, fibronectin and fibrin have been considered as the ECM coatings of choice when attempting to endothelialize vascular prostheses.9 The present study demonstrates that
2(VIII) collagen is a potent attachment factor for ECs, significantly enhancing EC retention both in vitro and in vivo. In addition, we have identified that EC attachment to recombinant 2(VIII) substrata occurs via the 2ß1 integrin. These data suggest that recombinant 2(VIII) collagen has significant potential to improve both initial cell attachment and retention of ECs to prosthetic vascular grafts and open new avenues of research in the development of both single-stage EC-seeded prostheses and the next generation of tissue-engineered vascular grafts. |
Acknowledgments |
|---|
The help of Sue Craig in optimizing the cell-attachment assays and Simon Stephan for advice on
2(VIII) collagen is gratefully acknowledged. Sources of Funding
Funding was from the UK Centre for Tissue Engineering (Biotechnology and Biological Sciences Research Council, Medical Research Council and Engineering and Physical Sciences Research Council) and the Wellcome Trust.
Disclosures
None.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionsReferences |
|---|
1. Ballermann BJ, Dardik A, Eng E, Liu A. Shear stress and the endothelium. Kidney Int. 1998; 54: s100–s108.[CrossRef]
2. Rosenman JE, Kempczinski RF, Pearce WH, Silberstein EB. Kinetics of endothelial cell seeding. J Vasc Surg. 1985; 2: 778–784.[CrossRef][Medline] [Order article via Infotrieve]
3. Baguneid M, Murray D, Salacinski HJ, Fuller B, Hamilton G, Walker M, Seifalian AM. Shear-stress preconditioning and tissue-engineering-based paradigms for generating arterial substitutes. Biotechnol Appl Biochem. 2004; 39 (pt 2): 151–157.[CrossRef][Medline] [Order article via Infotrieve]
4. Pu FR, Williams RL, Markkula TK, Hunt JA. Effects of plasma treated PET and PTFE on expression of adhesion molecules by human endothelial cells in vitro. Biomaterials. 2002; 23: 2411–2428.[CrossRef][Medline] [Order article via Infotrieve]
5. Miller DC, Webster TJ, Haberstroh KM. Technological advances in nanoscale biomaterials: the future of vascular graft design. Expert Rev Med Devices. 2004; 1: 259–268.[CrossRef][Medline] [Order article via Infotrieve]
6. Meinhart JG, Deutsch M, Fischlein T, Howanietz N, Froschl A, Zilla P. Clinical autologous in vitro endothelialization of 153 infrainguinal ePTFE grafts. Ann Thorac Surg. 2001; 71 (suppl): S327–S331.[CrossRef][Medline] [Order article via Infotrieve]
7. Vohra R, Thomson GJ, Carr HM, Sharma H, Walker MG. Comparison of different vascular prostheses and matrices in relation to endothelial seeding. Br J Surg. 1991; 78: 417–420.[Medline] [Order article via Infotrieve]
8. Desgranges P, Tardieu M, Loisance D, Barritault D. Extracellular matrix covered biomaterials for human endothelial cell growth. Int J Artif Organs. 1992; 15: 722–726.[Medline] [Order article via Infotrieve]
9. Bujan J, Garcia-Honduvilla N, Bellon JM. Engineering conduits to resemble natural vascular tissue. Biotechnol Appl Biochem. 2004; 39 (pt 1): 17–27.[CrossRef][Medline] [Order article via Infotrieve]
10. Sage H, Pritzl P, Bornstein P. A unique, pepsin-sensitive collagen synthesized by aortic endothelial cells in culture. Biochemistry. 1980; 19: 5747–5755.[CrossRef][Medline] [Order article via Infotrieve]
11. Shuttleworth CA. Type VIII collagen. Int J Biochem Cell Biol. 1997; 29: 1145–1148.[CrossRef][Medline] [Order article via Infotrieve]
12. Kittelberger R, Davis PF, Greenhill NS. Immunolocalization of type VIII collagen in vascular tissue. Biochem Biophys Res Commun. 1989; 159: 414–419.[CrossRef][Medline] [Order article via Infotrieve]
13. Greenhill NS, Ruger BM, Hasan Q, Davis PF. The alpha1(VIII) and alpha2(VIII) collagen chains form two distinct homotrimeric proteins in vivo. Matrix Biol. 2000; 19: 19–28.[CrossRef][Medline] [Order article via Infotrieve]
14. Sage H, Iruela-Arispe ML. Type VIII collagen in murine development: association with capillary formation in vitro. Ann N Y Acad Sci. 1990; 580: 17–31.[Medline] [Order article via Infotrieve]
15. Iruela-Arispe ML, Sage EH. Expression of type VIII collagen during morphogenesis of the chicken and mouse heart. Dev Biol. 1991; 144: 107–118.[CrossRef][Medline] [Order article via Infotrieve]
16. Sinha S, Kielty CM, Heagerty AM, Canfield AE, Shuttleworth CA. Upregulation of collagen VIII following porcine coronary artery angioplasty is related to smooth muscle cell migration not angiogenesis. Int J Exp Pathol. 2001; 82: 295–302.[CrossRef][Medline] [Order article via Infotrieve]
17. Saelman EU, Nieuwenhuis HK, Hese KM, de Groot PG, Heijnen HF, Sage EH, Williams S, McKeown L, Gralnick HR, Sixma JJ. Platelet adhesion to collagen types I through VIII under conditions of stasis and flow is mediated by GPIa/IIa (alpha 2 beta 1-integrin). Blood. 1994; 83: 1244–1250.
18. Turner NJ, Kielty CM, Walker MG, Canfield AE. A novel hyaluronan-based biomaterial (Hyaff-11) as a scaffold for endothelial cells in tissue engineered vascular grafts. Biomaterials. 2004; 25: 5955–5964.[CrossRef][Medline] [Order article via Infotrieve]
19. Jaffe EA, Nachman RL, Becker CG, Minick CR. Culture of human endothelial cells derived from umbilical veins: identification by morphologic and immunologic criteria. J Clin Invest. 1973; 52: 2745–2756.[Medline] [Order article via Infotrieve]
20. Humphries MJ. Cell adhesion assays. Mol Biotechnol. 2001; 18: 57–61.[CrossRef][Medline] [Order article via Infotrieve]
21. Tuckwell DS, Smith L, Korda M, Askari JA, Santoso S, Barnes MJ, Farndale RW, Humphries MJ. Monoclonal antibodies identify residues 199–216 of the integrin alpha2 vWFA domain as a functionally important region within alpha2beta1. Biochem J. 2000; 350: 485–493.[CrossRef][Medline] [Order article via Infotrieve]
22. Akiyama SK, Yamada SS, Chen WT, Yamada KM. Analysis of fibronectin receptor function with monoclonal antibodies: roles in cell adhesion, migration, matrix assembly, and cytoskeletal organization. J Cell Biol. 1989; 109: 863–875.
23. Miyamoto S, Akiyama SK, Yamada KM. Synergistic roles for receptor occupancy and aggregation in integrin transmembrane function. Science. 1995; 267: 883–885.
24. Rashid ST, Salacinski HJ, Hamilton G, Seifalian AM. The use of animal models in developing the discipline of cardiovascular tissue engineering: a review. Biomaterials. 2004; 25: 1627–1637.[CrossRef][Medline] [Order article via Infotrieve]
25. Kannan RY, Salacinski HJ, Butler PE, Hamilton G, Seifalian AM. Current status of prosthetic bypass grafts: a review. J Biomed Mater Res B Appl Biomater. 2005; 74: 570–581.[Medline] [Order article via Infotrieve]
26. Bernacca GM, Gulbransen MJ, Wilkinson R, Wheatley DJ. In vitro blood compatibility of surface-modified polyurethanes. Biomaterials. 1998; 19: 1151–1165.[CrossRef][Medline] [Order article via Infotrieve]
27. Aldenhoff YB, Koole LH. Platelet adhesion studies on dipyridamole coated polyurethane surfaces. Eur Cell Mater. 2003; 5: 61–67.[Medline] [Order article via Infotrieve]
28. Ben Hamida F, M’hibik S, Karoui C, Abderrahim E, Kaaroud H, Beji S, Zouaghi K, Barbouche S, Goucha R, Hedri H, El Younsi F, Ben Abdallah T, Ben Moussa F, Kheder A, Ben Maiz H. Indications, complications and cost of internal jugular catheters in hemodialysed patients: study of 533 cases [in French]. Tunis Med. 2005; 83: 519–523.[Medline] [Order article via Infotrieve]
29. Chan BP, Liu W, Klitzman B, Reichert WM, Truskey GA. In vivo performance of dual ligand augmented endothelialized expanded polytetrafluoroethylene vascular grafts. J Biomed Mater Res B Appl Biomater. 2005; 72: 52–63.[CrossRef][Medline] [Order article via Infotrieve]
30. Tiwari A, Salacinski HJ, Punshon G, Hamilton G, Seifalian AM. Development of a hybrid cardiovascular graft using a tissue engineering approach. FASEB J. 2002; 16: 791–796.
31. Jun HW, West J. Development of a YIGSR-peptide-modified polyurethaneurea to enhance endothelialization. J Biomater Sci Polym Ed. 2004; 15: 73–94.[CrossRef][Medline] [Order article via Infotrieve]
32. Fernandez P, Deguette A, Pothuaud L, Belleannee G, Coste P, Bordenave L. Quality control assessment of ePTFE precoating procedure for in vitro endothelial cell seeding. Biomaterials. 2005; 26: 5042–5047.[CrossRef][Medline] [Order article via Infotrieve]
33. Berger K, Sauvage IR, Rao AA, Wood S. Healing of arterial prostheses in man: its incompleteness. Ann Surg. 1972; 175: 119–127.
34. Stephan S, Sherratt MJ, Hodson N, Shuttleworth CA, Kielty CM. Expression and supramolecular assembly of recombinant alpha1(viii) and alpha2(viii) collagen homotrimers. J Biol Chem. 2004; 279: 21469–21477.
35. Illidge C, Kielty C, Shuttleworth A. The alpha1(VIII) and alpha2(VIII) chains of type VIII collagen can form stable homotrimeric molecules. J Biol Chem. 1998; 273: 22091–22095.
36. Tulla M, Pentikainen OT, Viitasalo T, Kapyla J, Impola U, Nykvist P, Nissinen L, Johnson MS, Heino J. Selective binding of collagen subtypes by integrin alpha 1I, alpha 2I, and alpha 10I domains. J Biol Chem. 2001; 276: 48206–48212.
37. Ruegg C, Dormond O, Mariotti A. Endothelial cell integrins and COX-2: mediators and therapeutic targets of tumor angiogenesis. Biochim Biophys Acta. 2004; 1654: 51–67.[Medline] [Order article via Infotrieve]
38. Iivanainen E, Kahari VM, Heino J, Elenius K. Endothelial cell-matrix interactions. Microsc Res Tech. 2003; 60: 13–22.[CrossRef][Medline] [Order article via Infotrieve]
39. Luckman SP, Rees E, Kwan AP. Partial characterization of cell-type X collagen interactions. Biochem J. 2003; 372: 485–493.[CrossRef][Medline] [Order article via Infotrieve]
40. Hou G, Mulholland D, Gronska MA, Bendeck MP. Type VIII collagen stimulates smooth muscle cell migration and matrix metalloproteinase synthesis after arterial injury. Am J Pathol. 2000; 156: 467–476.
41. Siljander PR, Hamaia S, Peachey AR, Slatter DA, Smethurst PA, Ouwehand WH, Knight CG, Farndale RW. Integrin activation state determines selectivity for novel recognition sites in fibrillar collagens. J Biol Chem. 2004; 279: 47763–47772.
|
Footnotes |
|---|
The online-only Data Supplement can be found with this article at http://circ.ahajournals.org/cgi/content/full/CIRCULATIONAHA.106.635292/DC1.
This article has been cited by other articles:
 |
|
 |
|
  E. Adiguzel, P. J Ahmad, C. Franco, and M. P Bendeck Collagens in the progression and complications of atherosclerosis Vascular Medicine, February 1, 2009; 14(1): 73 - 89. [Abstract] [PDF]
|
|
|
|
 |
|
 Q. Lin, D. Gourdon, C. Sun, N. Holten-Andersen, T. H. Anderson, J. H. Waite, and J. N. Israelachvili Adhesion mechanisms of the mussel foot proteins mfp-1 and mfp-3 PNAS, March 6, 2007; 104(10): 3782 - 3786. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||