HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to Personal Folders Download to citation manager Author home page(s): Mika Lahtinen Giedrius Baliulis Vitas Zemgulis Permission Requests Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Lahtinen, M. Articles by Zemgulis, V. Search for Related Content PubMed PubMed Citation Articles by Lahtinen, M. Articles by Zemgulis, V. Related Collections Coronary disease Peripheral vascular

Eur J Cardiothorac Surg 2007;31:383-390. doi:10.1016/j.ejcts.2006.11.048
Copyright © 2007, European Association for Cardio-Thoracic Surgery. Published by Elsevier B.V. All rights reserved

In vivo h-VEGF₁₆₅ gene transfer improves early endothelialisation and patency in synthetic vascular grafts

^a Department of Surgical Sciences, Uppsala University Hospital, Uppsala, Sweden
^b Department of Clinical Chemistry, Uppsala University Hospital, Sweden
^c Clinical Research Center, Karolinska Institutet, Huddinge University Hospital, Huddinge, Sweden
^d Department of Cardiothoracic Surgery, Southampton University Hospitals, Southampton, UK
^e Department of Genetics and Pathology, Uppsala University, Uppsala, Sweden
^f Statistical Consulting Center, Wright State University, Dayton, OH, USA

Received 20 March 2006; received in revised form 12 September 2006; accepted 22 November 2006.

^_* Corresponding author. Address: Department of Clinical Chemistry, Entrance 61, 3rd floor, Uppsala University Hospital, SE-751 85 Uppsala, Sweden. Tel.: +46 708 901090; fax: +46 18 6113703. (Email: mika.lahtinen{at}medsci.uu.se).

	Abstract

Top Abstract 1. Introduction 2. Materials and methods 3. Results 4. Discussion References

 
Objectives: Small-diameter synthetic vascular graft performance is inferior to autologous vein grafts. This study tested the hypotheses that local in vivo administration of plasmids encoding for human vascular endothelial growth factor (VEGF), or co-administration of plasmids encoding for human vascular endothelial growth factor/plasmids encoding for fibroblast growth factor-2 in the tissues surrounding a porous synthetic vascular graft would enhance graft endothelialisation and, consecutively, graft patency. Methods: First, optimal gene for small-diameter synthetic graft endothelialisation was studied in rat abdominal aorta model (n = 132): plasmids encoding for human vascular endothelial growth factor; co-administration of plasmids encoding for human vascular endothelial growth factor/plasmids encoding for fibroblast growth factor-2; or control plasmids were injected around 60 µm ePTFE graft. Second, optimal small-diameter synthetic graft design for endothelialisation was explored in rabbit abdominal aorta model (n = 90). Various ePTFE grafts or pre-clotted polyester grafts were used with/without plasmids encoding for human vascular endothelial growth factor. Third, clinically used medium-size synthetic grafts were investigated with/without plasmids encoding for human vascular endothelial growth factor in dog carotid (n = 20) and femoral arteries (n = 15). Endothelialisation was assessed in midgraft area with scanning electron microscopy. Results: In rats, plasmids encoding for human vascular endothelial growth factor enhanced endothelialisation; whereas co-administration of plasmids encoding for human vascular endothelial growth factor/plasmids encoding for fibroblast growth factor-2 had worst outcome at 1 week (NS), 2 weeks (P = 0.01) and 4 weeks (P = 0.02). In rabbits, pre-clotted polyester grafts had a trend for faster endothelialisation than ePTFE grafts (P = 0.08); whereas plasmids encoding for human vascular endothelial growth factor enhanced endothelialisation compared to controls at 2 weeks (P = 0.06), however, the effect reversed at 4 weeks (P = 0.03). In dogs, synthetic graft patency was improved by plasmids encoding for human vascular endothelial growth factor in femoral position (P = 0.103); whereas all carotid grafts were patent at 6 weeks. Conclusions: Thus, these data suggested that endothelialisation was fastest in pre-clotted polyester grafts; and that local application of plasmids encoding for human vascular endothelial growth factor had a potential to improve early endothelialisation and patency in synthetic vascular grafts.

Key Words: Angiogenesis • Endothelium • Gene therapy • Patency • Prosthesis • VEGF

	1. Introduction

Top Abstract 1. Introduction 2. Materials and methods 3. Results 4. Discussion References

 
Although the first synthetic vascular grafts made of polyester were used in humans some 50 years ago, their performance is inferior to autologous human vein grafts as small-diameter blood conduits (<6 mm) [1]. With biocompatible synthetic vascular grafts capable of replacing use of autologous vein grafts, wound complications associated with leg incisions could be avoided and operation time shortened. However, lack of endothelialisation of foreign surfaces limits synthetic vascular graft biocompatibility. In humans, the synthetic vascular graft endothelialisation is limited to anastomoses, the remaining part being covered by pseudointima; whereas the synthetic surface becomes completely endothelialised in animals.

Synthetic graft endothelialisation could theoretically be accomplished in humans by enhancing longitudinal growth from the anastomotic area (transanastomotic growth); enhancing ingrowth of capillary endothelial cells through the interstices in the synthetic graft wall (transgraft growth); providing endothelialised surface ex vivo by seeding of mature endothelial cells or their precursors; or enhancing in vivo seeding of circulating endothelial precursor cells [2]. Subsequently, to improve transgraft growth in vivo, synthetic graft porosity has been manipulated [3,4] and angiogenic proteins have been administered [5,6]; to cover foreign synthetic surfaces with cellular surface prior to implantation, endothelial cells [1,7], bone marrow [8], adipose tissue [8], or gene-transfected endothelial cells [9,10] and adipose tissue [11] have been seeded and sodded on the foreign surface; and to build complete blood vessels ex vivo consisting of all three cellular layers, tissue engineering has been used [8].

Angiogenic genes and proteins are used experimentally to treat cardiovascular diseases. In humans, genes encoding for vascular endothelial growth factor (VEGF) and fibroblast growth factor (FGF) are administered for peripheral and coronary artery disease [12]. In animal models, naked plasmid DNA encoding for human VEGF (h-VEGF) improves angiogenesis in hind limb, inhibits stent restenosis [13], and seeding of ex vivo VEGF-transfected adipose tissue enhances synthetic graft endothelialisation [11]. The angiogenesis is active in and around synthetic vascular grafts with associated capillary infiltration through synthetic graft interstices to luminal surface not only in animals [14], but also in cyanotic infants [15]; whereas in human adults, the inadequacy of angiogenic capacity is indicated by the unhealed synthetic flow surface.

This study assessed the hypotheses that local in vivo administration of plasmid encoding for h-VEGF₁₆₅ (pNGVL3-VEGF₁₆₅), or co-administration of plasmids encoding for h-VEGF₁₆₅/FGF-2 (pNGVL3-VEGF₁₆₅/pNGVL7-FGF-2) in the tissues surrounding the porous synthetic graft would enhance graft endothelialisation and, consecutively, graft patency. A series of experiments were performed; first, optimal gene for small-diameter synthetic graft endothelialisation, pNGVL3-VEGF₁₆₅ or combination of pNGVL3-VEGF₁₆₅/pNGVL7-FGF-2, was determined in rats; second, optimal small-diameter synthetic graft design to be used in combination with pNGVL3-VEGF₁₆₅ transfection, pre-clotted polyester or various porous ePTFE grafts, was investigated in rabbits; third, clinically used medium-size grafts were investigated in combination with pNGVL3-VEGF₁₆₅ in dogs.

	2. Materials and methods

Top Abstract 1. Introduction 2. Materials and methods 3. Results 4. Discussion References

 
2.1 Manufacture of expression plasmids
2.1.1 pNGVL3-VEGF₁₆₅ expression plasmid
As human VEGF₁₆₅ contains signal sequences for secretion, the coding sequence was directly cloned into pNGVL3 (National Gene Vector Laboratories, The University of Michigan Medical Center, Ann Arbour, MI). h-VEGF₁₆₅ cDNA (Medline accession # M32977) was amplified by PCR. Template cDNA was prepared from total RNA isolated from HEK293 cells. cDNA was synthesized by reverse transcription (RT) and PCR amplified with the sense primer GATCGAATTCGTTAACCA-TGAACTTTCTGCTGTCTTGG containing an EcoRI site and the antisense primer GATCGGATCCGTTAACTCACCGCCTCGGCTTGTCACATC with an engineered BamH1 site. The purified cDNA and pNGVL3 were digested with a combination of EcoRI and BamHI. cDNA was directionally ligated into pNGVL3 (Ligation Express^TM Kit, Clontech, Palo Alto, CA). Chemically competent DH5 E. coli were used for plasmid propagation. Plasmids were purified with QIAprep spin miniprep kit (Qiagen GmbH, Hilden, Germany). The correctness of pNGVL3-VEGF₁₆₅ was verified with ABI automated DNA sequencer. h-VEGF₁₆₅ expression was detected in transfected cells with Western blotting, and chorioallantoic chick membrane assay verified biologic activity (Data not shown).

2.1.2 pNGVL7-FGF-2 expression plasmid
As FGF-2 lacks secretion signal sequence, it was cloned into pNGVL7 (National Gene Vector Laboratories) in frame with the secretion signal sequence for tissue plasminogen activator (tPA). Human FGF-2 cDNA (Genbank accession # NJ04513) was PCR amplified as above. Sense primer ATACTCTAGAATGGCAGCCGGGAGCATCACCACGCTG, containing an XbaI restriction site, was used in combination with the antisense primer GATCAGATCTTCAGCTCTTAGCAGACATTGGAAGAAA containing a BglII restriction site. The product was digested with XbaI and BglII and directionally ligated into the expression vector pNGVL7 (digested with XbaI and BamHI). Correctness of pNGVL7-FGF-2 and h-FGF-2 expression in transfected cells were verified as above (Data not shown).

2.1.3 pNGVL3-EGFPLuc expression plasmid
The EGFPLuc gene from pEGFPLuc (BD, Palo Alto, CA) was inserted into pNGVL3. Plasmid pNGVL3 was linearized by digestion with XbaI and dephosphorylated. pEGFPLuc was propagated in a dam(–)-strain (E. coli S1179) and digested with NheI and XbaI. The linearized vector and the fragment containing the EGFPLuc fragment were purified by gel electrophoresis and eluted with QIAquick spin columns (Qiagen). The purified fragments were ligated and DH5 cells were transformed with the ligation-mixture. Restriction enzyme digestion, together with cell expression studies with fluorescent microscopy (Zeiss Axiovert 200M, Oberkochen, Germany) and luciferase assay (Promega Corp., Madison, WI) verified the correctness of pNGVL3-EGFPLuc (Data not shown).

2.1.4 pNGVL1-nt-ß-gal expression plasmid
The pNGVL1-nt-ß-gal expression plasmid encodes nuclear targeted ß-galactosidase (Data not shown).

2.1.5 Preparation and check of endotoxin-free plasmid DNA
Plasmid DNA was prepared and purified with the EndoFree^TM Plasmid Mega Kit (Qiagen) (h-VEGF₁₆₅, h-FGF-2 and ß-galactosidase) or manufactured (h-VEGF₁₆₅, h-FGF-2 and GFPLuc) by BayouBiolabs (Harahan, LA). Plasmid DNA quality was verified as above (Data not shown).

2.2 In vivo expression and persistence studies
All animals received human care in compliance with European Convention on Animal Care and the studies were approved by institutional ethics committees.

2.2.1 Verification of in vivo gene transfer with RT-PCR
pNGVL3-VEGF₁₆₅ was injected around native abdominal aorta in rat. Anaesthesia was with intraperitoneal mixture of midazolam, fluanisone and fentanyl. After 7 days, the tissue was harvested and PCR performed with the vector-derived sense primer CGCGCGCGCCACCAGACATAATAGCTG based on vector sequence 111 bp upstream of the multiple cloning site and the h-VEGF₁₆₅ specific antisense primer GCAAGTACGTTCGTTTAACTCAAGCTG 21 bp from the carboxyterminal end of the h-VEGF₁₆₅ sequence. The PCR reaction mixture was electrophoresed and cDNA visualised. GAPDH primers served as positive control and cDNA synthesis without addition of RT as negative controls.

2.2.2 Verification of in vivo gene expression with luciferase assay
ePTFE graft piece (2 cm long; 1/4 of the 2 mm graft radius) was inoperated to the periaortic retroperitoneal fat tissue in rat. pNGVL3-EGFPLuc (400 µg) was administered around the graft piece with a syringe (n = 15); controls (n = 11) received either nothing or pNGVL1-nt-ß-gal. Luciferase activity was measured according to previously described methods at 1, 3, 7 and 14 days [16]. The results were expressed as picogram luciferase per milligram protein (pg/mg protein) in the tissue surrounding the graft piece.

2.2.3 Verification of plasmid persistence in the tissue surrounding the foreign material
PCR detection of the pNGVL3-VEGF₁₆₅ and pNGVL7-FGF-2 were performed using above-mentioned periaortic fat tissue model. Samples were explanted at 1, 3, 7, 14 and 28 days.

2.3 In vivo endothelialisation and patency studies
2.3.1 Rat study: effects of pNGVL3-VEGF₁₆₅ and co-administration of pNGVL3-VEGF₁₆₅/pNGVL7-FGF-2 on 60 µm ePTFE graft endothelialisation
2.3.1.1 Groups
Sixty-micrometer internodal distance ePTFE grafts (Impra, Tucson, AZ), previously demonstrated to be optimal for transgraft growth in animals [3,4], were used to select optimal gene for synthetic graft endothelialisation. Male Sprague–Dawley rats (n = 132) were divided in three groups: controls (n = 51); pNGVL3-VEGF₁₆₅ (100–800 µg; n = 66); co-administration of pNGVL3-VEGF₁₆₅/pNGVL7-FGF-2 (400 µg; n = 15). Endothelialisation was analysed with scanning electron microscopy (SEM) at 1 week (6–8 days; n = 30), 2 weeks (13–15 days; n = 47), 4 weeks (28–31 days; n = 41) and 1.5 years (n = 10) (Table 1 ).

2.3.3 Sacrifice and graft explantation
After anaesthesia as above, 500 units of heparin (Lövens, Ballerup, Denmark) were administered i.v. PBS was perfused to the left ventricle while synchronously exsanguinating the rat through right atrium. Then, the rat was perfusion-fixed with 2% paraformaldehyde and 0.2% glutaraldehyde in PBS. The graft was explanted and divided either in five equal parts and mid-part preserved, or, divided in to two halves, and the proximal half preserved for SEM analysis in 2% paraformaldehyde/2% glutaraldehyde in PBS.

2.4 Rabbit study: pNGVL3-VEGF₁₆₅ effects on endothelialisation in various small-diameter grafts
2.4.1 Groups
pNGVL3-VEGF₁₆₅ effects on additional three commercially available small-diameter synthetic graft materials (2 cm x 3 mm) were assessed in rabbits, because these grafts were not available with internal diameter suitable for rat aorta (2 mm). Male New Zealand rabbits (2.5–4 kg) (n = 90) received either knitted polyester (Sulzer Vaskutec, Winterthur, Switzerland) (n = 25), 60 µm ePTFE (Impra) (n = 23), Hybrid ePTFE (Atrium Medical Corporation, Hudson, NH) (n = 33) or 30 µm ePTFE (Impra) (n = 9). Rabbits were divided to controls (n = 44) and pNGVL3-VEGF₁₆₅ (400–600 µg, n = 46). SEM was performed at 2 weeks (n = 32), 4 weeks (n = 24), 12 weeks (n = 18) and 43 weeks (n = 16) (Table 1).

2.4.2 Surgical arterial reconstruction and gene transfer
ASA in drinking water gave an estimated daily dose 10 mg. Anaesthesia was with fentanyl/fluanosine mixture s.c. and midazolam i.m. Antibiotics were as above. The polyester grafts (3 cm x 3 mm), but none of the ePTFE grafts, were pre-clotted during 30 min in unheparinised venous blood and cut to the length of 2 cm. After 500 U of heparin i.v., 20 mL dextran (Mw 70 000; 60 g/L; Pharmacia, Uppsala, Sweden) was infused i.v. over 15 min. Infrarenal aorta was replaced end-to-end with 7-0 sutures. Plasmid administration was as above.

2.4.3 Sacrifice and graft explantation
Anaesthesia was as above. 1000 U heparin were given i.v. Rabbit sacrifice and graft explantation were as above.

2.5 Dog study: pNGVL3-VEGF₁₆₅ effects on clinically used medium-size synthetic grafts
2.5.1 Groups
In contrast to experimental microsurgical grafts in rats and rabbits, clinically used medium-size synthetic grafts were investigated in dogs. Ten beagles (9–13 kg) received two grafts in carotid position (Knitted bovine-collagen-coated polyester (7 cm x 5 mm; n = 10; Intervascular, Cedex, France) or 60 µm ePTFE (7 cm x 4 mm; n = 10; Impra)) and two grafts in femoral position (Hybrid ePTFE (7 cm x 5 mm; n = 15; Atrium)) (Table 1). All grafts in each dog were separately treated with either pNGVL3-VEGF₁₆₅ (2000 µg) or pNGVL3-EGFPLuc (2000 µg). Dogs were sacrificed at 6 weeks (42–52 days).

2.5.2 Surgical arterial reconstruction and gene transfer
Dogs received ASA 325 mg/day from day –4 and onward and dipyridamole 75 mg/day from day –1 and onward. Acepromazin, metadon and atropin were given as sedatives. Anaesthesia was induced with propofol and maintained with isoflurane, N₂O, O₂ and 60% air. 100 U/kg heparin were given i.v. prior to first anastomoses and then 40 U after one hour. Cloxacillin was given i.v. First, bilateral end-to-side anastomoses were created to common carotids with polyester or 60 µm ePTFE grafts using continuous 6-0 sutures. Then, bilateral femoral anastomoses were performed with Hybrid ePTFE grafts. The arterial segment between the anastomoses was ligated and divided in carotids, and ligated but not divided in femorals. The plasmid solution was injected around the graft using a syringe and needle before wound closure in carotids, but after wound closure in femorals. Metadon and carprofen were used as analgesics during 4 days.

2.5.3 Sacrifice and graft explantation
After anaesthesia as above, 3000 U heparin were given i.v. Sternotomy and perfusion fixation were as above. After explantation, the graft was divided in four pieces (part A, B, C and D) and examined for thrombosis. Part A and part C were preserved for SEM analysis.

2.5.4 Graft preparation to SEM and SEM analysis
The fixed grafts were prepared for SEM by washing in PBS, post-fixating in 1% OsO₄, washing in PBS, dehydrating in ethanol, critical point drying in CO₂, mounting on stubs with carbon conducting cement, and finally sputter coating with gold at 22 nm thickness. Images were taken at 25x, 100x, 800x, 1600x and 3200x magnification (Philips XL30 SEM at 10 kV) (FEI Electron Optics, Eindhoven, Netherlands).

The midgraft endothelialisation was graded: unendothelialised (0); partly endothelialised (1); completely endothelialised (2) (Fig. 1 ). The results were expressed as percentage (%) of grafts that were rated as ‘2’ for rats and rabbits, and ‘1’ for dogs. Assessing overall percentage of total surface area covered with cells with SEM at low magnification has an inherent risk of misinterpreting luminal areas covered by inflammatory cells or pseudointima as endothelialised surfaces, or, reporting predominantly transanastomotic growth instead of transgraft growth. Therefore, successful complete midgraft endothelialisation was used as the endpoint.

View larger version (137K): [in this window] [in a new window]   Fig. 1. Scanning electron microscopy (SEM). SEM picture reveals a completely endothelialised surface of ePTFE graft (internodal distance 60 µm).

	3. Results

Top Abstract 1. Introduction 2. Materials and methods 3. Results 4. Discussion References

 
3.1 In vivo expression and persistence studies
In native rat aorta, a fragment with the expected size of 666 bp was amplified after pNGVL3-VEGF₁₆₅ administration; whereas neither negative controls nor pNGVL-ß-gal plasmid gave any PCR product, and amplification with GAPDH primers resulted in a product of the expected size.

In the fat pad model, PCR detection revealed that pNGVL3-VEGF₁₆₅ and pNGVL7-FGF-2 remained in two of four samples at day 28 (data not shown).

In the fat pad model, luciferase remained close to zero at all time points in controls; whereas luciferase values were higher at all times with pNGVL3-EGFPLuc: 70-fold at day 1 (0.14 ± 0.22 vs 0.002 ± 0.002 pg/mg protein); 20-fold at 3 days (0.02 ± 0.02 vs 0.001 ± 0 pg/mg protein); 2-fold at 7 days (0.002 ± 0.0008 vs 0.001 ± 0 pg/mg protein); and 2900-fold at 14 days (5.84 ± 11.7 vs 0.002 ± 0.002 pg/mg protein).

3.2 In vivo endothelialisation and patency studies
3.2.1 Rat study: effects of pNGVL3-VEGF₁₆₅ and co-administration of pNGVL3-VEGF₁₆₅/pNGVL7-FGF-2 on 60 µm ePTFE graft endothelialisation
pNGVL3-VEGF₁₆₅ improved midgraft endothelialisation in rats compared to controls and combination of pNGVL3-VEGF₁₆₅/pNGVL7-FGF-2 (Fig. 2 ).

View larger version (22K): [in this window] [in a new window]   Fig. 2. Effects of pNGVL3-VEGF165 and pNGVL3-VEGF165/pNGVL7-FGF-2 transfection on ePTFE graft endothelialisation in rat aortic grafts. Endothelialisation was enhanced by pNGVL3-VEGF165 transfection compared to controls and co-administration of pNGVL3-VEGF165/pNGVL7-FGF-2: (25% vs 14% vs 0%; P = NS; n = 30)1w; (35% vs 0% vs 0%; P = 0.01; n = 47)2w; and (61% vs 41% vs 0%; P = 0.024; n = 41)4w. At 18 months all pNGVL3-VEGF165 grafts (n = 10) and 75% of controls (n = 4) were endothelialised (P = 0.29). Data shown as % completely endothelialised midgraft area.

View larger version (10K): [in this window] [in a new window]   Fig. 3. Effects of pNGVL3-VEGF165 transfection on endothelialisation in rabbit aortic grafts. pNGVL3-VEGF165 transfection enhanced endothelialisation at 2 weeks compared to controls (53% vs 19%; P = 0.062; n = 32); whereas the relationship reversed at 4 weeks (42% vs 92%; P = 0.03; n = 24). At later time points, there was no difference between the groups ((80% vs 75%; P = NS; n = 18)12w; (75% vs 100%; P = NS; n = 16)43w). Data shown as % completely endothelialised midgraft area.

View larger version (42K): [in this window] [in a new window]   Fig. 4. Association between graft type and endothelialisation in rabbit aortic grafts. Midgraft endothelialisation occurred in all graft types: preclotted polyester grafts at 2 weeks (56%), 4 weeks (100%) and 43 weeks (78%); 60 (m ePTFE grafts at 2 weeks (38%) and 3 months (70%); Hybrid ePTFE grafts at 2 weeks (10%), 4 weeks (50%), 3 months (88%) and at 43 weeks (100%); and 30 (m ePTFE grafts at 4 weeks (56%). Pre-clotted polyester grafts had a trend for faster endothelialisation at the early time points (P 2w = 0.11; n = 32, P 4w = 0.08; n = 24); whereas at the later time points there was no significant difference between the groups (P 12w = NS; n = 18, P 43w = NS; n = 16). Data shown as % completely endothelialised midgraft area.

View larger version (8K): [in this window] [in a new window]   Fig. 5. Effects of pNGVL3-VEGF165 transfection on graft patency in dog. Hybrid ePTFE graft patency was enhanced with pNGVL3-VEGF165 transfection at 6 weeks in femoral position (P = 0.103; n = 15), whereas in carotid position all collagen-coated polyester grafts (n = 10) and 60 µm ePTFE grafts (n = 10) were patent. Data shown as % patent grafts.

	4. Discussion

Top Abstract 1. Introduction 2. Materials and methods 3. Results 4. Discussion References

 
In the present animal experiments the effects of genes encoding for angiogenic growth factors were evaluated in rat, rabbit and dog models by determining patency and endothelialisation of various synthetic vascular graft surfaces with SEM. In contrast to humans, animals have angiogenetic capacity to endothelialise porous synthetic vascular graft spontaneously via transgraft growth mechanism [2]. During an observation period of 1 to 77 weeks, synthetic graft surfaces became completely endothelialised in rats and local treatment with pNGVL3-VEGF₁₆₅ resulted in highest endothelialisation rate at every observed time point compared to control and pNGVL3-VEGF₁₆₅/pNGVL7-FGF-2 plasmids. During an observation period of 2–43 weeks, rabbits completely endothelialised graft surfaces and local treatment with pNGVL3-VEGF₁₆₅ resulted in higher probability of endothelialisation at 2 weeks, whereas the relationship reversed at 4 weeks. There was a trend for improved graft endothelialisation at earlier time points in more porous polyester grafts. During an observation period of 6 weeks in dogs, local treatment of grafts in femoral position with pNGVL3-VEGF₁₆₅ resulted in trend for improved graft patency. Results from the present study reveal a novel concept for enhancement of synthetic vascular graft endothelialisation and patency.

Improvement of synthetic graft patency in coronary [1] and peripheral position [7] with cell seeding underlines the importance of endothelialised surface for synthetic vascular graft biocompatibility; whereas the complete in vivo healing of the synthetic graft surface with endothelium does not occur in adult humans due to incapacity to extend pannus growth more than 6-10 mm from the anastomosis and capillary incapacity for transinterstitial growth. The reason for difference in the synthetic vascular graft healing ability between animals and humans is unknown.

In the present study, both transanastomotic and transgraft growth were observed with SEM in all three animal species. The observed hierarchy of endothelialisation between species concurred with previous reports [17,18]. Unexpectedly, there was spontaneous endothelialisation in some untreated grafts in rats at 1 week; whereas no complete endothelialisation was observed in controls at 2 weeks. We have no valid explanation for this phenomenon.

Polyester and ePTFE are the most often used synthetic vascular graft materials today with equal clinical performance in comparative studies [19]. Prior studies suggest that 60 µm internodal distance is optimal for ePTFE graft endothelialisation [3,4]. In the present study, as expected, most porous pre-clotted polyester grafts had highest rate of completely endothelialised grafts in rabbits at the earlier time points, whereas, unexpectedly, at the latest time point less porous Hybrid ePTFE was highest. The greater porosity in knitted polyester grafts compared to ePTFE grafts could cause decreased endothelial stability at later time points in the present study, similar to decreased endothelial stability with increased internodal distance from 60 to 90 µm in ePTFE grafts in a previous study [4]. In dog carotids greater proportion of 60 µm ePTFE grafts than collagen-coated polyester grafts had at least partial endothelialisation, indicating indirectly, and confirming the prior observation [20], that pre-clotted polyester grafts have better endothelialisation properties than collagen-coated polyester grafts.

In the previous studies, there is both indirect and direct evidence for the role of angiogenic growth factors in synthetic graft endothelialisation and blood vessel healing: VEGF expression is elevated in anastomosis area [21] and in transposition tissue around the synthetic graft [14]; VEGF protein application enhances anastomosis healing [22]; VEGF [5,23] and FGF proteins, and their combinations [6], stimulate endothelial cell growth on synthetic grafts; seeding of ex vivo VEGF-transfected adipose tissue improves synthetic graft endothelialisation [11]. However, VEGF affects also smooth muscle cell (SMC) behaviour: majority of studies suggest that VEGF predominantly inhibits SMC growth; whereas one study reports that the 30 µm ePTFE graft impregnated with combination VEGF-extracellular matrix enhances SMC proliferation [23]. However, there is to our knowledge no data indicating that administration of VEGF protein/gene would increase neointimal thickness in association to implantation of vascular implants.

The local in vivo gene delivery in a solution in the present study, instead of using extracellular matrix/other matrixes [23] as a carrier for the genes on the graft surface, provided the endothelial cells with uncomplicated access to the pores in the graft and, thus, avoided the theoretical problem of obstructing the pores with excessive foreign material. Alternatively, similar to proteins [24], plasmids could have been bound directly on the graft surface for local release.

In the present study, the favourable effects of in vivo administration of pNGVL3-VEGF₁₆₅ in combination with 60 µm ePTFE grafts were consistent in both rats and rabbits at all time points. The inferior results with pNGVL3-VEGF₁₆₅ transfection at 4 weeks in rabbits, after pooling the graft subgroups, derived from 30 µm ePTFE and Hybrid ePTFE subgroups. Subsequently, the graft material- and architecture-specific properties could affect endothelial stability, or susceptibility for h-VEGF₁₆₅-induced transgraft or transanastomotic growth.

In the present study, VEGF₁₆₅/FGF-2 co-transfection resulted in worst outcome at every observed time point; whereas h-VEGF₁₂₁/FGF-2 co-transfection of tumour tissue before implantation in nude mice enhances angiogenesis in previous reports [25]. Subsequently, the excessive VEGF₁₆₅ and FGF-2 co-stimulation of healthy tissues may activate an inhibitory feedback loop, whereas tumour tissue could have lost the sensitivity for inhibition.

The improved patency in clinical size vascular grafts in femoral position in dogs with pNGVL3-VEGF₁₆₅ was an important finding. The high patency rate in control grafts in carotid position in the present study was concordant with a previous study in dogs [26]; whereas the lower patency rate in femoral arteries compared to carotid arteries could be explained by differences in blood flow and surgical graft positioning (loop vs straight). The small sample sizes and high rate of occlusions in control grafts prevented comparisons between the groups regarding graft endothelialisation. However, the improved graft patency could be mediated not only by pNGVL3-VEGF₁₆₅-enhanced endothelialisation, but also by induction of NO production [27].

In the present study, it was recognized that many of the contingency tables upon which the chi-squared tests are based were quite sparse. Therefore, where appropriate, the Fisher Exact Test was used to determine statistical significance. In a post-hoc power analysis it was found that the highest powers for the observed effect sizes were approximately 70% for dogs, 60% for rabbits, and 45% for rats. In many of the tables, however, the power was well under 50%. Given the generally low statistical power in the present studies, the reported statistically significant results can be regarded as being based on real treatment effects.

Although the overall results were promising, some caution needs to be used when transferring the results to clinical coronary artery surgery. First, due to the diastolic perfusion pattern in myocardium, the blood flow pattern is substantially different in coronary artery bypass grafts compared to the both interposition and bypass grafts in systemic arteries. Therefore, the present experimental results need to be confirmed in coronary position; and the design of synthetic coronary artery grafts may need to be modified compared to peripheral artery grafts. The same applies for synthetic grafts positioned in venous system due to lower pressure in venous system. Second, the gene transfer with VEGF primarily affects the endothelialisation of the foreign surface and, thus, theoretically decreases stimuli for neointima formation [13]. However, the present methods were not designed to affect two other important stimuli for neointima formation: compliance mismatches between the synthetic graft and the native vessel; and shear stress and flow disturbances due to the angle in the end-to-side anastomoses. Third, the end-to-end interposition grafts were 2 cm long in rat aorta, whereas the end-to-side clinical coronary artery bypass graft length is 5–20 cm; and the observation time of 6 weeks with clinical size peripheral arterial synthetic grafts in dogs was primarily designed to detect differences in short term patency. Therefore, to demonstrate the applicability to clinical coronary artery surgery, a longer observation time with small-diameter synthetic coronary grafts of 5–20 cm length will be required in a large animal model.

In summary, pNGVL3-VEGF₁₆₅ resulted in improved endothelialisation of synthetic vascular grafts in rat aorta, contradictory results in rabbit aorta, and trend towards improved synthetic graft patency in dog femorals; whereas combination of pNGVL3-VEGF₁₆₅ and pNGVL7-FGF-2 had unfavourable outcome in rat aorta. There was a trend for faster endothelialisation for pre-clotted polyester grafts in rabbits at earlier time points. Subsequently, the angiogenic protein production with expression plasmids may be sufficient for healing of porous synthetic vascular grafts and a higher transfection efficacy and longer expression achieved with viral vectors may not be needed. Further studies are required; to increase our understanding concerning VEGF-transfection effect on histology and restenosis in synthetic vascular grafts; to investigate if this novel concept is advantageous in combination with synthetic grafts in humans.

	Acknowledgments

 
The authors are grateful for goodwill of Professor David Bergqvist for coaching the manuscript and Professor Göran Hedenstierna for providing facilities for rabbit operations; Professor Per Artursson and Associate Professor Jarmo Wahlfors for fruitful discussions regarding gene expression; Magnus Koping-Hoggard PhD for supervising of luciferase analysis; practical assistance of Matts Jennerberg; and data collection by Eneli Lahtinen. This study was supported by Xenerate Ab.

	Footnotes

 
^\#9734; An abstract concerning the work has been presented at International Meeting on Tissue Engineered Blood Vessels, April 23th, 2005, in Gothenburg, Sweden; and accepted, but not presented, as a poster at American Society of Gene Therapy Meeting in St. Louis, June 1–5, 2005.

	References

Top Abstract 1. Introduction 2. Materials and methods 3. Results 4. Discussion References

 

1. Laube HR, Duwe J, Rutsch W, Konertz W. Clinical experience with autologous endothelial cell-seeded polytetrafluoroethylene coronary artery bypass grafts. J Thorac Cardiovasc Surg 2000;120(1):134-141.[Abstract/Free Full Text]
2. Burkel WE, Vinter DW, Ford JW, Kahn RH, Graham LM, Stanley JC. Sequential studies of healing in endothelial seeded vascular prostheses: histologic and ultrastructure characteristics of graft incorporation. J Surg Res 1981;30(4):305-324.[CrossRef][Medline]
3. Tsuchida H, Kashyap A, Cameron BL, Peng SK, Wilson SE. In vivo study of a high-porosity polytetrafluoroethylene graft: endothelialization, fluid leakage, and the effect of fibrin glue sealing. J Invest Surg 1993;6(6):509-518.[Medline]
4. Golden MA, Hanson SR, Kirkman TR, Schneider PA, Clowes AW. Healing of polytetrafluoroethylene arterial grafts is influenced by graft porosity. J Vasc Surg 1990;11(6):838-844discussion 845.[CrossRef][Medline]
5. Stone D, Phaneuf M, Sivamurthy N, LoGerfo FW, Quist WC. A biologically active VEGF construct in vitro: implications for bioengineering-improved prosthetic vascular grafts. J Biomed Mater Res 2002;59(1):160-165.[CrossRef][Medline]
6. Masuda S, Doi K, Satoh S, Oka T, Matsuda T. Vascular endothelial growth factor enhances vascularization in microporous small caliber polyurethane grafts. ASAIO J 1997;43(5):M530-M534.[Medline]
7. 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(5 Suppl.):S327-S331.[Abstract/Free Full Text]
8. Hoenig MR, Campbell GR, Rolfe BE, Campbell JH. Tissue-engineered blood vessels: alternative to autologous grafts?. Arterioscler Thromb Vasc Biol 2005;25(6):1128-1134.[Abstract/Free Full Text]
9. Dunn PF, Newman KD, Jones M, Yamada I, Shayani V, Virmani R, Dichek DA. Seeding of vascular grafts with genetically modified endothelial cells. Secretion of recombinant TPA results in decreased seeded cell retention in vitro and in vivo. Circulation 1996;93(7):1439-1446.[Abstract/Free Full Text]
10. Huber TS, Welling TH, Sarkar R, Messina LM, Stanley JC. Effects of retroviral-mediated tissue plasminogen activator gene transfer and expression on adherence and proliferation of canine endothelial cells seeded onto expanded polytetrafluoroethylene. J Vasc Surg 1995;22(6):795-803.[CrossRef][Medline]
11. Yamamoto K, Takanashi Y, Dzeu V, Pratt R. Human VEGF gens transfer into vascular prosthesis. Circulation 1996;637:3721.
12. Bekeredjian R, Shohet RV. Cardiovascular gene therapy: angiogenesis and beyond. Am J Med Sci 2004;327(3):139-148.[CrossRef][Medline]
13. Van Belle E, Tio FO, Chen D, Maillard L, Kearney M, Isner JM. Passivation of metallic stents after arterial gene transfer of phVEGF165 inhibits thrombus formation and intimal thickening. J Am Coll Cardiol 1997;29(6):1371-1379.[Abstract]
14. Hazama K, Nishibe T, Shimada T, Miura H, Watanabe S, Kondoh S, Takahashi T, Okuda Y, Katoh H. Effects of omental wrap on performance of small-caliber high-porosity expanded polytetrafluoroethylene grafts. J Surg Res 1999;81(2):174-180.[CrossRef][Medline]
15. Tomizawa Y, Takanashi Y, Noshikiki Y, Nishida H, Koyanagi H. Evaluation of small caliber vascular prostheses implanted in small children: activated angiogenesis and accelerated calcification. ASAIO J 1998;44(5):M496-M500.[Medline]
16. Koping-Hoggard M, Tubulekas I, Guan H, Edwards K, Nilsson M, Varum KM, Artursson P. Chitosan as a nonviral gene delivery system. Structure-property relationships and characteristics compared with polyethylenimine in vitro and after lung administration in vivo. Gene Ther 2001;8(14):1108-1121.[CrossRef][Medline]
17. Sauvage LR, Berger KE, Wood SJ, Yates 2nd SG, Smith JC, Mansfield PB. Interspecies healing of porous arterial prostheses: observations, 1960 to 1974. Arch Surg 1974;109(5):698-705.[Abstract/Free Full Text]
18. Noishiki Y, Yamane Y, Ichikawa Y, Yamazaki I, Yamamoto K, Kosuge T, Manabe T, Mo M. Age dependency of neointima formation on vascular prostheses in dogs. Artif Organs 2000;24(9):718-728.[CrossRef][Medline]
19. Lau H, Cheng SW. Is the preferential use of ePTFE grafts in femorofemoral bypass justified?. Ann Vasc Surg 2001;15(3):383-387.[CrossRef][Medline]
20. Guidoin R, Marois Y, Deng X, Chakfe N, Marois M, Roy R, King MW, Douville Y. Can collagen impregnated polyester arterial prostheses be recommended as small diameter blood conduits?. ASAIO J 1996;42(6):974-983.[Medline]
21. Hamdan AD, Aiello LP, Misare BD, Contreras MA, King GL, LoGerfo FW, Quist WC. Vascular endothelial growth factor expression in canine peripheral vein bypass grafts. J Vasc Surg 1997;26(1):79-86.[CrossRef][Medline]
22. Seipelt RG, Backer CL, Mavroudis C, Stellmach V, Seipelt IM, Cornwell M, Hernandez J, Crawford SE. Topical VEGF enhances healing of thoracic aortic anastomosis for coarctation in a rabbit model. Circulation 2003;108(Suppl. 1):II150-II154.[Medline]
23. Randone B, Cavallaro G, Polistena A, Cucina A, Coluccia P, Graziano P, Cavallaro A. Dual role of VEGF in pretreated experimental ePTFE arterial grafts. J Surg Res 2005;127(2):70-79.[CrossRef][Medline]
24. Crombez M, Chevallier P, Gaudreault RC, Petitclerc E, Mantovani D, Laroche G. Improving arterial prosthesis neo-endothelialization: application of a proactive VEGF construct onto PTFE surfaces. Biomaterials 2005;26(35):7402-7409.[CrossRef][Medline]
25. Giavazzi R, Sennino B, Coltrini D, Garofalo A, Dossi R, Ronca R, Tosatti MP, Presta M. Distinct role of fibroblast growth factor-2 and vascular endothelial growth factor on tumor growth and angiogenesis. Am J Pathol 2003;162(6):1913-1926.[Abstract/Free Full Text]
26. Pasic M, Muller-Glauser W, von Segesser L, Odermatt B, Lachat M, Turina M. Endothelial cell seeding improves patency of synthetic vascular grafts: manual versus automatized method. Eur J Cardiothorac Surg 1996;10(5):372-379.[Abstract]
27. Kroll J, Waltenberger J. VEGF-A induces expression of eNOS and iNOS in endothelial cells via VEGF receptor-2 (KDR). Biochem Biophys Res Commun 1998;252(3):743-746.[CrossRef][Medline]

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to Personal Folders Download to citation manager Author home page(s): Mika Lahtinen Giedrius Baliulis Vitas Zemgulis Permission Requests Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Lahtinen, M. Articles by Zemgulis, V. Search for Related Content PubMed PubMed Citation Articles by Lahtinen, M. Articles by Zemgulis, V. Related Collections Coronary disease Peripheral vascular