| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
J Thorac Cardiovasc Surg 1994;108:540-548
© 1994 Mosby, Inc.
SURGERY FOR ACQUIRED HEART DISEASE |
Sheffield and Cardiff, United Kingdom
Supported by grants from the National Heart Research Fund and the British Heart Foundation.
Received for publication Oct. 13, 1993. Accepted for publication Jan. 28, 1994. Address for reprints: Department of Cardiac Surgery, University of Bristol, Bristol Royal Infirmary, Bristol BS2 8HW, United Kingdom.
Abstract
Intimal smooth muscle cell proliferation and superimposed atheroma are the main causes of late failure of saphenous vein bypass grafts. It has been suggested that these reactions are caused by the production of growth factors from the cells of the vessel wall. To test this hypothesis, we cultured segments of pig venous arterial grafts, removed 1 and 4 weeks after implantation, in serum-free medium for 24 hours. Tissue viability as assessed by adenosine triphosphate concentration was maintained throughout the 24-hour culture period (239±21 nmol/gm wet weight [standard error of the mean], n = 26, 0 hours; 240±24 nmol/gm wet weight, n = 17, 24 hours). Cell proliferation occurred and autoradiography showed proliferating cells to be located in the neointimal and medial layers. These cells were identified as smooth muscle cells by means of a monoclonal antibody to
-actin. Graft-conditioned media were tested for mitogenic activity by means of a fibroblast proliferation assay. Media conditioned for 24 hours produced significant stimulation of cell growth (284%±30%, n = 17) above that obtained in culture medium alone (100%). This mitogenic activity was inhibited by 61%±9%, n = 8, with a polyclonal-neutralizing antibody to platelet-derived growth factor. Reverse-transcription polymerase chain reaction analysis and Northern blots demonstrated platelet-derived growth factor B messenger ribonucleic acid (mRNA) in vein grafts but not in ungrafted vein. Analysis of graft tissue sections by in situ hybridization demonstrated an abundance of platelet-derived growth factor B mRNA positive cells in the endothelial and neointimal layers, as well as in the endothelial cells of the adventitial vessels. These data constitute direct evidence for active growth factor production within the cells of the vein graft. They also suggest that endogenously produced platelet-derived growth factor may play a role in regulating smooth muscle cell proliferation in this model. (J THORACCARDIOVASCSURG1994;108:540-8)
Autologous saphenous vein continues to be widely used for coronary and peripheral artery bypass grafting despite its disappointing long-term patency rate. 
1-4 Early occlusions occur by thrombosis on a damaged endothelial surface whereas late occlusions are due to narrowing of the graft lumen by intimal vascular smooth muscle cell proliferation. The resulting thickened intima then appears to encourage superimposed atheroma formation. 2-4 Therefore considerable interest has been generated in defining the factors responsible for intimal vascular smooth muscle cell proliferation.
Endothelial damage during surgical preparation promotes platelet and leukocyte adhesion 5 which, in turn, may play a role in early vascular smooth muscle cell proliferation. It is now clear, however, that intimal proliferation continues beyond the time taken for endothelial repair 6 and is not influenced by the degree ofinitial endothelial injury 6 or by antithrombotic treatments. 3 This leads to the hypothesis that grafting, through either altered hemodynamics 2,7,8 or injury, 9-11 induces growth factor production from the venous endothelium or smooth muscle cell layers, which then causes intimally directed vascular smooth muscle cell proliferation. A number of such growth factors have been isolated and characterized in cell culture studies 12 in lesions of atherosclerosis 13-17 in models of angioplasty and at the anastomotic sites of grafts constructed with synthetic conduits. 18 Nevertheless, there is no direct evidence as to which, if any, of these growth factors may be produced in venous arterial bypass grafts.
This problem was investigated in pig saphenous vein–carotid artery bypass grafts, obtained 1 and 4 weeks after implantation, that were cultured for 24 hours in serum-free medium. Release of growth factor activity was assessed by the ability of conditioned media to stimulate the proliferation of quiescent 3T3 fibroblasts. Neutralizing antibodies were used to characterize the growth-promoting activity. To confirm that cells intrinsic to the graft were responsible for growth factor production, we also investigated growth factor messenger ribonucleic acid (mRNA) by reverse-transcription polymerase chain reaction (PCR) analysis using heterologous primers based on human gene sequences.
Finally, we have shown by in situ hybridization of mRNA that platelet-derived growth factor (PDGF) expression in graft tissue is localized predominantly to the cells of the endothelial and neointimal layers with some expression in the vasa vasorum. Thus PDGF B chain mRNA expression occurs in the same place as the cellular proliferative response characteristic of progressive vein graft intimal thickening.
MATERIALS AND METHODS
Surgical procedures
Studies were performed with 15 Yorkshire White pigs weighing 20 to 25 kg. Premedication, anesthesia, and carotid artery bypass grafting with autologous saphenous vein were performed as detailed previously. 5 In brief, a longitudinal incision was made on the outer aspect of the hind leg to expose approximately 12 cm of the long saphenous vein. The vein was dissected free of surrounding tissue by a "no touch" technique, 19 and all side branches were secured with a 6-0 Prolene ligature (Ethicon, Inc., Somerville, N.J.). After this, the vein was removed and gently irrigated with heparinized isosmotic sodium chloride solution (0.9 gm/L). The vein was stored in the same solution at room temperature for 10 to 30 minutes until needed. The common carotid artery was isolated through a longitudinal neck incision and a 3 to 4 cm segment, isolated between vascular clamps, was excised. Both cut ends were then beveled to approximately 45 degrees. A segment of the reversed saphenous vein was similarly beveled and then anastomosed end to end to the carotid artery with a continuous 7-0 Prolene suture. The neck and leg wounds were then closed in layers and the animals allowed to recover. After 1 and 4 weeks the animals were reanesthetized as described earlier, the neck was opened, and the graft was identified. The carotid artery was transected distal to the graft and the absence of blood flow was taken to indicate graft occlusion. Patent grafts were then transported to the laboratory in sterile HEPES-buffered RPMI-1640 medium supplemented with penicillin (100 µg/L), streptomycin (100 U/ml), amphotericin (2.5 µg/ml), gentamicin (2.5 µg/ml), glutamine (2 mmol/L) (all from Flow Labs, High Wycombe, United Kingdom), and heparin (4 U/ml) (CP Pharmaceuticals, Wrexham, United Kingdom). The transfer time to the laboratory was approximately 30 minutes.
Tissue culture procedure
Grafts were transferred into Petri dishes containing culture medium that consisted of RPMI-1640 medium supplemented with sodium bicarbonate 2.0 gm/L, in place of HEPES, and antibiotics as described earlier. The grafts were cleaned of excess fat and adventitia, opened along their upper aspect, and washed carefully several times. Then approximately 1 cm segments were pinned out, endothelial surface uppermost, onto polyester mesh in the base of the Petri dish. Segments were then rested for 15 minutes, after which time the conditioned media was removed. Fresh culture medium supplemented with [ 3H]-thymidine (1µCi/ml, specific activity 25 Ci/mmol; Amersham International, Amersham, United Kingdom) was then added to the cultures, which were maintained at 37° C in a humidified atmosphere with 5% (vol/vol) carbon dioxide in air. At the end of the 24-hour culture period the conditioned medium was collected, aliquoted into sterile tubes, and stored at -80° C.
In a separate set of experiments, conditioned media were collected at several times during a 28-hour period. Fresh medium prewarmed to 37° C was added at each time point.
Measurement of purine metabolites by high-performance liquid chromatography, 20 deoxyribonucleic acid (DNA) by fluorimetry, 21 and total thymidine incorporation by liquid scintillation spectrometry was carried out as previously described. 22
Electron and light microscopy
Morphologic characteristics of the endothelium were studied by en face scanning electron microscopy and morphologic characteristics of the graft with transmission electron microscopy, both carried out as previously described. 22 Graft segments were also fixed in 10% buffered formalin for 24 hours, processed, and embedded in paraffin. Transverse sections (4 µm) were stained with alcian blue–Miller's elastic–van Gieson stain and the mean medial and intimal thickness determined from measurements taken at 20 equidistant points along the section length with a digital image analyzer (Seescan, Cambridge, United Kingdom).
Autoradiography was performed as previously described 22 to localize cell proliferation occurring in culture. Immunocytochemistry on paraffin sections of graft was also performed to identify macrophages. Deparaffinized, trypsinized sections were incubated for 30 minutes with a 1:100 dilution of Mac-387 antisera (Dako, High Wycombe, United Kingdom). After the sections were washed in Tris-buffered saline solution, biotinylated secondary antisera (1:50) was added for 30 minutes. Sections were washed twice and streptavidin/biotin complex (1:100) (Dako) was added for a further 30 minutes. The substrate diaminobenzidine was used to visualize the staining and nuclei were counterstained with Mayer's hematoxylin stain.
Cell proliferation assay
Swiss mouse 3T3 fibroblasts were used to assess the mitogenic activity of the pig vein graft–conditioned media. Cells were passaged twice a week in Dulbecco's modified Eagle's medium with bicarbonate (2 gm/L) and supplemented with 10% fetal calf serum (Northumbria Biologicals, Northumberland, United Kingdom), penicillin (100 µg/ml), streptomycin (100 U/ml), amphotericin (2.5 µg/ml), gentamicin (2.5 µg/ml), and glutamine (2 mmol/L). Cells were plated at a density of 10 4 cells per well in a 96-well plate (Costar, High Wycombe, United Kingdom) in Dulbecco's modified Eagle's medium supplemented with antibiotics and 10% fetal calf serum. After 24 hours, the cells were made quiescent by incubation in medium containing 2% pig plasma–derived serum 23 for a further 2 days before the addition of undiluted test samples and standards. [ 3H]-Thymidine (1 µCi/ml, specific activity 25 Ci/mmol) was added to the cells simultaneously with the test agents, and its incorporation into DNA was determined over a 20-hour interval.
After exposure, the cell layers were washed three times with phosphate-buffered saline solution and fixed for 3 minutes with ice-cold 10% wt/vol trichloracetic acid. The trichloracetic acid–precipitable material was harvested by aspirating the media and solubilizing the remaining trichloracetic acid–insoluble material in 100 µl of 1 mol/L sodium hydroxide overnight at 37° C. This solubilized material was then added to 2 ml of scintillation fluid (Ultima gold, Packard, Pangbourn, United Kingdom), and radioactivity was determined by liquid scintillation counting. The data are expressed as disintegrations per minute (DPM) per well, and as percentage stimulation over that in culture medium alone. The response to 10% calf serum was taken as a positive control. Porcine PDGF (British Biotechnology, Oxford, United Kingdom, 10 ng/ml in culture medium) was also used in the bioassay as a pure source of mitogen. In a separate set of experiments, a polyclonal antibody to PDGF that neutralized all forms of PDGF (British Biotechnology) was added at a final concentration of 100 µg/ml to the conditioned medium, and the results were compared with those of the same sample of medium incubated with nonimmune immunoglobulin G (Sigma Chemicals, Poole, United Kingdom) at the same concentration. Media were preincubated with antibody or nonimmune immunoglobulin G for 1 hour at 37° C before addition to the fibroblasts.
Reverse-transcription PCR
Cellular RNA was isolated from snap-frozen segments of freshly isolated saphenous vein and venous arterial graft by means of a one-step phenol chloroform method. 24 In brief, vessels were homogenized in RNAzol (Biogenesis Ltd., Bournemouth, United Kingdom), RNA was precipitated with ice-cold isopropanol (BDH Chemicals, Poole, Dorset, United Kingdom) and washed in ice-cold 75% vol/vol ethanol (BDH Chemicals). Total RNA (1 µg) was reverse transcribed in a 20 µl reaction containing 1.25 U/ml avian myeloblastosis virus reverse transcriptase and 2.5 µmol/L random hexanucleotide primers as described by di Giovine and associates. 25
Primers for PDGF B chain were derived from the human nucleotide sequences, designed by a commercial software program (OLIGO, National Biosciences, Hamel, Minn.) and screened for specificity by the SERC facility Seqnet VAX 3600 computer on node DVLH (Janet 000001009302). Sequences used for PDGF B were in the 3' untranslated region of the gene 5'CG CAC CAA CGC CAA CTT CC 3'(sense) and 5'TTT GGC TCG CTG CTC CTG GG 3'(antisense) corresponding to nucleotides 1318 to 1569, respectively. The predicted size for the PDGF B complementary deoxyribonucleic acid (cDNA) product was 271 bp on the cDNA from mRNA.
PCR of cDNA was performed as described by di Giovine and colleagues. 25 The cycling parameters were 1 minute at 95°C, 1 minute at 62° C, and extension at 72° C for 1 minute for 35 cycles, with a final extension period of 6 minutes at 72° C. PCR products were size fractionated and separated from unincorporated primers by electrophoresis through 1.5% agarose gels and amplified DNA visualized by ethidium bromide staining under ultraviolet light transillumination. Control primers were designed to ß-actin Genbank:M10278 5'TTC TAC AAT GAG CTG CGT GTG G 3'(antisense) and 5'CTC GGT CAG GAT CTT CAT GAG G 3' predicted size 318 bp on cDNA from mRNA.
In all experiments, to rule out any false mRNA signal from any genomic DNA contaminating the RNA, separate reactions were done both with and without reverse-transcription enzyme to check for amplification of genomic DNA from intronless copies of the PDGF and actin genes.
Sequencing of reverse-transcription PCR generated probe and Northern blot analysis
It was necessary to sequence the cDNA product generated by reverse-transcription PCR to validate it before use in Northern blots. In brief, the downstream primer was end labeled with biotin to purify single-stranded DNA for sequencing with streptavidin-coated beads. After reverse-transcription PCR, the biotinylated PCR products (100 µl) were purified by ethanol precipitation and single-stranded DNA isolated by affinity separation on streptavidin-coated magnetic beads (Dynal UK, Wirral, Cheshire, United Kingdom) and resuspended in water. Sequencing reactions were performed by dideoxy chain termination and T7 polymerase (Sequenase V2.0, U.S. Biochemical Corp., Cleveland, Ohio) with deoxyadenosine triphosphate, deoxycitidine triphosphate, deoxythymidine triphosphate, and deoxyguanosine triphosphate, and [- 35S] dATP (5 µCi, 1000 Ci/mmol, Amersham, United Kingdom). Samples were heated at 80° C before being loaded on a TBE (Tris–boric acid–ethylenediaminetetraacetic acid) buffer gradient (7.5x to 0.5x), 6% acrylamide 0.4 mm x 21 cm x 80 cm gel, and run for 10 hours at a constant voltage of 2800 V. The gel was then dried for 2 hours and applied to film (Hyperfilm ß max, Amersham) for autoradiography.
For Northern blot analysis, total RNA was extracted from ungrafted and grafted vein as described by Chomczynski and Sacchi. 24 Northern blots using 10 µg total RNA were performed as described by di Giovine and colleagues, 25 and membranes were then exposed to film (Hyperfilm ß max, Amersham) for 24 hours at -70° C with double intensifying screens. Blots were stripped, reexposed, and rehybridized to a control cDNA probe for ß-actin.
In situ hybridization
Tissue segments for in situ hybridization were fixed for 24 hours in 10% buffered formalin and embedded in paraffin wax. Tissue sections collected onto aminopropyltriethoxysilane-coated slides (4 µm) were deparaffinized and pretreated sequentially with 4% paraformaldehyde (10 minutes at 4° C), Proteinase K (1 µg/ml in 500 mmol/L NaCl in 10 mmol/L Tris HCl, pH 8.0) (7.5 minutes at 37° C), and 0.25% acetic anhydride in triethanolamine (0.1 mol/L, pH 8.0). The slides were prehybridized for 2 hours and then hybridized with the 35S-labeled riboprobe (see below) at 0.6 x 10 6 cpm/µl according to themethod described by Wilcox and colleagues. 16 The tissue sections coated with emulsion were left in the dark at 4° C for 10 weeks. The probe used for in situ hybridization was generated by reverse-transcription PCR, cloned in to the pCR vector (Invitrogen, Oxford, United Kingdom), and subsequently used as a template for in vitro transcription of antisense and sense probes by SP6 or T7 polymerase, respectively (Promega) in the presence of uridine 5'[(35S)thio triphosphate (>1000 Ci/mmol, Amersham).
Statistical methods
Values shown are mean ± standard error of the mean. Data were considered significant if p was less than 0.05 according to the Student's t test for paired and unpaired data.
RESULTS
Graft morphologic characteristics
The time course of medial and intimal thickening in pig venous arterial grafts has been reported. 6 Consistent with this previous work, the internal elastic laminae and the punctate elastic material distributed throughout the tunica media became more separated in grafts than in veins. During the first week, there was an increase in medial thickness as compared with ungrafted vein and a detectable neointima developed (Fig. 1, a). Between 1 and 4 weeks a further increase in medial thickness (423 ± 78 versus 828 ± 51 µm, n = 5, p x 0.005) and intimal thickening also occurred (70 ± 7 versus 177 ± 12 µm, n = 5, p x 0.005). The neointimal layer was composed of large amounts of extracellular matrix and smooth muscle–like cells containing an abundance of rough endoplasmic reticulum and few actin filaments suggesting that they were in a synthetic rather than a contractile phenotype (Fig. 1, b). Immunocytochemistry with a monoclonal antibody against -actin confirmed the presence of vascular smooth muscle cells in the neointima (Fig. 1, c). A few macrophages were seen by immunostaining with the antibody Mac-387 in the medial layers with occasional positive cells in the neointima (data not shown). Cell proliferation occurred in the grafts during culture as assessed by [3H]-thymidine incorporation (893 ± 113 DPM/µg DNA, n = 17). Autoradiography of transverse sections showed dividing labeled cells in the neointimal and medial layers (Fig. 1, d).
|
Scanning electron microscopy of grafts removed after 1 and 4 weeks showed that the endothelial cells were largely undamaged, and this appearance was maintained after culture (data not shown). This argues against the possibility that mitogens were released into the medium as a result of cell injury or death. Scanning electron microscopy also showed no evidence of adherent platelets or microthrombi on the luminal surface of grafts prepared for culture. It therefore seems unlikely that the platelets were the source of the mitogenic activity detected. Further evidence for endogenous mitogenic activity was provided by testing conditioned media from grafts rested for 15 minutes, which produced only minimal proliferation of 3T3 fibroblasts as assessed by 3H-thymidine incorporation (DPM/well) when compared withserum-free culture medium (11,565 ± 543, n = 8, versus 10,124 ± 1,227, n = 12, no significant difference).
Conditioned media from 1- and 4-week grafts maintained in serum-free culture for 24 hours caused a significantly higher fibroblast proliferation of 26,339 ± 1,786 DPM/well (n = 10) and 30,436 ± 4,200 DPM/well (n = 7), respectively (both p < 0.005 versus serum-free culture media). Media containing 10% fetal calf serum produced a cell proliferation of 48,918 ± 5,248 DPM/ well (n = 12) (p < 0.002 versus serum-free culture media). In a separate set of experiments the release of mitogens into the media was evaluated at seven time points during the culture period. The release of mitogens appeared maximal at 6 to 8 hours and then declined to a lower but constant level between 16 and 28 hours (Fig. 2).
|
Expression of PDGF B chain mRNA in venous arterial grafts
Reverse-transcription PCR was used to assess mRNA levels for PDGF B chain in venous arterial grafts. With this technique, higher levels of PDGF B chain were observed in grafts than in ungrafted veins (Fig. 3, A and B). These results were confirmed by Northern blot analysis using the amplified cDNA, 271 bp as a probe (Fig. 3, C and D). The sequence of the reverse-transcription PCR-generated cDNA probe was confirmed to be identical to that of human PDGF B chain by dideoxy chain termination sequencing.
|
|
In this series of pig venous arterial grafts, a rapid intimal thickening occurred as a result of vascular smooth muscle cell proliferation and synthesis of extracellular matrix. Consistent with previous reports, 5,6 intimal thickening apparently took place in the presence of an intact endothelium and after termination of platelet activation, 5,6 which implies that growth factors released from platelet granules do not play an important role. Production of endogenous growth factors from the venous endothelium, vascular smooth muscle cells, or from invading inflammatory cells seems likely therefore to be responsible for the intimally directed proliferation of vascular smooth muscle cells. The results of this study provide direct evidence for production of growth factor activity by the cells intrinsic to the vein graft. Furthermore, the study demonstrates the release of a PDGF-like mitogen and increased expression of mRNA for one form of PDGF. Thus PDGF, which is a potent vascular smooth muscle cell mitogen, has a presumptive role in regulating vascular smooth muscle cell proliferation in this model. In this respect, the role of PDGF may be similar to that proposed in the development of human 12,13 and experimental atherosclerotic lesions. 14-17 Indeed, recent work with direct gene transfer techniques in porcine arteries has implicated recombinant PDGF B gene expression in inducing intimal hyperplasia in vivo. 26
For release of growth factors from vein grafts to be measured, a short-term serum-free organ culture had to be developed, and the primary concern was to demonstrate maintenance of intimal and medial cell viability. Scanning electron microscopy of the intimal surface of cultured grafts showed a confluent endothelium with occasional gaps between cells. Transmission electron microscopy confirmed the findings of healthy looking endothelial cells covering the neointimal layers, which were made of abundant extracellular matrix and smooth muscle cells in a predominantly synthetic phenotype. Tissue ATP concentration (which has previously been used to assess the viability of vascular smooth muscle cells in tissue 22) was also maintained during culture. Taken together, these observations demonstrate that a high degree of cell viability was maintained and argue against the possibility that the release of growth factor activity into the conditioned media was from dying cells. Scanning electron microscopy also showed no evidence of adherent platelets on the intimal surface of grafts that had been prepared for culture, and it therefore seems unlikely that platelets were the source of the released mitogens detected.
A continuing release of mitogenic activity from the graft into the culture media was observed. Release of mitogens was low when grafts were simply rinsed in medium and rested for 15 minutes. Release was higher initially, but settled to a constant level between 16 and 28 hours. The time course is consistent with the interpretation that cells intrinsic to the graft continue to synthesize and release growth factors after being placed in culture. PDGF B chain gene expression was demonstrated in the graft by means of reverse-transcription PCR and Northern blot analysis. This provides important supplementary evidence that arterialization of the vein segment over the course of 1 week activates growth factor gene expression.
In situ hybridization analysis of venous arterial graft tissue revealed an intense hybridization signal for PDGF B chain mRNA in the cells of the graft endothelial and neointimal layers. We have previously demonstrated (using immunostaining and transmission electron microscopy) that these cells have the characteristics of smooth muscle cells. The increased expression of PDGF B mRNA observed in these cells can therefore be attributed to the implantation of saphenous vein into the arterial circulation. Increased expression of PDGF B mRNA was also observed in the vasa vasorum. This is interesting, because we and others 12 have observed the perivascular proliferation of the microvasculature or vasa vasorum in venous arterial bypass grafts. Because increased vasa vasorum formation is usually confined to patent grafts, this would imply that the process of angiogenesis is important in maintaining graft patency by encouraging smooth muscle cell migration toward the adventitial edge of the vessel, as well as stimulating collateral blood flow. PDGF 26 and other factors, such as basic fibroblast growth factor, 27-29 have been implicated in angiogenesis.
Further work should aim to determine the precise cellular source of the mitogens that mediate intimal hyperplasia in this model. Both endothelial cells and smooth muscle cells are known to be capable of producing growth factors, including PDGF, in vitro. 30 Infiltrating macrophages are also capable of producing PDGF, 31 but these were detected infrequently in the vein grafts. In addition, because only part of the growth-promoting activity was related to PDGF-like proteins, other mitogens must be present and identification of other active growth factor and cytokine genes is a research priority.
Footnotes
From the Departments of Cardiac Surgery,a Molecular Medicine,b and Pathology,c University of Sheffield, Sheffield, and the Department of Cardiology,d University of Wales College of Medicine, Cardiff, United Kingdom. 
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
X174-Hae III DNA markers: Lane 1, Positive control (human atheromatous coronary artery); lane 2, porcine venous arterial graft (1 week); lane 3, ungrafted saphenous vein; lane 4, negative control (without reverse transcriptase). Third panel, Total RNA was extracted from porcine ungrafted vein and from a 1-week graft tested by Northern blot for PDGF B chain as described in the text. Lane 1, Venous arterial graft; lane 2, ungrafted saphenous vein. Third panel, C, An increased level of PDGF B transcript is seen in the graft. Fourth panel, D, Signal obtained from rehybridization of the blot with a beta-actin probe that detects a single species of 2.1 kb.
