Author + information
- Received April 15, 1996
- Revision received February 14, 1997
- Accepted February 26, 1997
- Published online June 1, 1997.
- Noel M Caplice, MRCPI, FRACPA,
- Constantine N Aroney, MD, FRACPB,
- J.H.Nicholas Bett, FRACPB,
- James Cameron, FRACPB,
- Julie H Campbell, PhDC,
- Nancy Hoffmann, BScA,
- Paul T McEniery, MBBS, FACCB and
- Malcolm J West, MBBS, PhD, FRACPA,*
- ↵*Dr. Malcolm J. West, Department of Medicine, Level B1, Pathology, Prince Charles Hospital, Rode Road, Chermside, Brisbane, Queensland 4032, Australia.
Objectives. This study sought to 1) assess in vivo release of platelet-derived growth factor (PDGF) and basic fibroblast growth factor (bFGF) into the coronary circulation after vascular injury in human subjects; and 2) evaluate mitogenic effects of PDGF and bFGF on the patient’s own vascular smooth muscle cells (VSMCs).
Background. Circumstantial evidence suggests involvement of PDGF and bFGF peptides in the neointimal response to vascular injury. To date, no study has shown biologically active growth factors within the coronary circulation after vascular injury in human subjects.
Methods. In 18 patients, plasma PDGF AB, platelet factor 4 (PF4) and beta-thromboglobulin (beta-TG) levels were measured in coronary sinus blood obtained before and up to 30 min after angioplasty. In five patients undergoing atherectomy, coronary sinus serum was added to cultured VSMCs derived from atherectomy tissue to assess the mitogenic potential of the serum. Mitogenicity attributable to PDGF and bFGF was determined using neutralizing antibodies to these factors. PDGF A, PDGF B and bFGF were localized within the atherectomy tissue using immunocytochemical analysis.
Results. Before angioplasty, PDGF AB, PF4 and beta-TG levels were elevated threefold in patients scheduled for angioplasty compared with those in control patients (p < 0.01). Within 5 min of angioplasty, PDGF AB levels increased twofold and returned toward preangioplasty levels at 30 min; PF4 and beta-TG levels remained elevated. Serum obtained at 30 min after atherectomy showed a sixfold increase in mitogenicity compared with preatherectomy serum (p = 0.01). This increase in mitogenicity was reduced by 20%, 40% and 65% in the presence of neutralizing antibodies to PDGF, bFGF and PDGF + bFGF, respectively. PDGF A, PDGF B and bFGF were visualized within the intima of the atherectomy tissue.
Conclusions. The change in plasma PDGF level is consistent with first-phase release of PDGF after vascular injury. The increase in mitogenicity of serum suggests that PDGF and bFGF are biologically active.
(J Am Coll Cardiol 1997;29:1536–41)
Neointimal proliferation and subsequent restenosis is a common outcome after treatment of obstructive coronary artery lesions with percutaneous transluminal coronary angioplasty (PTCA) or atherectomy in human subjects ([1, 2]). This neointima is composed of proliferating vascular smooth muscle cells (VSMCs), platelets, organized thrombus and extracellular matrix (). After arterial injury, release of mitogenic and chemotactic factors from intracellular and extracellular elements within the lesion and from aggregated platelets on the damaged intimal surface is thought to initiate the neointimal response ().
A number of studies suggest that growth factors mediate neointimal proliferation of VSMCs, but the mechanisms have not been clarified. Immunocytochemical and in situ hybridization techniques have demonstrated ([5–7]) increased expression of platelet-derived growth factor (PDGF) and basic fibroblast growth factor (bFGF) within atherosclerotic lesions. In the rat balloon injury model ([8, 9]), neutralizing antibodies to PDGF or bFGF have successfully inhibited neointimal formation. Trapidil, a competitive PDGF receptor blocking agent, has significantly reduced the rate of restenosis after angioplasty in both animal and human studies ([10, 11]). However, despite these data, the contribution of different cellular and extracellular elements in the injured lesion to the dynamics of growth factor release after angioplasty in humans is uncertain. Furthermore, the relative importance of different growth factors in lesion development, and the time during which mitogen release occurs, is unknown.
The purpose of the present study was to search for evidence of growth factor release in human subjects that would equate with the first phase of the response to injury pattern in the rat balloon injury model (). To identify the existence and biologic significance of first-phase growth factor release in human subjects, we devised two experiments.
In the first of these we assessed plasma levels of PDGF AB together with the platelet activation markers platelet factor 4 (PF4) and beta-thromboglobulin (beta-TG), collected from the coronary sinus before and after angioplasty. In a second series of experiments, we collected coronary sinus serum before and after atherectomy. The mitogenic potential of each patient’s serum was assessed by measuring its proliferative effect on cultured VSMCs derived from the same patient’s tissue obtained at the time of atherectomy. Using neutralizing antibodies to PDGF and bFGF, the relative contribution of growth factor to patient serum mitogenicity was determined. Immunocytochemical analysis was used to localize PDGF A, PDGF B and bFGF within recovered atherectomy tissue.
The first study group included 18 patients with single-vessel coronary artery disease (15 men, 3 women; mean [±SD] age 54 ± 12 years, range 35 to 73) referred for PTCA and 10 control patients (8 men, 2 women; mean age 57 ± 12 years, range 39 to 70) with angiographically normal coronary arteries. The second study group included five male patients (mean age 45 ± 8 years, range 36 to 58) undergoing directional atherectomy for proximal left anterior descending coronary artery stenosis. After femoral sheath access, all patients received 10,000 to 15,000 U of heparin intraarterially. All patients were taking aspirin and antianginal medications (beta-adrenergic blocking agents, calcium channel blocking agents and nitrates). Patients with diseases that might influence platelet function, such as severe anemia, chronic renal failure, recent unstable angina and recent myocardial infarction, were excluded from the study. All subjects had given written informed consent in accordance with hospital ethical guidelines.
1.2 Coronary Sinus Sampling.
A 7F Cournand catheter was placed through the left antecubital vein into the coronary sinus. After 30 min, a control blood sample was collected. Blood was again taken at 5, 10, 15, 20 and 30 min after successful angioplasty or atherectomy with the balloon or atherectomy device withdrawn into the guiding catheter. Three milliliter samples of blood were obtained at each time point and those samples not obtained smoothly were discarded.
1.3 Plasma and Serum Preparation.
Blood for plasma samples was collected in ice-cold glass tubes (Becton-Dickinson) containing citrate, theophylline, adenosine and dipyridamole. Blood for serum samples was allowed to clot in SST clot activator tubes (Becton Dickinson Vacutainer Systems, Meylan, France). Plasma and serum was separated and centrifuged at 3,000 rpm for 30 min at 4°C and then immediately frozen and stored at −80°C.
1.4 Plasma Growth Factor and Platelet Marker Assays.
Plasma assays of PDGF AB, PF4 and beta-TG were performed within 1 month of collection using commercially available enzyme-linked immunosorbent kits (PDGF AB: R. & D. Systems; PF4: Behring Diagnostics, Marburg, Germany; beta-TG: Diagnostica Stago, Asierses-sur-Seine, France).
1.5 VSMC Culture and Serum Mitogenicity.
Tissue obtained during atherectomy was explanted on 35-mm fibronectin-coated (10 μg/cm2) plastic culture plates in the presence of Dulbecco’s modified Eagles’ medium (DMEM) and 10% fetal calf serum. Medium was changed every fourth day. After 4 weeks, VSMCs were subcultured using trypsin (0.125%). VSMCs were identified by their “hill and valley” morphology and positive staining with alpha-smooth muscle actin (HHF 35) antibody.
VSMCs between passages 2 and 6 were used for mitogenicity experiments and were seeded at a density of 1 × 104cells/well in a 96-well plate. Cells were allowed to attach overnight and then growth arrested for 48 h in serum-free medium. Cells were then released from growth arrest using 5% (vol/vol) of the patient’s own serum in DMEM. The wells were incubated with the following culture media: 1) DMEM + 5% patient serum from samples obtained before and at previously described times after atherectomy (n = 5); 2) DMEM + 5% patient serum + a neutralizing antibody (R. & D. Systems) to PDGF (10 μg/ml) (n = 5); 3) DMEM + 5% patient serum + a neutralizing antibody (R. & D. Systems) to bFGF (10 μg/ml) (n = 3); and 4) DMEM + 5% patient serum + neutralizing antibodies to both PDGF and bFGF (n = 3). The dose of PDGF and bFGF antibodies used was 10 times that required to produce neutralization according to the manufacturer. Antibodies were stated to be specific for PDGF and bFGF and with minimal cross-reactivity with other cytokines. Cells were also incubated with pooled control serum from human donors. 3H-Thymidine (0.5 μCi/well) was pulsed between 24 and 48 h after release from growth arrest. Cells from triplicate wells were harvested with trypsin at 48 h onto glass fiber filters. The filters were dried overnight, immersed in scintillation fluid and analyzed in a beta-counter. Counts were expressed as cpm/well.
1.6 Immunocytochemical Analysis.
Frozen sections (5 μm) were obtained from one atherectomy tissue fragment from each patient. Tissue was lightly fixed in cold acetone and incubated overnight at 4°C with primary antibodies to PDGF A, PDGF B (Genzyme), bFGF (Santa Cruz) and mouse immunoglobulin G (Amersham) over a concentration range of 0.1 to 2.0 μg/ml in phosphate buffer containing 1% bovine serum albumin. A secondary anti-mouse biotinylated system with streptavidin–horseradish peroxidase and 3,3-diaminobenzidine for visualization was used. Sections were counterstained with hematoxylin.
Results are expressed as mean value ± SEM. Clinical and angiographic characteristics were compared using a Wilcoxon signed rank test. Two-way analysis of variance with repeated measures and the Scheffé S test were used to test differences between levels of plasma PDGF AB, PF4 or beta-TG at each time point after angioplasty and the preangioplasty level (). An unpaired ttest was used for comparison between plasma values for control subjects and for angioplasty-group patients before angioplasty. For serum mitogenicity experiments, analysis of variance and the Scheffé S test were used to test differences between 3H-thymidine incorporation at each time point. An unpaired ttest was used for comparison between pooled donor serum values and preatherectomy values; p < 0.05 was considered significant for all statistical tests.
2.1 Clinical Data.
A summary of clinical data is shown in Table 1. There was no significant difference in coronary risk factors between control, angioplasty and atherectomy groups. No subject had diabetes mellitus. In subjects undergoing PTCA, stenosis severity was reduced from 84 ± 2% to 22 ± 2%. In those undergoing atherectomy, stenosis was reduced from 84 ± 2% to 16 ± 5%.
2.2 Coronary Sinus Plasma PDGF AB, PF4, beta-TG.
Plasma PDGF A and B, PF4 and beta-TG concentrations after angioplasty are shown in Fig. 1. Control patients had significantly lower (p < 0.01) plasma levels of PDGF AB, PF4 and beta-TG than angioplasty-group patients before PTCA, indicating a low level of platelet activation in the control group. After angioplasty, there was a twofold increase in PDGF AB levels that peaked at 5 min and returned to pre-PTCA levels by 30 min. Both PF4 and beta-TG continued to be elevated during the 30 min after angioplasty, although both levels dropped over time.
2.3 Coronary Sinus Serum Mitogenicity.
The mitogenicity of serum obtained from atherectomy-group patients and from human donors is shown in Fig. 2. The mitogenicity of serum obtained before atherectomy, measured by VSMC 3H-thymidine incorporation, was low and similar to that of pooled donor serum (1,942 ± 275 vs. 1,092 ± 42 cpm/well, n = 5 for each group). However, serum obtained at each time point after atherectomy produced a sequential sixfold increase in 3H-thymidine incorporation (3,250 ± 378 cpm/well at 5 min vs. 11,603 ± 324 cpm/well at 30 min, p = 0.01, n = 5).
2.4 Effect of Neutralizing Antibodies to PDGF and bFGF on Mitogenicity of Serum.
Neutralizing antibody to PDGF caused a 20% reduction in the mitogenicity of serum obtained at each time point after atherectomy. Neutralizing antibody to bFGF caused a 40% reduction in the mitogenicity of the same serum after atherectomy. In the presence of PDGF and bFGF neutralizing antibodies combined, the mitogenicity of serum obtained at each time point after atherectomy was reduced by ∼65% (Fig. 2).
2.5 Localization of PDGF A and B and bFGF Within Plaque.
PDGF A and B staining was seen in atherectomy specimens from all patients. PDGF A staining was localized to the subendothelial and intimal cells, with little staining of the endothelium. PDGF B staining was strongly localized to endothelial cells as well as to the cytoplasm of some subendothelial cells. There was also extracellular staining of PDGF B. bFGF staining was prominent in the subendothelial region of all plaque specimens (Fig. 3).
3.1 Coronary Sinus Plasma PDGF AB and Platelet Activation Markers.
The first observation from our study was that baseline pre-PTCA plasma levels of PDGF AB and platelet activation markers were elevated in angioplasty-group patients with coronary disease compared with those in control subjects. After PTCA a further twofold elevation in plasma PDGF AB was seen, which returned to pre-PTCA levels 30 min after PTCA. High plasma PF4 and beta-TG levels were maintained after angioplasty compared with control levels. Repeated sampling of blood in control patients over the same time period as that in patients undergoing angioplasty was not possible. The measures of platelet activation after angioplasty differ in this study compared with those of Scharf et al. () who demonstrated a drop in PF4 levels and an elevation in beta-TG levels after angioplasty. However, in the study group of Scharf et al., preangioplasty levels of platelet activation were very low. Because PF4 is known to bind more strongly to matrix components (i.e., heparan sulfate proteoglycans) within the vessel wall than PDGF, it is unlikely that PDGF AB and PF4 are equivalent in their distribution between tissue and plasma compartments downstream from the angioplasty site. Therefore, PF4 levels in the coronary sinus cannot be linked to the extent of platelet release of PDGF AB at the site of injury. However, sustained elevation of beta-TG in our study suggests that ongoing platelet activation contributed to PDGF AB release. Release of PDGF AB from other sources, such as the vessel wall, cannot be excluded.
3.2 Coronary Sinus Serum Mitogenicity and Source of Serum Mitogens.
The main observation from our study was that the mitogenicity of serum derived from the coronary sinus over the first 30 min after atherectomy increased sixfold compared with preatherectomy or control donor serum. The mitogenic effect of each patient’s serum on his or her own VSMCs in culture was due predominantly to PDGF and bFGF mitogens. Autologous serum was used for culture experiments because in our experience single-patient serum was toxic to VSMCs derived from other human subjects. Pooled control donor serum does not have these problems and has been used in a previous study (). Autologous serum coincubated with the patient’s own VSMCs may also better mimic conditions in vivo. PDGF A, PDGF B and bFGF were demonstrated using immunocytochemical analysis in the subendothelial and endothelial regions of plaque obtained at the time of atherectomy, and these findings are consistent with other immunocytochemical studies ([6, 7]).
In the present study, growth factor release may have been from activated platelets, as indicated by our observations in the first series of experiments or from damaged endothelial () and intimal smooth muscle cells (). Serum was used instead of plasma for the mitogenic assay because plasma produced coagulation of the VSMC culture medium. Serum derived from clotted blood contains products of maximal platelet activation that are present in each of the samples. It is therefore likely that the increased serum mitogenicity after intervention is from a source other than platelets, possibly the vessel wall. Crowley et al. () recently demonstrated, using immunoblot analysis and neutralization studies, that mechanically injured bovine VSMCs in culture immediately released biologically active growth factors bFGF, PDGF and epidermal growth factor (EGF), presumably from preformed stores in the cell cytoplasm. Heparin, which was given to all patients undergoing atherectomy during the procedure, is also known to release PDGF and bFGF from matrix-bound stores, especially in injured vessels without an endothelium ([17–20]). PDGF may also be released from endothelial cells by proteases such as thrombin, which may be a mechanism for active mobilization of growth factors by injured endothelium ().
Our data provide evidence of first-phase growth factor release in human subjects after vascular injury. PDGF and bFGF appear to be released in sufficient quantities within the first 30 min after atherectomy to be strongly mitogenic to each patient’s VSMCs. It is possible that the neutralizing antibodies used blocked autocrine PDGF or bFGF produced by the explanted smooth muscle cells. The low level of mitogenicity obtained with control serum-treated cells suggests that autocrine growth factor production was small and that postatherectomy serum-derived growth factors provided the bulk of the increased mitogenic stimulus. The mitogenic effects of serum collected after atherectomy were not completely blocked by antibodies to PDGF and bFGF, suggesting that mitogens other than PDGF and bFGF are present. Factors that might have contributed to this include mitogens, such as EGF and transforming growth factor-beta, as well as nonpeptide factors, such as serotonin and adenosine diphosphate, which are known to be present in the coronary sinus blood after angioplasty (). These factors may have synergistic effects maximizing the cellular growth response to locally released mitogens (). A synergism between increased levels of these factors and autocrine PDGF or bFGF, or both, cannot be fully excluded.
3.3 Study Limitations.
There are limitations in the interpretation of the results of our study. In control subjects it was not felt ethically justifiable to examine the effects of coronary sinus catheterization over a similar time period to that in patients undergoing angioplasty. Strict time controls are therefore lacking in this study. We assumed that serum obtained from the coronary sinus reflects events occurring at the site of atherectomy. It is likely the degree of platelet degranulation and growth factor release at the site of injury was underestimated by measurement of the concentration of mitogen downstream in the coronary sinus. Increased flow within the coronary circulation after stenosis dilation may cause these substances to be further diluted downstream. We measured the spillover of mitogens into coronary sinus blood that were either released as a consequence of the injury or were weakly bound within the atherosclerotic plaque and washed out as a result of exposure after the atherectomy. We can only speculate that similar or greater concentrations of growth factors are available locally at the site of injury or within the vessel wall. Assessment of serum mitogenicity after atherectomy involved measurement of DNA synthesis during serum incubation with VSMCs for 48 h. Although the index of serum mitogenicity remained elevated at 30 min after atherectomy, it is not possible to determine whether these mitogens are elevated in vivo for sufficiently long periods to initiate VSMC proliferation at the lesion site. Continued sampling of coronary sinus blood over a longer period would help to address this question. Withdrawal of blood through a long catheter may cause platelet activation (). The use of heparin in our study and the low levels of platelet activation markers in control plasma suggest that significant platelet activation was not induced during sampling.
Characterization of coronary sinus plasma containing elevated levels of PDGF AB after angioplasty and serum showing strong mitogenic properties after atherectomy is consistent with first-phase growth factor release as described in the balloon injury animal model. Early release of biologically active growth factors, even in small quantities, may act locally in a synergistic fashion to produce a significant mitogenic signal (). The present study provides a basis for the recently described reduction in restenosis rate after PTCA after the use of the PDGF receptor blocking drug trapidil (). It indicates that a combined approach to blocking both PDGF and bFGF may have added benefit in reducing neointimal proliferation.
We thank Diana Steenberg for preparation of the manuscript and Maria Kay for assistance in sample collection.
☆ Dr. Caplice was supported by the National Heart Foundation of Australia, Canberra, Australian Capital Territory and the Prince Charles Hospital Private Practice Research and Education Trust Fund, Brisbane, Queensland, Australia.
This study was presented in part at the Annual Scientific Meeting of the Cardiac Society of Australia and New Zealand, Canberra, Australian Capital Territory, August 1995.
- basic fibroblast growth factor
- Dulbecco’s modified Eagle’s medium
- epidermal growth factor
- platelet factor 4
- platelet-derived growth factor
- percutaneous transluminal coronary angioplasty
- vascular smooth muscle cells
- Received April 15, 1996.
- Revision received February 14, 1997.
- Accepted February 26, 1997.
- The American College of Cardiology
- Nobuyoshi M,
- Kimuna T,
- Nosaka H,
- et al.
- Johnson DE,
- Hinohara T,
- Selmon MR,
- Braden LJ,
- Simpson JB
- Schwartz SM,
- de Blois D,
- O’Brien ER
- Barrett TB,
- Benditt EP
- ↵Wilcox JN, Smith KM, Williams LT, Schwartz SM, Gordon D. Platelet-derived growth factor mRNA detection in human atherosclerotic plaques by in situ hybridization. J Clin Invest 1988;82:1134–3.
- Hughes SE,
- Crossmann D,
- Hall PE
- Ferns GA,
- Raines EW,
- Sprugel KH,
- et al.
- Lindner V,
- Reidy MA
- Liu MW,
- Roubin GS,
- Robinson KA,
- et al.
- Maresta A,
- Barducelli M,
- Cantini L,
- et al.
- Fisher LD,
- van Belle G
- Scharf RE,
- Power A,
- Marzec UM,
- et al.
- Dartsch PC,
- Voisard R,
- Bauriedel G,
- Hofling B,
- Betz E
- Soyombo AA,
- Di Corleto PE
- Crowley ST,
- Ray CJ,
- Nawaz D,
- Majack RA,
- Horwitz LD
- Raines EW,
- Ross R
- Khachigian IM,
- Owensby DA,
- Chesterman CN
- Kelly JL,
- Sanchez A,
- Brown GS,
- Chesterman CN,
- Sleigh MJ
- Nichols AB,
- Owen J,
- Grossman BA,
- et al.