Vascular Smooth Muscle Cell Cytotoxicity and Sustained Inhibition of Neointimal Formation by Fibroblast Growth Factor 2–Saporin Fusion Protein
Abstract Basic fibroblast growth factor (FGF2) is an important mediator of smooth muscle cell (SMC) proliferation following arterial injury that results in neointimal growth. The present study was designed to explore the effects of recombinant FGF2 linked to the ribosome-inactivating protein saporin-6 (rFGF2-SAP) on vascular SMC cytotoxicity and neointimal formation following arterial injury. Cultured rat aortic SMCs were exposed to various concentrations of rFGF2-SAP, FGF2, and saporin-6 (SAP). Incubation with rFGF2-SAP resulted in a decreased number of SMCs beginning at a concentration of 10−9 mol/L. Significant cytotoxicity was observed with as little as a 30-minute exposure of SMCs to rFGF2-SAP. To evaluate the ability of rFGF2-SAP in an in vivo model to reduce neointimal formation, Sprague-Dawley rats underwent carotid artery balloon denudation and received an intravenous bolus of vehicle or 5, 10, 15, or 20 μg/kg rFGF2-SAP on 0, 3, 6, and 9 days after injury. Rats were euthanized at 14 days, and carotid arteries were analyzed by computerized morphometry. The threshold dose for a significant reduction in neointimal area by rFGF2-SAP was 15 μg/kg (47% reduction in neointima). When dosing was extended to include days 16, 19, and 22, the neointima was reduced 33% at 28 days (P=.048). rFGF2-SAP reduced neointima without associated medial thinning or arterial wall dilatation. To determine if rFGF2-SAP directly targets SMCs in vivo, rats underwent carotid injury and received either 15 μg/kg rFGF2-SAP or vehicle on day 0 and at 72 hours, with euthanasia at 78 hours after balloon denudation. Medial SMC number was reduced 46% in the rFGF2-SAP group. Tissue sections from arteries 3 days after balloon injury demonstrated rFGF2-SAP binding to medial SMCs and adventitial cells. Staining for fibroblast growth factor receptor 1 revealed a high level of expression in ballooned arteries 3 and 14 days after injury. Taken together, these results provide a molecular and cellular basis for the observed specificity. Prolonged delivery of rFGF2-SAP can affect the natural history of arterial repair after injury.
The pathophysiology of restenosis involves the complex interaction of cellular elements (platelets, inflammatory cells, endothelial cells, and SMCs) with cytokines and growth factors. Previous experimental studies have suggested that FGF2 is an important mediator of the vascular response following arterial injury. Injured endothelial cells release FGF2.1 2 In the rat carotid artery, an infusion of exogenous FGF2 resulted in a marked increase in medial SMC proliferation measured 41 hours and 2 weeks after balloon arterial injury.3 The mitogenic effect of FGF2 on SMCs appears to require arterial injury, since exogenous FGF2 caused no increase in SMC proliferation when the endothelial surface was intact.3 4 Antibody to FGF2 administered before balloon injury in the rat was associated with an 80% reduction in medial smooth muscle proliferation at 41 hours.5 Although FGF2 has been shown to be an important mediator of medial SMC replication early and late after injury, Olsen et al6 were unable to show a reduction in SMC proliferation when the antibody to FGF2 was infused 4 and 5 days after injury, suggesting that other growth factors mediate the response as well. SMCs that constitute the intimal lesion may remain a target for stimulation by FGF2, since mRNA for FGFR1 has been shown to be present on replicating and quiescent intimal SMCs 5 days and 6 weeks, respectively, after injury.7
The novel strategy of cytotoxins linked to growth factors to inhibit SMC proliferation has been demonstrated in in vitro studies using a mutated Pseudomonas exotoxin linked to transforming growth factor-α.8 9 SAP, an alternative cytotoxin, is a ribosome-inactivating protein that cleaves adenine from ribose in the 28S RNA of the 60S subunit.10 SAP enzymatically inhibits protein synthesis by preventing binding of elongation factor 2 to the 60S subunit, resulting in cell death.11 SAP is nontoxic to cells when it remains extracellular, and mammalian cells have no receptors for the plant protein. When SAP is chemically conjugated to FGF2 to produce FGF2-SAP and internalized by cells expressing FGF receptors, FGF2-SAP acts as a mitotoxin and is lethal to cells. The cytocidal properties of FGF2-SAP have been used to inhibit solid tumor growth12 and lens reepithelialization.13 Previous short-term studies using the chemical conjugate of FGF2 and SAP in injured rat carotid arteries demonstrated reduced numbers of medial SMCs and inhibition of intimal growth,3 14 but optimal concentrations, optimal duration of dosing, and the duration of the inhibitory response were not established. Furthermore, a recombinant fusion protein, rFGF2-SAP (a homogenous single-chain monomer compared with the more heterogeneous and variable chemical conjugate), has been developed15 but has not been evaluated in a balloon injury experimental model.
The objectives of the present study were (1) to further investigate the kinetics of cytotoxicity using the rFGF2-SAP fusion protein in cultured rat aortic SMCs, (2) to identify dosing requirements for sustained inhibition of neointimal proliferation in the rat carotid artery balloon injury model, and (3) to identify targets of rFGF2-SAP in the damaged arterial wall.
Materials and Methods
Preparation and Purification of rFGF2-SAP Fusion Protein
rFGF2-SAP was expressed and purified as described by Lappi et al15 with modifications by McDonald et al.16 In brief, the soluble mitotoxin expressed in Escherichia coli was purified to homogeneity from cell lysates via expanded bed absorption chromatography followed by cation-exchange, heparin-affinity, and size-exclusion chromatography. The purified protein was formulated in 50 mmol/L sodium citrate, pH 6, containing 80 mmol/L NaCl and 0.1 mmol/L EDTA. The final materials were shown to be >98% pure by SDS-PAGE, reverse-phase HPLC, and size-exclusion HPLC and were stored at −80°C until used. Endotoxin levels were <0.05 EU/mg protein.
Cytotoxicity of rFGF2-SAP on Rat Aortic SMCs in Culture
Rat aortic SMCs were prepared from adult female Sprague-Dawley rats (Hazelton, Pa) weighing 200 to 250 g by the explant method as previously described.17 Briefly, explants were obtained from the thoracic aorta, the adventitial layer was dissected away with a scalpel blade, and the endothelium was removed with a cotton swab. The tissue explants were maintained in medium 199 supplemented with 10% FBS, 4 mmol/L l-glutamine, 100 U/mL penicillin G sodium, and 100 μg/mL streptomycin sulfate in a humidified atmosphere of 5% CO2/95% air. The medium was replaced every third day. SMCs were allowed to grow out from the tissue, which was subsequently removed after 9 to 11 days. After confluence, the cells were subcultured using 0.05% trypsin and 0.053 mmol/L EDTA. Cells were routinely subcultured at a 1:5 ratio and used between passages 4 and 6.
SMCs were plated in 24-well tissue culture plates (10 000 cells per well) containing growth arrest medium (medium 199) containing 1 mmol/mL insulin, 5 mg/mL transferrin, and 0.1% bovine serum albumin. The following day, SMCs were exposed, for 72 hours, to media alone, rFGF2-SAP, SAP, or FGF2 at concentrations of 10−11 to 10−7 mol/L. At the end of the incubation period, SMCs were washed in PBS and trypsinized. Cell number was determined with a Coulter particle counter. Experiments were performed in duplicate and repeated three times.
Effect of Incubation Time on Cytotoxicity
SMCs were plated as described above. The following day, SMCs in duplicate wells were incubated for 0.5, 1, 4, or 72 hours in rFGF2-SAP at concentrations of 10−11 to 10−7 mol/L. SMCs in media alone served as a control. At the end of the incubation period, SMCs were rinsed in PBS, and wells were filled with medium. Cell number was determined 72 hours after initial exposure to rFGF2-SAP. For each incubation time, a line plot was generated that related the percent reduction in SMC number to the concentration rFGF2-SAP. The ID50 was defined as the inhibitory dose of rFGF2-SAP needed to reduce the average cell number of duplicate wells by 50%. This experiment was performed three times.
The Effect of rFGF2-SAP on Neointimal Formation Following Arterial Injury
Anesthetized (ketamine/xylazine) male Sprague-Dawley rats (250 to 350 g) underwent left carotid artery endothelial injury by three passes of an inflated (0.3-mL) 2F Fogarty balloon catheter introduced via the external carotid artery. Animals (n=51) were randomized to receive 5, 10, 15, or 20 μg/kg of rFGF2-SAP diluted in phosphate-buffered saline containing 0.1% bovine serum albumin or an equal volume of vehicle (phosphate-buffered saline containing 0.1% bovine serum albumin) in control rats on days 0, 3, 6, and 9 after balloon denudation. rFGF2-SAP and vehicle were administered intravenously via the tail vein. Pulse dosing every 3 days was chosen in an effort to widen the therapeutic window for rFGF2-SAP. Rats were euthanized 14 days after balloon denudation and perfusion-fixed with Histochoice fixative (Amresco). Twenty-four hours and 1 hour before euthanasia, rats received an intraperitoneal injection of BrdU (30 mg/kg). One hour before euthanasia, 0.3 mL Evans blue dye was injected intravenously to mark the denuded arterial segment.
Carotid arteries were removed and immersion-fixed with Histochoice for 24 hours. The entire carotid artery was dehydrated in a series of graded alcohols, cleared with xylene, sectioned into 2.0-mm segments, embedded in paraffin, cut at 4 μm, and stained with hematoxylin-eosin and Movat pentachrome stains. All arterial segments showed intimal thickening, and segments were magnified and digitized. Computerized morphometry (IP Lab Spectrum software) was performed (with the observer blinded to treatment group) on arterial segments to measure the area within the EEL, IEL, media, and neointima. Proliferating intimal and medial SMCs were identified after incubating the slides with a monoclonal antibody directed against BrdU. The total number intimal and medial cells (BrdU positive and negative) were counted in eight randomly selected oil immersion fields (×400) from the arterial segment that demonstrated the largest neointimal area.
In a set of experiments to establish the duration of mitotoxin effect, rats (n=42) were randomized to treatment with intravenous 15 μg/kg rFGF2-SAP or vehicle on days 0, 3, 6, and 9 after carotid injury, with euthanasia on day 28. Another group of rats received either intravenous 15 μg/kg rFGF2-SAP or vehicle on days 0, 3, 6, 9, 16, 19, and 22, with euthanasia 28 days after carotid artery injury. Perfusion fixation, artery harvest, tissue processing, and analysis of sections were performed as described above.
Assessment of Medial SMC Toxicity
Male Sprague-Dawley rats underwent carotid balloon endothelial denudation as described above. Animals (n=14) were randomized to receive 15 μg/kg of rFGF2-SAP or vehicle intravenously or vehicle at 0 and 72 hours with euthanasia at 78 hours. On the day of arterial injury and daily until euthanasia, rats received an intraperitoneal injection of BrdU (30 mg/kg). One hour before euthanasia, 0.3 mL Evans blue dye was injected intravenously. Rats were perfusion-fixed, and carotid arteries were processed for light microscopy as described above. The arterial segment demonstrating the greatest degree of medial injury was selected with the observer blinded to the treatment. The section was magnified (×400), and all medial SMC nuclei (BrdU positive and negative) were counted. IEL, EEL, and medial areas were measured by computerized morphometry. Eight randomly selected high-power (×400) fields from the adventitia were digitized, and all adventitial cells (fibroblasts, neutrophils, macrophages, and lymphocytes) within the field were counted. The area of the adventitia was measured, and adventitial cell density (cells/mm2) was calculated. All medial SMC nuclei were counted from an arterial segment from the central portion of the uninjured right carotid artery.
rFGF2-SAP Binding to Tissue Sections
Direct binding of rFGF2-SAP and SAP was performed on sections from a separate group of rat arteries (n=4) obtained 3 days after injury; sections of an uninjured artery from the same animal were used as controls. An antibody against SAP was raised in goats (Chemicon International, Inc) and purified using a protein A column. Frozen sections cut at 4 μm were allowed to dry overnight at room temperature. Sections were hydrated in PBS and incubated in either 0.2 μg/mL rFGF2-SAP or SAP alone for 45 minutes at 37°C. The sections were washed in PBS and incubated with 15 μg/mL of goat anti-SAP for 45 minutes at 37°C to detect bound rFGF2-SAP or SAP. The goat antibody was detected with peroxidase-conjugated donkey anti-goat IgG (Jackson ImmunoResearch) diluted 1:100. The rFGF2-SAP, SAP, and primary and secondary antibodies were diluted in PBS containing 0.2% gelatin from cold water fish skin (Sigma Chemical Co) as a protein carrier. Diaminobenzidene tablets (SigmaFast, Sigma) were used as the chromogen.
FGF Receptor Immunohistochemistry
Immunohistochemistry was performed on Histochoice-fixed, paraffin-embedded, 4-μm-thick sections obtained from tissues collected on days 3 and 14 after arterial injury (n=4 at each time point). A rabbit polyclonal antibody to the human FGFR1 was purchased from Santa Cruz Biotechnology. This antibody was raised against C-terminal peptide residues 808 to 822, PRHPAQLANGGLKRR. Immunostaining was performed on the TekMate 2000 automated immunostainer (BioTek Solutions), as described previously.18 Sections were placed in sodium citrate buffer and heated in the microwave oven for antigen retrieval. The primary antibody was diluted in PBS containing 0.2% gelatin from cold water fish skin as a protein carrier. The primary antibody was detected using the ABC kit provided by BioTek Solutions. Cognate peptides were used to demonstrate staining specificity. A second monoclonal FGFR1 antibody, generated against the extracellular domain of FGFR1, confirmed the specificity of the staining.
Data are expressed as mean±SEM. Differences between groups were compared using an unpaired Student’s t test. A two-tailed probability of P<.05 was considered significant.
Cytotoxicity of rFGF2-SAP on Rat Aortic SMCs
The incubation of SMCs with varying concentrations of rFGF2-SAP for 72 hours resulted in a significant decrease in SMC number that was first detected at a concentration of 10−9 mol/L. At higher concentrations of rFGF2-SAP (10−7 mol/L), 90% SMC cytotoxicity was achieved. SAP alone did not reduce SMC number from a concentration of 10−11 to 10−8 mol/L (Fig 1⇓).
The efficacy of rFGF2-SAP–induced SMC cytotoxicity in vitro was also dependent on the duration of exposure (Fig 2⇓). Concentrations of rFGF2-SAP of ≈22 nmol/L were effective in producing 50% SMC death within 30 minutes of exposure to the drug, indicating that a prolonged exposure is not required to kill SMCs. Increasing the exposure time to rFGF2-SAP (0.5 to 72 hours) markedly reduced the concentration of rFGF2-SAP required to produce 50% lethal cell injury. For example, a 72-hour incubation time resulted in an ≈7-fold decrease in the ID50 of rFGF2-SAP compared with the 30-minute time period.
The Effect of rFGF2-SAP on Neointimal Formation and Cellular Proliferation Following Arterial Injury
Pulse dosing every third day (days 0, 3, 6, and 9) with rFGF2-SAP (5, 10, 15, or 20 μg/kg) resulted in a significant (47%, P<.03) reduction in neointimal formation in rats receiving drug treatment at 15 and 20 μg/kg (Figs 3⇓ and 4⇓). rFGF2-SAP did not affect the areas enclosed by the EEL or IEL or the area of the media assessed 14 days after injury (Fig 3⇓). Medial repair was complete in carotid arteries 14 days after balloon injury in rats receiving rFGF2-SAP; only occasional foci of medial SMC loss were present, and no acute cellular necrosis or inflammation or aneurysm formation was present. rFGF2-SAP was well tolerated by rats through 15 μg/kg, with no clinical evidence of toxicity. There was a 40% mortality (hepatic toxicity) in rats receiving 20 μg/kg, occurring after a mean of three doses of rFGF2-SAP was administered. In rats receiving 15 μg/kg rFGF2-SAP, the BrdU labeling index at day 14 (Fig 5A⇓) in the neointima was similar to that in control rats (6.7±1.5% versus 5.5±1.4%, respectively; P=NS). There was a trend toward greater SMC proliferation in the media (1.8±0.8%) in rFGF2-SAP–treated rats compared with control rats (0.4±0.2%, P=.076; Fig 5A⇓). Cell density (cells/mm2) in the neointima (9116±510 versus 8722±712, P=NS) and in the media (3185±180 versus 3098±159, P=NS) was similar in rFGF2-SAP–treated and vehicle-treated animals, respectively. Since the amount of neointima was significantly less in rFGF2-SAP–treated rats, the total SMC number was correspondingly less as well.
To determine whether the inhibition of neointimal formation observed at 2 weeks was maintained at 4 weeks after surgery, rats were randomized to treatment with intravenous 15 μg/kg rFGF2-SAP or vehicle on days 0, 3, 6, and 9 (short-term treatment) or days 0, 3, 6, 9, 16, 19, and 22 (long-term treatment) after carotid injury, with euthanasia on day 28. There was no reduction in neointimal area in the rFGF2-SAP group (0.136±0.006 mm2) versus the control group (0.126±0.010 mm2) with short-term treatment (Fig 6A⇓). Medial areas were similar in both groups (0.152±0.007 versus 0.151±0.003 mm2), and there were no significant differences in the areas enclosed by the IEL (0.437±0.017 versus 0.454±0.011 mm2) and EEL (0.589±0.022 versus 0.605±0.013 mm2) in rFGF2-SAP–treated and vehicle-treated rats, respectively. The neointimal BrdU labeling index (Fig 5B⇑) was similar in rFGF2-SAP–treated rats (2.4±0.7%) compared with control rats (1.9±0.6%, P=NS). There was no difference in the medial BrdU labeling index (0.5±0.2% in rFGF2-SAP–treated rats versus 0.9±0.5% in control rats, P=NS; Fig 5B⇑).
With long-term administration of rFGF2-SAP, there was a significant (P=.048) reduction in neointimal area in the rFGF2-SAP–treated rats (0.068±0.012 mm2) compared with control rats (0.102±0.010 mm2, Fig 6B⇑). There were no differences in medial area (0.145±0.007 versus 0.148±0.005 mm2) or areas enclosed by the EEL (0.586±0.031 versus 0.630±0.032 mm2) and IEL (0.447±0.028 versus 0.483±0.029 mm2) in the rFGF2-SAP–treated group compared with the vehicle-treated group, respectively. There were no differences in the BrdU labeling index (Fig 5C⇑) in the neointima (2.1±0.4% versus 1.4±0.5%, P=NS) or media (0.1±0.1% versus 0.05±0.05%, P=NS) between rFGF2-SAP–treated and vehicle-treated rats, respectively. Cell density (cells/mm2) in the intima (8229±695 versus 7094±438, P=NS) and in the media (3085±178 versus 2719±266, P=NS) was similar in rFGF2-SAP–treated and vehicle-treated animals, respectively. These results indicate that there were fewer neointimal SMCs at 28 days in this dosing regimen of rFGF2-SAP compared with the control condition. There was no mortality in rats receiving rFGF2-SAP.
Is the Medial SMC the Target of rFGF2-SAP In Vivo?
To determine if rFGF2-SAP is capable of directly targeting medial SMCs, rats were subjected to balloon injury, drug administration on days 0 and 3, and euthanasia 6 hours after the last dose (78 hours). The mean number of medial SMCs in the most injured arterial segment was 148±25 cells in the rFGF2-SAP–treated rats compared with 274±19 cells in the control rats (46% reduction, P=.002; Table⇓). Expressed as SMC density per unit medial area, the mean number of medial SMCs/mm2 in the most injured arterial segment was 1150±265 cells/mm2 in rFGF2-SAP–treated rats versus 2065±236 cells/mm2 in control rats (44% reduction, P=.024). The rFGF2-SAP–treated arteries also had extensive full medial thickness SMC necrosis compared with focal necrosis in control arteries (Fig 7⇓). The mean number of BrdU-positive medial SMCs in the most injured arterial segment was 25±7 in the rFGF2-SAP–treated group versus 34±8 in the control group (26% reduction, P=.46). Notably, in the adventitia, the mean number of cells/mm2 was 246±57 in drug-treated rats versus 556±100 in control rats (56% reduction, P=.02). The mean number of medial SMCs per arterial segment in the uninjured right carotid arteries was similar in rFGF2-SAP–treated rats (257±42) compared with vehicle-treated rats (278±49, P=.41), indicating the specificity of the response to the injured arterial segment.
Identification of Binding Sites for rFGF2-SAP
rFGF2-SAP was incubated with tissue sections from ballooned and uninjured arteries harvested and processed 3 days after injury to assess rFGF2-SAP binding. Extensive binding was observed in the media (Fig 8⇓). Interestingly, adventitial cells from injured vessels also demonstrated considerable binding of rFGF2-SAP. When SAP alone was incubated with tissue sections, no staining was observed using anti-SAP antibodies, indicating that rFGF2 conferred binding specificity to SAP.
Expression of FGFR1 on Intimal and Medial SMCs After Arterial Injury
Because FGF2 can bind FGFR1 with high affinity, sections of rat arteries harvested 3 and 14 days after injury were analyzed for FGFR1 distribution. In uninjured arteries, receptor immunoreactivity was detected in SMCs localized throughout the media (Fig 9A⇓). Competition with the antigen eliminated staining (Fig 9B⇓). In arteries collected 3 days after balloon injury and vehicle treatment, FGFR1 immunoreactivity was found in the densely cellular adventitia (Fig 9C⇓) as well as in medial SMCs. In rFGF2-SAP–treated animals at 3 days, no FGFR1 immunoreactivity was detected in the adventitia, and considerably fewer cells were present in the adventitia (Fig 9D⇓). Fourteen days after injury, in both vehicle and rFGF2-SAP–treated arteries, immunoreactive FGFR1 was detected in SMCs in the neointima and media, but the adventitia was negative for FGFR1 staining and contained markedly fewer cells than at earlier time points (Fig 9E⇓ and 9F⇓). In aggregate, FGFR1 staining was less in the rFGF2-SAP–treated group because these animals had a smaller neointima and less cellularity in the adventitia.
The findings described in the present study demonstrate that the mitotoxin rFGF2-SAP inhibits the neointimal proliferation after arterial injury and extend the work of Lindner, Casscells, and colleagues.3 14 In the study by Lindner et al,3 there was a 53% reduction in medial SMCs 41 hours after injury in rats treated with 120 μg of FGF2 chemically conjugated to SAP. Casscells et al14 demonstrated a significant reduction in the ratio of the area of the intima to the media at 14 days after a single treatment with 75 μg/kg of FGF2 chemically conjugated to SAP. Initial studies with the recombinant fusion protein rFGF2-SAP have been performed in a canine carotid endarterectomy model19 and in canine arteriovenous conduits.20 Mattar et al19 demonstrated a 46% reduction in post–carotid endarterectomy intimal area at 14 days with continuous local delivery of 2 μg/kg per day rFGF2-SAP. There was a 40% reduction in the intimal area at the arteriovenous graft anastomotic site at 14 days with continuous local delivery of 2.7 μg/kg per day rFGF2-SAP.20 No previous studies have examined the acute as well as chronic biological effects of repeated doses of the rFGF2-SAP.
In the present report, we have characterized the biological response to the recombinant fusion protein, rFGF2-SAP, as opposed to the chemical conjugate that was used in previous studies.3 14 SAP alone does not inhibit SMC growth, supporting the concept that the toxin must be internalized to elicit a cytotoxic effect. The potency of the in vitro cytotoxic effect is also proportional to the duration of exposure of SMCs to rFGF2-SAP. In vivo experiments further demonstrate that pulse bolus dosing every 3 days with rFGF2-SAP inhibits neointimal formation following arterial injury. A significant reduction in neointimal size at 14 days is seen with a dose of ≥15 μg/kg administered on days 0, 3, 6, and 9 after balloon denudation. Importantly, rFGF2-SAP does not affect arterial size (area within the IEL and EEL) or medial area, and neither acute inflammation nor SMC necrosis is seen at 14 and 28 days. Necrosis of SMCs may be an important early event, since it was observed only at day 3. Sustained inhibition of neointimal formation 28 days after balloon injury required subsequent dosing (days 0, 3, 6, and 9 plus days 16, 19, and 22) of rFGF2-SAP, raising the possibility that intimal growth may ultimately “catch up” to control arteries in the absence of continued treatment.
The mechanism of rFGF2-SAP’s effect on neointimal formation in vivo is probably via its cytotoxic effect on medial SMCs. Three days after arterial denudation, the cytotoxic effect of rFGF2-SAP is seen, with a 46% reduction in medial SMC number. At 14 days (with four doses of rFGF2-SAP) and at 28 days (with seven doses of rFGF2-SAP), intimal SMC density was similar in drug-treated and control animals. Thus, the smaller intimal area in rFGF2-SAP–treated animals corresponds to a reduction in the total number of intimal SMCs and not cell density. The reason for the trend toward increased medial SMC proliferation that was seen on day 14 after four doses of rFGF2-SAP is uncertain and may reflect rebound from the cytotoxic effects of initial drug treatment. It should be noted, however, that the 14-day time point is 5 days past the last rFGF2-SAP treatment; thus, the lack of detectable reduction in SMC proliferation (Fig 5A⇑) may have been obviated by analysis nearer to the time of dosing, since marked effects were observed at day 3 when animals were exposed to the drug on days 0 and 3. Further, these proliferating cells may be targets for cytotoxicity by subsequent doses (on days 16, 19, and 22) of rFGF2-SAP, resulting in the chronic inhibition observed at day 28. This finding may explain why further doses of rFGF2-SAP are required to ultimately suppress neointimal growth at 28 days.
The results of the present study showing sustained inhibition of neointimal growth with rFGF2-SAP are supportive of an important role of FGF2 beyond the early post–arterial injury period. These data differ from previous work by Lindner and Reidy,5 in which a polyclonal antibody directed against FGF2, administered immediately before arterial injury, inhibited SMC proliferation measured 41 hours after injury but produced no significant reduction in intimal area at 8 days. This same antibody to FGF2, given 4 and 5 days after injury, resulted in no significant reduction in intimal SMC proliferation.6 These findings suggested that FGF2 influences the proliferative response early but has a lesser role in the continued intimal SMC proliferation that occurs 4 to 14 days after injury.6 In contrast, the present study demonstrates that repeated injections of rFGF2-SAP produce a smaller neointima at 14 and 28 days, suggesting that SMCs continue to respond to FGF2 signals beyond the peak period of SMC proliferation (0 to 4 days) in the long-term neointimal responses to injury. These different results may reflect different experimental approaches (antibody to FGF2 versus FGF2-directed cytotoxicity with rFGF2-SAP) to reduce neointimal growth.
The present study also shows that the SMCs that migrate and proliferate within the injured arterial segment express FGFR1. Intense FGFR1 expression seen on intimal and medial SMCs through at least up to 14 days after injury may explain the long-term suppression of neointimal growth with repeated dosing of rFGF2-SAP. Four classes of FGF receptors have been identified.21 mRNA for FGFR1 has been shown to be present on endothelial cells 8 hours after injury and on both replicating and quiescent SMCs 5 days and 6 weeks, respectively, after injury.7 Casscells et al14 demonstrated binding of radiolabeled FGF2 to medial SMCs 2 days after injury and on neointimal SMCs at 4 days; by immunohistochemistry, there was positive antibody staining for FGFR1 on neointimal SMCs at 14 days. Together with earlier results, finding FGFR1 on medial and intimal SMCs at 3 and 14 days and on adventitial cells on day 3 in the present study suggests that these migrating and proliferating cells are targets for rFGF2-SAP and that reduced neointimal size is due to internalization of the cytotoxin and cell death during the period of intimal growth.
There was a significant (56%) reduction in adventitial cellularity at 3 days in rFGF2-SAP–treated animals. This adventitial effect may also be important in neointimal growth inhibition because adventitial cells also possess FGFR1 early after injury (Fig 9C⇑) and may be an additional source of growth factors and cytokines. The diminished adventitial FGFR1 staining and cellularity observed in rFGF2-SAP–treated arteries on day 3 suggests that rFGF2-SAP affects cells besides SMCs involved in the responses to vascular injury, particularly inflammatory cells, and that these adventitial cells play an important role in the regulation of neointimal growth.
The effect of rFGF2-SAP also appeared to be specific for injured arteries, since medial SMC numbers were similar in noninjured arteries in rats that received rFGF2-SAP or vehicle. Our results and those of others7 indicate that uninjured SMCs also express FGFR1 yet are unresponsive to exogenous rFGF2-SAP or exogenous FGF2 until they are injured. The reason for this lack of response despite receptor expression is unclear but may be secondary to receptor upregulation in the injury response. Positive intracellular staining with an antibody to FGF2 has been shown in uninjured arterial SMCs,6 and FGF2 has been shown to be localized in the nucleus.5 22 FGF2 is not secreted via classical signal transduction pathways23 but is believed to be released via a distinct export pathway24 and in the setting of cell injury or death.6 A reservoir of FGF2 may also be present in the extracellular connective tissue matrix,25 26 27 and extracellular FGF2 may increase via the release of intracellular stores as a result of SMC injury.28 Thus, although uninjured quiescent SMCs contain FGF2 and express FGF receptors, they remain unresponsive to stimuli until injured.
The data in the present study suggest that a targeted therapy for reducing SMC number and neointimal formation may be accomplished by using a mitotoxin such as rFGF2-SAP. However, the results also support the hypothesis that therapies that limit restenosis in the short term may only delay neointimal formation; in order to achieve sustained inhibition of neointimal growth, prolonged contact of drug with the previously damaged arterial wall may be necessary. It should be noted that in the present report, the optimal dose of rFGF2-SAP for inhibition of neointima formation was not identified, since higher doses resulted in systemic toxicity. Major targets of toxicity include the liver and kidney, the sites of metabolism and elimination, respectively (authors’ unpublished data, 1996). Thus, a shorter duration of treatment with a higher local concentration may result in sustained suppression of neointimal growth. In addition, local delivery of rFGF2-SAP would also be expected to have a favorable influence on the therapeutic index.
Selected Abbreviations and Acronyms
|EEL||=||external elastic lamina|
|FGF2||=||basic fibroblast growth factor|
|FGFR1||=||fibroblast growth factor receptor 1|
|IEL||=||internal elastic lamina|
|SMC||=||smooth muscle cell|
Reprint requests to Renu Virmani, MD, Department of Cardiovascular Pathology, Armed Forces Institute of Pathology, Washington, DC 20306-6000.
The opinions and assertions contained herein are the private views of the authors and are not to be construed as official or reflecting the views of the Department of the Army or the Department of Defense.
- Received November 7, 1996.
- Accepted December 30, 1996.
- © 1997 American Heart Association, Inc.
McNeil P, Muthukrishnan L, Warder E, D’Amore PA. Growth factors are released by mechanically wounded endothelial cells. J Cell Biol.. 1989;109:811-822.
Lindner V, Lappi DA, Baird A, Majack RA, Reidy MA. Role of basic fibroblast growth factor in vascular lesion formation. Circ Res.. 1991;68:106-113.
Edelman ER, Nugent MA, Smith LT, Karnovsky MJ. Basic fibroblast growth factor enhances the coupling of intimal hyperplasia and proliferation of vasa vasorum in injured rat arteries. J Clin Invest.. 1992;89:465-473.
Lindner V, Reidy MA. Proliferation of smooth muscle cells after vascular injury is inhibited by an antibody against basic fibroblast growth factor. Proc Natl Acad Sci U S A.. 1991;88:3739-3743.
Lindner V, Reidy MA. Expression of basic fibroblast growth factor and its receptor by smooth muscle cells and endothelium in injured rat arteries: an en face study. Circ Res.. 1993;73:589-595.
Epstein SE, Siegall CB, Biro S, Fu YM, FitzGerald D, Pastan I. Cytotoxic effects of a recombinant chimeric toxin on rapidly proliferating vascular smooth muscle cells. Circulation.. 1991;84:778-787.
Biro S, Siegall CB, Fu Y-M, Speir E, Pastan I, Epstein SE. In vitro effects of a recombinant toxin targeted to the fibroblast growth factor receptor on rat vascular smooth muscle and endothelial cells. Circ Res.. 1992;71:640-645.
Beitz JG, Davol P, Clark JW, Kato J, Medina M, Frackelton AR Jr, Lappi A, Baird A, Calabresi P. Antitumor activity of basic fibroblast growth factor-saporin mitotoxin in vitro and in vivo. Cancer Res.. 1992;52:227-230.
Casscells W, Lappi DA, Olwin BB, Wai C, Siegman M, Speir EH, Sasse J, Baird A. Elimination of smooth muscle cells in experimental restenosis: targeting of fibroblast growth factor receptors. Proc Natl Acad Sci U S A.. 1992;89:7159-7163.
Lappi DA, Ying W, Barthelemy I, Martineau D, Prieto I, Benatti L, Soria M, Baird A. Expression and activities of a recombinant basic fibroblast growth factor-saporin fusion protein. J Biol Chem.. 1994;269:12552-12558.
Danilenko DM, Ring BD, Tarpley JE, Morris B, Van GW, Morawiecki A, Callahan W, Goldenberg M, Hershenson S, Pierce GF. Growth factors in porcine full and partial thickness burn repair: differing targets and effects of keratinocyte growth factor, platelet-derived growth factor-BB, epidermal growth factor, and neu differentiation factor. Am J Pathol.. 1995;147:1261-1277.
Chen C, Mattar SG, Hughes JD, Pierce GF, Cook JE, Ku DN, Hanson SR, Lumsden AB. Recombinant mitotoxin basic fibroblast growth factor-saporin reduced venous anastomotic intimal hyperplasia in the arteriovenous graft. Circulation.. 1996;94:1989-1995.
Hughes SE, Hall PA. Overview of the fibroblast growth factor and receptor families: complexity, functional diversity, and implications for future cardiovascular research. Cardiovasc Res.. 1993;27:1199-1203.
Abraham JA, Mergia A, Whang JL, Tumolo A, Friedman J, Hjerrild KA, Gospodarowicz D, Fiddes JC. Nucleotide sequence of a bovine clone encoding the angiogenic protein, basic fibroblast growth factor. Science.. 1986;233:545-548.
Gonzalez AM, Buscaglia M, Ong M, Baird A. Distribution of basic fibroblast growth factor in the 18-day rat fetus: localization in the basement membranes of diverse tissues. J Cell Biol. 1990:110:753-765.
Vlodavsky I, Folkman J, Sullivan R, Friedman R, Ishai-Michaeli R, Sasse J, Klagsbrun M. Endothelial cell-derived basic fibroblast growth factor: synthesis and deposition into subendothelial extracellular matrix. Proc Natl Acad Sci U S A.. 1987;84:2292-2296.