Tranilast Prevents Activation of Transforming Growth Factor-β System, Leukocyte Accumulation, and Neointimal Growth in Porcine Coronary Arteries After Stenting
N(3,4-Dimethoxycinnamoyl) anthranilic acid (tranilast) prevents the synchronous upregulation of isoforms and receptors of the transforming growth factor (TGF)-β system after arterial injury and reduces restenosis after human coronary angioplasty. However, the effects of tranilast and the importance of the TGF-β system in stent restenosis, in which inward remodeling is unimportant but inflammatory cell stimulation of neointima formation is exaggerated, are uncertain. Boston minipigs, treated with tranilast or vehicle, were subjected to endoluminal stenting, and the expression of TGF-β1 and TGF-β3, the expression of their signaling receptors ALK-5 and TβR-II, leukocyte numbers around the stent struts, and neointima development were assessed over 28 days. Stenting greatly increased early (5-day) mRNA expression of the 2 TGF-β isoforms and their receptors. Immunohistochemical localization later showed that their concentrations were greatest in regions adjacent to stent struts, where leukocytes and collagen deposition were prevalent. Tranilast suppressed these elevations in TGF-β mRNAs and reduced their immunoreactive peptides detectable around stent struts. The accumulation of leukocytes and deposition of collagen in these regions was also greatly inhibited by tranilast. These effects were associated with a 48% reduction in maximal neointimal cross-sectional area and 43% reduction in mean neointimal cross-sectional area at 28 days (P<0.05). We conclude that tranilast suppresses neointima development after stenting, effects that can be at least partly attributed to its ability to attenuate the induction of the TGF-β system and leukocyte accumulation around stent struts.
Restenosis after coronary angioplasty and stenting remains largely an unsolved problem, resulting in significant morbidity and health care costs to the western world. Several recent studies have demonstrated a consistent reduction in restenosis after nonstent coronary intervention with N(3,4-dimethoxycinnamoyl) anthranilic acid (tranilast) that is due to its ability to attenuate inward remodeling and intimal hyperplasia.1,2⇓ Tranilast potently inhibits the synchronous upregulation of isoforms and receptors of the transforming growth factor (TGF)-β system after injury, 3 and this may help explain its antirestenotic effect in unstented vessels, inasmuch as TGF-β potentiates inward remodeling and neointima formation after angioplasty.4,5⇓ However, stents are now implanted in the vast majority of coronary interventional cases, and the role of tranilast and TGF-β in stent restenosis is unclear.
Stent restenosis differs from that after angioplasty, inasmuch as inward remodeling is unimportant, whereas inflammatory cell stimulation of smooth muscle cell (SMC) migration and extracellular matrix production is exaggerated.6,7⇓ In addition, stenting usually results in several small intimal tears at the stent struts rather than 1 large dissection. Thus, the stent struts act as a local inflammatory stimulus and facilitate SMC migration into the intima. TGF-β is likely to influence all contributors to stent restenosis: inflammatory cell infiltration, SMC migration and proliferation, and extracellular matrix production. Studies in unstented vessels have shown that TGF-β promotes neointima formation predominantly by enhancing extracellular matrix deposition but also by its ability to enhance SMC migration by elevating the expression of integrins αvβ3 on SMCs.5,8,9⇓⇓ However, TGF-β is also chemotactic for monocytes10 and can confer adhesive properties on these cells by inducing the expression of urokinase receptors and MAC-1, cell surface receptors that bind fibrin and vitronectin, promoting infiltration of the area around the stent struts.11 In other types of wound healing, infiltrating macrophages are also a significant source of local TGF-β release.12 In theory, the multifactorial role of TGF-β may be central to neointima formation after stenting.
However, the roles of TGF-β and tranilast in restenosis after stenting are uncertain. In a small nonrandomized human study, tranilast reduced restenosis after coronary stenting, but this was not quite statistically significant.13 In a recent study of porcine coronary stenting, tranilast also reduced neointima formation but at doses (50 mg/kg twice daily) far in excess of those associated with clinically relevant toxicity in humans (>600 mg/d).14 This excess was likely unnecessary because plasma tranilast levels in the present study well exceeded those required to suppress TGF-β expression and SMC migration and proliferation. Consequently, we investigated the effects of more moderate doses of tranilast (10 mg/kg twice daily) on neointima formation after porcine arterial stenting and correlated these effects with suppression of activation of the TGF-β system and local inflammatory responses.
Animals, Procedures, and Treatments
Nineteen male Boston minipigs, aged 26 to 40 weeks and weighing 40 to 50 kg, were obtained from a breeding colony at Monash University, Melbourne, Australia. All pigs were maintained on standard pig chow throughout the studies. They all received aspirin, 300 mg orally per day, beginning 5 days before stenting and continuing through to euthanasia. Verapamil hydrochloride slow-release tablets (120 mg, Knoll Australia) were administered the night before the stenting procedure. Fourteen pigs were treated with vehicle (empty gelatin capsule): 4 were killed 5 days after stent implantation, and the vessels (n=8) used for reverse transcription (RT)-polymerase chain reaction (PCR) analysis, whereas morphology and immunohistochemistry were performed on vessels obtained from pigs killed at 5 days (n=2 pigs [4 vessels]), 14 days (n=2 pigs [4 vessels]), and 28 days (n=6 pigs [12 vessels]) after stent implantation. Five pigs were administered tranilast (10 mg/kg orally, twice daily, in gelatin capsules; Kissei Pharmaceuticals) beginning 2 days before the surgery and continuing until death: 3 pigs (6 vessels) were used for morphology and immunohistochemistry, and 2 pigs (4 vessels) were used for RT-PCR analysis.
On the day of the surgery, the pigs were premedicated by intramuscular injection of acepromazine (0.1 mg/kg), droleptan (10 mg), and atropine sulfate (1.2 mg). A 22-gauge intravenous cannula was inserted into the ear vein, and propofol (150 to 200 mg) was given to allow intubation and ventilation. After the administration of 15 000 U heparin, an 8F sheath was inserted into the right common carotid artery. Angiography was performed in standard planar views (25° left anterior oblique for the coronary arteries and straight anteroposterior for the iliac arteries) by using a Judkins left 4 (JL4) guiding catheter 10 seconds after injecting glyceryl trinitrate (200 μg), and recordings were made on super VHS tape. Low-osmolality ionic radiographic contrast (Ioxaglate and Hexabrix, Mallinckrodt) was used to avoid platelet activation. Measurements of arterial diameter and inflated balloon size were made from end-diastolic frames with hand-held digital calipers at the midpoint of the balloon by use of the guiding catheter as a reference.
Half Palmaz-Schatz PS-153 stents, cut at the articulation site, were implanted in the circumflex coronary artery and in the recurrent circumflex branch of the left external iliac artery (whose size is nearly identical to that of the coronary arteries) by using 15-mm-long oversized balloons inflated to 10 atm for 30 seconds. The degree of balloon oversizing for each animal was calculated from the diameter of an artery before injury and the manufacturer-specified balloon size, creating a balloon-to-artery ratio of 1.3 to 1.5:1. Angiography was repeated after removing the balloon catheter and the guidewire. The guiding catheter and the sheath were removed, the carotid artery was ligated, and the wound was closed. Any (infrequent) cardiac arrhythmias were treated with intravenous lidocaine (100 mg, for nonsustained ventricular arrhythmias) or electrical defibrillation (for sustained arrhythmias).
Vessel Collection and Processing
Arteries collected for RT-PCR analysis were snap-frozen in liquid nitrogen and stored at −80°C until RNA extraction. Control segments were taken from an uninjured portion of the same artery at least 10-mm remote from the edge of the stent. Arteries for morphology and immunohistochemistry were perfusion-fixed by using 4% formalin in PBS (pH 7.4) after isolation of the vascular bed. Iliac vessels were isolated by ligation of the aorta and inferior vena cava (just proximal to the aortic bifurcation) and of the origin of the internal iliac vessels and the external iliac vessels just distal to the recurrent circumflex iliac branch. The perfusate cannula was inserted into the distal aorta, and a venous drainage incision was made in the distal vena cava. The coronary vessels were isolated by cross-clamping the ascending aorta and superior and inferior venae cavae. A perfusate cannula was inserted into the aortic root, and venous drainage incisions were made in the atrial appendages. All vessels were initially perfused with saline until clear perfusate arose from the venous drainage incision, and then they were perfused with 4% formalin in PBS for 5 minutes at 100 to 150 mm Hg. The excised vessels were stored in 4% formalin at 4°C for 24 hours before being embedded in Epon (Araldyte GY250 Epon 812, Spi Chem). Subsequently, the tissues containing epoxy resin blocks were trimmed with a diamond bone saw (Kovacs Gems and Minerals) so that the end of the stent abutted the cutting face of the block. This surface was then exposed to 0.1 mol/L nitrohydrochloric acid for 120 minutes before being cut into 2-μm sections with a glass knife (Latta and Hartman). This process was then repeated at intervals along the stented segment (3 to 5 times per vessel) to measure intimal area at the site of greatest lumen loss and at all sites to create mean data. The sections were mounted onto poly-l-lysine- or 3-amino-triethoxysilane-coated slides, and the resin was removed by immersion in NaOH-saturated ethanol. After rehydration in graded aqueous ethanol solutions, the sections were stained by use of Masson’s trichrome with orcein and hematoxylin-eosin or were subjected to immunohistochemistry.
Rehydrated sections were incubated in 10% horse serum for 30 minutes, washed in 0.1 mol/L PBS, and then incubated for 1 hour at room temperature in PBS containing the relevant primary antibody or control IgG (1:1000). The sections were again washed in PBS and incubated with the appropriate biotinylated secondary antibody (1:200) for 1 hour. Then, after further washings, the antigen of interest was visualized by using a streptavidin-alkaline phosphatase system (Silenus) and naphthol AS-BI/fuchsin/nitrite/levamisole solution (Sigma Chemical Co), except for CD45 immunostaining, for which an avidin-biotin-peroxidase complex with a diaminobenzidine colorimetric system was used. The sections were counterstained with hematoxylin. The intensity of immunostaining around the stent struts was scored as follows: 0, no staining; 1, minor staining only; 2, moderate staining; and 3, heavy staining. Scoring was performed on every third strut in each vessel beginning with the strut closest to the top of the slide by an investigator blinded to the treatment allocation.
The primary antibodies used were as follows: a TGF-β1-specific polyclonal purified chicken IgG raised against human TGF-β1 (Becton-Dickinson), a polyclonal TGF-β3 rabbit IgG raised against amino acid residues 303 to 315 of human TGF-β3, a polyclonal ALK-5 antibody raised against amino acids 158 to 179 of human ALK-5, a polyclonal TβR-II antibody raised against amino acids 246 to 266 of human TβR-II (Santa Cruz Biotechnologies), and a monoclonal mouse anti-pig CD45 antibody (Serotec). Secondary antibodies were a biotinylated goat anti-turkey/chicken IgG (Zymed) and a biotinylated goat anti-rabbit IgG (Vector Laboratories).
The neointima that developed in the stented vessels from the different animals was analyzed by planimetry, as previously described,15 with the use of a projecting microscope and digitizing tablet (Complot Series 7000 digitizer, Bausch and Lomb) by an investigator blinded to treatment group. Data are presented separately for the site in each vessel with the smallest luminal area (and maximal intimal area) and as the mean of all sites sampled. The neointima was defined by its inner border with the lumen and its outer border, the internal elastic lamina (IEL). Intimal area was adjusted for vessel area (area within the external elastic lamina [EEL]), yielding a measure independent of vessel size (intimal area/vessel area). The severity of strut-induced injury was averaged over all struts in each vessel and scored as described by Schwartz et al16: 0, IEL intact; 1, IEL fractured by strut; 2, lacerated media; and 3, strut-ruptured EEL.
Inflammation associated with the stent struts of control and tranilast-treated pigs was graded by counting the mean number of CD45-positive cells associated with each stent strut from hematoxylin-eosin-stained sections, as previously described.17 No attempt was made to quantitatively assess the relative contribution of lymphocytes, neutrophils, or other monocytic cells to the leukocyte infiltrates.
Neointimal collagen density was quantified by using a computer-interfaced color imaging system (Optimus Bioscan 2, Thomas Optical Measurement System, Inc) from sections stained with Masson’s trichrome, as previously described.15
Measurement of mRNA Levels
mRNA levels in vessels stented 5 days earlier were measured by using a standardized RT-PCR assay, exactly as previously described.8 Briefly, after extracting total RNA from the vessel segments by using a single-step guanidinium thiocyanate-phenol-chloroform procedure, the RNA was treated with DNAse. Then 200 ng total RNA was subjected to RT-PCR with conditions and cycle number optimized for each primer pair, such that the amount of PCR product was reproducibly proportional to mRNA levels in the tissue.9 The reliability of the assay in quantification of mRNA levels was demonstrated by showing that the PCR product generated was linearly related to total RNA, over the RNA range (100 to 400 ng). Linear regression equations describing the relationships between the amounts of PCR products generated, measured as optical density [P(DNA)], and the amounts (in nanograms) of RNA used (RNA) in this standardized RT-PCR procedure were as follows: for TGF-β1, P(DNA)=0.037 (RNA)+0.3, R2=0.99; for TGF-β3, P(DNA)=0.032 (RNA)−0.50, R2=0.96; for ALK-5, P(DNA)=0.059 (RNA)+1.0, R2=0.98; for TβR-II, P(DNA)=0.051 (RNA)−2.0, R2=0.96; and for L7, P(DNA)=0.149 (RNA)−2.0, R2=0.987. Reproducibility was estimated by repeating the assay on 3 separate occasions with the same RNA, with coefficients of variation all <10%.9 The numbers of PCR cycles used for quantification, together with the size of cDNA fragments generated and their characterization by restriction enzyme digestion, are listed in Table 1.
Each assay is standardized by expressing the PCR product for each target mRNA relative to the amount of PCR product from L7 mRNA, which encodes a constitutively expressed noninducible cell cycle-independent ribosomal protein.9 The result for each stented segment was then expressed relative to that from the uninjured portion of the same vessel. Thermal cycling conditions and reagents used in the RT-PCR reactions were identical to those used in our previous studies.3,9⇓ All PCR products were electrophoresed in 2% agarose gels (Progen), together with HaeIII-digested ΦX174 DNA size markers (Promega,). They were photographed under UV light with positive/negative film (Polaroid 665), and the intensities on the negatives were quantified by laser densitometry.
Oligonucleotide primers used in the PCR reactions were designed with the Primer Detective program (Clontech Laboratories) with the use of the following GenBank cDNA sequences: for TGF-β1 (accession No. M23703), sense 5′-TGCTAATGGTGGAAAGCGGCAACC-3′, antisense 5′-GTTATCTTTGCTGTCACAGGAACAGTGGGC-3′ (bp 332 to 690); for TGF-β3 (accession No. X14150), sense 5′-GAGATCCATAAATTC-GACATGATCCAGGGG-3′, antisense 5′-ATTTCCAGACCCA-AGTTGGACTCTCT-3′ (bp 708 to 1266); for ALK-5 (accession No. L11695), sense 5′-TCTTGCCCATCTTCACATGGAGATTGTTGG-3′, antisense 5′-TACATTTTGATGCCTTCCTTHGGCTGAGC-3′ (bp 930 to 1505); and for TβR-II (accession No. M85079), sense 5′-GTGGCTGTCAAGATCTTCCCCTACG-3′, antisense 5′-TCGCTGTT CCCACCTGCCCGTTGTTG-3′ (bp 820 to 1267).
Results are expressed as mean±SEM. Comparisons between 2 groups were made by using either an unpaired t test or Mann-Whitney rank sum test, depending on whether the data were normally distributed, as assessed by using the Kolmogorov-Smirnov test. For multiple comparisons, 1-way ANOVA/ANOVAR was used, followed by a post hoc pairwise t test or Mann-Whitney rank sum test. Results were considered statistically significant at P<0.05.
Neointima Structure and TGF-β Expression
The size of the neointima at the site of minimal luminal area increased progressively over the 28 days after stenting; the increase from 5 to 28 days was nearly 3-fold (Figure 1). The mean neointimal area also increased over time (0.18±0.04 mm2 at 5 days, 0.57±0.18 mm2 at 14 days, and 0.95±0.11 mm2 at 28 days). In contrast to our findings in unstented vessels,18 neither maximal (1.61±0.29 [coronary] versus 1.39±0.21 [iliac] mm2, P=0.35; n=6 vessels each) nor mean (1.08±0.26 [coronary ] versus 0.82±0.22 [iliac] mm2, P=0.25) neointimal area was significantly affected by the vessel stented. Sections of stented vessels stained with Masson’s trichrome with orcein from the latter time indicated that the largely fibrocellular neointima was thickest where the IEL was most damaged and that extracellular matrix had accumulated particularly around the stent struts, whereas neointimal SMCs were more prevalent in regions away from the struts (Figure 1). Although small breaks in the IEL were consistently associated with the stent struts, the EEL of the stented vessels appeared intact. There was substantial fibrous thickening of the adventitia, which 28 days after the stenting was ≈3-fold greater than in control vessels (Figure 1).
Stenting also increased arterial mRNAs encoding TGF-β isoforms and receptors. Five days after endoluminal stenting, TGF-β1 mRNA levels were nearly 6-fold greater than in control (uninjured) vessels (P<0.05, Figure 2); increases in TGF-β3 were nearly 4-fold (P<0.05), whereas mRNAs encoding the TGF-β type I and type II signaling receptors (ALK-5 and TβR-II) were both increased nearly 6-fold (P<0.05, Figure 2). Two weeks after stenting, TGF-β1 immunopeptides were localized predominantly around the stent struts, with highest concentrations on the luminal side in the neointima (Figure 3). TGF-β3 was also mostly localized to the luminal side of the neointima adjacent to the stent struts (Figure 3). High concentrations of the TGF-β signaling receptors, ALK-5 and TβR-II, were also present in the vicinity of the stent struts, mostly localized to regions expressing high concentrations of the TGF-β isoforms (Figure 3). Only minor staining was seen in the adventitia for isoforms and receptors. At 4 weeks, an essentially similar pattern was observed, although the relative concentrations of TGF-β3 and ALK-5 appeared to be less (Figure 3).
Tranilast and TGF-β Expression in Stented Coronary Arteries
We have previously shown that tranilast administration markedly but reversibly attenuates TGF-β isoform and receptor mRNA expression after rat carotid balloon injury.3 To evaluate whether lower doses would suppress the synchronous upregulation of isoforms and receptors of the TGF-β system after stenting in pigs, we measured mRNA levels in 4 tranilast-treated vessels 5 days after stenting. Tranilast suppressed the early (5-day) elevation in TGF-β1 by nearly 65% (P<0.05, Table 2). TβR-II mRNA levels were reduced by nearly 80% (P<0.05). Although TGF-β3 and ALK-5 mRNA levels were also reduced with tranilast treatment, this was not statistically significant (P>0.05, Table 2). Treatment with tranilast for 4 weeks markedly reduced detectable TGF-β1 immunopeptide levels in the vicinity of the stent struts (mean score 2.4±0.2 for vehicle versus 0.8±0.4 for tranilast) for its extracellular and intracellular concentrations (Figure 4). TGF-β3 levels were also greatly reduced (1.4±0.2 for vehicle versus 0.4±0.2 for tranilast), as were the concentrations of TGF-β receptors, ALK-5 (1.6±0.2 for vehicle versus 0.6±0.2 for tranilast) and TβR-II (2.2±0.2 for vehicle versus 1.0±0.3 for tranilast), in these regions of the stented arteries (Figure 4).
Tranilast and Intimal Hyperplasia After Coronary Artery Stenting
Balloon oversizing (balloon-to-artery ratio) was similar in the tranilast- and vehicle-treated groups, averaging 1.49±0.04 and 1.52±0.02, respectively (P>0.05). The severity of injuries inflicted by stenting (ie, strut injury scores) were also similar, averaging 1.6±0.4 in the tranilast-treated group and 1.4±0.6 in the vehicle-treated group (P>0.05). Treatment for 4 weeks after injury reduced maximal neointimal cross sectional area by 48% (P<0.05, Figure 5) and by 45% when corrected for size of the vessel (intimal area/vessel area, 0.16±0.02 for tranilast versus 0.29±0.03 for vehicle). Mean intimal area was also reduced by 43% (0.54±0.07 mm2 for tranilast versus 0.95±0.11 mm2 for vehicle, P<0.05) and by 42% when corrected for vessel area (0.10±0.03 mm2 for tranilast versus 0.18±0.03 mm2 for vehicle, P<0.05). Examination of vessel sections stained with Masson’s trichrome with orcein revealed that the reduction in intimal area appeared to be due to a significant reduction in extracellular matrix around the stent struts and in SMCs accumulating on the luminal side of the stent struts (Figure 5).
Accumulation of Leukocytes in Stented Vessels During Tranilast Therapy
Because accentuated neointima formation after stenting has been linked to inflammatory infiltrates around the struts19 and because TGF-β is a potent chemotactic agent for many mononuclear leukocytes, including neutrophils and monocytes,12 we next investigated how tranilast treatment affected leukocyte numbers in the region of the stent struts. Four weeks after stenting, the frequency of leukocytes in the immediate vicinity of the stent struts was reduced by tranilast treatment by nearly 70% (P<0.05, Figure 6). The distribution of these mononuclear cells was always highest in the neointima, on the side of the strut facing the lumen of the vessel, and was rarely observed on the side facing the media (Figure 6).
In the present study, we demonstrate that at doses comparable to those used in human trials, tranilast reduces the neointima after stenting and suppresses the synchronous upregulation of isoforms and receptors of the TGF-β system and leukocyte infiltration around the stent struts. This association suggests a specific role for TGF-β in coordination of the augmented inflammation and SMC migration around stent struts, to which the exaggerated intimal hyperplastic response after stenting has been attributed.
Although stents are implanted in most coronary interventional procedures, very little is known about stent restenosis compared with restenosis after angioplasty. Histopathological studies of in-stent restenosis have shown that the degree of inflammatory cell infiltration around stent struts is at least as important as the severity of arterial injury in determining the amount of neointima formation.6,7,19–21⇓⇓⇓⇓ To this end, much attention has been focused on stent design and stent coatings to attempt to reduce the localized trauma and inflammation associated with stenting.22 However, little is known about the cytokines and growth factors that are specifically associated with the focal trauma and inflammation induced by each stent strut. The present study demonstrates that activation of the TGF-β system after arterial stenting is associated with accentuated local expression of TGF-β1 and TGF-β3 and their 2 signaling receptors, ALK-5 and TβR-II, around the stent struts. Such site-specific coordinated activation indicates that TGF-βs may play an important role in the accentuated inflammatory response and in SMC migration around the stent struts, to which the exaggerated neointima formation of stent restenosis has been attributed, in addition to their well-known role in injury-induced neointima formation. TGF-β activation after arterial injury stimulates the expression of integrins αv and β3 and matrix receptors critical for SMC migration9 and would likely enhance early migration of SMCs through the small breaks in the IEL around stent struts. The expression of TGF-β isoforms in the vicinity of the stent struts may also contribute to the recruitment of macrophages because TGF-β is a potent chemotactic factor for monocytes and facilitates their movement through the extracellular matrix by inducing their secretion of urokinase plasminogen activator and other proteolytic enzymes.10,23⇓ Preferential accumulation of collagen around stent struts, which is strongly induced by TGF-β, reflects localization of TGF-β bioactivity to these regions.
In our previous studies, we demonstrated that the synchronous upregulation of isoforms and receptors of the TGF-β system after arterial injury is profoundly but reversibly inhibited by administration of tranilast.3 In the present study, lower doses of tranilast also significantly suppressed the early elevations in mRNAs encoding TGF-β1 and TβR-II and later attenuated immunoreactivity around the stent struts for TGF-β1, TGF-β3, ALK-5, and TβR-II, confirming that significant suppression of the TGF-β system also occurs in an animal model, which better represents human coronary restenosis, and in doses relevant to human trials. mRNA levels and immunohistochemical intensity are only relative measures of TGF-β isoform and receptor expression, and such changes may not necessarily be associated with important changes in bioactivity. However, the reduction in collagen deposition around the struts with tranilast therapy is reasonable circumstantial evidence that TGF-β bioactivity is important and that the suppression is significant.
Greater inhibition of induction of TβR-II than ALK-5 by tranilast may have reduced the receptor ratio. Some investigators have suggested that such changes in the receptor ratio are important in promoting atherogenesis,24 findings questioned by others.25 In restenosis, however, specific TβR-II inhibition reduces collagen production and neointima formation,4 and such effects are much more consistent with our findings in the present study.
Early reported summaries of results of the Prevention of Restenosis With Tranilast and Its Outcomes (PRESTO) study,26 in which 300 or 450 mg tranilast was administered for 1 or 3 months after percutaneous coronary intervention, indicate that in these doses tranilast was ineffective in preventing restenosis. The choice of doses in that study was curious because in previous studies, 300 mg per day was ineffective, whereas only 600 mg per day was effective in preventing restenosis after angioplasty. In the only previous study of tranilast after stenting, the dose used was 600 mg per day.13 The full results of that study are anxiously awaited to ascertain whether the discrepancy between this and previously positive Japanese trials1,2,13⇓⇓ can be attributed to dose or racial differences.
Tranilast also attenuated the infiltration of mononuclear leukocytes around the stent struts. Although attenuation of TGF-β bioactivity would be expected to reduce inflammatory cell accumulation, tranilast has also recently been shown to prevent inflammatory cell release of several cytokines, including TGF-β.27 It is possible that a generalized damping of the inflammatory response may be attributable to prevention of TGF-β production and its effects.
Long-term treatment with tranilast (20 mg/kg per day) also significantly suppressed neointimal growth in the stented vessels. Although tranilast has previously been reported to inhibit porcine stent restenosis at doses of 100 mg/kg per day,14 11% of the patients administered 600 mg/kg per day in a recent human restenosis trial experienced liver dysfunction or abdominal discomfort,1 and it is likely that doses of 100 mg/kg per day would be poorly tolerated. In addition, it is unlikely that such high doses are necessary because plasma tranilast levels achieved with this dosing were ≈1 mmol/L,14 whereas maximal effectiveness in inhibiting TGF-β induction3 and SMC migration, proliferation, and collagen production28 occurs between 100 and 300 μmol/L. The present study, which demonstrates efficacy in the suppression of TGF-β activation, neointima formation, and leukocyte infiltration at lower doses, confirms that such high doses are indeed unnecessary. Although the present study did not examine the possibility that once the drug is stopped, there may be a rebound increase in intimal hyperplasia, human clinical studies with the drug have shown that this is not a functionally important consequence of drug cessation because the angiographic benefits of the drug at 3 months (when the drug was stopped) were qualitatively similar to clinical results at 12 months.2
Tranilast attenuates the bioactivity of many other pathways implicated in neointima formation and restenosis; however, many of these pathways involve the TGF-β system either directly or indirectly. For example, the ability of tranilast to suppress intimal hyperplasia has been linked to its induction of p21,29 a well-known effect of TGF-β itself.30 Recently, tranilast has also been reported to suppress chymase activity and inhibit neointima formation in the injured carotid artery of the dog; these effects were attributed to the inhibition by tranilast of chymase-dependent angiotensin II formation.31 Because angiotensin II also stimulates TGF-β1 production in vascular SMCs32 and because chymase is known to efficiently release matrix-bound latent TGF-β1,33 it is likely that a component of the effects of tranilast in these circumstances is due to the inhibition of TGF-β1. Similarly, platelet-derived growth factor and fibroblast growth factor induce TGF-β activation, and the inhibition of their actions by tranilast may involve TGF-β suppression.9 Irrespective of the relative importance and interrelationships between the different pathways, the ability of tranilast to specifically attenuate events specifically linked to intimal hyperplasia after stenting, including activation of the TGF-β system and inflammatory cell infiltration around the stent struts, means that it is likely to be at least as effective (if not more so) in the prevention of restenosis after stenting as it has been after angioplasty.
In summary, the present study suggests that activation of the TGF-β system after stenting occurs specifically around the struts. Tranilast, an agent that has successfully reduced restenosis rates after angioplasty2 and directional atherectomy,1 suppresses this the synchronous upregulation of isoforms and receptors of the TGF-β system and also reduces inflammation around the struts. Further characterization of the role of TGF-β in stent restenosis may lead to therapies that are more effective and better tolerated than tranilast for the prevention of restenosis.
Received March 5, 2002; revision accepted April 12, 2002.
- ↵Smith JD, Bryant SR, Couper LL, Vary CP, Gotwals PJ, Koteliansky VE, Lindner V. Soluble transforming growth factor-beta type II receptor inhibits negative remodeling, fibroblast transdifferentiation, and intimal lesion formation but not endothelial growth. Circ Res. 1999; 84: 1212–1222.
- ↵Nabel EG, Shum L, Pompili VJ, Yang ZY, San H, Shu HB, Liptay S, Gold L, Gordon D, Derynck R, Nabel GJ. Direct transfer of transforming growth factor beta 1 gene into arteries stimulates fibrocellular hyperplasia. Proc Natl Acad Sci U S A. 1993; 90: 10759–10763.
- ↵Kearney M, Pieczek A, Haley L, Losordo DW, Andres V, Schainfeld R, Rosenfield K, Isner JM. Histopathology of in-stent restenosis in patients with peripheral artery disease. Circulation. 1997; 95: 1998–2002.
- ↵Farb A, Sangiorgi G, Carter AJ, Walley VM, Edwards WD, Schwartz RS, Virmani R. Pathology of acute and chronic coronary stenting in humans. Circulation. 1999; 99: 44–52.
- ↵Ignotz RA, Heino J, Massague J. Regulation of cell adhesion receptors by transforming growth factor-beta: regulation of vitronectin receptor and LFA-1. J Biol Chem. 1989; 264: 389–392.
- ↵Ward MR, Agrotis A, Kanellakis P, Dilley R, Jennings G, Bobik A. Inhibition of protein tyrosine kinases attenuates increases in expression of transforming growth factor-beta isoforms and their receptors following arterial injury. Arterioscler Thromb Vasc Biol. 1997; 17: 2461–2470.
- ↵Wahl SM, Allen JB, Weeks BS, Wong HL, Klotman PE. Transforming growth factor beta enhances integrin expression and type IV collagenase secretion in human monocytes. Proc Natl Acad Sci U S A. 1993; 90: 4577–4581.
- ↵Simon DI, Rao NK, Xu H, Wei Y, Majdic O, Ronne E, Kobzik L, Chapman HA. Mac-1 (CD11b/CD18) and the urokinase receptor (CD87) form a functional unit on monocytic cells. Blood. 1996; 88: 3185–3194.
- ↵Hsu Y-S, Tamai H, Ueda K, Ono S, Kosuga K, Tanaka S, Matsui S, Motohara S, Uehata H. Efficacy of tranilast on restenosis after coronary stenting. Circulation. 1996; 94 (suppl I): I–620.Abstract.
- ↵Ward MR, Kanellakis P, Ramsey D, Funder J, Bobik A. Epleronone suppresses collagen accumulation and constrictive remodeling after balloon angioplasty in porcine coronary arteries. Circulation. 2001; 104: 467–472.
- ↵Edelman ER, Seifert P, Groothuis A, Morss A, Bornstein D, Rogers C. Gold-coated NIR stents in porcine coronary arteries. Circulation. 2001; 103: 429–434.
- ↵Komatsu R, Ueda M, Naruko T, Kojima A, Becker AE. Neointimal tissue response at sites of coronary stenting in humans: macroscopic, histological, and immunohistochemical analyses. Circulation. 1998; 98: 224–233.
- ↵Rogers C, Edelman ER, Simon DI. A mAb to the beta2-leukocyte integrin Mac-1 (CD11b/CD18) reduces intimal thickening after angioplasty or stent implantation in rabbits. Proc Natl Acad Sci U S A. 1998; 95: 10134–10139.
- ↵Edelman ER, Rogers C. Hoop dreams: stents without restenosis. Circulation. 1996; 94: 1199–1202.
- ↵McCaffrey TA, Consigli S, Du B, Falcone DJ, Sanborn TA, Spokojny AM, Bush HL Jr. Decreased type II/type I TGF-beta receptor ratio in cells derived from human atherosclerotic lesions: conversion from an antiproliferative to profibrotic response to TGF-beta1. J Clin Invest. 1995; 96: 2667–2675.
- ↵Bobik A, Agrotis A, Kanellakis P, Dilley R, Krushinsky A, Smirnov V, Tararak E, Condron M, Kostolias G. Distinct patterns of transforming growth factor-beta isoform and receptor expression in human atherosclerotic lesions: colocalization implicates TGF-beta in fibrofatty lesion development. Circulation. 1999; 99: 2883–2891.
- ↵SoRelle R. Late-breaking clinical trials at the American Heart Association’s Scientific Sessions 2001. Circulation. 2001; 104: E9046–E9048.
- ↵Capper EA, Roshak AK, Bolognese BJ, Podolin PL, Smith T, Dewitt DL, Anderson KM, Marshall LA. Modulation of human monocyte activities by tranilast, SB 252218, a compound demonstrating efficacy in restenosis. J Pharmacol Exp Ther. 2000; 295: 1061–1069.
- ↵Takahashi A, Taniguchi T, Ishikawa Y, Yokoyama M. Tranilast inhibits vascular smooth muscle cell growth and intimal hyperplasia by induction of p21(waf1/cip1/sdi1) and p53. Circ Res. 1999; 84: 543–550.
- ↵Pardali K, Kurisaki A, Moren A, ten Dijke P, Kardassis D, Moustakas A. Role of smad proteins and transcription factor Sp1 in p21Waf1/Cip1 regulation by transforming growth factor-beta. J Biol Chem. 2000; 275: 29244–29256.
- ↵Shiota N, Okunishi H, Takai S, Mikoshiba I, Sakonjo H, Shibata N, Miyazaki M. Tranilast suppresses vascular chymase expression and neointima formation in balloon-injured dog carotid artery. Circulation. 1999; 99: 1084–1090.
- ↵Gibbons GH, Pratt RE, Dzau VJ. Vascular smooth muscle cell hypertrophy vs. hyperplasia: autocrine transforming growth factor-beta 1 expression determines growth response to angiotensin II. J Clin Invest. 1992; 90: 456–461.
- ↵Taipale J, Lohi J, Saarinen J, Kovanen PT, Keski-Oja J. Human mast cell chymase and leukocyte elastase release latent transforming growth factor-beta 1 from the extracellular matrix of cultured human epithelial and endothelial cells. J Biol Chem. 1995; 270: 4689–4696.