Introduction

Top Abstract Introduction Experimental Procedures Results Discussion References

Percutaneous transluminal coronary angioplasty (PTCA) has become increasingly important in the management of coronary artery disease. However, restenosis in 30 to 40% of PTCA patients limits its usefulness (Kuntz and Baim, 1993). Although the pathogenesis of this process is multifactorial, the formation of a neointima is common after balloon injury. Histological studies have shown that intimal thickening in the restenotic lesion contains smooth muscle cells in an abundant extracellular matrix. Although an increasing amount of experimental evidence suggests that the proliferation of smooth muscle cells after balloon injury is the most likely mechanism for the development of postangioplasty restenosis, there is controversy regarding the degree of the proliferative response in biopsies from clinical postangioplasty restenotic lesions (Schwartz et al., 1992; Isner et al., 1994). Furthermore, in spite of encouraging results in animal models, no systemic pharmacological agent has been shown conclusively to produce a clinically worthwhile reduction in restenosis after PTCA (Franklin and Faxon, 1993; Moliterno and Topol, 1996).

Most of the therapies for preventing restenosis tested so far have been directed against neointimal hyperplasia. However, clinical observations have questioned prior assumptions that neointima formation solely correlates with luminal narrowing (Luo et al., 1996; Mintz et al., 1996). This implies that neointimal hyperplasia is not the sole mechanism of restenosis after conventional balloon angioplasty in humans. Recent experimental and clinical studies have suggested that late vessel constriction and vasospasm may be the major factor (Ardissino et al., 1991; Kakuta et al., 1994; Mattsson and Clowes, 1995; Andersen et al., 1996).

Pigs have been used as models for postangioplasty restenosis with some success. Injury to porcine coronary arteries by PTCA using an overexpanded balloon catheter stimulates the formation of vascular lesions morphologically similar to those seen in human postangioplasty restenosis (Muller et al., 1992). An advantage of this model is the ability of the investigator to study coronary arteries rather than peripheral vessels. The findings in this model have shown that coronary injury results in a significant vasoconstriction, induced by autacoid from activated platelets, and in a significant remodeling of the adventitia, accompanied by the proliferation and differentiation of adventitial fibroblasts into myofibroblasts, which acquire -smooth muscle actin, during the formation of neointima (Kadokami et al., 1996; Scott et al., 1996; Shi et al., 1996a,b).

On the basis of the fact that there are striking similarities between the process of wound healing and the adventitial response of the arterial wall to injury and because stimulation of adventitia to proliferate and constrict after angioplasty is thought to be one of the typical wound-healing processes (Darby et al., 1990; Clark, 1993; Yokozeki et al., 1997), we developed TAS-301, a new synthesized constrictive remodeling regulator, and tested it on three-dimensional (3D) collagen gel contraction, which is a frequently used and previously established in vitro model for wound repair process (Bell et al., 1979; Grinnell et al., 1994). TAS-301 showed a potent inhibitory effect on type-I collagen gel contraction caused by fibroblasts precultured with TGF-₁, which cells acquired a strong gel-contracting ability under low-serum culture conditions (E. Sasaki, Y. Yamasaki, Y. Tanahashi, Y. Muranaka, H. Terakawa, K. Miyoshi, N. Oda, H. Miyake, and N. Matsuura, submitted). We have previously indicated that this compound potently inhibits the intimal thickening after balloon injury to rat common carotid artery due to inhibition of both migration of medial smooth muscle cells and proliferation of medial and intimal smooth muscle cells (Muranaka et al., 1998).

In the present study, we undertook an examination of the possible effect of TAS-301 on stenosis after balloon overstretch injury of porcine coronary arteries by measuring the diameter of vessels by angiography and conducting histological analysis, based on the results of the above-mentioned in vitro experiments on the proliferation of and 3D collagen gel contraction by adventitial fibroblastic cells.

	Experimental Procedures

Top Abstract Introduction Experimental Procedures Results Discussion References

Twenty-eight male Yucatan micropigs (18.4-29.2 kg; Charles River Laboratories, Wilmington, MA) were used in this study. The animals were housed in constant-temperature facilities and given standard lab chow and water ad libitum. All experiments were carried out according to protocols approved by the Institutional Animal Care and Use Committee.

Animal Preparation. Animal experiments were performed as previously reported with minor modification (Hata et al., 1995). Each animal was randomly assigned to one of the following five groups: control group (n = 5); TAS-301 10-mg/kg-treated group (n = 5); TAS-301 30-mg/kg-treated group (n = 5); TAS-301 100-mg/kg-treated group (n = 5); or normal group (n = 5). In all groups except for the normal group, balloon dilation was performed in the left anterior descending coronary artery as described below.

Experimental Protocol. Coronary arteriography was performed 3 min after intracoronary administration of nitroglycerin (20 µg/ml), which was given before balloon dilation, immediately after the balloon dilation, and 4 weeks after the balloon dilation procedure to document changes in lumen size. Nitroglycerin treatment was performed to assess organic stenosis in maximally dilated state. Furthermore, coronary arteriography was also performed 3 min after intracoronary administration of serotonin (10 µg/ml), 4 weeks after the balloon dilation procedure, to document changes in lumen size. TAS-301 (10, 30, and 100 mg/kg) was administered daily as a dietary admixture, from 1 day before the balloon dilation until 4 weeks after it.

Data Analysis. The photocopies in end-diastolic frame were made for measurements of coronary artery diameter by use of commercially available software (CAM-1000; Nishimoto, Tokyo, Japan). The size of the catheter was used to calibrate the actual vessel diameter in millimeters. The percentages of increase in and narrowing of the coronary diameter after dilation procedure were calculated as follows: luminal dilation (%) = diameter immediately after balloon dilation (mm)/diameter before balloon dilation (mm) × 100, and luminal stenosis (%) = (1  diameter 4 weeks after balloon dilation (mm)/diameter immediately after balloon dilation (mm)) × 100.

Histopathological Approach. After the final arteriography, the arteries were perfused with 10% buffered formalin for 10 min at 100 mm Hg pressure. Four or five cross-sectional segments in the lesion area were embedded in paraffin. Sections were then cut (5 µm) and stained with H&E or with Van Gieson elastin stain. From photographs of all sections, the areas for adventitia, media, intima, and lumen were measured; and the ratio for intimal area to medial area, that for adventitial area to medial area, and that for luminal area to intimal plus luminal areas were calculated and compared between groups.

Cell Culture. Adventitial fibroblast cells from normal and injured coronary artery were isolated by an enzyme dispersion method using collagenase, as previously reported (Chamley-Campbell et al., 1981). The cells were maintained in Dulbecco's modified Eagle's medium (DMEM), supplemented with 10% fetal bovine serum (FBS), 100 U/ml penicillin, 100 U/ml streptomycin, and grown in a moist atmosphere of 95% air and 5% CO₂ at 37°C. For immunocytochemistry using antibody against -smooth muscle actin (Boehringer Mannheim Biochemicals, Tokyo, Japan), these cells isolated from normal artery were faintly positive, whereas they were markedly positive after having been precultured with TGF-₁ for 7 days (data not shown). It has been demonstrated that TGF-₁ can induce -smooth muscle actin expression in fibroblasts and cause these cells to change phenotypically into myofibroblasts (Arora and McCulloch, 1994). Therefore, we identified these cells as fibroblasts. Cells were passaged by trypsinization (0.1% trypsin/0.02% EDTA in PBS; Kojin-Bio, Saitama, Japan) and were used at a subconfluent stage in passage 5 to 9.

In Vitro Proliferation Assay. Cell proliferation was determined by incorporation of 5-bromo-2'-deoxyuridine (BrdU) by quiescent cells, as described previously (Magaud et al., 1988). Adventitial fibroblast cells were seeded at 3 × 10³ cells/well in 96-well plates in DMEM containing 10% FBS. Three days after the seeding, their growth was minimally arrested for 5 days in DMEM containing 1% FBS. Then, the DMEM was removed and serum-free DMEM containing 0.1% BSA with or without TAS-301 was added to the quiescent cells 2 h before treatment with the desired growth factor, i.e., TGF-₁ (1 pg/ml) or bFGF (10 ng/ml). Sixteen hours after stimulation, BrdU (10 µM) was added to the cultures, and 2 h later the cells were fixed. An enzyme-linked immunosorbent assay system was used according to the manufacturer's recommendations (RPN250; Amersham, Amersham, England) to detect and to quantify the incorporated BrdU. The drugs were present during the entire time of the experiments.

Preparation of 3D Collagen Gel Culture. Collagen gels (3D) were prepared essentially as described previously (Zhang et al., 1996; Yokozeki et al., 1997). Adventitial fibroblasts-populated collagen gels were prepared by rapidly mixing together a suspension of fibroblast cells in DMEM with porcine skin type I collagen stock solution (3 mg/ml) and pouring the mixture into a 24-well plate (Iwaki Co., Tokyo, Japan) that had been precoated with 2% BSA dissolved in PBS. The plates were immediately incubated at 37°C in 95% air + 5% CO₂ to promote collagen fibrillogenesis. Under these conditions, the pH of the media remained neutral except for a few seconds immediately upon addition of the collagen stock solution. The cell-containing solution usually gelled within 1 h after pouring. In the experiments described below, the final FBS concentration was 0.1%; and the final cell concentration, 1.5 × 10⁵/ml. The final collagen concentration was always 0.75 mg/ml. After gelation was complete, the gels were loosened or detached from the well surface by vigorous shaking. To determine the efficacy of TAS-301 on collagen gel contraction, this agent was incubated with the fibroblast cell suspension for 2 h before the cells were mixed with collagen stock solution. Collagen gel contraction was determined by using NIH Image software to measure the area of the detached gel on the plates at 24 h after gel formation. For these in vitro assays, drugs were dissolved in dimethyl sulfoxide and diluted in medium.

Materials. TAS-301 was synthesized by Taiho Pharmaceutical Co., Ltd. (Saitama, Japan). The following reagents with their source in parentheses, were used: serotonin (Sigma Chemical Co., St. Louis, MO), bFGF and TGF-₁ (Life Technologies, Grand Island, NY), and type I collagen stock solution (porcine skin; Wako Pure Chemical, Osaka, Japan).

Statistical Analysis. All data are expressed as means ± S.E.M. Multiple comparisons with vehicle were tested by Dunnett's multiple comparison test. Statistically significant differences between two groups were calculated by (two-tailed) Aspin-Welch's t test. A P value <.05 was considered to indicate statistical significance.

	Results

Top Abstract Introduction Experimental Procedures Results Discussion References

Group Characteristics. There were no significant differences in animal weight among the control and three experimental groups at the time of the balloon injury. Three of 28 micropigs died during the initial procedure. These deaths were attributed to ventricular arrhythmia. Heart rate and arterial pressure under anesthesia did not differ among the four groups (data not shown).

Angiographic Analysis. Angiographic analysis was performed with a computer-based system to determine the degree of vessel stretch, which was measured as the ratio of artery diameter during and after balloon inflation to diameter before balloon inflation (Table 1). The mean LAD coronary artery diameter was not significantly different among the groups. Similarly, the luminal dilation after balloon injury did not differ among groups: there was a greater than 40% average increase in vessel diameter after the injury (control group, 45.0 ± 1.5%; TAS-301 at 10 mg/kg, 41.0 ± 2.2%; at 30 mg/kg, 42.3 ± 4.3%; and at 100 mg/kg, 43.5 ± 5.2%). This large stretch was associated with greater injury, as shown later.

View larger version (39K): [in this window] [in a new window]   Fig. 1.   Effects of TAS-301 on stenosis ratio (%) 4 weeks after the balloon injury. Data show stenosis % (details under Experimental Procedures, mean ± S.E.M., n = 5). **P < .01 versus control (Dunnett's multiple test). Note the significant reduction of stenosis % by treatment with TAS-301.

View larger version (94K): [in this window] [in a new window]   Fig. 2.   Coronary angiogram of left coronary artery after overstretch of the vessel in a micropig. A, immediately after the balloon injury. B, 4 weeks after the injury in control micropig. C, 4 weeks after the injury in a TAS-301 (100 mg/kg)-treated micropig. Note the remarkable reduction in the diameter of LAD coronary artery in the control 4 weeks after the injury and preservation of coronary artery diameter by the treatment with TAS-301.

Coronary Vasoreactivity. Table 2 shows the magnitude of vasoconstriction 4 weeks after angioplasty. Four weeks after the balloon injury, the vasoreactivity of the coronary artery was evaluated arteriographically after intracoronary administration of serotonin. At a dose of 10 µg/kg, serotonin caused significant hyper-reactive vasoconstriction at the angioplasty sites in control micropigs, whereas it did not induce remarkable vasoconstriction at control sites in normal micropigs, as previously described (Kadokami et al., 1996). The treatment with TAS-301 increased the diameter of coronary artery in a dose-dependent manner before administration of serotonin. Administration of the monoamine induced vasoconstriction at the angioplasty sites in TAS-301-treated micropigs. This serotonin-induced coronary vasoreactivity (constriction ratio of diameter) in each group of TAS-301-treated micropigs was similar to that seen in the control micropigs. This means that TAS-301 did not have any antagonistic effect against serotonin and did not change the vasoreactivity of the coronary artery.

Histological Analysis. Histopathological analysis was performed on all injured segments (four to five sections per vessel). As previously reported (Shi et al., 1996a), there was rupture of the internal elastic lamina, with neointima growth replacing the disrupted media 4 weeks after injury in all animals. The degree of injury achieved in TAS-301 treatment and control groups was probably rather uniform because all vessels included in the analysis demonstrated rupture of the internal elastic lamina and eccentric lesions. Furthermore, adventitial thickness increased compared with that of normal vessels. As depicted in Fig. 3, the most striking changes in adventitial dimensions were found in the regions adjacent to the site of medial injury. In contrast, the thickness of the adventitia opposite the medial injury or in other uninjured sections from the same vessels remained largely unaffected. Several independent morphometric measurements of the vessels were performed to assess the response to injury.

View larger version (64K): [in this window] [in a new window]   Fig. 3.   Cross-section of micropig coronary artery 4 weeks after balloon injury. A, control. B, TAS-301, 100-mg/kg treatment. A, large neointimal lesion that fills the gap in the disrupted media is present. The adventitia exhibits thickening in the vicinity of the medial injury compared with the thickness on the opposite site of the same section in control. B, note the reduction in adventitial thickness by TAS-301 treatment. A, adventitia; I, intima; M, media; L, lumen. Scale bar, 200 µm.

View larger version (27K): [in this window] [in a new window]   Fig. 4.   Effects of TAS-301 on histological parameter 4 weeks after balloon injury to micropig coronary artery. A, ratio of intimal area to medial area (%). B, ratio of adventitial area to medial area (%). C, residual lumen (%). Data are expressed as means ± S.E.M. of five micropigs. *P < .05, **P < .01 versus control (Dunnett's multiple test). Note the significant reduction in the ratio of adventitial area to medial area and increase in residual lumen (%) by treatment with TAS-301.

Adventitial Fibroblast Proliferation (In Vitro Assay). The hypercellularity of the adventitial layer and proliferation of fibroblasts are believed to be associated with the development of the thickened adventitia in the coronary artery injury model in micropigs, and TAS-301 caused a significant reduction in adventitial thickness 4 weeks after overstretch injury in this study. As bFGF and TGF-₁ are well known mitogens for fibroblasts, the effect of TAS-301 on the proliferation of micropig adventitial fibroblasts stimulated by these cytokines in vitro was examined (Fig. 5).

View larger version (31K): [in this window] [in a new window]   Fig. 5.   Effects of TAS-301 on bFGF- (A) and TGF-1- (B) induced BrdU incorporation into adventitial fibroblast cells. Data show BrdU incorporation (O.D.) induced by bFGF or TGF-1 treatment (mean ± S.E.M., n = 8 except for basal condition, n = 4). *P < .05, **P < .01 versus control (Dunnett's multiple test). ##P < .01 versus control (Aspin-Welch's t test). Note the potent inhibitory effect of TAS-301 on BrdU incorporation into the cells.

Gel Contraction by Adventitial Fibroblast (In Vitro Assay). Because the alteration of the adventitial fibroblast phenotype to the myofibroblastic one is believed to be associated with the development of a thickened adventitia and because the constrictive ability of myofibroblasts contributes to the narrowing of the vessel lumen after coronary artery injury, we decided to examine the 3D collagen gel contraction model using adventitial fibroblasts from the angioplasty sites 4 weeks after injury in micropigs (INJ-fibroblasts) as an extrapolation model for the in vivo constrictive remodeling resulting in restenosis after angioplasty. In this model, we found that the degree of gel contraction was dependent on both serum concentration and cell density; and we determined the optimal conditions to be an FBS concentration of 0.1% and a cell concentration of 1.5 × 10⁵ cells/ml (data not shown).

View larger version (30K): [in this window] [in a new window]   Fig. 6.   Effect of TAS-301 on collagen gel contraction by INJ-fibroblasts. Collagen gels were prepared by mixing 1.5 × 105 cells/ml of INJ-fibroblasts (after preincubation of the cells with various doses of TAS-301 for 2 h) and 0.75 mg/ml type I collagen and then incubated in the presence of 0.1% FBS for 24 h. Data are expressed as means ± S.E.M. of six collagen gels. Note the potent reduction in the contractile ability of INJ-fibroblasts by the treatment with TAS-301. *P < .05, **P < .01 versus control (Dunnett's multiple test). ##P < .01 versus control (Aspin-Welch's t test).

	Discussion

Top Abstract Introduction Experimental Procedures Results Discussion References

Recent advances in interventional cardiology have led to more aggressive strategies to relieve coronary obstruction. Thus, a deep medial injury that may potentially affect the adventitia appears to be common in clinical practice. The possibility of myofibroblast formation and the deposition of extracellular matrix in the adventitia after coronary artery injury in humans may lead to vascular tissue retraction (den Heijer et al., 1994; Shi et al., 1996a). In fact, recent findings obtained by the intravascular ultrasound technique appear to corroborate this possibility (Di Mario et al., 1994). The failure of many pharmacological approaches to reduce restenosis in clinical settings has stimulated considerable interest in site-specific therapy after coronary angioplasty. Furthermore, the successful application of a vascular stent to the atherosclerotic plaque, resulting in less frequency of restenosis, has shown the importance of preserving the vessel in the dilated state; but its use is still limited, by few sites conductive to application, induction of coagulation, and restenosis due to intimal hyperplasia (Marso et al., 1999). The involvement of the adventitia in the vascular repair process may require the development of strategies allowing for the administration of potentially active compounds to this site (Edelman et al., 1993; Hadeishi et al., 1994).

On the basis of this background, we synthesized TAS-301 as a first trial drug to target constrictive remodeling and developed it for the prevention of restenosis after angioplasty. The major finding of this study is that TAS-301 reduced the vascular response to balloon injury without changing vasoreactivity, as demonstrated mainly by a larger lumen diameter and decreased A/M ratio as judged by angiographical and histopathological analysis 4 weeks after injury.

Several recent studies on the vessel response to injury as it pertains to clinical restenosis have suggested that two measurements of this response should be independently evaluated: the growth of neointima, as a marker of the biological response to injury, and the follow-up luminal diameter, as a marker to the clinical significance of the result. Because there was no significant reduction in intimal area or I/M ratio in this study, we cannot relate the increase in lumen diameter observed to a change in smooth muscle cell proliferation and intimal mass. Recent analysis of the relation between intimal mass and lumen diameter has failed to show a significant correlation between these variables. Scott et al. (1996) indicated that cell proliferation at the earliest time point after angioplasty of porcine coronary arteries was greater in the adventitia than in the media and that the adventitia may be an important region with respect to the first wave of growth after angioplasty of coronary arteries. Furthermore, they hypothesized that the cells that proliferate in the adventitia may also contribute to vascular lesion formation by synthesizing growth and/or differentiation factors. Their study suggests that the adventitia may play a role in vascular lesion formation by contributing to the cellular mass of the neointima and the synthesis of growth factors and also the adventitia may contribute to vascular remodeling and constriction of the external elastic lamina through an accumulation of myofibroblasts in the adventitia surrounding the injury site.

In view of the striking similarities between the process of wound healing and the response of the adventitial fibroblasts to injury, we tested TAS-301 on the basis of this wound-healing restenosis hypothesis by using adventitial fibroblast-populated collagen gel matrix that forms a three-dimensional lattice using adventitial fibroblasts. TAS-301 inhibited the gel contraction by fibroblasts taken from the overstretched coronary artery 4 weeks after injury, to the same extent as neutralizing antibody against TGF-_1,2,3 (E. Sasaki, Y. Yamasaki, Y. Tanahashi, Y. Muranaka, H. Terakawa, K. Miyoshi, N. Oda, H. Miyake, and N. Matsuura, submitted). The change of phenotype from fibroblast to myofibroblast may have contributed to the constrictive remodeling resulting in renarrowing of the coronary artery after angioplasty, and TAS-301 might have prevented this change.

Furthermore, TAS-301 showed potent inhibition of mitogen (bFGF and TGF-₁, well known mitogen for fibroblast)-induced proliferation of adventitial fibroblasts. Previously, we reported that TAS-301 potently inhibited the migration of rat vascular smooth muscle cells induced by PDGF-BB, insulin-like growth factor-1, and heparin-binding epidermal growth factor-like growth factor and also inhibited the proliferation induced by bFGF in vitro (Muranaka et al., 1998). When rat vascular smooth muscle cells were stimulated with PDGF-BB, TAS-301 inhibited the transient and sustained increase in free intracellular Ca²⁺ concentration, as monitored by Fura-2 fluorescence (Sasaki et al., 2000). This finding indicates that TAS-301 inhibits intracellular Ca²⁺ mobilization via intracellular Ca²⁺ release and extracellular Ca²⁺ influx induced by PDGF-BB. Considering that TGF- and bFGF induce extracellular Ca²⁺ influx in fibroblasts (Muldoon et al., 1988; Munaron et al., 1995), we guess that TAS-301 inhibits the proliferation of adventitial fibroblasts by regulating TGF- and bFGF-induced intracellular Ca²⁺ mobilization.

Preliminarily, we determined that the plasma level of TAS-301 given to micropigs at the dose of 100 mg/kg was maintained at over 3 µM during the experimental period, because TAS-301 was administered daily as a dietary mixture. Because the concentration of TAS-301 we used in the in vitro experiment is almost consistent with its plasma concentration in the micropigs, these in vitro effects also might have resulted in the reduction in the adventitial thickness and increment of residual lumen diameter of the coronary artery 4 weeks after the angioplastic injury. TGF-₁ is involved in tissue repair by modulating the growth of mesenchymal cells, augmenting the synthesis of several extracellular matrix proteins, and facilitating both migration and proliferation of fibroblasts (Miyazaki et al., 1998). Myofibroblasts represent highly specialized mesenchymal cells that play a central role in tissue repair. It has also been suggested that TGF-₁ provides the signal for fibroblasts to acquire a differentiated phenotype that imparts synthetic and mechanical properties (Desmouliere et al., 1993; Arora and McCulloch, 1994; Shi et al., 1996c). Previously, we found that the strong 3D gel-contracting ability of fibroblasts from injured coronary artery could be blocked by antibody against TGF-_1,2,3. Further investigation on the contribution of TGF-₁ in this model is needed to understand the precise mechanism of TAS-301 action in vascular remodeling.

In summary, our new synthesized TAS-301, a first compound developed for targeting constrictive remodeling, inhibited coronary artery stenosis of micropigs after injury as judged angiographically. This effect of TAS-301 appears to be due to a reduction in adventitial fibroblast proliferation and lessened contractile ability of myofibroblasts. This highly potent activity of TAS-301 suggests the potential use of this drug for the prevention of restenosis after angioplasty via regulation of arterial remodeling by adventitial fibroblasts and the need to examine the therapeutic usefulness of this drug in clinical trials.