Inhibitory effects of tranilast on expression of transforming growth factor-β isoforms and receptors in injured arteries

Atherosclerosis 137 (1998) 267 – 275

Inhibitory effects of tranilast on expression of transforming growth factor-b isoforms and receptors in injured arteries Michael R. Ward *, Takayuki Sasahara, Alex Agrotis, Rodney J. Dilley, Garry L. Jennings, Alex Bobik Cell Biology Laboratory and Alfred-Baker Medical Unit, Baker Medical Research Institute, Commercial Rd, Prahran, VIC 3181, Australia Received 21 May 1997; received in revised form 14 October 1997; accepted 10 November 1997

Abstract Tranilast (N(3,4-dimethoxycinnamoyl)anthranilic acid), an agent which in cell culture inhibits transforming growth factor-b (TGF-b) secretion and antagonises the effects of TGF-b and platelet-derived growth factor (PDGF) on cell migration and proliferation, has been reported to reduce the incidence of restenosis after angioplasty in angiographically validated human clinical trials. We investigated in a rat model of balloon angioplasty whether tranilast’s effects in vivo could be attributed to inhibition of expression of TGF-b and/or its receptor types. Using a standardised reverse transcriptase-polymerase chain reaction (RT-PCR) assay, we examined the effects of three doses of tranilast (25, 50 and 100 mg/kg) on the expression of two TGF-b isoforms, the types I and II TGF-b receptors and two putative TGF-b responses, induction of integrins av and b3 mRNA, 2 h after oral administration and 26 h after vessel injury. Tranilast attenuated in a dose-dependent and reversible manner the injury-induced increases in mRNA levels encoding TGF-b1, TGF-b3, two type I TGF-b receptors ALK-5 and ALK-2, and the type II receptor TbRII. At the highest dose mRNA levels encoding TGF-b1 and TbRII were attenuated to levels approaching or below those observed in uninjured vessels. Messenger RNAs encoding TGF-b3, ALK-5 and ALK-2 were all attenuated by between 70 and 74% (all P B0.05). Tranilast also attenuated in a reversible manner the elevations in mRNA levels for integrins av and b3 observed after vessel injury, by 90 and 72%, respectively. We also investigated, in cultured smooth muscle cells derived from injured carotid arteries, the extent to which tranilast (300 mg/l) attenuated any increases in expression of type I and type II receptors stimulated by PDGF-BB and TGF-b1, growth factors implicated in smooth muscle cell migration and proliferation in injured vessels. Increases in mRNA levels of the type I receptors ALK-5 and ALK-2 induced by PDGF-BB and TGF-b1 were almost completely prevented by tranilast. Tranilast also prevented the PDGF-BB induced increases in TbRII but only partially inhibited the TGF-b1 induced upregulation of TbRII. We conclude that tranilast can inhibit transcriptional mechanisms associated with the upregulation of TGF-b and its receptor types in balloon catheter injured vessels. It is possible that these mechanisms contribute to its ability to reduce the frequency of restenosis after angioplasty. © 1998 Elsevier Science Ireland Ltd. All rights reserved. Keywords: TGF-bs; TGF-b receptors; Tranilast; Angioplasty; Integrins; Restenosis

1. Introduction Transforming growth factor-beta1 (TGF-b1) has been implicated in the development of human restenosis after angioplasty [1–3] as well as in the neointimal fibrocellular lesions which develop in balloon catheter injured * Corresponding author. Tel.: + 61 3 95224333; fax: + 61 3 95211362; e-mail: [email protected]

arteries of experimental animals [4–9]. In animals the rise in TGF-b1 expression is apparent early after injury and remains elevated during healing and remodelling of the vessel [4,8,9]. Pharmacological studies involving intravenous administration of TGF-b1 to animals immediately after balloon catheter injury or transfection of TGF-b1 cDNA into the injured vessel wall indicate that TGF-b1 can promote both the formation and growth of the neointima [5,7]. In humans mRNA encoding TGF-

0021-9150/98/$19.00 © 1998 Elsevier Science Ireland Ltd. All rights reserved. PII S0021-9150(97)00275-X

268

M.R. Ward et al. / Atherosclerosis 137 (1998) 267–275

b1 is also increased in restenotic lesions compared to both normal vessels and de novo atherosclerotic lesions [1,2]. These increases are most striking in recurrent restenotic lesions and saphenous vein graft restenotic lesions where restenosis rates are highest [3]. Despite these studies linking TGF-b with restenosis and the development of neointimal fibrocellular lesions in balloon injured vessels, only limited direct interventional trials examining the effects of inhibiting TGF-b actions on neointima formation in injured vessels in experimental animals and on restenosis rates in humans have been carried out. In experimental animals TGF-b1 neutralising antibodies partially attenuated neointima formation [6], whilst in humans tranilast (N(3,4dimethoxycinnamoyl) anthranilic acid), an agent reported to inhibit TGF-b secretion by keloid fibroblasts, reduces the incidence of angiographic restenosis when administered for 3 months after angioplasty and stenting [10–12]. Tranilast also exerts multiple inhibitory effects on cultured vascular smooth muscle cells which have been stimulated by TGF-b1 or platelet-derived growth factor (PDGF-BB) [13,14]. Thus, it is possible that its mechanism of action in intact injured vessels may include the inhibition of expression of TGF-b isoforms and their signalling systems. In the present study we tested this hypothesis by examining how tranilast affected the expression of mRNA encoding TGF-b1, TGF-b3, and their types I and II receptors after balloon catheter injury to the rat carotid artery. Since TGF-b1 has been reported to elevate expression of integrins av and b3 in cultured smooth muscle cells [15,16], the ability of tranilast to inhibit their expression after injury of the vessel was also examined. We demonstrate that tranilast can prevent activation of the TGF-b system in the injured vessel by impairing expression of the genes coding for TGF-b1 and TGF-b3 and the types I and II TGF-b receptors; putative early effects of TGF-bs, elevations in expression of integrins av and b3, are also attenuated 26 h after balloon catheter injury to the rat carotid, a time when TGF-b1 mRNA is already markedly upregulated [4,9]. We also demonstrate in cell culture that tranilast’s ability to inhibit induction of TGF-b receptors in smooth muscle is due to inhibitory actions on TGF-b and growth factors activating receptor tyrosine kinases, such as PDGF.

Research Institute and Alfred Hospital Animal Experimentation Committee. Rats were administered either tranilast (25, 50 or 100 mg/kg) (Kissei Pharmaceuticals, Hotaka, Japan), suspended in vehicle (2% Dihydromethoxycellulose solution; Sigma Pharmaceuticals, Melbourne Australia), or vehicle alone by gavage 2 h prior to and 24 h after balloon injury. Animals were sacrificed 2 h after administration of the second dose of tranilast when plasma concentrations of the drug after oral administration are at their peak (100–130 mg/ml, personal communication, Kissei Pharmaceutics, Japan). To assess reversibility of tranilast’s effects on the different mRNAs expressed after balloon catheter injury, four additional animals were sacrificed 6 and 12 h after administration of the 100 mg/kg dose of tranilast. Relative mRNA levels from the vessels of these animals were compared with those from animals given vehicle alone.

2.2. Balloon catheter injury and tissue collections Injury to the rat carotid artery was carried out as has been described in detail previously using a balloon catheter [17]. Briefly, after anaesthetising the rats with pentobarbitone (30 mg/kg), methohexitone (40 mg/kg) and atropine sulphate (3 mg/kg), administered by intraperitoneal (i.p.) injection, a midline neck incision was made and blunt dissection to the carotid bifurcation was performed. Through an arteriotomy in the external carotid a 2F Fogarty Arterial Embolectomy catheter (Baxter, CA) was passed to the aortic arch. The balloon was then inflated with 25 ml of saline and withdrawn with a rotating action to the bifurcation. This procedure was performed three times before the balloon catheter was removed and the external carotid artery ligated. The incision was then closed and the animals allowed to recover from the surgery in a humidified warmed chamber for 1–2 h. At the end of the experiments all animals were sacrificed 26–36 h after the operation, by administering pentobarbitone (60 mg/ kg i.p.). The left carotid artery of all animals and also the right carotid artery from vehicle only treated animals were rapidly dissected free of connective tissue, snap frozen in liquid nitrogen and stored at − 70°C for mRNA analysis.

2.3. Re6erse transcriptase polymerase chain reaction (RT-PCR) 2. Methods

2.1. Animals, study design and drug administrations Male Sprague Dawley rats weighing 400 – 500 g were obtained from a colony maintained at the Baker Medical Research Institute (BMRI), Melbourne, Australia. All procedures were approved by the Baker Medical

2.3.1. RNA Extraction Total RNA was isolated from vessels using the phenol-chloroform extraction method of Chomczynski and Sacchi [18], then resuspended in 20 ml sterile distilled water. The absence of any contaminating genomic DNA in the RNA was ensured by incubating the vessel extracts with 2 units of DNase (Stratagene) at 37°C for

M.R. Ward et al. / Atherosclerosis 137 (1998) 267–275

269

Table 1 Oligonucleotide primers used in RT-PCR to amplify specific mRNAs together with the number of cycles used for quantitation and sizes of the amplified products Target regiona

Product sizeb

Cycles

784-1224

441

36

708-1266

359

28

79-519

443

34

128-459

332

34

1214-1552

339

30

42-531

490

30

S: CAC TAC TAT GGA TTA CCC ATC TCT GG AS: GTT GTT GAG GCA GGT GGC ATT GAA GG

168-447

280

34

S: CCT GAG GAA GAA GTT TGC CC AS: CTT GTT GAG CTT CAC AAA GGT GCC

143-428

286

23

Target mRNA

Primers

TGFb1 [22]

S: CAA AGA CAT CAC ACA CAG TA AS: AG GTG TTG AGC CCT TTC CAG

TGFb3 [23]

S: GAG ATC CAT AAA TTC GAC ATG ATC CAG GGG AS: ATT TCC AGA CCC AAG TTG GAC TCT CT

ALK-5 [24]

S: GCG AAG GCA TTA CAG TGT TTC TGC AS: CTC TGA AAT GAA AGG GCG ATC TAG TGA TGG

ALK-2 [25]

S: TCT GTG CTA ATG ATG ATG GCT CTC C AS: TTC TGC GAT CCA GGG AAG GAT TTC C

TbRII [26]

S: ACC CTA CTC TGT CTG TGG ATG ACC AS: TCA GTG TCT CAC ACA CGA TCT GGA TGC C

Integrin av [27]

S: TAT TGG GGA TGA CAA CCC TCT GAC C AS: CTC ATA GAT GTG CTG AAC AGG C

Integrin b3 [27] L7 [28]

S, sense primer; AS, antisense primer.a Nucleotides of published sequence.b Base pairs.

30 min. RNA was then reprecipitated with 2 ml 2 M sodium acetate and an equal volume (24 ml) of isopropanol and the RNA sedimented by centrifugation at 14000 rpm (Eppendorf, Germany) for 15 min at 4°C. After aspiration of the supernatant, the RNA pellet was washed by resuspension in 70% aqueous ethanol and centrifugation at room temperature. Upon removing the ethanolic supernatant the pellet was dried at 37°C for 20 min. The purified RNA was dissolved in sterile water, quantitated by spectrophotometry at 260 nm and its concentration adjusted to 266 ng/ml by addition of sterile water.

2.3.2. RT-PCR Standardised RT-PCR was performed in a manner which ensured that PCR product amount reflected mRNA levels in the original tissue as previously described [19]. Using this method the amount of product amplification to RNA relationship was linear (r 2 values ranging from 0.94 to 1.00) for the cycle number used over the range 100 – 400 ng of total RNA [20]. The overall procedure was further standardised by expressing the amount of PCR product for each target mRNA relative to the amount of PCR product formed for L7, a ribosomal protein which is encoded by a non-inducible cell cycle-independent gene [21]. Primers for the TGF-b isoforms, TGF-b receptors and integrins were designed according to the published cDNA sequences [22 – 28] using the program Primer Detective (Clontech Labs, CA) and the follow-

ing criteria: GC content 45–55%, melting point 76– 83.5°C, filtering hairpins and 3% homologies. The sizes of the regions amplified together with the oligonucleotides used and number of cycles performed to amplify and quantitate each target mRNA are summarised in Table 1; in addition the identity of each amplified DNA fragment was confirmed using either restriction enzyme digestion or nucleotide sequencing after cloning into pGEM-T vectors (Promega, WI). Conditions for the RT-PCR reactions were as follows: each RT incubation mixture contained 1 ml 25 mM MgCl2, 0.5 ml 10 × PCR buffer, 2 ml dNTP mix (containing 2.5 mM of each of dATP, dCTP, dGTP and dTTP), 0.25 ml of 50 mM Random Hexamers, 0.25 ml 20 U/ml RNase Inhibitor, 0.25 ml 50 U/ml MuLV Reverse Transcriptase, and 0.75 ml of 0.27 mg/ml total RNA (200 ng). After equilibration at room temperature for 10 min reverse transcription was performed on a Hybaid Omnigene Thermal Cycler at 42°C for 15 min, followed by a 95°C incubation for 5 min. Samples were then placed on ice. PCR was performed with each reaction mixture containing 5 ml of RT product, 1 ml primers (containing 10 mM of the sense and antisense primers), 0.5 ml 25 mM MgCl2, 2 ml 10×PCR buffer, 0.125 ml 5 U/ml Amplitaq ® DNA Polymerase and 0.125 ml 1 mg/ml Anti-Taq DNA Polymerase Antibody (MAb 8C1C Technogene, Moscow) and 16.25 ml sterile distilled water. All components except for the antibody were from a GeneAmp ® RNA PCR Core Kit (Perkin Elmer, NJ).

270

M.R. Ward et al. / Atherosclerosis 137 (1998) 267–275

Fig. 1. Dose-dependency of the effects of tranilast on TGF-b isoform mRNA levels in the rat carotid artery 26 h after injury and 2 h after tranilast administration. Graphs represent TGF-b1 and TGF-b3 mRNA levels relative to the ribosomal protein L7 mRNA and are expressed as a percentage of those obtained from injured vessels in vehicle treated rats. V represents injured vessels from vehicle treated rats and N uninjured vessels from vehicle treated rats. Results are the means9 S.E.M. of five animals in each group. *PB 0.05 from V; **PB 0.05 from N. Representative gels are shown in the top right corner of each graph. M shows FX174 molecular weight markers, V is the RT-PCR product from a balloon injured vessel from a vehicle treated rat, and T is the RT-PCR product from a balloon injured vessel treated with 100 mg/kg tranilast.

Each cycle consisted of the following stages: 94°C for 30 s, 60°C for 1 min and 72°C for 2 min with a prolonged extension stage after the final cycle of 72°C for 8 min. PCR products were then electrophoresed on 2% Agarose gels (Progen, Darra, QLD, Australia) together with HaeIII digested FX174 DNA as size markers (Promega, WI) at 120 mV. They were photographed under ultraviolet light with positive/negative film (Polaroid 665) and the intensities on the negatives were quantitated using a laser densitometer (LKB 2222-010 Ultrascan XL, LKB, Bromma, Sweden).

2.4. Cell culture Vascular smooth muscle cell cultures were prepared by enzyme digestion using rat carotid arteries injured 24 h earlier with a balloon catheter as previously described [29]. Briefly, after removal of the adventitia from the vessels with the aid of a dissecting microscope, the medial smooth muscle layers were subjected to enzyme digestion with collagenase and elastase. Cells were harvested from the resultant suspension by centrifugation. They were resuspended in Dulbecco’s Modified Eagle’s Medium (DMEM) containing 10% fetal calf serum (FCS-DMEM) prior to plating onto 30 mm tissue culture plates. In culture these cells exhibited the characteristic hills and valleys pattern typical of many cultured vascular smooth muscle cells [30]. Near confluent layers of cells were deprived of serum for 24 h prior to exposure for 24 h to either DMEM alone, or DMEM containing platelet-derived growth factor-BB (PDGF-BB, 200 ng/ml, Sigma, St Louis, MO) or TGF-b1 (2 ng/ml, Celltrix, Palo Alto, CA), both without and with tranilast 300 mg/ml.

2.5. Statistical methods Differences or effects of the various treatments on mRNA levels in tissues or cells were detected using a Kruskal-Wallis One Way Analysis of Variance on Ranks and then individual pairs of groups compared on a post-hoc basis using the Mann-Whitney rank sum test. Normality was analysed using the KolmogorovSmirnov test (Sigmastat, Jandel Scientific). Differences were considered statistically significant if PB 0.05.

3. Results

3.1. Tranilast and TGF-b isoform expression after 6essel injury Initially we examined the effects of increasing doses of tranilast on the elevations in mRNAs encoding TGF-b1 and TGF-b3 which occurred 26 h after balloon injury. In uninjured vessels TGF-b1 mRNA levels were low but 26 h after injury they were elevated approximately six-fold (Fig. 1). Tranilast, administered 2 h earlier attenuated this increase in a dose dependent manner; after administering the highest dose (100 mg/ kg) relative levels of mRNA encoding TGF-b1 were reduced by 91% compared with levels in vessels of the vehicle treated animals (PB 0.05; Fig. 1). TGF-b3 mRNA levels were also elevated 26 h after the injury by approximately five-fold and these were reduced after administering tranilast (100 mg/kg) by 74% (PB 0.05). Resultant mRNA levels encoding these two TGF-b isoforms in the balloon injured vessels of the tranilast (100 mg/kg) treated animals were not significantly dif-

M.R. Ward et al. / Atherosclerosis 137 (1998) 267–275

271

Fig. 2. Relative TGF-b receptor mRNA levels 2 h after different tranilast doses administered 26 h after balloon carotid injury. Graphs represent ALK-5, ALK-2 and TbRII mRNA levels relative to the ribosomal protein L7 mRNA and are expressed as a percentage of those obtained from injured vessels in vehicle treated rats. V represents injured vessels from vehicle treated rats and N uninjured vessels from vehicle treated rats. Results are the means9 S.E.M. of five animals in each group. *PB0.05 from V; **PB0.05 from N. Representative gels are shown in the top right corner of each graph. M shows FX174 molecular weight markers, V is RT-PCR product from a balloon injured vessel from a vehicle treated rat, and T is RT-PCR product from a balloon injured vessel treated with 100 mg/kg tranilast.

ferent from mRNA levels in uninjured carotid vessels (P \ 0.05; Fig. 1). Lower doses of tranilast induced smaller reductions in mRNA levels (Fig. 1).

3.2. Tranilast and the expression of TGF-b receptors Tranilast has been reported to attenuate multiple TGF-b responses in SMCs, including collagen production, proliferation and migration [13,14]. Accordingly we examined in vivo in the balloon catheter injured carotid artery whether tranilast administration might reduce the expression of either type I and/or type II TGF-b receptors. Twenty six hours after injuring the carotid artery relative mRNA levels encoding the type I receptors ALK-5 and ALK-2 were elevated 2.5-fold and 4-fold, respectively compared with uninjured arteries (Fig. 2). The highest dose (100 mg/kg) of tranilast, administered 2 h earlier, suppressed relative ALK-5 mRNA levels by 71% (PB 0.05) and relative ALK-2 mRNA levels by 74% (P B0.05). In addition, the greater than 10-fold increases in relative mRNA levels encoding TbRII observed in vehicle treated animals 26 h after injury were completely (\99%) suppressed by the 100 mg/kg dose of tranilast (P B0.05). Relative TbRII mRNA levels in vessels of these tranilast treated animals were not different from those in uninjured vessels (P\0.05; Fig. 2). Lower doses of tranilast reduced mRNA levels encoding the types I and II receptors to a lesser extent, in a dose-dependent manner (Fig. 2). Since elevations in mRNAs encoding the type I (ALK-5) and type II (TbRII) receptors can be induced by either TGF-b1 or growth factors which activate receptor tyrosine kinases [20,31,32], we examined in cell culture the extent to which the ability of tranilast to suppress the induction of the mRNAs in vivo might

reflect inhibition of receptor tyrosine kinase and/or receptor serine threonine kinase activities. Incubation of the cultured smooth muscle cells with PDGF-BB for 24 h elevated mRNA levels encoding ALK-5 by 5.8fold and ALK-2 mRNAs 2.4-fold (Fig. 3). Similarly, relative mRNA levels encoding TbRII were elevated four-fold by PDGF-BB (Fig. 3). Simultaneous incubation of the smooth muscle cells with tranilast (300 mg/ml) suppressed the increases in mRNAs encoding ALK-5, ALK-2 and TbRII by 92, 89 and 96%, respectively (Fig. 3). Twenty four hour incubation of the smooth muscle cells with TGF-b1 also induced similar elevations in the relative mRNA levels encoding ALK5, ALK-2 and TbRII, 8.3-, 2.5- and 5.2-fold, respectively. Co-incubation of the cells for 24 h with tranilast (300 mg/ml) suppressed the elevations in mRNAs encoding ALK-5, ALK-2 and TbRII by 100, 88 and 48%, respectively (Fig. 3). In all instances mRNA encoding the ribosomal protein L7 was unaltered, either by the two growth factors or by the simultaneous presence of tranilast (not shown).

3.3. Effect of tranilast on integrin mRNA le6els Since TGF-b1 elevates the expression of integrins av and b3 in cultured vascular smooth muscle cells [15,16], and tranilast suppressed elevations in mRNAs encoding TGF-b1 and its types I and II receptors in the injured vessels, we examined whether tranilast might also suppress the expression of mRNAs encoding these two integrins in the balloon catheter injured vessels. Twenty six hours after balloon catheter injury to the carotid artery the mRNAs encoding integrins av and b3 were increased by approximately 8- and 4.5-fold, respectively (PB 0.05 from uninjured). Tranilast (100 mg/kg) administered 2 h earlier greatly suppressed these mRNA

272

M.R. Ward et al. / Atherosclerosis 137 (1998) 267–275

Fig. 3. Effects of tranilast on growth factor induced expression of TGF-b receptors in cultured smooth muscle cells. Graphs represent ALK-5, ALK-2 and TbRII mRNA levels relative to the ribosomal protein L7 mRNA. Results are expressed as a percentage of those obtained from cultured smooth muscle cells incubated with DMEM alone. P represents those incubated with PDGF-BB (200 ng/ml in DMEM), T those incubated with TGF-b1 (2 ng/ml in DMEM), PT those incubated with PDGF-BB and tranilast (300 mg/ml) in DMEM and TT those incubated with TGF-b1 and tranilast in DMEM. C represents those incubated with DMEM alone. Results are the means 9S.E.M. of three experiments (*PB0.05 vs. C; †P B 0.05 vs. growth factor stimulated cells in the absence of tranilast).

elevations in integrins av and b3, by 89 and 70%, respectively (both P B 0.05 from controls) while the lower doses of tranilast, 25 and 50 mg/kg, suppressed the rises in integrin av and integrin b3 mRNA levels to a lesser extent (Fig. 4).

3.4. Re6ersibility of tranilast actions To ensure that the vascular actions of tranilast on TGF-b isoform and receptor mRNA levels were not due to toxic effects on the smooth muscle cells, we also evaluated the reversibility of its inhibitory effects on the different mRNAs. Six hours after administering the 100 mg/kg dose of tranilast, the levels of mRNA encoding TGF-b3, ALK-5, ALK-2, TbRII, and integrins av and b3 in carotid arteries injured 30 h earlier were similar to those in injured arteries of vehicle treated animals; only a 30% reduction in mRNA encoding TGF-b1 persisted (data not shown). Twelve hours after administering tranilast all seven mRNA species were at levels observed in injured vessels of vehicle treated animals (Fig. 5).

4. Discussion In this study we have demonstrated that tranilast can down-regulate components of an activated TGF-b system after balloon catheter injury to the rat carotid artery. Its effects include the suppression of elevations in mRNAs encoding TGF-b1, TGF-b3 and their receptors ALK-5, ALK-2 and TbRII; elevations in integrins av and b3 mRNA are also reduced. Many of these actions of tranilast appear to be mediated either through inhibition of receptor tyrosine kinases and

possibly receptor serine threonine kinase activities. While the response to injury in the normal rat carotid artery does not take into account the influence of atheromatous components in the process, it is likely that the ability of tranilast to prevent activation of the TGF-b system in mechanically injured normal blood vessels would also occur in atherosclerotic lesions and contribute to its ability to reduce the frequency of restenosis after coronary artery angioplasty. In the injured rat carotid artery upregulation of TGF-b1 and TGF-b3 are dependent on increased levels of gene transcription [4,20]. Maximal increases in mRNA levels become apparent 24–48 h after the injury and their protein products promote smooth muscle cell migration and neointima development [5,33]. Administration of tranilast rapidly suppresses (within 2 h) these elevations in mRNA levels. Both reductions in the rates of gene transcription and/or accelerated degradation of the respective mRNAs could account for suppression of these mRNAs. In this investigation we did not examine the extent to which these two mechanisms contribute to reductions in the different mRNAs, although their functional consequences are likely to be similar. In vivo the elevations in vascular smooth muscle TGF-b1 and TGF-b3 mRNA have been shown to be dependent on tyrosine kinase activities [20] and in cultured smooth muscle cells growth factors activating receptor tyrosine kinases, such as PDGF-BB or FGF-2, induce large elevations in TGF-b1 [34]. Evidence indicating that tranilast is capable of inhibiting smooth muscle cell migration induced by both PDGF-BB and TGF-b1 suggest that inhibition of tyrosine kinases might be important in this regulation [14]. Clearly further studies will be required to support this suggestion.

M.R. Ward et al. / Atherosclerosis 137 (1998) 267–275

273

Fig. 4. Integrin mRNA levels 2 h after different tranilast doses and 26 h after balloon carotid injury. Graphs represent integrin av and integrin b3 mRNA levels relative to the ribosomal protein L7 mRNA and are expressed as a percentage of those obtained from injured vessels in vehicle treated rats. V represents injured vessels from vehicle treated rats and N uninjured vessels from vehicle treated rats. Results are the means9S.E.M. of five animals in each group. *PB 0.05 from V; **PB 0.05 from N. Representative gels are shown in the top right corner of each graph. M shows FX174 molecular weight markers, V is RT-PCR product from a balloon injured vessel from a vehicle treated rat, and T is RT-PCR product from a balloon injured vessel treated with 100 mg/kg tranilast.

Tranilast also has the potential to directly prevent the effects of the TGF-b isoforms on smooth muscle cells within the injured vessels by suppressing any elevations in mRNAs encoding its various receptors. TGF-b isoforms interact with several surface membrane signalling and binding proteins. Both the types I and II receptors binding TGF-bs are transmembrane serine threonine kinases which interact to form heteromeric complexes containing either one or two molecules of each of the types I and II receptors bound to each TGF-b molecule [35]. Within this complex it is generally assumed that the type II receptor transphosphorylates the type I receptor, and it is this phosphorylated receptor which determines the nature of the cellular responses [36]. Tranilast was found to suppress, in a dose dependent manner, the elevations in mRNAs encoding the type I receptors and also the type II receptor in the balloon injured arteries. The highest dose of tranilast (100 mg/ kg) was most effective, completely preventing these elevations in the respective mRNAs. The more striking reduction in mRNA levels of the type II receptor by tranilast when compared with its effects on mRNA levels of the type I receptors may potentially be important considering recent suggestions that the ratio of type I to type II TGF-b receptors is critical in determining the biological responses [37]. Multiple mechanisms are likely to contribute to this ability of tranilast to suppress the elevations in mRNAs encoding the receptors in injured vessels; in cell culture tranilast abolished PDGF-BB and TGF-b1 induced increases in ALK-5 and ALK-2 mRNA levels. The PDGF induced elevations in TbRII mRNA were also largely abolished but those induced by TGF-b1 were not significantly

reduced. We did not investigate the concentration-dependency of these tranilast inhibitory responses in the cultured smooth muscle cells. These inhibitory effects were relatively short in duration, being highly prominent 2 h after administration and largely undetectable after 6 h. Hence, to reduce TGF-b bioactivity, and putatively neointima formation, over a full 24-h period would require a larger dose than 100 mg/kg/day, and would probably be most effective when administered in multiple doses or as a sustained release preparation. As drug pharmacokinetics in other species are likely to be different to those in the rat, these relationships would need to be determined in each animal to optimally suppress the effects of TGF-bs. Recent studies in humans suggest significant roles for TGF-b in restenotic lesions. TGF-b1 and TGF-b3 mRNA levels have been reported to be significantly increased in restenotic lesions [1–3]. There is also evidence that the TGF-b system is highly active in such lesions. For example, the expression of big-h3, a TGFb inducible cell adhesion molecule, is also elevated in restenotic lesions [38]. In addition antagonism of other proteins implicated in vascular smooth muscle cell migration, such as the TGF-b inducible integrins av and b3, has also been reported to reduce the frequency of restenosis [39]. Here we also provide evidence that tranilast is capable of suppressing the involvement of integrins in the formation of fibrocellular lesions after vessel injury by preventing induction of their mRNAs, most probably via inhibition of receptor tyrosine kinases and/or receptor serine threonine kinase actions.

274

M.R. Ward et al. / Atherosclerosis 137 (1998) 267–275

Fig. 5. Agarose gels demonstrating the reversibility of tranilast’s effects on mRNA levels 12 h after its administration. Each panel depicts FX174 molecular weight markers (left lane) with PCR product from the carotid mRNA of vehicle treated animals (middle lane) and tranilast treated animals (right lane). Tranilast and vehicle were administered 24 h after balloon injury was performed. The RNA (200 ng) from the injured vessels was used for RT-PCR as described.

5. Conclusion In summary, we have demonstrated that tranilast, an agent which reduces the incidence of restenosis after percutaneous translumenal coronary angioplasty, causes dose-dependent and reversible reductions in mRNAs encoding two TGF-b isoforms and their receptors after balloon injury of the rat carotid artery. Part of its ability to reduce the incidence of restenosis may be mediated by inhibition of growth factor responses in injured vessels such as increases in integrins av and b3. Other yet-to-be defined effects dependent on receptor tyrosine kinases and/or receptor serine threonine kinases may also contribute to its efficacy in reducing the incidence of restenosis after angioplasty.

[3]

[4]

[5]

[6]

[7]

Acknowledgements Dr Michael Ward is a recipient of an NH and MRC postgraduate medical scholarship. These studies have in part been funded by an NH and MRC institute block grant and an NHF Project grant. Tranilast was a gift from Kissei Pharmaceutical Company, Hotaka, Nagano, Japan.

References [1] Nikol S, Isner JM, Pickering G, Kearney M, Leclerc G, Weir L. Expression of transforming growth factor-b1 is increased in human vascular restenosis lesions. J Clin Invest 1992;90:1582 – 92. [2] Ueda H, Imazu M, Sumil K, Yamamoto H, Ono K, Tadehara F, Hayashi Y, Yasui W, Yamakido M. Association of trans-

[8]

[9]

[10]

[11]

forming growth factor-beta isoforms with coronary plaque in patients with coronary ischaemia. J Am Coll Cardiol 1997;29(Supp I):155A. Nikol S, Ramee SR, Benenati JF, Leclerc G, Weir L. Transforming growth factor-b1 expression is significantly increased in saphenous vein graft with multiple versus single restenosis events (Abstract). Circulation 1992;86(Supp I):I – 21. Majesky MW, Lindner V, Twardzik D, Schwartz SM, Reidy MA. Production of transforming growth factor b1 during repair of arterial injury. J Clin Invest 1991;88:904 – 10. Nabel EG, Shum L, Pompili VJ, Yang Z-Y, San H, Shu HB, Liptay S, Gold L, Gordon D, Derynck R, Nabel GJ. Direct transfer of transforming growth factor-b1 gene into arteries stimulates fibrocellular hyperplasia. Proc Natl Acad Sci USA 1993;90:10759– 63. Wolf YG, Rasmussen LM, Ruoslahti E. Antibodies against transforming growth factor-b1 suppress intimal hyperplasia in a rat model. J Clin Invest 1994;93:1172 – 8. Kenzaki T, Tamura K, Takahashi K, Saito Y, Akikusa B, Oohashi H, Kasayuki N, Ueda M, Morisaki N. In vivo effect of TGF-b1. Enhanced intimal thickening by administration of TGF-b1 in rabbit arteries injured with a balloon catheter. Arterioscler Thromb Vasc Biol 1995;15:1951 – 7. Wysocki SJ, Zheng M-H, Fan Y, Lamawansa MD, House AK, Norman PE. Expression of transforming growth factor-b1 (TGF-b1) and urokinase-type plasminogen activator (u-PA) genes during arterial repair in the pig. Cardiovasc Res 1996;31:28 – 36. Kim S, Kawamura M, Wanibuchi H, Ohta K, Hamaguchi A, Omura T, Yukimura T, Miura K, Iwao H. Angiotensin II type I receptor blockade inhibits the expression of immediate-early genes and fibronectin in rat injured artery. Circulation 1995;92:88 – 95. The TREAT Study Investigators. The impact of tranilast on restenosis following coronary angioplasty: the Tranilast Restenosis Following Angioplasty Trial (TREAT). Circulation 1994;90:I – 652. The TREAT-2 Study Investigators. The impact of tranilast on restenosis following coronary angioplasty: the second Tranilast Restenosis Following Angioplasty Trial (TREAT 2). Circulation 1996;94:I – 620.

M.R. Ward et al. / Atherosclerosis 137 (1998) 267–275 [12] 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:I– 620. [13] Tanaka K, Honda M, Kuramochi T, Morioka S. Prominent inhibitory effects of tranilast on migration and proliferation of and collagen synthesis by vascular smooth muscle cells. Atherosclerosis 1994;107:179–85. [14] Miyazawa K, Kikuchi S, Fukuyama J, Hamano S, Ujiie A. Inhibition of PDGF- and TGF-b1-induced collagen synthesis, migration and proliferation by tranilast in vascular smooth muscle cells from spontaneous hypertensive rats. Atherosclerosis 1995;118:213 – 21. [15] Brown SL, Lundgren CH, Nordt T, Fujii S. Stimulation of migration of human aortic smooth muscle cells by vitronectin: implications for atherosclerosis. Cardiovasc Res 1994;28:1815 – 20. [16] Janat MF, Argraves WS, Liau G. Regulation of vascular smooth muscle cell integrin expression by transforming growth factor-b1 and by platelet-derived growth factor-BB. J Cell Physiol 1992;151:588 – 95. [17] Clowes AW, Reidy MA, Clowes MM. Kinetics of cellular proliferation after arterial injury. Lab Invest 1983;49:327–33. [18] Chomczynski P, Sacchi N. Single step method of RNA isolation by acid Guanidinium Thiocyanate-Phenol-Chloroform extraction. Anal Biochem 1987;162:152–9. [19] Wong J, Rauhoft C, Dilley R, Agrotis A, Jennings GL, Bobik A. Angiotensin converting enzyme inhibition abolishes medial smooth muscle PDGF-AB biosynthesis and attenuates their proliferation in injured carotid arteries. Relationship to neointima formation. Circulation 1997;96:1631–40. [20] Ward M, Agrotis A, Kanellakis P, Dilley R, Jennings G, Bobik A. Inhibition of protein tyrosine kinases attenuate increases in expression of transforming growth factor-beta isoforms and their receptors following arterial injury. Arterioscler Thromb Vasc Biol 1997;17:2461 – 70. [21] Wick M, Burger C, Brusselbach S, Lucibello FC, Muller R. Identification of serum inducible genes: different patterns of gene regulation during G0 “ S and G1 “ S progression. J Cell Sci 1994;107:227 – 39. [22] Qian SW, Kondiah P, Roberts AB, Sporn MB. cDNA cloning by PCR of rat transforming growth factor b-1. Nucl Acids Res 1990;18:3059. [23] Wang J, Kuliszewski M, Yee W, Sedlackova L, Xu J, Tseu I, Post M. Cloning and expression of glucocorticoid-induced genes in fetal rat lung fibroblasts. J Biol Chem 1995;270:2722– 8. [24] He WW, Gustafson ML, Hirobe S, Donahoe P. Developmental expression of four novel serine/threonine kinase receptors homologous to the activin/transforming growth factor-beta type II receptor family. Dev Dyn 1993;196:133–42. [25] Tsuchida K, Mathews LS, Vale WW. Cloning and characterisation of a transmembrane serine kinase that acts as an activin type I receptor. Proc Natl Acad Sci USA 1993;90:11242 – 6.

.

275

[26] Tsuchida K, Lewis KA, Mathews LS, Vale WW. Molecular characterisation of rat transforming growth factor-b type II receptor. Biochem Biophys Res Commun 1993;191:790–5. [27] Shinar DM, Schmidt A, Halperin D, Rodan GA, Weinreb M. Expression of av and b3 integrin subunits in rat osteoclasts in situ. J Bone Min Res 1993;8:403 – 14. [28] Hemmerich P, von Mikecz A, Neumann F, Sozeri O, Wolff-Vorbeck G, Zoebelein R, Krawinkel U. Structural and functional properties of ribosomal protein L7 from humans and rodents. Nucl Acids Res 1993;21:223 – 31. [29] Little PJ, Cragoe EJ, Bobik A. Na-H exchange is a major pathway for Na influx in rat vascular smooth muscle. Am J Physiol 1986;251:C707– 12. [30] Neylon CB, Little PJ, Cragoe EJ, Bobik A. Intracellular pH in human arterial smooth muscle. Regulation by Na + /H + Exchange and a novel 5-(N-ethyl-N-isopropyl)amiloride-sensitive Na + - and HCO3− -dependent mechanism. Circ Res 1990;67:814 – 25. [31] Lastres P, Letamendia A, Zhang H, Rius C, Almendro N, Raab U, Lopez LA, Lang C, Fabra A, Letarte M, Bernabeu C. Endoglin modulates cellular responses to TGF-beta 1. J Cell Biol 1996;133:1109– 21. [32] Sakata M, Kurachi H, Ikegami H, Jikihara H, Morishige K, Miyake A, Terakawa P, Tanizawa O. Autocrine growth mechanism by transforming growth factor (TGF)-b1 and TGF-b1-receptor regulation by epidermal growth factor in a human endometrial cancer cell line IK-90. Int J Cancer 1993;54:862–7. [33] Koyama N, Koshikawa T, Morisaki N, Saito Y, Yoshida S. Bifunctional effects of transforming growth factor-b on migration of cultured smooth muscle cells. Biochem Biophys Res Commun 1990;169:725 – 9. [34] Agrotis A, Saltis J, Bobik A. Transforming growth factor-b1 gene activation and growth of smooth muscle from hypertensive rats. Hypertension 1994;23:593 – 9. [35] Yamashita H, ten Dijke P, Franzen P, Miyazono K, Heldin C-H. Formation of hetero-oligomeric complexes of type I and type II receptors for transforming growth factor-b. J Biol Chem 1994;269:20172– 8. [36] Lin HY, Moustakis A. TGF-b receptors: structure and function. Cell Mol Biol 1994;40:337 – 49. [37] McCaffrey TA, Consigli S, Du B, Falcone DJ, Sanborn TA, Spokojny AM, Bush HL. Decreased type II/type I TGF-b receptor ratio in cells derived from human atherosclerotic lesions. J Clin Invest 1995;96:2667 – 75. [38] O’Brien ER, Bennett KL, Garvin MR, Zderic TW, Hinohara T, Simpson JB, Kimura T, Noboyushi M, Mizgala H, Purchio A, Schwartz SM. big-h3, a transforming growth factor-b-inducible gene, is overexpressed in atherosclerotic and restenotic human vascular lesions. Arterioscler Thromb Vasc Biol 1996;16:576–84. [39] Topol EJ, Califf RM, Weisman HF, Ellis SJ, Tcheng TE, Worley S, Ivanhoe R, George BS, Fintel D, Weston M. Randomised trial of coronary intervention with antibody against platelet IIb/IIIa integrin for reduction of clinical restenosis: results at six months. Lancet 1994;343:881 – 6.