Left ventricular hypertrophy: a new approach for fibrosis inhibition

Hypertensive heart disease involves a structural remodelling of cardiomyocytes and non-myocytes. In addition, fibroblasts contribute to perivascular fibrosis, which initially surrounds intramural coronary arteries and subsequently moves into the interstitial space [1–3]. This fibrosis consists of an increase of fibrillar collagen types I and III and may be responsible for progressive abnormalities of diastolic ventricular filling and relaxation, and of systolic dysfunction and cardiac rhythm and conduction disturbances, therefore contributing to the risk associated with left ventricular hypertrophy (LVH) [4]. The excess of ventricular collagen may be the result of collagen synthesis, but also of insufficient collagen degradation by interstitial collagenase and, in particular, metalloproteinase 1 (MMP-1), whose activity is regulated by specific tissue inhibitors of metalloproteinases (TIMP) [5]. The initiation of reactive fibrosis is triggered, at least in part, by the local activation of trophic peptides and, in particular, the renin–angiotensin system (angiotensin II, aldosterone and/or catecholamines) resulting in the sequential expression of transforming growth factor (TGF)-β1 and fibronectin, and a relative increase of collagen I [5]. Angiotensin II may also regulate collagen degradation by reducing interstitial matrix MMP-1 activity in adult human cardiac fibroblasts and by enhancing the production of TIMP-1 in endothelial cells. TGF-β1 is considered a primary mediator of the cardiovascular effects of angiotensin II. In the heart, three isoforms of TGF are present, although the level of the type 1 isoform seems to be particularly related to the development of LVH [6]. TGF-β1 is secreted by most cell types and may exert a range of complex actions (mainly related to the type of cells involved), inhibiting the proliferation of many cells and, at the same time, stimulating the growth of mesenchymal cells and extracellular matrix formation [6]. In-vitro experiments have shown that angiotensin II increases cardiac fibroblast-mediated collagen synthesis and mRNA levels of collagen type I, type III, pro alfa1(I) collagen, proalfa1(III) collagen and fibronectin, inhibits matrix metalloproteinase 1 (MMP-1) activity [7] and increases the concentration of TIMP-1 [8]. These effects are mediated by an increased TGF-β1 production and are greatly attenuated by an angiotensin II type 1 receptor antagonist. Angiotensin II stimulates TGF-β1 gene expression and promotes the conversion of latent TGF-β1 to the active form. Angiotensin II has been shown to stimulate TGF-β1 production in neonatal and adult cardiac fibroblasts, although it was recently suggested that a (soluble?) myocyte-generated factor may play a role in the stimulation of collagen formation, in association with angiotensin II [9]. In-vivo studies have shown that angiotensin II increase is related to TGF-β1 expression in infarcted hearts and stimulates TGF-β1 production via binding to AT1 receptors [10]. An increase in TGF-β1 expression is also observed during LVH development in TGR (mRen2)27 (Ren2) rats, a monogenetic model of hypertension with low plasma and renal renin and high expression of renin in cardiac tissue [11]. The role of TGF-β1 in the progression from compensated cardiac hypertrophy to cardiac dysfunction and heart failure is still not completely understood. The expression of TGF-β1 was not increased in non-failing spontaneously hypertensive rats (SHR) [12], while an increased expression of TGF-β1 was observed in non-failing hypertrophy after abdominal aortic coartation [13], suggesting that the importance of TGF-β1 in the development of cardiac hypertrophy and fibrosis may vary according to the different model of experimental hypertension. Moreover, the different effect of TGF-β1 on the heart remains to be completely established as the increased expression of TGF-β1-induced atrial, but not ventricular fibrosis in transgenic mice [14]. The regression of structural remodelling in hypertensive heart disease represents a goal of pharmacological intervention as a reduction in blood pressure in arterial hypertension may be unable to reduce fibroblast growth and collagen turnover. Therapeutic strategies have included the use of angiotensin-converting enzyme (ACE)-inhibitors, angiotensin II type 1 receptor antagonists, antagonists of endothelin A and B receptors and antagonists of aldosterone receptors and, more recently, metalloproteinases activity modulators have been proposed. Treatment with ACE-inhibitors reversed or even prevented myocardial accumulation of fibrous tissue in SHR [5]. The regression in fibrosis with lisinopril was independent of the antihypertensive effect and partially due to enhanced collagen degradation by MMP-1. In humans, a recent trial showed that administration of an ACE-inhibitor, but not of a diuretic, in essential hypertensive patients might reduce the collagen content, measured by intramyocardial biopsies and expressed as collagen volume fraction, with a consequent improvement of diastolic filling [15]. Chronic treatment with losartan (an antagonist of angiotensin II type 1 receptor) resulted in the normalization of collagen synthesis and reversal of left ventricular fibrosis in SHR, confirming a role for angiotensin II in increased collagen synthesis in that hypertension model [16]. A non-hypotensive dose of the angiotenin II type 1 receptor antagonist blunted the expression of left ventricular TGF-β1, reducing collagen production and perivascular fibrosis in the hypertensive Ren2 rats [11]. Recently, the effect of tranilast (N (3,4dimethoxycinnamoyl)-anthranilic acid), a drug inhibiting TGF-β1 mediated collagen synthesis, has been evaluated in animal models of hypertension and atherosclerosis. Tranilast has been used for the prevention and treatment of keloids and hypertrophic scarring and for the treatment of allergic diseases, including bronchial asthma [17,18]. Tranilast has been shown to inhibit collagen synthesis in fibroblasts derived from hypertrophic tissues and keloids [19], to suppress mast cell degranulation and histamine release [20] and to suppress oxygen free radical production from neutrophils. Furthermore, tranilast suppresses vascular chymase (synthesized mainly in mast cells) expression [21] and possibly chymase-dependent angiotensin II production, contributing to the prevention of neointima formation. In-vivo studies using either carotid, peripheral or coronary arteries of small and large animals showed that tranilast reduced the neointimal area on histomorphometry. The oral administration of tranilast in the setting of preliminary clinical studies has given promising results, decreasing the rate of coronary restenosis after angioplasty [22]. The increasing interest in the potential use of this compound in cardiovascular disease treatment is demonstrated by an ongoing clinical study on coronary restenosis involving more than 10 000 patients (PRESTO) [21]. In this issue of the journal, Hocher et al. [23] evaluated the effect of treatment with tranilast for 12 weeks in rats with renovascular hypertension compared to an ACE-inhibitor or no treatment, and demonstrated that tranilast reversed TGF-β1 mediated cardiac fibrosis, independently of blood pressure, while treatment with an ACE-inhibitor reduced blood pressure and left ventricular fibrosis. This work provides evidence that TGF-β1 plays a central role in the process of cardiac fibrosis, particularly when the renin–angiotensin system is activated, confirming results obtained in previous studies. Tranilast mainly inhibits proliferation of non-cardiomyocyte cells, but does not affect the hypertrophic process of myocytes; in fact, a small, not statistically significant, reduction of left ventricular mass was observed in the present study. This functional antagonist was particularly effective in reducing interstitial fibrosis, with minor effects on perivascular fibrosis; this effect was previously observed in transgenic Ren(2) rats [11], another model of experimental hypertension with an elevated activity of the cardiac renin–angiotensin system. The hypothesis that haemodynamic factors may have a prevalent role in inducing perivascular more than interstitial fibrosis in these models of hypertension [24] is confirmed by the observed effects of an ACE-inhibitor, given at a hypotensive dose, on both perivascular and interstitial fibrosis. The association of existing treatment with ACE-inhibitors or angiotensin II type 1 receptor antagonists with anti-fibrotic drugs such as tranilast would represent a promising strategy to successfully prevent or regress cardiac fibrosis; the combination of both drugs, not addressed and evaluated in this study, could block the effect of angiotensin II, produced by ACE and chymase-dependent pathways. An unexpected and rather disappointing finding in the study by Hocher et al. [23] is the lack of the effect of both tranilast and the ACE-inhibitor on media–lumen ratio of intracardiac arteries, evaluated by an histological approach. It has been repeatedly shown in SHR, in experimental models of right ventricular hypertrophy, and in clinical studies, that structural changes in small arteries may be influenced by drugs interfering with the renin–angiotensin system, even at subhypotensive dosages [25–27]. However, vascular structure and geometry may be best analysed by alternative techniques, such as in-vivo perfusion fixation or micromyography [28], and the effect of tranilast on structural changes in small arteries should be assessed in the future using these more reliable approaches. There is some concern about the potential side-effects of tranilast, including other organ and systemic toxicity, due to it limiting the fibrotic tissue reaction in other organs. Other studies have shown that tranilast inhibits collagen sysnthesis in cultured fibroblats derived from scar tissue, but not in those taken from healthy skin [19]. Moreover, in animal studies, as well as in human clinical studies (when the drug was administered for the treatment of dermatological or allergic conditions), no changes in serum markers of kidney and liver function or adverse cardiopulmonary or gastrointestinal side-effects were reported [17,18]. The importance of humoral factors, and in particular of circulating or tissue angiotensin II, in the accumulation of collagen in the heart has gained increasing evidence. TGF-β1 represents a principal mediator of angiotensin II and plays a central role in promoting fibrosis and regulating cardiac tissue structure during the development of cardiac hypertrophy and in the transition to heart failure. The possibility of inhibiting TGF-β1 expression using tranilast offers a new and unique approach to the treatment of cardiovascular diseases, complementary to the administration of ACE-inhibitors, angiotensin II type 1 receptor antagonists and anti-aldosterone diuretics. The safety and tolerability of this drug, already proven in the treatment of dermatological and allergic diseases, stimulates the possibility of testing its effect in future clinical trials.