AIM: To examine whether the administration of atorvastatin and rosuvastatin would prevent experimentally-induced hepatic cirrhosis in rats.

METHODS: Liver cirrhosis was induced by injections of thioacetamide (TAA). Rats were treated concurrently with TAA alone or TAA and either atorvastatin (1,10 and 20 mg/kg) or rosuvastatin (1, 2.5, 5, 10 and 20 mg/kg) given daily by nasogastric gavage.

RESULTS: Liver fibrosis and hepatic hydroxyproline content, in the TAA-treated group was significantly higher than those of the controls [11.5 ± 3.2 vs 2.6 ± 0.6 mg/g protein (P = 0.02)]. There were no differences in serum aminotransferase levels in the TAA controls compared to all the groups treated concomitantly by statins. Both statins used in our study did not prevent liver fibrosis or reduce portal hypertension, and had no effect on hepatic oxidative stress. Accordingly, the hepatic level of malondialdehyde was not lower in those groups treated by TAA + statins compared to TAA only. In vitro studies, using the BrdU method have shown that atorvastatin had no effect of hepatic stellate cells proliferation. Nevertheless, statin treatment was not associated with worsening of liver damage, portal hypertension or survival rate.

CONCLUSION: Atorvastatin or rosuvastatin did not inhibit TAA-induced liver cirrhosis or oxidative stress in rats. Whether statins may have therapeutic applications in hepatic fibrosis due to other etiologies deserve further investigation.

Liver cirrhosis is one of the leading causes of morbidity and mortality worldwide. Hepatic stellate cells (HSC) play a major role in the pathogenesis of hepatic fibrosis[1]. Injury to hepatocytes results in generation of lipid peroxides, which may have a direct stimulatory effect on matrix production by activated HSC[2]. It has been suggested that aldehyde-protein adducts, including products of lipid peroxidation, modulate collagen gene expression in human fibroblasts[3,4] and may be a link between tissue injury and hepatic fibrosis[2,5]. Several studies demonstrated that activation of HSC in culture is provoked by generation of free radicals and is blocked by anti-oxidants. This activation may involve the transcription factor c-myb and nuclear factor kappa B (NF-κB)[6,7]. Accordingly, antioxidants have been suggested as therapeutic modalities in experimental models[8-10], and in patients with chronic liver injury[11].

3-Hydroxy-3-methylglutaryl-coenzyme A (HMG-CoA) reductase inhibitors (statins) are used extensively to reduce serum cholesterol in an effort to reduce atherosclerotic cardiovascular morbidity and mortality. In addition to their cholesterol-lowering effect, statins demonstrate other biological effects (pleiotropic effects) that some of them may lead to clinical benefits. Those include anti-inflammatory[12] and antioxidant effects[13], inhibition of PDGF-stimulated proliferation and upregulation of tumor growth factor (TGF)-β signaling in murine mesangial cells and cultured heart cells[14-16].

Over the past decade the potential effect of statins as anti-fibrotic agents has received increasing attention. The rationale for the anti fibrotic efficacy is based on the ability of statin compounds to: (1) decrease the growth of human Ito cells in vitro, independently of their effect on cholesterol synthesis[17]; (2) inhibit the proliferation rate of HSC and reduction of the collagen protein steady state levels by both simvastatin and lovastatin[18]; (3) inhibit steatosis, hepatic fibrosis and carcinogenesis in a rat model of non-alcoholic steatohepatitis (NASH)[19]; and (4) enhance hepatic nitric oxide production and decrease hepatic vascular resistance in patients with cirrhosis and portal hypertension[20]. In addition to inhibiting stellate cell activation, the anti-oxidative activity and the attenuation of inflammation by specific statin derivatives may also contribute to the inhibition of fibrosis. Despite the existing data suggesting that statin derivatives can inhibit fibrosis by various mechanisms, treatment with simvastatin or pravastatin did not decrease fibrosis-induced by bile duct ligation (BDL) or carbon tetrachloride (CCl₄) in rats[21,22]. However, one recent study showed that very early atorvastatin treatment inhibits HSC activation and fibrosis in the BDL model in vivo[23].

To further elucidate the anti-fibrotic activity of statins in the liver, we examined the effects of both atorvastatin and rosuvastatin in a well characterized model of chronic thioacetamide (TAA)-induced administration in which cirrhosis is mainly produced via the formation of reactive oxygen species. TAA undergoes an extensive metabolism to acetamide shortly after administration, and to the hepatotoxic reactive metabolite thioacetamide-S-oxide by the mixed function oxidase system[24-26]. We hypothesized that inhibition of HSC activity in addition to the anti inflammatory and anti oxidative effects induced by statins may prevent the hepatic damage induced by TAA in rats.

Our results indicate that both atorvastatin and rosuvastatin did not diminish neither oxidative stress nor the development of TAA-induced cirrhosis in rats, and also had no effect on the proliferation of cultured HSC.

Our major finding in the present study is that atorvastatin and rosuvastatin do not have a therapeutic value as a potential anti-oxidant or anti-fibrotic agents targeting increased oxidative stress or liver fibrosis induced by TAA in rats.

There are various experimental observations regarding the direct anti-fibrotic activity of the statins by the inhibition of stellate cell proliferation; lovastatin inhibits pancreatic stellate cell activation and alpha-smooth muscle actin expression[31] and both simvastatin and lovastatin interferes with HSC activation in vitro[20,21]. Fluvastatin reduces renal fibroblast proliferation and collagen type III production[32] and suppresses oxidative stress and kidney fibrosis after ureteral obstruction[33]. It is also possible that the anti-fibrotic effects of statins are mediated through mechanisms that stimulate fibroblast apoptosis in vitro and in vivo as it was shown in lung and renal fibroblasts[32,34]. Alternatively, the anti-fibrotic effects of statins may be mediated through their newly recognized anti inflammatory[12] and antioxidant mechanisms[35-37]. These include the inhibition of myeloperoxidase derived and nitric oxide derived oxidants[36], S-nitrosylation and activation of thioredoxin in endothelial cells[37], decreased expression of essential NAD(P)H oxidase subunits and upregulation of catalase expression in vascular smooth muscle cells[35]. Statins also induce the expression of a protein with antioxidant and anti-inflammatory functions, heme oxygenase-1 (HO-1), in vitro and in vivo[33,38]. However, this effect is tissue specific demonstrating significant increase in liver HO-1 with simvastatin and lovastatin but not atorvastatin and rosuvastatin. The involvement of hydroxyl radicals and oxidative stress in TAA induced cirrhosis[39], the anti-oxidative effects of different statins, including atorvastatin, and the antifibrotic effect of atorvastatin in a rat model if BDL as was reported recently by Trebicka et al[23], provided a good rationale for the assumption that statins may reduce or prevent fibrogenesis in this specific animal model.

The explanation for the unexpected failure of atorvastatin and rosuvastatin to inhibit fibrosis in this model of cirrhosis are not clear. Two studies demonstrated anti fibrogenic effects of atorvastatin in a CCl₄ induced fibrosis[40,41]. Gardner et al[41] have shown that atorvastatin exhibited statistically significant, although modest, suppression of CCl₄-induced fibrogenesis after 3 wk of treatment. This was shown only using a novel technique for measuring hepatic collagen synthesis in vivo through metabolic labeling with heavy water (²H₂O). Histopathology of the same tissues revealed no significant differences in fibrosis scores among groups that received cotreatment with atorvastatin, emphasizing the importance of the fibrosis measuring method.

Nevertheless the evidence that fibrosis was not inhibited by atorvastatin and with a more potent cholesterol metabolism pathway inhibitor, rosuvastatin, suggests that this effect is not selective and occurs independently of their HMG-CoA inhibition. Since we used much higher doses than those used clinically (up to 20 mg/kg) it does not seem that we are dealing with under dose treatment. Appropriate timing for the administration of the statins may also be critical. Later therapy in the BDL model, for example, lacked significant effects on fibrosis with no change in hepatic inflammation[23]. The statin treatment in our study began in parallel to the administration of TAA. Although, it is possible that pretreatment by statins would be more effective, the fact that there was no improvement, neither in MDA levels nor in fibrosis after three months, do not support this hypothesis. Another possibility to explain these negative results might be the selection of the wrong statins. Indeed, the two statins that inhibited HSC proliferation in vitro were lovastatin and simvastatin[18] and atorvastatin in vivo. In addition pitavastatin was the one that inhibited hepatic fibrosis in a choline-deficient L-amino acid defined diet liver fibrosis[19]. However, despite these effects, even the latter mechanisms were not efficient to prevent or inhibit liver fibrosis, as it was demonstrated with simvastatin or pravastatin in two different animal models of cirrhosis (bile duct ligation and CCl₄), in rats[21,22]. Moreover, antioxidant therapy in human clinical trials also lack or have minimal effect in chronic liver disease[42,43]. Finally, statin-induced protein kinase C (PKC) activation in activated HSCs may interrupt statin-induced HSC apoptosis, thereby reducing its antifibrotic efficacy. Indeed Yang et al[44] recently demonstrated that simultaneous treatment with pravastatin and PKC inhibitor may synergistically enhance antifibrotic efficacy in hepatic fibrosis induced by intraperitoneal injection of carbon tetrachloride or thioacetamide in mice. Finally, we also have to consider the possibility that the negative results may be due to the small numbers (type 2 error).

To further explore the effect of statins on liver fibrosis we examined the effect of several atorvastain concentrations on the proliferation of primary HSC. We observed that atorvastatin had no inhibitory effect on HSC proliferation either in the presence or in the absence of PDGF (Figure 1). This finding is in discordance with several previous studies that demonstrated decreased HSC activation by statins[17,18]. Moreover, using western blot analysis we found that atorvastatin (5 × 10^-8-10^-9 mol/L) had no effect on the expression of α smooth muscle actin further supporting the lack of effect shown in the proliferation assay.

It is of interest that treatment with atorvastatin or rosuvastatin in all doses did not reduce spleen weights, a parameter of portal hypertension. This is in contrast to the recently described therapeutic effects of simvastatin in patients with cirrhosis and portal hypertension. In patients with cirrhosis and portal hypertension simvastatin enhanced hepatic nitric oxide production and decreased hepatic resistance[20]. Similarly, in cirrhotic rats, induced by BDL, atorvastatin has been shown to lower intrahepatic resistance and to decrease portal hypertension[45].

Finally, oral atorvastatin and rosuvastatin were both well tolerated with no side effects, toxicity or mortality. Furthermore, transaminase levels, hepatic MDA and hydroxyproline and liver histopathology score did not aggravate. These data suggest that the use of atorvastatin and rosuvastatin is not associated with an increased risk of hepatoxicity in damaged liver.

In summary, our results show that atorvastatin and rosuvastatin have no effect on TAA-induced liver cirrhosis. Obviously further studies are required to evaluate whether statins may have therapeutic applications, in the development of hepatic fibrosis induced by other etiologies.