Rodríguez-Vilarrupla A, Laviña B, García-Calderó H, Russo L, Rosado E, Roglans N, et al. PPARα activation improves endothelial dysfunction and reduces fibrosis and portal pressure in cirrhotic rats. J Hepatol 2012;56:1033–1039 (Reprinted with permission.)
Peroxisome proliferator-activated receptor α (PPARα) is a transcription factor activated by ligands that regulates genes related to vascular tone, oxidative stress, and fibrogenesis, pathways implicated in the development of cirrhosis and portal hypertension. This study aims at evaluating the effects of PPARα activation with fenofibrate on hepatic and systemic hemodynamics, hepatic endothelial dysfunction, and hepatic fibrosis in CCl(4)-cirrhotic rats. Mean arterial pressure (MAP), portal pressure (PP), and portal blood flow (PBF) were measured in cirrhotic rats treated with oral fenofibrate (25mg/kg/day, n=10) or its vehicle (n=12) for 7 days. The liver was then perfused and dose-relaxation curves to acetylcholine (Ach) were performed. We also evaluated Sirius Red staining of liver sections, collagen-I mRNA expression, and smooth muscle actin (α-SMA) protein expression, cyclo-oxygenase-1 (COX-1) protein expression, and cGMP levels in liver homogenates, and TXB(2) production in perfusates. Nitric oxide (NO) bioavailability and eNOS activation were measured in hepatic endothelial cells (HEC) isolated from cirrhotic rat livers. CCl(4) cirrhotic rats treated with fenofibrate had a significantly lower PP (-29%) and higher MAP than those treated with vehicle. These effects were associated with a significant reduction in hepatic fibrosis and improved vasodilatory response to acetylcholine. Moreover, a reduction in COX-1 expression and TXB(2) production in rats receiving fenofibrate and a significant increase in NO bioavailability in HEC with fenofibrate were observed. PPARα activation markedly reduced PP and liver fibrosis and improved hepatic endothelial dysfunction in cirrhotic rats, suggesting it may represent a new therapeutic strategy for portal hypertension in cirrhosis.
Portal pressure (PP) is the simple product of two factors, flow and resistance. Thus PP = Q (portal flow) × R (liver/portal resistance). In portal hypertension, both flow and resistance are known to be increased. Thus, there are two mechanisms: the “forward flow” and the “backward” resistance. That forward flow is increased has been extensively demonstrated in numerous studies showing mesenteric hyperemia and increased portal flow in animal models and humans with portal hypertension (reviewed1, 2). The backward component has both static and dynamic factors. The static factor is the presence of extensive fibrosis that distorts the architecture and increases vascular resistance principally at the sinusoidal level. The dynamic factor is increased intrahepatic vascular tone. Contractile elements in the sinusoids such as activated hepatic stellate cells (HSCs) under the influence of dilators/constrictors such as nitric oxide (NO) and endothelins can dramatically increase liver vascular tone and resistance. Since the structural deformation due to fibrosis is not easily reversible, researchers are generally focused on intrahepatic vascular tone, and thus endothelial dysfunction is a key issue.3
In cirrhosis the imbalance between the hyperresponsiveness to and overproduction of vasoconstrictors (mainly endothelin-1 and cyclooxygenase-derived prostaglandins, such as TXA2) and the hyporesponsiveness to and impaired production of vasodilators (mainly NO) is the main mechanism responsible for the increased vascular tone in the sinusoidal/postsinusoidal area.4, 5 Rodríguez-Villarupla et al.6 in this study have now examined fenofibrate, a synthetic peroxisome proliferator-activated receptor α (PPARα) receptor ligand, to examine the potential beneficial effects on both the static and dynamic resistance factors causing portal hypertension. They demonstrated that this drug improves hepatic hemodynamics and liver fibrosis in a carbon tetrachloride (CCl4) rat model of cirrhosis.6 The PPARα activation improved endothelial dysfunction and NO bioavailability; it also significantly decreased cyclo-oxygenase-1 (COX-1) protein expression and diminished TXB2 content in the fibrotic liver6 (Fig. 1). A strength of this study is the chronic (1 week) administration of fenofibrate which allowed enough time not only for improving NO bioavailability in endothelium and inhibiting COX-1, but also demonstrating a potential effect on liver fibrosis.
The authors suggested that the reduction of portal vein pressure by fenofibrate is mainly due to the improvement of architectural abnormalities of the cirrhotic liver. Did the “forward flow” mechanism play any role in their study? PPAR-α is highly expressed in hepatocytes, cardiomyocytes, enterocytes, and renal proximal tubular cells.7 PPAR-α activation in the liver has beneficial effects on portal hypertension and hyperdynamic circulation. Since oral administration of fenofibrate activates PPARα and increased NO in the whole body,8 the NO bioavailability was improved not only intrahepatically but also systemically, including the gut circulation. In that regard, it is noteworthy that fenofibrate reduced portal blood flow by a mean of 21%, which although not statistically significant (P = 0.21 by the nonparametric Mann-Whitney test), approached the extent of reduction in portal pressure (−29%). It is possible that methodological issues such as the greater accuracy of the portal pressure measurement method compared to transonic flowmetry to measure portal flow, as well as the possibility of a type II statistical error, may account for the lack of statistical significance. Moreover, if that result is reanalyzed using a two-tailed unpaired Student t test which assumes a normal distribution, interestingly the 21% difference between the two groups becomes highly statistically significant (P = 0.0014). Thus, we believe that it is premature to conclude that PPARα reduces portal pressure by only reducing fibrosis/resistance and that the forward flow component is unaffected.
Limitations of this study included studying just two groups, both with cirrhosis, but no noncirrhotic controls. Thus, when the authors demonstrated that fenofibrate increased mean arterial pressure (MAP), they were unable to conclude whether MAP was totally reversed back to normal or just partially normalized. Also, PPARα has pleiotropic effects including on lipid metabolism, antiinflammation,9 antioxidative stress,10 and increasing NO. The authors did not evaluate oxidative and inflammatory mechanisms, both of which can contribute to fibrogenesis; thus, the initial pathway of the antifibrotic effects of PPARα activation remains unclear.
Another limitation is that the authors did not evaluate the effects of fenofibrate on HSCs, which is the key cell involved in fibrogenesis.11 The liver is rich in many nonparenchymal cell types (40% of the total number of liver cells,12 such as HSCs, Kupffer cells, sinusoidal endothelial cells, and bile duct epithelial cells). Whether the effects of fenofibrate on the improvement of fibrosis are mediated by only endothelial cells or via other cell types, such as HSCs, remains unanswered. Previous studies have demonstrated that the antifibrotic effect of a different PPAR subtype, PPARγ, is mediated via HSCs.11 The consensus view is that PPARα is not expressed in quiescent or activated HSCs and has not been shown to affect HSC biology in vitro or hepatic fibrosis.13 However, not all studies agree with this notion: Ip et al.14 demonstrated that a PPARα agonist reduced the number of activated HSCs. Fenofibrate also has antioxidative and antiinflammatory effects which may indirectly affect HSC behavior.
Overall, numerous experimental animal studies in portal hypertension have yet to be translated into clinical studies. Thus, to date several studies demonstrated that some agents decrease portal pressure via the reduction of hepatic vascular tone. Agents targeted to intrahepatic endothelial dysfunction, such as an endothelin-A receptor antagonist, decrease portal hypertension.15 Removing increased O(2)(−) levels by transferring superoxide dismutase gene improves NO bioavailability and, therefore, reduces portal pressure.16 Besides NO, carbon monoxide (CO) also plays a role in the maintenance of intrahepatic vascular tone. Heme-oxygenase (HO) and CO are decreased in the cirrhotic liver, and the HO-inhibitor zinc protoporphyrin-IX increases intrahepatic vascular tone in a dose-dependent manner.17 However, all of these studies are in animal models, whereas in clinical practice nonselective β-blockers have been the mainstay of pharmacologic therapy. The adverse effects of β-blockers include dizziness, fatigue, cardiac insufficiency, and bronchospasm.. Moreover, decreased survival in patients with ascites treated with β-blockers has been reported.18 A PPARα activator, such as fenofibrate, has been extensively used in the treatment of dyslipidemia, so it is considered safe and ready to use in other diseases, especially for those patients with nonalcoholic fatty liver disease, an increasing cause of cirrhosis worldwide. Therefore, we agree wholeheartedly with the authors that fenofibrate and other PPARα activators deserve further study in clinical trials in cirrhosis patients.