Liver-Specific Peroxisome Proliferator–Activated Receptor α Target Gene Regulation by the Angiotensin Type 1 Receptor Blocker Telmisartan Free

HepG2, AML12, and COS7 cells were cultured following the manufacturer's guidelines. Cells were serum deprived for 16 h before stimulation with the vehicle (DMSO) or different effectors.

Real-time PCR was performed as previously described using an ABI 7000 and Stratagene 3000 MXP PCR cycler with either the SybrGreen or FAM-TAMRA detection system (3). Primers and probes are shown in the online appendix (available at http://dx.doi.org/10.2337/db07-0839).

Transient transfection and luciferase assays were performed as previously described (3). Cos7 cells were transfected with the use of Lipofectamine 2000 (Invitrogen), with pGal4-hPPAR (human PPARα) αDEF (hPPARα LBD fused to Gal4 DBD [DNA binding domain]) and pGal5-TK-pGL3, and 10 ng pRL-CMV (Renilla-Cytomegalovirus), a renilla luciferase control reporter vector. After 4 h, transfection medium was replaced by 10% fetal bovine serum DMEM plus the indicated ARBs, Wy14.643, fenofibric acid, or vehicle (DMSO), and luciferase activity was measured after 24 h.

Electrophoresis and blotting following protein isolation from murine liver tissue was performed as previously described (3). Blots were incubated with an anti-PPARα (Sigma) or anti-ACSL1 (kindly provided by Rosalind A. Coleman, Chapel Hill, NC) antibody.

The small interfering RNA (siRNA) targeting human PPARα was purchased from Dharmacon, and two sequences (d-003434–01 and d-003434–02) were used simultaneously according to the manufacturer's instructions. The siRNA negative control from Dharmacon (d-001810–01) was used to test nonspecific effects on gene expression. Overnight-starved HepG2 cells were transfected using HiPerfect (Qiagen), according to the manufacturer's instructions, in 24-well plates containing 10⁵ cells/well with 5 nmol/l siRNA/well (each sequence 2.5 nmol/l). Thirty minutes after the start of transfection, cells were treated for 48 h with telmisartan (50 μmol/l; 0.5% serum) before RNA analysis.

Male C57BL/6J mice were treated as previously described (3). Mice aged 4–5 weeks, were purchased from Harlan Winkelmann (Borchen, Germany). All mice were housed in a temperature-controlled (25°C) facility with a 12-h light/dark cycle. Mice were fed with a high-fat diet (60% kcal from fat) for 16 weeks, followed by randomization to either a vehicle-treated (n = 6) (0.5% Tween 80/H₂O), a telmisartan-treated (n = 6) (3 mg · kg⁻¹ · day⁻¹), or a pioglitazone-treated (n = 6) (10 mg · kg⁻¹ · day⁻¹) group. Telmisartan was provided by Boehringer Ingelheim, and piogliatzone was extracted from tablets. Animals were treated by oral gavage for 10 weeks. Before and after treatment, blood samples were collected from overnight-fasted animals by retroorbital venous puncture under isoflurane anesthesia for analysis of serum triglycerides (enzymatic-colorimetric test; Cypress Diagnostics). After the experiment, animals were killed and organs were dissected. All animal procedures were in accordance with institutional guidelines and were approved.

Triglyceride content in liver was measured as described previously (19). Briefly, tissues were homogenized in liquid nitrogen and treated with ice-cold chloroform/methanol/water mixture (2:1:0.8) for 2 min. After centrifugation, the aqueous layer was removed and the chloroform layer was decanted. The mixture was incubated at 70°C for chloroform clearance, and the residues were dissolved in isopropanol and assessed for the triglyceride content using an enzymatic-colorimetric test (Cypress Diagnostics), according to the manufacturer's instructions. For immunohistochemical studies, organs were fixed in 4% formalin, embedded in paraffin, and stained with hematoxylin/eosin.

ANOVA, followed by multiple comparison testing or t test, was performed for statistical analysis as appropriate. Statistical significance was designated at P < 0.05. Values are expressed as means ± SE or as indicated.

To evaluate whether telmisartan regulates “classical” PPARα target genes, mRNA expression of CPT1A, the rate-limiting enzyme of fatty acid oxidation, in HepG2 cells was examined. Telmisartan induced CPT1A mRNA expression after 48 h in a dose-dependent manner starting at 10 μmol/l (1.5 ± 0.2–fold vs. vehicle-treated cells, P < 0.05) and reaching a maximum at 50 μmol/l telmisartan (2.84 ± 0.63–fold vs. vehicle-treated cells, P < 0.001 vs. control) (Fig. 1A and B). Treatment of HepG2 cells with a classical PPARα agonist Wy-14643 for 24 h resulted in a 6.6 ± 1.6–fold induction of CPT1A mRNA (100 μmol/l; P < 0.05 vs. vehicle-treated cells; data not shown).

To prove that telmisartan mediates its actions via PPARα activation, HepG2 cells were transfected with PPARα-specific siRNA. After siRNA treatment of HepG2 cells, PPARα mRNA expression was significantly reduced (Fig. 1C, small graph). This siRNA knockdown of PPARα resulted in a significant reduction of telmisartan-induced CPT1A mRNA expression compared with control siRNA, indicating a PPARα-dependent mechanism of action (Fig. 1C).

To determine whether PPARα target gene regulation by telmisartan was gene or species specific, we next studied the expression of ACSL1, the key player in lipid biosynthesis and fatty acid degradation, in the murine hepatic cell line AML12. ACSL1 mRNA was markedly induced by telmisartan (Fig. 1D). In contrast to CPT1A induction in HepG2 cells, maximum ACSL1 mRNA upregulation in AML12 cells was already achieved at 10 μmol/l 2.4 ± 0.1–fold after 48 h incubation with telmisartan 10 μmol/l (P < 0.001 vs. vehicle) (Fig. 1D). Wy-14643 (10 μmol/l) resulted in a 1.46 ± 0.12–fold (P < 0.01 vs. vehicle), and fenofibrate (100 μmol/l) in a 1.55 ± 0.38–fold, induction of ACSL1 mRNA (P < 0.05 vs. vehicle) (Fig. 1D). The present data demonstrate that telmisartan induces the PPARα target genes CPT1A in HepG2 cells and ACSL1 in AML12 cells.

In order to prove whether the induction of hepatic PPARα target genes by telmisartan resulted from a direct activation of PPARα, we examined its ability to directly activate the PPARα LBD by using a chimeric Gal4-DBD-hPPARα-LBD fusion protein on a Gal4-dependent luciferase reporter. Telmisartan induced the activation of the PPARα LBD in a concentration-dependent manner, reaching a maximum at 50 μmol/l (Fig. 2), with 22.5% of the maximum response induced by the reference PPARα agonist Wy-14643 identifying telmisartan as a partial PPARα agonist. No activation of the PPARα LBD was achieved with the ARBs irbesartan or losartan (Fig. 2). The computed EC50 values for activation are as follows: telmisartan EC50, 21.8 μmol/l; fenofibric acid EC50, 18.2 μmol/l; and Wy-14643 EC50, 6.4 μmol/l. Here, we identify telmisartan as a partial PPARα agonist inducing activation in the micromolar range.

It has been shown previously that after a single oral administration of telmisartan at a dose of 1 mg/kg in rats, telmisartan prominently concentrates in the liver (tissue distribution of ¹⁴C-telmisartan 4 h after application: liver, 10,673.06 ± 1,274.93 ng eq/ml; plasma, 218.85 ± 6.08 ng eq/ml; and skeletal muscle, 17.55 ± 1.18 ng eq/ml), implicating that the high concentrations required for PPARα activation and target gene regulation observed in vitro might be achieved in vivo (18). To demonstrate that hepatic PPARα activation by telmisartan occurs in vivo and translates into metabolic changes, high-fat diet–fed obese mice were treated with telmisartan 3 mg · kg⁻¹ · day⁻¹ or vehicle for 10 weeks and hepatic/muscular PPARα target gene expression, hepatic triglyceride accumulation, and serum triglyceride levels were determined.

ACSL1 protein expression in liver tissue was markedly increased in telmisartan-treated mice compared with the vehicle-treated animals (Fig. 3A). Accordingly, relative ACSL1 mRNA expression increased 2.5 ± 0.3–fold in livers of telmisartan-treated animals (P < 0.01 vs. vehicle) (Fig. 3B). In consonance, telmisartan treatment led to a 3.2 ± 0.4–fold increase of hepatic CPT1A mRNA compared with vehicle-treated mice (P < 0.01 vs. vehicle) (Fig. 3B). Interestingly, telmisartan had no effect on PPARα target gene expression in skeletal muscle (Fig. 3C). Together, these results indicate that telmisartan markedly induces hepatic PPARα target genes involved in fatty acid catabolism in obese mice.

Previously, telmisartan has been characterized as a partial PPARγ agonist (4). To evaluate the role of hepatic PPARγ activation in the actions of telmisartan, mRNA expression of the PPARγ target gene CD36 and PPARγ2 was analyzed in liver tissue from telmisartan-treated mice (Fig. 3D). Hepatic CD36 and PPARγ2 mRNA levels were not significantly regulated by telmisartan treatment, suggesting the absence of major hepatic PPARγ activation by telmisartan (Fig. 3D). Furthermore, we studied the effect of the full PPARγ agonist pioglitazone in livers from high-fat diet–fed mice. Mice treated with pioglitazone (10 mg · kg⁻¹ · day⁻¹) for 10 weeks exhibited no regulation of hepatic CPT1A mRNA expression, whereas the PPARγ target gene CD36 was significantly induced by 2.8 ± 0.3–fold compared with vehicle-treated high-fat diet–fed mice (P < 0.01 vs. vehicle-treated high-fat diet–fed mice) (Fig. 3E).

Increased expression of hepatic CPT1A and ACSL1, and subsequent higher rates of hepatic fatty acid oxidation in telmisartan-treated animals, should result in decreased accumulation of triglycerides in liver and serum. Telmisartan treatment prominently reduced liver triglyceride content in obese mice (29.7 ± 11.7μmol/g wet wt) when compared with vehicle-treated mice (79.5 ± 17.4 μmol/g wet wt, P < 0.005) (Fig. 4A and B). In accordance, serum triglycerides in high-fat diet–fed mice declined from 122.3 ± 28.4 mg/dl before treatment to 53.9 ± 4.5 mg/dl after telmisartan treatment (P < 0.005) (Fig. 4C), implicating that hepatic PPARα gene induction by telmisartan translates into a lowering of systemic and local triglyceride level.

In humans, aspartate aminotransferase (AST)-to-alanine aminotrasferase (ALT) ratios are indicative for the extent and etiology of liver damage (20). Whereas mild liver disease like nonalcoholic fatty liver disease and uncomplicated virus hepatitis are affiliated with ratios <1, severe liver disease like chronic hepatitis, liver cirrhosis, and alcoholic fatty liver disease result in increased ratios >1 (20). First, we compared AST-to-ALT ratios in mice on a low-fat diet (10% kcal from fat) with high-fat diet–fed animals. The mean AST-to-ALT ratio in low-fat diet–fed control mice was 1.1 ± 0.1, which was significantly decreased by high-fat diet to 0.5 ± 0.1 (P < 0.05). In line with the reduction of liver steatosis, telmisartan treatment restored impaired the AST-to-ALT ratio to normal levels (1.1 ± 0.3), indicating that reduction in hepatic triglycerides by telmisartan also improved high-fat diet–mediated deterioration of liver function.

To explore additional mechanisms of PPARα activation by telmisartan, which might be additive to LBD-dependent activation, we investigated the regulation of PPARα by telmisartan. Liver PPARα protein expression markedly increased with telmisartan treatment (Fig. 5A). In accordance, hepatic PPARα mRNA was upregulated 1.9 ± 0.2–fold in the telmisartan-treated animals compared with vehicle-treated mice (P < 0.01) (Fig. 5B). Moreover, PPARα mRNA induction by telmisartan was observed in HepG2 cells, with a maximum induction of 3.4 ± 0.4–fold (P < 0.05 vs. vehicle) after 48 h at high concentrations of telmisartan (50 μmol/l) (Fig. 5C). The ARB eprosartan had no effect on PPARα mRNA expression (Fig. 5C). These data show that, in addition to LBD activation, telmisartan is capable of inducing PPARα expression, which might contribute to the observed target gene regulation.

The present data demonstrate that the ARB telmisartan induces PPARα target genes in human and murine hepatic cells and acts as a partial PPARα agonist in the higher micromolar range in vitro. As a result of its particular pharmacokinetic profile with high concentrations in liver, telmisartan potently induces hepatic PPARα target genes involved in fatty acid catabolism in obese mice, which was associated with a significant reduction of systemic and local triglyceride level.

Derosa et al. (5) could demonstrate that telmisartan, when compared with eprosartan, significantly reduced serum triglycerides in hypertensive type 2 diabetic patients by 24.8% compared with baseline, an effect that may not be fully explained by its AT1-blocking/PPARγ-modulating actions. Additional studies (10,11) in diabetic and nondiabetic hypertensive patients have confirmed significant lower plasma triglyceride levels after telmisartan treatment. In the present study, we identify telmisartan as a weak PPARα agonist in vitro. Furthermore, telmisartan treatment in vivo induced PPARα-regulated genes involved in hepatic fatty acid oxidation at relatively low doses, suggesting that hepatic PPARα activation by telmisartan might be clinically relevant. It is well known that treatment of hypertriglyceridemic patients with fibrates results in potent lowering of triglyceride levels, which is, at least in part, mediated via hepatic PPARα activation, subsequent induction of fatty acid oxidation in the liver, and decreased VLDL particle production and plasma triglycerides (21). In consonance with this, telmisartan also significantly reduced circulating triglyceride levels in our animal model, which is consistent with recent data observed in the leptin receptor–deficient Zucker rat treated with telmisartan (22). Together, these results underscore that PPARα activation in the liver may contribute to telmisartans action on dyslipidemia in patients.

Telmisartan has been recently characterized as a selective PPARγ modulator (3,4). Hepatic activation of PPARγ contributes to the actions of PPARγ agonists on lipid and glucose metabolism in mice (23–25). To characterize the relevance of hepatic PPARγ activation for telmisartan's effects, CD36 and PPARγ2 expression were investigated and compared with the full PPARγ agonist pioglitazone. Telmisartan did not regulate CD36 or PPARγ2 in livers from high-fat diet–fed mice, excluding a major role of hepatic PPARγ in telmisartan's action. In contrast, pioglitazone failed to induce CPT1A but stimulated CD36 mRNA expression in liver. These data are consistent with previous findings of distinct gene expression profiles in adipocytes treated with telmisartan or piogliatzone (3). In liver, telmisartan may mainly act as a partial PPARα agonist, whereas pioglitazone also activates PPARγ pathways.

Telmisartan reduced hepatic triglyceride content in high-fat diet–fed mice. Nonalcoholic steatohepatitis (NASH) frequently develops during obesity as a result of insulin resistance and subsequent hepatic triglyceride overload (26). NASH is considered a hepatic component of the metabolic syndrome, leading to liver fibrosis and cirrhosis (26). Currently, no drug therapy has been established for the treatment of NASH. Recently, a number of small clinical trials (27–29) have demonstrated that the PPARγ agonists rosiglitazone and pioglitazone improve liver histology and aminotransferase levels in patients with NASH. PPARα activation by fibrates has been recently reported to reduce the development of NASH in different animal models (30,31). Therefore, combination of hepatic PPARα activation and systemic PPARγ modulation by telmisartan, together with the observed reduction of liver triglyceride content, may provide a new therapeutic option for the future treatment of NASH. Telmisartan's action on hepatic pathology has already been described by Sugimoto et al. (32). In rats fed a high-fat, high-carbohydrate diet, telmisartan, but not valsartan, significantly reduced hepatic triglyceride (32). Along this line, Fujita et al. (33) could show that telmisartan application to a rat model of NASH improved numerous pathological features of the disease, including liver steatosis, liver inflammation, and liver fibrosis, underlining the potential role of telmisartan for NASH therapy. The hepatic actions of telmisartan will gain clinical importance in light of previous data demonstrating an increased prevalence of fatty liver disease in hypertensive patients (34). The antihypertensive actions combined with hepatoprotective actions of telmisartan could be of added clinical value in these patients.

The pharmacokinetic profile of telmisartan seems to be highly important for its PPARα-activating properties. In preclinical studies with telmisartan, Shimasaki et al. (18) have shown that telmisartan prominently concentrates in the liver with ∼40 times higher levels compared with plasma and skeletal muscle, which opens the possibility that the high concentrations required for PPARα activation and target gene regulation observed in vitro might be achieved in vivo in a tissue-specific manner. The high liver content of telmisartan is likely caused by binding of the compound to the glutathione-S-transferase type 1–1 (ligandin) protein, which is present at high concentrations in the liver (35). Hepatic storage may be further supported by the lipophilic characteristic of telmisartan. Compared with the liver, telmisartan concentrations in skeletal muscle are extremely low 4 h after administration, which makes it unlikely that concentrations required for muscular PPARα activation are reached (18). In accordance with the pharmacokinetic data, telmisartan did not affect PPARα target gene expression in skeletal muscle of obese mice, strongly supporting a tissue-specificity for telmisartan-induced PPARα activity.

The liver specificity of telmisartan-mediated PPARα activation does not only provide a mechanism for its beneficial effects on triglyceride levels but also plays a major role for potential PPARα-mediated side effects. One of the most common toxic side effect during fibrate therapy are myopathies associated with myalgias, in particular in combination with statin therapy (36). Since muscular PPARα target genes are not activated by telmisartan, and telmisartan concentrations in skeletal muscle are minor, the occurrence of PPARα-mediated muscular side effects under telmisartan therapy are very unlikely. Moreover, the unique pharmacokinetic profile of telmisartan allows beneficial liver-specific PPARα activation in the absence of common PPARα-mediated side effects.

In addition to activation of the PPARα LBD, telmisartan-mediated induction of protein and mRNA expression of PPARα could be detected in vitro and in vivo. This data are in accordance with previously published studies, demonstrating a stabilization of the PPARα protein after ligand binding (37). Here, we identify that also PPARα mRNA expression is positively regulated by telmisartan, suggesting an additional transcriptional mechanism of ligand (telmisartan)-mediated receptor regulation. This is in line with previous reports demonstrating an induction of hepatic PPARα mRNA expression by fibrate treatment in rodents and human hepatocytes (38–40). Nevertheless, future studies are required to investigate whether PPARα mRNA regulation by telmisartan depends on telmisartan PPARα LBD interactions or whether this might be mediated via blockade of the AT1 receptor. As yet, it seems that telmisartan positively regulates the PPARα pathway by two different mechanisms: 1) LBD activation and 2) receptor upregulation.

In summary, the present study identifies the ARB/PPARγ modulator telmisartan as a partial PPARα agonist. As a result of its particular pharmacokinetic profile with high concentration in liver, PPARα activation by telmisartan seems to be restricted to the liver. Hepatic PPARα activation by telmisartan may provide an explanation for its antidyslipidemic actions observed in clinical trials and prevents simultaneously potential danger from systemic PPARα-mediated adverse effects. The multimodal mechanism of action of telmisartan, including AT1-receptor blockade/PPARγ modulation and hepatic PPARα activation, characterizes this compound as a therapeutic option for the treatment of patients suffering from multiple cardiometabolic disorders such as hypertension, glucose intolerance, and dyslipidemia.

This study was supported by Bayer Schering Pharma and Boehringer Ingelheim Pharma. M.C. is supported by the Deutsche Forschungsgemeinschaft (DFG-KI 712/3-1). C.B. is supported by the Deutsche Forschungsgemeinschaft (DFG-GK 754 III). R.G. is supported by the Deutsche Forschungsgemeinschaft (DFG-GU-285/7-1). T.U. is supported by the Deutsche Forschungsgemeinschaft (DFG-GK 754-III, DFG-GK 865-II). U.K. is supported by the Deutsche Forschungsgemeinschaft (DFG-GK 754-III, DFG-GK 865-II, DFG-KI 712/3-1).

We thank Rosalind A. Coleman (Department of Nutrition, University of North Carolina, Chapel Hill, NC) for providing the anti mouse ACSL1 antibody. We are deeply in debt to Christiane Sprang for excellent technical assistance.