Elsevier

Toxicology Letters

Volume 139, Issue 1, 20 March 2003, Pages 67-75
Toxicology Letters

Short communication
Troglitazone but not rosiglitazone induces G1 cell cycle arrest and apoptosis in human and rat hepatoma cell lines

https://doi.org/10.1016/S0378-4274(02)00468-XGet rights and content

Abstract

Rosiglitazone (RSG), an agonist of peroxisome proliferator-activated receptor γ (PPARγ), induces minor toxicity in humans relative to another PPARγ agonist, troglitazone (TRO). In contrast, recent reports suggest that RSG causes growth arrest and apoptosis of normal and cancerous cells. Therefore, in this study, we investigated the relative toxicities of TRO and RSG on three different hepatoma cell lines, and observed that TRO, but not RSG, was cytotoxic. Additionally, we studied the mechanism by which TRO induced damage to HepG2 hepatoma cells. Our results indicated that TRO increased the levels of p53, p27, and p21, while it reduced the levels of cyclin D1 and phospho-Rb in a time-dependent manner. Increased p27 and p21 levels coincided with reduced activities of cell cycle dependent kinases (cdk) such as cdk2- and cyclin A-protein kinases 24 h after TRO treatment. These results demonstrate that TRO, but not RSG, causes G1 arrest of hepatoma cells, most likely through changing the levels of cell cycle regulators. Furthermore, because RSG did not affect the levels of cell cycle regulators, TRO-mediated growth inhibition appears independent of PPARγ activation.

Introduction

Peroxisome proliferator-activated receptor γ (PPARγ) is a member of the steroid/thyroid nuclear hormone receptor superfamily (Evans, 1988) and plays an important role in cellular physiology and metabolism. Because of its physiological role in glucose and lipid metabolism, many natural and synthetic agonists of PPARγ are used to treat adult onset non-insulin dependent diabetic patients (Day, 1999, Levovitz et al., 2002). A naturally occurring lipid analog, 15-deoxy-Δ12,14-prostaglandin J2 is an endogenous agonist to PPARγ (Forman et al., 1995). In contrast, troglitazone (TRO), rosiglitazone (RSG) and pioglitazone are synthetic thiazolidinedione ligands for PPARγ (Day, 1999). These PPARγ agonists are known to sensitize the target cells to insulin, thus improving the impaired metabolic conditions associated with adult onset diabetes. Although the precise mechanism of action of PPARγ agonists is still not fully understood, the rank order of agonist binding affinities to PPARγ closely matches the order of their anti-diabetic potencies (Wilson et al., 1996, Adams et al., 1997, Levovitz et al., 2002).

Besides sensitizing cells to insulin, some of the PPARγ agonists have been shown to cause growth arrest or apoptosis in cultured cells and in animal models (Hirase et al., 1999, Ohta et al., 2001, Toyoda et al., 2001). During the preclinical testing phase of TRO efficacy, approximately 1.9% of patients receiving TRO developed severe hepatic problems with elevated serum transaminase activities (Watkins and Whitcomb, 1997). In the severe cases, TRO caused fulminant hepatic failures, leading to multiple human deaths. Due to the severity of TRO-induced hepatotoxicity and the availability of its structural derivatives such as RSG and pioglitazone (Day, 1999, Levovitz et al., 2002), TRO was removed from the market in 2000.

Recently, a report has also suggested that RSG may cause hepatotoxicity in humans (Forman et al., 2000), although the incidence of RSG-related hepatotoxicity is considered extremely rare (Levovitz et al., 2002). Additional studies have revealed that RSG also causes cell growth arrest or death in normal and cancerous cells: vascular smooth muscle cells (Gouni-Berthold et al., 2001, Wakino et al., 2000), myeloid leukemia cells (Sugimura et al., 1999), differentiated human macrophages (Chinetti et al., 1998), intestinal epithelial cells (Kitamura et al., 2001), and human papillary thyroid carcinoma cells (Ohta et al., 2001). In addition, pioglitazone affects various cancer cells similarly (Dubey et al., 1993, Sugimura et al., 1999, Goke et al., 2001). Together, these results suggest the possibility that RSG and pioglitazone may also damage the cells of hepatic origin.

Cyclin dependent kinases (CDKs) are serine-threonine protein kinases that regulate cell cycle progression. CDKs are activated by various cyclins and inhibited by natural inhibitors such as p21, p27, and p18 (for review, see Sherr and Roberts, 1999). These CDKs, cyclins, and CDK inhibitory proteins are tightly controlled by complex mechanisms of transcriptional and post-translational modifications. Due to the critical roles CDK and CDK suppressor proteins play in cell cycle progression and arrest, the effect of PPARγ agonists on CDKs and growth arrest has been studied. Results from these studies varied, however, depending on target cell type, particular PPARγ agonist used, exposure time, and the presence of other mitogenic factors. For instance, Motomura et al. (2000) demonstrated that TRO inhibited human pancreatic carcinoma cell growth by selective up-regulation of p27 without elevation of p21 and p18. In contrast, Koga et al. (2001) showed that TRO treatment for 24 h increased the levels of p21, p27, and p18 in human hepatoma cells, although the effect of RSG was not studied. Other studies showed that both TRO and RSG selectively inhibited the expression of the cyclin D1 gene (Kitamura et al., 2001), while both of these compounds prevented the induction of p21 in a PPARγ dependent manner (Wakino et al., 2000). On the other hand, TRO was shown to stimulate growth of osteosarcoma cells (Lucarelli et al., 2002). Based on these different results, the mechanism for TRO- or RSG-mediated change in growth arrest or stimulation is still unclear. Furthermore, the direct effect of RSG on hepatocytes or hepatoma cells had not been adequately investigated compared to that of TRO. Therefore, the time-dependent comparative effects of TRO and RSG on growth rate of HepG2 and Chang liver human hepatoma cells and McA-RH7777 rat hepatoma cells were investigated in this study. In addition, to study the mechanism of toxicity by TRO and RSG, levels of various cell cycle regulators in HepG2 hepatoma cells were determined.

Section snippets

Materials

Propidium iodide, dimethyl sulfoxide (DMSO, tissue culture grade), and other chemicals were purchased from Sigma Chemicals (St. Louis, MO). TRO was kindly provided by the Parke-Davis Company (Ann Arbor, MI). RSG was kindly provided by Dr Joong-Kwon Choi at the Korea Research Institute of Chemical Technologies (Daejon, Korea). All tissue culture media and other agents including fetal bovine serum were procured from InVitrogen (Carlsbad, CA). Specific antibodies to the protein analyzed were

Differential effects of TRO and RSG on apoptosis of cultured hepatoma cells

To evaluate the cytotoxic effects of TRO and RSG on cultured hepatoma cells, we treated HepG2 human hepatoma and McA-RH7777 rat hepatoma cells with varying concentrations (up to 100 μM) of TRO or RSG for different times and then measured the cell viability by the MTT cell proliferation assay. TRO caused cell death of HepG2 and McA-RH7777 hepatoma cells in a time- and TRO concentration-dependent manner (Fig. 1). At 12.5 and 25 μM TRO, less than 10% of HepG2 and McA-RH7777 hepatoma cells died at

Discussion

PPARγ agonists play important roles in fat and glucose metabolism in the liver, muscle, and adipose tissue and differentiation of adipocytes (Forman et al., 1995, Day, 1999, Levovitz et al., 2002). Despite the insulin-sensitizing action of PPARγ agonists, TRO and RSG were also shown to cause growth arrest and apoptosis of various normal and cancerous cells. These cells include: hepatocytes (Toyoda et al., 2001, Toyoda et al., 2002), hepatoma cells (Koga et al., 2001), monocyte-derived

Acknowledgements

We thank Dr Norman Salem, Jr. for his support during this study. We also appreciate Drs Ihn-Kyung Jang and Joong-Kwon Choi for technical help and provision of RSG, respectively. We are also grateful to Drs Young-Ho Kim and Van-Anh Nguyen for critical reading of our manuscript.

References (37)

  • S. Wakino et al.

    Peroxisome proliferator-activated receptor γ ligands inhibit retinoblastoma phosphorylation and G1→S transition in vascular smooth muscle cells

    J. Biol. Chem.

    (2000)
  • M. Adams et al.

    Activators of peroxisome proliferator-activated receptor gamma have depot-specific effects on human preadipocyte differentiation

    J. Clin. Invest.

    (1997)
  • M.A. Bae et al.

    Acetaminophen induces apoptosis of C6 glioma cells by activating the c-Jun N-terminal protein kinase related cell death pathway

    Mol. Pharmacol.

    (2001)
  • C. Day

    Thiazolidinediones: a new class of antidiabetic drugs

    Diabet. Med.

    (1999)
  • R.K. Dubey et al.

    Pioglitazone attenuates hypertension and inhibits growth of renal arteriolar smooth muscle in rats

    Am. J. Physiol.

    (1993)
  • R.M. Evans

    The steroid and thyroid hormone receptor superfamily

    Science

    (1988)
  • L.M. Forman et al.

    Hepatic failure in a patient taking rosiglitazone

    Ann. Intern. Med.

    (2000)
  • R. Goke et al.

    Pioglitazone inhibits growth of carcinoid cells and promotes TRAIL-induced apoptosis by induction of p21waf1/cip1

    Digestion

    (2001)
  • Cited by (50)

    • Riccardin D-26, a synthesized macrocyclic bisbibenzyl compound, inhibits human hepatocellular carcinoma growth through induction of apoptosis in p53-dependent way

      2013, Cancer Letters
      Citation Excerpt :

      For example, activated p53 is able to regulate the transcription of its downstream targets, the cell cycle regulator gene p21Waf1/Cip1 and pro-apoptotic gene bax [28,29]. High expression of p21Waf1/Cip1 leads to the inhibition of cyclin-dependent kinases and consequently to the G1 arrest [30], while overexpression of bax results in mitochondrial cytochrome c release and apoptosis [31,32]. Owing to these potent activities, p53 has thus been considered as an attractive target for cancer therapy.

    • Down-regulation of RhoA is involved in the cytotoxic action of lipophilic statins in HepG2 cells

      2010, Atherosclerosis
      Citation Excerpt :

      Yamamoto et al. [28] have shown that troglitazone and one of its metabolites cause concentration-dependent cell injury characterized by internucleosomal DNA fragmentation and nuclear condensation in HepG2 cells, although such an action is not mimicked by other thiazolidinedione derivatives such as pioglitazone and rosiglitazone. The PPARγ-independent hepatotoxicity of troglitazone has also been reported, in which troglitazone but not rosiglitazone causes G1 arrest in HepG2 cells by increasing the levels of p53, p27, and p21 and reducing the levels of cyclin D1 [29]. Inhibition of RhoA/ROCK pathway is one of pleiotropic effects of statins that show beneficial actions on angiogenesis, atherosclerosis and coronary artery disease.

    • Loss of XIAP sensitizes colon cancer cells to PPARγ independent antitumor effects of troglitazone and 15-PGJ2

      2008, Cancer Letters
      Citation Excerpt :

      More recently, large body of evidence indicates that specific ligands for PPARγ may exert their effect independent of PPARγ receptor [13,24–29]. The fact that the same PPARγ ligand may exert different effect in different subtypes of a particular cell line [30], or different PPARγ ligands exert different effects in the same cell line [31,32] further suggest that mechanisms other than PPARγ receptor may be responsible for the anticancer effect of PPARγ ligands. We were interested in exploring whether combinational targeting XIAP and PPARγ would offer a better therapeutic benefit for colon cancer.

    View all citing articles on Scopus
    View full text