Peroxisome proliferator-activated receptor gamma ligand-mediated apoptosis of hepatocellular carcinoma cells depends upon modulation of PI3Kinase pathway independent of Akt

Abstract

Background: Ligands of Peroxisome proliferator-activated receptor gamma (PPARγ) can inhibit growth and promote apoptosis in various cancer cells, and thus have the potential to be utilized as anticancer drugs. This potential however, has been seriously challenged by observations that they can lead to tumor promotion in some cancer models, possibly due to activation of different signaling mechanisms in various tumor environments. Elucidation of the specific signaling events that modulate PPARγ ligand-mediated events is thus critical to increase their efficacy. The studies described here were designed to elucidate the signaling pathway(s) that modulate the apoptotic potential of Troglitazone (TRG), an artificial PPARγ ligand in hepatocellular carcinoma (HCC) cells.

Results: Our results indicate that the apoptotic potential of TRG was regulated by the presence or absence of serum in the media. When added in serum-containing media, TRG inhibited proliferation and cyclin D1 expression, but was unable to induce any apoptosis. However, TRG's apoptotic potential was induced significantly when added in serum deficient media, as indicated by increased PARP and Caspase-3 cleavage and results from apoptosis assay. Furthermore, TRG-induced apoptosis in serum deficient media was associated with a dramatic reduction in PI3Kinase downstream target AktSer473 and FoxO1Thr24/FoxO3aThr32 phosphorylation. On the contrary, there was an increase of PI3K-induced AktSer473 and FoxO1Thr24/FoxO3aThr32 phosphorylation involving Pak, when TRG was added in serum-containing media. Pharmacological inhibition of PI3Kinase pathway with LY294002 inhibited Aktser473 phosphorylation and sensitized cells towards apoptosis in the presence of serum, indicating the involvement of PI3K in apoptosis resistance. Interestingly, pharmacological inhibition or siRNA-mediated knockdown of Akt or inhibition of Pak was unable to sensitize cells towards TRG-induced apoptosis in the presence of serum. Similarly, TRG was unable to induce apoptosis in the Akt1-KO, Akt1&2-KO MEFs in serum-containing media.

Conclusion: These studies indicate that TRG-induced apoptosis is modulated by PI3K pathway in a novel Akt-independent manner, which might contribute to its tumor promoting effects. Since PI3K activation is linked with various cancers, combination therapy utilizing TRG and PI3K inhibitors has the potential to not only increase the efficacy of TRG as a chemotherapeutic agent but also reduce its off target effects.

Figure 1

Effect of TRG on HCC cell proliferation. (A) Subconfluent Huh-7 cells were plated on 6 well plates in regular growth medium. Next day, they were treated with either DMSO (-) or 25 μM TRG (+) and harvested at the indicated time intervals. The cell numbers were determined and represented as % control considering the DMSO-treated sample of 24 hours as 100%. Cells were plated in triplicate for each time point and each experiment was repeated at least twice. (B) Cells were treated as in A for the indicated time periods, following which they were harvested and total protein was extracted. Western Blot analysis of the cell extracts was then performed with antibodies against Cyclin D1, PCNA and GAPDH (as control). (C) The cells were treated as in B and cell extracts analyzed by Western Blots with the indicated antibodies. (D) Huh-7 cells were treated with either DMSO or increasing concentration of TRG for 48 hours, followed by Western Blot analysis with the indicated antibodies.

Figure 2

Role of PI3Kinase pathway on TRG-induced cell growth arrest. (A) Subconfluent Huh-7 cells were treated as in 1B or (B) as in 1D, followed by Western Blot analysis of the cell extracts with antibodies against pAkt^Ser473, Akt and GAPDH. (C) Huh-7 cells were treated in the absence (-) or presence (+) of 25 μM TRG for 24 hours following a 1 hour pretreatment with either none (lanes 1 & 2), or 1 μM Wortmannin (lanes 3 & 4) or 5 μM LY294002 (lanes 5 & 6). Western Blot analysis was then performed with the antibodies indicated. (D) Subconfluent Huh-7 cells were treated with DMSO, 25 μM TRG, DMSO + 1 μM Wortmannin, or TRG + 1 μM Wortmannin for the indicated time intervals, following a 1 hour pretreatment with Wortmannin. Cell proliferation assay was performed as described in 1A.

Figure 3

Effect of TRG on HCC cell apoptosis in the presence or absence of serum. (A) Subconfluent Huh-7 cells were treated with 25 μM TRG in serum-containing media for the indicated time intervals. Western Blot analysis was performed with antibodies against PARP, Caspase-3 and GAPDH (as control). (B) Huh-7 cells were treated with increasing concentrations of TRG in serum-containing media for 9 hours and subjected to Western Blot analysis as in A. (C) Huh-7 cells were treated with 25 μM TRG in serum deficient media for the indicated time intervals. At the end of incubation cells were harvested and apoptosis assays were performed using cell death detection ELISA^PLUS kit. The data in each set represents the mean ± S.D. of 4 independent experiments. (D) Western Blot analysis of cell extracts treated with TRG in serum deficient media for the indicated periods of time and with antibodies against PARP, Caspase-3, cleaved Caspase-3 (detects only the cleaved form) and GAPDH. (E) &(F) Huh-7 cells treated with increasing concentrations of TRG in serum-deficient media were subjected to apoptosis assays using cell death detection ELISA^PLUS kit (E) or Western Blot analysis (F).

Figure 4

Effect of TRG on PI3K/Akt Pathway in the presence or absence of serum. (A) Huh-7 cells were treated with 25 μM TRG in the absence (-) or presence (+) of serum for the indicated periods of time. Equal amounts of cell extracts were analyzed by Western Blot analysis utilizing antibodies against pAkt^Ser473, Akt or GAPDH. (B) Western Blot analysis of the cell extracts treated with TRG in serum-deficient media following normalization of total AKT levels. (C) Cells treated with increasing concentrations of TRG for 6 hours in serum deficient media were subjected to Western Blot analysis as in A.

Figure 5

Effect of TRG on FoxO1/FoxO3a phosphorylation in the presence or absence of serum. Huh-7 cells were treated with 25 μM TRG for the indicated time intervals in serum deficient (A) or serum-containing (B) media. Western Blot analyses were performed with antibodies against phospho-FoxO1^Thr24/FoxO3a^Thr32, FoxO1, FoxO3a and GAPDH.

Figure 6

Effect of PI3K pathway inhibition on TRG-induced apoptosis resistance in serum-containing media. (A) Huh-7 cells were treated in the absence (-) or presence (+) of 25 μM TRG in serum-containing media for 24 hours following a 1 hour pretreatment with none (lanes 1 & 2), 50 μM LY294002 (lanes 3 & 4), or 20 μM Akt inhibitor VIII (lanes 5 & 6). Western Blot analyses were performed with the antibodies indicated. (B) Cells were treated with TRG as in A following a 1 hour pretreatment with none (lane 2), 50 μM LY294002 (lanes 3), or 20 μM LY303511 (lane 4) and analyzed by Western Blot. (C) Huh-7 cells were treated with TRG as in A, following a 1 hour pretreatment with none (lanes 1 & 2), 10 μM Pak inhibitor (lanes 3 & 4), or 100 nM Rapamycin (lanes 5 & 6). Western Blot analyses with the indicated antibodies were performed next. (D) Huh-7 cells were transfected with either a control-siRNA (lanes 1 & 2), or PPARγ-siRNA (lanes 3 & 4) for 72 hours, followed by TRG treatment in serum-containing media for an additional 24 hours. Western Blot analysis was then performed with the indicated antibodies. TRG-treated Huh-7 cell extract in serum-deficient media was used as positive control for PARP cleavage.

Figure 7

Effect of Akt inhibition on TRG-induced apoptosis resistance in serum-containing media. (A) Subconfluent Huh-7 cells were transfected with either control-siRNA (lanes 1 & 2), AKT-siRNA (lanes 3 & 4), or Akt-3m-siRNA (lanes 5 & 6) for 72 hours followed by treatment with 25 μM TRG for 24 hours in serum-containing media. Western Blot analysis was then performed with the antibodies indicated. (B) MEFs from Wild type (lanes 1 & 2), Akt1 KO (lanes 3 & 4) or Akt1/2 KO (lanes 5 & 6) mice were treated with 25 μM TRG in serum-containing media for 24 hours followed by Western Blot analysis. TRG-treated Huh-7 cell extracts in serum-deficient media were used as positive controls for PARP and Caspase-3 cleavage in A & B and WT-MEF extract as positive control for Akt3 in A.

Figure 8

Effect of PI3K and Akt inhibition on TRG-induced apoptosis resistance in Hep3B cells. (A) Subconfluent Hep3B cells were treated with 25 μM TRG in the presence (+) or absence (-) of serum for the indicated periods of time. Western Blot analyses were performed with the antibodies indicated. (B) Western analysis of Hep3B extracts treated with TRG in serum-containing media for 24 hours following a 1 hour pretreatment with none (lane 2), LY294002 (lane 3), LY303511 (lane 4), or Akt inhibitor VIII (lane 5).

Figure 9

Model representing the signaling pathway of TRG-induced cellular effects in HCC cells . Incubation of the HCC cells with TRG in serum-containing media leads to a decrease in the expression of cyclin D1 resulting in cell growth arrest. TRG-induced reduction of cyclin D1 was shown to involve inhibition of CREB pathway in our earlier studies [32]. However, incubation with TRG under these conditions shows no apoptosis and leads to an increase in Akt^Ser473 and FoxO1^Thr24/3a^Thr32 phosphorylation involving PI3K and Pak pathways, which might lead to tumor progression. Inhibition of PI3K pathway but not Pak or Akt pathways sensitizes cells towards apoptosis. In addition, incubation with TRG in serum-deficient media antagonizes Akt^Ser473 phosphorylation and leads to potent apoptosis.