ELSEVIER
available at www.sciencedirect.com
ScienceDirect
-
Experimental CELL RESEARCH
Research Article
Rosiglitazone induces autophagy in H295R and cell cycle deregulation in SW13 adrenocortical cancer cells
Lidia Cerquettia,b, Camilla Sampaolia,b, Donatella Amendolab, Barbara Buccib, Laura Masuelli“, Rodolfo Marcheseb, Silvia Misitia,b, Agostino De Venanziª, Maurizio Poggiª, Vincenzo Toscanoª, Antonio Stiglianoa,b,*
ªEndocrinology, Department of Clinical and Molecular Medicine, Sant’Andrea Hospital, Faculty of Medicine and Psychology “Sapienza” University of Rome, Via di Grottarossa, 1035-00189 Rome, Italy bResearch Center S. Pietro Hospital, Via Cassia, 600-00189 Rome, Italy “Department of Experimental Medicine, “Sapienza” University of Rome, Rome, Italy
ARTICLE INFORMATION
Article Chronology: Received 5 October 2010 Revised version received 17 February 2011 Accepted 25 February 2011 Available online 3 March 2011
Keywords: Thiazolidinediones Autophagy Cell cycle Adrenocortical carcinoma
ABSTRACT
Thiazolidinediones, specific peroxisome proliferator-activated receptor- y (PPAR-y) ligands, used in type-2 diabetes therapy, show favourable effects in several cancer cells. In this study we demonstrate that the growth of H295R and SW13 adrenocortical cancer cells is inhibited by rosiglitazone, a thiazolidinediones member, even though the mechanisms underlying this effect appeared to be cell- specific. Treatment with GW9662, a selective PPAR-y-inhibitor, showed that rosiglitazone acts through both PPAR-y-dependent and -independent mechanisms in H295R, while in SW13 cells the effect seems to be independent of PPAR-y. H295R cells treated with rosiglitazone undergo an autophagic process, leading to morphological changes detectable by electron microscopy and an increased expression of specific proteins such as AMPKa and beclin-1. The autophagy seems to be independent of PPAR-y activation and could be related to an increase in oxidative stress mediated by reactive oxygen species production with the disruption of the mitochondrial membrane potential, triggered by rosiglitazone. In SW13 cells, flow cytometry analysis showed an arrest in the G0/G1 phase of the cell cycle with a decrease of cyclin E and cdk2 activity, following the administration of rosiglitazone. Our data show the potential role of rosiglitazone in the therapeutic approach to adrenocortical carcinoma and indicate the molecular mechanisms at the base of its antiproliferative effects, which appear to be manifold and cell-specific in adrenocortical cancer lines.
@ 2011 Elsevier Inc. All rights reserved.
Introduction
Adrenocortical carcinoma (ACC) is a rare and malignant endocrine tumour with a worldwide incidence of approximately two cases per million persons per year [1].
The long-term therapeutic results are devastating and largely dependent on the stage of the tumour. Currently, different kinds of treatment are available for ACC. Surgery represents the treatment of choice for patients with respective tumours in the primary and secondary stages and for local recurrence [2,3]. Moreover, several
* Corresponding author at: Endocrinology, Department of Clinical and Molecular Medicine, Sant’Andrea Hospital, Faculty of Medicine and Psychology, “Sapienza” University of Rome, San Pietro Hospital Research Center, Via Cassia, 600-00189 Rome, Italy. Fax: +39 6 33251278.
E-mail address: antonio.stigliano@uniroma1.it (A. Stigliano).
Abbreviations: ACC, adrenocortical carcinoma; TZDs, thiazolidinediones; PPAR-y, peroxisome proliferator-activated receptor-y; RGZ, rosiglitazone; AMPK, AMP-activated protein kinase.
0014-4827/$ - see front matter @ 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.yexcr.2011.02.014
cytotoxic agents have been used as monotherapy or in combina- tion to treat the disease in the advanced stages [4].
The lack of effective medical treatment is due to the fact that the mechanisms leading to the malignant transformation of adrenocortical cells have yet to be clarified; thus, the search for better medical treatment protocols for ACC is a continuing challenge.
Thiazolidinediones (TZDs) are a new class of antidiabetic drugs which attenuate the insulin resistance associated with obesity and hypertension, so that they are used as a new approach of type 2 diabetes treatment. TZDs act as ligands for the peroxisome proliferator-activated receptor-y (PPAR-y), a member of the nuclear receptor superfamily of ligand-dependent transcription factors, which is predominantly expressed in adipose tissue, but also in other tissues at much lower levels [5,6], involved in a variety of physiological processes, including the regulation of the metabolism, inflammation, cellular growth and differentiation [7-10]. Upon activation by its ligands, the PPAR-y acts as a transcription factor, forming a heterodimer complex with the retinoid X receptor, binding with the peroxisome proliferator response element (PPRE) within the promoter of target genes, modulating their expression [11-13].
Additionally some authors have reported that TZDs have tumour-suppressive effects on breast, prostate, colon, renal cell and non-small lung cancers through their effect on cell apoptosis and proliferation [5,14,15]. Moreover, the exact mechanism of the PPAR ligand inducing growth inhibition is not yet known. It has been demonstrated that growth inhibition could often be associated with the induction of cell cycle arrest and apoptosis and seemed unrelated to the PPAR-y signal [14,16,17].
Autophagy is a cellular process, regulating normal cytoplasmic and organelle turnover, characterized by the formation of double- or multiple-membrane surrounded cytoplasmic vesicles, named autophagosomes, surrounding cytoplasmic organelles such as mitochondria and endoplasmic reticulum. Subsequently, autopha- gosomes fuse with lysosomes and their contents are degraded by the lysosomal acidic hydrolases [18,19]. Initially, the autophagy was described as a survival strategy in times of nutrient limitation to produce ATP. Recently, this process has received much attention because when it is prolonged, proteins and organelles essential for basic homeostasis and cell survival are degraded, which can lead to cell death (autophagic cell death or programmed cell death type II, PCD II) [18,19].
AMP-activated protein kinase (AMPK) belongs to a conserved family of protein kinases activated by ATP depletion and consequent AMP accumulation, and plays an important role in regulating both fatty acid and glucose metabolism. Nevertheless, more recent reports clearly indicate that AMPK may contribute to growth inhibition, interfering with the mTOR (mammalian target of rapamycin) pathway and thus may exert a positive effect on autophagy. Han and Roman [20] found that rosiglitazone (RGZ), a member of the TZDs, increasing the phosphorylation of AMPKa, induces a decrease in the phosphorylation of p70 ribosomal protein S6 kinase (p70S6K), a downstream target of mTOR. The p70S6K is an important factor linking growth/proliferation signals to the regulation of protein synthesis and cell size [21]. It suggests that the treatment with RGZ may involve a molecular pathway independent from the PPAR-y receptor.
Reactive oxygen species (ROS) are molecules or ions that are formed by the incomplete one-electron reduction of oxygen. The
majority of the species of ROS include superoxide, hydrogen peroxide, and hydroxyl radical. ROS regulation of autophagy has also been demonstrated in many reports [22,23]. Several studies have also explained the role of TZDs in the induction of ROS and in the alteration of the mitochondrial membrane potential [24,25].
In this work we examined the effect of RGZ on cell proliferation using H295R and SW13 ACC cell lines. We found that the events responsible for the growth-inhibitory effects were remarkably heterogeneous depending on cell type sensitivity and on the induction of the activation of different molecular pathways. In particular we demonstrate that RGZ in the H295R cell line induces cell growth inhibition through extensive vacuolisation and autophagic cell death; instead, in the SW13 cell line the inhibitory effect on cell proliferation was due to cell cycle perturbation, which promotes a G0/G1 delay, affecting cell cycle related proteins.
Materials and methods
Cell culture and treatment
Human H295R ACC cell line was maintained in a humidified 5% CO2, 37 ℃ incubator in a culture medium consisting of a mixture of 1:1 DMEM/F12, enriched with insulin/transferring/selenium, 10% fetal bovine serum, 2 mM glutamine, 100 U/ml penicillin, and 100 µg/ml streptomycin. The ACC SW13 cell line was cultured in Leibovitz’s medium supplemented with 10% fetal bovine serum, 100 U/ml penicillin and 100 µg/ml streptomycin. The two cell lines, H295R and SW13, were treated with RZG (Vinci-Biochem) at concentration of 20 and 10 uM, respectively (supplied by Glax- oSmithKline, Munich, Germany). RGZ was administered every 3 and 6 h, throughout the entire study. The multiple administration regimen is due to the short half-life of the drug (3-4 h) [26]. The two cell lines also were treated with 20 µM GW9662 (purchased from Calbiochem, Merck Darmstadt, Germany), dissolved in dimethylsulfoxide and was given 1 h before RZG treatment. Final concentration of dimethylsulfoxide in cell culture medium was adjusted to 0.03%.
H295R cell line was treated with 5 mM 3-methyladenine (3- MA) (Sigma-Aldrich, Milan, Italy) directly dissolved in the culture medium and was given 1 h before RGZ treatment.
Trypan blue analysis
Cell number was determined using a hemocytometer. The presence of necrosis or apoptosis was determined by trypan blue staining. After trypsinization, cells were suspended in PBS and mixed with equal volume of 0.4% trypan blue in PBS, and the percentage of stained cells was determined.
MTT assay
H295R and SW13 cell lines were seeded on 96-well plates and treated with RZG, GW9662, or both; cell proliferation was evaluated by MTT (Cell 96 Non-Radioactive Cell proliferation Assay) (Promega Corporation, Madison, WI, USA), based on the transformation and colorimetric quantification of 3-(4,5- dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide after 24, 48, and 72 h from treatment. Cells were incubated with dye
solution for 4 h at 37 ℃, and then the conversion reaction of the tetrazolium salt in formazan was arrested by incubating the cells with Stop Solution for 1 h at 37 ℃. The absorbance of the solution was measured using an ELISA plate reader (Bio-Rad Laboratories, Hercules, CA) at a wavelength of 570 nm.
Cell cycle analysis
As to BrdUrd incorporation assay, both H295R and SW13 cell lines were exposed to RGZ for 72 h. BrdUrd was added to the medium at the dose of 10 UM during the last 30 min before analysis. At indicated times, the cells were harvested, washed once in PBS, fixed in 70% ethanol, and stored at 4 ℃ before the analysis. Then, samples were incubated with 4N HCl for 20 min at room temperature to partially denature DNA. Cells were washed once in borax-borate buffer (pH 9.1) to neutralize the acid pH and then once in PBS. Cells were then permeabilized by incubation with complete medium plus 20% FCS and 0.5% Tween 20 for 10 min. The samples were then incubated with mouse monoclonal antibody anti-BrdUrd (Roche Diagnostics, Milan, Italy) in complete medium containing 20% FCS and 0.06% Tween 20 (Calbiochem, San Diego, CA) at room temperature for 1 h. After washing in PBS, cells were incubated with FITC-conjugated F(abV) rabbit anti-mouse IgG (DAKO, Glostrup, Denmark) in PBS for 1 h. Finally, after washing in PBS, cells were stained with a solution containing 5 µg/ml PI and 75 KU/ml RNase in PBS for at least 3 h. Twenty thousand events per sample were acquired by using a FACScan cytofluorimeter. The top region of the cytograms represents BrdUrd-positive cells. The percentages of the cell cycle distribution were estimated on linear PI histograms by using the MODFIT software.
Electron microscopy
Cell samples for electron microscopy were fixed in 2.5% glutaral- dehyde in PBS, pH 7.4, at 48 ℃. They were then washed in PBS and postfixed in 1.33% osmium tetroxide for 2 h at 48 ℃. After several washes in PBS, the samples were dehydrated in graded alcohol, transferred into toluene, and embedded in Epon 812 resin. The resin was allowed to polymerize in a dry oven at 60 ℃ for 24 h. Thin sections were cut with a glass knife on a Reichert microtome, stained with toluidine blue, and examined on Axioscope micro- scope (Zeiss Jena GmbH, Germany). Ultrathin sections were cut on a Richert microtome using a diamond knife, stained with uranyl acetate-lead citrate and evaluated on a Philips electron micro- scope Morgagni 268D (Philips, Endhoven, The Netherlands). Each observation was carried out independently 6-7 times per sample [27].
Measurement of ROS content
For the reactive oxygen species content analysis, control and treated adherent cells were first assayed for viability by trypan blue dye exclusion, incubated with 4 µM dihydroethidium (Mo- lecular Probes, Eugene, OR) for 45 min at 37 °℃, and then analyzed by using flow cytometry at the indicated times.
Determination of mitochondrial membrane potential
Mitochondrial membrane potential on control and RGZ treated cells has been evaluated using the lipophilic cationic probe JC-1
(5,5’6,6’-tetrachloro-1,1’,3,3’-tetraethylbenzimidazolcarbocya- nine iodide; Molecular Probes). It is mitochondria-selective and it has been used successfully for flow cytometric measurements of mitochondrial potential because of its dual emission character- istics that are sensitive to membrane potential. JC-1 changes reversibly its color because of its dual emission characteristics, which are sensitive to membrane potential. JC-1 is mitochondria- selective and in normal polarized mitochondria results in a green- orange/red emission after excitation at 488 nm. Green fluores- cence represents the monomeric form of JC-1 and it is collected on FL-1 channel at 530 nm while orange /red, collected on FL-2 channel at 585 nm, corresponds to the J-aggregate form of JC-1. The color of the dye changes from orange/red to green fluores- cence as the mitochondrial membrane becomes depolarized. Briefly, cells were washed once in PBS and resuspended in 0.5 ml PBS containing JC-1 at the final concentration of 1 µM. After 15 min of incubation at 37 ℃ the reaction was stopped on ice and samples were analyzed by flow cytometry at the indicated times. Valinomycin (10 µM)-induced depolarization was used as positive control. Data were analyzed utilizing a computer quadrant region to quantify the percentages of orange/red and green fluorescence. 1 quadrant region represents normal potential mitochondrial membrane; 2 quadrant region represents intermediate mitochon- drial membrane; 3 quadrant region represents low mitochondrial membrane and 4 quadrant region represents very low mitochon- drial membrane.
Western blotting analysis
Cellular lysates were sonicated on ice, clarified by centrifugation at 20,000 x g and stored at -80 ℃. An aliquot of the cell lysates was used to evaluate the protein content by colorimetric assay. Total protein content (70 µg) was electrophoresed on 10% polyacryl- amide gel in the presence of SDS and transferred onto a nitrocellulose membrane. Blots were blocked for 1 h at room temperature with 5% non-fat dry milk in T-PBS buffer. Treated and untreated cells were incubated with the anti- PPAR-y 1:100 (Santa Cruz Biotechnology, CA, USA), Beclin-1 1:100 (Santa Cruz Biotechnology, CA, USA), cyclin E 1:500 (BD Biosciences, Franklin Lakes, NJ, USA), cdk2 1:800 (BD Biosciences, Franklin Lakes, NJ, USA), pRB (Ser 780) 1:500 (Santa Cruz Biotechnology, CA, USA), anti-pan-cytokeratin 1:100 (Sigma-Aldrich, Saint Louis, MO, USA), p-AMPK& 1:200 (Santa Cruz Biotechnology, CA, USA), ß-actin 1:10,000 (Sigma-Aldrich, Saint Louis, MO, USA), and vinculin 1:10,000 (Sigma-Aldrich, Saint Louis, MO, USA). The visualization of the antigens was done by enhanced chemiluminescent detection reagents by ECL. The analysis of bands was performed with Image] (Image Processing and analysis in Java) software program.
Immunofluorescence analysis
H295R cell lines were seeded on coverslip and treated with 20 uM RGZ. At the time points indicated, 24-48-72h, the cells were washed twice in phosphate buffer saline (PBS) and fixed in 4% paraformaldehyde for 15 min at room temperature. Fixed cells were blocked in 3% bovine serum albumin (BSA, Sigma-Aldrich, Germany) for 15 min. RT prior to incubation with primary antibody. Antibody anti-Lamp1 1:40 was diluted in BSA. After a 1 h incubation at RT, the slides were washed twice with PBS and
incubated with secondary polyclonal antibody rabbit anti-mouse TRITC (DakoCytomation) for 1 h at room temperature. Then cells were washed twice in PBS and incubated for 20 min with Hoechst 33258 (50 ng/ml). Stained cells were washed, then mounted in 50% (v/v) glycerol-PBS and examined by fluorescence microscopy.
Statistical analysis
All experiments were repeated at least three times and each experiment was carried out at least by duplicates. The data were presented as means ±SD (standard deviation). A comparison of the individual treatment was conducted by using Student’s t test employing at least three independent data points. Statistical correlation of data was also checked for significance by the ANOVA test. A p value <0.05 was considered significant. The software used was Excel and ImageJ.
Results
RGZ induces cell growth inhibition in both SW13 and H295R cell lines with different mechanisms
So far three reports have examined the effect of TZDs on ACC cell lines, demonstrating an antiproliferative effect of these drugs [28-30]. In this study, we investigated the effects of the treatment with RGZ in H295R and in SW13 ACC cell lines. As shown in Fig. 1 the ability of RGZ to arrest cell proliferation was evident since 48 h from the beginning of treatment (- 40% and - 31% in H295R and SW13 cells, respectively, compared with control cells) (p<0.01) and reached its maximum after 72 h, exceeding 50% of growth inhibition for both cell lines (p <0.01). Although RGZ was known to be a selective ligand for PPAR-y, the use of GW9662, its selective inhibitor, at a concentration of 20 uM, only partially restored cell proliferation in H295R cells. In fact after 72 h we observed 34% of inhibition with GW9662 vs. 55% with RGZ alone (Fig. 1a). Treatment did not induce cellular toxicity as shown in Fig. 1c.
On the contrary, in SW13 cell line, GW9662 20 uM did not affect the inhibition of growth induced by RGZ. As expected, cells treated with RGZ and GW9662 in combination showed the same trend of cells treated with RGZ alone (51% of cell growth inhibition compared with control after 72 h), suggesting that RGZ exerted its antiproliferative effect through a mechanism which does not involve PPAR-y receptor, as GW9662 alone did not alter neither cell growth nor vitality (Fig. 1b and d). The effect of RGZ and GW9662 on cell growth was confirmed by MTT assay on both cell lines (Fig. 1e and f).
This data indicates that the drug acts through both PPAR- y-dependent and -independent mechanisms in H295R cell line. By the contrary, in SW13 cell line the effect of the drug seems to be completely independent of PPAR-y.
Expression levels of PPAR-y protein were evaluated by Western blotting and densitometric analysis, using a specific anti-PPARy antibody. The results shown in Fig. 2 indicate that RGZ is able to induce the expression of PPAR-y at 24 and 48 h compared to control (p<0.01) in SW13 cell line, and at 48 h (p<0.01) in H295R cell line. Moreover GW9662 administration was able to inhibit the receptor both alone and in combination with RGZ (p<0.01) (Fig. 2a and b).
These findings therefore suggest that the mechanism of action of RGZ is highly cell-specific, and that some PPAR-y-independent signals may mediate the effects of this drug on cell growth.
RGZ impairs DNA synthesis in both SW13 and H295R cell lines
The cell cycle analysis was performed by BrdUrd incorporation in both H295R and SW13 cell lines. Fig. 3a and b shows that RGZ affected cell cycle in both cell lines. In SW13 cells, RZG treatment was able to completely impair DNA synthesis. Indeed, the analysis of cytograms already at 24 h after treatment clearly showed that BrdUrd incorporation has been completely inhibited (6%) compared with control cells (28%) (Fig. 3b). On the contrary, in H295R cells, a significant BrdUrd depletion occurred only at 72 h (10%) (Fig. 3a) when a fraction of cells, which were unable to re-enter in the cell cycle, were going to die for autophagy. The inhibition of BrdUrd incorporation observed in RGZ exposure was consistent with a reduced cell percentage in S phase observed in both cell lines, as estimated by applying to each PI histogram the MODFIT software. This analysis also showed the G1 accumulation.
RGZ induces cell cycle deregulation in SW13 cell line
Cell cycle analysis performed by BrdUrd incorporation permitted to observe a marked G1 accumulation in treated SW13 cell line (Fig. 3b).
We thus investigated the effect of RGZ on proteins which are known to regulate G1/S transition in SW13 cells. As detected by Western blotting, the drug was able to induce a marked reduction of cyclin E expression, as well as of Cdk2, its associated kinase (Fig. 4b and c). Moreover, we evaluated the levels of activation of retinoblastoma protein, which is one of the targets of Cdk2. Activating phosphorylation of Rb at Serine 780 was seen to be diminished after RGZ administration, thus confirming a decreased functionality of the cyclin E-Cdk2 complex (Fig. 4d). The intensity of the bands was quantified using Image] and B-actin was used for normalization.
RGZ induces vacuolization and autophagy in H295R cell line
To determine the effect of RGZ on cell morphology, we observed cells by contrast light microscope. In H295R cells, vacuoles were recognizable in the cytoplasm approximately 24 h after RGZ administration and then we observed an enlargement in their volume and density. The percentage of cells with a visible vacuole increased exponentially and at 48 h, and about 65% of cells were vacuolated (data not shown).
To assess whether cellular vacuolisation induced by RGZ had reference to authophagy, we performed an ultrastructural analysis by transmission electron microscopy. As shown in Fig. 5, the majority of treated cells, but not control cells, enclosed many autophagic vacuoles surrounded by double membrane and containing almost intact organelles, thus demonstrating the occurrence of autophagy.
In order to evaluate the specificity of RGZ effect on the autophagic process, we administered 3-MA, a compound able to inhibit autophagy before autophagosome development, to cell culture. As shown in Fig. 5, samples treated with 3-MA did not show autophagic vacuoles in the cytoplasm.
a
H295R
6
5
nr cells x 10
4
C
- GW9662
3
-RGZ
- GW9662 + RGZ
2
1
0
24
48
72
96
time (h)
b
SW13
1,8
1,5
nr cells x 16
1,2
C
- GW 9662
0,9
- RGZ
- GW9662 + RGZ
0,6
0,3
0,0
0
24
48
72
96
time (h)
c
d
100
100
90
90
% of trypan blue positive cells
80
% of trypan blue positive cel $
80
70
70
60
24 h ☐
60
☐ 24 h
50
☒ 48 h
50
☒ 48 h
40
☒ 72 h
40
☒ 72 h
30
30
20
T
-
20
I
E
I
1
I
1
I
10
E
I
I
10
I
I
I
I
I
I
I
I
0
0
C
GW9662
RGZ
GW9622 + RGZ
c
GW9662
RGZ
GW9622 + RGZ
e
f
0,30
0,07
**
**
0,25
**
0
T
0,06
L
1
**
I L
Y
1
1
L
**
0,05
Y
T
0,20
**
L
L
f
1
1
☐
C
T
T
I
☐
C
0,04
T L
T
T L
1
8
☐ GW9662
0,15
1
L
T
@ GW9662
L
I
1
8
1
RGZ
0,03
☐ RGZ
GW9662 +RGZ
GW9662 + RGZ
0,10
0,02
0,05
0,01
0,00
0,00
24
48
72
time (h)
24
48
72
time (h)
a
H295R
24 h
48 h
72 h
RGZ 20UM
-
-
+
+
-
-
+
+
-
-
GW9662 20UM - +- + - + - + -+ - +
+
+
67 KDa
PPARY
42 KDa
B-actin
**
2,8
1
2,4
2
I
1,6
T
☐
C
☐
GW9662
1,2
☒
ROZ
r I
GW9662-ROZ ☒
0,8
I
0,4
£
1
I
0
24
48
72
tine (1)
b
SW13
24 h
48 h
72 h
RGZ 10UM -
-
+
+
-
-
+
+
-
-
GW9662 20uM - + - + - + - + - + - +
+
+
67 KDa
PPARY
42 KDa
B-actin
**
**
2,1
1,8
I
L
T
**
1,5
1
1,2
Ţ
☐ c
1
T
☐ GW9662
I
Z
0,9
I
1
I
RGZ
-
T
GW9662-ROZ
0,6
Z
L
0,3
0
24
48
72
time (1)
To evaluate if RGZ was able to induce autophagic cell death also in SW13 cells we also performed electron microscopy analysis. Nevertheless, in this case it appeared that RGZ did not trigger autophagy, as treated cells did not show evident morphological differences from control cells and no vacuolisation was detectable (data not shown).
The ultrastructural analysis of cells also showed absence of apoptosis, as neither chromatin condensation nor nuclear frag- mentation were visible in treated cells of both lines.
Since an intact cytoskeleton was necessary in autophagic cell death, we analyzed the expression level of cytokeratin by Western blotting after RGZ treatment. As shown in Fig. 6a, the treatment cause
b SW13
Control
RGZ
H295R
Control
RGZ
28%
%
6%
%
G1 60
G1 75
21%
%
%
26%
S 27
G1 56
S 12
G1 58
S 24
G2 13
S 26
G2 13
24h
G2 16
G2 20
24h
%
5%
%
24%
%
%
25%
G1 64
G1 86
FL1-H
G1 63
12%
G1 75
FL1-H
S 22
S 22
S 11
G2 15
G2 14
G2 14
S 8
G2 6
48h
48h
25%
%
10%
%
G1 61
G1 78
20%
%
1%
%
S 23
S 9
G1 72
G1 91
G2 16
G2 13
72h
S 16
G2 12
S 1
G2 8
72h
FL3-H
FL3-H
a
24 h
48 h
72 h
RGZ 10UM
+
+
+
-
-
-
50 KDa
cyclin E
33 KDa
Cdk2
116 KDa
pRb
42 KDa
B-actin
b
2
cyclin E
**
1,6
I
I
1,2
rd u
IC
0,8
*
RGZ
I
0,4
T
1
0
24
48
72
áme (h)
c
2
Cdk2
**
1,6
1
1.2
**
£
T
1
C
0,8
RGZ
0,4
-
0
24
48
72
time (h)
d
1
pRb
3KS
**
0,S
I
**
0.6
C
A
0,4
I
RGZ
0,2
0
24
4S
72
time (h)
any modification in the expression pattern of cytokeratins, thus supporting our previous data relative to autophagy; in fact, during this process, microtubules and intermediate filaments are preserved, in order to allow the formation of autophagic vacuoles and to permit the transport of cytoplasmic organelles to the site of their degradation.
Autophagy RGZ-induced involves AMP-activated protein kinase activation in H295R cell line
As autophagy appears to be a process subjected to many regulatory steps and that can be triggered by a lot of different stimuli, we investigated the role of AMPK in the RGZ-induced autophagy observed in H295R cell line. It is indeed well documented that RGZ is able to promote the phosphorylation of the @-subunit, that is the
catalytic one, of AMPK and that this kinase is involved in the induction of autophagic process, either directly or through the inhibition of cell survival cascades.
Western blot performed on H295R cells treated with rosigli- tazone showed an increase of 1.5-fold of the phosphorylated form of AMPKa with respect to control cells (p<0.01), which main- tained for the entire duration of experiments, thus confirming the ability of the drug to activate this protein (Fig. 6b).
RGZ promotes ROS formation and disrupts normal mitochondrial membrane potential in H295R cells
Some authors [22,23,31] reported that ROS were necessary for autophagy inducing a selective degradation of damaged mitochondria
H295R
+ 3-MA
Control 72h
-
RGZ 24h
@
RGZ 48h
RGZ 72h
and peroxisome. The autophagy is a major catabolic pathway by which eukaryotic cells degrade and recycle macromolecules and organelles as it is considered a type of programmed cell death under certain conditions. This pathway is activated under environmental stress conditions, during development and in several pathological situations. In this study, we describe the role of ROS as signalling molecules in autophagy. Thus, in the attempt to understand where ROS production fits in the autophagic signalling pathway in H295R cell line, we monitored the RGZ effects. As reported in Fig. 7 the RGZ treatment significantly increased, in time-dependent manner, the level of ROS, from 28% at 24 h from treatment until 58% at 72 h.
We also evaluated the effect of rosiglitazone on ROS production in SW13 cells. The amount of ROS was measured by FACS analysis and showed no significant differences between control and treated cells (8% and 6%, respectively after 72 h) (data not shown). Therefore we concluded that growth inhibition induced by RGZ in SW13 cell line is not related to autophagy nor to an increase in ROS generation.
To delineate the mechanism of RGZ cytotoxicity in H295R cell line, the lipophilic cation JC-1 was used to determine whether RGZ treatment induced alterations in mitochondrial membrane poten- tial in these cells [32]. The cells were untreated and treated with RGZ for 24-48 h, stained with JC-1, and analyzed by FACS at the indicated times. Fig. 8 shows dot plot analysis of JC-1 mitochondria membrane potential by flow cytometry measurement. As evident in this figure, JC-1 is present both in monomeric and in aggregated forms in control polarized mitochondrial membranes resulting in a basal green vs. orange/red fluorescence emission (Fig. 8a and b, quadrant 1). On the contrary, in RGZ treated samples, the JC-1 monomeric green-emitting form increased with a marked loss of the orange/red fluorescence, indicating that these cells have
depolarized the mitochondrial membrane potential. Indeed, as shown in the figure, it is evident as RGZ clearly increased the percentage of cells that emitted higher green fluorescence associated to the lost of orange/red fluorescence (45%), compared to control cells (3%) (Fig. Se and f, quadrant 4) (p<0.01). This value reached 55% at 48 h after treatment (p<0.01). To verify the sensitivity of the JC-1 experiments, we used the valinomycin- induced depolarization (Fig. 8c and d).
RGZ stimulates proteins involved in different phases of the autophagic process
Finally, we investigated the effect of RGZ on Beclin-1 and LAMP-1, two proteins involved in different steps of the autophagic process. Beclin-1 is a cytoplasmatic protein required for the formation of autophagosome, which occurs during the first phases of autop- hagy. During this event, levels of expression of Beclin-1 increase; thus, we performed a Western blot analysis in our cells, in order to confirm the occurrence of autophagy after giving RGZ. Fig. 9 shows as in H295R cell line expression of Beclin-1 increased, reaching its maximal during the first 48 h following RGZ administration. After this time, instead, we noticed a decrease in the expression level of this protein.
Modulation of Beclin-1 was investigated also in SW13 cell line, and the results showed no difference between controls and treated cells at any of the times considered (24, 48 and 72 h from treatment) (data not shown).
Since autophagy is a catabolic process involving the degrada- tion of cell through the lysosomal machinery, the state of lysosome-associated membrane glycoprotein 1 (LAMP-1) was investigated by immunofluorescence experiments. We found an increase of LAMP-1 signal in subcellular compartment of H295R
a
24 h
48 h
72 h
RGZ 20UM
+
+
+
-
-
-
69 KDa
59 KDa
pan-cytokeratin
58 KDa
-
4
117 KDa
vinculin
:
1,6
1,2
1
0,5
T
I
T
Oc
ROZ
1
0,4
0
24
4S
72
tin (1)
b
24 h
48 h
72 h
RGZ 20µM
+
-
+
-
+
-
63 KDa
p-AMPKa
117 KDa
vinculin
2
3333
1,6
1,2
ã
*
*
Oc
0,s
ROZ
I
T
0,4
L
T
0
24
4S
2
time (1)
cells, in fluorescent staining at 48 and 72 h after RGZ administra- tion with respect to control cells (Fig. 10). This result was consistent with previous data, showing that the treatment induced autophagy by increasing lysosome organelles in number and size. As expected, localization of LAMP-1 in SW13 cells appeared not to be altered by RGZ treatment (data not shown).
Discussion
It has previously been demonstrated that the PPAR-y agonist induces cell death in a variety of cancer cells [13-15]. Although growth inhibition in ACC cells by the PPAR-y agonist had already been demonstrated by some authors [28-30], there was little data regarding its molecular mechanism of action.
In this study, we have evaluated the effects of RGZ, a synthetic PPAR-y ligand, belonging to the family of TZDs, in ACC cell lines,
with respect to the mechanism of the action of the drug. We can report that RGZ induces cell cycle deregulation in SW13 and our findings report autophagic cell death in the H295R ACC cell lines for the first time.
We report that RGZ inhibits cell proliferation on H295R and SW13 cells, involving both PPAR-y dependent and independent signals. The receptor dependence of the antineoplastic effect induced by TZDs is still under debate. Ferruzzi et al. [28] and Betz et al. [29] demonstrated the expression of PPAR-y in some ACC and in the H295R cell line. In this paper, we demonstrate, in both cell lines the expression of the receptor and its induction after the administration of its ligand. Conversely, the GW9662 antagonizes the effect of RGZ, reducing the expression of PPAR-y compared with cells treated only with rosiglitazone and with rosiglitazone and GW9662 in combined treatment (Fig. 2a and b). This data suggests a proper function of the ligand receptor system in these cellular models of ACC.
H295R
Control
RGZ
100 150 200 250
250
200
FSC-H
FSC-H
150
100
24h
8
2%
3
28%
0
0
10
0
10
10€
O
FL3-H
10
104
10
10
102
FL3-H
10
104
100 150 200 250
250
200
FSC
FSC-H
FSC-H
150
100
8
1%
48h
3
56%
0
0
0
10
10
102
103
104
10
0
FL3H
10
102
FL3-H
103
104
50 100 150 200 250
250
200
FSC-H
FSC-H
150
100
72h
1%
8
58%
0
10
0
10
102
0
FL3-H
10°
10
0
10°
10
102
FL3H
10%
104
ROS content
Control
Valinomicyn
RGZ
a
104
c
e
1
2
104
1
2
104
1
103
2
103
103
FL2H
102
FL2H
102
FL2H
10?
4
24h
red/orange fluorescence
3
3
3
4
3
3
4
0
100
3%
8
65%
45%
100
10 1
102
FL1-H
103
104
1.00
100
101
102
FL1-H
103
104
100
101
102
FL1-H
103
10
b
B
d
104
f
1
2
1
2
104
1
2
103
103
FL2H
3
102
FL2H
102
FL2H
102
3
4
3
4
3
3
48h
4
0
3
9%
3
100
100
70%
100
55%
100
101
102
FL1-H
103
104
100
101
102
10
4
FL1-H
103
100
101
102
FL1-H
103
104
green fluorescence
24 h
48 h
72 h
RGZ 20uM
+
+
+
-
-
-
60 KDa
BECN-1
42 KDa
B-actin
2
*
1,6
**
E
1,2
**
oc
0,s
ROZ
Y
I
0,4
1
0
24
4S
72
time (1)
However, the majority of authors retain these effects mediated by other PPAR-y pathways. Many human cancer cell lines, indeed, have been reported to exhibit high levels of PPAR-y expression. In vitro exposure of these cancer cell types to high doses (≥10 uM) of TZDs induces a suppression of tumour cell proliferation and the induction of a more differentiated tumour phenotype, suggesting a link between PPAR-y signalling and the antineoplastic activities of the TZDs [14,15,18]. Furthermore, other types of cancer, such as
LNCaP prostate cancer and MCF-7 breast cancer cells, exhibiting a low level of expression of PPAR-y, were more sensitive to the effects of TZDs on growth inhibition than their PPAR-y counter- parts, PC3 and MDA-MB-231, respectively [16,17]. Moreover, there is a large discrepancy between the concentration required to induce antineoplastic effects and that for PPAR-y activation. In our work, the use of the GW9662 receptor antagonist is only partially able to reduce the inhibition of proliferation in the H295R cell line.
a
b
c
d
Control
e
f
g
h
RGZ 48h
i
j
k
1
RGZ 72h
This data strongly supports the hypothesis that the TZDs interfere with multiple signalling mechanisms independent of PPAR-y activation, as also happens in our cellular models. The ability of TZDs to induce an antiproliferative effect on tumours belonging to different lineages and with a heterogeneous genotypic profile (such as H295R and SW13 ACC cell lines) is an encouraging discovery for the field of oncology.
Regarding the effect induced by TZDs on the cell cycle in other malignant cell lines, many studies report a modification in the cell cycle checkpoint [33,34]. Our data shows that RGZ induces cell cycle deregulation interfering in the G0/G1 phase of the cell cycle by cyclin E and cdk2 modulation in the SW13 cell line.
Conversely, we found that RGZ in the H295R ACC cell line leads to the activation of autophagy-mediated cell death in particular versus mitochondrial autophagy. The autophagic process was shown using different approaches: an ultrastructural study to evaluate the development of cytoplasmic vesicles, Western blot analysis to demonstrate the molecular pathway of autophagy (Beclin-1, cytokeratin and p-AMPK& expression), flow cytometry to assess ROS content and mitochondrial membrane potential, and immunofluorescence to detect Lamp-1 lysosomal protein.
Many studies have now clearly shown that autophagy correlates with the ROS intracellular increase. ROS are generally small molecules with a short half-life. They include O2 anions, free radicals and peroxides. These species are produced by ionizing radiation, products of mito- chondrial respiration and enzymes of the NADPH family [22,25,31]. ROS accumulation is related to oxidative stress to which cells reply by activating various defences or, finally, by dying. Autophagy, a process by which cells degrade and recycle macromolecules, indeed, is also closely related to the cellular response to oxidative stress [22,31]. The role of mitochondria in autophagy induction is fundamental in autophagoso- mal biogenesis. This process is linked to ROS generation causing a loss of mitochondria membrane permeability [24,25,35]. ROS derived from mitochondria, indeed, function as signalling molecules in the autophagic process, leading the cells in many instances to survival or death. This data is quite consistent with the finding that TZDs induce the release of ROS from mitochondria [25]. Indeed, we report that RGZ treatment in the H295R cell line induces ROS production in a time-dependent manner (Fig. 7), accompanied by a loss of mitochondrial membrane potential as suggested by Rodriguez-Enriquez et al. (Fig. 8) [24]. This data is consistent with that produced by some authors who argue that cell death induced by TZDs in some types of cancer is linked to ROS generation and associate this phenomenon with the autophagic process [22,23,25].
In addition, since AMPK (AMP-activated protein kinase) is considered a sensor of cellular energy, it seems plausible that this protein may play a role in the regulation of intracellular protein degradation. It is known, indeed, that the yeast ortholog of AMPK (Snf1) stimulates autophagy in yeast. In accordance with this, we found an increased level of expression of p-AMPK protein after RGZ treatment, indicating that RGZ mediates the induction of the autophagic process [36,37].
Autophagy is a process leading to a continuous turnover of intracellular constituents by means of the removal of damaged or unwanted products, through an autophagosomic-lysosomal path- way that finally leads to protein degradation [36]. This process is characterized by multiple steps which lead to the final event of the autophagolysosome formation. In this biological process, we demonstrate the involvement, as mentioned above, of some proteins with a pivotal role in the autophagic cell death, such as beclin-1, which participates in the formation of autophagosomes
and the Lamp-1 protein, a highly glycosylated protein associated with the lysosome membrane, involved in the autophagolisosomal formation. Our results suggest that the RGZ treatment is involved in autophagy process by altering the level of expression of proteins involved in the major events associated with the autophagic pathway.
We hypothesize that the mechanism by which RGZ triggers the autophagic pathway in the H295R cell line depends on the phosphorylation of & subunit of AMPK with ROS production followed by an alteration of the permeability of the mitochondrial membrane. At molecular level, RGZ induces the expression of beclin-1 and the aggregation of Lamp-1 (Figs. 9 and 10).
Numerous reasons are attributed to the autophagic process. This acts in particular situations of metabolic distress to preserve cell viability. Sometimes autophagy, also known as programmed cell death type II, is irreversible and represents a real mechanism of cell death [38,39].Given the series of events triggered by RGZ, these should be interpreted as significant of a programmed cell death in H295R ACC cells.
Betz et al. [30] describe an apoptotic cell death in the H295R cell line induced by TZDs. However, the integrity of the cytoskeleton, investigated by the expression of cytokeratin, needed in the autophagic process but not in apoptosis, clarifies the autophagy which took place in our cells. The seemingly contradictory aspect of this data really fits with the latest acquisitions concerning the mechanisms of cell death in which apoptosis and autophagy might be mechanisms chained together [38,39]. In this regard we report a typical example deriving from two recent reports describing RGZ as able to induce apoptosis and autophagy in the same breast cancer cell line [40,41].
On the basis of this data, further investigation of the way in which RGZ affects intracellular signalling in different ways in the two H295R and SW13 ACC cell lines are needed. Our results reveal a novel biological function for the RGZ compound and point to new possibilities in developing innovative therapeutics in ACC .
Knowledge of the biological mechanisms of the PPAR-y ligands is increasingly urgent in the light of the vast amounts of data accumulated concerning the inhibitory effect in cancer. Further- more, we retain that these results represent an interesting source of speculation not only on the mechanisms of the action of ligands of PPAR-y but also on those of switch in cell death.
Acknowledgments
This work was financed by Ateneo ‘06 research grant number: 8.1.1.1.2.16 from ‘Sapienza’ Università di Roma. The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the research reported.
REFERENCES
[1] M.B. Lipsett, R. Hertz, G.T. Ross, Clinical and pathophysiologic aspects of adrenocortical carcinoma, Am. J. Med. 35 (1963) 374-383.
[2] K. Cohn, L. Gottesman, M. Brennan, Adrenocortical carcinoma, Surgery 100 (1986) 1170-1177.
[3] R.F. Pommier, M.F. Brennan, An eleven-year experience with adrenocortical carcinoma, Surgery 112 (1992) 963-970.
[4] B. Allolio, S. Haner, D. Weismann, M. Fassnacht, Management of adrenocortical carcinoma, Clin. Endocrinol. 117 (2004) 1142-1146.
[5] C.R. Kahn, L. Chen, S.E. Cohen, Unraveling the mechanism of action of thiazolidinediones, J. Clin. Invest. 106 (2000) 1305-1307.
[6] O. Braissant, F. Foufelle, C. Scotto, M. Dauça, W. Wahli, Differential expression of peroxisome proliferator-activated receptors (PPARs):tissue distribution of PPAR-alpha, -beta, and -gamma in the adult rat, Endocrinology 137 (1996) 354-356.
[7] S.A. Kliewer, B.M. Forman, B. Blumberg, E.S. Ong, U. Borgmeyer, D.J. Mangelsdorf, K. Umesono, R.M. Evans, Differential expression and activation of a family of murine peroxisome proliferator-activated receptor, P.N.A.S. USA 91 (1994) 7355-7359.
[8] T.M. Willson, P.J. Brown, D.D. Sternbach, B.R. Henke, The PPARs: from orphan receptors to drug discovery, J. Med. Chem. 43 (2000) 527-550.
[9] M. Stumvoll, H.U. Haring, Glitazones: clinical effects and molecular mechanism, Ann. Med. 34 (2002) 217-224.
[10] E.D. Rosen, B.M. Spiegelman, PPARgamma: a nuclear regulator of metabolism, differentiation, and cell growth, J. Biol. Chem. 276 (2001) 37731-37734.
[11] S.A. Kliewer, K. Umesono, D.J. Noonan, R.A. Heyman, R.M. Evans, Convergence of 9-cis retinoic acid and peroxisome proliferator signalling pathways through heterodimer formation of their receptor, Nature 358 (1992) 771-774.
[12] S.A. Kliewer, K. Umesono, D.J. Mangelsdorf, R.M. Evans, Retinoid X receptor interacts with nuclear receptors in retinoic acid, thyroid hormone and vitamin D3 signalling, Nature 355 (1992) 446-449.
[13] P. Tontonoz, S. Singer, B.M. Forman, P. Sarraf, J.A. Fletcher, C.D. Fletcher, R.P. Brun, E. Mueller, S. Altiok, H. Oppenheim, R.M. Evans, B.M. Spiegelman, Terminal differentiation of human liposarcoma cells induced by ligands for peoxisome proliferator-activated receptor gamma and the retinoid X receptor, P.N.A.S. USA 94 (1997) 237-241.
[14] V.G. Keshamouni, R.C. Reddy, D.A. Arenberg, B. Joel, V.J. Thannickal, G.P. Kalemkerian, T.J. Standiford, Peroxisome proliferator-activated receptor-gamma activation inhibits tumor progression in non-small-cell lung cancer, Oncogene 23 (2004) 100-108.
[15] E. Mueller, M. Smith, P. Sarraf, T. Kroll, A. Aiyer, D.S. Kaufman, W. Oh, G. Demetri, W.D. Figg, X.P. Zhou, C. Eng, B.M. Spiegelman, P.W. Kantoff, Effects of ligand activation of peroxisome proliferator-activated receptor gamma in human prostate cancer, P.N.A.S. USA 97 (2000) 10990-10995.
[16] C.W. Shiau, C.C. Yang, S.K. Kulp, K.F. Chen, C.S. Chen, J.W.Huang, C.S. Chen, Thiazolidinediones mediate apoptosis in prostate cancer cells in part through inhibition of Bcl-xL/Bcl-2 functions indipendently of PPARgamma, Cancer Res. 65 (2005) 1561-1569.
[17] J.W. Huang, C.W. Shiau, Y.T. Yang, S.K. Kulp, K.F. Chen, R.W. Brueggemeier, C.L. Shapiro, C.S. Chen, Peroxisome proliferator-activated receptor gamma-independent ablation of cyclin D1 by thiazolidinediones and their derivates in breast cancer cells, Mol. Pharm. 67 (2005) 1342-1348.
[18] A.S. Alva, S.H. Gultekin, E.H. Baehrecke, Autophagy in human tumors: cell survival or death? Cell Death Diff. 11 (2004) 1046-1048.
[19] E.H. Baehrecke, Autophagy: dual roles in life and death, Nat. Rev. Mol. Cell Biol. 6 (2005) 505-510.
[20] S. Han, J. Roman, Rosiglitazone suppresses human lung carcinoma cell growth through PPARgamma-dependent and PPARgamma-independent signal pathways, Mol. Cancer Ther. 5 (2006) 430-437.
[21] H.A. Lane, A. Fernandez, N.J. Lamb, G. Thomas, p70s6k function is essential for G1 progression, Nature 363 (1993) 170-172.
[22] Y. Chen, M.B. Azad, S.B. Gibson, Superoxide is the major reactive oxygen species regulating autophagy, Cell Death Diff. 16 (2009) 1040-1052.
[23] R.R. Nair, M.F. Emmons, A.E. Cress, R.F. Argilagos, K. Lam, W.T. Kerr, H.G. Wang, W.S. Dalton, L.A. Hazlehurst, HYD1-induced
increase in reactive oxygen species leads to autophagy and necrotic cell death in multiple myeloma cells, Mol. Cancer Ther. 8 (2009) 2441-2451.
[24] S. Rodriguez-Enriquez, L. He, J.J. Lemasters, Role of mitochondrial permeability transition pores in mitochondrial autophagy, Int. J. Biochem. Cell Biol. 36 (2004) 2463-2472.
[25] J.M. Perez-Ortiz, P. Tranque, M. Burgos, C.F. Vaquero, J. Llopis, Glitazones induce astroglioma cell death by releasing reactive oxygen species from mitochondria: modulation of cytotoxicity by nitric oxide, Mol. Pharm. 72 (2007) 407-417.
[26] P.J. Cox, D.A. Ryan, F.J. Hollis, A .- M. Harris, A.K. Miller, M. Vousden, H. Cowley, Absorption, disposition, and metabolism of rosiglitazone, a potent thiazolidinedione insulin sensitizer, in human, Drug Metab. Dispos. 28 (2000) 772-780.
[27] L. Masuelli, C. Focaccetti, V. Cereda, F. Lista, D. Vitolo, P. Trono, P. Gallo, A. Amici, P. Monaci, M. Mattei, M. Modesti, G. Forni, M.H. Kraus, R. Muraro, A. Modesti, R. Bei, Gene-specific inhibition of breast carcinoma in BALB-neuT mice by active immunization with rat Neu or human ErbB receptors, Int. J. Oncol. 30 (2007) 381-392.
[28] P. Ferruzzi, E. Ceni, M. Tarocchi, C. Grappone, S. Milani, A. Galli, G. Fiorelli, M. Serio, M. Mannelli, Thiazolidinediones inhibit growth and invasiveness of the human adrenocortical cancer cell line H295R, J.C.E.M. 90 (2005) 1332-1339.
[29] M.J. Betz, I. Shapiro, M. Fassnacht, S. Hahner, M. Reincke, F. Beuschlein, Peroxisome proliferator-activated receptor-gamma agonist suppress adrenocortical tumor cell proliferation and induced differentiation, J.C.E.M. 90 (2005) 3886-3896.
[30] G. Cantini, A. Lombardi, E. Piscitelli, G. Poli, E. Ceni, S. Marchiani, T. Ercolino, A. Galli, M. Serio, M. Mannelli, M. Luconi, Rosiglitazone inhibits adrenocortical cancer cell proliferation by interfering with the IGF-IR intracellular signaling, PPAR Res. 28 (2008) 1-10.
[31] R. Scherz-Shouval, Z. Elazar, ROS, mitochondria and the regulation of autophagy, Trends Cell Biol. 17 (2007) 422-427.
[32] J. Dörrie, H. Gerauer, Y. Wachter, S.J. Zunino, Resveratrol induces extensive apoptosis by depolarizing mitochondrial membranes and activating caspase-9 in acute lymphoblastic leukemia cells, Cancer Res. 61 (2001) 4731-4739.
[33] M. Lu, T. Kwan, C. Yu, F. Chen, B. Freedman, J.M. Schafer, E.J. Lee, J.L. Jameson, V.C. Jordan, V.L. Cryns, Peroxisome proliferator-activated receptor y agonists promote TRAIL-induced apoptosis by reducing survivin levels via cyclin D3 repression and cell cycle arrest, J. Biol. Chem. 280 (2005) 6742-6751.
[34] J.R. Weng, C.Y. Chen, J.J. Pinzone, M.D. Ringel, C.S. Chen, Beyond peroxisosome proliferator-activated receptor y signalling: the multi-facets of the antitumor effect of thiazolidinediones, Endocr. 13 (2006) 401-413.
[35] R. Scherz-Shouval, E. Shvets, Z. Elazar, Oxidation as a post-translational modification that regulates autophagy, Autophagy 3 (2007) 371-373.
[36] R.A. Lockshin, Z. Zakeri, Apoptosis, autophagy, and more, Int. J. Biochem. Cell Biol. 36 (2004) 2405-2419.
[37] J.M. Ye, N. Dzamko, A.J. Hoy, M.A. Iglesias, B. Kemp, E. Kraegen, Rosiglitazone treatment enhances acute AMP-activated protein kinase-mediated muscle and adipose tissue glucose uptake in high-fat-fed rats, Diabetes 55 (2006) 2797-28804.
[38] B. Loos, A.M. Engelbrecht, Cell death a dynamic response concept, Autophagy 5 (2009) 590-603.
[39] E. Morselli, L. Galluzzi, O. Kepp, J.M. Vicencio, A. Criollo, M.C. Maiuri, G. Kroemer, Anti- and pro-tumor functions of autophagy, Biochim. Biophys. Acta 1793 (2009) 1524-1532.
[40] D. Bonofiglio, S. Gabriele, S. Aquila, H. Qi, M. Belmonte, S. Catalano, S. Andò, Peroxisome proliferator-activated receptor (gamma) activates fas ligand gene promoter inducing apoptosis in human breast cancer cells, Breast Cancer Res. Treat. 113 (2009) 423-434.
[41] J. Zhou, W. Zhang, B. Liang, M.C. Casimiro, D. Whitaker-Menezes, M. Wang, M.P. Lisanti, S. Lanza-Jacoby, R.G. Pestell, C. Wang, PPARy activation induces autophagy in breast cancer cells, Int. J. Biochem. Cell Biol. 26 (2009) 1-10.