Pharmacological Reports. 2016 Feb;68(1):144-54. doi: 10.1016/j.pharep.2015.08.001. Epub 2015 Aug 20.

Pioglitazone, an anti-diabetic drug requires sustained MAPK activation for its anti-tumor activity in MCF7 breast cancer cells, independent of PPAR-γ pathway


Labanyamoy Kole, Mrinmoy Sarkar, Anwesha Deb, Biplab Giri

Experimental Medicine & Stem Cell Research Laboratory, Department of Physiology, West Bengal State University, Kolkata 700126, India.

Discovery Research SBU, Dr. Reddy’s Laboratories Ltd., Hyderabad, India.



Background: The thiazolidinedione (TZD) class of peroxisome proliferator-activated receptor gamma (PPAR-g) ligands are known for their ability to induce adipocyte differentiation, to increase insulin sensitivity including anticancer properties. But, whether or not upstream events like MAPK activation or PPAR-g signaling are involved or associated with this anticancer activity is not well understood in breast cancer cells. The role of MAPK and PPAR pathways during the pioglitazone (Pio) induced PPAR-g independent anticancer activity in MCF7 cells has been focused here.

Methods: The anticancer activity of Pio has been investigated in breast cancer cells in vitro. Anti-tumor effects were assessed by alamar blue assay, Western blot analysis, cell cycle analysis, and annexin V-FITC/PI binding assay by flow cytometry, Hoechst staining and luciferase assay.

Results: The anticancer activity of Pio is found to be correlating with the up regulation of CDKIs (p21/p27) and down regulation of CDK-4. This study demonstrates that the induction of CDKIs by Pio is due to the sustained activation of MAPK. The Pio-mediated activation of MAPK is transmitted to activate ELK-1 and the related anti-proliferation is blocked by MEK inhibitor (PD-184352).

Conclusions: Pio suppresses the proliferation of MCF7 cells, at least partly by a PPAR-g-independent mechanism involving the induction of p21 which in turn requires sustained activation of MAPK. These findings implicate the utility of Pio in the treatment of PPAR positive or negative human cancers and the development of a new class of compounds to enhance the effectiveness of Pio.



Findings till date: Peroxisome Proliferator Activated Receptor (PPAR) is a family of nuclear receptors with known subtypes- α, β and γ. In-vivo, these receptors bind an array of biological ligands such as steroids, thyroid hormone, retinoic acid, vitamin-D etc. and are responsible for initiating different downstream processes within different kinds of cells in our body. Under basal conditions, PPAR-γ is found in an inactive form, bound by its co-repressors. When a ligand comes in its vicinity, PPAR-γ binds with it, undergoes configurational modifications, releases its co-repressors and in turn, binds with its co-activators. Active PPAR-γ forms a hetero-dimer with another nuclear receptor called Retinoid X Receptor-α (RXR-α) and this complex translocates inside the nuclear membrane, to bind with promoters of the genes whose expression they control. The result is varied like cell cycle arrest, apoptosis and DNA repair response.

In previous studies, it had been observed that a mutated PPARG gene is associated with human colon cancer and an oncogengic fusion protein PAPX8-PPAR-γ is associated with thyroid follicular carcinoma [1]. Everything put together, indicate towards the vital role of PPAR-γ as a key player in tumorigenesis (PPAR-γ positive cancers). In contrary, studies in MCF-7 breast cancer cells and HT-29 colon cancer cells, suggested that there is no mutation in PPAR-γ gene, though there is an increased availability of PPAR-γ due to the presence of caveolin-1 protein [2]. Another study showed beneficial therapeutic response of enhanced PPAR-γ production in case of non-small cell lung cancer [3]. Hence, it could be concluded that in case of breast cancer, no mutation or modulation of expression (up or down) of PPAR-γ was responsible for tumorigenesis (PPAR-γ negative cancers) .

As mentioned before, there are several ligands for PPAR-γ including the TZD class of insulin sensitizing drugs such as Troglitazone (Tro), Rosiglitazone (Rosi) and Pioglitazone (Pio) as well as particular non-steroidal anti-inflammatory drugs. In muscles and adipose tissue, these compounds bring about decreased insulin resistance via activation of PPAR-γ, which in turn lead to increased production of proteins which are responsible for uptake of free glucose present in circulation. In the liver, they cause increased insulin sensitivity and thereby create a drop in further glucose production in the body [4]. That is the reason why these drugs are popularly prescribed for the treatment of diabetes mellitus and have also proved themselves valuable for vascular and atherogenic complications. There are existing studies to highlight that PPAR-γ has positive roles in adipocyte differentiation and also, cell cycle arrest and terminal differentiation in case of liposarcomas and metastatic breast adenocarcinomas [5]. Thus in case of cancer, TZDs can activate PPAR-γ receptor to inhibit cancer cell migration and angiogenesis, thereby killing them.

MAPK/ERK signalling cascades mediate cell survival and proliferation signals. In different types of cells, a variety of regulatory mechanisms involve phosphorylation of PPAR-γ by MAPKs and this forms the heart of such signalling mechanisms. Milieu of proteins like ERK, JNK and p38 phosphorylate PPAR-γ within a MAPK motif, thereby inactivating it in such a way that there is lowered basal and ligand based transactivation through PPAR-γ [6]. Therefore, when agonists of MAPK/ERK pathway activators are charged, there is inhibition of differentiation function of PPAR-γ due to its phosphorylation.

By another mechanism, PPAR-γ interacts with other transcription factors at the DNA level, leading to PPRE independent genomic actions of PPAR-γ and its ligands. When ERK pathway is activated, it phosphorylates the latter transcription factors which were destined to interact with PPAR-γ and thus inhibit their action. There is also a third pathway which commences at the stage of nuclear export. A MAPK pathway intermediate called MEK1 interacts with PPAR-γ, resulting in off-DNA interaction of PPAR-γ with distinct protein partners such as cytoskeleton, lipid droplets, kinases etc., in turn leading to cytoplasmic signalling [7]. In earlier studies, it has been noted that PPAR-γ agonists such as Pio and Tro rapidly activate MAPKK/ERK pathway. Such activation of ERK is associated with several phenomenons such as cell proliferation, differentiation, apoptosis and cell cycle arrest depending on the availability and intensity of downstream target proteins. The magnitude and duration (sustained or transient) of ERK 1/2 activation, partially determines the cell’s response to extracellular stimuli. Stimulation of MEK/ERK pathway brings about cell cycle arrest in G1 phase. These events are associated with ERK dependent induction of p21 (CDK2 inhibitor) in fibroblasts, hepatocytes and PC-12 cells [8]. Such p53 dependent/independent activation of p21 is due to activation of ERK and proved strongly because MEK inhibitor blocks induction of p21 in Raf-1 over expressed cells [9].

When it comes to discussing the probability of crosstalk between PPAR-γ and MAPK pathway, it has already been demonstrated that PPAR-γ ligands such as Pio, Tro and Rosi can function via intracellular signalling such as ERK cascade, by a PPAR-γ independent mechanism which is derived from an exogenous application of ligands that bind to receptors on plasma membrane [10].

TZD class of anti-diabetic drugs have the ability to induce CDKI (p21) expression in many cancer cells but whether or not upstream events like MAPK activation or PPAR-γ signalling are involved or associated with this anticancer activity is not well understood [11]. Therefore, our present efforts are concentrated on the role of MAPK and PPAR-γ pathways during Pio induced anti-cancer activity in MCF-7 cells.

Findings from this study: Post activation of MAPK via diverse mitotic and non-mitotic signals, its role in downstream events of cell differentiation, apoptosis and anti-proliferation is not well understood till now. In previous investigations, it was reported that with the treatment of Tro (a PPAR-γ ligand) in non-small cell lung cancer (A549) cells, there was sustained stimulation of ERK1/2 and further activation of differentiation. Through the series of experiments undertaken in this current study, we could show that the anti-proliferative action of Pio on MCF-7 cells required sustained induction of MAPK. With increasing concentrations of Pio, more and more cells were seen to withdraw from the active cell cycle and this could be verified by Anexin-V binding assay and cell morphology studies via Hoechst staining. This cell death could also be prevented if the activation of ERK could be inhibited by the use of a MEK inhibitor (PD184352).

In the presence of PMA (phorbol 12-myristate 13-acetate), there occurs prolonged phosphorylation of MAPK and subsequent macrophage-like differentiation in human myeloid leukemia cells and cisplatin induced apoptosis. Depending on the cell type and kind of stimulus, the ERK pathway led to a variety of anti-proliferative symptoms like apoptosis, autophagy and senescence. All this is contrary to the popular age-old belief that ERK is pro-survival in function.

In an earlier study conducted by Roovers and Assoian, it was shown that continued activation of ERK1/2 is often associated with cell differentiation and/or growth arrest in a cell and tissue specific manner. The MEK/ERK pathway could induce p21 in an ERK dependent manner and consequently bring about cell cycle arrest. Keeping this hypothesis in mind, we delved into the status of ERK1/2 phosphorylation in the same Pio treated sample as that used for induction of p21. From our investigation, it was found that Pio is a potent inducer of ERK1/2 phosphorylation in both breast and colon cancer cell lines and that phosphorylation was dose dependent in nature. This could also be inhibited dose dependently by the MEK inhibitor PD at 10 and 100nM concentrations. The dose dependent inhibition of proliferation of Pio was halted when PD was charged at 10nM (1/20 of IC50), along with different concentrations of Pio. It was observed that the activation of ERK was important for the induction of Pio mediated anti-proliferation in breast cancer cells. Once phosphorylated, ERK remained in this condition for 6 to 72 hours. Down regulation of ERK activity could also be accomplished by the use of MEK inhibitors and an inhibition of anti-proliferation could be achieved. The use of MEK inhibitor PD184352 is significant because it selectively blocks MEK1/2 only and does not influence MEK5 which is known to promote apoptosis in meduloblastoma cells. Pio mediated induction of ERK phosphorylation induces CDKIs (p21 and p27). A popularly known mitogen and inducer of MAPK called TPA (tumor-promoting phorbol ester) was observed to have an opposite impact on MCF-7 cells, although MAPK activation was present. Similar activation of MAPK was also observed after MCF-7 cells were treated with Pio. Thus, it could be inferred that MAPK activation plays a vital role in growth inhibition of MCF-7 cells.

When EGF or PDGF ligands activate MAPK, it is seen that it decreases transcription by PPAR-γ by phosphorylating it at Ser 82, thereby inhibiting adipocyte and insulin sensitivity in 3T3-Li-adipocytes [12]. From one study, it was found that EGF terminated any functional implication of PPAR-γ in cell death and in another; it was revealed that it was not absolutely critical to the phosphorylation of ERK1/2 [13]. Thus, the use of EGF to enhance signal transmission of ELK-1 was justified and these findings were in line with our assumptions and observations in this study. EGF enhanced Pio mediated ELK-1 trans-activation and this could be blocked by PD within 6 hours. However, after 48 hours, luciferase activity was seen to drop because of cell death ushered by long hours of incubation after transfection using lipofectamine. The transfected cells thus obtained were used for EGF study for duration of 48 hours.

Simultaneously, another study was conducted wherein MCF-7 cells were charged with Pio and PD at various concentrations. Here, it was seen that though Pio dose dependently induced PPAR-γ expression, there were no alterations in its level even after co-treatment with PD up to 100nM alongside Pio at 10µM. This was a clear indication that anti-proliferative effect of Pio was PPAR independent. At higher concentrations of Pio, PPRE activation did not increase much and the activation of MAPK by TGF-α potentiated PPRE trans-activation. Therefore, these data do not fall in line with the conventional knowledge of PPAR mediated action in 3T3-Li cell line.

Binding affinity studies of Pio with purified PPAR-γ showed that it was much weaker when compared to other PPAR-γ ligands like Rosi. All these findings point towards the involvement of a non-PPAR pathway in the resultant anti-proliferation by TZDs. To support this, Palakurthi et al. (2001) showed identical anti-proliferation in PPAR+/+ and PPAR -/- mouse ES-cell by TZDs [14]. Consistent with these findings, we have also provided evidence for Pio mediated anti-proliferation of MCF7 breast cancer cells through sustained ERK phosphorylation in a PPAR independent manner (Fig.1).

Altogether, the results obtained successfully show that TZDs could be potentially used for the treatment of PPAR positive and negative human cancers in future. This study also highlights new therapeutic applications of Pio and provides evidence that ERK activity could be useful in predicting the type of tumor, which in turn would help in optimizing treatment using TZDs.




Figure 1: The figure depicts the conceptual part of our study which demonstrates that the ERK 1/2 phosphorylation (part of MAPK pathway) is capable to inhibit the cellular proliferation and induce apoptogenic activities in MCF7 breast cancer cell line by the treatment of PPARg agonist Pioglitazone, irrespective of the regulation of PPARg itself. This has been observed to be accomplished by inducing cell cycle arrest at G0/G1 phase of the cell cycle and starting apoptogenic activities in MCF7 breast cancer cells. Higher intracellular expression of cell cycle inhibitory protein p21 and suppressed expression of CDK4 has been seen to be associated with the relevant cancer cell growth inhibition and death observed in the study.



[1] Raman P, Koenig RJ. Pax-8-PPAR-γ fusion protein in thyroid carcinoma. Nat Rev Endocrinol. 2014 Oct;10(10):616-23. doi: 10.1038/nrendo.2014.115. Epub 2014 Jul 29. Review. PubMed PMID: 25069464; PubMed Central PMCID: PMC4290886.

[2] Burgermeister E, Tencer L, Liscovitch M. Peroxisome proliferator-activated receptor-gamma upregulates caveolin-1 and caveolin-2 expression in human carcinoma cells. Oncogene 2003;22:3888–900.

[3] Keshamouni VG, Reddy RC, Arenberg DA, Joel B, Thannickal VJ, Kalemkerian GP, et al. Peroxisome proliferator-activated receptor-g activation inhibits tumor progression in non-small-cell lung cancer. Oncogene 2004;23(1):100–8.

[4] Lowel BB. PPARgamma: an essential regulator of adipogenesis and modulator of fat cell function. Cell 1999;99:239–42.

[5] Chang TH, Szabo E. Induction of differentiation and apoptosis by ligands of PPARg in non-small cell lung cancer. Cancer Res 2000;60:1129–38.

[6] Camp HS, Tafuri SR. Regulation of peroxisome proliferator-activated receptor gamma activity by mitogen-activated protein kinase. J Biol Chem 1997;272:10811–16.

[7] Burgermeister E, Seger R. PPARg and MEK interactions in cancer. PPAR Res 2008;2008:309469.

[8] Liu Y, Martindale JL, Gorospe M, Holbrook NJ. Regulation of p21WAF1/CIP1 expression through MAPK signaling pathway. Cancer Res 1996;56:31–5.

[9] Pumiglia KM, Decker SJ. Cell cycle arrest mediated by the MEK/MAPK pathway. Proc Natl Acad Sci USA 1997;94:448–52.

[10] Papageorgiou E, Pitulis N, Msaouel P, Lembessis P, Koutsilieris M. The nongenomic crosstalk between PPAR-g ligands and ERK1/2 in cancer cell lines. Expert Opin Ther Targets 2007;11(8):1071–85.

[11] Yin F, Wakino S, Liu Z, Kim S, Hsueh WA, Collins AR, et al. Troglitazone inhibits growth of MCF-7 breast carcinoma cells by targeting G1 cell cycle regulators. Biochem Biophys Res Commun 2001;286(5):916–22.

[12] Camp HS, Tafuri SR. Regulation of peroxisome proliferator-activated receptor gamma activity by mitogen-activated protein kinase. J Biol Chem 1997;272: 10811–16.

[13] Chbicheb S, Yao X, Rodeau JL, Salamone S, Boisbrun M, Thiel G, et al. EGR1 expression: a calcium and ERK1/2 mediated PPARg-independent event involved in the antiproliferative effect of 15-deoxy-Iˆ0012,14-prostaglandin J2 and thiazolidinediones in breast cancer cells. Biochem Pharmacol 2011;81:1087–97.

[14] Palakurthi SS, Aktas H, Grubissich LM, Mortensen RM, Halperin JA. Anticancer effects of thiazolidinediones are independent of PPARg and mediated by inhibition of translation initiation. Cancer Res 2001;61:6213–8.



This work is partially supported by the grant of B Giri from SERB, Department of Science and Technology, Government of India (Grant No. SR/FT/LS-132/2010) and Dr. Reddy’s Laboratories Ltd., Hyderabad.



Biplab Giri, Ph.D.

Experimental Medicine & Stem Cell Research Laboratory,

Department of Physiology,

West Bengal State University,

Kolkata 700126, India.



Present address of Labanyamoy Kole, Ph.D.:

Premas Biotech

Manesar, Gurgaon,

Haryana 122050, India.



Multiselect Ultimate Query Plugin by InoPlugs Web Design Vienna | Webdesign Wien and Juwelier SchönmannMultiselect Ultimate Query Plugin by InoPlugs Web Design Vienna | Webdesign Wien and Juwelier Schönmann