Obesity lies at the crossroads of various metabolic diseases

Obesity, which is recognized as one of the most significant global health concerns, is a chronic condition of excess body fat (Hill et al. 2003). Because current human living conditions tend to encourage high-calorie food intake with less energy expenditure, the obesity epidemic will show continuous increases in the future (Yach et al. 2006). In addition, obesity contributes to several metabolic diseases, including various cardiovascular diseases, insulin resistance, type 2 diabetes, impaired immunity, and the risk of cancer (Kopelman 2000; Pi-Sunyer 2003). The treatment of obesity has focused on behavioral changes, and education to consume a healthy diet with physical activity can be successful, but it is difficult to maintain a healthy life style. Other strategies such as surgical procedures and medications are limited due to safety, economic reasons, and effectiveness (Fisher and Schauer 2002). During the last few decades, more players in food intake, energy expenditure, and nutrient storage have been elucidated (Turner et al. 2014). With the identification of new molecular targets and better understanding of energy metabolism, new therapies may become available to keep pace with the obesity prevalence in future.

PPARγ plays a central role in obesity, immunity, and cancer

Three distinct peroxisome proliferator-activated receptor types—PPARα (NR1C1), PPARβ/δ (NR1C2), and PPARγ (NR1C3)—have been identified. The most extensively studied PPARγ is highly expressed in adipose tissues and plays a role in glucose and lipid metabolism, whereas PPARα is a major player in the mitochondrial and peroxisomal β-oxidation pathway expressed most abundantly in liver (van Raalte et al. 2004). Ubiquitously expressed PPARβ/δ in hypertension, insulin resistance, and anti-inflammation has been documented (Fredenrich and Grimaldi 2005). Although PPAR family members seem to contain similar functions in inflammation, energy metabolism, and vascular tone, their affinities for lipid metabolites and their expression patterns can be discriminated.

Metabolic PPARγ

At the intersection of obesity, insulin resistance, inflammation, and cardiovascular disease, the nuclear receptor peroxisome proliferator-activated receptor gamma (PPARγ) is a critical metabolic regulator (Tontonoz and Spiegelman 2008) (Fig. 1). PPARγ, a ligand-activated transcription factor, heterodimerizes with RXR (retinoic X receptor) and translocates into the nucleus to regulate downstream target gene expression upon activation by its agonists (Tontonoz and Spiegelman 2008). PPARγ induces the expression of genes involved in lipid synthesis and adipocyte differentiation. These target genes include CD36, fatty acid binding protein 4, adiponectin, and CCAAT/enhancer binding protein α. PPARγ is most famous for its roles in adipogenesis. Defects in PPARγ result in lipodystrophy in mice and humans. Forced expression of PPARγ in fibroblasts can induce adipocyte-like phenotypic changes. PPARγ can incorporate various signaling cascades of WNT, IGF, FGF, TGF-β, and BMP, cell density, and the composition of the extracellular matrix to coordinate adipogenic processes (Hong and Park 2010). PPARγ also orchestrates adipogenesis through its interaction with other transcriptional mediators such as the C/EBP family (α, β, and δ), SREBP, PRDM, PGC-1, ZFP423, REV-ERBα, GATA3, and several miRNAs. Thus, it is believed that PPARγ is the orchestral conductor in adipogenesis through harmonizing the actions of these multiple players (Cristancho and Lazar 2011).

Fig. 1
figure 1

Pleiotropic effects of PPARγ in metabolism, the immune system, and cancer. The lipid metabolic sensor PPARγ conveys the cellular environment to control metabolic gene expression in adipocytes, hepatocytes, the immune systems, and cancer. In these tissues, PPARγ behaves as a key signaling integrator of growth factors, nutrients, and transcriptional cascades to coordinate metabolism, immunity, cell proliferation, and differentiation

Full size image

Investigations of the molecular actions of insulin sensitizing chemical drugs have revealed that PPARγ can play a significant role in systemic glucose homeostasis. Troglitazone a class of thiazolidinediones (TZDs) drug were first identified in animal models as an insulin sensitizer. Subsequent studies demonstrated that PPARγ in adipose tissues unexpectedly acts as the biological receptor for TZDs that include ciglitazone, englitazone, and pioglitazone (Lehmann et al. 1995). These studies were further verified by the identification of mutations in PPARγ associated with insulin resistance and diabetes in both animal models and humans (Tontonoz and Spiegelman 2008). Although TZDs are being questioned due to their adverse effects (Nissen and Wolski 2007; Graham et al. 2010), they have broadened the understanding of PPAR actions in systemic glucose metabolism beyond adipogenesis.

There are at least two different types of adipocyte tissues: white adipose tissue (WAT), which stores excess calories for energy partitioning into peripheral organs under energy-deprived conditions, and brown adipose tissue (BAT), which dissipates heat in response to thermogenic stimuli (Peirce et al. 2014). Recent studies have shown that multilocular brown adipocytes containing more mitochondria with uncoupling protein-1 expression generate energy expensive heat resulting in increased energy expenditure, body weight decrease, and improved insulin sensitivity. Therefore, BAT is considered as health beneficial fats that counteract obesity and metabolism. Most players in white adipogenesis described above, including PPARγ, similarly act to induce brown adipogenesis (Bartelt and Heeren 2014). However, BAT- or WAT-specific players are also described, suggesting the existence of alternate routes for the generation of the two different fats. For example, PRDM16 and PGC-1α play critical roles only for brown adipogenesis, whereas the players RIP140 and TIF2 induce white adipogenesis selectively (Peirce et al. 2014). Further studies also suggest functional links between the immune system and thermogenic adipocytes (Wernstedt Asterholm et al. 2014). Eosinophil and type 2 cytokine signaling in macrophages can translate cold sensing into thermogenesis in adipocytes, leading to body weight decrease (Qiu et al. 2014). Elucidation of the molecular mechanisms that generate the brown-like phenotype in white adipocytes can provide a new defense against obesity and metabolic diseases. Likewise, further dissection of immune and thermogenic adipocytes are intriguing new avenues for targeting obesity and related diseases.

PPARγ also functions in other metabolic tissues. PPARγ is detected at a low level in skeletal muscles compared with that in fat tissues but indirectly affects skeletal muscle by systemic lipid partitioning and secretion of adipokine expression (Tontonoz and Spiegelman 2008). Similarly, PPARγ is expressed in the liver, but its level compared with the expression in adipose tissues is also significantly lower. However, the expression of PPARγ is induced in hepatic fatty liver, indicating the pathophysiological roles of PPARγ in the liver (Gavrilova et al. 2003). TZD treatments in diabetic mice exacerbate hepatic steatosis in a PPARγ-dependent manner (Yu et al. 2003). Thus, overcoming this obstacle can lead to better insulin sensitization under diabetic conditions.

Global PPARγ deletion caused embryonic lethality due to the defects in placenta (Barak et al. 1999). Subsequently, the placental defects rescued PPARγ deleted embryos demonstrated the roles of PPARγ in lipodystrophy and insulin resistance (Barak et al. 1999; Duan et al. 2007). Mice lacking PPARγ in muscle developed insulin resistance (Hevener et al. 2003; Norris et al. 2003). Hepatocyte specific disruption of PPARγ improved fatty liver but decreased insulin sensitivity in diabetic mice (Matsusue et al. 2003). The effects of PPARγ in whole body metabolism and insulin sensitivity can be also attributed by the adipose PPARγ. Adipocyte specific deletion by the adipocyte protein2 (aP2) promoter driven Cre line on a high fat diet showed the roles of adipose PPARγ in fat mass and fatty liver (He et al. 2003; Jones et al. 2005). The critical role of adipose PPARγ in lipid metabolism and insulin resistance were recently verified by the generation of another fat specific knockout model using adiponectin promoter driven Cre (Wang et al. 2013). This fat specific deletion of PPARγ exhibited lipoatrophy, insulin resistance, increased bone mass, and abnormal marry glands, further indicating that the adipose PPARγ is necessary for fat development, fat related tissues, and whole body metabolisms. Therefore, it is reasonable to consider that the PPARγ in various metabolic tissues crosstalk to maintain whole body insulin sensitivity and lipid metabolism.

Immunogenic PPARγ

Obesity is often considered as a low-grade, chronically inflamed state in addition to involving excessive nutrient stores. In obese and insulin-resistant mice, increased macrophage infiltration has been observed in fat tissues (Weisberg et al. 2003; Xu et al. 2003). Enhanced cytokine production from the infiltrated macrophages contributes to insulin resistance (Ferrante 2013). In addition, PPARγ also play a role in macrophage biology and insulin sensitivity. Macrophages can be activated to classically activated inflammatory (M1) macrophages or to more metabolism-friendly alternatively activated (M2) macrophages (Odegaard and Chawla 2011). Infiltrated macrophages in adipose tissues can also lead to alteration of alternatively activated macrophages to classically activated macrophages. Macrophages defective in PPARγ were shown to fail to differentiate into M2 macrophages and are associated with insulin resistance (Odegaard et al. 2007).

Additionally, PPARγ exerts an anti-inflammatory response in murine and human macrophages. Although the mechanism of repression needs to be further clarified, liganded PPARγ, instead of binding to DNA elements, inhibits the inflammatory activities of activator protein-1 (AP-1), signal transducers and activators of transcription 1 (STAT-1), nuclear factor κB (NF-κB), and nuclear factor of activated T cells (NFAT) transcription factor by physical interaction, referred to as transrepression (Pascual et al. 2005). Furthermore, it has been shown that defects in NF-κB signaling in macrophages improve insulin sensitivity (Shoelson et al. 2003). Thus, anti-inflammatory roles of PPARγ in macrophages can participate, at least in part, in insulin-sensitizing actions.

Macrophage PPARγ also plays important roles in atherosclerosis. 9-Hydroxy octadecadienoic acid (9-HODE) and 13-HODE catabolized from atherogenic oxidized LDL cholesterols activate PPARγ and increase lipid uptake through the induction of CD36, a scavenger receptor for LDL, ultimately converting into atherogenic foam cells (Nagy et al. 1998; Tontonoz et al. 1998). It was also reported that PPARγ ligands could be anti-atherogenic by suppressing cytokine release from macrophages (Ricote et al. 1998). This effect was shown to be mediated by the induction of another nuclear receptor liver X receptor (LXR) followed by the increase of its target gene ATP-binding cassette transporter A1 (ABCA1), a reverse cholesterol transporter with reduction of SR-A, a second LDL scavenger receptor in mouse macrophages (Chawla et al. 2001). Nevertheless, PPARγ acts to regulate cholesterol uptake in macrophages, and activation of PPARγ by agonists reduces atherosclerosis in rodents and human patients.

Regarding its anti-inflammatory actions in macrophages, PPARγ also negatively affects the maturation of dendritic cells (DCs) (Szanto and Nagy 2008; Kiss et al. 2013). Activation of PPARγ in DCs decrease cytokine IL-12, chemokine CXCL10 and CCL5, and costimulatory molecules CD80, CD83, and CD40 with induction of the coinhibitory molecule B7H1 in cultured human DCs (Nencioni et al. 2002; Szatmari et al. 2006). These negative effects on the immunostimulatory capacity in DCs by PPARγ activation are shown to be mediated through inhibition of the signaling cascade for the NF-κB family of transcription factors (Appel et al. 2005). Activated DCs downregulate chemokine receptor CCR7 by PPARγ activation, which in turn responds poorly to CCL19 and CCL21 for homing DCs to secondary lymphoid tissues (Nencioni et al. 2002; Hammad et al. 2004; Appel et al. 2005). Together, these data show the roles of PPARγ in innate immune cells to further shape immune responses.

In addition to the roles in the innate immune system, several studies have demonstrated that PPARγ also tunes adaptive immune responses (da Rocha Junior et al. 2013; Kidani and Bensinger 2012). PPARγ negatively regulates T cell survival and activation by inhibiting IL-2 production. PPARγ is also involved in the balance of Th1/Th2 by increasing Th2 cytokine and IL-4 production but decreasing IFN-γ and Th1 cytokine production, although Th2 preference by PPARγ activation can be dependent on the cell type and selective ligand (Tontonoz and Spiegelman 2008; da Rocha Junior et al. 2013). PPARγ activation also impairs Th17 cell development and affects Th17-dependent autoimmune diseases such as experimental autoimmune encephalomyelitis, inflammatory bowel disease, and arthritis (Straus and Glass 2007; Klotz et al. 2009). Regulatory T cells (Treg) are decreased in T-cell-specific PPARγ-deleted mice, suggesting that both the survival of regulatory T cells and their effects on effector CD4+ T cells can be controlled by PPARγ (Hontecillas and Bassaganya-Riera 2007; Guri et al. 2010). Similar to the effects in T cells, PPARγ activation in B cells is anti-proliferative, but the effects are not consistent with those reported in the literature (da Rocha Junior et al. 2013). Given that various autoimmune diseases are associated with the aberrant activation of adaptive immunity, further understanding of PPARγ effects in the context of immune diseases will be required.

Anti-cancerous PPARγ

PPARγ is expressed in various cancer tissues, as well as in metabolic and immune cells (Campbell et al. 2008). Several reports have shown that PPARγ displays anti-tumorigenic functions (Koeffler 2003). Many studies have identified that the activation of PPARγ represses lung, breast, colon and prostate cancer (Bren-Mattison et al. 2008; Bonofiglio et al. 2006; Cesario et al. 2006; Sikka et al. 2012). Christen A. et al. suggested that depletion of PPARγ increases colon cancer in APCmin/+ mice (McAlpine et al. 2006). Corroborating these observations, the survival rate of colorectal cancer patients was found to be increased more under the positive expression of PPARγ (Ogino et al. 2009). Adipocyte-specific knockout PPARγ mice display more susceptibility toward DMBA-induced breast cancer than wild-type mice (Skelhorne-Gross et al. 2012). In prostate cancer, upregulated NcoR1, a PPARγ negative cofactor, demonstrates negative effects on tumorigenesis (Battaglia et al. 2010).

Additionally, agonists of PPARγ regulate the tumorigenesis of various cancers. For example, TZDs have anti-tumorigenic effects on cancer tissues by inducing cell cycle arrest and apoptosis, as well as by suppressing EMT. One PPARγ agonist, troglitazone, suppresses pancreatic carcinoma cell growth through the increased p27 protein levels (Motomura et al. 2000). The proliferation of the breast cancer cell line MCF-7 was blocked by troglitazone, resulting in the reduction of Rb phosphorylation and leading to decreased CDK2 and -4 activities (Yin et al. 2001). Another TZD, ciglitazone, can also increase p27 protein levels, leading to the suppression of cell proliferation (Chen and Harrison 2005).

Several studies have shown that activation of PPARγ by its ligands could modulate apoptosis via various signaling mechanisms. For example, TZD, upon binding PPARγ, can inhibit post-translational modification such as phosphorylation of β-catenin, ultimately suppressing Wnt pathways and inducing cell death (Sharma et al. 2004; Wei et al. 2007). Activation of PPARγ by rosiglitazone extensively increases PTEN levels, suppressing growth factor-dependent AKT activation (Farrow and Evers 2003). PPARγ can also directly regulate phosphorylation of AKT and PI3K activities, inducing apoptosis (Kim et al. 2006; Yan et al. 2010). Furthermore, PPARγ activation is able to inhibit BCL family function, accelerate the turnover rate of FLIP, and increase the expression of BAX and BAD, resulting in apoptosis (Shiau et al. 2005; Kim et al. 2002; Zander et al. 2002; Bae and Song 2003). Finally, many cancers with a related inflammation mechanism are also regulated by PPARγ activation. PPARγ agonists suppress the expression of pro-inflammatory proteins, IL-6, TNF, and MCP1 (McKinnon et al. 2012; Hwa et al. 2011; Wang et al. 2011; Nguyen et al. 2012). Agonists also contribute to PPARγ sumoylation, which negatively interfere with the activation of NF-κB activity, leading to cell death (Ramkalawan et al. 2012). On the other hand, some of the studies report that patients of diabetes mellitus treated with TZD or rosiglitazone displayed a significant positive correlation with a diagnosis of cancer (Ramos-Nino et al. 2007). TZD treatment also accelerated polyp formation in the colon of APCMin mice (Saez et al. 1998). However, a population-based cohort study in the patients with diabetes mellitus treated with TZD displayed reduction in lung cancer risk but not in colorectal and prostate cancer (Govindarajan et al. 2007). It is not clear at the moment whether the systemic activation of PPARγ using its activators could reduce the risk of cancer development. Thus, the effects of PPARγ agonists on relieving the symptom of diabetes mellitus and controlling the risk of cancer are still to be pursued. In summary, these data suggest that PPARγ activation by its agonist could have repressive roles in tumor formation by inducing cell cycle arrest, accelerating apoptosis and suppressing the production of pro-inflammatory proteins. However, its systemic effects on risk of cancer in patients with diabetes mellitus will require further investigations.

Miscellaneous PPARγ

Bone homeostasis is controlled by bone formation by osteoblasts and resorption by osteoclasts (Sims and Gooi 2008; Manolagas 1998). PPARγ inhibits the generation of bone-forming osteoblasts at the expense of adipocytes from common precursor cells. PPARγ silencing in bone marrow skews toward osteoblasts with significant reduction of osteoclasts and adipocytes (Park et al. 2008a). Conversely, aging and PPARγ activation increase lipid droplet formation and osteoclast activity and reduce osteoblast formation in bone marrow cells (Moerman et al. 2004). A number of signaling events, including the Wnt and hedgehog pathways, control the inverse balance between osteogenic and adipogenic differentiation (Park et al. 2008a). The reciprocal decision of adipogenic or osteogenic fate was further elucidated by a high bone mass resulting from the loss of bone marrow fat in adipose-specific PPARγ knockout mice (Wang et al. 2013). PPARγ is also involved in neuronal death, differentiation, and axon polarity (Quintanilla et al. 2014). Taken together, the data show that PPARγ, a lipid metabolic sensor, exerts its functions through survival, proliferation, and differentiation and is a regulator of metabolic homeostasis in various tissues (Fig. 1).

PPARγ as a therapeutic target

Drugs of the TZD class are PPARγ agonists that have been therapeutically proven to be useful in the treatment of type 2 diabetes (Lehmann et al. 1995). Genetic evidence, including gain-of function and loss-of-function experiments, showed that PPARγ in adipose tissues is the primary target for the glucose-lowering effects of TZDs (Tontonoz and Spiegelman 2008). TZDs activate a set of genes to increase lipid synthesis and lipid uptake in adipocytes. Thus, activation of PPARγ is thought to repartition insulin-resistant lipids from insulin-acting tissues through the induction of lipid flux in adipocytes, which in turn improves insulin sensitivity in the liver and muscles (Waki and Tontonoz 2007). Unfortunately, the use of TZD is questioned by unfavorable side effects such as weight gain, fluid retention, edema, and a higher risk of congestive heart failure (Nesto et al. 2004; Nissen and Wolski 2007). Therefore, the limitation of TZD requires other therapeutic approaches to treat type 2 diabetes.

To circumvent the issue of TZDs, significant efforts are being made for differential PPARγ activation. To separate the beneficial metabolic effects from the unwanted side effects, highly potent and selective PPARα modulators (SPPARMα) and PPARγ modulators (SPPARMγ) are being investigated (Balint and Nagy 2006; DePaoli et al. 2014; Feldman et al. 2008). Dual PPAR-α/γ agonists were also studied and have displayed some promising metabolic effects. Partial agonists, pan agonists, and antagonists are also being developed for the generation of safe and effective drugs. Similarly, significant efforts have led to the identification of SPPARMγ, and partial or dual agonists from herbal natural products (Wang et al. 2014a). These selective compounds have already shown improvement regarding metabolic parameters in diabetic animals with few side effects compared with TZD treatments.

Since the identification of PPARγ as a biological receptor of anti-diabetic TZDs, PPARγ regulation has been actively investigated. Recently, newly identified small-molecule inducers of PPARγ have proposed alternative PPARγ activation for the anti-diabetic effects. Harmine appears to act by inducing PPARγ mRNA expression in preadipocytes. Chronic treatment with harmine in diabetic mice lowered serum glucose levels and improved insulin sensitivity without the detrimental effects of weight gain and liver failure (Waki et al. 2007). Further mechanistic studies showed that harmine induced adipocyte differentiation through acting on the Wnt signaling pathway (Park et al. 2008b). Phenamil was also reported to induce PPARγ expression and adipocyte differentiation (Park et al. 2010). Thus, chemicals that regulate PPARγ expression can represent novel approaches without the side effects as shown in TZD treatments.

PPARγ activation through post-translational modifications of PPARγ can also be viewed as new modulators for PPARγ activity (Hauser et al. 2000; van Beekum et al. 2009). Indeed, recent studies have shown that the phosphorylation of PPARγ by CDK5 lowers glucose levels, and non-TZD drugs post translationally modify PPARγ through acting on CDK5 (Choi et al. 2010; 2011). In addition, PPARγ can be ubiquitinated and sumoylated (Kilroy et al. 2009; Floyd and Stephens 2004). Sumoylation of PPARγ seems to mediate anti-inflammatory activities by repressing inflammatory gene expression in macrophages (Pascual et al. 2005). The ubiquitin ligase Siah2 affects PPARγ protein stability and activity (Kilroy et al. 2012). Similarly, MKRN1 affects PPARγ stability through direct association (Kim et al. 2014). These studies bring new insights into the improvement of insulin sensitivity by managing PPARγ post-translational modification. Furthermore, newly developed PPARγ modulators at the transcriptional and post-translational levels can be used in cancer and immune diseases.

PPARγ are expressed in various cancer tissues, including the breast, colon, lung, prostate and bladder (Campbell et al. 2008). Many investigations are now aimed at determining whether PPARγ and its ligands could be used as therapeutics for cancer treatment. Jeong et al. showed that PPARγ-positive lung cancer cell lines are more sensitive to the anti-cancerous effects of its ligands, implicating its possible usage for cancer therapy (Jeong et al. 2012). In a phase 2 clinical trial, troglitazone showed that it could inhibit the growth of prostate cancer cells (Mueller et al. 2000). Troglitazone also had a negative effect on renal cell carcinoma by inducing p38 MAPK-mediated-cell cycle arrest (Fujita et al. 2011). Rosiglitazone, in a phase II trial, raised the radioiodine uptake in thyroid cancer patients (Kebebew et al. 2006). Recently, many trials have been conducted regarding the combination treatment of PPARγ ligands with other putative cancer drugs such as RXRα agonists, chemotherapeutic agents, statins, or some of cell-to-cell signaling molecules to synergistically cure cancers (Skelhorne-Gross and Nicol 2012). For example, IFN-β-treated pancreatic cancer cells were more affected by co-treatment with troglitazone, which represses the NF-κB-related survival pathway (Vitale et al. 2012). A PPARγ agonist, DIM-C, makes bladder cancer cells more susceptible to EGFR inhibition (Mansure et al. 2013). The combination treatment of troglitazone and the RXR agonist RA inhibits the proliferation of gastric cancer cells by increasing apoptosis (Liu et al. 2013). These results indicate that targeting PPARγ alone or with combination treatments could be beneficial for curing cancer in the future.

Future directions

In this review, the pleiotropic effects of PPARγ in multiple cell types were discussed (Fig. 1). PPARγ is one of the most important lipid sensors that plays a communicative role at the crossroads between metabolism and immunity or cancer. An increasing number of publications have addressed the roles of PPARγ in various settings. However, it is clear that several issues, as noted below, remain to be addressed.

As described above, TZDs are the principal class of drugs available for the improvement of insulin sensitivity in obese diabetic subjects. However, from a therapeutic standpoint of diabetes, treatment with TZD is hampered by many side effects. Thus, the limitation of TZDs requires other therapeutic approaches. To develop alternatives, identification of novel chemicals followed by structural optimization and assignments of new molecular targets need to be performed. New pharmacological targets can be elucidated by dissecting the molecular mechanisms of bioactive small molecules. Studies on resveratrol, an anti-obese and anti-oxidant bioactive compound, have identified the novel targets Sirtuin 1, AMP-activated protein kinase, and phosphodiesterase 4 (Picard et al. 2004; Park et al. 2012). The anti-diabetic small molecule harmine has led to the identification of Id2 as a new factor in adipogenesis (Park et al. 2008b). Similarly, phytochemicals from various herbal products can also be used for the elucidation of new metabolic players (Wang et al. 2014b). Therefore, such studies have revealed potential chemicals for the treatment of insulin resistance, have expanded our knowledge of certain unidentified pathways in biology, and have offered new molecular targets for insulin resistance. Based on these studies, efforts to screen for novel chemical regulators, along with the elucidation of their molecular pathways, should be investigated more extensively in the future.

PPARγ agonists exert anti-inflammatory effects in macrophages. Transrepression is primarily responsible for the anti-inflammatory actions of PPARγ activation. However, the detailed mechanistic explanation regarding how PPARγ mediates anti-inflammatory effects in macrophages remains elusive. Transrepression, as well as the related PPARγ-dependent and -independent effects, in innate immune systems need to be further clarified.

Given the complexity of adaptive immunity, it is not surprising that activation of PPARγ in subsets of T cells and B cells produces inconsistent results in proliferation and activation. Pharmacological interventions targeted at a single cell type or molecule are thus prone to be ineffective for autoimmune diseases. Regarding the alteration of immune responses toward the genesis of a specific subset of immune cells such as Treg or Th17 through the modulation of metabolic sensors, PPARγ can be an alternative or can be used in new combinatory strategies.

Concerning cancer therapy, using of PPARγ has numerous merits. However, some reports have suggested that PPARγ or its agonists could accelerate tumorigenesis in colon and bladder cancer. For example, in APCmin/+ mice, troglitazone-mediated PPARγ activation could increase polyp formation in colon cancer (Saez et al. 1998). Additionally, rosiglitazone displayed more eruption of urinary bladder cancer induced by OH-BBN (Lubet et al. 2008). Based on these observations, the roles of PPARγ in various physiological contexts should be pursued extensively. One other problem in cancer therapy targeting PPARγ is the side effects caused by its ligands such as heart failure, bone fractures, and obesity (Erdmann et al. 2009; Govindarajan et al. 2007). To circumvent these problems, putative therapeutic drugs with fewer side effects are actively being investigated and pursued worldwide. Several recent reports have highlighted the new development of non-TZD chemicals and natural extracts developed for cancer therapy.

In conclusion, it is obvious that the metabolic sensor PPARγ interprets information from various cellular environments of available growth factors, lipid metabolites, and synthetic ligands to control metabolic gene expression in metabolic, immune, and cancer cells. Figure 1 summarizes the functions of PPARγ and further delineates PPARγ as a key integrator of cellular signals in various tissues. Due to the multifaceted roles of PPARγ, complicated and inconsistent biological actions of PPARγ in different settings are somewhat expected. Once we clearly understand the cellular and molecular complexity of PPARγ, the context-dependent pleiotropic roles of PPARγ will be better resolved. Subsequently, more defined strategies targeting PPARγ in clinical trials that explore possible treatments using PPARγ for various associated human diseases will be proposed in the future.