Update on Pparγ and Nonalcoholic Fatty Liver Disease

Nonalcoholic fatty liver disease (NAFLD) is the most common initial presentation of obesity and insulin resistance. Uninterrupted progression of hepatic lipid accumulation often leads to fatty liver disease and eventually cirrhosis. Insulin resistance is one of the characteristics of type 2 diabetes. Several types of treatment have been employed against type 2 diabetes some of which ameliorate NAFLD. The frequent line of treatment to improve insulin sensitivity is the use of thiazolidinediones (TZD) which activate the nuclear receptor, peroxisome proliferator activated receptor gamma (Pparγ). Although TZDs are proven to be very effective in promoting insulin sensitivity, its actions on Pparγ have been complicated, specifically on NAFLD. According to studies in different models, Pparγ manifests both beneficial and undesirable effects on NAFLD. This paper will focus on the current knowledge of Pparγ and its effect on NAFLD.


Introduction
Hepatic steatosis without excessive alcohol intake, called nonalcoholic fatty liver disease (NAFLD), is commonly associated with obesity and insulin resistance [1,2]. NAFLD affects the general population and its incidence is linked with the epidemics of obesity and type 2 diabetes [3]. The metabolic pathways leading to hepatic steatosis include enhanced nonesterified fatty acid release from the adipose tissue, increased de novo lipogenesis, decreased β-oxidation and reduced VLDL export [4,5]. Steatosis in the liver is characterized by a large intracytoplasmic fat droplet or welldefined droplets displacing the nucleus to the cell periphery [6]. The accumulation of hepatic lipids could be due to elevated peripheral fatty acids, de novo lipogenesis and defective apolipoprotein biosynthesis [7,8]. The progression of hepatic steatosis often leads to liver inflammation or steatohepatitis and, if unchecked, will worsen liver fibrosis and cirrhosis [9]. Several mouse models are used to elucidate the mechanisms of fatty liver disease [10,11]. The bulk of studies on NAFLD have been through the administration of high fat diets or methionine-choline-deficient diets and the use of genetically leptin-deficient (ob/ob) or leptin-receptordeficient (db/db) mouse models [12]. These models exhibit insulin resistance which is one sequelae of NAFLD [13].
Increased insulin secretion has been directly implicated in the development of fatty liver disease [14]. The therapy used for insulin resistance frequently include the administration of TZDs which are agonists for the nuclear receptor peroxisome proliferator activated receptor gamma (Pparγ) [15]. However, studies have demonstrated that TZDs have also exhibited deleterious side effects that warrant their withdrawal from the market [16]. Of interest, selective Pparγ modulators (SPPARMs) that do not have the undesirable effects of TZDs have been recently identified [17,18]. Since NAFLD often leads to insulin resistance, this paper focuses on the relationship between Pparγ and NAFLD.

Ppar-gamma (Pparγ)
Pparγ is a member of the nuclear receptor superfamily of ligand-activated transcription factors [19] highly expressed in adipocytes [20,21] and plays a role in improving glucose homeostasis and adipocyte differentiation [22]. It increases insulin sensitivity by upregulating the glucose transporter 4 (Glut4) [23]. Pparγ also enhances the transcription factors adipocyte determination and differentiation-dependent factor 1 (Add1) and sterol regulatory element binding protein 1 (Srebp1) which results in the expression of lipogenic genes such as fatty acid synthase (Fas) [24]. Pparγ is expressed as 2 PPAR Research 2 major isoforms, γ1 and γ2, generated from the same gene by alternate promoter usage and RNA splicing [25]. Both isoforms could stimulate adipogenesis when introduced to fibroblasts [26]. Adipocytes treated with the Pparγ ligand, TZD, stimulate expression of uncoupling protein 2 (Ucp2) and therefore increase energy expenditure [27]. Another mechanism of action by TZDs in the adipose tissue is to upregulate the expression of AMP-activated protein kinase (Ampk) which increases fatty acid oxidation while decreasing lipogenesis via downregulation of Srebp-1c and carbohydrate response element binding protein (Chrebp) [28]. The upregulation of hepatic Pparγ is frequently observed in mice fed a high fat diet [29]. In addition, liver specific deletion of Pparγ in mice established its role as a prosteatotic factor in the development of NAFLD [30]. Of importance is that Pparγ activation by TZDs promotes efflux of free fatty acids from the liver and muscle while increasing fat mass which consequently improves insulin sensitivity [31]. Therefore, whether the upregulation of Pparγ causes steatosis or vice versa remains unclear.

The Effect of Pparγ Variants in the Development of NAFLD
Variants in the Pparγ gene found in human genotyping studies have been reported to affect hepatic steatosis. A Japanese cohort was first reported to have a polymorphism in the peroxisome proliferator activator receptor gamma coactivator 1 alpha (Pparγc1a) gene [32].The polymorphism in the T allele of rs2290602 was found in patients with nonalcoholic steatosis which was further confirmed by quantitative real time PCR [32]. In addition, single nucleotide polymorphisms (SNPs) in the C161T genotype in the Pparγ gene found in a Chinese population was associated NAFLD possibly through the adiponectin pathway [33]. Moreover, the Pro12Ala variant in the Pparγ gene was found to be associated with pathogenesis of NAFLD in Indian, Chinese, and North American cohorts but not in German and Italian cohorts [34][35][36][37][38]. Taken together, polymorphisms in the Pparγ gene could be useful to identify individuals that are at high risk for NAFLD but should not be considered as the main factor for the disease.

The Role of Pparγ in the Development of Hepatic Steatosis
The mode of action of Pparγ in liver was suggested to promote insulin sensitivity but with concomitant development of fatty liver. High-fat diet fed mice develop hepatic steatosis and have increased Pparγ expression [39]. This could be due to the suppression cAMP response element binding protein (Creb) levels, the upstream regulator of Pparγ, in high fat diet fed mice [39]. In hepatic overexpression studies, Yu et al. showed that Pparγ1 leads to adipogenic hepatic steatosis [40]. This group employed the Pparα deleted (Pparα-KO) mouse model and then injected the mice with adenovirus overexpressing Pparγ1 [40]. They showed that hepatic overexpression of Pparγ1 induced adipocyte specific gene expression patterns in the livers of Pparα-KO mice [40]. Therefore, they propose that excess Pparγ activity can lead to the development of adipogenic hepatic steatosis [40]. In addition, hepatic adenoviral overexpression of Pparγ2 in lean mice increased liver triglyceride content and induced hypertension [41]. This occurrence was reported to involve the target of Pparγ, fat specific protein 27 (Fsp27) [42] and its actions on the afferent vagal signals in the liver [41]. In a liver specific Pparγ deletion study, Gavrilova et al. reported that the A/ZIP/F-1 mouse model, which develops severe lipoatrophic diabetes, exhibited attenuation of hepatic steatosis but compromised triglyceride clearance [43]. The same group also showed that the liver Pparγ is essential for the effects of a Pparγ agonist, rosiglitazone, to improve glucose metabolism [43]. Moran-Salvador et al. reported that hepatocyte specific deletion of Pparγ in mice protected high fat diet fed mice from accumulation of lipids and, therefore, further implicated its role in the development of hepatic steatosis [30]. They also showed that the Pparγ in Kupffer cells might not be involved in the development of hepatic steatosis [30]. In addition, a mouse model of dyslipidemia showed that hepatic Pparγ2 upregulation induced hepatic de novo lipogenesis [44]. Zhang et al. fed a western-type diet to mice that express the human apolipoprotein B and lack the brown adipose tissue (apoB/BATless) [44]. These mice are obese, insulin resistant and have hepatic steatosis [44]. They showed that hepatic Pparγ2 expression is increased due to elevated rates of lipogenesis via the upregulation of de novo lipogenic genes Fas and acetyl-CoA carboxylase (Acc) [44]. Taken together, these studies strongly implicate Pparγ in the development of hepatic steatosis.

The Role of Pparγ in the Reduction of Hepatic Steatosis
Diet induced hepatic fibrosis mouse models that were either treated with rosiglitazone or administered with adenovirus overexpressing Pparγ were shown to ameliorate hepatic steatosis [45,46]. Mice fed a methionine-cholinedeficient (MCD) diet developed severe hepatic steatosis, inflammation, and fibrosis with downregulation of Pparγ levels [45]. Meanwhile, mice that were fed the same diet supplemented with rosiglitazone were protected from the adverse effects of the MCD diet [45]. The protection from nutritional fibrosing steatohepatitis by Pparγ could be due to the inhibition of hepatic stellate cell activation which is one of the main causes for fibrosis [45]. Similarly, the hepatic adenoviral overexpression of Pparγ in MCD dietfed mice elicited protection from fibrotic steatohepatitis [46]. This could be explained by the genetic upregulation of adiponectin (adipoQ) and hemeoxygenase 1 (Hmox1) and the downregulation of inflammatory markers such as tumor necrosis factor alpha (Tnfα) and interleukin 6 (Il-6) [46]. In a model of hepatic steatosis involving alcohol, mice that were fed ethanol showed amelioration of hepatic steatosis following administration of rosiglitazone due to stimulation of fatty acid oxidation in the liver [47]. In addition, the Long Evans rats, which exhibit moderate obesity and insulin resistance, were given rosiglitazone and subsequently ameliorated hepatic steatosis which could be modulated by Sirtuin 6 (Sirt6) and its target genes Pparγc1a, forkhead box protein O1 (Foxo1), liver kinase B1 (Lkb1) and 5 adenosine monophosphate-activated protein kinase (Ampk) [48]. In vitro experiments, confirmed the function of Sirt6 by using the free fatty acid stimulated mouse hepatocyte cell line, AML 12, which also showed the protection from hepatic steatosis following rosiglitazone treatment [48]. Similarly, Sprague-Dawley rats that were given high sucrose and high fat diet showed amelioration of hepatic steatosis following treatment with rosiglitazone [49]. The decrease in liver triglycerides in these rats could be due to the effect the Pparγ agonist in increasing serum adiponectin and the upregulation of fatty acid oxidation genes, carnitine palmitoyl transferase 1 (Cpt1) and acyl coenzyme A oxidase (Aco) [49]. Furthermore, dietary methionine restriction (MR) in F344 rats upregulated hepatic Pparγ expression, improved insulin sensitivity, and increased fatty acid oxidation [50][51][52]. Overall, these sets of data suggest that Pparγ ameliorated hepatic steatosis due to increased fatty acid oxidation.

The Development of Selective Pparγ Modulators (SPPARMS)
Although Pparγ agonists have direct actions to improve insulin sensitivity, this line of treatment also has undesirable side effects. For example, rosiglitazone was reported to reduce bone mass in mice [53]. In addition, mice that are obese and diabetic develop hepatic steatosis following treatment with TZDs [54]. More recently, a meta-analysis of type 2 diabetes patients showed that pioglitazone is associated with increased risk for urinary bladder cancer [55]. The development of SPPARMS could potentially reduce these negative effects. Modifications in the Pparγ ligands showed direct effects on insulin sensitivity but not on adipogenesis [56,57]. The ligand FMOC-L-Leucine, a chemically distinct ligand for Pparγ, was reported to improve insulin sensitivity but did not affect hepatic lipid metabolism in db/db mice [58]. Telmisartan, an angiotensin receptor blocker that acts as a Pparγ ligand, enhanced insulin sensitivity and decreased body fat in high fat diet fed mice [59]. A synthetic Pparγ ligand, nTZDpa, ameliorated fasting hyperglycemia and hyperinsulinemia and caused decrease in weight gain and adipose tissue size in high fat diet fed mice [60]. In addition, results using a gene expression-based screening identified N-acetylfarnesylcysteine (AFC) as a full and partial agonist of Pparγ [61]. The compound upregulated Pparγ agonist target genes adipose differentiation-related protein (Adrp), angiopoietin-related protein 4 (Angptl4) and adipoq, but was only a partial agonist of adipocyte fatty acid binding protein 2 (ap2) [61]. The AFC also improved glucose homeostasis and reduced adipose tissue inflammation and expansion in diet-induced obese mice [61]. Furthermore, a synthetic Pparγ antagonist, SR-202, decreased expression of Pparγ target genes and promoted insulin sensitivity in diet-induced obese mice as well as in ob/ob mice [62]. Moreover, a partial Pparγ agonist, INT131, was designed to mitigate insulin sensitivity while minimizing the side effects of thiazolidinediones [63]. It was reported that INT131 reduced fasting plasma glucose in humans and also increased insulin sensitivity in db/db and diet induced obesity in mice [64,65]. Taken together, recent data on the use of SPPARMS maintain the effects of Pparγ as an insulin sensitizing agent but with decreased risks for undesirable effects.

Conclusion
With all these data surrounding the effect of Pparγ on the development of NAFLD and improvements in insulin sensitivity in several models, it is still not conclusive as to whether the nuclear receptor is beneficial or detrimental. Therefore, further investigations are necessary to elucidate the effect of specific conformational and structural differences between the nuclear receptor and its ligands. The advent of the development of SPPARMS points to the direction of specifically eliciting the desirable effects of Pparγ activation.

Conflict of Interests
The author reports no conflict of interests.