Adipocyte-specific Gene Expression and Adipogenic Steatosis in the Mouse Liver Due to Peroxisome Proliferator-activated Receptor (cid:1) 1 (PPAR (cid:1) 1) Overexpression*

Peroxisome proliferator activated-receptor (PPAR) isoforms, (cid:2) and (cid:1) , function as important coregulators of energy (lipid) homeostasis. PPAR (cid:2) regulates fatty acid oxidation primarily in liver and to a lesser extent in adipose tissue, whereas PPAR (cid:1) serves as a key regulator of adipocyte differentiation and lipid storage. Of the two PPAR (cid:1) isoforms, PPAR (cid:1) 1 and PPAR (cid:1) 2 generated by al-ternative splicing, PPAR (cid:1) 1 isoform is expressed in liver and other tissues, whereas PPAR (cid:1) 2 isoform is expressed exclusively in adipose tissue where it regulates adipogenesis and lipogenesis. Since the function of PPAR (cid:1) 1 in liver is not clear, we have, in this study, investigated the bio-logical impact of overexpression of PPAR (cid:1) 1 in mouse liver. Adenovirus-PPAR (cid:1) 1 injected into the tail vein induced hepatic steatosis in PPAR (cid:2)

Peroxisome proliferator-activated receptor ␥ (PPAR␥), 1 a member of the nuclear receptor superfamily, is a key regulator of adipogenesis (1)(2)(3)(4). The PPAR subfamily consists of three isotypes, namely PPAR␣, PPAR␥, and PPAR␤/␦, and like all other nuclear receptors, PPARs possess a highly conserved DNA-binding domain that recognizes peroxisome proliferator response elements (PPREs) in the promoter regions of target genes (5)(6)(7)(8)(9). After ligand binding PPARs heterodimerize with retinoid-X-receptors (RXR) and PPAR-RXR heterodimers bind to PPRE to initiate the transcriptional regulation of target genes, in particular those involved in lipid homeostasis (5)(6)(7)(8)(9)(10). The three PPAR isotypes are products of separate genes, and they exhibit distinct patterns of tissue distribution (7). PPAR␣ is expressed at a relatively high concentration in liver, plays a central role in regulating enzymes involved in the oxidation of fatty acids, and is essential for the pleiotropic responses induced in liver by structurally diverse chemicals known as peroxisome proliferators (8 -10). PPAR␥ is present in two isoforms, PPAR␥1 and PPPAR␥2, resulting from alternate promoter usage (1,8,9). PPAR␥2 contains an additional 30 amino acids at the N-terminal end relative to PPAR␥1. PPAR␥2 expression is limited exclusively to adipose tissue where it play a key role in adipogenesis (1,2). On the other hand, PPAR ␥1 is expressed at relatively low levels in many tissues including liver, but the function of this isoform in non-adipose tissue locations is not well delineated (1,9,13). Forced expression of PPAR␥2 or PPAR␥1 can initiate the differentiation of fibroblasts to adipocytes, and in the process the fibroblasts express adipocytespecific genes and accumulate lipid (2,14). Although there is some suggestion that PPAR␥ ligands may induce adipocytespecific gene expression in certain tumor cells, it is uncertain as to whether PPAR␥1 or PPAR␥2 expression in vivo, in nonfibroblast mesenchymal cells, or in epithelial tissues can lead to adipocyte-specific gene expression and adipogenesis (15).
It is well known that CCAAT/enhancer binding family of transcription factors C/EBP␣, -␤, and -␦ have been shown to play an important role in adipogenesis in that high expression of each member of this family will direct fibroblasts to differentiate into adipocytes, and this conversion is mediated through down-stream regulator PPAR␥ and PPAR coactivator PBP/TRAP220/DRIP205 (4, 16 -19). C/EBPs and PBP are expressed in liver, but the significance of this expression in terms of adipogenesis and or lipid accumulation in liver cells remains unclear. The inability of C/EBP and PBP to induce adipogenesis in normal hepatocytes may be due to the fact that downstream regulator PPAR␥ (PPAR␥1 in liver) may be rate-limiting in hepatocytes (19 -21). In this study, we used an adenoviral gene delivery system to overexpress PPAR␥1 in mouse liver to determine whether this would trigger the expression of adipocyte-specific genes and lipid accumulation (steatosis) in liver cells. To avoid the confounding effect of PPAR␣, we used PPAR␣ Ϫ/Ϫ mice (22), and found that overex-pression of PPAR␥1 in these livers leads to adipocyte-specific gene expression and lipid accumulation. We also demonstrate that fatty liver induced by starvation or that developing after feeding a diet deficient in choline is not associated with the induction of genes associated with adipogenesis unlike that accompanying PPAR␥1 overexpression in liver. These results strongly suggest that the low level of PPAR␥1 appears to prevent liver cells from becoming adipocytes despite the prominence of C/EBP␣ gene expression in these cells and that overexpression of PPAR␥1 leads to adipogenic hepatic steatosis.

MATERIALS AND METHODS
Mice and Treatment-Wild type (C57BL/6J) mice and PPAR␣ Ϫ/Ϫ mice (22), 3 to 4 months of age and weighing 25-35 g, were used in this study. PPAR␣ Ϫ/Ϫ mice were maintained on powdered chow with or without troglitazone (0.1% w/w) for 5 days prior to adenovirus injection and killed on day 2, 3, 4, 5, or 6 after injections while still on the same diet. For dose response of Ad/PPAR␥1, mice were maintained on powdered chow, injected with different concentrations of virus, and killed 6 days after injection. For the induction of fatty liver, PPAR␣ Ϫ/Ϫ mice were either fasted for 96 h (23) or fed a choline-deficient diet (Dyets, Bethlehem, PA) for 15 days (24,25). All animal procedures used in this study were reviewed and preapproved by the Institutional Review Boards for Animal Research of the Northwestern University.
Adenoviral Gene Transfer-Construction of recombinant adenovirus containing the mouse PPAR␥1 cDNA (Ad/mPPAR␥1) was as follows. Mouse PPAR␥1 cDNA (8) was cloned into pShuttle-CMV expression vector at SalI site (Quantum Biotechnologies, Inc.). The linearized shuttle vector and AdEasy vector (Quantum Biotechnologies, Inc.) were then co-transformed into Escherichia coli strain BJ5183. Positive recombinant plasmid Ad/mPPAR␥1 was selected. The Ad/mPPAR␥1 virus was then generated as described previously (26). Adenoviral construct of Ad/LacZ was the generous gift of Dr W. El-Deiry (University of Pennsylvania, Philadelphia, PA) and has been described (27). Mice were intravenously injected (tail vein) in a volume of 200 l with 1 ϫϫ 10 11 virus particles of Ad/LacZ or Ad/mPPAR␥1 and killed 6 days later. Mice injected with PBS served as controls in some cases.
Morphology-Tissue fixed in 10% neutral buffered formalin was embedded in paraffin by using standard procedures. Sections (4-m thick) were cut and stained with hematoxylin and eosin. For visualizing ␤-galactosidase activity, sections of liver were incubated in PBS containing 5 mM potassium ferricyanate, 2 mM MgCl 2 , and 1/20 volume of 20 mg/ml 5-bromo-4-chloro-3-indolyl ␤-D-galactoside in dimethylformamide at 37°C (26,27). Immunohistochemical localization of PPAR␥ was performed as described previously using polyclonal anti-PPAR␥ antibodies (Santa Cruz, CA). Frozen sections of formalin-fixed liver (5-m thick) were stained with Oil Red O and counterstained with Giemsa. Histological analysis and image processing were carried out using Leica DMRE microscope equipped with Spot digital camera as described (24, 26 -30).
Northern and Immunoblot Procedures-Total RNA isolated from liver using Trizol reagent (Invitrogen) was glyoxylated, electrophoresed on 0.8% agarose gel, and then transferred to nylon membrane. These nylon membranes were then hybridized at 42°C in 50% formamide hybridization solution using 32 P-labeled cDNA probes. Equal loading was verified by the intensity of methylene blue-stained 18 S and 28 S RNA or by hybridizing the filters for glyceraldehyde-3-phosphate dehydrogenase. For immunoblotting, liver extracts were subjected to 7.5% or 10% SDS-PAGE, transferred to nitrocellulose membranes, and immunoblotted as previously described (26,28,29). The aP2 antibody was a generous gift from Dr. G. S. Hotamisligil (Harvard School of Public Health, Boston, MA).
Microarray Approach-Total RNA was isolated from the livers of PPAR␣ Ϫ/Ϫ mice injected with Ad/LacZ or Ad/mPPAR␥1 and killed 6 days after injection (26). Reverse transcription, second-strand synthesis, and probe labeling were all performed using 10 g of total liver RNA as a template for cDNA synthesis. Biotin-labeled cRNA was produced using the above cDNA as a template, purified, fragmented, and hybridized to U74Av2 arrays (Affymetrix, Santa Clara, CA). After hybridization, bound cRNA was fluorescently labeled using R-phycoerythrinstreptavidin (Molecular Probes), and the fluorescence was intensified by the antibody-amplification method. Data were collected and analyzed using the Affymetrix Microarray suite 4.01 software.

Fatty Liver Resulting from PPAR␥1 Overexpression-We
have investigated the morphological changes in the liver of PPAR␣ Ϫ/Ϫ mice injected with 1 ϫ 10 11 adenoviral-PPAR␥1 particles intravenously. PPAR␥1 overexpression in the liver of PPAR␣ Ϫ/Ϫ mice caused extensive lipid accumulation in hepatocytes located in the periportal and midzonal regions of liver lobules at 6 days (Fig. 1). The accumulation of fat in liver was grossly evident between 3 and 6 days after Ad/PPAR␥1 injection. The degree and zonality of lipid accumulation in liver in mice overexpressing PPAR␥1 appeared essentially similar whether or not they were on troglitazone, the PPAR␥1 ligand ( Fig. 1, B and C). Mice given troglitazone and injected with either PBS or with Ad/LacZ failed to reveal lipid accumulation in hepatocytes (Fig. 1, A and D). Oil red O staining of liver sections obtained from PPAR␣ Ϫ/Ϫ mice infected with Ad/mP-PAR␥1 confirmed hepatic lipid accumulation (Fig. 1E), whereas no appreciable lipid accumulation was evident in the liver of mice treated with Ad/LacZ (Fig. 1F). Immunohistochemical analysis revealed PPAR␥ nuclear staining in ϳ70% of hepatocytes with microvesicular steatosis between 4 and 6 days following Ad/mPPAR␥1 injection into the PPAR␣ Ϫ/Ϫ mice (Fig.  1G). No detectable immunostaining for PPAR␥ was noted in the livers of Ad/LacZ-injected mice (Fig. 1H). As expected, ϳ60 -70% of hepatocytes stained positively for ␤-galactosidase after Ad/LacZ injection ( Fig. 1, I). Wild type (PPAR␣ ϩ/ϩ ) mice dosed with Ad/mPPAR␥1 also exhibited hepatic lipid accumulation but not as severe as that seen in PPAR␣ Ϫ/Ϫ livers with PPAR␥1 overexpression (not illustrated), suggesting that the presence of PPAR␣ in the liver facilitates the up-regulation of fatty acid oxidation systems and reduces lipid accumulation (11,23).
FIG. 3. Expression of aP2 in PPAR␥-induced adipogenic hepatic steatosis but not in other forms of fatty liver. A, immunoblot analysis of aP2 expression in PPAR␣ Ϫ/Ϫ PPAR␣ ϩ/ϩ wild type mouse liver. Mice were injected with PBS, Ad/LacZ, or Ad/PPAR␥1 through tail vein with or without treatment with troglitazone and killed 6 days later. Liver samples from each group were immunoblotted for PPAR␥ and aP2. The non-responsive gene catalase is used as loading control. B, expression of PPAR␥ and aP2 genes in fatty livers. Steatosis that developed in PPAR␣ Ϫ/Ϫ mouse liver after infection with Ad/PPAR␥1 for 6 days (lanes 1 and 2, following fasting for 4 days (lanes 3 and 4), or fed a choline-deficient diet for 15 days (lanes 5 and 6). Representative liver samples were immunoblotted (two mice in each group) for PPAR␥, aP2 (PPAR␥-responsive gene), and catalase (nonresponsive gene). C, Northern blot analysis of RNA obtained from fatty livers that developed in PPAR␣ Ϫ/Ϫ mice after infection with Ad/PPAR␥1 (6 days), starvation for 4 days, or after 2 weeks of feeding a choline-deficient diet. Expression of adipogenesis-associated genes is confined to PPAR␥-induced adipogenic hepatic steatosis and not with other forms of fatty liver.  6.0 Highly similar to sodium-dependent multivitamin transporter expression of PPAR␥2 is restricted mainly to adipocytes and is the key regulator of adipogenesis, although forced expression of PPAR␥1 isoform can also induce adipogenesis in fibroblasts (2,14). Since we noted a striking degree of lipid accumulation in hepatocytes following PPAR␥1 overexpression, it appeared necessary to investigate the adipogenic action of overexpressed PPAR␥1 in liver (Fig. 2). Northern analysis of RNA isolated from PPAR␣ Ϫ/Ϫ mouse liver revealed dramatic induction mRNAs for fat differentiation markers aP2, adipsin and adiponectin (adipoQ/acrp30) (see Refs. 31,32) in PPAR␥1 overexpressing livers but not after Ad/LacZ infection (Fig. 2, A and B).
Since these mice are PPAR␣ Ϫ/Ϫ , it is reasonable to conclude that PPAR␣ plays no role in the observed induction of these adipogenic genes. The mRNA of the adipsin gene reached peak level at 3 days postinjection, whereas aP2 mRNA level was maximally expressed at 6 days after Ad/PPAR␥1 injection ( Fig.  2A). The induction of adiponectin appeared somewhat delayed when compared with aP2 and adipsin. Glucose-6-phophatase (Glc-6-P) mRNA level in liver increased 2 days after Ad/ PPAR␥1 injection ( Fig. 2A). We also determined the hepatic mRNA levels of several genes 6 days after Ad/PPAR␥1 injection to assess the adipogenic and lipogenic profile. Noticeable increases in CD36, glucokinase, malic enzyme, low density lipoprotein receptor, microsomal triglyceride transfer protein, and ⌬9d mRNA levels were detected at 6 days after Ad/PPAR␥ injection (Fig. 2B). We did not observe perceptible increases in the levels of C/EBP␣, SREBP, glucokinase, phospho(enol)pyruvate carboxykinase, and Glut-2 mRNAs in liver (Fig. 2B). We also assessed the mRNA levels of PPAR␣-regulated peroxisomal fatty acid ␤-oxidation system genes and found modest increases in straight chain fatty acyl-CoA oxidase, L-PBE, and peroxisomal 3-ketoacyl-CoA thiolase mRNAs ( Fig. 2A), suggesting that PPAR␥1 at very high levels can transcriptionally activate PPAR␣ target gene expression and regulate fatty acid ␤-oxidation in liver in the absence of PPAR␣. Immunoblotting confirmed the induction of white adipose tissue marker protein aP2 in PPAR␥1-overexpressing liver beginning at day 3 (Fig. 2C). This protein was not detected in normal liver or Ad/LacZ-injected mouse livers (Fig. 3C). Increase in PPAR␥1 and in L-PBE protein levels were also seen in Ad/PPAR␥1-expressing livers (Fig. 2C). As expected, no change in the amount of catalase protein, the peroxisomal marker enzyme, was discerned.

FIG. 5. Northern blot analysis for caveolin-1, CIDE-A, and nur77.
Total RNA isolated from liver of PPAR␣ Ϫ/Ϫ mice at 2, 3, 4, 5, and 6 days after Ad/mPPAR␥1 injection or 6 days after Ad/LacZ injection. Note marked increases in caveolin-1 and CIDE-A mRNAs at 5 and 6 days after Ad/mPPAR␥1 injection. Increases in nur77 mRNA are also evident. aP2 Protein Gene Expression in Liver is PPAR␥-dependent-We observed that the gene expression of aP2 in liver was in a PPAR␥ dose-dependent manner (Fig. 2C). We also established that the induction of aP2 expression in the liver of PPAR␣ Ϫ/Ϫ and wild type (PPAR␣ ϩ/ϩ ) mice is dependent upon PPAR␥1 overexpression whether or not these mice were on troglitazone, the synthetic PPAR␥ ligand (Fig. 3A). Since PPAR␥1 overexpression induced the adipogenic aP2 protein as well as hepatic steatosis in PPAR␣ Ϫ/Ϫ mice, it appeared necessary to study the relationship if any of PPAR␥1 and aP2 gene expression with hepatic steatosis and to distinguish hepatic adipogenesis (hepatic adiposis) from the typical non-alcoholic hepatic steatosis (30). We induced fatty liver in PPAR␣ Ϫ/Ϫ mice (Fig. 4) by either fasting these mice for 96 h (23) or feeding them a diet deficient in choline for 15 days (24). Immunoblotting failed to reveal PPAR␥ and aP2 protein in fatty livers induced by starvation or by a choline-deficient diet, but these two proteins appeared prominent in the fatty liver induced by PPAR␥1 overexpression (Fig. 3B). These observations clearly demonstrate that hepatic fatty change itself is not enough to induce either PPAR␥ or its target gene aP2, adipsin, and adiponectin and that PPAR␥1 overexpression leads to a novel type of adipogenic hepatic steatosis designated hepatic adiposis to distinguish it from the common hepatic steatosis.
Endogenous PPAR␥ Expression-We have ascertained that forced expression of PPAR␥1 does not up-regulate endogenous PPAR␥ (Fig. 6). PPAR␥2 sense primer was designed to include the N-terminal region of PPAR␥2 to distinguish it from endogenous PPAR␥1. For ectopic PPAR␥1 the sense primer was from the cytomegalovirus promoter. The RT-PCR data show no increase in endogenous PPAR␥1 or PPAR␥2 mRNA in the liver 6 days after Ad/PPAR␥1 injection.

DISCUSSION
The nuclear receptor PPAR␥ is essential for adipogenesis and lipid storage (1)(2)(3)(4). PPAR␥ is present in two isoforms, PPAR␥1and PPAR␥2, generated by alternate promoter usage and splicing (9). PPAR␥2 isoform, which is expressed exclusively in adipocytes, plays a pivotal role in adipocyte differentiation and adipocyte-specific gene expression, but there is some controversy as to whether PPAR␥1 isoform can initiate and sustain adipogenic gene expression (2,14,48). Tontonoz et al. (2) reported that both PPAR␥1and PPAR␥2 could stimulate adipogenesis when introduced into fibroblasts, whereas Ren et al. (48), using engineered zinc finger repressors to inhibit the expression of PPAR␥ isoforms, have concluded that PPAR␥1 overexpression exerted no adipogenic effect. Recently, Mueller et al. (14) using PPAR␥ null fibroblasts demonstrated convincingly that both PPAR␥1and PPAR␥2 isoforms have intrinsic ability to induce robust adipogenesis. While most of the work on PPAR␥-induced adipogenesis involved the use of fibroblast cell lines, the key question as to whether PPAR␥1 isoform, which is the only isoform expressed in liver and many other tissues, can stimulate fat cell differentiation or transformation of hepatocytes into adipocytes. In this study, we have shown that overexpression of PPAR␥1 in mouse liver induced adipocyte-specific gene expression as well as microvesicular steatosis in this organ. Furthermore, we have demonstrated that hepatic steatosis resulting from PPAR␥1 overexpression is associated with adipogenic gene expression, whereas hepatic steatosis induced by starvation or developing after feeding a cholinedeficient diet failed to stimulate adipogenic gene expression. Accordingly, we propose that PPAR␥ overexpression leads to the development of a novel form of hepatic steatosis, which is designated adipogenic hepatic steatosis or simply "hepatic adiposis" to distinguish this entity from the common forms of hepatic steatosis.

PPAR␥1 Overexpression and Adipogenic Hepatic Steatosis
Our results showed the expression of adipocyte-specific genes in liver upon overexpression of PPAR␥1 whether or not these mice were treated with troglitazone. This may be due to the availability of putative endogenous ligands in liver generated as part of the normal lipid metabolism (49,50). Because PPAR␥1 is endogenously expressed at a low level in mouse liver it is unlikely that this receptor exerts any adipogenic effect under normal physiological conditions. Furthermore, the relative abundance of PPAR␣ in normal liver serves as a key regulator of fatty acid catabolism thereby minimizing the need for adipogenesis in liver to store lipids (11,30). In that sense functionally active PPAR␣ and fatty acid oxidation systems keep the PPAR␥1 in check (30). Our results clearly establish that the overexpression of PPAR␥1 isoform in liver in PPAR␣ null background triggers the expression of adipogenesis-related genes and fatty change in liver. This process essentially represents conversion of hepatocytes into cells with active adipogenic gene expression profile, and in that regard this represents transformation or transdifferentiation of hepatocyte into adipocytes, i.e. the conversion of fat-burning hepatocytes into fat storage cells (31,32,51). It is important to note that the expression of adipocyte-specific genes and microvesicular steatosis observed in liver is not associated with overexpression of the endogenous PPAR␥ gene. RT-PCR analysis showed no induction of endogenous PPAR␥1 or PPAR␥2 isoforms (Fig. 6). Thus all the changes observed in gene expression patterns and morphological changes are the result of the exogenous expression of adenovirally driven PPAR␥1 overexpression. These results suggest that maintenance of low levels of PPAR␥ in liver is crucial for preventing hepatocytes from encountering an adipogenesis fate. The maintenance of marginal PPAR␥1 gene expression in liver may be achieved by unknown mechanisms designed to prevent positive feedback between C/EBP␣ and PPAR␥ because the downstream adipogenic events could manifest successfully in the presence of abundant PPAR␥ protein (4).
We performed Northern analysis and global transcriptional profiling to define the pattern of genes expressed in liver as a result of PPAR␥1 overexpression. Our studies indicate that overexpression of PPAR␥1 isoform in liver results in the upregulation of many genes known to be up-regulated during adipocyte differentiation of 3T3-L1 fibroblasts (31,32,(52)(53)(54). Adipogenesis genes up-regulated in liver include PPAR␥ target genes adipsin, aP2, adiponectin, caveolin, fat-specific protein 27, and others (Table I). Caveolae and caveolin-1 protein expression are most abundant in adipocytes (39,55). The overexpression of caveolin-1 mRNA in liver expressing PPAR␥ is further indication of adipogenic transformation of hepatocytes in that caveolin-1 appears to participate in facilitating the conversion of triglycerides in lipoprotein form to triglycerides in lipid droplet storage from (39). Caveolin-1 appears functionally necessary to maintain lipid droplet integrity, and the absence of this protein in caveolin-1-deficient mice leads to abnormalities in adipocyte function (39). Increase in caveolin-1 gene expression in PPAR␥1-overexpressing livers is further indication of attaining an adipocyte phenotype. In addition, numerous adipocyte-enriched genes involved in lipogenesis and lipid metabolism, in particular E-FABP (keratinocyte lipid binding protein) (40), cyp4a10, cyp4A14 (25), fasting-induced adipose factor (angiopoietin-like 4) (56, 57), CD36 (58), glycerophosphate dehydrogenase (52), and hormone-sensitive lipase (42), were markedly up-regulated in liver following PPAR␥1 overexpression. Thus, our in vivo observations clearly establish the adipogenic conversion of liver when PPAR␥1 is overexpressed in this organ.
In this study, we noted marked up-regulation of three genes that have recently been shown to participate in apoptosis. These include CIDE-A (44), cyclophilin C (45), and nur77 (46,47). CIDEs belong to a novel family of recently identified cell death-inducing proteins that induce apoptosis when overexpressed (59). CIDE is not normally expressed in liver, but increased levels of CIDE mRNA have been noted in livers of mice that were treated with Wy-14,643, a PPAR␣ ligand (43). However, the CIDE expression was not dependent upon PPAR␣ (43). The functional implications of up-regulation of three apoptotic genes in liver with adipogenic hepatic steatosis suggests that the acquisition of adipogenic phenotype places hepatocytes at greater risk for apoptosis. Additional studies are needed to ascertain if the CIDE, cyclophilin C, and nur77 genes are PPAR␥ targets for regulation. Studies are also needed to ascertain if DNA synthesis is a prerequisite for the adipogenic conversion of hepatocytes. It is of interest that this new form of hepatic adiposis or adipogenic hepatic steatosis resulting from PPAR␥1 overexpression may have potential clinical implications. Individuals with relatively low hepatic levels of PPAR␣ and hyperactive PPAR␥1 resulting from single nucleotide polymorphism or by some other mechanism could develop adipogenic hepatic steatosis similar to that described in this report. It is also worth noting that ob/ob mice exhibit increased levels of PPAR␥ mRNA in their livers (60). Increased expression levels of aP2 and CD36 mRNA were seen in these livers when these mice were given PPAR␥ ligand troglitazone (61). Thus, it is important to consider PPAR␥1 overexpression or hyperactivity as a possible molecular mechanisms responsible for a special type of non-alcoholic hepatic steatosis, designated as hepatic adiposis or adipogenic hepatic steatosis.
Finally, we have identified a number of novel genes as modestly up-regulated in PPAR␥1-overexpressing liver, which remain to be characterized (Table I). This includes a gene we designated as promethin (AY167031), which is induced in liver by PPAR␥ overexpression. Additional studies are needed to further characterize these and other genes to gain appreciation of the role of PPAR␥ in liver in relation to glucose homeostasis, energy utilization, and insulin resistance (61).