Dual regulation of adipose triglyceride lipase by pigment epithelium-derived factor: A novel mechanistic insight into progressive obesity

Molecular and Cellular Endocrinology 377 (2013) 123–134

Contents lists available at SciVerse ScienceDirect

Molecular and Cellular Endocrinology journal homepage: www.elsevier.com/locate/mce

Dual regulation of adipose triglyceride lipase by pigment epitheliumderived factor: A novel mechanistic insight into progressive obesity Zhiyu Dai a,1, Weiwei Qi a,1, Cen Li a, Juling Lu a, Yuling Mao a, Yachao Yao a, Lei Li a, Ting Zhang a, Honghai Hong a, Shuai Li a, Ti Zhou a, Zhonghan Yang a, Xia Yang a,b, Guoquan Gao a,c,⇑, Weibin Cai a,⇑ a b c

Department of Biochemistry, Zhongshan Medical School, Sun Yat-sen University, Guangzhou 510080, China China Key Laboratory of Tropical Disease Control (Sun Yat-sen University), Ministry of Education, Guangzhou 510080, China Key Laboratory of Functional Molecules from Marine Microorganisms (Sun Yat-sen University), Department of Education of Guangdong Province, Guangzhou 510080, China

a r t i c l e

i n f o

Article history: Received 4 February 2013 Received in revised form 31 May 2013 Accepted 2 July 2013 Available online 10 July 2013 Keywords: ATGL PEDF Lipolysis Obesity Proteasome degradation pathway Free fatty acid

a b s t r a c t Both elevated plasma free fatty acids (FFA) and accumulating triglyceride in adipose tissue are observed in the process of obesity and insulin resistance. This contradictory phenomenon and its underlying mechanisms have not been thoroughly elucidated. Recent studies have demonstrated that pigment epithelium-derived factor (PEDF) contributes to elevated plasma FFA and insulin resistance in obese mice via the activation of adipose triglyceride lipase (ATGL). However, we found that PEDF downregulated adipose ATGL protein expression despite of enhancing lipolysis. Plasma PEDF and FFA were increased in associated with a progressive high-fat-diet, and those outcomes were also accompanied by fat accumulation and a reduction in adipose ATGL. Exogenous PEDF injection downregulated adipose ATGL protein expression and elevated plasma FFA, while endogenous PEDF neutralization significantly rescued the adipose ATGL reduction and also reduced plasma FFA in obese mice. PEDF reduced ATGL protein expression in a time- and dose-dependent manner in differentiated 3T3-L1 cells. Small interfering RNA-mediated PEDF knockdown and antibody-mediated PEDF blockage increased endogenous ATGL expression, and PEDF overexpression downregulated ATGL. PEDF resulted in a decreased half-life of ATGL and regulated ATGL degradation via ubiquitin-dependent proteasomal degradation pathway. PEDF stimulated lipolysis via ATGL using ATGL inhibitor bromoenol lactone, and PEDF also downregulated G0/G1 switch gene 2 (G0S2) expression, which is an endogenous inhibitor of ATGL activation. Overall, PEDF attenuated ATGL protein accumulation via proteasome-mediated degradation in adipocytes, and PEDF also promoted lipolysis by activating ATGL. Elevated PEDF may contribute to progressive obesity and insulin resistance via its dual regulation of ATGL. Ó 2013 Elsevier Ireland Ltd. All rights reserved.

1. Introduction Adipose tissue represents an individual’s pool of excess energy that is stored in the form of triglycerides deposited in lipid droplets. During starvation and exercise, adipose tissue mobilizes to meet the energy demands of the body by hydrolyzing triglycerides into free fatty acids (FFA) and glycerol (Rosen and Spiegelman, 2006). Over the past several years, adipose tissue has been emphasized as the master regulatory organ that controls the whole-body lipid flux. Deregulation of adipose tissue function may lead to excessive circulating FFA and ectopic fat accumulation in non-adipose tissues, such as the liver and skeletal muscle, which results in ⇑ Corresponding authors. Address: Department of Biochemistry, Zhongshan School of Medicine, Sun Yat-sen University, 74 Zhongshan 2nd Road, Guangzhou 510080, China. Tel./fax: +86 20 87332128. E-mail addresses: [email protected] (G. Gao), [email protected] (W. Cai). 1 These authors contributed equally to this work. 0303-7207/$ - see front matter Ó 2013 Elsevier Ireland Ltd. All rights reserved. http://dx.doi.org/10.1016/j.mce.2013.07.001

obesity and type 2 diabetes (Guilherme et al., 2008; Rosen and Spiegelman, 2006; Savage et al., 2007). Hence, studies of adipose lipid regulation may yield treatments for obesity and type 2 diabetes. Lipolysis is a catabolic process that breaks down triglycerides stored in adipose tissue and releases non-esterified fatty acid and glycerol, and the process is precisely controlled by a number of enzymes and factors. For several decades, hormone-sensitive lipase (HSL) was believed to catabolize the initiate cleavage of triglycerides, until adipose triglyceride lipase (ATGL) was identified as a lipase that hydrolyzes triglycerides by three different groups independently (Jenkins et al., 2004; Villena et al., 2004; Zimmermann et al., 2004). It became clear that ATGL firstly hydrolyzes triglycerides, thereby initiating the degradation of stored lipids. Then, diglycerides are cleaved by HSL, and monoglyceride lipase contributes to the final step of lipolysis (Duncan et al., 2007). Moreover, in vivo studies have demonstrated that ATGL-deficiency mice exhibit adiposity and impaired lipolysis both in basal and isoproterenol

124

Z. Dai et al. / Molecular and Cellular Endocrinology 377 (2013) 123–134

(ISO)-stimulated states, as well as reduced plasma FFA levels (Zimmermann et al., 2004). Human subjects with mutated ATGL developed a neutral lipid storage disease (NLSD) (Fischer et al., 2006), which indicated that ATGL is the rate-limiting enzyme that controls the release of cellular FFA. ATGL may be a drug target based on the evidence that inhibition of ATGL reduces the concentration of plasma FFA, which increases insulin sensitivity (Zimmermann et al., 2009). Both plasma FFA and cellular FFA levels are positively correlated with increased insulin resistance and obesity (Boden and Shulman, 2002; Gordon, 1960; Nielsen et al., 2004). Elevated plasma FFA comes from enhancing basal lipolysis in obesity. However, downregulation of adipose ATGL, the key player in lipolysis, is often observed in obese and insulin resistance humans (Jocken et al., 2007; Steinberg et al., 2007) and rodents (Kim et al., 2006). The reasons for these paradoxical phenomena remain unknown, and it is unclear why adipose ATGL levels are not always positively associated with plasma FFA in obesity. Investigation of the mechanism of decreased adipose ATGL and elevated plasma FFA will help us to understand the discrepancy. Despite sequestering lipids, adipose tissue also regulates body energy homeostasis through the secretion of cytokines or adipokines, such as adiponectin and leptin (Guilherme et al., 2008; Halaas et al., 1995; Maeda et al., 2002), which improve insulin sensitivity. However, in obesity, adipocytes also secrete adipokines that contribute to the development of insulin resistance and obesity, including tumor necrosis factor-a (TNF-a) (Hotamisligil, 1999; Miyazaki et al., 2003), monocyte chemoattractant protein-1 (Sartipy and Loskutoff, 2003), resistin (Steppan and Lazar, 2004), and retinol-binding protein (Yang et al., 2005). PEDF is a well-known inhibitor of angiogenesis (Dawson et al., 1999; Gao et al., 2001), as well as a neuroprotective factor (TombranTink and Barnstable, 2003) and an anti-tumor factor (Fernandez-Garcia et al., 2007; Xu et al., 2011; Yang et al., 2009; Zhang et al., 2011). Clinical studies have elucidated that plasma PEDF is elevated in patients who are obese (Wang et al., 2008), and who have metabolic syndrome (Yamagishi et al., 2006) and type 2 diabetes (Akın et al., 2012). Further studies showed that PEDF inhibited adipogenesis via inhibition of the MAPK/ERK pathway and mitotic clonal expansion (Wang et al., 2009). Recent work reported that PEDF released from adipose tissue contributed to the pathogenesis of insulin resistance in obesity and also enhanced adipocyte lipolysis (Crowe et al., 2009). In addition to governing lipolysis in adipocytes, ATGL has been shown to bind with PEDF in the plasma membrane of the retina as a receptor (Notari et al., 2006) and to interact with PEDF in hepatocytes (Chung et al., 2008). Another study also showed that PEDF induces lipolysis in adipocytes and reduces fatty acid oxidation in skeletal muscle in an ATGL-dependent manner (Borg et al., 2011). Our study elucidated that elevated adipose PEDF decreases ATGL protein expression in adipose tissue in obesity, as well as induces basal lipolysis, which explains the concurrent phenomena of progressive fat accumulation and enhanced lipolysis in obesity.

2. Materials and methods

use committee of Sun Yat-sen University (IACUC SYSU, NO. 20061211005). 2.2. Experimental animals and protocols Male C57BL/6J mice (eight-weeks-old) were obtained from the Center of Experimental Animals, Sun Yat-sen University (Guangzhou, China). For the HFD-induced obese mouse model, mice (n = 6) were allowed to acclimate to local conditions for 1 week and then were fed a normal chow diet or a HFD (60% calories from fat; animal center of Guangdong Province, Guangzhou, China) for 4 weeks, 8 weeks, 12 weeks or 16 weeks (Crowe et al., 2009). Mice were anesthetized by intraperitoneal injection of urethane after 8 h of fasting, and they were subsequently sacrificed for tissues and plasma collection. For the in vivo experiment of prolonged PEDF administration, PEDF protein (primary concentration: 7.5 mg/ml) was diluted with PBS to 1 mg/ml prior to injection. Mice (n = 8) were allowed ad libitum access to food and were injected intraperitoneally with 1  PBS or PEDF daily (50 lg; totally 50 ll) for 7 days, and they were sacrificed for tissue and plasma collection after 8 h of fasting (Borg et al., 2011; Crowe et al., 2009). For the in vivo experiment of acute PEDF administration, mice (n = 5) were injected with 1  PBS or PEDF (50 lg) prior to 8 h of fasting, and they were then maintained for 0.5 h or 8 h until being sacrificed (Borg et al., 2011; Crowe et al., 2009). For the in vivo acute PEDF (50 lg) combined ISO (6 lg, Sigma, St. Louis, MO, USA) injection, mice (n = 5) were maintained for 8 h until being sacrificed. For PEDF antibody blockage in obese mice, mice (n = 3) that were fed the HFD for 16 weeks were injected intraperitoneally with IgG (Beyotime, China) or PEDF antibody (Genscript, China) for 7 days, and they were later sacrificed for adipose tissue and plasma collection after 8 h of fasting. All blood and tissue samples isolated from the mice were immediately stored on ice. 2.3. Cell culture The 3T3-L1 cell line was purchased from the Cell Bank of China Science Academy (Shanghai, China). The cells were maintained in Dulbecco’s modified Eagle’s medium (DMEM, GIBCO, Gaithersburg, MD, USA) supplemented with 10% (v/v) fetal bovine serum (FBS, GIBCO, Gaithersburg, MD, USA), and incubated at 37 °C in a humidified incubator at 5% CO2 until use. 3T3-L1 cells were grown to confluence and induced to differentiate as described previously (Borg et al., 2011). Briefly, 2 days post-confluence (defined as day 0), cells were incubated with differentiation medium containing 0.5 mmol/ l isobutylmethylxanthine (IBMX), 1 lM dexamethasone (DEX), 10 lg/ml insulin and 10% FBS for 3 days (IBMX, DEX and insulin were obtained from Sigma, St. Louis, MO, USA). Then, cells were exposed to DMEM with 10 lg/ml insulin and 10% FBS for another 2 days, they were then switched to DMEM with 10% FBS. Maturation of adipocytes was confirmed by BODIPY493/503 (Molecular Probes, Eugene, OR, USA) staining of the lipid droplets that were visualized under an Axioskop microscope (Carl Zeiss, Oberkochen, Germany) at day 9. All cells were starved of serum for 10 h prior to the experiment. 2.4. Purification of recombinant PEDF and construction of mouse ATGL plasmid

2.1. Ethics statement Care, use and treatment of all animals in the present study were in strict agreement with the institutionally approved protocol that followed the guidelines set forth in the Care and Use of Laboratory Animals by the Sun Yat-sen University. The experiment procedures were reviewed and approved by the institutional animal care and

Recombinant PEDF was expressed and purified as described previously (Xu et al., 2011). In brief, the pET30a(+)/PEDF construct was expressed by the BL21 (DE3) Escherichia coli strain (Novagen, Madison, WI, USA) and purified with Ni–NTA His-Bind resin (Novagen, Madison, WI, USA) using FPLC. Recombinant PEDF was confirmed by SDS–PAGE and western blot analysis. Plasmids

Z. Dai et al. / Molecular and Cellular Endocrinology 377 (2013) 123–134

expressing mouse ATGL were constructed into a pTriEx1.1 vector (a gift from Dr. Jian-xing Ma, Oklahoma University) as previously described (Dai et al., 2013). 2.5. Quantitative real-time PCR Total RNA was extracted from adipose tissue of the mice or cultured cells according to the manufacturer’s instructions for Trizol reagent (Invitrogen, CA, USA). Total RNA (500 ng) was used for reverse transcription using PrimeScriptÒ RT reagent Kit Perfect Real Time kit (Takara Bio Inc., Japan). The cDNA was used for quantitative real-time PCR analysis (qPCR) using SYBRÒ Premix Ex Taq™ (Perfect Real Time) (Takara Bio Inc., Japan) and an Roche’s capillary-based Light CyclerÒ 2.0 Systems (Roche Diagnostics Corporation, Indianapolis, IN, USA). Mouse cDNA was amplified with specific primers for PEDF (sense primer: GCTTGGACTCTGATCTCAACTG, antisense primer: AGGGGCAGGAAGAAGATGAT), ATGL (sense primer: TGTGGCCTCATTCCTCCTAC, antisense primer: TCG TGGATGTTGGTGGAGCT), HSL (sense primer: GCTGGGCTGTCAAGCACTGT, antisense primer: GTAACTGGGTAGGCTGCCAT), CGI-58 (sense primer: TGTGCAGGACTCTTACTTGGCAGT, antisense primer: GTTTCTTTGGGCAGACCGGTTTCT), Perilipin (sense primer: AGGGGACTAGACAAATTGG, antisense primer: GCTTCTCCGACTTGCC), G0S2 (sense primer: AGTGCTGCCTCTCTTCCCAC, antisense primer: TTTCCATCTGAGCTCTGGGC) and b-actin (sense primer: ACTCTTCC AGCCTTCCTTC, antisense primer: ATCTCCTTCTGCATCCTGTC) (Invitrogen, CA, USA). Target mRNA was determined using the comparative cycle threshold method of relative quantitation. The calibrator sample was selected from PBS-treated tissue or cell samples, and b-actin was used as an internal control. 2.6. Western blot analysis Adipose tissue samples that weighed 100 mg from four mice were randomly selected from six mice in the different groups and were lysed with 1 ml 1  SDS buffer for total protein extraction. For cell samples, the cells were harvested and lysed after triple PBS washings. Protein concentration was determined using Bio-Rad DC protein assay kit (Bio-Rad Laboratories, Hercules, CA, USA) according to the manufacturer’s protocol. Aliquots of equal amounts of protein (80 lg) from the lysate underwent western blot analysis for the PEDF (rabbit anti-human polyclonal antibody made by our lab as described previously (Gao et al., 2001)), ATGL (rabbit polyclonal antibody from Cayman, Madison, MI, USA) and ubiquitin (mouse monoclonal antibody, Santa Cruz Biotechnology, Santa Cruz, CA, USA) as described previously (Dai et al., 2013). The same membrane was stripped and reblotted with an antibody specific to b-actin or GADPH (mouse monoclonal antibody, Sigma, St. Louis, MO, USA). For quantitative analysis, the bands were selected and quantified using Image J software, and the data were transformed and normalized relative to b-actin or GADPH as the integral optical density (IOD) ratio. 2.7. Measurement of secreted PEDF The plasma of the mice was collected from the retro orbital plexus under urethane anesthesia after 8 h of fasting and was centrifuged for 5 min at 2000 rpm. The supernatant was collected, and the PEDF level was determined using commercial ELISA kit (Cusabio, China) according to the manual. 2.8. Measurement of free fatty acids For the in vivo study, the plasma of the mice was collected and treated as described above. The supernatant was collected and was subjected to FFA measurement using the NEFA assay kit (Wako,

125

Japan). For differentiated 3T3-L1 cells, the cells were incubated with DMEM-2% BSA without phenol red and treated with PEDF (500 nM) or ISO (1 lM) or ATGL inhibitor bromoenol lactone (BL, 25 lM, Cayman, MI, USA) for the time indicated. The supernatant was collected and was subjected to FFA measurement.

2.9. SiRNA transfection and ATGL overexpression SiRNA oligonucleotides matching the selected regions of the mouse PEDF (siPEDF1: 50 -CAGAATTGTGTTTGAGAGGAA-30 and siPEDF2: 50 -CCCGCTAGACTATCACCTTAA-30 ) were obtained from Qiagen, Hilden, Germany. Differentiated 3T3-L1 cells were transfected with siRNA oligonucleotides at a final concentration of 100 nM by electroporation using the Invitrogen Neon transfection system at 1300 V, 20 ms and 2 pulses. Twenty-four hours later, the cells were supplemented with DMEM for 6 h. Next, the cells were harvested for the analysis of PEDF and ATGL expression. For ATGL overexpression, 293A cells that were seeded in a 60-mm well were transfected with the appropriate plasmids of ATGL, EGFP or PEDF (2 lg totally for each well). Total protein was collected 36 h after transfection and was subjected to western blot analysis.

2.10. Protein half-life determination Differentiated 3T3-L1 cells were incubated with cycloheximide (CHX, 10 lg/ml, Sigma, St. Louis, MO, USA) for the time indicated and were analyzed by western blot analysis. The intensity of the bands was quantified using Image J software, and the data were log-transformed. A linear fit was performed to calculate the slope (constant, k) and the half-life (T1/2) using the equation T1/ 2 = ln (2)/k (Belle et al., 2006).

2.11. Ubiquitin conjugated assay Differentiated 3T3-L1 cells were pretreated with 10 lM MG132 (Merck, Germany) for 2 h and then incubated with PBS or 500 nM PEDF for 6 h. The cells were then washed with cold PBS and lysed in the radio immunoprecipitation assay (RIPA) buffer (150 mM NaCl, 1% NP-40, 0.1% SDS, 50 mM Tris–HCl, pH 8.0, 0.5 mM EDTA, 0.25% deoxysodiumcholate, 1 mM phenylmethylsulfonyl fluoride (PMSF), 1  phosphatase inhibitor (Geneapply, Beijing, China), and 1  cocktail (Merck, Germany)) at 4 °C for 4 h. After determining the total protein concentration, aliquots of equal amounts of protein were incubated with rabbit polyclonal antibody against ATGL overnight at 4 °C. Next, protein G-Sepharose beads (Invitrogen, CA, USA) were added and incubated for 4 h at 4 °C. The beads were then centrifuged and washed with pre-cool basic RIPA buffer (150 mM NaCl, 1% NP-40, 0.1% SDS, 50 mM Tris–HCl, pH 8.0). After releasing with 2  SDS buffer, the precipitated proteins were subjected to western blot analysis with total cell lysates.

2.12. Statistical analysis All data were expressed as mean ± standard deviation. SPSS 13.0 software was used for statistical evaluation using one-way ANOVA for comparison of more than two groups, followed by LSD-t test for multiple comparisons amongst groups. Student’s t test was used for comparing between two groups. The correlation analyses were determined by the Pearson correlation test. A p value less than 0.05 was considered to be significant.

126

Z. Dai et al. / Molecular and Cellular Endocrinology 377 (2013) 123–134

3. Results 3.1. Increased PEDF and decreased ATGL in the adipose tissue of HFDfed mice To investigate the expression pattern of PEDF and ATGL in adipose tissue during the progression of obesity, we fed mice with a chronic high-fat diet (HFD), and we fed the control group a normal chow diet. The mRNA and protein levels were assessed at 4 weeks, 8 weeks, 12 weeks and 16 weeks after the feedings began. As shown in Table 1, compared to the control group at 4 weeks, the HFD-fed mice exhibited a nearly 4-fold increase in epididymal adipose tissue weight and an 18% increase in body weight. The epididymal fat and body weight accumulated progressively in HFD-fed mice, which suggested the successful establishment of the obese mouse model. The protein level of PEDF in the HFD-fed mice increased at 4 weeks and remained elevated from 8 to 16 weeks. However, the ATGL expression in epididymal adipose tissue of the HFD-fed mice decreased at 4 weeks and was barely detectable from 8 to 16 weeks (Fig. 1A–D). Moreover, a similar increase in PEDF transcripts and a decrease in ATGL transcripts were observed by qPCR assay in the HFD-fed mice at all time points (Fig. 1E and F). We also found that plasma PEDF was upregulated in the HFD-fed mice compared to that of the chow-fed mice at all stages using ELISA assay (Fig. 1G), and the high level of plasma PEDF was consistent with high levels of plasma FFA in the HFD-mice (Fig. 1H). Furthermore, correlation analysis showed that adipose PEDF protein expression was positively associated with plasma FFA (Fig. 1I) and negatively correlated with adipose ATGL protein expression (Fig. 1J). Additionally, adipose ATGL protein was negatively correlated with plasma FFA (Fig. 1K).

3.2. PEDF decreased adipose ATGL in vivo To mimic the long-term effect of elevated PEDF in obesity, mice were administered recombinant PEDF for 7 days. As shown in Fig. 2A and B, prolonged PEDF administration significantly increased the plasma FFA levels and glycerol level compared to PBS injection. This suggests that lipolysis was enhanced by PEDF. Next, the effect of PEDF on the ATGL mRNA level was evaluated using qPCR assay. Seven-day PEDF treatment did not result in ATGL mRNA alteration compared to the control group (p = 0.186, Fig. 2C). Additionally, ATGL protein levels of adipose tissue were reduced following prolonged PEDF treatment (Fig. 2D). To determine the acute effect of PEDF on ATGL expression, mice were injected with PEDF or PBS and maintained for either 0.5 h or 8 h. The 0.5-h PEDF administration was sufficient to induce elevation of plasma FFA levels, and the enhanced lipolysis was consistently observed after 8 h (Fig. 2E). Furthermore, we tested whether ATGL protein levels were altered using western blot

Table 1 Body weight and epididymal fat mass of mice fed chow diet and HFD during 4– 16 weeks.

Body weight (g)

Epididymal fat mass (g)

Time (week)

Chow

HFD

p value

4⁄ 8⁄ 12⁄ 16⁄ 4⁄ 8⁄ 12⁄ 16⁄

26.52 ± 2.65 28.43 ± 1.33 31.47 ± 1.34 32.58 ± 1.74 0.15 ± 0.08 0.26 ± 0.04 0.31 ± 0.08 0.19 ± 0.12

33.44 ± 3.44 38.27 ± 7.62 43.57 ± 5.91 47.60 ± 5.52 1.38 ± 0.65 1.98 ± 0.86 2.11 ± 0.44 2.01 ± 0.42

<0.0001 <0.0001 <0.0001 <0.0001 0.0028 0.0008 <0.0001 <0.0001

The initial body weight was 24.95 ± 1.01 g (n = 6, p < 0.05).

analysis. As shown in Fig. 2F, acute PEDF administration similarly caused ATGL protein levels to decrease both at both 0.5 h and 8 h. Taken together, these results suggested that PEDF decreases ATGL protein expression at the post-transcriptional level in vivo. 3.3. PEDF neutralization restored adipose ATGL downregulation in obese mice To further confirm the role of PEDF on ATGL expression, obese mice were injected with PEDF monoclonal antibody or IgG for 7 days. As shown in Fig. 3A and B, PEDF neutralization attenuated the plasma FFA level and the glycerol level in obese mice compared to the IgG group, which was demonstrated in a previous study (Crowe et al., 2009). ATGL mRNA and protein expression in adipose tissue were then determined by qPCR and western blot, and we found that PEDF antibody did not change the level of ATGL transcription when compared to the IgG group (Fig. 3C), while PEDF antibody rescued the decrease in the level of adipose ATGL protein (Fig. 3D), which suggests that the downregulation of ATGL in obesity might due to elevated PEDF. Our in vivo studies indicate that elevated circulating PEDF contributes to elevated plasma FFA and low levels ATGL in adipose tissue in the development of obesity. 3.4. PEDF downregulated ATGL protein levels in vitro To evaluate the effects of PEDF on ATGL regulation in vitro, we treated differentiated 3T3-L1 cells with PEDF. We did not observe a significant change in ATGL transcription in differentiated 3T3L1 cells at different time points (Fig. 4A). While PEDF decreased ATGL protein levels in a time-dependent manner (Fig. 4B). In addition, a dose-concentration of the PEDF-treated experiment showed that 10 nM of PEDF was sufficient to downregulate ATGL (Fig. 4C). However, there was no striking difference in the presence of ISO (Fig. S1). These results indicate that PEDF downregulates ATGL protein expression without affecting ATGL transcription. To further demonstrate the effects of endogenous PEDF on ATGL, we delivered the PEDF-target siRNA to knock down endogenous PEDF in differentiated 3T3-L1 cells. As shown in Fig. 4D, both siPEDF1 and siPEDF2 oligonucleotides significantly reduced the endogenous PEDF expression, while also increasing the protein levels of cellular ATGL. Furthermore, the endogenous ATGL protein was enhanced in the presence of specific PEDF antibody, which was used to block the PEDF secreted by 3T3-L1 cells (Fig. 4E). To determine whether the PEDF-induced downregulation of ATGL was restricted in adipocytes, ATGL and PEDF plasmids were cotransfected into 293A cells. As shown in Fig. 4F, ATGL expression was attenuated in the PEDF transfected 293A cells. Therefore, PEDF decreased ATGL protein expression in vitro. 3.5. PEDF induced ATGL ubiquitin-dependent degradation To determine how ATGL was regulated by PEDF in differentiated 3T3-L1 cells, we hypothesized that PEDF might have some effects on the protein stability of ATGL. To confirm the hypothesis, the half-life of ATGL was measured using cycloheximide (CHX), which blocks protein synthesis in vitro. As shown in Fig. 5A and B, ATGL had a half-life of approximately 69 min. The amount of ATGL was dramatically decreased at 15 min and was sustained for 120 min under CHX treatment. The ATGL degradation was further accelerated in the presence of PEDF with CHX (Fig. 5C). To evaluate which pathway contributes to ATGL protein degradation, we tested the degradation rate of ATGL in the presence of different protease inhibitors. As shown in Fig. 5D, there was no change of ATGL in differentiated 3T3-L1 cells with the treatment of the general protease inhibitor leupeptin. However, a small amount of ATGL accumulated in cells treated with lysosomal inhibitor ammonium chloride

Z. Dai et al. / Molecular and Cellular Endocrinology 377 (2013) 123–134

127

Fig. 1. Expression pattern of adipose PEDF and ATGL in progressively obese mice Expression of adipose PEDF and ATGL protein at 4 weeks (A), 8 weeks (B), 12 weeks (C) and 16 weeks (D) in HFD-fed mice and chow-fed mice by immunoblotting analysis. b-actin was used as a loading control. The bands were quantified using Image J software, and data were transformed and normalized relative to b-actin (n = 4). Chow as chow-fed mice group, HFD as HFD-fed mice (p < 0.05). PEDF (E) and ATGL (F) mRNA expression of adipose tissue at 4, 8,12 and 16 weeks in HFD-fed mice and chow-fed mice by qPCR analysis (n = 6). b-actin was used as an internal control (p < 0.05). Plasma PEDF (G) and FFA (H) levels at 4, 8, 12 and 16 weeks in HFD-fed mice and chow-fed mice were determined using a PEDF ELISA kit and NEFA kit as described in the Methods. (n = 6,  p < 0.05). (I) (J) (K) Correlation analysis between adipose PEDF protein expression, plasma FFA levels and adipose ATGL protein expression. (n = 32).

128

Z. Dai et al. / Molecular and Cellular Endocrinology 377 (2013) 123–134

Fig. 2. The effect of PEDF on ATGL expression of adipose tissue in vivo Plasma FFA (A) and glycerol (B) were determined in mice that were fasted for 8 h and underwent prolonged PEDF and PBS injection (7 days, n = 8, p < 0.05). (C) ATGL transcription of adipose tissue between prolonged PEDF- and PBS-treated mice (7 days) was compared using qPCR analysis. b-actin mRNA expression was used as an internal control (n = 8). (D) ATGL protein expression of adipose tissue was compared by western blot analysis in prolonged PEDF- and PBS-treated mice (7 days). b-actin was used as a loading control. The bands were quantified and normalized relative to b-actin (n = 4, p < 0.05). (E) Plasma FFA was measured in mice that were fasted for 8 h and underwent acute PEDF- and PBS-injection (0.5 h and 8 h) (n = 5, p < 0.05). (F) Adipose ATGL proteins of acute PEDF- and PBS-treated mice (0.5 h and 8 h) were assessed by immunoblotting. b-actin was used as a loading control. The bands were quantified and normalized relative to bactin (n = 3, p < 0.05).

(NH4Cl), and a great amount of ATGL accumulated in the presence of the proteasome specific inhibitor MG132. These results indicate that ATGL degradation is regulated by the proteasome-mediated degradation pathway. To further investigate whether ATGL undergoes proteasomedependent degradation regulated by PEDF, MG132 was applied with PEDF. As shown in Fig. 5E and F, PEDF-induced reduction of

ATGL protein was completely restored in differentiated 3T3-L1 cells in the presence of MG132 in both the time- and concentration–response experiments. This indicates that PEDF downregulates ATGL by proteasome-dependent protein degradation. Furthermore, we employed an ubiquitination assay to test whether ATGL was ubiquitinated prior to protein degradation induced by PEDF. The differentiated 3T3-L1 cells were treated with MG132

Z. Dai et al. / Molecular and Cellular Endocrinology 377 (2013) 123–134

129

Fig. 3. PEDF blockage upregulated ATGL expression in obese mice Comparisons between IgG and PEDF antibody treatment (7 days) with regard to FFA levels (A) and glycerol (B), ATGL mRNA levels (C) and ATGL protein levels (D) in obese mice. (n = 3, p < 0.05).

Fig. 4. PEDF attenuated adipose ATGL expression in vitro (A) A time-response experiment of PEDF on ATGL transcription was assessed in differentiated 3T3-L1 cells using a qPCR assay. Cell were treated with PEDF (500 nM) for the indicated time (0.5, 1, 3, 6, 12 h), then ATGL mRNA was analyzed relative to b-actin mRNA by qPCR (n = 6). (B) A time-response experiments of PEDF on ATGL expression were assessed in differentiated 3T3-L1 cells. Cells were incubated with PEDF (500 nM) for the indicated time (15, 30 and 60 min). (C) A dose–response study of PEDF on ATGL protein level was verified in differentiated 3T3-L1 cells. Cells were incubated with different doses of PEDF for 6 h. (D) ATGL expression was evaluated by PEDF knockdown. Cells were transfected with 100 nM PEDF-target siRNA (siPEDF1 and siPEDF2) using electroporation as described in the Method, at 30 h post-transfection, cells were subjected to western blot analysis. (E) ATGL expression was tested by PEDF antibody blockage. Cells were starved, and PEDF specific antibody (2 lg/ml) was added for 24 h, at which point the cells were subjected to western blot analysis. (F) ATGL expression in 293A cells was decreased by PEDF transfection. 293A cells were co-transfected with ATGL and PEDF plasmids as described in the Methods, and 36 h after transfection, the cells were collected and subjected to western blot analysis.

130

Z. Dai et al. / Molecular and Cellular Endocrinology 377 (2013) 123–134

Fig. 5. PEDF-induced ATGL degradation in adipocytes is proteasome-dependent. (A) ATGL stability was evaluated under CHX treatment for the indicated time (15, 30, 60, 120 and 240 min). The ATGL protein level was determined using western blot analysis, and b-actin was used as an internal control. (B) Quantitative analysis of the ATGL level using Image J software. The data were log-transformed and calculated as described in the Methods; the half-life of ATGL was as T(1/2). (C) ATGL stability was evaluated under CHX or PEDF treatment for the indicated time (1, 2, 4 and 8 h), and DMSO was used as a negative control. (D) Differentiated 3T3-L1 adipocytes were pretreated with DMSO, 5 lg/ml leupeptin, 10 lM NH4Cl or 10 lM MG132 for 30 min, and the ATGL protein level was evaluated using western blotting after 6 h. (E) (F) Differentiated 3T3-L1 cells were incubated with PEDF (500 nM) for the indicated time (15, 30 and 60 min) (E) or for the different doses of PEDF (6 h) (F) in the absence or presence of MG132. Comparable DMSO was used as a negative control. ATGL protein was measured by western blot, and b-actin was used as an internal control. (G) Differentiated 3T3-L1 cells were treated with PEDF or MG132 or DMSO for 6 h, then immunoprecipitated with ATGL antibody. ATGL ubiquitination was determined using anti-ubiquitin antibody. IP: immunoprecipitation. (Ub)n-ATGL: polyubiquitinated ATGL.

and PEDF. ATGL was pulled down by immunoprecipitation with a specific anti-ATGL antibody. As shown in Fig. 5G, MG132 pretreatment significantly resulted in a typical protein ladder. Therefore, ATGL was polyubiquitinated. Importantly, PEDF in combination with MG132 treatment enhanced the levels of polyubiquitinated ATGL. This suggests that PEDF induced ATGL to undergo ubiquitin-mediated proteasome degradation.

3.6. PEDF induced lipolysis in adipose tissue via ATGL Our studies found plasma PEDF to be positively associated with plasma FFA levels during the progression of obesity (Fig. 1I). Our further studies also found that recombinant PEDF induced basal lipolysis in adipose tissue (Fig. 2A and E), but not ISO-stimulated lipolysis in vivo (Fig. 6A) or in differentiated 3T3-L1 cells (Fig. 6B), which is

Z. Dai et al. / Molecular and Cellular Endocrinology 377 (2013) 123–134

131

Fig. 6. PEDF promoted adipose tissue lipolysis (A) Plasma FFA was determined among mice treated with acute PEDF (50 lg) and ISO (6 lg) injection (8 h) (n = 8, p < 0.05). (B) Differentiated 3T3-L1 cells were treated with PEDF (500 nM) or ISO (1 lM) for 6 h, FFA in the media of differentiated 3T3-L1 cells was measured (n = 3, p < 0.05). (C) Differentiated 3T3-L1 cells were pretreated with ATGL inhibitor bromoenol lactone (BL, 25 lM) for 2 h, then they were stimulated with PEDF (500 nM) for another 6 h, FFA in the media of differentiated 3T3-L1 cells was measured (n = 3, p < 0.05, vs DMSO, #p < 0.05, vs PEDF). (D) Adipose HSL, CGI-58, G0S2 and perilipin 1 mRNA were analyzed relative to b-actin mRNA by qPCR in the prolonged PEDF treatment experiment (n = 8, p < 0.05). (E) Differentiated 3T3-L1 cells were incubated with PEDF (500 nM) or PBS for 6 h, then HSL, CGI-58, G0S2 and perilipin 1 mRNA were analyzed relative to b-actin mRNA by qPCR (n = 3, p < 0.05).

consistent with previous reports (Borg et al., 2011; Crowe et al., 2009). Furthermore, PEDF neutralization attenuated plasma FFA and glycerol levels in obese mice (Fig. 3A and B). Previous studies demonstrated that PEDF stimulated lipolysis in adipose tissue via ATGL using an isolated adipose tissue explant from ATGL/ mice (Borg et al., 2011). Similarly, our studies demonstrated that the ATGL inhibitor bromoenol lactone (BL) blocked PEDF-stimulated lipolysis in differentiated 3T3-L1 cells (Fig. 6C). Moreover, we also found that PEDF downregulated ATGL-inhibited factor G0/G1 switch gene 2 (G0S2) and lipid droplet associated protein perilipin 1 expression in the prolonged PEDF treatment experiment and in differentiated 3T3-L1 cells (Fig. 6D and E), but it did not affect the HSL or CGI-58 transcripts. The results indicate that PEDF enhances lipolysis via ATGL in spite of the degradation of ATGL induced by PEDF. 4. Discussion Although a number of studies have shown that plasma levels of PEDF are elevated in human subjects with obesity, type 2 diabetes

and metabolic syndrome, the potential role of PEDF in metabolism remains unknown. Recent studies revealed that PEDF enhanced basal lipolysis and impaired insulin sensitivity in obesity by modulating ATGL (Borg et al., 2011; Crowe et al., 2009; Famulla et al., 2010). Elevated plasma PEDF in obese subjects induces basal lipolysis in adipocytes and liberates more FFA into the circulation, which contributes to ectopic triglyceride accumulation in the peripheral organs (Crowe et al., 2009). Regarded as the controller of lipolysis and due to its close relationship with PEDF, ATGL is important in mediating the action of PEDF in adipocytes. Our study focused on post-translational regulation of ATGL by PEDF. Despite of the increasing lipolytic rate by PEDF, the protein level of ATGL was reduced by PEDF both in vivo and in vitro. We also found that ATGL underwent proteasome-mediated degradation regulated by PEDF. Previous studies have demonstrated that the level of ATGL determines the rate of lipolysis (Zimmermann et al., 2004; Zimmermann et al., 2009), and decreased adipose ATGL protein expression has often been observed in obese humans (Jocken

132

Z. Dai et al. / Molecular and Cellular Endocrinology 377 (2013) 123–134

Fig. 7. The schematic overview of the potential mechanism involved in PEDF-mediated regulation of ATGL in adipocytes PEDF promotes ubiquitin–proteasome-mediated protein degradation of ATGL and consequently downregulates the protein level of ATGL in the setting of possessive obesity and insulin resistance. Furthermore, PEDF enhances lipolysis in adipocytes and promotes the release of FFA by activating the ATGL hydrolase. Therefore, the dual regulation of ATGL by PEDF may be responsible for the observation of contradictory phenomena in the progression of obesity and insulin resistance. PEDF, pigment epithelium-derived factor; ATGL, adipose triglyceride lipase; FFA, free fatty acids; TGs, triglycerides; Ub, ubiquitin.

et al., 2007; Steinberg et al., 2007) and rodents (Kim et al., 2006). Our data, which are consistent with these reports, show that the protein levels of ATGL in the adipose tissue of HFD-fed mice was dramatically decreased from 4 weeks to 16 weeks compared to that of chow-fed mice. In contrast with ATGL, PEDF transcription and protein levels were elevated in HFD-fed mice from 4 weeks to 16 weeks. Another group observed parallel observations from 12 weeks HFD-fed and ob/ob mice (Crowe et al., 2009). Additionally, we found that the negative correlation between PEDF and ATGL protein expression emerged as early as 4 weeks in HFD-fed mice (Fig. 1). This indicates that the time point of modulating PEDF and ATGL expression in obese subjects needs to be considered and further explored. The negative correlation between PEDF and ATGL expression in obesity strongly suggests that PEDF might regulate ATGL expression in adipocytes. Actually, our in vivo study showed that prolonged and acute PEDF exposure decreased ATGL protein levels in adipose tissue (Fig. 2), while PEDF neutralization restored the downregulation of adipose ATGL in obese mice (Fig. 3). Similar results from in vitro studies also indicated that ATGL protein levels were reduced in response to PEDF in differentiated 3T3-L1 cells. Endogenous ATGL was upregulated by PEDF knockdown and PEDF antibody blockage (Fig. 4). Previous studies demonstrated that adipose ATGL overexpression is protective against diet-induced obesity and improves insulin resistance (Ahmadian et al., 2009). On the contrary, the reduced ATGL protein expression in adipose tissue associated with PEDF provides an indirect hint that the PEDF-mediated reduction in ATGL availability impaires absolute ATGL enzyme activity and lipolysis, which might result in progressive fat accumulation in adipose tissue in the setting of excessive lipid influx. However, further investigation is required to confirm this hypothesis conclusively. Our study and previous studies have demonstrated that PEDF stimulates lipolysis in adipocytes and contributes to elevated circulating FFA in obesity. Our in vitro studies also confirm that the stimulated effect of PEDF occurs via ATGL, which paralleled the results of another report (Borg et al., 2011). ATGL is the rate-limiting factor governing basal lipolysis, and its regulation is complex (Zechner et al., 2012). The level of ATGL does not always

correlate with its cellular hydrolase activity (Zechner et al., 2012). In addition to the cellular protein level, the full hydrolase activity of ATGL is regulated by its cofactors, such as comparative gene identification-58 (CGI-58), as well as G0S2 and its subcellular localization in lipid droplets (Zechner et al., 2012). Although we observed that the total cellular ATGL was downregulated by PEDF, it is still possible that the remaining ATGL translocates to lipid droplets in the presence of PEDF, as ATGL could traffic to the lipid droplets in hepatocytes with PEDF treatment (Chung et al., 2008). We also observed that PEDF resulted in the downregulation of G0S2 transcription (Fig. 6D and E). This observation suggests that decreased G0S2 might release G0S2bound ATGL and facilitate lipolysis. In addition, we found that PEDF reduced perilipin 1 transcription (Fig. 6D and E). Decreasing perilipin 1 had been shown to release CGI-58, which enhances lipolysis (Brasaemle, 2000; Brasaemle et al., 2008). Collectively, although ATGL expression was downregulated, the total ATGL hydrolase activity and lipolysis might be upregulated by PEDF (Fig. 7). Our data present a pronounced ATGL protein decrease induced by short-term PEDF treatment. Although ATGL transcription of adipose tissue was decreased in obese mice, we did not observe the transcriptional regulation of ATGL in PEDF-injected mice, in the neutralization experiment or in differentiated 3T3-L1 cells. This might due to other factors, such as insulin or TNF-a (Chakrabarti and Kandror, 2009; Kim, 2006), which have been demonstrated to regulate ATGL transcription. Additionally, we demonstrated for the first time that ATGL undergoes proteasome-degradation that is induced or activated by PEDF (Fig. 7). Based on the direct interaction of ATGL and PEDF found by other groups (Chung et al., 2008; Notari et al., 2006), the underlying mechanism of ATGL degradation might be regulated by PEDF in post-translational modification, and further investigation is needed to answer both of these questions. In summary, our findings have elucidated the dual regulation of ATGL by PEDF: PEDF downregulated ATGL expression and activated lipolysis via ATGL in obesity. PEDF-induced ATGL downregulation occurred via proteasome-degradation. Fat accumulation in adipose tissue resulted in increased adipose PEDF and plasma

Z. Dai et al. / Molecular and Cellular Endocrinology 377 (2013) 123–134

PEDF, which induced ATGL degradation and FFA release in adipose tissue and, in turn, led to progressive obesity and insulin resistance. Acknowledgements This study was supported by National Nature Science Foundation of China (30971208, 30973449, 81070746, 81172163, 81272338, 81272515, 81200706), National Key Sci-Tech Special Project of China (2013ZX09102-053), Program for Doctoral Station in University (20100171110049), Guandong natural science foundation (10151008901000007, S2012010009250, S2012040006986 and Key Sci-tech Research Project 10251008901000009, 2011B031200006), Fundamental Research Funds for the Central Universities of China (Youth Program 09YKPY73, 10YKPY28), Changjiang Scholars and Innovative Research Team in University (985 project PCSIRT 0947), Guangzhou Science and Technology Project (2011Y1-00017-8 & 12A52061519) and Yixian innovation personnel program of Sun Yat-sen University. Appendix A. Supplementary material Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.mce.2013. 07.001. References Ahmadian, M., Duncan, R.E., Varady, K.A., Frasson, D., Hellerstein, M.K., Birkenfeld, A.L., Samuel, V.T., Shulman, G.I., Wang, Y., Kang, C., Sul, H.S., 2009. Adipose overexpression of desnutrin promotes fatty acid use and attenuates dietinduced obesity. Diabetes 58, 855–866. Akın, S., Aksoy, D.Y., Cınar, N., Aydın, K., Karaag˘aog˘lu, E., Arıyürek, M., Gülçelik, N.E., Usman, A., Gürlek, A., 2012. Pigment epithelium-derived factor increases in type 2 diabetes after treatment with metformin. Clin. Endocrinol. 77, 852– 856. Belle, A., Tanay, A., Bitincka, L., Shamir, R., O’Shea, E.K., 2006. Quantification of protein half-lives in the budding yeast proteome. Proc. Natl. Acad. Sci. USA. 103, 13004–13009. Boden, G., Shulman, G.I., 2002. Free fatty acids in obesity and type 2 diabetes: defining their role in the development of insulin resistance and beta-cell dysfunction. Eur. J. Clin. Invest. 32 (Suppl 3), 14–23. Borg, M.L., Andrews, Z.B., Duh, E.J., Zechner, R., Meikle, P.J., Watt, M.J., 2011. Pigment epithelium-derived factor regulates lipid metabolism via adipose triglyceride lipase. Diabetes 60, 1458–1466. Brasaemle, D.L., Rubin, B., Harten, I.A., Gruia-Gray, J., Kimmel, A.R., Londos, C., 2000. Perilipin A increases triacylglycerol storage by decreasing the rate of triacylglycerol hydrolysis. J. Biol. Chem. 275, 38486–38493. Brasaemle, D.L., Subramanian, V., Garcia, A., Marcinkiewicz, A., Rothenberg, A., 2008. Perilipin A and the control of triacylglycerol metabolism. Mol. Cell. Biochem. 326, 15–21. Chakrabarti, P., Kandror, K.V., 2009. FoxO1 controls insulin-dependent adipose triglyceride lipase (ATGL) expression and lipolysis in adipocytes. J. Biol. Chem. 284, 13296–13300. Chung, C., Doll, J.A., Gattu, A.K., Shugrue, C., Cornwell, M., Fitchev, P., Crawford, S.E., 2008. Anti-angiogenic pigment epithelium-derived factor regulates hepatocyte triglyceride content through adipose triglyceride lipase (ATGL). J. Hepatol. 48, 471–478. Crowe, S., Wu, L.E., Economou, C., Turpin, S.M., Matzaris, M., Hoehn, K.L., Hevener, A.L., James, D.E., Duh, E.J., Watt, M.J., 2009. Pigment epithelium-derived factor contributes to insulin resistance in obesity. Cell Metab. 10, 40–47. Dai, Z., Lu, L., Yang, Z., Mao, Y., Lu, J., Li, C., Qi, W., Chen, Y., Yao, Y., Li, L., Chen, S., Zhang, Y., Cai, W., Yang, X., Gao, G., 2013. Kallikrein-binding protein inhibits LPS-induced TNF-a by upregulating SOCS3 expression. J. Cell. Biochem 114, 1020–1028. Dawson, D.W., Volpert, O.V., Gillis, P., Crawford, S.E., Xu, H., Benedict, W., Bouck, N.P., 1999. Pigment epithelium-derived factor: a potent inhibitor of angiogenesis. Science 285, 245–248. Duncan, R.E., Ahmadian, M., Jaworski, K., Sarkadi-Nagy, E., Sul, H.S., 2007. Regulation of lipolysis in adipocytes. Annu. Rev. Nutr. 27, 79–101. Famulla, S., Lamers, D., Hartwig, S., Passlack, W., Horrighs, A., Cramer, A., Lehr, S., Sell, H., Eckel, J., 2010. Pigment epithelium-derived factor is one of the most abundant proteins secreted by human adipocytes and induces insulin resistance and inflammatory signaling in muscle and fat cells. Int. J. Obes., 1– 11.

133

Fernandez-Garcia, N.I., Volpert, O.V., Jimenez, B., 2007. Pigment epithelium-derived factor as a multifunctional antitumor factor. J. Mol. Med. 85, 15– 22. Fischer, J., Lefèvre, C., Morava, E., Mussini, J.-M., Laforêt, P., Negre-Salvayre, A., Lathrop, M., Salvayre, R., 2006. The gene encoding adipose triglyceride lipase (PNPLA2) is mutated in neutral lipid storage disease with myopathy. Nat. Genet. 39, 28–30. Gao, G., Li, Y., Zhang, D., Gee, S., Crosson, C., Ma, J., 2001. Unbalanced expression of VEGF and PEDF in ischemia-induced retinal neovascularization. FEBS Lett. 489, 270–276. Gordon, E.S., 1960. Non-esterified fatty acids in the blood of obese and lean subjects. Am. J. Clin. Nutr. 8, 740–747. Guilherme, A., Virbasius, J.V., Puri, V., Czech, M.P., 2008. Adipocyte dysfunctions linking obesity to insulin resistance and type 2 diabetes. Nat. Rev. Mol. Cell Biol. 9, 367–377. Halaas, J.L., Gajiwala, K.S., Maffei, M., Cohen, S.L., Chait, B.T., Rabinowitz, D., Lallone, R.L., Burley, S.K., Friedman, J.M., 1995. Weight-reducing effects of the plasma protein encoded by the obese gene. Science 269, 543–546. Hotamisligil, G.S., 1999. The role of TNFalpha and TNF receptors in obesity and insulin resistance. J. Intern. Med. 245, 621–625. Jenkins, C.M., Mancuso, D.J., Yan, W., Sims, H.F., Gibson, B., Gross, R.W., 2004. Identification, cloning, expression, and purification of three novel human calcium-independent phospholipase A2 family members possessing triacylglycerol lipase and acylglycerol transacylase activities. J. Biol. Chem. 279, 48968–48975. Jocken, J.W.E., Langin, D., Smit, E., Saris, W.H.M., Valle, C., Hul, G.B., Holm, C., Arner, P., Blaak, E.E., 2007. Adipose triglyceride lipase and hormone-sensitive lipase protein expression is decreased in the obese insulin-resistant state. J. Clin. Endocrinol. Metab. 92, 2292–2299. Kim, J.-Y., Tillison, K., Lee, J.-H., Rearick, D.A., Smas, C.M., 2006. The adipose tissue triglyceride lipase ATGL/PNPLA2 is downregulated by insulin and TNF-a in 3T3L1 adipocytes and is a target for transactivation by PPARc. AJP: Endocrinol. Metab. 291, E115–E127. Maeda, N., Shimomura, I., Kishida, K., Nishizawa, H., Matsuda, M., Nagaretani, H., Furuyama, N., Kondo, H., Takahashi, M., Arita, Y., Komuro, R., Ouchi, N., Kihara, S., Tochino, Y., Okutomi, K., Horie, M., Takeda, S., Aoyama, T., Funahashi, T., Matsuzawa, Y., 2002. Diet-induced insulin resistance in mice lacking adiponectin/ACRP30. Nat. Med. 8, 731–737. Miyazaki, Y., Pipek, R., Mandarino, L.J., DeFronzo, R.A., 2003. Tumor necrosis factor a and insulin resistance in obese type 2 diabetic patients. Int. J. Obes. 27, 88– 94. Nielsen, S., Guo, Z., Johnson, C.M., Hensrud, D.D., Jensen, M.D., 2004. Splanchnic lipolysis in human obesity. J. Clin. Invest. 113, 1582–1588. Notari, L., Baladron, V., Aroca-Aguilar, J.D., Balko, N., Heredia, R., Meyer, C., Notario, P.M., Saravanamuthu, S., Nueda, M.L., Sanchez-Sanchez, F., Escribano, J., Laborda, J., Becerra, S.P., 2006. Identification of a lipase-linked cell membrane receptor for pigment epithelium-derived factor. J. Biol. Chem. 281, 38022– 38037. Rosen, E.D., Spiegelman, B.M., 2006. Adipocytes as regulators of energy balance and glucose homeostasis. Nature 444, 847–853. Sartipy, P., Loskutoff, D.J., 2003. Monocyte chemoattractant protein 1 in obesity and insulin resistance. Proc. Natl. Acad. Sci. USA. 100, 7265–7270. Savage, D.B., Petersen, K.F., Shulman, G.I., 2007. Disordered lipid metabolism and the pathogenesis of insulin resistance. Physiol. Rev. 87, 507–520. Steinberg, G.R., Kemp, B.E., Watt, M.J., 2007. Adipocyte triglyceride lipase expression in human obesity. AJP: Endocrinol. Metab. 293, E958–E964. Steppan, C.M., Lazar, M.A., 2004. The current biology of resistin. J. Intern. Med. 255, 439–447. Tombran-Tink, J., Barnstable, C.J., 2003. PEDF: a multifaceted neurotrophic factor. Nat. Rev. Neurosci. 4, 628–636. Villena, J.A., Roy, S., Sarkadi-Nagy, E., Kim, K.-H., Sul, H.S., 2004. Desnutrin, an adipocyte gene encoding a novel patatin domain-containing protein, is induced by fasting and glucocorticoids: ectopic expression of desnutrin increases triglyceride hydrolysis. J. Biol. Chem. 279, 47066–47075. Wang, P., Smit, E., Brouwers, M.C.G.J., Goossens, G.H., van der Kallen, C.J.H., van Greevenbroek, M.M.J., Mariman, E.C.M., 2008. Plasma pigment epitheliumderived factor is positively associated with obesity in Caucasian subjects, in particular with the visceral fat depot. Eur. J. Endocrinol. 159, 713– 718. Wang, M., Wang, J.J., Li, J., Park, K., Qian, X., Ma, J.X., Zhang, S.X., 2009. Pigment epithelium-derived factor suppresses adipogenesis via inhibition of the MAPK/ ERK pathway in 3T3-L1 preadipocytes. AJP: Endocrinol. Metab. 297, E1378– E1387. Xu, Z., Fang, S., Zuo, Y., Zhang, Y., Cheng, R., Wang, Q., Yang, Z., Cai, W., Ma, J., Yang, X., Gao, G., 2011. Combination of pigment epithelium-derived factor with radiotherapy enhances the antitumor effects on nasopharyngeal carcinoma by downregulating vascular endothelial growth factor expression and angiogenesis. Cancer Sci. 102, 1789–1798. Yamagishi, S.-I., Adachi, H., Abe, A., Yashiro, T., Enomoto, M., Furuki, K., Hino, A., Jinnouchi, Y., Takenaka, K., Matsui, T., Nakamura, K., Imaizumi, T., 2006. Elevated serum levels of pigment epithelium-derived factor in the metabolic syndrome. J. Clin. Endocrinol. Metab. 91, 2447–2450. Yang, Q., Graham, T.E., Mody, N., Preitner, F., Peroni, O.D., Zabolotny, J.M., Kotani, K., Quadro, L., Kahn, B.B., 2005. Serum retinol binding protein 4 contributes to insulin resistance in obesity and type 2 diabetes. Nature 436, 356– 362.

134

Z. Dai et al. / Molecular and Cellular Endocrinology 377 (2013) 123–134

Yang, H., Cheng, R., Liu, G., Zhong, Q., Li, C., Cai, W., Yang, Z., Ma, J., Yang, X., Gao, G., 2009. PEDF inhibits growth of retinoblastoma by anti-angiogenic activity. Cancer Sci. 100, 2419–2425. Zechner, R., Zimmermann, R., Eichmann, T.O., Kohlwein, S.D., Haemmerle, G., Lass, A., Madeo, F., 2012. FAT SIGNALS – lipases and lipolysis in lipid metabolism and signaling. Cell Metab. 15, 279–291. Zhang, Y., Han, J., Yang, X., Shao, C., Xu, Z., Cheng, R., Cai, W., Ma, J., Yang, Z., Gao, G., 2011. Pigment epithelium-derived factor inhibits angiogenesis and growth of gastric carcinoma by down-regulation of VEGF. Oncol. Rep. 26, 681–686.

Zimmermann, R., Strauss, J.G., Haemmerle, G., Schoiswohl, G., Birner-Gruenberger, R., Riederer, M., Lass, A., Neuberger, G., Eisenhaber, F., Hermetter, A., Zechner, R., 2004. Fat mobilization in adipose tissue is promoted by adipose triglyceride lipase. Science 306, 1383–1386. Zimmermann, R., Lass, A., Haemmerle, G., Zechner, R., 2009. Fate of fat: the role of adipose triglyceride lipase in lipolysis. BBA – Mol. Cell Biol. Lipids 1791, 494– 500.