Am J Physiol Endocrinol Metab 302: E705–E713, 2012. First published January 10, 2012; doi:10.1152/ajpendo.00237.2011.
Overexpression of a short human seipin/BSCL2 isoform in mouse adipose tissue results in mild lipodystrophy Xin Cui,1* Yuhui Wang,1* Lingjun Meng,2 Weihua Fei,3 Jingna Deng,4 Guoheng Xu,4 Xingui Peng,5 Shenghong Ju,5 Ling Zhang,1 George Liu,1 Liping Zhao,2 and Hongyuan Yang3 1
Institute of Cardiovascular Sciences and Key Laboratory of Molecular Cardiovascular Sciences, Ministry of Education, Peking University Health Science Center; 2Transgenic Animal Center, National Institute of Biological Science, Beijing, China; 3School of Biotechnology and Biomolecular Sciences, the University of New South Wales, Sydney, New South Wales, Australia; 4Department of Physiology and Key Laboratory of Molecular Cardiovascular Sciences, Ministry of Education, Peking University Health Science Center, Beijing; and 5Department of Radiology, Zhong-Da Hospital, Southeast University, Jiangsu Key Laboratory of Molecular Imaging and Functional Imaging, Nanjing, China Submitted 10 May 2011; accepted in final form 4 January 2012
Cui X, Wang Y, Meng L, Fei W, Deng J, Xu G, Peng X, Ju S, Zhang L, Liu G, Zhao L, Yang H. Overexpression of a short human seipin/BSCL2 isoform in mouse adipose tissue results in mild lipodystrophy. Am J Physiol Endocrinol Metab 302: E705–E713, 2012. First published January 10, 2012; doi:10.1152/ajpendo.00237.2011.— Berardinelli-Seip congenital lipodystrophy type 2 (BSCL2) is a recessive disorder characterized by an almost complete loss of adipose tissue, insulin resistance, and fatty liver. BSCL2 is caused by lossof-function mutations in the BSCL2/seipin gene, which encodes seipin. The essential role for seipin in adipogenesis has recently been established both in vitro and in vivo. However, seipin is highly upregulated at later stages of adipocyte development, and its role in mature adipocytes remains to be elucidated. We therefore generated transgenic mice overexpressing a short isoform of human BSCL2 gene (encoding 398 amino acids) using the adipocyte-specific aP2 promoter. The transgenic mice produced ⬃150% more seipin than littermate controls in white adipose tissue. Surprisingly, the increased expression of seipin markedly reduced the mass of white adipose tissue and the size of adipocytes and lipid droplets. This may be due in part to elevated lipolysis rates in the transgenic mice. Moreover, there was a nearly 50% increase in the triacylglycerol content of transgenic liver. These results suggest that seipin promotes the differentiation of preadipocytes but may inhibit lipid storage in mature adipocytes. Berardinelli-Seip congenital lipodystrophy type 2; fatty liver
(CGL, also known as Berardinelli-Seip congenital lipodystrophy, or BSCL), is an autosomal recessive disorder characterized by a near total loss of adipose tissue, severe insulin resistance, hypertriglyceridemia, and fatty liver (2, 3). Genome-wide linkage analysis has identified two loci for CGL. CGL type 1 (CGL1) is caused by mutations in the 1-acylglycerol-3-phosphate-O-acyltransferase-2 (AGPAT2) gene and CGL2 by mutations in the BSCL2 gene that encodes seipin (1, 20). Recently, a homozygous nonsense mutation in caveolin-1 has been linked to CGL in a candidate gene approach (16). Cavin has subsequently been shown to play an important role in adipogenesis because it is essential for the function of the caveolins (14, 18). Both AGPAT2 and caveolin-1/cavin have clear cellular functions; AGPAT2 catalyzes the formation of phosphatidic acid (PA) CONGENITAL GENERALIZED LIPODYSTROPHY
* X. Cui and Y. Wang contributed equally to this work. Address for reprint requests and other correspondence: H. Yang, School of Biotechnology and Biomolecular Sciences, Univ. of New South Wales, Sydney, NSW, 2052, Australia (e-mail:
[email protected]). http://www.ajpendo.org
and appears to control adipogenesis through modulating the synthesis of phospholipids; caveolin-1/cavin has a defined role in caveola formation. In contrast, little is known about the molecular function of seipin despite the fact that CGL2/BSCL2 represents the most severe form of human lipodystrophy (2). Two independent studies have confirmed the role of seipin in adipogenesis in cultured cells (6, 21). Recently, we have generated the first murine model [seipin knockout (SKO) mouse] of seipin research by targeted deletion of seipin from the mouse genome (7). The SKO mice display an over 80% reduction in adipose tissue mass, severe hepatic steatosis, and insulin resistance. The loss of adipose tissue in SKO mice is not as severe as that in human BSCL2 patients, but the essential in vivo role of seipin in adipocyte development is nonetheless apparent. It has been demonstrated that the human BSCL2/seipin gene has three transcripts: 1.6, 1.8, and 2.2 kb. The 1.8-kb mRNA is exclusively expressed in brain and testis, but the other two mRNA transcripts are ubiquitously expressed (26). Recent studies have revealed that seipin is highly expressed in adipose tissue and is strongly induced during adipocyte differentiation (6, 21). The basal level of seipin expression in the early stages of 3T3-L1 differentiation appears to be critical to adipocyte development, as PPAR␥ expression was inhibited within 24 h of adipogenic induction in seipin knockdown cells (6). Given the critical role of seipin in early stages of adipocyte differentiation, it is somewhat surprising that seipin expression is not significantly induced until almost 4 days after induction of adipogenesis, much later than that of PPAR␥ (6). It is therefore likely that seipin may also play important roles in maintaining the normal function of mature adipocytes. Seipin and its orthologs have also been found to regulate the cellular dynamics of lipid droplets (LDs) (5, 10, 23). In yeast, deletion of FLD1 (the yeast seipin homolog) resulted in the clustering of small LDs and the formation of supersized LDs, as well as increased triacylglycerol (TAG) accumulation (10). The absence of Drosophila seipin reduced fat body formation but increased TAG synthesis and LD formation in the salivary gland (24). In HeLa and NIH3T3 cells, knocking down seipin increased TAG synthesis and LD proliferation, whereas overexpressing tagged forms of seipin inhibited TAG accumulation (9). These studies established a strong link between seipin and glycerolipid metabolism; however, little is known about the molecular function of seipin. It has been suggested that Fld1p/ seipin may regulate the metabolism of phospholipids (PA in
0193-1849/12 Copyright © 2012 the American Physiological Society
E705
E706
SEIPIN IN MATURE ADIPOCYTES
particular) or may play a structural role in organizing the biogenesis of the LDs (4, 8). Despite the important role of seipin in both LD formation and adipocyte differentiation (two important aspects of human lipid storage and obesity), the in vivo function of seipin in mature adipocytes, where it is most highly expressed, remains to be elucidated. We have generated and characterized a transgenic mouse model with enhanced seipin expression in mature adipocytes. Our results suggest an important yet surprising role for seipin in regulating lipid storage in mature adipocytes in vivo. MATERIALS AND METHODS
Animals. Human seipin cDNA (hSeipin) was amplified by polymerase chain reaction (PCR) from human cDNA (GenBank, BC012140) by using human seipin-specific primers (forward 5=TATGCGGCCGCTCAGGAACTAGAGCAGG-3=, reverse 5=-TATGATATCATGGTCAACGACCCTCC-3=). The PCR product was subcloned into the pEASY-Blunt Simple Cloning Vector (TransGen Biotech, Beijing, China). A 1.2-kb NotI/EcoRV fragment containing the full-length hSeipin cDNA was cloned downstream of the 5.4-kb aP2-promoter with NotI/Sma1 (Addgene plasmid 11424). The transgene as a Kpn1/SacII fragment was isolated from the vector and purified. Transgenic mice were produced by oocyte injection using FVB mice-derived eggs. Transgenic (Tg) mice were identified by PCR (forward 5=-ACTTCTTCGCGTTCGGTGATGC-3=, reverse 5=CTGGATGCGCTTGCTGTGGAT-3=) and confirmed by Southern blotting. Copy number was determined by qPCR. All experiments involving mice were approved by the Institutional Animal Care Research Advisory Committee of the National Institute of Biological Science (NIBS) and Animal Care Committee of Peking University Health Science Center. Animals were housed and allowed free access to tap water and standard laboratory chow. The aP2hSeipin-Tg mice used in these studies were back-crossed with C57BL/6 for two to seven generations. All mice were maintained on a 12:12-h light-dark cycle and were fed ad libitum with regular mouse chow (10% of kilocalories from fat) or Western type diet (40% kilocalories from fat for 10 wk). Blood samples were taken from 4-h-fasted mice. For tissue sample collection, mice were fasted for 4 h, and tissue samples were then harvested, snap-frozen in liquid nitrogen, and stored at ⫺80°C for real-time PCR and Western blot analysis. RNA isolation and quantitative real-time PCR. Total RNA from adipose and other tissues were extracted using Tri Reagent (Molecular Research Center), and first-strand cDNA was generated using an RT kit (Invitrogen). Quantitative real-time PCR was performed using
primer sets as shown in Table 1. Amplifications were performed in 35 cycles using an opticon continuous fluorescence detection system (MJ Research) with SYBR Green fluorescence (Molecular Probes, Eugene, OR). Each cycle consisted of heating denaturation for 30 s at 94°C, annealing for 30 s at 55°C, and extension for 30 s at 72°C. All samples were quantitated using the comparative CT method for relative quantitation of gene expression, normalized to GAPDH (12). Western blot analysis. Tissues for western blot analysis of total lysate were lysed in RIPA buffer containing 25 mmol/l Tris (pH 7.4), 50 mmol/l NaCl, 0.5% sodium deoxycholate, 2% NP-40, 0.2% SDS, and Complete protease inhibitor cocktail (Roche). The protein content of cell lysates was determined using a bicinchoninic acid protein assay kit (Pierce). Mouse adipose tissue homogenate (20 g protein extracted by RIPA solution) was subjected to electrophoresis on 10% SDS-PAGE and transferred to nitrocellulose membranes (Sigma, St. Louis, MO). After blocking with 5% bovine serum albumin for 1 h, the membrane was probed with 1:2,000 mouse anti-seipin polyclonal antibody (Abnova, cat. no. H00026580-A02), followed by horseradish peroxidaseconjugated secondary antibody (goat anti-mouse, 1:5,000). The reaction was detected by chemiluminescence and exposed to Kodak X-Omat film (Kodak, Rochester, NY). Seipin protein level was normalized to that of GAPDH by using a goat anti-mouse monoclonal antibody. Antibodies, against GAPDH (Millipore), -tubulin (Santa Cruz Biotechnology), adipose triglyceride lipase (ATGL), perilipin, and adipose differentiation-related protein (ADRP) were also used for Western blot analysis. Densitometric reading of relative levels of the indicated proteins in Western blot analyses involved the use of NIH Image software. Plasma lipid analysis. Plasma total cholesterol (TC), total triacylglycerols (TAG), and glucose were determined y using enzymatic methods (Sigma kits). Serum insulin, leptin, and adiponectin were measured by ELISA (Linco Research, St. Charles, MO), and free fatty acids were measured using a free fatty acid kit (Wako). Glucose tolerance tests. For glucose tolerance tests, animals were fasted overnight for 12 h, and a basal blood sample was taken, followed by intraperitoneal injection of glucose (2 g/kg body wt; Abbott). Blood samples were drawn at 15, 30, 60, and 120 min after the injection for measurement of glucose (Sigma kits). For differences in plasma glucose levels during the glucose tolerance tests, an ANOVA was performed and significance determined using Tukey’s post hoc test. A P value ⬍ 0.05 was considered statistically significant. Food consumption measurement and core body temperature. For food consumption measurement, mice were caged individually. The amount of normal diet consumed was carefully monitored at 10:00 AM local time every morning for a period of 2 wk. Core body
Table 1. List of primers used and their sequences SREBP-1c PPAR␥ FAS SCD1 ATGL Leptin Adiponectin ADRP
Forward Reverse Forward Reverse Forward Reverse Forward Reverse Forward Reverse Forward Reverse Forward Reverse Forward Reverse
TGGAGACATCGCAAACAAG GGTAGACAACAGCCGCATC GACCACTCGCATTCCTTT CCACAGACTCGGCACTCA GGGTCTATGCCACGATTC GTGTCCCATGTTGGATTTG TGACCTGAAAGCCGAGAA ATGTGCCAGCGGTACTCA CGTCACCTGTGCCTTACTC GGGCTCAAACTCCCAAAC CACAGTCTGGAGCGAAGG CACAATCTGGGAACAAGC CTCCTGCTTTGGTCCCTC GCCAGTGCTGCCGTCATA CACAAACGATGACCCTC GCATGTTGCGACTGC
LPL ACC DGAT HSL TNF-␣ Resistin Perilipin TIP47
Forward Reverse Forward Reverse Forward Reverse Forward Reverse Forward Reverse Forward Reverse Forward Reverse Forward Reverse
See text for definitions. AJP-Endocrinol Metab • doi:10.1152/ajpendo.00237.2011 • www.ajpendo.org
ACTAGGTCCCACAGGACTG GACTTCCAGAAGTAACCAACTTTG CCAGACCCTTTCTTCAGC TTGTCGTAGTGGCCGTTC ATCTGAGGTGCCATCGTC ATGCCATACTTGATAAGGTTCT GACTCACCGCTGACTTCCT CTGTCTCGTTGCGTTTGTAG CTGTGAAGGGAATGGGTGTT CAGGGAAGAATCTGGAAAGGTC TCCTTGTCCCTGAACTGC ACGAATGTCCCACGAGCC CCATCTCCGCAGTGACCT GCTGACGCCTCGGTTTT TCCTAAGCCTGATGGAAACT GCACCTGGTCCTTCACATT
E707
SEIPIN IN MATURE ADIPOCYTES
temperature was measured using a rectal probe (Thermalet TH-5) at ambient room temperature. Food and water were provided ad libitum. Histological studies and determination of adipocyte cell size. A segment of liver was cut and cryostat sectioned at a thickness of 7 m onto poly-L-lysine slides for lipid deposition analysis by Oil red O and hematoxylin staining. Adipose tissue was prepared and subjected to hematoxylin and eosin (H&E) staining, and adipocyte area was quantified using computer-assisted planimetry (Image-Pro Plus, Media Cybernetics). The non-TG littermates were used as controls. Liver lipid analysis. Approximately 100 mg of of liver (wet weight) was weighed and homogenized in 1 ml of PBS. Lipids were extracted as described by Folch et al. (13) and dissolved in 100 l of 3% Triton X-100. The determination of TAG was carried out using enzymatic methods as described above. Magnetic resonance imaging. For MRI acquisition, anesthesia was induced by inhalation of a mixture of oxygen and 5% isoflurane and maintained by a mixture of oxygen containing 1–2% isoflurane. Mice were positioned and immobilized prone inside the tomograph with either the thoracic or the abdominal region in the center of the field of view (FOV). All MRI experiments were carried out using a 7T small animal magnetic resonance tomograph with Bruker Pharmascan 7.0 T/16 cm spectrometer equipped with a mini imaging gradient coil system (gradient strength, 375 mT/m) and a 1H transmit-receive quatrature coil with 31-mm inner diameter. T1-weighted (T1W) images were acquired with a respiratory-gated spin echo sequence, FOV 3.5 cm ⫻ 3.5 cm, matrix size 256 ⫻ 256, slice thickness 2.0 mm, repetition time (TR) 500 ms and echo times (TE) 15 ms, and a number of repetitions (NEX) of 4. The T1W images were used to study the distribution of fat stores. MRI images for adipose tissue area (at renal hilum) and volume (1 cm thick transaxial slice, total of 5 2-mm-thick slices, from inferior pole of kidney) were quantified using computerassisted planimetry (Image-Pro Plus, Media Cybernetics). Lipolysis. Lipolysis in isolated primary adipocytes was measured essentially as described (30). Gonadal fat pads were removed from anesthetized animals in nonsurvival surgery, under aseptic conditions, minced, and digested with collagenase I to isolate the stromal vascular cells (composed of preadipocytes and other cell types), and the adipocytes were allowed to float by gravity. Aliquots of the floating
adipocyte layer were washed twice with DMEM and then incubated for 1 h in serum-free DMEM (without phenol red) containing 2% fatty acid-free BSA in the presence or absence of 1 M isoproterenol. Conditioned medium was then removed for glycerol analyses. Glycerol released into the primary adipocytes culture medium served as an index of lipolysis. The level of glycerol was measured by colorimetric assay (GPO Trinder reaction) from absorption at 490 nm. Lipolysis data were expressed as micromoles per milligram of protein. Adipocytes were rinsed with PBS and harvested in lysis buffer for the analysis of protein content as described above. Statistical analysis. All data are presented as means ⫾ SE. Statistical comparison between groups was performed using Student’s t-test or two-way ANOVA. A value of P ⬍ 0.05 was considered statistically significant. RESULTS
Generation of aP2-hSeipin-Tg mice. Human and mouse seipin are both predicted to have long and short isoforms due to alternative translation initiation sites (19). The long isoforms of human (hSeipinL, 462 aa) and mouse seipin (mSeipinL, 443 aa) share ⬃84% identity. The short isoforms (hSeipinS, 398 aa vs. mSeipinS, 383 aa) share ⬃88% identity. It is not known which isoform(s) of seipin is translated in mouse adipose tissue. To investigate the physiological function of seipin in vivo, we generated Tg mice expressing the short isoform of human seipin (hSeipinS) exclusively in adipose tissue by using the adipocyte-specific aP2 promoter/enhancer (Fig. 1A) (22). To confirm the expression of the transgene, we tested a number of seipin antisera and found that a polyclonal antibody from Abnova (cat. no. H00026580-A02) could specifically detect mouse seipin and overexpressed human seipin (hSeipin) (Fig. 1B). The bands detected by the antibody in wild-type (Non-Tg) mice could be endogenous mouse seipin, as the same antibody failed to detect the same band in WAT from the SKO mice. Expression of total seipin (human and mouse seipin) in white
Fig. 1. Generation and characterization of transgenic (Tg) mice overexpressing human seipin (hSeipin) in adipose tissue. A: diagram of construct used to generate aP2-hSeipin-Tg mice. The lengths of each segment in the transgene are indicated below the construct. B and D: determination of Tg expression of seipin protein and mRNA levels in white adipose tissue (WAT), brown adipose tissue (BAT), liver, kidney, and/or heart by Western blot (B) or RT-PCR (D, n ⫽ 6; *aP2-hSeipin-Tg vs. Non-Tg, ***P ⬍ 0.001). C: densitometric reading of relative seipin levels from WAT in Western blot analysis (as in B) was quantitated. Relative protein level in Non-Tg mice was designated as 1.0 (male, n ⫽ 3, *aP2-hSeipin-Tg vs. Non-Tg, *P ⬍ 0.05). All values are means ⫾ SE.
AJP-Endocrinol Metab • doi:10.1152/ajpendo.00237.2011 • www.ajpendo.org
E708
SEIPIN IN MATURE ADIPOCYTES
adipose tissue (WAT) of the Tg mice was ⬃1.5-fold greater than that of the Non-Tg mice both at the level of protein (Fig. 1, B and C) and mRNA (Fig. 1D). The increase in human seipin expression was most significant in WAT; other tissues we examined included brown adipose tissue, liver, heart, and kidney (Fig. 1D). The Tg mice are referred to as the aP2hSeipin-Tg mice. The aP2-hSeipin-Tg mice were viable throughout adulthood without apparent abnormalities in growth. On chow diet, the aP2-hSeipin-Tg mice demonstrated normal rates of growth. No difference in total body weight, plasma chemistry, or glucose tolerance was observed between aP2-hSeipin-Tg mice and Non-Tg littermates (Fig. 2). Decreased adipose tissue mass in aP2-hSeipin-Tg mice. Previous studies showed that loss-of-function mutations of BSCL2/seipin caused complete loss of adipose tissue, suggesting that BSCL2 plays an important role in the development of adipocytes (3, 7). However, the exact function of seipin in mature adipocytes remains to be elucidated. The aP2-hSeipin-Tg mice expressing hSeipin specifically in mature adipocytes allow us to examine the role of seipin in postdifferentiation states, as the expression of endogenous seipin is greatly increased in mature adipocytes compared with preadipocytes (6). We first measured the mass of adipose depots by MRI via Bruker Pharmascan 7.0 T/16 cm spectrometer. Visual comparison of MRI images of 6-mo-old mice at the renal hilum of
Fig. 2. Unaltered body weight, plasma lipid levels, food consumption, body temperature, and glucose tolerance. A and B: body weight of female (n ⫽ 10) and male (n ⫽ 11) mice from 4 to 20 wk. C: plasma levels of triacylglycerol (TAG), total cholesterol (TC), and glucose (Glu) from 8-mo-old Non-Tg and aP2-hSeipin-Tg mice after 4-h fast (male n ⫽ 8). D and E: unaltered food consumption and body temperature in Non-Tg and aP2-hSeipin-Tg mice (male n ⫽ 6). F: unchanged glucose tolerance test of Tg mice and Non-Tg littermates (male n ⫽ 10).
Non-Tg and Tg mice demonstrated that the mass of adipose tissue was considerably less in the Tg group (Fig. 3A). Interestingly, the area and volume of adipose tissue was greatly decreased in the Tg mice (Fig. 3, C and D; n ⫽ 6; *P ⬍ 0.05). These results were further validated by the measurement of fat depots in individual dissected tissues from euthanized animals (Table 2). Examination of major WAT depots in 8-mo-old Tg mice revealed an ⬃30% decrease in adipose tissue wet weight, including subcutaneous, mesenteric, and retroperitoneal fat pads, and total fat weight (Table 2, n ⫽ 6) as well as in adipocyte cell size (hSeipin-Tg vs. Non-Tg mice: 2,485 ⫾ 223 and 3,635 ⫾ 165 m2; Fig. 3, B and E, n ⫽ 6; **P ⬍ 0.01). Although statistically insignificant, the gonadal and inguinal WAT of the Tg mice was also slightly reduced compared with Non-Tg littermates (Table 2; hSeipin-Tg vs. Non-Tg mice: 578.8 ⫾ 52.7 vs. 873.9 ⫾ 158.5 mg, and 44.5 ⫾ 4.4 vs. 55.8 ⫾ 5.4 mg, respectively). There was no sex-specific difference in adipose tissue mass. Increased lipolysis in aP2-hseipin-tg mice. To determine the mechanism for the decrease in adipose tissue mass, we investigated whether there were any changes in lipolysis and/or lipogenesis. Spontaneous glycerol release (basal lipolysis) was found to be significantly increased (⬃35%) in the Tg mice (Fig. 4A, n ⫽ 4). Isoprotenerol-stimulated lipolysis was also significantly higher in the aP2-hSeipin-Tg mice. Western blot analysis showed a significantly increase (⬃1.5-fold) in level of
F
AJP-Endocrinol Metab • doi:10.1152/ajpendo.00237.2011 • www.ajpendo.org
SEIPIN IN MATURE ADIPOCYTES
E709
Fig. 3. Magnetic resonance imaging (MRI) and histological analysis of Non-Tg and aP2-hSeipin-Tg mice. A: abdominal MRI of aP2-hSeipin-Tg mice with decreased subcutaneous and intra-abdominal fat compared with Non-Tg littermates with normal fat distribution. Fat is shown white in these MRI images (scale bar, 1 cm). B: histology of WAT from 8-mo-old Non-Tg and aP2-hSeipin-Tg mice (scale bar, 20 m). Note: adipocytes of WAT from Tg mice were markedly diminished (hematoxylin & eosin stain). C: MRI measurements of adipose fat area of renal hilum. D: adipose tissue volume of a 1-cm-thick slice transaxial from inferior pole of kidney. E: cell size quantitation of gonadal WAT from Non-Tg and hSeipin-Tg mice on chow diet. (n ⫽ 6: male 3, female 3; *aP2-hSeipin-Tg vs. Non-Tg, *P ⬍ 0.05, **P ⬍ 0.01).
ATGL in the Tg mice, which is a TAG lipase primarily catalyzing the first step of TAG hydrolysis (Fig. 4, B and C) (29). Perilipin, which is required for inducing the translocation of hormone-sensitive lipase (HSL) from the cytosol to the LDs of adipocytes and triggering the lipolytic reaction, also had an approximately twofold upregulation (Fig. 4, B and C) (27). Another lipid droplet protein, ADRP, which is involved in the Table 2. Fat pad weights and phenotypic comparison of 8mo-old Non-Tg and aP2-hSeipin-Tg mice Body Weight, g Subcutaneous fat, mg Inguinal fat, mg Mesenteric fat, mg Retroperitoneal fat, mg Gonadal fat, mg Total fat, mg Liver, mg Heart, mg Kidney, mg
Non-Tg
aP2-hSeipin-Tg
P
30.5 ⫾ 1.5 313.8 ⫾ 30.6 55.8 ⫾ 5.4 335.4 ⫾ 32.7 251.7 ⫾ 26.7 873.9 ⫾ 158.5 1,830.6 ⫾ 223.4 848.7 ⫾ 38.4 115.1 ⫾ 7.0 141.5 ⫾ 7.3
29.7 ⫾ 1.2 211.1 ⫾ 19.1 44.5 ⫾ 4.4 221.0 ⫾ 33.0 143.2 ⫾ 21.2 578.7 ⫾ 52.7 1,198.6 ⫾ 104.9 1025.7 ⫾ 26.3 109.3 ⫾ 2.5 148.7 ⫾ 6.2
0.69 0.02* 0.13 0.03* 0.008** 0.10 0.03* 0.004** 0.5 0.4
All values are means ⫾ SE. Tg, transgenic; hSeipin, human seipin. Eightmonth-old male mice were fed standard rodent chow. Statistical analysis was done with two-tailed unpaired Sstudent’s t-test; n ⫽ 6. *P ⬍ 0.05, **P ⬍ 0.01.
formation and maturation of LDs in adipocytes (27), had an ⬃1.5-fold increase in the aP2-hSeipinTg mice (Fig. 4, B and C). Similar changes for ATGL, perilipin, and ADRP were also observed at the level of transcription (Fig. 4D). Notably, the mRNA level of HSL was upregulated almost threefold. The increase in ATGL and HSL may be responsible for the enhanced lipolysis in the Tg mice (Fig. 4A). We also examined the expression of adipokines secreted by adipocytes (Fig. 4D, n ⫽ 6). Despite a reduction of WAT mass, the level of expression for some of the adipokines was elevated (15). Leptin, which hypothalamically modulates body weight, food intake, and fat stores, had a twofold higher expression compared with Non-Tg littermates. The level of another antiobesity adipokine, adiponectin, was significantly increased (⬃2.3-fold) as well. Although the proinflammatory cytokine tumor necrosis factor-␣ (TNF-␣) and resistin had 1.8- and 1.3-fold increases, respectively, the differences were not statistically significant. Surprisingly, the plasma levels of leptin and adiponectin were not significantly changed (Fig. 4, E and F). This may have been due to reduced secretion of the adipokines in the Tg mice. The reduced WAT mass in the aP2-hSeipin-Tg mice could also have been caused by reduced lipogenesis. We used qRT-
AJP-Endocrinol Metab • doi:10.1152/ajpendo.00237.2011 • www.ajpendo.org
E710
SEIPIN IN MATURE ADIPOCYTES
Fig. 4. Increased lipolysis in aP2-hSeipin-Tg mice. A: in vitro lipolysis rates (glycerol release) in gonadal adipose depots of 3-mo-old Non-Tg and aP2-hSeipin-Tg mice. Basal refers to no isoproterenol (ISO, 1 M) treatment (male, n ⫽ 4; *aP2-hSeipin-Tg vs. Non-Tg, *P ⬍ 0.05, **P ⬍ 0.01). B and C: Western blot analysis of ATGL, perilipin, and ADRP (B) and densitometric quantitation (C) from WAT of Non-Tg and hSeipin-Tg mice (male, n ⫽ 3; *aP2-hSeipin-Tg vs. Non-Tg, *P ⬍ 0.05). D: relative mRNA expression of HSL, ATGL, TNF-␣, leptin, resistin, adiponectin, perilipin, ADRP, and TIP47 in WAT of Non-Tg and hSeipin-Tg mice (male, n ⫽ 6; *aP2-hSeipin-Tg vs. Non-Tg, *P ⬍ 0.05). E and F: plasma leptin (E) and adiponectin (F) level (male, n ⫽ 6). G: real-time RT-PCR analysis of adipogenic genes LPL, PPAR␥, SREBP-1c, FAS, ACC1, SCD-1, and DGAT expression in WAT taken from 8-mo-old Non-Tg and aP2-hSeipin-Tg mice (male, n ⫽ 6 for each genotype). See text for definitions.
PCR to examine the expression of key genes responsible for lipogenesis in adipose tissue, including sterol regulatory element-binding protein-1c (SREBP-1c), acetyl-coA carboxylase-1 (ACC1), fatty acid synthase (FAS), diacylgycerol acyltransferase-2 (DGAT2), and stearoyl-CoA desaturase-1 (SCD1) and found no significant change in mRNA levels (Fig. 4G, n ⫽ 6). This is consistent with a recent study using cultured mammalian cells (9). At the level of transcription, the short isoform of seipin does not seem to increase lipogenesis in mouse WAT. However, lipogenesis could still be affected since seipin may directly regulate the flux of fatty acids. Abnormal lipid accumulation in the liver of aP2-hSeipin-Tg mice. Changes in adipose tissue mass are frequently associated with alterations in whole body lipid homeostasis. We therefore measured the amount of TAG in the liver and skeletal muscle of hSeipin-Tg mice and examined the liver morphology by oil red O staining to determine whether lipid homeostasis was affected by hSeipin expression in mature adipocytes. Liver TAG content of the aP2-hSeipin-Tg mice was ⬃40% higher than those of Non-Tg littermates (hSeipin-Tg vs. Non-Tg mice: 4.0 ⫾ 0.4 and 2.8 ⫾ 0.3 mg/g; n ⫽ 7, *P ⬍ 0.05; Fig. 5, A and B), whereas no significant difference in the amount of skeletal muscle TAG was observed between hSeipin-Tg and Non-Tg mice (3.2 ⫾ 0.6 vs. 2.9 ⫾ 0.5 mg/g, n ⫽ 5; Fig. 5B). Morphologically, the liver of Tg mice contained more LDs than that of Non-Tg mice (Fig. 5A). The increased accumulation of TAG in the liver could result from reduced lipid storage in the adipose tissue. Indeed, the nonesterified fatty acid
(NEFA) of Tg mice was slightly but significantly increased (hSeipin-Tg vs. Non-Tg mice, n ⫽ 10: 2.0 ⫾ 0.1 vs. 2.3 ⫾ 0.07 meq/l fasted and 1.7 ⫾ 0.1 vs. 1.4 ⫾ 0.1 meq/l fed) when fasted or fed (Fig. 5C); in contrast, hSeipin overexpression did not alter the levels of TAG or cholesterol in the plasma of mice from either genotype (Fig. 2). These results indicate that the increased levels of plasma NEFA and liver TAG in the aP2hSeipin-Tg mice may have resulted from the increased lipolysis in the adipose tissue. aP2-hSeipin-Tg mice are resistant to high-fat-induced obesity but not protected from impairment in glucose metabolism. We next examined the response of aP2-hSeipin-Tg mice to high-fat diet. For this purpose, the aP2-hSeipin-Tg mice were back-crossed into the C57BL/6 background. On chow diet, the loss of WAT was evident for aP2-hSeipin-Tg mice of the C57BL/6 background (Fig. 6A, n ⫽ 8), similar to those of a mixed background (Table 2). The aP2-hSeipin-Tg mice of C57BL/6 background were then subjected to a high-fat diet for 10 wk. As shown in Fig. 6B, the Tg mice were more resistant to high-fat-induced weight gain. Consistent with this, the Tg mice looked thinner, with less subcutaneous and gonadal fat (Fig. 6C, 948.0 ⫾ 130.6 vs. 530.1 ⫾ 54.5 mg and 125.6 ⫾ 15.9 vs. 71.7 ⫾ 2.9 mg, respectively, n ⫽ 8). The Tg mice on high-fat diet did not demonstrate significantly improved glucose clearance in a glucose tolerance test (Fig. 6D, n ⫽ 8, Non-Tg-HF vs. Tg-HF; P ⫽ 0.99, 0.99, 0.69, 0.38, and0.24 for different time points, respectively). No significant differences in plasma lipids and glucose were observed between the
AJP-Endocrinol Metab • doi:10.1152/ajpendo.00237.2011 • www.ajpendo.org
E711
SEIPIN IN MATURE ADIPOCYTES
Fig. 5. Abnormal lipid accumulation in the liver of aP2-hSeipin-Tg mice. A: images of liver sections stained with Oil red O. Nuclei were counterstained with hematoxylin. Red droplets represent lipid droplets (LDs) in the liver (scale bar, 50 m). B: Liver (male, n ⫽ 7) and skeletal muscle (male, n ⫽ 5) TAG contents from 8-mo-old Tg mice and non-Tg littermates. C: plasma NEFA of fasted or fed aP2-hSeipin-Tg and Non-Tg mice (male, n ⫽ 10). Each bar represents the mean ⫾ SE values (*aP2-hSeipin-Tg vs. Non-Tg, *P ⬍ 0.05).
Non-Tg and aP2-hSeipin-Tg mice on a high-fat diet (Fig. 6, E–G, n ⫽ 8). Surprisingly, the liver weight of the Tg mice decreased on high-fat diet (Fig. 6H, 1,858.9 ⫾ 209.1 vs. 1,317.7 ⫾ 74.5 mg, n ⫽ 8). DISCUSSION
Recent studies have revealed important roles of seipin/BSCL2 in lipid storage at both the cellular level (lipid droplets) and the whole body level (adipose tissue development) (2, 10, 23). We and others (6, 7, 21) have established the essential role for seipin in adipogenesis both in vitro and in vivo. However, seipin is upregulated primarily at later stages of adipocyte development, and its role in mature adipocytes remains to be elucidated. There are two isoforms of seipin (398 and 462 aa, respectively) due to different translation initiation sites. Here, we overexpressed the short human seipin specifically in the adipose tissue and generated aP2-hSeipinS-Tg mice. We found that overexpression of hSeipinS in mature adipocytes significantly reduced the size of adipocytes and LDs therein, causing a reduction in adipose tissue mass. This effect may be partially due to enhanced lipolysis as a result of seipin overexpression. Our results therefore suggest that seipin functions to inhibit lipid storage in mature adipocytes. What is the function of seipin in fully differentiated adipocytes, where its expression level can be upregulated almost 70-fold (6)? Our results here suggest that seipin may function to reduce lipid storage, in part by promoting lipolysis in mature adipocytes (Figs. 4 and 5). This would seem highly counterintuitive due to the lipodystrophic condition (no adipocytes and therefore no lipid storage) caused by loss of seipin. How could deletion and overexpression of seipin both reduce lipid storage in adipocytes? Lipid storage in adipose tissue may be regulated by at least two means: one is to control the differentiation of adipocytes, which involves the activation of a transcriptional cascade (25); the other is to directly control lipogenesis. As discussed earlier, seipin has a clear
role early during adipocyte differentiation, but this alone does not explain its function in fully differentiated adipocytes,where it is most highly expressed. In this study, seipin was expressed under the control of the aP2 promoter, which overexpresses seipin only in differentiated mature adipocytes. Our results clearly suggest that seipin can inhibit lipid storage in mature adipocytes. We hypothesize that seipin may have one biochemical function that is required for the proper induction of the adipogenic program, and that is also needed to limit lipid storage in mature adipocytes. Limiting TAG synthesis (lipogenesis) and storage in adipocytes may serve to protect them from “overdifferentiation”, i.e., unlimited TAG synthesis and enlargement of LDs, and to prevent them from exceeding their storage capacity (17). Uncapped synthesis of TAG and growth of LDs would compromise other cellular functions, triggering cell death and inflammation. Like seipin, both Insig1 and -2 are highly upregulated in fully differentiated adipocytes to limit lipogenesis and TAG storage (17). Consistent with a role for seipin in inhibiting lipid storage, deletion of the yeast seipin enhanced lipogenesis and lipid storage and reduced lipolysis (10, 11, 28). It was proposed that seipin/Fld1p might be required for sequestering Tgl3 (a yeast TAG lipase with a patatin domain) on LD surface to facilitate lipolysis (28). In addition, seipin deletion in Drosophila causes accumulation of ectopic LDs in the salivary gland, a nonadipose tissue (24). Importantly, this accumulation can be suppressed by expressing dSeipin specifically within the salivary gland, suggesting a tissue-autonomous role for seipin in inhibiting lipogenesis. Moreover, we have recently reported that overexpression of seipin in cultured mammalian cells reduces TAG storage (9). Therefore, the results of this paper, i.e., that seipin overexpression reduces lipid storage in adipose tissue possibly by enhancing lipolysis, are consistent with the latest results from other experimental systems.
AJP-Endocrinol Metab • doi:10.1152/ajpendo.00237.2011 • www.ajpendo.org
E712
SEIPIN IN MATURE ADIPOCYTES
Fig. 6. aP2-hSeipin-Tg mice are resistant to diet-induced obesity. A: fat pad weight of Non-Tg and ap2-hSeipin-Tg mice on chow or high-fat (HF) diet. B: body weight curve of Non-Tg and aP2-hSeipin-Tg mice on chow or HF diet. C: whole body view and diminished subcutaneous and epididymal white fat (black arrow)/testis from aP2-hSeipin-Tg mice (right) vs. Non-Tg mice (left) after 10-wk HF diet. D: impaired glucose tolerance test under HF diet but no difference between Non-Tg and aP2-hSeipin-Tg mice. E–G: plasma levels of TAGs, TC, and glucose, respectively, from 4-h-fasted Non-Tg and aP2-hSeipin-Tg mice after 10-wk HF diet. H: tissue weight of Non-Tg and aP2-hSeipin-Tg mice on chow or HF diet. Each bar represents the mean ⫾ SE values (male, n ⫽ 8 for each group; *chow vs. HF, *P ⬍ 0.05, **P ⬍ 0.01, ***P ⬍ 0.001, #Tg vs. Non-Tg, #P ⬍ 0.05, ##P ⬍ 0.01).
Similar to the SKO lipodystrophic model, ectopic liver fat deposition was also found in the aP2-hSeipin-TG mice, albeit to a lesser extent. Likewise, the aP2-hSeipin-TG mice were protected from diet-induced weight gain but not from glucose deregulation. The lack of weight gain is probably due to limited fat storage in the adipose tissue, and the deregulation of glucose metabolism probably results from lipid overflow from adipose tissue to muscle and liver, causing lipotoxicity and insulin resistance. There are a number of limitations to the current study, wherein only the short isoform of human seipin was overexpressed. It would be interesting to also test the long isoform in the future. The loss of WAT mass of the aP2-hSeipin-Tg mice could also have resulted from reduced lipogenesis, but we were not able to directly measure TAG synthesis using isolated adipocytes. Although there were no significant changes in the expression of lipogenic genes in the transgenic mice (Fig. 4), seipin may directly control the flux of substrates (e.g., fatty acids) used for TAG synthesis as shown recently (9). The increase in lipolysis observed in the transgenic WAT could be attributable to the production of more lipases, but we cannot rule out other possibilities. For instance, the smaller droplets in the transgenic mice provided more surface area for
lipases to act, thereby leading to increased lipolysis. Nonetheless, we have generated the first transgenic mouse model for seipin research, and we have uncovered a novel role of seipin in limiting lipid storage in mature adipocytes in vivo. Our findings are consistent with recent results from other experimental systems and will lead to a better understanding of the molecular and physiological function of seipin, a key protein in adipogenesis and cellular lipid homeostasis. ACKNOWLEDGMENTS We thank the members of the Yang and Liu laboratory for helpful discussions. We thank Dr. Peng Li for providing reagents and Dr. Jinkuk Choi for critical reading of this manuscript. H. Yang is a Future Fellow of the Australian Research Council. GRANTS This work was supported in part by National Natural Science Foundation of the People’s Republic of China (nos. 30821001 and 30930037), Major National Basic Research Program of the People’s Republic of China (G2006CD503801) to G. Liu, National Natural Science Foundation of the People’s Republic of China (no. 30971102) to Y. Wang, and a research grant from the National Health and Medical Research Council of Australia (no. 568725) to H. Yang.
AJP-Endocrinol Metab • doi:10.1152/ajpendo.00237.2011 • www.ajpendo.org
SEIPIN IN MATURE ADIPOCYTES DISCLOSURES No conflicts of interest, financial or otherwise, are declared by the author(s). AUTHOR CONTRIBUTIONS Author contributions: X.C., Y.W., G.X., S.J., G.L., L. Zhao, and H.Y. conception and design of research; X.C., Y.W., L.M., W.F., J.D., X.P., and L. Zhang performed experiments; X.C., Y.W., J.D., and X.P. analyzed data; X.C., Y.W., J.D., G.X., X.P., S.J., G.L., L. Zhao, and H.Y. interpreted results of experiments; X.C. and W.F. prepared figures; X.C., W.F., J.D., G.X., X.P., S.J., L. Zhao, and H.Y. drafted manuscript; X.C., Y.W., G.L., and H.Y. edited and revised manuscript; X.C., G.L., and H.Y. approved final version of manuscript. REFERENCES 1. Agarwal AK, Arioglu E, De Almeida S, Akkoc N, Taylor SI, Bowcock AM, Barnes RI, Garg A. AGPAT2 is mutated in congenital generalized lipodystrophy linked to chromosome 9q34. Nat Genet 31: 21–23, 2002. 2. Agarwal AK, Garg A. Genetic basis of lipodystrophies and management of metabolic complications. Annu Rev Med 57: 297–311, 2006. 3. Agarwal AK, Garg A. Seipin: a mysterious protein. Trends Mol Med 10: 440 –444, 2004. 4. Binns D, Lee S, Hilton CL, Jiang QX, Goodman JM. Seipin is a discrete homooligomer. Biochemistry 49: 10747–10755, 2010. 5. Boutet E, El Mourabit H, Prot M, Nemani M, Khallouf E, Colard O, Maurice M, Durand-Schneider AM, Chretien Y, Gres S, Wolf C, Saulnier-Blache JS, Capeau J, Magre J. Seipin deficiency alters fatty acid Delta9 desaturation and lipid droplet formation in Berardinelli-Seip congenital lipodystrophy. Biochimie 91: 796 –803, 2009. 6. Chen W, Yechoor VK, Chang BH, Li MV, March KL, Chan L. The human lipodystrophy gene product BSCL2/seipin plays a key role in adipocyte differentiation. Endocrinology, 2009. 7. Cui X, Wang Y, Tang Y, Liu Y, Zhao L, Deng J, Xu G, Peng X, Ju S, Liu G, Yang H. Seipin ablation in mice results in severe generalized lipodystrophy. Hum Mol Genet, 2011. 8. Fei W, Du X, Yang H. Seipin, adipogenesis and lipid droplets. Trends Endocrinol Metab, 2011. 9. Fei W, Li H, Shui G, Kapterian TS, Bielby C, Du X, Brown AJ, Li P, Wenk MR, Liu P, Yang H. Molecular characterization of seipin and its mutants: implications for seipin in triacylglycerol synthesis. J Lipid Res 52: 2136 –2147, 2011. 10. Fei W, Shui G, Gaeta B, Du X, Kuerschner L, Li P, Brown AJ, Wenk MR, Parton RG, Yang H. Fld1p, a functional homologue of human seipin, regulates the size of lipid droplets in yeast. J Cell Biol 180: 473–482, 2008. 11. Fei W, Shui G, Zhang Y, Krahmer N, Ferguson C, Kapterian TS, Lin RC, Dawes IW, Brown AJ, Li P, Huang X, Parton RG, Wenk MR, Walther TC, Yang H. A role for phosphatidic acid in the formation of “supersized” lipid droplets. PLoS Genet 7: e1002201, 2011. 12. Fink L, Seeger W, Ermert L, Hanze J, Stahl U, Grimminger F, Kummer W, Bohle RM. Real-time quantitative RT-PCR after laserassisted cell picking. Nat Med 4: 1329 –1333, 1998. 13. Folch J, Lees M, Sloane Stanley GH. A simple method for the isolation and purification of total lipids from animal tissues. J Biol Chem 226: 497–509, 1957. 14. Hayashi YK, Matsuda C, Ogawa M, Goto K, Tominaga K, Mitsuhashi S, Park YE, Nonaka I, Hino-Fukuyo N, Haginoya K, Sugano H, Nishino I. Human PTRF mutations cause secondary deficiency of caveolins resulting in muscular dystrophy with generalized lipodystrophy. J Clin Invest 119: 2623–2633, 2009. 15. Kershaw EE, Flier JS. Adipose tissue as an endocrine organ. J Clin Endocrinol Metab 89: 2548 –2556, 2004.
E713
16. Kim CA, Delepine M, Boutet E, El Mourabit H, Le Lay S, Meier M, Nemani M, Bridel E, Leite CC, Bertola DR, Semple RK, O’Rahilly S, Dugail I, Capeau J, Lathrop M, Magre J. Association of a homozygous nonsense caveolin-1 mutation with Berardinelli-Seip congenital lipodystrophy. J Clin Endocrinol Metab 93: 1129 –1134, 2008. 17. Li J, Takaishi K, Cook W, McCorkle SK, Unger RH. Insig-1 “brakes” lipogenesis in adipocytes and inhibits differentiation of preadipocytes. Proc Natl Acad Sci USA 100: 9476 –9481, 2003. 18. Liu L, Brown D, McKee M, Lebrasseur NK, Yang D, Albrecht KH, Ravid K, Pilch PF. Deletion of Cavin/PTRF causes global loss of caveolae, dyslipidemia, and glucose intolerance. Cell Metab 8: 310 –317, 2008. 19. Lundin C, Nordstrom R, Wagner K, Windpassinger C, Andersson H, von Heijne G, Nilsson I. Membrane topology of the human seipin protein. FEBS Lett 580: 2281–2284, 2006. 20. Magre J, Delepine M, Khallouf E, Gedde-Dahl T Jr, Van Maldergem L, Sobel E, Papp J, Meier M, Megarbane A, Bachy A, Verloes A, d’Abronzo FH, Seemanova E, Assan R, Baudic N, Bourut C, Czernichow P, Huet F, Grigorescu F, de Kerdanet M, Lacombe D, Labrune P, Lanza M, Loret H, Matsuda F, Navarro J, NivelonChevalier A, Polak M, Robert JJ, Tric P, Tubiana-Rufi N, Vigouroux C, Weissenbach J, Savasta S, Maassen JA, Trygstad O, Bogalho P, Freitas P, Medina JL, Bonnicci F, Joffe BI, Loyson G, Panz VR, Raal FJ, O’Rahilly S, Stephenson T, Kahn CR, Lathrop M, Capeau J. Identification of the gene altered in Berardinelli-Seip congenital lipodystrophy on chromosome 11q13. Nat Genet 28: 365–370, 2001. 21. Payne VA, Grimsey N, Tuthill A, Virtue S, Gray SL, Dalla Nora E, Semple RK, O’Rahilly S, Rochford JJ. The human lipodystrophy gene BSCL2/seipin may be essential for normal adipocyte differentiation. Diabetes 57: 2055–2060, 2008. 22. Soloveva V, Graves RA, Rasenick MM, Spiegelman BM, Ross SR. Transgenic mice overexpressing the beta 1-adrenergic receptor in adipose tissue are resistant to obesity. Mol Endocrinol 11: 27–38, 1997. 23. Szymanski KM, Binns D, Bartz R, Grishin NV, Li WP, Agarwal AK, Garg A, Anderson RG, Goodman JM. The lipodystrophy protein seipin is found at endoplasmic reticulum lipid droplet junctions and is important for droplet morphology. Proc Natl Acad Sci USA 104: 20890 –20895, 2007. 24. Tian Y, Bi J, Shui G, Liu Z, Xiang Y, Liu Y, Wenk MR, Yang H, Huang X. Tissue-autonomous function of Drosophila seipin in preventing ectopic lipid droplet formation. PLoS Genet 7: e1001364, 2011. 25. Tontonoz P, Spiegelman BM. Fat and beyond: the diverse biology of PPARgamma. Annu Rev Biochem 77: 289 –312, 2008. 26. Windpassinger C, Auer-Grumbach M, Irobi J, Patel H, Petek E, Horl G, Malli R, Reed JA, Dierick I, Verpoorten N, Warner TT, Proukakis C, Van den Bergh P, Verellen C, Van Maldergem L, Merlini L, De Jonghe P, Timmerman V, Crosby AH, Wagner K. Heterozygous missense mutations in BSCL2 are associated with distal hereditary motor neuropathy and Silver syndrome. Nat Genet 36: 271–276, 2004. 27. Wolins NE, Brasaemle DL, Bickel PE. A proposed model of fat packaging by exchangeable lipid droplet proteins. FEBS Lett 580: 5484 –5491, 2006. 28. Wolinski H, Kolb D, Hermann S, Koning RI, Kohlwein SD. A role for seipin in lipid droplet dynamics and inheritance in yeast. J Cell Sci 124: 3894 –3904, 2011. 29. Zimmermann R, Strauss JG, Haemmerle G, Schoiswohl G, BirnerGruenberger R, Riederer M, Lass A, Neuberger G, Eisenhaber F, Hermetter A, Zechner R. Fat mobilization in adipose tissue is promoted by adipose triglyceride lipase. Science 306: 1383–1386, 2004. 30. Zu L, He J, Jiang H, Xu C, Pu S, Xu G. Bacterial endotoxin stimulates adipose lipolysis via toll-like receptor 4 and extracellular signal-regulated kinase pathway. J Biol Chem 284: 5915–5926, 2009.
AJP-Endocrinol Metab • doi:10.1152/ajpendo.00237.2011 • www.ajpendo.org
