Lipolytic actions of secretin in mouse adipocytes[S]

Secretin (Sct), a classical gut hormone, is now known to play pleiotropic functions in the body including osmoregulation, digestion, and feeding control. As Sct has long been implicated to regulate metabolism, in this report, we have investigated a potential lipolytic action of Sct. In our preliminary studies, both Sct levels in circulation and Sct receptor (SctR) transcripts in adipose tissue were upregulated during fasting, suggesting a potential physiological relevance of Sct in regulating lipolysis. Using SctR knockout and Sct knockout mice as controls, we show that Sct is able to stimulate lipolysis in vitro in isolated adipocytes dose- and time-dependently, as well as acute lipolysis in vivo. H-89, a protein kinase A (PKA) inhibitor, was found to attenuate lipolytic effects of 1 μM Sct in vitro, while a significant increase in PKA activity upon Sct injection was observed in the adipose tissue in vivo. Sct was also found to stimulate phosphorylation at 660ser of hormone sensitive lipase (HSL) and to bring about the translocation of HSL from cytosol to the lipid droplet. In summary, our data demonstrate for the first time the in vivo and in vitro lipolytic effects of Sct, and that this function is mediated by PKA and HSL.

fat pads from Wt, SctR Ϫ / Ϫ , and Sct Ϫ / Ϫ mice were placed in Krebs-Ringer bicarbonate buffer with 30 mM HEPES (KRBH buffer) supplemented with 3% FA-free BSA, 500 nmol/l adenosine, and 1 mg/ml collagenase type I (Worthington Biochemical, Lakewood, NJ) and agitated for 60 min at 37°C. The cells were washed three times and then resuspended in KRBH buffer with 3% FAfree BSA. Suspended adipocytes (10 5 cells/ml) were used for a lipolysis assay by incubating them with or without Sct (1 pM to 1 M) in the presence or absence of a nonselective ␤ -adrenergic agonist, isoproterenol (Iso) (1 M) or ␤ 3 -AR-specifi c agonist CL-316243 (5 M). After 60 min, glycerol release was measured with a kit from BioAssay Systems. For pathway analysis, cells were pretreated with 10 M H-89 (PKA inhibitor), 1 M R0-31-8220 (PKC inhibitor), 10 M SP 600125 (JNK inhibitor), or 10 M or 1 M CAY (HSL inhibitor) in the presence or absence of 1 M Sct. For time-dependent effects, cells were treated with or without 1 M Sct, and glycerol release was measured 0, 10,20,30,40,50,60, and 90 min after stimulation.

Serum analysis
For in vivo lipolysis assay, baseline serum samples were collected from 18 h-fasted Wt, SctR

PKA assay
A PepTag PKA activity assay kit (Promega) was used. Mice were fasted for 18 h and ip injected with either PBS or Sct (0.5 mg/ kg). Fifteen minutes after ip injection, epididymal fat pads were removed and processed as per the manufacturer's instructions.

Quantitative real-time PCR
Total RNA from epididymal fat tissue was extracted using Trizol reagent (Invitrogen). Total RNA (1 g) was used for the synthesis of fi rst strand cDNA. The cDNA was diluted 3-fold and real-time PCR was performed either with specifi c TaqMan probes (GAPDH, 4352339E; SctR, Mm01290790_m1) or with a SYBR PCR Master Mix kit (Applied Biosystems) and primers (sequences listed in Table 1 ). The specifi city of the SYBR Green PCR signal measured by the 7300 real-time PCR system (Applied Biosystems) was confi rmed by melting curve analysis and agarose gel electrophoresis. The threshold cycle ( C t ) value was used for calculating the ratio change in the target gene relative to the GAPDH control gene which was determined by the 2 Ϫ ⌬ ⌬ Ct method ( 31 ).

Western blot analysis
Mice were treated with an ip injection of either PBS or Sct (0.5 mg/kg) and 15 min later epididymal fat depots (100 mg) were as yet on the underlying cellular mechanism, secondary messenger pathway, and mode of action on the actions of Sct on adipocytes. In this report, using secretin receptor knockout (SctR Ϫ / Ϫ ) and secretin knockout (Sct Ϫ / Ϫ ) mice as controls in comparison with wild-type (Wt) animals, we sought to study the function of Sct in lipolysis and investigate the molecular mechanisms involved in this process in mouse adipocytes.

Animals
Procedures of animal care and handling were in accordance with the protocols approved by the Committee on the Use of Live Animals in Teaching and Research of the University of Hong Kong. All experiments were carried out using adult mice (23-26 g) of at least N5 generation, which were kept in a temperaturecontrolled room with a 12 h light-dark cycle. Mice were fed ad libitum with standard rodent chow (#5010; Test Diet, Richmond, IN) and water unless otherwise stated.

Fasting experiments
Age-and weight-matched mice (8-9 weeks old) were used for all the experiments. Food was removed from a cohort of mice at the beginning of the dark cycle and epididymal fat pads were collected from fasted and control mice at 0, 12, 18, and 24 h, and were immediately snap-frozen with liquid nitrogen for real-time quantitative measurements of secretin receptor (SctR) transcripts. Blood was collected from mice at the same time points and plasma Sct concentrations were measured by enzyme immunoassays (Phoenix Pharmaceuticals Inc.).

Adipocyte isolation and in vitro lipolysis assay
Adipocyte isolation and in vitro lipolysis assay were done as described ( 29,30 ), with minor modifi cations. Briefl y, epididymal

Immunofl uorescence imaging and immunohistochemical staining
Immunocytochemical staining was performed as described ( 34 ) with minor modifi cations. Isolated adipocytes were treated subsequently homogenized in a buffer composed of 25 mM Tris-HCl (pH 7.4), 25 mM NaCl, 1 mM MgCl 2 , 2.7 mM KCl, and protease and phosphatase inhibitors (0.5 mM Na 3 VO 4 , 1 mM NaF, 1 M leupeptin, 1 M pepstatin, 1 M okadaic acid, and 0.1 mM PMSF). For isolated adipocytes, cells were stimulated with or without Sct (1 M) or Iso (1 M) for 30 min before being lysed with buffer as described above. Western blot analysis of the lysates was performed as described ( 32 ) with dilutions of primary antibodies as suggested by their respective companies.

Continuous Sct infusion
Adult male mice (8-9 weeks old) were treated with PBS or Sct [2.5 nmol/kg/day, a dose similar to that published previously ( 33 )] by ip implantation of Alzet ® osmotic minipumps (Alzet, Cupertino, Mice were starved and blood samples or fat tissues were obtained for measurements of Sct peptide and SctR transcript levels, respectively, at specifi ed hours. A: Sct peptide concentrations in the circulation measured by an EIA kit were found to be signifi cantly increased after starvation . B: SctR transcript levels measured by real-time PCR were found to be signifi cantly increased after starvation. For (A) and (B), n = 9-11 (* P < 0.05, ** P < 0.0005 versus 0 h): C: SctR protein was localized on the adipose cell membrane of Wt mice by immunohistochemical staining using SctR Ϫ / Ϫ mice as negative controls.

, and SctR
Ϫ / Ϫ mice were found to exhibit similar basal lipolytic activities and responded similarly to 1 M Iso or 5 M CL-316243 treatment for 1 h. Data are means ± SEM of three separate experiments performed in triplicate. * P < 0.0001.
Bannockburn, IL) according to the recommended procedure using a rabbit anti-mouse SctR antibody [1:300; raised in our laboratory using a synthetic peptide (R-A-E-C-L-R-E-L-S-E-E-K-K) that is present in the mouse SctR] ( 17,35 ).

Statistical analysis
All data are shown as means ± SE. The deviations between groups were analyzed using Prism 3.0 software (GraphPad Software Inc., San Diego, CA). An unpaired t -test was performed when two groups were under consideration, whereas data from >2 groups were analyzed by one-way ANOVA, followed by Dunnett's test.
with or without 1 M Sct or 1 M Iso for 20 min, fi xed, and then blocked for 1 h in 5% goat serum at room temperature. Cells were then incubated overnight at 4°C with the primary antibody, anti-HSL (Cell Signaling Technology) (1:50 dilution). This was followed by a wash and 1 h incubation at room temperature with Alexa Fluor 488 goat anti-rabbit secondary antibody (1:300 dilution; Molecular Probes). After washing, the cells were placed in SlowFade Gold antifade reagent (Invitrogen) and confocal images were obtained using a Carl Zeiss LSM 710-NLO (Carl Zeiss, Oberkochen, Germany). For immunohistochemical staining, epididymal adipose tissue was fi xed in 3.7% formalin, embedded in paraffi n, and sectioned (7 m). Immunostaining was performed with a Leica Bond-Max automatic immunostainer (Leica,

Sct level in circulation and SctR expression in epididymal adipose tissue are increased during starvation
To determine whether Sct plays a physiological role in lipid mobilization, we fi rst studied the effects of fasting on the Sct level in the circulation and SctR expression in the epididymal adipose tissue of mice. Interestingly, we found that the Sct level in the circulation ( Fig. 1A ) and SctR ( Fig. 1B ) transcript in the adipose tissue increased in a time-dependent manner, and the most signifi cant increases were observed after 18 h fasting (3.2 ± 1.1-fold and 9.4 ± 1.6-fold increase in Sct peptide in blood and SctR transcript in adipose tissue, respectively). Expression of SctR protein in the adipocyte cell membrane was confi rmed by immunohistochemical staining ( Fig. 1C ) using the same tissue from SctR Ϫ / Ϫ mice as a negative control. In summary, the increase of both the ligand in the circulation and the receptor in the fat tissue suggests potential activity of Sct in fat metabolism in response to starvation.

Sct stimulates lipolysis in isolated adipocytes from mice via SctR and PKA
To determine whether Sct stimulates lipolysis, we investigated the in vitro rate of lipolysis in the presence of graded concentrations of Sct in isolated adipocytes in Wt mice, using SctR Ϫ / Ϫ and Sct Ϫ / Ϫ as negative and positive controls, respectively. In this study, the rate of glycerol release from isolated adipocytes into the culture medium was determined to refl ect its lipolytic activity. Sct was able to stimulate a dosedependent release of glycerol in Wt adipocytes ( Fig. 2A ) with an EC 50 of 37.5 nM. This lipolytic response of Sct was found to be specifi c to its receptor, as this effect was completely abolished in SctR Ϫ / Ϫ adipocytes ( Fig. 2A ), but could be reproduced in Sct Ϫ / Ϫ adipocytes ( Fig. 2A ). Sct also stimulated lipolysis in Wt adipocytes in a time-dependent manner ( Fig. 2B ). Isolated adipocytes from Wt, Sct Ϫ / Ϫ , and SctR Ϫ / Ϫ mice had similar basal lipolytic activity (glycerol release in mol/10 5 cells/h), as well as no signifi cant differences in their lipolytic responses to 1 M Iso (a nonselective ␤ -adrenergic agonist) or 5 M CL-316243 ( ␤ 3 -AR-specifi c agonist) ( Fig. 2C ), indicating that SctR Ϫ / Ϫ adipocytes are able to respond normally to other lipolytic agents.
After showing the lipolytic actions of Sct, we next investigated its secondary messenger pathway. In this study, the specifi c inhibitor of PKA (10 M H-89), PKC (1 M Ro-31-8220), or JNK (10 M SP 600125) was preincubated with adipocytes before stimulation by 1 M Sct; it was found that only H-89, but not R0-31-8220 or SP 600125, could completely abolish lipolytic actions of Sct in Wt and Sct Ϫ / Ϫ adipocytes ( Fig. 3A ). The rates of glycerol release in SctR Ϫ / Ϫ adipocytes were similar to basal levels in all treatment groups due to the lack of SctR in these cells. Taken together, our data show that the lipolytic effect of Sct is mediated by the PKA pathway. Additionally, incubation with the HSL specifi c inhibitor, 10 M or 1 M CAY, completely abolished the glycerol release stimulated by 1 M Sct ( Fig. 3A ). Furthermore, FFA release by 1 M Sct was attenuated to around 33-35% on coincubation with 10 M or 1 M CAY ( Fig. 3B ). Such a response is similar to that of catecholamine in HSLnull mouse adipocytes, wherein catecholamine stimulation completely abolished glycerol release and severely reduced FFA release ( 36,37 ). This study suggests that HSL is the primary downstream enzyme involved, while delineating the role for other enzymes to be minimal or none.

Sct activates phosphorylation of 660 ser and 563 ser in HSL and translocation of HSL from cytosol to lipid droplets in adipocytes
To confi rm the involvement of HSL as a downstream regulator, we measured changes in the phosphorylation of the serine residues at positions 660, 563, and 565 of HSL after Sct

Sct induces an acute lipolytic response through SctR and PKA in vivo
After showing the in vitro lipolytic actions of Sct through PKA and HSL in isolated adipocytes, our next objective was to investigate whether Sct could induce lipolysis in vivo. After 5, 10, and 15 min ip injections of Sct (0.5 mg/kg), we found signifi cant elevation of circulating FFAs in the blood ( Fig. 5A ). This lipolytic effect of Sct is specifi c to SctR because the rise in the FFA level was not observed in SctR Ϫ / Ϫ , while it could be reproduced in Sct Ϫ / Ϫ mice ( Fig. 5B ). In addition, both the Wt and SctR Ϫ / Ϫ mice showed a normal lipolytic response to ip CL-316243 (0.1 mg/kg) as controls ( Fig. 5C ), indicating the normal functioning of these animals in response to the ␤ 3 -AR-specifi c agonist.
To test the in vivo involvement of PKA and HSL in carrying out the lipolytic effects of Sct, mice were ip injected with Sct (0.5 mg/kg) or control PBS, and epididymal adipose tissue was removed 15 min after for PKA measurements. It was found that PKA activity in epididymal adipose tissue was signifi cantly upregulated in Sct-injected mice when compared with control mice ( Fig. 6A ). Similar to the in vitro studies, Sct injection resulted in increased phosphorylation of 660 ser in HSL ( Fig. 6B ), while there were no signifi cant changes in the phosphorylation status of 563 ser and 565 ser , as well as protein levels of HSL. In summary, our in vivo data confi rm that Sct activates lipolysis through PKA to stimulate phosphorylation of HSL-660 ser . stimulation. While the protein levels of GAPDH and total HSL of the Wt adipocytes were similar with or without Sct stimulation ( Fig. 3C ), Sct incubation led to a 4.26 ± 0.75-fold ( P < 0.05) and a 1.99 ± 0.32-fold ( P < 0.05) increase in phosphorylation at positions 660 and 563, respectively, but not Ser 565 ( Fig. 3C ). Perilipin phosphorylation is essential for HSL-stimulated lipolysis, and Sct consistently stimulated phosphorylation of perilipin ( Fig. 3C ). Additionally, stimulation with Sct did not have any effect on the triglyceride lipase, ATGL, and its positive and negative regulators ABDH5 and G0S2, respectively ( Fig. 3C ), thus confi rming that HSL is the primary enzyme involved in the lipolytic effect of Sct. In summary, our fi ndings suggest that Sct-activated lipolysis in adipocytes could be mediated by the phosphorylation of HSL at 660 ser and 563 ser .
To understand further the cellular mechanism of Sct on lipolysis, translocation of HSL from the cytosol to the lipid droplet in isolated adipocytes was studied. Through immunofl uorescent imaging, in the positive control, treatment of isolated adipocytes with Iso clearly resulted in the translocation of HSL from the cytosol to the phospholipid monolayer surface of the lipid droplet in Wt and SctR adipocytes ( Fig. 4 ). This translocation of HSL was also observed in Sct-treated Wt cells, but not in Sct-treated SctR Ϫ / Ϫ adipocytes or in unstimulated Wt and SctR Ϫ / Ϫ cells ( Fig. 4 ). Our fi ndings suggest that the lipolytic action of Sct is mediated via the translocation of HSL from the cytosol to the lipid droplet. involved in fasting-induced lipolysis, while glucocorticoids have also been shown to play a role ( 6 ). Among hormones that stimulate lipolysis, such as catecholamines, insulin, glucagon, growth hormone, and thyroid hormone ( 5, 6 ), the lipolytic effect of Sct remains elusive (23)(24)(25)(26)(27)(28). Being a recently recognized anorectic hormone ( 20 ) with elevated plasma concentrations during fasting, as shown in our study and other studies (38)(39)(40), along with increased SctR expression in the epididymal adipose tissue during fasting ( Fig. 1B ), we therefore hypothesized the potential function of Sct in lipid metabolism. Consistent with this hypothesis, using Sct Ϫ / Ϫ and SctR Ϫ / Ϫ mouse models, we show here that Sct is able to stimulate dose-and timedependent lipolysis in both in vitro adipocytes and in in vivo animals. This lipolytic effect is mediated by SctR expressed on the fat cells, which stimulates the cAMPdependent protein kinase, PKA, which leads to phosphorylation of HSL, the rate-limiting enzyme for lipolysis ( 41 ), to bring about translocation of HSL to fat droplets for the activation lipolysis ( Fig. 8 ). PKA is known to phosphorylate HSL at residues 565 ser , 563 ser , 660 ser , and 659 ser ; where 565 ser is considered the basal phosphorylation site, while the other three serines are regulatory sites. Among them,

Continuous Sct infusion does not increase circulating FFA levels
Our in vivo and in vitro data consistently showed an acute lipolytic effect of Sct, therefore we sought to investigate its long-term effect on lipid metabolism by continuous infusion of Sct (2.5 nmol/kg/day) with a mini-osmotic pump. Sct infusion for 18 to 24 h was found to be unable to signifi cantly change levels of circulating FFAs ( Fig. 7A ). When we measured expression changes of lipogenic and lipolytic genes, interestingly, a 24 h Sct infusion was found to elevate both lipolytic HSL as well as lipogenic gene cluster of differentiation 36 (CD 36) transcript levels ( Fig. 7B ). These data suggest a dual lipolytic and lipogenic role of long-term infusion of Sct that needs to be explored and clarifi ed in future studies.

DISCUSSION
Catabolizing triacylglycerols stored in lipid droplets is a simple biochemical reaction; the process, however, is highly regulated and required for energy homeostasis (4)(5)(6). Catecholamine is known, so far, to be the key hormone ser was shown to be important for lipolysis because mutation of this residue resulted in a loss of enzymatic activity ( 42,43 ). Consistent with this, we show that Sct is able to stimulate phosphorylation of HSL-660 ser in both in vitro and in vivo studies. Similar to catecholamines ( 42 ) and glucagon ( 44 ) that act via the PKA/HSL pathway to promote translocation of HSL from the cytosol to the phospholipid monolayer of the lipid droplet, we were also able to observe an accumulation of HSL surrounding the lipid droplet after Sct stimulation in Wt, but not in SctR Ϫ / Ϫ , adipocytes . It seems that catecholamines, glucagon, and Sct use the same intracellular pathway to stimulate lipolysis in the fat cells. A working model summarizing the molecular mechanism in the lipolytic effect of Sct is shown in Fig. 8 .
While catecholamines are mainly responsible for acute lipolysis, other stimulants such as TNF-␣ and glucocorticoids have more chronic lipolytic effects ( 45 ). As shown in this study, Sct is able to stimulate an acute lipolytic response in vivo, evidenced by an increase in circulating FFA levels upon Sct injection ( Fig. 5A ). A 24 h continuous infusion of Sct through a mini-osmotic pump had no signifi cant effect on levels of circulating FFAs, but could stimulate expressions of lipolytic HSL and lipogenic CD 36 genes, suggesting that effects of long-term Sct on lipid metabolism is different from its acute response. In fact, it has been reported that chronic elevated levels of Sct can increase FA uptake and triglyceride storage in fat cells, as well as upregulation of FA uptake genes including CD 36 in adipose cells in vitro ( 25 ). In addition, the expression of SctR has been found to be increased in the omental adipose tissue of obese individuals ( 46 ) and has been positively correlated with body mass index ( 25 ). This evidence, along with our fi ndings, suggests a potential lipogenic role for longterm Sct, although this remains to be confi rmed and the molecular mechanisms elucidated in future.
The balance between triacylglycerol hydrolysis and FFA esterifi cation controls lipid homeostasis in our body. A slight alteration in these processes could therefore lead to metabolic disorders such as type 2 diabetes ( 47 ), hepatic steatosis ( 48 ), cardiovascular diseases ( 49 ), lipotoxicity ( 50 ), dyslipidemia ( 51 ), and certain types of cancer ( 52 ). In the case of obesity, higher levels of FFAs in the circulation due to changes in lipolytic rates result in the development Fig. 8. Working model summarizing the lipolytic effect of Sct. Sct binds SctR on the plasma membrane of adipocytes to activate the cAMP-dependent PKA. PKA then phosphorylates HSL at 660 ser leading to translocation of HSL from the cytosol to the phospholipid monolayer of the lipid droplet. The translocated HSL is responsible for the hydrolysis of stored triacylglycerol to release glycerol and FFAs. Norepinephrine (a catecholamine) and glucagon both activate the PKA pathway similar to Sct. TNF-␣ and glucocorticoids stimulate lipolysis by downregulating perilipin and enhancing the expression of ATGL, respectively. Insulin inhibits lipolysis by reducing the cAMP level through activation of PDE via the PI3K/Akt pathway. AMPK, AMP-activated protein kinase. of insulin resistance ( 53 ). In familial hyperlipidemia, decreased expression or function of HSL causes an impaired lipolytic function of catecholamines ( 54 ). It is therefore not surprising that proteins involved in lipid metabolism are targeted as pharmacological strategies against disorders such as obesity and metabolic syndrome ( 10,55 ). This reinforces the relevance and importance of the need for more research on Sct and lipid metabolism. Our group and others have recently highlighted the importance of reevaluating the metabolic effects of Sct ( 20,56 ). In this report, we have provided evidence to conclude that Sct has an acute lipolytic effect in vitro and in vivo . The relationship between lipid metabolism and Sct should warrant further studies in the future to provide an alternative therapeutic means to tackle various metabolic diseases.