BBRC

Biochemical and Biophysical Research Communications 326 (2005) 744751 www.elsevier.com/locate/ybbrc

Lysophosphatidylcholine enhances glucose-dependent insulin secretion via an orphan G-protein-coupled receptorq

Takatoshi Soga*, Takahide Ohishi, Tetsuo Matsui, Tetsu Saito, Mitsuyuki Matsumoto, Jun Takasaki, Shun-ichiro Matsumoto, Masazumi Kamohara, Hideki Hiyama, Shigeru Yoshida, Kazuhiro Momose, Yoshitaka Ueda, Hitoshi Matsushime, Masato Kobori, Kiyoshi Furuichi
Institute for Drug Discovery Research, Yamanouchi Pharmaceutical Co., Ltd, 21 Miyukigaoka, Tsukuba, Ibaraki 305-8585, Japan Received 8 November 2004 Available online 8 December 2004

Abstract A lysophospholipid series, such as lysophosphatidic acid, lysophosphatidylserine, and lysophosphatidylcholine (LPC), is a bioactive lipid mediator with diverse physiological and pathological functions. LPC has been reported to induce insulin secretion from pancreatic b-cells, however, the precise mechanism has remained elusive to date. Here we show that an orphan G-protein-coupled receptor GPR119 plays a pivotal role in this event. LPC potently enhances insulin secretion in response to high concentrations of glucose in the perfused rat pancreas via stimulation of adenylate cyclase, and dose-dependently induces intracellular cAMP accumulation and insulin secretion in a mouse pancreatic b-cell line, NIT-1 cells. The Gs-protein-coupled receptor for LPC was identied as GPR119, which is predominantly expressed in the pancreas. GPR119-specic siRNA signicantly blocked LPC-induced insulin secretion from NIT-1 cells. Our ndings suggest that GPR119, which is a novel endogenous receptor for LPC, is involved in insulin secretion from b-cells, and is a potential target for anti-diabetic drug development. 2004 Elsevier Inc. All rights reserved.
Keywords: G-protein-coupled receptor; Lysophosphatidylcholine; Insulin secretion; Adenylate cyclase; cAMP; Pancreatic b-cell

A lysophospholipid (LPL) series, such as lysophosphatidic acid (LPA), lysophosphatidylcholine (LPC), and lysophosphatidylserine (LPS), is a bioactive lipid with diverse physiological and pathological functions. LPA evokes various biological eects, including neuroq Abbreviations: GPCR, G-protein-coupled receptor; LPL, lysophospholipid; LPC, lysophosphatidylcholine; LPI, lysophosphatidylinositol; LPE, lysophosphatidylethanolamine; RT-PCR, reverse transcription-polymerase chain reaction; HPRT, hypoxanthine-guanine phosphoribosyl transferase; IBMX, 3-isobutyl-1-methylxanthine; cAMP, cyclic adenosine monophosphate; BSA, bovine serum albumin; siRNA, short interference RNA; EC50, median eective concentration; SE, standard error; AUC, area under curve. * Corresponding author. Fax: +81 29 852 5444. E-mail address: soga.takatoshi@yamanouchi.co.jp (T. Soga).

genesis, angiogenesis, and carcinogenesis [1]. LPC plays an etiological role in atherosclerosis [2], and promotes inammatory eects, including the induction of the expression of chemotactic factors in endothelial cells [3] and macrophage activation [4]. LPS induces mast cell activation [5]. The recent discovery of G-protein-coupled receptors (GPCRs) for LPA and LPC has established the status of LPLs as endogenous mediator. LPA1/Edg2, LPA2/Edg4, LPA3/Edg7, and LPA4/ GPR23 have been identied as LPA receptors [6]. G2A [7] and GPR4 [8] have been identied as LPC receptors (recent studies [9,10] incorrectly reported GPR4 and G2A to be proton-sensing GPCRs). One of LPCs reported physiological actions is the induction of insulin secretion from pancreatic b-cells

0006-291X/$ - see front matter 2004 Elsevier Inc. All rights reserved. doi:10.1016/j.bbrc.2004.11.120

T. Soga et al. / Biochemical and Biophysical Research Communications 326 (2005) 744751

745

[11,12], however, to date, the precise mechanism has remained elusive. Here we show that LPC induces insulin secretion through cAMP production not only in the pancreatic b-cell lines but also in the intact pancreas. Incretin hormones, such as glucagon-like peptide-1 (GLP-1) and glucose-dependent insulinotropic polypeptide (GIP), are well known to bind to their endogenous Gs-protein-coupled receptors expressed in pancreatic b-cells, inducing glucose-dependent insulin secretion by accumulating intracellular cAMP via adenylate cyclase activation [13,14]. Assuming that a Gs-protein-coupled receptor for LPC is expressed in pancreatic b-cells, the physiological phenomenon described above could be explained. Although the presence of a Gs-protein-coupled receptor for LPC has been suggested [15], previously identied LPC receptors, G2A and GPR4, are Gi-protein-coupled receptors and are not expressed in the pancreas [16]. The recent progress in genetic analysis technologies enables us to identify and characterize numbers of novel orphan GPCRs, and we have succeeded in clarifying novel ligandreceptor systems and their function on several orphan GPCRs by demonstrating their endogenous ligands, such as cysteinyl leukotriene receptor CysLT2 [17], ADP receptor P2TAC [18], prokineticin receptors PK-Rs [19], and nicotinic acid receptor HM74b [20]. In line with this course of characterizing and de-orphaning research, we could identify an orphan GPCR, GPR119 [21], as a novel Gs-protein-coupled receptor for LPC. We demonstrated here that GPR119 is predominantly expressed in pancreatic b-cells and that activation of GPR119 by LPC leads to glucose-dependent insulin secretion.

Materials and methods

Materials and cells. 1-Palmitoyl (16:0), -stearoyl (18:0), -oleoyl (18:1)-LPCs, LPS, lysophosphatidylinositol (LPI), lysophosphatidylethanolamine (LPE), lyso-platelet-activating factor (lyso-PAF), stearic acid, oleic acid, arachidonic acid, sphingosylphosphorylcholine (SPC), psychosine, fatty acid-free bovine serum albumin (BSA), MDL12330A, 3-isobutyl-1-methylxanthine (IBMX), and anti-FLAG M2 antibody were purchased from Sigma. 1-Oleoyl-LPA and sphingosine-1-phosphate (S-1-P) were purchased from Cayman Chemical. Phosphatidic acid, phosphatidylcholine, phosphatidylserine, phosphatidylinositol, phosphatidylethanolamine, cardiolipin, and sphingomyelin were obtained from Doosan Serdary Research Laboratories. Other chemical reagents were of analytical grade. A rat hepatoma cell line, RH7777 cells were kindly provided by Dr. M. Finnen (Yamanouchi Research Institute, Littlemore, Oxford, UK). A mouse pancreatic b-cell line, NIT-1 cells were purchased from ATCC. Perfusion of isolated rat pancreas. Pancreas was isolated from male Wistar rats (CLEA) weighing 300400 g and perfused according to the method described previously [22], with slight modications. In brief, the pancreas and associated spleen, stomach, and duodenum were isolated under sodium pentobarbital anesthesia (50 mg/kg i.p.). The isolated pancreas was perfused through the coeliac artery at a ow rate

of 1 ml/min with Hepes-balanced KrebsRinger bicarbonate buer (pH 7.4) (KRBB) containing 2.8 or 16.8 mM glucose and 0.1% fatty acid-free BSA. The preparation was placed in an acrylic chamber lled with KRBB kept at 37 C. The euent perfusate from a portal vein cannula was collected into fraction tubes containing aprotinin (500 U/ tube) at 1- or 2-min intervals. Collected samples were stored at 20 C until the insulin concentration was measured by radioimmunoassay (Amersham Biosciences), using rat insulin as standard. Insulin secretion assay. NIT-1 cells were seeded in 1 ml of Dulbeccos modied Eagles medium (DMEM) containing 25 mM glucose and 10% fetal bovine serum in a 24-well plate at 2 105 cells/well for 48 h. The cells were washed with Hepes-balanced KRBB containing 2 mM glucose and 0.1% fatty acid-free BSA, and preincubated for 60 min at 37 C in the same medium. After preincubation, the cells were stimulated with 18:1-LPC in Hepes-balanced KRBB containing 12 mM glucose at 37 C for 25 min. MDL12330A was added 10 min before the incubation period. Insulin secreted into the supernatant was measured by radioimmunoassay (Amersham Biosciences) using rat insulin as standard. Molecular cloning. The ORF of human GPR119 (GenBank Accession No. BD169091) was amplied by PCR from human pancreas cDNA (Clontech) with the following XbaI restriction-site-containing primers: sense primer, 5 0 -AAAATCTAGAATGGAATCA TCTTTCTCATTTG-3 0 ; antisense primer, 5 0 -CGGCTCTAGATT AGCCATCAAACTCTGAGCTGG-3 0 . This amplied product was subcloned into the mammalian expression vector plasmid, pEF-BOS [23], with slight modications (the cloned cDNA is a tagged FLAGepitope DYKDDDDK with a N-terminus), using XbaI sites. This plasmid was named pEF-hGPR119. Mouse and rat counterparts (mGPR119, rGPR119) against human GPR119 were cloned using the homology-based PCR method (GenBank Accession Nos. mGPR119: AY288423, rGPR119: BD169104). pEF-hGPR119 and pcDNA3.1/Zeo (Invitrogen) were cotransfected into RH7777 cells. The candidate clones stably expressing hGPR119 were selected by resistance to 80 lg/ml zeocin. The RH7777 cells stably expressing hGPR119 were conrmed by FACS scan analysis using anti-FLAG M2 antibody. cAMP assay. NIT-1 cells were seeded in 96-well plates at 2 104 cells/well, and hGPR119-RH7777 cells and mock transfected RH7777 cells (vector-RH7777) were seeded in collagen type I-coated 96-well plates at 3 103 cells/well, for 48 h. The cells were exposed to stimulants dissolved in DMEM containing 1 mM IBMX. After incubation for 15 min, the cells were harvested with 0.2% Triton X-100 in phosphate-buered saline. MDL12330A (Sigma) was added 10 min before the incubation period. Intracellular cAMP was measured using a cAMP homogeneous time-resolved uorescence kit (CIS Bio International). Quantitative analysis of GPR119 transcripts by reverse transcriptionpolymerase chain reaction. Poly(A)+ RNAs from various human and rat tissues were purchased (Clontech and Biochain), and total RNAs from rat islets and cultured cell lines were isolated using RNeasy kits (Qiagen). cDNAs were synthesized as described previously [24]. We quantied GPR119 transcripts by means of a Prism 7700 Sequence Detector (Applied Biosystems) with the specic primer sets (human sense primer; 5 0 -TCTCGGCCCACACAGAAGA-3 0 , human antisense primer; 5 0 -GCTGCGGAGGAAGTGACAAA-3 0 , rat and mouse sense primer; 5 0 -AGCTCTGCTCAGCACACACAG-3 0 , rat and mouse antisense primer; 5 0 -GAATGCCATCCGAAGGCTAC-3 0 ). Reverse transcription-polymerase chain reaction (RT-PCR) was carried out in a 25 ll reaction mixture prepared with a 2 SYBR Green Master Mix (Applied Biosystems) containing an appropriately diluted cDNA solution and 0.2 lM of each primer. To obtain a calibration curve, we amplied a known amount of human, rat, and mouse genomic DNA in the same manner as the samples. The number of copies of genomic DNA per microliter was calculated with the equation described previously [25]. Human, rat, and mouse b-actin mRNAs were also measured as an internal control.

746

T. Soga et al. / Biochemical and Biophysical Research Communications 326 (2005) 744751

Short interfering RNA. Five short interfering RNAs (siRNAs) (21 nucleotides each) were designed from the mGPR119 sequence and they were purchased (Dharmacon). In our experiments to suppress mGPR119 expression in NIT-1 cells, we used the most eective siRNA from among them (target sequence; 5 0 -GAUGGCAUUUGUCACU UCUdTdT-3 0 ). We used scramble I duplex siRNA (Dharmacon) as a control. NIT-1 cells were seeded in 500 ll DMEM containing 25 mM glucose and 10% fetal bovine serum in a 24-well plate at 1.5 105 cells/ well. After overnight incubation (20 h), medium was exchanged for 500 ll/well DMEM containing 25 mM glucose and pre-mixed transfection reagent (5 ll/well of 20 lM siRNA, 1.5 ll/well Lipofectamine2000 (Invitrogen), and 45 ll/well Opti-MEM). After 4 h incubation, supernatant was exchanged for 1 ml/well DMEM containing 25 mM glucose and 10% fetal bovine serum. Two days after transfection, we treated with LPC, and mGPR119 mRNA expression was determined as described above.

Results and discussion Previous studies have reported that LPC is involved in insulin secretion in primary cultured pancreatic islets and pancreatic b-cell lines [11,12], but glucose dependency has not been demonstrated. In an attempt to address this issue directly, isolated rat pancreas was perfused and stimulated with LPC (18:1-LPC) in the presence of 2.8 or 16.8 mM glucose. At basal (2.8 mM) glucose concentration, LPC (10 or 30 lM) had no eect, but it potently enhanced insulin secretion in a dose-dependent manner in the presence of 16.8 mM glucose [areas under curve (AUC) for 10 min of 16.8 mM glucose only, 16.8 mM glucose + 10 lM LPC, and 16.8 mM glucose + 30 lM LPC were 43.8, 76.7, and 91.0 ng/10 ml, respectively] (Fig. 1A). The observation that LPC potentiates insulin secretion in response to a high concentration of glucose suggests that LPC functions in a similar way to incretin hormones in vivo [13,14]. This result is not consistent with previous studies [11,12] reporting that LPC induces insulin secretion at basal levels of glucose (0 or 1.7 mM) in perfused islets and a pancreatic b-cell line. We suggest two reasons for this discrepancy: rst, the extremely high concentration (about 100200 lM) of LPC used in previous studies, and second the dierence between whole tissue and cultured cells. Incretin hormones, such as GLP-1 and GIP, are well known as glucose-dependent insulinotropic peptides that fulll their function by causing accumulation of intracellular cAMP via their Gs-protein-coupled receptors expressed on pancreatic b-cells [13,14]. To investigate the hypothesis that LPC, like the incretin hormones, potentiates glucose-dependent insulin secretion by causing accumulation of intracellular cAMP, we examined the eect of an adenylate cyclase inhibitor, MDL12330A [26]. MDL12330A (10 lM) markedly suppressed the potentiation of insulin secretion by LPC (30 lM) (AUC for 10 min of 16.8 mM glucose only, 16.8 mM glucose + 30 lM LPC, and 16.8 mM glucose + 30 lM LPC + 10 lM MDL12330A were 57.2,

Fig. 1. Eect of LPC (18:1-LPC) on insulin secretion in perfused rat pancreas. (A) Eect of LPC (10 and 30 lM) on insulin secretion at 2.8 mM glucose and 16.8 mM glucose (n = 3). (B) Eect of MDL12330A (10 lM) on potentiation of glucose-dependent insulin secretion by LPC (30 lM) (n = 3). These assays were performed using assay buer containing 0.1% BSA to retain the function of the isolated pancreas. The data are means SE.

101.8, and 64.3 ng/10 ml, respectively) (Fig. 1B), indicating that LPC potentiates insulin secretion via intracellular cAMP production in the pancreas. Next, to conrm that LPC acts directly on pancreatic b-cells, we examined the eect of LPC (18:1-LPC) on insulin secretion and intracellular cAMP concentration in the mouse pancreatic b-cell line NIT-1. LPC dose-dependently stimulated intracellular cAMP accumulation (Fig. 2A) and induced insulin secretion (Fig. 2B). To determine whether cAMP is an essential signal for LPC-potentiated insulin secretion, we examined the effect of MDL12330A on LPC-induced insulin secretion and cAMP accumulation. MDL12330A (10 lM) significantly blocked cAMP accumulation evoked by LPC (10 lM), but did not aect the cAMP level in the absence of LPC (Fig. 2C). Similarly, MDL12330A (10 lM) signicantly inhibited insulin secretion by

T. Soga et al. / Biochemical and Biophysical Research Communications 326 (2005) 744751

747

Fig. 2. Analysis of LPC (18:1-LPC)-induced insulin secretion mechanism in the mouse pancreatic b-cell line, NIT-1. (A) Doseresponse analysis of the eect of LPC on intracellular cAMP accumulation (n = 4). (B) Doseresponse analysis of the eect of LPC on insulin secretion (n = 4). (C) Eect of MDL12330A (10 lM) on stimulation of cAMP accumulation by LPC (10 lM) (n = 4). (D) Eect of MDL12330A (10 lM) on induction of insulin secretion by LPC (10 lM) (n = 4). (E) Eect of siRNA on expression of GPR119 mRNA in NIT-1 cells. Data represent the copy ratio of GPR119 and HPRT to control siRNA (n = 3). (F) Eect of siRNA on LPC-induced insulin secretion (n = 4). The data are means SE. *p < 0.05, **p < 0.01; Students t test. The graphs shown are representative of at least three experiments.

LPC (10 lM), but did not aect insulin secretion in the absence of LPC (Fig. 2D). These data show that LPC induces insulin secretion through cAMP production, not only in the intact pancreas but also in the pancreatic b-cell line. The results shown in Figs. 1B and 2D indicate that LPC-potentiated insulin secretion is closely associated with activation of adenylate cyclase. Adenylate cyclase can be activated directly, by agents such as forskolin, or via the receptor-linked stimulatory Gs-protein [27]. A previous study [28] has demonstrated that LPC has

a minimal eect on adenylate cyclase at concentrations less than 100 lg/ml (about 200 lM), suggesting that the concentrations of LPC (less than 30 lM) used in the current study cannot directly activate adenylate cyclase. These observations suggest that LPC might bind to and activate an unknown membrane-surface Gs-protein-coupled receptor to induce insulin secretion. To clarify the identity of the molecule involved in LPC-potentiated insulin secretion, we searched among orphan GPCRs for a Gs-protein-coupled receptor and identied that GPR119 induces intracellular cAMP

748

T. Soga et al. / Biochemical and Biophysical Research Communications 326 (2005) 744751 Table 1 Lipid-derivative potency to induce intracellular cAMP accumulation in RH7777 cells stably expressing human GPR119 EC50 (lM) Phosphatidic acid Phosphatidylcholine Phosphatidylethanolamine Phosphatidylserine Phosphatidylinositol 1-Oleoyl-lysophosphatidic acid (18:1-LPA) 1-Palmitoyl-lysophosphatidylcholine (16:0-LPC) 1-Stearoyl-lysophosphatidylcholine (18:0-LPC) 1-Oleoyl-lysophosphatidylcholine (18:1-LPC) Lysophosphatidylethanolamine Lysophosphatidylserine Lysophosphatidylinositol Lyso-platelet-activating factor (lyso-PAF) Cardiolipin Sphingomyelin Sphingosine-1-phosphate Sphingosylphosphorylcholine Sphingosine Psychosine Arachidonic acid Stearic acid Oleic acid >30 >30 >30 >30 >30 >30 1.6 0.52 3.3 0.80 1.5 0.17 5.7 0.50 >30 5.7 0.56 >30 >30 >30 >30 >30 >30 >30 >30 >30 >30

accumulation by LPC. LPC (18:1-LPC, 18:0-LPC, and 16:0-LPC) evoked intracellular cAMP accumulation in RH7777 rat hepatoma cells stably expressing human GPR119 (hGPR119-RH7777) in a dose-dependent manner, on the one hand, it did not aect mock transfected RH7777 cells (vector-RH7777) (Figs. 3A and B). As well as LPC, some endogenous lysophospholipids (LPLs), including lysophosphatidylinositol (LPI) and lysophosphatidylethanolamine (LPE), signicantly induced intracellular cAMP accumulation in a dose-dependent manner (Fig. 3A). The rank order of potency of the LPLs was 18:1-LPC, 16:0-LPC > 18:0-LPC > LPE, and LPI, with EC50 values of 1.5 0.17, 1.6 0.52, 3.3 0.80, 5.7 0.50, and 5.7 0.56 lM, respectively. Other endogenous candidate bioactive molecules including lipid-derivatives, amines, nucleotides, peptides, and hormones did not activate GPR119 (Table 1 and data not shown). These results demonstrate that GPR119 is a Gs-protein-coupled receptor for LPC as well as for LPI and LPE. To detect the direct interaction between these LPLs and GPR119, a radioligand binding study was performed using cell membranes from hGPR119RH7777 cells. Unlabelled LPC, LPE, and LPI were used in attempts to displace [3H]16:0-LPC. However, these binding experiments failed, most likely due to the low anity of the receptor for 16:0-LPC, as indicated by its high EC50 value (1.6 lM, Table 1). Therefore, proper synthetic compounds with high anities are required to analyze the direct interaction between them. Tissue distribution of GPR119 was quantitatively studied using RT-PCR-based analysis. Human and rat GPR119 are predominantly expressed in the pancreas (Figs. 4A and B), and the expression of rat GPR119 is about 20 times higher in the islets of pancreas than in the whole pancreas (Figs. 4B and C). In addition, the analysis of GPR119 mRNA in various cell lines revealed that GPR119 is expressed in the only pancreatic b-cell

Values were expressed as means SE of at least three separate experiments performed in n = 4.

lines (NIT-1, MIN6, and RIN5) (Fig. 4C). Taken together, GPR119 is predominantly localized in the pancreas, especially in the islets. These results suggest that GPR119 is expressed in pancreatic b-cells in vivo, strongly supporting our contention that LPC potentiates insulin secretion via GPR119. Finally, we performed experiments to inhibit the expression of mouse GPR119 (mGPR119) in NIT-1 cells by small interfering RNA (siRNA). The treatment with mGPR119-specic siRNA reduced mGPR119 mRNA to about half the level of that in NIT-1 cells, while it

Fig. 3. Activation of GPR119 by several lysophospholipids. Doseresponse analysis of eects of 16:0-LPC (d), 18:0-LPC (m), 18:1-LPC (j), LPI (n), and LPE (s) on cAMP accumulation (A) in RH7777 cells stably expressing human GPR119 and (B) in mock transfected RH7777 cells. (C) Eect of BSA on LPC-evoked intracellular cAMP accumulation in RH7777 cells stably expressing human GPR119 (n = 4). The data are means SE. The graphs shown are representative of at least three experiments.

T. Soga et al. / Biochemical and Biophysical Research Communications 326 (2005) 744751

749

Fig. 4. Expression analysis of GPR119 mRNA. (A) Quantitative RT-PCR analysis of GPR119 transcripts in normal human tissues. (B) Quantitative RT-PCR analysis of GPR119 transcripts in normal rat tissues. (C) Quantitative RT-PCR analysis of GPR119 transcripts in pancreatic b-cell lines and rat islets. NIT-1, mouse pancreatic b-cell line; MIN6, mouse pancreatic b-cell line; RIN5, rat pancreatic b-cell line; Neuro2a, mouse neuroblastoma cell line; RBL-2H3, rat basophilic/mast cell line; RAW264, mouse macrophage cell line; C6, rat glioma cell line; 3T3-L1, mouse embryonic broblasts; RH7777, rat hepatoma cell line; and HEK293, human embryonic kidney cell line. The data represent the ratio of GPR119 to b-actin mRNA.

had no eect on the expression level of the control housekeeping gene, hypoxanthine-guanine phosphoribosyl transferase (HPRT) and b-actin (Fig. 2E and data not shown). mGPR119-specic siRNA signicantly blocked 18:1-LPC (5 lM, 10 lM)-potentiated insulin secretion from NIT-1 cells (Fig. 2F). These data indicate that the LPC, at least partly, promotes insulin secretion via GPR119, resulting from cAMP production in NIT-1 cells. Physiological concentration of LPC in serum or plasma has been reported to be very high (120180 lM) [29,30]. It should be noted, however, that this concentration does not mean free LPC concentration. LPC is mainly present in the albumin- and lipoprotein-bound form [31]. Certainly, bovine serum albumin (BSA) greatly diminished the LPC-evoked intracellular cAMP accumulation in hGPR119-RH7777 (Fig. 3C). Intracellular cAMP accumulation with LPC (10 lM) almost disappeared by 0.5% BSA (LPC:BSA molar ratio of about 1:7.5). In vivo, the molar ratio of albumin (5% in plasma) to LPC (120 lM in plasma) is about 6.3-fold, which theoretically indicates that there would be little free LPC in plasma. However, many reports have described a signicant elevation of LPC levels in cells and tissues in different diseases [30,32] Therefore, we speculate that

GPR119 would be activated by LPC locally elevated via some kind of stimulation in vivo. We need further investigations to solve this issue. The unique tissue distribution of GPR119 is very similar to that of GPR40 for long chain fatty acids, including arachidonic acid and oleic acid [33]. The long chain fatty acids amplify glucose-stimulated insulin secretion from pancreatic b-cells by activating GPR40, as demonstrated in the case of GPR119. GPR119 diers from GPR40 in terms of coupled G-protein and ligand-selectivity. GPR40 couples mainly with Gq-protein and partially with Gi-protein, on the other hand, GPR119 coupled with Gs-protein (Fig. 3A). GPR40 is activated by arachidonic acid and oleic acid at a few micromolar, while GPR119 was not activated by them (Table 1). In view of the fact that both LPLs and free fatty acids potentiate glucose-stimulated insulin secretion, phospholipase-A2 (PLA2), which hydrolyzes the sn-2 ester bonds of phospholipids to yield LPLs and free fatty acids, may be involved in this function. Indeed, PLA2 has been shown to stimulate insulin secretion [11,34], activators of PLA2 have stimulated insulin release [35], and inhibitors of PLA2 generally reduce insulin secretion [35,36]. In summary, we have identied GPR119 as a novel LPC receptor involved in insulin secretion. We would

750

T. Soga et al. / Biochemical and Biophysical Research Communications 326 (2005) 744751 [16] Y. Xu, Sphingosylphosphorylcholine and lysophosphatidylcholine: G protein-coupled receptors and receptor-mediated signal transduction, Biochim. Biophys. Acta 1582 (2002) 8188. [17] J. Takasaki, M. Kamohara, M. Matsumoto, T. Saito, T. Sugimoto, T. Ohishi, H. Ishii, T. Ota, T. Nishikawa, Y. Kawai, Y. Masuho, T. Isogai, Y. Suzuki, S. Sugano, K. Furuichi, The molecular characterization and tissue distribution of the human cysteinyl leukotriene CysLT(2) receptor, Biochem. Biophys. Res. Commun. 274 (2000) 316322. [18] J. Takasaki, M. Kamohara, T. Saito, M. Matsumoto, S. Matsumoto, T. Ohishi, T. Soga, H. Matsushime, K. Furuichi, Molecular cloning of the platelet P2T(AC) ADP receptor: pharmacological comparison with another ADP receptor, the P2Y(1) receptor, Mol. Pharmacol. 60 (2001) 432439. [19] T. Soga, S. Matsumoto, T. Oda, T. Saito, H. Hiyama, J. Takasaki, M. Kamohara, T. Ohishi, H. Matsushime, K. Furuichi, Molecular cloning and characterization of prokineticin receptors, Biochim. Biophys. Acta 1579 (2002) 173179. [20] T. Soga, M. Kamohara, J. Takasaki, S. Matsumoto, T. Saito, T. Ohishi, H. Hiyama, A. Matsuo, H. Matsushime, K. Furuichi, Molecular identication of nicotinic acid receptor, Biochem. Biophys. Res. Commun. 303 (2003) 364369. [21] R. Fredriksson, P.J. Hoglund, D.E. Gloriam, M.C. Lagerstrom, H.B. Schioth, Seven evolutionarily conserved human rhodopsin G protein-coupled receptors lacking close relatives, FEBS Lett. 554 (2003) 381388. [22] G.M. Grodsky, R.E. Fanska, The in vitro perfused pancreas, Methods Enzymol. 39 (1975) 364372. [23] S. Mizushima, S. Nagata, pEF-BOS, a powerful mammalian expression vector, Nucleic Acids Res. 18 (1990) 5322. [24] M. Matsumoto, M. Kamohara, T. Sugimoto, K. Hidaka, J. Takasaki, T. Saito, M. Okada, T. Yamaguchi, K. Furuichi, The novel G-protein coupled receptor SALPR shares sequence similarity with somatostatin and angiotensin receptors, Gene 248 (2000) 183189. [25] M. Kamohara, J. Takasaki, M. Matsumoto, T. Saito, T. Ohishi, H. Ishii, K. Furuichi, Molecular cloning and characterization of another leukotriene B4 receptor, J. Biol. Chem. 275 (2000) 27000 27004. [26] C. Lippe, C. Ardizzone, Actions of vasopressin and isoprenaline on the ionic transport across the isolated frog skin in the presence and the absence of adenyl cyclase inhibitors MDL12330A and SQ22536, Comp. Biochem. Physiol. C 99 (1991) 209211. [27] W.J. Tang, A.G. Gilman, Adenylyl cyclases, Cell 70 (1992) 869 872. [28] W.T. Shier, J.H. Baldwin, M. Nilsen-Hamilton, R.T. Hamilton, N.M. Thanassi, Regulation of guanylate and adenylate cyclase activities by lysolecithin, Proc. Natl. Acad. Sci. USA 73 (1976) 15861590. [29] M. Kriat, J. Vion-Dury, S. Confort-Gouny, R. Favre, P. Viout, M. Sciaky, H. Sari, P.J. Cozzone, Analysis of plasma lipids by NMR spectroscopy: application to modications induced by malignant tumors, J. Lipid Res. 34 (1993) 10091019. [30] M. Okita, D.C. Gaudette, G.B. Mills, B.J. Holub, Elevated levels and altered fatty acid composition of plasma lysophosphatidylcholine (lysoPC) in ovarian cancer patients, Int. J. Cancer 71 (1997) 3134. [31] M. Croset, N. Brossard, A. Polette, M. Lagarde, Characterization of plasma unsaturated lysophosphatidylcholines in human and rat, Biochem. J. 345 (2000) 6167. [32] A.A. Murphy, N. Santanam, A.J. Morales, S. Parthasarathy, Lysophosphatidylcholine, a chemotactic factor for monocytes/Tlymphocytes is elevated in endometriosis, J. Clin. Endocrinol. Metab. 83 (1998) 21102113. [33] Y. Itoh, Y. Kawamata, M. Harada, M. Kobayashi, R. Fujii, S. Fukusumi, K. Ogi, M. Hosoya, Y. Tanaka, H. Uejima, H. Tanaka

expect that this nding would facilitate the understanding of the insulin secretion mechanism and the development of an improved anti-diabetic drug.

Acknowledgments We thank Drs. K. Suzuki, K. Tsunoyama, and M. Naitou for their bioinformatic analysis, Ms. M. Isshiki, Y. Abe, A. Matsuo, and E. Watanabe for their expert technical assistance, and Drs. T. Shimokawa and M. Yokono for helpful advice. We are grateful to Dr. S. Nagata for providing the pEF-BOS vector.

References
[1] W.H. Moolenaar, LPA: a novel lipid mediator with diverse biological actions, Trends Cell Biol. 4 (1994) 213219. [2] A.J. Lusis, Atherosclerosis, Nature 407 (2000) 233241. [3] G. Murugesan, M.R. Sandhya Rani, C.E. Gerber, C. Mukhopadhyay, R.M. Ransoho, G.M. Chisolm, K. Kottke-Marchant, Lysophosphatidylcholine regulates human microvascular endothelial cell expression of chemokines, J. Mol. Cell. Cardiol. 35 (2003) 13751384. [4] N. Yamamoto, S. Homma, I. Millman, Identication of the serum factor required for in vitro activation of macrophages. Role of vitamin D3-binding protein (group specic component, Gc) in lysophospholipid activation of mouse peritoneal macrophages, J. Immunol. 147 (1991) 273280. [5] A. Bruni, G. Monastra, F. Bellini, G. Toano, Autacoid properties of lysophosphatidylserine, Prog. Clin. Biol. Res. 282 (1988) 165179. [6] E. Kostenis, Novel clusters of receptors for sphingosine-1-phosphate, sphingosylphosphorylcholine, and (lyso)-phosphatidic acid: new receptors for old ligands, J. Cell. Biochem. 92 (2004) 923936. [7] J.H. Kabarowski, K. Zhu, L.Q. Le, O.N. Witte, Y. Xu, Lysophosphatidylcholine as a ligand for the immunoregulatory receptor G2A, Science 293 (2001) 702705. [8] K. Zhu, L.M. Baudhuin, G. Hong, F.S. Williams, K.L. Cristina, J.H. Kabarowski, O.N. Witte, Y. Xu, L.Q. Le, Sphingosylphosphorylcholine and lysophosphatidylcholine are ligands for the G protein-coupled receptor GPR4, J. Biol. Chem. 276 (2001) 4132541335. [9] M.G. Ludwig, M. Vanek, D. Guerini, J.A. Gasser, C.E. Jones, U. Junker, H. Hofstetter, R.M. Wolf, K. Seuwen, Proton-sensing G-protein-coupled receptors, Nature 425 (2003) 9398. [10] N. Murakami, T. Yokomizo, T. Okuno, T. Shimizu, G2A is a proton-sensing G-protein-coupled receptor antagonized by lysophosphatidylcholine, J. Biol. Chem. 279 (2004) 4248442491. [11] W.Y. Fujimoto, S.A. Metz, Phasic eects of glucose, phospholipase A2, and lysophospholipids on insulin secretion, Endocrinology 120 (1987) 17501757. [12] W.Y. Fujimoto, J. Teague, Phasic eects of glucose, p-hydroxymercuribenzoate, and lysophosphatidylcholine on insulin secretion from HIT cells, Diabetes 38 (1989) 625628. [13] D.J. Drucker, Glucagon-like peptides, Diabetes 47 (1998) 159169. [14] R.G. Yip, M.M. Wolfe, GIP biology and fat metabolism, Life Sci. 66 (2000) 91103. [15] Y. Yuan, S.M. Schoenwaelder, H.H. Salem, S.P. Jackson, The bioactive phospholipid, lysophosphatidylcholine, induces cellular eects via G-protein-dependent activation of adenylyl cyclase, J. Biol. Chem. 271 (1996) 2709027098.

T. Soga et al. / Biochemical and Biophysical Research Communications 326 (2005) 744751 M. Maruyama, R. Satoh, S. Okubo, H. Kizawa, H. Komatsu, F. Matsumura, Y. Noguchi, T. Shinohara, S. Hinuma, Y. Fujisawa, M. Fujino, Free fatty acids regulate insulin secretion from pancreatic beta cells through GPR40, Nature 422 (2003) 173176. [34] S. Yamamoto, T. Nakaki, T. Nakadate, R. Kato, Insulinotropic eects of exogenous phospholipase A2 and C in isolated pancreatic islets, Eur. J. Pharmacol. 86 (1982) 121124.

751

[35] S. Yamamoto, T. Nakadate, T. Nakaki, K. Ishii, R. Kato, Tumor promotor 12-O-tetradecanoylphorbol-13-acetate-induced insulin secretion: inhibition by phospholipase A2-and lipoxygenaseinhibitors, Biochem. Biophys. Res. Commun. 105 (1982) 759765. [36] S.G. Laychock, Fatty acid incorporation into phospholipids of isolated pancreatic islets of the rat. Relationship to insulin release, Diabetes 32 (1983) 613.