Ghrelin Modulates the Downstream Molecules of Insulin Signaling in Hepatoma Cells*

Ghrelin was identified in the stomach as an endogenous ligand specific for the growth hormone secretagogue receptor (GHS-R). GHS-R is found in various tissues, but its function is unknown. Here we show that GHS-R is found in hepatoma cells. Exposure of these cells to ghrelin caused up-regulation of several insulin-induced activities including tyrosine phosphorylation of insulin receptor substrate-1 (IRS-1), association of the adapter molecule growth factor receptor-bound protein 2 with IRS-1, mitogen-activated protein kinase activity, and cell proliferation. Unlike insulin, ghrelin inhibited Akt kinase activity as well as up-regulated gluconeogenesis. These findings raise the possibility that ghrelin modulates insulin activities in humans.

up of 50 mM Tris-HCl, 75 mM KCl, 3 mM MgCl 2 , 10 mM dithiothreitol, pH 8.3, 1 mM dNTPs, 15 pmol of antisense primer, and 200 units of Moloney murine leukemia virus RT (M-MLV, Promega). The PCR amplifications were performed in 50 l of medium containing 50 mM Tris-HCl, 50 mM KCl, 1 mM MgCl 2 , pH 9.0, 0.2 mM dNTPs, 15 pmol of sense and antisense primers, and 2.5 units of Taq DNA polymerase (Promega). The reaction mixtures were subjected to 40 cycles of PCR amplification of GHS-R cDNA consisting of denaturation for 60 s at 94°C, annealing for 60 s at 45°C, and elongation for 90 s at 72°C. The oligonucleotide primers (5Ј-CTCTGCATGCCCCTGGACCTCGTTCG-C-3Ј sense and 5Ј-CTGCCGATGAGACTGTAGAGGACCGTGAGAC-3Ј antisense) allowed amplification of a 429-bp GHS-R cDNA product and were verified with DNA sequencing (model 310, PerkinElmer Life Sciences). The reaction mixtures were subjected to 40 cycles of PCR amplification of PEPCK cDNA consisting of denaturation for 60 s at 94°C, annealing for 60 s at 56°C, and elongation for 120 s at 72°C. The oligonucleotide primers (5Ј-AGCCTCGACAGCCTGCCCCAGG-3Ј sense and 5Ј-CCAGTTGTTGACCAAAGGCTTTT-3Ј antisense) allowed amplification of a 575-bp PEPCK cDNA product (10), and the product was verified with DNA sequencing. The ␤-actin sequence of the upstream primer was 5Ј-TGACCCAGATCATGTTTGAGAGACC-3Ј and of the downstream primer was 5Ј-CCATACCCAAGAAGGAAGGC-3Ј. After an initial denaturation for 30 s at 94°C, PCR was performed for 40 cycles The conditions for PCR were 94°C for 60 s, 60°C for 60 s, 72°C for 60 s. The rat ␤-actin-coding region was obtained by RT-PCR and verified with DNA sequencing.
Western Blot Analysis-Cells were lysed in 20 mM Tris-HCL, pH 7.4, 137 mM NaCl, 10% glycerol, 1% Nonidet P-40, 1 mM Na 3 VO 4 , 1 mM phenylmethylsulfonyl fluoride, 1 mg/ml leupeptin, and 1 mg/ml aprotinin for 30 min on ice and cleared by centrifugation for 20 min at 4°C. Immunoprecipitation was performed by incubating the supernatant fraction with antibodies specific for IRS-1, IR␤, IRS-2, GRB2, or p85 PI3K overnight at 4°C, followed by binding to protein A-agarose (Amersham Life Sciences). After a 2-h incubation at 4°C, the incubation was centrifuged briefly at 10000 ϫ g, and the agarose was washed with 20 mM Hepes, pH 7.0, 10% glycerol, 0.1% Triton X-100, 150 mM NaCl, 1 mM Na 3 VO 4 and recentrifuged. This was repeated three times after which Laemmli medium was added, and the agarose was heated at 100°C for 5 min. Immunoprecipitated proteins were fractionated on 7.5% acrylamide gel and blotted to polyvinylidene difluoride membrane for 30 min. Immunoblot was blocked by incubation in TBST (10 mM Tris-HCl, pH 7.5, 150 mM NaCl, 0.1% Tween 20) containing 5% skim milk for 1 h. Blots were incubated with horseradish peroxidase-conjugated antiphosphotyrosine antibodies RC20 for 1 h to detect tyrosinephosphorylated IRS-1, IRS-2, and IR␤. Blots were incubated with anti-IRS-1 for 1 h and then with horseradish peroxidase-conjugated antirabbit IgG for 1 h to detect association of GRB2 or PI3K with IRS-1. A chemiluminescent peroxidase substrate (ECL, Amersham Biosciences, Inc.) was applied according to the manufacturer's instructions, and the membranes were exposed briefly to x-ray film.
MAPK in Vitro Kinase Assays-MAPK activity of HepG2 cells was measured by using a MAPK assay kit according to the manufacturer's instructions. In short, the cells were washed with phosphate-buffered saline, lysed in lysis buffer (20 mM Tris (pH 7.5), 150 mM NaCl, 1 mM EDTA, 1 mM EGTA, 1% Triton X-100, 2.5 mM sodium pyrophosphate, 1 mM ␤-glycerophosphate, 1 mM Na 3 VO 4 , 1 l/ml leupeptin, 1 mM phenylmethylsulfonyl fluoride), scraped off the dishes, transferred to microcentrifuge tubes, sonicated, and microcentrifuged for 10 min at 4°C. The supernatant was used for immunoprecipitation of activated MAPKs. The supernatant was incubated with immobilized phospho-p44/42 MAPK (Thr 202 /Tyr 204 ) monoclonal antibody for 4 h and microcentrifuged for 30 s at 4°C. The pellets were washed twice with ice-cold lysis buffer and twice with kinase buffer (25 mM Tris (pH 7.5), 5 mM ␤-glycerophosphate, 2 mM dithiothreitol, 0.1 mM Na 3 VO 4 , 10 mM MgCl 2 ). The pellets were incubated with 200 M ATP and 2 g of Elk-1 fusion protein for 30 min at 30°C. The reaction was terminated by 25 l of 3ϫ SDS sample buffer (187.5 mM Tris-HCl (pH 6.8), 6% w/v SDS, 30% glycerol, 150 mM dithiothreitol, 0.3% w/v bromphenol blue). Samples were boiled, separated by electrophoresis through a 10% SDS-polyacrylamide gel, and transferred to polyvinylidene difluoride membrane for 30 min. Immunoblot was blocked by incubation in TBST containing 5% skim milk at room temperature for 1 h and then probed with 1:1000 dilution of anti-phospho-Elk-1 antibody. The membranes were washed three times in TBST, incubated with a 1:1000 dilution of horseradish peroxidase-linked anti-rabbit IgG antibody, and then reacted with Lu-miGLO reagent according to the manufacturer's instructions. The membranes were exposed briefly to x-ray film.
Akt in Vitro Kinase Assays-Akt kinase activity of HepG2 cells was measured by using Akt kinase assay kit according to the manufacturer's instructions. In short, the cells were washed with phosphate-buffered saline, lysed in lysis buffer (20 mM Tris (pH 7.5), 150 mM NaCl, 1 mM EDTA, 1 mM EGTA, 1% Triton X-100, 2.5 mM sodium pyrophosphate, 1 mM ␤-glycerophosphate, 1 mM Na 3 VO 4 , 1 l/ml leupeptin, 1 mM phenylmethylsulfonyl fluoride), scraped off the dishes, transferred to microcentrifuge tubes, sonicated, and microcentrifuged for 10 min at 4°C. The supernatant was used for immunoprecipitation of activated Akt kinase. The supernatant was incubated with immobilized Akt antibody for 2-3 h and microcentrifuged for 30 s at 4°C. The pellets were washed twice with ice-cold lysis buffer and twice with kinase buffer (25 mM Tris (pH 7.5), 5 mM ␤-glycerophosphate, 2 mM dithiothreitol, 0.1 mM Na 3 VO 4 , 10 mM MgCl 2 ). The pellets were incubated with 200 M ATP and 1 g of GSK-3 fusion protein for 30 min at 30°C. The reaction was terminated by 20 l of 3ϫ SDS sample buffer (187.5 mM Tris-HCl (pH 6.8), 6% w/v SDS, 30% glycerol, 150 mM dithiothreitol, 0.03% (w/v) bromphenol blue). Samples were boiled, separated by electrophoresis through a 12% SDS-polyacrylamide gel, and transferred to polyvinylidene difluoride membrane for 30 min. Immunoblot was blocked by incubation in TBST containing 5% skim milk at room temperature for 1 h and then probed with 1:1000 dilution of anti-phospho-GSK-3 antibody. The membranes were washed three times in TBST, incubated with a 1:1000 dilution of horseradish peroxidase-linked antirabbit IgG antibody, and then reacted with LumiGLO reagent according to the manufacturer's instructions. The membranes were exposed briefly to x-ray film. Phosphatidylinositol 3-Kinase Activity-In vitro phosphorylation of PtdIns was carried out in the immune complexes as described previously (11). Subconfluent HepG2 cells grown in 100-mm dishes were made quiescent by overnight incubation in Dulbecco's modified Eagle's medium with 0.1% BSA. The quiescent cells were incubated with ghrelin for 20 min, insulin for 3 min, or ghrelin for 20 min and then insulin for 3 min, and washed once with ice-cold phosphate-buffered saline and twice with 20 mM Tris-HCl, pH 7.5, containing 137 mM NaCl, 1 mM MgCl 2 , 1 mM CaCl 2 , and 100 M Na 3 VO 4 (buffer A). The cells were solubilized in 1 ml of buffer A containing 1% Nonidet P-40, 10% glycerol, and 0.35 mg/ml phenylmethylsulfonyl fluoride, and insoluble material was removed by centrifugation at 13,000 ϫ g for 10 min. Supernatant was incubated with anti-IRS-1 antibody for 4 h at 4°C, and immune complexes were precipitated from the supernatant with protein A-Sepharose and washed successively in PBS containing 1% Nonidet P-40 and 100 M Na 3 VO 4 (three times); 100 mM Tris-HCl, pH 7.5, containing 500 mM LiCl and 100 M Na 3 VO 4 (three times); and 10 mM Tris-HCl, pH 7.5, containing 100 mM NaCl, 1 mM EDTA, and 100 M Na 3 VO 4 (two times). The pellets were resuspended in 50 l of 10 mM Tris-HCl, pH 7.5, containing 100 mM NaCl and 1 mM EDTA and combined with 10 l of 100 mM MnCl 2 and 10 l of 2 g/l PtdIns sonicated in 10 mM Tris-HCl (pH 7.5) containing 1 mM EGTA. The phosphorylation reaction was started by adding 10 l of 440 M ATP containing 30 Ci of [␥-32 P]ATP. After 10 min at 22°C, the reaction was stopped with 20 l of 8 N HCl and 160 l of CHCl 3 :methanol (1:1). The samples were centrifuged, and the lower organic phase was removed and applied to a silica gel TLC plate that had been coated with 1% potassium oxalate. TLC plates were developed in CHCl 3 :CH 3 OH:H 2 O:NH 4 OH (60:47:11.3: 2), dried, and visualized by autoradiography.
Northern Blot Analysis-Total RNA was prepared from cultured H4-II-E cells using Trizol. Twenty g of total RNA was heat-denatured for 5 min at 70°C in buffer containing 50% formamide, 2.4 M formaldehyde, 1ϫ 0.02 MOPS, 50 mM sodium acetate, and 10 mM Na 2 EDTA and then loaded onto a 1.2% agarose/formaldehyde gel made up in 1ϫ MOPS, pH7.4, at 50 V for 3-4 h. After electrophoresis, RNA was transferred onto a Hybond N membrane by capillary action and fixed onto the filters by ultraviolet light cross-linking. The blot was probed with either a rat PEPCK or a rat ␤-actin probe labeled by a Gene Images random prime labeling module and detected by a Gene Images CDP-star detection module according to the manufacturer's instructions, and the membranes were exposed to x-ray film for 30 -60 min.
Cell Proliferation Assays-Cell proliferation assays were performed by CellTiter 96 AQueous One Solution Cell Proliferation Assay kit according to the manufacturer's instructions (Invitrogen). Briefly, ϳ48 -72 h before the assay, HepG2 cells were seeded into 96-well plates at 1-3 ϫ 10 5 cells/well. When cells reached 60 -70% confluence, the medium was replaced with Dulbecco's modified Eagle's medium containing 0.1% BSA and then with 100 nM ghrelin, 100 nM insulin, both 100 nM ghrelin and 50 M PD98059, or both 100 nM insulin and 50 M PD98059. The cultures were incubated in a CO 2 incubator for 48 h at 37°C, and then 20 l/well of CellTiter 96 AQueous One Solution Reagent, which contains a tetrazolium compound, methanethiosulfonate, was added to the culture medium. The methanethiosulfonate tetrazolium compound is bioreduced by cells into a colored formazan product that is soluble in tissue culture medium. This conversion is presumably accomplished by NADPH or NADH produced by dehydrogenase enzymes in metabolically active cells. After incubating the plate 2-4 h in a CO 2 incubator, the absorbance at 490 nm was recorded with a 96-well plate reader. All samples were assayed in duplicate, and each experiment was repeated at least three times.
Statistical Analysis-The data were expressed as the mean Ϯ S.E. unless noted otherwise. Statistics were analyzed by one-way repeated measures analysis of variance with a significance level of 0.05.

RESULTS
To investigate the possible effects of ghrelin on insulin-regulated responses, we looked for cell lines expressing a GHS-R.
Various cell lines derived from the liver, adipose tissue, and muscle were screened by RT-PCR with oligonucleotides that have been reported previously (12). A human hepatocellular carcinoma cell line, HepG2 cells, provided one PCR product, for which identity with human GHS-R mRNA was confirmed by DNA sequencing. The same product was obtained by RT-PCR from a human pituitary cDNA library, a human liver cDNA library, and a rat hepatoma cell line, H4-II-E cells (Fig. 1A), and its identity was confirmed by sequencing.
We next investigated the effect of ghrelin on the profile of tyrosine-phosphorylated cellular proteins. HepG2 cells were treated with 100 nM ghrelin for 0 -20 min, and cellular proteins were analyzed by immunoblot analysis with antibodies to phos- photyrosine (anti-Tyr(P)). Ghrelin treatment for 10 -20 min caused a significant increase in the amount of tyrosine-phosphorylated IRS-1 in HepG2 cells compared with those without ghrelin treatment (354 Ϯ 42 versus 100 Ϯ 5% basal level at 10 min, p Ͻ 0.01; 364 Ϯ 47 versus 106 Ϯ 5% basal level at 20 min, p Ͻ 0.01) (Fig. 1B). When we cultured HepG2 cells for 20 min with varying concentrations of ghrelin, 10 -100 nM ghrelin caused a significant and dose-dependent increase in the amount of tyrosine-phosphorylated IRS-1 as shown in Fig. 1C. Furthermore, ghrelin-induced tyrosine phosphorylation of HepG2 cells were treated with 100 nM ghrelin for 20 min followed by stimulation with 100 nM insulin for 1 min, and cellular proteins were analyzed by immunoblot analysis with anti-Tyr(P). Ghrelin and insulin significantly increased tyrosine-phosphorylated IRS-1 levels up to 172 Ϯ 10 and 262 Ϯ 15 densitometric units (DU), respectively, compared with vehicle alone (99 Ϯ 9 DU). Combined stimulation with ghrelin and insulin resulted in an additive increase of tyrosine-phosphorylated IRS-1 levels up to 442 Ϯ 23 DU (p Ͻ 0.01, n ϭ 5) (Fig. 2A). The tyrosine phosphorylation of IR␤ chain was not stimulated by the presence of ghrelin (Fig. 2B). Ghrelin also increased tyrosine phosphorylation of IRS-1 in H4-II-E cells (Fig. 2D). To investigate whether the effect of ghrelin on tyrosine phosphorylation is specific for IRS-1, we tested the effect of ghrelin on tyrosine phosphorylation of IRS-2. HepG2 cells were treated with 100 nM ghrelin for 20 min followed by stimulation with 100 nM insulin for 1 min. Ghrelin did not increase tyrosine phosphorylation of IRS-2 (Fig. 2C).
Furthermore, because MAPKs and Akt are the downstream substrates for GRB2 and PI3K, respectively, we tested whether MAPKs and Akt are involved in cellular responses to ghrelin. Phosphorylated active MAPK was collected from cell lysates using anti-phospho-MAPK antibody, and its enzyme activity was determined by the amount of phosphorylated Elk-1 fusion protein. HepG2 cells were pretreated with 100 nM ghrelin for 20 min followed by stimulation with 100 nM insulin for 3 min and were then analyzed for MAPK activity. MAPK activity in vehicle-treated cells (75 Ϯ 14 DU) was increased either by ghrelin (246 Ϯ 24 DU, p Ͻ 0.01) or by insulin (389 Ϯ 36 DU, p Ͻ 0.01). Combined treatment with ghrelin and insulin caused an additive increase (546 Ϯ 58 DU, p Ͻ 0.01, n ϭ 5) (Fig. 4A). These response patterns of MAPK activity to either insulin, ghrelin, or the combination of both were compatible with those of GRB2-associated IRS-1 (Fig. 3A). We also measured the Akt kinase activity determined by the amount of phosphorylated glycogen synthase kinase-3 (GSK-3) fusion protein. Similarly treated cells were analyzed for Akt kinase activity. Akt kinase activity in vehicle-treated cells (154 Ϯ 24 DU) was increased by insulin (524 Ϯ 126 DU, p Ͻ 0.01) and, in contrast, decreased by ghrelin (29 Ϯ 8 DU, p Ͻ 0.01) as well as by a combined treatment with ghrelin and insulin (82 Ϯ 17 DU, p Ͻ 0.01, n ϭ 5) (Fig. 4B). These findings were not correlated with the association of PI3K with IRS-1. We therefore measured the amount of PI3K activity in IRS-1 immunoprecipitates. IRS-1associated PI3K activity in vehicle-treated cells (64 Ϯ 10 DU) was increased either by ghrelin (200 Ϯ 19 DU, p Ͻ 0.01) or by insulin (203 Ϯ 24 DU, p Ͻ 0.01). However, no additive increase in IRS-1-associated PI3K activity was found by combined treatment with ghrelin and insulin (226 Ϯ 7 DU, p Ͻ 0.01, n ϭ 4) (Fig. 4C).
We also investigated whether ghrelin affects glucose homeostasis in cell culture. Hepatic and renal gluconeogenesis is crucially important in maintaining glucose homeostasis. The rate-limiting enzyme of gluconeogenesis is phosphoenolpyruvate carboxykinase (PEPCK). This enzyme has no known allosteric control and is down-regulated by insulin at the transcriptional level. The rat hepatoma cell line, H4-II-E cells, has been used successfully to study the regulation of PEPCK expression, whereas HepG2 cells do not express PEPCK efficiently (14 -16). The amount of PEPCK mRNA in H4-II-E cells treated first with 8-CPT-cAMP and then with insulin was reduced compared with cells treated with 8-CPT-cAMP alone. Surprisingly, incubation of the cells with ghrelin for 1 to 2 h before adding insulin partially reversed the down-regulating effect of insulin on PEPCK mRNA levels (Fig. 5). Another main regulator of PEPCK gene expression is cAMP. Hence, we investigated the intracellular cAMP using the enzyme immunoassay for cAMP. Ghrelin did not increase the intracellular cAMP in H4-II-E cells (data not shown).
Next we tested the effects of ghrelin on cell proliferation. On the basis of the analogous data between ghrelin and insulin with regard to the stimulation of MAPK activity in HepG2 cells, we hypothesized that ghrelin causes cell proliferation in HepG2 cells via MAPK pathway. Ghrelin stimulated proliferation of HepG2 cells, and a MAPK kinase-1-specific inhibitor, PD98059, completely blocked both ghrelin-and insulin-induced cell proliferation (Fig. 6). DISCUSSION We have demonstrated for the first time that ghrelin treatment causes stimulation of the IRS-1-GRB2-MAPK pathway as well as cell proliferation and that ghrelin inhibits Akt kinase activity as well as up-regulates PEPCK gene expression.
The effect of ghrelin on IRS-1 phosphorylation is unlikely to be mediated via IR, because tyrosine phosphorylation of the IR␤ chain was not stimulated by the presence of ghrelin. However, it would be exerted independently by an activation of a GHS-R signaling cascade. We found that ghrelin-induced tyrosine phosphorylation of IRS-1 was blunted by an antagonist for GHS-R, [D-Lys-3]GHRP-6, and that other GHSs, GHRP-6 and GHRP-2, induced a significant increase in the amount of tyrosinephosphorylated IRS-1 in HepG2 cells (data not shown). Therefore, the downstream molecules of GHS-R signaling can crosstalk with the insulin-signaling pathway. Indeed, it has been reported that IRS-1 is phosphorylated by growth factors and cytokines, including insulin-like growth factor-I, interferon-␣, interleukin-4, and interleukin-9 as well (11,(17)(18)(19)(20). It is unique, however, that GHS-R is the G protein-coupled receptor that cross-talks with the insulin-signaling pathway. Hence, it remains to be elucidated what molecules in the GHS-R signaling pathway affect the tyrosine phosphorylation of IRS-1 (Fig. 7).

FIG. 4. Ghrelin up-regulates MAPK activity and down-regulates Akt kinase activity, whereas it up-regulates IRS-1-associated PI3K
activity. HepG2 cells were treated as described in the legend for Fig. 3A. A, the basal level of MAPK activity, as determined by the amount of phosphorylated Elk-1, was increased both by ghrelin and by insulin and was further increased by a combined treatment with both ghrelin and insulin. Statistical significance is shown by asterisks: *, p Ͻ 0.01 versus control; **, p Ͻ 0.01 versus insulin-treated cells. B, the basal amount of Akt kinase activity, as determined by the amount of phosphorylated GSK-3, was increased by insulin and, in contrast, was decreased both by ghrelin and by a combined treatment with ghrelin and insulin. Statistical significance is shown by asterisks: *, p Ͻ 0.01 versus control; **; p Ͻ 0.01 versus insulintreated cells. C, phosphorylation of Pt-dIns was carried out in the immune complexes as described under "Experimental Procedures." The basal amount of IRS-1associated PI3K activity was increased both by ghrelin and by insulin as well as by a combined treatment with both ghrelin and insulin. The representative result is shown in the left panel. In the right panel, all values are the mean Ϯ S.E. of the amount of IRS-1-associated PI3K activity after densitometric analysis (n ϭ 4). Statistical significance is shown by asterisks: *, p Ͻ 0.01 versus control.
IRS-1 is characterized to possess the 20 -22 potential tyrosine phosphorylation sites that are conserved between IRS-1 homologs. The surrounding amino acid residues are also highly conserved, and several of these represent potential binding sites for proteins that contain Src homology 2 (SH2) domains (21,22). IRS-1 interacts with many SH2 proteins with diverse phosphotyrosine motif requirements including PI3K and GRB2 (13,21). In the present study, ghrelin increased the tyrosine phosphorylation of IRS-1, association of GRB2 with IRS-1, and MAPK activity, indicating up-regulation of the IRS-1-GRB2-MAPK pathway. Furthermore, ghrelin-stimulated proliferation of HepG2 cells and PD98059 completely blocked ghrelininduced cell proliferation, indicating that MAPKs were essential in HepG2 cell proliferation caused by ghrelin. However, ghrelin suppressed Akt kinase activity, despite of the presence of insulin, as well as up-regulating the amount of PEPCK mRNA expression, although it increased not only the association of PI3K with IRS-1 but also IRS-1-associated PI3K activity.
Furthermore, the activity of effector proteins that are dependent on PI3K activation can be negatively regulated by PTEN (phosphate and tensin homolog deleted on chromosome PI3K enzymes can phosphorylates the 3Ј position of PtdIns, PtdIns(4)P, and PtdIns(4,5)P 2 to produce PtdIns-3P, PtdIns(3,4)P 2 , and PtdIns(3,4,5)P 3 , respectively. PTEN and SHIP dephosphorylate the 3Ј and 5Ј positions of the inositol ring of phosphoinositides, respectively, and PTEN reverses the reaction catalyzed by PI3K. PtdIns(3,4)P 2 and PtdIns(3,4,5)P 3 recruit the protein-serine/ threonine kinases Akt and PDK-1 to the membrane and induce a conformational change in Akt, exposing the activation loop. Phosphorylation of Akt at Thr-308 of the activation loop by PDK-1 turn on the protein kinase activity. Phosphorylation of Akt at the C-terminal site (by PDK-2?) causes further activation. PTEN turns off the pathway by dephosphorylating the 3Ј position of PtdIns(3,4)P 2 and PtdIns(3,4,5)P 3 . ten) and SHIP (SH2-containing inositol 5Ј-phosphatase), two phosphoinositide-specific phosphatases that dephosphorylate the 3Ј and 5Ј positions of the inositol ring of phosphoinositides, respectively, leading to inhibition of cellular responses mediated by PI3K products (26,27). In the present study, the dissociation of the downstream molecules in the IRS-1-PI3K-Akt pathway remains difficult to explain, but the presence of a potent inhibiting mechanism of Akt kinase activity by ghrelin, even under full activation by the PI3K-IRS-1 association, is likely. It is possible that these enzymes, such as PDK-1, PDK-2, SHIP, and PTEN, may be affected by GHS-R-mediated signaling molecules. Hence, it remains to be elucidated what molecules in the GHS-R signaling pathway affect the IRS-1-PI3K-Akt pathway (Fig. 7).
The repression of the PEPCK gene by insulin has been studied in detail. There is conflicting evidence as to which signaling pathways may be involved in insulin repression of PEPCK gene expression, the activation of the PI3K pathway, or the activation of the MAPK pathway in which GRB2 engages IRS-1, IRS-2, Shc, or SHP2 (28 -30). Recently, however, accumulating evidence suggests that the signaling pathways in insulin repression of PEPCK gene expression may involve the activation of PI3K pathway but not the MAPK pathway (30 -32). Our findings that ghrelin inhibits Akt activity may be in agreement with the recent data that PI3K pathway is involved in the insulin-induced repression mechanism of PEPCK expression. Another main regulator of PEPCK gene expression is cAMP. However, ghrelin did not increase the intracellular cAMP in H4-II-E cells (data not shown). Therefore, the signaling pathway by which ghrelin stimulates PEPCK gene expression remains as yet unknown. We propose that ghrelin can affect gluconeogenesis, at least in the H4-II-E cells, by attenuating the effect of insulin on the expression of PEPCK. Considering the effect of ghrelin on glucose homeostasis, it is of note that GHSs are diabetogenic in rats (33). Physiological significance of ghrelin in vivo animals is to be clarified.
In conclusion, we found two novel actions of ghrelin in addition to its GH-releasing action: one is insulin-like action stimulating the IRS-1-GRB2-MAPK pathway, which in turn activates cell proliferation; and the other is anti-insulin action suppressing Akt activity and up-regulation of gluconeogenesis. Although the mechanism by which ghrelin affects insulin signaling pathways remains not fully understood, our findings obtained in hepatoma cells strongly implicate the peripheral actions of ghrelin in glucose homeostasis and in mitogenic processes in humans.