Advertisement

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

Open AccessPublished:November 27, 2001DOI:https://doi.org/10.1074/jbc.M103898200
      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.
      Growth hormone (GH)
      The abbreviations used are:
      GH
      growth hormone
      GHS
      growth hormone secretagogue
      GHS-R
      growth hormone secretagogue receptor
      GHRP
      growth hormone-releasing peptide
      IRS
      insulin receptor substrate
      PI3K
      phosphatidylinositol 3-kinase
      GRB2
      growth factor receptor-bound protein 2
      GSK-3
      glycogen synthase kinase-3
      IRβ
      insulin receptor β
      MAPK
      mitogen-activated protein kinase
      PtdIns
      phosphatidylinositol
      8-CPT-cAMP
      8-(4-chlorophenylthio)adenosine 3′,5′-cyclic monophosphate
      BSA
      bovine serum albumin
      RT
      reverse transcriptase
      PEPCK
      phosphoenolpyruvate carboxykinase
      MOPS
      4-morpholinepropanesulfonic acid
      anti-Tyr(P)
      antibodies to phosphotyrosine
      DU
      densitometric units
      PDK-1
      phosphoinositide-dependent kinase-1
      PTEN
      phosphate and tensin homolog deleted on chromosome ten
      SH2
      Src homology 2
      1The abbreviations used are:GH
      growth hormone
      GHS
      growth hormone secretagogue
      GHS-R
      growth hormone secretagogue receptor
      GHRP
      growth hormone-releasing peptide
      IRS
      insulin receptor substrate
      PI3K
      phosphatidylinositol 3-kinase
      GRB2
      growth factor receptor-bound protein 2
      GSK-3
      glycogen synthase kinase-3
      IRβ
      insulin receptor β
      MAPK
      mitogen-activated protein kinase
      PtdIns
      phosphatidylinositol
      8-CPT-cAMP
      8-(4-chlorophenylthio)adenosine 3′,5′-cyclic monophosphate
      BSA
      bovine serum albumin
      RT
      reverse transcriptase
      PEPCK
      phosphoenolpyruvate carboxykinase
      MOPS
      4-morpholinepropanesulfonic acid
      anti-Tyr(P)
      antibodies to phosphotyrosine
      DU
      densitometric units
      PDK-1
      phosphoinositide-dependent kinase-1
      PTEN
      phosphate and tensin homolog deleted on chromosome ten
      SH2
      Src homology 2
      is synthesized and secreted from the anterior pituitary under complex regulation mechanisms. The two hypothalamic peptides, GH-releasing hormone and somatostatin, coordinately exert the positive and negative control of GH release, respectively (
      • Muller E.E.
      • Locatelli V.
      • Cocchi D.
      ). On the other hand, GH secretagogues (GHSs) were discovered as a series of small peptide derivatives of pentapeptides Leu- and Met-enkephaline, which selectively stimulated GH secretion from pituitary cells. The prototype of this class of GHSs, GHRP-6, was found to be extremely potent and specific in mammals and particularly in humans (
      • Momany F.A.
      • Bowers C.Y.
      • Reynolds G.A.
      • Hong A.
      • Newlander K.
      ). Non-peptidyl GHSs, L-692,429 (
      • Smith R.G.
      • Chen K.
      • Schoen W.R.
      • Pong S.-S.
      • Hickey G.J.
      • Jacks T.M.
      • Batler B.S.
      • Chan W.W.-S.
      • Chaung L.-Y.P.
      • Judith F.
      • Taylor A.M.
      • Wyvratt M.J., Jr.
      • Fisher M.H.
      ) and L-163,191 (MK-0677) (
      • Patchett A.A.
      • Nargund R.P.
      • Tata J.R.
      • Chen M-H.
      • Barakat K.J.
      • Johnston D.B.R.
      • Chen K.
      • Chan W.W.-S.
      • Butler B.S.
      • Hickey G.J.
      • Jacks T.M.
      • Scleim K.
      • Pong S-S.
      • Chaung L.-Y.P.
      • Chen H.Y.
      • Fraizier E.
      • Leung K.H.
      • Chui S.-H.L.
      • Smith R.G.
      ), were also manufactured to improve oral bioavailability. The GHS receptor (GHS-R) was cloned by the robust expression system that pig pituitary poly(A)+RNA was coinjected into Xenopus oocytes together with cDNA encoding Gα11 (
      • Howard A.D.
      • Feighner S.D.
      • Cully D.F.
      • Arena J.P.
      • Liberator P.A.
      • Rosenblum C.I.
      • Hamelin M.J.
      • Hreniuk D.L.
      • Palyha O.C.
      • Anderson J.
      • Paress P.S.
      • Diaz C.
      • Chou M.
      • Liu K.
      • McKee K.K.
      • Pong S.-S.
      • Chang L.-Y.
      • Elbrecht A.
      • Dashkevicz M.
      • Heavens R.
      • Rigby M.
      • Sirinathsinghji D.J.S.
      • Dean D.C.
      • Mellilo D.G.
      • Patchett A.A.
      • Nargund R.
      • Griffin P.R.
      • DeMartino J.A.
      • Gupta S.K.
      • Schaeffer J.M.
      • Smith R.G.
      • Van der Ploeg L.H.T.
      ) and subsequently in rats (
      • McKee K.K.
      • Palyha O.C.
      • Feighner S.D.
      • Hreniuk D.L.
      • Tan C.P.
      • Phillips M.S.
      • Smith R.G.
      • Van der Ploeg L.H.
      • Howard A.D.
      ). GHS-R was prominently expressed in several hypothalamic nuclei and also in the dentate gyrus and CA2 layers of the hippocampus (
      • Howard A.D.
      • Feighner S.D.
      • Cully D.F.
      • Arena J.P.
      • Liberator P.A.
      • Rosenblum C.I.
      • Hamelin M.J.
      • Hreniuk D.L.
      • Palyha O.C.
      • Anderson J.
      • Paress P.S.
      • Diaz C.
      • Chou M.
      • Liu K.
      • McKee K.K.
      • Pong S.-S.
      • Chang L.-Y.
      • Elbrecht A.
      • Dashkevicz M.
      • Heavens R.
      • Rigby M.
      • Sirinathsinghji D.J.S.
      • Dean D.C.
      • Mellilo D.G.
      • Patchett A.A.
      • Nargund R.
      • Griffin P.R.
      • DeMartino J.A.
      • Gupta S.K.
      • Schaeffer J.M.
      • Smith R.G.
      • Van der Ploeg L.H.T.
      ). In searching an endogenous ligand for GHS-R, however, all efforts to use the brain extracts proved fruitless.
      Recently, an endogenous ligand for GHS-R, named ghrelin, was purified from the extracts of the stomach and found to be abundant exclusively in the stomach (
      • Kojima M.
      • Hosoda H.
      • Date Y.
      • Nakazato M.
      • Matsuo H.
      • Kangawa K.
      ). GHS-R mRNA is expressed not only in the pituitary and brain but also in other tissues such as pancreas (
      • Guan X.-M., Yu, H.
      • Palyha O.C.
      • Feighner S.D.
      • Sirinathsinghji D.J.S.
      • Smith R.G.
      • Van der Pleog L.H.T.
      • Howard A.D.
      ), suggesting that ghrelin may have other physiological functions in addition to the regulation of GH release. Furthermore, a very recent report showed that ghrelin caused hyperphagia and obesity (
      • Tschàp M.
      • Smiley D.L.
      • Heiman M.L.
      ). These findings let us to explore the possibility that ghrelin may play some role in glucose homeostasis and metabolism and modulate insulin action.

      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 (
      • Skinner M.M.
      • Nass R.
      • Lopes B.
      • Laws E.R.
      • Thorner M.O.
      ). 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. 1 A), and its identity was confirmed by sequencing.
      Figure thumbnail gr1
      Figure 1The ghrelin receptor GHS-R is found in hepatoma cells, and ghrelin up-regulates tyrosine phosphorylation of IRS-1. A, RT-PCR of mRNA from human pituitary (lane 1), human liver (lane 2), HepG2 cells (lane 3), H4-II-E cells (lane 4), 3T3-L1 adipocytes (lane 5), and L6 myocytes (lane 6).B, time-dependent effects of ghrelin on tyrosine phosphorylation of IRS-1. Serum-starved HepG2 cells were treated with 100 nm ghrelin for 0, 3, 10, and 20 min. Cell extracts were immunoprecipitated (IP) with antibodies to IRS-1 and immunoblotted with anti-Tyr(P) (Anti-pTyr). The representative result is shown in the upper panel. In thelower panel, all values are expressed as the mean ± S.E. of tyrosine-phosphorylated IRS-1 after densitometric analysis (n = 5). HepG2 cells were treated with 100 nm ghrelin (○) or untreated (●). An arbitrary value of 100 was assigned to the basal level before treatment. Statistical significance is shown by asterisks: *, p < 0.01 versus vehicle control. C, dose-dependent increase by ghrelin of tyrosine phosphorylation of IRS-1. HepG2 cells were treated with 0.1–100 nm ghrelin for 20 min. Control cells (lane 1) gave 96 ± 6 DU. Cells treated with 0.1, 1, 10, and 100 nm ghrelin gave 102 ± 7, 104 ± 8, 215 ± 20, and 324 ± 16 DU, respectively. The representative result is shown in the upper panel. In the lower panel, all values are the mean ± S.E. of tyrosine phosphorylation of IRS-1 after densitometric analysis (n = 5). Statistical significance is shown by asterisks: *, p < 0.01 versus control, **, p < 0.01versus 10 nm ghrelin-treated cells.D, HepG2 cells were treated for 20 min with vehicle alone (103 ± 10 DU), 100 nm ghrelin (178 ± 15 DU), 100 nm ghrelin + 25 μm[d-Lys-3]GHRP-6 (97 ± 10 DU), and 25 μm [d-Lys-3]GHRP-6 alone (97 ± 6 DU). The representative result is shown in the upper panel. In the lower panel, all values are the mean ± S.E. of tyrosine phosphorylation of IRS-1 after densitometric analysis (n = 5). Statistical significance is shown byasterisks: *, p < 0.01 versusvehicle alone. The amount of IRS-1, determined in the same blot by anti-IRS-1, was not changed in panels B–D.
      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 phosphotyrosine (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 ± 42versus 100 ± 5% basal level at 10 min,p < 0.01; 364 ± 47 versus 106 ± 5% basal level at 20 min, p < 0.01) (Fig.1 B). 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. 1 C. Furthermore, ghrelin-induced tyrosine phosphorylation of IRS-1 was canceled by an antagonist for GHS-R, [d-Lys-3]GHRP-6 (Fig. 1 D).
      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 ± 10 and 262 ± 15 densitometric units (DU), respectively, compared with vehicle alone (99 ± 9 DU). Combined stimulation with ghrelin and insulin resulted in an additive increase of tyrosine-phosphorylated IRS-1 levels up to 442 ± 23 DU (p < 0.01,n = 5) (Fig.2 A). The tyrosine phosphorylation of IRβ chain was not stimulated by the presence of ghrelin (Fig. 2 B). Ghrelin also increased tyrosine phosphorylation of IRS-1 in H4-II-E cells (Fig. 2 D). 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. 2 C).
      Figure thumbnail gr2
      Figure 2Ghrelin up-regulates both tyrosine phosphorylation and insulin-dependent tyrosine phosphorylation of IRS-1. A, serum-starved HepG2 cells were treated with 100 nm ghrelin for 20 min followed by 100 nm insulin for 1 min. Cell extracts were immunoprecipitated (IP) with antibodies to IRS-1 and immunoblotted with anti-Tyr(P) (Anti-pTyr). The basal amount of tyrosine phosphorylated IRS-1 was increased both by ghrelin and by insulin and was further increased by a combined treatment with both ghrelin and insulin. The representative result is shown in the upper panel. In the lower panel, all values are expressed as the mean ± S.E. of tyrosine phosphorylation of IRS-1 after densitometric analysis (n = 5). Statistical significance is shown by asterisks: *, p < 0.01 versus control; **, p < 0.01versus insulin-treated cells. The amount of IRS-1, determined in the same blot by anti-IRS-1, was not changed.B, extracts of HepG2 cells, treated as described inA, were immunoprecipitated with antibodies to human insulin receptor (IR) and immunoblotted with anti-Tyr(P) (Anti-pTyr). Control cells (lane 1) gave 22 ± 3 DU. Insulin-treated cells (lane 3) gave 402 ± 16 DU. Cells treated with ghrelin and insulin (lane 4) gave 420 ± 17 DU. Cells treated with ghrelin alone (lane 2) gave 23 ± 2 DU. The representative result is shown in theupper panel. In the lower panel, all values are expressed as the mean ± S.E. of anti-pTyr after densitometric analysis (n = 5). Statistical significance is shown byasterisks: *, p < 0.01 versuscontrol. The amount of IR, determined in the same blot by anti-IR, was not changed. C, extracts of HepG2 cells, treated as described in A, were immunoprecipitated with antibodies to IRS-2 and immunoblotted with anti-Tyr(P) (Anti-pTyr). Control cells (lane 1) gave 59 ± 8 DU. Insulin-treated cells (lane 3) gave 231 ± 36 DU. Cells treated with ghrelin and insulin (lane 4) gave 246 ± 27 DU. Cells treated with ghrelin alone (lane 2) gave 62 ± 9 DU. The representative result is shown in theupper panel. In the lower panel, all values are expressed as the mean ± S.E. of tyrosine phosphorylation of IRS-2 after densitometric analysis (n = 5). Statistical significance is shown by asterisks: *, p < 0.01 versus control. The amount of IRS-2 was determined in the same blot by anti-IRS-2. D, serum-starved H4-II-E cells were treated with 100 nm ghrelin for 20 min followed by 100 nm insulin for 1 min. Cell extracts were immunoprecipitated with antibodies to IRS-1 and immunoblotted with anti-Tyr(P). Control cells (lane 1) gave 59 ± 9 DU. Insulin-treated cells (lane 3) gave 285 ± 26 DU. Cells treated with ghrelin and insulin (lane 4) gave 405 ± 27 DU. Cells treated with ghrelin alone (lane 2) gave 210 ± 19 DU. The representative result is shown in the upper panel. In thelower panel, all values are expressed as the mean ± S.E. of tyrosine phosphorylation of IRS-1 after densitometric analysis (n = 5). Statistical significance is shown byasterisks: *, p < 0.01 versuscontrol; **, p < 0.05 versusinsulin-treated cells.
      Downstream signaling of IRS-1 is mediated by several associated proteins including GRB2 and PI3K (
      • Cheatham B.
      • Kahn C.R.
      ). We therefore tested the effect of ghrelin on the interaction of GRB2 or PI3K with IRS-1. When HepG2 cells were pretreated with 100 nm ghrelin for 20 min followed by stimulation with 100 nm insulin for 3 min, the amount of GRB2-associated IRS-1 (basal, 114 ± 10 DU) was increased either by ghrelin (256 ± 34 DU, p < 0.01) or by insulin (288 ± 33 DU, p < 0.01). Combined treatment with ghrelin and insulin resulted in an additive increase (418 ± 44 DU, p < 0.01,n = 5) (Fig.3 A). Similarly treated cells were analyzed to examine the association of PI3K with IRS-1. PI3K-associated IRS-1 in vehicle-treated cells (114 ± 10 DU) was increased either by ghrelin (220 ± 28 DU, p < 0.01) or by insulin (275 ± 32 DU, p < 0.01). However, no additive increase in PI3K associated with IRS-1 was found by a combined treatment with ghrelin and insulin (211 ± 24 DU, p < 0.01, n = 5) (Fig.3 B).
      Figure thumbnail gr3
      Figure 3Ghrelin up-regulates the association of GRB2 with IRS-1 and the association of PI3K with IRS-1. Serum-starved HepG2 cells were treated with 100 nm ghrelin for 20 min followed by 100 nm insulin for 3 min. A, cell extracts were immunoprecipitated (IP) with polyclonal anti-GRB2 and immunoblotted with anti-IRS-1. The basal amount of GRB2 associated with IRS-1 was increased both by ghrelin and by insulin and was further increased by a combined treatment with both ghrelin and insulin. The representative result is shown in the upper panel. In the lower panel, all values are the mean ± S.E. of the amount of GRB2-associated IRS-1 after densitometric analysis (n = 5). Statistical significance is shown byasterisks: *, p < 0.01 versuscontrol; **, p < 0.01 versusinsulin-treated cells. The amount of GRB2, determined in the same blot by anti-GRB2, was not changed. B, cell extracts were immunoprecipitated with polyclonal anti-PI3K and immunoblotted with anti-IRS-1. The basal amount of PI3K associated with IRS-1 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 upper panel. In the lower panel, all values are the mean ± S.E. of the amount of PI3K-associated IRS-1 after densitometric analysis (n = 5). Statistical significance is shown by asterisks: *, p < 0.01 versus control. The amount of PI3K, determined in the same blot by anti-PI3K, was not changed.
      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 ± 14 DU) was increased either by ghrelin (246 ± 24 DU, p < 0.01) or by insulin (389 ± 36 DU, p < 0.01). Combined treatment with ghrelin and insulin caused an additive increase (546 ± 58 DU, p < 0.01, n = 5) (Fig. 4 A). 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. 3 A). 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 ± 24 DU) was increased by insulin (524 ± 126 DU,p < 0.01) and, in contrast, decreased by ghrelin (29 ± 8 DU, p < 0.01) as well as by a combined treatment with ghrelin and insulin (82 ± 17 DU,p < 0.01, n = 5) (Fig.4 B). 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-1-associated PI3K activity in vehicle-treated cells (64 ± 10 DU) was increased either by ghrelin (200 ± 19 DU, p < 0.01) or by insulin (203 ± 24 DU, p < 0.01). However, no additive increase in IRS-1-associated PI3K activity was found by combined treatment with ghrelin and insulin (226 ± 7 DU, p< 0.01, n = 4) (Fig. 4 C).
      Figure thumbnail gr4
      Figure 4Ghrelin 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. A. 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 < 0.01 versus control; **,p < 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 byasterisks: *, p < 0.01 versuscontrol; **; p < 0.01 versusinsulin-treated cells. C, phosphorylation of PtdIns was carried out in the immune complexes as described under “Experimental Procedures.” The basal amount of IRS-1-associated 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 ± S.E. of the amount of IRS-1-associated PI3K activity after densitometric analysis (n = 4). Statistical significance is shown by asterisks: *,p < 0.01 versus control.
      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 (
      • Granner D.
      • Andreone T.
      • Sasaki K.
      • Beale E.
      ,
      • Xing L.
      • Quinn P.G.
      ,
      • Cohen B.
      • Novick D.
      • Rubinstein M.
      ). 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).
      Figure thumbnail gr5
      Figure 5Ghrelin up-regulates PEPCK expression in rat hepatoma H4-II-E cells. Serum-starved rat hepatoma H4-II-E cells were pretreated with 8-CPT-cAMP (0.5 mm, 3 h) followed by treatment with ghrelin and insulin as described in the legend for Fig. A. Northern blot analysis was done on cytoplasmic RNA with a DNA probe corresponding to rat PEPCK mRNA. The amount of cytoplasmic PEPCK mRNA in cells treated with insulin (100 nm, 2 h) was reduced (519 ± 68 DU,p < 0.01, lane 2) compared with control cells or ghrelin-treated cells (100 nm, 2 h) (1831 ± 88 and 2402 ± 179 DU, lanes 1 and3, respectively). Pretreatment of cells with ghrelin (100 nm) for 0.5, 1, or 2 h followed by treatment with insulin (100 nm, 2 h) partially reversed the insulin-induced down-regulation of PEPCK mRNA expression (726 ± 62 DU, p < 0.01, lane 4; 1246 ± 81 DU, p < 0.01, lane 5; 1448 ± 86 DU,p < 0.01, lane 6). The same amount of RNA was present in each lane as shown by reblotting the membrane with a probe corresponding to rat actin cDNA. Statistical significance is shown by asterisks: *, p < 0.01versus control; **, p < 0.01 versus lane 2.
      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).
      Figure thumbnail gr6
      Figure 6Ghrelin induces cell proliferation of HepG2 cells, and at 50 mm, PD98059 completely inhibits both ghrelin- and insulin-induced cell proliferation for 48 h.Ghrelin accelerated cell proliferation for 48 h by cell proliferation assay using an methanethiosulfonate solution. Control cells (lane 1) gave 0.60 ± 0.01 absorbance. Cells treated with ghrelin alone (lane 2) gave 1.15 ± 0.08 absorbance. Insulin-treated cells (lane 4) gave 1.23 ± 0.08 absorbance. Cells treated with ghrelin and PD 98059 (lane 3) gave 0.61 ± 0.01 absorbance. Cells treated with insulin and PD 98059 (lane 5) gave 0.58 ± 0.01 absorbance.

      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 IRβ 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 tyrosine-phosphorylated IRS-1 in HepG2 cells (data not shown). Therefore, the downstream molecules of GHS-R signaling can cross-talk 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-α, interleukin-4, and interleukin-9 as well (
      • Myers M.G., Jr.
      • Sun X.J.
      • Cheatham B.
      • Jachna B.R.
      • Glasheen E.M.
      • Backer J.M.
      • White M.F.
      ,
      • Yin T.
      • Tsang M.L.S.
      • Yang Y.C.
      ,
      • Yin T.
      • Keller S.R.
      • Quelle F.W.
      • Witthuhn B.A.
      • Tsang M.L.S.
      • Lieenhard G.E.
      • Ihle J.N.
      • Yang Y.C.
      ,
      • Uddin S.
      • Yenush L.
      • Sun X.J.
      • Sweet M.E.
      • White M.F.
      • Platanias L.C.
      ,
      • Permis A.
      • Witthuhn B.
      • Keegan A.D.
      • Nelms K.
      • Garfein E.
      • Ihle J.N.
      • Paul W.E.
      • Pierce J.H.
      • Rothman P.
      ). 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).
      Figure thumbnail gr7
      Figure 7Putative signal transduction molecules of IRS-1-PI3K-Akt pathway activated by insulin and ghrelin. PI3K enzymes can phosphorylates the 3′ position of PtdIns, PtdIns(4)P, and PtdIns(4,5)P2 to produce PtdIns-3P, PtdIns(3,4)P2, and PtdIns(3,4,5)P3, respectively. PTEN and SHIP dephosphorylate the 3′ and 5′ positions of the inositol ring of phosphoinositides, respectively, and PTEN reverses the reaction catalyzed by PI3K. PtdIns(3,4)P2 and PtdIns(3,4,5)P3 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 3′ position of PtdIns(3,4)P2 and PtdIns(3,4,5)P3.
      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 (
      • Sun X.J.
      • Rothenberg P.
      • Kahn CR.
      • Backer J.M.
      • Araki E.
      • Wilden P.A.
      • Cahill D.A.
      • Goldstein B.J.
      • White M.F.
      ,
      • Araki E.
      • Sun X.J.
      • Haag B.L.
      • Zhang Y.
      • Chuang L.M.
      • Zhang Y.
      • Yang-Feng T.
      • White M.F.
      • Kahn C.R.
      ). IRS-1 interacts with many SH2 proteins with diverse phosphotyrosine motif requirements including PI3K and GRB2 (
      • Cheatham B.
      • Kahn C.R.
      ,
      • Sun X.J.
      • Rothenberg P.
      • Kahn CR.
      • Backer J.M.
      • Araki E.
      • Wilden P.A.
      • Cahill D.A.
      • Goldstein B.J.
      • White M.F.
      ). 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 ghrelin-induced 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.
      The phospholipid kinase PI3K is activated by virtually all receptor tyrosine kinases. Activated PI3K phosphorylates PtdIns(4)P and PtdIns(4,5)P2 to generate the membrane-embedded second messengers PtdIns(3,4)P2 and PtdIns(3,4,5)P3. These lipids play a crucial role in the activation of Akt. PtdIns(3,4,5)P3 mediated membrane translocation of a variety of signaling proteins, including the protein-serine/threonine kinases, 3′-phosphoinositide-dependent kinase-1 (PDK-1), and Akt. Akt is phosphorylated by PDK-1 on Thr-308 in its activation loop. Phosphorylation of Thr-308 is a prerequisite for kinase activation, but phosphorylation of the C-terminal hydrophobic residue is required as well for full activation of Akt kinase. The Akt Ser-473 kinase (hypothetical PDK-2) remains to be identified (
      • Toker A.
      • Cantley L.C.
      ,
      • Toker A.
      • Newton A.C.
      ,
      • Toker A.
      • Newton A.C.
      ).
      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 ten) and SHIP (SH2-containing inositol 5′-phosphatase), two phosphoinositide-specific phosphatases that dephosphorylate the 3′ and 5′ positions of the inositol ring of phosphoinositides, respectively, leading to inhibition of cellular responses mediated by PI3K products (
      • Bolland S.
      • Pearse R.N.
      • Kurosaki T.
      • Ravetch J.V.
      ,
      • Maehama T.
      • Dixon J.E.
      ). 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 (
      • Sutherland C.
      • O'Brien R.M.
      • Granner D.K.
      ,
      • Nakajima T.
      • Fukamizu A.
      • Takahashi J.
      • Gage F.H.
      • Fisher T.
      • Blenis J.
      • Montminy M.R.
      ,
      • Gabbay R.A.
      • Sutherland C.
      • Gnudi L.
      • Kahn B.B.
      • O'Brien R.M.
      • Granner D.K.
      • Flier J.S.
      ). 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 (
      • Gabbay R.A.
      • Sutherland C.
      • Gnudi L.
      • Kahn B.B.
      • O'Brien R.M.
      • Granner D.K.
      • Flier J.S.
      ,
      • Sutherland C.
      • Waltner-Law M.
      • Gnudi L.
      • Kahn BB.
      • Granner DK.
      ,
      • Agati J.M.
      • Yeagly D.
      • Quinn P.G.
      ). 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 (
      • Clark R.G.
      • Thomas G.B.
      • Mortensen D.L.
      • Won W.B., Ma, Y.H.
      • Tomilinson E.E.
      • Fairhall K.M.
      • Robinson I.C.A.F.
      ). 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.

      Acknowledgments

      We are grateful to Chika Ogata for excellent technical assistance.

      REFERENCES

        • Muller E.E.
        • Locatelli V.
        • Cocchi D.
        Physiol. Rev. 1999; 79: 511-607
        • Momany F.A.
        • Bowers C.Y.
        • Reynolds G.A.
        • Hong A.
        • Newlander K.
        Endocrinology. 1984; 114: 1531-1536
        • Smith R.G.
        • Chen K.
        • Schoen W.R.
        • Pong S.-S.
        • Hickey G.J.
        • Jacks T.M.
        • Batler B.S.
        • Chan W.W.-S.
        • Chaung L.-Y.P.
        • Judith F.
        • Taylor A.M.
        • Wyvratt M.J., Jr.
        • Fisher M.H.
        Science. 1993; 260: 1640-1643
        • Patchett A.A.
        • Nargund R.P.
        • Tata J.R.
        • Chen M-H.
        • Barakat K.J.
        • Johnston D.B.R.
        • Chen K.
        • Chan W.W.-S.
        • Butler B.S.
        • Hickey G.J.
        • Jacks T.M.
        • Scleim K.
        • Pong S-S.
        • Chaung L.-Y.P.
        • Chen H.Y.
        • Fraizier E.
        • Leung K.H.
        • Chui S.-H.L.
        • Smith R.G.
        Proc. Natl. Acad. Sci. U. S. A. 1995; 92: 7001-7005
        • Howard A.D.
        • Feighner S.D.
        • Cully D.F.
        • Arena J.P.
        • Liberator P.A.
        • Rosenblum C.I.
        • Hamelin M.J.
        • Hreniuk D.L.
        • Palyha O.C.
        • Anderson J.
        • Paress P.S.
        • Diaz C.
        • Chou M.
        • Liu K.
        • McKee K.K.
        • Pong S.-S.
        • Chang L.-Y.
        • Elbrecht A.
        • Dashkevicz M.
        • Heavens R.
        • Rigby M.
        • Sirinathsinghji D.J.S.
        • Dean D.C.
        • Mellilo D.G.
        • Patchett A.A.
        • Nargund R.
        • Griffin P.R.
        • DeMartino J.A.
        • Gupta S.K.
        • Schaeffer J.M.
        • Smith R.G.
        • Van der Ploeg L.H.T.
        Science. 1996; 273: 974-977
        • McKee K.K.
        • Palyha O.C.
        • Feighner S.D.
        • Hreniuk D.L.
        • Tan C.P.
        • Phillips M.S.
        • Smith R.G.
        • Van der Ploeg L.H.
        • Howard A.D.
        Mol. Endocrinol. 1997; 11: 415-423
        • Kojima M.
        • Hosoda H.
        • Date Y.
        • Nakazato M.
        • Matsuo H.
        • Kangawa K.
        Nature. 1999; 402: 656-660
        • Guan X.-M., Yu, H.
        • Palyha O.C.
        • Feighner S.D.
        • Sirinathsinghji D.J.S.
        • Smith R.G.
        • Van der Pleog L.H.T.
        • Howard A.D.
        Mol. Brain Res. 1997; 48: 23-29
        • Tschàp M.
        • Smiley D.L.
        • Heiman M.L.
        Nature. 2000; 407: 908-913
        • Rajas F.
        • Croset M.
        • Zitoun C.
        • Montano S.
        • Mithieux G.
        Diabetes. 2000; 49: 1165-1168
        • Myers M.G., Jr.
        • Sun X.J.
        • Cheatham B.
        • Jachna B.R.
        • Glasheen E.M.
        • Backer J.M.
        • White M.F.
        Endocrinology. 1993; 132: 1421-1430
        • Skinner M.M.
        • Nass R.
        • Lopes B.
        • Laws E.R.
        • Thorner M.O.
        J. Clin. Endocrinol. Metab. 1998; 83: 4314-4320
        • Cheatham B.
        • Kahn C.R.
        Endocrinol. Rev. 1995; 16: 117-142
        • Granner D.
        • Andreone T.
        • Sasaki K.
        • Beale E.
        Nature. 1983; 305: 549-551
        • Xing L.
        • Quinn P.G.
        Mol. Endocrinol. 1993; 7: 1484-1494
        • Cohen B.
        • Novick D.
        • Rubinstein M.
        Science. 1996; 274: 1185-1188
        • Yin T.
        • Tsang M.L.S.
        • Yang Y.C.
        J. Biol. Chem. 1994; 269: 26614-26617
        • Yin T.
        • Keller S.R.
        • Quelle F.W.
        • Witthuhn B.A.
        • Tsang M.L.S.
        • Lieenhard G.E.
        • Ihle J.N.
        • Yang Y.C.
        J. Biol. Chem. 1995; 270: 20497-20502
        • Uddin S.
        • Yenush L.
        • Sun X.J.
        • Sweet M.E.
        • White M.F.
        • Platanias L.C.
        J. Biol. Chem. 1995; 270: 15938-15941
        • Permis A.
        • Witthuhn B.
        • Keegan A.D.
        • Nelms K.
        • Garfein E.
        • Ihle J.N.
        • Paul W.E.
        • Pierce J.H.
        • Rothman P.
        Proc. Natl. Acad. Sci. U. S. A. 1995; 92: 7971-7975
        • Sun X.J.
        • Rothenberg P.
        • Kahn CR.
        • Backer J.M.
        • Araki E.
        • Wilden P.A.
        • Cahill D.A.
        • Goldstein B.J.
        • White M.F.
        Nature. 1991; 352: 73-77
        • Araki E.
        • Sun X.J.
        • Haag B.L.
        • Zhang Y.
        • Chuang L.M.
        • Zhang Y.
        • Yang-Feng T.
        • White M.F.
        • Kahn C.R.
        Diabetes. 1993; 42: 1041-1054
        • Toker A.
        • Cantley L.C.
        Nature. 1997; 387: 673-676
        • Toker A.
        • Newton A.C.
        J. Biol. Chem. 2000; 275: 8271-8274
        • Toker A.
        • Newton A.C.
        Cell. 2000; 103: 185-188
        • Bolland S.
        • Pearse R.N.
        • Kurosaki T.
        • Ravetch J.V.
        Immunity. 1998; 8: 509-516
        • Maehama T.
        • Dixon J.E.
        J. Biol. Chem. 1998; 273: 13375-13378
        • Sutherland C.
        • O'Brien R.M.
        • Granner D.K.
        J. Biol. Chem. 1995; 270: 15501-15506
        • Nakajima T.
        • Fukamizu A.
        • Takahashi J.
        • Gage F.H.
        • Fisher T.
        • Blenis J.
        • Montminy M.R.
        Cell. 1996; 86: 465-474
        • Gabbay R.A.
        • Sutherland C.
        • Gnudi L.
        • Kahn B.B.
        • O'Brien R.M.
        • Granner D.K.
        • Flier J.S.
        J. Biol. Chem. 1996; 271: 1890-1897
        • Sutherland C.
        • Waltner-Law M.
        • Gnudi L.
        • Kahn BB.
        • Granner DK.
        J. Biol. Chem. 1998; 273: 3198-3204
        • Agati J.M.
        • Yeagly D.
        • Quinn P.G.
        J. Biol. Chem. 1998; 273: 18751-18759
        • Clark R.G.
        • Thomas G.B.
        • Mortensen D.L.
        • Won W.B., Ma, Y.H.
        • Tomilinson E.E.
        • Fairhall K.M.
        • Robinson I.C.A.F.
        Endocrinology. 1997; 138: 4316-4323