- Email: [email protected]
Ghrelin promotes pancreatic adenocarcinoma cellular proliferation and invasiveness
BBRC Biochemical and Biophysical Research Communications 309 (2003) 464–468 www.elsevier.com/locate/ybbrc
Ghrelin promotes pancreatic adenocarcinoma cellular proliferation and invasiveness Mark S. Duxbury, Talat Waseem, Hiromichi Ito, Malcolm K. Robinson, Michael J. Zinner, Stanley W. Ashley, and Edward E. Whang* Department of Surgery, Brigham and Women’s Hospital, Harvard Medical School, Boston, MA 02115, USA Received 11 August 2003
Abstract Ghrelin, a newly described potent orexigenic peptide, may have therapeutic potential in patients with cachexia. We assessed whether pancreatic adenocarcinoma, commonly associated with marked cachexia, is a ghrelin-responsive malignancy. Pancreatic adenocarcinoma cells were exposed to ghrelin (0–100 nM). Proliferation was determined by MTT assay. Ghrelin, ghrelin 1a and 1b receptor expression and Akt phosphorylation were assessed. The effects of ghrelin ( its antagonist D -Lys-GHRP6, or the PI3-K inhibitor Wortmannin) on cellular motility and invasiveness were quantified by Matrigel Boyden chamber assay. All cell lines expressed ghrelin 1a and 1b receptor transcript and protein, but only PANC1 weakly expressed ghrelin transcript. Ten nanomolar ghrelin increased proliferation, motility, invasiveness, and Akt phosphorylation in all cell lines. Proliferation was affected dosedependently, being suppressed at higher ghrelin concentrations. D -Lys-GHRP6 suppressed ghrelin-induced proliferation, invasion, and Akt phosphorylation. Wortmannin abolished the effects of ghrelin on motility and invasiveness. Pancreatic adenocarcinoma is a ghrelin-responsive malignancy. Ó 2003 Elsevier Inc. All rights reserved. Keywords: Ghrelin; Cachexia; Akt; Pancreatic adenocarcinoma; Cancer; GHS-R; PANC1; MIAPaCa2; BxPc3; Capan2; Invasion
Pancreatic cancer is an extremely aggressive malignancy characterized by extensive invasion, early metastasis, and marked cachexia [1,2]. Cachexia is a strong independent predictor of mortality, poor therapeutic response, diminished functional capacity, and reduced quality of life [3–5]. Cancer patients commonly have derangements in basal metabolic rate as well as impaired appetite and food intake [5,6], but cancer cachexia remains difficult to manage. Even moderate weight loss is associated with psychological stress and lower quality of life [5]. The processes mediating pancreatic cancer cachexia are poorly understood, although numerous cytokines including interferon c, interleukin 6, leukemia inhibitory factor, protein and lipid mobilizing factors, tumor necrosis factor alpha, and proteolysis inducing factor have been implicated as mediators of this catabolic process [7–9]. * Corresponding author. Fax: 1-617-739-1728. E-mail address: [email protected] (E.E. Whang).
0006-291X/$ - see front matter Ó 2003 Elsevier Inc. All rights reserved. doi:10.1016/j.bbrc.2003.08.024
Ghrelin is a recently discovered 28 amino acid brain– gut peptide hormone initially isolated from rat and human stomach [10]. In its octonylated form, ghrelin activates the 366 amino acid G protein-coupled type 1a growth hormone secretagogue receptor (GHS-R1a). The 289 amino acid type 1b receptor is a splice variant for which a function has not yet been ascribed. GHS-R mRNA is widely expressed in tissues including the pancreas [11]. Ghrelin is among the most potent stimulants of food intake and weight gain in humans [12,13]. Levels are increased by fasting and rapidly decrease following meals [14]. Ghrelin is therefore of interest as a potential treatment for cachexia. Administration of ghrelin is reported to increase food intake and promote weight gain in normal and melanoma-bearing nude mice [15] but, in order for ghrelin to be a rational approach for the treatment of cancer cachexia, it must not adversely affect disease progression. We hypothesized that pancreatic adenocarcinoma may be a ghrelin-responsive malignancy. We characterized
M.S. Duxbury et al. / Biochemical and Biophysical Research Communications 309 (2003) 464–468
cellular ghrelin, GHS-R1a and 1b expression, and investigated the response of pancreatic adenocarcinoma to ghrelin treatment in vitro. Our results indicate that caution will be required in the evaluation of ghrelin as a treatment strategy for pancreatic cancer cachexia.
Materials and methods Cell lines. Poorly differentiated human pancreatic cancer cell lines PANC1 and MIAPaCa2 and well-differentiated lines BxPC3 and Capan2 were obtained from ATCC (Rockville, MD). PANC1, MIAPaCa2, and BxPC3 were maintained in Dulbecco’s modified Eagle’s medium (DMEM) containing 10% fetal bovine serum (FBS) and Capan2 was maintained in McCoy’s 5A medium containing 10% FBS in a humidified (37 °C, 5% CO2 ) incubator. All cell lines were sub-cultured before confluence and seeded at a density of 1 104 cells/cm2 . Reverse transcriptase PCR. At 70% confluence, total RNA was extracted from cultured cells using Trizol (Life Technologies) according to the manufacturer’s instructions. RNA integrity was confirmed by denaturing formaldehyde gel electrophoresis. RT-PCR was preformed using the One-Step RT-PCR kit (Stratagene, Cedar Creek, TX) according to the manufacturer’s instructions. The primer sequences for ghrelin were: forward 50 -TGAGCCCTGAACACCAGAGAG-30 and reverse 50 -AAAGCCAGATGAGCGCTTCTA-30 (expected product size 327 bp). The same forward primer was used for GHS-R1a and 1b: 50 -TCGTGGGTGCCTCGCT-30 . Reverse primers for GHS-R1a and 1b were 50 -CACCACTACAGCCAGCATTTTC-30 (expected product size 65 bp) and 50 -GCTGAGACCCACCCAGCA-30 (expected product size 66 bp), respectively [16]. b-Actin primers were forward 50 -GT GGGGCGCCCCAGGCACCA-30 and reverse 50 -TTGGCCTTGGG GTTCAGGGG-30 . Thermal cycling (Hybaid, Ashford, UK) consisted of a 30 min 50 °C RT stage, followed by 5 min at 95 °C then 35 cycles of 1 min at 94 °C, 1 min at 60 °C, and 1 min at 72 °C. The sequence terminated following 10 min at 72 °C. Western blot analysis. Cells (2 106 ) were harvested and rinsed twice with phosphate buffered saline (PBS), pH 7.4. Cell extracts were prepared with lysis buffer (20 mM Tris, pH 7.5, 0.1% Triton X, 0.5% deoxycholate, 1 mM phenylmethylsulfonyl fluoride, 10 lg/ml aprotinin, and 10 lg/ml leupeptin) and cleared by centrifugation at 12,000g, 4 °C. Total protein concentration was measured using the BCA assay kit (Sigma, St. Louis, MO) with bovine serum albumin as a standard, in accordance with the manufacturer’s instructions. Cell extracts containing 30 lg total protein were subjected to 10% SDS/PAGE and the resolved proteins transferred electrophoretically to PVDF membranes (Invitrogen, Carlsbad, CA). After blocking with PBS containing 0.2% casein for 1 h at room temperature, membranes were incubated with 5 lg/ml affinity purified goat anti-ghrelin (Santa Cruz Biotechnology, Santa Cruz, CA), rabbit anti-GHS-R1a or rabbit anti-GHS-R1b antibody (Phoenix Pharmaceuticals, Belmont, CA) in PBS containing 0.1% Tween 20 overnight at 4 °C. Following secondary antibody incubation, the Vecstain ABC kit and DAB liquid substrate were used for chromogenic detection according to the manufacturer’s instructions. Equal protein loading was confirmed by probing with 3 lg/ml anti-actin monoclonal antibody (Lab Vision, Freemont, CA). Phosphorylation of Akt was assessed using anti-Akt and anti-phospho-(Thr-308)-Akt (pAKT) antibodies (Santa Cruz, Santa Cruz, CA), and normalized to total Akt. Proliferation assays. Cell proliferation was quantified by MTT assay and confirmed by cell counting. Cells were seeded into 96-well plates at 104 cells per well. Cells were cultured for 3 days in appropriate medium containing 10% FBS and human n-octonylated ghrelin (Phoenix Pharmaceuticals) at concentrations from 1 to 100 nM ghrelin (dissolved in PBS). Negative controls received the same medium con-
465
taining vehicle only (PBS). Medium was replaced every 24 h. In order to determine the mechanism by which ghrelin influenced proliferation, cells were pre-treated for 1 h with the GHS-R-specific peptide antagonist D -Lys-GHRP6 (0.1 mM). After 72 h, the cells were quantified by 3-(4,5-dimethylthiazolyl-2yl)-2,5-diphenyltetrazolium (MTT) assay (Trevigen, Gaithersburg, MD) according to the manufacturer’s instructions. The plates were read using a Vmax microplate spectrophotometer (Molecular Devices, Sunnyvale, CA) at a wavelength of 570 nm, referenced to 650 nm and normalized to controls. Each independent experiment was performed three times. At the same time point, cell counting was performed. Cells were trypsinized to form a single cell suspension. Intact cells, determined by Trypan blue exclusion, were counted using a Neubauer hemocytometer (Hausser scientific, Horsham, PA), the number of cells per milliliter calculated and compared to the control group for each cell line. Invasion and migration assay. The BD BioCoat Matrigel invasion chamber (BD Bioscience, Bedford, MA) was used according to the manufacturer’s instructions. Pancreatic cancer cells (2.5 104 ) in serum-free media, containing 10 nM ghrelin or PBS, were seeded onto Matrigel-coated filters. In the lower chambers, 5% FBS was added as a chemoattractant. After 24 h incubation, the filters were stained with Diff-Quik kit (BD Bioscience) and the number of cells that had invaded through the filter was counted under magnification (randomly selected high-power fields). The counting was performed for three fields in each sample and values from three independent experiments were used. Invasiveness was normalized to cell proliferation under the same conditions to control for the effect of increases in cell number. The effect of D -Lys-GHRP6 was determined by pre-treatment of the cells for 1 h prior to ghrelin treatment. To assess cellular migratory potential the protocol was identical to that of the invasion assay, with the exception that no Matrigel was used. In order to confirm the functional importance of the phosphoinositol 3-kinase (PI3-K)/Akt pathway in ghrelin’s effect on migratory potential and invasiveness, the corresponding assays were performed, in the presence or absence of 10 nM ghrelin, following 1 h pre-incubation of the cells in medium containing 50 nM Wortmannin (Calbiochem, EMD Biosciences, San Diego, CA), an irreversible PI3-K inhibitor. Statistical analysis. Differences between groups were analyzed using Student’s t test, ANOVA, and Mann Whitney U test for non-parametric data, as appropriate, using Statistica 5.5 software (StatSoft, Tulsa, OK). P < 0:05 was considered statistically significant.
Results Ghrelin transcript was detected at very low levels in only PANC1 (Fig. 1). Ghrelin peptide was not detected in any cell line examined (Fig. 2). GHS-R1a and b transcripts were present in all cell lines tested. Single RT-PCR products were detected for the ghrelin 1a (65 bp) and 1b (66 bp) receptor. Expression of both the functional GHS-R1a isoform and GHS-R1b was confirmed at the protein level by Western analysis. Ghrelin treatment for 72 h increased cellular proliferation relative to control cells in a dose-dependent fashion, peaking at 10 nM, in all four cell lines (Fig. 3A). Above this concentration, the proliferative effect decreased. GHS-R antagonism by pre-treatment with 100 lm D -LysGHRP6 for 1 h abrogated this proliferative response. In cells pre-treated with D -Lys-GHRP6, exposure to ghrelin at concentrations greater than 10 nM inhibited proliferation (Fig. 3B).
466
M.S. Duxbury et al. / Biochemical and Biophysical Research Communications 309 (2003) 464–468
Fig. 1. Single RT-PCR products were amplified from PANC1 (P), MIAPaCa2 (M), BxPC3 (B), and Capan2 (C) pancreatic adenocarcinoma cell lines using primers for GHS-R1a and GHS-R1b. Ghrelin transcript was detected weakly in PANC1 and was absent from the other pancreatic adenocarcinoma cell lines tested. ()) no template negative control and (+) positive control HT29 colonic adenocarcinoma cells.
Fig. 2. Western immunoblot of whole cell lysates from PANC1 (P), MIAPaCa2 (M), BxPC3 (B), and Capan2 (C) using (a) anti-GHS-R1a, (b) anti-GHS-R1b, and (c) anti-ghrelin antibodies. Human ghrelin was used as a positive control.
Cellular invasiveness was significantly increased by exposure to 10 nM ghrelin for 24 h (Fig. 4B). The mean increase, expressed relative to PBS-treated cells (SD), was 60 5% in the case of PANC1, 35 3% for MIAPaCa2, 30 3% for BxPC3, and 15 3% for Capan2 (P < 0:05 in each case). Cellular migration was similarly increased by ghrelin treatment (Fig. 4A). Akt activation, for which phosphorylation at threonine 308 is a requirement, is reported to be associated with increased pancreatic cancer invasiveness [17–20]. We therefore examined the effect of ghrelin on Akt phosphorylation by phospho-Akt-specific immunoblot. Akt phosphorylation was increased by 10 nM ghrelin in all cell lines (Fig. 5). The increase in Akt phosphorylation induced by ghrelin was abolished by D -LysGHRP6. The role of increased Akt activity in the stimulatory effects of ghrelin on migratory potential and invasiveness was confirmed using the PI3-K inhibitor Wortmannin. Pre-incubation of cells in medium containing 50 nM Wortmannin for 1 h prior to performing the migration and invasion assays abolished the stimu-
Fig. 3. The effect of ghrelin on cellular proliferation was determined by MTT assay in the absence (A), or presence (B), of the GHS-R antagonist D -Lys-GHRP6. Absorbance at 560 nm has been converted to percentages above control (SD). Data represent one of three identical experiments, n ¼ 16 in each case. (A) Ghrelin promoted cellular proliferation in a dose-dependent manner, peaking at 10 nM (*P < 0:01 versus control). Above this concentration, ghrelin had a negative effect upon proliferation. (B) The pro-proliferative effect of ghrelin was abolished by pre-treatment with D -Lys-GHRP6. *P < 0:05 versus 10 nM ghrelin.
Fig. 4. Cells were treated with 10 nM ghrelin or PBS vehicle. Invasion was quantified after 24 h by Matrigel Boyden chamber assay. Motility was quantified in the same way with the exception that Matrigel was omitted. Ten nanomolar ghrelin increased invasiveness and motility in all cell lines tested. *P < 0:05.
M.S. Duxbury et al. / Biochemical and Biophysical Research Communications 309 (2003) 464–468
Fig. 5. Western immunoblot using anti-phospho-Akt (threonine 308) antibody. Total Akt was probed using an anti-Akt antibody. Treatment with 10 nM ghrelin increased Akt phosphorylation at threonine 308 in all cell lines.
Fig. 6. Pre-treatment for 1 h with the PIK-3 inhibitor Wortmannin abolished the stimulatory effects of 10 nM ghrelin on cellular motility and invasiveness.
latory effect of 10 nM ghrelin on cellular migration and invasiveness (Fig. 6). Discussion In this study, we have demonstrated expression of GHS-R1a and 1b isoforms in pancreatic adenocarci-
467
noma cells. Although ghrelin transcript was only weakly present in PANC1 and ghrelin peptide was not detected, cellular proliferation, invasiveness, and motility were stimulated by exposure to 10 nM ghrelin, indicating that pancreatic adenocarcinoma is ghrelin-responsive. In addition, we have demonstrated that the PI3-K/Akt pathway is an important mediator of ghrelin’s stimulatory effects on migration and invasiveness in pancreatic adenocarcinoma cells. To our knowledge, this is the first characterization of ghrelin receptor expression and response to ghrelin exposure in human pancreatic adenocarcinoma cells. Ghrelin is secreted principally by the stomach, and is present in only low levels in normal pancreatic tissue [11], where it is expressed principally by beta cells [21]. The absence of expression of ghrelin peptide in the adenocarcinoma cell lines examined is consistent with a reported series of micro-dissected pancreatic adenocarcinomas in which ghrelin immunoreactivity was not detected [21]. In the case of PANC1, post-transcriptional modulation of ghrelin expression may occur. It is also conceivable that ghrelin secreted by normal islets could act in a paracrine fashion to stimulate pancreatic adenocarcinoma proliferation in vivo. GHS-R1a and 1b expression has been confirmed in a variety of neoplasms including prostate adenocarcinoma [22], pancreatic endocrine tumors [21], somatotroph and other central nervous system tumors [23]. While GHSR1a is a functional G protein-coupled receptor, the role of GHS-R1b is not currently known. Stimulation of prostate adenocarcinoma cellular proliferation is reported following exposure to ghrelin, with a maximal proliferative effect at 5 nM that diminished above this concentration [22]. This is comparable to the maximal stimulation of pancreatic adenocarcinoma growth we observed with 10 nM ghrelin. This growth-promoting effect also declines at higher ghrelin concentrations. The growth-promoting effects of ghrelin on malignant cells are not universal. Suppression of breast, lung, and thyroid adenocarcinoma proliferation has been observed in response to ghrelin treatment [16,24,25]. These antiproliferative effects were generally seen at ghrelin concentrations greater than 10 nM: the level above which the growth-promoting effects of ghrelin subside in pancreatic and prostatic adenocarcinoma cells [22]. The GHS-R antagonist D -Lys-GHRP6 suppresses proliferation in response to 10 nM ghrelin and, at higher ghrelin concentrations, this decline in proliferation becomes more exaggerated in D -Lys-GHRP6-treated cells. Together, these findings suggest first that the growth-promoting effect of ghrelin is mediated via the GHS-R. Second, it raises the possibility that another, lower affinity nonGHS-R receptor may be responsible for mediating the anti-proliferative effects of ghrelin observed at higher concentrations. Evidence is mounting for the existence of a non-GHS-R ghrelin receptor. This hypothesis is
468
M.S. Duxbury et al. / Biochemical and Biophysical Research Communications 309 (2003) 464–468
supported by experiments in which administration of synthetic peptidyl ghrelin analogues which lack the in vivo action of ghrelin, but not administration of ghrelin itself, inhibits DNA synthesis and proliferation in human lung carcinoma cells [25]. Further credence is given to the existence of a non-GHS-R ghrelin receptor by the demonstration that ghrelin binds to normal and neoplastic thyroid tissue, inhibiting proliferation, despite the absence of GHS-R mRNA expression [16]. Cellular invasiveness and motility were promoted by ghrelin. These responses were associated with increased phosphorylation of Akt at threonine 308, which was used as a marker of Akt activation [26]. Akt activation has been associated with increased malignant cellular invasiveness, both in vitro and in vivo [17–20]. Invasiveness, motility, and threonine phosphorylation of Akt are all suppressed by D -Lys-GHRP6, indicating that the stimulatory effects of ghrelin on pancreatic adenocarcinoma cells are mediated via the GHS-R. While these in vitro observations may not directly replicate the response of pancreatic adenocarcinoma to ghrelin in vivo, these data provide clear evidence that pancreatic adenocarcinoma cells are ghrelin-responsive, which may have implications for the use of ghrelin in the setting of cancer cachexia. In conclusion, we have shown that ghrelin increases proliferation, motility, and invasiveness of pancreatic adenocarcinoma cells through an Akt-dependent mechanism, indicating that pancreatic adenocarcinoma is a ghrelin-responsive malignancy. There remains a need for selective non-toxic treatments to attenuate cachexia. However, the use of ghrelin as a treatment for human pancreatic cancer cachexia will require cautious evaluation.
Acknowledgments This study was supported by a grant from the National Pancreas Foundation and by National Institute of Health Grants DK02786 (E.E.W.) and DK47326 (S.W.A.). The authors gratefully acknowledge the technical support of Jan Rounds.
References [1] E.P. DiMagno, H.A. Reber, M.A. Tempero, Gastroenterology 117 (1999) 1464–1484.
[2] S. Keleg, P. Buchler, R. Ludwig, M.W. Buchler, H. Friess, Mol. Cancer 2 (2003) 14. [3] M.D. Barber, J.A. Ross, K.C. Fearon, Nutr. Cancer 35 (1999) 106–110. [4] W.D. DeWys, C. Begg, P.T. Lavin, P.R. Band, J.M. Bennett, J.R. Bertino, M.H. Cohen, H.O. Douglass Jr., P.F. Engstrom, E.Z. Ezdinli, J. Horton, G.J. Johnson, C.G. Moertel, M.M. Oken, C. Perlia, C. Rosenbaum, M.N. Silverstein, R.T. Skeel, R.W. Sponzo, D.C. Tormey, Am. J. Med. 69 (1980) 491–497. [5] L. Ovesen, J. Hannibal, E.L. Mortensen, Nutr. Cancer 19 (1993) 159–167. [6] J.S. Falconer, K.C. Fearon, C.E. Plester, J.A. Ross, D.C. Carter, Ann. Surg. 219 (1994) 325–331. [7] M.J. Tisdale, J. Natl. Cancer Inst. 89 (1997) 1763–1773. [8] M.J. Tisdale, J. Nutr. 129 (1999) 243S–246S. [9] M.J. Tisdale, Support Care Cancer 11 (2003) 73–78. [10] M. Kojima, H. Hosoda, K. Kangawa, Horm. Res. 56 (Suppl. 1) (2001) 93–97. [11] S. Gnanapavan, B. Kola, S.A. Bustin, D.G. Morris, P. McGee, P. Fairclough, S. Bhattacharya, R. Carpenter, A.B. Grossman, M. Korbonits, J. Clin. Endocrinol. Metab. 87 (2002) 2988. [12] J. Pinkney, G. Williams, Lancet 359 (2002) 1360–1361. [13] A.M. Wren, L.J. Seal, M.A. Cohen, A.E. Brynes, G.S. Frost, K.G. Murphy, W.S. Dhillo, M.A. Ghatei, S.R. Bloom, J. Clin. Endocrinol. Metab. 86 (2001) 5992. [14] J. Eisenstein, A. Greenberg, Nutr. Rev. 61 (2003) 101–104. [15] T. Hanada, K. Toshinai, N. Kajimura, N. Nara-Ashizawa, T. Tsukada, Y. Hayashi, K. Osuye, K. Kangawa, S. Matsukura, M. Nakazato, Biochem. Biophys. Res. Commun. 301 (2003) 275–279. [16] M. Volante, E. Allia, E. Fulcheri, P. Cassoni, E. Ghigo, G. Muccioli, M. Papotti, Am. J. Pathol. 162 (2003) 645–654. [17] J.Q. Cheng, B. Ruggeri, W.M. Klein, G. Sonoda, D.A. Altomare, D.K. Watson, J.R. Testa, Proc. Natl. Acad. Sci. USA 93 (1996) 3636–3641. [18] S.J. Grille, A. Bellacosa, J. Upson, A.J. Klein-Szanto, F. van Roy, W. Lee-Kwon, M. Donowitz, P.N. Tsichlis, L. Larue, Cancer Res. 63 (2003) 2172–2178. [19] D. Kim, S. Kim, H. Koh, S.O. Yoon, A.S. Chung, K.S. Cho, J. Chung, FASEB J. 15 (2001) 1953–1962. [20] S. Tanno, S. Tanno, Y. Mitsuuchi, D.A. Altomare, G.H. Xiao, J.R. Testa, Cancer Res. 61 (2001) 589–593. [21] M. Volante, E. Allia, P. Gugliotta, A. Funaro, F. Broglio, R. Deghenghi, G. Muccioli, E. Ghigo, M. Papotti, J. Clin. Endocrinol. Metab. 87 (2002) 1300–1308. [22] P.L. Jeffery, A.C. Herington, L.K. Chopin, J. Endocrinol. 172 (2002) R7–R11. [23] M. Korbonits, M. Kojima, K. Kangawa, A.B. Grossman, Endocrine 14 (2001) 101–104. [24] P. Cassoni, M. Papotti, C. Ghe, F. Catapano, A. Sapino, A. Graziani, R. Deghenghi, T. Reissmann, E. Ghigo, G. Muccioli, J. Clin. Endocrinol. Metab. 86 (2001) 1738–1745. [25] C. Ghe, P. Cassoni, F. Catapano, T. Marrocco, R. Deghenghi, E. Ghigo, G. Muccioli, M. Papotti, Endocrinology 143 (2002) 484–491. [26] J. Downward, Curr. Opin. Cell Biol. 10 (1998) 262–267.
Ghrelin promotes pancreatic adenocarcinoma cellular proliferation and invasiveness Mark S. Duxbury, Talat Waseem, Hiromichi Ito, Malcolm K. Robinson, Michael J. Zinner, Stanley W. Ashley, and Edward E. Whang* Department of Surgery, Brigham and Women’s Hospital, Harvard Medical School, Boston, MA 02115, USA Received 11 August 2003
Abstract Ghrelin, a newly described potent orexigenic peptide, may have therapeutic potential in patients with cachexia. We assessed whether pancreatic adenocarcinoma, commonly associated with marked cachexia, is a ghrelin-responsive malignancy. Pancreatic adenocarcinoma cells were exposed to ghrelin (0–100 nM). Proliferation was determined by MTT assay. Ghrelin, ghrelin 1a and 1b receptor expression and Akt phosphorylation were assessed. The effects of ghrelin ( its antagonist D -Lys-GHRP6, or the PI3-K inhibitor Wortmannin) on cellular motility and invasiveness were quantified by Matrigel Boyden chamber assay. All cell lines expressed ghrelin 1a and 1b receptor transcript and protein, but only PANC1 weakly expressed ghrelin transcript. Ten nanomolar ghrelin increased proliferation, motility, invasiveness, and Akt phosphorylation in all cell lines. Proliferation was affected dosedependently, being suppressed at higher ghrelin concentrations. D -Lys-GHRP6 suppressed ghrelin-induced proliferation, invasion, and Akt phosphorylation. Wortmannin abolished the effects of ghrelin on motility and invasiveness. Pancreatic adenocarcinoma is a ghrelin-responsive malignancy. Ó 2003 Elsevier Inc. All rights reserved. Keywords: Ghrelin; Cachexia; Akt; Pancreatic adenocarcinoma; Cancer; GHS-R; PANC1; MIAPaCa2; BxPc3; Capan2; Invasion
Pancreatic cancer is an extremely aggressive malignancy characterized by extensive invasion, early metastasis, and marked cachexia [1,2]. Cachexia is a strong independent predictor of mortality, poor therapeutic response, diminished functional capacity, and reduced quality of life [3–5]. Cancer patients commonly have derangements in basal metabolic rate as well as impaired appetite and food intake [5,6], but cancer cachexia remains difficult to manage. Even moderate weight loss is associated with psychological stress and lower quality of life [5]. The processes mediating pancreatic cancer cachexia are poorly understood, although numerous cytokines including interferon c, interleukin 6, leukemia inhibitory factor, protein and lipid mobilizing factors, tumor necrosis factor alpha, and proteolysis inducing factor have been implicated as mediators of this catabolic process [7–9]. * Corresponding author. Fax: 1-617-739-1728. E-mail address: [email protected] (E.E. Whang).
0006-291X/$ - see front matter Ó 2003 Elsevier Inc. All rights reserved. doi:10.1016/j.bbrc.2003.08.024
Ghrelin is a recently discovered 28 amino acid brain– gut peptide hormone initially isolated from rat and human stomach [10]. In its octonylated form, ghrelin activates the 366 amino acid G protein-coupled type 1a growth hormone secretagogue receptor (GHS-R1a). The 289 amino acid type 1b receptor is a splice variant for which a function has not yet been ascribed. GHS-R mRNA is widely expressed in tissues including the pancreas [11]. Ghrelin is among the most potent stimulants of food intake and weight gain in humans [12,13]. Levels are increased by fasting and rapidly decrease following meals [14]. Ghrelin is therefore of interest as a potential treatment for cachexia. Administration of ghrelin is reported to increase food intake and promote weight gain in normal and melanoma-bearing nude mice [15] but, in order for ghrelin to be a rational approach for the treatment of cancer cachexia, it must not adversely affect disease progression. We hypothesized that pancreatic adenocarcinoma may be a ghrelin-responsive malignancy. We characterized
M.S. Duxbury et al. / Biochemical and Biophysical Research Communications 309 (2003) 464–468
cellular ghrelin, GHS-R1a and 1b expression, and investigated the response of pancreatic adenocarcinoma to ghrelin treatment in vitro. Our results indicate that caution will be required in the evaluation of ghrelin as a treatment strategy for pancreatic cancer cachexia.
Materials and methods Cell lines. Poorly differentiated human pancreatic cancer cell lines PANC1 and MIAPaCa2 and well-differentiated lines BxPC3 and Capan2 were obtained from ATCC (Rockville, MD). PANC1, MIAPaCa2, and BxPC3 were maintained in Dulbecco’s modified Eagle’s medium (DMEM) containing 10% fetal bovine serum (FBS) and Capan2 was maintained in McCoy’s 5A medium containing 10% FBS in a humidified (37 °C, 5% CO2 ) incubator. All cell lines were sub-cultured before confluence and seeded at a density of 1 104 cells/cm2 . Reverse transcriptase PCR. At 70% confluence, total RNA was extracted from cultured cells using Trizol (Life Technologies) according to the manufacturer’s instructions. RNA integrity was confirmed by denaturing formaldehyde gel electrophoresis. RT-PCR was preformed using the One-Step RT-PCR kit (Stratagene, Cedar Creek, TX) according to the manufacturer’s instructions. The primer sequences for ghrelin were: forward 50 -TGAGCCCTGAACACCAGAGAG-30 and reverse 50 -AAAGCCAGATGAGCGCTTCTA-30 (expected product size 327 bp). The same forward primer was used for GHS-R1a and 1b: 50 -TCGTGGGTGCCTCGCT-30 . Reverse primers for GHS-R1a and 1b were 50 -CACCACTACAGCCAGCATTTTC-30 (expected product size 65 bp) and 50 -GCTGAGACCCACCCAGCA-30 (expected product size 66 bp), respectively [16]. b-Actin primers were forward 50 -GT GGGGCGCCCCAGGCACCA-30 and reverse 50 -TTGGCCTTGGG GTTCAGGGG-30 . Thermal cycling (Hybaid, Ashford, UK) consisted of a 30 min 50 °C RT stage, followed by 5 min at 95 °C then 35 cycles of 1 min at 94 °C, 1 min at 60 °C, and 1 min at 72 °C. The sequence terminated following 10 min at 72 °C. Western blot analysis. Cells (2 106 ) were harvested and rinsed twice with phosphate buffered saline (PBS), pH 7.4. Cell extracts were prepared with lysis buffer (20 mM Tris, pH 7.5, 0.1% Triton X, 0.5% deoxycholate, 1 mM phenylmethylsulfonyl fluoride, 10 lg/ml aprotinin, and 10 lg/ml leupeptin) and cleared by centrifugation at 12,000g, 4 °C. Total protein concentration was measured using the BCA assay kit (Sigma, St. Louis, MO) with bovine serum albumin as a standard, in accordance with the manufacturer’s instructions. Cell extracts containing 30 lg total protein were subjected to 10% SDS/PAGE and the resolved proteins transferred electrophoretically to PVDF membranes (Invitrogen, Carlsbad, CA). After blocking with PBS containing 0.2% casein for 1 h at room temperature, membranes were incubated with 5 lg/ml affinity purified goat anti-ghrelin (Santa Cruz Biotechnology, Santa Cruz, CA), rabbit anti-GHS-R1a or rabbit anti-GHS-R1b antibody (Phoenix Pharmaceuticals, Belmont, CA) in PBS containing 0.1% Tween 20 overnight at 4 °C. Following secondary antibody incubation, the Vecstain ABC kit and DAB liquid substrate were used for chromogenic detection according to the manufacturer’s instructions. Equal protein loading was confirmed by probing with 3 lg/ml anti-actin monoclonal antibody (Lab Vision, Freemont, CA). Phosphorylation of Akt was assessed using anti-Akt and anti-phospho-(Thr-308)-Akt (pAKT) antibodies (Santa Cruz, Santa Cruz, CA), and normalized to total Akt. Proliferation assays. Cell proliferation was quantified by MTT assay and confirmed by cell counting. Cells were seeded into 96-well plates at 104 cells per well. Cells were cultured for 3 days in appropriate medium containing 10% FBS and human n-octonylated ghrelin (Phoenix Pharmaceuticals) at concentrations from 1 to 100 nM ghrelin (dissolved in PBS). Negative controls received the same medium con-
465
taining vehicle only (PBS). Medium was replaced every 24 h. In order to determine the mechanism by which ghrelin influenced proliferation, cells were pre-treated for 1 h with the GHS-R-specific peptide antagonist D -Lys-GHRP6 (0.1 mM). After 72 h, the cells were quantified by 3-(4,5-dimethylthiazolyl-2yl)-2,5-diphenyltetrazolium (MTT) assay (Trevigen, Gaithersburg, MD) according to the manufacturer’s instructions. The plates were read using a Vmax microplate spectrophotometer (Molecular Devices, Sunnyvale, CA) at a wavelength of 570 nm, referenced to 650 nm and normalized to controls. Each independent experiment was performed three times. At the same time point, cell counting was performed. Cells were trypsinized to form a single cell suspension. Intact cells, determined by Trypan blue exclusion, were counted using a Neubauer hemocytometer (Hausser scientific, Horsham, PA), the number of cells per milliliter calculated and compared to the control group for each cell line. Invasion and migration assay. The BD BioCoat Matrigel invasion chamber (BD Bioscience, Bedford, MA) was used according to the manufacturer’s instructions. Pancreatic cancer cells (2.5 104 ) in serum-free media, containing 10 nM ghrelin or PBS, were seeded onto Matrigel-coated filters. In the lower chambers, 5% FBS was added as a chemoattractant. After 24 h incubation, the filters were stained with Diff-Quik kit (BD Bioscience) and the number of cells that had invaded through the filter was counted under magnification (randomly selected high-power fields). The counting was performed for three fields in each sample and values from three independent experiments were used. Invasiveness was normalized to cell proliferation under the same conditions to control for the effect of increases in cell number. The effect of D -Lys-GHRP6 was determined by pre-treatment of the cells for 1 h prior to ghrelin treatment. To assess cellular migratory potential the protocol was identical to that of the invasion assay, with the exception that no Matrigel was used. In order to confirm the functional importance of the phosphoinositol 3-kinase (PI3-K)/Akt pathway in ghrelin’s effect on migratory potential and invasiveness, the corresponding assays were performed, in the presence or absence of 10 nM ghrelin, following 1 h pre-incubation of the cells in medium containing 50 nM Wortmannin (Calbiochem, EMD Biosciences, San Diego, CA), an irreversible PI3-K inhibitor. Statistical analysis. Differences between groups were analyzed using Student’s t test, ANOVA, and Mann Whitney U test for non-parametric data, as appropriate, using Statistica 5.5 software (StatSoft, Tulsa, OK). P < 0:05 was considered statistically significant.
Results Ghrelin transcript was detected at very low levels in only PANC1 (Fig. 1). Ghrelin peptide was not detected in any cell line examined (Fig. 2). GHS-R1a and b transcripts were present in all cell lines tested. Single RT-PCR products were detected for the ghrelin 1a (65 bp) and 1b (66 bp) receptor. Expression of both the functional GHS-R1a isoform and GHS-R1b was confirmed at the protein level by Western analysis. Ghrelin treatment for 72 h increased cellular proliferation relative to control cells in a dose-dependent fashion, peaking at 10 nM, in all four cell lines (Fig. 3A). Above this concentration, the proliferative effect decreased. GHS-R antagonism by pre-treatment with 100 lm D -LysGHRP6 for 1 h abrogated this proliferative response. In cells pre-treated with D -Lys-GHRP6, exposure to ghrelin at concentrations greater than 10 nM inhibited proliferation (Fig. 3B).
466
M.S. Duxbury et al. / Biochemical and Biophysical Research Communications 309 (2003) 464–468
Fig. 1. Single RT-PCR products were amplified from PANC1 (P), MIAPaCa2 (M), BxPC3 (B), and Capan2 (C) pancreatic adenocarcinoma cell lines using primers for GHS-R1a and GHS-R1b. Ghrelin transcript was detected weakly in PANC1 and was absent from the other pancreatic adenocarcinoma cell lines tested. ()) no template negative control and (+) positive control HT29 colonic adenocarcinoma cells.
Fig. 2. Western immunoblot of whole cell lysates from PANC1 (P), MIAPaCa2 (M), BxPC3 (B), and Capan2 (C) using (a) anti-GHS-R1a, (b) anti-GHS-R1b, and (c) anti-ghrelin antibodies. Human ghrelin was used as a positive control.
Cellular invasiveness was significantly increased by exposure to 10 nM ghrelin for 24 h (Fig. 4B). The mean increase, expressed relative to PBS-treated cells (SD), was 60 5% in the case of PANC1, 35 3% for MIAPaCa2, 30 3% for BxPC3, and 15 3% for Capan2 (P < 0:05 in each case). Cellular migration was similarly increased by ghrelin treatment (Fig. 4A). Akt activation, for which phosphorylation at threonine 308 is a requirement, is reported to be associated with increased pancreatic cancer invasiveness [17–20]. We therefore examined the effect of ghrelin on Akt phosphorylation by phospho-Akt-specific immunoblot. Akt phosphorylation was increased by 10 nM ghrelin in all cell lines (Fig. 5). The increase in Akt phosphorylation induced by ghrelin was abolished by D -LysGHRP6. The role of increased Akt activity in the stimulatory effects of ghrelin on migratory potential and invasiveness was confirmed using the PI3-K inhibitor Wortmannin. Pre-incubation of cells in medium containing 50 nM Wortmannin for 1 h prior to performing the migration and invasion assays abolished the stimu-
Fig. 3. The effect of ghrelin on cellular proliferation was determined by MTT assay in the absence (A), or presence (B), of the GHS-R antagonist D -Lys-GHRP6. Absorbance at 560 nm has been converted to percentages above control (SD). Data represent one of three identical experiments, n ¼ 16 in each case. (A) Ghrelin promoted cellular proliferation in a dose-dependent manner, peaking at 10 nM (*P < 0:01 versus control). Above this concentration, ghrelin had a negative effect upon proliferation. (B) The pro-proliferative effect of ghrelin was abolished by pre-treatment with D -Lys-GHRP6. *P < 0:05 versus 10 nM ghrelin.
Fig. 4. Cells were treated with 10 nM ghrelin or PBS vehicle. Invasion was quantified after 24 h by Matrigel Boyden chamber assay. Motility was quantified in the same way with the exception that Matrigel was omitted. Ten nanomolar ghrelin increased invasiveness and motility in all cell lines tested. *P < 0:05.
M.S. Duxbury et al. / Biochemical and Biophysical Research Communications 309 (2003) 464–468
Fig. 5. Western immunoblot using anti-phospho-Akt (threonine 308) antibody. Total Akt was probed using an anti-Akt antibody. Treatment with 10 nM ghrelin increased Akt phosphorylation at threonine 308 in all cell lines.
Fig. 6. Pre-treatment for 1 h with the PIK-3 inhibitor Wortmannin abolished the stimulatory effects of 10 nM ghrelin on cellular motility and invasiveness.
latory effect of 10 nM ghrelin on cellular migration and invasiveness (Fig. 6). Discussion In this study, we have demonstrated expression of GHS-R1a and 1b isoforms in pancreatic adenocarci-
467
noma cells. Although ghrelin transcript was only weakly present in PANC1 and ghrelin peptide was not detected, cellular proliferation, invasiveness, and motility were stimulated by exposure to 10 nM ghrelin, indicating that pancreatic adenocarcinoma is ghrelin-responsive. In addition, we have demonstrated that the PI3-K/Akt pathway is an important mediator of ghrelin’s stimulatory effects on migration and invasiveness in pancreatic adenocarcinoma cells. To our knowledge, this is the first characterization of ghrelin receptor expression and response to ghrelin exposure in human pancreatic adenocarcinoma cells. Ghrelin is secreted principally by the stomach, and is present in only low levels in normal pancreatic tissue [11], where it is expressed principally by beta cells [21]. The absence of expression of ghrelin peptide in the adenocarcinoma cell lines examined is consistent with a reported series of micro-dissected pancreatic adenocarcinomas in which ghrelin immunoreactivity was not detected [21]. In the case of PANC1, post-transcriptional modulation of ghrelin expression may occur. It is also conceivable that ghrelin secreted by normal islets could act in a paracrine fashion to stimulate pancreatic adenocarcinoma proliferation in vivo. GHS-R1a and 1b expression has been confirmed in a variety of neoplasms including prostate adenocarcinoma [22], pancreatic endocrine tumors [21], somatotroph and other central nervous system tumors [23]. While GHSR1a is a functional G protein-coupled receptor, the role of GHS-R1b is not currently known. Stimulation of prostate adenocarcinoma cellular proliferation is reported following exposure to ghrelin, with a maximal proliferative effect at 5 nM that diminished above this concentration [22]. This is comparable to the maximal stimulation of pancreatic adenocarcinoma growth we observed with 10 nM ghrelin. This growth-promoting effect also declines at higher ghrelin concentrations. The growth-promoting effects of ghrelin on malignant cells are not universal. Suppression of breast, lung, and thyroid adenocarcinoma proliferation has been observed in response to ghrelin treatment [16,24,25]. These antiproliferative effects were generally seen at ghrelin concentrations greater than 10 nM: the level above which the growth-promoting effects of ghrelin subside in pancreatic and prostatic adenocarcinoma cells [22]. The GHS-R antagonist D -Lys-GHRP6 suppresses proliferation in response to 10 nM ghrelin and, at higher ghrelin concentrations, this decline in proliferation becomes more exaggerated in D -Lys-GHRP6-treated cells. Together, these findings suggest first that the growth-promoting effect of ghrelin is mediated via the GHS-R. Second, it raises the possibility that another, lower affinity nonGHS-R receptor may be responsible for mediating the anti-proliferative effects of ghrelin observed at higher concentrations. Evidence is mounting for the existence of a non-GHS-R ghrelin receptor. This hypothesis is
468
M.S. Duxbury et al. / Biochemical and Biophysical Research Communications 309 (2003) 464–468
supported by experiments in which administration of synthetic peptidyl ghrelin analogues which lack the in vivo action of ghrelin, but not administration of ghrelin itself, inhibits DNA synthesis and proliferation in human lung carcinoma cells [25]. Further credence is given to the existence of a non-GHS-R ghrelin receptor by the demonstration that ghrelin binds to normal and neoplastic thyroid tissue, inhibiting proliferation, despite the absence of GHS-R mRNA expression [16]. Cellular invasiveness and motility were promoted by ghrelin. These responses were associated with increased phosphorylation of Akt at threonine 308, which was used as a marker of Akt activation [26]. Akt activation has been associated with increased malignant cellular invasiveness, both in vitro and in vivo [17–20]. Invasiveness, motility, and threonine phosphorylation of Akt are all suppressed by D -Lys-GHRP6, indicating that the stimulatory effects of ghrelin on pancreatic adenocarcinoma cells are mediated via the GHS-R. While these in vitro observations may not directly replicate the response of pancreatic adenocarcinoma to ghrelin in vivo, these data provide clear evidence that pancreatic adenocarcinoma cells are ghrelin-responsive, which may have implications for the use of ghrelin in the setting of cancer cachexia. In conclusion, we have shown that ghrelin increases proliferation, motility, and invasiveness of pancreatic adenocarcinoma cells through an Akt-dependent mechanism, indicating that pancreatic adenocarcinoma is a ghrelin-responsive malignancy. There remains a need for selective non-toxic treatments to attenuate cachexia. However, the use of ghrelin as a treatment for human pancreatic cancer cachexia will require cautious evaluation.
Acknowledgments This study was supported by a grant from the National Pancreas Foundation and by National Institute of Health Grants DK02786 (E.E.W.) and DK47326 (S.W.A.). The authors gratefully acknowledge the technical support of Jan Rounds.
References [1] E.P. DiMagno, H.A. Reber, M.A. Tempero, Gastroenterology 117 (1999) 1464–1484.
[2] S. Keleg, P. Buchler, R. Ludwig, M.W. Buchler, H. Friess, Mol. Cancer 2 (2003) 14. [3] M.D. Barber, J.A. Ross, K.C. Fearon, Nutr. Cancer 35 (1999) 106–110. [4] W.D. DeWys, C. Begg, P.T. Lavin, P.R. Band, J.M. Bennett, J.R. Bertino, M.H. Cohen, H.O. Douglass Jr., P.F. Engstrom, E.Z. Ezdinli, J. Horton, G.J. Johnson, C.G. Moertel, M.M. Oken, C. Perlia, C. Rosenbaum, M.N. Silverstein, R.T. Skeel, R.W. Sponzo, D.C. Tormey, Am. J. Med. 69 (1980) 491–497. [5] L. Ovesen, J. Hannibal, E.L. Mortensen, Nutr. Cancer 19 (1993) 159–167. [6] J.S. Falconer, K.C. Fearon, C.E. Plester, J.A. Ross, D.C. Carter, Ann. Surg. 219 (1994) 325–331. [7] M.J. Tisdale, J. Natl. Cancer Inst. 89 (1997) 1763–1773. [8] M.J. Tisdale, J. Nutr. 129 (1999) 243S–246S. [9] M.J. Tisdale, Support Care Cancer 11 (2003) 73–78. [10] M. Kojima, H. Hosoda, K. Kangawa, Horm. Res. 56 (Suppl. 1) (2001) 93–97. [11] S. Gnanapavan, B. Kola, S.A. Bustin, D.G. Morris, P. McGee, P. Fairclough, S. Bhattacharya, R. Carpenter, A.B. Grossman, M. Korbonits, J. Clin. Endocrinol. Metab. 87 (2002) 2988. [12] J. Pinkney, G. Williams, Lancet 359 (2002) 1360–1361. [13] A.M. Wren, L.J. Seal, M.A. Cohen, A.E. Brynes, G.S. Frost, K.G. Murphy, W.S. Dhillo, M.A. Ghatei, S.R. Bloom, J. Clin. Endocrinol. Metab. 86 (2001) 5992. [14] J. Eisenstein, A. Greenberg, Nutr. Rev. 61 (2003) 101–104. [15] T. Hanada, K. Toshinai, N. Kajimura, N. Nara-Ashizawa, T. Tsukada, Y. Hayashi, K. Osuye, K. Kangawa, S. Matsukura, M. Nakazato, Biochem. Biophys. Res. Commun. 301 (2003) 275–279. [16] M. Volante, E. Allia, E. Fulcheri, P. Cassoni, E. Ghigo, G. Muccioli, M. Papotti, Am. J. Pathol. 162 (2003) 645–654. [17] J.Q. Cheng, B. Ruggeri, W.M. Klein, G. Sonoda, D.A. Altomare, D.K. Watson, J.R. Testa, Proc. Natl. Acad. Sci. USA 93 (1996) 3636–3641. [18] S.J. Grille, A. Bellacosa, J. Upson, A.J. Klein-Szanto, F. van Roy, W. Lee-Kwon, M. Donowitz, P.N. Tsichlis, L. Larue, Cancer Res. 63 (2003) 2172–2178. [19] D. Kim, S. Kim, H. Koh, S.O. Yoon, A.S. Chung, K.S. Cho, J. Chung, FASEB J. 15 (2001) 1953–1962. [20] S. Tanno, S. Tanno, Y. Mitsuuchi, D.A. Altomare, G.H. Xiao, J.R. Testa, Cancer Res. 61 (2001) 589–593. [21] M. Volante, E. Allia, P. Gugliotta, A. Funaro, F. Broglio, R. Deghenghi, G. Muccioli, E. Ghigo, M. Papotti, J. Clin. Endocrinol. Metab. 87 (2002) 1300–1308. [22] P.L. Jeffery, A.C. Herington, L.K. Chopin, J. Endocrinol. 172 (2002) R7–R11. [23] M. Korbonits, M. Kojima, K. Kangawa, A.B. Grossman, Endocrine 14 (2001) 101–104. [24] P. Cassoni, M. Papotti, C. Ghe, F. Catapano, A. Sapino, A. Graziani, R. Deghenghi, T. Reissmann, E. Ghigo, G. Muccioli, J. Clin. Endocrinol. Metab. 86 (2001) 1738–1745. [25] C. Ghe, P. Cassoni, F. Catapano, T. Marrocco, R. Deghenghi, E. Ghigo, G. Muccioli, M. Papotti, Endocrinology 143 (2002) 484–491. [26] J. Downward, Curr. Opin. Cell Biol. 10 (1998) 262–267.