Ghrelin Is Produced by the Human Erythroleukemic HEL Cell Line and Involved in an Autocrine Pathway Leading to Cell Proliferation
http://www.100md.com
内分泌学杂志 2005年第3期
Department of Biochemistry and Nutrition, Faculty of Medicine, Université Libre de Bruxelles, B-1070 Brussels, Belgium
Address all correspondence and requests for reprints to: Dr. Christine Delporte, Department of Biochemistry and Nutrition, Faculty of Medicine, Université Libre de Bruxelles, Bat G/E, CP 611, 808 route de Lennik, B-1070 Brussels, Belgium. E-mail: cdelport@ulb.ac.be.
Abstract
Ghrelin, a ligand of the GH secretagogue receptor (GHS-R 1a), is a 28-amino acid peptide with an unusual octanoyl group on Ser3, crucial for its biological activity. For the first time, ghrelin and GHS-R 1b, a truncated variant of the receptor resulting from alternative splicing, but not GHS-R 1a, mRNAs were detected in the human erythroleukemic cell line HEL. Two antibodies, used for RIA, were directed against octanoylated and total (octanoylated and desoctanoylated) ghrelin, and the recognized epitopes were characterized. Using reverse phase HPLC analysis followed by RIA, we demonstrated that octanoylated and desoctanoylated ghrelins were present in HEL cells and their culture medium, of which more than 90% was octanoylated. The ghrelin levels were not affected after 24 h treatment with sodium butyrate, phorbol 12-myristate 13-acetate, or forskolin, but a significant 3-fold increase in desoctanoylated ghrelin was detected in the culture medium after 48 h treatment with sodium butyrate. The antighrelin SB801 and SB969 antisera inhibited HEL cell proliferation by 24% and 39%, respectively, after 72 h. Taken together, these data suggested that endogenous ghrelin stimulated HEL cell proliferation by an autocrine pathway involving an unidentified receptor, distinct from GHS-R1a, and that the HEL cell line represents a unique model to study the octanoylation of ghrelin.
Introduction
GHRELIN IS A novel gastrointestinal hormone that was recently purified and identified from rat stomach (1). Ghrelin is a 28-amino acid peptide containing an n-octanoyl modification on the Ser3 residue, which is essential for its biological activity, and a C-terminal acid. Its C-terminal sequence can be shortened without loss of biological activity (2). Since its discovery, other forms of ghrelin were purified from the stomach: des-Gln14-ghrelin, resulting from an alternative splicing of rat ghrelin gene (3); and decanoyl ghrelin-(1–28), decenoyl ghrelin-(1–28), octanoyl ghrelin-(1–27), decanoyl ghrelin-(1–27), and des-acyl ghrelin-(1–27), produced by posttranslational processing of the ghrelin precursor in human stomach (4). Ghrelin stimulates GH release from the pituitary (5) and regulates food intake and energy metabolism (6, 7). These biological actions of ghrelin are mainly mediated by the GH secretagogue receptor (GHS-R) (8). Two GHS-R subtypes, generated by alternative splicing of a single gene, have been cloned: GHS-R type 1a (GHS-R 1a) and GHS-R type 1b (GHS-R 1b). The human GHS-R 1a consists of 366 amino acids with seven transmembrane domains. Its activation leads to inositol triphosphate generation and Ca2+ release through activation of the G protein G11. The GHS-R 1b consists of 289 amino acids with only the first five transmembrane domains, followed by a 24-amino acid sequence encoded by an alternatively spliced intronic sequence. GHS-R 1b fails to bind GHS or respond to GHS and is not known to exhibit any ghrelin-mediated biological activity.
Circulating ghrelin is mainly derived from the stomach and is influenced by the feeding state (9). The most abundant circulating form of ghrelin is des-acyl ghrelin (10), a form of ghrelin that does not act on the GHS-R 1a. Only acylated forms of ghrelin bind to the GHS-R 1a and exert endocrine actions. Recently, however, des-acyl ghrelin was shown to modulate cell proliferation in prostate carcinoma cell lines (11), to stimulate adipogenesis (12), to induce cardiovascular effects (13), and to inhibit apoptosis in cardiomyocytes and endothelial cells (14). These effects could be mediated by an as yet unidentified ghrelin receptor.
In addition to the stomach, ghrelin has been detected in several tissues, such as intestine, kidney, pituitary, pancreas, placenta, lung, testis, and ovary (15); in tumors, such as pituitary adenomas (16), gastrointestinal carcinoids (17), and endocrine pancreatic tumors (18); and in cell lines, such as prostate neoplasms (11, 19) and medullary thyroid carcinoma (20). Ghrelin mRNA was also found in T cells, B cells, and neutrophils as well as in leukemic B, T, and myeloid cell lines (21).
The human erythroleukemic cell line HEL is representative of the erythroblastic stage of differentiation of hemopoietic cells, retains a number of features of the megakaryocyte/platelet lineage (22), and can differentiate into megakaryocytes, macrophages, or erythrocytes (23, 24, 25, 26).
The aims of the present study were to examine, in the human erythroleukemic HEL cell line, the presence of ghrelin and GHS-R subtypes, ghrelin expression during cellular differentiation, and the effects of ghrelin on cell proliferation. To achieve this, we obtained and characterized two antibodies directed against human ghrelin to set up specific RIAs and developed extraction and chromatographic procedures to separate ghrelin from des-acyl ghrelin.
Materials and Methods
Peptide synthesis
All peptides used were synthesized by solid phase methodology using the 9-fluorenyl-methoxy-carbonyl (fmoc) strategy (27) with a Symphony Multiplex apparatus (Protein Technologies, Inc., Tucson, AZ) and were purified by reverse phase HPLC (RP-HPLC) on Amberchrome-type resin 78 CG-162sd (15.9 x 2.5 cm), Vydac 259HP510A (25 x 1 cm; Alltech, Laarne, Belgium) and Vydac 259VHP54 (25 x 0.46 cm; Alltech). The Ser3 hydroxyl group of ghrelin was acylated with n-octanoic acid using 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide in the presence of 4-(dimethylamino)pyridine according to the method of Bednarek et al. (2). Selective protection of Ser3 was obtained using fmoc Ser-Trt, instead of fmoc Ser-Tbu for Ser2, to allow its selective deprotection and using a solution of 1% trifluoroacetic acid (TFA) in dichloromethane before subsequent octanoylation. All ghrelin analogs were based on the human sequence, unless specified otherwise, and those possessing a C-terminal acidic function instead of an amide were noted by -OH. Peptide purity (>95%) was assessed by capillary electrophoresis (P/ACE MDQ, Beckman Coulter, Fullerton, CA), and molecular weight was verified by electrospray mass spectrometry using a VG Platform ns 8230E (Waters Corp., Milford, MA).
Preparation of antighrelin sera
[Cys12]Ghr-(1–11) (1.5 mg) and [Cys0]Ghr-(13–28)-OH (1.5 mg) were conjugated separately to keyhole limpet hemocyanin using m-maleimidobenzoyl-N-hydroxysuccinimide ester (Eurogentec, Herstal, Belgium). Each conjugate was emulsified with Freund’s adjuvant. Two batches of antisera were obtained from New Zealand White rabbits after four monthly intradermic injections of conjugate (Eurogentec, Seraing, Belgium).
Peptide radioiodination
[Tyr24]Ghr-(1–23) and [Tyr0]Ghr-(13–28)-OH were radioiodinated on tyrosine by the Iodogen method (28) and were purified on a Sep-Pak C18 cartridge (Waters Corp.).
RIA for ghrelin
Assays were performed in duplicate at 4 C in a 300-μl final volume containing 100 μl standard human Ghr-(1–28)-OH or unknown sample, 100 μl antiserum (diluted at 1/10,000) containing 0.1% normal rabbit serum, and 100 μl 125I-labeled tracer (10,000 cpm) diluted in RIA buffer [10 mM sodium phosphate buffer (pH 7.4), 0.05% Tween 20, 150 mM NaCl, 10 mM EDTA, and 0.1% sodium azide]. The antigen and antibody were preincubated overnight, followed by the tracer addition. After an additional 18-h incubation, 1 ml 4% polyethylene glycol 6000 diluted in RIA buffer containing 1% sheep antirabbit serum was added. After a 30-min incubation at room temperature, free and bound tracers were separated by centrifugation for 20 min at 2600 x g. After aspiration of the supernatant, the radioactive pellet was counted into a -counter. Nonspecific binding was determined in the absence of antiserum. For SB801 and SB969 antibodies, the limits of detection for ghrelin were 3 and 7 fmol/assay; the intraassay coefficients of variation were 3% and 1%; and the peptide recoveries were 97% and 98%, respectively.
Ghrelin (and ghrelin receptors) mRNA detection by RT-PCR
HEL total RNA was extracted using the SV RNA extraction kit (Promega Corp., Leiden, The Netherlands), which includes a deoxyribonuclease treatment. cDNA synthesis was performed using 1 μg total RNA with the Expand Reverse Transcriptase (Roche, Brussels, Belgium). The resulting cDNA was subjected to PCR amplification with 0.4 μM of the sense and antisense primers and 0.25 U Goldstar DNA polymerase (Eurogentec, Seraing, Belgium). The PCR primers used were: human ghrelin: sense, 5'-AAGGAGTCGAAGAAGCCACCA-3' (nucleotides 148–168); and antisense, 5'-GCCAGATGAGCGCTTCTAAACTTA-3' (nucleotides 416–439 in accession no. AB029434, GenBank); human GHS-R 1a and GHS-R 1b: sense, 5'-TCTTCCTTCCTGTCTTCTATC-3' (nucleotides 662–682 in accession no. U60179 and U60181, GenBank); human GHS-R 1a: antisense, 5'-AGTCTGAACACTGCCACC-3' (nucleotides 993-1010 in accession no. U60179, GenBank); human GHS-R 1b: antisense, 5'-TCAGAGAGAAGGGAGAAGG-3' (nucleotides 852–870 in accession no. U60181, GenBank); and human ?-actin: sense, 5'-TGACGGGGTCACCCACACTGTGCCCGTC-3' (nucleotides 539–566); and antisense, 5'-CTAGAAGCATTAGCGGTGGACGATGGAGG-3' (nucleotides 1171–1199; in accession no. BC002409, GenBank). Amplification of ?-actin served as a quality control for the RNA. For ghrelin cDNA amplification, 35 cycles were performed (10 sec at 94 C, 10 sec at 57 C, and 1 min at 72 C). For ghrelin receptor cDNA amplification, 40 cycles were performed (10 sec at 94 C, 20 sec at 50 and 58 C for, respectively, GHS-R 1a and GHS-R 1b, and 45 sec at 72 C). For ?-actin amplification, 25 cycles were performed (1 min at 94 C, 1 min at 60 C, and 1 min at 72 C). Five microliters of each PCR product were submitted to electrophoresis on a 1.4% agarose gel stained with ethidium bromide and visualized under UV light. For ghrelin, GHS-R 1a, and GHS-R 1b PCRs, positive controls included, respectively, a pSG5 vector containing human ghrelin cDNA (29), a pEGFP-N1 vector containing human GHS-R 1a cDNA (30), and human placenta cDNA.
Cell culture, cell treatment, and cell lysate preparation
HEL cells were grown in RPMI 1640 medium (BioWhittaker Europe, Verviers, Belgium) supplemented with 10% inactivated fetal bovine serum, 100 IU/ml streptomycin-penicillin, and 4 mM glutamine and were routinely passaged twice a week. In some experiments HEL cells were treated, or not, for 24 or 48 h with 33 nM phorbol 12-myristate 13-acetate (PMA), 10 μM forskolin, or 1 mM sodium butyrate, then submitted to peptide extraction and RP-HPLC separation, followed by RIA. HEL cell lysate was prepared as follows. HEL cells were sonicated in distilled water for 30 sec and centrifuged at 4 C at 1000 x g for 10 min. The supernatant was collected and used for determination of butyrylcholinesterase and carboxylesterase activities.
Peptide extraction and RP-HPLC separation
HEL cells were centrifuged at 500 x g for 10 min. The medium was collected and acidified with HCl to pH 4–5. The cell pellet was resuspended into 10 mM HCl, frozen in liquid nitrogen, heated at 100 C for 5 min, cooled, and centrifuged at 20,000 x g for 30 min at 4 C. The resulting supernatant was collected. The acidified medium and supernatant were loaded onto a Sep-Pak C18 column preequilibrated with 3% CH3CN/0.1% TFA. After washing with 10% CH3CN/0.1% TFA, the peptides were eluted with 60% CH3CN/0.1% TFA. The eluates were lyophilized in a Speed-Vac concentrator, subjected to RP-HPLC analysis on a C18 Vydac 218TP54 column (25 x 0.46 cm; Alltech), and equilibrated with 3% CH3CN/0.1% TFA, using a linear gradient of CH3CN from 3–80% in 0.1% TFA for 50 min. The OD was monitored at 226 nm. RP-HPLC fractions were collected, lyophilized, and submitted to RIA.
Degradation of synthetic ghrelin exogenously added to culture medium in the presence of HEL cells
One microgram of synthetic ghrelin was added to 10 ml culture medium in the presence of 2.5 x 106 HEL cells. The culture media were collected at various times, then submitted to peptide extraction and RP-HPLC separation, followed by RIA.
Assay for butyrylcholinesterase activity
Butyrylcholinesterase activity was measured by the method of Ellman et al. (31), using an LKB Ultrospec Plus 4054 UV/visible spectrophotometer (LKB, Bromma, Sweden). One hundred microliters of HEL cell lysate or inactivated fetal bovine serum (FBS) were added to 100 mM butyrylthiocholine iodide and 0.25 mM 5',5'-dithiobis-2-nitrobenzoic acid in 50 mM Tris-HCl, pH 7.4. The absorbance was read at 405 nm every 30 sec for up to 6 min. The enzyme activity was calculated as micromoles of the product per minute (as unit, U) after correction for nonenzymatic hydrolysis of the substrate using the extinction coefficient (13,300 M–1 · cm–1) of the product.
Assay for carboxylesterase activity
Carboxylesterase activity was determined by measuring the hydrolysis of -naphtylacetate (32, 33). One hundred microliters of HEL cell lysate or inactivated FBS were preincubated at 37 C for 20 min with 10 μM eserine to inhibit acetyl- and butyrylcholinesterases and with 10 mM EDTA to inhibit paraoxonase, then 10 μl 0.02 M -naphtylacetate were added in a 100 mM phosphate buffer, pH 7.0. The absorbance was measured at 321 nm every 10 min for up to 60 min. Enzyme activity was calculated as micromoles of the product per minute using the extinction coefficient (2200 M–1 · cm–1) of the product.
HEL cell proliferation studies
HEL cells were grown in six-well plates (5 x 105 cells/ml) in RPMI 1640 medium (BioWhittaker Europe) supplemented with 2.5% inactivated fetal bovine serum, 100 IU/ml streptomycin-penicillin, and 4 mM glutamine and were incubated at 37 C for up to 72 h with or without 1 μM octanoylated ghrelin, 1 μM des-acyl ghrelin, 1% SB801, 1% SB969, or 1% of both SB801 and SB969. Appropriate controls were performed for each condition. The addition of 1% or 2% preimmune rabbit serum did not significantly modify HEL cell proliferation. After 0-, 24-, 48-, and 72-h incubation, cells were counted in triplicate in a Multisizer III (Coulter Electronics Ltd., Luton, UK).
Protein assay
The protein concentration was determined using Bradford’s method (34).
Data analysis
Data are summarized as the mean ± SEM. Results were statistically analyzed using the paired t test or one-way ANOVA, followed by Tukey-Kramer multiple comparisons test. All statistical values reported were obtained using GraphPad InStat version 3.02 for Windows (GraphPad, Inc., San Diego, CA). P < 0.05 was considered significant.
Results
Characterization of the ghrelin epitope recognized by SB801 antiserum
SB801 antiserum was obtained after rabbit immunization using a synthetic human ghrelin peptide corresponding to the N-terminal 11 amino acids [Ghr-(1–11)], including the n-octanoylated Ser3 (Fig. 1). The sequence of this peptide differs from its rat homologous peptide by the presence of Arg11, instead of Lys11.
FIG. 1. Polyclonal antibodies raised against human ghrelin. SB801 and SB969 antisera were obtained by immunization with the synthetic human peptides Ghr-(1–11) and Ghr-(13–28), respectively, as described in Materials and Methods. Epitopes recognized by the antisera are delimited by the boxes.
SB801 antiserum recognized with similar high affinity human and rat Ghr-(1–28)-OH, suggesting that the amino acid in position 11 was not important for recognition of the peptide by the antiserum. It also recognized Ghr-(1–28)-OH, rat Ghr-(1–28)-OH, Ghr-(1–23), Ghr-(1–14), Ghr-(1–7)-OH, and Ghr-(1–5) with similar high affinities, whereas des-acyl ghrelin analogs and Ghr-(3–23) were not recognized (Figs. 2 and 3 and Table 1). Compared with Ghr-(1–28)-OH, decanoyl Ghr-(1–28)-OH had a 2.8-fold lower affinity for SB801 antiserum, suggesting that octanoyl was the optimal acyl group for antibody recognition (Fig. 2 and Table 1).
FIG. 2. Competition curves of ghrelin analogs for the binding of [125I]Tyr24-Ghr-(1–23) to SB801 antiserum. RIA was performed as described in Materials and Methods. The experiments were performed in duplicate, and the results are expressed as the percentage of specific binding (B) in the absence of peptides (B0; mean ± SEM; n = 3). rGhr, Rat Ghr.
FIG. 3. Competition curves of Ala-scan analogs of ghrelin-(1–14) for the binding of [125I]Tyr24-Ghr-(1–23) to SB801 antiserum. RIA was performed as described in Materials and Methods. The experiments were performed in duplicate, and the results are expressed as a percentage of specific binding (B) in the absence of peptides (B0; mean ± SEM; n = 3).
TABLE 1. Ki values for several ghrelin analogs on [125I]Tyr24-Ghr-(1–23) binding to SB801 antiserum
Because Ghr-(1–14) was recognized with similar affinity as Ghr-(1–28)-OH, each of its amino acids was systematically substituted by Ala to determine the amino acids involved in recognition by SB801 antiserum. The SB801 antiserum did not discriminate [Ala1]-, [Ala6]-, [Ala7]-, [Ala8]-, [Ala9]-, [Ala10]-, [Ala11]-, [Ala12]-, [Ala13]-, or [Ala14]-Ghr-(1–14) from Ghr-(1–14) (Fig. 3 and Table 2), whereas [Ala2]-, [Ala4]-, and [Ala5]-Ghr-(1–14) were recognized with 8-, 66-, and 1.7-fold lower affinities, and [Ala3]-Ghr-(1–14) was not recognized (Fig. 3A and Table 2). These results suggested that the ghrelin epitope recognized by SB801 antiserum includes positions 2–5.
TABLE 2. Ki values for Ala scans ghrelin analogs on [125I]Tyr24-Ghr-(1–23) binding to SB801 antiserum
Characterization of the ghrelin epitope recognized by SB969 antiserum
SB969 antiserum was obtained after rabbit immunization using a synthetic human ghrelin peptide corresponding to the C-terminal 16 amino acids of human ghrelin [Ghr-(13–28); Fig. 1]. The sequence of this peptide is identical in both human and rat ghrelins.
SB969 antiserum recognized Ghr-(1–28)-OH, des-acyl Ghr-(1–28)-OH, decanoyl Ghr-(1–28)-OH, Ghr-(1–23), and des-acyl Ghr-(1–23) with Ki values of 303, 180, 194, 647, and 877 pM, respectively (Fig. 4A and Table 3). N-Terminal shortening of Ghr-(1–23) [Ghr-(3–23), Ghr-(4–23), Ghr-(5–23), and Ghr-(8–23)] did not affect antibody recognition (Fig. 4A and Table 3). N-Terminal-shortened analogs of Ghr-(1–28) ([Tyr0]Ghr-(13–28)-OH, [Cys0]Ghr-(13–28)-OH, and Ghr-(13–28)-OH) had somewhat higher affinities than Ghr-(1–28) for the SB969 antiserum, suggesting that the N-terminal ghrelin sequence impaired recognition of the C terminus of the antibody (Fig. 4B and Table 3). Amidated [Tyr0]Ghr-(13–28), [Cys0]Ghr-(13–28), as well as C-terminal truncated Ghr-(13–28)-OH [Ghr-(13–27)-OH, Ghr-(13–26)-OH, Ghr-(13–25)-OH, and Ghr-(13–24)-OH] increased nondisplaceable binding and decreased the peptide affinity for the antibody (Fig. 4B and Table 3). Taken together, these results indicated that the antibody recognized at least the five C-terminal amino acids of ghrelin in addition to its C-terminal acid.
FIG. 4. Competition curves of ghrelin analogs for the binding of [125I]Tyr0-Ghr-(13–28)-OH to SB969 antiserum. RIA was performed as described in Materials and Methods. The experiments were performed in duplicate, and the results are expressed as the percentage of specific binding (B) in the absence of peptides (B0; mean ± SEM; n = 3).
TABLE 3. Ki values for several ghrelin analogs on [125I]Tyr0-Ghr-(13–28)-OH binding to SB969 antiserum
GHS-R subtypes and ghrelin mRNA detection by RT-PCR in HEL cells
RT-PCR products of the size expected for GHS-R 1b and ghrelin (209 or 292 bp, respectively), were obtained from HEL cells (Fig. 5). Direct sequencing of the PCR products confirmed that they corresponded to GHS-R 1b and ghrelin (data not shown). In contrast, no RT-PCR product of the size expected for GHS-R 1a (349 bp) could be obtained from HEL cells (Fig. 5).
FIG. 5. GHS-R subtypes and ghrelin mRNA expression in HEL cells. RNA extraction, RT, and PCR were performed as described in Materials and Methods. Positive controls were performed using pEGFP-N1 vector containing human GHS-R 1a cDNA diluted at 100 pg/μl, human placenta cDNA, or PSG5 vector containing human ghrelin diluted at 10 pg/μl. Negative controls were performed by omitting cDNA in the PCR mixture.
Ghrelin identification in HEL cells by HPLC followed by RIA
To identify ghrelin, HEL cells and their culture medium were subjected to peptide extraction and subsequent RP-HPLC analysis, followed by RIA using both SB801 and SB969 antisera. An immunoreactive (IR) ghrelin peak, detected in RIA using SB801 antiserum, eluted at a position identical to that of human octanoylated ghrelin (Fig. 6, A and B). Two IR peaks, detected by RIA using SB969 antiserum, eluted at positions identical to human octanoylated ghrelin and human desoctanoylated ghrelin (Fig. 6, C and D).
FIG. 6. Identification of IR-ghrelin in HEL cells and their culture medium. After 24 h of culture, peptide extracts from HEL cells (A and C) and their culture medium (B and D) were submitted to RP-HPLC on a C18 Vydac 218TP54 column using a linear gradient of 3–80% CH3CN containing 0.1% TFA for 50 min (see Materials and Methods). HPLC fractions of 1 ml were lyophilized, then submitted to RIA using SB801 (A and B) or SB969 (C and D). IR-ghrelin was expressed as femtomoles per fraction. Arrows indicate the elution positions of human desoctanoylated (1 ) and octanoylated (2 ) ghrelins.
Using SB801 antiserum, levels of IR-ghrelin eluting at the octanoylated ghrelin position were 543 ± 17 fmol/106 cells (1784 ± 56 fmol/mg protein; n = 3) and 97 ± 32 fmol/106 cells (320 ± 106 fmol/mg protein; n = 3) in HEL cells and their culture medium after 24 h of culture, respectively. Using SB969 antiserum, IR-ghrelin levels eluting at desoctanoylated ghrelin position were 39 ± 18 fmol/106 cells (128 ± 57 fmol/mg protein; n = 3) and 11 ± 4 fmol/106 cells (36 ± 13 fmol/mg protein; n = 3) in HEL cells and their culture medium after 24 h of culture, respectively.
To determine whether inactivated FBS could contribute to IR-ghrelin found in the culture medium of HEL cells after 24 h, IR-ghrelin was measured in inactivated FBS. IR-ghrelin levels assayed with the SB801 and SB969 antisera were 25 and 115 fmol/ml inactivated FBS, respectively.
Degradation of synthetic ghrelin exogenously added to culture medium in the presence of HEL cells
To determine whether secreted ghrelin was degraded, synthetic ghrelin (1 μg) was exogenously added to culture medium in the presence of HEL cells. After 1- and 24-h incubation, IR-ghrelin assayed with the SB801 antiserum represented 46 ± 17% and 0.79 ± 0.13%, respectively (n = 3), of ghrelin at time zero.
Determination of butyrylcholinesterase and carboxylesterase activities in inactivated FBS and HEL cell lysate
To test whether butyrylcholinesterase and carboxylesterase activities from inactivated FBS and/or HEL cells participated in ghrelin degradation, both esterase activities were measured in inactivated FBS and in HEL cell lysate. Butyrylcholinesterase activity, calculated for one culture flask of HEL cells containing 50 ml medium, was 0.320 and 0.054 U for inactivated FBS and HEL cells, respectively. Carboxylesterase activity, calculated for one culture flask of HEL cells containing 50 ml medium, was 0.024 and 0.057 U for inactivated FBS and HEL cells, respectively.
HEL cell treatments with differentiating agents
HEL cells were untreated or treated for 24 or 48 h with 33 nM PMA, 1 mM sodium butyrate, or 10 μM forskolin, then submitted to peptide extraction and RP-HPLC separation, followed by RIA. After 24 h of treatment, octanoylated and desoctanoylated ghrelins levels were not modified. After 48 h of treatment, octanoylated and desoctanoylated ghrelins levels were not modified by PMA and forskolin, whereas sodium butyrate significantly increased desoctanoylated ghrelin in culture medium by 346 ± 29% (n = 3; P < 0.05).
Determination of butyrylcholinesterase and carboxylesterase activities in HEL cell lysate and HEL cell medium after treatment with sodium butyrate
To test whether butyrylcholinesterase and carboxylesterase activities were modified after 48 h treatment with 1 mM sodium butyrate, both esterase activities were measured in HEL cell lysate and HEL cell medium. In both preparations, butyrylcholinesterase activity was not different in control conditions or after sodium butyrate treatment. In HEL cell lysate, carboxylesterase activity was 1.7-fold higher in HEL cells treated with sodium butyrate. In HEL cell medium, carboxylesterase activity was similar in control and sodium butyrate-treated HEL cells.
Effects of sodium butyrate on purified butyrylcholinesterase and carboxylesterase enzymes
To test whether butyrylcholinesterase and carboxylesterase activities were modified by a direct effect of sodium butyrate, both enzyme activities were measured in the absence or presence of 1 mM sodium butyrate. Butyrylcholinesterase and carboxylesterase activities were not affected in the presence of sodium butyrate.
HEL cell proliferation studies
To explore the effect of ghrelin on HEL cell proliferation, HEL cells were incubated with 2.5% FBS for up to 72 h with or without 1 μM octanoylated ghrelin, 1 μM des-acyl ghrelin, 1% SB801, 1% SB969, or 1% of both SB801 and SB969. Appropriate controls were performed for each condition. The addition of 1% or 2% preimmune rabbit serum did not significantly modify HEL cell proliferation. The ability of the SB801 and SB969 antisera to inhibit the biological activity of ghrelin was verified by performing a dose-effect curve of ghrelin on the intracellular calcium increase in Chinese hamster ovary cells coexpressing the recombinant GHS-R1a and aequorin (30). Under these conditions, SB801 and SB969 antisera increased by 78- and 15-fold, respectively, the 50% effective concentration of ghrelin on intracellular calcium increase.
Octanoylated and des-acyl ghrelin had no significant effect on HEL cell proliferation (data not shown). After 48 h treatment, SB969 significantly decreased HEL cell proliferation by 24%. After 72 h treatment, SB801 and SB969 decreased HEL cell proliferation by 24% and 39%, respectively. HEL cell proliferation measured after 48 and 72 h in the simultaneous presence of both antibodies was not statistically different from that measured in the presence of SB969 alone and was statistically different from that measured in the presence of SB801 alone (Fig. 7).
FIG. 7. HEL cell proliferation. HEL cells (5 105 cells/ml) grown in RPMI 1640 medium supplemented with 2.5% inactivated FBS, 100 IU/ml streptomycin-penicillin, and 4 mM glutamine were incubated for up to 72 h with 1% or 2% preimmune rabbit serum, 1% SB801, 1% SB969, or 1% of both SB801 and SB969 and counted in triplicate every 24 h as described in Materials and Methods. The results of four separate experiments performed in triplicate are expressed as the percentage of HEL cell number at time zero (mean ± SEM). Data significantly different compared with control: *, P < 0.001; compared with SB801: #, P < 0.05.
Discussion
Expression of the GHS-R 1b mRNA is widespread, whereas GHS-R 1a receptor expression is more restricted. It is predominantly expressed in the pituitary and (at much lower levels) in the thyroid, pancreas, spleen, myocardium, and adrenal gland (35). Ghrelin is expressed in many tissues and cell lines (10, 11, 15, 16, 17, 18, 19, 20). For blood cells, a previous study reported the presence of both ghrelin and ghrelin receptors mRNAs in human T cell, B cell, and myeloid leukemic cell lines as well as in normal T cells, B cells, and neutrophils with large individual variation (21). There is evidence indicating that several cancer cells express the components of the ghrelin/GHS-R axis and that the axis may have an important autocrine/paracrine role in regulating cancer cell proliferation.
In the present study ghrelin and GHS-R 1b mRNAs were first detected by RT-PCR in erythroleukemic HEL cells. In contrast, mRNA encoding the functional GHS-R 1a receptor, specific binding of [125I]Y24-ghrelin-(1–23), or an increase in free intracellular Ca2+ concentration in response to ghrelin (data not shown) was not observed in HEL cells.
To detect ghrelin synthesis, we developed two RIAs using SB801 and SB969 antisera. SB801 antiserum is probably capable of recognizing all biologically active forms of ghrelin, because C-terminal-shortened acyl ghrelins [down to Ghr-(1–5)] have been shown to retain their full biological activity on the GHS-R 1a, without an important change in their potency (2), and SB969 antiserum allows measurements of ghrelin, des-acyl ghrelin, as well as N- and, to some extent, C-terminal-shortened analogs.
Many studies have evaluated plasma, serum, or tissue ghrelin levels in human and animal models and attempted to correlate the results with several parameters, such as diet-induced weight loss or fasting (36, 37). The specificity of our antisera implies that SB801 does not measure solely acylated Ghr-(1–28), and that SB969 probably underestimates the quantity of total biologically active ghrelin if C-terminally truncated ghrelin fragments exist. Therefore, it is inappropriate to estimate the nonacylated ghrelin concentrations by merely subtracting the octanoylated ghrelin concentration from the total ghrelin concentration. Besides, as previously reported, great care should be taken when assimilating N-terminal IR-ghrelin to octanoylated ghrelin, and C-terminal IR-ghrelin to total ghrelin, because the specificities of the available antibodies have not been studied in detail.
Using RP-HPLC analysis followed by RIA, we demonstrated for the first time that HEL cells produced an important amount of octanoylated ghrelin (543 ± 17 fmol/106 cells = 1784 ± 56 fmol/mg protein; n = 3), a level only 2-fold lower than that in rat stomach (377 fmol/mg wet tissue = 3770 fmol/mg protein) (10) and 44-fold higher than that in medullary thyroid carcinoma TT cells (8.9 fmol/106 cells = 40.5 fmol/mg protein) (20). Over a 24-h period, HEL cells secreted octanoylated ghrelin into their culture medium (18% the amount of ghrelin present in the cells). The proportion of octanoylated ghrelin represented about 90% of the total ghrelin in HEL cells and their culture medium, compared with 17% in rat stomach (10). To our knowledge, desoctanoylated ghrelin has always been reported to be by far the major form of ghrelin present in tissues and cells (10, 20). Because of its capacity to synthesize high quantities of octanoylated ghrelin, the HEL cell line represents a unique model to study the octanoylation of ghrelin.
We recently showed that octanoylated ghrelin was degraded by butyrylcholinesterase in human serum and carboxylesterase in rat serum (38). A rapid and important degradation of ghrelin exogenously added to the culture medium (46% after 1 h) was observed in the presence of HEL cells by RIA using the SB801 antiserum. This degradation could result from the action of esterases from both HEL cells and FBS. Indeed, both butyrylcholinesterase and carboxylesterase activities were detected in inactivated FBS and HEL cell lysate. Besides, ghrelin was a better substrate for carboxylesterase than for butyrylcholinesterase. Indeed, in 1 h, 0.1 U purified carboxylesterase degraded 50% of ghrelin, whereas 2.5 U purified butyrylcholinesterase degraded only 20% of ghrelin (our unpublished observations). The butyrylcholinesterase activity in FBS and in HEL cell lysate represented 0.320 and 0.054 U/flask of HEL cells, respectively. This activity accounts for only 2.6% and 0.4% of ghrelin degradation after 1 h in the culture medium and is negligible compared with the 50% ghrelin degradation observed. In contrast, per flask of HEL cells, carboxylesterase activity in FBS and in HEL cell lysate represented 0.024 and 0.057 U, respectively. This is sufficient to account for the degradation of 12.0% and 28.5% of the added ghrelin after 1 h. Therefore, the carboxylesterase activity of HEL cells is probably responsible for most of the ghrelin degradation observed after 1 h. These data imply that we underestimated the octanoylated ghrelin secretion in the culture medium. When considering the possible proteolysis and the rapid degradation of added ghrelin in the culture medium, the amounts of octanoylated and desoctanoylated ghrelins in FBS would be negligible compared with those measured in the HEL culture medium after 24 h. The unexpected low amount of immunoreactive desoctanoylated ghrelin in the culture medium (10% of the amount of total ghrelin) could result from a C-terminal proteolysis of ghrelin and/or desoctanoylated ghrelin. Indeed, if ghrelin degradation occurs at the C-terminal end of ghrelin, the resulting fragments will be recognized by SB801 [if acylated, down to Ghr-(1–5)], but not, or poorly, by SB969 antiserum. This hypothesis is supported by the lower IR-ghrelin, corresponding to the elution position of octanoylated ghrelin (see peak 2 of the RP-HPLC analysis of HEL cell and medium; Fig. 6), detected by SB969 antiserum compared with that detected by SB801 antiserum. This suggests that some C-terminal-shortened ghrelin forms might indeed coelute with Ghr-(1–28) in our RP-HPLC system.
HEL cells are a poorly differentiated triphenotypic cell line constitutively expressing an erythroid phenotype, but also expressing antigens of other lineages (25, 39, 40). HEL cells increase their erythroid phenotype after stimulation with agents such as sodium butyrate (23, 24, 26) and their macrophage phenotype or megakaryocyte/platelet phenotypes after stimulation with phorbol esters or other agents (23, 24, 25). We did not observe any modification of ghrelin or desoctanoylated ghrelins levels after 24 h treatment with sodium butyrate, PMA, or forskolin. We only observed a significant 3-fold increase in desoctanoylated ghrelin into the culture medium after 48 h treatment with sodium butyrate. Therefore, it is unlikely that ghrelin can be used as a differentiation marker or plays a crucial role during the differentiation processes. The increase in desoctanoylated ghrelin observed after 48 h treatment with sodium butyrate could be linked to a concomitant increase in carboxylesterase production by HEL cells. Indeed, a 1.7-fold increase in carboxylesterase activity, but no increase in butyrylcholinesterase activity, was detected in treated HEL cell lysate, suggesting that only carboxylesterase could participate in the desoctanoylation of ghrelin in the culture medium. Because carboxylesterase isoforms possess distinct cellular distribution (intracellular, secreted into the medium, or bound to the extracellular surface of the membrane), their presence at the cell surface could explain the increase in desoctanoylated ghrelin after 48 h HEL cell treatment with sodium butyrate.
Hattori et al. (21) recently showed that T and B lymphocytes as well as neutrophils express ghrelin and GHS-R mRNA transcripts, but did not investigate the presence of the peptide or the receptor protein. Dixit et al. (41) recently demonstrated that ghrelin is endogenously produced and secreted by human T lymphocytes and monocytes, but without distinguishing octanoylated and desoctanoylated ghrelins. In humans, ghrelin levels declined only by 35–50% for 1 wk postgastrectomy and increased thereafter, suggesting that other tissues participate in ghrelin production (4). Those data and ours showing the production of ghrelin by undifferentiated and differentiated (into megakaryocytes, macrophages, or erythrocytes) HEL cells strongly suggest that several blood cell types express and produce ghrelin.
It is interesting to observe that HEL cells expressed the ghrelin mRNA transcript, contained extremely high quantities of octanoylated ghrelin, and secreted this peptide. Because ghrelin is known to modulate cell proliferation in several cell types (11, 12, 19), HEL cell proliferation was evaluated by adding exogenous ghrelin, des-acyl ghrelin, or SB801 and SB969 antisera used as ghrelin antagonists. HEL cell proliferation was inhibited at 72 h by the anti-ghrelin SB801 and SB969 antisera, with a greater effect of SB969 compared with SB801, and was unaffected by exogenous ghrelin and des-acyl ghrelin, suggesting that endogenous octanoylated and des-acyl ghrelins are sufficient to stimulate HEL cell proliferation. This autocrine effect could involve a specific ghrelin receptor distinct from the GHS-R1a receptor; evidence for its absence in HEL cells has been demonstrated. Besides this autocrine proliferative effect, ghrelin might also be involved in cytokine and/or chemokine regulation in pathological conditions associated with inflammation. Indeed, ghrelin was recently shown to inhibit TNF--induced IL-8 and monocyte chemoattractant protein-1 secretion in human endothelial cells and in a rat model of endotoxic shock as well as mononuclear cell adhesion (42) and exerts inhibitory effects on the expression and production of the inflammatory cytokines IL-1?, IL-6, and TNF- by human T cells and monocytes upon cellular activation and leptin exposure (41). Also, ghrelin was shown to attenuate the development of acute pancreatitis in rats by reducing inflammatory infiltrates of pancreatic tissues, vacuolization of acinar cells, plasma lipase activity, and IL-1? concentration (43). In experimental arthritis in rats and in rheumatoid arthritis in humans, decreased ghrelin levels could contribute in part to weight loss (44). In contrast, cytokines, such as TNF- and IL-1?, have been shown to induce anorexia (for review, see Ref. 45). Also, obesity was proposed to be a low grade systemic inflammation disease characterized by elevated serum levels of C-reactive protein, IL-6, TNF-, and leptin (46).
In summary, ghrelin and GHS-R 1b mRNA were first detected by RT-PCR in erythroleukemic HEL cells. Octanoylated ghrelin was produced in extremely high quantities in undifferentiated and differentiated HEL cells, was secreted in the culture medium, and stimulated, as des-acyl ghrelin, HEL cell proliferation by an autocrine pathway involving an unidentified receptor distinct from GHS-R1a. Moreover, the HEL cell line represents a unique model to study the octanoylation of ghrelin.
Acknowledgments
We thank Profs. A. Delforge, A. Herchuelz, S. Meuris, M. Svoboda, and M. Waelbroeck for helpful discussion, and M. Stiévenart for secretarial assistance.
References
Kojima M, Hosoda H, Date Y, Nakazato M, Matsuo H, Kangawa K 1999 Ghrelin is a growth-hormone-releasing acylated peptide from stomach. Nature 402:656–660
Bednarek MA, Feighner SD, Pong SS, McKee KK, Hreniuk DL, Silva MV, Warren VA, Howard AD, Van Der Ploeg LH, Heck JV 2000 Structure-function studies on the new growth hormone-releasing peptide, ghrelin: minimal sequence of ghrelin necessary for activation of growth hormone secretagogue receptor 1a. J Med Chem 43:4370–4376
Hosoda H, Kojima M, Matsuo H, Kangawa K 2000 Purification and characterization of rat des-Gln14-ghrelin, a second endogenous ligand for the growth hormone secretagogue receptor. J Biol Chem 275:21995–22000
Hosoda H, Kojima M, Mizushima T, Shimizu S, Kangawa K 2003 Structural divergence of human ghrelin. Identification of multiple ghrelin-derived molecules produced by post-translational processing. J Biol Chem 278:64–70
Kojima M, Hosoda H, Matsuo H, Kangawa K 2001 Ghrelin: discovery of the natural endogenous ligand for the growth hormone secretagogue receptor. Trends Endocrinol Metab 12:118–122
Tsch?p M, Smiley DL, Heiman ML 2000 Ghrelin induces adiposity in rodents. Nature 407:908–913
Nakazato M, Murakami N, Date Y, Kojima M, Matsuo H, Kangawa K, Matsukura S 2001 A role for ghrelin in the central regulation of feeding. Nature 409:194–198
Howard AD, Feighner SD, Cully DF, Arena JP, Liberator PA, Rosenblum CI, Hamelin M, Hreniuk DL, Palyha OC, Anderson J, Paress PS, Diaz C, Chou M, Liu KK, McKee KK, Pong SS, Chaung LY, Elbrecht A, Dashkevicz M, Heavens R, Rigby M, Sirinathsinghji DJ, Dean DC, Melillo DG, Van der Ploeg LH, Patchett AA, Nargund R, Griffin PR, DeMartino JA, Gupton SK, Schaeffer JM, Smith RG 1996 A receptor in pituitary and hypothalamus that functions in growth hormone release. Science 273:974–977
Ariyasu H, Takaya K, Tagami T, Ogawa Y, Hosoda K, Akamizu T, Suda M, Koh T, Natsui K, Toyooka S, Shirakami G, Usui T, Shimatsu A, Doi K, Hosoda H, Kojima M, Kangawa K, Nakao K 2001 Stomach is a major source of circulating ghrelin, and feeding state determines plasma ghrelin-like immunoreactivity levels in humans. J Clin Endocrinol Metab 86:4753–4758
Hosoda H, Kojima M, Matsuo H, Kangawa K 2000 Ghrelin and des-acyl ghrelin: two major forms of rat ghrelin peptide in gastrointestinal tissue. Biochem Biophys Res Commun 279:909–913
Cassoni P, Ghe C, Marrocco T, Tarabra E, Allia E, Catapano F, Deghenghi R, Ghigo E, Papotti M, Muccioli G 2004 Expression of ghrelin and biological activity of specific receptors for ghrelin and des-acyl ghrelin in human prostate neoplasms and related cell lines. Eur J Endocrinol 150:173–184
Thompson NM, Gill DA, Davies R, Loveridge N, Houston PA, Robinson IC, Wells T 2004 Ghrelin and des-octanoyl ghrelin promote adipogenesis directly in vivo by a mechanism independent of the type 1a growth hormone secretagogue receptor. Endocrinology 145:234–242
Bedendi I, Alloatti G, Marcantoni A, Malan D, Catapano F, Ghe C, Deghenghi R, Ghigo E, Muccioli G 2003 Cardiac effects of ghrelin and its endogenous derivatives des-octanoyl ghrelin and des-Gln14-ghrelin. Eur J Pharmacol 476:87–95
Baldanzi G, Filigheddu N, Cutrupi S, Catapano F, Bonissoni S, Fubini A, Malan D, Baj G, Granata R, Broglio F, Papotti M, Surico N, Bussolino F, Isgaard J, Deghenghi R, Sinigaglia F, Prat M, Muccioli G, Ghigo E, Graziani A 2002 Ghrelin and des-acyl ghrelin inhibit cell death in cardiomyocytes and endothelial cells through ERK1/2 and PI 3-kinase/AKT. J Cell Biol 159:1029–1037
Gualillo O, Lago F, Gomez-Reino J, Casanueva FF, Dieguez C 2003 Ghrelin, a widespread hormone: insights into molecular and cellular regulation of its expression and mechanism of action. FEBS Lett 552:105–109
Korbonits M, Kojima M, Kangawa K, Grossman AB 2001 Presence of ghrelin in normal and adenomatous human pituitary. Endocrine 14:101–104
Papotti M, Cassoni P, Volante M, Deghenghi R, Muccioli G, Ghigo E 2001 Ghrelin-producing endocrine tumors of the stomach and intestine. J Clin Endocrinol Metab 86:5052–5059
Volante M, Allia E, Gugliotta P, Funaro A, Broglio F, Deghenghi R, Muccioli G, Ghigo E, Papotti M 2002 Expression of ghrelin and of the GH secretagogue receptor by pancreatic islet cells and related endocrine tumors. J Clin Endocrinol Metab 87:1300–1308
Jeffery PL, Herington AC, Chopin LK 2002 Expression and action of the growth hormone releasing peptide ghrelin and its receptor in prostate cancer cell lines. J Endocrinol 172:R7–R11
Kanamoto N, Akamizu T, Hosoda H, Hataya Y, Ariyasu H, Takaya K, Hosoda K, Saijo M, Moriyama K, Shimatsu A, Kojima M, Kangawa K, Nakao K 2001 Substantial production of ghrelin by a human medullary thyroid carcinoma cell line. J Clin Endocrinol Metab 86:4984–4990
Hattori N, Saito T, Yagyu T, Jiang BH, Kitagawa K, Inagaki C 2001 GH, GH receptor, GH secretagogue receptor, and ghrelin expression in human T cells, B cells, and neutrophils. J Clin Endocrinol Metab 86:4284–4291
Poncz M, Surrey S, LaRocco P, Weiss MJ, Rappaport EF, Conway TM, Schwartz E 1987 Cloning and characterization of platelet factor 4 cDNA derived from a human erythroleukemic cell line. Blood 69:219–223
Hong Y, Martin JF, Vainchenker W, Erusalimsky JD 1996 Inhibition of protein kinase C suppresses megakaryocytic differentiation and stimulates erythroid differentiation in HEL cells. Blood 87:123–131
Long MW, Heffner CH, Williams JL, Peters C, Prochownik EV 1990 Regulation of megakaryocyte phenotype in human erythroleukemia cells. J Clin Invest 85:1072–1084
Papayannopoulou T, Nakamoto B, Yokochi T, Chait A, Kannagi R 1983 Human erythroleukemia cell line (HEL) undergoes a drastic macrophage-like shift with TPA. Blood 62:832–845
Kruh J 1982 Effects of sodium butyrate, a new pharmacological agent, on cells in culture. Mol Cell Biochem 42:65–82
Gourlet P, Vandermeers A, Vertongen P, Rathe J, De Neef P, Cnudde J, Waelbroeck M, Robberecht P 1997 Development of high affinity selective VIP1 receptor agonists. Peptides 18:1539–1545
Fraker PJ, Speck Jr JC 1978 Protein and cell membrane iodinations with a sparingly soluble chloroamide, 1,3,4,6-tetrachloro-3,6-diphrenylglycoluril. Biochem Biophys Res Commun 80:849–857
Tomasetto C, Karam SM, Ribieras S, Masson R, Lefebvre O, Staub A, Alexander G, Chenard MP, Rio MC 2000 Identification and characterization of a novel gastric peptide hormone: the motilin-related peptide. Gastroenterology 119:395–405
Van Craenenbroeck M, Gregoire F, De Neef P, Robberecht P, Perret J 2004 Ala-scan of ghrelin (1–14): interaction with the recombinant human ghrelin receptor. Peptides 25:959–965
Ellman GL, Courtney KD, Andres Jr V, Feather-Stone RM 1961 A new and rapid colorimetric determination of acetylcholinesterase activity. Biochem Pharmacol 7:88–95
Yang ZP, Dettbarn WD 1998 Prevention of tolerance to the organophosphorus anticholinesterase paraoxon with carboxylesterase inhibitors. Biochem Pharmacol 55:1419–1426
Duysen EG, Li B, Xie W, Schopfer LM, Anderson RS, Broomfield CA, Lockridge O 2001 E vidence for nonacetylcholinesterase targets of organophosphorus nerve agent: supersensitivity of acetylcholinesterase knockout mouse to VX lethality. J Pharmacol Exp Ther 299:528–535
Bradford MM 1976 A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 72:248–254
Gnanapavan S, Kola B, Bustin SA, Morris DG, McGee P, Fairclough P, Bhattacharya S, Carpenter R, Grossman AB, Korbonits M 2002 The tissue distribution of the mRNA of ghrelin and subtypes of its receptor, GHS-R, in humans. J Clin Endocrinol Metab 87:2988
Cummings DE, Weigle DS, Frayo RS, Breen PA, Ma MK, Dellinger EP, Purnell JQ 2002 Plasma ghrelin levels after diet-induced weight loss or gastric bypass surgery. N Engl J Med 346:1623–1630
Bagnasco M, Kalra PS, Kalra SP 2002 Ghrelin and leptin pulse discharge in fed and fasted rats. Endocrinology 143:726–729
De Vriese C, Grégoire F, Lema-Kisoka R, Waelbroeck M, Robberecht P, Delporte C 2004 Ghrelin degradation by serum and tissue homogenates: identification of the cleavage sites. Endocrinology 145:4997–5005
Martin P, Papayannopoulou T 1982 HEL cells: a new human erythroleukemia cell line with spontaneous and induced globin expression. Science 216:1233–1235
Long MW, Heffner CH 1988 Detection of human megakaryocyte antigens by solid-phase radioimmunoassay. Exp Hematol 16:62–70
Dixit VD, Schaffer EM, Pyle RS, Collins GD, Sakthivel SK, Palaniappan R, Lillard JW Jr, Taub DD 2004 Ghrelin inhibits leptin- and activation-induced proinflammatory cytokine expression by human monocytes and T cells. J Clin Invest 114:57–66
Li WG, Gavrila D, Liu X, Wang L, Gunnlaugsson S, Stoll LL, McCormick ML, Sigmund CD, Tang C, Weintraub NL 2004 Ghrelin inhibits proinflammatory responses and nuclear factor-B activation in human endothelial cells. Circulation 109:2221–2226
Dembinski A, Warzecha Z, Ceranowicz P, Tomaszewska R, Stachura J, Konturek SJ, Konturek PC 2003 Ghrelin attenuates the development of acute pancreatitis in rat. J Physiol Pharmacol 54:561–573
Otero M, Nogueiras R, Lago F, Dieguez C, Gomez-Reino JJ, Gualillo O 2004 Chronic inflammation modulates ghrelin levels in humans and rats. Rheumatology 43:306–310
Stenvinkel P, Pecoits-Filho R, Lindholm B 2003 Leptin, ghrelin, and proinflammatory cytokines: compounds with nutritional impact in chronic kidney disease? Adv Ren Replace Ther 10:332–345
Das UN 2001 Is obesity an inflammatory condition? Nutrition 17:953–966(Carine De Vriese, Fran?oi)
Address all correspondence and requests for reprints to: Dr. Christine Delporte, Department of Biochemistry and Nutrition, Faculty of Medicine, Université Libre de Bruxelles, Bat G/E, CP 611, 808 route de Lennik, B-1070 Brussels, Belgium. E-mail: cdelport@ulb.ac.be.
Abstract
Ghrelin, a ligand of the GH secretagogue receptor (GHS-R 1a), is a 28-amino acid peptide with an unusual octanoyl group on Ser3, crucial for its biological activity. For the first time, ghrelin and GHS-R 1b, a truncated variant of the receptor resulting from alternative splicing, but not GHS-R 1a, mRNAs were detected in the human erythroleukemic cell line HEL. Two antibodies, used for RIA, were directed against octanoylated and total (octanoylated and desoctanoylated) ghrelin, and the recognized epitopes were characterized. Using reverse phase HPLC analysis followed by RIA, we demonstrated that octanoylated and desoctanoylated ghrelins were present in HEL cells and their culture medium, of which more than 90% was octanoylated. The ghrelin levels were not affected after 24 h treatment with sodium butyrate, phorbol 12-myristate 13-acetate, or forskolin, but a significant 3-fold increase in desoctanoylated ghrelin was detected in the culture medium after 48 h treatment with sodium butyrate. The antighrelin SB801 and SB969 antisera inhibited HEL cell proliferation by 24% and 39%, respectively, after 72 h. Taken together, these data suggested that endogenous ghrelin stimulated HEL cell proliferation by an autocrine pathway involving an unidentified receptor, distinct from GHS-R1a, and that the HEL cell line represents a unique model to study the octanoylation of ghrelin.
Introduction
GHRELIN IS A novel gastrointestinal hormone that was recently purified and identified from rat stomach (1). Ghrelin is a 28-amino acid peptide containing an n-octanoyl modification on the Ser3 residue, which is essential for its biological activity, and a C-terminal acid. Its C-terminal sequence can be shortened without loss of biological activity (2). Since its discovery, other forms of ghrelin were purified from the stomach: des-Gln14-ghrelin, resulting from an alternative splicing of rat ghrelin gene (3); and decanoyl ghrelin-(1–28), decenoyl ghrelin-(1–28), octanoyl ghrelin-(1–27), decanoyl ghrelin-(1–27), and des-acyl ghrelin-(1–27), produced by posttranslational processing of the ghrelin precursor in human stomach (4). Ghrelin stimulates GH release from the pituitary (5) and regulates food intake and energy metabolism (6, 7). These biological actions of ghrelin are mainly mediated by the GH secretagogue receptor (GHS-R) (8). Two GHS-R subtypes, generated by alternative splicing of a single gene, have been cloned: GHS-R type 1a (GHS-R 1a) and GHS-R type 1b (GHS-R 1b). The human GHS-R 1a consists of 366 amino acids with seven transmembrane domains. Its activation leads to inositol triphosphate generation and Ca2+ release through activation of the G protein G11. The GHS-R 1b consists of 289 amino acids with only the first five transmembrane domains, followed by a 24-amino acid sequence encoded by an alternatively spliced intronic sequence. GHS-R 1b fails to bind GHS or respond to GHS and is not known to exhibit any ghrelin-mediated biological activity.
Circulating ghrelin is mainly derived from the stomach and is influenced by the feeding state (9). The most abundant circulating form of ghrelin is des-acyl ghrelin (10), a form of ghrelin that does not act on the GHS-R 1a. Only acylated forms of ghrelin bind to the GHS-R 1a and exert endocrine actions. Recently, however, des-acyl ghrelin was shown to modulate cell proliferation in prostate carcinoma cell lines (11), to stimulate adipogenesis (12), to induce cardiovascular effects (13), and to inhibit apoptosis in cardiomyocytes and endothelial cells (14). These effects could be mediated by an as yet unidentified ghrelin receptor.
In addition to the stomach, ghrelin has been detected in several tissues, such as intestine, kidney, pituitary, pancreas, placenta, lung, testis, and ovary (15); in tumors, such as pituitary adenomas (16), gastrointestinal carcinoids (17), and endocrine pancreatic tumors (18); and in cell lines, such as prostate neoplasms (11, 19) and medullary thyroid carcinoma (20). Ghrelin mRNA was also found in T cells, B cells, and neutrophils as well as in leukemic B, T, and myeloid cell lines (21).
The human erythroleukemic cell line HEL is representative of the erythroblastic stage of differentiation of hemopoietic cells, retains a number of features of the megakaryocyte/platelet lineage (22), and can differentiate into megakaryocytes, macrophages, or erythrocytes (23, 24, 25, 26).
The aims of the present study were to examine, in the human erythroleukemic HEL cell line, the presence of ghrelin and GHS-R subtypes, ghrelin expression during cellular differentiation, and the effects of ghrelin on cell proliferation. To achieve this, we obtained and characterized two antibodies directed against human ghrelin to set up specific RIAs and developed extraction and chromatographic procedures to separate ghrelin from des-acyl ghrelin.
Materials and Methods
Peptide synthesis
All peptides used were synthesized by solid phase methodology using the 9-fluorenyl-methoxy-carbonyl (fmoc) strategy (27) with a Symphony Multiplex apparatus (Protein Technologies, Inc., Tucson, AZ) and were purified by reverse phase HPLC (RP-HPLC) on Amberchrome-type resin 78 CG-162sd (15.9 x 2.5 cm), Vydac 259HP510A (25 x 1 cm; Alltech, Laarne, Belgium) and Vydac 259VHP54 (25 x 0.46 cm; Alltech). The Ser3 hydroxyl group of ghrelin was acylated with n-octanoic acid using 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide in the presence of 4-(dimethylamino)pyridine according to the method of Bednarek et al. (2). Selective protection of Ser3 was obtained using fmoc Ser-Trt, instead of fmoc Ser-Tbu for Ser2, to allow its selective deprotection and using a solution of 1% trifluoroacetic acid (TFA) in dichloromethane before subsequent octanoylation. All ghrelin analogs were based on the human sequence, unless specified otherwise, and those possessing a C-terminal acidic function instead of an amide were noted by -OH. Peptide purity (>95%) was assessed by capillary electrophoresis (P/ACE MDQ, Beckman Coulter, Fullerton, CA), and molecular weight was verified by electrospray mass spectrometry using a VG Platform ns 8230E (Waters Corp., Milford, MA).
Preparation of antighrelin sera
[Cys12]Ghr-(1–11) (1.5 mg) and [Cys0]Ghr-(13–28)-OH (1.5 mg) were conjugated separately to keyhole limpet hemocyanin using m-maleimidobenzoyl-N-hydroxysuccinimide ester (Eurogentec, Herstal, Belgium). Each conjugate was emulsified with Freund’s adjuvant. Two batches of antisera were obtained from New Zealand White rabbits after four monthly intradermic injections of conjugate (Eurogentec, Seraing, Belgium).
Peptide radioiodination
[Tyr24]Ghr-(1–23) and [Tyr0]Ghr-(13–28)-OH were radioiodinated on tyrosine by the Iodogen method (28) and were purified on a Sep-Pak C18 cartridge (Waters Corp.).
RIA for ghrelin
Assays were performed in duplicate at 4 C in a 300-μl final volume containing 100 μl standard human Ghr-(1–28)-OH or unknown sample, 100 μl antiserum (diluted at 1/10,000) containing 0.1% normal rabbit serum, and 100 μl 125I-labeled tracer (10,000 cpm) diluted in RIA buffer [10 mM sodium phosphate buffer (pH 7.4), 0.05% Tween 20, 150 mM NaCl, 10 mM EDTA, and 0.1% sodium azide]. The antigen and antibody were preincubated overnight, followed by the tracer addition. After an additional 18-h incubation, 1 ml 4% polyethylene glycol 6000 diluted in RIA buffer containing 1% sheep antirabbit serum was added. After a 30-min incubation at room temperature, free and bound tracers were separated by centrifugation for 20 min at 2600 x g. After aspiration of the supernatant, the radioactive pellet was counted into a -counter. Nonspecific binding was determined in the absence of antiserum. For SB801 and SB969 antibodies, the limits of detection for ghrelin were 3 and 7 fmol/assay; the intraassay coefficients of variation were 3% and 1%; and the peptide recoveries were 97% and 98%, respectively.
Ghrelin (and ghrelin receptors) mRNA detection by RT-PCR
HEL total RNA was extracted using the SV RNA extraction kit (Promega Corp., Leiden, The Netherlands), which includes a deoxyribonuclease treatment. cDNA synthesis was performed using 1 μg total RNA with the Expand Reverse Transcriptase (Roche, Brussels, Belgium). The resulting cDNA was subjected to PCR amplification with 0.4 μM of the sense and antisense primers and 0.25 U Goldstar DNA polymerase (Eurogentec, Seraing, Belgium). The PCR primers used were: human ghrelin: sense, 5'-AAGGAGTCGAAGAAGCCACCA-3' (nucleotides 148–168); and antisense, 5'-GCCAGATGAGCGCTTCTAAACTTA-3' (nucleotides 416–439 in accession no. AB029434, GenBank); human GHS-R 1a and GHS-R 1b: sense, 5'-TCTTCCTTCCTGTCTTCTATC-3' (nucleotides 662–682 in accession no. U60179 and U60181, GenBank); human GHS-R 1a: antisense, 5'-AGTCTGAACACTGCCACC-3' (nucleotides 993-1010 in accession no. U60179, GenBank); human GHS-R 1b: antisense, 5'-TCAGAGAGAAGGGAGAAGG-3' (nucleotides 852–870 in accession no. U60181, GenBank); and human ?-actin: sense, 5'-TGACGGGGTCACCCACACTGTGCCCGTC-3' (nucleotides 539–566); and antisense, 5'-CTAGAAGCATTAGCGGTGGACGATGGAGG-3' (nucleotides 1171–1199; in accession no. BC002409, GenBank). Amplification of ?-actin served as a quality control for the RNA. For ghrelin cDNA amplification, 35 cycles were performed (10 sec at 94 C, 10 sec at 57 C, and 1 min at 72 C). For ghrelin receptor cDNA amplification, 40 cycles were performed (10 sec at 94 C, 20 sec at 50 and 58 C for, respectively, GHS-R 1a and GHS-R 1b, and 45 sec at 72 C). For ?-actin amplification, 25 cycles were performed (1 min at 94 C, 1 min at 60 C, and 1 min at 72 C). Five microliters of each PCR product were submitted to electrophoresis on a 1.4% agarose gel stained with ethidium bromide and visualized under UV light. For ghrelin, GHS-R 1a, and GHS-R 1b PCRs, positive controls included, respectively, a pSG5 vector containing human ghrelin cDNA (29), a pEGFP-N1 vector containing human GHS-R 1a cDNA (30), and human placenta cDNA.
Cell culture, cell treatment, and cell lysate preparation
HEL cells were grown in RPMI 1640 medium (BioWhittaker Europe, Verviers, Belgium) supplemented with 10% inactivated fetal bovine serum, 100 IU/ml streptomycin-penicillin, and 4 mM glutamine and were routinely passaged twice a week. In some experiments HEL cells were treated, or not, for 24 or 48 h with 33 nM phorbol 12-myristate 13-acetate (PMA), 10 μM forskolin, or 1 mM sodium butyrate, then submitted to peptide extraction and RP-HPLC separation, followed by RIA. HEL cell lysate was prepared as follows. HEL cells were sonicated in distilled water for 30 sec and centrifuged at 4 C at 1000 x g for 10 min. The supernatant was collected and used for determination of butyrylcholinesterase and carboxylesterase activities.
Peptide extraction and RP-HPLC separation
HEL cells were centrifuged at 500 x g for 10 min. The medium was collected and acidified with HCl to pH 4–5. The cell pellet was resuspended into 10 mM HCl, frozen in liquid nitrogen, heated at 100 C for 5 min, cooled, and centrifuged at 20,000 x g for 30 min at 4 C. The resulting supernatant was collected. The acidified medium and supernatant were loaded onto a Sep-Pak C18 column preequilibrated with 3% CH3CN/0.1% TFA. After washing with 10% CH3CN/0.1% TFA, the peptides were eluted with 60% CH3CN/0.1% TFA. The eluates were lyophilized in a Speed-Vac concentrator, subjected to RP-HPLC analysis on a C18 Vydac 218TP54 column (25 x 0.46 cm; Alltech), and equilibrated with 3% CH3CN/0.1% TFA, using a linear gradient of CH3CN from 3–80% in 0.1% TFA for 50 min. The OD was monitored at 226 nm. RP-HPLC fractions were collected, lyophilized, and submitted to RIA.
Degradation of synthetic ghrelin exogenously added to culture medium in the presence of HEL cells
One microgram of synthetic ghrelin was added to 10 ml culture medium in the presence of 2.5 x 106 HEL cells. The culture media were collected at various times, then submitted to peptide extraction and RP-HPLC separation, followed by RIA.
Assay for butyrylcholinesterase activity
Butyrylcholinesterase activity was measured by the method of Ellman et al. (31), using an LKB Ultrospec Plus 4054 UV/visible spectrophotometer (LKB, Bromma, Sweden). One hundred microliters of HEL cell lysate or inactivated fetal bovine serum (FBS) were added to 100 mM butyrylthiocholine iodide and 0.25 mM 5',5'-dithiobis-2-nitrobenzoic acid in 50 mM Tris-HCl, pH 7.4. The absorbance was read at 405 nm every 30 sec for up to 6 min. The enzyme activity was calculated as micromoles of the product per minute (as unit, U) after correction for nonenzymatic hydrolysis of the substrate using the extinction coefficient (13,300 M–1 · cm–1) of the product.
Assay for carboxylesterase activity
Carboxylesterase activity was determined by measuring the hydrolysis of -naphtylacetate (32, 33). One hundred microliters of HEL cell lysate or inactivated FBS were preincubated at 37 C for 20 min with 10 μM eserine to inhibit acetyl- and butyrylcholinesterases and with 10 mM EDTA to inhibit paraoxonase, then 10 μl 0.02 M -naphtylacetate were added in a 100 mM phosphate buffer, pH 7.0. The absorbance was measured at 321 nm every 10 min for up to 60 min. Enzyme activity was calculated as micromoles of the product per minute using the extinction coefficient (2200 M–1 · cm–1) of the product.
HEL cell proliferation studies
HEL cells were grown in six-well plates (5 x 105 cells/ml) in RPMI 1640 medium (BioWhittaker Europe) supplemented with 2.5% inactivated fetal bovine serum, 100 IU/ml streptomycin-penicillin, and 4 mM glutamine and were incubated at 37 C for up to 72 h with or without 1 μM octanoylated ghrelin, 1 μM des-acyl ghrelin, 1% SB801, 1% SB969, or 1% of both SB801 and SB969. Appropriate controls were performed for each condition. The addition of 1% or 2% preimmune rabbit serum did not significantly modify HEL cell proliferation. After 0-, 24-, 48-, and 72-h incubation, cells were counted in triplicate in a Multisizer III (Coulter Electronics Ltd., Luton, UK).
Protein assay
The protein concentration was determined using Bradford’s method (34).
Data analysis
Data are summarized as the mean ± SEM. Results were statistically analyzed using the paired t test or one-way ANOVA, followed by Tukey-Kramer multiple comparisons test. All statistical values reported were obtained using GraphPad InStat version 3.02 for Windows (GraphPad, Inc., San Diego, CA). P < 0.05 was considered significant.
Results
Characterization of the ghrelin epitope recognized by SB801 antiserum
SB801 antiserum was obtained after rabbit immunization using a synthetic human ghrelin peptide corresponding to the N-terminal 11 amino acids [Ghr-(1–11)], including the n-octanoylated Ser3 (Fig. 1). The sequence of this peptide differs from its rat homologous peptide by the presence of Arg11, instead of Lys11.
FIG. 1. Polyclonal antibodies raised against human ghrelin. SB801 and SB969 antisera were obtained by immunization with the synthetic human peptides Ghr-(1–11) and Ghr-(13–28), respectively, as described in Materials and Methods. Epitopes recognized by the antisera are delimited by the boxes.
SB801 antiserum recognized with similar high affinity human and rat Ghr-(1–28)-OH, suggesting that the amino acid in position 11 was not important for recognition of the peptide by the antiserum. It also recognized Ghr-(1–28)-OH, rat Ghr-(1–28)-OH, Ghr-(1–23), Ghr-(1–14), Ghr-(1–7)-OH, and Ghr-(1–5) with similar high affinities, whereas des-acyl ghrelin analogs and Ghr-(3–23) were not recognized (Figs. 2 and 3 and Table 1). Compared with Ghr-(1–28)-OH, decanoyl Ghr-(1–28)-OH had a 2.8-fold lower affinity for SB801 antiserum, suggesting that octanoyl was the optimal acyl group for antibody recognition (Fig. 2 and Table 1).
FIG. 2. Competition curves of ghrelin analogs for the binding of [125I]Tyr24-Ghr-(1–23) to SB801 antiserum. RIA was performed as described in Materials and Methods. The experiments were performed in duplicate, and the results are expressed as the percentage of specific binding (B) in the absence of peptides (B0; mean ± SEM; n = 3). rGhr, Rat Ghr.
FIG. 3. Competition curves of Ala-scan analogs of ghrelin-(1–14) for the binding of [125I]Tyr24-Ghr-(1–23) to SB801 antiserum. RIA was performed as described in Materials and Methods. The experiments were performed in duplicate, and the results are expressed as a percentage of specific binding (B) in the absence of peptides (B0; mean ± SEM; n = 3).
TABLE 1. Ki values for several ghrelin analogs on [125I]Tyr24-Ghr-(1–23) binding to SB801 antiserum
Because Ghr-(1–14) was recognized with similar affinity as Ghr-(1–28)-OH, each of its amino acids was systematically substituted by Ala to determine the amino acids involved in recognition by SB801 antiserum. The SB801 antiserum did not discriminate [Ala1]-, [Ala6]-, [Ala7]-, [Ala8]-, [Ala9]-, [Ala10]-, [Ala11]-, [Ala12]-, [Ala13]-, or [Ala14]-Ghr-(1–14) from Ghr-(1–14) (Fig. 3 and Table 2), whereas [Ala2]-, [Ala4]-, and [Ala5]-Ghr-(1–14) were recognized with 8-, 66-, and 1.7-fold lower affinities, and [Ala3]-Ghr-(1–14) was not recognized (Fig. 3A and Table 2). These results suggested that the ghrelin epitope recognized by SB801 antiserum includes positions 2–5.
TABLE 2. Ki values for Ala scans ghrelin analogs on [125I]Tyr24-Ghr-(1–23) binding to SB801 antiserum
Characterization of the ghrelin epitope recognized by SB969 antiserum
SB969 antiserum was obtained after rabbit immunization using a synthetic human ghrelin peptide corresponding to the C-terminal 16 amino acids of human ghrelin [Ghr-(13–28); Fig. 1]. The sequence of this peptide is identical in both human and rat ghrelins.
SB969 antiserum recognized Ghr-(1–28)-OH, des-acyl Ghr-(1–28)-OH, decanoyl Ghr-(1–28)-OH, Ghr-(1–23), and des-acyl Ghr-(1–23) with Ki values of 303, 180, 194, 647, and 877 pM, respectively (Fig. 4A and Table 3). N-Terminal shortening of Ghr-(1–23) [Ghr-(3–23), Ghr-(4–23), Ghr-(5–23), and Ghr-(8–23)] did not affect antibody recognition (Fig. 4A and Table 3). N-Terminal-shortened analogs of Ghr-(1–28) ([Tyr0]Ghr-(13–28)-OH, [Cys0]Ghr-(13–28)-OH, and Ghr-(13–28)-OH) had somewhat higher affinities than Ghr-(1–28) for the SB969 antiserum, suggesting that the N-terminal ghrelin sequence impaired recognition of the C terminus of the antibody (Fig. 4B and Table 3). Amidated [Tyr0]Ghr-(13–28), [Cys0]Ghr-(13–28), as well as C-terminal truncated Ghr-(13–28)-OH [Ghr-(13–27)-OH, Ghr-(13–26)-OH, Ghr-(13–25)-OH, and Ghr-(13–24)-OH] increased nondisplaceable binding and decreased the peptide affinity for the antibody (Fig. 4B and Table 3). Taken together, these results indicated that the antibody recognized at least the five C-terminal amino acids of ghrelin in addition to its C-terminal acid.
FIG. 4. Competition curves of ghrelin analogs for the binding of [125I]Tyr0-Ghr-(13–28)-OH to SB969 antiserum. RIA was performed as described in Materials and Methods. The experiments were performed in duplicate, and the results are expressed as the percentage of specific binding (B) in the absence of peptides (B0; mean ± SEM; n = 3).
TABLE 3. Ki values for several ghrelin analogs on [125I]Tyr0-Ghr-(13–28)-OH binding to SB969 antiserum
GHS-R subtypes and ghrelin mRNA detection by RT-PCR in HEL cells
RT-PCR products of the size expected for GHS-R 1b and ghrelin (209 or 292 bp, respectively), were obtained from HEL cells (Fig. 5). Direct sequencing of the PCR products confirmed that they corresponded to GHS-R 1b and ghrelin (data not shown). In contrast, no RT-PCR product of the size expected for GHS-R 1a (349 bp) could be obtained from HEL cells (Fig. 5).
FIG. 5. GHS-R subtypes and ghrelin mRNA expression in HEL cells. RNA extraction, RT, and PCR were performed as described in Materials and Methods. Positive controls were performed using pEGFP-N1 vector containing human GHS-R 1a cDNA diluted at 100 pg/μl, human placenta cDNA, or PSG5 vector containing human ghrelin diluted at 10 pg/μl. Negative controls were performed by omitting cDNA in the PCR mixture.
Ghrelin identification in HEL cells by HPLC followed by RIA
To identify ghrelin, HEL cells and their culture medium were subjected to peptide extraction and subsequent RP-HPLC analysis, followed by RIA using both SB801 and SB969 antisera. An immunoreactive (IR) ghrelin peak, detected in RIA using SB801 antiserum, eluted at a position identical to that of human octanoylated ghrelin (Fig. 6, A and B). Two IR peaks, detected by RIA using SB969 antiserum, eluted at positions identical to human octanoylated ghrelin and human desoctanoylated ghrelin (Fig. 6, C and D).
FIG. 6. Identification of IR-ghrelin in HEL cells and their culture medium. After 24 h of culture, peptide extracts from HEL cells (A and C) and their culture medium (B and D) were submitted to RP-HPLC on a C18 Vydac 218TP54 column using a linear gradient of 3–80% CH3CN containing 0.1% TFA for 50 min (see Materials and Methods). HPLC fractions of 1 ml were lyophilized, then submitted to RIA using SB801 (A and B) or SB969 (C and D). IR-ghrelin was expressed as femtomoles per fraction. Arrows indicate the elution positions of human desoctanoylated (1 ) and octanoylated (2 ) ghrelins.
Using SB801 antiserum, levels of IR-ghrelin eluting at the octanoylated ghrelin position were 543 ± 17 fmol/106 cells (1784 ± 56 fmol/mg protein; n = 3) and 97 ± 32 fmol/106 cells (320 ± 106 fmol/mg protein; n = 3) in HEL cells and their culture medium after 24 h of culture, respectively. Using SB969 antiserum, IR-ghrelin levels eluting at desoctanoylated ghrelin position were 39 ± 18 fmol/106 cells (128 ± 57 fmol/mg protein; n = 3) and 11 ± 4 fmol/106 cells (36 ± 13 fmol/mg protein; n = 3) in HEL cells and their culture medium after 24 h of culture, respectively.
To determine whether inactivated FBS could contribute to IR-ghrelin found in the culture medium of HEL cells after 24 h, IR-ghrelin was measured in inactivated FBS. IR-ghrelin levels assayed with the SB801 and SB969 antisera were 25 and 115 fmol/ml inactivated FBS, respectively.
Degradation of synthetic ghrelin exogenously added to culture medium in the presence of HEL cells
To determine whether secreted ghrelin was degraded, synthetic ghrelin (1 μg) was exogenously added to culture medium in the presence of HEL cells. After 1- and 24-h incubation, IR-ghrelin assayed with the SB801 antiserum represented 46 ± 17% and 0.79 ± 0.13%, respectively (n = 3), of ghrelin at time zero.
Determination of butyrylcholinesterase and carboxylesterase activities in inactivated FBS and HEL cell lysate
To test whether butyrylcholinesterase and carboxylesterase activities from inactivated FBS and/or HEL cells participated in ghrelin degradation, both esterase activities were measured in inactivated FBS and in HEL cell lysate. Butyrylcholinesterase activity, calculated for one culture flask of HEL cells containing 50 ml medium, was 0.320 and 0.054 U for inactivated FBS and HEL cells, respectively. Carboxylesterase activity, calculated for one culture flask of HEL cells containing 50 ml medium, was 0.024 and 0.057 U for inactivated FBS and HEL cells, respectively.
HEL cell treatments with differentiating agents
HEL cells were untreated or treated for 24 or 48 h with 33 nM PMA, 1 mM sodium butyrate, or 10 μM forskolin, then submitted to peptide extraction and RP-HPLC separation, followed by RIA. After 24 h of treatment, octanoylated and desoctanoylated ghrelins levels were not modified. After 48 h of treatment, octanoylated and desoctanoylated ghrelins levels were not modified by PMA and forskolin, whereas sodium butyrate significantly increased desoctanoylated ghrelin in culture medium by 346 ± 29% (n = 3; P < 0.05).
Determination of butyrylcholinesterase and carboxylesterase activities in HEL cell lysate and HEL cell medium after treatment with sodium butyrate
To test whether butyrylcholinesterase and carboxylesterase activities were modified after 48 h treatment with 1 mM sodium butyrate, both esterase activities were measured in HEL cell lysate and HEL cell medium. In both preparations, butyrylcholinesterase activity was not different in control conditions or after sodium butyrate treatment. In HEL cell lysate, carboxylesterase activity was 1.7-fold higher in HEL cells treated with sodium butyrate. In HEL cell medium, carboxylesterase activity was similar in control and sodium butyrate-treated HEL cells.
Effects of sodium butyrate on purified butyrylcholinesterase and carboxylesterase enzymes
To test whether butyrylcholinesterase and carboxylesterase activities were modified by a direct effect of sodium butyrate, both enzyme activities were measured in the absence or presence of 1 mM sodium butyrate. Butyrylcholinesterase and carboxylesterase activities were not affected in the presence of sodium butyrate.
HEL cell proliferation studies
To explore the effect of ghrelin on HEL cell proliferation, HEL cells were incubated with 2.5% FBS for up to 72 h with or without 1 μM octanoylated ghrelin, 1 μM des-acyl ghrelin, 1% SB801, 1% SB969, or 1% of both SB801 and SB969. Appropriate controls were performed for each condition. The addition of 1% or 2% preimmune rabbit serum did not significantly modify HEL cell proliferation. The ability of the SB801 and SB969 antisera to inhibit the biological activity of ghrelin was verified by performing a dose-effect curve of ghrelin on the intracellular calcium increase in Chinese hamster ovary cells coexpressing the recombinant GHS-R1a and aequorin (30). Under these conditions, SB801 and SB969 antisera increased by 78- and 15-fold, respectively, the 50% effective concentration of ghrelin on intracellular calcium increase.
Octanoylated and des-acyl ghrelin had no significant effect on HEL cell proliferation (data not shown). After 48 h treatment, SB969 significantly decreased HEL cell proliferation by 24%. After 72 h treatment, SB801 and SB969 decreased HEL cell proliferation by 24% and 39%, respectively. HEL cell proliferation measured after 48 and 72 h in the simultaneous presence of both antibodies was not statistically different from that measured in the presence of SB969 alone and was statistically different from that measured in the presence of SB801 alone (Fig. 7).
FIG. 7. HEL cell proliferation. HEL cells (5 105 cells/ml) grown in RPMI 1640 medium supplemented with 2.5% inactivated FBS, 100 IU/ml streptomycin-penicillin, and 4 mM glutamine were incubated for up to 72 h with 1% or 2% preimmune rabbit serum, 1% SB801, 1% SB969, or 1% of both SB801 and SB969 and counted in triplicate every 24 h as described in Materials and Methods. The results of four separate experiments performed in triplicate are expressed as the percentage of HEL cell number at time zero (mean ± SEM). Data significantly different compared with control: *, P < 0.001; compared with SB801: #, P < 0.05.
Discussion
Expression of the GHS-R 1b mRNA is widespread, whereas GHS-R 1a receptor expression is more restricted. It is predominantly expressed in the pituitary and (at much lower levels) in the thyroid, pancreas, spleen, myocardium, and adrenal gland (35). Ghrelin is expressed in many tissues and cell lines (10, 11, 15, 16, 17, 18, 19, 20). For blood cells, a previous study reported the presence of both ghrelin and ghrelin receptors mRNAs in human T cell, B cell, and myeloid leukemic cell lines as well as in normal T cells, B cells, and neutrophils with large individual variation (21). There is evidence indicating that several cancer cells express the components of the ghrelin/GHS-R axis and that the axis may have an important autocrine/paracrine role in regulating cancer cell proliferation.
In the present study ghrelin and GHS-R 1b mRNAs were first detected by RT-PCR in erythroleukemic HEL cells. In contrast, mRNA encoding the functional GHS-R 1a receptor, specific binding of [125I]Y24-ghrelin-(1–23), or an increase in free intracellular Ca2+ concentration in response to ghrelin (data not shown) was not observed in HEL cells.
To detect ghrelin synthesis, we developed two RIAs using SB801 and SB969 antisera. SB801 antiserum is probably capable of recognizing all biologically active forms of ghrelin, because C-terminal-shortened acyl ghrelins [down to Ghr-(1–5)] have been shown to retain their full biological activity on the GHS-R 1a, without an important change in their potency (2), and SB969 antiserum allows measurements of ghrelin, des-acyl ghrelin, as well as N- and, to some extent, C-terminal-shortened analogs.
Many studies have evaluated plasma, serum, or tissue ghrelin levels in human and animal models and attempted to correlate the results with several parameters, such as diet-induced weight loss or fasting (36, 37). The specificity of our antisera implies that SB801 does not measure solely acylated Ghr-(1–28), and that SB969 probably underestimates the quantity of total biologically active ghrelin if C-terminally truncated ghrelin fragments exist. Therefore, it is inappropriate to estimate the nonacylated ghrelin concentrations by merely subtracting the octanoylated ghrelin concentration from the total ghrelin concentration. Besides, as previously reported, great care should be taken when assimilating N-terminal IR-ghrelin to octanoylated ghrelin, and C-terminal IR-ghrelin to total ghrelin, because the specificities of the available antibodies have not been studied in detail.
Using RP-HPLC analysis followed by RIA, we demonstrated for the first time that HEL cells produced an important amount of octanoylated ghrelin (543 ± 17 fmol/106 cells = 1784 ± 56 fmol/mg protein; n = 3), a level only 2-fold lower than that in rat stomach (377 fmol/mg wet tissue = 3770 fmol/mg protein) (10) and 44-fold higher than that in medullary thyroid carcinoma TT cells (8.9 fmol/106 cells = 40.5 fmol/mg protein) (20). Over a 24-h period, HEL cells secreted octanoylated ghrelin into their culture medium (18% the amount of ghrelin present in the cells). The proportion of octanoylated ghrelin represented about 90% of the total ghrelin in HEL cells and their culture medium, compared with 17% in rat stomach (10). To our knowledge, desoctanoylated ghrelin has always been reported to be by far the major form of ghrelin present in tissues and cells (10, 20). Because of its capacity to synthesize high quantities of octanoylated ghrelin, the HEL cell line represents a unique model to study the octanoylation of ghrelin.
We recently showed that octanoylated ghrelin was degraded by butyrylcholinesterase in human serum and carboxylesterase in rat serum (38). A rapid and important degradation of ghrelin exogenously added to the culture medium (46% after 1 h) was observed in the presence of HEL cells by RIA using the SB801 antiserum. This degradation could result from the action of esterases from both HEL cells and FBS. Indeed, both butyrylcholinesterase and carboxylesterase activities were detected in inactivated FBS and HEL cell lysate. Besides, ghrelin was a better substrate for carboxylesterase than for butyrylcholinesterase. Indeed, in 1 h, 0.1 U purified carboxylesterase degraded 50% of ghrelin, whereas 2.5 U purified butyrylcholinesterase degraded only 20% of ghrelin (our unpublished observations). The butyrylcholinesterase activity in FBS and in HEL cell lysate represented 0.320 and 0.054 U/flask of HEL cells, respectively. This activity accounts for only 2.6% and 0.4% of ghrelin degradation after 1 h in the culture medium and is negligible compared with the 50% ghrelin degradation observed. In contrast, per flask of HEL cells, carboxylesterase activity in FBS and in HEL cell lysate represented 0.024 and 0.057 U, respectively. This is sufficient to account for the degradation of 12.0% and 28.5% of the added ghrelin after 1 h. Therefore, the carboxylesterase activity of HEL cells is probably responsible for most of the ghrelin degradation observed after 1 h. These data imply that we underestimated the octanoylated ghrelin secretion in the culture medium. When considering the possible proteolysis and the rapid degradation of added ghrelin in the culture medium, the amounts of octanoylated and desoctanoylated ghrelins in FBS would be negligible compared with those measured in the HEL culture medium after 24 h. The unexpected low amount of immunoreactive desoctanoylated ghrelin in the culture medium (10% of the amount of total ghrelin) could result from a C-terminal proteolysis of ghrelin and/or desoctanoylated ghrelin. Indeed, if ghrelin degradation occurs at the C-terminal end of ghrelin, the resulting fragments will be recognized by SB801 [if acylated, down to Ghr-(1–5)], but not, or poorly, by SB969 antiserum. This hypothesis is supported by the lower IR-ghrelin, corresponding to the elution position of octanoylated ghrelin (see peak 2 of the RP-HPLC analysis of HEL cell and medium; Fig. 6), detected by SB969 antiserum compared with that detected by SB801 antiserum. This suggests that some C-terminal-shortened ghrelin forms might indeed coelute with Ghr-(1–28) in our RP-HPLC system.
HEL cells are a poorly differentiated triphenotypic cell line constitutively expressing an erythroid phenotype, but also expressing antigens of other lineages (25, 39, 40). HEL cells increase their erythroid phenotype after stimulation with agents such as sodium butyrate (23, 24, 26) and their macrophage phenotype or megakaryocyte/platelet phenotypes after stimulation with phorbol esters or other agents (23, 24, 25). We did not observe any modification of ghrelin or desoctanoylated ghrelins levels after 24 h treatment with sodium butyrate, PMA, or forskolin. We only observed a significant 3-fold increase in desoctanoylated ghrelin into the culture medium after 48 h treatment with sodium butyrate. Therefore, it is unlikely that ghrelin can be used as a differentiation marker or plays a crucial role during the differentiation processes. The increase in desoctanoylated ghrelin observed after 48 h treatment with sodium butyrate could be linked to a concomitant increase in carboxylesterase production by HEL cells. Indeed, a 1.7-fold increase in carboxylesterase activity, but no increase in butyrylcholinesterase activity, was detected in treated HEL cell lysate, suggesting that only carboxylesterase could participate in the desoctanoylation of ghrelin in the culture medium. Because carboxylesterase isoforms possess distinct cellular distribution (intracellular, secreted into the medium, or bound to the extracellular surface of the membrane), their presence at the cell surface could explain the increase in desoctanoylated ghrelin after 48 h HEL cell treatment with sodium butyrate.
Hattori et al. (21) recently showed that T and B lymphocytes as well as neutrophils express ghrelin and GHS-R mRNA transcripts, but did not investigate the presence of the peptide or the receptor protein. Dixit et al. (41) recently demonstrated that ghrelin is endogenously produced and secreted by human T lymphocytes and monocytes, but without distinguishing octanoylated and desoctanoylated ghrelins. In humans, ghrelin levels declined only by 35–50% for 1 wk postgastrectomy and increased thereafter, suggesting that other tissues participate in ghrelin production (4). Those data and ours showing the production of ghrelin by undifferentiated and differentiated (into megakaryocytes, macrophages, or erythrocytes) HEL cells strongly suggest that several blood cell types express and produce ghrelin.
It is interesting to observe that HEL cells expressed the ghrelin mRNA transcript, contained extremely high quantities of octanoylated ghrelin, and secreted this peptide. Because ghrelin is known to modulate cell proliferation in several cell types (11, 12, 19), HEL cell proliferation was evaluated by adding exogenous ghrelin, des-acyl ghrelin, or SB801 and SB969 antisera used as ghrelin antagonists. HEL cell proliferation was inhibited at 72 h by the anti-ghrelin SB801 and SB969 antisera, with a greater effect of SB969 compared with SB801, and was unaffected by exogenous ghrelin and des-acyl ghrelin, suggesting that endogenous octanoylated and des-acyl ghrelins are sufficient to stimulate HEL cell proliferation. This autocrine effect could involve a specific ghrelin receptor distinct from the GHS-R1a receptor; evidence for its absence in HEL cells has been demonstrated. Besides this autocrine proliferative effect, ghrelin might also be involved in cytokine and/or chemokine regulation in pathological conditions associated with inflammation. Indeed, ghrelin was recently shown to inhibit TNF--induced IL-8 and monocyte chemoattractant protein-1 secretion in human endothelial cells and in a rat model of endotoxic shock as well as mononuclear cell adhesion (42) and exerts inhibitory effects on the expression and production of the inflammatory cytokines IL-1?, IL-6, and TNF- by human T cells and monocytes upon cellular activation and leptin exposure (41). Also, ghrelin was shown to attenuate the development of acute pancreatitis in rats by reducing inflammatory infiltrates of pancreatic tissues, vacuolization of acinar cells, plasma lipase activity, and IL-1? concentration (43). In experimental arthritis in rats and in rheumatoid arthritis in humans, decreased ghrelin levels could contribute in part to weight loss (44). In contrast, cytokines, such as TNF- and IL-1?, have been shown to induce anorexia (for review, see Ref. 45). Also, obesity was proposed to be a low grade systemic inflammation disease characterized by elevated serum levels of C-reactive protein, IL-6, TNF-, and leptin (46).
In summary, ghrelin and GHS-R 1b mRNA were first detected by RT-PCR in erythroleukemic HEL cells. Octanoylated ghrelin was produced in extremely high quantities in undifferentiated and differentiated HEL cells, was secreted in the culture medium, and stimulated, as des-acyl ghrelin, HEL cell proliferation by an autocrine pathway involving an unidentified receptor distinct from GHS-R1a. Moreover, the HEL cell line represents a unique model to study the octanoylation of ghrelin.
Acknowledgments
We thank Profs. A. Delforge, A. Herchuelz, S. Meuris, M. Svoboda, and M. Waelbroeck for helpful discussion, and M. Stiévenart for secretarial assistance.
References
Kojima M, Hosoda H, Date Y, Nakazato M, Matsuo H, Kangawa K 1999 Ghrelin is a growth-hormone-releasing acylated peptide from stomach. Nature 402:656–660
Bednarek MA, Feighner SD, Pong SS, McKee KK, Hreniuk DL, Silva MV, Warren VA, Howard AD, Van Der Ploeg LH, Heck JV 2000 Structure-function studies on the new growth hormone-releasing peptide, ghrelin: minimal sequence of ghrelin necessary for activation of growth hormone secretagogue receptor 1a. J Med Chem 43:4370–4376
Hosoda H, Kojima M, Matsuo H, Kangawa K 2000 Purification and characterization of rat des-Gln14-ghrelin, a second endogenous ligand for the growth hormone secretagogue receptor. J Biol Chem 275:21995–22000
Hosoda H, Kojima M, Mizushima T, Shimizu S, Kangawa K 2003 Structural divergence of human ghrelin. Identification of multiple ghrelin-derived molecules produced by post-translational processing. J Biol Chem 278:64–70
Kojima M, Hosoda H, Matsuo H, Kangawa K 2001 Ghrelin: discovery of the natural endogenous ligand for the growth hormone secretagogue receptor. Trends Endocrinol Metab 12:118–122
Tsch?p M, Smiley DL, Heiman ML 2000 Ghrelin induces adiposity in rodents. Nature 407:908–913
Nakazato M, Murakami N, Date Y, Kojima M, Matsuo H, Kangawa K, Matsukura S 2001 A role for ghrelin in the central regulation of feeding. Nature 409:194–198
Howard AD, Feighner SD, Cully DF, Arena JP, Liberator PA, Rosenblum CI, Hamelin M, Hreniuk DL, Palyha OC, Anderson J, Paress PS, Diaz C, Chou M, Liu KK, McKee KK, Pong SS, Chaung LY, Elbrecht A, Dashkevicz M, Heavens R, Rigby M, Sirinathsinghji DJ, Dean DC, Melillo DG, Van der Ploeg LH, Patchett AA, Nargund R, Griffin PR, DeMartino JA, Gupton SK, Schaeffer JM, Smith RG 1996 A receptor in pituitary and hypothalamus that functions in growth hormone release. Science 273:974–977
Ariyasu H, Takaya K, Tagami T, Ogawa Y, Hosoda K, Akamizu T, Suda M, Koh T, Natsui K, Toyooka S, Shirakami G, Usui T, Shimatsu A, Doi K, Hosoda H, Kojima M, Kangawa K, Nakao K 2001 Stomach is a major source of circulating ghrelin, and feeding state determines plasma ghrelin-like immunoreactivity levels in humans. J Clin Endocrinol Metab 86:4753–4758
Hosoda H, Kojima M, Matsuo H, Kangawa K 2000 Ghrelin and des-acyl ghrelin: two major forms of rat ghrelin peptide in gastrointestinal tissue. Biochem Biophys Res Commun 279:909–913
Cassoni P, Ghe C, Marrocco T, Tarabra E, Allia E, Catapano F, Deghenghi R, Ghigo E, Papotti M, Muccioli G 2004 Expression of ghrelin and biological activity of specific receptors for ghrelin and des-acyl ghrelin in human prostate neoplasms and related cell lines. Eur J Endocrinol 150:173–184
Thompson NM, Gill DA, Davies R, Loveridge N, Houston PA, Robinson IC, Wells T 2004 Ghrelin and des-octanoyl ghrelin promote adipogenesis directly in vivo by a mechanism independent of the type 1a growth hormone secretagogue receptor. Endocrinology 145:234–242
Bedendi I, Alloatti G, Marcantoni A, Malan D, Catapano F, Ghe C, Deghenghi R, Ghigo E, Muccioli G 2003 Cardiac effects of ghrelin and its endogenous derivatives des-octanoyl ghrelin and des-Gln14-ghrelin. Eur J Pharmacol 476:87–95
Baldanzi G, Filigheddu N, Cutrupi S, Catapano F, Bonissoni S, Fubini A, Malan D, Baj G, Granata R, Broglio F, Papotti M, Surico N, Bussolino F, Isgaard J, Deghenghi R, Sinigaglia F, Prat M, Muccioli G, Ghigo E, Graziani A 2002 Ghrelin and des-acyl ghrelin inhibit cell death in cardiomyocytes and endothelial cells through ERK1/2 and PI 3-kinase/AKT. J Cell Biol 159:1029–1037
Gualillo O, Lago F, Gomez-Reino J, Casanueva FF, Dieguez C 2003 Ghrelin, a widespread hormone: insights into molecular and cellular regulation of its expression and mechanism of action. FEBS Lett 552:105–109
Korbonits M, Kojima M, Kangawa K, Grossman AB 2001 Presence of ghrelin in normal and adenomatous human pituitary. Endocrine 14:101–104
Papotti M, Cassoni P, Volante M, Deghenghi R, Muccioli G, Ghigo E 2001 Ghrelin-producing endocrine tumors of the stomach and intestine. J Clin Endocrinol Metab 86:5052–5059
Volante M, Allia E, Gugliotta P, Funaro A, Broglio F, Deghenghi R, Muccioli G, Ghigo E, Papotti M 2002 Expression of ghrelin and of the GH secretagogue receptor by pancreatic islet cells and related endocrine tumors. J Clin Endocrinol Metab 87:1300–1308
Jeffery PL, Herington AC, Chopin LK 2002 Expression and action of the growth hormone releasing peptide ghrelin and its receptor in prostate cancer cell lines. J Endocrinol 172:R7–R11
Kanamoto N, Akamizu T, Hosoda H, Hataya Y, Ariyasu H, Takaya K, Hosoda K, Saijo M, Moriyama K, Shimatsu A, Kojima M, Kangawa K, Nakao K 2001 Substantial production of ghrelin by a human medullary thyroid carcinoma cell line. J Clin Endocrinol Metab 86:4984–4990
Hattori N, Saito T, Yagyu T, Jiang BH, Kitagawa K, Inagaki C 2001 GH, GH receptor, GH secretagogue receptor, and ghrelin expression in human T cells, B cells, and neutrophils. J Clin Endocrinol Metab 86:4284–4291
Poncz M, Surrey S, LaRocco P, Weiss MJ, Rappaport EF, Conway TM, Schwartz E 1987 Cloning and characterization of platelet factor 4 cDNA derived from a human erythroleukemic cell line. Blood 69:219–223
Hong Y, Martin JF, Vainchenker W, Erusalimsky JD 1996 Inhibition of protein kinase C suppresses megakaryocytic differentiation and stimulates erythroid differentiation in HEL cells. Blood 87:123–131
Long MW, Heffner CH, Williams JL, Peters C, Prochownik EV 1990 Regulation of megakaryocyte phenotype in human erythroleukemia cells. J Clin Invest 85:1072–1084
Papayannopoulou T, Nakamoto B, Yokochi T, Chait A, Kannagi R 1983 Human erythroleukemia cell line (HEL) undergoes a drastic macrophage-like shift with TPA. Blood 62:832–845
Kruh J 1982 Effects of sodium butyrate, a new pharmacological agent, on cells in culture. Mol Cell Biochem 42:65–82
Gourlet P, Vandermeers A, Vertongen P, Rathe J, De Neef P, Cnudde J, Waelbroeck M, Robberecht P 1997 Development of high affinity selective VIP1 receptor agonists. Peptides 18:1539–1545
Fraker PJ, Speck Jr JC 1978 Protein and cell membrane iodinations with a sparingly soluble chloroamide, 1,3,4,6-tetrachloro-3,6-diphrenylglycoluril. Biochem Biophys Res Commun 80:849–857
Tomasetto C, Karam SM, Ribieras S, Masson R, Lefebvre O, Staub A, Alexander G, Chenard MP, Rio MC 2000 Identification and characterization of a novel gastric peptide hormone: the motilin-related peptide. Gastroenterology 119:395–405
Van Craenenbroeck M, Gregoire F, De Neef P, Robberecht P, Perret J 2004 Ala-scan of ghrelin (1–14): interaction with the recombinant human ghrelin receptor. Peptides 25:959–965
Ellman GL, Courtney KD, Andres Jr V, Feather-Stone RM 1961 A new and rapid colorimetric determination of acetylcholinesterase activity. Biochem Pharmacol 7:88–95
Yang ZP, Dettbarn WD 1998 Prevention of tolerance to the organophosphorus anticholinesterase paraoxon with carboxylesterase inhibitors. Biochem Pharmacol 55:1419–1426
Duysen EG, Li B, Xie W, Schopfer LM, Anderson RS, Broomfield CA, Lockridge O 2001 E vidence for nonacetylcholinesterase targets of organophosphorus nerve agent: supersensitivity of acetylcholinesterase knockout mouse to VX lethality. J Pharmacol Exp Ther 299:528–535
Bradford MM 1976 A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 72:248–254
Gnanapavan S, Kola B, Bustin SA, Morris DG, McGee P, Fairclough P, Bhattacharya S, Carpenter R, Grossman AB, Korbonits M 2002 The tissue distribution of the mRNA of ghrelin and subtypes of its receptor, GHS-R, in humans. J Clin Endocrinol Metab 87:2988
Cummings DE, Weigle DS, Frayo RS, Breen PA, Ma MK, Dellinger EP, Purnell JQ 2002 Plasma ghrelin levels after diet-induced weight loss or gastric bypass surgery. N Engl J Med 346:1623–1630
Bagnasco M, Kalra PS, Kalra SP 2002 Ghrelin and leptin pulse discharge in fed and fasted rats. Endocrinology 143:726–729
De Vriese C, Grégoire F, Lema-Kisoka R, Waelbroeck M, Robberecht P, Delporte C 2004 Ghrelin degradation by serum and tissue homogenates: identification of the cleavage sites. Endocrinology 145:4997–5005
Martin P, Papayannopoulou T 1982 HEL cells: a new human erythroleukemia cell line with spontaneous and induced globin expression. Science 216:1233–1235
Long MW, Heffner CH 1988 Detection of human megakaryocyte antigens by solid-phase radioimmunoassay. Exp Hematol 16:62–70
Dixit VD, Schaffer EM, Pyle RS, Collins GD, Sakthivel SK, Palaniappan R, Lillard JW Jr, Taub DD 2004 Ghrelin inhibits leptin- and activation-induced proinflammatory cytokine expression by human monocytes and T cells. J Clin Invest 114:57–66
Li WG, Gavrila D, Liu X, Wang L, Gunnlaugsson S, Stoll LL, McCormick ML, Sigmund CD, Tang C, Weintraub NL 2004 Ghrelin inhibits proinflammatory responses and nuclear factor-B activation in human endothelial cells. Circulation 109:2221–2226
Dembinski A, Warzecha Z, Ceranowicz P, Tomaszewska R, Stachura J, Konturek SJ, Konturek PC 2003 Ghrelin attenuates the development of acute pancreatitis in rat. J Physiol Pharmacol 54:561–573
Otero M, Nogueiras R, Lago F, Dieguez C, Gomez-Reino JJ, Gualillo O 2004 Chronic inflammation modulates ghrelin levels in humans and rats. Rheumatology 43:306–310
Stenvinkel P, Pecoits-Filho R, Lindholm B 2003 Leptin, ghrelin, and proinflammatory cytokines: compounds with nutritional impact in chronic kidney disease? Adv Ren Replace Ther 10:332–345
Das UN 2001 Is obesity an inflammatory condition? Nutrition 17:953–966(Carine De Vriese, Fran?oi)