Ghrelin Is Expressed in a Novel Endocrine Cell Type in Developing Rat Islets and Inhibits Insulin Secretion from INS-1 (832/13) Cells

Abstract

Ghrelin is produced mainly by endocrine cells in the stomach and is an endogenous ligand for the growth hormone secretagogue receptor (GHS-R). It also influences feeding behavior, metabolic regulation, and energy balance. It affects islet hormone secretion, and expression of ghrelin and GHS-R in the pancreas has been reported. In human islets, ghrelin expression is highest pre- and neonatally. We examined ghrelin and GHS-R in rat islets during development with immunocytochemistry and in situ hybridization. We also studied the effect of ghrelin on insulin secretion from INS-1 (832/13) cells and the expression of GHS-R in these cells. We found ghrelin expression in rat islet endocrine cells from mid-gestation to 1 month postnatally. Islet expression of GHS-R mRNA was detected from late fetal stages to adult. The onset of islet ghrelin expression preceded that of gastric ghrelin. Islet ghrelin cells constitute a separate and novel islet cell population throughout development. However, during a short perinatal period a minor subpopulation of the ghrelin cells co-expressed glucagon or pancreatic polypeptide. Markers for cell lineage, proliferation, and duct cells revealed that the ghrelin cells proliferate, originate from duct cells, and share lineage with glucagon cells. Ghrelin dose-dependently inhibited glucose-stimulated insulin secretion from INS-1 (832/13) cells, and GHS-R was detected in the cells. We conclude that ghrelin is expressed in a novel developmentally regulated endocrine islet cell type in the rat pancreas and that ghrelin inhibits glucose-stimulated insulin secretion via a direct effect on the β-cell.

Keywords

ghrelin islet hormones islet cells islet development insulin β-cells INS-1 (823/13) cells Brn 4

Ghrelin is a 28-amino-acid octanoylated peptide isolated from rat stomach as a ligand of the growth hormone (GH) secretagogue receptor (GHS-R) (Kojima et al. 1999). Ghrelin is mainly produced by a population of endocrine cells, the A-like cells, which are abundant in the oxyntic mucosa in the stomach, less common in the gastric antrum, and still fewer in the small intestine (Date et al. 2000; Dornonville de la Cour et al. 2001). Ghrelin expression has also been detected in the hypothalamus (Kojima et al. 1999), placenta (Gualillo et al. 2001), kidney (Mori et al. 2000), testis (Tena-Sempere et al. 2002), and developing lung (Volante et al. 2002b). It has been suggested that ghrelin is secreted primarily from the stomach, acting as a hormone to stimulate pituitary GH release (Kojima et al. 2001). Recent studies have shown additional actions of ghrelin. Ghrelin stimulates food intake in rats and humans (Tschöp et al. 2000; Wren et al. 2000,2001a) and causes body weight gain with increased adiposity due to changes in energy balance, including a reduction in fat utilization, as studied in rats and mice (Tschöp et al. 2000; Wren et al. 2001b). There are reports indicating expression of ghrelin and GHS-R in the pancreas (Guan et al. 1997; Date et al. 2002; Volante et al. 2002a; Wierup et al. 2002; Colombo et al. 2003). In one report, ghrelin expression was found in the β-cells of human islets (Volante et al. 2002a). In another report, ghrelin expression was localized to the α-cells of human and rat islets (Date et al. 2002). We recently reported that ghrelin cells in the human pancreas are devoid of any of the “classical” islet hormones and constitute a novel endocrine islet cell type. We also found that the islet ghrelin cells are developmentally regulated in that they are numerous pre- and neonatally but few in adults (Wierup et al. 2002). Ghrelin has been reported to influence insulin secretion. In some studies, ghrelin was shown to inhibit insulin secretion, as observed in humans (Broglio et al. 2001) and rodents (Egido et al. 2002; Colombo et al. 2003; Reimer et al. 2003). However, stimulating effects on insulin secretion in rats have also been reported (Date et al. 2002; Lee et al. 2002). We decided to examine the possibility of ghrelin and GHS-R expression in the developing and mature rat pancreas with immunocytochemistry (ICC) and in situ hybridization (ISH). We also studied the effect of ghrelin on glucose-stimulated insulin secretion from INS-1 (832/13) cells (Hohmeier et al. 2000) and the possibility of GHS-R expression in these cells. Further, we examined the possibility of co-expression of ghrelin and hormones known to be present in rat islets during development [insulin, glucagon, somatostatin, pancreatic polypeptide (PP), peptide YY (PYY), islet amyloid polypeptide (IAPP), and gastrin]. We also used cell- or cell lineage-specific markers, including Brn4, a POU domain transcription factor, specific for cells belonging to the α-cell lineage (Hussain et al. 1997), and the homeodomain transcription factor PDX-1 as a marker for the β-cell lineage (Oster et al. 1998). Cytokeratin 20 (CK20) was used as a duct cell/islet cell precursor marker (Bouwens et al. 1994) and Ki67 as a marker of proliferating cells. Chromogranin A (CgA) was used as a general marker of endocrine cells.

Materials and Methods

Animals and Tissue

Rats were housed under alternate 12-hr periods of light and dark with free access to standard rat food and tapwater. Sprague-Dawley rats (B&K; Stockholm, Sweden) of both genders, 8–20 weeks of age (n = 20), fetuses at embryonic (E) days 15, 17, 20, and 21 (n = 8–12 of each stage), and newborn and young at postnatal (P) days 0, 1, 2, 3, 5, 12, 17, 20, 25, 30, and 40 (n = 5 of each stage) were used. The pancreas and stomach were dissected out, fixed overnight in Stefanini's solution (2% paraformaldehyde and 0.2% picric acid in 0.1 M PB, pH 7.2), rinsed thoroughly in Tyrode solution containing 10% sucrose, and frozen on dry ice. Sections (10-μm thickness) were cut and thaw-mounted on slides. In postnatal and adult rats, head and tail portions of the pancreas were processed separately. The experiments were approved by the Animal Ethics Committee, Lund and Malmö.

Table 1

Details of the antibodies used for immunocytochemistry

Antigen	Code	Raised against	Raised in	Dilution	Source
Ghrelin	H-031–31	Rat ghrelin	Rabbit	1:3200	Phoenix; Belmont, CA
Ghrelin	FG-G-031–30	Rat ghrelin	Rabbit	1:1000	Phoenix
Proinsulin	9003	Human proinsulin	Guinea pig	1:2560	Eurodiagnostica; Malmö, Sweden
Glucagon	8708	Porcine glucagon	Guinea pig	1:5210	Eurodiagnostica
Somatostatin	V 1169	Human somatostatin	Mouse	1:400	Biomeda; Foster City, CA
Pancreatic polypeptide	AHP 515	Human pancreatic polypeptide	Sheep	1:640	Serotec; Oxford, UK
Chromogranin-A	LK2H10	Human chromogranin	Mouse	1:160	Boehringer; Mannheim, Germany
Brn4	4R3	Recombinant protein	Rabbit	1:640	Own
GLUT2	AB 1342	Synthetic peptide	Rabbit	1:640	Chemicon; Temecula, CA
IAPP	F 055–24	Human IAPP 1–10	Mouse	1:2560	Amylin Pharmaceuticals; San Diego, CA
CK20	MKs20.8	Human CK	Mouse	1:160	Eurodiagnostica
Ki67	M7187	Human Ki67	Mouse	1:80	DAKO; Glostrup, Denmark
GHS-R1a	H-001–61	Human GHS-R1a	Rabbit	1:500	Phoenix
PDX-1	PDX-1	Carboxy terminus of mouse pdx-1	Rabbit	1:2000	Generous gift from Dr M. German, University of California San Francisco

Immunocytochemistry

Antibodies were diluted in PBS, pH 7.2, containing 0.25% bovine serum albumin and 0.25% Triton X-100. Sections were incubated with primary antibodies in moisturized chambers (Table 1) overnight at 4C, followed by rinsing in PBS with Triton X-100 twice for 10 min. Thereafter secondary antibodies with specificity for rabbit, guinea pig, sheep, or mouse IgG coupled to either fluorescein isothiocyanate (FITC) (DAKO, Copenhagen, Denmark; Jackson, West Grove, PA; Sigma, St Louis, MO), Texas Red (Jackson), or 7-amino-4-methyl coumarin-3-acetic acid (AMCA) (Jackson), were applied on the sections. Incubation was for 1 hr at room temperature (RT). Sections were again rinsed in Triton X-100-enriched PBS twice for 10 min and then mounted in PBS:glycerol 1:1. In addition, ghrelin antibodies directly conjugated with 5-(and 6)-carboxyfluorescein (FAM) were used (Table 1). INS-1 (832/13) cells grown on coverslips were fixed in 4% paraformaldehyde for 10 min, washed with PBS twice for 10 min, and PBS with 0.25% Triton X-100 for 15 min. Primary antibody was applied and incubation was for 1 hr at RT. The cells were washed with PBS containing 0.25% Triton X-100, incubated with secondary antibody for 1 hr at RT, and again washed with PBS containing 0.25% Triton X-100. The specificity of immunostaining was tested using primary antisera pre-absorbed with homologus antigen (100 μg of peptide per ml antiserum at working dilution) or by omission of the primary antibodies. Double or triple immunofluorescence was also used with combinations of primary antibodies (rabbit, guinea pig, sheep, or monoclonal antibodies). The two or three primary antibodies were incubated simultaneously overnight at 4C, followed by rinsing in PBS with Triton X-100 twice for 10 min. Then the two or three secondary antibodies were incubated simultaneously for 1 hr at RT. In these studies the controls included tests for inappropriate binding of the secondary antibodies. When directly conjugated ghrelin antibodies in combination with conventional indirect immunostaining were used, the incubations were in sequence, ending with the directly conjugated antibody.

In Situ Hybridization

The ghrelin probe was complementary to the sequence 172–201 of rat ghrelin cDNA (GenBank accession number NM 021669) (Kojima et al. 1999). The GHS-R probe was complementary to the sequence 162–194 of rat GHS-R cDNA (GenBank accession number NM 032075.1) (McKee et al. 1997) and covering GHS-R subtypes 1a and 1b (Guan et al. 1997). BLAST searches were run, demonstrating lack of significant sequence similarity with any other mammalian cDNAs. The probes were synthesized by the Biomolecular Resource Facility of the University of Lund or by DNA Technology A/S (Aarhus, Denmark), and were 3'-endtailed by [³⁵S]-dATP (NEN; Stockholm, Sweden) (Mulder et al. 1993). The ISH protocol has been described previously (Mulder et al. 1993). In brief, sections were rapidly air-dried, fixed in 4% paraformaldehyde for 15 min, washed twice for 5 min in PBS, and acetylated with 0.25% acetic anhydride in 0.1 M triethanolamine for 10 min. Then the sections were dehydrated in graded ethanols, treated with chloroform for 5 min, ethanol (95.5%) for 5 min, and again air-dried. Hybridization was carried out in sealed moisturizing chambers at 37C overnight, using probe concentrations of approximately 1 pmol/ml for ghrelin and 3 pmol/ml for GHS-R, followed by stringent post-hybridization washing (1 × SSC; 0.15 M NaCl, 0.015 M Na citrate). The slides were dipped in NTB 2 emulsion (Eastman Kodak; Rochester, NY) and stored in light-sealed boxes at 4C for 10–18 days. They were then developed in Kodak D-19, fixed in Kodak Polymax, and mounted in Kaiser's glycerol gelatin. As controls, sense probe or excess unlabeled probe was used, as previously described (Wierup et al. 2002).

Imaging and Cell Quantification

Immunofluorescence was examined in an epifluorescence microscope (Olympus BX60). By changing filters, the location of the different secondary antibodies in double and triple staining was determined. ISH radiolabeling was examined in brightfield. Images were captured with a digital camera (Olympus DP50). Ghrelin-immunoreactive (IR) cells were counted in immunostained sections of pancreas and stomach from fetal (E17 and E20) (n = 5 of each stage) and postnatal rats (P0, 1, 2, 3, 5, 12, 17, 20, 25, 30, and 40) (n = 5 of each stage), using a fluorescence microscope (×250 magnification; objective ×25, eyepiece ×10). All ghrelin-IR cells were counted within five to ten visual fields in one section from each specimen.

Insulin Secretion Assay and Radioimmunoassay on INS-1 (832/13) Cells

INS-1 (832/13) cells were seeded in 24-well plates and grown to confluence in RPMI 1640 with 11.1 mM glucose supplemented with 10% fetal calf serum, 100 U/ml penicillin, 100 μg/ml streptomycin, 10 mM HEPES, 2 mM glutamine, 1 mM sodium pyruvate, and 50 μM β-mercaptoethanol. All incubations were at 37C in a humidified atmosphere containing 5% CO₂. Twelve hours before the assay the medium was changed to complete RPMI 1640 with 5 mM glucose. Next, the cells were pre-incubated in HEPES-buffered saline solution (HBSS; 114 mM NaCl, 4.7 mmol KCl, 1.2 mM KH₂PO₄, 1.16 mM MgSO₄, 20 mM HEPES, 2.5 mM CaCl₂, 25.5 mM NaHCO₃, pH 7.2, with 3 mM glucose) for 2 hr. For assay of insulin secretion, the cells were incubated for 1 hr in HBSS with 3 mM glucose or 15 mM glucose with the addition of 0.1–100 nM of ghrelin (Phoenix; Belmont, CA). Insulin released into the buffer during the 1-hr static incubation was determined with a human insulin RIA kit (Coat-a-count; DPC, Los Angeles, CA).

Statistical Analysis

Results are shown as means ± SEM. Data were analyzed by a one-way ANOVA followed by Bonferroni's correction. Differences with a value of p<0.05 were considered significant.

Results

Immunocytochemistry and Cell Quantification

Ghrelin-IR cells were readily seen in the pancreas from E15 to P25. The numbers of ghrelin-IR cells peaked between E17 and P5 (Figure 1A). After P12 the number of cells gradually declined to only a single cell per visual field at P25. Only an occasional ghrelin-IR cell could be detected at P30 and P40 (Figure 1A). In the adult rat, ghrelin IR cells were only rarely detected in the islets or in the exocrine parenchyma. In the stomach, ghrelin IR cells were undetectable at E17 but demonstrable at E20. With the gastric ghrelin cell density at E20 assigned as 1, the cell density had increased threefold at P12 and 10-fold at P20 (Figure 1B). In the pancreas, the ghrelin-IR cells were located at the periphery of the islets, usually as single cells but sometimes forming small clusters. The ghrelin-IR cells often had a rounded shape but occasional cells displayed one or two cytoplasmic extensions. To verify that the ghrelin-IR cells were endocrine cells, CgA was used as an endocrine cell marker. Double staining for ghrelin and CgA showed that the ghrelin-IR cells were weakly to moderately CgA-positive.

To investigate if the ghrelin IR cells co-expressed any of the classical islet hormones, triple staining for ghrelin/somatostatin/insulin and double staining for ghrelin/PP and for ghrelin/glucagon were performed. Ghrelin-IR cells were often located in the immediate vicinity of somatostatin-IR and/or glucagon-IR cells and/or PP-IR cells. Ghrelin-IR cells did not coexpress insulin or somatostatin (Figures 2A and 2B) but occasionally expressed PP or glucagon in minor subpopulations. Ghrelin-IR cells co-expressing PP could be detected from late stages of gestation (E20) (Figure 2E) to P5. Later during development, PP-IR cells and ghrelin-IR cells were separate populations (Figure 2F). A minor subpopulation of the ghrelin-IR cells co-expressed glucagon from E17 (Figure 2C) to P5. However, the cells were fewer than those co-expressing PP. At later stages, glucagon-IR cells and ghrelin-IR cells were separate populations (Figure 2D). Because glucagon IR was detected in some ghrelin-IR cells around birth, we investigated whether they also expressed the α-cell lineage marker Brn4. Brn4 IR was found in a large population of peripheral islet cells both pre- and postnatally and was localized to the nucleus. Double immunostaining with glucagon revealed that the vast majority of the Brn4-expressing cells were glucagon IR. Double staining for ghrelin and Brn4 showed that a minority of the ghrelin-IR cells also expressed Brn4 (Figures 3A and 3B). Next, we used PDX-1 as a marker of β-cell lineage. As expected, double staining for PDX-1 and insulin showed that all insulin-IR cells were PDX-1-positive. Conversely, ghrelin-IR cells consistently lacked PDX-1 (Figure 3C). To investigate if the ghrelin-IR cells, like islet endocrine cells in general, develop from cells of duct epithelium, we used CK20 as a duct cell/islet cell precursor marker. Staining for CK20 was detected in duct epithelium and in peripheral islet cells, as previously reported (Bouwens et al. 1994). Double staining for ghrelin and CK20 revealed that a quite prominent subpopulation of the ghrelin-IR cells were CK20 IR (Figure 3D). To investigate whether the ghrelin-IR cells proliferate, expression of Ki67 was examined. Double staining for ghrelin and Ki67 revealed that a subpopulation of the ghrelin-IR cells were Ki67 IR (Figures 3E and 3F). We then studied some regulatory peptides known to be expressed in rat islets during development. We detected IAPP expression to extend beyond insulin and somatostatin IR cells, in the developing islets, as previously reported (Mulder et al. 1997), and early postnatally some ghrelin-IR cells harbored IAPP IR. Double immunostainings revealed that ghrelin and gastrin or ghrelin and PYY only rarely coexisted in the islets during development.

Figure 1

Density of ghrelin cells in the developing rat pancreas (A) and stomach (B). Ghrelin cells in the pancreas precede ghrelin cells in the stomach. The decrease in pancreatic ghrelin cell density coincides with an increase in gastric ghrelin cell density.

In Situ Hybridization for Ghrelin and GHS-R

In the stomach of adult rats, many scattered cells in the mucosal epithelium were strongly labeled for ghrelin mRNA, confirming that the probe was effective. In the pancreas, labeling for ghrelin mRNA was detected in a few cells at the periphery of islets from late prenatal stages (E20) to P25 (Figures 4A and 4B). In addition, scattered labeling for GHS-R mRNA was detected over larger islet areas of all stages examined (Figure 4C). Single densely labeled cells were randomly distributed in islets between late gestation (E20) and P12 (Figure 4D). No or only very weak background labeling was detected in control sections when the sense probe or an excess of unlabeled probe was used.

Effect of Ghrelin on Insulin Secretion

To examine whether ghrelin affects insulin secretion by a direct effect on the β-cell, the clonal INS-1 (832/13) line was employed. Cells were incubated at low (3 mM) or high (15 mM) glucose with increasing concentrations of ghrelin. Raising the glucose in the medium from 3 to 15 mM provoked a fivefold increase in insulin secretion during the 1-hr static incubation. At 3 mM glucose, addition of ghrelin (0.1–100 nM) did not affect insulin secretion. However, at 15 mM glucose, addition of ghrelin dose-dependently inhibited insulin secretion. The maximal effect was observed at 100 nM ghrelin and reached a 40% reduction (p<0.05) (Figure 5). A physiological effect of ghrelin requires expression of a cell surface receptor to which the peptide can bind and elicit its effects. Therefore, immunostaining for GHS-R type 1a in the 832/13 line was performed. Indeed, GHS-R type 1a IR was observed in all INS-1 (832/13) cells studied (Figure 6).

Figure 2

Fluorescence photomicrographs of double- or triple-immunostained rat islets. A (E20) and B (P25) stained for ghrelin (red), insulin (blue), and somatostatin (green). C (E20) and D (P25) stained for ghrelin (red) and glucagon (green). E (E20) and F (P25) stained for ghrelin (red) and PP (green). Ghrelin is expressed in a separate cell type except for a period around birth, when expression of glucagon or PP is seen in a minor subpopulation of the ghrelin cells. Co-expression (yellow) is indicated with arrows. Islet border is indicated with a dashed line in F. Bars = 20 μm.

Figure 3

Fluorescence photomicrographs of double-immunostained neonatal (P2) rat islet cells. (A,B) Ghrelin (red) and Brn4 (green). A sub-population of the ghrelin cells express Brn4. (C) Ghrelin (red) and PDX-1 (green). Ghrelin cells are devoid of PDX-1. (D) Ghrelin (red) and CK20 (green). A major proportion of the ghrelin cells also harbored CK20. (E,F) Ghrelin (red) and Ki67 (green). A subpopulation of the ghrelin cells is proliferating. Co-expression (yellow) is indicated by arrows. Bars = 20 μm.

Figure 4

ISH autoradiographs of fetal and neonatal rat islets. A (E20) and B (P25), islets with labeling for ghrelin mRNA in cells at the islet periphery. (C,D) Sections of islets with labeling for GHS-R mRNA. (C) P20: weak diffuse labeling of central islet cells. (D) P3: densely labeled peripheral islet cell. Islet border indicated with dashed line. Bars = 20 μm.

Discussion

In the present study we show that ghrelin is expressed in a separate and novel cell population in developing rat islets. We found cells displaying ghrelin IR and ghrelin mRNA labeling in the developing islets from E15 to 1 month postnatally. At later stages of development and in the adult rat, cells displaying ghrelin peptide or ghrelin mRNA were only rarely seen in the pancreas. Date et al. (2002) have presented data indicating that ghrelin is expressed in the α-cells in adult human and rat islets. However, in this study we were unable to confirm the co-expression of ghrelin and glucagon in the islets of adult rats. There is no obvious explanation for the divergent results, but the previously reported lack of ghrelin mRNA in the adult rat pancreas (Lee et al. 2002), together with our observations that α-cells in the adult rat and the vast majority of the glucagon IR cells in the neonatal rat lacked both ghrelin mRNA and ghrelin peptide, do not favor any significant ghrelin expression in these cells. In fact, our data indicate that islet ghrelin cells in rats, as in humans (Wierup et al. 2002), largely constitute a separate cell population, with the exception (in rats) of a short perinatal period when expression of glucagon or PP was seen in a minor subpopulation of the ghrelin cells. However, these latter findings suggest that ghrelin cells, α-cells, and PP cells belong to the same cell lineage. Our finding that some ghrelin cells co-express Brn4 further strengthens a developmental relationship between ghrelin cells and α-cells. In contrast, the lack of PDX-1 expression in the ghrelin cells clearly indicates that they belong to a cell lineage different from the β-cell. We also found that the ghrelin cells expressed CK20 and that Ki67 also occurred in a sub-population of the ghrelin cells. Together, these data suggest that the ghrelin cells, like islet cells in general, develop from the duct epithelium (Bouwens et al. 1994) and that they proliferate perinatally. The finding of cells co-expressing ghrelin and IAPP is in line with a broad expression pattern of IAPP during rat islet development (Mulder et al. 1997). During a period around birth IAPP is found in several different islet cell types, as documented by ICC.

Figure 5

Insulin secretion from on INS-1 (832/13) cells treated with 0.1–100 nM ghrelin for 1 hr. Insulin was measured with RIA as medium concentration after incubation. Ghrelin dose-dependently inhibits glucose-stimulated insulin secretion. ∗p<0.05 vs control. Data are from five independent experiments performed in quadruplicate.

It is of interest that the onset of islet ghrelin expression precedes that in the stomach. In the present study, islet ghrelin cells were detected as early as at E15, whereas gastric ghrelin cells were undetectable at E17 but demonstrable at E20. These latter data are in agreement with those recently presented by Hayashida et al. (2002), who found the first ghrelin-IR cells in the stomach at E18. This developmental profile of ghrelin cells in the rat pancreas and stomach is reminiscent of that in human fetuses, in which pancreatic ghrelin cells precede those in the stomach (Wierup et al. 2002). Previous data also indicate that the levels of ghrelin peptide and ghrelin mRNA remain low during the first week after birth in rat and mouse stomach (Lee et al. 2002; Liu et al. 2002), and not until 2–3 weeks postnatally is there a steep rise in gastric ghrelin expression. Our observations agree with these data in that the frequency of ghrelin cells in the rat stomach increases most rapidly during the weaning period. The density of gastric ghrelin cells increased fourfold between P12 and P20, with only a limited further increase between P20 and P40. Interestingly, this rapid increase in gastric ghrelin cell density coincides in time with the gradual decline in density of pancreatic ghrelin cells, suggesting a relationship between ghrelin production in pancreas and stomach during development. In preliminary studies we have made similar observations in mice (Wierup et al., unpublished data). This may suggest a hormonal role for pancreatic ghrelin, e.g., as a GH regulator, during development before gastric ghrelin comes into play. Ghrelin may also have a role as a local paracrine regulator of hormone secretion and/or of islet growth or differentiation. The documented GHS-R expression in the human fetal pituitary (Shimon et al. 1998) indicates that the ghrelin-GH axis is also active during development, and it is not inconceivable that islet ghrelin may contribute ligand. In humans this may also be the case in adults, because after gastrectomy 35% of circulating ghrelin remains (Ariyasu et al. 2001), and islet ghrelin cells remain in adults (Wierup et al. 2002). The corresponding figure in adult rats is lower, about 20% (Dornonville de la Cour et al. 2001). This lower percentage may be related to the lack of pancreatic ghrelin cells in adult rats.

Figure 6

Fluorescence photomicrograph of INS-1 (832/13) cells stained for GHS-R type 1a. All cells express GHS-R type 1a to various extents. Bar = 10 μm.

Recent studies have reported ghrelin to inhibit insulin secretion both in humans (Broglio et al. 2001) and in rodents (Egido et al. 2002; Reimer et al. 2003). However, stimulating effects on insulin secretion in rats have also been reported (Date et al. 2002; Lee et al. 2002). These data derive from studies either in vivo, on isolated perfused pancreas, or on isolated islets. Therefore, they do not allow any conclusion about whether ghrelin acts directly on the β-cell or indirectly via the release of β-cell modulators. While this study was in progress, Colombo et al. (2003) reported that ghrelin inhibits glucose-stimulated insulin secretion from INS-1E cells. Our present observation that ghrelin dose-dependently inhibits glucose-stimulated insulin secretion from INS-1 (832/13) cells corroborates these findings, and together these observations indicate that ghrelin acts directly on the β-cell. This notion is further substantiated by the present finding of GHS-R expression in the INS-1 (832/13) cells. Ghrelin can therefore be added to an array of inhibitors of insulin secretion that are overexpressed in the islets during development. Such overexpression may serve to keep the β-cells glucose-unresponsive prenatally and could therefore be an important protective mechanism against hypoglycemia in the fetus. Among such inhibitory peptides are NPY, expressed in rat β-cells perinatally (Moltz and McDonald 1985; Myrsen-Axcrona et al. 1997; Ahren 2000), and the structurally related peptide PYY, expressed mainly in the α-cells during islet ontogeny (Böttcher et al. 1989; Upchurch et al. 1994; Myrsen-Axcrona et al. 1997).

Cells expressing mRNA for GHS-R were detected in the islets throughout development. This may suggest an autocrine or paracrine role of islet ghrelin. The cells in the islets expressing GHS-R have not yet been identified. Recent data indicate low levels of expression in β-cells as well as α-cells (Colombo et al. 2003). Further studies are needed to clarify the cellular location of GHS-R in the islets and their role in islet function. The possibility that receptors other than the cloned GHS-R are involved must also be considered (Bodart et al. 2002).

In conclusion, this study has shown that ghrelin cells are present in developing rat islets. The ghrelin cells constitute a separate cell population except for a short period around birth, when a minor subpopulation of the ghrelin cells co-express PP or glucagon. These findings in developing rat islets, together with our previous findings in human islets, establish the ghrelin cell as a novel islet cell type with a possible lineage relationship to the α-cells and PP cells. Ghrelin is an inhibitor of insulin secretion and exerts its effect directly on the β-cells. The ontogenetic appearance of islet ghrelin cells precedes that of gastric ghrelin cells, which may indicate a developmental role for islet ghrelin.

Footnotes

Acknowledgements

Supported by grants from the Swedish Medical Research Council (project no. 4499), the Swedish Diabetes Association, the Royal Physiographic Society, and the Påhlsson, Hedberg, and Gyllenstiernska Krapperup Foundations.

We thank Dr Michael German (University of California, San Francisco, CA) for kindly providing the PDX-1 antibody. We thank Eva Hansson, Karin Jansner, Ann-Christin Lindh, and Doris Persson for expert technical assistance.

References

Ahren

(2000) Autonomic regulation of islet hormone secretion—implications for health and disease. Diabetologia 43:393–410

Ariyasu

Takaya

Tagami

Ogawa

Hosoda

Akamizu

Suda

. (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

Bodart

Febbraio

Demers

McNicoll

Pohankova

Perreault

Sejlitz

. (2002) CD36 mediates the cardiovascular action of growth hormone-releasing peptides in the heart. Circ Res 90:844–849

Böttcher

Ahren

Lundquist

Sundler

(1989) Peptide YY: intrapancreatic localization and effects on insulin and glucagon secretion in the mouse. Pancreas 4:282–288

Bouwens

Wang

De Blay

Pipeleers

Klöppel

(1994) Cytokeratins as markers of ductal cell differentiation and islet neogenesis in the neonatal rat pancreas. Diabetes 43:1279–1283

Broglio

Arvat

Benso

Gottero

Muccioli

Papotti

Lely

. (2001) Ghrelin, a natural GH secretagogue produced by the stomach, induces hyperglycemia and reduces insulin secretion in humans. J Clin Endocrinol Metab 86:5083

Colombo

Gregersen

Xiao

Hermansen

(2003) Effects of ghrelin and other neuropeptides (CART, MCH, orexin A and B, and GLP-1) on the release of insulin from isolated rat islets. Pancreas 27:161–166

Date

Kojima

Hosoda

Sawaguchi

Mondal

Suganuma

Matsukura

. (2000) Ghrelin, a novel growth hormone-releasing acylated peptide, is synthesized in a distinct endocrine cell type in the gastrointestinal tracts of rats and humans. Endocrinology 141:4255–4261

Date

Nakazato

Hashiguchi

Dezaki

Mondal

Hosoda

Kojima

. (2002) Ghrelin is present in pancreatic alpha-cells of humans and rats and stimulates insulin secretion. Diabetes 51:124–129

10.

Dornonville de la Cour

Björkqvist

Sandvik

Bakke

Zhao

Chen

HÅkanson

(2001) A-like cells in the rat stomach contain ghrelin and do not operate under gastrin control. Regul Pept 99:141–150

11.

Egido

Rodriguez-Gallardo

Silvestre

Marco

(2002) Inhibitory effect of ghrelin on insulin and pancreatic somatostatin secretion. Eur J Endocrinol 146:241–244

12.

Gualillo

Caminos

Blanco

Garcia-Caballero

Kojima

Kangawa

Dieguez

. (2001) Ghrelin, a novel placental-derived hormone. Endocrinology 142:788–794

13.

Guan

Palyha

McKee

Feighner

Sirinathsinghji

Smith

. (1997) Distribution of mRNA encoding the growth hormone secretagogue receptor in brain and peripheral tissues. Brain Res Mol Brain Res 48:23–29

14.

Hayashida

Nakahara

Mondal

Date

Nakazato

Kojima

Kangawa

. (2002) Ghrelin in neonatal rats: distribution in stomach and its possible role. J Endocrinol 173:239–245

15.

Hohmeier

Mulder

Chen

Henkel-Rieger

Prentki

Newgard

(2000) Isolation of INS-1-derived cell lines with robust ATP-sensitive K+ channel-dependent and -independent glucose-stimulated insulin secretion. Diabetes 49:424–430

16.

Hussain

Lee

Miller

Habener

(1997) POU domain transcription factor brain 4 confers pancreatic alpha-cell-specific expression of the proglucagon gene through interaction with a novel proximal promoter G1 element. Mol Cell Biol 17:7186–7194

17.

Kojima

Hosoda

Date

Nakazato

Matsuo

Kangawa

(1999) Ghrelin is a growth-hormone-releasing acylated peptide from stomach. Nature 402:656–660

18.

Kojima

Hosoda

Matsuo

Kangawa

(2001) Ghrelin: discovery of the natural endogenous ligand for the growth hormone secretagogue receptor. Trends Endocrinol Metab 12:118–122

19.

Lee

Wang

Englander

Kojima

Greeley

Jr (2002) Ghrelin, a new gastrointestinal endocrine peptide that stimulates insulin secretion: enteric distribution, ontogeny, influence of endocrine, and dietary manipulations. Endocrinology 143:185–190

20.

Liu

Yakar

Otero-Corchon

Low

Liu

(2002) Ghrelin gene expression is age-dependent and influenced by gender and the level of circulating IGF-I. Mol Cell Endocrinol 189:97–103

21.

McKee

Palyha

Feighner

Hreniuk

Tan

Phillips

Smith

. (1997) Molecular analysis of rat pituitary and hypothalamic growth hormone secretagogue receptors. Mol Endocrinol 11:415–423

22.

Moltz

McDonald

(1985) Neuropeptide Y: direct and indirect action on insulin secretion in the rat. Peptides 6:1155–1159

23.

Mori

Yoshimoto

Takaya

Hosoda

Ariyasu

Yahata

Mukoyama

. (2000) Kidney produces a novel acylated peptide, ghrelin. FEBS Lett 486:213–216

24.

Mulder

Ekelund

Ekblad

Sundler

(1997) Islet amyloid polypeptide in the gut and pancreas: localization, ontogeny and gut motility effects. Peptides 18:771–783

25.

Mulder

Lindh

Sundler

(1993) Islet amyloid polypeptide gene expression in the endocrine pancreas of the rat: a combined in situ hybridization and immunocytochemical study. Cell Tissue Res 274:467–474

26.

Myrsen-Axcrona

Ekblad

Sundler

(1997) Developmental expression of NPY, PYY and PP in the rat pancreas and their coexistence with islet hormones. Regul Pept 68:165–175

27.

Oster

Jensen

Serup

Galante

Madsen

Larsson

(1998) Rat endocrine pancreatic development in relation to two homeobox gene products (Pdx-1 and Nkx 6.1). J Histochem Cytochem 46:707–715

28.

Reimer

Pacini

Ahren

(2003) Dose-dependent inhibition by ghrelin of insulin secretion in the mouse. Endocrinology 144:916–921

29.

Shimon

Yan

Melmed

(1998) Human fetal pituitary expresses functional growth hormone-releasing peptide receptors. J Clin Endocrinol Metab 83:174–178

30.

Tena-Sempere

Barreiro

Gonzalez

Gaytan

Zhang

Caminos

Pinilla

. (2002) Novel expression and functional role of ghrelin in rat testis. Endocrinology 143:717–725

31.

Tschöp

Smiley

Heiman

(2000) Ghrelin induces adiposity in rodents. Nature 407:908–913

32.

Upchurch

Aponte

Leiter

(1994) Expression of peptide YY in all four islet cell types in the developing mouse pancreas suggests a common peptide YY-producing progenitor. Development 120:245–252

33.

Volante

Allia

Gugliotta

Funaro

Broglio

Deghenghi

Muccioli

. (2002a) Expression of ghrelin and of the GH secretagogue receptor by pancreatic islet cells and related endocrine tumors. J Clin Endocrinol Metab 87:1300–1308

34.

Volante

Fulcheri

Allia

Cerrato

Pucci

Papotti

(2002b) Ghrelin expression in fetal, infant, and adult human lung. J Histochem Cytochem 50:1013–1021

35.

Wierup

Svensson

Mulder

Sundler

(2002) The ghrelin cell: a novel developmentally regulated islet cell in the human pancreas. Regul Pept 107:63–69

36.

Wren

Seal

Cohen

Brynes

Frost

Murphy

Dhillo

. (2001a) Ghrelin enhances appetite and increases food intake in humans. J Clin Endocrinol Metab 86:5992

37.

Wren

Small

Abbott

Dhillo

Seal

Cohen

Batterham

. (2001b) Ghrelin causes hyperphagia and obesity in rats. Diabetes 50:2540–2547

38.

Wren

Small

Ward

Murphy

Dakin

Taheri

Kennedy

. (2000) The novel hypothalamic peptide ghrelin stimulates food intake and growth hormone secretion. Endocrinology 141:4325–4328