Leptin, a recently discovered hormone secreted mainly from adipose tissue, was first described as a regulator of adiposity, food intake and energy metabolism. It is now apparent that leptin physiology is much more complex and is likely to play an important role in many other systems including reproduction, haematopoiesis and immunity. Leptin levels have been shown to be well correlated with body fat in both humans and rodents, with administration of exogenous leptin to rats and mice resulting in loss of body fat. Leptin is, therefore, likely to be an important humoral signal to the central nervous system on body composition and regulation of food consumption. Due to the limited cross-reactivity of leptin from other species in the current assays for leptin, physiological research on leptin has, to a large extent, been restricted to rodents and humans. The aim of this study was to develop a leptin immunoassay suitable for use with sheep, enabling the investigation of the basic physiology of leptin in an animal larger than rats or mice, thus allowing repeated blood sampling. Using this assay we investigated the short-term effects of insulin, adrenaline and glucagon (all modulators of blood glucose) on plasma leptin levels. Antiserum to bovine recombinant leptin (brLeptin) raised in chickens was used to develop a competitive ELISA. Using brLeptin as standard, the assay has a sensitivity of 0. 5 ng/ml with inter- and intra-assay variation of 15% and 7% respectively. The cross-reactivity of human recombinant leptin was 36.5%, while mouse leptin showed no cross-reactivity. Plasma samples from ewes, male castrate animals and rams (n=4-5) diluted in parallel to the standard with mean leptin concentrations of 6.0+/-2. 9, 3.3+/-0.4 and 3.1+/-1.3 ng/ml respectively. Leptin levels in rams were significantly lower than in ewes. The non-significant difference in leptin levels between rams and male castrate animals suggests that testosterone may not be responsible for the lower levels of leptin. Four groups of 3-4 ewes were given intravenous insulin (1 iu/kg), adrenaline (65 microg/kg), glucagon (24 iu/kg) or saline. Blood samples were taken at 1, 3, 5, 10, 20, 30, 60, 90 and 120 min after injection. As expected, glucose levels declined within 10 min of the insulin injection and rose after 3 min following both adrenaline and glucagon injections. Leptin levels, however, remained relatively unchanged for the 2 h following the treatments. Finally, a bolus intravenous dose of glucose (240 mg/kg) was given and sequential blood samples taken. Despite plasma glucose levels rising to over 200 mg/dl, leptin levels did not significantly change over the three hours following treatment. These data indicate that plasma leptin levels in sheep, in contrast to rodents, are not responsive to short-term changes in blood glucose or insulin, as has been shown in humans.
K Kauter, M Ball, P Kearney, R Tellam and McFarlane JR
G. R. MILNER, M. DE GASPARO, R. KAY and R. D. G. MILNER
Foetal rat pancreatic rudiments explanted on day 14 of gestation were grown for 6 days in organ culture in medium containing glucose (5·5 (1G) or 16·5 (3G) mmol/l) and amino acids at the 'physiological' (1AA) or seven times the 'physiological' (7AA) concentration. Cultures were also performed in medium to which zinc sulphate had been added at 10−7 to 10−5 mol/l concentration. At the end of the period of culture the diameters of insulin, glucagon and zymogen granule profiles in the rudiments were compared with those in normal 20-day foetal pancreas by quantitative morphology. The β cell volume, the number of granules per β cell, the insulin granular volume fraction and the area of insulin granule core and halo were also measured under selected experimental conditions.
Zymogen granule profiles were largest in vivo, intermediate in diameter when grown in 1G × 7AA medium and smallest in 1G × 1AA medium. The mean diameter of glucagon granule profiles remained constant for growth in vivo, in 1G × 1AA medium or in 1G × 7AA medium. Insulin granule profiles were largest in 1G ×7AA medium, smallest in 3G × 1AA medium and of intermediate diameter in vivo. Amino-acid enrichment increased the diameter of insulin granules and glucose enrichment decreased it. The addition of zinc to the culture medium had no effect on insulin granule diameter. In 1G × 7AA cultures the β cells were of similar size to those in vivo, but there were 29% fewer insulin granules per cell. The increased size of the insulin granules in 1G × 7AA cultures resulted in the insulin granule volume fraction in 1G × 7AA being 17·6 compared with 10·8% in vivo. Insulin granule cores were made larger by amino-acid enrichment of the culture medium but they were unaffected by glucose. The haloes were larger in 7AA medium and smaller in 3G medium. Glucose and amino-acid enrichment had a significant interaction on halo area, the mean area in 3G × 7AA medium being less than would have been expected from the summation of the effects of the two conditions.
E Bojanowska and B Stempniak
To date, glucagon-like peptide 1(7-36) amide (tGLP-1) has been found to affect the neurohypophysial and cardiovascular functions in normotensive and normovolaemic rats. The aim of the present study was to investigate possible effects of tGLP-1 on the mean arterial blood pressure and the release of vasopressin and oxytocin under conditions of blood volume depletion in the rat. In the first series of experiments, the animals were injected i.p. with either 0.15 M saline or 30% polyethylene glycol (PEG). PEG caused an 18% reduction of blood volume 1 h after injection. No significant changes in the mean arterial blood pressure were found in either normo- or hypovolaemic rats during the experiment. tGLP-1 injected i.c.v. at a dose of 1 microg/5 microl 1 h after the i.p. injection increased similarly the arterial blood pressure in normo- and hypovolaemic rats. The plasma vasopressin/oxytocin concentrations were markedly elevated in hypovolaemic animals and tGLP-1 further augmented the release of both hormones. In the second study, hypovolaemia was induced by double blood withdrawal. The haemorrhage resulted in a marked decrease of the mean arterial blood pressure and in the elevated plasma vasopressin/oxytocin concentrations. tGLP-1 injected immediately after the second blood withdrawal increased the arterial blood pressure. In parallel, tGLP-1 enhanced significantly vasopressin and oxytocin secretion when compared with haemorrhaged, saline-injected rats. The results of this study indicate that tGLP-1 may affect the arterial blood pressure and the secretion of neurohypophysial hormones under pathological conditions brought about by blood volume depletion.
S Saifia, AM Chevrier, A Bosshard, JC Cuber, JA Chayvialle and J Abello
The neuropeptide galanin is widely distributed in the gastrointestinal tract and exerts several inhibitory effects, especially on intestinal motility and on insulin release from pancreatic beta-cells. The presence of galanin fibres not only in the myenteric and submucosal plexus but also in the mucosa, prompted us to investigate the regulatory role of galanin, and its mechanism of action, on the secretion of the insulinotropic hormone glucagon-like peptide-1 (GLP-1). Rat ileal cells were dispersed through mechanical vibration followed by moderate exposure to hyaluronidase, DNase I and EDTA, and enriched for L-cells by counterflow elutriation. A 6- to 7-fold enrichment in GLP-1 cell content was registered after elutriation, as compared with the crude cell preparation (929 +/- 81 vs 138 +/- 14 fmol/10(6) cells). L-cells then accounted for 4-5% of the total cell population. Bombesin induced a time-(15-240 min) and dose- (0.1 nM-1 microM) dependent release of GLP-1. Glucose-dependent insulinotropic peptide (GIP, 100 nM), forskolin (10 microM) and the phorbol ester 12-0-tetradecanoylphorbol-13-acetate (TPA, 1 microM) each stimulated GLP-1 secretion over a 1-h incubation period. Galanin (0.01-100 nM) induced a dose-dependent inhibition of bombesin- and of GIP-stimulated GLP-1 release (mean inhibition of 90% with 100 nM galanin). Galanin also dose-dependently inhibited forskolin-induced GLP-1 secretion (74% of inhibition with 100 nM galanin), but not TPA-stimulated hormone release. Pretreatment of cells with 200 ng/ml pertussis toxin for 3 h, or incubation with the ATP-sensitive K+ channel blocker disopyramide (200 microM), prevented the inhibition by galanin of bombesin- and GIP-stimulated GLP-1 secretion. These studies indicate that intestinal secretion of GLP-1 is negatively controlled by galanin, that acts through receptors coupled to pertussis toxin-sensitive G protein and involves ATP-dependent K+ channels.
N Dachicourt, P Serradas, D Bailbe, M Kergoat, L Doare and B Portha
The effects of glucagon-like peptide-1(7-36)-amide (GLP-1) on cAMP content and insulin release were studied in islets isolated from diabetic rats (n0-STZ model) which exhibited impaired glucose-induced insulin release. We first examined the possibility of re-activating the insulin response to glucose in the beta-cells of the diabetic rats using GLP-1 in vitro. In static incubation experiments, GLP-1 amplified cAMP accumulation (by 170%) and glucose-induced insulin release (by 140%) in the diabetic islets to the same extent as in control islets. Using a perifusion procedure, GLP-1 amplified the insulin response to 16.7 mM glucose by diabetic islets and generated a clear biphasic pattern of insulin release. The incremental insulin response to glucose in the presence of GLP-1, although lower than corresponding control values (1.56 +/- 0.37 and 4.53 +/- 0.60 pg/min per ng islet DNA in diabetic and control islets respectively), became similar to that of control islets exposed to 16.7 mM glucose alone (1.09 +/- 0.15 pg/min per ng islet DNA). Since in vitro GLP-1 was found to exert positive effects on the glucose competence of the residual beta-cells in the n0-STZ model. we investigated the therapeutic effect of in vivo GLP-1 administration on glucose tolerance and glucose-induced insulin release by n0-STZ rats. An infusion of GLP-1 (10 ng/min per kg; i.v.) in n0-STZ rats enhanced significantly (P < 0.01) basal plasma insulin levels, and, when combined with an i.v. glucose tolerance and insulin secretion test, it was found to improve (P < 0.05) glucose tolerance and the insulinogenic index, as compared with the respective values of these parameters before GLP-1 treatment.
BD Green, MH Mooney, VA Gault, N Irwin, CJ Bailey, P Harriott, B Greer, FP O'Harte and PR Flatt
Glucagon-like peptide-1(7-36)amide (GLP-1) possesses several unique and beneficial effects for the potential treatment of type 2 diabetes. However, the rapid inactivation of GLP-1 by dipeptidyl peptidase IV (DPP IV) results in a short half-life in vivo (less than 2 min) hindering therapeutic development. In the present study, a novel His(7)-modified analogue of GLP-1, N-pyroglutamyl-GLP-1, as well as N-acetyl-GLP-1 were synthesised and tested for DPP IV stability and biological activity. Incubation of GLP-1 with either DPP IV or human plasma resulted in rapid degradation of native GLP-1 to GLP-1(9-36)amide, while N-acetyl-GLP-1 and N-pyroglutamyl-GLP-1 were completely resistant to degradation. N-acetyl-GLP-1 and N-pyroglutamyl-GLP-1 bound to the GLP-1 receptor but had reduced affinities (IC(50) values 32.9 and 6.7 nM, respectively) compared with native GLP-1 (IC(50) 0.37 nM). Similarly, both analogues stimulated cAMP production with EC(50) values of 16.3 and 27 nM respectively compared with GLP-1 (EC(50) 4.7 nM). However, N-acetyl-GLP-1 and N-pyroglutamyl-GLP-1 exhibited potent insulinotropic activity in vitro at 5.6 mM glucose (P<0.05 to P<0.001) similar to native GLP-1. Both analogues (25 nM/kg body weight) lowered plasma glucose and increased plasma insulin levels when administered in conjunction with glucose (18 nM/kg body weight) to adult obese diabetic (ob/ob) mice. N-pyroglutamyl-GLP-1 was substantially better at lowering plasma glucose compared with the native peptide, while N-acetyl-GLP-1 was significantly more potent at stimulating insulin secretion. These studies indicate that N-terminal modification of GLP-1 results in DPP IV-resistant and biologically potent forms of GLP-1. The particularly powerful antihyperglycaemic action of N-pyroglutamyl-GLP-1 shows potential for the treatment of type 2 diabetes.
R. M. Elliott, L. M. Morgan, J. A. Tredger, S. Deacon, J. Wright and V. Marks
The acute effects of different macronutrients on the secretion of glucagon-like peptide-1(7–36)amide (GLP-1(7–36)amide) and glucose-dependent insulinotropic polypeptide (GIP) were compared in healthy human subjects. Circulating levels of the two hormones were measured over a 24-h period during which subjects consumed a mixed diet. In the first study, eight subjects consumed three equicaloric (375 kcal) test meals of carbohydrate, fat and protein. Small increases in plasma GLP-1(7–36) amide were found after all meals. Levels reached a maximum 30 min after the carbohydrate and 150 min after the fat load. Ingestion of both carbohydrate and fat induced substantial rises in GIP secretion, but the protein meal had no effect. In a second study, eight subjects consumed 75 g glucose or the equivalent portion of complex carbohydrate as boiled brown rice or barley. Plasma GIP, insulin and glucose levels increased after all three meals, the largest increase being observed following glucose and the smallest following the barley meal. Plasma GLP-1(7–36)amide levels rose only following the glucose meal. In the 24-h study, plasma GLP-1(7–36)amide and GIP concentrations were increased following every meal and remained elevated throughout the day, only falling to fasting levels at night. The increases in circulating GLP-1(7–36)amide and GIP levels following carbohydrate or a mixed meal are consistent with their role as incretins. The more sustained rises observed in the daytime during the 24-h study are consistent with an anabolic role in lipid metabolism.
Journal of Endocrinology (1993) 138, 159–166
Y Watanabe, K Kawai, S Ohashi, C Yokota, S Suzuki and K Yamashita
To examine the structure–activity relationships in the insulinotropic activity of glucagon-like peptide-1(7–36) amide (GLP-1(7–36)amide), we synthesized 16 analogues, including eight which were designed by amino acid substitutions at positions 10 (Ala10), 15 (Serl5), 16 (Tyr16), 17 (Arg17), 18 (Lys18), 21 (Gly21), 27 (Lys27) and 31 (Asp31) of GLP-1(7–36)amide with an amino acid of GH-releasing factor possessing only slight insulinotropic activity, and three tentative antagonists including [Glu15]-GLP-1(8–36)amide. Their insulinotropic activities were assessed by rat pancreas perfusion experiments, and binding affinity to GLP-1 receptors and stimulation of cyclic AMP (cAMP) production were evaluated using cultured RINm5F cells.
Insulinotropic activity was estimated as GLP-1(7–36)amide = Tyr16>Lys18, Lys27>Gly21>Asp31⪢Ser15,Arg17>Ala10⪢GRF>[Glu15]-GLP-1(8–36) amide. Displacement activity against 125I-labelled GLP-1 (7–36)amide binding and stimulatory activity for cAMP production in RINm5F cells correlated well with their insulinotropic activity in perfused rat pancreases.
These results demonstrate that (1) positions 10 (glycine), 15 (aspartic acid) and 17 (serine) in the amino acid sequence of GLP-1(7–36)amide, in addition to the N-terminal histidine, are essential for its insulinotropic activity through its binding to the receptor, (2) the amino acid sequences for the C-terminal half of GLP-1(7–36)amide also contribute to its binding to the receptor, although they are less important compared with those of the N-terminal half, and (3) [Glu15]-GLP-1(8–36)amide is not an antagonist of GLP-1(7–36)amide as opposed to des-His1 [Glu9]-glucagon amide which is a potent glucagon antagonist.
Journal of Endocrinology (1994) 140, 45–52
P-M Bourlon, A Faure-Dussert, B Billaudel, B Ch J Sutter, G Tramu and M Thomasset
The pancreatic B cell is equipped with specific receptors for 1,25-dihydroxyvitamin D3 (1,25-(OH)2D3) and contains vitamin D-dependent calcium binding proteins (calbindin-D). Insulin secretion is impaired by vitamin D deficiency and is restored by 1,25-(OH)2D3 (concomitantly with an improved calcium handling within B cells) but the effect of 1,25-(OH)2D3 on the pancreatic B cell via calbindin-D is unclear. Therefore we examined the relationship between calbindin-D28K or calbindin-D9K and the activity of the endocrine pancreas in normal (N), four week vitamin D-deficient (−D) and one week 1,25-(OH)2D3-replete (+D) rats. Calbindin-D9K was not found in the pancreas, neither in the islets nor in the exocrine part, of any of the groups of rats (N, −D, or +D). Surprisingly, total islet calbindin-D28K content was increased by vitamin D deficiency and partly restored by 1,25-(OH)2D3. Calbindin-D28K immunostaining was observed only on A and B cells in the endocrine part of the pancreas, the greatest staining being found in A cells. This difference in staining density was increased by vitamin D deficiency and decreased by 1,25-(OH)2D3 treatment. In vitro, 1,25-(OH)2D3 also produced a negative influence on calbindin-D28K staining in A cells, as demonstrated using pieces of pancreas incubated with the steroid for 2 h. No significant influence on labeling intensity of B cell calbindin-D28K could be shown. Plasma insulin and islet insulin release in response to 10 mm arginine stimulation were decreased in −D rats and enhanced in +D rats towards N values. In contrast, plasma glucagon and the amount of glucagon secretion, stimulated in vitro by 10 mm arginine or by low (1·7 mm) glucose concentration, was increased in −D rats and attenuated by 1,25-(OH)2D3.
Thus there appears to be no relationship between the steady state level of B cell calbindin-D28K and the regulation of insulin secretion by 1,25-(OH)2D3 in vitamin D-deficient rats. However there is a correlation between A cell calbindin-D28K and glucagon secretion, which are both negatively regulated by 1,25-(OH)2D3. The predominance of calbindin-D28K in A cells raises the question as to how A and B cells interact and the role of calbindin-D28K in calcium handling.
Journal of Endocrinology (1996) 148, 223–232
P. Langer, O. Földes and K. Gschwendtová
Polyethylene tubes were inserted into the bile duct and femoral vein of rats under pentobarbital anaesthesia and bile was collected for three 2-h periods. After the first (control) period the animals were infused intravenously at a rate of 1·2 ml/h with the following compounds: (1) 0·9% (w/v) NaCl (control group), (2) glucagon (1200 ng/h), (3) vasopressin (1200 ng/h) or (4) angiotensin II (600 ng/h). The concentrations of thyroxine (T4), tri-iodothyronine (T3) and reverse tri-iodothyronine (rT3) in the bile were estimated by radioimmunoassay. No significant differences between groups were found in the biliary excretion of T4 and T3, while the excretion of rT3 after the infusion of all the hormones used was significantly (P < 0·001 at 2 to 4 h of the infusion) increased, no such increase being found in the controls. It may be concluded therefore that the administration of the above hormones resulted in some changes in iodothyronine metabolism in the liver. These may be explained by an inhibition of iodothyronine 5′-monodeiodination related to the glycogenolytic and gluconeogenetic effects of these hormones.
Journal of Endocrinology (1989) 121, 299–302