The nature of big plasma somatostatin: implications for the measurement of somatostatin-like immunoreactivity in human plasma

The Enigmatic N-Terminal Domain of Proglucagon; A Historical Perspective

Enteroglucagon refers to the predominant peptide with glucagon-like immunoreactivity (GLI) that is released by the intestine into the circulation in response to nutrients. Development of a radioimmunoassay for glucagon revealed issues that were not apparent in applications of the insulin radioimmunoassay. The fact that some antisera raised against glucagon recognized glucagon-related peptides in extracts of both pancreas and gut whereas others recognized only components in the pancreas remained a mystery until it was realized that the "gut GLI cross-reactive" antisera were directed against an epitope in the N-terminal to central region of glucagon whereas the "pancreatic glucagon specific" antisera were directed against an epitope in the C-terminal region. Unlike the cross-reactive antisera, the glucagon specific antisera did not recognize components in which glucagon was extended from its C-terminus by additional amino acids. Initial attempts to purify enteroglucagon from porcine ileum led to the erroneous conclusion that enteroglucagon comprised 100 amino acids with an apparent molecular mass of 12,000 Da and was consequently given the name glicentin. Subsequent work established that the peptide constituted residues (1-69) of proglucagon (M_r 8128₎. In the 40 years since the structural characterization of glicentin, attempts to establish an unambiguous physiological function for enteroglucagon have not been successful. Unlike the oxyntomodulin domain at the C-terminus of enteroglucagon, the primary structure of the N-terminal domain (glicentin-related pancreatic peptide) has been poorly conserved among mammals. Consequently, most investigations of the bioactivity of porcine glicentin may have been carried out in inappropriate animal models. Enteroglucagon may simply represent an inactive peptide that ensures that the intestine does not release equimolar amounts of a hyperglycemic agent (glucagon) and a hypoglycemic agent (GLP-1) after ingestion of nutrients.

Comparison of incretin immunoassays with or without plasma extraction: Incretin secretion in Japanese patients with type 2 diabetes

Aims/Introduction:  The effectiveness of incretin‐based therapies in Asian type 2 diabetes requires investigation of the secretion and metabolism of glucose‐dependent insulinotropic polypepide (GIP) and glucagon‐like peptide 1 (GLP‐1). Plasma extractions have been suggested to reduce variability in intact GLP‐1 levels among individuals by removing interference that affects immunoassays, although no direct demonstration of this method has been reported. We have evaluated the effects of ethanol and solid‐phase extractions on incretin immunoassays. We determined incretin levels during meal tolerance tests in Japanese patients with type 2 diabetes and characterized predictors for incretin secretion.

Materials and Methods:  Japanese patients with type 2 diabetes (23 anti‐diabetic drug‐naïve and 18 treated with sulfonylurea [SU] alone) were subjected to meal tolerance tests, and incretin levels were determined by immunoassays with or without extraction.

Results:  Intact GLP‐1 levels determined by an intact GLP‐1 immunoassay with ethanol and solid‐phase extractions were lower than those determined without extraction. Intact GLP‐1 levels determined by the extractions were highly correlated with each other, much more so than the levels with and without extraction. Total GLP‐1 was unaffected by extractions, showing that extractions remove interference only in the case of intact GLP‐1. Incretin secretion after meal ingestion was similar between drug‐naïve and SU‐treated patients. Fasting and postprandial GLP‐1 levels were correlated positively with fasting free fatty acids and negatively with dipeptidyl peptidase‐4 activity.

Conclusions:  Ethanol and solid‐phase extractions remove interference for intact GLP‐1 immunoassay. SU showed little effect on incretin secretion. GLP‐1 and GIP secretion were predicted by different factors. (J Diabetes Invest, doi: 10.1111/j.2040‐1124.2011.00141.x, 2012)

Evidence that vasoactive intestinal polypeptide is a parasympathetic neurotransmitter in the endocrine pancreas in dogs

Vasoactive intestinal polypeptide (VIP) has been found in pancreatic nerves in several species. Studies were conducted to determine if VIP could be a parasympathetic neurotransmitter in the canine endocrine pancreas. To verify that VIP is localized in pancreatic parasympathetic nerves, sections of canine pancreas were immunostained for VIP. VIP staining was identified in the majority of neuronal cell bodies in intrapancreatic parasympathetic ganglia. In addition. VIP was localized in nerve fibers innervating pancreatic islets in the proximity of alpha cells. Next, to determine if VIP is released during electrical stimulation of parasympathetic nerves, pancreatic spillover of VIP was measured during vagal nerve stimulation (VNS) in anesthetized dogs. VIP spillover increased from a baseline of 630+/-540 pg/min to 2580+/-540 pg/min (delta = +1950+/-490 pg/min, p <0.01). Pancreatic VIP release during VNS was not affected by atropine, whereas ganglionic blockade with hexamethonium nearly abolished the VIP response to VNS (p<0.005 vs control), suggesting that VIP is a postganglionic neurotransmitter in the dog pancreas. To examine the effects of VIP on pancreatic hormone secretion, synthetic VIP was infused locally into the pancreatic artery. VIP, at a low dose (5 pmol/min), increased glucagon secretion from 1750+/-599 to 3800+/-990 pg/min (delta = +2060+/-870 pg/min, p<0.05), but did not affect insulin secretion (delta = -1030+/-760 microU/min, NS). Thus, VIP is contained in and released from pancreatic parasympathetic nerves in proximity to islet alpha cells and exogenous VIP, at a dose which approximates the increase of VIP spillover during VNS, preferentially stimulates glucagon vs insulin secretion. Therefore, VIP is likely to function as a parasympathetic neurotransmitter in the endocrine pancreas in dogs. We hypothesize that VIP could mediate the glucagon response to parasympathetic activation which has been shown to resistant to cholinergic blockade with atropine in several species.

Loxiglumide inhibits cholecystokinin stimulated somatostatin secretion and simultaneously enhances gastric acid secretion in humans

In vitro studies have demonstrated that cholecystokinin releases somatostatin from the gastric mucosa. To date, there is no information about the in vivo significance of this finding in man. Therefore, we have studied the effect of infusion of cholecystokinin resulting in plasma concentrations within the range found after meal-stimulation, on somatostatin release and on gastric acid secretion. In addition we have studied these functions during infusion of the type A cholecystokinin receptor antagonist loxiglumide. In eight healthy subjects, basal gastric acid secretion was distinctly stimulated by cholecystokinin. The effect of cholecystokinin on gastric acid secretion was markedly enhanced by loxiglumide. Cholecystokinin also significantly stimulated somatostatin output into the gastric lumen, but not into the systemic circulation. Somatostatin output into the gastric lumen during infusion of cholecystokinin was abolished by loxiglumide. The data indicate that on the one hand circulating cholecystokinin, like gastrin, stimulates gastric acid secretion probably by binding to less specific type B receptors on parietal cells that are not blocked by loxiglumide, but on the other hand that cholecystokinin, in contrast to gastrin, also inhibits gastric acid secretion probably by binding to specific type A receptors present on somatostatin producing D-cells in the gastric mucosa, that are blocked by loxiglumide.

Contribution of galanin to stress-induced impairment of insulin secretion in swimming mice

This study examines the potential role of the neuropeptide, galanin, in stress-induced inhibition of insulin secretion in swimming mice. Firstly, the pancreatic and adrenal content of galanin-like immunoreactivity was determined in mice after swimming stress. It was found that pancreatic content was significantly lower in stressed mice than in resting controls, both after 2 (P less than 0.05) and 6 (P less than 0.025) minutes of swimming, suggesting partial release of pancreatic galanin during stress. In contrast, the adrenal content of galanin-like immunoreactivity did not change during the swimming stress. Gel filtration of tissue extracts indicated that (1) mouse pancreas contains two forms of galanin-like immunoreactivity; one co-eluting with synthetic porcine galanin (centered on Kav of 0.70) and another with a larger molecular weight (centered on Kav of 0.30), and (2) mouse adrenal contains a small void volume-peak and a larger peak of immunoreactivity, the latter co-eluting with synthetic galanin. Secondly, the effects of swimming stress on plasma glucose and insulin levels were compared in mice that received high titre rabbit anti-galanin serum with those in mice receiving normal rabbit serum. In normal rabbit serum-pretreated swimming mice, glucose-induced insulin levels were only 50% of resting controls (P less than 0.01). Immunoneutralization of galanin with specific antiserum abolished this swimming stress-induced inhibition of glucose-stimulated insulin levels. This was accompanied by a modestly enhanced rate of glucose disappearance. These findings suggest that pancreatic galanin is released during swimming stress in mice and that endogenous galanin makes a major contribution to stress-induced impairment of insulin secretion.

Somatostatin cells in rat antral mucosa: qualitative and quantitative ultrastructural analyses in different states of gastric acid secretion

In the gastrointestinal tract somatostatin is localized in endocrine cells and in neurons. The antral somatostatin (D-) cell shares features of both cell types. The activity of the antral D-cell is regulated by intragastric pH. Therefore different states of gastric acidity were induced experimentally in order to study D-cell morphology at the electron microscopical level. The morphological findings were related to measurements of plasma and tissue concentrations of the peptide. The D-cell is characterized by extensive membrane interdigitations with neighbouring cells. Changes in the activity of antral D-cells are reflected by an increase in cytoplasmic secretory granule density and a shift of secretory granules towards basal cell processes. Direct endocrine cell contacts at the level of the perikarya were rarely observed. The intracellular distribution of secretory granules suggests that cell communication is more likely to take place at the level of the strongly immunoreactive cytoplasmic processes. No evidence for endocrine or exocrine (luminar) secretion was observed morphologically. This is in agreement with the concept of paracrine secretion of the antral D-cell.

Circulating prosomatostatin-derived peptides. Differential responses to food ingestion

Prosomatostatin (pro-S) and its bioactive posttranslational products, somatostatin-14 (S-14), somatostatin-13 (S-13), and somatostatin-28 (S-28), were measured in human plasma by the use of immunoglobulins to the NH2-terminus of S-28 conjugated with agarose to separate them and, thereafter, by RIA with an antiserum recognizing the COOH-terminus of pro-S, and by specific RIA for the NH2-terminus of S-14 and pro-S. In healthy men, mean basal levels of pro-S were 4 pg equivalent S-14/ml; S-14/S-13 combined were 9 pg equivalent S-14/ml; and S-28 levels were 16 pg/ml. After a 700-kcal meal, pro-S, S-14, and S-14/S-13 did not change, whereas S-28 levels doubled by 120 min and remained elevated for 240 min. To evaluate the origins of these peptides, their levels were compared in peripheral, portal, gastric, and mesenteric veins of anesthetized patients and in patients with total resection of stomach and pancreas before and after nutrient intake. The stomach and small intestine were sources of both peptides; however, most S-28 originated in the small intestine. These findings suggest that, in contrast to S-14, S-28 is a hormone and may modulate postprandial nutrient absorption and use.

Galanin release during pancreatic nerve stimulation is sufficient to influence islet function

To determine if galanin is released during pancreatic neural activation, we measured galanin-like immunoreactivity (GLIR) in pancreatic venous and peripheral arterial plasma during 10 min of electrical stimulation of the mixed autonomic pancreatic nerves in halothane-anesthetized dogs, using a sensitive and specific radioimmunoassay. During mixed pancreatic nerve stimulation (MPNS), pancreatic venous GLIR increased by 174 +/- 20 fmol/ml, whereas arterial GLIR did not change. By use of the arteriovenous concentration difference and measurements of pancreatic venous blood flow, pancreatic spillover of GLIR was calculated and found to increase by 640 +/- 90 fmol/min during MPNS. This MPNS inhibited the output of immunoreactive insulin (IRI; delta = -53 +/- 9%) and somatostatin-like immunoreactivity (SLI, delta = -49 +/- 13%) and stimulated that of immunoreactive glucagon (IRG, delta = +600 +/- 200%). To determine if the amount of GLIR released during MPNS was sufficient to elicit these changes of pancreatic hormone secretion, we compared the effect of MPNS on IRI, SLI, and IRG output with the effect of synthetic galanin infused directly into the pancreatic artery at a rate that reproduced the MPNS-induced spillover of GLIR. Exogenous infusion of synthetic galanin (2.7 pmol/min) increased pancreatic venous levels of GLIR by 169 +/- 38 fmol/ml, did not change arterial GLIR levels, and thus increased calculated spillover (appearance) by 550 +/- 160 fmol/min, which was nearly identical to the increment produced by MPNS. This matched infusion of galanin inhibited IRI (delta = -58 +/- 3%) and SLI output (delta = -35 +/- 3%) and modestly stimulated IRG output (delta = +62 +/- 10%).(ABSTRACT TRUNCATED AT 250 WORDS)

The effects of epinephrine on islet hormone secretion in the dog

We investigated the direct effects of physiological levels of epinephrine on the basal and arginine-stimulated secretion of insulin, glucagon, and somatostatin from the in situ pancreas in halothane-anaesthetized dogs. An IV infusion of 20 ng/kg/min of epinephrine increased plasma epinephrine levels to 918 +/- 103 pg/ml (P less than 0.001), and increased the baseline pancreatic output of insulin (P less than 0.05), glucagon (P less than 0.05) and somatostatin (P less than 0.05). The acute insulin response (AIR) to 2.5 g of arginine during this infusion of epinephrine was significantly higher (P less than 0.05) than in controls as were the acute glucagon response (AGR) (P less than 0.05) and the acute somatostatin response (ASLIR) (P less than 0.05). Plasma glucose levels increased slightly and transiently during infusion of epinephrine from 99 +/- 2 mg/dl to a maximum of 110 +/- 3 mg/dl (P less than 0.05). An IV infusion of 80 ng/kg/min of epinephrine produced plasma epinephrine levels of 2,948 +/- 281 pg/ml, and increased the baseline pancreatic output of insulin (P less than 0.05) and glucagon (P less than 0.05). In contrast, baseline somatostatin output decreased transiently during this high dose infusion of epinephrine. The AIR and ASLIR to arginine were both significantly lower (P less than 0.05) than those during the infusion of epinephrine at the low dose. The AGR to arginine remained potentiated (P less than 0.05). Plasma glucose levels increased from 99 +/- 3 mg/dl to 119 +/- 4 mg/dl (P less than 0.01). We conclude that the effect of epinephrine on islet hormone secretion is dependent on the plasma level of epinephrine. At stress levels of 900-1000 pg/ml, both insulin and somatostatin secretion are stimulated; only at near pharmacologic, or extreme stress levels, does epinephrine produce net inhibition.

Somatostatin response to glucose before and after prolonged fasting in lean and obese non-diabetic subjects

Insulin, glucagon, and somatostatin concentrations were measured in 7 lean and 7 obese non-diabetic subjects over 7 days of fasting. In addition each subject was given a 75 g oral glucose tolerance test after fasts of 12 h and 7 days. In lean subjects complete food deprivation induced a significant decrease in the circulating levels of both insulin and somatostatin, while glucagon nearly doubled by 48 h and then remained constant for the duration of starvation. Refeeding with oral glucose suppressed the increased plasma glucagon, but insulin and somatostatin responses were enhanced in comparison with the prefast values, as assessed by the integrated areas of change. In obese subjects peripheral insulin and somatostatin levels were significantly lowered, but plasma glucagon level was unchanged at the end of the starvation period. In the same group glucose-induced insulin and somatostatin release were greater than in the fed state. Suppression of plasma glucagon by glucose appeared less complete in obese than in lean subjects. It is concluded that prolonged starvation enhances D-cell responsiveness to glucose in lean and obese subjects.

Effects of pancreatic noradrenaline infusion on basal and stimulated islet hormone secretion in the dog

We investigated the direct pancreatic effects of noradrenaline in vivo on the secretion of insulin, glucagon, and somatostatin from the in situ pancreas in halothane-anaesthetized dogs. Noradrenaline was infused into the superior pancreatic artery at 12 ng min-1, a rate that did not alter systemic glucose or noradrenaline levels nor heart rate or blood pressure. This pancreatic infusion of noradrenaline did not affect the basal pancreatic output of insulin, yet did markedly inhibit arginine-stimulated insulin secretion. The acute insulin response (AIR) to an intravenous injection of arginine (2.5 g), which was 4293 +/- 1260 microM min-1 under control conditions, was reduced to 1054 +/- 396 microU min-1 by noradrenaline (P less than 0.01). Noradrenaline increased basal pancreatic glucagon output from 321 +/- 130 pg min-1 to 876 +/- 309 pg min-1 after 20 min of infusion (P less than 0.05) and the acute glucagon response (AGR) to arginine, being 1033 +/- 203 pg min-1 under control conditions and 1746 +/- 249 pg min-1 during noradrenaline infusion (P less than 0.05). The basal output of somatostatin did not change during noradrenaline infusion, but arginine-stimulated somatostatin secretion was impaired. The acute somatostatin response (ASLIR) to arginine was 473 +/- 124 fmol min-1 under control conditions and was decreased to 140 +/- 80 fmol min-1 by noradrenaline (P less than 0.01).(ABSTRACT TRUNCATED AT 250 WORDS)

Effect and mechanism of vagal nerve stimulation on somatostatin secretion in dogs

To investigate the effect of vagal nerve stimulation on the release of pancreatic somatostatin, we electrically stimulated (10 Hz, 5 ms, 13.5 mA, and 10 min) the thoracic vagi just below the heart in halothane anesthetized dogs (n = 15). The stimulation increased the pancreatic output of somatostatinlike immunoreactivity (SLI) (delta = +248 +/- 81 fmol/min, P less than 0.005; base-line levels = 455 +/- 150 fmol/min). min). Arterial plasma SLI levels increased as well (delta = +16 +/- 3 fmol/ml, P less than 0.001; base-line levels = 65 +/- 3 fmol/ml), reflecting stimulation of extrapancreatic SLI secretion. Significant vagal activation was verified by a fivefold increase of pancreatic output of pancreatic polypeptide (PP) (delta = +31.4 +/- 5.9 ng/min, P less than 0.001; base-line levels = 7.8 +/- 0.9 ng/min). Atropine pretreatment (n = 6) inhibited partially both the PP response (delta = +7.9 +/- 3.8 ng/min after atropine) and the pancreatic SLI response (delta = +92 +/- 29 fmol/min) to vagal nerve stimulation. However, atropine pretreatment did not modify the arterial SLI response (delta = +20 +/- 7 fmol/ml). Hexamethonium pretreatment (n = 9) completely abolished all three responses. We conclude that 1) electrical stimulation of the vagus stimulates pancreatic SLI, extrapancreatic SLI, and PP release in vivo in the dog; 2) both muscarinic and nonmuscarinic mechanisms mediate the PP and pancreatic SLI responses; 3) a nonmuscarinic mechanism mediates the extrapancreatic SLI response; and 4) all three responses are mediated via ganglionic nicotinic receptors.

Effect of caerulein administration on portal plasma somatostatin in dogs

The effect of administration of synthetic caerulein on canine plasma somatostatin concentration in the portal vein was investigated using specific radioimmunoassay of acid-acetone extracted plasma. Gel chromatography revealed that the main component of the plasma somatostatin in the portal vein was identical with somatostatin-14. After administration of caerulein (5 ng/kg), the somatostatin level increased significantly while the gastrin level remained unchanged. These results suggested that caerulein affects D cell function but not gastrin secretion.

Contribution of the pancreas to circulating somatostatin-like immunoreactivity in the normal dog

These studies were performed to assess the contribution of the pancreas to the somatostatin-like immunoreactivity (SLI) circulating in arterial and portal venous plasma. Basal SLI concentrations in arterial, pancreatic venous, and portal venous plasma were 95 +/- 9, 277 +/- 32, and 130 +/- 12 pg/ml, (means +/- SEM), respectively. Measurement of pancreatic and portal venous blood flow (5 +/- 1 vs. 365 +/- 46 ml/min) and hematocrit allowed calculation of net, base-line SLI output from the right lobe of the pancreas (521 +/- 104 pg/min) and from the gastrointestinal tract (8,088 +/- 1,487 pg/min), which suggested that the contribution of the pancreas to circulating SLI was minor when the D cells were not stimulated. To stimulate the secretion of SLI from both pancreatic and nonpancreatic sources, isoproterenol, a beta-adrenergic agonist, was infused intravenously for 1 h into six anesthetized dogs. Arterial SLI increased by 52 +/- 9 pg/ml; superior pancreatico-duodenal venous SLI increased by 380 +/- 95 pg/ml; portal venous SLI increased by 134 +/- 14 pg/ml. Pancreatic venous blood flow remained unchanged at 5 +/- 1 ml/min, but portal venous blood flow increased to 522 +/- 62 ml/min. SLI output from the right lobe of the pancreas increased by 684 +/- 227 pg/min and that from the gastrointestinal tract increased by 23,911 +/- 3,197 pg/min, again suggesting that the pancreas was a minor source of circulating SLI even when the D cells were stimulated. We conclude that the measurement of arterial-venous SLI concentrations, in the absence of measurements of organ blood flow, can give a false impression of the organ's contributions of circulating SLI. To verify that the contribution of the pancreas was negligible, six dogs received an acute pancreatectomy and then an intravenous infusion of isoproterenol at the same rate. In these dogs, both the base-line level of SLI in arterial plasma (109 +/- 12 pg/ml) and the increment during isoproterenol (56 +/- 8 pg/ml) were similar to those of normal dogs. Likewise, in pancreatectomized dogs both the base-line level of SLI in portal venous plasma (129 +/- 16 pg/ml) and the increment during isoproterenol (174 +/- 34 pg/ml) were similar to those of normal dogs. We conclude that, in normal dogs, the pancreas makes a negligible contribution to the basal and stimulated level of SLI in arterial and portal venous plasma and therefore that these levels should not be used as an index of secretory activity of the pancreatic D cells.