John I. Stagner

Sustained Oscillations of Insulin, Glucagon, and Somatostatin from the Isolated Canine Pancreas during Exposure to a Constant Glucose Concentration

Canine pancreata were perfused in vitro to examine whether hormone cycles could be demonstrated without hepatic or central nervous influence. Insulin, glucagon, and somatostatin demonstrated regular sustained cyclic secretion from the in vitro canine pancreas. Oscillations were noted for over 200 min during the infusion of a constant glucose concentration. Insulin demonstrated a 10-min period with a range of 8-12 min/cycle. Somatostatin had a 10-min period with a range of 8-11 min. Glucagon had a period of 8.6 min with range of 6-10 min. These periods do not allow glucagon to be consistently 90 degrees out of phase with insulin and somatostatin. When glucose was increased from 88 to 200 mg/dl, insulin cycles persisted but on an elevated base line, demonstrating that cycles react to glucose changes but are not dependent upon them. Cycles were disrupted by infusions of dopamine, apomorphine, epinephrine, and acetylcholine, but were reestablished. Autonomic blockade by both single and combined infusions of atropine (cholinergic), propranolol, and dibenzyline (adrenergic) had no effect on cycles. These results suggest that, in vitro, there is an intrinsic rhythm of hormone secretion by the pancreas despite a constant glucose level. The production of in vitro cycles requires the presence of a driving oscillator or pacemaker within the pancreas and the coordination of islets by pace-maker-islet communication, presumably by a non-adrenergic neural system. In vitro oscillations may Indicate that the pancreas is the driver or Zeitgeber of in vivo glucose-insulin cycles.

The order of islet microvascular cellular perfusion is B----A----D in the perfused rat pancreas.

In order to determine whether microvascular blood flow is important in the regulation of intra-islet cellular interactions, rat pancreata were isolated and perfused in vitro, both anterogradely or retrogradely, with and without anti-insulin or anti-somatostatin gamma-globulin. Expressed as percent change, anterograde infusion of insulin antibody increased efflux concentrations of glucagon (110 +/- 20%, P less than 0.0005) and somatostatin (2,112 +/- 73%, P less than 0.0005) above their respective control. Retrograde infusion of insulin antibody did not affect efflux concentrations of glucagon (P less than 0.50) or somatostatin (P less than 0.50). The anterograde infusion of anti-somatostatin antibody had no effect upon insulin (P less than 0.50) or glucagon (P less than 0.50) efflux concentrations, whereas retrograde anti-somatostatin antibody infusion produced immediate increases in efflux concentrations of both insulin (115 +/- 33%, P less than 0.0005) and glucagon (77 +/- 8%, P less than 0.0005). These results strongly suggest that (a) the vascular compartment is important in the regulation of intra-islet cellular interactions and further suggest that (b) the order of islet cellular perfusion and interaction is from the B cell core outward to the mantle, and (c) the mantle is further subordered with the majority of D cells downstream or distal to the majority of A cells. Thus, in the vascular compartment, B cells inhibit A-cell secretion and A cells stimulate D-cell secretion.

The Vascular Order of Islet Cellular Perfusion in the Human Pancreas

The vascular order of pancreatic islet cellular perfusion is important in the intraislet regulation of hormone secretion. Establishment of the sequence of interaction is fundamental to understanding the physiology and pathophysiology of the human islet. Intraislet insulin from the β-cell regulates both net hormone secretion and pulsatile secretion from α- and δ-cells. In terms of vascular perfusion, the δ-cell is perfused last and does not directly affect α- or β-cells in humans.

α-Adrenergic Blockade Improves Glucose-Potentiated Insulin Secretion in Non-Insulin-Dependent Diabetes Mellitus

The impairment of glucose-potentiated insulin secretion present in non-insulin-dependent diabetes mel I it us (NIDDM) can be approximated in normal subjects by an epinephrine infusion. Therefore, we sought to determine the role of the endogenous sympathetic nervous system in glucose-potentiated insulin secretion in both NIDDM (n = 6) and normal (n = 6) subjects. Glucose-potentiated insulin secretion was calculated as the slope of the curve relating increasing ambient glucose levels to the acute insulin response to an intravenous pulse of 5 g of L-arginine. Glucose-potentiated insulin secretion was determined on separate days during α-, β-, and combined α- plus β-adrenergic blockade and compared with a saline control. In. normal subjects, there was no effect of α-, β-, or α- plus β-blockade on the slope of glucose potentiation. In NIDDM, the initially decreased slope of glucose potentiation (0.25 ± 0.06 μU · ml−1 · mg−1 · dl, mean ± SE; P < .01) was not affected by β-blockade but increased during α-blockade (0.91 ± 0.22 μU · ml−1 · mg−1 · dl; P < .05). However, this improvement was abolished by combined α- plus β-blockade (0.32 ± 0.07 μU · ml−1 · mg−1 dl). Plasma norepinephrine was increased above basal levels in both normal (+ 260 ± 89 pg/ml) and NIDDM (+ 438 ± 162 pg/ml) subjects during α-blockade (P < .05 for both). This increase in plasma norepinephrine strongly suggests that there is an increase in synaptic cleft norepinephrine concentration. These results suggest that the increase in glucose-potentiated insulin secretion during α-adrenergic blockade was a result of both a decrease in an endogenous overactive α-adrenergic stimulation and an increase in synaptic cleft norepinephrine levels, which resulted in an increase in islet β-adrenergic stimulation. We suggest that a chronic decrease in islet α-adrenergic stimulation may prove to be a useful adjunct to NIDDM management.

β → α → δ Pancreatic Islet Cellular Perfusion in Dogs

Intraislet communication between β-, α-, and δ-cells and their secretory products may theoretically occur via the paracrine (interstitial) and/or vascular routes. Recently, we have shown that there is a directed microvascular circulation in the rat islet with a cellular order of perfusion of β → α → δ. The direction of microvascular perfusion of cells within the dog islet has been controversial. Anterograde (arterial) perfusion and retrograde (reversed or venous) perfusion of a segment of isolated dog pancreas with potent insulin antibodies yielded results similar to those found in the rat pancreas (anterograde, 158 ±44% increase in glucagon and 65 ± 20% increase in somatostatin; retrograde, no change in glucagon or somatostatin). Anterograde infusion of glucagon antibody (no change in insulin, −33.5 ± 3% decrease in somatostatin) or somatostatin antibody (no change in insulin or glucagon) also yielded the same results as in the rat pancreas. Anterograde infusion of 500 pg/ml glucagon caused a larger increase in insulin secretion (245 ± 10%) than retrograde infusion (45 ± 4%), whereas somatostatin was stimulated more retrogradely (339 ± 17%) than anterogradely (121 ± 9%). Anterograde infusion of somatostatin produced a larger decrease in insulin and glucagon than did retrograde perfusion (P < .0001 for both comparisons). The retrograde infusion of 0.3 mU/ml insulin caused a decrease in glucagon but was without effect anterogradely. The results from the infusion of exogenous hormones suggest that the sensitivity of the α-, β-, and δ-cells to insulin, glucagon, and somatostatin is determined by the β → α → δ order of perfusion. The antibody studies indicate that directed microvascular perfusion is central to intraislet regulation of insular secretions in dogs.

beta----alpha----delta pancreatic islet cellular perfusion in dogs.

Intraislet communication between alpha-, beta-, and delta-cells and their secretory products may theoretically occur via the paracrine (interstitial) and/or vascular routes. Recently, we have shown that there is a directed microvascular circulation in the rat islet with a cellular order of perfusion of beta----alpha----delta. The direction of microvascular perfusion of cells within the dog islet has been controversial. Anterograde (arterial) perfusion and retrograde (reversed or venous) perfusion of a segment of isolated dog pancreas with potent insulin antibodies yielded results similar to those found in the rat pancreas (anterograde, 158 +/- 44% increase in glucagon and 65 +/- 20% increase in somatostatin; retrograde, no change in glucagon or somatostatin). Anterograde infusion of glucagon antibody (no change in insulin, -33.5 +/- 3% decrease in somatostatin) or somatostatin antibody (no change in insulin or glucagon) also yielded the same results as in the rat pancreas. Anterograde infusion of 500 pg/ml glucagon caused a larger increase in insulin secretion (245 +/- 10%) than retrograde infusion (45 +/- 4%), whereas somatostatin was stimulated more retrogradely (339 +/- 17%) than anterogradely (121 +/- 9%). Anterograde infusion of somatostatin produced a larger decrease in insulin and glucagon than did retrograde perfusion (P less than .0001 for both comparisons). The retrograde infusion of 0.3 mU/ml insulin caused a decrease in glucagon but was without effect anterogradely. The results from the infusion of exogenous hormones suggest that the sensitivity of the alpha-, beta-, and delta-cells to insulin, glucagon, and somatostatin is determined by the beta----alpha----delta order of perfusion.(ABSTRACT TRUNCATED AT 250 WORDS)

Local growth factors are beneficial for the autonomic reinnervation of transplanted islets in rats

Introduction Transplanted islets, being avascular and denervated, receive blood vessels and nerves from the recipient. Reinnervation may account in part for the normalization of islet function in islet transplants. Whether reinnervation is possible to augment is not known. Aims and Methodology To explore whether reinnervation of transplanted islets is augmented by local addition of growth factors to the graft, syngeneic islets were transplanted to the pancreas of streptozotocin-diabetic Lewis rats with or without pellets locally releasing nerve growth factor (NGF) and vascular endothelial growth factor (VEGF), alone or in combination. The pellets released growth factors for 14 days at a rate of 20 ng/day. After 7 weeks, pancreatic tissue was processed for immunofluorescence of insulin and the neural markers neuropeptide Y (NPY) and tyrosine hydroxylase (TH). Results Islets were larger and more numerous after treatment with NGF (p = 0.024) and with NGF in combination with VEGF (p = 0.044). Similarly, immunostaining for TH and the C-terminal flanking peptide of NPY (C-PON) was more pronounced after treatment with NGF in combination with VEGF than in controls (both p < 0.05). Conclusion Local growth factor treatment has a beneficial effect on autonomic reinnervation as well as islet integrity and survival of the graft after islet transplantation in rats.

The anterograde and retrograde infusion of glucagon antibodies suggests that A cells are vascularly perfused before D cells within the rat islet

SummaryWe have suggested that the order of cellular vascular perfusion within the islet is important in the regulation of islet hormone secretion. Anatomically, the A and D cells appear to be randomly dispersed throughout the mantle. Although islet capillary blood flow is known to be from the B-cell core to the A- and D-cell mantle, it has not yet been established whether the cells of the mantle may influence one another vascularly. Rat pancreata were perfused in vitro anterogradely and retrogradely with or without glucagon antibody in order to determine the order of cellular perfusion and interaction between the A and D cells in the islet mantle. Anterograde infusion of glucagon antibody did not affect insulin secretion, but rapidly decreased somatostatin secretion −46±8%, (p<0.005). Retrograde infusion of glucagon anti body decreased insulin secretion (−27±8%, p<0.005) but had no effect upon somatostatin secretion. This study not only confirms a core to mantle islet perfusion but also establishes that the A cell precedes the D cell in the terms of vascular perfusion. Thus within the islet, vascular borne insulin regulates the release of glucagon, which in turn, regulates the release of somatostatin. Somatostatin is vascularly neutral owing to its downstream position in the sequence (B to A to D) of cellular perfusion.

Modulation of Insulin Secretion by Pancreatic Ganglionic Nicotinic Receptors

Autonomie ganglia may be regulated, in part, by nicotinic receptors. To test whether basal insulin secretion may be modulated by an endogenous pancreatic ganglionic mechanism, the effects of ganglionic pre- and postsynaptic nicotinic receptor antagonism were studied in the in vitro canine pancreas. Combined infusion of atropine, phentolamine, and propranolol had no affect on insulin secretion (P < .30). Presynaptic nicotinic receptor blockade by β-bungarotoxin (β-BuTX) in combination with atropine and phentolamine reduced mean insulin secretion (78 ± 18 U/ml, P < .0025) from preinfusion concentrations (287 ± 43 U/ml). The decrease in insulin secretion resulting from BuTX, atropine, and phentolamine was prevented by the addition of either specific postsynaptic nicotinic receptor blockade by α-bungarotoxin (P < .05) or propranolol (P < .005). Because it is known that postsynaptic nicotinic receptor agonism may stimulate the intragan-glionic release of norepinephrine, these results suggest that nicotinic receptors are present at the ganglionic level in the pancreas and modulate insulin secretion by a complex intraganglionic mechanism. The postulated ganglionic nicotinic receptor-mediated mechanism may operate by the interaction of a β-adrenergic inhibitory component, which may be activated by intraganglionic norepinephrine, and a stimulatory nonmuscar-inic nonadrenergic (possibly peptidergic) component, which may be activated in the absence of intraganglionic norepinephrine.

Induction of angiogenesis by growth factors: relevance to pancreatic islet transplantation

Biodegradable pellets releasing 20 ng/day of endothelial cell growth factor alpha (alpha ECGF) or a- or b-fibroblast growth factor (FGF) and 90 micrograms/day of heparin were implanted beneath the renal capsule in rats and dogs and the muscularis/serosal border of the pyloric stomach in dogs to test for angiogenesis in a potential pancreatic islet transplant site. These factors were also tested in vitro to determine whether the capillary bed of the isolated islet could be preserved. alpha ECGF was superior to a- or bFGF in promoting endothelial cell growth and capillary formation in isolated islets. Both a- or bFGF and alpha ECGF induced the development of a dense capillary bed in the dog stomach, whereas in the kidney site alpha ECGF was more effective in the rat than was a- or bFGF. Priming the isolated islet as well as the transplant site prior to islet transplantation resulted in islet blood flow being established within 3 days in contrast to 7-14 days in controls.