2'9 Illllllllllllllllll This is to certify that the thesis entitled Effects of Acetylcholine and Norepinephrine on Glucose-Induced Insulin Secretion from Ob/Ob and Lean Mouse Pancreatic Islets presented by Twylla M. Tassava has been accepted towards fulfillment of the requirements for Masters Human Nutrition degree in Mama Major professor Date g/Rgr/QC] 0-7639 MS U is an Affirmative Action/Equal Opportunity Institution LIBRARY Michigan State University '\ A PLACE IN RETURN BOX to remove this checkout from your record. TO AVOID FINES return on or before date due. DATE DUE DATE DUE DATE DUE fl =___=I_ lfifil MSU I. An Affirmative Action/Equal Opportunity lnditution emulation EFFECTS OF ACETYLCHOLINE AND NOREPINEPHRINE ON GLUCOSE-INDUCED INSULIN SECRETION FROM OB/OB AND LEAN MOUSE PANCREATIC ISLETS BY Twylla Tassava A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Food Science and Human Nutrition 1989 ABSTRACT Effects of Acetylcholine and Norepinephrine on Glucose- Induced Insulin Secretion from Ob/ob and Lean Mouse Pancreatic Islets BY Twylla Tassava Objectives of this study were to determine dynamics of 1) glucose-induced insulin secretion, and 2) acetylcholine potentiation and norepinephrine inhibition of glucose-induced insulin secretion, from pancreatic islets of 8-9 wk old ob/ob and lean mice. Ob/ob mouse islets were larger and were hypersensitive and hyperresponsive to glucose stimuli. In the presence of 15 mM glucose, ob/ob islets were 1) hyperresponsive but equally sensitive to acetylcholine stimulation, and 2) hypersensitive and hyperresponsive to norepinephrine inhibition compared to lean islets. In the presence of 5 mM glucose, acetylcholine potentiated insulin secretion from ob/ob but not lean islets. In ob/ob mice, both glucose and acetylcholine may be important contributing factors to hyperinsulinemia. In addition, there may be decreased sympathoadrenal inhibition of ob/ob islet B cells. Such decreased inhibition could contribute to hyperinsulinemia in ob/ob mice, at least prior to compensatory increases in islet sensitivity to catecholamines. I. A. B. Obesity C. LITERATURE REVIEW Introduction TABLE OF CONTENTS 1. Hyperinsulinemia in Ob/ob Mice. Ob/ob Mice as Models for Human Obesity D. Regulation of Plasma Insulin Concentration . . . . Physiology of Insulin Synthesis and Secretion . Glucose as a Regulator of Insulin Secretion . . F. II. A. B. C. 1. 2. 3. Neurotransmitters as Regulators of Insulin Secretion. 4. Insulin Clearance Relationship Between Central Nervous Hyperinsulinemia l. VMH- -Lesioned Rats and Obesity 2. 3. Ob/ob Mice. . Zucker (fa/fa) Rats . Hypothesis and Obiectives. MATERIALS AND METHODS. Animals and Materials. Islet Preparation. Experimental Design. System, 1. 2. General Design. . . . . Experiment 1 - Effects of Glucose on Insulin Secretion 3. 4. Experiment 2 - Effects of Neurotransmitters Glucose- Induced Insulin Secretion. . . . . . Experiment 3 - Effects of Neurotransmitter Antagonists on Insulin Secretion . D. E. III. RESULTS A. Experiment I Secretion Sample Analysis. Statistical Analysis Effects of Glucose on Insulin B. Experiment 2 Effects 9; Neurotransmitters 9Q Glucose-Induced Insulin Secretion C. Experiment ; - Effects 9; Neurotransmitter Antagonists 9Q Insulin Secretion. IV. DISCUSSION . . . . V. RECOMMENDATIONS FOR FUTURE EXPERIMENTS. A. Epinephrine in Obzob Mice B. C. D. E. VI. Interactive Effects 9: Neurotransmitters. Pancreatic Insulin 9f Pre-Obese Ob/ob Mice. Insulin Secretion from Pre-Obese Ob/ob Mouse Islets Islets 9f Adrenalectomized Ob/ob Mice REFERENCES . . . . \lmmubUP-‘l—‘H g.) ham 12 14 16 17 20 22 22 23 23 23 27 27 28 29 29 30 30 35 4O 42 51 51 52 52 52 53 54 Figure Figure Figure Figure Figure Figure Figure 1. 2. 3. 4. Plasma insulin and glucose of 8-9 wk old ob/ob LIST OF FIGURES and lean mice. Distribution of islet size from ob/ob and lean mice . . . . Islet and pancreatic insulin content from ob/ob and lean mice. Glucose-induced insulin secretion during 30 min incubation of 8 wk old ob/ob and lean mouse islets Insulin secretion from ob/ob and lean mouse islets in response to 15 mM glucose, or 5 mM glucose and acetylcholine. Insulin secretion from ob/ob and lean mouse islets in response to 15 mM glucose and norepinephrine . Insulin secretion from ob/ob and lean mouse islets in response to neurotransmitter antagonists. iv 31 32 33 34 36 38 41 I. LITERATURE REVIEW A. Introduction This literature review is intended to briefly introduce the problem of human obesity and the genetically obese (ob/ob) mouse as a model to study obesity. Since this research focuses on hyperinsulinemia in the ob/ob mouse as a major contributor to obesity, I have provided background information on general physiology of insulin secretion and clearance. I included alterations in physiology specific to the ob/ob mouse at the end of each section. I then focused more in detail on control of insulin secretion via circulating secretagogues such as glucose and via neural signals originating in the central nervous system (CNS). Specific evidence of altered glucose and neural regulation of insulin secretion is presented for only 3 well studied animal models of obesity, the ventral medial hypothalamic (VMH)- lesioned rat, the Zucker fa/fa rat, and the ob/ob mouse. This evidence forms the basis for my hypothesis and research objectives. B. Obesity Approximately 26% of U.S. adults are overweight and about 1/3 of these are considered extremely overweight 1 (VanItallie, 1985). Obesity in humans is a risk factor for hypertension, hypercholesterolemia, and type II diabetes, and is associated with increased morbidity and mortality (Kral, 1985; VanItallie, 1985). Although the etiology of obesity remains to be elucidated, there is increasing evidence that obesity may be a partially heritable trait (for review, see Dulloo and Miller, 1987). Most convincing evidence of a genetic component to obesity are studies with adopted children and/or adults. Stunkard et al, 1986, classified adopted subjects by body mass index into weight classes and correlated the weight class of the adopted subject with the body mass index of the biological or adoptive parents. He found a strong relationship between weight class of adopted subjects and body-mass index of biological parents; whereas, there was no relationship between weight class of adopted subjects and body-mass index of adoptive parents, indicating that a genetic component may determine later body weight. Given the prevalence of obesity and the potential for a genetic contribution to the disease, studies on the mechanism of genetic obesity are important to the elucidation of treatments for obesity. However, since co-existance of genetic and dietary variability make human studies difficult to conduct and interpret, most obesity research has been conducted on obese animals as models for the human condition. The ob/ob mouse is one of the most common obese animals studied because of the extensive data base already established and because the obesity is genetic rather than chemically or electrically induced. Thus, since these mice were available in our lab, I chose to study the etiology of obesity using the genetically obese ob/ob mouse. C. Ob/ob Mice ss Models for Human Obesity In genetically obese C57BL/6J (ob/ob) mice, the obese phenotype is inherited as a Mendelian recessive trait on chromosome 6 (Coleman, 1978). Metabolic abnormalities displayed by ob/ob mice are similar to those of obese humans including hyperphagia, hyperinsulinemia, and peripheral insulin resistance (Karam et a1, 1963; Bray and York, 1979; Olefsky and Kolterman, 1980; Dulloo and Miller, 1987). Additionally, ob/ob mice are characterized by mild hyperglycemia, decreased brown adipose tissue thermogenesis, increased plasma corticosterone, increased energy efficiency and many more metabolic abnormalities (Bray and York, 1979). Since hyperinsulinemia is 1) common to both obese humans and obese animals, 2) one of the first measurable metabolic abnormalities in ob/ob mice, and 3) implicated as an important causative factor in obesity, I chose to focus my efforts toward determining the cause of hyperinsulinemia in the ob/ob mouse. The remainder of this literature review will be focused on the role of hyperinsulinemia in contributing to obesity and then on the possible causes of hyperinsulinemia in ob/ob mice. 1. Hyperinsulinemia i_ Oblob Mice Although the primary defect responsible for obesity in ob/ob mice has not been elucidated, hyperinsulinemia is apparent as early as 6 days of age (Dubuc, 1981), before increased adiposity (7 days; Boissonneault et a1, 1978), increased plasma corticosterone (21 days; Dubuc, 1976a), hyperphagia (28 days; Lin et al, 1977), hyperglycemia (30 days; Westman, 1968), and peripheral insulin resistance (6 wks; Batchelor et a1, 1975). Early hyperinsulinemia has been implicated as the primary cause of excess adiposity in ob/ob mice since insulin is known to increase activity of lipogenic enzymes, decrease lipolysis, and promote adipocyte hypertrophy and adipose tissue cell proliferation (Bray and York, 1979; Geloen et a1, 1989). Hyperinsulinemia could also contribute to obesity by causing hypoglycemia with resultant hyperphagia (MacKay et a1, 1940). Artificial lowering of plasma insulin in ob/ob mice using streptozotocin reduces adiposity through a decrease in lipogenesis (Loten et a1, 1974). In fact, in every case where plasma insulin is lowered in ob/ob mice, adiposity is also decreased. Restricted feeding (Dubuc, 1976b), and treatment with B adrenergic agonist cimateral (Walker and Romsos, 1988) both lower plasma insulin and decrease severity of obesity. Adrenalectomy also lowers plasma insulin in ob/ob mice, but to various degrees depending on the diet fed. Decreases in adiposity correspond to decreases in plasma insulin (Warwick and Romsos, 1988). Again, since hyperinsulinemia is one of the first metabolic defects that contributes to obesity and the primary cause of hyperinsulinemia in ob/ob mice is unknown, it is an important area to study in an attempt to elicit the etiology of obesity in ob/ob mice. Plasma insulin concentrations are controlled by 2 factors, secretion, and clearance; therefore, I begin the next section with a brief description of the general physiology of insulin secretion, discussing alterations in physiology observed in the ob/ob mouse at the end of each section. I then focus on glucose and neurotransmitters as factors that influence insulin secretion, and finally, I discuss the role of insulin clearance in contributing to hyperinsulinemia. D.Regulation 9; Plasma Insulin Concentration 1. Physiology 9f Insulin Synthesis and Secretion The pancreas is composed of both exocrine, and endocrine tissue. Exocrine cells secrete digestive enzymes into a duct system which empties into the duodenum via the common bile duct. In contrast, endocrine cells are localized into clusters of 1000-2000 cells known as the islets of Langerhans and secrete hormones directly into the circulatory system. The islets of Langerhans, which make up only 1-2% of total pancreatic tissue, are composed of a core of B cells which secrete insulin, surrounded by A, D, and F cells which secrete glucagon, somatostatin and pancreatic polypeptide, respectively. Approximately 60-80% of islet cells are B cells, 15-20% are A cells, 15-20% are F cells and 5-10% are D cells (Bonner-Weir, 1989). Islets are surrounded by a dense network of capillaries which supply nutrients and transport secretory products via portal blood directly to the liver, then into general circulation. Insulin is a peptide hormone, synthesized within the islet B cell as single chain pro-insulin (MW 9000), which is packaged into secretory granules, then cleaved into C-peptide and insulin (MW 6000). Secretory granules serve as storage for insulin and upon stimulus, granules release their contents by exocytosis. Pancreatic islets from adult ob/ob mice (> 8 wks of age) are distinctly different from islets of lean littermates. Not only are there more islets per pancreas in adult ob/ob mice (Coleman and Hummel, 1973), but ob/ob mouse islets tend to be larger (Hellman et a1, 1961; Coleman and Hummel, 1973), contain a greater percentage B cells (>90%: Gepts et al, 1960; Baetens et al, 1978), and are supplied by a much more dense system of capillaries than islets of lean littermates (Rooth et a1, 1985). Furthermore, whereas normal islets are highly granulated, ob/ob islets tend to be degranulated and display alterations in rough endoplasmic reticulum and golgi apparatus indicative of a constantly stimulated pattern of insulin synthesis and secretion (Gepts et al, 1960; Diani et al, 1984). Alterations in islet morphology appear to occur developmentally since at 4 wks of age, islets of ob/ob mice are hypertrophied and degranulated, but at 8 wks of age these alterations in islet morphology are even more exaggerated (Coleman and Hummel, 1973). The exact mechanism of observed islet changes is unknown. There are many factors that control insulin secretion including glucose, amino acids, hormones within the islet, islet blood flow, neuropeptides and neurotransmitters. These signals reaching the islet can be carried through the circulatory system or can be initiated in the brain and travel through neural pathways. Additionally, cells within the islet can directly influence B cell secretion through paracrine control systems. If insulin secretion is increased in ob/ob mice, the factors that may contribute to insulin hypersecretion must be understood. I will discuss in detail only glucose and neurotransmitters since these are proposed to be important regulators of insulin secretion, and since my experiments involved these two regulators. 2. Glucose ss s Regulator Q: Insulin Secretion The primary stimulus for insulin secretion from the islet B cell is glucose. Upon glucose stimulus, insulin is released in a biphasic manner. Glucose metabolism is required for stimulus/secretion coupling; however, the exact mechanism whereby glucose stimulates insulin secretion is unclear. Glucose metabolism results in ATP production. ATP sensitive K+ channels appear to be involved in initial membrane depolarization (Arkhammer et al, 1987) which then ++ ++ activates voltage sensitive Ca channels and allows Ca entry (Atwater et a1, 1980; Wollheim and Sharp, 1981; Henquin and Meissner, 1984). Increased cellular Ca"+ is believed to be involved in translocation of secretory granules from the cell interior to the plasma membrane for exocytosis; however, mechanisms beyond Ca++ entry into the B cell remain to be elucidated (Draznin and Dahl, 1989). Islets from adult ob/ob mice secrete insulin in a typical biphasic manner but are both hyperresponsive and hypersensitive to glucose stimulus (Lavine et a1, 1977). Mechanisms involved in this altered secretory response are unknown. This hypersecretion in response to glucose could partially explain the observed hyperinsulinemia in ob/ob mice; however, glucose may not be the sole cause of insulin hypersecretion since at the earliest time point of hyperinsulinemia (6 days of age) and until weaning, ob/ob mice are hypoglycemic yet are still hyperinsulinemic (Dubuc, 1976c; 1981). 3. Neurotransmitters ss Regulators sf Insulin Secretion Neural pathways have been proposed as an important system to modulate glucose-induced insulin secretion. In fact, meal-induced insulin release in rats has been estimated to be approximately 26% neurally mediated. (Berthoud, 1984). Pancreatic islets are innervated by both parasympathetic and sympathetic fibers which have been observed in close vicinity of B cell membranes (Esterhuizen et al, 1968). Sympathetic fibers innervating the pancreas originate in the spinal cord, travel through the splanchnic nerve and release norepinephrine at the B cell membrane (Bereiter et al, 1981; Miller, 1981). Epinephrine, released by the adrenal medulla into the bloodstream, provides additional sympathetic control of insulin secretion. In normal rats, there appears to be a tonic inhibition of insulin secretion through the sympathetic nervous system. Some researchers have estimated that the sympathetic nervous system may tonically inhibit i vivo insulin secretion in fed rats by as much as 47% (Curry, 1983). Thus, the sympathetic nervous system may be an important regulatory system for overall insulin response. Although both norepinephrine and epinephrine can potentially act at B (stimulatory) or a (inhibitory) receptors, sympathetic innervation in rats or mice results in overall inhibition of insulin secretion (for review, Miller, 1981). Inhibitory effects appear to be mediated by the 0:2 subclass of a receptors since effects of epinephrine or norepinephrine on insulin secretion from mouse islets can be blocked by a nonspecific a antagonist, phentolamine (Coll- Garcia and Gill, 1968), and more specifically blocked by an 02 antagonist yohimbine (Nakadate et al, 1980). The mechanism of norepinephrine inhibition of insulin secretion is not completely understood. Binding at a2 10 receptors was originally believed to act through an inhibitory G protein which decreased cellular cAMP; however, many studies have dissociated the effects of norepinephrine on cAMP from its effects on insulin release (Ullrich and Wollheim, 1984). More recent evidence suggests that binding to a2 receptors may act through a more distal site in stimulus-secretion coupling, perhaps involving desensitizing the secretory apparatus to Ca++ (Morgan and Montague, 1985; Metz, 1988; Nilsson et al, 1988). Parasympathetic fibers innervating the pancreas appear to originate in the nucleus ambiguus and dorsal motor nucleus of the main brainstem, travel through the vagus nerve, and release acetylcholine at B cell membranes (Bereiter et al, 1981). Parasympathetic nerves innervating islets release acetylcholine which acts at muscarinic receptors and results in potentiation of glucose-induced insulin secretion (Bergman and Miller, 1973; Loubatieres-Mariani et al, 1973; Miller, 1981). Parasympathetic nervous system activity at the pancreas is believed to be responsible for cephalic phase insulin secretion (insulin secretion that occurs just prior to a meal and before an increase in plasma glucose) since cephalic insulin secretion can be blocked by vagotomy (Storlien, 1985). In fact, parasympathetic input to the pancreas may be responsible for over 26% of the insulin response to a meal (Berthoud, 1984). Potentiation of glucose-induced insulin secretion by acetylcholine appears to require the presence of a glucose 11 concentration above threshold for insulin release (Hermans et al, 1987). Furthermore, effects of acetylcholine are proposed to be greater at higher concentrations of glucose (Campfield and Smith, 1980). The mechanism of acetylcholine potentiation of glucose- induced insulin secretion is also not completely understood. Binding of acetylcholine to muscarinic receptors on the B cell membrane is believed to result in activation of the phosphoinositide second messenger system with resultant release of intracellular Ca++ (Best and Malaisse, 1984). Increased cellular Ca++ then initiates insulin release. Recent evidence suggests that acetylcholine mediates an early transient Ca++ entry required for its potentiating effect (Henquin, et al, 1988; Sanchez-Andres et al, 1988). Yet, exact mechanisms of acetylcholine action remain unclear. Alterations in insulin secretion, controlled by glucose, neurotransmitters, and other stimuli could contribute to hyperinsulinemia; however, insulin clearance must also be considered as a possible contributing factor. 4. Insulin Clearance Approximately 50% of insulin leaving the pancreas is immediately cleared during the first pass through the liver (Karakash et al, 1976). Since plasma insulin is so excessive in ob/ob mice, decreased clearance of insulin has been suggested as a potential contributing factor. Although insulin clearance is decreased in ob/ob compared to lean 12 livers, decreased clearance appears to be secondary to hyperinsulinemia rather than a primary contributing factor to the hyperinsulinemia. This was concluded because reduction of plasma insulin using streptozotocin in ob/ob mice resulted in increased insulin clearance, indicating that ob/ob livers were normal in their ability to clear insulin when plasma insulin was normalized (Karakash et al, 1976). As a result, increased insulin secretion must be the primary initiator of hyperinsulinemia in ob/ob mice. Since neural input appears to play a major role in control of insulin secretion, alterations in parasympathetic or sympathetic activity could potentially contribute to hyperinsulinemia in ob/ob mice. Additionally, because neural input to the pancreas is originally signalled from the brain, evidence of alterations in neural regulation of insulin secretion from ob/ob islets may indicate a defect in the brain. This would support a current hypothesis that a central neural defect is the primary etiology of obesity in ob/ob mice. This central neural defect hypothesis is discussed in the next section. Evidence in support of this hypothesis has been collected on 3 models of obesity; thus, specific evidence will be provided on the VMH-lesioned rat, the Zucker fa/fa rat and the ob/ob mouse. E. Relationship Between Central Nervous SystemL HyperinsulinemiaL and Obesity Several animal models of obesity have been studied in an attempt to locate a primary defect responsible for widespread 13 metabolic abnormalities seen in obese animals. Genetically obese animals, ob/ob mice, and Zucker fa/fa rats, share many common metabolic abnomalities with rats made obese by electrical lesioning of the ventromedial hypothalamus (VMH- lesioned rats). 0b/ob mice, fa/fa rats and VMH-lesioned rats are all characterized by hyperinsulinemia, hyperphagia, obesity, decreased thermoregulatory thermogenesis (through decreased sympathetic neural activity at brown adipose tissue), and hypertrophy and hypersecretion of pancreatic islets (Bray and York, 1979; Jeanrenaud, 1985). A single gene defect in genetically obese rodents results in such diverse abnormalitities, which so closely resemble those of rodents made obese through a lesion in the CNS, that researchers have suggested all 3 obese states may result from a primary central neural abnormality. A central neural defect or lesion could alter neural regulation of many organs including the pancreas which could explain existing hyperinsulinemia and resultant obesity (Jeanrenaud, 1985). The VMH and lateral hypothalamic regions of the hypothalamus are proposed to reciprocally control metabolism through the sympathetic and parasympathetic nervous systems, respectively (Bray, 1986). Alterations in these brain areas could lead to excess insulin secretion by B cells through increased stimulation by acetylcholine or decreased inhibition by norepinephrine and epinephrine, or both. Techniques used to measure whether obese animals display altered neural regulation of insulin secretion include: 1) 14 effects of vagotomy or neurotransmitters and neurotransmitter antagonists on is yiyg insulin secretion, 2) effects of neurotransmitters on insulin secretion from isolated pancreatic islets is vitro, and 3) norepinephrine turnover as a measure of sympathetic activity in whole pancreas. 1. VMH-Lesioned Rats Islets isolated from VMH-lesioned rats 8-10 wks after lesioning are hypertrophied and hypersecrete insulin in response to both low (5 mM glucose) and high (20 mM) concentrations of glucose (Inoue et al, 1977). By 8-10 wks after lesioning islet hypertrophy could be the result of hyperphagia; however, there is some evidence that this is not the case. When VMH-lesioned, hypophysectomized rats were pair-fed to control rats, islets of VMH-lesioned, hypophysectomized rats were still larger than islets of control rats (Han et al, 1970). Additionally, islets from VMH-lesioned rats are hyperresponsive to glucose stimulus within 1 day after lesioning, before increased food intake or increased islet size (Campfield et al, 1986). Thus, the lesion itself must somehow be signaling the islets such that they become hyperresponsive to glucose stimulus and increase in size. Altered neural signalling to the islets may be one cause of these observed islet changes. There is considerable evidence for alterations in the parasympathetic and sympathetic neural activity to the pancreas in the VMH—lesioned rat. Within minutes after 15 lesioning there is an increase in plasma insulin which can be reversed by vagotomy (Berthoud and Jeanrenaud, 1979). These results suggest that increases in parasympathetic nerve activity after VMH-lesioning mediate the hyperinsulinemia. Evidence for alterations in both sympathetic and parasympathetic innervation of islets from VMH-lesioned rats is presented by Campfield and Smith, 1983. By incubating isolated pancreatic islets from VMH and control rats in varying concentrations of neurotransmitters and 10 mM glucose, sensitivity to neurotransmitters was measured. VMH- lesioned rat islets were less responsive and less sensitive to acetylcholine, and more responsive and more sensitive to norepinephrine than islets of non-lesioned rats. These data suggest possible up-regulation of norepinephrine receptors and down-regulation of acetylcholine receptors, perhaps in response to decreased sympathetic and increased parasympathetic tone at the islets i_ vivo. Additional time course studies demonstrated simultaneous increases in glucose and norepinephrine responsiveness at day 1 after lesion; whereas, acetylcholine responsiveness decreased at day 2. Authors suggest a possible role for norepinephrine in altering glucose responsiveness (Campfield et al, 1986). An additional study by Campfield et al, 1984, suggests that the parasympathetic nervous system may play a role in regulating islet glucose sensitivity since vagotomy in normal rats increased islet sensitivity to glucose. These studies provided indirect estimates of neural 16 activity to the islet; however, more direct studies performed in vivo support the is vitro results in suggesting that alterations in islet sensitivity and responsiveness to neurotransmitters are compensatory responses to decreased sympathetic and increased parasympathetic tone. Infusion of an acetylcholine analog into VMH-lesioned rats 1 wk after lesioning resulted in less of an increase in plasma insulin in lesioned rats compared to sham-operated rats. Again this indicated decreased responsiveness to acetylcholine after VMH-lesioning. After infusion of norepinephrine and 0.5 g glucose/kg body weight, plasma insulin decreased much more in VMH-lesioned rats compared to control rats. This confirms increased responsiveness to norepinephrine in VMH-lesioned rats (Smith and Campfield, 1986). Finally, direct recordings from pancreatic nerves have shown that after VMH lesion, firing rate of sympathetic nerves is decreased and firing rate of parasympathetic nerves is increased (Yoshimatsu et al, 1984). 2. Zucker ifa/fai Rats Zucker (fa/fa) rats are similar to ob/ob mice in that their obesity is inherited as an autosomal recessive trait. Plasma insulin of fa/fa rats has been reported at 3-30X greater than that of lean littermates (Curry and Stern, 1985). As with ob/ob mice, pancreatic islets of fa/fa rats are both hypertrophied (Shino et al, 1973) and hyperplastic (Larsson et a1, 1977) and are hypersensitive and 17 hyperresponsive to glucose stimulus (Schade and Eaton, 1975; Pansini and Tolman, 1981; Curry and Stern, 1985). CNS alterations have been suggested as contributing to insulin hypersecretion in fa/fa rats. Glucose infusion into pre-obese, 17 day old fa/fa rats results in hypersecretion of insulin compared to lean littermates. This increased responsiveness to glucose in pre-obese rats could be blocked by pre-treatment with atropine suggesting increased vagal activity in fa/fa rats (Rohner-Jeanrenaud et al, 1983). Additionally, electrical stimulation of the vagus nerve in 6- 9 wk old fa/fa and lean rats resulted in much greater stimulation of insulin secretion from fa/fa rats than lean littermates (Rohner-Jeanrenaud et al, 1983). This evidence is supported by recent research using a fully innervated pancreatic perfusion method on Zucker fa/fa rats. Insulin secretion in response to 200 mg glucose/d1 was greatest from perfused pancreas of CNS-intact fa/fa rats and was decreased by 50% in CNS-ablated fa/fa rats. Vagotomy of CNS-intact rats decreased insulin secretion to levels of CNS-ablated rats. No effect of ablating CNS or vagotomy were observed from lean rats (Lee et al, 1989). This supports the view that increased activity of the vagus nerve in fa/fa rats is responsible for much of the hyperinsulinemia. 3. 0b/ob Mice There are several reasons to postulate that neurotransmitter mediated insulin secretion may be altered in 18 ob/ob mice. Ahren and Lundquist, 1982, administered 26 umol methylatropine (a muscarinic cholinergic antagonist) per kg body weight intraperitoneally (IP) to ob/ob and control NMRI mice. Measurements of plasma insulin at 15, 30 and 60 mins post-injection showed significantly greater absolute decreases in plasma insulin in ob/ob mice compared to controls at all time points. In addition, when expressed as a percentage of baseline insulin concentration, which is significantly higher in ob/ob than controls, by 15 mins plasma insulin had decreased 60% in ob/ob mice compared to 30% in controls. Furthermore, phentolamine and L— propranolol, a and B adrenergic antagonists respectively, were administered to determine possible alterations in insulin response to sympathetic activity. Phentolamine administration IP at 53 umol/kg resulted in greater absolute increases in plasma insulin concentration in ob/ob mice compared to controls. Percentage increases in plasma insulin were 200% over baseline in both ob/ob and controls. L- propranolol (68 umol/kg) IP resulted in a 40 % decrease in plasma insulin concentration in both ob/ob and controls: however, absolute decreases in plasma insulin were greater in ob/ob than controls. Although these results support the hypothesis of altered insulin response to neurotransmitters in ob/ob mice, several limitations must be considered. Control mice used by Ahren and Lunquist, 1982, were of different genetic background (NMRI mice) than ob/ob mice used; thus, differences in 19 response to adrenergic and cholinergic blockade may have resulted from genetic background differences rather than differences in expression of the ob/ob gene itself. Furthermore, drugs were administered based on body weight. Since ob/ob mice were up to 40 g heavier than controls, absolute amounts of drug administered were greater in ob/ob than controls. Furthermore, phentolamine has recently been shown to have a direct stimulatory effect on insulin secretion (Schulz and Hasselblatt, 1988). A recent report from Kuhn et al, 1987, supports the data of Ahren and Lunquist, 1982, as evidence for possible abnormalities in neural control of insulin secretion in ob/ob mice. Epinephrine, an adrenergic agonist, injected subcutaneously (3 ug/lo g body weight), resulted in a rapid 50% decrease in plasma insulin in ob/ob mice; whereas, no change in plasma insulin was observed in lean mice. To evaluate islet insulin response to endogenous neural activity, a nonspecific a adrenergic antagonist, phentolamine, was administered. Absolute increases in plasma insulin were greater in ob/ob than controls supporting results seen with exogenous epinephrine administration. Furthermore, ob/ob mice exhibited greater decreases in plasma insulin than controls in response to immobilization stress, implying exaggerated responsiveness to catecholamines or sympathetic hypertonicity with stress (Kuhn et al, 1987). Again, a problem with this work is that phentolamine has a direct stimulatory effect on islet insulin secretion (Schulz 20 and Hasselblatt, 1988). Norepinephrine turnover, a measure of sympathetic nervous system activity, in ob/ob mouse pancreas is comparable to control mouse pancreas (Knehans and Romsos, 1983). Although a normal norepinephrine turnover in whole pancreas would appear to argue for normal sympathetic activity to the islets, islets make up only 1—2% of pancreatic tissue. Thus, sympathetic activity to islets could be altered and remain undetectable when measuring whole pancreas. Additionally sympathetic epinephrine from the adrenal medulla could be decreased as has been shown for rats with hypothalamic knife cuts (Vander Tuig et al, 1987). Additional evidence for altered neural regulation of insulin secretion in ob/ob mice is provided by studies on islet blood flow. Norepinephrine injection resulted in immediate inhibition of blood flow to ob/ob islets with little effect on lean islets (Rooth et al, 1985). Furthermore, in response to epinephrine, ob/ob mouse islets secreted glucagon at 4X basal levels; whereas, the same dose of epinephrine had no effect on lean islets (Beloff—Chain et al, 1977). F. Hypothesis and Obigctives I hypothesized that islets of ob/ob mice would hypersecrete insulin in response to glucose stimulus, and would be less sensitive to acetylcholine and more sensitive to norepinephrine stimulus compared to islets of lean 21 littermates. Throughout this report, I use the term hypersensitive to refer to islets that secrete insulin in response to lower concentrations of secretagogue when compared to control islets. 0n the other hand, I use hyperresponsive to refer to islets that secrete larger amounts of insulin in response to a given concentration of secretagogue as compared to control islets. My objectives were to determine 1) the dynamics of glucose-induced insulin secretion, and 2) the dynamics of norepinephrine inhibition and acetylcholine potentiation of glucose-induced insulin secretion, from islets of 8-9 wk old ob/ob and lean mice. II. MATERIALS AND METHODS A. Animals and Materials Female obese (ob/ob) mice and lean littermates (ob/+ or +/+) from our breeding colony (C57BL/6J - ob/+) were group- housed in a temperature controlled room (25°C) with a 12 hr light-dark cycle (lights on at 0700 hr). Wood shavings were provided for bedding. Stock diet (Wayne Lab-Blocks, Continental Grain, Chicago, IL) and water were allowed sg libitum from weaning at 3 wks of age until mice were used at 8-9 wks of age. Collagenase, type V, lot # 18F-6840, bovine serum albumin (fraction V, radioimmunoassay grade), yohimbine hydrochloride, prazosin hydrochloride, acetylcholine chloride, norepinephrine (arterenol hydrochloride), eserine, atropine methyl nitrate, and HEPES, were from Sigma Chemical Company, St. Louis, MO; 1251 insulin was from ICN Biomedicals, Irvine, CA; ascorbic acid was from Fisher Scientific Company, Fair Lawn, NJ; anti-porcine insulin guinea pig serum and rat insulin standard were from Novo BioLabs, Danbury, CT. Krebs Ringer bicarbonate buffer for isolation of islets and islet incubations was freshly oxygenated, pH 7.4. 22 23 B. Islet Preparation We isolated pancreatic islets using a modification of the method by Lacy and Kostianovsky, 1967. Mice, (8-9 wks of age) were killed by cervical dislocation in the fed state between 1000 and 1200 hr. Upon opening the abdominal cavity, we ligated the common bile duct at the duodenal entrance, and by inserting a 30 guage needle into the hepatic portion of the bile duct, we inflated the pancreas with 3 mls of 37°C Krebs Ringer bicarbonate buffer containing 2.5 mg collagenase/ml and 2.5 mM glucose. We quickly dissected the pancreas and without chopping, placed it in a small glass tube containing an additional 0.5 ml of a 10 mg collagenase/ml solution. Each tube containing 1 pancreas was gently shaken by hand in a 37°C water bath for 3-4 mins, then briskly shaken about 10 times to loosen islets from surrounding connective tissue. To stop the digestion, we added ice-cold buffer containing 2.5 mM glucose, then washed the islets several times to remove digested acinar tissue and collagenase. Islet yield averaged 60 islets per lean mouse and 100 islets per ob/ob mouse. C. Experimental Design 1. General Design Each experiment consisted of mean results from 5 or 6 experimental days, (n=5 or 6). On one experimental day, 3 lean and 2 ob/ob mice were alternately killed and their 24 islets isolated. Islets from the 3 lean mice were pooled and distributed 10 islets per dish with 1 or 2 dishes per treatment. 0b/ob mouse islets were also pooled and distributed across identical treatments. For treatments with 2 dishes of islets, the 2 values for insulin secretion for that experimental day were averaged and considered a single replication (n). Otherwise, insulin secretion from the one dish (10 islets) for each treatment was considered a single replication (n). A stereoscopic microscope and a 200 pl micropipetter were used to transfer islets into small, black bottom petri dishes. Care was taken to distribute islets across dishes such that all dishes within one phenotype received 10 representative islets of approximately equal size. Islets for all experiments underwent 3, consecutive, 30 min incubations at 37°C under a continuous 95% 02, 5% C02 atmosphere. For the first 30 min, islets were pre-incubated in 1 ml of 2.5 mM glucose Krebs Ringer bicarbonate buffer with 1 mg/ml bovine serum albumin. Islets were then transferred, using a 200 pl micropipetter and visualizing under a stereoscopic microscope, to 1 ml of fresh but identicle incubation buffer. After a second 30 min incubation, 0.5 ml of buffer was sampled for determination of basal, non-stimulated immunoreactive insulin (referred to as insulin throughout this paper) release. This 30 min incubation in 2.5 mM glucose was included in each experiment as a method of determining islet damage incurred during collagenase digestion. The 0.5 mls of sample buffer 25 collected for determination of islet damage was then replaced with 0.5 m1 of Krebs Ringer bicarbonate buffer containing the desired concentrations of glucose, neurotransmitters, or antagonists, depending on the experiment. Islets were then incubated for another 30 min. Following this last, 30 min incubation, medium was collected for determination of insulin released. At the end of each experiment, representative islets were extracted for determination of islet insulin content (Curry, 1986). Since ob/ob islets secrete more insulin under some conditions than lean islets and this would influence final islet insulin content, we only extracted islets that were incubated in 2.5 mM glucose, or 15 mM glucose plus 3 or 30 uM norepinephrine. Under these conditions, insulin secretion from ob/ob and lean islets was similar. As stated earlier, insulin secreted during the second, 2.5 mM glucose incubation was used as an index of islet damage. Hahn et al, 1976, demonstrated that over-digested islets secreted up to 20 fold more insulin under basal conditions than intact islets and then up to 2 fold more insulin than intact islets in response to 20 mM glucose stimulus. In our preliminary tests, both ob/ob and lean islets that released greater than 0.3 ng insulin/islet ( 10 fold greater than normal) during basal conditions also secreted excessive quantities of insulin when challenged with glucose; therefore, non-stimulated secretion over 0.3 ng insulin/islet was considered an indication of islet damage 26 and these data were discarded. As an additional control, islet viability was evaluated by islet ability to secrete insulin in response to glucose (Krause et al, 1973; Lacy and Kostianovsky, 1967). If islets from an experimental day showed no secretory response to glucose, data was discarded. On each of 5 randomly chosen experimental days, we measured diameters of approximately 20 representative islets. These data were cumulated to determine the distribution of various islet sizes used from ob/ob and lean mice. A total of 98 ob/ob islets and 92 lean islets were measured. Non- spherical islets were measured at their longest and shortest axes and the average of the two was recorded. Additional mice were killed between 1000-1200 hr for determination of pancreatic insulin content, plasma insulin and plasma glucose. For pancreatic extraction of insulin, each carefully dissected pancreas was chopped, sonicated for 5 min, and incubated overnight at 4°C in 10 ml acid ethanol (composition 7.5 ml 12 N HCl to 492.5 ml 75% ethanol). Trasylol, 1000 K/pancreas, was added to inhibit proteolytic enzymes. Samples were centrifuged at approximately 3000 g: Supernatants were removed and pellets re-extracted (Curry, 1986). Following three overnight extractions, all samples were diluted in Krebs Ringer bicarbonate buffer and stored at -20°C for subsequent insulin determination. 27 2. Experiment 1 - Effects 9: Glucose gs Insulin Secretion For experiment 1, islets were incubated in 2.5, 5, 7, 10, 12, 15, or 20 mM glucose during the last 30 min period to characterize the glucose/insulin dose response curves for both ob/ob and lean mouse islets. Based on these results, we selected the concentrations of glucose to use for studies with acetylcholine and norepinephrine. Acetylcholine may be more effective at higher concentrations of glucose. Therefore, although alterations in sensitivity to glucose of ob/ob islets is well documented (Lavine et al, 1977), we felt it was important to characterize the effects of glucose on insulin secretion from our ob/ob and lean mouse islets. Due to alterations in sensitivity, what may be a high concentration of glucose for ob/ob islets (eliciting near 100% of maximum insulin secretion), may be a low concentration of glucose for lean islets (eliciting less than 50% of maximum secretion). Thus, in an attempt to eliminate this potential source of bias, we selected a concentration of glucose that had approximately the same effect on both ob/ob and lean islets (as a % of maximum secretion). 3. Experiment ; - Effects sf Neurotransmitters 9g Glucose-Induced Insulin Secretion Experiment 2 was conducted in 3 parts. It was designed to determine islet responsiveness and sensitivity to 1) varying concentrations of acetylcholine in the presence of 28 high (15 mM) glucose, 2) varying concentrations of 'acetylcholine in the presence of low (5 mM) glucose, and 3) varying concentrations of norepinephrine in the presence of high (15 mM) glucose. Thus, for the 2 trials with acetylcholine, during the last 30 min incubation, each dish contained either 15 or 5 mM glucose and 0, 0.005, 0.01, 0.05, 0.1, 1, or 10 uM acetycholine. All treatment dishes including the control (0 uM acetylcholine) dishes contained 10 uM eserine to antagonize acetylcholine esterase activity. For the trial with norepinephrine, 0, 0.0003, 0.003, 0.03, 0.3, 3, or 30 uM norepinephrine in addition to 15 mM glucose was used for the last, 30 min incubation. All treatment dishes including the control (0 uM norepinephrine) dishes contained 1 mM ascorbic acid to prevent oxidation of norepinephrine. 4. Experiment 3 - Effects sf Neurotransmitter Antagonists gs Insulin Secretion Experiment 3 was conducted in 2 parts and employed acetylcholine and norepinephrine antagonists to determine if, in previous experiments, neurotransmitters were interacting with specific receptors to produce the observed stimulatory and inhibitory effects. Treatment groups included; 15 mM glucose alone; 15 mM glucose plus neurotransmitter; 15 mM glucose plus neurotransmitter plus antagonist; and 15 mM glucose plus antagonist. First, 10 MM atropine (a muscarinic acetylcholine antagonist) was used to block the stimulatory 29 effects of 1 pM acetylcholine. And second, 30 pM yohimbine (an a2 adrenergic antagonist) or 30 uM prazosin (an al adrenergic antagonist) was used to block the inhibitory effects of 3 uM norepinephrine. D. Sample Analysis Insulin in incubation buffer and plasma was determined using radioimmunoassay (Novo Research Laboratories, Bagsvaerd, Denmark). Intra-assay and inter-assay coefficients of variation were 5 and 12%, respectively. Plasma glucose was determined using the glucose oxidase method (Boehringer Mannheim, Indianapolis, Indiana). E. Statistical Analysis Data were analyzed by two factor factorial analysis of variance (phenotype x concentration of glucose or neurotransmitter) or by one way analysis of variance. Dunnett's t-test was used to detect the lowest effective dose of drug on insulin secretion, and Students t-test to detect differences in insulin secretion between ob/ob and lean mouse islets at a given drug dose (Gill, 1981). Means were considered statistically significant at P< .05. III. RESULTS At 8-9 wks of age, ob/ob and lean mice weighed approximately 35 and 22 g, respectively. Plasma insulin and glucose concentrations of 8-9 wk old ob/ob and lean mice are presented in Figure 1. Plasma insulin was 95 fold higher and plasma glucose 30% higher in ob/ob compared to lean mice. Additionally, pancreatic islets of ob/ob mice were on average larger in diameter than islets of lean littermates, and were distributed across a wider size range (Figure 2). Mean islet 3, were 29.6 x 10- volumes, calculated using the formula 4/3nr 3 for ob/ob islets and 7.2 x 10"3 mm3 for lean islets. Although ob/ob islets were significantly larger, they contained the same quantity of insulin as lean islets (Figure 3). Total pancreatic insulin content was 25% greater in ob/ob than lean mice (Figure 3). A. Experiment 1 - Effects pf Glucose ps Insulin Secretion Ob/ob and lean mouse islets responded to glucose in a dose-dependent manner (Figure 4). Basal insulin secretion in response to 2.5 mM glucose was not significantly different between ob/ob and lean islets. However, 5 and 7 mM glucose stimulated insulin secretion from ob/ob islets, while having no stimulatory effect on insulin secretion from lean islets (Figure 4). The glucose dose response curve for ob/ob islets 3O 31 .mmhuoconm no uomuum ucmofluflcmflm mmumowccfl gownmums .AS n :v mm H mamma who once .m0fl8 Gama can n0\no 6H0 x3 mum Ho mmoosaw cam cwazmcfl mammam .H musmwm O . zummno 30mm .Amuomuocmm you w .mumHmH you mm H :v mm H mammfi zfim o no mm>uocmnm mo uomumm unmowuwcvflm campficcfl mHonfi>m Mowuopmm paw mmwuocwsm Casufl3 coflumuomm AmwOUSHm SE m.~v mcflammmn m>onm coflumuomm cflasmCA wwmmuocfl >Hucmoflmacvfim ou cofipwuucmucoo mmOOSHm ummzoa on» mumowpcfl maonsxm mmouu .Ahne u :v mm H names can puma .muwamw mwsoe coma can n0\no cao x3 w mo coflumnsocfl CHE om mcflusp coflumuomm :aasmcw meSUCAImmoozao EE .3326 cu me or m c .IIOHHHR. + 1.. 1 0m. I e nu I *n u_. me w. a I 1 .. / e on e nu e I I." .N w 1* e I: U in I I." Zuocwnm mo pomumm unmofimacmwm mumoflpcw maonahm Mofiumumm can .mm>uocmnm Gwnufl3 coaumuomm mafiawmmn m>onm :ofiumuomm awasmcw mmmmuocw >HUGMOAMHcmwm ou cowumuucmocoo mafiaonoa>umom umo3oH muooflccfl maonfi>m mmouo .AH u : .cmma .m n c .QO\no “U can é CH onwaonoH>uwom 2: 0H you ummoxm pie n CV mm H some was once .AU.onm :oflumuomm :flasmcw cw mmmmuocfl muddomnm no can Ao.¢v coaumuomm cflasmcfl munaomnc mm pmmmmumxo mum muasmmm .mcwHOSOHaumom can AQCUV mmoosHo SE m Ho Am.¢v mmOOSHm :5 ma ou uncommon :fl humane mmsoa coma cam £0\no Eouu cowumuowm awasmcH .m ousvflm .21 65.65362 Po. cw F w. r. we. to. to. ad ad to to N6 N6 n «d u I no 5 m c U , v- I I n 0 e o m... u H 3 m. S ”u P. F. [Inn 8 W 8 H n l o a 0 u I em ‘ w. u u w. n u e . I." e o I o o , m 2 I m 2 37 insulin secretion due to acetylcholine (Figure 5,B). Acetylcholine (0.05 pH and higher) resulted in much greater absolute increases in insulin secretion from ob/ob islets than from lean islets. Acetylcholine, (1 uM), potentiated insulin secretion by approximately 7 ng/islet from ob/ob islets, compared to approximately 0.9 ng/islet from lean islets (Figure 5,B). Since acetylcholine acts 1 vivo to stimulate insulin secretion just prior to a meal (termed cephalic phase insulin secretion; Storlien, 1985), when plasma glucose would be low, 5 mM glucose was selected to further study the effects of acetylcholine on glucose-induced insulin secretion. Only ob/ob islets responded in a dose-dependent manner to acetylcholine and 5 mM glucose (Figure 5,C). The lowest effective dose of acetylcholine on insulin secretion from ob/ob islets was 1 uM. Baseline insulin secretion, in response to 5 mM glucose, 0 acetylcholine (Figure 5,C) was again significantly different between ob/ob and lean islets; therefore, baseline secretion was subtracted from all secretion values. Thus, when expressed as absolute increase in insulin secretion over baseline secretion, acetylcholine at concentrations of 0.01 uM and above resulted in significantly greater potentiation of insulin secretion from ob/ob than lean islets (Figure 5,D). Data on effects of norepinephrine on glucose-induced insulin secretion are presented in Figure 6. Norepinephrine 38 .Hcmfiummhu :HnuH3 wmauocmnm mo pummum unmonHchm mumoHccH maonexm quHmumm can .coHumuowm mcHmemn aonn :oHumuomm :HHsmcH mmmmuoop maucmoHuHcmHm on mcHuzmmchmuoc mo :oHumuucmocou ummsoa on» mumoHccH maone>m mmouu .Amlm n :V mm H some mum puma .Ad amoumv mcHunmmCHmmHo: SE 0 .wmoudaq :5 ma CH COHumuuom CHHsmcH mm pochmc mH mcHHmmmm .Amv coHumuumm mcHHommn 30Hmn ommmuomp muaaomnm no Aév COHHuHUmw :HadmcH muzaomnm mm commonmxm mcHHanchmuo: can mmoosam 2E ma 0» uncommon CH mumHmH omsoa coma can n0\no Eouw :oHumuoom cHazmcH .21 6553:5952 n m- on a n. no. moo. 88. M an n n. 8. mac. 88. c O m . m H In. _ _ _ . + n o . 3 N... l a . I." + .. om Fl: * * _. g n . n., o . N w . 25.. III In. F , a .m musmHm osnalsufiu ‘uunsm UIlLI 39 inhibited glucose-induced insulin secretion from both ob/ob and lean islets in a dose-dependent manner. Norepinephrine, (0.0003 uM), significantly inhibited glucose-induced insulin secretion from ob/ob islets while having no effect on lean islets. Insulin secretion from lean islets was not significantly inhibited until 0.03 uM norepinephrine was used (Figure 6,A). Norepinephrine, 0.0003 uM, resulted in a 0.8 ng/islet decrease in insulin secretion from ob/ob islets which is a 4 fold greater decrease in insulin secretion than observed when lean islets were exposed to the maximally effective dose of norepinephrine (30 uM). Additionally, despite large differences in baseline insulin secretion, both 3 and 30 uM norepinephrine completely inhibited insulin secretion from ob/ob islets to the same absolute level of secretion as lean islets (Figure 6,A). Data from Figure 6,B are absolute decreases in insulin secretion below baseline secretion (15 mM glucose, 0 norepinephrine). When exposed to maximally effective doses (30 pM) of norepinephrine, insulin secretion from ob/ob islets was decreased by approximately 2.4 ng/islet; whereas, the same concentration of norepinephrine decreased secretion by only approximately 0.2 ng/islet for lean islets (Figure 6,B). All concentrations of norepinephrine resulted in greater absolute decreases in insulin secretion from ob/ob islets than from lean islets. 40 C. Experiment 3 - Effects of Neurotransmitter Antagonists o Insulin Secretion Finally, in experiment 3, atropine (louM) completely blocked the potentiating effect of 1 pM acetylcholine on 15 mM glucose-induced insulin secretion (Figure 7,A,C). Atropine itself had no effect on glucose-induced insulin secretion (Figure 7,A,C). The inhibitory effect of norepinephrine (3 pM) on 15 mM glucose-induced insulin secretion was completely blocked by 30 uM yohimbine (a2 antagonist), while yohimbine itself had no significant effect on glucose-induced insulin secretion (Figure 7,B,D). Prazosin (30 uM; al antagonist) had no effect on norepinephrine inhibition of glucose-induced insulin secretion from lean or ob/ob islets (Figure 7,B,D). 41 .omhuocmnm :HSHHS msoum Houucoo mmoosau SE ma Eouw mocmummqu unmoHuHcmHm wumoHch maonESm onuoumd .AumHnowmucm onumcmuUm d 1 S: on can Amzv mcHunmmchmuo: S: m .mmoosao SE ma :H 91m How 6cm .Auchommucw onuocHHono “megv ochouum S: on can ASUSV mcHHocoa>uoow S: H .mmoous SE ma :H mEHE on you pwumnsocH who; mumHmH .014 now .muchommucm umuuHEmcmuuouswc ou uncommon CH mumeH mMDOE AD1UV Emma paw Am1¢V Q0\no Eoum :oHumuowm :HHDmcH .b musva <5. o> :5 + + + <5. uz o> m2 m: + + + + + E.< zw< :w< 8.0 2.6 use use 2.6 8.0 + . use use use 3.0 oo 1.. 1.. Insulin, ng/islet/30 min. 'uuu 08/13ls!/5U ‘unnsm OQ‘DQNO ,- IV. DISCUSSION I conclude from these results that when compared to lean islets, ob/ob mouse islets are 1) both hyperresponsive and hypersensitive to glucose stimulus (Figure 4), 2) hyperresponsive to acetylcholine stimulus in the presence of 15 or 5 mM glucose (Figure 5), and 3) hyperresponsive and hypersensitive to norepinephrine inhibition in the presence of 15 mM glucose (Figure 6). These alterations in insulin secretion from ob/ob islets in response to neurotransmitters may contribute to the extremely high plasma insulin concentrations observed in 8-9 wk old ob/ob mice (Figure 1). Hyperinsulinemia as well as moderate hyperglycemia (Figure 1) observed in ob/ob mice in the present experiment, are consistent with previous reports (Coleman and Hummel, 1973; Dubuc, 1976c). Even though islets from ob/ob mice were on average larger (Figure 2), insulin content of ob/ob mouse islets was similar to that of lean mouse islets (Figure 3). Total pancreatic insulin was higher in ob/ob compared to lean mice (Figure 3). In previous studies, both ob/ob islet insulin content and pancreatic insulin content have been reported as either higher or lower than that of lean littermates (Findley et al, 1973; Loten et al, 1974; Dunbar and Walsh, 1979; Kakita et al, 1982; Black et al, 1988). These discrepencies 42 43 in insulin content of islets or pancreas may be due to differences in sex, age, condition of mice (fed vs. fasted), technique used to isolate islets, technique used to determine total insulin content, or part of the pancreas studied (dorsal vs. ventral). For example, Black et al, 1988, using islets from only the splenic portion of the pancreas of 8-12 wk old male mice reported islet insulin contents of 20 ng per ob/ob islet and 32 ng per lean islet; contrary to reports by Dunbar, on islets from whole pancreas of 24-32 week old fasted mice of approximately 280 ng per ob/ob islet and 60 ng per lean islet. Results in Figure 3 can be used to calculate approximate islet number per pancreas by dividing pancreatic insulin content by islet insulin content. Thus, on average, one ob/ob mouse pancreas contains approximately 200 islets compared to approximately 150 islets per lean mouse pancreas. This corresponds to 25% more islets per ob/ob mouse pancreas, which is consistent with histological studies by Gepts et al, 1960, who found 15% more islets in the pancreata of 16 wk old ob/ob mice compared to lean littermates. Islets from ob/ob mice were larger than islets from lean littermates (Figure 2), consistent with previous work (Lavine et al, 1977; Dunbar and Walsh, 1980; Black et al, 1988). The etiology of islet hypertrophy in ob/ob mice is unknown and has not been well studied. However, in general, pancreatic islet hypertrophy is consistently observed when excessive demands are placed on insulin secretion, such as 44 the case of partial pancreatectomy in rats (Chen et al, 1989), or constant stimulation of islet insulin secretion with glucose, (King et al, 1978). Islet hypertrophy also occurs in rats as a result of VMH-lesion. In fact, islet hypertrophy in VMH-lesioned rats occurs after increased insulin responsiveness to glucose stimulus, and after increases in parasympathetic activity and decreases in sympathetic activity to VMH-lesioned rat islets (Campfield et al, 1986). This suggests excess stimulus for insulin secretion as a contributing factor to islet hypertrophy. Islet hypertrophy in ob/ob mice is also likely to be the result of excessive demands on insulin secretion. Ob/ob mouse islets may be constantly stimulated to secrete insulin under both fasted (5 mM glucose) and fed (15 mM glucose) conditions, unlike lean islets, which are stimulated to secrete insulin only at 10 mM glucose or above (Figure 4). The 15 fold higher or 5.5 fold higher insulin secretion from ob/ob islets compared to lean islets at 10 and 15 mM glucose, respectively, cannot be fully explained by an average 4 fold greater ob/ob islet volume. I conclude from these results that increased insulin secretion from ob/ob islets is not due exclusively to increased islet size, nor is it due to increased islet insulin content, as discussed previously. Therefore, there must be some as yet unidentified factor that is responsible for altered glucose-induced insulin secretion from ob/ob mouse islets. Increased sensitivity and responsiveness to glucose of 45 ob/ob islets may be an important contributing factor to hyperinsulinemia in adult ob/ob mice. These alterations could also play a role in the development of hyperinsulinemia in ob/ob mice, especially if hypersensitivity to glucose is present prior to weaning, when ob/ob mice are hypoglycemic, but plasma insulin is increased (Dubuc, 1981). Unfortunately studies on islets of younger ob/ob mice have not been conducted. However, results from studies with VMH-lesioned rats clearly demonstrate a role for islet hyperresponsiveness in initiating hyperinsulinemia. Islets of VMH-lesioned rats hypersecrete insulin in response to glucose as early as one day after lesioning, before any islet hypertrophy can occur (Campfield et al, 1986). The hypersensitivity to glucose observed in ob/ob mice may contribute to hyperinsulinemia in other ways, perhaps by making the B cells of ob/ob mice more susceptable to parasympathetic stimulation than are B cells of lean mice. Since stimulatory concentrations of glucose appear to be needed before acetylcholine can potentiate insulin secretion (Hermans et al, 1987), lean islets may not respond to acetylcholine at 5 or 7 mM glucose; whereas, ob/ob islets may be responsive to acetylcholine stimulus at both 5 and 7 mM glucose. This idea will be discussed further with regard to experiments performed with 5 mM glucose and acetylcholine. Even if ob/ob mouse pancreata contain 25% more islets, and those islets are up to 15 fold more responsive to glucose stimulus, this cannot completely explain the 95 fold higher 46 plasma insulin concentration in ob/ob compared to lean mice. Alterations in neural control of insulin secretion may be a second contributing factor. The parasympathetic nervous system has tremendous potential to cause excess insulin secretion from pancreata of ob/ob mice (Figure 5). In fact, high concentrations of acetylcholine (1 uM) in the presence of 15 mM glucose potentiated insulin secretion by 7 ng per ob/ob islet compared to 0.9 ng per lean islet, corresponding to an 8 fold greater potentiation of insulin secretion from ob/ob islets (Figure 5,B). Furthermore, even under glucose conditions representative of the fasting state (5 mM glucose), acetylcholine is able to potentiate glucose-induced insulin secretion only from ob/ob islets (Figure 5,D); whereas, lean islets do not respond to acetylcholine at 5 mM glucose. The reason for this difference in response to acetylcholine may be that 5 mM glucose alone is able to stimulate insulin secretion from ob/ob islets but not from lean islets. The potentiating effect of 1 pM acetylcholine on ob/ob islets in the presence of 5 mM glucose (+0.34 ng/islet; Figure 5,D) is as great as the stimulatory effect of 15 mM glucose alone on lean mouse islets (+0.32 ng/islet; Figure 4). Again establishing that acetylcholine could be playing a major role in hyperinsulinemia of ob/ob mice. Data from Figure 5 are also consistent with is yiyp studies using acetylcholine antagonists in ob/ob mice. Ahren and Lundquist, 1982, found that cholinergic blockade with 47 atropine resulted in much greater decreases in plasma insulin in ob/ob mice compared to control NMRI mice. This suggests the possibility of increased parasympathetic nerve activity in ob/ob mice, or increased responsiveness to existing parasympathetic nerve activity. In addition to alterations in responsiveness to acetylcholine, ob/ob islets may be slightly less sensitive to acetylcholine stimulus. Higher concentrations of acetylcholine were needed to achieve significant potentiation of glucose-induced insulin secretion from ob/ob islets compared to lean islets (Figure 5,A). However, contrary to these statistical results, the dose response curve of ob/ob islets does not appear to be shifted to the right compared to leans which suggests sensitivity is not altered. Additionally, in Figure 5,B, absolute increase in insulin secretion from ob/ob islets at 0.01 MM acetylcholine is actually 2 fold higher than that of lean islets. Therefore, I conclude that ob/ob and lean islets are probably equally sensitive to acetylcholine stimulus. As a result, I suggest that parasympathetic tone at the islets of ob/ob mice is probably not altered compared to islets of lean mice. This is in contrast with what has been observed with other obese rodents. In both Zucker fa/fa rats and VMH- lesioned rats, research suggests that increased parasympathetic nerve activity to the pancreas contributes to hyperinsulinemia (Rohner-Jeanrenaud et al, 1983; Yoshimatsu et al, 1984). 48 0b/ob mouse islets are also hyperresponsive to norepinephrine inhibition of insulin secretion (Figure 6). Norepinephrine (30 uM) had a much greater effect on ob/ob islets than on lean islets. Even though baseline secretion (15 mM glucose, 0 norepinephrine) was much different between ob/ob and lean islets, when incubated in 30 pM norepinephrine, secretion from ob/ob and lean islets was the same. This was a decrease of 2.4 ng per ob/ob islet, but only 0.2 ng per lean islet, corresponding to a 12 fold greater effect on ob/ob islets (Figure 6,B). 0b/ob mouse islets are also more sensitive to norepinephrine inhibition, as seen by the leftward shift of the ob/ob islet norepinephrine dose response curve (Figure 6,B). Furthermore, ob/ob islet insulin secretion was inhibited by 0.0003 uM norepinephrine, a 100 fold lower concentration of norepinephrine than was required to produce inhibition in lean mouse islets, 0.03 MM. In support of these data, Kuhn et al, 1987, also found that ob/ob mice were hyperresponsive to catecholamines. Kuhn and coworkers injected epinephrine into ob/ob and lean mice and found a large decrease in plasma insulin of ob/ob mice with no effect on plasma insulin of lean mice. Increased ob/ob islet sensitivity to norepinephrine is indicative of up-regulation of adrenergic receptors or hypersensitization of the second messenger system for adrenergic receptors. Such increased sensitivity could be in response to decreased is vivo sympathetic stimulus at the 49 ob/ob islet B cell. In ob/ob mice, if there is a decreased sympathetic inhibition of insulin secretion, hyperinsulinemia could result, at least prior to compensatory increases in islet sensitivity to catecholamines. Alterations in islet response to sympathetic stimuli has been demonstrated with other obese rodents. Campfield and Smith, 1983, demonstrated that islets from VMH-lesioned rats were hypersensitive to norepinephrine. In this case, increased sensitivity to norepinephrine could be confirmed by studies is yiyp as a compensatory mechanism for decreased is yiyp sympathetic input to the islets. (Yoshimatsu, 1984; Smith and Campfield, 1986). In contrast to my data, which suggests decreased sympathetic tone at islets of ob/ob mice, evaluation of data on norepinephrine turnover in pancreas of ob/ob mice suggests that whole pancreas sympathetic activity is the same in ob/ob and lean mice (Knehans and Romsos, 1983). However islet cells comprise only 1-2% of total pancreatic tissue. Therefore, sympathetic activity specifically at the ob/ob islet cells, could still be decreased. An alternative explanation could be that there is less epinephrine from the adrenal medulla reaching the islets of ob/ob mice compared to lean littermates, resulting in the observed hypersensitivity to norepinephrine stimulus. Evidence of decreased epinephrine in obese animals is provided by experiments on rats made obese by hypothalamic knife-cuts. Urinary epinephrine was measured and found to be lower in obese 50 knife-cut rats compared to lean rats (Vander Tuig et al, 1987). One could also argue that B cells of ob/ob mice have an intrinsic defect which results in alterations in sensitivity to norepinephrine. However, this is probably not the case since the ob/ob B cell is not the only site where increased sensitivity to catecholamines is observed. Epinephrine inhibits islet blood flow to a much greater extent in ob/ob islets than in lean islets (Rooth et al, 1985). Additionally, epinephrine results in a 4 fold increase in glucagon secretion from islets of ob/ob mice with no effect on glucagon secretion from islets of lean littermates (Beloff-Chain et al, 1977). Thus, alterations in sensitivity to norepinephrine are not isolated to an intrinsic defect in the ob/ob islet B cell. Another explanation could be that in ob/ob mice both a1 and a2 adrenergic receptors rather than just a2 receptors may be functioning to enhance the signals produced by norepinephrine. Results in Figure 7 dispel this theory since only “2 receptor antagonism blocked effects of norepinephrine in both lean and ob/ob islets. Intrinsic islet defects have been suggested by Black et al, 1988, who demonstrated that ob/ob islets may have impaired function of voltage dependent Ca++ channels. Also, the electrical pattern produced in B cells of ob/ob mice upon glucose stimulus is quite different from that observed with lean mouse islets (Rosario et al, 1985). However, these authors did not establish whether these islet alterations are 51 primary or whether they may be secondary to the multiple metabolic abnormalities in ob/ob mice. In summary, alterations in sensitivity to neurotransmitters in the present experiment suggest normal parasympathetic tone and decreased sympathetic tone or sympathoadrenal activity at the islets of ob/ob mice compared to islets of lean littermates. Additionally, these results suggest a role for both parasympathetic and sympathetic nervous systems in the hyperinsulinemia of ob/ob mice. The parasympathetic nervous system, via release of acetylcholine, can result in much greater increases in insulin secretion and at lower glucose concentrations from ob/ob than lean islets. The sympathetic nervous system, via decreased sympathetic or sympathoadrenal inhibition of insulin secretion, could contribute to hyperinsulinemia in ob/ob mice, at least prior to compensatory increases in islet sensitivity to catecholamines. V. RECOMMENDATIONS FOR FURTHER STUDIES To better understand the present results, and to continue searching for the cause of hyperinsulinemia and obesity in the ob/ob mouse, I propose the following studies. A. Epinephrine is Ob/ob Mice Plasma epinephrine turnover or urinary epinephrine should be measured in 8-9 wk old ob/ob and lean mice to determine if decreased plasma epinephrine in ob/ob mice may 52 be responsible for increased ob/ob islet sensitivity to catecholamines. B. Interactive Effects pi Neurotransmitters The interaction profile of acetylcholine and norepinephrine on islet insulin secretion should be measured in ob/ob and lean islets to establish the relative importance of inhibitory and stimulatory input. Sympathetic dominance could indicate that increases in sensitivity and responsiveness to norepinephrine of the ob/ob islet B cell might be compensating for increased acetylcholine potentiated insulin secretion. C. Pancreatic Insulin pi Pre-Obese Ob/ob Mice As a preliminary study to experiment D, pancreata of 2 wk old ob/ob mice should be extracted for total pancreatic insulin and compared with total pancreatic insulin of lean littermates. Differences in pancreatic insulin content would be indicative of altered islet insulin content or islet number early in development. This would suggest that insulin secretion from islets may also be altered by 2 wks of age in ob/ob mice. D. Insulin Secretion from Pre-Obese Ob/ob Mouse Islets Pancreatic islets should be isolated from pancreata of 2 wk old ob/ob and lean mice. Dose response of insulin to glucose should be performed for these islets to determine if 53 during a period of i vivo hypoglycemia, islets are hypersensitive or hyperresponsive to glucose stimulus. If islets are hypersensitive to glucose at this early age, we would suspect that altered islet insulin secretion in response to glucose is a primary factor initiating hyperinsulinemia and obesity in ob/ob mice. Selected concentrations of norepinephrine should also be used to determine islet sensitivity to norepinephrine. If alterations in sensitivity to norepinephrine is apparent at 2 wks of age without alterations in glucose sensitivity, we would suspect that alterations in glucose sensitivity could be a consequence of alterations in sympathetic input to the islets. E. Islets pi Adrenalectomized Ob/ob Mice Since adrenalectomy normalizes plasma insulin in ob/ob mice, the effects of adrenalectomy on islet size and insulin secretion should be studied. Ob/ob mice should be adrenalectomized at 4 wks of age. At 8 wks of age, islets should be isolated and sensitivity to glucose and norepinephrine measured. If adrenalectomy normalizes islet size, sensitivity to glucose and sensitivity to norepinephrine, we would suspect that the increased glucocorticoid levels in ob/ob mice may be the cause of alterations in islet response to glucose and altered neural activity to the islets of ob/ob mice. REFERENCES REFERENCES AHREN, B., AND I. LUNDQUIST. Modulation of basal insulin secretion in the obese, hyperglycemic mouse. Metabolism 31: 172-179, 1982. ARKHAMMAR, P., T. NILSSON, P. RORSMAN, AND P.-O. BERGGREN. Inhibition of ATP-regulated K channels precedes depolarization-induced increase in cytoplasmic free Ca2+ concentration in pancreatic B-cells. J. Biol. Chem. 262: 5448-5454, 1987. ATWATER, I., C.M. DAWSON, A. SCOTT, G. EDDLESTONE, AND E. ROJAS. The nature of the oscillatory behaviour in electrical activity from pancreatic B-cell. Horm. Metab. Res. Suppl 10: 100-107, 1980. BAETENS, 0., y. STEFAN, M. RAVAZZOLA, F. MALAISSE-LAGAE, D.L. COLEMAN, AND L. ORCI. Alteration of islet cell populations in spontaneously diabetic mice. Diabetes 27: 1-7, 1978. BATCHELOR, B.R., J.S. STERN, P.R. JOHNSON, AND R.J. MAHLER. Effects of streptozotocin on glucose metabolism, insulin response, and adiposity in ob/ob mice. Metabolism 24: 77-91, 1975. BELOFF-CHAIN, A., M.E. NEWMAN, AND K.R.L. MANSFORD. Factors influencing insulin and glucagon secretion in lean and genetically obese mice. Horm. metab. Res. 9: 33-37, 1977. BEREITER, D.A., F. ROHNER-JEANRENAUD, H.-R. BERTHOUD, AND B. JEANRENAUD. CNS modulation of pancreatic endocrine function: multiple modes of expression. Diabetologia 20: 417-425, 1981. BERGMAN, R.N., AND R.E. MILLER. Direct enhancement of insulin secretion by vagal stimulation of the isolated pancreas. Am. J. Physiol. 225: 481-486, 1973. BERTHOUD, H.R., AND B. JEANRENAUD. Acute hyperinsulinemia and its reversal by vagotomy after lesions of the ventromedial hypothalamus in anesthetized rats. Endocrinology 105: 146-151, 1979. BERTHOUD, H.R. The relative contribution of the nervous system, hormones, and metabolites to the total insulin response during a meal in the rat. Metabolism 33: 18-25, 1984. 54 55 BEST, L., AND W.J. MALAISSE. Nutrient and hormone- neurotransmitter stimuli induce hydrolysis of polyphosphoinositides in rat pancreatic islets. Endocrinology 115: 1814-1820, 1984. BLACK, M.A., L.A. FOURNIER, H.M. HEICK, AND N. BEGIN-HEICK. Different insulin-secretory responses to calcium channel blockers in islets of lean and obese (ob/ob) mice. Biochem. J. 249: 401-407, 1988. BOISSONNEAULT, G.A., M.J. HORNSHUH, J.W. SIMONS, D.R. ROMSOS, AND G.A.LEVEILLE. Oxygen consumption and body fat content of young lean and obese (ob/ob) mice. Proc. Soc. Exp. Biol. Med. 157: 402-406, 1978. BONNER-WEIR, S. Pancreatic Islets: Morphology, Organization, and Physiological Implications. In: Molecular and Cellular Biology pi Diabetes Mellitus 1Vol. 1): Insulin Secretion, edited by B. Draznin, S. Melmed, and D. LeRoith. New York: Alan R. Liss, Inc., 1988, p. 1-11. BRAY, G.A., AND D.A. YORK. Hypothalamic and genetic obesity in experimental animals: an autonomic and endocrine hypothesis. Physiol. Rev. 59: 719-809, 1979. BRAY, G.A. Autonomic and endocrine factors in the regulaion of energy balance. Fed. Proc. 45: 1404-1410, 1986. CAMPFIELD, L.A., AND F.J. SMITH. Modulation of insulin secretion by the autonomic nervous system. Brain Res. Bull. 5: 103-107, 1980. CAMPFIELD, L.A., AND F.J. SMITH. Alteration of islet neurotransmitter sensitivity following ventromedial hypothalamic lesion. Am. J. Physiol. 244 (Regulatory Integrative Comp. Physiol. 13): R635-R640, 1983. CAMPFIELD, L.A., F.J. SMITH, AND R.E. ESKINAZI. Glucose responsiveness and acetylcholine sensitivity of pancreatic B-cells after vagotomy. Am. J. Physiol. 246 (Regulatory Integrative Comp. Physiol. 15): R985-993, 1984. CAMPFIELD, L.A., F.J. SMITH, AND C. LARUE-ACHAGIOTIS. Temporal evolution of altered islet neurotransmitter sensitivity after VMH lesion. Am. J. Physiol. 251 (Regulatory Integrative Comp. Physiol. 20): R63-R69, 1986. CHEN, L., I KOMIYA, L. INMAN, K. MCCORKLE, T. ALAM, AND R.H. UNGER. Molecular and cellular responses of islets during perturbations of glucose homeostasis determined by is situ hybridization histochemistry. Proc. Natl. Acad. Sci. USA 86: 1367-1371, 1989. 56 COLEMAN, D.L. AND K.P. HUMMEL. The influence of genetic background on the expression of the obese (ob) gene in the mouse. Diabetologia 9: 287-293, 1973. COLEMAN, D.L. Obese and diabetes: two mutant genes causing diabetes-obesity syndromes in mice. Diabetologia 14: 141-148, 1978. COLL-GARCIA, E., AND J.R. GILL. Insulin release by isolated pancreatic islets of the mouse incubated in vitro. Diabetologia 5: 61-66, 1969. CURRY, D.L. Direct tonic inhibition of insulin secretion by central nervous system. Am J. Physiol. 244 (Endocrinol. Metab. 7): E425-E429, 1983. CURRY, D.L., AND J.S. STERN. Dynamics of insulin hypersecretion by obese Zucker rats. Metabolism 34: 791- 796, 1985. CURRY, D.L. Insulin content and insulinogenesis by the perfused rat pancreas: effects of long term glucose stimulation. Endocrinology 118: 170-175, 1986. DIANI, A.R., T. PETERSON, G.A. SAWADA, B.M. WYSE, B.J. GILCHRIST, A.E. HEARRON, AND A.Y. CHANG. Ciglitazone, a new hypoglycaemic agent. 4. effect on pancreatic islets of C57BL/6J ob/ob and C57BL/KsJ-db/db mice. Diabetologia 27: 225-234, 1984. DRAZNIN, B., AND R. DAHL. Cell Biology of Insulin Secretion. In: Molecular and Cellular Biology of Diabetes Mellitus LYC1.1): Insulin Secretion, New York: Alan R. Liss, Inc., 1988, p. 37-47. DUBUC, P.U. Effects of limited food intake on the obese- hyperglycemic syndrome. Am. J. Physiol. 230: 1474-1479, 1976a. DUBUC, P.U. Basal corticosterone levels of young ob/ob mice. Horm. Metab. Res. 9: 95-97, 1976b. DUBUC, P.U. The development of obesity, hyperinsulinemia, and hyperglycemia in ob/ob mice. Metabolism 25: 1567-1574, 1976C. DUBUC, P.U. Non-essential role of dietary factors in the development of diabetes in ob/ob mice. J. Nutr. 111: 1742-1748, 1981. DULLOO, A.G., AND D.S. MILLER. Obesity: a disorder of the sympathetic nervous system. Wld. Rev. Nutr. Diet. 50: 1- 56, 1987. 57 DUNBAR, J.C., AND M.F. WALSH. Glucagon and insulin secretion by islets of lean and obese (ob/ob) mice. Horm. Metab. Res. 12: 39-40, 1980. ESTERHUIZEN, A.C., T.L.B. SPRIGGS, AND J.D. LEVER. Nature Of islet-cell innervation in the cat pancreas. Diabetes 17: 33-36, 1968. FINDLAY, J.A., K.A. ROOKLEDGE, A. BELOFF-CHAIN, AND J.D. LEVER. A combined biochemical and histological study on the islets of Langerhans in the genetically obese hyperglycaemic mouse and in the lean mouse, including observations on the effects of streptozotocin treatment. J. Endocrinol. 56: 571-583, 1973. GEPTS, W., J. CHRISTOPHE, AND J. MAYER. Pancreatic islets in mice with the obese-hyperglycemic syndrome: lack of effect of carbutamide. Diabetes 9: 63-69, 1960. G LOEN, A., A.J. COLLET, G. GUAY, AND L.J. BUKOWIECKI. Insulin stimulates in vivo cell proliferation in white adipose tissue. Am. J. Physiol. 256 (Cell Physiol. 25): C190-C196, 1989. GILL, J.L. Design and Analysis pi Experiments i_ the Animal n and. Medical Sciences. Ames: Iowa State Univ. Press, 1981. HAHN, H.-J., M. ZIEGLER, H. JAHR, R. BUTTER, AND K.D. KOHNERT. Investigations on isolated islets of Langerhans in vitro XIII. Experiments concerning the preparation conditions with collagenase. Endokrinologie 67: 67-78, 1976. HAN, P.W., Y.-K. YU, AND S.L. CHOW. Enlarged pancreatic islets of tube-fed hypophysectomized rats bearing hypothalamic lesions. Am. J. Physiol. 218: 769-771, 1970. HELLMAN, B., S. BROLIN, C. HELLERSTRflM, AND K. HELLMAN. The distribution pattern of the pancreatic islet volume in normal and hyperglycaemic mice. Acta Endocrinol. 36: 609-616, 1961. HENQUIN, J.C., M.-C. GARCIA, M.BOZEM, M.P. HERMANS, AND M. NENQUIN. Muscarinic control of pancreatic B cell function involves sodium-dependent depolarization and calcium influx. Endocrinology 122: 2134-2142, 1988. HENQUIN, J.C., AND H.P. MEISSNER. Significance of ionic fluxes and changes in membrane potential for stimulus- secretion coupling in pancreatic B-cells. Experientia 40: 1043-1051, 1984. 58 HERMANS, M.P., W. SCHMEER, AND J.C. HENQUIN. Modulation of the effect of acetylcholine on insulin release by the membrane potential of B cells. Endocrinology 120: 1765- 1772, 1987. INOUE, S., L.A. CAMPFIELD, AND G.A. BRAY. Comparison of metabolic alterations in hypothalamic and high fat diet- induced obesity. Am. J. Physiol. 233 (Regulatory Integrative Comp. Physiol. 2): R162-R168, 1977. JEANRENAUD, B. An hypothesis on the aetiology of obesity: dysfunction of the central nervous system as a primary cause. Diabetologia 28: 502-513, 1985. KARAKASH, C., F. ASSIMACOPOULOS-JEANNET, AND B JEANRENAUD. An anomaly of insulin removal in perfused livers of obese- hyperglycemic (ob/ob) mice. J. Clin. Invest. 57:1117- 1124, 1976. KARAM, J.H., G.M. GRODSKY, AND P.H. FORSHAM. Excessive insulin response to glucose in obese subjects as measured by immunochemical assay. Diabetes 12: 197-204, 1963. KAKITA, K., K. O'CONNELL, AND M.A. PERMUTT. Pancreatic content of insulins I and II in laboratory rodents: analysis by immunoelectrophoresis. Diabetes 31: 841-845, 1982. KING, D.L., K.C. KITCHEN, AND W.L. CHICK. Pancreatic fi-cell replication: relation to insulin secretion. Endocrinology 103: 1321-1327, 1978. KNEHANS, A.W., AND D.R. ROMSOS. Norepinephrine turnover in obese (ob/ob) mice: effects of age, fasting and acute cold. Am. J. Physiol. 244 (Endocrinol. Metab. 7): E567- E574, 1983. KRAL, J.G. Morbid obesity and related health risks. Ann. Int. Med. 103: 1043-1047, 1985. KRAUSE, U.H., PUCHINGER, AND A. WACKER. Inhibition of glucose-induced insulin secretion in trypsin-treated islets of Langerhans. Horm. Metab. Res. 5: 325-329, 1973. KUHN, C.M., C. COCHRANE, M.N. FEINGLOS, R.S. SURWIT. Exaggerated peripheral responses to catecholamines contributes to stress-induced hyperglycemia in the ob/ob mouse. Pharm. Biochem. Behav. 26: 491-495, 1987. LACY, P.E., AND M. KOSTIANOVSKY. Method for the isolation of intact islets of Langerhans from the rat pancreas. Diabetes 16: 35-39, 1967. 59 LARSSON, L.-I., G.B. BODER, AND W.N. SHAW. Changes in the islets of Langerhans in the obese Zucker rat. Lab. Invest. 36: 593-598, 1977. LAVINE, R.L., N. VOYLES, P.V. PERRINO, AND L. RECANT. Functional abnormalities of islets of Langerhans of obese hyperglycemic mouse. Am. J. Physiol. (Endocrinol. Metab. Gastrointest. Physiol. 2): E86-E90, 1977. LEE, H.C., D.L. CURRY, AND J.S. STERN. Direct effect of CNS on insulin hypersecretion in obese Zucker rats: involvement of vagus nerve. Am. J. Physiol. 256: (Endocrinol. Metab. 19): E439-E444, 1989. LIN, P.-Y., D.R. ROMSOS, AND G.A. LEVEILLE. Food intake, body weight gain, and body composition of the young obese (ob/ob) mouse. J. Nutr. 107: 1715-1723, 1977. LOTEN, E.G., A. RABINOVITCH AND B. JEANRENAUD. In Vivo studies on lipogenesis in obese hyperglycaemic (ob/ob) mice: possible role of hyperinsulinaemia. Diabetologia 10: 45-52, 1974. LOUBATIERES-MARIANI, M.M., J. CHAPAL, R. ALRIC, AND A. LOUBATIERES. Studies of the cholinergic receptors involved in the secretion of insulin using isolated perfused rat pancreas. Diabetologia 9: 439-446, 1973. MACKAY, E.M., J.W. CALLAWAY, AND R.H. BARNES. Hyperalimentation in normal animals produced by protamine insulin. J. Nutr. 20: 59-66, 1940. METZ, S.A. Epinephrine impairs insulin release by a mechanism distal to calcium mobilization: similarity to lipoxygenase inhibitors. Diabetes 37: 65-73, 1988. MILLER, R.E. Pancreatic neuroendocrinology: peripheral neural mechanisms in the regulation of the islets of Langerhans. Endocrine Rev. 2: 471-494, 1981. MORGAN, N.G., AND W. MONTAGUE. Studies on the mechanism of inhibition of glucose-stimulated insulin secretion by noradrenaline in rat islets of Langerhans. Biochem. J. 226: 571-576, 1985. NAKADATE, T., T. NAKAKI, T. MURAKI, AND R. KATO. Regulation of plasma insulin level by a -adrenergic receptors. Eur. J. Pharmacol. 65: 421-424, 1 80. 60 NILSSON, T., P. ARKHAMMAR, P. RORSMAN, AND P.-O. BERGGREN. Inhibition of glucose-stimulated insulin release by a2- adrenoceptor activation is parallelled by both a repolarization and a reduction in cytoplasmic free Ca concentration. J. Biol. Chem. 263: 1855-1860, 1988. 2+ OLEFSKY, J.M., AND O.G. KOLTERMAN. In-vivo studies of insulin resistance in human obesity. In: Recent Advances in Obesity Research III, edited by P. Bjorntorp, M. Cairella, and A.H. Howard. London: John Libbey, 1980, p. 254-267. PANCINI, A.R., AND E.L. TOLMAN. Enhanced insulin secretion and calcium uptake by Zucker "fatty" rat islets. Horm. Metab. Res. 13: 430-433, 1981. ROHNER-JEANRENAUD, F., A.-C. HOCHSTRASSER, AND B. JEANRENAUD. Hyperinsulinemia of preobese and obese fa/fa rats is partly vagus nerve mediated. Am J. Physiol. 244 (Endocrinol. Metab. 7): E317-E322, 1983. ROOTH, P., K. GRANKVIST, AND I.-B. TALJEDAL. is vivo fluorescence microscopy of blood flow in mouse pancreatic islets: adrenergic effects in lean and obese- hyperglycemic mice. Microvasc. Res. 30: 176-184, 1985. ROSARIO, L.M., I. ATWATER, AND E. ROJAS. Membrane potential measurements in islets of Langerhans from oblob obese mice suggest an alteration in [Ca ]i-activated K permeability. Quart. J. Exp. Physiol. 70: 137-150, 1985. SANCHEZ-ANDRES, J.V., C. RIPOLL, AND B. SORIA. Evidence that muscarinic potentiation of insulin release is initiated by an early transient calcium entry. FEBS (Fed. Eur. Biochem. SOC.) Lett. 231: 143-147, 1988. SCHADE, D.S., AND R.P. EATON. Insulin secretion by perfused islets from the obese Zucker rat. Proc. Soc. Exp. Biol. Med. 149: 311-314, 1975. SCHULZ, A., AND A. HASSELBLATT. Phentolamine, a deceptive tool to investigate sympathetic nervous control of insulin release. Arch. Pharmacol. 337: 637-641, 1988. SHINO, A., T. MATSUO, H. IWATSUKA, AND 2. SUZUOKI. Structural changes of pancreatic islets in genetically obese rats. Diabetologia 9: 413-421, 1973. SMITH, F.J., AND L.A. CAMPFIELD. Pancreatic adaptation in VMH obesity: in vivo compensatory response to altered neural input. Am. J. Physiol. 251 (Regulatory Integrative Comp. Physiol. 20): R70-R76, 1986. 61 STORLIEN, L.H. The ventromedial hypothalamic area and the vagus are neural substrates for anticipatory insulin release. J. Auto. Nerv. Syst. 13: 303-310, 1985. STUNKARD, A.J., T.I.A. SORENSEN, C.HANIS, T.W. TEASDALE, R. CHAKRABORTY, W.J. SCHULL, AND F. SCHULSINGER. An adoption study of human obesity. N. Engl. J. Med. 314: 193-198, 1986. ULLRICH, 8., AND C.B. WOLLHEIM. Islet cyclic AMP levels are not lowered during a -adrenergic inhibition of insulin release: studies witE epinephrine and forskolin. J. Biol. Chem. 259: 4111-4115, 1984. VAN ITALLIE, T.B. Health implications of overweight and obesity in the United States. Ann. Int. Med. 103: 983- 988, 1985. VANDER TUIG, J.G., K.A. CRIST, AND D.R. ROMSOS. Temporal adjustments in sympathoadrenal activity in rats with obesity-producing hypothalamic knife cuts. Proc. Soc. Exp. Biol. Med. 185: 134-140, 1987. WALKER, H.C., AND D.R. ROMSOS. Effects of cimaterol, a fi- adrenergic agonist, on energy metabolism in ob/ob mice. Am. J. Physiol. 255 (Regulatory Integrative Comp. Physiol. 24): R952-R960, 1988. WARWICK, B.P., AND D.R. ROMSOS. Energy balance in adrenalectomized ob/ob mice: effects of dietary starch and glucose. Am J. Physiol. 255 (Regulatory Integrative Comp. Physiol. 24): R141-R148, 1988. WESTMAN, S. Development of the obese-hyperglycaemic syndrome in mice. Diabetologia 4: 141-149, 1968. WOLLHEIM, C.B., AND G.W.G. SHARP. Regulation of insulin release by calcium. Physiol. Rev. 61: 914-973, 1981. YOSHIMATSU, H., A. NIIJIMA, Y. OOMURA, K. YAMABE, AND T. KATAFUCHI. Effects of hypothalamic lesion on pancreatic autonomic nerve activity in the rat. Brain Res. 303: 147-152, 1984. MICHIGAN STATE UNIV. LIBRARIES lllll"NI"HIIWIIHWIWll1WWWHIHIVIWll 31293007903218