.1- I! :11‘4 n ‘ inn . I .H "Earn 1. «5. ( Hi .q I. 1 6.. a... L! 31.: ‘f . .3. I. .11!!! 1.39.321. {n.zulch)... ». .s I Alsi £3 :4::..N:E¢r.v II: «tuna, .q. . n 1.. .. 3,5”. 1.935. t... . , u. .1 !. Irv: II I) x Euswo. 35‘...“ .| 511.1 -s' 5A1. . I 4. .Hp‘lfi I. .17! ~4AruA;A..) 3.1.15.1!» :0. ~b! ..\5. Av Lin!’ y ‘ {t z. .3.‘ \‘..)'q.w. , .1. :1: \X..V.n:7. {mks-s & RSITY UBRARIES ” llllllllllllll "LLLL'LL l LLLLLL LLLLLLLLL 131293 This is to certify that the dissertation entitled CELLULAR MECHANISMS OF INSULIN SECRETION IN OB/OB MICE presented by Neng-Guin Chen has been accepted towards fulfillment of the requirements for l Ph.D. Nutritional Biochemistry ' degree in 7! éajor professor 3-11—96 Date MS U it an Affirmative Action/Equal Opportunity Institution 0-12771 LIBRARY Michigan State University PLACE It RETURN “”00 remove this modicum your «cord. TO AVOID FINES mum on or baton dot. duo. DATE DUE DATE DUE DATE DUE inflmjml f L i _ 3mijr‘1 MSU I. An Affirmative WM Opportunity lm WM! CELLULAR MECHANISMS OF INSULIN SECRETION IN OB/OB MICE BY Neng-Guin Chen A DISSERTATION Submitted to Michigan State university in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Food Science and.numan Nutrition 1996 ABSTRACT CELLULAR MECHANISMS OF INSULIN SECRETION IN OB/OB MICE BY Neng-Guin Chen Obesity is a major nutritional disorder in the Western world. Hyperinsulinemia caused by hypersecretion of insulin from the pancreas is one of the earliest detected abnormalities in the development of obesity. The aim of the present work was to identify the cellular basis for these early-onset abnormalities in regulation of insulin secretion in a genetically obese animal, the ob/ob mouse. Pancreatic islets of 2-wk-old ob/ob mice and their lean littermates were perifused with various insulin secretagogues. First, islets were perifused with glucose (10 or 20 mm). Islets of both phenotypes responded to glucose similarly. Acetylcholine (ACh) and cholecystokinin (CCK) potentiate glucose-induced insulin secretion via activation of the phospholipase Cfprotein kinase C (PKC) signaling pathway. Insulin secretion from islets of ob/ob mice was abnormally enhanced in response to ACh or CCK. Islet responsiveness to ACh was greater in islets from ob/ob mice than in islets from lean mice even after islets were cultured for up to 12 days. This phenotype-specific effect of ACh was mimicked by phorbol-lz-myristate-13-acetate (PMA, a PRC agonist) . PKC enhances insulin release by activating voltage-dependent Ca2+ channels (VDCCs) as well as by post-VDCC mechanisms that directly enhance the exocytotic machinery. Insulin secretion from islets of both phenotypes perifused with BAY K8644 (a L-type, VDCC agonist) increased similarly, suggesting that L-type VDCCs in islets of ob/ob mice function normally when directly activated. Addition of ACh or PMA to islets that had been directly activated with BAY K8644 (lo uM) caused a further equal increase in insulin secretion in both phenotypes. This suggests that the mechanism whereby PRC activates L-type, VDCCs is altered in islets from ob/ob mice. This PRC-mediated alteration in insulin secretion persists even when islets are cultured for up to 12 days. I dedicate my whole heart to Jesus Christ, without whose encouragement and love, this work would. not have been possible. 'By' wisdom the Lord laid the earth's fOundations, by understanding he set the heavens ianlace; by his knowledge the deeps were divided, and the clouds let drop the dew." Proverbs 3:19-20 iv ACKNOWLEDGMENTS The research.presented in this dissertation would not have been possible without the assistance and support of numerous individuals whose direct and indirect contribution are an integral part of this work. First, I would like to thank my mentor, Dr. Dale R. Romsos, for his outstanding help and direction. His patience as a father and encouragement as a great teacher inspire my interest in studyingzmedical research and complete this dissertation. I will always hold his advice and insight in the highest regard, as I have the deepest respect and admiration for Dr. Ramsos as both an outstanding scientist and a good friend. I am also indebted to my committee members. Dr. Maurice R. Bennink contributed his expertise in my dissertation research. Dr. William. G. Helferich has been most generous with his time and his great support throughout the whole process. Dr. Laryssa N. Kaufman provided her expertise in my research study and her great assistance. I wish to give special recognition to Dr. werner G. Bergen who served as a friend and provided consistent encouragement to me. In addition, I am.grateful Dr. Twylla. M. Tassava for her help and freindship, specially in the early years when I was only a young pup still wet behind the ears. She impressed upon me the importance of attitudes and priciples to conduct laboratory work. Special thanks go to my long time friend Ross Santell for sharing the successes and failures that go along with.being a graduate student. Furthermore, I thank.my parents and.my brothers and sisters-in-law for their love and support as well as for having confidence in me and giving me the freedom.to do as I choose. Finally, I acknowledge the grace of God for giving me the motivation and ability to initiate and finish this project and this dissertation. vi TABLE OF CONTENTS CHAPTER I. INTRODUCTION 0 0° 0- 1 CHAPTER II. REVIEW OF LITERATURE A. Eyperinsulinemia - a common abnormality in obese animals B. Approaches to study insulin secretion . .. .. 10 1. In vivo study . .. .. 10 2. Pancreas perfusion . .. .. 11 3. Islet preparations and long term culture of islets . .. .. 12 4. B-cell preparations . .. .. 14 5. Insulin producing cell lines . .. -- 15 C. Mechanism.of action of nutrient insulin secretagogues 1. Effects of glucose on insulin secretion .... .. 17 1.1 The general mechanism of glucose action . . . . . 17 1.2 Glucose-induced hypersecretion of insulin in genetically obese rodents .... .. 19 2. Effects of other nutrients on insulin secretion . .. .. 22 2.1 Mannose and Fructose Action . .. .. 22 2.2 Leucine and Arginine Action . .. .. 24 2.3 Malonyly-CoA and Long chain fatty acid action . .. .. 25 3. Summary of nutrient action on insulin secretion . .. .. 27 D. Mechanisms of action of neurohormone insulin secretagogues . .. .. 27 1. Effects of ACh and CCK on glucose-induced insulin secretion . .. .. 27 vii 1.1 The general mechanism.of ACh and CCK action . .. .. 27 1.2 ACh-, and CCK-mediated hypersecretion of insulin in obese rodents . . .. . 31 2. Effects of GIP and GLP on glucose-induced insulin secretion . . .. . 35 2.1 The general mechanism.of SIP and GDP action .. . .. 35 2.2 GIP-, and GLP-mediated glucose-induced insulin secretion in obese rodents . . .. . 37 3. Summary of neurohormonal action on insulin secretion . .. 0- 38 CHAPTER III. ENHANCED SENSITIVITY OP PANCREATIC ISLETS PROM PREOBESE Z-WEER-OLD OB/OB NICE TO NEUROHORNONAL STIMULATION OF INSULIN SECRETION A. Abstract . .. .. 40 3. Introduction .. . .. 41 C. Materials and methods . .. .. 45 1. Animals ooooo 45 2. Materials . .. .. 45 3. Experimental design . .. .. 47 4. Methods . .. .. 49 5. Sample and statistical analysis . . .. . 52 D. Results . .. .. 53 E. Discussion . .. .. 55 CHAPTER IV. PERSISTENTLY ENHANCED SENSITIVITY TO PROTEIN KINASE C STIMULATED INSULIN SECRETION IN CULTURED PANCREATIC ISLETS FROM OB/OB MICE A. Abstract . .. .. 72 B. IntrOduCtion o o o o o 73 C. Materials and.methods . .. .. 75 1. Animals . .. .. 75 2. Materials 0 .. .. 77 viii BIB] 3. Experimental design . . . . . 78 4. Methods . . . . . 79 5. Sample and statistical analysis . . . . . 81 D. Results . . . . . 82 E. Discussion CHAPTER V. SUMMARY AND RECOMMENDATIONS FOR FUTURE STUDY oooooloo BIBLIOGRAPHY 0 0 0° 0 108 ix Table Table Table 1. Table 2. LIST OF TABLES Body weight, fat pad, plasma insulin and glucose, pancreatic islet diameter, islet DNA content and islet insulin content in 2-wk-old ob/ob and lean mice ooooo 55 Diameter, DNA content, and insulin content of cultured pancreatic islets from 2-week-old ob/ob and lean mice ..... 83 Figure Figure Figure Figure Figure Figure Figure Figure Figure 7. LIST OF FIGURES Schematic representation of cellular events in the regulation of insulin secretion in pancreatic B-cells by glucose and neurohormones . .... .. 7 Insulin secretion from.pancreatic islets perifused in 1.7 mM glucose for 30 min and 20 mM glucose for 60 min . . . . . . 56 Threshold for glucose-induced insulin secretion .. . .. . 57 Threshold for ACE-and CCR-induced insulin secretion . 0- . .. 63 Synergistic effects of ACh and GIP on glucose-induced insulin secretion . .. - .. 64 Glucose, ACh and potassiumrinduced insulin secretion . ..,. .. 84 PRC potentiation of glucose-induced insulin secretion . .. . .. 87 L-type, VDCC potentiation of glucose-induced insulin secretion .... .. . 89 Effects of direct L-type, VDCC activation on ACh and PMA enhanced insulin secretion .. .... . 90 Figure 10. Effects of perifusate Ca2+ concentration on glucose-induced insulin secretion . .... .. 92 Figure 11. Forskolin and IBMx potentiation of glucose-induced insulin secretion . .. . .. 94 xi ACb CC! DAC GI] IE IP PR PI LIST OF ABBREVIATIONS ACh.......................... acetylcholine ccx......................... cholecystokinin DAG ..........................diacy1g1ycero1 GIP . . . . . . . . glucose-dependent insulinotropic polypeptide Im-oo-o-ooooooo-o-oooooisobutylmethylxanthine 11:3....................ooinositoltriphosphate pm .........................proteinkinuseg pxc.........................proteinkinnsec ch..........................phospholipasec PMAo-o-o-oo-oooo-oophorbol-lZ-myristate-l3-acetate VDCCs . . . . . . . . . . . . voltage-dependent calcium channels xii CHAPTER I . INTRODUCTION Obesity is a nutrition-related chronic disease and highly prevalent in western societies. No universal definition of obesity or overweight exists (Van Itallie 1985). It is, however, generally agreed that a body mass index (BMI) (weight in kilograms divided by the square of the height in meters) greater than 27 corresponds to overweight, and that a BMI greater than 30 corresponds to obesity. In the USA, the overall prevalence of obesity is about 12% for both men and women based on a BMI of greater than 30. (Gray 1989). In other words, almost 3 million people are affected by obesity. There are strong epidemiological associations of obesity’ with insulin resistance, glucose intolerance and hyperinsulinemia, which are all recognized risk factors for the subsequent development of non-insulin-dependent diabetes mell‘itus (NIDDM). Morbid obesity is also associated with a decreased life span and with an increased incidence of cardiovascular diseases, certain cancers, hepatic disorders, and respiratory In 2 disorders (Lew and Garfinkel 1979). During the past decade, there has been a 33.4 % increase in the incidence of overweight among US adults 20 years of age or older. These increases appear to be occurring throughout all race/sex groups, rather than being limited to certain subgroups. Factors such as dietary knowledge, attitudes, and practices, physical activity levels, and perhaps social, demographic, and health behavior factors are the likely candidates responsible for these increases in the prevalence of overweight (Ruczmarski et a1. 1994). Obese, patients often have increased fasting insulin concentrations. This increase in plasma insulin likely reflects B-cell hypersecretion rather than significant alterations in hepatic insulin clearance (Polonsky et a1. 1988). In agreement with this, pancreatic islet hyperplasia and enhanced glucose-induced insulin secretion are often observed in obese patients. The mechanisms behind this hypersecretion are unclear. An understanding of these mechanisms should aid in the-development of approaches to help treat obese subjects. Obese rodents are widely used as a model to study obesity. An advantage of these animals is that they have a 3 similar phenotype to human obesity. Such obese animals include ob/ob mice, db/db mice, and fa/fa Zucker rats. In ob/ob mice, the cause of obesity is an ob gene mutation. This ob gene has been cloned recently, and codes for a adipose-secreted hormone (leptin), presumably a satiety factor (Zhang et a1. 1994). 0b/ob mice can not produce a functional leptin, therefore, they develop metabolic alterations leading to gross obesity. Eyperinsulinemia is one of the early-onset abnormalities in these mice and is possibly caused by insulin hypersecretion, as likely occurs in human obesity. To further investigate the mechanisms behind this insulin hypersecretion, obese rodent models are much easier to study than human patients. In Zucker fa/fa rats (Blonz et a1. 1985), genetically obese (ob/ob) mice (Dubuc 1976) and at least in some obese humans (Stunff and. Bougneres 1994) the hyperinsulinemia precedes insulin resistance, which develops after significant obesity is evident. The primary cause of increased insulin secretion in fa/fa rats and ob/ob mice is unclear. Pancreatic islets obtained from adult fa/fa rats (Ruffert et al.1988) and ob/ob mice (Tassava et a1. 1992, Chen et a1. 1993) are enlarged, and their glucose-induced insulin secretion exhibits enhanced sensitivity and responsiveness. 4 Further, acetylcholine (ACh) potentiates insulin secretion from islets of these fa/fa rats and ob/ob mice to a much greater extent than observed in their lean counterparts (Tassava et al. 1992, Chen et al. 1993, Lee et al. 1993). The stimulatory effects of ACh on glucose-induced insulin secretion are much greater in islets of these obese animals than in islets of lean counterparts, even though their plasma insulin concentrations are similar. This early-onset abnormality in insulin secretion also persists in adult fa/fa rats and ob/ob mice (Tassava et al. 1992, Chen et al. 1993, Lee et al. 1993), suggesting that ACh potentiation of insulin secretion may play primary roles in both development and) maintenance of hyperinsulinemia in these animals. However, it is difficult to determine the primary mechanisms responsible for hypersecretion of insulin in these animals because of their marked pre-existing hyperinsulinemia. ACh is believed to potentiate glucose-induced insulin secretion via the phospholipase C (PLC) signaling pathway. Cholecystokinin (CCR), another insulin secretion potentiator that shares a common post—receptor signaling pathway with ACh in potentiation of glucose-induced insulin secretion, may also play a role in the development of insulin hypersecretion. This ra: I81 in< in: the pla wk- muc (At :0] gen ear in 1) 2)? 5 raises a further question of whether the enhanced responsiveness of islets to ACh reported in preobese fa/fa rats and in adult ob/ob mice is restricted to ACh per se, or includes enhanced responsiveness to CCK as well. Glucose-dependent insulinotropic polypeptide (GIP), an insulin secretion potentiator, mediates insulin secretion via the cAMP signal trasduction pathway. In adult ob/ob mice, plasma GIP concentrations are elevated and.may contribute to their hyperinsulimnemia (Flatt et al. 1989). In fa/fa rats (5- wk-old), GIP potentiates glucose-induced insulin secretion at much lower glucose concentrations than observed in lean rats (Atef et al. 1991). These observations suggest a potential role for GIP in the hyperinsulinemia characteristic of genetically obese rodents. My overall research objective is to identify the possible early-onset abnormalities in regulation of insulin secretion in ob/ob mice. Specific objectives and hypotheses are: 1) To understand effects of glucose on insulin secretion W1 Glucose-induced insulin secretion from islets of ob/ob mice is abnormally enhanced. 2) To understand effects of ACh and CCK on glucose-induced 3) 4) insulin secretion amnesia Pancreatic islets from.ob/ob mice exhibit enhanced responsiveness and sensitivity to the PLC signaling pathway. To understand effects of ACh-stimulated PLC signal transduction on glucose-induced insulin secretion W Pancreatic islets from.ob/ob:mice exhibit enhanced PRC and VDCCs activation via the ACh-stimulated PLC signal transduction pathway. To understand effects of GIP-stimulated cAMP signal transduction on glucose-induced insulin secretion. W Pancreatic islets from ob/ob mice exhibit enhanced GIP stimulation and PRA activation via the cAMP-mediated signaling pathway. Caz. Kt Glucose hannel I \. v PKA K” ‘filucose \ ATP metabolism AC P i. " r tein-P f [Ca2+], 4 GIP :// Protein-P l 69 PLC J J 0 ACh, ch Insulin secretion Figure 1. Schematic representation of cellular events in the regulation of insulin secretion in pancreatic B-cells by glucose and neurohormones Glucose metabolism leads to formation of ATP. ATP closes the ATP-sensitive K+ channels, resulting in depolarization, opening of VDCCs and then increases in [Ca“11. ACh and CCK activate the PLC signaling pathway by acting through G— proteins, resulting in the generation of DAG. DAG activates PRC, which phosphorylates VDCCs, and then increases [Ca“]1. GIP activates the cAMP signaling pathway by acting through G- proteins, resulting in the formation of cAMP. cAMP activates PKA, which phosphorylates VDCCs, and then elevates [Ca“1i. An increase of [Caifli is necessary for the initiation and maintenance in the insulin secretory process. PKC and PKA also exert main effects in insulin secretion by direct interaction with the exocytotic machinery in a manner that is not directly correlated to changes in [Ca“]1. CHAPTER II. REVIEW OF LITERATURE A. Hyperinsulinemia - a common abnormality in obese animals One of the main abnormalities in genetically obese rodents with recessive single gene mutations is hyperinsulinemia. Hyperinsulinemia in these genetically obese animals has been proposed as a key factor in the etiology of their obesity (Jeanrenaud 1985, Loten et a1. 1974) . For example, genetically ob/ob mice have elevated plasma insulin concentrations and pancreatic islet hyperplasia and hypertrophy (Dubuc 1976a). Hyperinsulinemia in ob/ob mice is apparent as early as 6 days of age (Dubuc 1981), and precedes other abnormalities such as increased adiposity (Boissonneault et a1. 1978), increased plasma corticosterone (Dubuc 1976b), hyperphagia (Lin et al. 1977) and peripheral insulin resistance (Bachelor et a1. 1975) . The early-onset hyperinsulinemia could play an important role in the development of obesity. For example, in ventromedial hypothalamic(VME)-lesioned obese rats (Berthoud 9 and Jeanrenaud 1979), Zucker fa/fa rats (Blonz et al. 1985), genetically obese (ob/ob) mice (Dubuc 1976a) and at least in some obese humans (Stunff and Bougneres 1994) the hyperinsulinemia precedes insulin resistance, which develops after significant obesity is evident. The hyperinsulinemia is likely secondary to increased rates of insulin secretion from the pancreas. Increased rates of insulin secretion in mm- lesioned rats (Campfield and Smith 1983) obviously originate from primary alterations induced by the lesion, likely including an increased parasympathetic drive to pancreatic B- cells with resultant enhanced ACh potentiation of glucose- induced insulin secretion. However, the primary cause of increased insulin secretion in ob/ob mice and fa/fa rats is unclear. Pancreatic islets obtained from adult ob/ob mice (Tassava et a1. 1992) and fa/fa rats (Ruffert et al. 1988) are enlarged, and they exhibit enhanced sensitivity and responsiveness to glucose-induced insulin secretion. It is. however, difficult to determine the primary mechanisms responsible for hypersecretion of insulin in these animals with marked pre-existing hyperinsulinemia. Preobese fa/fa rats (Atef et al. 1991 and Rohner- Jeanrenaud 1983) and adrenalectomized ob/ob mice (Mistry et 10 al. 1995) fed a high carbohydrate stock diet do not exhibit marked hyperinsulinemia and therefore have been used to examine early-onset abnormalities in insulin secretion. These preobese fa/fa rats and adrenalectomized ob/ob mice do not hypersecrete insulin in response to high concentrations of glucose (16-20 mM glucose). This implies that the hyperresponsiveness of islets from adult fa/fa rats and ob/ob mice to glucose is secondary to the prolonged hypersecretion of insulin that existed in these adult animals prior to study. Studies are needed to determine the mechanisms responsible for the initial hypersecretion of insulin in obesity' prone animals . B. Approaches to study insulin secretion 1. In vivo study A common approach to study insulin secretion is to measure plasma insulin concentrations after a meal or a glucose load. Insulin and C-peptide are secreted in equal amounts. Varying amounts of insulin are extracted from the portal blood by the liver. The amount of insulin extracted dependends on the physiological state of the subject. In contrast, C-peptide 11 undergoes no significant hepatic extraction (Licinio-Paixio et al. 1986) . Consequently peripheral plasma C-peptide concentrations are a more reliable indicator of insulin secretion than peripheral insulin concentrations, which reflect a balance between secretion and hepatic extraction. The advantage of this approach is that it is relatively easy to obtain blood samples. The disadvantage of this approach is that it does not directly measure insulin secretion, but rather it measures the overall balance of secretion and clearance. This approach is usually used in clinical trials for examining the hyperinsulinemic-related diseases. 2. Pancreas perfusion The pancreas can be perfused by cannulating the aortic segment containing the celiac artery. This permits one to maintain an intact innervation of the pancreas and thus to study the interaction of the vagus nerve and pancreas for example (Blonz et al. 1985). By stimulating or inhibiting these nerve endings, effects of the nervous system on secretion of glucagon, insulin or other pancreatic hormones can be studied. The advantage of this approach is that an intact pancreas can be studied. A disadvantage is that the 12 entire pancreas is used to make one observation. Consequently. a large number of animals is needed if various treatments are utilized. Additionally, there is uncertainity about the number of islets within a pancreas. It is thus difficult to compare phenotypic differences in insulin secretion without knowledge of the number of islets per animal. 3. Islet preparations and long term culture of islets The endocrine portion of the rat pancreas consists of individual islets scattered throughout the acinar parenchyma. The total volume of the islets comprises only a small percentage of the entire pancreas. For this reason a simple method for isolation of intact islets from the normal rat pancreas was developed. This method is based upon disruption of the acinar parenchyma by injecting Banks solution into the pancreas followed by incubation of the pancreas in collagenase. The isolated islets release insulin in vitro and appear intact and normal by light and electron microscopy after incubation (Lacy et al. 1967). Use of isolated islets has become a very common method to study insulin secretion because it is possible to directly measure insulin secretion, and because sufficient islets are often isolated from a single 13 animal to apply multiple treatments. One complication in use of intact islets is that islets contain cells other than B- cells. Freshly isolated islets have been used in.most studies of islet function in rodents. These studies have usually been confined to acute experiments lasting only a few hours, where it is difficult to separate in vitro treatment effects from all the prior complex influences that occurred in vivo. Use of cultured pancreatic islets to study glucose-induced insulin secretion provides an experimental approach with many advantages. First, by culturing islets it is possible to diminish carry over effects of the multiple influences of hormones that occurred in vivo, which for example may cause glucose memory in freshly isolated rat islets (Zawalich et al. 1988a, Zawalich et al. 1989a). Second, consumption of the low carbohydrate milk diet by animals before weaning might suppress glucose-induced insulin secretion from. freshly isolated islets. Islets from.preweaned rodents might not have completely developed a functional response to glucose. Therefore, prolonged culture of islets from.preweaned animals would avoid some of these problems. The duration of culturing islets is critical. Glucose- be Ve me l4 stimulated insulin release from rat islets cultured for 1,2 or 7 days showed nonmal glucose-concentration-dependent insulin release when compared to freshly isolated islets (Liang et al. 1992). Long term (wks) exposure of islets at high glucose concentrations (above 10 mM) causes glucose desensitization and toxicity, but this desensitization and toxicity can be avoided if shorter term exposures (several days) and lower glucose concentrations are used (Robertson et al. 1994) . Therefore, it is necessary to examine effects of a short term exposure of pancreatic islets from mice or rats to stimulatory concentrations of glucose (10 or 30 mM glucose) to determine the appropriate glucose concentrations for islet culture. 4. B-cell preparations Pancreatic islets isolated by freehand microdissection or by collagenase treatment of the pancreas are extensively used for studying mechanisms of insulin secretion. However, the isolated islet is still a complex model, not only because it contains at least four types of endocrine cells but also because other structural elements are present, such as blood vessels, a surrounding connective tissue capsule and basement membranes. The latter structures constitute possible barriers 15 which could well interfere with molecules entering or leaving the incubated, isolated islets. By using a single B-cell, it is possible to study the direct effects of any substance on the isolated cells; this approach permits more detailed investigations of transport kinetics uncomplicated by consideration of diffusion in an extracellular space (Lermnark 1974). There are two common methods to prepare single cells from isolated pancreatic islets. One method uses dispase, a new proteolytic enzyme (Ono et al. 1977), another uses EGTA and Ca-free HEPES-buffer Krebs-Ringeeredium.(Lernmark 1974). The single chells are difficult to separate from other islet cells, therefore, B-cell preparations are often contaminated with other cell types. Another limitation is that B-cells often do not secrete as much insulin as islets. 5. Insulin-producing cell lines During the last decade, some permanent insulin-producing tissue culture cell lines such as RIN mSF, RINr, and.AtT-20ins have been developed for study of insulin synthesis and secretion (Lenzen and Tiedge 1992). The advantage of this approach is that it avoids the need to handle animals and it PE fo be are 16 is easy to obtain a huge number of these cells. However, a disadvantage of these tumour cell lines is that, in contrast to normal pancreatic B-cells, they usually show an abnormal insulin secretorprattern in response to glucose stimulation. These cells are often unresponsive to glucose stimulation in the normal physiological concentration range. Only cells derived from a hamster insulinoma (HIT-T15) cell line retain the ability to respond to glucose, therefore, these cells are widely used as a model for B-cells. HIT-T15 cells actually respond at lower concentrations of glucose than normal B-cells (Santerre et al. 1981). This concentration difference results from the expression of the low affinity GLUT-2 glucose transporter in B-cells and both GLUT-2 and high.affinity GLUT- 1 transporters in HIT cells (Inagaki et al. 1992) . These permanent insulin-producing cell lines are an excellent model for the manipulation of gene expression. For example, they can be produced in large quantities through tissue culture, and are suitable for transplantation into diabetic animals, so the function of such genetically modified insulin-producing cells, for normalization of a diabetic state, can be elucidated. 3t the Pla f0: 17 C. Mechanisms of action of nutrient insulin secretagogues 1. Effects of glucose on insulin secretion 1.1 The general mechanism of glucose action Glucose exhibits a dual function in pancreatic B-cells. Glucose is both a fuel and, at millimolar concentrations, a physiological stimulus for insulin secretion and insulin biosynthesis. This dual function of glucose has been the theoretical basis for the concept of a signal function of fuel metabolism for the initiation of insulin secretion (Lenzen 1992) . In this system, a device to translate changes in the blood glucose concentrations into corresponding signal- generating metabolic flux rates is required for initiation of insulin secretion in the B-cells. Glucose uptake by the B- cells via facilitated diffusion is usually not rate limiting for glucose utilization and therefore, cannot serve a signal- generating function for initiation of insulin secretion (Lenzen 1992) . Glucokinase catalyzes the first rate-limiting step in glycolysis, the phosphorylation of glucose, and is therefore in a prime position to sense and respond to ambient plasma glucose concentrations and serve as a signal generator for the initiation of glucose-induced insulin secretion in B- 18 cells (Lenzen and Panten 1988) . Hexokinase with a low Rll also catalyzes glucose phosphorylation. However, it is inhibited to a large extent by glucose-G-P, leaving glucokinase to play the major role in phosphorylation of glucose (Schaftingen et al. 1994). Simultaneous phosphorylation of glucose to glucose-G-P, catalyzed by glucokinase, and dephosphorylation of glucose-6-P to glucose, catalyzed by glucose-6-phosphatase, has been termed glucose cycling (Rhan et a1. 1995). Glucose cycling may also play a role in regulation of insulin secretion. The enhanced glucose cycling may contribute to the increase insulin secretory process. It should be noted that although glucose plays an important role in insulin secretion, glucose per se is not the stimulus for insulin secretion. Rather, the end product of glucose metabolism, ATP is the primary signal messenger in the B-cells (Zawalich and Rasmussen 1990). When extracellular glucose rises to high values (around 10 mM or higher), the ATP content of the B-cell increases. This rise in ATP content inhibits Rt efflux through specific ATP sensitive R’ channels in the plasma membrane. As a consequence, the membrane depolarizes and voltage-dependent, dihydropyridine-sensitive Ca" channels (VDCCs) open. The resulting influx of Ca“ leads 19 to an increase in intracellular Ca2t which is one of the major signal transduction messengers generated in response to this increase in the glucose concentrations. A rise in intracellular Cah'is necessary for initiation and maintenance of insulin granule mobilization and exocytosis. Furthermore, the rise in Ca" also acts to cause the opening of Ca“- sensitive Rt channels in the membrane, resulting in an increase R? efflux, and a repolarization of the membrane to thereby close the voltage-dependent Ca channels. This would help control the magnitude of insulin secretion. 1.2 Glucose-induced hypersecretion of insulin in genetically obese rodents Isolated.pancreatic islets from genetically obese (ob/ob) adult mice secrete 1 to 12 fold more insulin in response to glucose than their lean counterparts (Lavine et al.1977). At 8 wks of age, ob/ob mouse islets are both hypersensitive and hyperresponsive to glucose stimulation even when comparing ob/ob islets of the same size as their lean counterparts (Chen et. al. 1993). Islets from 8-wk-old ob/ob mice possess a lower glucose threshold (1.9 1 0.1 mM glucose) than islets from their lean counterparts (4.8 i: 0.1 mM glucose) as 20 determined by a glucose gradient (Chen et al. 1993). In addition, islets from ob/ob mice exhibit a greater capacity for insulin secretion (4.1 1 0.1 fmole insulin secretion . islet'1 . min") than islets from their lean counterparts (2.1 1 0.1 fmole . islet'l- min") . This increased responsiveness of islets from ob/ob mice to glucose was determined by challenging islets with 20 mM glucose. Even after food deprivation (24 hr), islets from ob/ob mice still exhibit a 56% lower glucose threshold and a 64% greater capacity for insulin secretion than islets from lean mice (Chen et a1. 1993). Therefore, increased islet sensitivity and responsiveness to glucose is an obvious contributing factor to hyperinsulinemia in adult ob/ob mice. However, it is unclear to what extent these alterations in. pancreatic insulin secretion contribute to early development of hyperinsulinemia and obesity in ob/ob mice. Glucose cycling is greater in islets from.ob/ob:mice than in islets from lean mice (Rhan et al. 1990). The increased glucose cycling is attributed to increased glucose-6- phosphatase activity which is observed in permeabilized ob/ob islets (Rhan et a1. 1995). It remains unclear whether this glucose cycling contributes to the enhanced insulin secretion 21 from islets of ob/ob mice. In islets of fa/fa rats, glucose transporter (GLUT 2), glucokinase, glycolytic intermediates. and end product ATP, key elements in glucose metabolism, are all abnormal. Furthermore, VDCCs, another major site for glucose action on insulin secretion, have enhanced activity in adult ob/ob mice (Black et al. 1985). Whether the abnormal VDCCs activities are primary or secondary to the prolonged elevations in insulin secretion characteristic of adult ob/ob mice is unclear. However, ATP-sensitive Rt channels activity is normal in pancreatic islets from adult ob/ob mice (Fournier et al. 1990) and ob/ob pancreatic B-cells cultured for 2-5 days (Rukulian et al. 1990) . Overall, the mechanism behind the glucose-induced hypersecretion of insulin by islets from genetically obese rodents is unresolved. It is clear that the hypersecretion of insulin presists even in culture. B-cells from adult ob/ob mice secreted significantly more insulin than their lean counterparts in response to a 20 mM glucose stimulus, even after being maintained in culture for a prolonged time (14 to 20 days) (Fournier et al. 1992) . This finding is consistent with the results from adult obese Zucker rats where it was shown that the islets from obese rats do not normalize their insulin 22 secretory patterns even after a three week culture period (Hayek 1979). These results imply that the persistently high secretory rates in islets from obese Zucker rats are possibly a phenomenon primary to the obesity. To verify whether hypersecretion of insulin. is a primary or a secondary metabolic alteration to obesity, islets from young preobese animals have been examined. Pancreatic islets from preobese fa/fa rats do not hypersecrete insulin in response to high concentrations of glucose (16 mM glucose)(Atef et al. 1991). This implies that the previously observed hyperresponsiveness of islets from fa/fa rats to glucose is secondary to the prolonged hypersecretion of insulin that existed in these animals prior to study. Likewise, the sensitivity and responsiveness of islets from adult ob/ob mice to glucose might be altered secondary to the prolonged hypersecretion of insulin that exists in these animals. Studies are needed to examine islets of younger ob/ob mice. 2. Effects of nutrients other than glucose on insulin secretion 2.1 Mannose and fructose action 23 Besides glucose, other sugars such as mannose and fructose are capable of stimulating insulin release from rat pancreatic B-cells. These sugars generate metabolic signals arising from their catabolism. Mannose is indeed less potent than is glucose in inducing insulin release. Glucose and mannose induce insulin release at threshold levels of 4 and 10 mM, half-maximal levels of 8 and 15 mM and maxinal levels of 15 and 20 mM. respectively (Zawalich et al. 1977) . Fructose alone has no effect on insulin secretion (Zawalich et al. 1977) . It may be surmised that the inability of fructose to provide the necessary signal for insulin release is due to the fact that metabolic flux never reaches an apparently critical value of about 50 pmol/islet/h. However, fructose is capable of augmenting insulin secretion in the presence of a substimulatory or stimulatory glucose concentration. Therefore, fructose is clearly a potentiator instead of an initiator. Recently, it has been found that, in islets as in hepatocytes. the activity of glucokinase is modulated by a regulatory protein which mediates the antagonistic effects of frucose-l-P and fructose-6-P upon glucose phosphorylation (Malaisse et al.1990) . Since islets are found to display limited but sizeable fructokinase 24 activity (Malaisse et al. 1989), the generation of fructose-1- P from exogenous fructose could conceivably favour, to a limited extent, glucose phosphorylation by glucokinase. Moreover, fructose-l-P may also be generated in islets from a triose and triose phosphate, e.g. from glyceraldehyde and glycerol - 3 -phosphate . 2.2 Leucine and arginine action Leucine , a branched chain amino acid, stimulates insulin secretion. It enhances insulin secretion from mouse and rat islets in vitro. Leucine metabolism shares part of a common pathway with glucose metabolism (citric acid cycle) , enhancing the ATP content of the B-cells. Leucine is transported into the B-cells, is deaminated, and generates 2-ketoisocaproic acid. The action of leucine and 2-ketoisocaproic acid on inducing insulin secretion is directly linked to phosphatidylinositol hydrolysis (Zawalich 1988c) . The events coupled to the hydrolysis of membrane inositol-containing phospholipids induced by leucine and 2-ketoisocaproic acid participate not only in their acute insulin stimulatory action, but also in their ability to induce time-dependent potentiation (memory) in isolated islets. Whether the ability of the amino acid leucine and its keto acid to induce insulin 25 release is altered in islets of ob/ob mice is unclear. Arginine, a positivly charged amino acid, also plays a role in nutrient-induced insulin secretion. Arginine alone failed to elicit an insulin response, however, the combination of arginine plus glucose could stimulate glucose-induced insulin secretion to a great extent in rat islets (Heinze and Steinke 1971). In the study of 5-day-old genetically fa/fa rats, arginine (20 mM) was able to potentiate glucose (16.6 mMD-induced insulin secretion 3-fold higher than in the basal state. No significant differences between the time course and the increase in insulin secretion were observed between preobese and lean rats (Atef et al. 1991). However, pancreatic islets of 17-day-old obese fa/fa rats were hyperresponsive to arginine when tested in vivo (Rohner-Jeanrenaud and Jeanrenaud 1985) or in vitro (Blonz 1985). The results suggest that the effect of arginine on glucose-induced insulin secretion is secondary to the development of obesity. 2.3 Malonyl-CoA and long chain fatty acids action Stimulation of insulin secretion by glucose is associated Lwith.inhibition of fatty acid oxidation as a consequence of a rise in the concentration of malonyl-CoA and increased lipid 26 synthesis (e.g. long chain acyl-CoA esters) in pancreatic islets. Exogenous fatty acids (ie B-hydroxy butyrate, palmitate) potentiate glucose-induced insulin release, possibly by providing the acyl groups for lipid synthesis (Goberna et al. 1974 ). Exogenous long chain fatty acids are known to cause insulin release and potentiate glucose-induced insulin secretion in isolated pancreatic islets (Vara et al. 1988) and the perfused pancreas (Campillo et al. 1979). Long chain fatty acids and in particular myristate and palmitate markedly potentiated glucose-induced insulin secretion in HIT cells. Both.malonyl-CoA and long chain acyl-CoAs esters serve as metabolic coupling factors when pancreatic B-cells are stimulated with glucose and other nutrient secretagogues (Corkey et al. 1989., Prentki et al. 1991). A very recent finding suggests that long chain fatty acids could induce functional, morphologic and metabolic abnormalities in. pancreatic B-cells consistent with the hypersecretion of insulin characteristic of islets of obese Zucker fa/fa rats. Nbrmal rat islets cultured for 1 week with free fatty acid showed enhanced glucose metabolism and B-cell hyperplasia which might further contribute to hyperinsulinemia (Milburn et al. 1995). 27 3. Summary of nutrient action on insulin secretion Glucose plays a major role on insulin secretion. Glucose metabolism, not glucose per se. controls the whole secretory process. Three major sites for coupling of glucose and changes in the rate of insulin secretion are glucokinase/glucose-6- phosphatase, the ATP-sensitive potassium channels and VDCCs. In islets from adult ob/ob mice, glucose-induced insulin secretion is abnormally enhanced. These alterations in glucose metabolism possibly involve impaired glucose cycling and VDCCs, rather than ATP-sensitive R channels. However, 1 whether these alterations are primary or secondary to the hypersecretion of insulin is unclear. Whether these alterations are found in younger ob/ob mice is still unknown. D. Mechanisms of action of neurohormone insulin secretagogues 1. Effects of ACh and CCR on glucose-induced insulin secretion 1.1 The general mechanism.of ACh and CCR action Acetylcholine is released upon stimulation of the vagal nerve or the mixed autonomic innervation of the pancreas 28 (Ahren et al. 1986). These nerves have their terminals in close proximity to B-cells. ACh can directly stimulate insulin secretion via its muscarinic receptor on the B-cell membrane. Glucose, a nutrient insulin secretagogue, plays a priming role on ACh-stimulated insulin secretion because the potentiation of ACh on insulin secretion is dependent on glucose concentrations. When islets are exposed to basal glucose concentrations, ACh is not able to initiate sustained insulin secretion. ACh only stimulates insulin release in the presence of physiological glucose concentrations (Zawalich et al 1989). The stimulatory effect of this neurotransmitter can be blocked by atropine, indicating its muscarinic nature. When muscarinic receptors are activated by ACh, glucose-induced insulin secretion is potentiated by PLC-mediated signal transduction pathway. The hydrolysis of phosphatidylinositol biphosphate (PIP2) by PLC leads to the generation of two important messengers, inositol triphosphate (IP,) and diacylglycerol (DAG). The IP3 released from PIP, promotes a rise in the cytosolic-free Ca2+ concentration by inducing endogenous Ca2+ mobilization. This process is caused by the release of Ca” from internal Ca” pools, which are 1P3 sensitive. 29 Another consequence of PLC-catalyzed phospholipid breakdown is the generation of DAG, which is considered an important signal molecule to activate protein kinase C (PRC). The major isoenzymes of PRC in mouse pancreatic islets are o- and 81,. When the enzyme exists in the cytoplasm, a pseudosubstrate is thought to bind to the substrate-binding site, rendering the kinase inactive. Binding of DAG produces a conformational change that results in dislocation of the pseudosubstrate from the active site, an event that actives the PRC. PRC is translocated from the cytosol to the plasma membrane to regulate several downstream effectors by phosphorylation. First, PRC phosphorylates the a subunit of the VDCCs to prolong VDCCs opening time and further increase Ca" influx. PRC also exerts stimulatory effects on B-cell stimulus-secretion coupling by direct interaction with the exocytotic machinery in a manner that is not directly correlated to changes in intracellular Ca" concentrations (Ammala et al. 1994). One of the target proteins phosphorylated by PRC is MARCRS , a serine/threonine kinase. IMARCRS is a calmodulinrbinding protein, the phosphorylation of which results in rapid release of calmodulin, which can then 3O activate calmodulin-dependent protein kinase (CAMPR). a component of the B-cell cytoskeleton. Activation of CAMPR could phosphorylate components of the exocytosis system (Liang and Matschinsky 1994). PRC can also directly exert feedback inhibition of the PLC system. Overall, PRC plays an important role in the regulation of second phase insulin secretion in islets and the time-dependent potentiation and. proemial sensitization of insulin secretion. Cholecystokinin is secreted by endocrine cells in the gut after feeding (Mutt 1980). CCR also serves as a neurotransmitter, and.is abundant in nerve fibers innervating the islets (Rehfeld et al. 1980). Various fragments of CCR, such as CCR-4. CCR-8, CCR-12, CCR-22, CCR-33, CCR-39. and CCR- 58 are formed by degradation of the CCR peptide precursor. The smaller fragments (CCR-4 and CCR-8) probably act as neurotransmitters. CCR-8, or larger fragments, with an intact C-terminus, enhance basal insulin release and potentiate the response to glucose and other secretagogues. The mechanism.of action of CCR is similar to ACh action except the specific CCR receptors and the different G-proteins to PLC on B-cells (Versphol et al. 1986, Schnefel et al. 1988). 31 1.2 ACh-, and CCR—mediated hypersecretion of insulin in obese rodents Ob/ob nice. The parasympathetic nervous system release of ACh within the pancreas has tremendous potential to contribute to hyperinsulinemia in ob/ob mice. Freshly isolated pancreatic islets from obese (ob/ob) adult mice are hyperresponsive to ACh. Islet responsiveness to ACh is several-fold greater rather in ob/ob mice than their lean counterparts in the presence of stimulatory glucose concentrations (15 or 20 mM) . even when islet size is standardized (Tassava et al. 1992, Chen et al. 1993) . In addition, islets of ob/ob mice have a lower ACh-stimulated glucose threshold when compare to lean mice. Mechanisms whereby ACh potentiates glucose-induced insulin secretion are not fully understood, but likely involve phosphoinositide hydrolysis, activation of protein kinase C, and downstream effectors. Any of these ACh-mediated signal transduction systems might be altered in islets from ob/ob mice. Adrenalectomized ob/ob mice. Adrenalectomy arrests further development of obesity in hyperinsulinemic genetically obese (ob/ob) mice; this probably involves the adrenalectomy- induced lowering of their plasma insulin concentrations. The If 32 composition of the diet is critical in lowering plasma insulin concentrations and retarding development of obesity in adrenalectomized ob/ob mice. In adrenalectomized ob/ob mice fed a nonpurified, high carbohydrate commercial diet, plasma insulin concentrations are markedly lowered to the same extent as in lean counterparts. However, consumption of a purified high carbohydate diet or a purified high glucose diet attenuate effects of adrenalectomy on plasma insulin concentrations (Okuda and Romsos 1994. Rang et al. 1992) . The lowering of plasma insulin concentrations in adrenalectomized ob/ob mice fed a nonpurified, high carbohydrate commercial diet is probably caused by diminished pancreatic insulin secretion. In adrenalectomized ob/ob mice fed this conmerical diet, insulin secretion from pancreatic islets still remains normal but ACh further potentiated glucose-induced insulin release when compared to their adrenalectomized lean counterparts. In contrast, insulin secretion from islets of adrenalectomized ob/ob mice fed a purified high glucose diet was enhanced much more than these ob/ob mice fed with this commercial diet, and addition of ACh even further potentiated the diet effect (Okuda and Romsos 1994; Mistry et al. 1995). Thus, adrenalectomy did not 33 completely normalize insulin secretion from pancreatic islets obtained from ob/ob mice even though plasma insulin concentrations were normalized in these mice. The regulation of insulin secretion within islets of ob/ob mice is possibly persistently defective in response to diet or ACh. Fa/fa Zucker rats. The most pronounced abnormality in insulin secretion reported in preobese fa/fa rats is islet responsiveness to ACh (Atef et al. 1991). The stimulatory effects of ACh on glucose-induced insulin secretion are much greater in islets of these obese rats than in islets of lean counterparts, even though their plasma insulin concentrations are similar. This abnormality in insulin secretion also persists in intact adult fa/fa rats, suggesting that ACh potentiation of insulin secretion may play primary roles in both development and maintenance of hyperinsulinemia in these Zucker fa/fa rats. Since ACh responsiveness of islets from preobese fa/fa rats (Atef et al. 1991) and adrenalectomized ob/ob mice (Mistry et al. 1995) is clearly enhanced, mechanisms of action now need to be examined. It is unknown if the sensitivity of these islets to ACh is altered, or whether ACh receptor or post-receptor events are altered. CCR shares a common post- 34 receptor signal transduction pathway with ACh in potentiation of glucose-induced insulin secretion. Effects of CCR on insulin release from these ob/ob mice or fa/fa Zucker rats have not been reported. This raises a further question of whether the enhanced responsiveness of islets to ACh reported in preobese fa/fa rats and in adrenalectomized ob/ob mice is restricted to ACh per se, or includes enhanced responsiveness to CCR as well. VMH-lesioned rats. Isolated pancreatic islets from ventromedial hypothalamic (VMH)-lesioned rats exhibit an enhanced glucose-induced insulin release, as observed in ob/ob mice, but marked decreases in sensitivity and responsivesness to ACh stimulation of insulin release (Campfield and Smith 1983), unlike what is observed in islets from ob/ob mice. Diminished responsiveness and sensitivity of VMH-lesioned islets to ACh has been interpreted as a secondary metabolic consequence to increased parasympathetic nervous system activity in these islets. These rats develop marked hyperinsulinemia, like ob/ob mice, and offer the advantage that the site of the primary defect is known to be in the central nervous system. However, no evidence is available for an enhanced parasymapthetic nervous system activity in islets 35 of ob/ob mice, suggesting fundamental differences in the neural-mediated contribution to hyperinsulinemia in ‘VMH- lesioned rats and ob/ob mice. 2. Effects of GIP and GLP on glucose-induced insulin secretion 2.1 The general mechanism.of GIP and GLP-1 action Insulin secretion is greater when glucose is given orally compared to intravenously. This has been attributed to the insulinotropic effect of certain gastrointestinal hormones that are released by oral glucose. GIP serves to potentiate the stimulatory effect of glucose on insulin release. Stimulation of insulin release by GIP administration has been shown in vivo. In vitro, GIP addition to isolated rat islets has been shown to potentiate glucose-induced insulin secretion. GIP potentiates insulin secretion via a signal transduction pathway involving adenylate cyclase coupled to GIP receptors via G-proteins, resulting in the formation of cAMP. By activating protein kinase A (PRA), cAMP promotes phosphorylation of the VDCCs and thereby to some extent increases Ca2+ influx (Zawalich and Rasmussen 1990). In 36 addition, PRA may directly interact with the exocytotic machinery in a manner that is not directly correlated to changes in intracellular Ca" concentrations (Berggren and Larsson 1993). Further, GIP also acts in concert with the PLC agonists ACh and CCR to synergistically potentiate glucose-induced insulin secretion (Zawalich 1988b and Zawalich et al. 1989b) . The expression of intracellular messengers generated by the combined action of the two classes of neurohormonal agonists (ACh, GIP) are observed only at high glucose concentrations (at least 7 mM glucose). GIP has been the most extensively investigated of the Lincretins. However, it is clear that GIP is not the sole mediator of the endocrine arm of the entero-insular axis. A number of glucagon-like peptides (GLP) are now recognized which have the ability to stimulate insulin secretion (Morgan 1992) . GLP-1 and glucagon are structurally related peptides arising from the tissue-specific processing of proglucagon; glucagon is the primary hormone secreted from pancreatic or- cells while GLP-1 is secreted from the intestinal L-cells in response to oral glucose. GLP-1 is cleaved to GLP-1(7-36) and this truncated form of GLP-1 is the major circulating form 37 following a meal in man (Qrskov et al.1987). The molar equivalent insulin secretory potency of GLP-1(7-36) is more powerful than GIP in human volunteers, although its circulating level does not rise as high as GIP in response to an oral glucose load or test meal( Rreymann et al. 1988; Takahashi et al. 1990) . GLP-I(7-36) has been found to potentiate glucose-induced insulin secretion in a manner that is similar to the GIP-stimulated signaling pathway (Lu et al. 1993) . However, the role of GLP-1(7-36) in the cellular mechanism of insulin secretion is still unclear. 2.2 GIP-,and GLP-mediated glucose-induced insulin secretion in obese rodents Obese animals appear to be particularly sensitive to the insulinotropic effect of gastrointenstinal hormones, as exaggerated insulin responses have been obeserved following 'the administration of many of these hormones, including GIP, GLP-1, CCR. Intestinal and plasma GIP concentrations are elevated in adult ob/ob mice and may contribute to their hyperinsulinemia. Not all genetically obese rodents exhibit high circulating concentrations of GIP. Zucker fatty (fa/fa) rats, whose degree of hyperinsulinemia is mild compared with 38 ob/ob mice, have normal plasma GIP concentrations in response to nutritional stimuli. In fa/fa rats (5-wk-old), GIP potentiates glucose-induced insulin secretion at much lower glucose concentrations than observed in lean rats (Chan et al. 1993). This result is possible due to the lower glucose threshold in these fa/fa rats rather than to islet sensitivity to GIP. However, GLP-1(7-36) lowers the glucose threshold more in fa/fa rats than their lean rats in perfused pancreas, but islet responsiveness to GLP-1(7-36) is similar in both phenotypes (Jia et al. 1995). These observations suggest a potential role for GLP-1(7-36) in the hyperinsulinemia characteristic of genetically obese rodents. However, whether effects of GLP-1(7-36) on insulin secretion are primary or secondary to early-onset abnormality of obesity is unclear. 3. Summary of neurohormonal action on insulin secretion Neurohormones play important roles in potentiating glucose-induced insulin secretion. These neurohormones regulate the insulin secretory process via distinct signaling transduction pathways. ACh and CCR act via the PLC signaling pathway and GIP acts via the cAMP. ACh alters glucose-induced insulin secretion in ob/ob 39 mice. However. the mechanism behind ACh action is unclear. Whether CCR shares the same PLC pathway as ACh in enhancing insulin release in ob/ob mice is unresolved. Mechanisms beyond the initial receptor actions of ACh and CCR may also be altered in islets of ob/ob mice. 0n the other hand, GIP via the cAMP pathway also potentiates glucose-induced insulin secretion. Whether this pathway also affects the hypersecretion of insulin in islets of ob/ob mice has been underinvestigated. Chapter III. Enhanced sensitivity of pancreatic islets from preobese 2-wk-old ob/ob mice to neurohormonal stimulation of insulin secretion (Published in Endocrinology 136:505- 511,1995) A. Abstract Insulin secretion from perifused islets of preobese, 2- week-old, genetically obese (ob/ob) mice and their lean littermates was examined to identify early-onset abnormalities in regulation of insulin secretion by ob/ob mice. The ob/ob mice were slightly hyperinsulinemic(+20%) and hypoglycemic (- 12%) at 2 weeks of age. Pancreatic islet size, DNA content, and insulin content were similar in ob/ob and lean mice. The responsiveness of islets to glucose, as determined by 20 mM glucose-induced insulin secretion, and the sensitivity of islets to glucose, as determined by the glucose threshold for insulin secretion, were unaffected by phenotype, but two insulin secretagogues that potentiate glucose-induced insulin secretion via activation of the phospholipase-C signal 40 41 transduction pathway (i.e.acetylcholine and cholecystokinin) were more effective in stimulating insulin secretion from islets of ob/ob mice than from islets of lean mice. Both responsiveness and sensitivity to acetylcholine and cholecystokinin potentiation of glucose-induced insulin secretion were enhanced in islets from ob/ob mice. Further, glucose-dependent insulinotropic polypeptide, which stimulates glucose-induced insulin secretion via activation of adenylate cyclase, interacted with acetylcholine to further augment differences in insulin secretion between islets from ob/ob and lean mice. The signal traansduction pathway common to acetylcholine and cholecystokinin, and cross-talk between this pathway and the glucose-dependent insulinotropic polypeptide singal transduction pathway are loci for early-onset defects in control of insulin secretion from islets of ob/ob mice (Endocrinology 13 6': 505-511, 1995) B . In troduc ti on Hyperinsulinemia and insulin resistance often co-exist in obese animal models and in obese humans. In ventromedial hypothalamic (VMH)-lesioned obese rats (Berthoud and Jeanrenaud 1979), Zucker fa/fa rats (Blonz et al. 1985), 42 genetically obese (ob/ob) mice (Dubuc 1976) and at least in some obese humans (Stunff and Bougneres 1994) the hyperinsulinemia precedes insulin resistance, which develops , after significant obesity is evident. The hyperinsulinemia is likely secondary to increased rates of insulin secretion from the pancreas. Increased rates of insulin secretion in mm- lesioned rats obviously originate from primary alterations induced by the lesion, likely including an increased parasympathetic drive to pancreatic B-cells with resultant enhanced acetylcholine (ACh) potentiation of glucose-induced insulin secretion (Campfield and Smith 1983). The primary cause of increased insulin secretion in fa/fa rats and ob/ob mice is unclear. Pancreatic islets obtained from adult fa/fa rats (Ruffert et al.1988) and ob/ob mice (Tassava et al. 1992, Chen et al. 1993) are enlarged, and they exhibit enhanced sensitivity and responsiveness to glucose-induced insulin secretion. Further, acetylcholine (ACh) potentiates insulin secretion from islets of these fa/fa rats and ob/ob mice to a much greater extent than observed in their lean counterparts (Tassava et al. 1992, Chen et al. 1993, Lee et al. 1993) . However, it is difficult to determine the primary mechanisms responsible for 43 hypersecretion of insulin in these animals because of their marked pre-exis ting hyperinsulinemia . Preobese fa/fa rats (Atef et al. 1991, Rohner- Jeanrenaud et al. 1983) and adrenalectomized ob/ob mice (Mistry et al. 1995) fed a high carbohydrate stock diet do not exhibit marked hyperinsulinemia and thereby have been used to examine early-onset abnormalities in insulin secretion. Pancreatic islets from these preobese fa/fa rats and adrenalectomized ob/ob mice do not hypersecrete insulin in response to high concentrations of glucose (16-20 mM glucose) (Atef et al. 1991, mistry et al. 1995) . This implies that the previously observed hyperresponsiveness of islets from adult fa/fa rats and ob/ob mice to glucose is secondary to the prolonged hypersecretion of insulin that existed in these animals prior to study. The most pronounced abnormality in insulin secretion reported in preobese fa/fa rats and in adrenalectomized ob/ob mice is in their responsiveness to ACh (Atef et al. 1991, Rohner-Jeanrenaud et al. 1983, Mistry et al. 1995). The stimulatory effects of ACh on glucose-induced insulin secretion are much greater in islets of these obese animals than in islets of lean counterparts, even though their plasma 44 similar . This early-onset insulin concentrations are abnormality in insulin secretion also persists in adult fa/fa rats and ob/ob mice (Tassava et al. 1992, Chen et al. 1993, Lee et al. 1993), suggesting that ACh potentiation of insulin secretion may play primary roles in both development and maintenance of hyperinsulinemia in these animals. The present aim of research work was conducted to examine insulin secretion in perifused pancreatic islets from 2-week- old ob/ob and lean mice. At 2 wk of age ob/ob mice are not are also only slightly yet visually obese . They hyperinsulinemic, and are hypoglycemic (Dubuc 1976) . Therefore, their islets have not yet experienced long term hypersecretion of insulin, and the secondary complications of pre-existing insulin resistance are not yet present. Islets were first challenged with 20 mM glucose to determine islets responsiveness to glucose. Our expectation was that glucose responsiveness of islets from preobese ob/ob mice would not be enhanced, which would be consistent with observations in normoinsulinemic, adrenalectomized ob/ob mice (Mistry et al. 1995) . Next, the minimum threshold for glucose-induced insulin secretion was examined. Again our expection, based on earlier observations in adrenalectomized ob/ob mice (Mistry et al. 45 1995), was that islets from 2-week-old ob/ob and lean mice would exhibit similar sensitivities to glucose. Then the involvements of ACh, cholecystokinin (CCR) and glucose- dependent insulinotropic polypeptide (GIP) in glucose-induced insulin secretion was examined; measurements included effects of ACh, CCR and GIP on the minimum threshold for glucose- induced insulin secretion, on the ACh and GIP potentiation of 10 mM glucose-induced insulin secretion. and on the sensitivity of islets to ACh- and CCR-induced insulin secretion. CCR shares a common postreceptor phospholipase-C signal transduction pathway with ACh in potentiation of gluocse-induced insulin secrtion (Zawalich and Rasmussen 1991), whereas GIP mediates insulin secretion via a signal transduction pathway involving adenylate cyclase (Zawalich 1988, Zawalich et al. 1989). Finally, the potential synergism of GIP and ACh in stimulation of insulin secretion from islets of 2-week-old ob/ob mice was explored. C. Materials and Methods 1. Animals Female preobese (ob/ob) mice and lean littermates (ob/+ or +/+) from our breeding colony (C57BL/6J-ob/+) were used at 2 46 wk of age. Mice were housed at 24°C and with a 12 h light, 12-h dark cycle (lights on at 0700 h). Wood shavings were provided for bedding. A nonpurified diet (Rodent Laboratory Chow 5001; Purina Mills, Inc, St. Louis, MO) and water were provided. Litter size was reduced to 6 pups per litter at 2-4 day of age by killing male pups. Only litters with at least three and usually four or five female pups were used. All the female mice in a litter were killed, and the abdominal fat pads were collected and compared. A preobese ob/ob and lean littermate pair was selected on the following basis: the abdominal fat pad weight of the preobese ob/ob mouse had to exceed the fat pad weight of the lean littermate mouse by at least 100% and body weights of the two mice had to be similar. No more than one pair of mice was selected from a single litter, and some litters failed to produce an ob/ob and lean mouse pair. 2. Materials Collagenase (type v, lot 100E6851), bovine serum albumin (fraction v, radioimmunoassay grade), acetylcholine chloride (ACh), cholecystokinin (CCR-8S ; fragment 26-33 amide, sulfated on the tyrosine residue) , glucose-dependent insulinotropic polypeptide (GIP), and Hoechst H 33258 were 47 from Sigma Chemical Co. (St. Louis, MO); rabbit anti-guinea pig IgG was from E.Y Labs (San Mateo, CA); anti-porcine insulin guinea pig serum was from Linco Research Inc. (St. Louis, MO) ; rat insulin standard was from Novo Biolabs (Danbury, CT); Bio- Gel (P-2 Gel, 45-90 mm) was from Bio-Rad (Richmond, CA). Krebs-Ringer bicarbonate buffer (pH 7.4) for isolation of islets and islet incubation was freshly oxygenated. 3. Experimental Design Experiment 1 - Responsiveness of islets to glucose. Responsiveness of pancreatic islets to glucose was measured by incubating islets in 20 mM glucose. Islets were first perifused with 1.7 mM glucose during a 30 min adaptation period. A second 30 min perifusion with 1.7 mM glucose was conducted to establish basal rates of insulin secretion. After this 30 min basal period, the perfusate was switched to 20 mM glucose for 60 min. Samples were collected at 5-min intervals. Experiment 2 - Sensitivity of islets to glucose. The lowest concentrations of glucose that stimulated insulin secretion above basal secretion (i.e. the glucose threshold) was measured by employing a glucose gradient. Effects of ACh, CCR-8s and GIP on the glucose thresholds were 48 also assessed. During the 30 min preincubation and the first 15 min of the basal perifusion periods, islets were exposed to 0.5 mM glucose. For the second 15 min of the basal period islets were exposed to 0.5 mM glucose 1 ACh (10 pH). 1 CCR-8S (1 in!) or 1 GIP (1 uM). Samples were collected at a 5-min intervals. Then a linear glucose gradient was initiated, starting with 0.5 mM glucose and increasing to 13 mM glucose over 70 min (average slope, 0.21 mM glucose per min ). ACh (110 Mt), CCR- 83 (11 in!) or GIP (11 in!) was present throughout the 70-min period. Samples were collected at 2-min intervals. Experiment 3 - Sensitivity of islets to ACh-induced and CCR- 8S-induced insulin secretion. Sensitivities of pancreatic islets to ACh and CCR-8S were quantified by exposing perifused islets to linear ACh or CCR- 8S gradients. Islets were first perifused with 0.5 mM glucose for a 30-min preincubation period; this was followed by a 15 min basal perifusion period in 0.5 mM glucose. Islets were then switched to 10 mM glucose for 30-min. Samples were collected at 5-min intervals. Linear ACh and CCR-8S gradients from 0 to 100 nM ACh or from 0 to 50 nM CCR-8S and in the presence of 10 mM glucose developed over 60 min. During this 49 60 min period" samples were collected at 2-min intervals. Experiment 4 - Synergistic effects of ACh and GIP on-glucose- induced insulin secretion. Islets were again perifused with 0.5 mM glucose for a 30- min preincubation period followed.by a 15-min basal perifusion in 0.5 mM glucose. Islets were then stimulated with 10 mM glucose for a 30-min period. .ACh (10 uM) alone or GIP (1 uM) alone was added in the continued.presence of 10 mM glucose for a second 30-min perifusion period. In the third 30 min stimulatory period islets were exposed to the combination of ACh plus GIP. Samples were collected at 5-min intervals. 4. Methods Islet Preparation. Islets were isolated by the method of Lacy and Rostianovsky 1967 as modified by Tassava et. al. 1992. Mice were killed by cervical dislocation. Each pancreas was inflated in situ via the common bile duct, or via direct injection into the pancreas, with 2 mL of Rrebs Ringer bicarbonate buffer (37°C) containing 1.0 mg collagenase/mL and 0.5 mM glucose. Each pancreas was then quickly removed and placed in a small glass tube containing an additional 0.25 mL of a 10 mg collagenase/mL solution. Tubes were gently shaken by hand in a water bath for 1-2 min. Then they were briskly 50 shaken several times to loosen islets from surrounding connective tissue. To stOp the digestion, 10 mL of ice-cold buffer containing 0.5 mM glucose was added. Islets were then washed several time to remove digested acinar tissue and collagenase. After digestion, isolated islets were selected with the aid of a pipetman under a stereoscopic microscope, and islet diameter was measured. Islets (20 islets/perifusion chamber) that secreted more than 2 fmole insulin . islet'1 . min") under basal glucose conditions (0.5 or 1.7 mM glucose) were considered damaged by the collagenase digestion; these islets responded poorly to elevated glucose. Approximately 10 % of the islets preparations were excluded on this basis. When data from one mouse within a littermate pair were excluded, data from the corresponding littermate were also excluded. Islet Perifusion. The perifusion chamber was a shortened 3-mL plastic syringe with nylon mesh in the bottom of the syringe barrel. Twenty islets were placed between two biogels (0.2 mL) on the mesh, and the syringe plunger, with an 18- gauge needle inlet for buffer entry, was then inserted. The Rrebs-Ringer bicarbonate perifusate (maintained at 37°C) was pumped through the chamber at a flow rate of 0.4 mL/min. 51 An islet preparation from a mouse was only used in a single experiment. ELISA (Enzyme-Linked Immunosorbent Assay) insulin assay. Insulin secreted by islets was assayed using a modified technique by Rekow et al. (1988) . Microtiter plates with 96 round-bottomed wells were coated by a sandwich principle. First, the plates were coated with the rabbit anti-guinea pig antibody. This antibody was used in a dilution of 1:1000, 150 ul and incubated at 20° for 12 hrs. Each well was then washed three times with washing buffer and then incubated with the 1:100 diluted anti-insulin antibody, 120 ul and the plates were incubated at 4°C for two days. The plates were then washed three times. Standards were prepared from 0.125 ng/ml to 10 ng/ml. The standards or the test samples, 120 ul, were removed by repeated washing. The plates were incubated at 37° C, for 50 min. Insulin peroxidase conjugate (100 ul) was added to each well. Plates were again incubated at 37° C for 40 min. Plates then washed three times. Next, ABT solution (100 ul) were added and then incubated about 1 hour to develop reasonable color change at room temperature. Finally, stopping solution (100 ul) was added. Then the optical density (OD) was measured. A graph was made for insulin standard curve (0.125- 52 10 ng/ml) and samples were calculated from insulin standard curve. Islet insulin extraction. Each islet was sonicated for 10 s and incubated overnight at 4°C in 0.5 ml acid-ethanol (7.5 ml 12 N HCl + 492.5 ml 75% ethanol). All samples were then diluted in Krebs-Ringer bicarbonate buffer and stored at -20°C for subsequent insulin determination. 5. Sample and Statistical Analysis Insulin in perifused buffer, islets, and plasma was determined using an Enzyme-linked immunosorbent assay method (Rekow et al. 1988) . The intra-assay coefficient of variation averaged 5%, and ob/ob and lean littermate samples were always assayed at the same time. Glucose was determined with a YSI 2300 STAT Glucose Analyzer II. Stock solutions of ACh and CCR-83 for development of the respective perifusate gradients contained a known ratio of ACh or CCR-8S to 3H30. Appearance of ’H,O in the perifusate was used to calculate ACh and CCR-8S concentrations in the perifusate gradients. DNA content in the islets was measured by DNA assay with Hoechst H 33258 fluorescent compound (Labarca and Paigen 1980) . The threshold for glucose-induced insulin secretion was determined as described by Brelje and Sorenson 1988. The first 53 incidence of five consecutive points with at least four points with rates of insulin release faster than the basal range was determined. The glucose concentration of the first of these five points was taken as the minimal detectable glucose threshold. The same general approach was employed to estimate the minimal thresholds for ACh- and CCR-8S-induced insulin secretion. Data were analyzed by the Student paired t-test (ob/ob and lean mice within litter), except for effects of ACh, CCR-8S or GIP treatments on glucose thresholds as well as insulin secretion in response to ACh and GIP combinations which were assessed by TWO-WAY ANOVA in conjunction with the Tukey's test. Differences with P<0.05 were considered statistically significant. D. Resul ts Although body weights of ob/ob mice were only 13% greater than their lean littermates at 2 weeks of age and the ob/ob mice were not yet visually obese, they had a nearly 3-fold increase in abdominal fat content (Table 1). The 2-week-old ob/ob mice were slightly hyperinsulinemic (+20 %) and hypoglycemic (~12%) when compared with lean littermates. 54 Pancreatic islet size, islet DNA content, and islet insulin content were similar in ob/ob and lean littermate mice (Table 1). Basal rates of insulin secretion from islets perifused in 1.7 mM glucose were low and similar in ob/ob and lean littermate mice (Fig 2) (average of 0.64 1 0.01 fmole insulin release . islet" . min") in ob/ob mice versus 0.59 1 0.01 in lean mice, p>0.05) . Within minutes after changing the perifusion solution from 1.7 to 20 mM glucose, insulin secretion approximately doubled, and the elevated rates of secretion were maintained for the 60 min period (Fig 2). Perifused islets from ob/ob mice did not secrete any more insulin in response to 20 mM glucose than islets from lean mice (average of 1.34 1 0.04 fmole insulin release . islet" . min") in ob/ob mice versus 1.33 1 0.02 in lean mice), indicating that islet responsiveness to 20 mM glucose was not altered in ob/ob mice at 2 weeks of age. To determine whether islets from 2-week-old ob/ob mice hypersecrete insulin in response to lower physiological glucose concentrations, islets were exposed to a linear glucose gradient ranging from 1 to 13 mM glucose (Fig 3, upper panel). The lowest glucose concentration that induced insulin secretion above basal was 55 Table 1. Body weight, fat pad, plasma insulin and glucose, pancreatic islet diameter, islet DNA content and islet insulin content in 2-wk-old ob/ob and lean mice‘. mice IteL QhLQb lean Body weight (7)-g 7.210.07' 6.410.05 Fat pad (7)-mg 17.010.5‘ 6.610.3 Plasma insulin (10)-nM 0.1210.003‘ 0.1010.001 Plasma glucose (10)-mM 7.210.07‘ 8.210.l Islet diameter2 (8)-mm 0.1410.002 0.1410.003 Islet DNA2 (5)-ng/islet 28.412.5 27.213.2 Islet insulin2 (7)- 7.910.5 8.210.4 pmole/islet 1Values are means 1 SEM ; numbers of animals are indicated in () following each item. Data were analyzed by student paired t-test with * indicating significant differences (P<0.05) . ’Values were obtained by measuring 20, 18, and 10 islets from each mouse for determination of islet diameter, DNA, and insulin, respectively. These averaged values were used to obtain group means . 56 20mM Glucose ...... son ' LEAN 1.7 mM Glucose <——>— < ........... INSULIN SECRETION (nonceounnonnoncnoc (is '1'5'2'5 5515165955595 TIME-MIN Figure 2. Insulin secretion from pancreatic islets perifused in 1.7 mM glucose for 30 min and 20 mM glucose for 60 min. Samples were collected at 5 min intervals. Data points represent means + SEM for 6 mice. Average rates of insulin secretion in the basal state (1.7 mM glucose) and in the stimulated state (20 mM glucose) were calculated and presented in the text. Phenotype did not influence insulin secretion as determined by the student paired t-test (p>0.05) . INSULIN SECRETION - fmoie - islet'1 - min"I 0,5 Glucose Gradient mM 0.5 mM to 13 mM 2.Glucose < ' <—> 1. T = 361 0'3 oooooooo: LEAN T = 8.3‘ 0 01505 310515 316315.515 0" 0.5 Glucose Gradient mM 0.5 mM to 13 mM 3Glucose “‘7 > 77-:- 10 uM ACh 5 OB 2 E E 545“”; E T = 5.2 6555 1‘ 3 I LEAN W __v __ 5 IT = 6.4 50:5 0.5 310 5:5 8:0 16.5 13.0 - 0.5 Glucose Gradient mM 0.5 mM to 13 mM 1+ ' 0.5 0.5 3.0 5.5 s..o10513.0 0.5 Glucose Gradient li 0.5 mM to 13 mM . Glucose: 4 Ar ' :<' ‘l M GIP : T = 7.9 08 . WWI LEAN 5 IT = 8.0 v'vvvv vvv'vvvv'vvvv'v'vv‘vvvv'vvvv 0.5 0.5 3.0 5.5 3.0105 13.0 GLUCOSE - mM 57 Figure 3. Threshold for glucose- induced insulin secretion. Islets were perifused in 0.5 mM glucose for 15 min, then with 0.5 mM glucose 1 ACh (10 uM), 1 CCK—88 (1 uM) or 1 GIP (1 pill) for a second 15 min period before initiating a glucose gradient (0.5 to 13 mM glucose over a 70 min period). Samples were collected at 2 min intervals. There were 7 mice per group. For ease of comparisons, the upper panel presents the pooled glucose threshold values for all islets not exposed to ACh, CCR or GIP. The upper panel thus represents means 1 SEM for 21 mice. The corresponding control glucose threshold values for the ACh,CCK or GIP groups are presented in the text. "T" within each panel indicates the average minimal detectable threshold for glucose- induced insulin secretion (mM glucose). The rate of increase in insulin secretion from the slopes of the line (threshold to 13 mM glucose) for each group of mice was analyzed by linear regression. Data were statistically analyzed by TWO- WAY ANOVA (phenotype x treatment) . Significant effects of phenotype, treatment and phenotype x treatment interaction were evident for the glucose threshold for ACh and CCK, and for the rates of increases in insulin. secretion beyond the glucose threshold concentration for ACh. 58 termed the glucose threshold. Basal insulin secretion from ob/ob and lean mouse islets in 0.5 mM glucose was low and identical (average of 0.48 1 0.02 fmole insulin release . islet" . min") in ob/ob mice versus 0.49 1 0.01 in lean mice). The threshold for glucose- induced insulin secretion in ob/ob mice averaged 8.1 1 0.1 mM glucose, and was similar to the threshold in lean mice (8.3 1 0.1 mM glucose, P>0.05, n-21, Fig 3, upper panel). Insulin secretion increased linearly once the glucose threshold had been reached. This rate of increase in insulin secretion (from the slopes of the line, threshold to 13 mM glucose) for each group of mice was calculated. These rates of change were unaffected by phenotype; values averaged 0.13 1 0.01 and 0.13 1 0.01 fmole insulin secretion . islet" . min" per each mM glucose increase above the glucose threshold in ob/ob and lean mice, respectively (Fig 3, upper panel). The ability of Ach to modify the threshold for glucose-induced insulin secretion was examined. Basal rates of insulin secretion from islets of both phenotypes with ACh present (0.59 1 0.04 fmole . islet" . min") in ob/ob mice versus 0.56 1 0.04 in lean mice) was slightly greater than their respective control groups (0.48 1 0.02 fmole . islet" . min") in ob/ob control group versus 59 (0.46 1 0.01) in lean control group) (Fig 3, second panel). ACh treatment lowered the glucose threshold in both phenotypes, but with greater responses in ob/ob mice than in lean mice. The average glucose threshold values were 5.2 1 0.07, and 8.0 1 0.01 mm glucose in islets fron ob/ob mice + ACh and from control ob/ob islets, respectively, and 6.4 1 0.07land 8.2 1 0.04 mm glucose in islets from lean mice + ACh and in control lean islets, respectively (p<0.05, Pig 3, second panel). Significant effects of phenotype, ACh and phenotype x ACh interaction were evident for the glucose threshold and for rates of increases in insulin secretion beyond the glucose threshold. ACh not only lowered the threshold for glucose-induced insulin secretion to a greater extent in islets from ob/ob mice than in islets from lean mice, but also approximately doubled the rate of increase (slope of the line) in insulin secretion per each m)! increase in glucose from the point of the threshold to 13 mm glucose compared to values in lean mice (Fig 3, second panel). In the presence of ACh, rates of insulin secretion averaged 0.23 1 0.007 and 0.13 1 0.005 fmole insulin secreted . islet" . min") per each mM glucose increase in islets from ob/ob and lean mice, respectively (P<0.05, Pig 6O 3, second panel) . Addition of CCK-8S to the perfusate did not affect insulin secretion in 0.5 mm glucose, but lowered the glucose thresholds with greater responsiveness in ob/ob mice than in lean mice (Fig 3, third panel). The average glucose thresholds were 4.2 1 0.12, and 8.2 1 0.2 mM glucose in islets from ob/ob mice 4» CCK-88 and in control ob/ob islets, respectively, and 5.9 1 0.2 and 8.4 1 0.25 ml! glucose in islets from lean mice + CCK-8S and control lean islets, respectively (P<0.05, Fig 3, third panel). Significant effects of phenotype, CCK and phenotype x CCK interaction were evident for the glucose threshold. However, the rates of increase in insulin secretion above the threshold were unaffected by CCK-BS addition. Average rates of insulin secretion with CCK-83 present were 0.12 1 0.02 and 0.10 1 0.02 fmole insulin secreted . islet" . min" per each mM glucose increase in islets from ob/ob and lean mice, respectively (p>0.05, Fig 3, third panel). GIP treatment did not affect glucose thresholds or rates of increase in insulin secretion at glucose concentrations above the threshold in either ob/ob or lean mice (Fig 3, fourth panel) . High concentrations of ACh and CCK-83 caused greater 61 changes in insulin secretion from islets of ob/ob mice than from islets of lean mice (Fig 3, second and third panels). To determine the sensitivity of these islets to ACh and CCK, linear ACh and CCK gradients was employed. Since ACh and CCK- 83 function by potentiating glucose-induced insulin secretion, a glucose concentration within the physiological range that stimulated insulin secretion was used (i.e. 10 mm glucose) (Fig 4). Basal insulin secretion rates in 0.5 mm glucose were similar in islets from ob/ob and lean mice (0.53 1 0.02 fmole . islet" - min") in ob/ob mice versus 0.53 1 0.02 in lean mice) (combined data from upper and lower panels in Fig 4). Increasing glucose to 10 mm increased insulin secretion similarly in both phenotypes (0.59 1 0.01 fmole . islet" . min‘ 1) in ob/ob mice versus 0.57 1 0.01 in lean mice), providing further evidence that glucose-induced insulin secretion is not altered in these young ob/ob mice. The lowest ACh concentration that induced insulin secretion from islets perifused in 10 mm glucose was termed the ACh threshold. In islets from ob/ob mice the ACh threshold for stimulation of glucose-induced insulin secretion averaged 21 1 2 nM ACh, and was much lower than in islets from 62 lean mice where the stimulation threshold averaged 67 1 3 nM ACh (P<0.05, Fig 4, upper panel). The sensitivity of ob/ob islets to CCK-BS was also examined via a linear CCK-8S gradient from 0 to 50 nM CCK-88. A rapid-onset, transient insulin secretion response was observed in both phenotypes when the CCK-BS gradient was initiated (Fig 4, lower panel). A sustained increase in insulin secretion occurred subsequent to this transient phase; CCK-8S thresholds were determined for this second phase of insulin secretion. The first incidence of five consecutive points with at least four points with rates of insulin secretion greater than the basal 10 mm glucose range was determined, and the CCK-88 concentration corresponding to the first of these five points taken as the minimal detectable threshold. The CCK-8S threshold in ob/ob islets averaged 13 1 0.9 nM CCK-BS and was lower than in lean islets where the threshold averaged 27 1 0.8 n1! CCK-BS (P<0.05, Fig 4, lower panel) . Finally, synergistic effects of ACh and GIP in potentiating insulin secretion in the presence of 10 mM glucose were examined (Fig 5) . ACh (10 all) alone evoked a significant insulin secretory response with a 2-fold greater INSULIN SECRETION - fmole ~ lslet‘1 ~mln'1 63 10mMGMmme 1.2‘ 0.8 ‘ 0.5 mM Glucose 1: ll ACh Gradient' < 0 to 100 nM f LEAN 0.4* T=57 0 “When"- .................... 0 0 25 50 75 100 ACh-nM :4 10 mM Glucose o :f CCK Gradient . 8a 0t050nM 0 I < > 3 I 1.6- ('5 1 2 l 1']. E . ca .4 .. a l "a . J O I 1.2- : 3‘ : {V T=13 TT-J. , - ' "l l ." - : ' l, 1 4- 0.8" : v... n -.- ‘03" * I..'*°'.o-7o‘ hi‘t LEAN : I - I I I 0 -----! ............................... 0 0 25 50 Figure 4. Threshold for ACh- and. CCK-induced. insulin secretion. Islets were perifused in 0.5 mM glucose for 15 min and 10 mM glucose for 30 min. Then the .ACh- or‘ CCK- gradients (from 0 to 100 nM ACh or from 0 to 50 nM CCK- 88) in the presence of 10 mM glucose were developed over a 60 min period. Samples were collected at 2 min intervals. Points represent means 1 SEM for 6 mice. "T" within each panel indicates the average minimal detectable threshold for .ACh- or CCK-induced insulin secretion. Data were statistically analyzed by student paired t-test. Significant effects of phenotype was evident for both ACh (upper panel) and CCK (lower panel) thresholds. INSULIN SECRETION - fmole - islet"l - min'1 64 0.5 "WI" 10 mM Glucose > Glucose A 10 11M AChAV “ ’2 1pm GIP l<——> 8.. 6.. 4.1 2.. o IIIT‘IIIIIIIIIIIIIIIII 15 45 75 105 '35 "WP 10 mM Glucose V Glucose 1 11M GIP t 1 10pm ACh ~<——> 8.. 6.. 4. . 2] , 0 WE - .. 15 45 75 105 TIME - MIN Figure 5. Synergistic effects of ACh and GIP on glucose-induced insulin secretion. Islets were perifused in 0.5 mM glucose for 15 min and 10 mM glucose for 30 min. ACh (10 ”M; upper panel) or GIP (1 11M; lower panel) alone was then added in the continued presence of 10 mM glucose for 30 min. Finally, islets were exposed to both .ACh and. GIP. Samples were collected at 5 min intervals. Points represent means 1 SEM for 6 mice. Significant effects of phenotype were observed with ACh alone, and when ACh and GIP were present together. 65 insulin secretion response from ob/ob islets (2.42 1 0.25 fmole . islet" . min") than from lean islets (1.2 1 0.16) (P<0.05, Fig 5, upper panel). GIP alone only evoked a small insulin secretory response, and both phenotypes responded similarly (P>0.05, Fig 5, lower panel). Insulin secretion from ob/ob and lean islets was amplified when ACh and GIP were present together in the perifusate, with effects in islets from ob/ob mice exceeding those in islets from lean mice. In ob/ob islets, insulin release rates averaged 6.9 1 0.6 fmole . islet" . min") when ACh and GIP were present (average of upper and lower panels in Fig 5), whereas in lean islets insulin release rates averaged 4.3 1 0.4 fmole . islet" - min") (P<0.05) . 5 . Discussion The present results demonstrate that pancreatic islets from preobese, 2-week-old, ob/ob mice exhibit normal responsiveness and sensitivity to glucose-induced insulin secretion, but show enhanced responsiveness and sensitivity to ACh and CCK as well as enhanced ACh plus GIP potentiation of insulin secretion compared to lean littermates. At 2 weeks of age, ob/ob and lean mice consume similar amounts of milk 66 and therefore of nutrient insulin secretagogues (Lin et al. 1979). The 2-week-old ob/ob mice also have slightly lower plasma glucose concentrations than lean littermates. consistent with an earlier report (Dubuc 1981) . Therefore, the initial onset of hyperinsulinemia in these 2-week-old ob/ob mice cannot be explained by primary alterations in glucose-induced insulin secretion. This conclusion is consistent with observations in isolated islets from adrenalectomized ob/ob mice (Mistry et al. 1995) and in preobese fa/fa rats (Atef et al. 1991) where glucose alone also fails to cause hypersecretion of insulin. The responsiveness and sensitivity of islets from ob/ob mice to glucose-induced insulin secretion does change markedly after weaning as these animals begin to develop gross obesity (Chen et a1. 1993), suggesting that these changes in insulin secretion are secondary to the onset of obesity. Two factors contribute to these substantial differences in responsiveness and sensitivity of islets from 2-week-old ob/ob mice to glucose-induced insulin secretion versus values in adult ob/ob mice. First, there is a well established developmental progression in insulin secretion as animals transition from the neonatal period, when high-fat and low-carbohydrate milk 67 is consumed to adulthood when standard practice is to feed a low-fat, high-carbohydrate diet (Lin et al. 1979) . The 2-week- old lean mice secreted only «60% as much insulin in response to 20 ml! glucose as observed in 8-week-old lean mice (1.33 1 0.02 vs 2.11 1 0.07 fmole insulin release . islet" . min") (Chen et al. 1993 and present study). The threshold for glucose-induced insulin secretion was also much higher in 2- week-old lean mice than in 8-week-old lean mice (8.3 1 0.1 vs 4.8 1 0.1 mm glucose) (Chen et al. 1993 and present study). These developmental and diet dependent changes in lean mice are not sufficient to totally explain the changes in glucose- induced insulin secretion observed in ob/ob mice as they develop. A second factor, probably associated with development of obesity, causes more dramatic changes in glucose-induced insulin secretion in ob/ob mice than predicted from the normal age and diet dependent transitional changes that occur in lean littermates. Islets from adult ob/ob mice secrete ~10095 more insulin in response to 20 mm glucose and have a ~609s lower glucose threshold than adult. lean mice, even when comparing islets of similar size (Chen et al. 1993). But, because these developmental associated changes in islet responsiveness and sensitivity to glucose in ob/ob mice are 68 almost completely prevented by adrenalectomy, they do not likely represent inherent defects in the islets of ob/ob mice. Enhanced potentiation of glucose-induced insulin secretion by ACh is one of the earliest detectable changes in islets of ob/ob mice, consistent with findings in preobese fa/fa rats (Atef et al. 1991) . This enhanced insulin secretion response to ACh is even present in islets from adrenalectomized, normoinsulinemic ob/ob mice, suggesting that there may be inherent defects in this signal transduction pathway in islets from ob/ob mice. In addition to the enhanced responsiveness of islets from 2-week-old ob/ob mice to ACh, these islets also exhibited enhanced sensitivity to ACh. These early onset alterations in ACh sensitivity in ob/ob islets are consistent with increases in affinity and/or number of ACh receptors on the B-cell membrane. But our findings with CCK suggest that post-receptor events are more likely to be involved. CCK, which shares a common post-receptor and post-G protein-coupled signalling system with ACh (Prentki and Matshinsky 1987, Schnefel et al. 1988), also potentiates glucose-induced insulin secretion with enhanced responsiveness and sensitivity in islets from 2-wk old ob/ob mice. Therefore, the ACh and CCK 69 post-receptor effector system responsible for coupling signal transduction and modulation of insulin secretion within the B- cell may be responsible for these alterations in sensitivity. There are a host of possible candidate sites in the ACh and CCK signal transduction pathway where potentiation of glucose-induced insulin secretion might be enhanced in islets from ob/ob mice. For example, phenotype-specific alterations at the level of phospholipase C enzyme activity (Prentki and Matshinsky 1987) could cause increased diacylglycerol and inositol triphosphate concentrations in islets from ob/ob mice. Diacylglycerol via activation of protein kinase C and inositol triphosphate via increased Ca‘2 mobilization would be expected to facilitate insulin secretion (Zawalich and Rasmussen 1990) . More distal sites in the signal transduction pathway are also potential sources of the enhanced sensitivity of islets from ob/ob mice to ACh aand CCK. For examle, modulation of voltage-dependent calcium channels via effectors including arachidonic acid may lower the threshold rise in membrane potential required for activation of voltage- dependent calcitnn channels in the B-cell membrane (Ramanadham and Turk 1994) more in ob/ob mice than in lean mice, thereby 70 facilitating amplification of the response to glucose plus ACh/CCK secretagogues. Experiments are needed to focus on these and other possibilities for the enhanced sensitivity of islets from ob/ob mice to ACh and CCK. GIP is a less potent stimulator of glucose-induced insulin secretion than ACh or CCK (Zawalich 1988, Brelje and Sorenson 1988). In the present study GIP actually failed to significantly lower the threshold for glucose-induced insulin secretion (Fig 3, fourth panel), or to elevate insulin secretion in the presence of 10 mM glucose. Possibly the GIP signal transduction pathway is not yet fully functional in islets from 2-week-old mice. Alternatively a higher glucose concentration in the perifusate (i.e. > 10-13 ml!) may be necessary to elicit glucose-induced insulin secretion in response to GIP in islets from young mice. But clearly there is not an absolute deficit in GIP responsiveness in these islets because GIP functioned synergistically with ACh to potentiate glucose-induced insulin secretion, as has also been observed in islets from adult rats (Mccullough et a1. 1985 ) . GIP potentiated insulin secretion in the presence of ACh to a greater extent in islets from ob/ob mice than in islets from lean mice. The mechanism of this action in ob/ob 71 mice is unknown. One candidate would be modulation by GIP of voltage-dependent calcium channels to increase extracellular Ca" influx (Lu et al. 1993) and thereby enhance insulin release. The interplay between glucose and neurohormones in potentiation of insulin release is clearly altered in islets from ob/ob mice. In the early onset of hyperinsulinemia in ob/ob mice this abnormality is expressed via enhanced potentiation of glucose-induced insulin secretion by several neurohormones. Heightened sensitivity of a component of the intracellular signal transduction pathway common to ACh and CCK appears central to hypersecretion of insulin by islets from young ob/ob mice. Crosstalk between the ACh and GIP signal transduction. pathways further' potentiates insulin secretion from islets of these ob/ob mice. Identification of specific loci within these signalling pathways responsible for the enhanced insulin secretion from islets of ob/ob mice should help in understanding the genetic basis for hyperinsulinemia in these animals. CHAPTER IV. PERSISTENTLY ENHANCED SENSITIVITY OF CULTURED PANCREATIC ISLETS FROM 08/ OB NICE TO PROTEIN KINASE C- STIMULATED INSULIN SECRETION A. Abstract Islets from 2-wk-old ob/ob and lean littermate mice were cultured for 4 to 12 days and then perifused with various insulin secretagogues to identify early-onset abnormalities in the regulation of insulin secretion in ob/ob mice. Islets from ob/ob and lean mice increased insulin secretion similarly in response to glucose (10 or 20 ml!) whereas responsiveness to glucose plus acetylcholine (ACh, 10 ul!) was greater in islets from ob/ob mice than lean mice. ACh potentiates glucose- induced insulin secretion through the phospholipase C -protein kinase C (PKC) signal transduction pathway. This phenotype- specific effect of ACh was mimicked by PMA (100 nl!), a PRC agonist. PKC enhances insulin release by activating voltage- dependent Ca channels (VDCCs) as well as by post-VDCCs mechanisms that directly enhance the exocytotic machinery. 72 73 Islets from ob/ob and lean mice perifused in glucose plus the L-type, VDCC agonist BAY R8644 (2,10,20 ul!) increased insulin secretion similarly, suggesting normal functioning of directly activated L-type, VDCCs in islets of ob/ob mice. After activation of these VDCCs by BAY R8644 (10 ul!), addition of ACh or PMA now stimulated insulin secretion equally from islets of ob/ob and lean mice. Protein kinase A (PRA) activation by either forskolin or IBM! dramatically and equally potentiated glucose-induced insulin secretion from islets of ob/ob and lean mice. We propose that the mechanism whereby PRC activates L-type, VDCCs is altered in islets from ob/ob mice and that this alteration persists even when islets are cultured for up to 12 days. B. Introduction Hyperinsulinemia is an early-onset characteristic of ob/ob mice and fa/fa rats although pancreatic islets obtained from these genetically obese rodents at 5-14 days of age do not yet hypersecrete insulin in response to glucose (Chen and Romsos 1995, Atef et al. 1991). However, the stimulatory effects of acetylcholine (ACh) on glucose-induced insulin secretion are already much greater in islets of these young 74 obese animals than in islets of lean littermates. Possibly this.ACh.stimulatory system.contributes to the development of hyperinsulinemia in these animals. CCR shares a common post- receptor signal transduction pathway with ACh in potentiation of glucose-induced insulin secretion (Schnefel et al. 1988). In 2-week-old ob/ob mice, islet responsiveness to CCR is also altered in a manner similar to ACh (Chen and Romsos 1995) . This implies that hypersecretion of insulin in these young obese rodents is associated with the ACh- and CCR-stimulated post-receptor PLC signaling pathway. PLC activation generates inositol IP3 which mobilizes stored Ca" from the endoplasmic reticulum and DAG which activates PRC. Increase in mObilization of intracellular Ca“2 and activation of PRC both enhance glucose-induced insulin secretion (Prentki and Matshinsky 1987, Zawalich and Rasmussen 1990) and thus are candidates to explain the greater increase in ACh-induced insulin secretion from islets of young ob/ob mice than from islets of lean mice. A second signal transduction pathway involving activation of adenylate cyclase by GIP also enhances glucose-induced insulin secretion (Zawalich and Rasmussen 1990). Islet responsiveness to GIP is difficult to detect in either 2-wk- 75 old ob/ob or lean mice (Chen and Romsos 1995) . Possibly the GIP signal transduction pathway is not yet fully functional in islets from these young mice. But clearly there is not an absolute abnormality in GIP responsiveness in these islets because GIP functions synergistically with ACh to potentiate glucose-induced insulin secretion more from islets of 2-wk-old ob/ob mice than from islets of their lean littermates (Chen and Romsos 1995) . Thus, it remains possible that the adenylate cyclase- PRA signal transduction pathway in islets of young ob/ob mice is altered. The recent discovery that the primary genetic defect in ob/ob mice is expressed only in white adipose tissue (Zhang et a1. 1994) raises the possibility that residual effects of the in vivo environment explain the enhanced ACh- induced insulin secretion observed in freshly-isolated islets from young ob/ob mice (Chen and Romsos 1995) . The first aim of the present study was to examine this possibility by culturing islets from 2-wk-old ob/ob mice for up to 12 days before they were perifused with ACh. Contrary to the expectation that maintenance of the islets from ob/ob and lean mice in a controlled in vitro environment would equalize their responsivenss to ACh, the enhanced insulin secretion response 76 to ACh persisted in islets from ob/ob mice maintained in culture. Additional aims of the present study were thus designed to further examine the metabolic basis for the enhanced ACh stimulation of insulin secretion from cultured islets of ob/ob mice. Effects of PRC and L-type, VDCCs activation on insulin secretion were examined as were effects of PRA activation. C. Materials and Methods 1. Animal and diet Fanale ob/ob mice and lean littermates (ob/+- or +/+) from our breeding colony (C57BL/6J-ob/+) were used at 2 wk of age. Mice were housed at 24°C and with a 12 h light, 12-h dark cycle (lights on at 0700 h). Wood shavings were provided for bedding. A nonpurified diet (TERIJID Laboratory Diet 7005; Harlan, Inc, Bartonville, IL) and water were provided. Litter size was reduced to 6 pups per litter at 2-4 day of age by removing male pups. Only litters with at least three and usually four or five female pups were used. All the female mice in a litter were killed at 2 wk of age, and the abdominal fat pads were collected and compared. An ob/ob and lean littermate pair was selected on the following basis: the 77 abdominal fat pad weight of the ob/ob mouse had to exceed the fat pad weight of the lean littermate mouse by at least 100% and body weights of the two mice had to be similar. No more than one pair of mice was selected from a single litter, and some litters failed to produce an ob/ob and lean mouse pair. 2. Materials Collagenase (type v, lot 32H6803), bovine serum albumin (fraction v, radioimmunoassay grade), acetylcholine chloride (ACh), phorbol-l2-myristate-13-acetate (PMA), forskolin, isobutylmethylxanthine (IBEX) , RPMI medium and Hoechst H 33258 were from Sigma Chemical Co.(St. Louis, HO); HA! R8644 was from RBI (Nhtick, MA); rabbit anti-guinea pig IgG was from E.Y Labs (San Mateo, CA); anti-porcine insulin guinea pig serum.was from Linco Research Inc. (St. Louis, MO); rat insulin standard was from.Novo Biolabs (Danbury, CT); Bio-Gel (P-2 Gel, 45-90 am) was from Bio-Rad (Richmond, CA); fetal bovine serum. was from Gibco (Grand Island, NY). Petri dishes (Falcon 3002) were from Fisher Co. (Itasca, IL). Rrebs-Ringer bicarbonate buffer (pH 7.4) for isolation and perifusion of islets was freshly oxygenated. l!gCl2 was substituted for CaCl, to prepare calcuimrfree Rrebs-Ringer bicarbonate medimm for one study. 78 3. Experimental design Experiment 1 - Glucose-, ACh- and potassium-induced insulin secretion. Islets were perifused with 0.5 (or 1.7) ml! glucose for 15 min, switched to 10 (or 20) ml! glucose for 30 min, and subsequently exposed to glucose plus 10 ul! ACh or 45 ml! R+ for the last 30 min of the 75 min perifusion. Experiment 2 - Protein kinase C activation of glucose-induced insulin secretion. Islets were perifused in 10 ml! glucose and Pl!A, a PRC agonist (Prentki and Matshinsky 1987, Zawalich and Rasmussen 1990). At 10 min intervals, concentrations of PMA in the perifusate were increased stepwise from 0 to 1000 nl!. Other islets were perifused in 1 ul! mm for 60 min, or in 100 nl! PMA for 30 min and then 100 nl! Pl!A plus 10 ul! ACh for another 30 min. Experiment 3 - L-type, VDCC activation of glucose-induced insulin secretion. Islets were perifused in 10 ml! glucose and BAY R8644, a L-type, VDCC agonist (Prentki and Matshinsky 1987, Zawalich and Rasmussen 1990) . At 10 min intervals, concentrations of BAY R8644 in the perifusate were increased stepwise from 0 to 79 100 uM. Other islets were perifused in 10 uM BAY R8644 for 30 min before addition of 10 ul! ACh or 100 nM PMA to the perifusate, or conversely islets were first perifused in 10 ul! ACh for 30 min before addition of 10 ul! BAY R8644. Additional islets were exposed at 10 min intervals to stepwise increases in Ca” concentrations from 0 to 2.5 ml! in the presence of glucose (10 nl!), BAY R8644(10 ul!) and ACh (10 ul!) . Experiment 4 - Protein kinase A actviation of glucose-induced insulin secretion. After a 20 min perifusion period in 10 ml! glucose and at 10 min intervals, islets were exposed to stepwise increases in forskolin, an adenylate cyclase activator (Malaisse et al. 1984, Liang and Matschinsky 1994), from 0 to 100 nl! or in IBM, phosphodiesterase inhibitor (Liang and Matschinsky 1994), from 0 to 500 uM. Other islets were perifused in 10 ml! glucose and then in 10 ml! glucose plus 0.1 ul! IBM for 60 min. 4. Methods Islet preparation. Islets were isolated by the method of Lacy and Rostianovsky 1967 as modified by Tassava et. al. 1992 . After islets were freed from the pancreatic tissue they were harvested with the aid of a pipetman under a 80 stereoscopic microscope, and placed in petri dishes. Islet culture. Islets from ob/ob and lean mice were cultured in petri dishes (25-30 islets per dish) containing 6 ml medium (RPMI 1640 medium containing 10 ml! glucose, 10,000 units/ml penicillin and 10 mg/ml streptomycin). Islets were maintained undisturbed for 4 days at 37°C in an atmosphere of 59‘ CO2 in humidified air. Fetal bovine serum was not included in the medium.during these first 4 days of culture because it caused islets to firmly attach to the bottom of dish. These attached islets were easily damaged when harvested for perifusion. A preliminary study showed that glucose-induced insulin secretion was similar from islets cultured for 4 days in the presence or absence of 10 % fetal bovine serum (data not shown). Islets cultured for 12 days were maintained between days 4 and 12' in medium containing 5 % fetal bovine serum; medium.was changed at 2-3 day intervals between days 4 and 12. Islet perifusion. Islets were harvested from the culture dishes with a pipetman and perifused as previously described (Chen and Romsos 1995). The perifusion chamber was a shortened 3-mL plastic syringe with nylon mesh in the bottom of the syringe barrel. Twenty islets were placed between two biogels 81 (0.2 mL) on the mesh, and the syringe plunger, with an 18- gauge needle inlet for buffer entry, was then inserted. The Rrebs-Ringer bicarbonate perifusate (maintained at 37%” was pumped through the chamber at a flow rate of 0.4 mL/min. During a 30 min pre-experimental period islets were perifused in medium containing 0.5 mM glucose (Fig 6, panel A,C), 1.7 mM glucose (Fig 6, panel B) or 10 mM glucose for the remain of figures. An islet preparation from a mouse was only used in a single experiment. 5. Sample and statistical Analysis For extraction of insulin, islets were sonicated for 10 s and incubated overnight at 4°C in 0.5 ml acid-ethanol (7.5 ml 12 N HCl + 492.5 ml 75% ethanol). All samples were then diluted in Rrebs-Ringer bicarbonate buffer and stored at -20%: for subsequent insulin determination. Insulin in perifusate buffer and in islet extracts was measured using an enzyme- linked imunosorbent assay method (Rekow et al. 1998). The intra-assay coefficient of variation averaged 6%, and ob/ob and lean littermate samples were always assayed at the same time. DNA content in the islets was measured with an assay utilizing Hoechst H 33258 fluorescent compound (Labarca and Paigen 1980). 82 Data were analyzed by TWO-WAY ANOVA (phenotype and treatment) with Turkey's post-hoc test, or by a Split-plot design (the main effects : phenotype and treatment) with the subplot (time), or by the Student's paired t-test (ob/ob and lean mice within litter), as indicated in figure legends. Differences with P<0.05 were considered statistically significant. D. Results Pancreatic islets of 2-week-old ob/ob and lean littermate mice cultured for 4 or 12 days in medium containing 10 mM glucose exhibited similar sizes, DNA content, and insulin content (Table 2); these values are also similar to those obtained from freshly isolated islets of these mice (Chen and Romsos 1995). Islets cultured for 4 or 12 days and then perifused in 10 or 20 mM glucose increased (P<0.05) insulin secretion above basal rates of secretion (Fig 6). These glucose-induced increases in insulin secretion were unaffected by phenotype (P>0.05) . Glucose metabolism depolarizes islet cells by inhibiting an ATP sensitive R-channel (Prentki and Matshinsky 1987, Zawalich and Rasmussen 1990) . Depolarization of these islets by 45 ml! R‘ further increased glucose-induced 83 Table 2. Diameter, DNA content, and insulin content of cultured pancreatic islets from 2-week-old ob/ob and lean mice Mice W M lean— Islet diameterIlO), mm 4 0.1310.002 0.1310.004 12 0.1310.003 0.1310.004 Islet DNA(6), ng/islet 4 22.411 20.312 12 27.613 28.214 Islet insulin(5), pmole/islet 4 8.110.7 9.111.6 12 11.71242 12.112.3 Values were obtained by measuring 20, 10, and 10 islets from each mouse for determination of islet diameter, DNA, and insulin, respectively. These averaged values were used to obtain group means 1 SEM; the number of animals is indicated in parentheses following each item. No significant differences were found as determined by Student's paired t test (P>0.05). INSULIN SECRETION - fmole - islet'1 - ml '1 84 A ' —<»—os l —6—LEAN 6 I i l I 10 mM.Glucose . 1 0.5 mu : 10 0M ACh 5- Glucose ; I E 4‘ l E I E 3" I : 2 I : . a I I I I C l a I I fi I I I I 01020304050607080 B |¢ 20 mM Glucose 1.7 mM 5 1' 0 "M AC". 4-GMase 5 . a 3" I E ' E 2- I : I I I I I I I I 0 10 20 30 40 50 60 70 so C 10 mM Glucose ‘ .l . : 10uMACII 0.5 ml! H 4- Glucose 'L “WE—“59L. 3'1 g 45mMK+ 1_ I G 15! l I III 01020304050607080 MINUTES Fig. 6. Glucose, ACh and potassium-induced insulin secretion. Islets were cultured for 4 (panels A and B) or 12 (panel C) days in. 10 mM glucose and then were perifused in 0.5 or 1.7 mM glucose for 15 min, switched to 10 or 20 mM glucose for 30 min, and subsequently exposed to 10 uM ACh for the last 30 min of the 75 min perifusion. Data in panel D are from islets cultured for 4 days and then perifused in 10 mM glucose before exposure to 45 mM K‘. Data points represent means 1 SEM for samples collected at 5 min intervals from five mice. Phenotype did not influence insulin secretion (P>0.05) at basal (0.5 or 1.7 mM), 10 mM or' 20 ‘mM glucose. Significant effects of phenotype (P<0.05) on insulin. secretion. 'were observed in islets exposed to 10 uM ACh in the presence either 10 mM glucose (panel A) or 20 mM glucose (panel B) as well as after 4 (panel A) or 12 days (panel C) in culture. Phenotype did not influence (P>0.05) the increase in insulin secretion that occurred in islets exposed to 45 mM K+ (panel D). 85 insulin secretion independent of phenotype (Fig 6, panel D). Aa potentiation of insulin secretion. After addition of ACh to the perifusate, islets from ob/ob mice increased insulin secretion more than did islets of their lean littermates. This greater responsiveness of islets from ob/ob mice to ACh-stimulated insulin secretion occured in islets cultured for 4 days in either 10 ml! glucose (+80 9a greater secretion in ob/ob vs lean, P<0.05, Fig 6, panel A) or 20 ml! glucose (+30 % greater secretion in ob/ob vs lean, P<0.05, Fig 6, panel B) as well as after 12 days culture (+115 is greater secretion in ob/ob vs lean, P<0.05, Fig 6, panel C). Again, these results are comparable to those in freshly isolated islets from young ob/ob and lean mice (Chen and Romsos 1995), suggesting that islets of ob/ob mice exist a persistent abnormality in response to ACh-stimulated insulin secretion. Next, we determined whether direct stimulation of PRC, an enzyme activated via ACh stimulation of the PLC signaling transduction pathway, would also cause greater insulin secretion from islets of ob/ob mice than from islets of lean mice. PRC ac ti va ti on of insulin secretion . PMA at concentrations of 1,10,100 nl! PMA stimulated insulin secretion 86 more from islets of ob/ob mice than islets of lean mice (Fig 7, panel A; P<0.05) . However, at a PMA concentration of 1 ul! islets of both ob/ob and lean mice secreted similar amounts of insulin (Fig 7, panel A). To confirm whether the responsiveness of islets from ob/ob and lean mice were comparable when exposed to 1 uM PMA for more than 10 min, islets were perifused in 1 ul! PMA for one hour (Fig 7, panel B). The rate of insulin secretion increased during the first 30 min to reach a maximal release of ~9 1 2 fmole . islet" . min" in both phenotypes. This rate of release is twice as high as observed in islets of ob/ob mice exposed to a maximal stimulatory concentrations of ACh (10 uM) (Fig 6) . PRC activation with a lower concentration PMA (100 nl!) effectively mimicked the magnitude and phenotype-specificity of 10 ul! ACh on glucose-induced insulin secretion (Fig 7, panels A and C, a Fig 6, panel A). I Islet responsiveness to the addition of ACh to the PMA containing perifusate was also examined. Islets from lean mice (p<0.05), but not from ob/ob mice (P>0.05), secreted more insulin in the presence of PMA and ACh than to PMA alone (Fig 7, panel C). Consequently, in the presence of PMA and ACh rates of insulin secretion from islets of ob/ob and lean mice INSULIN SECRETION - fmole - IsIet'1 - min"1 * 'k 0 1 10 100 500 1000 nM PMA 1 10mMGlucose r :4 1uMPMA » 14- '1” LEAN 12- Io~ a- 5.. 4- 2- o '. . . . I . I 01020304050607080 10 mM Glucose 100 nM PMA _ 10 all ACh C <—. IA 7‘ " . ‘ l -..-..... Cl TIIIIII 010m304050607080 MINUTES 87 Fig 7. PKC potentiation of glucose-induced insulin secretion. Islets were cultured for 4 days in 10 mM glucose and then perifused in 10 mM glucose and PMA, a PKC agonist. Panel A - At 10 min intervals , concentrations of PMA in the perifusate were increased stepwise from 0 to 1000 um. Bars represent means 1 SEM for 5 mice of rates of insulin secretion for each of the 10 min intervals. Significant effects (P<0.05) of phenotype on. insulin secretion. were evident at 1, 10 and 100 nM PMA as determined by student's paired t test. Panel B - Islets were perifused in 1 uM PMA for 60 min. Values represent means 1 SEM for 5 mice of samples collected at 5 min intervals. Phenotype (P>0.05) did not influence rates of insulin secretion. Panel C - Islets were perifused in 100 nM PMA for 30 min; 10 uM ACh was then added to the perifusate. values represent means 1 SEM for 5 mice of samples collected at 5 min intervals. Significant effects (P<0.05) of phenotype and treatment were observed as determined by TWO-WAY ANOVA. Turkey's test indicated significant effects of phenotype in the presence of PMA alone (P<0.05), but not in the presence of PMA plus ACh. 88 was comparable (P>0.05). VDCC activation of insulin secretion. Islets from ob/ob and lean mice increased insulin release shmilarly in response to the increasing BAY R8644 concentrations (Fig 8). Maximal insulin release was attained at BAY R8644 concentrations between 10-50 uM in both phenotypes. A higher concentration of BAY R8644 (100 uM) stimulated the rate of insulin secretion less in both phenotypes than did the lower concentrations. This has also been reported that higher concentrations of BAY R8644 inhibit insulin secretion (Henquin et al. 1985). To confirm.that direct activation of VDCC by BAY R8644 increases insulin secretion equally from islets of ob/ob and lean mice, islets were perifused with 10 uM BAY R8644 for 30 min (Fig 9, panel A and B). Insulin secretion increased 81 t in islets from.ob/ob mice and 72 % in islets from lean mice when 10 uM BAY R8644 was added to the 10 mM glucose perifusate (P>0.05 for phenotype effect, Fig 9 , panel A and B). Addition of 10 uM ACh to the 10 uM BAY R8644 containing perifusate further and equally enhanced insulin secretion from islets of ob/ob and lean mice (P>0.05 for phenotype effect; Fig 9, panel A). The initial insulin secretion response of islets from ob/ob mice to 100 nM PMA, after exposure to 10 uM HAY R8644, was 89 '3; E 5 .7' 03 *5; LEAN 272 4‘ s O E 3- ’5' F: 2‘ 1.1.] a: 8 a; 1‘ Z ...I :3 J (D 0 . .2. o 2 10 20 so 100 1.1M BAY K8644 Fig 8. L-type, VDCC potentiation of glucose-induced insulin secretion. Islets were cultured for 4 days in 10 mM glucose and then perifused in 10 mM glucose and BAY K8644, a L-type, VDCC agonist. At 10 min intervals, concentrations of BAY K8644 in the perifusate were increased stepwise from 0 to 100 uM. Bars represent means 1 SEM for 5 mice of rates of insulin secretion at 10 min intervals. BAY K8644 (P<0.05), but not phenotype, influenced insulin secretion, as determined by TWO-WAY ANOVA. INSULIN SECRETION - mele - islet" - min"1 A -—o—-OB -—+—4£AN 10 mM Glucose 7 - 4 + 6 _ l‘ 10 uM BAY K8644 I ; 10 uM ACh ,<———> 5 - ' : I a 4‘ l I a - ' 5 I . 2« I 1. 5 0 I I I I I I l I I 0 10 z3:m 40 also W)!” A 10 mM Glucose l 10 uM BAY K8644 ~ I 5 100 nM PMA ae—————>~ 7 - I g l I 6 "I ' E s - I 5 I a 4 - I : 3‘ l l 2- l 1- I o I J I I W L l I I 70 10 £1304” anew Nlao 6 (i 10 ml! Glucose + !A 10 uM ACh A 5 - " 3 10 an I ; BAY K8644 4 ~ I . 3‘ l l 2- l I . 1~ 1' E o I l I I T 1 I I I 0 10 33:» «15w a) a) MINUTES 90 Fig. 9. Effects of direct L- type, VDCC activation on ACh and PMA enhanced insulin secretion. Islets were cultured for 4 days in 10 mM glucose and then perifused. Values are means 1 SEM for 5 mice Of samples collected at 5 min intervals. Data were analyzed as a split- plOt design with the main plots (phenotype and treatment) and subplot (time). Panel A - Islets were perifused in 10 uM BAY K8644 for 30 min before addition Of 10 uM ACh to the perifusate. Treatment (P<0.05) , but not phenotype, influenced insulin secretion. Panel B - Identical to panel A, except that 100 nM PMA replaced 10 uM ACh . Treatment (P<0 . 05) , but not phenotype, influenced insulin secretion. Panel C - Islets were exposed to 10 ul! ACh for 30 min before exposure to 10 uM BAY K8644. Significant effects Of phenotype (P<0.05), but not treatment , influenced insul in secretion. 91 more pronounced than observed in islets from lean mice, however, mean rates of insulin secretion during the entire 30 min period from islets of ob/ob and lean mice exposed to BAY R8644 and PMA were comparable (P>0.05, Fig 9, panel B). Next, the effect of BAY R8644 on insulin secretion was investigated in islets where 10 ul! ACh was added to the perifusate before, not after, addition of BAY R8644 (Fig 9, panel C). As expected, ACh increased insulin secretion more from islets of ob/ob 'mice than from islets of lean mice (P<0.05). Subsequent addition of BAY R8644 failed to further enhance insulin secretion from islets of ob/ob mice (P>0.05) , but approximately doubled insulin secretion from islets Of lean mice (P<0.067) . The presence of BAY R8644 thus abolishes the phenotype-specific differences in insulin secretion caused by ACh. In the presence of glucose, BAY R8644 and ACh, the availability of Ca2+ in the perifusate limits insulin secretion from islets. Increasing the Ca“ concentrations stepwise from 0 to 2.5 mM enhanced insulin secretion equally from islets of ob/ob and lean mice (Fig 10), indicating that the islets from Ob/ob and lean mice do not have a different sensitivity to Ca” . 92 'c E 4 CB .5 LEAN 2'5 .7." 3- .32 O E I 2‘ Z 2 I'- E 1- o - NJ (I) z I I I :1 o- D ‘3 o 0.010.05 0.1 0.5 1 2.5 mM CALCIUM Fig 10. Effects Of perifusate Ca2+ concentration on glucose- induced insulin secretion. Islets were cultured 4 days and then perifused in Ca“~ free medium with 10 mM glucose for 30 min. Islets were then exposed at 10 min intervals to stepwise increases in Ca2+ concentrations from 0 tO 2.5 mM in the presence Of glucose (10 mM), BAY K8644(10 uM) and ACh (10 uM). Bars represent means 1 SEM for five mice of rates of insulin secretion for each 10 ‘min. interval. A. significant treatment (P<0.05), but not phenotype, effect was evident as determined by TWO-WAY ANOVA. 93 RKA activation of insulin secretion. Islets from both ob/ob and lean mice responded to increasing forskolin or IBM! concentrations with increased insulin secretion (Fig 11, panels A and B), suggesting an active PKArmediated signaling pathway in.mice at 2 weeks of age. Only at IBM! concentrations of 0.1 and 1 um was insulin secretion in ob/ob mice enhanced when compared to their lean littermates (Fig 11 panel B). This enhanced insulin secretion in islets of ob/ob mice exposed.to 0.1 um IBMX versus values in islets of lean mice was not evident when the islets were perifused with 0.1 um IBEX for 1 h (Fig 11, panel C). INSULIN SECRETION - fmole - isIet'1 - min'1 94 1a ' 16- 14- 12- 10~ 8- 6"! 4.. 2' I I I 0- 0 0.01 0.1 1 10 pM FORSKOLIN A as LEAN 100 O 0.010.1 1 10 100 500 pMIBMX <3 -—o——oe +LHN 10mM Glucose 2-s‘ 4+ 1‘ 0.1uMIBMX > I I 1.. I I I o I F I I I l I 0 1O 20 30 4O 50 60 70 80 MINUTES Fig. 11. Forskolin and IBMX potentiation of glucose- induced insulin secretion. Islets were cultured for 4 days in 10 mM glucose and then perifused in 10 mM glucose. Panel A. and B - After a 20 min peri fus ion period and at 10 min intervals, islets were exposed to stepwise increases in forskolin from O to 100 uM or in IBMX from 0 to 500 uM. Bars represent means 1 SEM for 5 mice of rates of insulin secretion at 10 min intervals . No phenotype differences in insulin secretion were evident for either forskolin (panel A) or IBMX (panel B) treated islets, except at IBMX concentrations of 0.1 and 1 uM where values from islets of ob/ob mice exceeded values from islets of lean mice (P<0.05) as determined by TWO-WAY ANOVA and Student's paired t-test. Panel C - Islets were exposed to 0.1 uM IBMX for 60 min. Data points represent the mean 1 SEM for five 'mice of samples collected at 5 min intervals. Phenotype did not influence insulin secretion (P>0.05). 95 E. Discussion The main findings of the present report are as follows. The enhanced ACh-induced insulin secretion characteristic of islets from ob/ob mice versus islets from lean.mice persists even when islets are cultured for up to 12 days. Activation of PKC with the phorbol ester PMA mimicked this phenotype differential response to ACh. However, prior activation of L- type, VDCCs by BAY K8644 abolished the ability of either ACh or PMA to differentially enhance insulin secretion from islets of ob/ob versus lean mice. ACh via activation of PKC thus appears to differentially regulate L-type, VDCCs in islets of ob/ob versus lean mice to enhance insulin secretion more from islets of ob/ob mice than from.islets of lean mice. Addition of micromolar concentrations of PMA, forskolin or IBMX to the perifusion buffer enhanced insulin secretion equally in islets from ob/ob and lean mice, and to a substantially greater extent than observed when ACh was added (Figs 6,7 and 11). This implies that the overall capacity for insulin secretion is not limiting in islets of lean mice versus ob/ob mace, and that the PKC and PRA signal transduction system each possess considerable potential to enhance glucose-induced insulin secretion from these islets. 96 Based on the insulin secretion responses, a maximal stimulatory concentration of ACh (Tassava et al. 1992) appears to only partially activate the PRC system in islets. It is in this situation where the PRC signal transduction pathway is only-partially activated (ie when islets are exposed to 10 uM ACh or 100 nM PMA) that islets from ob/ob mice hypersecrete insulin relative to islets from lean mice. Sensitivity of the PRC signal transduction system that regulates insulin secretion, not responsiveness of this PRC system, is thus altered in islets from ob/ob mice. In contrast to the minimal responsiveness of islets from 2-wk- old ob/ob and lean mice to PRA activation via GIP (Chen and Romsos 1995). forskolin and IBM-induced PRA activation increased insulin secretion substantially. Apparently the coupling of GIP to the PRA system limits stimulation of insulin secretion from islets of 2-wk-old mice. The ability of forskolin and IBM}! to each enhance insulin secretion agrees with earlier reports (Malaisse et al. 1984, Liang and Matschinsky 1994) . Alterations in PRA-induced insulin secretion have been noted in islets of adult ob/ob mice (Black et al. 1986) but the present results obtained in islets from young ob/ob mice would suggest that these reported 97 alterations in adult mice may be a secondary consequence of prolonged hypersecretion of insulin. The origin of the enhanced sensitivity of the PRC signal transduction system to differentially stimulate insulin secretion from islets of ob/ob mice versus lean mice remains unclear. Activation of PRC influences insulin secretion via a number of mechanisms including phosphorylation of VDCCs to increase Ca“ uptake (Prentki and Matshinsky 1987, Zawalich and Rasmussen 1990) When this action of PRC was bypassed by directly activating the VDCCs with BAY R8644, insulin secretion increased equally in islets from ob/ob and lean mice (Fig. 9). Subsequent addition of 10 um ACh or 100 nM PMA to the perifusate to activate PRC further increased insulin secretion equally in islets from ob/ob and lean mice. Thus, direct activation of VDCCs masked the differential sensitivity of PRC activation to enhance insulin secretion in a phenotype- specific manner. Possibly the regulation of VDCC activation by PRC is directly or indirectly influenced in a differential way in islets of ob/ob versus lean mice. Alternatively, PRC may influence another component of the insulin secretory process, such as regulation of ca“ efflux, that can be overriden by direct activation of the VDCCs. Because high concentrations of 98 PMA also override the differential sensitivity of islets from ob/ob versus lean mice to ACh, it is likely that the physiological balance between regulatory effects of PRC on the phosphorylation of cellular proteins and counter-regulatory effects of protein phosphatases on dephosphorylation of these proteins is altered in islets of ob/ob mice. Based on the recent report that the primary genetic defect in ob/ob mice resides only in white adipose tissue, pancreatic islets from these mice would not be expected to possess any inherent defects. Rather, altered insulin secretion from islets of ob/ob mice must be secondary to a lack of functional leptin, the ob gene product, synthesized and secreted from white adipose tissue of ob/ob mice. In freshly-isolated islets, it is possible that residual effects of the in vivo environment would lead to differences in insulin secretion between islets obtained from ob/ob mice versus lean mice. It is interesting that this alteration in insulin secretion in islets from ob/ob mice is confined to the ACh and CCR coupled PLC signal transduction pathway and not to a more generalized alteration in glucose-induced insulin secretion. Islets utilized in the present study were cultured in the absence of serum and therefore in the absence of 1! persi ob/ok diff: duriz durir maint SECIE 99 of leptin for 4 days. Thus it would seem unlikely that the persistent alterations in insuin secretion in islets from ob/ob mice after 4 days in culture could be attributed to differences in availibity of leptin per se to the islets during culture. The possibility that an absence of leptin during early development. programs islets to chronically maintain an enhanced sensitivity to PRC-induced insulin secretion warrants study. CRAP incl hype: is o: role hype hype adul Pane CHAPTER V. SUMMARY AND RECOMMENDATIONS FOR FUTURE STUDY SUMMARY Genetic ob/ob mice display metabolic abnormalities including hyperinsulinemia, pancreatic islet hypertrophy and hyperplasia (Dubuc 1976). The marked hyperinsulinemia , which is one of the earliest abnormalities, could play an important role in the development of obesity. Pancreatic islets hypersecrete insulin from these ob/ob mice and this hypersecretion is likely primary to the hyperinsulinemia. In adult ob/ob mice (Tassava et al. 1992, Chen et al. 1993), pancreatic islets are enlarged, and they exhibit enhanced sensitivity and responsiveness to glucose-induced and ACh- potentiated insulin secretion. However, it is difficult to determine the primary mechanisms responsible for hypersecretion of insulin in these animals with marked pre- existing hyperinsulinemia. I therefore used 2-wk-old preobese (ob/ob) and lean mice to examine the possible primary cause of hypersecretion of insulin by freshly isolated islets of 100 ob/o] insui hype: Panc: simii of i: insui as d1 were secr secr PathI insu 0f L CCK enha GIP, acti furtI isle Path Path. 101 ob/ob mice (Chapter 3). I first examined plasma glucose and insulin. I found that ob/ob mice were slightly hyperinsulinemic and hypoglycemic at 2 weeks of age. Pancreatic islet size, DNA content, and insulin content were similar in ob/ob and lean mice. Secondly, the responsiveness of islets to glucose, as determined by 20 mm glucose-induced insulin secretion, and the sensitivity of islets to glucose, as determined.by the glucose threshold for insulin secretion, were unaffected by phenotype. Thirdly, two insulin secretagogues that. potentiate glucose-induced. insulin secretion via activation of the PLC signal transduction pathway (i.e. ACh.and CCR) were more effective in stimulating insulin secretion from islets of ob/ob mice than from islets of lean mice. Both responsiveness and sensitivity to ACh and CCR potentiation of glucose-induced insulin secretion were enhanced in islets from ob/ob mice. Further, I also examined GIP, which stimulates glucose-induced insulin secretion via activation of adenylate cyclase. GIP interacted with ACh to further augment differences in insulin secretion between islets from ob/ob and lean mice. The signal transduction pathway common to ACh and CCR, and cross-talk between this pathway and the GIP singal transduction pathway are possible 102 loci for early-onset defects in control of insulin secretion from islets of ob/ob mice. To advance our understanding of this early-onset defect in insulin secretion, I further examined. whether islet hyperresponsiveness to ACh in ob/ob mice was maintained even after long-term. removal from. their in vivo environment (Chapter 4). The results showed that a persistent alteration existed in 4 or 12 day-cultured islets of ob/ob mice in response to ACh , but not glucose. Therefore, I looked at cellular elements corrosponding to ACh-stimulated PLC signal transduction. First, ACh potentiates glucose-induced insulin secretion through the PLC - PRC signal transduction pathway. This phenotype-specific effect of ACh was mimicked by PMA (100 nl!), a PRC agonist. PRC enhances insulin release by activating voltage-dependent Ca channels (VDCCs) as well as by post-VDCCs mechanisms that directly enhance the exocytotic machinery. Islets from ob/ob and lean mice perifused in glucose plus the L-type, VDCC agonist BAY R8644 (2,10,20 uM) increased insulin secretion similarly, suggesting a normal functioning of directly activated L-type, VDCCs in islets of ob/ob mice. After activation of these VDCCs by BAY R8644 (10 uM), addition of ACh or PMA now stimulated insulin secretion __I equa pote aden acti equa isle PKC mic cul REC sea dev are :39- 103 equally from islets of ob/ob and lean mice. Secondly, GIP potentiates glucose-induced insulin secretion via the adenylate cyclase - PRA signal transduction pathway. PRA activation by either forskolin or IBMX dramatically and equally potentiated glucose-induced insulin secretion from islets of ob/ob and lean mice. 'Finally, I propose that the cellular mechanism whereby PRC activates L-type, VDCCs is altered in islets from ob/ob mice and that this alteration persists even when islets are cultured for up to 12 days. RECOMNDATIONS FOR FUTURE STUDY To better understand the present results, and to continue searching for the cellular mechanisms of insulin secretion in development of obesity in ob/ob mice, the following studies are proposed . 1. Determine whether the alteration of PRC action directly acts on VDCC gated Ca“ influx. Our present findings suggest that the action of PRC on regulation of VDCCs in islets of ob/ob mice is possibly alte: furt? furt meas int: gluc of COD: enh alt unc P08 Ca2 de: Wh is 104 altered, therefore, enhances intracellular Ca" levels and further contributes to the hypersecretion of insulin. To further investigate this possibility, intracellular Ca“ measurement should be applied. I hypothesize that intracellular Ca” levels will be similar in response to glucose in both phenotypes, but abnormally enhanced in islets of ob/ob mice in response to ACh. If so, this would be consistent with altered regulation of VDCCs in ob/ob mice. Whether the alteration of PRC action is due to abnormally enhanced translocation of PRC to plasma membrane or due to altered phosphorylation of VDCCs by PRC would still be unclear. Further work would be needed to address these possibilities. 2 . Studies are needed to determine if other PRC actions may also affect intracellular Ca“ levels. For example, PRC may play a role in governing Ca-ATPase (Ca pump), one of transporting systems which can extrude Ca" from the B-cells (Zawalich and Rasmussen 1990) . Ca-ATPase derives its energy from ATP hydrolysis. The cellular mechanism whereby PRC action activates Ca-ATPase is unclear. Ca-ATPase is a Ca-calmodulin dependent enzyme. Whether PRC plays a direct under hypod mice: than compa In hypc uM) indt in 1 how. K85 whe: fur to Act 105 direct or indirect role in regulating this enzyme is underinvestigated. It then raises a possibility that my hypothesis about abnormally elevated [Ca”], in islets of ob/ob mice may also be attributed to PRC action on Ca-ATPase rather than VDCC. One may need to use “Ca“ to study Ca“ efflux and compare the possible phenotype difference. 3. Determine whether ACh action on glucose-induced insulin secretion after BAY R8644 activiation is independent of Ca2+ influx. Again, intracellular Ca“ measurement would be applied; I hypothsize that after activation of VDCC by BAY R8644 (10 uM) PRC action would directly enhances exocytotic machinery, independent of Ca” influx. One study has shown that [Ca’fli in pancreatic B-cells increased after BAY R8644 stimulation, however, further stimulated with the combination of BAY R8644 plus forskolin, [Ca’fli did not signifincantly increase when compared to BAY R8644 alone (Ammala et al. 1993) . This implies that when [Ca”’]1 reaches maximum levels by BAY R8644, further action of PRA might depend upon PRA acting directly to enhance the exocytotic machinery. I assume that a similar action may exist for PRC. First, islets would be treated with -__.___I BAY K8 ACh. phenoI 88 ea nutri alter of r. 1993: 6t 32 durm Perm neur mate and Sec: are a alte Befo 106 BAY R8644 (10 uM), and then the combination of BAY R8644 plus ACh. We“); should be similar in both treatments and both phenotypes . 4. Studies to determine if persistently enhanced insulin secretion in islets of ob/ob mice is associated with an early developmental alteration. There are a number of examples in which early developmental nutritional manipulations have lead to subsequent persistent alterations in insulin secretion including preweaning exposure of rat pups to a high carbohydrate diet (Vadlamudi et al. 1993) , intrauterine exposure to mild hyperglycemia (Gauguier et a1. 1991) , and consumption of a low protein diet by the dam during gestation (Dahri et al. 1991) . These findings indicate permanent influences on islet morphology and on neuroregulation of insulin secretion. To characterize the metabolic basis for this defect, islets from neonatal ob/ob and lean mice need to be examined. A study of insulin secretion in mice at birth may help determine whether there are altertions in islets of ob/ob mice at birth or whether the alterations are acquired between birth and 2 weeks of age. Before studying that, identity of phenotypes (ob/ob versus 107 lean) by Southern blotting or RT-PCR would need to be developed because there is no other method to determine phenotype of these newborn mice. Whether this defect is directly or indirectly related to the absence of ob protein in ob/ob mice remains to be scrutinized. BIBLIOGRAPHY BIBLIOGRAPHY Ammala, C., Ashcroft, F.M., a Rorsman, P. (1993) Calcium- independent potentiation of insulin release by cyclic AMP in single B-cells. Nature 363:356-358 Ammala, C., Eliasson, L., Bokvist, R., Berggren, P-O, Bonkanen, R.E., Sjoholm, A., a Rorsman, P. (1994) Activation of protein kinase and inhibition of protein phosphatase play a central role in the regulation of exocytosis in mouse pancreatic 8 cells. Proc Natl Acad Sci USA 91:4343-4337 Anderson, A., Westman, J., & Hellerstrom, C. (1974) Effect of glucose on the ultrastructure and insulin biosynthesis of isolated mouse pancreatic islets maintained in tissue culture. Diabetologia 10:743-53 Atef, N., Brule, C., Bihoreau, M., & Rtorza, A. (1991) Enhanced insulin secretory response to acetylcholine by perifused pancreas of S-day-old preobese Zucker rats. Endocrinology 129:2219-2224 Bachelor, B.R., Stern, J.S., Johnson, P.R., & Mahler, R.J. (1975) Effects of streptozotocin on glucose metabolism, insulin response, and adiposoty in ob/ob mice. Metabolism 24:77-91 Berggren P-O., Larsson O. (1994) Ca”'and pancreatic B-cell function. Biochem Society Transactions 22:12-17 Berthoud ER, Jeanrenaud B (1979) Acute hyperinsulinemia and it reversal by vagotomy after lesions of the ventromedial hypothalamus in anesthetized rats. Endocrinology 105:146- 151 108 109 Black, M., Beick, B.M., & Begin-Beick, N. (1986) Abnormal regulation of insulin secretion in the genetically obese (ob/ob) mouse. Biochem J 238:863-869 Black, M.A., Fournier, L.A., Beick, B.M., & Begin-Beick, N. (1988) Different insulin-secretory responses to calcium- channel blockers in islets of lean and ob/ob (ob/ob) mice. Biochem J 249:401-407 Blonz, B.R., Stern, J.S., & Curry, D.L. (1985) Dynamics of pancreatic insulin release in young Zucker rats: a heterozygote effect. Am J Physiol 248 (Endocrinol Metab 11): E188-El93 Boissonneault, G.A., Bornshuh, M.J., Simons, J.M., Romsos, D.R., & Leveille G.A. (1978) Oxygen consumption and body fat content of young lean and obese)ob/ob) mice. Proc Soc Exp Biol Med 157:402-406 Brelje, T.C., & Sorenson, R.L. (1988) Nutrient and hormonal regulation of the threshold of glucose-stimulated insulin secretion in isolated rat pancreases. Endocrinology 123:1582-1589 Brouwer, A.E., Carroll, P.B., & Atwater, I.J. (1991) Effects of leucine on insulin secretion and beta cell membrane potential in mouse islets of langerhans. Pancreas 6:221- 228 Campfield, L.A., & Smith FJ (1983) Alteration of islet neurotransmitter sensitivity following ventromedial hypothalamic lesion..mm J Physiol 244 (Regulatory Integrative Comp Physiol) l3:R635-R640 Campillo, J.E., Luyckx, A.S., Torres, M.D., & Lefebvre, P.J. (1979) Effect of oleic acid on insulin secretion by the isolated perfused rat pancreas. Diabetologia 16:267-273 Chan, C.B., Macphail, R.M., & Mitton R (1993) Evidence for defective glucose sensing by islets of fa/fa obese Zucker rats. Can J Pharmocol 71:34-39 Chen, N-G, Tassava, T.M., & Romsos, R.D. (1993) Threshold Cl F. 110 for glucose-stimulated insulin secretion in pancreatic islets of genetically obese (ob/ob ) mice is abnormally low. J Nutr 123:1567-1574 Chen, N-G, a Romsos, D.R. (1995) Enhanced sensitivity of pancreatic islets from preobese 2-week-old ob/ob mice to neurohormonal stimulation of insulin secretion. Endocrinology 136:505-511 Corkey, B.E., Glennon, M.C., Chen, R.S., Deeney, J.T., Matschinsky, F.M., a Prentki, M. (1989) A role for malonyl-CoA in glucose-stimulated insulin secretion from clonal pancreatic B-cells. J biol Chem 264:21608-21612 Dahri, S., Snoeck, A., Reusens-Billen, B., Ramacle, C.,a Hoet, J.J. (1991) Islet function in offspring of mothers on low-protein diet during gestation. Diabetes 40(Suppl 2):115-120 Dubuc, P.U. (1976a) The development of obesity. hyperinsulinemic, and hyperglycemia in ob/ob mice. Metabolism 25:1567-1574 Dubuc, P.U. (1976b) Basal corticosterone levels of young ob/ob mice. Horm Metab Res 9:95-97 Dubuc, P.U. (1981) Non-essential role of dietary factors in the development of diabetes in ob/ob mice. J Nutr 111:1742-1748 Flatt, P.R., Bailey, C.J., Rwasowski, P., a Page, T. (1984) Plasm imunoreactive gastric inhibitory polypeptide on obese hyperglucemia (ob/ob) mice. J Endocr 101:249-256 Flatt, P.R., Bailey, C.J., Hampton, S.M., Swanston-Flatt, S.R., & Marks, V. '(1987) Immunoreactive C-peptide in spontaneous syndromes of obesity and diabetes in mice. Horm metabol Res 19:1-5 Fournier, L.A., Heick, H.M., a Begin-Heick, N. (1990) The influence of R*-induced membrane depolarization on insulin secretion in islets of lean and obese (ob/ob) mice. Biochem Cell Biol 68:243-248 111 Fournier, I... Begin-Heick N., Whitefield, J.F., a Schwartz, J-I. (1992) Comparison of the properties of the ATP- sensitive R‘ channels of pancreatic B-cells of lean and obese(Ob/0b) C57BL/6J Mice. J Membrane Biol 129:267-276 Gao, z-Y., Gilon, P. a Henquin, J-C (1994) The role of protein kinase-C in signal transduction through vasopressin and acetylcholine receptors in pancreatic B-cells from normal mouse. Endocrinology 135:191-199 Gauguier, D., Bihoreau, M-T.,Picon, B., & Rtarza, A. (1991) Insulin secretion in adult rats after intrauterine exposure to mild hyperglycemia during late gestation. Diabetes 40(Suppl. 2) 109-114 Gilon, P., Jonas, J.C., a Henquin, J.C. (1994) Culture duration and conditions affect the oscillations of cytoplasmic calcium concnetration induced by glucose in mouse pancreatic islets. Diabetologia 37:1007-1014 Goberna, R., Tamarit, J., Osorio, J., Fussganger, R., Tamarit, J., & Pfeiffer, E.F. (1974) Action of B-hydroxy butyrate, acetoacetate, and palmitate in the insulin release in the perfused isolated rat pancreas. Horm Metab Res 6:256-260 Gray, D.S. (1989) Diagnosis and prevalence of obesity. Med Clin North Am 73:1-13 Grodsky, G.M. & Bolaffi, J.L. (1992) Desensitization of the insulin-secreting beta cell. J Cell Biochem 48:3-11 Hayek, A. (1980) Insulin release in long-term culture from isolated islets of obese and lean Zucker rats. Horm Metab Res 12:85-86 Henquin, J.C., Schmeer, W., Nenquin, M., & Meissner, H.P. (1985) Effects of a calcium channel adonist on the electrical, ionic and secretory events in mouse pancreatic B-cells. Biochem and Biophys Res Commun 131:980-986 Jeanrenaud, B. (1985) An hypothesis on the etiology of 112 obesity: dysfunction of the central nervous system as a primary cause. Diadetologia 28:502-513 Rang, J.S., Pilkington,J.D., Ferguson, D., Rim ,H.-R. & Romsos,D. R. (1992) Dietary glucose and fat attenuate effects of adrenalectomy on energy balance in ob/ob mice. J Nutr 122:895-905 Rhan, A., Chandramouli, V., Ostenson, C-G, Ahren, B., Schumann, W.C., Low, H. Landau, B.R., a Efendic, S. (1989) Evidence for the presence of glucose cycling in pancreatic islets of the ob/ob mouse. J Biol Chem 264: 9732-9733 Rhan, A., Cao, H-L, & Landau B.R. (1995) Glucose-6- phosphatase activity in islets from.ob/ob and lean mice and the effect of dexemethasone. Endocrinology 136: 1934-1938 Rekow, J., Ulrichs, R., Muller-Ruchholtz, W., 5 Gross, W.C. (1988) Measurement of rat insulin : Enzyme-linked immunosorbent assay with increased sensitivity, high accuracy, and greater practicability than established radioimmunoassay. Diabetes 37:321-326 Romatsu, M., Yokokawa, N., Takedas T., Nagasawa, Y., Aizawa, T. & Yamada, T. (1989) Pharmacological characterization of the voltage-dependent calcium.channel of pancreatic B- cell Endocrinology 125:2008-2014 Ruczmarski, R.J., Flegal, R.M., Campbell, S.M., & Johnson, C.L. (1994) Increasing prevalence of overweight among US adults. JAMA 272:205-211 Ruffert, A., Stern, J.S., & Curry, D.L. (1988) Pancreatic hypersensitivity to glucose by young obese Zucker (fa/fa) rats. Metabolism 37:952-957 Rukulian, M., Li, M. Y., & Atwater, I. (1990) Characterization of potassium.channels in pancreatic B-cells from ob/ob mice. FEBS lett 266: 105-108 113 Labarca, C., & Paigen, R. (1980) A simple, rapid and sensitive DNA assay procedure. Anal Biochem 102:344-352 Lacy, P.E., & Rostianovsky, M. (1967) Method for the isolation of intact islets of langerhans from the rat pancreas. Diabetes 16:35-39 Lavine, R.L., Voyles, N., Perrino, P.V., & Recant L. (1977) Functional abnormalities of islets of langerhans of obese hyperinsulinemic mouse. Am.J Physiol E86-E90 Lee, H.C., Curry, D.L., a Stern, J.S. (1993) Selective muscarinic sensitivity in perfused pancreata of obese Zucker rats. Intern J Obesity 17:569-577 Lenzen, S. (1992) Glucokinase: signal recognition enzyme for glucose-induced insulin secretion. In: Nutrient Regulation of Insulin Secretion. edited by P.R. Flatt. Portland press, London, D.R., pp.101-124 Lenzen, S. a Panten, U. (1988) Signal recognition by pancreatic B-cells. Biochem Pharmacol 42:1385-1391 Lenzen, S. & Tiedge, M. (1992) Regulation of pancreatic B- cell glucokinase and GLUT2 glucose transporter gene expression.In: Nutrient Regulation of Insulin Secretion. edited by P.R. Flatt. Portland press, London, D.R.. PP. 1-6. Lernmark, A. (1974) The preparation of, and studies on, free cell suspensions from.mouse pancreatic islets. Diabetologia 10:431-438 Lew, E.A. & Garfinkel, L. (1979) Variations in mortality by weight among 750000 men and women. J Chronic Dis 32:563- 576 Liang, Y., Najafi, H., Smith, R.M., Zimmemman, E.C., Magnuson, M. A., Tal, M., E Matschinsky, F.M. (1992) Concordant glucose induction of glucokinase, glucose usage, and glucose-stimulated insulin release in pancreatic islets maintained in organ culture. Diabetes 41:792-806 114 Liang, Y., & Matschinsky, F.M. (1994) Mechanisms of action of nonglucose insulin secretagogues. Annu Rev Nutri 14:59-81 Licinio-Paixao, J., Polonsky, R.S., Given, R.D., Pugh, W., Ostrega, D., Frank, B.F. & Rubenstein, A.H. (1986) Ingestion of a mixed meal does not affect the metabolic clearance rate of biosynthetic human C-peptide. J Clin Endocrinol Metab 63: 401-403 Lin, P.Y., Romsos, D.R., & Leveille, G.A. (1977) Food intake, body weight gain, and body composition of the young obese (ob/ob) mouse. J Nutr 107:1715-1723 Lin, P.Y., Romsos, D.R., Vander Tuig, J.G., & Leveille, G.A. (1979) Maintenance energy requirements, energy retention and heat production of young obese (ob/ob) and lean mice fed a high-fat or a high-carbohydrate diet. J Nutr 109:1143-1153 Loten, E.G., Rabinovitch A., & Jeanrenaud, B. (1974) In vivo studies on lipogenesis in obese-hyperglycaemic (ob/ob) mice: Possible role of hyperinsulinemia. Diabetologia 10:45-52 Lu, M., Wheeler M.B., Leng, x-n, & Boyd A.E. (1993) The role of the free cytosolic calcium level in S-cell signal transduction. by' gastric inhibitory' polypeptide and glucagon-like peptide I(7-37). Endocrinology 132:94-100 Malaisse, W.J., Garcia-Marales, P., Dufrane, S.P., Sener, A., & Valverde, I. (1984) Forskolin induced activation of adenylate cyclase, cyclic AMP production and insulin release in rat pancreatic islets. Endocrinology 115:2015- 2020 Malaisse, W.J., Malaisse-Lagae, F., Davies, D.R., & Van Schaftingen, E. (1989) Presence of fructokinase in pancreatic islets. FESS Lett 255:175-178 Malaisse, W.J., Malaisse-Lagae, F., Davies, D.R., Vandercammen, A. a Van Schaftingen, E. (1990) Regulation of glucokinase by a fructose-l-phosphate- 115 sensitive protein in pancreatic islets. Eur J Biochem 190:539-545 Malaisse-Lagae, F., Mathias P.C.F., & Malaisse, W.J. (1984) Gating and.blocking of calcium.channels by dihydropyridines in the pancreatic B-cell. Biochem Biophys Res Comm 123:1062-1068 Mccullough, A.J., Marshall, J.B., Bingham, C.P., Rice, B.L., Manning, L.D., Ralhan, S.C. (1985) Carbachol modulates GIP-mediated insulin release from rat pancreatic lobules in vitro. Am J Physiol 248 (Endocrinol Metab 11):E299- E303 Milburn, J.L., Hirose, H., Lee, Y.H., Nagasawa, Y., Ogawa, A., Ohneda, M., Beltrandelrio, H., Newgard, C.B., Johnson, J.B., & Unger, R.H. (1995) Pancreatic B-cells in obesity : evidence for induction of functional, morphologic, and.metabolic abnormalities by increased long chain fatty acids. J Biol Chem 270:1295-1299 Mistry, A.M., Chen, N-G, Lee, Y-S, & Romsos, D.R. (1995) Dietary glucose enhances the sensitivity of pancreatic islets from.adrenalectomized genetically obese (ob/ob) mice to glucose-induced insulin secretion. J Nutr 125:503-511 Morgan, L.M. (1992) Insulin secretion and the entero-insular axis. In: Nutrient Regulation of Insulin Secretion. edited by P.R. Flatt. Portland press, London,‘U.R., pp. 1-22. Nishizuka, Y. (1984) The role of protein kinase C in cell surface signal transduction and tumor promotion. Nature 308:693-698 Ono, J., Takaki, R., & Fukuma, M; (1977) Preparation of single cells from pancreatic islets of adut rat by the use of dispase. Endocrinol Japon 24:265-270 Plant, T.D. (1988) Properties and calcium-dependent inactivation of calcium currents in cultured mouse pancreatic B-cells. J Physiol 404:731-74 116 Polonsky, R.S., Given, B.D., Hirsch L., Shapiro, E.T., Tillil H., Beebe, C., Galloway, J.A., Frank, B.E., Rarrison, T., vanCauter, E. (1988) Quantitative study of insulin secretion and clearance in normal and obese subjects. J Clin Invest 81:435-441 Prentki, M., & Matshinsky, F.M. (1987) Ca", cAMP and phospholipid-derived messengers in coupling mechanisms of insulin secretion. Physiol Rev 67:1185-1248 Prentki, M., Vischer, S., Glennon, M.C., Regazzi, R., Deeney, J.T., & Corkey, B.E. (1992) Malonyl-CoA and long chain acyl-CoA esters as metabolic coupling factors in nutrient-induced insulin secretion. J Biol Chem 267:5802- 5810 Ramanadham, S. & Turk, J. (1994) w-Conotoxin inhibits glucose-induced and arachidonic acid-induced rises in intracellular [Ca“] in rat pancreatic islet B-cell. Cell Calcimm 15:259-264 Robertson, R.P., Olson, L.R., & Zhang, H-J (1994) Differentiating glucose toxicity from glucose desensitization : a new message from the insulin gene. Diabetes 43:1085-1089 Rohner-Jeanrenaud, F., Hochstrasser, A.C., & Jeanrenaud, B. (1983) Hyperinsulinemia of preobese and obese fa/fa rats is partly vagus nerve mediated. Am.J Physiol 244 (Endocrinol Metab 7):E317-E322 Rorsman, P., Ashcroft, F.M., & Trube, G. (1988) Single Ca channel currents in mouse pancreatic B-cells. Pflugers Arch 412:597-603 Schaftingen, E.V., Detheux, M., 5 Da Cunha, M.V.(l994) Short-term control of glucokinase activity: role of a regulatory protein. FASEB J 8:414-419 Schnefel, S., Banfic, H., Eckhardt, L., Schultz, G., & Schulz,I. (1988) Acetylcholine and cholecystokinin receptors functionally couple by different G-proteins to phospholipase C in pancreatic acinar cell. FEBS lett 117 230:125-130 Stunff, C.L., & Bougneres, P. (1994) Early changes in postprandial insulin secretion, not in insulin sensitivity, characterize juvenile obesity. Diabetes 43:696-702 Tassava, M.T., Okuda, T., & Romsos, R.D. (1992) Insulin secretion from ob/ob mouse pancreatic islets: effects of neurotransmitters. Am J Physiol 262 :E338-E343 Vadlamudi, S., Hiremagalur, B.R., Tao, L., Ralhan, S.C., Ralaria, R.N., Raung, H-L. C., a Patel, M.S. (1993) Long- term effects on pancreatic fuction of feeding a BC formula to rats during the preweaning period. Am J Physiol 265 (Endocrinol Metab 28):E565-571 Van Itallie, a T.B., (1985) Health implication of overweitght and obesity in the United States. Ann Intern Med 103:983-988 Vara, B., Fernandez-Martin, C., Garcia, C., & Tamarit- Rodriguez, J. (1988) Palmitate dependence of insulin secretion, 'de novo' phospholipid synthesis and ”Ca“- turnover in glucose stimulated rat islets. Diabetologia 31:687-693 Zawalich, M.T., Rognstad,R., Pagliara, A.S., & Matschinsky, F.M. (1977) A comparison of the utilization rates and hormone-releasing actions of glucose, mannose, and fructose in isolated pancreatic islets. J Biol Chem 252:8519-8523 Zawalich, W.S. & Zawalich R. C. (1988a) Induction of memory in rat pancreatic islets by tolbutamide. Diabetes 37:816- 23 Zawalich, W.S. (1988b) Synergistic impact of cholecystokinin and gastric inhibitory polypeptide on the relation of insulin secretion. Metabolism 37 :778-781 Zawalich, W.S. (1988C) Time-dependent potentiation of insulin release induced by alpha-ketoisocaproate and 118 leucine in rats: possible involvement of phosphoinositide hydrolysis. Diabetologia 31:435-442 Zawalich, W.S., Diaz, V.A., & Zawalich R. C. (1989a) Role of phosphoinositide metabolism.in duration of memory in isolated perifused rat islets..mm J Physiol 254:E609- E616 Zawalich W.S., Zawalich R.C. & Rasmussen H. (1989b) Interaction between cholinergic agonists and enteric factors in the regulation of insulin. Acta Endocrinol (Copenh) 120:702-707 Zawalich, W.S., & Rasmussen, H. (1990) Control of insulin secretion: a model involving Ca“, cAMP and diacylglycerol. Mol Cell Endocrinol 70:119-137 Zhang, Y., proenca, R., Maffei, M., Barone, M., Leopold, L. & Friedman, J.M. (1994) Positional cloning of the mouse obese gene and its human homologue. Nature 372:424-432 "IIIII'IIJIIIIIIIIIIIIIII“