UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Sensitization of isolated rat islets to stimulation with glucose by prior exposure to glucose-dependent… Fell, Charlene Deanne 1995

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_1995-0136.pdf [ 6.29MB ]
Metadata
JSON: 831-1.0086872.json
JSON-LD: 831-1.0086872-ld.json
RDF/XML (Pretty): 831-1.0086872-rdf.xml
RDF/JSON: 831-1.0086872-rdf.json
Turtle: 831-1.0086872-turtle.txt
N-Triples: 831-1.0086872-rdf-ntriples.txt
Original Record: 831-1.0086872-source.json
Full Text
831-1.0086872-fulltext.txt
Citation
831-1.0086872.ris

Full Text

S E N S I T I Z A T I O N O F I S O L A T E D R A T I S L E T S T O S T I M U L A T I O N W I T H G L U C O S E B Y P R I O R E X P O S U R E T O G L U C O S E - D E P E N D E N T I N S U L L N O T R O P I C P O L Y P E P T I D E (GIP) by CHARLENE DEANNE FELL B.Scv The University of British Columbia, 1992 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in THE FACULTY OF GRADUATE STUDIES Department of Physiology We accept this thesis as cortforming to the required standard UNIVERSITY OF BRITISH COLUMBIA February 1995 © Charlene Deanne Fell, 1995 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department The University of British Columbia Vancouver, Canada DE-6 (2/88) ABSTRACT This thesis describes the relationship between glucose- and glucose dependent insulinotropic polypeptide (GlP)-stimulated insulin release from the pancreatic islet and P cell. Protocols were developed to test the hypothesis that the intact islet contains heterogeneous populations of (3 cells and that GIP can sensitize these populations to subsequent stimulation by glucose. Changes in intracellular free calcium- concentration in response to glucose or GIP administered prior to glucose were used to test the hypotheses, respectively. Culture conditions were developed which allowed the measurement of free intracellular calcium from individual (3 cells within the intact islet. Due to technical difficulties and questionable physiological relevance of the islet cultures developed to test these hypotheses, attempts to measure intracellular free calcium were abandoned and the hypotheses not tested. An alternate approach to investigating the relationship between glucose- and GIP-stimulated insulin secretion was developed by measuring insulin secretion from perifused whole islets. Insulin secretion in response to 1 nM GIP administered 1 h immediately prior to 2 h of 11.0 mM glucose was found to be significantly. (p<0.05) greater than that secreted in response to 11.0 mM glucose without prior application of GIP. This priming effect of GIP persisted when the interval between GIP and glucose application was increased to 20 or 40 minutes. No significant difference in the priming and potentiating effects of GIP on glucose-stimulated insulin secretion from islets was observed. Although these studies do not provide direct evidence for the mechanisms of glucose- and GIP-stimulated insulin secretion, they provide indirect evidence for crosstalk between second messengers generated by glucose and GIP within the P cell. TABLE OF CONTENTS ABSTRACT i i TABLE OF CONTENTS iii LIST OF TABLES v LIST OF FIGURES v ACKNOWLEDGEMENTS vii CHAPTER 1 Introduction 1 Insulin 1 Discovery of Insulin 1 Role of Insulin in Whole Body Homeostasis 2 Regulation of Insulin Release 3 p cell physiology 4 Glucose-Stimulated Insulin Release 4 Glucose Uptake and Metabolism 5 P cell Electrophysiology 6 Intracellular Calcium 6 Exocytosis of Insulin Secretory Vesicles 7 The Adenylate Cyclase/cAMP Cascade 8 The DAG/IP3 Cascade 8 The Enteroinsular Axis 9 Identification of the Enteroinsular Axis 9 Glucose-Dependent Insulinotropic Polypeptide 10 Other Incretins 13 Effects of Incretin Interaction on Insulin Release 14 P cell Heterogeneity in Response to Glucose 15 Proemial Sensitization 17 Hypotheses and Rationale 19 CHAPTER 2 The Priming Effects of GIP on Glucose-Stimulated Increases of Intracellular Free Calcium in Individual Pancreatic P cells Within the Intact Islet 23 Introduction 23 Materials and Methods 23 Islet Isolation 23 Subconfluent Islet Monolayer Culture 25 Immunocytochemistry 26 C a 2 + Imaging 28 CHAPTER 2 (continued) Data Analysis and Statistical Analysis 30 Results 31 Subconfluent Monolayer Islet Culture. 31 Measurement of Intracellular Free Calcium 32 Glucose Gradients at 20 °C 33 1 or 2 Minute Pulses of High Glucose at 20 °C 34 Sustained High Glucose at 20 °C 34 Sustained High Glucose at 37 °C 35 Step-Wise Increases to High Glucose 35 CHAPTER 3 The Priming Effects of GIP on Glucose-Stimulated Insulin Secretion from Perifused Rat Pancreatic Islets 52 Introduction 52 Materials And Methods 52 Islet Isolation ; 52 Islet Perifusion... 53 Insulin determination 55 Data Analysis..... 57 Statistical Analysis 57 Protocol Development 58 2 x 20 Minute Pulse Protocol 58 2x2 Hour Pulse Protocol 58 1x2 Hour Pulse Protocol.. 60 Results 60 The Priming Effects of GLP on Glucose-Stimulated Insulin Secretion 60 Latency of the Priming Effects of GIP on Glucose-Stimulated Insulin Secretion 60 CHAPTER 4 Discussion 68 Development Of Subconfluent Monolayer Islet Cultures 69 Measurement of Intracellular Free Calcium from Individual (3 Cells within the Intact Islet..... ..; 70 The Priming Effects of GLP on Glucose-Stimulated Insulin Secretion from Perifused Islets... 76 Future Directions : 82 Conclusions..... :'. 83 REFERENCES 85 LIST OF TABLES V Table I Protocols developed to elicit [Ca2+]i responses from individual B cells within the intact islet 32 Table II. Islet perifusion protocols 59 LIST OF FIGURES Figure 1. Schematic diagram of the second messenger systems involved in glucose- and agonist-induced insulin secretion from the pancreatic B cell 22 Figure 2. Formation of a subconfluent monolayer islet culture of an isolated islet ; 36 Figure 3. Subconfluent monolayer islet cultures immunostained for insulin (brown) and glucagon (blue) 38 Figure 4. Schematic diagram of the equipment used to measure [Ca 2 +]; 43 Figure 5. A: A subconfluent monolayer islet culture immunostained for insulin (DAB, brown) and glucagon (DAB-Ni, blue). B. Captured image of the same islet using Attofluor software 45 Figure 6. Change in background corrected 334/380 fluorescence excitation ratio of a single B cell within an intact islet to a graded increase of glucose from 20 mM to 25 mM 47 Figure 7. Mean change in background corrected 334/380 fluorescence excitation ratio of individual islet cells to 25 mM glucose, 25 mM glucose + 10-7 M GIP, arm 25 mM glucose + 20 ug/ml Br-A231875. !.. 48 vi Figure 8. Mean change in background corrected 334/380 fluorescence excitation ratio of islet cells to 25.0 mM glucose at 37°C 49 Figure 9. Mean change in background corrected 334/380 fluorescence excitation ratio of islet cells to sustained 25.0 mM glucose at 37°C 50 Figure 10. Mean change in background corrected 334/380 fluorescence excitation ratio of islet cells to changes in glucose concentration from 2.0 mM glucose to 4.4 mM glucose and 25.0 mM glucose 51 Figure 11. Schematic diagram of perifusion equipment 62 Figure 12. Insulin secretory response of perifused rat islets to glucose and GIP 64 Figure 13. Integrated insulin secretory resp'onse of perifused rat islets to (a) 11.0 mM glucose, (b) 11.0 mM glucose plus 1 nM GIP, and (c) 1 nM GIP administered prior to 11.0 mM glucose 65 Figure 14. Latency effects of a priming dose of 1 nM GIP on glucose stimulated insulin secretion from perifused rat islets 66 Figure 15. Integrated insulin secretory response of perifused rat islets to 1 nM GIP administered 0,20, or 40 minutes prior to 11.0 mM glucose 67 ACKNOWLEDGEMENTS vii I extend a sincere thank-ybu to the people who helped me complete this thesis and who made the last three years special: I am grateful to my committee members: Drs. Alison Buchan, John Church, Ken Curry, Steve Kehl, and Roger Brownsey, for guidance through a sometimes difficult project. I thank in particular to Alison Buchan; John Church, and Steve Kehl for critical reading of the manuscript and for the use of their equipment. I am grateful for the expert technical assistance of Leslie Checknita. I also thank Lydia Taylor, Robert Pauly, and Xlda Ng for technical assistance. I thank John Sanker, Joe Teh, and Christina Fell for their assistance' in the completion of figures for this thesis. I thank Jack Lewis for timely repairs to bur equipment. I extend a sincere thank you to Dave Phelan for expert animal care. I thank Marie Langton for excellent administrative assistance which went above and beyond the call of duty, and for her ever-ready giggle. I am grateful for the support and friendship of my fellow graduate students. In particular, I thank Tim Kieffer for his patience, humour, and meticulous instruction in the lab. I thank Khaled Abdul-Hamid for sharing his expertise in calcium imaging, and for our interesting and varied cbriversatiBhs. I thank Gabrielle Weichert for encouraging me both academically and athletically, and for her friendship. I also thank Kristi Mcintosh for her support and friendship throughout my time at UBC. I would like to thank Dr. Card! Ann Courneya for introducing me to teaching, TIPS, and triathlons. I thank her also for her enthusiastic support and friendship, and for serving as an unsuspecting role rftodel. I extend a special thank you to* my supervisor, Dr. Ray Pederson. I thank Ray for his friendship, guidance, and uitwai^ering support throughout the course of my time in Physiology. I will always rem'e'm'bisr Gfad Retreats at Mayne Island, potlucks at the viii Pederson's, and Physiology Christmas parties. I thank Ray for making Graduate Studies so much more than just trie completion of an M.Sc. thesis. Special thanks to my faniily, whose support and encouragement was always there, always felt, and always appreciated. CHAPTER 1 Introduction This thesis investigates the relationship between glucose- and glucose-dependent insulinotropic polypeptide (GlP)-stimulated insulin secretion from the pancreatic p" cell. It is well established that f3 cells secrete insulin in response to elevated plasma glucose levels, and that GIP potentiates glucose-stimulated insulin secretion. However, although much is known about the mechanisms responsible for glucose- and GIP-stimulated insulin secretion, little is known about the effects of the interaction of these mechanisms on the generation of second messengers in the |3 cell. INSULIN Discovery of Insulin The discovery of insulin stems from work by Minkowski and von Merring in 1889 which showed that pancreatectomized dogs exhibited symptoms similar to diabetes mellitus in humans (Minkowski, 1989). This suggested that the pancreas plays a key role in the development of this disease. In 1921, Banting and Best isolated a pancreatic extract, later named insulin, which was found to lower plasma glucose when injected into diabetic dogs. The plasma glucose-lowering effect of insulin was later demonstrated in humans (Banting et al., 1922). Early production of large quantities of insulin was problematic. A reliable method for the extraction of insulin was devised by MacLeod and Collip (Bliss, 1982). Banting and MacLeod won the Nobel Prize in Medicine or Physiology in 1923 for their pioneering work on insulin. 2 Role of Insulin in Whole Body Homeostasis The maintenance of plasma glucose levels within a narrow range is of critical importance in mammals as both hyper- and hypoglycemia can lead to tissue damage over time. Insulin is the primary hormone involved in maintaining glucose homeostasis and in regulating the storage and utilization of nutrients. Insulin has both short term and long term effects on target tissues. Rapid effects of insulin include the activation of glucose transport systems in muscle and fat, resulting in a lowering of plasma glucose levels. Long term effects of insulin include the stimulation of cell proliferation and differentiation. The metabolic effects of insulin are mediated through a heterotetramer tyrosine kinase receptor (Kahn and White, 1988; Ullrich and al., 1985). Cells with the highest concentrations of the insulin receptor, and thus the most responsive to the hormone, are liver, muscle, and fat cells (Kahn & White, 1988). The insulin receptor consists of two extracellular A chains, which contain the insulin binding domain of the receptor, and two B chains, which contain a ligand-activated tyrosine kinase function (Kahn & White, 1988; Ullrich & al., 1985). Binding of insulin to its receptor induces auto-phosphorylation of several tyrosine residues on the B chain (Kasuga et al., 1982; Rosen, 1987). The activated receptor then catalyses the phosphorylation or dephosphorylation of several cellular proteins. The result is a cascade of protein phosphorylation/dephosphorylation reactions within target tissues. Diabetes mellitus is a disease characterized by an inability to assimilate a glucose load. There are two major types of diabetes mellitus: insulin-dependent diabetes mellitus (IDDM, or juvenile onset diabetes) and non-insulin dependent diabetes mellitus (NIDDM, or adult or late onset diabetes). Insufficient insulin production or complete lack of the hormone results in IDDM (Dotta and Eisenbarth, 1989; Kahn and Rosenthal, 1979). Lack of insulin in this pathological state is due to the selective destruction of p cells by the immune system by an as yet unknown mechanism (Kahn & Rosenthal, 1979). NIDDM is the result of peripheral resistance to the actions of insulin, and is often accompanied by a loss of the insulin secretory response of p cells to a glucose challenge (Weir et al., 1986). Unlike type I diabetes mellitus, there is no loss of p cells in type II diabetes mellitus; rather, insufficient insulin production is thought to be due to desensitization of the p cells to chronic mild hyperglycemia (Genuth, 1973). Regulation of Insulin Release Insulin secretion is regulated by a variety of nutritive, neural, and hormonal factors acting on the p cell. The primary stimulus for insulin secretion is glucose, although other nutritive factors such as amino acids also stimulate insulin release (Rasmussen et al., 1990). Vagal stimulation of the p cells also leads to insulin secretion. Vagal afferents to the p cell release acetylcholine (ACh), which acts via M3 muscarinic receptors (Ahren et al., 1986). Vasoactive intestinal peptide (VIP) and gastrin releasing peptide (GRP) are also released from vagal afferents to the p cells, comprising non-cholinergic vagal stimulation of insulin secretion (Ahren et al., 1986; Bailey et al., 1989). cc2-adrenergic stimulation of p cells inhibits insuUn release, as do neuropeptide Y (NPY) and galanin (Ahren et al., 1986). Hormones which stimulate insulin secretion include gastrin, secretin, cholesystokinin (CCK), glucagon, gastric inhibitory polypeptide (GIP), and glucagon-like peptide (GLP-1) (Rasmussen et al., 1990). Somatostatin inhibits insulin release (Rasmussen et al., 1990). Insulin secretion may also be influenced by the other hormones secreted by the islet: glucagon, somatostatin, and pancreatic polypeptide. To this extent, islet morphology has an important role in the regulation of insulin release. Islets are composed of glucagon (a)-, insulin (P)-, somatostatin (6)-, and pancreatic polypeptide (F)-secreting cells, p cells comprise 60-80% of the total cell population of the islet with ex, 5, and F cells forming a discontinuous mantle around a p cell core in rats (Orci, 1986). Secretions from a and 6 cells may have a paracrine effect on p cells which are located on the periphery of the p cell core. Likewise, peripheral p cells may influence glucagon and somatostatin secretion. Ionic and metabolic coupling of p cells within the islet has been shown to play an important role in the regulation of insulin secretion (Gylfe et al., 1991; Orci, 1986; Orci et al., 1975; Pipeleers et al., 1985a; Pipeleers et al., 1985b; Valdeolmillos et al., 1993). Electron microscopy (Orci et al., 1975) and injection of fluorescent dyes (Micheals and Sheridan, 1981) revealed the existence of gap junctions between like and unlike cell types within the islet. Considering the effects of glucagon and somatostatin on insulin release, ionic and metabolic coupling between p cells and either a or 6 cells may also have an important function in the regulation of insuHn secretion (Pipeleers et al., 1985a; Pipeleers et al., 1985b). Insulin is synthesized and packaged into secretory vesicles as proinsulin which consists of linked A, B, and C peptides (Orci, 1986). Within the secretory vesicle, C peptide is cleaved from proinsulin to yield insulin. Insulin and C peptide are thus secreted in equimolar concentrations from the p cell. Although C peptide does not have any known physiological function, it serves as a useful index of the rate of endogenous insulin production in individuals who are receiving insulin therapy (Polonski and Rubenstein, 1986). Normal, fasted individuals secrete insulin at a rate of approximately 40 fig/hour giving rise to plasma insulin concentrations of approximately 0.5 ng/ ml in the peripheral circulation (Schade et al., 1983). B CELL PHYSIOLOGY Glucose-Stimulated Insulin Release Insulin secretion results from the synthesis of various intracellular second messenger signals which are generated by stimulation of the p cell by one or more insulin secretagogues. Activation of second messengers in the p cell ultimately results in an increase in intracellular calcium. It is this rise in [Ca2"1"]^ through an as yet unknown mechanism, which triggers the release of insuUn from the p cell. Most insulin secretagogues exert their effects on the p cell by activating either the adenlyate cyclase or phospholipase C second messenger systems. Glucose, however, stimulates insulin secretion by activating both systems. For this reason, p cell physiology will be discussed with respect to glucose-stimulated insulin release. A schematic diagram illustrating the various second messenger systems involved in insulin release is given in Figure 1. Glucose Uptake and Metabolism Glucose enters the p cell by facilitated diffusion through low affinity, high capacity GLUT 2 transporters (Km = 17 mM) (Purrello et al., 1993) which enable p cells to respond to a wide range of plasma glucose concentrations. Glucose is metabolized to produce ATP. The rate limiting step in this process in normal cells is the phosphorylation of glucose to glucose-6-phosphate by glucokinase (Purrello et al., 1993). Because the capacity of GLUT 2 transporters to transport glucose into the p cell exceeds the activity of glucokinase, this enzyme, and not the GLUT 2 transporter, plays a key role in the regulation of glucose-stimulated insulin secretion from normal p cells (Purrello et al., 1993). The generation of ATP within the p cell leads to a complex pattern of electrophysiological events. Dean and Matthews first reported glucose-induced membrane depolarizations and action potentials in p cells (Dean and Matthews, 1970b). It is now well established that the uptake and metabolism of glucose (Dean et al., 1975) and subsequent generation of ATP within the p cell leads to a reduction of the open probability of ATP-sensitive K + channels, decrease in K + conductance across the membrane, and membrane depolarization (Ashcroft et al., 1988; Dean et al., 1975). P cell Electrophysiology Pancreatic p cells have a resting membrane potential of approximately -70 mV. Closure of ATP-sensitive K+ channels depolarizes p cells to approximately -40 mV (Dean & Matthews, 1970b). This change in membrane potential is sufficient to open voltage dependent C a 2 + channels (VDCC) in the p cell membrane, which are activated at membrane potentials between -50 and 50 mV (Keahey et al., 1989; Rorsman and Trube, 1986). The rapid influx of calcium into the p cell during VDCC activation is the predominant cause of glucose-induced electrical activity and action potentials within these cells (Dean and Matthews, 1970a). P cells show neither electrical activity nor membrane depolarization in response to glucose levels which are substimulatory for insulin secretion (Meissner and Schmelz, 1974). In response to 5.5 mM - 16.6 mM glucose, electrical activity in p cells has been described. Bursts of action potential-like spikes were initiated by a depolarization to a plateau level from which spiking activity arose (Meissner & Schmelz, 1974; Atwater et al., 1978). Repolarization to the resting membrane potential ended each burst (Meissner & Schmelz, 1974). The electrical activity exhibited by p cells is glucose concentration-dependent; with increasing glucose concentrations, the duration of each burst increased and the interval between each burst decreased (Meissner & Schmelz, 1974). Cyclic changes in p cell electrical activity have also been described in terms of fluctuations in p cell membrane resistance to K + and Ca 2 + (Atwater et al., 1978). Intracellular Calcium P cell electrical activity in response to glucose has been correlated to oscillations in [Ca 2 +]j measured with the fluorescent calcium-binding dye, fura-2, (Grapengiesser et al., 1989, 1992; Gylfe, 1988; Leech et a l , 1994; Sakurada et al., 1993; Valdeolmillos et al., 1993; Wang et al., 1992, 1993; Yada et al., 1992). Single p cells exhibit an initial decrease followed by an increase in [Ca 2 +] ; in response to glucose (Gylfe, 1988; Yada et al., 1992). The initial decrease in [Ca 2 +]; is thought to be due, in part, to sequestering of 7 [ C a 2 + ] 2 by the endoplasmic reticulum. Moderate levels of glucose lead to the generation of [Ca 2 +]j oscillations in single (3 cells with a periodicity of 2-6 minutes (Grapengiesser et al., 1989; Herchulez et al., 1991; Valdeolmillos et al., 1993). At higher glucose concentrations,[Ca2+]; oscillations develop into a sustained increase in [Ca 2 +]j (Grapengiesser et al., 1989). Calcium oscillations have also been correlated with synchronous oscillations in IP3 in murine islets (Barker et al., 1994). Several studies have demonstrated an ATP-dependent C a 2 + store in the endoplasmic reticulum (ER) which can be mobilized by IP3 (Biden et al., 1984; Hellman etal., 1986; Nilsson et al., 1987; Streb et al., 1983). The ATP-dependent uptake of calcium by the ER is thought to buffer intracellular calcium. Binding of IP3 to an IP3 receptor/ C a 2 + channel on the IP3-sensitive calcium store releases calcium from the ER. The ER may also contain an IP3-insensitive store of calcium (Nilsson et al., 1987). Calcium may be released from these stores by elevated intracellular calcium (Ca2+-induced Ca2+release) (Berridge and Rapp, 1979). Exocytosis of Insuhn Secretory Vesicles Although the link between the rise in [Ca 2 +]j and insulin secretion is poorly understood, it has been postulated that an increase in [Ca 2 +]j concentrations lead to changes in the function of Ca 2 +/calmodulin regulated proteins which may be associated with exocytosis of insulin secretory vesicles (Efendic et al., 1991). Although a relationship between oscillations on [Ca 2 + ] i and insulin secretion has often been assumed, simultaneous oscillations of [Ca 2 +]/ and insulin secretion have only recently been demonstrated (Gilon et al., 1993). Using changes in membrane capacitance as an index for exocytosis, (Ammala et al, 1993) found a strong correlation between increased intracellular calcium and exocytosis. Furthermore, these investigators found the concentration of calcium at secretory sites to be in the micromolar range. Inhibitors of C a 2 + / calmodulin dependent protein kinases caused a decrease in depolarization-induced exocytosis (Ammala et al., 1993), suggesting a relationship between protein phosphorylation and exocytosis. The diameter of insulin secretory granules, estimated from changes in membrane capacitance, is approximately 250 nm (Ammala et al., 1993). The Adenylate Cyclase/ cAMP Cascade cAMP has been shown to be generated in the p cell in response to glucose by two mechanisms: Ca2+/calmodulin-dependent activation of adenylate cyclase and activation of adenylate cyclase by G proteins which are coupled to neurohormone receptors on the p cell plasma membrane (Prentki and Matschinsky, 1987). cAMP activates a cAMP-dependent protein kinase (PKA) which phosphorylates latent VDCCs, rendering them responsive to changes in membrane potential (Ammala et al., 1993). In this way, cAMP enhances the Ca 2 +signal by increasing the number of functional VDCCs in the plasma membrane. cAMP has also been postulated to modulate the activity of ATP-sensitive potassium channels (KATP)hi the plasma membrane (Holz et al., 1993). cAMP has been shown to interact directly with the exocytotic machinery (Ammala et al., 1993; Gillis and Misler, 1993). cAMP is also believed to restrict agonist-induced hydrolysis of phosphoinositols, and thus the generation of DAG and IP3 (Zawalich, 1988). The DAG /IP3 Cascade DAG and IP3 are generated in the p cell by the hydrolysis of phosphoinositols by phospholipase C (PLC) in response to glucose (Prentki & Matschinsky, 1987). In addition to being activated by glucose, PLC is activated by G-proteins coupled to neurohormone receptors in the p cell plasma membrane. IP3 mobilizes [Ca 2 +]i from internal stores; the first evidence for this phenomenon came from studies using permeabilized pancreatic acinar cells (Streb et al., 1983). It has subsequently been shown that IP3 mobilizes [Ca 2 +]/ from an IP3-sensitive calcium pool in the endoplasmic reticulum (Biden et al., 1984; Hellman et al., 1986; Nilsson et al., 1987). In many cell types, including the pancreatic p cell, fluxes of C a 2 + in and out of the endoplasmic reticulum are largely responsible for oscillations of [Ca 2 +]/ (Berridge, 1987; Mertniti et al., 1992). Theoretical studies have suggested that influx of C a 2 + through VDCCs (ie., in glucose-stimulated insulin release) may have an effect on the amplitude and/or the frequency of IP3-generated [Ca 2 +]/ oscillations (Kiezer and De Young, 1993). These investigators suggest that the combined effects of increasing [Ca 2 +]; via influx through VDCCs and from fluxes in and out of the endoplasmic reticulum may play a role in the potentiation of glucose-stimulated insuUn release by several insulin agonists. DAG converts protein kinase C to a Ca 2 +-sensitive kinase and has a role in the translocation of cytosolic PKC to the p cell plasma membrane (Berridge, 1987; Metz, 1988; Zawalich and Rasmussen, 1990). PKC is thought to phosphorylate a subset of p cell proteins which may contribute to insulin secretion in an as yet undefined manner. Although the exact role of PKC in insulin secretion is unclear, the translocation of the kinase to the plasma membrane seems to be an important step in its mechanism of action. THE ENTEROINSULAR AXIS Identification of the Enteroinsular Axis The primary stimulus for insulin secretion is elevated plasma glucose. In 1964 (Mclntyre et al., 1964) reported that despite the lower plasma glucose levels it produced, an intrajejunal infusion of glucose resulted in higher plasma insulin levels than an intravenous infusion of glucose. That year Elrick and colleagues published the results of an elegant study which compared the insulin secretory response to oral and intravenous doses of glucose which yielded the same level of plasma glucose. They reported a mean increase in plasma insulin concentrations in response to both routes; however, the mean increase in plasma insulin was 37% greater in response to oral glucose than in response to the intravenous glucose infusion. These studies suggested 10 that an additional gastrointestinal signal stimulated insulin secretion upon oral ingestion of glucose. The term "enteroinsular axis" was originally proposed by Unger and Eisentraut (1969) to describe this phenomenon. Labarre coined the term "incretin" (Labarre, 1932) to describe the gastrointestinal substance which could stimulate insulin secretion. A l l gastrointestinal hormones have at one time or another been considered as possible incretins. Of these, GIP and GLP-1 are the only peptides which are secreted in response to nutrients in the small intestine and stimulate insulin secretion at physiological concentrations under hyperglycemic conditions. Glucose-Dependent Insulinotropic Polypeptide GIP is a 42 amino acid peptide which is secreted from K-cells in the upper small intestine in humans and from cells distributed in the small upper intestine to ileum in the rat (Brown et al., 1989; Pederson, 1992) in response to glucose, fat, and some amino acids (Cataland et al., 1974; Falko et al., 1975; Pederson et al., 1975; Thomas et al., 1978). It is a member of the glucagon superfamily of peptides, and shares sequence homology with glucagon, vasoactive intestinal peptide (VIP), gastrin releasing factor (GRF), and glucagon-like peptides I and II (GLP-I and -II) (Dockray, 1989). The discovery of GIP arose from attempts to purify CCK from duodenal mucosa extracts. It was shown that a 10% pure preparation of CCK was a more potent inhibitor of pentagastrin-stimulated gastric acid secretion in the dog than a 40% pure preparation (Pederson, 1971). It was hypothesized that an inhibitor of acid secretion had been removed during the CCK purification process. Purification of this inhibitory substance from duodenal extracts by Brown and colleagues (Brown et al., 1970; Brown et al., 1969) resulted in the isolation of Gastric Inhibitory Polypeptide (GIP). The impure CCK extract was also tested for its insulinotropic effects (Rabinovitch and Dupre, 1972). The insulinotropic properties of a 10% pure CCK preparation were removed upon further purification of the preparation. Dupre and colleagues (Dupre et 11 al., 1973) demonstrated a greater increase in circulating insulin in response to infusion of 10 fig GIP/min plus 0.5 g glucose/min compared to glucose alone. The insulfnotropic effect was not observed in fasted individuals. Further evidence that the insulinotropic effects of GIP were glucose-dependent was given by the demonstration that plasma insulin levels did not increase in response to the stimulation of GIP secretion by ingestion of fat (Pederson et al., 1975). Thus, GIP was given the alternate designation, Glucose-dependent Insulinotropic Polypeptide. The insulinotropic effects of GIP have been demonstrated in man (Andersen et al , 1978; Dupre et al., 1973; Elahi et al, 1979), dog (Pederson et al., 1975), rat (Pederson and Brown, 1976; Pederson and Brown, 1978), isolated islet (Verchere, 1991), and purified p-cell preparations (Kieffer et al., 1993). In vivo studies have demonstrated that post-prandial levels of plasma GIP (200-600 pM) stimulate insulin secretion during hyperglycemia in man (Andersen et al., 1978; Dupre et al., 1973) and dog (Pederson et al., 1975). The observation by (Dupre et al., 1973) that GIP did not stimulate insulin release under euglycemic conditions suggested that glucose regulates GIP-stimulated insulin release. Glucose-dependent GIP-stimulated insuMn secretion has been described in vitro in the dog (Pederson et al., 1975), in man (Elahi et al., 1979), and in the perfused rat pancreas (Pederson & Brown, 1976), where the glucose threshold for the insulinotropic effects of GIP was determined to be 5 mM. Additional studies have shown that the ingestion of fat does not lead to insulin release unless plasma glucose is elevated (Brown, 1974; Cleator and Gourlay, 1975; Ross and Dupre, 1978). Recently, Opara and Go (1993) demonstrated that GIP may attenuate insulin secretion under euglycemic conditions. Together, these studies suggest that the glucose-dependent insulinotropic effects of GIP may safeguard against hypoglycemia by preventing inappropriate insulin secretion after a high fat, low carbohydrate meal. 12 The insulinotropic effects of GIP are mediated through specific G protein-coupled receptors identified as part of the secretin/VIP family of G protein-coupled receptors (Cristophe et al., 1986). The rat GIP receptor was cloned by Usdin and colleagues (1993). GIP-specific binding sites have been identified on isolated rat islets (Verchere, 1991), hamster B-cell tumor membrane preparations (Amiranoff et al., 1984; 1985), a tumor-derived mouse B-cell line (Kieffer et al., 1993), and a human insulinoma membrane preparation (Maletti et al., 1984,1987). [ 1 2 5I]GIP bound to cultured rat islets was displaced by 1 nM natural GIP (Verchere, 1991). Both high affinity and low affinity GIP binding sites have been identified on pancreatic p cells. The dissociation constant (K D ) for each type of receptor varies with the preparation studied. High affinity GIP binding sites were reported on transformed B cell membrane preparations (Krj> = 7nM), (Amiranoff et al., 1984; Amiranoff et al., 1985), a hamster p cell tumor membrane preparation (K D = 2 nM), (Maletti et al., 1984), and a human insulinoma membrane preparation (K D = 2.23nM) (Maletti et al., 1987). Low affinity binding sites with dissociation constants of 800 nM, 39 nM, and 8.4 nM, respectively, were identified on the same preparations. GIP binding sites were identified in the pancreas, in the glandular portion of the stomach, throughout the intestine, and in muscle by Whitcomb and colleagues (1984) using [ 1 2 5I]GIP. More recent studies using Northern blots, reverse transcription polymerase chain reaction (PCR), and in situ hybridization have demonstrated GIP receptor mRNA in these loci as well as in the heart, brain, and adrenal cortex (Usdin et al., 1993). Identification of GIP receptor mRNA in these sites suggests that GIP may have previously undescribed effects on these tissues or that a novel peptide may exist which shares binding sites with GIP. Binding of GIP to its receptor was shown to cause a concentration-dependent increase in cAMP in a hamster p cell line (Amiranoff et al., 1984). Subsequent studies in cultured rat islets (Siegel and Creutzfeldt, 1985), human insulinoma cells (Maletti et al., 13 1987), and rat insulinoma cells (Gallwitz et al., 1993) supported the hypothesis that GIP acts via activating G protein-coupled adenylate cyclase. Truncated GLP-1 (tGLP-l(7-36)) was also shown to cause an increase in cAMP in p cells (Druker et al., 1987; Gallwitz et al., 1993; Goke and Cordon, 1988) Both tGLP-l(7-36) and GIP lead to an increase in C a 2 + influx through VDCCs (Lu et al., 1993b; Yada et al., 1993). GIP-stimulated insulin release did not occur in the absence of external calccium ([Ca 2 +] 0) (Wahl et al., 1992). The effects of GIP and tGLP-1(7-36) o n the p cell do not involve hydrolysis of phosphoinositols: neither peptide caused a change in phosphoinositol metabolism in a hamster p cell line (Lu et al., 1993b) and GIP did not increase IP3 in mouse islets (Wahl et al., 1992). GIP-stimulated insulin release was suppressed by inhibitors of phospholipase A2, lipoxygenase, and cyclooxygenase, suggesting that some of the effects of GIP on the p cell may be mediated by these enzymes (Lardinois et al., 1990). GIP was also shown to increase insuhn mRNA levels in a hamster B-cell line (Lu et al., 1993a). In addition to its effects on insulin secretion, GIP also stimulates glucagon secretion. In the presence of low glucose, GIP stimulated glucagon release from the perfused rat pancreas (Pederson & Brown, 1978). The effect could be suppressed by increasing concentrations of glucose. This effect of GIP is important because glucagon is a potent stimulus for insulin secretion and its effects on the pancreatic p cell must be taken into account when interpreting GIP-stimulated insulin secretion from the whole islet. Other Incretins GIP is not the only incretin to affect p cell function. In their study of the relative importance of GIP in the enteroinsular axis, Ebert and colleagues (Ebert et al., 1979) demonstrated that immunoneutralization of endogenous GIP resulted in an initial inhibition of insulin release, followed by the return of normal insulin levels within 20-30 minutes. Further studies showed that GIP antisera (Ebert and Creutzfeldt, 1982; Ebert 14 et al., 1983) did not fully block the incretin effect of GIP. This suggested that other gut factors influence insulin secretion. Gastrin, secretin, CCK, and tGLP-l(7_36) have all been proposed as possible incretins (Brown, 1988; Creutzfeldt, 1979). Perhaps the most important of these is tGLP-1(7-36). In 1987 Kreyman and colleagues (Kreyman et al., 1987) demonstrated the presence of tGLP-l(7-36) in the small intestine in man and reported an increase in circulating tGLP-l(7_36) after oral glucose or a mixed meal in man. tGLP-l(7_36) was shown to have potent insulinotropic effects on the perfused rat pancreas (Mojsov et al., 1987). Like GIP, the insulinotropic effects of tGLP-l(7_36) are glucose dependent (Weir et al., 1989). Simultaneous infusion of physiological concentrations of tGLP-l(7_36) and glucose resulted in an enhanced insulin secretion and reduced peak plasma glucose concentrations. CCK acts on the p cell via a G-protein linked receptor (Rosenzweig et al., 1983; Wank et al., 1992; Williams and McChesney, 1987). Binding of CCK to its receptor causes activation of PLC, resulting in hydrolysis of membrane phospholipids and the production of IP3 and DAG. Effects of Incretin Interaction on Insulin Release It is well documented that simultaneous application of insulintropic agents which act via different second messenger systems results in a potentiated insulin secretory response to increased glucose. This effect has been demonstrated with CCK and GIP on the perfused rat pancreas (Sandberg et al., 1988; Zawalich, 1988) and isolated islet (Zawalich et al., 1989b). The potentiation of insulin secretion by cholinergic agonists and GIP has been shown in the perfused human pancreas (Brunicardi et al., 1990), perfused rat pancreas (Verchere, 1991), and isolated islet (Zawalich et al., 1989b). The addition of cholinergic agonists and CCK to islets resulted in a potentiated insulin response (Zawalich et al., 1989b). Simultaneous application of GIP and tGLP-1(7-36) yielded an additive effect on insulin secretion (Fehmann et al., 15 1989). An additive effect was also seen with the simultaneous application of arginine and GIP to the perfused rat pancreas (Pederson & Brown, 1978). Interactions between arginine and GIP were also demonstrated in man (Elahi et al., 1982) Rasmussen and colleagues (Rasmussen et al., 1990) have suggested that interactions between glucose and other insulinotropic agents occur at the post receptor level, and that glucose acts as a facilitator for the effects of these agents. The fact that these agents are glucose-dependent for insulin secretion and act through common second messenger systems as glucose in the p* cell supports this hypothesis. In addition, the effects of the simultaneous application of stimulants which activate different signaling systems suggests that potentiation of insulin release also occurs at the post receptor level. B CELL HETEROGENEITY IN RESPONSE TO GLUCOSE Heterogeneity among p cells in response to glucose was first reported by Dean and Matthews (1970a) in their work on electrical activity in p cells. These investigators hypothesized that the increase in the number of p cells which fired action potentials as the stimulating glucose concentration was increased indicated that p cells have differing glucose sensitivities. Heterogeneous patterns of insulin release from single p cells in response to glucose were reported by Salomon and Meda (1986) using a reverse hemolytic plaque assay. Differences in the glucose concentration required to stimulate B cells, as measured by observing changes [Ca 2 +]/ with fura-2, also suggest heterogeneity among single rat p cells (Gylfe et al., 1991; Herchulez et al., 1991), auto-fluorescence-activated cell sorted (FACS) rat p cells (Wang et al., 1992), and clusters of rat p cells (Gylfe et al., 1991). Holz and colleagues (Holz et al., 1993) reported heterogeneous patterns of membrane depolarization in single rat p cells in response to glucose. They classified subpopulations of p cells in their preparation as either 16 responsive or non-responsive to glucose. Interestingly, they found that prior exposure of a subset of the non-responsive (3 cells to tGLP-l(7_36) rendered them responsive to a subsequent stimulation with glucose. In their work with p cell clusters, Gylfe and colleagues (1991) demonstrated that p cells which differed in their [Ca 2 +]j response to 11 mM glucose exhibited synchronous oscillations of [Ca 2 +]j at 20 mM glucose. They hypothesized that coordination of individual p cells within a cluster is due to electrical coupling of the cells. It may be possible that electrical coupling of p cells may induce non-responsive cells to respond to glucose. There are several points along the glucose-induced signal cascade leading to insulin release in which post-receptor signaling systems may differ in different p cells. Current evidence points to differences among p cells in the expression of glucokinase (Heimberg et al., 1993; Jetton and Magnuson, 1992) and proinsulin biosynthesis (Schuit et al., 1988). Glucose transport does not appear to contribute to p cell heterogeneity, as GLUT 2 mRNA levels and the rate of glucose transport across the p cell membrane were similar in subpopulations of p cells which differed in their responses to glucose (Heimberg et al., 1993). Indirect evidence which supports the role of glucokinase in p cell heterogeneity comes from work by Holz and coworkers (1993) which shows that sensitization of non-responsive p cells with tGLP-l(7-36) prior to stimulation with glucose renders them glucose-competent. This suggests that non-responsive p cells have an intact adenylate cyclase signaling system, and that heterogeneity in this population may result from the inability of glucose to activate this system. With the exception of the early studies by Dean and Matthews (1970b), p cell heterogeneity has been reported in single p cells or p cell clusters. It is not yet known whether p cell heterogeneity is an inherent feature of pancreatic p cells or a result of the isolation of p cells from the islet micro environment. A goal of the present study was to examine the possibility that heterogeneity is an inherent feature of the pancreatic p cell 17 by measuring the [Ca 2 +], response of individual p cells within the intact islet using microfluorimetry and the fluorescent calcium indicator, fura-2. PROEMIAL SENSITIZATION In their work with heterogeneous populations of single p cells, Holz and colleagues (1993) demonstrated that non-responsive p cells were rendered glucose-competent by a prior exposure to tGLP-l(7-36). This seems to contradict evidence that the effects of incretins such as tGLP-l(7-36) are glucose-dependent (see above). However, many studies have demonstrated that islets can be primed to a subsequent stimulation with glucose by prior exposure to one of several insulinotropic factors, including CCK, ACh, tGLP-l(7-36), and glucose itself. Zawalich, Zawalich, and Rasmussen (1989a) coined the term proemial sensitization (preparing the way) to describe this phenomenon. Grodksy and coworkers were the first to demonstrate that when the rat pancreas is stimulated with two consecutive pulses of glucose, the first pulse of glucose greatly enhances the insulin response to the second pulse (Grodsky et al., 1969). This proemial sensitization of islets to glucose stimulation was later shown to be dependent on glucose metabolism and the generation of second messengers within the p cell, although increased activity of the adenylate cyclase/cAMP second messenger system did not seem to be involved in the sensitization process (Grill et al., 1977). Glucose-induced proemial sensitization was found to persist longer than the time of induction: a five minute pulse of high glucose sensitized the perfused rat pancreas for up to thirty minutes, a ten minute pulse for up to 60 minutes (Grill, 1981). One hypothesis for the physiological significance of glucose-induced proemial sensitization states that the effect may be one of many adaptations of organisms to regulate plasma glucose levels (Grill, 1981). 18 Glucose is not the only agent that can have a sensitizing effect on glucose-stimulated insulin release. Zawalich and Diaz (1987) demonstrated that prior exposure to CCK also sensitizes islets to subsequent stimulation by glucose. Proemial sensitization by CCK increased with both increasing concentrations of the peptide and increasing durations of exposure of the islets to the hormone (Zawalich & Diaz, 1987). As well, the sensitizing effects of CCK were dependent on continued association of CCK with its receptor on the p cell membrane and on glucose metabolism (Zawalich & Diaz, 1987). CCK-induced proemial sensitization was associated with increased phosphoinositol hydrolysis in perifused islets (Zawalich et al., 1987), suggesting the activation of PLC and the generation of DAG and/or IP3 may contribute to this effect. It has also been hypothesized that the sensitization effects of CCK are due to activation of PKC and its translocation to the p cell plasma membrane (Zawalich et al., 1989a). Other insulinotropic agents which act via the PLC signal transduction cascade also have a sensitizing effect on glucose-stimulated insulin secretion. The cholinergic agonists acetylcholine and carbachol were found to sensitize perifused rat islets to subsequent stimulation by glucose, arginine, or tolbutamide (Zawalich et al., 1989a). ACh- and carbachol-induced proemial sensitization were also associated with increased phosphoinositol turnover, suggesting the mechanism through which CCK and cholinergic agonists sensitize islets is similar (Zawalich et al., 1989a). Both CCK and acetylcholine are present in cholinergic nerve fibers which "synapse" on the pancreatic p cell, thus their ability to sensitize islets to subsequent stimulation by glucose and other nutrients may play a significant role in the cephalic stage of insulin secretion (Zawalich et al., 1989a). Although early studies by Grill and coworkers (1977) suggested that adenylate cyclase and cAMP do not play a role in glucose-stimulated proemial sensitization of islets, recent studies suggest that insulinotropic agents which act via the adenylate cyclase/cAMP signal cascade are able to sensitize p cells to subsequent stimulation by 19 glucose (Holz et al., 1993). In these studies, single rat (3 cells which did not exhibit membrane depolarization in response to glucose exhibited pronounced depolarization when tGLP-l (7-36) was applied to the cells prior to glucose. The cAMP second messenger system was found to be necessary for this effect, as cAMP is thought to contribute to the regulation of K A T P channels in the (3 cell membrane (Holz et al., 1993). The results of this study suggest that these effects of tGLP-l(7-36) are mediated by a synergistic effect of glucose and cAMP on the K A T P channel (Holz et al., 1993). These effects of tGLP-l(7-36) were investigated in (3 cells which did not initially respond to glucose. It is possible that the ability of tGLP-l(7-36) to recruit non-responsive p cells is due to the activation of cAMP, much like the proemial sensitization due to CCK and cholinergic agonists is thought to be due to activation of PKC. In light of the evidence of heterogeneity among |3 cells, it would be interesting to determine if tGLP-l(7-36) or other insulinotropic agents which act via cAMP, such as GIP, had the same sensitizing effects on p cells which respond normally to glucose. It has been established that glucose and insulinotropic agents which act via the PLC signal transduction system are able to sensitize islets to subsequent stimulation by glucose. The evidence for insulinotropic agents which act via adenylate cyclase and cAMP is conflicting, although tGLP-l(7-36) was found to render non-responsive p cells glucose-responsive. There is no evidence that GIP, which acts via the same second messenger systems as tGLP-l(7-36), can prime islets to glucose. A goal of this study was to determine if GIP is able to prime islets to subsequent stimulation with glucose. HYPOTHESES AND RATIONALE Insulin is the predominant hormone involved in glucose metabolism and the maintenance of glucose homeostasis. It is secreted from pancreatic p cells primarily in response to glucose, although a number of nutritive, hormonal, and neural factors also 20 stimulate insulin secretion. Despite the importance of glucose in stimulating insulin release, not all (3 cells respond to it in the same way. Heterogeneous p cell responses to glucose have been demonstrated by measuring insulin secretion, membrane potential, and intracellular free Ca 2 +. Although most studies which demonstrate heterogeneity among p cells have been performed in single p cells or clusters of p cells, one study suggests heterogeneity among p cells within the intact islet. We hypothesize that p cell heterogeneity is an inherent characteristic of p cells and not a function of the procedures used to purify p cells from islets. The hormone GIP has been shown to potentiate glucose-stimulated insulin secretion from the perfused pancreas, isolated islet, and purified p cell. This effect of GIP may be due to interaction of signaling systems invoked by GIP (cAMP) and by glucose (cAMP, DAG, IP3) resulting in a substantial increase in [Ca2+],. A number of insulinotropic factors, including CCK, cholinergic agonists, and glucose, sensitize p cells to subsequent stimulation by glucose. tGLP-l(7_36), which acts via the same second messenger system as GIP, has also been shown to have a priming effect on p cells. In a population of single p cells which were non-responsive to glucose, tGLP-l(7_36) given prior to glucose was shown to render these cells "glucose-competent". We hypothesize that GIP also has a priming effect on p cells. Furthermore, we hypothesize that, like tGLP-l(7_36), GIP is able to render non-responsive p cells "glucose competent". To test for heterogeneity among p cells within the intact islet, changes in [Ca 2 +]; of individual p cells in response to incremental changes in glucose were measured using fura-2. In order to measure [Ca 2 +] z from single p cells within the islet, culture conditions which encourage islets to form subconfluent monolayers were developed. Changes in [Ca 2 +] z in response to glucose after a priming dose of GIP were also measured to determine whether GIP is able to prime individual p cells within the intact islet to subsequent stimulation by glucose. 21 During the course of the experiments, several difficulties with attempts to identify heterogeneous populations of p cells within the intact islet were encountered. Due to technical difficulties, changes in [Ca 2 +]; from islets or p cells within the islet were not observed. Unreliable production of subconfluent islet monolayer cultures and the physiological relevance of this preparation were also of concern. Faced with these difficulties in obtaining C a 2 + data and in the interpretation of the physiological relevance of these results, attempts to identify subpopulations of p cells within the whole islet and attempts to test whether GIP is able to render non-responsive p cells "glucose-competent" were abandoned. Instead, the investigation was limited to determining the ability of GIP to prime whole islets to a subsequent glucose stimulation. The final goal of this study was to determine whether GIP is able to prime whole islets to a subsequent stimulation with glucose and to test the latency of this effect. An Endotronics Acusyst/APS-10 cell culture/perifusion system was used to measure the insulin response from whole islets to GIP given prior to glucose. Although it will not provide information on the effects of GIP on subpopulations of p cells within the intact islet, this study will contribute new information on the effects of a priming dose of GIP on glucose-stimulated insulin secretion. Ca2+ glucose Figure 1. Schematic diagram of the second messenger systems involved in glucose-and agonist-induced insulin secretion from the pancreatic p cell. AC: adenylate cyclase. G: G-protein; Gi: inhibitory G-protein; Gs: stimulatory G protein. Pi: activation via phosphorylation R: receptor. 23 CHAPTER 2 The Priming Effects of GIP on Glucose-Stimulated Increases of Intracellular Free Calcium in Individual Pancreatic 0 cells Within the Intact Islet INTRODUCTION The main goals of these experiments were to determine if p cells within the islet are heterogeneous and to examine the effects of a priming dose of GIP on glucose-stimulated increases in [Ca 2 +]; in individual p cells within the intact islet. A methodology was developed which allowed the simultaneous measurement of [Ca 2 +]i from individual p cells within the intact islets and insulin secretion from these islets. We hypothesized that there are subpopulations of p cells within the islet which have different glucose thresholds and that GIP is able to prime p cells to subsequent stimulation by glucose. Difficulties with the C a 2 + imaging, methods and subconfluent islet monolayer cultures, as described below, led to the abandonment of these experiments. An alternate approach was adopted in which isolated islets were perifused to determine the effects of a priming dose of GIP on glucose-stimulated insulin release. MATERIALS AND METHODS Islet Isolation Islets were isolated by collagenase digestion of the pancreas followed by purification in a discontinuous dextran gradient using a modified version of the 24 methods described by Van der Vliet et al. (1988). Hank's Balanced Salt Solution (HBSS) was prepared as a 5x concentrate from powder (Gibco) and supplemented with 18.5 g/1 NaHCC»3 (Sigma), sterile filtered (Falcon Bottle Top Filters, 0.22 micron) into sterile glass bottles and stored at 4°C until use. The concentrate was diluted 1:5 with distilled water as needed, supplemented with 1% BSA (Fraction V, Sigma), and adjusted to pH 7.4 with HC1. Type XI collagenase (Sigma, lot 40H6803) solution was prepared by dissolving collagenase aliquots into an appropriate volume of l x HBSS to achieve a final concentration of 0.25 mg/ml. The discontinuous dextran gradient consisted of layers of 29, 23, and 11% dextran.. A stock solution of 29% dextran was prepared by dissolving 145g dextran T-70 (Pharmacia) in 500 ml lx HBSS. From this, 23% and 11% dextran were prepared by diluting 29% dextran 10:3 and 5:8 with lx HBSS, respectively. HBSS, collagenase, and dextran solutions were kept on ice for the duration of the islet isolation procedure. Islets were isolated from two rats on each experimental day. Male Wistar rats weighing 250-275 g were anaesthetized with an intraperitoneal injection of 60 mg/kg sodium pentobarbital (Somnotol, MTC Pharmaceuticals). A midline incision was made from pubis to xiphisternum and the bowel retracted to expose the pancreas and common bile duct. The sphincter of Oddi was clamped, the common bile duct was cannulated, and the cannula secured with a single ligature. The pancreas was inflated slowly with 15 ml of ice cold collagenase solution. The pancreas was excised, trimmed of excess fat, washed with 5-10 ml HBSS and placed in a 50 ml centrifuge tube (Falcon) on ice until the second pancreas was inflated and excised. The pancreata were combined and digested at 37°C for 14-17 minutes with intermittent vigorous shaking. Digestion was terminated by the addition of ice-cold HBSS to a total volume of 40 ml. Pancreatic tissue was mechanically dispersed by pipetting through a 10 ml serological pipet (Falcon). The tissue was washed three times by centrifugation (4 minutes, 40 x g) 25 in 40 ml ice-cold HBSS. Plastic mesh was used to remove large particles (>1 mm diameter) of undigested tissue and the filtrate was washed once more in 40 ml HBSS. Islets were separated from exocrine tissue and fat by purification in a discontinuous dextran gradient. After the final wash in HBSS, the pellet was suspended in 10 ml 29% dextran. A 4.0 ml layer of 29% dextran was carefully added under this. A layer of 6.0 ml 23% dextran was carefully added above the 29% dextran layer, and a layer of 5.0 ml 11% dextran was added above the 23% dextran layer. The gradient was centrifuged for 5 minutes at 40 x g followed by 10 minutes at 500 x g. Exocrine tissue and undigested pancreatic tissue not removed by mesh filtration formed a pellet in the 29% layer. Islets and some exocrine tissue migrated to the 23% layer, and fat rose to the top of the 11% layer. Using a siliconized pasteur pipet, the 11% and 23% layers were removed from the gradient to a petri dish. The bottoms of the petri dishes were painted black to aid visibility. With the aid of a dissecting microscope, the digested tissue was examined and islets were removed from the dextran solution and placed in a second petri dish filled with 5-8 ml ice cold HBSS. Islets were purified from non-islet tissue in the petri dish by alternately removing non-islet tissue and islet fragments from the dish or transferring single islets to another HBSS-filled dish. Typically, 300-500 islets were isolated from the combined pancreata of two rats. Subconfluent Islet Monolayer Culture Subconfluent islet monolayer cultures were developed to be used in the proposed C a 2 + imaging experiments. This islet culture method was designed to allow the measurement of intracellular C a 2 + from individual A and p cells within the intact islet. Several methods to encourage subconfluent monolayer formation were tested, including mechanical and enzymatic disruption of the islet capsule and overdigestion of the excised, inflated pancreas (please refer to Results). Slight overdigestion of the excised pancreas was the method used to encourage subconfluent monolayer formation. 26 Islets which grew into subconfluent monolayers had a distinctive "fried egg" appearance in which a roughly spherical mass of islet cells (primarily p" cells) was surrounded by a monolayer of cells which had distended and flattened from the islet mass (primarily a cells). Islets were cultured in Dulbecco's Modified Eagle's Medium (DME) (prepared at the Terry Fox Labs, Vancouver, B.C.) containing 4.4 mM glucose and supplemented with 5% FBS (Gibco), 50 TJ/ml penicillin (Gibco), and 50 /ig/ml streptomycin (Gibco) at 37°C in 5% CO2 in air. Isolated islets in HBSS were placed in a 15 ml centrifuge tube (Falcon) and centrifuged 4 minutes at 40 x g. The pellet was resusupended in 2.0 ml DME and washed by centrifugation three times. The final pellet was resuspended in 1.2 - 3.6 ml DME and 200 ul aliquots of the suspension were seeded onto the center of collagen coated 22 mm diameter glass coverslips (Fisher Scientific) which were placed in 6 well tissue culture plates (Costar). The plates were gently transferred to a 37°C, 5% CC»2/air incubator. After 1 hour, 800 ul DME was added to form a dome over the coverslips. The islets were returned to the incubator, and after an additional hour, 2.0 ml DME was added to each well. In this manner, islets were encouraged to adhere to the center of each coverslip. Approximately 12-30 islets were seeded onto each coverslip. Approximately 10 - 20% of the attached islets formed subconfluent monolayers after 3-7 days in culture (Figure 2). Coverslips which had islets adhered in the center with a well-defined monolayer within 10 days after isolation were selected for calcium imaging experiments. This represented approximately 17% of all coverslips. Immunocytochemistry Immunocytochemical techniques were used to identify islet cells in the subconfluent islet monolayer. The islets were double-stained for insulin and glucagon (Figure 3). Diaminobenzedene (DAB, brown) and diaminobenzedene-nickel (DAB-Ni, 27 blue) were used to identify and differentiate between B and A cells, respectively. Islets were first stained for insulin, then for glucagon. Phosphate buffered saline (PBS) supplemented with either 0.3% triton or 0.1% sodium azide was used throughout. Primary antibodies, guinea pig anti-insulin (MRC Regulatory Peptide Group) and mouse anti-glucagon (Gregor, Berlin), were diluted 1:1000 in PBS-triton. Secondary antibodies, biotin-conjugated goat anti-guinea pig and biotin-conjugated horse anti-mouse (Vector), were diluted 1:300 in PBS-triton. The avidin-biotin complex was prepared using a Vectastain ABC Kit. DAB was prepared by adding 200 mg D -glucose (Sigma), 40 mg NH 3C1 (Sigma), 0.03 mg glucose oxidase (Sigma) and 12.5 mg diaminobenzedene to 100 ml 0.5 M Tris (pH 7.6). DAB-Ni was prepared by adding 150 mg NiNH 4S0 4,10 mg DAB, and 7.5 fd H 2 0 2 to 50 ml 0.5 M Tris (pH7.6) Collagen-coated coverslips with adhered islets which had been in culture or had been used in C a 2 + imaging experiments were placed in 2.0 ml Bouin's fixative (75% saturated picric acid, 25% formaldehyde) for 5 minutes. Islets were incubated with 1.0 ml 0.03% H2O2 in methanol for 1 hour at 4°C to remove any endogenous peroxidase activity, washed three times with PBS-azide, then incubated with 1.0 ml primary antibody for 48 hours at 4°C. Islets were washed three times with PBS-azide and incubated with secondary antibody for 2 hours at room temperature. The islets were washed three times with PBS-azide and incubated with 1.0 ml avidin-biotin complex for 1 hour at room temperature. The islets were then washed again and incubated with 1.0 ml DAB (for insulin) or DAB-Ni (for glucagon) 3-15 minutes. Development of the stain was halted by removal of DAB or DAB-Ni and washing the islets with PBS-azide. After staining, the islets were examined under transmitted light with a Ziess Axiovert microscope. 28 Calcium Imaging Calcium imaging of islets was achieved with the Attofluor Imaging System (Carl Ziess Canada). This system allows the simultaneous measurement of [Ca 2 +]j from many cells in a mixed population of cells, such as the subconfluent islet monolayer. After imaging, islets were fixed in Bouin's solution and immunostained as described. Measured changes in [Ca 2 +] z could then be matched to immunostained A and 0 cells. Fura-2 A M (Molecular Probes, Inc., Eugene, Oregon) was dissolved in DMSO to make a 1 mg/ ml solution and this was divided into 20 ul aliquots and stored at -70°C. Fura-2 A M was prepared by slowly adding one 20 ul aliquot of 1M fura-2 A M to 4.0 ml C a 2 + imaging medium while vortexing to give a final concentration of 5 fiM. The 5/iM solution was shielded from light with foil and warmed to 37°C in an incubator. Each 4.0 ml preparation of fura-2 was sufficient to load islets on two coverslips. The C a 2 + imaging medium consisted of Dulbecco's Modified Eagle's Medium (DMEM; Sigma) supplemented with 25 mM HEPES (Sigma), 2 mM 1-glutamine (Sigma), 1 mM sodium pyruvate (Sigma), 2 mM NaHC03 (Sigma), 23 mM sodium chloride (Sigma), and 1% FCS (Gibco). Sufficient D-glucose (Sigma) was added to achieve media with concentrations of 2.0, 4.4, 7.8, 11.0, and 25.0 mM. The pH was adjusted to 7.3 with sodium hydroxide and the medium filtered (Falcon Bottle Top Filters, 0.22 micron) into sterile bottles. GIP (IO'7 M) was prepared from porcine GIP (EG III, MRC Regulatory Peptide Group) by adding 2.0 ml 25 mM glucose calcium imaging medium to a 5 fig aliquot of GIP. The calcium ionophore Br-A23187 was used to generate a maximum calcium response from the islet cells. Br-A23187 was prepared as a 1 mg / ml stock solution in DMSO, divided into 40 fil aliquots, and stored at -70°C until required. The ionophore aliquots were reconstituted as needed in 2.0 ml calcium imaging medium. The final concentration of ionophore used was 36 fiM. Experiments were performed at room temperature (approximately 20°C) except where noted. 29 Coverslips with islets selected for calcium imaging studies were washed with 2.0 ml calcium imaging buffer and incubated with 2.0 ml 5 uM fura-2 A M at 37°C in 5% CO2/ air for 45 minutes. Coverslips with fura-2 loaded islets were placed in a plexiglass laminar flow through chamber such that the coverslip formed the bottom of the chamber (Figure 4, inset). The chamber was inserted into a temperature-regulated aluminum holder and the entire assembly mounted onto the stage of a Ziess Axiovert 135 microscope. A peristaltic pump was used to deliver and remove medium from the chamber at 1 ml/ min. Fractions of the perfusate were collected every 5 minutes over the duration of each experiment and later analyzed for insulin content. A diagram of the equipment configuration is given in Figure 4. The Attofluor Imaging System was used to measure changes in [Ca 2 +]i from individual cells within an intact islet. The system operates by capturing images of the islets within the field of view and measuring the change in the ratio of fluorescence emissions generated by exciting fura-2 alternately with 334 and 380 nm light. Using the system, boxes can be placed over discrete areas of the field of view (for example, over individual cells or over an area of background), and the system records the changes in the intensity of fluorescence emissions from within the boxes. Box size can be adjusted so that only one cell, or part of a cell, is contained within each box. In this way, data can be collected from individual cells within a mixed cell population. After the islets had been loaded with fura-2 and placed on the microscope, an image of the islets was taken, stored by the imaging system, and displayed on a video screen. Using the computer-generated image and a hand-held mouse to manipulate the image on the screen, small boxes, approximately 12 x 12 pixels, were placed over individual cells in the islet monolayer. Larger boxes, approximately 75 x 75 pixels, were placed over the islet mass and over an area of the field of view which did not contain any cells to measure fluorescence emissions from the whole islet and background, respectively. Islets were alternately excited with 334 and 380 nm light and emissions at 30 510 nm from the islets collected every 30 seconds during baseline periods and every 10 seconds while the islets were being stimulated with glucose, GIP, or Br-A23187. The emission fluorescence of fura-2 from each box in response to excitation at 334 and 380 nm was converted to a 334/380 emission ratio by the Attofluor software package. An increase in the emission ratio indicates an increase in [Ca 2 +h concentration. This is due to the inherent fluorescence properties of fura-2. The peak emission fluorescence of fura-2 changes when the dye is bound to C a 2 + (Grynkiewicz et al., 1985). In low calcium solutions, the peak emission fluorescence from fura-2 excited with 334 nm is less intense than the peak obtained when the dye is excited with 380 nm. The converse is true for peak emission fluorescence obtained using high calcium solutions. As calcium concentration increases, emission fluorescence from 380 nm excitation decreases and emission fluorescence from 334 nm increases. Thus, an increase in calcium concentration results in an increase in the 334/ 380 emission ratio. Following calcium imaging, islets were fixed with Bouin's solution and cells within the monolayer identified by immunostaining for islet peptides, as described. The coverslip was then returned to the imaging system and the stained islets examined by transmitted light. Stored islet images were then matched to the immunostained islets (Figure 3). Fluorescence ratio data were retrieved from the computer for each boxed cell which corresponded to a selected immunostained cell. Data Analysis and Statistical Analysis Background fluorescence was subtracted from each data point using software developed and kindly provided by Dr. Khaled Abdel-Hamid. This program subtracts the background fluorescence emissions at 334 and 380 nm from the fluorescence emissions from each box at 334 and 380 nm, and then recalculates the 334/380 ratio. The program is also capable of calculating [Ca 2 +]j concentrations based on the 334/ 380 ratio; however, the Attofluor Imaging System was not calibrated and results are 31 expressed as background-corrected 334/380 emission ratios. Results are therefore qualitative: an increase in the 334/ 380 ratio indicates a relative increase in [Ca 2 +]j, and vice versa. No statistical analysis of the data was performed, as data which could be used to test the hypotheses could not be obtained. RESULTS Subconfluent Monolayer Islet Culture Several methods were used to disrupt the islet capsule, a manoeuver thought to aid in the formation of subconfluent islet monolayers. Disruption of the islet capsule by mechanical methods, such as tearing at the capsule with needle tips under a dissecting microscope or by aspiration of isolated islets through a 28.5 gauge needle, resulted in no difference in the average rate of islet flattening compared to islets which were not manipulated after isolation. Disruption of the islet capsule by enzymatic digestion with either trypsin/ EDTA (0.5 units/ ml trypsin+180 /ig/ ml EDTA, Sigma) or 20 mg/ml type XI collagenase either resulted in complete digestion of the islet into single cells and clusters of cells or in no change to untreated islets. After several failed attempts to find the right combination of enzyme concentration and digestion time, this method was abandoned. Treatment of isolated islets with growth hormone has been reported to decrease the time required for neonatal mouse islets to form subconfluent monolayers (Hans Kofod, Novo Nordisk, Denmark, personal communication). The addition of 1 /ig/ml recombinant bovine growth hormone (Sigma) to islet culture medium resulted in no difference in the rate of islet flattening at 4 and 10 days after isolation. Overdigestion of the excised pancreas during the islet isolation procedure resulted in inconsistent results. In some cases, islets formed monolayers within 3-4 days of isolation; in others, there was no difference in the rate of monolayer formation with controls. 32 Despite inconsistencies observed in the formation of "fried egg" islet cultures using overdigestion of the excised pancreas, this method was used to culture islets for all experiments. Isolated islets formed subconfluent monolayers after 3-7 days in culture (Figure 2). Approximately 10 - 20% of adhered islets formed subconfluent monolayers within this period. Approximately 80% (by visual inspection) of the cells in the subconfluent monolayer were a cells (Figure 3). Measurement of Intracellular Free Calcium Changes in [Ca 2 +]; of individual p cells within the intact islet were measured in order to determine (a) whether there is heterogeneity among individual B cells within the islet and (b) whether GIP is able to prime p cells to subsequent stimulation by glucose. Several protocols were developed to stimulate increases of [Ca 2 +], in islet p cells (Table I). With a few exceptions, changes in [Ca 2 +]; concentrations were not observed in response to stimulants which have been shown to increase [Ca2+]z- in other pancreatic p cells preparations (Grapengiesser et al., 1988; Grapengiesser et al., 1989; Herchulez et al., 1991; Valdeolmillos et al., 1989). Table I. Protocols developed to elicit [Ca 2 +] z responses from individual p cells within the intact islet 1. Protocol Temp. C Q n # islets with responsive cells 2 glucose gradients 20 13 2 1 or 2 min pulses of high glucose alone or + IO"7 M GIP or 20 ug/ ml Br-A21875 20 14 4 sustained high glucose 20 2 1 sustained high glucose 37 3 3 step-wise increases in glucose 37 8 1 1. Refer to text for a more detailed description of protocols 2. "Responsive cells" were those cells which responded to stimuli with an increase in [Ca 2 +]/, as determined by visual inspection of raw data. 33 Glucose Gradients at 20°C It has been demonstrated by others that single p cells exhibit heterogeneous increases in [Ca2+],- in responses to glucose. In order to test whether p cells within the intact islet also exhibit heterogeneous calcium responses to glucose, changes in [Ca 2 +], were measured from isolated subconfluent islet monolayers in response to gradients of glucose. Use of a continuous gradient of from 0 mM to either 11.0 mM or 20 mM glucose provided a method for determining the concentration of glucose to which individual p cells respond. Initially, glucose gradients from 0 mM to 25 mM were generated over a period of 20 minutes by a modification of a previously described methodology (Pederson et al., 1982). Two flasks containing 20.0 ml of either 0 mM or 25.0 mM glucose DMEM were connected in series with a peristaltic pump. The DMEM was drawn from the flasks and delivered to the islets at a rate of 1 ml/ min. The slope of the generated gradient could be varied by altering the volume or glucose concentration of either solution. Individual P cells within the intact islet responded to the 0 mM to 25.0 mM glucose gradient only at higher concentrations of glucose (-20 mM or greater) (Figure 6). A methodology was also developed to deliver glucose gradients to islets within 1 - 2 minutes. This time frame approaches the stimulation times used by (Holz et al., 1993) in their studies demonstrating priming of p cells by GLP-1. An Endotronics APS-10 computer-controlled peristaltic pump and an Endotronics MMCI mixing chamber were used to generate the gradients. The computer was programmed to deliver pulses of 0 mM and either 11.0 mM or 25.0 mM glucose to the mixing chamber such that the media emerging from the chamber consisted of a gradient from 0 mM to either 11.0 mM or 25.0 mM glucose. The entire gradient (from low to high glucose) was generated within 1-2 minutes and delivered to the islets at a rate of 1 ml/ min. p cells within the intact islet did not respond to either gradient generated using this method. 34 1 or 2 Minute Pulses of High Glucose at 20°C With the failure of 1-2 minute glucose gradients to produce changes in [Ca 2 +]; in islet p cells, 1 or 2 minute pulses of 25.0 mM glucose were delivered to the islets in an attempt to elicit a response. Levels of p cell [Ca 2 +]j during islet stimulation were compared to levels during perifusion with 4.4 mM glucose alone. There was no change in the [Ca 2 +], of individual p cells when islets were perifused with 25.0 mM glucose or with 25.0 mM glucose plus I O 7 M GIP. EGTA in the presence of 4.4 or 25.0 mM glucose was administered to the islets in order to determine the effects of removing extracellular calcium on [Ca 2 +]j during high and low glucose. No increase in [Ca2+],' in response to 25.0 mM glucose plus EGTA was expected to be seen; no change from stimulation with 4.4 mM glucose alone was observed. T3r-A23187 was used to elicit a "maximum" calcium response from p cells in both high and low glucose. Under conditions where [Ca 2 +] 0 is high, the ionophore transports Ca 2 +into the cell and protons out of the cell (Pfieffer et al., 1978), Because [Ca 2 +]; levels resulting from the administration of Br-A23187 are dependent on levels of extracellular calcium and not glucose concentration per se, an increase in [Ca 2 +] z was expected when the ionophore was delivered in either 4.4 mM or 25.0 mM glucose. Individual p cells within the intact islet exhibited a transient increase in [Ca 2 +]j in response to ionophore in the presence of 25.0 mM (Figure 7) but not 4.4 mM glucose. Sustained High Glucose at 20°C In the glucose gradient experiments, some p cells had been shown to respond to the gradients at higher glucose concentrations. However, p cells did not respond to high glucose administered in 1-2 minute pulses. A possible explanation for this discrepancy in results was that a 1-2 minute stimulation period is not long enough to elicit an increase in [Ca 2 +]j from p cells within the intact islet. Therefore, 25.0 mM glucose was administered to the islets for 30 minute periods. There was no change in 35 [Ca 2 +] z when islets were stimulated with 25.0 mM glucose for 30 minutes from [Ca 2 +], levels when the islets were stimulated with 4.4 mM glucose. Sustained High Glucose at 37°C A temperature-controlled calcium chamber was made available by Dr. John Church (Dept. of Anatomy, University of British Columbia, Vancouver) to test if temperature had an effect on [Ca 2 +]j levels in islet p cells in response to high glucose. At 37°C, there was an increase of intracellular calcium in response to 25.0 mM glucose which returned to baseline when the islets were returned to 4.4 mM glucose (Figure 8). Oscillating levels of [Ca2+]j- were observed in response to sustained stimulation with 25.0 mM glucose (Figure 9). The oscillations had a period of approximately 5 minutes. Step-Wise Increases to High Glucose After observing increases in [Ca 2 +]j in islets exposed to 25.0 mM glucose at 37°C, lower concentrations of glucose were used to stimulate islets. To determine if individual p cells within the intact islet have different thresholds in their responses to stimulation with glucose, 20 minute pulses of 2.0 mM, 4.4 mM, 7.8 mM, 11.0 mM and 25.0 mM glucose were administered to islets in a stepwise formation. Islet p cells responded to an increase in glucose from 2.0 mM to 4.4 mM; however, further increases in glucose concentration resulted in no change in [Ca 2 +], from levels observed in response to 4.4 mM glucose (Figure 10). 36 Figure 2. Formation of a subconfluent monolayer islet culture of an isolated islet, (a) An islet in culture 4 days after isolation, (b) The same islet, 7 days after isolation. 37 38 Figure 3. Subconfluent monolayer islet cultures immunostained for insulin (brown) and glucagon (blue) (A,f3,C,D). Approximately 80% of the cells in the monolayer are a cells. A 3 9 40 B 41 43 Figure 4. Schematic diagram of the equipment used to measure [Ca 2 +],. A: UV light source. B: excitation filter. C: neutral density filter. D: dichroic mirror. E: mirror F: 40x objective. G: Temperature regulated chamber. H: camera. I: computer housing Attofluor software and hardware. Inset: a: collagen-coated coverslip with adhered islets, b: plexiglass coverslip holder, c: inflow tubing, d: outlfow suction, e: temperature-regulated imaging chamber. 45 Figure 5. A: A subconfluent monolayer islet culture immunostained for insulin (DAB, brown) and glucagon (DAB-Ni, blue). B. Captured image of the same islet using Attofluor software. Numbered boxes are placed over individual cells. Data showing an increase in [Ca 2 +] z in response to elevated glucose from a single p cell (arrowhead, Box #27) is given in Figure 6. 46 '•'•V''. »••''•* f>,#••*•/£/i'Srv * 18 19 • 2 9 3 0 J . \ fia. :". - - • . ' P 3 9 31 • v.::?-;.::-.v L • r—i 4 3 1 • ' 1 1 te. ' U 12-27 • L J ^ J r o j EZJ . ... ^. {A|»l • f'jf Jt'• J", .... ^ • a 47 20mM Cs- 25 mM 1.3 time (min) Figure 6. Change in background-corrected 334/380 fluorescence excitation ratio of a single p cell within an intact islet to a graded increase of glucose from 20 mM to 25 mM (open bar). The immunostained islet and captured image of the islet are shown in Figure 3. 48 time (min) Figure 7. Mean change in background-corrected 334/380 fluorescence excitation ratio of individual islet cells to 25 mM glucose (open bar), 25 mM glucose + 10-7 M GIP (hatched bar), and 25 mM glucose + 20 fig/ml Br-A231875 (shaded bar) (n=40 boxes). 49 1.7 ^ -I 1 — i 1 — 1 0 10 20 30 time (min) Figure 8. Mean change in background corrected 334/380 fluorescence excitation ratio of islet cells caused by changing glucose from 4.4 mM to 25.0 mM (open bar) at 37°C (n=35 boxes). 50 CD CO CO "-^ •=? CO CO T 1 1 1 1 1 1 r 10 15 20 25 30 35 40 45 i r 60 65 70 time (min) Figure 9. Mean change in background corrected 334/380 fluorescence excitation ratio of islet cells to sustained 25.0 mM glucose (open bar) at 37°C (n=27 boxes). Oscillations occurred after the islets were exposed 25.0 mM glucose for approximately 7 minutes and occured with an approximate frequency of 0.2 min"1. Figure 10. Mean change in background corrected 334/380 fluorescence excitation ratio of islet cells to changes in glucose concentration from 2.0 mM glucose to 4.4 mM glucose (open bar) and 25.0 mM glucose (hatched bar) (n=64 boxes). CHAPTER 3 52 The Priming Effects of GIP on Glucose-Stimulated Insulin Secretion from Perifused Rat Pancreatic Islets INTRODUCTION The original goal of this thesis was to determine if p cells within the intact isolated islet differ in their sensitivity to glucose. Attempts to assess p cell heterogeneity by measuring changes in intracellular calcium from individual p cells within the intact isolated islet in response to glucose met with limited success. Attempts to determine if GIP was able to prime p cells to subsequent glucose stimulation by measuring changes in intracellular calcium were also unsuccessful. Faced with these difficulties, attempts to measure intracellular calcium from individual P cells were abandoned and subsequent experiments were limited to investigating the priming effects of GIP on insulin secretion from perifused rat islets. Experimental data presented in this chapter represent the effects of a priming dose of GIP on glucose-stimulated insulin secretion from perifused isolated islets. The effects of increasing the period between GIP and glucose application were also tested. We hypothesized that GIP is able to prime islets to a subsequent stimulation by glucose, and that this priming effect is at least as great as the potentiating effects of GIP on glucose-stimulated insulin secretion. Furthermore, we hypothesized that the priming effects of GIP are dependent on the interval between GIP and glucose administration. MATERIALS AND METHODS Islet Isolation Islets were isolated as described in Chapter 2. Islet Perifusion Islets were perifused to determine the effects of a priming dose of GIP on the insulin secretory response of islets to glucose. The experimental apparatus is diagrammed in Figure 11. Islets were maintained in perifusion chambers and placed in a temperature- and gas-controlled perifusion system (Acusyst-S, Endotronics). A cross-section of the chambers is shown in Figure 11 (insert). The islets were sandwiched between approximately 200 ul layers of cytodex beads within the 500 ul reservoir of the each chamber. A wire mesh and .200 micron filter provided physical support for the cytodex and islets and helped prevent contamination of the islets while in the system. Perifusion chambers were placed in a water reservoir within the Acusyst-S and media was warmed by passing over a water-jacketed heating block prior to reaching the islets. Temperature was regulated by a remote thermistor at 34-37°C at the islets with a water bath (Haake) linked in series to the Acusyst-S. Delivery of 5% CO2 in air was set at 40 ml/ minute, which was shown by us to maintain the pH of the medium at 7.4 when the media flow rate was 6 ml/hour. The medium was gassed and warmed as it passed through gas permeable tubing. Media delivery to the islets was controlled by a microprocessor linked to two peristaltic pumps (APS-10, Endotronics) and 'fine tuned' by adjusting the tension of the tubing over a second set of peristaltic pumps on the Acusyst-S. GIP solutions were injected into the system with a programmable infusion pump (Harvard Apparatus). Fractions of the perfusate were collected at 1 hour, 30 minute, or 10 minute intervals with an LKB fraction collector. Cytodex was prepared by adding 50 ml PBS-CMF (Sigma) to 1 g cytodex 3 (Pharmacia) and allowing the beads to swell overnight. Beads were washed twice by centrifugation (40 x g, 4 minutes) in glucose-free RPMI 1640 (Gibco) medium. The pellet of cytodex from the final wash was suspended in 10 - 15 ml glucose-free RPMI 1640, the slurry divided into 1.0 ml aliquots, autoclaved, and stored at 4°C for later use. RPMI 1640 (Gibco) was supplemented with 0.07% human serum albumin (Sigma), 0.0025% human transferrin (98% pure, iron free, Sigma), 2.5 x 10~8 M sodium selenite (Sigma), 2 x IO*5 M ethanolamine hydrochloride (Sigma), 1% (vol/vol) PBS-CMF (Sigma); sufficient D-glucose (Sigma) was added to achieve 4.4 and 11.0 mM glucose solutions. Synthetic human GIP (Peninsula) was prepared as a 1 x IO - 6 M stock solution in PBS-CMF and stored at -20°C. It was diluted to 1 x 10~7 M in glucose- and supplement-free RPMI 1640. GIP was infused at a rate of 600/d/h into the perifusion system. With the system flow rate at 6 ml/h, the final concentration of GIP at the islets was estimated to be 1 nM. A l l tubing, glassware, and chambers were autoclaved prior to use. Islets were loaded directly into perifusion chambers from FTBSS immediately after being isolated. Four to six chambers were loaded sequentially for each experiment. A l l lines were primed with 4.4 mM glucose RPMI 1640 and bubbles were cleared from the system just prior to loading islets into the chambers. A 10 cc syringe was filled with 2-3 ml glucose- and supplement- free RPMI 1640 and attached to the outflow tubing from the bottom of the perifusion chamber. Inflow tubing was removed from tne top of the chamber and set aside, as was the top of the chamber, the top wire mesh, and the top filter. Medium was withdrawn from the chamber using the syringe until the bottom filter paper could be seen. An aliquot of cytodex was resuspended and 200 ul of the slurry added to the chamber while excess medium was simultaneously withdrawn. This method of withdrawing excess medium with the syringe while adding cytodex or islets served to pack the beads and islets into the chambers and was used each time either was added to the chambers. Approximately 75-80 islets were picked from the isolated islets with a pasteur pipet and added to the chamber. Cytodex slurry was then added until it formed a wet dome over the top of the chamber. The wire mesh, filter paper, and top of the chamber were replaced and medium from the syringe forced through the chamber to remove any air bubbles which might have collected during the loading procedure. When all the bubbles were removed, the inflow tubing was replaced and the syringe removed. After all chambers had been loaded, the flow rate through them was fine tuned to 6 ml/hour by adjusting the tension of the tubing passing over the peristaltic pumps on the Acusyst-S. The APS-10 apparatus was then set to run the experimental program. At the end of each experiment, islets from each chamber were extracted with 3N acetic acid and the insulin content of each extract determined. The inflow tubing was detached from the top of each chamber and the chamber removed from the perifusion system. The top filter paper was removed and placed in a borosilicate glass tube. The chamber was inverted and held over the glass tube. Using a 5 cc syringe, 3 ml 3N acetic acid was injected into the chamber via the outflow tubing and the acid, along with the islets and cytodex, was collected in the tube. The bottom filter paper was placed in the tube and the slurry was boiled for 10 minutes. The extracts were then centrifuged 5 min at 40 x g at 4°C and the supernatant stored at -4°C for later insulin determination. Each chamber of 75-80 islets constituted an n of 1. Insulin determination The amount of immunoreactive insulin (IRI) secreted by islets in response to glucose and GIP was determined by radioimmunoassay (RIA) using a method similar to that developed previously (Albano et a l , 1972). Samples were assayed in duplicate. Insulin assay buffer consisted of 5% charcoal extracted plasma (vol/ vol) and 10% 0.4 M phosphate buffer (vol/ vol) pH 7.4, in distilled water. The buffer was prepared on the day of the assay and kept at 4 °C. Insulin standards were prepared from lyophilized rat insulin (NOVO Research Institute) and serially diluted to obtain concentrations from 160 u\J/ml to 5 /iU/ml. A standard curve was constructed and insulin concentrations of test samples were read from the curve. The standard curve was accurate between 10 and 140 /dJ/ml. Controls were prepared from lyophilized rat insulin (NOVO Research Institute) and diluted to a final concentration of 50 uU/ ml for use in the RIA. An inter-assay control was employed in each assay performed. Control values which deviated more than +10 fdJ/ml from the 50 fdJ/ml value rendered the assay invalid. The mean measured value of control samples was 53.8 ± 0.7 //U/ml (n=80) for the studies included in this thesis. Antisera (guinea pig anti-rat insulin, GP 01) was prepared by the MRC Regulatory Peptide Group (University of British Columbia, Vancouver) and diluted to 1:1x10^ in assay buffer. The antisera had previously been found not to cross react with other known gastrointestinal hormones. Porcine insulin was iodinated with 125J. by the chloramine T method (Hunter and Greenwood, 1963) by Dr. R. A.. Pederson (Department of Physiology, University of British Columbia, Vancouver). 125l-i ab ei ecl insulin was diluted to a final concentration of 1500-2000 cpm/100 ul in assay buffer for use in the RIA. Separation of bound and unbound insulin was achieved by the addition of excess dextran coated charcoal (5 g dextran T-70 (Sigma), 50 g activated charcoal (Sigma), and 1.0 1 of 0.04 M PO4 buffer, pH 7.5) to the assay samples and centrifugation at 40,360 x gravity and 4 °C for 30 minutes (Damon/ IEC centrifuge). The insulin RIA was performed on a refrigerated table at approximately 4 °C. Assays of standards of known concentration of insulin were performed in triplicate; control amounts of insulin and samples of perfusate were assayed in duplicate. Release media, standards, or controls were added as 100 ul samples and incubated with insulin antisera for 24 hours at 4 °C. 1 2 5I-labeled insulin was then added to each tube and the assays were allowed to incubate for a further 24 hours at 4 °C. Bound and unbound insulin were then separated by the addition of dextran-coated charcoal and centrifugation. The supernatant (containing bound insulin) was decanted and gamma emissions from the pellets were counted for 1.0 minute with an LKB gamma counter (Stockholm, Sweden). An RIA software program (RIACalc vl.0, Pharmacia) was used to analyze the data. The amount of bound non-radioactive insulin was estimated for each sample by dividing the difference between sample counts per minute (cpm) and non-specific binding (NSB) cpm by the difference between 0 insulin cpm and NSB cpm. A spline-smoothed, linear standard curve was constructed using data from the standards assayed in the RIA. This curve was displayed as a log/linear curve; however, curve fitting and estimation of the insulin content of each sample was calculated using the linear curve. 57 Data Analysis Insulin secretory responses were normalized by being expressed as a percentage of total islet insulin content (%TIC). The amount of insulin in each sample was divided by the total content of insulin in the extract for the chamber from which the sample came and multiplied by 100%. Insulin secretory responses are expressed as the mean percent change over baseline secretory responses + S.E.M. Baseline secretory responses were determined as the average %TIC secreted during the four hour period immediately prior to the administration of either 11.0 mM glucose or 1 nM GIP. The percent change over baseline was calculated by dividing the %TIC of each data point by the baseline response. Experimental hypotheses were tested by comparing integrated insulin secretory responses. Integrated insulin secretory responses were calculated by adding the insulin secretory response of each chamber from the initiation of glucose stimulation (t=0h) to the end of the 5 hour recovery period (t=5h). In this way, insulin secretory responses to glucose which did not occur strictly within the two-hour glucose stimulation time were included in the data analysis. Integrated responses are expressed as the mean integrated response ± SEM. Statistical Analysis Statistical analyses were performed using one-tailed and two-tailed single-factor ANOVA or Student's t test where appropriate. In all cases, P <. 0.05 was considered statistically significant. PROTOCOL DEVELOPMENT 58 Isolated rat islets were perifused to compare the effects of 11.0 mM glucose, 11.0 mM glucose plus 1 nM GIP, and a priming dose of 1 nM GIP administered immediately prior to 11.0 mM glucose on insulin secretion. Four separate protocols were developed in attempts to test these effects. 2 x 20 Minute Pulse Protocol These experiments were designed to mimic the protocol used to measure the effects of glucose and GIP on [Ca 2 +]j from individual p cells within the intact islet. Islets were isolated and loaded into chambers placed in the Acusyst-S perifusion system. Islets were cultured for 24 hours in 4.4 mM glucose RPMI to allow for recovery from any damage incurred during the isolation procedure. After the 24 hour culture period, islets were stimulated with a 20 minute pulse of 11.0 mM glucose RPMI, followed by 1 hour of 4.4 mM glucose RPMI, followed by a second 20 minute pulse of 11.0 mM glucose RPMI. 1 nM GIP was either infused for 20 minutes simultaneously with the second pulse of high glucose (GIP potentiation studies) or for 20 minutes immediately prior to the second pulse of high glucose (GIP priming studies). The first pulse of glucose was intended to serve as an internal control demonstrating the islet insulin secretory response to glucose alone. Results from these experiments were erratic. Islets in most chambers did not respond to either pulse of glucose or to GIP. Insulin secreted from islets in response to glucose or GIP was minimal. Table II outlines the success rate of the 2 x 20 minutes pulse protocol. 2x2 Hour Pulse Protocol The limited success of the 2 x 20 minute pulse protocols was thought to be due to the short period of time the islets were exposed to 11.0 mM glucose. Examination of results obtained with similar experiments using the Acusyst-S perifusion system (Robert Pauly, 1994, unpublished observations) revealed that islets in this system responded to 11.0 mM glucose after approximately one hour of continuous stimulation. For this reason, the 2 x 20 minute protocol was modified to a 2 x 2 hour pulse protocol. Isolated islets were loaded into chambers in the Acusyst-S perifusion system and cultured in 4.4 mM glucose for 24 hours. After the equilibration period, the islets were stimulated for 2 hours with 11.0 mM glucose, followed by 4.4 mM glucose for 5 hours. This was followed by a second 2 hour pulse of 11.0 mM glucose and 5 hour period of 4.4 mM glucose. In a similar set of experiments testing the priming effects of GIP on glucose-stimulated insulin secretion, 1 nM GIP was infused for 1 hour immediately prior to the second pulse of 11.0 mM glucose. The success rate for these experiments is given in Table II. Again, results from these experiments were erratic and varied considerably between experiments and between chambers within an experiment. Of the islets that did respond to glucose, the insulin secretory response to the second pulse of glucose was observed to be greater than to the first. It was postulated that the first pulse of glucose may sensitize islets to the second pulse, although it occurs five hours before the second. Thus, the 2 x 2 hour pulse protocol was modified to include only one pulse of glucose. Table II. Islet Perifusion Protocols Protocol 1 n # responsive chambers2 Success Rate 2 x 20 minutes 11.0 mM glucose 1 nM GIP potentiation 33 10 30% 1 nM GIP priming 14 3 21% 2x2 hours 11.0 mM glucose glucose control 15 6 40% 1 nM GIP priming 14 2 14% 1. Refer to text for protocol details. 2. Chambers were classified as "responsive" or "non-responsive" by visual inspection of a graph of the insulin secretory responses vs. time. A "responsive" chamber was one whose insulin secretory response was greater than background secretory levels. 1x2 Hour Pulse Protocol Isolated islets were loaded into chambers in the Acusyst-S perifusion apparatus and cultured for 24 hours in 4.4 mM glucose. Islets were stimulated for 2 hours with 11.0 mM glucose followed by a 5 hour period of 4.4 mM glucose. 1 nM or 10 nM GIP was infused either prior to or simultaneously with the 11.0 mM glucose pulse to determine the potentiating or priming effects of the hormone on glucose-stimulated insulin secretion. Islet perifusion data which were presented as figures and analyzed for statistical significance were generated using this protocol. RESULTS The Priming Effects of GIP on Glucose-Stimulated Insulin Secretion Isolated islets were perifused for 24 h in 4.4 mM glucose, followed by 2 h in 11.0 mM glucose in the absence (glucose control) or presence (GIP potentiation) of 1 nM GIP or immediately preceded by a 1 h perifusion with 4.4 mM glucose plus 1 nM GIP (GIP priming) (Figure 12). Islets responded to glucose in all protocols with increased insulin secretion which returned to baseline when glucose was reverted to 4.4 mM. The insulin secretory response to glucose alone, however, was preceeded by an uncharacteristic decrease followed by an increase in insulin secretion. Comparison of the insulin secretory response of perifused islets was achieved using the integrated perfusion data. Perifusion of islets with 1 nM GIP administered either with or prior to 11.0 mM glucose resulted in significant insulin secretion compared to 11.0 mM glucose alone (Figure 13). However, there was no significant difference in the insulin secretory response between GIP potentiation and GIP priming (Figure 13). Latency of the Priming Effects of GIP on Glucose-Stimulated Insulin Secretion The latency effects of a priming dose of GIP on glucose-stimulated insulin secretion were tested by perifusing isolated islets for 1 hour with 4.4 mM glucose and 1 nM GIP administered 0, 20, or 40, minutes prior to perifusion for 2 h with 11.0 mM glucose (Figure 14). Statistical analysis of the integrated responses revealed that there was no difference in the insulin secretory response to glucose when GIP was administered 0, 20, or 40 minutes prior to glucose (Figure 15). Figure 11. Schematic diagram of perifusion equipment. Inset: Schematic diagram islet perifusion chamber. 63 64 time (hours) Figure 12. Insulin secretory response of perifused rat islets to glucose (open bars) and GIP (hatched bars), (a) 11.0 mM glucose (n=16), (b) 11.0 mM glucose plus 1 nM GIP (n=7), and (c) 1 nM GIP administered prior to 11.0 mM glucose (n=9). 65 o e m 5000 4000-J 3000 J 2000 4 1000-J Figure 13. Integrated insulin secretory response of perifused rat islets to (a) 11.0 mM glucose, (b) 11.0 mM glucose plus 1 nM GIP, and (c) 1 nM GIP administered prior to 11.0 mM glucose. *P^ 0.05 compared to (a) 11.0 mM glucose alone. 66 3000 • time (hours) 2500 H 2000 H -4 -3 -2 -1 0 1 2 3 4 5 6 7 time (hours) Figure 14. Latency effects of a priming dose of 1 nM GIP on glucose stimulated insulin secretion from perifused rat islets. GIP (hatched bars) was administered at (a) 0 minutes (n=9), (b) 20 minutes (n=5), and (c) 40 minutes (n=4) prior to 11.0 mM glucose (open bars). 67 15000 10000 A 5000 4 0 min 20 min 40 min Figure 15. Integrated insulin secretory response of perifused rat islets to 1 nM GIP administered 0, 20, or 40 minutes prior to 11.0 mM glucose. 68 C H A P T E R 4 Discussion This thesis investigates the relationship between glucose- and GIP-stimulated insulin secretion from the p cell and pancreatic islet. It is well established that glucose is the most important stimulant for mammalian insulin secretion. Despite its importance, experimental evidence shows that isolated p cells do not all respond to glucose with the same sensitivity, p cell heterogeneity in response to glucose has been demonstrated by measuring insulin secretion (Salomon and Meda, 1986), changes in [Ca 2 +] z (Gylfe et al., 1991; Herchulez et al., 1991; Wang et al., 1992), and membrane potential (Holz et al., 1993) in single pancreatic p cells and p cell clusters. As well, heterogeneity has been shown among p cells within intact islets (Dean and Matthews, 1970). A goal of the present study was to test the hypothesis that the intact islet contains heterogeneous populations of B cells. Glucose-stimulated insulin secretion is potentiated by several gastrointestinal hormones which are released upon ingestion of glucose (incretins). Evidence has accumulated to show that the peptides, GIP and tGLP-l(7-36), are involved as incretins in the enteroinsular axis. The actions of both hormones on the p cell have also been shown to be glucose-dependent. Recent studies on the effects of tGLP-l(7_36) on insulin secretion from subpopulations of p cells suggest that this hormone, when administered prior to glucose, is able to elicit glucose-stimulated insulin secretion from subpopulations of p cells previously unresponsive to glucose. Because GIP and tGLP-1(7-36) act via similar G-protein linked receptors on the p cell and via similar second messenger signaling systems in the p cell, we hypothesized that GIP is also able to prime individual p cells to subsequent stimulation by glucose. A second goal of the present study was to determine the priming effects of GIP on glucose-stimulated changes in [Ca 2 +] 2 from individual p cells within the intact islet. Measurement of changes in [Ca 2 +]; levels of individual p cells within the intact islet was used to test the hypotheses. There is a large body of experimental evidence which shows that glucose stimulates an increase in [Ca 2 +]j in p cells (Grapengiesser et al., 1988; Grapengiesser et al., 1989; Grapengiesser et al., 1992; Gylfe, 1988; Leech et al., 1994; Sakurada et al., 1993; Valdeolmillos et al, 1993; Valdeolmillos et al., 1989; Wang et al., 1992; Wang et al., 1989; Wang et al., 1993; Wang et al., 1991; Yada et al., 1992). As well, GIP has been shown to increase [Ca 2 +]j in p cells in the presence of glucose (Lu et al., 1993; Wahl et al., 1992; Yada et al., 1993). DEVELOPMENT OF SUBCONFLUENT MONOLAYER ISLET CULTURES In order to measure changes in [Ca 2 +], from individual p cells within the islet, culture conditions were developed which encouraged islets to form subconfluent monolayers. The methodology used to encourage subconfluent islet monolayer formation was based on the theory that disruption of the isolated islet capsule aids the formation of subconfluent monolayers. Attempts to mechanically disrupt the capsules of isolated islets via aspiration through a 28.5 gauge needle or by tearing the capsule with needle tips were unsuccessful. Enzymatic disruption of isolated islet capsules with trypsin/ EDTA, a method used in the isolation of p cells from islets (Pipeleers et al., 1985b; Wang et al., 1992), or with 20 mg/ml type XI collagenase resulted in either complete disruption of the islet into single cells and clusters of cells or in no change from controls. After several failed attempts to determine a workable combination of enzyme concentration and digestion time, this methodology was abandoned. Treatment of isolated islets with growth hormone was also tested as a method to encourage subconfluent islet monolayer formation. Treatment of isolated neonatal murine islets with growth hormone has been shown to encourage subconfluent monolayer formation within 14 days of islet isolation (Hans Kofod, Novo Nordisk, Denmark, personal communication). Using recombinant bovine growth hormone, this method did not result in subconfluent monolayer formation within 10 days of islet isolation. Because islets were used for calcium imaging experiments within a maximum of 10 days after isolation, experiments to determine whether growth hormone encourages subconfluent islet formation after 10 days of treatment were not pursued. Disruption of the islet capsule by enzymatic digestion during the islet isolation process was also used to encourage "fried egg" islet formation. Results using this method were inconsistent. In some cases, islets formed subconfluent monolayers within 3-7 days of isolation; in others, there was no difference in the rate of monolayer formation with controls. The discrepancies in the results obtained with this method may be attributed to a number of factors, including differences in the volume of type XI collagenase used to inflate the pancreas, the degree to which the pancreas was inflated, and the period of time the first inflated pancreas was kept on ice while the second was being isolated. Despite the inconsistencies observed using this method, over digestion of the excised pancreas was used to encourage subconfluent islet formation for all the islets used in calcium imaging experiments. Approximately 10% - 20% of islets which adhered to the collagen-coated coverslips formed subconfluent monolayers within 7 days of culture. Roughly 80% of the cells in the monolayer were a cells, as determined by visual inspection of immunostained "fried egg" islets. MEASUREMENT OF INTRACELLULAR FREE CALCIUM FROM INDIVIDUAL B CELLS WITHIN THE INTACT ISLET The hypotheses that the islet contains subpopulations of |3 cells and that GIP is able to prime (3 cells to subsequent stimulation by glucose were to be tested by stimulating islets with glucose and GIP and observing changes in [Ca 2 +]j from p cells within the islet subconfluent monolayer. An Attofluor Imaging System, which allows measurement of [Ca 2 +], from individual cells in a mixed population of cells, was used for these experiments. Despite the capabilities of the equipment, changes in [Ca 2 +], were observed from individual cells on only one occasion. In all other experiments, changes in [Ca 2 +]j from individual cells within the islet monolayer were recorded, but the cells in which these changes occur could not be identified. For this reason, data from these experiments was expressed as the change in 334/380 emission fluorescence ratio from islet cells within the subconfluent monolayer. An approximate identification of islet cells could have been achieved by a box-by-box examination of the data. Because the majority of the cells in the subconfluent monolayer are a cells, most of the boxes should not record increases in [Ca 2 +]; in response to glucose. In theory, only those boxes which lie over p cells should record an increase in [Ca 2 +] r in response to glucose. However, the responses of all the boxes were similar for each separate experiment. As well, a box-by-box examination of the data revealed no difference in the pattern of emission fluorescence from the boxes. This indicates that either all the boxes were placed over cells of the same type (either all a or P cells) or that a and p cells in the subconfluent islet monolayer were responding in an uncharacteristically similar way to stimulation by glucose. It is possible that for some experiments, no boxes were placed over p cells, as boxes are placed over the captured image before islets are immunostained and cells identified as either a or p cells. We were unable to test the hypothesis that the intact islet contains heterogeneous subpopulations of p cells. Islet cells responded to a 20 minute continuous gradient of 0 mM to 25.0 mM glucose only at higher glucose concentrations (20.0 mM to 25.0 mM glucose). These data are consistent with studies by (Jia et al., in press) which demonstrated an insulin response from the perfused pancreas in response to 40 minute glucose gradients of 2.8 mM to 11.0 mM glucose only at higher glucose concentrations (approximately 8.5 mM glucose). Islet cells did not respond to glucose gradients generated within 1-2 minutes. Because of the poor response of the islet cells to glucose gradients, 1-2 minute pulses of high glucose were delivered to the islets in an attempt to elicit a response. This pulse duration approaches the stimulation time used by (Holz et al., 1993) in their studies of the effects of tGLP-l(7_36) on membrane potential in isolated p cells. Islet cells did not respond to 1-2 minute pulses of 25.0 mM glucose, nor did they respond to 1-2 minute pulses of 25.0 mM glucose plus IO - 7 M GIP. These data contradict the large body of evidence which shows that raising glucose levels results in an increase in [Ca2"1"]/. Although some investigators used short stimulation periods (5-7 minutes) (Gylfe, 1988; Gylfe, 1989; Valdeolmillos et al., 1993; Wang et al., 1992) and others used extended stimulation periods (>10 minutes) (Grapengiesser et al., 1989; Grapengiesser et al., 1992; Gylfe et al., 1991), p cells in either protocol responded to the onset of elevated glucose within 3-5 minutes. One possible explanation for the discrepancy between the present data and established patterns of glucose-stimulated increase in [Ca 2 +]j is the very short period of time the islets were exposed to high glucose (1-2 minutes). This short period may not have allowed sufficient stimulation of the p cells within the subconfluent monolayer. In addition, the volume of the fluid over the islets and within the islet imaging chamber may have had a diluting effect on the pulse of high glucose, so that the actual concentration of glucose at the islets was lower than the expected 11.0 mM or 25.0 mM. EGTA was administered to the islets in order to obtain a "minimum" level of [Ca 2 +]j within the islet cells. EGTA is a divalent cation chelator and is used to remove extracellular calcium from perfusate bathing islets or p cells. Because the rise in [Ca 2 +], in response to glucose in p cells is thought to be due to influx of calcium through VDCCs, the addition of EGTA to the 25.0 mM glucose perfusate was expected to result in no increase in intracellular calcium in response to glucose. However, because no increase in [Ca2+],' was observed in response to 25.0 mM glucose alone, the lack of response to EGTA is inconclusive. The calcium ionophore, BR-A23187, was used to elicit a "maximum" increase in [Ca 2 +]j within the islet cells. The ionophore transports C a 2 + across the plasma membrane in exchange for protons. Increases in [Ca 2 +]j in response to BR-A23187 are thus independent of the level of glucose in the medium bathing the islet cells. It was therefore expected that the [Ca 2 +]j response to BR-A23187 in either 4.4 mM or 25.0 mM glucose would be approximately equal. However, islet cells within the intact islet responded to BR-A23187 in the presence of 25.0 mM glucose with a transient rise in [Ca2+]„ but did not respond to the ionophore in 4.4 mM glucose. One possible explanation for this anomalous result could be that the islet cells were responding to glucose and not to the ionophore. The [Ca 2 +]j response of cells to Br-A23187 is dependent not only on the presence of calcium in the extracellular medium, but also on the concentration of ionophore administered to the cells. The lack of calcium response to Br-A23187 plus 4.4 mM glucose may indicate that the concentration of ionophore applied to the cells was insufficient to cause an increase in [Ca 2 +]j. If this were the case, then the calcium signal seen in response to Br-A23187 plus 25.0 mM glucose may in fact be due to stimulation of the islet cells by glucose, not by the ionophore. However, the lack of response to 1 - 2 minute pulses of high glucose in this islet preparation make this explanation unlikely. In addition, the transient nature of the calcium response to Br-A23187 plus 25.0 mM glucose, and the lack of response to ionophore in 4.4 mM glucose, may be due to washout of the ionophore from the islet imaging chamber. During initial experiments, p cells within the intact islet responded to higher glucose concentrations generated with a 20 minute 0 mM - 25.0 mM gradient, and did not respond to 1 - 2 minute pulses of 25.0 mM glucose. This suggested that islet cells require longer stimulus times in order to respond to glucose. Sustained 25.0 mM glucose at 20°C elicited no response from islet cells. However, sustained 25.0 mM glucose at 37°C resulted in an increase in [Ca 2 +], and in some cases, oscillations of [Ca?+]f. The rise in [Ca 2 +]j in response to 25.0 mM glucose at 37°C is consistent with previous reports of the effects of glucose on [Ca 2 +]; (Grapengiesser et al., 1989; Gylfe, 1988). The observed oscillations in [Ca 2 +]j are also consistent with previous reports of calcium oscillations in p cells in response to glucose (Grapengiesser et al., 1989; Grapengiesser et al., 1992; Gylfe et al., 1991; Valdeolmillos et al., 1993). Calcium oscillations in the present study occurred with a period of approximately 5 minutes, which is similar to the oscillations recorded by Grapengiesser and colleagues and Valdeolmillos and colleagues (Grapengiesser et al., 1989; Grapengiesser et al., 1992; Gylfe et al., 1991; Valdeolmillos et al., 1993). Once it had been established that islet cells respond to sustained high glucose at 37°C with an increase in [Ca 2 +] z, attempts were made to test the hypothesis that the intact islet contains subpopulations of p cells. To determine if individual p cells within the islet have different thresholds for glucose, 20 minute pulses of 2.0 mM, 4.4 mM, 7.8 mM, 11.0 mM, and 25.0 mM glucose were administered to the islets in a stepwise manner. Islet cells responded to an increase in glucose from 2.0 mM to 4.4 mM; however, further increases in glucose concentration resulted in no change from [Ca 2 +]i levels in response to 4.4 mM glucose. This result is unusual, because in previous experiments islets responded to an increase in glucose from 4.4 mM to 25.0 mM. In light of the extensive literature reporting changes in [Ca 2 +], from individual p cells using calcium imaging methods similar to those described in this thesis, it is surprising that we were unable to reliably observe changes in [Ca 2 +]j from our islet preparation. Although technical difficulties, including the maintenance of constant fluid levels over the islets and adequate loading of fura-2 within the islet cells, sometimes hampered experiments, the lack of any consistent response from the subconfluent islet preparation suggests the culture itself does not behave in predictable ways. Islets in vivo have a highly ordered morphology which consists of a core of p cells surrounded by a, 6 and PP cells (Orci, 1986). The "fried egg islet" has a quite different morphology: a core of p cells (the egg "yolk") surrounded by a monolayer consisting of a cells interspersed with a few p cells (the egg "white"). The p cells within the monolayer, from which intracellular calcium was being measured, exist in a microenvironment which is very different from that of an typical p cell within the centre of a normal, spherical islet, p cells within the subconfluent monolayer are surrounded by a cells, whereas within the centre of a normal islet they are surrounded by other p cells. Islet structural integrity has been shown to be important in the regulation of msulin release. Comparison of the insulin secretory response of islet cell cultures which consisted of either 90% single p cells, 90% coupled p cells, or 60-65% single p cells enriched with 20-25% a cells shows that coupling among p cells and the presence of a cells resulted in an increased insulin secretory response to 20.0 mM glucose (Pipeleers et al., 1982). This enhanced effect was thought to be due to the number of p cells coupled together and a paracrine effects of glucagon released from a cells. Later studies with FACS-sorted islet cells (Pipeleers et al., 1985a) support these findings and suggest that glucagon acts to increase cAMP production in p cells, thus enhancing the insulin secretory response to glucose. The enhanced secretory capacity of p cell aggregates is thought to be due to cooperation of p cells resulting in a more organized secretory effector unit. Identification of both tight and gap junctions between islet p cells supports this concept (Orci et al , 1975). Isolation of p cells in the subconfluent monolayer results in a loss of p cell-to-cell contact. This "uncoupling" of p cells within the islet may result in the p cell behaving as an isolated p cell, despite its location within the subconfluent monolayer of the islet. In addition, the ratio of A:p cells within the subconfluent monolayer together with the isolation of p cells may result in coupling of A and p cells in the monolayer. Orci (Orci et al., 1975) has shown that A and p cells are coupled with both tight and gap junctions within the intact islet. It is possible that coupling of A and p cells within the subconfluent monolayer has an adverse effect on the responsiveness of (3 cells within the islet monolayer. In summary, of a total of 40 calcium imaging experiments performed, 11 islets had cells within the subconfluent monolayer which responded to high glucose or high glucose plus BR-A23187. No islets responded to GIP. We were unable to test the hypothesis that whole islets contain populations of p cells which vary in their sensitivity to glucose. Because we were unable to determine p cell heterogeneity in the whole islet, and because we were unable to either (a) obtain a reliable glucose response from islet cells or (b) obtain a response to GIP in the presence of high glucose, we did not test the hypothesis that GIP can prime individual p cells to subsequent stimulation by glucose. Therefore, attempts to test the hypotheses by measuring changes in [Ca 2 +]j from individual p cells within the intact islet were abandoned. An alternate approach to testing the hypothesis that GIP can prime p cells to subsequent stimulation by glucose was achieved by measuring insulin secreted from perifused isolated islets. THE PRIMING EFFECTS OF GIP ON GLUCOSE-STIMULATED INSULIN SECRETION FROM PERIFUSED ISLETS The final goal of the present study was to determine if GIP is able to prime islets to subsequent stimulation by glucose. Previous reports have shown that glucose, Ach, and CCK are able to potentiate the insulin secretory response to glucose when given to islets prior to glucose (Grodsky et al., 1969; Zawalich and Diaz, 1987; Zawalich et al., 1989a). As well, Holz and colleagues (1993) have shown through measurement of p cell membrane potential that tGLP-l(7-36) is able to prime p cells to subsequent stimulation by glucose. We hypothesized that GIP is able to prime islets to glucose. We also hypothesized that the priming effects of GIP are dependent on the interval between application of GIP and glucose to the islets. An Endotronics Acusyst/APS-10 cell culture/perifusion system was used to measure the insulin response from whole islets. This system has been previously used in our laboratory to measure insulin secretion from whole islets (Robert Pauly and Patrick Kerridge, unpublished observations). Initial attempts to test the hypotheses were unsuccessful. A 2 x 20 minute pulse protocol was developed to mimic the protocol that achieved successful results when measuring changes of [Ca 2 +] z from subconfluent islet monolayers. The 2 x 20 minute pulse protocol did not result in insulin secretion from isolated islets. The lack of response may be due to the relatively short stimulation period. Although 20 minutes is an adequate stimulation time for observing changes in [Ca2+]„ previous studies in our laboratories have shown that glucose-stimulated insulin secretion from isolated islets in the Acusyst-S/APS-10 does not occur until after approximately 1 hour of stimulation with high glucose (Robert Pauly, unpublished observations). The 2 x 20 minute pulse protocol was therefore altered to a 2 x 2 hour pulse protocol. Results from the 2 x 2 hour pulse protocol were also erratic. The insulin secretory response of the islets was greater to the second pulse of glucose than to the first. This suggests that the first pulse of glucose had a priming effect on the second pulse. In studies of the priming effects of glucose on glucose-stimulated insulin release from the perfused rat pancreas, Grill and colleagues (Grill et al., 1977) found that a 20 to 30 minute pulse of 27.7 mM glucose potentiated the insulin secretory response to a second glucose stimulus, even when the two pulses were separated by 60 minutes of exposure to 3.3 mM glucose. In later studies, (Grill, 1981) showed that the priming effects of a 10 minute pulse of 27.7 mM glucose persisted after 60 minutes but not after a 90 minute interval. These results suggest that the priming effect of glucose is dependent on the duration of the first pulse as well as the duration of the interval between pulses. Although the intervals between pulses of 11.0 mM glucose administered in the present study were 5 hours long, it is possible that the priming effects of a 2 hour stimulation period could persist for a 5 hour period. Thus, the 2 x 2 hour pulse protocol was modified to a 1 x 2 hour pulse protocol. The 1 x 2 hour pulse protocol was used to test the hypothesis that GIP has a priming effect on insulin secretion from isolated islets. The insulin secretory response to 11.0 mM glucose alone, the control experiment, resulted in a pattern of insulin secretion not characteristically observed using the cell perifusion system (Robert Pauly, unpublished observations). A decrease in insulin secretion upon application of 11.0 mM glucose was observed, followed by an increase in insulin secretion. This depression in insulin secretion may have been due to technical difficulties experienced on the particular day of the experiments, and may be alleviated by performing additional experiments. Because of this dip in the insulin secretory response, glucose-stimulated insulin secretion could have been under estimated in the present study. Perifusion of islets with 1 nM GIP administered for 1 hour immediately prior to 11.0 mM glucose resulted in a significant increase in insulin secretion compared to stimulation by glucose alone. There was no significant difference between administering 1 nM GIP either immediately prior or concurrently with 11.0 mM glucose. There was no difference in the insulin secretory response to glucose when the interval between GIP and glucose application was increased from 0 minutes to either 20 or 40 minutes. These results are consistent with previous reports that insulinotropic agents can prime islets to subsequent stimulation by glucose. The priming effects of glucose have been described (see above). The priming effects of CCK-8 have been well documented (Zawalich & Diaz, 1987; Zawalich et al., 1987). When administered to perifused isolated islets for 10 minutes followed by a 10 minute interval, 200 nM CCK has been shown to potentiate both the first and second phases of insulin secretion in response to 7.5 mM glucose (Zawalich & Diaz, 1987). The priming effects of CCK were shown to be dependent on both the dose of CCK-8 administered and the duration of the interval between CCK-8 and glucose application (Zawalich & Diaz, 1987). Priming effects were observed when 10, 25, or 50 nM CCK-8 was used to stimulate the islets. Likewise, the priming effects of a 10 minute 200 nM dose of CCK-8 were observed when the interval between CCK and glucose application was less than 40 minutes; no effect was observed when the interval was extended to 60 minutes (Zawalich & Diaz, 1987). Acetylcholine and the cholinergic agonist carbachol were also found to prime islets to subsequent stimulation by glucose (Zawalich et al., 1989b). These agonists also primed islets to subsequent stimulation by glucose plus 5 nmol/1 CCK-8 and 50 /ig/1 GIP. Like CCK, the priming effects of Ach and carbachol were dependent on the duration of the interval between application of the stimulants and glucose. Although the stimulation times used in the present study are longer than those reported in previous studies, the priming effect of GIP observed is consistent with the reported priming effects of glucose, CCK-8, and cholinergic agonists. An inconsistent finding of the present study is the lack of a time-dependent effect of the priming effects of GIP. This may be due to the extended period of stimulation with priming doses of GIP. In previous studies, priming effects lasting up to 40 minutes could be achieved with as little as a 10 minute stimulation period with the priming agent. In the present study, the extended stimulation period with GIP could have resulted in priming effects which lasted longer than the 20 and 40 minute intervals which were tested. If so, results from the present study are consistent with the theory that islet priming is dependent on the duration of the priming dose (Zawalich & Diaz, 1987). Current theories on the mechanisms of priming point to the generation of second messengers within the p* cell. The priming effects of glucose were found to be dependent on glucose metabolism but not on the generation of cAMP or on the presence of extracellular calcium (Grill et al., 1977). In addition, the priming effects of glucose did not affect de novo insulin biosynthesis (Grill et al., 1977). The priming effects of CCK are also dependent on glucose metabolism (Zawalich & Diaz, 1987). As well, persistent binding of CCK-8 to CCK receptors on the p cell has been shown to be necessary for its priming effects (Zawalich & Diaz, 1987). This suggests that second messengers are needed to translate a message from outside the cell (CCK bound to its receptor) to inside the cell (exocytosis of insulin vesicles). Experimental evidence shows that the priming effects of CCK-8, ACh, and carbachol are correlated with an increase in phosphoinositol turnover which persists long after the removal of these priming agents (Zawalich et al., 1987; Zawalich et al., 1989a). Since phosphoinositol turnover is associated with the production of DAG and IP3, one theory suggests that either one or both of these second messengers may be responsible for the priming effects of CCK, Ach, and carbachol. The generation of second messengers has also been implicated in the priming effects of tGLP-l(7_36) (Holz et al., 1993). The priming effects of tGLP-l(7-36) were dependent on binding of tGLP-l(7_36) to specific tGLP-l(7_36) receptors on the p cell and subsequent generation of cAMP. Furthermore, a cAMP antagonist, Rp-cAMPS, was found to inhibit tGLP-l(7-36) priming, and a cAMP agonist, Sp-cAMPS, potentiated the priming effects of tGLP-l(7_36) (Holz et al., 1993). It has been hypothesized that glucose and tGLP-l(7_36) act synergistically to close K A T P channels, possibly through cAMP-dependent phosphorylation of the channel, suggesting crosstalk between the glucose and tGLP-1(7.36) signaling cascades (Holz et al., 1993). In light of the evidence suggesting that islet priming is due to the generation of second messengers in the p cell, it could be speculated that the priming effects of GIP are due to the generation of cAMP. GIP and tGLP-l(7-36) act via similar G-protein coupled receptors (Cristophe et al., 1986) and both act to stimulate cAMP production by adenylate cyclase, further supporting this hypothesis. Like tGLP-l(7_36), GIP may have an inhibitory effect on the K A T P channel, also possibly via cAMP-dependent phosphorylation of the channel. Another mechanism which may contribute to the priming effects of GIP on the p cell is GIP-stimulated glucagon release from a cells within the islet. GIP has been shown to stimulate glucagon release in low glucose from the perfused rat pancreas (Pederson and Brown, 1978). This effect is attenuated by high glucose. Glucagon has been shown to potentiate glucose-stimulated insulin secretion in man (Samols et al., 1965). It is possible that GIP administered to the islets in 4.4 mM glucose stimulates glucagon secretion from islet a cells, which in turn potentiates insulin secretion from p cells in response to glucose. It is unknown whether glucagon has a priming effect on glucose-stimulated insulin secretion from the islet. Islet priming has been suggested to be an important part of the coordination of neural, hormonal and nutritive signals which regulate insulin release. CCK, which has been shown to both potentiate and prime glucose-stimulated insulin secretion, is not itself secreted in response to glucose. The release of CCK during the cephalic phase of digestion or in response to a mixed meal may sensitize the islets to subsequent postprandial glucose levels (Zawalich et al., 1987). Priming by Ach has been suggested to play a similar role in the regulation of insulin release (Zawalich et al., 1989a). The role of GIP priming in the regulation of insulin release is not yet clear. It is well established that GIP is released in response to elevated plasma glucose. However, there is no clear evidence suggesting that GIP is released during the cephalic phase of digestion. Both sham feeding and direct vagal stimulation do not elicit GIP release (Berthoud et al., 1982; Taylor and Feldman, 1982). In this respect it is difficult to envisage a physiological role for the priming effects of GIP on glucose-stimulated insuUn secretion. However, GIP is released in response to fat (Brown, 1974; Cleator and Gourlay, 1975) and amino acids (Schultz et al., 1982). In situations where fat or amino acids are ingested prior to glucose, GIP priming may play a role in preparing the islet for subsequent glucose stimulation. One therapeutic role for the priming effects of GIP could be in the treatment of NIDDM. Holz and colleagues (1993) have suggested that the priming effects of tGLP-1(7-36), especially in its ability to render glucose-insensitive p cells "glucose competent", would be advantageous in the treatment of NIDDM. Unlike the sulphonylureas, which indiscriminately close K A T P channels, the effects of tGLP-l(7-36) on the K A T P channel are dependent on glucose. This reduces the chances of producing hypoglycemia during attempts to increase insulin release. If the priming effects of GIP are indeed via the same mechanisms as those of tGLP-l(7_36), this hormone could also be used in the treatment of NIDDM. In summary, 1 nM GIP was found to prime perifused islets to stimulation with 11.0 mM glucose. The insulin secretory response of the islets to GIP priming was significantly greater than to glucose alone, but did not differ to 1 nM GIP administered simultaneously with 11.0 mM glucose. No time-dependent effect of priming was observed. FUTURE DIRECTIONS To further investigate the priming effects of GIP on glucose-stimulated insulin secretion from islets, several lines of study may be pursued. First, the experiments described in the present study should be replicated in order to decrease the large expermental error observed in several experiments. In addition to decreasing experimental error, replication would smooth out irregularities in the pattern glucose-stimulated insulin secretion that was observed in the present study. Second, the roles of glucagon and somatostatin in GIP priming of insulin secretion should be determined. Glucagon has been shown to be released from a cells in response to GIP under euglycemic conditions (Pederson and Brown, 1978). Glucagon has also been shown to potentiate glucose-stimulated insulin secretion via a G-protein linked receptor which activates adenylate cyclase (Samols et al., 1965; van Schravendijk et al., 1985). It is possible that the priming effects of GIP may be mediated by glucagon, and are not the result of a direct effect of GIP on the p cell. To test this theory, the priming effects of glucagon on glucose-stimulated insulin secretion should be determined, as well as the effects of including a glucagon antagonist, such as a specific anti-glucagon antibody, on GIP priming of glucose-stimulated insulin secretion. Somatostatin may also affect sensitization of islets by GIP. Somatostatin secretion is stimulated by both GIP and glucose and inhibits insuUn secretion (Brown et al., 1989; Samols et al., 1986). It may therefore have an inhibitory effect on sensitization of islets to glucose by GIP. To test this theory, specific somatostatin antibodies could be used to antagonize somatostatin. This approach has been successful in studies of glucose-induced GIP secretion from a mixed population of transformed GIP- and somatostatin-secreting cells (Kieffer et al., 1993). Third, the mechanism of islet priming of GIP should be elucidated. Current work with the priming effects of tGLP-l(7_36) suggest that this incretin primes islets by activating adenylate cyclase (Holz et al., 1993). Because of the similarities between GIP and tGLP-l(7-36) in their mechanism of action on the p cell, GIP may also prime islets by activating adenylate cyclase. This theory may be tested by determining whether analogues of cAMP, such as Rp-cAMP, forskolin, or IBMX, have a priming effects on islets. The effects of cAMP inhibitors included in the perfusate during priming experiments should also be tested. Finally, the latency effects of GIP priming should be further investigated. In the present study, the priming effects of GIP were found to persist for up to 40 minutes. However, in previous studies, the priming effects of CCK and cholinergic agonists were found to be dependent on the duration of both the period of application of the priming agent and interval between the priming agent and glucose administration (Zawalich and Diaz, 1987; Zawalich et al., 1987; 1989a; 1989b). The relationship between the duration of GIP application and interval between GIP and glucose with respect to islet sensitization should be further examined. CONCLUSIONS The goal of this thesis was to investigate the relationship between glucose- and GIP-stimulated insulin secretion from the pancreatic p cell. Glucose is the most important stimulant for insulin release; however, experimental evidence suggests that not all p cells respond to glucose in the same way. A hypothesis of this study was that P cells within the intact islet were heterogeneous in their response to glucose. The incretin GIP has been shown to play an important role in potentiating glucose-stimulated insulin release from the perfused pancreas, isolated islet, and purified p cell. Several other insulinotropic factors, including tGLP-l(7_36), which is similar to GIP in the way it interacts with and stimulates insulin release from p cells, have been shown to sensitize islets to stimulation by glucose. A second hypothesis of this study was that GIP is able to prime p cells to subsequent stimulation by glucose. Due to technical difficulties, unreliable production of subconfluent islet monolayer cultures, and questionable physiological relevance of these cultures, attempts to test these hypotheses using calcium imaging techniques were abandoned. Instead, the priming effects of GIP on glucose-stimulated insulin secretion from whole islets was investigated. GIP was found to significantly prime islets to a subsequent stimulation by glucose. The priming effects of GIP were found to persist up to 40 minutes after the removal of GIP from the perfusate. Although these studies do not directly demonstrate interaction between glucose- and GIP-induced signalling systems within the p cell, they suggest that the priming effects of GIP are due to crosstalk between the two systems. 85 REFERENCES Ahren, B., Taborsky, G., and Porte, D. J. (1986). REVIEW: Neuropeptidergic versus cholinergic and adrenergic regulation of islet hormone secretion. Diabetologia, 29: 827-36. Albano, J., Ekins, R., Maritz, G., and Turner, R. (1972). A sensitive, precise radioimmunoassay of serun insulin relying on charcoal separation of bound and free hormone moieties. Acta Endocrinologica, 70:487-509. Amiranoff, B., Vauclin-Jacques, N., and Laburthe, M. (1984). Functional GIP receptors in a hamster pancreatic beta cell line, In III: Specific binding and biological effects. Biochemical and Biophysical Research Communications, 123: 671-76. Amiranoff, B., Vauclin-Jacques, N., and Laburthe, M. (1985). Interaction of gastric inhibitory polypeptide (GIP) with the insulin-secreting pancreatic beta-cell line In- I l l : characteristics of GIP binding sites. Life Sciences, 36: 807. Ammala, C., Ashcroft, F., and Rorsman, P. (1993). Calcium-independent potentiation of insulin release by cAMP in single p cells. Nature, 363: 356-58. Andersen, D. K., Elahi, D., Brown, J. C, Tobin, J. D., and Andres, R. (1978). Oral glucose augmentation of insulin secretion: Interactions of gastric inhibitory polypeptide with ambient glucose and insulin levels. The Journal of Clinical Investigation, 62:152-61. Ashcroft, F. M., Ashcroft, S. J. H., and Harrison, D. E. (1988). Properties of single potassium channels modulated by glucose in rat pancreatic 6-cells. Journal of Physiology, 400: 501-527. Atwater, I., Ribalet, B., and Rojas, E. (1978). Cyclic changes in potential and resistance of the p cell membrane induced by glucose in islets of Langerhans from mouse. Journal of Physiology, 278:117-139. Bailey, C, Wilkes, L., Cordon, J., and Armstrong, P. (1989). Effects of gastric inhibitory polypeptide, vasoactive intestinal polypeptide and peptide histidine isoleucine on the secretion of hormones by isolated mouse pancreatic islets. Journal of Endocrinology, 125: 375-79. Banting, F., Best, C, Collip, J., Campbell, W., and Fletcher, A. (1922). Pancreatic extracts in the treatment of diabetes mellitus. Canadian Medical Association Journal, 12:141-46. Barker, C., Nilsson, T., Kirk, C, Michell, R., and Berggren, P.-O. (1994). Simultaneous oscillations of cytoplasmic free C a 2 + concentration and Ins(l,4,5)P3 concentration in mouse pancreatic p cells. Biochemical Journal, 297: 265-68. 86 Berridge, M. (1987). REVIEW: Inositol triphosphate and diacylglycerol: two interacting second messengers. Annual Reviews in Biochemistry, 56:159-93. Berridge, M., and Rapp, P. (1979). REVIEW: A comparative survey of the function, mechanism, and control of cellular oscillators. Journal of Experimental Biology, 81: 217-79. Berthoud, H., Trimble, E., and Moody, A. (1982). Lack of gastric inhibitory polypeptide (GIP) response to vagal stimulation in the rat. Peptides, 3. Biden, T., Prentki, M., Irvine, R., Berridge, M., and Wollheim, C. (1984). Inositol 1,4,5-triphosphate mobilizes intracellular C a 2 + from permeabilized insulin-secreting cells. Biochemical Journal, 223:467-73. Bliss, M. (1982). The discovery of insulin. Chicago: University of Chigcago Press. Brown, J. (1974). Gastric inhibitory polypeptide (GIP). In S. Taylor (Eds.), Endocrinology (pp. 276-84). London: Heinemann. Brown, J., Buchan, A., Mcintosh, C, and Pederson, R. (1989). REVIEW: Gastric Inhibitory Polypeptide. In S. Schultz,G. Makhlouf, & B. Rauner (Eds.), Handbook of Physiology, Section 6, the Gastrointestinal System (pp. 403-430). Bethesda, MD: American Physiological Society. Brown, J., Mutt, V., and Pederson, R. (1970). Further purification of a polypeptide demonstrating enterogastrone activity. Journal of Physiology (London), 209: 57-64. Brown, J., Pederson, R., Jorpes, J., and Mutt, V. (1969). Preparation of a highly active enterogastrone. Canadian Journal of Physiology and Pharmacology, 47:113-114. Brown, J. C. (1988). Enteroinsular axis. Bailliere's Clinical Endicrionology and Metabolism, 2:359-73. Brunicardi, F., Druck, P., Seymour, N., Sun, Y., Elahi, D., and Andersen, D. (1990). Selective neurohormonal interactions in islet cell secretion in the isolated human pancreas. Journal of Surgical Research, 48: 273-78. Cataland, S., Crockett, S., Brown, J., and Mazzaferri, E. (1974). Gastric inhibitory polypeptide (GIP) stimulation by oral glucose in man. The Journal of Clinical Endocrinology and Metabolism, 39: 223-28. Cleator, I., and Gourlay, R. (1975). Release of immunoreactive gastric inhibitory polypeptide (IR-GIP) by oral ingestion of food substances. American Journal of Surgery, 130:128-35. Creutzfeldt, W. (1979). REVIEW: The incretin concept today. Diabetologia, 16: 75-85. Cristophe, J., Svoboda, M., Lambert, M., Waelbroek, M., Winand, J., Dehaye, J., Vandermeers-Piret, M., Vandermeers, A., and Robberecht, P. (1986). Effector mechanisms of peptides of the VIP family. Peptides [suppl. 1], 7:101-7. Dean, P., and Matthews, E. (1970a). Electrical activity in pancreatic islet cells: effect of ions. Journal of Physiology, 210: 265-275. Dean, P., and Matthews, E. (1970b). Glucose-induced electrical activity in pancreatic islet cells. Journal of Physiology, 210: 255-64. Dean, P., Matthews, E., and Sakamoto, Y. (1975). Pancreatic islet cells: effects of monosaccharides, glycolytic intermediates and metabolic inhibitors on membrane potential and electrical activity. Journal of Physiology, 246:459-78. Dockray, G. (1989). Comparative neuroendocrinology of gut peptides. In S. Schultz,G. Makhlouf, & B. Rauner (Eds.), Handbook of Physiology (pp. 133-70). Bethesda: American Physiological Socitey. Dotta, F., and Eisenbarth, G. (1989). Type I diabetes Mellitus: a predictable autoimmune disease with interindividual variation in the rate of p cell destruction. Clinical Immunology, 50: 585-95. Druker, D., Phillipe, J., Mosjov, S., Chick, W., and Habener, J. (1987). Glucagon-like peptide I stimulates insulin gene expression and increases cyclic AMP levels in a rat islet cell line. Proceedings of the National Academy of Science, USA, 84: 3434-38. Dupre, J., Ross, S., Watson, D., and Brown, J. (1973). Stimulation of insulin secretion by gastric inhibitory polypeptide in man. Journal of Clinical Endocrinology and Metabolism, 37: 826-828. Ebert, R., and Creutzfeldt, W. (1982). Influence of gastric inhibitory polypeptide antiserum on glucose-induced insulin secretion in rats. Endocrinology, 111: 1601-06. Ebert, R., Illmer, K., and Creutzfeldt, W. (1979). Release of gastric inhibitory polypeptide (GIP) by intraduodenal acidification in rats and humans and abolishment of the incretin effect by GIP-antiserum in rats. Gastroenterology, 76: 515-23. Ebert, R., Unger, H., and Creutzfeldt, W. (1983). Preservation of incretin activity after removal of gastric inhibitory polypeptide (GIP) from rat gut extracts by immunoadsorption. Diabetologia, 24:449-54. Efendic, S., Kindmark, H., and Berggren, P.-O. (1991). Mechanisms involved in the regulation of the insulin secretory process. Journal of Internal Medicine, 229: 9-22. Elahi, D., Anderson, D., Brown, J., Debas, H., Hershopf, R., Raizes, G., Tobin, J., and Andres, R. (1979). Pancreatic a and p cell responses to GIP infusion in normal man. American Journal of Physiology, 237: E185-91. Elahi, D., Raizes, G., Andres, R., Hershcopf, R., Muller, D., Tobin,J., and Anderson, D. (1982). Interaction of arginine and gastric inhibitory polypeptide on insulin release in man. American Journal of Physiology, 242: E343-51. Falko, J., Crockett, S., Cataland, S., and Mazzaferri, E. (1975). Gastric inhibitory polypeptide (GIP) stimulated by fat ingestion in man. Journal of Clinical Endocrinology and Metabolism, 4 1 : 260. Fehmann, H., Goke, B., Goke, R., Truatmann, M., and Arnold, R. (1989). Synergistic stimulatory effect of glucagon-like peptide-1 (7-36) amide and glucose-dependent insulin-releasing polypeptide on the endocrine rat pancreas. Federation of European Biochemical Society, 252:109-12. Gallwitz, B., Witt, M., Folsch, U., Creutzfeldt, W., and Schmidt, W. (1993). Binding specificity and signal transduction of receptors for glucagon-like peptide-1 (7-36) amide and gastric inhibitory polypeptide on RINm5F insulinoma cells. Journal of Molecular Endocrinology, 10: 259-68. Genuth, S. (1973). Plasma insulin and glucose profiles in normal, obese, and diabetic persons. Annals of Internal Medicine, 79: 812-22. Gillis, K., and Misler, S. (1993). Enhancers of cytosolic cAMP augment depolarization-induced exocytosis from pancreatic B-cells: evidence for effects distal to C a 2 + entry. Pflugers Archives, 424:195-97. Gilon, P., Sheperd, R., and Henquin, J. (1993). Oscillations of secretion driven by oscillations of cytoplasmic C a 2 + as evidences in single pancreatic islets. Journal of Biological Chemistry, 268: 22265-68. Goke, R., and Conlon, J. M. (1988). Receptors for glucagon-like peptide-1 (7-36) amide on rat insulinoma-derived cells. Journal of Endocrinology, 116: 357-62. Grapengiesser, E., Glyfe, E., and Hellman, B. (1988). Glucose-induced oscillations of cytoplasmic C a 2 + in the pancreatic |3 cell. Biochemical and Biophysical Research Communications, 151:1299-1304. Grapengiesser, E., Glyfe, E., and Hellman, B. (1989). Three types of cytoplasmic Ca oscillations in stimulated pancreatic f3 cells. Archives of Biochemistry and Biophysics, 268:404-07. Grapengiesser, E., Gylfe, E., and Hellman, B. (1992). Glucose sensing of individual pancreatic fi-cells involves transitions between steady-state and oscillatory cytoplasmic Ca 2 +. Cell Calcium, 13:219-26. Grill, V. (1981). Time and dose dependencies for priming effect of glucose on insulin secretion. American Journal of Physiology, 240: E24-E31. 89 Grill, V., Adamson, U., and Cerasi, E. (1977). Immediate and time-dependent effects of glucose an insulin release from rat pancreatic tissue. Journal of Clinical Investigation: 1034-43. Grodsky, G., Curry, D., Landahl, H., and Bennet, L. (1969). Further studies on the dynamic aspects of insulin release in vitro with evidence for a two-compartmental storage system. Acta Diabetologia, 6: 554-79. Grynkiewicz, G., Poenie, M., and Tsien, R. (1985). A new generation of C a 2 + indicators with greatly improved fluorescence properties. The Journal of Biological Chemistry, 260:3440-50. Gylfe, E. (1988). Glucose-induced early changes in cytoplasmic calicum of pancreatic (3 cells studied with time-sharing dual-wavelength fluorometry. The Journal of Biological Chemistry, 263:5044-48. Gylfe, E. (1989). Glucose-induced buffering of cytoplasmic Ca2+ in the pancreatic p cell - an artifact or a physiological phenomenon? Biochemical and Biophysical Research Communications, 159: 907-12. Gylfe, E., Grapengiesser, E., and Hellman, B. (1991). Propagation of cytoplasmic Ca 2 + oscillations in clusters of pancreatic 6-cells exposed to glucose. Cell Calcium, 12: 229-240. Heimberg, H., De Vos, A., Vandercammen, A., Van Schaftingen, E., Pipeleers, D., and Schuit, F. (1993). Heterogeneity in glucose sensitivity among pancreatic 6 cells is correlated to differences in glucose phosphorylation rather than glucose transport. The EMBO Journal, 12: 2873-79. Hellman, B., Gylfe, E., and Wesslen, N. (1986). ABSTRACT: Inositol 1,4,5-triphosphate mobilizes glucose-incorporated calcium from pancreatic islets. Biochemistry International, 13:383-89. Herchulez, A., Pochet, R., Pastiles, C, and Van Praet, A. (1991). Heterogeneous changes in [Ca 2 +]. induced by glucose, tolbutamide, and K + in single rat pancreatic |3 cells. Cell Calcium,'12:577-86. Holz, G., and Habener, J. (1992). REVIEW: Signal transduction crosstalk in the endocrine system: pancreatic 6-cells and the glucose competence concept. TIBS, 17: 388-93. Holz, G., Kuhtreiber, W., and Habener, J. (1993). Pancreatic beta-cells are rendered glucose-competent by the insulinotropic hormone glucagon-like peptide-l(7-37). Nature, 361: 362-365. Jetton, T., and Magnuson, M. (1992). Heterogeneous expression of glucokinase among pancreatic p cells. Proceedings of the National Acadamy of Science, USA, 89: 2619-23. 90 Jia, X., Brown, J., Ma, P., Pederson, R., and Mcintosh, C. (in press). The effects of glucose dependent insulinotropic polypeptide and glucagon-like peptide-1 (7-36) on insulin secretion. American Journal of Physiology. Kahn, C, and Rosenthal, A. (1979). Immunologic reations to insulin: insulin allergy, insulin resistance, and the autoimmune syndrome. Diabetes Care, 2: 283-95. Kahn, C, and White, M. (1988). The insulin receptor and the molecular mechanism of insulin action. Journal of Clinical Investigation, 82:1151-56. Kasuga, M., Karlsson, F., and Kahn, C. (1982). Insulin stimulates the phosphorylation of the 95,000 dalton subunit of its own receptor. Science, 215:185-87. Keahey, PL, Rajan, A., Boyd, A. I., and Kunze, D. (1989). Characterization of voltage-dependent Ca2+ channels in beta-cell line. Diabetes, 38:188-93. Kieffer, T. J., Verchere, C. B., Fell, C. D., Huang, Z., Brown, J. C, and Pederson, R. A. (1993). Gastric inhibitory polypeptide (GlP)-stimulated insulin release from a tumor-derived 6-cell line. Canadian Journal of Pharmacology and Physiology, 71: 917-22. Kiezer, J., and De Young, G. (1993). Effect of voltage-gated plasma membrane C a 2 + fluxes on IP3-hnked Ca 2 + oscillations. Cell Calcium, 14: 397-410. Kreyman, B., Williams, G., Ghatei, M. A., and Bloom, S. R. (1987). Glucagon-like peptide-1 (7-36): a physiological incretin in man. Lancet, 2:1300-04. Labarre, J. (1932). Sur les possibilites d'un traitment du diabete par l'incretine. Bulletin Academie Royale Society Medicine Belgium, 12: 620-634. Lardinois, C, Richeson, R. I., Starich, G., Mazzu, D., and Mazzaferri, E. (1990). Gastric inhibitory polypeptide mechanisms of augmenting insulin secretion. Life Sciences, 47: 1015-22. Leech, C, Holz, G. I., and Habener, J. (1994). Voltage-independent calcium channels mediate slow oscillations of cytosolic calcium that are glucose dependent in pancreatic B-cells. Endocrinology, 135: 365-72. Lu, M., Wheeler, M., Leng, X.-H., and Boyd, A. I. (1993a). Stimulation of Insulin secretion and insulin gene expression by gastric inhibitory polypeptide. Clinical Research41: 201. Lu, M., Wheeler, M. B., Leng, X.-H., and Boyd III, A. E. (1993b). The role of the free cytosolic calcium level in £-cell signal transduction by gastric inhibitory polypeptide and glucagon-like peptide I (7-37). Endocrinology, 132: 94-100. Maletti, M., Altman, J.-J., Hui Bon Hoa, D., Carlquist, M., and Rosselin, G. (1987). Evidence of functional gastric inhibitory polypeptide (GIP) receptors in human insulinoma: binding of synthetic human GIP 1-31 an activation of adenylate cyclase. Diabetes, 36:1336-40. Maletti, M., Portha, B., Carlquist, M., Kergoat, M., Laburthe, M., Marie, J. C., and Rosselin, G. (1984). Evidence for and characterization of specific high affinity binding sites for the gastric inhibitory polypeptide in pancreatic 6-cells. Endocrinology, 115: 1324-31. Mclntyre, N., Holdsworth, C., and Turner, D. (1964). New interpretation of oral glucose tolerance. The Lancet, i i : 20-21. Meissner, H., and Schmelz, H. (1974). Membrane potential of p cells in pancreatic islets. Pflugers Archivs, 351:195-206. Menniti, F., Bird, G., Glennon, M., Obie, J., Rossier, M., and Putney, J. J. (1992). Inositol phosphates and calcium signaling. Molecular and Cellular Neurosciences, 3:1-10. Metz, S. (1988). Membrane phospholipid turnover as an intermediary step in insulin secretion: putative roles of phospholipases in cell signalling. The American Journal of Medicine, 85: 9-21. Micheals, R., and Sheridan, J. (1981). Islets of Langerhans: Dye coupling among immunocytochemically distinct cell types. Science, 214: 801-802. Minkowski, O. (1989). Historical development of the theory of pancreatic diabetes. Diabetes, 38:1-6. Mojsov, S., Weir, G., and Habener, J. (1987). Insulinotropin: glucagon-like peptide (7-37) co-encoded in the glucagon gene is a potent stimulator of insulin release in the perfused rat pancreas. Journal of Clinical Invesigation, 9: 616-19. Nilsson, T., Arkhammar, P., Hallberg, A., Hellman, B., and Berggren, P.-O. (1987). Characterization of the inositol 1,4,5-triphosphate-induced C a 2 + release in pancreatic p cells. Biochemical Journal, 248: 329-36. Opara, E., and Go, V. (1993). Dual effects of gastric inhibitory polypeptide on insulin secretion. Pancreas, 8: 39-43. Orci, L. (1986). The insulin cell: Its cellular environment and how it processes (pro)Insulin. Diabetes /Metabolism Reviews, 2: 71-106. Orci, L., Malaisse-Lagae, F., Ravazzola, M., Rouiller, D., Renold, A., Perrelet, A., and Unger, R. (1975). A morphological basis for intercellular communication between A and P cells in the endocrine pancreas. Journal of Clinical Investigation, 56:1066-1070. Pederson, R. (1992). Gastric Inhibitory Polypeptide. In Gut Peptides: Biochemistry and Physiology. Walsh, J. H., and Dockray, G.J., Eds. Raven Press, New York: 217-59 92 Pederson, R., and Brown, J. (1976). The insulinotropic actions of gastric inhibitory polypeptide in the perfused rat pancreas. Endocrinology, 99: 780-785. Pederson, R., and Brown, J. (1978). Interaction of gastric inhibitory polypeptide, glucose, and arginine on insulin and glucagon secretion from the perfused rat pancreas. Endocrinology, 103: 610-15. Pederson, R., Buchan, A., Zahedi-Asl, S., Chan, C, and Brown, J. (1982). Effect of jejunoileal bypass in the rat on the enteroinsular axis. Regulatory Peptides, 5: 53-63. Pederson, R., Schubert, H., and Brown, J. (1975). Gastric inhibitory polypeptide: Its physiologic release and insulinotropic action in the dog. Diabetes, 18: 733-738. Pipeleers, D., In't Veld, P., Maes, E., and Van de Winkel, M. (1982). Glucose-induced insulin release depends on functional cooperation between islet cells. Proceedings of the National Academy of Science, 79: 7322-25. Pipeleers, D., Schuit, F., Veld, P., Maes, E., Hooghe-Peters, E., Van De Winkel, M., and Gepts, W. (1985a). Interplay of nutrients and hormones in the regulation of insulin release. Endocrinology, 117: 824-832. Pipeleers, D., Veld, P., Van De Winkel, M., Maes, E., Schuit, F., and Gepts, W. (1985b). A new in vitro model for the study of pancreatic a and p cells. Endocrinology, 117: 806-815. Polonski, K., and Rubenstein, A. (1986). Current approaches to measurement of insulin secretion. Diabetes and Metabolism Reviews, 1: 23-26. Prentki, M., and Matschinsky, F. (1987). Ca 2 +' cAMP, and phospholipid-derived messengers in coupling mechanisms of insulin secretion. Physiological Reviews, 67: 1185-1248. Purrello, F., Buscema, M., Rabuazzo, A., Caltabiano, V., Forte, F., Vinci, C, Vetri, M., and Vigneri, R. (1993). Glucose modulates glucose transporter affinity, glucokinase activity, and secretory response in rat pancreatic Q cells. Diabetes, 42:199-205. Rabinovitch, A., and Dupre, J. (1972). Insulinotropic and glucagonotropic activities of crude preparation of cholesystokinin-pancreozymin. Clinical Research, 20: 945. Rasmussen, H., Zawalich, K., Ganesan, S., Calle, R., and Zawalich, W. (1990). Physiology and pathophysiology of insulin secretion. Diabetes Care, 13: 655-666. Rorsman, P., and Trube, G. (1986). Calcium and delayed potassium currents in mouse pancreatic C-cells under voltage clamp conditions. Journal of Physiology, 374: 531-550. Rosen, O. (1987). After insulin binds. Science, 237:1452-58. Rosenzweig, S., Miller, L., and Jamieson, J. (1983). Identification and localization of cholecystokinin-binding sites on rat pancreatic plasma membranes and acinar cells: a biochemical and autoradiographic study. Journal of Cell Biology, 96:1288-97. Ross, S., and Dupre, J. (1978). Effects of ingestion of triglyceride or galactose on secretion of gastric inhibitory polypeptide and on responses to intravenous glucose in normal and diabetic subjects. Diabetes, 27: 327-33. Sakurada, M., Kanatsuka, A., Saitoh, T., Makino, H., Yamamura, I., Miyazaki, J.-L, Kikuchi, M., and Yoshida, S. (1993). Relation between glucose-stimulated insulin secretion and intracellular calcium accumulation studied with a superfusion sustem of a glucose-responsive pancreatic C-cell line MLN6. Endocrinology, 132: 2659-65. Salomon, D., and Meda, P. (1986). Heterogeneity and contact-dependent regulation of hormone secretion by individual 13 cells. Experimental Cell Research, 162: 507-20. Samols, E., Marri, C, and Marks, V. (1965). Promotion of insulin secretion by glucagon. The Lancet: 415-16. Sandberg, E., Ahren, B., Tendler, D., and Efendic, S. (1988). Cholecystokinin-33 potentiates and vasoactive intestinal peptide inhibits gastric inhibitory polypeptide-induced insulin secretion in the perfused rat pancreas. Acta Endocrinologia, 117: 545-51. Schade, D., Santiago, J., Skyler, J., and Rizza, R. (1983). Intensive Insulin Therapy. Amsterdam: Medical Examination Publishing Co. Schuit, F., In't Veld, P., and Pipeleers, D. (1988). Glucose stimulates proinsulin biosynthesis by a dose-dependent recruitment of pancreatic beta cells. Proceedings of the National Academy of Science, USA, 85: 3865-69. Schultz, T., Jorde, R., Burhol, P., and Christensen, O. (1982). Augmented release of gastric inhibitory polypeptide into the portal vein in response to intraduodenal glucose and amino acids in anesthetized rats treated with methylprednisolone of alloxan. Scandinavian Journal of Gastroenterology, 17: 357-62. Siegel, E., and Creutzfeldt, W. (1985). Stimulation of insulin release in isolated rat islets by GIP in physiological concentrations and its relation to islet cyclic AMP content. Diabetologia, 28:857-61. Streb, Irvine, Berridge, M., and Schultz, I. (1983). Release of C a 2 + from a nonmitochondrial intracellular store in pancreatic acinar cells by inositol-1,4,5-trisphosphate. Nature, 306: 67-69. Taylor, I., and Feldman, M. (1982). Effect of cephalic-vagal stimulation on insulin, gastric inhibitory polypeptide, and pancreatic polypeptide release in humans. Journal of Clinical Endocrinology and Metabolism, 55:1114-17. Thomas, F., Sinar, D., Mazzaferri, E., Cataland, S., Mekhjian, H., Caldwell, J., and Fromkes, J. (1978). Selective release of gastric inhibitory polypeptide by intraduodenal amino acid perfusion in man. Gastroenterology, 74:1261. Ullrich, A., and al., e. (1985). Human insulin receptor and its relationship to the tyrosine kinase family of oncogenes. Nature, 313: 756-61. Unger, R., and Eisentraut, A. (1969). Entero-insular axis. Archives of Internal Medicine, 123: 261-266. Usdin, T., Mezey, E., Button, D., Brownstein, M., and Bonner, T. (1993). Gastric inhibitory polypeptide receptor, a member of the secretin-vasoactive intestinal peptide receptor family, is widely distributed in peripheral organs and the brain. Endocrinology, 133: 2861-70. Valdeolmillos, M., Nadal, A., Soria, B., and Garcia-Sancho, J. (1993). Fluorescence digital image analysis of glucose-induced [Ca2+] oscillations in mouse pancreatic islets of Langerhans. Diabetes, 42:1210-14. Valdeolmillos, M., Santos, R., Contreras, D., Soria, B., and Rosario, L. (1989). Glucose-induced oscillations of intracellular C a 2 + concentration resembling bursting electrical activity on single mouse islets of Langerhans. FEBS, 259:19-23. Van der Vliet, J., Meloche, R., Field, M., Chen, D., Kaufman, D., and Sutherland, D. (1988). Pancreatic islet isolation in rats with ductal collagenase distension, stationary digestion, and dextran separation. Transplantation, 45: 493-95. Verchere, C. (1991) The role of Gastric Inhibitory Polypeptide in the regulation of pancreatic endocrine secretion. Ph.D Thesis, University of British Columbia. Wahl, M. A., Plehn, R. J., Landsbeck, E. A., Verspohl, E. J., and Amnion, H. P. T. (1992). Are ionic fluxes of pancreatic beta cells a target for gastric inhibitory polypeptide? Molecular and Cellular Endocrinology, 90:117-23. Wang, J., Baimbridge, K., and Brown, J. (1992). Glucose- and acetylcholine-induced increase in intracellular free Ca in subpopulations of individual rat pancreatic fi cells. Endocrinology, 131. Wang, J., Baimbridge, K., and Leung, P. (1989). Changes in cytosolic free calcium ion concentrations in individual rat granulosa cells: Effect of leutinizing hormone-relaeasing hormone. Endocrinology, 124:1912-17. Wang, J., Corbett, J., Marshall, C, and McDaniel, M. (1993). Glucose-induced insulin secretion from purified C-cells: A role for modulation of Ca2+ influx by cAMP and protein kinase C dependent signal transduction pathways. The Journal of Biological Chemistry, 268: 7785-91. Wang, X., Treistman, S., and Lemos, J. (1991). Direct identification of individual vasopressin-containing nerve terminal of the rat neurohypophysis after 'whole cell' patch-clamp recordings. Neuroscience Letters, 124:125-28. Wank, S., Pisegna, J., and deWeerth, A. (1992). Brain and gastrointestinal cholecystokinin receptor family: structure and functional expression. Proceedings of the National Academy of Sciences, USA, 89: 8691-95. Weir, G., Leahy, J., and Bonner-Weir, S. (1986). Experimental reduction of p* cell mass: implication for the pathogenesis of diabetes. Diabetes and Metabolism Reviews, 2:125-61. Weir, G. C, Mosjov, S., Hendrik, G. K., and Habener, J. F. (1989). Glucagon-like peptide-1 (7-37) actions on endocrine pancreas. Diabetes, 38: 338-42. Whitcomb, D., O'Dorisio, T., Cataland, S., and Nishikawara, M. (1984). In vivo binding of insuUn to target tissues using a new in vivo radioreceptor assay. Dig. Dis. Sci., 29: 95 S. Williams, J., and McChesney, D. (1987). Cholecystokinin induces the interaction of its receptor with a guanine nucleotide binding protein. Regulatory Peptides, 18:109-17. Yada, T., Itoh, K., and Nakata, M. (1993). Glucagon-like peptide-l-(7-36) amide an a rise in cyclic adenosine 3',5'-monophosphate increase cytosolic free C a 2 + in rat pancreatic p cells by enhancing Ca 2 + channel activity. Endocrinology, 133:1685-92. Yada, T., Kakei, M., and Tanaka, H. (1992). Single pancreatic p cells from normal rats exhibit an initial decrease and subsequent increase in cytosolic free C a 2 + in response to glucose. Cell Calcium, 13: 69-78. Zawalich, W. (1988). Synergistic impact of cholecystokinin and gastric inhibitory polypeptide on the regulation of insulin secretion. Metabolism, 37: 778-81. Zawalich, W., and Diaz, V. (1987). Prior choloecystokinin exposure sensitizes Islets of Langerhans to glucose stimulation. Diabetes, 36:118-22. Zawalich, W., Diaz, V., and Zawalich, K. (1987). Cholecystokinin-induced alterations in B-cell activity: Duration, specificity, and involvement of phosphoinositide metabolism. Diabetes, 36:1420-24. Zawalich, W., and Rasmussen, H. (1990). REVIEW: Control of insulin secretion: a model involving C a 2 + , cAMP, and diacylglycerol. Molecular and Cellular Endocrinology, 70:119-37. Zawalich, W., Zawalich, K., and Rasmussen, H. (1989a). Cholinergic agonists prime the B-cell to glucose stimulation. Endocrinology, 125: 2400-06. Zawalich, W., Zawalich, K., and Rasmussen, H. (1989b). Interactions between cholinergic agonists and enteric factors in the regulation of insulin secretion from isolated perfused rat islets. Acta Endocrinologica, 120: 702-07. 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0086872/manifest

Comment

Related Items