Citation
Characterization and physiological significance of brain insulin receptors

Material Information

Title:
Characterization and physiological significance of brain insulin receptors
Creator:
Boyd, Frederick Tilghman, 1959-
Publication Date:
Language:
English
Physical Description:
vii, 149 leaves : ill. ; 29 cm.

Subjects

Subjects / Keywords:
Astrocytes ( jstor )
Brain ( jstor )
Cells ( jstor )
Cultured cells ( jstor )
Insulin ( jstor )
Neuroglia ( jstor )
Neurons ( jstor )
Norepinephrine ( jstor )
Rats ( jstor )
Receptors ( jstor )
Binding Sites ( mesh )
Brain Chemistry ( mesh )
Dissertations, Academic -- Physiology -- UF ( mesh )
Physiology thesis Ph.D ( mesh )
Receptor, Insulin -- physiology ( mesh )
Genre:
bibliography ( marcgt )
non-fiction ( marcgt )

Notes

Thesis:
Thesis (Ph.D.)--University of Florida, 1985.
Bibliography:
Bibliography: leaves 139-148.
General Note:
Typescript.
General Note:
Vita.
Statement of Responsibility:
by Frederick Tilghman Boyd, Jr.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
The University of Florida George A. Smathers Libraries respect the intellectual property rights of others and do not claim any copyright interest in this item. This item may be protected by copyright but is made available here under a claim of fair use (17 U.S.C. §107) for non-profit research and educational purposes. Users of this work have responsibility for determining copyright status prior to reusing, publishing or reproducing this item for purposes other than what is allowed by fair use or other copyright exemptions. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder. The Smathers Libraries would like to learn more about this item and invite individuals or organizations to contact the RDS coordinator (ufdissertations@uflib.ufl.edu) with any additional information they can provide.
Resource Identifier:
021252667 ( ALEPH )
16865407 ( OCLC )

Downloads

This item has the following downloads:


Full Text

PAGE 1

CHARACTERIZATION AND PHYSIOLOGICAL SIGNIFICANCE OF BRAIN INSULIN RECEPTORS By Frederick Tilghman Boyd, Jr. A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF IX)CTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 1985

PAGE 2

To Marisa, Mom and Dad

PAGE 3

ACKNOWLEDGEMENTS I would like to thank my major adviser and mentor, Dr. Mohan K. Raizada, for his guidance, support, patience, impatience and inspiration. He has taught me the excitement of discovery and the beauty of science. I would also like to thank the members of my supervisory conunittee individually and collectively for their insights and guidance over the course of this project. Thanks to Dr. Ed Meyer for sharing some of his experimental magic and seeing the significance of points that were, at times, beyond my line of sight. Thanks to Dr. Colin Sumners for paving the way for some of these experiments and showing me how to do things I could not make work. I wish to extend my appreciation to Dr. M. Ian Phillips for his ongoing support and encouragement during my graduate career. My sincere thanks to Dr. Sidney Cassin for his criticisms of my research based on long experience, they have come home to roost in some interesting ways. My thanks to these gentlemen for providing me with a rich and exciting education. Thanks also go to Dr. Michael Kappy, a former member of my conunittee, and to Dr. Derrell Clarke for stimulating collaborations and the generous use of their facilities. I wish to express my appreciation to Viviana Perez, Esen Momol, Lynn Holmes and Muriel Turlington for their hard work and tolerance of my methods. i i i

PAGE 4

TABLE OF CONTENTS PAGE ACKNOW!JEOOEMENTS iii ABSTRACT Vi CHAPTER / INTRODUCTION ........... 1 Insulin and the Brain .................................. 1 ~he Insulin Receptor ................................... 3 / Kinetics of Insulin Binding 5 Regulation of Insulin Receptors 6 The Role of Glycosylation and Protein Synthesis in Insulin Receptor Expression 8 Recognized Actions of Insulin in the Brain 9 I I METHODS ....... 15 Animals and Solutions 15 Neuron-Enriched Brain Cell Cultures 16 Glial-Enriched Brain Cell Cultures 18 Fibroblast Culty25s::: Measurement of I-Insulin Binding 27 Neuronal Cultures .................................. 27 Glial and Fibroblast Cultures . 28 Light Microscopic Autoradiography 28 Measurement of Protein Synthesis and Glycosylation 29 Measurement of Monoamine Uptake 29 III KINETIC AND AUTORADIOGRAPHIC CHARACTERIZATION OF NEURONAL AND GLIAL INSULIN RECEPTORS ................ 34 Experimental Procedures 35 Time Courses of Association and Dissociation 35 Specificity ........................................ 35 Saturability and Affinity 36 Light Microscopic Autoradiography 36 Results .............................................. 36 Neuronal Insulin Binding 36 Glial Insulin Binding ............................. 46 /iscussion 53 iv

PAGE 5

IV REGULATION OF NEURONAL AND GLIAL INSULIN RECEPTOR S BY INSULIN AND INHIBITORS OF GLYCOSYLATION AND PROTEIN SYNTHESIS . 62 Results ................................. 63 Effect of Insulin on Neuronal and Astrocyte Glial Insulin Binding ... 63 Effect of Tunicamycin on Neuronal and Astrocyte Glial Insulin Binding . 70 Effect of Cycloheximide on Neuronal and Astrocyte Glial Insulin Binding . 79 Discussion ........................................... 99 V MODULATION OF NEURONAL MONOAMINE UPTAKE BY INSULIN . 103 Experimental Procedures . 105 Results ............................................. 105 Discussion .......................................... 125 VI CONCLUSIONS ................ 128 APPENDIX ..................... 13 4 BIBLic:x:;RAPHY . 13 9 Bl(X;RAPHICAL SKETCH ............. 149 V

PAGE 6

Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy CHARACTERIZATION AND PHYSIOLOGICAL SIGNIFICANCE OF BRAIN INSULIN RECEPTORS By Frederick Tilghman Boyd, Jr. Chairman: Mohan K. Raizada Major Department: Physiology August, 1985 The brain has been considered an insulin-independent organ because of the limited penetration of insulin into the central nervous system (CNS). However, brain insulin and insulin binding sites and behavioral effects of centrally administered insulin have recently been reported. Differential cell culture techniques were used to determine which cell types from the CNS possess specific insulin receptors and what the actions of insulin might be on those cell types. It was determined that both cultured neurons and astrocyte glia possess specific, high affinity insulin receptors. Scatchard analysis of competition-inhibition experiments yielded biphasic plots for both neurons and astrocyte glia. Astrocyte glial cells exhibited no negative cooperativity as dissociation of bound labeled insulin proceeded at the same rate whether or not unlabeled insulin was present. vi

PAGE 7

Neuronal and astrocyte glial insulin receptors were found to be regulated in distinctly different manners. Neuronal cells in culture were able to change their regulation of insulin binding in response to insulin under different culture conditions. In one culture condition insulin induced up regulation of insulin binding and in another insulin induced down regulation. Inhibition of glycosylation or protein synthesis had no effect on neuronal insulin binding. Astrocyte glial cells consistently exhibited down regulation of insulin binding in response to insulin with decreased binding being due to fewer high affinity receptors on the cell membrane. Inhibition of glycosylation reduced glial insulin binding and inhibition of protein synthesis increased insulin binding in manners typical of the responses of peripheral target tissues of insulin. These findings suggest that neuronal and glial insulin receptors may be structurally distinct and have different life cycles within the cell. The effect of insulin on neuronal monoamine uptake was studied to determine if insulin directly affects neuronal function. Insulin was capable of inhibiting more than 95% of specific norepinephrine uptake and was shown to be 600% more potent an inhibitor than the specific pharmacological norepinephrine uptake inhibitor, maprotiline. In addition, insulin stimulated serotonin uptake and had no effect on sodiuminsensitive uptake. These findings demonstrate that insulin directly affects neuronal function and support the conclusion that insulin is a neuromodulator in the CNS. vii

PAGE 8

CHAPTER I INTRODUCTION Insulin and the Brain Insulin was first isolated from the dog pancreas and used in the treatment of diabetes mellitus by Banting and Best in 1922 (1). Within six years Schmidt (2) reported beneficial effect s of insulin, co-administered with carbohydrates to prevent hypoglycemic shock, in the treatment of psychoses. In 1935 Sakel (3) reported successful treatment of psychotic patients with insulin (hypoglycemic) shock therapy. This became the treatment of choice for schizophrenia, depression and psychosis until the advent of electroconvulsive shock therapy and the development of various psychotropic drugs (4). Thus insulin has an early history as a psychoactive drug. However, insulin's role as one of the primary hormones of glucose homeostasis has dictated that insulin research be focused on its actions in classical peripheral target tissues such as liver, muscle and adipose tissue. Blood glucose levels are acutely regulated by insulin via its stimulatory actions on liver gluconeogenesis and glucose uptake and metabolism in muscle and adipose tissue (5). Glucose supply to the brain must be 1

PAGE 9

2 maintained at adequate levels because the mature brain is a glucose obligate organ. Its sole energy substrate under normal circumstances is glucose and its metabolism is almost entirely aerobic (6). This is emphasized by clinical evidence that the prognosis of patients with acute brain ischemia is influenced by the level of carbohydrate present in the brain (7). When ischemia occurs in patients with high brain lactate levels, a lesion is produced with astrocyte glial and endothelial necrosis as well as neuronal death and cerebral edema. At lower lactate levels tissue damage is limited to selective neuronal damag e ; astrocytes and endothelial cells are spared and no cerebral edema develops. Apparently astrocytes can function to defend brain tissue against the effects of anoxia if sufficient glucose is available. However, in vivo studies looking at the effect of insulin on brain glucose uptake or metabolism have consistently found no effect of insulin (8-11). As Morgan in Best and Taylor' classic text characterized the issue of insulin actions in the brain, "(brain) cells are notable for their lack of responsiveness to insulin"(l2, p. 79). Despite the classical notion of the brain as an insulin insensitive organ, Havrankova et al. (13) reported concentrations of insulin in the brain averaging 25 times the concentration of insulin in the blood. They demonstrated concentrations greater than 55 ng per gram wet tissue in the hypothalamus and olfactory bulb. This finding, along with their report of insulin receptors with similar distributions in the central nervous system (14),

PAGE 10

3 suggested that an endogenous brain insulin system might exist. Considerable attention has been devoted to determining the source of brain insulin, that is whether insulin is concentrated by the brain from the blood or synthesized in the brain. It has been determined that in vivo brain insulin concentrations are independent of plasma insulin and that insulin is present in neurons cultured in the absence of insulin (15-17). Insulin content in cultured neurons is reduced by inhibition of protein synthesis (18) and recent studies have isolated an mRNA species in neuronal cultures which hybridizes to cDNA probes for insulin mRNAs (19), demonstrating that insulin is synthesized in neuronal cells. These findings necessitate the reevaluation of the role of insulin in the brain. These present studies have examined the responsiveness of neuronal cells to insulin and report here those findings. The Insulin Receptor The first step in the action of insulin in all tissues, including the brain, is its interaction with cell surface membrane receptors. Receptors are proteins which fulfill certain criteria which distinguish them from "binding sites." Binding to receptors displays kinetics of binding consistent with biological effects associated with the peptide, is saturable, temperature and pH dependent and specific for certain peptides. After interaction with the peptide message, receptors initiate specific actions which are characteristic of the peptide's biological effect.

PAGE 11

4 Insulin receptors have been isolated from many different peripheral tissue types and characterized physically as well as kinetically and biologically. The insulin receptor is very similar among all peripheral tissues studied. Among the tissue types where insulin receptors have been identified are adipocytes (20), lymphocytes (21), fibroblasts (22-23), monocytes (22), erythrocytes (24), hepatocytes (25), liver cells (26), heart cells (27), and brain cells (14, 28). Isolated receptor complexes aggregate into clusters of from 300,000d to 3,000,000d. The single unit of the insulin receptor seems to have a molecular weight of 300,000d and is composed of two subunits, linked by disulfide bonds, called the a and S subunits. The 300,000d receptor is thought to exist in the membrane as (as)2 that is with 2 a subunits and 2 S subunits (29-31). The a subunit has a molecular weight of 135,000d and can be labeled with 125I-labeled insulin by various methods and so is thought to be the binding subunit of the receptor. The S subunit has a molecular weight of 95,000d and is not labeled by 125I-insulin (29-32), although it seems to be associated with the kinase activity of the insulin receptor and thus is thought to be responsible for the initiation of the second messenger for insulin (33). The insulin receptor kinase associated with the S subunit is an autophosphorylase which stimulates phosphorylation primarily of tyrosine residues of the S subunit, although some stimulation of phosphoserine and possibly phosphothreonine occurs as well (34).

PAGE 12

5 Both the aand S subunits are glycoproteins and are exposed to the external cell surface. Both subunits are adsorbed to lectin affinity columns and tunicamycin, an antibiotic which inhibits glycosylation, also inhibits insulin receptor expression (35-36). In addition, cell proteins biosynthetically labeled with carbohydrate are precipitated with anti-insulin receptor antibodies (37). Kinetics of Insulin Binding Binding properties of insulin to the insulin receptor are similar among the various peripheral tissues. Association of insulin with its receptor is rapid and consistently reaches equilibrium binding within two hours at room temperature. Binding is reversible, dilution of the labeled ligand or replacement of the binding ligand with an excess of unlabeled ligand results in reduction of specific binding. Typically more than 75% of cell-associated label is displaceable within two hours at room temperature. The remaining cell-associated ligand is interpreted as representing labeled ligand which has been internalized within the cell and thus cannot be dissociated. Insulin binding to all peripheral tissues is saturable, which means there are limited numbers of insulin receptors on any cell. Insulin from various animal species and peptides structurally related to insulin compete for porcine insulin binding (the standard assay condition) to extents related to their biological potency (26). Unrelated peptides do not compete for insulin binding, in fact there are no insulin antagonists

PAGE 13

6 other than insulin receptor antibodies. Thus, the insulin receptor is specific for insulin and fulfills all the requirements of a peptide receptor. Regulation of Insulin Receptors Since the initiation of insulin's cellular actions is dependent on its interaction with insulin receptors on the cell surface, regulation of insulin receptor numbers is critical to determining the cell's responsiveness to insulin. Reduction of insulin binding was first observed in animal models of obesity, the obese/obese (ob/ob) and the diabetic/diabetic (db/db) mice (38, 39). Human studies also showed that obese subjects (162-352% of ideal body weight) had markedly reduced insulin binding to circulating lymphocytes (40). Furthermore, insulin binding could be normalized with dietary restriction. It was proposed that changes in insulin binding in obese subjects was a result of high concentrations of circulating insulin and that insulin itself could reduce insulin binding (41). This process is known as "down regulation" and has been confirmed to occur in many tissues including cultured human lymphocytes (42), rat adipocytes, (43), rat hepatocytes (44), human fibroblasts (45), hepatoma cells (46), human IM-9 lymphocytes (47), and in vivo (48). Down regulation of the insulin receptor has been demonstrated to involve a reduction in the number of cell surface insulin receptors via increased degradation and/or internalization of the insulin receptor-insulin complex (42, 45, 47, 49). Down regulation of insulin receptors has been

PAGE 14

7 demonstrated in all peripheral tissues studied with the exceptions of cultured chondrosarcoma chondrocytes and possibly 3T3-Ll lymphocytes (50-51), two transformed or transforming cell types. Thus insulin clearly regulates its own peripheral receptors with high concentrations of circulating insulin resulting in reduced cell surface insulin receptor numbers. Other physiological signals have also been demonstrated to regulate insulin receptor expression. In particular, thyroid hormone and glucocorticoids have been demonstrated to increase insulin receptor numbers (up regulation) (52-55). Concanavalin A has been shown to reduce insulin binding of frog erythrocytes by lowering the affinity of the high affinity receptor (56). Changes in intracellular cAMP can induce down regulation of insulin receptors independently of cell surface signals interacting with insulin receptors (57), and adenosine itself has been shown to alter cellular sensitivity to insulin by a postreceptor mechanism (58). These studies clearly show that regulation of cell surface insulin binding is an important physiological mechanism.

PAGE 15

8 The Role of Glycosylation and Protein Synthesis in Insulin Receptor Expression The insulin receptor is a glycoprotein. Glycosylation is an important step in the insertion of newly synthesized recepto/ s or latent receptors in an intracellular pool into the cell membrane. Tunicamycin is an antibiotic which inhibits N-acetyl glycosylation of nascent peptides (59). Treatment of peripheral tissues with tunicamycin has been demonstrated to result in a 70%-90% loss of insulin binding (60-63). This effect is due to a decrease in the number of insulin receptors on the cell surface in response to tunicamycin treatment. However, Kadle (39) has demonstrated that intracellular pools of insulin receptors are not reduced to as great an extent. This suggests that tunicamycin has effects on several processes involved in the expression of the fibroblastic insulin receptor, namely, required glycosylation of the insulin receptor itself, and glycosylation of proteins which are required for the surface expression of intracellular receptors. Inhibition of protein synthesis has variable effects on insulin receptor expression depending on the studied tissue. Incubation of various tissues with cycloheximide, a specific inhibitor of protein synthesis has increased insulin binding (mouse fibroblasts, 39); had no effect (human fibroblasts, 49); and decreased binding (3T3-Ll fatty fibroblasts, 59).

PAGE 16

9 Recognized Actions of Insulin in the Brain Most investigators have approached the question of the role of insulin in the brain as "how does the action of insulin in the periphery affect the brain?" Many researchers have focused on the effect of insulin on blood concentrations of amino acids which are precursors to various neurotransmitters. Either insulin or glucose administration decreases blood concentrations of free tryptophan (60) (the amino acid precursor of serotonin), tyrosine (61) and phenylalanine (62) (the amino acid precursors of the catecholamines). Insulin and glucose increase brain concentrations of these amino acids (61, 63). The!se findings and others led to the hypothesis that insulin stimulates the transport of these amino acids into the brain (61). An alternative explanation to these findings has been that the stimulatory effect of insulin on peripheral uptake of the neutral amino acids which compete with tryptophan, tyrosine and phenylalanine for the brain transport system reduces competition for the brain uptake mechanism, thereby resulting in increased brain levels of these amino acids (64-65). Since the first step in synthesis of serotonin and catecholamines is rate limiting and normally unsaturated, the brain concentrations of these amino acid precursors influences the rate of synthesis of these transmitters and their concentrations in the brain (66-67). Both serotonin and catecholamine synthesis are influenced by meal content and the resultant insulin, glucose and amino acid concentrations in the blood (68-71). In conclusion,

PAGE 17

10 insulin affects brain content of serotonin and catecholamines by altering the rate of uptake of their precursors into the brain, either by stimulating uptake directly or by altering the relative concentrations of neutral amino acids competing for a common uptake system. Many investigators have suggested that insulin interacts specifically with neurons of the hypothalamus and circumventricular organs to signal the brain about the current metabolic state of the body (72). Several lines of evidence suggest that insulin acts in the ventromedial hypothalamus to control feeding. The electrical activity of neurons in the hypothalamus is altered by injection of low quantities of insulin (73). Autoradiography with systemically injected 1251-insulin revealed localization of insulin binding sites on neuronal processes within the arcuate nucleus/median eminence (74). Infusion of insulin into the hypothalamus suppresses feeding in rats and baboons (75-76). Lesions of the ventromedial hypothalamus result in hypersecretion of pancreatic insulin (77). Established neural connections from the insulin receptive neurons in the hypothalamus and insulin concentrating neurons in the area postrema to the vagal nuclei may provide information regarding the insulin state of the animal to these parasympathetic centers (78). These centers, in turn, regulate insulin secretion by the pancreas, thus providing a feedback control loop for insulin secretion (79).

PAGE 18

11 Lesions of the ventromedial hypothalamus, in addition to resulting in insulin hypersecretion, also result in an increase in the duration of sleep (80). This finding has spawned research into possible brain actions of insulin in regulating sleep. Administration of insulin into the lateral ventricles increased the daily duration of slow wave sleep with no effect on paradoxical (REM) sleep (81). Intravenous or intraventricular administration of anti-insulin antibodies caused a decrease in daily slow wave sleep duration. Neutralization of endogenous insulin resulted in more frequent waking from slow wave sleep. Paradoxical sleep was unchanged by any of these treatments. In addition to interactions of insulin with the brain in the context of classical metabolic roles of insulin, several studies have suggested that insulin acts directly as a neuromodulator within the central nervous system. Barbaccia et al. (82) suggested a role for insulin as a neuromodulator interacting with the dopamine system active in olfactory bulbs. Sauter et al. (83) reported that insulin at high doses stimulates release of dopamine, norepinephrine and epinephrine from hypothalamic slices. Saller and Chiodo (84) have demonstrated changes in firing rates of nigrostriatal dopamine neurons by insulin and Palovcik et al. (85) have demonstrated changes in firing rates of hippocampal neurons by insulin. These studies suggest that insulin may have direct effects on neural function.

PAGE 19

12 In conclusion, insulin is recognized to have many actions within the brain, but little synthesis has occurred to explain the significance of brain insulin and brain insulin binding sites. In addition, studies of the role of insulin in the brain have been complicated by the use of mixed populations of cells. Researchers have assumed that the brain is a homogeneous tissue with respect to the effects of insulin studied. Thus, contrasting effects of insulin on distinct cell types in the central nervous system might go unnoticed. In fact, the brain is a heterogeneous population of diverse cell types which have distinct functions within the central nervous system and can be regulated differentially by different factors. Although neuronal cells are the primary excitable cells in the brain, there are glial cells which serve important metabolic functions as well as modulating neurotransmission by various mechanisms. Glial cells serve to support and compartmentalize the neuronal network within the brain. Radial glia are important structures leading to the migration of specific neurons and the development of certain neuronal circuits. Astrocyte glial cells are sources of Nerve Growth Factor (NGF), which may serve as an important trophic factor in brain development. Thus glial cells appear to be important in the development of specific neural connections. In addition to a role in development, astrocyte glial cells are morphologically suited to control local concentrations of metabolic substances and neurotransmitters. Astrocytes have a

PAGE 20

13 large number of processes which serve to compartmentalize the brain. These processes tend to terminate with enlargements on blood vessels or around synaptic elements of neurons (86). Astrocytes contain significant stores of glycogen which is depleted from astrocytes during anoxia. It has been demonstrated that glycogen is broken down in response to elevated cAMP levels within the cell and glial cells have been demonstrated to respond to catecholamines via B receptors to stimulate cAMP accumulation. It is conceivable, then, that neurons can signal astrocyte glial cells to liberate pyruvate or glucose from stores of glycogen i n emergency situations. In addition, astrocytes have been demonstrated to modulate neurotransmitter metabolism. In particular, astrocyte glial cells have been shown to actively take up y-aminoisobutyric acid (GABA) in several preparations (87, 88, 89). Astrocytes can also synthesize and release GABA in response to potassium-evoked depolarization (90, 91). Schrier and Thompson (91) suggest that astrocytic glioma cells either secrete or take up GABA to maintain external GABA concentration around 0.7 uM. The interactions of neurons and glia have been extensively studied and reviewed comprehensively by Varon and Somjen (92). Glial cells clearly have significant roles in the control of local concentrations of important metabolic and neuroactive substances and need to be considered when actions of putative neuroactive substances are studied.

PAGE 21

14 In these studies, differential culture techniques were developed and utilized to investigate the differential responsiveness to insulin of specific cell types from the central nervous system. This method allows study of responses of a single cell type without the interferences of other cells and other variables. Specifically, these studies were designed to determine which cell types from the central nervous system possessed specific insulin receptors, how those receptors were regulated and what effect insulin exerts via high affinity specific insulin receptors. These studies were carried out in the context of an ongoing project in the laboratories of Drs. Raizada, Kappy and Clarke elucidating the role of insulin as an important metabolic and neuroactive substance in the brain.

PAGE 22

CHAPTER II METHODS Differential culture techniques were developed to selectively grow neurons or astrocyte glial cells from the rat brain. Generally accepted methods were adapted to tissue culture to determine experimental parameters. Animals and Solutions One-day-old Sprague-Dawley rats were obtained from our breeding stock. Dulbecco's Modified Eagle's Medium (DMEM), Temin's Modification of Eagle's Medium (TMEM), fetal bovine serum (FBS) and horse serum (HS) were purchased from GIBCO, Grand Island, New York. Plasma derived horse serum (PHS) was from HyClone Laboratories, Logan, Utah. Deoxyribonuclease I and lx crystallized trypsin (190 U/mg) were from Worthington Biochemicals, Freehold, New Jersey. Bovine serum albumin, Fraction V (BSA), poly-L-lysine (MW 150,000), cytosine arabinoside, 2-Deoxy D-glucose, HEPES (N-2 hydroxyethylpiperazineN1-2 ethane-sulfonic acid) and tunicamycin were from Sigma Chemicals, St. Louis, Missouri. Crystalline porcine insulin, guinea pig insulin, proinsulin and glucagon were gifts of Lilly 1 5

PAGE 23

16 Research Laboratories, Indianapolis, Indiana. Chicken insulin, human growth hormone and desoctapeptide insulin were gifts of the NIH and NIADDK pituitary hormone distribution program. Purified IGF-II was kindly provided by Dr. Peter Nissley of NIH. 2-(G-3H) Deoxy D-glucose (specific activity=8 Ci/mmole), D(6-3H-(N)) glucosamine hydrochloride (specific activity 19.0 Ci/mmol), ( 3 H} valine (specific activity 56.8 Ci/mmol), (-)-(7-3H(N)}norepinephrine (specific activity 16.9-24.0 Ci/mmol), (-)-(3 H} serotonin (specific activity 16.9-24.0 Ci/mmol) and (-)-(3 H} dopamine (specific activity 16.9-24.0 Ci/mmol) were obtained from 125 New England Nuclear, Boston, Massachussets. The I was purchased as labeled NaI from the Amersham Corp., Arlington Heights, Illinois. Porcine insulin was iodinated by the method of Freychet et al., (93) and separated from NaI on coarse grade cellulose (Whatman) as described by Lesniak (94) using dilute chloramine T. The 125I-insulin produced was 90-96% trichloroacetic acid (TCA) precipitable and had a specific activity of 150-180 uCi/ug. Iodinated insulin obtained by this method was used for insulin binding experiments without further purification of individual insulin isomers. 125 B 26-labelled I-porcine insulin was kindly provided by Dr. Bruce Frank of Eli Lilly Laboratories, Indianapolis, Indiana. Neuron-Enriched Brain Cell Cultures Primary neuron-enriched cultures were prepared as described previously (16) from brains of one-day-old rats. Brains were

PAGE 24

17 removed from the cranium at the leve l of the medulla oblongata and placed in isotonic salt solution (Appendix, Solution D). All pia mater and blood vessels were carefully rem o ved from the brains, which were then washed twice with isotonic salt solutjon and minced to about 1-to 2-mm3 pieces. The minced tjssue was subjected to trypsin and deoxyribonuclease treatments to dissociate single cells (Appendix, Trypsin and Deoxyribonuclease). Dissociated cells were suspended in Dulbecco's Modified Eagles Medium (Appendix, DMEM) containing 10% fetal bovine serum (FBS) and sedimented at 500 g for 10 min. After washing the cell pellet once, they were suspended in DMEM containing 10% FBS. Cells were plated onto Falcon 60 mm plastic tissue culture dishes which had been coated with poly-L-lysine 6 (Appendix, Poly-L-lysine) at a density of 8 x 10 cells/plate. Dishes were incubated at 37c in a humidified incubator with 5% co2 -95% air. Cells were allowed to establish in culture for 3 days. Medium was replaced with DMEM containing 10% horse serum (HS) with 10 uM cytosine arabinoside (Appendix, Cytosine Arabinoside) to inhibit the multiplication of dividing cells, predominantly glial cells. After 48 h, medium containing cytosine arabinoside was replaced with DMEM containing 10% HS. Cells were maintained in culture for 14 days with the medium replaced every 5 days before being used for experiments. Microscopic examination of these cultures reveals two populations of cells. The majority of established cells have small, phase dark cell bodies with extensive processes (Fig. 1).

PAGE 25

18 Many of these cells aggregate together in small clusters while others exist as single cells. The processes form an extensive network of interconnections between cells. Immunocytochemical characterization of these cultures with antibodies to the enzyme, neuron-specific enolase (NSE), a specific marker for neurons, has revealed that these cells are neurons (16). These cells represent more than 80% of the total populations of established cells in these cultures. These cells overlay a layer of large flat cells of glial origin which constitute the second population of cells in these cultures. Glial-Enriched Brain Cell Cultures Astrocyte glial cultures were prepared as previously described (95) from brains of one-day-old rats. Primary cultures are prepared as described above with the following exceptions. After sedimentation cells were resuspended in DMEM with 10% FBS to a density of 1.0 x 10 6 cells per ml. Cells were inoculated in 10 ml of growth medium onto Falcon 100 mm plastic tissue culture dishes which had been coated with poly-L-lysine at a density of lOx 10 6 cells/plate. Dishes were incubated at 37 Cina humidified incubator with 5% co2 -95% air. Cells began to attach to the culture dishes immediately and 90-96% were attached to dishes within 30-60 min. Medium was replaced with fresh DMEM with 10% FBS every 3 days until cultures were confluent

PAGE 26

Figure 1: Phase contrast photomicrograph of neuronal cultures. Neuronal cultures were prepared as described in the text. Neurons are small phase dark cells with projections extending out from the cell soma. Cell soma are 6 to 10 um across. A glial cell can be seen on the left edge of the photomicrograph. Glial cells are large, flat transparent cells underlying neurons. Bar represents 10 um.

PAGE 27

20

PAGE 28

21 (as determined by microscopic examination). Examination of these cultures revealed heterogeneous populations of cells consisting predominantly of large, flat, phase light cells which have been identified as nonneuronal, glial cells and a subpopulation of small, phase dark cells with extensive processes which have been identified as neurons (16, 95). These cultures were washed once with 5 ml of Solution D and dissociated by mild trypsin treatment for approximately 10 min. Trypsin digestion was terminated with DMEM containing 10% FBS, cells were transferred to sterile centrifuge tubes and sedimented at 500 g for 10 min Dissociated cells were resuspended in DMEM containing 10% FBS a nd inoculated in 10 ml of growth medium onto 60 mm plastic tissue culture dishes at a density of 5 x 10 5 cells per plate. Cells were maintained in a humidified incubator at 37c in 5% co2 95% air, fed every 5 days with DMEM containing 10% FBS and used for experiments when confluent as determined by microscopic examination, usually within three weeks of primary culturing Phase contrast examination of cultures demonstrated a monolayer of large flat cells which have been identified as of glial origin (18, 28) (Fig. 2) In addition, irnrnunocytochemical analysis with antibodies to glial fibrillary acidic protein (GFAP), a marker specific for astrocytes, demonstrated that these cultures contain more than 95% astrocyte glial cells (Fig. 3).

PAGE 29

Figure 2: Phase contrast photomicrograph of astrocyte glial cultures. Astrocyte glial cultures were prepared as described in the text.

PAGE 30

23

PAGE 31

Figure 3: Immunocytochemical staining of astrocyte glial cells with glial fibrillary acidic protein antibody. Astrocyte glial cultures were fixed in 3.7% paraformaldehyde in PBS for 2 h, then in Triton X-100. Fixed cells were incubated with rabb~t anti-GFAP antibody for 24 h in a humidified chamber at 4 C. Cultures were then allowed to warm to room temperature and labeled with mouse anti-rabbit second antibody and prepared for observation by the biotin-avidin m ethod.

PAGE 32

25

PAGE 33

26 Fibroblast Cultures 2 A 2 cm patch of skin was removed from the back of ether anesthetized mice. This piece of skin was soaked in TMEM containing 200 units of penicillin G and 200 ug of streptomycin per ml of medium. After 10 min the tissue wa s cut into 2 mm2 pieces and 7-10 pieces were placed in a 100 mm culture dish, outer side up. A sterile, 3 mm2 piece of glass slide was pressed on top of each skin piece. Six ml of TMEM with 15% FBS and 100 units of penicillin G and 100 ug of streptomycin per ml was added to the culture dishes. Cultures were incubated at 37c in a humidified incubator with 5% co2 and 95% air. Cultures were fed with fresh TMEM and 15% FBS every third day. Fibroblasts grew out of the skin explants within two days and reached confluency in 10-12 weeks. When cultures reached confluency, cells were dissociated off the dishes with a mild trypsin treatment. Trypsinization was halted by addition of TMEM with 15% FBS, cells were transferred to a sterile centrifuge tube and pelleted at 500 x g for 10 min. The pellet was resuspended in TMEM with 15% FBS and inoculated 6 onto 100 mm tissue dishes at a density of 1 x 10 cells per dish in a volume of 8 ml of TMEM. Alternatively, cells were 5 inoculated into 35 mm dishes at a density of 1 x 10 cells per dish in 2 ml of TMEM. When 35 mm plates were confluent, cultures were used for experiments.

PAGE 34

2 7 f 125 l. d. Measurement o I-Insu in Bin ing Neuronal Cultures d f 125 1. t 1 1 11 d Bin ing o I-insu in o g ia ce s was measure as described previously (96). Neurons cultured as described above were washed twice with 4 ml of phosphate buffered saline (Appendix, PBS), pH 7.4, containing 0.9 mM calcium chloride and 0.49 mM magnesium chloride. Triplicate culture dishes were incubated at 24c with 1 ml of HEPES buffer (Appendix, HEPES Buffer), pH 7.4, containing 1.6% BSA and either 0.4-0.5 nM of a racernic mixture of 1251-insulin (150,000-200,000 cpm) or 150,000-200,000 cpm of B26 labeled 1251-insulin to determine total binding. Triplicate cultures were also incubated simultaneously h 1 1 f ff 1 6 125 1. d 16 7 wit rn o HEPES bu er, % BSA, I-insu in an uM unlabeled porcine insulin to determine nonspecific binding. Following incubation, the cultures were rapidly washed four times with ice-cold PBS, pH 7.4. The attached cells were dissolved in 1.0 ml of 0.2 N NaOH and transferred to counting tubes. The dishes were rinsed with 1 ml of water which was combined with the original samples. Radioactivity was determined in a gamma counter with a counting efficiency of 73% for 125r. Specific 125r-insulin binding was calculated by subtracting nonspecific binding from total binding and normalized to the protein content of each dish as measured in aliquots of the solution by the method of Lowry et al. (97)

PAGE 35

2 8 Glial and Fibroblast Cultures d' f 125 1 1 1 11 d Bin ing o I-1nsu 1n tog 1a ce s was measure as described previously (96) and essentially as described above for neuronal cultures except pH 7.6 was used instead of pH 7.4. d' f 125 1 1 t f'b bl t it d Bin ing o -insu 1n o i ro as cu ures was measure a s described above except pH 8.0 was used instead of p H 7.4. Light Microscopic Autoradiography Binding of 1251-insulin to neuronal cultures was localized by light microscopic autoradiography. Insulin was bound to neuronal cultures as described above except that cultures were incubated with 500,000 cpm 1251-insulin per dish. Following the incubation, cultures were washed twice with ice cold PBS and were then fixed with 3.7% glutaraldehyde in PBS, pH 7.2, for 30 min a t 4c. After fixation, cultures were washed twice with PBS and then dehydrated through graded alcohols and air dried. The dishes were coated with liquid nuclear emulsion (Ilford KS), diluted 1:1 with distilled water containing 2% glycerol, according to the method of Rogers (98). Plates were dried overnight over silica gel at room temperature and then stored at s0c in light tight boxes. After 3 weeks of exposure, the emulsions were developed for 2 min in D-19 diluted 1:1 with water, rinsed with water and fixed for 2 min with 20% sodium thiosulphate. The dishes were finally washed for 10 min with distilled water and the cells examined by phase contrast microscopy

PAGE 36

29 Measurement of Protein Synthesis and Glycosylation f 3 1 3 1 . Incorporation o H-va 1ne or H-g ucosamine into protein was measured as indices of protein synthesis or glycosylation, respectively. One uCi/ml of labeled compound was added to the medium in which cultures were growing. After incubation for 60 min. at 37c, cells were washed twice with ice cold PBS, twice with ice cold 10% (w/v) trichloroacetic acid (TCA), once with ice cold 5% TCA, twice with ethanol/ether (3:1, v/v), then dissolved in 0.2 N NaOH. Dissolved cells were transferred to plastic tubes and each dish was rinsed with 1.0 ml distilled water which was combined with the original sample. A 1.0 ml aliquot of each sample was combined with 10 ml LiquiscintR a n d radioactivity was determined in an LKB Liquid Scintillation counter (model 1215) with a counting efficiency of 57% for tritium. A 1.0 ml aliquot of each sample was taken for protein determination by the method of Lowry et al. (97). Measurement of Monoamine Uptake Uptake of 3tt-monoamines into cultured brain cells was determined in cells attached to 60 mm culture dishes. Growth medium was removed and cells were washed once with PBS, 37c, pH 7.4, containing Cac1 2 (0.75 mM), MgC12 (0.75 mM) and glucose (30 mM). Uptake was determined by incubating cultures with 1.0 ml of PBS containing pargyline (100 uM), ascorbate (200 uM) and 0.1 uM of the appropriate {-)-(3 HJ-monoamine for 2 min. Previous studies (99) have shown that inclusion of 100 uM pargyline in the

PAGE 37

30 uptake reaction almost completely prevents deamination of N E t o dihydroxy-phenylglycol after NE is taken up int o the cells. Following the incubation, the reaction mixture was removed and dishes were washed twice with ice cold PBS, pH 7.4. Cells were dissolved in 0.2 N NaOH and were transferred to plastic tubes. Each dish was rinsed with 1.0 ml distilled water which was combined with the original sample. A 1.0 ml aliquot of each sample was combined with 10 ml LiquiscintR and radioacti v ity determined in an LKB Liquid Scintillation counter (Model 1 2 15) with a counting efficiency of 57% for tritium. uptake was expressed as pmol/mg protein. 3 H-monoa mine All monoamines (norepinephrine (NE), epinephrine (EPI), dopamine (DA), and serotonin (5-HT) are taken up by the various amine uptake systems with relatively high affinities. It i s necessary to determine specific uptake of a particular monoamine by subtracting uptake in the presence of a specific monoamine uptake inhibitor from total uptake in analogy with determining specific binding by subtracting nonspecific binding from total binding to determine specific binding (Fig. 4). In these studies maximally effective doses of specific uptake inhibitors were used in the nonspecific uptake reaction mixtures to determine nonspecific uptake. The right panel of Fig. 4 illustrates the experimental protocol to determine specific NE uptake. Maprotiline was used as a specific blocker of NE uptake, and fluvoxamine as a specific 5-HT uptake blocker. In addition, since all specific monoamine uptake sites are sodium sensitive,

PAGE 38

3 1 nonspecific uptake was determined by measuring 3 e-monoamine uptake in the absence of sodium in the reaction mixture.

PAGE 39

Figure 4: Schematic of monoamine uptake experiments. All monoamines are taken up with high affinity by the various monoamine uptake systems. To determine specific uptake it is necessary to subtract nonspecific uptake (in the presence of specific uptake inhibitors) from total uptake. Maprotiline is a specific inhibitor of norepinephrine (NE) uptake.

PAGE 40

33 TOTAL w NE +MAPROTILINE ..... a.. ::> w EPI EPI z I DA DA I Cf) ....... 5-HT 5-HT lL 0 z w -Na+ z -Na+ 0 a.. 0 0

PAGE 41

CHAPTER III KINETIC AND AUTORADIOGRAPHIC CHARACTERIZATION OF NEURONAL AND GLIAL INSULIN RECEPTORS In order to determine which specific cell types from the rat brain might respond to insulin, it was necessary to determine which cell types possess specific insulin receptors. Receptors are proteins which fulfill certain criteria which distinguish them from "binding sites." Binding to receptors displays kinetics of binding consistent with biological effects associated with the peptide, is saturable, temperature and pH dependent and specific for certain peptides. The receptor must also have sufficient affinity for the proposed ligand to explain physiological effects of the ligand. In these studies, the insulin binding assay described in Chapter II was modified in ways to determine if astrocyte glial and neuronal insulin binding sites satisfy the major kinetic criteria to be termed insulin receptors. Specifically, the time courses of association and dissociation, ligand specificity, saturability and the affinity of the binding sites for porcine insulin were determined for both cell types, using differential cell cultures. 34

PAGE 42

3 5 Experimental Procedures Time Courses of Association and Dissociation To determine the time course of association, triplicate plates of neuronal or glial cultures were incubated with 150,000 cpm/ml of B 26 labeled 1251-insulin with and without 16.7 uM unlabeled porcine insulin for increasing time periods from 5 min to 3 h. At the appropriate time, cultures were washed with ice-la PBS t h lt b d' d b d 125 1 1 co o a in ing an remove un oun -insu in, dissolved in 0.2 N NaOH, scraped and counted as described in Chapter II. The time course of dissociation was determined by incubating neuronal or glial cultures with 1251-insulin with or without 16.7 uM unlabeled insulin as described above for the period determined to result in equilibrium binding. The binding reaction mixture was then aspirated from the plates labeled with 12 5 I 1 1 d 1 d th 1 f b ff d 1 6 % -insu in a one an rep ace wi one m o u er an BSA either with or without 16.7 uM unlabeled porcine insulin. At times from 2 min to 3 h, triplicate plates were washed four times with ice cold PBS, cells dissolved in 0.2 N NaOH, plates scraped and samples counted. Specificity Specificity of the astrocyte glial or neuronal binding site for insulin was determined by incubating glial or neuronal cultures with various peptides with known abilities to compete f 125 1 b d . h 1 1 or I-insu in in ing at perip era insu in receptors. Triplicate cultures were incubated for 2 h with 150,000 cpm/ml of B26 labeled 1251-insulin and increasing concentrations of chicken insulin, human insulin, porcine insulin, guinea pig insulin,

PAGE 43

36 proinsulin, desoctapeptide insulin, glucagon, g rowt h hormone, and relaxin. Cultures were washed four times with ice cold PBS dissolved in 0.2 N NaOH, plates scraped and samples counted and assayed for protein content as described in Chapter II. Saturability and Affinity Saturability of insulin binding to glia and neurons was t t db t' for 125 1 1 b d th inves iga e y compe ing -insu in in ing wi increasing concentrations of porcine insulin as described above in the specificity experiments. Saturability was determined b y the inability of increasing concentrations of unlabeled p orcine insulin to further displace 1251-insulin binding. These dat a were then plotted by the method of Scatchard (100) to determine the affinities of the neuronal and glial insulin binding sites for porcine insulin. Light Microscopic Autoradiography 125 1 1. b. d. 1 1. d 1 1 b -insu 1n in ing was oca ize on neurona cu tures y light microscopic autoradiography as described in Chapter II. Results Neuronal Insulin Binding Binding of 1251-insulin to neuronal cultures was time and H d d t Th t f 125 1 1 b' d' t p epen en. e ime course o -insu in in ing o neuronal cultures showed that equilibrium binding was reached after 2 hat 24 C (Fig. 5a). The T 112 of association was 35 min. Total binding of 1251-insulin to neuronal culture s at equilibrium was 6.3 ~0.1% (mean~ S.E .M. of total radioactivity/ mg protein); and nonspecific binding was 0.9 + 0.1%

PAGE 44

Figure 5: t' d d. t. f 125 1' 1 Associa ion an issocia ion O I-insu in to n1 ~ 5ona cultures. Neuronal cultures were incubated with Iinsulin for indicated time periods as described in the text. When binding had reached a steady state, the binding reaction mixtures was aspirated from the cells and replaced with 16.7 uM unlabeled porcine insulin. Data are representative of two experiments and are expressed as mean+ S.E.M.

PAGE 45

120 a. b. C, z 0 80 z al eo1 /f I \ I w 00 -X c( '#-40 20 40 80 120 160 0 40 80 120 TIME

PAGE 46

39 More than 90% of bound 1251-insulin could be dissociated within 2 h after addition of a large excess (16.6 uM) of unlabeled porcin e insulin (Fig. Sb). The T112 of dissociation of 1251-insulin was 15 min. The optimum pH for binding was 7.4. C t t f 1251 1 b d' b h d ompe i ion or -insu in in ing y ot er pepti es structurally or physiologically related to insulin showed competition for binding proportional to their biological activity at peripheral insulin receptors (Fig. 6). The order of potency d 1 125 1 h k 1 h 1 in isp acing I-insu in was c ic en insu in> uman i nsu in= porcine insulin> proinsulin > guinea pig insulin. Growth hormone and glucagon did not compete for insulin b inding. Cells were incubated with 1251-insulin in the presence of increasing concentrations of unlabeled porcine insulin for 2 h (Fig. 7). The rc50 for inhibition of 1251-insulin b inding b y porcine insulin was 16.7 nM. These data were analyzed by the method of Scatchard (100) and resulted in a curvilinear plot (Fig. 8). The apparent affinity constants were calculated using a model for two classes of binding sites. The apparent high affinity site has a dissociation constant (Kd) of 11 nM and the low affinity site a Kd of 65 nM. The high affinity binding capacity (B0 ) was 0.4 pmol/mg protein and total binding capacity was 1.8 pmol/mg protein. Autoradiographic microscopy was performed to determine h h 12 S 1 b d t th 11 wet er I-insu in was in i ng to neurons or o e sma percentage of glial cells which persist in these cultures.

PAGE 47

Figure 6: Competition for neuronal 1251-insulin binding by 12~ated peptides. Neuronal cultures were incubated with !insulin and indicated concentrations of related peptides: human insulin (()), porcine insulin (~), guinea pig insulin (i'.'.l), proinsulin (~), glucagon (()) and growth hormone (4t) for 2 hat 24 C. Data are representative of three experiments and are expressed as mean+ S.E.M.

PAGE 48

41 0 0 0 'r" C -0 w 'r" z 0 a: 0 :c 'r" 0 IO O IO O IO C\I C\I 'r" 'r" 0 (OOl X) 33~:I/ONnoe

PAGE 49

Figure 7: Competition-i~bition curve of unlabeled porcine insulin for neuronal 125insulin binding. Neuronal cultures were incubated with I-insulin and the indicated concentrations of unlabeled porcine insulin for 2 hat 24 C. Data are representative of four experiments and are expressed as mean+ S.E.M.

PAGE 50

4 3 z w .... 0 a: a. C, 2 ...... C z ::> 0 co 'iP, 1 0 L----'-------'------.1.-----'-------'-------'------' 1 1 10 9 8 7 6 -LOG INSULIN CONCENTRATION (M) w

PAGE 51

Figure 8: Scatchard plot of insulin binding to neuronal cultures. Data from Fig. 7 were replotted as Bound/Free vs. Bound. The slopes of the lines are related to the Kd of binding and the X-intercept is the binding capacity {B). The data were analyzed by a model for two receptor0sites and revealed a high affinity Kd of 11 nM and a low affinity Kd of 65 nM. The binding capacity was 1.8 pmol/mg protein.

PAGE 52

4 3 -0 0 'P'" X -w w 2 I ------I ,:,.. V, 0 z :::, 0 m 1 o-------.&. ______ _._ ______ ......_ _____ 0 400 800 1200 BOUND (FMOL/MG PROTEIN)

PAGE 53

46 Exposure of Ilford KS emulsion for three weeks to 1251-insulin labeled neuronal cultures revealed discrete grains specifically associated with neurons in these cultures (Fig. 9). Neurites had a higher density of grains than neuronal soma (Fig. 9a-b). Not all neurons exhibited autoradiographic grains (Fig. 9c). No f. 1 1 t. f 1 b 1 d h 1251 1' speci ic oca iza ion o a e occurre wen -insu in was bound to cultures in the presence of 16.6 uM unlabeled porcine insulin. Glial Insulin Binding Astrocyte glial cultures specifically bound 1251-insulin. d. f 125 1. 1. 1 11 24 d dl Bin ing o I-insu in tog ia ce sat C increase rapi y for the first 60 min. and reached a plateau by 90-120 min (Fig. 10a). The T112 of association was 10 min. Binding remained constant for up to 150 min, when insulin degradation (20% of the total insulin) interfered with accurate binding determinations. At equilibrium, approximately 1% of the total 1251-insulin was specifically bound to glial cultures per mg protein. When cultures were incubated with 1251-insulin for 2 h to reach equilibrium, the reaction mixture aspirated and replaced with 16 7 1 b 1 d 1 125 1. d . t d f 1. 1 uM un a e e insu in I-insu in issocia e rom g ia cultures in a time-dependent manner (Fig. 10b). Approximately 95% f 1251 1. d. d h. 75 . th T f 8 o -insu in issociate wit in min wi a 112 o min indicating rapid reversibility of insulin binding. However, the rate of dissociation of 1251-insulin was not decreased by addition of buffer alone at the initiation of the dissociation period (Fig. 11).

PAGE 54

Figure 9: h t d' h f 125 1 b d L1g t microscopic au ora 1ograp yo I-1nsu 1n oun to neuronal cul1 ~5es. Neuronal cultures wer5 incubated with 500,000 cpm I-insulin for 60 min at 24 C, washed four times with ice-cold PBS, then fixed with 3.7% glutaraldehyde for 30 min. Cultures were washed twice with ice cold PBS, then coated with Ilford KS emulsion. Emulsion was exposed for two weeks and developed for visualization by light microscopy. Panel A shows discrete autoradiographic grains over soma of neuronal cells, Panel B shows discrete grains on neurites from neuronal cells, and Panel C shows control autoradiograms where binding was carried out in the presence of 16.7 uM unlabeld porcine insulin.

PAGE 55

48 A B C

PAGE 56

. 10 t. d d. t. f 1251 1 t Figure : Associa ion an issocia ion o -insu in o astrocyte glial12~ltures. Astrocyte glial cultures were incubated with I-insulin for indicated time periods as described in the text. When binding had reached a steady state, the binding reaction mixtures was aspirated from the cells and replaced with 16.7 uM unlabeled porcine insulin. Data are representative of three experiments and are expressed as mean+ S.E.M.

PAGE 57

0 CII ... 0 so 0 0 0 0 0 o e e N ... (wnw1xer.~) punoe u11nsu1-1 vza IO 0 -IO .... IO IO 0 0 C IO e -....., GI 0 E CII i= 0 GI 0 e 0 ., 0

PAGE 58

Figure 11: Dissociation of 1251-insulin from astrocyte glial cultures with and without unlabeled porcine l~~ulin. Astrocyte glial cultures were incubated with I-insulin for 2 hat 24 C. Binding reaction mixture was aspirated and replaced with either 16.7 uM unlabeled porcine insulin~) or PBS (()). Data are representative of two experiments and expressed as mean+ S.E.M.

PAGE 59

0 0 ,0 CX) 52 0 co 0 C\I 8NIONl8 r.Jnll::ISn1no3 % 0 0) 0 co 0 ('I') 0 0 -z -w ....

PAGE 60

53 B d f 125 1 1. 1 1 bl d in ing o I-insu in tog ia cu tures was satura e an specific for insulin. Binding of 1251-insulin was inhibited by unlabeled insulins and related peptides (Fig. 12). Chicken insulin was equipotent with porcine insulin in inhibiting the b d. f 1251 1. . 1. . 1 in ing o -insu in, guinea pig insu in, porcine proinsu in, insulin-like growth factor II competed with less potency consistent with their insulin-like biological activity in other tissues, whereas desoctapeptide insulin, glucagon, relaxin and human growth hormone did not inhibit binding. C 11 b d h 12 5 1 th f e s were incu ate wit I-insu in in e presence o increasing concentrations of unlabeled porcine insulin for 2 h. (Fig. 13). The rc50 for inhibition of 1251-insulin binding by porcine insulin was 30 nM. These data were analyzed by the method of Scatchard (100) and resulted in a curvilinear plot (Fig. 14). The apparent affinity constants were calculated using a model for two classes of binding sites. The apparent high affinity site has a dissociation constant (Kd) of 13.7 nM and high affinity binding capacity of 0.2 pmol/mg protein and the low affinity Ka of 357 nM with a total binding capacity (B0 ) of 3.83 pmol/rng protein. Discussion Cultured cells from newborn rat brain were used to kinetically characterize insulin binding sites on astrocyte glia d B d' f 125 1 1 t b th 11 t an neurons. in ing o -insu in o o ce ypes is consistent with classical binding of ligand with receptor. In particular, binding is rapid, reversible, saturable, and specific for various insulins and related peptides.

PAGE 61

Figure 12: Competition for astrocyte glial 1251-insulin binding by relat1 ~5peptides. Astrocyte glial cultures were incubated with I-insulin and the indicated concentrations of related peptides: human insulin (~), porcine insulin (() ) guinea pig insulin (~), proinsulin (~), relaxin, human growth hgrmone, glucagon, and desoctapeptide insulin for 2 hat 24 C. Data are representative of two experiments and expressed as mean+ S.E.M.

PAGE 62

RelaJC. n 1,G H 100 G lucogon D O P in, a l i n I ---:::::::::::: ...--. e so -C: 0 (.) L ~'\ \ \ j Vl "O 60 Vl C: :J 0 aJ c: 40 :J "' C: I ""20 ('lj -0 0 1 1 0 1 0 100 1000 Insulin (ng)

PAGE 63

Figure 13: Competition-inhibiti~~5curve of unlabeled porcine insulin for astrocyte glial I-insul~~ binding. Astrocyte glial cultures were incubated with I-insulin and the indicated concentrations of unlabeled porcine insulin for 2 hat 24 C. Data are representative of four experiments and expressed as mean+ S.E.M.

PAGE 64

57 (\I .... Nl310~d ~~taNnoa% 0 -z ...J :::, Cl) z co C, 0 _J I 0)

PAGE 65

Figure 14: Scatchard plot of insulin binding to astrocyte glial cultures. Data from Fig. 7 were replotted as Bound/Free vs. Bound. The slopes of the lines are related to the Ka o f binding and the X-intercept is the binding capacity (B). The data were analyzed by a model for two receptor si~es and revealed a high affinity Kd of 13.7 nM and a low affinity Kn of 357 nM. The binding capacity was 3.83 pmol/mg protein.

PAGE 66

0 C\I 59 T"9 coo~ X)33~.:1,aNnoa z w t o a: 0. (!J ...... 0 ..J ~o 0. "" C z :::> 0 al

PAGE 67

60 B d f 125 1. 1. 1 1 in ing o I-insu in tog ia cu tures was more rapid than binding to neuronal cultures (T112 of associations, 10 min vs 35 min, glia vs neurons) although binding reached equilibrium within 2 h in both culture systems. Dissociation times were approximately equivalent. Neuronal cells in culture have greater ff t f 125 1 th 1 1 11 1 a ini y or I-insu in an astrocyte g ia ce sin cu ture (11 nM and 65 nM vs 13.7 and 357 nM, high and low affinity Kds for neurons and glia, respectively), although this difference is not marked. Considerable variability exists between independent neuronal cultures which may account for observed differences in binding kinetics. At d h t h d t t that 1251-insuli'n u ora iograp ic ec niques emons ra e binding sites exist on both neurites and cell soma of neurons in culture, confirming that the observed insulin binding sites in neuronal cultures are primarily due to neuronal insulin binding, not to the small proportion of astrocyte glia which contaminate the preparation. In general, insulin binding to both neuronal and glial cultures is similar to insulin binding to other known peripheral target tissues for insulin (32). The only consistently observable difference in insulin binding to these cultures as compared to their peripheral counterparts appears to be lack of negative cooperativity of glial insulin receptors. That is, the rate of dissociation of insulin from glial cultures is not increased when dissociation occurs in the presence of excess insulin as compared to dilution with buffer alone. This finding

PAGE 68

61 has been previously reported in whole brain preparations (101) and is confirmed here to occur in astrocytes. Both neurons and astrocyte glial cells in culture possess binding sites for insulin with kinetics consistent with those of classical insulin receptors. These findings suggest that neurons and astrocyte glia cells may be responsive to insulin, whether of central or peripheral origin. Subsequent chapters address the questions of how insulin binding to neurons and glia is regulated and what effect insulin may have on neurons via interaction with these specific insulin binding sites.

PAGE 69

CHAPTER IV REGULATION OF NEURONAL AND GLIAL INSULIN RECEPTORS BY INSULIN AND INHIBITORS OF GLYCOSYLATION AND PROTEIN SYNTHESIS Regulation of insulin binding to a cell is one of the primary physiological methods for altering a cell's responsiveness to insulin. Down regulation of insulin receptors by insulin is ubiquitous among peripheral tissues (42-48). Since the insulin receptor is a glycoprotein, inhibition of glycosylation affects insulin receptor expression. Treatment with tunicamycin, an inhibitor of glycosylation has been demonstrated to result in a reduction in insulin binding in peripheral tissues (60-63). However, inhibition of protein synthesis has had variable effects on insulin binding to peripheral tissues. It seems that treatment with cycloheximide, an inhibitor of protein synthesis, results in an increase in insulin binding, presumably by decreasing the rate of loss of receptors from the cell surface (39). Experiments were carried out to determine what effect incubation with insulin, tunicamycin, and cycloheximide had on neuronal and astrocyte glial insulin binding. 62

PAGE 70

63 Results Effect of Insulin on Neuronal and Astrocyte Glial Insulin Binding Neuronal and glial cultures were incubated at 37c for 24 h with increasing concentrations of porcine insulin in PBS to determine how neuronal and glial insulin binding were affected by increased extracellular insulin. Contrasting results were obtained from neuronal cultures prepared in slightly different ways. Neurons cultured in 10% fetal bovine serum (FBS) for the first three days of culture (FBS/HS neurons) responded to elevated concentrations of insulin (up to to 3 uM insulin) by increasing insulin binding by a maximum of 86% with 3 uM. The Eo 50 of insulin stimulated increase in FBS/HS neurons was 100 nM (Fig. 15). However, when neurons were cultured in 10% plasma derived horse serum (PHS) for the first three days of culture (PHS/HS or PHS/PHS neurons) elevated concentrations of insulin resulted in reduced insulin binding (Fig. 15). A maximum decrease in PHS/HS neuronal insulin binding of 53% was observed with incubation of cells with 166 nM insulin with an ED50 of 5 nM insulin. Decreased insulin binding in PHS/HS neurons appears to be due to a reduction in maximum binding capacity and not a reduction in the rate of association (Fig. 16). Astrocyte glial cultures (cultured continuously in 10% FBS) responded to increasing concentrations of insulin by decreasing specific glial insulin binding. Maximum reduction of 125Iinsulin binding to 35% of control binding was observed with incubation of cells for 24 h with 166 nM insulin. The ED50 for reduction of insulin binding by insulin was 16 nM (Fig. 17).

PAGE 71

Figure 15: Effect of insulin on neuronal insulin binding. Neuronal cultures prepared for the first three days of culture in 10% FBS, and subsequently in 10% HS (FBS/HS,~), initially in 10% Plasma Derived Horse Serum, then 10% HS (PHS/HS,~), or continuously in PHS (PHS/PHS,()) were incubated with the indbcated concentrations of porcine insulin for 24 hat 37 C. Insulin binding was then assayed as described in the text. Data are representative of ten experiments and presented as mean+ S.E.M.

PAGE 72

65 LO -z HH
PAGE 73

. 16 . f 125 1 1 1 / Figure : Association o I-insu in to neurona cu tures. PHS HS neuronal cultures were either incubated (~) are not incubated (()) with 166.7 nM insuli~sfor 24 hat 37 C. Cultures were then incubated with I-insulin for the indicated time periods and binding measured. Data are expressed as mean+ S.E.M.

PAGE 74

(!) z 0 z co 0 iL 0 w 0.. en '#. 5 4 3 I 2 1 0 0 I 40 80 TIME (MIN) I CTI -...J 120

PAGE 75

Figure 17: Effect of insulin on astrocyte glial insulin binding. Astrocyte glial cultures were incubated with the indicated concentrations of porcine insulin for 24 hat 37 c. Insulin binding was then assayed as described in the text. Data are representative of six experiments and presented as mean+ S.E.M.

PAGE 76

69 C\J 0 Nl310Hd E)LAJ/GNnos % 00 6 O') z :J :::> U) z C.!J 9 I

PAGE 77

70 Addition of 166 nM insulin resulted in significant reduction of insulin binding within 4 h with maximum reduction of binding observed by 30 h {Fig. 18). Insulin binding was assayed in insulin-treated {down regulated) and untreated astrocyte glial cells in the presence of increasing concentrations of unlabeled porcine insulin {Fig. 19). The Ic50 for inhibition of 125Iinsulin binding by porcine insulin was unchanged by treatment with insulin (30 nM control vs 40 nM insulin treated). These data were analyzed by the method of Scatchard (100). The major change in the binding parameters between control and insulintreated glial cells was a reduction in the number of high affinity binding sites by insulin treatment (0.25 pmol/mg protein vs 0.12 pmol/mg protein, control vs insulin treated) {Fig. 20). Effect of Tunicamycin On Neuronal and Astrocyte Glial Insulin Binding Neuronal and astrocyte glial cultures were incubated with increasing doses of tunicamycin {TM) to determine effective doses for inhibition of glycosylation as determined by inhibition of incorporation of 3tt-glucosamine into TCA insoluble material. It was determined that 0.28 uM TM was an effective dose for both neuronal and astrocyte glial cultures, inhibiting 73% of neuronal glycosylation {data not shown) and 60% of glial glycosylation {Fig. 21). Incubation of neuronal cultures with varying concentrations f f 24 h h d ff 1 125 1. b. d o TM or a no e ect on neurona I-insu in in ing

PAGE 78

Figure 18: Time course of down regulation of glial insulin binding by insulin. Astrocyte glial cultures were incubated with 16.7 uM insulin for the indicated time periods and insulin binding assayed as described in the text. Data are representative of two experiments and presented as mean+ S.E.M.

PAGE 79

72 8NIGNl8 Ol.:1103dS % 0 M 0 C\I 0 'r"" 0 0 :r: -w ....

PAGE 80

Figure 19: Competition-inhibition by porcine insulin of 125I-insulin binding to normal and down regulated astrocyte glial cultures. Astrocyte glial cultures were either incubated (~) or not incubated (()) with 16.7 uM porcine insulin for 2f2g at 37 C. Subsequently cultures were incubated with I-insulin and the indicated conc~ntrations of unlabeled porcine insulin for 2 hat 24 C. Insulin binding was assayed as described in the text. Data are representative of four experiments and and are expressed as mean+ S.E.M.

PAGE 81

0 C'? 74 C\I .... Nl3l0l:ld DVHONnoa% :E -z ..J ::> Cf) co z 0) O> 0 T

PAGE 82

Figure 20: Scatchard plots of normal and down regulated astrocyte glial cells. Data from Figure 20 were replotted by the method of Scatchard and reveal that the decrease in insulin binding observed by incubation of astrocyte glial cells with insulin is due to a loss of high affinity insulin receptors and not a marked change in the affinities of the glial insulin receptor.

PAGE 83

76 0 ('I') N .... L----L----...-.::i..c::1~::h!..::...__ __ ~-=illlllL-1-______ .......JO 0 O C? N .... coo i x)33~:1,0Nnoa C Cl) 0 0. C> E ..... 0 E 0. -0 z ::::> 0 m z :J ::::> u, z -

PAGE 84

Figure 21: Effect of tunicamycin on astrocyte glial glucosamine incorporation. Astrocyte glial cultures were incubated with the indicated concentrations of tunicamycin (TM) for 24 hat 37 C. Glucosamine incorporation into TCA insoluble substance was measured as described in the text and used as an index of glycosylation of proteins. Data are expressed as mean~ S.E.M.

PAGE 85

0 0 T""" 0 co 78 0 V 0 C\I 0 NOll'v'tfOdtfOONI 3Nl1"J'v'S00nl8 lOtflNOO % I'-co a, 0 ..,-..,-..,z 0 a: .._ z w (.) z 0 (.) z (.) ><{ (.) z :::::> .._ (!) 0 ...J I

PAGE 86

79 (Fig. 22). In contrast, incubation of both astrocyte glial and fibroblastic cultures for 24 h with TM caused a significant reduction in insulin binding (Fig. 22). A maximum reduction of insulin binding of 60% was achieved with both astrocyte glial cultures and fibroblastic cultures with 0.28 uM TM with an ED50 of 20 nM. Astrocyte glial cultures were incubated for 24 h with varying concentrations of insulin with or without 0.28 uM TM to determine whether TM affected the process of down regulation. Incubation of cultures with TM blocked the reduction in insulin binding normally resulting from incubation of cultures with insulin (Fig. 23). Effect of Cycloheximide On Neuronal and Astrocyte Glial Insulin Binding Cycloheximide (CHX) effectively inhibits protein synthesis in both neuronal cultures and astrocyte glial cultures with 0.35 uM CHX inhibiting 73% of neuronal protein synthesis (data not shown) and 70% of astrocyte glial protein synthesis (Fig 24) as measured by inhibition of incorporation of 3H-valine into TCA insoluble material. Incubation of neuronal cultures with CHX resulted in dose-independent increase in insulin binding, with maximum increase in neuronal insulin binding occuring with 0.35 nM CHX (Fig. 25). This concentration of CHX has no effect on neuronal protein synthesis. Incubation of astrocyte glial cultures with increasing concentrations of CHX for 68 h resulted in a dose-dependent

PAGE 87

Figure 22: Effect of TM on neuronal, astrocyte glial and fibroblastic insulin binding. Neuronal (~), astrocyte glial (()) and fibroblastic (II) cultures were incubated with the indicated concentrations of TM for 24 hat 37 C. Insulin b inding was assayed as described in the text and data are representative of six experiments and expressed as mean+ S.E.M.

PAGE 88

81 co (\I 0 ..-~ ('! :::, oz (.) >'It <( ~(.) O...... 0 0 Z :::> .... '------'----:::o~-+--""'f---=------'---=-------:::0 (\I .... E>NIGNl8 lOi::llNO~ %

PAGE 89

Figure 23: Effect of inhibition of glycosylation on down regulation of astrocyte glial insulin receptors by insulin. Astrocyte glial cultures were incubated with indicated concentrations of insulin with (~) or without (()) 0.28 uM TM for 24 hat 37 C. Insulin binding was assayed as described in the text and data are representative of two experiments and expressed as mean+ S.E.M.

PAGE 90

83 L----~0i+--+-"-'-t-~L01-----;o 0 8NIGNl8 ~l.:f1~3dS % ,..... co 0) -z 0 a: .,_ z w 0 z 0 0 z :J ::> Cf) z CJ 0 ...J I

PAGE 91

Figure 24: Effect of cycloheximide on valine incorporation into protein in astrocyte glial cultures. Astrocyte glial cultures were incubated with the indicated concentrations of CHX for 24 hat 37 C. Valine incorporation into TCA insoluble substance was measured as described in the text and used as a measure of protein synthesis. Data are representative of two experiments and expressed as mean+ S.E.M.

PAGE 92

z 0 a: 100 0 a.. a: 0 (.) z 75 -w z ::J <( 50 II co > lJ1 ...J 0 a: .... z 25 0 (.) ?fi 0 9 8 7 LOG CYCLOHEXIMIDE CONCENTRATION

PAGE 93

Figure 25: Effect of inhibition of protein synthesis on neuronal insulin binding. Neuronal cultures were incubated with the indicated concentrations of CHX for 24 hat 37 C. Insulin binding was assayed as described in the text and data are representative of three experiments and and expressed as mean+ S.E.M.

PAGE 94

87 LO z 0 (0 a: I-z w (.) z ,...... 0 (.) w 0 -:E co X w I 0 _J 0) (.) CJ 0 0 _J ,.I C\J ,.0 . E:>NIONl8 81.:U83dS %

PAGE 95

88 increase in insulin binding, with 0.35 uM CHX resulting in a 70% increase in glial insulin binding (Fig. 26). Incubation of astrocyte cultures with 0.35 uM CHX resulted in a time-dependent increase in binding, with a significant increase in binding occuring by 38 hand a maximum increase of 480% of control binding occuring at 84 h, although reduced cell viability resulted in large experimental error at time periods greater than 72 h (Fig. 27). Insulin binding was assayed in astrocyte glial cultures treated with or without 0.35 uM CHX for 72 h. The IC 50 of 20 nM for inhibition of 125I-insulin binding by unlabeled porcine insulin was unchanged by CHX treatment (Fig. 28). When these data were analyzed by the method of Scatchard (100), it was apparent that the CHX induced increase in astrocyte glial insulin binding was due to an increase in the number of high affinity sites (Fig. 29). Astrocyte glial cultures were incubated with increasing concentrations of insulin in the presence or absence of CHX to determine whether protein synthesis is important for the process of insulin-induced downregulation of ast.rocyt.e glial insulin receptors. Incubation of cultures with 0.35 uM CHX blocked the insulin-induced down regulation of insulin binding (Fig. 30).

PAGE 96

Figure 26: Effect of inhibition of protein synthesis on astrocyte glial insulin binding. Astrocyte glial cultures were incubat6d with the indicated concentrations of CHX for 48 hat 37 C. Insulin binding was assayed as described in the text and data are representative of two experiments and expressed as mean percentage of total radioactivity bound per mg protein+ S.E.M.

PAGE 97

90 z 0 ....... a: .... z w (.) z 0 (.) w 0 CX) X w I 0 ...J (.) C!J 0 ...J 0) ('I) C\J . 0 E>NIGNl8 81.:U83dS %

PAGE 98

Figure 27: Time course of increase in astrocyte glial insulin binding by inhibition of protein synthesis. Astrocyte glial cells were incubated with 0.35 uM CHX at 37 C for the indicated time periods. Insulin binding was determined as described in the text and data presented as mean+ S.E.M.

PAGE 99

0 0 lO 92 0 0 0 0 0 0 'l:t' C"') C\I 8NIONl8 lO~lNO~ % 0 0 ,... 0 co 0
PAGE 100

. 28 t' 'nh'b't' f . 1 f 125 Figure : Compe ition-i ii ion curve o porcine insu in or !-insulin binding to normal and CHX-treated astrocyte glial cultures. Astrocyte glial cultures were ingubated with 0.35 uM CHX (()) or not (tt) fo12~a hat 37 C. Cultures were washed and incubated with I-insulin and the indicated concentrations of porcine insulin. Insulin binding was measured as described in the text.

PAGE 101

94 (0 -z 0 CI: tz w CX) (.) z 0 (.) z ::J :::> Cl) z 0 CJ .,0 _J I LO 0 LO C\J . .,. 0 0 8NIGNl8 ~l.~1~3dS %

PAGE 102

Figure 29: Scatchard plots of normal and CHX-treated astrocyte glial cells. Data from Figure 28 were replotted by the method of Scatchard and demonstrate that the increase in glial insulin binding after incubation of cells with CHX is due to an increase in surface high affinity insulin receptors.

PAGE 103

2.0 -0 0 ,... X -w w a: LL 1.0 Cl PY I I.O z ::) 0 m 0 .......__----L-----'-----'----..._-----JL....-_.1 o .33 .66 to 1.33 t66 BOUND (PMOL/MG PROTEIN)

PAGE 104

Figure 30: Effect of inhibition of protein synthesis on down regulation of astrocyte glial cells by insulin. Astrocyte glial cultures were incubated with (~) or without (()) 0.35 uM CHX ang the indicated concentrations of insulin for 24 hat 37 C. Insulin binding was measured as described in the text and data presented as mean+ S.E.M.

PAGE 105

C"') 0 C\I 0 98 .,... Q 8NIONl8 81.:U83dS % co r,... a:> O> 0 .,... 0 :E -z 0 a: Iz w t) z 0 () z :J ::> (f) z C, 9

PAGE 106

99 Discussion Neuronal insulin receptors in culture are unique in their responses to extracellular insulin. Neuronal cells cultured in 10% FBS for the first three days of culture respond to insulin by increasing specific insulin binding in contrast to all other normal tissues in which insulin binding has been studied. However, if neurons are cultured initially in plasma derived horse serum (PHS), neurons regulate their insulin binding in a manner typical of other tissues. Neurons apparently have the novel ability to change their responsiveness to insulin during the early stages of culture. Since FBS has many more growth factors than PHS, this evidence suggests that there is a factor present in FBS but absent in PHS which irreversibly alters the regulation of neuronal insulin binding. Alternatively, there may be interactions established between astrocyte glia cells persisting in neuronal cultures which are dependent on some factor to be established. Type II diabetes mellitus is thought to be, in large part, due to a defect in regulation of insulin binding. Further investigation of these changes in neuronal insulin binding regulation with culturing conditions may have clinical relevance in the understanding and treatment of Type II diabetes. These observations have important implications for the usefulness of neuronal cultures as models for normal neuron function in the central nervous system. They suggest that normal neurons may have a degree of plasticity and can change essential points of their metabolic responses to physiologic stimuli.

PAGE 107

100 However, it is essential to determine what the normal responses of neurons in vivo are to these stimuli. Neuronal cultures must be recognized as models, and not normal situations, to study neuronal function. Astrocyte glial cells in culture down regulate their insulin receptors in response to extracellular insulin in a manner similar to down regulation observed in other tissues. This down regulation occurs over a short time course suggesting that it may be physiologically significant in regulating astrocyte glial responsiveness to insulin. Glial down regulation results from a reduction in the number of high affinity insulin receptors in a manner analogous to that of insulin responsive peripheral tissues. Inhibition of glycosylation has different effects on insulin receptor expression in neuronal and astrocyte glial cultures. Neuronal insulin binding is not affected by incubation with tunicamycin (TM), an inhibitor of glycosylation. This suggests that glycosylation is not an important step in the processing of the neuronal insulin receptor. Other investigators have found that the brain insulin receptor is smaller than peripheral receptors and have suggested that this difference may be due to reduction in the carbohydrate moities present on the receptor protein (101). Previous studies have suggested glycosylation is an essential step for the receptor protein to be able to be inserted into the cell membrane. Thus, TM induced reductions in insulin binding are due to reduced insertion of

PAGE 108

101 synthesized receptors into the cell membrane. This may be the mechanism of TM induced reduction of insulin binding in astrocyte glial cultures, however, it is also apparent that TM eliminates insulin-induced down regulation, suggesting that glycosylation is also essential for some aspect of ligand-induced internalization in astrocyte glial cells. Treatment of neuronal and astrocyte glial cultures with cycloheximide (CHX) also produced different results. Neuronal insulin binding was increased by CHX treatment, however this increase in binding occurred at a dose which does not inhibit protein synthesis and no further increase in binding was noted at higher concentrations. This suggests that protein synthesis is not as important in the maintenance of normal levels of neuronal insulin binding as has been found in other tissues although this effect could also be due to reduced turnover of neuronal membrane proteins or inhibition of synthesis of an enzyme essential to internalization at extremely low concentrations of CHX. Astrocyte glial insulin binding is increased by CHX treatment at concentrations similar to those concentrations at which CHX inhibits protein synthesis. This increase in insulin binding is due to increased high affinity binding sites. These findings are consistent with the hypothesis that protein synthesis is essential to the normal internalization (turnover) of unoccupied receptors. However, protein synthesis also seems to be essential for insulin-induced internalization of the glial insulin receptor as well. CHX treatment blocks insulin-induced down regulation of glial insulin receptors.

PAGE 109

102 These findings indicate that neuronal and glial insulin receptors are regulated differently. Further study of the newly differences in neuronal and astrocyte glial regulation of insulin binding may have important implications for the understanding of insulin's physiological significance in the brain as well as the responses of peripheral cells to insulin.

PAGE 110

CHAPTER V MODULATION OF NEURONAL MONOAMINE UPTAKE BY INSULIN The central nervous system has been considered insulin independent because of the limited penetration of pancreatic insulin into the brain. However, with the demonstration of insulin and insulin binding sites in the brain (13-17), and subsequent characterization of specific insulin receptors on neurons and astrocyte glial cells in culture (Chapter III) the possible role of insulin in brain metabolism and function needs to be reconsidered. Several studies have suggested that brain insulin plays a role in neuromodulation. Barbaccia et al. (82), suggested a role for insulin as a neuromodulator interacting with the dopamine system active in olfactory bulbs. Sauter et al. (83), reported that insulin at high doses stimulates release of dopamine, norepinephrine and epinephrine from hypothalamic slices. Saller and Chiodo (84) have demonstrated changes in firing rates of nigrostriatal dopamine neurons by insulin and Palovcik et al. (85) have demonstrated changes in firing rates of hippocampal neurons by insulin. Clearly, these studies are complicated by the use of various whole brain preparations, making it extremely difficult 103

PAGE 111

104 to prove a direct effect of insulin on neurotransmitter processes. Neuronal cultures are a good model for studying direct actions of insulin on neurons. Sumners et al. have shown that neuronal cultures contain catecholamines (102) and possess specific uptake mechanisms for catecholamines which are inhibited by specific norepinephrine (NE) uptake blockers, such as maprotiline, but not by serotonin uptake inhibitors (99). The direct effect of insulin on monoamine (catecholamine and serotonin) uptake was studied to determine if insulin directly affects neurotransmitter metabolism. Uptake is the primary route of inactivation of the monoamines, so changes in the rate of uptake of these neurotransmitters results in changes in the duration of action of the transmitter at the synapse. Although the various uptake systems are specific for their particular amine in the sense that there are specific inhibitors of each uptake system, each monoamine is taken up by each uptake system with relatively high affinity. To determine insulin's effect on a particular uptake system, it was necessary to examine changes in specific uptake, that is the differences between total uptake and uptake which was insensitive to specific blockers. In particular, insulin's effect on total catecholamine uptake, specific norepinephrine uptake, specific serotonin uptake and. sodium-independent uptake. The findings suggest that insulin may exert neuromodulatory effects directly on neurons by altering the inactivation of synaptic transmitter by reuptake.

PAGE 112

105 Experimental Procedures Uptake of 3 H-monoamines into cultured neurons was determined as described in Chapter II. To determine the effect of insulin on monoamine uptake, cultures were pretreated with various concentrations of unlabeled porcine insulin with or without maximum effective concentrations of uptake inhibitors for 10 min at 37 c. The same concentration of insulin and uptake inhibitors was included in the reaction mixture during the uptake period. The time course of insulin's effect on NE uptake in neuronal cultures was studied by preincubating cultures with a maximally effective concentration (167 nM) for varying times from Oto 60 min. Specific inhibitors used to assay specific norepinephrine and serotonin uptake were rnaprotiline and fluvoxarnine, respectively. Results Preincubation of neuronal cultures with insulin for 10 min resulted in a dose-dependent reduction in total NE uptake with maximum inhibition of 30% of total uptake at 1.67 x 10-7 M insulin (Fig. 31). At this concentration of insulin inhibition of NE uptake was evident by 5 min of preincubation, reached maximum inhibition by 10 min and remained unchanged over a one hour time course (Fig. 32). The specific NE uptake inhibitor, maprotiline, inhibited NE uptake in a dose-dependent manner with maximal inhibition of 30% of total NE uptake occurring at 100 uM rnaprotiline (Fig. 33). Insulin was capable of inhibiting over 95% of maprotiline-

PAGE 113

Figure 31: Effect of insulin on neuronal monoamine uptake. Neuronal cultures were preincubated with the indicased 3 concentrations of insulin for 10 min at 37 C. Hnorepinephrine uptake was then assayed as described in the text. Data are representative of six experiments and presented as mean~ S.E.M.

PAGE 114

107 Cl .... Cl CD U) a> i 0 .... 0 in N .... -C ::::, en C C) 0 I

PAGE 115

Figure 32: Time course of inhibition of neuronal monoamine uptake by insulin. Neuronal cultures were incubated with 16.7 uM insulin for the indicated time periods and monoamine uptake assayed as described in the text. Data are expressed as mean~ S.E.M.

PAGE 116

I I I I I ... I . ) I I -..A-. 0 C\I 1 0 9 I I I 0 ,---0 (0 0 0 C\I 0 Nl310tid 8~/3)fv'ldn 3NltiHd3Nld3tiON-H % -z -w

PAGE 117

Figure 33: Effect of maprotiline on neuronal norepinephrine uptake. Neuronal cultures were incubated with the indicated concentrations of maprotiline, a specific inhibitor of norepinephrine uptake, for 10 min at 37 C. Norepinephrine uptake was assayed as described in the text and data are representative of two experiments and expressed as mean~ S.E.M.

PAGE 118

0 0 ,.... 111 0 LO I e 1 LO C\I 0 3>iv'ldn 3NltlHd3Nld3tlON-H lv'!Ol % V LO (0 -z 0 a: Jz w (.) z 0 (.) w z :J a: a.. < C, 0 ...J I

PAGE 119

112 sensitive uptake (Fig. 34). This effect was statistically significant by 1.67 x 10-11 M with an Ic50 of 8 x 10-11 M. Maximum inhibition occurred at 1.67 x 10-7 M. In addition, in experiments where specific neuronal NE uptake sites wer e maximally inhibited by 100 uM maprotiline, insulin was capable of stimulating maprotiline-insensitive uptake by 30%. The serotonin specific uptake inhibitor, fluvoxamine, inhibited uptake of 3H-serotonin in a dose-dependent manne r with maximal inhibition of 38% of total serotonin uptake occuri ng a t 10 uM (Fig. 35). Insulin stimulated specific serotonin uptake with maximum stimulation of 148% of control uptake occuring wi t h 1.6 X 10-7 M insulin (Fig. 36) Sodium ion in the uptake buffer was replaced with choline ion and NE uptake experiments performed. Removal of sodium ion reduced total NE uptake by 39%. Increasing doses of insulin had no effect on sodium insensitive uptake (Fig. 37). M t l t d f 125 I 1 b. d. d apro i ine compe e or -insu in in ing in a ose dependent manner. Maximum inhibition of insulin binding was observed with 1 uM maprotiline (Fig. 38). Cells were incubated t h 125 I 1 . h f . f wi -insu in int e presence o increasing concentrations o unlabeled porcine insulin with or without 100 uM maprotiline. This competition data was plotted by the method of Scatchard (100) and demonstrated that insulin bindi ng in the presence of 100 uM maprotiline has only one apparent binding site with an affinity (Kd) of 90 nM (Fig. 39).

PAGE 120

Figure 34: Effect of insulin on specific neuronal norepinephrine uptake. Neuronal cultures were incubated with the indicated concentrations of insulin with or without 100 uM maprotiline. Specific neuronal uptake was determined by subtracting nonspecific uptake (in the presence of maprotiline) from total uptake (in the absence of maprotiline). Data are representative of six experiments and presented as mean+ S.E.M.

PAGE 121

11 4 ----4-----+-----+----t----r---""""1 co 0 0 .... C 'a) C co C C CII a) 0 .... -C ::, u, C 8' I

PAGE 122

Figure 35: Effect of fluvoxamine on neuronal serotonin uptake. Neuronal cultures were incubated with the indicated concentrations of fluvoxamine, a sgecific inhibitor of serotonin uptake, for 10 min at 37 c. Serotonin uptake was assayed as described in the text and data are expressed as mean~ S.E.M.

PAGE 123

116 Cf) z 0 a: Iz w (.) z 0 (.) LO w z <( X co 0 > ::::> _J LL ,..... (!} 0 _J I 0 0 0 . ('I T"'" Nl310Hd 8fl\J/3)fv'ldn NIN0l0H3S-H o/o

PAGE 124

Figure 36: Effect of insulin on specific serotonin uptake. Neuronal cultures were incubated with the indicated concentrations of insulin with or without 100 uM fluvoxamine. Specific uptake was determined by subtracting nonspecific uptake (in the presence of fluvoxamine) from total uptake (in the absence of fluvoxamine). Data are representative of two experiments and presented as mean+ S.E.M.

PAGE 125

118 co C :J C/J C 0) 0 I T ~--J _ -----'-'--+-t ~--;-~---~---' ___ I 0 CO CO -::t" C\J 0 0 O 0 0

PAGE 126

Figure 37: Effect of insulin on sodium-independent monoamine uptake. Neuronal cultures were incubated with the indicated concentrations of insulin for 10 min at 37 C in uptake buffer in which sodium ion had been replaced by choline ion. Norepinephrine uptake was assayed as described in the text and data are representative of two experiments and presented as mean~ S.E.M.

PAGE 127

120 co 0) -C: ::i Ult c 0 0) 'r'" 0 I 'r'" 'r'" C\J 'r'"

PAGE 128

Figure 38: Competition by maprotiline for neuronal 1251-ins~~~n binding. Neuronal cultures were incubated with !insulin and the indicated concentrations of maprotiline for 2 hat 37 C. Data are representative of four experiments and expressed as mean+ S.E.M.

PAGE 129

1.0 C, 0.8 z 0 z CD 0.6 (.) u::: (.) w 0.4~ I ..... CL rv en rv -;:R. 0 0.2 0 9 8 7 6 5 4 LOG MAPROTILINE CONCENTRATION (M)

PAGE 130

Figure 39: Scatchard plots of neuronal insulin binding with and withoy25rnaprotiline. Neuronal cultures were incubated with I-insulin and varying concentrations of porcine insulin with (()) and without (~) 100 uM maprotiline. These data were plotted by the method of Scatchar~2 ~nd reveal that 100 uM maprotiline completely blocks !insulin binding to the high affinity neuronal insulin receptor.

PAGE 131

,.... 0 0 .,... X ...... 4 0 3.0 w w a: u. 2 0 -C z ::> 0 m 1.0 O'-----------'----------'---------~~ 0 0.5 1.0 1 5 INSULIN BOUND (PMOL/MG PROTEIN) I-' N J:,.

PAGE 132

125 Discussion Norepinephrine uptake has been well characterized in this culture system. Neuronal NE uptake is sodium, temperature, time and concentration dependent. Uptake is inhibited by specific inhibitors of neuronal NE uptake such as maprotiline and desmethylimipramine. Autoradiographic characterization of this uptake system has shown that NE uptake is specific to a subpopulation of neurons in culture. This implies that cultured neurons maintain specific monoamine uptake mechanisms. The neuronal culture amine uptake systems are identical to the in vivo uptake mechanisms in these respects and is a good model system for investigating modulation of monoamine uptake into neurons. We have used this system to study the effect of insulin on monoamine uptake. Insulin has effects in neuronal cultures at concentrations which suggest that the neuronal insulin receptor mediates certain physiological actions of brain insulin. Insulin inhibits total norepinephrine uptake into neurons at a concentration as low as 1.67 x 10-11 M. This concentration of insulin binds less than 5% of total neuronal insulin receptors which implies that neurons are very sensitive to insulin inhibition of NE uptake. Furthermore, insulin is capable of inhibiting over 95% of specific NE uptake at 1.67 x 10-7 M suggesting that insulin is a very potent inhibitor of neuronal NE uptake. Insulin is also capable of significant stimulation of norepinephrine uptake occurring by other than specific norepinephrine uptake sites. This reflects, at least partially,

PAGE 133

126 stimulation of serotonin uptake by insulin. The effect of insulin on specific dopamine and epinephrine uptake has not been studied yet because of the lack of good pharmacological inhibitors of DA or EPI uptake. However, insulin has no effect on sodium insensitive uptake suggesting that all the effects of insulin seen are due to direct effects on specific uptake mechanisms. Insulin and maprotiline inhibit NE uptake to the same extent. However, insulin is 600 times more potent in inhibiting NE uptake than maprotiline. Maprotiline also competes for high affinity insulin binding and maximally effective doses at blocking NE uptake completely block the high affinity insulin receptor. This is the first demonstration of competition for high affinity insulin binding by a synthetic, pharmacological compound. This suggests that insulin and maprotiline may both be inhibiting NE uptake via interaction with a high affinity insulin receptor. Maprotiline is a clinically effective antidepressant drug with its mechanism of action thought to be related to its inhibition of specific NE uptake. As was noted in Chapter I, insulin had an early history as an antidepressant treatment. When taken with the inhibition of NE uptake by low concentrations of insulin and competition by maprotiline for high affinity insulin receptors these findings suggest that insulin may be an endogenous substance with antidepressant properties. This may have important implications for the development of future

PAGE 134

127 antidepressant treatments as insulin is so much more potent at inhibiting NE uptake than rnaprotiline. More potent antidepressants may be found by studying structure-function relationships of insulin on inhibiting NE uptake and synthetically mimicking the important domain of insulin.

PAGE 135

CHAPTER VI CONCLUSIONS These studies have utilized unique cell culture techniques to investigate the brain insulin system. The brain has classically been considered an insulin independent tissue, largely because of the limited penetration of peripheral insulin beyond the blood-brain barrier. However, within the last seven years, insulin has been isolated from the brain (13, 15-18), insulin binding sites have been demonstrated in the brain (14), and numerous studies have demonstrated in vivo effects of insulin acting in the central nervous system (CNS) (72-85). However, the use of preparations of mixed cells made it difficult to determine whether the components of the brain insulin system were of neuronal or glial origin. A major component of these studies has been the development and standardization of culture techniques which enable the investigator to preferentially grow either neuronal cells or astrocyte glial cells from the newborn rat brain. Neuronal cultures develop extensive networks of neurites with soma of approximately 6-10 um, have spontaneous electrical activity, and 128

PAGE 136

129 possess components of various neurotransmitter systems. Astrocyte glial cultures express the astrocyte glial specific marker, glial fibrillary acidic protein. These cultured cells exhibit many of the characteristics of neuronal or glial cells in vivo and are good models to study possible physiological actions of various agents on either neurons or astrocyte glial cells. However, this is a model system and findings from models must be confirmed, where possible, in the "real", in vivo situation. These studies have demonstrated that both neurons and astrocyte glial cells have specific insulin receptors. These receptors have binding characteristics similar to insulin binding in peripheral tissues. Binding of 1251-insulin to both neurons and astrocyte glial cells is rapid and reversible, specific for insulin, saturable and yield curvilinear Scatchard plots, suggesting two classes of receptors, one of high affinity (Kd=ll nM, neurons, 13.7 nM, glia) and one of low affinity (Kd=65 nM, neurons, 357 nM, glia). Light microscopic autoradiography of 125 1 b' d' 1 1 1 d th 1 I-insu in in ing to neurona cu tures revea e at insu in binding to neuronal cultures was localized to soma and neurites of neuronal cells and not primarily due to binding to the small percentage of glial cells which persist in neuronal cultures. The presence of specific receptors on neurons and astrocyte glial cells suggests that these cell types have preserved mechanisms for recognizing an insulin-like ligand as a signal in the brain. Regulation of neuronal insulin receptor expression is controlled somewhat differently from insulin receptor expression

PAGE 137

130 in the periphery. In neuronal cultures grown for the first three days of culture in 10% FBS, incubation of cultures with high concentrations of insulin does not down regulate insulin receptors. Indeed, insulin binding is increased by 86% with 3 uM insulin. However, in neuronal cultures grown for the first three days in 10% plasma derived horse serum, insulin receptors are down regulated by insulin. This suggests that neurons might have the capability to change their responsiveness to brain insulin early in development. Whether either or both of these responses to insulin exist in vivo in the brain remains to be varified. It has been reported that exposure of mouse cortical brain cells to high concentrations of insulin increases cortical cell insulin binding (103). In addition, inhibition of neuronal glycosylation or protein synthesis does not markedly affect neuronal insulin binding. In peripheral tissues studied inhibition of glycosylation has resulted in decreased insulin binding, presumably by preventing insertion of newly synthesized receptors into the cell membrane. Inhibition of protein synthesis has been reported to result in increased insulin binding in peripheral tissues. This is thought to be due to the inhibition of synthesis of some protein important to the process of internalization, thus resulting in accumulation of receptors on the cell surface. These findings in neuronal cells suggest that glycosylation is not an important step in the processing of newly synthesized neuronal insulin receptor and indeed, the neuronal

PAGE 138

131 receptor may not be glycosylated or may have reduced carbohydrate content. In addition, it appears that cycling of the neuronal receptor is different from cycling of the insulin receptor of peripheral tissues. In contrast, astrocyte glial cells consistently down regulate their insulin receptors in response to high concentrations of insulin, exhibit reduced insulin binding in response to inhibition of glycosylation and increased binding in response to inhibition of protein synthesis. All of these responses are similar to those seen in classical peripheral target tissues of insulin. These differences in the responses of neuronal and astrocyte glial cells with respect to regulation and cycling of insulin receptors are also apparent when other classical actions of insulin are studied. It has been previously reported by Clarke et al. (95) and Boyd et al. (104) that 2-deoxy-glucose (2-dGlc) uptake into astrocyte glial cells in culture is stimulated by insulin in a manner analagous to insulin-stimulated glucose uptake into peripheral target tissues, but neuronal 2-dGlc uptake is not affected by insulin. These studies suggest that neuronal and astrocyte glial insulin receptors are physiologically different, as well as structurally different (differences in glycosylation) and regulated differently. Derek LeRoith and others working at the National Institutes of Health (personal communication) have found that the subunit structure of neuronal and glial insulin receptors are slightly different. The binding subunit of the neuronal insulin receptor

PAGE 139

132 is substantially smaller than the glial a subunit or livera subunit (115K, neuron vs 125K, glia and 135K, liver). This difference in molecular weight is not a function of newborn tissue vs mature tissue as newborn liver insulin receptor was / studied as well as liver from mature animals. This difference in a subunit molecular weight may reflect the reduced carbohydrate of the neuronal receptor. Although the neuronal insulin receptor does not mediate classical actions of insulin such as stimulation of 2-dGlc uptake, it does mediate changes in neurotransmitter systems which may be the basis of many of the actions of insulin in the brain. Insulin inhibits norepinephrine uptake in the same manner as classical antidepressant agents such as maprotiline. Significant inhibition of norepinephrine uptake occurs at 1.67 x 10-11 M insulin, a concentration which occupies fewer than 5% of neuronal insulin receptors, although maximum inhibition occurs in the range of the Ka of the the high affinity insulin receptor. Insulin is 600 times more potent at inhibiting norepinephrine uptake than the specific uptake inhibitor, maprotiline. Maprotiline competes for 125I-insulin binding and 100 uM maprotiline completely blocks 125I-insulin binding to the high affinitiy receptor. These finding suggest that insulin and maprotiline may both be acting to inhibit norepinephrine uptake by binding to a common receptor. Further improvements in antidepressant drugs may result from studying the structurefunction relationships of insulin's inhibition of NE uptake.

PAGE 140

133 Studies using neuronal cultures have satisfied many of the criteria for classifying insulin as a neurotransmitter. Insulin is synthesized in neurons (19). Insulin is released from neurons in response to depolarization (105). There are specific insulin receptors on neurons (Chapter III), which mediate actions of insulin which affect neurotransmission (Chapter V). The anatomical relationships of insulin-containing neurons and neurons with insulin receptors need to be studied in the intact brain, but it seems likely that insulin is a neurotransmitter in the intact brain. In addition, insulin may act locally on glial insulin receptors to regulate local glucose availability and may thus be important in maintaining brain metabolism. These studies suggest that insulin may have significant actions within the CNS.

PAGE 141

APPENDIX Cytosine Arabinoside cytosine arabinoside Dissolve in Phosphate Buffered Saline, pH 7.4. Filter in 20 um filters and freeze. Dilutt~ 1:100 for use. Deoxyribonuclease DNAse Dissolve in Solution D. Filter in 20 um filters and freeze. Use 1.5 ml per 10 brains. 134 grams/liter .2797 .160

PAGE 142

plate. 135 grams/liter Poly-L-lysine Poly-L-lysine .011 Dissolve in H 2o. Filter in 20 um filters and freeze. Plate dishes with 2 ml/30 mm, 4 ml/60 mm, 8 ml/100 mm Trypsin Trypsin 2.5 Dissolve in Solution D. Filter in 20 um filter and freeze in 50 ml aliquots. Use 25 ml per 10 brains.

PAGE 143

136 grams/liter HEPES Buffer HEPES 26.21 NaCl 1. 75 Glucose 1.8 CaC12 0.1 MgC12 0.1 Adjust pH to 7.4 for neuronal binding, 7.8 for glial binding. Must be refrigerated to prevent contamination.

PAGE 144

137 Norepinephrine Uptake Buffer CaC12 MgC12 I
PAGE 145

138 grams/liter Solution D NaCl 8.0 KCl 0.402 Na2HP04 0.02412 KH2P04 0.0299 Glucose 0.99 Sucrose 20.17 Penicillin 1.0 X 10 5 uni ts /1 i ter Streptomycin 0.1 Fungizone 0.00025 Phenol Red 0.2 ml/liter Adjust pH to 7.2 and filter through 20 um filter. Freeze in 500 ml aliquots.

PAGE 146

BIBLIOGRAPHY 1. Humbel, R.E., Bosshard, H.R., and Zahn, H. Chemistry of insulin. In: aandbook of Physiology, Sect. 7: Endocrinology, Vol. I. Endocrine Pancreas. R.O. Greep and E.B. Astwood, Eds., American Physiological Society, Washington, D.C., 1972. 2. Schmidt, P. Uber Organtherapie und insulinbehandlung bei endogenen Geistesstorungen. Klin. Wochenschr., 7:839, 1928. 3. 4. 5. Sakel, M. Neue Behanlunsmethode der Schizophrenie. Verlag Moritz Perles, Vienna. 1935. Prange, A.J., Jr., and Loosen, P.T. Peptides in Depression. In: Frontiers in Biochemical and Pharmacological Research in Depression. E. Usein, Ed. Raven Press, New York, 1984. Czech, M.P. Cellular dynamics of insulin action. Fed. Proc. 41:2717-2718, 1982. 6. Miller, A.L. Brain intermediary metabolism in vivo: Changes with carbon dioxide, development, and seizures. Int. Rev. of Neurobiol. 22:47-82, 1981. 7. Plum, F. What causes infarction in ischemic brain?: The Robert Wartenberg Lecture. Neurology (NY) 33(2):222-233, 1983. 8. Belloff-Chain, A., Catanzaro, R., Chain, E.B.,1,asi, I., and Pocchiari, F. Fate of uniformly labelled C-glucose in brain slices. Proc. Roy. Soc. B., 144:22-28, 1956. 9. Sloviter, H.A., and Yamada, H. Absence of direct action of insulin on metabolism of the isolated perfused rat brain. J. Neurochem. 18:1269-1274, 1971. 10. Goodner, C.J., and Berrie, M.A. hypothalamic tissues to take up to respond to insulin in vivo. 1977. The failure of rat labeled insulin in vivo or Endocrin. 101:605-612, 11. Hom, F.G., Goodner, C.J., and Berrie, M. A (3H)-2-deoxyglucose method for comparing rates of glucose metabolism and insulin responses among rat tissues in vivo. Diabetes 33(2) :141-152, 1984. 139

PAGE 147

140 12. Morgan, H.E. Hormonal control of the pancreatic islets. In: Best and Taylor's Physiological Basis of Medical Practice, 10th Ed. John R. Brobeck, Ed. The Williams and Wilkins Company, Baltimore, pp. 77-80, 1979. 13. Havrankova, J., Schrnechel, D., Roth, J., and Brownstein, M. Identification of insulin in rat brain. Proc. Natl. Acad. Sci. USA 75:5737-5741, 1978. 14. Havrankova, J., Roth, J., and Brownstein, M. Insulin receptors are widely distributed in the central nervous system of the rat. Nature 272: 827-829, 1978. 15. Havrankova, J., Roth, J., and Brownstein, M. Concentrations of insulin and of insulin receptors in the brain are independent of peripheral insulin levels. J. Clin. Invest. 64: 636-642, 1979. 16. Raizada, M.K. Localization of insulin-like irnrnunoreactivity in the neurons from primary cultures of rat brain. Exp. Cell Res. 143: 351-357, 1983. 17. Birch, N.P., Christie, D.L., and Renwick, A.G.C. Irnrnunoreactive insulin from mouse brain cells in culture and whole rat brain. Biochem. J. 218: 19-27, 1984. 18. Raizada, M.K., Stamler, J.F., Quinlan, J.T., Landas, S. and Phillips, M.I. Insulin receptor containing cells in primary cultures from rat brain. Cell. Mol. Neurobiol. 2: 47-52, 1982. 19. Cooper, D.L., Clarke, D.W., Raizada, M.K., Stein, J.L., and Stein, G.S. Arn. J. Physiol. In press. 20. Cuatrecasas, P. Insulin receptor interactions in adipose tissue cells: Direct measurement and properties. Proc. Natl. Acad. Sci. USA. 68:1264-1268, 1971. 21. Gavin, J.R.,III, Mann, D.L., Buell, D.N., and Roth, J. Preparation of solubilized insulin receptors from human lymphocytes. Biochem. Biophys. Res. Comm. 49: 870-876, 1972. 22. Gavin, J.R., Ill, Roth, J., Jen, P., and Freychet, P. Insulin receptors in human circulating cells and fibroblasts. Proc. Natl. Acad. Sci. USA. 69:747-751, 1972. 23. Raizada, M.K. and Perdue, J.F. Mitogen receptors in chick embryo fibrob},~ts: Kinetics, specificity, unmasking and synthesis of I-insulin binding sites. J. Biol. Chem. 251:6445-6455, 1976.

PAGE 148

141 24. Ginsberg, B.H., Kahn, C.R., and Roth, J. The insulin receptors of the turkey erythrocyte: Similarity to mammalian insulin receptors. Endocrinology 100:82-90, 1977. 25. Olefsky, J.M., Joh~gn, J., Liu, F., Edwards, P. and Baur, s. Comparison of I-insulin binding and degradation to isolated rat hepatocytes and liver membranes. Diabetes 24:801-810, 1975. 26. Freychet, P., Roth J., and Ney~5le, D.M. Insulin receptors in liver-specific binding of I-insulin to plasma membranes and its relation to bioactivity. Proc. Natl. Acad. Sci. USA. 68:1833-1837, 1971. 27. Santora, A.C., II, Wheeler, F.B., DeHaan, R.L. and Elsas, L.J. Relationship of insulin binding to amino acid transport by cultured 14 day embryonic chick heart cells. Endocrinology 104:1059-1068, 1979. 28. ,~zada, M.K., Yang, J.W. and Fellows, R.E. Binding of I-insulin to specific receptors and stimulation of nucleotide incorporation in cells cultured from rat brain. Brain Res. 200:389-400, 1980. 29. Jacobs, s., and Cuatrecasas, P. Insulin Receptor:Structure and Function. Endocrine Reviews 2(3) :251-263, 1981. 30. Jacobs, S., Shechter, Y., Bissel, K., and Cuatrecasas, P. Purification and properties of insulin receptors from rat liver membranes. Biochem. Biophys. Res Commun. 77:981-988, 1977. 31. Kasuga, M., Hedo, J.A., Yamada, K.M., and Kahn, C.R. The structure of insulin receptor and its subunits: evidence for multiple nonreduced forms and a 210,000 possible proreceptor. J. Biol. Chem. 257(17) :10392-10399, 1982. 32. Gammeltoft, S. Insulin receptors: binding kinetics and structure-function relationship of insulin. Physiological Reviews 64(4) :1321-1378, 1984. 33. Kasuga, M., Karlsson, F.A., and Kahn, C.R. Insulin stimulates the phosphorylation of the 95,000-dalton subunit of its own receptor. Science 215:185-187, 1982. 34. Kasuga, M., Zick, Y., Blith, D.L., Karlsson, F.A., Haring, H.U., and Kahn, C.R. Insulin stimulation of phosphorylation of the B subunit of the insulin receptor: Formation of both phosphoserine and phosphotyrosine. J. Biol. Chem. 257(17) :9891-9894, 1982.

PAGE 149

142 35. Cuatrecasas, P. and Tell, G.P.E. Insulin-like activity of conconavalin A and wheat germ agglutinin: direct interactions with insulin receptors. Proc. Natl. Acad. Sci. USA. 70:485-489, 1973. 36. Rosen, O.M., Chia, G.H. and Rubin, C.S. Tunicamycin mediated depletion of insulin receptors in 3T3-Ll adipocytes. J. Cell Physiol. 99:37-42, 1979. 37. Hedo, J.A., Kasuga, M., Van Obberghen, E., Roth, J. and Kahn, C.R. Direct demonstration of glycosylat.ion of insulin receptor subunits by biosynthetic and external labeling:evidence for heterogeneity. Proc. Natl. Acad. Sci. 78:4791-4795, 1981. 38. Stauffacher, w., Orci, L., Cameron, D, Burr, I.M. and Ronold, A.E. Spontaneous hyperglycemia and/or obesity in laboratory rodents: an example of the possible usefulness of animal disease models with both genetic and genetic and environmental components. Rec. Prog. Horm. Res. 27:41-95, 1971. 39. Kadle, R., Regulation of insulin receptors in fibroblastic cultures from nondiabetic and diabetic mice. Thesis submitted to the Graduate College of the University of Iowa, 1983. 40. Archer, J.A., Gorden, P. and Roth, J. Defect in insulin binding to receptors in obese man. Amelioration with calorie restriction. J. Clin. Invest. 55:166-174, 1975. 41. Roth, J., Kahn, C.R., Lesniak, M.A., Gorden, P., Megyesi, K., Neville, D.M., Jr., Gavin, J.R III, Sall, A.M., Freychet, P., Goldfine, I.D., Bar, R.S. and Archer, J. Receptors for insulin, NSILA-S and growth hormone: application to disease states in man. Rec. Prog. Horm. Res. 31:95-139, 1975. 42. Gavin, J.R., III, Gorden, P., Roth, J., Archer, J.A. and Buell, D.N. Characteristics of the human lymphocyte insulin receptor. J. Biol. Chem. 248(6):2202-2207, 1973. 43. Marshall, S. and Olefsky, J.M. Characterization of insulininduced receptor loss and evidence for internalization of the insulin receptor. Diabetes 30:746-753, 1981. 44. Petersen, B., Beckner, S. and Blecher, M. Characteristics of insulin receptors in a new line of cloned neonatal rat hepatocytes. Biochim. Biophys. Acta 542:470-485, 1978. 45. Mott, D.M., Howard, B.V. and Bennett, P.H. Stoichiometric binding and regulation of insulin receptors on human diploid fibroblasts using physiologic insulin levels. J. Biol. Chem. 254(18) :8762-8767, 1979.

PAGE 150

143 46. Heaton, J.H. and Gelehrter, T.D. Desensitization of hepatoma cells to insulin action. Evidence for a postreceptor mechanism. J. Biol. Chem. 256:12257-12262, 1981. 47. Kosmakos, F. and Roth, J. Insulin-induced loss of the insulin receptor in IM-9 lymphocytes: a biological process mediated through the insulin receptor. J. Biol. Chem. 255(20) :9860-9869, 1980. 48. Harrison, L.C., Martin, F.I.R., and Melick, R. A. Correlation between insulin receptor binding in isolated fat cells and insulin sensitivity in obese human subjects. J. Clin. Invest. 58:1435-1441, 1976. 49. Prince, M.J., Baldwin, D., Tsai, P. and Olefsky, J.M. Regulation of insulin receptor in cultured human fibroblasts. Endocrinol. 109:1754-1759, 1981. 50. Stevens, R.L., Austen, K.F., and Nissley, S.P. Insulininduced increase in insulin binding to cultured chondrosarcoma chondrocytes. J. Biol. Chem. 258(5) :2940-2944, 1983. 51. Reed, B.C., Kaufmann, S.H., Mackall, J.C., Student, A.K. and Lane, M.D. Alterations in insulin binding accompanying differentiation of 373-11 preadipocytes. Proc. Natl. Acad. Sci. USA. 74:4876-4880, 1977. 52. Heise, E., Joost, H.G. and Hasselblatt, A. Insulin binding and response to insulin of adipocytes from thyroxinetreated rats. Endocrinol. 110:955-960, 1982. 53. DeRuyter, H., Burman, K.D., Wartofsky, L. and Taylor, S.I. Effects of thyroid hormone on the insulin receptor in rat liver membranes. Endocrinol. 110:1922-1925, 1982. 54. Knutson, V.P., Ronnett, G.V. and Lane, M.D. Control of insulin receptor level in 3T3 cells: Effect of insulininduced down-regulation and dexamethasone-induced upregulation on rate of receptor inactivation. Proc. Natl. Acad. Sci. USA. 79:2822-2826, 1982. 55. Fantus, I.G., Saviolakis, G.A., Hedo, J.A. and Gorden, P. Mechanism of glucocorticoid-induced increase in insulin receptors of cultured human lymphocytes. J. Biol. Chem. 257(14) :8277-8283, 1982. 56. Vale, R.D., De Lean, A., Lefkowitz, R.J. and Stadel, J.M. Regulation of insulin receptors in frog erythrocytes by insulin and concanavalin A. Mol. Pharmacol. 22:619-626, 1982.

PAGE 151

144 57. Steinfelder, H.J. and Joost, H.G. In vitro effects of theophylline on insulin receptors in adipocytes: correlation with the lipolytic action of the agent. Biochem. Biophys. Res. Commm. 104:45-51, 1982. 58. Joost, H.-G. and Steinfelder, H.-J. Modulation of insulin sensitivity by adenosine. Effects on glucose transport, lipid synthesis, and insulin receptors of the adipocyte. Mol. Pharmacol. 22:614-618, 1982. 59. Karlsson, F.A., Grunfeld, C., Kahn, C.R. and Roth, J. Regulation of insulin receptors and insulin responsiveness in 3T3-Ll fatty fibroblasts. Endocrinol. 104:1383-1392, 1979. 60. Fernstrom, J.D. and Wurtman, R.J. Brain serotonin content:physiological dependence on plasma tryptophan levels. Science 173:149-152, 1971. 61. Tagliamonte, A., De Montis, M.G., Olianas, M.C., Onali, P.L. and Gessa, G.L. Possible role of insulin in the transport of tyrosine and tryptophan from blood to brain. Adv. Exp. Med. Biol. 69:89-94, 1976. 62. De Montis, M.G., Olianas, M.C., Haber, B. and Tagliamonte, A. Increase in large neutral amino acid transport into brain by insulin. J. Neurochem. 30:121-124, 1978. 63. Lipsett, D., Madras, B.K., wurtman, R,.J. and Munro, H.N. Serum tryptophan level after carbohydrate ingestion: Selective decline in non-albumin-bound tryptophan coincident with reduction in serum free fatty acids. Life Sci. 12:57-64, 1973. 64. Pratt, O.E. Kinetics of tryptophan transport across the blood-brain barrier. J. Neural Trans. Suppl. 15:29-43, 1979. 65. Daniel, P.M., Love, E.R., Moorhouse, S.R. and Pratt, O.E. Effect of insulin upon levels of amino acids in the blood and upon the influx of tryptophan into the brain. J. Physiol. 289:87P-88P, 1978. 66. Cooper, J.R.; Bloom, F.E., and Roth, R.H. The Biochemical Basis of Neuropharmacology. Oxford University Press, New York, 1978. 67. WUrtman, R.J., Larin, F., Mostafapour, S., and Fernstrom, J.D. Brain catechol synthesis: control by brain tyrosine concentration. Science 185:183-184, 1974. 68. Fernstrom, J.D. and Wurtman, R.J. Brain serotonin content: increase following ingestion of carbohydrate diet. Science 174:1023-1025, 1971.

PAGE 152

145 69. Fernstrom, J.D. Effects of the diet on brain neurotransmitters. Metabolism 26:207-223, 1977. 70. Sved, A.F., Fernstrom, J.D., and Wurtman, R.J. Tyrosine adminstration reduces blood pressure and enhances brain norepinephrine release in spontaneously-hypertensive rats. Proc. Natl. Acad. Sci, USA 76:3511-3514, 1979. 71. Gibson, C.J. and Wurtman, R.J. Physiological control of brain norepinephrine synthesis by brain tyrosine concentration. Life Sci. 22:1399-1406, 1978. 72. Van Houten, M. and Posner, B.I. Cellular basis of direct insulin action in the central nervous system. Diabetologia 20:255-267, 1981. 73. Oomura, Y. Significance of glucose, insulin and free fatty acids on the hypothalamic feeding and satiety neurons. In: Hunger: Basic mechanisms and clinical implications. Sacher, E.J., Ed. Raven Press, New York, pp. 145-157, 1976. 74. Woods, S.C., Lotter, E.C., McKay, L.D. and Porte, D., Jr. Chronic intracerebroventricular infusion of insulin reduces food intake and body weight of baboons. Nature 282:503-505, 1979. 75. Van Houten, M., Posner, B.I., Kopriwa, B.M., and Brawer, J.R. Insulin binding sites localized to nerve terminals in rat median eminence and arcuate nucleus. Science 207:1081-1083, 1980. 76. Hatfield, J.S., Millard, W.J., and Smith, C.J.V. Shortterm influence of intraventromedial hypothalamic administration of insulin in normal and diabetic rats. Pharmacol. Biochem. Behav. 2:223-226, 1974. 77. Steffans, A.B., Mogensen, G.J. and Stevenson, J.A.F., Boood glucose, insulin and free fatty acids after stimulations and lesions of the hypothalamus. Am. J. Physiol. 22:1446-1462, 1972. 78. Ricardo, J.A., and Koh, E.T. Anatomical evidence of direct projections from the nucleus of the solitary tract to the hypothalamus, amygdala and other forebrain structures in the rat. Brain Res. 153:1-26, 1978. 79. Woods, s.c. and Porte,D., Jr., Neural control of the endocrine pancreas. Physiol. Rev. 54:596-619, 1974. 80. Danguir, J. and Nicolaidis, s. Sleep and feeding patterns in the ventromedial hypothalamic lesioned rat. Physiol. Behav. 21:769-777, 1978.

PAGE 153

146 81. Danguir, J. and Nicolaidis, S., Chronic intracerebroventricular infusion of insulin causes selective increase of slow wave sleep in rats. Brain Res. 306:97-103, 1984. 82. Barbaccia, M., Chuang, D. and Costa, E. Is insulin a neuromodulator? Adv. Biochem Psychopharm. 33, 511-518, 1982. 83. 84. 85. 86. 87. 88. 89. Sauter, A., Goldstein, M., Engel, J. and Ueta, K. of insulin on central catecholamines. Brain Res. 333, 1983. Effect 260:330-Saller, C.F. and Chiodo, L.A. Glucose suppresses basal firing and galoperidol-induced increases in the firing rate of central dopaminergic neurons. Science, 210:1269-1271, 1980. Palovcik, R.A., Phillips, M.I., Kappy, M.S., and Raizada, M.K. Insulin inhibits pyramidal neurons in hippocampal slices. Brain Res. 309:187-191, 1984. Palay, S.L. and Chan-Palay, V. General morphology of neurons and neuroglia. In: Handbook of Physiology. Section 1: The Nervous System, Vol. I. Cellular Biology of Neurons, Part 1. Brookhart, J.M., Mountcastle, V.B., Kanel, E.R., and Geiger, S.R., Eds. Bethesda, Md., American Physiological Society, pp. 5-37, 1977. Henn, F.A., and Hamberger, A. Glial cell function: Uptake of transmitter substances. Proc. Natl. Acad. Sci. 68:2686-2690, 1971. Snodgrass, S.R., and Iversen, L.L. Amino acid uptake into human brain tumors. Brain Res. 76:95-107, 1974. Hutchison, H.T., Werrback, K., Uptake of neurotransmitters by and neuroblastoma in culture. aminobutyric acid. Brain Res. Vance, c., and Haber, B. clonal lines of astrocytoma I. Transport of gamma-66:265-274, 1974. 90. Kanazawa, I., Iverson, L.L., and Kelly, J.S. Glutamate decarboxylase activity in the rat posterior pituitary, pineal gland, dorsal root ganglion and superior cervical ganglion. J. Neurochem. 27:1267-1269. 1976. 91. Schrier, B.K., and Thompson, E.J. On the role of glial cells in the mammalian nervous system. Uptake, excretion, and metabolism of putative neurotransmitters by cultured glial tumor cells. J. Biol. Chem. 249:1769-1780, 1974. 92. Varon, S.S., and Somjen, G.G. Neurosciences Res. Prog. Bull. Neuron-glial interactions. 17(1):1-239, 1979.

PAGE 154

147 93. Freychet, P., Roth, J., and Neville, D.M. Monoiodoinsulin: Demonstration of its biological activity and binding to fat cells and liver membraines. Biochem. Biophys. Res. Comm. 43:400-408, 1971. 94. Lesniak, M. Iodination of insulin. In: Seventh Training Course in Radio Ligand Assay Techniques. Endocrine Society, Bethesda, MD. pp. 57-58, 1978. 95. Clarke, D.W., Boyd, F.T., Jr., Kappy, M.S., and Raizada, M.K. Insulin binds to specific receptors and stimulates 2-deoxy D-glucose uptake in cultured glial cells from rat brain. J. Biol. Chem. 259:11672-11675, 1984. 96. Raizada, M.K., Tan, G. G., and Fellows, R.E. Fibroblastic cultures from the diabetic db/db mouse: demonstration of decreased insulin receptors and impaired responses to insulin. J. Biol. Chem. 255:9149-9155, 1980. 97. Lowry, O.H., Rosebrough, N.J., Farr, A.L., and Randall, R.J. Protein measurement with the Folin phenol reagent. J. Biol. Chem. 193:265-275, 1951. 98. Rogers, A.W. Techniques of Autoradiography. Elsevier/North Holland, Amsterdam pp. 368-370, 1979. 99. Sumners, c., Muther, T.F., and Raizada, M.K. Altered norepinephrine uptake in neuronal cultures from Spontaneously Hypertensive rat brain. Arn. J. Physiol. 248:c488-c497, 1985. 100. Scatchard, G. The attractions of proteins for small molecules and ions. Ann. N.Y. Acad. Sci. 51:660-672, 1949. 101. Gammeltoft, S., Staun-Olsen, P., Fahrenkrug, J., Ottesen, B. Insulin receptors in liver and brain: Complexity of binding kinetics and neurocrine function of insulin. In: Serono Symposium No. 41. Current Views on Insulin Receptors. Andreani,D., de Pirro, R., Lauro,R., Olefsky, J., and Roth J., Eds. New York, Academic Press, pp. 275-289, 1981. 102. Sumners, C., Phillips, M.I. and Raizada, M.K. Rat brain cells in primary culture: Visualization and measurement of catecholamines. Brain Res. 264:267-275, 1983. 103. Van Schravendijk, C.F.H., Hooghe-Peters, E.L., De Meyts, P., and Pipeleers, D.F. Identification and characterization of insulin receptors on foetal-mouse brain-cortical cells. Biochem. J. 220:165-172, 1984.

PAGE 155

148 104. Boyd, F.T., Jr., Clarke, D.W., and Raizada, M.K., Insulin receptors and modulation of norepinephrine uptake by insulin in neuronal cells in culture. J. Biol. Chem. In Press. 105. Clarke, D.W., Boyd, F.T., and Raizada, M.K. Insulin is released from cultured neurons by depolarization. Submitted.

PAGE 157

I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Mohan K. Raizada, Chairman Associate Professo r of Physiology I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Si~ COA~ Professor of Physiology I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. M. Ian Phillips Professor of Physiology I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. (7 ,.1!. /', 'I (_.J?,, 1 J fr': 1 J-(l. ( I I f Edwin M Meyer Assistant Professor of Pharmacology and Therapeutics

PAGE 158

I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. QL -~:~-~/-_-.J;..----_ Colin Sumners Assistant Professor of Physiology This dissertation was submitted to the Graduate Faculty of the College of Medicine and to the Graduate School and was accepted as partial fulfillment of the requirements for the degree of Doctor of Philosophy. August, 1985