INFORMATiON TO USERS

This manuscript has been reproduced from the microfilm master. UMI films the text directly from the original or copy submitted. Thus, some thesis and dissertation copies are in typewriter face, while others may be from any type of computer printer.

The quality of this reproduction is dependent upon the quality of the copy submitted. Broken or indistinct print, colored or poor quality illustrations and photographs, print bleedthrough, substandard margins, and improper alignment can adversely affect reproduction.

In the unlikely event that the author did not send UMI a complete manuscript and there are missing pages, these will be noted. Also, if unauthorized copyright material had to be removed, a note will indicate the deletion.

Oversize materials (e.g., maps, drawings, charts) are reproduced by sectioning the original, beginning at the upper left-hand comer and continuing from left to right in equal sections with small overlaps. Each original is also photographed in one exposure and is included in reduced form at the back of the book.

Photographs included in the original manuscript have been reproduced xerographically in this copy. Higher quality 6" x 9" black and white photographic prints are available for any photographs or illustrations appearing in this copy for an additional charge. Contact UMI directly to order.

Bell & Howell Information and Learning 300 North Zeeb Road, Ann Arbor, MI 48106-1346 USA 800-521-0600

THE ROLES OF -LIKE -1 (GLP-1) IN TEE MOUSE BRAIN

Julie Kim

A thesis submitted in conformity with the requirements for the degree of Master of Science Graduate Department of Physiology University of Toronto

O Copyright by Julie Kim (1 998) National Library Bibliotheque nationale 1+1 , du Canada Acquisitions and Acquisitions et Bibliographic Services services bibliographiques 395 Wellington Street 395, rue Wellington Ottawa ON KIA ON4 Ottawa ON KIA ON4 Canada Canada your lire Vorre refence

Our Ne Notre refdrence

The author has granted a non- L'auteur a accorde une licence non exclusive licence allowing the exclusive pernettant a la National Library of Canada to Bibliotheque nationale du Canada de reproduce, loan, distribute or sell reproduire, preter, distribuer ou copies of this thesis in microfom, vendre des copies de cette these sous paper or electronic formats. la forme de microfiche/fh, de reproduction sur papier ou sur format electronique.

The author retains ownership of the L'auteur conserve la propriete du copyright in this thesis. Neither the droit d'auteur qui protege cette these. thesis nor substantial extracts fiom it Ni la these ni des extraits substantiek may be printed or otherwise de celle-ci ne doivent 6eimprimes reproduced without the author's ou autrernent reproduits sans son permission. autorisation. THE ROLES OF GLUCAGON-LIKE PEPTIDE-1 (GLP-1) IN THE MOUSE BRAIN

Master of Science, 1998 Julie Kim Department of Physiology, University of Toronto

ABSTRACT

Glucagon-like peptide- 1 (GLP- 1) is a gut-brain synthesized by peptidergic cleavage of in the pancreatic A cells and the L cells of the ileal mucosa. GLP-1 is involved in the regulation of pancreatic and gastric functions, stimulating secretion and inhibiting gastric secretion, respectively. Although GLP-1 has been postulated to regulate food intake, mice with targeted disruption of the GLP- 1 receptor (GLP-1 R -/- mice) exhibit normal body weights, suggesting that GLP-1 is not an exclusive determinant of the central mechanisms controlling appetite. The objective of this thesis was to further explore potential central sites of GLP-1 action by mapping the distribution of its receptors in the mouse brain as well as by studying the endocrine consequences of the loss of GLP-1 receptor mediated responses, in the GLP-IR -1- mouse. Quantitative autoradiography demonstrates extensive distribution of GLP-I receptors in the diencephalon and in the limbic system, suggesting that rather than specifically regulating appetite, GLP-I may in fact play a role in a number of neuronal systems that influence neuroendocrine hction and behavior. In studying GLP-1R -1- mice from birth to adulthood, we found that while they appear phenotypically similar from the control wildtype mice (GLP- 1R +I+), their reproductive development is subtly impaired. In addition, loss of GLP-I signaling results in slight deficits in their stress responses. Responses to mild stress are subnormal, although responses to severe stress remain intact in GLP-1R -/- animals. These data suggest that GLP-1 may play a subtle modulatory role in the CNS in regulating neuroendocrine mechanisms involved in regulating reproductive function and the control of adrenal function, in addition to its postulated involvement in the regulation of food intake. Several people have been cited for having made contributions to this thesis: Drs. Daniel J. Drucker and Louise A. Scrocchi were responsible for generating mice with targeted disruption of the GLP- I receptor and also for supplying both wildtype and GLP-1R -/- mice; Dr. Patricia L. Brubaker synthesized radiolabeled [125~]-~~~-1-(7-36)and ['25~]-exendin-4; Dr. Richard Hochberg provided radiolabeled 1 1P-rnethoxy 16a[125~]-iodo WE2); Dr. Sylvia Asa performed histological examinations. In addition, there are a number of people I wish to acknowlegde for making this thesis possible. First, I owe a debt of gratitude to my supervisor Dr. Neil J. MacLusky for his instructed guidance and constant encouragement. I would also like to thank Dr. Theodore J. Brown for his generosity of time and helpfid assistance in the lab. To Alexandra Kollara and Andreas Evangelou, I owe special thanks for their friendship and unlimited supply of energy- boosting Hersheys hugs & kisses. To Deboleena Roy and Shirya Rashid, I thank them for all their encouragement. I also thank Heather Edwards and Arawn Therrien for their practical wisdom and amusing insights that men are from mars. A special thanks to my cousidroommate Moon-Sook Kyun for Girlfriends' Nites and for always being there for me, especially when soot almost gave us the BIG boot! Special thanks to Davey Chung for his unparalleled sense of humor, sushi and Garth Brooks that only he can sing to. To my faithfid friend, Cei-Whan Kim, I thank him for 'peas & carrots', for all his prayers, and especially for his classical guitar playing - Cei, you've got it nailed! To Byron Carter, Shelley MacDonald and the rest of my family at the Toronto Baptist Fellowship, I thank them for all their prayers and support. Finally, to my parents, Dr. Kwang Oh Kim and Myung Ja Paik, I owe them a debt of gratitude for being unfkilingly supportive and immensely loving. And to Jurny, Juan and Justin, my awesome .lK3 siblings, I thank God for blessing me with them. Last but certainly not least, I give eternal thankfulness to God for being my FaiWl Rock and for guiding me to The Meeting Place (Toronto)!

(iv.) Emulsion A utoradiography...... 22 2.5 DISCUSSION ...... 42

GLP-1R -/- MOUSE MODEL ...... 45 PRELIMINARY OBSERVATIONS IN GLP-I R -/- MICE...... 45 OBJECTIVE...... 46 MATERIALS AND METHODS ...... -47 (i.) Animal Breeding ...... 47 (ii.) Vaginal Opening Inspection ...... 47 ... (rrr.) Tissue Dissection ...... -47 (iv.) GLP-I Receptor Autoradiography ...... 48 (v.) Gonadal Histology ...... 48 (vi.) Radioirnrnunoassay ...... -48 .. (vzr.) Statistical Analysis ...... -49 RESULTS ...... 50 (i.) Deficits in GLP-I R -1- Reproductive Functions...... 50 (ii.) Examining GLP-I R -1- Testes and ...... 51 (iii.) Esrradiol. . and Corticosterone Radioimmunoassuy...... 51 DISCUSSION ...... 66

CHAPTER LV . lRlPLICATIONS OF GLP-1 IN THE CONTROL OF APPETITE...... ,...... 70 ROLE OF CCK-8 IN APPETITE CONTROL ...... 70 OBJECTIVE...... 72 MATERIALS AND METHODS...... 73 (i.) Chemicals...... -73 (ii.) Animal Preparation ...... -73 ... .. (rrr.) Tissue Sectzonzng...... 73 (iv.) CCK-8 Receptor A uiorndiography...... -73 (v.) Computer-assisted Image Analysis of Receptor Autoradiograms ...... 74 (vi) Statistical Analysis ...... -74 RESULTS ...... 75 (i.) Summary of CCK-8 Receptor Distribution ...... -75 (ii.) Comparison of CCK-8 Receptor Concentrations...... 75 DISCUSSION ...... 80 CHAPTER V .SEXUAL DIFFERENTIATION IN TEIE GLP-1R 4- MOUSE BRAIN...... 82 5.1 SEXUAL DIFFERENTIATION OF THE CNS...... 82 5.2 OBJECTIVE ...... -84 5.3 MATERIALS AND METHODS ...... 85 (i.) Chemicals...... 85 .. (11.) Animal Preparation ...... 85 ... . . (r z z.) Tissue Sectzonmg ...... -85 (iv.) Receptor A utoradiopphy ...... -86 (v.) Statistical Analysis ...... 86 5.4 RESULTS ...... 87 5.5 DISCUSSION ...... 92

OVERALL DISCUSSION...... 93 REFERENCES...... -97 FrGURE 1 Schematic representation of the post-translational processing of the proglucagon molecule...... -.....2

FIGURE 2 Model depicting the proposed physiologic actions of GLP-1 ...... 5

FIGURE 3 Model depicting the proposed GLP-1 signal transduction pathways in the pancreatic P cell ...... -..- .. .-... -... .7

FIGURE 4 Representative of the glucagon superfamily of ...... 15

FIGURE 5 Autoradiograms demonstrating identical binding patterns of radioiodinated GLP- 1 and exendin-4 in coronal mouse brain sections...... 24

FIGURE 6 Autoradiograms demonstrating identical binding patterns of radioiodinated GLP- 1 and exendin-4 in coronal mouse brain sections...... 25

FIGURE 7 Autoradiograms demonstrating binding patterns of radioiodinated [12'1]-~~~-1 in sagittal mouse brain sections taken from wildtype GLP- 1R +/+ and GLP- 1R -/- mice ...... 26

FIGURE 8 Photomicrograph of ['*'I] exendin-4 binding in the dentate gyrus of the hippocampus of an adult male mouse ...... 27

FIGURE 9 Photomicrograph of [Iz51] exendin-4 binding in the medial habenula of an adult male mouse ...... 28

FIGURE 10 Distribution of GLP-1R binding sites in forebrain of male and female GLP- 1R +I+ mice ...... 29

FIGURE 11 Distribution of GLP-1R binding sites in hindbrain of male and female GLP- I R +/+ mice ...... -. -...... - -3 0

FIGURE 12 Distribution of GLP- 1 R binding sites in diencephalon of male and female GLP- f R +I+ mice...... , ...... 3 1

FIGURE 13 Distribution of GLP- 1R binding sites in diencephalon of male and female GLP- 1R +I+ mice ...... -...... 32

FIGURE 14 Distribution of GLP- 1R binding sites in limbic system of male and female GLP- 1R +I+ mice...... - ...... 3 3

vii FIGURE 15 Distribution of GLP-IR binding sites in limbic system of male and female GLP-1R +I+ mice ...... 34

FIGURE 16 Series of autoradiograms demonstrating identical [12'~]-~~~-1 binding patterns in intact male and gonadectomized male rat brain sections...... -36

FIGURE 17 Distribution and comparison of GLP- I R binding sites measured in forebrain and hindbrain of intact and gonadectornized male rats ...... 37

FIGURE 18 Distribution and comparison of GLP- I R binding sites measured in diencephalon of intact and gonadectomized male rats ...... 38

FIGW 19 Distribution and comparison of GLP- 1R binding sites measured in diencephalon of intact and gonadectomized male rats ...... 39

FIGURE 20 Distribution and comparison of GLP- I R binding sites measured in limbic system of intact and gonadectomized male rats ...... 40

FIGURE 2 1 Distribution and comparison of GLP- 1R binding sites measured in limbic system of intact and gonadectomized male rats ...... 41

FIGURE 22 Profile of body weight gain in male and female GLP- 1R +/+ and GLP- I R -1- mice...... 53

FIGURE 23 Ages in days at onset of pubertal maturation in female GLP- 1R +/+ and GLP- I R -/- mice ......

FIGURE 24 Composite graph showing TBW, pituitary gland and weights of GLP- 1R +I+ and GLP- 1R -1- mice ......

FIGURE 25 Composite graph showing mean ovarian and uterine weights of proestrous GLP- 1R +I+ and GLP- I R -1- females at 6 weeks of age...... 5 7

FIGURE 26 Composite graph showing mean weights of testes, prostate, epididymis and seminal vesicle taken fiom male GLP- 1R +/+ and GLP- 1R -1- mice at 6 weeks of age...... 58

FIGURE 27 Number of developing follicles and corpora Iutea per fiom female GLP- 1R +I+ and GLP- 1R -1- mice...... 59

FIGURE 28 Composite graph showing mean semestradiol levels in proestrous and diestrous female GLP- 1R +I+ and GLP- 1R -1- mice...... 6 1 FIGURE 29 Composite graph showing mean serum testosterone levels in male GLP- 1R +/+ and GLP- 1R -/- under different stress conditions and with 1 rn& glucocorticoid treatment ...... 63

FIGURE 30 Composite graph showing mean serum corticosterone levels in male and female GLP- I R +/+ and GLP- 1R -/- mice under different stress conditions and with 1 mg/d glucocorticoid treatment...... 65

FIGURE 3 1 Autoradiograms depicting no significant difference in CCK-8 receptor levels between GLP-1 R +/+ and GLP- 1R -1- mouse brain section ...... 77

FIGURE 32 Comparison of CCK-SR levels measured in discrete brain regions of male and female GLP-1 R +/+ and GLP-1R 4-mice ...... 78

FIGURE 33 Comparison of CCK-8R levels measured in discrete brain regions of male and female GLP- 1R +/+ and GLP- 1R -1- mice...... 79

FIGURE 34 Autoradiograrn depicting no significant difference in estrogen receptor levels between male and femaIe GLP-1 R +/+ and GLP-1 R -/- mouse preoptic area brain sections...... 89

FIGURE 35 Autoradiogram depicting no significant difference in estrogen receptor levels between male and female GLP- 1R +/+ and GLP- 1R -/- mouse ventromedial nucleus brain sections...... 90

FIGURE 36 Comparison of estrogen receptor levels measured in discrete brain regions of male and female GLP- 1R +/+ and GLP- 1R -/- mice ...... 9 1 TABLE 1 Quantification of [12S~]-~~~-1 binding in intact and gonadectomized GLP- 1R +/+ mice at different regions of the brain ...... 23

TABLE 2 Quantification of [12'~]-~~~-1binding in intact and gonadectomized male rats at different levels of the brain ...... 35

TABLE 3 Total body weight and dissected tissue weight measurements taken from female and male GLP- 1R +/+ and GLP- 1R -1- mice...... 55

TABLE 4 Table summarizing differences in serum estradiol levels in proestrous and diestrous female GLP- 1R +/+ and GLP- I R -1- mice ...... -60

TABLE 5 Table summarizing differences in serum testosterone levels in adult male GLP- 1R +/+ and GLP- 1R -/- mice when mildly stressed and . . glucocortico~d-treated...... -62

TABLE 6 Table summarizing differences in serum corticosterone levels of male and female GLP- 1R +/+ and GLP- 1R -/- mice under different stress conditions and with glucocorticoid treatment...... -64

TABLE 7 Quantitative comparison of [125~]-~~~binding in male and female GLP- 1R +/+ and GLP- f R -/- mice ...... -76

TABLE 8 Quantitative comparison of ['25~-~~2binding in male and female GLP- 1R +/+ and GLP- 1R -/- mice at sexual dimorphic regions of the brain ...... -8 8 LIST OF ABBREVIATIONS ac anterior commissure aAMY anterior amygdala ABP androgen binding protein ACB accumbens nucleus ADT anterodorsal thalamic nucleus AHiP am ygdalohippo campal area ANS autonomic nervous system ARC arcuate nucleus of the BLA basolateral amygdala BST bed nucleus of stria terminalis CAMP cyclic adenosine monop hosphate cAMY central amygdaloid nucleus CA 1,2, and 3 cornu arnmonis 1,2, and 3 subregions of the hippocampus CCK-8 -octapeptide cDNA complimentary deoxyribonucleic acid cmAMY corticomedial amygdala CN caudate nucleus CNS central nervous system CRF Corticotrophin-Releasing Factor CTX cerebral cortex DG dentate gyms of the hippocampus ER estrogen receptor FSH follicle stimulating hormone FSTR fundus striati GDX gonadectomized GIP glucose-dependent insulin-releasing peptide GLP- I glucagon-like peptide- 1 GLP-2 glucagon-like peptide-2 GLP- 1R GLP-1 receptor GLP- 1R +/+ normal wildtype mice GLP- I R -/- mutant mice with targeted disruption of the GLP-1 R GnRH Gonadotrophin-Releasing Hormone GP globus pallidus m habenula HPA Hypothalamic-Pituitary-Adrenal axis HPG Hypothalamic-Pituitary-Gonadal axis ICV intracerebroventricular IP intraperitoneal LH lateral nucleus of the hypothalamus LS lateral septum LSD dorso-lateral septum ME median eminence MIE-, I 1P-methoxy- 16a[125~] iodo estradiol MBH mediobasal hypothalamus MFB medial forebrain bundle rnRNA messenger ribonucleic acid MS medial septum NTS nucleus solitarius tractus OX optic chiasrn PAG periaqueductal grey PKA protein kinase A PHI peptide histidine isoleucine arnide PLC phospholipase C POA preoptic area of the hypothalamus PVN paraventricular nudeus of the hypo thalamus RIA radioimmunoassay RT reticuIar nucleus of the thalamus SF0 subfornical organ

xii SB subiculurn SON supraoptic nucleus of the hypothalamus TB W total body weight VIP vasoactive intestinal peptide VMN ventromedial nucleus of the hypothalamus ZI zona incerta

... Xlll INTRODUCTION

1.1 GLUCAGON-LIKE PEPTIDE-1 (GLP-1) (i.) Qverview Glucagon-like peptide- 1 (GLP- 1) is a synthesized by peptidergic cleavage of the proglucagon molecule in the endocrine L-cells of the gastrointestinal mucosa. Mammalian proglucagon is encoded by a single gene that contains six exons and five introns and it is expressed in the intestinal L-cells, the pancreatic A-cells and in some neurons in the CNS 11,22,71,111,145 . Proteolytic cleavage of proglucagon is tissue- specific23,125,146 . In pancreatic A-cells, the catabolic hormone glucagon is the predominant bioactive peptide formed, and GLP-1 and GLP-2, a peptide functionally distinct fkom GLP-1, are retained in the major proglucagon fragment (MPGF), a iarge prohormone fragment of unknown biological function 141.144 . Conversely, in the intestinal L-cells, both GLP-I and GLP-2 are processed from proglucagon, and the glucagon sequence is contained within a large prohormone fiagment called glicentin that has unknown biological actions (figure

The processing by proteolytic cleavages of GLP-1 (amino acid 72-1 08) in the intestine is somewhat complicated. At least four different isoforms of GLP-1 have been identified: GLP- 1-(I-37), GLP- 1-(7-3 7), GLP- 1-(I-3 6) arnide and GLP- 1-(7-36) amide90. 123.125. The physiologic activities appear to be restricted to the truncated GLP-1-(7- 37) and GLP-1-(7-36) aide peptides13g. Furthermore, no studies reported to date have distinguished differences between the biological potencies of these two peptides. The GLP- 1-(I-37) and GLP-1-(1-36) arnide have little, if any, biological actions. Henceforth, for the purpose of this thesis, the active forms of truncated GLP- 1, which are also called GLP- 1-(78- 108) and GLP-1-(78-107) arnide will be referred to as GLP-1. PREPROGLUCAGON

1 61 tZ ~a-eas(alpha cells) GRPP GL~~OOI MPF

FIGURE 1. A schematic representation of the post-translational processing of the proglucagon molecule as it occrus in the pancreatic islets and in the intestinal mucosa. In the pancreatic a-cells, GRPP (glicentin-related polypeptide), glucagon and MPF (major proglucagon fragment) are the main products generated by cleavage of the proglucagon molecule at the sites indicated by the heavy vertical bars. In intestinal L-cells, glicentin, GLP-1, GLP-2 and are the main products. IP refers to intervening peptide. The numbers indicate positions of amino acid residues in proglucagon. (Adapted from Holst, J.J., 1994). (2.) Re-mlation of GLP-I Secretion The development of specific radioimmunoassay systems allowed study of the secretory pattern of GLP-1 into the circulation177. During food ingestion, an increase in GLP- 1 plasma concentration fiom 1- 10 pmoVL to 30-50 pmol/L was foundIs3. Under certain pathological conditions, such as increased gastric emptying and operations on the gastrointestinal tract, concentrations exceeding 100 pmol/d were found12l. In addition to a meal-induced GLP- 1 secretion, intrduminal glucose, both metabolized and unrnetabolized forms, stimulated the release of GLP-1 73,137. The response of GLP-1 to a meal or an oral glucose challenge is rapid. Significant increases occur after a few minutes, and peak values were reached within 15-30 minutes after oral intake of a stimulus. The mechanism by which GLP- 1 is released from the intestinal mucosa could involve interactions of digested nutrients with the rnicrovillous plasma membrane on the cytoplasmic processes of the L-cells, which have been shown to reach the etlumen (figure 2)72,78,79,141,145 , particularly gastric inhibitory polypeptide (GIP), -releasing- peptide (GRP), and substance P are potent releasers of GLP-1 in rats 52,72,152. However, in humans, gastroduodenal hormones, including gastrin, , cholecystokinin, GIP, GRP, and , do not increase GLP-L secretion when inftiseci in physiological amounts 52,131,152

(iii.) Pancreatic Actions ofGLP-/ GLP-1 is the most potent insulinotropic hormone identified to date, both in vitro and in vivo 47,48,52,93,137 . GLP-1 is shown to stimulate insulin secretion fiom cultured insulinoma cell lines and from isolated pefised rat, porcine and canine pancreas28.49.62.1 14. Stimulation of insulin secretion in these experimental systems is observed at concentrations of GLP-1 as low as 1 to 10 picomolar47,48 . The Krn for activation of insulin secretory response is 0.1 to

1.0 nanomolar GLP-1 '88. The insulin secretory response to GLP- I is glucose-dependent. In the presence of substimulatory concentrations of glucose (2.8mM), GLP-1 does not induce insulin secretion. However, with slightly elevated glucose concentrations (6.6 mM), GLP-1 induces a very strong potentiation of insulin secretion which is also observed at higher glucose concentrations of up to 16.7 d7,48,188 . This glucose-dependent stimulatory action of GLP-I is a characteristic feature of hormones. The incretin effect of GLP-1 on insulin secretion has been demonstrated in normal healthy humans66.84, 153 and even in patients with severe type 2 (insulin-independent) diabetes mellitus59,66,~30 An important role of GLP-1 in pancreatic p-cells is the transcriptional activation of the insulin gene. Treatment of insulin-producing RIN 1046-38 and HIT-TI 5 insulinoma cells with GLP-1 induces elevated levels of proinsulin messenger RNA67,133 - Stimulation of proinsulin gene transcription by GLP-1 is also observed in transient transfection experiments of hamster insulinoma HIT-T15 cells and mouse insulinoma PTC-1 cells67,133 . In summary, GLP-I is not only a powerfd physiological insulin secretatogue, but it can also restore the levels of insulin by stimulating transcription. GLP-1 stimulates the release of from pancreatic 8-cells and inhibits glucagon release From pancreatic a-cells in rats, dogs, and humans 19,s 1,239,135. The release of somatostatin is due to the interaction of GLP-1 with its receptors on the 6-cells. GLP-1 receptors have been identified on the somatostatin-producing islet cell line RIN1027-~2'~.In contrast, the GLP-1-mediated inhibition of glucagon secretion is thought to occur by an indirect paracrine mechanism since glucagon-secreting a-cell line InRl G9 does not possess receptors for GLP- 1 to interact directly with28.50.122 . This paracrine mechanism is known to be mediated by increased insulin and somatostatin secretion since both insulin and somatostatin are known as physiological inhibitors of a-cell f~nction~~'.These studies suggest that insulin, by a paracrine mode of action, suppresses the release of glucagon and that the suppression of glucagon secretion by GLP- I is a consequence of the GLP- 1-mediated stimulation of insulin secretion (figure 2). Upon infusion of GLP-I , the plasma concentration of insulin increases, the glucagon concentration decreases, and as a result, blood glucose concentration decreasess4. The hypoglycemic effect of GLP- I is self-limiting since GLP- l 's effect on insulin and glucagon secretion wanes with decreasing glucose concentration. This may be of particular importance in the treatment of diabetes where GLP-I, unlike insulin, does not cause dangerous hypoglycemia131,138,179,192

D GLUCOSE mm-oooomo-o@>@ FOOD -

FIGURE 2. Model depicting the proposed physiologic actions of GLP- 1. Nutrients in the intestinal lumen (food, glucose) stimulate the release of GLP-1, which increases insuIin and somatostatin secretion and suppresses glucagon secretion via paracrine action. Insulin is an anabolic hormone that stimulates glucose uptake by liver, muscle, and fat during feeding. Glucagon is a catabolic hormone that stimulates glucose production during fasting. Thus, alternative proteolytic processing of proglucagon produces GLP-1 in the intestines during feeding and glucagon in the pancreas during fasting. (Adapted from Fehmann, H.C., 1992). (iv.) GLP- I Si-gnalina Pathwqs GLP-l -mediated insulin secretion is glucose-dependent and requires dual inputs fi-om the glucose-signaling pathways and the GLP-I receptor-mediated CAMP protein kinase A (PKA) signaling pathways87, I80 . The key event in insulin secretion is the influx of calcium ions &om the extracellular fluid into the cell, a process that triggers exocytosis and release of insulin into extracellular fluid or blood stream. The influx of calcium is largely dependent upon the opening of voltage-sensitive calcium channels (ca2+-vs), whose activation is dependent on the voltage potential between the inside and outside of the cell controlled by the ATP-sensitive potassium channels (K'-ATP). The closure of K'-ATP channels leads to the depolarization of the P cell and consequent insulin secretion. The closure of the K+-ATP channels requires the dual input of the ATP signal generated by glycolysis and the CAMP signal produced by the activation of GLP-1 receptor (figure 3)41.156,166,168,185,199 The GLP-1 receptor could also be coupled to the phospholipase C (PLC) pathway and related to increases in intracellular calcium. In receptor-transfected Cos-7 cells (cells that do not express endogenous GLP-I receptor gene), an accumulation of inositol phosphates and cytoplasmic calcium was monitored 196,197 . Furthermore, in receptor-transfected fibroblasts and Cos-7 cells, GLP-I binding lead to adenylyl cyclase activation and CAMP production. The activation of adenylyl cyclase and increase in intracellular levels of CAMP were shown to be correlated to the stimulation of the proinsulin gene expression in both insulinoma cell lines and isolated isletsljO. Together, these findings suggest that the effects of GLP-1 on insulin secretion and gene transcription are attributable to CAME' production and PKA acti~ation'~~. Depol-tion D-Glucose K+ Ca* Transporter

CAMP PKA

Vesicular Krebs Cycle

1 1

00 0 a** Insulin .. . . a . Secretion - . ..om - -0 a . 0 0. 0

FIGURE 3. Model depicting the proposed GLP-1 signal transduction pathways in the pancreatic p cell. The key elements of the model are the requirement for the dual inputs to the glucose-glycolysis signalling pathway and the GLP- I /GLP- 1R-mediated CAMP protein kinase A (PKA) signalling pathways to effect closure of ATP-sensitive potassium channels (K-ATP). The closure of these channels results in depolarization of the P cell leading to opening of voltage-sensitive calcium channels (Ca-VS). The influx of calcium through the opened Ca-VS triggers vesicular insulin secretion by the process of exocytosis. (Adapted from Habener, J.F., 1992). (v.) Gastrointestinal Actions of GLP-l In addition to having a cardinal role in the regulation of systemic glucose homeostasis, GLP-1 inhibits gastric acid secretion fiom parietal cells in rats, pigs and humans 7,45,46,140.165,194 - In rats, GLP-1 exerts its effect indirectly by stimulating somatostatin release fiom gastric D-cells which, in turn, inhibits acid secretion46. In humans and pigs, the effect of GLP-1 on parietal cells is mediated by interacting directly to a specific receptor of identical structure as the cloned pancreatic p-cell receptor 165,166,167 In addition to controlling acid secretion, GLP-1 inhibits gastric emptying68,182,194 This effect can be abolished by vagotorny which suggests that the gastrointestinal actions of GLP-1 involve the inhibition of efferent, stirnulatory vagal pathways 193.195 . Although the molecular basis for this mechanism is not fully understood, the inhibition of gastric functions is important in restricting the rate of nutrient absorption by the intestinal mucosa.

(vi.) Other Peripheral Actions of GLP-I Based on a series of studies conducted in rats, GLP-1 &As and receptors have , lUng88.96. 156,157 been identified in the kidney162, heart9'j1 , skeletal muscle3', and the brain22,60,98 . While GLP-1 and its receptors have been localized to these extrapancreatic sites, the physiological relevance of their presence is not yet fully understood. However, it is known that the GLP-1 receptors at these sites are of identical sequence 11,71,16G.190 GLP-1 is known to have an effect on adipocyte metabolism. When adipocytes are exposed GLP-1, fatty acid synthesis occurs44,134 . GLP-1 affects the kidneys as well 136,162- Preliminary studies show that the half-life of GLP-1 ifised intravenously into hurnans is about 4.5 minutes, and the metabolic clearance rate has been calculated to approximately 13 mLrkg"'min' I 135. Thus the peptide seems to be rapidly and effectively removed from the circulation. The kidneys may possibly have a role in effectively metabolizing and removing GLP-1 from the circulation, since in nephrectomized rats, the rate of GLP-1 metabolism seem to decreaseI6*. Furthermore, elevated plasma GLP-1 levels in patients with failure indicate that the kidneys may contribute to clearance of GLP-1 in humans as well136. Besides the GLP-1-mediated actions cited above, some cardiovascular effects of the peptide have also been reported. GLP-I is found to induce an increase in rate and in mean arterial blood

(vii.) Role ofGLP- I in the ClVS Using autoradiography60,6 1, immunohist~chemistry~~,RNAse protection analysis22, Northern Blot analysisu, in situ hybridizationu,s1,171 , and RT-PCR~~,the cellular localization and distribution of GLP-l/GLP-1 R and their mRNA expression have been identified in the rat brain. GLP-1 and its receptors are foaid in the rat brain primarily dong the hypothalamus, thalamus and structures of the limbic system. The presence of GLP-1 and its receptors in these sparse but well-defined autonomic and neuroendocrine regions of the brain have opened a variety of experimental fields outlining the potential roles of GLP-1 as a neurotransmitter, neuromodulator, and/or a behavior determinant. Recently, it has been suggested that GLP-1 in the brain may be a novel mediator in the regulation of food intake 101.132 . Evidence stems from an experiment conducted by Turton el ul. (1 996) who demonstrated that intracerebroventricular (ICV) administration of GLP-I, rather than intraperitoneal (IP) administration of GLP- I, profoundly inhibits feeding in 24- hour fasted rats, exerting greater effects at higher doses. This effect was blocked when GLP- 1 and its antagonist, exendin-(9-39), were co-injected. The injection of exendin-(9-39) alone more than doubled food intake in satiated rats1''. Consistent with these findings are reports showing GLP-1 binding sites in distinct hypothalamic regions that are implicated in the regulation of appetite6O,l7 1 . Despite all the evidence, the mechanism by which GLP-1 controls appetite in the CNS is not yet fully understood. The role of GLP-1 as a satiety agent has also been extended to the mouse species. By engineering a mouse model harboring a disruption in the gene encoding the GLP-1 receptor (GLP-1R -/- mice), Scrocchi et al. (1996) were able to demonstrate a GLP-1-mediated inhibition of food intake in normal wildtype (GLP-IR +I+) mice, but not in GLP-1 R -1- mice170. Furthermore, in comparing body weights of male and female GLP-1R +/+ to age- matched GLP-IR -/- mice at different ages, they found that there was no significant differences in body weightsl7'. The normal food intake and body weights of GLP-IR -/- mice suggest that the GLP-l/GLP-1R system may not be the exclusive component in controlling appetite. As is the case for many other multi-factorial systems in the CNS, there may be redundancy in the control of appetite. Therefore, when the GLP- IIGLP- 1R system is disrupted, other systems that regulate appetite may compensate as a safe-guard mechanism. In many studies examining effects of centrally administered GLP-1 on feeding, the dependent variable was the weight of food taken over a period of time. However, mammalian food intake comprises complex behavior and constitutes a discontinuous process in which periods of eating alternate with periods of non-eating. These qualitative features draw attention to the distinction between food intake (usually assessed by measuring the weight of food consumed) and feeding behavior which can be addressed through analysis of the feeding response. In keeping with this hypothesis, when starved GLP-I R -1- and GLP-1 R +/+ mice were placed in a novel cage with free access to food, both exhibited different feeding behaviors and responses to the change in the external environment (Scrocchi and Brown, unpublished observations). While GLP-1R +I+ mice showed a normal response, initially being hesitant to approach the food container in the novel environment, GLP-1R -/- mice did not appear at all apprehensive and began feeding immediately. These observations suggest that the reduction of food intake in GLP-1R +I+ mice could be attributable to an adaptive behavioral response related to the novel testing environment rather than a specific central GLP-I effect on appetite. There is already evidence in the literature that GLP-1 may regulate stress responses.

Larsen ei aL(1997) demonstrated that central administration of GLP-1 activates the Hypothalamic-Pituitary-Adrenal (HPA) axis primarily through stimulation of the CRF- containing PVN neurons9*. Furthermore, in considering potential central sites of GLP-1 action, Goke el al. (1995) observed GLP-1 receptors in the hypothalamic PVN neurons6'. This combination of evidence raises the possibility that GLP-1 might have a physiological role within the HPA axis. GLP- 1 may also be involved in the regulation of the Hypothalamic-Pituitary-Gonadal (HPG) axis. Evidence for this stems from a study by Beak et al. (1998), who have recently demonstrated a dose-dependent increase in Gonadotrophin-Releasing Hormone (GnRH) secretion following GLP-1 administration to both the rat hypothalamus and the immortalized hypothalamic GnRH-producing GTI-7 cells10. The notion that GLP-1 might be involved in the regulation of GnRH release is consistent with preliminary observations by Scrocchi and Drucker (unpublished) that there is a reduced litter size in GLP-1R -/- mice. These lines of evidence provided the impetus to firrther examine the effects of the GLP-1 R -/- mutation on neuroendocrine function and behavior.

1.2 HYPOTHESIS We hypothesized that GLP-1 may have a general neuromodulatory role in the CNS, as is the case for many other gut-brain peptide systems. The localization of GLP-1 receptors in both the autonomic and neuroendocrine regions of the brain is consistent with the view that GLP-1 might have effects on a number of systems that influence neuroendocrine control of the pituitary gland, such as the HPA and HPG axes.

1.3 OBJECTIVES Some clues as to the potential role of any peptide or neurotransmitter in the brain can be gleaned from an analysis of its target sites of action. One of our first objectives was to determine the potential targets of GLP-1 action in the mouse brain, both in normal wild-type GLP-1 R +/+ mice and in the "knock-out" GLP-1R -/- mice. The latter provided us with a potentially powerful tool to elucidate the physiological role(s) of GLP-1 in the brain; but use of this model was only valid if a number of conditions could be fulfilled. (1) We had to be sure that the mouse did indeed normally have CNS receptor systems for GLP-1. (2) It remained to be determined whether the targeted disruption of the GLP-1 receptor gene did indeed eliminate functional GLP-1 receptors from the brain. (3) It remained to be established whether other receptor systems in the brain might also be sensitive to the effects of GLP-I. If such systems existed, we might not expect to see any marked effect of deleting the normal GLP- I receptor. To achieve these objectives, we mapped the GLP-1 receptors in the mouse brain using in vitro ['2i~]-~~~-1receptor autoradiography Previous studies have established the location of GLP-1 target sites in the rat brain (Goke et al., 1995). No previous studies, however, have examined the distribution of GLP-1 receptors in the mouse brain. By studying thz distribution of [12S~]-~~~-lbinding in mouse brain, we hoped to determine both the normal target sites of action of this hormone in the mouse brain and possible remaining sites of action in the GLP- 1R -/- mouse model. Having established the consequences of the GLP-IR -1- mouse model on CNS [12S~]- GLP-1 binding, the next step was to determine the effects of the loss of GLP- 1 signaling on reproductive fbnctions and in functions related to stress/novel environment responses. To establish whether GLP-1 is involved in functions related to the HPG system, several reproductive parameters were examined. While inspecting GLP- I R -/- and GLP- 1R +/+ mice from birth to adulthood, we compared litter size and examined onset of puberty, cyclicity, and gonadal weight and histology. By radioimmunoassay, we also compared serum estradiol and testosterone concentrations in female and male GLP-1R +/+ and GLP-IR -/- mice, respectively. To determine potential roles of GLP-1 within the HPA axis, serum corticosterone levels were measured by radioimmunoassay, and compared between GLP- 1R - /- and GLP-1 R +/+ mice when subjected to three different stress conditions: non-stress, mild stress, and severe stress. Senun corticosterone levels were also measured in GLP- 1R +/+ and GLP-1R -/- mice that have been subjected to chronic treatment with glucocorticoid in their drinking water. Finally, we addressed the possibility that disruption of the GLP-1 receptor gene might indirectly alter other CNS systems involved in the control of neuroendocrine function and behavior. The apparent lack of any overt effect of the GLP-IR -1- mutation on body weight regulation might have been explained by a concomitant increase in the sensitivity of other systems involved in the regulation of satiety. One such hypothalamic system is the cholecystokinin-octapeptide (CCK-8) system in the hypothalamus. By quantitative [12'1]- CCK-8 receptor autoradiography, we tested whether loss of GLP-1 signaling causes a compensatory up-regulation of CCK-8 receptor expression. Another possible mechanism through which alteration in GLP-1 sensitivity might impact on both appetite regulation and neuroendocrine fhction is through interference with normal sexual differentiation of the brain. If GLP-1 is involved in regulation of hypothalamic GnRH output, as Beak et aL1* have suggested, then it is possible that there might be a deficit in GnRH production early in development and therefore in the differentiation and function of the embryonic testicular Leydig cells. This could resuIt in inadequate testosterone production and, hence, incomplete masculinization, which would secondarily impact on hctions that are normally sexually differentiated, such as the control of appetiteu3 and responses to a novel environment. One of the most obvious biochemical sequelae of sexual differentiation in the brain is a reduction in estrogen receptor concentrations in the hypothalamic ventromedial nucleus and periventricular preoptic area of the male. Therefore, we analyzed the expression of estrogen receptors in the brain, to assess whether GLP-1R -/- mice have undergone normal CNS sexual differentiation. GLP-1 RECEPTOR DISTRIBUTION IN THE MOUSE BRAIN

2.1 CHARACTERIZATION OF GLP-1 RECEPTOR (i.) Shwcture qf the GLP- I Receptor The primary structure of the GLP-1 receptor has been molecularly characterized by isolating cDNA clones fiom the rat and human pancreatic islets3'. GLP-1 receptor belongs to a subfamily of seven-helix transmembrane receptors that are coupled to one or more intracellular signaling pathway via the heterotrimeric guanine nucleotide-binding protein (G protein) 77,166 . Included in this subfamily of G-protein-coupled receptors are glucagon, GLP- 2, exendin-4, GIP, VIP, secretin, GRF, and PHI (figure 4)'j2. These receptors are characterized by 8 hydrophobic regions, the first of which is a leader sequence required for translocation of the receptor peptide in the endoplasrnic reticulum during its biogenesis. The remaining 7 hydrophobic regions are transmembrane domains 1 J7.96 .

(iU GLP- I Binding Pro-oerties Receptor studies in insulin-producing islet cells, rat insulinoma cells, and GLP-1 receptor transfected cells, show a high binding affinity of radiolabeled GLP-I for its re~e~tors~'~.With the development of GLP-I mutant peptides that have single alanine substitution at each amino acid position, the binding properties of GLP-I have been determined. Results reveal that histidine at position 1, glycine at position 4, phenyalanine, threonine and aspartic acid at position 6, 7 and 9 as well as phenyalanine and isoleucine at position 22 and 23, are all important for efficient binding and coupling to CAMP'. Furthermore, the development of a glucagodGLP-1 chimera indicates that the carboxy- terminal end of the GLP-I peptide is essential for recognition of the peptide by its receptor, whereas the amino-terminal end is more important for proper activation of the receptor 2 I,j4.76 KNDWK H

FIGURE 4. Representative hormones of the glucagon superfamily of peptides. GLP-1 sequence is shown at the top. The amino acids in the other peptides that are identical to GLP-1 are shaded. Greatest similarities in structures of the peptides reside in the amino- terminal regions. Exendin-4 is a GLP-1 receptor agonist and is found in the venom of Heloderma Suspectum. GIP, glucose-dependent insulin-releasing peptide; VIP. Vasoactive intestinal peptide; PACAP-3 8, pituitary adenylyl cyclase activating peptide-3 8; PHI, peptide histidine isoleucine; GRF; -releasing factor. (Adapted from Fehmonn, H.C. and Habener, J.F., 1992). The sequence of GLP-1 is entirely conserved in dl mammalian species investigated to

date1"71. The active products of the proglucagon peptide (glucagon, GLP- 1, and GLP-2), together with GIP, VIP, PHI, and secretin, have a high degree of sequence similarity and form the super family of glucagon-related peptides (figure 4)". Despite the high degree of similarity, none of the above mentioned peptides show any affinity for the GLP-1 receptor, except glucagon152. Glucagon was able to displace GLP-1 receptor binding with an affinity 2-3 orders of magnitude lower than that of GLP-1, and this is consistent with the demonstration that glucagon is 100-1000 times less potent at the pancreatic p-cell than GLP- 154,55.76.80.122 . Surprisingly, exendinll, a 39 amino acid peptide isolated from the venom of Eeieloderma S.rlspecrum, has been recody sho-;;n to be a very potent agonisi of CLP-I wkde exendin-(9-39) was found to be an antagonist of GLP-1. Exendin-4 and exendin-(9-39) differ because of the absence of the first 8 amino acids in the latter. In addition, the amino- terminus of exendin-4 demonstrates high similarity with GLP-1 since 7 out of 8 amino acids are conserved in both peptides. This suggests that binding to the receptor is most likely directed by the last 30 amino acids of the peptide, while the agonist property requires interaction of the first amino-terminal region of the peptide with the receptor 96,122163. Both have high sequence similarity to GLP-1 and can bind both rat, pig, and human GLP-1 receptors with relatively high affinity76.92,154

(iii.) GLP-I Receptors in the CNS As deduced by Wei et al. (1995)Ig0, the structure of the brain GLP-1 receptor is identical to the pancreatic p-cell receptor. By receptor autoradiography, the anatomical distribution of GLP-1 receptor in the rat brain has been thoroughly mapped by Goke et al. (1995f0. They have reported a high concentration of [125~]-~~~-~binding in regions of the diencephalon and limbic system. In the diencephalon, high densities of GLP-1 receptors were identified in the paraventricular nucleus (PVN), ventromedial nucleus 0, dorsomedial nucleus @MN) and the preoptic area (POA). GLP-1 receptors were also visualized in the median eminence (ME) and arcuate nucleus (ARC). In the limbic system, the highest densities of GLP-1 receptors were reported in the subfomical organ (SFO), lateral septum (LS), and the central arnygdala (cAMY). In the pituitary, GLP-1 receptors have been reported in the lobe, and fewer in the , but none were detected in the intermediate lobe6 The localization of GLP-1 receptors in distinct regions of the brain can be correlated to its function. For example, the presence of GLP-1 receptors in the hypothalamic VMN, DMN and PVN provides a basis for GLP-1's involvement in appetite regulation 178,181,198 The presence of GLP-1 receptors in the nucleus solitarius tractus and the area postrema is relevant to GLP-1 's role in the cardiovascular Furthermore, the demonstration of GLP- 1 receptors in the pituitary gland and in various structures of the autonomic and limbic systems, provides indirect evidence for a role of GLP in the control of neuroendocrine ~cticns5 1 .

2.2 OBJECTIVES The current knowledge of GLP-1 hction in the mouse is Limited since most reports have emerged from studies conducted in rats. The objective of the present study, therefore, was to explore the roles of GLP-1 in mice. In doing so, we have first aimed to identify areas of the brain where GLP-1 receptors might be present by [12S~]-~~~-1receptor autoradiography. If the regional distribution of GLP-1 receptors in the mouse were similar to that already reported in the rat, this would be consistent with a potential role for GLP-1 in the regulation of responses mediated through the hypothalamus and limbic system6'. Since, as will become evident during the presentation of our results, below, we actually observed a different pattern of [12S~]-~~~-lbinding in the mouse than that reported by Goke el al. (19951~' for the rat, we also re-examined the distribution of [12S~]-~~~-1receptor binding in the rat to control for possible methodological artifacts. Since exendin-4 is an excellent high- affinity ligand for the GLP-1 receptor, we examined the binding of [125~]-exendin-4in parallel with that of GLP-1, in part to determine whether use of this ligand might reveal additional binding sites that were not apparent with GLP-1. Finally, in all binding studies, GLP-1R 4- mice were also examined, to determine whether there might be expression of alternate potential receptor targets for GLP-1 in the brains of these mice, despite the targeted disruption of the GLP-1 receptor gene. 2.2 MATERIALS AND METHODS (i.) Chemical3 Radiolabeled [12'1]-~~~-1-(7-36)(specific activity = 440 Ci/mmol) and [12'1]- exendin-4 (specific activity = 620 Ci/rnmol) were prepared and supplied by Dr. Patricia Brubaker (Department of Physiology, University of Toronto). Synthetic GLP-1 (7-36) was purchased from S igrna (S igma-G8 147). Methoxyflurane anesthesia (Metofaney was purchased from Pitman-Moore, USA (Washington Crossing, NJ, USA). Autoradiographic film (Amersharn ~~~erfilm-~~)was purchased from Amersharn Canada (Oakville, Ontario, Canada).

(ii) Animal Preparation Age-matched and sex-matched wild type mice (GLP-1 R +/+, Charles River Breeding laboratories, St. Constance, Quebec, Canada) and GLP- I R -/- mice were housed five per cage by genotype and sex phenotype and were given access to Purina Rat Chow and water ad libitum. Male Wistar rats (Charles River Breeding laboratories, St. Constance, Quebec, Canada) were housed under standard environmental conditions (12hr Iight/l2hr dark cycle, lights on at 0700 hrs). Surgical removal of bilateral gonads (gonadectomy) was performed under Metofane". Animals were sacrificed by decapitation seven days post-gonadectomy, and their brains and pituitaries immediately removed and frozen onto cryostat chucks over liquid nitrogen vapor. Brains and pituitaries were stored in air-tight containers at -80'~ensuring stability of GLP-1 receptors for at least 12 months.

(iii.) Tissue Sectioning Brains stored at -80°C were transferred on dry ice to a Reichert HistoSTAT cryostat (Scientific Instruments, Inc., Buffalo, NY) maintained at -27'~. Once brains were allowed to equilibrate to cryostat temperature for 30 minutes, serial sagittal and coronal sections (20 j.~ m) were rostral-caudally cryosectioned throughout the brain from the joining of the anterior commissure (Bregma 0.70 mm, Interaural 9.70; c.f. Paxinos and ats son'^^) to the ventral CA-3 region of the hippocampus (Bregma -4.52, Intraural 4.45; c.f. Paxinos and p at son'^'). Sections were thaw-mounted onto pre-cleaned glass slides (VWR Canlab) coated with 0.02 mg/rnL poly-L-lysine, and immediately re-fiozen on a cold metal platform at -27°C. All slides were then stored dessicated at -80°C until the day of the assay.

(iv.) GLP-I Receptor A utoradiopaplq In vitro radio-labeling of GLP- I receptors using either [12'1]-~~p-1-(7-3 6) or [12'1]- exendin-4 was performed in the mouse brain sections essentially as described by Turton et a/. (1997)18'. Slide cassettes were quickly transferred on dry ice from the -80°C freezer to a cold room maintained at 4"C, and slides were then arranged flat across rowed straws placed in a tissue culture tray lined with wet pper towel. -Mouse brain sections were preincubated in 25 rnM HEPES assay buffer (400yVslide), pH 7.4, containing 2rnM MgC12, 0.1% bacitracin, 0.05% Tween-20 and 1% bovine serum albumin for 30 minutes at room temperature. Excess liquid was drained fiom the slides and then incubated in the assay buffer containing radio- labeled [125~]-~~~-1-(7-36) or [125~]-exendin-4for 2 hours at 20°C (500pVslide). Slides were brought back to the cold room, and excess incubation buffer was drained off slides. The slides were then loaded onto metal slide racks and rinsed 3 X 1 minute each in ice-cold HEPES buffer and dipped briefly in ice-cold double deionized distilled water. Slides were then dried overnight at 4'~under air stream. Once dried, slides were exposed for 14 days against Arnersham hyperfil~n-~~and then developed with Kodak D- 19 developer, diluted 1:3 with water. Autoradiography for [125~]-exendin-4was performed using identical procedures. GLP-L binding displacement assay was carried out by co-incubating non-radiolabeled GLP- 1 with both [IZ51]- exendin-4 and [125~]-~~~-l.To test reliability of our experimental methodologies, we have used identical autoradiography procedures in rat brain to verify whether results match those reported in the literatures.

(v.) Cornouter-assisted Image A natvsis of Rece-ntor A utoradiomams Rat and mouse brain images generated from [12S~]-~~~-1receptor autoradiography were analyzed with a computer-assisted densitometry (MCID system; Imaging Research, St. Catherines, Ontario) after calibration with the autoradiographic standards. Autoradiographic standards were obtained by a gradient of highest to lowest non-specific radiolabeling of brain tissues. The densities produced by the standard sections were standardized in terms of ferntomoles [125~]-~~~-1bound per rng protein by scraping standardized brain sections into 12 x 75 mm culture tubes and digesting scrapings with 100~1of 0.3N KOH in a 42°C water bath. Protein content was measured by the method of Lowry et a2 (1951)110. [125~]-~~~-1 content was determined at 70% efficiency using an ICN Micromedic 4/600 Plus Gamrna- counter.

(vie) Emulsion A utoradiowa~hv For higher resolution analysis to distinguish individual cells, Kodak NTB-2 liquid autoradiography emulsion was used. Under 2 sodium vapor lamp, the emulsion was diluted with 600 nN ammonium acetate, and slides dipped while emulsion was maintained at 42°C. Slides were dried at room temperature, and stored at 4°C ion a light-proof container for 15 days. Development of emulsion-dipped slides was carried out for 3 minutes at 20°C in Kodak D-19 developer, diluted 1:3 with water. Slides were then washed for 30 seconds in 2% acetic acid and fixed for 4 minutes in 5% sodium thiosulfate. Slides were washed in 20" C water for 20 minutes, and dehydrated by 2-minute dips in SO%, 70%, 95% and 100% ethanols. The slides were lift in another 100% ethanol dip for 1 hour, followed by immersion in cresyl violet stain for 8 minutes, and a brief dip in 95% and 100% ethanol. The slides were then cleared in 3X xylene and coverslipped with Permount (Fisher).

(vii.) Statistical Analysis All data are presented as the mean + SEM. Statistical analysis was performed using PC-compatible microcomputer programs [SPSS for Windows; and Sigmastat (Jandel Scientific)]. Two group comparisons were made using Student's t-test. Multiple group comparisons were made using ANOVA, followed by Duncan's multiple range test. Differences between means were considered statistically significant at the p < 0.05 level (two-tailed). 2.4 RESULTS (i.} GLP-I Rece-ntor Autoradio,waph-v in the Mouse Braiq Autoradiograms of brain sections taken fiom GLP-IR 4- mice showed a complete absence of [125~]-~~~-1labeling (figure 5, top row). In GLP-I R +/+ mice, intense GLP-I receptor labeling was observed in different areas of the forebrain and hindbrain. An extensive distribution of GLP-1 receptors was also observed in the various regions of the diencephalon and the limbic system. The density of receptors was categorized according to the following levels: 0-1 0 fmoVmg of tissue protein, low; 10-1 5 finoVmg, moderate; > 15 fmol/mg, high. A high density of receptors in the diencephalon was observed in the hypothalamic medial preoptic nucleus (figure 5- 1c), the supraoptic nucleus (figwe 6- 1e), the arcuate ilucleus (figure 6-2f), and mediobasal hypothalamus (figure 6-2e). In the telencephalon, high densities of ['25~]-~~~-~labeling were identified in the dentate gyms of the hippocampus (figure 6-le,f), the lateral septum (figure 5-lc), the subfornical organ (figure 6-ld), and the basolateral amygdala (figure 6-2e). The distribution GLP-1 receptors in coronal and sagittal mouse brain sections is visually represented in figures 5 to 7. Resdts obtained fiom both male and female GLP-1R +/+ mice were indistinguishable, indicating a lack of sex difference in the distribution of GLP-1 binding (2 way ANOVA F=0.009; d.f 1,4Q1 ; p=0.924). Furthermore, gonadectomy did not affect GLP- 1 receptor expression. Quantification of GLP-I receptor concentrations in different brain regions are summarized in table 1 and are graphically represented in figures I0 through 15.

(ii.) C;L P-I Receptor Autoradiomaphv in the Rat Brain Autoradiographic quantification of GLP-1 receptor distribution in the rat brain was indistinguishable from the data previously reported by Goke el al. (1 995l6', confirming that the regional differences we observed between binding in the mouse brain and the previously reported data in the rat6' did not represent a methodological artifact, but likely reflect a real species difference in the distribution of GLP-1 receptors. The density of receptors was classified as follows: 0-5 finoVmg of tissue protein, low; 6-10 fmol/rng, moderate; >I0 hdmg, high. In the diencephalon, highest density of [125~]-~~~-~was observed in the arcuate nucleus (figure 16-1 b), median eminence (figure 16- 1b), and supraoptic nucleus (figure 16-2c). In the telencephalon, a high density of [125~]-~~~-~was detected in the subfomical organ (figure 16- la) and the lateral septum. The density of [125~]-~~~-1binding was moderate in the basolateral amygdala (figure 16-2b). The dentate gyrus of the hippocampus was free of ['**I]-GLP-I receptors. No quantitative differences in receptor concentrations were detected between gonadectomized and intact rats (2 way ANOVA F=0.082; d.f 1,125; p=0.776). Results are quantified in figures 17 to 21 and in table 2.

(iii) A Shared Receptor fir GLP-I and Exendin-4 Autoradiography of [12S~]-exendin-4binding in blain sections taken fiom GLP- 1R +/+ mice revealed a labeling pattern identical to that using ['25~]-~~~-i,except that the labelling observed with [12S~]-exendin-4was stronger, resulting in more contrast images (figures 5 and 6, column 3). In GLP-1R -/- mice, we did not detect any labeling with [125~]-exendin-4 (figure 7) suggesting that no other related high-afFmity binding site is expressed in the brains of GLP-1R -/- mice. In all experiments, in wild type animals, ['2S~]-exendin-4binding was completely displaced by co-incubation with excess non-radiolabeled GLP-I, consistent with the hypothesis that the two radiolabeled ligands interact with the same high-affinity binding site (data not shown).

(iv.) Emulsion A utoraciio-waa~h_v High resolution autoradiography using ['25~]-exendin-4demonstrated a heterogeneous distribution in binding sites in areas of the brain that appeared to be target sites for this peptide. Photomicrographs of the hippocampal dentate gyms taken under light- and dark- field illumination showed that binding sites are not localized directly over the granule cell layer. Instead, binding was concentrated over the relatively cell poor region which contains the granule cell dendrites (figure 8). Photomicrographs taken from the medial habenula, also showed considerable anatomical specificity in binding site distribution. For example, the periventricular region of the nucleus had a much lower concentration of binding than the more cell-dense areas that bordered the lateral habenula (figure 9). Similar results were obtained in the septum, in which the dorsolaterd apex of the lateral septal nucleus was found to have by far the highest labeling density (data not shown). I GLP-1 RECEPTOR DISTRIBUTION IN THE MOUSE BRAIN 1

male mice (fmoYmg protein) female mice (fmoVmg protein) MFB 13.94 + 3.13 1 1.29 -t 0.55 ACB 8.36 k 0.56 8.09 k 0.41 l3-b 25.29 f: 1.31 26.74 k 1.25 VmB 16.76 + 3.36 ;~H~~~+@p::~w2p,~8"'p;<4y&qg.$$.k.k.A. .-'<.."" ""..-, , L.3.. . .Y+.,; : /"":~~&. y;43y&$@&::5,XXZ6,L$.d@$$$~5~ . .' ' ' -.,..~..::<~.~~::Yk~ .w ,vH. ?<*'*k; ..>M....x*...... s<<.s+:.:.,..+ :*,*:.;*:,*$$7;m%$&?&&*% ;&$ +.:*$&$&&>~?*&~&$t@.;:riv,s~~~$;@&~~$@i$~#@g@g&~~&$$ Brain Regions I Specitic Binding of GLP-1 in 1 Specific Binding of GLP-I in male mice (fmol/mg protein) I female mice (fmol/mg protein) Cerebellum 15.74 + 0.87 f 15.22 k 0.72 Medulla NTS ZI

male mice (fmoymg protein) female mice (fmoVmg protein) POA 19.58 + 1.57 21.35 + 1.57 SON PVN VMN ARC MBH RT LH

Brain Regions I male mice (fmoVmg protein) I female mice (fmoVmg protein) ., n F I CA 3 FSTR BST LSD MS DG SF0 aAMY BLA cAMY AHiP

TABLE 1. Quantification of ['2S~]-~~~-lreceptor binding indicates no labeling differences in different brain regions between female and male GLP-IR +/+ mice. GLP-1R DISTRIBUTION IN THE MOUSE BRAIN

GLP- 1: Female GLP-1: Male

FIGURE 5. Series of autoradiograms demonstrating identical binding patterns of radioiodinated GLP-1 (columns 1 and 2) and exendin-4 (column 3) in coronal mouse brain sections. Sections are taken at the level of the lateral and medial septum (LS, MS; row b) and at the level of the anterior commissure (ac; row c). Autoradiograms taken fkom GLP- I R -/- mouse brain (row a) reveal neither [I2SI]-GLP-l or [*2?]-exendin-4 labeling and confirm absence of GLP-1 receptors. ACB, accurnbens nucleus; MFB, medial forebrain bundle; DB, diagonal band; LS, lateral septum; MnPO, median preoptic nucleus; AVPO, anteroventral preoptic nucleus; FSTR, fundus striati. -snmepwodLq aqljo snapnu a~vnam'm fwaxq suoz '12 fsnmeppoUq palel 'H? :E syomnu03 fepp8hp~a~qoseq '~78 :snumpqodrlq pseqo!pam '~m fsnapnu ogdowdns 'NOS fsnapnu lapqua~md'~d fsMaJ8)nap 'ga fopuaqeq 'w fspupmal eysjo snapnu paq '1~8-0 MOJ :3~)snapnu a~en3leaqjo IaAal aq, ~epue (a MOJ :~a)smB aJ,nap pdmooddy aqjolaAq ag, JV '(p MOJ toss) W%JO ppojqnsawjo laqatp r nam aram suoyoag -suopoasqwq asnom puoroo p (E mloa)v-~puaxa pm (2 pae I sm~oa)1-d79 paJempo!o!prujo srua~edBqp~q papuap! %up~suomapsmfio!pao~nu~o sauas .g -3~ GLP-1 R DISTRIBUTION IN THE MOUSE BRAIN

. --.... ,. Cerebellum

FIGURE 7. Autoradiograms demonstratingbinding patterns ofradioiodinated [ lZ51]-GLP-1 in sagittal mouse brain sections taken fiom wildtype GLP- I R +/+ mice @anel a) and GLP- 1 R -/- mice (panel b). aAMY, anterior arnygdala; DG, dentate gyms of the hippocampus; LS, lateral septum; MBH, mediobasal hypothalamus; NTS, nucleus tractus solitarius; 21, zona incerta. FIGURE 8. High resolution autoradiography of ["*I] exendin-4 binding in the dentate ams of the hippocampus of an adult male mouse. Photomicrographs of the same cell fields were taken under light- (A) and dark- (B) field illumination to depict the distribution of both cell nuclei (counterstained with cresyl violet) and the reduced silver grains (white dots under darkfield) resulting from exposure to [125~]labeled exendin-4. FIGURE 9. High resolution autoradiography of ['=I] exendin-4 binding in the medial habenula of an adult male mouse. Photomicrographs of the same cell fields were taken under light- (C) and dark- @) field illumination to depict the distribution of both cell nuclei (counterstained with cresyl violet) and the reduced silver grains (white dots under darkfield) resulting from exposure to ['''I] labeled exendin-4. GLP-1 RECEPTORS IN GLP-1R +I+ FOREBRAIN

Male (N=ll) U Female (N=ll)

20 12 MFB ACB I

VIHDB

FIGURE 10. Distribution of GLP- 1 R binding sites in discrete brain regions of the GLP- 1 R +/+ mice shows no significant differences between males and females (2 way ANOVA F=0.0009;d.f 1.40 1; P=0.924). Data are expressed in femtomoles of labeled GLP-1 retained per mg tissue equivalent. MFB, medial forebrain bundle; ACB accumbens nucleus; Hb, habenula; VIHDB, ventral / horizontal diagonal band. GLP-1 RECEPTORS IN GLP-1R +I+ HINDBRAIN

Male (N=4) Female (N=3)

Cerebellum 2o ! T Medu"a

NTS

FIGURE 11. Distribution of GLP- 1R binding sites in hscrete brain regions of the GLP- IR +/+ mice shows no significant differences between males and females (2 way ANOVA F=0.0009; d.f 1.401; P=0.924). Data are expressed in femtomoles of labeled GLP- 1 retained per mg tissue equivalent. NTS, nucleus tractus solitarius; 21, zona incerta. GLP-1 RECEPTORS IN GLP-1 R +I+ DIENCEPHALON

Male (N=ll) Female (N=11)

30 3 5 1 POA SON

PVN VMN - I

FIGURE 12. Distribution of GLP- 1R binding sites in discrete brain regions of the GLP- 1R +/+ mice shows no significant differences between males and females (2 way ANOVA F=0.0009; d.f 1-401 ; P=0.924). Data are expressed in femtomoles of labeled GLP- 1 retained per mg tissue equivalent. POA, preoptic area of the hypothalamus; SON supraoptic nucleus; PVN, paraventricular nucleus; VMN, ventromedial nucleus. GLP-1 RECEPTORS IN GLP-1 R +I+ DIENCEPHALON

Male (N=l1) Female (N=l1)

20 ARC MBH

T

FIGURE 13. Distribution of GLP- 1R binding sites in discrete brain regions of the GLP- 1 R +/+ mice shows no significant differences between males and females (2 way ANOVA F=0.0009;d.E 1.401;P=0.924). Data are expressed in ferntomoles of labeled GLP- 1 retained per mg tissue equivalent. ARC, arcuate nucleus; MBH, mediobasal hypothalamus; RT, reticular thalamus; LH, lateral hypothalamus. GLP-1 RECEPTORS IN GLP-1R +I+ LIMBIC SYSTEM Male (N=ll) Female (N=ll) 10 12 CA 1,2 CA 3

I* I FSTR BST

15 1 1 LSD MS

FIGURE 14. Distribution of GLP- I R binding sites in discrete brain regions of the GLP-1 R +/+ mice shows no significant differences between males and females (2 way ANOVA F=0.0009;d.f. 1.40 1 ; P=0.924). Data are expressed in ferntomoles of labeled GLP- I retained per mg tissue equivalent. CA 1, CA2, and CA3, cornu amrnonis 1, 2 and 3 subregions of the hippocampus; BST, bed nucleus of stria terminalis: FSTR, fundus striati; LSD, lateral dorsal septum; MS, medial septum. GLP-1 RECEPTORS IN GLP-1R +I+ LIMBIC SYSTEM IMale (N=l 1) Female (N=l 1)

PAHiP

FIGURE 15. Distribution of GLP- 1 R binding sites in discrete brain regions of the GLP- 1 R +I+ mice shows no significant differences between males and females (2 way ANOVA F=0.0009; d.f 1.401; P=0.924). Data are expressed in ferntomoles of labeled GLP-1 retained per mg tissue equivalent. DG? dentate gyms of the hippocampus; SFO, subfornical organ; aAMY, anterior arnygdala; BLA, basolateral amygdala; cAMY, central amygdala; AKiP, amygdalohippocarnpal area. I GLP-I RECEPTOR DISTRIBUTION IN TEIE RAT BRAIN I Brain Regions intact rats (frnol/mg protein) GDX rats (frnoUmg protein) 10.25 + 0.94 8.86 k 0.84 8.01 + 0.52 10.27 f 1.37 8.37 + 1.21 6.75 k 0.63 7.40 1: 0.48 7.23 k 0.50

Brain Regions Specific Binding of GLP-I in / intact rats (fmoYmg protein) GDX rats (fmoVmg protein) POA 10.83 + 1-05 8.63 k 0.74 SON PVN VMN ARC ME ADT RT

intact rats (fmoVmg protein) GDX rats (fmoVmg protein) CA 1,2 6.47 + 0.78 CA 3 LS MS DG SF0 BST BLA

TABLE 2. Quantification of [125~]-~~~-1binding in intact and gonadectomized (GDX) maIe rats in different areas of the brain reveals no regional labeling differences between two treatment groups. Data were assessed by quantitative in vitro autoradiography and are presented in femtomole of labeled GLP-1 retained per mg of tissue equivalent. GLP-1 R DISTRIBUTION IN THE MALE RAT BRAIN

INTACT RATS GDX RATS

FIGURE 16. Series of autoradiograms demonstrating identical [ '"I]-GLP- 1 binding patterns in intact male (column 1) and in gonadectomized (GDX) male (column 2) rat brain sections. Sections were taken at the level of the subfomical organ (SFO; row a), at the level of the arcuate nucleus (ARC; row b) and at the level ofthe supraoptic nucleus (SON; row c). CP, caudate putamen; MBH, mediobasd hypothalamus; ADT, anterodorsal thalamus; PVN, paraventricular nucleus; ME, median eminence; Hb, habenula; BLA, basolateral amygdala; 21, zona incerta; CA3, corm amrnonis 3. GLP-IR DISTRIBUTION IN THE RAT FOREBRAIN AND HINDBRAIN

Intact Male (N=3) 3 GDX Male (N=4)

FIGURE. 17. GLP-1R binding sites measured in discrete brain regions of intact and gonadectomized (GDX) male rats by quantitative in vitro autoradiography. There is no significant difference in GLP- 1R distribution between intact and GDX male rats (2 way ANOVA F=0.082; d.f 1.125; p=0.776). Data are expressed in ferntomoles of labeled GLP-1 retained per mg tissue equivalent. DB, diagonal band; Hb, habenula; OX, optic chiasm; ZI, zona incerta GLP-1 R DISTRIBUTION IN THE RAT DIENCEPHALON

IIntact Male (N-3) GDX Male (N=4)

POA SON

15 10 A PVN VAllN 3

T I

FIGURE 18. GLP- 1 R binding sites measured in discrete hypothalamic brain regions of intact and gonadectornized (GDX) male rats by quantitative in vitro autoradiography. There is no significant difference in GLP- 1R distribution between intact and GDX male rats (2 way ANOVA F=0.082; d.f 1.125; p=0.776). Data are expressed in femtornoles of labeled GLP- I retained per mg tissue equivalent. POA, preoptic area; SON, supraoptic nucleus; PVN, paraventricular nucleus; VMN, ventromedial nucleus. GLP-1R DISTRIBUTION IN THE RAT DIENCEPHALON

Intact Male (N=3) e7 GDX Male (N=4)

ARC

10 - 10 ADT RT

FIGURE 19. GLP-1 R binding sites measured in discrete brain regions of intact and gonadectomized (GDX) male rats by quantitative in vitro autoradiography. There is no significant difference in GLP-IR distribution between intact and GDX male rats (2 way ANOVA F4.082;d.f 1.125; ~4.776).Data are expressed in ferntomoles of labeled GLP- 1 retained per mg tissue equivalent. ARC, arcuate nucleus; ME, median eminence; ADT, anterodorsal thalamus; RT, reticular nucleus of the thalamus- GLP-1 R DISTRIBUTION IN THE RAT LIMBIC SYSTEM

Intact Male (N=3)

-k GDX Male (N=4)

FIGURE 20. GLP- 1R binding sites in discrete brain regions of intact and gonadectornized (GDX) male rats, assessed by quantitative in vitro autoradiography. There is no significant difference in GLP- 1 R distribution between intact and GDX male rats (2 way ANOVA F=0.082;d.f. 1.125; p=0.776). Data are expressed in ferntomoles of labeled GLP- 1 retained per mg tissue equivalent. CA 1,2 and 3, corm arnmonis 1,2 and 3 of the hippocampus; LS, lateral septum; MS, medial septum. GLP-1 R DISTRIBUTION IN THE RAT LIMBIC SYSTEM

IIntact Male (N=3) += += GDX Male (N=4)

BST BLA - T

FIGURE 21. GLP-IR binding sites measured in discrete brain regions of intact and gonadectomized (GDX) male rats by quantitative in vitro autoradiography. There is no significant difference in GLP- 1 R distribution between intact and GDX male rats (2 way ANOVA F=0.082; d.f 1.125; p=0.776). Data are expressed in femtomoles of labeled GLP-1 retained per rng tissue equivalent. DG, dentate gyms of the hippocampus; SFO, subfomical organ; BST. bed nucleus stria terminalis; BLA, basolateral amygdala. 2.5 DISCUSSION With little information available on GLP-1 action in the mouse, we have attempted to explore potential roles of GLP-I in mice, by examining the neuroanatomical localization of its receptors. We have presented for the frst time, a comprehensive visualization of GLP-1 receptor binding in different brain regions of the mouse brain. From this, we have set out to understand physiological relevance of central sites of GLP-1 action as it may potentially have important roles in various neuronal systems. In examining autoradiograms of GLP- 1R +/+ mouse brain, we observed that there are both similarities and differences in GLP-1 receptor distribution pattern between mice and rats. For example, while previous studies report an absence of GLP-1 receptors in the rat hippocampal dentate gyms, in our study, we observed a high density of GLP-1 receptors in the mouse hippocampal dentate gyms. To ascertain whether this difference is due to experimental error, we used the same autoradiographic procedures in rat brain sections to test whether they yield results that are consistent to the ones previously described by Goke et al. (1995)~'. Autoradiograms obtained from rat brain were identical with the ones described by Goke et d. (1995)60. These findings confirm that our experimental methodology is valid. As well, they suggest that GLP-1 receptor distribution is species-specific. Several similarities and dissimilarities are evident when comparing GLP- 1 receptors distribution between the mouse and rat brain. In both mice and rats, GLP-I receptors are present in the diencephalon and in different structures of the limbic system, though in varying amounts. In the diencephalon, GLP-1 receptor distribution is similar in both species. There is a dense accumulation of [i2S~]-~~~-lbinding in the preoptic area, supraoptic nucleus, paraventricular nucleus, ventrornedial nucleus, and the arcuate nucleus. A moderate density of ['25~]-~~~-1binding was observed in the mediobasal hypothalamus. The demonstration of intense ['25~]-~~~-1labeling in hypothalamic areas that are known to regulate appetite is consistent with previous studies that show a centrd GLP- 1-mediated inhibition of food intake in mice I70 . In the limbic system of both species, the highest densities of ['25~]-~~~-1binding were detected in the lateral septum and in the subfomical organ. Moderate densities of [12'1]- GLP-1 binding were detected in the bed nucleus of stria terminalis, nucleus solitarius tractus, hdus striati, and the basolateral arnygdala. Furthermore, unlike the rats, where GLP-1 receptors are known to be restricted only to the CA3 subregion of the hippocampus, GLP-1 receptors are present in all CA1,2 and 3 subregions of the mouse hippocampus, though in varying amounts. Autoradiograms obtained from GLP- 1R +/+ mouse brain sections showed widespread and well-defined distribution of GLP-1 receptors in the diencephalon and various structures of the limbic system. These results are consistent with the hypothesis that GLP-1, as a neuromodulator, might have a role in influencing neuroendocrine control of the pituitary gland, specifically in relation to the HPA and HPG systems. GLP-1 receptor binding is high in the region of the CRF-producing parvocellular PVN neurons and in the GnRH-producing anterior POA neurons. While these results do not constitute proof that GLP-1 is involved in the regulation of the neuroendocrine peptides produced by these two regions of the brain, the data are clearly consistent with previous observations that both GnRH and hypothalamic CRF release can be modulated by GLP-1 10'8. Our data also appear to validate the GLP-1R -1- mouse model. Autoradiograms of brain sections taken from GLP-I R -/- mice show a complete absence of [12S~]-~~~-lbinding (figure 5, top row). The absence of ['Z5~]-~~~-linteraction to its own receptor or to alternate receptors indicates that GLP-1 signal is lost and therefore, the GLP-1 R -/- mouse model is a reliable negative control tool to Merexamine the central roles of GLP- 1. Previous studies speculating that GLP-I and exendin-4 act on the same receptor system are confirmed in the present study. The evidence that [125~]-exendin-4binding is absent in GLP-1R -/- mouse brain sections, but present in the same anatomical regions as [125~]-~~~-1in GLP- 1 R +/+ mice (figures 5 and 6, column 3), supports the view of a shared receptor for GLP-1 and exendin-4. Our results fail to demonstrate any distinct receptor population specific for exendin-4 since [125~]-exendin-4binding was completely displaced by co-incubation with non-radiolabeled GLP- 1. High resolution autoradiography using [12'1]-exendin-4 to label presumptive GLP- 1 receptors demonstrated that even in areas of the brain that appear to be target sites for this class of peptides, there is considerable anatomic specificity in the distribution of the binding sites. In the hippocampus, strong labeling is observed in the dentate gyms; but this is not localized directly over the granule cell layer. Instead, binding is concentrated over the relatively cell poor region which contains the granule cell dendrites. No labelling at all is observed over the pyramidal cells of the CA 3 fields of the hippocampus (figure 8). In the medial habenula, the periventricular region of the nucleus has a much lower concentration of binding than the more cell-dense areas that border on the lateral habenula (figure 9). Similar results were obtained in the septum, in which the dorsolateral apex of the lateral septa1 nucleus was found to have by far the highest labeling density (data not shown). This heterogeneity in binding site distribution suggests that GLP-1 may have specific, discretely localized actions within the limbic system. The association of particularly high GLP-I binding site concentrations with structures involved in afferent (dentate gyms) and efferent (septum) connections to the hippocampus raises the possibility that GLP-1 may have an important role in modulating hippocampal activity. The hippocampus has a vital role in the regulation of responses to stress, and is also involved in the regdation of gonadotrophin secretion, further strengthening the hypothesis that GLP-1 might be involved in regulating both these functions in the mouse. In summary, GLP-1 receptors are primarily located in welI-defined regions of the autonomic and limbic systems, suggesting that it may have a role as a neuromodulator, neurotransmiaer and/or behavior determinant. We have also detected GLP- 1 receptors in the region of the CRF-producing parvocellular PVN neurons and the GnRH-producing mPOA neurons. These findings are in agreement with our hypothesis that GLP-1 is involved in the neuroendocrine control of the pituitary gland. There appears to be a shared receptor for GLP- 1 and its agonist, exendin-4. Finally, the complete lack of GLP-I binding in the brain of the GLP-1R -/- mouse is consistent with the view that this animal does indeed lack functional receptors for GLP-1 and is therefore a valid tool with which to fiuzher explore the roles of GLP-1 in the central nervous system. CHAPTER III CHARACTERIZATION OF THE GLP-1R 4- MOUSE MODEL

3.1 PRELIMINARY OBSERVATIONS IN GLP-1R -1- MICE The GLP-IR -/- mouse model was developed by Dr. Louise A. Scrocchi and Dr. Daniel J. Drucker (Division of Bmting and Best Diabetes Center, The Toronto Hospital Research Institute). The GLP-1R -1- mouse model is a valid negative control since GLP-1 cannot interact with its own receptor or, apparently, with dternate high-affinity receptors in the brain to transduce a GLP-I signal. In addition to impaired responses in standardized tests of feeding behavior, GLP-IR - /- mice may also have slightly impaired reproductive functions. Preliminary observation of a reduced litter size invariably seen in GLP-1R -/- mice, suggests that loss of GLP-1 function might result in subtle loss of normal reproductive function. The recent demonstrations of a GLP- 1-induced GnRH release following GLP- 1 administration to both the rat hypothalamus and the immortalized hypothalamic GnRH-producing GTI-7 cellst0,confirm that GLP-1 has a regulatory role within the HPG axis. The evidence for strong [125~]-~~~-1receptor binding in the anterior preoptic nucleus and in the parvocellular region of the PVN (see chapter 2) is also consistent with a neuroendocrine role for GLP-1. To Merexplore this hypothesis, we examined the consequences of the GLP-1R -1- mutation for normal development and the regdation of the HPG and HPA axes.

3.2 OBJECTIVE We characterized the growth rates and gross wet weights of a number of organs in GLP-1 R -/- and GLP- 1R +/+ mice. In addition, to investigate whether the loss of GLP-1 receptors causes overt abnormalities in the functions of the HPG and HPA axes, we determined the onset of puberty, and examined gonadal histology in both sexes and cyclicity in the females. By radioimmunoassay (RIA), we measured circulating estradiol concentrations in the females at proestrus and diestrus. Testosterone and corticosterone levels were measured by RIA, and compared in GLP-IR -/- and GLP-IR +/+ mice subjected to three different stress conditions: non-stress, mild stress, and severe stress. Serum testosterone and corticosterone levels were also measured in mice subjected to chronic glucocoaicoid treatment, to determine whether there was any effect of eliminating GLP-I sensitivity on the feedback control of HPA hction. 3.3 MATERIALS AND METHODS (i) Animal Breedinz Male and female GLP- 1R +/+ mice and GLP- 1R -/- mice were housed under standard environment conditions (5 per cage) and according to genotype. For breeding purposes, two females were housed per male for four weeks according genotype, and subsequently removed and housed individually. At parturition, number of homozygous GLP-1R +/+ and GLP-1R - /- pups were counted and allowed to remain in their respective cages to be nursed by their appropriate mothers until time of weaning (three weeks post-parturn). Purina Rat Chow and water were fieety accessible, and a schedule of 12-hour light, 12-hour dark cycle was maintained with lights on at 0700 hrs.

(ii) Pup Inmectinn The growth of pups was closely monitored and their weights measured daily until they reached adulthood (6 weeks of age). Female pups were regularly inspected for timing of vaginal opening, which is a useful indicator of pubertd mset. Vaginal smears were taken daily at 0900 hrs for a period of 24 days (6 complete estrous cycles) to monitor cyclicity. Smears were obtained by flushing 50 p1 of deionized double distilled water gently into the vagina and immediately retrieving the water with a plastic pipette tip. Droplets of vaginal fluid were smeared onto pre-cleaned glass slides (VWR Cadab) and dried for microscopic examination.

(iii) Tissue Dissection Total body weights (TB W) of aged-matched and sex-matched adult GLP-LR +/+ and GLP-1R -1- mice were measured immediately before being sacrificed by decapitation. Females were selectively sacrificed at the proestrous and diestrous stage of the estrous cycle, as established by vaginal smears taken 0900 hrs on the day of sacrifice. Once sacrificed, male and female gonads and steroid-dependent accessory sex glands (epididymis, prostate gland, seminal vesicle, and uteri) were dissected free of surrounding fatty tissue, frozen over a sheet of aluminum foil covering dry ice and finally weighed on a Sartorius Supermicro Balance. (iv.) GLP-1 rece~torA utoradiomap& To determine whether GLP-1 receptors are present at gonadal sites, we performed autoradiography on testicular and ovarian sections as previously described (see chapter two). Testes and Ovaries were dissected fiee hand, frozen and embedded in OCT embedding matrix and then allowed to equilibrate to a temperature of -27°C. Sections of 14pm thickness were collected for ['25~]-~~~-1 receptor autoradiography

(v.) Gonadal Hisloloq Frozen ovaries collected from proestrous GLP-1R +/+ and GLP-1R -/- mice were embedded in OCT embedding matrix, transferred to a cryostat (Reichert-Jung Frigocut 2800E) and allowed to equilibrate to a temperature of -27°C for 30 minutes. Once ovary- embedded matrix had solidified, it was then mounted onto cryostat specimen holder and 20p m sections were cryosectioned, thaw-mounted onto pre-cleaned glass slides coated in 0.02 mg/ml poly-r-lysine (Sigma-P 1666, High MW), and immediately re-frozen on a cold metal platform at -27OC. After sections were collected, slides were dehydrated in 100% ethanol for 5 minutes, followed by a 2-minute dehydration in 95%, 80% and 70% ethanol and a 2- minute hydration period in deionized double distilled water. Slides were then dipped in cresyl violet for 10 minutes (cresyl violet stock was prepared by dissolving 0.5g of cresyl violet in 500 ml of double distilled water and then by filtering appropriate amount for use), briefly dipped in deionized double distilled water, followed by dehydration in 70% ethanol for 1 minute and in acid alcohol (1 50 ml of 70% with added 3 drops of glacial acetic acid) for 30 seconds. Slides were immersed in 80% ethanol for 1 minute and then 2 minutes in 95% and 100% ethanol. Finally, slides were submerged in xylene for 5 minutes and when dried, coverslipped with Paramount (Fisher).

(vi.) Radioimrnunoassav Serum corticosterone levels of male and female GLP-1R -/- and GLP-1R +/+ mice were measured using Coat-A-Count Rat Corticosterone kit (Diagnostic Products Corporation, LA, CA). Blood was collected by intracardiac puncture, maintained at 4OC, and centrifuged at 3000 rpm for 15 minutes. Mer centrifugation, the serum portion was separated and stored in sterile 1 mL eppendorf tubes at -20°C until the day of assay. Blood was collected from GLP-1 R -/- and GLP-1R +/+ rnice under three different stress conditions: non-stressed, mild stress, and severe stress. Severe stress was achieved using volatile anesthesia, which strongly activates the KPA axis in small rodents. Mice were anesthetized with ~etofane~and, 10 minutes later while they were still under the anesthetic, blood was collected by venipuncture of the lateral tail vein. Severely stressed mice were allowed to recuperate for 7 days after which, they were sacrificed under mild stress conditions. Mild stress was defined as transport by elevator (a mild stress factor), in their home cages, fkom the animal housing room to our laboratories on the 4th floor of the Max Bell Research Center of the Toronto Hospital Research Institute. Non-stressed animals were sacrificed in the animal housing facilities within 60 seconds of removal from their home cages. Blood was also collected fkom rnice who have been chronically treated with a high dose (1 mg/mL) of glucocorticoid (hydrocorticosterone) in their drinking water for 7 days. Females were selectively sacrificed at the proestrous and diestrous stages of the estrous cycle, as established by vaginal smears taken at 0900 hrs on the day of sacrifice. Serum testosterone and estradiol levels of male and female GLP-1R -/- and GLP-1R +/+ mice were also measured by Coat-A-Count Rat Testosterone and Estradiol kits (Diagnostic Products Corporation, LA, CA) .

(vii.) Statistical AnaZvsis All data are presented as means + SEM. Statistical analysis was performed using PC- compatible microcomputer programs [SPSS for Windows; and Sigmastat (Jandel Scientific)]. Two group comparisons were made using Student's t-test. Multiple group comparisons were made using ANOVA, followed by Duncan's multiple range test. Differences between means were considered statistically significant at the p c 0.05 level (two-tailed). 3.4 RlESUlLTS (i.) Deficits in GLP-I R 4- reproductivefunction Adult male and female GLP-1R -/- mice appeared normal and were phenotypically indistinguishable fiom GLP-IR +/+ mice. When monitored fiom birth, GLP-1R -/- mice were viable and developed normally until adulthood. When total body weights (TBW) of wild-type pups were measured until adulthood, the males reached a slightly larger TI3 W than the females, which is normal for mice. Male GLP-1 R -/- and GLP- 1R +/+ mice exhibited TB W of and 3 1SO + 0.83 g and 30.96 + 0.81 g (t=OS 19; p=0.610), respectively, whereas female GLP-1R 4- and GLP- 1R +/+ mice exhibited TBW of 25.40 f 0.95 g and 27.69 + 0.43 (t=2.45; p=0.0209), respectively (figure 24, table 3). Growth rates in GLP-IR -/- and GLP- 1R +/+ mice were similar (figure 22). In females however, at the time when growth rates in both GLP- 1R -/- and GLP- 1R +/+ mice were indistinguishable, reproductive hction appears to be subtly impaired in GLP-1 R -/- animals. While female GLP-1R +/+ mice reached puberty (indicated in this species by vaginal opening and the onset of the first reproductive cycle) at post-partum Day 38 (the expected time for pubertal maturation in female mouse), a 2-day delay was consistently observed in GLP-IR -/- females (figure 23). A reduced litter six in GLP-1R -/- mice was also observed, so that approximately 60% of all pups born were homozygous GLP- 1R +/+. Despite these subtle impairments, cyclicity in female GLP- 1R -/- mice was normal once vaginal opening had occurred and was indistinguishable from that of female GLP- 1R +/+ mice. Paired gonadal weights, expressed per g TB W, were slightly reduced in GLP-1 R -/- mice as compared to GLP-1 R +/+ animals. Testicular weights of GLP- I R -/- and GLP- 1R +/+ mice were 5.29 + 0.19 mg and 6.83 F 0.33 mg, respectively (k4.26; p=0.0004), whereas ovarian weights of proestrous GLP-1R -/- and GLP-1R +/+ mice were 0.74 + 0.067 mg and 0.89 + 0.057 mg, respectively (t=1.48; p=0.1503). Male accessory sex glands [epididyrnis (t=1.06; p=0.3005), prostate gland (P1.33; p=0.2006), and seminal vesicle (~4.26; p=0.0004)] weights were all slightly reduced in GLP-IR -1- mice as compared to GLP-1R +/+ controls, but uterine weights at proestrus were greater in female GLP- 1R -/- mice (6.87 + 0.87 mg) than in GLP-1R +/+ females (4.84 + 0.28 mg) (t=2.86; p=0.008). Results are graphically represented in figures 25 and 26, and summarized in table 3. (ii.) HistoZop qf Testes and Ovaries To elucidate possible pathways of GLP-1 action in the control of reproductive function, we examined whether GLP-1 interacts directly with receptors in gonadal tissues. Autoradiograms revealed no detectable ['"I]-GLP- 1 binding, suggesting that GLP- I receptor concentrations in these tissues are likely to be low and therefore that GLP-I probably does not affect testicular and ovarian function through a direct action. Histological analysis of gonads reveal normal morphology of testicular leydig and sertoli cells and ovarian grandosa and thecal cells in GLP- 1R -1- mice. Ovaries taken fiom GLP-1 R -1- female mice, however, contained fewer developing follicles, Graafkm follicles and corpora lutea (figure 27).

(iii) Estradiol. Testosterone. and CorficosteroneRadioimmunoassay Using RIA, we measured serum estradiol levels from GLP-1R -/- and GLP-1R +I+ females at times when circulating levels of estradiol are at maximum (proestrus) and minimum (diestrus). At proestrus, estradiol levels were reduced in GLP-1R -/- females (13.43 k 8.4 1 pg estradioVmL serum) in comparison to GLP- 1R +/+ females (46.18 t 19.71 pg/mL), although considerable variability was observed in these measurements, in both genotypes (t=7.10; p=0.0 138). Conversely, at diestrous, estradiol levels in GLP- I R -/- (8.80 2 2.88 pg/mL) and GLP-1R +/+ females (6.60 t 3.62 pg/mL) were indistinguishable (e0.744; p=0.4633). Results of serum estradiol measurements are graphically presented in figure 28 and summarized in table 4. Testosterone levels in male GLP-1R -1- rnice were dramatically reduced in comparison to male GLP- 1R +I+ mice. Under non-stress condition, testosterone levels were lower in male GLP-1R -/- rnice (1 -24 rt 0.10 ng/mL) than in male GLP- 1R +/+ mice (5.52 + 0.53 ng/mL) (~4.83;p=0.0003). Similarly, under mild stress condition, testosterone levels were lower in male GLP- 1R -1- mice (1.28 + 0.5 1 nglmL) than in male GLP- 1R +/+ mice (10.8 + 2.12 ng/rnL) (e5.11; p=0.000 1). With chronic glucocorticoid treatment, testosterone levels were also reduced in male GLP-IR -1- mice (0.91 2 0.06 ng/mL) as compared to male GLP-1R +I+ mice (5.09 + 0.23 ng/mL) (H.34; p=0.0018). Results of serum testosterone measurements are graphically presented in figure 29 and summarized in table 5. To determine whether loss of GLP-1 receptors afZected hctions related to stress, we measured senun corticosterone levels in GLP-1 R -/- and GLP- 1R +I+ mice under different stress conditions. Under non-stress conditions, circulating levels of corticosterone were markedly reduced in GLP-1R -1- males (60.7 + 11.8 ng steroid/ml serum) as compared to GLP-1R +/+ males (243.9 2 29.4 ng/mL) (1~4.50;p=0.0001). In diestrous GLP-1R -1- females, corticosterone levels were slightly reduced (143.7 + 27.4 ng/mL) in comparison to GLP-1R +I+ diestrous females (1 89.3 + 46.3 ng/mL) (e0.648; p=0.5292). Under mild stress condition, corticosterone levels in male GLP-1R -1- mice (82.7 i 13.5 ng/rnL) were relatively lower than in male GLP-IR +/+ mice (227.2 t 28.4 ng/mL) (~1.06;p=0.0018). Similarly, corticosterone levels were lower in proestrous GLP-1R -1- females (166.7 + 27.4 ng/rnL) than in proestrous GLP-1R +I+ females (207.3 + 56.3 ng/mL) (H.33; p=0.0138). The deficit in circulating corticosterone levels in GLP-1R -1- mice was overcome under conditions of severe stress. Under anesthetic-induced stress, corticosterone levels were higher in male GLP-1R -1- mice (529.4 + 150.5 ng/rnL) than in male GLP-1R +I+ mice (166.5 + 11.0 ng/rnL) (t=3.45; p=0.0001). Chronic treatment with glucocorticoids caused an overall reduction in basal corticosterone levels, but the effects were greater in GLP-1R -/- males (22.6 + 13.0 ng/mL) than in GLP-I R +I+ males (74.8 2 13.2 ng/mL) (~2.79;p=0.0209). Results of serum corticosterone measurements are grzphically presented in figure 30 and summarized in table 6. ANIMAL GROWTH CURVES

0 5 10 15 20 25 300 5 I0 15 20 25 30

Days post-weaning

FIGURE 22. Body weight gain in male and female GLP-1 R +/+ and GLP- 1R -/- mice. Body weights were measured regularly at time of weaning (3 weeks post-parturn) until mice reached adulthood. There are no significant differences in total body weight gain between GLP- 1 R +/+ mice and GLP- 1R -/- mice in males (p=0.6 10) and females (p=O.716). ONSET OF PUBERTAL MATURATION

Female GLP-1 R +/+ (N=26) 3 Female GLP-I R -1- (N=12)

GLP-I R +/+ GLP-I R 4-

FIGURE 23: Ages in days at onset of pubertal maturation. In rodents, onset of puberty is indicated by vaginal opening and occurs at approximately 6 weeks of age. In the present study, female pups were regularly inspected for vaginal opening. This graph illustates that GLP-l.R -/- females reach puberty later (by a 2 day delay) than GLP- 1R +/+ females. I WEIGHT MEASUREMENTS 1

Male GLP- 1R -/- (n= 13) 3 1.50 k 0.83 FemaIe GLP- 1R +/+ (n=22) 27.69 _+ 0.43 Female GLP- 1R -/- (n=8) 25.40 k 0.95 .:*...... x.:. ..-Xn,..,,...... , .> ,.. .*. ..A ~*~@p~~~>~;~~p$g3;g$~$~4x$x~$yy$v.....,....A. ,. . .>.., .*. / ...... -...... Male GLP- 1R +/+ (n=8) I 45.8 + 6.5 X lo-;' bide GLP-1R -/- (n=13) Female GLP- 1R +/+ (n=22) Female GLP- 1R -/- (n=8)

1 Male GLP-1R +/+ (n=8) I 0.40 t 0.06 I Male GLP- I R -/- (n=13) Female GLP- 1R +/+ (n=22) Female GLP- 1R -/- (n=8)

Female GLP- 1R +/+ (n=22) 0.89 + 0.057 I Female GLP- 1R -/- (n=8) I 0.74 + 0.067 I

I Female GLP- 1R -/- (n=8) I 6.87 + 0.87 I

I Male GLP- 1R -/- (n= 13) I 5.29 k 0.19 I

1 Male GLP- 1R -/- (n=13) I 0.98 k 0.08 I

Male GLP- 1R -/- (n= 13) 2.43 + 0.16

I Male GLP- 1R -/- (n=13) I 4.80 + 0.35 I

TABLE 3. Total body weight (TBW) and dissected tissue weight measurements taken from female and male GLP-1 R +/+ and GLP-1 R -/- mice. All weights were taken when animals were 6 weeks of age and tissue weights are expressed as weighed per TBW. WEIGHT MEASUREMENTS

Male GLP-I R +I+ (N=8) 7Male GLP-1R -1- (N=13)

Female GLP-IR +I+ (N=22) Female GLP-IR -1- (N=8)

Total Body Weight I

0.09 0.09 Pituitary Gland

Adrenal Gland I -

FIGURE 24. Composite graph showing mean @EM) total body, pituitary gland and adrenal gland weights of male and female GLP-IR +/+ and GLP-1 R -/- mice, as measured at 6 weeks of age. *, Statistically significant difference (p10.05, Student's t-test). Pituitary and adrenal glands were weighed and are expressed per total body weight. OVARIAN AND UTERINE WEIGHTS

m Female GLP-1 R +I+ (N=22) i Female GLP-1R 4- (N=8)

Ovary Uterus *-

FIGURE 25. Composite graph showing mean @EM) ovarian and uterine weights of proestrous GLP- 1 R +/+ and GLP- 1 R -/- females at 6 weeks of age. *, Statistically significant difference (p

Male GLP-1 R +I+ (N-8) r- ! Male GLP-1R 4-(N-13)

Testes I Prostate 1

Epididymis Seminal Vesicle I a3

FIGURE 26. Composite graph showing mean &SEMI weights of testes and androgen-dependent accessory sex glands taken fiom male GLP- 1 R +/+ and GLP- 1R -/- mice, as measured at 6 weeks of age. *, Statistically significant difference (pc0.05,Student's t-test). All weights were measured and are expressed per total body weight. Follicular and Corpora Luteal Counts

Female GLP-I R +I+ (n=5)

Female GLP-I R -1- (n=6)

Developing Corpora Follicles Lutea

FIGURE 27. Number of developing follicles and corpora lutea per ovary, taken &om female GLP- I R +/+ and GLP- 1R -/- mice at proestrus of the estrous cycle. In GLP- 1 R -/- females, ovaries contain fewer follicles suggesting that the reduction in litter size results fiom production of a reduced number of ova per cycle. *, Statistically significant difference (p

Genotv~e Estradiol Levels h@mL serum)

GLP- 1 R +I+ (n=5) GLP- 1 R -/- (n=5)

TABLE 4. Table summarizing differences in serum estradiol levels in proestrous and diestrous female GLP- 1R +/+ and GLP- 1R -1- mice. Estradiol concentrations were measured by radioirnmunoassay and data are presented as mean + S.E.M. ESTRADIOL RADIOIMMUNOASSAY

Female GLP-1 R +I+ !7Female GLP-1R -1-

Proestrus

15 Diestrus

FIGURE 28. Composite graph showing mean @EM) serum estradiol levels of proestrous and diestrous female GLP- 1R +/+ and GLP- I R -/- mice, as measured by radioimmunoassay. *, Statistically significant difference (~4.05,Student's t-test). I TESTOSTERONE RADIOLMRlUNOASSAY I GenotvDe Testosterone Levels (@mL serum) 1 Male GLP- 1R +I+ (n=6) I Male GLP- 1R -1- (n=5)

-- eno ow I Testosterone Levels (ndmL serum) Male GLP- 1R +/+ (n=8) 10.8 t 2.12 Male GLP-1 R -1- (n=7) 1.28 + 0.5 1 Testosterone Levels !n mL serum) I Male GLP- 1R +/+ (n=4) Male GLP- 1R -1- (n=4)

TABLE 5. Table summarizing differences in serum testosterone levels of adult male GLP- 1R +/+ and GLP- 1R -1- mice when mildly stressed and glucocorticoid-treated. Testosterone concentrations were measured by radioimmunoassay and data are presented as mean + S.E.M. TESTOSTERONE RADIOIMMUNOASSAY Male GLP-1R +/+ 0 Male GLP-1R -1-

Non Stress (8 weeks old)

-! 15 l5 1 Mild Stress ! I 12 -j T (I3weeks Old' 12 1 i-

10 10 Glucocorticoid-Treated

FIGURE 29. Composite graph showing mean @EM) serum testosterone levels of male GLP- 1 R +/+ and GLP- 1R -/- mice, as measured by radioimmunoassay, under non stress and mild stress conditions, and with chronic 1 mg/mL glucocorticoid treatment. *, Statistically significant difference (pc0 .O5,Student's t-test). CORTICOSTERONE RADIOIMMUNOASSAY

Genotye Corticosterone Levels !nghnL serum)

Male GLP- 1R +I+ (n=6) Male GLP- 1R -/- (n=5) Diestrous Female GLP- 1R +I+(n=7) Diestrous Female GLP-Z R -/- (n=7)

Genotvpe Corticosterone Levels !n~/mLserum)

Male GLP- 1R tlt (n= 12) Male GLP-1R -/- (n=l1) Proestrous Female GLP- 1R +/+ (n=7) Proestrous Female GLP-1 R -1- (n=9)

Corticosterone Levels [ng/mL serum)

Male GLP-1 R +/+ (n=6) Male GLP- 1R -/- (n=5)

Corticosterone Levels (ng/mL serum)

MaIe GLP- 1R +/+ (n=6) Male GLP- 1R -/- (n=6)

TABLE 6. Table summarizing differences in serum corticosterone levels of male and diestrous and proestrous female GLP-1R +/+ and GLP-1R -1- mice, as measured by radioimmunoassay, under three different stress (non-stress, mild stress, severe stressed) conditions and with chronic 1 mg/mL glucocorticoid treatment. CORTICOSTERONE RADIOIMMUNOASSAY

Male GLP-1 R +I+ i7- Male GLP-1 R -1- Diestrous Female GLP-1 R +I+ I Diestrous Female GLP-1R -1- Proestrous Female GLP-1 R +I+ Proestrous Female GLP-1 R -1-

T Non Stress Mild Stress I T

300 750 Glucocorticoid-Treated Severe Stress 250 - T

200 -

roo 1 r *

FIGURE 30. Composite graph showing mean @EM) serum corticosterone levels of male and &estrous and proestrous female GLP- I R +/+ and GLP- 1 R -/- mice, as measured by radioimmunoassay, under three different stress conditions (non stress, mild stress, severe stress) and with chronic 1 rng/mL glucocorticoid treatment. *, Statistically significant difference (~~0.05,Student's t-test). 3.5 DISCUSSION The elimination of GLP-1 signaling in the GLP-1R -1- mouse model results in subtle impairments to the male and female reproductive systems. Impairments in reproductive functions could conceivably be attributed to many factors, including loss of GLP-L action at the CNS and/or at peripheral, including gonadal, sites. However, autoradiograms of testes and ovaries incubated with ['2S~]-~~~-lshowed no evidence of specific, saturable [lZ51]- GLP-1 binding, suggesting that GLP-1 receptors are either not present or are present at very low concentrations in these organs. This suggests that GLP-I is unlikely to exert its effect directly at the peripheral level. The pituitary gland secretes (LH) and follicle stimulating hormone (FSH) which, in turn, are known to coordinate progression of the estrous cycle. The secretory profiles of LH and FSH in post-pubertal female mice are known to be associated with various phases of their estrous cycle48. The 4-day estrous cycle is characterized by spontaneous ovulation followed by luteal development, luteal regression, and a proestrus period. During the estrous stage of the cycle, there is a large preovulatory surge of LH shortly before ovulation which usually coincides with the onset of sexual receptivity by females. The perkc! preceding this surge is usually called proestrus, and is characterized by increasing secretion of estradiol from rapidly growing ovarian follicles. Moreover, the pattern of FSH secretion during the estrous cycle consists of two peaks, one coincident with the LH surge and a secondary peak that occurs 24 hours later. The period after ovulation and after the second FSH peak is known as diestrus which lasts 48 hours. To determine whether loss of GLP-1 sensitivity interferes with normal cyclical LH and FSH secretion, vaginal smears fiom both female GLP-1R +I+ and GLP-IR -/- mice were obtained on a daily basis to monitor cyclicity. In rodents, different stages of the estrous cycle can be identified by microscopic observation of different vaginal cell types obtained fiom the vaginal lumen. During vaginal estrous, comified epithelial cells are present for 36 hours and then the vaginal epithelium is invaded by leukocytes, resulting in a vaginal smear pattern consisting of leukocytes with a few nucleated epithelial cells which lasts through metestrus and diestrus, together lasting about 48 hours. The subsequent stage of the estrous cycle, proestrus, is characterized by an increase in the number of nucleated epithelial cells, as well as a sharp decrease in the number of leukocytes. Vaginal smears obtained during 24 consecutive days, spanning six complete estrous cycles, revealed a completely normal cycling pattern in GLP-1R -/- females. The normal cyclicity in female GLP-1R -/- mice suggests that biological actions of gonadotrophins are not overtly altered by the loss of GLP- 1 signaling. To further examine whether GLP-1 has a role in regulating gonadotrophin action, the histology of gonads taken from GLP-1 R +I+ and GLP-I R -1- mice was examined. FSH and LH are important for gonadal development. For example, loss of FSH and LH resulting fiom hypophysectomy (removal of the pituitary gland) leads to testicular atrophy. Restoration of testicular hction can be effected by administration of pituitary extracts of pituitary gonadotrophins. We found a slight reduction in testicular size and weight in GLP-1R -/- males, suggesting that gonadotrophin secretion might be impaired in these animals. Consistent with this hypothesis, we also observed reduced serum testosterone and a reduction in the organ weights of the epididymis, prostate gland and seminal vesicles, all testosterone target tissues. However, the histology of the interstitial Leydig cells and the tubular Sertoli cells in the testes was apparently normal, suggesting that if there is indeed any deficit in the regulation of gonadotrophin release it is probably subtle. The actions of LH and FSH on the female gonads are very complex, due mainly to the variation in ovarian cell types and differing hormone secretion by these cell types as the female progresses through the various phases of her reproductive cycle. Estradiol secretion involves both LH and FSH acting differentially on the thecal and granulosa cells of the follicle. Thecal cells respond to LH stimulation with increased synthesis and secretion of androgen. The secreted androgens diffuse across the basement membrane and are taken up by the granulosa cells. Under tfi2 influence of FSH, these granulosa cells convert the androgens into estradiol which becomes secreted into the systemic circulation. Circulating estradiol will consequently regulate production and secretion of pituitary LH and FSH. While high levels of estradiol functions to inhibit further release of LH and FSH, low levels of estradiol promotes release of LH and FSH. In female GLP-1R -1- mice, it seems that actions of LH and FSH are subtly impaired as well. In proestrus, the occurrence of an estradiol concentration below physiological levels in the GLP-1R females suggests that there might be an interference in the mechanism by which ovarian cell types function to synthesize estradiol. The finding of a reduced number of developing follicles in the ovaries at proestrus is also consistent with the hypothesis that there is reduced ovarian stimulation. The reduction in litter size probably, therefore, results fiom release of a reduced number of ova per cycle. Although GLP-1R -/- males have slightly smaller testes, spermatogenesis continues nonetheless, so it is unlikely that any normally released ova remain unfertilized after mating. The most reasonable conclusion, overall, is that there is a subtle impairment of reproductive hction in both sexes in GLP-1R -1- mice; but this impairment is not sufficient to abrogate normal reproductive activity. All that it does is to slightly retard the time of normal vaginal opening and reduce the trophic stimulus from the pituitary to the gonads, in both sexes, without seriously disrupting overall patterns of reproductive function. Exactly what the deficit is in the mechanisms regulating gonadal function in GLP-1R -/- rnice remains to be established. In view of the work of Beak et al. (1998)1°, it is tempting to speculate that loss of GLP-IR results in impaired GnRH release, which in turn reduces FSH and LH secretion. In view of the pulsatile nature of GnRH secretion, however, this is something that would be very difficult to characterize illy. Loss of GLP-1 input to the GnRH pulse generator could conceivably alter the frequency and/or amplitude of GnKH pulses. Such effects would be difficult to examine definitively in intact mice. It is likely that further work on these mechanisms will have to focus on in vitro models, which would allow examination of the effects of GLP-1 and GLP-1 antagonists on pulsatile GnRH release under more defined conditions. GLP-1 has also been postulated to have a role in the HPA axis. The HPA axis involves three different hierarchies of hormones: CRF, ACTH, and corticosterone. CRF is a hormone synthesized in the cell bodies of the hypothalamic paraventricular nucleus, transported within the axons of these neurons to the median eminence where it is released into the portal blood vessels supplying the anterior lobe of the pituitary gland. The output of CRF is increased dramatically by stress, through pathways that involve both the brain stem and higher CNS centers including the temporal cortex and hippocampus. CRF in turn, triggers the release of adrenocorticotrophin hormone (ACTII), which will induce the adrenal cortical cells to secrete glucocorticoids - corticosterone in rodents and in humans. Typically, when rodents are exposed to severe acute stress, they respond initially by a dramatic increase in corticosterone production. This rise in corticosterone is short-lived as it is counteracted by a regulatory mechanism. Our results show that there is a near 5-fold reduction in basal levels of corticosterone in non-stressed GLP-1R -1- mice in comparison to GLP-1R +I+ mice. A reduced corticosterone concentration was also detected in GLP- 1R -1- mice, but not in GLP-1 R +/+ mice under mild stress conditions (transporting mice to another environment). Under conditions of severe (anesthetic) stress, the opposite pattern was observed, corticosterone levels rising much higher in GLP-1R -1- mice than in GLP-1R +I+ controls, despite continuous exposure to the stressful stimulus (MetofaneTM). A possible interpretation of these apparently conflicting data is that the rise in corticosterone secretion may have been delayed in the GLP-1R -1- animals, which would have also delayed the compensatory negative feedback response in terms of pituitary ACTH release - resulting in higher observed corticosterone levels at 10 minutes after the onset of anesthesia. It is also conceivable that there might be reduced levels of testosterone-induced CBG (Corticosterone Binding Globulin) in the male GLP-1R -1- mice, thereby giving rise to increased levels of circulating free corticosterone - resulting in negative feedback response to lower corticosterone concentrations. Alternatively, it is possible that the reserves of CRF and ACTH might be greater in GLP-IR -1- animals, allowing for a greater rise in the levels of these hormones once a stress response is elicited. Additional studies on the time course of the stress response in both genotypes will be required to resolve this issue. In any case, whatever the mechanism it is clear from the data available that loss of GLP-1 function in some way impairs normal stress responses. Thus, GLP-1 R -1- mice are unresponsive and fail to exhibit a normal stress response when placed in a novel environment; but "over-react" when severely stressed by exposure to volatile anesthetic. In summary, we have qualitatively and quantitatively demonstrated that there are deficits in both the HPG and HPA axis when GLP-1 signals are lost. Our results are consistent with the hypothesis that, in mice, GLP-1 has a role in the neuroendocrine control of the reproductive system as well as stress responses. CHAPTER IV IMPLICATIONS OF GLP-1 IN THE CONTROL OF APPETITE

. -- - - - . 4.1 ROLE OF CCK-8 IN APPETITE CONTROL Cholecystokinin-octapeptide (CCK-8) is a 33 amino acid peptide released by the I- cells of the jejunal mucosa in response to fatty substance in the intestinal chyme. Since its isolation, there has been extensive investigations on the possible function of CCK-8 as it is found in both the brain and the periphery of several animal species, including humans3536,313 Among its role in regulating gastric functions20.30, I04 and controlling pancreatic secretions20 and gall bladder contractions106,CCK-8 is best known to control food intake centrally and is therefore referred to as a gut-brain hormone8,127 . CCK-8 was first proposed as a satiety agent by Gibbs et a[. (1973) who noted that CCK-8 decreased food intake in starved ratss7. Several studies since then have supported this view by reporting similar reductions in food intake with exogenously administered CCK-8 in numerous species including the hamster2, human9', rhesus monkey56,dogs 26.1 55, pig 4.143, rat5,6,33 ,d ,heep32,34 As is the case for other , CCK-8 exerts its biological actions by binding to specific receptors on its target tissues. Receptors for CCK-8 are classified into two receptor subtypes: subtypes-A (CCK-8RA) and -B (ccK-~R~)~~.The development of highly selective antagonists for CCK-8RA and for CCK-8RB has not only provided specific functions for specific CCK-8 receptor subtypes, but they have also been useful tools to delineate their distributions in the CNS39,69,72,85.94 . For example, through the use of (a CCK-8RA antagonist) to selectively displace [125~]-~~~-8from CCK-8RA, Hill et al. (1987) were able to demonstrate that CCK-8RA are found primarily in the periphery 73,75 . Although CCK-8RA have been traditionally classified as peripheral receptors, they have been shown to be present in few selective areas of the CNS (medial posterior nucleus accurnbens, anterior pituitary, dorsal horn of the spinal cord) and the PNS (vagus nerve). Radiolabeled devazepide has been used to directly label CCK-8RA autoradiographically 1 6,43.74,129.183 By a similar approach, both (21-988 and L-365,260 (CCK-8RB antagonists) have provided autoradiographic mapping of CCK-8RB to be localized predominantly throughout the CNS in a pattern that parallels the distribution of CCK-8 and mRNA 24,70,109,202 . CCK-8RB have been traditionally classified as central receptors since the majority of CCK-8 receptors in the brain are of the CCK-8RB type. In the rat brain, CCK-8RB are found in highest concentrations in layers II and I11 of the cerebral cortex. They are also found in the arnygdaloic! complex, hippocampus and in the hypothalamic dorsomedial and ventromedial nucleus 112,147 In an attempt to determine brain regions that are sensitive to the suppressive feeding effect of centrally injected CCK-8, the central distribution of CCK-8 and its receptors as well as the relevance of specific brain areas for feeding regulation have been considered. Previous localization studies have clearly demonstrated the presence of both CCK-8 and receptors in the hypothalamus 120,164 . The hypothalamus itself represents a brain area at which various signals involved in the control of feeding behavior are likely integratedlo5. Support for this stems fi-om a series of classical experiments which have demonstrated that the destruction of the hypothalamic VMN results in hyperphagia, whereas electric stimulation of the VMN profoundly inhibits feeding 15,176 . In addition to VMN, the hypothalamic DMN is also known to remarkably inhibit feeding since CCK-%mediated suppression of feeding was attenuated in DMN lesioned rats12. The effects of CCK-8 have been looked at in subjects with different eating abnormalities. For example, in the genetically obese Zucker (fa/fa) rats, depolarization- evoked release of CCK-8 from the hypothalamus is much greater than that in lean littermates 118,1l9,131,173. In humans, administration of CCK decreased food intake with greater effects in obese than in non-obese humans. There is strong evidence that levels of endogenous CCK-8 are altered in various nutritional states. For example, in both rats and humans, hypothalamic CCK-8 concentrations have been reported to greatly increase in a satiated stated rather than in a fasted state91,169 . 4.2 OBJECTIVE Reports that GLP-1 is synthesized in the mouse hypothalamus and that its receptors are discretely localized in the hypothalamic feeding centers, have raised the possibility that GLP-1 may be involved in the central regulation of appetite. This is best demonstrated as centrally injected GLP- 1 causes marked reduction in food intake in GLP- 1R +/+ mice but not in GLP-1R -/- mice. Unexpectedly, no significant differences in TBW were detected in comparing weights of age-matched and sex-rnatched GLP-1R -/- and GLP-I R +/+ mice, suggesting that while disruption of GLP-1 receptors abolishes decrease in food intake, it is not associated with disturbances in TBW. One obvious mechanism to explain this apparent anomaly is that other systems involved in regulation of appetite might undergo compensatory adjustment in GLP-1R -/- mice, offsetting the loss of GLP-1 sensitivity. From the evidence briefly reviewed above, CCK-8 is an excellent candidate as a mechanism that might undergo such a compensatory modification. We therefore performed quantitative receptor autoradiography for [12S~]-~~~-8binding in the GLP-1R -/- mouse brain, to determine whether CCK-8 receptors might be up-regulated in these animals as compared to wild-type controls. 4.3 MATERIALS AND METHODS (i.) Chemicals Radiolabeled ['25~]-~oltonHunter Cholecystokinin octapeptide (CCK-8; specific activity = 2,200 CVmmol) was purchased fiom DuPont Canada Inc. (Markham, Ontario, Canada). Methoxyflurane anesthesia (MetofaneTM)was purchased fiom Pitrnan-Moore USA (Washington Crossing, NJ, USA). Autoradiographic film (Amersham ~~~erfilrn-~~)was purchased fkom Amersham Canada (Oakville, Ontario, Canada).

(ii.) Animal Preparation Age-matched male and femde GLP- 1R -/- and GLP- 1R +/+ mice were housed five per cage according to genotype and sex and were given fiee access to Purina Rat Chow and water and a schedule of 12-hour light, 12-hour dark cycle was maintained with lights on at 0700 hrs. Animals were sacrificed by decapitation and brains were immediately removed and frozen onto cryostat chucks over liquid nitrogen vapor. Pituitaries were collected and snap- frozen in liquid nitrogen. Brains and pituitaries were stored in air-tight containers at -80°C. All experimental protocols were pre-approved by the Institutional Animal Care and Use Committee of the Toronto General Hospital Research Institute.

(iii.) Tissue Sectioning Brains stored at -80°C were transferred over dry ice to a Jung Reichert HistoSTAT cryostat (Scientific Instruments, Inc., Buffalo, NY) maintained at -27°C. Once brains were allowed to equilibrate to cryostat temperature for 30 minutes, serial coronal sections of 20 p m were cryosectioned and thaw-mounted onto pre-cleaned glass slides (VWR Canlab) coated with 0.02 mg/ml poly-L-lysine, and immediately re-fiozen on a cold metal platform at -27'~. All slides were then stored dessicated in slide cassettes at -80'~until the day of the assay.

(iv.) CCK-8 Rece- On the day of the assay, slide cassettes were quickly transferred over dry ice fiom the -80°C freezer to a 4OC cold room, and slides were then arranged flat across rowed straws placed in a tissue culture tray lined with wet paper towel. Sections were pre-incubated in 50mM Tris-HC1 buffer (400pl/slide), pH 7.7, containing 5mM MgC12, 0.2% bovine serum albumin, imM dithiothreitol and 0.2% bacitracin for 15 minutes at room temperature. Excess liquid was drained from slides, which were then incubated in Tris-HC1 buffer containing 0.28nM of radio-labeled [i25~]-~~~-8for 2 hours at room temperature (350~ Vslide). Slides were then rinsed 3 X 5 minutes each in circulating ice-cold 5mM MgC12- added Tris-HC1 buffer, briefly dipped in ice-cold double-deionized distilled water, and finally dried in a stream of cold air overnight. Autoradiograms were prepared by exposing slide-rnounted sections to Arnersham hyperfi1rn-3~for 5 days. Films were subsequently developed at 20" for 5 minutes using Kodak D-19 developer, and then water-rinsed, fixed in 0.5% Kodak Photoflow, and finally dried.

(v.) Cornwuter-assisted Image Anahis of Receptor A utoradiog~arns Autoradiographic images generated following [125~]-~~~-8receptor autoradiography were analyzed using computerized image analysis system (MCID system; Imaging Research, St. Catherines, Ontario). The densities produced by the standard sections were characterized by ferntomol of [125~]-~~~-8bound per rng protein by scraping standard sections into 12 x 75 mm culture tubes and digesting scrapings with 100p1 of 0.3N KOH. Protein content was measured by the method of Lowry er al. (195 1)'I0. [125~]-~~~-8content was determined at 70% efficiency using an ICN Micromedic 4/600 Plus Gamma-counter.

(vi) Statistical Anahis All data are presented as the mean + SEM. Statistical analysis was performed using PC-compatible microcomputer programs [SPSS for Windows; and Sigmastat (Jandel Scientific)]. Two group comparisons were made using Student's t-test. Multiple group comparison were made using ANOVA, followed by Duncan's multiple range test. Differences between means were considered statistically significant at the p < 0.05 level (two-tailed). 4.4 RESULTS (L) Summap ofCCK-8 Receptor Distribution CCK-8 receptors were localized in highest concentrations in the cerebral cortex, the limbic regions of the amygdala and hippocampus, basal ganglia, nucleus accumbens, olfactory tubercle. In the cerebral cortex, CCK-8 receptors were highly concentrated in layers 11, In and VI. Moderate levels of CCK-8 binding were found localized to the hypothalamus and the lateral and third ventricles. In the hypothalamus, CCK-8 receptors were discretely localized in the DMN, VMN and PVN, nuclei that are involved in the integration of feeding behavior. This distribution of CCK-8 receptor localization is consistent with previously reported data for the rat.

(ii) Com~arironof CCK-8 Receptor Concentrations Quantification of CCK-8 receptor concentrations at sites that are sensitive to the suppressive feeding effects indicates that there are no significant differences in CCK-8 receptor concentrations between GLP- 1R -/- and GLP- 1R +/+ mice in males (2 way ANOVA F=1.841; d.f. 1.88; p=0.178) and in females (2 way ANOVA F=1.791; d.f. 1.80; p=0.185). Furthermore, surprisingly, we detected no sex differences in the GLP-1 R +/+ mouse group (2 way ANOVA F4.096; d.f. 1.960; p=0.790) and in the GLP- I R -/- mouse group (2 way ANOVA F=2.772; d. f. 1.72; p=0.8 94). IAutoradiograms depicting CCK-8 receptor concentrations in GLP- 1R +/+ and GLP- 1R -1- mice are represented in figure 3 1 through 33. Quantification of CCK-8 receptors is summarized in Table 7. CCK-8 RECEPTORS IN GLP-1R +/+ AND GLP-1R 4- MICE 1

Male GLP- 1 R -/- (n=6) Female GLP- 1R +I+ (n=7) Female GLP- 1R -1- (n=5)

Male GLP- 1R +/+ (n=7) 3.59 + 0.52 Male GLP- 1R -/- (n=6) Female GLP-1 R +I+(n=7) Female GLP- 1R -1- (n=5)

Male GLP- 1R -1- (n=6) Female GLP- 1R +/+ (n=7) Female GLP- 1R -1- (n=5)

Male GLP- 1R -/- (n=6) Female GLP- 1R +I+ (n=7) Female GLP- 1R -1- (n=5)

Male GLP-1 R -/- (n=6) Female GLP- 1R +I+ (n=7) Female GLP- 1R -1- (~5)

Male GLP- 1R -/- (n=6) Female GLP- 1R +/+ (n=7) Female GLP- 1R 4- (n=5)

Male GLP- 1R +I+ (n=7) 5.40 + 0.48 Male GLP-1 R -/- (n=6) Female GLP-IR +I+ (n=7) Female GLP- 1R -1- (n=5)

Male GLP-1 R -1- (n=6) 3.60 k 0.39 Female GLP- 1R +I+ (n=7) 3.34 f 0.36 Female GLP- 1R -1- (n=5) 3.13 k 0.30 . i TABLE 7. Quantification of [1L'~]-~~K-8receptor binding reveals no differences between in male and female GLP- 1R +I+ and GLP- 1R -1- mice. COMPARISON OF CCK-8 RECEPTORS IN GLP-1R +I+ AND GLP-1 R -I- MOUSE BRAIN

GLP- 1R -/-

FIGURE 31. Autoradiograms depicting no significant difference in CCK-8 receptor levels between male GLP- 1 R +/+ and GLP- 1 R -1- mouse brain. Sections were taken at the level of the paraventricular nucleus (PVN). CCK-8 RECEPTORS IN GLP-1R +I+ AND GLP-1 R -1- MICE

Male GLP-1R +I+ (N=7) a Male GLP-1R 4-(N=6)

5 Female GLP-1R +I+ (N=7) - i- I FemaIeGLP-lR%(N=5)

CTX

FIGURE 32. Comparison of CCK-8R levels measured in discrete brain regions reveals no significant difference between male and female GLP- I R +/+ and GLP-1R -/- mice (2 way ANOVA P<0.05). Data are expressed in ferntomoles of labeled CCK-8 retained per mg tissue equivalent. CA 3, cornu ammonis 3; CN, caudate nucleus; CTX, cerebral cortex; GP, globus pallidus. CCKB RECEPTORS IN GLP-1 R +I+ AND GLP-1R -1- MICE

Male GLP-IR +I+ (N=7) i! i! - Male GLP-1R 4-(N=6)

Female GLP-I R +I+ (N=7) Fernak GLP-I R 4- (W5)

PAG PVN

VMN

FIGURE 33. Comparison of CCK-8R levels measured in discrete brain regions reveals no significant difference between male and female GLP- I R +/+ and GLP-1R -1- mice (2 way ANOVA p<0.05). Data are expressed in ferntomoles of labeled CCK-8 retained per mg tissue equivalent. PAG, periaqueductal grey; PVN, paraventricular nucleus; SB, subicdum; VMN, ventromedial nucleus. 4.5 DISCUSSION GLP-1 has been postulated to represent a satiety agent in the CNS 101,132 . Supportive of this view are the expression of GLP-1 and its receptors in regions of the hypothalamus involved in control of food intake. In addition, the fmdings that centrally administered GLP- 1 caused a potent inhibition of food intake only in starved GLP-1R +/+ mice, but not GLP- 1R -/- mice, support the view that GLP-1 is involved in the central regulation of food intake. Unexpectedly however, in comparing TBW of age-matched male and female GLP-1R +/+ and GLP- 1R -/- mice, no significant differences were detected, suggesting that GLP- 1/GLP- 1R system may not be an exclusive mechanism in regulating feeding behavior. These fmdings raised the possibility that other systems that regulate appetite might be overcompensated for the loss of GLP-I function. In this chapter, we have reviewed the CCK-8 system, a system recognized to powerfidly suppress food intake. We postulated that CCK-8 sensitivity might be increased when the GLP-l/GLP-1R system is disrupted. If this were to be the case, the similarity in TBW between GLP-IR -/- mice and GLP-1R +/+ mice can be considered an effect of an enhanced CCK-8 sensitivity. To test this, we quantitatively compared CCK-8 receptor concentrations from both GLP- I R +/+ and GLP- 1 R -1- brain autoradiograms. Autoradiograms revealed no significant differences in CCK-8 receptor concentrations, indicating that the CCK system is not over-compensated when GLP-1 signals are lost. However, by comparing male and female mouse brain autoradiograms, we detected no sex differences in CCK-8 receptor concentrations. This finding was unexpected because, in rats, the situation is completely different. Concentrations of CCK-8 and its receptors vary according to sex, and sex steroidal milieu. Several areas of the rat brain have been reported to have sex differences in CCK concentration. in a study by Larriva-Sahd et al. (1986), it was shown by RIA that CCK concentrations within the medial preoptic area are higher in males than in femalesg7. Micevych et al. (1987) further demonstrated that the periventricular preoptic nucleus and the dorsal medial preoptic area on the female rat contain more CCK immunoreactive perikarya than those in the male'". However, the CCK immunoreactive cells in these nuclei were larger in male than in female. Conversely, male rats had more CCK immunoreactive cells in the in the medial arnygdala, bed nucleus of stria terminalis, and the posterior magnocellular paraventricular nucleus than did females. The reason for the apparent disparity between mice and rats with respect to the expression of CCK-8 receptors remains to be established. One possibility is that it may reflect the difference between these two species in the consequences of sexual differentiation: in rats, there is a profound sex difference in the morphology of several regions of the hypothalamus and preoptic area, including structures that contain substantial amounts of cCK97,113,120 . By contrast, the same structures do not show any overt sexual dimorphism in the mouse l I3 . It is therefore possible that the lack of a sex difference in CCK-binding in the mouse may reflect lack of any sex difference in this species in hypothalamic content or release rates for CCK. Further studies would be required to test this hypothesis. In any case, whatever the basis for the species difference, it seems clear that in mice there is no compensatory up-regulation in CCK-8 receptor levels in response to the loss of GLP-I sensitivity in GLP-1R 4- mice. This, in turn, suggests that the lack of any overt difference in TBW regulation between GLP-IR -/- and GLP-IR +/+ mice cannot be ascribed to a generalized overcompensation of other hypothalamic receptor systems involved in the control of appetite. It is possible that there might be changes in other systems involved in appetite regulation (e.g. catecholamine, NPY) but, if so, these changes would have to be selective, rather than generalized. I SEXUAL DIFFERENTIATION IN THE GLP-1R -/- MOUSE BRAIN

5.1 SEXUAL DIFFERENTIATION OF THE CNS

The gonads, the testes and ovaries, secrete androgens and into the systemic circulation. Gonadal steroids play an important role in the sexual differentiation of the CNS. Hormones secreted by the gonads are important for sexual differentiation of reproductive and non-reproductive behaviors such as taste preference, body weight regulation, feeding, learning, memory and scent marking113. Furthermore, sex differences in the neural control of a number of endocrine interactions are also the result of differential effects of gonadal hormones. In rodents, sex differences in the regulation of gonadal and adrenal functions are all primarily the result of sex differences in the hormones secreted by the gonads during early fetal and neonatal development113.

Gonadal steroids have both organizational and activationd effects in the developing CNS. Neural tissue is sufficiently plastic to respond permanently and irreversibly to androgens and estrogens during a restricted critical late fetal or early post-natal period of neural differentiation. This critical period varies somewhat between species. During this period testicular androgens exert an organizational influence on the developing CNS. Consequently, manipulation of hormone levels prenatally, or early in development, can result in irreversible neuronal development and structural differences between male and female on later adult CNS functions. During puberty or adulthood, circulating gonadal hormones can also effect sexual differentiation by impermanently and reversibly activating or deactivating specific regions of the CNS. Activational effects are relatively short-term and are dependent on the presence or absence of androgens or estrogens.

Both male and female brains have the enzyme aromatase, and therefore have the ability to aromatize testosterone into 17P-estradiol (estrogen). While a majority of the actions of androgens on the brain are mediated by estrogen aromatized in the brain fkom testosterone, androgens are directly responsible for some actions. Testosterone-derived estrogen plays a profound role in regulating the structure and fimction of many estrogen- sensitive neuronal systems in mammalian brain. Structural differences between addt male and female brains are sex-steroid sensitive, and have been reported in many species including the rat, zebra finch and humanH3. The sexually dimorphic nucleus of the preoptic area (POA) was first described in the rat brain by Gorski et al. (1978)~~.Due to differences in perinatal steroid levels, the POA in the male rats was found to be thee to eight times larger than in the female rat. On the basis of lesion experiments in rats it was found that the POA seems to be involved in aspects of male sexual behavior (e-g., mounting, intromission and ejac~lation)~~.However, the effects of lesions on sexual behavior are only slight, so it might well be that the major functions of the POA are still unknown at present. Classical estrogen receptors (ER) have been identified in discrete regions of the brain. In the neonatal rat and mouse a transient population of estrogen concentrating cells is found in the cerebral cortex187.Following this period, at postnatal day 10, the regional distribution of ER changes and becomes concentrated within the sexually dimorphic structures of the preoptic area, ventromedial nucleus, arcuate nucleus and corticomedial amygdala 113,149.187 A corollary of the hypothesis that lack of GLP-1 receptors might affect GnRH secretion is that impaired GLP-1 signaling during development might secondarily affect neuroendocrine and behavioral systems that are involved in the regulation of reproduction and the HPA axis in addthood. If, for example, development of the GnRH neurons was delayed in GLP-IR -/- mice, this could result in a delay in testicular development which would be disastrous in terms of brain sexual differentiation because this process is so temporally cued to specific "critical periods" of brain deve~o~rnent"~.The data we have accumulated suggest that there is probably no gross impairment of normal testicular differentiation in GLP-1R -/- animals: testicular morphology is normal, although the testes are small, and the prostate, seminal vesicles and epididymis have all obviously differentiated, as have the external genitalia. However, the possibility that testicular androgen production might have been sub-normal during early life, as it apparently is in adulthood, prompted us to examine the question of whether the brains of GLP-IR -/- mice had indeed undergone normal sexual differentiation. We suspected that they had, because the normal sex difference in body weight regulation was also observed in GLP-IR -/- mice. However, in view of the possibility that subtle differences in both regulation of pituitary hction and behavior might secondarily result from impairment of normal CNS sexual differentiation, it seemed important to verify that indeed there was no obvious effect of loss of GLP-1 sensitivity on sexual differentiation of the brain. As mentioned above, many of the primary morphological endpoints of sexual differentiation in the rat (e-g. the sexually dimorphic nucleus of the preoptic area113) are not sexually differentiated in mice. Therefore, these simple histological endpoints cannot be used in studies of mice. However, brain estrogen receptors are sexually dimorphic in mice as they are in rats, so we therefore decided to use this endpoint as a means of determining whether GLP- 1R -/- male mice had indeed undergone normal CNS masculinization.

5.2 OBJECTIVE We used in vituo autoradiography to compare ER concentrations sexually dimorphic brain structures in GLP- 1 R +/+ and GLP-1R -/- mice, to verify whether the loss of functional GLP- 1 receptors interferes with sexual differentiation. 5.3 MATERIALS AND METHODS (i.) Chemicals Radiolabeled 1 1p-rnethoq- 16a['25~] iodo estradiol (ME2; specific activity = 2,200 Ci/mmol) was synthesized and purified by HPLC as previously described (Zielinsky el al. 1986)~'~and was diluted to approximately 440 Ci/mmol by adding unlabeled MEt. Methoxyflurane anesthesia (Metofanem) was purchased from Pitman-Moore USA (Washington Crossing, NJ, USA). Autoradiographic film (Amersham ~~~erfilrn-~~)was purchased from Amersham Canada (Oakville, Ontario, Canada).

(ii.) Animal Preparation Age-matched adult male and female GLP-1 R +/+ and GLP- 1R -/- mice were housed five per cage according to genotype and sex phenotype, and were given access to rodent chow (Purina Rat Chow) and water ad libitum, and a schedule of 12-hour light, 12-hour dark cycle was maintained with lights on at 0700 hrs. Animals were sacrificed under MetofaneTM anesthesia, followed by decapitation. Brains were immediately removed and fiozen onto cryostat chucks over liquid nitrogen vapor and stored in air-tight containers at -80'~. All experimental protocols were pre-approved by the Institutional Animal Care and Use Committee of the Toronto General Hospital Research Institute.

(iii.) Tissue Sectioning Brains stored at -80'~were transferred over dry ice to a Jug Reichert HistoSTAT cryostat (Scientific Instruments, Inc., Buffalo, NY) maintained at -27OC. Once brains were allowed to equilibrate to cryostat temperature for 30 minutes, alternate serial coronal sections of 20 pm were cut and thaw-mounted onto pre-cleaned slides coated with 0.02 mg/mL poly- L-lysine. Slides were immediately re-fiozen on a cold metal platform at -27OC to avoid possible loss of estrogen receptor binding when exposed briefly to temperatures above O°C. All slides were then stored dessicated in slide cassettes at -80'~until the day of the assay. (iv.) Estrogen Receptor A utor~diogruvh-p Quantitative in vitro autoradiography of ER in brain sections was conducted essentially as described by Yuan et al. (199~)~~~.Quickly, brain sections were transferred over dry ice from the -80°C freezer to a 4'~cold room and were immediately treated with ice-cold [125~]-~~~2-addedTEGDB buffer, pH 7.4, containing 10 mM Tris, 1.5 rnM EDTA, 10% glycerol, 1rnM DDT, 0.5 mM bacitracin and 1 mg/ml protamine sulfate. When all sections were covered with incubation buffer containing ['25~]-~~~2(400pl/slide), the trays were closed and place in a humidified incubator (National Appliance Co., Portland, OR), preheated to 30'~. After 30 minutes incubation, slides were transferred to the cold room and excess incubation beer was drained off the slides. The slides were then loaded into slide racks and rinsed in ice-cold circulating PM Buffer (1 -0 mM KH2P04and 3.0 mM MgC12, pH 6.8) containing O.lg/L protamine sulfate for 5 minutes, and then fixed using 4% paraformaldehyde in O.lM Phosphate buffer, pH 7.2, for 5 minutes. After fixation, slides were rinsed twice in cold, circulating PMTx buffer (PM buffer with added 0.1% Triton X- 100) for 10 and 5 minutes, then rinsed twice in PM buffer for 10 and 5 minutes, and fmally dipped in ice-cold double-deionized distilled water. When fan dried at 4'~overnight, slides were exposed against Amersham hyperfilm-3~for 24 hours at 20°C and developed with Kodak D-19 developer, diluted with 1:3 water. Autoradiograms were analyzed with a computer-assisted densitometry after calibration with the autoradiographic standards. The densities produced by the standard sections were characterized by hol of [125~]-~~2bound per mg protein by scraping standard sections into 12 x 75 mrn culture tubes and digesting scrapings with 1OOp.l of 0.3N KOH. Protein content was measured by the method of Lowry et al. (1951)"~. [125~]-~~~2content was assessed at 70% efficiency using a Micromedic 4/600 Plus Gamma-counter.

(v.) Statistical Anahis All data are presented as the mean + SEM. Statistical analysis was performed using PC-compatible microcomputer programs [SPSS for Windows; and Sigmastat (Jandel Scientific)]. Two group cornparisons were made using Student's t-test. Multiple group comparison were made using ANOVA, followed by Duncan's multiple range test. Differences between means were considered statistically significant at the p < 0.05 level (two-tailed).

5.4 RESULTS Autoradiographic results show that there are no differences in ER concentration between GLP- I R +I+ and GLP- I R -/- mice in males (2 way ANOVA, F= 1 1 -40; d.f. 1.32; p=0.8479) and in females (2 way ANOVA F=0.0695; d.f. 1.32; p=0.7938). However, there are differences in ER concentrations between male and female GLP-IR +I+ mice (2 way ANOVA F=87.4; d.f. 1.32; p=0.0001) and GLP-1R -/- mice (2 way ANOVA F=115.2; d.f. 1.32; p=0.0001). For example, in the sexually dimorphic nucleus of the POA there is a greater ER concentrations in female than in males. No statistically significant difference was observed between the data for GLP-1R +/+ and GLP-1R -1- mice. Autoradiograms depicting differences in ER concentrations between GLP- I R +/+ and GLP-1 R -1- hypothalamic POA and VMN are schematically represented in figures 34 and 35, respectively, and quantified in table 8. I ESTROGEN RECEPTOR IN GLP-1R +/+ AND GLP-1R 4- MICE I

Male GLP- 1R +I+ (n=5) Male GLP- 1R -/- (n=5) Female GLP- 1R +/+ (n=5) Female GLP- 1R -1- (n=5)

Male GLP- 1R +I+ (n=5) Male GLP- 1R -/- (n=5) Female GLP- I R +I+ (n=5) Female GLP- 1R -1- (n=5)

Male GLP- 1R +I+ (n=5) Male GLP- 1R -1- (n=5) Female GLP-1R +/+ (n=5) Female GLP- 1R -1- (n=5)

I I Bound MIE2 !fmoVm~protein) Male GLP- 1R +/+ (n=5) Male GLP- 1R -/- (n=5) Female GLP- 1 R +/+ (n=5) Female GLP- 1R -/- (n=5)

TABLE 8. Quantification of estrogen receptor binding in sexual dimorphic regions of the mouse brain reveals no regional labeling differences between male and female GLP-I R +I+ and GLP- 1R -/- mice. Data were assessed by quantitative in vitro autoradiography and are presented in ferntomole of radiolabeled MI& retained per mg of tissue equivalent. COMPARISON OF ESTROGEN RECEPTORS IN THE GLP-1 R +I+AND GLP-1 R 4-PREOPTIC AREA

GLP- 1R +I+ GLP- 1R 4-

FIGURE 34. Autoradiograms depicting no significant differences in estrogen receptor levels between female GLP-1R +/+ and GLP-IR -/- (top row) and male GLP- 1 R +/+ and GLP-1 R 4- (bottom row) mouse brain. Autoradiograms show estrogen-dependentsexual dimorphism in the preoptic area of the hypothalamus (POA) between male and female animals. COMPARISON OF ESTROGEN RECEPTORS IN THE GLP-1 R +I+ AND GLP-1 R -I- VENTROMEDIAL NUCLEUS

GLP- 1R +/+ GLP- 1R -/-

FIGURE 35. Autoradiograms depicting no significant differences in estrogen receptor levels between female GLP- I R +/+ and GLP- 1R -/- (top row) and male GLP- I R +/+ and GLP- 1 R -/- (bottom row) mouse brain. Coronal brain sections were taken at the level of the ventromedial nucleus of the hypothalamus 0. ESTROGEN RECEPTORS IN GLP-1R +I+ AND GLP-1R -1- MICE

Male GLP-1R +!+ (N=5) - -i i Male GLP-1R -I- (N=5)

Female GLP4 R +I+ (N=5) a Female GLP-IR 4- (N=5)

POA VMN

FIGURE 36. Comparison of ER levels measured in discrete brain regions reveals a significant difference between male and female GLP- 1 R +/+ and GLP- 1 R 4- mice (2 way ANOVA P4.05). Data are expressed in ferntomoles of labeled ME2 retained per mg tissue equivalent. ARC, arcuate nucleus of the hypothalamus; cmAMY, corticomedial amygdala; POA, preoptic nucleus of the hypothalamus; VMN, ventromedial nucleus. 5.5 DISCUSSION The process of sexual differentiation is a complex cascade of events. It starts with the determination of chromosomal (or genotypic) sex at conception. Whether the individual has an XX (female) or XY (male) genotype will determine the gonadal sex. The SRY gene located in the Y chromosome determines the differentiation of the undifferentiated gonad into a testis. If the SRY gene is missing, as is the case with a genetic female, the gonad differentiates into an ovary. With regard to the testes, endocrine activity begins soon after its morphological differentiation during prenatal life. However, if the developing individual is female, the onset of gonadal endocrine activity is delayed until post-partum, and hence the early female-type genital differentiation occurs autonomously without induction of gonadal hormoness3. The fmding that there is a difference in ER concentration between male and female GLP-1R +/+ and GLP- 1 R 4-mice, suggests that loss of GLP- 1 sensitivity does not interfere with normal sexual differentiation. This, in turn, implies that early testicular development is probably quite normal, since any overt reduction in early testosterone secretion might secondarily be reflected in an impairment of sexual differentiation, both centrally and peripherally. Therefore, the altered behavioral responses described by others in GLP-1 R -1- mice are probably not due to abnormal sexual differentiation of the hypothalamus. Moreover, we can conclude that while GLP-1 may be involved in modulation of the GnRH system it is probably not critical for the early development and function of the GnRH neurons in the hypothalamus: if it was, it is likely that some abnormality of sexual differentiation would ensue, since testicuIar androgen production is dependent on pituitary LH secretion even during embryonic development. OVERALL DISCUSSION

In the periphery, glucagon-like peptide-1 (GLP-1) plays an important role in the stimulation of insulin47.48.93 , suppression of glucagon28,89,t25 and delay of gastric e~n~t~in~~~~'~'~~~~.in addition to these peripheral actions, GLP-1 has several other important activities in the CNS. Among these is its ability to induce satiety and reduce food intake101,132

In the present study, we have explored the possible roles of GLP-1 in the mouse brain. In an attempt to determine brain regions that are targeted sites for GLP-1 action, we have for the fust time mapped the GLP-I receptors in the mouse brain. In doing so, we have found that GLP-1 receptors are primarily distributed in discrete regions of the diencephalon and in various structures of the limbic system. GLP-1 having an extensive neuroendocrine and autonomic connections suggests that it may be implicated in the integrative control of the ANS and several endocrine glands in response to changes in the internal and externd environment. In keeping with the observation of GLP-1 receptors being present in distinctive neuroendocrine brain regions, the impetus of the present study was to explore the roles of GLP-1 in the neuroendocrine control of the HPG and HPA axis. By characterizing the reproductive parameters of the GLP-1 R -/- mouse model, we were able to identify several reproductive impairments when GLP-1 signals are lost. Included as reproductive impairments are delayed onset of puberty, reduced litter size, slightly reduced ovarian size/weight, fewer number of follicles per ovary, reduced levels of estradiol in proestrous females, and reduced levels of testosterone in males. However, not all reproductive functions are impaired. For example, we have found that the estrous cycle is normal, in addition to a normal histology of interstitial ovarian cells (thecal and granulosa cells) and testicdar cells (Ieydig and sertoli cells). Despite ample evidence showing impaired reproductive functions, the data are nevertheless somewhat conflicting. For example, while the RIA data suggest that there is reduced drive to gonadal steroidogenesis, in both sexes, a finding that is consistent with the reduced organ weights for the androgen-dependent male accessory organs (prostate gland, epididymis and seminal vesicles), in females, we observed no reduction in uterine weight and only very mild effects on ovarian fimction. As result, it is difficult to be certain of the possible extent of GLP-1 involvement in the HPG axis. We suggest that GLP-1 does not exert its action directly at the peripheral gonadal sites since ['25~]-~~~-1autoradiograms of testicular and ovarian tissue sections did not display any GLP-1 binding sites. It is possible that the apparent impairment of ovarian estradiol production we observed may have reflected the time of day at which the animals were killed. Uterine weight would have increased primarily in response to circulating estradiol levels the night before, so if there were subtle differences between GLP- 1R +/+ and GLP- 1R -/- mice that were time-dependent, these would not necessarily have been reflected in gross uterine weight. We have also shown that loss of GLP-1 activity alters the regulation of the CRF- producing hypothalamic PVN neurons, but this alteration is also subtle and difficult to interpret. An action of GLP-1 at the CRF-containing parvocellular PVN neurons is perhaps to be expected, given the high concentration of GLP-1 receptors in this nucleus. The lack of any overt corticosterone response to moving the animals from their home environment (the animal quarters) to the laboratory, situated 5 floors up in the same building argues in favor of the idea that the GLP-1R -/- mice are stress hyporesponsive. However, severe stress can clearly overcome this deficit, as shown by the exaggerated response of the GLP-1R -/- animals to a short period of Metofanem anesthesia. Further work is required to elucidate the basis of these responses, but a reasonable hypothesis is that GLP-1 may provide a weak positive input to the CRF neurons which is important for regulating normal CRF and hence ACTH activity, but less important in terms of the massive activation of CRF release that is observed under conditions of severe stress. It is important to recognize the potential for interaction between different systems with the hypothalamus176.Effects on the HPA and HPG axes cannot therefore be considered in isolation, because it is entirely possible that the consequences of the GLP- 1 R -/- condition for one of these systems could secondarily impact on the other82,114 . In particular, impairment in the neuroendocrine control of the HPA system can cause imbalances in other systems. Of important physiological relevance to our study, stress-related hormones can influence reproductive functions at all three levels of the HPG ad3: the brain, to inhibit GnRH secretion; the pituitary, to interfere with GnRH-induced LH release, and the gonads, to alter the stirnulatory effect of gonadotrophins on sex steroid secretion. In keeping with previous reports that CRF axons terminate on the GnRH-secreting neurons114,and that CRF lowers serum testosterone le~els"~,it is possible that reproductive impairment in GLP-1 R -/- mice might secondarily result from changes in the HPA system. It is difficult to see a direct connection, because a reduction in bad CRF release would actually be expected to enhance reproductive function, if anything, since increased CRF release decreases the frequency of GnRH pulses. However, it is possible that there may be some contribution from the HPA to the HPG axis. For example, the effects of the loss of GLP-1 input to the GnRH neurons may be partially attenuated by the simultaneous reduction in CRF input. Another level of interaction that is important to consider is the possibility that changes in HPA activity may have secondary impact on appetite. Stress-related hormones are also known to influence food intake and feeding behavior. For example, in the study by Hoaa et aL(199 it was concluded that chronic ICV infusion of CRF induced continuous inhibition of food intake. Circulating glucocorticoids also have important effects on appetite, both directly through their receptors in the brain and indirectly through their actions on glucose homeostasis loo. This is important when examining feeding behavior of GLP- 1R -/- mice. A GLP-1-mediated inhibition of food intake was evident in GLP- 1R +/+ mice, but completely abolished in GLP-1R -/- mice. However, in comparing the TB W of sex-matched and age-matched GLP- 1R -/- mice to GLP-1 R +/+ mice, no differences in body weights were ~bserved"~.The normal feeding behavior described in GLP-1R -/- mice raises the question as to whether this is a consequence of an impaired CRF action to suppress feeding and/or the result of losing the suppressive feeding effects of central GLP-I. The indistinguishable body weights and growth curves of GLP-1R -/- and GLP-1R +/+ mice suggest that GLP-l/GLP-1R system is likely not an essential component of mechanisms controlling appetite. Our data for CCK-8 binding also suggest that loss of GLP- 1 signaling does not cause a generalized compensation in other system involved in satiety. More specifically, we were interested in seeing whether the loss of GLP-1 signaling enhances CCK-8 sensitivity in GLP- 1R -/- mice. Results depict no significant differences in CCK-8R concentration between GLP-1R -/- and GLP-1R +I+ mice, indicating that the CCK-8 system is not overcompensated when the GLP-IIGLP-1R system is disrupted. Furthermore, no sex differences were detected in comparing male to female, suggesting that lack of sexual differentiation in the brain might be responsible for impairments in reproductive or stress- related functions during prenatal development andor during adult life. Sexual differentiation profoundly affects feeding behavior. Hence, if reproductive functions are impaired during prenatal development, this can lead to disruption in the process of sexual differentiation which, in turn, have indirect effects in altering appetite responses. Within this framework, we have undertaken autoradiographic procedures for estrogen receptor binding in brain sections taken fiom both GLP-1R -1- ad GLP-1 R +/+ mice to see whether the lack for sexual differences is responsible for unique feeding behavior seen in GLP-1 R -1- mice. Autoradiograms results indicate that GLP- 1R -1- mice, like GLP- 1R +I+ mice, have successfully sexually differentiated. In the tuberal region of the POA, a gross difference in ER concentration was evident between males and femaIes, but indistinguishable between GLP-1R -1- and GLP-1R +I+ mice. From these resuIts, it can be deduced that disruption of the GLP-1 receptor does not interfere with normal processes of sexual differentiation. We have therefore eliminated the possibility of altered sexual differentiation giving rise to the unique feeding behavior described seen in GLP- 1R -1- mice. In summary, we have found that there are subtle impairments in reproductive functions and fimctions related to stresshovel environment responses in GLP-1R -/- mice. Although the pathway(s) by which GLP-1 exert its effect is not yet hlly understood, it may involve a complex interactions of multiple factors known to have roles in each of these systems. In support of this hypothesis, the autoradiographic data on the distribution of GLP- 1 receptors in mice and rats demonstrate that many different target regions of the brain are likely sensitive to the effects of GLP-1, including many structures within the diencephalon and limbic system that are know to be involved in the regulation of stress responses, food intake and the regulation of neuroendocrine hction. The extent to which GLP-1 regulates the functions of each of these neuronal centers under normal physiological circumstances remains a subject for future research. 1. Adelhorst, K., Hedegaard, B.B., Knudsen, L.B., and Ole, K. Structure-Activity studies of glucagon-like peptide- 1. The Journal of Biological Chemistry 269:6275-6278, 1993.

2. Adrian, T.E., Bilchick, A.J., Zucker, K.A., and Modlin, I.M. CCK receptor blockade increases hamster body weight and food intake. FASEB Journal 2:A73 7-A743, 1988.

3. Alvarez, E., Roncero, I., Chowen, LA., Thorens, B., and Blazquez, E. Expression of the glucagon-like peptide-l receptor gene in rat brain. Journal of Neurochemishy 66:920-927, 1996.

4. Anika, S.M., Houpt, T.R., and Houpt, K.A. Cholecystokinin and satiety in pigs. American Journal of Physiology 240:R3 10423 18, 198 1.

5. Anika, S.M., Houpt, T.R.,and Houpt, K.A. Satiety elicited by cholecystokinin in intact and vagotomized rats. Physiology of Behaviour 19:76 1-766, 1977.

6. Antin, J., Gibbs, J., Holt, J., Young, R.C., and Smith, G.P. Cholecystokinin elicits the complete behavioral sequence of satiety in rats. J Comp.Physio. Psychd 89:784-790, 1975.

7. Baldissera, F.G. and Holst, I.J. Glucagon-related peptides in the human gastrointestinal mucosa. Diabetologia 26:223-228, 1984.

8. Banks, W.A. Evidence for a cholecystokinin gut-brain axis with modulation by bombesin. Peptides 1547-351, 2980.

9. Barragan, J.M., Rodriguez, R.E., and Blazquez, E. Changes in arterial blood pressure and heart rate induced by glucagon-like peptide-1-(7-36) arnide in rats. American Journal of Physiology 266:E459-E466, 1994.

10. Beak, S.A., Heath, M.M., Small, C.J., Morgan, D.G.A., Ghatei, M.A., Taylor, A.D., Buckingham, J.C., Bloom, S.R., and Smith, D.M. Glucagon-like peptide-1 (GLP-1) stimulates lutehizing hormone releasing hormone secretion in a rodent hypothalamic neuronal cell line (GTl-7). Journal of Clinical Investigation 101:1334-1341, 1998.

11 . Bell, G.I., Santerre, R.F., and Mullenbach, G.T. Hamster preproglucagon contains the sequence of glucagon and two related peptides. Nature 302:7 16-7 18, 1983.

12. Bellinger, L.L. and Bemadis, L.L. Suppression of feeding by cholecystokinin but not bombesin is attenuated in dorsomedial hypothalamic nuclei lesioned rats. Peprides 5:547- 552,1984. 13. Billington, C., Levine, A.S., and Morley, J.E. Are peptides truly satiety agents? A method of testing for neuro humoral satiety effects. American Journal of Physiology 245:R920-R926, 1983.

14. Blundel, J.E., Rogers, P.J., and Hill, A.J. Behavioural structure and mechanisms of anorexia: calibration of natural and abnormal inhibition of eating. Brain Research Bulletin 15:371-376, 1985.

15. Boden, P. and Hill, R.G. Effects of cholecystokinin and related peptides on neuronal activity in the ventromedial nucleus of the rat hypothalamus. British Journal of Pharmacology 94:246-252, 1988.

16. Branchereau, P., Bohme, G.A., Champagnat, I., Morin-Sm, M.P., Durieux, C., Blanchard, J.C., Roques, B .P., and Devanit-Saubie, M. Cholecy~tokinin~and chole~ystokinin~receptors in neurons of the brainstem solitary complex of the rat: pharmacological identification. Journal of Pharmacology & Experimental Therapy 260: 1433-1440, 1992.

17. Brown, T.J., Scherz, B., Hochberg, R.B., and MacLusky, N.J. Regulation of estrogen receptor concentrations in the rat brain: effects of sustained androgen and estrogen exposure. Neuroendocrinology 63 :53-60, 1996.

18. Brown, T.J., Naftolin, F., and MacLwky, N.J. Sex differences in estrogen receptor binding in the rat hypothalamus: Effects of subsaturating pulses of estradiol. Brain Research 578: 129- 134, 1992.

19. Brubaker, P.L., Efendic, S., and Greenberg, G.R. Truncated and hll-length glucagon-like peptide- 1 (GLP- I) differentially stimulate intestinal somatostatin release. Endocrine 6:91 -95, 1997.

20. Bruley des Varannes, S., Mizrahi, M., Curran, P., Solomon, T., Turkelson, C., and Dubois, A. Role of cholecystokinin in the regulation of gastric emptying and motility. Gastroenterology 98:A33 1-A339, 1990.

21. Buggy, J.J., Livingston, J.N., Rabin, D.U., and Yoo-Warren, H. Glucagon-glucagon-like peptide 1 receptor chimeras reveals domains that determine specificity of glucagon binding. The Journal of Biological Chemistry 270:7474-7478, 1995.

22. Bullock, B.P., Heller, R.S., and Habener, J.F. Tissue distribution of messenger ribonucleic acid encoding the rat glucagon-like peptide- l receptor. Endocrinology 1372968- 2978, 1996.

23. Carnpos, R.V., Lee, Y.C., and Drucker, D.J. Divergent tissue-specific and developmental expression of receptors for glucagon and glucagon-like peptide-1 in the mouse. Endocrinology 134:2156-2 164, 1994. 24. Carlberg, M., Gundlach, A.L., Mercer, L.D., and Beart, P.M. Autoradiographic localization of cholecystokinin A and B receptors in rat brain using [125I]D-Tyr25-CCK 25- 33s. European Journal ofNezrroscience 4563-573, 1992.

25. Chatterton, R.T. The role of stress in female reproduction: Animal and human considerations. International Journal of Fertility 35% 13, 1990.

26. Cheng, C.A., Geoghegan, J.G., Lawson, D.C., Berlangieri, S.U., Akwari, O., and Pappas, T.N. Central and peripheral effects of CCK receptor antagonists, L-356,260 and MK-329, on satiety in dogs. American Journal of Physiology 247:E4S 3 -E45 8, 1984.

27. Christophe, J. Glucagon receptors: from genetic structure and expression to effector coupling and bioIogicd responses. Biachimica et Biophysics Acta 1241 :45-57, 1995.

28. dlAlessio, D.A., Fuj imoto, W.Y., and Ensinck, J. W. Effects of glucagon-like peptide- 1- (7-36) on release of insulin, glucagon, and somatostatin by rat pancreatic islet cell monolayer cultures. Diabetes 38:1534, 1989.

29. Deacon, C.F., Nauck, M.A., Toft-Nielsen, M., Pridal, L., Willms, B., and Holst, J.J. Both subcutaneously and intravenously administered glucagon-like peptide 1 are rapidly degraded from the NH2-terminus in type I1 diabetic patients and in healthy subjects. Diabetes 44: 1 126- 1131, 1995.

30. Debas, H.T., Farooq, O., and Grossman, M.I. Inhibition of gastric emptying is a physiological action of cholecystokinin. Gastroenterology 68: 121 1 - 12 17, 1975.

3 1. Delgado, E., Luque, M.A., Alcantara.A., Trapote, M.A., Clemente, F., Galera, C., Valverde, I., and Villanueva-Penacarillo, M.L. Glucagon-like peptide-l binding to rat skeletal muscle. Peptides 16:225-229, 1995.

32. Della-Fera, M.A. and Baille, C.A. Cholecystokinin octapeptide: continuous picomole injections into the cerebral ventricles of sheep suppress feeding. Science 9:729-733, 1988.

33. Della-Fera, M.A., Coleman, B.D., Doubek, C.A., Marshall, H.N., Miner, J., Paterson, J., Gingerich, R.L., and BailIe, C.A. Cholecystokinin concentration in specific brain areas of rats fed during the light or dark phase of the circadian cycle. Physiolu~and Behavior 45:801- 807,1989.

34. Della-Fera, M.A., BaiIle, C.A., Schneider, B.S., and Grinker, J.A. Cholecystokinin antibody injected in cerebra1 ventricles stimulates feeding in sheep. Science 2 12:687-689, 1981.

35. Dietl, M.M. and Palacios, J.M. The distribution of cholecystokinin receptors in the vertebrate brain: species differences studied by receptor autoradiography. Journal of Chemical Neuroanatomy 2:149- 162, 1989. 36. Dietl, M.M., Probst, A., and Palacios, J.M. On the distribution of cholecystokinin binding sites in the human brain: an autoradiographic study. Synapse 1: 169- 183, 1987.

37. Dillon, J.S., Tanizawa, Y., Wheeler, M.B., Leng, X.H., Ligon, B.B., Rabin, D.U., Warren, H.Y., Permutt, M.A., and Boyd, A.E. Cloning and functional expression of the human glucagon-like peptide- l (GLP-1) receptor. Endocrinology 133: 1907-2 9 12, 1993.

3 8. Dockray, G.J. Irnmunochernical evidence of cholecystokinin-like peptides in brain. Nature 364568-570, 1976.

39. Dourish, C.T., Rycroft, W., and Iversen, S.D. Postponement of satiety by blockade of brain cholecystokinin (CCK-B) receptors. Science 245: t 509- 15 11, 1989.

40. Dourish, C.T. and Hill, D.R. Classification and function of cholecystokinin receptors. Trends in Pharmacological Sciences 8:207-208, 1987.

41. Drucker, D.J., Phillippe, J., Mojsov, S., Chick, W.L., and Habener, J.F. Glucagon-like peptide I stimulates insulin gene expression and increases cyclic AMP levels in a rat islet cell line. Proceedings of the National Academy of Science USA 84:3434-343 8, 1987.

42. Dupre, J., Behme, M.T., Hramiak, I.M., McFarlane, P., Wiliiamson, M.P., Zabel, P., and McDonald, T.J. Glucagon-like peptide 1 reduces postprandial glycemic excursions in IDDM. Diabetes 44526-630, 1995.

43. Durieux, C., Pelaprat, D., Charpentier, B., Morgat, J.L., and Roques, B.P. Characterization of pH]-CCK binding sites in mouse and rat brain. Neuropeptides 12:141 - 148,1988.

44. Egan, J.M., Montrose-Rafizadeh, C., Wang, Y., Bernier, M., and Roth, J. Glucagon-like peptide- 1(7-3 6)amide (GLP- 1) enhances insulin-stimulated glucose metabolism in 3T3 -L 1 adipocytes: one of several potential extrapancreatic sites of GLP-1 action. Endocrinology l35:2070-2075, 1994.

45. Eissele, R., Bothe-Sandfort, E., Goke, B., Eng, J., Arnold, R., and Koop, H. Rat gastric somatostatin and gastrin release: interactions of exendin-4 and truncated glucagon-like peptide-1 (GLP-I) arnide. Life Sciences 55:629-634, 1994.

46. Eissele, R., Goke, R., Wiilerner, S., Harthus, H.P., Verrneer, H., Arnold, R., and Goke, B. Glucagon-like peptide-l cells in the gastrointestinal tract and pancreas of rat, pig and man. European Journal of Clinical Investigation 22:283-291, 1992.

47. Fehrnan, H.C. and Habener, I.F. Insulinotropic glucagon-like peptide- 1(7-37)(&- 36)amide: A new incretin hormone. Trends in Endocrinology and Metabolism 3 :158, 1992. 48. Fehmann, H.C., Goke, R., and Goke, B. Glucagon-like peptide-l(7-37)/(7-36)amide is a new incretin. Molecular and Cellular Endocrinology 85:C39, 1992.

49. Fehmann, H.C. and Habener, J.F. Insulinotropic hormone glucagon-like peptide-1(7-37) stimulation of proinsulin gene expression and proinsulin biosynthesis in iosulinoma bTC-1 cells. Endocrinology 130:159- 166, 1992.

50. Fehmann, H.C. and Habener, J.F. Functional receptors for the insulinotropic hormone glucagon-like peptide-1 (7-37) on somatostatin secreting cell line. FEBS Letters 279:279-335, 1991.

51. Ferguson, A.V. The area postrema: a cardiovascular control centre at the blood-brain interface? Canadian Journal of Physiology and Phurrnacology 69 :1 026- 103 4, 199 I.

52. Fieseler, P., Bridenbaugh, S., Nustede, R., Martell, J., Orskov, C., and Nauck, M.A. Physiological augmentation of amino acid-induced insulin secretion by GIP and GLP-I but not by CCK. American Journal of Physiology 265:E949-E955, 1995.

53. Furuse, M., Matsumoto, M., Mori, R., Sagahara, K., Kano, K., and Hasegawa, S. Influence of fasting and Y on the suppressive food intake induced by intracerebroventricular injection of glucagon-like peptide-1 in the neonatal chick. Brain Research 764:289-292, 1997.

54. Gallwitz, B., Witt, M., Paetzold, G., Morys-Wortmann, C., Zimmerman, B., Eckart, K., Folsch, U.R., and Schmidt, W.E. Structure/Activity characterization of glucagon-like peptide-1 . European Journal of Biochemistry 225: 115 1- 1 156, 1994.

55. Ghiglione, M., Uttenthal, S.K., George, S.K., and Bloom, S.R. How glucagon-Like is glucagon-like peptide- 1? Diabetologia 27599-600, 1984.

56. Gibbs, J., Fdasco, J.D., and McHugh, P.R. Cholecystokinin decreased food intake in rhesus monkeys. American Journal of Physiology 230:15- 18, 1976.

57. Gibbs, J., Young, R.C., and Smith, G.P. Cholecystokinin decreases food intake in rats. Journal of Comparative Physiology & Psycho logy 84:48 8-495, 1 973.

58. Gilles, C., Lotstra, F., and Vanderhaeghan, J.J. CCK nerve terminals in the rat striatal and limbic areas originate partly in the brain stem and partly in telencephalic structures. Life Sciences 32: 1683-1690, 1983.

59. Goke, B., Goke, R., Holst, J.J., and Orskov, C. Glucagon-like peptide-1: on its way to clinical application. 54:339-3 97, 1993. 60. Goke, R., Larsen, P.J., Mikkelsen, J.D., and Sheikh, S.P. Distribution of GLP-1 binding site in the rat brain: Evidence that exendin-4 is a ligand of brain GLP-1 binding sites. European Journal of Neuroscience 7:2294-2300, 1995.

61. Goke, R., Larsen, P.J., Mikkelsen, J.D., and Sheikh, S.P. Identification of specific binding sites for glucagon-like peptide- 1on the posterior lobe of the rat pituitary. Neuroendocrinology 62: 130-134, 1995.

62. Goke, R., Cole, T., and Conlon, J.M. Characterization of the receptor for glucagon-like peptide-1 (7-36) arnide on plasma membranes from rat insulinoma-derived cells by covalent cross-linking. Journal of Molecular Endocrinology 2:93-102, 1989.

63. Gorski, R.A., Gordon, J.H., Shryne, J.E., and Southam, A.M. Evidence for a morphological sex difference within the medial preoptic area of the rat brain. Brain Research 1481333-346,1978.

64. Gros, L., Thorens, B., Bataille, D., and Kervan, A. Glucagon-like peptide-1 -(7-36)Amide, Oxyntomodulin, and Glucagon interact with a common receptor in a somatostatin-secreting ceIl line. Endocrinology 133:63 1-638, 1993.

65. Gunn, I., O'Shea, D., and Bloom, S.R. Control of appetite - the role of glucagon-like peptide- l(7-36) arnide. (Review). Journal of Endocrinology 155: 197-200, 1997.

66. Gutniak, M., Orskov, C., Holst, J.J., Ahren, B., and Efendic, S. Antidiabetogenic effect of Glucagon-like peptide- l(7-3 6)amide in normal subjects and patients with diabetes rnellitus. The New EngZandJournul of Medicine 326:13 16-1322, 1992.

67. Hamrnonds, P., Schofield, P.N., and Ashcrofi, S.J.H. Glucose regulates preproinsulin messenger RNA levels in a clonal cell line of simian virus 40-transformed b cells. FEBS Letter 2 13 :149- 154, 1987.

68. Hansen, A.B., Gespach, C.P., and Rosslin, G.E. Effect of truncated glucagon-like peptide- 1 on CAMP in rat gastric glands and HGT- 1 human gastric cancer cells. FEBS Letters 236: 119-123, 1988.

69. Hanson, H. and Strouse, J. Effects of the CCK antagonist, L-364,718, on food intake and of the blockade of feeding produced by exogenous CCK in rat. FedProc. 46: 1480, 1987.

70. Hays, S.E., Beinfeld, M.C., Jensen, R.T., Goodwin, F.K., and Paul, S.M. Demonstration of a putative receptor site for cholecystokinin in the rat brain. Neuropeptides 153-62, 1980.

71. Heinrich, G., Gros, P., and Habener, J.F. Glucagon gene sequence. Four of six exons encode separate functional domain of preproglucagon. Journal of Biological Chemisiry 259: 14082-14087, 1984. 72. Hemann, C., Voge, A., and Goke, B. Regulation of glucagon-like peptide-1 (GLP-1) release fiom the isolated perfused rat by nutrients, peptides and neuromediators. Digestion 54:267-272, 1993.

73. Hill, D.R., Shaw, T.M., Graham, W., and Woodruff, G.N. Autoradiographic detection of cholecystokinin-A receptors in primate brain using 1251-Bolton Hunter CCK-8 and 3H-M.- 329. Journal ofNeuroscience 10:1070-1081, 1990.

74. Hill, D.R., Shaw, T.M., Dourish, C.T., and WoodrufXG-N. CCK-A receptors in the rat interpeduncular nucleus: evidence for a presynaptic location. Brain Research 454: 101- 105, 1988.

75. Hill, D.R., Campbell, N.J., Shaw, T.M., and Woodruff, G.M. Autoradiographic localization and biochemical characterization of peripheral type CCK receptors in rat CNS using highly selective nonpeptide CCK antagonists. Journal of Neuroscience 7:2967-2976, 1987.

76. Hjorth, S.A., Adelhorst, K., Pedersen, B.B., Kirk, O., and Schwarz, T. W. Glucagon and glucagon-like peptide 1: Selective receptor recognition via distinct peptide epitopes. The Journal Biological Chemistry 269:30 12 1-30 124, 1994.

77. Hoebel, B.G. Neuroscience and appetitive behavior research: 25 years. (Review). Appetite 29(2): 1 19- 133, 1987.

78. Holst, J.J.? Orskov, C., Vagn Nielsen, O., and Schwartz, T.W. Truncated glucagon-like peptide- 1, an insulin-releasing hormone from the distal gut. FEBS Letter 2 1 1: 169- 174, 1987.

79. Holst, J.J. Gut glucagon, enteroglucagon, gut GLI, glicentin - current status. Gastroenterology 84: 1602- 16 13, 1983.

80. Holst, J.J. Radioreceptor assays for glucagon. Handbook of Experimental Pharmacology 60:245-26 1, 1983.

8 1. Horsch, D.J., Goke, R., Eissele, R., Michel.B., and Goke, B. Reciprocal cellular distribution of glucagon-like peptide- 1 (GLP- I) irnmunoreactivity and GLP- 1 receptor mRNA in pancreatic islets of rat. Pancreas l4:290-294, 1997.

82. Hotta, M., Shibasaki, T., Yarnauchi, N., Ohno, H., Benoit, R., Ling, N., and Demura, H. The effects of chronic central administration of corticotropin-releasing factor on food intake, body weight, and hypothalamic-pituitary-adrenocortical hormones. Life Sciences 48:1483- 1491,1991.

83. Huhtaniemi, I. Molecular aspects of the ontogeny of the pituitary-gonadal axis. Reproduction, Fertility and Development 7: 1025- 1035, 1995. 84. Hvidberg, A., Nielsen, M.T., Hilsted, J., Orskov, C., and Holst, J.J. Effect of glucagon- like peptide-1 (proglucagon 78-107 amide) on hepatic glucose production in healthy man. Metabolism: Clinical & Experimental 43 :104-1 08, 1994.

85. Innis, R.B. and Snyder, S.H. Distinct cholecystokinin receptors in brain and pancreas. Proceedings of the National Academy of Sciences, USA 77:69 1 7-692 1, 1980.

86. Jacobson, C.D., Shryne, J.E., Shapiro, F., and Gorski, R.A. Ontogeny of the sexually dimorphic nucleus of the preoptic area. Journal of Comparative Neurology 193:54 1-548, 1980.

87. Jelinek, L., Lok, S., Rosenberg, G.B., Smith, R.A., Grant, F.J., Biggs, S., Bensch, P.A., Kuijper, I.L., Sheppard, P.A., Sprecher, C.A., O'Hara, P.J., Foster, D., Walker, K.M., Chen, L.H. J., McKernan, P. A., and Kindsvogel, W. Expression Cloning and Signaling Properties of the Rat . Science 259: 16 14- 1622, 1993.

88. Kame, S.M., Kreyrnann, B., Ghatei, M.A., and Bloom, S.R. Identification and characterization of glucagon-like peptide-l(7-36)amide binding sites in the rat brain and lung. FEBS Letter 241 :209-212, 1988.

89. Kawai, K., Suzuki, S., Ohashi, S., Mukai, H., Ohrnori, H., Murayama, A*,and Yamzshita, K. Comparison of the effects of glucagon-like peptide-1 -(I-37) and -(7-37) and glucagon on islet hormone release from isolated perfused canine and rat . Endocrinology 124: 1768-1773, 1989.

90. Kieffer, T.J., McIntosh, C.H.S., and Pederson, R.A. Degradation of glucose-dependent insulinotropic polypeptide and truncated glucagon-like peptide 1 in vitro and in vivo by dipeptidyl peptidase IV. Endocrinology 136(8):3585-3597, 1995.

91. Kissileff, H.R., Pi-Sunyer, F.X., Thornton, J., and Smith, G.P. C-terminal octapeptide of cholecystokinin decreases food intake in man. American Journal of Clinical Nutrition 34:154-160, 1981..

92. Kolligs, F., Fehrnann, H.C., Goke, R., and Goke, B. Reduction of the Incretin effect in rats by the glucagon-like peptide 1 receptor antagonist exendin (9-39) amide. Diabetes 44:465-469, 1995.

93. Kreyrnann, B., Williams, G., Chatei, M.A., and Bloom, S.R. Glucagon-like peptide- 1(7- 36): a physiological incretin in man. Lancet 2: 1300- 1304, 198%

94. Laferrere, B., Nguyen, M., Bonhomme, G., Legall, A., Basdevant, A., and Guy-Grand, B. Effect of BIM- 18216, a novel antagonist, on food intake reduction induced by cholecystokinin. Behavioral Neuroscience 105:707-71 I, 1991. 95. Lankat-Buttgereit, B., Goke, R., Stockrnann, F., Jiang, J., Fehmann, H.C., and Goke, B. Detection of the Human Glucagon-like peptide 1 (7-36) amide receptor on insulinoma- derived cell membranes. Digestion 55:29-33, 1994.

96. Lankat-Buttgereit, B., Goke, R-, Fehmann, H.C., Richter, G., and Goke, B. Molecular cloning of a cDNA encoding for the GLP-1 receptor expressed in rat lung. Experimental and Clinical Endocrinology 102:34 1-347, 1994.

97. Larriva-Sahd, J., Gorski, R. A., and Micevych, P.E. Cholecystokinin synapses in the sexually dimorphic central and peripheral nervous. Experimental Neurology 92:639-650, 1986.

98. Larsen, P.J., Tang-Christensen, M., and Jessop, D.S. Central administration of glucagon- like peptide-1 activates hypothalamic neuroendocrine neurons in the rat. Endocrinology 138:4445-4455, 1997.

99. Larsen, P.J., Tang-Christensen, M., Hoist, J.J., and Orskov, C. Distribution of glucagon- like peptide-1 and other preproglucagon-derived peptides in the rat hypothalamus and brainstem. Neuroscience 77:257-270, 1997.

100. Leibowitz, S .F., Sladek.C., Spencer, L., and Tempel, D. , epinephrine, neuroepinephrine in the paraventricular nucleus: Stimulation of feeding and the release of corticosterone, , and glucose. Brain Research 2 1:9OW 12, 1988.

101. Leibowitz, S.F., Morgane, P.J., and Panksepp, J. Neurochemical systems of the hypothalamus in control of feeding and drinking behavior and water-electrolyte excretion. Annals of New York Academy of Sciences 3:299-307, 1980.

102. Leibowitz, S.F. Brain neurotransmitters and appetite regulation. PsychopharrnacologicaZ Bulletin 21 :412-418, 1980.

103. Leibowitz, S.F. Paraventricular nucleus: A primary site mediating adrenergic stimulation of feeding and drinking. Pharmacology, Biochemistry & Behavior 8:163-1 75, 1978.

204. Liberge, M., Armebo, M.P., and Bueno, L. Role of hypothalamic cholecystokinin octapeptide in the colonic motor response to a meal in rats. Gastroenterology 100:441-449, 1991.

105. Liberge, M., Armebo, P., and Bueno, L. CCK neurons of the ventromedial (VMH) hypothalamus mediate the upper gut motor changes associated with feeding in rats. Brain Research 502:1036-1041, 1989. 106. Liddle, R.A., Gertz, B.J., and Kanayama, S. Effects of a novel cholecystokinin (CCK) receptor antagonist, MK-329, on gallbladder contraction and gastric emptying in humans. Journal of Clinical Investigation 84: 1220-1225, 1989. 107. Linn, T., Schneider, K., Goke, B., and Federlin, K. Glucagon-like peptide-l(7-36) amide improves glucose sensitivity in beta-cells of NOD mice. Actu Diabetologia 3 3 :1 9-24, 1996.

108. Lorenz, D.,N. and Goldman, S.A. Vagal mediation of the cholecystokinin satiety effect in rats. Physiology of Behavior 29:599-604, 1982.

109. Lotti, V.J. and Chang, R.S.L. A new potent and selective non-peptide gastrin antagonist and cholecystokinin receptor (CCK-B) ligand: L-365,260. European Journal of Pharmacology 162273-280, 1989.

110. Lowry, O.H., Rosebrough, N.J., Farr, A.L., and Randall, J.R. Protein measurement with the Folin phenol reagent. Journal of Biological Chemistry 193:265-275, 195 1.

111. Lu, F., Jin, T., and Drucker, D.J. Proglucagon gene expression is induced by gastrin- releasing peptide in a mouse line. Endocrinology 13737 10-3 71 6, 1996.

1 12. MacDonald, J.L., Parnavelas, J.G., Karamalidis, A.N., Rosenquist, G., and Brecha, N. The morphology and distribution of peptide containing neurones in the adult and developing visual cortex of the rat 111. Cholecystokinin. Journal of Neurocytology 11 :8 8 1-895, 1982.

113. MacLusky, N.J. and Naftolin, F. Sexual differentiation of the Central Nervous System. Science 21:1294-1303, 1981.

114. MacLusky, N.J., Naftolin, F., and Leranth,C. IrnrnunocytochemicaI evidence for direct synaptic connections between corticotrophin-releasing factor (CRF) and gonadotrophin- releasing hormone (GnRH)-containing neurons in the preoptic area of the rat. Brain Research 439:39 1-395, 1988.

115. Maeda, K., Nagatani, S., Estacio.M.A., and Tsukamura, H. Novel estrogen feedback sites associated with stress-induced suppression of luteinizing hormone secretion in female rats. Cellular and Molecular Nettrobiology 16311-324, 1996.

116. Magiakou, M.A., Mastorakos, G., Webster, E., and Chrousos, G.P. The hypothalamic- pituitary-adrenal axis and the female reproductive system. Annals of Ner York Academy of Sciences 8:42-56, 1980.

117. Matsumoto, A. and Arai, Y. Sex differences in volume of the ventromedid nucleus of the hypothalamus in the rat. Endocrinologica Japonica 3 0 :277-280, 1983.

118. McLaughIin, C.L., Baile, C.A., and Buonomo, F.C. Effect of CCK antibody on food intake and weight gain in Zucker rats. Physiology of Behavior 35:215-220, 1985. 119. McLaughlin, C.L. and Baile, C.A. Decreased sensitivity of obese Zucker rats to the putative satiety agent cholecystokinin. Physiology of Behavior 25:543-548, 1980.

120. Micevych, P.E., Park, S.S., Akesson, T.R., and Elde, R. Distribution of cholecystokinin immunoreactive cell bodies in the male and female rat hypothalamus. Journal of Comparative Neurology 269:38 1-3 9 I, 1987.

121. Miholic, J., Orskov, C., Holst, J.J., Kotzer, J., and Meyer, H.J. Emptying of the gastric substitute, glucagon-like peptide-l (GLP-I), and reactive after total gastrectomy. Digestive Diseases and Science 3 6: 136 1-1370, 199 1.

122. Moens, K., Heimberg, H., Flamez, D., Huypens, P., and Quartier, E. Expression and hctiond activity of glucagon, GLP-I and GIP receptors in rat pancreatic islets cells. Diabetes 45:257-26 1, 1996.

123. Mojsov, S., Kopzjnski, M.G., and Habener, J.F. Both amidated and non-arnidated forms of glucagon-like peptide-l are synthesized in the intestine and pancreas. Journal of Biological Chemistry 265:8OO 1-8008, 1990.

124. Moj sov, S., Weir, G.C., and Habener, J.F. Insulinotropin: glucagon-like peptide- 1(7-3 7) co-encoded in the glucagon gene is a potent stimulator of insulin release in the perfused pancreas. Journal of Clinical Investigation 79:6 16-6 19, 1987.

125. Mojsov, S., Heinrich, G., Wilson, I.B., Ravauola, M., and 0rci.L. Preproglucagon gene expression in pancreas and intestine diversifies at the level of post-translational processing. Journal of Biological Chemisny 261 :1 1880-1 1889, 1986.

126. Montrose-Rafizadeh, C., Yang, H., Rodgers, B.D., Beday.A., Pritchette, L.A., and Eng, J. High potencies antagonists of the pancreatic glucagon-like peptide-1 receptor. Journal of BioZogicul Chemistry 272:2 1201-2 1206, 1997.

127. Moran, T.H. and McHugh, P.R. Cholecystokinin suppresses food intake by inhibiting gastric emptying. American Journal of Physiology 242:R49 1-R497, 1982.

128. Morin, M.P., Demarchi, P., Champagnat, I., Vanderhaeghen, J.. Rossier, J., and Denavit-Saubie, M. Inhibitory effects of cholecystokinin on neurons in the nucleus tractus solitarius. Brain Research 265:333-338, 1983.

129. Morley, J.E. Neuropeptide regulation of appetite and weight. Endocrine Reviews 8:256- 287,1987.

130. Nauck, M.A., Kleine, N., Orskov, C., Holst, J.J., Willms, B., and Creutdeldt, W. Normalization of fasting hyperglycaemia by exogenous glucagon-like peptide 1 (7-36 amide) in type 2 (non-insulin-dependent) diabetic patients. Diabetologia 36:74 1-744, 1993. 131. Nauck, M.A., Heimsaat, M.M., Orskov, C., Holst, J.J., Ebert, R., and Creutzfeld, W. Preserved incretin activity of glucagon-like peptide 1 [7-361 but not of synthetic human gastric inhibitory polypeptide in patients with type-2 diabetes mellitus. Journal of Clinical Investigation 91 :301-307, 1993.

132. Navarro, M., Rodriquez de Fonseca, F., Alvarez, E., Chowen, J.A., Zueco, J.A., Gomez, R., Eng, J., and Blazquez, E. Colocalization of glucagon-like peptide-1 (GLP-1) receptors, glucose transporter GLUT-2, and glucokinase mRNAs in rat hypothalamic cells: evidence for a role of GLP-1 receptor agonists as an inhibitory signal for food and water intake. Journal of Neurochemistry 67: 1982-199 1, 1996.

133. Nielsen, D.A., Welsh, M., Cassadaban, M.J., and Steiner, D.F. Control of insulin gene expression in pancreatic b cells and in an insulin-producing cell line, RM-SF cells. Effects of glucose and CAMP on transcription of insulin mRNA. Journal of Biological Chemishy 260: 11585-1 1589, 1985.

134. Oben, J., Morgen, L., Fletcher, J., and Marks, V. Effect of entero-pancreatic hormone, gastric inhibitory polypeptide and glucagon-like polypeptide-1-(7-36)amide on fatty acid synthesis in explants of rat . Journal of Endocrinology l30:267-272, 199 1.

135. Orskov, C.. Wettergren, A., and Holst, J.J. The metabolic rate and the biological effects of GLP- 1(7-3 6)amide and GLP- 1(7-3 7) in healthy volunteers are identical. Diabetes 42: 6 5 8- 661, 1993.

136. Orskov, C., Andreasen, J., and Holst, J.J. All products of proglucagon are elevated in plasma fiom uremic patients. Journal of Clinical Endocrinology and Metabolism 74:379- 384, 1992.

137. Orskov, C. Glucagon-like peptide-1, a new hormone of the enteroinsular axis. Diabetologia 3 5 :70 1-7 1 1, 1992.

138. Orskov, C., Jeppesen, J., Madsbad, S., and Holst, J.J. Proglucagon products in plasma of noninsulin-dependent diabetics and nondiabetic controls in the fasting state and after oral glucose and intravenous arginine. Journal of Clinical Investigation 87:4 1 5-423, 199 1.

139. Orskov, C., Bersani, M., Johnsen, A.H., Hqjrup, P., and Holst, J.J. Complete sequences of glucagon-like peptide-l fiom human and pig small intestine. Journal of Biological Chemistry 264: 12826-12824, 1989.

140. Orskov, C., Holst, J.J., and Nielsen, O.V. Effect of truncated glucagon-like peptide-1 Cproglucagon-(78-107)amidel on endocrine secretion fiom pig pancreas, antrum, and non- antral . Endocrinology 123:2009-20 13, 1988. 141. Orskov, C., Holst, J.J., Knuhtsen, S., Baldissera, F.G.A., Poulsen, S.S., and Nielsen, O.V. Glucagon-like peptides GLP-1 and GLP-2, predicted products of the glucagon gene, are secreted separately fkom the pig small intestine, but not pancreas. Endocrinology 119: 1467- 1475,1986.

142. Ortega, E., Ruiz, E., Rodriguez, E., and Frias, I. Effect of Corticotrophin-Releasing Factor (CRF) in the median eminence on in ovariectomized rats with or without steroid priming: Dose-response study. Neurochemical Research 19: 1225- 1230, 1994.

143. Parrott, R.F. and Baldwin, B.A. Operant feeding and drinking in pigs following intracerebroventricular injection of synthetic cholecystokinin octapeptide. Physiology of Behavior 26:4 19-422, 198 1.

144. Patzelt, C. and Schiltz, E. Conversion of proglucagon in pancreatic alpha cells: the major endproducts are glucagon and a single peptide, the major proglucagon fragment, that contains two glucagon-like sequences. Proceedings of [he National Academy of Science, USA 8 1 :5OO7-5O 12, 1984.

145. Patzelt, C. and Schug, G. The major proglucagon fragment: An abundant islet protein and secretory product. FEBS Letter 129: 127, 198 1.

146. Patzelt, C., Tager, H.S., Carroll, R.J., and Steiner, D.F. Identification and processing of proglucagon in pancreatic islets. Nature 282:260-767, 1979.

147. Paxinos,G. and Watson,C. The rat brain in Stereotaxic Coorrdinates. Academic Press, New York, 1990.

148. Petraglia, A., Sutton, S., Vale, W., and Plotsky, P. Corticotrophin-Releasing Factor decreases plasma Luteinizing Hormone levels in female rats by inhibiting Gonadotrophin- Releasing Hormone release into hypophysial-portal circulation. Endocrinology 120: 1083- 1088,1987.

149. Pfaff, D.W. and Keiner, M. Atlas of estradiol-concentrating cells in the central nervous system of the female rat. Journal of Comparative Neurology 15 1: 1 2 1- 15 8, 1973.

150. Philippe, J. and Missotten, M. Functional characterization of a CAMP-responsive element of the rat insulin 1 gene. Journal of Biological Chemistry 265: 1465- 1469, 1990.

15 1. Plaisancie, P., Bernard, C., Chayvialle, J.A., and Cuber, J.C. Regulation of glucagon- like peptide-1-(7-36) amide secretion by intestinal neurotransmitter and hormones in the isolated vascularly perfused rat colon. Endocrinology l35:2398-240 1, 1 994.

152. Plisetskaya, E.M. Glucagon and related peptides. An overview. Progress in Clinical and Biological Research 342:67-72, 1992. 153. Qualrnann, C., Nauck, M.A., Holst, J.J., and Orskov, C. Insulinotropic actions of intravenous glucagon-like peptide-1 (GLP-1) (7-36)amide in the fasting state in healthy subjects. Acta Diabetologica 32: 13-1 6, 1995.

154. Raufmann, J.P., Singh, L., Singh, G., and Eng, J. Truncated glucagon-like peptide-1 interacts with exendin receptors of dispersed acini fiom guinea pig pancreas. The Journal of Biological Chemistry 267:2 1432-2 1437, 1992.

155. Reidelberger, R.D., Kalogeris, T.J., and Solomon, T.E. Plasma CCK levels after food intake and infusion of CCK analogues that inhibit feeding in dogs. American Journal of Physiology 256:R1148-R1154, 1989.

156. Richter, G., Goke, R., and Goke, B. Characterization of glucagon-like peptide- l(7- 36)amide receptors on rat lung membranes by covalent cross-linking. FEBS Letter 280:247- 251, 1991.

157. Richter, G., Goke, R., and Goke, B. Characterization of receptors for glucagon-like peptide-1(7-36)amide on rat lung membranes. FEBS Letter 267:78-84, 1990.

158. RitzeI, R., Orskov, C., Holst, J.J., and Nauck, M.A. Pharmokinetic, insuhotropic, and glucagonostatic properties of GLP-1[7-36 amide] after subcutaneous injection in healthy volunteers. Dose-response-relationships. Diabetologia 38: 720-725, 1995.

159. Rivier, C. and Rivest, S. Effect of Stress on the activity of the hypothalamic-pituitary- gonadal axis: Peripheral and Central Mechanisms. Biology of Reproduction 45 :523-532, 1991.

160. Rivier, C. and Vale, W. Diminished responsiveness of the hypothalamic-pituitary- adrenal axis of the rat during exposure to prolonged stress: A pituitary-mediated mechanism. Endocrinology 121: 1320-1328, 1987.

161. Rowland, N.E., Morien, A., and Li, B.H. The physiology and brain mechanisms of feeding. Review. Nutrition 12:626-639, 1996.

162. Ruiz-Grande, C., Alarcon, C., Alcantara, A., Castilla, C., Lopez Novoa, J.M., Villanueva-Penacarillo, M., and Valverde, I. Renal catabolism of truncated glucagon-like peptide- 1. Hormone and Metabolism Research 25:6 12-6 16, 1993.

163. Schepp.W., Schmidtler, J., Riedel, T., Dehne, K., Schusdziarra, V., Holst, J.J., Eng, J., Raufman, J.P., and Classen, M. Exendin-4 and exendin(9-39)M2: agonist and antagonist, respectively, at the rat parietal cell receptor for glucagon-like peptide-l (7-36) NH2. European Journal of Pharmacology 269: 183- 191, 1994. 164. Schick, R.R., Hardy, G.J., Yaksh, T.L., and Go, V.L.W. Sites in the brain at which cholecystokiain octapeptide (CCK-8) acts to suppress feeding in rats. A mapping study. Neuropharmacology 29: 109-1 18, 1990.

165. Schjoldager, B.T.G., Mortensen, P.E., Christiansen, J., Orskov, C., and Holst, J.J. GLP- 1 (glucagon-like peptide- 1) and truncated GLP- 1, fragments of human proglucagon, inhibit gastric acid secretion in man. Digestive Disease and Sciences 3 5:703-708, 1980.

166. Schrnidtler, J., Dehne, K., Offermanns, S., Rosenthal, W., Classen, M., and Schepp, W. Stimulation of rat parietal cell fimction by histamine and GLP-1-(7-36) amide is mediated by G,a. American Journal of Physiology 266:G775-G783, 1994.

167. Schmidtler, J., Dehne, K., Allescher, H.D., Schusdziarra, V., Classen, M., Holst, J.J., Polack, A., and Schepp, W. Rat parietal cell receptors for GLP- 1-(7-36)amide: Northern blot, cross-linking, and radioligand binding. American Journal of Physiology 267:G423-G432, 1994.

168. Schmidtler, J., Schepp, W., Janczewska, I., Weigert, N., Furlinger, C., Schusdziarra, V., and Classen, M. GLP- 1-(%36)amide, -(I-37), and -(I-3 6)amide: potent CAMP-dependent stimuli of rat parietal cell function. American Journal of Physiology 260:G940-G950, 1991.

169. Schneider, B.S., Monahan, J.W., and Hirsh, J. Brain cholecystokinin and nutritional status in rats and mice. Journal of Clinical investigation 64: 1348- 1356, 1979.

170. Scrocchi, L.A., Brown, T.J., MacLusky, N.J., Brubaker, P.L., Auerbach, A.B., Joyner, A.L., and Drucker, D.J. Glucose intolerance and normal satiety in mice with a null mutation in the glucagon-like peptide 1 receptor gene. Nature Medicine 2: 1254- 1258, 1996.

171. Shughrue, P.J., Lane, MY., and Merchanthaler, I. Glucagon-like peptide-l receptor (GLP 1-R) rnRNA in the rat hypothalamus. Endocrinology 137:5 159-5 162, 1996.

172. Steinbusch, H.W.M. and Niewenhuys, R. The raphe nuclei of the rat brainstem. A cytoarchitectonic and irnmunohistochemical study. Chemical Neuroanafomy 13 1:20 1-207, 1983.

173. Straus, E. and Yalow, R.S. Cholecystokinin in the brains of obese and nonobese mice. Science 203 :68-69, 1979.

174. Stricker-Krongrad, A., Max, J.P., Musse, N., Nicolas, J.P., Burlet, C., and Beck, B. Increased threshold concentrations of neuropeptide Y for a stirnulatory effect on food intake in obese Zucker rats - changes in the microstructure of the feeding behavior. Brain Research 660(1): 162- 166, 1994.

175. Strohrnayer, A.J. and Smith, G.P. A sex difference in the effect of CCK-8 on food and water intake in obese (ob/ob) and lean (+/+) mice. Peptides 89345-848, 1987. 176. Swaab, D.F., Hofman, M.A., Lucassen, J.S., Purba, J.S., Raadsheer, F.C., and Van d.: Nes, J.A.P. Functional neuroanatorny and neuropathology of the hypothalamus. Anatomy and Embryologv l87:3 17-330, 1993.

177. Takahashi, H., Manaka, H., and Suda, K. Radioimrnunoassay for glucagon-like peptide- 1 in human plasma using N-terminal and C-terminal directed antibodies: a physiologic insulinotropic role of GLP- l(7-36 arnide). Biomedical Research 1 1:99-108, 1990.

178. Tang-Christensen, M., Larsen, P.J., Goke, R., Fink-Jensen, A., Jessop, D.S., Moller, M., and Sheikh, S.P. Central administration of GLP-1-(7-36) amide inhibits food and water intake in rats. American Journal of Physiology 27 1: R848--R856, 1996.

179. Todd, J.F., Wildling, J.P., Edwards, C.M., Khan, F.A., Ghatei, M.A., and Bloom, S.R. Glucagon-like peptide-1 (GLP-1); a trial of treatment in non-insulin-dependent diabetes mellitus. European Journal of Clinical Investigation 27:S 3 -53 6, 1997.

180. Tseng, C., Boylan, M.O., Jarboe, L.A., Usdin, T.B., and Wolfe, M.M. Chronic desensitization of the glucose-dependent insulinotropic polypeptide receptor in diabetic rats. American Journal of Physiology 270:E66 1-E666, 1996.

181. Turton, M.D., O'Shea, D., Gum, I., Beak, S.A., Edwards, C.M., Meeran, K., Choi, S.J., Taylor, G.M., Heath, M.M., Lambert, P.P., Wilding, J.P., Smith, D.M., Ghatei, M.A., Herbert, J., and Bloom, S.R. A role for glucagon-like peptide-1 in the central regulation of feeding. Nature 379:69-72, 1997.

182. Uttenthal, L.O. and Blazquez, E. Characterization of high-affinity receptors for truncated glucagon-like peptide- 1 in rat gastric glands. FEBS Letter 262: 139- 148, 1990.

183. van der Kooy, D. Area postrema: Site where cholecystokinin acts to decrease food intake. Brain Research 295:345-357, 1984.

184. Van Dij k, A., Richards, J.G., Treciak, A., Gillessen, D., and Mohler, H. Cholecystokinin receptors: biochemical demonstration and autoradiographic localization in rat brain and pancreas using [3H]-cholecystokinin as radioligand. Journal of Nezaoscience 4:1021 - 1033, 1984.

185. Van Eyll, B., Lankat-Buttgereit, B., Bode, H.P., Goke, R., and Goke, B. Signal transduction of the GLP-1 receptor cloned from a human insulinoma. FEBS Letters 348:7-13, 1994.

186. Vaxillaire, M., Viomet, N., Vigouroux, C., Sun, F., Espinosa 111, R., Lebeau, M.M., Stoffel, M., Lehto, M., Beckmann, J.S., Detheux, M., Passa, P., Cohen, D., Van Schaftingen, E., Velho, G., Bell, G.I., and Froguel, P. Search for a third susceptibility gene for maturity- onset diabetes of the young. Diabetes 43 :3 89-395, 1994. 187. Walters, M.J., Brown, T.J., and MacLusky, N.J. In vitro autoradiographic visualization of occupied receptors in the rat brain with an iodinated estrogen ligand. Journal of Histochemistry and Cytochemistry 4 1: 1279- 1290, 1993.

188. Wang, Z., Wang, R.M., Owji, A.A., Smith, D.M., Ghatei, M.A., and Bloom, S.R. Glucagon-like peptide 1 is a physiological incretin in rat. Journal of Clinical Investigation 95:4 17421, 1995.

189. Wank, S.A., Pisegna, J.R., and De Weerth, A. Brain and gastrointestinal cholecystokinin receptor family: structure and functional expression. Proceedings of the National Academy of Sciences, USA 89:869 1-8695, 1992.

190. Wei, Y. and Mojsov, S. Tissue-specific expression of the human receptor for glucagon- like peptide-1: brain, heart and pancreatic forms have the same deduced amino acid sequences. FEBS Letters 358:2 19-224, 1995.

191. Weir, G.C., Bomer, and Weir, S. Islets of Langerhans: The puzzle of intraislet interactions and their relevance to diabetes. Journal of Clinical Investigation 85983-987, 1990.

192. Weir, G.C., Mojsov, S., Hendrick, G.K., and Habener, J.F. Glucagon-like peptide- l(7- 37) actions on endocrine pancreas. Diabetes 38:338-342, 1989.

193. Wettergren, A., Petersen, H., Orskov, C., Christiansen, J., Shiekh, S.P., and Holst, J.J. Glucagon-like peptide-1 (GLP-1) 7-36amide and peptide YY from the L-cell in the ileal mucosa are potent inhibitors of vagally induced gastric acid in man. Scandanuvian Journal of GustroenteroZogy 29:5O 1-505, 1994.

194. Wettergren, A., Schjoldager, B., Mortensen, P.E., Myhre, I., and Christiansen, J. Truncated GLP- 1 (proglucagon 72- 107amide) inhibits gastric and pancreatic functions in man. Digestive Diseases & Sciences 3 8:665-673, 1993.

195. Wettergren.A., Wojdemann, M., Meisner, S., Stadil, F., and Ho1st.J.J. The inhibitory effect of glucagon-like peptide-l (GLP-1) 7-36 amide on gastric acid secretion in humans depends on an intact vagal innervation. Gut 40597-601, 1997.

196. Wheeler, M.B., Lu, M., Dillon, J.S., Leng, X.H., Chen, C., and Boyd, A.E. Functionalexpression of the rat glucagon-like peptide- l receptor, evidence for coupling to both adenylyl cyclase and phospholipase C. Endocrinology 133:57-62, 1993.

197. WidmannC., Burke, E., Dolci, W., and Thorens, B. Signal transduction by the cloned glucagon-like peptide-1 receptor. Comparison with signalling by the endogenous receptors of b cell lines. Molecular Pharmacology 45: 1029- 103 5, 1993. 198. Woods, S.C., Taborsky, G.J., and Porte, D. Central nervous system control of nutrient homeostasis. Handbook of Phyiology, The Nervous System. 1 :365-411, 1986.

199. Yada, T., Itoh, K., and Nakata, M. Glucagon-like peptide-1 -(7-36) arnide and a rise in cyclic Adenosine 3',5'-Monophosphate increase cystolic fiee Ca2+ in rat pancreatic b-cells by enhancing Ca2+ channel activity. Endocrinology 133: 1685- 1692, 1993.

200. Yuan, H., Bowlby, D.A., Brown, T.J., Hochberg, R.B., and MacLusky, N.J. Distribution of occupied and unoccupied estrogen receptors in the rat brain: effects of physiological gonadal steroid exposure. Endocrinology 136:96- 105, 1995.

201. Zaborski, L., Beinfeld, M.C., Palkovits, J., and Mandheimer, L. Brainstem connection to the hypothalamus ventrornedial nucleus in the rat: a CCK-containing pathway. Brain Research 303:225-23 1, 1984.

202. Zarbin, M.A., Innis, R.B., Wamsley, J.K., Snyder, S.,H., and Kuhar, M.J. Autoradiographic localization of cholecystokinin receptors in rodent brain. Journal of Neuroscience 3 :877-906, 1983.

203. Zarbin, M.A., Wamsley, J.K., Innis, R.B., and Kuhar, M.J. Cholecystokinin receptors: presence and axonal flow in the rat vagus nerve. Life Sciences 29:697-673, 198 1.

204. Zielinski, J.E., Yabuki, H., Pahuja, S.L., Lamer, J.M., and Hochberg, R.B. 16a- [125~]iodo-l1P-methoxy-17P-est~adiol: A radiochemical probe for estrogen-sensitive tissues. Endocrinolo~119:130-139, 1986.