@prefix vivo: . @prefix edm: . @prefix ns0: . @prefix dcterms: . @prefix dc: . @prefix skos: . vivo:departmentOrSchool "Medicine, Faculty of"@en, "Cellular and Physiological Sciences, Department of"@en ; edm:dataProvider "DSpace"@en ; ns0:degreeCampus "UBCV"@en ; dcterms:creator "Hinke, Simon Amadeus"@en ; dcterms:issued "2009-11-13T00:00:00"@en, "2002"@en ; vivo:relatedDegree "Doctor of Philosophy - PhD"@en ; ns0:degreeGrantor "University of British Columbia"@en ; dcterms:description "Insulin secretory responses to oral glucose are compromised in type 2 diabetes. GIP receptor desensitization and internalization were studied as possible mechanisms for the blunted responsiveness to GIP in the human disease, employing in vitro cellular models. Using clonal insulin producing tumour cells (βTC-3) and rat GIP receptor transfected CHO-K1 cells, it was possible to characterize important aspects of receptor regulation. GIP receptor desensitization appeared to be slower than for other related receptors, and the rate appeared to parallel receptor internalization. Phosphorylation of receptor carboxyl terminal serine residues was implicated in both processes. Using co-transfection techniques and pharmacological agents, it was possible to partially delineate cellular proteins involved in GIP receptor desensitization and internalization. Dipeptidyl peptidase IV cleaves dipeptides from the N-termini of GIP, GLP-1 and glucagon, all of which are insulinotropic peptides involved in glucose homeostasis. Using cells transfected with the cognate receptors for these hormones, it was possible to demonstrate the importance of this enzyme in the modulation of hormone bioactivity. Structure-activity relationships for the peptides were designed to characterize the N-terminally truncated peptides, as well as design enzyme resistant molecules predicted to have superagonist activity in vivo. Such analogues with enhanced bioactivity may have a use in the treatment of diabetic states (GIP and GLP-1) or cardiovascular complications (glucagon). In vivo bioassay of these peptides confirmed their increased potency and highlighted their therapeutic potential. Additionally, fragment analysis was performed on GIP in an attempt to minimize the bioactive domain of the molecule, thus generating small molecular weight GIP receptor agonists."@en ; edm:aggregatedCHO "https://circle.library.ubc.ca/rest/handle/2429/14867?expand=metadata"@en ; dcterms:extent "12235793 bytes"@en ; dc:format "application/pdf"@en ; skos:note "M O D U L A T I O N O F I N S U L I N O T R O P I C H O R M O N E B I O A C T I V I T Y W I T H A F O C U S O N G L U C O S E - D E P E N D E N T I N S U L I N O T R O P I C P O L Y P E P T I D E (GIP) A N D ITS R E C E P T O R B y S I M O N A M A D E U S H I N K E B . Sc. (Hon.), The University of Bri t ish Columbia , 1997 A T H E S I S S U B M I T T E D I N P A R T I A L F U L F I L L M E N T O F T H E R E Q U I R E M E N T S F O R T H E D E G R E E O F D O C T O R O F P H I L O S O P H Y In T H E F A C U L T Y O F G R A D U A T E S T U D I E S A N D T H E F A C U L T Y O F M E D I C I N E , D E P A R T M E N T O F P H Y S I O L O G Y W e accept this,thesis as conforming to the required standard U N I V E R S I T Y O F B R I T I S H C O L U M B I A November 2002 © Simon Amadeus Hinke , 2002 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department The University of British Columbia Vancouver, Canada DE-6 (2/88) ABSTRACT Insulin secretory responses to oral glucose are compromised in type 2 diabetes. GIP receptor desensitization and internalization were studied as possible mechanisms for the blunted responsiveness to GIP in the human disease, employing in vitro cellular models. Using clonal insulin producing tumour cells ((3TC-3) and rat GIP receptor transfected CHO-K1 cells, it was possible to characterize important aspects of receptor regulation. GIP receptor desensitization appeared to be slower than for other related receptors, and the rate appeared to parallel receptor internalization. Phosphorylation of receptor carboxyl terminal serine residues was implicated in both processes. Using co-transfection techniques and pharmacological agents, it was possible to partially delineate cellular proteins involved in GIP receptor desensitization and internalization. Dipeptidyl peptidase IV cleaves dipeptides from the N-termini of GIP, GLP-1 and glucagon, all of which are insulinotropic peptides involved in glucose homeostasis. Using cells transfected with the cognate receptors for these hormones, it was possible to demonstrate the importance of this enzyme in the modulation of hormone bioactivity. Structure-activity relationships for the peptides were designed to characterize the N-terminally truncated peptides, as well as design enzyme resistant molecules predicted to have superagonist activity in vivo. Such analogues with enhanced bioactivity may have a use in the treatment of diabetic states (GIP and GLP-1) or cardiovascular complications (glucagon). In vivo bioassay of these peptides confirmed their increased potency and highlighted their therapeutic potential. Additionally, fragment analysis was performed on GIP in an attempt to minimize the bioactive domain of the molecule, thus generating small molecular weight GIP receptor agonists. i i Table of Contents A B S T R A C T ii T A B L E O F C O N T E N T S iii L I S T O F F I G U R E S v L I S T O F T A B L E S vi L I S T O F A B B R E V I A T I O N S vii P R E F A C E A N D D E D I C A T I O N ix A C K N O W L E D G E M E N T S x E P I G R A P H xi C H A P T E R 1: I N T R O D U C T I O N 1 1.1 OVERVIEW 1 1.2 THE DISCOVERY OF G I P 2 1.2.1 History: Enterogastrone Effects & Gastric Inhibitory Polypeptide 2 1.3 G I P SEQUENCE 3 1.3.1 Peptide Sequence and mRNA Isolation 3 1.3.2 The GIP Gene 5 1.3.3 Gene Regulation and GIP Expression 7 1.4 G I P MEASUREMENT AND RELEASE 9 1.4.1 Radioimmunoassay and Complications 9 1.4.2 Stimuli for Release 10 1.5 BIOLOGICAL EFFECTS 14 1.5.1 Gastrointestinal and Pancreatic Effects 14 1.5.2 The Enteroinsular Axis & Incretin Concept 16 1.5.3 Effects on Nutrient Storage and Metabolism 20 1.5.4 Other Attributed Biological Functions 23 1.6 G I P BINDING SITES 24 1.6.1 GIP lodination and Binding Studies 24 1.6.2 Cloning the GIP Receptor, Gene Expression & mRNA Distribution 29 1.7 G I P RECEPTOR SIGNAL TRANSDUCTION 33 1.8 THE PROGLUCAGON GENE PRODUCTS AND RECEPTORS 36 1.9 THESIS INVESTIGATION 39 C H A P T E R 2: M A T E R I A L S A N D M E T H O D S 40 2.1 REAGENTS 40 2.2 RECEPTOR PLASMID CONSTRUCTS 40 2.3 CELL CULTURE AND TRANSFECTION 41 2.4 PEPTIDES 45 2.5 PEPTIDE IODINATION 46 2.6 BINDING STUDIES 48 2.7 CYCLIC A M P MEASUREMENTS 49 2.8 INSULIN RELEASE EXPERIMENTS 52 2.9 RECEPTOR INTERNALIZATION 52 2.10 FLUORESCENCE MICROSCOPY 54 2.11 PEPTIDE DEGRADATION STUDIES AND CELL D P I V ACTIVITY 55 2.12 ANIMALS AND PEPTIDE BIOASSAY 56 2.13 HORMONE RADIOIMMUNOASSAYS 57 2.14 ANALYTICAL METHODS 58 C H A P T E R 3: R E G U L A T I O N O F G I P R E C E P T O R F U N C T I O N 59 3.1 INTRODUCTION 59 3.1.1 G-Protein Coupled Receptor Regulation 59 3.1.2 Regulation of Family B G-Protein Coupled Receptors 61 i i i 3.1.3 Potential Physiological Relevance of GIP receptor Desensitization 63 3.1.4 Thesis Objective 64 3.2 RESULTS 64 3.2.1 Desensitization offiTC-3 Cells to GIP 64 3.2.2 Desensitization of the Transfected GIP Receptor 71 3.2.3 Internalization of the Transfected GIP Receptor 78 3.3 DISCUSSION 106 3.3.1 Insulinoma Cell Desensitization 106 3.3.2 Desensitization of Transfected Cells 110 3.3.3 GIP Receptor Internalization in Transfected Cells 114 3.3.4 Conclusion 117 CHAPTER 4: ANALOGUES OF INSULINOTROPIC HORMONES 119 4.1 INTRODUCTION I '9 4.1.1 Structure-Activity Relationships of GIP 119 4.1.2 Metabolism of GIP 121 4.1.3 Metabolism of GLP-1 and Glucagon 123 4.1.4 Thesis Objective 125 4.2 RESULTS •.• 126 4.2.1 GIP Fragment Analysis '26 4.2.2 GlPiJ2 and Studies on Cellular DP1V in vitro 140 4.2.3 Design of DP IV-resistant GIP Analogues 144 4.2.4 Characterization of [D-Ala2]G1P',_42 in Vitro and in Vivo 147 4.2.5 Parallel Comparison of [Ser2] and [(P)Ser2] Substituted GIP and GLP-1 157 4.2.6 DPIV Degradation of Glucagon and DPIV-resistant Glucagon Analogues 165 4.3 DISCUSSION 177 4.3.1 GIP Fragments 177 4.3.2 DPIV-resistant lncretin Analogues 184 4.3.3 DPIV Degradation of Glucagon and DPIV-resistant Analogues 191 4.3.4 Conclusion 194 CHAPTER 5: SUMMARY AND FUTURE DIRECTIONS 196 APPENDIX A 201 REFERENCES 202 iv List of Figures Figure 1: The human G I P gene, m R N A and post-translational processing , 6 Figure 2: Schematic of the two dimensional topography of the rat G I P receptor 32 Figure 3: Post-translational processing of the proglucagon gene products 38 Figure 4: H P L C profiles of iodinated synthetic peptides 48 Figure 5: Effect of glycemic conditions on cycl ic A M P production in (3TC-3 cells 68 Figure 6: Time-course of homologous desensitization of GIP-stimulated c A M P production and effect of various inhibitors on desensitization in (3TC-3 cells 69 Figure 7: Insulin release from (3TC-3 cells 70 Figure 8: A representative saturation binding curve for w tGIPR cells 73 Figure 9: Effect of G I P prestimulation on the concentration-response curve of w tGIPR and r G I P R - L 2 cells 74 Figure 10: Time-course of c A M P accumulation in w tGIPR cells with and without I B M X 75 Figure 11: Time-course of desensitization in r G I P R - L 2 cells 76 Figure 12: Desensitization of C-terminal mutant G I P receptors 77 Figure 13: Time-course of 1 2 5 I -GIP binding to w tGIPR cells 79 Figure 14: Effect of acid stripping alone or in combination with 60 min 100 n M G I P pretreatment on G I P receptor binding on wtGIPR cells 80 Figure 15: G I P receptor internalization by agonist and antagonist 82 Figure 16: Effect of sucrose and monensin on receptor internalization in w t G I P R cells 87 Figure 17: Ligand-independent G I P receptor internalization 88 Figure 18: Effect of co-transfection of the G I P receptor with G R K - 2 90 Figure 19: GIP-stimulated c A M P production in C T Ser to A l a substitution mutant receptors expressed in C H O - K 1 cells 93 Figure 20: Competitive-binding studies on C H O - K 1 cells stably transfected with C-terminal Ser to A l a mutant G I P receptors 94 Figure 21: Internalization kinetics of C-terminal serine to alanine mutant receptors in transfected C H O - K 1 cells 95 Figure 22: Binding competition experiments with C-terminal green fluorescent protein (GFP) tagged G I P receptors in transfected C H O - K 1 cells 98 Figure 23: C y c l i c A M P production by subclones of G f P R - G F P cell lines 99 Figure 24: Internalization of G I P R - G F P in transfected C H O - K 1 cells 100 Figure 25: Fluorescence microscopy of G I P R - G F P distribution in transfected C H O - K 1 cells.. 101 Figure 26: Bind ing and c A M P production of fluorescein-conjugated G I P 103 Figure 27: Loss of w tGIPR binding sites on incubation with [Fluo-Trp 2 5 ]GIP 104 Figure 28: Fluorescence microscopy of w tGIPR cells incubated with [Fluo-Trp 2 5 ]GIP 105 Figure 29: Competition-binding displacement curves of synthetic G I P fragments on w tGIPR cells 128 Figure 30: Competition-binding of modified G I P M 4 analogues on w tGIPR cells 129 Figure 31: C y c l i c A M P production in wtGIPR cells by selected bioactive truncated peptides.. 132 Figure 32: C y c l i c A M P production by 20 u M of substituted G I P M 4 peptides 133 Figure 33: Concentration-response curves of intracellular cycl ic A M P production in w tGIPR cells by N-terminally modified GIP , . , 4 peptides 134 Figure 34: Antagonism of native G I P by C-terminal G I P fragments 135 Figure 35: Pancreatic perfusion of G I P fragments in rats 137 Figure 36: Bioassay of G I P , . 4 2 in anesthetized male Wistar rats 138 Figure 37: Glucose lowering effects of G I P fragments in anesthetized Wistar rats 139 Figure 38: Competitive inhibition of G I P , . 4 2 by G I P 3 . 4 2 142 v Figure 39: C e l l associated D P I V activity 143 Figure 40: Binding affinity and cAMP-s t imula t ing ability of modified G I P , . 3 0 N H 2 analogues.... 146 Figure 41: Incubation of 1 2 5 I - G I P , . 4 2 or l 2 5 I - [D-Ala 2 ]GIP ,_ 4 2 with D P I V 148 Figure 42: Binding studies using , 2 5 I - G I P M 2 or l 2 5 I - [ D - A l a 2 ] G I P , . 4 2 as tracer 149 Figure 43: Bioactivity of [D-Ala 2 ]GIP i_ 4 2 in vitro 150 Figure 44: Bioassay of G I P M 2 and [ D - A l a 2 ] G I P , . 4 2 in Wistar rats 153 Figure 45: Bioassay of G I P 3 . 4 2 in conscious Wistar rats 154 Figure 46: Bioassay of [ D - A l a 2 ] G I P M 2 in conscious lean V D F Zucker rats 155 Figure 47: Bioassay of [D-Ala 2 ]GIP,_ 4 2 in conscious obese V D F Zucker rats 156 Figure 48: Binding and c A M P stimulation in wtGIPR cells by G I P , . 4 2 , [Ser^GIP, [(P)Ser 2]GIP 1. 3 0 N H 2 159 Figure 49: Binding and c A M P stimulation in w t G L P - l R cells by G L P - 1 7 . 3 6 N H 2 , [ S e r 2 ] G L P - l 7 . 3 6 N H 2 and [ ( P ) S e r 2 ] G L P - l 7 . 3 6 N H 2 .\". 160 Figure 50: Bioassay of native and Ser 2-substituted incretin analogues in Wistar rats 163 Figure 51: Bioassay of Phosphosei^-substituted incretin analogues in Wistar rats 164 Figure 52: Inhibition of D P I V degradation of glucagon by Ile-thia monitored by bioassay 166 Figure 53: Bioactivi ty of glucagon and N-terminally truncated analogues in vivo 167 Figure 54: H P L C separation of 1 2 5I-glucagon incubated with porcine D P I V 168 Figure 55: Concentration-dependent stimulation of c A M P production in h G l u c R cells by glucagon and synthetic fragments 170 Figure 56: Antagonist properties of N-terminally truncated forms of glucagon 171 Figure 57: Competitive-binding of synthetic glucagon fragments on hGlucR cells 172 Figure 58: In vitro characterization of N-terminally modified glucagon analogues 174 Figure 59: Bioactivi ty of glucagon analogues in vivo 176 Figure 60: Predicted secondary structure of G I P 180 List of Tables Table 1: A m i n o acid sequence alignments of G I P from different species 4 Table 2: Comparison of binding constants measured using transformed cells and transfected cell models 28 Table 3: Summary of cycl ic A M P production in (3TC-3 cells in response to G I P and forskolin.65 Table 4: Effect of signal transduction cascade activators/inhibitors on basal and GIP-stimulated c A M P accumulation in w tGIPR cells 84 Table 5: Effect of signal transduction cascade activators/inhibitors on G I P receptor expression and internalization in w tGIPR cells 85 Table 6: Binding and internalization characteristics of C H O - K 1 cells stably transfected with G I P receptor C-terminal serine mutants 92 Table 7: Summary statistics for studies on synthetic G I P fragments using w t G I P R cells 127 Table 8: Bind ing and c A M P statistics for modified G I P | _ 3 0 N H 2 analogues 145 Table 9: Integrated glucose and insulin profiles for G I P and [ D - A l a 2 ] G I P in vivo 157 Table 10: Integrated glucose and insulin profiles for GIP , G L P - 1 , [Ser 2]- and [(P)Ser 2]-substituted analogues in vivo 162 Table 11: Summary of glucagon receptor binding and activation by synthetic peptides 175 Table 12: Predicted and measured molecular masses of synthetic peptides 201 vi List of Abbreviations Amino Acids: 3 Letter Code 1 Letter Code Alanine Ala A Arginine Arg R Asparagine Asn N Aspartate Asp D Cysteine Cys C Glutamine Gin Q Glutamate Glu E Glycine Gly G Histidine His H Isoleucine He I Leucine Leu L Lysine Lys K Methionine Met M Phenylalanine Phe F Proline Pro P Serine Ser S Threonine Thr T Tryptophane Try W Tyrosine Tyr Y Valine Val V PCR = polymerase chain reaction RT-PCR = reverse-transcription PCR ORF = open reading frame mRNA = messenger ribonucleic acid cDNA = complementary DNA DNA = deoxyribonucleic acid Da = Dalton (molecular mass) HPLC = high pressure liquid chromatography/high performance liquid chrmatography MALDI-TOF MS = matrix-assisted laser-desorption ionization-time of flight mass spectrometry CT = carboxyl terminal (-COOH-terminal) NT = amino terminal (-NH2-terminal) sp = synthetic porcine np = natural porcine sh = synthetic human wt = wild type E C 5 0 - effective concentration to achieve 50% of the maximal biological response IC 5 0 = concentration inhibiting 50% of measured parameter (e.g. radioligand binding) cpm = counts per minute (radioactivity) AUC = area under the curve (integrated) CHO - Chinese hamster ovary cells i CHL = Chinese hamster lung cells HEK = Human embryonal kidney cells COS = green monkey kidney cells PTC = mouse insulinoma cells EDTA = ethylenediaminetetraacetate DMSO = dimethylsulfoxide HEPES = A'-2-hydroxyethylpiperazine-A''-2-ethane sulfonic acid IBMX = 3-isobutyl-l-methylxanthine BSA = bovine serum albumin Ab = antibody DPIV = dipeptidyl peptidase IV (CD26) vii A V P = arginine vasopressin (anti-diuretic hormone, A D H ) G R P = gastrin releasing peptide GLP-1 = glucagon-like peptide-1 GLP-2 = glucagon-like peptide-2 GIP = glucose-dependent insulinotropic polypeptide/gastric inhibitory polypeptide VIP = vasoactive intestinal peptide C C K = cholecystokinin G H R H / G R F = growth hormone releasing hormone/growth hormone releasing factor PHI /PHM = peptide histidine isoleucine/peptide histidine methionine E G F = epidermal growth factor PTH = parathyroid hormone/parathormone T H R H = thyroid hormone releasing hormone L H = luteinizing hormone F S H = follicle stimulating hormone \" R \" following these abbreviations denote the hormone's receptor G P C R = G-protein coupled receptor G R K = G-protein coupled receptor kinase G-protein = heterotrimeric guanosine triphosphate (GTP) binding proteins A T P = adenosine 5'-triphosphate A D P = adenosine 5'-diphosphate c A M P = adenosine 3',5'-cyclic monophosphate C a 2 + = cellular calcium PIP 2 = phophotidylinositol 4,5,-bisphosphate IP 3 = inositol 1,4,5-trisphosphate D A G = diacylglycerol C R E = cyclic A M P response element A d C = adenylyl cyclase (forms c A M P from ATP) P L C = phospholipase C (forms IP 3 and D A G from PIP 2) M A P K / E R K = mitogen activating protein kinase/extracellular regulated kinase PI3K = phosphotidylinositol 3' kinase P K A = protein kinase A/cyclic A M P dependent protein kinase P K C = protein kinase C G E F = guanine-nucleotide exchange factor G A P = GTPase-activating protein P D E = phosphodiesterase (forms A M P from c A M P ) IV = intravenous ICV = intracerebroventricular IP = intraperitoneal ID = intraduodenal SC = subcutaneous V D F = Vancouver diabetic fatty (Zucker rat) Fa/? = homozygous/heterozygous dominant for functional leptin receptor = lean Zucker rat fa/fa = homozygous recessive for non-functional leptin receptor = obese Zucker rat v i i i Preface and Dedication This thesis is dedicated to my high school teachers for their understanding and patience. In particular, M r s . Broome, M s . VanHeteren, and M r . McCar thy inspired me to pursue science, and Jackie M c D o n a l d , with her friendship, instilled in me enthusiasm and faith in humanity. Some of the work contained in this thesis has been published: Hinke , S A , Gel l ing , R , Manhart, S, Lynn , F , Pederson, R A , Kuhn-Wache, K , Rosche, F , Demuth, H - U , and C H S Mcintosh , 2002. Structure-activity relationships of glucose-dependent insulinotropic polypeptide (GIP). Biol. Chem. (in press). Hinke , S A , Gel l ing , R W , Pederson, R A , Manhart, S, Nian , C , Demuth, H - U , and C H S Mcin tosh , 2002. Dipeptidyl peptidase IV-resistant [D-Ala 2]glucose-dependent insulinotropic polypeptide (GIP) improves glucose tolerance in normal and obese diabetic rats. Diabetes 51:652-661. Hinke , S A , Manhart, S, Pamir, N , Demuth, H - U , Gel l ing , R W , Pederson, R A , and C H S Mcin tosh , 2001. Identification of a bioactive domain in the amino-terminus of glucose-dependent insulinotropic polypeptide (GIP). Biochim. Biophys. Acta 1547:143-155. Pospisi l ik, J A , Hinke , S A , Hoffmann, T , Rosche, F , Schlenzig, D , Heiser, U , Glund , K , Mcin tosh , C H S , Pederson, R A and H - U . Demuth, 2001. Metabol ism of glucagon by dipeptidyl peptidase I V . Regul. Pept. 96:133-141. Hinke , S A , Pauly, R P , Ehses, J , Kerridge, P, Demuth, H - U , Mcin tosh , C H S , and R A Pederson, 2000. Role of glucose in chronic desensitization of isolated rat islets and mouse insulinoma ((3TC-3) cells to glucose-dependent insulinotropic polypeptide. J. Endocrinol. 165:281-291 Hinke , S A , Pospisil ik, J A , Demuth, H - U , Manhart, S, Ki ihn-Wache, K , Hoffmann, T , Nishimura, E , Pederson, R A and C H S Mcin tosh , 2000. Dipeptidyl Peptidase I V Degradation of Glucagon : Characterization of Glucagon Degradation Products and D P I V Resistant Analogs. J. Biol. Chem. 275:3827-3834 Wheeler, M B , Gel l ing , R W , Hinke , S A , T u , B , Pederson, R A , L y n n , F , Ehses, J , and C H S Mcin tosh , 1999. Characterization of the carboxyl-terminal domain of the rat glucose-dependent insulinotropic polypeptide (GIP) receptor. A role for serines 426 and 427 in regulating the rate of internalization. J. Biol. Chem. 274:24593-24624. Acknowledgements I would l ike to thank Drs . Chris Mcin tosh , Ray Pederson, U l i Demuth and M i k e Wheeler. Each of them have characteristics that I admire, that have helped me during my learning process under their tutelage. W i t h Chris as a supervisor, I have learned what academic freedom really means, and I have valued every opportunity that arose because of the freedom I was given. I was allowed to develop as a scientist at my own rate, and Chris was a constant resource to draw on. Ray was l ike a co-supervisor to me. Al though this was not an official title, his interest and support for the work, as wel l as the friendship that has developed over the years I have known him are valued. I think that all of us who have been graduate students with Ray and Chris as supervisors have been given the greatest chance to succeed. A l l of our future work w i l l be indebted to them. Collaboration with U l i and M i k e has been very fruitful, and their enthusiasm and drive has often inspired me. Without them, much of this work would not have been possible. T w o people have helped me during this degree more than any others. Cu i l an N i a n and Madeleine Speck have worked with me side-by-side, and in some cases, as extra pairs of hands for me. I appreciate every bit of help you have offered me, and every bit of help I had to ask for. Certainly, this thesis would have been much shorter without your help! I also would l ike to thank Dr . Susanne Manhart for the shared interest in the G I P structure-activity relationship project. Our fruitful discussions led to many fruitful studies, and answered many questions. Without her, none of it would have happened. I should also thank the students that worked under me, Mary Grace Miraf lor and Paul Sanders. I think we all learned from those experiences. Irene Bremsak should also be given thanks for keeping things in order while she was with us. I would also l ike to thank my compadres and co-workers. It has been a long haul, and I am glad that we all worked together. Once we all found our niches, we worked we l l together. Intellectual discussions, beers and friendship were always appreciated. M y family has been a tremendous support for me throughout my extended education. Thank you for everything. . . I love you a l l . F ina l ly , last but not least, I have to thank my girls, Christine and Sasha. These two have taught me a thing or two about love, life, relationships, and above al l , have given me the much needed break from my work to enjoy life outside of the lab. Simon Hinke , November 8 t h, 2002 x EPIGRAPH CORPORA NON AGUNT NISI FIXATA - P. Ehrlich Chapter 1: Introduction 1.1 Overview G I P is a p o l y p e p t i d e h o r m o n e h a v i n g s t ruc tu ra l s i m i l a r i t i e s to the glucagon/secretin/vasoactive intestinal polypeptide superfamily of hormones. It is thought that G I P arose from a series of gene duplications occurring over the last 1,000 mi l l ion years, derived from the ancestral gene for pituitary adenylyl cyclase activating polypeptide [1]. T o date, G I P has yet to be identified in fish or birds, suggesting that the final gene duplication, resulting in the ancestral G I P gene, must have been prior to the divergence of modern mammals, 72 m i l l i o n years ago [1; 2]. Protein purification methods expanded in the 1960s and 1970s, and interest in gut hormones increased. Initially, G I P was identified for its enterogastrone (acid inhibitory) effects, hence the first term for the hormone, gastric inhibitory polypeptide. Large scale purification from pork intestine a l lowed the elucidation of the amino acid sequence of G f P . Based on sequence similarity to glucagon, it was hypothesized that G I P might have a role in glucose homeostasis. In contrast to the diabetogenic actions of glucagon, G I P acted to lower glycemia by augmenting nutrient stimulated insul in release from the pancreatic Islets of Langerhans. Thus came the second designation for GIP : glucose-dependent insulinotropic polypeptide, retaining the original acronym for the hormone [3]. Init ial b iochemical characterization of the G I P receptor was hindered by the apparent difficulty radiolabell ing the hormone. Measurement of G f P by radioimmunoassay did not require biological ly active G I P tracer, however, receptor binding studies did. Identification of the biological ly active iodinated G I P product separated by high pressure l iquid chromatography allowed these studies to begin in the mid 1980s [4]. Dur ing the same time, the first G-protein coupled receptor ( G P C R ) was cloned [5; 6]. Nearly a decade later, molecular biology techniques had advanced to the point where low-stringency screening of c D N A libraries was al lowing the 1 cloning of families of receptors, based on conserved structural motifs. In 1991, the receptor for secretin was first c loned [7], and shortly thereafter, the receptor c D N A s for glucagon-like peptide-1 ( G L P - 1 ) [8], glucagon [9; 10], and G I P [11] were also obtained. The cloning of the G I P receptor has al lowed three significant areas of research to advance: (1) Structure-activity relationships of synthetic G I P could be examined in an isolated model , (2) the signal transduction cascades uti l ized by the receptor could be dissected more easily, and (3) site-directed mutagenesis of the receptor allowed the molecular characterization of receptor binding, activation, and regulation. 1.2 The Discovery of GIP 1.2.1 History: Enterogastrone Effects & Gastric Inhibitory Polypeptide Ear ly studies on impure hog intestine mucosal extracts containing cholecys tok in in-pancreozymin ( C C K - P Z ) established the role of C C K - P Z in stimulating enzyme release from the exocrine pancreas and contraction of the gall bladder, however, when looking at gastrointestinal (GI) motility or acid secretion, contradictory results were obtained [12]. When two preparations of C C K - P Z with differing purities, 200 Ivy dog units/mg (\"10% pure\") and 1,500 Ivy dog units/mg (\"40% pure\"), were examined in a multiparameter study in conscious dogs, evidence for a second contaminating hormone emerged. When C C K - P Z activity was held constant (0.2 IDTJ/kg/min) at sub-maximal doses for gall bladder contraction (measured using a bioassay in guinea pig), infusion into dogs gave equivalent changes in intra-gallbladder pressure, antral motor activity, and possibly pepsin secretion, however, significant differences in acid secretion from denervated fundic (Heidenhain) pouches were measured [13]. B r o w n and Pederson proposed two possibilities for these observations: either they had concentrated a gastric stimulant during purification or an acid inhibitory molecule was proportionately removed during the purification protocol. These hypotheses were later tested using a similar canine model with 2 mucosal extracts having undergone a series of gel chromatography separations and precipitation steps; fractions were tested for C C K - P Z activity using a bioassay, and all fractions not having significant guinea pig gallbladder contracting ability were pooled. Thus, it was demonstrated that it was possible to purify a polypeptide having the abi l i ty to potently inhibi t both acetylcholine- and pentagastrin-stimulated acid secretion [12; 14; 15]. W h i l e results were preliminary, the authors suggested that this purified polypeptide was the enterogastrone (gastric inhibitory) substance described by Kosaka and L i m some thirty-nine years earlier [16]. It wasn't until B r o w n published the amino acid composi t ion and tryptic peptides o f the purif ied polypeptide that he officially dubbed the molecule \"Gastric Inhibitory Polypeptide,\" giving the acronym G I P [17]. 1.3 GIP Sequence 1.3.1 Peptide Sequence and mRNA Isolation Not long after the initial isolation of G I P from pig intestinal mucosa, B r o w n published the first amino acid sequence data for the newly discovered peptide [17]. U s i n g gel filtration chromatography and high voltage paper electrophoresis, fragments o f G I P generated by enzymatic cleavage with trypsin or chymotrypsin, or chemically cleaved with cyanogen bromide ( C N B r ) were separated, and amino acid analyses and partial peptide sequencing were performed. Quantitative amino acid analysis suggested a molecule of 42 or 43 amino acid residues - it was uncertain as to the number of G l x residues present. Later the same year, the complete amino acid sequence was published; the Edman degradation sequencing method had been applied to all o f the G I P fragments [18]. Unfortunately, the uncertainty in number of G i n residues was not entirely resolved; the sequence in question, tryptic fragment Tr-3b, was difficult to analyze due to low yields and amino acid composition. Comparisons between natural and early synthetic analogues were unable to confirm the original published structure, and the hypothesis was put 3 forward that the or ig inal sequence of B r o w n and Dryburgh was incorrect [19]. U p o n reinvestigation, Jornval l and colleagues used H P L C to separate G I P fragments generated by C N B r or N-chlorosuccinimide (cleaving G I P after each of its two tryptophane residues), as wel l as trypsin, prior to Edman sequencing. Hence, it was confirmed that indeed, one too many G i n residues was inserted at position 30 of the molecule, and in fact, G I P was only a 42 amino acid polypeptide [19]. Isolation and peptide sequencing of human G I P showed that only minor species variations existed between pig and man - A r g 1 8 and Ser 3 4 from the porcine sequence were replaced with His and A s n in Homo sapiens, respectively [20]. The same year, the sequence of cow G I P was also reported, differing at only one residue from the pig isoform (Table 1) [21]. The sequence of human G I P was later confirmed upon the isolation of an intestinal c D N A encoding the human isoform [22], and the cloning of the human G I P gene [23]. Sequences of rodent G I P isoforms have been entirely deduced from D N A sequences; three groups published the rat c D N A sequence in 1992 and 1993 [24-27], and the mouse c D N A sequence was published in 1996 [28]. Comparing isoforms from all species, very little variation exists - the sequences sharing greater than 90% sequence identity among mammals. G I P has yet to be isolated from fish, amphibians, birds or invertebrates, thus little is known about the molecular evolution of the molecule [1; 2]. Table 1: Amino acid sequence alignments of GIP from different species Y - A - E - G - T - F - I - S - D - Y - S - I - A - M - D - K - I - H - Q - Q - D - F - V - N - W - L - L - A - Q - K - G - K - K - N - D - W - K - H - N - I - T - Q Y - A - E - G - T - F - I - S - D - Y - S - I - A - M - D - K - I - R - Q - Q - D - F - V - N - W - L - L - A - Q - K - G - K - K - S - D - W - K - H - N - I - T - Q Y - A - E - G - T - F - I - S - D - Y - S - I - A - M - D - K - I - R - Q - Q - D - F - V - N - W - L - L - A - Q - K - G - K - K - S - D - W - I - H - N - I - T - Q Y - A - E - G - T - F - I - S - D - Y - S - I - A - M - D - K - I - R - Q - Q - D - F - V - N - W - L - L - A - Q - K - G - K - K - N - D - W - K - H - N - I - T - Q Y - A - E - G - T - F - I - S - D - Y - S - I - A - M - D - K - I - R - Q - Q - D - F - V - N - W - L - L - A - Q - K - G - K - K - N - D - W - K - H - N - L - T - Q Y - A - E - G - T - F - I - S - D - Y - S - I - A - M - D - K - I - R - Q - Q - D - F - V - N - W - L - L - A - Q - R - G - K - K - S - D - W - K - H - N - I - T - Q H u m a n [20], porcine [17-19] and bovine [21] sequences were first obtained by protein sequencing. Molecular cloning techniques allowed the deduction of rodent G I P sequences from genomic [25] or m R N A sequences [24; 26-29]. The human sequence was confirmed with the cloning of the G I P gene and isolation of an intestinal c D N A encoding G I P [22; 23]. Shadowed amino acids represent variations from the porcine sequence. 4 1.3.2 The GIP Gene Inagaki and colleagues [23] were the first to report the structure o f the human G I P gene, isolating clones from a human genomic phage library using a c D N A probe. D N A sequencing of three overlapping clones indicated that the human G I P gene consisted of six exons encoding G I P m R N A , with five intervening introns [23]. Mature processed G I P i _ 4 2 was encoded within exons 3 and 4, whereas the unprocessed translated product of G I P m R N A , preproGIP, was encoded by exons 2, 3, 4 and 5, and the 5'-untranslated region and 3'untranslated region/polyadenosine tail were in exons 1 and 6, respectively (Figure 1). Chromosomal localization indicated the G I P gene mapped to chromosome 17q [23; 30; 31]. Analysis of human c D N A clones showed a 459 base pair open reading frame encoding a 153 amino acid protein (preproGIP) with a predicted molecular weight of approximately 17 K D a [22]. Removal of the signal peptide fol lowing G l y 2 1 and proteolytic processing to mature G I P by unidentified convertases fol lowing single arginine residues indicate production of a putative 21 amino acid signal peptide, a 30 residue N H 2 -terminal peptide and a 60 residue COOH-te rmina l peptide, with G I P ' s 42 amino acid sequence in between (Figure 1). The rat genomic G I P sequence has also been shown to consist of 6 exons and 5 introns, with the m R N A encoded within exons 2-5, and mature G I P M 2 split between exons 3 and 4 [25], fo l lowing a pattern similar to the human gene (Figure 1). Analysis of rodent G I P m R N A sequences suggest a putative preproGIP similar to that predicted in humans: a 432 base pair open reading frame encoding the unprocessed 144 amino acid product, with a 21 amino acid signal sequence, a 22 amino acid NH 2 - te rmina l peptide, the 42 amino acid hormone, and a 59 amino acid COOH- te rmina l peptide. Due to the existence of a G l y 3 ' - L y s 3 2 - L y s 3 3 sequence in all forms of mature GIP,_ 4 2 , it has been suggested that cel lular processing may result in the production of G I P , . 3 0 N H 2 in vivo [28; 32], although its existence has not been confirmed experimentally. 5 A . g e n o m i c D N A 1 Kb Exon 1 Exon 2 Exon 3 Exon 4 Exon 5 Exon 6 Figure 1: The human GIP gene, mRNA and post-translational processing Adapted from [20; 22; 23]. 1.3.3 Gene Regulation and GIP Expression The 5'-upstream sequence from the human G I P gene was also determined, in order to gain insight into the potential regulation of G I P gene transcription; a T A T A motif and consensus sequences for transcription factors S p l , A p - 1 , A p - 2 and a C R E element were identified, implicating regulation by cycl ic A M P and protein kinases A and C [23]. Basal promoter activity of human G I P m R N A transcription was found to require the D N A sequence -180 to +14 (relative to the transcriptional initiation site), whereas inducible transcription was primarily mediated by one of two C R E elements in the regions -164 to -149, and c-Jun was capable of repressing transcription [33]. Several transcription factor motifs were identified in the rat G I P promoter [25]; delineation of regulatory domains of the rat promoter indicated a proximal and distal promoter, wi th cell-specific expression mediated by a G A T A mot i f in the distal promoter, between -193 and -182 from the transcriptional start site [34]. Rat duodenal G I P m R N A transcripts have been detected as early as days 18-21 of embryogenesis [35; 36], but it is unclear as to whether low levels are present until birth, followed by a post-natal increase in transcription [36], or whether G I P m R N A transcripts steadily increase until birth, whereupon they reach adult levels [35]. It has been reported that in rats, 2 day starvation either nearly doubled [27] or halved [37] G I P m R N A levels. Clarif icat ion of this finding is necessary, as the question remains whether absence of nutrient stimulation acts as an anticipatory signal to promote biosynthesis of G I P for the next meal , or alternately, that biosynthesis of G I P during starvation is an expendible process during a time where energy should be conserved. Glucose has been reported to promote G I P transcript abundance in cell lines and rodents, with mi ld increases measured in immunoreactive G I P content after oral or ID glucose [28; 37; 38]. However , in hyperglycemic diabetic animals no changes in either G I P m R N A or protein content could be measured [39]. Gastric administration of peptone was found to have no influence on duodenal G I P gene transcription, but was able to increase intestinal G I P 7 content; this effect was reversed by omeprazole, indicating that peptone-induced effects were mediated v ia stimulation of acid secretion. Confirmation of this hypothesis was shown by perfusing rat duodenum wi th 0.1 M H C 1 , wh ich increased both G I P transcription and immunoreactive G I P content [38; 40]. Dietary fat has also been reported to increase G I P m R N A levels [26; 37], but d id not result in increased duodenal G I P peptide levels [37]. Hence, G I P expression is regulated developmentally and by dietary intake. Early histological studies defined the intestinal mucosal cells containing immunoreactive (IR) G I P . These cells were determined to belong to the A P U D (amine precursor uptake and decarboxylation) type, but distinct from other enteroendocrine cells; G I P containing cells were identified to be the K - c e l l [41-43]. Original studies found K-ce l l s to have a distribution with greatest quantities in the duodenum and upper jejunum, dist inctly different from L-ce l l s containing gut glucagons, which were primari ly in the jejunum and i leum [41; 43]. Recent studies have suggested that a population of L-ce l l s in the jejunum exists that are co-local ized with K-ce l l s [44; 45], but this remains to be confirmed. Radioimmunoassay of extractable G I P from tissues compares wel l with m R N A expression profiles. G I P m R N A and I R - G I P are found in highest amounts in the proximal small intestine, and are present in decreasing amounts distally in the bowel [3]. The only convincingly confirmed regions in which G I P has been detected outside of the gut are in ductal cells of the submandibular salivary gland and in a specific population of endocrine cells in the stomach [26; 29; 35]. In addition to using the traditional methods of P C R detection, in situ hybridization and immunocytochemistry to define the stomach and intestinal populations of GIP-posi t ive cells, Yeung et al [29] generated mice with a herpes simplex thymidine kinase gene driven by the human G I P promoter region, and examined tissue enzyme activity. This technique revealed significant enzyme activity in the stomach and pancreas, however, activity was not detected in the small intestine. The absence of enzyme activity in the duodenum, a wel l 8 established G I P expressing tissue, may be due to species differences controlling gene expression (i.e. human promoter in mice), or the G I P promoter construct was incomplete, and additional 5' upstream or intronic sequence is necessary for intestinal transgene expression. A similar transgenic approach employing the rat G I P promoter resulted in transgene expression in both stomach and duodenum [46]. Both the rat and human promoters were able to drive gene transcription in the insulinoma cel l line, H1T-T15, to a certain degree, but this was extremely small compared to the level seen in the enteroendocrine tumor line, S T C - 1 [33; 34]. M o s t studies show that G I P is not expressed in esophagus, stomach, gastric antrum, gastric fundus, gastric cardia, colon, cecum, rectum, liver, gall bladder, pancreas, kidney, spleen, lung, heart, muscle, ovary, brain, hypothalamus, pituitary, spinal cord, or thyroid [22-24; 26; 27; 29; 39; 46; 47]. Apart from enriched primary cultures of K-ce l l s [48] or E. coli transformed with a synthetic G I P expression plasmid [49], expression of G f P in vitro has been l imited to only human embryonic intestinal 407 cells, STC-1 cells or derivatives of this tumor line, among a host of cell types tested, including cells derived from the salivary gland [28; 34; 50]. 1.4 GIP Measurement and Release 1.4.1 Radioimmunoassay and Complications Since the first development of a radioimmunoassay for G I P in 1974 [51], studies have been performed in a variety of species, including python, amphibians, pig, dog, rat, cow, sheep, goat, horse, and human, but many of these are complicated by the cross-reactivity of antibodies raised against natural source G I P [3; 52; 53]. Generally, antibodies have been raised against an epitope of G I P within amino acids 15-42 of the primary sequence; unfortunately, this same sequence is the region in which the greatest species differences in G I P occur (Table 1), thus resulting in inconsistent results when used for measuring G I P levels in other animals [52; 54]. Furthermore, these antibodies differentially recognize 4 or 5 immunoreactive species, only one of which is the 9 5 K D a peptide responsible for the known biological effects of G I P [3; 52; 53; 55; 56]. In a study designed to compare several antibodies in parallel, Jorde et al [56] reported human fasting basal G I P levels fell within the range 12-92 p M , and rose to 35-235 p M postprandially. Generally speaking, most studies report between a 2 X and 10X increase in immunoreact ive G I P concentrations in response to appropriate stimuli regardless of the antibody employed or study subject. M o r e recent studies have shown that G f P is inactivated by dipeptidyl peptidase I V ( D P I V ; discussed in Chapter 4), and that G I P 3 . 4 2 represents a significant proportion of circulating I R - G I P , further hindering interpretation of the data [19; 53; 57-61]. Another complication when comparing literature values of G f P release in response to different st imuli are the different methodologies used. Commonly , oral nutrients are given in solid or l iquid form, with varied composit ion (glucose/carbohydrate, fat, protein/amino acids), but it is we l l documented that meal composition can dramatically affect gastric emptying, and thus nutrient delivery to the site of G I P release. Furthermore, salivary and gastric enzymes may metabolize nutrients prior to reaching the K - c e l l , or nutrients may stimulate release of secondary agents, such as gastric acid, making it difficult to conclusively determine factors capable of releasing G I P . For these reasons, emphasis is put on studies using intra-duodenal (ID) infusion of nutrients for in vivo release experiments and in vitro models for G I P secretion. 1.4.2 Stimuli for Release 1.4.2.1 Carbohydrates: It is generally accepted that G I P granule exocytosis from the K - c e l l is by direct dose-dependent simulation with specific sugars, fats, and amino acids, as wel l as duodenal acid; hence ingestion of a mixed meal is a potent stimulus for hormone release [62; 63]. Glucose is perhaps the best characterized stimulant for G I P release, as shown by ID perfusion studies in rat, dog, pig and human [55; 63-74]. A n elegant study by Thomas et al [75] using positional perfusion of the 10 human small intestine with proximal occlusion, demonstrated glucose-stimulated G I P secretion was greatest from the duodenum and proximal jejunum, compared to the mid-jejunum or i leum. It was shown that G I P release required intestinal sodium-dependent sugar transport and was specific for ID glucose, galactose, maltose or sucrose - fructose, mannose, 6-deoxygalactose, 2-deoxyglucose, myoinositol, sorbitol and lactose were without effect [64; 70; 76; 77]. 1.4.2.2 Lipids: Similar ly , ID fat has been shown to be a potent G I P secretagogue [55; 67; 74]. Specifically, long chain fatty acids were required to stimulate G I P secretion, whereas medium chain fatty acids had no effect [78; 79]. The differential stimulation by medium versus long chain fatty acids is thought to be l inked to chylomicron formation; medium chain fatty acids are directly absorbed, whereas absorption of long chain fatty acids requires formation of chylomicrons. Blockade of chylomicron formation completely ablated the G I P response to intra-duodenal fat and the associated rise in plasma triglycerides [80; 81]. In comparing oral fat and oral glucose in dogs, it was demonstrated that lipids caused a greater and sustained release of G I P compared to glucose, an effect l ikely due to the slower gastric emptying of fat [82]. Another study comparing isocaloric doses of intra-duodenal glucose or l ip id in humans indicated fat to be the more potent stimulus on both weight and molar basis [55]. Results from studies in swine suggested that fat alone was a weak G I P stimulus, but when in administered in combination with glucose, it was able to stimulate greater G I P release than glucose alone [67]. 1.4.2.3 Protein and pH: Intra-gastric peptone (protein hydrolysate) stimulated G I P release in rats [40], but this was l ike ly indirect v ia stimulation of acid release, as indicated by inhibit ion with omeprazole; in another study, ID lactalbumin hydrolysate was capable of stimulating G I P secretion [74]. Duodenal acidification has been shown to stimulate G I P secretion in rats and humans in most [40; 64; 72], but not in all cases [66]. Contrary to the dose-dependent effect of HC1 on G I P 11 release, duodenal alkalinization with N a O H or perfusion with comparable ionic strength salt solutions were unable to change plasma G I P levels [64; 67; 70]. A m i n o acids alone also appear to stimulate hormone release from the K - c e l l directly [78; 83-85]. Infusion of a solution containing A r g , H i s , He, Leu , L y s , and Thr to the small intestine of human volunteers increased plasma G I P by about 7-fold, whereas an isotonic solution containing amino acids known to potently stimulate cholecystokinin ( C C K ) release potently (Met, Phe, Trp, and Va l ) was only able to cause a 2-fold rise in circulating G I P [83]. Hence, it appears that G I P is differentially released depending on the meal content, such that its biological effects act to maintain plasma nutrient homeostasis under the specific dietary conditions. 1.4.2.4 Feedback Regulation: Modulat ion of G I P release from small intestine has been studied with respect to hormonal and nutrient feedback, as wel l as neural input. Extensive study has been directed at the role of circulat ing glucose and insul in in attenuating G I P release by luminal nutrients; the close relationship between these parameters make conclusions drawn from these studies difficult to interpret [68; 86; 87]. In a study of insulinopenic type 1 diabetes mellitus patients, both glucose and insulin independently inhibited the G I P response to oral fat but not oral glucose [88]. In one study, Dryburgh et al [89] found intravenous infusion of C-peptide ablated G I P release in response to I D Intralipid emulsion, while insulin and glucose levels were unchanged by the fat perfusion, possibly indicating negative hormonal feedback from the endocrine pancreas. G I P is k n o w n to st imulate both gastric and pancreatic somatostatin release [50; 90-92] . Immunoneutralization of somatostatin augmented G I P release in vitro [48; 50], whereas somatostatin has been shown to inhibit G I P secretion in vivo and in vitro [50; 71]. A recent study has shown that inhibition of D P I V in dogs, thus increasing the biological half-life of G I P and the related incretin glucagon-like peptide-1 ( G L P - 1 ; discussed below), as wel l as other hormonal substrates of D P I V , resulted in diminished G I P secretion in dogs, implicating possible 12 autocrine, paracrine or endocrine feedback [93]. However , this may be mediated by enhanced somatostatin and/or insulin and C-peptide release. Other studies have examined the effects o f infusion of other hormones in vivo [64; 94-96], although the design o f these studies prevents dissecting influences from changes in blood glucose, insulin, or exogenous hormones on G I P release. Hence, it is important that studies such as these are duplicated using in vitro K - c e l l models and the vascularly perfused small intestine to establish primary versus secondary effects on G I P release. Mos t studies agree that there is little i f any evidence for modulation of K-ce l l s by the autonomic nervous system [63; 65; 69; 73], although gastrin-releasing peptide ( G R P ) has been shown to stimulated G I P release in vitro [48]. Calcitonin-gene related polypeptide ( C G R P ) may negatively regulate G I P release via the enteric nervous system [97]. Hence, G I P cells also express receptors for neurotransmitters, although the significance is unclear. 1.4.2.5 GIP Release in Vitro: Very few studies have been undertaken to examine G I P release in vitro, stemming from the diffuse distribution of enteroendocrine cells in the gut. Use of elutriation centrifugation of canine intestinal mucosa has been shown to allow enrichment of canine K-ce l l s to a sufficient degree to allow release studies [48]. In this manner, it was possible to demonstrate glucose- and GRP-st imulated G I P secretion, and the inhibitory effects of somatostatin. Furthermore, it was shown that cell depolarization effected G I P exocytosis, and that secretion could be modulated by signal transduction cascades involving intracellular calc ium or cyc l ic A M P [48]. The expense and difficulty of this method prompted the development of an alternative system for studying G I P secretion. Subcloning the intestinal endocrine tumour ce l l l ine, S T C - 1 , resulted in a derivative ce l l l ine, S T C 6-14, that contained ~30% I R - G I P containing cells and a s imilar percentage of somatostatin containing cells [50]. T o date, studies using these models have been l imited to those on the effect of glucose on G I P release in vitro and the role of somatostatin in control of G I P release. Further studies using these methodologies are necessary to clarify the 13 roles of other nutrients and hormones with respect to G I P secretion, particularly for non-glucose nutrients and hormonal modulation. 1.5 Biological Effects 1.5.1 Gastrointestinal and Pancreatic Effects 1.5.1.1 Enterogastrone Action: F o l l o w i n g Bay l i s s and Star l ing 's demonstration that the gastrointestinal tract was an endocrine organ [98], many studies were performed in attempts at explaining physiological phenomena with responses to isolated humoural factors from gut extracts. One such study was that of Kosaka and L i m [16], who coined \"enterogastrone\" as the name of the mucosal hormone secreted from the small intestine in response to fat and resulting in gastric acid inhibition. A s mentioned previously, peptide purification techniques allowed isolation, purification and testing of mucosal hormones, and several are candidates for enterogastrone activity; among candidates are peptide Y Y , secretin, C C K , glicentin/oxyntomodulin, and G I P [3; 99-102]. Recent progress in elucidation of G I P ' s enterogastrone action has fallen from popularity, primarily due to the question as to the physiological relevance of this function and the shifted emphasis towards the other biological functions of the hormone. Ear ly studies examining inhibit ion of acid secretion were confl ict ing, apparently due to differing methodologies employed. Original studies were performed in denervated Heidenhain gastric pouches [12-15], and when studies were duplicated in models employing intact innervated stomach, the potency of G I P as a physiological enterogastrone was difficult to demonstrate [103-106]. Clarification of these discrepant results came when the role of the vagus nerve was taken into account [107-109], although some studies have been able to show an effect of endogenous G I P immunoneutralization on acid secretion from the innervated stomach [64; 110]. Studies on the mode of action of G I P in the stomach suggest that somatostatin is the final 14 mediator of G I P ' s inhibitory ability [92; 107], although this pathway can be negatively regulated by acetylcholine and endogenous opioids [90; 111; 112]. G I P is one of the most potent stimuli for gastric somatostatin release, suggesting that the inhibitory activity of G I P on exogenous pentagastrin- and histamine-stimulated acid secretion was via paracrine somatostatin activity at the parietal ce l l , although G I P may also mediate paracrine regulation of gastrin-producing G -cells, histamine-releasing enterochromaffin cells and possibly chief cells [12-14; 90; 107; 113]. W i t h specific regard to gastrin, some studies have reported G I P to be a stimulant [92; 114; 115], while others report G I P to diminish gastrin release [109; 110; 116; 117]; regardless, the net inhibi t ion of acid secretion indicates that the effect of somatostatin is dominant. G i v e n the strong neural influence of the vagus on somatostatin release from gastric D-cel ls [90; 107; 109; 115; 118], the inconclusive results regarding the physiological importance of G I P in acid secretion are not surprising. Indeed, inhibit ion of GIP-stimulated somatostatin by exogenous acetylcholine and vagal stimulation was demonstrated, and it was suggested that in order for G I P to play a role in acid regulation, parasympathetic tone must be reduced, possibly involving a catecholamine mediated enterogastric reflex [107; 109]. Mos t recently, Rossowski et al [119] developed a potent somatostatin antagonist which , when administered intravenously to rats prepared with gastric fistulae, resulted in a potent blockade of the enterogastrone activity of G I P , confirming earlier studies. In all l ikel ihood, the enterogastrone effect proposed by Kosaka and L i m [16] is mediated by a number of gastrointestinal hormones with overlapping functions, only one of which is GIP . 1.5.1.2 Other GI Effects: Investigation of other possible actions of G I P on the gastrointestinal system have been extremely l imited. Prior to demonstration of expression in the salivary gland [26; 35], G I P was considered to act as a regulator of salivary duct electrolyte transport [120]. Similar ly , G I P was shown to be a local regulator of mucosal alkalinization in the proximal duodenum [121] and to 15 control water and electrolyte absorption in the jejunum [122; 123]. Early studies indicating G I P had a stimulatory effect on exocrine pancreatic acinar cells were later suggested to be due to contaminating C C K in natural source preparations of hormone [124; 125]. Later studies employing synthetic or highly purified peptide preparations indicated that G I P was capable of potentiatiating acetylcholine- and CCK-s t imu la t ed amylase excretion [126; 127], whereas it inhibited bombesin-stimulated amylase secretion [32j; it is thought that G I P ' s action on the exocrine pancreas may be a secondary effect mediated via insulin. The wel l designed studies of Cheeseman and co-workers indicated G I P upregulated phloridzin-insensitive hexose transport, thus augmenting absorption of sugars from the small intestine [128; 129]. G I P appears to have only weak effects on gastrointestinal motility during the digestive [130-132] and interdigestive phases [133; 134]. In some animals (but not humans) high concentrations of G I P stimulated hormone release from intestinal L-cel ls [45; 135-140]. Addit ional ly , G I P has also been shown to affect blood flow to the mesentery [132; 141-143]. A s with biological functions attributed to any hormone, the distinction between physiological and pharmacological effects must be made. Whi le the enterogastrone and incretin effects of G I P are we l l established, uncorroborated studies must be carefully scrutinized with respect to hormone concentration and experimental design in order to discriminate between physiological and pharmacological effects. Often the tendency is to administer larger hormone doses to convincingly (i.e. statistically) demonstrate an effect - this is a two edged sword that must be regarded with caution. 1.5.2 The Enteroinsular Axis & Incretin Concept The seminal articles demonstrating an augmented insul in response to nutrients when administered orally versus intravenously [144-147] prompted Unger and Eisentraut [148] to propose the \"Entero-insular A x i s , \" a concept later expanded by Creutzfeldt [80; 149] to describe 16 all gut derived input - direct substrate, neural and hormonal signals - on release of any hormone (insulin, glucagon, somatostatin, and pancreatic polypeptide) from the Islets of Langerhans on nutrient ingestion. Prior to the discovery of insulin, Moore , Edie and Abram attempted to purify a hormone from duodenal mucosa to treat type 1 diabetes [150]. After Banting and Best [151] succeeded in isolating insulin, the internal secretion of the pancreas, L a Barre termed mucosal hormones able to stimulate insulin secretion \"rincretine\" [152-154]. Currently, two criteria define an incretin: (1) it must be released by luminal nutrients, particularly carbohydrates, and (2) at physiological levels, it must stimulate insulin secretion in the presence of elevated blood glucose levels [80]. O f potential gastrointestinal hormones, G I P may be the only uncontested candidate as a true physiological incretin [155-157]; glucagon-like peptide-1 ( G L P - 1 ; discussed below) has potent antihyperglycemic effects, and has also been considered a potential incretin, however, recently it has been suggested that glucose-lowering effects attributed to this peptide may be primari ly derived from slowing gastric emptying and/or local activation of afferent neural fibres, rather than incretin activity by definition [158; 159]. The first indication that G I P had insulin stimulating ability came from the research group of Dupre . Impure preparations of C C K - P Z improved glucose tolerance when infused intravenously, however, results could not be duplicated with highly purified C C K - P Z [160; 161], much l ike the early findings of B r o w n and Pederson on gastric acid secretion. Hence, Dupre reasoned G I P was responsible for the insulinotropic action in impure C C K - P Z , and when this hypothesis was tested, it was shown to be correct [162]. Further human studies using an intravenous glucose infusion to maintain a hyperglycemic clamp, with a concurrent oral glucose load [163] or G I P infusion [164], unequivocally established G I P as a physiological incretin in man, and suggested that G I P could be responsible for the entire incretin-mediated insulin response. Importantly, the insulinotropic action of G I P was dependent upon prevailing glycemic conditions, such that G I P was unable to stimulate insulin during euglycemia, thus preventing the 17 risk o f hypoglycemia [82; 87; 162; 165]. Cons ider ing that fat is the most potent G I P secretagogue, the glucose-dependency of its insulinotropic activity is crucial . In the isolated vascularly perfused rat pancreas, in order for G I P to stimulate insulin release, a threshold glucose concentration of 5.5 m M was necessary, and maximum potentiation occurred with 16.7 m M glucose in the perfusate [165]. G I P ' s actions on the (3-cell were also concentration dependent [57; 165]; a linear peptide gradient in the perfused pancreas resulted in significant insulin release at concentrations as low as 70 p M [57], consistent wi th the physiological postprandial concentrations of G I P . The insulinotropic actions of G I P have been replicated using isolated islets of Langerhans [166-173], purified (3-cells [174], and clonal (3-cell models [175-182]. Confirmation of the physiological role of G I P as an incretin has been found using methodologies to ablate G I P signalling: immunoneutralization of G I P [64; 183; 184], immunoneutralization of the G I P receptor [185], injection of peptide antagonists of the G I P receptor [186-188], and generation of a transgenic mouse with a null mutation in the G I P receptor gene [189]. These studies generally indicate that 50-70% of the postprandial insul in response results from stimulation by GIP . The mode of action of G I P is mediated via specific activation of a cel l surface receptor present on the (3-cell. Receptor activation results in generation of second messengers cyc l ic A M P , intracellular ca lc ium flux, and arachidonic acid release (discussed below), wh ich ultimately results in greater insulin secretion. The dogma of (3-cell activation is that cellular glucose uptake and metabolism increases the cellular ratio of A T P : A D P , resulting in the closure of K + A T P channels and thus membrane depolarization; the opening of voltage-dependent calcium channels ( V D C C ) allows influx of C a 2 + , a requirement for granule exocytosis. Hence, activation of the G I P receptor must accelerate the stimulus-exocytosis cascade or recruit additional (un)responsive (3-cells to potentiate insulin release. Whi le the dependence of G I P ' s proximal intracellular signals on prevailing glucose conditions is unclear, G I P has been shown to decrease 18 conductance of ATP-sens i t ive K + channels responsible for (3-cell depolarization [190-193]. Al though electrophysiological studies were not able to show a direct action of G I P on calcium channels [190], other studies indicate G I P does augment influx of extracellular calcium via these channels [4; 176; 194]. Addi t ional ly , studies have shown that G I P is able to further stimulate exocytosis at a level distal to the rise in intracellular calcium [190; 191], as wel l as via a K + A T P channel independent pathway [195]. It has been proposed that the \"glucose sensor\" a l lowing incretin-induced insulin release only in the presence of elevated glucose may be simply the requirement of PKA-s t imula ted granule mobilization of a high A T P : A D P ratio [192]; notably, oxidative metabolic intermediates have also been shown to permit the insulinotropic activity of G I P [57; 168]. The |3-cell integrates various external stimuli, such that G I P and G L P - 1 have additive effects on insulin release [196-199], and G I P potentiates the glucose-dependent insulinotropic action of C C K , acetylcholine and arginine [200-204]. Cholinergic augmentation of GIP-stimulated insulin release has been implicated [200; 204-206], whereas somatostatin, enterostatin, galanin, V I P and islet amyloid polypeptide ( I A P P , amylin) were able to inhibit G I P ' s action on the (3-cell [201; 207-210]; pancreastatin may or may not play an inhibitory role [207; 211; 212]. In addition to enhancing insulin release, G I P further acts as an insulinotropic agent by stimulating proinsulin gene transcript ion and translation [213-216] and up-regulating p lasmalemmal glucose transporters and hexokinase in the |3-cell [216]. Islet hormone regulation by G I P is not l imited to the (3-cell. Early studies in the perfused rat and dog pancreas indicated that G I P stimulated release of glucagon, somatostatin, and pancreatic polypeptide [170; 202; 217; 218]. The effect of G I P on glucagon release was only under low glucose conditions and decreased upon increasing glucose concentration in the perfusate, consistent with the physiological release of glucagon [202; 217]. In purified rat a-cells, G I P and G L P - 1 were able to promote exocytosis v ia a cyc l ic AMP-dependent potentiation of calc ium 19 currents. U s i n g the same preparation, somatostatin was able to restrain the glucagonotropic effects of G I P and G L P - 1 , whereas insulin was without effect [219]. Whi le the ability of G I P to stimulate glucagon release has been duplicated in isolated perfused human cadaver pancreata [206], in c l in ica l studies, no effect of G I P on serum glucagon is observed [87; 164; 220-222]. Hence, it is l i ke ly that the somatostatinotropic activity o f G I P in the pancreas is more physiological ly relevant, resulting in suppression or no change in glucagon in humans during exogenous G I P infusion. 1.5.3 Effects on Nutrient Storage and Metabolism 1.5.3.1 Liver: Few studies have been performed regarding the b io logy o f G I P and the hepatocyte. Expression of the known G I P receptor m R N A present in other tissues has not been detected in hepatic tissue [223; 224], nor were G I P binding sites detected [175; 225; 226]. However , some work has indicated G I P does have biological effects on liver. A gut factor was considered to be responsible for a decrease in hepatic extraction of insulin [227-229], however, other studies suggested that G I P was unlikely to be the mediator [230]. Canine experiments using chronically implanted catheters and Doppler - f low probes confirmed that a gut factor reduced insul in extraction by the liver, but comparison of oral versus intraportal glucose administration indicated hepatic glucose uptake was not influenced by gut hormones; indeed, when co-infused with IV glucose, G I P had no influence on hepatic glucose uptake [231]. In isolated cultured rat hepatocytes, G I P acted in a catabolic fashion, stimulating glycogenolysis and a slight but significant increase in gluconeogenesis. Somatostatin was able to reduce these glucagon-like effects on the l iver cells [232]. In the perfused rat liver, G I P acted anabolically in concert with insul in to potently suppress hepatic glucose production induced by glucagon [233; 234]. Similar ly , in dog and human euglycemic clamp experiments, G I P acted to inhibit hepatic glucose 20 production only in the presence of insulin [235-237]. Studies examining both G I P receptor expression and biological effects in l iver are required to clarify the hormone's action on this tissue. 1.5.3.2 Skeletal Muscle: One research group reported the ability to detect specific binding sites for G f P on striated muscle [226], whereas others have failed [225]. W h i l e this tissue was not tested for receptor expression using molecular cloning techniques, muscle has been considered to be a potential target organ for G I P in a l imi ted number of experiments. Prel iminary studies aimed at establishing whether the glucose lowering effects of G I P were entirely mediated by insulin or whether G I P had insulin-independent actions were in i t ia l ly negative. G I P was unable to stimulate glucose uptake in the rat hemidiaphragm preparation [238]. In a more recent systematic approach to the subject, O'Harte and co-workers found native G I P alone to be nearly as potent as equimolar insulin in stimulating glucose uptake, glucose oxidation, glycogenesis, and lactate formation in mouse abdominal muscle slices, and effects on these parameters were exerted at physiological concentrations of hormone [239]. 1.5.3.3 Adipose Tissue: A s fat is the most potent G I P secretagogue, many studies have been performed examining effects of G I P on fat metabolism. Specific G I P receptors have been demonstrated on adipocytes by a number of techniques [11; 240; 241]. In rats, immunoneutralization of endogenous G I P caused an increase in plasma triglycerides and exogenous G I P resulted in the opposite effect after intra-duodenal fat [234; 242]. However, in studies designed to examine G I P ' s effect on an intravenous fat tolerance test with a commercial fat emulsion (Intralipid), it was concluded that G I P had no effect on plasma triglycerides in man or dog [243; 244]. In a unique approach, Wasada et al [245] collected chyle from thoracic duct fistulae in donor dogs fed a fatty meal. When chyle was infused IV with or without exogenous G I P in recipient dogs, the rise in plasma 21 triglycerides was significantly lower when G I P was co-infused. Thus chylomicron formation may not only be necessary for fat-stimulated G I P release [80; 81], but also for GIP-stimulated triglyceride removal from the circulation. In this regard, G I P has been linked to stimulation of lipoprotein lipase activity and incorporation of fatty acids in 3T3-L1 mouse preadipocytes and rat epididymal fat pad explants, acting in synergy with insulin [246-251]. G I P has also been causally related to de novo l ipogenesis, acting with insul in to increase glucose uptake in adipocytes and stimulating fatty acid synthesis (incorporation of glucose and acetate into extractable lipids) [251-254]. The role of G I P in l ipolysis is currently unclear: it has been reported to exert weak or no l ipoly t ic effects on its own in some reports on isolated rat adipocytes [253; 255], whereas other studies using 3T3-L1 cells have found G I P to exert l ipolytic effects v ia a cycl ic A M P mediated pathway similar to adrenergic agonists [241]. Ebert and Creutzfeldt [234] reported that alone, G I P was equally l ipolytic as glucagon in isolated rat fat cells, but G I P dose dependently inhibited glucagon-stimulated l ipolysis with a concurrent reduction of cel lular c A M P ; the same study showed similar effects on secretin-stimulated l ipolysis , albeit to a lesser degree. In the study by Mcin tosh et al [241], insulin was shown to block G I P - and isoproterenol-stimulated lipolysis, whereas in earlier studies [253; 255], G I P was found to attenuate the l ipolyt ic effects of glucagon and isoproterenol, but not those of V I P or secretin. The mode of G I P ' s inhibitory activity is not certain as most of its actions are thought to occur through the cycl ic A M P pathway, the same pathway that glucagon, V I P , secretin and isoproterenol also exert their effects on the adipocyte [234; 241; 253; 255]. It was suggested that G I P may directly compete for glucagon binding to fat cells [234; 255], however, the ligand specificity of the cloned glucagon and G I P receptors does not support this contention. It is hypothesized that G I P has a dual action on adipocytes depending on the fed state of the animal: during fasting, G I P is l ipolytic and during the fed state, G I P acts to promote nutrient storage via 22 lipogenesis, both dependent upon ambient circulating insulin levels. In this case, methodological differences may help explain the contradictory results. 1.5.4 Other Attributed Biological Functions The G I P receptor has been localized to many tissues of the body, for which no characterized function of G I P has yet to be described. This section w i l l be limited to tissues and organs where b io logica l effects have been described. Central injection (1CV) o f G I P dose- and time-dependently inhibi ted plasma fo l l i c le - s t imula t ing hormone ( F S H ) concentration wh i l e stimulating growth hormone ( G H ) release in ovariectomized rats (and not affecting luteinizing hormone ( L H ) , prolactin or thyroid stimulating hormone); in the same study, G I P injection IV was without effect on any pituitary hormone measured [256]. Peripheral (SC) administration of G I P to anesthetized male rats dose-dependently decreased G H levels [257]. In primary dispersed cultured anterior pituitary cells, GIP-stimulated both F S H and L H accumulation in the culture medium [256]. G iven the different experimental designs, it is difficult to determine how and where G I P was acting to modify pituitary hormone secretion. The significance of exogenous G I P effects on pituitary hormones is presently unclear; G I P receptors have been wel l described in brain [11; 225], however, the receptors are localized in areas inaccessible to blood borne hormone, and neither G I P m R N A nor G I P peptide have been found in brain [4]. G I P has been reported to regulate blood flow in gut vascular beds [132; 141-143], and functional receptors have been described in endothelial cel ls; it has been proposed that G I P ' s activity on the vasculature may be present to maximize nutrient absorption [11; 258; 259]. G I P stimulated Cortisol secretion in a type of food-dependent Cushing's syndrome has been wel l characterized [260; 261]. The G I P receptor was reported to be expressed in the rat adrenal gland [11], but studies on the human disease found G I P receptor expression l imited to the cortical adenoma responsible for the condit ion, but not in normal adrenal cells [260; 262]. A recent study 23 examining the corticosterone response to bolus IP GIP injection found it to be a stimulant in normal rats [263], possibly indicating species differences. Recently, the unique hypothesis of an 'entero-osseous axis' has been proposed, whereby GIP has been shown to have osteotropic effects on bone osteoblast cells via cyclic AMP and calcium second messenger cascades coupled to the GIP receptor; furthermore, chronic GIP treatment in ovariectomized rats enhanced bone density in vivo [259; 264]. Use of GIP receptor knockout mice or receptor antagonists have yet to confirm most extra-intestinopancreatic effects of GIP. 1.6 GIP Binding Sites 1.6.1 GIP Iodination and Binding Studies The development of an iodination protocol for purified porcine (p)GIP,_42, using the chloramine-T method with subsequent high-pressure liquid chromatography (HPLC) purification, allowed proof of the existence of specific GIP receptors present on the pancreatic (3-cell [175; 265; 266]. Fractions collected from two peaks identified on the HPLC chromatogram were found to have similar binding ability to hamster insulinoma membranes and insulin-releasing ability from the perfused rat pancreas, and preliminary evidence suggested both were iodinated on Tyr10 [265; 266]. Comparison of porcine and bovine GIP sequences (Table 1, Introduction section 1.3.1) indicate potential iodination of residues Tyr1 or Tyr10 (or possibly His38); iodination of bovine GIP4_42 (lacking Tyr') resulted in only one biologically active iodinated product, presumably labelled on Tyr10 [267]. Later studies confirmed [mono-[125I]iodo-Tyr'°]-pGIP|_42 to be the preferred biologically active iodinated GIP form to use in experimentation [4; 59; 182]. Iodination of synthetic human GIP using the same protocol results in a more complex HPLC profile - specific peaks are biologically active, but their molecular identities have not been characterized (unpublished observations). 24 T o date, G I P receptor binding sites have been limited to brain and specific in vitro cell types (transformed ce l l models of various tissues) or receptor transfected cells . A l though only published in abstract form, Whitcomb et al [226] provided evidence for specific binding sites for G I P in pancreas, the glandular portion of the stomach, the duodenum, jejunum, ileum, colon and various muscle groups, using a novel in vivo radioreceptor assay for the anesthetized rat described elsewhere in detail [268; 269]. In the same abstract, it was reported that specific receptors were not found in liver, adrenal gland, submandibular gland, spleen, kidney, testis, epididymus, prostate or seminal vesicles. In a later autoradiographic survey of rat tissues it was only possible to detect specific binding in regions of the brain, including cerebral cortex, anterior olfactory nucleus, lateral septal nucleus, subiculum, inferior coll iculus, and inferior ol ive of the medulla oblongata, however, saturable binding sites were not detected in spinal cord, pituitary, stomach, small intestine colon, pancreas, liver, heart or skeletal muscle [225]. Similar ly , the first studies employing monocomponent iodinated G I P failed to detect specific binding to various tissue membranes, including kidney cortex, l iver, brain, adrenal cortex, adipose tissue and intestinal epithelium, but could measure binding to (transformed) hamster (3-cell membranes [175; 266]. A specific analysis of the adrenal gland using autoradiography resulted in the detection of binding in the inner cortical layers [263]. Destruction of receptors during tissue preparation and degradation of iodinated ligand are possible explanations for the mixed success of these studies. Binding experiments have since demonstrated receptors on whole isolated islets and purified (3-cells [3; 174], transplantable hamster insulinoma membranes [21; 266; 267; 270; 271], cultured insulinoma/transformed cells: In 11.1 [175; 272], (3TC-3 [178], human insulinoma [182], R I N m 5 F cells [198; 273; 274], B R I N - D 1 1 and INS-1 cells [275], differentiated 3T3-L1 preadipocytes [240; 241], and human osteoblast-like cells: SaOS2 and M G 6 3 [264]. Scatchard analysis of binding competition curves on hamster and human insulinoma cells and transformed hamster (3-cells indicated a two site model with high-capacity/low-affinity and low-25 capacity/high-affinity binding sites; ( K D high affinity: 0.2-7 n M , low affinity: 0.039-8.93 u M ; Table 2) [175; 182; 266; 272]. Membrane binding dissociation studies, by dilution with and without added cold G I P , demonstrated disassociation constants ( K o f f ) to be 5.3 X 1 0 3 min 1 for high affinity sites and 2.3 X 10\"2 min\" 1 for low affinity sites [272]. Scatchard analysis of autoradiographys of brain slices identified a single high affinity binding site, K D = 16-62 p M [225]; however, because of the associated inaccuracies with calculating K D values by Scatchard analysis compared to direct curvilinear regression analysis of saturation binding curves, these data should be interpreted with caution. Saturation binding has been performed on insulinoma cells ((3TC-3 and INS-1 cells), differentiated 3T3-L1 mouse adipocyte cells, and osteoblast-like cells to accurately measure binding constants (Table 2). Optimization of G I P binding study parameters showed an increase in binding capacity with decreasing temperature. Th is correlat ion also reduced l igand and membrane receptor degradation, but required a longer incubation time to reach equilibrium. Optimal p H for binding was between 6.5 and 8.0; C a C l 2 (5 m M ) completely blocked receptor binding, whereas binding was moderately reduced by hypertonic salts (1-2 M N a C l or KC1) , but enhanced by inclusion of 5 m M M g C l 2 or M n C l 2 in the binding buffer. Binding of l 2 5 I - G I P was specifically displaced only by G I P , and not glucagon, insul in , V I P , secretin, C C K - 8 , G H R H , P H I or E G F [266; 272]. Further biochemical analysis of the G I P receptor by covalent cross-l inking indicated it was a monomeric glycoprotein with an electrophoretic mobility corresponding to a predicted receptor mass of 59 K D a , however, incubation with dithiothreitol proved that intrachain disulfide bonds existed and resulted in a slower mobility, predicting a receptor mass of 68 K D a ; disulfide bonds were not required for binding of l 2 5 I - G I P to its receptor, however. Elu t ion of lectin bound membranes from wheat germ agglutinin and concanavalin A suggested glycosylat ion with o l i g o s a c c h a r i d e moie t ies p o s s i b l y c o n t a i n i n g N - a c e t y l g l u c o s a m i n e , me thy l -c i -D-mannopyranoside and/or sialic acid [270; 271]. 26 Fo l lowing the cloning of the rat G I P receptor, hamster and human isoforms were isolated a l lowing study of the G I P receptor using transfected cell models (described below). B ind ing analyses were performed to confirm the specificity of G I P binding sites, and compare the binding affinities measured using various G I P labels and/or G I P preparations. In general, all studies identified a single class of specific G I P binding sites, with K D values (from Scatchard analysis or saturation binding) in the range 180 p M to 19.3 n M (Table 2). Studies have examined transfected rat [276], human [277; 278] and hamster [223] G I P receptors expressed in C H O - K 1 [223; 276], C O S - 7 [276] or C H L [277; 278] cells, using either iodinated synthetic porcine or human G I P label (Table 2). Only two studies employed saturation binding isotherms to measure K D values more accurately, both using synthetic pGIP,_ 4 2 monocomponent tracer, but in one case measuring whole cell binding to human G I P R transfected C H L cells [277], and in the other, whole C H O or C O S - 7 cells, or C H O membrane preparations from rat receptor transfected cells [276] (Table 2). Whi l e methodological differences (time, temperature, receptor isoform, cell type, expression level , label concentration, molecular species o f label, competing ligand) certainly must be considered, in properly designed experiments using transfected cells with synthetic human or pork G f P , . 4 2 , I C 5 0 values should be in the low n M range, and accurate determinations of K D values to be in the high p M range. The remarkable binding specificity of the transfected G I P receptor was demonstrated in all studies: displacement of G I P label by high concentrations (1 u M ) of glucagon, G L P - 1 , G L P - 2 , V I P , secretin, P A C A P - 2 7 , P A C A P - 3 8 , or P H I could not be detected - only the G i l a monster venom peptides exendin-4 and exendin-(9-39) were found to displace significant amounts of specific 1 2 T - G I P binding [223; 276-278]. In an analysis of hybrid G I P / G L P - 1 receptor chimeras, it was shown that only the N-terminal 132 amino acids of the G I P receptor (Figure 2, section 1.6.2) were necessary for high affinity l 2 5 I - G I P binding, however, this was enhanced i f the length was extended to 151 ( N T and T M 1 ) or 222 (NT , T M 1 , 27 I d , T M 2 , E C 1 , and T M 3 ) amino acids [279; 280]. Hence it was concluded that the primary binding determinants were within the extracellular N-terminus of the G I P receptor, however additional contact sites may be present in the \"binding pocket\" wi th in the transmembrane domains, and possibly in the first extracellular loop. The latter possibility may be supported by the reduced functionality of a naturally occurring mutant human G I P receptor bearing a missense mutation ( G l y l 9 8 C y s ) in the first extracellular loop [281]. Table 2: Comparison of binding constants measured using transformed cells and transfected cell models. Cell Line Receptor Iodinated Peptide Binding Constants Ref. Insulinoma Hamster pGIP,. 4 , K D a 2.05 n M (high), 39 n M (low) 1 In 111 Hamster pGIP,. 4 2 K D a 7 nM (high), 0.8 u M (low) 2 Insulinoma Human pGIP,. 4 2 K Da 223 p M (high), 8.93 u M (low) 3 Brain slices Rat p G I P M 2 K Da 16-62 p M 4 |3TC-3 cells Mouse p G I P M 2 K D b 277 p M 5 INS-1 cells Rat PGIP,.42 K Db 5 3 1 p M 5 3T3-L1 cells Mouse pGIP,. 4 2 K Db 46 p M 6 SaOS2 cells Human hGIP,. 4 2 K D b 362 p M 7 MG63 cells Human hGIP,. 4 2 K D b 320 p M 7 CHO-K1 cells' Hamster h G I P M 2 I C 5 0 9.6 n M K D a 18.2 n M 8 C H L cells c Human p G I P w „ K Da 11.3 nM (vs. pGIP,. 3 ( 1) K D a 19.3 nM (vs. h G I P M 2 ) 9 C H L cells1' Human (short)11 PGIP,.42 K Db 180 p M 10 C H L cells1' Human (long)d pGIP,. 4 2 K D \" 650 p M 10 CHO-K1 cells' Rat Islet PGIP,.42 I C 5 0 7.6 nM (vs. p G I P M 2 ) IC, 0 8.9 nM (vs. h G I P M 2 ) 11 COS-7 cells' Rat Islet PGIP,.42 IC 5 0 2.6-3.7 n M (vs. several preparations of h G I P M 2 and p G I P M 2 ) 11 CHO-K1 cells' Rat Islet PGIP,. 42 K Db 204 p M (whole cell) K D b 334 p M (cell membranes) 11 a : K D values calculated from competitive binding studies by the Scatchard method. b : K D values accurately measured from saturation binding analysis. ': Transfected cell model d : The short form corresponds to the other published human GIPR sequences, and better aligns with other species isoforms, the long form contains a 27 amino acid C-terminal insertion due to incomplete m R N A splicing. References: (1) [266], (2) [175], (3) [182], (4) [225], (5) J. Ehses, unpublished, (6) S. Hinke, unpublished, (7) [264], (8) [223], (9) [278], (10) [277], (11) [276]. Cel l lines: In 111: hamster insulinoma, [3TC-3: mouse insulinoma, INS-1, Rat insulinoma, 3T3-L1 cells, mouse adipocyte, SaOS2 and M G 6 3 cells, human osteoblast-like cells, C H O - K 1 cells, Chinese hamster ovary cells, C H L cells, Chinese hamster lung cells, COS-7 cells, green monkey kidney fibroblast cells. 28 1.6.2 Cloning the GIP Receptor, Gene Expression & mRNA Distribution In 1993, Usdin et al described the cloning of the rat GIP receptor from a rat cerebral cortex cDNA library [11]. Initially, they used degenerate primers based on sequences in the third and seventh transmembrane regions of the secretin and PTH receptors to isolate a unique 507 bp PCR fragment that was used as a probe to screen cDNA libraries from rat brain and RINm5F cells. Upon isolation of the complete receptor, sequence analysis revealed a potential 52,223 Da protein made up of 455 amino acids with substantial similarity to cloned receptors from the secretin-VIP family (Figure 2); the likely cleavage of an 18 amino acid signal peptide would result in a protein with a mass of 50,063 Da. However, consensus sequences for N-linked glycosylation were also found at amino acids 59, 69, and 74 of the external N-terminal region, and an additional site in the first extracellular loop, confirming earlier biochemical analyses indicating that the GIPR was a glycoprotein, and suggesting that the molecular weight was likely substantially greater than 50 KDa [11]. Comparison to structurally homologous receptors indicated that greatest sequence similarity was found with the glucagon (~44%) and the GLP-1 (~40%) receptors from rat, and that the GIP receptor belongs to the Family B class of G-protein coupled receptors which contains all GPCRs for small peptide ligands. Not long after Usdin's initial report, papers were published describing the cloned GIP receptors for human pancreatic islet [277], human insulinoma [278], human insulinoma/lung [282], as well as hamster insulinoma [223] and isolated rat islets [276]. Comparison of interspecies sequence differences shows ~86% identity between rat and hamster receptors, and the human to be 79-81% identical to both rodent receptors. Hydropathy plots suggest the presence of seven possible transmembrane domains, characteristic of G-protein coupled receptors (Figure 2). Use of molecular biology techniques has allowed testing of different tissues for the cloned GIP receptor, particularly those where receptor transcript abundance might be low. Northern (mRNA) blotting revealed receptor transcripts of 5.5, 3.8 and 2.2 Kb in length from RNA 29 prepared from rat telencephalon and R I N m 5 F cells, but m R N A degradation precluded detection in pancreatic tissue [11]. In the fol lowing year, Yasuda et al [223] was able to detect a 3.8 K b transcript in rat islets and HIT-T15 cells, but not other hamster tissues. V o l z et al was able to detect the 5.5 K b transcript in human insulinoma, but not stomach carcinoma (HGT-1) or colon m R N A , whereas M o e n s et al detected bands corresponding to G I P receptor m R N A in insulinoma, glucagonoma, isolated islets, as we l l as purified (3-cells and purified non-(3-cells consisting of > 80% ct-cells, but none in lung, stomach, intestine, l iver or brain [224; 278]. Us ing in situ hybridization in rat embryos, G I P receptor expression was detected in pancreas, cardiac endothelium, blood vessel endothelium, lung endothelium, inner adrenal cortex, adipose, stomach and intestine, but not spleen or liver [11]. Us ing P C R based methodologies, the rat G I P receptor has been detected in pancreas, stomach, duodenum, proximal small intestine, fat, adrenal and pituitary, and brain regions: telencephalon, diencephalon, brain stem and cerebellum, and particularly abundant in the olfactory bulb, however not in l iver or spleen [11]. A more recent publication examining human tissues reported G I P receptor in weak amplicon abundance in liver and adrenal cortex, as wel l as in cultured pancreatic islets and adipose, but not in muscle or pancreatic ductal tissue [283]. Whi le many of these tissues exhibit characterized responses to G I P , several tissues were never considered as targets for G I P , thus a l lowing other avenues of research on G I P physiology. The genomic structure of the G I P receptor has been described for both rat and human [282; 284]. The intron/exon pattern is very similar, with the exception of one additional exon for the rat - the human gene spans 13.8 K b on chromosome 19q (bands 13.2-13.3) comprised of 14 exons, 13 of which encode receptor protein; splicing out intronic sequence yields a 1389 bp open reading frame ( O R F ) predicting a 466 amino acid protein [277; 282]. The rat receptor gene is only 10.2 K b in length, but when spliced, yields the 1365 O R F / 4 5 5 amino acid receptor predicted by the cloned c D N A sequences [11; 276; 284]. Splice variants have been detected in 30 various tissues, however, most encode non-functional/truncated receptors [277; 278; 283; 284]. Comprehensive functional analysis of the G I P receptor promoter has yet to be published, however, a preliminary report of sequence analysis of the rat G I P receptor gene suggests the presence of three SP-1 sites, and one each of O C T - 1 , AP-1 and C R E sites, which may regulate G I P receptor expression v i a transcription factor binding. N o T A T A box was identified immediately upstream of the transcriptional initiation site, however distal T A T A (-1001 bp) and C A A T (-1071 bp) motifs were found, although R N A s e protection assays suggested that they may not be used to regulate expression. Dr iv ing luciferase enzyme expression by various lengths of G I P receptor 5'upstream promoter sequence showed basal expression in R I N 3 8 cells was p r imar i ly conferred by p rox imal sequences between -100 and +1 bp, relative to the transcriptional start site; in contrast, a sequence upstream of -181 bp resulted in cel l specific transgene expression [284]. Recently, functional studies in clonal |3-cells (INS-1 and B R I N -D l 1), L y n n et al (In press) were able to show regulation of G f P receptor expression by fat and glucose, and found supportive evidence that expression level was dependent on the transcription factor PPARc t . 31 Figure 2: Schematic of the two dimensional topography of the rat G I P receptor A m i n o acid numbers of putative transmembrane domain junctions are indicated, as wel l as potential N - l i n k e d glycosylation sites (N*) and disulfide bridge [11; 276]. C O O H - t e r m i n a l residues mutated in the current thesis are indicated by the symbols. 32 1.7 GIP Receptor Signal Transduction A preliminary report by Frandsen and M o o d y [285] demonstrated GIP-stimulated adenylyl cyclcase activity in mouse islet membrane preparations, however, the first complete study to show activation of intracellular messenger cascades in response to G I P was that of Szecowka et al [170] using whole isolated rat islets. In this paper, they described the correlation between glucose and G I P with tritiated cyc l ic A M P accumulation and insulin release. Under elevated glycemic conditions (> 6.7 m M ) , 30 n M G I P produced significant effects on both c A M P and insul in , whereas no effect was observed under low glucose (3.3 m M ) conditions [170]. A duplication of this study found that inclusion of the phosphodiesterase inhibitor, f B M X , allowed detection of c A M P responses to G I P at concentrations as low as 200 p M , which falls within the normal physiological range of the hormone [172]. The first report using a homogeneous cel l system was the study of Gespach, E m a m i and Rossel in [286] employing the human gastric cancer cell line, H G T - 1 . C y c l i c A M P stimulation by G I P in these cells was found to be dose-and time-dependent, with plateau c A M P levels achieved within 10-15 min (in the presence of I B M X ) , and half-maximal stimulation achieved at 8 n M G I P ; under these conditions, G I P was able to stimulate a maximal 5-fold stimulation of c A M P levels over basal [286]. A s G I P was unable to directly stimulate cycl ic A M P production in isolated antral and fundic glands [287], Gespach and colleagues hypothesized that G I P ' s acid inhibitory effect was mediated by cycl ic AMP-dependent stimulation of somatostatin release [286; 288], consistent wi th the current understanding of the hormone's action. A t the same time, researchers from the same institute published a report of GIP-stimulated cycl ic A M P formation in the transformed hamster (3-cell model, In 111. In these cells, G I P was similarly dose- and time-dependent, with plateau cycl ic A M P levels (+ I B M X ) achieved after 20 min, and half-maximal c A M P production occurring at 30 n M G I P ; a significant rise above basal c A M P levels was detected at 0.3 n M peptide, and 33 maximal GlP-st imuiated c A M P accumulation was 4-times basal [175]. In light of these results, the G I P receptor was proposed to be coupled to the stimulatory \"alpha subunit of the guanyl regulatory protein\" (GTP-b ind ing protein subunit G a s ) to transduce receptor activation upon ligand binding to stimulation of adenylyl cyclase [289], although this hypothesis has not been confirmed directly. G I P receptor mutagenesis studies have suggested that putative G-protein coupl ing occurs v ia specific residues in the 3 l d intracellular loop (IC3) and the intracellular COOH-te rmina l tail [290-292]. C y c l i c A M P is currently considered to be the primary intracellular s ignal l ing pathway through which G I P mediates most of its biological actions [4], and in general, G I P action is mimicked by c A M P raising agents such as forskolin. GIP-stimulation of cycl ic A M P production has been detected in many cel l models, inc luding isolated (3- and a - cells [224], human insulinoma [182], HIT-T15 [176], (3TC-3 [214], R I N [198; 214; 273; 274], INS-1 and B R I N -D l l (F. L y n n and J . Ehses, unpublished observations), as we l l as differentiated 3T3-L1 preadipocytes [240; 241], and SaOS2 osteoblast-like cells [264]. M a n y of the actions of G I P , including its stimulatory action on exocytosis, may partially or whol ly result from activation of c y c l i c - A M P dependent protein kinase ( P K A ) , as indicated by use of P K A inhibitors Rp-8 -Br -c A M P S or H 8 9 [136; 190; 191; 219; 263]. M o r e recently, high concentrations of G I P were shown to affect the cellular distribution of P K A in a specific endothelial cell line [258], and G I P activated the P K A / C R E B signalling module in insulin secreting INS-1 cells at physiological levels [293]. C y c l i c A M P stimulation by G I P was shown in all cell lines transfected with the cloned G I P receptors from all species [11; 223; 276-278], and with normal E C 5 0 values ranging between 0.3 and 0.9 n M [11; 276; 277]. N o c A M P production was detected in response to any peptide hormones other than G I P , but the venom peptide exendin-4 produced very small responses at high concentrations [11; 223; 276-278]. It should be noted that there are P K A independent activities of cycl ic A M P and G I P is able to activate other signalling cascades, and 34 thus P K A likely does not mediate all actions of GIP . Recently GIP-stimulated increases in cycl ic A M P were reported to activate c A M P binding protein G E F I I - R i m 2 complexes as a means of augmenting (3-cell insulin release by a P K A independent mechanism [294]. Early studies on islets indicated that G I P is not involved in phosphoinositol hydrolysis [295], and this finding was later confirmed with the cloned hamster receptor transfected in C H O cells [223]. In more recent work, Ehses et al [296] were able to demonstrate dose- and time-dependent GIP-stimulated release of arachidonic acid from (3TC-3 and receptor transfected C H O cells, l ike ly v ia G(3y subunit activation of Ca 2 +-independent phospholipase A 2 , confirming an earlier hypothesis [297]. Mos t evidence indicates that cycl ic A M P / P K A activation by G I P is not affected by prevai l ing glucose conditions. However , with ion currents in glucose sensitive excitable cells, elevated glycemia is necessary for closure of K + A T P channels and subsequent opening of V D C C s , although electrophysiological maneuvers may allow glucose-independent current movement [176; 190; 191; 193; 194; 219]. In the first comprehensive examination of G I P induced intracellular calcium ([Ca 2 +]|) fluxes, using the calcium sensitive fluorophore fura-2 in H I T - T 1 5 insulinoma cells, glucose alone modulated [ C a 2 + ] h and, in order to exert a dose-dependent effect ( E C 5 0 = 0.2 n M ) , G I P required the presence of glucose; addition of extracellular E G T A completely blocked G I P ' s ability to increase [Ca 2 + ] ; and involvement of the V D C C was confirmed by use of the specific blocker nimodipine. In the presence of nimodipine, A V P was stil l able to mobi l ize intracellular stores of calcium, whereas G I P could not [176]. Fo l lowing cloning of the rat G I P receptor by Usdin et al, they showed receptor activation was able to cause an increase in calc ium entry, when transfected into reporter H E K - 2 9 3 cells co-expressing apo-aequorin [11], however, later studies on the human receptor in C H L fibroblasts were unable to detect a calc ium signal in response to G I P [277; 278]. Us ing rat G I P receptors transfected in monkey kidney C O S - 7 cells , GIP-stimulated [Ca 2 + ] ; was shown to originate pr imari ly from mobil izat ion of intracellular pools, but also via a reversible plasma membrane calcium flux not 35 mediated by V D C C [276]. It is presently unclear whether differences in calcium signalling are receptor (species isoform) or cell type specific. One of the frontiers of G I P research is elucidation of pleiotropic signal transduction cascades. G I P has recently been suggested to regulate protein kinase modules other than the prototypical c A M P / P K A pathway. Although results are l imited, Straub and Sharp [298] reported inhibition of GIP-st imulated insulin release from H I T - T 1 5 cells by wortmannin, a relatively selective inhibitor of phosphatidylinositol 3-kinase (PI3K). The fol lowing year, G I P was shown to time-and dose-dependently activate mitogen activated protein kinase ( M A P K ; peak at 10 min, E C 5 0 = ~590 p M ) , a key enzyme in the regulation of cell proliferation, differentiation and metabolism, by both wortmannin-sensitive and -insensitive pathways in human receptor transfected C H O cells [299]. Us ing broad screening approaches, Triimper et al [293] and Ehses et al [300] have been able to show G I P coupl ing to pleiotropic s ignal l ing modules including c A M P / P K A , M A P K , P I 3 K / P K B as wel l as upstream and downstream effectors of these modules. These prel iminary findings may have important consequences for G I P ' s role in 6-cel l survival , proliferation, and apoptosis [301; 302]. 1.8 The Proglucagon Gene Products and Receptors Upon the first isolation and sequencing of proglucagon m R N A from vertebrates (anglerfish), it was apparent that the transcript encoded additional glucagon-like peptides within the same open reading frame [303-305]. When mammalian proglucagon m R N A s were subsequently sequenced, it was revealed that in addition to glucagon, two glucagon-like peptides, G L P - 1 and G L P - 2 , were also contained in the transcript [306-310]. It was clear from c loning and hybridizat ion studies that the same m R N A transcript gives rise to both pancreatic and gut glucagons. Subsequent work on posttranslational processing revealed that tissue specific cleavage by prohormone convertases results in formation of G L P - 1 l 7 . 3 6 N H 2 j / [ 7 . 3 7 J and G L P - 2 , along 36 with glicentin and/or glucagon-related polypeptide ( G R P P ) and oxyntomodulin, in intestinal L -cells and brain, whereas in pancreatic a-cel ls , G R P P , glucagon, and the major proglucagon fragment are generated (Figure 3) [100]. In contrast to glucagon, the primary counter-regulatory hormone responsible for maintaining normoglycemia in the fasted state by promoting gluconeogenesis and glycogenolysis [99; 311], based on sequence homology to G I P , G L P - 1 was predicted to have insulin releasing effects as a potential incretin [303; 304]. N o w , it is generally accepted that G L P - 1 acts as a glucose lowering hormone in vivo. Al though it has been questioned whether G L P - 1 acts as an incretin by definition (see section 1.5.2), it may act to combat g lycemic excursions by promoting enteroinsular-neural reflexes, slowing gastric emptying, and acting as a satiety factor in the brain [100; 155; 158; 159; 312]. G L P - 2 , in contrast, appears not to be involved in regulation of glucose homeostasis, but rather acts as a potent intestinotrophic factor to maintain bowel mass [312]. The G-protein coupled receptors for glucagon, G L P - 1 and G L P - 2 have all been cloned from various species, and shown to be coupled to cycl ic A M P production, inositol-trisphosphate (IP 3) generation and ca lc ium mobi l iza t ion [8-10; 176; 313-323]. Hence , it appears that biological divergence after gene duplications resulting in the glucagon-like peptides and their receptors, has al lowed unique tissue specific effects mediated by common signal transduction cascades. 37 P R O G L U C A G O N 61 64 69 72 7H KIN III 123 126 | GLUCAGON j IF-1 j K K K H K K 3132 6263 7071 J\"''2! GLP-2 j-C K K K K 1IW1J0 124125 P A N C R E A S MAJOR PROGLUCAGON FRAGMENT | I N T E S T I N E and B R A I N K K 159 160 G R P P , . , 0 Glucagon-,.,^, IP-164-M M P G F , , . „ G R P P , 3 ( ) Glicentin, 6 9 Oxyntomodulin 3 3 6 9 G L P - 1 7 8 . 1 0 8 1P-2 G L P - 2 i . . . Figure 3: Post-translational processing of the proglucagon gene products Prohormone convertase cleavage at specific pairs of basic residues ( K , R ) results in tissue specific production of bioactive hormones. In the pancreas, glicentin-related pancreatic polypeptide ( G R P P ) , glucagon, intervening peptide-1 (IP-1) and the major proglucagon fragment ( M P G F ) are generated. In intestine and brain, G R P P , glicentin, oxyntomodulin, glucagon-like peptide-1 ( G L P - 1 7 . 3 6 N H 2 / 7 . 3 7 ) , IP-2, and glucagon-like peptide-2 ( G L P - 2 ) are produced. Adapted from [100]. 38 1.9 Thesis Investigation In summary, the peptide G I P is generally considered to be an anabolic hormone, acting to promote nutrient storage and utilization by insulin-dependent and -independent actions, as wel l as acting as a gut-derived gastric acid inhibitory factor. G I P ' s role as an integrative hormone regulating the postprandial handling of nutrients is exemplified by the explicit requirements for its release, and the existence of specific receptors on target tissues. The cloning and functional expression of the G I P receptor has provided the opportunity to examine G I P and its receptor at the molecular level in a relatively isolated experimental model. W i t h this means to an end, one of the primary goals of the current work was to compare the cellular regulation of G I P receptor function in clonal (3-cells and receptor transfected cells, and to characterize the process of homologous receptor desensitization. The second aim herein was to examine the structural requirements of the ligand necessary for G I P receptor activation. F ina l ly , modulation of G I P , GLP -1 and glucagon bioactivity by the ubiquitous serine protease, dipeptidyl peptidase I V , was probed with the strategy of designing enzyme-resistant peptide analogues, which were examined in vitro and in vivo. Hypothesis 1: The G I P receptor undergoes homologous desensitization and internalization in clonal (3-cells and/or transfected cell models, and regulation differs from related homologous receptors. Hypothesis 2: The G I P peptide can be dissected into smaller peptide fragments with either agonist and antagonist properties. Hypothesis 3: DPIV-degradat ion of G I P , G L P - 1 and glucagon w i l l a l low generation of superactive enzyme resistant analogues by N-terminal modification. 39 Chapter 2: Materials and Methods 2.1 Reagents All chemicals were of reagent, analytical or higher grade, from Canadian distributors. Common chemicals and salts were from Gibco (Life Technologies Inc., Burlington, O N ) , Sigma-Ald r i ch (Oakvil le , O N ) , B D H Inc. (Toronto, O N ) , Fisher Scientific (Nepean, O N ) , V W R Canlab (Mississauga, O N ) , Merck (Darmstadt, Germany), Amersham Pharmacia Biotech (Mississauga, O N ) and Perk in-Elmer /Mandel S c i e n t i f i c / N E N L i f e Scientific C o . (Guelph, O N ) . Specif ic sources for chemicals are indicated in brackets in the fol lowing sections describing experimental methodology. 2.2 Receptor Plasmid Constructs The rat pancreatic islet G I P receptor was previously isolated by R T - P C R by Dr . R. W . Ge l l ing [276; 324]. This construct was provided in the mammalian expression vector, p c D N A 3 (Invitrogen, Carlsbad, C A ) , subcloned in the Hindlll/Xhol (Gibco) restriction endonuclease sites, and the plasmid was the template for the creation of mutant G I P receptors, generated in collaboration with Dr . M . B . Wheeler (Dept. of Physiology, Universi ty of Toronto). A l l mutant G I P receptor constructs were made using the QuikChange site-directed mutagenesis kit (Stratagene, L a Jolla, C A ) , wi th a P C R based methodology and complementary mutant oligonucleotide primers (25 to 32 bp in length) to amplify the entire plasmid. Degradation of methylated and hemi-methylated parental (wild-type) D N A was accomplished by incubation with Dpnl (G ibco) ; chemical ly competent T o p i OF' E. coli cells ( ini t ia l ly obtained from Invitrogen; rendered competent in house by the R b C l method) were transformed with resultant D N A , prior to spreading on LB/agar petri dishes with appropriate (100 p g / m L ampic i l l in or kanamycin) selection agent. Individual colonies were grown overnight in suspension (Lur ia-Bertani broth with appropriate antibiotic) and plasmid D N A was isolated by \"mini-prep\" [325]. 40 Introduction of the intended mutation (and not additional P C R errors) to the D N A sequences was confirmed using B i g Dye chain terminator sequencing reactions (done primarily in Toronto, and confirmed a second time in some cases at U B C ; sequencing kindly performed by Dr . Ivan Sadowski , Dept. of Biochemistry) . Large scale (midi- or maxi-preparations) plasmid D N A preparations were then prepared using kits provided by Qiagen (Mississauga, O N T ) , resulting in sufficiently pure D N A suitable for mammalian cel l transfection (spectrophotometry A 2 6 0 / A 2 8 0 ratios between 1.8 and 2.0). Nine receptor constructs with mutations in the carboxyterminal tail were generated by this method: G I P R - S 3 9 8 A , G I P R - S 4 0 6 A , G I P R - C 4 1 1 A , G I P R - S 4 2 6 A , G I P R - S 4 2 7 A , G I P R -S426/427A, G I P R - S 4 4 0 A , G I P R - S 4 5 3 A , GIPR-S398/406/426/427/440/453A (Figure 2), as wel l as a G I P receptor chimera with green fluorescent protein ( p G I P R - G F P ) . The latter construct was generated by removal of the stop codon of the G I P receptor and subcloning into the vector p E G F P - N 2 (Clontech, Palo A l t o , C A ) , bearing an in-frame 3 ' O R F for the red-shifted (F64L/S65T) mutant of A . victoria jel lyfish green fluorescent protein. The resulting p G I P R - G F P p lasmid encoded the G I P receptor and G F P intervened by a 14 amino acid sequence ( K P N S A D G I H R P V A T ) derived from the polyl inker of the vector used for mutagenesis and p E G F P - N 2 . Due to extenuating circumstances, the precise sequences of oligonucleotides used for mutagenesis of the G I P receptor cannot be reproduced here, however, mutant constructs w i l l be made available upon request. 2.3 Cell Culture and Transfection T w o cel l types were used in experiments contained in this thesis: Chinese hamster ovary ( C H O ) fibroblast cells (strain K l ; American type tissue collection: C C L - 6 1 ) and (3TC-3 cells, a clonal mouse (3-cell insulinoma cel l line derived from transgenic mice expressing a hybrid insulin/oncogene [326]. Non-transfected C H O - K 1 cells were propagated until passage 15. 41 Transfected cells were discarded after 30-40 passages. (3TC-3 cells were tested between passages 20 to 25. Qualitative assesment of ce l l growth and health was accomplished by observation with an inverted microscope. It is important to maintain low passage number for clonal (3-cells, as the age-related decline in glucose-responsiveness and insulin content is wel l characterized [327], although the (3TC-3 cells may be responsive to insulin secretagogues up to passage 39 [178], or even higher [326]. C H O - K 1 cells were grown in a 1:1 mixture of Dulbecco 's modified Eagle medium ( D M E M ) and Ham's F12 nutrient solution powder provided containing 25 m M glucose, 2 m M glutamine and 110 mg/L pyruvate (Gibco), 50 U / m L penici l l in G (Sigma), 50 [tg/mL streptomycin (Sigma), and 10% newborn bovine calf serum (Cansera, Rexdale, O N ) . Growth medium of stably transfected cells was additionally supplemented with Geneticin (G418, Gibco ; 800 pg/mL) to ensure maintenance of the neomycin cassette containing plasmids, present in p c D N A 3 and p E G F P - N 2 . (3TC-3 cells were propagated in low glucose (5.5 m M ) D M E M with 2 m M glutamine and 110 mg /L pyruvate (Gibco), with antibiotics (as above), 12.5% horse serum and 2.5% fetal bovine calf serum (Cansera). The p H of the medium was adjusted to 7.2-7.3, such that upon vacuum-filter steril ization, physiological p H would be achieved. A l l tissue culture operations were performed in an aseptic laminar flow biohazard cabinet to minimize the chances of microbe contamination. Cel ls were grown at 37°C under a humidified atmosphere of 95% air/5% C 0 2 . Cel ls were fed as necessary, until 80-90% confluence was attained and then harvested using 0.25% trypsin/0.3% E D T A solution (w/v) made up in C a 2 + - and Mg 2 + - f ree Hank ' s balanced salt solution ( H B S S ; Gibco) . Cel ls were mixed with an equal volume of growth medium, centrifuged, and the cell pellet vigorously resuspended by repeated passage through a 2 m L serological pipette. One-tenth of the cells was added to 10 m L of fresh growth medium in a new 75 c m 2 T-flask culture vessel (Fa l con , B e c k t o n - D i c k i n s o n , Mis s i s sauga , O N ) , and the remainder used for experimentation. Cel ls were counted using a haemocytometer: C H O - K 1 cells were seeded at 50-42 70,000 cel ls /wel l and (3TC-3 cells at 500,000 cel ls /well into 24-wel l plates (Falcon) in 1 m L growth medium, and maintained for 48 hours before testing. For stable transfection of C H O - K 1 cells, the C a P 0 4 co-precipitation method was employed. Br ie f ly , this method consisted of plating C H O - K 1 cells in 10 cm diameter culture dishes (Falcon), such that they were 60-80% confluent on the day of transfection. M e d i a were replaced with fresh growth media (6 m L ) at least 2-4 hours prior to transfection. The co-precipitation reaction was performed in a sterile microcentrifuge tube: 10 ug Qiagen purified plasmid D N A were added to 62.5 u L of 2.5 m M C a C l 2 , and brought up to a final volume of 500 u.L with sterile de-ionized H 2 0 . To the D N A / C a l c i u m mixture, 500 u L of 2 X H B S (for 100 m L : 1.6 g N a C l , 0.074 g K C 1 , 0.02 g N a 2 H P 0 4 (anhydrous), 0.2 g dextrose, 1 g H E P E S , p H adjusted to precisely 6.95) were added, the solution vortex mixed and the precipitate was al lowed to form for 30 minutes at room temperature. The entire 1 m L of D N A / C a l c i u m / H B S solution was added to the cells in 6 m L growth medium and incubated for 4 hours at 37°C . M e d i a were aspirated, and cells subjected to a 90 second glycerol shock by addition of 2 m L 15% glycerol/1 X H B S . Glycero l was removed, 6 m L fresh growth medium added, and cells were allowed to recuperate overnight. Cel l s were split 1:2 into two 10 cm dishes with growth medium supplemented with 800 (xg/mL G418; medium was changed as necessary, and after 7-10 days, colonies were visible to the naked eye. In some cases, single colonies were subcloned and expanded for further characterization (where noted), or dishes were simply trypsinized, and pooled clones were used. It was established that pooled subclones of C H O - K 1 cells transfected with the G I P receptor showed similar levels of surface expression as the wel l characterized high-expressing subclone rGIP-15 (also known as w t G I P R or GIPR-455) [276; 280; 290]; in the current report, it was found that such pooled clones d id show s imilar levels of receptor b inding, cyc l i c A M P production, and could be used in many types of experiments. However, the heterogenous nature of these cel l populations presented some difficulties for other experiments, and subcloning was 43 performed where feasible. Transfection by the C a P 0 4 method initially resulted in approximately 10-12% transfection efficiency; fo l lowing selection with G 4 1 8 , greater than 85% of cells expressed the protein contained in the plasmid (as estimated in experiments with p G I P R - G F P ) . Subcloning was accomplished by aspirating the media from the 10 cm dishes and using a P20 Pipetman (Gilson Inc., Middleton, W l ) with sterile tips to carefully pick clusters of cells from the centers o f vis ible colonies; colonies were transferred to individual wells o f 24-wel l plates (Falcon) to expand. Subclones were then screened for receptor expression by measuring l 2 5 I - G I P binding in the presences and absence of unlabelled G I P (see below). Transient transfection was used in several experiments when it was desired to transfect more than one plasmid into the same cell (e.g. the G I P receptor and G-protein receptor kinase 2). This was accomplished using the LipoFectamine 2000 reagent (Gibco) and a modified manufacturer's protocol. Br ief ly , cells were plated in 10 cm dishes such that 90-95% confluence was achieved on the day of transfection; for C H O - K l cells, this was achieved by plating ~2 X 10 6 cells in 7 m L growth media, and culturing for 48 hours. In an autoclaved microcentrifuge tube, 10 u.g of total D N A (e.g. 3 ug p G I P R and 7 ug of p c D N A 3 or p G R K - 2 ) were diluted in 250 u L of sterile room-temperature D M E M / F 1 2 (antibiotic and serum-free) and in a second tube, 20 u.L of LF2000 reagent were added to 250 u L D M E M / F 1 2 . Solutions were vortex mixed, incubated for 5 min at room-temperature, and then combined and mixed. The D N A / l i p o s o m e complexes were allowed to form for 30 min at room-temperature; the entire 500 u.L solution was then added to cells in 3 m L fresh growth media (final volume 3.5 m L ) , and cells were incubated at 37°C overnight. The fo l l owing day, cells were harvested and seeded in 24-wel l plates for experimentation 48 hours later. This protocol resulted in transfection efficiencies between 40 and 45%. Whi l e this method was more time-consuming than generating pooled stable clones, it was less labour intensive than subcloning stable cel l lines, and a l lowed control of receptor 44 expression level by keeping total plasmid D N A constant (to maintain transfection efficiency) and varying the proportion of receptor plasmid. Stable transfected C H O - K1 cell lines expressing the human G L P -1 receptor ( h G L P - l R cells) or human glucagon receptor (hGlucR cells) were generated by Dr . R. Ge l l ing , and previously described in the literature [280; 328]. 2.4 Peptides Synthetic human and porcine G I P , . 4 2 , G L P - 1 7 . 3 6 N H 2 , and [Glu 9 ]glucagon 2 . 2 9 were purchased from Bachem (Torrance, C A ) . A l l other peptides described in the current report were synthesized by Dr . Susanne Manhart , Probiodrug A G , Ha l l e (Saale), Germany; additional batches of GIP,_ 4 2 and G L P - 1 7 _ 3 6 N H 2 were also prepared. These peptides were synthesized using an automated peptide synthesizer, Ra in in Symphony, according to published methods [329]. Crude peptides were purified by H P L C using a C H 3 C N / H 2 0 gradient in the presence of 0.1 % trifluoroacetic acid ( T F A ) , and subjected to mass spectrometry ( M A L D I - T O F ) and analytical H P L C to confirm identity and purity. A table of peptides is provided in Appendix A , with expected and measured masses. Peptide content of lyophi l ized peptides was determined to be approximately 70% for all syntheses (including commercial ly bought peptides), but was not factored in during testing of the peptides described here, as is common practice. Peptide handling was according to standard protocol. Basic peptides were appropriately dissolved in a small volume of 0.1 M C H 3 O H , and made up to volume with sterile de-ionized water, whereas 0.1 M N H 4 O H was used for acidic peptides; G I P , . 4 2 , human and porcine, were the only basic peptides, of all those presented. Peptides were dispensed at 2 nmol or 20 nmol in si l iconized tubes, and freeze dried in vacuo, using either a lyophil izer or Speed-Vac. Tubes containing dried peptides were sealed with Parafilm, and kept at -20°C until use. 45 2.5 Peptide Iodination Radioactive labelling of peptides was accomplished using established techniques developed in our laboratory. Brief ly , 1 m C i o f carrier-free iodine-125 (Perkin-Elmer) was added to 1 nmol o f peptide dissolved in 100 u l o f 0.4 M phosphate buffer (pH 7.5), and iodination initiated by addition o f 10 p i chloramine-T (14.2 m M in 0.4 M P 0 4 buffer; B D H ) for 15 seconds and the reaction was quenched with 20 u l sodium metabisulphite (66.3 m M in 0.4 M PO4 buffer; Fisher). Rad io labe led peptide was separated from free iodine by gel fdtration (Sephadex G-10 or G-15; Pharmacia, Uppsala, Sweden) in 0.2 M acetic acid with 0.5% B S A (R1A fraction V ) and 2% Trasylol® (aprotinin; Bayer, Etobikoke, Canada). Labelled peptides were then purified by H P L C to yie ld single molecular species; the C t L C N ^ O (+ 0.1 % T F A ) gradients used depended on the specific peptide, and are shown in Figure 4. Solvents were filtered and degassed on each day o f purification. The H P L C apparatus consisted o f two Beckinan H O B solvent delivery module pumps (flow rate = 1 mL/min) with a programmable Beckinan 4 2 1 A controller, a pBondapak-C18 column (Waters, M i l f o r d , M A ) , and a model 170 in-line Radioisotope Detector connected to a chart recorder. Peaks were manually collected, aliquotted at 3-4 X 10 6 cpm into tubes containing 5 p L o f 1% B S A (w/v) and 50% Trasylol (v/v); tubes were freeze-dried, sealed with Parafilm, and stored at -20°C until used. Peak II o f the 1 2 5I-GIP,. 42 chromatogram is known to be [ l 2 5 l - T y r 1 0 ] G I P i_ 42 [59], and is a b io logica l ly active product [174; 266]. For [D-Ala 2 ]GIP|.42, G L P - I7.36NH2, and G I P U M O H , a l l peaks were empirically tested for receptor binding using appropriate methodologies. For the first two peptides, two biological ly active iodinated products were detected for each (Figure 4), but only the first more abundant product was used for experimentation, whereas no bio logica l ly active iodinated GIPI-MOH product could be detected. Using the same iodination protocol, but with non-radioactive N a l at the same concentration, mass spectrometric analysis detected only oxidized forms o f mono- and di-iodinated GIPI-MOH (data not shown). Use o f this method for 46 iodination and purification routinely resulted in specific radioactivities in the range o f -250-350 u.Ci/ug (-45 to 65 M B q / n m o l ) [59; 178; 276; 324; 330]. Glucagon label (3 - [ 1 2 5 I ] iodo ty rosy l ' ° -g lucagon) was purchased from Amersham. E Q. E a. A . Purification of 1 2 5 I - G I P 1 ^ 2 O H 15 20 25 Time (Min) B. Purification of 1 2 5 l - [ D - A l a 2 ] G I P l J l 2 0 H 15 20 25 Time (Min) C. Purification of 1 2 5 I - G I P 1 . 1 4 0 H 15 20 Time (Min) D. Purification of 1 2 5 I - G L P - 1 7 . 3 6 N H 2 42% 36% ^ H * 1 / \\ A \\ : 1 1 1 1 1 1 1 — u 10 15 20 Time (Min) 25 30 35 •10 47 Figure 4: HPLC profiles of iodinated synthetic peptides Peptides were iodinated by the chloramine-T method (see text) using 1 m C i carrier free iodine and 1 nmole of peptide. Iodinated product was separated from free iodine by gel filtration, and then separated using a C H 3 C N : H 2 0 + 0.1% T F A gradient (right Y axis on graphs) at a flow rate of 1 m L / m i n on a uBondapak C-18 column. Peaks were empirically tested for specific binding to w t G I P R or h G L P - l R cells, and those producing acceptable specific binding with low non-specific binding are indicated by (*). T w o biological ly active peaks were identified for each peptide except G I P M 4 , which had none and, invariably, the earlier eluting peak was used in all further experiments. The second bioactive peak for p G I P M 2 has been previously reported [174; 266]. Mass spectrometric analysis indicated that iodinated G I P , . , 4 consisted only of oxidized forms of mono- and di-iodo peptides (data not shown). 2.6 Binding Studies A l l transfected C H O - K 1 cells were tested for receptor expression and receptor binding affinity by competition binding studies. Cel ls were prepared in 24-well plates, and had been grown for 48 hours to approximately 2-5 X 10 5 cells/well (~90% confluent). Cel ls were washed twice with ice-cold 15 m M HEPES-buf fe red (pH 7.4) D M E M / F 1 2 supplemented with 0 .1% B S A . Cel ls were then incubated 12-16 hours (overnight) at 4 ° C in 200 u.L of the same buffer additionally supplemented with 1% Trasylol , 50,000 cpm iodinated peptide and concentrations of unlabelled peptides indicated in the figures (1 p M to 1-40 p M , depending on the experiment and peptide tested; triplicate determinations). M e d i u m was then aspirated from the wells, which were washed twice with ice-cold buffer, prior to solubilization of cell-associated radioactivity using 1 m L of 0.1 M N a O H and transfer to borosilicate tubes for counting. Non-specific binding was defined as cpm measured in the presence of 1 u M of native peptide, G I P j . 4 2 , G I P , . 3 0 N H 2 , G L P -1 7 . 3 6 N H 2 , or glucagon,_ 2 9, where appropriate. Non-l inear regression analysis of competitive-binding curves followed.algorithms included with the Prism 3 software package (GraphPad, San Diego, C A ) . Data were fitted to a single site model: Y = Bottom + [(Top - Bottom)/(l + 1 0 ( X - L o g I C 5 0 ) )] E q . 1 Where Y is the bound label and X is the log value of the concentration of unlabelled competitor, T o p and Bot tom are high and low measured cpm values (respectively), and I C 5 0 is the 48 concentration of unlabelled competitor that displaces 50% of bound label. Specific binding in the absence of competitor (B0) is proportional to maximal binding ( B m a x ) by the relationship: B 0 = B m a x X / ( X + K d ) E q . 2 Where X is the label concentration (molar), and is constant for a given experiment, and K d is the equi l ibr ium dissociation constant for the receptor and ligand, also in molar units. Hence B 0 could be used in competitive-binding experiments to estimate B m a x , and determine relative levels of expression, provided affinity was constant. Saturation b ind ing isotherm experiments were also performed in some cases to experimentally determine B m a x values accurately. In these studies, cells in 24-well plates were prepared similar to above, except that serial 1:2 dilutions of 2 X 10 6 cpm were added to wells in the presence or absence of 1 uJVI unlabelled peptide (in triplicate). Total label concentrations were calculated using the specific radioactivity of the label and measured cpm values of added label and specific binding. Data were fitted to the relationship: Y = B m a x X/(X + K d ) E q . 3 Where Y is experimentally measured specific binding in units of cpm/cell and X is the molar concentration of added label, thus al lowing calculation of B m a x (converted to sites/cell) and K d . Scatchard plots were not used, due to the wel l documented inaccuracies associated with that method [331]. 2.7 Cyclic AMP Measurements C y c l i c A M P studies performed in C H O - K 1 cells were performed on cells plated 48 hours prior in 24-well plates, having grown to ~2-5 X 10 5 cells per wel l . Cel ls were washed 2 X and preincubated for one hour in warm H E P E S - b u f f e r e d D M E M / F 1 2 wi th 0 .1% B S A . Subsequently, cells were stimulated for 30 min in the presence or absence of peptides (at the concentrations shown in figures, in triplicate), in the same buffer additionally supplemented with 49 1% Trasylol and 0.5 m M isobutyl-methylxanthine ( I B M X ; Research Biochemicals Intl., Natick, U S A ) . I B M X was dissolved at 0.5 M in dimethylsulfoxide ( D M S O ; B D H ) , dispensed into 25 uJL aliquots and frozen at -20°C until use. (Gel l ing [324] reported that maximal cyc l ic A M P levels varied between experiments; it is thought that a contributing factor to this variability was in part because in the earlier work by Ge l l i ng , I B M X was weighed and dissolved in warm isopropanol on the day of each individual experiment). F o l l o w i n g stimulation, media was decanted from the wells, and cells were lysed in 1 m L ice-cold 70% ethanol. Cellular debris was removed by centrifugation (5-10 min, 4 ° C , 12-15,000 rpm), and cel l contents were dried by vacuum centrifugation. Samples were reconstituted in an appropriate volume of sodium acetate buffer (0.05 M , p H 6.2) and assayed for cycl ic A M P content by radioimmunoassay kit according to the manufacturer's instructions (Biomedical Technologies Inc, Stoughton, U S A ) . Analysed data are presented as pmol cyl ic A M P / w e l l , femtomol/1000 cells, or as a percentage of maximal GIP-stimulated c A M P production. Mod i f i ed versions of the protocol above were designed to characterize desensitization of the G I P receptor in transfected cells. The effect of receptor expression level , a cycl ic A M P accumulation time-course and a time-course of desensitization were established. Generally, 100 n M G I P with 1% Trasylol was included for various times (in the absence of I B M X ) prior to a 10 minute washout period, during which time media were changed twice, and subsequently a 30 min stimulation with G I P in the presence of I B M X was carried out. Concentration and measurement of intracellular cycl ic A M P proceded unchanged. C y c l i c A M P measurements in (3TC-3 cells were performed essentially as described with minor modifications. T w o days after plating, cells were incubated in standard growth media, but with only 1 m M glucose, for 6 hours prior to experimentation. Cel l s were washed twice with 3 7 ° C buffer ( D M E M , 25 m M H E P E S , 0 .1% B S A , p H 7.4, and glucose: 0, 5.5 or 11.0 m M ) . Ce l l s were then incubated for 1 hour in c A M P buffer of varying glucose concentrations, supplemented with 1% Trasylol , with or without 100 n M GIP . Treated cells were rinsed twice 50 with appropriate glucose-containing c A M P buffer over a 10 minute washout period. Cel ls were then stimulated for 30 minutes with a range of G I P concentrations in triplicate, as indicated in the figures, in (3TC buffer supplemented with 0.5 m M I B M X . Forskol in (10 p M ; Sigma) was used as a positive control. C y c l i c A M P content was measured as above (70% ethanol lysis , centrifugation, drying, and c A M P R I A ) . Further control experiments were performed in Kreb ' s Ringer Bicarbonate H E P E S ( K R B H ) buffer (120 m M NaCI , 5.7 m M K C 1 , 1.2 m M N a H 2 P 0 4 , 15.5 m M N a H C 0 3 , 1.2 m M M g C l 2 , 15 m M H E P E S , 0.1% B S A , 2.5 m M C a C l 2 ) . Cel ls were similarly incubated for 1 hour in 0 or 5.5 m M glucose K R B H , prior to a 30 minute test period measuring basal, 100 n M GIP-stimulated or 10 p M forskolin-stimulated c A M P production. The effect of glucose on c A M P production and desensitization were examined in parallel, such that one independent experiment consisted of 6 experimental conditions on cells prepared all at the same time. C y c l i c A M P data were expressed as pmol per wel l or normalized to production of cycl ic A M P stimulated by 10 u.M forskolin. Further experiments on |3TC-3 cells were conducted on the effect of dipeptidyl peptidase I V ( D P I V ) inhibit ion, pertussis toxin, phosphodiesterase inhibition, or protein kinases A and C on desensitization of the cycl ic A M P response of (3TC-3 cells to G I P . A s desensitization could be explained by degradation of extracellular G I P by induction of D P I V or pertussis toxin-sensitive G-proteins, these hypotheses were tested. Cel ls were prepared as above, pretreated with 100 n M G I P in HEPES-buffered 5.5 m M glucose D M E M , and subsequently stimulated with 10 n M G I P in the presence or absence of 50 u M isoleucine-thiazolidide (Ile-thia, D P I V K, = 130 n M ; Probiodrug, Hal le , Germany) or pertussis toxin (100 p g / m L during washout and 500 p g / m L during stimulation; Sigma). The effect of I B M X (phosphodiesterase I C 5 0 = 2-50 p M ) on G I P desensitization was similarly tested. Cel ls were pretreated in the presence or absence of 100 n M G I P fol lowed by a 10 minute washout period; during the subsequent stimulation period with 10 n M G I P , I B M X concentration was varied between 0 and 4 m M . Blockade of protein kinases A 51 or C was accomplished using 5 u M H89 ( P K A K; = 48 n M ) or 100 n M staurosporine ( P K C K ; = 0.7 n M ; both from Calbiochem, L a Jolla, C A ) , included 15 minutes prior to and during the 100 n M GIP /60 min desensitization period. Inhibitor concentrations were based on inhibitory constants and commonly used values. The effect of these inhibitors on basal and GIP-stimulated c A M P production was also measured. 2.8 Insulin Release Experiments [3TC-3 cells were washed twice with appropriate glucose containing K R B H and preincubated for 1 hour in the presence or absence of 100 n M G I P (in quadruplicate) in K R B H supplemented with 1% Trasylol at 3 7 ° C / 5 % C 0 2 . A s with the c A M P studies, cells were washed over a 10 minute period and then either stimulated a further 30 minutes in K R B H with glucose (0, 5.5 m M , or 11 m M ) with or without 10 n M G I P or 10 u M forskolin. M e d i a were removed for insul in radioimmunoassay (described below) and untreated cells were extracted with 2.0 M acetic acid for determination of total cell insulin content. 2.9 Receptor Internalization G I P receptor sequestration in transfected C H O - K 1 cells was measured using modifications of two methods previously described: the first examined loss of surface binding sites, designed by combining the methods of [332] and [333], and was similar to methods published later [334], and the second was an altered protocol examining internalization of iodinated peptide [335]. The first method measured remaining surface receptor binding fo l lowing exposure to G I P for set durations. Cel ls in 24-well plates were washed 2 X with 37°C assay buffer (15 m M H E P E S -buffered D M E M / F 1 2 + 0.1% B S A ) , and covered with 180 u L of the same buffer. Cel ls were held on a shallow stage in a 37°C waterbath, and at designated time-points (counting down from 60 min), 20 U.L of peptide were added (in triplicate or quadruplicate) to give a final concentration 52 of 100 n M in the wel l . A t t = 0, cells were rinsed with ice-cold acid stripping buffer (2 X 1 m L ; 150 m M N a C I , 50 m M glycine [analytical grade; M e r c k ] , brought to p H 3 with glacial acetic acid) and incubated on ice in this buffer for 5 minutes to remove surface bound (non-internalized) unlabelled G I P . Cel l s were again washed two times wi th co ld assay buffer, resuspended in 200 u L of buffer containing 50,000 cpm l 2 5 I - G I P with 1% Trasylol . Remaining surface binding was measured after a 4 hr incubation at 4 ° C , as per the radioligand binding studies. Non-specific binding was determined by inclusion of excess unlabelled peptide; data were normalized to non-internalized surface binding (t = 0 min). In some experiments, different concentrations o f peptide or pharmacological agents were used (0.5 m M I B M X , 10 p M forsko l in , 100 p M M D L - 1 2 , 3 3 0 A { C a l b i o c h e m } , 100 p M 2 ' , 5 ' - D i d e o x y a d e n o s i n e {Calbiochem}, 5 u M H89 , 400 n M P M A {4-phorbol-12-myristate-13-acetate; Calbiochem}, 100 n M staurosporine, added 15 min prior to GIP) , and compared at a single time-point (i.e. 30 min G I P alone, 45 min agent alone versus 30 min G I P + 45 min agent), relative to control untreated cell surface receptor expression. The second method involved measurment of internalized l 2 5 I - G I P , and was only used for w t G I P R cells, and C H O - K 1 cells transfected with G I P R / G F P constructs. Br ie f ly , cells in 24 wel l plates were washed twice in ice-cold assay buffer, and incubated with 50,000 cpm of 1 2 5 I -G I P at 4 ° C for 60 min. Plates were heated to 37°C for the indicated times; half of the cells were treated by acid stripping on ice (as above), the other half washed with cold buffer (total binding). Non-specif ic binding measured in the presence of excess unlabelled G I P was determined separately for acid-stripped and total binding cells. B ind ing that is resistant to acid stripping is generally accepted as radioligand which has been internalized by receptor mediated endocytosis, and expressed as % acid resistant binding/total binding. Acid-resistant binding at t = 0 (cells not warmed to 3 7 ° C ) was between 10-15%, and represents the efficiency of the acid stripping 53 protocol at removing surface bound label - many studies subtract this value, however, it is included as important control data. 2.10 Fluorescence Microscopy Cel ls were seeded onto 3-aminopropyl triethoxy silane ( A P E S ; Sigma) coated 18 mm glass coverslips (Fisher) in 12 wel l plates at densities of 12.5, 25, 50 and 100 X 10 5 cells and grown for two days. For G I P R / G F P studies, transfected cells were washed twice with assay buffer, and incubated for 60 min ± 100 n M G I P M 2 at 3 7 ° C . Cel l s were washed twice with phosphate-buffered saline (PBS , p H 7.4), and fixed with 4% paraformaldehyde (Merck) freshly prepared in P B S at room temperature. Coverslips were rinsed again with P B S , and mounted on slides in 30% glycero l /PBS and adhered with rubber cement. For [Fluorescein-Trp 2 5 ]GIP l _ 3 0 N H 2 studies, cells were washed twice with Earle 's balanced salt solution (140 m M N a C l , 5 m M K C 1 , 1.8 m M C a C l 2 , 0.9 m M M g C l 2 , 25 m M H E P E S , 0.09% glucose, 0.2% B S A ; adjusted to p H 7.4 with 500 m M trisma base), and incubated with 25, 50 and 100 n M [Fluo-Trp 2 5 ]GIP for 60 min at either 4 ° C as a control, or 37°C to study receptor internalization. Cel l s were washed with P B S and fixed with paraformaldehyde as above. Coverslips were mounted on slides in 30% glycerol /PBS and adhered with nailpolish. Separate sets of coverslips were additionally permeabilized with 0.1% Tr i ton -X 100 (5 min), blocked with 5% B S A (60 min), and then treated with the A l e x a -fluor 488 fluorescein amplification protocol (according to the manufacturer's directions: 25 |xg/mL A F 4 8 8 rabbit anti-fluorescein IgG, fol lowed by 15 u.g/mL A F 4 8 8 goat anti-rabbit IgG; Molecu la r Probes, Eugene, O R ) . Images were captured using a Zeiss Axiopho t microscope (Thornwood, N Y ) and a F I T C filter set (excitation X = 488 nm, emission X = 520 nm). 54 2.11 Peptide Degradation Studies and Cell DPIV Activity The hydrolysis o f iodinated peptides by purified D P I V (specific activity = 31.2 U / m g ; prepared by Leona Wagner, Probiodrug A G , Halle (Saale), Germany) was studied as previously described [59; 330]. In brief, radiolabeled peptides (750,000 cpm/150 u L ) were incubated in H B S (40 m M H E P E S , 154 m M NaCI , p H 7.6) with or without D P I V (10 m U ) for 16 h. Samples were resolved by H P L C using a C H 3 C N / 0 . 1 % T F A (v/v) and H 2 O / 0 . 1 % T F A solvent system at a flow rate o f 1 m L / m i n , according to published methods. Iodinated G I P peptides (approx. 50,000 cpm/20 u L injection) were separated by a protocol consisting o f 14 minutes at constant 32% C H 3 C N / 0 . 1 % T F A and fol lowing, a linear gradient to 38% C H 3 C N / 0 . 1 % T F A over 10 min , fol lowed by a further 5 minutes at 38% CH3CN/0.1%> T F A . Fractions were collected at 15 second intervals between minutes 5 and 27 o f this protocol. Data were normalized to the total radioactivity recovered in a l l o f these fractions. Between each sample injection, the C-18 u,-Bondapak H P L C column was rinsed for 20 minutes in 100% C H 3 C N / 0 . 1 % T F A , and any residual radioactivity was removed with 3 consecutive injections o f 300 p i D M S O ; the sample port was rinsed wi th 400 u L o f H2O and the column was al lowed to re-equilibrate at 32%o C H 3 C N / 0 . 1 % T F A for 5 min prior to the next injection. Under these conditions, a second peak elut ing earlier than intact l 2 5 I - G l P i_42 has been conf i rmed to be 1 2 5 f-GfP3-42 by Edman degradation analysis [59]. A single experiment was conducted to establish i f in vitro cell models display D P I V - l i k e activity. C H O - K 1 or |3TC-3 cells were harvested in 1 m M E D T A / P B S (pH 7.4) and washed 2 X with 0.9% NaCI + 40 m M H E P E S + 0.1% B S A . Cel ls were resuspended in 250 u L buffer ± 100 p M Ile-thia (after substrate addition, 50 p M final) at ~ 2 X 10 6 cells/tube. Substrate (250 p L ; Gly-Pro-parani t roani l ine; Sigma) in buffer (defined above) was added to cells, for a final [substrate] = 400 u M and cells were rotated slowly at 30°C for 30 min. Cel ls were centrifuged 55 (1200 rpm, 5 min, 4C) and 100 u.L of the supernatant was added in duplicate to a 96 wel l plate. Absorbance was read at 405 n M on a M R X plate reader (Dynatech Laboratories, Chanti l ly, V A ) . D P I V activity was determined from a standard curve, and normalized for cell number. 2.12 Animals and Peptide Bioassay M a l e Wistar rats were purchased from the Universi ty o f Br i t i sh Co lumbia A n i m a l Care Faci l i ty (Vancouver, Canada); male V D F Zucker rats were obtained from the colony maintained in the Department o f Physiology, University o f Bri t ish Columbia (Vancouver, Canada). Animals were held in group housing with free access to rat chow and tap water, with a 12 hour l ightdark cycle. Animals were fasted overnight (15-18 hours) prior to experimentation. Anesthesia, where indicated, was achieved with intraperitoneal (IP) Somnotol® (65 m g / K g sodium pentobarbital; M T C Pharmaceuticals, Cambridge, Canada). A n i m a l experiments conformed to the guidelines set forth by the Universi ty o f Bri t ish Columbia Committee on A n i m a l Care and the Canadian Counc i l on A n i m a l Care. Ear ly bioassay experiments were performed using intravenous infusion o f peptide and intraperitoneal injection of glucose. Cannulae were inserted into the jugular vein and the carotid artery of anesthetized fasted male Wistar rats (150-250 g). Basal blood samples were withdrawn from the carotid artery (500 u.1), and fasted blood glucose was measured via the tail vein using a SureStep® glucose analyser (LifeScan Canada L td . , Burnaby, B . C . ) . Intravenous (IV) saline or peptide infusion (2.3 mL/h r to deliver 1 pmol/min/100 g bw or 100 pmol/min/100 g bw) was then started (t = -5 min). A t t = 0 min, an IP glucose injection was given (1 g / K g body weight, bw). B lood samples were taken at 10, 20, 30 and 60 min, and blood glucose was monitored at 10 minute intervals throughout the experiment. Plasma was separated by centrifugation (10,000 rpm, 20 min , 4 ° C ) and immunoreactive insul in was measured by radioimmunoassay, as described below. 56 A bioassay was later developed for the conscious unrestrained rat, in order to min imize surgical labour and animal sacrifice. M a l e Wistar rats (approx. 290 g) were fasted overnight, and then given an oral glucose tolerance test ( O G T T ; 1 g glucose/Kg body weight) wi th concurrent subcutaneous (SC) saline or peptide (8 n m o l / K g bw in most cases) injection (500 p L ) . B l o o d samples were taken at indicated times from the tail vein using heparinized capillary tubes, and the plasma separated by centrifugation for radioimmunoassay. B l o o d glucose was measured every 10 minutes using a SureStep® glucose analyser (LifeScan Canada Ltd . , Burnaby, Canada). Similar experiments were subsequently carried out using age matched 16-20 week old lean and obese V D F Zucker rats (approx. 335 g and 575 g, respectively). Obese Zucker rats have a homozygous recessive defect in their leptin receptor (fa/fa) causing hyperphagia leading to hyperglycemia, hyperl ipidemia, hyperinsulinemia, and insul in resistance, s imilar to human disorders o f obesity and diabetes. The dominant genotype (Fa/Fa or Fa/fa) results in a normal lean Zucker rat. 2.13 Hormone Radioimmunoassays Insulin measurement by R1A was done by incubating samples diluted (usu. 1:3 v/v) in insulin R I A buffer (0.5% charcoal extracted human donor plasma, in 40 m M phosphate buffer, p H 7.5), wi th insulin antisera (GP01; 1:1,000,000 final dilution) under disequilibrium conditions using rat insulin (Linco Research Inc., St. Charles, U S A ) as a standard. After 24 h at 4 ° C , 2000 cpm o f chloramine-T iodinated porcine insulin (Sigma) was added to al l tubes; subsequent to another 24 h 4 ° C incubation, antibody bound and free radioactive insulin were separated by centrifugation with dextran-coated charcoal. This protocol has been previously reported [336; 337], and has a lower detection limit o f 0.125 ng/ml. Measurement o f G I P by R I A was performed as previously described using a C-terminally directed antibody [336]. Plasma samples were appropriately diluted (usually 1:5 v/v) in G I P 57 R1A buffer (5% charcoal extracted human donor plasma, 2% Trasylol , 40 m M phosphate buffer, p H 6.5), and incubated with G I P antiserum ( R K 3 4 3 F , 1:300,000 final dilution) at 4 °C . 1 2 5 I - G I P (5000 cpm) was added 24 h later and, fol lowing a second 24 h incubation, antibody bound and free r a d i o l a b e l e d G I P were separated by P E G - 8 0 0 0 precipi ta t ion (12 .5% w / v f ina l concentration; B D H ) and centrifugation. The lower detection limit for this assay is 7.8 pg/ml. Measurement o f immunoreactive G L P - 1 was accomplished using a commercial ly available kit (Linco) that employed an antibody recognizing C-terminal ly amidated forms o f G L P - 1 independent o f whether the N-terminus was intact (i.e. GLP-17.36NH2, GLP-I9 .36NH2, G L P - 1 1 . 36NH2)- Non-extracted plasma samples were treated according to the manufacturer's instructions, al lowing the assay to measure G L P - 1 in samples at concentrations above 10 pg/ml. 2.14 Analytical Methods For in vitro experiments, in all studies, a minimum of three independent experiments were performed. Each independent experiment was conducted on a different day using triplicate or quadruplicate determinations for each condition. A n i m a l experiments were performed with a minimum of 4-6 animals in each group, assigned treatment in a random fashion. N o data points were excluded, except in cases where human error was noted. Data shown in figures represents the mean ± standard error of the mean ( S . E . M . ) , and the number of independent experiments is indicated in the figure legends or text. Statistical significance was assessed using Pr ism data analysis software (GraphPad, San Diego, C A ) . Student's t test or analysis of variance ( A N O V A ) were performed where appropriate, fol lowed by the post-hoc tests, Newman-Keuls multiple comparison test, Dunnet 's t-test or the Tukey test (where indicated). P < 0.05 was considered significant; F tests confirmed variances were equal. Integrated glucose and insulin responses were calculated using the trapezoidal method, with the algorithm included in the Prism software package, and basal levels as a baseline. 58 Chapter 3: Regulation of GIP Receptor Function 3.1 Introduction 3.1.1 G-Protein Coupled Receptor Regulation Therapeut ical ly , it may be beneficial in specific disease states to either activate or antagonize/inactivate a given receptor, thus it is useful to characterize the phys io logica l mechanisms of receptor regulation. Exposure of a receptor to high and/or continuous agonist usually leads to dimunition of the cellular response. There are several levels at which the signal can be modulated: (1) extracellular degradation of the ligand, (2) ligand removal by cellular uptake, (3) s te r ica l ly prevent ing G-pro te in c o u p l i n g by receptor phosphory la t ion (desensitization), (4) modulation of G T P hydrolysis by G-proteins, (5) modulation of receptor effectors, (6) receptor sequestration, (7) receptor degradation, and (8) down-regulation of receptor gene transcription. Mechanisms of desensitization have been reviewed recently [338-350]. Classical ly, desensitization refers to the rapid effect of receptor phosphorylation functioning to block heterotrimeric G-proteins from re-coupling to the receptor, thus b locking further signall ing. In order to completely block the signal, (3-arrestin proteins are recruited to the receptor's intracellular face. The time scale of receptor desensitization by phosphorylation is generally a matter of seconds [338; 340; 341; 343]. Desensitization is further classified into two types: homologous and heterologous desensitization. The first type refers to desensitization of a receptor by its own ligand, and generally occurs by phosphorylation of ligand-bound receptors by G-protein coupled receptor kinases ( G R K s ) , although depending on the signalling pathways used by the receptor, this could also theoretically be mediated by signal transduction kinases, such as P K A and P K C . Heterologous desensitization, however, is not as specific - both ligand-bound and unbound receptors may be phosphorylated by signal transduction kinases activated by alternate receptors. This latter means of desenstization has been less extensively studied than 59 homologous desensitization [338; 340; 341; 343]. Heterologous desensitization was not examined in the current manuscript, and only homologous desensitization w i l l be considered further. The discovery of receptor endocytosis stems from work on the adrenergic system whereby, fo l lowing agonist treatment, binding sites were lost from plasma membranes of erythrocytes [351]; later, b inding sites were discr iminated by separation o f intracellular membrane compartments from the plasma membrane by sucrose gradient centrifugation or by using hydrophi l ic versus hydrophobic ligands [352; 353]. Subsequent work was able to show internalization of adrenergic receptors using epitope tagged receptors [354] and in real-time using receptor/green-fluorescent protein chimeras [355]. Al though there is controversy as to whether phosphorylation is necessary for receptor endocytosis, it does appear to be an antecedant in most cases, and may result in the wide variability observed in kinetics of internalization, when comparing different types o f receptors. W h i l e phosphorylation may not be an absolute prerequisite for internalization of many receptors, phosphorylation does appear to promote receptor endocytosis. In general, the time-scale of receptor endocytosis is in the order of minutes to hours [338]. Once sequestered into the intracellular compartment known as an \"endosome\", the l igand is degraded and receptor may undergo dephosphorylation (facilitated by the low intravesicular pH) and be recycled to the cell surface (i.e. re-sensitization), or the receptor may also be degraded [345]. Recent studies have indicated that receptor sequestration is a requirement for dephosphorylation, recycling and resensitization [356; 357]. When a receptor agonist persists for hours or days, the reduced number of receptors expressed appears to extend beyond protein degradation [345]. Currently, there is evidence that prolonged stimulation results in down-regulation of steady-state m R N A levels, and can influence m R N A stability [339]. 60 3.1.2 Regulation of Family B G-Protein Coupled Receptors The p-adrenergic receptor is often considered to be the prototypical G P C R , and thus the paradigm of desensitization and internalization, as characterized using the (3-adrenergic system has been broadly applied to al l G P C R s . However , in a side-by-side comparison of the |3 2-adrenergic (family A ) and secretin (family B ) receptor, it was found that the family B receptor behaved differently from the family A counterpart [358; 359]. Whi le the secretin receptor was a substrate for G R K - 2 and -5, and phosphorylation was shown to promote desensitization [359], G R K s and (3-arrestin had no influence on receptor internalization, but P K A did [358]. Hence, it is clear that the dogma of receptor desensitization does apply to most G protein-coupled receptors, however, each receptor must be considered individual ly with respect to the precise mechanisms of desensitization. A review of desensitization of receptors of the secretin/VIP superfamily of receptors is found elsewhere [360], and only desensitization of receptors closely related to the G I P receptor w i l l be considered further, namely the G L P - 1 and glucagon receptors. Desensitization and internalization of the cloned glucagon receptor have been characterized less than these processes for the G L P - 1 receptor. Treatment of CHO-K1/g lucagon receptor cells wi th glucagon induced a rapid time- and concentration-dependent phosphorylation o f the receptor; phosphorylation was independent of P K A and P K C activity [361]. Phosphorylation of serines in the intracellular C-terminus was correlated with receptor endocytosis [335; 361]. Pretreatment of baby hamster kidney ( B H K ) cells transfected with the human glucagon receptor wi th glucagon or carbachol desensitized the ca lc ium response to a subsequent glucagon stimulation, indicating that the receptor undergoes homologous and heterologous desensitization [322], however, the involvement of G R K s or signal transduction kinases has not been proven. The first report of desensitization of the G L P - 1 receptor examined attenuation of cycl ic A M P and insulin responses to G L P - 1 fol lowing preperifusion of HIT-T15 cells with G L P - 1 , G I P or glucagon [177]. A rapid, reversible desensitization of both c A M P and insulin responses to G L P -61 1 were observed; a 10 min desensitization period with 100 n M G L P - 1 , fol lowed by a 10 minute washout period, and subsequent 30 min stimulation reduced the c A M P response to 49% of control [177]. Further work in insulinoma (3TC-3 cells, found evidence for PKA-independent homologous desensitization of GLP-1-s t imulated calc ium transients, and that P K C activation reduced calc ium responses by 75%, but that PKC-independent pathways were also involved [362]. Fehmann et al [363] suggested that G L P - 1 receptor responsiveness was determined by surface expression; a reduced G L P - 1 binding capacity was observed in R I N m 5 F cells with G L P -1 treatment, and on a longer time-scale, chronic treatment with P K A activators reduced surface receptor expression and m R N A stability. Internalization of 1 2 5 I - G L P - 1 has been shown for insu l inoma and transfected cells [364; 365]. G L P - 1 R homologous desensitization and internalization were correlated with specific phosphorylation of C-terminal tail serine doublets, but differential quantitative impairment of desensitization or internalization in mutant receptors indicated different ce l lu lar mechanisms cont ro l l ing these processes [366]. Howeve r , heterologous desensitization of the G L P - 1 receptor with phorbol esters s imilar ly involved phosphorylation of serine doublets in the intracellular tail of the receptor [367; 368]. L i k e the glucagon receptor, involvement of G R K s in desensitization and/or receptor endocytosis is unknown. O n ini t ia t ing work for the current thesis, only two reports regarding G I P receptor desensitization existed. The first reported a single experiment demonstrating homologous desensitization of GIP-stimulated insulin release from perifused HIT-T15 cells, as part of a larger study of G L P - 1 receptor regulation [177]. The other reported chronic desensitization of the G I P receptor in diabetic rats resulting from elevated serum levels of G I P in these animals [369]. Unfortunately, experimental flaws in this latter report prevent the extraction of results it claimed to report. Preliminary work on G I P receptor mutants with C-terminal truncations indicated that the C-terminus played a role in receptor sequestration [290; 324]. Dur ing preparation of our own 62 reports on G I P desensitization in (3TC-3 cells [370], internalization of G I P receptor mutants in transfected cells [290], and further studies contained in the present thesis, Tseng and Zhang released a series o f publicat ions impl ica t ing C- terminal residues (S406 and C411) in desensitization and surface expression [291], and the involvement of regulator of G protein signalling ( R G S ) proteins [371] and G protein-receptor kinases [372] in the attenuation of G I P receptor activation. 3.1.3 Potential Physiological Relevance of GIP receptor Desensitization Studies in insulin-resistant type 2 diabetic patients, suggested that there was a reduced incretin effect in the disease [373]. Subsequent results have indicated that alterations in both G I P and G L P - 1 are involved: a secretory defect appears to result in diminished G L P - 1 secretion, while the responsiveness of the (3-cell to G I P is blunted [155; 374]. A s such, physiological concentrations of exogenous G L P - 1 can still act as a potent antidiabetic agent, whereas G I P is much less potent [220; 375-378]. Poor preparations o f early batches o f synthetic G I P contributed to the reduced potency of G I P in both normal and type 2 diabetic patients [92; 276]. Regardless, there is little doubt that there is a blunted response to G I P by the (3-cell in type 2 diabetes. The root of the problem is difficult to ascertain, considering that G f P levels in diabetics have been reported to be elevated, normal, or low, however, these differences are l ikely due to complications with radioimmunoassay of G I P [52; 53]. Recently, it was proposed that the blunted responsiveness to G I P in type 2 diabetes patients may be due to decreased expression of the G I P receptor [374]. When testing this hypothesis in an animal model of type 2 diabetes, the Vancouver diabetic fatty ( V D F ) rat, L y n n and colleagues [379] discovered a reduction in G I P m R N A and protein expression in isolated islets, and correlated the diminished expression to decreased ability to stimulate'cyclic A M P and insulin release from islets, and inability of G I P to reduce glycemic excursions in vivo when infused at physiological concentrations. 63 3.1.4 Thesis Objective In the current study, homologous desensitization and internalization of the G I P receptor were characterized, as it is possible that these two cellular processes are involved in the diminished incretin effect observed in type 2 diabetes. W i t h this goal in mind, 6 T C - 3 cells (Methods 2.3) were employed to examine regulat ion o f GIP-s t imula ted c y c l i c A M P by glucose and homologous desensitization of endogenously expressed receptors (Methods 2.7), as wel l as desensitization of insulin release (Methods 2.8). W i l d type and mutant (Methods 2.2) G I P receptor transfected C H O - K 1 cel ls (Methods 2.3) were used to further characterize desensitization (Methods 2.7) and internalization by radioactive (Methods 2.9) or fluorescent means (Methods 2.10). Furthermore, using pharmacological inhibitors of signal transduction molecules, cellular processes involved in desensitization of GIP-stimulated c A M P in (3TC-3 cells (Methods 2.7) and internalization in transfected cells was accomplished (Methods 2.9). 3.2 Results 3.2.1 Desensitization of pTC-3 Cells to GIP Possible mechanisms of desensitization to G I P were studied using (3TC-3 cells as a (3-cell model. G I P receptors have been demonstrated to be expressed in this cell line, and stimulation of insulin release by G I P is glucose-dependent [178]. C y c l i c A M P production in response to G I P was found to increase in a concentration-dependent manner (Figure 5 A - D ) , and was only slightly affected by glycemic conditions. The maximal cycl ic A M P produced in response to G I P was moderately blunted with increasing ambient glucose conditions (Table 3), while sensitivity to G I P was not affected ( E C 5 0 values: 0 glucose: 12.5 ± 4.8 n M ; 5.5 m M glucose: 9.5 ± 2.0 n M ; 11 m M glucose: 10.9 ± 4 . 4 n M ) . A t each glucose concentration, however, the desensitized c A M P response was similar fol lowing pretreatment of cells with 100 n M G I P for 1 hour (Table 3). C y c l i c A M P production stimulated by 10 u M forskolin was neither altered by glycemic 64 condi t ion nor pretreatment with 100 n M G I P for 1 hour (Table 3). A time-course for desensitization of GIP-stimulated c A M P indicated a mi ld but significant reduction in c A M P production could be observed with 10 min of 100 n M G I P pretreatment, and the apparent desensitization was more pronounced with increasing duration of pretreatment (Figure 6 A ) . Table 3: Summary of cyclic AMP production in f$TC-3 cells in response to GIP and forskolin Data represent mean ± S . E . M . of 3 independent experiments. Cel ls were incubated with G I P (320 pM-1 p M ) or 10 p M forskolin for 30 minutes in the presence of 0.5 m M I B M X , and c A M P was measured by R I A (see Figure 5). Refer to Methods section 2.3 and 2.1 for more detail. Glucose Control 100 n M GIP-Pretreated* (mM) (pmol c A M P / w e l l ) (pmol c A M P / w e l l ) M a x i m a l G I P Response 0.0 70.5 ± 7.9 52.4 ± 10.3 5.5 66.3 ± 8 . 1 55.7 ± 9 . 1 11.0 55.2 ± 7 . 8 44.4 ± 7.6 10 p M Forskolin 0.0 331.0 ± 4 0 . 8 347.1 ± 2 9 . 1 5.5 346.3 ± 16.8 328.3 ± 2 5 . 8 11.0 337.3 ± 30.0 342.6 ± 36.6 *: Cells were prestimulated with 100 n M GIP for 1 hour, followed by a 10 min washout period Previous studies have reported failure of G L P - 1 or G I P to stimulate c A M P production in the absence of glucose in rodent insulinoma cell lines when performed in Kreb ' s Ringer solution [176; 177]. Upon finding that G I P was able to produce a concentration dependent production of c A M P in |3TC-3 cells in zero glucose D M E M (Figure 5 A , B ) , experiments were repeated in K R B to rule out the involvement of more complex constituents. G I P (100 n M ) and forskolin (10 p M ) yielded similar responses in the presence of either 0 m M or 5.5 m M glucose, respectively (P > 0.05), and were similar in magnitude to experiments performed in D M E M (data not shown). Degradation o f extracellular l igand is a we l l accepted mechanism for attenuation of stimulation. Induction of dipeptidyl peptidase I V , the primary enzyme responsible for incretin inactivation [58; 59; 380], could contribute to the desensitized state observed in |3TC-3 cells with static incubations. Thus, the potent reversible non-hydrolysable transition state inhibitor, 65 isoleucine-thiazolidide [381; 382] was employed to test this hypothesis. Ile-thia did not affect basal, stimulated or desensitized G I P induced cycl ic A M P responses (Figure 6B) . Dur ing the preparation of this thesis, Kesper et al [383] found that chronic exposure of INS-1 cells to G I P resulted in induction of G u i G-proteins. A n increase of inhibitory G-protein complement could also diminish the c A M P response to G I P . However , pertussis toxin had no effect on basal, stimulated or desensitized GIP-induced cycl ic A M P responses in (3TC-3 cells (Figure 6C) . Prior work examining desensitization of the glucagon receptor has suggested that alterations in phosphodiesterase ( P D E ) activity could also lead to an apparent reduction in c A M P production by hormones [384]. It was hypothesized that, i f P D E were modulated by pretreatment with G I P , then a change in sensitivity to I B M X would be observed relative to control cells . T o test the possibi l i ty of modulation o f phosphodiesterase activity in the desensitized state, the effect of I B M X concentration was tested in both the control and G I P -pretreated conditions (Figure 6D) . Intracellular cyc l ic A M P levels increased with increasing concentration of I B M X over the range tested, however, maximal inhibition of phosphodiesterase was not achieved. When c A M P was normalized to the maximal value observed in either control or desensitized state, no alteration in sensitivity to I B M X was observed. Higher concentrations of I B M X would probably result in non-specific effects. Signal transduction kinases P K C and P K A have been shown to contribute to receptor desensitization in other hormone models. These kinases can form a negative feedback loop whereby surface receptors are phosphorylated, thus blunting their response to further stimulation. Broad specificity inhibitors o f P K C (staurosporine) and P K A (H89) were used during the pretreatment period to block this feedback, i f present (Figure 6 E and F) . Staurosporine acted to significantly reduce the basal c A M P concentrations in non-stimulated cells (P < 0.05) and to enhance GIP-stimulated c A M P production (1.9-fold; P < 0.01), but did not inhibit the attenuation 66 of the c A M P response by pretreatment with G I P . In contrast, H 8 9 had no effect on basal or stimulated c A M P values, or desensitization of the c A M P response to GIP . The degree of glucose stimulated insulin release from (3TC-3 cells in K R B was similar to that previously observed [178], with significant insulin released by 5.5 and 11 m M glucose (163 ± 16 % and 182 ± 21 % basal release in 0 glucose, respectively; p < 0.05 and p < 0.01 versus basal) (Figure 7 A ) . In the absence of glucose, G I P failed to stimulate insulin release (102 ± 10 % basal), but forskolin stimulated a small but significant release (140 ± 18 % basal; p < 0.05). In the presence of elevated glucose, 10 n M G I P potentiated insulin release (5.5 m M glucose: 223 ± 26 % basal; 11 m M glucose: 247 ± 28 % basal; p < 0.05), as d id 10 u.M forskolin (5.5 m M glucose: 397 ± 17 % basal; 11 m M glucose: 472 ± 26 % basal; p < 0.001) (Figure 7 A ) . In contrast to the weak effect of G I P pretreatment on cycl ic A M P production, pretreatment of cells with 100 n M G I P completely abolished the potentiating effect of G I P on insulin secretion under elevated glucose conditions (5.5 m M glucose: 113 ± 4 % basal; 11 m M glucose: 101 ± 3 % basal; p < 0.001 versus euglycemic controls). Furthermore, while G I P pretreatment yielded no effect on forskolin-stimulated c A M P production, the same treatment significantly reduced forskolin-stimulated insulin release to values roughly half the magnitude of euglycemic controls (5.5 m M glucose: 168 ± 17 % basal; 11 m M glucose: 218 ± 14 % basal; p < 0.001 versus euglycemic controls). 67 A . Glucose Comparison B . 0.0 mM Glucose 30-, c .5 .2 o o m 20-•o 0 0 U-CL 5 0. =*-S o 10-< *~ O 0 -Basal 0 Glucose 5.5 Glucose 11 Glucose -9 30-c .E .2 o o % Z 5 2 0 o u. CL 2 CL =*-5 o 10-< ' r \" • Control o 100 nM GIP Pre-stim Log 1 0 [GIP] Basal -9 -8 -7 Log 1 0 [GIP] „ 30-c .= .2 o •5 £ 20-T3 O O LL w Q- 5 CL I 2 o 10-< Basal 5.5 mM Glucose Control 100 nM GIP Pre-stim -8 Log 1 0 [GIP] D. —. 30-O w o u. CL 5 CL --5 O 10-< T~ 11 mM Glucose • Control o 100 nM GIP Pre-stim Basal -9 -8 -7 Log 1 0 [GIP] -6 Figure 5: Effect of glycemic conditions on cyclic AMP production in (3TC-3 cells (A) Comparison of GIP-stimulated c A M P production under various glucose conditions. ( B - D ) Effect of pretreating cells with 100 n M G I P for 60 min prior to performing concentration-response studies. Glucose had no effect on forskolin-stimulated c A M P production. Data represent the mean ± S . E . M . of 3 independent experiments. Refer to Methods sections 2.3 and 2.7 for more detail. 68 A. GIP Desensitization Timecourse B. Effect of DPIV Inhibition I I Basal ^ 1 0 nM GIP Control 10 Min 30 Min 60 Min Duration of GIP Pretreatment 100-50-g o T3 O CL a. < o 10 nM GIP 50 p.M lle-Thia Control mm Desensitized c o C. Effect of Pertussis Toxin • • Control ESSS Desensitized D. Effect of IBMX 10nMGIP 500 ng/mL PTX --4.2 -3.9 -3.6 -3.3 -3.0 -2.7 L o g 1 0 [IBMX] c o 200-IOOH •a ~ n2 O 0- o Q. ^ < o 10 nM GIP 100 nM Stauro E. Effect of Staurosporine • Control SSI Desensitized F. Effect of H89 c o = f °- o 0_ • • Control E H 3 Desensitized Figure 6: Time-course of homologous desensitization of GIP-stimulated cAMP production and effect of various inhibitors on desensitization in (5TC-3 cells (A) Cel ls were pretreated with or without 100 n M G I P for indicated times, followed by a 10 min washout period and subsequent stimulation with 10 n M G I P . Subsequent figures show G I P -stimulated c A M P production with or without pretreatment with 100 n M G I P for 60 min , fol lowed by a 10 min washout period and subsequent stimulation with 10 n M G I P . 0.5 m M I B M X was used during the stimulation period, except where indicated. (* = p < 0.05). (B) Effect of isoleucine-thiazolidide (Ile-thia), a potent D P I V inhibitor, administered during the stimulation period. (C) Effect of pertussis toxin (100 ng /mL during washout period and 500 n g / m L during stimulation) on GIP-st imulated c A M P production. (D) Effect of I B M X on apparent desensitization of GIP-stimulated c A M P production. Data were normalized to the response observed in the control or desensitized state at 4 m M I B M X . Inset shows the same data 69 as pmol c A M P / w e l l . N . B . I B M X concentration is plotted using a logarithmic scale. (E and F) Effect of 100 n M staurosporine or 5 u M H89 on homologous desensitization of GIP-stimulated c A M P formation, administered 15 min prior to and during the G I P pretreatment. (a = p < 0.05, b = p < 0.01). Data represent the mean ± S . E . M . of 4 independent experiments. Experiments shown here were in the presence of 5.5 m M glucose. Refer to sections 2.3 and 2.7 for specific methods. B 0 Glucose 5.5 mM Glucose 11 mM Glucose 0 Glucose 5.5 mM Glucose 11 mM Glucose Figure 7: Insulin release from pTC-3 cells ( A ) Comparison of glucose, G I P and forskolin-stimulated insulin release. (B) The effect of pretreating cells with 100 n M G I P prior to 10 n M G I P stimulation. (C) The effect of pretreating cells with 100 n M G I P prior to 10 m M Forskolin stimulation. Data represent mean ± S . E . M . of at least 4 independent experiments. (* = P < 0.05, ¥ = P < 0.01, § = P < 0.001). Statistical comparisons of G I P and forskolin-stimulated insulin release versus euglycemic controls were analyzed, whereas glucose-stimulated insulin release was compared to basal release in the absence of glucose. Refer to sections 2.3 and 2.8 for specific methods. 70 3.2.2 Desensitization of the Transfected GIP Receptor Experiments designed to examine the regulation of the G I P receptor were continued using a transfected ce l l model . Unfortunately, transfected cel l models are usually over-expression systems, which complicated demonstration of G I P R desensitization. Fo l lowing work presented by the research group of Thorens on the G L P - 1 receptor [366], only low expressing clones of G I P receptor transfected C H O - K 1 cells were able to be desensitized. Subcloning a pooled clone of w i l d type G I P receptor transfected cells yielded two cell lines, r G I P R - L 2 and r G I P R - L 1 3 , which had sufficiently low receptor expression to allow receptor desensitization studies to be performed. Saturation binding (Figure 8) on transfected cells indicated maximum binding values (Bmax) of 76.5 ± 16.9 cpm/1000 cells (wtGIPR) , 18.6 ± 6.7 cpm/1000 cells ( r G I P R - L 2 ) and 4.35 ± 2.93 cpm/1000 cells ( r G I P R - L 1 3 ) , and K d values between 208-838 p M . N o further studies beyond preliminary desensitization and saturation binding were done on r G I P R - L 1 3 due to the intrinsic variability of responses with this extremely low expressing cell line. Figure 9 shows a concentration-response curve of G I P i _ 4 2 (with I B M X ) on w t G I P R and r G I P R - L 2 cells with or without a 60 minute pretreatment with 100 n M G I P in the absence of I B M X . The normal G I P curve for w t G I P R cells displayed an E C 5 0 of 22.6 ± 3.8 p M , and maximal cyc l i c A M P accumulation was 255 ± 54 fmol/1000 cells; it appeared that the 100 n M G I P could not be washed off in 10 min in the case of pretreatment of w tGIPR cells, as a constant high cycl ic A M P production level persisted even in the absence of G I P in the medium (Figure 9). In contrast, r G I P R - L 2 cells were only able to maximal ly increase intracellular cyc l ic A M P to 38.4 ± 5.0 fmol/1000 cells in response to G I P , which was reduced to 25.2 ± 1.6 fmol/1000 cells after G I P pretreatment (P < 0.05). However , a small but significant persistent elevation of basal cyc l ic A M P was noted in the pretreated cells: 2.85 ± 0.12 fmol/1000 cells (control) versus 8.98 ± 0.55 fmol/1000 cells (pretreated) (P < 0.05), indicating that it was not possible to wash away all of the residual G I P (Figure 9). Half-maximal cycl ic A M P stimulation by G I P occurred at 607 ± 57 p M 71 G I P in the case of the control, whereas the E C 5 0 for cells pretreated with 100 n M G I P for 60 min prior to the concentration-response curve was significantly right shifted: 1.27 ± 0.38 n M GIP. Some investigations have used cyc l i c A M P accumulation time-course experiments as indicating desensitization, as cycl ic A M P levels plateau. Others have argued that the plateau may arise from overcoming the I B M X blockade of c A M P degradation by phosphodiesterase due to such high levels of cycl ic nucleotide. Wi th this in mind, a 1 n M GIP-stimulated cycl ic A M P time-course experiment was performed in w tGIPR cells with and without 0.5 m M I B M X (Figure 10). In both cases, plateau levels of cyc l ic A M P were approached by 15 min. Repetitive stimulation experiments were performed on r G I P R - L 2 cells. In a similarly designed experiment, a time-course of c A M P accumulation in r G I P R - L 2 cells in response to 1 n M G I P (in the presence of 0.5 m M I B M X ) was measured, with and without a 60 min 100 n M G I P pretreatment (in the absence of I B M X ) (Figure 11 A ) . In control cells, c A M P levels reached a plateau by 30 min, despite the fact that r G I P R - L 2 cells displayed much lower maximal GIP-stimulated c A M P production than w t G I P R cells ( r G I P R - L 2 : 38.4 ± 5.0 versus wtGIPR: 255 ± 54 fmol/1000 cells; Figure 9). When r G I P R - L 2 cells were pretreated with 100 n M G I P for 60 min (+ 10 min washout), during the subsequent 1 n M G I P stimulation, the plateau was simply reset to a lower level, approximately 63% of the maximal control value (Figure 11A). T o establish the time scale during which receptor desensitization was occurring, r G I P R - L 2 cells were prestimulated for various times (60, 25 and 10 min) with 100 n M G I P in the absence of I B M X , followed by a washout period, and a 30 min stimulation in the presence of I B M X with or without 1 n M G I P (Figure 1 IB) . G I P receptor desensitization in transfected C H O - K 1 cells was slow and moderate, with a profile similar to that observed in (3TC-3 cells (Figure 6). A series of nine G I P receptor mutants was generated (refer to Figure 2) to examine the role of potential phosphorylation sites in receptor sequestration (using high expressing pooled cel l l ines). A t the same time, two mutant lines were selected to subclone for examination o f 72 desensitization in low expressing cell lines. In the standard 60 min/100 n M G I P prestimulation protocol, receptors with a serine doublet mutated to alanines ( rGIPR-S426/427A) desensitized to a similar degree as w i ld type receptors, ~72% of control, whereas a receptor mutant with all six C-terminal serines mutated to alanines (rGIPR-S398/406/426/427/440/453A) still displayed 94% of control c A M P production. These results indicate that specific serines in the C-terminal tail permit receptor desensitization, l ikely v ia phosphorylation, but the residues are unlikely to be o-l 1 1 1 1 1 0 250 500 750 1000 1250 1 2 5 I-GIP (pM) Figure 8: A representative saturation binding curve for wtGIPR cells Cel ls were incubated in the presence of l 2 5 I - G I P with or without 1 p M G I P at 4 ° C for 12-16 hours. See Methods sections 2.2 and 2.6 for details. 73 A. wtGIPR 400n i i i i i i r r 0 -12 -11 -10 -9 -8 -7 -6 L o g 1 0 [GIP] B. rGIPR-L2 50n C o w I 1 1 1 1 1 1 r 0 -12 -11 -10 -9 -8 -7 -6 L o g 1 0 [GIP] Figure 9: Effect of GIP prestimulation on the concentration-response curve of wtGIPR and rGIPR-L2 cells Cells were pretreated with 100 nM GIP in the absence of IBMX for 60 min in buffer containing 1% Trasylol at 37°C. Cells were washed twice over 10 min with warm buffer, followed by a concentration-response experiment with varied GfP concentrations in the presence of 0.5 mM fBMX. Data represent mean ± S.E.M.; n = 3-4. See sections 2.2, 2.3, and 2.7 for detailed methods. 74 I 1 1 1 1 1 1 -0 5 10 15 20 25 30 Time (Min) Figure 10: Time-course of cAMP accumulation in wtGIPR cells with and without IBMX Cells were preincubated in 37°C buffer for 60 min, prior to adding 1 nM GIP to wells with without 0.5 mM IBMX in buffer containing 1% Trasylol. Data represent mean ± S.E.M.; n = See sections 2.2, 2.3, and 2.7 for detailed methods. A _^ 100-o c 80-o o 60-Q. 40-< 20-o 0-B 100-, trol) 80-c o 60-o CL 40-< o 20-0-- o - Control - • - Desensitized i — i — i — i — 0 5 10 15 i— 30 \"eio Time (min) Basal Desensitized 10 25 Time (Min) Figure 11: Time-course of desensitization in rGIPR-L2 cells (A) Cel l s were preincubated in 37°C buffer (with 1% Trasylol) for 60 min with or without 100 n M GIP , prior to a 10 minute washout period, and adding 1 n M G I P to wells with 0.5 m M I B M X for various times. (B) Cel ls were preincubated for various times with 100 n M G I P (no I B M X ) , prior to washout and subsequent stimulation with 1 n M G I P for 30 min; \"Basa l \" refers to cells which were incubated in the absence of G I P during the 30 min stimulation period. Data represent mean ± S . E . M . ; n = 4. See sections 2.2, 2.3, and 2.7 for detailed methods. 76 A. Wild Type Rat GIP Receptor Basal Control Des. Bas Des. Test B. rGIPR-S426/427A Basal Control Des. Bas Des. Test C. rGIPR-CT Serine Knockout Basal Control Des. Bas Des. Test Figure 12: Desensitization of C-terminal mutant GIP receptors L o w expressing subclones of C H O - K 1 cells transfected with (B) GIPR-S426 /427A or (C) G I P R -S398/406/426/427/440/453A were compared to control low expressing w i ld type G I P R subclone r G I P R - L 2 (A) , when treated with 100 n M G I P for 60 min, prior to a 10 min washout and subsequent 30 min stimulation in the absence (Des. Bas.) or presence (Des. Test) of 1 n M G I P in the presence of I B M X . Data represent mean ± S . E . M . ; n = 4. See sections 2.2, 2.3, and 2.7 for detailed methods. 77 3.2.3 Internalization of the Transfected GIP Receptor 3.2.3.1 Development of an Internalization Protocol Efforts were made to develop a protocol to measure receptor internalization. O f several methods attempted, the one with the most success involved adding 100 n M G I P at various time-points to w t G I P R cells (or mutants), stripping off surface bound peptide with a hypertonic acidic solution, washing wi th buffer, and then performing a binding experiment to measure the remaining surface receptors (Methods section 2.9). To validate the internalization protocol, it was necessary to ensure that receptor binding affinity was not affected by G I P treatment or the acid stripping protocol. B ind ing to w t G I P R cells was allowed to proceed for 4 hours at 4 ° C , as this achieved near equi l ibr ium (Figure 13); acid treatment alone reduced total binding (B 0 ) to 64.0 ± 1.6% of control values, and 100 n M G I P for 1 hour, prior to the acid wash resulted in a B 0 of 36.5 ± 2.7% of control (Figure 14). Bind ing affinity was not significantly affected by G I P pretreatment or acid stripping procedure ( IC 5 0 values: 3.64 ± 0.26 n M , control untreated; 5.73 ± 0.55 n M , acid alone; 5.87 ± 1 . 1 1 n M , G I P pretreatment and acid wash), validating the use of B 0 as an estimate of B m . l x in these studies (Eq. 2, Methods section 2.6). Non-specific binding was sl ightly, but not significantly elevated in both acid treated groups (% of total label added: control, 0.85%; acid alone, 1.27%; internalized, 1.09%), and was on the order of 400-600 cpm. Hence, from this control experiment, the estimated internalization caused by 60 min treatment with 100 n M G I P was a ~27.5% loss of surface binding. 78 120n Time (min) 6' ' 30 50 100 150 200 240 Time (min) Figure 13: Time-course of 125I-GIP binding to wtGIPR cells GIP tracer (50,000 cpm/weli) was incubated with attached wtGIPR cells in 24 well plates for indicated times at 4°C. Wells were washed twice with cold buffer, prior to solubilization with 0.1 M NaOH and transfer to borosilicate tubes for counting. Non-specific binding was measured at each time-point in wells containing excess unlabelled GIP. Data represent mean ± S.E.M., normalized to the maximal specific binding observed at 4 h; n = 4. Refer to Methods sections 2.3, 2.5 and 2.6. 79 A . • 60 min Internalized i 1 1 1 1 1 1 r 0 -12 -11 -10 -9 -8 -7 -6 Log [GIP] B . i 1 1 1 1 1 1 r 0 -12 -11 -10 -9 -8 -7 -6 Log [GIP] Figure 14: Effect of acid stripping alone or in combination with 60 min 100 n M GIP pretreatment on GIP receptor binding on wtGIPR cells Acid stripping (5-10 min with 150 mM NaCl/50 mM glycine, pH 3) is commonly performed to remove surface bound non-internalized peptide. This experiment was designed to establish whether acid stripping had any effect on receptor expression or affinity alone. (A) Binding data normalized to B0 of control untreated cells. (B) Binding data normalized to B 0 of each condition. Data represent mean ± S.E.M.; n = 4. The control is a normal competitive-binding inhibition curve with non-treated cells. Refer to Methods sections 2.3, 2.5, 2.6, and 2.9. 80 3.2.3.2 Agonist and Antagonist Stimulated Receptor Internalization Further characterization of G I P receptor internalization continued with a sequestration time-course in response to 100 n M G I P or the G I P receptor antagonist (see Chapter 4, section 4.2.1, cf. Figure 34), 1 p M G I P 7 . 3 0 N H 2 (Figure 15). Six ty min o f G I P , . 4 2 treatment resulted in internalization of 38.6 ± 1.1% of the surface receptors, whereas a ten-times greater concentration of GIP 7 . 3 O N H 2 resulted in the internalization of 18.8 ± 3 . 1 % (Figure 15A). The threshold for GIP , . 4 2 stimulated loss of surface binding was between 10 and 20 n M , and above this concentration, concentration-dependent internalization was observed (Figure 15B). G I P 7 . 3 0 N H 2 did not stimulate measurable c A M P production at concentrations as high as 10 p M , and exhibited approximately a 10-fold lower binding affinity, when compared to native G I P (Chapter 4, section 4.2.1, cf. Table 7). Significant GIP 7 . 3 0 N H 2 - s t imula ted loss of surface receptors was observed at a concentration of 1 p M , and at 10 p M , this peptide caused a 72.4 ± 1.3% reduction in surface binding. The concentration at which 50% of surface receptors were lost was between 0.5 and 1 p M for G I P M 2 , and 5 and 7.5 p M for G I P 7 . 3 0 N H 2 . Hence, it would appear that ligand binding by the receptor is a more important determinant of internalization than receptor activation, with half-maximal internalization potencies correlating wel l with binding affinities. 81 A. Agonist and Antagonist Stimulated Receptor Internalization B. Agonist and Antagonist Internalization Dose Response m o o o o o o o T - m m m o [Peptide] Figure 15: GIP receptor internalization by agonist and antagonist (A) wtGIPR cells were treated for the indicated times with 100 nM GIPi_42 or 1 uM GIP7_30NH2, prior to acid stripping and measurement of remaining surface binding as per methods described in the text. (B) wtGIPR cells were treated with increasing concentrations of GIP,.42 or GIP7.30NH2 for 60 minutes, prior to acid stripping and surface expression determination. Data represent mean ± S.E.M.; n = 4. Refer to Method section 2.9 for more details. 82 3.2.3.2 Role of Signal Transduction in GIP Receptor Internalization Elucidation of the pathways necessary for internalization required various pharmacological inhibitors. Hypertonic sucrose has been used to block clathrin-coated pit-mediated endocytosis, as has concanavalin A (a glycoprotein binding lectin) [385]. Treatment of wtGIPR cells with 0.5 M sucrose and 100 nM GIP time dependently increased surface receptor binding (47.6 ± 7.3% at 60 min); this was not due to sucrose alone, as all cells were treated with sucrose for the duration of the experiment, and GIP was added at the various time-points (Figure 16). It was hypothesized that GIP was stimulating a receptor recycling pathway, hence monensin, an inhibitor of vesicular transport acting to increase intravesicular pH [386; 387], was inlcuded in conjunction with sucrose and GIP, and reversed the increase in surface binding (Figure 16). In contrast, when cells were treated with just monensin and GIP, the degree of receptor internalization was not significantly different from cells treated with GIP alone (60 min GIP treatment: 39.8 ± 2.4% versus GIP/monensin treatment: 34.8 ± 9.5% loss of surface binding), suggesting GIP stimulation of the endosomal recycling pathway is very minor (Figure 16). Preliminary experiments indicated that treatment of cells with concanavalin A also blocked GIP receptor internalization (data not shown). While both GIP receptor agonists and antagonists were able to cause receptor internalization, the role of receptor signalling was probed using cyclic AMP pathway activators and blockers, as well as phospholipase C/protein kinase C modulators. Table 4 shows the effect of these inhibitors on basal,1 nM and 100 nM GIP-stimulated cAMP production; the 1 nM concentration was chosen to establish if the agents had an effect on the GIP concentration-response curve, and 100 nM to compare to the results of the internalization studies. A 45 minute treatment of cells with adenylyl cyclase inhibitors (MDL-12,330 and 2',5'dideoxyadenosine, DDA), a protein kinase A blocker (H89), or an activator (phorbolmyristic acid) or inhibitor (staurosporine) of protein kinase C, had no measurable effect on basal cAMP levels in wtGIPR cells. Only the 83 adenylyl cyclase activator, forskolin, was able to increase cAMP on its own (Table 4). Protein kinase A and C appeared not to affect GIP-stimulated cAMP greatly, but PMA was able to double the cAMP response to 100 nM GIP. Adenylyl cyclase inhibitors were able to significantly reduce both 1 nM and 100 nM GIP-stimulated cAMP formation, although inhibition ranged between 22 and 60%. 10 \\iM forskolin is often used as an agent to demonstrate maximal cyclic AMP production. Here, 1 nM or 100 nM GIP combined with 10 uM forskolin, more than doubled the cAMP response to forskolin alone, and the two agents were not acting in an additive manner, but rather synergistically (Table 4). Table 4: Effect of signal transduction cascade activators/inhibitors on basal and GIP-stimulated cAMP accumulation in wtGIPR cells Cells were treated with pharmacological agents 15 min prior to addition of GIP. Peptide stimulation for 30 minutes followed; data are expressed as picomoles cyclic AMP/well (* = P < 0.05; n > 3, shown in brackets). Basal 1 nM GIP 100 nM Control\" 1.67 ±0.20 (12) 65.0 ± 10.5 (6) 105.6 ±5.8 (6) 100 uM MDL-12,330 1.01 ±0.20 (3) 26.4 ±2.7* (3) 79.2 ±7.5* (3) 100 uM2',5' DDA 1.12 ± 0.13 (3) 50.5 ±4.7* (3)' 76.5 ±7.7* (3) 10 uM Forskolin 246.0 ±51.1* (3) 531.9 ±28.3* (3) 558.4 ±64.9* (3) 5 uM H89 1.55 ±0.21 (5) 74.1 ±5.2 (5) 122.6 ±8.5 (5) 400 nM PMA 1.73 ±0.66 (3) 57.4 ± 10.3 (3) 201.4 ± 12.7* (3) 100 nM Staurosporine 1.40 ±0.21 (3) 70.1 ±5.2 (3) 117.5 ±5.1 (3) A l l measurements were made in the presence of 0.5 m M I B M X , a concentration that completely blocks phosphodiesterase activity, thus values represent c A M P production over the 30 minute stimulation period. For comparison, control values without I B M X were 0.49 ± 0.01 (basal), 20.4 ± 6.9 (1 nM GIP) and 60.5 ± 1.3 (100 nM GIP), n = 3. 84 Table 5: Effect of signal transduction cascade activators/inhibitors on GIP receptor expression and internalization in wtGIPR cells Cells were treated with pharmacological agents 15 min prior to addition of GIP. Stimulation for 30 minutes followed; data are expressed as % control receptor expression (* = P < 0.05; n = 4-6). The % loss of surface binding is shown in brackets versus appropriate control. Agent: 45 min Agent Alone 30 min 100 nM GIP Agent + GIP 100 pMMDL-12,330 8.2 ±0.8* 78.9 ±3.7 (21.1) 15.9 ±0.6* (84.1a) 100 pM2',5' DDA 75.5 ±2.4* 79.2 ± 3.4 (20.8) 70.6 ± 2.5* (29.4a) 500 pM IBMX 105.8 ±2.7' 69.9 ±0.8 (30.1) 78.9 ±4.3 (26.9) 10 pM Forskolin 103.9 ±5.4 74.5 ± 3.3 (25.5) 78.6 ±4.0 (25.3) 5 pM H89 97.4 ±2.1 73.9 ±2.5 (26.1) 77.6 ± 1.8 (19.8) 400 nM PMA 74.6 ±4.9* 71.1 ±0.8 (28.9) 63.8 ± 2.8* (36.2a) 100 nM Staurosporine 112.4 ± 1.8* 78.7 ± 1.7 (21.3) 82.3 ± 4.4 (17.7a) A l l measurements were made in the absence of 0.5 m M I B M X , except where noted. a : Agent alone caused changes in receptor expression; internalization difference is versus control without agent. Effect of pharmacological agents on receptor expression and sequestration followed a similar protocol to that employed for examining effects of agents on basal and GIP-stimulated cAMP production. Briefly, experiments consisted of four study groups: (1) control surface expression (i.e. untreated cells), which was used to normalize other data, (2) cells treated with 100 nM GIP for 30 min, i.e. \"control internalized\", (3) cells treated with pharmacogical agent for 45 minutes to see the effect of the agent on receptor expression alone, and (4) cells treated with agent for 45 minutes and with GIP for 30 min (i.e. GIP added 15 min after agent), to see the effect of the agent on GIP-stimulated receptor endocytosis. In all four groups, cells were acid stripped and surface receptor binding measured as described in Methods section 2.9. Cyclic AMP raising agents, IBMX and forskolin, alone appeared to slightly (but not significantly) increase cell surface binding but, in combination with GIP, did not affect the degree of receptor internalization after 30 min (~25-30%; Table 5). In contrast, when cells were treated with inhibitors of adenylyl cyclase alone, GIP receptor surface expression was dramatically reduced (MDL-12,330: 91.2 ± 0.8% reduction; 2',5'-DDA: 24.5 ± 2.4% reduction). When GIP was administered in combination with adenylyl cyclase inhibitors, it did not produce a greater effect, however, as the inhibitors had effects of their own, it makes interpretation of these data very difficult (Table 5). 85 These data may suggest that adenylyl cyclase inhibitors may also be able to stimulate sequestration of the enzyme, and that the enzyme is in close proximity with the GIP receptor, so that they are co-internalized, or alternatively, this result may be an artifact of receptor overexpression in these cells. PKA inhibition did not alter surface receptor expression alone, and did not affect GIP-stimulated receptor endocytosis (Table 5). Activation of PKC with phorbol esters alone caused a 25.4 ± 4.9% reduction in surface receptor expression, and in combination with GIP, yielded a 36.2 ± 2.8% loss (P < 0.05; Table 5), indicating PKC is able to stimulate receptor internalization, and is additive with that stimulated by GIP. PKC inhibition with staurosporine, however, did not alter the degree of loss stimulated by GIP (GIP control: 21.3 ± 1.7%; GIP + staurosporine: 17.7 ± 4.4%); staurosporine alone appeared to moderately increase surface receptor expression (12.4 ± 1.8%). Together, these results indicate that PKC can be a negative modulator of GIP receptor expression, and when activated by other pathways, can contribute to GIP-stimulated receptor endocytosis. A short study was undertaken to characterize ligand-independent GIP receptor internalization further. Internalization time-courses and dose-response relationships for MDL-12330 and PMA are shown in Figure 17. MDL treatment resulted in a rapid internalization profile that reached a nadir between 25 and 60 min; the effect of PMA was latent and much less pronounced, and did not significantly reduce surface receptor expression until 45-60 min of treatment. Both drugs appeared to have concentration dependent effects, with PMA causing significant internalization at 4 nM, and MDL at 10 uM (Figure 17). 86 Time (min) Figure 16: Effect of sucrose and monensin on receptor internalization in wtGIPR cells Cells were pretreated with 0.5 M sucrose or 100 pM monensin 15 minutes prior to addition of GIP. The experiment was designed such that sucrose and/or monensin were present with the cells for the duration of the experiment, and the only variable was time exposed to GIP; control cells are treated with GIP alone (in the absence of sucrose or monensin). Cells were acid stripped and cell surface receptor expression was estimated by radioligand binding (see Methods 2.9). Data represent mean ± S.E.M.; n = 3-4. 87 c Is oo £ f/j o a 20 « § ° * 10H m 1 1— 0 5 10 15 —I— 30 45 Time (min) —i 60 G I P - R - 4 5 5 G I P - R - S 3 9 8 A G I P - R - S 4 0 6 A G I P - R - C 4 1 1 A G I P - R - S 4 5 3 A B. 40-t — 30 3 (0 W fc «- ** O Q. 20 « g 8 » .J * 10 G I P - R - 4 5 5 G I P - R - S 4 2 6 A G I P - R - S 4 2 7 A G I P - R - S 4 4 0 A G I P - R - S 4 2 6 / 4 2 7 A G I P - R - S 3 9 8 - 4 5 3 A c. 30-, OJ 3 tf) 20-W fc O tf) (0 o a g 10-a: 04 i i i i i 0 5 10 15 30 Time (min) 45 60 10 15 GIP-GIP-GIP-GIP-GIP-GIP-GIP-GIP-GIP-GIP-R - 4 5 5 R - S 3 9 8 A R - S 4 0 6 A R - C 4 1 1 A R - S 4 2 6 A R - S 4 2 7 A R - S 4 4 0 A R - S 4 5 3 A R - S 4 2 6 / 4 2 7 A R - S 3 9 8 - 4 5 3 A Time (min) Figure 21: Internalization kinetics of C-terminal serine to alanine mutant receptors in transfected CHO-K1 cells (A) Mutants not differing from the wild type GIP receptor. (B) Mutants showing altered internalization kinetics. (C) Examination of the first 15 min of internalization. Data represent mean ± S.E.M.; n = 3-10; * = P < 0.05, ** = P < 0.01. Refer to Methods sections 2.2, 2.3, and 2.8 for more details. 95 3.2.3.5 Internalization of GFP-tagged GIP Receptor In an attempt to validate results obtained from radiometric analysis of GIP receptor internalization, a plasmid construct encoding a GIPR/GFP chimera was generated to allow internalization to be studied by fluorescent microscopy (refer to Methods section 2.2). Four stable transfected cell lines were used in the current set of experiments. Wild type GIP receptor expressing CHO-K1 cells (wtGIPR) were used for control purposes, in order to establish effects of adding GFP to the carboxyl terminus of the receptor on receptor expression, binding affinity, and signal transduction. Upon killing non-transfected cells with G418, isolated colonies of GIPR-GFP expressing cells arose, and were transferred to culture vessels for further characterization. One hundred lines were screened, and three were selected based on their expression level relative to wtGIPR cells: GIPR/GFP-1 (rGIPR/GFP-23S: 0.49 ± 0.12), GIPR/GFP-2 (rGIPR/GFP-8L: 0.68 ± 0.09) and GIPR/GFP-3 (rGIPR/GFP-20G: 2.41 ± 0.24). Hence, the GIPR/GFP chimera was still able to be expressed at comparable levels to the native receptor, and C-terminal extension did not prevent correct intracellular sorting of the receptor to the cell surface. Binding competition experiments similarly showed that binding affinity was not adversely affected by the large C-terminal addition (IC50 values: 3.21 ± 0.83 nM [wtGIPR], 4.21 ± 0.90 nM [GIPR-GFP-1], 3.79 ± 0.64 nM [GIPR-GFP-2], and 5.54 ± 1.30 nM [GIPR-GFP-3]) (P> 0.05; Figure 22). Concentration-response curves of GIP,.42 on transfected cells, measuring cyclic AMP production over 30 minutes, indicated major differences between wild type and tagged receptors (Figure 23). Only GIPR-GFP-3 displayed similar sensitivity to GIP as the native receptor (EC50 values: 446 ± 80 pM versus 174 ± 45 pM, respectively; P > 0.05), however there was a proportional association between maximal cAMP production and expression level: wtGIPR (338.1 ± 39.1 fmol/1000 cells), GIPR-GFP-1 (43.0 ± 3.2 fmol/1000 cells), GIPR-GFP-2 (115.7 ± 6.7 fmol/1000 cells), GIPR-GFP-3 (288.7 ± 39.6 fmol/1000 cells); GIPR-GFP-1 and GIPR-GFP-96 2 were significantly different from wtGIPR, P < 0.05. The sensitivities of the lower expressing tagged receptors was significantly right shifted (EC50 values: GIPR-GFP-1, 8.86 ± 1.88 nM, GIPR-GFP-2, 1.52 ± 0.20 nM; P < 0.05). Thus it appears that the GFP tag may have interfered with G-protein coupling to a degree, but this could be somewhat overcome by overexpression. For GIPR-GFP cells, it was possible to examine receptor sequestration by two validated independent methods. The first method employed involved measurement of remaining surface binding following treatment with 100 nM GIP for various times, and a hypertonic acid wash. Using this method, it was readily apparent from the two higher expressing GIPR-GFP cell lines that receptor internalization kinetics were blunted (Figure 24A). At t=60 min, there was a 31.8 ± 0.8% reduction in wild type receptor surface binding, whereas GIPR-GFP-2 and GIPR-GFP-3 displayed reductions of 10.3 ± 2.6% and 11.1 ± 1.5%, respectively. The expression level of GIPR-GFP-1 was too low to permit measurement of internalization by this method with any sensitivity, and it appeared to not internalize at all (Figure 24A). The second method used to examine receptor endocytosis measured internalization of 125I-GIP (acid resistant cell associated radioactivity). This method indicated that the acid stripping protocol used in all internalization studies was approximately 85-90% effective at removing surface bound ligand. As with the first method, GIPR-GFP transfected cells showed reduced sequestration kinetics. Acid resistant binding of wild-type receptors was 43.3 ± 1.6% (accounting for efficiency of the acid stripping protocol) after warming to 37°C for 60 min (Figure 24B). In contrast, the acid resistant binding of the three clones of GIPR-GFP cells at 60 min ranged between 26.7 and 29.0% (P < 0.05). Despite the fact that alterations in signalling and sequestration were observed with GFP tagged receptors, studies were continued using fluorescence microscopy. In the non-stimulated state, all three cell lines transfected with GFP-tagged receptor showed diffuse apparent surface fluorescence, with micropili-like projections (Figure 25A); thus receptors were appropriately processed and sorted to the cell surface. Most cells appeared to have normal morphology to non-97 transfected CH0-K1 cells (> 95% fibroblast-like). Upon stimulation for 60 min with 100 nM GIP, the fluorescence appeared to redistribute: surface fluorescence was less intense, and punctate peri-nuclear fluorescence was observed, characteristic of the endosomal degradation pathway (Figure 25B). In a small proportion of the cells (< 15%), changes in cell morphology were observed following treatment with GIP in addition to the redistribution of fluorescence. 120 n i 1 1 1 1 1 1 1 1 0 -12 -11 -10 -9 -8 -7 -6 -5 Log 1 0 [GIP] Figure 22: Binding competition experiments with C-terminal green fluorescent protein (GFP) tagged GIP receptors in transfected CHO-K1 cells Three stable subclone cell lines were selected based on their surface expression level (in brackets) relative to wtGIPR cells. No significant difference was observed in IC50 values. Data are mean ± S.E.M. of 4-6 experiments. Refer to Methods sections 2.2, 2.3, and 2.6 for more details. 98 i i i i i i i r Basal -12 -11 -10 -9 -8 -7 -6 L o g 1 0 [GIP] Figure 23: Cyclic AMP production by subclones of GIPR-GFP cell lines Both GIPR-GFP-1 and GIPR-GFP-2 showed significant differences in maximal cAMP production and half-maximal activation statistics (P < 0.05). GIPR-GFP-3 showed slightly reduced maximal cAMP and right shifted receptor sensitivity to GIP, but these were not significantly changed from wtGIPR cells (P > 0.05). Data represent mean ± S.E.M.; n > 3. Refer to Methods sections 2.2, 2.3, and 2.7 for more details. 99 A. Loss of Surface Receptors 110-1 o o 100-Q. c 0) o 90-O o o 80-Q: a 70-u CO c Surf indi 60-Surf ca 50-40-0 wtGIPR (1.00) GIPR/GFP-1 (0.49) G I P R / G F P - 2 (0.68) G I P R / G F P - 3 (2.41) ~ i — i — 10 15 25 Time (min) 60 Figure 24: Internalization of GIPR-GFP in transfected CHO-K1 cells (A) Loss of surface receptor expression on treatment with 100 nM GIP. (B) Internalization of l25I-GIP versus time. Data represent mean ± S.E.M.; n = 4-8; * = P < 0.05. Refer to Methods sections 2.2, 2.3, and 2.8 for more details. 100 Figure 25: Fluorescence microscopy of GIPR-GFP distribution in transfected CHO-K1 cells (A) Fixed control GIPR-GFP-3 cells in the unstimulated state. (B) GIPR-GFP-3 cells, fixed following 60 min with 100 nM GIP at 37°C. wtGIPR cells did not show specific fluorescence above autofluorescence level of non-transfected CHO-K1 cells. Figures are representative images of the entire cell populations observed in each case. Refer to Methods sections 2.2, 2.3, and 2.9 for more details. 101 3.2.3.6 Internalization of [Fluorescein-Trp 1GIP , . 3 m m in wtGIPR cells Although useful for fluorescence imaging, tagging the GIP receptor resulted in retarded internalization kinetics, thus an alternative approach was made to examine internalization by fluorescence, namely, conjugating fluorescein to tryptophane25 of GIP. Competitive-binding inhibition experiments with GIPi. 3 0 N H 2 [n] and [Fluo-Trp25]GIP, -30NH2 ffl v e r s u s \"I_GIP|.42 on wtGIPR cells showed similar IC50 values (Figure 26A), with the fluorescent conjugate having slightly greater affinity: 1.68 ± 0.07 nM [n] and 1.14 ± 0.08 nM [f] (P < 0.05). Similarly, [Fluo-Trp25]GIP was slightly more potent than native GIP when examining maximal cAMP production (430.6 ± 52.5 and 390.4 ± 54.2 fmol/1000 cells, respectively; P > 0.05; Figure 26B), however, receptor sensitivity to these peptides was not different (EC50 values: 101 ±33 pM [n] and 104 ± 18 pM [f]). When measuring loss of GIP surface binding sites on wtGIPR cells on exposure to 100 nM GIP,. •30NH2 o r fluorescein-conjugated peptide, the latter peptide caused significantly greater receptor internalization at all time-points used (Figure 27). At t = 60 min, wtGIPR cells exposed to [Fluo-Trp25]GIP had ~10% fewer binding sites remaining than those exposed to native GIP. Thus, it appears that [Fluo-Trp25]GIP suffers from the opposite problem to that observed with GIP-GFP receptors. Study of cells using [Fluo-Trp25]GIP by fluorescence microscopy was somewhat more difficult than anticipated. Photobleaching and weak fluorescence prevented the production of usable images with cells fixed after incubating for 60 min with 25-100 nM peptide either on ice or at 37°C (data not shown). However, use of a fluorescein amplification kit consisting of an anti-fluorescein antibody, an Alexafluor-488 conjugated secondary antibody, and permeabilized cells produced satisfactory images (Figure 28), although the signal-to-noise ratio appeared to be lower than that observed with GIPR-GFP cells. Incubation of cells with 25-100 nM [Fluo-Trp25]GIP at 37°C indicated primarily intracellular fluorescence in the area around the nucleus (Figure 28B), whereas fluorescence of control cells with the same procedure at 4°C appeared to be mainly on the cell surface (Figure 102 28A). As with studies using GIPR/GFP, to definitively establish the location cellular fluorescence, confocal microscopy would have to be performed. i 1 1 1 1 1 1 r 0 -12 -11 -10 -9 -8 -7 -6 \" 120-i i i i i i i i r 0 -12 -11 -10 -9 -8 -7 -6 Log [Peptide] Figure 26: Binding and cAMP production of fluorescein-conjugated GIP (A) Competitive-binding inhibition of '25I-GIP,.42 on wtGIPR cells incubated with increasing concentrations of GIP,.30NH2 or [Fluo-Trp25]GIP,.30NH2. (B) Cyclic AMP accumulation in wtGIPR cells incubated with GIP,.30NH2 or [Fluo-Trp25]GIP,.30NH2. Data represent mean ± S.E.M. of 4-6 experiments. Refer to Methods sections 2.2, 2.3, 2.4, 2.6 and 2.7 for more details. 103 i 1 1 1 1 1 r 0 1 0 2 0 3 0 4 0 5 0 6 0 Time (min) Figure 27: Loss of wtGIPR binding sites on incubation with [Fluo-Trp ]GIP Cells were incubated with 100 nM GIP,.30NH2 or fluorescein-conjugated equivalent for various times at 37°C. Non-internalized surface bound peptide was stripped with low pH hypertonic saline, and remaining surface receptors estimated by radioligand binding. Data represent mean ± S.E.M. of 6 experiments. Refer to Method section 2.8 for more details. 104 Figure 28: Fluorescence microscopy of wtGIPR cells incubated with [Fluo-Trp^JGIP (A) Cells incubated with 25-100 nM peptide on ice for 60 min, prior to washing, fixing, and treatment with fluorescein amplification kit, as per text. (B) The same protocol as in (A), but performed at 37°C. The dashed line indicates the cell perimeter. Figures are representative images of the entire cell populations observed in each case. Refer to Method section 2.9 for more details. 105 3.3 Discussion 3.3.1 Insulinoma Cell Desensitization In order to understand the basis of desensitization more clearly at the level of intracellular signalling cascades, the (3TC-3 insulinoma cell line was employed. The effect of glucose on homologous desensitization of (3-cells to GIP was studied with respect to cyclic AMP production, and the subsequent effect, insulin release. In contrast to previous work on transformed [3-cells [176; 177], 6TC-3 cells exhibited GIP-concentration dependent effects on cAMP production in the absence of glucose, as well as at physiological glycemic levels (Figure 5). Glucose alone appeared to have very mild effects on cyclic AMP production, slightly reducing cellular responsiveness to GIP with increasing glucose. Prior work on glucose dependence of incretin-stimulated cAMP in transformed cells found that neither GLP-1 nor GIP could elevate intracellular cAMP in the absence of glucose, while forskolin stimulated cAMP production was unaffected [176; 177]. The apparent conflict between the glucose dependence of the cyclic AMP data from HIT, RIN and |3TC beta cell models may be species dependent and will likely be clarified by similar experiments on alternative insulinoma cells [327] or purified |3-cells [224], although previously glucose-independent GIP-stimulated cAMP has been shown to occur in human insulinoma tissue [182]. Regardless, the necessity for the presence of glucose for GIP to potentiate insulin release was observed in (3TC-3 cells (Figure 7), indicating that the glucose-dependency of GIP was intact. Homologous desensitization of GIP mediated effects was first described by Fehmann and Habener [177]. In their study, only effects of GIP pretreatment on insulin release from HIT cells were examined, not second messenger cascades. Desensitization of GIP-stimulated cAMP production in (3TC-3 cells resulting from GIP pretreatment was at best moderate (Figure 5). The magnitude of the desensitized response was similar under all glycemic conditions although, due to the slight inhibitory effect of glucose on GIP-stimulated cAMP formation, the desensitization 106 appears to be greater in the absence of glucose. Using an identical protocol, the degree of desensitization observed in the absence of glucose corresponds well with that observed in non-glucose sensitive CHO-K1 cells transfected with the pancreatic GIP receptor (cf. Figures 5 and 9, and Figures 6A and 1 IB). However, the time-course of GIP receptor desensitization (Figures 6A and 1 IB) does not match the rapid potent desensitization that the GLP-1 receptor underwent in HIT-T15 cells, using a very similar protocol [177]. In order to elucidate the point in the signal transduction cascade at which desensitization of the cAMP response to GIP was occurring, various inhibitors were used (Figure 6). The apparent reduction in cAMP produced after GIP pretreatment could be explained by degradation of extracellular ligand, activation of pertussis-sensitive G-proteins, modulation of adenylyl cyclase or phosphodiesterase (PDE) activity, or negative feedback by signal transduction kinases (PKC and PKA). Inhibition of DPIV, the primary enzyme responsible for degradation of GIP in vivo [59], had no effect on GIP-stimulated cAMP levels or desensitization; neither did pertussis toxin. GIP pretreatment reduced only GIP-stimulated cAMP production, not forskolin-stimulated cAMP production, suggesting that adenylyl cyclase activity was unaffected by GIP pretreatment. Furthermore, PDE sensitivity to IBMX was unchanged in the desensitized state, ruling out an increase in PDE activity resulting in a reduction of intracellular cAMP. Staurosporine did affect basal and GIP-stimulated cAMP accumulation in control cells, indicating that PKC probably has a role in regulating normal (3-cell responsiveness to GIP, but did not affect homologous desensitization to GIP. PKA inhibition affected neither responsiveness nor desensitization of (3TC-3 cells to GIP. These results, taken together, suggest that homologous desensitization of GIP-stimulated cAMP formation occurs at the receptor level, and may involve mechanisms such as RGS (regulators of G protein signalling) proteins [371], GRK-2 [372], receptor sequestration [290], or unidentified processes. 107 Both glucose and GIP produced similar changes in insulin release from [3TC-3 cells to those previously published [178]. Parallel experiments on desensitization of insulin release showed that the same protocol used for the cAMP studies yielded a significant reduction in insulin release. Pretreatment of cells with GIP dramatically returned 10 nM GIP-stimulated insulin to basal levels in either 5.5 and 11 mM glucose conditions. It would be difficult to attribute the dramatic alteration in insulin release to the mild attenuation of GIP-stimulated cAMP by GIP pretreatment, possibly implicating desensitization of distal steps in the stimulus-exocytosis coupling cascade. This hypothesis is supported by similar experiments utilizing forskolin. Forskolin potentiates glucose-induced insulin release via direct activation of adenylyl cyclase and cyclic AMP production. In contrast to GIP-stimulated cAMP production, pretreatment with GIP had no effect on forskolin-stimulated cAMP accumulation (Table 3). Nevertheless, GIP pretreatment significantly blunted forskolin-mediated insulin release to near basal levels (Figure 7). GIP has been found to augment depolarization-induced exocytosis from individual mouse (3-cells via a protein kinase A (PKA)-dependent mechanism [190]. Consistent with results using cyclic AMP analogues and forskolin, GIP exerts its action on exocytosis at a level distal to elevation in intracellular calcium via PKA [190; 388]. Evidence also exists that cyclic AMP interacts directly with the secretory machinery, sensitizing it to [Ca2+], [388; 389]. Modulation of kinases activated by GIP [293; 298-300] or any element of the stimulus-secretion machinery by these kinases, or other uncharacterized mediators of GIP effects could account for the profound effect of GIP prestimulation on both GIP and forskolin-stimulated insulin responses. Indeed, the related incretin hormone, GLP-1, has been implicated in the tyrosine phosphorylation of several proteins, one of which was identified as the synaptic-associated protein of 25 kDa (SNAP-25) [390], supporting the notion of direct modulation of the exocytotic machinery. Further support of this hypothesis is the recent finding that chronic 9 hour pretreatment of INS-1 cells with GIP also blunted glucose induced insulin release [383]. 108 Insulin release from |3TC-3 cells was reported to be reduced by 42-55% by a 10 min exposure to 100 nM GIP preincubation, followed by a 10 min washout period [371]. While desensitization of GIP-stimulated cAMP production in pTC-3 cells was not demonstrated in this report, it was shown that GIP induced an increase in the steady-state level of RGS2 mRNA and that transfection of |3TC-3 cells with RGS2 attenuated GIP-stimulated but not glucose-stimulated insulin release [371]; further, Tseng and Zhang presented evidence that RGS2 was capable of co-precipitating with Gcts from transfected cells in vitro. In a separate report, they also found evidence for expression of GRK-2 and (3-arrestin-1 in (3TC-3 cells, and that transfection of these proteins diminished GIP-stimulated insulin release [372]. The insulin data from those reports do not compare well with previous data on the same cells [178] or that presented here, likely because Tseng and Zhang used (3TC-3 cells over the high passage range 62-70 [371; 372], for which insulin content and cell response to secretagogues have not been characterized. Lower passage ranges are generally preferable due to the well characterized decline in insulin content and glucose responsiveness over time in transformed (3-cells [327]. RGS2 is ubiquitously expressed [391], and thus does represent a potential mechanism for (3TC-3 desensitization to GIP, however, the increased rate of GTP hydrolysis by RGS proteins should in theory also affect cyclic AMP levels. As stated previously, the dramatic desensitization of the insulin response in (3TC-3 cells (Figure 7) did not correlate well with homologous desensitization of cAMP responses to GIP in these cells (Figure 5 and 6). This is particularly important considering the 10 minute time-point of the cAMP desensitization time-course (Figure 6A), and the 10 min desensitization period reported by Tseng and Zhang [371] with respect to insulin. Another question is raised considering that RGS proteins were identified 20 years ago, and exhibit interactions with Got, and Ga q strong enough to facilitate G-protein-RGS complex crystallography structure determinations, yet all attempts by those who have focused their 109 research on RGS proteins have failed to show any interaction with Ga s [349; 350; 391]. Nevertheless, reports have shown that specific Gas RGS proteins exist [392], and also that RGS proteins may inhibit cAMP formation via direct interaction with adenylyl cyclase [393], rather than by accelerating G-protein GTP hydrolysis. Hence, GRK-2 and RGS2 attenuation of GIP signalling may be a feasible desensitization mechanism, however, given the points raised, independent duplication by other researchers should be done to validate these findings. It is likely that desensitization to GIP occurs at both proximal (receptor-effector) and distal (exocytotic machinery) stages of the stimulus secretion cascade. 3.3.2 Desensitization of Transfected Cells Difficulties in development of a desensitization protocol for the transfected rat GIP receptor hampered early work on this subject. Once it was found that overexpression of the receptor artefactually prevented measurement of desensitization (Figure 9), specific studies could be performed. Some studies have used cyclic AMP accumulation time-course experiments as indicating desensitization, as cyclic AMP levels plateau (e.g. [358; 359]). Others have argued that the plateau may arise from overcoming the IBMX blockade of cAMP degradation by phosphodiesterase due to such high levels of cyclic nucleotide [394]. In the absence of IBMX a plateau was achieved, and cyclic AMP levels did not begin to return to basal values (Figure 10). Hence an equilibrium between GIP-stimulated cAMP production and PDE-mediated cAMP degradation was achieved, indicating that the GIP receptor is fairly resistant to desensitization (or the plateau would have declined towards basal levels), and this is not simply an artifact of overexpression. In the low expressing rGIPR-L2 cell line, which produces much lower maximal GIP-stimulated cAMP levels (Figure 9), time-course experiments on cAMP production also revealed a steady state cAMP level with time in the presence of IBMX (Figure 11A); as the cAMP content of these cells is much lower than wtGIPR cells, it is unlikely that the plateau 110 results from overcoming inhibition of PDE by IBMX. In cells pretreated with 100 nM GIP for 60 min prior to a cAMP accumulation time-course, a plateau was also achieved, except this time it was set at a lower level (Figure 11 A). Thus it was shown that plateau steady-state cyclic AMP levels are not simply due to an equilibrium being achieved between phosphodiesterase, cAMP and IBMX, and that prolonged (chronic) stimulation induces a functionally distinct mode of desensitization from repetitive stimulation (Figures 10 and 11 A). Notably, the desensitization observed in all experiments in transfected CHO-K1 cells (Figures 10 and 11) also indicated that the GIP receptor is relatively resistant to desensitization, and data compares well with the desensitization observed in 6TC-3 cells (Figures 5 and 6). The first study of GIP receptor desensitization in transfected cells was published in 1996 [369]. In this report, a reporter cell line driving expression of (3-galactosidase with a cyclic AMP response element (CRE) containing promoter, transfected with the GIP receptor (LGIPR2) was used. Cells were cultured in the presence of 2 nM GIP for periods between 0 and 24 hours, and it was shown that cAMP-dependent (3-galactosidase activity peaked with 4 hours of incubation, but then returned to near baseline within 16 to 24 hours [369]. Hence it was concluded that this represented a chronic desensitization of the receptor. As experiments in which cells were incubated with GIP were performed in the presence of 5% fetal bovine serum, it is likely that degradation of the ligand by serum proteases contributed a large degree towards the attenuation of the signal; cell culture medium and radioimmunoassay buffers containing serum products were shown to degrade GIP by DPIV-like activity in vitro [330]. Furthermore, enzyme protein turnover and stability were not considered. Given that cAMP accumulation time-courses presented here and elsewhere (cf. Introduction section 1.7) produce rapid peak responses, reaching a plateau between 15 and 30 min (Figures 10 and 11), the time-course of (3-galactosidase activity as a reporter indicates that this method is invalid for examining desensitization. I l l In the current thesis, strong evidence is presented indicating receptor expression level is critical for demonstration of desensitization. In two prior papers reporting GIP receptor desensitization [291; 371], the authors reported a constant 45% transfection efficiency using 0.75 pg receptor plasmid with 4 pL Lipofectamine and cells in 12-well plates. Given the current results, and the fact that they were able to show desensitization, this transfection protocol must have yielded low level receptor expression. Binding studies were performed, which should have allowed comparison, however, the first step of the binding protocol they used involved detaching cells from culture vessels with trypsin, likely explaining the 10-fold right-shift in binding affinities published in the same report, as many trypsin consensus sites are present in the GIP receptor sequence [291]. Regardless, two desensitization protocols were used: a 24 hour 100 nM GIP preincubation [291] or a 10 min 100 nM GIP preincubtion [371], and in either case followed by a 10 min exposure to a range of GIP concentrations. Apparently, in both cases, this produced approximately a 90-95% reduction in GIP-stimulated cAMP formation in the pretreated cells. The cause of the disparate results from those reports, compared to the current thesis are unclear. While cell specific and/or methodological differences may be the root, GIP receptor desensitization in CHO-K1 cells (Figures 9 and 11) and HEK-293 cells (data not shown), was nearly identical to that observed in (3TC-3 cells (Figures 5 and 6). No studies of the mechanisms of desensitization were pursued in transfected cells in the current report (these were limited to (3TC-3 cells), however, it was shown that removal of all potential phosphorylation sites in the C-terminal tail by site-directed mutagenesis completely blocked the desensitization of GIP-stimulated cAMP production in low-expressing cell lines with this construct (Figure 12). In the same set of experiments, the role of the serine doublet at amino acids 426 and 427 in homologous desensitization was assessed, as serine doublet phosphorylation is a well characterized regulatory mechanism for the GLP-1 receptor [366-368]. However, cells transfected with the GIPR-S426/427A mutant receptor desensitized identically to 112 control receptors (Figure 12). Taken together, the results implicate phosphorylation of specific C-terminal serine residues during homologous desensitization, however, apparently S426 and S427 are not involved. Tseng and Zhang [291] found that rat GIP receptors bearing a S406 mutations were slightly protected from chronic desensitization (~20% mutant response compared to ~5% wild type response when preincubated for 24 hours with 100 nM GIP). However, the prolonged treatment with GIP resulted in profound receptor downregulation - 30-32% of mutant receptor binding sites remained following the preincubation, compared to 13% for wild type receptors, so signalling was proportional to these values. These chronic experiments cannot really be compared to the acute signalling data presented here; thus phosphorylation of S406 alone or in combination of other residues (possibly not including S426 and S427) may mediate acute desensitization. Co-transfection of GRK-2 or RGS2 blunted steady-state GIP-stimulated cyclic AMP in transfected cells [291; 372]. This is corroborated by a similar experiment shown for GRK-2 in Figure 18. It was reported that RGS co-transfection did not affect receptor expression [291], whereas GRK constructs were transfected into stable LGIPR2 cells with approximately 40% efficiency [372], but effects on receptor expression were not measured. As co-transfection with two plasmids theoretically gives proportional DNAdiposome complexes with a given transfection efficiency, all cells successfully transfected will contain both plasmids. When transfecting a stable cell line, where during the original transfection all non-transfected cells were killed by the selection agent, the second transient transfection can only co-express in the percentage of cells determined by transfection efficiency; the majority of cells secondarily transfected by this method will express only the first stably transfected plasmid. In preliminary experiments, transient transfection of wtGIPR cells with 10 pg GRK-2 produced only a minor effect (data not shown), relative to that observed when 3 pg of pGIPR was co-transfected with 7 pg pGRK-2 (Figure 18). Given that co-transfection of GIPR and GRK-2 resulted in diminished 113 surface expression, but with unchanged transfection efficiency, the reduced signalling observed here (Figure 18) and elsewhere [372], probably results from enhanced internalization (Figure 18) mediated by GRK-2 phosphorylation of the GIP receptor [372]. 3.3.3 GIP Receptor Internalization in Transfected Cells Receptor sequestration in response to acute treatment with GIP likely occurs via a clathrin-coated pit mediated pathway, as hypertonic sucrose blocks this pathway (Figure 16) [395]. Receptor signalling appeared not to be a requirement for receptor internalization, as GIP7 3 0 N H 2, a receptor antagonist was also able to promote receptor internalization in a time- and dose-dependent manner (Figure 15). Adenylyl cyclase inhibitors were also able to stimulate the loss of specific GIP surface binding sites, independently of GIP (Table 5, Figure 17). However, the specificity of adenylyl cyclase inhibitors to internalize the GIP receptor was possibly non-specific, as the opposite effect was not observed with cAMP raising agents forskolin or IBMX; perhaps the inhibitors caused cyclase endocytosis, and due to close proximity, the GIP receptor was also sequestered. Both PKA activation and inhibition affected GIP receptor sequestration in the presence or absence of GIP, and likely represents a modulator of this process (Table 5, Figure 17). A role for phosphorylation in the process of receptor sequestration is implied by the altered internalization kinetics of G1PR-S426A, GIPR-S427A, GIPR-S440A, GIPR-S426/427A and GIPR-S398-453A (Figure 21). Signalling and expression of these constructs was not different from wild type (Table 6, Figure 19 and 20). The complete removal of all C-terminal serines (GIPR-S398-453A) resulted in slower internalization kinetics than all other mutants (Figure 2IB). Single mutation of S426, S427 and S440 produced a moderate reduction in internalization rate, but not different from the double mutant, GIPR-S426/427A (Figure 2IB). Thus, at least for S426 and S427, the reduction in internalization was not additive. As the complete serine 114 knockout had slower kinetics, then perhaps phosphorylation of S440 in combination with S426 or S427 may be additive, but it cannot be ruled out that additive effects from phosphorylation of residues having no effect alone (S398, S406, or S453) may contribute to the slowed internalization of the GIPR-S398-453A mutant. Notably, mutant receptor GIPR-S426/427A was shown not to resist homologous desensitization (Figure 12), but did display reduced internalization kinetics (Figure 21). In contrast, GIPR-S398-453A was both resistant to desensitization (Figure 12) and had a slowed sequestration time-course (Figure 21). These results suggest that the residues required for desensitization and internalization may bear some overlap, but are functionally distinct processes. The chronic downregulation observed by Tseng and Zhang [291], likely represents a different process than receptor sequestration, as identical mutants GIPR-C411A and GIPR-S406A showed identical internalization kinetics compared to wild type receptors (Figure 21, Table 6), but during chronic 24 hour GIP treatments, both of these mutant receptors were resistant to downregulation [291]. Direct phosphorylation of the GIP receptor has been demonstrated, although the precise residues phosphorylated were not identified [372]. Further, it was reported that GRK-2 and |3-arrestin had no effect on receptor sequestration [372]. Only a single time-point of acid resistant binding was presented, and dose- or time-dependent internalization has never been shown by this group. The most plausible explanation for the lack of effect is that Tseng and Zhang transiently transfected stable LGIPR2 cells with GRKs and or (3-arrestin, and the 40% transfection efficiency did not allow demonstration of an effect, as previously discussed; co-transfection of the GIPR with GRK-2 clearly showed an increased kinetics of internalization (Figure 18C). Fluorescent tracking of G protein coupled receptors has been reported by many different groups for different receptors. Use of fluorescently labelled ligands is limited, primarily due to the cost of synthesis, and resulting low yields. The discovery of jellyfish green fluorescent 115 protein provides a unique opportunity to use fluorescent microscopy to track intracellular receptor trafficking using receptor chimeras [396; 397]. The primary limitation of this approach is the large size of GFP (27 KDa); effects of GFP tagging on surface expression, binding affinity, and receptor activation must be carefully performed prior to extensive use of such a construct. Adrenergic, THRH, CCK and PTH receptors were reported to be unaltered by C-terminal tagging with GFP [332; 355; 398-401]. However, subsequent use of GFP-tagged GLP-1 receptors and secretin receptors, omitted receptor activation control experiments with receptor-GFP chimeras [358; 402], In our hands, GLP-1R-GFP (the same construct used by Salapatek et al. [402]) showed dramatically reduced GLP-1 stimulated cAMP production relative to wild type receptors (data not shown). Similarly, a GIPR-GFP chimera also showed reduced ability to activate adenylyl cyclase (Figure 23), but normal cell surface expression (Figure 22). Unfortunately, GFP-tagging also slowed receptor internalization kinetics (Figure 24), possibly by preventing GRK phosphorylation and/or (3-arrestin binding by steric hindrance. Although qualitative, GFP-tagged GIP receptor internalization appeared normal when examined by fluorescence microscopy (Figure 25), and appeared to be similar to that observed for other receptors [332; 355; 398; 399]. Hence, the GIPR/GFP chimeric protein has severe limitations, but may prove useful in select applications, such as showing cell surface expression or marking the endosomal degradation pathway. It is possible that introduction of a more flexible linker or a longer linker between the GIP receptor and GFP may improve the signalling characteristics of this chimera, facilitating further applications. Because of the drawbacks of GFP-tagged GIP receptors, a fluorescein-conjugated peptide ligand of the GIP receptor was synthesized. Fluorescent peptide ligands have been reported for many peptide hormones, and have proven useful for monitoring receptor sequestration indirectly, via internalization of the ligand [398; 403; 404]. Binding, signalling and receptor sequestration kinetics were favourable for [Fluo-Trp25]GIP,. 3ONH2» w i m only small, but significant, differences 116 from native peptide (Figures 26 and 27). While signalling was unaltered, [Fluo-Trp25]GIPl.30NH2 displaced l2T-GIP binding slightly more efficiently, and was somewhat more potent at stimulating receptor internalization, monitored by radioligand assays. These changes likely would not affect the utility of this compound as a tracking molecule during receptor endocytosis. The real drawback was not discovered until fluorescent microscopy images were obtained. Only low fluorescence could be observed with [Fluo-Trp25]GIP,.30NH2 alone - photobleaching and limited shelf life of fluorescent compounds are likely responsible. However, despite this setback, use of a Alexafluor-488 fluorescein amplification kit permitted satisfactory images to be obtained (Figure 28). Hence, perhaps directly conjugating GIP to an alternate, more stable fluorophore may produce better results when examining receptor sequestration by fluorescent microscopy, however, binding, signalling and internalization controls will need to be repeated, and may not match those for [Fluo-Trp25]GIP,_30NH2. 3.3.4 Conclusion In summary, homologous desensitization to GIP has been demonstrated in (3TC-3 cells and CHO-K1 cells stably expressing the GIP receptor at low levels. Desensitization of GIP-stimulated cyclic AMP production appeared to be relatively slow, and paralleled GIP receptor sequestration kinetics. Desensitization was shown to not involve extracellular ligand degradation, induction of inhibitory G proteins, modulation of adenylyl cyclase or phosphodiesterase activity, or via signal transduction kinases A or C. In transfected cells, homologous desensitization could be dissected into repetitive versus continuous stimulation, as well as acute and chronic desensitization. Receptor internalization was found to be mediated by a clathrin-coated pit pathway, and did not require receptor activation per se. Modulation of internalization was not greatly affected by. activators or inhibitors of signal transduction cascades, however inhibition of adenylyl cyclase or activation of protein kinase C was able to 117 sequester surface receptors independently of ligand binding. Phosphorylation was implicated in both receptor desensitization and internalization, however, the specific sites were not fully identified. Evidence indicated that phosphorylation sites for desensitization and internalization may overlap, but are not necessarily identical for both processes. Use of fluorescent methods to track internalization was accomplished by creating a GIP receptor/green fluorescent protein chimera and by conjugating fluorescein to GIP, however, both methods had some drawbacks. The further characterization of GIP receptor desensitization, internalization and downregulation may provide insight into the diminished responsiveness of type 2 diabetic patients to GIP. 118 Chapter 4: Analogues of Insulinotropic Hormones 4.1 Introduction 4.1.1 Structure-Activity Relationships of GIP Use of cyanogen bromide to chemically cleave GIPM 2 into GIPM 4 and GIP,5.42 [405] for use in peptide sequencing [17; 18] ultimately resulted in the first structure-activity study of the newly discovered hormone [165]. Peptide purification methods had not advanced to the widespread use of HPLC, and as such, early studies must be considered with caution due to the possibility of peptide contamination (for example, later HPLC assessment of the purity of \"pure\" natural porcine GIP preparations found as much as 32% GIP3_42 as well as CCK contamination [19; 166]). Nevertheless, Pederson and Brown [165] demonstrated that pig GIP15.42 retained partial insulinotropic activity in the perfused rat pancreas, whereas [Homoserinelactone14]GIPM4 did not. These findings were supported by the finding that bovine GIP,7_42 [21] also retained 32% insulin releasing ability compared to native hormone using the same experimental model [267]. Use of enterokinase, trypsin and S. aureus V8 protease allowed isolation of GIP fragments corresponding to amino acids 1-3, 1-16, 4-42, 17-42 and 19-30 [21; 267]. On transplantable hamster insulinoma membranes, only GIP4.42 and GIP,7.42 were able to significantly displace l25I-GIP binding when tested up to 10 u,M. Pancreas perfusions with GIP4.42 indicated that this molecule had little, if any, insulinotropic activity, and it was suggested that it may act as a GIP antagonist, while GIP,.,6 was not tested based on prior results with GIPM 4 [165; 267]. Conflicting data were obtained, however, when GIPn_42 was tested on RINm5F insulinoma cells: native GIP was able to stimulate cyclic AMP production, whereas 10 u.M GIPI7.42 could not [198]. Neither GIP,. 18, nor GIP,9.42 were able to potentiate glucose-induced insulin release from isolated islets [406]. Another development arose during testing of synthetic GIP preparations. Yanaihara et al [407] synthesized a 43 amino acid peptide, based on the original reported structure of GIP [17; 119 18], and found that a partially purified preparation contained ~30% insulin releasing ability from the perfused rat pancreas. However, purified synthetic GIP,_38 was equally insulinotropic to the natural source GIP, but lacked potency when examining acid inhibitory action [408]; the insulinotropic properties of GIP]_39 have been independently confirmed [409]. Three possibilities were proposed to explain the discrepancy: (1) the last 5 amino acids (of the 43 a.a. sequence) are important for acid inhibitory activity (i.e. there are two separable bioactive domains for acid inhibition and insulin stimulation), (2) the original peptide sequence contained an error, or (3) acid inhibitory activity of natural source peptide was due to contamination with another molecule. Support has been found for the first two suggestions. Upon reassessment of GIP's primary sequence, lornvall et al [19] found that GIP was only 42 amino acids long, and an extra Gin residue was included at position 30. Synthetic GIP,_31 was found to be an equally potent stimulant of cyclic AMP production in human insulinoma compared to natural GIP,_42 [182]; the potency of GIP,_30 has been confirmed with transformed insulinoma and transfected cell models [198; 276; 278]. Similarly, in the perfused rat pancreas, synthetic porcine (sp) GIP,.30NH2 and GIP,_42 exhibited equipotent activation of insulin release, but in the perfused stomach, spGIP,. 3 ( ) N H 2 lacked somatostatinotropic activity [337; 410]. Bioassay of spGIP,.30NH2 in isolated perfused organs in rats showed it lacked gastric acid inhibitory activity, but retained inhibitory action on exocrine secretions, likely mediated via insulin [32]. Taken together, fragment analysis suggested the insulinotropic domain resided between the overlap of GIP|_3(INH2 and GIP, 5-42 0 r G1P,M2; when tested, GIP,5.30, GIP,7.30 and GIP,9.30 did possess weak insulin stimulating ability in the perfused pancreas [337]. Residues 27-30 may be important for biological activity, as GIP,.27 and GIP,.28 lost insulinotropic potency [198; 411; 412]. Antagonism of the GIP receptor has been demonstrated with N-terminally truncated peptides, GIP6.30NH2, GIP7.30NH2, and GIP,0.30NH2, and the complete high affinity binding domain of GIP resides between residues 6 and 30 [186; 413]. 120 4.1.2 Metabolism of GIP 4.1.2.1 Enzymatic Inactivation When the sequence of GIP was reassessed, during HPLC purification of natural porcine GIP, an additional peptide corresponding to GIP3.42 was identified as a major component of the preparation, and it was suggested that enzymatic cleavage may have resulted in the N-terminally truncated molecule [19]. Preliminary reports using the perfused rat pancreas model suggested that the peptide was devoid of insulin releasing ability [414; 415]; complete examination in the perfused stomach and pancreas indicated that GIP3.42 was not effective in stimulating somatostatin or insulin release [57]. Subsequent studies on isolated rat islets confirmed that GIP3.42 was not insulinotropic, and further, that it likely was not an antagonist of the GIP receptor [166; 416]. The seminal paper of Mentlein et al [58] on degradation of GIP,.42 by purified dipeptidyl peptidase IV (DPIV) and serum confirmed that GIP inactivation by cleavage of a dipeptide from the N-terminus was enzymatic in nature. Using HPLC, it was possible to identify the released Tyr1-Ala2 molecule, as well as separate GIP3.42 from GIP,_42 on chromatograms, facilitating kinetic enzymology studies. Use of the DPIV inhibitor, Lys-Pyrrolidide, in serum, indicated that DPIV was the major degrading enzyme present, and kinetic constants supported a physiological role for cleavage [58]. In vivo relevance was demonstrated independently, where 125I-GIP,_42 was shown to be degraded to 12T-GIP3.42 by purified DPIV, serum, and during IV injection at physiological concentrations into anesthetized rats [59; 330]. Subsequent studies have shown that GIP3.42 is actually a weak antagonist of the cloned GIP receptor; as such, GIP3 4 2 was unable to stimulate cAMP accumulation in transfected cells and retained high affinity receptor binding [324]. Kinetics of GIP degradation by purified DPIV and serum were reassessed by mass spectrometry, and confirmed GIP3.42 as the primary degradation product in both cases [380]. Specific inhibition of DPIV in vivo was able to increase the proportion of N-terminally intact bioactive peptide when exogenous GIP,_42 was infused [417]. 121 Development of N-terminally specific immunoassay techniques has been critical in showing clinical relevance of DPIV-mediated degradation of GIP in humans [60; 61]. Because of its primary role in inactivating incretin hormones, specific DPIV inhibition has been proposed as a therapeutic approach for type 2 diabetes, to improve glucose tolerance by increasing the biological half-life of endogenously secreted incretins [381; 382; 418; 419]. GIP has been reported to be degraded by neutral endopeptidase (NEP), however, it was a poor substrate for the enzyme [420]. Furthermore, GIP fragments corresponding to amino acids 7-42, 10-42, 11-42, 17-42 and 1-39 have been isolated from natural sources [20; 409; 421], indicating that further cleavage of the peptide is possible, although the enzymes responsible and the physiological relevance have not been established. 4.1.2.2 Pharmacokinetics Early studies on GIP pharmacokinetics in humans were performed with a C-terminally directed GIP antibody, and resulted in an estimated a t,/2 value in the order of 20 min [60; 82; 164; 221; 422]. The similarity between metabolic clearance rate (MCR, man 2.6-8.7 mL/Kg/min; pig 8.3 mL/Kg/min) and the glomerular filtration rate suggests that GIP is likely freely filtered, and implicates the kidneys for extraction of GIP, as measured with C-terminal specific assays [60; 221; 422]. This is consistent with elevated total IR-GIP levels in uremia, renal failure, and experimental nephrectomy, as well as renal arteriovenous differences in man, dog and rat [423-426]. As it was realized that C-terminally directed immunoassays were overestimating bioactive GIP, upon development of an N-terminally specific assay, pharmacokinetic parameters of GIP were reassessed in swine and man [60; 417]. When measuring intact GIP,.42, the t,/2 disappearance rate for exogenously infused GIP ranged from 3.3 (pig) to 7.3 min (human) [60; 417], corresponding well to t,/2 values for physiological conversion of 125I-GIP,.42 to l25I-GIP3.42 in rats [59]. When tissue extraction was reassessed using N- and C-122 terminally directed GIP radioimmunoassays, extraction of C-terminal immunoreactivity was only affected in the kidney by DPIV inhibition [417]. While the renal brush borders are rich in DPIV [427], in order to extract C-terminal immunoreactivity, either secondary degradation of the epitope must occur, or the peptide is removed in urine; as DPIV inhibition reduced C-terminal clearance of GIP [417], these events likely take place after dipeptide cleavage. Further, from these studies was the finding that liver and muscle contribute to the DPIV-mediated N-terminal degradation of GIP [417], whereas prior examination of total IR-GIP extraction by the liver suggested that it was not a major site of GIP removal [230; 428]. Notably, DPIV is present as a soluble isoform in plasma and on the surface of lymphocytes; as such, the bloodstream is likely a major site of GIP cleavage [380]. 4.1.3 Metabolism of GLP-1 and Glucagon In contrast to the long clearance half time for C-terminal GIP immunoreactivity (GIP,.42 + GIP3.42), use of side-viewing antibodies to GLP-1 (recognizing N-terminally extended and truncated forms) suggest the molecule is rapidly removed from plasma. In all species tested (rat, pig, dog and human), the tl/2 for clearance of total GLP-1 immunoreactivity is approximately 5 min [58; 429-435]. Correspondingly, the metabolic clearance rate in man is on the order of 10 mL/Kg/min [429-432]. Study of tissue specific degradation of GLP-1 in anesthetized pigs showed, like GIP, renal catabolism is the main site of removal of total GLP-1 immunoreactivity, but also lungs, liver and hindleg were implicated for N-terminal truncation [433]. Earlier experimental surgery procedures to rats indicated the kidneys as the major site of GLP-1 clearance; total GLP-1 levels were elevated during nephrectomy and ureteral ligation, whereas GLP-1 was extracted from the perfusate of isolated kidney [435]. Similarly, uremia increases plasma concentrations of GLP-1 in humans [426]. 123 The circulating half-life of glucagon has been reported to be in the range of 5 to 6 min in dogs and humans [436; 437]. Degradation of glucagon by the liver is controversial. It has been reported that glucagon is metabolized by the liver by the cytosolic enzyme dipeptidyl peptidase I (DPI/Cathepsin C) [438-440]. In reviewing the contribution of the liver to glucagon degradation, Hoist [441] concluded that in dogs as well as humans, there is low hepatic extraction. This conclusion is supported by findings that glucagon was not degraded by passage through the perfused rat liver [442]. Consensus exists that the kidney plays a major role in the metabolic clearance and degradation of glucagon [441], Studies by several groups indicated that after glomerular filtration in the kidney, glucagon is hydrolyzed by brush border enzymes in the proximal tubule [443-445]. Like GIP, there was preliminary evidence showing N-terminal truncation in serum [446], prior to establishment of DPIV as the enzyme responsible for GLP-1 degradation by purified enzyme and serum incubations in vitro [58]. GLP-19_36NH2 was formed by in vitro plasma GLP-I 7 - 3 6 N H 2 incubation and found to be a primary metabolite in humans [447]; exogenously infused GLP-1 7 - 3 6 N H 2 w a s rapidly converted to this truncated form [448]. In the anesthetized rat, IV injections of physiological concentrations of iodinated GLP-17.36NH2 were converted to 125I-GLP-I 9 - 3 6 N H 2 W l t ;h a half time less than two minutes [59]. Administration of specific DPIV inhibitors has further shown a primary role for DPIV-mediated hydrolysis, as inhibition results in preservation of N-terminally intact GLP-1 whether it is endogenous GLP-1 measured [381; 382; 418], or exogenously infused [449]. In vitro studies have indicated that GLP-19.36NH2 may also act as a weak antagonist of the GLP-1 receptor, however, given the proportions of endogenous levels of N-terminally intact to N-terminally degraded peptide, physiological antagonism is unlikely [450; 451]. In pigs, it was reported that the majority of GLP-1 released is degraded by DPIV in capillaries before it reaches target tissues [452], questioning the role of GLP-1 as a physiological incretin. Recently, we have published reports that DPIV is also capable of 124 degrading glucagon by purified enzyme and serum incubations in vitro [328; 453]. As such, previous reports indicating the remarkable stability of glucagon incubated in serum [437; 454] were likely flawed in that antibodies cross-reacted with N-terminally truncated glucagon. Further studies are required to establish the physiological role of DPIV in inactivation of glucagon, although kinetic constants compare well with those for GIP and GLP-1, i.e. physiological substrates, and thus it is likely that DPIV does normally play a role [453]. Hence, studies on the pharmacokinetics of glucagon and GLP-1 must also consider soluble DPIV present in serum and DPIV expressed on the surface of lymphocytes as sites of peptide metabolism. 4.1.4 Thesis Objective In the current study, two main goals were established: (1) domains of GIP responsible for receptor activation and binding were delineated, and (2) the modulation of GIP, GLP-1 and glucagon bioactivity by DPIV cleavage was examined, with a specific aim of generating superactive DPIV resistant peptide analogues for use in vivo. Hence, studies required the use of CHO-K1 cells stably transfected with the GIPR, GLP-1 R or glucagon receptor (Methods sections 2.2 and 2.3), coupled with synthetic peptides (Methods section 2.4). Analogues were tested for cyclic AMP stimulating ability (Methods section 2.7) and receptor binding affinity (Methods sections 2.5 and 2.6). For some peptides, degradation studies (Methods section 2.11) and/or in vivo bioassay in rats (Methods sections 2.13 and 2.14) were performed. The rationale for these studies was the potential use of small molecular weight peptides and/or superactive analogues for the treatment of human type 2 diabetes, or other therapeutic applications for these hormones. 125 4.2 Results 4.2.1 GIP Fragment Analysis 4.2.1.1 Competitive-binding Studies Binding affinities of synthetic peptides were determined by binding displacement assays on CHO-K1 cells transfected with the wild type rat pancreatic GIP receptor (Figure 29). A summary of statistics is shown in Table 7. Of the peptides tested, only GIP,_42, GIP|.30NH2 and GIP7.3ONH2 were able to fully displace l25I-GIP binding (IC,0 values: 3.17 ± 0.3 nM, 2.04 ± 0.73 nM and 23.7 ± 3.7 nM, respectively; n = 4). Truncation of the amino-terminus by 14, 15 and 16 amino acids resulted in C-terminal fragments (GIP,5_30NH2, GIP,6.3()NH2, and GIPRi7.30NH2) that produced half-maximal displacement values in the low micromolar range (Table 7); further truncation at the amino-terminus resulted in IC50 values greater than 10 pM, and thus cannot be determined from the range of peptides tested. Short amino-terminal peptides were less potent at displacing specific 125I-GIP binding: GIP,_42 > G I P ^ ^ H , » > GIP,.140H > GIP,.,4NH2 > GIP,.,3NH2 = GIPM 5 N H2 (Figure 29, Table 7). Peptides GIP,.6NH2, GIP,.7NH2, GIP,.130H and GIP,. , 5 0 H failed to displace significant 125I-GIP under the given assay conditions (Table 7). In order to elucidate key residues contained within amino acids 1-14 of GIP conferring biological activity, an alanine scan was performed. Also because DPIV degradation inactivates GIP, and likely GIPM4, three predicted DPIV-resistant analogues as well as a molecule with a synthetic (3-turn induced in the peptide backbone (BTD) were generated. Binding displacement curves were performed for all peptides, and full curves are presented for selected peptides in Figure 30. Substitution of any residue of the 1-14 primary sequence resulted in significantly reduced binding affinity/ability to displace 1 2 5 I -GIP i_ 4 2 , with the notable exception of [Tyr13]GIP,_ 140H (Figure 30); GIP normally has alanines in position 2 and 13, thus these residues were replaced with those found in glucagon. Residues which may be particularly important for favouring structure for binding or forming contacts with residues of the GIP receptor are 126 suggested by the complete loss of binding potency for GIPM 4 0 H analogues with Ala1, Ala3, Ala4 or Ala5 substitutions (Figure 30B). [D-Ala2] substitution of GIPM 4 0 H was not well tolerated, whereas reduction of the scissile bond between residues 2 and 3 ([Tyr'-Ala2xjj(CH2NH)]GIPM40H gave a similar degree of displacement compared to native sequence, and [L-Pro3]GIPM40H and GIP,. • I I ( B T D )12- I 4 O H were somewhat less potent (Figure 30). Table 7: Summary statistics for studies on synthetic GIP fragments using wtGIPR cells Data represent mean ± S.E.M. of at least 3 independent experiments. Synthetic cAMP Production Receptor Binding Peptide: (Fold Basal\") 10 uM 20 uM % Displacement at 10 uM IC50 (nM) GIP(1-42OH) 119 ± l l * b b 100\" 3.2 ±0.3 1-6NH2 1.27 ±0.18 1.08 ±0.03 . -3.6 + 7.8 -1-7NH2 0.92 ± 0.05 1.06 ±0.06 -6.1 ±3.4 -1-13OH 1.03 ±0.06 1.15 ±0.07 -0.2 ±3.4 -1-13NH2 6.51 ± 1.33* 15.7 ±3.0* 5.0 ± 1.1* -1-140H 88.9 ±9.5* 85.2 ±7.6* 51.3 ± 1.2* -1-14NH2 75.4 ± 10.7* 88.3 ±5.9* 27.9 ±2.8* -1-150H 0.97 ±0.06 0.91 ±0.05 -3.1 ±4.3 -1-15NH2 2.26 ±0.32* 4.37 ±0.51* 4.2 ± 1.7* -1-30NH2 108 ±12*b b 99.8 ± 1.2* 2.0 ±0.7 7 - 3 0 N H 2 0.89 ±0.06 0.85 ±0.03 99.3 ± 1.0* 23.7 ±3.7 15-42OH 1.02 ± 0.10 1.01 ±0.03 83.3 ±0.7* 1270 ±150 15 -30NH2 1.24 ±0.28 1.01 ±0.11 82.7 ± 1.0* 1400 ±310 16 -30NH2 1.04 ±0.06 0.80 ±0.02 82.1 ± 1.9* 2530 ± 450 17 -30NH2 1.13 + 0.09 1.12 ±0.05 81.9 ±2.1* 1540 ±550 19-30NH2 20.1 ± 1.3* 45.0 ± 1.6* 52.3 ±0.6* -a: Basal cyclic AMP = 2.96 ± 0.03 fmol/1000 cells. b: cAMP and Binding experiments with G I P i _ 4 2 and GIP|.30NH2 were only tested at concentrations as high as 1 u.M. *: p < 0.05 versus control conditions, denoting significant increase in cyclic AMP over basal levels or significant displacement of 125I-GIP. 127 o CQ CO 100 80H 60 4 0 -2 0 -0 -0 •11 -9 -7 L o g 1 0 [Peptide] o • • • A • V o 1 - 4 2 o H 1 -140H 1-14NH2 1-30NH2 7-30NH2 1 5 - 4 2 0 H 1 5 - 30NH2 16 - 30NH2 1 7 - 30NH2 1 9 - 3 0 N H 2 Figure 29: Competition-binding displacement curves of synthetic GIP fragments on wtGIPR cells Data represent the mean ± S.E.M. of 3-8 independent experiments. Refer to Table 7 for binding statistics. Refer to sections 2.4, 2.5 and 2.6 of the Methods section for more detail. 128 100 — 80 o 60 CQ CQ 40-20-0-G'Pl-140H J 3 . [Tyr^GIPLMOH [D-Ala^GIPL^oH [Pro^GIPLuoH Y1A2Th(CH2NH)GIP1.140H -*-G\\P 1-11(BTD)12-140H B. S 100 -5.8 -5.5 -5.2 -4.9 -4.6 Log™ [Peptide] -4.3 T ' - N n N - i n i D N M O O O < c o < < < < < < < T - M n t < < < >• < CM < I Q CO o m LU X C N X o < >-Figure 30: Competition-binding of modified GIP,.,4 analogues on wtGIPR cells (A) ConcentraUon-dependent displacement of l25I-GIP,_42 binding by GIP,.140H and substituted peptides. (B) Displacement of GIP tracer by 50 |iM peptide. Peptides are a series of GIP analogues based on amino acids 1-14 of the primary sequence with single amino acid substitutions (to Ala, Ser or Tyr, where indicated), modified N-terminal amino acids (D-Ala2, Pro3), a reduced scissile bond between amino acids 2 and 3, or introduction of a synthetic P-turn mimetic (BTD) beween residues 11 and 12. Data represent the mean ± S.E.M. of > 3 experiments (# = significantly non-zero, * = significantly different from GIPM 4 0 H; P < 0 .05 ) . Refer to sections 2.4, 2.5 and 2.6 of the Methods section for more detail. 129 4.2.1.2 Cyclic AMP studies All peptides were screened on wtGIPR cells initially with a concentration-response curve over the range of 1 pM to 20 pM. Table 7 shows the summary of peptides tested for the 10 and 20 pM responses for non-substituted GIP fragments; peptides showing no cyclic AMP stimulation at high concentration did not demonstrate any activity at lower concentrations (data not shown). The basal cyclic AMP level in wtGIPR cells was 2.96 ± 0.03 fmol/1000 cells (n > 30). Of peptides tested, only GIPM 4 (amidated or free acid) and GIP,9.30 produced large cyclic AMP responses. GIP M 3 N H 2 and GIP,. i 5 N H 2 also showed significant increases in cyclic AMP from basal levels, however, responses were weak compared to GIPM4. Full concentration-response curves for bioactive GIP fragments are shown in Figure 31. GIP,_42 had an EC50 of 377 ± 25 pM for cyclic AMP production, and maximal cyclic AMP generation was 358 ± 46 fmol/1000 cells (n = 6); similarly, GIP,.30NH2 had an EC5 0 of 139 ± 77 pM for cyclic AMP production, and maximal cyclic AMP generation was 326 ± 35 fmol/1000 cells (n = 3). GIPM 4 and GIP19.30 did not stimulate maximal cyclic AMP levels over the concentration range tested, and thus it is difficult to estimate EC50 values. However, given the maximal cAMP production by GIP,_42, the EC50 of GIP,.,4 (C-terminal amide or free acid) was in the micromolar range, and both amidated and hydroxylated forms of GIP,.I4 appear to be full agonists of the receptor (Figure 31). Bioactive GIP fragments had no effect on non-transfected CHO-K1 cells (data not shown). During bioactivity testing of alanine substituted GIPM 4 0 H analogues, a broader concentration range of peptides was used (Figure 32). Thus it was shown that GIP,.140H was indeed a full agonist of the GIP receptor (EC50 = 780 ± 190 nM) although significantly lower potency than full length GIP (EC5() = 239 ± 13 pM), but still producing similar maximal cAMP accumulation in wtGIPR cells (GIP,.42: 312 + 18 fmol/1000 cells; GIPM 4 0 H: 323 ± 50 fmol/1000 cells; Figure 33). DPIV-resistant GIPM 4 0 H analogues, [Pro3]GIPM40H and [Tyr'-Ala2ip(CH2NH)]GIPM40H were also 130 able to stimulate maximal cAMP production, the former being less potent and the latter more potent compared to GIPM 4 0 H (Figure 33). Reversing chirality of position 2 or inducing a (3-turn completely ablated ability to activate the GIP receptor. A concentration of 20 pM was the highest tested for alanine scanning mutants of GIP|.140H (Figure 32). At this concentration, almost all residues appeared to be important for signalling, although this may partially reflect the reductions in binding displacement, which were weak for many peptides tested at 50 pM (Figure 30). Only Ala1, Ser2, Ala3, Ala4 and Tyr13 substituted peptides were able to increase cAMP levels above basal (P < 0.05), although Ala1 and Ala3 were extremely weak. 20 pM native GIPM 4 0 H, as well as Ser2 and Tyr13 substituted analogues were more potent than 1 nM GIP,_42 when examining cAMP production in wtGIPR cells (Figure 33). GIP fragments truncated at the amino terminus have previously been shown to act as potent receptor antagonists [413], and thus the shorter C-terminal fragments were similarly tested for inhibition of GIP-stimulated cAMP production (Figure 34A). The potency of GIP fragments as antagonists paralleled their binding affinities (Table 7). Peptides GIP|5.420H, GIP15_30NH2, GIP,6. 30NH2 a n d GIP|7.30NH2 acted as weak antagonists of the GIP receptor; GIP7.30NH2 was included as a positive control [413]. To further delineate structure-activity relationships of GIP, and as an important control, the ability of GIP17.3UNH2 to antagonize GIP,_|40H and GIP,Q -30NH2 w a s examined. It was hypothesized that the non-overlapping C-terminal GIP17.30NH2 fragment would not antagonize GIP,.140H, whereas it should antagonize GIP,9.30NH2. The results of these experiments are found in Figures 34B and 34C (N.B. data were normalized to 20 pM concentrations of agonist for better comparison, as GIPM 4 0 H is more potent than GIP19.3()NH2; see Figure 31). In confirmation of the proposed hypothesis, 20 pM GIP17.30NH2 was not able to significantly reduce cyclic AMP production stimulated by 20 pM GIP,.|40H, whereas 1 pM GIP7.30NH2 reduced 131 production by 76.6 ± 1.9% (p < 0.05). Furthermore, 20 uM GIP,7.30NH2 reduced 20 uM GIP19. 30NH2-stimulated cAMP by 48.9 ± 4.7% and 1 uM GIP„0 N H 2 by 89.0 ± 1.1% (p < 0.05). i 1 1 1 1 1 1 1 1 1 Basal -12 -11 -10 -9 -8 -7 -6 -5 -4 L o g 1 0 [Peptide] Figure 31: Cyclic AMP production in wtGIPR cells by selected bioactive truncated peptides Data represent mean ± S.E.M. of at least 3 independent experiments. Refer to Table 7 for cAMP data for all GIP peptides tested. Refer to sections 2.4 and 2.7 of the Methods section for more detail. 132 Figure 32: Cyclic AMP production by 20 uM of substituted GIP,.,4 peptides An alanine scan of GIPM 4 0 H was performed (where endogenous Alanines were found, they were substituted with residues found at the corresponding positions of glucagon). Peptides were tested over a concentration range between 0 and 20 uM. Only the highest concentration is shown. (# = significantly greater than basal, * = significantly greater than 1 nM GIP|_42, P < 0.05). Refer to sections 2.4 and 2.7 of the Methods section for more detail. 133 A. i 1 1 1 1 ' 1 1 r Basal -11 -9 -7 -5 L o g 1 0 [Peptide] 0 50 100 150 200 Cyclic AMP (XBasal) Figure 33: Concentration-response curves of intracellular cyclic AMP production in wtGIPR cells by N-terminally modified GIP M 4 peptides (A) Full concentration-response curve for N-terminally modified analogues. (B) Fold-basal responses for maximal cAMP production values measured. Data represent mean ± S.E.M. of at least 3 independent experiments (# = significantly greater than basal, * = significantly different than 1 nM GIP|_42, P < 0.05). Refer to sections 2.4 and 2.7 of the Methods section for more detail. 134 A. Figure 34: Antagonism of native G I P by C-terminal G I P fragments (A) Antagonism of 1 nM GIP,_42 stimulated cAMP production in wtGIPR cells by N- and C-terminally truncated GIP fragments. On average, 1 nM GIP,.42 produced 278 ± 32 femtomoles of cyclic AMP/1000 cells (cf. Figure 31). Antagonism of (B) 20 uM GIP,.,40H or (C) 20 uM GIP19. 3ONH2 by 20 uM GIP,7.30NH2 or 1 uM GIP7.30NH2. Data represent mean ± S.E.M. of >3 independent experiments; * = P < 0.05 versus control (1 nM GIP,.42 or 20 uM GIPM 4 0 H or GIP19.30NH2). Refer to sections 2.4 and 2.7 of the Methods section for more detail. 135 4.2.1.3 Perfused Rat Pancreas and in Vivo Bioassay Given the ability to stimulate cyclic AMP production, GIPM 4 was selected for testing in the perfused rat pancreas. The effect of GIP|9.30 on insulin secretion in the perfused pancreas has already been reported; this peptide exhibited a small but significant response from the endocrine pancreas [337], consistent with its ability to weakly activate the GIP receptor (Figure 31). A peptide gradient from 0-5 nM under high glucose conditions (16.7 mM) was perfused over 40 min. Immunoreactive insulin release profiles are shown for GIP,.30NH2 and GIP,.14NH2 as well as the glucose control (Figure 35). GIPM 4 was found to induce a small, but significant, increase in insulin release (p < 0.05). Integrated responses for insulin release from the perfused pancreas were: GIP,.30NH2: 167.5 ± 62.7 mU, GIPM 4 N H 2: 43.0 ± 8.5 mU, glucose control: 17.1 ± 9.3 mU, over the 40 minute perfusion period (n = 3-4). A bioassay was developed based on the principal that GIP acts to lower blood glucose mainly through an insulin-dependent mechanism (Methods section 2.12). Hence, blood glucose was monitored during intravenous peptide infusion following an IP glucose tolerance test. Infusion of 1 pmol/min/100 g bw of GIP,.420H significantly reduced circulating glucose levels relative to control animals (Figure 36A). Concurrent insulin measurements revealed that GIP,.420H infusion induced a rapid peak in circulating insulin within 10 minutes of glucose injection, followed by a return to baseline at 60 min. In contrast, saline control animals demonstrated a slow rise in insulin, peaking at 30 minutes, and similarly returning to baseline at 60 min (Figure 36B). Because of the sampling times used, it was not possible to determine if GIP treatment enhanced phase-I versus phase-II insulin release, or both; however, recently it was demonstrated that GIP primarily reduces postprandial glucose excursions via augmenting the early phase of insulin release [185]. Synthetic GIP peptides shown to have biological activity on transfected cells and in the perfused pancreas were tested, monitoring only blood glucose. GIP,.420H and GIP,.30NH2 were equally effective in reducing excursions in glycemia, relative to saline controls (Figure 37; 136 p < 0.05 at all time-points). A much greater dose of GIP,_i4NH2 (100 pmol/min/100 g bw) was required to achieve the same effect (p < 0.05 relative to controls at all time-points), whereas the same dose of GIP ]9.30NH2 only slightly reduced the glucose response after IP glucose (Figure 37; p < 0.05 at 20 and 30 minute time-points). Time (min) Figure 35: Pancreatic perfusion of GIP fragments in rats (A) Immunoreactive insulin release from the perfused rat pancreas in response to 16.7 mM glucose, with or without linear peptide gradients of GIPi_30NH2 or GIP,_,4NH2 (0 to 5 nM). (B) Integrated insulin responses for the data shown in (A). Data represent mean ± S.E.M. of 4 experiments; * = P < 0.05 versus glucose control. Perfusions were kindly performed by Nathalie Pamir (M.Sc.) and published in [455]; used with permission. 137 A. (0 (0 CQ fl) (0 o o (D TS O O CQ B. Q . 100-8 0 4 6 0 4 * 40-^ ^ - /P Glucose Injection •—Saline Control (BIP-I-420H 20 30 40 Time (min) • -Sa l i ne Control GIP1-420H IV Peptide Infusion Basal 10 20 30 40 Time (min) 50 60 Figure 36: Bioassay of G I P ^ in anesthetized male Wistar rats Glucose (A) and immunoreactive plasma insulin (B) in anesthetized male Wistar rats on intravenous infusion of saline or GIP,.42, with concurrent intraperitoneal glucose challenge (1 g/Kg). Data represent mean ± S.E.M. of 4 animals; * = P < 0.05 versus saline control. Refer to section 2.12 of the Methods for more details. 138 A . IV Peptide Infusion 1 1 1 1 1 1 r -5 0 10 20 30 40 50 60 Time (min) B . I IV Peptide Infusion I—i 1 1 1 1 1 -5 0 10 20 30 40 50 60 Time (min) Figure 37: Glucose lowering effects of GIP fragments in anesthetized Wistar rats (A) Peptides infused at a dosage of 1 pmol/min/100 g bw versus saline control. (B) Peptides infused at a dosage of 100 pmol/min/100 g bw versus saline control. Peptides in phosphate-buffered saline were infused intravenously, while glucose was administered by intraperitoneal injection. Glucose was monitored by tail vein measurements using a SureStep blood glucose analyzer. Data represent mean ± S.E.M. of 4 animals; * = P < 0.05 versus saline control. Refer to section 2.12 of the Methods for more details. 139 4.2.2 GIP 3 4 2 and Studies on Cellular DPIV in vitro Several experiments were performed to examine the activity of N-terminally truncated GIP3. 4 2, the peptide resulting from DPIV hydrolysis of native GIP. Synthetic GIP3_42 was able to completely displace 12T-GIP,_42 binding, with an IC50 value of 58.4 ± 18.8 nM, significantly right-shifted from the half maximal binding displacement observed for native GIP,_42 at 3.56 ± 0.81 nM (P < 0.05). Cyclic AMP-stimulation by GIP3.42 was absent at concentrations as high as 10 pM (Figure 38A). When preincubated for 15 min at various concentrations, followed by addition of 1 nM GIP,_42 for a 30 min stimulation period, GIP3.42 was able to concentration-dependently antagonize native GIP-stimulated cAMP production by greater than 90% at 10 pM (Figure 38A). Furthermore, when 10 pM GIP342 was included with varied concentrations of GIP, 4 2 in a cAMP dose-response curve, the EC50 value was right shifted 9-fold, but did not affect maximal cAMP production (Figure 38). Taken together, results indicate that GIP3.42 behaves as a pure antagonist (over the concentration ranges tested), and acts in a competitive-reversible fashion (with an irreversible antagonist, a reduction in the maximal agonist stimulated cAMP level would be expected). Addition of dipeptides H2N-Tyr-Ala-OH, H2N-His-Ala-OH or H2N-Ffis-Ser-OH (corresponding to N-terminal dipeptides cleaved from GIP, GLP-1 and glucagon, respectively) in conjunction with GIP3.42 did not produce measurable cAMP production (Figure 38B). In this experiment, the basal cAMP level was 1.2 ± 0.1% of that stimulated by 1 nM GIP,.42, whereas 10 pM GIP3.42 and the dipeptides (10 or 100 pM), alone or in combination produced cAMP levels between 1.2 and 1.6% of 1 nM GIP,.42 cAMP production in wtGIPR cells. Hence, simply the presence of the correct dipeptide is insufficient to restore biological activity to the truncated peptide. The importance of this N-terminal dipeptide was also demonstrated by the complete lack of biological activity of [Ala'-Tyr2]GIP,_42, a full length peptide with only the first two residues switched [324; 456]. 140 In order to quantitate any DPIV activity that may be present on CHO-K1 cells or |3TC-3 cells commonly used for testing GIP bioactivity, a DPIV activity assay was performed. Figure 39A shows a DPIV standard curve using purified porcine DPIV with or without 50 uM isoleucine-thiazolidide; inclusion of the DPIV inhibitor completely blocked production of the chromogenic nitroaniline from Gly-Pro-para-nitroaniline to background levels. Cells were incubated with DPIV substrate in HEPES-buffered physiological saline with or without Ile-thia, (3TC-3 cells were also additionally tested in the presence of depolarizing concentrations of KC1 (30 mM) to test if DPIV activity was present within the secretory granules. When DPIV was inhibited, a non-specific level of substrate hydrolysis was observed in all cases, corresponding to approximately 8-13 u.U/106 cells (Figure 39B). In CHO-K1 cells, specific DPIV activity was 56.5 ± 7.4 u.U/106 cells (transfected and non-transfected cells showed no difference), and (3TC-3 cells varied between 44.7 and 51.9 u.U/106 cells, irrespective of extracellular KC1. Thence, to evaluate the importance of this specific DPIV activity during GIP bioactivity analysis, a concentration-response curve of GIP,.42 was performed on wtGIPR cells with or without 50 u,M Ile-thia; no significant difference was observed with complete DPIV inhibition (Figure 39C), suggesting that this degree of activity is extremely low and can be considered negligible during experimentation. 141 3 - 4 2 10 uM His-Ala - - - + - - + - - - -10 uM His-Ser . . . . + . . + . . . 100 |xM Tyr-Ala + - -100 uM His-Ala + -100 uM Ser-Ala + Figure 38: Competitive inhibition of GIP,^ by GIP3.4 2 (A) Inhibition of 1 nM GIP-stimulated cAMP production by varied concentrations of GIP342 (dashed line) and effect of 10 uM GIP3.42 on GIP,.42's concentration-response curve. Antagonists were added 15 minutes prior to the 30 min stimulation with GIP,_42. Stimulation of cells with 30 min of GIP3.42 alone is also indicated. (B) Effect of 30 min co-incubation of GIP3.42 with N-terminal dipeptides. Each data point represents the mean ± S.E.M. of 3-7 independent experiments, normalized to the cAMP level stimulated by 1 nM GIP,_42. Refer to Methods sections 2.4 and 2.7 for more details. 142 A. DP IV Standard Curve c •j= 120 o ^ £ ° 100 S E Q. O 0 . ° < 5 80-60-40-20H 0-C. Cyclic AMP Production i wtGIPR ( E C 5 0 : 330 ± 127 pM) i + 50 nM lle-Thia ( E C 5 0 : 291 ± 82 pM) \" i 1 -12 -11 -10 -9 -8 -7 Log [GIP] Figure 39: Cell associated DPIV activity (A) A DPIV standard curve with and without 50 uM isoleucine-thiazolidide. (B) Hydrolysis of Gly-Pro-paranitroaniline substrate by cells in the presence or absence of Ile-thia. (C) A concentration-response curve of GIP,.42 on wtGIPR cells in the presence or absence of Ile-thia. Each data point represents the mean ± S.E.M. of 3 independent experiments. Refer to sections 2.7 and 2.11 of the Methods for more detail. 143 4.2.3 Design of DPIV-resistant GIP Analogues Nine modified peptides were designed based on the human GIP,.3nNH2 sequence. Peptides were designed to not conform to the substrate specificity requirements of DPIV. Modified peptides include [D-Ala~]GIPi_30NH2, [Tyr'-Ala2xjj(CH2NH)]GIP, 3ONH2 (i-e- a reduced peptide bond between Ala2 and Glu3), [(P)Ser2]GIP,.30NH2 (phosphoserine modified), [N-MeGlu3]GIP|.30NH2 (N-methylated), and a cyclized peptide, [cyclo(Lys16, Asp21)]GIP|_30NH2, with a 6 amino acid ring. Substituted analogues were also generated with glycine, serine or valine in position two, or proline in position 3. All peptides were tested for DPIV-resistance by MALDI-TOF mass spectrometry (K. Kiihn-Wache, Probiodrug AG, Halle, Germany) [329]. Peptides could be divided into three categories: (1) those with degradation kinetics similar to native GIP,.30NH2, which included only [cyclo(Lysl6,Asp2')]GIP,.30NH2 (t1/2 = ~2 min with purified enzyme), (2) moderately resistant analogues, [Gly2]GIP|.30NH2, [Ser2]GIP].30NH2, and [Val2]GIP,.30NH2 (t,/2 = 137-298 min with purified enzyme), and (3) stable peptides completely resisting degradation by purified enzyme, [D-Ala2]GIPU30NH2, [Tyr'-Ala2ip(CH2NH)]GIP1.30NH2, [(P)Ser2]GIP,.30NH2, [Pro3]GIP|.30NH2 and [N-MeGlu3]GIPl.30NH2. Binding competition and cyclic AMP stimulation studies were performed on wtGIPR cells with all 1-30NH2 based peptides (Figure 40, Table 8). In general, all peptides were able to displace 100% of bound GIP tracer, and showed rank IC50 values: GIP,.30NH2 < [SerJGIP,.^ < [Tyr'-Ala^KCfTNH^GIP,.^^ < [D-Ala2]GIP,.30NH2 < [Gly2]GIP,.30NH2 < [Val2]GIP,.30NH2 < [N-MeGlu3]GIP,.30NH2 < [Pro3]GIP,.30NH2 < [(P)Ser2]GIP,.30NH2 < [cyclo(Lys,6,Asp2')]GIP,. 3 O N H 2 (Figure 40A). However, signalling ability did not correlate well with binding affinity (Figure 40B). [Tyr'-Ala2(CH2NH)]GIP,.30NH2, [Pro3]GnY30NH2, and [N-MeGlu3]GIP, 3 0 N H 2 were all unable to stimulate maximal cyclic AMP production in wtGIPR cells (Figure 40B, Table 8). Most peptides showed variable reductions in cAMP-stimulating potency, whereas only [D-Ala2] or [Ser2] substituted analogues were not significantly different from native GIP. 144 Table 8: Binding and cAMP statistics for modified GIPi. 3 0 N H 2 analogues Analogue IC5„ (nM) EQo (nM) cAMP (%Max) GTP I-30NH2 3.75 ±0.55 0.230 ±0.039 100 [D-Ala2]GIP,.3()NH2 8.52 ±2.3 0.680 ±0.021 94.5 ± 3.7 [Y'A2ib(CH2NH)]GIP,30NH2 12.7 ± 1.9 ND 17.6 ±2.0* [Gly2]GIP].30NH2 11.8 ± 1.6 3.18 =t 0.8* 96.2 ±2.5 [SerJGlP^NH, 4.59 ±0.34 0.289 ± 0.042 91.8 ± 5.1 [(P)Ser2]GIP,.30NH2 93.1 ±8.6* 106 ±7* 92.5 ± 8.4 [Val2]GIP,.3()NH, 17.4 ± 1.8* 11.2 ± 1.3* 91.4 ±3.6 [L-Pro3]GlP,.30NH2 25.1 ±3.6* ND 36.0 ± 6.4* [L-(N-Me)Glu3]GIP1.30NH2 23.6 ±2.8* ND 10.5 ±0.4* Cyclo[Lys,6,Asp2,]GIP,.30NH2 94.9 ±7.9* 3.11 ± 1.43* 112.7 ± 14.7 IC50 and EC50 values were determined by nonlinear regression analysis of curves shown in Figure 40\"(n = 3-7). ND = Not determined (half maximal stimulation was not achieved). * = differ from responses to GIP|.30NH2 by at least P < 0.05 as determined by one way ANOVA. 145 0 -12 -11 -10 -9 -8 -7 -6 -5 • GIP a [D-Ala2]GIP A [Tyr1-Ala2^(CH2NH)]GIP • [Gly2]GIP • [Ser2] GIP o [(P)Ser2]GIP • [val2]GIP o [Pro3]GIP A [N-MeGlu^GIP * [cyclo(Lys16,Asp21)]GIP Figure 40: Binding affinity and cAMP-stimulating ability of modified GIP,. 3 0 N H 2 analogues ( A ) Competitive displacement of l25I-GIP binding from wtGIPR cells. (B) Concentration-response curves on wtGIPR cells. Each data point represents the mean ± S.E.M. of 3-7 independent experiments. Refer to Methods sections 2.4, 2.6 and 2.7 for more details. 146 4.2.4 Characterization of [D-Ala^GIP, .42 in Vitro and in Vivo 4.2.4.1 In Vitro Characterization Given the strong DPIV resistance of [D-Ala2]GIP|.30NH2, and minimal changes in receptor potency, this N-terminally modified analogue was resynthesized as a full length peptide. Initial studies were designed to reassess the DPIV resistance of this molecule and examine its effects on the transfected GIP receptor. Following the procedure for iodination of synthetic porcine GIP,.42, it was possible to label [D-Ala2]GIP,.42 radioactively and identify peaks with intrinsic binding ability (Figure 4). Using the methodology of Kieffer et al [59], it was possible to separate 125I-GIP,.42 from l2T-GIP342 by HPLC (Figure 41). When incubated with purified porcine DPIV, l25I-GIP,_42 (RT = 14.3 ± 0.3 min) was completely degraded to l25I-GIP3.42 (RT = 10.3 ± 0.1 min), as resolved by HPLC (Figure 41A). Similar studies using monocomponent 125I-[D-Ala2]GIPM2 (RT = 16.4 ± 0.2 min) indicated that it was not a substrate of DPIV (Figure 4IB), as no peak corresponding to 125I-GIP3.42 eluted. Receptor binding and biological activity of GIPM 2 and [D-Ala2]GIP,_42 in vitro were not significantly different, fn binding competition assays, regardless of whether the radioligand used was l25I-GIP,.42 or 125I-[D-Ala2]GIP,.42, IC50 values were equivalent (Figure 42). [D-Ala2]GIPM2 showed nearly equal cyclic AMP stimulating potency as native GIP on wtGIPR cells (EC50 values: GIPM2, 183 ± 18 pM, [D-Ala2]GIP,.42, 630 ± 119 pM, p < 0.05; Figure 43). Given the in vitro data, we hypothesized that [D-Ala2]GIP,_42 may have enhanced bioactivity in vivo relative to native GIP, resulting from its DPIV resistance. 147 i 1 1 1 1— i 1 1 1 r~ 5 10 15 20 25 5 10 15 20 25 Time (min) Time (min) Figure 41: Incubation of , 2 SI-GIP 1 4 2 or 1 2 5I-[D-Ala2]GIP1 4 2 with DPIV HPLC separation of l2T-GIP342 from (A) l2T-GIP,.42 or (B) l2T-[D-Ala2]GIP,.42 incubated with (dashed lines) or without (solid lines) 10 mU of dipeptidyl peptidase IV for 16 h. Iodinated peptides were separated by a protocol consisting of 14 minutes at 32% CH3CN/0.1% TFA and a linear gradient to 38% CH3CN/0.1% TFA over 10 min, followed by a further 5 minutes at 38% CH3CN/0.1% TFA. Each trace represents the compiled results from 3-4 chromatograms. Refer to sections 2.5 and 2.11 of the Methods for more details. 148 i 1 1 1 1 1 1 r 0 -12 -11 -10 -9 -8 -7 -6 L o g 1 0 [Peptide] B c i 1 1 1 1 1 1 r 0 -12 -11 -10 -9 -8 -7 -6 L o g 1 0 [Peptide] Figure 42: Binding studies using 1 2 5I-GIPM 2 or 1 2 SI-[D-Ala2]GIPM 2 as tracer Competitive radioligand binding studies on wtGIPR cells using (A) '^I-GIP^ or (B) l25I-[D-Ala2]GIP,.42 as tracer. (A) solid circles, GIP,_42 (IC50 = 2.23 ±0.18 nM), solid diamonds, [D-Ala2]GIP,.42 (ICJO = 3.48 ± 0.20 nM). (B) solid circles, GIP,_42 (IC50 = 2.33 ± 0.18 nM), solid diamonds, [D-Ala2]GIP,_42 (IC50 = 2.45 ± 0.24 nM). Each data point represents the mean ± S.E.M. of 4-5 independent experiments. Refer to Methods sections 2.4, 2.5 and 2.6 for details. 149 i 1 1 1 1 1 1 r Basal -12 -11 -10 -9 -8 -7 -6 L o g 1 0 [Peptide] Figure 43: Bioactivity of [D-Ala2]GIPi.42 in vitro Cyclic AMP production in wtGIPR cells by GIP,.42 (solid circles) and [D-Ala2]GIP,.42 (solid diamonds). EC50s: GIP,.42, 183 ± 18 pM, [D-Ala2]GIPM2, EC50 = 630 ± 119 pM (p < 0.05). Each data point represents the mean ± S.E.M. of 6-8 independent experiments. Refer to Methods sections 2.4 and 2.7 for details. 150 4.2.4.2 In Vivo Bioassay of GIP and [D-Ala2]GIP in Lean and Obese Rats. Initial experiments were performed with conscious Wistar rats (287 ± 6.5 g bw, fasting glycemia = 4.7 ± 0.1 mM, fasting insulin = 87.6 ±11.6 pM, fasting GIP = 606 ± 82 pg/mL; n > 20). Subcutaneous injection of 8 nmol/Kg bw GIP,_42 with a concurrent oral glucose tolerance test significantly reduced the glycemic profile relative to the saline control, and this was associated with increased circulating insulin levels (Figure 44). Measurement of GIP levels by RIA indicated that this dose of GIP resulted in a 10-fold greater peak GIP level (control: 1.68 ± 0.17 ng/mL, treated: 16.6 ± 3.0 ng/mL) during the OGTT. The exogenous GIP appeared to be rapidly absorbed and to follow a similar elimination profile to the endogenous GIP in control animals (Figure 44C). In contrast, the same dosage of GIP3.42 had no effect on postprandial glycemia or insulin release (Figure 45). Subcutaneous injection of [D-Ala2]GIPM2 resulted in a more pronounced reduction in the glycemic profile and an enhanced insulin time-course than native GIP during an oral glucose tolerance test (Figure 44). Notably, both GIP and D-Ala2-modified GIP appeared to exert their glucose lowering effects by significant enhancement of the early phase of insulin release, while the latter peptide displayed more protracted bioactivity. Integrated glucose and insulin profiles can be found in Table 9. Studies in humans with type 2 diabetes and rodent disease models have suggested that the GIP effect on insulin release and glycemia is blunted when infused at physiological concentrations [220; 375-379]. However, these studies did not address the possible therapeutic application of GIP in diabetes when administered at higher concentrations. Thus, comparison of GIP and [D-Ala2]GIP was subsequently studied in the VDF Zucker animal model of type 2 diabetes [379]. Age matched obese animals (fa/fa; 576.1 ± 9.1 g bw) displayed significantly higher fasting glycemia than their lean (Fa/?; 335.6 ± 14.9 g bw) littermates (7.3 ± 0.3 mM versus 4.8 ± 0.1 mM) and, similarly, a fasting hyperinsulinemia (979 ± 109 pM versus 8.5 ± 2.4 pM), typical of this animal model (p < 0.05, n > 20). Fasting GIP levels in these animals were 151 not significantly different (lean: 954 ± 72 pg/mL, obese: 926 ±110 pg/mL; n ^ 13), whereas following an OGTT in control animals, postprandial GIP levels (the mean of samples taken at t = 10, 20 and 30 min) in obese rats (1730 ±170 pg/mL, n = 15) were significantly greater (p < 0.05) than levels in lean animals (1150 ± 120 pg/mL, n = 19). Injection of 8 nmol/Kg bw GIP|_42 resulted in 15.4-fold (lean) and 9.6-fold (obese) greater peak GIP levels in vivo after the OGTT, relative to peak values in saline control animals. In lean animals, GIP injection produced moderate reductions in postprandial glycemic levels (16.8% reduction, compared to saline control at t = 40 min using fold-basal values), whereas [D-Ala2]GIPi.42 was more potent (46.8% reduction at the same time-point) (Figure 46A). Similarly, in obese animals, comparison at the t = 40 min time-point indicated that GIP reduced glycemia by 18.7% and [D-Ala2]GIP by 41.5%, relative to the saline control (Figure 47A). In lean VDF Zucker rats, both peptides appeared to augment insulin release similarly, with [D-Ala2]GIP resulting in more elevated insulin levels at the first time-point (3.5 min) (Figure 46B). However, in obese rats, differences in the potencies of GIP and DPIV-resistant GIP were more evident, with insulin levels remaining at near peak values at the 60 min time-point for the [D-Ala2]GIP treated group, while insulin levels approached control values for the GIP treated group after one hour (Figure 47B). Integrated glucose and insulin profiles for bioassay data can be found in Table 9. 152 I 1 1 1 1 1 1 1 0 10 20 30 40 50 60 70 Figure 44: Bioassay of GIP 1 4 2 and [D-Ala2]GIP, .42 in Wistar rats Bioassay of G I P M 2 (solid circles) and [ D - A l a 2 ] G I P , . 4 2 (solid diamonds) in conscious unrestrained male Wistar rats, compared to a saline control (solid squares). (A) Whole blood glycemia measured from tail vein samples. (B) Immunoreactive plasma insul in levels from tail vein samples. (C) Comparison of immunoreactive G I P profiles in control animals (endogenous G I P , solid squares, right axis) and those receiving SC G I P injection (endogenous + exogenous G I P , solid circles, left axis). Peptides (8 n m o l / K g bw in 500 p L saline) were injected SC at time 0, immediately fol lowing an O G T T (1 g / K g bw). Each data point represents the mean ± S . E . M . of 8 animals; * = p < 0.05 versus saline control, # = p < 0.05 between peptides. Refer to sections 2.12 and 2.13 for details. 153 A 4 J 0 10 20 30 4 0 50 6 0 Time (min) 600 -1 0 10 2 0 30 4 0 50 6 0 Time (min) Figure 45: Bioassay of GIP3.4 2 in conscious Wistar rats (A) Whole blood glycemia measured from tail vein samples. (B) Immunoreactive plasma insulin levels from tail vein samples. GIP 3. 4 2 (open squares, 8 nmol/Kg bw in 500 uL saline) or saline (solid squares) was injected SC at time 0, immediately following an OGTT (1 g/Kg bw). Each data point represents the mean ± S.E.M. of a5 animals. Refer to sections 2.12 and 2.13 of the Methods section for details. 154 A T3 O O 1 2 J § 10\" o i 8 ^ CD 6^ 4 ^ # *# *# B 800-, 0 10 20 30 40 Time (min) 50 60 Q_ 600-c 400-(/) i or 200H o J r i 1 1 1 1 1 r 0 10 20 30 40 50 60 Time (min) Figure 46: Bioassay of [D-Ala2]GIP, . 4 2 in conscious lean VDF Zucker rats Bioassay of GIP|.42 (solid circles) and [D-Ala2]GIP,.42 (solid diamonds) in conscious lean (Fa/?) VDF Zucker rats versus injection of saline (solid squares). (A) Whole blood glycemia measured from tail vein samples. (B) Immunoreactive plasma insulin levels from tail vein samples. Peptides (8 nmol/Kg bw in 500 uL saline) were injected SC at time 0, immediately following an OGTT (1 g/Kg bw). Each data point represents the mean ± S.E.M. of 6-8 animals; * = p < 0.05 versus saline control, # = p < 0.05 between peptides. Refer to sections 2.12 and 2.13 of the Methods section for details. 155 X. 2 0 - 1 E 1 8 -8 16\" o o 3 1 4 -O 1 2 -1 0 -8 -6 -•o o o CQ B 5 0 0 0 - , s 4 0 0 0 -C 3 0 0 0 -\" 5 c 2 0 0 0 -1 0 0 0 -0 -I—I—I 0 1 0 2 0 3 0 4 0 Time (min) r 0 1 0 2 0 ~ 3 0 ~ 4 0 Time (min) 5 0 6 0 5 0 6 0 Figure 47: Bioassay of [D-Ala2]GIP, ,42 in conscious obese VDF Zucker rats Bioassay of GIP,.42 (solid circles) and [D-Ala2]GIP,_42 (solid diamonds) in conscious obese (fa/fa) VDF Zucker rats versus saline injection (solid squares). (A) Whole blood glycemia measured from tail vein samples. (B) Immunoreactive plasma insulin levels from tail vein samples. Peptides (8 nmol/Kg bw in 500 uL saline) were injected SC at time 0, immediately following an OGTT (1 g/Kg bw). Each data point represents the mean ± S.E.M. of 6-8 animals; * = p < 0.05 versus saline control, # - p < 0.05 between peptides. Refer to sections 2.12 and 2.13 of the Methods section for details. 156 Table 9: Integrated glucose and insulin profiles for GIP and [D-Ala2]GIP in vivo Animal Model Test Glucose Profile (mM X 60 min) Early Insulin Response ( p M X 10 min) Complete Insulin Profile (pM X 60 min) Wistar Rat Saline GIP [D-Ala 2 ]GlP 289 ± 21 1 9 5 ± 2 6 t 145 ± 17f 2538 ± 545 4271 ± 4 9 3 t 4979 ± 848 f 13258 ± 2 3 0 5 14385± 1225 20589 ± 2936 Lean Zucker Rat Saline GIP [D-Ala 2 ]GlP 315 ± 21 2 5 8 ± 15 f 113± 13 t J 1433 ± 2 0 2 3225 ± 5 3 2 t 4168 ± 5531\" 15984 ± 1512 18570 ± 1094 21178 ± 3759 Fat Zucker Rat Saline GIP [D-Ala 2 ]GlP 503 ± 68 3 6 8 ± 4 3 t 2 2 6 ± 3 2 n 4463 ± 1584 1 6 7 6 3 ± 2 6 4 l f 20467± 1177f 4 4 9 1 2 ± 15378 9 3 4 8 8 ± 1 5 6 7 4 t 165573± 14792 n *Data represent the area under the curve (AUC) of data shown in Figures 44, 46, and 47. p < 0.05 compared to saline control. { : p < 0.05 compared to GIP test. 4.2.5 Parallel Comparison of [Ser2] and [(P)Ser2] Substituted GIP and GLP-1 Although [Ser'jGIP,. 30NH2 w a s o n ' y moderately resistant to DPIV degradation [329], it had favourable binding and signalling characteristics at the cloned GIP receptor (Figure 40, Table 8). [Ser]GLP-l 7 - 3 6 N H 2 has been previously reported to display enhanced antidiabetogenic activity owing to its reduced degradation by DPIV [457-459]. In our examination of degradation of glucagon, which normally has a serine in position 2, we found that modification of serine with a phosphate group, [(P)Ser], resulted in complete resistance to purified DPIV, and this molecule was time-dependently dephosphorylated in serum (see next section). Hence, a study was designed to examine GIP, GLP-1, along with [Ser2] and [(P)Ser2] substituted analogues in parallel. Analogues were tested in vitro for DPIV resistance (K. Kiihn-Wache; t,/2 values: native peptides, ~2 min; [Ser2]incretins, ~137-686 min; [(P)Ser2]incretins, completely resistant), and potency on cells transfected with the GIP or GLP-1 receptors. In vivo studies followed the conscious unrestrained male Wistar rat bioassay protocol with respect to glycemic excursions and insulin profiles (Methods section 2.12). 157 4.2.5.1 ln Vitro Binding and Bioactivity Initial studies were directed at characterizing peptide analogues in vitro. Using CHO-K1 cells transfected with the GIP or GLP-1 receptor (GIPR and GLP-1R cells, respectively), it was possible to compare [Ser] and [(P)Ser]-substituted incretin analogues, with respect to receptor binding affinity and ability to activate the adenylyl cyclase/cyclic AMP cascade. Native GIP|.42 was able to displace 50% of 125I-GIP binding to GIPR cells at a concentration of ~4.5 nM (Figure 48A). The IC50 of [SerJGIP was slightly greater (1.4X; P > 0.05), and the affinity of [(P)Ser]GIP was significantly reduced 21X (P < 0.05; Figure 48A). All three peptides were able to fully displace specific 125I-GIP binding. When examining GIP receptor activation by these peptides, a similar pattern was observed. GIP and [Ser]GIP displayed EC5 0 values ranging between 245-289 pM (P > 0.05), but the concentration-response curve for [(P)Ser]GIP was significantly right shifted (433X; P < 0.05; Figure 48B). However, the GIP analogues were able to produce maximal cyclic AMP stimulation in GIPR cells similar to that of native hormone (Figure 48B). Receptor binding of GLP-1 and analogues to the transfected GLP-1 receptor are shown in Figure 49A. The binding affinity of GLP-1 for the GLP-1 receptor was similar to that of GIP for the GIP receptor, having an IC50 on the order of ~4.3 nM. The concentration of [Ser2]GLP-l required to displace 50% of l25I-GLP-l binding was ~8.1 nM, or 1.9X lower affinity than native hormone (P > 0.05; Figure 49A). As was the case with [(P)Ser]GIP peptide, [(P)Ser]GLP-l showed significantly lower affinity for its respective receptor, with an IC50 value right shifted by 284X (P < 0.05). GLP-1, [Ser2]GLP-l and [(P)Ser2]GLP-l were all able to fully compete with 125I-GLP-1 binding to the GLP-1 receptor (Figure 49A). The cyclic AMP EC50 values parallel results for binding affinity, with rank potency being: GLP-1 > [Ser]GLP-l > [(P)Ser2]GLP-l. While the potency of [Ser^GLP-l was found to be 3.3-fold lower than native GLP-1, statistical 158 X significance was not reached (P > 0.05), however, the EC5 0 value of [(P)Ser]GLP-l was significantly reduced by three orders of magnitude (P < 0.05; Figure 49B). i 1 1 1 1 1 1 1 r 0 - 1 2 -11 - 1 0 -9 -8 -7 -6 -5 Log [Peptide] Figure 48: Binding and cAMP stimulation in wtGIPR cells by GIP, 4 2, [Ser2]GIP, 3 0 N H 2 and [(P)Ser2]GIP1.30NH2 (A) Competitive-binding inhibition of 1 2 5 I -GIP i_ 4 2 . IC50 values (nM): G1P,42, 4.49 ± 0.55; [Ser^GIP^oNH,, 6.40 ± 0.73; [(P)Ser2]GIP,.30NH2, 93.1 ± 8.6. (B) Cyclic AMP\" stimulation in wtGIPR cells. ECJO values (nM): GIP,.42, 0.245 ± 0.063; [Ser]GIP,.30NH2, 0.289 ± 0.042; [(P)Ser2]GIP, -30NH2' 106 ± 7. Each data point represents the mean ± S.E.M. of > 4 independent experiments. Refer to sections 2.4, 2.5, 2.6 and 2.7 of the Methods for details. 159 i 1 1 1 1 1 1 i r 0 -12 -11 -10 -9 -8 -7 -6 -5 Log [Peptide] Figure 49: Binding and cAMP stimulation in wtGLP-lR cells by GLP-17 3 6 N H 2 , [Ser2]GLP-I7 .36NH2 and [(P)Ser2]GLP-l7.36NH2 (A) Competitive-binding inhibition of l 2 5 I -GLP-l 7 _ 3 6 N H 7 . IC 5 0 values (nM): GLP-1 7 _ 3 6 N H 2 , 4.26 ± 1.19; [Ser 2]GLP-l 7 . 3 6 N H 2 , 8.08 ± 3 . 1 1 ; [(P)Ser2]GLP-\"l7.35NH2, 1210 ± 430. (B) Cyclic AMP stimulation in wtGLP-lR cells. EC, 0 values (nM): GLP-1 7 . 3 6 N H 2 , 0.155 ± 0.042; [Ser2]GLP-l7. 3 6 N H 2 , 0.509 ± 0.119; [(P)Ser]GLP-l 7-36NH2) 513 ± 107. Each data point represents the mean ± S.E.M. of > 4 independent experiments. Refer to sections 2.4, 2.5, 2.6 and 2.7 of the Methods for details. 160 4.2.5.2 In Vivo Bioassay Initial experiments were performed with conscious Wistar rats (312 ± 20 g bw, fasting glycemia = 4.3 ± 0.1 mM, fasting insulin = 72.2 ± 7.5 pM, n = 54; fasting GIP = 897 ± 49 pg/mL, fasting GLP-1 = 13.4 ± 1.2 pg/mL; n a 9). Subcutaneous injection of 8 nmol/Kg bw GIPi.42 or GLP-17.36NH2 with a concurrent oral glucose tolerance test significantly reduced the glycemic profile relative to the saline control, and this was associated with increased \"early phase\" circulating insulin levels (Figure 50); GIP reduced the integrated glycemic excursion by 27.7% and GLP-1 was slightly more effective, reducing the area under the curve by 33.9% (P > 0.05, n = 6; Table 10). Measurement of hormone levels by RIA indicated that this dose of peptide resulted in a 9-fold greater peak GIP level (control: 1.60 ± 0.10 ng/mL, treated: 14.3 ± 0.4 ng/mL) and a 2-fold greater peak GLP-1 level (control: 24.0 ± 5.7 pg/mL, treated: 48.4 ± 6.0 pg/mL) during the OGTT. The difference in the values for GIP and GLP-1, despite identical dosages likely reflects the differing pharmacokinetics of the two hormones. Given that their insulinotropic potency was similar, it can be inferred that the mode of action of GIP and GLP-1 is rapid, as is their inactivation by N-terminal dipeptide cleavage, such that peptide pharmacokinetics do not noticeably affect their biological activity in this assay system. Subcutaneous injection of [Ser]GIP,.30NH2 resulted in a slightly more pronounced reduction in the glycemic profile and an enhanced insulin time-course during an oral glucose tolerance test, as did [Ser2]GLP-l (Figure 50, Table 10), however, insulin profiles were not remarkably different from those observed during native hormone administration. Equivalent doses (8 nmol/Kg) of [(P)Ser] substituted incretin analogues reduced the blood glucose profile between 31 and 44%, with [(P)Ser2]GIP,.30NH2 being more potent than [(P)Ser]GLP-l7 -36NH2' however, insulin responses at the first time-point sampled (3.5 min) were greatly reduced compared to other peptides tested, and were not significantly different from the control values. Increasing the dosage of phosphoserine2 incretin analogues to 80 nmol/Kg replaced this deficit, allowing greater reduction 161 in the glycemic profile, and significantly enhanced insulin responses (Figure 51, Table 10). Differences between GIP and GLP-1 analogues with [(P)Ser2] substitution were less pronounced than for native or [Ser2] substituted peptides. Despite the difference in shape of the glycemic profile for GIP and GLP-1, the overall antidiabetic potency of the peptides appeared to be very similar. Table 10: Integrated glucose and insulin profiles for GIP, GLP-1, [Ser2]- and [(P)Ser2]-substituted analogues in vivo Test Glucose Profile Early Insulin Response Complete Insulin Profile (mM X 60 min) (nM X 10 min) (nM X 60 min) Saline 292+ 16 1.91 ±0.35 13.5 ±2.0 GIP,.42 211 ± 18* 3.35 ±0.38* 15.7 + 3.4 [Ser^GIP.jo 188 ±24* 3.31 ±0.55* 16.9 ±2.4 [(P)Sei2]GIP1.M 163 ±29* 2.21 ±0.50 15.2 ± 1.2 10X [(P)Ser2]GIP,.30 138 ±30* 3.35 ±0.75* 19.5 ±3.9* GLP- 117-36NH2I 193 ±26* 2.73 ±0.60 14.0 ±2.4 [Ser]GLP-l l 7. 3 f ) N H 2, 156 ±27* 2.99 ±0.39* 17.2 ±4.3 [(P)Sei 2]GLP-l„. 3 6 N H2i 202 ± 30* 2.29 ± 0.45 14.6 ±2.4 10X [(nSe^GLP-lp^Hj, 123 ± 16* 3.65 ±0.69* 16.7 ±5.5 Data represent the area under the curve (AUC) of data shown in Figures 50 and 51. = P < 0.05 compared to saline control 162 A . i 1 1 1 1 1 r 0 10 20 30 40 50 60 B . § 600- * * o-J i 1 1 1 1 1 r 0 10 20 30 40 50 60 Time (min) Figure 50: Bioassay of native and Ser2-substituted incretin analogues in Wistar rats Bioassay of GIP,.42, [Ser]GIP|_30NH2, GLP-17.3 6 N H 2 and [Ser2]GLP-l7.36NH2 in conscious unrestrained male Wistar rats, compared to a saline control. (A) Whole blood glycemia measured from tail vein samples. (B) Immunoreactive plasma insulin levels from tail vein samples. Peptides (8 nmol/Kg bw in 500 uE saline) were injected SC at time 0, immediately following an OGTT (1 g/Kg bw). Each data point represents the mean ± S.E.M. of 6 animals; * = p < 0.05 versus saline control, # = p < 0.05 between peptides. Refer to sections 2.12 and 2.13 of the Methods for details. 163 A . — 12-0) 10-4J Saline -\"-[(P)Ser2]GIP a-[(P)Ser 2]GLP-1 - • -10X [(P)Ser2]GIP -o- 10X [(P)Ser2]GLP-1 0 i— 10 20 — i — 30 —i— 40 —i— 50 ~60 B. * o-J i 1 1 1 1 1 r 0 10 20 30 40 50 60 Time (min) Figure 51: Bioassay of Phosphoser2-substituted incretin analogues in Wistar rats Bioassay of [(P)Ser2]GIPL.3NNH2 and [(P)Ser2]GLP-l7.36NH2 in conscious unrestrained male Wistar rats, compared to a saline control. (A) Whole blood glycemia measured from tail vein samples. (B) Immunoreactive plasma insulin levels from tail vein samples. Peptides (8 or 80 nmol/Kg bw in 500 pL saline) were injected SC at time 0, immediately following an OGTT (1 g/Kg bw). Each data point represents the mean ± S.E.M. of 6 animals; * = p < 0.05 versus saline control, # = p < 0.05 between peptides. Refer to sections 2.12 and 2.13 of the Methods for details. 164 4.2.6 DPIV Degradation of Glucagon and DPIV-resistant Glucagon Analogues Mass spectrometry studies with serum and purified DPIV suggested that glucagoni_29 was also a substrate for DPIV [328; 453]. Purified pork kidney DPIV hydrolyzed glucagon,.^ to glucagon3.29 and glucagon,_29 in vitro and, in human serum, it was converted first to glucagon3.29, and subsequently its amino terminus is cyclized by a serum enzyme (possibly to pyroglutamyl-glucagon3.29 {[pGlu3]glucagon3_29}), thus preventing further DPIV degradation. Thus, in order to characterize further the potential role these N-terminally truncated peptides might have, they were bioassayed in vivo and tested for binding and stimulatory action at the cloned transfected human glucagon receptor. Furthermore, a series of N-terminally modified glucagon analogues was generated to test if DPIV resistance may confer enhanced in vivo bioactivity. 4.2.6.1 DPIV Cleavage Inactivates Glucagon In an initial study into the effects of purified porcine DPIV on glucagon, 2 nmol of glucagon,_29 dissolved in 1 mL PBS, was incubated with purified DPIV (0.31 Units) with or without 50 pM Isoleucine-thiazolidide at 37°C for 3.25 hours. When injected IV (~7.1 nmol/Kg) into anesthetized male Wistar rats (225-275 g; not fasted), glycemic profiles for control glucagon injection (incubation in saline alone) and that observed for injection of glucagon incubated with DPIV and Ile-thia were superimposable (Figure 52). In contrast, injection of peptide incubated with DPfV alone showed little effect on the blood glucose profile. Mass spectrometry identified glucagon3.29 and glucagon5.29 as degradation products with purified enzyme, and glucagon3.29 and [pGlu3]glucagon3.29 were identified in serum incubations [328; 453]. Bioassay of these peptides by subcutaneous injection (71 nmol/Kg) produced no significant alteration from basal glycemia, while glucagon,_29 (7.1 nmol/Kg) caused mobilization of glycogen stores to potently raise blood sugar levels (Figure 53). Incubation of monocomponent l25I-glucagon,_29 with purified DPIV without Ile-thia in PBS for 2 hours at 37°C prior to separation by HPLC allowed identification of 165 a single peak with a slightly longer retention time to that of control tracer incubation or incubation in the presence of Ile-thia (Figure 54). It is likely that the gradient conditions were not suitable for the separation of glucagon3.29 and glucagon,_29. 0) CO y CO O °? -a 2 o o O LL ca \" 1.5-1.4-1.3-1.2-1.1-1.0-0.9-0 \" T \" 5 r-10 - 1 — 15 20 Time (min) 25 Glucagon + DPIV + DPIV + lle-Thia 30 Figure 52: Inhibition of DPIV degradation of glucagon by Ile-thia monitored by bioassay Glucagon (2 nmol) was incubated (37°C, 3.25 h) in saline with purified DPIV in the presence or absence of Ile-thia (50 uM), prior to intravenous injection in anesthetized male Wistar rats. Data are mean ± S.E.M. (n = 4); * p < 0.05. Refer to section 2.12 of the methods for more details. 166 i 1 1 1 1 1 r 0 10 20 30 40 50 60 Time (min) Figure 53: Bioactivity of glucagon and N-terminally truncated analogues in vivo Glucagon (7.1 nmol/Kg) or N-terminally truncated glucagon analogues (71 nmol/Kg) were injected SC into conscious unrestrained (fed, not-fasted) male Wistar rats. Blood glucose was monitored via the tail vein and a hand-held glucometer. Data are mean ± S.E.M. (n = 6). Experimental work for this figure was performed by J. Andrew Pospisilik (B.Sc.) and published in [453]; used with permission. 167 HPLC Profile of pure 125l-Glucagon 0 10 20 30 40 Time (min) HPLC Profile 125l-Glucagon Incubated With Purified Porcine DPIV RT+ lle-Thia (Peak 1): 22.97 ± 0.05 Min (n = 5) Glucagon + DPIV (2 hr, 37C) RTPeak 1: 23.03 ± 0.05 Min (n = 6) + 50 [iM lle-Thia \\ RJ p e a k 2: 23.75 ± 0.06 Min (n = 6) i 1 1 1 r -0 10 20 30 40 Time (min) Figure 54: HPLC separation of 125I-glucagon incubated with porcine DPIV HPLC separation of l25I-glucagon incubated with DPIV in the presence or absence of 50 uM Ile-thia was performed as described in the text. Traces are representative chromatograms of 5-6 HPLC runs; compiled (mean ± S.E.M.) retention times (RT) are indicated on the figure. Methodology was similar to that used for separation of GIP fragments by HPLC (Methods section 2.11). 168 4.2.6.2 In Vitro Characterization of N-terminally Truncated Glucagon Fragments Stimulation of cyclic AMP production in CHO-K1 cells transfected with the human glucagon receptor (hGlucR cells) by glucagon,.,,,, glucagon3.29, [pGlu3]glucagon3.29 and glucagon5_29 is shown in Figure 55. A summary of the statistical analysis is shown in Table 11. Fragments were all partial agonists of the glucagon receptor, with the rank of potency being: [pGlu3]glucagon3.29 > glucagon3_29 > glucagon,_29. [Glu9]glucagon2.29 was included in antagonism experiments as a positive control since it is a well characterized antagonist [460]; [Glu9]glucagon2_,9 exhibited a small but significant concentration-dependent increase in intracellular cyclic AMP content of hGlucR cells (2.7 times basal). Since the glucagon fragments were only partial agonists, they were also tested for possible antagonist activity (Figure 56). Glucagon5.29 was found to antagonize cAMP stimulation by 1 nM glucagon,.29(2 pM: p< 0.05; 10 pM: p < 0.01), although, to a lesser degree than [Glu9]glucagon2.29. Glucagon5_,9 was an approximately 11-fold weaker antagonist. Neither glucagon3.29 nor [pGlu3]glucagon3.29 were found to antagonize glucagon,.29 activity. Competition binding experiments on hGlucR cells are shown in Figure 57 and affinity statistics in Table 11. Glucagon,.29 and [Glu9]glucagon2.29 exhibited approximately equal affinity for the glucagon receptor. All other truncated peptides showed significantly lower affinity for the human glucagon receptor than glucagon,.,9 under the given assay conditions. [pGlu3]glucagon3.29 and glucagon,.,, both had approximately 5-fold lower affinity in binding competition experiments, whereas glucagon3.29 had 18-fold lower affinity for the receptor. 169 4 0 r I I I I I I I I L Basal -11 -10 -9 -8 -7 -6 -5 L o g 1 0 [Peptide] Figure 5 5 : Concentration-dependent stimulation of cAMP production in hGlucR cells by glucagon and synthetic fragments CHO-K1 cells stably transfected with the human glucagon receptor were incubated in the presence or absence of varying concentrations of peptides with 0.5 mM IBMX for 30 min at 37°C. Intracellular cAMP content was measured by radioimmunoassay. Data are the mean ± S.E.M. of 3-4 independent experiments. Refer to Table 11 for statistics. See Methods section 2.7 for details. 170 CD _ JO o 2 i S u- *: \"o a> CO o o \" e o o O ~o o o 2.0-1.84 1.6-1.44 1.2-5 1 C H Glucagon-,.29 [D-Ser2]Glucagon [(P)Ser2]Glucagon [Gly2]Glucagon [D-Gln3]Glucagon Basal 10 20 30 45 Time (min) 60 Figure 59: Bioactivity of glucagon analogues in vivo Equimolar doses (7.1 nmol/Kg) of glucagon or analogue were injected subcutaneously into unrestrained conscious rats. Whole blood glucose was measured from the tail vein using a SureStep glucose analyzer. Data are the mean ± S.E.M. of a minimum of 4 animals (* = P < 0.05). Experimental work for this figure was performed by J. Andrew Pospisilik (B.Sc.) and published in [328]; used with permission. 176 4.3 Discussion 4.3.1 GIP Fragments The explicit prerequisite for incretin-induced insulin release is the need for hyperglycemic conditions. Thus, unlike other non-endogenous insulinotropic agents used in the treatment of type 2 diabetes, the incretins are unable to act inappropriately to stimulate insulin release during euglycemia. It is this unique feature which has led to recent interest in the incretins as a novel therapy for diabetes. Clinical trials have been restricted to GLP-1 [461], but administration of peptide analogues of both GLP-1 [457-459] and GIP [462; 463] with prolonged circulating half-lives, as well as inhibition of dipeptidyl peptidase IV [381; 382; 417-419; 449], a physiological regulator of incretin activity, have both been shown to produce improved glucose tolerance in experimental models. Although some populations of type 2 diabetic patients have been reported to show decreased responsiveness to GfP, while responses to GLP-1 were greater [220; 373; 375; 376; 378], |3-cell sensitivity to GIP improved with glyburide treatment [464]. Even though GIP and GLP-1 are proposed to improve glucose tolerance via insulin release, recently there has been some interest in use of these peptides in treatment of type 1 diabetes. The enteroinsular axis was preserved in type 1 diabetes [465], and exogenous GIP [378] or GLP-1 [466; 467] improved glucose tolerance to a degree, presumably via enhancing glucose uptake, delaying gastric emptying and/or inhibiting glucagon release. Therefore, GIP analogues may be useful in treatment of diabetes, and it is important to develop a full understanding of the mode of action of GIP. In the last three decades, considerable effort has been targeted at structural determinations of small peptide hormones. Because short peptides are in general flexible, conformation depends on peptide concentration, solvent composition and other molecules present in solution [468]. X-ray diffraction studies have been limited to glucagon, although these studies compare well with solution structure determinations [468]. Solution structure determination techniques, circular 177 dichroism (CD) and nuclear magnetic resonance (NMR), have been performed on most members of the glucagon/secretin/VIP family [468-474], with the exception of GIP and PHI (peptide histidine isoleucine). In order to stabilize secondary structure, all studies have been performed in the presence of organic solvents or micelles, as there was little evidence of stable structure in water alone [468; 470]. The general structural features of the glucagon superfamily appear to be a disordered N-terminal region of 6 to 8 amino acids, followed by a helical region of 18 ± 2 amino acids, that can be a continuous helix, or broken into two segments by a hinge region of 2 amino acids [468; 470-474]. The exendin peptides from Helodermatidae venom represent a family of peptides structurally related to the glucagon superfamily [475]. Structural determination has been completed for helodermin (exendin-2), which acts as an agonist at VIP and secretin receptors, indicating that it retains significant secondary structure in water, and this is enhanced by organic solvents [476]; helodermin has a core a-helix of 15 amino acids in water, and this extends to 21 amino acids with the addition of trifluoroethanol (TFE). It is thought that helical structure is the preferred conformation for receptor binding, since changing the experimental conditions from aqueous to organic approximately mimics the situation in vivo for a blood borne hormone going from solution to a membrane bound receptor-hormone complex [476]. Notably, related Helodermatidae venom peptide exendin-4 (a GLP-1 receptor agonist) and exendin-4(9_39)NH2 act as weak GIP receptor antagonists [276-278]. A long-term goal of structural studies is to compare NMR solution structures and effects of peptide deletions and substitutions on hormone bioactivity to determine whether there are common or specific structural features relevant to the modes of peptide action [476]. In a unique study, Inooka et al [All] were able to use NMR spectroscopy to determine the structure of PACAP in the receptor bound state; from their results, they hypothesized a two-step ligand transportation model whereby the a-helix is induced by non-specific binding to cell membranes, followed by two-dimensional diffusion leading to specific binding to its receptor. To date, GIP structural analysis 178 has been limited to structure-function relationships using enzyme or chemically cleaved peptides or synthetic peptides and rudimentary computational methods (Figure 60, chapter 4 introduction and described below). The current body of experimental evidence indicates that there are four dissociable domains in GIP,_42. The high affinity binding domain, GIP6.30NH2, is a potent GIP receptor antagonist [413]. Initially, two bioactive domains in GIP were indicated. Truncation of 12 amino acids from the carboxyl terminus of GIP,_42 resulted in a peptide with equivalent insulinotropic activity, but lacking somatostatinotropic activity [32; 337; 410]. Part of the GIP molecule in the carboxyl-terminus is therefore critical for its acid inhibitory (enterogastrone) activity [337; 405; 410], although it is unclear whether this is due to the existence of a second GIP receptor, an alternatively spliced receptor, or differential ligand recognition or coupling of the existing receptor in gastric cells. The insulinotropic domain of GIP was initially hypothesized to be contained between residues 19 to 30, consistent with partial retention of insulinotropic activity of GIP,9.30, GIP,542 and GIP,7.42 [165; 267; 337]. However, this was inconsistent with the importance of the amino-terminus in GIP signal transduction and regulation of GIP activity by DPIV [57-59; 166] (Figure 38). In following this hypothesis, evidence presented in the current studies suggests a third bioactive domain of GIP, residing in residues 1-14 (Figures 31, 35, 37, and Table 7). 179 D P I V Figure 60: Predicted secondary structure of GIP Primary structure of GIP,_42 with predicted a-helical regions (grey; Gascuel and Golmard Basic Statistical Method) and enzymatic cleavage sites used for structure-function relationships. Amino acid positions are indicated above the residues prior to cleavage sites. Abbreviations: DPIV, dipeptidyl peptidase IV, V8, Staphylococcus V8 Protease, CNBr, cyanogen bromide, Trypsin, EK, enterokinase. Computer assisted secondary structure analysis of GIP predicts an alpha helical region between residues 10 and 29 (Figure 60; PCGENE, IntelliGenetics; GGBSM Method [478]). Bioactivity of N-terminal GIP fragments was limited to GIPM 4 (amide or free acid) and amidated forms of GIP,.,3 and GIP,.,,. Hence, it is possible that, along with an intact amino terminus (Tyr'-Ala2), preservation of this helical structure is also important for biological activity. In the absence of Met14 the helix may be unable to form, but the charged residue Asp15 destabilizes it. Amidation of the carboxyl-terminus appears to have minor effects on bioactivity, only noticeable for GIP1.13NH2 and GIPM 5 N H 2, where it may partially stabilize the secondary structure (Table 8). Contrary to data in the current report, GIP,.,4 created by cyanogen bromide cleavage was not insulinotropic in the perfused rat pancreas [165]. However, conversion of Met14 to homoserine lactone by this cleavage method, when this region appears particularly sensitive to structural perturbations, likely generates a biologically inactive N-terminal peptide (or alternatively, the sensitivity of the perfused pancreas may have been too low to detect bioactivity of this fragment). 180 Study of the bioactivity of amino terminal peptide fragments of the secretin-glucagon family is not limited to GIP. Data now exists for parathyroid hormone (PTH), secretin, glucagon and GLP-1 and vasoactive intestinal polypeptide (VIP), and PACAP; however, bioactivity of these fragments appears to be largely dependent on the peptides examined. In a recently published report, micromolar concentrations of PTH,.,4 were able to stimulate cyclic AMP production in cells transfected with the human or rat PTH-1 receptors [479]. While it was not possible to demonstrate 125I-PTH,.34 displacement by PTH,.,4 using radioligand binding assays [479], photoaffinity cross-linking experiments have indicated that the amino terminal region of PTH interacts with the PTH receptor in the area of the junction of the extracellular tail and the first transmembrane domain [480-482]. Further work of Gardella and colleagues found that secretin,. ,3 activated the transfected secretin receptor [479]. In the course of our experiments, we have also tested glucagon,.,40H and GLP-1 [ 7. 2 0 O H | on CHO-K1 cells transfected with human isoforms of their respective receptors, but neither peptide showed any stimulation of cAMP production at concentrations as high as 20 pM (data not shown). GLP-117.20J was previously tested in the perfused rat and canine pancreas, and had little or no effect [483; 484]. Carboxyl-terminal truncation of glucagon dramatically reduced bioactivity [485], and the shortest glucagon fragment reported to retain receptor binding and activation has been glucagon,.,7 [486]. PACAP,_21 was found to be a full agonist of its receptor [477]. Synthetic VIP,.,4 has also been reported to be inactive [487], as was GIP,.,8 [406]. Thus, it appears that structural similarities of amino-termini across all members of the secretin/vasoactive intestinal polypeptide/glucagon superfamily of hormones do not necessarily confer biological activity. Sequential alanine substitution of the GIP,.,4 amino acid sequence confirmed the sensitivity of this peptide to structural perturbations (Figures 30 and 32). Only replacement of amino acids 2 (Ala2Ser) and 13 (Ala13Tyr) with those found in glucagon failed to produce dramatic reductions in receptor binding and activation; this said, given the explicit selectivity of glucagon receptors 181 for glucagon, and GIP receptors for GIP, the results indicate that either residues 2 and 13 are not important ligand-receptor contacts or the domain responsible for receptor selectivity resides in the C-termini of the peptide hormones. The latter postulate is unlikely, as GIP,.^ was not able to activate transfected glucagon or GLP-1 receptors (data not shown). There is some evidence allowing separation of residues important for receptor binding and activation. [Tyr13]GIP|_14 potently bound and activated the GIP receptor, however, [Ser2]GIP,.|4 displayed comparatively lower displacement of l25I-GIP, with similar receptor activation. Similarly, the only two other peptides from the alanine scanning experiment to show significant cAMP stimulation were those substituted in position 1 and 3, however, these peptides also showed little radioligand displacement, compared to peptides which were unable to activate the receptor (Figures 30 and 32). Alanine scanning of the bioactive PTHM 4 N-terminus identified residues 3, 10, 11, 12, 13 and 14 as being tolerant to substitution (i.e. identity not important for receptor activation) [479]. Furthermore, alanine scanning of the entire primary sequence of GLP-1 revealed positions 1, 4, 6, 7, 9, 22 and 23 as being particularly important either for maintenance of secondary structure or ligand-receptor interactions [488; 489]. Comparison of Figure 32 to the data of Luck et al [479], clearly indicates that although both GIP,_i4 and PTHM 4 are bioactive, important residues for the two ligands differ, and those of GIP apparently are also dissimilar from key residues of GLP-1. Due to the structural sensitivity of G I P i _ , 4 , in order to conclusively determine important residues, alanine scanning must be performed for the entire sequence of GIP,_42 or at least GIP|.30. Extensive molecular characterization of PTFfM4 by multisite substitution and modification suggested it was possible to produce small peptide fragments with improved potency [490-493]. Only four modified GIPM 4 peptides were generated (Figures 30 and 33). Introduction of a reduced peptide bond between Ala2 and Glu3 resulted in a peptide with improved receptor potency compared to GIPM4, while at the same time this modification conferred DPIV resistance. 182 Hence, systematic screening of modified GIP,.14 peptides will likely produce potent peptide ligands that may be applied to in vivo experimentation. While structural data for GIP are lacking, it is possible to propose a mechanistic model for ligand binding and receptor activation. A large body of evidence has accumulated regarding receptor ligand interactions for small ligands, including small peptides [494], however, the types of analyses performed on small ligands, such as thyrotropin (TRH; pyroGlu-His-ProNH2) are not possible for larger polypeptide ligands due to their complexity. Use of chimeric ligands and receptors has made it somewhat possible to dissect the molecular domains of peptide hormones necessary for binding and activation. Based on some of this work, Hjorth and Schwartz proposed production of a pseudo-tethered intermediate involving the large extracellular amino-terminus of the cognate receptors for polypeptide ligands, prior to a conformational change drawing the peptide toward the transmembrane helices [495]. Gelling et al demonstrated the importance of the N-terminus of the GIP receptor for ligand binding and activation [280]. Given that GIP^i4 and GIP19.30 both demonstrate receptor binding ability (Figure 29 and Table 7), and that the high affinity binding domain of GIP resides within residues 6 to 30 [413], it is likely that multiple contact residues contribute to high affinity receptor binding. Furthermore, the body of evidence demonstrating the importance of the two N-terminal residues of GIP [57-59; 166] combined with the bioactivity of G I P i 9 . 3 0 suggest an interaction or close proximity of the amino terminus of GIP and its core region (possibly indicating the presence of a functional hinge in the alpha helices), resulting in receptor activation. Until the solution structure for GIP is known and/or the contact residues are established by photoaffinity cross-linking, it is not possible to extend this mechanistic hypothesis without extensive testing of further GIP analogues. 183 4.3.2 DPIV-resistant Incretin Analogues Recognition of the importance of the structure of the N-terminus for biological activity of peptides in the secretin-glucagon superfamily has resulted in the development of numerous analogues with reduced in vivo catabolism and increased biological activity. Substitution of L-Tyr1 with the D-isomer in growth hormone releasing hormone (GRH) [496-498], L-His1 with D-His' in glucagon [499] and GLP-1 [458], or D-amino acids in P2 of glucagon (see below) or GLP-1 [500-502] resulted in peptides with increased in vivo potency. Although conformational changes in the molecules may play a role in increasing biological activity [499], a more prolonged biological half-life as a result of their resistance to enzymatic degradation is probably the more important factor. Frohman et al [496; 503] first showed the importance of DPIV in the physiological degradation of members of the secretin-glucagon superfamily, by demonstrating that GRH is metabolized to biologically inactive GRH3.44 by DPIV both in vitro and in vivo. Additionally, it was shown that amino-terminal substituted analogues, including des-amino-Tyr'-, D-Tyr'-, and D-Ala2-GRH, were resistant to DPIV cleavage [496]. More recently, studies on GLP-1 have shown a similar resistance to DPIV degradation with analogues containing D-amino acids in the P2 N-terminal position [500-502]. The current study was targeted at developing long-acting analogues of the second important incretin, GIP. Flatt and colleagues have reported that [Tyr'-Glucitol]GIP, a peptide analogue modified post-synthesis, displays both DPIV-resistance and improvement of glucose tolerance in a diabetic mouse model [181; 239; 462; 463]. Removal of the first two amino-terminal residues of GIP (GIP3.42) resulted in a peptide which displayed reduced receptor affinity in competition binding studies and that was devoid of the ability to stimulate cAMP production in wtGIPR cells at concentrations as high as 10 uM (Figure 38). This supports the early claim that GIP3.42 isolated from porcine intestinal extracts lacked insulinotropic activity in the perfused rat pancreas [57]. GIP3.42 was an antagonist of GIPi-42-induced cyclic AMP production in the uM range, inhibiting cAMP production by over 90% 184 (Figure 38). Schmidt et al tested GIP3.42 for antagonism of GIP1-42 action on isolated rat islets and found that it was unable to reduce GIP-stimulated insulin release when administered in equal or 10-fold greater concentrations [166]; Figure 38 indicates at least a 1000-fold greater concentration of GIP3-42 is necessary to reduce native GIP action on the cloned receptor. The claim that GIP3-42 lacks insulinotropic activity is further supported by lack of effect on glucose excursions and insulin profde when bioassayed in conscious Wistar rats (Figure 45). In this bioassay, excess exogenous GIP3-42 injected subcutaneously, was unable to block or even shift the insulinotropic effect of endogenously released GIP 1.42 resulting from the oral glucose load; thus it is extremely unlikely that GIP3.42 plays an antagonistic role in vivo. Given the sensitivity of the N-terminus of GIP to inactivation by DPIV we sought to generate peptide analogues resistant to this enzyme for use in vivo. Initial in vitro studies looked at modification or substitution of amino acids in positions 2 and 3 of GIP, using peptides based on the shorter 30 amino acid bioactive core of the hormone (1-30). In general, the amino-terminus of GIP was fairly tolerant of amino acid substitution or modification, when examining binding affinity only, however, when examining bioactivity at the GIP receptor, modification of the N-terminus can have dramatic effects (Figure 40). The data presented here compares well to the earlier limited structure-function analysis of the GIP N-terminus, testing 1-30NH2 analogues [desamino-Tyr'(phenylproprionic acid)]GIP, [D-Tyr']GIP, [D-Ala2]GIP, [D-Glu3]GIP and [D-Ala4]GIP [324] - modifications caused little (less than 10-fold reduced binding for most peptides) or no change in ligand affinity, but specific analogues showed significantly reduced bioactivity. Thus, [D-Ala2]GIP, [Y'A2i|)(CH2NH)]GIP, [Gly2]GIP, and [Ser^ GIP were well tolerated with respect to binding affinity, and [Val2]GIP, [Pro3]GIP and [N-MeGlu3]GIP were only modestly reduced. However when looking at cyclic AMP production, [Y'A>(CH2NH)]GIP, [Pro3]GIP and [N-MeGlu3]GIP were not even able to stimulate maximal levels such that EC50 values could not even be calculated; other analogues showed minor or 185 moderate alterations in receptor activation. [(P)Ser]GIP showed both dramatically reduced binding affinity and cAMP production (Table 8, Figure 40). Cyclic analogues of small peptide hormones generated by lactam bridge formation between endogenous or substituted Lys and Asp or Glu residues have been reported in the literature [504-507]. The current study, to our knowledge, is the first report of a cyclic GIP analogue; this peptide has allowed testing of the effects of further structural constraints on the molecule. By creating a lactam bridge between Lys16 and Asp21, a derivative of GIP was generated with a six amino acid ring, spanning the predicted hinge-domain within the alpha-helical binding core. Remarkably, this peptide was able to stimulate maximal cyclic AMP production, with an increase in EC50 of one order of magnitude relative to native hormone (Figure 40, Table 8). In contrast, the binding affinity of cyclo[Lys16,Asp2l]GIP was disproportionately reduced, resulting in an IC50 value for cyclized GIP that was 25.3-times greater than the unmodified peptide. Results from cyclo[Lys16,Asp2l]GIP have shown the feasibility of using cyclic incretin analogues, as well as confirming hypotheses generated from previous structure-activity relationship studies of GIP. It was shown earlier that the high affinity binding domain resided in the predicted core a-helical domain between residues 6 and 30 of GIP [413]; however, two dissociable bioactive domains were discovered: G I P i _ 1 4 and GIP19.30, both of which also possessed weak binding ability (Figures 29 and 31). Solution structure data for related hormones has shown the presence of a common 16-20 amino acid helical core, either as a continuous helix or with a two amino acid hinge region. Hence, the lactam bridge in cyclo[Lysl6,Asp21]GIP may interfere with receptor binding (Figure 40, Table 8) either by disrupting helix formation or interfering with the flexibility of the hinge domain, without causing proportional reductions in bioactivity. From these data, substitution with D-Ala in position 2 was shown to have the greatest potential for further development; preliminary trials found that this molecule was completely 186 resistant to DPIV degradation for over 24 hours, and had minimal changes in receptor activity (Figure 40, Table 8). Unfortunately, all other peptides exhibiting complete DPIV resistance displayed dramatically reduced cyclic AMP stimulating ability at the GIP receptor (Figure 40, Table 8). When testing full length [D-Ala2]GIP,.42, these results were corroborated and studies continued in animal models. While DPIV-resistance did little to the effectiveness of the analogue in vitro (consistent with the negligible DPIV activity in CHO and BTC-3 cells; Figure 39), when tested in vivo, [D-Ala2]GIP reduced glycemic excursions in all animal models to a greater extent than native GIP. This was associated with enhanced early phase insulin release in lean animals (Figures 44 and 46, Table 9), and in diabetic rats, where the first phase of insulin release is compromised, an augmentation of the entire insulin time-course was observed (Figure 47). The latter finding is of particular interest, as GIP's effectiveness in type 2 diabetes mellitus and animal models of the disease has been questioned, and remains controversial. Lean Zucker rats showed significant differences in glycemic profiles between GIP and [D-Ala2]GIP, while both peptides appeared equally insulinotropic except for the first time-point (t = 3.5 min) (Figure 46). These data are consistent with either an increased insulin sensitivity in these animals, as noted by Pederson et al [381], and/or enhanced ability of this compound to stimulate glucose. uptake in peripheral tissues [239; 251]. Although GIP,.42 has been shown to exhibit equivalent insulinotropic activity to GLP-1 [336] there have been no previous reports targeted at developing long acting analogues with therapeutic potential in diabetes until very recently, and these have been limited. The reason for the lack of such studies probably originates in the report of Nauck et al [220], in which they observed that human GIP was almost devoid of insulinotropic activity in diabetic patients. It had been shown earlier that there was a reduced incretin response in type 2 diabetes, characterized by a reduction in the component of (3-cell secretion resulting from oral glucose relative to that obtained with intravenous glucose [377]. Additionally, the responsiveness of insulin resistant 187 diabetic patients to exogenous porcine GIP, at concentrations resulting in physiological [376], or near physiological [378] circulating levels, was blunted. It was also observed that porcine GIP,. 42 stimulated insulin secretion under fasting conditions [508; 509] whereas normal controls did not respond, presumably due to the fact that circulating fasting glucose levels in type 2 diabetes mellitus patients reach the required threshold for the insulinotropic activity of GIP. Therefore, there appears to be little doubt that insulin responses to exogenous GIP are reduced in diabetic patients. Nevertheless, the pancreas still retains some GIP sensitivity. The reason for the almost complete lack of response to human GIP observed in the study of Nauck et al [220] may lie elsewhere. In their study, responses of normal controls to the human peptide were also extremely weak, unlike those described by the same group in an earlier study with GIP from a different commercial source [221], suggesting that the synthetic human peptide used in the diabetic study exhibited only weak biological activity. It has recently been shown that some [336], but not all [276], commercial preparations of human GIP exhibit very low biological activity, and it is therefore critical that further clinical trials with human GIP and GIP analogues, are performed with peptide of established biological activity. It is also important to define the origin of the resistance to GIP. One possibility is increased receptor desensitization/down-regulation or altered signal-transduction pathways (see Chapter 3). Alternative explanations include antag onism of GIP,.42 action by GIP3_42, as suggested for GLP-1 [448], or a genetic defect resulting in reduced receptor expression [374]. Antagonism by GIP3_42 appears unlikely in view of its low binding affinity for the receptor and strength as an antagonist, with at least 1000-fold higher concentrations being required to show a significant effect. However, some support for the latter hypothesis has been obtained recently, with the finding that obese VDF Zucker rats have compromised GIP receptor expression, both at the mRNA and protein level in isolated islets of Langerhans [379]. During an IP glucose tolerance test, when GIP was infused at minimum threshold levels necessary to obtain a biological response in lean 188 animals (4 pmolmin'Kg\"1), obese animals were unable to respond to the same dose. This was linked to a defect in ability to stimulate cyclic AMP in isolated islets, and quantitative PCR and immunoblotting suggested a reduction in GIP receptor expression [379]. The etiology of the diminished expression, however, has not been elucidated, and may result from elevated postprandial GIP levels and subsequent desensitization/internalization/down-regulation of the receptor. In the current study, fasting GIP levels in lean and obese VDF Zucker rats were not significantly different, confirming the result of Lynn and colleagues [379], however, elevated levels of GIP were detected in obese animals following an OGTT, lending support for the down-regulation hypothesis. In order to directly compare the potency of DPIV-resistant GIP and GLP-1 analogues, several peptides were tested in parallel in vitro and in vivo. For this study, [Ser] substituted peptides were chosen, as this substitution produced no alteration in receptor binding or activation (Figure 40, Table 8), and [Ser2]GLP-l has been reported to show enhanced in vivo bioactivity owing to its DPIV resistance [457-459]. Both Ser2 substituted incretins showed relatively normal receptor binding and activation parameters (Figures 48 and 49), and moderate, but not complete DPIV resistance to purified enzyme. Accordingly, bioassay of these peptides demonstrated that they were slightly more effective than native peptides in vivo (Figure 50, Table 10). Comparison of native peptides (i.e. GIP versus GLP-1) or Ser2 substituted peptides showed very similar potencies, particularly when considering the integrated responses, which were not significantly different (Table 10). The glucagonostatic property of GLP-1 has been an often touted benefit of GLP-1 over GIP as a therapeutic possibility [374; 510]. However, both GIP and GLP-1 are glucagonotropic in vitro [190; 219], but GIP produces no alteration in glucagon release in humans [164; 220]. To date, no experimental study has been performed demonstrating that GLP-1's glucagon lowering effects actually contribute to its glucose lowering effects independently of its insulinotropic action in normal humans or patients with type 2 189 diabetes mellitus (although in insulinopenic type 1 diabetes, it was proposed that inhibition of glucagon secretion contributed to the antihyperglycemic effect of exogenous GLP-1 in these individuals [466]). In fact, counter-intuitively, transgenic overexpression of the glucagon receptor in mice actually improved glucose tolerance (Dr. R. Gelling, personal communication) and glucagon confers (3-cell glucose competence [283]. Furthermore, GLP-1 is a well characterized inhibitor of gastric emptying, whereas the effect of GIP on gastric emptying is not well characterized [511]. Thus, despite the potential differences in their modes of action, their antidiabetic action is quite similar, consistent with their equal insulin-releasing ability [336]. Phospho-Serine2 modification was found to render glucagon completely resistant to purified DPIV, but this peptide was time-dependently dephosphorylated in serum. We reasoned that [(P)Ser2] modification of incretins would provide an extra degree of protection from enzymatic degradation during absorption after SC injection, as these peptides were also completely resistant to purified DPIV, and would likely also be dephosphorylated in vivo to the moderately resistant [Ser2] incretin analogues. One potential problem was their dramatically reduced potency at their respective receptors (Figures 48 and 49). However, on subcutaneous injection into rats, with a concurrent glucose challenge, these peptides produced equivalent or more potent antidiabetic effects than their non-phosphorylated counterparts, at the same dosage (Figure 51, Table 10). It was noted, however, that phosphoserine2 substitution ablated incretin effects on the early phase of insulin release, but this could be overcome by administration of a greater dosage (Figure 51, Table 10). Hence, [(P)Ser2] incretin analogues represent the first generation of \"pro-drug\" forms of long acting metabolically stable peptides. 190 4.3.3 DPIV Degradation of Glucagon and DPIV-resistant Analogues Several lines of evidence have resulted in the necessity for re-assessment of glucagon degradation in vivo. Controversy in the past regarding glucagon degradation, with respect to specific enzymes and organs involved needs to be clarified. Evidence presented indicate that dipeptidyl peptidase IV is a prime candidate for enzymatic inactivation of glucagon [328; 453]. The recent finding demonstrating DPIV in the secretory granules of the pancreatic islet a-cell compels one to question how much of the pancreatic glucagon enters the circulation intact [512]. Grondin and colleagues further argue that the low pH of the secretory granule would not permit activity of DPIV, and thus DPIV would not be active until granule contents are secreted. The discovery that glucagon is successively hydrolyzed by dipeptidyl peptidase IV into N-terminally truncated peptides raises a number of questions. The first question is the role of the hydrolyzed peptides - are they simply degradation products, or do they have a physiological role? The emergence of the \"mini-glucagon\" story in the pancreas suggests the hypothesis of a local action of N-terminally truncated glucagon. Processing of glucagon by MGE (miniglucagon-generating endopeptidase) to glucagon,9.29, results in a peptide having differential effects on cardiac myocytes [513], and with the ability to inhibit insulin release in the picomolar range [514]. This report forms the foundation for further work on glucagon degradation products and their possible function in vivo. Preliminary studies using an in vivo bioassay suggested that DPIV inactivates glucagon. Incubation of glucagon with purified DPIV abolished its hyperglycemic effects, but not if a specific DPIV inhibitor was included (Figure 52). Similarly, injection of large doses of cleavage products, that were identified by mass spectrometry, demonstrated no biological activity (Figure 53). It is likely that glucagon degradation fragments are present in vivo, and thus measurements of glucagon immunoreactivity by \"side-viewing\" or C-terminally directed antibodies [99] would likely cross-react with these peptides. Studies using antibodies for the measurement of glucagon 191 may therefore be misleading and this is likely the reason that glucagon was previously reported to be stable in plasma [437; 454]. Frandsen et al [515] found that glucagon5_29 cross-reacted with at least two antisera tested , and it is likely that antibodies also cross react with glucagon3_29 and [pGlu3]glucagon3.29. Thus, as for GIP and GLP-1 [60; 448], measurements of glucagon in serum overestimate biologically active peptide. Fragments similar to those described here have been tested for agonism and antagonism in other biological systems. Glucagon,_29 was found to have <0.001% of the potency of native glucagon in the rat hepatocyte membrane adenylyl cyclase activity assay [515], and it was found that this fragment also acted as an antagonist in this tissue. In the current study, using cells over-expressing the human glucagon receptor, it was found that glucagon5.29 has 28.5% of the potency of glucagon,_29 (Figure 55, Table 11), and indeed, it does act as a weak antagonist on these cells (Figure 56). Similar glucagon analogues to those tested here, [Glu9]glucagon3.29 and [Glu9]glucag on5_29, have also been previously characterized [460]. The binding affinities reported for the Glu9 substituted analogues [460] are consistent with the trend observed with native fragments on transfected cells (Figure 57), however, the amino acid substitution at position 9 alone ([Glu9]glucagon,_29) was also shown to have dramatic effects on binding affinity [460]. Similarly, [Glu9]gl ucagon,_29 had significantly reduced potency, and N-terminally truncated (desHis1) peptides showed negligible adenylyl cyclase stimulating activity [460]. In light of the finding that only glucagon5.29 showed antagonism on hGlucR cells, it is likely that the Glu9 substitution was responsible for the antagonism observed for [Glu9]glucagon3.29, and resulted in antagonism potencies similar to [Glu9]glucagon2.29, as was also the case for [Glu9]glucagon5.29 [460]. Native glucagon5.29 showed only weak antagonism compared to [Glu9]glucagon2.29 (Figure 56). The activities of the fragments tested support the importance of the amino terminus in glucagon signal transduction. Cyclization of the side chain of Gin3 to form [pGlu3]glucagon3.29 both increased binding affinity and potency as compared to glucagon3.29 192 (Figures 55 and 57). Glucagon5.29 also retained high affinity binding (greater than glucagon3.29), but showed lower potency when compared to either [pGlu3]glucagon3.29 or glucagon3.29. Characterization of DPIV-resistant, N-terminally modified glucagon analogues is consistent with published literature. The general conclusion from random molecular mutagenesis screening was that modification of the amino terminus of glucagon reduces biological activity, implicating it as an important domain necessary for receptor activation [516]. Robberecht and colleagues found that altering the chirality at positions 2 and 3 of glucagon had minor effects on potency in the hepatocyte adenylyl cyclase assay; [D-Ser]glucagon was equivalent to native glucagon in terms of cAMP formation and binding affinity, whereas reversing the chirality of position 3 had significant effects on both parameters [517]. Unson and Merrifield [518] also substituted the D-isomer of serine in position 2, however, they found it dramatically reduced affinity and potency of this analogue in vitro. Our work using cells transfected with the human glucagon receptor are consistent with the earlier studies using hepatocyte membranes [517]. The [D-Ser2] substitution was better tolerated than [D-Gln3], when looking at in vitro cAMP stimulatory activity and receptor binding affinity (Figures 58, Table 11). Previous studies on [D-Ser2] and [D-Gln3] were limited to in vitro structure-function studies. With the objective of generating DPIV resistant glucagon analogues, to support the hypothesis of DPIV degradation of glucagon, an in vivo assay system was necessary. The [D-Ser2] substitution was the only analogue which possessed enhanced ability to increase circulating glucose levels relative to native glucagon. The greater potency in vivo can be attributed to the lack of degradation by DPIV, as the in vitro potency was found to be moderately reduced (Figure 58). However, this substitution rendered the peptide more susceptible to degradation by trypsin-like enzymes [328]. Other N-terminally modified glucagon analogues were not suitable for demonstrating the contribution of DPIV to the degradation of glucagon, as they possessed reduced biological activity in vitro, and in vivo, or were susceptible to DPIV degradation. 193 4.3.4 Conclusion The current work has verified the importance of the amino-terminus of GIP for bioactivity. GIP,.14 is a unique GIP fragment that displays specific GIP receptor binding, and cyclic AMP production in transfected cells, in addition to insulinotropic activity in the perfused pancreas and improvement of glucose tolerance in vivo. Consistent with prior studies in the perfused pancreas, synthetic GIP,9.3() was able to stimulate cAMP production weakly and had intrinsic receptor binding ability. Hence, it appears that the two insulinotropic domains exhibit secondary structure within GIP,.,4 and GIP,9.30, while a putative enterogastrone/somatostatinotropic domain lies in the C-terminus of the molecule. The significance of GIP as a physiological incretin has been emphasized in studies using specific GIP antagonists in rat [186] and inhibition of DPIV activity in GLP-1 receptor knockout mice [519]. With the important findings that sulfonylureas improve 6-cell sensitivity to GIP [464], that smaller fragments of GIP are bioactive, and in the present report, that even in animals with compromised GIP receptor expression, bolus injections of GIP and analogues with improved plasma stability are still capable of improving glucose tolerance, the pharmacological potential of GIP in treatment of human diabetic states is preserved. In order to truly recognize the potential of GIP and degradation resistant analogues, the extent of the reduced sensitivity to GIP in diabetic patients needs to be quantitatively assessed using a wide range of peptide concentrations, as has been done for GLP-1 [510], rather than single low dosage protocols, biased to show a lack of effectiveness. No matter what underlying cause is ultimately determined to be responsible for the reduced responsiveness to GIP in human diabetes, analogues based on the DPIV-resistant forms of GIP and GLP-1 described here may be useful in stimulating insulin secretion through the residual islet capacity to respond to incretins. Several structure-activity relationships of glucagon have been assessed in vitro and in vivo, with specific reference to degradation of glucagon by dipeptidyl peptidase IV. N-terminally 194 truncated glucagon fragments were all weak partial agonists of the human glucagon receptor, and showed no glycemic effect in vivo. The role of DPIV degradation in glucagon metabolism was also studied using amino-terminally modified glucagon analogues. DPIV-resistant [D-Ser2] exhibited enhanced biological activity relative to native glucagon in a bioassay. Thus further evidence that DPIV is likely a primary enzyme involved in glucagon degradation has been provided. 195 Chapter 5: Summary and Future Directions Results have been presented regarding the potential regulatory mechanisms modulating insulinotropic peptide hormone bioactivities, namely receptor desensitization and enzymatic inactivation, with a focus on glucose-dependent insulinotropic/gastric inhibitory polypeptide. Studies on insulinotropic hormones glucagon and GLP-1 were limited to design of synthetic analogues. On the one hand, GIP receptor desensitization, sequestration, degradation and down-regulation were considered as a potential mechanism for the attenuated responsiveness of the 6-cell to GIP in type 2 diabetes and animal models of the disease [290; 370; 374; 379]. Results indicated that the GIP receptor was fairly resistant to desensitization compared to other receptor models, including the GLP-1 and glucagon receptors (Chapter 3). Thus it is likely that if receptor desensitization/down-regulation is the culprit for the ablated GIP effect in diabetes, it is likely a chronic effect involving degradation of receptors and/or transcriptional down-regulation of GIP receptor expression. Furthermore, the relative resistance of the receptor to rapid desensitization is a favourable characteristic, if GIP receptor agonists are to be used as potential therapies. Enzymatic inactivation of GIP, GLP-1 and glucagon by dipeptidyl peptidase IV has been reported [328; 427; 453]. The development of specific inhibitors of DPIV and enzyme resistant incretin analogues have been well documented as potential therapies for type 2 diabetes mellitus [456; 461; 511; 520; 521]. Presented here was the first report of systematic design of DPIV-resistant GIP analogues [329; 456], 3 of which were selected for in vivo studies (Chapter 4). Although responsiveness to GIP in diabetes is reduced, this does not necessarily preclude the pharmaceutical potential of analogues alone or in combination with traditional oral therapy. Sulphonylureas improve 6-cell responsiveness to GIP [464], lending support for combination therapy, while simply raising the dosage will likely overcome the diminished effect of GIP alone. The finding that DPIV also degrades the counter-regulatory hormone, glucagon [328; 453], 196 raises some interesting issues. While DPIV-resistant analogues were employed in the current report in order to support the hypothesis that DPIV-hydrolysis of glucagon may be a normal physiological process, therapeutic potential of hyperglycemic agents is limited. Glucagon is currently used clinically in cases presenting hypoglycemia (including but not limited to insulin or sulfonylurea overdose), during refractory bradycardia (during cardiogenic shock or (3-blocker overdose) and as a GI relaxant during endoscopy, among other uses in emergency medicine [522; 523], and superactive analogues of glucagon such as [D-Ser2]glucagon].29 described here may be used at lower doses to achieve the same results. Findings suggest that DPIV inhibition will likely also preserve bioactive glucagon; the net effect of postprandial DPIV inhibition is to reduce glycemic excursions [381; 382; 418], however, the effect of DPIV during fasting has not been examined, nor has the relative contribution of glucagon to the postprandial glycemic profile during DPIV inhibition. In point form, the salient results from the current thesis are: 1. Desensitization of (5TC-3 cells to GIP occurs at the level of the receptor, and at distal steps of the stimulus-secretion cascade. 2. Desensitization of the GIP receptor parallels receptor sequestration kinetics in transfected cells. 3. Receptor phosphorylation was implicated in both receptor desensitization and internalization, however, the specific sites appeared to be different, but may overlap. 4. Protein kinase C was implicated in the regulation of GIP responsiveness in |3TC-3 cells and of the transfected receptor. 5. Fluorescent methods were developed to track GIP receptor internalization, and may be further used to delineate endocytic and recycling pathways. 197 6. Inactivation of GIP by dipeptidyl peptidase IV was confirmed: GIP3.42 bound to the GIP receptor with high affinity, but was unable to promote cyclic AMP production at concentrations as high as 10 pM. 7. A series of DPIV-resistant GIP analogues were generated, which serve to further our understanding of the importance of GIP's amino terminus for receptor activation, as well as permitting demonstration of enhanced bioactivity of enzyme resistant peptides in vivo. [D-Ala2]GIP,. .42 was antidiabetogenic in normal and obese diabetic rats. 8. Parallel comparison of GIP and GLP-1 analogues in vitro and in vivo indicated that the potency of these peptides were not dramatically different, lending support for the use of both peptides as therapeutic agents. 9. Inactivation of glucagon by DPIV was shown. Fragments arising from DPIV cleavage were partial agonists in vitro, but ineffective in vivo. DPIV-resistant analogues were generated, and [D-Ser2]glucagon showed similar in vitro potency to native GIP, but enhanced in vivo biological activity. From an academic standpoint, there is still much we don't know about GIP receptor physiology. Its regulation has only begun to be examined. Phosphorylation was implicated by the current work as a potential regulatory mechanism. A more detailed examination of individual phosphorylation sites would further clarify the importance of each site in either desensitization or internalization; alanine substitution was employed here, but substitution of negatively charged residues may mimic phosphorylation, and provide some insight. G-protein coupled receptor kinases, in particular GRK-2, have been implicated in the phosphorylation of the GIP receptor, although specific phosphorylation sites were not identified. Perhaps more interesting is the potential cross-talk between receptors, possibly mediating heterologous desensitization, a subject which has not been studied for the GIP receptor. The diversity of GIP 198 receptor signalling has been recently addressed [293; 294; 296; 300; 302], however, full characterization of these pathways has yet to be done. The role of receptor sequestration and/or interaction with cytoskeletal scaffold proteins in the activation of pleiotropic signal transduction cascades should be performed; MAP kinases may feedback on signalling molecules GRK-2 and/or 6-arrestin to alter desensitization [524-535]. It is not clear whether the recent demonstration of G-protein switching for the adrenergic receptor [536; 537] and interacting molecules capable of altering the selectivity of the CGRP receptor [538; 539] may be applied to other G-protein coupled receptors such as those for incretin hormones. The role of GIP receptor down-regulation in the etiology of type 2 diabetes needs to be further characterized. Supportive evidence has been found from in vivo studies in animal models of type 2 diabetes [379]. In order to fully understand the reasons for greater responsiveness to GLP-1, but diminished responsiveness to GIP may be accomplished by gene expression array technology comparing donor tissue from normal and diabetic islets. Furthermore, the underlying mechanisms for improved [3-cell sensitivity in sulphonylurea treated diabetics may similarly be assessed [464]. Study of GIP receptor expression will likely yield agents capable of increasing GIP receptor expression, and thus reverse the ineffectiveness of the hormone in diabetes. In the meantime, a full pharmacokinetic study should be undertaken to characterize fully the dose-response relationship of GIP in healthy and diabetic humans. This will provide useful information regarding the feasibility of GIP and GIP analogues for the potential treatment of diabetes. In the current thesis, comparison of a completely resistant GIP analogue ([D-Ala2]GIP) to a partially resistant analogues ([Ser2]GIP) clearly demonstrated the benefit of greater DPIV resistance for enhanced in vivo potency (Chapter 4). Unfortunately, most of the completely DPIV-resistant analogues generated in the current study suffered compromised activity at the cloned receptor. Thus effort to obtain completely resistant GIP analogues with complete DPIV resistance would be worthwhile. Clinical testing of [D-Ala2]GIP in healthy individuals and type 199 2 diabetes seems clear to proceed; recently, a preliminary report of use of DPIV-resistant [Tyr1-glucitoljGIP in humans has been presented in abstract form [540]. With the finding that dipeptidyl peptidase IV is also capable of degrading glucagon [453; 456], there are many potential future directions. Perhaps the first and most important direction would be the development of analytical techniques to simultaneously measure glucagon and truncated fragments in serum. Use of immunoprecipitation combined with mass spectroscopy may be the most rapid method at this point [61], however, development of N-terminally specific antibodies would be equally effective, but perhaps more difficult [60; 417; 433; 448]. Thus it would be possible to examine the effects of DPIV inhibition on normal glucagon metabolism. Combination of glucagon antagonists and DPIV inhibitors may allow quantification of the relative contribution of glucagon to the postprandial glycemic profile in animals thus treated. The role that peptide hormones play in the regulation of glucose homeostasis continues to be better understood with well designed experimental evidence. However, with each question that is answered, many more questions arise. It is hoped that this thesis has both answered many questions regarding GIP receptor physiology and regulation of insulinotropic hormones by enyzmatic inactivation, as well as provided insight for future studies in the field. GIP was the only known physiological incretin during the 1970s [3], but upon the discovery of GLP-1 [100], GLP-1 has been the focus of intense study by many groups for two decades, while GIP has been relatively neglected [511]. It seems now that we are in the beginning of a renaissance of interest in GIP in normal physiology, disease and as a therapeutic possibility. 200 Appendix A Table 12: Predicted and measured molecular masses of synthetic peptides Mass spectrometry was performed by Dr. S. Manhart, Probiodrug, Halle (Saale), Germany. Peptides were purified by HPLC to greater than 92-95%. Synthetic Molecular Weight Synthetic Molecular Weight Peptide: (Daltons) Peptide: (Daltons) Expected Measured Expected Measured GIP(1-420H) 4984.3 4984.7 [Ala 1 4 ] l -140H 1507.7 1506.0 1-6NH2 685.7 686.9 [D-Ala 2 ] l -140H 1567.8 1570.6 1-7NH2 798.9 800.2 [Pro 3 ] l-140H 1535.8 1536.0 1-130H 1436.5 1438.2 [ Y ' A \\ p ( C H 2 N H ) ] l - 1 4 0 H 1552.9 1554.2 1-13NH2 1435.6 1435.6 [BTD]1-140H 1766.8 1769.5 1-140H 1567.8 1569.3 1-14NH2 1566.8 1569.7 [D-Ala 2 ] l -420H 5002.7 4998.6 1-150H 1682.8 1680.3 [D-Ala 2 ] l -30NH2 3552.0 3553.8 1-15NH2 1681.9 1682.6 [Y'A 2a)>(CH 2NH)]l-30NH2 3537.1 3539.0 1-30NH2 3552.0 3553.3 [Gly 2 ] l -30NH2 3537.0 3539.1 7-30NH2 2882.3 2886.9 [Ser 2]l-30NH2 3567.1 3568.0 15-420H 3433.8 3434.4 [(P)Ser]l-30NH2 3647.1 3646.9 15-30NH2 2001.3 2003.3 [Val 2 ] l -30NH2 3579.1 3580.7 16-30NH2 1886.3 1887.6 [Pro 3]l-30NH2 3519.1 3522.9 17-30NH2 1758.1 1761.1 [N-MeGlu 3 ] l -30NH2 3565.1 3566.1 19-30NH2 1488.7 1489.8 [Cyclo(K\"\\D 2 ' ) ] l -30NH2 3533.1 3524.6 3-420H 47494 4751.4 [Ala ' ] l -140H 1475.7 1475.2 GLP-1 |7 .36NH2| 3297.7 3298.4 [ S e r 2 ] ! - K O H [Ala 3 ] l -140H 1583.8 1583.2 [ S e r 2 ] G L P - l | M 6 N H 2 , 3313.9 3314.3 1509.7 1511.4 [ ( P ) S e r 2 ] G L P - l l „ 6 N H 2 | 3393.7 3394.3 [Ala 4]l-140H 1581.8 1586.8 [Ala 5]l-140H 1537.7 1533.2 Glucagon,. 2 9 3482.8 3482.9 [Ala 6 ] l -140H 1491.7 1487.7 Glucagon 3. 2 9 3258.6 3256.7 [Ala 7 ] l -140H 1525.7 1526.1 [pGlu 3]glucagon 3. 2 9 3241.6 3242.2 [Ala 8 ] ! -140H 1551.8 1546.6 Glucagon5_2 9 [D-Ser]glucagon,_2 9 [(P)Ser2]glucagon,.29 3073.4 3074.9 [Ala 9 ] l -140H 1523.7 1523.3 3482.8 3485.2 [Ala '°] l -140H 1475.7 1477.8 3562.8 3564.1 [A la\" ] l -140H 1551.8 1553.6 [Gly2]glucagon,_2 9 [D-Gln3]glucagon,.29 3451.8 3452.0 [ A l a l 2 ] l - 1 4 0 H 1525.7 1525.4 3482.8 3483.9 [Tyr 1 3 ] l -140H 1659.9 1648.2 201 References [I] Sherwood, N. M., Krueckl, S. L., and McRory, J. E. (2000). The origin and function of the pituitary adenylate cyclase-activating polypeptide. Endocr Rev, 21:619-670. [2] Campbell, R. M., and Scanes, C. G. (1992). Evolution of the growth hormone-releasing factor (GRF) family of peptides. Growth Regulation, 2:175-191. [3] Pederson, R. A. (1994). Gastric Inhibitory Polypeptide, in J. H. Walsh, and G. J. Dockray, eds., Gut Peptides: Biochemistry and Physiology: New York, Raven Press Ltd., p. 217-259. [4] Mcintosh, C. H. S., Wheeler, M. B., Gelling, R. W., Brown, J. C, and Pederson, R. A. (1996). GIP receptors and signal-transduction mechanisms. Acta Physiol Scand, 157:361-365. [5] Lefkowitz, R. J. (2000). The superfamily of heptahelical receptors. Nat Cell Biol, 2:E133-E116. [6] Dixon, R. A., Kobilka, B. K., Strader, D. J., Benovic, J. L., Dohlman, H. G., Frielle, T., Bolanowski, M. A., Bennett, C. D., Rands, E., and Diehl, R. E. (1986). Cloning of the gene and cDNA for mammalian beta-adrenergic receptor and homology with rhodopsin. Nature, 321:75-9. [7] Ishihara, T., Nakamura, S., Kaziro, Y., Takahashi, T., Takahashi, K., and Nagata, S. (1991). Molecular cloning and expression of a cDNA encoding the secretin receptor. EMBO J, 10:1635-1641. [8] Thorens, B. (1992). Expression cloning of the pancreatic beta cell receptor for the gluco-incretin hormone glucagon-like peptide 1. Proc Natl Acad Sci USA, 89:8641-8645. [9] Svoboda, M., Ciccarelli, E., Tastenoy, M., Robberecht, P., and Christophe, J. (1993). A cDNA construct allowing the expression of rat hepatic glucagon receptors. Biochem Biophys Res Commun, 192:135-142. [10] Jenlinek, L. J., Lok, S., Rosenberg, G. B., Smith, R. A., Grant, F. J., Biggs, S., Bensch, P. A., Kuijper, J. L., Sheppard, P. O., and Sprecher, C. A. (1993). Expression cloning and signalling properties of the rat glucagon receptor. Science, 259:1614-1616. [II] Usdin, T. B., Mezey, E., Button, D. C, Brownstein, M. J., and Bonner, T. I. (1993). Gastric inhibitory polypeptide receptor, a member of the secretin-vasoactive intestinal peptide receptor family, is widely distributed in peripheral organs and the brain. Endocrinology, 133:2861-2870. [12] Brown, J. C, Pederson, R. A., Jorpes, E., and Mutt, V. (1969). Preparation of highly active enterogastrone. Can J Physiol Pharmacol, 47:113-114. [13] Brown, J. C, and Pederson, R. A. (1970). A multiparameter study on the action of preparations containing cholecystokinin-pancreozymin. Scand J Gastroenterol, 5:537-541. 202 [14] Brown, J. C, Mutt, V., and Pederson, R. A. (1970). Further purification of a polypeptide demonstrating enterogastrone activity. J Physiol, 209:57-64. [15] Pederson, R. A., and Brown, J. C. (1972). Inhibition of histamine-, pentagastrin-, and insulin-stimulated canine gastric secretion by pure \"gastric inhibitory polypeptide\". Gastroenterology, 62:393-400. [16] Kosaka, T., and Lim, R. K. S. (1930). Demonstration of the humoral agent in fat inhibition of gastric secretion. Proc Soc Exptl Biol NY, 27:890-891. [17] Brown, J. C. (1971). A gastric inhibitory polypeptide. I. The amino acid composition and the tryptic peptides. Can J Biochem, 49:255-261. [18] Brown, J. C, and Dryburgh, J. R. (1971). A gastric inhibitory polypeptide II: The complete amino acid sequence. Can J Biochem, 49:867-872. [19] Jornvall, H., Carlquist, M., Kwauk, S., Otte, S. C, Mcintosh, C. H. S., Brown, J. C, and Mutt, V. (1981). Amino acid sequence and heterogeneity of gastric inhibitory polypeptide (GIP). FEBS Lett, 123:205-210. [20] Moody, A. I., Thim, L., and Valverde, I. (1984). The isolation and sequencing of human gastric inhibitory peptide (GIP). FEBS Lett, 172:142-148. [21] Carlquist, M., Maletti, M., Jornvall, H., and Mutt, V. (1984). A novel form of gastric inhibitory polypeptide (GIP) isolated from bovine intestine using a radioreceptor assay. Fragmentation with staphylococcal protease results in GIP 1-3 and GIP 4-42, fragmentation with enterokinase in GIP 1-16 and GIP 17-42. Eur J Biochem, 145:573-577. [22] Takeda, J., Seino, Y., Tanaka, K-L, Fukumoto, H., Kayano, T., Takahashi, H., Mitani, T., Kurono, M., Suzuki, T., Tobe, T., and Imura, H. (1987). Sequence of an intestinal cDNA encoding human gastric inhibitory polypeptide precursor. Proc Natl Acad Sci USA, 84:7005-7008. [23] Inagaki, N., Seino, Y., Takeda, J., Yano, H., Yamada, Y., Bell, G. I., Eddy, R. L., Fukushima, Y., Byers, M. G., Shows, T. B., and Imura, H. (1989). Gastric inhibitory polypeptide: structure and chromosomal localization of the human gene. Mol Endocrinol, 3:1014-1021. [24] Higashimoto, Y., Simchock, J., and Liddle, R. A. (1992). Molecular cloning of rat glucose-dependent insulinotropic peptide (GIP). Biochim Biophys Acta, 1132:72-74. [25] Higashimoto, Y., and Liddle, R. A. (1993). Isolation and characterization of the gene encoding rat glucose-dependent insulinotropic polypeptide. Biochem Biophys Res Commun, 193:182-190. [26] Tseng, C-C, Jarboe, L. A., Landau, S. B., Williams, E. K., and Wolfe, M. M. (1993). Glucose-dependent insulinotropic peptide: structure of the precursor and tissue-specific expression in rat. Proc Natl Acad Sci USA, 90:1992-1996. 203 [27] Sharma, S. K., Austin, C, Howard, A., Lo, G., Nicholl, C G., and Legon, S. (1992). Characterization of rat gastric inhibitory peptide cDNA. J Mol Endocrinol, 9:265-272. [28] Schieldrop, P. J., Gelling, R. W., Elliot, R., Hewitt, J., Kieffer, T. J., Mcintosh, C. H. S., and Pederson, R. A. (1996). Isolation of a murine glucose-dependent insulinotropic polypeptide (GIP) cDNA from a tumor cell line (STC6.14) and quantification of glucose-induced increases in GIP mRNA. Biochim Biophys Acta, 1308:111-113. [29] Yeung, C-M., Wong, C. K. C, Chung, S. K., Chung, S. S. M., and Chow, B. K. C. (1999). Glucose-dependent insulinotropic polypeptide gene expression in the stomach: revealed by a trangenic mouse study, in situ hybridization and immunohistochemical staining. Mol Cell Endocrinol, 154:161-170. [30] Lewis, T. B., Saenz, M., O'Connell, P., and Leach, R. J. (1994). Localization of glucose-dependent insulinotropic polypeptide (GIP) to a gene cluster on chromosome 17q. Genomics, 19:589-591. [31] Anderson, L. A., Friedman, L., Osborne-Lawrence, S., Lynch, E., Weissenbach, J., Bowcock, A., and King, M. C. (1993). High-density genetic map of the BRCA1 region of chromosome 17ql2-q21. Genomics, 17:618-623. [32] Rossowski, W. L, Zacharia, S., Mungan, Z., Ozmen, V., Ertan, A., Baylor, L. M., Jiang, N-Y., and Coy, D. H. (1992). Reduced gastric acid inhibitory effect of a pGIP(l-30)NH2 fragment with potent pancreatic amylase inhibitory activity. Regul Pept, 39:9-17. [33] Someya, Y., Inagaki, N., Maekawa, T., Seino, Y., and Ishii, S. (1993). Two 3',5'-cyclic-adenosine monophosphate response elements in the promoter region of the human gastric inhibitory polypeptide gene. FEBS Lett, 317:67-73. [34] Boylan, M. O., Jepeal, L. I., Jarboe, L. A., and Wolfe, M. M. (1997). Cell-specific expression of the glucose-dependent insulinotropic polypeptide gene in a mouse neuroendocrine tumor cell line. J Biol Chem, 272:17438-17443. [35] Tseng, C-C, Boylan, M. O., Jarboe, L. A., Williams, E. K., Sunday, M. E., and Wolfe, M. M. (1995). Glucose-dependent insulinotropic polypeptide (GIP) gene expression in the rat salivary gland. Mol Cell Endocrinol, 115:13-19. [36] Higashimoto, Y, and Liddle, R. A. (1994). Developmental expression of the glucose-dependent insulinotropic polypeptide gene in rat intestine. Biochem Biophys Res Commun, 201:964-972. [37] Higashimoto, Y., Opara, E. C, and Liddle, R. A. (1995). Dietary regulation of glucose-dependent insulinotropic peptide (GIP) gene expression in rat small intestine. Comp Biochem Physiol, 110C:207-214. [38] Tseng, C-C, Jarboe, L. A., and Wolfe, M. M. (1994). Regulation of glucose-dependent insulinotropic peptide gene expression by a glucose meal. Am J Physiol, 266:G887-G891. [39] Berghofer, P., Peterson, R. G., Schneider, K., Fehmann, H-C, and Gdke, B. (1997). Incretin hormone expression in the gut of diabetic mice and rats. Metabolism, 46:261-267. 204 [40] Wolfe, M. M., Zhao, K-B., Glazier, K. D., Jarboe, L. A., and Tseng, C-C. (2000). Regulation of glucose-dependent insulinotropic polypeptide release by protein in the rat. Am J Physiol, 279:G561-566. [41] Buchan, A. M. J., Polak, J. M., Capella, C, Solcia, C. E., and Pearse, A. G. E. (1978). Electronimmunocytochemical evidence for the K cell localization of gastric inhibitory polypeptide (GIP) in man. Histochemistry, 56:37-44. [42] Polak, J. M., Bloom, S. R., Kuzio, M., Brown, J. C, and Pearse, A. G. E. (1973). Cellular localization of gastric inhibitory polypeptide in the duodenum and jejunum. Gut, 14:284-288. [43] Buffa, R., Polak, J. M., Pearse, A. G. E., Solcia, C. E., Grimelius, L., and Capella, C. (1975). Identification of the intestinal cell storing gastric inhibitory peptide. Histochemistry, 43:249-255. [44] Mortensen, K., Petersen, L. L., and 0rskov, C. (2000). Colocalization of GLP-1 and GIP in human and porcine intestine. Ann NY Acad Sci, 921:469-472. [45] Damholt, A. B., Kofod, H., and Buchan, A. M. J. (1999). Immunocytochemical evidence for a paracrine interaction between GIP and GLP-1-producing cells in canine small intestine. Cell Tissue Res, 298:287-293. [46] Cheung, A. T., Dayanandan, B., Lewis, J. T., Korbutt, G. S., Rajotte, R. V., Bryer-Ash, M., Boylan, M. O., Wolfe, M. M., and Kieffer, T. J. (2000). Glucose-dependent insulin release from genetically engineered K cells. Science, 290:1959-1962. [47] Steinhoff, M., Hesse, H., Goke, B., Steinhoff, A., Eissele, R., and Slater, E. P. (2001). Indirect RT-PCR in-situ hybridization: a novel non-radioactive method for detecting glucose-dependent insulinotropic peptide. Regul Pept, 97:187-194. [48] Kieffer, T. J., Buchan, A. M. J., Barker, H., Brown, J. C, and Pederson, R. A. (1994). Release of gastric inhibitory polypeptide from cultured canine endocrine cells. Am J Physiol, 267:E489-E496. [49] Chow, B. K.-C, Morrow, G. W., Ho, M., Pederson, R. A., Mcintosh, C. H. S., Brown, J. C, and MacGillivray, R. T. A. (1990). Expression of recombinant human glucose-dependent insulinotropic polypeptide in Escherichia coli by sequence-specific proteolysis of a protein A fusion protein. Peptides, 11:1069-1074. [50] Kieffer, T. J., Huang, Z., Mcintosh, C. H. S., Buchan, A. M. J., Brown, J. C, and Pederson, R. A. (1995). Gastric inhibitory polypeptide release from a tumor-derived cell line. Am J Physiol, 269:E316-E322. [51] Kuzio, M., Dryburgh, J. R., Malloy, K. M., and Brown, J. C. (1974). Radioimmunoassay for gastric inhibitory polypeptide. Gastroenterology, 66:357-364. [52] Alam, M. J., and Buchanan, K. D. (1993). Conflicting gastric inhibitory polypeptide data: possible causes. Diabetes Res Clin Prac, 19:93-101. [53] Krarup, T. (1988). Immunoreactive gastric inhibitory polypeptide. Endocr Rev, 9:122-134. 205 [54] Amland, P. F., Jorde, R., Revhaug, A., Myhre, E. S., Burhol, P. G., and Giercksky, K.-E. (1984). Fasting and postprandial GIP values in pigs, rats, dogs, and man measured with five different GIP antisera. Scand J Gastroent, 19:1095-1098. [55] Krarup, T., Hoist, J. J., and Larsen, K. L. (1985). Responses and molecular heterogeneity of IR-GIP after intraduodenal glucose and fat. Am J Physiol, 249:E195-E200. [56] Jorde, R., Burhol, P. G., and Schulz, T. B. (1983). Fasting and postprandial plasma GIP values in man measured with seven different antisera. Regul Pept, 7:87-94. [57] Brown, J. C., Dahl, M., Kwauk, S., Mcintosh, C. H. S., Otte, S. C , and Pederson, R. A. (1981). Actions of GIP. Peptides, 2 (Suppl. 2):241-245. [58] Mentlein, R., Gallwitz, B., and Schmidt, W. E. (1993). Dipeptidyl-peptidase IV hydrolyses gastric inhibitory polypeptide, glucagon-like peptide-1 (7-36)amide, peptide histidine methionine and is responsible for their degradation in human serum. Eur J Biochem, 214:829-835. [59] Kieffer, T. J., Mcintosh, C. H. S., and Pederson, R. A. (1995). Degradation of glucose-dependent insulinotropic polypeptide and truncated glucagon-like peptide 1 in vitro and in vivo by dipeptidyl peptidase IV. Endocrinology, 136:3585-3596. [60] Deacon, C. F., Nauck, M. A., Meier, J. J., Hiicking, K., and Hoist, J. J. (2000). Degradation of endogenous and exogenous gastric inhibitory polypeptide in healthy and in type 2 diabetic subjects as revealed using a new assay for the intact peptide. J Clin Endocrinol Metab, 85:3575-3581. [61] Wolf, R., Rosche, F., Hoffmann, T., and Demuth, H-U. (2001). Immunoprecipitation and liquid chromatographic-mass spectrometric determination of the peptide glucose-dependent insulinotropic polypeptides GIP1-42 and GIP3-42 from human plasma samples. New sensitive method to analyse physiological concentrations of peptide hormones. J Chromatogr A, 926:21-27. [62] Amland, P. F., Jorde, R., Burhol, P. G., and Giercksky, K-E. (1984). Similar plasma GIP responses in obese and lean subjects after an oral test meal and after intraduodenal stimulation with fat and glucose. Jnt J Obes, 8:649-653. [63] Greenberg, G. R., and Pokol-Daniel, S. (1994). Neural modulation of glucose-dependent insulinotropic peptide (GIP) and insulin secretion in conscious dogs. Pancreas, 9:531-535. [64] Ebert, R., Ulmer, K., and Creutzfeldt, W. (1979). Release of gastric inhibitory polypeptide (GIP) by intraduodenal acidification in rats and humans and abolishment of the incretin effect of acid by GIP-antiserum in rats. Gastroenterology, 76:515-523. [65] Flaten, O. (1981). The effect of adrenoceptor blockade on the release of gastric inhibitory polypeptide after intraduodenal glucose in humans. Scand J Gastroent, 16:641-645. 206 [66] Flaten, O. (1981). Radioimmunoassay of gastric inhibitory polypeptide (GIP) and the effect of intraduodenal acidification on glucose-stimulated and unstimulated GIP release in humans. Scand J Gastroent, 16:545-554. [67] Knapper, J. M. E., Heath, A., Fletcher, J. M., Morgan, L. M., and Marks, V. (1995). GIP and GLP-1 (7-36)amide secretion in response to intraduodenal infusions of nutrients in pigs. Comp Biochem Physiol, lllC:445-450. [68] Bryer-Ash, M., Cheung, A., and Pederson, R. A. (1994). Feedback regulation of glucose-dependent insulinotropic polypeptide (GIP) secretion by insulin in conscious rats. Regul Pept, 51:101-109. [69] Nelson, R. L., Go, V. L. W., McCullough,. A. J., Ilstrup, D. M., and Service, F. J. (1986). Lack of a direct effect of the autonomic nervous system on glucose-stimulated gastric inhibitory polypeptide (GIP) secretion in man. Dig Dis Sci, 31:929-935. [70] O'Dorisio, T. M., Spaeth, J. T., Martin, E. W. Jr., Sirinek, K. R., Thomford, N. R., Mazzaferri, E. L., and Cataland, S. (1978). Mannitol and glucose: effects on gastric acid secretion by endogenous gastric inhibitory polypeptide. Dig Dis Sci, 23:1079-1083. [71] Lavin, J. H., Wittert, G. A., Andrews, J., Yeap, B., Wishart, J. M., Morris, H. A., Morley, J. E., Horowitz, M., and Read, N. W. (1998). Interaction of insulin, glucagon-like peptide 1, gastric inhibitory polypeptide, and appetite in response to intraduodenal carbohydrate. Am J Clin Nutr, 68:591-598. [72] LeRioth, D., Spitz, I. M., Ebert, R., Liel, Y., Odes, S., and Creutzfeldt, W. (1980). Acid-induced gastric inhibitory polypeptide secretion in man. J Clin Endocrinol Metab, 51:1385-1389. [73] Ohneda, A., Kobayashi, T., Nihei, J., Imamura, M., Naito, H., and Tsuchiya, T. (1985). Role of vagus nerve in secretion of gastric inhibitory polypeptide in dogs. Tohuku J Exp Med, 147:183-190. [74] Lucey, M. R., Fairclough, P. D., Wass, J. A. H., Kwasowski, P., Medbak, S., Webb, J., and Rees, L. H. (1984). Response of circulating somatostatin, insulin, gastrin and GIP, to intraduodenal infusion of nutrients in normal man. Clin Endocrinol, 21:209-217. [75] Thomas, F. B., Shook, D. F., O'Dorisio, T. M., Cataland, S., Mekhjian, H. S., Caldwell, J. H., and Mazzaferri, E. L. (1977). Localization of gastric inhibitory polypeptide release by intestinal glucose perfusion in man. Gastroenterology, 72:49-54. [76] Sykes, S., Morgan, J., English, J., and Marks, V. (1980). Evidence for preferential stimulation of gastric inhibitory polypeptide secretion by the rat by actively transported carbohydrates and their analogues. J Endocrinol, 5:210-217. [77] Sirinek, K. R., Levine, B. A., O'Dorisio, T. M., and Cataland, S. (1983). Gastric inhibitory polypeptide (GIP) release by actively transported, structurally similar carbohydrates. Proc Soc Exp Biol Med, 173:379-385. 207 [78] O'Dorisio, T. M., Cataland, S., Stevenson, M., and Mazzaferri, E. L. (1976). Gastric inhibitory polypeptide (GIP): intestinal distribution and stimulation by amino acids and medium-chain triglycerides. Dig Dis Sci, 21:761-765. [79] Sirinek, K. R., Crockett, S. E., Mazzaferri, E. L., Cataland, S., and Thomford, N. R. (1974). Release of gastric inhibitory polypeptide. Comparison of glucose and fat as stimuli. Surg Forum, 25:361-363. [80] Creutzfeldt, W., and Ebert, R. (1985). New developments in the incretin concept. Diabetologia, 28:565-573. [81] Ebert, R., and Creutzfeldt, W. (1984). Evidence that fat induced secretion of gastric inhibitory polypeptide is coupled to chylomicron formation (abstract). Eur J Clin Invest, 14:9A. [82] Pederson, R. A., Schubert, H. E., and Brown, J. C. (1975). Gastric inhibitory polypeptide, its physiologic release and insulinotropic action in dog. Diabetes, 24:1050-1056. [83] Thomas, F. B., Sinar, D., Mazzaferri, E. L., Cataland, S., Mekhjian, H. S., Caldwell, J. PL, and Fromkes, J. J. (1978). Selective release of gastric inhibitory polypeptide by intraduodenal amino acid perfusion in man. Gastroenterology, 74:1261-1265. [84] Schulz, T. B., Jorde, R., and Burhol, P. G. (1982). Gastric inhibitory polypeptide release into the portal vein in response to intraduodenal amino acid loads in anesthetized rats. Scand J Gastroenterol, 17:709-713. [85] Thomas, F. B., Mazzaferri, E. L., Crockett, S. E., Mekhjian, H. S., Gruemer, H. D., and Cataland, S. (1976). Stimulation of secretion of gastric inhibitory polypeptide and insulin by intraduodenal amino acid perfusion. Gastroenterology, 70:523-527. [86] Sirinek, K. R., Pace, W. G., Crockett, S. E., O'Dorisio, T. M., Mazzaferri, E. L., and Cataland, S. (1978). Insulin-induced attenuation of glucose-stimulated gastric inhibitory polypeptide secretion. Am J Surg, 135:151-155. [87] Verdonk, C. A., Rizza, R. A., Nelson, R. L., Go, V. L. W., Gerich, J. E., and Service, F. J. (1980). Interaction of fat-stimulated gastric inhibitory polypeptide on pancreatic alpha and beta cell function. J Clin Invest, 65:1119-1125. [88] Creutzfeldt, W., Talaulicar, M., Ebert, R., and Willms, B. (1980). Inhibition of gastric inhibitory polypeptide (GIP) release by insulin and glucose in juvenile diabetes. Diabetes, 29:140-145. [89] Dryburgh, J. R., Hampton, S. M., and Marks, V. (1980). Endocrine pancreatic control of the release of gastric inhibitory polypeptide. A possible physiological role for C-peptide. Diabetologia, 19:397-401. [90] Mcintosh, C. H. S. (1985). Gastrointestinal somatostatin: distribution, secretion and physiological significance. Life Sci, 37:2043-2058. [91] Schmid, R., Schusdziarra, V., Aulehner, R., Weigert, N., and Classen, M. (1990). Comparison of GLP-1 (7-36amide) and GIP on release of somatostatin-like 208 immunoreactivity and insulin from the isolated rat pancreas. Z Gastroenterol, 28:280-284. [92] Jia, X., Brown, J. C, Kwok, Y. N., Pederson, R. A., and Mcintosh, C. H. S. (1994). Gastric inhibitory polypeptide and glucagon-like peptide-1(7-36) amide exert similar effects on somatostatin secretion but opposite effects on gastrin secretion from the rat stomach. Can J Physiol Pharmacol, 72:1215-1219. [93] Deacon, C. F., Wamberg, S., Bie, P., Hughes, T. E., and Hoist, J. J. (2002). Preservation of active incretin hormones by inhibition of dipeptidyl peptidase IV suppresses meal-induced incretin secretion in dogs. J Endocrinol, 172:355-362. [94] Greenberg, G. R., Chan, B., McDonald, T. J., and Alleyne, J. (1985). The role of vagal integrity in gastrin releasing peptide stimulated gastroenteropancreatic hormone release and gastric acid secretion. Regul Pept, 10:179-187. [95] Williams, R. H., and Champagne, J. (1979). Effects of cholecystokinin, secretin, and pancreatic polypeptide on secretion of gastric inhibitory polypeptide, insulin, and glucagon. Life Sci, 25:947-956. [96] Sirinek, K. R., Cataland, S., O'Dorisio, T. M., Mazzaferri, E. L., Crockett, S. E., and Pace, W. G. (1977). Augmented gastric inhibitory polypeptide response to intraduodenal glucose by exogenous gastrin and cholecystokinin. Surgery, 82:438-442. [97] Kraenzlin, M. E., Ch'ng, J. L. C, Mulderry, P. K., Ghatei, M. A., and Bloom, S. R. (1985). Infusion of a novel peptide, calcitonin gene-related peptide (CGRP) in man. Pharmacokinetics and effects on gastric acid secretion and on intestinal hormones. Regul Pept, 10:189-197. [98] Bayliss, W. M., and Starling, E. H. (1902). The mechanism of pancreatic secretion. J Physiol, 28:235-334. [99] Hoist, J. J., and 0rskov, C. (1994). Glucagon and other proglucagon-derived peptides, in J. H. Walsh, and G. J. Dockray, eds., Gut Peptides: Biochemistry and Physiology: New York, Raven Press Ltd., p. 305-340. [100] Kieffer, T. J., and Habener, J. F. (1999). The glucagon-like peptides. Endocr Rev, 20:876-913. [101] Zhao, X. T., Walsh, J. H., Wong, H., Wang, L., and Lin, H. C. (1999). Intestinal fat-induced inhibition of meal-stimulated gastric acid secretion depends on CCK but not peptide YY. Am J Physiol, 276:G550-G555. [102] Lloyd, K. C. K., Maxwell, V., Kovacs, T. O. G., Miller, J., and Walsh, J. H. (1992). Cholecystokinin receptor antagonist MK-329 blocks intestinal fat-induced inhibition of meal-stimulated gastric acid secretion. Gastroenterology, 102:131-138. [103] Yamagishi, T., and Debas, H. T. (1980). Gastric inhibitory polypeptide (GIP) is not the primary mediator of the enterogastrone action of fat in the dog. Gastroenterology, 78:931-936. 209 [104] El-Munshid, H. A., Hakanson, R., Lindberg, G., and Sundler, F. (1980). Effects of various gastrointestinal peptides on parietal cells and endocrine cells in the oxyntic mucosa of rat stomach. J Physiol, 305:249-265. [105] Soon-Shiong, P., Debas, H. T., and Brown, J. C. (1979). The evaluation of gastric inhibitory polypeptide (GIP) as the enterogastrone. J Surg Res, 26:681-686. [106] Maxwell, V., Shulkes, A., Brown, J. C, Solomon, T. E., Walsh, J. H., and Grossman, M. I. (1980). Effect of gastric inhibitory polypeptide on pentagastrin-stimulated acid secretion in man. Dig Dis Sci, 25:113-116. [107] Mcintosh, C. Ff. S., Pederson, R. A., Koop, H., and Brown, J. C. (1981). Gastric inhibitory polypeptide stimulated secretion of somatostatinlike immunoreactivity from the stomach: inhibition by acetylcholine or vagal stimulation. Can J Physiol Pharmacol, 59:468-472. [108] Soon-Shiong, P., Debas, H. T., and Brown, J. C. (1984). Bethanechol prevents inhibition of gastric acid secretion by gastric inhibitory polypeptide. Am J Physiol, 247:G171-G175. [109] Hoist, J. J., Jensen, S. L., Knuhtsen, S., Nielsen, O. V., and Rehfeld, J. F. (1983). Effect of vagus, gastric inhibitory polypeptide, and HCl on gastrin and somatostatin release from perfused pig antrum. Am J Physiol, 244:G515-G522. [110] Wolfe, M. M., Hocking, M. P., Maico, D. G., and McGuigan, J. E. (1982). Effects of antibodies to gastric inhibitory peptide on gastric acid secretion and gastrin release in the dog. Gastroenterology, 84:941-948. [Ill] Mcintosh, C. H. S., Kwok, Y. N., Mordhorst, T., Nishimura, E., Pederson, R. A., and Brown, J. C. (1983). Enkephalinergic control of somatostatin secretion from the perfused rat stomach. Can J Physiol Pharmacol, 61:657-663. [112] Mcintosh, C. H. S., Jia, X., and Kwok, Y. N. (1990). Characterization of the opioid receptor type mediating inhibition of rat gastric somatostatin secretion. Am J Physiol, 259:G922-G927. [113] Lloyd, K. C, and Walsh, J. H. (1994). Gastric Secretion, in J. H. Walsh, and G. J. Dockray, eds., Gut Peptides: Biochemistry and Physiology: New York, Raven Press Ltd., p. 633-654. [114] Pederson, R. A., Mcintosh, C. H. S., Meuller, M. K., and Brown, J. C. (1981). Absence of a relationship between immunoreactive-gastrin and somatostatin-like immunoreactivity secretion in the perfused rat stomach. Regul Pept, 2:53-60. [115] Schusdziarra, V., Bender, H., and Pfeiffer, E. F. (1983). Release of bombesin-like immunoreactivity from the isolated perfused rat stomach. Regul Pept, 7:21-29. [116] Villar, H. V., Fender, H. R., Rayford, P. L., Bloom, S. R., Ramus, N. L , and Thomson, J. C. (1976). Suppression of gastrin release and gastric secretion by gastric inhibitory polypeptide (GIP) and vasoactive intestinal polypeptide (VIP). Ann Surg, 184:97-102. [117] Wolfe, M. M., and Reel, G. M. (1986). Inhibition of gastrin release by gastric inhibitory peptide mediated by somatostatin. Am J Physiol, 250:G331-G335. 210 [118] Chiba, T., and Yamada, T. (1994). Gut Somatostatin, in J. H. Walsh, and G. J. Dockray, eds., Gut Peptides: Biochemistry and Physiology: New York, Raven Press Ltd., p. 123-145. [119] Rossowski, W. L, Cheng, B-L., Jiang, N-Y., and Coy, D. H. (1998). Examination of somatostatin involvement in the inhibitory action of GIP, GLP-1, amylin and adrenomedullin on gastric acid release using a new SRIF antagonist analogue. Brit J Pharmacol, 125:1081-1087. [120] Denniss, A. R., and Young, J. A. (1978). Modification of salivary duct electrolyte transport in rat and rabbit by physaleamin, VIP, GIP and other enterohormones. Pflugers Arch Eur J Physiol, 376:73-80. [121] Konturek, S- J-, Bilski, J., Tasler, J., and Laskiewicz, J. (1985). Gut hormones in stimulation of gastroduodenal alkaline secretion in conscious dogs. Am J Physiol, 248:G687-G691. [122] Helman, C. A., and Barbezat, G. O. (1977). The effect of gastric inhibitory polypeptide on human jejunal water and electrolyte transport. Gastroenterology, 72:376-379. [123] Barbezat, G. O. (1986). The effect of luminal and hormonal factors on small intestinal water and electrolyte transport. Schweiz Med Wochenschr, 116:946-949. [124] Jensen, R. T., Lemp, G. F., Beinfeld, M. C, and Gardner, J. D. (1982). Actions of natural gastric inhibitory peptide on pancreatic acinar cells: due to contamination with cholecystokinin. Gastroenterology, 82:20-25. [125] Mueller, M. K., Demol, P., Goebell, H., Fladrich, G., and Brown, J. C. (1983). Natural purified porcine gastric inhibitory polypeptide (GIP) stimulates exocrine pancreas secretion due to contamination with a cholecystokinin-like substance. Regul Pept, 5:307-315. [126] Muller, M. K., Scheck, T., Demol, P., and Goebell, H. (1986). Interaction of acetylcholine and gastric inhibitory polypeptide on endocrine and exocrine rat pancreatic secretion: augmentation of acetylcholine-induced amylase and volume secretion by the insulinotropic action of gastric inhibitory polypeptide. Digestion, 33:45-52. [127] Mueller, M. K., Scheck, T., Dreesmann, V., Miodonski, A., and Goebell, H. (1987). GIP potentiates CCK stimulated pancreatic enzyme secretion: effect of GIP and CCK on amylase secretion. Pancreas, 2:106-113. [128] Cheeseman, C. I., and Tsang, R. (1996). The effect of GIP and glucagon-like peptides on intestinal basolateral membrane hexose transport. Am J Physiol, 271:G477-G482. [129] Cheeseman, C. I., and O'Neill, D. (1998). Basolateral D-glucose transport activity along the crypt-villus axis in rat jejunum and upregulation induced by gastric inhibitory peptide and glucagon-like peptide-2. Exp Physiol, 83:605-616. 211 [130] Keinke, O., Ehrlein, H. J., and Wulschke, S. (1987). Mechanical factors regulating gastric emptying examined by the effects of exogenous cholecystokinin and secretin on canine gastroduodenal motility. Can J Physiol Pharmacol, 65:287-292. [131] Young, A. A., Gedulin, B. R., and Rink, T. J. (1996). Dose-responses for the slowing of gastric emptying in a rodent model by glucagon-like peptide (7-36)NH2, amylin, cholecystokinin, and other possible regulators of nutrient uptake. Metabolism, 45:1-3. [132] Fara, J. W., and Salazar, A. M. (1978). Gastric inhibitory polypeptide increases mesenteric blood flow. Proc Soc Exp Biol Med, 158:446-448. [133] Thor, P., Laskiewicz, J., Konturek, S. J., and Creutzfeldt, W. (1987). Role of GIP and insulin in glucose-induced changes in intestinal motility patterns. Am J Physiol, 252:G8-G12. [134] Castresana, M., Lee, K. Y., Chey, W. Y., and Yajima, H. (1978). Effects of motilin and octapeptide of cholecystokinin on antral and duodenal myoelectric activity in the interdigestive state and during inhibition by secretin and gastric inhibitory polypeptide. Digestion, 17:300-308. [135] Roberge, J. N., and Brubaker, P. L. (1993). Regulation of intestinal proglucagon-derived peptide secretion by glucose-dependent insulinotropic peptide in a novel enteroendocrine loop. Endocrinology, 133:233-240. [136] Damholt, A. B., Buchan, A. M. J., and Kofod, H. (1998). Glucagon-like-peptide-1 secretion from canine L-cells is increased by glucose-dependent-insulinotropic peptide but unaffected by glucose. Endocrinology, 139:2085-2091. [137] Brubaker, P. L. (1991). Regulation of intestinal proglucagon-derived peptide secretion by intestinal regulatory peptides. Endocrinology, 128:3175-3182. [138] Dumoulin, V., Dakka, T., Plaisancie, P., Chayvialle, J. A., and Cuber, J. C. (1995). Regulation of glucagon-like peptide-1-(7-36) amide by intestinal neurotransmitters and hormones in the isolated vascularly perfused rat colon. Endocrinology, 136:5182-5188. [139] Herrmann-Rinke, C , Voge, A., Hess, M., and Goke, B. (1995). Regulation of glucagon-like peptide-1 secreion from rat ileum by neurotransmitters and peptides. J Endocrinol, 147:25-31. [140] Brubaker, P. L., Schloos, J., and Drucker, D. J. (1998). Regulation of glucagon-like peptide-1 synthesis and secretion in the GLUTag enteroendocrine cell line. Endocrinology, 139:4108-4114. [141] Kogire, M., Inoue, K., Sumi, S., Doi, R., Takaori, K., Yun, M., Fujii, N., Yajima, H., and Tobe, T. (1988). Effects of synthetic human gastric inhibitory polypeptide on splanchnic circulation in dogs. Gastroenterology, 95:1636-1640. [142] Kogire, M., Inoue, K., Sumi, S., Doi, R., Yun, M., Kaji, H., and Tobe, T. (1992). Effects of gastric inhibitory polypeptide and glucagon on portal venous and hepatic arterial flow in conscious dogs. Dig Dis Sci, 37:1666-1670. 212 [143] Svensson, A. M., Efendic, S., Ostenson, C. G., and Jansson, L. (1997). Gastric inhibitory polypeptide and splanchnic blood perfusion: augmentation of the islet blood flow increase in hyperglycemic rats. Peptides, 18:1055-1059. [144] Elrick, H., Stimler, L . , Hlad, C. J. Jr., and Arai, Y. (1964). Plasma insulin response to oral and intravenous glucose administration. J Clin Endocrinol, 24:1076-1082. [145] Mclntyre, N., Holdsworth, C. D., and Turner, D. S. (1965). Intestinal factor in the control of insulin secretion. J Clin Endocrinol Metab, 25:1317-1324. [146] Perley, M. J., and Kipnis, D. M. (1967). Plasma insulin responses to oral and intravenous glucose. J Clin Invest, 46:1954-1962. [147] Mclntyre, N., Holdsworth; C. D., and Turner, D. S. (1964). New interpretation of oral glucose tolerance. Lancet, 2:20. [148] Unger, R. H., and Eisentraut, A. M. (1969). Entero-insular axis. Arch Intern Med, 123:261-266. [149] Creutzfeldt, W. (1979). The incretin concept today. Diabetologia, 16:75-85. [150] Moore, B., Edie, E. S., and Abram, J. H. (1906). On the treatment of diabetes mellitus by acid extract of duodenal mucous membrane. Biochem J, 1:28. [151] Banting, F. G., and Best, C. H. (1922). The internal secretion of the pancreas. J Lab Clin Med, 7:251-264. [152] La Barre, J., and Still, E. V. (1930). Studies on the physiology of secretin. Am J Physiol, 91:649-653. [153] La Barre, J. (1932). Sur les possibilites d'un traitement due diabete par l'incretine. Bull Acad Roy Med Belg, 12:620-634. [154] Zunz, E., and La Barre, J. (1929). Contributions a l'etude der variations physiologiques de la secretion interne du pancreas: relations entre les secretion externe du pancreas. Arch Int Physiol Biochem, 31:20-44. [155] Hoist, J. J., and 0rskov, C (2001). Incretin hormones - an update. Scand J Clin Lab Invest, 61 (Suppl. 234):75-86. [156] Rehfeld, J. F. (1972). Gastrointestinal hormones and insulin secretion. Scand J Gastroent, 7:289-292. [157] D'Alessio, D. (1997). Peptide hormone regulation of islet cells. Rec Prog Horm Res, 29:297-300. [158] Nauck, M. A. (1999). Is glucagon-like peptide 1 an incretin hormone? Diabetologia, 42:373-379. [159] Balkan, B., and Li, X. (2000). Portal GLP-1 administration in rats augments the insulin response to glucose via neuronal mechanisms. Am J Physiol, 279:R1449-R1454. 213 [160] Dupre, J., and Beck, J. C. (1966). Stimulation of release of insulin by an extract of intestinal mucosa. Diabetes, 15:555-559. [161] Rabinovitch, A., and Dupre, J. (1972). Insulinotropic and glucagonotropic activities of crude preparations of cholecysokinin-pancreozymin (abstract). Clin Res, 20:945. [162] Dupre, J., Ross, S. A., Watson, D., and Brown, J. C. (1973). Stimulation of insulin secretion by gastric inhibitory polypeptide in man. J Clin Endocrinol Metab, 37:826-828. [163] Andersen, D. K . , Elahi, D., Brown, J. C, Tobin, J. D., and Andres, R. (1978). Oral glucose augmentation of insulin secretion. Interactions of gastric inhibitory polypeptide with ambient glucose and insulin levels. J Clin Invest, 62:152-161. [164] Elahi, D., Andersen, D. K . , Brown, J. C, Debas, H. T., Hershcopf, R. J., Raizes, G. S., Tobin, J. D., and Andres, R. (1979). Pancreatic a- and 6-cell responses to GIP infusion in normal man. Am J Physiol, 237:E185-E191. [165] Pederson, R. A., and Brown, J. C. (1976). The insulinotropic action of gastric inhibitory polypeptide in the perfused isolated rat pancreas. Endocrinology, 99:780-785. [166] Schmidt, W. E., Siegel, E. G., Kiimmel, H., Gallwitz, B., and Creutzfeldt, W. (1987). Commercially available preparations of porcine glucose-dependent insulinotropic polypeptide (GIP) contain a biologically inactive GIP-fragment and cholecystokinin-33/-39. Endocrinology, 120:835-837. [167] Schauder, P., Brown, J. C, Frerichs, H., and Creutzfeldt, W. (1975). Gastric inhibitory polypeptide: effect on glucose-induced insulin release from isolated rat pancreatic islets in vitro. Diabetologia, 11:483-484. [168] Schauder, P., Schindler, B., Panten, U., Brown, J. C, Frerichs, H., and Creutzfeldt, W. (1977). Insulin release from isolated rat pancreatic islets induced by alpha-ketoisocapronic acid, L-leucine, D-glucose or D-glyceraldehyde: effect of gastric inhibitory polypeptide or glucagon. Mol Cell Endocrinol, 7:115-123. [169] Fujimoto, W. Y. (1981). Effect of gut peptides on glucose-stimulated insulin release by monolayer cultures of neonatal rat islet cells. Horm Met Res, 13:135-138. [170] Szecowka, J., Grill, V., Sandberg, E., and Efendic, S. (1982). Effect of GIP on the secretion of insulin and somatostatin and the accumulation of cyclic AMP in vitro in the rat. Acta Endocrinol, 99:416-421. [171] Mazzaferri, E. L., Ciofalo, L., Waters, L. A., Starich, G. H., Groshong, J. C, and DePalma, L. (1983). Effects of gastric inhibitory polypeptide on leucine- and arginine-stimulated insulin release. Am J Physiol, 245:E114-E120. [172] Siegel, E. G., and Creutzfeldt, W. (1985). Stimulation of insulin release in isolated rat islets by GIP in physiological concentrations and its relation to islet cyclic AMP content. Diabetologia, 28:857-861. 214 [173] Fujimoto, W. Y., Ensinck, J. W., Merchant, F. W., Williams, R. H., Smith, P. H., and Johnson, D. G. (1978). Stimulation by gastric inhibitory polypeptide of insulin and glucagon secretion by rat islet cultures. Proc Soc Exp Biol Med, 157:89-93. [174] Verchere, C. B. (1991). The role of gastric inhibitory polypeptide in the regulation of pancreatic endocrine secretion [Ph.D. Dissertation], University of British Columbia, Vancouver, 224 p. [175] Amiranoff, B., Vauclin-Jacques, N., and Laburthe, M . (1984). Functional GIP receptors in a hamster pancreatic beta-cell line In 111: specific binding and biological effects. Biochem Biophys Res Commun, 123:671-676. [176] Lu, M . , Wheeler, M . B., Leng, X-H. , and Boyd, A. E., 3rd (1993). The role of the free cytosolic calcium level in |3-cell signal transduction by gastric inhibitory polypeptide and glucagon-like peptide 1(7-37). Endocrinology, 132:94-100. [177] Fehmann, H - C , and Habener, J. F. (1991). Homologous desensitization of the insulinotropic glucagon-like peptide-I(7-37) receptor on insulinoma (HIT-T15) cells. Endocrinology, 128:2880-2888. [178] Kieffer, T. J., Verchere, C. B., Fell, C. D., Huang, Z., Brown, J. C , and Pederson, R. A. (1993). Glucose-dependent insulinotropic polypeptide stimulated insulin release from a tumour-derived |3-cell lin ((3TC3). Can J Physiol Pharmacol, 71:917-922. [179] Swanston-Flatt, S. K., and Flatt, P. R. (1988). Effects of amino acids, hormones and drugs on insulin release and 4 5 Ca uptake by transplantable rat insulinoma cells maintained in tissue culture. Gen Pharmacol, 19:239-242. [180] Montrose-Rafizadeh, C , Egan, J. M . , and Roth, J. (1994). Incretin hormones regulate glucose-dependent insulin secretion in RIN 1046-38 cells: mechanisms of action. Endocrinology, 135:589-594. [181] O'Harte, F. P. M . , Abdel-Wahab, Y. H., Conlon, J. M . , and Flatt, P. R. (1998). Amino terminal glycation of gastric inhibitory polypeptide enhances its insulinotropic action on clonal pancreatic B-cells. Biochim Biophys Acta, 23:319-327. [182] Maletti, M . , Altman, J. J., Hoa, D. H., Carlquist, M . , and Rosselin, G. (1987). Evidence of functional gastric inhibitory polypeptide (GIP) receptors in human insulinoma. Binding of synthetic human GIP 1-31 and activation of adenylate cyclase. Diabetes, 36:1336-1340. [183] Ebert, R., Unger, H., and Creutzfeldt, W. (1983). Preservation of incretin activity after removal of gastric inhibitory polypeptide (GIP) from rat gut extracts by immunoadsorption. Diabetologia, 24:449-454. [184] Ebert, R., and Creutzfeldt, W. (1982). Influence of gastric inhibitory polypeptide antiserum on glucose-induced insulin secretion in rats. Endocrinology, 111:1601-1606. 215 [185] Lewis, J. T., Dayanandan, B., Habener, J. F., and Kieffer, T. J. (2000). Glucose-dependent insulinotropic polypeptide confers early phase insulin release to oral glucose in rats: demonstration by a receptor antagonist. Endocrinology, 141:3710-3716. [186] Tseng, C - C , Kieffer, T. J., Jarboe, L. A., Usdin, T. B., and Wolfe, M . M . (1996). Postprandial stimulation of insulin release by glucose-dependent insulinotropic polypeptide (GIP). Effect of a specific glucose-dependent insulinotropic polypeptide receptor antagonist in the rat. J Clin Invest, 98:2440-2445. [187] Tseng, C-C , Zhang, X. Y. , and Wolfe, M . M . (1999). Effect of GIP and GLP-1 antagonists on insulin release in the rat. Am J Physiol, 276:E1049-E1054. [188] Gault, V. A., O'Harte, F. P., Harriott, P., and Flatt, P. R. (2002). Characterization of the cellular and metabolic effects of a novel enzyme-resistant antagonist of glucose-dependent insulinotropic polypeptide. Biochem Biophys Res Commun, 290:1420-1426. [189] Miyawaki, K., Yamada, Y., Yano, H., Niwa, H., Ban, N. , Ihara, Y., Kubota, A., Fujimoto, S., Kajikawa, M . , Kuroe, A., Tsuda, K., Hashimoto, H., Yamashita, T., Jomori, T., Tashiro, F., Miyazaki, J., and Seino, Y. (1999). Glucose intolerance caused by a defect in the entero-insular axis: a study in gastric inhibitory polypeptide receptor knockout mice. Proc Natl Acad Sci USA, 96:14843-14847. [190] Ding, W. G., and Gromada, J. (1997). Protein kinase A-dependent stimulation of exocytosis in mouse pancreatic 6-cells by glucose-dependent insulinotropic polypeptide. Diabetes, 46:615-621. [191] Gromada, J., Bokvist, K., Ding, W. G., Hoist, J. J., Nielsen, J. H., and Rorsman, P. (1998). Glucagon-like peptide 1 (7-36) amide stimulates exocytosis in human pancreatic 6-cells by both proximal and distal regulatory steps in stimulus-secretion coupling. Diabetes, 47:57-65. [192] Gromada, J., Hoist, J. J., and Rorsman, P. (1998). Cellular regulation of islet hormone secretion by the incretin hormone glucagon-like peptide 1. Pfliigers Arch Eur J Physiol, 435:583-594. [193] Beguin, P., Nagashima, K., Nishimura, M. , Gonoi, T., and Seino, S. (1999). PKA-mediated phosphorylation of the human K(ATP) channel: separate roles of Kir6.2 and SUR1 subunit phosphorylation. EMBO J, 18:4722-4732. [194] Wahl, M . A., Plehn, R. J., Landsbeck, E. A., Verspohl, E. J., and Ammon, H. P. (1992). Are ionic fluxes of pancreatic beta cells a target for gastric inhibitory polypeptide? Mol Cell Endocrinol, 90:117-123. [195] Yajima, H., Komatsu, M. , Schermerhorn, T., Aizawa, T., Kaneko, T., Nagai, M. , Sharp, G. W., and Hashizume, K. (1999). cAMP enhances insulin secretion by an action on the ATP-sensitive K+ channel-independent pathway of glucose signaling in rat pancreatic islets. Diabetes, 48:1006-1012. [196] Suzuki, S., Kawai, K., Ohashi, S., Watanabe, Y., and Yamashita, K. (1992). Interaction of glucagon-like peptide-l(7-36) amide and gastric inhibitory polypeptide or 216 cholecystokinin on insulin and glucagon secretion from the isolated perfused rat pancreas. Metabolism, 41:359-363. [197] Fehmann, H - C , Goke, B., Goke, R., Trautmann, M . E., and Arnold, R. (1989). Synergistic stimulatory effect of glucagon-like peptide-1 (7-36) amide and glucose-dependent insulin-releasing polypeptide on the endocrine rat pancreas. FEBS Lett, 252:109-112. [198] Gallwitz, B., Witt, M. , Folsch, U. R., Creutzfeldt, W., and Schmidt, W. E. (1993). Binding specificity and signal transduction of receptors for glucagon-like peptide-l(7-36)amide and gastric inhibitory polypeptide on RINm5F insulinoma cells. J Mol Endocrinol, 10:259-268. [199] Siegel, E. G., Schulze, A., Schmidt, W. E., and Creutzfeldt, W. (1992). Comparison of the effect of GIP and GLP-1 (7-36amide) on insulin release from rat pancreatic islets. Eur J Clin Invest, 22:154-157. [200] Verchere, C. B., Kwok, Y. N. , and Brown, J. C. (1991). Modulation of acetylcholine-stimulated insulin release by glucose and gastric inhibitory polypeptide. Pharmacology, 42:273-282. [201] Sandberg, E., Ahren, B., Tendler, D., and Efendic, S. (1988). Cholecystokinin-33 potentiates and vasoactive intestinal polypeptide inhibits gastric inhibitory polypeptide -induced insulin secretion in the perfused rat pancreas. Acta Endocrinol, 117:545-551. [202] Pederson, R. A., and Brown, J. C. (1978). Interaction of gastric inhibitory polypeptide, glucose, and arginine on insulin and glucagon secretion from the perfused rat pancreas. Endocrinology, 103:610-615. [203] Rasmussen, H., Zawalich, K. C , Ganesan, S., Calle, R., and Zawalich, W. S. (1990). Physiology and pathophysiology of insulin secretion. Diabetes Care, 13:655-666. [204] Zawalich, W. S., Zawalich, K. C , and Rasmussen, H. (1989). Interactions between cholinergic agonists and enteric factors in the regulation of insulin secretion from isolated perifused rat islets. Acta Endocrinol, 120:702-707. [205] McCullough, A. J., Marshall, J. B., Bingham, C. P., Rice, B. L., Manning, L. D., and Kalhan, S. C. (1985). Carbachol modulates GIP-mediated insulin release from rat pancreatic lobules in vitro. Am J Physiol, 248:E299-E303. [206] Brunicardi, F. C , Druck, P., Seymour, N. E., Sun, Y. S., Elahi, D., and Andersen, D. K. (1990). Selective neurohormonal interactions in islet cell secretion in the isolated perfused human pancreas. J Surg Res, 48:273-278. [207] Ma, H. T., Kato, M. , and Tatemoto, K. (1996). Effects of pancreastatin and somatostatin on secretagogues-induced rise in intracellular free calcium in single rat pancreatic islet cells. Regul Pept, 61:143-148. [208] Silvestre, R. A., Salas, M . , Garcfa-Hermida, O., Fontela, T., Degano, P., and Marco, J. (1994). Amylin (islet amyloid polypeptide) inhibition of insulin release in the perfused rat pancreas: implication of the adenylate cyclase/cAMP system. Regul Pept, 50:193-199. 217 [209] Rodrfguez-Gallardo, J., Silvestre, R. A., and Marco, J. (1999). Inhibitory effect of enterostatin on the beta cell response to digestive insulinotropic peptides. Int J Obes, 23:787-792. [210] Miralles, P., Peiro, E., Silvestre, R. A., Villanueva, M . L., and Marco, J. (1988). Effects of galanin on islet cell secretory responses to VIP, GIP, 8-CCK, and glucagon by the perfused rat pancreas. Metabolism, 37:766-770. [211] Peiro, E., Miralles, P., Silvestre, R. A., Villanueva, M, L., and Marco, J. (1989). Pancreastatin inhibits insulin secretion as induced by glucagon, vasoactive intestinal peptide, gastric inhibitory peptide, and 8-cholecystokinin in the perfused rat pancreas. Metabolism, 38:679-682. [212] Ishizuka, J., Tatemoto, K., Cohn, D. V., Thompson, J. C , and Greeley, G. H., Jr. (1991). Effects of pancreastatin and chromogranin A on insulin release stimulated by various insulinotropic agents. Regul Pept, 34:25-32. [213] Lu, M . , Wheeler, M . B., Leng, X.-H. , and Boyd, A. E., 3rd (1993). Stimulation of insulin secretion and insulin gene expression by gastric inhibitory polypeptide. Trans Assoc Am Phys, 106:42-53. [214] Fehmann, H - C , and Goke, B. (1995). Characterization of GIP(l-30) and GIP(l-42) as stimulators of proinsulin gene transcription. Peptides, 16:1149-1152. [215] Schafer, R., and Schatz, H. (1979). Stimulation of (Pro-)insulin biosynthesis and release by gastric inhibitory polypeptide in isolated islets of rat pancreas. Acta Endocrinol, 91:493-500. [216] Wang, Y., Montrose-Rafizadeh, C , Adams, L., Raygada, M . , Nadiv, O., and Egan, J. M . (1996). GIP regulates glucose transporters, hexokinases, and glucose-induced insulin secretion in RIN 1046-38 cells. Mol Cell Endocrinol, 116:81-87. [217] Adrian, T. E., Bloom, S. R., Hermansen, K., and Iversen, J. (1978). Pancreatic polypeptide, glucagon and insulin secretion from the isolated perfused canine pancreas. Diabetologia, 14:413-7. [218] Ipp, E., Dobbs, R. E., Harris, V., Arimura, A., Vale, W., and Unger, R. H. (1977). The effects of gastrin, gastric inhibitory polypeptide, secretin, and the octapeptide of cholecystokinin upon immunoreactive somatostatin release by the perfused canine pancreas. J Clin Invest, 60:1216-1219. [219] Ding, W. G., Renstrom, E., Rorsman, P., Buschard, K., and Gromada, J. (1997). Glucagon-like peptide I and glucose-dependent insulinotropic polypeptide stimulate Ca2+-induced secretion in rat a-cells by a protein kinase A-mediated mechanism. Diabetes, 46:792-800. [220] Nauck, M . A., Heimesaat, M . M . , Orskov, C , Hoist, J. J., Ebert, R., and Creutzfeldt, W. (1993). Preserved incretin activity of glucagon-like peptide 1 [7-36 amide] but not of synthetic human gastric inhibitory polypeptide in patients with type-2 diabetes mellitus. J Clin Invest, 91:301-307. 218 [221] Nauck, M . , Schmidt, W. E., Ebert, R., Strietzel, J., Cantor, P., Hoffmann, G., and Creutzfeldt, W. (1989). Insulinotropic properties of synthetic human gastric inhibitory polypeptide in man: Interactions with glucose, phenylalanine, and cholecystokinin-8. J Clin Endocrinol Metab, 69:654-662. [222] Elahi, D., Raizes, G. S., Andres, R., Hershcopf, R. J., Muller, D. C , Tobin, J. D., and Andersen, D. K. (1982). Interaction of arginine and gastric inhibitory polypeptide on insulin release in man. Am J Physiol, 242:E343-351. [223] Yasuda, K., Inagaki, N. , Yamada, Y., Kubota, A., Seino, S., and Seino, Y. (1994). Hamster gastric inhibitory polypeptide receptor expressed in pancreatic islets and clonal insulin-secreting cells:' its structure and functional properties. Biochem Biophys Res Commun, 205:1556-1562. [224] Moens, K., Heimberg, H., Flamez, D., Huypens, P., Quartier, E., Ling, Z., Pipeleers, D., Gremlich, S., Thorens, B., and Schuit, F. (1996). Expression and functional activity of glucagon, glucagon-like peptide I, and glucose-dependent insulinotropic peptide receptors in rat pancreatic islet cells. Diabetes, 45:257-261. [225] Kaplan, A. M . , and Vigna, S. R. (1994). Gastric inhibitory polypeptide (GIP) binding sites in rat brain. Peptides, 15:297-302. [226] Whitcomb, DC, O'Dorisio, TM, Nishikawara, MT, Shetzline, M , and Cataland, S (1984). Identification of target organs for gastric inhibitory polypeptide (GIP) with a new in vivo radioreceptor assay (abstract). Dig Dis Sci, 29 (August Supplement):95S. [227] Madsbad, S., Kehlet, H., Hilsted, J., and Tronier, B. (1983). Discrepancy between plasma C-peptide and insulin response to oral and intravenous glucose. Diabetes, 32:436-438. [228] Nauck, M . A., Homberger, E., Siegel, E. G., Allen, R. C , Eaton, R. P., Ebert, R., and Creutzfeldt, W. (1986). Incretin effects of increasing glucose loads in man calculated from venous insulin and C-peptide responses. J Clin Endocrinol Metab, 63:492-498. [229] Ikeda, T., Yoshida, T., Honda, M . , Ito, Y., Murakami, I., Mokuda, O., Tominaga, M . , and Mashiba, H. (1987). Effect of intestinal factors on extraction of insulin in perfused rat liver. Am J Physiol, 253:E603-E607. [230] Hanks, J. B., Andersen, D. K., Wise, I. E., Putnam, W. S., Meyers, W. C , and Jones, R. S. (1984). The hepatic extraction of gastric inhibitory polypeptide and insulin. Endocrinology, 115:1011-1018. [231] Chap, Z., Ishida, T., Chou, J., Lewis, R., Hartley, C , Entman, M . , and Field, J. B. (1985). Effects of atropine and gastric inhibitory polypeptide on hepatic glucose uptake and insulin extraction in conscious dogs. J Clin Invest, 76:1174-1181. [232] Matsumura, M . , Akiyoshi, H. , and Saito, S. (1980). Effects of somatostatin on gastrointestinal hormone-induced glycogenolysis and gluconeogenesis in cultured liver cells. Gastroenterologia Japonica, 15:439-443. 219 [233] Hartmann, H., Ebert, R., and Creutzfeldt, W. (1986). Insulin-dependent inhibition of hepatic glycogenolysis by gastric inhibitory polypeptide (GIP) in perfused rat liver. Diabetologia, 29:112-114. [234] Ebert, R., and Creutzfeldt, W. (1987). Metabolic effects of gastric inhibitory polypeptide. Front Horm Res, 16:175-185. [235] Andersen, D. K., Sun, Y. S., Brunicardi, F. C , Berlin, S. A., Lebovitz, H. E., and Elahi, D. (1984). Regulation of hepatic glucose production by gastric inhibitory polypeptide (GIP), insulin (INS) and glucagon (GLUC). Dig Dis Sci, 29:A5. [236] Elahi, D., Meneilly, G. S., Minaker, K. L., Rowe, L. D., and Andersen, D. K. (1986). Regulation of glucose production by gastric inhibitory polypeptide in man (Abstracts presented at the Sixth International Symposium on Gastrointestinal Hormones. Vancouver, British Columbia, Canada, July 6-10, 1986). Can J Physiol Pharmacol, 65 (Suppl.): 18. [237] Andersen, D. K., Putnam, W. S., Hanks, J. B., Wise, J. E., Lebovitz, H. E., and Jones, R. S. (1982). Gastric inhibitory polypeptide (GIP) suppression of hepatic glucose production [abstract]. Regul Pept, 1 (Suppl. 2):4. [238] Pederson, R. A., Schubert, H. E., and Brown, J. C. (1975). The insulinotropic action of gastric inhibitory polypeptide. Can J Physiol Pharmacol, 53:217-223. [239] O'Harte, F. P. M . , Gray, A. M . , and Flatt, P. R. (1998). Gastric inhibitory polypeptide and effects of glycation on glucose transport and metabolism in isolated mouse abdominal muscle. J Endocrinol, 156:237-243. [240] Yip, R. G., Boylan, M . O., Kieffer, T. J., and Wolfe, M . M . (1998). Functional GIP receptors are present on adipocytes. Endocrinology, 139:4004-4007. [241] Mcintosh, C. H. S., Bremsak, I., Lynn, F. C , Gill, R., Hinke, S. A., Gelling, R., Nian, C , McKnight, G., Jaspers, S., and Pederson, R. A. (1999). Glucose-dependent insulinotropic polypeptide stimulation of lipolysis in differentiated 3T3-L1 cells: wortmannin-sensitive inhibition by insulin. Endocrinology, 140:398-404. [242] Ebert, R., Nauck, M . , and Creutzfeldt, W. (1991). Effect of exogenous or endogenous gastric inhibitory polypeptide (GIP) on plasma triglyceride responses in rats. Horm Met Res, 23:517-521. [243] Jorde, R., Pettersen, J. E., and Burhol, P. G. (1984). Lack of effect of exogenous or endogenous gastric inhibitory polypeptide on the elimination rate of Intralipid in man. Acta Med Scand, 216:19-23. [244] Ohneda, A., Kobayashi, T., and Nihei, J. (1983). Effect of endogenous gastric inhibitory polypeptide (GIP) on the removal of triacylglycerol in dogs. Regul Pept, 6:25-32. [245] Wasada, T., McCorkle, K., Harris, V., Kawai, K., Howard, B., and Unger, R. H. (1981). Effect of gastric inhibitory polypeptide on plasma levels of chylomicron triglycerides in dogs. J Clin Invest, 68:1106-1107. 220 [246] Eckel, R. H., Fujimoto, W. Y., and Brunzell, J. D. (1981). Effect of in-vitro lifespan of 3T3-L1 cells on hormonal responsiveness of lipoprotein lipase activity. Int J Obes, 5:571-577. [247] Eckel, R. H., Fujimoto, W. Y., and Brunzell, J. D. (1979). Gastric inhibitory polypeptide enhanced lipoprotein lipase activity in cultured preadipocytes. Diabetes, 28:1141-1142. [248] Knapper, J. M . , Puddicombe, S. M . , Morgan, L. M . , and Fletcher, J. M . (1995). Investigations into the actions of glucose-dependent insulinotropic polypeptide and glucagon-like peptide-l(7-36)amide on lipoprotein lipase activity in explants of rat adipose tissue. J Nutr, 125:183-188. [249] Knapper, J. M . , Puddicombe, S. M . , Morgan, L. M. , Fletcher, J. M . , and Marks, V. (1993). Glucose dependent insulinotropic polypeptide and glucagon-like peptide-l(7-36)amide: effects on lipoprotein lipase activity. Biochem Soc Trans, 21:135S. [250] Beck, B., and Max, J. P. (1983). Gastric inhibitory polypeptide enhancement of the insulin effect on fatty acid incorporation into adipose tissue in the rat. Regul Pept, 7:3-8. [251] Miyawaki, K., Yamada, Y. , Ban, N. , Ihara, Y., Tsukiyama, K., Zhou, H., Fujimoto, S., Oku, A., Tsuda, K., Toyokuni, S., Hiai, H., Mizunoya, W., Fushiki, T., Hoist, J. J., Makino, M . , Tashita, A., Kobara, Y. , Tsubamoto, Y. , Jinnouchi, T., Jomori, T., and Seino, Y. (2002). Inhibition of gastric inhibitory polypeptide signalling prevents obesity. Nat Med, 8:738-742. [252] Starich, G. H., Bar, R. S., and Mazzaferri, E. L. (1985). GIP increases insulin receptor affinity and cellular sensitivity in adipocytes. Am J Physiol, 249:E603-E607. [253] Hauner, H. , Glatting, G., Kaminska, D., and Pfeiffer, E. F. (1988). Effects of gastric inhibitory polypeptide on glucose and lipid metabolism of isolated rat adipocytes. Ann Nutr Metab, 32:282-8. [254] Oben, J., Morgan, L., Fletcher, J., and Marks, V. (1991). Effect of the entero-pancreatic hormones, gastric inhibitory polypeptide and glucagon-like polypeptide-1(7-36) amide, on fatty acid synthesis in explants of rat adipose tissue. J Endocrinol, 130:267-272. [255] Dupre, J., Greenidge, N. , McDonald, T. J., Ross, S. A., and Rubinstein, D. (1976). Inhibition of actions of glucagon in adipocytes by gastric inhibitory polypeptide. Metabolism, 25:1197-1199. [256] Ottlecz, A., Samson, W. K., and McCann, S. M . (1985). The effects of gastric inhibitory polypeptide (GIP) on the release of anterior pituitary hormones. Peptides, 6:115-119. [257] Murphy, W. A., Lance, V. A., Sueiras-Diaz, J., and Coy, D. H. (1983). Effects of secretin and gastric inhibitory polypeptide on human pancreatic growth hormone-releasing factor(l-40)-stimulated growth hormone levels in the rat. Biochem Biophys Res Commun, 112:469-474. 221 [258] Zhong, Q., Bollag, R. J.', Dransfield, D. T., Gasalla-Herraiz, J., Ding, K. H., Min, L., and Isales, C. M . (2000). Glucose-dependent insulinotropic peptide signaling pathways in endothelial cells. Peptides, 21:1427-1432. [259] Bollag, R. J., Zhong, Q., Ding, K. FL, Phillips, P., Zhong, L., Qin, F., Cranford, J., Mulloy, A. L., Cameron, R., and Isales, C. M . (2001). Glucose-dependent insulinotropic peptide is an integrative hormone with osteotropic effects. Mol Cell Endocrinol, 177:35-41. [260] Lacroix, A., Ndiaye, N. , Tremblay, J., and Hamet, P. (2001). Ectopic and abnormal hormone receptors in adrenal Cushing's syndrome. Endocr Rev, 22:75-110. [261] Nussdorfer, G. G., Bahcelioglu, M . , Neri, G., and Malendowicz, L. K. (2000). Secretin, glucagon, gastric inhibitory polypeptide, parathyroid hormone, and related peptides in the regulation of the hypothalamus- pituitary-adrenal axis. Peptides, 21:309-324. [262] Chabre, O., Liakos, P., Vivier, J., Bottari, S., Bachelot, I., Chambaz, E. M . , Feige, J. J., and Defaye, G. (1998). Gastric inhibitory polypeptide (GIP) stimulates Cortisol secretion, cAMP production and DNA synthesis in an adrenal adenoma responsible for food-dependent Cushing's syndrome. Endocr Res, 24:851-856. [263] Mazzocchi, G., Rebuffat, P., Meneghelli, V., Malendowicz, L. K., Tortorella, C , Gottardo, G., and Nussdorfer, G. G. (1999). Gastric inhibitory polypeptide stimulates glucocorticoid secretion in rats, acting through specific receptors coupled with the adenylate cyclase-dependent signaling pathway. Peptides, 20:589-594. [264] Bollag, R. J., Zhong, Q., Phillips, P., Min, L., Zhong, L., Cameron, R., Mulloy, A. L., Rasmussen, H., Qin, F., Ding, K. Ff., and Isales, C. M . (2000). Osteoblast-derived cells express functional glucose-dependent insulinotropic peptide receptors. Endocrinology, 141:1228-1235. [265] Maletti, M . , Amiranoff, B., Laburthe, M . , and Rosselin, G. (1983). Demonstration of specific receptors for gastric inhibitory peptide (GIP) [French]. C R Acad Sc Paris(III), 297:563-565. [266] Maletti, M . , Portha, B., Carlquist, M . , Kergoat, M . , Laburthe, M . , Marie, J. C , and Rosselin, G. (1984). Evidence for and characterization of specific high affinity binding sites for the gastric inhibitory polypeptide in pancreatic 6-cells. Endocrinology, 115:1324-1331. [267] Maletti, M . , Carlquist, M . , Portha, B., Kergoat, M . , Mutt, V., and Rosselin, G. (1986). Structural requirements for gastric inhibitory polypeptide (GIP) receptor binding and stimulation of insulin release. Peptides, 1:75-78. [268] Whitcomb, D. C , O'Dorisio, T. M . , Cataland, S., and Nishikawara, M . T. (1985). Theoretical basis for a new in vivo radioreceptor assay for polypeptide hormones. Am J Physiol, 249:E555-E560. [269] Whitcomb, D. C , O'Dorisio, T. M. , Cataland, S., Shetzline, M . , and Nishikawara, M . T. (1985). Identification of tissue insulin receptors: use of a unique in vivo radioreceptor assay. Am J Physiol, 249:E561-E567. 222 [270] Couvineau, A., Amiranoff, B., Vauclin-Jacques, N., and Laburthe, M. (1984). The GIP receptor on pancreatic beta cell tumor: molecular identification by covalent cross-linking. Biochem Biophys Res Commun, 122:283-288. [271] Amiranoff, B., Couvineau, A., Vauclin-Jacques, N., and Laburthe, M. (1986). Gastric inhibitory polypeptide receptor in hamster pancreatic beta cells. Direct cross-linking, solubilization and characterization as a glycoprotein. Eur J Biochem, 159:353-358. [272] Amiranoff, B., Vauclin-Jacques, N., and Laburthe, M. (1985). Interaction of gastric inhibitory polypeptide (GIP) with the insulin-secreting pancreatic beta-cell line In 111: characteristics of GIP binding sites. Life Sci, 36:807-813. [273] Gallwitz, B., Witt, M., Paetzold, G., Morys-Wortmann, C , Folsch, U. R., and Schmidt, W. E. (1995). Binding characteristic of N-terminal GIP/GLP-1 hybrid peptides. Endocrinol Metab, 2:39-46. [274] Gallwitz, B., Witt, M., Morys-Wortmann, C , Folsch, U. R., and Schmidt, W. E. (1996). GLP-1/GIP chimeric peptides define the structural requirements for specific ligand-receptor interaction of GLP-1. Regul Pept, 63:17-22. [275] Lynn, F. C , Thomson, S. A., Pospisilik, J. A., Ehses, J. A., Hinke, S. A., Pamir, N., Mcintosh, C. H. S., and Pederson, R. A. (2002). A novel pathway for regulation of glucose-dependent insulinotropic polypeptide (GIP) receptor expression in (3-cells. FASEB J, In press. [276] Wheeler, M. B., Gelling, R. W., Mcintosh, C. H., Georgiou, J., Brown, J. C , and Pederson, R. A. (1995). Functional expression of the rat pancreatic islet glucose-dependent insulinotropic polypeptide receptor: ligand binding and intracellular signaling properties. Endocrinology, 136:4629-4639. [277] Gremlich, S., Porret, A., Hani, E. H., Cherif, D., Vionnet, N., Froguel, P., and Thorens, B. (1995). Cloning, functional expression, and chromosomal localization of the human pancreatic islet glucose-dependent insulinotropic polypeptide receptor. Diabetes, 44:1202-1208. [278] Volz, A., Goke, R., Lankat-Buttgereit, B., Fehmann, H-C, Bode, H. P., and Goke, B. (1995). Molecular cloning, functional expression, and signal transduction of the GIP-receptor cloned from a human insulinoma [erratum appears in FEBS Lett 1996 Mar 4;381(3):271.]. FEBS Lett, 373:23-29. [279] Xiao, Q., Jeng, W., and Wheeler, M. B. (2000). Characterization of glucagon-like peptide-1 receptor-binding determinants. J Mol Endocrinol, 25:321-335. [280] Gelling, R. W., Wheeler, M. B., Xue, J., Gyomorey, S., Nian, C , Pederson, R. A., and Mcintosh, C. H. (1997). Localization of the domains involved in ligand binding and activation of the glucose-dependent insulinotropic polypeptide receptor. Endocrinology, 138:2640-2643. [281] Kubota, A., Yamada, Y., Hayami, T., Yasuda, K., Someya, Y., Ihara, Y., Kagimoto, S., Watanabe, R., Taminato, T., Tsuda, K., and Seino, Y. (1996). Identification of two 223 missense mutations in the GIP receptor gene: a functional study and association analysis with NIDDM: no evidence of association with Japanese NIDDM subjects. Diabetes, 45:1701-1705. [282] Yamada, Y., Hayami, T., Nakamura, K., Kaisaki, P. J., Someya, Y., Wang, C. Z., Seino, S., and Seino, Y. (1995). Human gastric inhibitory polypeptide receptor: cloning of the gene (GIPR) and cDNA. Genomics, 29:773-776. [283] Huypens, P., Ling, Z., Pipeleers, D., and Schuit, F. (2000). Glucagon receptors on human islet cells contribute to glucose competence of insulin release. Diabetologia, 43:1012-1019. [284] Boylan, M . O., Jepeal, L. I., and Wolfe, M . M . (1999). Structure of the rat glucose-dependent insulinotropic polypeptide receptor gene. Peptides, 20:219-228. [285] Frandsen, E. K., and Moody, A. J. (1980). Glucagon, GIP, VIP, and secretin activation of mouse islet adenylate cyclase [abstract]. Diabetologia, 19:27'4. [286] Gespach, C , Emami, S., and Rosselin, G. (1984). Gastric inhibitory peptide (GIP), pancreatic glucagon and vasoactive intestinal peptide (VIP) are cAMP-inducing hormones in the human gastric cancer cell line HGT-1. Homologous desensitization of VIP receptor activity. Biochem Biophys Res Commun, 120:641-649. [287] Gespach, C , Bataille, D., Dutrillaux, M . C , and Rosselin, G. (1982). The interaction of glucagon, gastric inhibitory peptide and somatostatin with cyclic AMP production systems present in rat gastric glands. Biochim Biophys Acta, 720:7-16. [288] Emami, S., Chastre, E., Bodere, H., Gespach, C , Bataille, D., and Rosselin,.G. (1986). Functional receptors for VIP, GIP, gIucagon-29 and -37 in the HGT-1 human gastric cancer cell line. Peptides, 1:121-127. [289] Rosselin, G. (1986). The receptors of the VIP family peptides (VIP, secretin, GRF, PHI, PHM, GIP, glucagon and oxyntomodulin). Specificities and identity. Peptides, 1:89-100. [290] Wheeler, M . B., Gelling, R. W., Hinke, S. A., Tu, B., Pederson, R. A., Lynn, F., Ehses, J., and Mcintosh, C. H. S. (1999). Characterization of the carboxyl-terminal domain of the rat glucose-dependent insulinotropic polypeptide (GIP) receptor. A role for serines 426 and 427 in regulating the rate of internalization. J Biol Chem, 274:24593-24601. [291] Tseng, C - C , and Zhang, X. Y. (1998). The cysteine of the cytoplasmic tail of glucose-dependent insulinotropic peptide receptor mediates its chronic desensitization and down-regulation. Molecular & Cellular Endocrinology, 139:179-186. [292] Tseng, C-C , and Lin, L. (1997). A point mutation in the glucose-dependent insulinotropic peptide receptor confers constitutive activity. Biochem Biophys Res Commun, 232:96-100. [293] Trumper, A., Trumper, K., Trusheim, H., Arnold, R., Goke, B., and Horsch, D. (2001). Glucose-dependent insulinotropic polypeptide is a growth factor for beta (INS-1) cells by pleiotropic signaling. Mol Endocrinol, 15:1559-1570. 224 [294] Kashima, Y. , Miki, T., Shibasaki, T., Ozaki, N. , Miyazaki, M . , Yano, H., and Seino, S. (2001). Critical role of cAMP-GEFII-Rim2 complex in incretin-potentiated insulin secretion. J Biol C/i6:14\\-144. [378] Krarup, T., Saurbrey, N. , Moody, A. J., Kiihl, C , and Madsbad, S. (1987). Effect of porcine gastric inhibitory polypeptide on beta-cell function in type I and type II diabetes mellitus. Metabolism, 36:677-682. [379] Lynn, F. C , Pamir, N. , Ng, E. H., Mcintosh, C. H., Kieffer, T. J., and Pederson, R. A. (2001). Defective glucose-dependent insulinotropic polypeptide receptor expression in diabetic fatty Zucker rats. Diabetes, 50:1004-1011. [380] Pauly, R. P., Rosche, F., Wermann, M . , Mcintosh, C. H. S., Pederson, R. A., and Demuth, H. U . (1996). Investigation of glucose-dependent insulinotropic polypeptide^ 1-42) and glucagon-like peptide-1-(7-36) degradation in vitro by dipeptidyl peptidase IV using matrix-assisted laser desorption/ionization-time of flight mass spectrometry. A novel kinetic approach. J Biol Chem, 271:23222-23229. [381] Pederson, R. A., White, H. A., Schlenzig, D., Pauly, R. P., Mcintosh, C. H. S., and Demuth, H-U. (1998). Improved glucose tolerance in Zucker fatty rats by oral administration of the dipeptidyl peptidase IV inhibitor isoleucine thiazolidide. Diabetes, 47:1253-1258. 231 [382] Pauly, R. P., Demuth, H-U., Rosche, F., Schmidt, J., White, H. A., Lynn, F., Mcintosh, C. FL, and Pederson, R. A. (1999). Improved glucose tolerance in rats treated with the dipeptidyl peptidase IV (CD26) inhibitor Ile-thiazolidide. Metabolism, 48:385-389. [383] Kesper, S., Rucha, J., Neye, H., Mazenot, C , and Verspohl, E. J. (1999). Galpha(i2)-mRNA and -protein regulation as a mechanism for heterologous sensitization and desensitization of insulin secretion. Cell Signal, 11:759-768. [384] Attramadal, H., Eikvar, L., and Hansson, V. (1988). Mechanisms of glucagon-induced homologous and heterologous desensitization of andenylate cyclase in membranes and whole Sertoli cells of the rat. Endocrinology, 123:1060-1068. [385] Mundell, S. J., and Kelly, E. (1998). The effect of inhibitors of receptor internalization on the desensitization and resensitization of three Gs-coupled receptor responses. Brit J Pharmacol, 125:1594-1600. [386] Tartakoff, A. M . (1983). Perturbation of vesicular traffic with the carboxylic ionophore monensin. Cell, 32:1026-1028. [387] Mollenhauer, H. H., Morre, D. J., and Rowe, L. D. (1990). Alteration of intracellular traffic by monensin; mechanism, specificity and relationship to toxicity. Biochim Biophys Acta, 1031:225-246. [388] Ammala, C , Ashcroft, F. M . , and Rorsman, P. (1993). Calcium-independent potentiation of insulin release by cyclic AMP in single (3-cells. Nature, 363:356-358. [389] Renstrom, E., Eliasson, L., and Rorsman, P. (1997). Protein kinase A-dependent and -independent stimulation of exocytosis by cAMP in mouse pancreatic B-cells. J Physiol, 502:105-118. [390] Zhou, J., and Egan, J. M . (1997). SNAP-25 is phosphorylated by glucose and GLP-1 in RIN 1046-38 cells. Biochem Biophys Res Commun, 238:297-300. [391] De Vries, L., Zheng, B., Fischer, T., Elenko, E., and Farquar, M . G. (2000). The regulator of G protein signaling family. Annu Rev Phamacol Toxicol, 40:235-271. [392] Zheng, B., Ma, Y. C , Ostrom, R. S., Lavoie, C , Gill , G. N. , Insel, P. A., Huang, X . Y., and Farquhar, M . G. (2001). RGS-PX1, a GAP for GalphaS and sorting nexin in vesicular trafficking. Science, 294:1939-1942. [393] Sinnarajah, S., Dessauer, C. W., Srikumar, D., Chen, J., Yuen, J., Yilma, S., Dennis, J. C , Morrison, E. E., Vodyanoy, V., and Kehrl, J. H. (2001). RGS2 regulates signal transduction in olfactory neurons by attenuating activation of adenylyl cyclase III. Nature, 409:1051-1055. [394] Houslay, M . D. (1991). Crosstalk': a pivotal role for protein kinase C in modulating relationships between signal transduction pathways. Eur J Biochem, 195:9-27. [395] Heuser, J. E., and Anderson, R. G. W. (1989). Hypertonic media inhibit receptor-mediated endocytosis by blocking clathrin-coated pit formation. J Cell Biol, 108:389-400. 232 [396] Milligan, G. (1999). Exploring the dynamics of regulation of G protein-coupled receptors using green fluorescent protein. Brit J Pharmacol, 128:501-509. [397] Ferguson, S. S. G. (1998). Using green fluorescent protein to understand the mechanisms of G-protein-coupled receptor regulation. Braz J Med Biol Res, 31:1471-1477. [398] Tarasova, N . I., Stauber, R. H., Choi, J. K., Hudson, E. A., Czerwinski, G., Miller, J. L., Pavlakis, G. N. , Michejda, C. J., and Wank, S. A. (1997). Visualization of G protein-coupled receptor trafficking with the aid of the green fluorescent protein: endocytosis and recycling of cholecystokinin receptor type A. J Biol Chem, 272:14817-14824. [399] Conway, B. R., Minor, L. K., Xu, J. Z., D1 Andrea, M . R., Ghosh, R. N. , and Demarest, K. T. (2001). Quantitative analysis of agonist-dependent parathyroid hormone receptor trafficking in whole cells using a functional green fluorescent protein conjugate. J Cell Physiol, 189:341-355. [400] Drmota, T., Gould, G. W., and Milligan, G. (1998). Real time visualization of agonist-mediated redistribution and internalization of a green fluorescent protein-tagged form of the thyrotropin-releasing hormone receptor. J Biol Chem, 272:24000-24008. [401] Awaji, T., Hirasawa, A., Kataoka, M . , Shinoura, H., Nakayama, Y., Sugawara, T., Izumi, S., and Tsujimoto, G. (1998). Real-time optical monitoring of ligand-mediated internalization of alb-adrenoceptor with green fluorescent protein. Mol Endocrinol, 12:1099-1111. [402] Salapatek, A. M . F., MacDonald, P. E., Gaisano, H. Y., and Wheeler, M . B. (1999). Mutations to the third cytoplasmic domain of the glucagon-like peptide 1 (GLP-1) receptor can functionally uncouple GLP-1-stimulated insulin secretion in HIT-T15 cells. Mol Endocrinol, 13:1305-1317. [403] Gaudriault, G., Nouel, D., Dal Farra, C , Beaudet, A., and Vincent, J-P. (1997): Receptor-induced internalization of selective peptidic p and 5 opioid ligands. J Biol Chem, 272:2880-2888. [404] Grady, E. F., Slice, L. W., Brant, W. O., Walsh, J. H., Payan, D. G., and Bunnett, N . W. (1995). Direct observation of endocytosis of gastrin releasing peptide and its receptor. J Biol Chem, 270:4603-4611. [405] Brown, J. C , and Pederson, R. A. (1970). Cleavage of a gastric inhibitory polypeptide with cyanogen bromide and the physiological action of the C-terminal fragment. J Physiol (Lond), 210:52P. [406] Schmidt, W. E., Siegel, E. G., Gallwitz, B., Kiimmel, H., Ebert, R., and Creutzfeldt, W. (1986). Characterization of the insulinotropic activity of fragments derived from gastric inhibitory polypeptide (abstract). Diabetologia, 29:591 A. [407] Yanaihara, N. , Mochizuki, T., Yanaihara, C , Sakura, N. , Hashimoto, T., Sakagami, M . , Kaneko, T., Kanetoh, A., and Brown, J. C. (1978). Synthesis of GIP, in S. R. Bloom, ed., Gut Hormones: New York, Churchill Livingstone, p. 271-276. 233 [408] Moroder, L., Hallett, A., Thamm, P., Wilschowitz, L., Brown, J. C , and Wiinsch, E. (1978). Studies on gastric inhibitory polypeptide: synthesis of the octatricontapeptide GIP(l-38) with full insulinotropic activity (abstract). Scand J Gastroenterol, 13 (Suppl. 49): 129. [409] Sandberg, E., Ahren, B., Tendler, D., Carlquist, M., and Efendic, S. (1986). Potentiation of glucose-induced insulin secretion in the perfused rat pancreas by porcine GIP (gastric inhibitory polypeptide), bovine GIP, and bovine GIP(l-39). Acta Physiol Scand, 127:323-326. [410] Pederson, R. A., Mochizuki, T., Yanaihara, C , Yanaihara, N., and Brown, J. C. (1990). Reduced somatostatinotropic effect of a GIP fragment (pG!P[l-30]-NH2) with insulinotropic activity. Digestion, 46:86. [411] Taminato, T., Seino, Y., Goto, Y., Inoue, Y., and Kadowaki, S. (1977). Synthetic gastric inhibitory polypeptide. Stimulatory effect on insulin and glucagon secretion in the rat. Diabetes, 26:480-484. [412] Yajima, H., Kai, Y., Ogawa, H., Kubota, M., and Mori, Y. (1977). Structure-activity relationships of gastrointestinal hormones: motilin, GIP, and [27-TYR]CCK-PZ. Gastroenterology, 72:793-796. [413] Gelling, R. W., Coy, D. H., Pederson, R. A., Wheeler, M. B., Hinke, S., Kwan, T., and Mcintosh, C. H. S. (1997). GIP6.,0amide contains the high affinity binding region of GIP and is a potent inhibitor of GIPI4, action in vitro. Regul Pept, 69:151-154. [414] Moody, A. J., Damm J0rgensen, K., and Thim, L. (1981). Structure-function relationships in porcine GIP (abstract). Diabetologia, 21:306. [415] Brown, J. C , Dahl, M., Kwauk, S., Mcintosh, C. H. S., Muller, M., Otte, S. C , and Pederson, R. A. (1981). Properties and actions of GIP, in S. R. Bloom, and J. M. Polak, eds., Gut Hormones: New York, Churchill Livingstone, p. 248-255. [416] Schmidt, W. E., Siegel, E. G., Ebert, R., and Creutzfeldt, W. (1986). N-terminal tyrosine-alanine is required for the insulin-releasing activity of glucose-dependent insulinotropic polypeptide (abstract). Eur J Clin Invest, 16: A9. [417] Deacon, C. F., Danielsen, P., Klarskov, L., Olesen, M., and Hoist, J. J. (2001). Dipeptidyl peptidase IV inhibition reduces the degradation and clearance of GIP and potentiates its insulinotropic and antihyperglycemic effects in anesthetized pigs. Diabetes, 50:1588-1597. [418] Pauly, R. P., Demuth, H-U., Rosche, F., Schmidt, J., White, H. A., Mcintosh, C. H. S., and Pederson, R. A. (1996). Inhibition of dipeptidyl peptidase IV (DP IV) in rat results in improved glucose tolerance (abstract). Regul Pept, 64:148. [419] Balkan, B., Kwasnik, L., Miserendino, R., Hoist, J. J., and Li , X. (1999). Inhibition of dipeptidyl peptidase IV with NVP-DPP728 increases plasma GLP-1 (7-36 amide) concentrations and improves oral glucose tolerance in obese Zucker rats. Diabetologia, 42:1324-1331. 234 [420] Hupe-Sodmann, K, McGregor, GP, Bridenbaugh, R, Goke, R, Goke, B, Thole, H, Zimmermann, B, and Voigt, K (1995). Characterisation of the processing by human neutral endopeptidase 24.11 of GLP-1 (7-36) amide and comparison of the substrate specificity of the enzyme for other glucagon-like peptides. Regul Pept, 58:149-156. [421] Agerberth, B., Boman, A., Andersson, M . , Jornvall, H., Mutt, V., and Boman, H. G. (1993). Isolation of three antibacterial peptides from pig intestine: gastric inhibitory polypeptide (7-42), diazepam-binding inhibitor (32-86) and a novel factor, peptide 3910. Eur J Biochem, 216:623-9. [422] S arson, D. L., Hayter, R. C , and Bloom, S. R. (1982). The pharmacokinetics of porcine glucose-dependent insulinotropic polypeptide (GIP) in man. Eur J Clin Invest, 12:457-461. [423] Sirinek, K. R., O'Dorisio, T. M . , Gaskill, H. V., and Levine, B. A. (1984). Chronic renal failure: effect of hemodialysis on gastrointestinal hormones. Am J Surg, 148:732-735. [424] Jorde, R., Burhol, P. G., Gunnes, P., and Schulz, T. B. (1981). Removal of IR-GIP by the kidneys in man, and the effect of acute nephrectomy on plasma GIP in rats. Scand J Gastroent, 16:469-471. [425] O'Dorisio, T. M . , Sirinek, K. R., Mazzaferri, E. L., and Cataland, S. (1977). Renal effects on serum gastric inhibitory polypeptide (GIP). Metabolism, 26:651-656. [426] 0rskov, C , Andreasen, J., and Hoist, J. J. (1992). All products of proglucagon are elevated in plasma of uremic patients. J Clin Endocrinol Metab, 74:379-384. [427] Mentlein, R. (1999). Dipeptidyl-peptidase IV (CD26)—role in the inactivation of regulatory peptides. Regul Pept, 85:9-24. [428] Chap, Z., O'Dorisio, T. M . , Cataland, S., and Field, J. B. (1987). Absence of hepatic extraction of gastric inhibitory polypeptide in conscious dogs. Dig Dis Sci, 32:280-284. [429] Kreymann, B., Ghatei, M . A., Williams, G., and Bloom, S. R. (1987). Glucagon-like peptide-1 (7-36): a physiological incretin in man. Lancet, 2:1300-1304. [430] Schjoldager, B. T., Mortensen, P. E., Christiansen, J., 0rskov, C , and Hoist, J. J. (1989). GLP-1 (glucagon-like peptide 1) and truncated GLP-1, fragments of human proglucagon, inhibit gastric acid secretion in humans. Dig Dis Sci, 34:703-708. [431] Wettergren, A., Schjoldager, B. T., Mortensen, P. E., Myhre, J., Christiansen, J., and Hoist, J. J. (1993). Truncated GLP-1 (proglucagon 78-107amide) inhibits gastric and pancreatic functions in man. Dig Dis Sci, 38:665-673. [432] 0rskov, C , Wettergren, A., and Hoist, J. J. (1993). Biological effects and metabolic rates of glucagon-like peptide-1 7-36 amide and glucagon-like peptide-1 7-37 in healthy subjects are indistinguishable. Diabetes, 42:658-661. [433] Deacon, C. F., Pridal, L., Klarskov, L., Olesen, M . , and Hoist, J. J. (1996). Glucagon-like peptide 1 undergoes differential tissue-specific metabolism in the anesthetized pig. Am J Physiol, 271:E458-E464. 235 [434] Pridal, L., Deacon, C. F., Kirk, O., Christensen, J. V., Carr, R. D., and Hoist, J. J. (1996). Glucagon-like peptide-1(7-37) has a larger volume of distribution than glucagon-like peptide-l(7-36)amide in dogs and is degraded more quickly in vitro by dog plasma. Eur J Drug Met Pharmacokinet, 21:51-59. [435] Ruiz-Grande, C , Alarcon, C , Alcantara, A., Castilla, C , Lopez Novoa, J. M . , Villanueva-Penacarrillo, M . L., and Valverde, I (1993). Renal catabolism of truncated glucagon-like peptide-1. Horm Metab Res, 25:612-616. [436] Jaspan, J., Polonsky, K., Lewis, M. , Pensler, J., Pugh, W., Moossa, A., and Rubinstein, A. (1981). Hepatic metabolism of glucagon in the dog: contribution of the liver to overall metabolic disposal of glucagon. Am J Physiol, 240:E233-E244. [437] Alford, F., Bloom, S., and Nabarro, J. (1976). Glucagon metabolism in man. Studies on the metabolic clearance rate and the plasma acute disappearance time of glucagon in normal and diabetic subjects. J Clin Endocrinol Metab, 42:830-838. [438] McDonald, J., Callahan, P., Zeitman, B., and Ellis, S. (1969). Inactivation and degradation of glucagon by dipeptidyl aminopeptidase I (cathepsin C) of rat liver. J Biol Chem, 244:6199-6208. [439] McDonald, J., Callahan, P., and Ellis, S. (1971). Polypeptide degradation by dipeptidyl aminopeptidase I (cathepsin C) and related peptidases, in A. Barrett, and J. Dingle, eds., Tissue Proteinases: Amsterdam, North-Holland Publishing, p. 69-107. [440] Rao, N . , Rao, G., and Hoidal, J. (1997). Human dipeptidyl peptidase I: gene characterization, localization and expression. J Biol Chem, 272:10260-10265. [441] Hoist, J. J. (1991). Degradation of glucagons, in J. Henriksen, ed., Degradation of bioactive substances: physiology and pathophysiology: Boca Raton, CRC Press, p. 167-180. [442] Hildebrandt, W., Blech, W., and Lohnert, K. (1991). Kinetic studies on hepatic handling of glucagon using the model of non recirculating perfused rat livers. Horm Metab Res, 23:410-413. [443] Lefebvre, P., and Luyckx, A. (1983). The renal handling of glucagon, in P. J. Lefebvre, ed., Handbook of Experimental Pharmacology 66/11: New York, Springer- Verlag, p. 389-396. [444] Peterson, D. R., Green, E. A., Oparil, S., and Hjelle, J. T. (1986). Transport and hydrolysis of glucagon in the proximal nephron. Am J Physiol, 251:F460-F467. [445] Talor, Z., Emmanouel, D. S., and Katz, A. I. (1983). Glucagon degradation by luminal and basolateral rabbit tubular membranes. Am J Physiol, 244:F297-F303. [446] Buckley, D. I., and Lundquist, P. (1992). Analysis of the degradation of insulinotropin [GLP-1 (7-37)] in human plasma and production of degradation resistant analogs (abstract). Regul Pept, 40:117. 236 [447] Deacon, C. F., Johnsen, A. FL, and Hoist, J. J. (1995). Degradation of glucagon-like peptide-1 by human plasma in vitro yields an N-terminally truncated peptide that is a major endogenous metabolite in vivo. J Clin Endocrinol Metab, 80:952-957. [448] Deacon, C. F., Nauck, M . A., Toft-Nielsen, M . , Pridal, L., Willms, B., and Hoist, J. J. (1995). Both subcutaneously and intravenously administered glucagon-like peptide I are rapidly degraded from the NH2-terminus in type II diabetic patients and in healthy subjects. Diabetes, 44:1126-1131. [449] Deacon, C. F., Hughes, T. E., and Hoist, J. J. (1998). Dipeptidyl peptidase IV inhibition potentiates the insulinotropic effect of glucagon-like peptides 1 in the anesthetized pig. Diabetes, 47:764-769. [450] Wettergren, A., Wojdemann, M. , and Hoist, J. J. (1998). The inhibitory effect of glucagon-like peptide-1 (7-36)amide on antral motility is antagonized by its N-terminally truncated primary metabolite GLP-1 (9-36)amide. Peptides, 19:877-882. [451] Knudsen, L. B., and Pridal, L. (1996). Glucagon-like peptide-1-(9-36) amide is a major metabolite of glucagon-like peptide-1-(7-36) amide after in vivo administration to dogs, and it acts as an antagonist on the pancreatic receptor. Eur J Pharmacol, 318:429-435. [452] Hansen, L., Deacon, C. F., Orskov, C , and Hoist, J. J. (1999). Glucagon-like peptide-l-(7-36)amide is transformed to glucagon-like peptide-l-(9-36)amide by dipeptidyl peptidase IV in the capillaries supplying the L cells of the porcine intestine. Endocrinology, 140:5356-5363. [453] Pospisilik, J. A., Hinke, S. A., Pederson, R. A., Hoffmann, T., Rosche, F., Schlenzig, D., Glund, K., Heiser, U. , Mcintosh, C. H. S., and Demuth, H-U. (2001). Metabolism of glucagon by dipeptidyl peptidase IV (CD26). Regul Pept, 96:133-141. [454] Marki, F. (1983). Stability of endogenous immunoreactive glucagon (IRG) in animal blood. Horm Metab Res, 15:307-308. [455] Hinke, S. A., Manhart, S., Pamir, N. , Demuth, H-U., Gelling, R. W., Pederson, R. A., and Mcintosh, C. H. S. (2001). Identification of a bioactive domain in the amino-terminus of glucose-dependent insulinotropic polypeptide (GIP). Biochim Biophys Acta, 1547:143-155. [456] Hinke, S. A., Gelling, R. W., Pederson, R. A., Manhart, S., Nian, C , Demuth, H-U., and Mcintosh, C. H. S. (2002). Dipeptidyl peptidase IV-resistant [D-Ala2]glucose-dependent insulinotropic polypeptide (GIP) improves glucose tolerance in normal and obese diabetic rats. Diabetes, 51:652-661. [457] Siegel, E. G., Scharf, G., Gallwitz, B., Mentlein, R., Morys-Wortmann, C , Folsch, U. R., and Schmidt, W. E. (1999). Comparison of the effect of native glucagon-like peptide 1 and dipeptidyl peptidase IV-resistant analogues on insulin release from rat pancreatic islets. Eur J Clin Invest, 29:610-614. [458] Deacon, C. F., Knudsen, L. B., Madsen, K , Wiberg, F. C , Jacobsen, O., and Hoist, J. J. (1998). Dipeptidyl peptidase IV resistant analogues of glucagon-like peptide-1 which 237 have extended metabolic stability and improved biological activity. Diabetologia, 41:271-278. [459] Ritzel, U., Leonhardt, U., Ottleben, M. , Ruhmann, A., Eckart, K., Spiess, J., and Ramadori, G. (1998). A synthetic glucagon-like peptide-1 analog with improved plasma stability. J Endocrinol, 159:93-102. [460] Unson, C. G., Gurzenda, E. M . , Iwasa, K., and Merrifield, R. (1989). Glucagon antagonists: contribution to binding and activity of the amino-terminal sequence 1-5, position 12, and the putative alpha-helical segment 19-27. J Biol Chem, 264:789-794. [461] Hoist, J. J. (1999). Glucagon-like peptide-1, a gastrointestinal hormone with pharmaceutical potential. Curr Med Chem, 6:1005-1017. [462] O'Harte, F. P. M . , Mooney, M . H., and Flatt, P. R. (1999). NH2-terminally modified gastric inhibitory polypeptide exhibits amino-peptidase resistance and enhanced antihyperglycemic activity. Diabetes, 48:758-765. [463] O'Harte, F. P. M . , Mooney, M . H., Kelly, C. M . N. , and Flatt, P. R. (2000). Improved glycaemic control in obese diabetic ob/ob mice using N-terminally modified gastric inhibitory polypeptide. J Endocrinol, 165:639-648. [464] Meneilly, G. S., Bryer-Ash, M. , and Elahi, D. (1993). The effect of glyburide on B-cell sensitivity to glucose-dependent insulinotropic polypeptide. Diabetes Care, 16:110-114. [465] Nauck, M . A., Busing, M. , Orskov, C , Siegel, E. G., Talartschik, J., Baartz, A., Baartz, T., Hopt, U. T., Becker, H. D., and Creutzfeldt, W. (1993). Preserved incretin effect in type 1 diabetic patients with end-stage nephropathy treated by combined heterotopic pancreas and kidney transplantation. Acta Diabetol, 30:39-45. [466] Creutzfeldt, W. O., Kleine, N. , Willms, B., Orskov, C , Hoist, J. J., and Nauck, M . A. (1996). Glucagonostatic actions and reduction of fasting hyperglycemia by exogenous glucagon-like peptide 1(7-36) amide in type I diabetic patients. Diabetes Care, 19:580-586. [467] Dupre, J., Behme, M . T., Hramiak, I. M. , McFarlane, P., Williamson, M . P., Zabel, P., and McDonald, T. J. (1995). Glucagon-like peptide I reduces postprandial glycemic excursions in IDDM. Diabetes, 44:626-630. [468] Blundell, T., and Wood, S. (1982). The conformation, flexibility, and dynamics of polypeptide hormones. Ann Rev Biochem, 51:123-154. [469] DaCambra, M . P., Yusta, B., Sumner-Smith, M . , Crivici, A., Drucker, D. J., and Brubaker, P. L. (2000). Structural determinants for activity of glucagon-like peptide-2. Biochemistry, 39:8888-8894. [470] Wray, V., Nokihara, K., and Naruse, S. (1997). Solution structure comparison of the VIP/PACAP family of peptides by NMR spectroscopy. Ann NY Acad Sci, 865:37-44. [471] Barden, J. A., and Cuthbertson, R. M . (1993). Stabilized NMR structure of human parathyroid hormone(l-34). Eur J Biochem, 215:315-321. 238 [472] Pellegrini, M, Royo, M, Rosenblatt, M, Chorev, M, and Mierke, DF (1998). Addressing the tertiary structure of human parathyroid hormone-(l-34). J Biol Chem, 273:10420-10427. [473] Barden, J. A., and Kemp, B. E. (1993). NMR solution structure of human parathyroid hormone(l-34). Biochemistry, 32:7126-7132. [474] Thornton, K, and Gorenstein, DG (1994). Structure of glucagon-like peptide(7-36) amide in a dodecylphosphocholine micelle as determined by 2D NMR. Biochemistry, 33:3532-3539. [475] Raufman, J-P. (1996). Bioactive peptides from lizard venoms. Regul Pept, 61:1-18. [476] Blankenfeldt, W., Nokihara, K., Naruse, S., Lessel, U., Schomburg, D., and Wray, V. (1996). NMR spectroscopic evidence that helodermin, unlike other members of the secretin/VIP family of peptides, is substantially structured in water. Biochemistry, 35:5955-5962. [477] Inooka, H., Ohtaki, T., Kitahara, O., Ikegami, T., Endo, S., Kitada, C , Ogi, K., Onda, H., Fujino, M., and Shirakawa, M. (2001). Conformation of a peptide ligand bound to its G-protein coupled receptor. Nat Struct Biol, 8:161-165. [478] Gascuel, O., and Golmard, J. L. (1988). A simple method for predicting the secondary structure of globular proteins: implications and accuracy. Comput Appl Biosci, 4:357-365. [479] Luck, MD, Carter, PH, and Gardella, TJ (1999). The (1-14) fragment of parathyroid hormone (PTH) activates intact and amino-terminally truncated PTH-1 receptors. Mol Endocrinol, 13:670-680. [480] Carter, P. H., Shimzu, M., Luck, M. D., and Gardella, T. J. (1999). The hydrophobic residues phenylalanine 184 and leucine 187 in the type-1 parathyroid hormone (PTH) receptor functionally interact with the amino-terminal portion of PTH-(l-34). J Biol Chem, 274:31955-31960. [481] Behar, V., Bisello, A., Bitan, G., Rosenblatt, M., and Chorev, M. (2000). Photoaffinity cross-linking identifies differences in the interactions of an agonist and an antagonist with the parathyroid hormone/parathyroid hormone-related protein receptor. J Biol Chem, 275:9-17. [482] Adams, A. E., Bisello, A., Chorev, M., Rosenblatt, M., and Suva, L. J. (1998). Arginine 186 in the extracellular N-terminal region of the human parathyroid hormone 1 receptor is essential for contact with position 13 of the hormone. Mol Endocrinol', 12:1673-1683. [483] Ohneda, A, Ohneda, K, Ohneda, M, Koizumi, F, Ohashi, S, Kawai, K, and Suzuki, S (1991). The structure-function relationship of GLP-1 related peptides in the endocrine function of the canine pancreas. Tohuku J Exp Med, 165:209-221. [484] Suzuki, S., Kawai, K., Ohashi, S., Mukai, H., and Yamashita, K. (1989). Comparison of the effects of various C-terminal and N-terminal fragment peptides of glucagon-like 239 peptide-1 on insulin and glucagon release from the isolated perfused rat pancreas. Endocrinology, 125:3109-3114. [485] Epand, R. M., and Grey, V. (1973). Conformational and biological properties of partial sequences of glucagon. Can J Physiol Pharmacol, 51:243-248. [486] Carrey, E. A., and Epand, R. M. (1983). Conformational and biological properties of glucagon fragments containing residues 1-17 and 19-29. Int J Peptide Protein Res, 22:362-370. [487] Fournier, A., Saunders, J. K., and St.-Pierre, S. (1984). Synthesis, conformational studies and biological activities of VIP and related fragments. Peptides, 5:169-177. [488] Gallwitz, B., Witt, M., Paetzold, G., Morys-Wortmann, C , Zimmermann, B., Eckart, K., Folsch, U. R., and Schmidt, W. E. (1994). Structure/activity characterization of glucagon-like peptide-1. Eur J Biochem, 225:1151-1156. [489] Adelhorst, K., Hedegaard, B. B., Knudsen, L. B., and Kirk, O. (1994). Structure-activity studies of glucagon-like peptide-1. J Biol Chem, 269:6275-6278. [490] Shimizu, N., Guo, J., and Gardella, T. J. (2001). Parathyroid hormone (PTH)-(1-14) and -(1-11) analogs conformational^ constrained by alpha-aminoisobutyric acid mediate full agonist responses via the juxtamembrane region of the PTH-1 receptor. J Biol Chem, 276:49003-49012. [491] Shimizu, M., Carter, P. H., Khatri, A., Potts, J. T., Jr., and Gardella, T. J. (2001). Enhanced activity in parathyroid hormone-(l-14) and -(1-11): novel peptides for probing ligand-receptor interactions. Endocrinology, 142:3068-3074. [492] Shimizu, M., Potts, J. T., Jr., and Gardella, T. J. (2000). Minimization of parathyroid hormone. Novel amino-terminal parathyroid hormone fragments with enhanced potency in activating the type-1 parathyroid hormone receptor. J Biol Chem, 275:21836-21843. [493] Carter, P. H., and Gardella, T. J. (2001). Zinc(II)-mediated enhancement of the agonist activity of histidine-substituted parathyroid hormone(l-14) analogues. Biochim Biophys Acta, 23:2-3. [494] Gershengorn, M. C , and Osman, R. (2001). Minireview: insights into G protein-coupled receptor function using molecular models. Endocrinology, 142:2-10. [495] Hjorth, S. A., and Schwartz, T. W. (1996). Glucagon and GLP-1 receptors: lessons from chimeric ligands and receptors. Acta Physiol Scand, 157:343-345. [496] Frohman, L. A., Downs, T. R., Heimer, E. P., and Felix, A. M. (1989). Dipeptidyl peptidase IV and trypsin-like enzymatic degradation of human growth hormone-releasing hormone in plasma. J Clin Invest, 83:1533-1540. [497] Heiman, M. L., Nekola, M. V., Murphy, W. A., Lance, V. A., and Coy, D. H. (1985). An extremely sensitive in vitro model for elucidating structure-activity relationships of growth hormone-releasing factor analogs. Endocrinology, 116:410-415. 240 [498] Lance, V. A., Murphy, W. A., Sueiras-Diaz, J., and Coy, D. H. (1984). Super-active analogs of growth hormone-releasing factor (l-29)-amide. Biochem Biophys Res Commun, 119:265-272. [499] Sueiras-Diaz, J., Lance, V. A., Murphy, W. A., and Coy, D. H. (1984). Structure-activity studies on the N-terminal region of glucagon. J Med Chem, 27:310-315. [500] Xiao, Q., Giguere, J., Parisien, M . , Jeng, W., St-Pierre, S. A., Brubaker, P. L., and Wheeler, M . B. (2001). Biological activities of glucagon-like peptide-1 analogues in vitro and in vivo. Biochemistry, 40:2860-2869. [501] Siegel, E. G., Gallwitz, B., Scharf, G., Mentlein, R., Morys-Wortmann, C , Folsch, U. R., Schrezenmeir, J., Drescher, K., and Schmidt, W. E. (1999). Biological activity of GLP-1 analogues with N-terminal modifications. Regul Pept, 79:93-102. [502] Joseph, J. W., Kalitsky, J., St-Pierre, S., and Brubaker, P. L. (2000). Oral delivery of glucagon-like peptide-1 in a modified polymer preparation normalizes basal glycaemia in diabetic db/db mice. Diabetologia, 43:1319-1328. [503] Frohman, L. A., Downs, T. R., Williams, T. C , Heimer, E. P., Pan, Y. C , and Felix, A. M . (1986). Rapid enzymatic degradation of growth hormone-releasing hormone by plasma in vitro and in vivo to a biologically inactive product cleaved at the NH 2 terminus. J Clin Invest, 78:906-913. [504] Bitar, K. G., Somogyvari-Vigh, A., and Coy, D. H. (1994). Cyclic lactam analogues of ovine pituitary adenylate cyclase activating polypeptide (PACAP): discovery of potent type II receptor antagonists. Peptides, 15:461-466. [505] Barbier, J. R., MacLean, S., Morley, P., Whitfield, J. F., and Willick, G. E. (2000). Structure and activities of constrained analogues of human parathyroid hormone and parathyroid hormone-related peptide: implications for receptor-activating conformations of hormones. Biochemistry, 39:14522-14530. [506] Ahn, J. M. , Gitu, P. M . , Medieros, M. , Swift, J. R., Trivedi, D., and Hruby, V. J. (2001). A new approach to search for the bioactive conformation of glucagon: positional cyclization scanning. J Med Chem, 44:3109-3116. [507] Parker, J. C , Andrews, K. M . , Rescek, D. M. , Massefski, W. Jr, Andrews, G. C , Contillo, L. G., Stevenson, R. W., Singleton, D. H., and Suleske, R. T. (1998). Structure-function analysis of a series of glucagon-like peptide-1 analogs. J Peptide Res, 52:398-409. [508] Amland, P. F., Jorde, R., Aanderud, S., Burhol, P. G., and Giercksky, K.-E. (1985). Effects of intravenously infused porcine GIP on serum insulin, plasma C-peptide, and pancreatic polypeptide in non-insulin-dependent diabetes. Scand J Gastroenterol, 20:315-320. [509] Jones, I. R., Owens, D. R., Luzio, S., and Hayes, T. M . (1989). Glucose dependent insulinotropic polypeptide (GIP) infused intravenously is insulinotropic in the fasting state in type 2 (non-insulin dependent) diabetes mellitus. Horm Metabol Res, 21:23-26. 241 [510] Ritzel, R., 0rskov, C, Hoist, J. J., and Nauck, M . A. (1995). Pharmacokinetic, insulinotropic, and glucagonostatic properties of GLP-1 [7-36 amide] after subcutaneous injection in healthy volunteers: dose-resopnse relationships. Diabetologia, 38:720-725. [511] Meier, J. J., Nauck, M . A., Schmidt, W. E., and Gallwitz, B. (2002). Gastric inhibitory polypeptide: the neglected incretin revisited. Regul Pept, 107:1-13. [512] Grondin, G., Hooper, N . , and LeBel, D. (1999). Specific localization of membrane dipeptidase and dipeptidyl peptidase IV in secretion granules of two different pancreatic islet cells. J Histochem Cytochem, 47:489-497. [513] Pavoine, C , Brechler, V., Kervran, A., Blache, P., Le-Nguyen, D., Laurent, S., Bataille, D., and Pecker, F. (1991). Miniglucagon [glucagon-(19-29)] is a component of the positive inotropic effect of glucagon. Am J Physiol, 260:C993-C999. [514] Dalle, S., Smith, P., Blache, P., Le-Nguyen, D., Le Brigand, L., Bergeron, F., Ashcroft, F. M . , and Bataille, D. (1999). Miniglucagon (glucagon 19-29), a potent and efficient inhibitor of secretagogue-induced insulin release through a Ca2+ pathway. J Biol Chem, 274:10869-10876. [515] Frandsen, E. K., Gr0nvald, F. C , Heding, L. G., Johansen, N. L. , Lundt, B. F., Moody, A. J., Markussen, J., and Volund, A. (1981). Glucagon: structure-function relationships investigated by sequence deletions. Hoppe-Seyler's Z Physiol Chem, 362:665-678. [516] Smith, R. A., Sisk, R., Lockhart, P., Mathewes, S., Gilbert, T., Walker, K., and Piggot, J. (1993). Isolation of glucagon antagonists by random molecular mutagenesis and screening. Mol Pharmacol, 43:741-748. [517] Robberecht, P., Damien, C , Moroder, L., Coy, D. H., Wunsch, E., and Cristophe, J. (1988). Receptor occupancy and adenylate cyclase activation in rat liver and heart membranes by 10 glucagon analogs modified in position 2, 3, 4, 25, 27, and/or 29. Regul Pept, 21:117-128. [518] Unson, C. G., and Merrifield, R. B. (1994). Identification of an essential serine residue in glucagon: implication for an active site triad. Proc Natl Acad Sci USA, 91:454-458. [519] Marguet, D, Baggio, L, Kobayashi, T, Bernard, A - M , Pierres, M , Nielsen, PF, Ribel, U, Watanabe, T, Drucker, DJ, and Wagtmann, N (2000). Enhanced insulin secretion and improved glucose tolerance in mice lacking CD26. Proc Natl Acad Sci USA, 97:6874-6879. [520] Creutzfeldt, W. (2001). The entero-insular axis in type 2 diabetes — incretins as therpeutic agents. Exp Clin Endocrinol Diabetes, 109 (Suppl. 2):S288-S300. [521] Hoist, J. J., and Deacon, C. F. (1998). Inhibition of the activity of dipeptidylpeptidase IV as a treatment for type 2 diabetes. Diabetes, 47:1663-1670. [522] White, C. M . (1999). A review of potential cardiovascular uses of intravenous glucagon administration. J Clin Pharmacol, 39:442-447. 242 [523] Pollack, C. V., Jr. (1993). Utility of glucagon in the emergency department. J Emerg Med, 11:195-205. [524] Fan, G., Shumay, E., Wang, H., and Malbon, C. C. (2001). The scaffold protein gravin (cAMP-dependent protein kinase-anchoring protein 250) binds the beta 2-adrenergic receptor via the receptor cytoplasmic Arg-329 to Leu-413 domain and provides a mobile scaffold during desensitization. J Biol Chem, 276:24005-24014. [525] Pierce, K. L„ Tohgo, A., Ahn, S., Field, M. E., Luttrell, L. M. , and Lefkowitz, R. J. (2001). Epidermal growth factor (EGF) receptor-dependent ERK activation by G protein-coupled receptors: a co-culture system for identifying intermediates upstream and downstream of heparin-binding EGF shedding. J Biol Chem, 276:23155-23160. [526] Jordan, B. A., Trapaidze, N . , Gomes, I., Nivarthi, R., and Devi, L. A. (2001). Oligomerization of opioid receptors with beta 2-adrenergic receptors: a role in trafficking and mitogen-activated protein kinase activation. Proc Natl Acad Sci USA, 98:343-348. [527] Tang, H., Nishishita, T., Fitzgerald, T., Landon, E. J., and fnagami, T. (2000). Inhibition of ATI receptor internalization by concanavalin A blocks angiotensin fl-induced ERK activation in vascular smooth muscle cells. Involvement of epidermal growth factor receptor proteolysis but not ATI receptor internalization. J Biol Chem, 275:13420-13426. [528] Elorza, A., Sarnago, S., and Mayor, F., Jr. (2000). Agonist-dependent modulation of G protein-coupled receptor kinase 2 by mitogen-activated protein kinases. Mol Pharmacol, 57:778-783. [529] Maudsley, S., Pierce, K. L., Zamah, A. M. , Miller, W. E., Ahn, S., Daaka, Y., Lefkowitz, R. J., and Luttrell, L. M . (2000). The beta(2)-adrenergic receptor mediates extracellular signal-regulated kinase activation via assembly of a multi-receptor complex with the epidermal growth factor receptor. J Biol Chem, 275:9572-9580. [530] Pierce, K. L., Maudsley, S., Daaka, Y., Luttrell, L. M. , and Lefkowitz, R. J. (2000). Role of endocytosis in the activation of the extracellular signal-regulated kinase cascade by sequestering and nonsequestering G protein-coupled receptors. Proc Natl Acad Sci USA, 97:1489-1494. [531] Pitcher, J. A., Tesmer, J. J., Freeman, J. L., Capel, W. D., Stone, W. C , and Lefkowitz, R. J. (1999). Feedback inhibition of G protein-coupled receptor kinase 2 (GRK2) activity by extracellular signal-regulated kinases. J Biol Chem, 274:34531-34534. [532] Schramm, N. L., and Limbird, L. E. (1999). Stimulation of mitogen-activated protein kinase by G protein-coupled alpha(2)-adrenergic receptors does not require agonist-elicited endocytosis. J Biol Chem, 274:24935-24940. [533] Lin, F. T., Miller, W. E., Luttrell, L. M . , and Lefkowitz, R. J. (1999). Feedback regulation of (3-arrestinl function by extracellular signal-regulated kinases. J Biol Chem, 274:15971-15974. 243 [534] Budd, D. C , Rae, A., and Tobin, A. B. (1999). Activation of the mitogen-activated protein kinase pathway by a Gq/11-coupled muscarinic receptor is independent of receptor internalization. J Biol Chem, 274:12355-12360. [535] Daaka, Y., Luttrell, L. M . , Ahn, S., Delia Rocca, G. J., Ferguson, S. S., Caron, M . G., and Lefkowitz, R. J. (1998). Essential role for G protein-coupled receptor endocytosis in the activation of mitogen-activated protein kinase. J Biol Chem, 273:685-688. [536] Luo, X. , Zeng, W„ Xu, X. , Popov, S., Davignon, I., Wilkie, T. M . , Mumby, S. M . , and Muallem, S. (1999). Alternate coupling of receptors to Gs and Gi in pancreatic and submandibular gland cells. J Biol Chem, 274:17684-17690. [537] Daaka, Y., Luttrell, L. M. , and Lefkowitz, R. J. (1997). Switching of the coupling of the pVadrenergic receptor to different G proteins by protein kinase A. Nature, 390:88-91. [538] Foord, S. M . , and Marshall, F. Ff. (1999). RAMPs: accessory proteins for seven transmembrane domain receptors. Trends Pharmacol Sci, 20:184-187. [539] Sexton, P. M . , Albiston, A., Morfis, M . , and Tilakaratne, N . (2001). Receptor activity modifying proteins. Cell Signal, 13:73-83. [540] Lindsay, J. R., Au, S. T. B., Kelley, C. M . , O'Harte, F. P., Flatt, P. R., and Bell, P. M . (2002). A novel amino-terminally glycated analogue of glucose-dependent insulinotropic polypeptide (GIP), with prolonged insulinotropic activity in type 2 diabetes mellitus (abstract 1394-P). Diabetes, 51 (Suppl. 2):A341. 244 f "@en ; edm:hasType "Thesis/Dissertation"@en ; vivo:dateIssued "2002-11"@en ; edm:isShownAt "10.14288/1.0091443"@en ; dcterms:language "eng"@en ; ns0:degreeDiscipline "Physiology"@en ; edm:provider "Vancouver : University of British Columbia Library"@en ; dcterms:publisher "University of British Columbia"@en ; dcterms:rights "For non-commercial purposes only, such as research, private study and education. Additional conditions apply, see Terms of Use https://open.library.ubc.ca/terms_of_use."@en ; ns0:scholarLevel "Graduate"@en ; dcterms:title "Modulation of insulinotropic hormone bioactivity with a focus on clucose-dependent insulinotropic polipeptide (GIP) and its receptor"@en ; dcterms:type "Text"@en ; ns0:identifierURI "http://hdl.handle.net/2429/14867"@en .