AU2005217198A1

AU2005217198A1 - Peptide analogues of GIP for treatment of diabetes, insulin resistance and obesity

Info

Publication number: AU2005217198A1
Application number: AU2005217198A
Authority: AU
Inventors: Peter Raymond Flatt; Victor A. Gault; Nigel Irwin; Finbarr Paul Mary O'harte
Original assignee: Ulster University
Current assignee: Ulster University
Priority date: 2004-02-25
Filing date: 2005-02-25
Publication date: 2005-09-09
Also published as: JP2008500280A; CA2557151A1; GB0404124D0; EP1730188A2; WO2005082928A2; CA2557151C; WO2005082928A3

Description

WO 2005/082928 PCT/GB2005/000710 PEPTIDE ANALOGUES OF GIP FOR TREATMENT OF DIABETES, INSULIN RESISTANCE AND OBESITY CROSS-REFERENCE TO RELATED APPLICATIONS 5 This application claims the benefit of U.K. Application No. GB 0404124.0, filed on 25 February 2004. The entire teachings of the above application are incorporated herein by reference. FIELD OF THE INVENTION 10 The present invention relates to the release of insulin and the control of blood glucose concentration. More particularly the invention relates to antagonists of gastric inhibitory peptide (GIP) as pharmaceutical preparations for treatment of type 2 diabetes. 15 BACKGROUND Obesity and diabetes are predicted to reach epidemic proportions throughout the world in the next 20 years and current treatments do not restore normal insulin sensitivity or glucose homeostasis, therein resulting in debilitating diabetic complications and premature death. 20 Gastric inhibitory polypeptide (GIP) and glucagon-like peptide- 1 (7-36)amide (truncated GLP-1; tGLP-1) are two important insulin-releasing hormones secreted from endocrine cells in the intestinal tract in response to feeding. Together with autonomic nerves they play a vital supporting role to the pancreatic islets in the control of blood glucose homeostasis and nutrient metabolism. 25 GIP is released from intestinal endocrine K-cells into the bloodstream following ingestion of carbohydrate, protein and particularly fat (Meier, J.J. et al., 2002, Regul. Pept. 107:1-13). GIP was initially discovered through its ability to inhibit gastric acid secretion (Brown, J.C. et al. 1969, Can. J. Physiol. Pharmacol 47:113-114) but its major physiological role is now generally believed to be that of an 30 incretin hormone that targets pancreatic islets to enhance insulin secretion and help reduce postprandial hyperglycemia (Creutzfeldt, W., 2001, Exp. Clin. Endocrinol. Diabetes 109:S288-S303). GIP acts through binding to specific G-protein coupled GIP receptors located on pancreatic beta-cells (Wheeler, M.B. et aL, 1995, Endocrinology 136:4629-4639). Like its sister incretin hormone, glucagon-like WO 2005/082928 PCT/GB2005/000710 peptide-1 (GLP-1), this ability to stimulate insulin secretion plus other potentially beneficial actions on pancreatic beta-cell growth and differentiation have led to much interest in using GIP or GLP-1 receptor agonists in the treatment of type 2 diabetes (Creutzfeldt, W., 2001, Exp. Clin. Endocrinol. Diabetes 109:S288-S303; Holz, G.G. 5 et al., 2003, Curr. Med. Chemn. 10:2471-2483). Since GIP functions as a potent and natural stimulator of insulin secretion released from the intestine by feeding, it is widely expected that antagonists opposing GIP action will block the insulin-releasing actions of GIP and impair both oral glucose tolerance and the glycemic response to nutrient ingestion. In fact, all studies 10 published to date indicate that GIP is a key physiological component of the enteroinsular axis and that functional ablation of GIP leads to impaired glucose homeostasis moving the metabolic characteristic towards a type 2 diabetes phenotype (Gault, V.A. et al., 2002, Biochemn. Biophys. Res. Common. 290:1420-1426). Dipeptidyl peptidase IV (DPP IV; EC 3.4.14.5) has been identified as a key 15 enzyme responsible for inactivation of GIP and tGLP- 1 in serum. This occurs through the rapid removal of the N-terminal dipeptides Tyr'-Ala 2 and His7-Alas giving rise to the main metabolites GIP(3-42) and GLP-l (9-36)amide, respectively. These truncated peptides are reported to lack biological activity or to even serve as antagonists at GIP or tGLP-1 receptors. The resulting biological half-lives of these 20 incretin hormones in vivo are therefore very short, estimated to be no longer than 5 minutes. DPP IV is completely inhibited in serum by the addition of diprotin A (DPA, 0.1 mmol/1). In situations of normal glucose regulation and pancreatic B-cell sensitivity, this short duration of action is advantageous in facilitating momentary adjustments to 25 homeostatic control. However, the current goal of a possible therapeutic role of incretin hormones, particularly tGLP- I in non-insulin dependent diabetes (NIDDM) therapy is frustrated by a number of factors in addition to finding a convenient route of administration. Most notable of these are rapid peptide degradation and rapid absorption (peak concentrations are reached in 20 minutes) and the resulting need for 30 both high dosage and precise timing with meals. Recent therapeutic strategies have focused on precipitated preparations to delay peptide absorption and inhibition of GLP-1 degradation using specific inhibitors of DPP IV. A possible therapeutic role is also suggested by the observation that a specific inhibitor of DPP IV, isoleucine -2- WO 2005/082928 PCT/GB2005/000710 thiazolidide, lowered blood glucose and enhanced insulin secretion in glucose-treated diabetic obese Zucker rats presumably by protecting against catabolism of the incretin hormones tGLP- 1 and GIP. Studies have indicated that tGLP-1 infusion restores pancreatic B-cell 5 sensitivity, insulin secretory oscillations and improved glycemic control in various groups of patients with impaired glucose tolerance (IGT) or NIDDM. Longer term studies also show significant benefits of tGLP-1 injections in NIDDM and possibly IDDM therapy, providing a major incentive to develop an orally effective or long acting tGLP-1 analogue. Several attempts have been made to produce structurally 10 modified analogues of tGLP-1 which are resistant to DPP IV degradation. A significant extension of serum half-life is observed with His 7 -glucitol tGLP-1 and tGLP-1 analogues substituted at position 8 with Gly, Aib (amino isobutyric acid), Ser or Thr. However, these structural modifications seem to impair receptor binding and insulinotrophic activity thereby compromising part of the benefits of protection from 15 proteolytic degradation. In recent studies using His 7 glucitol tGLP-1, resistance to DPP IV and serum degradation was accompanied by severe loss of insulin releasing activity. GIP shares not only the same degradation pathway as tGLP-1 but many similar physiological actions, including stimulation of insulin and somatostatin 20 secretion, and the enhancement of glucose disposal. These actions are viewed as key aspects in the antihyperglycemic properties of tGLP-1, and there is therefore good expectation that GIP may have similar potential as NIDDM therapy. Indeed, compensation by GIP is held to explain the modest disturbances of glucose homeostasis observed in tGLP- I knockout mice. Apart from early studies, the anti 25 diabetic potential of GIP has not been explored and tGLP- 1 may seem more attractive since it is viewed by some as a more potent insulin secretagogue when infused at so called physiological concentrations estimated by radioimmunoassay (RIA). There is therefore a need for a diabetes treatment that includes an analogue of GIP which can cause release of insulin, yet also be resistant to rapid degradation by 30 DPP IV.

WO 2005/082928 PCT/GB2005/000710 SUMMARY OF THE INVENTION Disclosed herein are GIP antagonist peptides which are resistant to rapid degradation by DPP IV. The invention includes a peptide analogue of GIP(l -42) (SEQ ID NO:1), 5 which includes at least 12 amino acid residues from the N-terminal end of GIP(3-42). The invention also includes a peptide analogue of GIP(1-42) (SEQ ID NO: 1), which includes at least 12 amino acid residues from the N-terminal end of GIP(1-42) and having an amino acid substitution at Glu 3 . The amino acid substituted at Glu 3 can be selected from the group consisting 10 of: proline, hydroxyproline, lysine, tyrosine, phenylalanine and tryptophan. Specifically, a proline can be substituted for Glu 3 . The peptide analogue can further include modification by fatty acid addition at an epsilon amino group of at least one 16 37 lysine residue. The lysine residue can be Lys , or Lys The peptide analogue of GIP(1-42) (SEQ ID NO:1) can include at least 12 15 amino acid residues from the N-terminal end of GIP(1-42), and an amino acid modification at amino acid residues 1, 2 or 3. The N-terminal amino acid residue can be acetylated. It can further comprising modification by fatty acid addition at an epsilon amino group of at least one lysine residue. The modification can be the linking of, e.g., a C-8, a C-10, a C-12, a C-14, a C-16, a C-18 or a C-20 palmitate 20 group to the epsilon amino group of a lysine residue. The lysine residue can be Lys 1, or Lys 37 . The invention also includes a peptide analogue of GIP(1-42) (SEQ ID NO:1), wherein the analogue comprises a base peptide consisting of one of the following: GIP(1-12), GIP(1-13), GIP(1-14), GIP(1-15), GIP(1-16), GIP(1-17), GIP(l-18), 25 GIP(1-19), GIP(1-20), GIP(1-21), GIP(1-22), GIP(1-23), GIP(1-24), GIP(1-25), GIP(1 -26), GIP( 1-27), GIP(1 -28), GIP(l1-29), GIP(1 -30), GIP(1 -31), GIP( 1-32), GIP(1-33), GIP(1-34), GIP(1-35), GIP(1-36), GIP(1-37), GIP(1-38), GIP(1-39), GIP(1-40), GIP(1-41) and GIP(1-42), where the base peptide possesses one or more of the following modifications: (1) an amino acid substitution at Glu 3 ; (2) a 30 modification by fatty acid addition at an epsilon amino group of at least one lysine residue; and (3) a modification by N-terminal acetylation. Such a peptide analogue can have a proline substituted for Glu. It can also have a modification in the form of a C- 16 palmitate group linked to the epsilon amino group of a lysine residue. The -4- WO 2005/082928 PCT/GB2005/000710 modification can be the linking of, e.g., a C-8, a C-10, a C-12, a C-14, a C-16, a C-18 or a C-20 palmitate group to the epsilon amino group of a lysine residue. The lysine residue can be Lys 16 , or Lys 37 The invention further includes a peptide analogue of GIP(1-42) (SEQ ID 5 NO:1), comprising at least 12 amino acid residues from the N-terminal end of GIP(3 42), wherein the peptide analogue is resistant to degradation by enzyme DPP IV when compared to naturally-occurring GIP. Also included is a peptide analogue of GIP(1-42) (SEQ ID NO:1), comprising at least 12 amino acid residues from the N-terminal end of GIP(1 -42) and having an 10 amino acid substitution at Glu 3 , wherein the peptide analogue is resistant to degradation by enzyme DPP IV when compared to naturally-occurring GIP. In addition, the invention includes a peptide analogue of GIP(1-42) (SEQ ID NO: 1), comprising at least 12 amino acid residues from the N-terminal end of GIP(3 42), wherein the peptide analogue modulates insulin secretion. 15 The invention also includes a peptide analogue of GIP(1-42) (SEQ ID NO:1), comprising at least 12 amino acid residues from the N-terminal end of GIP(l -42) and having an amino acid substitution at Glu 3 , wherein the peptide analogue modulates insulin secretion. The invention also includes use of any of the analogues in the preparation of a 20 medicament for the treatment of obesity, insulin resistance, insulin resistant metabolic syndrome (Syndrome X) or type 2 diabetes. The invention also includes a pharmaceutical composition including the peptide analogues. The pharmaceutical composition can further comprise a pharmaceutically acceptable carrier. The peptide analogue can be in the form of a 25 pharmaceutically acceptable salt, or a pharmaceutically acceptable acid addition salt. In a further aspect, the invention includes a method of treating insulin resistance, obesity, or type 2 diabetes, where the method comprises administering to a mammal in need of such treatment a therapeutically effective amount of the pharmaceutical composition. 30 According to the present invention there is provided an effective peptide analogue of the biologically active GIP(1-42) which has improved characteristics for treatment of Type 2 diabetes wherein the analogue comprises at least 15 amino acid -5- WO 2005/082928 PCT/GB2005/000710 residues from the N terminus of GIP(1-42) and has at least one amino acid substitution or modification at position 1-3 and not including TyrI glucitol GIP(1-42). The structures of human and porcine GIP(1 -42) are shown below. The porcine peptide differs by.just two amino acid substitutions at positions 18 and 34. 5 The analogue may include modification by fatty acid addition at an epsilon amino group of at least one lysine residue. The invention includes Tyri glucitol GIP(1-42) having fatty acid addition at an epsilon amino group of at least one lysine residue. Analogues of GIP(1-42) may have an enhanced capacity to stimulate insulin 10 secretion, enhance glucose disposal, delay glucose absorption or may exhibit enhanced stability in plasma as compared to native GIP. They also may have enhanced resistance to degradation. Any of these properties will enhance the potency of the analogue as a therapeutic agent. 15 Analogues having D-amino acid substitutions in the 1, 2 and 3 positions and/or N-glycated, N-alkylated, N-acetylated or N-acylated amino acids in the 1 position are resistant to degradation in vivo. Various amino acid substitutions at second and third amino terminal residues are included, such as GIP(1-42)Gly 2 , GIP(1-42)Ser 2 , GIP(1-42)Abu 2 , GIP(1-42)Aib 2 , 20 GIP(l-42)D-Ala 2 , GIP(1-42)Sar 2 , and GIP(1-42)Pro 3 . Amino-terminally modified GIP analogues include N-glycated GIP(1-42), N alkylated GIP(1-42), N-acetylated GIP(1-42), N-acetyl-GIP(1-42) and N-isopropyl GIP(1-42). Other stabilized analogues include those with a peptide isostere bond between 25 amino terminal residues at position 2 and 3. These analogues may be resistant to the plasma enzyme dipeptidyl-peptidase IV (DPP IV) which is largely responsible for inactivation of GIP by removal of the amino-terminal dipeptide Tyr'-Ala 2 . In particular embodiments, the invention provides a peptide which is more potent than human or porcine GIP in moderating blood glucose excursions, said 30 peptide consisting of GIP(1-42) or N-terminal fragments of GIP(1-42) consisting of up to between 15 to 30 amino acid residues from the N-terminus (i.e., GIP(1-15) GIP( 1-3)) with one or more modifications selected from the group consisting of: (a) substitution of Ala 2 by Gly; -6- WO 2005/082928 PCT/GB2005/000710 (b) substitution of Ala 2 by Ser; (c) substitution of Ala 2 by Abu; (d) substitution of Ala 2 by Aib; (e) substitution of Ala 2 by D-Ala; 5 (f) substitution of Ala 2 by Sar sarcosinee); (g) substitution of Glu 3 by Pro; (h) modification of Tyr' by acetylation; (i) modification of Tyr' by acylation; (j) modification of Tyr' by alkylation; 10 (k) modification of Tyr' by glycation; (1) conversion of Ala 2 - Glu 3 bond to a psi [CH 2 NH] bond; (in) conversion of AIa-Glu 3 bond to a stable peptide isotere bond; and (n) (n-isopropyl-H) IGIP. The invention also provides the use of Tyr'-glucitol GIP in the preparation of 15 a medical ment for the treatment of diabetes. The invention further provides improved pharmaceutical compositions including analogues of GIP with improved pharmacological properties. Other possible analogues include certain commonly encountered aino acids, which are not encoded by the genetic code, for example, beta-alanine (beta-ala), or 20 other omega-amino acids, such as 3-amino propionic, 4-amino butyric and so forth, ornithine (0m), citrulline (Cit), homoarginine (Har), t-butylalanine (t-BuA), t buty(glycine (t-BftG), N-methylisoleucine (N-Mele), phenylglycine (Phg), and cyclohexylalanine (Cha), norleucine (Nle), cysteic acid (Cya) and methionine sulfoxide (MSO), substitution of the D form of a neutral or acidic amino acid or the D 25 form of tyrosine for tyrosine. According to the present invention there is also provided a pharmaceutical composition useful in the treatment of diabetes type wt which comprises an effective amount of the peptide as described herein, in admixture with a pharmaceutically acceptable excipient. 30 The invention also provides a method of N-terminally modifb- ing GIP or analogues thereof the method comprising the steps of synthesizing the peptide from the C terminal to the penultimate N terminal amino acid, adding tyrosine to a bubbler system as a P-moc protected Tyr(tBu)-Wang resin, deprotecting the N-terminus of the -7- WO 2005/082928 PCT/GB2005/000710 tyrosine and reacting with the modifying agent, allowing the reaction to proceed to completion, cleaving the modified tyrosine from the Wang resin and adding the modified tyrosine to the peptide synthesis reaction. Preferably the agent is glucose, acetic anhydride or pyroglutamic acid. 5 The invention will now be demonstrated with reference to the following non limiting examples and the accompanying figures wherein. BRIEF DESCRIPTION OF THE DRAWINGS Fig. 1 a illustrates degradation of GIP by DPP IV. 10 Fig. lb illustrates degradation of GIP and Tyr' glucitol GIP by DPP IV. Fig. 2a illustrates degradation of GIP human plasma. Fig. 2b illustrates degradation of GIP and Tyr' glucitol GIP by human plasma. Fig. 3 illustrates electrospray ionization mass spectrometry of GIP, Tyr' glucitol GIP and the major degradation fragment GIP(3-42). 15 Fig. 4 shows the effects of GIP and glycated GIP on plasma glucose homeostasis. Fig. 5 shows effects of GIP on plasma insulin responses. Fig. 6 illustrates DPP-IV degradation of GIP (1-42). Fig. 7 illustrates DPP-IV degradation of GIP (Abu 2 ). 20 Fig. 8 illustrates DPP-IV degradation of GIP (Sar 2 ). Fig. 9 illustrates DPP-IV degradation of GIP (Ser 2 ). Fig. 10 illustrates DPP-IV degradation of N-Acetyl-GIP. Fig. 1I illustrates DPP-IV degradation of glycated GIP. Fig. 12 illustrates human plasma degradation of GIP. 25 Fig. 13 illustrates human plasma degradation of GIP (Abu 2 ). Fig. 14 illustrates human plasma degradation of GIP (Sar 2 ). Fig. 15 illustrates human plasma degradation of GIP (Ser 2 ). Fig. 16 illustrates human plasma degradation of glycated GIP. Fig. 17 shows the effects of various concentrations of GIP 1-42 and GIP 30 (Abu 2 ) on insulin release from BRIN-BDI 1 cells incubated at 5.6mM glucose. Fig. 18 shows the effects of various concentrations of GIP 1-42 and GIP (Abu 2 ) on insulin release from BRIN-BDI I cells incubated at 16.7mM glucose. -8- WO 2005/082928 PCT/GB2005/000710 Fig. 19 shows the effects of various concentrations of GIP 1-42 and GIP (Sar 2 ) on insulin release from BRIN-BDl 1 cells incubated at 5.6mM glucose. Fig. 20 shows the effects of various concentrations of GIP 1-42 and GIP (Sar 2 ) on insulin release from BRIN-BD 11 cells incubated at 16.7mM glucose. 5 Fig. 21 shows the effects of various concentrations of GIP 1-42 and GIP (Ser 2 ) on insulin release from BRIN-BD 11 cells incubated at 5.6mM glucose. Fig. 22 shows the effects of various concentrations of GIP 1-42 and GIP (Ser 2 ) on insulin release from BRIN-BD 11 cells incubated at 16.7mM glucose. Fig. 23 shows the effects of various concentrations of GIP 1-42 and N-Acetyl 10 GIP 1-42 on insulin release from BRIN-BD1 I cells incubated at 5.6mM glucose. Fig. 24 shows the effects of various concentrations of GIP 1-42 and N-Acetyl GIP 1-42 on insulin release from BRIN-BD 11 cells incubated at 16.7mM glucose. Fig. 25 shows the effects of various concentrations of GIP 1-42 and glycated GIP 1-42 on insulin release from BRIN-BD1 1 cells incubated at 5.6mM glucose. 15 Fig. 26 shows the effects of various concentrations of GIP 1-42 and glycated GIP 1-42 on insulin release from BR[N-BD1 I cells incubated at 16.7mM glucose. Fig. 27 shows the effects of various concentrations of GIP 1-42 and GIP (Gly 2 ) on insulin release from BRIN-BD1 cells incubated at 5.6mM glucose. Fig. 28 shows the effects of various concentrations of GIP 1-42 and GIP 20 (Gly 2 ) on insulin release from BRIN-BD I1 cells incubated at 16.7mM glucose. Fig. 29 shows the effects of various concentrations of GIP 1-42 and GIP (Pro 3 ) on insulin release from BRIN-BD1 I cells incubated at 5.6mM glucose. Fig. 30 shows the effects of various concentrations of CIP 1-42 and GIP (Pro 3 ) on insulin release from BRIN-BD1 I cells incubated at 16.7mM glucose. 25 Fig. 31 a shows the primary structure of human gastric inhibitory polypeptide (GIP) (SEQ ID NO:1), and Fig. 31b shows the primary structure of porcine gastric inhibitory polypeptide (GIP) (SEQ ID NO:2). Figs. 32A and 32B are a line graph and a bar graph, respectively, showing the effects of (Pro 3 )GIP on GIP-stimulated cyclic AMP generation and insulin secretion 30 in vitro. Figs. 33A - 33F are a set of six bar graphs showing the effects of Glu 3 . substituted forms of GIP and GIP(3-42) on GIP-stimulated insulin secretion in vitro. -9- WO 2005/082928 PCT/GB2005/000710 Figs. 34A through 34D are a set of two line graphs (Figs. 34A, 34C) and two bar graphs (Figs. 34B, 34D) showing that acute administration of (Pro')GIP completely antagonises the actions of GIP on glucose tolerance (Figs. 34A, 34B) and plasma insulin (Figs. 34C, 34D) responses in obese diabetic ob/ob mice. 5 Figs. 35A through 35D are a set of two line graphs (Figs. 35A, 35C) and two bar graphs (Figs. 35B, 35D) showing that acute administration of (Pro 3 )GIP impairs physiological meal-stimulated insulin release and worsens glycemic excursion in obese diabetic ob/ob mice. Figs. 36A and 36B are a set of two bar graphs showing that chronic 10 administration of (Pro 3 )GIP for 11 days decreases plasma glucose and insulin concentrations of obese diabetic ob/ob mice. Figs. 37A through 37C are a set of three bar graphs showing that chronic administration of (Pro 3 )GIP for 11 days decreases glycated HbAIc, pancreatic insulin content and associated islet hypertrophy of obese diabetic ob/ob mice. 15 Figs. 38A through 38D are a set of two line graphs (Figs. 38A, 38C) and two bar graphs (Figs. 38B, 38D) showing that chronic administration of (Pro 3 )GIP for 11 days improves glucose tolerance of obese diabetic ob/ob mice without change of circulating insulin. Fig. 39 is a line graph showing that chronic administration of (Pros)GIP for 11 20 days improves insulin sensitivity in obese diabetic ob/ob mice. Fig. 40 is a line graph showing that the beneficial effects of chronic administration of (Pro 3 )GIP for 11 days in obese diabetic ob/ob mice are reversed 9 days after cessation of treatment. Figs. 41A and 41B are a set of two line graphs showing that chronic 25 administration of (Pro 3 )GIP for 11 days causes glucose intolerance in normal mice with reversal by 9 days after cessation of treatment. Figs. 42A through 42D are a set of two line graphs (Figs. 42A, 42C) and two bar graphs (Figs. 42B, 42D) showing the effects of (Pro 3 )GIP on plasma glucose and insulin response to native GIP 4 hours after administration. 30 Figs. 43A through 43D are a set of two line graphs and two bar graphs showing the effects of daily (Pro 3 )GIP administration on food intake (Fig. 43A), body weight (Fig. 43B), plasma glucose (Fig. 43C) and insulin (Fig. 43D) concentrations in ob/ob mice. -10- WO 2005/082928 PCT/GB2005/000710 Figs. 44A through 44D are a set of four line graphs with inset bar graphs showing the effects of daily (Pro 3 )GIP administration on glucose tolerance and plasma insulin response to glucose in ob/ob mice. Figs. 45A through 45D are a set of two line graphs (Figs. 45A, 45C) and two 5 bar graphs (Figs. 45B, 45D) showing the effects of daily (Pro 3 )GIP administration (A; black bars) or saline (c; white bars) on glucose (Figs. 45A, 45B) and insulin (Figs. 45C, 45D) responses to feeding in ob/ob mice fasted for 18 hours. Figs. 46A through 46D are a set of two line graphs (Figs. 46A, 46C) and two bar graphs (Figs. 46B, 46D) showing the effects of daily (Pro 3 )GIP administration on 10 insulin sensitivity in ob/ob mice. Figs. 47A through 47D are a set of four bar graphs showing the effects of daily (Pro 3 )GIP administration on pancreatic weight (Fig. 47A), insulin content (Fig. 47B), islet number (Fig. 47C) and islet diameter (Fig. 47D) in ob/ob mice. Figs. 48A through 48F are a set of two bar graphs (Figs. 48A, 48D) and four 15 photomicrographs (Figs. 48B, 48C, 48E, 48F), showing the effects of daily (Pro 3 )GIP administration on islet size and morphology in ob/ob)mice. Fig. 49 is an illustration of how the GIP receptor ("GIP-R") antagonist, (Pro 3 )GIP, counters beta cell hyperplasia, hyperinsulinemia and insulin resistance lead to improved glucose intolerance .and diabetes control. 20 Figs. 50A and 50B are a line graph and a bar graph, respectively, showing intracellular cyclic AMP production (Fig. 50A) by GIP (A) and fatty acid derivatised GIP analogues N-AcGIP(LysPAL 16 ) (o) and N-AcGIP(LysPAL 37 ) (e), and insulin releasing activity of glucose (5.6 mmo/l glucose; white bars), GIP (lined bars) and fatty acid derivatised GIP analogues (Fig. 50B) N-AcGIP(LysPAL 16 ) (grey bars) and 25 N-AcGIP(LysPAL 37 ) (black bars) in the clonal pancreatic beta cell line, BRIN-BD 1l. Figs. 51A through 51D arc a set of two line graphs (Figs. 51A, 51C) and two bar graphs (Figs. 51B, 51D) showing glucose lowering effects (Figs. 51A, 51B) and insulin-releasing activity (Figs. 51 C, 51 D) of GIP and fatty acid derivatised GIP analogues in 18 hour-fasted ob/ob mice. 30 Figs. 52A and 52B are a pair of bar graphs showing dose-dependent effects of GIP and N-AcGIP(LysPAL 37 ) in ob/ob mice fasted for 18 hours. Figs. 53A through 53E are a set of graphs showing the effects of daily N AcGIP(LysPAL 37 ) (*; black bars) administration on food intake (Fig. 53A), body - 11 - WO 2005/082928 PCT/GB2005/000710 weight (Fig. 531B), plasma glucose (Fig. 53C), insulin (Fig. 53D) and glycated hemoglobin N-AcGIP(LysPAL 37 ) (12.5 nmoles/kg/day) (Fig. 53E). Figs. 54A through 54D are a set of two line graphs (Figs. 54A, 54C) and two bar graphs (Figs. 54B, 54D) showing the effects of daily N-AcGIP(LysPAL 37 ) 5 administration on glucose tolerance (Figs. 54A, 54B) and plasma insulin response (Figs. 54C, 54D) to glucose. Figs. 55A through 55D are a line graph and three bar graphs showing the effects of daily N-AcGIP(LysPAL 37 ) administration on insulin sensitivity (Figs. 55A, 55B) and pancreatic weight (Fig. 55C) and insulin content (Fig. 55D). 10 Figs. 56A through 56D are a set of two line graphs (Figs. 56A, 56C) and two bar graphs (Figs. 56B, 56D) showing the retention of glucose lowering (Figs. 56A, 56B) and insulin releasing (Figs. 56C, 56D) activity of N-AcGIP(LysPAL 37 ) and native GIP after daily injection for 14 days. Figs. 57A through 57D are a set of two line graphs (Figs. 57A, 57C) and two 15 bar graphs (Figs. 57B, 57D) showing the acute glucose lowering (Figs. 57A, 57B) and insulin releasing (Figs. 57C, 57D) effects of N-AcGIP(LysPAL 37 ) after 14 daily injections of either N-AcGIP(LysPAL 3 7 ) (12.5 nmoles/kg/day; *; black bars), native GIP (12.5 nmoles/kg/day; A; lined bars) or saline vehicle (control; n; white bars). 20 DETAILED DESCRIPTION The peptide analogues disclosed herein display resistance to degradation by the enzyme DPP IV. These analogues include those with alterations at residues 1, 2 and/or 3 of the native GIP(1-42) peptide, where the alterations interfere with or delay cleavage by DPP IV. The alterations can include chemical modification of one or 25 more of the first three amino acids, such as by acylation, acetylation, alkylation, glycation, conversion of a bond between two amino acids, such as to a psi-[CH 2 NH] bond, or to a stable isotere bond, or addition of an isopropyl group. The alterations can also include amino acid substitutions at the 1, 2, and/or 3 position, to either a different naturally-occurring amino acid, or an amino acid not encoded by the genetic 30 code. Other alterations are also possible, and the object is to prevent cleavage of the peptide by DPP IV, yet still allow for insulin secretion. This may be accomplished by alterations at other regions of the peptide, such as by alterations that alter the three dimensional structure to prevent DPP IV cleavage, yet still allow insulin secretion. - 12 - WO 2005/082928 PCT/GB2005/000710 Preferred alterations include chemical modifications of residues 1, 2, and 3, amino acid substitutions at residues 1, 2, and 3, and chemical modifications of lysine residues throughout the protein. Particularly preferred alterations include acetylation of Tyri and linkage of a 5 C-16 palmitate group to the epsilon amino group of a lysine residue (especially Lys 16 or Lys 37 ), or substitution of Glu 3 , especially by proline. The modification can also be the linking of, e.g., a C-8, a C-10, a C-12, a C-14, a C-18 or a C-20 palmitate group to the epsilon amino group of a lysine residue. It has been found that longer-term, as opposed to acute, GIP receptor 10 antagonism using Glu3-substituted forms of GIP, such as (Pro 3 )GIP, improve obesity related insulin resistance and associated glucose intolerance. This has uncovered an unexpected approach to the therapy of obesity, insulin resistance and type 2 diabetes. As described in Example 1 below, an N-terminally modified version of the GIP protein was prepared, as were analogues of the modified protein. The protein and 15 its analogues were then evaluated in Example 2 for their antihyperglycemic and insulin-releasing properties in vivo, and were found to exhibit a substantial resistance to amino peptidase degradation and increased glucose lowering activity relative to native GIP. The 42 amino acid GIP is an important incretin hormone released into the 20 circulation from endocrine K-cells of the duodenum and jejunum following ingestion of food. The high degree of structural conservation of GIP among species supports the view that this peptide plays an important role in metabolism. Secretion of GIP is stimulated directly by actively transported nutrients in the gut lumen without a notable input from autonomic nerves. The most important stimulants of GIP release are 25 simple sugars and unsaturated long chain fatty acids, with amino acids exerting weaker effects. As with tGLP- 1, the insulin-releasing effect of GIP is strictly glucose-dependent. This affords protection against hypoglycemia and thereby fulfills one of the most desirable features of any current or potentially new antidiabetic drug. The present results demonstrate for the first time that Tyr'-glucitol GIP 30 displays profound resistance to serum and DPP IV degradation. Using ESI-MS the present study showed that native GIP was rapidly cleaved in vitro to a major 4748.4 Da degradation product corresponding to GIP(3-42), which confirmed previous findings using matrix-assisted laser desorption ionization time-of-flight mass - 13 - WO 2005/082928 PCT/GB2005/000710 spectrometry. Serum degradation was completely inhibited by diprotin A (Ile-Pro Ile), a specific competitive inhibitor of DPP IV, confirming this as the main enzyme for GIP inactivation in vivo. In contrast, Tyr]-glucitol GIP remained almost completely intact after incubation with serum or DPP IV for up to 12 hours. This 5 indicates that glycation of GIP at the amino-terminal Tyr' residue masks the potential cleavage site from DPP IV and prevents removal of the Tyr'-Ala 2 dipeptide from the N-terminus preventing the formation of GIP(3-42). Consistent with in vitro protection against DPP IV, administration of Tyr' glucitol GIP significantly enhanced the antihyperglycemic activity and insulin 10 releasing action of the peptide when administered with glucose to rats. Native GIP enhanced insulin release and reduced the glycemic excursion as observed in many previous studies. However, amino-terminal glycation of GIP increased the insulin releasing and antihyperglycemic actions of the peptide by 62% and 38% respectively, as estimated from insulin area under the curve (AUC) measurements. Detailed kinetic 15 analysis is difficult due to necessary limitation of sampling times, but the greater insulin concentrations following Tyr 1 -glucitol GIP as opposed to GIP at 30 minutes post-injection is indicative of a longer half-life. The glycemic rise was modest in both peptide-treated groups and glucose concentrations following injection of Tyr 1 -glucitol GIP were consistently lower than after GIP. Since the insulinotropic actions of GIP 20 are glucose-dependent, it is likely that the relative insulin-releasing potency of Tyr' glucitol GIP is greatly underestimated in the present in vivo experiments. In vitro studies in the laboratory of the present inventors using glucose responsive clonal B-cells showed that the insulin-releasing potency of Tyr'glucitol GIP was several orders of magnitude greater than GIP and that its effectiveness was 25 more sensitive to change of glucose concentrations within the physiological range. Together with the present in vivo observations, this suggests that N-terminal glycation of GIP confers resistance to DPP IV degradation whilst enhancing receptor binding and insulin secretory effects on the B-cell. These attributes of Tyr]glucitol GIP are fully expressed in vivo where DPP IV resistance impedes degradation of the peptide 30 to GIP(3-42), thereby prolonging the half-life and enhancing effective concentrations of the intact biologically active peptide. It is thus possible that glycated GIP is enhancing insulin secretion in vivo both by enhanced potency at the receptor as well as improving DPP IV resistance. Thus numerous studies have shown that GIP (3-42) - 14 - WO 2005/082928 PCT/GB2005/000710 and other N-terminally modified fragments, including GIP(4-42), and GIP(17-42) are either weakly effective or inactive in stimulating insulin release. Furthermore, evidence exists that N-terminal deletions of GIP result in receptor antagonist properties in GIP receptor transfected Chinese hamster kidney cells [9], suggesting 5 that inhibition of GIP catabolism would also reduce the possible feedback antagonism at the receptor level by the truncated GIP(3-42). In addition to its insulinotopic actions, a number of other potentially important extrapancreatic actions of GIP may contribute to the enhanced antihyperglycemic activity and other beneficial metabolic effects of Tyr 1 -glucitol GIP. These include the 10 stimulation of glucose uptake in adipocytes, increased synthesis of fatty acids and activation of lipoprotein lipase in adipose tissue. GIP also promotes plasma triglyceride clearance in response to oral fat loading. In liver, GIP has been shown to enhance insulin-dependent inhibition of glycogenolysis. GIP also reduces both glucagon-stimulated lipolysis in adipose tissue as well as hepatic glucose production. 15 Finally, recent findings indicate that GIP has a potent effect on glucose uptake and metabolism in mouse isolated diaphragm muscle. This latter action may be shared with tGLP- 1 and both peptides have additional benefits of stimulating somatostatin secretion and slowing down gastric emptying and nutrient absorption. This study demonstrates that the glycation of GIP at the aminoterminal Tyr 20 residue limits GIP catabolism through impairment of the proteolytic actions of serum peptidases and thus prolongs its half-life in vivo. This effect is accompanied by enhanced antihyperglycemic activity and raised insulin concentrations in vivo, suggesting that such DPP IV resistant analogues are potentially useful therapeutic agents for NIDDM. Tyr 1 -glucitol GIP appears to be particularly interesting in this 25 regard since such amino-terminal modification of GIP enhances rather than impairs glucose-dependent insulinotropic potency as was observed recently for tGLP-1. As shown in Table 1 in Example 3, glycated GIP, acetylated GIP, GIP(Ser 2 ) are GIP(Abu 2 ) more resistant than native GIP to in vitro degradation with DPP IV. From these data GIP(Sar 2 ) appears to be less resistant. As shown in Table 2, all 30 analogues tested exhibited resistance to plasma degradation, including GIP(Sar 2 ) which from DPP IV data appeared least resistant of the peptides tested. DPA substantially inhibited degradation of GIP and all analogues tested with complete - 15- WO 2005/082928 PCT/GB2005/000710 abolition of degradation in the cases of GIP(Abu 2 ), GIP(Ser 2 ) and glycated GIP. This indicates that DPP IV is a key factor in the in vivo degradation of GIP. As shown in Figs. 17-30, the glycated GIP analogue exhibited a considerably greater insulinotropic response relative to native GIP. N-terminal acetylated GIP 5 exhibited a similar pattern and the GIP(Ser 2 ) analogue also evoked a strong response. From these tests, GIP(Gly 2 ) and GIP(Pros) appeared to the least potent analogues in terms of insulin release. Other stable analogues tested, namely GIP(Abu 2 ) and GIP(Sar 2 ), exhibited a complex pattern of responsiveness dependent on glucose concentration and dose employed. Thus, very low concentrations were extremely 10 potent under hyperglycemic conditions (16.7 mM glucose). This suggests that even these analogues may prove therapeutically useful in the treatment of type 2 diabetes where insulinotropic capacity combined with in vivo degradation dictates peptide potency. A major limitation to the possible therapeutic use of both GIP and GLP-1 as 15 insulin-releasing agents for the treatment of diabetes is their rapid degradation in vivo by dipeptidylpeptidase-IV (DPP-IV; EC 3.4.14.5). This enzyme rapidly removes the amino-terminal dipeptide from the two peptides producing GIP(3 -42) and GLP- 1(9 36), which lack biological activity (Gault, V.A. et al., 2002, Biochem. Biophys. Res. Commun. 290:1420-1426). In searching for stable amino-terminally modified forms 20 of GIP and GLP- 1, it was discovered that a novel synthetic GIP analogue with a single proline substitution at position 3 close to the cleavage site, (Pro 3 )GIP, functioned as a potent GIP receptor antagonist. As shown in Example 4, below, (Pro 3 )GIP, other Glu 3 -substituted forms of GIP and GIP(3-42) are potent GIP receptor antagonists both in vitro and in vivo. 25 Experiments evaluating the effects of chronic GIP receptor antagonism in normal mice using (Pros)GIP demonstrated a substantial but reversible deterioration of glucose tolerance. This is entirely consistent with the widely recognised physiological role of GIP as an important insulin-releasing intestinal hormone involved in the regulation of glucose disposal following feeding (Meier, J.J. et al., 30 2002, Regul. Pept. 107:1-13). Most notably, and in complete contrast to normal mice, the experiments disclosed herein show that chronic (Pro 3 )GIP administration to obese diabetic ob/ob mice for 11 days does not worsen glucose intolerance and diabetes status at all. - 16 - WO 2005/082928 PCT/GB2005/000710 Surprisingly, GIP receptor antagonism in this obese insulin resistant model was associated with highly substantial improvements of glycated HbA I, plasma glucose and insulin concentrations, glucose tolerance and insulin sensitivity. Pancreatic insulin content was also decreased and the characteristic islet hypertrophy of the 5 obese mutant was partially reversed. These latter observations indicate a decreased secretory demand for endogenous insulin following (Pro 3 )GIP as a result of improved insulin resistance. Indeed, insulin sensitivity tests conducted in ob/ob mice 11 days into (Pro 3 )GIP treatment revealed a substantial improvement in tissue insulin insensitivity, 10 which more than compensated for the functional ablation of the insulin-releasing GIP component of the enteroinsular axis. The exact mechanism responsible for this effect on insulin sensitivity is unknown but ablation of direct action of circulating GIP on adipose tissue metabolism is a likely candidate. Also noteworthy was the fact that all these beneficial actions of (Pro 3 )GIP in obese diabetic ob/ob mice were reversed 15 within 9 days cessation of treatment. These results clearly indicate that (Pro 3)GIP and other analogues based on Glu 3 -substituted or N-terminally truncated forms of the gastrointestinal hormone GIP can offer an important therapeutic means of alleviating insulin resistance for the treatment of obesity, the so-called insulin resistant (metabolic) syndrome and type 2 20 diabetes in humans. Some studies have attempted to enhance incretin action using DPP IV inhibitors or stable analogs of GLP- I and GIP for the treatment of type 2 diabetes (Green, B.D. et al., 2004, Curr. Pharm. Des. 10:In Press; Drucker, D.J. et al., Diabetes Care 10:2929-2940). Such an approach is reliant on the possibility that 25 incretin action is defective in diabetes and that the underlying defects responsible for metabolic disarray might be over-ridden by exogenous GLP--1 or GIP administration. There is some evidence for a beneficial and possibly therapeutic role of both GLP-1 and GIP analogs in diabetes (Meier, J.J. et al., 2002, Regul. Pept. 107:1-13; Gault, V.A. et al., 2003, Biochem Biophys Res Commun 308:207-213; Holst, J.J. et al., 2004, 30 Am. J. Physiol. Endocrinol. Metab. 287:E1 99-E206; Green, B.D. et al., 2004, Curr. Pharm. Des. 10:In Press; Drucker, D.J. et al., Diabetes Care 10:2929-2940). Nevertheless, understanding of the possible involvement of incretin hormones in the pathophysiology of diabetes is lacking, partly due to cross-reaction of classical GLP-1 - 17- WO 2005/082928 PCT/GB2005/000710 and GIP radioimmunoassays with the predominant DPP IV-generated truncated peptide forms, GLP-1(9-36) and GIP(3-42), which circulate at particularly high concentrations (Meier, J.J. et al., 2002, Regul. Pept. 107:1-13). Some clinical studies seems to suggest existence of a defect in the secretion of GLP- 1 and a defect in the 5 action of GIP in type 2 diabetes (Holst, J.J. et al., 2004, Am. J Physiol. Endocrinol. Metab. 287:E199-E206). However, the basis for such a conclusion is not impressive given the many previous contradictory human studies (Morgan, L.M., "Insulin secretion and the enteroinsular axis," In: Nutrient regulation of insulin secretion, Flatt, P.R., ed.; London, Portland Press, 1992, p. 1-22), and the likelihood that the 10 reported insensitivity of pancreatic beta cells to GIP (Vilsboll, T. et al., 2002, Diabetologia 45:1111-1119) may reflect a generalized secretory dysfunction rather than a specific cellular defect (Meier, J.J. et al., 2003, Metabolism 52:1579-1585). Indeed, the insulin secretory response to all secretagogues, including GLP-1 is compromised in type 2 diabetes (Kjems, L.L. et al., 2003, Diabetes 52:380-386; Flatt, 15 P.R. et al., "Defective insulin secretion in diabetes and insulinoma," in Nutrient regulation of insulin secretion, Flatt P.R., ed. London, Portland Press, 1992, p. 341 386). Thus the proposed use of GLP- 1 and GIP for diabetes therapy is reliant on peptide engineering to provide analogs of incretin hormones with improved potency due to DPP IV resistance, decreased renal clearance and/or enhanced GIP receptor 20 and post-receptor activity (Gault, V.A. et al., 2003, Biochem Biophys Res Commun 308:207-213). Although no single animal model can match the complex etiology of type 2 diabetes in man, studies of the ob/ob syndrome in mice have highlighted notable abnormalities of GIP in relation to the interplay between hyperphagia, 25 hyperinsulinemia and the metabolic demise associated with progressive obesity diabetes (Flatt, P.R. et al., 1983, Diabetes 32:433-435; Flatt, P.R. et al., 1984, J Endocrinol. 101:249-256; Bailey, C.J. et al., 1986, Acta Endocrinol. (Copenh) 112:224-229). These animals constitute a model of non-insulin dependent diabetes associated with gross obesity and severe insulin resistance, driven by leptin deficiency 30 (Bailey, C.J. et al., "Animal syndromes resembling type 2 diabetes," in Textbook of Diabetes, 3rd ed. Pickup J.C. & Williams G., eds. Oxford, Blackwell Science Ltd., 2003, p. 25.1-25.30). Furthermore, recent research suggests an interaction between leptin and the enteroinsular axis (Anini, Y. et al., 2003, Diabetes 52:252-259) and that - 18- WO 2005/082928 PCT/GB2005/000710 over-stimulation of the GIP receptor ("GIP-R") on adipocytes appears to be an important contributor to fat deposition in ob/ob mice (Miyawaki, K. et al., 2002, Nat. Med. 8:738-742). As shown in Example 5, below, daily injections of the stable and specific GIP 5 R antagonist, (Pro 3 )GIP can be used to chemically ablate the GIP-R and evaluate the role of endogenous circulating GIP in obesity-diabetes as manifested in ob/ob mice. The results reveal a cardinal role for GIP in insulin resistance and associated metabolic disturbances, and provide the first experimental evidence that GIP-R antagonists might provide a novel and effective means of treating obesity-driven 10 forms of type 2 diabetes. Knock-out of the GIP-R in normal mice has been shown to result in significant impairment of glucose tolerance and meal-induced insulin secretion without appreciable effects on food intake, body weight or basal glucose or insulin concentrations (Miyawaki, K. et al., 1999, Proc. Nat. Acad. Sci. USA 96:14843 15 14847). More recent studies with genetic GIP-R knockout mice have corroborated these findings and additionally shown that GIP has a significant involvement in the enteroinsular axis (Pederson, R.A. et al., 1998, Diabetes 47:1046-1052; Pamir, N. et al., 2003, Am. J. Physiol. Endocrinol. Metab. 284:E931-939). However, double knockout of receptors for GLP- 1 and GIP results in g surprisingly modest 20 deterioration of glucose homeostasis (Hansotia, T., et al., 2004, Diabetes 53:1326 1335; Preitner, F., et al., 2004, J. Clin. Invest. 113:635-645), indicating possible up regulation of compensatory mechanisms during life-long deletion of GLP-1 and GIP receptors. The analogue (Pro 3 )GIP can be used as a specific and potent antagonist of the 25 GIP-R that is highly stable and resistant to DPP IV-mediated degradation (Gault, V.A. et al., 2002, Biochen. Biophys. Res. Commun. 290:1420-1426). Using (Pro')GIP acutely, the results disclosd herein highlight the involvement of GIP in the plasma insulin response to feeding and the enteroinsular axis of ob/ob mice (Gault, V.A. et al., 2003, Diabetologia 46:222-230). Comparison with the effects of the GLP-1-R 30 antagonist, exendin(9-39), indicates that GIP contributes substantially to the insulin releasing actions of the enteroinsular axis and represents the major physiological incretin (Gault, V.A. et al., 2003, Diabetologia 46:222-230). Once daily administration of (Pro 3 )GIP to normal mice for 11 days results in the reversible - 19- WO 2005/082928 PCT/GB2005/000710 impairment of glucose tolerance associated with decreased insulin sensitivity (Irwin, N., 2004, Biol. Chem. 385:845-852). Basal and postprandial insulin secretion together with pancreatic insulin content and islet morphology were unchanged. Thus longer-term chemical ablation of GIP-R function with daily (Pro 3 )GIP can mimic the 5 phenotype induced by genetic GIP-R knockout in mice with the exception of revealing a potentially important additional effect of endogenous GIP on insulin action, which appears to be independent of enhanced insulin secretion. Far from reproducing this predicted scenario and the metabolic deterioration observed following genetic or chemical knockout of the GIP-R in normal mice 10 (Miyawaki, K. et al., 1999, Proc. Nat. Acad. Sci. USA 96:14843-14847; Irwin, N., 2004, Biol. Chem. 385:845-852), ob/ob mice treated with daily (Pro 3 )GIP for 11 days exhibited a marked improvement in diabetic status. This included decreased fasting and basal hyperglycemia, lowered glycated hemoglobin, improved glucose tolerance and a significantly diminished glycemic excursion following feeding. Notably, basal 15 and glucose-stimulated plasma insulin concentrations were decreased, suggesting that insulin sensitivity must have improved significantly following (Pro 3 )GIP in order to restrain the hyperglycemia. Indeed, insulin sensitivity tests conducted after 11 days of (Pro 3 )GIP administration revealed a 57% increase in the glucose-lowering action of exogenous insulin. Bearing in mind that the severity of the ob/ob syndrome 20 represents a tough test for current antidiabetic drugs, including insulin, sulfonylureas, metformin and thiazolidenediones (Flatt, P.R. et al., "Defective insulin secretion in diabetes and insulinoma," in Nutrient regulation of insulin secretion, Flatt P.R., ed. London, Portland Press, 1992, p. 341-386; Stevenson, R.W. et al., 1995, The Diabetes Annual 9:175-191; Wiernsperger, N.F., "Preclinical pharmacology of biguanides," 25 Handbook of Experimental Pharmacology 119:305-358, 1996), induction of such rapid and reversible changes by GIP-R blockade using (Pro 3 )GIP is unprecedented. It is important to note that the above effects were observed independently of any change in food intake or body weight in (Pro 3 )GIP treated ob/ob mice. This accords with the view that endogenous GIP lacks effects on feeding activity (Meier, 30 J.J. et al., 2002, Regul. Pept. 107:1-13). However, the observation on body weight contrasts with findings in ob/ob mice cross-bred to genetically knockout GIP-R function (Miyawaki, K. et al., 2002, Nat. Med. 8:738-742). Thus in these transgenic mice, life-long depletion of GIP-R function was associated with decreased body - 20 - WO 2005/082928 PCT/GB2005/000710 weight gain and significant amelioration of both adiposity and insulin resistance compared with control (Lepob/Lepob) mice (Miyawaki, K. et al., 2002, Nat. Med. 8:738-742). In this previous study, the improvement of insulin sensitivity may have been a simple consequence of reduced adipose tissue mass as this would significantly 5 enhance peripheral glucose disposal (Bailey, C.J. et al., "Animal syndromes resembling type 2 diabetes," in Textbook ofDiabetes, 3rd ed. Pickup J.C. & Williams G., eds. Oxford, Blackwell Science Ltd., 2003, p. 25.1-25.30). However, the present results observed in rapid time and without effects on feeding or body weight clearly indicate the involvement of an alternative mechanism. 10 The most plausible explanation for the present data stem from appreciation of the key milestones in the age-dependent progression of the ob/ob syndrome on the Aston background as depicted in Fig. 49, which is an illustration of how the GIP-R antagonist, (Pro 3 )GIP, counters beta cell hyperplasia, hyperinsulinemia and insulin resistance lead to improved glucose intolerance and diabetes control. Possible longer 15 term direct actions of GIP on adipocyte function and fat stores, suggested by studies in GIP-R knockout ob/ob mice have been omitted. Due to double recessive ob mutation and resulting leptin deficiency, young ob/ob mice develop a profound early hyperphagia (Bailey, C.J., et al., 1982, lit. J. Obes. 6:11-21). Substantial enteroendocrine stimulation results in K-cell hyperplasia 20 and markedly elevated concentrations of intestinal and circulating GIP (Flatt, P.R. et al., 1983, Diabetes 32:433-435; Flatt, P.R. et al., 1984, J Endocrinol. 101:249-256; Bailey, C.J. et al., 1986, Acta Endocrinol. (Copenh) 112:224-229). This in turn promotes islet hypertrophy and beta cell hyperplasia (Bailey, C.J., et al., 1982, Int. J. Obes. 6:11-21) together with marked hyperinsulinemia and mounting insulin 25 resistance (Flatt, P.R., et al., 1981, Horin Metab Res 13:556-560). This process manifests itself in terms of rising basal hyperglycemia and glucose intolerance. A vicious spiral is thus established wherein beta cell compensation results in marked hyperinsulinemia which attempts to moderate increasing insulin resistance (Bailey, C.J., et al., 1982, Int. J. Obes. 6:11-21; Flatt, P.R., et al., 1981, Horm Metab Res 30 13:556-560). Viewed in this context, it is clear that chemical ablation of GIP-R function with daily (Pro 3 )GIP will decrease beta cell stimulation and hyperinsulinemia. However, instead of causing further impairment of glucose homeostasis, a preferentially marked improvement of insulin sensitivity results in a -21- WO 2005/082928 PCT/GB2005/000710 substantial improvement of the metabolic syndrome. Further support for this scenario, is the partial amelioration of islet hypertrophy and beta cell hyperplasia in (Pro 3 )GIP treated ob/ob mice (Fig. 48). Notably, average islet diameter was diminished with the largest islets (>15 mm) being replaced by a greater proportion 5 with small or medium diameters (0.1-15mm). These effects were largely reversed by 9 day cessation of treatment, supporting the idea of active islet and beta cell growth in adult ob/ob mice (Bailey, C.J., et al., 1982, Int. J Obes. 6:11-21). Recent observations indicate that GIP acts as a mitotic stimulus and anti-apoptotic agent to the beta cell (Pospisilik, J.A. et al., 2003, Diabetes 52:741-750; Trumper, A. et al., 10 2001, Mol. Endocrinol. 15:1559-1570; Ehses, J.A. et al., 2003, Endocrinology 144:4433-4445, Trumper, A. et al., 2002, J Endocrinol. 174:233-246). Thus, it is believed that negative effects of (Pro 3 )GIP on islet size reflects a combination of decreased proliferation and increased apoptosis of beta cells. The results shown in Example 5 have demonstrated for the first time that daily 15 administration of the GIP-R antagonist, (Pro 3 )GIP, improves glucose tolerance and ameliorates insulin resistance and abnormalities of islet structure and function in ob/ob mice. Notably, these effects were reversed by discontinuation of (Pro 3 )GIP for 9 days. Freedom from any obvious side effects also accords with earlier observations in normal mice (Irwin, N., 2004, Biol. Chem. 385:845-852) and mice genetically 20 engineered with life-long GIP-R deficiency (Miyawaki, K. et al., 2002, Nat. Med. 8:738-742). The present observations point to a cardinal role of endogenous GIP in the pathogenesis of obese-insulin resistant-diabetes. More importantly, the data indicate that GIP-R antagonists, such as (Pro 3 )GIP, provide a novel, physiological and effective means to treat obese type 2 diabetes through the alleviation of insulin 25 resistance. In Example 6, fatty acid derivatisation was used to develop two novel long acting, N-terminally modified GIP analogues (N-AcGIP(LysPAL 16 ) and N AcGIP(LysPAL 37 )). Degradation studies were carried out with dipeptidylpeptidase IV (DPP IV). 30 Cyclic AMP production was assessed using GIP receptor transfected CHL fibroblasts. In vitro insulin release was assessed in BRIN-BD1 1 cells. Insulinotropic and glycaemic responses to acute and long-term peptide administration were evaluated in obese diabetic (ob/ob) mice. - 22 - WO 2005/082928 PCT/GB2005/000710 In contrast to GIP both analogues displayed resistance to DPP IV degradation. The analogues also stimulated cyclic AP production and exhibited significantly improved in vitro insulin secretion compared to control. Administration of N AcGIP(LysPAL 16 ) or N-AcGIP(LysPAL 3 7 ) together with glucose in ob/ob mice 5 significantly reduced the glycaemic excursion and improved the insulinotropic response compared to GIP. Dose-response studies with N-AcGIP(LysPAL 3 ) revealed highly significant decreases in the overall glycaemic excursion and increases in circulating insulin even with 6.25 nmoles/kg. Once daily injection of ob/ob mice with N-AcGIP(LysPAL 37 ) over 14 days significantly decreased plasma glucose, 10 glycated haemoglobin and improved glucose tolerance compared with saline or native GIP. Plasma and pancreatic insulin were significantly increased, together with a significant enhancement in the insulin response to glucose and a notable improvement of insulin sensitivity. No evidence was found for GIP-receptor desensitization and the metabolic effects of N-AcGIP(LysPAL 37 ) were independent of any change in feeding 15 or body weight. These results show that novel fatty acid derivatised, N-terminally modified analogues of GIP such as N-AcGIP(LysPAL 37 ), may have significant potential for the treatment of type 2 diabetes. One approach to counter both renal clearance and enzyme degradation of GIP 20 concerns the utilisation of fatty acid derivatisation together with N-terminal modification. Fatty acid derivatisation has previously been shown to prolong the half life of insulin (Kurtzhals, P. et al., 1995, Biochem. J. 312: 725-73 1) and the sister incretin glucagon-like peptide-1 (GLP-1) (Knudsen, L.B. et al., 2000, J Med. Chem. 43: 1664-1669; Green, B.D. et aL., 2004, Biol. Chem. 385: 169-177; Kim, J.G. et al., 25 2003, Diabetes 52: 751-759). A number of N-terminally modified GIP analogues have been developed which exhibit profound resistance to DPP IV (Hinke, S.A. et al., 2002, Diabetes 51: 656-66 1; Gault, V.A. et al., 2002, Biochem. J. 367: 9 13-920; Gault, V.A. et al., 2003, J Endocrinol. 176: 133-141; O'Harte, F.P.M. et al., 1999, Diabetes 48: 758-765). Several of these, most notably those modified at Tyr, of GIP 30 with an addition of an acetyl, glucitol, pyroglutamyl or Fmoc adduct, exhibit enhanced activity at the GIP receptor in vitro (Gault, V.A. et aL., 2002, Biochem. J. 367: 913-920; O'Harte, F.P.M. etal., 1999, Diabetes 48: 758-765; O'Harte, F.P.M. et al., 2002, Diabetologia 45: 1281-1291). As a result of degradation resistance and - 23 - WO 2005/082928 PCT/GB2005/000710 enhanced cellular activity, these analogues display enhanced and protracted antihyperglycaemic and insulin-releasing activity when administered acutely to animals with obesity-diabetes (Hinke, S.A. et al., 2002, Diabetes 51: 656-66 1; Gault, V.A. et al., 2002, Biochem. J. 367: 913-920; Gault, V.A. et al., 2003, J Endocrinol. 5 176: 133-141; O'Harte, F.P.M. et al., 1999, Diabetes 48: 758-765; O'Harte, F.P.M. et al., 2002, Diabetologia 45: 1281-1291). Of these, N-AcGIP has emerged as the most effective DPP IV-resistant analogue, substantially augmenting the plasma insulin response and curtailing the glycaemic excursion following conjoint administration with glucose to ob/ob mice (O'Harte, F.P.M. et al., 2002, Diabetologia 45: 1281 10 1291). Example 6 was designed to evaluate the metabolic stability, biological activity and antidiabetic potential of novel second generation fatty acid derivatised, N terminally modified N-AcGIP analogues, N-AcGIP(LysPAL1 6 ) and N AcGIP(LysPAL 37 ). Both GIP analogues contain a C-16 palmitate group linked to the 15 epsilon-amino group of Lys at positions 16 or 37, in combination with an N-terminal (Tyr') acetyl group (O'Harte, F.P.M. et al., 2002, Diabetologia 45: 1281-1291). The relative stability to DPP IV degradation, insulin secretion and cyclic AMP properties were examined in vitro together with acute and dose-response studies in obese diabetic ob/ob mice. The most effective analogue, N-AcGIP(LysPAL 37 ) was 20 administered to ob/ob mice by once daily intraperitoneal injection for 14 days prior to evaluation of glucose homeostasis, pancreatic beta cell function and insulin sensitivity. Possible desensitization of GIP receptor action by prolonged exposure to elevated concentrations of N-AcGIP(LysPAL 37 ) was also examined. The results indicate the particular promise of the novel second generation N-terminally acetylated 25 GIP analogue, N-AcGIP(LysPAL 37 ), as a potential therapeutic agent for the treatment of type 2 diabetes. Despite their many attributes, DPP IV-resistant analogues of GIP and GLP-1, like their native counterparts, are still subject to renal filtration. To circumvent this problem, fatty acid derivatisation has been used to improve the duration of action of 30 GLP-l (Knudsen, L.B. et al., 2000, J. Med. Chem. 43: 1664-1669; Green, B.D. et al., 2004, Biol. Chem. 385: 169-177; Kim, J.G. et al., 2003, Diabetes 52: 751-759). The most promising analogue, NN221 1 (Liraglutide), appears effective in improving blood glucose control in type 2 diabetic subjects despite a tendency towards - 24 - WO 2005/082928 PCT/GB2005/000710 promotion of nausea possibly due to slowing of gastric emptying (Agerso, H. et al., 2002, Diabetologia 45: 195-202). Example 6 describes the results of introducing two specific modifications to the native GIP hormone, namely N-terminal acetylation and C-terminal fatty acid 5 derivatisation. N-terminal acetylation was employed, as previously described (O'Harte, F.P.M. et al., 2002, Diabetologia 45: 1281-1291), to significantly enhance stability to DPP IV. In contrast, conjugation of a C-16 palmitate residue at the epsilon-amino group of Lys 16 or Lys 37 was introduced to extend the biological half life through binding to circulating proteins (Kurtzhals, P. et al., 1995, Biochem. J. 10 312: 725-731). Unlike the native peptide, both GIP analogues appeared to be completely resistant to enzymatic breakdown by DPP IV, which corroborates previous observations with N-AcGIP (O'Harte, F.P.M. et al., 2002, Diabetologia 45: 1281-1291). Furthermore, both analogues displayed similar or slightly better insulin releasing and cyclic AMP generating properties to native GIP and N-AcGIP when 15 tested in the in vitro cellular systems (O'Harte, F.P.M. et al., 2002, Diabetologia 45: 1281-1291). To assess the antihyperglycaemic and insulinotropic potential of the fatty acid derivatised GIP analogues in vivo, obese diabetic (ob/ob) mice were employed. The ob/ob syndrome is an extensively studied model of spontaneous obesity and diabetes, 20 exhibiting hyperphagia, marked obesity, moderate hyperglycaemia and severe hyperinsulinemia (Bailey, C.J. et al., 1982, Int. J. Obesity 6: 11-21). As described in previous studies (Gault, V.A. et al., 2002, Biochen. J 367: 9 13-920; Gault, V.A. et al., 2003, J Endocrinol. 176: 133-141), native GIP only modestly reduced the glycaemic excursion in ob/ob mice reflecting the severe insulin resistance of this 25 mutant animal model (Bailey, C.J. et al., 1982, Int. J. Obesity 6: 11-21). In sharp contrast, both N-acetylated GIP analogues additionally substituted with a palmitate molecule at Lys or Lys" (N-AcGIP(LysPAL 16) and N-AcGIP(LysPAL )) significantly lowered plasma glucose levels compared to the native peptide. This was accompanied by significantly enhanced insulin-releasing activity, especially in the 30 case of N-AcGIP(LysPAL 37 ). The significantly protracted insulinotropic response to both fatty acid derivatised GIP analogues at 60 minutes despite substantially lower plasma glucose is indicative of an extended plasma half-life. This may be due to binding of both palmitate derivatised GIP analogues to serum albumin, therefore - 25 - WO 2005/082928 PCT/GB2005/000710 significantly impairing their clearance via the kidneys (Meier, J.J. et al., 2004, Diabetes 53: 654-662). However, further studies including establishment of sensitive and specific immunoassays for the novel GIP analogues would be needed to confirm such actions. 5 N-AcGIP(LysPAL 37 ) appeared to be the best fatty acid derivatised analogue displaying a more protracted, significantly enhanced insulin-releasing potency over N AcGIP(LysPAL1 6 ) in vivo. Reasons for the increased potency ofN AcGIP(LysPAL3 7 ) remain unclear, but one explanation is an extended half-life. Another possibility may be that a fatty acid chain linked to the Lys closer to the C 10 terminus of the peptide may have less of a detrimental effect upon the bioactive region of the molecule known to be located within the N-terminus (Gault, V.A. et al., 2002, Biosci. Rep. 22: 523-528; Hinke, S.A. et al., 2001, Biochii. Biophys. Acta 1547: 143-55; Manhart, S. et al., 2003, Biochenistry 42: 3081-3088). However, similarities between the in vitro biological activities of the two palmitate substituted 15 analogues make this less likely. Given that N-AcGIP(LysPAL 37 ) was the more potent of the two analogues in vivo, it was further utilised in dose-response studies. Considering that native GIP itself has only very modest effects in ob/ob mice, as sometimes observed with type 2 diabetic subjects (Nauck, M.A. et al., 1993, J Clin. Invest. 91: 301-307; Meier, J.J. et 20 al., 2004, Diabetes 53: 220-224; Vilsbll, T. et al., 2002, Diabetologia 45: 1111 1119), it is remarkable that N-AcGIP(LysPAL 37 ), even at the lowest dose of 6.25 nmoles/kg, exhibited significant glucose-lowering and insulinotropic activity when administered with glucose. Considering N-AcGIP(LysPAL 37 ) is subject to albumin binding, the fact that it is still highly biologically active even at lower concentrations 25 indicates striking potency. Daily administration of N-AcGIP(LysPAL 37 ) to young adult ob/ob mice by intraperitoneal injection (12.5 nmoles/kg) resulted in a progressive lowering of plasma glucose concentrations and a significant decrease of glycated haemoglobin by 14 days. This was associated with a substantial improvement of glucose tolerance. 30 Importantly food intake and body weight were unchanged ruling out the possibility that improvement of glucose homeostasis was merely the consequence of body weight loss. These observations also indicate that N-AcGIP(LysPAL ) did not exert any untoward toxic actions affecting feeding over the study period. This is in harmony - 26 - WO 2005/082928 PCT/GB2005/000710 with recent studies showing that GIP does not inhibit gastric emptying (Meier, J.J. et aL, 2003, ,Am. Ji Physiol. Endocrinol. Metab. 286: 621-625). Daily administration of native GIP to ob/ob mice for 14 days had no effect on any of the parameters measure, consistent with the very short half-life of the native GIP in vivo. 5 As expected, a key component of the beneficial action of N-AcGIP(LysPAL") concerned effects on beta-cells. Thus although native GIP is a weak stimulus to insulin secretion in ob/ob mice at the age studied, plasma and pancreatic insulin concentrations were raised in ob/ob mice receiving the novel fatty acid derivatised analogue. This is consistent with the action of GIP as a promoter of proinsulin gene 10 expression (Wang, Y. et al., 1996, Mol. Cell. Endocrinol. 116:81-87) and exemplifies the increased potency reported for N-terminally modified GIP analogues in this animal model of diabetes (Hinke, S.A. et al., 2002, Diabetes 51: 656-661; Gault, V.A. et al., 2002, Biochen. J 367: 913-920; Gault, V.A. et al., 2003, J Endocrinol. 176: 133-141; O'Harte, F.P.M. et al., 1999, Diabetes 48: 758-765; O'Harte, F.P.M. et al., 15 2002, Diabetologia 45: 1281-1291). Furthermore, the insulin response to glucose was significantly enhanced in ob/ob mice receiving N-AcGIP(LysPAL1 7 ). This ability to augment or restore pancreatic beta cell glucose responsiveness has been similarly observed with GLP-1 (Holz, G.G. et al., 1993, Nature 28: 362-365; Flamez, D. et al., 1998, Diabetes 47: 646-652). As with observations on glycaemic control, none of 20 these attributes were reproduced by daily injections of native GIP. Results of insulin sensitivity tests conducted after 14 days treatment indicate that the improvement of diabetic status achieved in ob/ob mice with N AcGIP(LysPAL 37 ) was not solely due to the potentiation of insulin secretion. Thus, these animals also exhibited a significant improvement of insulin sensitivity compared 25 to the GIP or saline treated groups. Given that hyperinsulinemia is generally believed to down-regulate insulin receptor function (Marshall, S. et al., 1981, Diabetes 30: 746-753), this suggests that N-AcGIP(LysPAL 37 ) may exert other compensatory effects. Further study is necessary to evaluate this aspect but possibilities include inhibition of counter-regulatory hormones and effects on extrapancreatic sites such as 30 muscle, adipose tissue and liver (Morgan, L.M. et al., 1996, Biochem. Soc. Trans. 24:585-591; O'Harte, F.P.M. et al., 1998, J. Endocrinol. 156: 237-243; Yip, R.G. et al., 1998, Endocrinology 139: 4004-4007). - 27 - WO 2005/082928 PCT/GB2005/000710 Irrespective of knowledge of the full range of actions contributing to the antihyperglycaemic effect of N-AcGIP(LysPAL 37 ), a currently envisaged problem of long-term treatment with stable analogues of GIP or GLP-1 concerns desensitization of hormone receptor action (Delmeire, D. et al., 2004, Biochemn. Pharmacol. 68: 33 5 39). Although this has been observed during prolonged exposure of pancreatic beta cells to GIP in rats (Tseng, C.C. et al., 1996, Am. J Physiol. 270: E661-E666), there was no evidence that treatment with N-AcGIP(LysPAL 37 ) for 14 days compromised the glucose lowering or insulin releasing actions of N-AcGIP(LysPAL"). Thus the antidiabetic actions of N-AcGIP(LysPAL 37 ) were clearly evident when the analogue 10 was administered acutely together with glucose. Furthermore, the acute effects of N AcGIP(LysPAL 37 ) in such experiments were identical in groups of ob/ob mice receiving either N-AcGIP(LysPAL 37 ), native GIP or saline injections for 14 days. Such data clearly indicate that prolonged exposure to N-AcGIP(LysPAL 37 ) does not induce and possibly overcomes inherent GIP receptor desensitization in 15 ob/ob mice. Given the high circulating concentrations of GIP in these obese-diabetic rodents (Flatt, P.R. et al., 1983, Diabetes 32: 433-435; Flatt, P.R. et al., 1984, J Endocrinol. 101: 249-256), it is tempting to link beta cell refractoriness to GIP evident in ob/ob mice to simple receptor desensitization at the hands of inappropriate secretion and metabolism of GIP. 20 The data shown herein demonstrate that N-terminally acetylated GIP carrying a palmitate group linked to Lys at position 37 displays resistance to DPP IV and an impressive profile of bioactivity manifested by potent and long-acting glucose lowering activity in a commonly employed animal model of obesity-diabetes. This activity profile provides strong encouragement for the development of long-acting 25 fatty acid derivatised N-terminally modified analogues of GIP for the once-daily treatment of type 2 diabetes. The peptide analogues of the present invention have use in treating diseases and conditions caused by improper modulation of insulin levels, including diabetes, type 2 diabetes, insulin resistance, insulin resistant metabolic syndrome (Syndrome 30 X), and obesity. A peptide analogue produced by the methods of the present invention can be used in a pharmaceutical composition, wherein the analogue is combined with a pharmaceutically acceptable carrier. Such a composition may also contain (in - 28 - WO 2005/082928 PCT/GB2005/000710 addition to the analogue and a carrier) diluents, fillers, salts, buffers, stabilizers, solubilizers, and other materials well known in the art. The term "pharmaceutically acceptable" means a non-toxic material that does not interfere with the effectiveness of the biological activity of the active ingredient(s). The characteristics of the carrier 5 will depend on the route of administration. Administration of the peptide analogue of the present invention used in the pharmaceutical composition or to practice the method of the present invention can be carried out in a variety of conventional ways, such as by oral ingestion, inhalation, topical application or cutaneous, subcutaneous, intraperitoneal, parenteral or 10 intravenous injection. Administration can be internal or external; or local, topical or systemic. The compositions containing a peptide analogue of this invention can be administered intravenously, as by injection of a unit dose, for example. The term "unit dose" when used in reference to a therapeutic composition of the present 15 invention refers to physically discrete units suitable as unitary dosage for the subject, each unit containing a predetermined quantity of active material calculated to produce the desired therapeutic effect in association with the required diluent, i.e., carrier or vehicle. Formulations suitable for parenteral administration include aqueous and non 20 aqueous sterile injection solutions which may contain anti-oxidants, buffers, bacteriostats and solutes which render the formulation isotonic with the blood of the intended recipient; and aqueous and non-aqueous sterile suspensions which may include suspending agents and thickening agents. The fonnulations may be presented in unit-dose or multi-dose containers, for example, sealed ampules and vials, and may 25 be stored in a freeze-dried (lyophilized) condition requiring only the addition of the sterile liquid carrier, for example, water for injections, immediately prior to use. Extemporaneous injection solutions and suspensions may be prepared from sterile powders, granules and tablets of the kind previously described. When a therapeutically effective amount of the composition of the present 30 invention is administered orally, the composition of the present invention will be in the form of a tablet, capsule, powder, solution or elixir. When administered in tablet form, the pharmaceutical composition of the invention may additionally contain a solid carrier such as a gelatin or an adjuvant. The tablet, capsule, and powder contain - 29 - WO 2005/082928 PCT/GB2005/000710 from about 5 to 95% protein of the present invention, and preferably from about 25 to 90% protein of the present invention. When administered in liquid form, a liquid carrier such as water, petroleum, oils of animal or plant origin such as peanut oil, mineral oil, soybean oil, or sesame oil, or synthetic oils may be added. The liquid 5 form of the pharmaceutical composition may further contain physiological saline solution, dextrose or other saccharide solution, or glycols such as ethylene glycol, propylene glycol or polyethylene glycol. When administered in liquid form, the pharmaceutical composition contains from about 0.5 to 90% by weight of the composition of the present invention, and preferably from about 1 to 50% of the 10 composition of the present invention. When a therapeutically effective amount of the composition of the present invention is administered by intravenous, cutaneous or subcutaneous injection, the composition of the present invention will be in the form of a pyrogen-free, parenterally acceptable aqueous solution. The preparation of such parenterally 15 acceptable protein solutions, having due regard to pH, isotonicity, stability, and the like, is within the skill in the art. A preferred pharmaceutical composition for intravenous, cutaneous, or subcutaneous injection should contain, in addition to the composition of the present invention, an isotonic vehicle such as Sodium Chloride Injection, Ringer's Injection, Dextrose Injection, Dextrose and Sodium Chloride 20 Injection, Lactated Ringer's Injection, or other vehicle as known in the art. The pharmaceutical composition of the present invention may also contain stabilizers, preservatives, buffers, antioxidants, or other additives known to those of skill in the art. Use of timed release or sustained release delivery systems are also included. 25 A sustained-release matrix, as used herein, is a matrix made of materials, usually polymers, which are degradable by enzymatic or acid/base hydrolysis or by dissolution. Once inserted into the body, the matrix is acted upon by enzymes and body fluids. The sustained-release matrix desirably is chosen from biocompatible materials such as liposomes, polylactides (polylactic acid), polyglycolide (polymer of 30 glycolic acid), polylactide co-glycolide (co-polymers of lactic acid and glycolic acid) polyanhydrides, poly(ortho)esters, polyproteins, hyaluronic acid, collagen, chondroitin sulfate, carboxylic acids, fatty acids, phospholipids, polysaccharides, nucleic acids, polyamino acids, amino acids such as phenylalanine, tyrosine, -30- WO 2005/082928 PCT/GB2005/000710 isoleucine, polynucleotides, polyvinyl propylene, polyvinylpyrrolidone and silicone. A preferred biodegradable matrix is a matrix of one of either polylactide, polyglycolide, or polylactide co-glycolide (co-polymers of lactic acid and glycolic acid). 5 The therapeutic compositions can include pharmaceutically acceptable salts of the components therein, e.g., which may be derived from inorganic or organic acids. By "pharmaceutically acceptable salt" is meant those salts which are, within the scope of sound medical judgement, suitable for use in contact with the tissues of humans and lower animals without undue toxicity, irritation, allergic response and the like and 10 are commensurate with a reasonable benefit/risk ratio. Pharmaceutically acceptable salts are well-known in the art. For example, S. M. Berge, et al., describe pharmaceutically acceptable salts in detail in J Pharmaceutical Sciences (1977) 66:1 et seq., which is incorporated herein by reference in its entirety. Pharmaceutically acceptable salts include the acid addition salts (formed with the free amino groups of 15 the polypeptide) that are formed with inorganic acids such as, for example, hydrochloric or phosphoric acids, or such organic acids as acetic, tartaric, mandelic and the like. Salts formed with the free carboxyl groups can also be derived from inorganic bases such as, for example, sodium, potassium, ammonium, calcium or ferric hydroxides, and such organic bases as isopropylamine, trimethylamine, 2 20 ethylamino ethanol, histidine, procaine and the like. The salts may be prepared in situ during the final isolation and purification of the compounds of the invention or separately by reacting a free base function with a suitable organic acid. Representative acid addition salts include, but are not limited to acetate, adipate, alginate, citrate, aspartate, benzoate, benzenesulfonate, bisulfate, butyrate, 25 camphorate, camphorsufonate, digluconate, glycerophosphate, hemisulfate, heptonoate, hexanoate, fumarate, hydrochloride, hydrobromide, hydroiodide, 2 hydroxymethanesulfonate (isethionate), lactate, maleate, methanesulfonate, nicotinate, 2-naphthalenesulfonate, oxalate, pamoate, pectinate, persulfate, 3-phenylpropionate, picrate, pivalate, propionate, succinate, tartate, thiocyanate, phosphate, glutamate, 30 bicarbonate, p-toluenesulfonate and undecanoate. Also, the basic nitrogen-containing groups can be quaternized with such agents as lower alkyl halides such as methyl, ethyl, propyl, and butyl chlorides, bromides and iodides; dialkyl sulfates like dimethyl, diethyl, dibutyl, and diamyl sulfates; long chain halides such as decyl, -31 - WO 2005/082928 PCT/GB2005/000710 lauryl, myristyl and stearyl chlorides, bromides and iodides; arylalkyl halides like benzyl and phenethyl bromides and others. Water or oil-soluble or dispersible products are thereby obtained. Examples of acids which may be employed to form pharmaceutically acceptable acid addition salts include such inorganic acids as 5 hydrochloric acid, hydrobromic acid, sulphuric acid and phosphoric acid and such organic acids as oxalic acid, maleic acid, succinic acid and citric acid. As used herein, the terms "pharmaceutically acceptable", "physiologically tolerable" and grammatical variations thereof as they refer to compositions, carriers, diluents and reagents, are used interchangeably and represent that the materials are 10 capable of administration to or upon a mammal with a minimum of undesirable physiological effects such as nausea, dizziness, gastric upset and the like. The preparation of a pharmacological composition that contains active ingredients dissolved or dispersed therein is well understood in the art and need not be limited based on formulation. Typically such compositions are prepared as injectables either 15 as liquid solutions or suspensions, however, solid forms suitable for solution, or suspensions, in liquid prior to use can also be prepared. The preparation can also be emulsified. The active ingredient can be mixed with excipients which are pharmaceutically acceptable and compatible with the active ingredient and in amounts 20 suitable for use in the therapeutic methods described herein. Suitable excipients include, for example, water, saline, dextrose, glycerol, ethanol or the like and combinations thereof. In addition, if desired, the composition can contain minor amounts of auxiliary substances such as wetting or emulsifying agents, pH buffering agents and the like which enhance the effectiveness of the active ingredient. 25 The amount of peptide analogue of the present invention in the pharmaceutical composition of the present invention will depend upon the nature and severity of the condition being treated, on the nature of prior treatments which the patient has undergone, and on a variety of other factors, including the type of injury, the age, weight, sex, medical condition of the individual. Ultimately, the attending physician 30 will decide the amount of the analogue with which to treat each individual patient. Initially, the attending physician will administer low doses of peptide analogue and observe the patient's response. Larger doses of peptide analogue may be - 32 - WO 2005/082928 PCT/GB2005/000710 administered until the optimal therapeutic effect is obtained for the patient, and at that point the dosage is not increased further. Additional guidance on methods of determining dosages can be found in standard references, for example, Spilker, Guide to Clinical Studies and Developing 5 Protocols, Raven Press Books, Ltd., New York, 1984, pp. 7-13 and 54-60; Spilker, Guide to Clinical Trials, Raven Press, Ltd., New York, 1991, pp. 93-101; Craig et al., Modern Pharmacology, 2d ed., Little Brown and Co., Boston, 1986, pp. 127-133; Speight, Aveiy's Drug Treatment: Principles and Practices of Clinical Pharmacology and Therapeutics, 3d ed., Williams and Wilkins, Baltimore, 1987, pp. 10 50-56; Tallarida et al., Principles in General Pharmacology, Springer-Verlag, New York, 1998, pp. 18-20; and Olson, Clinical Pharmacology Made Ridiculously Simple, MedMaster, Inc., Miami, 1993, pp. 1-5. EXAMPLES 15 Example 1. Preparation of N-terminally modified GIP and analogues thereof. The N-terminal modification of GIP is essentially a three step process. Firstly, GIP is synthesized from its C-terminal (starting from a Fmoc-Gln (Trt)-Wang resin (Calbiochem Novabiochem,.Beeston, Nottingham, UK) up to the penultimate N terminal amino-acid (Ala 2 ) on an automated peptide synthesizer (Applied Biosystems, 20 California, USA). The synthesis follows standard Fmoc peptide chemistry protocols. Secondly, the N-terminal amino acid of native GIP (Tyr) is added to a manual bubbler system as a Fmoc-protected Tyr(tBu)-Wang resin. This amino acid is deprotected at its N-terminus (piperidine in DMF (20% v/v)) and allowed to react with a high concentration of glucose (glycation, under reducing conditions with sodium 25 cyanoborohydride), acetic anhydride (acetylation), pyroglutamic acid (pyroglutamyl) etc. for up to 24 hours as necessary to allow the reaction to go to completion. The completeness of reaction is monitored using the ninhydrin test which determines the presence of available free a-amino groups. Thirdly (once the reaction is complete), the now structurally modified Tyr is cleaved from the Wang resin (95% TFA, and 5% 30 of the appropriate scavengers. N.B. Tyr is considered to be a problematic amino acid and may need special consideration) and the required amount of N-terminally modified-Tyr consequently added directly to the automated peptide synthesiser, which will carry on the synthesis, thereby stitching the N-terminally modified-Tyr to the a- WO 2005/082928 PCT/GB2005/000710 amino of GIP (Ala 2 ), completing the synthesis of the GIP analogue. This peptide is cleaved off the Wang resin (as above) and then worked up using the standard Buchner filtering, precipation, rotary evaporation and drying techniques. 5 Example 2. Preparation of Tyr 1 -Glucitol GIP and Its Properties in vivo. The following example investigates preparation of TyrIglucitol GIP together with evaluation of its antihyperglycemic and insulin-releasing properties in vivo. The results clearly demonstrate that this novel GIP analogue exhibits a substantial resistance to aminopeptidase degradation and increased glucose lowering activity 10 compared with the native GIP. Research Design and Methods Materials. Human GIP was purchased from the American Peptide Company (Sunnyvale, California, USA). HPLC grade acetonitrile was obtained from Rathburn 15 (Walkersburn, Scotland). Sequencing grade trifluoroacetic acid (TFA) was obtained from Aldrich (Poole, Dorset, UK). All other chemicals purchased including dextran T-70, activated charcoal, sodium cyanoborohydride and bovine serum albumin fraction V were from Sigma (Poole, Dorset, UK). Diprotin A (DPA) was purchased from Calbiochem-Novabiochem (UK) Ltd. (Beeston, Nottingham, UK) and rat insulin 20 standard for RIA was obtained from Novo Industria (Copenhagen, Denmark). Reversed-phase Sep-Pak cartridges (C-18) were purchased from Millipore-Waters (Milford, MA, USA). All water used in these experiments was purified using a Milli Q, Water Purification System (Millipore Corporation, Milford, Massachusetts, USA). 25 Preparation of Tyrl-glucitol GIP. Human GIP was incubated with glucose under reducing conditions in 10 mmol/l sodium phosphate buffer at pH 7.4 for 24 hours. The reaction was stopped by addition of 0.5 mol/l acetic acid (30 ptl) and the mixture applied to a Vydac (C18)(4.6 x 250mm) analytical HPLC column (The Separations Group, Hesperia, California, USA) and gradient elution conditions were established 30 using aqueous/TFA and acetonitrile/TFA solvents. Fractions corresponding to the glycated peaks were pooled, taken to dryness under vacuum using an AES 1000 Speed-Vac concentrator (Life Sciences International, Runcorn, UK) and purified to -34- WO 2005/082928 PCT/GB2005/000710 homogeneity on a Supelcosil (C-8) (4.6 x 150mm) column (Supelco Inc., Poole, Dorset, UK). Degradation of GIP and Tyr'-glucitol GIP by DPP IV. HPLC-purified GIP or Tyr' 5 glucitol GIP were incubated at 37 0 C with DPP-IV (5mU) for various time periods in a reaction mixture made up to 500 pl with 50 mmol/l triethanolamine-HC1, pH 7.8 (final peptide concentration 1 umol/l). Enzymatic reactions were terminated after 0, 2, 4 and 12 hours by addition of 5 pl of 10% (v/v) TFA/water. Samples were made up to a final volume of 1.0 ml with 0.12% (v/v) TFA and stored at -20'C prior to HPLC 10 analysis. Degradation of GIP and Tyr'-glucitol GIP by human plasma. Pooled human plasma (20 p1) taken from six healthy fasted human subjects was incubated at 37'C with GIP or Tyr'-glucitol GIP (10 pg) for 0 and 4 hours in a reaction mixture made up to 500 15 pl, containing 50 mmol/l triethanolamine/HCL buffer pH 7.8. Incubations for 4 hours were also performed in the presence of diprotin A (5 mU). The reactions were terminated by addition of 5 [1 of TFA and the final volume adjusted to 1.0 ml using 0.1% v/v TFA/water. Samples were centrifuged (13,000g, 5 minutes) and the supernatant applied to a C- 18 Sep-Pak cartridge (Millipore-Waters) which was 20 previously primed and washed with 0.1% (v/v) TFA/water. After washing with 20 ml 0.12% TFA/water, bound material was released by elution with 2 ml of 80% (v/v) acetonitrile/water and concentrated using a Speed-Vac concentrator (Runcorn, UK). The volume was adjusted to I.Oml with 0.12% (v/v) TFA/water prior to HPLC purification. 25 -35- WO 2005/082928 PCT/GB2005/000710 HPLC analysis of degraded GIP and Tyr'-glucitol GIP. Samples were applied to a Vydac C-18 widepore column equilibriated with 0.12% (v/v) TFA/H 2 0 at a flow rate of 1.0 ml/minute. Using 0.1% (v/v) TFA in 70% acetonitrile/H 2 0, the concentration of acetonitrile in the eluting solvent was raised from 0% to 31.5% over 15 min, to 5 38.5% over 30 minutes and from 38.5% to 70% over 5 minutes, using linear gradients. The absorbance was monitored at 206 nm and peak areas evaluated using a model 2221 LKB integrator. Samples recovered manually were concentrated using a Speed-Vac concentrator. 10 Electrospray ionization mass spectroinetry (ES-MS). Samples for ESI-MS analysis containing intact and degradation fragments of GIP (from DPP IV and plasma incubations) as well as Tyr'-glucitol GIP, were further purified by HPLC. Peptides were dissolved (approximately 400 pmol) in 100 gl of water and applied to the LCQ benchtop mass spectrometer (Finnigan MAT, Hemel Hempstead, UK) equipped with 15 a microbore C-1 8 HPLC column (150 x 2.0mm, Phenomenex, Ltd., Macclesfield, UK). Samples ( 3 0lV direct loop injection) were injected at a flow rate of 0.2ml/min, under isocratic conditions 35% (v/v) acetonitile/water. Mass spectra were obtained from the quadripole ion trap mass analyzer and recorded. Spectra were collected using full ion scan mode over the mass-to-charge (m/z) range 150-2000. The 20 molecular masses of GIP and related structures were determined from ESI-MS profiles using prominent multiple charged ions and the following equation Mr = iMi - iMi 1 where Mr = molecular mass; Mi = m/z ratio; i = number of charges; MI, = mass of a proton. 25 In vivo biological activity of GIP and Tyr'-glucitol GIP. Effects of GIP and Tyr glucitol GIP on plasma glucose and insulin concentrations were examined using 10 12 week old male Wistar rats. The animals were housed individually in an air conditioned room and 22 2 0 C with a 12 hour light/12 hour dark cycle. Drinking 30 water and a standard rodent maintenance diet (Trouw Nutrition, Belfast, Northern Ireland) were supplied ad libitui. Food was withdrawn for an 18 hour period prior to intraperitoneal injection of glucose alone (1 8mmol/kq body weight) or in combination with either GIP or TyrIglucitol GIP (10 nmol/kg). Test solutions were administered - 36 - WO 2005/082928 PCT/GB2005/000710 in a final volume of 8 ml/kg body weight. Blood samples were collected at 0, 15, 30 and 60 minutes from the cut tip of the tail of conscious rats into chilled fluoride/heparin microcentrifuge tubes (Sarstedt, Numbrecht, Germany). Samples were centrifuged using a Beckman microcentrifuge for about 30 seconds at 13,000 g. 5 Plasma samples were aliquoted and stored at -20'C prior to glucose and insulin determinations. All animal studies were done in accordance with the Animals (Scientific Procedures) Act 1986. Analyses. Plasma glucose was assayed by an automated glucose oxidase procedure 10 using a Beckman Glucose Analyzer II [33]. Plasma insulin was determined by dextran charcoal radioimmunoassay as described previously [34]. Incremental areas under plasma glucose and insulin area under the curve (AUC) were calculated using a computer program (CAREA) employing the trapezoidal rule [35] with baseline subtraction. Results are expressed as mean = SEM and values were compared using 15 the Student's unpaired t-test. Groups of data were considered to be significantly different if P<0.05. Degradation of GIP and Tyr'-glucitol GIP by DPP IV Fig. 1 illustrates the typical peak profiles obtained from the HPLC separation of the products obtained from the 20 incubation of GIP (Fig I a) or Tyr'-glucitol GIP (Fig 1 b) with DPP IV for 0, 2, 4 and 12 hours. The retention times of GIP and Tyr]-glucitol GIP at t=0 were 21.93 minutes and 21.75 minutes respectively. Degradation of GIP was evident after 4 hours incubation (54% intact), and by 12 hours the majority (60%) of intact GIP was converted to the single product with a retention time of 21.61 minutes. Tyr 1 -glucitol 25 GIP remained almost completely intact throughout 2-12 hours incubation. Separation was on a Vydac C-18 column using linear gradients of 0% to 31.5% acetonitrile over 15 minutes, to 38.5% over 30 minutes and from 38.5 to 70% acetonitrile over 5 minutes. 30 Degradation of GIP and Tyr'-glucitol GIP by human plasma. Fig. 2 shows a set of typical HPLC profiles of the products obtained from the incubation of GIP or Tyr' glucitol GIP with human plasma for 0 and 4 hours. GIP (Fig 2a) with a retention time of 22.06 minutes was readily metabolised by plasma within 4 hours incubation giving -37- WO 2005/082928 PCT/GB2005/000710 rise to the appearance of a major degradation peak with a retention time of 21.74 minutes. In contrast, the incubation of Tyr 1 -glucitol GIP under similar conditions (Fig 2b) did not result in the formation of any detectable degradation fragments during this time with only a single peak being observed with a retention time of 21.77 minutes. 5 Addition of diprotin A, a specific inhibitor of DPP IV, to GIP during the 4 hours incubation completely inhibited degradation of the peptide by plasma. Peaks corresponding with intact GIP, GIP (3-42) and Tyr 1 -glucitol GIP are indicated. A major peak corresponding to the specific DPP IV inhibitor tripeptide DPA appears in the bottom peanels with retention time of 16-29 minutes. 10 Identification of GIP degradation fragments by ESI-MS. Fig. 3 shows the monoisotopic molecular masses obtained for GIP (Fig. 3A), Tyr'-glucitol GIP (Fig. 3B) and the major plasma degradation fragment of GIP (Fig. 3C) using ESI-MS. The peptides analyzed were purified from plasma incubations as shown in Fig. 2. Peptides 15 were dissolved (approximately 400 pmol) in I 00d of water and applied to the LC/MS equipped with a microbore C-i 8 HPLC column. Samples (30R1 direct loop injection) were applied at a flow rate of 0.2m1/min, under isocratic conditions 35% acetonitrile/water. Mass spectra were recorded using a quadripole ion trap mass analyzer. Spectra were collected using full ion scan mode over the mass-to-charge 20 (m/z) range 150-2000. The molecular masses (Mr) of GIP and related structures were determined from ESI-MS profiles using prominent multiple charged ions and the following equation Mr = iMi - iM 1 ,. The exact molecular mass (Mr) of the peptides were calculated using the equation Mr = iMi - iMp, as defined above in Research Design and Methods. After spectral averaging was performed, prominent multiple 25 charges species (M+3H )3+ and (M+4H )4+ were detected from GIP at m/z 1661.6 and 1246.8, corresponding to intact Mr 4981.8 and 4983.2 Da, respectively (Fig. 3A). Similarly, for TyrI-glucitol GIP ( (M+4H )+ and (M+5H) 5 *) were detected at m/z 1287.7 and 1030.3, corresponding to intact molecular masses Of Mr 5146.8 and 5146.5 Da, respectively (Fig. 3B). The difference between the observed molecular 30 masses of the quadruply charged GIP and the N-terminally modified GIP species (163.6 Da) indicated that the latter peptide contained a single glucitol adduct corresponding to Tyr'-glucitol GIP. Fig. 3C shows the prominent multiply charged species (M+3H )3+ and (M+4H) 4 + detected from the major fragment of GIP at m/z -38- WO 2005/082928 PCT/GB2005/000710 1583.8 and 1188.1, corresponding to intact Mr 4748.4 and 4748 Da, respectively (Fig. 3C). This corresponds with the theoretical mass of the N-terminally truncated form of the peptide GIP(3-42). This fragment was also the major degradation product of DPP IV incubations (data not shown). 5 Effects of GIP and Tyr -glucitol GIP on plasma glucose homeostasis. Figs. 4 and 5 show the effects of intraperitoneal (ip) glucose alone (1 8mmol/kg) (control group), and glucose in combination With GIP or Tyr' glucitol GIP (10nmol/kg) on plasma glucose and insulin concentrations. 10 Fig. 4A shows plasma glucose concentrations after i.p. glucose alone (1 8mmol/kg) (control group), or glucose in combination with either GIP of Tyr 1 glucitol GIP (1Onmol/kg). The time of injection is indicated by the arrow (0 minutes). Fig. 4B shows plasma glucose AUC values for 0-60 minutes post injection. Values are mean SEM for six rats. **P<0.0, <0.001 compared with GIP and Tyr' 15 glucitol GIP; tP<0.05, ttP<0.01 compared with non-glucated GIP. Fig. 5A shows plasma insulin concentrates after i.p. glucose along (18 mmol/kg) (control group), or glucose in combination with either with GIP or glycated GIP (10nmol/kq). The time of injection is indicated by the arrow. Fig. 5B shows plasma insulin AUC values were calculated for each of the 3 groups up to 90 minutes post injection. The time of 20 injection is indicated by the arrow (0 minutes). Plasma insulin AUC values for 0-60 minutes post injection. Values are mean ± SEM for six rats. *P<0.05, **P<0.001 compared with GIP and Tyr'glucitol GIP; tP<0.05, ttP<0.01 compared with non glycated GIP. Compared with the control group, plasma glucose concentrations and area 25 under the curve (AUC) were significantly lower following administration of either GIP or Tyr 1 -glucitol GIP (Figs 4A, B). Furthermore, individual values at 15 and 30 minutes together with AUC were significantly lower following administration of Tyr'-glucitol GIP as compared to GIP. Consistent with the established insulin releasing properties of GIP, plasma insulin concentrations of both peptide-treated 30 groups were significantly raised at 15 and 30 minutes compared with the values after administration of glucose alone (Fig. 5A). The overall insulin responses, estimated as AUC were also significantly greater for the two peptide-treated groups (Fig. 5B). Despite lower prevailing glucose concentrations than the GIP-treated group, plasma - 39 - WO 2005/082928 PCT/GB2005/000710 insulin response, calculated as AUC, following Tyr 1 -glucitol GIP was significantly greater than after GIP (Fig. 5B). The significant elevation of plasma insulin at 30 minutes is of particular note, suggesting that the insulin-releasing action of Tyr I glucitol GIP is more protracted than GIP even in the face of a diminished glycemic 5 stimulus (Figs. 4A, 5A). Example 3. Additional N-Terminal Structural Modifications of GIP. This example further looked at the ability of additional N-terminal structural modifications of GIP in preventing inactivation by DPP and in plasma and their 10 associated increase in both the insulin-releasing potency and potential therapeutic value. Native human GIP, glycated GIP, acetylated GIP and a number of GIP analogues with N-terminal amino acid substitutions were tested. Materials and Methods. High-performance liquid chromatography (HPLC) grade 15 acetonitrile was obtained from Rathbum (Walkersburn, Scotland). Sequencing grade trifluoroacetic acid (TFA) was obtained from Aldrich (Poole, Dorset, UK). Dipeptidyl peptidase IV was purchased from Sigma (Poole, Dorset, UK), and Diprotin A was purchased from Calbiochem Novabiochem (Beeston, Nottingham, UK). RPMI 1640 tissue culture medium, foetal calf serum, penicillin and streptomycin were all 20 purchased from Gibco (Paisley, Strathclyde, UK). All water used in these experiments was purified using a Milli-Q, Water Purification System (Millipore, Milford, Massachusetts, USA). All other chemicals used were of the highest purity available. 25 Synthesis of GIP and N-terminally modified GIP analogues. GIP, GIP(Abu 2 ), GIP(Sar2), GIP(Ser2), GIP(Gly 2 ) and GIP(Pro) were sequentially synthesized on an Applied Biosystems automated peptide synthesizer (model 432A) using standard solid-phase Fmoc procedure, starting with an Fmoc-Gln-Wang resin. Following cleavage from the resin by trifluoroacetic acid: water, thioanisole, ethanedithiol 30 (90/2.5/5/2.5, a total volume of 20 ml/g resin), the resin was removed by filtration and the filtrate volume was decreased under reduced pressure. Dry diethyl ether was slowly added until a precipitate was observed. The precipitate was collected by low speed centrifugation, resuspended in diethyl ether and centrifuged again, this -40 - WO 2005/082928 PCT/GB2005/000710 procedure being carried out at least five times. The pellets were then dried in vacuo and judged pure by reversed-phase HPLC on a Waters Millennium 2010 chromatography system (Software version 2.1.5.). N-terminal glycated and acetylated GIP were prepared by minor modification of a published method. 5 Electrospray ionization-mass spectrometry (ESI-MS) was carried out as described in Example 2. Degradation of GIP and novel GIP analogues by DPP IV and human plasma was carried out as described in Example 2. Culture of insulin secreting cells. BRIN-BD 11 cells [30] were cultured in sterile 10 tissue culture flasks (Corning, Glass Works, UK) using RPMI- 1640 tissue culture medium containing 10% (v/v) foetal calf serum, 1% (v/v) antibiotics (100 U/ml penicillin, 0.1 mg/ml streptomycin) and 11.1 mM glucose. The cells were maintained at 37 0 C in an atmosphere of 5% CO 2 and 95% air using a LEEC incubator (Laboratory Technical Engineering, Nottingham, UK). 15 Acute testsfor insulin secretion. Before experimentation, the cells were harvested from the surface of the tissue culture flasks with the aid of trypsin/EDTA (Gibco), seeded into 24-multiwell plates (Nunc, Roskilde, Denmark) at a density of 1.5 x 10s cells per well, and allowed to attach overnight at 37'C. Acute tests for insulin release 20 were preceded by 40 minutes pre-incubation at 37'C in 1.0 ml Krebs Ringer bicarbonate buffer (115 mM NaCI, 4.7 mM KC1, 1.28 mM CaC1 2 , 1.2 mM KH 2

PO

4 , 1.2 mM MgSO 4 , 10 mM NaIC0 3 , 5 g/l bovine serum albumin, pH 7.4) supplemented with 1.1 mM glucose. Test incubations were performed (n=12) at two glucose concentrations (5.6 mM and 16.7 mM) with a range of concentrations (10-1 to 10-8 25 M) of GIP or GIP analogues. After 20 minutes incubation, the buffer was removed from each well and aliquots (200 pl) were used for measurement of insulin by radioimmunoassay [31]. Statistical analysis. Results are expressed as mean S.E.M. and values were 30 compared using the Student's unpaired t-test. Groups of data were considered to be significantly different if P< 05. -41- WO 2005/082928 PCT/GB2005/000710 Structural identification of GIP and GIP analogues by ESI-MS. The monoisotopic molecular masses of the peptides were determined using ESI-MS. After spectral averaging was performed, prominent multiple charged species (M+3H) 3 + and (M+4H)* were detected for each peptide. Calculated molecular masses confirmed 5 the structural identity of synthetic GIP and each of the N-terminal analogues. Degradation of GIP and novel GIP analogues by DPP-IV. Figs. 6-11 illustrate the typical peak profiles obtained from the HPLC separation of the reaction products obtained from the incubation of GIP, GIP(Abu 2 ), GIP(Sar 2 ), GIP(Ser 2 ), glycated GIP 10 and acetylated GIP with DPP IV, for 0, 2, 4, 8 and 24 hours. The results summarized in Table 1 indicate that glycated GIP, acetylated GIP, GIP(Ser 2 ) are GIP(Abu 2 ) more resistant than native GIP to in vitro degradation with DPP IV. From these data GIP(Sar 2 ) appears to be less resistant. 15 Table 1. Percent intact peptide remaining after incubation with DPPIV. Peptide % Intact peptide remaining after time (h) 0 2 4 8 24 GIP 1-42 100 52 i 1 23 1-1 0 0 Glycated GIP 100 100 100 100 100 GIP (Abu) 100 3811 28 2 0 0 GIP (Ser) 100 77+2 60 1 32+4 0 GIP (Sar) 100 28 2 8 0 0 N-Acetyl-GIP 100 100 100 100 Table represents the percentage of intact peptide (i.e., GIP 1-42) relative to the major degradation product GIP 3-42. Values were taken from HPLC traces performed in triplicate and the mean and S.E.M. values calculated. DPA is diprotin A, a specific inhibitor of DPPIV. 20 Degradation of GIP and GIP analogues by human plasma. Figs. 12-16 show a representative set of HPLC profiles obtained from the incubation of GIP and GIP analogues with human plasma for 0, 2, 4, 8 and 24 hours. Observations were also made after incubation for 24 hours in the presence of DPA. These results are 25 summarized in Table 2 are broadly comparable with DPP IV incubations, but - 42 - WO 2005/082928 PCT/GB2005/000710 conditions which more closely mirror in vivo conditions are less enzymatically severe. GIP was rapidly degraded by plasma. In comparison, all analogues tested exhibited resistance to plasma degradation, including GIP(Sar 2 ) which from DPP IV data appeared least resistant of the peptides tested. DPA substantially inhibited 5 degradation of GIP and all analogues tested with complete abolition of degradation in the cases of GIP(Abu 2 ), GIP(Ser 2 ) and glycated GIP. This indicates that DPP IV is a key factor in the in vivo degradation of GIP. Table 2. Percent intact peptide remaining after incubation with human plasma. Peptide % Intact peptide remaining after incubations with human plasma 0 2 4 8 24 DPA GIP 1-42 100 52 ±1 23 ±1 0 0 68± 2 Glycated GIP 100 100 100 100 100 100 GIP (Abu2) 100 38± 1 2812 0 0 100 GIP(Serz) 100 77+2 60 1 32 ±4 0 63± 3 GIP (Sar2) 100 28A2 8 0. 0 100 10 Table represents the percentage of intact peptide (i.e., GIP 1-42) relative to the major degradation product GIP 3-42. Values were taken from HPLC traces performed in triplicate and the mean and S.E.M. values calculated. DPA is diprotin A, a specific inhibitor of DPPIV. 15 Dose-dependent effects of GIP and novel GIP analogues on insulin secretion. Figs. 17-30 show the effects of a range of concentrations of GIP, GIP(Abu 2 ), GIP(Sar 2 ), GIP(Ser 2), acetylated GIP, glycated GIP, GIP(Gly 2 ) and GIP(Pro 3 ) on insulin secretion from BRIN-BD 11 cells at 5.6 and 16.7 mM glucose. Native GIP provoked a prominent and dose-related stimulation of insulin secretion. Consistent with 20 previous studies [28], the glycated GIP analogue exhibited a considerably greater insulinotropic response compared with native peptide. N-terminal acetylated GIP exhibited a similar pattern and the GIP(Ser2) analogue also evoked a strong response. From these tests, GIP(Gly 2 ) and GIP(Pro 3 ) appeared to be the least potent analogues in terms of insulin release. Other stable analogues tested, namely GIP(Abu 2 ) and 25 GIP(Sar 2 ), exhibited a complex pattern of responsiveness dependent on glucose - 43 - WO 2005/082928 PCT/GB2005/000710 concentration and dose employed. Thus very low concentrations were extremely potent under hyperglycemic conditions (16.7 mM glucose). This suggests that even these analogues may prove therapeutically useful in the treatment of type 2 diabetes where insulinotropic capacity combined with in vivo degradation dictates peptide 5 potency. Example 4. Glu 3 substituted GIP improves obesity-related insulin resistance and associated glucose intolerance. This example examines GIP receptor antagonism and obesity-related insulin 10 resistance and associated glucose intolerance using a Glu 3 -substituted form of GIP, namely, (Pro 3 )GIP. Cell lines and animals. In vitro insulin secretion was evaluated using the clonal pancreatic beta-cell line, BRIN-BDI 1 (McClenaghan, N.H. et al., 1996, Diabetes 15 45:1132-1140). In vitro cyclic AMP generation was measured using Chinese hamster lung (CHL) fibroblast cells stably transfected with the human GIP receptor (Gremlich, S. et al., 1995, Diabetes 44:1202-1208). In vivo studies were conducted in 8-12 week-old obese diabetic ob/ob mice (Bailey C.J. et al., 1982, Int. J Obesity 6:11-21) and normal control mice. 20 Peptide synthesis and characterisation. Glu 3 -substituted analogues were sequentially synthesised on an Applied Biosystems automated peptide synthesiser (Model 432A) using standard solid-phase Fmoc peptide chemistry (Fields, G.B. et al., 1990, int. J Pept. Protein Res. 35:161-214), from a pre-loaded Fmoc-Gln-Wang resin. The 25 synthetic peptides were judged pure by reversed-phase HPLC on a Waters Millenium 2010 chromatography system (Software version 2.1.5). The molecular masses of the purified peptide analogues were determined using Matrix Assisted Laser Desorption lonisation-Time of Flight (MALDI-TOF) mass spectrometry. Samples were dissolved in 10 pl H 2 0 (approximately 40 pmol/l), placed on a stainless steel sample 30 plate and allowed to dry at room temperature. Samples were then mixed with a matrix solution (10 mg/ml solution of a-cyano-4-hydroxycinnamic acid) in acetonitrile/ethanol (1/1) and allowed to dry at room temperature. The molecular masses were then recorded as mass-to-charge (m/z) ratio versus relative peak intensity - 44 - WO 2005/082928 PCT/GB2005/000710 and compared using theoretical values on a Voyager-DE BioSpectrometry Workstation (PerSeptive Biosystems, Framingham, MA, USA). Tissue culture. Chinese hamster lung (CHL) fibroblast cells stably transfected with 5 the human GIP receptor were cultured in DMEM tissue culture medium containing 10% (v/v) foetal bovine serum, 1% (v/v) antibiotics (100 U/il penicillin, 0.1 mg/ml streptomycin). BRIN-BD 11 cells were cultured using RPMI- 1640 tissue culture medium containing 10% (v/v) foetal bovine serum, 1% (v/v) antibiotics (100 U/ml penicillin, 0.1 mg/ml streptomycin). Cells were maintained in sterile tissue culture 10 flasks (Corning Glass Works, Sunderland, UK) at 37*C in an atmosphere of 5% CO 2 and 95% air using an LEEC incubator (Laboratory Technical Engineering, Nottingham, UK). Acute studies of insulin release. Insulin release from BRIN-BD I1 cells was 15 determined using cell monolayers (McClenaghan, N.H. et al., 1996, Diabetes 45:1132-1140). Cells were harvested with the aid of trypsin/EDTA (Gibco), seeded into 24-multiwell plates (Nunc, Roskilde, Denmark) at a density of 1.0 x 105 cells per well, and allowed to attach overnight at 37'C. Prior to acute test, cells were preincubated for 40 minutes at 37'C in 1.0 ml Krebs Ringer bicarbonate buffer (115 20 mM NaCl, 4.7 mM KCl, 1.28 mM CaCl 2 , 1.2 mM KH 2

PO

4 , 1.2 mM MgSO4, 10 mM NaHCO 3 , 0.5% (w/v) bovine serum albumin, pH 7.4) supplemented with 1.1 mM glucose. Acute tests for insulin release were performed for 20 minutes at 37'C at 5.6 mM glucose using various concentrations of Glu 3 -substituted analogues and GIP(3 42) in the presence of native GIP (10-7 M) as indicated in the Figures. After 25 incubation, aliquots of buffer were removed and stored at -20 0 C for insulin radioimmunoassay (Flatt, P.R. et al., 1981, Diabetologia 20:573-577). Acute studies ofcyclic AMP generation. GIP receptor transfected CHL cells were seeded into 12-well plates (Nunc, Roskilde, Denmark) at a density of 1.0 x 105 cells 30 per well. The cells were then allowed to grow for 48 hours before being loaded with tritiated adenine (2 pCi; TRK3 11, Amersham, Buckinghamshire, UK) and incubated at 37*C for 6 hours in 1 ml DMEM, supplemented with 0.5% (w/v) foetal bovine serum. The cells were then washed twice with HBS buffer (130 mM NaCI, 20 mM -45 - WO 2005/082928 PCT/GB2005/000710 HEPES, 0.9 mM NaHPO4, 0.8 mM MgSO 4 , 5.4 mM KCl, 1.8 mM CaCl2, 25 mM glucose, 25 RM phenol red, pH 7.4). The cells were then exposed for 10 minutes at 37'C to forskolin (FSK, 10 pM) or varying concentrations of (Pro 3 )GIP in the absence (control) or presence of native GIP (10-7 M). After removal of the medium, 5 cells were lysed with I ml of 5% trichloroacetic acid (TCA) containing 0.1 mM unlabelled cAMP and 0.1 mM unlabelled ATP. The intracellular tritiated cAMP was then separated on Dowex and alumina exchange resins as previously described (Widmann, C. et al., 1993, Mol. Pharmacol. 45:1029-1035). 10 Acute in vivo effects of (Pro )GIP administration in obese diabetic ob/ob mice. Plasma glucose and insulin responses were evaluated using 8- to 12-week old obese diabetic ob/ob mice following intraperitoneal (i.p.) injection of native GIP, (Pro 3 )GIP (25 nmol/kg body weight) or saline (0.9% (w/v) NaCl; control) immediately following the combined injection of GIP (25 nmol/kg body weight) with glucose (18 15 mmol/kg body weight). All test solutions were administered in a final volume of Sml/kg body weight. Blood samples were collected from the cut tip of the tail of conscious mice into chilled fluoride/heparin microcentrifuge tubes (Sarstedt, Numbrecht, Germany) immediately prior to injection and at 15, 30 and 60 minutes post injection. Blood samples were immediately centrifuged using a Beckman 20 microcentrifuge (Beckman Instruments, UK) for 30 seconds at 13000g and stored at 200 prior to glucose and insulin determinations. -46 - WO 2005/082928 PCT/GB2005/000710 Acute in vivo effects of (Pro 3 )GIP on plasma glucose and insulin responses tofeeding in obese diabetic ob/ob mice. Plasma glucose and insulin responses were evaluated using 8- to 12-week old ob/ob mice where food was withdrawn for an 18-hour period prior to intraperitoneal injection of saline (0.9% (w/v) NaCl; control) or (Pro')GIP (25 5 nmol/kg body weight). Following injection, the mice were allowed to re-feed for 15 minutes. Blood samples were collected from the cut tip of the tail of conscious mice into chilled fluoride/heparin microcentrifuge tubes (Sarstedt, Numbrecht, Germany) immediately prior to injection and at 15, 30, 60 and 120 minutes post injection. Blood samples were immediately centrifuged using a Beckman microcentrifuge (Beckman 10 Instruments, UK) for 30 seconds at 13000g and stored at -20' prior to glucose and insulin determinations. Effects of chronic (Pro)GIP administration on plasma glucose, insulin and glycated HbAI in obese diabetic ob/ob mice and normal mice. Obese diabetic ob/ob mice and 15 normal control mice aged 8-12 weeks were randomly divided into groups which received once daily subcutaneous injections (17:00h) of either saline (0.9% w/v NaCI) or (Pro 3 )GIP (25 nmol/kg body weight in saline). After 11 days, treatment was ceased. Food intake and body weight were recorded daily. Blood samples were collected from the cut tip of the tail of conscious mice into chilled fluoride/heparin 20 coated glucose microcentrifuge tubes (Sarstedt, Numbrecht, Germany). Blood samples were immediately centrifuged using a Beckman microcentrifuge (Beckman Instruments, UK) for 30 seconds at 13000g prior to glucose, insulin and HbAc determinations. 25 Effects ofchronic treatment with (Pro)GIP on glucose tolerance in ob/ob mice and normal mice. Plasma glucose and insulin concentrations were measured following intraperitoneal administration of glucose (18 mmol/kg body weight) in ob/ob and normal mice treated with either saline (0.9% w/v NaCI) or (Pro 3 )GIP (25 nmol/kg body weight/day) for 11 days. This test was repeated 9 days after cessation of chronic 30 (Pro 3 )GIP treatment. Blood samples were collected from the cut tip of the tail of conscious mice into chilled fluoride/heparin microcentrifuge tubes (Sarstedt, Numbrecht, Germany) immediately prior to injection and at 15, 30 and 60 minutes post injection. Blood samples were immediately centrifuged using a Beckman - 47 - WO 2005/082928 PCT/GB2005/000710 microcentrifuge (Beckman Instruments, UK) for 30 seconds at 13000g and stored at 200 prior to glucose and insulin determinations. Effects of chronic treatment with (Pro 3 )GIP on the glucose lowering effects of 5 exogenous insulin in ob/ob mice. The glucose lowering effects of insulin were evaluated by measuring plasma glucose response in 11-day saline (0.9% w/v NaCl) and (Pro 3 )GIP (25 nmol/kg body weight/day) treated ob/ob mice following acute intraperitoneal administration of insulin (50 U/kg bodyweight). Blood samples were collected from the cut tip of the tail of conscious mice into chilled fluoride/heparin 10 microcentrifuge tubes (Sarstedt, Numbrecht, Germany) immediately prior to injection and at 30 and 60 minutes post injection. Blood samples were immediately centrifuged using a Beckman microcentrifuge (Beckman Instruments, UK) for 30 seconds at 13000g and stored at -20' prior to glucose determination. 15 Effects of chronic treatment with (Pro 3 )GIP on pancreatic insulin content and associated islet hypertrophy in ob/ob mice. Pancreatic tissue was excised from non fasted ob/ob mice after 11 days treatment with either saline (0.9% w/v NaCl) or (Pro 3 )GIP (25 nmol/kg body weight/day). Pancreatic samples were individually wrapped in aluminium foil and snap frozen in liquid nitrogen. Individual excised 20 pancreatic samples were then either embedded, sectioned and immunohistochemically stained for insulin or permeabilised for determination of pancreatic insulin content. Determination of HbA 1 i, plasma glucose and insulin concentrations. HbAl was measured in whole blood by ion-exchange high-performance liquid chromatography 25 using the Menari HA-8140 kit (BIOMEN, Berkshire, UK). Plasma glucose was assayed by an automated glucose oxidase procedure using a Beckman Glucose Analyzer II (Stevens, J.F., 1971, Clinica ChemicaActa 32:199-201) and plasma insulin was determined by RIA (Flatt, P.R. et al., 1981, Diabetologia 20:573-577). Incremental areas under plasma glucose and insulin curves (AUC) were calculated 30 using a computer generated program (CAREA) employing the trapezoidal rule (Burington, R.S., 1973, Handbook ofMathematical Tables and Formulae, New York, McGraw Hill) with baseline subtraction. - 48- WO 2005/082928 PCT/GB2005/000710 Statistical analysis. Results are expressed as means ± SEM. Values were compared using Student's unpaired t-test and groups of data were considered to be significantly different if P<0.05. 5 Results GIP-stimulated cyclic AMP production and insulin secretion were inhibited in dose-dependent fashion by (Pro 3 )GIP, showing that (Pro 3 )GIP is a potent functional GIP receptor antagonist. GIP receptor transfected Chinese hamster lung (CHL) fibroblasts were 10 incubated with 10- to 10-6 M (Pro 3 )GIP in the presence of native GIP (10-' M). The results are shown in Figs. 32A and 32B. Fig. 32A is a line graph showing 3 H-cAMP production as a percent of maximal response (y-axis) with increasing peptide concentration (M) (x-axis). Fig. 32B is a bar graph showing insulin secretion (y-axis) with increasing peptide concentration (M) (x-axis) for 5.6 mM glucose (control) 15 (white bar), GIP (gray bars), (Pro 3 )GIP (lined bars) and (Pro 3 )GIP+GIP(10 7 M) (black bars). *P<0.05, **P<0.01, ***P<0.001 compared to glucose control. 'AP<0.01,

AAAP<

0

.

0 0 1 compared with native GIP at the same concentration. Values are means SEM for 3-8 observations. (Pro)GIP inhibited GIP-induced cAMP formation with an IC 50 value of 2.6 20 RM. Insulin-releasing activity of BRIN-BD1 1 cells exposed to native GIP and (Pro 3 )GIP (in the absence and presence of 10-7 M GIP). GIP-stimulated insulin secretion was inhibited in a dose-dependent fashion by GIP(3-42), (Hyp 3 )GIP, (Lys 3 )GIP, (Tyr 3 )GIP, (Trp 3 )GIP, and (Phe 3 )GIP, as shown in Figs. 33A through 33F, which are bar charts. Fig. 33A shows 3 H-cAMP production 25 as a percent of 10~ 7 M GIP (y-axis) versus logo of GIP (10- 7 M) (white bar, control) and GIP (10- 7 M)+GIP(3-42) (black bars). Figs. 33B through 33F show insulin secretion (in ng/106 cells/20 minutes) (y-axis) as a function of peptide concentration (M) (x-axis) for GIP (10- 7 M) (white bar, control) and a Glu 3 -substituted form of GIP (black bars), including (Hyp 3 )GIP (Fig. 33B), (Lys 3 )GIP (Fig. 33C), (Tyr 3 )GIP (Fig. 30 33D), (Trp 3 )GIP (Fig. 33E), and (Phe 3 )GIP (Fig. 33F). *P<0.05, **P<0.01 compared to GIP (10~7 M) control. Values are means ±SEM for 3-8 observations. Figs. 34A through 34D are a set of two line graphs (Figs. 34A, 34C) and two bar graphs (Figs. 34B, 34D) showing that acute administration of (Pro 3 )GIP - 49 - WO 2005/082928 PCT/GB2005/000710 completely antagonises the actions of GIP on glucose tolerance (Figs. 34A, 34B) and plasma insulin (Figs. 34C, 34D) responses in obese diabetic ob/ob mice. Figs. 34A and 34C are line graphs show plasma glucose levels (Fig. 34A, y-axis) and plasma insulin levels (Fig. 34C, y-axis) over time (x-axis) for glucose (control; V), glucose + 5 GIP (*) and glucose + (GIP+Pro 3 GIP)) (A). Figs. 34B and 34D are bar graphs showing plasmia glucose AUC for glucose alone (white bars), GIP (grey bars) and glucose + (GIP+Pro 3 GIP)) (black bars). Plasma glucose and insulin concentrations after i.p. administration of glucose alone (18 mmol/kg body weight) or in combination with either native GIP or native 10 GIP plus (Pro 3 )GIP (25 nmol/kg body weight). The time of injection is indicated by the arrow (0 minutes). Plasma glucose and insulin AUC values are given for 0-60 minutes post-injection. Values are means ± SEM for 8 mice. *P<0.05, **P<0.01, ***P<0.001 compared with glucose alone. AAP<0.01, """P<0.001 compared with native GIP. 15 Acute administration of (Pro 3 )GIP completely antagonised the insulin releasing action of GIP and the associated improvement of glucose tolerance in ob/ob mice. Indeed, the glycemic excursion following (Pro 3 )GIP (A) was worse than when glucose was administered alone (V). Figs. 35A through 35D show the effects of (Pro 3 )GIP on physiological meal 20 stimulated insulin release and glycemic excursion in obese diabetic ob/ob mice. Plasma glucose and insulin concentrations were measured in mice allowed to re-feed for 15 minutes prior to i.p. administration of saline (0.9% (w/v) NaCl) as control or (Pro 3 )GIP (25 nmol/kg body weight). The time of injection is indicated by the arrow (15 minutes). 25 The results are shown in Figs. 35A through 35D, which are a set of two line graphs (Figs. 35A, 35C) and two bar graphs (Figs. 35B, 35D). The figures show plasma insulin (Figs. 35A) and plasma glucose (Fig. 35C) over time for saline control (V) and (Pro 3 )GIP (0), and plasma insulin AUC (Fig. 35B) and plasma glucose AUC (Fig. 35D) for saline control (white bars) and (Pro 3 )GIP (black bars), respectively. 30 Values are means ± SEM for 8 mice. *P<0.05, **P<0.01, ***P<0.001 compared with saline alone. Acute administration of (Pros)GIP decreased the insulin response to feeding and worsened the associated glycemic excursion in ob/ob mice. These effects of - 50 - WO 2005/082928 PCT/GB2005/000710 functional ablation of endogenous GIP by the (Pro 3 )GIP antagonist are fully consistent with the accepted role of GIP in the regulation of insulin secretion and glycemic excursion following feeding. The effects of chronic administration of (Pro 3 )GIP for 11 days on plasma 5 glucose and insulin concentrations of obese diabetic ob/ob mice were also studied. According to classical thinking and the experiments described above and the results shown in Figs. 32-35, functional ablation of endogenous GIP by daily administration of (Pro 3 )GIP over 11 days would be expected to inhibit insulin secretion and cause a marked deterioration in glucose tolerance. 10 However, the exact opposite occurred during chronic treatment with (Pros)GIP in ob/ob mice. This is shown in Fig. 36, which is a set of two bar graphs showing plasma glucose (Fig. 36A) and insulin (Fig. 36B) concentrations after daily subcutaneous administration of saline alone (0.9% (w/v) NaCI; as control; white bars) or (Pro 3 )GIP (25 ninol/kg body weight; black bars) for 11 days. Values are means ± 15 SEM for 6 mice and *P<0.05 compared with saline alone. Chronic administration of (Pro 3 )GIP (black bars) for 11 days decreases plasma glucose and plasma insulin concentrations of obese diabetic ob/ob mice, relative to controls. The effects of chronic administration of (Pro 3 )GIP for 11 days on HbAic, (Fig. 37A), pancreatic insulin content (Fig. 37B) and associated islet hypertrophy (Fig. 20 37C) were examined in obese diabetic ob/ob mice treated with saline (control, white bars) and (Pro 3 )GIP were examined. HbA 1 c, pancreatic insulin content and average islet diameter were measured after 11 daily subcutaneous injections of either saline alone (white bars) or (Pro 3 )GIP (25 nmol/kg body weight; black bars) to obese diabetic ob/ob mice. Values are means ± SEM for 6 mice and *P<0.05, ***P<0.001 25 compared with saline-treated group. Beneficial effects of chronic (Pro 3 )GIP administration in ob/ob mice were associated with significant decreases in HbA 1 e and pancreatic insulin stores, with partial correction of the marked islet hypertrophy of the ob/ob mutant. There was also an approximate 7% decrease in body weight in (Pro 3 )GIP-treated ob/ob mice without 30 any change in food intake. This effect did not achieve significance over the short study period, but this observation clearly suggests that GIP antagonism may also have a longer-term anti-obesity action. - 51 - WO 2005/082928 PCT/GB2005/000710 The effects of chronic administration of (Pro 3 )GIP for 11 days on glucose tolerance and plasma insulin in obese diabetic ob/ob mice is shown in Figs. 38A-38D, which are a set of line graphs (Figs. 38A, 38C) and bar graphs (Figs. 38B, 38C) showing plasma glucose levels (Figs. 38A, 38B) and plasma insulin levels (Figs. 38C, 5 38D) in obese diabetic ob/ob mice treated with saline (control, white) or (Pro 3 )GIP (black). Plasma glucose and insulin concentrations were measured prior to and at intervals after intraperitoneal administration of glucose (18 mmol/kg body weight). Arrow indicates time of injection (t-0). Values are means ± SEM for 6 mice and *P<0.05, **P<0.01, ***P<0.001 compared with saline-treated group. 10 After 11 days treatment with (Pro')GIP, glucose tolerance of ob/ob mice was substantially improved without change of circulating insulin (Fig. 38). Fig. 39 shows the effects of chronic administration of (Pro 3 )GIP for 11 days on insulin sensitivity in obese diabetic ob/ob mice. Plasma glucose concentrations of saline and (Pro)GIP treated ob/ob mice were measured prior to and at intervals after 15 intraperitoneal administration of exogenous insulin (50 U/kg body weight; t=0). Values are means ± SEM for 6 mice and *P<0.05 compared with saline-treated group. As shown in Fig. 39, chronic administration of (Pro 3 )GIP caused a significant improvement of insulin sensitivity. Interestingly, the beneficial effects of chronic administration of (Pro 3 )GIP for 20 11 days in obese diabetic ob/ob mice was reversed 9 days after cessation of treatment. This is consistent with a physiological effect, and is shown in Fig. 40. Plasma glucose concentrations were measured prior to and after intraperitoneal administration of glucose (18 mmol/kg body weight) for mice that had been treated with saline (control, n) or (Pro 3 )GIP (A). Arrow indicates time of injection (t=0). Values are means 25 SEM for 6. Figs. 41 A and 41 B are a pair of line graphs showing the effects of chronic administration of (Pro 3 )GIP for 11 days on glucose tolerance in normal mice. Plasma glucose concentrations were measured prior to and after intraperitoneal administration of glucose (18 mmol/kg body weight). Arrow indicates time of injection (t=O). 30 Values are means ± SEM for 6 and *P<0.05, **P<0.01 compared to saline-treated group. In total contrast to beneficial actions in ob/ob mice, chronic daily treatment of normal mice with (Pro 3 )GIP (A) for 11 days resulted in a marked deterioration of - 52 - WO 2005/082928 PCT/GB2005/000710 glucose tolerance (Fig. 41A) relative to controls (m), which was reversed 9 days after cessation of treatment (Fig. 41 B). Example 5. Chemical Ablation of Gastric Inhibitory Polypeptide Receptor Action By 5 Daily (Pro3)GIP Administration Improves Glucose Tolerance and Ameliorates Insulin Resistance and Abnormalities of Islet Structure in Obesity-Diabetes. Gastric inhibitory polypeptide (GIP) is an important incretin hormone secreted by endocrine K-cells in response to nutrient ingestion. This study investigated the effects of chemical ablation of GIP receptor (GIP-R) action on aspects of obesity 10 diabetes using a stable and specific GIP-R antagonist, (Pro 3 )GIP. Young adult ob/ob mice received once daily i.p. injections of saline vehicle or (Pro 3 )GIP over an 11-day period. Non-fasting plasma glucose levels and the overall glycemic excursion (AUC) to a glucose load were significantly reduced (1.6-fold; P <0.05) in (Pro 3 )GIP-treated mice compared to controls. GIP-R ablation also significantly lowered overall plasma 15 glucose (1.4-fold; P < 0.05) and insulin (1.5-fold; P < 0.05) responses to feeding. These changes were associated with significantly enhanced (1.6-fold; P <0.05) insulin sensitivity in the (Pro 3 )GIP-treated group. Daily injection of (Pro')GIP reduced pancreatic insulin content (1.3-fold; P <0.05) and partially corrected the obesity-related islet hypertrophy and beta cell hyperplasia of ob/ob mice. These 20 comprehensive beneficial effects of (Pro 3 )GIP were reversed following 9 days cessation of treatment and were independent of food intake and body weight, which were unchanged. These studies highlight a role for GIP in obesity-related glucose intolerance and emphasize the potential of specific GIP-R antagonists as a new class of drugs for the alleviation of insulin resistance and treatment of type 2 diabetes. 25 Research Design And Methods Animals. Obese diabetic (ob/ob) mice derived from the colony maintained at Aston University, UK (Bailey, C.J., et aL, 1982, Int. J Obes. 6:11-21) were used at 12-16 weeks of age. Animals were age-matched, divided into groups and housed 30 individually in an air-conditioned room at 22-2'C with a 12 hour light: 12 hour dark cycle. Drinking water and a standard rodent maintenance diet (Trouw Nutrition, Cheshire, UK) were freely available. All animal experiments were carried out in - 53 - WO 2005/082928 PCT/GB2005/000710 accordance with the UK Animals (Scientific Procedures) Act 1986. No adverse effects were observed following administration of (Pro 3 )GIP. Synthesis, purification and characterization of (Pro)GIP. (Pro')GIP was 5 sequentially synthesized on an Applied Biosystems automated peptide synthesizer (Model 432 A). (Pro 3 )GIP was purified by reversed-phase HPLC on a Waters Millenium 2010 chromatography system (Software version 2.1.5) and subsequently characterized using electrospray ionization mass spectrometry (ESI-MS). 10 Experimental protocolsfor ob/ob mouse studies. Initially, extended biological activity of (Pro 3 )GIP was examined in 18-hour fasted ob/ob mice 4 hours after administration. Thereafter, over an 11-day period, mice received once daily i.p. injections (17:00 hours) of either saline vehicle (0.9% (w/v), NaCl) or (Pro 3 )GIP (25 nmol/kg body wt). During a subsequent 9-day period, observations were continued 15 following discontinuation of (Pro 3 )GIP administration. Food intake and body weight were recorded daily whilst plasma glucose and insulin concentrations were monitored at intervals of 2-6 days. Whole blood for the measurement of glycated hemoglobin was taken on days 11 and 20. Intraperitoneal glucose tolerance (18 mmol/kg body wt), metabolic response to native GIP (25 nmol/kg body wt) and insulin sensitivity 20 (50 U/kg body wt) tests were performed on days 11 and 20. Mice fasted for 18 hours were used to examine the metabolic response to 15 minutes feeding. In a separate series, pancreatic tissues were excised at the end of the 11-day treatment period or 9 days following discontinuation of (Pro 3 )GIP and processed for immunohistochemistry or measurement of insulin following extraction with 5 ml/g of ice-cold acid ethanol 25 (750 ml ethanol, 235 ml water, 15 ml concentrated HCl). Blood samples taken from the cut tip of the tail vein of conscious mice at the times indicated in the Figures were immediately centrifuged using a Beckman microcentrifuge (Beckman Instruments, UK) for 30 seconds at 13,000 g. The resulting plasma was then aliquoted into fresh Eppendorf tubes and stored at -20'C prior to glucose and insulin determinations. 30 Biochemical analysis. Plasma glucose was assayed by an automated glucose oxidase procedure (Stevens, J.F., 1971, Clin. Chem. Acta 32:199-201) using a Beckman Glucose Analyzer II (Beckman Instruments, Galway, Ireland). Plasma and pancreatic - 54 - WO 2005/082928 PCT/GB2005/000710 insulin were assayed by a modified dextran-coated charcoal radioimnunoassay (Flatt, P.R. et al., 1981, Diabetologia 20:573-577). Glycated hemoglobin was determined using cation-exchange columns (Sigma, Poole, Dorset, UK) with measurement of absorbance (415 nm) in wash and eluting buffer using a VersaMax Microplate 5 Spectrophotometer (Molecular Devices, Wokingham, Berkshire, UK). Immunocytochenistry. Tissue fixed in 4% paraformaldehyde/PBS and embedded in paraffin was sectioned at 8 pin. After de-waxing, sections were incubated with blocking serum (Vector Laboratories, CA, USA) prior to exposure to insulin antibody. 10 Tissue samples were then incubated consecutively with secondary biotinylated universal, pan-specific antibody (Vector Laboratories, CA, USA) and streptavidin/peroxidase preformed complex (Vector Laboratories, CA, USA). Following washing, the stained pancreatic tissue was counterstained with hematoxylin (BDH Chemicals, Dorset, UK) and then plunged into acid methanol (500 ml 15 methanol, 500 ml H 2 0 and 2.5 ml concentrated HCI) prior to dehydration and mounting in Depex (BDH Chemicals, Dorset, UK). The stained slides were viewed under a microscope (Nikon Eclipse E2000, Diagnostic Instruments Incorporated, Michigan, USA) attached to a JVC camera Model KY-F55B (JVC, London, UK) and analyzed using Kromoscan imaging software (Kinetic Imaging Limited, Faversham, 20 Kent, UK). The average number and diameter of every islet in each section was estimated in a blinded manner using an eyepiece graticule calibrated with a stage micrometer (Graticules Limited, Tonbridge, Kent, UK). The longest and shortest diameters of each islet were determined with the graticule. Half of the sum of these two values was then considered to be the average islet diameter. Approximately 60 25 70 random sections were examined from the pancreas of each mouse. Statistics. Results are expressed as mean ± SEM. Data were compared using ANOVA, followed by a Student-Newman-Keuls post hoc test. Area under the curve (AUC) analyzes were calculated using the trapezoidal rule with baseline subtraction 30 (Burington, R.S., Handbook ofMatheinatical Tables and Formulae, New York, McGraw-Hill, 1973). P < 0.05 was considered to be statistically significant. - 55 - WO 2005/082928 PCT/GB2005/000710 Results Effects of (Pro 3 )GIP on plasma glucose and insulin concentrations 4 hours after administration were examined. The results are shown in Figs. 42A through 42D, which are a set of two line graphs (Figs. 42A, 42C) and two bar graphs (Figs. 42B, 5 42D) showing the effects of (Pro 3 )GIP on plasma glucose and insulin response to native GIP 4 hours after administration. Tests were conducted 4 hours after administration of (Pro)GIP (25 nmoles/kg body weight) or saline (0.9% NaCl) in 18 hour-fasted ob/ob mice. Plasma glucose and insulin concentrations were measured prior to and after i.p. administration of glucose (18 mmoles/kg body weight) in 10 combination with native GIP (25 nmoles/kg body weight). The incremental area under the glucose or insulin curves (AUC) between 0 and 60 min are shown in the right panels. Values represent means ± SEM for 8 mice. *P <0.05 and **P < 0.01 compared with saline alone group. As shown in Figs. 42A through 42D, administration of (Pro 3 )GIP for 4 hours 15 previously impaired the plasma glucose and insulin responses to native GIP, given together with glucose. AUC glucose and insulin values were increased by 151% (P < 0.05) and decreased by 25% (P < 0.05); respectively, compared with saline-treated controls. This supports a protracted biological half-life and forms the basis of the once-daily injection. 20 The effects of (Pro 3 )GIP on food intake, body weight and non-fasting plasma glucose and insulin concentrations were studied. The results are shown in Figs. 43A through 43D, which are a set of two line graphs and two bar graphs showing the effects of daily (Pro 3 )GIP administration on food intake (Fig. 43A), body weight (Fig. 43B), plasma glucose (Fig. 43C) and insulin (Fig. 43D) concentrations in ob/ob mice. 25 Parameters were measured for 5 days prior to, 11 days during (indicated by black bar) and 9 days after treatment with saline or (Pro 3 )GIP (25 nmol/kg bw/day). Values are mean ± SEM for eight mice. *P < 0.05 compared with saline group. Administration of (Pro 3 )GIP had no effect on food intake and body weight (Fig. 43A and 43B). On day 11, plasma glucose had declined to significantly reduced 30 (P < 0.05) concentrations in ob/ob mice receiving (Pro 3 )GIP (Fig. 43C). Cessation of treatment returned plasma glucose concentrations towards control levels. Consistent with this pattern, glycated hemoglobin was significantly lower (P <0.05) after 11 days treatment with (Pro 3 )GIP than either before or 9 days following cessation of - 56 - WO 2005/082928 PCT/GB2005/000710 daily injection (8.0 ± 0.3%, 6.9 ± 0.2%, 7.7 ± 0.4%, respectively). No significant changes in plasma insulin levels were noted during or after the treatment period. However, there was a general trend for plasma insulin concentrations to decrease progressively during (Pro 3 )GIP treatment (Fig. 43D). 5 The effects of (Pro 3 )GIP on glucose tolerance are shown in Figs. 44A through 44D, which are a set of four line graphs with inset bar graphs showing the effects of daily (Pro 3 )GIP administration on glucose tolerance and plasma insulin response to glucose in ob/ob mice. Tests were conducted after daily treatment with (Pros)GIP (25 nmoles/kg body weight/day; A; black bars) or saline (control; u; white bars) for 11 10 days (Fig. 44A, 44C) or 9 days after cessation of treatment (Fig. 44B, 44B). Glucose (18 mmoles/kg body weight) was administered at the time indicated by the arrow. Plasma glucose (Fig. 44A, 44B) and insulin (Fig. 44C, 44D) AUC values for 0-60 minutes post injection, with identical baseline subtractions in each case to demonstrate the complete effect of (Pro 3 )GIP treatment, are shown in insets. Values 15 are mean SEM for eight mice. *P < 0.05, **P < 0.01 and ***P < 0.001 compared with saline group. Daily administration of (Pro 3 )GIP for 11 days resulted in significantly reduced (P < 0.001) plasma glucose concentrations at 15, 30 and 60 minutes following intraperitoneal glucose (Fig. 44A). This was corroborated by a significantly 20 decreased 0-60 minutes AUC value (Fig. 44A) which was 2.1-fold reduced (P < 0.01) compared to controls. Plasma insulin concentrations were also significantly (P < 0.05) reduced 15, 30 and 60 minutes following intraperitoneal glucose injection in the (Pro 3 )GIP treated group (Fig. 44A). AUC, 0-60 minutes values were also significantly decreased (P < 0.001). Interestingly, an almost identical pattern was 25 observed when 11 day treated ob/ob mice were administered glucose together with native GIP (25 nmoles/kg body weight) (data not shown). This supports the view that GIP action was effectively antagonized in the (Pro 3 )GIP treated group. Discontinuation of (Pro 3 )GIP treatment for 9 days (day 20 of study) resulted in almost identical plasma glucose and insulin responses to intraperitoneal glucose (Fig. 44), 30 with lower glucose-mediated plasma insulin concentrations noted at one time point (15 minutes; P <0.05). The effects of (Pro 3 )GIP on metabolic response to feeding and insulin sensitivity are shown in Figs. 45 and 46. Figs. 45A through 45D are a set of two line -57- WO 2005/082928 PCT/GB2005/000710 graphs (Figs. 45A, 45C) and two bar graphs (Figs. 45B, 45D) showing the effects of daily (Pro 3 )GIP administration (A; black bars) or saline (o; white bars) on glucose (Figs. 45A, 45B) and insulin (Figs. 45C, 45D) responses to feeding in ob/ob mice fasted for 18 hours. Tests were conducted after daily treatment with (Pro')GIP (25 5 nmol/kg body weight/day) or saline for 11 days. The arrow indicates the time of feeding (15 minutes). AUC values for 0 105 minutes post-feeding are also shown. Values are meantSEM for eight mice. *P <0.05 compared with saline group. Figs. 46A through 46D are a set of two line graphs (Figs. 46A, 46C) and two bar graphs (Figs. 46B, 46D) showing the effects of daily (Pro 3 )GIP administration on 10 insulin sensitivity in ob/ob mice. Tests were conducted after daily treatment with (Pro 3 )GIP (25 nmol/kg body weight/day; V; black bars) or saline (0; white bars) for 11 days (Fig. 46A, 46B) or 9 days after cessation of treatment (Fig. 46C, 46D). Insulin (50 U/kg body weight) was administered by intraperitoneal injection at the time indicated by the arrow. AUC values for 0 60 minutes post-injection are also 15 shown. Values are mean+SEM for eight mice. *P <0.05 compared with saline group. Plasma glucose and insulin responses to 15 minutes feeding were significantly lowered (P < 0.05) at 30 and 60 minutes in ob/ob mice treated with (Pro 3 )GIP for II days (Fig. 45). Similarly, AUC glucose and insulin were significantly (P < 0.05) decreased in (Pro 3 )GIP treated ob/ob mice, despite similar food intakes of 0.3 - 0.5 20 g/mouse/15 minutes. As shown in Fig. 46A and 45B, the hypoglycemic action of insulin was significantly (P < 0.05) augmented in terms of AUC measures and post injection values in ob/ob mice treated with (Pro 3 )GIP for 11 days. The responses following 9 days discontinuation of (Pro 3 )GIP treatment were similar to saline treated controls (Fig. 45C, 45D). 25 The effects of (Pro 3 )GIP on pancreatic insulin and islet morphology are shown in Figs. 47A through 47D, and 48A through 48F. Figs. 47A through 47D are a set of four bar graphs showing the effects of daily (Pro 3 )GIP administration on pancreatic weight (Fig. 47A), insulin content (Fig. 47B), islet number (Fig. 47C) and islet diameter (Fig. 47D) in ob/ob mice. Parameters were measured after daily treatment 30 with (Pros)GIP (25 nmol/kg body weight/day; black bars) or saline (white bars) for 11 days and 9 days after cessation of treatment (day 20). Values are mean SEM for eight mice. *P < 0.05 and ***P < 0.001 compared with saline group. Figs. 48A through 48F are a set of two bar graphs (Figs. 48A, 48D) and four photomicrographs - 58 - WO 2005/082928 PCT/GB2005/000710 (Figs. 48B, 48C, 48E, 48F), showing the effects of daily (Pro 3 )GIP administration on islet size and morphology in ob/ob)mice. (Pro )GIP treatment had no effect on pancreatic weight (Fig. 47A). However, pancreatic insulin content was significantly (P < 0.05) decreased in ob/ob mice 5 receiving (Pro3)GIP for 11 days compared to controls (Fig. 47B). No significant differences were observed in islet number per pancreatic section (Fig. 47C), but average islet diameter was markedly and significantly reduced (P <0.001) in (Pro3)GIP treated ob/ob mice (Fig. 47D). These effects were effectively reversed by discontinuation of (Pro 3 )GIP on day 20, however average islet diameter was still 10 significantly reduced (P <0.05). As shown in Fig. 48A, more detailed analysis revealed that the reduction is islet diameter on day 11 was due to a significant decrease (P < 0.001) in the percentage of larger diameter (>0.15 mm) islets with increases in the proportion of islets with small (< 0.10 mm) and medium (0.1 - 0.15 mm) diameters. Figure 48D presents similar analysis following cessation of 15 treatment, with a significant (P < 0.05) increase in the percentage of small islets still apparent. Representative images (x40 magnification) of pancreata immunohistologically stained for insulin from 11-day (Pro 3 )GIP treated ob/ob mice (Fig. 48B) and saline treated controls (Fig. 48C) illustrate the dramatic changes in pancreatic islet morphology induced by (Pro3)GIP treatment. Pancreata 20 immunohistologically stained for insulin on day 20 are also shown (Fig. 48E, 48F). Parameters were measured after daily treatment with (Pros)GIP (25 nmol/kg body weight/day) or saline for 11 days (Fig. 48A) and 9 days after cessation of treatment (Fig. 48D). Proportion of islets classified as large (> 0.15 mm) diameter, medium (0.1 - 0.15 mm) diameter and small (< 0.1 mm) diameter are shown. Values 25 are mean=zSEM for eight mice Figs. 48B, 48C, 48E and 48F are representative images (x 40 magnification) of pancreata stained for insulin following 11 days treatment with (Pro3)GIP (Fig. 48B) or saline (Fig. 48C). Corresponding images 9 days after cessation of treatment with (Pros)GIP (Fig. 48E) or saline (Fig. 48F) are also shown. The arrows indicate islets. 30 Example 6. N-Terminally Acetylated and Ly and Lys3 7 -substituted GIP This example examines the metabolic stability, biological activity and antidiabetic potential of fatty acid derivatized N-terminally modified GIP analogues. -59- WO 2005/082928 PCT/GB2005/000710 These are N-AcGIP(LysPAL 16 ) and N-AcGIP(LysPAL 37 ), which have an N-terminal Tyri acetyl group, and a C- 16 palmitate group linked to the epsilon-amino group of the lysine at either position 16 or position 37 of the GIP protein. 5 Materials and methods Animals. Obese diabetic (ob/ob) mice derived from the colony maintained at Aston University, UK were used at 12-17 weeks of age. The genetic background and characteristics of the colony used have been outlined in detail elsewhere (Bailey, C.J. et al., 1982, Int. J Obesity 6:11-21; Gault, V.A. et al., 2003, J. Endocrinol. 176: 133 10 141). Animals were housed in an air-conditioned room at 22+2'C with a 12 hours light:12 hours dark cycle. Drinking water and standard rodent maintenance diet (Trouw Nutrition, Cheshire, UK) were freely available. All test solutions were administered by i.p. injection in a final volume of 5 ml/kg bw. Blood was collected from the cut tip of the tail vein of conscious mice into chilled fluoride/heparin 15 microcentrifuge tubes immediately prior to injection and at the times indicated in the Figures. Plasma was separated using a Beckman microcentrifuge (Beckman Instruments, UK) at 13,000 g for 30 second and stored at -20'C prior to glucose and insulin determinations. All animal experiments were carried out in accordance with the UK Animals (Scientific Procedures) Act 1986. No adverse effects were observed 20 following acute or long-term administration of any of the peptides. Materials. High performance liquid chromatography (HPLC) grade acetonitrile was obtained from Rathburn (Walkersburn, UK). Trifluoroacetic acid (TFA) and trichloroacetic acid (TCA) were obtained from Aldrich (Poole, Dorset, UK). DPP IV, 25 isobutylmethylxanthine (IBMX), alpha-cyano-4-hydroxycinnamic acid, cyclic AMP and ATP were all purchased from Sigma (Poole, Dorset, UK). Fmoc-protected amino acids were from Calbiochem Novabiochem (Nottingham, UK). RPMI-1640 and DMEM tissue culture medium, foetal bovine serum, penicillin and streptomycin were all purchased from Gibco (Paisley, Strathclyde, UK). The chromatography columns 30 used for cyclic AMP assay, Dowex AG50 WX and neutral alumina AG7 were obtained from Bio-Rad (Life Science Research, Alpha Analytical, Larne, UK). All water used in these experiments was purified using a Milli-Q Water Purification -60- WO 2005/082928 PCT/GB2005/000710 System (Millipore, Milford, MA, USA). All other chemicals used were of the highest purity available. Synthesis, purification and characterisation of GIP and related analogues. Native 5 GIP was sequentially synthesised using standard solid-phase Fmoc peptide chemistry (ABI 432A Peptide Synthesiser) as described previously (O'Harte, F.P.M. et al., 2002, Diabetologia 45: 1281-1291). N-AcGIP(LysPAL 16 ) and N-AcGIP(LysPAL 37 ) were synthesised in the same way as native GIP but with the exception that the epsilon-amino groups of Lys at positions 16 or 37 were conjugated with a C-16 10 palmitate fatty acid. In addition, an acetyl adduct was incorporated at the N-terminal Tyr'. The synthetic peptides were judged pure by reversed-phase HPLC on a Waters Millenium 2010 chromatography system (Software version 2.1.5) and subsequently characterised using matrix assisted laser desorption ionisation-time of flight mass spectrometry (MALDI-ToF MS) as described previously (Gault, V.A. et al., 2002, 15 Cell. Biol. Int. 27: 41-46). DPP IV degradation studies. GIP and fatty acid derivatised GIP analogues were incubated at 37 0 C with purified porcine dipeptidylpeptidase IV (5 mU in 50 mmol/l triethanolamine-HCI; pH 7.8) for 0, 2, 4, 8 and 24 hours (final peptide concentration 2 20 mmol/1). The reactions were subsequently terminated by addition of 10% (v/v) TFA/water and the reaction products separated using HPLC. Reaction products were applied to a Vydac C-4 column (4.6 x 250 mm; The Separations Group, Hesparia, CA) and the major degradation product GIP(3-42) separated from intact GIP. The column was equilibrated with 0.12% (v/v) TFA/water at a flow rate of 1.0 ml/minute 25 using 0.1% (v/v) TFA in 70% acetonitrile/water with the concentration of acetonitrile in the eluting solvent being raised from 0% to 40% over 10 minutes, and then from 40% to 75% over 35 minutes. The absorbance was monitored at 206 nm using a SpectraSystem UV 2000 Detector (Thermoquest Limited, Manchester, UK) and the peaks collected manually prior to MALDI-ToF MS analysis. HPLC peak area data 30 were used to calculate % intact peptide remaining throughout the incubation. Cells and cell culture. Chinese hamster lung (CHL) fibroblasts stably transfected with the human GIP receptor (Gremlich, S. et al., 1995, Diabetes 44: 1202-1208) -61 - WO 2005/082928 PCT/GB2005/000710 were cultured in DMEM tissue culture medium containing 10% (v/v) FBS, 1% (v/v) antibiotics (100 U/ml penicillin, 0.1 mg/mI streptomycin). Clonal pancreatic BRIN BD I1 cells (McClenaghan, N.H. et al., 1996, Diabetes 45: 1132-1140) were cultured using RPMI- 1640 culture medium containing 10% (v/v) FBS, 1% (v/v) antibiotics 5 (100 U/ml penicillin, 0.1 mg/ml streptomycin) and 11.1 mmol/l glucose. Cells were maintained at 37'C in an atmosphere of 5% CO 2 and 95% air using an LEEC incubator (Laboratory Technical Engineering, Nottingham, UK). In vitro biological activity. Intracellular cyclic AMP production was measured using 10 GIP-receptor transfected CHL fibroblasts (O'Harte, F.P.M. et al., 2002, Diabetologia 45: 1281-1291). In brief, CHL cells were seeded into 12-well plates (NOnc, Roskilde, Denmark) at a density of 105 cells per well and allowed to grow for 48 hours before being loaded with tritiated adenine (2 piCi; TRK3 11; Amersham, Buckinghamshire, UK). The cells were then incubated at 37 0 C for 6 hours in 1 ml DMEM 15 supplemented with 0.5% (w/v) BSA and subsequently washed twice with HBS buffer (pH 7.4). Cells were then exposed to GIP/GIP analogues (10~1 to 10- mol/l) in HBS buffer in the presence of I mmol/l IBMX for 15 minutes at 37'C. The medium was subsequently removed and the cells lysed with 1 ml of 5% TCA containing 0.1 mmol/l unlabelled cyclic AMP and 0.1 mmol/l unlabelled ATP. The intracellular 20 cyclic AMP was then separated on Dowex and alumina exchange resins as described previously (O'Harte, F.P.M. et al., 2002, Diabetologia 45: 1281-1291). Insulin-release studies were carried out using clonal pancreatic BRIN-BD 11 cells as described previously (O'Harte, F.P.M. et al., 2002, Diabetologia 45: 1281 1291). Briefly, BRIN-BD 11 cells were seeded into 24-well plates at a density of 105 25 cells per well, and allowed to attach overnight at 37'C. Acute tests for insulin release were preceded by 40 minutes pre-incubation at 37 0 C in 1.0 ml Krebs Ringer bicarbonate buffer supplemented with 1.1 mmol/l glucose. Test incubations were performed in the presence of 5.6 mmol/l glucose with a range of concentrations (1013 to 106 mol/l) of GIP and GIP analogues. After 20 minutes incubation, the buffer was 30 removed from each well and aliquots (200 pd) used for measurement of insulin. Effects ofN-AcGIP(LysPAL 1) and N-A cGIP(LysPAL") in ob/ob mice. Metabolic and dose-response effects of GIP and N-AcGIP(LysPAL) analogues (at 6.25 - 25 - 62 - WO 2005/082928 PCT/GB2005/000710 nmoles/kg bw) following glucose administration (18 mmoles/kg bw) were examined in mice fasted for 18 hours. To evaluate long-term effects, groups of ob/ob mice received once daily intraperitoneal injections (17:00 h) for 14 days of either saline vehicle (0.9%, w/v, NaCl), native GIP or N-AcGIP(LysPAL1 7 ) (both at 12.5 5 nmoles/kg body weight/day). Food intake and body weight were recorded daily. Plasma glucose and insulin concentrations were monitored at 2-6 day intervals. At 14 days, groups of animals were used to evaluate intraperitoneal glucose tolerance (18 mmoles/kg) and insulin sensitivity (50 U/kg). In a separate series, two experimental protocols were employed to examine the possibility of GIP receptor desensitization 10 after 14 days treatment. Acute metabolic effects of the usual injection of either saline, GIP or N-AcGIP(LysPAL 37 ) were monitored when administered together with glucose (18 mmoles/kg). In the second, acute effects of N-AcGIP(LysPAL 3 7 ) given together with glucose were examined in all 3 groups of mice. At the end of the 14 day treatment period, pancreatic tissues were excised for measurement of insulin 15 following extraction with 5 ml/g ice-cold acid ethanol (75% ethanol, 2.35% H20, 1.5% HCl). Whole blood was taken for determination of glycated hemoglobin. Biochemical analyses. Plasma glucose was assayed by an automated glucose oxidase procedure (Stevens, J.F., 1971, Clin. Chem. Acta 32:199-201) using a Beckman 20 Glucose Analyser II (Beckman, Galway, Ireland). Plasma insulin was determined by dextran-charcoal RIA as described previously (Flatt, P.R. et al., 1981, Diabetologia 20: 573-577). Glycated hemoglobin was determined using cation-exchange columns (Sigma, Poole, Dorset, UK) with measurement of absorbance (415 urn) in wash and eluting buffers using a VersaMax microplate spectrophotometer (Molecular Devices, 25 Wokingham, Berkshire, UK). Statistics. Results are expressed as mean L SEM. Data were compared using the unpaired Student's t-test. Where appropriate, data were compared using repeated measures ANOVA or one-way ANOVA, followed by the Student-Newman-Keuls 30 post hoc test. Incremental areas under plasma glucose and insulin curves (AUC) were calculated using a computer-generated program employing the trapezoidal rule (Burington, R.S., 1973, Handbook ofMathematical Tables and Formulae, McGraw - 63 - WO 2005/082928 PCT/GB2005/000710 Hill, New York) with baseline subtraction. Groups of data were considered to be significantly different ifp<0.05. Results 5 Structural characterisation by MALDI-ToF MS. Following synthesis and HPLC purification, the molecular masses were obtained for GIP, N-AcGIP(LysPAL1 6 ) and N-AcGIP(LysPAL 37 ) using MALDI-ToF MS (Table 3, below). The mass-to-charge (m/z) ratio for native GIP was calculated to be 4983.7 Da, corresponding very closely to the theoretical mass of 4982.4 Da. Similarly, the m/z ratios for N 10 AcGIP(LysPAL 16 ) and N-AcGIP(LysPAL 3 7 ) were 5268.9 Da and 5267.7 Da, respectively. These values correlate very closely to the theoretical mass (5266.1 Da), therefore, confirming the correct structures for each of the synthetic peptides. Table 3. Structural characterisation of GIP and GIP analogues by MALDI-ToF MS. Peptide Experimental Theoretical Difference Mr (Da) M, (Da) (Da) GIP 4983.7 4982.4 1.3 N-AcGIP(LysPAL1 6 ) 5268.9 5266.1 2.8 N-AcGIP(LysPAL 3 7) 5267.7 5266.1 1.6 15 Peptide samples were mixed with matrix (alpha-cyano-4-hydroxycinnamic acid) and m/z ratio vs. relative peak intensity recorded using a Voyager-DE BioSpectrometry Workstation. Degradation by DPP IV. Table 4, below, illustrates the % intact peptide remaining 20 after incubation with DPP IV. Degradation of native GIP was evident after just 2 hours with only 52±3% of the peptide remaining intact. After 8 hours incubation the native peptide was entirely degraded to GIP(3-42). In contrast, both N AcGIP(LysPAL 1 6 ) and N-AcGIP(LysPAL 37 ) remained completely intact (no degradation fragment evident) even after 24 hours incubation with DPP IV. 25 Table 4. Percentage intact peptide remaining after incubation with DPP IV. Peptide % Intact peptide remaining after time (hours) - 64 - WO 2005/082928 PCT/GB2005/000710 0 2 8 24 Native GIP 100 52 A 3 0 0 N-AcGIP(LysPALl1 6 ) 100 100 100 100 N-AcGIP(LysPAL 3 7 ) 100 100 100 100 Values represent the % intact peptide remaining relative to the major degradation product GIP(3-42) following incubation with DPP IV as determined from HPLC peak area data. The reactions were performed in triplicate and the means+SEM values calculated. 5 Changes in Cyclic AMP production. Fig. 50A shows intracellular cyclic AMP production by GIP (A) and fatty acid derivatised GIP analogues N-AcGIP(LysPAL 16 ) (o) and N-AcGIP(LysPAL 3 7 ) (e), as determined by column chromatography, in CHL cells stably expressing the human GIP receptor. Each experiment was performed in 10 triplicate (n=3) and the results expressed (means ± SEM) as a percentage of the maximum GIP response. A concentration-dependent (10-13 to 10-6 mol/l) increase in cyclic AMP production was observed with native GIP (EC 5 0 value 18.2 nmol/1) using CHL cells transfected with the human GIP receptor (Fig. 50A). Likewise, both N 15 AcGIP(LysPAL 6 ) and N-AcGIP(LysPAL 37 ) followed a similar pattern of stimulation to that of native GIP with calculated EC 50 values of 12.1 and 13.0 nmol/l, respectively. The lower EC 5 0 values for both analogues suggest an enhanced cyclic AMP-stimulating potency. 20 In vitro insulin-releasing activity. Fig. 50B shows insulin-releasing activity of glucose (5.6 mmo/l glucose; white bars), GIP (lined bars) and fatty acid derivatised GIP analogues N-AcGIP(LysPAL 6) (grey bars) and N-AcGIP(LysPAL 37 ) (black bars) in the clonal pancreatic beta cell line, BRIN-BDl 1. After a pre-incubation (40 minutes), the effects of various concentrations of peptide were tested on insulin 25 release during a 20 minutes incubation. Values are means L SEM for 8 separate observations. *p<0.05, **p<0.01, ***p<0.00I compared to control (5.6 mmol/l glucose alone). Consistent with its role as a potent insulinotropic hormone, native GIP dose dependently stimulated insulin secretion (p<0.01 to p<0.001) compared to control (5.6 - 65 - WO 2005/082928 PCT/GB2005/000710 mmol/l glucose alone) (Fig. 50B). Likewise, both N-AcGIP(LysPAL 16 ) and N AcGIP(LysPAL 37 ) significantly stimulated glucose-induced insulin secretion (p<0.05 to p<0.00 l). On the basis of cyclic AMP and insulin secretory data, both GIP analogues appear to be at least equipotent to the native peptide. 5 Metabolic effects in ob/ob mice. Figs. 5lA through 51D are a set of two line graphs (Figs. 51A, 51C) and two bar graphs (Figs. 51B, 51D) showing glucose lowering effects (Figs. 51A, 51B) and insulin-releasing activity (Figs. 51C, 51D) of GIP and fatty acid derivatised GIP analogues in 18 hour-fasted ob/ob mice. Plasma glucose 10 and insulin concentrations were measured prior to and after i.p. administration of glucose alone (18 mmoles/kg bw; o; white bars) as a control, or in combination with GIP (A; lined bars) or GIP analogues N-AcGIP(LysPAL 16) (M; grey bars) and N AcGIP(LysPAL37) (*; black bars) (25 nmoles/kg bw). The incremental area under the glucose or insulin curves (AUC) between 0 and 60 minutes are shown in the right 15 panels. Values represent means + SEM for 8 mice. *p<0.05, **p<0.01, ***p<0.001 compared to glucose alone, Ap<0.05, p<0.01 and .AAp<0.001 compared to native GIP, mnp<0.001 compared with N-AcGIP(LysPAL 16). Basal blood glucose levels of the experimental groups were not significantly different at the start of the study (p>0.05). After injection of glucose alone, plasma 20 glucose levels increased rapidly, attaining values of 40.3±1.5 mmol/l at 60 min. Native GIP reduced plasma glucose at each of the time points monitored, however, this failed to reach significance in terms of overall glucose excursion as revealed by the AUC values (Fig. 52B). Administration of N-AcGIP(LysPAL 16 ) and N AcGIP(LysPAL 37 ) produced a significant reduction in plasma glucose at each time 25 point (p<0.01 to p<0.001) and significantly lowered glucose AUC (p<0.001 to p<0.001) when compared to glucose alone. Additionally, N-AcGIP(LysPAL 16 ) and N-AcGIP(LysPAL 37 ) decreased the overall glucose excursion (p<0.05 to p<0.001) when compared to native GIP. The corresponding plasma insulin responses are illustrated in Figs. 51 C and 30 51 D. After administration of glucose alone (control) the maximal rise in plasma insulin was observed at 15 minutes, which then fell towards basal levels over the remaining 45 minutes. Administration of native GIP significantly elevated the overall insulinotropic response (p<0.05) compared with glucose alone. When N - 66 - WO 2005/082928 PCT/GB2005/000710 AcGIP(LysPAL 16 ) or N-AcGIP(LysPAL1 7 ) where administered together with glucose, a maximum plasma insulin concentration was observed at 15 minutes. Protracted biological activity for both analogues was clearly evident from 30 to 60 minutes. Glucose-mediated plasma insulin concentrations were significantly higher compared 5 in both control (p<0.01 to p<0.001) and GIP-treated animals (p<0.05 top<0.001). The corresponding AUC values for N-AcGIP(LysPAL 16 ) and N-AcGIP(LysPAL 37 ) revealed significant enhancements in overall glucose-mediated insulin release compared to native GIP (1.5-fold and 2.3-fold, respectively; p<0.01 top<0.001). N AcGIP(LysPAL 37 ) was significantly more potent (1.5-fold: p<0.00 l) than N 10 AcGIP(LysPAL 16 ) at stimulating insulin secretion. Dose-dependent metabolic effects in ob/ob mice. Figs. 52A and 52B illustrate the dose-dependent antihyperglycaemic and insulinotropic effects of GIP and the more potent analogue N-AcGIP(LysPAL 37 ) when administered with glucose to ob/ob mice. 15 They are are a pair of bar graphs showing dose-dependent effects of GIP and N AcGIP(LysPAL 37 ) in ob/ob mice fasted for 18 hours. The incremental area under the curve (AUC) for glucose (Fig. 52A) and insulin (Fig. 52B) between 0 and 60 minutes after i.p. administration of glucose alone (18 mmoles/kg bw; white bars) or in combination with GIP (lined bars) or N-AcGIP(LysPAL 37 ) (each at 6.25, 12.5 and 25 20 nmoles/kg bw; black bars). Values represent means ± SEM for 8 mice. -*p<0.01 and ***p<0.001 compared to glucose alone. AAp<0.01 and "AAp<0.001 compared to native GIP at the same dose. Data are presented as overall AUC responses for convenience. Expressed in this manner, native GIP did not significantly affect AUC glucose and insulin at any of 25 the doses tested. N-AcGIP(LysPAL 37 ) was substantially more potent than native GIP (p<0.0I to p<0.001) and exhibited prominent dose-dependent antihyperglycaemic and insulinotropic actions at all doses administered (Figs. 52A, 52B). Remarkably, even the lowest concentration of N-AcGIP(LysPAL 37 ) tested (6.25 nmoles/kg) had highly significant antihyperglycaemic properties compared to glucose alone (p<0.00 1). 30 Consistent with this observation, 6.25 nmoles/kg N-AcGIP(LysPAL 37 ) elicited a prominent insulin response (2.0-fold; p<0.01) compared to glucose alone. - 67 - WO 2005/082928 PCT/GB2005/000710 Long-acting effects in ob/ob mice. The effects of daily injection of N AcGIP(LysPAL") for 14 days on food intake, body weight, glycated hemoglobin and non-fasting plasma glucose and insulin concentrations of ob/ob mice are shown in Figs. 53A through 53E, which are a set of graphs showing the effects of daily N 5 AcGIP(LysPAL 37 ) (*; black bars) administration on food intake (Fig. 53A), body weight (Fig. 53B), plasma glucose (Fig. 53C), insulin (Fig. 53D) and glycated hemoglobin N-AcGIP(LysPAL") (12.5 nmoles/kg/day) (Fig. 53E). Native GIP (12.5 nmoles/kg/day; A; lined bars) or saline vehicle (control; 0; white bars) were administered for the 14-day period indicated by the horizontal black bar. Values are 10 means E SEM for 8 mice. *p<0.05, **p<0.01 compared to control. A p <0.01 compared to native GIP. GIP or N-AcGIP(LysPAL 37 ) had no effect on body weight or food intake (Figs. 53A, 53B). Plasma glucose and insulin concentrations were also unchanged by treatment with native GIP for 14 days (Figs. 53C, 53D). In contrast, daily injection of 15 N-AcGIP(LysPAL 3 7 ) resulted in a progressive lowering of plasma glucose, resulting in significantly (p<0.05) lowered concentrations at 14 days (Fig. 53C). At this time, glycated hemoglobin was also significantly (p<0.01) decreased in N AcGIP(LysPAL 37 ) treated ob/ob mice (Fig. 53E). These changes were accompanied by a tendency towards elevated insulin concentrations, but these did not achieve 20 statistical significance over the time frame studies (Fig. 53D). -68- WO 2005/082928 PCT/GB2005/000710 Effects of long term treatment of ob/ob mice with N-AcGIP(LysPAL 3 ) on glucose tolerance. Figs. 54A through 54D are a set of two line graphs (Figs. 54A, 54C) and two bar graphs (Figs. 54B, 54D) showing the effects of daily N-AcGIP(LysPAL 3 7) administration on glucose tolerance (Figs. 54A, 54B) and plasma insulin response 5 (Figs. 54C, 54D) to glucose. Tests were conducted after 14 daily injections of either N-AcGIP(LysPAL 37 ) (12.5 nmoles/kg/day; e; black bars), native GIP (12.5 nmoles/kg/day; A; lined bars) or saline vehicle (control; o; white bars). Glucose (18 mmoles/kg) was administered by intraperitoneal injection at the time indicated by the arrow. Plasma glucose and insulin AUC values for 0-60 minutes post injection are 10 shown in the right panels. Values are means = SEM for 8 mice. *p<0.05, **p<0.01, **-*p<0.001 compared to control. 'p<0.05, "p< 0

.

0 1 , AAp< 0

.

0 0 1 compared to native GIP. Consistent with effects on glycated hemoglobin, treatment of ob/ob mice for 14 days with N-AcGIP(LysPAL 3 7 ) resulted in a significant improvement in glucose 15 tolerance (Figs. 54A, 54B). Plasma glucose concentrations throughout the test and the overall 0-60 minutes AUC values were decreased (p<0.01 to p<0.00 1). This was accompanied by increased insulin concentrations during the latter stages (p<0.05) and a greater (p<0.0 1) overall AUC insulin response (Figs. 54C, 54D). In contrast, daily administration of native GIP had no effect on glucose tolerance or the plasma insulin 20 response to glucose compared with control ob/ob mice receiving saline injections for 14 days (Fig. 54). Effects long term treatment of ob/ob mice with N-AcGIP(LysPAL3 7) on insulin sensitivity, and effects of long term treatment ofob/ob mice with N-AcGIP(LysPAL 3 1) 25 on pancreatic insulin content. Figs. 55A through 55D are a line graph and three bar graphs showing the effects of daily N-AcGIP(LysPAL3 7 ) administration on insulin sensitivity (Figs. 55A, 55B) and pancreatic weight (Fig. 55C) and insulin content (Fig. 55D). Observations were conducted after 14 daily injections of either N AcGIP(LysPAL 37 ) (12.5 nmoles/kg/day; e; black bars), native GIP (12.5 30 nmoles/kg/day; A; lined bars) or saline vehicle (control; ii; white bars). In Fig. 55A, insulin (50 U/kg) was administered by intraperitoneal injection at the time indicated by the arrow. Plasma glucose AUC values for 0-60 minutes post injection are shown - 69 - WO 2005/082928 PCT/GB2005/000710 in the right panels. Values are means I SEM for 8 mice. *p<0.05, **p<0.01 compared to control. Ap<0.05, Ap<0.01 compared to native GIP. Insulin sensitivity of the 3 groups of mice after 14 days treatment is shown in Figs. 55A, 55B. Compared with ob/ob mice receiving daily injections of saline or 5 native GIP, N-AcGIP(LysPAL 37 ) prompted a significant improvement of insulin sensitivity. Both the individual glucose concentrations and 0-60 minutes AUC values were significantly different (p<0.01) from the other two groups. In contrast, daily treatment with native GIP did not affect the characteristic insulin resistance of ob/ob mice (Fig. 55A, 55B). 10 Treatment of ob/ob mice for 14 days with native GIP or N-AcGIP(LysPAL1 7 ) did not affect pancreatic weight compared with saline-treated controls (Figs. 55C, 55D). Similarly, pancreatic insulin content was similar in the GIP and saline treated groups. However, daily administration of N-AcGIP(LysPAL 3 1) significantly increased (p<0.01) insulin content compared with each of the other groups (Figs. 55C, 15 55D). Evaluation of GIP receptor desensitization after long term treatment of ob/ob mice with N-AcGIP(LysPAL 37 ). Figs. 56A through 56D are a set of two line graphs (Figs. 56A, 56C) and two bar graphs (Figs. 56B, 56D) showing the retention of glucose 20 lowering (Figs. 56A, 56B) and insulin releasing (Figs. 56C, 56D) activity of N AcGIP(LysPAL 37 ) and native GIP after daily injection for 14 days. Glucose (18 mmoles/kg) was administered by intraperitoneal injection alone (0; white bars) or in combination with either N-AcGIP(LysPAL 37 ) (e; black bars) or native GIP (A; lined bars) (both at 25 nmoles/kg) at the time indicated by the arrow. Plasma glucose and 25 insulin AUC values for 0-60 minutes post injection are shown in the right panels. Values are means + SEM for 8 mice. *p<0.0 5 , **p<0.01 compared to glucose alone. p< 0

.

0 5 , p< 0

.

0 1 compared to native GIP. Figs. 57A through 57D are a set of two line graphs (Figs. 57A, 57C) and two bar graphs (Figs. 57B, 57D) showing the acute glucose lowering (Figs. 57A, 57B) and insulin releasing (Figs. 57C, 57D) effects of 30 N-AcGIP(LysPAL 3 7 ) after 14 daily injections of either N-AcGIP(LysPAL37) (12.5 nmoles/kg/day; *; black bars), native GIP (12.5 nmoles/kg/day; A; lined bars) or saline vehicle (control; o; white bars). N-AcGIP(LysPAL 37 ) (25 nmoles/kg) was administered by intraperitoneal injection with glucose (18 mmoles/kg) at the time - 70 - WO 2005/082928 PCT/GB2005/000710 indicated by the arrow. Plasma glucose and insulin AUC values for 0-60 minutes post injection are shown in the right panels. Values are means ± SEM for 8 mice. *p<0.05, **p<0.01 compared to mice receiving control injections. Ap<0.05, AAp<0.01 compared to group receiving injections of native GIP. 5 As shown in Figs. 56A through 56D, treatment of ob/ob mice with N AcGIP(LysPAL 37 ) for 14 days did not prevent the ability of the peptide to significantly moderate the glycaemic excursion (p<0.01) and enhance plasma insulin concentrations (p<0.01) when administered acutely with intraperitoneal glucose. In contrast, the responses of ob/ob mice to acute administration of native GIP were 10 almost identical in mice receiving treatment with GIP or saline for 14 days (Figs. 56A - 56D). To further substantiate the lack of GIP receptor desensitization following chronic treatment with N-AcGIP(LysPAL 37 ), the acute effects of the analogue, administered with glucose, were examined in each of the 3 groups after 14 days treatment with N-AcGIP(LysPAL 3 ), native GIP or saline (Figs. 57A - 57D). Apart 15 from lower basal values in the former group, the glucose and insulin responses were identical with similar 0-60 minutes AUC measures for both plasma glucose and insulin concentrations. While this invention has been particularly shown and described with 20 references to preferred embodiments thereof, it will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the scope of the invention encompassed by the appended claims. -71-

Claims

1. A peptide analogue of GIP(1-42) (SEQ ID NO:1), comprising at least 12 amino acid residues from the N-terminal end of GIP(3 -42). 5

2. A peptide analogue of GIP(l-42) (SEQ ID NO:1), comprising at least 12 amino acid residues from the N-terminal end of GIP(1-42) and having an amino acid substitution at Glu 3 10

3. The peptide analogue of claims 1 or 2, wherein the amino acid substituted at Glu 3 is selected from the group consisting of: proline, hydroxyproline, lysine, tyrosine, phenylalanine and tryptophan.

4. The peptide analogue of claim 3, wherein proline is substituted for Glu. 15

5. The peptide analogue of claims I or 2, further comprising modification by fatty acid addition at an epsilon amino group of at least one lysine residue.

6 The peptide analogue of claim 5, wherein the modification is the linking of a 20 C-16 palmitate group to the epsilon amino group of a lysine residue.

7. The peptide analogue of claim 6, wherein the lysine residue is Lys 6

8. The peptide analogue of claim 6, wherein the lysine residue is Lys. 25

9. A peptide analogue of GIP(1-42) (SEQ ID NO:1), comprising at least 12 amino acid residues from the N-terminal end of GIP(1-42), and having an amino acid modification at amino acid residues 1, 2 or 3. 30

10. The peptide analogue of claim 9, wherein the N-terminal amino acid residue is acetylated. - 72 - WO 2005/082928 PCT/GB2005/000710

11. The peptide analogue of claim 10, further comprising modification by fatty acid addition at an epsilon amino group of at least one lysine residue.

12. The peptide analogue of claim 11, wherein the modification is the linking of a 5 C-16 palmitate group to the epsilon amino group of a lysine residue.

13. The peptide analogue of claim 12, wherein the lysine residue is Lys m6 .

14. The peptide analogue of claim 12, wherein the lysine residue is Lys. 10

15. Use of an analogue as claimed in any of the preceding claims in the preparation of a medicament for the treatment of obesity, insulin resistance, the insulin resistant metabolic syndrome (Syndrome X) or type 2 diabetes. 15

16. A pharmaceutical composition comprising the peptide analogue of any of the preceding claims.

17. The pharmaceutical composition of claim 16, further comprising a pharmaceutically acceptable carrier. 20

18. The pharmaceutical composition of claim 16, wherein the peptide analogue is in the form of a pharmaceutically acceptable salt.

19. The pharmaceutical composition of claim 16, wherein the peptide analogue is 25 in the form of a pharmaceutically acceptable acid addition salt.

20. A method of treating insulin resistance, the method comprising administering to a mammal in need of such treatment a therapeutically effective amount of the composition of claim 16. 30

21. A method of treating obesity, the method comprising administering to a mammal in need of such treatment a therapeutically effective amount of the composition of claim 16. - 73 - WO 2005/082928 PCT/GB2005/000710

22. A method of treating type 2 diabetes, the method comprising administering to a mammal in need of such treatment a therapeutically effective amount of the composition of claim 16. 5

23. A peptide analogue of GIP(1-42) (SEQ ID NO: 1), wherein the analogue comprises: a base peptide consisting of one of the following: GIP(1-12), GIP(1 13), GIP(1-14), GIP(1-15), GIP(1-16), GIP(1-17), GIP(1-18), 10 GIP(1-19), GIP(1-20), GIP(1-21), GIP(1-22), GIP(1-23), GIP(1-24), GIP(1-25), GIP(1-26), GIP(1-27), GIP(1-28), GIP(1-29), GIP( 1-30), GIP(1-31), GIP(1-32), GIP(1-33), GIP(1-34), GIP(1 -35), GIP(1-36), GIP(1-37), GIP(1-38), GIP( 1-39), GIP(1-40), GIP( 1-41) and GIP(1 -42); 15 which possesses one or more of the following modifications: an amino acid substitution at one or more of the residues; an amino acid substitution of lysine for one or more or the residues; an amino acid substitution at Glu 3 ; a modification by fatty acid addition at an epsilon amino group of at 20 least one lysine residue; and a modification by N-terminal acetylation.

24. The peptide analogue of claim 23, wherein the analogue has a proline substituted for Glu3. 25

25. The peptide analogue of claims 23, further comprising modification by fatty acid addition at an epsilon amino group of at least one lysine residue.

26. The peptide analogue of claim 25, wherein the modification is the linking of a 30 C-16 palmitate group to the epsilon amino group of a lysine residue.

27. The peptide analogue of claim 26, wherein the lysine residue is Lys1 6 . - 74 - WO 2005/082928 PCT/GB2005/000710

28. The peptide analogue of claim 26, wherein the lysine residue is Lys 37 .

29. Use of an analogue of claims 23-28 in the preparation of a medicament for the treatment of obesity, insulin resistance, the insulin resistant metabolic 5 syndrome (Syndrome X) or type 2 diabetes.

30. A pharmaceutical composition comprising the peptide analogue of claims 23 28. 10

31. The pharmaceutical composition of claim 30, further comprising a pharmaceutically acceptable carrier.

32. The pharmaceutical composition of claim 30, wherein the peptide analogue is in the form of a pharmaceutically acceptable salt. 15

33. The pharmaceutical composition of claim 30, wherein the peptide analogue is in the form of a pharmaceutically acceptable acid addition salt.

34. A method of treating insulin resistance, the method comprising administering 20 to a mammal in need of such treatment a therapeutically effective amount of the composition of claim 30.

35. A method of treating obesity, the method comprising administering to a mammal in need of such treatment a therapeutically effective amount of the 25 composition of claim 30.

36. A method of treating type 2 diabetes, the method comprising administering to a mammal in need of such treatment a therapeutically effective amount of the composition of claim 30. 30

37. A peptide analogue of GIP(1-42) (SEQ ID NO:1), comprising at least 12 amino acid residues from the N-terminal end of GIP(3-42), wherein the - 75 - WO 2005/082928 PCT/GB2005/000710 peptide analogue is resistant to degradation by enzyme DPP IV when compared to naturally-occurring GIP.

38. A peptide analogue of GIP(l-42) (SEQ ID NO:1), comprising at least 12 5 amino acid residues from the N-terminal end of GIP(1-42) and having an amino acid substitution at Glu3, wherein the peptide analogue is resistant to degradation by enzyme DPP IV when compared to naturally-occurring GIP.

39. A peptide analogue of GIP(l-42) (SEQ ID NO:1), comprising at least 12 10 amino acid residues from the N-terminal end of GIP(3-42), wherein the peptide analogue modulates insulin secretion.

40. A peptide analogue of GIP(1-42) (SEQ ID NO:1), comprising at least 12 amino acid residues from the N-terminal end of GIP(1-42) and having an 15 amino acid substitution at Glu 3 , wherein the peptide analogue modulates insulin secretion. - 76 -