A fusion peptide for activation of the apj and / or glp-1 receptors

Abstract

A Fusion Peptide for Activation of the APJ and / or GLP-1 Receptors This invention relates to a fusion peptide comprising a GLP-1 receptor agonist (GLP-1 RA) and an APJ receptor agonist; wherein the GLP-1 RA is covalently linked to the APJ receptor agonist by a link; and wherein the fusion peptide retains at least some activity of at least one of the GLP-1 RA and the APJ receptor agonist.

Description

[0001] A Fusion Peptide for Activation of the APJ and / or GLP-1 Receptors

[0002] This invention relates to a fusion peptide comprising a GLP-1 receptor agonist (GLP-1 RA) and an APJ receptor agonist; wherein the GLP-1 RA is covalently linked to the APJ receptor agonist by a link; and wherein the fusion peptide retains at least some activity of at least one of the GLP-1 RA and the APJ receptor agonist.

[0003] Introduction

[0004] Glucagon-like peptide-1 (GLP-1) receptor activation via GLP-1 mimetics produces benefits for glucose homeostasis through actions on the pancreas as well as at extra-pancreatic sites. In humans, the incretin hormone GLP-1 is released physiologically in response to nutrient absorption from enteroendocrine L-cells of the lower small intestine and colon. GLP-1 promotes insulin secretion from the pancreatic beta-cells, promotes beta-cell function and blocks alpha-cell glucagon production. These actions along with its actions on slowing of gastric emptying, enhancing glucose uptake in various tissues, (all promote anti-hyperglycaemic responses) and effects on central satiety pathways make it a very good candidate for obesity related Type 2 diabetes therapy.

[0005] Apelin is an adipokine which is found in different molecular forms, with the apelin-13 being the one that retains full biological activity. Apelin-13 retains a wide range of biological activities by activating its cognate APJ receptor in various tissues. Apelin has many beneficial cardioprotective effects on lowering blood pressure, as well as reducing vascular tone, protects again vascular fibrosis and blunting of oxidative stress. Apelin, which is upregulated by insulin, is also known to improve insulin sensitivity and modulate food intake. More recently, apelin analogs have been found to have antidiabetic actions in animal models of diabetes.

[0006] On their own, both GLP-1 and apelin-13 are subject to rapid degradation in plasma and therefore stable analogs of both offer the best hope for therapeutic benefit in humans. Overall, the combination of this wide range of beneficial metabolic activities from both GLP-1 and apelin in a single molecular fusion peptide could have enormous benefits for both diabetic and obesity therapies.

[0007] Here, we explore the actions of unimolecular apelin / GLP-1 fusion peptides in an in vitro setting with pancreatic BRIN-BD11 cells, in acute glucose lowering activities in normal healthy mice, as well as, in high fat fed diet induced obese (DIO) mice. Finally, chronic once daily administration of fusion peptides using longer acting acylated and non-acylated formats for 21 days in DIO mice was assessed.

[0008] US2015 / 030594 concerns an apelin moiety fused to a half-life extending moiety, optionally via a linker. Paragraphs

[0066] and

[0067] define APJ and a 77 amino acid apelin preprotein, as well as apelin- 36, apelin-17, apelin-16, apelin-13 and apelin-12. Paragraphs

[0109] to

[0116] define various bioconjugates, wherein three (or four) of the amino acids X1 to X13 are different from the corresponding amino acids present in Pyr-1-apelin-13. Paragraphs

[0302] to

[0306] define second agents with which the bioconjugate of the invention can be used in combination - they include DPP- IV inhibitors, GLP-1 and GLP-1 agonists. US2015 / 030594 does not disclose a fusion peptide comprising a GLP-1 receptor agonist (GLP-1 RA) and an APJ receptor agonist; wherein the GLP-1 RA is covalently linked to the APJ receptor agonist by a linker; and wherein the fusion peptide retains at least some activity of at least one of the GLP-1 RA and the APJ receptor agonist.

[0009] US2018 / 104306 concerns an APJ receptor agonist for use in the treatment or the prevention of diabetes. Paragraphs

[0058] to

[0062] disclose that the APJ receptor agonist may be apelin itself, including any polypeptide that comprises the apelin-13 C-terminal fragment, such as the apelin-17 or apelin-36 fragments. Paragraphs

[0120] to

[0122] disclose kit-of-parts comprising (i) an APJ receptor agonist, as defined above, and (ii) at least one anti-diabetic drug, each of (i) and (ii) as a combined preparation for simultaneous, separate or sequential use in the treatment of diabetes. Paragraph

[0127] discloses that the anti-diabetic drug may a glucagon-like peptide-l (GLP-1) receptor agonist. US2018 / 104306 does not disclose a fusion peptide comprising a GLP-1 receptor agonist (GLP-1 RA) and an APJ receptor agonist; wherein the GLP-1 RA is covalently linked to the APJ receptor agonist by a linker; and wherein the fusion peptide retains at least some activity of at least one of the GLP-1 RA and the APJ receptor agonist.

[0010] US2011 / 305663 concerns "SABA" refers to a Serum Albumin Binding Adnectins™. Adnectins™ are ligand binding scaffold proteins based on the tenth fibronectin type III domain, i.e., the tenth module of Fn3, (10Fn3). Page 59 discloses four fusion proteins (i) (GLP-1 (7-37) with an N-terminal Met, fused to a (GGGGS)s linker, fused to SABA1 .4 ,(ii) GLP-1 (7- 37) with an N-terminal Met, fused to an (ED)5linker, fused to SABA1 .4 (iii) SABA1 .5 fused to a (GGGGS)s linker, fused to GLP-1 (7-37), (iv) SABA1 .5 fused to an (ED)s linker, fused to GLP-1 (7-37). Page 93 discloses SABA1 .7, fused to an (ED)5linker, fused to APLNv4, and (ii) SABA1 .6, fused to a 6XHis and (GS)? linker, fused to APLNv4. There are five further APLN-SABA conjugates on page 94. US2011 / 305663 does not disclose a fusion peptide comprising a GLP-1 receptor agonist (GLP-1 RA) and an APJ receptor agonist; wherein the GLP-1 RA is covalently linked to the APJ receptor agonist by a linker; and wherein the fusion peptide retains at least some activity of at least one of the GLP-1 RA and the APJ receptor agonist. Table 1 sets out the structures of the ELA and ALE co-agonist fusion peptides.

[0011] Table 1 : Primary structure of ELA and ALE co-agonist peptides

[0012] Primary structure of exendin-4(1-30)-LINK-Apelin-13 amide (ELA) co-agonist fusion peptide

[0013] His-Gly-Glu-Gly-Thr-Phe-The-Ser-Asp-Leu-Ser-Lys-GIn-Met-Glu Glu-Glu-Ala-Val-Arg-Leu-Phe-lle-

[0014] Glu-Trp-Leu-Lys-Asn-Gly-Gly- (SEQ ID NO: 1 )

[0015] [LINK]-

[0016] Gln-Arg-Pro-Arg-Leu-Ser-His-Lys-Gly-Pro-Met-Pro-Phe-amide (SEQ ID NO: 2)

[0017] Primary structure of apelin-13-LINK-exendin-4(1-30) amide (ALE) co-agonist fusion peptide

[0018] Gln-Arg-Pro-Arg-Leu-Ser-His-Lys-Gly-Pro-Met-Pro-Phe-(SEQ ID NO: 2)

[0019] [LINK]-

[0020] His-Gly-Glu-Gly-Thr-Phe-The-Ser-Asp-Leu-Ser-Lys-GIn-Met-Glu Glu-Glu-Ala-

[0021] Val-Arg-Leu-Phe-lle-Glu-Trp-Leu-Lys-Asn-Gly-Gly-amide (SEQ ID NO: 1 )

[0022] The LINK used herein is a dimer of 8Ado that is shown below:

[0023] 8Ado = 8-Amino-3,6-dioxaoctanoic acid

[0024] 8-Amino-3,6-dioxaoctanoic acid, > 99% (HPLC)

[0025] The dimer is called 8Ado-8Ado herein. The invention is not limited to 8Ado-8Ado as the Link.

[0026] As exemplified herein, Apelin-13 was assessed with a C-terminal amide, as was the ELA fusion peptide and the ALE fusion peptide. However, the invention is not limited to fusion peptides that are C-terminally amidated.

[0027] Glucagon-like peptide-1 receptor agonists are also known as GLP-1 receptor agonists, GLP-1 DAs, incretin mimetics, GLP-1 mimetics, or GLP-1 analogs. Glucagon-like peptide 1 agonists are analogs of GLP-1 , a gut-derived peptide hormone that exhibits a glucose-lowering effect via stimulation of insulin secretion from pancreatic islets in response to an oral glucose load, known as the incretin effect. Unlike the natural peptide hormone, synthetic GLP-1 analogs are resistant to degradation by the dipeptidyl peptidase 4 (DPP-4) enzyme, and, thus, have a longer half-life.

[0028] 3

[0029] SUBSTITUTE SHEET (RULE 26) There are several GLP-1 receptor agonists on the market - these include exenatide twice-daily, exenatide extended-release (ER) once-weekly, lixisenatide once-daily, liraglutide once-daily, dulaglutide once-weekly, semaglutide once-weekly, and oral semaglutide once-daily. Albiglutide was discontinued in 2017 due to a decrease in sales:

[0030] The FDA approved the first GLP-1 receptor agonist - exenatide - in 2005. The oral semaglutide tablet was approved by the FDA in 2019. Except for oral semaglutide, most commercially available GLP-1 receptor agonists are administered by subcutaneous injection.

[0031] Human GLP-1 cannot be used as a therapeutic treatment due to its extremely short half-life. Various GLP-1 derivatives, GLP-1 receptor agonists (GLP-1 RA), or GLP-1 analogs reproduce the effects of GLP-1 , while having a longer half-life. One such FDA-approved drug is Byetta® (exenatide) which is a synthetic exendin-4(1-39).

[0032] According to a first embodiment of the invention, there is provided a fusion peptide comprising a GLP-1 receptor agonist (GLP-1 RA) and an APJ receptor agonist; wherein the GLP-1 RA is covalently linked to the APJ receptor agonist by a linker; and wherein the fusion peptide retains at least some activity of at least one of the GLP-1 RA and the APJ receptor agonist. Optionally, the linker, before linkage, comprises at least one amino carboxylic acid. Further optionally, the amino carboxylic acid is selected from amino butanoic acid, amino pentanoic acid, amino hexanoic acid, amino heptanoic acid, amino octanoic acid, amino nonanoic acid, amino decanoic acid, amino undecanoic acid and amino dodecanoic acid.

[0033] In am embodiment, the linker, before linkage, comprises two amino carboxylic acids in series. Optionally, the two amino carboxylic acids, which can be the same or different, are independently selected from amino butanoic acid, amino pentanoic acid, amino hexanoic acid, amino heptanoic acid, amino octanoic acid, amino nonanoic acid, amino decanoic acid, amino undecanoic acid and amino dodecanoic acid.

[0034] Optionally, the amino group of one amino carboxylic acid is linked to the GLP-1 receptor agonist and the carboxylic group of the same, or another, amino carboxylic acid is linked to the APJ receptor agonist; wherein, optionally, the, or both, links are an amide link. Alternatively, wherein the amino group of one amino carboxylic acid is linked to the APJ receptor agonist and the carboxylic group of the same, or another, amino carboxylic acid is linked to the GLP-1 receptor agonist; wherein, optionally, the, or both, links are an amide link.

[0035] In an embodiment, the, or each, amino carboxylic acid further comprises a (PEG)nchain, wherein n is 1 to 3; optionally wherein n is 2. Optionally, the, or each, amino carboxylic acid is 8-amino-3,6- dioxaoctanoic acid. Alternatively, the at least one amino carboxylic acid is selected from non-proteinogenic amino acids such as, but not limited to, p-amino acids that the amine group bonded to the second carbon away, and y-amino acids that the amine group bonded to the third carbon away; and D-amino acids.

[0036] Non-alpha amino acids include p-alanine, GABA, and 6-aminolevulinic acid.

[0037] In an embodiment, the GLP-1 RA is “short-acting” or “prandial.” “Short-acting” or “prandial” is understood to mean GLP-1 RA whose apparent half-life of elimination after subcutaneous injection in humans is less than 8 hours, in particular less than 5 hours, preferably less than 4 hours or else less than 3 hours, such as, for example, exenatide (also known as exendin-4) or lixisenatide.

[0038] Optionally, the GLP-1 receptor agonist is selected from exendin-3, exendin-4 (1-30), exendin-4 (1- 39), exendin-4 (3-30), exendin-4 (3-39), lixisenatide, truncations, analogs or derivatives thereof and pharmaceutically acceptable amides and other salts thereof.

[0039] Optionally, the GLP-1 receptor agonist is exendin-4 (1-30) and the exendin-4 (1-30) further comprises a modification at the Lys at position 12 or 27 of exendin-4(1-30) such as a fatty acid addition at an epsilon amino group and I or the exendin-4 (1-30) comprises a substitution to Lys at another position and the modification by fatty acid addition at an epsilon amino group of the substituted Lys. Further optionally, the GLP-1 receptor agonist is modified by a fatty acid addition at an epsilon amino group,

[0040] Optionally, the fatty acid group is selected from a C12 to C20 fatty acid or diacid; optionally a C12 to Cis fatty acid or diacid; further optionally a C16 fatty acid (palmitate), linked to residue 8 by a bivalent linker, optionally, a y-glutamyl linker.

[0041] Optionally, the added fatty acid further comprises a second carboxylic acid and is a fatty diacid.

[0042] Optionally, the GLP-1 receptor agonist is semaglutide - it differs from human GLP-1 by two amino acid substitutions at positions 8 and 34, where alanine and lysine are replaced by 2-aminoisobutyric acid and arginine, respectively. It is thought the amino acid substitution at position 8 prevents chemical breakdown by dipeptidyl peptidase-4. In addition, the lysine at position 26 is acylated with stearic diacid.

[0043] In an embodiment, the GLP-1 RA is selected from the group consisting of exenatide (Byetta®, ASTRA-ZENECA), lixisenatide (Lyxumia®, SANOFI), analogs or derivatives thereof and pharmaceutically acceptable salts thereof. In an embodiment, the GLP-1 RA is exendin-4 (exenatide) or Byetta®, analogs or derivatives thereof and pharmaceutically acceptable salts thereof.

[0044] Exendin-4 is a 39-amino acid peptide that was originally isolated from Heloderma suspectum venom. Exendin-4 differs from Exendin-3 by the identity of the amino acids at N-terminal positions 2 and 3 - they are Gly-Glu, instead of Ser-Asp, respectively.

[0045] Exenatide (exendin-4) is a peptide having the sequence:

[0046] HGEGTFTSDL SKQMEEEAVR LFIEWLKNGG PSSGAPPPS (SEQ ID NO: 3)

[0047] In an embodiment, the GLP-1 RA is lixisenatide or Lyxumia®, analogs or derivatives thereof and pharmaceutically acceptable salts thereof.

[0048] Lixisenatide is a peptide having the sequence:

[0049] HGEGTFTSDL SKQMEEEAVR LFIEWLKNGG PSSGAPPSKK KKKK-[NH2] (SEQ ID NO: 4).

[0050] Exenatide and lixisenatide are described in US2004 / 0023871 and W00104156, whose whole contents are included herein by reference.

[0051] The GLP-1 receptor agonist may be selected from exendin-4 (exenatide) or lixisenatide.

[0052] Apelin belongs to the adipokines group; they are endogenous peptide ligands for their receptor, APJ receptor (APLNR, angiotensin receptor like-1). APJ is a G protein-coupled receptor identified from the human gene by O’Dowd in 1993 and consists of 380 amino acids with 40% to 50% homology to the hydrophobic region of the angiotensin type I receptor. In 1998, the endogenous ligand of APJ was extracted and purified from bovine gastric secretions, naming it apelin. The natural apelin peptide is processed from the C-terminal portion of a precursor protein consisting of 77 amino acid residues. The post-translational processing of the precursor protein involves cleavage by endopeptidases to form C-terminal biologically active peptides, including apelin-13, -16, -17, -19, and -36. The last 23 residues of the C-terminus of apelin are identical in mammals, implying an important physiological role for these peptides. Studies using synthetic peptides have revealed that apelin-13 and apelin-36 may be the most abundant and biologically active fragments. Apelin-12 (i.e., apelin-13 minus phenylalanine (Phe) at the C-terminus) has similar binding affinity and activity at the APJ receptor to native apelin-13.

[0053] Apelin-13 is susceptible to additional posttranslational modifications, which result in the production of its more stable, pyroglutaminated form, called [Pyr1]-apelin-13 (also known as (pGlu1)apelin-13), which is the most abundant form in cardiac tissue. [Pyr1]-apelin-13 is a thirteen-membered polypeptide consisting of p-Glu, Arg, Pro, Arg, Leu, Ser, His, Lys, Gly, Pro, Met, Pro and Phe residues joined in sequence.

[0054] In an embodiment, the APJ RA is “short-acting” or “prandial.” “Short-acting” or “prandial” is understood to mean an APJ receptor agonist whose apparent half-life of elimination after subcutaneous injection in humans is less than 8 hours, in particular less than 5 hours, preferably less than 4 hours or else less than 3 hours, such as, for example, apelin-13.

[0055] In an embodiment, the APJ RA is apelin-13, amides, analogs or derivatives thereof and pharmaceutically acceptable salts thereof:

[0056] QRPRLSHKGPMPF

[0057] Such analogs or derivatives thereof and pharmaceutically acceptable salts thereof include, but are not limited to, analogs comprising at least residues 2 - 13 of apelin-13 and further comprising a substitution and / or modification at residue 13 of apelin-13.

[0058] The substitution at residue 13 of apelin-13 may be selected from:

[0059] Tyr, Thi (p[2-thienyl]-alanine), 4-azido-Phe, 4-cyano-Phe, or Trp; or

[0060] Vai or Ala; and I or wherein residue 13 is the C-terminal residue and the modification at residue 13, if present, is the substitution of the terminal carboxyl group by an amide group.

[0061] The analog may further comprise a substitution at any residue of apelin-13, wherein the substitution is a D-amino acid substitution for the corresponding L-amino acid of apelin-13.

[0062] In the analog, at least one peptide bond between adjacent residues of the peptide analog is replaced with a psi [CH2NH] bond or a stable peptide isotere bond.

[0063] The analog may further comprise an N-terminal addition of pGlu (5-oxoproline) or an acetyl, acyl or acylated moiety; optionally pGlu (5-oxoproline). Such an N-terminal addition may be added to apelin-13.

[0064] The analog may further comprise a modification (optionally at residue 8 of apelin-13) such as a fatty acid addition at an epsilon amino group of residue 8 and I or the analog may comprise a substitution to Lys and the modification by fatty acid addition at an epsilon amino group of the substituted Lys. The fatty acid group may be a C12 to C20 fatty acid or diacid, optionally a C16 fatty acid (palmitate), linked to residue 8 by a bivalent linker, optionally, a y-glutamyl linker. Such APJ receptor agonists include, but are not limited to, (Tyr13)apelin-13, Apelin-13-amide, (pGlu)apelin-13, pGlu (Tyr13)apelin-13, (pGlu)apelin-13-amide, Lys8GluPAL(Tyr13)apelin-13, (Lys8GluPAL)apelin-13-amide, pGlu(Lys8GluPAL)apelin-13-amide; further optionally, from the group comprising (Lys8GluPAL)apelin-13-amide and pGlu(Lys8GluPAL)apelin-13-amide.

[0065] In an embodiment, there is provided a pharmaceutical composition comprising the fusion peptide of any one of Claims 1 to 26, and a pharmaceutically acceptable excipient and is, optionally, formulated for parenteral administration, such as, but not limited to, as subcutaneous (s.c.), intramuscular or intravenous (i.v.) injections.

[0066] In solid phase peptide synthesis, the peptide is synthesised from the C-terminus to the N-terminus and the links are stitched in just like any other protected amino acid group in a seamless process. Once the N-terminal Fmoc is removed chemically in the synthesis cycle, the exposed NH2 end is free to be then linked to the COOH terminal end of the next amino acid in the sequence.

[0067] The fusion peptides of the present invention may be synthesised by such solid phase peptide synthesis. For the ELA fusion peptide in which the linker is 8-Amino-3,6-dioxaoctanoic acid (8Ado) (this has an amino group (NH2) at one end and a COOH group at the other end), the APJ receptor agonist (in this case apelin-13 amide) is immobilised on the resin and, stepwise, at least one (in this case, two) protected linkers are attached to the immobilised apelin-13 amide. Thus, 8Ado is stitched into the immobilised sequence. During synthesis, the terminal NH2 is protected by Fmoc and the Fmoc is removed to allow the next 8Ado to be added, after which their protecting group is removed. The most C-terminal amino acid of the GLP-1 RA (in this case, Gly) may be immobilised onto the free end of peptide on the resin. Then, the growing peptide (now the C-terminal end of exendin-4(1- 30)) is linked to the NH2 end of this linker. Each other amino acid is then immobilised, in turn, onto the peptide on the resin. This is schematically illustrated as follows:

[0068] The final linker is 2 molecules of 8Ado-8Ado. The Fmoc-Ado is stitched into the peptide sequence in the same way that any Fmoc amino acid is as shown above. The synthesis takes place from the C- terminus towards the N-terminus attached to a resin. In the final steps, the peptide is removed from the resin and purified by reversed phase HPLC usually.

[0069] In the drawings:

[0070] Fig. 1a shows a reversed-phase HPLC trace of Exendin-LINKER-Apelin (ELA) indicating a retention time of 21 .32 min. The sample was run on a Vydac C-8 (250 x 4.6 mm analytical column) at a flow rate of 1.0 ml / min. The mobile phases were 0.12% TFA in HPLC grade water and 0.1 % TFA in 70% acetonitrile 29.9% water. The sample absorbance was measured at 214 nm and gradient elution was applied with increasing acetonitrile concentration indicated by the broken line. The peptide purity calculated as >95%.

[0071] Fig. 1 b shows a reversed phase HPLC trace of Apelin-LINKER-Exendin (ALE). The separation was performed on a Vydac C-8 analytical column (250 x 4.6 mm) to obtain ALE profile using a linear gradient from 0-100% acetonitrile over 30 min as shown. The ALE peak identified had a retention time of 22.01 min. The peptide purity calculated as >95%.

[0072] Fig 1 a shows ELA and Fig. 1 b shows ALE. The respective RP-HPLC retention times depend on how hydrophobic the peptide is and this parameter varies slightly between peptide candidates.

[0073] Fig. 2a shows matrix-assisted laser desorption ionisation-time of flight (MALDI-TOF) mass spectrum of Exendin-4-Linker-Apelin-13 (ELA) amide with molecular mass 5083.72 Da (theoretical Mr 5084.0). The peptide (1 pl) was mixed with 0.5 pl of a matrix solution (10 mg / ml solution of-cyano-4- hydroxycinnamic acid in acetonitrile / ethanol) and mass spectrum recorded using a Voyager-DE BioSpectrometry Workstation. Masses were recorded as mass-to-charge (m / z) ratio (X-axis) against relative peak intensity (Y-axis). Trace shows major peaks for the singly and doubly charged ions with molecular mass 5084.72 and 2543.19 Da, respectively.

[0074] Fig. 2b shows matrix-assisted laser desorption ionisation-time of flight (MALDI-TOF) mass spectrum of Apelin-13-LINKER-Exendin-4 (ALE) amide with molecular mass 5084.99 Da. The peptide (1 pl) was mixed with 0.5 pl of a matrix solution (10 mg / ml solution of-cyano-4-hydroxycinnamic acid in acetonitrile / ethanol) and mass spectrum recorded using a Voyager-DE BioSpectrometry Workstation. Masses were recorded as mass-to-charge (m / z) ratio (X-axis) against relative peak intensity (Y-axis). Trace shows major peaks for the singly and doubly charged ions with molecular mass 5084.99 and 2543.48 Da, respectively. Acute in vitro cell studies

[0075] Fig. 3 shows Apelin and Exendin-4 peptides and analog effects on insulin secretion from isolated BRIN-BD1 1 cells. BRIN-BD1 1 cells are immortalised clonal cells derived from RinM5f cells and NEDH rat beta cells produced by electrofusion. Cells and peptides were co-incubated for 20 min at a range of concentrations (10-12to 10-6M) prior to insulin RIA analysis. Values are mean ± S.E.M. (n=8), ***p<0.001 versus 5.6 mM glucose control (presented as dash line),AAAp<0.001 versus Exendin(1-30) andmp<0.001 versus Apelin-13. The five coloured columns (in the blue to purple range) for each peptide concentration, from left to right, comprise Apelin-13; a mixture of Apelin-13 & Exendin-4(1-30); a mixture of Apelin-13 & Exendin-4(1-39); Exendin-4(1-30); a fusion peptide comprising exendin-4(1-30)-LINKER-Apelin-13 amide (also known as ELA). All test peptides are in their amide form.

[0076] Exendin-4(1-30) was included in the fusion peptide because it was known to retain bioactivity as a GLP-1 receptor agonist and, from a synthetic viewpoint, it is easier to synthesise and cheaper than the larger exendin-4(1-39) amino acid counterpart with 9 less amino acids.

[0077] When peptide mixtures were administered, the same concentration of each peptide was used. So, for example, at 10-6M peptide in Fig. 3, the middle coloured bar represents Apelin- 13 amide at 10-6M and Exendin-4(1 -39) at 10-6M. So, both were at 10-6M.

[0078] The same is true for all peptide mixtures in the figures that follow.

[0079] Fig. 4 shows acute effect of apelin and exendin peptides and their analogs on insulin secretion from BRIN-BD1 1 pancreatic beta cells. Cells were co-incubated with peptides at varying concentrations (10-12to 10-6M) for 20 min followed by determination of insulin secretion using RIA. Values are expressed as mean ± S.E.M. (n=8). ***p<0.001 versus 5.6 mmol glucose control (dashed line),AAAp<0.001 versus exendin(1-30), andtp<0.001 versus apelin-13. The left-most column is KCI (30 nM). For each peptide concentration, the five columns are, from left to right, Apelin-13; Apelin- 13+Exendin-4(1-30); Apelin-13+Exendin-4(1 -39); Exendin-4(1 -30); Apelin-13-LINKER-Exendin-4(1- 30) (also known as ALE).

[0080] Figs. 3 and 4 show greater insulin secretion from clonal pancreatic BRIN-BD11 cells for the exendin - apelin fusion peptide (in either order) than when both of the peptides are given separately. Fig. 3 shows that ELA is much more effective than a mixture of Ex-4 (Exendin-4) + Apelin-13 administered separately. This is a little less marked in the case of ALE (Fig. 4). There might be explained by the relative stability of these fusion peptides. If ELA is much more stable than ALE, then it is possible that in vivo effects of the latter will be short lived, even though the in vitro insulin responses look quite good for ALE. Fig. 5 shows dose-dependent acute effects of ELA and GLP-1 R and APJR antagonists on insulin secretion from clonal pancreatic BRIN-BD1 1 cells in the presence of 5.6 mM glucose. The effect of the fusion ELA peptide alone and in combination with either the GLP-1 R antagonist (Exendin-4(9-39) or APJR antagonist (Val13)Apelin-13 (10-8M each) or both antagonists on insulin secretion is shown in BRIN-BD11 cells at a range on concentrations as shown. Cells were co-incubated for 20 min prior to RIA analysis of insulin secretion. Values are mean ± S.E.M. (n=8). ***p<0.001 versus 5.6 mM glucose control (shown as dashed line). The left-most columns are glucose (5.6 mM) and KCI (30 nM). For each peptide concentration, the four columns are, from left to right, ELA + Exendin-4(9-39) at 108M + (Val13)Apelin-13 at 108M; ELA + Exendin-4(9-39) at 108M; ELA + (Val13)Apelin-13 at 10’8M; Exendin-4(1-30)-LINKER-Apelin-13 amide (also known as ELA).

[0081] The knockdown shown in Fig. 5 shows that more activity is lost with the GLP-1 R antagonist Ex-4(9- 39) say 60% reduction from ELA response alone. For the Val13Apelin-13, the loss in insulinotropic activity was approx. 40% compared to ELA. Therefore, the ELA seems to be mostly a GLP-1 agonist and slightly less an APJ receptor agonist. Both receptor antagonists, given together, wipe out the ELA action completely.

[0082] Fig. 6 shows close-dependent acute effects of ALE and antagonists on insulin secretion from clonal pancreatic BRIN-BD11 cells 5.6 mM glucose. Acute effects of ALE peptide alone and in combination with either the GLP-1 R antagonist (Exend in-4(9-39)) or APJR antagonist (Val13)Apelin-13 (10-8M each) or both antagonists on insulin secretion in BRIN-BD1 1 cells at a range on concentrations as shown. Values represent mean ± SEM for n=8 where ***p<0.001 compared to 5.6 mM glucose (shown as dashed line). The left-most columns are glucose (5.6 mM) and KCI (30 nM). For each peptide concentration, the four columns are, from left to right, ALE+Exendin-4(9-39) 10-8M + (Val13)Apelin-13 108M; ALE+Exendin-4(9-39) 108M; ALE+(Val13)Apelin-13 108M; Apelin-13- LINKER-Exendin-4(1-30) amide (ALE).

[0083] The combination of both antagonists completely blocks the effect observed with the ALE fusion peptide. Both GLP-1 and APJ receptor antagonists seem to show about 50% reduction each.

[0084] Figs. 5 and 6 show dose-dependent acute effects of ELA and ALE and antagonists on insulin secretion from clonal pancreatic BRIN-BD11 cells (showing a probable mechanism of action). When both antagonists are used together, the response is back to the level of glucose alone (where no peptide was applied) on the extreme left of each graph. This means, if specific antagonists wiped out the response when used in combination, as is shown in both figures, that any insulinotropic activity could be completely explained by GLP-1 R and APJR actions. Thus, this shows that each part of insulinotropic effect of the fusion peptide can be knocked down when a specific GLP-1 or APJ receptor antagonist is used and the action is obliterated completely when both are used together (Fig. 5 and Fig. 6). Acute in vivo studies in mice

[0085] Fig. 7 shows the model for assessing food intake in trained mice.

[0086] Fig. 8 shows ELA and related peptide effects on food intake in high fat-fed NIH Swiss mice (A) immediately (t=0 h), and (B) 8 h prior to food presentation. Mice were fasted overnight (21 h) and peptides were administered by intraperitoneal (i.p.) injection at 25 nmol / kg BW. Values are Mean ± S.E.M. (n=6). *p<0.05, **p<0.01 , and ***p<0.001 versus saline vehicle control. At each time-point in Figs. 8A and 8B, the columns are, from left to right, vehicle, Apelin-13 amide, Exendin-4(1-39); Exendin-4(1-30); a mixture of Exendin (1-30) + Apelin-13 amide; ELA.

[0087] Fig.8 shows that ELA was more potent and long lasting than a mixture of exendin-4(1-30) and apelin-13 amide given together but separately.

[0088] Fig. 9 shows the acute effect of intraperitoneal injection of ELA alone or in combinations with specific APJR and GLP-1 R antagonists on food intake. Acute effects of ELA and the combinations of this peptide and its antagonists Exend in-4(9-39) and (Val13)apelin-13 at a dose of 25 nmol / kg on food intake in female C57BL / 6 mice. Peptides were administered by i.p. injection to 21 h fasted trained mice and cumulative food intake was monitored. Values are means ± SEM (n=7-8), ***p<0.001 compared to saline control. Fig. 9 shows likely receptor mediated effects of the fusion peptide and that specific receptor antagonists for GLP-1 and the apelin APJ receptors can block this action when both antagonists are tested together with the fusion peptide. At each time-point, the columns are, from left to right, saline, ELA, ELA+(Val13)Apelin-13+Exendin-4(9-39).

[0089] Earlier insulin data herein showed that each antagonist had a part to play in knocking down the ELA action. The combination removed activity completely back to glucose alone in the earlier data. The same is true here, where there is no significant difference between saline control and the mixture of ELA+(Val13)Apelin-13+Exendin-4(9-39).

[0090] Fig. 10 shows the acute effect of intraperitoneal injection of ELA and ALE on food intake in trained mice. Acute effects of two fusion peptides of apelin-13 and Exendin-4(1-30), each at a dose of 25 nmol / kg on food intake in female NIH Swiss mice. Peptides were administered by i.p. injection to 21 h fasted trained mice and cumulative food intake was monitored. Values are means ± SEM (n=8), ***p<0.001 compared to saline control. This verifies what was seen previously in terms of one fusion peptide (co-agonist) working (ELA) and one having no effect (ALE) in fasted mice on food intake in this model system. This was repeated in different mice compared to original study but produced the same result (data not shown). At each time-point, the columns are, from left to right, saline; Exendin- 4 / Apelin-13 (ELA); Apelin-13 / Exendin-4 (ALE).

[0091] Fig. 10 shows that the reverse fusion peptide (ALE) was almost completely ineffective in reducing food intake. Perhaps a very short half-life in vivo might help explain this. Some early in vitro plasma degradation was suggestive of this (not shown) and the delayed response GTT later gave some more insight on this stability issue.

[0092] Assessing glucose tolerance in healthy control mice

[0093] Fig. 11 shows the effect on blood glucose when a delayed ipGTT (18 mmol / kg BW glucose) was performed. Fig. 11 (A) compares fusion peptides of exendin-4 and apelin-13 (each order) and mixtures of exendin-4 and apelin-13 (given at 25 nmol / kg bodyweight (BW)) 2 hours in advance of the ip glucose load (2 h delayed) to 4 h fasted mice. Fig. 11 (B) shows ELA fusion peptide and mixtures of apelin-13 and exendin-4(1-30) (4 h delayed) were administered by i.p. injection to 4 h fasted mice. Values are mean ± SEM (n=8) where *p<0.05, **p<0.01 and ***p<0.001 compared to glucose control. The insert bar graph show glucose, ELA, ALE and the mixture of Apelin-13 plus EX- 4(1-30) from left to right.

[0094] Figs. 11A and B show that the hybrid peptide, ELA, was effective in significantly reducing glucose levels when an ipGTT (18 mmol / kg glucose) was performed 2 h and 4 h later, respectively.

[0095] Fig. 12 shows the effects on glucose tolerance in high fat-fed NIH Swiss mice from ELA and related peptides. Peptides were administered by i.p. injection at 25 nmol / kg BW. (A) immediately (t=0), (B) 8 h, (C) 24 h, and (D) 36 h prior to being injected i.p. with 18 mmol / kg BW glucose. Respective area under the curve data are shown below each graph (the columns, from left to right, are vehicle, apelin-13 amide, exendin-4(1-30), mixtures of exendin-4(1-30) + apelin-13 amide and exendin- 4 / apelin-13 (also known as ELA) fusion peptide. Values are mean ± S.E.M. (n=6). *p<0.05, **p<0.01 , and ***p<0.001 versus the glucose control. In Fig. 12A, the columns, from left to right, are vehicle, apelin-13 amide, exendin-4(1-30), mixtures of exendin-4(1-30)+apelin-13 amide, exendin- 4 / apelin-13 (ELA) fusion peptide. In Fig. 12B, the columns, from left to right, are vehicle, exendin- 4(1-30), mixtures of exendin-4(1-30)+apelin-13 amide, exendin-4 / apelin-13 (ELA) fusion peptide. In Figs. 12C and D, the columns, from left to right, are vehicle and exendin-4 / apelin-13 (ELA) fusion peptide.

[0096] The ELA fusion peptide was effective in reducing glucose levels when an ipGTT (18 mmol / kg glucose) was performed up to 24 hours later.

[0097] Figs. 11 and 12 show that the ELA peptide was the most effective and longest acting agent in improving glucose tolerance following an ipGTT in normal mice and was still effective after 24 h.

[0098] Fig. 13 compares of ELA, ALE and related peptides on glucose tolerance in high fat-fed male NIH Swiss mice. Animals were administered peptides i.p. at 25 nmol / kg BW immediately prior to 18 mmol / kg BW i.p. glucose injection, with area under the curve data directly beside. Values are expressed as mean ± S.E.M. (n=6). ***p<0.001 versus glucose-only control. The respective areas under the curve data are shown beside the graph (the columns, from left to right, are vehicle, ELA fusion peptide, mixture of exendin-4(1-30)+apelin-13 (referred to as “combination” in the graph), exendin-4(1-30), apelin-13 amide, ALE fusion peptide.

[0099] This effect on improving glucose tolerance was also observed (see Fig. 13) following an acute ipGTT in a mouse model of high fat feeding.

[0100] Assessing bioactivity of next generation acylated co-agonist peptides

[0101] The top line in the table above relates to ELA fusion peptide (in its amide form). The following 3 lines relate to acylated analogs of the fusion peptide, ELA amide; wherein the fusion peptide is acylated (with y-GluPAL, or C16 fatty acid (palmitate) linked by a bivalent linker that is a y-glutamyl linker) at Lys position 12, 27 or 38 of the fusion peptide, ELA amide.

[0102] Fig. 14 shows the effects of parent peptides on insulin release from BRIN-BD11 cells at 5.6 mM glucose. BRIN-BD11 cells were incubated (20 min) with a range of concentrations (10-12to 10-6mol / l) of test peptides in the presence of 5.6 mmol / l glucose, and insulin was measured using RIA. Values represent means ± SEM (n=8). *P<0.05, **P<0.01 and ***P<0.001 compared to respective glucose control.AP<0.05 andAAAP<0.001 compared to apelin-13 alone. At each peptide concentration, the columns, from left to right, are liraglutide, exendin-4(1-30), apelin-13 amide and a mixture of exendin-4 (1-30) and apelin-13 amide.

[0103] Fig. 15 shows the effects of ELA fusion peptides on insulin release from BRIN-BD11 cells at 5.6 mM glucose. BRIN-BD11 cells were incubated (20 min) with a range of concentrations (10-12to 10-6mol / l) of test peptides in the presence of 5.6 mmol / l glucose, and insulin was measured using RIA. Values represent means ± SEM (n=8). **P<0.01 and ***P<0.001 compared to respective glucose control.AP<0.05,AAP<0.01 andAAAP<0.001 compared to ELA (non-acylated). At each peptide concentration, the columns are, from left to right, non-acylated ELA fusion peptide, ELA fusion peptide acylated, respectively, at the Lys at position 12, 27 or 38 with y-GluPAL. Fig. 15 shows that ELA acylated at position 27 is less effective at stimulating insulin release than non-acylated ELA. ELA acylated at Lys12and Lys38retain activity in their acylated forms.

[0104] Fig. 16 shows the effects of ELA fusion peptides alone and in combination with antagonists on insulin release from BRIN-BD11 cells at 5.6 mM glucose. BRIN-BD11 cells were incubated (20 min) with a range of concentrations (10-12to 10-6M) of fusion peptides alone, or in combination with antagonist for apelin-13, exendin-4 or both, in the presence of 5.6 mmol / l glucose and insulin measured using RIA. Values represent means ± SEM (n=8). *P<0.05, **P<0.01 and ***P<0.001 compared to respective glucose control.AP<0.05,AAP<0.01 andAAAP<0.001 compared to test peptide alone.

[0105] The graphs are A to C (left to right); D and E are under A to C and are also left to right.

[0106] Fig 16A compares insulin release and, within each test peptide concentration, the columns are, from left to right, a mixture of exendin-4 (1-30) + apelin-13 amide, a mixture of exendin-4 (1 -30) + apelin-

[0107] 13 amide + a fixed concentration of Vai13apelin-13 (10-8M), a mixture of exendin-4 (1 -30) + apelin-

[0108] 13 amide + a fixed concentration of exend in-4(9-39) (10-8M), a mixture of exendin-4 (1-30) + apelin-

[0109] 13 amide + a fixed concentration of Vai13apelin-13 (10-8M) + a fixed concentration of exendin-4(9-

[0110] 39) (10-8M).

[0111] Fig 16A shows that including either or both of the APJR and GLP-1 R antagonists blocks the insulin secretion effect observed with mixtures of exendin-4 + apelin-13 amide. Thus, the fusion peptide would appear to be operating through the APJ and GLP-1 receptors since such antagonists reduce its efficacy.

[0112] Fig. 16B compares insulin release and, within each test peptide concentration, the columns are, from left to right, ELA fusion peptide, ELA fusion peptide + a fixed concentration of Vai13apelin-13 (10-8M), ELA fusion peptide + a fixed concentration of exendin-4(9-39) (10-8M), ELA fusion peptide + a fixed concentration of Vai13apelin-13 (10-8M) + a fixed concentration of exendin-4(9-39) (10-8M).

[0113] Fig. 16B shows that the GLP-1 R antagonist (exendin-4(9-39) is more effective at blocking insulin release that the APJR antagonist (Vai13apelin-13). It would appear that, from the in vitro insulin secretion, about 60% of the activity seems to be GLP-1 based and 40% is associated with APJ receptor activation.

[0114] Fig. 16C compares insulin release and, within each test peptide concentration, the columns are, from left to right, ELA fusion peptide that is acylated at position 12, ELA fusion peptide acylated at position 12 + a fixed concentration of Vai13apelin-13 (10-8M), ELA fusion peptide acylated at position 12 + a fixed concentration of exend in-4(9-39) (10-8M), ELA fusion peptide acylated at position 12+ a fixed concentration of Vai13apelin-13 (10-8M) + a fixed concentration of exendin-4(9-39) (10-8M). Fig 16C shows a dose-dependent effect of ELA fusion peptide acylated at position 12 on insulin release. Fig.16C also shows that the effect on insulin release is blocked by either the APJR antagonist or the GLP-1 R antagonist.

[0115] Fig. 16D compares insulin release and, within each test peptide concentration, the columns are, from left to right, ELA fusion peptide that is acylated at position 27, ELA fusion peptide acylated at position 27 + a fixed concentration of Vai13apelin-13 (10-8M), ELA fusion peptide acylated at position 27 + a fixed concentration of exend in-4(9-39) (10-8M), ELA fusion peptide acylated at position 27+ a fixed concentration of Vai13apelin-13 (10-8M) + a fixed concentration of exendin-4(9-39) (10-8M).

[0116] Fig 16D shows a dose-dependent effect of ELA fusion peptide acylated at position 27 on insulin release, albeit to a lesser extent than analogs acylated at Lys residues in position 12 or 38. Fig.16D also shows that the effect on insulin release is blocked by either the APJR antagonist or the GLP-1 R antagonist.

[0117] Fig. 16E compares insulin release and, within each test peptide concentration, the columns are, from left to right, ELA fusion peptide that is acylated at position 38, ELA fusion peptide acylated at position 38 + a fixed concentration of Vai13apelin-13 (10-8M), ELA fusion peptide acylated at position 38 + a fixed concentration of exend in-4(9-39) (10-8M), ELA fusion peptide acylated at position 38 + a fixed concentration of Vai13apelin-13 (10-8M) + a fixed concentration of exendin-4(9-39) (10-8M).

[0118] Fig 16E shows a dose-dependent effect of ELA fusion peptide acylated at position 38 on insulin release. Fig.16E also shows that the effect on insulin release is blocked by either the APJR antagonist or the GLP-1 R antagonist.

[0119] The combination of antagonists was a little less potent at completely removing bioactivity of ELA peptides as shown by the reduction in black bars (right-most bar in each figure) which was not completely down to glucose control levels.

[0120] Fig.17 shows the effects of ELA fusion peptides on proliferation and protection against apoptosis in rodent clonal pancreatic beta-cells (BRIN-BD1 1). BRIN-BD11 cells were incubated overnight (18 h) with test peptides (each at 10-8or 10-6M). (A) Proliferation was measured using Ki-67 immunocytochemistry. (B) TUNEL positive apoptotic cells were assessed following 2 h exposure to a cytokine cocktail (IL-1 p 100 U / mL, IFN-y 20 U / mL, TNF-a 200 U / mL) with or without co-culture in the presence of test peptides (each at 10-8and 10-6M). Values represent means ± SEM (n=3). (A) **P<0.01 and ***P<0.001 compared to respective media control.AAP<0.01 compared to ELA 10-6M. .++P<0.01 compared to ELA Lys1210-6M (B) *P<0.05, **P<0.01 and ***P<0.001 compared to respective media control.+P<0.05,++P<0.01 and+++P<0.001 compared to cytokine cocktail. The cytokine cocktail is designed to cause an inflammatory response and comprises IL-1 p 100 U / mL, IFN-y 20 U / mL, TNF-a 200 U / mL. Fig.17A shows a dose-dependent effect by ELA fusion peptide on cell proliferation that is either nonacylated or is acylated at position 12 or 38.

[0121] Fig.17B shows a dose-dependent effect by ELA fusion peptide at reducing rodent pancreatic betacell apoptosis that is either non-acylated or is acylated at position 12 or 38.

[0122] The bars show responses at 10-6M or 10-8M for each agent. ELA promotes beta cell proliferation over the media control in Fig. 17A. ELA reduced beta cell death (apoptosis) in Fig. 17B. The effects are both dose-dependent and work better at the higher 10-6M concentration.

[0123] ELA and related analogs - acute in vivo study results

[0124] Fig. 18 shows the effects of ELA fusion peptides on cumulative food intake in 21 -h fasted trained lean mice. Cumulative food intake was measured in 21 -h fasted trained lean mice at 30, 60, 90, 120, 150, and 180 min after i.p. injection at (A) 0.25 nmol / kg body weight, from left to right saline vehicle (0.9% w / v NaCI), ELA fusion peptide, ELA Lys12or ELA Lys38; at (B) 1 nmol / kg body weight from left to right saline vehicle (0.9% w / v NaCI), ELA fusion peptide, ELA Lys12, ELA Lys27or ELA Lys38; and at (C) 2.5 nmol / kg body weight, from left to right saline vehicle (0.9% w / v NaCI), ELA fusion peptide, ELA Lys12, ELA Lys27or ELA Lys38. Values represent means ± SEM (n=8). *P<0.05, **P<0.01 and ***P<0.001 compared to respective saline control.AAP<0.01 andAAAP<0.01 compared to ELA.

[0125] Fig. 18 shows i.p. injected ELA and ELA(Lys12) significantly reduced acute food intake (up to 180 min) in trained mice in a dose dependent manner. The ELA(Lys38) also was effective. However, ELA(Lys27) was not effective at 2.5 nmol / kg body weight and was thus not tested at the lower concentrations shown here (illustrated in Fig. 18C).

[0126] Fig. 19 shows dose-dependent effects on cumulative food intake by fusion peptides in 21-hfasted trained lean mice. Cumulative food intake was measured in 21 h fasted trained mice at 30, 60, 90, 120, 150, and 180 min following i.p. injection of saline vehicle (0.9% w / v NaCI) or test peptides) at (A) 0.25, (B) 2.5 or (C) 2.5 nmol / kg body weight. Values represent means ± SEM (n=6 / 8). *P<0.05, **P<0.01 and ***P<0.001 compared to respective saline control.AAP<0.01 andAAAP<0.001 compared to ELA (non-acylated).

[0127] Fig. 19A compares food intake for saline control, apelin-13 amide, exendin-4(1-30) and ELA fusion peptide at a lower dose of 0.25 nmol / kg body weight. Food intake was markedly reduced following administration of both ELA fusion peptide and exendin-4(1-30).

[0128] Fig. 19B compares, at 2.5 nmol / kg body weight, food intake for saline control, ALE fusion peptide, a mixture of exendin-4(1-30) + apelin-13 amide, and ELA fusion peptide. The ALE fusion peptide is much less effective than the ELA fusion peptide. Fig. 19C compares, at 2.5 nmol / kg body weight, food intake for saline control, ELA fusion peptide and its derivatives acylated at positions Lys12, Lys27and Lys38. Fig. 19C shows that ELA fusion peptide, ELA(Lys12) and ELA(Lys38) significantly reduced food intake over 180 min in trained mice. ELA(Lys27) was marginally effective but less so than the other 2 acylated fusion peptides.

[0129] Fig. 20 shows the effect of administration of ELA and its component peptides on food intake in trained fasted (21 h) male NIH Swiss mice immediately before (A) or 8 h (B) prior to food presentation. Immediately after administration (A), all peptides except apelin-13 amide produced a significant reduction in food intake at all time points. However, 8 h after administration, ELA as well as the exendin-4 peptides and the mixture of component peptides reduced food intake from 30-180 min but apelin-13 amide was ineffective. Values are represented as mean ± S.E.M. (n=8). *p<0.05, **p<0.01 , ***p<0.001 versus saline control.

[0130] Fig. 21 shows the effects of ELA fusion peptides in combination with antagonists on cumulative food intake in 21 -h fasted trained lean mice. Cumulative food intake was measured in 21 h fasted trained mice at 30, 60, 90, 120, 150, and 180 min after i.p. injection of saline vehicle (0.9% w / v NaCI) or test peptides (each at 2.5 nmol / kg body weight) in combination with its antagonists (each at 25 nmol / kg body weight). Values represent means ± SEM (n=6). **P<0.01 and ***P<0.001 compared to respective saline control.

[0131] Fig. 22 shows persistent effects of fusion peptides on cumulative food intake in 21 -hour fasted trained lean mice. Cumulative food intake was measured in 21 h fasted trained mice at 30, 60, 90, 120, 150, and 180 min following i.p. injection of saline vehicle (0.9% w / v NaCI) or fusion peptides (each at 25 nmol / kg body weight) provided at t=0 (A), -6 h (B), -12 h (C), -21 h (D), -42 h (E), or -63 h (F). Values represent means ± SEM (n=7 / 8). *P<0.05, **P<0.01 and ***P<0.001 compared to respective saline control.AP<0.05,AAP<0.01 andAAAP<0.001 compared to ELA (non-acylated).QP<0.05 compared to ELA Lys12.

[0132] Figs 22A and B compare cumulative food intake, at 0 h delay and 6 h delay time points, for saline control, non-acylated ELA, and ELA acylated at Lys 12, 27 or 38. Each fusion peptide, other than ELA acylated at Lys27reduced cumulative food intake at these two timepoints. At the 6 hour delay (Fig. 22B), ELA acylated at Lys12was much less effective than non-acylated ELA, or ELA acylated at Lys12or Lys38. The non-acylated ELA was not as effective as the acylated Lys12and Lys38analogs.

[0133] Figs 22C to F compare cumulative food intake, at 12 h, 21 h, 42 h, and 63 h delay time points, respectively, for saline control, non-acylated ELA, and ELA acylated at Lys12or Lys38. At the 12 h delay (Fig. 22C), 21 h delay (Fig. 22D) and 42 h delay (Fig. 22 E), non-acylated ELA, and ELA acylated at Lys12or Lys38significantly reduce cumulative food intake. The efficacy of ELA analogs on food intake was lost after 63 h. Fig. 23 shows the effects of fusion peptides on glucose concentrations in lean mice. (A) Blood glucose concentrations were measured immediately before and 15, 30, 60 and 105 min after i.p. injection of glucose alone (18 mmol / kg bw) and test peptides (each at 25 nmol / kg body weight) in 4 h fasted mice. (B) shows the glucose AUC values for 0-105 min post injection. Values represent mean ± SEM for 6 mice. *P<0.05, **P<0.01 and ***P<0.001 compared to glucose alone.

[0134] Figs. 23A and B show that exendin-4; non-acylated ELA; and ELA acylated at Lys 12, 27 or 38 each significantly reduced plasma glucose AUC.

[0135] Exendin-4 operates through the GLP-1 receptor. Exendin-4 is an antidiabetic drug that has potent effects upon glucose lowering. Apelin-13 amide is not that impressive on insulin secretion and hence likely glucose lowering on its own. It is possible that it is the exendin-4 part of ELA that is giving the most effect, and less so the apelin element.

[0136] Fig. 24 shows persistent glucose lowering effects of ELA fusion peptides in lean mice. Blood glucose concentrations were measured immediately before and 15, 30, 60 and 105 min following an i.p. glucose load (18 mmol / kg bw) in 21 h fasted mice injected with saline vehicle or test peptides 8 h previously. Blood glucose AUC values for 0-105 min post injection (the columns are, from left to right, glucose alone, apelin-13 amide, exendin-4, ELA, ELA Lys12, ELA Lys38, ELA Lys27). Values represent mean ± SEM for 6 mice. *P<0.05 compared to glucose alone.

[0137] Fig. 24 (A and B) shows that ELA Lys12and ELA Lys38yields persistent glucose lowering effects.

[0138] Fig. 25 (A) shows persistence of glucose lowering effects of ELA fusion peptides in lean mice. Blood glucose concentrations were measured immediately before and 15, 30, 60 and 105 min following i.p. glucose load (18 mmol / kg bw) in 24 h fasted mice injected with saline vehicle or test peptides 12 h previously. (B) Blood glucose AUC values for 0-105 min post injection (the columns are, from left to right, glucose alone, exendin-4, a mixture of apelin-13 amide + exendin-4; ELA Lys12, ELA Lys33). Values represent mean ± SEM for 6 mice. *P<0.05 compared to glucose alone.

[0139] Fig. 25 shows that ELA Lys12and ELA Lys38yields persistent glucose lowering effects 12 h after administration.

[0140] Fig. 26 shows persistent glucose lowering effects of ELA fusion peptides in lean mice. Blood glucose concentrations were measured immediately before and 15, 30, 60 and 105 min following i.p. glucose load (18 mmol / kg bw) in 21 h fasted mice injected with saline vehicle or test peptides 24 h previously. Blood glucose AUC values for 0-105 min post injection (the columns are, from left to right, glucose alone, ELA Lys12, ELA Lys38). Values represent mean ± SEM for 8 mice. *P<0.05 compared to glucose alone.

[0141] Fig. 26 shows that ELA Lys12yields persistent glucose lowering effects 24 h after administration. Fig. 27 shows persistent glucose lowering effects of ELA fusion peptides in lean mice. Blood glucose concentrations were measured immediately before and 15, 30, 60 and 105 min following i.p. glucose load (18 mmol / kg bw) in 4 h fasted mice injected with saline vehicle or test peptides 36 h previously. Blood glucose AUC values for 0-105 min post injection (the columns are, from left to right, saline, ELA Lys12, ELA Lys38). Values represent mean ± SEM for 8 mice.

[0142] Fig. 27 shows that neither ELA Lys12nor ELA Lys38lowers glucose levels when administered 36 hours previously.

[0143] ELA and related analogs - chronic in vivo studies

[0144] Fig. 28 shows chronic study design.

[0145] Treatment Groups:

[0146] Mice aged 6-8 weeks were initiated on this study with the high fat diet (45% fat) feeding for 84 days. Controls were fed on a normal rodent chow (10% fat). Following the initial feeding period, various treatment regimens were undertaken as shown in the diagram with additional elements below.

[0147] Male NIH Swiss mice (Harlan Ltd, Blackthorne, UK) were housed individually in an air-conditioned room (22°C), with relative humidity of 51% and a 12-hour light:dark cycle (8 AM to 8 PM). The mice were maintained on a high-fat diet (45% fat, 20% protein, 35% carbohydrate; percent of total energy 26.15 kJ / g; Dietex International Ltd, Witham, UK) for 84 days to produce a model of diet induced obesity (DIO) diabetes. Normal lean mice received a standard rodent diet (10% fat, 30% protein, 60% carbohydrate;percent of total energy 12.99 kJ / g, Trouw Nutrition, Cheshire, UK).

[0148] Treatment regime - Lean Saline Control - Once Daily (n=8), HFF Saline Control - Once Daily (n=7), Liraglutide - 25 nmol / kg - Once Daily (n=8), ELA - 25 nmol / kg - Once Daily (n=8) and ELA Lys 12 - 25 nmol / kg - Once Daily (n=8).

[0149] For GTT’s:

[0150] A dose of 18 mmol / kg glucose, in 18 h fasted mice was given by oral gavage, timed plasma samples collected, at time points: 0, 15, 30, 60 and 120 min. The Area Under the Curve (AUC) was calculated for this data (0-120 min) and displayed alongside the time course line graphs.

[0151] Pyruvate Tolerance Test:

[0152] To assess metabolic pathways related to hepatic glucose metabolism 2 g / kg sodium pyruvate was given to 18 h fasted mice by i.p. injection route. Monitoring blood glucose only, at various time points: 0, 15, 30, 60, 90, 120 min. The Area Under the Curve (AUC) was calculated for this data (0-120 min) and displayed alongside the time course line graphs. Insulin Sensitivity test:

[0153] Whole body insulin sensitivity was measure by administering 25 Units of insulin, to non-fasted mice, by i.p. injection route, and monitor blood glucose only, Time Points: 0, 15, 30 & 60 min. The Area Above the Curve (AAC) was calculated for this data (0-60 min) and displayed alongside the time course line graphs.

[0154] CLAMS: Metabolic cage study:

[0155] N=5 per group 3 h to acclimatise, 21 h monitoring = 24 h in total 10 am - 10 am following day.

[0156] Energy intake and body composition Comprehensive Laboratory Animal Monitoring System (CLAMS) metabolic chambers (Columbus Instrument, Columbus, Ohio) were used to measure indirect calorimetry and energy expenditure (EE) in the peptide-treated and lean control groups after 21 days of treatments described previously (O’Harte et al. 2016). Respiratory exchange ratio (RER) was calculated by dividing VCO2 by VO2. EE was calculated using the equation EE = (3.815 + 1 ,232XRER)XVO2. (Reference: O'Harte FPM, Ng MT, Lynch AM, Conlon JM, Flatt PR. Dogfish glucagon analogues counter hyperglycaemia and enhance both insulin secretion and action in diet- induced obese diabetic mice. Diabetes Obes Metab. 2016;18:1013-1024.)

[0157] Tissues taken and Snap Frozen: Pancreas (n=5), Liver, intestine - Duodenum, Jejunum & Ileum Adipose Tissue, Hypothalamus. For later use in immunocytochemistry staining of Pancreas.

[0158] Fig. 29 shows the effects of high fat feeding (HFF) diet (45% fat) and normal chow (lean controls) (10% fat) upon (A) non-fasting blood glucose, (B) cumulative food intake, (C) body weight and (D) acute glucose tolerance (after 84 days on respective diets) in mice. The inserted bar chart in Fig. 29D shows, from left to right, lean control, and High Fat Fed (HFF) mice. Values are mean ± SEM for n=8 mice in lean controls and n=40 mice in HFF group. *P<0.05, **P<0.01 and ***P<0.001 compared to lean controls. Fig. 29 shows the development of glucose intolerance and weight gain in the mice used for the subsequent interventions.

[0159] Figs. 29 A, C and D show that non-fasting glucose, body weight and glucose tolerance rise on a HFF diet. Fig. 29B shows that food intake was not very different in the lean and HFF groups but, nevertheless, the HFF calorific content is much higher in HFF fed mice than in lean controls. Thus, the calorie intake was much higher in the HFF mice.

[0160] Fig. 30 shows the effects of 21 -day treatment with Liraglutide, ELA and ELA Lys12, each at 25 nmol / kg body weight, on (A) blood glucose, (B) percentage body weight change, (C) energy intake, (D) water intake and (E) glycated haemoglobin HbAic in HFF mice. Parameters were measured during 21 -days treatment with test peptides (25 nmol / kg bw, i.p., once daily). Values are mean ± SEM for eight mice. *P<0.05, **P<0.01 and ***P<0.001 compared to lean controls.AP<0.05,AAP<0.01 andAAAP<0.001 compared to HFF saline controls.+P<0.05,++P<0.01 and+++P<0.001 compared to Liraglutide.QP<0.05 andQQP<0.01 compared to ELA (non-acylated). In a chronic study with high fat fed (HFF) (45% fat) mice, Figs. 30A and 30B show that ELA(Lys12) was the most effective analog at reducing non-fasting blood glucose and led to a 4% reduction in body weight after 21 days. Fig. 30B also shows that the non-acylated ELA fusion peptide was associated with some reduction in body weight versus saline treated HFF mice after 21 days.

[0161] Fig. 30E shows that the % HbA1C (glycated haemoglobin) was lower for both of ELA fusion peptide, ELA(Lys12) acylated fusion peptide, as well as, for liraglutide. Thus, both ELA fusion peptide and ELA(Lys12) acylated fusion peptide reduced blood glucose over the longer-term. This pattern was reflected the most marked effect on non-fasting blood glucose values by the acylated ELA Lys12 analog in Fig 30A.

[0162] Figs. 30C and 30D show cumulative energy intake in groups of mice over a 28 day intervention period; and water intake over the 28 day intervention period, respectively.

[0163] 12

[0164] Fig. 31 shows the effects of 21 -day treatment with Liraglutide, ELA and ELA Lys on glucose tolerance in HFF mice. Parameters were assessed following 21 -days treatment with test peptides (25 nmol / kg bw, i.p., once daily). Glucose (18 mmol / kg bw) was administered by i.p. injection at t=0 min in 18 h fasted mice. Glucose AUC values for 0-60 min post injection are shown in the righthand bar chart. Values are mean ± SEM for eight mice. *P<0.05 compared to lean controls.AP<0.05 andAAP<0.01 compared to saline treated HFF controls. QP<0.05 and QQQP<0.001 compared to ELA (non-acylated).

[0165] 12

[0166] Fig. 31 shows that chronic 21 day treatment with ELA Lys significantly improves glucose tolerance in HFF mice.

[0167] Fig. 32 shows the effects of 21 -day treatment with Liraglutide, ELA and ELA Lys12on pyruvate tolerance in HFF mice. Parameters were assessed following 21 -days treatment with test peptides (25 nmol / kg bw, i.p., once daily). (A) Pyruvate (2 g / kg body weight sodium pyruvate) was administered by i.p. injection at t=0 min in 18 hour fasted mice. (B) Blood glucose AUC values for 0- 120 min post injection is shown. Values are mean ± SEM for eight mice. *P<0.05 compared to lean controls.AAP<0.01 compared to HFF saline controls.QP<0.05 compared to ELA (non-acylated).

[0168] 12

[0169] Fig. 32 shows that ELA Lys significantly improves pyruvate tolerance in HFF mice. This shows that, following a pyruvate tolerance test, less glucose was produced which gives an indication of positive metabolic changes. This is thought to demonstrate less hepatic glucose output, a reduction in gluconeogenesis and lower glycogen breakdown, all of which is an indication of improved metabolic control. These actions can also be seen when the hyperglycaemic effects of glucagon hormone is suppressed. Glucagon concentrations are often elevated in Type 2 diabetes. Fig. 33 shows the effects of 21 -day treatment with Liraglutide, ELA and ELA Lys12on insulin sensitivity in HFF mice. Parameters were assessed following 21 -days treatment with test peptides (25 nmol / kg body weight, i.p., once daily). (A) Insulin was administered by i.p. injection at t=0 min in non-fasted mice. (B) Blood glucose area above the curve (AAC) values for 0-60 min post injection. The values obtained for AAC relates to insulin sensitivity testing. When insulin is injected, the glucose lowering achieved is an indicator of tissue insulin sensitivity. The larger the AAC, the better the insulin sensitivity of the mice. Insulin resistant mice (which is characteristic of Type 2 diabetes) will have a lower AAC value. Values are mean ± SEM for eight mice.

[0170] Fig. 33 shows that ELA and ELA Lys 12 are actually slightly better for AAC than liraglutide, although this improvement does not reach statistical significance.

[0171] Panels A and B of Fig. 34A shows the effects of 21 -day treatment with Liraglutide, ELA and ELA Lys12on locomotor activity in HFF mice.

[0172] Locomotor activity was assessed following 21-days treatment with test peptides (25 nmol / kg bw, i.p., once daily). Mice were placed in CLAMS metabolic chambers for 21 h. Activity counts in the X- axis was recorded every minute for 21 h. Values represent mean ± S.E.M. for 5 mice. **P<0.01 and ***P<0.001 compared to lean controls.AAP<0.01 compared to HFF saline controls.+P<0.05 compared to Liraglutide.

[0173] Panels A and B of Fig. 34 are measured during the dark phase and light phase, respectively. Panels A and B look at average X ambulatory (counts).

[0174] Fig. 34A shows that locomotor activity of fusion peptides of the invention was lower versus lean control mice, in the dark phase (Panel A). Despite this, ELA enhanced movement in the light phase (Panel B) versus the HFF saline control (P<0.01), as well as versus liraglutide (P<0.05).

[0175] Fig. 34B shows the effects of 21-day treatment with Liraglutide, ELA and ELA Lys12on locomotor activity in HFF mice. This is the average of locomotor activity over both dark and light phases.

[0176] Locomotor activity was assessed following 21-days treatment with test peptides (25 nmol / kg bw, i.p., once daily). Mice were placed in CLAMS metabolic chambers for 21 h. Activity counts in the Xaxis were recorded every minute for 21 h. Values represent average X total counts in both the combined dark and light phases of monitoring. Values are mean ± S.E.M. for 5 mice. *P<0.05 and **P<0.01 compared to lean saline controls.

[0177] Fig. 34B shows that locomotor activity of fusion peptides of the invention was lower versus lean control mice. Fig. 35 shows the effects of 21 -day treatment with Liraglutide, ELA and ELA Lys12on (A) cumulative food intake and (B) energy intake in HFF mice. Cumulative food intake and energy intake was assessed following 21 -days treatment with test peptides (25 nmol / kg bw, i.p., once daily). Mice were placed in CLAMS metabolic chambers for 21 h, with cumulative food intake measured or energy intake calculated. Values represent mean ± S.E.M. for 5 mice. *P<0.05 compared to lean controls.AP<0.05 andAAP<0.01 compared to HFF saline controls.

[0178] Fig. 35 shows that the non-acylated ELA fusion peptide significantly reduced energy intake. ELA fusion peptide was more effective than ELA Lys12at reducing food intake (and ELA was better than HFF saline treated mice from days 16 to 20). This may be because energy intake was better in all cases versus the HFF saline treated group.

[0179] Fig. 36 shows the effects of 21 -day treatment with Liraglutide, ELA and ELA Lys12on average respiratory exchange ratio (RER) and energy expenditure (EE) in HFF mice. Average RER (A) and EE (B) were assessed following 21-days treatment with test peptides (25 nmol / kg body weight, i.p., once daily). Mice were placed in CLAMS metabolic chambers for 21 h and RER and EE measured. Values represent mean ± S.E.M. for 5 mice. **P<0.01 and ***P<0.001 compared to lean controls.AAAP<0.001 compared to HFF saline controls.+P<0.05 and+++P<0.001 compared to Liraglutide.QQP<0.01 compared to ELA (non-acylated).

[0180] Fig. 36A shows that liraglutide, ELA fusion peptide and ELA Lys12fusion peptide significantly increase average respiratory exchange ratios compared to HFF saline controls. The effect seen with ELA Lys12fusion peptide was significantly higher than the effect seen with ELA fusion peptide.

[0181] Fig. 36B shows that liraglutide and ELA Lys12fusion peptides significantly increase energy expenditure while ELA (non-acylated) fusion peptide did not affect energy expenditure when compared with HFF saline control.

[0182] Fig. 37 uses immunohistochemistry to examine both pancreatic insulin and glucagon content. Values are mean ± SEM. Total islet area (A), beta cell area (B) and alpha cell area (C) for treated groups (Liraglutide, ELA and ELA Lys12) were lower than for lean controls. Values are mean ± SEM. *P<0.05, **P<0.01 and ***P<0.001 compared with lean controls.AAP<0.01 and compared with HFF control.

[0183] Fig 37, taken from immunocytochemistry data, shows that both ELA peptide treatments reduced islet cell area as well as beta- and alpha-cell area in treated mice. The acylated ELA Lys12fusion peptide was the most potent at reducing the glucagon producing alpha-cell area (Panel C).

[0184] Fig. 38 shows percentage beta and alpha cell area for each group (lean control, HFF, liraglutide, ELA, ELA Lys12). All treated groups had significantly lower % alpha cells versus the HFF saline treated group (Fig. 38). *P<0.05and ***P<0.001 compared with lean controls.AAAP<0.001 compared with HFF control.

[0185] Fig. 38 shows that all peptide treatments produce a better beta cell to alpha cell ratio in HFF treated mice which should have a beneficial effect for metabolic control in diabetes.

[0186] Fig. 39, taken from immunocytochemistry data, shows beta-cell proliferation for each group (lean control, HFF, liraglutide, ELA, ELA Lys12) following tissue staining for insulin and and a biomarker of proliferation Ki-67. Values are mean ± SEM. ***P<0.001 compared with lean controls.AAAP<0.001 , compared with HFF control and *P<0.05 compared with liraglutide.

[0187] Fig. 39 shows significantly increased beta-cell proliferation when ELA or ELA Lys12fusion peptides are administered compared to Liraglutide.

[0188] Fig. 40, taken from immunocytochemistry data, shows beta-cell apoptosis following tissue staining <p for insulin and TUNEL. Values are mean ± SEM. ***P<0.001 compared with HFF controls. P<0.05, compared with lean control

[0189] Fig. 40, taken from immunocytochemistry data, shows significantly decreased beta-cell apoptosis when ELA, ELA Lys12fusion peptide or Liraglutide is administered versus HFF control mice. ELA showed even less beta cell apoptosis when compared to the healthy lean controls.

[0190] Fig. 41 , taken from immunocytochemistry data, shows alpha cell apoptosis following tissue staining for glucagon and TUNEL. Values are mean ± SEM. ***P<0.001 compared with Lean controls. <PW

[0191] P<0.05 and P<0.01 compared with HFF control.

[0192] Fig. 41 shows significantly increased alpha-cell apoptosis when ELA or ELA Lys 12 fusion peptides are administered, when compared with HFF controls. Liraglutide does not work as well as ELA and ELA Lys12fusion peptides when compared back against HFF saline treated mice (P<0.05 to p<0.01).

[0193] Fig. 42 shows plasma cholesterol (A) and triglyceride (B) concentrations following 21 days treatment with test fusion peptides ELA and ELA Lys12(25 nmol / kg bw, by once daily i.p. injection). Panel C shows the HDL:LDL ratio and Panel D shows the Total Cholesterol:HDL ratio. Values represent mean ± S.E.M. *P<0.05, **P<0.01 and ***P<0.001 compared to HFF saline controls.AP<0.05 compared to lean controls.QP<0.05 andQQP<0.01 compared to Liraglutide.

[0194] Prolonged 21 day treatment with fusion peptides or with liraglutide showed raised HDL cholesterol and a reduction in LDL cholesterol in high fat fed (HFF) diet induced obese mice. Plasma triglycerides were reduced following either ELA or ELA Lys12treatments, both of which were more effective than liraglutide. Furthermore, the ratio of HDL / :LDL as well as the Total Chol:HDL was improved versus the HFF saline treated controls for both ELA and ELA Lys12treatment groups.

[0195] Fig. 43 shows plasma aspartate transaminase (AST) (A) and alanine transaminase (ALT) (B) concentrations following 21 days treatment with test peptides (25 nmol / kg bw, i.p., once daily). Test fusion peptides ELA and ELA Lys12are referred to as Hybrid P1 and Hybrid P12, respectively, in Figure 43. Values represent mean ± S.E.M. *P<0.05, **P<0.01 and ***P<0.001 compared to HFF saline controls.QP<0.05,QQP<0.01 andQQQP<0.001 compared to Liraglutide.

[0196] Prolonged 21 day treatment with GLP-1 R / APJR fusion peptides showed improved liver function and reduced circulating liver enzyme concentrations for AST and ALT versus liraglutide in high fat fed diet induced obese mice.

[0197] Methodology for the following data:

[0198] Treatment Groups: Mice C57BI / 6J mice (n=24) aged 6-8 weeks were initiated on this study with the high fat diet (45% fat) feeding for 18 weeks (126 days). Some controls (n=6) were fed on a normal rodent chow (10% fat). Following the initial feeding period, various treatment regimens were undertaken as shown in the diagram with additional elements below. Male NIH Swiss mice (Envivo, UK) were housed individually in an air-conditioned room (22°C), with relative humidity of 51% and a 12-hour light:dark cycle (8 AM to 8 PM). The mice were maintained on a high-fat diet (45% fat, 20% protein, 35% carbohydrate; percent of total energy 26.15 kJ / g; Research Diets Inc., New Brunswick, New Jersey, USA) for 126 days to produce a model of diet induced obesity (DIO) diabetes. Normal lean mice received a standard rodent diet (10% fat, 30% protein, 60% carbohydrate;percent of total energy 12.99 kJ / g, Research Diets Inc.).

[0199] Treatment regime - Lean Saline Control - Once Daily (n=6), HFF Saline Control - Once Daily (n=6), Liraglutide - 25 nmol / kg - Once Daily (n=6), ELA - 25 nmol / kg - Once Daily (n=6), ELA(Lys12) - 25 nmol / kg - Once Daily (n=6) and ELA(Lys38) - 25 nmol / kg - Once Daily (n=6).

[0200] Glucose Tolerance Test (GTT): A dose of 18 mmol / kg glucose, in 18 h fasted mice was given by oral gavage, timed plasma samples collected, at time points: 0, 15, 30, 60 and 120 min. The Area Under the Curve (AUC) was calculated for this data (0-120 min) and displayed alongside the time course line graphs.

[0201] Insulin Sensitivity Test: Whole body insulin sensitivity was measure by administering 25 Units of insulin, to non-fasted mice, by i.p. injection route, and monitor blood glucose only, Time Points: 0, 15, 30 & 60 min. The Area Above the Curve (AAC) was calculated for this data (0-60 min) and displayed alongside the time course line graphs. CLAMS: Metabolic cage study: n=3 per group 3 h to acclimatise, 21 h monitoring = 24 h in total 10 am - 10 am following day. Energy intake and body composition Comprehensive Laboratory Animal Monitoring System (CLAMS) metabolic chambers (Columbus Instrument, Columbus, Ohio) were used to measure indirect calorimetry and energy expenditure (EE) in the peptide-treated and lean control groups after 21 days of treatments described previously (O’Harte et al. 2016). Respiratory exchange ratio (RER) was calculated by dividing VCO2 by VO2. EE was calculated using the equation EE = (3.815 + 1 ,232xRER)xVO2. (Reference: O'Harte FPM, Ng MT, Lynch AM, Conlon JM, Flatt PR. Dogfish glucagon analogues counter hyperglycaemia and enhance both insulin secretion and action in diet-induced obese diabetic mice. Diabetes Obes Metab. 2016;18:1013— 1024.)

[0202] 21 -day treatment intervention after male mice were given free access to either a High Fat Fed (HFF) diet (45% fat) or normal diet (10% fat, lean control) for 18 weeks:

[0203] Fig. 44 Male mice were given free access to either a High Fat Fed (HFF) diet (45% fat) or normal diet (10% fat, lean control) for 18 weeks prior to commencement of this 21 -day treatment intervention. Cumulative food intake was measured during the 21-day treatment period with test peptides (25 nmol / kg bw, i.p., once daily) or saline controls. Values are mean ± SEM for six mice. *P<0.05 compared to HFF saline treated controls at the same timepoint. At each timepoint, the bars are, from left to right, lean control, HFF saline control, liraglutide, ELA Lys12and ELA Lys38. Fig. 44 shows the effects of 21-day treatment with Liraglutide, ELA Lys12, ELA Lys38, each at 25 nmol / kg body weight, on cumulative food intake. Parameters were measured during 21 -days treatment with test peptides (25 nmol / kg bw, i.p., once daily). Values are mean ± SEM for six mice. *P<0.05 compared to HFF saline controls. These data show that ELA(Lys12) was the most effective analogue at reducing cumulative food intake, as shown on day 18 above (P<0.05). The other acylated analogue ELA(Lys38) had a similar effect upon cumulative food intake but narrowly failed to reach statistical significance.

[0204] Fig 45 Male mice were given free access to either a High Fat Fed (HFF) diet (45% fat) or normal diet (10% fat, lean control) for 18 weeks prior to commencement of this 21-day treatment intervention.

[0205] (A) Body weight was measured during 21 -days treatment with test peptides (25 nmol / kg bw, i.p., once daily). (B) % change in body weight (C) % change in body weight from basal. Values are mean ± SEM for six mice. *P<0.05, **P<0.01 and ***P<0.001 compared to HFF controls.Ap<0.05 andAAp<0.01 compared to Liragutide. In Figure 45C, the bars are, from left to right, lean control, HFF saline control, liraglutide, ELA Lys12and ELA Lys38. In Figure 45B, ELA Lys38shows the greatest reduction in body weight, followed by ELA Lys12. Fig. 45 shows the effects of 21-day once daily i.p. injection either with saline controls or 25 nmol / kg bw Liraglutide, ELA Lys12or ELA Lys38. Body weight was measured at regular intervals during the 21-day treatment period with test peptides. (A) body weight, (B) % change in body weight (C) % change in body weight from basal. Values are mean ± SEM for six mice. *P<0.05, **P<0.01 and ***P<0.001 compared to HFF controls.Ap<0.05 andAAp<0.01 compared to Liraglutide. In this chronic study in high fat fed (HFF) (45% fat) diet induced obese (DIO) mice, these data show that ELA Lys38was the most effective analogue leading to a 5.5% reduction in body weight after 21 days. This was closely followed by ELA Lys12(4.8% reduction). However, the Liraglutide treatment group failed to reach significance over this time period.

[0206] Fig 46 Male mice were given free access to either a High Fat Fed (HFF) diet (45% fat) or normal diet (10% fat, lean control) for 18 weeks prior to commencement of this 21 -day treatment intervention.

[0207] (A) Non-fasted blood glucose was measured during 21 -day treatment with test peptides (25 nmol / kg bw), given by i.p. injection once daily. (B) Blood glucose excursion values based upon AUC vales from the upper panel A. In Figure 46B, the bars are, from left to right, lean control, HFF saline control, liraglutide, ELA Lys12and ELA Lys38. Fig. 46 shows the effects of high fat feeding (HFF) (45% fat) diet and normal chow (lean controls) (10% fat) diet upon (A) non-fasting blood glucose, (B) glucose excursion (after 18 weeks on respective diets) in male mice. Values are mean ± SEM for n=6 mice in lean controls and n=6 mice in HFF groups. *P<0.05 and ***P<0.001 compared to saline treated HFF controls. The present data shows that ELA Lys12was the most effective analogue in lowering blood glucose versus HFF saline-treated controls, over this 21 day intervention period (P<0.001). The anti-hyperglycaemic effect of Liraglutide, an established incretin mimetic treatment, failed to reach statistical significance.

[0208] Fig 47 Glucose tolerance was assessed following 21 -day treatment with test peptides (25 nmol / kg bw, i.p., once daily). (A) Glucose (18 mmol / kg bw) was administered by i.p. injection at t=O min in 18 h fasted mice. (B) Blood glucose AUC values for 0-120 min post injection. Values are mean ± SEM for six mice. ***P<0.001 compared to HFF controls given glucose + saline. In Figure 47B, the bars are, from left to right, lean control, HFF saline control, liraglutide, ELA Lys12and ELA Lys38. Fig. 47 shows the effects of 21 -day treatment with Liraglutide, ELA Lys12or ELA Lys38on glucose tolerance in HFF mice. Parameters were assessed following 21 -days of treatment with test peptides (25 nmol / kg bw, i.p., once daily). (A) Glucose (18 mmol / kg bw) was administered by i.p. injection at t=0 min in 18 h fasted mice. (B) Glucose AUC values for 0-120 min post injection are shown in the lower panel bar chart. Values are mean ± SEM for six mice. ***P<0.001 compared to saline treated HFF controls. Both ELA Lys12and ELA Lys38showed a marked improvement in glucose tolerance (42-44% reduction versus HFF control), but the established incretin mimetic, Liraglutide, failed to significantly improve glucose tolerance.

[0209] Fig 48 Insulin sensitivity was assessed following 21 -day treatment with test peptides (25 nmol / kg bw, i.p., once daily). (A) Insulin was administered by i.p. injection (25 Units) at t=O min in non-fasted mice. (B) Blood glucose AAC values for 0-60 min post injection. In Figure 48B, the bars are, from left to right, lean control, HFF saline control, liraglutide, ELA Lys12and ELA Lys38.

[0210] Values are mean ± SEM for six mice. *P<0.05 compared to HFF control. Fig. 48 shows the effects of 21 -day treatment with Liraglutide, ELA Lys12or ELA Lys38on insulin sensitivity in HFF mice. Parameters were assessed following 21-day treatment with test peptides (25 nmol / kg body weight, i.p., once daily). (A) Insulin was administered by i.p. injection (25 Units) at t=O min in non-fasted mice. (B) Blood glucose area above the curve (AAC) values for 0-120 min post injection. The values obtained for AAC are directly related to insulin sensitivity. When insulin is injected, the glucose lowering achieved is an indicator of tissue insulin sensitivity. The higher the AAC, the better the insulin sensitivity of the mice. Values are mean ± SEM for six mice. The saline treated mice on the normal 10% fat diet, were more insulin sensitive (P<0.05) than the HFF (45% fat) controls. Each of the peptide treatments improved insulin sensitivity versus the HFF saline-treated control group but none reached significance versus the HFF controls.

[0211] Fig. 49 Average respiratory exchange ratio (A) and energy expenditure (B) were assessed following 21 -day treatment with test peptides (25 nmol / kg bw, i.p., once daily). Mice were placed in CLAMS metabolic chambers for 21 h and RER and EE measured. In Figures 49A and 49B, the bars are, from left to right, lean control, HFF saline control, liraglutide, ELA Lys12and ELA Lys38. Values represent mean ± SEM for three mice. ***P<0.001 compared to HFF controls. Fig. 49 shows the effects of 21 -day treatment with Liraglutide, ELA Lys12or ELA Lys38on average respiratory exchange ratio (RER) and energy expenditure (EE) in HFF mice. Average RER (A) and EE (B) were assessed following 21 -day treatment with test peptides (25 nmol / kg body weight, i.p., once daily). Mice were placed in CLAMS metabolic chambers for 21 h and RER and EE measured. Values represent mean ± SEM for 3 mice. ***P<0.001 compared to HFF saline controls. This shows that Liraglutide, and coagonist fusion peptides ELA Lys12and ELA Lys38significantly change average respiratory exchange ratios (P<0.001) compared to HFF saline-treated controls. The Liraglutide treated group showed reduced energy expenditure (P<0.001), while both ELA acylated co-agonists did not affect energy expenditure when compared with HFF saline-treated controls.

[0212] Fig. 50 Locomotor activity was assessed following 21 -days treatment with test peptides (25 nmol / kg bw, i.p., once daily). Mice were placed in CLAMS metabolic chambers for 21 h. Activity counts in the X- and Z-axes were recorded every minute for 21 h. In Figure 7, the bars are, from left to right, lean control, HFF saline control, liraglutide, ELA Lys12and ELA Lys38. Values represent mean ± S.E.M. for three mice. **P<0.001 compared to saline controls,Qp<0.05 compared to ELA Lys38. Fig. 50 shows that both Liraglutide and ELA Lys12display less movement along the X-direction in the dark phase (P<0.01) versus saline treated lean control mice. In the light phase, Liraglutide treated mice move less in the Z-direction versus ELA Lys38(P<0.05)

[0213] Fig. 51 Bone mineral density and body composition were assessed following 21 -day treatment with test peptides (25 nmol / kg bw, i.p., once daily). In Figure 51 , the bars are, from left to right, lean control, HFF saline control, liraglutide, ELA Lys12and ELA Lys38. Values are mean ± SEM for five mice. Fig. 51 shows the effects of 21 -day treatment with Liraglutide, ELA Lys12or ELA Lys38on (A) bone mineral density (B) bone mineral content (C) total bone area and (D) % body fat in HFF DIO mice. Mice were anaesthesised and placed in a Dexa scanner (Piximus) and the various parameters were measured. Values represent mean ± SEM for 6 mice. No significant changes in bone parameters of % body fat were observed between any of the test peptides and controls. acute glucose tolerance in fasted mice fed a HFF for 28-30 weeks:

[0214] Fig. 52 Effect of hybrid ELA peptide or component peptides on acute glucose tolerance in fasted mice fed a HFF for 28-30 weeks. In Figure 52B, the bars are from left to right, glucose control, exendin (1-30), apelin-13, exendin (1-30) plus apelin-13 (given separately), and ELA. In a separate study to those provided above (i.e., Figs.44 to Fig. 51) glucose tolerance was examined in longer term (28-30 weeks) high fat fed male mice. Panel A: Blood glucose concentrations were measured immediately before and 15, 30, 60, 90 and 120 min following i.p. glucose load (18 mmol / kg bw) in 18 h fasted HFF C57BL / 6 mice injected alongside a saline vehicle (control) or peptide treatment at the lower dose of 10 nmol / kg. The hybrid co-agonist ELA peptide or its component peptides (apelin-13 or exendin-4(1-30) alone or in combination) were tested here. ELA showed improved lowering of plasma glucose from 30-120 min inclusive, versus the glucose control group. Panel B: Blood glucose AUC values for 0-120 min post injection from panel above. Values represent mean ± SEM for 6 mice. *P<0.05, **P<0.01 compared to glucose alone. Exendin-4(1-30) (P<0.05) and ELA (P<0.01) showed the best improvements in glucose tolerance versus the glucose control. The hybrid co-agonist peptide ELA (navy bar; right-most bar) was more effective than when mice were tested with an equimolar combined mixtures of each component peptides (lilac bar; second to right most bar).

[0215] Fig. 53 Effect of hybrid ELA co-agonist peptide on glucose tolerance in the presence of a GLP-1 or APJ receptor antagonist or the combination of both antagonists. In a separate study to those provided above (i.e., Fig. 44 to Fig. 51), glucose tolerance was examined in longer term (28-30 weeks) high fat fed male mice. Panel A: Blood glucose concentrations were measured immediately before and 15, 30, 60, 90 and 120 min following i.p. glucose load (18 mmol / kg bw) in 18 h fasted HFF C57BL / 6 mice given an i.p. injection (saline vehicle control) or peptide treatment as indicated (10 nmol / kg) in the presence and absence of a GLP-1 , APJ or both receptor antagonists (100 nmol / kg). Panel B: Blood glucose AUC values for 0-120 min post injection from above panel. Values represent mean ± SEM for 6 mice. *P<0.05, **P<0.01 and ***P<0.001 compared to glucose control given a saline injection. These results show that the greatest glucose lowering activity is achieved by the hybrid co-agonist ELA peptide (P<0.001 ; second from left bar). When a GLP-1 antagonist is applied, there was partial reduction of the ELA glucose lowering ability (green bar; middle bar) and similarly partial loss with an APJ receptor antagonist (grey bar; second from right most bar). When both component antagonists were separately applied, the glucose lowering ability of ELA was completely lost (pink bar; right most bar). These findings demonstrate that the GLP-1 and APJ receptors are wholly responsible for the glucose lowering actions observed here.

[0216] Summary of the main pre-clinical experimental findings

[0217] The ELA fusion peptide has significantly more insulinotropic activity than either component peptide or when each component is combined separately (Fig.3). The ALE fusion peptide shows dose-dependent insulin secretion but is not as effective as ELA (Fig. 4).

[0218] ELA and ALE both operate through the APJ receptor and GLP-1 R. Co-administration of both antagonists can obliterate the effect on insulin secretion (Fig.5 & Fig. 6).

[0219] ELA and exendin-4(1-39) significantly reduced food intake in trained mice with prolonged actions in vivo up to at least 8 h (Fig. 8). Actions of ELA were more potent and long lasting than when Ex-4(1- 30) and apelin-13 amide were both co-administered as a mixture.

[0220] The actions of ELA on food intake in mice were obliterated by the addition of both GLP-1 and APJ receptor antagonists (Fig. 9).

[0221] The reverse fusion peptide - ALE - was ineffective in reducing food intake in mice unlike the very potent actions of it counterpart ELA (Fig. 10).

[0222] The fusion ELA peptide was an effective and long acting agent in improving glucose tolerance following an ipGTT in normal mice (and was better than the actions of a mixture of Exendin-4(1-30) and apelin-13 amide co-administered) and was still effective at lowering glucose 24 h after injection (Fig. 11 & Fig. 12).

[0223] ELA remained very effective in lowering blood glucose following an acute ipGTT in a model of high fat feeding (HFF) in diet induced obese mice (Fig. 13).

[0224] Component peptides Exendin-4(1-30) and apelin-13 amide show a dose-dependent increase in insulin secretion from clonal pancreatic BRIN-BD11 cells with better effects when co-administered (as a mixture) (Fig 14).

[0225] Acylated analogs of ELA retained their insulinotropic activity in BRIN-BD11 cells with the ELA(Lys12) and ELA(Lys38) acylated forms having the most potent activity. Although ELA(Lys27) did show a dose-dependent insulinotropic action, it was less potent than the parent ELA hybrid peptide (Fig. 15).

[0226] The insulin releasing actions of a mixture of Exendin-4(1 -30) and apelin-13 amide were wiped out by using GLP-1 and APJ receptor antagonists (i.e., Exend in-4(9-39) and (Val13)apelin-13). Similarly, the insulinotropic actions of ELA, ELA(Lys12), ELA(Lys27) and ELA(Lys38) were removed by using combined receptor antagonists (Fig.16).

[0227] Pancreatic BRIN-BD11 cell proliferation was enhanced by 10-8and 10-6M component peptides, as well as ELA and two acylated fusion peptides ELA(Lys12), ELA(Lys38) and the higher concentration 10-6M of each of the peptides tested caused a reduction in apoptosis in BRIN-BD11 cells (Fig. 17). Intraperitoneal injected ELA as well as acylated fusion peptide analogs ELA(Lys12), ELA(Lys38) significantly reduced acute food intake (up to 180 min) in trained mice at doses as low as 1 nmol / kg and 2.5 nmol / kg (Fig. 18).

[0228] ELA and acylated analos ELA(Lys12) and ELA(Lys38) significantly reduced food intake over 180 min in trained mice but the analog, ELA(Lys27) was much less effective than ELA or the 2 other acylated analogs (Fig. 19).

[0229] ELA and its component peptides reduced acute food intake in trained mice but only ELA retained this activity when administered 8 h before resumption of food intake (Fig. 20).

[0230] Intraperitoneally injected ELA as well as acylated fusion peptide analogs ELA(Lys12) and ELA(Lys38) significantly reduced acute food intake (up to 180 min) in trained mice at a dose of 2.5 nmol / kg, but their actions were partially removed when combined GLP-1 and APJ receptor antagonists were administered simultaneously (Fig. 21).

[0231] The actions of ELA and the acylated analog ELA(Lys12) were effective in reducing 180 min food intake when applied 42 h prior to resumption of food intake in trained mice. ELA(Lys12) retained some appetite supressing activity up to 63 h prior to resumption of food intake (Fig. 22).

[0232] In an acute ipGTT in healthy mice, component peptides as well as ELA and its 3 acylated analogs, ELA(Lys12), ELA(Lys27) and ELA(Lys38), retained their glucose lowering effects in mice (Fig 23).

[0233] When administered 8 h in advance of an ipGTT in normal mice, ELA(Lys12) and ELA(Lys38) retained their significant glucose lowering abilities (Fig. 24).

[0234] When administered 12 h in advance of an ipGTT in normal mice, ELA(Lys12) and ELA(Lys38) retained their significantly glucose lowering abilities (Fig. 25) but only ELA(Lys12) retained this action when administered 24 h before the glucose load (Fig. 26) and this was lost when administered 36 h in advance of the glucose load (Fig. 27).

[0235] In a chronic study with high fat fed (HFF) (45% fat) mice, non-fasting blood glucose was elevated from day 28 to day 84 (Fig. 29A) and body weight was increased from day 7 to day 42 (Fig. 29C) and an acute ipGTT at 84 days (Fig. 29D) showed substantial impaired glucose tolerance versus lean normal chow fed mice (Fig. 29).

[0236] When mice had been fed on a high fat diet for 84 days, they were given once daily i.p. injections of various peptide treatments (25 nmol / kg) including Liraglutide, ELA and the acylated fusion peptide analog ELA(Lys12). ELA(Lys12) was the most effective analog at reducing non-fasting blood glucose and led to a 4% reduction in body weight after 21 days (Fig. 30B). Liraglutide, ELA and ELA(Lys12) all reduced glycated haemoglobin (HbAic) after 21 days of treatment (Fig. 30E). ELA(Lys12) demonstrated improved glucose tolerance and pyruvate tolerance in HFF mice after 21 days treatment whereas ELA and liraglutide failed to reach statistical significance (Fig. 31 & Fig. 32).

[0237] No significant improvement in insulin sensitivity was recorded following the use of 21 day injection with Liraglutide, ELA or the acylated fusion peptide analog ELA(Lys12) (Fig. 33).

[0238] Treatment with liraglutide, ELA and ELA(Lys12) reduced ambulatory activity of mice in the dark phase versus normal chow fed mice (Fig. 34A & Fig. 34B).

[0239] Administration of ELA for 21 days significantly reduced energy intake in mice from day 16-21 and was more effective than liraglutide in this action (Fig. 35).

[0240] Using Complete Laboratory Animal Monitoring System (CLAMS), it was found that mice given 21 day chronic peptide treatments (25 nmol / kg) including Liraglutide, ELA and the acylated fusion peptide analog ELA(Lys12) showed an elevated average respiratory exchange ratio with elevated energy expenditure observed with liraglutide and ELA(Lys12) (Fig. 36).

[0241] When pancreatic tissues were examined by immunocytochemistry after 21 day peptide treatment in HFF mice, the liraglutide, ELA and ELA(Lys12) treated groups all showed a reduced islet area, beta cell area and alpha cell area versus saline treated HFF control mice (Fig. 37).

[0242] When pancreatic tissues were examined by immunocytochemistry after 21 day peptide treatment in HFF mice, the liraglutide, ELA and ELA(Lys12) treated groups all showed increased beta-cell proliferation with both ELA derived peptides being more effective than liraglutide in their actions (Fig. 39).

[0243] When pancreatic tissues were examined by immunocytochemistry after 21 day peptide treatment in HFF mice, the liraglutide, ELA and ELA(Lys12) treated groups all showed reduced beta-cell apoptosis with ELA being more effective than the lean control mice (Fig. 40).

[0244] When pancreatic tissues were examined by immunocytochemistry after 21 day peptide treatment in HFF mice, the liraglutide, ELA and ELA(Lys12) treated groups all showed increased alpha-cell apoptosis versus lean controls with both ELA derived peptides showing higher rates of alpha-cell apoptosis than the HFF controls (Fig. 41).

[0245] Both liver enzymes (Fig. 43) and circulating lipid profiles (Fig. 42) were improved by fusion peptides versus liraglutide treatment of HFF diet induced obese mice after chronic 21 day treatments. Overall conclusions

[0246] In a chronic intervention head-to-head study, the antidiabetic and anti-obesity actions of the fusion

[0247] ELA peptide, and its acylated ELA analogs, is enhanced compared to the current standard of care drug of choice Liraglutide.

[0248] The combination of an APJ and GLP-1 receptor co-agonist is more powerful than separate agonists and could have marked potential for diabetes and obesity therapy in the future. The invention is not limited to the embodiments described herein but can be amended or modified without departing from the scope of the present invention.

Claims

Claims:

1. A fusion peptide comprising a GLP-1 receptor agonist (GLP-1 RA) and an APJ receptor agonist; wherein the GLP-1 RA is covalently linked to the APJ receptor agonist by a linker; and wherein the fusion peptide retains at least some activity of at least one of the GLP-1 RA and the APJ receptor agonist.

2. A fusion protein according to Claim 1 , wherein the linker, before linkage, comprises at least one amino carboxylic acid.

3. The fusion peptide according to Claim 2, wherein the amino carboxylic acid is selected from amino butanoic acid, amino pentanoic acid, amino hexanoic acid, amino heptanoic acid, amino octanoic acid, amino nonanoic acid, amino decanoic acid, amino undecanoic acid and amino dodecanoic acid.

4. A fusion protein according to Claim 2, wherein the linker, before linkage, comprises two amino carboxylic acids in series.

5. The fusion peptide according to Claim 4, wherein the two amino carboxylic acids, which can be the same or different, are independently selected from amino butanoic acid, amino pentanoic acid, amino hexanoic acid, amino heptanoic acid, amino octanoic acid, amino nonanoic acid, amino decanoic acid, amino undecanoic acid and amino dodecanoic acid.

6. The fusion peptide according to any one of Claims 2 to 5, wherein the amino group of one amino carboxylic acid is linked to the GLP-1 receptor agonist and the carboxylic group of the same, or another, amino carboxylic acid is linked to the APJ receptor agonist; wherein, optionally, the, or both, links are an amide link.

7. The fusion peptide according to any one of Claims 2 to 5, wherein the amino group of one amino carboxylic acid is linked to the APJ receptor agonist and the carboxylic group of the same, or another, amino carboxylic acid is linked to the GLP-1 receptor agonist; wherein, optionally, the, or both, links are an amide link.

8. The fusion peptide according to any one of Claims 2 to 7, wherein the, or each, amino carboxylic acid further comprises a (PEG)nchain, wherein n is 1 to 3; optionally wherein n is 2.

9. The fusion peptide of any one of Claims 2, 3, 6 or 7 when dependent on Claim 8, wherein the amino carboxylic acid is 8-amino-3,6-dioxaoctanoic acid.

10. The fusion peptide of any one of Claims 4 to 7 when dependent on Claim 8, wherein each amino carboxylic acid is 8-amino-3,6-dioxaoctanoic acid.11 . The fusion peptide of any one of Claims 1 to 10, wherein the GLP-1 receptor agonist (GLP-1 RA) is “short-acting” or “prandial” and wherein the GLP-1 receptor agonist is selected from exendin- 3, exendin-4 (1-30), exendin-4 (1-39), exendin-4 (3-30), exendin-4 (3-39), lixisenatide, truncations, analogs or derivatives thereof and pharmaceutically acceptable amides and other salts thereof.

12. The fusion peptide of any one of Claims 1 to 10, wherein the GLP-1 receptor agonist (GLP-1 RA) is “long-acting” and wherein the GLP-1 receptor agonist is selected from liraglutide, semaglutide, truncations, analogs or derivatives thereof and pharmaceutically acceptable amides and other salts thereof.

13. The fusion peptide of Claim 11 , wherein the GLP-1 receptor agonist is exendin-4 (1-30) and the exendin-4 (1-30) further comprises a modification at the Lys at position 12 or 27 of exendin-4(1- 30) such as a fatty acid addition at an epsilon amino group and I or the exendin-4 (1-30) comprises a substitution to Lys at another position and the modification by fatty acid addition at an epsilon amino group of the substituted Lys.

14. The fusion peptide of Claim 11 or 12, wherein the GLP-1 receptor agonist is modified by a fatty acid addition at an epsilon amino group.

15. The fusion peptide of Claim 13 or 14, wherein the fatty acid group is selected from a C12 to C20 fatty acid or diacid; optionally a C12 to C18 fatty acid or diacid; further optionally a C16 fatty acid (palmitate), linked to residue 8 by a bivalent linker, optionally, a y-glutamyl linker.

16. The fusion peptide of any one of Claims 13 to 15, wherein the added fatty acid further comprises a second carboxylic acid and is a fatty diacid and wherein, optionally, the added fatty acid is palmitic or stearic diacid.

17. The fusion peptide of any one of Claims 1 to 16, wherein the APJ receptor agonist is “shortacting” or “prandial”; and wherein the APJ receptor agonist is selected from apelin-13, apelin-13 amide, analogs or derivatives thereof and pharmaceutically acceptable salts thereof.

18. The fusion peptide of any one of Claims 1 to 16, wherein the APJ receptor agonist is “long- acting”; and wherein the APJ receptor agonist further comprises a modification at a Lys such as a fatty acid addition at an epsilon amino group and I or comprises a substitution to Lys at another position and the modification by fatty acid addition at an epsilon amino group of the substituted Lys.

19. The fusion peptide of Claim 17 or 18, wherein the APJ receptor agonist analog comprises at least residues 2 - 13 of apelin-13 and further comprises a substitution and I or modification at residue 13 of apelin-13.

20. The fusion peptide of Claim 19, wherein the substitution at residue 13 of apelin-13 may be selected from:Tyr, Thi (p[2-thienyl]-alanine), 4-azido-Phe, 4-cyano-Phe, or Trp; orVai or Ala; and I or wherein residue 13 is the C-terminal residue and the modification at residue 13, if present, is the substitution of the terminal carboxyl group by an amide group.21 . The fusion peptide of any one of Claims 17 to 20, wherein the APJ receptor agonist analog is apelin-13 and further comprises a modification at position 8 of apelin-13 such as a fatty acid addition at an epsilon amino group of position 8 and I or the APJ receptor agonist analog is apelin-13 and comprises a substitution to Lys at any one of position 1 to 7 or 9 to 13 of apelin-13 and the modification by fatty acid addition at an epsilon amino group of the substituted Lys.

22. The fusion peptide of Claim 18 or 21 , wherein the fatty acid group is selected from a C12 to C20 fatty acid or diacid; optionally a C12 to C18 fatty acid or diacid; further optionally a C16 fatty acid (palmitate), linked to residue 8 by a bivalent linker, optionally, a y-glutamyl linker.

23. The fusion peptide of any one of Claims 18, 21 or 22, wherein the added fatty acid further comprises a second carboxylic acid and wherein, optionally, the added fatty acid is palmitic or stearic diacid .

24. The fusion peptide of any one of Claims 1 to 23, wherein the fusion peptide is selected from exendin-4 - link - apelin-13; or a C-terminal amide thereof and wherein, optionally, the link comprises, before linkage, 8-amino-3,6-dioxaoctanoic acid or a dimer thereof.

25. The fusion peptide of any one of Claims 1 to 24, wherein the fusion peptide is exendin-4 - link - apelin-13 that is optionally acylated at one or more of residues 12, 27 and I or 38 of the fusion peptide, or an amide thereof; wherein, optionally, the fusion peptide is an acylated derivative of exendin-4 - link - apelin-13 that is acylated at residue 12 or 38, optionally position 12, of the fusion peptide, or an amide thereof.

26. The fusion peptide of any one of Claims 1 to 25, wherein the GLP-1 receptor agonist (GLP-1 RA) and the APJ receptor agonist simultaneously activate the APJ and GLP-1 receptors, optionally for treating diabetes and obesity.

27. A pharmaceutical composition comprising the fusion peptide of any one of Claims 1 to 26, and a pharmaceutically acceptable excipient; and is, optionally, formulated for parenteral administration, such as subcutaneous injection.

28. A method of treating diabetes and obesity by activating the APJ and GLP-1 Receptors, the method comprising administering the fusion peptide of any one of Claims 1 to 26.

29. A method of nonalcoholic steatohepatitis (NASH), non-alcoholic fatty liver disease NAFLD, and I or cardiovascular disease by activating one or both of the APJ and GLP-1 receptors, the method comprising administering the fusion peptide of any one of Claims 1 to 26.

Images