NOVEL COMPOUNDS CAPABLE OF ANTAGONIZING ISLET AMYLOID POLYPEPTIDE (IAPP) INDUCED BETA-CELL DAMAGE AND IMPAIRED GLUCOSE TOLERANCE

Abstract

Described are molecules specifically binding to human islet amyloid polypeptide (hIAPP) also known as amylin, particularly human-derived antibodies as well as fragments, derivatives and variants thereof for antagonizing islet amyloid polypeptide (IAPP) induced β-cell damage and impaired glucose tolerance which are symptoms typically associated with diabetes mellitus type 2 (T2D).

Claims

1-19. (canceled)

20. An antibody or a human islet amyloid polypeptide (hIAPP)-binding fragment thereof, wherein the antibody or the hIAPP-binding fragment thereof comprises an hIAPP-binding domain comprising: (i) a heavy chain variable (VH) region comprising three complementarity determining regions (CDRs) of the heavy chain variable (VH) sequence of SEQ ID NO:2 (as depicted in Fig. 1); and (ii) a light chain variable (VL) region comprising the amino acid sequence of SEQ ID NO:6.

21. The antibody or hIAPP-binding fragment thereof of claim 20, wherein the VH region comprises the amino acid sequence of SEQ ID NO:2.

22. The antibody or hIAPP-binding fragment thereof of claim 20, wherein the antibody or hIAPP-binding fragment thereof comprises a polypeptide sequence which is heterologous to the hIAPP-binding domain.

23. The antibody or hIAPP-binding fragment thereof of claim 22, wherein the polypeptide sequence comprises a constant domain of an antibody or part thereof which is heterologous to the hIAPP-binding domain.

24. The antibody or hIAPP-binding fragment thereof of claim 23, wherein the constant domain is of the IgG type.

25. The antibody or hIAPP-binding fragment thereof of claim 24, wherein the constant domain is of the IgG1 class or isotype.

26. The antibody or hIAPP-binding fragment thereof of claim 23, wherein the constant domain is a human constant domain.

27. The antibody or hIAPP-binding fragment thereof of claim 20, wherein the antibody or hIAPP-binding fragment thereof is (a) detectably labeled or (b) attached to a drug.

28. The antibody or hIAPP-binding fragment thereof of claim 27, which is detectably labeled, and wherein the detectable label is selected from the group consisting of an enzyme, a radioisotope, a fluorophore, and a heavy metal.

29. The antibody or hIAPP-binding fragment thereof of claim 20, wherein the antibody or hIAPP-binding fragment thereof selectively binds aggregated hIAPP in pancreas tissue of a human diabetic subject.

30. The antibody or hIAPP-binding fragment thereof of claim 20, wherein the antibody is a human-derived monoclonal antibody.

31. A pharmaceutical composition comprising the antibody or hIAPP-binding fragment thereof of claim 20.

32. The pharmaceutical composition of claim 31, wherein the composition further comprises an agent capable of preventing or reducing IAPP amyloid formation.

33. The pharmaceutical composition of claim 32, wherein the agent is selected from the group consisting of a flavonoid, an IAPP analog, metformin, and a thiazolidinedione.

34. The pharmaceutical composition of claim 33, wherein the thiazolidinedione is rosiglitazone.

35. The pharmaceutical composition of claim 31, further comprising a pharmaceutically acceptable carrier or diluent.

36. A method of protecting β-cells from hIAPP induced cell damage and/or islet amyloid toxic effects, the method comprising administering a therapeutically effective amount of an antibody or hIAPP-binding fragment thereof of claim 20 to a subject in need thereof.

37. One or more polynucleotide(s) encoding the antibody or hIAPP-binding fragment thereof of claim 20.

38. A cDNA encoding the antibody or hIAPP-binding fragment thereof of claim 20.

Description

DESCRIPTION OF THE DRAWINGS

[0056] FIG. 1: Amino acid sequences of the variable region, i.e. heavy chain variable region (SEQ ID NO:2) and kappa light chain variable region (SEQ ID NO:4) of human IAPP antibody NI-203.26C11 and kappa light chain variable region (SEQ ID NO:6) of human IARP antibody NI-203.26C11-PIMC. Framework (FR) and complementarity determining regions (CDRs) are indicated with the CDRs being underlined. The heavy chain joining region (JH) and light chain joining region (JK) are indicated as well. Due to the cloning strategy the amino acid sequence at the N-terminus of the heavy chain and light chain may potentially contain primer-induced alterations in FR1, which however do not substantially affect the biological activity of the antibody. In order to provide a consensus human antibody, the nucleotide and amino acid sequences of the original clone were aligned with and tuned in accordance with the pertinent human germ line variable region sequences in the database; see, e.g., Vbase hosted by the MRC Centre for Protein Engineering (Cambridge, UK). The amino acid sequence of human antibodies is indicated when N-terminus amino acids are considered to potentially deviate from the consensus germ line sequence due to the PCR primer and thus have been replaced by primer-induced mutation correction (PIMC).

[0057] FIG. 2: NI-203.26C11 antibody selectively and dose-dependently recognizes pathological hIAPP aggregates in the pancreas of patients diagnosed with diabetes mellitus type 2 (T2D). Thioflavin S (ThioS, left panels, green) staining of islet amyloid in pancreatic islets of T2D patients but not in healthy controls. NI-203.26C11 antibody shows a staining in T2D pancreatic islets loaded with islet amyloids (Subjects 1 and 2) but not in healthy control pancreatic islets lacking islet amyloids (Subjects 3 and 4). Detection of hIAPP aggregates on amyloid positive T2D pancreatic islets with recombinant mouse chimeric antibody NI-203.26C11 (NI-203.26C11-ch; brown) at 3, 10, 30 nM (strong staining), and 1 nM (weak staining). NI-203.26C11 antibody does not recognize physiological hIAPP on healthy control pancreases. Anti-IAPP antibody (1:100; anti-IAPP) and secondary donkey anti-mouse antibody only (anti-mouse secondary) were used as positive and negative controls, respectively. Counterstaining was performed to visualize cell nuclei (faint blue i.o., faint staining here). Scale bar: 100 .Math.m.

[0058] FIG. 3: NI-203.26C11 antibody selectively recognizes pathological hIAPP aggregates in T2D patients. NI-203.26C11 antibody shows a staining in T2D pancreatic islets but does not recognize physiological IAPP on healthy control pancreatic islets lacking pathological hIAPP aggregates. Detection of hIAPP aggregates on T2D pancreatic islets with recombinant mouse chimeric antibody NI-203.26C11 (NI-203.26C11-ch; blue; 100 nM; top panels). NI-203.26C11 staining of hIAPP aggregates is restricted to islet areas deprived of β-cells (no merge with insulin staining in red). Detection of insulin on islet β-cells with anti-insulin antibody (1:3; red). Anti-IAPP antibody (1:100; anti-IAPP; bottom panels) recognizes physiological hIAPP on T2D and healthy control pancreatic islet β-cells, as shown by co-localization with insulin staining (merge with insulin staining in red). Scale bars: 50 .Math.m.

[0059] FIGS. 4A-4B: Impaired glucose tolerance precedes the deposition of islet amyloids in a transgenic mouse model of type-2 diabetes (hIAPP transgenic mice). (FIG. 4A) Islet amyloids are visualized at 14 weeks of age but not 4 weeks of age in hIAPP transgenic mice. Wild-type mice are deprived of islet amyloids. Representative images of hIAPP transgenic and wild-type mouse pancreatic islets stained for islet amyloids (0.1% ThioS), insulin (anti-insulin antibody; 1:3; red) and glucagon (anti-glucagon antibody; 1:2500; blue). (FIG. 4B) Impaired glucose tolerance in 4 week- and 14 week-old transgenic mice. Blood glucose levels during an oral glucose tolerance test (oGTT) performed in hIAPP transgenic and wild-type mice at 4 weeks (top panel) and 14 weeks (bottom panel) of age. Blood glucose levels after oral glucose were higher in hIAPP transgenic mice compared with wild-type mice at 4 weeks (tg, n=13; wt, n=9) and 14 weeks of age (tg, n=27; wt, n=31). Fasting blood glucose levels were similar among groups at 4 weeks but increased in hIAPP transgenic mice at 14 weeks of age. ** p < 0.01 and *** p < 0.001 compared with wt group. Scale bar: 50 .Math.m.

[0060] FIG. 5: NI-203.26C11 antibody selectively binds hIAPP aggregates in hIAPP transgenic mice. NI-203.26C11 shows a staining on pancreatic islets of 4 week- and 16 week-old hIAPP transgenic mice (blue staining; left panels), with no staining on age-matched wild-type mice (absence of blue staining; right panels). Detection of hIAPP aggregates on transgenic islets with recombinant human antibody NI-203.26C11 (NI-203.26C11; blue; 100 nM). Detection of insulin on islet β-cells with anti-insulin antibody (1:3; red). Scale bar: 50 .Math.m.

[0061] FIG. 6: NI-203.26C11 antibody targets aggregated hIAPP after a single administration in hIAPP transgenic mice. Recombinant human NI-203.26C11 or isotype control (control IgG) antibodies were administered to 16 week-old hIAPP transgenic and wild-type mice at 10 mg/kg (i.p.) and antibody binding was evaluated 2 days after administration using an anti-human secondary antibody. Recombinant human NI-203.26C11 (brown staining here) targets transgenic islets showing ThioS-positive amyloids (green staining here) but not ThioS-negative wild-type islets. No staining was observed with the control IgG. Counterstaining was performed to visualize cell nuclei (faint blue i.o., faint staining here).

[0062] FIGS. 7A-7B: NI-203.26C11 antibody protects against β-cell loss in hIAPP transgenic mice. (FIG. 7A) Representative image of a mouse pancreatic islet stained with anti-insulin antibody to visualize β-cells (red i.o., strong staining here). (FIG. 7B) Once-weekly treatment with recombinant mouse chimeric NI-203.26C11 antibody in hIAPP transgenic mice (tg NI-203.26C11-ch, n=23; 10 mg/kg i.p. for 12 weeks) increases pancreatic insulin (insulin-positive area in % pancreas area and % islet area; top left and top middle panels), islet area (mean islet area; top right panel) and insulin secretion (plasma insulin levels; bottom left panel) compared with hIAPP transgenic mice receiving PBS (tg PBS, n=28). Islet density and pancreatic mass were unchanged after NI-203.26C11-ch treatment (bottom middle and bottom right panel, respectively). * p < 0.05 and ** p < 0.01 compared with tg PBS group. Scale bar: 50 .Math.m.

[0063] FIGS. 8A-8C: NI-203.26C11 antibody decreases fasting blood glucose, improves glucose tolerance and normalizes body weight gain in hIAPP transgenic mice. Recombinant mouse chimeric NI-203.26C11 antibody was administered once-weekly to hIAPP transgenic mice (tg NI-203.26C11-ch, n=24; 10 mg/kg i.p.). PBS was used as vehicle (tg PBS, n=27). (FIG. 8A) NI-203.26C11 antibody significantly decreases fasting blood glucose after 8 and 12 weeks of treatment in hIAPP transgenic mice compared with PBS group. Blood glucose levels were measured after overnight fasting. (FIG. 8B) NI-203.26C11 antibody significantly improves glucose tolerance in hIAPP transgenic mice compared with PBS group. Blood glucose levels during an oral glucose tolerance test (oGTT) performed in hIAPP transgenic mice after 10 weeks of treatment. (FIG. 8C) Incremental body weight (%) was normalized in NI-203.26C11-treated hIAPP transgenic mice over 12 weeks of treatment, as compared with wild-type mice injected with PBS (wt PBS, n=31). hIAPP transgenic mice under PBS treatment showed impaired body weight gain compared with wild-type mice. tg NI-203.26C11-ch compared with tg PBS: * p < 0.05 and ** p < 0.01; tg PBS compared with wt PBS: # p < 0.05 and ### p < 0.001; tg NI-203.26C11-ch compared with wt PBS: § p < 0.05.

[0064] FIGS. 9A-9C: NI-203.26C11 recognizes predominantly early fibrillar hIAPP aggregate species. (FIG. 9A) Thioflavin-T (Thio-T) aggregation assay with a classical sigmoidal aggregation curve for hIAPP (■) but no aggregation for rodent IAPP (□; rIAPP). (FIG. 9B) Dotblot analysis of samples (1:3 dilution serie) taken from the Thio-T experiment at the time points indicated revealed unselective binding of anti-IAPP antibody to all hIAPP samples collected as well as to rIAPP taken after 20 min. In contrast, NI-203.26C11 displayed a selective binding for hIAPP preparations taken at the growth phase of hIAPP aggregation (5′), with decreased binding to hIAPP samples taken at later time points (10′ and 20′). Importantly, NI-203.26C11 remained immuno-negative for non-aggregated hIAPP fractions (0′) and did not react to rIAPP. In filter retardation assay (FRA), the anti-IAPP antibody recognized SDS-resistant fibrillar hIAPP species at all time-points assessed, whereas no reactivity could be seen for NI-203.26C11 against these hIAPP species. (FIG. 9C) Transmission electron microscopy (TEM) analysis of samples taken from the Thio-T aggregation assay revealed a diverse morphological spectrum. No large fibrillar aggregates could be detected for hIAPP taken before aggregation (0′) neither for rIAPP taken at late stage (20′). Intermediate hIAPP species showed both, amorphous non-fibrillar as well as fibrillar characteristics (5′) while samples taken at the border to (10′) or within the equilibrium phase (20′) predominantly showed a fibrillar morphology.

[0065] FIGS. 10A-10B: NI-203.26C11 antibody normalizes glucose tolerance in a transgenic rat model of type-2 diabetes (hIAPP transgenic rats). Recombinant rat chimeric NI-203.26C11 antibody was administered once-weekly to hIAPP transgenic rats (tg NI-203.26C11-r, n=9; 3 mg/kg i.p.) and wild-type rats (wtNI-203.26C11-r, n=4; 3 mg/kg i.p.). PBS was used as vehicle in hIAPP transgenic rats (tg PBS, n=10) and wild-type rats (wt PBS, n=5). (FIG. 10A) Oral glucose tolerance test (oGTT) before treatment in 3 month-old hIAPP transgenic rats and wild-type rats showing equivalent blood glucose levels. (FIG. 10B) NI-203.26C11-r antibody normalizes blood glucose levels in hIAPP transgenic rats during an oGTT performed after 8 weeks of treatment, as compared with tg PBS and wt PBS rats. NI-203.26C11-r antibody does not affect blood glucose levels in wild-type rats (wt NI-203.26C11-r). tg NI-203.26C11-r compared with tg PBS: ** p < 0.01.

EXAMPLES

Methods

Antibody

[0066] Human-derived antibodies targeting aggregated species of human IAPP (hIAPP) were identified by high-throughput analyses of full complements of the human memory B-cell repertoire derived from clinically selected populations of aged human subjects. Antibody cDNAs derived from hIAPP-reactive memory B-cells were expressed for the determination of binding properties. To avoid neutralizing mouse and rat anti-human antibody responses directed against human IgG, chimeric antibodies consisting of human variable domains and mouse or rat constant regions were generated by protein engineering. hIAPP-reactive IgG clones were recombinantly produced in CHO for in vitro characterization and in vivo validation studies in transgenic mice and rats. Protein expression was scaled up to 20 liter wave reactors to allow for production of antibodies at 100 mg scale. Antibodies were purified endotoxin-free by affinity chromatography.

Mice

[0067] Mice were obtained from The Jackson Laboratory (Bar Harbor, ME). Hemizygous transgenic male mice (F1) with islet β-cell expression of hIAPP driven by the rat insulin II promoter and wild-type male mice (F1) on a FVB/NxDBA/2J background were generated by breeding hIAPP transgenic FVB/N (FVB/N-Tg(Ins2-IAPP)RHFSoel/J) males with DBA/2J wild-type female mice. Transgenic status was determined by PCR of genomic DNA using oligonucleotide primers directed against the hIAPP transgene.

[0068] The mice showed an early onset of the metabolic phenotype, i.e. the mice spontaneously developed diabetes characterized by impaired glucose tolerance and hyperglycemia already present at 1-month and 2-month of age, respectively. Mouse models previously described did not spontaneously develop diabetes and spontaneously develop diabetes with hyperglycemia and impaired glucose tolerance by 6 to 10 months of age, see e.g. Couce et al. Diabetes 45 (1996),1094-1101); Soeller et al. Diabetes 47 (1998), 743-750; Hull et al. Diabetes 52(2) (2003), 372-379; Hull et al. Am J Physiol Endocrinol Metab 289 (2005), 703-709; Butler et al. Diabetes 52 (2003), 2304-2314; Hoppener et al. Diabetologia 42(4) (1999), 427-434; and Janson et al. Proc. Natl. Acad. Sci. USA 93(14) (1996), 7283-7288; compared to the here described mouse model. In addition, the mouse model of the present invention showed an appearance of extracellular amyloid deposits at 2-month of age and extensive amyloidosis observed at 4-month of age, with associated β-cell loss, in comparison to the previously described mouse models where minimal amyloid deposition were observed in mice spontaneously developing diabetes, as well as showed an extracellular amyloid deposition at 12 months of age (amyloid severity = 1 to 5% and amyloid prevalence = 40 to 60%), with associated β-cell loss, and showed extracellular amyloid deposition at 16 to 19 months of age (amyloid prevalence = 25 to 85%), with associated β-cell loss; see e.g. Hull et al. Diabetes 52(2) (2003), 372-379; Hull et al. Am J Physiol Endocrinol Metab 289 (2005), 703-709; Hoppener et al. Diabetologia 42(4) (1999), 427-434.

[0069] The novel features of the mouse model with respect to previous models (see review: Matveyenko et al. (2006), ILAR J. 47(3): 225-233) include: [0070] 1) Different genetic background: [0071] Our mouse model: hemizygous hIAPP transgenics under a FVB/NxDBA/2J background. [0072] Previously described mouse models: hemizygous hIAPP transgenics under FVB/N (Couce et al. (1996), Diabetes 45: 1094-1101), FVB/N/A.sup.vy/A (Soeller et al. (1998), Diabetes 47: 743-750), C57BL/6J (Hull et al. (2005), Am J Physiol Endocrinol Metab 289: 703-709), DBA/2J (Hull et al. (2005), Am J Physiol Endocrinol Metab 289: 703-709), C57BL/6JxDBA/2J (Hull et al. (2003), Diabetes 52(2): 372-379; Hull et al. (2005), Am J Physiol Endocrinol Metab 289: 703-709), C57BL/6J/A.sup.vy/A (Butler et al. (2003), Diabetes 52: 2304-2314), and ob/ob (Hoppener et al. (1999), Diabetologia 42(4): 427-434) background. Homozygous hIAPP transgenics under a FVB/N background (Janson et al. (1996), Proc. Natl. Acad. Sci. USA 93(14): 7283-7288). [0073] 2) Early-onset metabolic phenotype: [0074] Our mouse model: mice spontaneously develop diabetes characterized by impaired glucose tolerance and hyperglycemia already present at 1-month and 2-month of age, respectively. [0075] Previously described mouse models: [0076] mice do not spontaneously develop diabetes: hemizygous hIAPP transgenics under FVB/N (Couce et al. (1996), Diabetes 45: 1094-1101), C57BL/6J (Hull et al. (2005), Am J Physiol Endocrinol Metab 289: 703-709), and DBA/2J background (Hull et al. (2005), Am J Physiol Endocrinol Metab 289: 703-709). [0077] mice spontaneously develop diabetes with hyperglycemia and impaired glucose tolerance by 6 to 10 months of age: hemizygous hIAPP transgenics under FVB/N/A.sup.vy/A (Soeller et al. (1998), Diabetes 47: 743-750), C57BL/6J/A.sup.vy/A (Butler et al. (2003), Diabetes 52: 2304-2314), C57BL/6JxDBA/2J (Hull et al. (2003), Diabetes 52(2): 372-379; Hull et al. (2005), Am J Physiol Endocrinol Metab 289: 703-709) and ob/ob background (Hoppener et al. (1999), Diabetologia 42(4): 427-434). [0078] 3) Early onset and more prominent islet pathology: [0079] Our mouse model: appearance of extracellular amyloid deposits at 2-month of age and extensive amyloidosis observed at 4-month of age, with associated β-cell loss. [0080] Previously described mouse models: minimal amyloid deposition observed in mice spontaneously developing diabetes. Hemizygous hIAPP transgenics under C57BL/6JxDBA/2J background (Hull et al. (2003), Diabetes 52(2): 372-379; Hull et al. (2005), Am J Physiol Endocrinol Metab 289: 703-709) showed extracellular amyloid deposition at 12 months of age (amyloid severity = 1 to 5% and amyloid prevalence = 40 to 60%), with associated β-cell loss. Hemizygous hIAPP transgenics under ob/ob background (Hoppener et al. (1999), Diabetologia 42(4): 427-434) showed extracellular amyloid deposition at 16 to 19 months of age (amyloid prevalence = 25 to 85%), with associated β-cell loss.

[0081] Mice were kept in a 12 h light schedule and given normal chow diet consisting (as a percentage of total calories [kcal]) of 7% fat and 18% protein (KLIBA NAFAG). All mice had free access to food and water.

Rats

[0082] Hemizygous transgenic rats with islet β-cell expression of hIAPP driven by the rat insulin II promoter (Matveyenko et al., Diabetes (2009), 1604-1615; Butler et al., Diabetes (2004), 1509-1516) and wild-type Sprague-Dawley male rats were obtained from Charles River Laboratories (Germany). Rats were kept in a 12h light schedule and given normal chow diet. All rats had free access to food and water.

Treatment

[0083] Transgenic mice received a once-weekly administration of recombinant mouse chimeric IgG2a antibody NI-203.26C11-ch (10 mg/kg body weight; i.p.) starting at 4 weeks of age and for the duration of the study. Vehicle-treated transgenic and wild-type mice were intraperitoneally administered with physiological saline (PBS).

[0084] Transgenic and wild-type rats received a once-weekly administration of recombinant rat chimeric IgG2b antibody NI-203.26C11-r (3 mg/kg body weight; i.p.) starting at 12 weeks of age and for the duration of the study. Vehicle-treated transgenic and wild-type rats were intraperitoneally administered with physiological saline (PBS).

Oral Glucose Tolerance Test, Fasting Blood Glucose, Plasma Insulin and Plasma hIAPP

[0085] For glucose tolerance testing, 5 hour-fasted mice and overnight-fasted rats were orally administered with a 2 g/kg glucose solution. Blood samples were collected from the mouse tail vein before and 10, 30, 60, 120 and 240 min after glucose injection. Rat blood was collected from the sublingual vein and under gas anesthesia before and 30, 60, 120 and 240 min after glucose injection. Blood glucose was measured using a glucometer (CONTOUR XT sensors, Bayer). Fasting blood glucose was measured from blood samples collected after overnight fasting. Non-fasting plasma insulin and hIAPP levels were determined using a mouse insulin (Mercodia) and a human amylin (Millipore) enzyme-linked immunosorbent assay (ELISA) kit.

Histology

[0086] Mice were sacrificed and pancreases were dissected, weighted (to calculate pancreatic mass), fixed in 4% (wt/vol.) phosphate-buffered paraformaldehyde, embedded in paraffin and 3-.Math.m sections were cut. Paraffin-embedded human tissues were obtained from the University Hospital Basel (UHB, Basel, Switzerland) and 3-.Math.m sections were cut. Sections were stained with 0.1% (wt/vol.) thioflavin S to visualize amyloid deposits, polyclonal guinea pig anti-insulin antibody (1:3; FLEX; Dako), mouse monoclonal anti-IAPP antibody (1:100; R10/99; Abcam), polyclonal rabbit anti-glucagon antibody, and recombinant NI-203.26C11 (human or mouse chimeric) antibody. Following primary antibody, slices were incubated with TRITC-conjugated donkey anti-guinea pig antibody (Jackson ImmunoResearch Europe Ltd., UK), Cy5-conjugated or biotinylated donkey anti-mouse antibody, Cy5-conjugated goat anti-rabbit antibody, Cy5-conjugated or biotinylated donkey anti-human antibody. Streptavidin-biotin-peroxidase reaction (Vectastain ABC kit, Vector Lab Inc., Burlingame, USA) was used with biotinylated secondary antibodies and sections were counterstained with hematoxylin to visualize cell nuclei. Image analysis was performed using Image J software. To quantify ThioS-positive amyloid area, insulin-positive area, mean islet area and islet density, all islets per pancreatic section (five sections per animal) were examined in detail at 20x magnification. ThioS-positive amyloid area and insulin-positive area were computed as the areas corresponding to fluorescence above a preset threshold. Islet area was determined by manually outlining the islet visualized on the ThioS channel, where the outline of the islet is clearly visible. ThioS staining was expressed in relation to the pancreas area and the insulin area. Insulin staining was expressed in relation to the pancreas area and the islet area.

Statistical Analysis

[0087] Data are expressed as means ± SEM. Significant differences between pairs of groups were determined using Student’s t test. Glucose tolerance data were analyzed using a repeated-measures ANOVA (with time on study and treatment as independent variables). All statistical analyses were conducted using Prism software (GraphPad Software Inc., San Diego, CA, USA). p < 0.05 was considered significant.

Example 1: Validation of Affinity and Selectivity of the Antibodies for Pathological hIAPP in Human Tissue

[0088] To test whether the NI-203.26C11 antibody selectively and dose-dependently recognizes pathological hIAPP aggregates human tissue, i.e. in the pancreas of patients diagnosed with type 2 diabetes mellitus (T2D), paraffin-embedded pancreas sections of patients diagnosed with T2D showing amyloid load in pancreatic islets observed upon ThioS staining (subject 1 and 2) were tested for NI-203.26C11 antibody binding. As control paraffin-embedded pancreas sections of a patients not diagnosed with T2D (subject 3 and 4) were used (FIG. 2).

[0089] After formic acid pretreatment, sections were incubated with the NI-203.26C11-ch antibody at different concentrations, i.e. 1 nM, 3 nM, 10 nM, and 30 nM, or a mouse monoclonal anti-IAPP antibody (1:100; Abcam, Cambridge, UK) and secondary donkey anti-mouse antibody were used as control, followed by incubation with biotinylated donkey anti-human secondary antibody (1:500; Jackson ImmunoResearch, Newmarket, UK) or biotinylated goat anti-mouse secondary antibody (1:500; Jackson ImmunoResearch, Newmarket, UK). Antibody signal was amplified with the Vectastain ABC-AP kit (Vector Laboratories, USA) and detected with diaminobenzidine substrate (Thermo Fisher Scientific, USA). Upon avidin/biotin blocking (Avidin/Biotin blocking kit, Vector Laboratories, USA), pancreatic islet β-cells were visualized using a polyclonal guinea pig anti-insulin antibody (1:5; Dako, USA) coupled to a biotinylated donkey anti-guinea pig secondary antibody (1:500; Jackson ImmunoResearch Laboratories, USA) and antibody signal was amplified with the Vectastain ABC-AP kit (Vector Laboratories, USA) and detected with alkaline phosphatase substrate (Vector Laboratories, USA).

[0090] NI-203.26C11 antibody showed dose-dependent a staining in T2D pancreatic islets loaded with islet amyloids (subjects 1 and 2) but not in healthy control pancreatic islets lacking islet amyloids (subjects 3 and 4) (FIG. 2). However, NI-203.26C11 antibody did not recognize physiological hIAPP on healthy control pancreases, as shown in FIG. 2 two lower rows.

[0091] In addition, the selectivity of the NI-203.26C11 antibody was tested by an immunofluorescence staining (FIG. 3). Tissue sections were labeled against IAPP and pancreatic islet β-cells utilizing NI-203.26C11 antibody, mouse monoclonal anti-IAPP antibody (1:100; Abcam, Cambridge, UK) and secondary donkey anti-mouse antibody, and polyclonal guinea pig anti-insulin antibody (1:5; Dako, USA) for pancreatic islet β-cells. The visualization was performed with fluorescently labeled secondary antibodies.

[0092] In brief, paraffin-embedded sections were labeled with NI-203.26C11 or mouse monoclonal anti-IAPP antibody (1:100; Abcam, Cambridge, UK) and subsequently by Cy5-labeled secondary donkey anti-mouse IgG. Pancreatic islet β-cells, i.e. insulin, colocalization with IAPP was determined using a polyclonal guinea pig anti-insulin antibody (1:5; Dako, USA) coupled to a TRITC-labeled donkey anti-guinea pig secondary antibody (1:500; Jackson ImmunoResearch Laboratories, USA). Stained samples were cover-slipped with Tris-buffered glycerol (a 3:7 mixture of 0.1 M Tris-HCl at pH 9.5 and glycerol supplemented with 50 mg/mL n-propyl-gallate).

[0093] As result, the recombinant mouse chimeric antibody NI-203.26C11 (NI-203.26C11-ch; blue; 100 nM; FIG. 3 top panels) detected hIAPP aggregates on T2D pancreatic islets. In particular, the staining was restricted to islet areas deprived of β-cells (no merge with insulin staining in red). However, no staining was visible on physiological IAPP in healthy control pancreatic islets lacking pathological hIAPP aggregates.

Example 2: Impaired Glucose Tolerance Precedes the Deposition of ThioS-Positive Material in a Transgenic Mouse Model of Type-2 Diabetes

[0094] To test the deposition of islet amyloids in the transgenic mouse model, islet amyloids were visualized utilizing an immunofluorescence staining, as described above utilizing hIAPP transgenic and wild-type mouse pancreatic islets stained for islet amyloids (0.1% ThioS), insulin (anti-insulin antibody; 1:3; red) and glucagon (anti-glucagon antibody; 1:2500; blue), at 14 weeks of age but not 4 weeks of age in hIAPP transgenic mice, see FIG. 4A and FIG. 4B. As shown in FIG. 4A the wild-type mice were deprived of islet amyloids compared to the transgenic mice.

[0095] In addition the glucose tolerance was measured in the transgenic mice, see FIG. 4B. In brief, glucose was administered intraperitoneally to a 5-hour fasting animal at 2 mg per gram of bodyweight and then the glucose level in the blood (peripheral blood, blood glucose concentration) of the animal was determined every 15 minutes over 240 minutes using a commercial measuring apparatus (Medisafe Reader, Terumo Co., Ltd.). The measurement utilizing a commercial measuring apparatus is based on colorimetric analysis. A measuring chip is prepared, and onto the chip are placed glucose oxidase and peroxidase as catalysts and 4-aminoantipyrine and N-ethyl-N(2-hydroxy-3-sulfopropyl)-m-toluidine as chromogenic agents. When a blood sample absorbed through capillary phenomenon is placed on this chip, here 4 .Math.l, and then glucose in the blood is oxidized by glucose oxidase. Then, the chromogenic agents on the chip are oxidized by hydrogen peroxide generated at this moment and peroxidase, which yields a red-purple color. The amount of glucose in the blood is calculated by measuring the degree of this color tone. The oral glucose tolerance test was performed in transgenic and wild-type mice at 4 weeks (top panel) and 14 weeks (bottom panel) of age, see FIG. 4B. The results showed that blood glucose levels after oral glucose were higher in hIAPP transgenic mice compared with wild-type mice at 4 weeks (tg, n=13; wt, n=9) and 14 weeks of age (tg, n=27; wt, n=31). In addition, the fasting blood glucose levels were similar among groups at 4 weeks but increased in hIAPP transgenic mice at 14 weeks of age.

Example 3: Validation of Affinity and Selectivity of the Antibodies for hIAPP Aggregates in hIAPP Transgenic Mice

[0096] To test the affinity and selectivity of the NI-203.26C11 antibody against hIAPP aggregates in hIAPP transgenic mice, immunofluorescence and immunohistochemical analysis utilizing recombinant human antibody NI-203.26C11 (NI-203.26C11; blue; 100 nM) for detection of hIAPP aggregates on transgenic islets and anti-insulin antibody (1:3; red) for the detection of insulin on islet β-cells, were performed, see FIG. 5. The results showed a NI-203.26C11 staining on pancreatic islets in 4 week- and 16 week-old hIAPP transgenic mice (blue staining; left panels, FIG. 5), with no staining on age-matched wild-type mice (absence of blue staining; right panels FIG. 5).

[0097] Additionally, the NI-203.26C11 antibody was administered to 16 week-old hIAPP transgenic and wild-type mice at a single administration (10 mg/kg (i.p.)) and its binding was evaluated 2 days after administration using an anti-human secondary antibody, see FIG. 6.

[0098] It was shown that the recombinant human NI-203.26C11 (brown staining, FIG. 6) targets transgenic islets showing ThioS-positive amyloids (green staining, FIG. 6) but not ThioS-negative wild-type islets. In addition, no staining was observed with the control IgG.

Example 4: Administration of the Antibodies of the Present Invention Protects Against β-Cell Loss in hIAPP Transgenic Mice

[0099] To test the effect of the NI-203.26C11 on β-cells in hIAPP transgenic mice, pancreatic insulin, islet area, and insulin secretion was measured after a once-weekly treatment with recombinant mouse chimeric NI-203.26C11 antibody in hIAPP transgenic mice (tg NI-203.26C11-ch, n=23; 10 mg/kg i.p. for 12 weeks); see FIG. 7A and FIG. 7B.

[0100] In brief, paraffin-embedded sections were labeled with polyclonal guinea pig anti-insulin antibody (1:5; Dako, USA) and subsequently by TRITC-labeled donkey anti-guinea pig secondary antibody (1:500; Jackson ImmunoResearch Laboratories, USA). Stained samples were cover-slipped with Tris-buffered glycerol (a 3:7 mixture of 0.1 M Tris-HCl at pH 9.5 and glycerol supplemented with 50 mg/mL n-propyl-gallate). Quantification of the insulin-positive area in relation to the pancreas and islet area, mean islet area and islet density is described in the methods section.

[0101] The results showed pancreatic insulin (insulin-positive area in % pancreas area and % islet area; top left and top middle panels), islet area (mean islet area; top right panel) and insulin secretion (plasma insulin levels; bottom left panel) was increased compared with hIAPP transgenic mice receiving PBS (tg PBS, n=28). However, the islet density and pancreatic mass were unchanged after NI-203.26C11-ch treatment (bottom middle and bottom right panel, respectively; FIG. 7A and FIG. 7B).

Example 5: Antibody of the Present Invention Recognizes Predominantly Early Fibrillary IAPP

[0102] To test to which hIAPP aggregate species the antibody NI-203.26C11 binds, a Thioflavin-T (Thio-T) aggregation assay was performed. In brief, spontaneous aggregation of synthetic hIAPP was assessed by monitoring amyloid fibril formation via the increase of fluorescence of the amyloid-specific dye Thioflavin-T (Thio-T). Lyophilized synthetic hIAPP peptide (Bachem, Switzerland) was reconstituted in pure DMSO and mixed in Thio-T solution (20 .Math.M Thio-T in 20 mM Tris-HCl, pH 8.5) to a final peptide concentration of 20 .Math.M. After filtration through a 0.22 .Math.m filter (Millipore), aggregation solution was immediately transferred into fluorescence quartz cuvettes and aggregation was recorded under stirring on a Cary Eclipse Fluorescence spectrophotometer (Agilent) measuring the fluorescence emission wavelength at 489 nm (excitation at 456 nm) every 1 min at RT. The assay showed a classical sigmoidal aggregation curve for hIAPP, but not for rodent IAPP (rIAPP) as shown in FIG. 9A.

[0103] Additionally, a DotBlot analysis were performed utilizing samples from the Thio-T experiment. The DotBlot analysis was performed as follows, hIAPP preparations from aggregation assays were serially diluted and filtered through a PBS-T (0.1% Tween-20 in PBS) pre-equilibrated nitrocellulose membrane (pore size 0.1 .Math.m). Wells were washed with PBS-T and samples were added. After complete filtration the membrane was washed three times. Subsequently, the membrane was shortly air-dried for 15 min at RT, incubated in blocking buffer (3% BSA, 0.1% Tween-20 in PBS) for 1 h at RT, and incubated with mouse-chimeric NI-203.26C11 antibody (5 .Math.g/ml in blocking buffer) for 1h at RT. As control antibody, a chicken anti-IAPP antibody (1:1000; P10997, Agrisera) was used. After washing, the membrane was incubated with HRP-conjugated anti-mouse and anti-chicken IgG secondary antibodies (1:10000 dilution; Jackson ImmunoResearch Laboratories) for 1 h at RT. Conversion of HRP substrate (ECL) was analyzed using ImageQuant LAS 4000 detection (GE Healthcare). As shown in FIG. 9B a selective binding for hIAPP preparations in early phases of the hIAPP aggregation could be shown, while a decreased binding was observed in later phases of growth. However, the antibody NI-203.26C11 remained immuno-negative for non-aggregated hIAPP fractions (0′) and did not react to rIAPP.

[0104] Furthermore, filter retardation assays (FRA) were performed using hIAPP preparation from aggregation assays mixed in 2% SDS. Samples were then filtered through a cellulose acetate membrane and membrane was blocked with blocking buffer (3% milk, 0.1% Tween-20 in PBS) for 1h at RT. Detection of SDS resistant hIAPP assemblies was done as described for the DotBlot analysis. The FRA showed that the anti-IAPP antibody recognize SDS-resistant fibrillar hIAPP species at all time-points assessed, whereas no reactivity could be observed for the antibody NI-203.26C11 against these hIAPP species, see e.g. FIG. 9B.

[0105] Morphology of hIAPP assemblies was assessed by Transmission electron microscopy (TEM) analysis. In brief, samples were adsorbed on glow-discharged carbon-coated copper grids and excess of sample was removed by blotting on filter paper. Grids were stained with 2% (w/v) uranyl acetate for 60 sec and excess of solution was removed by washing with ddH20. After air-drying, grids were imaged with a Philips CM100 transmission electron microscope with an acceleration voltage of 100 kV. As shown in FIG. 9C a diverse morphological spectrum could be assessed with TEM. In particular, at time point 0 min which indicates the time point before aggregation, no large fibrillar aggregates. After 5 min amorphous non-fibrillar as well as fibrillar characteristics could be observed. A fibrillar morphology was observed in samples taken at later time points, i.e. 10 min and 20 min.

Example 6: Administration of the Antibodies of the Present Invention Prevents From Symptoms Associated With Diabetes

[0106] To proof whether the administration of NI-203.26C11 antibody can improve symptoms associated with diabetes, the recombinant mouse chimeric NI-203.26C11 antibody was administered once-weekly to hIAPP transgenic mice (tg NI-203.26C11-ch, n=24; 10 mg/kg i.p.) and the blood glucose, glucose tolerance, body weight was assessed as already described above; see FIG. 8A, FIG. 8B and FIG. 8C.

[0107] After 8 and 12 weeks of treatment the NI-203.26C11 antibody showed a significant decrease of fasting blood glucose measured after overnight fasting in hIAPP transgenic mice compared with PBS group. In addition, the glucose tolerance after 10 weeks of treatment was significant improved.

[0108] Furthermore, a normalized body weight could be observed in hIAPP transgenic mice treated with NI-203.26C11, i.e. incremental body weight (%) was normalized in NI-203.26C11-treated over 12 weeks of treatment, as compared with wild-type mice injected with PBS (wt PBS, n=31). In particular, hIAPP transgenic mice under PBS treatment showed impaired body weight gain compared with wild-type mice; see FIG. 8A, FIG. 8B and FIG. 8C.

[0109] Therefore, as a result it has been shown that NI-203.26C11 antibody decreases fasting blood glucose, improves glucose tolerance and normalizes body weight gain in hIAPP transgenic mice.

Example 7: Antibodies of the Present Invention Normalize Glucose Tolerance in a hIAPP Transgenic Rat

[0110] In addition to further improve the results shown in transgenic mice, the effects of NI-203.26C11 antibody in a transgenic rat model of type-2 diabetes (hIAPP transgenic rats) were assessed, see FIG. 10A and FIG. 10B. In particular, a recombinant rat chimeric NI-203.26C11 antibody was administered once-weekly to hIAPP transgenic rats (tg NI-203.26C11-r, n=9; 3 mg/kg i.p.) and wild-type rats (wt NI-203.26C11-r, n=4; 3 mg/kg i.p.).

[0111] To test the glucose tolerance, an oral glucose tolerance (oGTT) test was performed before treatment in 3 month-old hIAPP transgenic rats and wild-type rats showing equivalent blood glucose levels. As a result, the NI-203.26C11-r antibody normalized blood glucose levels in hIAPP transgenic rats during an oGTT performed after 8 weeks of treatment, as compared with PBS-treated hIAPP transgenic rats (tg PBS, n=10) and PBS-treated wild-type rats (wt PBS, n=5). Additionally it was shown that the NI-203.26C11-r antibody does not affect blood glucose levels in wild-type rats (wt NI-203.26C11-r).