Self-assembled GLP-1/glucagon peptide nanofibrils prolong inhibition of food intake

Abstract

Introduction: Oxyntomodulin (Oxm) hormone peptide has a number of beneficial effects on nutrition and metabolism including increased energy expenditure and reduced body weight gain. Despite its many advantages as a potential therapeutic agent, Oxm is subjected to rapid renal clearance and protease degradation limiting its clinical application. Previously, we have shown that subcutaneous administration of a fibrillar Oxm formulation can significantly prolong its bioactivity in vivo from a few hours to a few days.

Methods: We used a protease resistant analogue of Oxm, Aib2-Oxm, to form nanfibrils depot and improve serum stability of released peptide. The nanofibrils and monomeric peptide in solution were characterized by spectroscopic, microscopic techniques, potency assay, QCM-D and in vivo studies.

Results: We show that in comparison to Oxm, Aib2-Oxm fibrils display a slower elongation rate requiring higher ionic strength solutions, and a higher propensity to dissociate. Upon subcutaneous administration of fibrillar Aib2-Oxm in rodents, a 5-fold increase in bioactivity relative to fibrillar Oxm and a significantly longer bioactivity than free Aib2-Oxm were characterized. Importantly, a decrease in food intake was observed up to 72-hour post-administration, which was not seen for free Aib2-Oxm.

Conclusion: Our findings provides compelling evidence for the development of long-lasting peptide fibrillar formulations that yield extended plasma exposure and enhanced in vivo pharmacological response.

Keywords: GLP-1/glucagon; depot formulations; metabolic diseases; nanofibrils; peptides; self-assembly.

Figure 1

Elongation of Aib2-Oxm and Oxm fibrils. a-b, Resonance frequency shift f3 as a function of time of the silicon dioxide QCM chip upon incubation of peptide. Elongation of Aib2-Oxm fibrils (dark blue) or Oxm (red) fibrils with 2 mg mL-1 Aib2-Oxm or Oxm, respectively (A). Linear approximation ± SD (n=3), R2 = 0.99. Control experiments: Aib2-Oxm fibrils incubated with 0.09% NaCl (blue); empty SiO2 chip incubated with 0.5 mg mL-1 Oxm (orange); Aib2-Oxm fibrils incubated with 0.5 mg mL-1 Oxm (violet); Oxm fibrils incubated with 0.5 mg mL-1 Oxm (red) (B). c-f, Representative AFM images of QCM-D sensors. Oxm fibrils before (C) and after incubation (D) with 2 mg mL-1 Oxm in 0.09% saline, and Aib2-Oxm fibrils before (E) and after (F) incubation with 2 mg mL-1 of Aib2-Oxm in 0.09% saline. Scale bar, 1 μm.

Figure 2

Morphological characterization of free and fibrillar Aib2-Oxm. a–c, AFM images of Aib2-Oxm peptide freshly dissolved at 10 mg mL-1 in 0.09% saline (A), after 5 to 6 days of incubation with orbital shaking and 2 to 3 days under quiescent conditions (B), and after incubation of 1 mg mL-1 fibrils with a 10 mg mL-1 free peptide solution in 0.09% saline for 12 days and diluted at 1 mg mL-1 in 0.09% saline. Scale bar, 1 μm (C). (D) CryoEM of fibrils at 1 mg mL-1 in 0.09% saline prepared in similar conditions as described in (C) Scale bar, 100 nm.

Figure 3

Structural properties of free and fibrillar Aib2-Oxm compared to Oxm species (A, B) and of free to fibrillar Aib2-Oxm (C, D), Free (black) and fibrillar (blue) Aib2-Oxm (line) and Oxm (dashed line) at 1 and 2 mg mL-1 in 0.09% saline. Far-UV CD (A) and ATR FT-IR (B) spectra, ThT (C) and Trp (D) fluorescence spectra.

Figure 4

Dissociation of fibrillar Aib2-Oxm. Peptide release (%) after incubation of 1 mg mL-1 fibrillar Aib2-Oxm in eight media and incubated for 4 h (green) and 48 h (black) at 37°C.

Figure 5

Characterization of released Aib2-Oxm. a-d, Far-UV CD spectrum (A) and analysis (B) of free (black) and fibrillar Aib2-Oxm (blue) in 0.09% saline and released Aib2-Oxm (green) in water, examples of a representative AFM image (C) mass spectrum (D) of released Aib2-Oxm in water. Scale bar, 1 μm.

Figure 6

Dissociation profile of Aib2-Oxm nanofibrils. (A) Phase changes as a function of time on top of the DPI sensor coated with collagen and fibrillar Aib2-Oxm in water (blue) and PBS (black). b-d, Representative AFM images of Aib2-Oxm nanofibrils deposited onto a collagen layer before (B) and after incubation in PBS (C) and water (D) for 18 h. Scale bar, 1 μm.

Figure 7

Agonist potency and cytotoxicity profiles of Aib2-Oxm species. a and b, In vitro potencies determined in cAMP accumulation assays in CHO cell lines expressing human GLP-1 (A) and GCG (B) receptors of free Aib2-Oxm (green), released Aib2-Oxm (violet), fibrillar Aib2-Oxm (red), free Oxm (blue), glucagon (black), GLP-1 (black, open circles). Data show representative curves of > 5 independent experiments. Curve data are the arithmetic mean ± s.d. of duplicate data points. (C) Forty-eight-hour cytotoxicity prolife in CHO-GLP-1R cells of vehicle (black), free Aib2-Oxm (green), released Aib2-Oxm (violet), fibrillar Aib2-Oxm (red), Staurosporine (brown). RFU, Relative fluorescence units. Data show representative curves from 3 independent experiments.

Figure 8

Pharmacokinetic profiles following administration of free and fibrillar Aib2-Oxm. Aib2-Oxm bioactivity in rat (A) or mouse (B) serum determined using in vitro cell-based cAMP bioassay for determining GLP-1 receptor agonist bioactivity after s.c. administration of 10 mg kg-1 of free Aib2-Oxm (red) or fibrillar Aib2-Oxm (blue). n=3 animals dosed with each material. Data shown as mean (line) and individual points above LOQ (dotted lines).

Figure 9

Effect of fibrillar Aib2-Oxm on food intake. a-d, Effect of fibrillar Aib2-Oxm (red, 15 mg.kg-1) vs vehicle (black), free form Aib2-Oxm (green, 15 mg.kg-1) and liraglutide (purple, 40 µg.kg-1) at single injection site on food intake in C57B6J lean mice. Accumulated food intake in lean mice (A), food intake from returning the food until end of the dark cycle [first 14-hour, (B)] and 2nd dark cycle (C), accumulate 48h food intake (D). Statistic was carried out with One-way ANOVA by Dunnett’s. * and *** are P ≤ 0.05 and 0.001 when compared to vehicle control respectively. p=0.089 vs vehicle by unpair T test. n=8 per group.

Figure 10

Effect of fibrillar Aib2-Oxm vs free form Aib2-Oxm at 40 or 80 mg.kg-1 at single or multiple injection sites on food intake in C57B6J lean mice. (A), Mean accumulated food intake in lean mice. From returning the food until end of the first dark cycle (First 14-hour, B) and second dark cycles (C) of food intake after returning food. Food intake at 48 hours (D) and 72 hours (E) after returning food. Statistics was carried out with One-way ANOVA by Dunnett's. *, **, *** and **** are P ≤ 0.05, 0.01, 0.001 and 0.0001 when compared to vehicle control respectively. &, &&, and &&&& are P ≤ 0.05, 0.01, 0.001 and 0.0001 when compared to the same doses of free Aib2-Oxm respectively by unpaired t test. Liraglutide was dosed at 40 µg.kg-1; Free Aib2-Oxm 40/3: free Aib2-Oxm at 40 mg.kg-1 over 3 sites. Fibrillar Aib2-Oxm 40/3: fibrillar Aib2-Oxm at 40 mg.kg-1 over 3 sites. Fibrillar Aib2-Oxm 40/1: fibrillar Aib2-Oxm at 40 mg.kg-1 over 1 site. Free Aib2-Oxm 80/3: free Aib2-Oxm at 80 mg.kg-1 over 3 sites. Fibrillar Aib2-Oxm 80/3: fibrillar Aib2-Oxm at 80 mg.kg-1 over 3 sites. n=8 per group.