Dapagliflozin and Liraglutide Therapies Rapidly Enhanced Bone Material Properties and Matrix Biomechanics at Bone Formation Site in a Type 2 Diabetic Mouse Model

Abstract

Aim: To compare head-to-head the effects of dapagliflozin and liraglutide on bone strength and bone material properties in a pre-clinical model of diabetes-obesity. Materials and methods: Combined low-dose streptozotocin and high fat feeding were employed in mice to promote obesity, insulin resistance and hyperglycaemia. Mice were administered daily for 28 days with saline vehicle, 1mg/kg dapagliflozin or 25 nmol/kg liraglutide. Bone strength was assessed by three-point bending and nanoindentation. Bone material properties were investigated by Fourier transform infrared microspectroscopy/imaging. Results: Although diabetic controls presented with dramatic reductions in mechanical strength, no deterioration of bone microarchitecture was apparent. At the tissue level, significant alterations in phosphate/amide ratio, carbonate/phosphate ratio, tissue water content, crystal size index, collagen maturity and collagen glycation were observed and linked to alteration of matrix biomechanics. Dapagliflozin and liraglutide failed to improve bone strength by 3-point bending or bone microarchitecture during the 28-day-treatment period. At bone formation site, dapagliflozin enhanced phosphate/amide ratio, mineral maturity, and reduced tissue water content, crystal size index and collagen glycation. Liraglutide had significant effects on phosphate/amide ratio, tissue water content, crystal size index, mature collagen crosslinks, collagen maturity and collagen glycation. At bone formation site, both drugs modulated matrix biomechanics. Conclusions: This study highlighted that these two molecules are effective in improving bone material properties and modulating matrix biomechanics at bone formation site. This study also highlighted that the resulting effects on bone material properties are not identical between dapagliflozin and liraglutide and not only mediated by lower blood glucose.

Biomechanical, Microstructural and Material Properties of Tendon and Bone in the Young Oim Mice Model of Osteogenesis Imperfecta

“…Right femurs were embedded undecalcified in pMMA after dehydration and infiltration, as previously reported [ 35 ]. One micrometre-thick sections of the midshaft femur were cut with an ultramicrotome (Leica EM UC7, Leica microsystems, Nanterre, France) and deposited on BaF2 windows.…”

“…The contribution of pMMA at ~1730 cm −1 was computed and a nanoporosity ratio (area ratio 1730 cm −1 /v1,v3 PO4) was calculated. After pMMA subtraction, the spectrum was vector-normalized for the v1, v3 phosphate band and analysed with a routine script in Matlab R2021b (The Mathworks, Natick, MA, USA), as previously reported [ 35 ]. Phosphate/amide I ratio (area of v1,v3 phosphate/area amide I), mineral crystallinity (1/full width at half-maximum of the v1 PO 4 peak at 960 cm −1 ), crystal size index (area ratio: ~1076 cm −1 /~1058 cm −1 ), acid phosphate content (area ratio: ~1127 cm −1 /~1096 cm −1 ), carbonate/phosphate ratio (intensity of v2 carbonate located at 850–900 cm −1 /v1,v3 PO4 band) and collagen maturity (area ratio: ~1660 cm −1 /1690 cm −1 ) were computed.…”

Efficacy and Safety of Dapagliflozin versus Liraglutide in Patients with Overweight or Obesity and Type 2 Diabetes Mellitus: A Randomised Controlled Clinical Trial in Tianjin, China

“…In patients receiving metformin-based background therapy who have an increased cardiovascular risk, specific GLP-1RAs and SGLT-2is have demonstrated favourable effects on certain cardiovascular outcomes [ 40 ]. However, studies have also shown that the effects of dapagliflozin and liraglutide on bone material properties are not identical and are not only mediated by low blood glucose levels [ 41 ]. Therefore, other effects of these two new hypoglycaemic drugs need to be confirmed in further studies.…”

[Gly²]-GLP-2, But Not Glucagon or [D-Ala²]-GLP-1, Controls Collagen Crosslinking in Murine Osteoblast Cultures

“…Post-processing was performed with a lab-made Matlab routine (R2020b, The Mathworks, Natick, CA) and included Mie scattering correction, second-derivative spectroscopy, and curve fitting routines. Trivalent collagen crosslink content (Area ratio 1660 cm -1 /Amide 1), divalent collagen crosslink content (Area ratio 1690 cm -1 /Amide 1), and collagen maturity (area ratio 1660/1690 cm -1 ) were computed as reported previously ( 23 ).…”