Matrix Vesicle-Embedded 3D-Collagen Bio-Scaffolds: An Innovation for Investigating Bone Mineralization Dynamics

Bone mineralization is a highly orchestrated and dynamic process that has been extensively studied from a structural, chemical, and biological point-of-view. The bone tissue can be described by the action of mineralization-competent cells embedded in an extracellular matrix composed mainly by collagen. Those cells release a special class of extracellular vesicles, named matrix vesicles (MVs) which are responsible for the deposition of an inorganic phase called biological apatite on collagen fibrils.Thus, the development of 3D-collagen scaffolds embedded with MVs offers alternative strategies to investigate bone mineralization in order to contribute for the development of cell-free therapies. Here, self-assembled 3D-matrices were obtained from the exposure of high concentrated type I collagen solutions to NH_3(g) for 24 hours, replicating the microenvironment of the native tissue. MVs were isolated by ultracentrifugation from a pre-osteoblastic cell line (MC3T3) and vascular smooth muscle (MOVAS) cell line cultivated in osteogenic medium. The average mean diameter of both the vesicles was 105 + 19 nm, as determined by nanoparticles tracking analysis (NTA), confirmed by atomic force microscopy. The specific activity of TNAP was 7.5 and 1.2 + 0.6 U.mg^-1 for MVs from MCT3T and MOVAS respectively, confirming their mineralizing potential. FTIR spectroscopy confirmed the formation of apatite after immersion of the vesicles in synthetic cartilage lymph. Morphological analysis developed by scanning electron microscopy (SEM) and spectroscopic analysis, revealed the ability of MVs to bind to the collagen scaffolds and propagate the mineral formation from the addition of ATP, simulating the process observed in vivo. Thus, reproducing the organization of collagen fibrils embedded with MVs in vitro is a suitable method to design mimetic models to study (patho)physiological mineralization processes. Acknowledgements: This work was supported by CNPq (200506/2022-6) and FAPESP (2019/25054-2).