Ex vivo phosphorus-31 solid state magnetic resonance spectroscopy identifies compositional differences between bone mineral and calcified vascular tissues.

Aim: Physiological bone mineralization and ectopic vascular calcification share similarities in the composition of calcium phosphate minerals. Evidence suggests a connection between the underlying biological mechanisms driving the deposition of bone mineral and cardiovascular calcification. Therefore, understanding the chemistry and composition of bone mineral and vascular calcification may be important for the development of effective treatments and diagnostic tools for cardiovascular diseases, as pharmacological interventions for the treatment of one process might affect the other. The goal of this study was to identify and compare compositional features of calcium phosphates in bone and calcified vascular tissues using phosphorus-31 (31P) solid state cross-polarization magic angle spinning (CPMAS) magnetic resonance (MR) spectroscopy, a specialized technique that provides compositional information unattainable through conventional chemical analysis. Methods: Solid state MR spectra were acquired from biological specimens of human trabecular bone (n = 1), human vascular plaque (n = 1), human calcified aortic valves (n = 5), as well as calcified aortic tissues of apolipoprotein E-deficient mice (n = 1) fed a high cholesterol diet. Synthetic hydroxyapatite [Ca10(OH)2(PO4)6] and synthetic brushite (CaHPO4·2H2O) were used to model the solid state 31P MR spectra of the phosphate ion PO4 -3 and hydrogen phosphate ion HPO4 -2, respectively. Qualitative spectral features and quantitative metrics derived using Herzfeld-Berger analysis were assessed to characterize mineral composition and maturity. Results: Solid state 31P MR spectra of all human specimens studied suggested a well-ordered crystal structure dominated by unprotonated phosphate (PO4 -3), consistent with mature bone-like mineral. These specimens exhibited long CP time constants (700-900 µs) and modest chemical shift anisotropy. In contrast, the calcified mouse aorta spectrum showed pronounced sidebands, a short CP time constant (∼270 µs), and a more prominent HPO4 -2 component-features indicative of immature, newly deposited mineral. Conclusion: 31P solid state MR spectroscopy reveals differences in the phosphate and hydrogen phosphate ion content among the calcified tissues studied. This technique could potentially be an important complement to basic studies of pathological calcification in atherosclerosis and related calcific disorders.