Atherosclerosis and bone loss in humans: results from deceased donors and from patients submitted to carotid endarterectomy

Abstract

Background and Aims: Atherosclerosis and osteoporosis share common risk factors, as well as inflammatory mechanisms. Our aim was to understand how atherosclerotic lesions are related with disturbances in bone. Methods: Gene expression of pro-inflammatory and bone metabolism related proteins (IL-1β, IL-6, IL-17A, TNF, RANKL, OPG, COL1, CTSK, OCL, TRAP, CBFA1, DKK1, SOST, ADIPOQ, and ADIPOR1) were analyzed in arteries and bones from 45 deceased donors and adipose tissue was used as control. Additionally, in 139 patients with advanced atherosclerosis submitted to carotid endarterectomy we compared calcium content (Alizarin red) and plaque inflammatory scores (CD3+, CD68+, and adiponectin) of patients with normal bone mineral density (BMD) with those with low BMD and explored the associations between gene expression in atherosclerotic plaques and BMD. Serum levels of pro-inflammatory and bone related proteins were measured both in donors and patients. Associations were investigated by the Pearson or Spearman correlation tests, and multivariate regression analyzes were performed when justified. Results: Gene expression of bone remodeling and pro-inflammatory proteins correlated positively in bone and aorta, independently of age and sex of donors, but not in adipose tissue. The expression of bone formation genes was significantly higher in atheroma plaques from endarterectomized patients with normal vs. low BMD as well as inflammatory CD68+ scores, regardless of patients' age and sex, but not of body mass index. No relationship was observed between serum levels and gene expression levels of pro-inflammatory or bone remodeling proteins. Conclusions: Our results suggest that the relationship between bones and vessels in the context of atherosclerotic disease and osteoporosis may rely on the intrinsic connection between the tissues involved, independently of disease stage. Serum measurements of pro-inflammatory and bone-remodeling proteins do not accurately translate tissue pathologic processes.