Introduction: Calcification of soft tissues is a common age-related pathology that primarily occurs within vascular tissue. The mechanisms underlying pathological calcification in humans and tissue specificity of the process is still poorly understood. Previous studies examined calcified tissues on one to one basis, thus preventing comparison of deregulated pathways across tissues.

Purpose: This study aimed to establish common and tissue-specific changes associated with calcification in aorta, artery tibial, coronary artery and pituitary gland in subjects from the Genotype-Tissue Expression (GTEx) dataset using its RNA sequencing and histological data.

Methods: We used publicly available data from the GTEx database https://gtexportal.org/home/aboutGTEx. All GTEx tissue samples were derived by the GTEx consorcium from deceased donors, with age from 20 to 79, both men and women. GTEx study authorization was obtained via next-of-kin consent for the collection and banking of de-identified tissue samples for scientific research. Hematoxylin and eosin (H&E) staining of arteries were manually graded based on the presence of calcification on a scale from zero to four, where zero designates absence of calcification and four designates severe calcification. Samples with fat contamination and mislabeled tissues were excluded, which left 430 aorta, 595 artery tibial, 124 coronary artery, and 283 pituitary samples for downstream gene expression analysis. Transcript levels of protein-coding genes were associated with calcification grade using sex, age bracket and cause of death as covariates, and tested for pathway enrichment using gene set enrichment analysis.

Results: We identified calcification deposits in 28 (6.5%) aortas, 121 (20%), artery tibials, 54 (43%), coronary arteries, and 24 (8%) pituitary glands of GTEx subjects. We observed an age-dependent increase in incidence of calcification in all vascular tissues, but not in pituitary. Subjects with calcification in the artery tibial were significantly more likely to have calcification in the coronary artery (OR = 2.56, p = 6.3e-07). Markers of calcification previously established in preclinical and in vitro studies, e.g., BMP2 and RUNX2, were deregulated in the calcified tibial and coronary arteries, confirming the relevance of these genes to human pathology. Differentially expressed genes associated with calcification poorly overlapped across tissues suggesting tissue-specific nuances in mechanisms of calcification. Nevertheless, calcified arteries unanimously down-regulated pathways of intracellular transport and up-regulated inflammatory pathways suggesting these as universal targets for pathological calcification. In particular, PD-1 and PD-L1 genes were up-regulated in calcified tissues but not in the blood of the same subjects, suggesting that localized inflammation contributes to pathological calcification.

Conclusion: Pathological calcification is a prevalent disease of aging that shares little changes in expression in individual genes across tissues. However, our analysis suggests that it potentially can be targeted by alleviating local inflammation of soft tissues.