UCLA

BACKGROUND: Previous studies suggested that vitamin D played a significant role in bone regeneration, facilitating the establishment of implant osseointegration. A whole genome microarray study further suggested that the vitamin D axis might involve circadian rhythm gene expression in the bone peripheral tissue.

OBJECTIVES: To identify key gene networks involved with vitamin D receptor in the bone regeneration process and to explore any correlation with circadian rhythm gene expression in bone marrow mesenchymal stromal/stem cells (BMSC).

METHODS: The available whole gene microarray data was analyzed using the weighted gene correlation network analysis (WGCNA) R package and Cytoscape software. Any gene expression correlation was examined for vitamin D receptor (VDR) as well as circadian rhythm genes. Separately, Per 1 luciferase transgenic Wistar rats were then applied for in vitro evaluation on BMSC circadian rhythm. Per1::luc BMSCs were seeded on 35mm dishes and forskolin-synchronized luminescence was recorded across different media conditions. The recording media included growth medium containing F12 (10% Fetal Bovine Serum, 1% Pen-Strep and antibiotic) supplemented with or without 1 nM 1,25D. Luminescence was also recorded in F12 growth medium containing bone differentiation factors (beta glycerophosphate, Ascorbic acid and Dexamethasone) supplemented with 0, 1 or 10 nM 1,25D.

RESULTS: A total of 47 gene network modules, based on the correlation between gene significance values and module membership, were generated. The "Blue" module contained VDR. With a threshold of 17 and median of 13, VDR exhibited 35 gene connections as well as correlation values of 0.31 and 0.55 for OSV+ (vitamin D sufficient condition) and OSV- (vitamin D insufficient condition) respectively. NPAS2, CLOCK, BMAL1 and casein kinase 1 were directly connected to VDR. Per1-3 and cry1-2 were each found in a different module with generally low correlation values. In vitro luminescence measures of Per1::luc rat BMSC showed a dose dependent sustained increase in Per 1 expression with the addition of 1,25D to the F12 growth medium as well as osteogenic differentiation medium. The baseline subtracted luminescence data indicated that the circadian pattern of BMSC was maintained with 1,25D supplementation.

CONCLUSION: The WGCNA data suggests that VDR may serve as the main coordinator, or hub gene, strongly correlated with the identified circadian rhythm genes. The forskolin-synchronized expression of Per1::luc exhibited circadian expression in BMSC in vitro. 1,25D supplementation significantly and dose dependently increased the baseline expression of Per1::luc, whereas the circadian rhythm was maintained. We postulate that the physiological regulation of vitamin D may mediate gene network organization via circadian rhythm gene hubs during the bone regeneration or remodeling process.