Due to a lack of spatial-temporal resolution at the single cell level, the etiologies of the bone dysfunction caused by diseases such as normal aging, osteoporosis, and the metabolic bone disease associated with chronic kidney disease (CKD) remain largely unknown.

To this end, flow cytometry and scRNAseq were performed on long bone cells from Sost-cre/Ai9⁺ mice, and pure osteolineage transcriptomes were identified, including novel osteocyte-specific gene sets.

Clustering analysis isolated osteoblast precursors that expressed Tnc, Mmp13, and Spp1, and a mature osteoblast population defined by Smpd3, Col1a1, and Col11a1. Osteocytes were demarcated by Cd109, Ptprz1, Ramp1, Bambi, Adamts14, Spns2, Bmp2, WasI, and Phex. We validated our in vivo scRNAseq using integrative in vitro promoter occupancy via ATACseq coupled with transcriptomic analyses of a conditional, temporally differentiated MSC cell line. Further, trajectory analyses predicted osteoblast-to-osteocyte transitions via defined pathways associated with a distinct metabolic shift as determined by single-cell flux estimation analysis (scFEA). Using the adenine mouse model of CKD, at a time point prior to major skeletal alterations, we found that gene expression within all stages of the osteolineage was disturbed.

In sum, distinct populations of osteoblasts/osteocytes were defined at the single cell level. Using this roadmap of gene assembly, we demonstrated unrealized molecular defects across multiple bone cell populations in a mouse model of CKD, and our collective results suggest a potentially earlier and more broad bone pathology in this disease than previously recognized.