Abstract
Osteolineage cells within the bone marrow microenvironment have been implicated in support and regulation of hematopoietic stem cells (HSCs). Recently, augmented hypoxia-inducible factor (HIF) signaling in osteoprogenitors has been shown to expand the HSC niche, and surprisingly these cells have also been demonstrated to express erythropoietin, the critical cytokine stimulating erythropoiesis. We therefore hypothesize that endosteal cells may represent an additional regulatory site for erythropoiesis. To further delineate the role of the osteolineage cells in the support of erythropoiesis, we isolated bone associated cells (BACs) with enzymatic digestion of adult C57bl/6 mice hind limbs after bone marrow flushing and depleted the BACs of CD45+ cells to enrich for osteogenic cells.
We suspected some contribution of erythroid cells to CD45- BACs, however we were surprised to find that ter119+ cells represented a large percentage of BACs after enzymatic digestion. After CD45 depletion, ter119+ cells constituted about 30% percent compared to approximately 0.85% of CD45+ cells (33 ± 4.4vs. 0.85 ± 0.26, p= 0.0018) by flow cytometric analysis. Additionally, CD45 depleted BACs had approximately 46 fold higher osteocalcin expression than CD45+ cells (1300 ± 120 vs. 28 ± 9.5, p < 0.0001), while CD45/Ter119/CD31 depleted BACs had approximately 2000 fold higher osteocalcin expression than CD45/Ter119/CD31 (+) cells (2000 ± 520 vs. 0.98 ± 0.02, p= 0.0044) by qRT-PCR, confirming enrichment of the osteoblastic lineage by this immunophenotypic panel. These data suggest that there are a large number of erythroid lineage cells associated with the BACs along the endosteum. In the bone marrow of adult mice, ter119 + cells represented approximately 85% in the CD45- pool as compared to 5% in the CD45+ cell pool.
To determine if the endosteum is an active site of erythropoiesis, we quantified erythroid progenitors and precursors in the BAC pool compared to whole bone marrow (wbm) and peripheral blood (pb) by both flow cytometric analysis and colony forming assays. Flow cytometric analysis demonstrated the presence of every phase of erythroid differentiation in the BAC pool, including the presence of phenotypic MEPs (wbm vs bac vs pb: 250 ± 30 vs 84 ± 22 vs 0), BFU-E (wbm vs bac vs pb: 300 ± 14 vs 110 ± 36 vs 0 ), CFU-E (wbm vs bac vs pb: 2900 ± 2 vs 430 ± 23 vs 1 ± 0.8) and proerythroblasts (wbm vs bac vs pb: 11000 ± 2500 vs 7600 ± 1600 vs 2300 ± 920) per million cells. The phenotypic frequency of CFU-E was particularly remarkable in the BACs (430 ± 23) as compared to peripheral blood (1 ± 0.8) , demonstrating that all stages of erythroid differentiation are found in tight association with the endosteum and are not due to contamination from circulating erythroid progenitors. Colony assays were performed for CFU-E (wbm vs. bac 108 ± 16 vs 6.3 ± 2 colonies per 20,000cells plated), BFU-E (wbm vs. bac 55 ±1.0 vs 2 ±1.0; colonies per 40,000 cells plated) and myeloid progenitors (wbm vs. bac 66 ± 28 vs 11 ± 2.5 ; colonies per 10,000 cells plated) also confirmed the presence of erythroid progenitors at endosteal sites.
Together these results identify the endosteal surface as a site for erythroid differentiation. The presence of all phases of erythroid lineage differentiation in the BACs suggests a potential role for osteolineage cells for maintenance and regulation of erythropoiesis. Whether osteolineage cells contribute to erythroid lineage homeostasis and/or stress response, and whether activation or damage to osteolineage cells alters local erythroid differentiation remains to be demonstrated. However our data suggest further study of the endosteum and osteolineage cells as a potential and unexpected site of erythroid regulation, which could potentially be targeted to accelerate erythropoiesis and treat anemia.
No relevant conflicts of interest to declare.
Author notes
Asterisk with author names denotes non-ASH members.
This feature is available to Subscribers Only
Sign In or Create an Account Close Modal