Abstract
Abstract 2581
The bone marrow (BM) microenvironment consists of a heterogeneous population including mesenchymal stem cells and as well as more differentiated cells like osteoblast and adipocytes. These cells are believed to be crucial regulators of hematopoetic cell development, however, so far, their identity and specific functions has not been well defined. We have by using Ebf2 reporter transgenic Tg(Ebf2-Gfp) mice found that CD45−TER119−EBF2+ cells are selectively expressed in non-hematopoietic cells in mouse BM and highly enriched with MSCs whereas the EBF2− stromal cells are very heterogenous (Qian, et al., manuscript, 2010). In the present study, we have subfractionated the EBF2− stromal cells by fluorescent activated cell sorter (FACS) using CD44. On contrary to previous findings on cultured MSCs, we found that the freshly isolated CD45−TER119−EBF2+ MSCs were absent for CD44 whereas around 40% of the CD45−TER119−EBF2− cells express CD44. Colony forming unit-fibroblast (CFU-F) assay revealed that among the CD45−LIN−EBF2− cells, CD44− cells contained generated 20-fold more CFU-Fs (1/140) than the CD44+ cells. The EBF2−CD44− cells could be grown sustainably in vitro while the CD44+ cells could not, suggesting that Cd44− cells represents a more primitive cell population. In agreement with this, global gene expression analysis revealed that the CD44− cells, but not in the CD44+ cells expressed a set of genes including connective tissue growth factor (Ctgf), collagen type I (Col1a1), NOV and Runx2 and Necdin(Ndn) known to mark MSCs (Djouad et al., 2007) (Tanabe et al., 2008). Furthermore, microarray data and Q-PCR analysis from two independent experiments revealed a dramatic downregulation of cell cycle genes including Cdc6, Cdca7,-8 and Ki67, Cdk4-6) and up-regulation of Cdkis such as p57 and p21 in the EBF2−CD44− cells, compared to the CD44+ cells indicating a relatively quiescent state of the CD44− cells ex vivo. This was confirmed by FACS analysis of KI67 staining. Furthermore, our microarray analysis suggested high expression of a set of hematopoietic growth factors and cytokines genes including Angiopoietin like 1, Kit ligand, Cxcl12 and Jag-1 in the EBF2−CD44− stromal cells in comparison with that in the EBF2+ or EBF2−CD44+ cell fractions, indicating a potential role of the EBF2− cells in hematopoiesis. The hematopoiesis supporting activity of the different stromal cell fractions were tested by in vitro hematopoietic stem and progenitor assays- cobblestone area forming cells (CAFC) and colony forming unit in culture (CFU-C). We found an increased numbers of CAFCs and CFU-Cs from hematopoietic stem and progenitor cells (Lineage−SCA1+KIT+) in culture with feeder layer of the EBF2−CD44− cells, compared to that in culture with previously defined EBF2+ MSCs (Qian, et al., manuscript, 2010), confirming a high capacity of the EBF2−CD44− cells to support hematopoietic stem and progenitor cell activities. Since the EBF2+ cells display a much higher CFU-F cloning frequency (1/6) than the CD44−EBF2− cells, this would also indicate that MSCs might not be the most critical regulators of HSC activity. Taken together, we have identified three functionally and molecularly distinct cell populations by using CD44 and transgenic EBF2 expression and provided clear evidence of that primary mesenchymal stem and progenitor cells reside in the CD44− cell fraction in mouse BM. The findings provide new evidence for biological and molecular features of primary stromal cell subsets and important basis for future identification of stage-specific cellular and molecular interaction pathways between hematopoietic cells and their cellular niche components.
No relevant conflicts of interest to declare.
Author notes
Asterisk with author names denotes non-ASH members.