Regulation of Hematopoietic Stem Cells by Bone Marrow Stromal Cells

Date of Award

Winter 5-15-2013

Author's School

Graduate School of Arts and Sciences

Author's Department

Biology & Biomedical Sciences (Molecular Cell Biology)

Degree Name

Doctor of Philosophy (PhD)

Degree Type



Under normal conditions, the great majority of hematopoietic stem cells (HSCs) reside in the bone marrow where they are tightly regulated by signals provided by bone marrow stromal cells. In particular, spindle-shaped N-cadherin+ osteoblasts (SNO cells) and CXCL12-abundant reticular cells (CAR cells) have been suggested to be important HSC niche-supporting cells. SNO cells have been proposed to support maintenance of HSCs through interactions between N-cadherin expressed on HSCs and N-cadherin expressed on osteoblasts. In order to test this hypothesis, we generated mice conditionally deficient in N-cadherin in their osteoblast compartment. N-cadherin-deleted mice exhibited normal baseline hematopoiesis as well as normal HSC cell cycle status, repopulating activity, and self-renewal activity. Moreover, homing and engraftment of wild type cells into N-cadherin-deleted recipients was normal. Finally, we tested the response to G-CSF, a potent HSC mobilizing stimulus, which leads to a profound loss of osteoblasts. N-cadherin-deleted mice showed normal mobilization of progenitors to the blood and spleen. Together, our data show that N-cadherin expression on SNO cells (and other osteoblast-lineage cells) is dispensable for HSC maintenance and trafficking. CXCL12 is an important chemokine that regulates a wide number of HSC functions including repopulating activity, self-renewal, homing, and quiescence. CXCL12 has been reported to be produced by many cells of the bone marrow stromal including endothelial cells, mineralizing osteoblasts, osteoprogenitors, and mesenchymal stem cells, and it is currently unclear which population is relevant for maintenance of HSCs. We first show that CXCL12 is primarily produced by reticular cells, which are derived from an osterix-expressing progenitor. We next we generated a CXCL12 conditional knockout mouse and targeted deletion of the gene in mature osteoblasts using the osteocalcin-cre, in osteoprogenitors using the osterix-cre, in all mesenchymal cells using the Prx1-cre, and in endothelial cells using the Tie2-cre. We demonstrate that cells of the mesenchymal lineage demonstrated differential effects on hematopoiesis: deletion in mineralizing osteoblasts affect multipotent progenitors, in osteoprogenitors affect short-term HSCs, while deletion in early mesenchymal cells (and endothelial cells) affect long-term HSCs. These data suggest two models for CXCL12 maintenance of hematopoiesis. Either many stromal cells cooperate to produce sufficient CXCL12 to support HSCs. Alternatively, the geographic location of CXCL12 production is important, and hematopoietic cells are supported by mesenchymal cells that mirror their stage of development.


English (en)

Chair and Committee

Daniel C Link

Committee Members

Kyunghee Choi, Roberto Civitelli, Timothy J Ley, Fanxin Long, David M Ornitz


Permanent URL: https://doi.org/10.7936/K7N877QV

This document is currently not available here.