An HSC stains with an antibody against the SLAM protein CD150 (arrow; red), but not antibodies against two other SLAM proteins (CD41 and CD48; green). This SLAM protein 'signature' enabled Kiel et al. to show that many HSCs contact blood vessel endothelial cells in the bone marrow (asterisk indicates lumen of blood vessel). Megakaryocyte in yellow. Credit: Reprinted with permission of Cell Press.

Stem cells are hard to find. Although they account for the output of new cells in the body, they are relatively rare, and appropriate molecular markers to pin them down have been difficult to identify. Such problems have plagued researchers studying hematopoietic stem cells (HSCs), the cells in the bone marrow that sustain lifelong blood formation and provide the material for many transplantation therapies for blood disorders.

A recent study in Cell (1016/j.cell.2005.05.026; 2005) goes a long way to alleviate this situation. Using microarray techniques Kiel et al. identify a set of three cell-surface proteins that precisely delineate HSCs and some of their more mature progeny within mouse bone marrow. With such markers in hand, the authors could also peer into stem cell 'niches'—where HSCs reside in cellular microenvironments. Although complex procedures using many markers have allowed the isolation of these cells, it had previously been difficult to visualize them in their normal bone marrow locations.

The proteins identified by Kiel et al. are members of the SLAM cell-surface receptor family, which regulate lymphocyte proliferation and function. The authors showed that HSCs in mouse bone marrow can be purified to near homogeneity by using antibody reagents directed against 3 different SLAM proteins, CD150, CD244 and CD48.

HSCs expressed the SLAM protein CD150, but not CD244 and CD48. They also examined bone marrow cells were CD150+ CD48− and CD41− (a non-SLAM family protein that is expressed on CD150+ CD48− cells); these CD150+CD48−CD41− cells encompassed all HSCs. This was shown by the ability of at least one in two of these cells to function after transplantation to produce all blood cell types for lifelong time intervals. These CD150+ CD48− CD41− cells meet all of the 'gold standard' criteria used to define HSCs, including extensive self-renewal potential. Multipotent, though not highly self-renewing, progenitor cells were found in the CD244+ CD150− CD48− compartment, whereas more restricted progenitors expressed CD48 but not the other two SLAM receptors. Collectively, a simple combination of three cell-surface markers reveals a SLAM 'code' that precisely defines three functionally distinct compartments of the HSC and progenitor cell lineage hierarchy.

Previous studies have defined distinct cellular compartments within the HSC hierarchy. What distinguishes the present work is the 'simplicity' of the definition, which allowed the investigators to purify HSCs, and to begin identifying their exact locations in situ. It has been shown that HSCs reside in proximity to the inner surfaces of the bone, probably in direct contact with osteoblast cells. This special relationship appears to be necessary to maintain HSCs in their largely quiescent or slowly proliferating state. There is also evidence that at least some HSCs contact blood vessel sinusoidal endothelial cells within the bone marrow—Kiel et al. suggest that most HSCs have this location. This may explain why HSCs can be mobilized into the circulation with very rapid kinetics.

It is unprecedented that three members of the same receptor family are so precisely and differentially expressed in three HSC and progenitor cell compartments. The genes encoding the 10 or 11 SLAM family members are linked; thus, future studies are likely to provide insight into the transcriptional regulatory mechanisms that distinguish closely related cell populations within the HSC hierarchy. The new work also sets the stage for a more detailed view of how the HSC niche influences the biology of these stem cells.