Stem cell markers: less is more!

Comment on Yilmaz et al, page 924

Yilmaz and colleagues present a simplified technique for the prospective identification and purification of hematopoietic stem cells; unlike previous methods, their method is effective in a variety of contexts, including old marrow, mobilized peripheral blood, and recipients of long-term engrafted transplants.

In this issue of Blood, Yilmaz and colleagues address a sine qua non of stem cell biology—the prospective identification and purification of viable hematopoietic stem cells. The Morrison lab determined that a simple combination of 2 cell-surface markers, CD150 and CD48, is sufficient to identify all of the long-term repopulating activity from not only young mouse bone marrow, but also, importantly, from bone marrow of old mice, from long-term engrafted radiation chimeras, and from mobilized stem cells found in the spleen. This feat has not been possible with other stem cell purification techniques. In a paper in Cell earlier this year,¹  this group detailed the use of members of this SLAM family of cell receptors to characterize specific developmental steps in the stem and progenitor cell hierarchy in bone marrow of young mice and arrived at the simple SLAM “code” for stem cells as CD150⁺ and CD48^-/CD244^- (see figure).

Heretofore, typical methods have relied on a combination of up to 10 cell-surface markers to characterize hematopoietic stem cells. Despite its difficulty and nuances, this technique has been in successful practice, and has advanced our understanding of stem cell biology, since its original publication in 1988.²  As interest in hematopoietic stem cells from different aged animals and from different hematopoietic sources has grown, limitations have become apparent, the most significant of which has been the issue of whether the stem cell markers that were used maintained their stem-cell fidelity. In earlier studies, Morrison et al showed that the same technique that yielded a population in which 1 in every 5 cells purified from young bone marrow had long-term engrafting potential yielded dramatically lower purities of functional stem cells from other sources.^3,4 

The disparity in apparent purities could be attributable to several possibilities. One is that expression of the traditional panel of markers is altered on cells either during aging, mobilization, or prior transplantation such that the staining profile is no longer inclusive of stem cells and/or is no longer exclusive of contaminating cells. A second possibility is that the requisite steps leading to long-term engraftment—that is, homing of stem cells to the marrow and the maintenance of long-term hematopoiesis—are less efficient in stem cells derived from contexts other than young bone marrow. Compelling data in the present paper demonstrate that the bulk of the disparity is accounted for by the presence of cells in the population identified as stem cells (Lin^-, Sca-1⁺, c-Kit⁺, Thy1^lo) that demonstrably lack stem cell function. In contrast, the population of CD150⁺, CD48^- cells from all hematopoietic sources tested showed dramatically improved stem-cell purities, although not quite to the highest level of enrichment obtained in populations purified from young bone marrow. With respect to aged stem cells, the difference between apparent young and old purities is probably accounted for by an approximately 3-fold less efficient homing of old stem cells to the marrow microenvironment.^3,5 

The findings of Yilmaz et al should significantly simplify stem cell purification, while at the same time improving purity. The results also raise important considerations for studies in which stem cells purified from different contexts are contrasted. For example, in comparing gene or protein expression profiles between stem cells from mice of different ages, it is crucial that the populations compared contain equivalent enrichments for cells with stem cell function. Purifications using the proper SLAM code should ensure that apples are compared to apples. ▪