You are here

| Haematopoetic Stem Cells

There Will be Blood: Isolation of Single Human Hematopoietic Stem Cells Capable of Long-Term Multilineage Engraftment

From Science
By Stuart P. Atkinson

A recent Stem Cells article (Woods et al) and associated review on the Stem Cells Portal described an optimized in vitro differentiation protocol for the generation of precursors of the hematopoietic lineage and primitive hematopoietic cells from human embryonic stem cells (ESC) and induced pluripotent stem cells (iPSCs). The authors themselves note however that this protocol failed to derive long-term repopulating haematopoietic stem cells (HSCs). This suggests either that the protocol is incomplete, maybe lacking certain key signals at necessary time-points, and/or that the cell surface markers used were insufficient to purify those cells capable of long-term repopulation and engraftment. In a recent paper in Science, researchers from the laboratory of John E. Dick at the Toronto Medical Discovery Tower, Toronto, Canada have identified key cell surface markers which now allow the purification of HSCs with long-term engraftment and serial transplantation ability and importantly, single-cell engraftment; the definitive assessment of HSC potential (Notta et al, 2011).

The group had previously optimized an HSC xenograft assay by using intrafemoral injection into female NOD-scid- IL2Rgc−/− (NSG) mice (Notta et al, 2010) and this was used to assay for cells capable of short-term and long-term engraftment at clonal resolution. Thy1 (CD90) was initially used in an attempt to segregate out HSCs (CD34+CD38CD45RAThy1+; Thy1+) from multipotent progenitors (MPPs) (CD34+CD38CD45RA Thy1; Thy1) collected from human cord blood samples (Majeti et al). Lymphomyeloid engraftment in NSG mice that persisted for at least 20 weeks was used as a definition of functional HSC existence. At non-limiting cell doses, recipients of Thy1+ and Thy1 cells had similar levels of human chimaerism and lineage distribution, and secondary transplants which were analysed after another 12-14 weeks showed that both populations could be serially transplanted, although the Thy1- subset did this at a lower efficiency. Limiting dilution analysis (LDA) showed that 5% of Thy1+ cells clonally initiated long-term haematopoiesis in NSG mice compared with only 1% of Thy1 cells. Overall this suggests that cells with extensive self renewal activity exist in both populations but that Thy1 can be used to enrich for HSCs.

Sorted cells were then grown on stromal cells, known to express HSC-supportive ligands, and while greater than 70% of both Thy1+ and Thy1 cells remained CD34+CD38-, intriguingly Thy1- cells consistently generated Thy1+ cells, a situation also observed in vivo within the bone marrow microenvironment of NSG mice that had received transplants. Further, these emerging Thy1+ cells showed robust activity in the engraftment assay, overall suggesting that the sorted Thy1- population is heterogeneous with small amounts of cells with repopulating activity and a larger fraction with MPP-like activity. This heterogeneity led to the search for another marker which could further enhance the purification of each population. Comparisons of several adhesion molecules known to be important in HSC-niche interactions in mouse found that ITGA6 (integrin a6 or CD49f) was differentially expressed between the Thy1+ and Thy1 populations. Therefore, sorted Thy1+Cd49f+ cells were compared to Thy1+Cd49f- in the engraftment assay and demonstrated that the Cd49f+ fraction had 6.7-fold greater chimaerism than the Thy1+Cd49f- and serial transplantation could be achieved only using the Thy1+CD49f+ population. LDA revealed that 9.5% of Thy1+Cd49f+ cells had repopulating activity compared to 0.9% in the Thy1+Cd49f- population. Interestingly Thy1-Cd49+ cells were also shown to be functional in the engraftment assay, with LDA showing that 4.5% of cells had long-term multilineage engraftment potential compared to 0.13% of Thy1-CD49f- dual negative cells. Differences in lineage potential between Thy1+Cd49f+ and Thy1-Cd49f+ were also found to be minimal, overall suggesting that human HSCs are marked with CD49f. Further analysis of peripheral blood and marrow of NSG mice after transplantation showed that Thy1-CD49f- cells had a higher engraftment and differentiation potential that HSCs immediately after transplant (2-4 weeks) and could give rise to B cells, monocytes, granulocytes and erythrocytes, together suggesting that Thy1-Cd49f- cells are in fact bona fide MPPs.

CD49f sorting therefore enables high purity human HSC isolation, but single-cell transplantation assays, the definitive assessment of HSC potential, still needed to be undertaken. It is known that high efflux of the mitochondrial dye rhodamine-123 (Rho) may enrich for HSCs within the LinCD34+CD38 fraction (McKenzie et al) and so cells were sorted for high Rho efflux (Thy1+Rholo) and low Rho efflux (Thy1+Rhohi) in the engraftment assay, and the high reflux cells showed 40-fold higher chimaerism in the injected femur and twice as many HSCs as compared to Thy1+ alone. The next obvious step was to sort on Thy1+CD49+Rholo and observe whether single sorted HSCs were capable of single cell engraftment. Indeed, 28% of NSG recipients transplanted with single cells displayed multilineage chimaerism after 20 weeks and serial transfer was successful in 50% (2 of 4) secondary recipients. A second experiment showed slightly lower numbers of engrafted cells, perhaps due to the heterogeneity of cord blood donors. Further, human cell engraftment was observed at distant marrow sites from the injected femur, overall demonstrating that Lin-CD34+CD38-CD45RA-Thy1+RholoCD49f+ cells are human HSCs.

This paper represents a significant step forward in the study of the haematopoietic system and has major implications towards those groups undertaking differentiation studies for possible clinical use. The discovery of this set of cell surface markers represents a new benchmark in the identification of HSCs capable of long term engraftment, and will likely accelerate new discoveries in this field, with the obvious next route for investigation being the use of such markers to assay cells derived from differentiating pluripotent cell sources. Perhaps even the purification of these cells can allow further more detailed study of the characteristics of these distinct populations, be that gene expression, cell surface markers or responses to external stimuli, and allow us to generate an in vitro system for propagating such cells.



Brief report: efficient generation of hematopoietic precursors and progenitors from human pluripotent stem cell lines.
Woods NB, Parker AS, Moraghebi R, Lutz MK, Firth AL, Brennand KJ, Berggren WT, Raya A, Belmonte JC, Gage FH, Verma IM.
Stem Cells. 2011 Jul;29(7):1158-64.

Isolation of Single Human Hematopoietic Stem Cells Capable of Long-Term Multilineage Engraftment
Faiyaz Notta, Sergei Doulatov, Elisa Laurenti, Armando Poeppl, Igor Jurisica, and John E. Dick
Science 8 July 2011: 218-221.

Engraftment of human hematopoietic stem cells is more efficient in female NOD/SCID/IL-2Rgc-null recipients.
Notta F, Doulatov S, Dick JE.
Blood. 2010 May 6;115(18):3704-7.

Identification of a hierarchy of multipotent hematopoietic progenitors in human cord blood.
Majeti R, Park CY, Weissman IL.
Cell Stem Cell. 2007 Dec 13;1(6):635-45.

Low rhodamine 123 retention identifies long-term human hematopoietic stem cells within the Lin-CD34+CD38- population.
McKenzie JL, Takenaka K, Gan OI, Doedens M, Dick JE.
Blood. 2007 Jan 15;109(2):543-5.