Genetic Regulation of Self-Renewal and Differentiation in Normal and Leukemic Hematopoiesis

The production of blood cells and lymphocytes is maintained throughout life from multipotential hematopoietic stem cells in the bone marrow. When stem cells enter active growth, they generate two kinds of progeny. Some are pre-programmed to differentiate. Other progeny, generated in a process known as self-renewal, remain as stem cells, thereby guaranteeing the persistence of stem cells and the permanence of blood cell formation. Competition between self-renewal and differentiation of stem cells is a feature common to the maintenance not only of most normal tissues but also of most cancers. Our work is concerned with the genes and mechanisms that specify self-renewal and differentiation in normal and leukemic hematopoietic stem cells.

We are interested in identifying known and novel genes that are specifically induced or downregulated in stem cells or subsequent early steps of differentiation. To do so, we need to be able to sample transcripts specifically from the various kinds of hematopoietic precursor cells. Our lab has made advances in three areas that now place such stage-specific transcripts at our disposal. First, we have developed technology for amplifying all transcripts present in a single sampled cell while preserving relative abundance relationships (9). Next, by exploiting the similarity of differentiation potential of sibling cells within nascent clones, we have built up an extensive archive of amplified cDNA samples representative of key stages in hematopoietic precursor development (1,4,6,). Finally, we have been able to purify the most primitive stem cells to near homogeneity, achieving in vivo reconstitution from each single cell injected into suitable murine hosts (2,10,13 and Figure).


10 of 12 mice, each injected with an average of 1.4 purified stem cells, regenerate significant numbers of red cells of donor origin 8 weeks after transplant. The figure shows electrophoresis of peripheral erythrocyte proteins. In each lane, corresponding to one mouse, the upper band is donor erythrocyte GPI1a and the lower band is host GPI1b. For a band to be visible (1% of total GPI1) the transplanted cell must generate at least 200 million erythrocytes.

We are using similar research strategies to unravel the genetic basis of maintenance of leukemic disease by leukemia stem cells. In human acute myeloblastic leukemia, leukemia cells in a patient differ widely from one another in potential for further growth. We have recently been able to demonstrate that proliferative potential is programmatically rather than randomly determined (Figure, and 14).

Single leukemia cells were allowed to grow to form clones containing 4 cells each. Two such clones are illustrated schematically at the top of the figure. From each clone, single sister cells were transferred to individual culture wells and given time to multiply. Each sister cell from the clone at the left grew little or not at all, whereas all sister cells from the clone at the right grew extensively. The figure summarizes the general results obtained from hundreds of clones analyzed in this way: sibling leukemia cells share the same pre-programmed potential for growth.

These advances are coupled with new technologies adapted and developed in the lab for massively parallel detection of expression differences by hybridization of globally amplified

A portion of a glass slide imprinted with 1760 individual cDNA fragments. Globally amplified cDNA from 15 single growth- competent cells was labelled with Cy3 (green in illustration). Pooled cDNA from 18 single growth-incompetent cells was labelled with Cy5 (red). A mixture of the two labelled cDNA pools was hybridized to the chip. Green-labelled genes were expressed most strongly by growth-competent cells, red-coloured genes by growth-incompetent cells. Yellow/orange spots were labelled to similar extents by the two cDNAs.

cDNAs to gene microarrays (Figure, and 9). Subtractive work allowed us to identify and functionally characterize previously unknown genes expressed in more advanced precursor cells. We are now addressing differences in patterns of gene expression between normal multipotent cells with long-term reconstitution potential and more advanced multipotent cells that lack long-term growth potential. We are also exploring expression differences between multipotent cells that have both myeloid and lymphoid potential, and multipotent cells that only have myeloid potential. Experiments with human leukemia cells similarly reveal significant differences in gene expression between samples differing in growth potential (Figure), and are directed at identifying key genes responsible for maintenance and extinction of proliferative potential in human leukemia. Novel genes have been identified in each of these projects and are in early stages of investigation of their nature and function using bioinformatics, structural, overexpression and gene knockout technologies.


Selected Publications

Click on the link above for a Pubmed search of our recent publications.

1. Analysis of gene expression in a complex differentiation hierarchy by global amplification of cDNA from single cells. G Brady, F Billia, J Knox, T Hoang, IR Kirsch, EB Voura, RG Hawley, R Cumming, M Buchwald, K Siminovitch, N Miyamoto, G Boehmelt, NN Iscove. Current Biology 5:909-922, 1995.

2. Cycle initiation and colony formation in culture by murine marrow cells with long-term reconstituting potential in vivo. M Trevisan, X-Q Yan, NN Iscove. Blood 88:4149-4158, 1996.

3. Hematopoietic stem cells expand during serial transplantation in vivo without apparent exhaustion. NN Iscove, K Nawa. Current Biology 7:805-808, 1997.

4. Expression of notch receptors, notch ligands, and fringe genes in hematopoiesis. N Singh, RA Phillips, NN Iscove, SE Egan. Exp Hematol 28:527-534, 2000.

5. An early take on "Stem Cell Plasticity" at a discussion convened at the NIH in March 2000 to identify relevant research issues.

6. Resolution of pluripotential intermediates in murine hematopoietic differentiation by global cDNA amplification from single cells: confirmation of assignments by expression profiling of cytokine receptor transcripts. F Billia, M Barbara, J McEwen, M Trevisan, NN Iscove. Blood 97:2257-2268, 2001.

7. Is plasticity here to stay? Inside Blood [Editorial]. NN Iscove. Blood 98:1999, 2001.

8. Hematopoietic competence is a rare property of neural stem cells that may depend on genetic and epigenetic alterations. CM Morshead, P Benveniste, D van der Kooy and NN Iscove. Nature Medicine 8:268-73, 2002.

9. Representation is faithfully preserved in global cDNA amplified exponentially from sub-picogram quantities of mRNA. NN Iscove, M Barbara, M Gu, M. Gibson, C. Modi, and N. Winegarden. Nature Biotechnology 20:940-943, 2002.

10. Hematopoietic stem cells engraft in mice with absolute efficiency. P Benveniste, C Cantin, D Hyam and NN Iscove. Nature Immunology 4:708-713, 2003.

11. In vitro and in vivo expansion of hematopoietic stem cells. G Sauvageau, NN Iscove and RK Humphries. Oncogene 23:7223-7232, 2004.

12. Identification of gene 3' ends by automated EST cluster analysis. E Muro, R Herrington, S Janmohamed, C Frelin, M Andrade-Navarro and NN Iscove. Proceedings of the National Academy of Sciences USA 105:20286-90 Dec 18, 2008.

13. Intermediate-term hematopoietic stem cells with extended but time-limited reconstitution potential. P Benveniste, C Frelin, S Janmohamed, M Barbara, R Herrington, D Hyam and NN Iscove. Cell Stem Cell 6:48-58, 2010.

14. Identification of a role for the nuclear receptor EAR-2 in the maintenance of clonogenic status within the leukemia cell hierarchy. CV Ichim, HL Atkins, NN Iscove and RA Wells. Leukemia (3 June 2011) | doi:10.1038/leu.2011.137, 2011