MATERIALS AND METHODS. Cells and cell separation INTRODUCTION. Li Lu,* Zhi-Hua Li,* Jie He,* Yue Ge,* Susan Rice, and Hal E.

Transcription

1 Transduction of recombinant human erythropoietin receptor cdna into daughter progenitors derived from single CD34 3 cord blood cells changes the differentiation profile of daughter progenitors Li Lu,* Zhi-Hua Li,* Jie He,* Yue Ge,* Susan Rice, and Hal E. Broxmeyer* Departments of *Microbiology and Immunology, and Medicine (Hematology/Oncology), the Walther Oncology Center, Indiana University School of Medicine, and the Walther Cancer Institute, Indianapolis Abstract: In this study, we tested the capacity to change the differentiation profile of progenitor cells by retroviral-mediated transduction of EpoR cdna into one of the paired daughter cells derived from single CD34 3 CB cells. Our results show that for the non-viral-treated daughter cells, the majority (99.6%) formed the same colony type. However, with cells transduced with viral vectors, 7.1% of the daughter cells transduced with the EpoR cdna formed either a burst forming unit-erythroid (BFU-E) or a colony-forming unit-granulocyte, macrophage, erythroid, megakaryocyte (CFU-GEMM) colony compared to the other daughter cell transduced with viral supernatant lacking EpoR cdna, which formed either a colony-forming unit granulocyte-macrophage (CFU-GM) or a high proliferative potential-colony forming cell (HPP-CFC) colony. Expression of the transduced EpoR cdna was confirmed in individual colonies by RT-PCR analysis. These results substantiate in a more rigorous fashion our previous results that it is possible to change the Epo-responsive differentiation profile of progenitor cells by transduction into these cells of an EpoR cdna and this change was apparent only in daughter cells derived from single CD34 3 kit cells transduced with EpoR cdna. J. Leukoc. Biol. 63: ; Key Words: viral vectors erythroid differentiation INTRODUCTION Hematopoiesis is regulated, at least in part, by hematopoietic growth factors. Erythropoietin (Epo), one such growth factor, is a major regulator of the proliferation and differentiation of erythroid progenitors/precursors [1]. Human and murine Epo receptors (R) have been cloned and identified on primary hematopoietic cells and established cultured cell lines [2, 3]. The EpoR is a member of the cytokine receptor superfamily and lacks a kinase domain, but binding of Epo to the EpoR induces rapid tyrosine phosphorylation of a series of cellular substrates that are involved in cell proliferation and differentiation [4 6]. We recently reported that transduction of CD34 3 cord blood (CB) cells with a retroviral vector encoding human (h) EpoR cdna, even at the single isolated cell level, changed the differentiation profile of stem/progenitor cells and resulted in increased numbers of Epo-dependent erythroid progenitor cells (BFU-E) [7, 8]. These studies, although done at the single transduced cell level, did not compare the growth and differentiations of daughter cells, which would allow a more rigorous evaluation of the capacity to change the differentiation profile of progenitor cells after insertion and expression of a gene for a growth factor receptor. Therefore, in this study, we have compared the growth and differentiation of single isolated daughter CB CD34 3 cells, one of which was transduced with EpoR cdna in the retroviral vector LhEpoRSN, and one that was transduced with a mock control vector (LXSN). We also evaluated CD34 3 c-kit 2, CD34 3 c-kit, and CD34 3 c-kit CB cells. The comparative growth and differentiation of these virally transduced daughter cells was also compared with the growth and differentiation of non-virally treated daughter cells. MATERIALS AND METHODS Cells and cell separation Cells were obtained from normal hcb scheduled for discard after delivery of the infant and after prior need for samples for clinical study had been satisfied. CD34 3 cells and their subsets (CD34 3 c-kit 2, CD34 3 c-kit, and CD34 3 c- kit ) were obtained after sorting nonadherent low-density T lymphocytedepleted cells on a Flow Cytometer (FacStar plus, Becton Dickinson, San Jose, CA) [7, 8]. These cells were directly sorted into wells with the use of an autoclone device as a single cell/well containing 50 µl culture medium containing recombinant (r) human steel factor (SLF; 50 ng/ml), human interleukin (IL)-3 (200 U/mL), human granulocyte-macrophage (GM) colonystimulating factor (CSF; 200 U/mL), and human Epo (1 U/mL) for prestimulation. SLF, IL-3, and GM-CSF were kind gifts of Immunex Corp. (Seattle, WA) and Epo was purchased from Amgen Corp. (Thousand Oaks, CA). Antibodies against CD34 and c-kit were purchased from Becton Dickinson and PharMingen (San Jose, CA), respectively. Abbreviations: CB, cord blood; Epo, erthropoietin; SLF, steel factor; IL, interleukin; GM-CSF, granulocyte-macrophage colony-stimulating factor; IMDM, Iscove s modified Dulbecco s medium; FBS, fetal bovine serum; RT-PCR, reverse transcriptase-polymerase chain reaction. Correspondence: Li Lu, M.D., Walther Oncology Center, Indiana University School of Medicine, 1044 West Walnut Street, Room 302, Indianapolis, IN Received July 8, 1997; revised November 11, 1997; accepted November 14, Journal of Leukocyte Biology Volume 63, March

2 Colony assay Semi-solid culture medium contained Iscove s modified Dulbecco s medium (IMDM, GIBCO-BRL, Gaithersburg, MD), 1% methylcellulose, 30% fetal bovine serum (FBS, Hyclone Laboratory, Logan, UT), and 0.1 mm hemin (Eastman Kodack Co., Rochester, NY) [9 11] and the same concentrations of growth factor used in the prestimulation studies noted above. Cells were incubated under humidified conditions at 37 C, 5% CO 2 and lowered (5%) O 2 for determination after 14 days of granulocyte-macrophage (CFU-GM), multipotential (CFU-GEMM), and BFU-E-colonies or after 21 days for high proliferative potential-colony forming cell (HPP-CFC) colonies. For the purposes of this study, we did not distinguish among CFU-GM colonies containing monocytes, macrophages, and/or different types of granulocytes. HPP-CFC colonies contain mainly or entirely cells of the macrophage lineage and have not in any of our previous studies [9] contained any erythroid cells, even after staining for erythroid cells. BFU-E colonies contained only hemoglobinized red cells, and CFU-GEMM colonies contained hemoglobinized red cells mixed with other cell types such as monocytes, macrophages, granulocytes, and to a lesser extent, megakaryocytes. Each colony was carefully visualized first at 40 and then at 100 magnification to make sure that colony types were accurately denoted. In our methylcellulose cultures, the red cell-containing colonies are clearly distinguishable as CFU-GEMM or BFU-E. CFU-GM and HPP-CFC colonies contained no noticeable red cells. Although we could not at this unstained cell microscopic level definitively rule out very early non-hemoglobinized cells of the erythroid lineage in some CFU-GM or HPP-CFC colonies, we consider this possibility unlikely because the cells in these colonies were morphologically characteristic of monocytes, macrophages, and granulocytes, and not of early erythroid cells. Retroviral vectors The retroviral vectors encoding hepor cdna (LhEpoRSN) and this vector lacking the hepor cdna (LXSN; as mock control) were used in the experiment as described previously [7, 8]. The titer for both viruses was to G418-resistant (G418 R ) particles/ml as assessed on NIH3T3 cells. Viral supernatant was harvested and filtered through and 0.22-µm filters and stored at 80 C until use. Retroviral transduction protocols After 2 days of prestimulation with cytokines, single CD34 3 CB cells or their subsets (CD34 3 c-kit 2, CD34 3 c-kit, and CD34 3 c-kit ) had divided into two daughter cells. One of two daughter cells was placed into another well containing the same culture medium. One of these daughter cells was transduced with 20 µl of retroviral supernatant containing LhEpoRSN plus polybrene (8 µg/ml) as described previously [7], whereas the other daughter cell was transduced with control LXSN viral supernatant plus polybrene. Another control consisted of separated daughter cells, neither of which were transduced with viral supernatant. The viability of cells transduced with either the hepor cdna or the control virus was 95% based on bulk cell gene transduction protocols. Also, total colony numbers forming from the transduced single cells were not significantly different from non-transduced cells. Thus, cell viability was not affected by gene transduction. The probability of significant differences between groups in the single cell evaluations was determined by 2 test [7, 8]. Reverse transcriptase-polymerase chain reaction (RT-PCR) analysis for gene expression RNA was extracted from individual colonies as described previously [7]. Individual colonies were removed and cell pellets were resuspended in 20 µl double-distilled water treated with dimethyl pyrocarbonate (DEPC), 100 U RNasin, and 1 µg trna, cooled in ice for 30 min, and pelleted for 30 s. A 9-µL supernatant of the resulting RNA solution was used in the reverse transcription reaction. To eliminate contaminating DNA, the RNA extracts were treated with one unit DNase I for 15 min at room temperature. The DNase was inactivated by adding 1 µl of 20 mm EDTA solution to the reaction mixture and heating at 65 C for 10 min before reverse transcription. A total of 20 µl of the RNA lysate in RT buffer [RT buffer: 25 mm Tris-HCl (ph 8.3), 37.5 mm KCl, 1.5 mm MgCl, 5 mm dithiothreitol, 10 mm dntp mixture, 0.5 µg oligo(dt)15 primer, 26 U RNasin, and 200 U murine Molony leukemia virus reverse transcriptase (Promega Corp., Madison, WI)] was used for reverse transcription. After cdna synthesis, 10 µl of the cdna solution was used for the RT-PCR reaction. The primers for RT-PCR reaction were as follows: 5 CCCCTACCCAC- CCCACCTAA3 of hepor cdna and 5 ACCTGCGTGCAATCCATCTTG3 corresponding to the neo gene of the plxsn vector. These primers thus allow amplification of only the transduced hepor gene, and not the endogenous EpoR. Each sample was amplified with AmpliTaq (Perkin-Elmer/Cetus, Roche Molecular System, Inc., Branchburg, NJ) using a DNA thermocycler (Perkin- Elmer/Cetus) for 45 cycles, including denaturation at 95 C for 25 s, annealing at 65 C for 30 s, and polymerization at 72 C for 2 min as described [7]. As a negative control for the RT-PCR reaction, the RNA extracts were used for PCR analysis without adding reverse transcriptase in order to ensure that the amplified products were mrna but not genomic DNA, which might contain integrated viral cdna. Ten microliters of reaction mixture was electrophoresed on a 1% agarose gel. RESULTS Influence of transduction of hepor cdna into a daughter progenitor derived from a single CD34 3 CB cell on the growth/differentiation of that cell To determine whether transduction of hepor cdna into one daughter of a pair of cells dividing from a single CD34 3 cell would change the differentiation profile of that cell compared to the other daughter cell transduced with a control virus (same virus but lacking hepor cdna), we utilized the LhEpoRSN virus and a mock control virus LXSN. Results from 878 paired daughter cells transduced with either hepor cdna or a control virus and 250 paired daughter cells with no viral treatment from a total of six experiments were pooled and summarized in Figure 1. As shown in this figure, although most daughter cells formed the same type of colony under the growth stimulation conditions supplied by Epo in combination with SLF, GM-CSF, and IL-3 (Fig. 1A), 7.1% of the daughter cells transduced with LhEpoRSN formed a BFU-E or CFU-GEMM colony, whereas the mock LXSN-transduced daughter cells formed either a CFU-GM or HPP-CFC colony (P for the pooled data; Fig. 1B). These differences were significant to at least P 0.05 for each of five of the individual experiments with changes in colony types ranging from 5.1 to 11.7%. One of the individual experiments did not show a significant difference with a 3.0% change in colony type. We did not detect a significant change from a BFU-E or CFU-GEMM colony to a CFU-GM colony on transducing the daughter cell with LhEpoRSN. As an added control, we evaluated the differentiation profile of daughter cells that were not transduced (non-viral control in Fig. 1). In greater than 99.5% of the cases, the same colony type that grew from one daughter cell also grew from the other daughter cell. Morphological examples of different colony types formed from two pairs of daughter cells in which one daughter cell was transduced with LhEpoRSN and the other with LXSN are shown in Figure 2. The transduction efficiency of CB myeloid progenitors by retroviral vectors is high, especially when this is done at the single cell level. Overall, mrna expression was detected in 65% (66/103) of colonies from LhEpoRSNtransduced cells. Figure 3 shows a sampling of the mrna expression of the inserted EpoR gene in cells from colonies in which the differentiation profile was changed. Transduced EpoR gene expression was detected only in colony cells 390 Journal of Leukocyte Biology Volume 63, March 1998

3 Fig. 1. Influence of transduction of hepor cdna on colony formation by single separated daughter pair progenitors derived from single CD34 3 CB cells. Separated daughter cells (878 pairs) were transduced with either hepor cdna (LhEpoRSN) or a control virus (LXSN; mock) or neither daughter cells (250 pairs) were treated with virus (non-viral control). These results are from a total of six separate experiments. Values in parentheses designate the percentage of wells in which a colony formed. For the virus-treated cells the colony type to the left of the arrow describes growth from the mock virus-treated cell and the colony type to the right of the arrow describes the colony type forming from the cell treated with LhEpoRSN. For the non-virus-treated separated daughter cells, the colony type to the left of the arrow reflects growth of one daughter cell, whereas that to the right of the arrow reflects growth of the other daughter cell. deriving from LhEpoRSN-, not in LXSN-transduced daughter cells. No colonies grew in the absence of growth factors and no erythroid cell-containing colonies grew in the absence of Epo (data not shown). These results confirm at a more rigorous daughter cell level our previous observations that insertion of the EpoR gene into single progenitor cells allows detection of increased numbers of Epo-dependent BFU-E and CFU-GEMM, Fig. 2. Examples of colony formation by two pairs of separated progenitor cells derived from CD34 3 cells transduced with LXSN mock (A, C) or LhEpoRSN (B, D) virus. (A and B) Growth from one pair of daughter cells; (C and D) growth of another pair of daughter cells. Colony types were as follows: (A) CFU-GM, (B) BFU-E, (C) HPP-CFC, and (D) CFU-GEMM. Lu et al. Epo receptor, gene transduction, hematopoietic daughter progenitors 391

4 Fig. 3. Examples of expression of viral EpoR cdna by RT-PCR analysis of individual colonies formed from daughter cell transduced with (A) LXSN mock control or (B) LhEpoRSN. Each number reflects RT-PCR analysis of daughter pair-derived colonies. Examples in (B) are from CFU-GM, HPP-CFC, or CFU-GEMM. LhEpoRSN plasmid was used as a positive control and the mixture of PCR reagents was used as a negative control. The RT-PCR reaction generates a 625-bp fragment. whose normal growth/differentiation pattern would have been either that of a CFU-GM or more immature (HPP-CFC) progenitor cell [7]. Influence of transduction of hepor cdna into a daughter progenitor derived from a single CD34 3 c-kit 2, CD34 3 c-kit,or CD34 3 c-kit CB cell CD34 3 cells can be subdivided into c-kit 2 cells, which are enriched for CFU-GEMM, and c-kit cells, which are enriched for CFU-GM [12, 13]. To determine whether insertion of the EpoR gene into certain subsets of CD34 cells could also change the differentiation profile of daughter cells, the same type of experiments done with CD34 3 cells (Fig. 1) were done for the CD34 3 c-kit 2, CD34 3 c-kit, and CD34 3 c-kit subsets of cells. As shown in Table 1, CD34 3 c-kit 2 cells contained mainly CFU-GEMM ( %) with the majority of the remaining progenitors being CFU-GM (21.1%). Few HPP-CFC ( %) progenitors were detected. The CD34 3 c- kit population contained mainly CFU-GM ( %). The third subset of cells, CD34 3 kit cells, contained different types of progenitors, including HPP-CFC ( %) with the remainder of progenitors being CFU-GM ( %) and CFU-GEMM ( %). For both the c-kit 2 and c-kit populations of cells, greater than 99.5% of the daughter cells formed the same colony type whether one daughter cell was transduced with LhEpoRSN and the other with LXSN or neither daughter cell was transduced. In contrast, 10.5% of the daughter cells from the CD34 3 c-kit subset transduced with LhEpoRSN formed a CFU-GEMM colony, whereas the mock LXSN-transduced daughter cells formed a CFU-GM colony (P 0.05). RT-PCR analysis of daughter cells demonstrated that 67.5% (56/83) of the colonies deriving from a daughter cell of a CD34 3 subset cell transduced with LhEpoRSN, but not with LXSN, expressed the virally transduced EpoR gene (data not shown). This suggests that not all CD34 3 cell populations, in this case CD34 3 c-kit 2 or CD34 3 c-kit cells, have their differentiation profile changed on insertion and expression of an EpoR gene. TABLE 1. Influence of hepor cdna on Colony Formation of Daughter Progenitors Derived from Single CD34 3 kit 2, CD34 3 kit and CD34 3 kit CB Cells Colony type Percent colonies CD34 3 c-kit 2 CD34 3 c-kit CD34 3 c-kit Transduced Non-transduced Transduced Non-transduced Transduced Non-transduced Same type of colony formed from daughter cells HPP-CFC to HPP-CFC CFU-GEMM to CFU-GEMM BFU-E to BFU-E CFU-GM to CFU-GM BFU-E to CFU-GEMM or CFU-GEMM to BFU-E HPP-CFC to CFU-GM or CFU-GM to HPP-CFC Change of colony phenotypes CFU-GM to CFU-GEMM CFU-GEMM to CFU-GM Results are shown for evaluation of CD34 3 c-kit 2, CD34 3 c-kit and CD34 3 c-kit cells in which respectively 479, 257, and 216 daughter pairs were transduced and 71, 63, and 52 daughter pairs were not transduced. The results are from a total of three experiments for CD34 3 kit 3 and CD34 3 kit cells and of two experiments for CD34 3 kit cells. 392 Journal of Leukocyte Biology Volume 63, March 1998

5 DISCUSSION The differentiation profile of hematopoietic stem and progenitor cells is intimately related to the presence of receptors on the cell surface for different cytokines. Although the differentiation profiles of immature cells is, under certain circumstances, believed to be a stochastic process [14, 15], it is possible to manipulate the system to direct these immature cells to a specific lineage. In this context, we have demonstrated at both the population and single cell level, that transduction of CD34 3 CB cells, highly enriched for stem and immature subsets of myeloid progenitor cells, with an EpoR gene by retroviral-mediated gene transduction can change the differentiation profile of these immature cells with enhanced detection of numbers of erythroid cell-containing Epo-dependent colonies [7, 8]. These studies, however, did not prove that a single cell can give rise to two daughter cells, one of whose differentiation profile could be changed by insertion and subsequent expression of a cytokine gene. This study has substantiated that this can occur; overall, 7.1% of CD34 3 daughter cells (with a range of 3.0 to 11.7% in individual experiments) transduced at the single level with LhEpoRSN formed BFU-E or CFU-GEMM colonies, whereas their counterpart daughter cells transduced with the control vector LXSN formed a CFU-GM or HPP-CFC colony. Although it is clear that only those colonies in which the differentiation profile was changed expressed the inserted EpoR gene as assessed by RT-PCR analysis, the actual amount of EpoR receptor protein on the surface of the transduced cells could not be determined. As noted previously [7], flow analysis of EpoR expression on the cell surface of total populations of transduced CD34 3 cells did not detect any EpoR. It is likely that the ability of the EpoR-transduced cells to respond to Epo with differentiation toward CFU-GEMM or BFU-E reflects, at least in part, the amount of receptor on the cell surface and not all transduced cells may express the same amount of EpoR mrna or the different cell populations may not process this protein to the cell surface in the same quantitative fashion. The level of hepor displayed on the cell surface is apparently lower than that which can be detected by flow cytometry analysis using an anti-hepor monoclonal antibody. The percentage of symmetric daughter cell divisions ( 99% in which the same colony type formed from the daughter cells) in our studies, using purified hcb CD34 3 cells that were either not transduced with viral vector or transduced with the control LXSN vector, is higher than that noted by the studies of others using mouse cells [16]. In that study the differentiation profile of daughter cells derived from blast cell colonies forming from spleen cells of five fluorouracil (5FU)-treated mice demonstrated about 17.6% asymmetric divisions (different colony types forming from daughter cells) [16]. Because all of our colonies were carefully evaluated at high magnification for morphological cell types, we consider it highly unlikely that we missed significant numbers of asymmetric divisions in which a few erythroid cells might have been present in colonies we designated as CFU-GM or HPP-CFC. We consider it more likely that the differences in symmetric vs. asymmetric divisions noted between our study and that of the other group [16] may in part reflect differences in species (human vs. mouse), organ (CB vs. spleen), and/or maturational or functional state of the cells (primary CB cells from untreated individuals that were purified to a CD34 3 population vs. cells from blast colonies derived from relatively unseparated cells of 5FU-treated mice). Our previous [7, 8] and present studies demonstrating enhanced erythroid differentiation of cells transduced with EpoR cdna are different from those of others in which insertion of EpoR cdna into cells from blast colonies formed from spleen cells of mice given 5FU [17] or bone marrow cells of 5FU-treated mice [18], had enhanced proliferative but not differentiation responses to Epo. Differences between our studies and the other studies [17, 18] may reflect the species and cell types used, as discussed above for spontaneous symmetrical vs. asymmetrical divisions noted with non EpoR gene-transduced cells. It is possible that the primary cord blood cells used in our study are of a more immature functional type able to respond to differentiation changes more easily. Primitive CB cells appear to be enhanced in their proliferative responsiveness to growth factor stimulation compared with adult cells [19] and it is possible that they may also be enhanced in their capacity to respond to differentiation stimuli. That not all EpoR gene-transduced cells can be induced to enhanced erythroid lineage differentiation is apparent from the information presented in Table 1. Our studies with CD34 3 c- kit 2 and CD34 3 c-kit cells suggest that not all CD34 3 cell subsets are as amenable to EpoR gene-transduced, Epodependent cell-induced multipotential, or erythroid differentiation. The CD34 3 c-kit 2 population is mainly composed of CFU-GEMM, and consistent with the results of the CD34 3 cells shown in Figure 1, we did not detect a change of CFU-GEMM to CFU-GM colonies (Table 1). The fact that the remainder of the progenitors in the CD34 3 c-kit 2 population that were CFU-GM, and the progenitors in the CD34 3 c-kit population, which were % CFU-GM, did not form BFU-E- or CFU-GEMM-colonies when the daughter cell was transduced with LhEpoRSN suggests that these CFU-GM may be different from those found in the total CD34 3 cell population or more specifically in the CD34 3 c-kit subset. Perhaps these CD34 3 c-kit 2 or c-kit cells are later more mature subsets of CFU-GM in which the differentiation profile cannot be changed, either because the inserted EpoR gene product cannot be processed to the cell surface in sufficient quantity (note that mrna for this gene was found by RT-PCR in the LhEpoRSN-transduced cells) or activation of EpoR in these cells is not sufficient for erythroid differentiation. These cells may lack the appropriate cascade of intracellular signaling molecules necessary for Epo-induced cell differentiation. Unraveling the complexities of these events, although beyond the scope of this study, could provide important insights into cytokine-induced differentiation of stem and progenitor cells. ACKNOWLEDGMENTS We thank Cynthia Booth, Rebecca Miller, and Linda Cheung for secretarial assistance. This study was supported by U.S. Public Health Service Grants R01 HL54037, R01 HL46416, R01 DK53674, and T32 Lu et al. Epo receptor, gene transduction, hematopoietic daughter progenitors 393

6 DK07519, a project in R01 HL53586 to H. E. B., and a grant from Phi Beta Psi Sorority to L. L. REFERENCES 1. Koury, M. J., Bondurant, M. C. (1992) The molecular mechanism of erythropoietin action. Eur. J. Biochem. 210, Broudy, V. C., Lin, N., Brice, M., Nakamoto, B., Papayannopoulou, T. (1991) Erythropietin receptor charateristics on primary human erythroid cells. Blood 77, Kitamura, T., Tojo, A., Kuwaki, T., Chiba, S., Miyazono, K., Urabe, A., Takaku, F. (1989) Identification and analysis of human erythropoietin receptors on a factor-dependent cell line, TF-1. Blood 73, Witthuhn, B. A., Quelle, F. W., Silvennoinen, O., Yi, T., Tang, B., Miura, O., Ihle, J. N. (1993) JAK2 associates with the erythropoietin receptor and is tyrosine phosphorylated and activated following stimulation with erythropoietin. Cell 74, Tauchi, T., Feng, G-S., Shen, R., Hoatin, M., Bagby, G. C., Kabat, D., Lu, L., Broxmeyer, H. E. (1995) Involvement of SH2-containing phosphotyrosine phosphatase SYP in erythropoietin receptor signal transduction pathways. J. Biol. Chem. 270, Tauchi, T., Damen, J. E., Toyama, K., Feng, G-S., Broxmeyer, H. E., Krystal, G. (1996) Tyrosine 425 within the activated erythropoietin receptor binds Syp, reduces the erythropoietin requried for Syp tyrosine phosphorylation, and promotes mitogenesis. Blood 87, Lu, L., Ge, Y., Li, Z-H., Keeble, W., Kabat, D., Bagby, G. C., Broxmeyer, H. E., Hoatin, M. E. (1996) Retroviral transfer of the recombinant human erythropoietin receptor gene into single hematopoietic stem/progenitor cells from human cord blood increases the number of erythropoietindependent erythroid colonies. Blood 87, Lu, L., Ge, Y., Li, Z-H., Xiao, M., Broxmeyer, H. E. (1996) Influence of retroviral-mediated gene transduction of both the recombinant human erythropoietin receptor and interleukin-9 receptor genes into single CD34 CD33 or low cord blood cells on cytokine-stimulated erythroid colony formation. Exp. Hematol. 24, Lu, L., Xiao, M., Shen, R-N., Grigsby, S., Broxmeyer, H. E. (1993) Enrichment, characterization and responsiveness of single primitive CD34 3 human umbilical cord blood hematopoietic progenitors with high proliferative and replating potential. Blood 81, Xiao, M., Broxmeyer, H. E., Horie, M., Grigsby, S., Lu, L. (1994) Extensive proliferative capacity of single isolated CD34 3 human cord blood cells in suspension culture. Blood Cells 20, Lu, L., Xiao, M., Clapp, D. W., Li, Z-H., Broxmeyer, H. E. (1993) High efficiency retroviral mediated gene transduction into single isolated immature and replatable CD34 3 hematopoietic stem/progenitor cells from human umbilical cord blood. J. Exp. Med. 178, Papayannopoulou, T., Brice, M., Broudy, V. C., Zsebo, K. M. (1991) Isolation of c-kit receptor-expression cells from bone marrow peripheral blood, and fetal liver: Functional properties and composite antigenic profile. Blood 78, Ashman, L. K., Cambareri, A. C., Bikto, L., Levinsky, R. T., Juttner, C. A. (1991) Expression of the YB5.B8 antigen (c-kit proto-oncogene product) in normal human bone marrow. Blood 78, Till, J. E., McCulloch, E. A., Siminovitch, L. (1964) A stochastic model of stem cell proliferation, based on the growth of spleen colony-forming cells. Proc. Natl. Acad. Sci. USA 51, Nakahata, T., Gross, A. J., Ogawa, M. (1982) A stochastic model of self-renewal and commitment to differentiation of the primitive hemopoietic stem cells in culture. J. Cell Physiol. 113, Suda, T., Suda, J., Ogawa, M. (1984). Disparate differentiation in mouse hemopoietic colonies derived from paired progenitors. Proc. Natl. Acad. Sci. USA 81, Pharr, P. N., Ogawa, M., Hofbauer, A., Longmore, D. (1994) Expression of an activated erythropoietin or a colony-stimulating factor 1 receptor by pluripotent progenitors enhances colony formation but does not induce differentiation. Proc. Natl. Acad. Sci. USA 91, Goncalves, F., Lacout, C., Dubart, A., Wendling, F., Vainchenker, W., Dumenil, D. (1997) Constitutive expression of MPL-R or R-Epo in pluripotent progenitors allows them to proliferate in response to Tpo or Epo but does not modify the orientation of their differentiation. Exp. Hematol. 25, 835a. 19. Broxmeyer, H. E. (1995) Cord blood, an alternative source for stem and progenitor cell transplantation. Curr. Op. Pediat. Hematol. Oncol. 7, Journal of Leukocyte Biology Volume 63, March 1998