Search

The ERCC1-XPF heterodimer is a structure-specific endonuclease involved in both nucleotide excision repair and interstrand crosslink repair. Mice carrying a genetic defect in Ercc1 display symptoms suggestive of a progressive, segmental progeria, indicating that disruption of one or both of these DNA damage repair pathways accelerates aging. In the hematopoietic system, there are defined age-associated changes for which the cause is unknown. To determine if DNA repair is critical to prolonged hematopoietic function, hematopoiesis in Ercc1-/- mice was compared to that in young and old wild-type mice. Ercc1-/- mice (3-week-old) exhibited multilineage cytopenia and fatty replacement of bone marrow, similar to old wild-type mice. In addition, the proliferative reserves of hematopoietic progenitors and stress erythropoiesis were significantly reduced in Ercc1-/- mice compared to age-matched controls. These features were not seen in nucleotide excision repair-deficient Xpa-/- mice, but are characteristic of Fanconi anemia, a human cancer syndrome caused by defects in interstrand crosslink repair. These data support the hypothesis that spontaneous interstrand crosslink damage contributes to the functional decline of the hematopoietic system associated with aging.

Recently it was shown that, analogous to normal hematopoietic cells, the level of CXC chemokine receptor 4 (CXCR-4) expression on acute myeloid leukemia (AML) cells correlates with stromal cell derived factor-1 alpha (SDF-1)-induced chemotaxis. As we speculated that an anomalous organ distribution of AML cells could affect cell survival and thus result in an altered fraction surviving chemotherapy, we examined a possible correlation between patient prognosis and CXCR-4 expression in AML patients. We found that patients with a high CXCR-4 expression in the CD34(+) subset had a significantly reduced survival and a higher probability of relapse, resulting in a median relapse-free survival (RFS) of only 8.3 months. CXCR-4 expression was significantly higher in fetal liver tyrosine kinase-3 (Flt3)/internal tandem duplication (ITD) AML than in Flt3/wild-type (wt) AML. Covariate analysis indicated that the prognostic significance of Flt3/ITDs with respect to RFS was no more apparent when analyzed in conjunction with the expression of CXCR-4 in the CD34(+) subset, suggesting that the poor prognosis of Flt3/ITD AML might be subordinate to the increased CXCR-4 expression. Using a granulocyte colony-stimulating factor receptor (G-CSF-R)-expressing 32D cell line, we observed that SDF-1/CXCR-4 interaction is required for the survival of myeloid differentiating cells, and it also induces a block in G-CSF-induced myeloid differentiation. These data suggest that the SDF-1/CXCR-4 axis may influence therapy responsiveness and defines unfavorable prognosis in AML.

Treosulfan (L-threitol-1,4-bismethanesulfonate) is an alkylating agent with routine clinical application in the treatment of ovarian cancer. In this murine study we show that this drug also has the ability to deplete primitive hematopoietic stem cells in a dose-dependent manner as determined by the cobblestone area-forming cell assay and is similar to its parent compound busulfan. Because busulfan is frequently used as part of the conditioning regimen before stem cell transplantation, we investigated an alternative nonmyeloablative protocol in an allogeneic bone marrow transplantation model in which low-dose treosulfan was added to an immune-suppressive regimen consisting of T cell-depleting antibodies, fludarabine, and thymic irradiation. Although this treatment protocol produced minimal myelosuppression, the addition of treosulfan proved to be important for allowing stable multilineage and mixed chimerism in C57BL/6 recipients of major histocompatibility complex-mismatched B10.A bone marrow without evidence of graft-versus-host disease. Donor lymphocyte infusion performed at 10 weeks after bone marrow transplantation had the effect of transforming the state of mixed chimerism to full donor-type cells, again without evidence of graft-versus-host disease. Donor-specific immunologic tolerance in the mixed chimeric animals was indicated by the acceptance of donor-type and rejection of third-party skin grafts. Thus, low-dose treosulfan may be considered as a useful component of a truly nonmyeloablative conditioning protocol in providing for mixed hematopoietic chimerism and, consequently, in establishing a platform for adoptive immunotherapy.

The stromal cell-derived factor-1 (SDF-1)/CXC chemokine receptor 4 (CXCR4) signaling pathway is thought to play an important role in the induction of neutrophil mobilization from the bone marrow in response to granulocyte-colony stimulating factor (G-CSF) treatment. CXCR4 belongs to the family of G protein-coupled receptors. Multiple members of this receptor family are desensitized by agonist-induced G protein-coupled receptor kinase (GRK)-mediated phosphorylation. Here, we demonstrate that in vitro SDF-1-induced chemotaxis of bone marrow-derived neutrophils from GRK6-deficient mice is significantly enhanced and that desensitization of the calcium response to SDF-1 is impaired in GRK6-/- neutrophils. CXCR4 activation by SDF-1 provides a key retention signal for hematopoietic cells in the bone marrow. It is interesting that we observed that in the absence of GRK6, the G-CSF-induced increase in circulating neutrophils is profoundly impaired. Three days after injection of pegylated-G-CSF, significantly lower numbers of circulating neutrophils were observed in GRK6-/- as compared with wild-type (WT) mice. In addition, early/acute neutrophil mobilization in response to G-CSF (3 h after treatment) was also impaired in GRK6-/- mice. However, blood neutrophil levels in untreated GRK6-/- and WT mice were not different. Moreover, the percentage of neutrophils in the bone marrow after G-CSF treatment was increased to the same extent in WT and GRK6-/- mice, indicating that neutrophil production is normal in the absence of GRK6. However, the increased chemotactic sensitivity of GRK6-/- neutrophils to SDF-1 was retained after G-CSF treatment. In view of these data, we suggest that the impaired G-CSF-induced neutrophil mobilization in the absence of GRK6 may be a result of enhanced CXCR4-mediated retention of PMN in the bone marrow.

Recent studies describe beneficial effects of bone marrow-derived mesenchymal stem cell infusion in animal models as well as in patients. However, data on the homing abilities of primary and culture-expanded MSC are lacking. In order to systematically investigate MSC homing we compared the fate of both primary and cultured MSC in a syngeneic mouse model. Twenty-four hours after transplantation of uncultured EGFP-transgenic MSC into sublethally irradiated mice, as many as 55-65% of injected CFU-F were recovered from the BM and 3.5-7% from the spleen. In the subsequent 4 weeks these donor CFU-F expanded 100-fold, which resulted in a normalization of femoral and splenic CFU-F numbers. This highly efficient homing of primary CFU-F contrasted with the defective homing of MSC following culture. Following their infusion immortalized multipotent syngeneic stromal cells were undetectable in BM, spleen, lymph nodes or thymus. Remarkably, following transplantation of primary MSC that had been cultured for only 24 h the seeding fraction in the BM was reduced to 10%, while after transplantation of 48 h cultured primary MSC no CFU-F were detected in the lymphohematopoietic organs. These data suggest that in vitro propagation of BM-derived MSC dramatically decreases their homing to BM and spleen.

Throughout life, the hematopoietic system requires a supportive microenvironment that allows for the maintenance and differentiation of hematopoietic stem cells (HSC). To understand the cellular interactions and molecules that provide these functions, investigators have previously established stromal cell lines from the late gestational stage and adult murine hematopoietic microenvironments. However, the stromal cell microenvironment that supports the emergence, expansion and maintenance of HSCs during mid-gestational stages has been largely unexplored. Since several tissues within the mouse embryo are known to harbor HSCs (i.e. aortagonads-mesonephros, yolk sac, liver), we generated numerous stromal cell clones from these mid-gestational sites. Owing to the limited cell numbers, isolations were performed with tissues from transgenic embryos containing the ts SV40 Tag gene (tsA58) under the transcriptional control of constitutive and ubiquitously expressing promoters. We report here that the growth and cloning efficiency of embryonic cells (with the exception of the aorta) is increased in the presence of the tsA58 transgene. Furthermore, our results show that the large panel of stromal clones isolated from the different embryonal subregions exhibit heterogeneity in their ability to promote murine and human hematopoietic differentiation. Despite our findings of heterogeneity in hematopoietic growth factor gene expression profiles, high-level expression of some factors may influence hematopoietic differentiation. Interestingly, a few of these stromal clones express a recently described chordin-like protein, which is an inhibitor of bone morphogenic proteins and is preferentially expressed in cells of the mesenchymal lineage.

The aorta-gonads-mesonephros (AGM) region autonomously generates the first adult repopulating hematopoietic stem cells (HSCs) in the mouse embryo. HSC activity is initially localized to the dorsal aorta and mesenchyme (AM) and vitelline and umbilical arteries. Thereafter, HSC activity is found in the urogenital ridges (UGs), yolk sac, and liver. As increasing numbers of HSCs are generated, it is thought that these sites provide supportive microenvironments in which HSCs are harbored until the bone marrow microenvironment is established. However, little is known about the supportive cells within these midgestational sites, and particularly which microenvironment is most supportive for HSC growth and maintenance. Thus, to better understand the cells and molecules involved in hematopoietic support in the midgestation embryo, more than 100 stromal cell lines and clones were established from these sites. Numerous stromal clones were found to maintain hematopoietic progenitors and HSCs to a similar degree as, or better than, previously described murine stromal lines. Both the AM and UG subregions of the AGM produced many supportive clones, with the most highly HSC-supportive clone being derived from the UGs. Interestingly, the liver at this stage yielded only few supportive stromal clones. These results strongly suggest that during midgestation, not only the AM but also the UG subregion provides a potent microenvironment for growth and maintenance of the first HSCs.

Objective: Busulfan (BU) is often used in conditioning regimens prior to bone marrow transplantation, but its mechanism of action remains to be resolved. We have examined the possibility that BU may exert part of its toxic effects via DNA alkylation at the O6 position of guanine as this might provide an approach to improving the conditioning regimen., Methods: Survival of LAMA-84 and RJKO cells was assessed by colony-forming assay and cell counting, respectively. O6-alkylguanine-DNA alkyltransferase (ATase) activity was assayed by transfer of radioactivity from [3H]-methylated DNA. Colony-forming potential of normal human bone marrow cells (BMC) was measured in the presence of appropriate growth factors as the formation of both granulocyte-macrophage colony-forming units (CFU-GM) or burst-forming unit erythroids (BFU-E) within the same assay. Murine hematopoietic precursors were grown under a bone marrow stromal cell line to allow measurement of the frequency of cobblestone area-forming cells (CAFC) that correspond to CFU-GM, spleen colony-forming units (CFU-S), and the primitive stem cells with long-term repopulating ability., Results: Inactivation of ATase by O6-benzylguanine (O6-BeG) sensitized a human erythromegakaryocytic cell line (LAMA-84) and normal human bone marrow progenitors to 1,3-bis(2-chloroethyl)-1-nitrosourea (BCNU) but not to BU toxicity. BCNU, but not BU, inactivated ATase in LAMA-84 cells. Overexpression of human ATase in cDNA transfected Chinese hamster cells attenuated the toxicity of BCNU but not BU. Finally, the in vivo treatment of mice showed that the depletion of primitive stem cells by BU as measured in the CAFC assay was not affected by addition of O6-BeG. O6-BeG did, however, dramatically potentiate BCNU toxicity in all CAFC subsets, leading to depletion of more than 99% stem cells., Conclusion: These data suggest that BU does not elicit toxicity via alkylation at the O6 position of guanine in DNA in a way that can be influenced by ATase modulation.

Busulfan (1,4-butanediol dimethanesulfonate, BU) is relatively unique among other standard chemotherapy compounds in its ability to deplete noncycling primitive stem cells in the host and consequently to allow for high levels of long-term, donor-type engraftment after bone marrow transplantation (BMT). Such a property explains why this drug can be used as an alternative to total body irradiation in preparative regimes for BMT. However, as with radiation, BU conditioning is still troubled by severe toxicities that limit its applications to suboptimal drug doses. These problems stress the need for other BMT-conditioning drugs that are better tolerated and more selectively targeted toward normal and malignant hematopoietic stem cells. We have therefore compared the effects of various novel dimethanesulfonate compounds (related to BU) in terms of their toxicity to different stem cell subsets in vivo and in vitro and their ability to provide for long-term donor bone marrow engraftment using the congenic glucose-6-phosphate isomerase type 1 marker. Introduction of a benzene or cyclohexane ring in some of these drugs affords rigidity to the molecule and restricts the spatial positioning of the alkylating groups. Among 25 different compounds thus far tested at single doses, PL63 [cis-1,2-(2-hydroxyethyl) cyclohexane dimethanesulfonate] proved to be the most effective in providing for hematopoietic engraftment. The transisomer of the same compound gave significantly less engraftment and was comparable with the effects of dimethylbusulfan and Hepsulfam. The engraftment data correlated well with the depletion of different bone marrow stem cell subsets in the host as measured using the cobblestone area forming cell assay. The extent of stem cell depletion could not be explained on the basis of the distance and orientation of the two alkylating groups. Pharmacokinetic data, however, indicate that there is a correlation between biological activity and plasma levels reached. The diverse cytotoxic effects shown by these novel analogues of BU have provided a basis for relating biological activity with pharmacokinetic properties rather than with structural properties such as distance and orientation of the two alkylating groups. The identification of highly active compounds such as PL63 offers an opportunity for further developing other closely related drugs for potential application in clinical BMT conditioning therapy.

Gap junctions (GJs) provide for a unique system of intercellular communication (IC) allowing rapid transport of small molecules from cell to cell. GJs are formed by a large family of proteins named connexins (Cxs). Cx43 has been considered as the predominantly expressed Cx by hematopoietic-supporting stroma. To investigate the role of the Cx family in hemopoiesis, we analyzed the expression of 11 different Cx species in different stromal cell lines derived from murine bone marrow (BM) or fetal liver (FL). We found that up to 5 Cxs are expressed in FL stromal cells (Cx43, Cx45, Cx30.3, Cx31, and Cx31.1), whereas only Cx43, Cx45, and Cx31 were clearly detectable in BM stromal cells. In vivo, the Cx43-deficient 14.5- to 15-day FL cobblestone area-forming cells (CAFC)-week 1-4 and colony-forming unit contents were 26%-38% and 39%-47% lower than in their wild-type counterparts, respectively. The reintroduction of the Cx43 gene into Cx43-deficient FL stromal cells was able to restore their diminished IC to the level of the wild-type FL stromal cells. In addition, these Cx43-reintroduced stromal cells showed an increased support ability (3.7-fold) for CAFC-week 1 in normal mouse BM and 5-fold higher supportive ability for CAFC-week 4 in 5-fluorouracil-treated BM cells as compared with Cx43-deficient FL stromal cells. These findings suggest that stromal Cx43-mediated IC, although not responsible for all GJ-mediated IC of stromal cells, plays a role in the supportive ability for hemopoietic progenitors and stem cells. (Blood. 2000;96:498-505)

Nonobese diabetic/severe combined immune deficiency (NOD/SCID) mouse repopulating cells (SRC) have been proposed to represent a more primitive human stem cell subset than the cobblestone area-forming cell (CAFC) week (wk) 6 or the long-term culture-initiating cell (LTC-IC) wk 5 on the basis of their difference in frequency, phenotype, transfectibility, and multilineage outgrowth potential in immunodeficient recipients. We have assessed the percentage of various progenitor cell populations (colony-forming cell [CFC] and CAFC subsets) contained in unsorted NOD/SCID BM nucleated cells (nc), human umbilical cord blood (UCB) nc, bone marrow (BM) nc, peripheral blood stem cells (PBSC), and CD34(+) selected UCB nc, seeding in the BM and spleen of NOD/SCID mice within 24 hours after transplantation. The seeding efficiency of NOD/SCID BM CAFC wk 5 was median (range) in the spleen 2.9% (0.7% to 4.0%) and in the total BM 8.7% (2.0% to 9.2%). For human unsorted UCB nc, BM nc, PBSC, and CD34(+) UCB cells, the seeding efficiency for CAFC wk 6 in the BM of NOD/SCID mice was 4.4% (3.5% to 6.3%), 0.8% (0.3% to 1.7%), 5.3% (1. 4% to 13.6%), and 4.4% (3.5% to 6.3%), respectively. Using flow cytometry, the percentage CD34(+) UCB cells retrieved from the BM of sublethally or supralethally irradiated NOD/SCID mice was 2.3 (1.4 to 2.8) and 2.5 (1.6 to 2.7), respectively. Because we did not observe any significant differences in the seeding efficiencies of the various stem cell subsets, it may be assumed that the SRC seeding efficiency in NOD/SCID mice is similarly low. Our data indicate that the seeding efficiency of a graft can be of great influence when assessing stem cell frequencies in in vivo repopulation assays.

In approximately 20% of cases of severe congenital neutropenia (SCN), mutations are found in the gene encoding the granulocyte colony-stimulating factor receptor (G-CSF-R). These mutations introduce premature stop codons, which result in truncation of 82-98 COOH-terminal amino acids of the receptor. SCN patients who develop secondary myelodysplastic syndrome and acute myeloid leukemia almost invariably acquired a GCSFR mutation, suggesting that this genetic alteration represents a key step in leukemogenesis. Here we show that an equivalent mutation targeted in mice (gcsfr-Delta715) results in the selective expansion of the G-CSF- responsive progenitor (G-CFC) compartment in the bone marrow. In addition, in vivo treatment of gcsfr-Delta715 mice with G-CSF results in increased production of neutrophils leading to a sustained neutrophilia. This hyperproliferative response to G-CSF is accompanied by prolonged activation of signal transducer and activator of transcription (STAT) complexes and extended cell surface expression of mutant receptors due to defective internalization. In view of the continuous G-CSF treatment of SCN patients, these data provide insight into why progenitor cells expressing truncated receptors clonally expand in vivo, and why these cells may be targets for additional genetic events leading to leukemia.

Purified CD34(+) and CD34(+)CD38(-) human umbilical cord blood (UCB) cells were transduced with the recombinant variant of Moloney murine leukemia virus (MoMLV) MFG-EGFP or with SF-EGFP, in which EGFP expression is driven by a hybrid promoter of the spleen focus-forming virus (SFFV) and the murine embryonic stem cell virus (MESV). Infectious MFG-EGFP virus was produced by an amphotropic virus producer cell line (GP+envAm12). SF-EGFP was produced in the PG13 cell line pseudotyped for the gibbon ape leukemia virus (GaLV) envelope proteins. Using a 2-day growth factor prestimulation, followed by a 2-day, fibronectin fragment CH-296-supported transduction, CD34(+) and CD34(+)CD38(-) UCB subsets were efficiently transduced using either vector. The use of the SF-EGFP/PG13 retroviral packaging cell combination consistently resulted in twofold higher levels of EGFP-expressing cells than the MFG-EGFP/Am12 combination. Transplantation of 10(5) input equivalent transduced CD34(+) or 5 x 10(3) input equivalent CD34(+)CD38(-) UCB cells in nonobese diabetic/severe combined immunodeficient (NOD/SCID) mice resulted in median engraftment percentages of 8% and 5%, respectively, which showed that the in vivo repopulating ability of the cells had been retained. In addition, mice engrafted after transplantation of transduced CD34(+) cells using the MFG-EGFP/Am12 or the SF-EGFP/PG13 combination expressed EGFP with median values of 2% and 23% of human CD45(+) cells, respectively, which showed that the NOD/SCID repopulating cells were successfully transduced. EGFP+ cells were found in all human hematopoietic lineages produced in NOD/SCID mice including human progenitors with in vitro clonogenic ability. EGFP-expressing cells were also detected in the human cobblestone area-forming cell (CAFC) assay at 2 to 6 weeks of culture on the murine stromal cell line FBMD-1. During the transduction procedure the absolute numbers of CAFC week 6 increased 5- to 10-fold. The transduction efficiency of this progenitor cell subset was similar to the fraction of EGFP+ human cells in the bone marrow of the NOD/SCID mice transplanted with MFG-EGFP/Am12 or SF-EGFP/PG13 transduced CD34(+) cells, ie, 6% and 27%, respectively. The study thus shows that purified CD34(+) and highly purified CD34(+)CD38(-) UCB cells can be transduced efficiently with preservation of repopulating ability. The SF-EGFP/PG13 vector/packaging cell combination was much more effective in transducing repopulating cells than the MFG-EGFP/Am12 combination.

IFN-alpha has been shown to prolong survival in chronic myeloid leukemia patients, but its mechanism of action is still not understood. The human cobblestone area-forming cell (CAFC) assay allows for the measurement of the concentration of normal as well as malignant stem cells, while their progeny can be measured in parallel long-term culture (LTC) in flasks. Using CAFC and LTC assays, we have examined direct effects of IFN-alpha (500; 5,000 IU/ml) on the maintenance and outgrowth of CD34-enriched normal and malignant stem cells, obtained from six patients with an established major cytogenetic response to IFN-alpha and from four nonresponding patients. CAFC concentrations were not affected by IFN-alpha. In contrast, IFN-alpha strongly inhibited the clonogenic output in flask LTC. Nucleated cells (NC) produced in LTC were evaluated by fluorescent in situ hybridization (FISH) for the presence of the Philadelphia (Ph) translocation. After 8 wk of LTC, the percentage of Ph+ NCs produced was significantly more inhibited by IFN-alpha in responding patients than in nonresponders. Control LTC without IFN-alpha showed no significant differences of Ph+ NC production between responders and nonresponders. These findings provide the first in vitro model for cytogenetic conversion and suggest that direct antiproliferative effects of IFN-alpha account for the cytogenetic response observed clinically.

In vivo expansion and multilineage outgrowth of human immature hematopoietic cell subsets from umbilical cord blood (UCB) were studied by transplantation into hereditary immunodeficient (SCID) mice. The mice were preconditioned with Cl2MDP-liposomes to deplete macrophages and 3.5 Gy total body irradiation (TBI). As measured by immunophenotyping, this procedure resulted in high levels of human CD45(+) cells in SCID mouse bone marrow (BM) 5 weeks after transplantation, similar to the levels of human cells observed in NOD/SCID mice preconditioned with TBI. Grafts containing approximately 10(7) unfractionated cells, approximately 10(5) purified CD34+ cells, or 5 x 10(3) purified CD34+CD38- cells yielded equivalent numbers of human CD45+ cells in the SCID mouse BM, which contained human CD34+ cells, monocytes, granulocytes, erythroid cells, and B lymphocytes at different stages of maturation. Low numbers of human GpA+ erythroid cells and CD41+ platelets were observed in the peripheral blood of engrafted mice. CD34+CD38+ cells (5 x 10(4)/mouse) failed to engraft, whereas CD34- cells (10(7)/mouse) displayed only low levels of chimerism, mainly due to mature T lymphocytes. Transplantation of graded numbers of UCB cells resulted in a proportional increase of the percentages of CD45+ and CD34+ cells produced in SCID mouse BM. In contrast, the number of immature, CD34+CD38- cells produced in vivo showed a second-order relation to CD34+ graft size, and mice engrafted with purified CD34+CD38- grafts produced 10-fold fewer CD34+ cells without detectable CD34+CD38- cells than mice transplanted with equivalent numbers of unfractionated or purified CD34+ cells. These results indicate that SCID repopulating CD34+CD38- cells require CD34+CD38+ accessory cell support for survival and expansion of immature cells, but not for production of mature multilineage progeny in SCID mouse BM. These accessory cells are present in the purified, nonrepopulating CD34+CD38+ subset as was directly proven by the ability of this fraction to restore the maintenance and expansion of immature CD34+CD38- cells in vivo when cotransplanted with purified CD34+CD38- grafts. The possibility to distinguish between maintenance and outgrowth of immature repopulating cells in SCID mice will facilitate further studies on the regulatory functions of accessory cells, growth factors, and other stimuli. Such information will be essential to design efficient stem cell expansion procedures for clinical use.

Stroma-supported long-term cultures (LTC) allow estimation of stem cell quality by simultaneous enumeration of hematopoietic stem cell (HSC) frequencies in a graft using the cobblestone area forming cell (CAFC) assay, and the ability of the graft to generate progenitors in flask LTC (LTC-CFC). We have recently observed that the number and quality of mobilized peripheral blood stem cells (PBSC) was low in patients having received multiple rounds of chemotherapy. Moreover, grafts with low numbers of HSC and poor HSC quality had a high probability to cause graft failure upon their autologous infusion. Because ex vivo culture of stem cells has been suggested to present an attractive tool to improve hematological recovery or reduce graft size, we have studied the possibility that such propagation may affect stem cell quality. In order to do so, we have assessed the recovery of different stem cell subsets in CD34+ PBSC after a 7-day serum-free liquid culture using CAFC and LTC-CFC assays. A numerical expansion of stem cell subsets was observed in the presence of interleukin-3 (IL-3), stem cell factor, and IL-6, while stroma-contact, stromal soluble factors, or combined addition of FLT3-ligand and thrombopoietin improved this parameter. In contrast, ex vivo culture severely reduced the ability of the graft to produce progenitors in LTC while stromal soluble factors partly abrogated this quality loss. The best conservation of graft quality was observed when the PBSC were cultured in stroma-contact. These data suggest that ex vivo propagation of PBSC may allow numerical expansion of various stem cell subsets, however, at the expense of their quality. In addition, cytokine-driven PBSC cultures require stroma for optimal maintenance of graft quality.

We studied the cell kill induced by granulocyte-macrophage colony-stimulating factor (GM-CSF ) fused to Diphtheria Toxin (DT-GM-CSF ) in acute myeloid leukemia (AML) samples and in populations of normal primitive hemopoietic progenitor cells. AML samples from three patients were incubated in vitro with 100 ng/mL DT-GM-CSF for 48 hours, and AML cell kill was determined in a proliferation assay, a clonogenic assay colony-forming unit-AML (CFU-AML) and a quantitative long-term bone marrow (BM) culture ie, the leukemic-cobblestone area forming cell assay (L-CAFC). To measure an effect on cells with in vivo leukemia initiating potential DT-GM-CSF exposed AML cells were transplanted into immunodeficient mice. In two out of three samples it was shown that all AML subsets, including those with long-term abilities in vivo (severe combined immunodeficient mice) and in vitro (L-CAFC assay) were reduced in number by DT-GM-CSF. Cell kill induced by DT-GM-CSF could be prevented by coincubation with an excess of GM-CSF, demonstrating that sensitivity to DT-GM-CSF is specifically mediated by the GM-CSF receptor. Therefore, binding and internalization of GM-CSF probably occur in immature AML precursors of these two cases of AML. The third AML sample was not responsive to either GM-CSF or DT-GM-CSF. The number of committed progenitors of normal bone marrow (burst-forming unit-erythroid, colony-forming unit granulocyte- macrophage, and cobble stone area forming cell [CAFC] week 2) and also the number of cells with long-term repopulating ability, assayed as week 6 CAFC, were unchanged after exposure to DT-GM-CSF (100 ng/mL, 48 hours). These studies show that DT-GM-CSF may be used to eliminate myeloid leukemic cells with long-term potential in vitro and in immunodeficient mice, whereas normal hemopoietic stem cells are spared.