BMT and Stem Cells

Multipotent neural precursors express neural and hematopoietic factors, and enhance ex vivo expansion of cord blood CD34+ cells, colony forming units and NOD/SCID-repopulating cells in contact and noncontact cultures

Article metrics


In view of the possible crosstalks between hematopoiesis and neuropoiesis, we evaluated two microenvironments, murine neonatal neural cell line C17.2 and primary embryonic aorta–gonad–mesonephros (AGM) stromal cells, on the ex vivo expansion of CD34+ cells from human cord blood. In a contact culture system, C17.2 or AGM cells significantly enhanced the expansion of CD34+ cells to a panel of early and committed hematopoietic progenitor cells. In a noncontact transwell system, pre-established C17.2 cells significantly increased the expansion of total nucleated cells, CD34+ cells and multilineage colony forming cells (P<0.01). Expanded cells were infused into nonobese diabetic/severe-combined immunodeficient mice. The engraftment of human (hu)CD45+ cells in the bone marrow of these mice was consistently higher in all the 10 experiments conducted with the support of C17.2 cells when compared with those in respective control groups (11.9 vs 2.43%, P=0.03). Using RT-PCR and Southern blot analysis, we showed that AGM and C17.2 cells expressed a panel of hematopoietic, bone morphogenetic and neurotrophic factors. Our data provided the first evidence on the promoting effects of a neural progenitor cell line on hematopoiesis at a noncontact condition. The mechanism could be mediated by the expression of multilineage regulatory factors.


Human umbilical cord blood has been established as an alternative source of hematopoietic stem cells for treatment of patients with malignant and nonmalignant conditions. The advantages of cord blood transplantation over bone marrow (BM) or mobilized peripheral blood stem cell (PBSC) transplantation include the ease of stem cell collection, less stringent requirement on the HLA blood group matching between donors and recipients, as well as the low severity of graft-versus-host disease.1, 2, 3 A major limiting factor, however, is the relatively low cell dose in a cord blood collection, which has resulted in the application of cord blood transplant mostly to patients with low body weight.1, 2 The associated delay in engraftment becomes a cause of morbidity, mortality and increased cost due to hospitalization, systemic infection and blood cell transfusion. A distinct survival advantage has been observed in cord blood transplants with a cell dose of over 4 × 107 nucleated cells per kg body weight of the recipient.1, 4 Thus, the generation of a large number of stem and progenitor cells by ex vivo expansion becomes a promising approach for improving the applicability and outcome of cord blood transplantation. Some clinical improvements have been observed in trials using expanded cord blood,5 BM6 and PBSC.7, 8

A major concern of culturing stem cells in the presence of hematopoietic growth factors is the promotion of lineage differentiation, possibly at the expense of multipotent stem cells with self-renewal and long-term engrafting potentials.9 Recently, there has been evidence suggesting the close relationship and possible crosstalks between hematopoiesis and neuropoiesis.10, 11 C17.2 is an immortalized, neural precursor cell line derived from the external germinal layer of neonatal mouse cerebellum.12 It can proliferate and differentiate into neurons, oligodendrocytes and astrocytes in vitro and in vivo, thus becoming a useful cell type for the investigation of neural regeneration and the transfer of therapeutic genes. However, its capacity as a microenvironment for promoting hematopoiesis has not been reported. Another logical candidate for the support of hematopoiesis and stem cell renewal is the embryonic aorta–gonad–mesonephros (AGM) microenvironment, from which the first definitive hematopoietic stem cells arise.13, 14 In this study, we provided evidence that these two microenvironments promoted the expansion of cord blood CD34+ cells to functional, early and multilineage hematopoietic cells, possibly mediated by the production of soluble regulatory factors.

Materials and methods

Collection and enrichment of human cord blood CD34+ cells

Umbilical cord blood samples were collected from cord veins during uncomplicated full-term, vaginal deliveries. These samples were kept in preservative-free heparin (10 IU/ml; David Bull Laboratories, Victoria, Australia) at room temperature and processed within 24 h. Informed consent was obtained for all blood collections and this study was approved by the Ethics Committee for Clinical Research of The Chinese University of Hong Kong. Cord blood mononucleated cells were prepared by density gradient centrifugation on Ficoll–Hypaque (1.077 g/ml) (Amersham Pharmacia, Uppsala, Sweden). CD34+ cells were enriched using the VarioMACS Isolation Kit (Miltenyi Biotec Inc., Gladbach, Germany) according to the manufacturer's instructions. The purity of enriched CD34+ cells evaluated by flow cytometry (FACSCalibur, Becton Dickinson (BD) San Jose, CA, USA) was 88.4±1.83% (mean±s.e.m., range 79.1–95.0%).

Mouse C17.2 cell line and primary AGM cells

The mouse C17.2 neural stem cell line was a gift from Dr David Walsh (Department of Anatomy, University of New South Wales, Australia). It was cultured in Iscove's modified Dulbecco's medium (IMDM; Gibco, Grand Island, NY, USA) containing 10% fetal calf serum (FCS; Gibco) at 37°C in a humidified atmosphere and 5% CO2. After the cells became confluent, they were harvested and cryopreserved until a week before coculturing with hematopoietic cells. We passaged these cells in 0.05% trypsin (Gibco) containing 0.53 mM ethylenediaminetetraacetic acid (EDTA; Gibco). Primary AGM cells were obtained from C57BL/6 fetal mice at 11.5 day of gestation. The region surrounding the aorta, genital ridge and mesonephros was collected under a dissecting microscope. After washing twice in IMDM, the tissues were treated with 0.04% collagenase (Sigma, St Louis, MO, USA) at 37°C for 1 h and pipetted gently to dissociate cell clumps every 15 min. The cells were then cultured in IMDM containing 10% FCS. Confluent adherent cells were harvested and cryopreserved in RPMI 1640 medium (Gibco) containing 25% FCS and 10% dimethyl sulfoxide (Sigma), until a week before coculturing with hematopoietic cells. All AGM stromal cells were used within five passages. AGM stromal cells were characterized by staining with anti-mouse fluorescein isothiocyanate (FITC)-labeled antibodies CD45, CD144 (VE cadherin), phycoerythrin (PE)-labeled CD34, CD31 (PECAM) or their respective isotypic controls (all from Pharmingen, San Diego, CA, USA). Trypsinized AGM cells were washed with PBS/0.2% bovine serum albumin (BSA, Gibco) and stained with specific antibodies for 20 min at room temperature. These cells were washed and 10 000 events were analyzed by the FACScan instrument (BD).

Contact and noncontact cocultures of CD34+ cells and stromal cells

At 1 day before the contact coculture of cord blood CD34+ cells and stromal cells, we trypsinized the stromal cells and resuspended them in IMDM containing 10% FCS. They were irradiated (30 Gy) in a Gammacell-1000 Elite Irradiator (MDS Nordion, Kanata, Ontario, Canada) and cultured in fibronectin-coated (20 μg/ml; Gibco) 24-well culture plates (BDA-3047; BD, Franklin Lakes, NJ, USA) at a density of 4 × 104/well in IMDM containing 10% FCS for 24 h. Enriched CD34+ cells at 2 × 104/well were added to the pre-established stromal cells, supplemented with thrombopoietin (TPO, 50 ng/ml), Flt-3 ligand (FL, 20 ng/ml) and interleukin (IL)-6 (20 ng/ml) and cultured at 37°C in a humidified atmosphere of 5% CO2.15 On days 7 and 10, four- and two-fold volumes of fresh medium and cytokines were added to the cultures, respectively. Expanded cells were harvested at days 7 and 14. First, nonadherent cells were collected aside. Adherent cells were recovered by incubation with 0.05% trypsin containing 0.53 mM EDTA at 37°C for 1 h. Both aliquots of cell suspensions were pooled for analysis of cell counts, viability by Trypan blue exclusion, and progenitor subpopulations by flow cytometry using fluorescence-conjugated antibodies against CD34 (BD), AC133 (CD133) (Miltenyi Biotec Inc.), CD61 and CD41 (Dako, Copenhagen, Denmark). We assayed functional progenitor cells as colony forming unit-granulocyte macrophage (CFU-GM), burst forming unit/colony forming unit-erythroid (BFU/CFU-E) and CFU-mixed (CFU-GEMM) in 1% methylcellulose (Sigma) cultures in IMDM supplemented with 30% FCS, 1% BSA and 0.1 mM β-mercaptoethanol (β-ME) in the presence of 3 IU/ml erythropoietin (EPO; Eprex, Cilag, Zug, Switzerland), 10 ng/ml granulocyte macrophage-colony stimulating factor (GM-CSF; Leucomax, Basle, Switzerland), 10 ng/ml IL-3 and 50 ng/ml stem cell factor (SCF) as described previously.15, 16 Enriched or expanded CD34+ cells at 3 × 103/ml were seeded in triplicate and incubated for 14 days. Colony forming unit-megakaryocytes (CFU-MK) were cultured in a plasma clot system. In all, 1 × 105 cells were cultured for 14 days in 35 mm plates with 1 ml IMDM containing 10% BSA, 0.1 mM β-ME, 10% FCS, 10% bovine plasma, 0.34 mg/ml CaCl2, 50 ng/ml TPO and 20 ng/ml IL-3 as described previously.15, 16 After staining with anti-human CD61-FITC antibody (Dako), CFU-MK was scored using fluorescence microscopy (LEITZ DMRD, Leica, Germany). All cytokines were purchased from Peprotech (Rocky Hill, NJ, USA) unless stated otherwise.

Noncontact cultures were performed on a transwell system using inserts (Costar, Corning, NY, USA) with a pore size of 0.4 μm and a pore density of 1 × 108 pores/cm2, thus allowing bidirectional diffusion across the membrane. C17.2 cells were cultured on six-well plates at a density of 4 × 104/ml in 2.8 ml IMDM containing 10% FCS. Enriched CD34+ cells (1.5 ml) at 2 × 104/ml were seeded in the transwell inserts placing on top of the pre-established C17.2 stromal cell layer. A control culture was performed without the addition of any stromal layer. The culture was supplemented with TPO (50 ng/ml), IL-6 (20 ng/ml), FL (20 ng/ml), with two-fold volumes of fresh medium and cytokines added at day 4. All cells at the inserts were harvested at day 8 for the analysis of stem and progenitor cells by flow cytometry, CFU assays and nonobese diabetic (NOD)/severe-combined immunodeficient (SCID) repopulating cell evaluation.

Engraftment of human cells in the NOD/SCID mouse model

NOD/LtSZ-scid/scid (NOD/SCID) mice were purchased from The Walter and Eliza Hall Institute of Medical Research (Victoria, Australia), bred and maintained in the Laboratory Animal Services Centre at The Chinese University of Hong Kong. All procedures were approved by the Animal Research Ethics Committee, The Chinese University of Hong Kong. Mice at 8–10 weeks of age were exposed to 280–320 cGy of total body irradiation from a 137Cs source (Gammacell-1000 Elite Irradiator). Ex vivo-expanded cells from individual cord blood (progenies of 3 × 104 CD34+ cells at day 0, n=10) cultured in the transwells with or without C17.2 cells were infused into sex- and age-matched mice. To prevent the loss of data due to animal mortality, the control and treatment groups each had two or three mice. These animals were killed 6 weeks post-transplantation. The engraftment parameters from each group were averaged as a single data for analysis, as described previously.15, 16

For the assessment of (hu)CD45+ cells and subsets, BM cells were flushed from both femurs of each mouse. Spleen cells were obtained by mincing and flushing separated cells from the tissue. We collected peripheral blood cells by heart puncture. For flow cytometric analysis, contaminating red blood cells were lysed with 0.83% ammonium chloride and washed with PBS/0.1% BSA. We then resuspended these cells at 5 × 105 cells/100 μl and incubated them with mouse IgG (Pharmingen, San Diego, CA, USA) and 5% human serum. These cells were then incubated with monoclonal antibody specific for huCD45 conjugated to phycoerythrin-cyanine 5-succinimidylester (PC5; Immunotech, Marseille, France) and propidium iodide (PI, 10 μg/ml; Sigma) for 20 min at room temperature. A total of 70 000 events were acquired, and for those BM samples that contained more than 1% human cells, we performed additional stainings using anti-human antibodies CD34-PE (BD), CD19-PE (Pharmingen), CD14-PE (Pharmingen), CD33-PE (Pharmingen), CD61-PE (Dako) and their isotypic controls. Nonviable cells (PI positive) were gated out during data analysis. We assayed human CFU in the BM of NOD/SCID mice that contained over 1% huCD45+ cells, using methylcellulose culture, and scored after 14 days.15 For CFU-MK assay, the plasma-clot system was performed in duplicate.

RT-PCR and Southern blot analysis of growth factors in mouse AGM and C17.2 cells

Total C17.2 and AGM cellular RNA were extracted by the RNeasy Mini Kit (Qiagen GmbH, Hilden, Germany) and the first-strand cDNA was synthesized using SuperScript II RNase H Reverse Transcriptase (Gibco) according to the manufacturer's protocols. Briefly, 1 μg of total cellular RNA was reverse-transcribed in the presence of 0.5 mmol/l each of dGTP, dTTP, dATP, dCTP, 100 ng of random hexamer and 200 U of SuperScript II enzymes in a volume of 20 μl. The RT reactions were performed at 42°C for 1 h, followed by inactivation of the enzyme at 70°C for 15 min. Subsequently, 1 μl of cDNA was amplified using 10 pmol of forward and reverse primers, 1 × PCR buffer, 2 mM MgCl2, 0.2 mM dNTP mix (Boehringer Mannheim, Mannheim, Germany) and 1.5 U Taq DNA polymerase (Gibco) in a final volume of 25 μl. The PCR reactions were performed in a Peltier Thermal Cycler PTC-200 (MJ Research, Watertown, MA, USA). PCR conditions for TPO, SCF, EPO, GM-CSF, IL-1β, IL-6, basic fibroblast growth factor (FGFb), vascular endothelial growth factor (VEGF), platelet-derived growth factor (PDGF), epidermal growth factor (EGF), nerve growth factor-β (NGF), brain-derived neurotrophic factor (BDNF), neurotrophin 3 (NT-3), insulin-like growth factor 1 (IGF-1), bone morphogenetic protein (BMP)-2, BMP-4, BMP-7 and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) began with an initial denaturation at 94°C for 10 min (except 5 min for TPO, SCF, IL-1β and GAPDH), followed by 30 cycles of denaturation at 94°C for 1 min, annealing at 55°C for 1 min, elongation at 72°C for 1 min, and ended with the final elongation at 72°C for 7 min. For FL, the denaturation, annealing and elongation steps were reduced to 30 s each. Conditions for EPO were 1, 2 and 3 min respectively for 40 cycles, whereas for granulocyte-colony stimulating factor (G-CSF), the denaturation was performed for 2 min, followed by 30 cycles of denaturation at 94°C for 1 min, annealing at 60°C for 45 s and elongation at 72°C for 45 s. All primers and probes were purchased from Gibco.

Southern blotting was performed as described previously17 according to the manufacturer's protocol using the ECL nucleic acid labeling and detection system (Amersham International, Buckinghamshire, England). PCR amplification products were fractionated on 1.5% agarose gels and transferred to Hybond-N+ membranes (Amersham) under an alkaline condition. The membranes were prehybridized at 42°C for 30 min and hybridized with 5 ng/ml of specific oligolabeled probes (Supplementary Table 1) at 42°C for 2 h. The membranes were then washed twice with 5 × SSC containing 0.1% (w/v) SDS for 5 min at room temperature, and then washed again twice with 1 × SSC containing 0.1% SDS for 15 min at 42°C. After blocking, the membranes were incubated with anti-fluorescein horseradish peroxidase conjugate (1:1000 dilution), detection reagents and subsequently exposed to autoradiograph films (Hyperfilm MP; Amersham).

Statistical analysis

We employed the SigmaStat software (Jandel Scientific Software, San Rafael, CA) to compare treatment and control groups. The analysis of variance and paired t-test or Wilcoxon's sign rank test were used, depending on data distribution. A P-value of 0.05 was considered as statistically significant. We compared survival rates of NOD/SCID mice using Fisher's exact test. All values were expressed as mean±standard error of the mean (s.e.m.).


Ex vivo expansion of CD34+ cells cocultured with C17.2 and AGM cells in contact cultures

C17.2 cell line and primary AGM cells were pre-established as the in vitro microenvironments for the expansion of cord blood CD34+ cells. The AGM cells were a fibroblast-like and heterogeneous population that expressed the endothelial cell marker CD144 (17.5±8.28%, n=5 from independent AGM collections) and hematopoietic marker CD45 (5.39±3.05%). These cells also expressed the CD31 (4.41±2.73%) and CD34 (4.21±2.41%) antigens, being detected on over 60% of the CD45+ population. In addition, there appeared a trend of decreased expressions of these markers with increased number of cell passages.

In the expansion culture, cell viability was high at day 7 (range 92.3–94.3%) and was well maintained at day 14 (87.2–90.9%). The coculture of enriched CD34+ cells with C17.2 stromal cells significantly increased the yield of CD34+ cells, CD34+AC133+ cells, BFU/CFU-E and CFU-MK at day 7 and all cell parameters and subsets at day 14 when compared with cultures that contained cytokines TPO, IL-6 and FL (T6F) alone (Figure 1). These increases were contributed not only by the total cell numbers but also the higher percentages of CD34+ cells (day 7: 19.4 vs 12.8%, P<0.01; day 14: 10.1 vs 6.04%, P<0.05) and CD34+ AC133+ cells (day 7: 17.2 vs 10.3%, P<0.01; day 14: 8.37 vs 4.91%, P<0.05) in the T6F+C17.2 group. Similar stimulating effects were observed when the expansions were performed with the support of primary AGM stromal cells (Figure 2). The number of CFU-MK was increased in the presence of either stromal layer (Figures 1 and 2). However, the percentages and total counts of CD61+CD41+ cells in the expansion cultures were not significantly increased by either C17.2 or AGM.

Figure 1

Ex vivo expansion of CD34+ cells in a contact coculture system with C17.2 cell line. Enriched CD34+ cells were cultured for 7 and 14 days in the presence of cytokines TPO (T), IL-6 (6) and FL (F), with or without pre-established C17.2 cells. C17.2 cells promoted the expansion of various stem and progenitor cell populations. The results are presented as mean and s.e.m. n=9. *P0.05, **P0.01.

Figure 2

Ex vivo expansion of CD34+ cells in a contact coculture system with primary AGM cells as the stromal layer. Enriched CD34+ cells were cultured for 7 and 14 days in the presence of cytokines T6F, with or without pre-established AGM cells. AGM cells promoted the expansion of various stem and progenitor cell populations. n=9. *P0.05, **P0.01.

Ex vivo expansion of CD34+ cells in noncontact cocultures with C17.2 cells and their engraftment in NOD/SCID mice

After 8 days of culture in the noncontact transwell system, cell viability was 90.6–98.7%. The expansion of CD34+ cells to multilineage progenitor cells was significantly increased in the presence of C17.2 stromal support (Figure 3). In the C17.2 cocultures, total nucleated cells, CFU-GM, CFU/BFU-E and CFU-MK were significantly increased by 17-, 20.3-, 9.5- and 24.6-fold, respectively, when compared with those by 9.2-, 9.1-, 4.2- and 9.4-fold in control cultures (P<0.01). Furthermore, C17.2 supported the expansion of early progenitors including CD34+ cells and CFU-GEMM by 3.1-fold (P=0.01) and 18-fold (P=0.01), respectively, when compared with those by 1.3- and 7.2-fold in the control group.

Figure 3

Ex vivo expansion of CD34+ cells in a noncontact transwell system with C17.2 cells. Enriched CD34+ cells were cultured for 8 days in a transwell system in the presence of cytokines T6F, with or without pre-established C17.2 cells. C17.2 cells promoted the expansion of various stem and progenitor cell populations. n=8. **P0.01.

Expanded cells were infused into sublethally irradiated NOD/SCID mice. At 6 weeks post-transplantation, the survival rates of mice in the control group, which received progenitor cells, expanded in cytokines alone (T6F) and those in the treatment group supported by T6F+C17.2 cells (92.3 and 74.0% respectively) were not significantly different (P=0.242). The proportion of huCD45+ cells in the BM of NOD/SCID mice was consistently higher in every one of the 10 experiments conducted in the presence of the noncontact stromal support of C17.2 cells when compared with those in the respective control group (11.9 vs 2.43%, P=0.03; Figure 4 and Supplementary Figure 1). There were trends of increased engraftment of huCD45+ cells in the spleen (2.26 vs 0.48%, P=0.07) and peripheral blood (2.46 vs 0.82%, P=0.11) of the mice in the C17.2 group. In BM of both groups of mice that demonstrated huCD45+ cell engraftment exceeding 1%, human progenitors (CD34+), myelocytic cells (CD14+, CD33+), B lymphocytes (CD19+) and megakaryocytic cells (CD61+) were consistently detected, but no significant difference was observed in the proportions of these subsets (Figure 5 and Supplementary Figure 1). Engraftments of total CFU (25.8±4.87 vs 10.5±3.08 per 105 BM cells, P=0.01) and CFU-GM (P=0.05) were significantly higher in the C17.2 group and trends of increased CFU/BFU-E (5.19 vs 2.48 per 105 BM cells, P=0.10) and CFU-GEMM (1.44 vs 0.50 per 105 BM cells, P=0.18) were also demonstrated in this group when compared with mice that received cells expanded in the absence of C17.2.

Figure 4

Engraftment of ex vivo-expanded human cells in the BM of NOD/SCID mice. In 10 independent experiments, enriched CD34+ cells were cultured for 8 days in a transwell system in the presence of cytokines T6F, with or without pre-established C17.2 cells. Expanded cells were infused into sublethally irradiated NOD/SCID mice and the engraftment of huCD45+ cells (mean values from 2–3 mice) in the BM of these animals was investigated by flow cytometry. Our data demonstrated that the proportions of these cells in the C17.2 group were significantly higher than those in the control mice. *P=0.03.

Figure 5

Engraftment of human hematopoietic cell subsets in NOD/SCID mice. No significant difference was observed in the engraftment of human progenitors (CD34+), myelocytic cells (CD14+, CD33+), B lymphocytes (CD19+) and megakaryocytic cells (CD61+) in the BM of the two groups of mice.

Expression of growth factors in C17.2 and AGM cells

Southern blot analysis of RT-PCR products demonstrated mRNA expression of G-CSF, TPO, SCF, FL, EPO, IL-6, BMP-2, BMP-4, BMP-7, FGFb, VEGF, PDGF, NGF, BDNF, NT-3 and IGF-1 in C17.2 cells (Supplementary Figure 2). Similar pattern of expression was observed in AGM cells with the exception of G-CSF, EPO and BMP-2, which were undetectable. We used extracts of primary lung, kidney and brain cells as well as the M2-10B4 mouse stromal cell line as positive controls. GM-CSF, IL-1β and EGF were undetectable in C17.2 or AGM cells.


In the presence of TPO, IL-6 and FL, a cytokine combination optimized in our previous study,15 cord blood CD34+ cells were significantly expanded to early progenitors (CFU-GEMM) and committed progenitors of the myeloid (CFU-GM), erythroid (BFU-E/CFU-E) and megakaryocytic (CFU-MK) lineages. In contact cultures, both C17.2 and AGM stromal cells enhanced the ex vivo expansion across the panel of progenitor cell populations, including CD34+AC133+ cells, which are known to possess a high NOD/SCID engraftment potential.18 The increased proportions of CD34+ and CD34+AC133+ subsets indicated that the microenvironments promoted not only differentiation but also the proliferation of these progenitor populations. Similar effects were observed in the noncontact cultures, although the overall magnitude of expansion appeared to be less efficient (comparing Figures 2 and 3, day 7). This observation could be attributed to the lack of cell-to-cell contact or merely a result of the slightly altered expansion protocol of the transwell system such as on the surface area, cell density and the schedule of medium changing. In this study, we employed a 6-week NOD/SCID mouse transplantation model to investigate the capacity of human hematopoietic stem and progenitor cells to home to BM, proliferate and differentiate into multilineage mature blood cells. Significantly, there was a consistent increase of NOD/SCID repopulating potential/cells in expansion cultures exposed to the noncontact support of C17.2 cells. It also appeared that the in vivo differentiating capacity of these NOD/SCID repopulating cells to the B-lymphoid, myeloid and megakaryocytic lineages was not altered by their exposure to C17.2.

The membrane antigen profile of the primary AGM cells is in line with those reported on some AGM cell lines.19 Our data on the supportive effects of AGM stromal cells on early progenitors are in agreement with studies that showed that some AGM cell lines enhanced hematopoietic stem cell expansion.19, 20, 21, 22, 23 Ohneda et al19 reported that an AGM-derived CD34+ cell line DAS 104-8 supported the expansion of mouse fetal liver stem cells, but a direct contact culture condition was required to provide BM reconstitution. Again, Xu et al20 and Nishikawa et al21 demonstrated that the promotion of hematopoiesis by AGM stromal cell lines was mediated by direct cell-to-cell contact. To our knowledge, our data provided the first evidence on the promoting effects of a neural progenitor cell line on the ex vivo expansion of cord blood hematopoietic stem and progenitor cells. This is in line with a recent study24 that demonstrated the stromal support of primary human brain endothelial cells in contact and noncontact cultures supplemented with GM-CSF, IL-3, IL-6, SCF and FL for the expansion of adult BM CD34+ cells and NOD/SCID repopulating cells.

The mechanism of supportive microenvironments on in vitro hematopoiesis has been attributable to cell-to-cell contact as well as the production of soluble regulatory factors. In this study, we demonstrated that the direct cell-to-cell contact of cord blood CD34+ cells with C17.2 cells was not a prerequisite for the expansion of the various subsets of stem and progenitor cells. We thus speculated that stromal cell-derived, soluble cytokines might play a role in regulating the proliferation, differentiation or homing capacity of these cells. In the primary AGM cells and C17.2 cell line, we detected the expression of a panel of cytokines, including hematopoietic, bone morphogenetic and neural growth factors. The early-promoting hematopoietic cytokines TPO, SCF, FL and IL-6 but not the more lineage-specific cytokines GM-CSF, EPO and IL-1β were detected in AGM cells. This profile of hematopoietic cytokine expression was slightly different from that of the AGM-S3 cell line,21 which did not express TPO and FL. Interestingly, the clonal C17.2 cells also expressed these hematopoietic cytokines as well as G-CSF and EPO. As TPO, FL and IL-6 were already included in the culture medium, it was unlikely that their expression by the stromal cells had contributed significantly to the observed enhancement of hematopoiesis. Our data demonstrated that C17.2 cells expressed BMP-2, BMP-4 and BMP-7, whereas AGM expressed the latter two factors. BMP are members of the transforming growth factor (TGF)-β family involved in skeletal development and homeostasis.25 However, recent studies have demonstrated that most BMP are pleiotropic regulators of embryonic and tissue development and functions, including hematopoiesis.26, 27, 28 Our previous study demonstrated that PDGF enhanced the ex vivo expansion of NOD/SCID repopulating cells from cord blood CD34+ cells.15 A group of pleiotropic factors was detected in both AGM and C17.2 cells. FGFb and VEGF are known factors for promoting neural, angiogenic and hematopoietic development.29, 30, 31, 32 NGF, BDNF and NT-3 are members of the neurotrophin family. They are polypeptides noted for their neurotrophic activity in the peripheral and central nervous system and might also be implicated in hematopoietic cells.33, 34 IGF-1 is a mediator of the action of growth hormone on normal tissues and it performs multiple functions, including cell proliferation and inhibition of apoptosis in hematopoietic cells.35 The similarity in the expression profiles of AGM cells and C17.2 is in line with recent proposals on the close relationship between hematopoiesis and neuropoiesis during embryonic development and in the manifestation of ‘adult stem cell plasticity’.36, 37, 38 The overlapping genetic programs of the two systems could imply that a reciprocal regulatory mechanism might be in place. Emerging evidence has shown that important hematopoietic growth factors such as EPO39 and TPO40 protect neurons from damages in models of brain ischemia. The possible ‘cross-talks’ between hematopoiesis and neuropoiesis could have potential implications in areas of transplantation and tissue regeneration.

In summary, we have demonstrated the role of microenvironments of the AGM region and neural precursor cell line C17.2 in the ex vivo expansion of cord blood CD34+ cells. We provided the first evidence that C17.2 cells enhanced the expansion of multilineage NOD/SCID repopulating cells in a noncontact culture system. The mechanism could be mediated by the expression of growth regulatory factors not commonly included in protocols for hematopoietic stem cell expansion. The lack of requirement for cell-to-cell contact in this system would be clinically advantageous as it eliminates the undesirable contamination of the hematopoietic graft with other cell types. The microenvironment support, particularly the individual growth factors, could be further developed for use in the expansion of cord blood stem cells for transplantation. The multifold expansion of early progenitor cells of the myeloid and megakaryocytic lineages would contribute to the reduction or abrogation of neutropenia and thrombocytopenia in post-transplant patients. The increase in stem cell dose might allow the application of cord blood transplant to recipients of large body weights. The optimized expansion strategy could also be extrapolated to other sources of hematopoietic stem cells such as PBSC and BM.


  1. 1

    Gluckman E . Current status of umbilical cord blood hematopoietic stem cell transplantation. Exp Hematol 2000; 28: 1197–1205.

  2. 2

    Rubinstein P, Carrier C, Scaradavou A, Kurtzberg J, Adamson J, Migliaccio AR et al. Outcomes among 562 recipients of placental-blood transplants from unrelated donors. N Engl J Med 1998; 339: 1565–1577.

  3. 3

    Rocha V, Wagner Jr JE, Sobocinski KA, Klein JP, Zhang MJ, Horowitz MM et al. Graft-versus-host disease in children who have received a cord-blood or bone marrow transplant from an HLA-identical sibling: Eurocord and International Bone Marrow Transplant Registry Working Committee on Alternative Donor and Stem Cell Sources. N Engl J Med 2000; 342: 1846–1854.

  4. 4

    Locatelli F, Rocha V, Chastang C, Arcese W, Michel G, Abecasis M et al. Factors associated with outcome after cord blood transplantation in children with acute leukemia: Eurocord-Cord Blood Transplant Group. Blood 1999; 93: 3662–3671.

  5. 5

    Jaroscak J, Goltry K, Smith A, Waters-Pick B, Martin PL, Driscoll TA et al. Augmentation of umbilical cord blood (UCB) transplantation with ex vivo-expanded UCB cells: results of a phase 1 trial using the AastromReplicell System. Blood 2003; 101: 5061–5067.

  6. 6

    Stiff P, Chen B, Franklin W, Oldenberg D, Hsi E, Bayer R et al. Autologous transplantation of ex vivo expanded bone marrow cells grown from small aliquots after high-dose chemotherapy for breast cancer. Blood 2000; 95: 2169–2174.

  7. 7

    Paquette RL, Dergham ST, Karpf E, Wang HJ, Slamon DJ, Souza L et al. Ex vivo expanded unselected peripheral blood: progenitor cells reduce posttransplantation neutropenia, thrombocytopenia, and anemia in patients with breast cancer. Blood 2000; 96: 2385–2390.

  8. 8

    Reichle A, Zaiss M, Rothe G, Schmitz G, Andressen R . Autologous tandem transplantation: almost complete reduction of neutropenic fever following the second transplantation by ex vivo expanded autologous myeloid postprogenitor cells. Bone Marrow Transplant 2003; 32: 299–305.

  9. 9

    McNiece IK, Almeida-Porada G, Shpall EJ, Zanjani E . Ex vivo expanded cord blood cells provide rapid engraftment in fetal sheep but lack long-term engrafting potential. Exp Hematol 2002; 30: 612–616.

  10. 10

    Terskikh AV, Easterday MC, Li L, Hood L, Kornblum HI, Geschwind DH et al. From hematopoiesis to neuropoiesis: evidence of overlapping genetic programs. Proc Natl Acad Sci USA 2001; 98: 7934–7939.

  11. 11

    Jay KE, Gallacher L, Bhatia M . Emergence of muscle and neural hematopoiesis in humans. Blood 2002; 100: 3193–3202.

  12. 12

    Snyder EY, Park KI, Flax JD, Liu S, Rosario CM, Yandava BD et al. Potential of neural ‘stem-like’ cells for gene therapy and repair of the degenerating central nervous system. Adv Neurol 1997; 72: 121–132.

  13. 13

    Sanchez MJ, Holmes A, Miles C, Dzierzak E . Characterization of the first definitive hematopoietic stem cells in the AGM and liver of the mouse embryo. Immunity 1996; 5: 513–525.

  14. 14

    Medvinsky A, Dzierzak E . Definitive hematopoiesis is autonomously initiated by the AGM region. Cell 1996; 86: 897–906.

  15. 15

    Su RJ, Zhang XB, Li K, Yang M, Li CK, Fok TF et al. Platelet-derived growth factor promotes ex vivo expansion of CD34+ cells from human cord blood and enhances LTC-IC, NOD/SCID repopulating cells and formation of adherent cells. Br J Haematol 2002; 117: 735–746.

  16. 16

    Zhang XB, Li K, Yau KH, Tsang KS, Fok TF, Li CK et al. Trehalose ameliorates the cryopreservation of cord blood in a preclinical system and increases the recovery of CFUs, long-term culture-initiating cells, and nonobese diabetic–SCID repopulating cells. Transfusion 2003; 43: 265–272.

  17. 17

    Chui CMY, Li K, Yang M, Chuen CKY, Fok TF, Li CK et al. Platelet-derived growth factor up-regulates the expression of transcription factors NF-E2, GATA-1 and c-Fos in megakaryocytic cell lines. Cytokine 2003; 21: 51–64.

  18. 18

    Handgretinger R, Gordon PR, Leimig T, Chen X, Bühring H-J, Niethammer D et al. Biology and plasticity of CD133+ hematopoietic stem cells. Ann NY Acad Sci 2003; 996: 141–151.

  19. 19

    Ohneda O, Fennie C, Zheng Z, Donahue C, La H, Villacorta R et al. Hematopoietic stem cell maintenance and differentiation are supported by embryonic aorta–gonad–mesonephros region-derived endothelium. Blood 1998; 92: 908–919.

  20. 20

    Xu M, Tsuji K, Ueda T, Mukouyame Y, Hara T, Yang F et al. Stimulation of mouse and human primitive hematopoiesis by murine embryonic aorta–gonad–mesonephros-derived stromal cell lines. Blood 1998; 92: 2032–2040.

  21. 21

    Nishikawa M, Tahara T, Hinohara A, Miyajima A, Nakahata T, Shimosaka A . Role of the microenvironment of the embryonic aorta–gonad–mesonephros region in hematopoiesis. Ann NY Acad Sci 2001; 983: 109–116.

  22. 22

    Kusadasi N, Oostendorp RAJ, Koevoet WJLM, Dzierzak EA, Ploemacher RE . Stromal cells from murine embryonic aorta–gonad–mesonephros region, liver and gut mesentery expand human umbilical cord blood-derived CAFCweek6 in extended long-term cultures. Leukemia 2002; 16: 1782–1790.

  23. 23

    Charbord P, Oostendorp R, Pang W, Hérault O, Noel F, Tsuji T et al. Comparative study of stromal cell lines derived from embryonic, fetal, and postnatal mouse blood-forming tissues. Exp Hematol 2002; 30: 1202–1210.

  24. 24

    Chute JP, Saini AA, Chute DJ, Wells MR, Clark WB, Harlan DM et al. Ex vivo culture with human brain endothelial cells increases the SCID-repopulating capacity of adult human bone marrow. Blood 2002; 100: 4433–4439.

  25. 25

    Canalis E, Economides AN, Gazzerro E . Bone morphogenetic proteins, their antagonists, and the skeleton. Endocr Rev 2003; 24: 218–235.

  26. 26

    Chadwick K, Wang L, Li L, Menendez P, Murdoch B, Rouleau A et al. Cytokines and BMP-4 promote hematopoietic differentiation of human embryonic stem cells. Blood 2003; 102: 906–915.

  27. 27

    Detmer K, Walker AN . Bone morphogenetic proteins act synergistically with haematopoietic cytokines in the differentiation of haematopoietic progenitors. Cytokine 2002; 17: 36–42.

  28. 28

    Maguer-Satta V, Bartholin L, Jeanpierre S, French M, Martel S, Magaud J-P et al. Regulation of human erythropoiesis by activin A, BMP2, and BMP4, member of the TGFβ family. Exp Cell Res 2003; 282: 110–120.

  29. 29

    Dono R . Fibroblast growth factors as regulators of central nervous system development and function. Am J Physiol Regul Integr Comp Physiol 2003; 284: R867–R881.

  30. 30

    Kashiwakura I, Takahashi TA . Basic fibroblast growth factor-stimulated ex vivo expansion of haematopoietic progenitor cells from human placental and umbilical cord blood. Br J Haematol 2003; 122: 479–488.

  31. 31

    Carmeliet P . Blood vessels and nerves: common signals, pathways and diseases. Nat Rev Genet 2003; 4: 710–720.

  32. 32

    Gerber H-P, Malik AK, Solar GP, Sherman D, Liang XH, Meng G et al. VEGF regulates haematopoietic stem cell survival by an internal autocrine loop mechanism. Nature 2002; 417: 954–958.

  33. 33

    Simone MD, De Santis S, Vigneti E, Papa G, Amadori S, Aloe L . Nerve growth factor: a survey of activity on immune and hematopoietic cells. Hematol Oncol 1999; 17: 1–10.

  34. 34

    Dormady SP, Bashayan O, Dougherty R, Zhang XM, Basch RS . Immortalized multipotential mesenchymal cells and the hematopoietic microenvironment. J Hermatother Stem Cell Res 2001; 10: 125–140.

  35. 35

    Zumkeller W, Burdach S . The insulin-like growth factor system in normal and malignant hematopoietic cells. Blood 1999; 94: 3653–3657.

  36. 36

    Terskikh AV, Easterday MC, Li L, Hood L, Kornblum HI, Geschwind DH et al. From hematopoiesis to neuropoiesis: evidence of overlapping genetic programs. Proc Natl Acad Sci USA 2001; 98: 7934–7939.

  37. 37

    Zhao L-R, Duan W-M, Reyes M, Keene CD, Verfaillie CM, Low WC . Human bone marrow stem cells exhibit neural phenotypes and ameliorate neurological deficits after grafting into the ischemic brain of rats. Exp Neurol 2002; 174: 11–20.

  38. 38

    Shih CC, DiGiusto D, Mamelak A, Lebon T, Forman SJ . Hematopoietic potential of neural stem cells: plasticity versus heterogeneity. Leuk Lymphoma 2002; 43: 2263–2268.

  39. 39

    Digicaylioglu M, Lipton SA . Erythropoietin-mediated neuroprotection involves cross-talk between Jak2 and NK-κB signalling cascades. Nature 2001; 412: 641–647.

  40. 40

    Yang M, Xia WJ, Li KKH, Chik KW, Pong NH, Li CK et al. TPO has neural regeneration effect. Blood 2003; 102: 393b.

Download references


We thank the late Dr David Walsh for his support on our research program, and nurses of the Labour Ward for their assistance in collecting umbilical cord blood. This study was financially supported by the Hong Kong Paediatrics Bone Marrow Transplant Fund, The Chinese University of Hong Kong and the Industrial Support Fund AF/203/98, Department of Industry, Hong Kong Government Special Administrative Region.

Author information

Correspondence to K Li.

Additional information

Supplementary Information

Supplementary Information accompanies the paper on the Leukemia website (

Supplementary information

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Li, K., Lee, S., Su, R. et al. Multipotent neural precursors express neural and hematopoietic factors, and enhance ex vivo expansion of cord blood CD34+ cells, colony forming units and NOD/SCID-repopulating cells in contact and noncontact cultures. Leukemia 19, 91–97 (2005) doi:10.1038/sj.leu.2403542

Download citation


  • ex vivo expansion
  • cord blood CD34+ cells
  • C17.2 cells
  • AGM stromal cells
  • NOD/SCID mouse

Further reading