Density Separation of Umbilical Cord Blood and Recovery of Hemopoietic Progenitor Cells: Implications for Cord Blood Banking

Transcription

1 Density Separation of Umbilical Cord Blood and Recovery of Hemopoietic Progenitor Cells: Implications for Cord Blood Banking Camillo Almici, Carinelo Carlo-Stella, Lina Mangoni, Daniela Garau, Luca Cottafavi, Alessandro Ventura,b Mirella Armanetti, John E. Wagnecd Vittorio Rizzoli Department of Hematology, BMT Unit, University of Parma, Parma, Italy; bdepartment of Obstetrics and Gynecology, Hospital of Reggio Emilia, Italy; Department of Obstetrics and Gynecology, University of Parma, Parma, Italy; ddepartment of Pediatrics, University of Minnesota, Minneapolis, Minnesota, USA Key Words. Cord blood Density separation Cryopreservation culture Stem cell factor CD34 Hemopoietic progenitors 0 Long-term Abstract. Umbilical cord blood (CB) has been evaluated as a potential source of hematopoietic stem cells suitable for clinical use in the transplantation setting. Previous reports have documented a significant loss of progenitor cells by any manipulation other than cryopreservation. We have evaluated the feasibility of fractionating and cryopreserving CB samples with minimal loss of progenitor cells. We have compared various separation procedures based on different density gradients in the attempt to obtain the highest depletion of red blood cells (RBC) while maintaining the highest recovery of progenitor cells. We compared three different densities of Percoll (1.069 g/ml, g/ml, g/ml), sedimentation over poligeline (Emagel) and sedimentation over poligeline followed by separation over FicolVHypaque (F/H). Separated samples (n = 25) were analyzed for recovery of CD34 cells and progenitor cells (CFU-GEMM, BFU-E, CFU-GM). Separation by sedimentation over poligeline followed by F/H allowed the highest depletion of RBC (hematocrit of the final cellular suspension 0.4 i 0.1%) while maintaining high recovery of CD34 cells (85.3 i 5.6%) and total recovery for CFU- GEMM, BFU-E and CFU-GM. After cryopreservation, recovery of clonogenic progenitors was 82% for CFU-GEMM, 94% for BFU-E, 82% for CFU-GM and 90% for colony-forming units (CFUs) after five weeks of long-term culture (LTC). We further evaluated the effect of stem cell factor (SCF) on the in vitro growth of hemopoietic progenitors and on replating efficiency. The presence Correspondence: Dr. Camillo Almici, Cattedra di Ematologia, Universith di Parma, Via Gramsci 14, I- Parma, Italy. Received February 17, 1995; provisionally accepted March 26, 1995; accepted for publication April 25, OAlphaMed Press /95/$5.00/0 of SCF significantly increased CFU-GEMM (14 * 4 versus 49 i 5,p IO.0005) and CFU-GM (112 i 18 versus 178 f 19, p ), as well as the replating efficiency of CB progenitor cells (21 i 3.5% versus 43.3 f 4.7%) and the number of CFC per replated colony (4.7 f 3.5 versus 12.6 i 5.3, p I 0.005). In conclusion, RBC depletion of umbilical CB can be accomplished with minimal loss of committed and primitive hemopoietic progenitors. This procedure may have important implications in the large-scale banking of CB as well as ex vivo expansiodgene therapy protocols. Introduction Cord blood (CB) has been used as a source of hemopoietic stem cells for clinical transplantation [ 1-51 and is proving to be an acceptable alternative to bone marrow (BM). Studies are under way to determine the feasibility of CB banking for use in unrelated transplants [6]. Preliminary data suggest that CB cells are less alloreactive than marrow cells [7]. A major concern for wider transplant application has been related to the low total number of progenitor cells in CB that could be obtained in a single collection. While several studies suggest that CB contains a higher proportion of primitive hemopoietic progenitor cells as compared to BM [8-121, it has been reported that significant numbers of CB hemopoietic progenitor cells are lost by any manipulation prior to cryopreservation [ In fact, gradient separation techniques currently used for BM separation, have produced poor results, both in terms of red blood cell (RBC) depletion and progenitor STEM CELLS 1995;13:

2 534 Cord Blood Density Separation Procedures cell recovery, when applied to CB [ 151. These findings have been attributed to differences in red cell density, red cell fragility, electrical waves on the surface of red cells [I31 and differences in cloning efficiency of CB progenitor cells in BM progenitor assays [ Further development of efficient methods for processing CB is a critical issue in determining the feasibility and wide applicability of CB progenitors in transplantation. It was therefore the aim of our study to compare the efficacy of separation techniques based on different density gradients, in order to identify the separation procedure capable of providing more successful depletion of RBC and at the same time, the highest recovery of progenitor cells. Materials and Methods Samples Human umbilical CB samples were obtained with institutional review board approval from healthy full-term neonates immediately after delivery [ 181. Briefly, immediately after delivery of the infant, the umbilical cord was doubly clamped 3-5 cm from the umbilicus and transected between the clamps. The infant was removed from the field and blood collected from the maternal end of the transected cord. The samples were collected in 200 ml jars containing 10 ml of ACD-A as anticoagulant, and issued to the laboratory within 24 h after collection. Each CB sample was analyzed for volume, cellular viability, nucleated cell numbers, immunophenotype and progenitor cell content. CB samples were diluted 1: 1 with calcium- and magnesium-free Hank s balanced salt solution (HBSS) (GIBCO Laboratories, Grand Island, NY), divided in five equal aliquots and separated using separation techniques based on different density gradients, as detailed below. After each separation procedure, cells were collected, washed and assayed for number and depletion of RBC, immunophenotype and content in progenitor cells. Separation Procedures An aliquot of each sample was layered over three different gradients of Percoll (Biochrom KG; Berlin, Germany) (d = g/ml; d = g/ml; and d = g/ml), centrifuged (400 g for 40 min at 4 C) and cells at the interface were collected. The fourth aliquot was separated over a poligeline gradient (Emagel, Behringwerke; Marburg, Germany). The sample was diluted with poligeline in order to obtain a solution at 33% of poligeline in a 50 ml syringe. The mixture was maintained at room temperature for 60 rnin until the RBC layer occupied 25% of the initial volume. The leukocyte-rich plasma layer was collected while the sedimented RBC were discarded. The fifth aliquot was first sedimented over poligeline and then the leukocyte-rich plasma fraction layered over FicolUHypaque gradient (F/H, d = g/ml; Sigma Chemical Co., St. Louis, MO), centrifuged (400 g for 40 min at 4 C) and cells at the interface were collected. Inzmunofluorescence Analysis Unseparated and separated CB samples were counted, their concentration adjusted to 0.5 x 106/ml and their surface antigen phenotype determined by immunofluorescence analysis. Cells were incubated for 30 min on ice with fluorescein isothiocyanate (F1TC)- and/or phycoerythrin (PE)-conjugated antibody. Cells were first acquired using a total cell gate that included both lymphocytes and monocytes with elimination of events with a very low forward and side scatter that were CD45-. Each fluorescence analysis included a double negative control (IgG,-FITC/IgG,PE). The monoclonal antibodies used to generate the data reported were purchased from Becton Dickinson (Mountain View, CA) and included: CD33, CD34 (HPCA-2), CD38, HLA/DR and CD45RA. The percentage of positive cells was determined by subtracting the percentage of fluorescent cells in the control from the percentage of cells positively stained with the different monoclonal antibodies. Phenotypic analysis was performed with a FACSort flow cytometer (Becton Dickinson). Data were processed with a Hewlett Packard 340 computer (Forth Collins, CO) using Lysis I1 software (Becton Dickinson). Cryopreservation and Thawing CB mononuclear cells (MNC) obtained after separation over poligeline and afterwards over F/H were cryopreserved in cryotubes (Nunc; Roskilde, Denmark) at 2 x lo7 nucleated cellslml in Iscove s modified Dulbecco s medium (IMDM) (Biochrom KG; Berlin, Germany) with 10% (vol/vol) dimethylsulfoxide

3 Almici et al. 535 (DMSO) (Tera Pharmaceuticals, Buena Park, CA) and 30% fetal calf serum (FCS) (Hyclone, Logan, UT). Rapid thawing was performed in a water bath at 37 C followed by slow step-wise dilution over 10 min with 10 times the volume of IMDM supplemented with 10% FCS. Cells were then washed, counted and the number of progenitor cells (CFU-GEMM, BFU-E, CFU-GM) determined in short- and long-term culture. Short-Tern? Culture Assay The assays for CFU-GEMM, BFU-E and CFU-GM were performed as described in detail elsewhere [ 191. Briefly, unseparated and separated cells were plated in 35 mm petri dishes in one ml aliquots of IMDM containing 30% fetal bovine serum (FBS), 5 x M 2-mercaptoethanol and 1.1% (wtlvol) methylcellulose. Cultures were stimulated with a mixture of human recombinant colony-stimulating factors (CSFs) including interleukin 3 (IL-3; 10 ng/ml, Behringwerke; Marburg, Germany), granulocyte-csf (G-CSF; 10 ng/ml, Amgen Inc.; Thousand Oaks, CA), granulocyte macrophage- CSF (GM-CSF; 10 ng/ml, Behringwerke), erythropoietin (Epo) (one U/ml, Amgen) and with and without the addition of stem cell factor ([SCF]; 50 ng/ml, Amgen). Progenitor cell growth was evaluated after incubation (37'C, 5% C02) for 14 days in a humidified atmosphere. Four dishes were set up for each individual data point per experiment. CFU-GEMM, defined as containing at least erythroid and granulocytic cells by their in situ appearance, BFU-E with 2500 cells and CFU-GM with 240 cells were all scored from the same plates [19]. Long-Term Culture (LTC) Assay LTC assay was performed according to the methods previously described [ with minor modifications. As feeder layer we utilized the murine stromal cell line M210B4 transfected with the IL-3 and G-CSF genes (kindly provided by Dr. Connie Eaves, Vancouver, BC, Canada) [22]. The stromal layer at confluence was irradiated at a dose of 80 Gy before inoculation of test cells. At initiation, 2 x lo6 MNC, obtained after sequential separation over poligeline and F/H, were seeded per flask. At seven day intervals, cultures were demidepopulated by removal of one-half of the culture volume followed by replacement with fresh medium. At the fifth week of culture, both nonadherent and adherent cells were harvested, counted and assayed in methylcellulose for colony growth, as indicated above. Single Colony- Transfer Experiments On day 14 of primary culture of MNC obtained after separation over poligeline and F/H, single, well-isolated colonies were removed under sterile conditions. Each colony, picked up in a volume not greater than 5 p1, was transferred to an individual well of a 96-well tissue-culture plate containing 0.1 ml of IMDM supplemented with 20% FBS, and dispersed into a single-cell suspension by gentle pipetting. The complete resuspension was verified by optical examination. Each cell suspension was then plated in methylcellulose culture medium into an individual well of a 12-well tissue-culture plate. Secondary cultures stimulated with IL-3, GM-CSF, G-CSF and Epo were incubated for 14 days and scored for colonies using an inverted microscope. Statistical Analysis Statistical analysis was performed with the statistical package Statview (Brainpower Inc.; Calabasas, CA) run on a Macintosh I1 personal computer (Apple Computer; Cupertino, CA). The Student's t-test for paired data was used to test for significance of changes in the Comparison of data involving counts. Results Cord Blood Collection The CB collection data, as well as the immunophenotype and the characteristics in short-term culture before any manipulation, are summarized in Table I. A total of 25 CB samples was collected and processed. The volume (mean * SD) was 58 k 11 ml, the number of nucleated cells was 13.1 f 3.4 x 106/ml, the hematocrit level was 47.8 k 3.2% and the viability was always >95%. Neither the total nucleated cell number nor the percentage of CD34' cells in the individual CB samples correlated with the volume collected (data not shown). Cell Separation Procedures Several methods of enriching nucleated cells by removal of RBC were compared. In Table I1 results are calculated as total numbers, as though the whole CB sample had been

4 ~ 536 Cord Blood Density Separation Procedures Table 1. General characteristics (mean i SD) of cord blood samples before manipulation Mean f SD Range Sample (n) 25 Viability (%) > WBC (x 106/ml) 13.1 i 3.4 ( ) Volume (ml) 58 f 11 (36-71) Hematocrit (a) 47.8 f 3.2 (43-49) CD34' (a) 0.7 f 0.2 ( ) CD33+(%) 12f3 ( ) CFU-GEMM* 2.6 f 0.9 (1-6) BFU-E* 33.6 f 9 (21-53) CFU-GM* 21.3 f 12 (14-35) ~~ *CFU-GEMM, BFU-E and CFU-GM are expressed as number of colonies per 5 x lo4 nucleated cells plated. Cultures were stimulated with IL-3 (10 ng/ml), G-CSF (10 ng/ml), GM-CSF (10 nglml) and Epo (1 U/ml). separated according to the indicated procedure, while in Table I11 results are reported as percentage of unseparated samples. Cell viability was >95% with all different separation procedures. The combination of poligeline sedimentation followed by separation over F/H (Em/FH) was able to achieve more complete RBC depletion, being the hematocrit level 0.4 f 0.1%. The nucleated cell recovery ranged between 11% and 17% for the different Percoll densities, 78% for poligeline and 23% for Em/FH. Imrnunofluorescence Analysis As shown in Table I, the frequency of CD34+ cells in CB, before any manipulation, was 0.7 * 0.2%, while the frequency of CD33+ cells was 12 k 370. The gradient separation over Percoll determined an important loss in the total number of CD34+ cells. On the contrary, the sedimentation over poligeline resulted in no loss in CD34' cells, while after separation over F/H, 85.3 i 5.6% of these cells were recovered (Table 111). Similar results were also obtained for the recovery of CD33' cells. In fact, the sedimentation over poligeline and the sequential separation over poligeline and F/H allowed the recovery of 99.4% and 87% of CD33+ cells, respectively. Progenitor Colony Forming Assay Progenitor assays were performed on unseparated CB, as well as after each separation procedure in order to calculate the recovery of hemopoietic progenitor cells. Results are reported in Table I1 as absolute numbers and in Table I11 as percentage of unseparated samples. Separation over Percoll determined a major loss of progenitor cells, probably due to the high number of RBC that interferes in the separation by tracking MNC and therefore lowering the recovery of progenitor cells. In fact, only 22.6%, 10.3% and 4% of CFU-GEMM were recovered after separation over Percoll of densities g/ml, g/ml and g/ml, respectively. Recovery results were similar for BFU-E. Higher recoveries were obtained for CFU-GM, amounting to 73.5%, 35.8% and 43.9% for Percoll of densities g/ml, g/ml and g/ml, respectively. In contrast, superior results were obtained with poligeline sedimentation either alone or as a first step for RBC depletion prior to separation over F/H. Both of these procedures permitted total recovery of CFU-GEMM and CFU-GM, with 67.7% and 28.8% of BFU-E recovery, respectively. Table II.Characteristics (mean i SD) of cord blood samples (n = 25) after separation over different density gradients Unseparated Hematocrit (%) 47.8 f 3.2 Cell Number (x lo6) CD34+ (x lo6) 5.3 f 0.8 CD33+ (x lo6) 91.2 f 5.3 CFU-GEMM (X 104). 3.9 f 0.7 BFU-E (X 104) 51.1 * 3.8 CFU-GM (X 104) 32.4 i 9.3 Emagel 10.1 f f f f f f f 2.7 EdFH 0.4 f f f f f f 1.7 Percoll (glml) i ~ i i A f f f fo f f f i f i * A f f f f 3.3 Results are expressed as total numbers, as though the whole CB sample had been separated according to the procedure indicated. Cultures were stimulated with IL-3 (10 nglml), G-CSF (10 nglml), GM-CSF (10 nglml) and Epo (1 Ulml).

5 Almici et al. 537 Table 111. Characteristics (%) of cord blood samples after separation over different density gradients Emagel EmlFH - Percoll (dml) Cell Recovery (%) CD34' (%) CD33+(%) CFU-GEMM (%) BFU-E (%) CFU-GM (%) MNC obtained after sequential separation over poligeline and F/H, were plated in methylcellulose stimulated with a standard combination of cytokines, as detailed above, in the presence or absence of SCF (50 ng/ml). The addition of SCF to cultures induced a statistically significant increase of the growth of CFU-GEMM ( versus 49 * 5, p ) and CFU-GM ( versus 178 * 19, p ) as shown in Figure 1. Addition of SCF resulted not only in significantly higher colony numbers but also in increased colony size. Cryopreservation As shown in Table IV, cryopreservation of CB samples obtained after sequential separation over poligeline and F/H, was associated with minimal loss of progenitor cells. Cell recovery CFWGM BFU-E CFU-GEMM Fig. 1. Effect of SCF on in vitro growth of hematopoietic progenitors from cord blood. 5 x lo4 mononuclear cells obtained after sequential separation over Emagel and FicoWHypaque were plated in methylcellulose. Cultures were stimulated with IL-3 (1 0 ng/ml), G-CSF (10 nglml), GM-CSF (10 nglml), Epo (1 Ulml) and with or without SCF (50 ng/ml). Presence of SCF (white bar), absence of SCF (striped bar). Vertical bars represent mean k SD. *p ; **p after freezing was 73 * 1 I%, with a viability >97%. Progenitor cell clonogenic capacity was determined, demonstrating a recovery of 82%, 94% and 82% for CFU-GEMM, BFU-E and CFU-GM, respectively. Recovery of primitive progenitor cells after cryopreservation was 90 8% (n = 4), as tested in long-term culture. Replating Potential To investigate the replating potential of CB MNC, single colonies were removed and replated individually into secondary cultures. Primary cultures were stimulated with a standard combination of CSFs (IL-3, G-CSF, GM-CSF and Epo) with or without the addition of SCF. Secondary cultures were stimulated with a standard combination of CSFs. The replating efficiency was determined as the percentage of primary colonies giving rise to at least one secondary colony. In Table V the results are summarized for nine paired experiments. Primary colonies cultured in the presence of SCF demonstrated greater potential for initiation of secondary cultures. A mean & SD of 43.3 k 4.7% of these transfers gave rise to secondary hemopoietic colonies while, in the absence of SCF in the primary cultures, only 21 f 3.5% of transferred colonies grew in secondary cultures. The mean f SD number of secondary colonies derived per each replated colony initiated without or with SCF was 4.7 f 3.5 and 12.6 * 5.3 (p I0.005), respectively. Secondary hemopoietic colonies were mainly of the CFU-GM lineage, although occasional BFU-E developed. Discussion It has recently been demonstrated that umbilical CB can be used as a source of transplantable

6 538 Cord Blood Density Separation Procedures Table IV. Clonogenic progenitor recovery (mean i SD) after cryopreservation of cord blood mononuclear cells (n = 13), separated sequentially over Emagel and FicolllHypaque Cryopreservation Pre Post Cell recovery CFU-GEMM BFU-E CFU-GM CFU Week 5* (%) (15 x 104) (15 x 104) (15 x 104) (J2 x lo6) f 5 53i7 178 i I9 204 i 12 73ill 40i3 50-t 11 I46 i 13 I84 i 16 Cultures in methylcellulose were stimulated with IL-3 (10 ng/ml), G-CSF (10 ng/ml), GM-CSF (10 ng/ml), Epo (1 U/ml) and SCF (50 nglml). *Colony-forming unit after five weeks in long-term culture; assay was performed in four different experiments. stem cells 11-4, 101. CB collected at delivery has been shown to contain hemopoietic progenitor cells at similar or higher frequencies than those in BM [12,23,24]. CB-derived progenitor cells may possess significant advantages in terms of proliferative capacity and immunologic reactivity [25,26]. These characteristics might have important implications for experimental programs involving gene transfer and unrelated transplantation [27]. CB, which is normally discarded, can be easily collected without any danger or inconvenience to the donor [6]. 'Therefore, CB is an attractive source of transplantable cells that can be used for the treatment of diseases potentially curable by bone marrow transplantation (BMT) [ l]. Since CB banks generally consist of unseparated samples, the advantage of an efficient separation technique would permit a reduction in the volume of samples to be cryopreserved, lowering the costs of banking and the need of spaces for storage. Volume reduction of the CB is only one of a number of benefits deriving from CB manipulation for transplants. In fact, the injection of a large volume of red cells and DMSO might be harmful for patients due to hemolysis products and DMSO toxicity. Moreover, the issue of RBC contamination holds a primary role for transplant application, since RBC depletion would reduce any possibility of ABO antigen incompatibility. In establishing a large-scale bank of cryopreserved CB samples, the issues of total volume and costs for cryopreservation need to be addressed. In fact, both of them are significantly decreased when the volume is reduced by RBC depletion [ 151. The separation of MNC would allow the storage of large numbers of CB samples with minimal space requirements, without the need for freezing unseparated blood bags. However, the major problem in separating CB is the high contamination of RBC, that interferes in the separation by tracking MNC and lowering the recovery of progenitor cells. Poor results have been reported when procedures derived from BM separation are directly applied to CB [15]. Interesting results to date have been reported using modified F/H and 3% gelatine separation procedure [24,28]. However, red cell contamination and progenitor cell loss have remained important problems. Therefore, we investigated other possible methods of separation in the attempt to ameliorate RBC depletion and maximize the progenitor cell recovery, showing, in contrast to earlier reports [13, 141, that the isolation of low-density CB cells was not accompanied by massive losses in numbers of Table V. Effect of SCF on the replating efficiency of cord blood progenitor cells Primary Colonies without SCF Primary Colonies with SCF No. Experiments 9 9 No. Primary Colonies Replated % Replates with at Least 1 Colony 21.0 f i 4.7 CFC per Replated Colony 4.7 i i 5.3* Primary cultures were stimulated with IL-3 (10 ng/ml), G-CSF (10 ng/ml), GM-CSF (10 nglml), Epo (1 U/ml) and with or without SCF (50 nglml). 'p S

7 Almici et al. 539 progenitor cells. We think that a sedimentation over a poligeline layer would deplete the majority of RBC. And coupling the sedimentation over poligeline with the separation over F/H would significantly reduce the final volume without affecting the recovery of progenitor cells. We report a recovery of CFU-GEMM and CFU-GM higher than loo%, caused probably by an underestimation of these progenitor cells in unseparated CB, due to the high contamination of RBC that interferes with the scoring of smaller colonies. Our results of separation over poligeline are similar to those obtained by separation over 3% gelatin [28]. However, the use of poligeline seems much easier and safer, since poligeline solution (Emagel, Behringwerke; Marburg, Germany) is commercially available and in contrast, gelatin comes in a powder form and therefore needs to be house-prepared and autoclaved for sterilization before use. The separation procedure over poligeline and F/H seems particularly useful when MNC separation is a first step toward CD34 cell purification. In fact, an excessively high contamination of RBC could interfere in the selection procedure by lowering the binding capacity to the specific devices and therefore decreasing the purity of the CD34 selected population. It was reported that cryopreservation of CB samples as whole or separated blood has minimal effects on cell recovery and viability [13, 24, 281. As we report, CB samples can be separated into MNC populations and frozen without any functional defects. Cryopreserved mononuclear cells were >97% viable and the in vitro colony assays demonstrated that these cells were also completely functional. Since CFU-GM colony number is not a reliable index of the content of stem cells, we have used long-term culture and replating assays to better measure the primitive progenitor cell content. CB cannot be established in primary longterm culture because it lacks sufficient stromal precursor cells to provide the microenvironment necessary for self-renewal and differentiation of hemopoietic cells while, on the contrary, if preformed marrow stroma is provided, the amplitude and length of progenitor cell production from CB is superior to that of normal bone marrow, suggesting that the proportion of stem cells is superior [9, 1 I]. We have reported an 82% recovery for CFU-GEMM and CFU-GM, a 94% recovery for BFU-E, and a 90% recovery of more immature progenitors in longterm culture. Furthermore, the colonies arising in cultures of CB are much larger than colonies arising from adult BM cultures, suggesting a greater proliferative potential for CB progenitors. This finding is even more evident by the addition of SCF to the CSFs mixture. SCF has been reported to influence early hematopoietic progenitors and to induce self-renewal [8, 12, 141. SCF in combination with other CSFs is evaluated as a factor for amplification of those cells capable of engraftment [29]. Results reported here confirm that SCF in combination with other CSFs significantly increases the growth of progenitor cells (CFU-GEMM and CFU-GM) and promote a significantly higher replating efficiency, as has been previously reported [8]. In conclusion, CB has proven to be an important source of hemopoietic stem cells suitable for clinical transplantation. Our data provide a rationale for cord blood banking as an alternative to volunteer marrow registries. Moreover, the large number of potential donors available would overcome the ethnic imbalance in the volunteer BM donors of the National Marrow Donor Program [30]. In view of adult transplantation, immunoselection of CD34 cells from CB [25,26, 3 1, 321 might provide an ideal material for studies of ex vivo stem cell expansion. Acknowledgments We are grateful to C. Eaves for providing us with the M210B4 murine stromal cell line. This work was supported in part by grants from Consiglio Nazionale delle Ricerche (Progetto Finalizzato A.C.R.O.) and Associazione Italiana per la Ricerca sul Cancro (AIRC). References Hirao A, Kawano Y, Takaue Y et al. Engraftment potential of peripheral and cord blood stem cells evaluated by a long-term culture system. Exp Hematol 1994;22: Gluckman E, Broxmeyer HE, Auerbach AD et al. Hematopoietic reconstitution in a patient with Fanconi anemia by means of umbilical cord blood from a HLA identical sibling. N Engl J Med 1989;321: Wagner JE, Broxmeyer HE, Byrd RL et al. Transplantation of umbilical cord blood after myeloablative therapy: analysis of engraftment. Blood 1992;79:

8 540 Cord Blood Density Separation Procedures 4 Apperley JF. Umbilical cord blood progenitor cell transplantation. Bone Marrow Transplant 1994; 14: Linch DC, Brent L. Can cord blood be used? Nature 1989;340: Rubinstein P, Rosenfield RE, Adamson JW et al. Stored placental blood for unrelated bone marrow reconstitution. Blood 1993;81: Harris DT, LoCascio J, Besencon FJ. Analysis of the alloreactive capacity of human umbilical cord blood: implications for graft-versus-host disease. Bone Marrow Transplant 1994;14: Carow CE, Hangoc G, Cooper SH et al. Mast cell growth factor (c-kit ligand) supports the growth of human multipotential progenitor cells with a high replating potential. Blood 1991 ;78: Hows JM, Bradley BA, Marsh JCV et al. Growth of human umbilical cord blood in long-term hemopoietic cultures. Lancet 1992;340: Nicol AJ, Hows JM, Bradley BA. Cord blood transplantation: a practical option? Br J Haematol 1994;87: Pettengel R, Luft T, Henschler R et al. Direct comparison by limiting dilution analysis of longterm culture-initiating cells in human bone marrow, umbilical cord blood, and blood stem cells. Blood l994;84: Broxmeyer HE, Hangoc G, Cooper S et al. Growth characteristics and expansion of human umbilical cord blood and estimation of its potential for transplantation in adults. Proc Natl Acad Sci USA 1992;89: Broxmeyer HE, Douglas GW, Hangoc G et al. Human umbilical cord blood as a potential source of transplantable hematopoietic stem/progenitor cells. Proc Natl Acad Sci USA 1989;86: Migliaccio G, Migliaccio AR, Druzin ML et al. Long-term generation of colony-forming cells in liquid culture of CD34 positive cord blood cells in the presence of recombinant human stem cell factor. Blood 1992;79: Newton I, Charbord P, Schaal JP et al. Toward cord blood banking: density separation and cryopreservation of cord blood progenitors. Exp Hematol 1993;21: Thierry D, Hervatin F, Traineau R et al. Hematopoietic progenitor cells in cord blood. Bone Marrow Transplant 1992;9(suppl 1): Almici C, Carlo-Stella C, Mangoni L et al. Enrichment and sensitivity to mafosfamide of cord blood derived progenitor cells separated sequentially over poligeline 33% and Ficoll/Hypaque. Exp Hematol 1994;22:7 11 a. 18 Wagner JE, Broxmeyer HE, Cooper S. Umbilical cord and placental blood hematopoietic stem cells: collection, cryopreservation and storage. J Hematother 1992;1: Carlo-Stella C, Cazzola M, Ganser A et al. Synergistic antiproliferative effect of recombinant interferon gamma with recombinant interferon alpha on chronic myelogenous leukemia hematopoietic progenitor cells (CFU-GEMM, BFU-E, CFU-GM). Blood 1988;72: Dexter TM, Allen TD, Lajtha LG. Conditions controlling the proliferation of hemopoietic stem cells in vitro. J Cell Physiol 1977;91: Gartner S, Kaplan HS. Long-term cultures of human bone marrow cells. Proc Natl Acad Sci USA 1980;77: Sutherland HJ, Eaves CJ, Eaves AJ et al. Characterization and partial purification of human marrow cells capable of initiating long-term hematopoiesis in vitro. Blood l989;74: Kinniburgh D, Russel NH. Comparative study of CD34-positive cells and subpopulations in human umbilical cord blood and bone marrow. Bone Marrow Transplant 1993;12: Harris DT, Schumacher MJ, Rychlik S et al. Collection, separation and cryopreservation of umbilical cord blood for use in transplantation. Bone Marrow Transplant 1994;13: Lansdorp PM, Dragowska W, Mayani H. Ontogeny related changes in proliferative potential of human hematopoietic cells. J Exp Med 1993;178: Mayani H, Dragowska W, Lansdorp PM. Characterization of functionally distinct subpopulations of CD34 positive cord blood cells in serum free long term cultures supplemented with hematopoietic cytokines. Blood 1993;82: Vormoor J, Lapidot T, Pflumio F et al. Immature human cord blood progenitors engraft and proliferate to high levels in severe combined immunodeficient mice. Blood 1994;83: Bertolini F, Lazzari L, Lauri E et al. A comparative study of different procedures for the collection and banking of umbilical cord blood. J Hematother 1995;4: Abboud MR, Xu F, Payne A et al. Effects of recombinant human steel factor (c-kit ligand) on early cord blood hematopoietic precursors. Exp Hematol 1994;22: Howard MR, Gore SM, Hows JM et al. A prospective study of factors determining the outcome of unrelated marrow donor searches. Bone Marrow Transplant 1995 (in press). 31 Lu L, Xiao M, Shen RN et al. Enrichment, characterization, and responsiveness of single primitive CD34+++ human umbilical cord blood hematopoietic progenitors with high proliferative and replating potential. Blood 1993;81: Traycoff CM, Abboud MR, Laver J et a]. Evaluation of the in vitro behavior of phenotipically defined populations of umbilical cord blood hematopoietic progenitor cells. Exp Hematol 1994:22: