Allogeneic umbilical cord blood (UCB) cells have recently been used for transplantation following high-dose chemotherapy. However, the numbers of total cells, including progenitor cells, harvested are low compared with bone marrow or peripheral blood progenitor cell harvests. Therefore, we evaluated the potential of UCB cells for their ability to expand granulocyte-macrophage colony-forming cells (GM-CFC) and burst-forming unit-erythroid (BFU-E) cells over 10 days. We used an ammonium chloride lysing buffer to eliminate the majority of contaminating red blood cells. An average recovery of 61% of the starting number of white blood cells was obtained, while retaining 100% of the CD34+ cells. Ex vivo expansion cultures were established in Teflon cell culture bags (American Fluoroseal Corp, Columbia, MD) in 25 ml defined medium (Amgen Inc, Thousand Oaks, CA) containing 100 ng/ml each of stem cell factor (SCF), granulocyte colony-stimulating factor (G-CSF), and megakaryocyte growth and development factor. Either unselected UCB cells or CD34+ UCB cells, selected with Magnetic Activation Cell Sorting technology (Miltenyi Biotech GmbH, Bergisch Gladbach, Germany), were incubated for 10 days at 37 degrees C without refeeding. Unselected UCB cells seeded at 1 X 10(6)/ml produced an average expansion of 1.4-fold in total cells, 0.8-fold in GM-CFC, and 0.3-fold in BFU-E cells. By contrast, CD34+ selected UCB cells seeded at 1.0 X 10(4)/ml produced an average expansion of 113-fold in total cells, 72.6-fold in GM-CFC, and 49-fold in BFU-E cells. These data demonstrate that CD34+ cell selection is necessary for optimal expansion of both GM-CFC and BFU-E cells. The cell numbers thus obtained postexpansion may be sufficient for transplantation in adults.

The Amgen Cell Selection Device (ACSD) is a fully automated system based on the research scale magnetic-activated cell separation (MACS) system (Miltenyi Biotech GmbH, Bergisch Gladbach, Germany) for the selection of CD34+ cells. Leukapheresis products (LP) (n = 30) from normal donors mobilized with recombinant human granulocyte colony-stimulating factor (rhG-CSF) were selected with the ACSD to evaluate the performance of this system. The starting LP contained a median of 0.51% CD34+ cells (range 0.21%-1.54%) and a median WBC count of 3.0 x 10(10) (range 1-4.7 x 10(10) cells). After selection on the ACSD a mean purity of 91.5% +/- 0.6% CD34+ cells was obtained, with a median purity of 95.5% CD34+ cells. A median of 98 x 10(6) total CD34+ cells were recovered postselection, with a range of 31-323 x 10(6) cells collected from the LP. This represented a mean recovery of 81.7% +/- 6% of CD34+ cells and a median of 78% compared with starting CD34+ cell numbers in the LP. FACS analysis of the selected products demonstrated a 4-5 log depletion of T cell subsets, including CD3, CD4, CD8, and CD56 subsets. These data demonstrate the high performance obtained with the ACSD resulting in a final product of greater than 90% purity of CD34+ cells. CD34+ cells selected with the ACSD represent an ideal product for clinical applications, such as tumor cell purging, T cell depletion for allogeneic transplant, ex vivo expansion, and gene therapy.

No abstract

Ex vivo expansion of cord blood (CB) cells requires CD34+ cell selection before expansion to obtain optimal numbers of progenitor cells. As a preliminary step to preclinical development of CB expansion, we have evaluated two clinical scale selection devices, the Isolex 300i (Baxter Healthcare, Immunotherapy Division) and the CliniMACS (Miltenyi Biotech Inc.), for CD34+ cell selection from frozen CB products. As expansion of CB results in differentiation of cells, there may be a depletion of stem cells. Therefore, only a fraction of the CB should be expanded while a portion of the CB is maintained unmanipulated for infusion. After thawing of 40% fractions of each CB product, we observed >95% viable cells, with a median total WBC count of 1.8 x 10(8) cells. Use of the Isolex 300i resulted in a median purity of 51% CD34+ cells (n=8) and a median recovery of 34% CD34+ cells. Use of the CliniMACS resulted in a median purity of 54% CD34+ cells (n=10) and a median recovery of 80% CD34+ cells. The absolute number of CD34+ cells recovered after selection varied with samples from 6.7 x 10(4) to 3.2 x 10(6) CD34+ cells. Expansion of CD34+ cells from both systems resulted in >20-fold expansion of CFU-GM, with a median of 44-fold expansion. These data demonstrate the feasibility of selecting small fractions of frozen CB products using clinical scale CD34+ cell selection devices.

In this study, we have compared the ability of recombinant human granulocyte colony-stimulating factor (rhG-CSF) alone and the combination of low doses of recombinant rat pegylated stem cell factor (rrSCF-PEG) plus rhG-CSF to mobilize peripheral blood progenitor cells (PBPCs) with long-term engrafting potential. Female recipient irradiated mice were transplanted with PBPCs from male mice that were mobilized with rhG-CSF alone (group A) or rrSCF-PEG plus rhG-CSF (group B). As previously shown, greater short-term survival resulted in group B compared with group A, with 80% and 40% survival at 30 days posttransplant, respectively. Both groups of animals showed long-term donor-derived engraftment in greater than 95% of animals, as determined by quantitative specific polymerase chain reaction amplification of a Y chromosome sequence from whole blood of the mice at 6 to 12 months posttransplantation. Analysis of individual granulocyte-macrophage colonies, picked up from semisolid methylcellulose culture of bone marrow cells from transplanted mice, resulted in detection of donor- derived DNA in 98% of colonies from group B mice compared with 81% from group A mice. These data show that cells with long-term potential are mobilized by rhG-CSF alone and the combination of rrSCF-PEG plus rhG- CSF. Furthermore, an increased number of cells with short-term and long- term engraftment potential was obtained with rrSCF-PEG plus rhG-CSF compared with rhG-CSF alone.