|
|
||||||||
a Department of Pediatrics,
b Obstetrics/Gynecology,
c Clinical Pathology,
d Hematology/Oncology,
e Medical Research Center, College of Medicine, Ewha Womens University;
f Department of Cell Biology and Immunology, Asan Institute for Life Sciences; University of Ulsan, Seoul, Korea;
g Department of Hematology, Chungnam University, Dae-Jeon, Korea;
h Graduate School of Biotechnology, Korea University, Seoul, Korea
Key Words. Cord blood CD34 • Ex vivo expansion • Adherent endothelial cell • Apoptosis • Long-term liquid culture
Chu-Myong Seong, M.D., Dept. of Hematology/Oncology, Ewha Wowens University, Mock-Dong Hospital, Yangchun-Ku, Mock-6-Dong 911-1, Seoul, Korea 158-056. Telephone: 82-2-650-5015; Fax : 82-2-650-5062; e-mail: cmseong{at}ewha.ac.kr.
| ABSTRACT |
|---|
|
|
|---|
| INTRODUCTION |
|---|
|
|
|---|
The feasibility of ex vivo expanding umbilical cord blood (UCB) CD34+ cells to provide increased cell doses for transplantation is of considerable interest [4]. We have previously shown that during ex vivo expansion of UCB CD34+ cells using thrombopoietin (TPO), flt-3 ligand (FL), and/or granulocyte colony-stimulating factor (G-CSF) in stroma-free liquid culture, distinct patterns of apoptosis are associated with modulation of CD44 [5], and phenotypic changes are dependent on the type of cytokines used [6]. Meanwhile, during UCB CD34+ cell expansion when using the same triple cytokines, we observed that a fibroblast-like adherent cell layer was found by week 4 or 5. Therefore, we have investigated the ability of UCBs to establish a stromal cell layer in a modified long-term liquid culture using immunohistochemistry and flow cytometry. We have also investigated whether preestablished stromal cells derived from CD34+ UCB can provide the beneficial effects on CD34+ cell expansion during ex vivo expansion, compared with those without adherent cell layers.
Recently, 7-amino-actinomycin D (7-AAD) has been introduced as a valuable fluorescent dye for assessing apoptosis [7]. The method of 7-AAD enables simultaneous staining of cell-surface antigens, as it does not require permeabilization of the cell membrane [8]. For assessment of apoptotic changes, flow cytometry was applied for quantification of apoptosis on an individual cell level. Finally, we compared the changes of apoptosis during ex vivo expansion with or without a preestablished adherent cell layer, which is derived from UCB CD34+ cells.
| MATERIALS AND METHODS |
|---|
|
|
|---|
Stroma-Free Liquid Culture
In 10 independent experiments from different cord blood donors, the purified CD34+ cell fractions were suspended in Iscoves-modified Dulbeccos medium ([IMDM] GIBCO; Grand Island, New York; http://www.lifetech.com) at a density of 5.0 x 105 cells/ml and then supplemented with 10% fetal bovine serum ([FBS] BioWhittaker; Walkersville, Maryland), recombinant humanized (rh) TPO (10 U/ml), rhFL (50 ng/ml), and rhG-CSF (100 U/ml) as previously described [5]. Custom-made coverslips were placed at the bottom of each 24-well culture plate and incubated at 37°C in 5% CO2. Twice a week, the cells were fed with the removal of one-half of the culture volume, which was replaced with fresh medium and growth factors. The cells were cultured for 5-6 weeks. During ex vivo expansion, an inverted light-microscopic examination was done. When a firmly adherent monolayer was fully formed, the adherent cell layers from the culture were harvested for morphological, immunophenotypic, and functional studies.
Immunohistochemistry
The adherent cell layers grown on customized coverslips were washed with PBS and then fixed in acetone or 4% paraformaldehyde for 15 minutes. Immunostaining was performed using a three-step streptoavidin-biotin complex immunoperoxidase technique. Briefly, endogenous peroxidase activity was blocked with 3% H2O2 in water for 5 minutes and washed in PBS three times for 5 minutes. Primary monoclonal antibodies at different dilutions were incubated with a fixed cell layer for 2 hours at room temperature and then the cells were washed in PBS for 10 minutes. Afterwards, the secondary biotinylated antibodies (goat anti-mouse IgG or goat anti-rabbit IgG [DAKO; Carpenteria, CA; http://www.dako.dk]) were applied for 30 minutes. The control was treated with a secondary antibody. Endogenous biotin activity was blocked by incubation of slides with avidin-biotin complex for 30 minutes and the staining was visualized with a 3-amino-9-ethylcarbazole (in N,N-dimethylformamide) substrate. The cells were then counterstained with hematoxylin solution. Next, slides were mounted in aqueous-based mounting media (DAKO). Primary antibodies were as follows: von Willebrand factor ([vWF] DAKO), human vascular cell adhesion molecule-1 ([VCAM-1] DAKO), human intracellular adhesion molecule-1 ([ICAM-1] Novocastra; Burlingame, CA;
http://www.novocastra.com
), human CD31 ([PECAM-1] DAKO), E-selectin (endothelial leukocyte adhesion molecule-1; DAKO), and human macrophage.
Cocultures Using Preestablished Adherent Cell Layers
Preestablished adherent cell monolayers from UCB were irradiated with 1,500 cGy. This was done in order to eliminate endogenous hematopoietic foci before the inoculation of the cells. In 10 independent experiments from different cord blood donors, freshly isolated UCB CD34+ cells in 24-well plates (BD Falcon) were cultured with and without the irradiated adherent cell layer. All experiments were duplicated. This was done at 5.0 x 104 cells/ml in IMDM supplemented with 10% FBS and cytokines (rhTPO [10 U/ml], rhFL [50 ng/ml], and rhG-CSF [100 U/ml]). These cells were cultured by the above methods for up to 5 weeks. At initiation and weekly intervals thereafter, the harvested cells were tested for a total cell number, CD34, colony-forming units granulocyte-macrophage (CFU-GM), and CFU-granulocyte-erythrocyte-macrophage-megakaryocyte (GEMM) with and without the association of an adherent cell layer, generated during the long-term in vitro culture of HCB CD34+ cells.
Fluorescence-Activated Cell Sorting (FACS) Analysis and Measurement of Apoptosis
Approximately 1 x 105 EDTA-detached cells in adherent cell layers were stained at 4°C for 30 minutes with FITC-anti-human CD34 (anti-HPCA-2; BD) and phycoerythrin (PE)-anti-human CD14 (TUK4; Chemicon; Temecula, CA; http://www.chemicon.com). During coculture on the adherent cell layer, the harvested cells were stained with 20 µg/ml 7-AAD (Sigma) at 4°C for 30 minutes [7, 8]. The negative control consisted of incubation with isotype-matched irrelevant antibodies (mouse IgG1-FITC [BD]; mouse IgG2a-PE [Chemicon]). Samples were analyzed on a FACSCalibur flow cytometer (BD). Ten thousand events were acquired for each analysis and data were analyzed using CellQuest (BD) or Attractors (BD) software. Results were expressed as mean ± standard deviation (SD) of percentage of positive cells from four separate samples.
Clonogenic Assays
Clonogenic assays were performed as follows: 1 x 103 CD34+ UCB cells were cultured at two plates per point in complete methylcellulose medium (HCC-4434; Methocult; Stem Cell Technologies, Vancouver, BC, Canada; http://www.stemcell.com). It was supplemented with 50 ng/ml stem cell factor (SCF), 10 ng/ml interleukin-3, GM-CSF, and 1 U/ml recombinant human erythropoietin at 37°C in a humidified atmosphere at 5% CO2. After 12 to 14 days of incubation, colonies were enumerated with the aid of an inverted microscope and counted by morphologic criteria.
Limiting Dilution Assay of UCB CD34+ Cells
Isolated UCB CD34+ cells using monoclonal antibody (QBEND10) and mini-MACS columns were seeded in 24-well plates at four dilutions, with 20 replicate wells for each. It was assumed in this model that there was homogenous suspension of endothelial cells in each dilution, and a single endothelial cell in a well resulted in a positive adherent cell. Wells that did not contain endothelial cells after 5 weeks of culture were counted and the number of endothelial cells in any other inoculum were determined by using Poisson distribution. A minimum of 1 x 102 cells/ml and a maximum of 1 x 104 CD34+ cells/ml were plated. The dilutions were chosen to be maximally informative at weeks 4 to 5 of the assay, yielding 10%-37% negative cells for fibroblastic adherent cell layers. Twice a week, the cells were fed with the removal of one-half of the culture volume, which was replaced with fresh medium and growth factors. After forming the fibroblastic adherent cell layer, we used immunohistochemistry using vWF antibody to confirm endothelial cells. Estimates of endothelial cell proportions were analyzed by standard limiting dilution assay techniques. Wells that did not contain endothelial cells after 5 weeks of culture were counted and the number of endothelial cells in any other inoculum were determined by using Poisson distribution.
Statistical Analysis
The results are presented as the mean ± SD of the data obtained from three or more experiments performed in duplicates. Statistical significance was determined using the Students t-test.
| RESULTS |
|---|
|
|
|---|
|
|
|
|
|
|
| DISCUSSION |
|---|
|
|
|---|
These endothelial cells may be generated from CD34+ cells using TPO, FL, and G-CSF because we used highly purified (93.6% ± 4.8%) CD34+ cells for culture (Fig. 1
), which were consistently shown in the previous two reports by our lab [5, 6], although the possibility that residual endothelial cell fractions from unfractionated UCB-formed endothelial cells cannot be ruled out. Despite the fact that large numbers of endothelial cells subsequently developed in the culture wells, mature endothelial cells were not observed on the cytospins (data not shown). Meanwhile, adherent fibroblast-like cells were negative for macrophage antigens. Estimates of endothelial cell proportions were analyzed by standard limiting dilution assay techniques. The number of tested cells (linear scale) producing 37% negative culture (logarithmic scale) is the number of cells containing one endothelial cell. Based on this method, one endothelial cell was found from 314 CD34+ cells after 5 weeks of culture. In addition, similar approaches of quantification of precursor cells for stromal cells in the CD34+ UCB should be done on more purified subsets of CD34+ cells (i.e., CD34+/CD38-/Thy1-+/Lin-, Rh123dim).
In our study, the established endothelial cell layers from UCB CD34+ cells supported to maintain and amplify the ex vivo expansion of CD34+ cells and CFU-GM (p < 0.05) and CFU-GEMM (p < 0.05, data not shown) during the 5 weeks of cultures compared with those without preestablished endothelial cell layers (Table 1
). By Philpott et al., the 7-AADdim cells have been demonstrated to be apoptotic by morphological observation, DNA gel electrophoresis, and terminal deoxynucleotidyl transferase-mediated deoxynucleotidyl (dUTP) nick end labeling [8]. On a scatter versus 7-AAD fluorescence, the three populations can be discriminated not only from one another but also from cell debris or clumps [5, 8]. The apoptosis of expanded hematopoietic progenitor cells were downregulated in preestablished endothelial cell cocultures groups (Table 2
). Traycoff et al. reported that the decline of primitive hematopoietic progenitor cell (HPC) activity during ex vivo expansion of human CD34+ cells was proliferation associated and might be a result of apoptosis [12]. Previously, we found some of the apoptotic fractions on at least myeloid differentiation of HCB CD34+ cells during ex vivo expansion were positive to CD64 and CD32 [6]. These results suggested that downregulation of apoptosis by endothelial cells that excrete endogenous cytokines during ex vivo expansion could be one of the mechanisms of more effective ex vivo expansion of HPCs. Human umbilical endothelial cells produce many cytokines: macrophage-colony stimulating factor, Kit ligand, fibroblast growth factor, transforming growth factor-ß, IL-1, tumor necrosis factor, IL-4, IL-6, GM-CSF, and G-CSF [13, 14]. However, which cytokines have key roles on apoptosis needs to be elucidated.
There may be some potential clinical applications of the stromal precursor cells for cellular therapies. Lazarus et al. and Liu et al. suggested that reconstitution of bone marrow stromal cells is enhanced by infusion of stromal progenitors [15, 16]. This observation has useful applications to cancer patients who have a severely damaged bone marrow microenvironment by cytotoxic drugs and radiation therapy after high-dose cytotoxic drugs [17]. Brandt et al. reported that the proliferation of HSCs in the presence of cytokines, without stromal cell support, may result in impairment of engraftment capacity, which may be overcome by coculturing with porcine endothelial cell lines [18]. However, further investigations are warranted to investigate if infusion of an ex vivo-generated adherent cell layer enhances the engraftment in patients transplanted with UCB CD34+ progenitors. Transplantation of the cultured donor adherent cells together with the ex vivo expanded CD34+ CB cells might not only accelerate hematologic recovery but also prevent acute graft-versus-host disease in the allogeneic setting. Moreover, the comparison of the effects produced by adherent cells with respect to other coculture systems (i.e., other stromal cell lines or primary bone marrow stromas) should be done in the future. In conclusion, our study found the establishment of the adherent endothelial cell layer, which does not express the macrophage-type antigens from UCB CD34+ cells during ex vivo expansion with TPO, FL, and G-CSF. These newly formed adherent endothelial cell layers seem to provide a better environment for UCB CD34+ cells during ex vivo expansion through downregulation of apoptosis. These observations may provide insight for the understanding of UCB transplantation and future cellular therapy or graft engineering.
| ACKNOWLEDGMENT |
|---|
|
|
|---|
| REFERENCES |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
R. M. Shepherd, B. J. Capoccia, S. M. Devine, J. DiPersio, K. M. Trinkaus, D. Ingram, and D. C. Link Angiogenic cells can be rapidly mobilized and efficiently harvested from the blood following treatment with AMD3100 Blood, December 1, 2006; 108(12): 3662 - 3667. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. A. Ingram, N. M. Caplice, and M. C. Yoder Unresolved questions, changing definitions, and novel paradigms for defining endothelial progenitor cells Blood, September 1, 2005; 106(5): 1525 - 1531. [Abstract] [Full Text] [PDF] |
||||
![]() |
W. Sun, L. Buzanska, K. Domanska-Janik, R. J. Salvi, and M. K. Stachowiak Voltage-Sensitive and Ligand-Gated Channels in Differentiating Neural Stem-Like Cells Derived from the Nonhematopoietic Fraction of Human Umbilical Cord Blood Stem Cells, August 1, 2005; 23(7): 931 - 945. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. A. Panepucci, J. L.C. Siufi, W. A. Silva Jr., R. Proto-Siquiera, L. Neder, M. Orellana, V. Rocha, D. T. Covas, and M. A. Zago Comparison of Gene Expression of Umbilical Cord Vein and Bone Marrow-Derived Mesenchymal Stem Cells Stem Cells, December 1, 2004; 22(7): 1263 - 1278. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. Herder, T. Tonn, R. Oostendorp, S. Becker, U. Keller, C. Peschel, M. Grez, and E. Seifried Sustained Expansion and Transgene Expression of Coagulation Factor VIII-Transduced Cord Blood-Derived Endothelial Progenitor Cells Arterioscler. Thromb. Vasc. Biol., December 1, 2003; 23(12): 2266 - 2272. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| STEM CELLS | THE ONCOLOGIST | CME | ALPHAMED PRESS JOURNALS |
