Biol Reprod Keystone Symposia Conference on Frontiers in Reproductive Biology & Regulation of Fertility.
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

BOR - Papers in Press, published online ahead of print October 12, 2005.
Biol Reprod 2005, 10.1095/biolreprod.105.043331
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Retraction
Right arrow A retraction has been published
Right arrow All Versions of this Article:
75/3/487    most recent
biolreprod.105.043331v2
74/2/295
biolreprod.105.043331v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Zhang, H.
Right arrow Articles by Yuo, A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Zhang, H.
Right arrow Articles by Yuo, A.
Agricola
Right arrow Articles by Zhang, H.
Right arrow Articles by Yuo, A.
BIOLOGY OF REPRODUCTION 74, 295–306 (2006)
DOI: 10.1095/biolreprod.105.043331
© 2006 by the Society for the Study of Reproduction, Inc.

Research Article

Efficient and Repetitive Production of Hematopoietic and Endothelial Cells from Feeder-Free Monolayer Culture System of Primate Embryonic Stem Cells1

Hong Zhang 3 4, Kumiko Saeki 3 4, Akiko Kimura 4, Koichi Saeki 4, Masako Nakahara 4, Masaru Doshi 4, Yasushi Kondo 5, Toru Nakano 6, and Akira Yuo 2 4

Department of Hematology,4 Research Institute, International Medical Center of Japan, Shinjuku-ku, Tokyo 162-8655, Japan Advanced Medical Research Laboratories,5 Tanabe Seiyaku Co. Ltd., Yodogawa-ku, Osaka, 532-8505, Japan Department of Molecular Cell Biology,6 Osaka University Research Institute for Microbial Diseases, Osaka 565-0871, Japan

ABSTRACT

We have established an innovative culture system for the efficient differentiation of hematopoietic and endothelial cells from primate embryonic stem (ES) cells without feeder cells, embryoid bodies, or cell-sorting processes. After several days' culture in murine stromal OP9-conditioned medium supplemented with a cytokine cocktail on collagen-coated dishes, ES cells differentiated into a very unique population of cells with a finger-like appearance. These finger-like cells were positive for mesodermal and/or hemangioblastic markers of kinase insert domain receptor (KDR) and T-cell acute lymphocytic leukemia 1 (TAL1), and produced large amounts of protein tyrosine phosphatase, receptor type, C-positive hematopoietic cells. These hematopoietic cells showed the morphology of immature hematopoietic cells, formed blast cell colonies with high efficiency, and were positive for CD34 antigen, KDR, TAL1, and GATA binding protein 1, suggesting that these blast cells are equivalent to the multipotent hematopoietic progenitor cells. Moreover, they produced functional macrophages in murine stromal MS-5-conditioned medium and primitive erythroblasts in the presence of erythropoietin. The finger-like cells, putative mesodermal progenitors and/or hemangioblasts, actively proliferated and repetitively produced hematopoietic cells as long as they were maintained on the original dish. By contrast, the majority of the finger-like cells differentiated into endothelial cells with specific markers and specific functions after transfer to fresh dishes, indicating that conditions established in the original dish supported the proliferation and hematopoietic differentiation of the finger-like cells. Our method provides a highly controllable culture protocol for repetitive production of hematopoietic and endothelial cells from feeder-free monolayer cultivation of primate ES cells.

developmental biology, embryo

INTRODUCTION

Embryonic stem (ES) cells are a valuable resource in regenerative medicine because of their high capacity to differentiate into a broad range of cell types. It has been demonstrated that primate ES cells have characteristics distinct from those of murine ES cells. For example, leukemia inhibitory factor is ineffective for maintaining the immaturity of human ES cells [1]. The gene expression patterns also differ between human and murine ES cells. For example, undifferentiated human ES cells already express kinase insert domain receptor (a type III receptor tyrosine kinase) (KDR, also known as Flk-1, a type 2 receptor for vascular endothelial growth factor [VEGF]), which is used as a marker of the immature mesoderm in murine cells [2]. Although the study of human ES cells is essential for clinical applications, basic research using monkey ES cells still has great importance because these cells provide good transplantation models that must be used in the preclinical studies [3]. Moreover, the use of monkey ES cells avoids ethical issues and therefore techniques for biotechnological manipulation of them, including gene transfer, can immediately be applied, which will contribute to advancing our understanding of human ES cells.

Currently, there are three major methods for the in vitro differentiation of ES cells into specific lineages. One is coculturing ES cells with stromal cells as their feeders. Because feeder cells have a large capacity to promote directed differentiation and to maintain the viability of differentiated cells, this system provides a powerful tool for efficient production of mature tissues. However, massive contamination of the final products by feeder cells is inevitable and an appropriate cell-sorting process is needed. In addition, the culture of mixed cell populations has a risk of generating misleading results due to a trace amount of cells of other lineages even after cell sorting. The second method is based on the feeder-free culture of embryoid bodies (EBs), which are generated by removing signals that are required for maintaining the immaturity of ES cells. However, the direction of differentiation in EBs is completely random and cannot be focused on a specific lineage. Thus, the efficiency of differentiation often becomes too low to perform subsequent analyses or to use the products for clinical purposes. To obtain sufficient amounts of the cells with desired phenotypes, an enormously large volume of the starting material is required, along with an appropriate cell-sorting process for concentration. Moreover, these two methods have an additional disadvantage: they do not provide clear microscopic fields for observation and manipulation due to coexisting feeder cells or the compact three-dimensional structures of EBs. This problem has been overcome in a third method [4]. In this system, ES cells are first differentiated into primitive progenitor populations in a feeder-free, monolayer culture. Then, the desired progenitor population is selected by fluorescence-activated cell sorting (FACS) as the starting material for tissue differentiation. Although this system has shown successful outcomes in particular cases, it cannot be directly applied to situations where the markers of specific progenitors have not been identified. In addition, the risk of losing cells that are important but susceptible to FACS-associated cell damage, not only during the cell-sorting process but also during cell-disaggregating steps using trypsin and/or EDTA treatment, still remains. Thus, all the current methods largely depend on the cell-sorting process. Accordingly, the establishment of a novel method that does not require feeder cells, EB formation, or a presorting process is especially important.

For the purpose of the hematopoietic differentiation of primate ES cells, such a method has not yet been developed. Although the availability of a number of excellent stromal cell lines such as OP9 has enabled effective production of mature hematopoietic cells [5], the coexistence of the live stromal cells is required. In the EB-using feeder-free culture system, the yields of the production of hematopoietic progenitor cells from human ES cells are less than 10% [6, 7]. Although the presorting of KDR-positive immature mesoderm cells is useful to concentrate the common progenitor of the hematopoietic/endothelial lineage in the murine case, the subsequent hematopoietic differentiation step per se requires the live stromal cells [4]. Moreover, KDR is expressed in undifferentiated human ES cells [2] and this technique cannot be directly applied to the human case.

There is yet another critical problem left concerning the generation of hematopoietic cells from ES cells. It has been reported that the murine ES-derived hematopoietic cells lack the ability to expand in liquid culture unless specific genes, such as homeo box B4 (HOXB4) [8] or an oncogenic BCR/ABL1 (a fusion gene of break point cluster region [BCR] and v-abl Abelson murine leukemia viral oncogene homolog 1 [ABL1]) [9], are introduced. The gene transfer technique has, however, a crucial drawback with respect to clinical purposes because it has a large risk of leukemogenesis [10]. Thus, the establishment of a culture method for the effective generation of hematopoietic cells from feeder-free, monolayer primate ES cells without any gene-transfer technique is an urgent task.

Here we report a novel method for the hematopoietic and endothelial differentiation of cynomolgus monkey ES cells in feeder-free monolayer culture without presorting of progenitor populations. Because our system provides clear microscopic fields, we identified for the first time a specific population of cells, which we called finger-like cells, in a very early phase of differentiation. These unique finger-like cells were considered to be cells of mesodermal lineage with a hemangioblastic nature, based on the culture conditions of these cells, the expression of several mesodermal and hemangioblastic markers, and the bidirectional development of both hematological and endothelial lineages from these cells. The biological significance of the finger-like cells and their usefulness in basic research on early-phase hematopoiesis will be discussed.

MATERIALS AND METHODS

Cells and Reagents

Murine embryonic fibroblasts (MEFs) that had been treated with Dulbecco modified Eagle medium (DMEM) containing mitomycin C (MMC; Sigma Chemical Co., St. Louis, MO) for 3 h were seeded on dishes coated with 0.1% gelatin. Cynomolgus monkey ES cells [11] were maintained on MMC-treated MEF-coated dishes in DMEM/F12 medium (Invitrogen Corp., Carslbad, CA) supplemented with 20% heat-inactivated fetal bovine serum (FBS) of selected lots (PAA Laboratories, GmbH, Linz, Austria), 8 ng/ml of fibroblast growth factor 2 (basic; FGF2) (Invitrogen Corp.), 20 ng/ml of recombinant human bone morphogenic protein 4 (rhBMP-4) (R&D Systems Inc., Minneapolis, MN), 1 mM ß-mercaptoethanol (Sigma Chemical Co.), 1 mM L-glutamine (Invitrogen Corp.), 10 U/ml of penicillin (Invitrogen Corp.), and 10 µg/ml of streptomycin (Invitrogen Corp.). The lots of FBS were selected based on their capacity to support the proliferation of immature monkey ES cells. ES cells were passaged every 2 days by treatment with 0.25% trypsin for 1 min and were seeded at a split ratio of 1:2 to 1:4 on new MEF-coated dishes. Human leukemic UT-7 cells were maintained and induced to undergo erythroid differentiation using erythropoietin as described [12].

Preparation of the Conditioned Media of OP9 and MS-5

Murine stromal OP9 cells [5] were maintained with {alpha}-MEM medium (Invitrogen Corp.) supplemented with 20% heat-inactivated FBS (PAA Laboratories GmbH), 0.1 mM ß-mercaptoethanol (Sigma Chemical Co.), 1 mM L-glutamine (Invitrogen Corp.), 10 U/ml of penicillin (Invitrogen Corp.), and 10 µg/ml of streptomycin (Invitrogen Corp.). Murine stromal MS-5 cells, which were purchased from the German Collection of Microorganisms and Cell Cultures (Braunschweig, Germany), were maintained with {alpha}-MEM medium (Invitrogen Corp.) supplemented with 10% heat-inactivated FBS (PAA Laboratories GmbH), 2 mM L-glutamine (Invitrogen Corp.), 2 mM sodium pyruvate (Invitrogen Corp.), 10 U/ml of penicillin (Invitrogen Corp.), and 10 µg/ml of streptomycin (Invitrogen Corp.). At 60–70% confluence, these cells were subjected to 46 Gy of {gamma}-irradiation. After a wash with phosphate-buffered saline (PBS), the cells were cultured with Iscove modified Dulbecco medium (IMDM) (Invitrogen Corp.) supplemented with 15% heat-inactivated FBS (PAA Laboratories GmbH), 0.1 mM ß-mercaptoethanol (Sigma Chemical Co.), 3 mM L-glutamine (Invitrogen Corp.), 5 µM hydrocortisone, 10 U/ml of penicillin (Invitrogen Corp.), and 10 µg/ml of streptomycin (Invitrogen Corp.). After another 12-h incubation at 37°C, culture supernatants were collected and passed through Acrodisc Syringe Filters (0.2 µm, Pall Corp., Ann Arbor, MI). All the supernatants were stored at 4°C and used within 1 wk.

Hematopoietic Differentiation

For feeder-free hematopoietic differentiation, the center of each immature ES colony on MEFs was harvested using a hand-made fine capillary without contamination by MEFs. About 100 colonies were transferred into six-well type-IV collagen-coated dishes (BD Biosciences, San Jose, CA) and cultured in 5 ml of the OP9-conditioned medium (OP9-CM) supplemented with 20 ng/ml of rhVEGF, 20 ng/ml of rhBMP-4, 20 ng/ml of rh KIT ligand (rhKITLG, stem cell factor [SCF]), 10 ng/ml of rh fms-related tyrosine kinase 3 ligand (rhFLT3LG), 20 ng/ml of rh interleukin 3 (rhIL3), and 10 ng/ml of rh interleukin 6 (rhIL6). These growth factors and cytokines and their concentrations were based on the method for efficiently inducing hematopoietic differentiation of monkey ES cells [13]. An alternative and technically simpler method was also used to obtain MEF-free immature ES cells. In brief, immature ES cells were treated with 0.25% trypsin for 30–40 sec at 37°C. After mild tapping of the dish, the ES cell suspension was collected, centrifuged, and resuspended in OP9-CM supplemented with a cytokine cocktail as described above.

For hematopoietic differentiation in a coculture system, a subconfluent culture of OP9 was {gamma}-irradiated (46 Gy) in advance and ES cells were seeded on it and cultured in IMDM (Invitrogen Corp.) supplemented with 15% heat-inactivated FBS (PAA Laboratories GmbH), 0.1 mM ß-mercaptoethanol, (Sigma Chemical Co.), 3 mM L-glutamine (Invitrogen Corp.), 5 µM hydrocortisone, 10 U/ml of penicillin (Invitrogen Corp.), and 10 µg/ml of streptomycin (Invitrogen Corp.) in the presence of 20 ng/ml of rhVEGF, 20 ng/ml of rhBMP-4, 20 ng/ml of rh KIT ligand (rhKITLG, stem cell factor [SCF]), 10 ng/ml of rh fms-related tyrosine kinase 3 ligand (rhFLT3LG), 20 ng/ml of rh interleukin 3 (rhIL3), and 10 ng/ml of rh interleukin 6 (rhIL6).

Morphological Examination

The viable cells were directly observed under an inverted phase-contrast light microscope (Olympus Optical Co. Ltd., Tokyo, Japan), or alternatively, cells were fixed on glass slides using a cytospin apparatus (Cytospin 2, SHANDON, Pittsburgh, PA), stained with Wright-Giemsa solution (Muto Pure Chemical Co., Tokyo, Japan), and then observed under a light microscope (Olympus Optical Co. Ltd.).

Examination of Cell-Surface Markers by Fluorescence-Activated Cell Sorting

Cell-surface markers analyzed by FACS were as follows: fucosyltransferase 4 (FUT4), also known as stage-specific embryonic antigen (SSEA)-1, SSEA-4, CD34 antigen (CD34), protein tyrosine phosphatase, receptor type, C (PTPRC, CD45), cadherin 5, type 2 (CDH5, vascular endothelial-cadherin; VE-cadherin), platelet/endothelial cell-adhesion molecule (CD31 antigen) (PECAM1, CD31), melanoma cell-adhesion molecule (MCAM, CD146), KDR, v-kit Hardy-Zuckerman 4 feline sarcoma viral oncogene homolog (KIT, CD117) and prominin 1 (PROM1, CD133). Hematopoietic cells were collected by gentle pipetting and adherent cells were collected by either trypsinization (for studies of the expression of FUT4, SSEA-4, CD34, and PTPRC) or 0.2% EDTA treatment (for studies of the expression of CDH5, PECAM1, MCAM, KDR, KIT, and PROM1). After a wash in PBS, 1 x 106 cells were reacted with first antibodies on ice for 30 min. In the cases of FUT4, SSEA-4, and CDH5, secondary antibody reactions were performed subsequently. The expression level of each protein was analyzed using a FACSCalibur (BD Biosciences). The primary antibodies used were mouse anti-human FUT4 (Chemicon International, Inc., Temecula, CA), mouse anti-human SSEA-4 (Chemicon International, Inc.), mouse anti-human CD34-R-phycoerythrin (PE; BD Biosciences), mouse anti-human PTPRC-FITC (BD Biosciences), mouse anti-human CDH5 (Chemicon International, Inc.), mouse anti-human KIT-PE (BD Biosciences), mouse anti-human KDR-PE (BD Biosciences), mouse anti-human PECAM1-FITC (Caltag Laboratories, An-Der-Grub, Austria), mouse anti-human MCAM-PE (Chemicon International, Inc., Temecula, CA), and mouse anti-human PROM1-PE (Miltenyi Biotech GmbH, Bergisch Gladbach, Germany). The secondary antibody used for the FUT4 expression study was a PE-labeled goat anti-mouse IgM (ICN Biomedicals, Inc., Aurora, OH) antibody and that for SSEA-4 and CDH5 was PE-labeled goat anti-mouse IgG antibody (ICN Biomedicals, Inc.).

Immunostaining

Cell-surface markers analyzed by immunostaining were as follows: CD34, PTPRC, CD14 antigen (CD14), and PECAM1. The cells were fixed on slide glasses with a cytospin apparatus (Cytospin 2) with further fixation with acetone/methanol solution (1:3). The immunostaining procedure was performed as described elsewhere [14] using mouse anti-human CD34-PE (BD Biosciences), mouse anti-human PTPRC-FITC (BD Biosciences), mouse anti-human CD14-FITC (Nichirei Bioscience Inc., Tokyo, Japan), and mouse anti-human PECAM1-FITC (Caltag Laboratories, An-Der-Grub, Austria) antibodies.

Cord Formation Assays

Matrigel (BD Biosciences) was loaded into the 24 multiwell dishes (95 µl/well). After the dishes were incubated for 30 min at 37°C, 1 x 104 cells per well were seeded in 1 ml of EGM-2 BulletKit medium supplemented with cytokines and growth factors according to the manufacturer's instructions (Takara Shuzo Co. Ltd., Shiga, Japan). The cobblestone cells and hematopoietic cells were treated with trypsin/EDTA solution, completely disaggregated, and subjected to the cord formation assay on Matrigel in 24 multiwell dishes. Cell morphology was observed under an inverted phase-contrast light microscope (Olympus Optical Co. Ltd.).

Nitroblue Tetrazolium-Reducing Activity

The respiratory burst activity of the cells was assessed by measuring nitroblue tetrazolium (NBT)-reducing activity. Five hundred thousand cells were collected, washed with PBS once, and suspended in 1 ml of differentiation medium. Next, 1 ml of NBT solution (consisting of 1 mg of NBT [Nacalai Tesque Inc., Tokyo, Japan] dissolved in 1 ml of differentiation medium at 37°C for 5 min with vigorous mixing) was added and the cells were incubated for 25 min at 37°C in the presence of 100 ng/ml phorbol myristate acetate to induce the respiratory burst of the cells. Then the cells were washed with PBS once and resuspended in 10 µl of PBS. The cell suspension was dropped on glass slides and covered with thin glass cover slips. The number of NBT-positive cells was counted under a light microscope (Olympus Optical Co. Ltd.).

Colony Assays

Colony assays were performed using Methocult TM GF+H4535 (Stemcell Technologies Inc., Vancouver, BC, Canada) according to the manufacturer's recommendations. In brief, 0.3 ml of cell suspension, which contained 10 cells, was mixed in 3 ml of methylcellulose solution consisting of 1% methylcellulose, 30% FBS, 1% bovine serum albumin, 0.1 mM ß-mercaptoethanol, 2 mM L-glutamine, 50 ng/ml rhKITLG, 20 ng/ml rh colony stimulating factor 2 (granulocyte-macrophage) (rhCSF2), 20 ng/ml rhIL3, 20 ng/ml rhIL6, and 20 ng/ml rh colony stimulating factor 3 (granulocyte; rhCSF3) in 3.5-cm culture dishes. After 2 wk, the number of colonies was counted. The morphology of the colonies was observed under an inverted light microscope (Olympus Optical Co. Ltd.).

Western Blotting

Western blotting was performed as described previously [14] using rabbit anti-human T-cell acute lymphocytic leukemia 1 (TAL1) antibody (H-20) (Santa Cruz Biotechnology Inc., Santa Cruz, CA) and rabbit anti-human KDR antibody (Santa Cruz Biotechnology Inc.).

Reverse Transcription-Polymerase Chain Reaction

RNA was extracted from 5 x 106 cells with an RNeasy Mini Kit (Qiagen K. K., Tokyo, Japan), and cDNA was synthesized from 250 ng of total RNA using a Superscript II Kit (Invitrogen Co., Carlsbad, CA) in a final reaction volume of 20 µl according to the manufacturer's protocol. The PCR procedure was performed using 2.5 µl of the reverse transcription reaction product as a template and Ex-Taq (Takara Shuzo Co. Ltd., Shiga, Japan) and a DNA Thermal Cycler PJ2000 (Perkins-Elmer Corp., Foster City, CA) with the following program: 98°C, 10 sec; 57°C, 30 sec; and 72°C, 45 sec for 40 cycles. The primers used were as follows: GATA binding protein 1 (globin transcription factor 1) (GATA1), 5'-gcctacagacactccccagt-3' (sense) and 5'-ctgtccctccgccacagt-3' (antisense); hemoglobin, 1 (HBE1), 5'-TGCATTTTACTGCTGAGGAGA-3' (sense) and 5'-AAGAGAACTCAGTGGTACTT-3'' (antisense); hemoglobin, {zeta} (HBZ), 5'-TTCCTCAGCCACCCGCAGAC-3' (sense) and 5'-AGCAGGCAGTGGGACAGGAG-3' (antisense); actin, ß (ACTB), 5'-gcaggagatggccacggcgcc-3' (sense) and 5'-tctccttctgcatcctgtcggc-3' (antisense). The products were separated by agarose gel electrophoresis and the DNA was visualized by ethidium bromide staining.

Uptake of Acetylated Low-Density Lipoprotein (Ac-LDL)

Cells were transferred into a 4-well chamber slide system (Nalge Nunc International Corp., Naperville, IL). After overnight culture, the cells were washed twice with Hanks balanced salt solution and incubated in serum-free medium containing 10 µg/ml of low-density lipoprotein from human plasma, acetylated, DiI complex (DiI Ac-LDL) (Molecular Probes, Eugene, OR) for 4 h. After the cells were washed with HBSS three times, they were observed under a fluorescence microscope (Olympus Optical Co. Ltd.).

RESULTS

Maintenance of Immature ES Cells and the Protocols for Starting Feeder-Free ES Differentiation

Cynomolgus monkey ES cells were maintained on MEFs. They formed flat round colonies consisting of cells with a high nucleus/cytoplasm ratio (Fig. 1, A and B). The fact that they showed high expression of SSEA-4 and low expression of FUT4 (also known as SSEA-1) (Fig. 1, C and D) indicates that the majority of the ES cells were maintained in the undifferentiated state.


Figure 1
View larger version (99K):
[in this window]
[in a new window]
  		FIG. 1. Morphology and surface-marker analysis of cynomolgus monkey ES cells. A, B) Phase-contrast microscopy of immature cynomolgus monkey ES cells cultured on MEFs. C, D) Cell-surface-marker analysis by FACS. The expression of an undifferentiated primate ES cell marker (SSEA-4) (C) and a differentiated marker (FUT4, also known as SSEA-1) (D) was examined. The black lines represent the results of the isotype control immunoglobulin reactions followed by the secondary antibody reactions in each case. The green lines represent the results of the anti-SSEA-4 (C) and anti-FUT4 (D) antibody reactions followed by the secondary antibody reactions. Original magnification x40 (A) and x100 (B). Bars = 40 µm

During our initial study, we found that the conditioned medium of murine bone marrow stromal OP9 cells, which are commonly used as feeders to induce hematopoietic differentiation of ES cells in a coculture system, could induce hematopoietic differentiation of monolayer cultures of cynomolgus monkey ES cells in the presence of a cytokine cocktail (20 ng/ml rhVEGF, 20 ng/ml rhBMP4, 20 ng/ml rhKITLG, 10 ng/ml rhFLT3LG, 20 ng/ml rhIL3, and 10 ng/ml rhIL6), although at low levels. On the basis of a search for ways to increase the differentiation efficiency, we found that there were four important points for effective differentiation into hematopoietic lineages.

First, the presence of MEFs, even a small number of them, significantly stimulated the growth of undifferentiated ES cells and strongly inhibited the hematopoietic differentiation. To exclude the MEFs from the starting materials, the very centers of ES colonies were manually picked with fine capillaries under a light microscope. This technically difficult method to obtain MEF-free immature ES cells, although capable of completely excluding MEF, could be replaced by a more simplified method using trypsin treatment followed by mild tapping of the dish (as described in Materials and Methods) because MEF contamination was almost completely ruled out by this simple method. Second, the cell density significantly affected the fate of differentiating ES cells: low cell numbers (<approximately 50 cells per well in a six-well-plate) led to the generation of spindle-like cells without proliferation activity, whereas high cell numbers (>approximately 200 cells per well in a six-well-plate) potentiated the proliferation of undifferentiated ES cells, while intermediate cell numbers (around 100 cells per well in a six-well-plate) promoted hematopoietic differentiation. Third, the size of ES cell aggregates affected the fate of differentiating ES cells: large cell aggregates (>approximately 20 cells per aggregate) biased the fate toward proliferation of undifferentiated ES cells and small cell aggregates (<approximately 4 cells per aggregate) biased the fate toward death of ES cells, while intermediate cell aggregates (5 to ~10 cells per aggregate) biased the fate toward hematopoietic differentiation. Last, we found that the hematopoietic differentiation capacity of OP9-CM waned within a few days.

On the basis of these findings, we established a protocol for the most efficient hematopoietic differentiation of feeder-free monolayer ES cells: undifferentiated ES colonies without contamination of MEFs were appropriately disaggregated by pipetting (5 to ~10 cells per aggregate), seeded into type-IV collagen-coated dishes at appropriate cell density (around 100 cells per well in a six-well-plate), and cultured in the fresh OP9-CM supplemented with rhVEGF, rhBMP4, rhKITLG, rhFLT3LG, rhIL3, and rhIL6.

Nonfeeder Differentiation of ES Cells Toward Hematopoietic Blast Cells via Finger-Like Cells

During the first few days of the differentiation culture, ES cells remained quiescent and some of them died. Then the remaining cells began to proliferate and produced a unique population of differentiating ES cells at 3–7 days. Because these cells morphologically resembled fingers, we call them finger-like cells (Fig. 2A). During our initial trial-and-error studies, we had found that the emergence of finger-like cells at this early phase was essential for the subsequent hematopoietic differentiation (23 experiments with finger-like cells and with hematopoietic differentiation versus 11 experiments without finger-like cells and without hematopoietic differentiation out of a total of 34 differentiation-inducing experiments). These cells actively proliferated and spread over the bottom surface of the culture dish in about 1 wk. At subconfluence, round cells began to emerge as if they had budded from the finger-like cells (Fig. 2B). After another 2 wk, the surface of the culture dish was filled with these round cells (Fig. 2C). We collected the round cells by gentle pipetting and examined their morphology by Wright-Giemsa staining. These cells showed a high nuclear/cytoplasmic ratio, two or three dark nucleoli, and dark blue cytoplasm without granule formation (Fig. 2D), all of which are known to be characteristics of hematopoietic blast cells.


Figure 2
View larger version (51K):
[in this window]
[in a new window]
  		FIG. 2. Differentiation of cynomolgus monkey ES cells into hematopoietic blast cells. A) The emergence of the finger-like cells starting several days after the induction of differentiation. Original magnification x100 and bars = 40 µm. B) The emergence of round cells on the finger-like cells about 1 wk after further culturing of the cells shown in A. C) Abundant production of the round cells after another week of culturing of the cells shown in B. D) Wright-Giemsa stain. The round cells in C were collected by gentle pipetting, fixed on a slide glass, and stained with Wright-Giemsa solution. Original magnification x400. E, F) Cell-surface-marker analysis by FACS. Total cells were collected by gentle pipetting and were then disaggregated by EDTA treatment. Then the expression of CD34 (E) and PTPRC (F) were examined. The black lines represent the results of isotype control IgG staining and the green lines represent those of the anti-CD34 or anti-PTPRC antibody staining. The percentages of CD34- and PTPRC-expressing cells were statistically analyzed from 13 independent experiments. G) Immunostaining. The round cells were collected by gentle pipetting, fixed on a glass slide with acetone/methanol solution, and stained with anti-CD34 or anti-PTPRC antibody as indicated (upper). The lower panels show the results of staining with the isotype control antibody. H, I) Phase-contrast microscopy of the primary and secondary colonies formed in methylcellulose semi-solid culture. J, K) Hematopoietic differentiation in the conventional OP9 coculture system. Cynomolgus monkey ES cells were cultured on confluent layers of irradiated OP9 cells in the presence of a cytokine cocktail. After 3 wk, the ES-derived cells were collected by gentle pipetting and subjected to FACS analysis. The expressions of CD34 (J) and PTPRC (K) were examined. The black lines represent the results of isotype control IgG staining and the green lines represent those of anti-CD34 or anti-PTPRC antibody staining as indicated. Similar results were obtained from three independent experiments and representative results of one experiment are shown

To confirm the hematopoietic blast cell differentiation of ES cells, we first tested the hematopoietic commitment by performing surface-marker analysis by FACS. Total cells were collected from the dish by gentle pipetting, and the expressions of CD34, a marker of stem cells of various tissues, including hematopoietic, neural, and endothelial cells, and PTPRC, a hematopoietic lineage-specific marker, were examined. As shown in Figure 2, the majority of the cells (86.8% ± 6.3%, n = 13) were positive for CD34 and more than two thirds of the cells (67.2% ± 13.8%, n = 13) were positive for PTPRC (Fig. 2, E and F). These results were further confirmed by immunostaining (Fig. 2G).

We next investigated whether the ES-derived hematopoietic cells had blast cell activities. The round cells were collected with the least-possible pipetting and subjected to colony-forming assays in methylcellulose semisolid culture in the presence of rhKITLG, rhCSF2, rhIL3, rhIL6, and rhCSF3. We found that the cells yielded small, slightly reddish compact colonies compatible with blast cell colonies [9, 15] with very high efficiency (80.6% ± 18.9%, n = 16) (Fig. 2H). Moreover, these cells generated secondary blast cell colonies, which resembled the primary colonies (Fig. 2I). All these findings together indicate that our feeder-free, monolayer culture system effectively generated hematopoietic blast cells with high self-renewal activities. Indeed, 1 x 102 undifferentiated ES cells yielded approximately 1 x 106 hematopoietic cells in about 4 wk.

Then, we compared the hematopoietic differentiation capacities between our system and the conventional OP9 coculture system [5]. As shown in Figure 2, J and K, the latter conventional method generated about 80% CD34-positive and 60% PTPRC-positive cells. Thus, our feeder-free monolayer culture system produces a comparable, or even better, outcome compared with the conventional coculture system for the generation of hematopoietic blast cells.

Molecular Analysis of Hematopoietic Blast Cells

To evaluate the differentiation stages of PTPRC-positive hematopoietic blast cells in our culture system, we studied the level of TAL1, a transcriptional factor that is expressed in hemangioblasts and hematopoietic stem cells and is closely associated with erythroid differentiation [16, 17, 18], in hematopoietic blast cells as compared with that in immature ES cells and finger-like cells. As shown in Figure 3A, both finger-like cells and hematopoietic blast cells but not immature ES cells expressed TAL-1 protein, suggesting that the blast cells had immature hematopoietic features with the potential for erythroid differentiation.


Figure 3
View larger version (11K):
[in this window]
[in a new window]
  		FIG. 3. Molecular characterization of hematopoietic blast cells. Differentiated ES cells like those shown in Figure 2C were analyzed for the expression of hematopoietic genes and/or proteins including, TAL1, GATA1, HBE, and HBZ. A) Western blotting was performed using rabbit anti-human TAL1 antibody in undifferentiated ES cells (lane 1), finger-like cells (lane 2), and hematopoietic cells (lane 3). B) RT-PCR was performed using primer pairs specific for GATA1 and ACTB in undifferentiated ES cells (lane 1), hematopoietic cells (lane 2), human erythroblastic K562 cells (lane 3), and human megakaryoblastic UT-7 cells (lane 4). C) RT-PCR was performed using primer pairs specific for HBE1, HBZ, and ACTB in undifferentiated ES cells (lane 1), hematopoietic blast cells (lane 2), and human erythroblastic K562 cells (lane 3)

To further confirm the hematopoietic potential of blast cells, we studied the expression of GATA1, a hematopoietic transcription factor critical for both erythroid and megakaryocytic differentiation [19]. As shown in Figure 3B, blastic cells express GATA1 at a level comparable with human erythroblastic K562 cells and human megakaryoblastic UT7 cells. On the other hand, immature ES cells did not express GATA1. To evaluate the erythropoietic potential of the blast cells, we then studied the expression level of globin genes in these cells. As shown in Figure 3C, the blast cells did not express the HBE1 gene and only marginally expressed the HBZ gene, whereas erythroblastic K562 cells expressed both HBE1 and HBZ genes abundantly. Thus, immature blast cells, while potentially erythropoietic, were maintained in an immature state at least during the short-term colony-forming assay.

These findings together suggest that PTPRC-positive blastic cells are very immature hematopoietic progenitor cells with differentiating potential toward the erythroid and megakaryocytic lineages.

Macrophage Induction from the Hematopoietic Blast Cells

To further evaluate the hematopoietic differentiation capacities of the ES-derived PTPRC-positive cells in our system, the cell samples shown in Figure 2C were further cultured in the conditioned medium of MS-5, a murine bone marrow stromal cell line with strong ability to support the production of mature hematopoietic cells [2], supplemented with a cytokine cocktail (20 ng/ml rhVEGF, 20 ng/ml rhSCF, 10 ng/ml rhFLT3LG, 20 ng/ml rhIL3, 10 ng/ml rhIL6, and 20 ng/ml rhCSF3). After a 2- or 3-wk culture, the round cells began to detach from the dishes en masse in a sheet-like form (Fig. 4, A and B, regions with red arrows). These detached cells were collected and morphologically examined. As shown in Figure 4C, they showed morphological characteristics of monocyte/macrophages with large irregular cytoplasm and eccentrically placed nuclei. Moreover, they were positive for CD14 (Fig. 4D), a component of the lipopolysaccharide receptor complex that is known to be a monocyte/macrophage-specific marker. For the functional analysis, we performed the NBT-reducing assay, which measures the superoxide-producing activity, a specific functional parameter of mature phagocytic leukocytes [20, 21]. As shown in Figure 4E, the ES-derived macrophages (M{Phi}) were positive for NBT-reducing activity. Thus, our feeder-free monolayer culture system effectively produced hematopoietic blast cells that can differentiate into mature functional macrophages.


Figure 4
View larger version (51K):
[in this window]
[in a new window]
  		FIG. 4. Monocyte/macrophage differentiation of the cynomolgus monkey ES cells. Differentiated ES cells like those shown in Figure 2C were further cultured in MS-5-CM supplemented with a cytokine cocktail. After about a 2-wk culture, some populations of the cells were detached en masse from the dish. A, B) Phase-contrast microscopy. The adhesiveness of the differentiated round cells, which were located at the marginal region of the colonies, became weak and some of them detached from the surface of the dish and disappeared from this microscopic field (shown by thin black arrows). The remaining round cells, which were located just inside the region from which the round cells had detached (shown by bold red arrows), were collected by gentle pipetting and subjected to morphological examination (C), immunostaining (D), and NBT-reducing assay (E). After pipetting, there were still round cells piled on the finger-like cells in more inner areas (shown by bold black arrows), which would differentiate and be collected by gentle pipetting after another few days of culture. C) The detached cells were collected, disaggregated by minimal trypsin/EDTA treatment, fixed on a glass slide, and stained with Wright-Giemsa solution. The cells showed large cytoplasm, occasionally forming vacuoles, with eccentric nuclei, features compatible with macrophages (M{Phi}). D) Immunostaining. The detached cells fixed on a glass slide as in C were stained with anti-CD14 antibody (upper). The lower panel shows the results of staining with the isotype control antibody. E) The detached cells were collected and NBT-reducing activities were examined. The cells were observed under an inverted light microscope. The NBT-reduction-positive cells contained dark purple granules in the cytoplasm, indicating superoxide-producing activity, a functional marker of macrophages (M{Phi}) (upper). For a negative control, human leukemic HL-60 cells were similarly subjected to the NBT-reducing assay (lower). Original magnification x40 (A) and x100 (B). Bars = 40 µm

Endothelial Cell Differentiation of Finger-Like Cells

To assess the possibility that ES-derived hematopoietic cells are originated from hemangioblasts or related cells, the angioblastic activities of the finger-like cells were tested. During our study, we incidentally found that, after transfer onto fresh collagen-coated dishes, the morphology of the finger-like cells changed into a cobblestone-like appearance (Fig. 5A), a characteristic of the endothelium [22].


Figure 5
View larger version (58K):
[in this window]
[in a new window]
  		FIG. 5. The spontaneous endothelial differentiation of the finger-like cells after trypsinization and transfer onto fresh dishes. The exponentially growing finger-like cells were trypsinized and transferred onto fresh collagen-coated dishes at a 1:3 dilution. The cells were further cultured in the same differentiation medium supplemented with the same cytokine cocktail. A) Phase-contrast microscopy of the transferred cells after 5 days. Note that the morphology of the cells had acquired a cobblestone appearance characteristic of endothelial cells. B–E) Cell-surface-marker analysis. The cobblestone cells in A were collected by vigorous pipetting followed by EDTA treatment. In some experiments, the cells in A were cultured for two passages in EGM-2 medium and were then collected similarly (D, E, and G). The expressions of CD34 (B), PTPRC (C), CDH5 (D), and PECAM1 (E) were examined by FACS analysis. The black lines represent the results of isotype control IgG staining and the green lines represent those of specific antibodies as indicated. Similar results were obtained from three independent experiments and the representative results of one experiment are shown in each case. F) Immunostaining. The cobblestone cells were collected by gentle pipetting, fixed on a glass slide with acetone/methanol solution, and stained with mouse anti-human PECAM1 antibody as indicated (upper). The lower panels show the results of staining with the isotype control antibody. G) Cell-surface-marker analysis. The cobblestone cells in A were collected by vigorous pipetting followed by EDTA treatment. In some experiments, the cells in A were cultured for two passages in EGM-2 medium and were then collected similarly (lower panel). The expression of MCAM was examined by FACS analysis. The black lines represent the results of isotype control IgG staining and the green lines represent those of specific antibodies as indicated. Similar results were obtained from three independent experiments and the representative results of one experiment are shown in each case. H) Ac-LDL uptake of cobblestone cells and finger-like cells with hematopoietic cells. I) The cord formation assay of cobblestone cells and hematopoietic cells. Original magnification x100 and bars = 50 µm in A, H, and I

These cobblestone cells were negative for CD34 (Fig. 5B), an immature cell-specific antigen, and PTPRC (Fig. 5C), a hematopoietic cell-specific antigen, and scarcely formed hematopoietic blast cell colonies (plating efficiency: 0.04 in cobblestone cells versus 0.8 in hematopoietic cells), suggesting that they had a relatively mature phenotype without hematogenic potential. Although these cobblestone cells were negative for CDH5 (Fig. 5D), a well-known marker of endothelial cells, such a phenomenon has already been reported in endothelial cells induced from monkey ES cells [23]. In addition, CDH5 became weakly but significantly positive in cobblestone cells after two passages in EGM-2 medium, a culture medium optimized for endothelial cell culture (Fig. 5E).

To confirm the endothelial nature of cobblestone cells, we studied the expression of PECAM1, a specific marker of mature endothelium. As shown in Figure 5F, about half of the cobblestone cells were positive for the cell-surface expression of PECAM1 whereas the finger-like cells and hematopoietic cells were not. The immunostaining study further showed that almost all of the cobblestone cells were positive for the total expression of PECAM1, including its cytoplasmic storage as well as its surface expression (Fig. 5G). We also found that cobblestone cells were highly positive for MCAM (Fig. 5H), another endothelial marker. Thus, the majority of the finger-like cells differentiated into endothelial lineages after transfer onto new dishes by an unknown mechanism.

To evaluate the functional capacity of cobblestone cells as endothelial cells, we performed the following two assays: Ac-LDL uptake and cord formation. As shown in Figure 5I, the cobblestone cells showed marked uptake of Ac-LDL in the EGM-2 medium, whereas finger-like cells and hematopoietic cells did not. In addition, the cobblestone-like cells eventually showed cord formation in a matrigel differentiation assay under the conditions used, whereas hematopoietic cells did not (Fig. 5J). These findings further confirm the endothelial nature of the cobblestone cells.

Thus, the ES-derived finger-like cells have the capacity to differentiate into the endothelial lineage in addition to the hematopoietic lineage.

Characterization of Finger-Like Cells

The hematopoietic cells are believed to be originated from either the hematogenic endothelium, which is positive for CDH5 [24], or more immature mesodermal progenitor populations including hemangioblasts, the common progenitors of the endothelial and hematopoietic lineages [15, 25, 26]. To assess whether the ES-derived hematopoietic blast cells generated in our system were originated from the hematogenic endothelium, cell-surface-marker analysis of the finger-like cells was performed. We found that the finger-like cells were negative for CDH5 (Fig. 6A). On the other hand, the finger-like cells contained fractions that were positive for several tissue stem cell markers, including CD34, KIT, PROM1, and KDR (Fig. 6A), all of which were negative or almost negative in undifferentiated ES cells (Fig. 6B). Among these molecules, KDR, a mesodermal or hemangioblastic marker, was also positive in cobblestone cells and hematopoietic cells (Fig. 6C). Similar findings for KDR expression during the differentiation process were made by Western blotting (Fig. 6D). In this assay, KDR expression was detectable in immature ES cells, consistent with the previous results showing that KDR was expressed in primate ES cells [2]. Thus, KDR was upregulated during the differentiation of immature ES cells toward finger-like cells and then downregulated during downstream pathway of differentiation into hematopoietic cells and endothelial cells. These findings together suggest that the hematopoietic blast cells did not originate from the hematopoietic endothelium but rather from mesodermal cells or hemangioblasts. Taken together with the expression levels of TAL1 (Fig. 3), a hemangioblastic marker, these findings suggest that the finger-like cells in our culture system correspond to hemangioblasts.


Figure 6
View larger version (26K):
[in this window]
[in a new window]
  		FIG. 6. Cell-surface-marker and Western blot analysis of the finger-like cells. A) The finger-like cells were collected 10 days after the induction of differentiation and subjected to FACS analysis. The expressions of CDH5, KDR, KIT, CD34, PROM1, and PTRPC were examined. The black lines represent the results of isotype control IgG staining and the green lines represent those of the specific antibodies as indicated. B) The undifferentiated ES cells were subjected to FACS analysis. The expressions of KDR, KIT, CD34, and PROM1 were examined. C) Hematopoietic cells and cobblestone cells were subjected to FACS analysis and the expression of KDR was examined. D) Western blot analysis of KDR was performed for undifferentiated ES cells (lane 1), finger-like cells (lane 2), hematopoietic blast cells (lane 3), and cobblestone cells (lane 4)

Repeated High Hematopoietic Production Potential of Finger-Like Cells

Next, we studied the fate of the finger-like cells that remained on the original dish after trypsinization. As shown in Figure 7A, a small number of finger-like cells remained on the dish even after trypsin treatment (0.25% trypsin for 30–40 sec at 37°C). Fresh OP9-CM supplemented with cytokine cocktail was added and the cells were further cultured. The finger-like cells actively proliferated (Fig. 7B) and massive numbers of hematopoietic cells were generated again after about 2 wk (Fig. 7C) with CD34 and PTPRC expression levels (Fig. 7D) comparable with those observed before (Fig. 2, E and F). This phenomenon was repeated again, and as a result, 100 undifferentiated ES cells yielded approximately 5 x 106 hematopoietic cells during about 2 mo. On the other hand, the cells transferred onto fresh dishes always differentiated into endothelial cells whether the dishes were coated with collagen or not (data not shown). All these findings together indicate that the conditions that were originally established during the generation and proliferation of the finger-like cells were appropriate for the self-renewal and subsequent hematopoietic differentiation of the finger-like cells, blocking their tendency to differentiate into the endothelial lineage.


Figure 7
View larger version (75K):
[in this window]
[in a new window]
  		FIG. 7. Repetitive production of hematopoietic cells in the liquid culture in the original dish. A–C) Phase-contrast microscopy. Trypsin treatment of either the subconfluent culture of the finger-like cells or the confluent culture of the finger-like cells containing abundant hematopoietic blast cells resulted in massive removal of the cells, and only a small number of the finger-like cells remained on the dish (A). The dish was again filled with the differentiation medium supplemented with a cytokine cocktail as before. Then the remaining finger-like cells began to proliferate (B), and after about 2 wk, the dish was again filled with abundant hematopoietic blast cells (C). D) The results of the FACS analysis indicated that CD34 and PTPRC expression levels were similar to those obtained before (Fig. 2, E and F). Original magnification x100 and bars = 40 µm

To explain these observations, we hypothesized the existence of a certain population of finger-like cells that had differentiated into hematopoiesis-supporting stromal cells attaching in a trypsin-resistant manner. In fact, large, round stroma-like cells with abundant cytoplasm, morphologically distinct from finger-like cells, were observed in the finger-like cell culture (Fig. 8A) and these cells remained on the dishes even after trypsin-treatment (Fig. 8B). In addition, collection of total cells (not only finger-like cells but also stroma-like cells) from the original dishes by vigorous pipetting with scraping and their total transfer to another dish resulted in the proliferation of finger-like cells but not cobblestone cells (Fig. 8C). Also in these secondary dishes, we could observe large stroma-like cells among the finger-like cells (Fig. 8C). This phenomenon might explain at least in part the reason why hematopoietic cells differentiated only from the original dish and why endothelial cells differentiated only from the fresh dishes.


Figure 8
View larger version (95K):
[in this window]
[in a new window]
  		FIG. 8. Presence of stroma-like cells in finger-like cell culture. A–C) Phase-contrast microscopy. Larger, flat stroma-like cells were observed in the finger-like cell culture (pink arrowhead) (A). These stroma-like cells were still present on the dishes even after harvesting of the finger-like cells by trypsin-treatment (B). The total collection of the cells (not only finger-like cells but also stroma-like cells) from the original dishes by vigorous pipetting without trypsin treatment and their total transfer to another dish resulted in the proliferation of finger-like cells along with the production of hematopoietic cells but not the emergence of cobblestone cells, and large stroma-like cells were also observed in these secondary dishes (pink arrow head) (C). Bars = 50 µm

Prolonged cultivation of finger-like cells with hematopoietic cells for more than 2 mo did not affect the speed of proliferation or the morphology of the cells. These cells, however, lost their capacity to differentiate into macrophages (data not shown). In contrast, prolonged cultivation did not attenuate but rather potentiated the capacity of the hematopoietic blast cells for erythroid differentiation. In the presence of erythropoietin, hematopoietic blast cells differentiate into erythroblasts of typical morphology with fetal globins (Fig. 9). Thus, the pluripotency of hematopoietic blasts decreased during prolonged culture of the finger-like cells and hematopoietic blast cells.


Figure 9
View larger version (39K):
[in this window]
[in a new window]
  		FIG. 9. Erythroid differentiation of hematopoietic blast cells. Prolonged cultivation of the finger-like cells with hematopoietic blast cells showed an apparent tendency of the hematopoietic blast cells to differentiate into erythroid hematopoietic cells. A) Wright-Giemsa stain. In the presence of erythropoietin, hematopoietic cells differentiated into erythroblasts with typical morphology. Bars = 20 µm. B) RT-PCR was performed using primer pairs specific for HBE1, HBZ, and ACTB in undifferentiated ES cells (lane 1), finger-like cells with (lane 2) or without (lane 3) erythropoietin, hematopoietic blast cells with (lane 4) or without (lane 5) erythropoietin, and human megakaryoblastic UT-7 cells with (lane 6) or without (lane 7) erythropoietin. Doublet bands of HBZ in lane 2 and 3 do not correspond to the expected PCR product size and were therefore considered to be nonspecific

DISCUSSION

We reported here an outstanding culture protocol for efficient hematopoietic and endothelial differentiation of feeder-free, monolayer primate ES cells without the need for cell sorting. The complete elimination of MEFs from the starting materials and the strict control of the number and size of cell aggregates guarantees differentiation of the ES cells into hematopoietic/endothelial precursors. It is also important to maintain ES cells, MEFs, and OP9 cells in their best conditions. In our protocol, ES cells first differentiated into a very unique cell population with a finger-like appearance after 3–7 days' culture in fresh OP9-CM supplemented with cytokine cocktail. These finger-like cells produced large amounts of hematopoietic blast cells. Even after massive removal by trypsinization, the finger-like cells actively and repetitively produced hematopoietic cells as long as they were maintained on the original dish. In contrast, the majority of the finger-like cells differentiated into endothelial cells after transfer to new dishes. These endothelial cells eventually showed cord formation in a matrigel differentiation assay, a specific indicator of functional maturation of the endothelial cells. All these features of our differentiation system are summarized in Figure 10.


Figure 10
View larger version (22K):
[in this window]
[in a new window]
  		FIG. 10. Schematic presentation of our feeder-free, monolayer differentiation process. CM: Conditioned medium; CKs, cytokines

In our monolayer culture system, we could clearly observe individual cells and this allowed us to identify for the first time the finger-like cells in the early phase of differentiation. The emergence of the finger-like cells is essential for the later hematopoietic and endothelial differentiation and a large proportion of the finger-like cells must possess the bidirectional differentiation potential. In this sense, the finger-like cells might possibly be considered as hemangioblasts, common precursors of hematopoietic and endothelial cells, although clonal analysis could not be performed due to severe loss of the hematopoietic-differentiating capacity of the finger-like cells as a result of cell-detaching procedures (unpublished results).

On the other hand, molecular characterization and cell-surface-marker analysis clarified that the finger-like cells were positive for KDR, KIT, and TAL1. This phenotype corresponds to that of mesodermal cells, particularly of hemangioblasts [5, 17, 18, 24, 25, 26]. Despite a substantial body of evidence for the in vivo existence of hemangioblasts, such as anatomical findings during embryogenesis [27], results of gene-targeting animal experiments [28] and clonogenic assays of human bone marrow hematopoietic stem cells [29], the characteristics of hemangioblasts have not been fully determined. The EB-derived blast-colony-forming cells (BL-CFCs) in semisolid cultures are considered to be the in vitro equivalents of hemangioblasts [15, 25], and critical roles of the transcription factors such as TAL1 [16] in the generation of BL-CFCs have been shown. As for the surface markers, it was reported that a certain population of CD34posKDRpos cells from human bone marrow contains hemangioblasts [29]. Indeed, the population of finger-like cells contains CD34-positive fractions as well as KDR-positive fractions (Fig. 4). It would be of great interest to elucidate by cell sorting what fraction of the finger-like cells contains the hemangioblast activity. However, finger-like cells that are forced to detach from the dish largely lose the capacity to generate hematopoietic cells, as described above. Future analysis with more sophisticated methodology that can be applied in situ with the precise monitoring of each cell fate for a sufficiently long time (more than 2 wk) will clarify this matter. In any case, our novel differentiation method will be highly useful for advancing our understanding of hemangioblasts as well as early-phase hematopoiesis.

It is also of great interest to know what decides the destiny of the finger-like cells. There are three possibilities regarding their fate: self-renewal, hematopoiesis, and endothelial differentiation. The reason why hematopoietic cells differentiated only from the original dish and why endothelial cells differentiated only from the fresh dishes may provide a hint. A certain population of finger-like cells might have differentiated into hematopoiesis-supporting stromal cells attaching in a trypsin-resistant manner. Alternatively, certain cell-surface molecules that are critical for the hematopoietic differentiation might have been destroyed during cell transfer processes and thus the cells were forced into endothelial differentiation. In this study, we obtained some evidence for the former possibility (Fig. 8), although further characterization of putative hematopoiesis-supporting stromal cells in the dishes with finger-like cells will be necessary for a better understanding of the mechanisms of self-renewal, hematopoiesis, and endothelial differentiation of the finger-like cells in our culture system.

Cobblestone cells were positive for endothelial marker expression, cord formation assay, and Ac-LDL uptake, all of which strongly suggest the endothelial nature of these cells. In contrast, cobblestone cells were completely negative for CD34 and PTPRC, clearly indicating that these cells are entirely different from hematopoietic blast cells. CDH5 negativity might suggest the immaturity of these cobblestone cells as cells of the endothelial lineage because these cells became positive for CDH5 in EGM-2 medium, which is optimized for endothelial cell culture. Alternatively, CDH5 negativity might reflect the diversity of endothelial cells, as it was reported by researchers of Dr. Thomson's group in Wisconsin that rhesus monkey ES cell-derived endothelial cells are negative for CDH5 [23].

The hematopoietic blast cells observed in this study showed immature morphology and expressed CD34, KDR, and TAL1. All these findings suggest that these cells represent a very immature stage in a hematopoietic lineage [16, 17, 24, 25]. In addition, the blast cells possessed potentiality for erythroid and macrophage development. Based on these findings, the blastic cells were considered to be pluripotent hematopoietic progenitor cells. These cells, however, were not adult-type dormant hematopoietic stem cells because they proliferated continuously and lost, at least in part, their pluripotency when cultured in vitro for more than 2 mo. In addition, these cells had the capacity to differentiate into erythroblasts with fetal hemoglobins, clearly indicating the primitive hematopoietic nature of the hematopoietic blastic cells in our culture system.

Nevertheless, our differentiation system has an advantage. Previous reports showed that ES-derived hematopoietic cells lack the abilities to expand in liquid culture and to reconstitute the bone marrow of lethally irradiated animals unless specific genes, such as HOXB4 [8] and BCR/ABL1 [9], are exogenously introduced. Our system has enabled an effective expansion of ES-derived hematopoietic cells in liquid culture without gene manipulations via an expansion of finger-like cells. Our system thereby has significant advantages in terms of safety and convenience for future clinical applications of ES cells for diseases involving hematopoietic disorders.

FOOTNOTES

1 Supported in part by a research grant from the Japan Health Science Foundation. Back

2 Correspondence: Akira Yuo, Department of Hematology, Research Institute, International Medical Center of Japan, 1-21-1, Toyama, Shinjuku-ku, Tokyo 162-8655, Japan. FAX: 81 3 3207 1038; yuoakira{at}ri.imcj.go.jp Back

3 These authors contributed equally to this work. The order of the names of the authors was arbitrarily arranged. Back

Received: 2 May 2005.

First decision: 4 June 2005.

Accepted: 12 October 2005.

REFERENCES

  1. Sato N, Meijer L, Skaltsounis L, Greengard P, Brivanlou AH, Maintenance of pluripotency in human and mouse embryonic stem cells through activation of Wnt signaling by a pharmacological GSK-3-specific inhibitor. Nat Med 2004 10:55-63[CrossRef][Medline]
  2. Vodyanik MA, Bork JA, Thomson JA, Slukvin II, Human embryonic stem cell-derived CD34+ cells: efficient production in the coculture with OP9 stromal cells and analysis of lymphohematopoietic potential. Blood 2005 105:617-626[Abstract/Free Full Text]
  3. Takagi Y, Takahashi J, Saiki H, Morizane A, Hayashi T, Kishi Y, Fukuda H, Okamoto Y, Koyanagi M, Ideguchi M, Hayashi H, Imazato T, Kawasaki H, Suemori H, Omachi S, Iida H, Itoh N, Nakatsuji N, Sasai Y, Hashimoto N, Dopaminergic neurons generated from monkey embryonic stem cells function in a Parkinson primate model. J Clin Invest 2005 115:102-109[CrossRef][Medline]
  4. Nishikawa SI, Nishikawa S, Hirashima M, Matsuyoshi N, Kodama H, Progressive lineage analysis by cell sorting and culture identifies FLK1+VE-cadherin+ cells at a diverging point of endothelial and hemopoietic lineages. Development 1998 125:1747-1757[Abstract]
  5. Nakano T, Kodama H, Honjo T, In vitro development of primitive and definitive erythrocytes from different precursors. Science 1996 272:722-724[Abstract]
  6. Chadwick K, Wang L, Li L, Menendez P, Murdoch B, Rouleau A, Bhatia M, Cytokines and BMP-4 promote hematopoietic differentiation of human embryonic stem cells. Blood 2003 102:906-915[Abstract/Free Full Text]
  7. Wang L, Li L, Shojaei F, Levac K, Cerdan C, Menendez P, Martin T, Rouleau A, Bhatia M, Endothelial and hematopoietic cell fate of human embryonic stem cells originates from primitive endothelium with hemangioblastic properties. Immunity 2004 21:31-41[CrossRef][Medline]
  8. Kyba M, Perlingeiro RC, Daley GQ, HoxB4 confers definitive lymphoid-myeloid engraftment potential on embryonic stem cell and yolk sac hematopoietic progenitors. Cell 2002 109:29-37[CrossRef][Medline]
  9. Perlingeiro RC, Kyba M, Daley GQ, Clonal analysis of differentiating embryonic stem cells reveals a hematopoietic progenitor with primitive erythroid and adult lymphoid-myeloid potential. Development 2001 128:4597-4604
  10. Hacein-Bey-Abina S, von Kalle C, Schmidt M, Le Deist F, Wulffraat N, McIntyre E, Radford I, Villeval JL, Fraser CC, Cavazzana-Calvo M, Fischer A, A serious adverse event after successful gene therapy for X-linked severe combined immunodeficiency. N Engl J Med 2003 348:255-256[Free Full Text]
  11. Suemori H, Tada T, Torii R, Hosoi Y, Kobayashi K, Imahie H, Kondo Y, Iritani A, Nakatsuji N, Establishment of embryonic stem cell lines from cynomolgus monkey blastocysts produced by IVF or ICSI. Dev Dyn 2001 222:273-279[CrossRef][Medline]
  12. McHale CM, Winter PC, Lappin TR, Erythroid gene expression is differentially regulated by erythropoietin, haemin and {delta}-aminolaevulinic acid in UT-7 cells. Br J Haematol 1999 104:829-837[CrossRef][Medline]
  13. Li F, Lu S, Vida L, Thomson JA, Honig GR, Bone morphogenetic protein 4 induces efficient hematopoietic differentiation of rhesus monkey embryonic stem cells in vitro. Blood 2001 98:335-342[Abstract/Free Full Text]
  14. Saeki K, Hong Z, Nakatsu M, Yoshimori T, Kabeya Y, Yamamoto A, Kaburagi Y, Yuo A, Insulin-dependent signaling regulates azurophil granule-selective macroautophagy in human myeloblastic cells. J Leukoc Biol 2003 74:1108-1116[Abstract/Free Full Text]
  15. Kennedy M, Firpo M, Choi K, Wall C, Robertson S, Kabrun N, Keller G, A common precursor for primitive erythropoiesis and definitive haematopoiesis. Nature 1997 386:488-493[CrossRef][Medline]
  16. Robertson SM, Kennedy M, Shannon JM, Keller G, A transitional stage in the commitment of mesoderm to hematopoiesis requiring the transcription factor SCL/tal-1. Development 2000 127:2447-2459[Abstract]
  17. Fehling H-J, Lacaud G, Kubo A, Kennedy M, Robertson S, Keller G, Kouskoff V, Tracking mesoderm induction and its specification to the hemangioblast during embryonic stem cell differentiation. Development 2003 130:4217-4227[Abstract/Free Full Text]
  18. Delabesse E, Ogilvy S, Chapman MA, Piltz SG, Gottgens B, Green AR, Transcriptional regulation of the SCL locus: identification of an enhancer that targets the primitive erythroid lineage in vivo. Mol Cell Biol 2005 25:5215-5225[Abstract/Free Full Text]
  19. Crispino JD, GATA1 in normal and malignant hematopoiesis. Semin Cell Dev Biol 2005 16:137-147[CrossRef][Medline]
  20. Jarstrand C, Akerlund B, Oxygen radical release by neutrophils of HIV-infected patients. Chem Biol Interact 1994 91:141-146[CrossRef][Medline]
  21. Yuo A, Differentiation, apoptosis and function of human immature and mature myeloid cells: intracellular signaling mechanism. Int J Hematol 2001 73:438-452[Medline]
  22. McGinn S, Poronnik P, Gallery ED, Pollock CA, A method for the isolation of glomerular and tubulointerstitial endothelial cells and a comparison of characteristics with the human umbilical vein endothelial cell model. Nephrology 2004 9:229-237[Medline]
  23. Kaufman DS, Lewis RL, Hanson ET, Auerbach R, Plendl J, Thomson JA, Functional endothelial cells derived from rhesus monkey embryonic stem cells. Blood 2004 103:1325-1332[Abstract/Free Full Text]
  24. Nishikawa SI, Nishikawa S, Kawamoto H, Yoshida H, Kizumoto M, Kataoka H, Katsura Y, In vitro generation of lymphohematopoietic cells from endothelial cells purified from murine embryos. Immunity 1998 8:761-769[CrossRef][Medline]
  25. Choi K, Kennedy M, Kazarov A, Papadimitriou JC, Keller G, A common precursor for hematopoietic and endothelial cells. Development 1998 125:725-732[Abstract]
  26. Huber TL, Kouskoff V, Fehling HJ, Palis J, Keller G, Haemangioblast commitment is initiated in the primitive streak of the mouse embryo. Nature 2004 432:625-630[CrossRef][Medline]
  27. Gilbert SF, Developmental Biology, 5th ed Sunderland, MA Sinauer Associates, Inc 2000 358-380
  28. Eichmann A, Corbel C, Nataf V, Vaigot P, Breant C, Le Douarin NM, Ligand-dependent development of the endothelial and hemopoietic lineages from embryonic mesodermal cells expressing vascular endothelial growth factor receptor 2. Proc Natl Acad Sci U S A 1997 94:5141-5146[Abstract/Free Full Text]
  29. Pelosi E, Valtieri M, Coppola S, Botta R, Gabbianelli M, Lulli V, Marziali G, Masella B, Muller R, Sgadari C, Testa U, Bonanno G, Peschle C, Identification of the hemangioblast in postnatal life. Blood 2002 100:3203-3208[Abstract/Free Full Text]



This article has been cited by other articles:


Home page
 Stem CellsHome page
D. Rajesh, N. Chinnasamy, S. M. Mitalipov, D. P. Wolf, I. Slukvin, J. A. Thomson, and A. F. Shaaban
Differential Requirements for Hematopoietic Commitment Between Human and Rhesus Embryonic Stem Cells
Stem Cells, February 1, 2007; 25(2): 490 - 499.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Retraction
Right arrow A retraction has been published
Right arrow All Versions of this Article:
75/3/487    most recent
biolreprod.105.043331v2
74/2/295
biolreprod.105.043331v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Zhang, H.