Stem Cells http://www.peprotech.com/
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Stem Cells Vol. 24 No. 6 June 2006, pp. 1496 -1504
doi:10.1634/stemcells.2005-0530; www.StemCells.com
© 2006 AlphaMed Press

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
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 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 Despars, G.
Right arrow Articles by O’Neill, H. C.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Despars, G.
Right arrow Articles by O’Neill, H. C.

research-article

Splenic Endothelial Cell Lines Support Development of Dendritic Cells from Bone Marrow

Geneviève Despars, Helen C. O’Neill

School of Biochemistry and Molecular Biology, The Australian National University, Canberra, Australian Capital Territory, Australia

Key Words. Spleen • Endothelial cells • Dendritic cell development • Microenvironment

Correspondence: Professor Helen C. O’Neill, B.Sc., Ph.D. School of Biochemistry and Molecular Biology, Building 41, Linnaeus Way, Australian National University, Canberra, ACT, Australia, 0200. Telephone: +61 2 6125 4720; Fax: +61 2 6125 0313; e-mail: helen.oneill{at}anu.edu.au

Received August 26, 2005; accepted for publication January 26, 2006.

   ABSTRACT
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 Disclosures
 References
 
Although growth factors are commonly used to generate dendritic cells (DCs) in vitro, the role of the microenvironment necessary for DC development is still poorly understood. The mixed splenic stromal cell population STX3 defines an in vitro microenvironment supportive of DC development. Dissection of cellular components of the STX3 stroma should provide information about a niche for DC development. STX3 was therefore cloned by single-cell sorting, and a panel of 102 splenic stromal cell lines was established. Four representative splenic stromal cell lines that support hematopoiesis from bone marrow are described here in terms of stromal cell type and DC production. All four stromal lines express the endothelial genes Acvrl1, Cd34, Col18a1, Eng, Flt1, Mcam, and Vcam1 but not Cd31 or Vwf. Three of the four lines form tube-like structures when cultured on Matrigel. Their endothelial maturity correlates with the ability to support myeloid DC development from bone marrow. A fourth cell line, unable to form tube-like structures in Matrigel, produced large granulocytic cells expressing CD11b and CD86 but not CD11c and CD80. Conditioned media from splenic stromal cell lines also support DC production, indicating that soluble growth factors and cytokines produced by stromal lines drive DC development. This article reports characterization of immature endothelial cell lines derived from spleen that are supportive of DC development and predicts the existence of such a cell type in vivo which regulates DC development within spleen.


   INTRODUCTION
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 Disclosures
 References
 
There is increasing evidence that splenic stroma plays an important role in dendritic cell (DC) development. Long-term stroma cultures (LTCs) of spleen support DC hematopoiesis [1]. These LTCs comprise adherent stromal cells, small DC progenitors attached to stroma, and large immature DCs in suspension [2]. The DCs produced in LTCs of spleen resemble immature myeloid-like DCs, both phenotypically and functionally [3]. In this system, stromal cells are required for longevity of LTCs and development of DC progenitors into immature DCs [2]. An essential role for stroma in DC hematopoiesis was evident from a study showing that splenic stroma and stroma-derived growth factors, but not growth factors like granulocyte macrophage-colony-stimulating factor, interleukin (IL)-3, stem cell factor, IL-6, tumor necrosis factor (TNF)-{alpha}, IL-1-ß, IL-7, and fetal liver tyrosine kinase-3 ligand (FLT3L), support DC development from progenitors maintained in LTCs [4]. The STX3 splenic stroma was derived from one LTC that ceased production of DCs and also supported DC development from overlaid bone marrow [5]. In another study, splenic stromal cells overlaid with linc–-kit+ progenitor cells from bone marrow supported the development of plasmacytoid DCs [6]. Endothelial-like splenic stromal cells can also influence the differentiation of DCs, generating regulatory DCs that inhibit proliferation of naïve T cells [7].

Although these studies point to the involvement of splenic stroma in DC development, very little has been done to characterize the cellular components of stroma. Indeed, stromal cells can encompass a wide range of cell types, including fibroblasts, osteoblasts, chondrocytes, smooth muscle cells, and endothelial cells. These cells all share a common embryonic origin from the mesenchyme [8]. In the study of Zhang et al. [7], endothelial-like splenic stromal cells were isolated by depletion of CD11b+ splenic cells followed by positive selection for CD106/Vcam1+ cells. Svensson et al. [6] isolated splenic stromal cells by collagenase digestion of spleen followed by culture of adherent cells, which were identified as a mixture of fibroblasts expressing ER-TR7 along with CD68+ resident macrophages. Clearly, the term "stromal cell" describes very different cell populations for these two examples. It is therefore critical to identify which cell type within the splenic stroma supports DC development.

We previously established an in vitro model for spleen stroma-dependent development of DCs [1]. Splenic stroma in LTCs supports several stages in DC development, including formation of foci of progenitors and release of large DCs into supernatant [9]. The STX3 stromal cell population was shown to support development of foci and release of DCs from overlaid bone marrow [5]. Because STX3 comprised a mix of endothelial and fibroblast-like cells, it was studied here to determine the type of stromal cells which support DC development. Here, we describe a panel of splenic stromal cell lines supportive of DC development and investigate their characteristics as endothelial cells.


   MATERIALS AND METHODS
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 Disclosures
 References
 
Animals
Specific pathogen-free BALB/c mice were obtained from the Animal Resource Centre (Perth, WA, Australia, http://www.arc.wa.gov.au) and the John Curtin School of Medical Research (Canberra, ACT, Australia, http://jcsmr.anu.edu.au). Mice were housed and handled according to protocols approved by the Animal Experimentation Ethics Committee at The Australian National University (Canberra, ACT, Australia, http://www.anu.edu.au).

Cell Lines
The STX3 splenic stroma was established from LTC that has ceased production of DCs [2] and was passaged by scraping attached cells. Cloned cell lines were passaged using trypsin-EDTA treatment to dissociate cells for transfer into a new flask. Stromal cells were cultured at 37°C, 5% CO2, in air in Dulbecco’s modified Eagle’s medium supplemented with 10% fetal calf serum, 5 x 10–4 M 2-mercaptoethanol, 10 mM Hepes, 100 U/ml penicillin, 100 µg/ml streptomycin, 4 mg/l glucose, 6 mg/l folic acid, 36 mg/l L-asparagine, and 116 mg/l L-asparagine HCl (complete medium).

Isolation of Cloned Splenic Stromal Cell Lines
STX3 stroma was treated with 0.25% trypsin-EDTA to dissociate cells from the flask for 1 minute at 37°C, 5% CO2, followed by addition of complete medium. The cell suspension was centrifuged, washed once with complete medium, and re-suspended to a final cell concentration of 5 x 105 cells per ml. The cell suspension was incubated at 37°C, 5% CO2, for 2 hours with gentle agitation every 10 minutes followed by flow cytometry using single-cell deposition into 96-well plates. Clones were expanded by successive transfer from 96-well plates to 24-well plates and then to 25-cm2 flasks. A total of 102 cloned cell lines were derived and tested for ability to support hematopoiesis from overlaid bone marrow.

Assays for DC Development
On the day prior to experiment, splenic stromal cell lines were seeded at 1.5 x 106 cells per 25-cm2 flask. On the day of experiment, stromal cells were irradiated at 1.5 Gy using a cobalt 60 (60Co) source (CSIRO, Canberra, Australia, http://www.csiro.au). Bone marrow from BALB/c mouse was collected from femurs, and a single-cell suspension was prepared, followed by red blood cell lysis. Bone marrow cells were overlaid on to stroma at a concentration of 105 cells per ml in a total volume of 5 ml of fresh medium. These cocultures were held at 37°C, 5% CO2, in air, and medium changes were performed when the culture supernatant became acidic. After 15–20 days of culture, nonadherent cells were collected by gently shaking the flask. Cells were centrifuged and resuspended in medium for analysis of cell surface marker expression by antibody staining and flow cytometry.

Bone marrow cultures involved conditioned medium collected from confluent splenic stromal cell lines and kept frozen until use. Bone marrow cells were resuspended at 105 cells per ml in complete medium supplemented with 50% conditioned medium, and cells in suspension were harvested at days 5 and 7 for analysis by flow cytometry.

Antibody Staining
Both direct and indirect antibody staining methods combined with flow cytometry were used to detect cell surface marker expression as described previously [2]. An LSRII FACS (fluorescence-activated cell sorter) was used to detect antibody binding (Becton, Dickinson and Company, Franklin Lakes, NJ, http://www.bd.com). Affinity-purified antibodies were specific for B220 (RA3-6B2: CyChrome [CyC]-conjugated rat immunoglobulin [Ig] G2a), CD8 (53-6.7: phycoerythrin [PE]-conjugated rat IgG2a), CD11b (M1/70: biotinylated rat IgG2b), CD11c (HL3: biotinylated hamster IgG), CD80 (16-10A1: fluorescein isothiocyanate [FITC]-conjugated hamster IgG2), CD86 (GL1:biotinylated rat IgG2a), and CD205 (MCA949: rat gG2a). Anti-bodies were purchased from BD Pharmingen (San Diego, http://www.bdbiosciences.com/pharmingen) except for CD205 (Serotec, Raleigh, NC, http://www.serotec.com). Fluorescent conjugates used included PE-conjugated streptavidin and CyC-conjugated streptavidin (both from Pharmingen) and FITC-conjugated goat (F[ab']2) anti-rat Ig from SouthernBiotech (Birmingham, AL, http://www.southernbiotech.com).

Reverse Transcription-Polymerase Chain Reaction
Total RNA was extracted from splenic stromal cell lines and human umbilical vein endothelial cells (HUVECs) using Trizol (Invitrogen Life Technologies, Mount Waverley, VIC, Australia, http://www.invitrogen.com). Reverse transcription (RT) was performed with 2 µg of total RNA in 20 µl using poly-dT primers and Superscript II according to the manufacturer’s instructions (Invitrogen Life Technologies). One microliter of the RT reaction was then used in polymerase chain reaction (PCR) performed in a 20-µl volume in reaction buffer (67 mM Tris-HCl [pH 8.8], 16.6 mM [NH4]2SO4, 0.45% Triton X-100, 0.2 mg/ml gelatin) together with 1.5 mM MgCl2, 0.2 mM dNTPs, 0.5 ìM of each primer, and 1 unit of Taq Polymerase (Fisher Biotech, Perth, WA, Australia, http://www.fisherbiotec.com). Primers for Acvrl1, Cd34, Col18a1, Eng, Flt1, and Mcam were designed from mouse mRNA reference sequences available in the National Center for Biotechnology Information (Bethesda, MD, http://www.ncbi.nih.gov) database and included at least one primer from each pair which overlapped an exon/exon boundary to ensure amplification from only cDNA. Design of primers amplifying both mouse and human mRNA for Cd31, Vcam1, and Vwf was done by alignment of mouse and human reference mRNA and selection of identical regions. These regions were submitted to Primer3 software (http://frodo.wi.mit.edu/cgi-bin/primer3/primer3_www.cgi/). Primer synthesis was performed by Proligo (Lismore, NSW, Australia, http://www.proligo.com). Primer sequences were Acvrl1_left 5'-CCTTCCAAGCTGGT-GAACTG-3' and Acvrl1_right 5'-TGGCCTCCAGCATCA-GAG-3'; Cd31_left 5'-AGCCAACTTCACCATCCAGA-3' and Cd31_right 5'-ATCCACCGGGGCTATCAC-3'; Cd34_left 5'-CACCAGAGCTATTCCCGAAA-3' and Cd34_right 5'-TTTTCTTCCCAACAGCCATC-3'; Col18a1_left 5'-AGCCT-TAGAGGCCCACGA-3' and Col18a1_right 5'-CTTCCTTC-CCAGGTACACCA-3'; Eng_left 5'-CTTCCAAGGACAGCC-AAGAG-3' and Eng_right 5'-GTGGTTGCCATTCAAGT-GTG-3'; Flt1_right 5'-TATAAGGCAGCGGATTGACC-3' and Flt1_left 5'-TCATACACATGCACGGAGGT-3'; Mcam_ left 5'-AGGTCTTCAAAGCCCCAGAG-3' and Mcam_right 5'-TCTTGCAAGGGCAGACTGTT-3'; Vcam1_left 5'-TCTT-GTTTGCCGAGCTAAAT-3' and Vcam1_right 5'-CTCGCTG-GAACAGGTCAT-3'; Vwf_left 5'-ATGGTGCTGTACGGCT-GGA-3' and Vwf_right 5'-TGGCAGATCCCACTGAAGG-3'. PCR conditions were 3 minutes at 95°C, followed by 30 cycles of 30 seconds at 95°C, 30 seconds at 60°C, 30 seconds at 72°C, with a final elongation step of 2 minutes at 72°C, except for Vcam1, which had an annealing temperature of 58°C. PCR products sizes were Acvrl, 235 base pairs (bp); Cd31, 442 bp; Cd34, 158 bp; Col18a1, 142 bp; Eng, 221 bp; Flt1, 159 bp; Mcam, 158 bp; Vcam1, 179 bp; and Vwf, 395 bp. Control reactions contained no RNA (negative control) and RNA without RT as a control for contaminating DNA. PCR products were resolved on 2% agarose-Tris Borate EDTA or 8% polyacrylamide-Tris Borate EDTA gels and were visualized by ethidium bromide staining.

Endothelial Cell Tube Formation
The endothelial cell tube formation assay was performed using a 96-well black/clear plate coated with Matrigel Basement Membrane matrix purchased from BD Biosciences (Bedford, MA, http://www.bdbiosciences.com). Splenic stromal cells were seeded at either 1 x 104 cells per well or 1.5 x 104 cells per well. HUVECs were used as a control and seeded at 2 x 104 cells per well. After 18 hours of culture, medium was discarded and the plate washed twice with Hanks’ balanced saline solution (HBSS). Cells were stained with calcein AM (Molecular Probes, Eugene, OR, http://probes.invitrogen.com) at 8 µg/ml in HBSS by incubation for 30 minutes at 37°C, 5% CO2, in air, followed by two washes with HBSS. Cell structures were observed under an inverted microscope (Olympus IX81; Olympus, Mount Waverley, VIC, Australia, http://www.olympus-global.com) under white light and photographed with a digital camera (Olympus DP70). Fluorescent images were acquired using a digital camera (Leica DC500; Leica Microsystems, Gladesville, NSW, Australia, http://www.leica.com) connected to a dissecting microscope (Leica MZFLIII; Leica Microsystems). The degree of branching was calculated as the ratio of total number of branching points to the total area of the field. A branching point was defined as a meeting point from which at least three cord structures emerged. Branching points were counted for one field for each cell line and cell number. The total area of the field was 4.66 mm2.

Electron Microscopy
The 2A8, 3B5, 5G3, and 10C9 splenic stromal cell lines were harvested and fixed in 2% glutaraldehyde in 0.1 M sodium cacodylate buffer (pH 7.4) for 2 hours at room temperature, followed by two washes of 15 minutes in 0.1 M sodium cacodylate buffer (pH 7.4). A second fixation was done in 1% osmium tetroxide in 0.1 M sodium cacodylate buffer (pH 7.4) for 1.5 hours. Two washes of 15 minutes in 0.1 M sodium cacodylate buffer (pH 7.4) were performed, followed by three washes of 5 minutes in ddH2O. Dehydration of the samples involved successive incubation in 30, 50, 70, 90, 95, and thrice 100% ethanol for 15 minutes each. Samples were then infiltrated in 50:50 Spurr’s resin/acetone for 2 hours. Three successive incubations of 2 hours were done in fresh 100% Spurr’s resin, followed by embedding in resin at –70°C overnight. Post-staining involved 2% uranyl acetate for 5 minutes, followed by three rapid rinses in ddH2O. Samples were incubated in lead citrate for 8 minutes, followed by three rapid rinses in ddH2O. The samples were observed using a transmission electron microscope Hitachi H-7000 (Schaumburg, IL, http://www.hitachi.com).


   RESULTS
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 Disclosures
 References
 
Splenic Stromal Cell Lines Support Hematopoiesis from Bone Marrow
The STX3 mixed stroma has been subcultured for several passages and is therefore considered immortalized. After cloning by single-cell deposition using flow cytometry, 102 splenic stromal cell lines were tested for ability to support DC development from bone marrow. Because many cell lines displayed morphological changes after 10 passages, the stromal cell lines used for coculture assays were obtained only from passages 1–8.

An example of hematopoiesis in cocultures of splenic stromal cell lines with bone marrow is shown (Fig. 1Go). The 2A8, 3B5, 5G3, and 10C9 cell lines overlaid with bone marrow resulted in hematopoiesis, whereas the 8A4 stroma did not support hematopoiesis (Fig. 1Go). Foci of small round cells were observed above stroma in all cocultures, except 8A4. A good example is seen for 2A8 overlaid with bone marrow for 7 days (Fig. 1Go). Foci appeared initially, expanded during the assay, regressed, and eventually disappeared. Foci are believed to contain a DC progenitor population as previously described [2]. A suspension population of small cells possessing dendrites arose above stroma during the assay (Fig. 1Go, all panels at day 14). Medium- to large-size cells with dendrites were observed on stroma at late stages of the assay. No hematopoietic cell development was observed in negative controls (Fig. 1Go, all panels, negative control). Lack of hematopoiesis on 8A4 stroma indicated that any bone marrow-derived stromal cells present in coculture did not contribute to hematopoiesis in this system. No stromal cell growth with medium alone was observed after 11 days of culture of bone marrow cells, suggesting that the percentage of stromal cells among the bone marrow preparation was very low.


Figure 1
View larger version (119K):
[in this window]
[in a new window]
  		Figure 1. Hematopoiesis in coculture of bone marrow over splenic stromal cell lines. Cloned stromal cell lines were overlaid with bone marrow depleted of red blood cells at a concentration of 105 cells per ml. Negative controls comprised stroma without bone marrow cells. Cocultures were photographed under phase contrast using an inverted microscope (Leitz Fluovert) and a digital camera (SPOT RT). Objective x20. Scale bar = 100 µm.

 
Cloned stromal cell lines maintained the hematopoietic support capacity of STX3. The production of suspension cells was quantified in the coculture assays at 15 days for four cell lines and STX3 (Table 1Go). All splenic stromal cell lines were better producers of suspension cells than STX3. The 2A8 and 10C9 lines were the most efficient producers. The production of suspension cells was consistently lower than the input number of bone marrow cells, indicating that DC precursors and progenitors are relatively rare within mouse bone marrow. The cloning of STX3 led to isolation of stromal cells, which were more efficient in support of hematopoiesis than the original population of STX3.


View this table:
[in this window]
[in a new window]
  		Table 1. Stromal cell support of hematopoiesis

 
The majority of splenic stromal clones displayed an endothelial cell phenotype with cobblestone or O-ring morphology (approximately 80% of tested cell lines) (Fig. 2AGo, clones 2F10 and 3C6). Ten percent of cell lines were fibroblast-like cells with elongated morphology in culture (Fig. 2AGo, clone 4G8). A small number of cell lines displayed a mixed phenotype of elongated as well as cobblestone cells (Fig. 2AGo, clone 6A2). Morphology after growth arrest was considered more definitive than morphology during routine culture, as shown for 2A8 and 3B5 at 14 days post-irradiation with 60Co (Fig. 2BGo). The 2A8 splenic cell line adopted a cobblestone morphology with no overgrowth on neighboring cells. In contrast, the 3B5 cell line retained elongated morphology and cells grew on top of each other. Weibel-Palade bodies are a characteristic of mature endothelial cells [10]. Extensive investigation for Weibel-Palade bodies by electron microscopy was performed for each of the 2A8, 3B5, 5G3, and 10C9 splenic cell lines. A representative section of the cytoplasm of 2A8 is shown (Fig. 2CGo). Similar morphology was observed in all other cell lines, and no Weibel-Palade bodies were found. Instead, the cytoplasm was found to be rich in mitochondria with enlarged endoplasmic reticulum, suggestive of intense metabolic activity.


Figure 2
View larger version (92K):
[in this window]
[in a new window]
  		Figure 2. Morphology of splenic stromal cell lines. (A): Examples of splenic stroma showing cobblestone (clone 2F10), O-ring (clone 3C6), fibroblastic (clone 4G8), and mixed (clone 6A2) morphology. Cell lines were photographed with an inverted microscope (Leitz Fluovert) equipped with a digital camera (SPOT RT). Objective x10. Scale bar = 100 µm. (B): Morphology of the splenic stromal cell lines 2A8 and 3B5 maintained for 14 days after irradiation (1.5 Gy, cobalt 60 source). Photographs were taken with an inverted microscope (Leitz Fluovert) equipped with a digital camera (SPOT RT). Objective x20. Scale bar = 100 µm. (C): Electron microscopy showing cytoplasmic ultrastructure in the splenic stromal cell line 2A8. Nucleus (nu), endoplasmic reticulum (er), and mitochondria (mi) are identified. Photography was performed with a Hitachi H-7000 transmission electron microscope. Magnification x12,000. Scale bar = 1 µm.

 
Splenic Stromal Cell Lines Are Immature Endothelial Cells
A selection of cloned cell lines was investigated further to define stromal cell type. Two criteria influenced the choice of splenic stromal cell lines: the enhanced capacity to support hematopoiesis (Table 1Go) and morphology under various culture conditions (Figs. 1Go and 2Go).

Initially, the expression of a battery of endothelial cell genes was investigated in each of the four selected lines (Fig. 3Go). Acvrl1/ALK1, Col18a1, and Mcam/Cd146 were previously identified as genes specifically expressed by STX3 and reflective of an endothelial cell component [11]. Other markers of circulating endothelial cells were also used to delineate cell type, including Cd34, endoglin/Cd105, Flt1, Cd31, Vcam1, and Vwf [12]. PCR products resolved on 8% acrylamide-Tris borate-EDTA were of expected size, indicating that amplification was done on cDNA and not genomic DNA (Fig. 3AGo). Acvrl1, Cd34, Col18a1, Endoglin, Flt1, and Mcam were all expressed by 2A8, 3B5, 5G3, and 10C9 despite morphological differences between cell lines (Fig. 3AGo). Expression of the endothelial markers Cd31, Vcam1, and Vwf was also investigated using HUVECs as a positive endothelial cell control (Fig. 3BGo). Vcam1 was expressed in all cell lines at a low level. No expression of Cd31 and Vwf was detected in the stromal cell lines, although it was expressed in HUVECs (Fig. 3BGo). The lack of expression of Vwf is consistent with the absence of Weibel-Palade bodies and the immature endothelial cell phenotype of the splenic stromal cell lines (Fig. 3CGo).


Figure 3
View larger version (16K):
[in this window]
[in a new window]
  		Figure 3. Expression of endothelial genes by splenic stromal cell lines. (A): PCR products amplified for Acvrl1, Cd34, Col18a1, Endoglin, Flt1, and Mcam transcripts were resolved on 8% polyacrylamide-Tris borate-EDTA gels including molecular weight markers. (B): PCR products for Cd31, Vwf, and Vcam1 were resolved on 2% agarose-Tris borate-EDTA gels using HUVECs as a positive control. Abbreviations: HUVEC, human umbilical vein endothelial cell; PCR, polymerase chain reaction.

 
Capacity for formation of tube-like structures in Matrigel is a well-documented characteristic of endothelial cells [13]. Tube-like structure formation within Matrigel was therefore compared for the four splenic stromal cell lines and the positive control HUVECs (Fig. 4Go). HUVECs grew and formed a network of tube-like structures with cells well-aligned in cord structures (Fig. 4AGo). By comparison, 2A8, 5G3, and 10C9 cells clustered in branching points from which cord structures emerged (Fig. 4AGo). These cords were well-organized in 2A8 but thinner in 5G3 and 10C9. 3B5 cells showed weak organization into clusters in a polarized manner. The number of branching points was counted and then expressed as a ratio of area (Fig. 4BGo). Positive control HUVECs formed approximately 15 branching points per mm2 within Matrigel. The low number of branching points per square millimeter for 2A8 could be explained by its wide ring structures (Fig. 4AGo). The 3B5 cell line is clearly the least organized in comparison with HUVECs, whereas 2A8, 5G3, and 10C9 showed intermediate organization.


Figure 4
View larger version (28K):
[in this window]
[in a new window]
  		Figure 4. The splenic stromal cell lines 2A8, 5G3, and 10C9 form tube-like structures in Matrigel, whereas 3B5 does not. HUVEC and splenic stromal cell lines were grown on a Matrigel-coated plate for 18 hours. (A): Cell structures were photographed with a digital camera (Olympus DP70) connected to an inverted microscope (Olympus IX81). 2A8, 1 x 104 cells per well; 3B5, 1.5 x 104 cells per well; 5G3, 1 x 104 cells per well; 10C9, 1 x 104 cells per well; and HUVEC, 2 x 104 cells per well. Objective x2. Scale bar = 500 µm. (B): Semiquantification of branching points. The number of branching points was counted in one field for each plating concentration and expressed as a ratio of the area of the field. Abbreviation: HUVEC, human umbilical vein endothelial cell.

 
Splenic Stromal Cell Lines Support DC Development from Bone Marrow
Coculture assays of supportive stromal cell lines and bone marrow cells were set up for analysis of hematopoietic cells type produced. Stromal cell lines were used for coculture at less than 10 passages to avoid morphological changes. Non-adherent cells in cocultures represented by both cells in suspension and cells deposited on stroma were harvested at day 14 and stained with antibodies to delineate cell type produced (Fig. 5Go). After 2 weeks of culture of bone marrow above 2A8, 5G3, and 10C9 cell lines, three subsets of cells were detectable by forward and side scatter analysis (Fig. 5AGo). A subset of small cells with low granularity was detected. Two subsets of large granulocytic cells were detected. The profile of forward and side scatter for 3B5-derived suspension cells was clearly distinct. The gated granulocytic cell populations were analyzed for expression of CD11c, CD11b, CD86, and CD80 (Fig. 5AGo). Cells developing above 2A8, 5G3, and 10C9 stroma showed similar staining profiles CD11clo CD11b+CD86+ CD80lo reflecting myeloid-like DCs. Cells developing above 3B5 stroma did not express CD11c and CD80 but did express high levels of CD11b and CD86, indicative of myeloid lineage. Cells obtained from coculture were also stained for CD205, CD8-{alpha}, and B220 (Fig. 5BGo). The gated subsets were consistently negative for CD205, CD8-{alpha}, and B220, indicating lack of evidence for lymphoid-like or plasmacytoid-like DCs.


Figure 5
View larger version (17K):
[in this window]
[in a new window]
  		Figure 5. The splenic stromal cell lines 2A8, 5G3, and 10C9, but not 3B5, support DC development from bone marrow. Nonadherent cells were harvested at day 14 of coculture, involving the stromal cell lines 2A8, 3B5, 5G3, and 10C9 overlaid with bone marrow. Fluorescent staining involved both directly conjugated antibodies and the use of secondary labeled conjugates for FACS analysis. Post-acquisition gating on side and forward scatter profiles was used to delineate the large granulocytic cell population (SSC and FSC panels). The large cells are expressed as a percentage of the total cell population (mean ± SD, n = 11). (A): Staining patterns of the gated large-cell subset are shown for CD11c, CD11b, CD86, and CD80. (B): Staining patterns of the gated large-cell subset are shown for CD205, CD8-{alpha}, and B220. Spleen cells were used as a positive control. Background staining of secondary reagent alone is shown by shading. The results presented are representative of at least two independent experiments. Abbreviations: DC, dendritic cell; FACS, fluorescence-activated cell sorting; FSC, forward scatter; SSC, side scatter.

 
Splenic Stromal Cell Lines Produce Soluble Factors That Support DC Development
Cells with the phenotype and morphology of DCs are released into supernatant during coculture of bone marrow cells above competent splenic stromal cell lines. Some variability has been noted in that 3B5 stroma is less competent than 2A8, 5G3, and 10C9 splenic cell lines in producing CD11c+DCs. This would appear to relate to stromal cell support of early hematopoiesis. One question is whether stromal cell lines also produce soluble growth factors that support DC development beyond the stage of cell-cell contact or cell-matrix contact required for progenitor cell differentiation. The extent to which each of the cloned stroma produces soluble factors that support DC development was investigated in bone marrow cultures.

Single-cell suspensions of bone marrow depleted of red blood cells were cultured in medium conditioned by splenic stromal cell lines (50% in fresh medium) (Fig. 6Go). As early as 24 hours after seeding, cells with dendrites were detectable in culture. This suggests proliferative support of existing DC precursors in bone marrow rather than production of new DC progenitors. After an initial proliferative burst, cells with very round cellular bodies and short dendrites reflecting immature or resting DCs settled on the bottom of flasks (Fig. 6Go, photographs). These cells did not flatten out over time and did not proliferate in foci. These nonadherent cells with dendrites were observed until day 11, after which time growth of a bone marrow stromal layer was evident and could influence the outcome of the assay. Previous studies have shown that bone marrow stroma alone does not support DC development [1].


Figure 6
View larger version (65K):
[in this window]
[in a new window]
  		Figure 6. Conditioned medium from splenic stromal cell lines supports DC development from bone marrow. A single-cell suspension of red blood-depleted bone marrow was cultured in 50% conditioned medium from stromal cell lines 2A8, 3B5, 5G3, and 10C9, or medium as a control. Photographs were taken with an inverted microscope (Leica DMIRE2). Cells deposited on the bottom of flasks were photographed at day 5 (2A8, 3B5, 5G3, and 10C9) or day 11 (medium only). Nonadherent cells were harvested at days 5 and 7 and stained with biotinylated CD11c followed by streptavidin-PE. Post-acquisition gating on large granulocytic cells delineates the DC population (SSC and FSC panels) expressed as a percentage of the total population (mean ± SD, n = 3). Background staining of secondary reagent is depicted by the shaded area. Results presented are representative of three independent experiments. 2A8, 3B5, 5G3, and 10C9 conditioned medium: objective x63, scale bar = 1 µm; media only: objective x20, scale bar = 50 µm. Abbreviations: DC, dendritic cell; FSC, forward scatter; PE, phycoerythrin; SSC, side scatter.

 
Nonadherent cells were harvested after 5 and 7 days of culture for analysis of CD11c expression by flow cytometry before expansion of stromal cells. Profiles of side and forward scatter indicated that conditioned media from four splenic stroma supported early proliferation of a subpopulation of large granulocytic cells from bone marrow precursors (Fig. 6Go, side scatter and forward scatter panels). This subset represented 8%–12% of the population. A subset of these large cells (23%–35%) expressed high levels of CD11c by day 5. This percentage increased up to 70%–80% during culture (Fig. 6Go, panels at day 7). Conditioned media from all four splenic stromal cell line were similar in their support capacity for DC production from bone marrow. Culture of bone marrow cells with medium alone did not produce large granular cells with DC characteristics or stromal cells (Fig. 6Go, medium only). Incubation of conditioned medium without bone marrow for up to 7 days did not produce stromal cells (data not shown).


   DISCUSSION
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
Disclosures
 References
 
Although some growth factors and cytokines are extensively used for in vitro production of DCs, very few reports address the question of stromal elements required for DC development. Our aim was to identify regulators of the microenvironment required for DC development. Here, we have established and characterized cloned stromal cell lines from the STX3 splenic stroma which supports DC development. Four cell lines were selected for further investigation based on capacity for support of hematopoiesis from bone marrow and their endothelial/fibroblast-like morphology. Splenic stromal cell lines regulate DC development from bone marrow cells by cell-cell contact of progenitors proliferating above stroma. Whereas 2A8, 5G3, and 10C9 supported the development of DCs expressing CD11c, CD11b, CD80, and CD86, the 3B5 cell line produced large granulocytic CD11c–CD80– cells expressing CD11b and CD86. Conditioned medium from all four cell lines drives DC development from bone marrow, indicating production of soluble factors and cytokines that also regulate DC development. DCs were identified by size, granularity, and cell surface marker expression using flow cytometry, which are standard procedures for characterization of DCs. In terms of identification of microenvironmental regulators of DC development, we showed here that splenic stromal cell lines can regulate DC development both by stromal cell contact with hematopoietic progenitors and by production of stromal soluble factors that influence later-stage DC precursors.

In both stromal and conditioned medium assays from bone marrow, there was no evidence for stromal contaminants from bone marrow contributing to DC development. Lack of stromal growth in bone marrow cultured with medium alone (Fig. 6Go) indicated that stromal cells comprise a very small fraction of total bone marrow cells. Lack of hematopoiesis above the non-supportive splenic stroma 8A4 (Fig. 1Go) indicated that stromal cells arising from bone marrow did not contribute to DC hematopoiesis. However, depletion of stromal cells from bone marrow cannot be used to confirm these results because of the difficulty of choosing appropriate markers and the possibility of depleting unknown DC precursors from the bone marrow preparation. Depletion using adhesive properties of stromal cells cannot be used, because it would result in loss of macrophages as well as semi-adherent DCs and DC precursors, both well-documented phenomena. Furthermore, a concentration of 105 bone marrow cells per ml overlaid onto stroma has been previously shown to exclude reseeding with colony-forming unit-fibroblast (CFU-F) derived from inoculate but allowing studies of hematopoietic capacity of stroma [14].

Both the 2A8, 5G3, and 10C9 stromal cell lines and their conditioned medium supported DC development. However, the 3B5 stromal cell line supported only development of a CD11c–myeloid cell type in bone marrow coculture after 14 days. The phenotype of these cells resembles previously described CD11b+CD80–CD86–Flt3+ DC progenitors [15]. However, conditioned medium from 3B5 fully supports DC development from bone marrow precursors by 5–7 days. It is possible that different hematopoietic progenitors/precursors within bone marrow respond in each assay. Early DC progenitors would be expected to require cell-cell contact with splenic stromal cells, shown here by formation of foci of small round cells above stroma. In the coculture assays, initial formation of foci was necessary for later release of DCs into supernatant. However, culture of bone marrow cells in conditioned medium from stromal cells generated an early proliferative burst of DC production with no evidence of foci formation. This assay appears to detect more differentiated bone marrow DC precursors requiring soluble growth factors and cytokines produced by splenic stroma for development. These results also support the involvement of soluble factors in the late stages of DC development. It was noted that the CD11c expression level on DCs developing in coculture was lower than that on DCs obtained from culture of bone marrow with conditioned medium. Splenic stromal cells may prevent full maturation of DCs during coculture assays, which needs to be tested. Studies on coculture assays that produce B lymphocytes and megakaryocytes have also reported that direct stromal cell contact can also maintain the immature phenotype of hematopoietic cells produced [16, 17].

The splenic stromal cell lines described here are of endothelial lineage. They express a panel of endothelial genes, including Acvrl1, Cd34, Col18a1, Eng, Flt1, Mcam, and Vcam1. Three cell lines, 2A8, 5G3, and 10C9, displayed characteristic cobblestone morphology in culture that matched their capacity to form tube-like structures within Matrigel. Their possible phenotype as immature endothelial cells is supported by evidence for absence of Weibel-Palade bodies and lack of expression of Cd31 and Vwf. Cd31 and Vwf are involved in adhesion and coagulation, which are properties of mature endothelial cells. Furthermore, inactivation of Cd31 or Vwf in the mouse genome does not alter angiogenesis in the embryo but does affect endothelial function in the adult [18, 19]. Weibel-Palade body formation is dependent on endothelial cell confluency in vitro [20].

Culture conditions used to derive STX3 and to clone stromal splenic cell lines have perhaps favored immortalization and account for the immature endothelial phenotype. Culture conditions for primary endothelial cells are quite different from those used for splenic stromal cell lines. HUVECs and human pulmonary microvascular endothelial cells (HPMECs) are commonly cultured on gelatin or fibronectin-coated dishes [21, 22]. Similarly, proper differentiation of endothelial progenitor cells requires coating of culture dishes with fibronectin or gelatin [23–25]. It has been reported that upon in vitro culture, endothelial cells tend to lose their specialized function [26]. Immortalization of HPMECs had a detrimental effect on expression of endothelial characteristics [21]. However, despite an immature endothelial phenotype, the splenic stromal cell lines are adequate supporters of DC development from bone marrow.

One possibility that needs to be investigated is that the splenic endothelial cell lines described here reflect an in vivo cell type in spleen. Although they express genes related to endothelial progenitor cells [12], heterogeneity in expression of endothelial markers among cells of the vascular system has been reported [26]. A comparison of morphology for HUVECs and microvascular placental endothelial cells indicated that the latter showed spindle-shape morphology quite different from the cobblestone standard typical of HUVECs [22]. Endothelial progenitor cells can display spindle-shaped or cobblestone morphology [27, 28]. Undifferentiated embryonic cell lines with endothelial potential have shown fibroblast morphology at subconfluency but could form cord-like structures at confluency [29]. Endothelial cells derived from hemangiomas displayed elongated morphology despite expression of endothelial markers [30]. Although a common functional assay for endothelial cells includes formation of tube-like structures within Matrigel [13], similar structures have also been reported for the bladder carcinoma cell line ECV-304 [31]. Some reports have indicated that endothelial progenitor cells from blood can form tube-like structures in Matrigel, whereas other reports were negative for this function [27–29]. Therefore, given a diversity of marker expression, morphology, and function among endothelial cells cultured in vitro, it is possible that the splenic endothelial cell lines described here reflect an in vivo subset of endothelial cells in spleen with distinct hematopoietic capacity.

Whereas the 2A8, 5G3, and 10C9 stromal cell lines show endothelial characteristics and support DC development, the 3B5 cell line reflects a less mature endothelial phenotype and is unable to support DC development in coculture assays. Splenic endothelial cells isolated by Zhang et al. [7] have some characteristics in common with the endothelial cell lines described here. They express high levels of Vcam1 but low levels of Vwf and Cd31. They support extended proliferation of DCs cultured out of bone marrow with IL-4 and TNF-{alpha}. Fully differentiated HUVECs do not support DC development, whereas HUVECs treated with TNF-{alpha} support DC development from CD34+ blood precursors [32]. DCs also mature by reverse transmigration through endothelium [33, 34]. All of these results confirm the involvement of endothelium in DC development and maturation. Cell surface markers expressed by the splenic stromal cell lines can be used to delineate a specific endothelial cell type in spleen, which could define a niche for DC development during steady-state conditions in spleen.


   DISCLOSURES
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 Disclosures
References
 
The authors indicate no potential conflicts of interest.


   ACKNOWLEDGMENTS
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 Disclosures
 References
 
We thank Sabine Gruninger for assistance in cell sorting, Anna Bezos for providing HUVECs, and Cathy Gillepsie and Roger McCarth for electron microscopy and microscopy of the Matrigel assay. This work was supported by grants from the National Health and Medical Research Council of Australia to H.C.O. G.D. was supported by a Ph.D. fellowship from the Fonds de la Recherche en Santé du Québec, Québec, QC, Canada.


   REFERENCES
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 Disclosures
 References
 

  1. Ni K, O’Neill HC. Long-term stromal cultures produce dendritic-like cells. Brit J Haematol 1997;97:710–725.[CrossRef][Medline]

  2. Wilson HL, Ni K, O’Neill HC. Identification of progenitor cells in long-term spleen stromal cultures that produce immature dendritic cells. Proc Natl Acad Sci U S A 2000;97:4784–4789.[Abstract/Free Full Text]

  3. Quah B, Ni K, O’Neill HC. In vitro hematopoiesis produces a distinct class of immature dendritic cells from spleen progenitors with limited T cell stimulation capacity. Int Immunol 2004;16:567–577.[Abstract/Free Full Text]

  4. Wilson HL, Ni K, O’Neill HC. Proliferation of dendritic cell progenitors in long term culture is not dependent on granulocyte macrophage-colony stimulating factor. Exp Hematol 2000;28:193–202.[CrossRef][Medline]

  5. Ni K, O’Neill HC. Spleen stromal cells support haemopoiesis and in vitro growth of dendritic cells from bone marrow. Brit J Haematol 1999;105:58–67.[CrossRef][Medline]

  6. Svensson M, Maroof A, Ato M et al. Stromal cells direct local differentiation of regulatory dendritic cells. Immunity 2004;21:805–816.[CrossRef][Medline]

  7. Zhang M, Tang H, Guo Z et al. Splenic stroma drives mature dendritic cells to differentiate into regulatory dendritic cells. Nature Immunol 2004;5:1124–1133.[CrossRef][Medline]

  8. Dennis JE, Charbord P. Origin and differentiation of human and murine stroma. STEM CELLS 2002;20:205–214.[Abstract/Free Full Text]

  9. O’Neill HC, Wilson HL, Quah B et al. Dendritic cell development in long-term spleen stromal cultures. STEM CELLS 2004;22:475–486.[Abstract/Free Full Text]

  10. Michaux G, Cutler DF. How to roll an endothelial cigar: The biogenesis of Weibel-Palade bodies. Traffic 2004;5:69–78.[CrossRef][Medline]

  11. Despars G, Ni K, Bouchard A et al. Molecular definition of an in vitro niche for dendritic cell development. Exp Hematol 2004;32:1182–1193.[CrossRef][Medline]

  12. Blann AD, Woywordt A, Bertolini F et al. Circulating endothelial cells: Biomarker of vascular disease. Thromb Haemost 2005;93:228–235.[Medline]

  13. Passaniti A, Taylor RM, Pili R et al. A simple, quantitative method for assessing angiogenesis and antiangiogenic agents using reconstituted basement membrane, heparin, and fibroblast growth factor. Lab Invest 1992;67:519–528.[Medline]

  14. Van Den Heuvel R, Schoeters G, Leppens H et al. Stromal cells in long-term cultures of liver spleen, and bone marrow at different developmental ages have different capacities to maintain GM-CFC proliferation. Exp Hematol 1991;19:115–121.[Medline]

  15. Hieronymus T, Gust TC, Kirsh RD et al. Progressive and controlled development of mouse dendritic cells from Flt3+CD11b+ progenitors in vitro. J Immunol 2005;174:2552–2562.[Abstract/Free Full Text]

  16. Koguma M, Matsuda K-I, Okuyama R et al. Selective proliferation of lymphoid cells from lineage–c-kit+ Sca-1+ cells by a clonal bone marrow stromal cell line. Exp Hematol 1998;26:280–287.[Medline]

  17. Zweegman S, Veenhof MA, Huijens PC et al. Regulation of megakaryocytopoiesis in an in vitro model: Preferential adhesion of megakaryocytic progenitors and subsequent inhibition of maturation. Exp Hematol 2000;28:401–410.[CrossRef][Medline]

  18. Denis C, Methia N, Frenette PS et al. A mouse model of severe von Willebrand disease: Defects in hemostasis and thrombosis. Proc Natl Acad Sci U S A 1998;95:9524–9529.[Abstract/Free Full Text]

  19. Schenkel AR, Chew TW, Muller WA. Platelet endothelial cell adhesion molecule deficiency or blockade significantly reduces leukocyte emigration in a majority of mouse strains. J Immunol 2004;173:6403–6408.[Abstract/Free Full Text]

  20. Howell GJ, Herbert SP, Smith JM et al. Endothelial cell confluence regulates Weibel-Palade body formation. Mol Membr Biol 2004;21:413–421.[CrossRef][Medline]

  21. Krump-Konvalinkova V, Bittinger F, Unger RE et al. Generation of human pulmonary microvascular endothelial cells lines. Lab Invest 2001;81:1717–1727.[Medline]

  22. Lang I, Pabst M, Hiden U et al. Heterogeneity of microvascular endothelial cells isolated from human term placenta and macrovascular umbilical vein endothelial cells. Eur J Cell Biol 2003;82:163–173.[CrossRef][Medline]

  23. Shi Q, Rafii S, Wu MH-D et al. Evidence for circulating bone-marrow derived endothelial cells. Blood 1998;92:362–367.[Abstract/Free Full Text]

  24. Bagley RG, Walter-Yohrling J, Cao X et al. Endothelial precursor cells as a model of tumor endothelium: Characterization and comparison with mature endothelial cells. Cancer Res 2003;63:5866–5873.[Abstract/Free Full Text]

  25. Bellik L, Ledda F, Parenti A. Morphological and phenotypical characterization of human endothelial progenitor cells in an early stage of differentiation. FEBS Lett 2005;579:2731–2736.[Medline]

  26. Garlanda C, Dejana E. Heterogeneity of endothelial cells. Arterioscler Thromb Vasc Biol 1997;17:1193–1202.[Abstract/Free Full Text]

  27. Gulati R, Jevremovic D, Peterson TE et al. Diverse origin and function of cells with endothelial phenotype obtained from adult human blood. Circ Res 2003;93:1023–1025.[Abstract/Free Full Text]

  28. Hur J, Yoon CH, Kim HS et al. Characterization of two types of endothelial progenitor cells and their different contributions to neovasculogenesis. Arterioscler Thromb Vasc Biol 2004;24:288–293.[Abstract/Free Full Text]

  29. Yin Y, Que J, The M et al. Embryonic cells lines with endothelial potential: An in vitro system for studying endothelial differentiation. Arterioscler Thromb Vasc Biol 2004;24:691–696.[Abstract/Free Full Text]

  30. Dosanjh A, Chang J, Bresnick S et al. In vitro characteristics of neonatal hemangioma endothelial cells: Similarities and differences between normal neonatal and fetal endothelial cells. J Cutan Pathol 2000;27:441–450.[CrossRef][Medline]

  31. Kishibe J, Yamada S, Okada Y et al. Structural requirements of heparin sulphate for the binding to the tumor-derived adhesion factor/angio-modulin that induces cord-like structures to ECV-304 human carcinoma cells. J Biol Chem 2000;275:15321–15329.[Abstract/Free Full Text]

  32. Moldenhauer A, Nociari M, Lam G et al. Tumor necrosis factor alpha-stimulated endothelium: An inducer of dendritic cell development from hematopoietic progenitors and myeloid leukemic cells. STEM CELLS 2004;22:144–157.[Abstract/Free Full Text]

  33. Randolph GJ, Beaulieu S, Lebecque S et al. Differentiation of monocytes into dendritic cells in a model of transendothelial trafficking. Science 1998;282:480–483.[Abstract/Free Full Text]

  34. D’Amico G, Bianchi G, Bernasconi S et al. Adhesion, transendothelial migration, and reverse transmigration of in vitro cultured dendritic cells. Blood 1998;92:207–214.[Abstract/Free Full Text]




This article has been cited by other articles:


Home page
 Molecular Cancer TherapeuticsHome page
E. Weisberg, R. D. Wright, D. W. McMillin, C. Mitsiades, A. Ray, R. Barrett, S. Adamia, R. Stone, I. Galinsky, A. L. Kung, et al.
Stromal-mediated protection of tyrosine kinase inhibitor-treated BCR-ABL-expressing leukemia cells
Mol. Cancer Ther., May 1, 2008; 7(5): 1121 - 1129.
[Abstract] [Full Text] [PDF]

Home page
 Stem CellsHome page
J. K. H. Tan and H. C. O'Neill
Concise Review: Dendritic Cell Development in the Context of the Spleen Microenvironment
Stem Cells, September 1, 2007; 25(9): 2139 - 2145.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
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 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 Despars, G.
Right arrow Articles by O’Neill, H. C.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Despars, G.
Right arrow Articles by O’Neill, H. C.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
STEM CELLS THE ONCOLOGIST CME ALPHAMED PRESS JOURNALS
http://www.peprotech.com/