Isolation and characterization of dendritic cells from common marmosets for preclinical cell therapy studies

Abstract

Dendritic cells (DCs) have important functions as modulators of immune responses, and their ability to activate T cells is of great value in cancer immunotherapy. The isolation of DCs from the peripheral blood of rhesus and African green monkeys has been reported, but the immune system in the common marmoset remains poorly characterized, although it offers many potential advantages for preclinical studies. In the present study, we devised methods, based on techniques developed for mouse and human DC preparation, for isolating DCs from three major tissue sources in the common marmoset: bone marrow (BM), spleen and peripheral blood. Each set of separated cells was analysed using the cell surface DC-associated markers CD11c, CD80, CD83, CD86 and human leucocyte antigen (HLA)-DR, all of which are antibodies against human antigens, and the cells were further characterized both functionally and morphologically as antigen-presenting cells. BM proved to be an excellent cell source for the isolation of DCs intended for preclinical studies on cell therapy, for which large quantities of cells are required. In the BM-derived CD11c⁺ cell population, cells exhibiting the characteristic features of DCs were enriched, with the typical DC morphology and the abilities to undergo endocytosis, to secrete interleukin (IL)-12, and to stimulate Xenogenic T cells. Moreover, BM-derived DCs produced the neurotrophic factor NT-3, which is also found in murine splenic DCs. These results suggest that BM-derived DCs from the common marmoset may be useful for biological analysis and for preclinical studies on cell therapy for central nervous system diseases and cancer.

Introduction

Dendritic cells (DCs) have the ability to prime T cells to produce immune responses against viruses, bacteria and tumours. When immature, DCs can capture and process exogenous antigens; following maturation, they enhance the expression of both major histocompatibility complex (MHC) class II and costimulatory molecules, and migrate to lymphoid organs, where they stimulate potent antigen-specific T cells.^1,2 Because of their high ability to activate cytotoxic T lymphocytes, DCs are regarded as a useful tool in cancer immunotherapy and are currently being used in human clinical studies.^3–6 Furthermore, we reported a new use of DCs for the treatment of central nervous system (CNS) diseases such as spinal cord injury (SCI), involving the activation of endogenous neural stem/progenitor cells (NSPCs).⁷

DCs have been extensively characterized in humans and rodents. The use of primates, instead of rodents, to examine the therapeutic effects of DC therapy is an important step towards future clinical studies on the treatment of SCI and cancer. Although DCs have been isolated from rhesus and African green monkeys,^8–12 the characteristics of DCs and other components of the immune system of the common marmoset (CM) remain unclear. Therefore, methods for the isolation of DCs from the CM and the subsequent characterization of these cells are needed for preclinical studies.

Compared with other monkeys, the CM offers many advantages for preclinical studies.^13,14 The average weight of an adult CM is between 200 and 300 g, making it possible to handle and breed them easily on a large scale and reducing the cost of experiments.¹⁴ Sequence analysis of the entire CM genome is progressing at Washington University and the National Institutes of Health (NIH) Intramural Sequencing Center, and the results of these efforts will clarify the genetic similarity between the CM and humans. Because of these advantages, CMs have been widely used in many studies involving gene therapy,^15,16 bacterial infection,¹⁷ toxicology¹⁸ and immunology.^19,20 The usefulness of a CM model for studies on CNS diseases has also been shown for Parkinson's disease,²¹ stroke,²² Huntington disease,²³ multiple sclerosis,^24,25 anxiety²⁶ and SCI.^27,28 In addition, human antibodies have been reported to be cross-reactive to CM peripheral blood cells,^29,30 and a CM anti-CD34 antibody has been produced for studying haematopoietic cells.^31,32 In this study, we established methods for isolating DCs from the bone marrow (BM), spleen and peripheral blood mononuclear cells (PBMC) of the CM; all of these tissues contain DC progenitor cells. Furthermore, in view of the excellent yield of DCs from BM, we focused on the characterization of these cells for use in preclinical studies on cell therapy.

Materials and methods

Animals

Healthy CMs (body weight 200–350 g; age 1–8 years) were selected from experimental stock at the Central Institute for Experimental Animals (Kawasaki, Japan) and were killed for isolation of DCs from tissues, as described below. All animal experiments were performed according to the guidelines of the Animal Care and Use Committee of the Keio University School of Medicine.

Antibodies

The cross-reactivities of the following fluorescein isothiocyanate (FITC)-, phycoerythrin (PE)- or allophycocyamin (APC)-conjugated anti-human monoclonal antibodies (mAbs) to CM were determined by flow cytometry and were found to agree with previous results:^15,29,30 CD1a (clone HI149; eBioscience, San Diego, CA), CD1c (clone ad5-8E7; MiltenyiBiotec, Bergisch Gladbach, Germany; clone 11·86; Becton Dickinson, San Jose, CA), CD3 (clone SP34; BD Pharmingen, San Diego, CA), CD4 (clone MT310; DAKO Cytomation, Glostrup, Denmark), CD8 (clone T8; Beckman Coulter, Miami, FL), CD11c (clone S-HCL-3; Becton Dickinson), CD14 (clone TUK4; DAKO Cytomation; clone M5E2; BD Pharmingen), CD34 (clone BIRMA-K3; DAKO Cytomation), CD80 (clone MAB104; Beckman Coulter), CD83 (clone HB15a; Beckman Coulter), CD86 (clone B-T7; Diaclone, Besançon, France), and HLA-DR (clone G46-6; BD Pharmingen).

Isolation of DCs from BM cells

Femurs and tibias were removed and left in 70% ethanol for a few minutes before washing in phosphate-buffered saline (PBS). Both ends were cut with scissors, and the marrow was flushed with RPMI-1640 (Sigma, St Louis, MO) using a plastic pipette. Cluster cells were dissociated by vigorous pipetting and were filtered through a cell strainer (100 µm; BD Falcon, Billerica, MA). Red blood cells were removed using an ACK lysis buffer (BioWhittaker, Walkersville, MD). BM cells were suspended in RPMI-1640 supplemented with 10% heat-inactivated fetal calf serum (FCS) at a cell density of 2 × 10⁷ cells/ml and were cultured at 37° in 5% CO₂. After overnight incubation, suspension cells were collected, adjusted to a cell density of 4 × 10⁵ cells/ml, and plated on six-well plates (Costar Corp., Cambridge, MA) in a complete medium (cRPMI), which consisted of RPMI-1640 supplemented with 10% FCS, penicillin and streptomycin (50 U/ml; Invitrogen, Carlsbad, CA), recombinant human granulocyte–macrophage colony-stimulating factor (rhGM-CSF) (100 ng/ml; PeproTech, Rocky Hill, NJ), and recombinant human interleukin-4 (rhIL-4) (100 ng/ml; PeproTech), based on the method for generating mouse BM-derived DCs.^33,34 The reactivity of human GM-CSF and IL-4 on CM cells has been previously demonstrated.^35,36 Half the supernatant was replenished with fresh cRPMI on culture day 4, and the floating cells were collected as a DC-enriched cell fraction on culture day 7 or 8. In this study, DCs were also generated from CD34⁺ BM progenitor cells based on methods previously reported.^9,11 Sorted CD34⁺ BM cells were plated RPMI-1640 medium supplemented with 10% FCS, 1% non-essential amino acids (Invitrogen), 1 mm sodium pyruvate, 10 mm HEPES, rhGM-CSF (100 ng/ml), penicillin and streptomycin (50 U/ml; Invitrogen), recombinant human Flt-3 ligand (rhFlt-3 ligand) (100 ng/ml; PeproTech), recombinant human stem cell factor (rhSCF) (100 ng/ml; PeproTech), and tumour necrosis factor-α (rhTNF-α) (5 ng/ml; PeproTech). On culture day 2, the same amounts of cytokines were added again to the medium. On day 5, cells were recultured in cRPMI supplemented with hrTNF-α (5 ng/ml), and further cultured for 1 week. For maturation, the BM cell culture was stimulated with 1 µg/ml Esherichia coli (055:B5)-derived lipopolysaccharide (LPS; Sigma) for 24 hr. To enrich the CD11c⁺ cell population, the floating cultured cells were labelled with PE-conjugated anti-human CD11c mAb and directly purified by cell sorting on Moflo (DAKO Cytomation) or further labelled with anti-PE immunomagnetic beads (Milteny Biotec, Bergisch Gladbach, Germany) for cell sorting on AutoMACS (Milteny Biotec).

Isolation of DCs from spleen

Splenocytes were dissociated with Type IV collagenase (1 mg/ml; Sigma) in Hanks' balanced salt solution (HBSS) for 20 min at 37° and filtered out with a cell strainer (100-µm pores; BD Falcon) after cell homogenization. These cells were suspended in a dense bovine serum albumin (BSA) solution (ρ = 1·080), overlaid with an equal volume of RPMI-1640 medium, and centrifuged in a swing bucket rotor at 9500 g for 15 min at 4°. The interface cell fraction was collected and analysed for cell surface antigens by flow cytometry. Splenic CD11c⁺ cells were further sorted using either AutoMACS or Moflo as described above. For maturation, the CD11c⁺ sorted cells were further incubated with LPS (1 µg/ml) in RPMI-1640 containing 10% FCS at 37° for 24 hr. These experiments were repeated at least five times.

Isolation of monocyte-derived DCs

PBMC were isolated from heparinized venous blood from CMs by gradient centrifugation using Lymphoprep (ρ = 1·077; Nycomed, Oslo, Norway). Using anti-human CD14 mAb, the monocytes were purified by Moflo and cultured in cRMPI at a cell density of 5 × 10⁵ cells/ml in a 48-well plate (Costar Corp) for 7 days. For maturation, the 7-day culture was stimulated with LPS (1 µg/ml) and interferon (IFN)-γ (100 ng/ml) for another 24 hr. These experiments were repeated at least five times.

Flow cytometric analysis

Cells (1–5 × 10⁵) were stained with the above-mentioned mAbs in PBS supplemented with 0·5% BSA for 30 min at room temperature and washed with PBS. A flow cytometric analysis was performed using an EPICS XL (Beckman Coulter, Miami, FL) or a FACS Calibur (BD Biosciences, San Jose, CA). Results are given as the percentage positive minus the background from appropriate isotype controls. Representative findings from several independent experiments were used.

Analysis of the xenogeneic mixed leucocyte reaction (MLR)

Adult human T cells were purified from PBMC using a magnetic microbeads separation kit [MACS human Pan T-cell isolation Kit; Miltenyi Biotec] as responder cells. In this study, xenogeneic human T cells were used because of difficulties in obtaining enough allogeneic CM T cells, referring to O'Doherty's work.⁸ These responder cells (6 × 10⁴) were seeded into a 96-well plate (Costar Corp.) together with titrated numbers of irradiated DCs as stimulators in 200 µl of RPMI-1640 supplemented with 10% human AB serum. After 5 days of coculturing, the cells were pulsed with 10 mm 5-bromo-2′ deoxyuridine (BrdU) for 24 hr and subjected to a BrdU incorporation assay using a cell proliferation enzyme-linked immunosorbent assay (ELISA) BrdU kit (Roche, Nutley, NJ) to measure newly synthesized DNA. Briefly, the cells were dried (2 hr at 60°), fixed in 70% ethanol in HCl (0·5 N) for 30 min at −20°, and incubated with peroxidase-conjugated mouse anti-BrdU mAb (30 min at room temperature). The reaction of the luminal substrate was measured using a luminometer (ARVO mx/Light Luminescence counter; PerkinElmer Life Sciences, Wellesley, MA). These experiments were repeated three times.

ELISA analysis

In the MLR experiments, the supernatants (day 2) were analysed for human IFN-γ (Endogene, Rockford, IL) and IL-4 (eBioscience, San Diego, CA) using the ELISA kits. The culture supernatants of the CM DCs stimulated with either LPS (1 µg/ml) or LPS (1 µg/ml) and IFN-γ (100 ng/ml, PeproTech) for 24 hr were analysed for IL-12 (p70) using an ELISA kit (R & D Systems, Minneapolis, MN). Lysates of BM-derived DCs and spleen tissue were assayed for neurotrophic factor NT-3 and brain-derived neurotrophic factor (BDNF) using an ELISA kit (Emax Immuno Assay System; Promega, Madison, WI). These experiments were repeated three times.

Analysis of endocytotic activity

The endocytotic activity of the DCs was measured as described previously.³³ BM-derived CD11c⁺ cells were incubated with dextran-FITC (1 mg/ml; Sigma) at either 4° or 37° for 30 min in cRPMI. After washing in PBS, the cells were analysed using a FACS Calibur. For the immunocytochemical analysis, PE-labelled CD11c⁺ cells were incubated with dextran-FITC (1 mg/ml; Sigma) at either 37° or 4° for 2 hr.

Immunocytochemical analysis

Cells were fixed in 4% paraformaldehyde in PBS for 20 min at room temperature and incubated with PE-conjugated anti-CD11c mAb and FITC-conjugated anti-HLA-DR mAb in antigen dilution solution (DAKO Cytomation) for 2 hr at 37° followed by counterstaining with 4′-6-diamidino-2-phenylindole (DAPI). Images were obtained using a confocal scanning laser microscope (LMS510; Carl Zeiss, Tokyo, Japan).

Statistical analysis

All statistical analyses were performed using the Student t-test.

Results

Generation of BM-derived DCs

BM cells were cultured in the presence of rhGM-CSF and rhIL-4. On day 7, 2–5% of the non-adherent cultured BM cells (non-adherent BM) were CD11c⁺ HLA-DR⁺, indicating that more than 1 × 10⁷ of the CD11c⁺ HLA-DR⁺ cells were isolated from the CM specimen (Fig. 1a). More CD11c⁺ HLA-DR⁺ cells were generated in the presence of rhGM-CSF and rhIL-4 than in the presence of rhGM-CSF alone (data not shown). Although we also generated DCs from CD34⁺ BM cells, the number of CD11c⁺ HLA-DR⁺ cells generated from CD34⁺ BM cells was less than one-eighth of that from non-adherent BM cells. We then examined the change in phenotype of the CD11c⁺ cells generated from non-adherent BM and CD34⁺ BM cells following maturation with LPS stimulation. As shown in Fig. 1(a), with both methods, the LPS-stimulated CD11c⁺ cells showed a higher expression of CD80, CD83, CD86 and HLA-DR than the non-stimulated CD11c⁺ cells and these CD11c⁺ cells from non-adherent BM cells and CD34⁺ BM cells showed a similar expression pattern. Although approximately 20% of the non-stimulated CD11c⁺ HLA-DR⁺ gated cell population from non-adherent BM cells expressed CD14, these cells expressed CD1a (78%) and CD1c (93%), which are known as human DC markers,^37,38 but did not contain a CD3⁺ population (Fig. 1b). Therefore, we generated DCs from non-adherent BM cells for further analyses.

Figure 1.

Characterization of bone marrow (BM)-derived dendritic cells (DCs) from common marmosets (CMs). (a) Expression of CD80, CD83, CD86 and human leucocyte antigen (HLA)-DR in CD11c⁺ cells generated from non-adherent cultured BM cells (non-adherent BM) and CD34⁺ BM cells on day 7 in culture. For maturation, cultured BM cells were treated with lipopolysaccharide (LPS; 1 µg/ml) for another 24 hr. The numbers within the dot blots represent the percentages within the quadrant. (b) Expression of CD3⁺, CD1a, CD1c and CD14 in non-stimulated, non-adherent BM-derived CD11c⁺ HLA-DR⁺ cells (red line). Isotype controls are shown by a blue line. (c) Immunocytochemical analysis of BM-derived DCs. After stimulation with LPS for 24 hr, CD11c (red) and HLA-DR (green) were expressed on the cell surface of the dendrites. Scale bar, 10 µm. (d) Culture supernatants of BM-derived DCs treated with LPS for 24 hr were analysed for interleukin (IL)-12 using an enzyme-linked immunosorbent assay (ELISA).

Confocal imaging showed the colocalization of CD11c and HLA-DR on the surface of cells with numerous dendrites, a morphological characteristic of DCs (Fig. 1c). To examine the functional characteristics of BM-derived CM DCs, cytokine production, the ability to stimulate Xenogenic human T cells, and endocytotic activity were analysed. As shown in Figs 1(d) and 2, the LPS-stimulated CD11c⁺ cells secreted IL-12 and caused a proliferation of Xenogenic human T cells in a dose-dependent fashion, indicating their potency as antigen-presenting cells. The BM-derived CD11c⁺ cells incubated at 37° incorporated more dextran-FITC than the cells incubated at 4°, and the LPS-stimulated CD11c⁺ cells (mature type) showed a lower endocytotic capacity than the non-stimulated CD11c⁺ cells (immature type), consistent with the functional features of DCs (Fig. 3). Furthermore, an ELISA analysis revealed that human T cells cocultured with the LPS-stimulated CD11c⁺ cells secreted IFN-γ but not IL-4, suggesting that BM-derived CD11c⁺ cells in CMs could induce T helper type 1 (Th1) immune responses similar to those induced by human DC1 (Fig. 4). Taken together, the results from the phenotypic and functional analyses suggest that BM-derived CD11c⁺ HLA-DR⁺ cells have the characteristic features of DCs.

Figure 2.

Mature bone marrow (BM)-derived dendritic cells (DCs) from common marmosets (CMs) stimulated the proliferation of xenogeneic human pan T cells. BM-derived CD11c⁺ cells with or without lipopolysaccharide (LPS) stimulation (1 µg/ml) were cocultured at the indicated ratio with xenogeneic human pan-T cells (6 × 10⁴) for 5 days. A 5-bromo-2′ deoxyuridine (BrdU) incorporation assay showed that the BM-derived CD11c⁺ cells treated with LPS induced higher proliferation of Xenogenic T cells than the untreated BM-derived CD11c⁺ cells. The CD11c⁺ HLA-DR⁺ cells constituted approximately 2% of the untreated CD11c⁺ cell fraction and 14% of the LPS-treated CD11c⁺ cell fraction, respectively. The mean ± standard deviation for duplicate wells is shown. **P < 0·01. CPS, counts per second.

Figure 3.

Immature bone marrow (BM)-derived dendritic cells (DCs) exhibit endocytotic activity. (a) BM-derived CD11c⁺ cells stimulated with lipopolysaccharide (LPS; 1 µg/ml) or left untreated were incubated with dextran-fluorescein isothiocyanate (FITC) for 30 min at either 37° (shaded blue histograms) or 4° (green line) as a control for background passive uptake. The red line shows an isotype control. (b) Immature BM-derived DCs labelled with a phycoerythrin (PE)-conjugated anti-CD11c monocolonal antibody (mAb) were incubated with dextran-FITC at 37° or 4° for 2 hr. Confocal microscopy imaging showed fluorescent microspheres (green) in the cytoplasm of immature BM-derived CD11c⁺ cells (red) incubated at 37°, but not in cells incubated at 4°. Scale bar, 10 µm.

Figure 4.

Interferon (INF)-γ secretion from human T cells in a xenogeneic mixed leucocyte reaction (MLR). The supernatants (2 days) of human pan T cells (6 × 10⁴ cells/well) cocultured with irradiated lipopolysaccharide (LPS)-treated bone marrow (BM)-derived CD11c⁺ cells (3 × 10³ cells/well) were assayed for INF-γ and interleukin (IL)-4 using an enzyme-linked immunosorbent assay (ELISA). Data are shown as the mean ± standard deviation.

Isolation of splenic DCs

After the splenocytes had been dissociated, a low density cell fraction was collected by centrifugation with dense BSA, based on a method used to isolate mouse splenic DCs.^39,40 For the CM specimens, 2–4 × 10⁵ CD11c⁺ cells were isolated from 1 × 10⁷ of the BSA fractioned cells. As shown in Fig. 5(a), the CD11c⁺ CD8⁺ population was present in the CM spleen; this cell population is found in mouse DCs. In mouse splenic DCs, the number of CD11c⁺ CD4⁺ DCs is approximately 2–3 times that of CD11c⁺ CD8⁺ DCs.^41,42 In contrast, the number of CD11c⁺ CD8⁺ cells was similar to the number of CD11c⁺ CD4⁺ cells in the CM spleen (Fig. 5a). The CD11c⁺ CD4⁺ cell population has been identified in human splenic DCs,⁴³ whereas the cells of the monocyte/macrophage lineage also express CD11c and CD4. Therefore, CD11c⁺ CD4⁺ cells in the CM spleen may contain some monocytes/macrophages. In the splenic CD11c⁺ cells, the expression of HLA-DR and CD86 was increased by LPS stimulation (Fig. 5b). Upon LPS stimulation, the splenic CD11c⁺ cells secreted higher amounts of IL-12 than the untreated CD11c⁺ cells (Fig. 5c). The MLR analysis revealed that these CD11c⁺ cells treated with LPS induced a stronger effect on the proliferation of xenogeneic human T cells (Fig. 6). Thus, the splenic CD11c⁺ cells exhibited the functional features of DCs.

Figure 5.

Characterization of splenic dendritic cells (DCs) from common marmosets (CMs). (a) Flow cytometry analysis of a low density cell fraction of splenocytes in dense bovine serum albumin (BSA). (b) The expression of human leucocyte antigen (HLA)-DR and CD86 increased in splenic CD11c⁺ cells stimulated in culture with lipopolysaccharide (LPS; 1 µg/ml) for 24 hr. (c) Culture supernatants of splenic CD11c⁺ cells treated with LPS for 24 hr were analysed for interleukin (IL)-12 using an enzyme-linked immunosorbent assay (ELISA).

Figure 6.

Xenogenic mixed leucocyte reaction (MLR) stimulatory activity of splenic dendritic cells (DCs) from common marmosets (CMs). Splenic CD11c⁺ cells treated with or without lipopolysaccharide (LPS) stimulation (1 µg/ml) were cocultured at the indicated ratio with 6 × 10⁴ human pan T cells for 5 days. The 5-bromo-2′ deoxyuridine (BrdU) incorporation assay showed that LPS-stimulated splenic CD11c⁺ cells increased the proliferation of Xenogenic human T cells. The mean ± standard deviation for duplicate wells is shown. **P < 0·01.

Generation of monocyte-derived DCs

Approximately 5 × 10⁵ of the CD14⁺ monocyte cells, less than 5% of the PBMC, were isolated from 10 to 15 ml of CM peripheral blood and were cultured in cRPMI including rhGM-CSF and rhIL-4 for 7 days. Although the yield depended on the individual CM specimens and approximately 20% of the cultured cells still expressed CD14⁺, 1–2 × 10⁵ CD11c⁺ HLA-DR⁺ cells showing the morphological characteristics of DCs were generated from one CM (data not shown). In monocyte-derived CD11c⁺ cells, the expression of both HLA-DR and CD86 was increased by LPS and IFN-γ costimulation (Fig. 7a). IL-12 was also secreted from these cultured cells after stimulation with LPS and IFN-γ (Fig. 7b). Moreover, these monocyte-derived cells stimulated the proliferation of xenogeneic human T cells in an MLR assay, indicating that the cells exhibited the functional features of DCs (Fig. 8).

Figure 7.

Characterization of monocyte-derived dendritic cells (DCs) from common marmosets (CMs). (a) Cultured CD14⁺ cells from peripheral blood mononuclear cells (PBMC) (day 7) were stimulated with lipopolysaccharide (LPS; 1 µg/ml) and interferon (IFN)-γ (100 ng/ml) for another 24 hr. The expression of human leucocyte antigen (HLA)-DR and that of CD86 in monocyte-derived CD11c⁺ cells were up-regulated by LPS and IFN-γ stimulation. (b) Cultured monocytes (day 7) were treated with LPS and IFN-γ for another 24 hr, and the supernatant was tested for interleukin (IL)-12 using an enzyme-linked immunosorbent assay (ELISA). Data are shown as the mean ± standard deviation. **P < 0·01.

Figure 8.

Monocyte-derived dendritic cells (DCs) from common marmosets (CMs) stimulated the xenogeneic mixed leucocyte reaction (MLR). Cultured monocyte-derived DCs (day 8) treated with or without lipopolysaccharide (LPS; 1 µg/ml) and interferon (IFN)-γ (100 ng/ml) for 24 hr were cocultured at the indicated ratio with 6 × 10⁴ Xenogenic human pan T cells for 5 days. The proliferation of human T cells was then assessed by 5-bromo-2′ deoxyuridine (BrdU) incorporation. The mean ± standard deviation of duplicate wells is shown. **P < 0·01.

Production of NT-3 in BM-derived DCs

We previously demonstrated that mouse splenic DCs secrete the neurotrophic factor NT-3.⁷ In this study, we found that BM-derived DCs from CMs also produced NT-3 (Fig. 9). In contrast, no production of BDNF was observed from BM-derived DCs from CMs (data not shown).

Figure 9.

Bone marrow (BM)-derived dendritic cells (DCs) produced neurotrophic factor NT-3. Lysates of BM-derived DCs and spleen tissue were assayed for NT-3 using an enzyme-linked immunosorbent assay (ELISA). Data normalized by the amount of protein are shown as the mean ± standard error of the mean.

Discussion

In this study, we developed methods to isolate DCs from three different CM tissues and described their phenotype and functional capability as antigen-presenting cells. In CMs, functional analysis of DCs isolated from different tissues has previously been limited. We demonstrated that CM DCs expressed CD11c, similar to the results obtained in rhesus monkeys and humans.^43–45 Most human DCs, with the exception of plasmacytoid DCs (pDCs), express CD11c.⁴¹ In addition, some populations of macrophages and B cells are stained with human anti-CD11c mAb.⁴³ Therefore, we purified CD11c⁺ HLA-DR⁺ cells from floating cultured BM cells to avoid contamination with fibroblasts and macrophages, which have a greater ability to attach to the dish surface than DCs. The BM-derived CD11c⁺ cells contained populations expressing CD1a, CD1c (BDCA-1) and CD83, typical markers of human DCs,^37,38 indicating the similarity of CM DCs to human DCs. However, the expression of HLA-DR in BM-derived CD11c⁺ cells was not high, even in mature cells. Two subsets of human DC precursors derived from cord blood CD34⁺ cells have been identified by the exclusive expression of CD1a and CD14 at early time-points (days 5–7) in culture, and both precursor subsets mature at days 12–14 into DCs with typical morphology and phenotype: CD1a⁺ CD14^– and CD1a^– CD14⁺ precursor cells differentiate into Langerhans-type and dermal-type DCs, respectively.⁴⁶ In this study, approximately 20% of BM-derived CD11c⁺ HLA-DR⁺ cells expressed CD14 at day 7 in culture. Taken together, these results suggest that the BM-derived CD11c⁺ HLA-DR⁺ cells in the CM may contain DC precursors that exhibit relatively low expression of HLA-DR, and it may be necessary to culture them for longer times to allow them to reach maturation. It is also possible that LPS stimulation alone was not enough to allow full maturation of BM-derived DCs. Further studies are required to optimize the method for enriching immature and mature DCs derived from BM cells in the CM. Regarding the DC subtypes in CMs, we could not analyse pDCs, which may be present in peripheral blood and spleen tissue, because of a lack of available CD123 mAb in our study. To analyse the DC subtypes in the CM, more mAbs are needed that are CM-specific.

In CM, we isolated DCs from the BM, spleen and PBMC, and these DCs exhibited the functional features of antigen-presenting cells. More than 1 × 10⁷ BM-derived DCs can be isolated from a CM specimen, and the yield from BM was 25–50 times higher than that obtained from spleen or PBMC. The yield of BM-derived DCs should be sufficient for preclinical studies on cell therapy in CM models, considering the number of DCs used for cancer immunotherapy in humans⁵ or for SCI therapy in mice.⁷ However, in vivo administration of Flt-3 ligand should be performed to determine whether this manipulation can increase the number of DCs, as reported in rhesus monkeys.⁴⁷ In view of its accessibility and lack of allograft problems, peripheral blood is a useful source of DCs; however, the amount of blood that can be isolated from a CM is limited. In this study, the maximum amount of blood obtained from a CM was 10 ml. The population of circulating human DCs in PBMC is less than 1%;⁴⁸ therefore, peripheral blood is not a good source of DCs isolated from CMs. Recently, DCs were generated from murine embryonic stem (ES) cells.⁴⁹ An ES cell line in CMs has been established;⁵⁰ thus, it may be possible to generate DCs from ES cells obtained from CMs.

Taken together, our results show that the characteristics of CM DCs resemble those of human DCs, suggesting the usefulness of CM DCs for preclinical studies on cell therapy. Moreover, we showed that BM-derived DCs from CMs also produce NT-3, an important neurotrophic factor for CNS regeneration. We are considering a preclinical study on cell therapy using BM-derived CD11c⁺ HLA-DR⁺ DCs for the treatment of SCI in CMs to evaluate the therapeutic effects and safety of this procedure.