SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. References

Murine bone marrow-derived dendritic cells (DC) can be generated by culture in the presence of granulocyte/macrophage colony-stimulating factor (GM-CSF) alone or GM-CSF in conjunction with interleukin-4 (IL-4). However, these two culture methods result in the production of heterogeneous DC populations with distinct phenotypic and stimulatory properties. In this study, we investigated the properties of DC generated under serum-free conditions in the presence or absence of IL-4 and compared their yield and phenotype to that of DC generated in the presence of fetal calf serum (FCS) (±IL-4). We did not observe a significant difference in the total cell yield between these four culture conditions, although the proportion of CD11c+ DC in cultures that received FCS was higher than that of their counterparts generated under serum-free conditions. Also, the four culture conditions generated CD11c+ DC with comparable levels of major histocompatibility complex (MHC) class II, CD40, CD80 and CD86 expression, with the exception of cells cultured under serum-free conditions in the absence of IL-4, which displayed suboptimal levels of these markers. Moreover, we compared the functional and stimulatory properties of DC generated under serum-free conditions in the presence or absence of IL-4. DC cultured in the presence of IL-4 were stronger stimulators of allogeneic splenocytes in a primary mixed lymphocyte reaction (MLR) and of naïve antigen-specific OT-II transgenic T cells when pulsed with the class II ovalbumin (OVA)323−339 peptide or whole OVA protein than DC cultured in the absence of IL-4. However, both DC populations displayed a similar capacity to take up fluorescein isothiocyanate (FITC)–albumin by macropinocytosis and FITC–Dextran by the mannose receptor and to secrete IL-12 in response to stimulation with lipopolysaccharide (LPS) or an agonistic anti-CD40 monoclonal antibody. Therefore, we conclude that although both DC culture methods result in the production of DC with similar functional abilities, under serum-free conditions, DC cultured in GM-CSF and IL-4 show an increased stimulatory potential over DC cultured in GM-CSF alone. This is an important consideration in the design of experiments where DC are being exploited as immunotherapeutic vaccines.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. References

Dendritic cells (DC) are extremely potent antigen-presenting cells capable of priming naïve T cells and directing primary immune responses [1, 2]. DC reside throughout peripheral tissues at very low concentrations, where they continuously sample the local antigenic environment. On encountering antigen and/or inflammatory cytokines [3], these immature DC traffic to regional lymph nodes, upregulate their surface expression of major histocompatibility complex (MHC) and costimulatory molecules and lose much of their ability to take up antigen as they migrate. Now phenotypically mature, the DC interact with T cells to produce antigen-specific tolerance or immunity depending on the antigen they are presenting and whether they are partially or fully functionally mature, as determined by their ability to release proinflammatory cytokines [4].

Advances in the generation of large numbers of DC from mouse bone marrow by using granulocyte/macrophage colony-stimulating factor (GM-CSF) alone [5, 6] or GM-CSF in conjunction with interleukin-4 (IL-4) [7] have allowed detailed studies into the biological function of DC. In particular, they are being exploited as vaccines for tumour therapy in several murine models [8–11]. Interestingly, unlike the human setting where IL-4 is essential for DC generation from DC precursors such as blood monocytes, murine studies still have a propensity to employ both culture techniques when using bone marrow precursors, despite important differences in the phenotype and functional capabilities of DC cultured in the presence or absence of IL-4 [7, 12–18]. DC generated in the absence of IL-4, for example, are often referred to as immature [13, 14, 16] and have been used in transplant studies, where they appear to prolong the survival of cardiac allografts in mice [16]. All of these studies, however, have been performed only with DC cultured in the presence of fetal calf serum (FCS).

The issues concerning FCS have been recently highlighted in two DC vaccination-based anti-tumour studies for melanoma in mice [19, 20]. These studies show that when both DC and B16 melanoma cells are cultured in the presence of FCS, the vaccination of mice with DC that have not been subject to prior loading with B16 antigens subsequently induces strong anti-B16 immune responses as a result of the association of B16 cells with FCS. Such responses may be wrongly interpreted as true tumour immunity [20]. Increasingly in human studies, the use of FCS to generate DC is being substituted with autologous plasma [21–23] or albumin [24], which results in the production of more clinically relevant data by reducing the incidence of nonspecific immune responses [21]. In the murine setting, FCS has been substituted for mouse serum, which, despite a lower DC yield, resulted in phenotypically and functionally normal DC capable of migratory activity in vivo[25].

In this study, we compared the yield, morphology and phenotype of DC generated from mouse bone marrow in either GM-CSF alone or GM-CSF and IL-4, in the presence or absence of immunogenic xenogenic serum antigens (FCS). Also, we compared the functional abilities of DC generated in the presence or absence of IL-4 under serum-free conditions. The four culture conditions did not produce a significant difference in cell yield after 7 days in culture; however, the proportion of CD11c+ DC within each culture was subject to considerable variation. DC populations cultured under serum-free conditions in the presence or absence of IL-4 displayed a similar ability to take up antigen by each of two different mechanisms and to produce IL-12 in response to maturation stimuli. However, they differed not only in their surface expression of MHC and costimulatory molecules, but also in their ability to stimulate naïve allogeneic splenocytes and prime naïve CD4+ transgenic T cells. These culture conditions will have implications for the interpretation of data when using DC for anti-tumour therapy and will allow a more realistic comparison when studying DC in human clinical trials.

Materials and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. References

Mice and cells C57BL/6 (H-2b) and C3H/HeN (H-2k) mice were purchased from Harlan (Oxford, UK). OT-II (H-2b) mice [26], which express a transgenic T-cell receptor (TCR) specific for MHC II/ovalbumin (OVA)323−339 peptide (ISQAVHAAHAEINEAGR) [27], were kindly donated by Prof Mike Kemeny and Dr Alistair Noble (Department of Immunology, Guy's, King's and St Thomas' School of Medicine, King's College London, London, UK). All mice used were 6- to 8-week-old females. Primary cells were cultured in serum-free X-VIVO-15 medium (Biowhittaker, Walkersville, MD, USA) containing 100 U/ml of penicillin and 100 μg/ml of streptomycin (both from Sigma, Poole, UK) or RPMI supplemented with 10% FCS, 100 U/ml of penicillin and 100 μg/ml of streptomycin.

Antibodies and other reagents Fluorescein isothiocyanate (FITC)-conjugated monoclonal antibodies (MoAbs) 16-10A1 (anti-CD80), AF6-120.1 (anti-MHC class II; I-Ab), RB6-8C5 (anti-Gr-1) and 3E2 (anti-ICAM-1) and phycoerythrin (PE)-conjugated MoAbs GL1 (anti-CD86), 3/23 (anti-CD40), rmC5-3 (anti-CD14) and M1/70 (anti-CD11b) were purchased from PharMingen (San Diego, CA, USA). Biotinylated HL3 (anti-CD11c) and streptavidin-peridinin chlorophyll protein (PerCP) were also purchased from PharMingen. FITC-conjugated CI:A3-1 (anti-F4/80) was purchased from Serotec (Oxford, UK). Non-labelled CT-17.1/CT-17.2 (mouse anti-mouse FcRγII/III-CD16/CD32) was obtained from Caltag Laboratories (Caltag-Medsystems, Silverstone, UK) and was detected using FITC-conjugated rabbit F(ab′)2 anti-mouse immunoglobulin G (IgG) (#F0313, Dako, Glostrup, Denmark). Chicken OVA323−339 peptide was kindly donated by Dr Alistair Noble and GAD65171−190 (IKTGHPRYFNQLSTGLDMVG) peptide by Dr Tim Tree (Department of Immunobiology, Guy's, King's and St Thomas' School of Medicine, New Guy's House, Guy's Hospital, London, UK).

Culture of bone marrow-derived dendritic cells Femurs and tibiae were removed from 6- to 8-week-old female C57BL/6 mice, scraped free of muscle tissue and placed in 70% ethanol for 1 min before being rinsed with medium. The ends of the bones were gently cut off, and the bone marrow was flushed out with medium using a 27.5-gauge Microlance-3 needle (Becton Dickinson, Oxford, UK) and a 5-ml syringe (Terumo, Leuven, Belgium), with the resulting cell suspension allowed to drip through a 70-μm cell strainer (Falcon, Becton Dickinson) to remove pieces of bone. Where comparisons needed to be made between culture conditions, PBS was used to flush out the bone marrow from one mouse/experiment, and after cell counting, the cells were separated into four equal aliquots, pelleted and resuspended in the appropriate media. Cells were plated out in 90-mm triple-vent bacteriological Petri dishes (Bibby Sterilin, Stafford, UK) at 7.5 × 105 cells/ml, 10 ml total volume. Plates received 5 ng/ml of GM-CSF (R&D Systems, Abingdon, UK), 10 ng/ml of IL-4 (PeproTech EC, London, UK) and 50 μm 2-mercaptoethanol (Sigma) and were incubated at 37 °C (day 0). On day 3, the plates received an additional 10 ml of fresh medium (containing 50 μm 2-mercaptoethanol) and a further 5 ng/ml of GM-CSF and 10 ng/ml of IL-4 (based on the final plate volume). On day 6, the plates received a further 5 ml of fresh medium and 1 ng/ml of GM-CSF (based on the final plate volume). Non-adherent DC were gently harvested for use on day 7.

Cell labelling and fluorescence-activated cell sorter analysis For the detection of cell-surface antigens, Fc-receptor activity was blocked by incubating DC with unlabelled anti-CD16/CD32 antibody for 10 min at room temperature. DC were then incubated with biotinylated anti-CD11c and the relevant FITC- or PE-conjugated antibodies for 20 min at 4 °C. The cells were then washed and incubated with streptavidin-PerCP for a further 20 min and analysed using a BD FACSCalibur (flow cytometer equipped with a 15-mW 488 nm air-cooled argon-ion laser, Becton Dickinson).

Phagocytosis of antigen DC were incubated at 37 °C for 45 min with 1 mg/ml of FITC–Dextran or 0.5 mg/ml of FITC–Albumin (both from Sigma). Controls were set up at 4 °C to detect any nonspecific adherence of FITC–Dextran or FITC–Albumin to the surface of the cells. Cells were then washed twice with HBSS/1% FCS and analysed by flow cytometry.

Allogeneic mixed lymphocyte reaction C57BL/6 DC were plated out in triplicate in a 96-well U-bottomed plate (Cellstar, Greiner Bio-One, Stonehouse, UK) and serially diluted 1 : 2 from 1 × 104 cells/well down to 6.25 × 102 cells/well. Splenocytes were harvested from the spleens of naïve C3H/HeN mice and were depleted of red blood cells by using Red Blood Cell Lysing Buffer (0.155 m ammonium chloride in 0.01 m Tris–HCl buffer, Sigma) and depleted of B cells by using B220 Dynabeads (Dynal, Oslo, Norway), according to the manufacturers' instructions. The residual splenocytes were then added to the plate at 1 × 105 cells/well (final well volume was 200 μl/well). Wells were set up as controls that contained either DC or splenocytes alone. The wells were kept at 37°C for 5 days, and splenocyte proliferation was assessed by [methyl-3H]thymidine (1 μCi/well, Amersham Pharmacia Biotech, Little Chalfont, UK) incorporation over 6 h. The cells were harvested onto filter discs and counted using a liquid scintillation analyser (TRI-CARB 2200CA, Packard, Downers Grove, IL, USA).

OT-II T-cell proliferation assay Naïve OT-II CD4+ T cells (3 × 104), negatively purified from the spleen, inguinal and mesenteric lymph nodes by using a CD4+ T-cell Isolation Kit (Miltenyi Biotech, Bisley, UK), were added to 1 × 104 DC in each well of a 96-well U-bottomed plate (Cellstar, performed in triplicate). OVA323−339 peptide (ISQAVHAAHAEINEAGR) (0.01–10 μg/ml) or whole OVA protein (1–1000 μg/ml) (Sigma) was added to the wells in a final volume of 200 μl/well. GAD65171−190 peptide (25 μg/ml) and bovine serum albumin (BSA) (1000 μg/ml, Sigma) were used as controls. Plates were incubated at 37 °C for 90 h (peptide) or 96 h (protein), with T-cell proliferation assessed by [methyl-3H]thymidine (1 μCi/well, Amersham Pharmacia Biotech) incorporation over the final 6 h (protein) or 18 h (peptide) of culture. The cells were harvested onto filter discs and counted using a liquid scintillation analyser (TRI-CARB 2200CA, Packard).

IL-12 analysis DC were resuspended in fresh medium at 1 × 106 cells/ml and stimulated with 500 ng/ml of lipopolysaccharide (LPS) (Sigma), 250 U/ml of tumour necrosis factor-α (TNF-α) (R&D Systems) or 10 μg/ml of 1C1O MoAb (agonistic anti-CD40, R&D Systems). Culture supernatant was removed after 24 h and frozen until ready for analysis. IL-12 was assessed by enzyme-linked immunosorbent assay (ELISA) using a kit from PeproTech EC with the following modifications of the manufacturer's protocol: A 96-well ELISA plate (F96 Maxisorp, Nunc, Roskilde, Denmark) was coated with 0.25 μg/ml of capture antibody diluted in PBS (100 μl/well), sealed and incubated overnight at room temperature. The following day, the plate was blocked with 1% BSA (Sigma) in PBS (250 μl/well) for 1 h at room temperature. Standards (0–2 ng/ml, serially diluted in diluent; 0.05% Tween-20, 0.1% BSA) and samples (including dilutions) were added in duplicate (100 μl/well) for 1.5 h at room temperature. The plate was then incubated with 0.25 μg/ml of the detection antibody (biotinylated antigen-affinity purified goat anti-mIL-12) in diluent (100 μl/well) for 1.5 h at room temperature. Later, the plates were incubated with streptavidin–horseradish peroxidase (HRP) (100 μl/well) (R&D Systems), diluted 1 : 200 in diluent, for 20 min at room temperature in the dark. Following each of these incubations, the plate was washed four times with wash buffer. Each well was then incubated with 100 μl of a 1 : 1 mixture of colour reagent A and colour reagent B (R&D Systems) for 20 min at room temperature in the dark to allow colour to develop. The reaction was stopped using 50 μl/well of 2 m H2SO4. The optical density of the plate was read on a Precision Microplate Reader (Molecular Devices, Menlo Park, CA, USA) at 450 nm, with wavelength correction set at 650 nm, and the data were analysed using softmax version 2.35 (Molecular Devices).

Statistical analysis Standard error of the mean (SEM) and significance values were calculated using graphpad prism version 3.02 for Windows (GraphPad Software, San Diego, CA, USA). Statistical comparisons were performed using an unpaired Student's t-test. P-values of <0.05 (*), <0.005 (**) and <0.001 (***) were considered significant.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. References

Morphology, yield and the proportion of CD11c+ cells

Dendritic cells were cultured from the bone marrow in GM-CSF in the presence (RPMI/FCS) or absence (X-VIVO) of FCS and in the presence (+IL-4) or absence (–IL-4) of IL-4. Clear differences could be seen in the morphology of cell cultures stimulated with GM-CSF alone or GM-CSF and IL-4 (observed in both RPMI/FCS and X-VIVO cultures; data not shown). DC cultures generated in GM-CSF alone comprised mainly of a monolayer of uniformly sized cells that displayed little evidence of protruding dendrites. By contrast, DC cultures generated in GM-CSF and IL-4 contained large clusters of cells, with clear evidence of protruding dendrites. Moreover, these cultures contained a small proportion of very small cells that appeared to be undifferentiated bone marrow precursors.

To investigate the impact of IL-4 and FCS on DC yield, we took bone marrow from a single C57BL/6 mouse and divided it into four equal aliquots, which were stimulated for 7 days with GM-CSF in the presence or absence of FCS and in the presence or absence of IL-4 (Fig. 1A). The total cell yield from each bone marrow culture was found to be consistently similar.

image

Figure 1. Total cell yield and proportion of CD11c+ cells in bone marrow dendritic cell (DC) cultures. Bone marrow from one C57BL/6 mouse was separated into four equal aliquots and stimulated with granulocyte/macrophage colony-stimulating factor ± interleukin-4 (GM-CSF ± IL-4) in the presence [RPMI/fetal calf serum (FCS)] or absence (X-VIVO) of FCS for 7 days. (A) Total cell counts following harvest of all non-adherent cells and (B) percentage of CD11c+ cells following subsequent analysis by flow cytometry. Data represent average values ± standard error of the mean derived from three separate experiments.

Download figure to PowerPoint

In the absence of FCS, the proportion of CD11c+ cells in the cultures (±IL-4) was found not to be significantly different (P = 0.1266, Fig. 1B); however, there were differences in the proportion of CD11c+ cells within cultures which did or did not receive FCS. In cultures which did not receive IL-4, a significantly higher proportion of CD11c+ cells were found in cultures that had received FCS when compared to cultures that did not receive FCS (P < 0.05). A similar correlation was observed between cultures which did or did not receive FCS in the presence of IL-4 (P < 0.01). Because there was no significant difference in the proportion of CD11c+ cells between cultures that were grown in X-VIVO and did not receive IL-4, and cultures that were grown in RPMI/FCS and did receive IL-4 (P = 0.5881), we conclude that the significantly higher proportion of CD11c+ cells present in cultures that received FCS when compared with cultures that did not receive FCS under the same cytokine conditions is not simply a reflection of the poor ability of X-VIVO medium to support the development of CD11c+ cells. Interestingly, it appears that IL-4 may inhibit the differentiation of some bone marrow precursors into CD11c+ cells because cultures that received FCS but not IL-4 produced a significantly higher proportion of CD11c+ cells than cultures that received both FCS and IL-4 (P < 0.01).

Influence of culture conditions on the phenotype of CD11c+ DC

To further characterize the CD11c+ DC generated under the four culture conditions, we analysed the cells by flow cytometry for the presence of costimulatory molecules and myeloid markers. In cultures grown in the absence of serum (X-VIVO), IL-4 had a potent influence on the level of CD40, CD80, CD86 and MHC class II expression (Fig. 2), resulting in higher levels of each of these markers. In contrast, the expression levels of these markers were very similar in cultures grown with or without IL-4 in the presence of FCS (RPMI/FCS). Cells cultured in the presence of FCS but in the absence of IL-4 also displayed a higher level of marker expression than cells cultured in the absence of both IL-4 and FCS, suggesting that FCS itself may also have a direct influence on the expression of CD40, CD80, CD86 and MHC class II. However, cells cultured in both IL-4 and FCS showed only a marginal increase in CD40, CD80 and CD86 expression and a marginal decrease in MHC class II expression when compared with cells cultured in IL-4 without FCS, suggesting that IL-4 and FCS do not have a cumulative effect on the expression level of these costimulatory molecules.

image

Figure 2. Surface expression of CD40, CD80, CD86 and major histocom patibility complex (MHC) II by bone marrow-derived CD11c+ dendritic cells (DC). DC were harvested following stimulation with granulocyte/macro phage colony-stimulating factor ± interleukin-4 (GM-CSF ± IL-4) in the presence [RPMI/fetal calf serum (FCS)] or absence (X-VIVO) of FCS and labelled with CD11c and CD40, CD80, CD86 or MHC class II monoclonal antibodies (MoAbs). Dot plots show all culture-derived cells after gating out dead cells and are representative of three separate experiments with similar results. The numbers in each profile indicate the percentage of total CD11c+ cells staining for each marker.

Download figure to PowerPoint

The expression of intercellular adhesion molecule-1 (ICAM-1) was also found to have a similar expression pattern to that of CD40, CD80, CD86 and MHC class II (Table 1). However, CD11c+ cells from each culture condition showed a similar level of CD11b expression and a similarly low level of CD14 (monocyte marker), Gr-1 (granulocyte marker) and F4/80 (macrophage marker) expression, confirming that these CD11c+ cells are DC. Interestingly, CD11c+ DC cultured in the absence of FCS, regardless of the presence or absence of IL-4, were found to express slightly elevated levels of FcRγII/III when compared to CD11c+ DC cultured in the presence of FCS.

Table 1.  Expression of adhesion and myeloid markers by bone marrow-derived CD11c+ dendritic cells (DC) cultured under different conditions
MarkerX-VIVO − IL-4X-VIVO + IL-4RPMI/FCS − IL-4RPMI/FCS + IL-4
  1. FCS, fetal calf serum; ICAM-1, intercellular adhesion molecule-1; IL-4, interleukin-4. Values are percentage of total CD11c+ cells ± standard error of the mean (SEM).

CD11b48 ± 146 ± 149 ± 146 ± 1
CD141 ± 11 ± 10 ± 00 ± 0
FcRγII/III14 ± 616 ± 32 ± 15 ± 1
Gr-111 ± 45 ± 23 ± 15 ± 1
ICAM-111 ± 320 ± 125 ± 524 ± 3
F4/801 ± 11 ± 10 ± 00 ± 0

Taken together, analysis of bone marrow cultures stimulated in the presence or absence of FCS and in the presence or absence of IL-4 showed that each culture method has the potential to induce the differentiation of CD11c+ DC. However, in an attempt to move away from nonspecific immune responses reportedly induced by the addition of FCS into bone marrow cultures, we decided to focus our investigation on the properties of bone marrow-derived DC cultured under serum-free conditions.

Functional properties of DC cultured under serum-free conditions

A major feature of DC is the ability to take up antigen. To assess the ability of DC cultured under serum-free conditions to take up antigen by two distinct mechanisms, we incubated, at 37 °C, DC cultured in the presence or absence of IL-4 with either FITC–Dextran or FITC–albumin (Fig. 3) for 45 min (controls were set up at 4 °C to monitor nonspecific uptake) and analysed DC by flow cytometry. DC from the two cultures were found to be consistent in their ability to take up FITC–albumin via macropinocytosis or take up FITC–Dextran via the mannose receptor [28].

image

Figure 3. Dendritic cells (DC) cultured in granulocyte/macrophage colony-stimulating factor ± interleukin-4 (GM-CSF ± IL-4) under serum-free conditions appear to exhibit a similar ability to take up antigen. DC cultured with or without IL-4 were removed from culture and incubated with fluorescein isothiocyanate (FITC)–Dextran or FITC–Albumin at 4 °C (shaded histograms) or 37 °C (black lines) for 45 min. Cells were then washed twice and assayed by flow cytometry. These graphs show a representative of three separate experiments with similar results.

Download figure to PowerPoint

Another important quality of DC is the ability to secrete the T-helper 1 (Th1)-polarizing cytokine IL-12. To investigate the capacity of DC cultured under serum-free conditions in the presence or absence of IL-4 to secrete IL-12 in response to maturation factors, we removed DC from culture, resuspended DC in fresh medium and then re-cultured DC for a further 24 h with LPS, TNF-α or an agonistic anti-CD40 MoAb. Culture supernatants were collected for the determination of IL-12 by ELISA (Fig. 4). There was no significant difference (P = 0.05) in the levels of IL-12 secreted by the unstimulated (control) DC from the two populations. In addition, although both DC populations produced a significant increase in IL-12 when stimulated with LPS or anti-CD40 MoAb when compared to unstimulated DC, the levels of IL-12 did not vary significantly between the two DC populations. However, DC generated in the absence of IL-4 produced a significant (P < 0.005) increase in IL-12 when stimulated with TNF-α (compared to control DC), whereas DC generated in the presence of IL-4 did not.

image

Figure 4. Interleukin-12 (IL-12) release in response to a 24-h coculture with lipopolysaccharide (LPS), tumour necrosis factor-α (TNF-α) or anti-CD40 monoclonal antibody (MoAb). Dendritic cells (DC) (1 × 106) cultured under serum-free conditions in the presence (▪) or absence (□) of interleukin-4 (IL-4) were incubated with 500 ng/ml of LPS, 250 U/ml of TNF-α or 10 μg/ml of anti-CD40 MoAb for 24 h. Supernatant samples were analysed using a standard IL-12 enzyme-linked immunosorbent assay (ELISA) kit. The graph represents average values of multiple experiments (n = 3 to 7), with error bars representing standard error of the mean (SEM). **P < 0.005. ***P < 0.001.

Download figure to PowerPoint

Stimulatory properties of DC cultured under serum-free conditions

The ability of the two DC populations to stimulate a proliferative response from naïve allogeneic B-cell–depleted splenocytes was studied in a primary mixed lymphocyte reaction (MLR) (Fig. 5). At DC-to-splenocyte ratios of 1 : 10 and 1 : 20, DC generated in the presence of IL-4 were significantly (P < 0.001 and P < 0.05, respectively) stronger stimulators of naïve allogeneic splenocytes than DC generated in the absence of IL-4. This result is in accordance with the finding of the increased expression levels of MHC class I (data not shown), MHC class II and costimulatory molecules displayed by DC generated in the presence of IL-4.

image

Figure 5. Dendritic cells (DC) cultured in the presence of interleukin-4 (IL-4) induce a stronger proliferative response by naïve allogeneic splenocytes than DC cultured in the absence of IL-4. C57BL/6 DC cultured in the presence (▪) or absence (□) of IL-4 were cocultured with naïve C3H/HeN B-cell–depleted splenocytes for 5 days at 37 °C. Proliferation was measured by the uptake of [methyl-3H]thymidine over 6 h. The graph represents combined data from three separate experiments carried out in triplicate (n = 9). Error bars represent standard error of the mean (SEM). *P < 0.05. ***P < 0.001.

Download figure to PowerPoint

To further compare the stimulatory properties of DC cultured with or without IL-4 under serum-free conditions, we compared their ability to process and present antigen to naïve T cells. DC were cocultured with negatively purified CD4+ OT-II transgenic T cells and various concentrations of OVA323−339 peptide or whole OVA protein. DC generated in the presence of IL-4 and pulsed with OVA323−339 peptide (Fig. 6A) were consistently significantly (P < 0.001) better at priming naïve OT-II T cells to proliferate than DC generated in the absence of IL-4. Similarly, DC generated in the presence of IL-4 were also superior at priming naïve OT-II T cells to proliferate (P < 0.05) than DC generated in GM-CSF alone when pulsed with various concentrations (1–100 μg/ml) of whole OVA protein (Fig. 6B). However, there was no significant difference (P = 0.1229) in the ability of the two DC populations to stimulate naïve OT-II T cells to proliferate when pulsed with 1000 μg/ml of OVA protein, which perhaps suggests that a high antigen dose may overcome the inferior priming capacity of DC cultured in the absence of IL-4.

image

Figure 6. Antigen processing and presentation by dendritic cells (DC) cultured in granulocyte/macrophage colony-stimulating factor ± interleukin-4 (GM-CSF ± IL-4) under serum-free conditions. DC cultured in the presence (▪) or absence (□) of IL-4 were cocultured with naïve CD4+ OT-II lymphocytes for 90 h [ A, ovalbumin (OVA)323−339 peptide] or 96 h (B, OVA protein) at 37 °C. Proliferation was measured by the uptake of [methyl-3H] thymidine over the final 6 (protein) or 18 (peptide) hours. Graphs represent combined data from two separate experiments carried out in triplicate. Error bars represent standard error of the mean (SEM). *P < 0.05. **P < 0.005. ***P < 0.001. BSA, bovine serum albumin.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. References

In the past, murine DC have been cultured primarily in RPMI containing FCS and GM-CSF; however, there is mounting evidence that this may lead to the generation of DC with suboptimal properties for use in immunotherapy studies. In this study, we investigated the effects of IL-4 on the generation of immunostimulatory DC from mouse bone marrow in the presence or absence of immunogenic xenogenic serum antigens (FCS).

The results presented in our study demonstrate that although the culture of bone marrow progenitors in the presence or absence of FCS and IL-4 resulted in the production of cells displaying classic DC morphology and phenotypic markers, the proportion of CD11c+ DC within each culture was subject to considerable variation. Also, we show that, when cultured under serum-free conditions, CD11c+ DC cultured in the absence of IL-4 express suboptimal levels of MHC class II, CD40, CD80 and CD86 when compared with DC cultured in the presence of IL-4. The expression of these molecules is of course of fundamental importance in the presentation of antigen and the delivery of signals to T cells. Other groups have observed that, in the presence of FCS, DC cultured in the absence of IL-4 also express suboptimal levels of MHC class II, CD40, CD80 and CD86 when compared with DC cultured in the presence of IL-4 [7, 12–16, 18], although we did not find this to be the case in this study.

Although DC cultured under serum-free conditions in the presence of IL-4 were found to express higher levels of MHC class II, CD40, CD80 and CD86 than DC cultured under serum-free conditions in the absence of IL-4, which would suggest that the addition of IL-4 results in the development of more mature cells, the similar ability of these two DC populations to take up antigen suggests that both of these culture methods result in the production of predominantly immature cells.

IL-12 is known to be a potent Th1-inducing cytokine that induces the production of interferon-γ and the activation of natural killer (NK) and NK T cells [29, 30]. It has been suggested that mature DC are less able to secrete IL-12 in response to CD40 ligation or bacterial products than immature DC in humans [31, 32] and in the NOD mouse [33]. As DC cultured under serum-free conditions in the presence of IL-4 in this study appeared phenotypically more mature than DC cultured under serum-free conditions in the absence of IL-4, we investigated whether there was a difference in the ability of the two DC populations to secrete IL-12 in response to TNF-α, LPS or anti-CD40 MoAb. When unstimulated, both DC populations secreted a similar baseline level of IL-12. In addition, both DC populations secreted increased levels of IL-12 in response to LPS or anti-CD40 MoAb. However, only DC cultured in the absence of IL-4 showed a significant rise in the level of IL-12 secretion in response to TNF-α. Although the reason for this is not immediately clear, it has been shown elsewhere that IL-4 downregulates the expression of TNFR2 on DC [34], which may account for the lack of IL-12 secretion by DC cultured in the presence of IL-4 in response to TNF-α. However, because IL-4 does not appear to have an effect on the expression of TNFR1 [34], this explanation may prove to be over-simplistic. Taken together, the similarities in the abilities of these two DC populations both to take up antigen and to secrete IL-12 strongly suggest that the increased marker expression associated with the addition of IL-4 into the culture medium is not representative of full DC maturation.

Although we, in this study, did not explore the ability of DC cultured in FCS to take up antigen, previous studies have reported that DC that are generated in the presence of IL-4 are less efficient at taking up FITC–Dextran [7, 12, 15] or mannosylated BSA [13] via the mannose receptor than DC cultured in the absence of IL-4. It is possible that serum has an effect on the levels of mannose receptor expression, which would affect the uptake of antigens. In this study, we did not attempt to quantify the number of mannose receptors on each DC population and hence cannot discount the possibility that the absence of serum results in an up- or downregulation in the expression levels of mannose receptors on either DC population. However, because the ability of DC cultured under serum-free conditions in the presence of IL-4 to take up FITC–albumin via macropinocytosis (a separate mechanism of antigen uptake from the mannose receptor) is also comparable to that of DC cultured under serum-free conditions in the absence of IL-4, it seems doubtful that the level of mannose receptor expression alone is responsible for the similar degree of competency displayed in antigen uptake. Another conceivable explanation could be that DC cultured in the presence of IL-4 might differentiate into cells capable of taking up antigen at an earlier time-point than DC cultured in the absence of IL-4, so that when FCS is present in the culture medium these DC may have already taken up an abundance of xenogenic serum antigens before they have been analysed for their antigen uptake ability in the laboratory and display a reduced capacity to take up further antigen in these studies as a consequence. Although the X-VIVO-15 medium used in this study, which was initially designed for use with human cells and was used here as a serum-free medium in an attempt to produce more clinically relevant data, does contain some amount of protein (human albumin and human insulin), it may be that these are present in such small quantities so as not to elicit the same DC-antigen-saturation effect.

DC generated in X-VIVO-15 in the presence of IL-4 were stronger stimulators of naïve allogeneic splenocytes in a primary MLR than DC generated in the absence of IL-4, a finding that is in accordance with that of previously published studies [7, 13–16, 18] using bone marrow DC generated in RPMI in the presence of FCS. It is likely that the increased expression of MHC class I (data not shown), MHC class II and costimulatory molecules displayed by DC cultured in the presence of IL-4 is directly related to their increased allostimulatory capacity.

We compared the ability of the two DC populations to process and present OVA323−339 peptide and OVA protein to naïve OT-II transgenic T cells. DC cultured in the presence of IL-4 were stronger stimulators of naïve OT-II CD4+ T-cell proliferation when pulsed with either peptide or protein than DC generated in the absence of IL-4. This is probably a direct result of the increased levels of MHC and costimulatory molecules on these cells. These results are strikingly similar to those of two previously published studies that used DC generated in the presence of FCS and the OVA-specific T-T hybridoma D011.1 [15] or naïve D011.10 CD4+ transgenic T cells [7].

In summary, we demonstrate that bone marrow DC cultured with IL-4 are phenotypically more mature and more potent at stimulating allogeneic and antigen-specific T cells than DC cultured in the absence of IL-4, yet both DC populations contain predominantly functionally immature DC that show a similar capacity to take up antigen and release IL-12 in response to maturation stimuli. We show that potent, fully functional mouse bone marrow DC can be generated in X-VIVO-15 in the absence of FCS. DC cultured in this way would provide clinically useful data for comparison with human clinical trials using DC for immunotherapy.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. References