Dr M. J. Clare-Salzler, PO Box 100275, Department of Pathology, Immunology, Laboratory Medicine, University of Florida, 1600 SW Archer Road, Gainesville, FL 32610, USA. E-mail: firstname.lastname@example.org
It is highly desirable that immature dendritic cells (DC) used for tolerance induction maintain steady immature state with predominant interleukin (IL)-10 production. In this study, we attempted to develop DC with durable immaturity and other tolerogenic features by using dexamethasone (Dex). We found DC derived from human monocytes in the presence of 10−7m Dex were negative for CD1a. Compared with control transduced DC (Ctrl-DC), Dex-DC expressed lower CD40, CD80 and CD86 but equivalent human leucocyte antigen-DR. Both immature Dex- and Ctrl-DC did not express CD83. Nevertheless, upon stimulation of lipopolysaccharide (LPS) or CD40 ligand, the expression of CD40, CD80, CD83 and CD86 was upregulated on Ctrl-DC but not on Dex-DC. The immaturity of Dex-DC was durable following Dex removal. Interestingly, Dex-DC maintained production of large amount of IL-10 and little IL-12 five days after Dex removed. Further study indicated that high-level IL-10 production by Dex-DC was associated with high-level phosphorylation of extracellular signal-regulated kinase (ERK) as blockade of this enzyme markedly attenuated IL-10 production. Furthermore, Dex-DC sustained the capability of high phosphorylation of ERK and IL-10 production 5 days after Dex removal. In addition, Dex-DC had significantly lower activity in stimulating T-cell proliferation. Neutralization of IL-10, to some extent, promoted DC maturation activated by LPS, as well as T-cell stimulatory activity of Dex-DC. The above findings suggest that IL-10-producing Dex-DC with durable immaturity are potentially useful for induction of immune tolerance.
Dendritic cells (DC) are important antigen-presenting cells (APCs) and play an essential role in regulation of immune responses . Different DC subsets have different functions in priming T cells [2, 3]. The same subset of DC may undergo different functional states during their maturation process [4–6]. It is widely accepted that mature DC have potent capacity to stimulate T cells, while immature DC poorly stimulate T cells and thereby induce T-cell tolerance [7, 8]. Nevertheless, due to functional flexibility and plasticity [2, 3], DC's function may be modified under certain circumstances. Immature DC may mature during interaction with T cells by maturation signals provided through CD40 ligand (CD40L) on activated T cells [7, 8] and consequently activate T cells instead of inducing T-cell tolerance. Therefore, immature DC in a steady state may manifest tolerogenic function during the steady state in vivo[9, 10].
Because of DC's central role in immune responses, DC have been studied extensively in immunotherapy including cancer immunotherapy by boosting T-cell responses [11–14] and T-cell tolerance induction for allogeneic transplantation and autoimmune diseases [15–17]. Dhodapkar et al.  demonstrated that injection of immature DC induced antigen-specific T-cell tolerance by inducing regulatory T cells. However, the potential of immature DC to become mature DC is a concern when using these APCs for tolerance induction, particularly under the conditions of allogeneic transplantation and autoimmune diseases where immature DC may encounter ‘danger’ signals [19–21].
Characterization of mouse tolerogenic DC demonstrated that these cells are immature even with potent stimulation and predominantly produce interleukin (IL)-10 with low level of IL-12 . Another recent study showed that murine regulatory DC maintained an immature state with a potent capacity to induce regulatory T cells . Application of these regulatory DC in allogeneic bone marrow transplantation for leukaemic animals attenuated graft-versus-host disease . These studies provide a model for preparing steady immature DC that predominantly produce IL-10 for the induction of transplantation tolerance and prevention of autoimmune diseases.
Previous studies demonstrated that anti-inflammatory glucocorticoids are promising agents for developing tolerogenic DC [25–30]. In this study, we attempted to develop human tolerogenic DC derived from human peripheral monocytes by using dexamethasone (Dex), and to determine whether Dex-induced (Dex)-DC possess durable immaturity. We found that Dex drove differentiation of monocytes into immature DC with resistance to the maturation stimuli, such as, lipopolysaccharide (LPS) or CD40L. Interestingly, Dex-DC produced high level of IL-10 and undetectable level of IL-12. It is of great interest that these DC sustain these features 5 days after removal of Dex from the culture. Our findings suggest Dex-DC are potentially to be used for tolerance induction in autoimmune diseases and allogeneic transplantation.
Materials and methods
Media and reagents. RPMI-1640 (Fisher Healthcare, Houston, TX, USA) was supplemented with 2 mm l-glutamine, 100 U/ml penicillin, 100 µg/ml streptomycin and 10 mm HEPES. Fetal calf serum was purchased from Invitrogen Life Technologies (Grand Island, NY, USA). Granulocyte-macrophage colony-stimulating factor (GM-CSF) and CD40L were obtained from Immunnex (Seattle, WA, USA). IL-4, interferon (IFN)-γ were purchased from PeproTech (Rocky Hill, NJ, USA). Dex, FITC-dextran and LPS were purchased from Sigma (St. Louis, MO, USA). Fluorescence-conjugated antibodies were purchased from BD PharMingen (San Diego, CA, USA). CD14-microbeads and CD45RA-microbeads were purchased from Miltenyi Biotech (Auburn, CA, USA). CD4+ T-cell negative isolation kits were obtained from Stem Cell Technology (Vancouver, Canada). Beadlyte human IL-10 and IL-12p70 kits were purchased from Upstate (Charlottesville, VA, USA). Anti-phosphorylated extracellular signal-regulated kinase (ERK) antibodies were purchased from CellSignal Inc. (Beverly, MA, USA).
Dendritic cell culture. Monocytes were isolated as previously described [31, 32]. Monocytes (5 × 105) were cultured in 12-well plate in 2 ml RPMI-1640 containing 10% FCS, 1000 U/ml GM-CSF and 20 ng/ml IL-4 in the absence or presence of indicated concentrations of Dex for 6 days. On day 3, half media were changed with fresh media containing indicated components. For some cultures, DC were washed thoroughly to get rid of Dex, stimulated with 100 ng/ml LPS or 1 μg/ml CD40L for 24 h to determine DC maturation.
For some experiments, Dex-DC and control transduced (Ctrl)-DC were generated using the method described above. Dex-DC were washed to remove Dex. Both Dex-DC and Ctrl-DC were cultured in the presence of GM-CSF and IL-4 for various length of time to ascertain Dex-DC immaturity and cytokine production after removal of this glycocorticoid.
In some experiments, Dex-DC were stimulated with LPS for 24 h with or without anti–IL-10 antibody (10 µg/ml), which could completely neutralize the IL-10 produced in culture. The influence of anti–IL-10 on DC maturation was determined by examining the expression of CD83 and CD86.
To assess the ability of DC in production of IL-10 and IL-12p70, DC (2.5 × 104) were incubated in 200 µl media with stimulation of 100 ng/ml LPS or LPS plus 50 ng/ml IFN-γ for 24 h. In some experiments, Dex-DC were stimulated with LPS with different doses of ERK inhibitor (PD98059) for 24 h. The supernatant in each incubation was examined for IL-10 and/or IL-12p70 production by Beadlyte human IL-10 and IL-12p70 beads kit using Luminex IS2.1 (Austin, TX, USA).
T-cell isolation. Peripheral mononuclear cells (PBMC) were isolated by Ficoll-hypaque gradient centrifugation. CD4+ T cells were purified by CD4+ T-cell negative isolation kit according to the protocol provided by the manufacturer (Stem Cell Technology). Briefly, PBMCs (5 × 107) in 1 ml of PBS were incubated with 100 µl of CD4+ T-cell enrichment cocktail antibodies in the presence of 1% rat serum at 4 °C for 15 min. Thereafter, 60 µl of magnetic colloid was added to each millilitre of PBMC for another 15 min of incubation. The cells were loaded onto a Stem-SepTM column placed in a magnet. CD4+ T cells were eluted by washing the column with 15 ml of PBS-1% FCS. The purity for CD4+ T cells was always around 95%.
FITC-dextran endocytosis by DC. To study the endocytosis activity of DC, a previously reported method [31, 32] was used with slight modifications. DC (5 × 104) were resuspended in 100 µl PBS-1% FCS and incubated with FITC-dextran (0.1 mg/ml) at 37 and 0 °C for 30 min. The incubations were stopped by adding 2-ml cold PBS-1% FCS. The cell were washed three times with cold PBS-1% FCS and analysed on a FACScan (Becton Dickson, San Diego, CA, USA).
Flow cytometry. For immunophenotypic characterization of DC, washed DC were incubated with indicated antibodies labelled with fluorescent dyes and diluted in PBS containing 1% BSA and 0.02% NaN3 at 4 °C for 25 min. Isotype-matched antibodies were used as negative controls. The cells were analysed by a FACScan with 10,000 events were collected for each sample, and the data were analysed by FCS express software (De Novo Software, Ontario, Canada).
Mixed leucocyte reaction (MLR). Different cell concentrations of DC were cultured with 1 × 105 allogeneic CD4+ T cells in each well of a U-bottom 96-well plate in 200 µl RPMI1640 media containing 10% FCS for 6 days. [3H]-thymidine (1 µCi) was added into each well for the last 16 h. The cells were harvested using a PHD cell harvester. Thymidine incorporation was determined by scintillation counting. All experiments were performed in triplicate incubations. In some experiments, anti–IL-10 (10 µg/ml) was added to the MLR as indicated.
Western immunoblotting. DC (2.5 × 105) were stimulated with LPS (100 ng/ml) for indicated durations at 37 °C. DC without any treatment were used as the control. After stimulation, DC were washed twice with cold PBS, solubilized in 100 µl of SDS-PAGE sample buffer and incubated at 95 °C for 3 min. Twenty microlitres of each sample (equivalent to 50,000 cells) were analysed by SDS-10% PAGE. The proteins in gels were transferred on to nitrocellulose membranes. The membranes were blocked with nonfat milk and probed separately with 1:1500 diluted primary antibody against phosphorylated ERK. Thereafter, the membrane was incubated with 1:1000 diluted secondary antibody conjugated to horseradish peroxidase. Chemiluminescence Reagent Plus (Amersham Bioscience, Piscataway, NJ, USA) was used for detection. Chemiluminescence was assayed with a Kodak Image Station (Kodak, Rochester, NY, USA).
Dex induces differentiation of monocytes into CD1a– DC in dose-dependent manner
In determining the optimal dose of Dex used for the present study, we found that Dex with increasing concentrations progressively suppressed CD1a+ DC development (Fig. 1). At 10−7-m concentration, more than 95% of DC were CD1a– cells. The various concentrations of Dex tested in these experiments did not induce cell death determined by Trypan Blue staining. Therefore, 10−7m Dex was chosen for the following studies.
Immunophenotype of Dex-DC
The phenotypic characteristics of Dex-DC developed in the presence or absence of Dex were determined by FACS. As shown in Fig. 2(A), both Dex-DC and Ctrl-DC were CD83 negative. Slight expression of CD14 was found on Dex-DC but not on Ctrl-DC. The expression of CD40, CD80 and CD86 was lower on Dex-DC than on Ctrl-DC. Interestingly, human leukocyte antigen (HLA)-DR expression on Dex-DC was higher than that on Ctrl-DC (Fig. 2A), suggesting that Dex-DC could serve as APC.
Dex-DC sustain immature state in response to maturation stimuli
It is expected that DC to be used for tolerance induction maintain immature state. To determine the applicability of Dex-DC for tolerance induction, we surprisingly found that Dex-DC were refractory to maturation induction by LPS or CD40L after removal of Dex (Fig. 2B). More interestingly, Dex-DC maintain this immaturity for a prolonged period of time after Dex removal (Fig. 3), suggesting that Dex-DC may maintain immature state in vivo after injection.
FITC-dextran endocytosis by DC
To determine the capacity of Dex-DC in capturing exogenous antigens, we examined the ability of Dex-DC in uptake of FITC-dextran. Similar to the previously reported results , we found that Dex-DC had higher capacity than Ctrl-DC in uptake of FITC-dextran. In contrast to Ctrl-DC, after LPS activation, the uptake of FITC-dextran by Dex-DC was only slightly decreased (Fig. 4), which could be associated with Dex-DC immaturity (Figs 2B and 3).
Dex-DC produce high levels of IL-10 and low levels of IL-12 and maintain this patterns of cytokine production after removal of Dex. Previous studies demonstrated that LPS was potent in stimulating DC for IL-10 production, while LPS plus IFN-γ was a potent stimulus for IL-12 [31, 34]. Therefore, we applied LPS or LPS plus IFN-γ as stimuli to investigate IL-10 and IL-12 production by DC. We found that upon LPS stimulation, Dex-DC produced high amount of IL-10 (Fig. 5A) with little IL-12p70. Even if LPS plus IFN-γ was used as a stimulus, IL-12p70 production by Dex-DC remained undetectable (Fig. 5A). However, Ctrl-DC produced high level of IL-12p70 upon stimulation of LPS and IFN-γ (Fig. 5A).
It is of great interest to learn whether Dex-DC maintain IL-10-producing feature after removal of Dex for a period of time. To this end, we cultured DC in the presence or absence of Dex for 5 days. DC removed from Dex were cultured in the media containing GM-CSF and IL-4 for 1, 3 and 5 days. The capacity of DC in production of IL-10 and IL-12 was examined by stimulating DC with LPS or LPS plus IFN-γ as described above. The results shown in Fig. 5(B) demonstrated that Dex-DC maintained IL-10-producing capacity after Dex removal. However, IL-12p70 production by Dex-DC was not recovered even if Dex was removed. These findings suggest that after development in the presence of Dex, DC IL-10 production becomes independent on this glycocorticoid.
High level of phosphorylation of ERK mitogen-activated protein kinases (MAPK) contributes to IL-10 production by Dex-DC
Earlier reports showed that IL-10 production by DC was associated with phosphorylation of ERK MAPK [34–36]. To determine whether production of IL-10 by Dex-DC is related to the high-level phosphorylation of ERK MAPK, we studied the phosphorylation of this kinase in LPS-activated DC. We found that phosphorylation of ERK was higher in Dex-DC than in Ctrl-DC at early time points (Fig. 6A). It was noted that Dex-DC maintained the enhanced potential for ERK phosphorylation when cultured for 5 days in the absence of Dex (Fig. 6B). To further determine whether ERK phosphorylation in Dex-DC contributes to high production of IL-10, we examined the effect of ERK blockade on LPS-stimulated IL-10 production by Dex-DC. Indeed, ERK blockade significantly reduced IL-10 production (Fig. 6C).
Neutralization of IL-10 partially reverses immaturity of Dex-DC
To determine whether IL-10 produced by Dex-DC suppresses Dex-DC maturation, we stimulated Dex-DC with LPS with or without anti – IL-10 antibody. Maturation markers, CD83 and CD86, were examined. The results shown in Fig. 7 indicate that neutralization of IL-10 allows some Dex-DC maturation induced by LPS. Nevertheless, the expression level of CD83 and CD86 was still significantly lower than that on mature Ctrl-DC (Fig. 7).
Stimulatory activity of DC in MLR
When Dex-DC and Ctrl-DC were studied for their capacity to stimulate proliferation of allogeneic CD4+ T cells, we found that Dex-DC had a significantly lower capacity than the Ctrl-DC in stimulating allogeneic CD4+ T-cell proliferation (Fig. 8A). We also found CD4+ T-cell proliferation stimulated by Dex-iDC was only partially recovered by neutralization of IL-10 (Fig. 8B).
Immunosuppressants, such as corticosteroids, cyclosporin, rapamycin and FK506 have been demonstrated to be effective in prevention of allogeneic transplant rejection and attenuation of autoimmune diseases [37–40]. However, nonspecific immunosuppression of these drugs renders the immune system compromised, thereby leading to an increased risk for tumours and infectious diseases . Therefore, it is the goal to induce antigen-specific tolerance in allogeneic transplantation and autoimmune diseases without impairing other important immune functions of the immune system. Application of tolerogenic DC would fulfil this task. A line of evidence has demonstrated the effect of immunosuppressants on DC [31–36, 41–43]. The application of immunosuppressants to modulate DC for tolerance induction has been postulated [40–42]. Thus, immunosuppressants could be used to develop tolerogenic DC. In addition, it has been demonstrated that vitamin D3 and antiestrogens can similarly modify the immunophenotypes and functions of monocyte-derived DC, which suggests other drugs can be employed to modify DC for treatment of autoimmune disorders [44, 45].
In evaluating immunosuppressants, Dex, rapamycin and FK506, commonly used in clinical practice for developing tolerogenic DC, in line with the previous studies [25, 26], we found that Dex induced development of CD1a– DC derived from human peripheral monocytes (Fig. 1). The suppression of CD1a expression on DC by Dex was dose-dependent (Fig. 1). Further phenotypic characterization demonstrated that Dex-DC expressed low levels of costimulatory molecules such as CD80, CD86 and CD40, but high level of HLA-DR (Fig. 2), suggesting that Dex-DC may induce T-cell tolerance during interaction with T cells by providing sufficient ‘signal 1’ elicited by high-level MHC-II-presented antigens and poor ‘signal 2’ by low level of costimulatory molecules [46, 47].
As mentioned earlier, immature DC have the potential to become mature under certain circumstances [7, 8]. For tolerance induction, it is perhaps desirable that DC maintain an immature state in the setting of maturation factors. Although it has been shown that Dex-DC express low level of costimulatory molecules such as CD40, CD80 and CD86 [26, 27], whether Dex-DC maintain their immaturity after removal of Dex for a period of time has not been addressed. Therefore, we evaluated the resistance of DC to LPS or CD40L stimulation. The results demonstrated that LPS or CD40L stimulation of Dex-DC, unlike Ctrl-DC, limits the upregulation of CD83, CD40, CD80 and CD86 on Dex-DC (Fig. 2). These results suggest that Dex-DC are resistant to maturation. However, concerns would be whether Dex-DC that are injected into a host maintain immaturity once removed from Dex. To address this issue, we tested the durability of immature state of Dex-DC several days after removing Dex from the culture. At different time points, the DC were stimulated with or without LPS, the expression of CD83 and CD86 was determined by flow cytometry. The results demonstrated that Dex-DC maintained an immature state for a prolonged period (Fig. 3). Further characterization suggested that high level of IL-10 production by LPS-stimulated Dex-DC partially contributed to maturation resistance of these cells (Fig. 7).
It has been established that cytokines secreted by DC play an important role in directing T-cell responses [48–50]. Kinetic study of IL-10 and IL-12 production by DC clearly showed that predominant IL-12-producing stage DC primed Th1 response, while IL-10-producing stage DC primed Th2 response . In vivo study also demonstrated distinct T-cell responses elicited by IL-10- and IL-12-producing DC . DC located in pulmonary tissue and intestines with high production of IL-10 have been suggested to be important for maintaining unresponsiveness of local T cells to commonly encountered antigens [48, 49]. Recently, IL-10-producing DC have been described to be able to induce regulatory T cells [22, 23]. It is therefore of interest to investigate IL-10 and IL-12 production by Dex-DC. To our surprise, Dex-DC produce large amount of IL-10 and almost undetectable IL-12 upon LPS stimulation (Fig. 5A). Even LPS plus IFN-γ, which is strong stimulus for IL-12, did not illicit production of this cytokine in Dex-DC (Fig. 5A). These findings suggest Dex-DC have a tolerogenic potential by secreting high level of IL-10, which may subsequently facilitate regulatory T-cell development [33, 22, 23, 53].
Of great interest is the potential of secreting high-level IL-10 maintained by Dex-DC after being removed from Dex for a prolonged period up to 5 days (Fig. 5B). Besides, the capacity of IL-12 production by Dex-DC did not recover (Fig. 5B). However, IL-12-producing capacity was maintained by Ctrl-DC, although reduced to some extent with time (Fig. 5B). These findings suggest that the cytokine-secreting profile by Dex-DC does not depend on the presence of Dex and applicability of Dex-DC for tolerance induction.
Our previous study, as of others, demonstrated that IL-10 production by activated DC was associated with phosphorylation of ERK MAPK [34–36]. Therefore, it is of interest to determine whether high-level IL-10 production by Dex-DC involves heightened phosphorylation of ERK. Indeed, LPS-stimulated Dex-DC displayed significantly higher level of phosphorylation of ERK (Fig. 6A). Furthermore, Dex-DC maintained this potential after being removed from Dex (Fig. 6B). Interestingly, Dex-DC sustained high level of ERK phosphorylation for a longer period at 5 days after Dex-removal (Fig. 6B) (60 min versus 20 min at day 1). These findings suggest that high level of ERK phosphorylation in LPS-activated Dex-DC might be correlated with high production of IL-10 and may play an important role in maintenance of high-level IL-10 production after removal of Dex. Our further studies demonstrating that ERK blockade markedly inhibited IL-10 production by LPS-stimulated Dex-DC (Fig. 6C) strongly support that high level of ERK phosphorylation in activated Dex-DC contributes to the high-level IL-10 production.
Immature DC possess a unique capacity to capture exogenous antigen . With regard to Dex-DC's antigen-capturing ability, we found that immature Dex-DC had higher antigen acquisition capacity compared with Ctrl-DC in terms of higher uptake of FITC-dextran (Fig. 4). After LPS activation, endocytosis of FITC-dextran by Ctrl-DC significantly reduced (Fig. 4). However, FITC-dextran endocytosis by LPS-activated Dex-DC only slightly decreased, which is consistent with maturation-resistance of Dex-DC. High-level phosphorylation of ERK (Fig. 6) could have contributed to the high endocytosis of FITC-dextran by Dex-DC (55). The above findings suggest that Dex-DC could be efficiently pulsed with specific antigens for inducing antigen-specific immune tolerance in immunotherapies for autoimmune diseases and allogeneic transplantation.
Another benefit of applying Dex-DC to induce immune tolerance is the weak T-cell stimulatory activity (Fig. 8A). IL-10 production by Dex-DC partially contributed Dex-DC' poor T-cell stimulatory function (Fig. 8B). Immature DC with low T-cell stimulatory activity appear to induce tolerance . Other features of Dex-DC, such as CD1a– phenotype (31), low level of surface costimulatory molecules, as well as durable immaturity may also have contributed to the lower capacity of Dex-DC in stimulation of T cells. It has been demonstrated that Dex-treated mouse DC can induce Tr1 cells with predominant IL-10 production . It is of interest to know whether human monocyte-derived Dex-DC can induce regulatory T cells. The studies of whether Dex-DC can promote regulatory T cells differentiation thereby suppress T-cell responses in vitro and in vivo are being conducted.
In conclusion, IL-10-producing Dex-DC with durable immaturity have a potential to induce immune tolerance. Dex-DC could be applied for preventing allograft rejection and autoimmune diseases by inducing antigen-specific tolerance.
This work was supported by NIH grant R21 DK063244 to MCS and American Diabetes Association Junior Faculty Award to CQX.