Published online before print October 21, 2005
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
RIIB1 and -B2 receptors in human dendritic cells
Brest Medical School, Cellular Therapy Laboratory, France
1 Correspondence: Brest Medical School, Cellular Therapy Laboratory, 22, avenue Camille Desmoulins, 29200 Brest, France. E-mail: nguriec{at}voila.fr
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
RII and elicit antigen presentation and protective antitumoral immune response in mice. Two protocols are commonly used to differentiate human monocyte-derived DC in vitro. They associate granulocyte macrophage-colony stimulating factor (CM-CSF) with interleukin (IL)-4 or IL-13. In this study, we first assessed the ability of the two types of DC to initiate an immune response against an IC-linked antigen. We evidenced that IL-4 and IL-13 DC display comparable lymphocyte stimulatory capacity and similar lifetimes. We next characterized FcRIIs expressed by pure populations of circulating myeloid DC (BDCA1+DC), IL-4, and IL-13 DC. We highlighted the expression of FcRIIA, -B1, and -B2 by pure populations of BDCA1 myeloid DCs and IL-4 and IL-13 DC. Moreover, IL-4 and IL-13 DC displayed greater FcRIIB expression than monocytes but a comparable FcRIIA. We next investigated the FcRIIB mechanism of action. We evidenced that deleting FcRIIB increased the ability of IC-pulsed DC to stimulate autologous lymphocytes. FcRIIB acted by lowering IC uptake, surface expression of costimulation molecules, and cytokine release. Finally, the balance between activating FcRIIA/inhibitory FcRIIB (B1+B2) could be modulated in vitro by inflammation mediators. By lowering FcRIIB expression without significantly affecting FcRIIA, prostaglandin E2 (PGE-2) appeared to be a major regulator of this balance. IL-1ß and tumor necrosis factor 
were also found to potentiate PGE-2 action. Altogether, our results evidence an inhibitory role for FcRIIB in human DC and provide an easy way to possibly improve in vitro the induction of immune response against IC-linked antigen.
Key Words: phagocytes • immunotherapy • inflammation • immunoglobulin • antigen presentation
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
A single class of FcR to immunoglobulin G (IgG); FcRII (CD32) is expressed by monocyte-derived DCs [4
, 5
]. IC capture by FcRII allows DC maturation and efficient presentation to naive CD8+ T cells of an exogenously derived antigen (e.g., IC-linked antigen [3
4
5
6
7
]). Furthermore, apoptotic tumor cells opsonized with antibodies have been shown to elicit DC maturation, generation of tumor-specific CD8+ T cells, and protective tumor immunity in vitro and in vivo [8
9
10
]. DC-based immunotherapies using IC-pulsed DC have, therefore, emerged as an attractive mean to stimulate protective antitumor immunity.
However, the amplitude of immune response is believed to depend on the ratio between activating and inhibitory FcR. In humans, FcRIIs exist as two major isoforms, FcRIIA and -B, which carry out divergent functions. FcRIIA contains an immunoreceptor tyrosine-based activating motif (ITAM) in its cytoplasmic tail, which mediates positive signaling. Activation of FcRIIA results in IC internalization as well as in initiation of immune response. By contrast, an immunoreceptor tyrosine-based inhibitory motif (ITIM) has been found in the FcRIIB cytoplasmic region. Engagement of FcRIIs by IC then leads to cellular activation or inhibition signals, depending on the engaged receptor. Two human FcRIIB inhibitory receptors are encoded by the FcRIIB gene in humans through an alternative splicing mechanism. FcRIIB1 and -B2 only differ by the lack of 19 amino acids in the intracellular domain of FcRIIB2 [11
]. Both are single-chain receptors with an ITIM domain and negatively regulate B cell receptor (BCR)-, TCR-, and FcR-dependent activation [12
13
14
]. Recent experiments have demonstrated a better efficacy of tumor eradication with humanized antibodies against tumors in FcRIIB knockout mice than in their wild-type counterparts [15
]. FcRIIB engagement on human monocytes is associated with the down-regulation of phagocytosis and decreases antigen-presenting capacities [16
, 17
]. FcRIIB also prevents spontaneous human DC maturation [10
]. However, no data are available about the expression of FcRIIB isoforms on in vitro-differentiated DC and their regulation. Elucidation of FcRII function and regulation in human DC thus appeared to us as a prerequisite to the development of tumor-specific vaccines. We studied the ability of monocyte-derived DC to induce lymphocyte proliferation depending on the DC differentiation protocol and examined FcRII expression, function, and regulation.
We found that the replacement of interleukin (IL)-13 by IL-4 in DC differentiation protocol allowed the generation of DC with similar lifetimes and comparable abilities to trigger an immune response when pulsed with IC. We observed distinct expression patterns for FcRIIs in blood myeloid DC, monocyte-derived DC. Yet, FcRIIA, -B1, and -B2 receptors were always expressed. Our study also provides clues of the mechanism of action of FcRIIB as well as regulations of this receptor by pro- and anti-inflammatory mediators.
Altogether, our data constitute the grounding for the development of new strategies based on the use of IC in immunotherapy protocols and for testing novel strategies aimed at evaluating the effective link between the humoral immunity and the cellular one.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Apoptosis quantitation by flow cytometry
Apoptosis was evidenced by flow cytometry analysis of annexin V and propidium iodide staining. The cells (5x105) were stained according to the manufacturer’s instructions (Beckman Coulter).
Immunoprecipitation, blotting, and reverse transcriptase-polymerase chain reaction (RT-PCR) analysis
Pure DC populations were suspended in lysis buffer (10 mM Tris, pH 7.4, 1 mM EDTA, 1 mM EGTA, 50 mM NaCl, 1 mM Na3VO4) for 10 min on ice and washed, and postnuclear extracts were solubilized further in immunoprecipitation buffer (10 mM Tris, pH 7.4, 1 mM EDTA, 1 mM EGTA, 150 mM NaCl, 2 mM Na3VO4, 100 mM NaF, 4 mM NaH2PO4, 1 µM phenylmethylsulfonyl fluoride, 1% Triton, 0.5% Nonidet P-40, 5 mg/ml leupeptin, 5 mg/ml aprotinin). The purity of membrane protein purification was controlled by caspase-3 Western blotting, as caspase-3 is known to be mostly cytosolic (Upstate Biotechnology, Lake Placid, NY, n=6, data not shown). Equal amounts of membrane proteins were immunoprecipitated by using anti-FcRII KB61 and AT10 antibodies (generous gift from Dr. Karen Pulford, John Radcliffe Hospital, Oxford, UK, and Abcyss SA, France, respectively), separated on 10% polyacrylamide gels, transferred onto polyvinylidene difluoride membranes (Hybond P, Amersham Biosciences, France), probed with anti-FcRII 2E1 antibody (Beckman Coulter), and developed by enhanced chemiluminescence [17
]; its signal was quantified with densitometry software (Bioprofil, Vilbert Lourmat, France). DC conditioned media were also tenfold concentrated using centricon (molecular weight cut-off=3000, Amicon, France) at day 6 for imDC and day 9 for mDC. Positive controls for Western blotting and immunoprecipitation experiments were the Raji cell line for FcRIIB expression, U937 cell line for FcRIIA, and K562 for soluble FcRIIA (sFcRIIA) [19
, 20
]. MRC5 cell line was used as a negative control.
Total RNA extraction from sorted DC or cell lines and RT were performed as described previously. Primers for FcRIIA, FcRIIB, and ß-actin and PCR conditions have been described previously [17
, 21
]. Positive and negative controls were the same as for immunoprecipitation experiments. PCR products were run on a 10% polyacrylamide gel. Equal efficiency of RT was checked by analysis of ß-actin expression, and results of FcRIIA and -B expression were normalized to ß-actin.
Transfection
A FcRIIB antisense and a sense construct were generated by RT-PCR using cDNA prepared from the Raji cell line as template. A 200-base pair fragment encompassing the second extracytoplasmic domain up to the first intracytoplasmic one and common to FcRIIB1 and -B2 was obtained by enzymatic digestion of the PCR product (BamHI+PstI). Then, the fragment was inserted into the PMG expression vector (Tebu, France). Constructs were under control of a cytomegalovirus promoter (encephalomyocarditis virus-internal ribosomal entry site). Plasmid DNA was prepared by alkaline lysis followed with Triton-X114 purification to remove LPS, and then endotoxins levels were checked as described previously [22
]. Plasmid DNA was dissolved in HEPES saline buffer (150 mM NaCl, 20 mM HEPES, pH 7.4). As mannose receptor is expressed by human monocytes after a 48-h incubation with GM-CSF + IL-4/IL-13 (data not shown [23
]), sense and antisense constructs were transfected at day 2 using mannose polyethylenimine conjugates (manPEI). Mannose was linked to commercially available PEI (Exgen 500, Euromedex, France) via a phenylisothiocyanate bridge made by using mannopyranosyphenyl isothiocyanate as a coupling reagent as described [23
]. Coupling efficiency and purification were controlled as reported [23
]. manPEI was suspended in HEPES saline buffer, and a 1/1 ratio between plasmid DNA and manPEI was used for transfection [23
]. Six-well plates were gently spinned before addition of DNA/manPEI complexes in complete medium. Transfection medium was replaced with complete medium after 4 h of incubation. Hygromycin selection of transfectants was started 24 h after transfection, and a Ficoll separation was performed at day 4 to remove dead cells, healthy were then incubated in complete medium with hygromycin for 2 additional days. At day 6, nonhealthy cells were removed by Ficoll separation. Viability was then checked by trypan blue exclusion and cytometry analysis of Annexin V and propidium iodide. As expected [23
], antibiotic selection gave rise to 15–20% resistant cells with more than 95% healthy cells. mRNA levels of FcRIIA, -B1, and -B2 were determined by specific RT-PCR. FcRIIA, -B1, and -B2 protein expression levels by nonadherent transfectants at day 6 were checked by immunoprecipitation, and maturation status was controlled by flow cytometry analysis of CD40, -80, -83, and -86 staining.
MLR assays
Autologous MLR were performed to evaluate antigen-presenting capacity of human DC for a soluble or an IC-linked antigen [24
]. U-bottom 96-well plates were coated with tetanus toxin IC (tetanus toxin, Roche Diagnostics, France; human polyclonal IgG against tetanus toxin, EFS, France), obtained as described previously (ratio 1/20, 10–8 M tetanus toxin [4
, 22
]). MLR assays were performed in triplicate by cocultivation of day 6 DC at various dilutions with lymphocytes (range 1/10–1/100) in X-vivo 15 medium (Bio-Whittakker, France). Proliferation was assessed by using tritiated thymidine and a ß counter (Tri-Carb, Packard, Downers Grove, IL).
Quantitation of IC uptake
Nonadherent, day 6 transfectants were washed twice in RPMI and incubated for 24 h in complete medium in plates coated overnight with tetanus toxin IC. Cells were then fixed with PBS with 3% paraformaldehyde and washed twice in PBS with 1% BSA and 5 mM EDTA to separate surface-bound IgG [5
]. They were further permeabilized with PBS containing 1% BSA and 0.3% saponin before staining with fluorescein isothiocyanate-conjugated goat anti-human antibody (Beckman Coulter [9
]). IC uptake from independent experiments was analyzed by flow cytometry on CD11c-positive cells [9
].
IL-10, p70 IL-12, and interferon- (IFN-) release by DC after FcR-mediated activation
Cytokine concentrations in frozen DC culture supernatants from various experiments were measured in duplicate by cytokine-specific sandwich enzyme-linked immunosorbent assay (ELISA) kits (R&D Systems).
Modulation of FcRII expression assays
Cytokines were purchased by R&D Systems and prostaglandin E2 (PGE-2) and polyIC (synthetic analog of double-strand RNA), by Sigma Chemical Co. Oligodeoxynucleotide (ODN) 2006 and ODN 2216 were synthesized and purified by Proligos (France). As described previously, the concentrations used ranged from 10 to 500 U/ml for IFN- and tumor necrosis factor (TNF-), 0.2 to 20 ng/ml IL-1ß, 1 to 100 ng/ml vascular endothelial growth factor (vEGF) and transforming growth factor-ß1 (TGF-ß1), 20 to 200 ng/ml IL-10, 10 to 500 ng/ml IL-6, 0.1 to 10 ng/ml PGE-2, 5 to 100 µg/ml polyIC, and 0.5 to 10 µg/ml ODN [25
26
27
28
29
30
31
32
33
]. To avoid any cytokine interference, DC were washed three times with RPMI medium before incubation for 24 h at 1 x 106 cells/ml in a medium containing RPMI, L-glutamine, 1% IgG-depleted autologous plasma, and mediators but neither GM-CSF nor IL-4/13 (such a deprivation affects neither FcRII expression levels nor patterns, n=6, data not shown). Finally, the cells were sorted on CD1a or CD83 according to their differentiation stage, and their purity was checked by flow cytometry.
Statistical analysis
Data are expressed as mean ± SEM. Nonparametric statistical methods (Wilcoxon test) and two-tailed P values were used for comparison.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
R expression other than FcRII (n=8, Fig. 1
). FcRI and FcRIII were not expressed by IL-4 imDC and mDC. IL-13 allowed DC to differentiate with an identical FcR expression profile. Moreover, FcRII expression was lower on mDC than on imDC. One should note that this difference was even more pronounced when maturation was induced by IC instead of LPS (Fig. 1)
.
 View larger version (16K): [in a new window] |
Figure 1. DC characterization and lymphocyte stimulatory capacity. Phenotypic follow-up of DC differentiation (A) with GM-CSF and IL-13 at day 6 (imDC) and day 9 (mDC). DC were pulsed with LPS to allow maturation (n=8). Similar staining was observed for IL-4 DC (n=8, data not shown). (B) FcR membrane expression by DC. First two plots: FcRIII and FcRI expression on imIL-13 DC at day 6 (n=8). Similar results were observed for IL-4 imDC and mDC (n=8, data not shown). Last two plots: comparative analysis of FcRII staining with 2E1 antibody for DC from the same donor according to the differentiation stage and the maturation protocol (n=8). (C) Lymphocyte proliferation as measured in autologous MLR when using IL-4 and IL-13 DC from the same donor pulsed with a soluble antigen (sol Ag) or with IC. Data presented are the means ± SEM of eight independent experiments. MLR were also performed in parallel without antigen (w/o Ag) or with antitetanus toxin IgG (IgG) as negative controls. dpm, Disintegrations per minute.
|
We also investigated DC lifetime by measuring the percentage of apoptotic DC at days 10, 12, and 14. IL-4 and IL-13 DC from the same donor were assessed by flow cytometry analysis of Annexin V-positive cells (n=8). No change in lifetime was noticed further to the use of IL-13 instead of IL-4 during the differentiation process (Fig. 2 ). Additionally, the use of LPS or IC as maturation agent resulted in similar lifetimes (Fig. 2) .
 View larger version (39K): [in a new window] |
Figure 2. Comparative analysis of IL-4 and IL-13 DC apoptosis. day 6 nonadherent cells, differentiated using GM-CSF with IL-4 or IL-13, were pulsed with IC or LPS. Values represent the means ± SEM of eight independent experiments.
|
RIIA, -B1, and -B2 receptors but not sFcRIIARII isoforms expressed by human DC, BDCA-1+ myeloid blood DC, CD1a+ imDC, and CD83+ mDC were sorted out and purity-checked by flow cytometry. As shown in Figure 3
after sorting, purity reached 98% CD1a+ cells for imDC and CD83+ cells for mDC. Immunophenotyping for costimulation and presentation molecules also confirmed that these cells were imDC and mDC, respectively (Fig. 3A
and 3B)
.
 View larger version (51K): [in a new window] |
Figure 3. Immunophenotype of pure DC populations. Cells were sorted using a positive selection for CD1a at day 6 to purify imDC (A) and a positive selection for CD83 at day 9 to sort mDC (B). Flow cytometry analysis was performed for eight donors. FcRII expression was analyzed using the 2E1 antibody. Results presented here are for IL-13 DC. Similar results were observed for IL-4 DC (n=8, data not shown).
|
RIIA, -B1, and -B2 were then performed on pure populations (n=8, Fig. 4
). These experiments showed the expression of FcRIIA, -B1, and -B2 receptors by human monocytes, BDCA-1+ blood DC, and DC. Moreover, DC differentiation with IL-4 or IL-13 resulted in a significant increase of FcRIIB1 and -B2 expression (P
0.05), whereas minor changes were observed for FcRIIA. Consistently, with flow cytometry analysis (Fig. 1)
, we found that FcRIIB expression on IL-13 DC was slightly higher than on IL-4 (P=0.05), whereas FcRIIA expression was similar (Fig. 4)
.
 View larger version (30K): [in a new window] |
Figure 4. FcRIIs expressed by human myeloid DC. RT-PCR and immunoprecipitation (IP) experiments (A) were performed on pure DC populations from the same donor (n=8). KB61 and AT10 antibodies were used to immunoprecipitate FcRIIs from postnuclear extracts and 2E1 antibody for Western blotting revelation. FcRIIA, -B1, and -B2 expressions are the results of the same exposure and were measured by densitometry (graphs). (B) Conditioned media of imDC at day 6 and mDC at day 9 were examined for sFcRIIA expression (n=12). Immunoprecipitations were performed for analysis of membrane expression of FcRIIs. Monocyte and BDCA-1+ were incubated for 24 h in RPMI with 1% IgG-depleted, autologous plasma after purification, and the conditioned media were collected.
|
RIIA, resulting from an alternative splicing of the FcRIIA gene. sFcRIIA acts as an inhibitory receptor by competing with membrane FcRIIs for IC binding and severely reduces Langerhans cell antigen-presenting capacity [19
, 20
]. We carried out immunoprecipitation experiments with DC conditioned media (normal and tenfold-concentrated) and cellular extracts from BDCA-1+ blood DC, sorted imDC, and mDC (n=12). The human K562 cell line and its conditioned medium were used as positive controls [18
, 19
]. None of these experiments allowed us to detect the sFcRIIA protein (Fig. 4)
.
FcRIIB deletion increases IC-pulsed–DC lymphocyte-stimulating capacity
To further investigate the role of FcRIIB in DC presentation function, we applied a knock-down strategy. Transfection of a FcRIIB antisense along with hygromycin selection allowed us to yield FcRIIB-deleted DC (DEL). Undeleted DC [wild-type (WT)] were obtained by using a sense construct. Fifteen to 20 percent of the cells initially plated were transfected efficiently and survived antibiotic selection as described previously [23
]. Flow cytometry analysis of costimulation molecules after transfection indicated no transfection-induced maturation (n=6, Fig. 5A
). Immunoprecipitation experiments confirmed that the antisense construct allowed a significant decrease in expression of both FcRIIBs, with no effect on FcRIIA as compared with the sense construct (n=6, Fig. 5B
). FcRIIB expression was reduced by 
75% for IL-13 DC and 85% for IL-4 (Fig. 5B)
. Autologous, primary MLR showed that lymphocyte proliferation was enhanced whenever activating FcRIIA had been selectively engaged through the use of FcRIIB-deleted DC (n=6, Fig. 5C
).
 View larger version (31K): [in a new window] |
Figure 5. DC deficient in FcRIIB expression are better lymphocyte stimulators when IC-pulsed. IL-4 or IL-13 differentiation was performed in parallel for each donor (n=6). Cell surface expression of costimulation molecules (A) revealed lack of maturation after transfection. Data presented here are for IL-4 DC, but similar results were obtained for IL-13 DC (n=6, data not shown). Immunoprecipitations (B) evidenced efficiency and specificity of the FcRIIB antisense construct (n=6). Densitometric analyses were performed for all experiments and are summarized in the graphic portion of B. Transfected IL-4 and IL-13 DC from the same donor were pulsed with IC in autologous MLR (C). Means ± SEM of six independent experiments are shown. Incorporated counts are plotted as a function of the stimulatory cell number, with IC amounts being constant. As controls, MLR were also performed with antitetanus toxin IgG, no DC (data not shown) or with soluble antigen (sol AG).
|
RIIB mechanism of action, we next assessed IC uptake by pure populations of WT and DEL imDC. IL-4 imDC and IL-13 imDC were pulsed with IC, and uptake was evidenced at t = 24 h by intracellular staining for human IgG as described previously [5
, 9
]. FcRIIB DEL imDC demonstrated a greater ability to internalize IC than WT DC (n=6, Fig. 6A
). DEL IL-4 and DEL IL-13 imDC also displayed similar phagocytic capacities (Fig. 6A)
.
 View larger version (53K): [in a new window] |
Figure 6. FcRIIB mechanism of action. Pure populations of DEL and WT IL-4 and IL-13 imDC from the same donor were obtained at day 6 and used for all experiments (n=6). IC internalization (A) was evidenced by intracellular staining for human IgG after 24 h incubation. The first four plots show controls without saponin with IC (open) and without saponin without IC (solid). The next two plots show intracellular staining for WT (shaded), DEL (solid line) DC, and control without IC (solid peak). Means ± SEM of fluorescence intensity (MFI) and percentages of positive cells (means±SEM) are indicated. One representative experiment of six is shown. (B) Surface expression of costimulation and presentation molecules on transfected IL-13 mDC. Incubation with IC was performed with pure populations of DC. Isotypic control fluorescence is shown as a solid peak; those of WT DC as a thin line; and DEL DC as a bold line. Similar results were observed for IL-4 DC. One representative experiment of six is shown. The graph summarizes variations of means of fluorescence intensity at day 9. Means of fluorescence intensity for WT IL-4 DC were taken as 1 for each staining. Data are shown as means ± SEM. Statistically significant differences between staining on WT and DEL DC are marked (*, P0.01). IL-10, p70 IL-12, and IFN- release (C) measured by ELISA in DC conditioned media at day 9. LPS was used to generate positive controls. Values represent means ± SEM of six experiments. No Ag, No antigen.
|
RIIB deletion had no significant impact on the expression levels of presentation and costimulation molecules by LPS mDC (n=6, data not shown).
According to literature data, DC maturation leads to the secretion of cytokines such as IL-10, p70 IL-12, and IFN-, which are known to play a key role in the regulation of immune response [1
2
3
4
]. We therefore used LPS or IC to pulse pure populations of WT/DEL IL-4 and WT/DEL IL-13 DC from the same donor (n=6). Measurement of cytokine amounts was performed at day 9 by ELISA. In the presence of LPS, WT/DEL IL-4 and IL-13 DC behaved similarly and displayed similar cytokine secretion levels (Fig. 6C)
. Conversely, this was not observed when pulsing with IC. No difference between WT and DEL DC was found for IL-10 secretion (Fig. 6C)
. However, FcRIIB deletion resulted in a greater secretion of p70 IL-12 and IFN- for both types of DC (n=6, Fig. 6C
).
Altogether, our results indicate that FcRIIB prevents DC maturation and lymphocyte proliferation by hampering IC uptake and limiting the surface expression of costimulation molecules and release of p70 IL-12 and IFN-.
Modulation of FcRII expression and lymphocyte-proliferative response
The combined action of pro- and anti-inflammatory compounds is thought to maintain immune homeostasis. Among proinflammation mediators, PGE-2, IL-1ß, IFN-, TNF-, ODN, and polyIC are known to increase DC activating receptor expression or to improve full DC maturation [25
26
27
, 32
, 33
]. Anti-inflammatory context or tumor development is often associated with increased levels of vEGF, TGF-ß, IL-6, and IL-10, which tend to counteract the effects of proinflammatory molecules [28
29
30
31
]. We investigated the effects of these mediators, alone or in combination, on a FcRII expression pattern on DC.
First, we observed that none of the tested compounds induced sFcRIIA release by imDC or mDC (IL-4 or IL-13 DC, n=8, data not shown). FcRIIA, -B1, and -B2 expression on IL-4 or IL-13 mDC also remained unaffected (n=8, data not shown). Second, we failed to detect any secretion of sFcRIIA by imDC or mDC or variation of FcRIIA, -B1, and -B2 expression on IL-4 or IL-13 mDC after incubation with mixtures (30 combinations were tested, n=8, data not shown). Some changes were, however, noticed on imDC.
IL-6 treatment, at concentrations beyond 100 ng/ml, resulted in overexpression of FcRIIB on IL-13 DC without affecting FcRIIA (Fig. 7
). Conversely, no effect was observed for IL-4 DC. Similar variations in the FcRII expression pattern were also found on both DC types with a high IL-10 concentration (100 ng/ml, Fig. 7 ). Together, IL-10 and IL-6 up-regulated FcRIIB1 and -B2 expression on IL-4 DC, whereas FcRIIA expression was unaffected (n=6, Fig. 7 ). Further addition of TGF-ß and vEGF resulted in an additional up-regulation of both forms of FcRIIBs, whereas FcRIIA expression remained constant (n=6, Fig. 7
).
 View larger version (49K): [in a new window] |
Figure 7. Regulation of FcRII expression by anti-inflammatory compounds. Membrane expression of FcRIIA, -B1, and -B2 proteins on imDC, as evidenced by immunoprecipitation after 24 h culture with mixtures of anti-inflammatory molecules and further sorting. FcRIIA, -B1, and -B2 expression is the result from the same exposure. Values represent means ± SEM of six independent experiments. Significant variations between control and treated cells are marked (*, P0.05).
|
RIIA was slightly increased, whereas FcRIIB1 and -B2 were strongly decreased, and that of FcRIIB2 became undetectable, and DC remained CD83-negative. The selective reduction of FcRIIB expression observed for IL-13 DC but not for IL-4 DC with PGE-2 alone is worth being noticed (Fig. 8)
. TNF- could also efficiently improve the ratio between activating FcRIIA and inhibitory FcRIIB (B1+B2), as down-regulation of FcRIIBs was more important than that of FcRIIA (Fig. 8)
. CD83 expression was observed when the TNF- concentration was beyond 500 U/ml, and the percentage of CD83-positive cells increased in a TNF- concentration-dependent manner (data not shown). IL-1ß and PGE-2 potentialized TNF- action, and cells remained CD83-negative as long as concentrations were kept low (100 U/ml TNF-, 2 ng/ml IL-1ß, 1 ng/ml PGE-2, Fig. 8
). The simultaneous addition of these compounds led to CD83 expression and simultaneous down-regulation of all FcRIIs, which suggests the initiation of the maturation process (Fig. 8)
. IFN- was unable to regulate expression of any FcRIIs at concentrations ranging from 10 to 500 U/ml (data not shown). Moreover, the down-regulation of FcRIIA and FcRIIB was observed after treatment of both DC types with ODN 2006, ODN 2216, and poly IC at concentrations beyond 5 and 50 µg/ml, respectively (Fig. 8)
. It was also accompanied with CD83 expression when compound concentrations were beyond 5 µg/ml ODN and 50 µg/ml polyIC (Fig. 8)
. The activating FcRIIA/inhibitory FcRIIB (B1+B2) ratio was thus not improved with those molecules.
 View larger version (43K): [in a new window] |
Figure 8. Proinflammation mediators regulate FcRII expression. Immunoprecipitation analysis of membrane expression of FcRIIs following 24 h incubation of imDC with proinflammatory compounds and further sorting (n=6). FcRIIA, -B1, and -B2 expression is the result from the same exposure. Data are displayed as means ± SEM, and statistically significant variations between control and treated DC are indicated (*, P0.05).
|
RIIA/inhibitory FcRIIB (B1+B2) balance were next assessed for several combinations using MLR. Treated DC were first washed twice with cell culture medium and then sorted before MLR were performed by pulsing these DC with IC or a soluble antigen. Treatments were active at modulating lymphocyte proliferation, albeit with different efficiencies. In all cases, the effects were more important when pulsing with IC than with a soluble antigen (n=6, Fig. 9
).
 View larger version (40K): [in a new window] |
Figure 9. Balance between activatory and inhibitory FcRIIs modulates the ability of DC to stimulate T cells. Some mediator-treated DC were sorted and used in MLR assays with IC or soluble antigen (sol Ag) as control (n=6). Negative controls included reactions with antitetanus toxin IgG or without antigen, respectively (table), and without DC (data not shown). Means ± SEM for six independent experiments are shown. Incorporated counts are plotted as a function of the stimulatory cell number, IC, or soluble antigen amount being constant. Table displays mean values ± SEM for the one-tenth ratio. (ND, Not done.)
|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
RIIs strongly modulate the efficiency of humanized antibodies as well as the action of the phagocyte system [7
, 15
, 36
]. In addition, DC express activating and inhibitory FcRIIs [10
], and opsonizing tumor cells with humanized antibodies result in efficient presentation and immune response in vitro and in vivo [6
, 8
, 9
]. As DC are mainly differentiated in vitro, according to two protocols using GM-CSF with IL-4 or GM-CSF with IL-13, we compared the ability of these two types of DC to initiate an immune response when pulsed with IC.
The lack of FcRI and FcRIII expression previously reported for IL-4 DC [4
, 5
, 25
] was confirmed by our experiments. We also found that these receptors are missing on IL-13 DC. Expression levels of FcRIIs are still a matter of debate [4
, 5
]. In agreement with previous data [4
, 5
], we found FcRII expression to be lower on mDC than on imDC. Shedding has been described to be more pronounced when IL-4 DC were pulsed with IC rather than with LPS [4
, 5
]. This was also evidenced in our experiments for IL-4 and IL-13 DC.
As expected, pulsing DC with IC resulted in the initiation of immune response. Tetanus toxin IC were preferred to antibody-coated tumor cells, as they allow the use of a fully autologous system [24 ]. Time-course analysis of apoptotic DC evidenced no difference in DC half-time according to the differentiation protocol or the maturation signal (soluble antigen or IC-linked antigen). Moreover, throughout the experiments performed in our study, the differences between IL-4 and IL-13 DC were only barely significant.
RT-PCR and immunoprecipitation experiments were performed on pure DC populations including BDCA-1+ circulating blood DC, IL-4, and IL-13 DC. In agreement with previous results, FcRIIA and FcRIIB were found to be expressed on human DC [10
]. We also evidenced for the first time the expression of FcRIIB1 and -B2 receptors. It should be noted that the ratios between activating FcRIIA and inhibitory FcRIIB (B1+B2) were constant from one donor to another. The relative constancy of FcRIIA expression on IL-4 and IL-13 DC agrees with previous studies about IL-4 DC [5
, 10
]. However, contrary to a recent report about DC and in agreement with data about monocytes and U937 cells [10
, 16
, 17
], we found that FcRIIB (B1+B2) expression on DC was high but less than FcRIIA. Differences between the techniques (cytometry vs. immunoprecipitation) used to investigate FcRIIB expression on DC may explain this discrepancy. Moreover, RT-PCR analysis allowed us to detect mRNA for the third FcRII expressed in humans, FcRIIC, where its protein is a hybrid gene product resulting from a crossing-over between FcRIIA and FcRIIB genes [37
]. FcRIIC extracellular domain is therefore identical with that of FcRIIB, whereas the intracytoplasmic region is the same as for FcRIIA and includes the ITAM-activating signaling motif [37
]. As a result, FcRIIB and FcRIIC can be detected with antibodies unable to recognize FcRIIA. In our experiments, FcRIIC transcripts and protein levels were low in IL-4 and IL-13 DC as well as in BDCA-1+ DC. Whereas FcRIIA, -B1, and -B2 receptors were readily detectable, different antibodies, extraction, and/or immunoprecipitation buffers may be required for FcRIIC detection. Whether FcRIIC may be functional in human DC also needs to be investigated further.
Furthermore, FcRIIB1 and -B2 were expressed on IL-4 DC. A greater expression of FcRIIB1 and FcRIIB2 on IL-4 DC compared with monocytes is in agreement with the already reported up-regulation of FcRIIB expression by IL-4 [5
, 16
, 17
]. Expression of FcRIIBs by IL-13 DC could also be expected, as IL-13 has been shown recently to control the activity of the FcRIIB promoter [36
]. However, the mechanisms underlying the alternative splicing of the FcRIIB gene product need to be explored further. Of interest is also the expression of FcRIIB1 and -B2, where the functional properties have been shown to differ in a B cell line model, especially regarding capping, internalization, and transduction [38
].
When IC internalization by the phagocytic system through engagement of FcRIIs allows antigen presentation, it requires a sufficient amount of free IC. The secretion of sFcRIIA by murine Langherans cells has been demonstrated to limit their ability to capture IC [18
, 19
]. Our experimental data highlighted a lack of sFcRIIA expression by human-circulating BDCA-1+ blood DC and DC. Immunoprecipitation experiments on one- and tenfold concentrated media were confirmed by analysis of cellular extracts to exclude a fast proteolysis of sFcRIIA in conditioned media detection. RT-PCR experiments also failed to detect sFcRIIA mRNA. Together, these data suggest that unlike their murine counterparts, human BDCA-1+ blood DC and DC cannot escape the maturation process triggered by IC through the secretion of sFcRIIA.
We next investigated more thoroughly whether the balance between FcRII activating and inhibitory receptors could be critical for the initiation of an immune response by IC-pulsed DC, which when deleted for FcRIIB expression, allowed us to show that FcRIIB hampered IC uptake by DC. Intracellular staining 24 h after pulsing DC with IC was more pronounced in DEL DC than in WT for IL-4 and IL-13 DC. After a 4- to 8-h coculture of DC with antibody-coated tumor cells, a recent study showed no improvement of uptake by antibody blockade of FcRIIB/C receptors [10
]. This discrepancy with our own data may result from the use of different strategies to bypass FcRIIBs, different times of analysis, and models. However, in both studies, the selective engagement of activating FcRIIA led to surface remodeling with up-regulation of costimulatory molecules (CD80/80/86) and no alteration of the expression of the presentation molecules (HLA class I and II). We also provided evidence for an up-regulation of CD40 and CD54. In addition, a significant increase in p70IL-12 and IFN- amounts was noted in both studies, and we showed IL-10 secretion not to vary significantly. Finally, the thought that a modulation in vitro of the balance between FcRII-activating FcRIIA/inhibitory FcRIIBs (B1+B2) could be of interest for immunotherapy protocols led us to investigate whether this ratio could be affected by a panel of immune mediators, known to alter DC functions. A marked increase of FcRIIB expression was observed after treatment of IL-4 and IL-13 imDC with a mixture of IL-6 and IL-10, in agreement with a previous report that described the regulation of the FcRIIB promoter activity by IL-10 [36
]. Moreover, the combinations of PGE-2 with IL-1ß or TNF- selectively down-regulated FcRIIB expression. These combinations enhanced the proliferation of lymphocytes in vitro. Such in vitro treatments may be of interest for immunotherapy protocols, although further investigations are still needed to assess these effects for potential therapeutic use.
The engagement of ITIM-containing receptors, Ig-like transcript 2 (ILT-2), ILT-4, and paired Ig-like receptor B (PIRB), has recently been described to render DC tolerogenic with the induction of anergic and immunosuppressive T cells [39
40
41
]. FcRIIB seems unable to act alike, as its activation appears to prevent spontaneous DC activation and "horor toxicus" [10
]. Moreover, the use of a human polyclonal antibody against tetanus toxin allowed us to generate IC able to bind FcRIIA, -B1, and -B2 as a result of the lack of selection regarding IgG subclasses. As pulsing DC with IC resulted, in our study and in those of others [5
6
7
8
9
10
], in the initiation of an immune response, we would then not dissociate the effect of FcRIIB from its expression level and that of FcRIIA. We would suggest that FcRIIB expression level and activation may determine the threshold required for DC activation. Consequently, we surmise that FcRIIB activation, by itself, is not sufficient to render DC tolerogenic. However, consequences of a selective engagement of FcRIIB on DC lifetime need further investigations.
Together, the results reported here evidence the ability of FcRIIB to act as an inhibitory receptor in human DC and describe its mechanism of action. They also provide clues for an improved efficiency of DC vaccination protocols. Engineering the Fc region of IgG molecules to preferentially engage activatory FcRIIA but not inhibitory FcRIIBs (B1+B2) could also allow one to turn this inhibition and improve the efficacy of immunization with IC.
Received March 18, 2005; revised July 19, 2005; accepted August 9, 2005.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
receptors and cross-presentation in dendritic cells J. Exp. Med. 195,F1-F3R-mediated induction of dendritic cell maturation and major histocompatibility complex class I-restricted antigen presentation after immune complex internalization J. Exp. Med. 189,371-380
-B signaling pathway J. Immunol. 170,3963-3970 receptors on dendritic cells J. Exp. Med. 195,1653-1659R provides efficient and versatile vaccination against tumors by dendritic cells J. Immunol. 170,1641-1648 receptor enables human dendritic cells maturation with IL-12p70 production and immunity to antibody-coated tumor cells Proc. Natl. Acad. Sci. USA 102,2910-2915 RIIB regulates negatively BCR-, TCR-, FcR-dependent activation Immunity 3,635-646[CrossRef][Medline] receptor-mediated phagocytosis by a non phagocytic Fc receptor Blood 91,1762-1768RIIB in human monocytic cells J. Biol. Chem. 277,5082-5089 receptors on human monocytes by Th1 and Th2 cytokines J. Immunol. 166,531-537RII/CD32) that inhibits the binding of immune complexes to FcR+ cells J. Immunol. 152,201-212[Abstract]RII/III modulate the antigen-presenting capacity of murine dendritic cells for IgG complexed antigens J. Immunol. 155,1725-1736[Abstract]RII Eur. J. Immunol. 32,530-540[CrossRef][Medline] receptor-mediated phagocytosis by human blood dendritic cells J. Immunol. 157,541-548[Abstract] receptors in human monocytes J. Leukoc. Biol. 77,767-776RII (CD32) isoforms expressed in B lymphocytes J. Immunol. 152,574-585[Abstract]
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
