The immunomodulator FTY720 interferes with effector functions of human monocyte-derived dendritic cells

Authors


Abstract

The potent immunomodulator FTY720 elicits immunosuppression via acting on sphingosine 1-phosphate receptors (S1PR), thereby leading to an entrapment of lymphocytes in the secondary lymphoid tissue. To elucidate the potential in vitro effects of this drug on human monocyte-derived DC, we used low nanomolar therapeutic concentrations of FTY720 and phosphorylated FTY720 (FTY720-P) and investigated their influence on DC surface marker expression, protein levels of S1PR and DC effector functions: antigen uptake, chemotaxis, cytokine production, allostimulatory and Th-priming capacity. We report that both FTY720 and FTY720-P reduce chemotaxis of immature and mature DC. Mature DC generated in the presence of FTY720 or FTY720-P showed an impaired immunostimmulatory capacity and reduced IL-12 but increased IL-10 production. T cells cultured in the presence of FTY720- or FTY720-P-treated DC showed an altered cytokine production profile indicating a shift from Th1 toward Th2 differentiation. In treated immature and mature DC, expression levels for two S1PR proteins, S1P1 and S1P4, were reduced. We conclude that in vitro treatment with FTY720 affects DC features that are essential for serving their role as antigen-presenting cells. This might represent a new aspect of the overall immunosuppressive action of FTY720 and makes DC potential targets of further sphingolipid-derived drugs.

Abbreviations:
CD40L:

CD40 ligand

f-actin:

Filamentous actin

FTY720-P:

Phosphorylated FTY720

LY:

Lucifer Yellow

MFI:

Mean fluorescence intensity

MIP-3β:

Macrophage inflammatory protein-3 beta

PI:

Propidium iodide

S1P:

Sphingosine 1-phosphate

S1PR:

S1P receptor

S1P1–4:

S1PR 1–4

SLC:

Secondary lymphoid tissue chemokine

Introduction

The novel immunomodulator FTY720 causes immunosuppression via enhancement of lymphocyte sequestration into secondary lymphoid organs, which results in a decrease of recirculating lymphocytes and reduced infiltration into sites of antigen challenge 13. This mechanism of action has been shown to be highly effective in suppressing allograft rejection and autoimmune diseases in a variety of animal models 4, 5 and in initial clinical trials 6, 7. As a synthetic analog of the lysophospholipid sphingosine 1-phosphate (S1P), phosphorylated FTY720 (FTY720-P) targets and activates G protein-coupled S1P receptors (S1PR), which are well characterized for their prominent functions in regulating growth-related and cytoskeleton-dependent cellular activities 810. The current concept of FTY720-induced lymphopenia is that FTY720 modulates lymphocyte trafficking in a dual manner: acceleration of migration into secondary lymphoid organs and blocking the egress into efferent lymphatics. Recent evidence suggests that these differing effects result from selective S1PR inactivation via receptor down-regulation, thereby trapping lymphocytes in secondary lymphatic organs 11, 12. In vascular endothelial cells, FTY720-P antagonizes VEGF-induced vascular permeability by inducing adherens junction assembly 13.

The S1PR family consists of five different isoforms (S1P1–S1P5), which influence a diversity of cellular processes. The S1PR have been identified, their isoform-specific differential tissue and cellular expression and their molecular signaling pathways have been characterized, and the elicited cellular responses have been examined 8, 1416. S1P1–3 are widely distributed, whereas S1P4 is mainly expressed in the lymphoid tissue, and S1P5 is restricted to the nervous system. Expression of S1PR in the immune system has been shown to vary among CD4+ and CD8+ T cells, B cells and monocytes 17.

Human dendritic cells (DC) express mRNA of the subtypes S1P1, S1P2, S1P3 and S1P418. S1P has been shown to affect both immature and mature DC by inducing chemotaxis in immature and modulating cytokine release in mature DC 18.

DC arise from bone marrow progenitors and migrate to peripheral tissues, where they serve their function as sentinels of the immune system by capturing and processing antigen at possible portals of entry. Activation of immature DC with signals like microbial products, members of the TNF family [e.g. CD40 ligand (CD40L)] or pro-inflammatory cytokines leads to the initiation of a phenotypic and functional maturation process upon which DC acquire high surface levels of MHC class II and costimulatory molecules and the capacity to produce distinct Th lineage-polarizing cytokines. After migrating into the draining lymph, mature DC potently activate and induce clonal expansion of naive T cells and therefore play a crucial role in initializing and directing primary adaptive immune responses. In addition, in the normal steady state, constantly trafficking DC play a critical role in the maintenance of peripheral immunotolerance. Transplant immunology has come to realize that donor-derived DC originally located in the graft and infiltrating recipient-derived DC are the responsible mediators of direct and indirect allorecognition, respectively. Furthermore, DC maturation within the graft is triggered by cytokine expression due to inflammatory lesions caused by allotransplantation itself. Several studies have shown that this system of APC is essential for instigating graft rejection 19. The fact that DC are the key regulators of alloimmunity has made them potential targets for immunosuppressive therapy 20, 21. This is underscored by ample evidence showing effects on DC functions induced by various immunosuppressive drugs at different stages of the DC life cycle 22.

Most previous studies to elucidate the mechanisms and possible target cells of the immunosuppressive effects of FTY720 have been focused on lymphocytes and endothelial cells. Here, we investigated whether FTY720 and its phosphate metabolite target human DC via S1PR engagement and report specific influences at low nanomolar concentrations on the functions of immature and mature DC. We show inhibition of DC chemotaxis and actin polymerization in immature DC and mature DC by FTY720 and FTY720-P. In mature DC, both FTY720 and FTY720-P reduce their T cell-stimulatory capacity in allogeneic MLR assays and change their cytokine production profile to promote Th2 differentiation. In parallel with these findings, FTY720- and FTY720-P-treated immature and mature DC revealed prominent reductions in the protein expression of the S1P1 and the S1P4 subtype, while the other S1PR were unaltered.

Results

Phenotypes of control DC and DC cultured with FTY720 or FTY720-P differ only in CD18 expression

Expression levels of surface markers involved in DC functions (Fig. 1) were analyzed on immature and mature FTY720- and FTY720-P-treated DC and compared to untreated controls. Upon maturation, expression of CD83 was induced and levels of CD80, CD86, HLA-ABC and HLA-DR molecules were up-regulated to high levels in control DC as well as in FTY720- and FTY720-P-treated DC. Of all analyzed molecules (CD11c, MHC class I, MHC class II, CD40, CD83, CD80, CD86 and CD11a, CD18, CD50, CD58), only CD18 expression was slightly but consistently reduced in both FTY720- and FTY20-P-treated immature and mature DC.

Figure 1.

Comparative flow cytometric analysis of FTY720-treated DC, FTY720-P-treated DC and control DC. Immature DC ± FTY720 (2 ng/ml) or FTY720-P (2 ng/ml) were stained for a panel of surface molecules. Shaded curves, treated DC; continuous lines, control DC; broken lines, isotypes. One experiment out of three with similar results is shown. CD18 expression is consistently reduced in FTY720-/FTY720-P-treated immature and mature DC.

Unchanged DC yields and antigen uptake capacities of FTY720-treated DC, FTY720-P-treated DC and control DC

To determine the yields of differentiated DC, equal numbers of monocytes were plated on day 0 and cultured with and without various concentrations of FTY720 or FTY720-P (2 ng/ml to 2 µg/ml) for 6 days. DC maturation was or was not induced from day 6 to day 8 by the defined cytokine mixture. No significant differences became apparent between the yields of immature and mature DC, FTY720-treated (2–200 ng/ml) DC and FTY720-P-treated (2 ng/ml to 2 µg/ml) DC. At 2 µg/ml, FTY720 strongly reduced DC yields by induction of apoptosis in differentiating cells (Fig. 2). Uptake of soluble antigens by receptor-mediated and fluid-phase endocytosis was measured in immature and mature DC (Fig. 3A, B). Immature DC demonstrated a high uptake of FITC-dextran and Lucifer Yellow (LY), whereas endocytosis was not detectable in mature DC. Treatment with FTY720 and FTY720-P did not significantly alter antigen uptake by immature DC. DC phagocytic capacity was studied by internalization of bacterial particles. Again, no differences were detected (Fig. 3C).

Figure 2.

Apoptosis in FTY720-treated DC, FTY720-P-treated DC and control DC. Apoptosis was detected on day 6 of DC culture by flow cytometry using annexin V-PI staining. Percentages are shown for annexin V+/PI and annexin V+/PI+ cells (n=6). C2-ceramide (20–80 µg/ml) was used for induction of DC apoptosis in positive controls.

Figure 3.

Endocytic activity of DC differentiated in the presence of FTY720 or FTY720-P. (A) Mannose receptor-mediated endocytosis was detected as uptake of FITC-dextran at 37°C: bold lines; as a negative control, cells were incubated at 4°C: dotted lines (n=5). (B) Fluid-phase endocytosis was evaluated as LY uptake. Results are expressed as mean fluorescence intensity (MFI) ± SD (n=5). Immature control DC, FTY720-treated DC and FTY720-P-treated DC were evaluated on day 6. Squares, control DC; triangles, 2 ng/ml; turned triangles, 20 ng/ml; diamonds, 200 ng/ml. (C) Phagocytic activity of FTY720- and FTY720-P-treated DC. Uptake of S. aureus bioparticles was evaluated in immature DC at the indicated time intervals at 37°C or 4°C (control). Cytochalasin D-treated control samples were used for subtraction of background fluorescence. Phagocytic activity is expressed as MFI ± SD (n=5). DC were evaluated on day 6. Squares, control DC; triangles, 2 ng/ml; turned triangles, 20 ng/ml; diamonds, 200 ng/ml.

Immature and mature FTY720- and FTY720-P-treated DC show reduced actin polymerization and chemotaxis in response to DC chemokines

Reorganization of the cytoskeleton is an essential step for the initiation of migration in leukocytes. We therefore investigated polymerization of actin by FITC-phalloidin staining. RANTES and SDF-1α were used to elicit actin polymerization in immature DC. Flow cytometry revealed a rapid response at 30 s, with a plateau at 60–120 s after stimulation. FTY720 and FTY720-P treatment of immature DC resulted in a markedly reduced cellular filamentous actin (f-actin) content at the indicated time intervals after chemokine stimulation, indicating diminished actin remodeling (Fig. 4A). In mature FTY720- and FTY720-P-treated DC that had been stimulated with macrophage inflammatory protein-3 beta (MIP-3β) and secondary lymphoid tissue chemokine (SLC), the relative f-actin content was also found to be lower when compared to mature DC (Fig. 4B).

Figure 4.

FTY720- and FTY720-P-treated DC show reduced actin polymerization in response to DC chemokines. Immature DC (A) and mature DC (B) were stimulated with RANTES; and MIP-3β, respectively. Square, control DC; triangles, FTY720-treated DC (2 ng/ml); turned triangles: FTY720-P-treated DC (2ng/ml). The relative f-actin content was detected after 25, 60, 120 s by NBD-phallacidin staining and subsequent FACS analysis. Results are shown as means ± SD (n=3). *, p<0.05; **, p<0.01; ***, p<0.001.

S1P has been described as a chemoattractant for immature DC, but not for mature DC. FTY720 modulates migratory responses to chemokines in T cells. Hence, we wondered whether FTY720 might modulate chemokine receptor-driven DC migration. DC were allowed to migrate into nitrocellulose filters using RANTES and SDF-1α as stimulators 23. Immature FTY720- and FTY720-P-treated DC showed a significantly decreased chemotactic response to both chemokines (Fig. 5A, left panel). Incubation of immature DC with 2×10–4–2×102 ng/ml FTY720 and FTY720-P dose-dependently inhibited DC migration starting at 2×10–4 ng/ml (Fig. 5B, upper panel). For mature DC, chemotaxis was induced by SLC and MIP-3β (Fig. 5A, right panel). Again, chemotaxis to both chemokines was significantly and dose-dependently reduced in mature FTY720- and FTY720-P-treated DC (Fig. 5B, lower panel).

Figure 5.

FTY720- and FTY720-P-treated DC display reduced chemotaxis towards DC chemokines. (A) Immature DC treated or not with 2 ng/ml FTY720 or FTY720-P were exposed to the chemokines RANTES (20 ng/ml) and SDF-1α (1 *g/ml). Mature DC were exposed to MIP-3β (10 nM) and SLC (1 µg/ml). FTY720- and FTY720-P-treated DC show impaired chemotaxis. Data are means of the chemotactic index (ratio between migration distance towards substance and control medium; n=3). *, p<0.05. (B) Suppression of DC chemotaxis with (sub)nanomolar concentrations of FTY720 and FTY720-P. On day 6, DC were incubated with the indicated concentrations of FTY720 or FTY720-P (45 min). Starting at subnanomolar concentrations, both FTY720 and FTY720-P dose-dependently inhibit immature and mature DC chemotaxis. Data are means of the chemotactic index (ratio between migration distance towards substance and control medium; n=3). *, p<0.05. (C) Suppression of blood DC chemotaxis with (sub)nanomolar concentrations of FTY720 and FTY720-P. Freshly isolated blood DC (CD1c+, CD19) were incubated with increasing doses of FTY720 or FTY720-P. Exposure to the chemokines RANTES (20 ng/ml) and SDF-1α (1 µg/ml) resulted in a similar dose-dependent suppression of chemotaxis as shown in Fig. 5B, upper panel, for immature monocyte-derived DC. Starting at subnanomolar concentrations, both FTY720 and FTY720-P dose-dependently inhibit blood DC chemotaxis. Data are means of the chemotactic index (ratio between migration distance towards substance and control medium; n=3). *, p<0.001.

To confirm the influence of FTY720 and FTY720-P on DC naturally present in human blood, DC were isolated from human blood samples of healthy volunteers. These DC reveal an immature phenotype (CD86+, CD83; data not shown) and were treated with (sub)nanomolar concentrations of FTY720 or FTY720-P for 60 min. Chemotaxis assays were then carried out using RANTES and SDF-1α as stimulators. Chemotaxis was dose-dependently reduced for both FTY720 and FTY720-P and for both chemokines. These results were similar to those obtained with immature DC.

DC chemokine receptor expression was examined by FACS analysis (CCR1, CCR3, CCR5, CCR7 and CXCR4). FTY720 or FTY720-P did not significantly modulate chemokine receptor expression of immature or mature DC (data not shown).

Reduced immunostimulatory capacity of FTY720- and FTY720-P-treated DC, and cytokine production: less IL-12 p70 but more IL-10 following stimulation via CD40 ligation

S1P acts as a potent suppressor of murine T cell proliferation and interferes with the cytokine production in these cells 24, and FTY720, at micromolar concentrations, is known to exert anti-proliferative effects on human T cells 2. We asked whether these effects could be partly attributed to S1PR engagement on APC. MLR with different combinations of stimulators and responders were performed. We examined the influence of FTY720 or FTY720-P treatment on the allogeneic T cell-stimulatory capacity of human DC. Both, FTY720 and FTY720-P treatment of DC markedly suppressed T cell proliferation (n=6; Fig. 6A), although levels of maturation markers and costimulatory molecules remained unchanged. Additional control MLR were performed in parallel to each experiment to exclude a possible carry-over effect of FTY720 or FTY720-P. Untreated DC were suspended in the supernatants from the last wash step of the FTY720-or FYT720-P-treated DC and subjected to MLR. No inhibitory effect on T cell proliferation became evident.

Figure 6.

Mature FTY720- and FTY720-P-treated DC show a reduced T cell stimulatory capacity and an altered cytokine production after stimulation with CD40L-expressing cells. (A) Treated and untreated DC were used as stimulators in an MLR. Upright triangles, mature control DC; turned triangles, immature control DC; open squares, mature FTY720-treated DC; diamonds, mature FTY720-P-treated DC; full squares, immature FTY720-treated DC; circles, immature FTY720-P-treated DC. Data are means ± SD (n=6). T cell proliferation in response to mature DC was significantly reduced when DC where pretreated with FTY720 or FTY720-P (2 ng/ml). (B) Cytokine production of 1×106 DC stimulated with 2.5×105 CD40L-expressing cells for 48 h was measured in the supernatants. Both FTY720- and FTY720-P-treated DC produced significantly less IL-12 but more IL-10 than untreated mature DC (n=5). p<0.05.

As a biological, highly active lipid mediator, S1P has also been reported to play an important role in regulating cytokine production in different cell types 25, 26. In human DC, S1P has been considered to inhibit their capacity to induce Th1 responses 18. Thus, we wondered whether FTY720- or FTY720-P-treated DC differ from control DC in their cytokine secretion profile. Ligation of CD40 in mature FTY720- and FYT720-P-treated DC induced significantly less IL-12 p70 production compared to mature control DC (Fig. 6B). The relative inhibition was 46% for FTY720-treated DC and 48% for FTY720-P-treated DC. In contrast, the IL-10 production of DC was increased by 65% and 70% for FTY720 and FTY720-P treatment, respectively.

Production of IL-4 is enhanced, but production of IFN-γ is reduced, in differentiating naive T cells when stimulated with FTY720- or FTY720-P-treated DC

To further investigate any possible influences of the sphingolipid analogs on the Th cell-polarizing properties of DC, naive T cell priming assays were carried out. CD4+CD45RA+ naive T cells were primed with either control DC, FT720-treated or FTY720-P-treated DC for 6 days. T cells stimulated with mature control DC were mostly CD69+ (Fig. 7A). Surface expression of this activation marker was clearly suppressed when T cells had been co-cultured with mature FTY720- or FTY720-P-treated DC. As expected, control immature DC, FTY720-treated DC or FTY720-P-treated DC poorly induced CD69 on T cells. Intracellular cytokine detection of the proliferated T cells showed a majority of INF-γ-producing T cells and a low amount of IL-4-producing T cells when co-cultured with mature control DC. In contrast, T cells primed with FTY720- or FTY720-P-treated DC differentiated into a markedly lower quantity of IFN-γ-producing T cells, but a significantly higher share of IL-4-producing T cells. In naive T cells stimulated with immature DC, production of cytokines was generally hardly detectable. Similar results were obtained by collecting the supernatants of the co-cultures and detecting cytokine concentrations by ELISA, confirming the lower capacity of FT720- and FTY720-P-treated DC to promote Th1 differentiation and a shift towards Th2 cell differentiation as seen for S1P (Fig. 7B) 18.

Figure 7.

T cell activation after co-culture with FTY720- and FTY720-P-treated DC. (A) CD69 surface expression and quantification of T cell-derived cytokines induced by mature DC ± FTY720 (2 ng/ml) or FTY720-P (2 ng/ml) were analyzed. CD4+CD5RA+ naive T cells (1×106) were co-cultured with 0.25×106 allogeneic mature control DC, FTY720- or FTY720-P-treated DC, as indicated. Fewer T cells were induced to express CD69 by FTY720- and FTY720-P-treated DC than by untreated DC. T cells gated for CD69 expression were analyzed for production of IFN-γ and IL-4. FTY720 and FTY720-P (2 ng/ml) treatment of DC resulted in a reduced production of IFN-γ and an enhanced production of IL-4. (B) Supernatants of the same cultures were analyzed for IFN-γ and IL-4. One representative experiment (upper panel) and a relative measurement of six experiments (lower panel) are shown. FTY720- as well as FTY720-P-treated DC induced the production of significantly lower concentrations of IFN-γ and higher concentrations of IL-4 (n=5). p<0.05.

Isoform-specific differential expression of S1P1–4 proteins in immature and mature DC: lower expression of S1P1 and S1P4 protein in FTY720- and FTY720-P-treated DC

Human T and B cells express different levels of S1PR proteins. mRNA expression of S1P1–4 has been reported for immature and mature human monocyte-derived DC 18. Expression patterns of the S1P1, S1P2, S1P3, and S1P4 proteins were examined in immature and mature control, FTY720- and FTY720-P-treated DC by Western blot analysis. We found subtype-specific differential expression of S1P1–4 proteins, with higher levels of S1P4 protein and slightly lower levels of S1P1 protein in immature DC when compared to mature control DC. Both differentiation stages showed similar expression of S1P2 and only traces of S1P3. Prominent differences in expression of S1P1 and S1P4 were detected for immature and mature FTY720- and FTY720-P-treated DC vs. control DC (Fig. 8). In immature and mature DC cultured in the presence of either drug, S1P1 proteins were reduced to low levels. FTY720-treated DC also showed a lower level of S1P4 protein, which was slightly more decreased in FTY720-P-treated DC. This was observed for immature and mature DC.

Figure 8.

Western blot analysis of S1P1, S1P2, S1P3 and S1P4 in human DC. Protein (50 µg) of immature and mature FTY720- (2 ng/ml) and FTY720-P-treated (2 ng/ml) DC as well as monocytes (MC, positive control) were analyzed for S1PR expression. Mature control DC expressed higher levels of S1P1 than immature DC. Immature and mature FTY720- (2 ng/ml) or FTY720-P-treated (2 ng/ml) DC express less S1P1 and S1P4. S1P3 is hardly detectable, and S1P2 remained unchanged (n=4).

Discussion

We show that in vitro treatment of monocyte-derived DC with therapeutically relevant doses of FTY720 and FTY720-P induces changes in three major DC effector functions: migration, cytokine production profile and T cell stimulation.

In addition to the observed delay in actin polymerization (Fig. 4), a prerequisite for cell migration, DC migration is affected in its capability to follow chemotactic stimuli. Both immature and mature FTY720- and FTY720-P-treated DC show a reduction in the chemotactic response to differentiation stage-specific chemokines (Fig. 5). To verify an effect of freshly isolated DC from peripheral blood, which are of immature phenotype, we exposed these cells to FTY720 and FTY720-P. Again, chemotaxis is dose-dependently reduced with subnanomolar concentrations of FTY720 and FTY720-P. These effects were not caused by modulation of DC chemokine receptor expression. The receptors for FTY720 and FTY720-P, S1P1–4, when engaged by their physiological ligand S1P, have been shown to link sphingolipid signaling to chemokine receptor activity via the multidrug resistance proteins MDR1 and MRP1. Both MDR1 and MRP1 physiologically transport sphingolipid analogs, and both transporter functions are known to be essential for effective mobilization of DC from the epidermis 2729. The decreased chemotactic response of DC upon engagement of the receptors by FTY720 and FTY720-P could be considered a consequence of blocking S1P1–4 for their physiologic ligand. It is still not clear, whether FTY720 exerts its biological effects on T cells mainly via S1PR agonism (i.e. mimicking endogenous S1P) or via other mechanisms that either activate or inhibit intrinsic S1PR activities 30. Recently, selective inactivation of several S1PR, with the most prominent effect on S1P1, has been reported for single S1PR-null cell transductants upon treatment with FTY720. These effects were shown to differ from the classic agonistic functions of the phosphorylated form, but have also added further complexity to the situation 12. One theory proposes a loss of responsiveness towards S1P in the plasma by FTY720, resulting in inhibition of lymphocyte re-entrance into the circulation. This conception was suspected from previous findings revealing that below-plasma concentrations (<0.1 µM) of S1P apparently enhance in vitro chemotactic responses of naive T cells towards CCL21, whereas plasma concentrations (0.3−1.0 µM) inhibit chemotaxis towards CCL21 31.

It has been shown that human DC express mRNA for S1P1–418. FTY720 and FTY720-P treatment induced a down-regulation of S1P1 and S1P4 proteins in human DC (Fig. 8). The altered S1PR expression profiles of FTY720- and FTY720-P-treated DC could therefore provide a possible explanation for the observed reduction of DC chemotaxis and actin polymerization.

In vitro treatment of human DC with S1P during maturation has been reported to impair IL-12 production and to augment IL-10 release, with a consequent skew to Th2 priming 18. The cytokine production profile of DC under the influence of FTY720 and FTY720-P changes from high IL-12/low IL-10 to low IL-12/high IL-10 production (Fig. 6B), with the expected consequence of biasing T cell responses toward Th2. Characteristic cytokine profiles of T cells stimulated with FTY720- and FTY720-P-treated DC are shown in Fig. 7. It is well established that most immune responses involved in allograft rejection and autoimmune diseases are of the Th1 type 19. Assuming that a biasing toward Th2 in the clinical application of FTY720 and FTY720-P takes place, the observed immunosuppressive effects could be at least partly attributed to this alteration in the outcome of an allograft-induced immune response.

The stimulation of T cells by mature DC treated with FTY720 and FTY720-P was impaired (Fig. 6A), although MHC and costimulatory molecule expression was not (Fig. 1). A possible explanation of this reduced capability to mount T cell responses is the decreased expression of CD18 on both immature and mature DC grown in the presence of FTY720 and FTY720-P. CD18 is the β2 subunit of three heterodimers, the α chains being CD11a, CD11b or CD11c. The heterodimer CD11a/CD18 forms the adhesion molecule LFA-1, the binding partner of ICAM-1 32. LFA-1 is one of the molecules enriched in the peripheral zone of the supra-molecular activation cluster, also known as immunologic synapse, formed between APC and T cells. If expression of this molecule is impaired, this might influence the extent of the stimulatory effect. CD18 can also form heterodimers with CD11b and CD11c, both of which are expressed on DC, and can bind to extracellular matrix proteins. A reduced expression of CD18 could therefore also interfere with the migratory activity of DC.

FTY720 pharmacokinetic profiles from initial clinical trials revealed that low nanomolar in vivo concentrations are sufficient to induce a striking lymphopenic status. This is thought to be due to noncompetitive inhibition of S1P effects by FTY720 and S1PR agonism by FTY720-P at these doses. Our data show that in vitro treatment with similar doses of FTY720 or FTY720-P impairs three major effector functions of DC, allowing the conclusion that the immunosuppressive effects of these agents are not limited to lymphocytes, but are also effective on APC.

Materials and methods

Generation of human CD14+ monocyte-derived DC

Human DC were prepared from peripheral blood monocytes essentially as described 33. Monocytes were obtained by isolating CD14+ cells by magnetic sorting (MACS; Miltenyi Biotec, Bergisch-Gladbach, Germany).

FTY720 and FTY720-P (Novartis, Basel, Switzerland; courtesy of V. Brinkmann) were added at 0.7–70 nmol/l (2–200 ng/ml) and kept at these concentrations during the culture period by adding the removed amounts at feeding times.

Isolation of human CD1c (BDCA-1)+ blood DC

Isolation of CD1c+ blood DC from whole blood samples was performed by two immunomagnetic sorting steps using a DC Isolation Kit (MACS; Miltenyi Biotec). The resulting highly purified blood DC were about 95% CD1c+ CD19.

Phenotypic analysis of DC by flow cytometry

FITC-conjugated mAb against the following molecules were used: CD40, CD80, CD86, CD83, HLA-ABC, HLA-DR, CD11a, CD18, CD50, CD58 (all BD PharMingen San Diego, CA), CD83 (Beckman-Coulter, Fullerton, CA) and CD11c (Dako/Immunotech, Glostrup, Denmark). Specimens were analyzed on a FACSCalibur instrument using CellQuest software (BD Systems, San Jose, CA).

Antigen uptake assays by flow cytometry

Cells (1×105) were incubated with 0.5 mg/ml FITC-dextran (Sigma Aldrich, St. Louis, MO) for 30 min at 37°C for detection of mannose receptor-mediated endocytosis; with 1 mg/ml LY dipotassium salt (Sigma Aldrich) for 30, 60 and 120 min to detect macropinocytosis; or with S. aureus particles with or without opsonization (S. aureus Bioparticles; Bioparticle opsonizing reagent, Molecular Probes, Eugene, OR) for 90 and 150 min to detect phagocytosis. Thereafter, samples were washed three times with PBS. For inhibition of phagocytosis, cytochalasin D was added to additional control samples that were similarly processed and used to subtract background fluorescence due to membrane-adherent extracellular particles. Specimens were immediately analyzed by flow cytometry using a FACSCalibur instrument.

Chemotaxis assay

DC migration into nitrocellulose filters towards gradients of chemoattractants [immature DC: RANTES (CCL5; 0.02 µg/ml), SDF-1α (CXCL12; 1 *g/ml); mature DC: MIP-3β (ELC/CCL19; 10 nM), SLC (CCL21; 1µg/ml), (PreproTech, London, UK)] was measured using a 48-well Boyden microchemotaxis chamber (Neuroprobe, Bethesda, MD) and a 5-µm pore filter (Sartorius, Göttingen, Germany) 35. Cells were incubated with various concentrations (1×10–4–1×102 ng/ml) of FTY720 and FTY720-P for 1 h prior to chemotaxis experiments. Cells were washed and subjected to chemotaxis for 4 h; migration depth of cells into the filter was quantified by microscopy, measuring the distance from the surface of the filter to the leading front of cells, before any cell had reached the lower surface (leading-front assay) 36. Data are expressed as the chemotaxis index: the ratio of the distance of stimulated and random migration of DC into the nitrocellulose filters.

Actin polymerization

Immature and mature DC (1×105) were added to 96-well, flat-bottom culture plates and stimulated with RANTES (20 ng/ml) or MIP-3β (10 nM). After 30, 60 and 120 s, cells were fixed in 4% formaldehyde. After 30 min, cells were washed and stained in a permeabilization medium [1 mg/ml lysophosphatidylcholine, 2% formaldehyde, 10 U/ml NBD-phallacidin (Eugene, Leiden, The Netherlands)]. Quantitation of f-actin was by FACS analysis.

MLR

On day 8 of DC culture, cells were harvested and washed. Graded doses of DC (1×104, 3×103, 3×102 and 1×102) were added to 2×105 allogeneic T cells in 96-well, flat-bottom culture plates and cultured for 6 days. Proliferation was detected by [3H] incorporation at 4 µCi–148 KBq/ml [3H]thymidine (specific activity 247.9 GBq/mmol=6.7 Ci/mmol; New England Nuclear, Boston, MA) over the last 16 h in 200 µl culture medium and subsequent measuring of incorporated radioactivity in a liquid scintillation counter (Wallac, Turku, Finland).

Purifying and priming of naive T cells in allogeneic differentiation assays

Bulk T cells were isolated from the rosettes that had formed with neuraminidase-treated sheep red blood cells (Dade Behring, Marburg, Germany) as described 37. To purify naive T cells, bulk T cells were incubated with the following purified mAb: anti-CD8, anti-CD16, anti-CD56, anti-CD45RO, anti-CD14, anti-CD19 and anti-HLA-DR (all from BD PharMingen). Petri dishes were coated for 1 h with AffiniPure goat anti-mouse IgG (10 µg/ml; Jackson ImmunoResearch Laboratories, Avondale, PA). CD4+CD45RA+ naive T cells were isolated using a panning technique as described 38. FACS analysis of the generated cells with anti-CD45RA mAb (BD PharMingen) showed >95% positive cells. Naive T cells were co-cultured with allogeneic DC (control DC vs. FT7720-/FT720-P-treated DC) in 24-well plates at a 4:1 ratio (1×106 T cells; 0.25×106 DC) for 6 days. This was followed by a restimulation period with 25 ng/ml PMA and 1.5 µg/ml ionomycin (Sigma Aldrich) for 4 h.

Intracellular cytokine detection by flow cytometry

After 6 days of DC-T cell co-culture, T cells were restimulated by simultaneous addition of 10 µg/ml of the protein transport inhibitor brefeldin A (GolgiPlugg; BD PharMingen) to prevent cytokine secretion. Cells were washed, fixed with 2% paraformaldehyde and permeabilized with 0.1% saponin. For three-color analysis, activated T cells were gated by CD69 expression via staining with CY5-conjugated mouse anti-CD69 mAb (BD PharMingen) and simultaneously stained with FITC-conjugated mouse anti-IFN-γ and PE-conjugated mouse anti-IL-4 mAb (FastImmune IFN-γ-FITC/IL-4-PE; BD Biosciences, San Jose, CA). Isotype-matched control Ab were used (FastImmune Control γ2a-FITC/γ1-PE). In non-activated T cells, staining of intracellular cytokines was consistently negative. Samples were analyzed on a FACSCalibur instrument using CellQuest software.

Cytokine assays

To determine T cell-derived cytokines, naive T cells were restimulated on day 6 of DC-T cell culture. Culture supernatants were frozen at –80°C until analysis for IFN-γ and IL-4. To measure cytokines secreted by DC, DC were washed three times to removeculture medium and co-cultured with murine myeloma cells transfected with the human CD154/CD40L molecule (P3 × TBA7 cells) in 24-well tissue culture plates at a 2:1 ratio (1×106 DC; 0.5×106 CD40L-expressing cells) 39. These cells were a gift from Dr. R. A. Krozek (Berlin, Germany). Supernatants were taken after 48 h and stored at –80°C until analysis by ELISA. All experiments were analyzed with commercial ELISA kits from Biosource (IFN-γ, IL-4) or BD PharMingen (IL-12 p70, IL-10).

Apoptosis detection

DC apoptosis was analyzed by staining of phosphatidylserine translocation with annexin V in combination with propidium iodide (PI) using the Annexin V-FITC apoptosis detection kit (Alexis Biochemicals, Lausen, Switzerland) according to the manufacturer's instruction. Cells were analyzed on a FACSCalibur instrument using CellQuest software. N-Acetyl-D-sphingosine (20–80 µm/ml; C2-ceramide, Sigma-Aldrich) was used for induction of DC apoptosis in additional positive control samples.

Western blot analyses

Protein expression of S1P1–4 was investigated in FTY720-treated DC, FT720-P-treated DC and in control DC. On day 8, protein was extracted from immature and mature DC using the following buffer: 150 mM NaCl, 1% Nonidet P40, 0.5% deoxycholate, 0.1% SDS, 50 nM Tris-HCl pH 8.0, 0.2 nM PMSF, proteinase inhibitors Complete Mini (Roche, Penzberg, Germany). For Western blot analysis, 50 µg of protein extracts were loaded on 10–20% gradient SDS-PAGE gels using the Mini-PROTEAN III electrophoresis system (Bio-Rad, Hercules, CA) and transferred to PVDF membranes (Amersham, Little Chalfont, UK). Immunodetection of S1PR expression was carried out using isoform-specific mouse anti-human (S1P2, S1P3, S1P4) or rabbit anti-human (S1P1) mAb (Exalpha, Watertown, MA) and secondary anti-mouse (-rabbit) horseradish peroxidase (HRP)-conjugated Ab (Pierce, Rockford, IL). After stripping the blots with Re-Blot Plus (Chemicon, Harrow, UK), a mouse anti-human actin mAb (Chemicon, Harrow, UK) was used. Signal detection was performed by chemiluminescence reaction using Super Signal West Femto Maximum Sensitivity Substrate (Pierce) and by visualizing the blots by a Fluor-S-Imager using Quantity One V4.1 software (Bio-Rad).

Statistical analysis

Data are expressed as means ± SD. Statistical evaluation was performed via Student's t-test and with Kruskal-Wallis analysis of variance (ANOVA) and post-hoc Mann-Whitney U test for chemotaxis experiments.

Acknowledgements

We thank Renate Sperk and Michaela Karches-Boehm for their excellent technical support. The study was supported by the Innsbruck Medical School Science Fund (project 70 to M.T.) the “Verein zur Förderung der Hämatologie, Onkologie und Immunologie” (to G.K.) and by the Austrian Science Fund (FWF; project P16021 to C.H.).

Footnotes

  1. 1

    WILEY-VCH

  2. 2

    WILEY-VCH

  3. 3

    WILEY-VCH

  4. 4

    WILEY-VCH

  5. 5

    WILEY-VCH

  6. 6

    WILEY-VCH

  7. 7

    WILEY-VCH

  8. 8

    WILEY-VCH

Ancillary