Sanglifehrin A (SFA) is a novel cyclophilin-binding immunosuppressant with an unknown mechanism of action. IL-12p70 plays a critical role in the pathogenesis of inflammation and autoimmune diseases. We discovered that SFA abrogates bioactive IL-12p70 production by human dendritic cells, the major producers of this cytokine. In direct comparison to the related calcineurin inhibitor cyclosporin A and the mammalian target of rapamycin inhibitor rapamycin, SFA acts uniquely within 1 h to inhibit (80–95%) IL-12p70 production by differentiated dendritic cells. Experiments with Toll-like receptor 3 and 4 ligands show a stimulus-independent suppression. Competitive experiments with a molar excess of cyclosporin A indicate a cyclophilin A-independent blockade of IL-12p70 production. We confirm potent inhibition of IL-12p70 production by SFA using purified human blood DC. Real-time RT-PCR reveals 84–94% suppression of IL-12p40, IL-12p35, and IL-23-specific p19 transcription. These novel insights into the immunosuppressive action of SFA are likely to impact on the clinical use of this agent.
Sanglifehrins, described originally by Sanglier et al. (1, 2), are novel immunophilin-binding immunosuppressants with an unknown mechanism of action that are produced by the actinomycetes strain Streptomyces A92-308110. The immunophilin-binding agent cyclosporin A (CsA)3 binds to cyclophilin and inhibits activity of the calcineurin phosphatase, whereas both FK506 and rapamycin (Rapa) bind to the FK506 binding protein but inhibit two different effector proteins: calcineurin phosphatase and mammalian target of Rapa (mTOR), respectively (3, 4). Although Sanglifehrin A (SFA), like CsA, binds with high affinity to cyclophilin, unlike the latter, SFA does not inhibit calcineurin phosphatase activity (5). SFA has been reported to exhibit lower immunosuppressive activity in the MLR when compared with CsA (1, 5). Studies of the immunosuppressive effects of SFA have been focused on T and B lymphocytes (5, 6). Thus, SFA inhibits IL-2-induced T cell proliferation and mitogen-induced B cell proliferation (5, 6). Similar to Rapa, SFA has been shown to inhibit cell cycle progression of T cells at the G1 phase (7). However, in contrast to Rapa, inhibition of cell cycle progression by SFA was suggested to be related to NF-κB-dependent activation of p53 (6). Unlike the calcineurin inhibitors CsA and FK506, SFA does not suppress IL-2 transcription (5, 6), nor does it inhibit IL-2 receptor expression (6).
Taken together, these data suggest that SFA exerts immunosuppressive effects on T and B cells via a novel mode of action that partially resembles the effects of Rapa. There is increasing evidence that immunophilin-binding agents, such as Rapa, exert immunosuppressive effects at the level of the APC. We reported recently that Rapa impairs Ag uptake of murine dendritic cells (DC) (8), a finding that was subsequently confirmed in human DC (9). Moreover, it was suggested that Rapa promotes in vitro apoptosis of human DC by interfering with GM-CSF signaling (10) and suppresses in vivo DC development and function (11). DC are professional APC with a unique ability to induce and control adaptive and innate immune responses (12, 13, 14, 15). Their maturation results in transition from Ag processing to APCs and is accompanied by up-regulation of costimulatory molecules (12). DC are unsurpassed in their capacity to produce bioactive IL-12, a heterodimeric proinflammatory cytokine that regulates expansion of Th1 and NK cells, bridging innate and adaptive immunity (16).
Given the pivotal role of DC in the induction and control of immune responses, it is necessary to understand how SFA affects their function. This knowledge may provide a framework for the selection and combination of immunosuppressive agents in different clinical settings. Our results show that SFA acts rapidly on human DC to abrogate IL-12p70 production and to a lesser extent, to suppress TNF-α production, while not affecting DC differentiation, phenotypic maturation, or cell viability. Thus, in comparison to the calcineurin inhibitor CsA and the mTOR inhibitor Rapa, SFA targets DC uniquely by rapid suppression of bioactive IL-12 production.
Materials and Methods
SFA, kindly provided by Novartis Pharmaceuticals (Basel, Switzerland), and Rapa and CsA, purchased from Sigma-Aldrich (Seelze, Germany), were used at the indicated concentrations and time points. Stock solutions were prepared in absolute ethanol (vehicle) and diluted on the day of the experiment with culture medium. Control DC were treated with drug vehicle.
Generation of DC
Human PBMC were isolated from buffy coats of healthy blood donors by Ficoll-Paque (Pharmacia, Freiburg, Germany) density gradient centrifugation. CD14+ monocytes were purified (>95%) using CD14 immunomagnetic microbeads (Miltenyi Biotec, Bergisch Gladbach, Germany) and 3 × 106 cells were cultured in six-well flat-bottom plates, in 3 ml of DC medium, comprising RPMI 1640, l-glutamine, penicillin/streptomycin, sodium-pyruvate, nonessential amino acids, 10% heat-inactivated FCS Gold (PAA Laboratories, Linz, Austria), 1000 IU/ml recombinant human (rh) GM-CSF (Novartis Pharmaceuticals), and 1000 IU/ml rhIL-4 (Promo Cell, Heidelberg, Germany). After 3 days, 50% supernatant was replaced with fresh cytokine-containing medium. On day 6, CD1a+ DC represented >90% of cultured cells. The studies of human blood samples were approved by the Institutional Review Board.
Isolation of peripheral blood DC
CD1c+ blood DC (purity >90%) were isolated by immunomagnetic bead sorting (AutoMACS) using the CD1c DC isolation kit (Miltenyi Biotec). Briefly, in a first step, PBMC from buffy coats of blood donors were depleted of B cells with CD19 microbeads and subsequently CD1c+ DC were positively selected after indirect labeling with biotinylated CD1c mAbs and antibiotin microbeads (17).
DC stimulation and detection of cytokine production by ELISA
DC maturation was stimulated with 100 ng/ml LPS (Sigma-Aldrich) in DC medium with cytokines. For stimulation of cytokine production, DC were incubated at 2 × 106/ml in 96-well plates in DC medium with cytokines and stimulated for 24 h with 1 μg/ml LPS (Escherichia coli 0111:B4; Sigma-Aldrich) or 100 μg/ml polyinosinic-polycytidylic acid (poly I:C; Sigma-Aldrich) plus 20 ng/ml rhIFN-γ (BD PharMingen, San Diego, CA). Human IL-12p70 and TNF-α were measured using BD OptEIA ELISA sets (BD PharMingen).
In vivo experiments
SFA stock solution was diluted freshly in 2.5% polysorbate 80, 51% polyethylene glycol (PEG) 300 (Sigma-Aldrich) and 46.5% sterile water. Eight- to 12-wk-old male C57BL10 mice (Charles River Breeding Laboratories, Sulzfeld, Germany) were injected i.p. with SFA (10 mg/kg/day, 3 days) or vehicle. On day 3, mice were injected i.p. with 0.5 μg of murine IL-4 and 10 μg of LPS, dissolved in PBS with 1% FCS. Four hours later, the animals were killed, blood was taken, and plasma IL-12p70 levels were measured by ELISA (R&D Systems, Wiesbaden, Germany).
Surface phenotype of cultured cells was analyzed on a FACSCalibur flow cytometer (BD Biosciences, San Jose, CA) with the following mAbs: HLA-DR-PE (clone L343), CD40-PE (clone 5C3), CD83-PE (clone HB15e), CD86PE (clone GL1), CD14-FITC (clone M5E2), and CD1a-CY (clone HI149). All mAbs, including isotype-matched controls, were purchased from BD PharMingen. Apoptosis and necrosis were analyzed by staining of phosphatidylserine translocation with Annexin VPE in combination with the vital dye 7-AAD (BD PharMingen).
The RNA of 5 × 106 cells was isolated by using the Qiagen Rneasy Mini kit (Hilden, Germany) and transcribed into cDNA using the Ready-To-Go You-Prime First-Strand Beads according to the manufacturer’s instructions (Amersham, Freiburg, Germany). Real-time RT-PCR was performed on an ABI-Prism 7000 PCR cycler (Applied Biosystems, Foster City, CA). The following validated PCR primers and TaqMan MGB probes (6FAM-labeled) were used: IL-12p35 (assay ID: HS00168405_m1), IL-12p40 (Hs00233688_m1), IL-23p19 (Hs00372324_m1). As endogenous controls, two independent primer sets with TaqMan probes were used: eukaryotic 18s ribosomal RNA (Hs99999901_s1) and GAPDH (Hs99999905_m1). PCR mix was prepared according to the manufacturer’s instructions (Assay on demand; Applied Biosystems) and thermal cycler conditions were as follows: 1 × 10 min 95°C, 40–50 cycles denaturation (15 s, 95°C) and combined annealing/extension (1 min, 60°C). Relative quantification was performed by comparison of threshold cycle values of samples with serially diluted standards.
Results and Discussion
SFA does not affect DC differentiation and phenotypic maturation
Analysis of GM-CSF/IL-4-expanded human monocyte-derived DC cultured from day 2 in the presence of SFA (100–1000 nM) and harvested at day 6 showed no significant effects of SFA on DC differentiation, as demonstrated by similar numbers of CD14− CD1a+ immature DC (Fig. 1,A). To induce DC maturation, immature cells were stimulated with LPS. SFA did not affect the up-regulation of costimulatory molecules (CD40, CD80, CD86) or HLA-DR (Fig. 1,B), nor did it affect de novo expression of CD83 (Fig. 1 B). Similar results were obtained after TNF-α (1000 U/ml) stimulation of DC.
SFA blocks bioactive IL-12 production by human DC independent of the nature of the eliciting stimulus
Addition of SFA to DC cultures on day 2 suppressed 80–90% of inducible IL-12p70 production after LPS stimulation (Toll-like receptor (TLR) 4 ligation) when compared with control DC (Fig. 1,C). Because IL-12p70 production can be induced by many different cytokines and microbial agents, we confirmed these findings by a second mode of stimulation through TLR3 with poly I:C and IFN-γ (Fig. 1,C). TNF-α production was also suppressed, but to a lesser extent (Fig. 1 D). To analyze whether these effects were related to apoptotic or necrotic cell death, we performed annexin V/7-AAD staining. In accordance with a recent publication suggesting that SFA can even act as an inhibitor of cell death (18), we consistently found a low incidence of DC death in SFA-treated cultures. In SFA-treated cultures, the mean incidence of apoptotic DC was 4.1 ± 1.1% (control DC 7.5 ± 1.6%, n = 3) and the mean incidence of necrotic DC was 0.9 ± 0.7% (control DC 1.6 ± 1%; n = 3).
SFA acts rapidly on already differentiated DC
To investigate whether SFA needed to be present during DC differentiation to block IL-12p70 production, we added SFA at different time points during DC generation (days 2–6). SFA blocked IL-12p70 production after LPS stimulation at all time points with similar potency (Fig. 2,A). SFA even blocked IL-12p70 production potently when added as late as day 6, 60 min before stimulation, suggesting that it was acting rapidly on differentiated DC (IC50 108 ± 45 nm; Fig. 2 B). Similar results were obtained after poly I:C/IFN-γ stimulation of DC (data not shown).
Suppression of human DC IL-12 production by SFA is unique compared with other immunophilin-binding immunosuppressants
The calcineurin-inhibitor CsA has been reported by different investigators to have a moderate or no suppressive effect on IL-12 production by DC (19, 20). The mTOR inhibitor Rapa has been shown to inhibit DC IL-12 production if present for several days during DC differentiation, either in vitro or in vivo (9, 11). We next questioned whether the rapid and potent suppressive effect of SFA on bioactive IL-12 production by differentiated DC was unique in direct comparison to the immunophilin-binding immunosuppressants CsA and Rapa. Neither Rapa nor CsA was able to inhibit bioactive IL-12 production by DC significantly when added shortly before the stimulation (Fig. 2,C). In direct contrast to CsA and Rapa, SFA was the only immunophilin-binding immunosuppressant that strongly inhibited IL-12p70 production of differentiated DC (Fig. 2 C).
Blockade of IL-12 production by SFA is unlikely to be dependent of cyclophilin binding
Both SFA and CsA bind to cyclophilin in cells with high affinity (5), but affected bioactive IL-12 production by DC in a different manner: SFA blocked IL-12 production, whereas CsA did not inhibit IL-12 production (Fig. 2,C). To assess whether suppression of IL-12 production was dependent on binding of SFA to cyclophilin, we performed competitive experiments with SFA and a molar excess of CsA. We found that SFA still blocked IL-12 production of DC even in the presence of a 20-fold molar excess of CsA indicating that the activity of SFA was independent of cyclophilin binding or at least not inhibited by CsA (Fig. 2 D).
Confirmation of the suppressive effect of SFA on bioactive IL-12 with sorted peripheral blood DC from healthy donors
Given the fact that monocyte-derived DC may display functional differences in comparison to preformed in vivo-generated DC (21), we questioned whether SFA was able to block IL-12 production by freshly isolated peripheral blood DC, purified from healthy blood donors. CD1c+ (BDCA 1) is expressed on CD11c+CD123− DC and represents the major subset of myeloid DC in human blood (17, 22). We purified CD1c+ peripheral blood DC (purity >90%) from blood donors after depletion of B cells by immunomagnetic cell sorting (17). DC were treated with SFA and stimulated with either LPS or poly I:C/IFN-γ. SFA-treated blood DC exhibited a striking decrease in bioactive IL-12 production, suggesting that SFA was effective on freshly isolated peripheral blood DC (Fig. 3 A).
SFA effectively suppresses IL-12 production in vivo
To investigate whether SFA suppresses IL-12 production in vivo, we injected mice with SFA (10 mg/kg/day). On day 3, IL-12 production was stimulated in vivo by injection of LPS/IL-4 and plasma IL-12p70 levels were measured 4 h later. Results demonstrate that a short course of SFA blocks 70% of bioactive IL-12 production in vivo in comparison to vehicle-injected animals (Fig. 3 B).
SFA blocks IL-12 expression by human DC at the transcriptional level
The calcineurin inhibitors CsA and FK506 exert potent immunosuppressive effects on the transcriptional level via inhibition of NFAT-induced cytokine gene expression (23), whereas mTOR inhibition by Rapa results in inhibition of translation initiation and cell cycle arrest (4). To investigate whether SFA inhibited IL-12 production by DC at the transcriptional or translational level, we performed real-time RT-PCR and quantified IL-12p35 and IL-12p40 mRNA expression. The results revealed a strong suppression of IL-12p35 and p40 mRNA expression, whereas expression of two independent housekeeping genes, 18s RNA and GAPDH was not affected (Fig. 4, A–D). Relative quantification of IL-12p35 and p40 transcripts in relation to diluted standards indicated ≥90% inhibition in the presence of 500 nM SFA (Fig. 4,F). Recently, IL-23, a novel member of the IL-12 cytokine family, has been discovered. IL-23 is a heterodimer, comprising IL-12p40 and the IL-23-specific p19 subunit (24). Because IL-23 is produced by DC and is suggested to play a unique role in the activation of memory T cells, as well as in autoimmune inflammation of the brain, we decided to analyze p19 mRNA in SFA-treated DC (24, 25). Short-term pretreatment of DC with SFA (1 h) blocked 84% of p19 transcription (Fig. 4, E and F).
In conclusion, these data provide evidence that the novel cyclophilin-binding immunosuppressant SFA acts rapidly on human DC and blocks bioactive IL-12 production at the transcriptional level. Additionally, we provide evidence that SFA suppresses expression of the IL-12-related, IL-23-specific p19 subunit. Direct comparison with the related drugs CsA and Rapa demonstrates that the rapid action of SFA on DC is unique among immunophilin-binding immunosuppressants. Furthermore, our in vivo experiments confirm the marked suppressive effect of SFA on IL-12p70 production. Given the clinical importance of CsA and Rapa in therapy of transplant rejection and autoimmune diseases (26, 27), these findings are likely to impact on the introduction of SFA into clinical therapy.
We thank Angelika Nockher and Gabriela Haley for excellent technical assistance.
This work was supported by the German Foundation of Hemotherapy Research and in part by National Institutes of Health Grants DK 49745 and AI 41011.
Abbreviations used in this paper: CsA, cyclosporin A; Rapa, rapamycin; mTOR, mammalian target of Rapa; SFA, Sanglifehrin A; DC, dendritic cell; rh, recombinant human; TLR, Toll-like receptor.