Upon stimulation with a wide range of concentrations of CpG oligodeoxynucleotide 2216 (CpG 2216), plasmacytoid DC are induced to produce type I IFN (IFN-α/β). In contrast, CpG 1668 shows a bell-shaped dose–response correlation, i.e. only intermediate but not high doses of CpG 1668 induce IFN-α/β. Interestingly, high-dose CpG 1668 completely inhibited IFN-α responses induced by CpG 2216. Experiments using supernatant of high-dose CpG-1668-treated cells indicated that secreted inhibitor(s) mediated the IFN-α shut-off. Among modulating cytokines, IL-10 turned out to be one important negative regulator. In line with this, supernatants of IL-10-deficient DC cultures stimulated with high-dose CpG 1668 did not inhibit IFN-α production. Interestingly, high-dose CpG 1668 also inhibited IFN-α responses induced by the DNA-encoded mouse cytomegalovirus, whereas IFN-α responses induced by negative-strand RNA-encoded vesicular stomatitis virus were only marginally affected. Experiments with DC cultures devoid of TLR9 indicated that TLR9 was critically required to mediate stimulatory and modulatory signals by low and high concentrations of CpG 1668, respectively. Analysis of purified DC subsets showed that conventional DC were the main IL-10 producers, whereas plasmacytoid DC hardly produced any IL-10.
Toll-like receptors (TLR) are widely expressed recognition receptors of the innate immune system. TLR are evolutionary conserved to recognize pathogen-associated molecular pattern (PAMP) from viruses, bacteria, fungi and parasites. In mice, currently 11 TLR have been identified that all signal in a similar manner due to the presence of a conserved toll/interleukin-1 receptor (TIR) domain in the cytosolic region (reviewed in 1, 2). Among the TLR, TLR9 is triggered by CpG-rich motifs of viral and bacterial unmethylated DNA. Signaling through TLR9, which is expressed primarily by subsets of B cells and plasmacytoid DC (pDC), can also be accomplished by small, synthetic oligodeoxynucleotides (ODN) containing CpG motifs 3. Such CpG ODN hold promise in being used as vaccine adjuvants, inducers of immune deviation toward Th1 responses, and inhibitors of Th2-mediated allergic reactions 4, 5.
Based on their function, immunostimulatory CpG ODN can be classified as CpG-A that induce strong IFN-α production by pDC and limited B-cell activation, CpG-B that induce low IFN-α production by pDC and strong B-cell activation, and CpG-C that combine strong induction of IFN-α with a robust B-cell activation 6–9. Furthermore, immunoregulatory CpG ODN have been described, which inhibit both B-cell and pDC responses to all classes of immunostimulatory CpG ODN 10, 11. Such immunoregulatory CpG ODN selectively block CpG-induced cytokine production but do not interfere with stimulatory activities of other pathogen-associated patterns such as the TLR4 ligand LPS or the lectin Concanavalin A 12. The observation that in co-cultivation experiments spleen cells pretreated with suppressive ODN did not block cytokine production of spleen cells incubated with immunostimulatory ODN suggested that soluble factors were not involved in CpG-mediated immune regulation 12. However, Duramad et al. found that CpG-stimulated pDC produced IL-10 that lowered IFN-α responses 13.
Similarly, virus-induced cytokine production can be modulated by certain stimuli. In this context, Payvandi et al. showed that exogenously added and endogenously produced IL-10 can reduce the frequency of IFN-α-producing cells and the absolute amount of IFN-α produced in responses to viruses such as HSV and vesicular stomatitis virus (VSV) 14. In the same line, other reports showed that monocyte-derived IL-10 suppressed IFN-α/β secretion of adenovirus-stimulated pDC 15, and TNF-α and IL-10 inhibited Sendai virus and HSV-induced IFN-α production of pDC 16. Furthermore, binding of the pDC-specific mAb 440c inhibits CpG-induced IFN-α responses by pDC in vitro17. In addition, vasoactive intestinal peptide 18, extracellular nucleotides 19, and antibody-mediated cross-linking of a variety of cell surface receptors 20 have been identified as negative regulators of type IFN responses by human pDC.
Inappropriate TLR9 signaling can result in immunotoxic effects 21 and there is evidence that chronically activated pDC produce IFN-α in response to TLR stimulation, which plays a role in the pathogenesis of systemic lupus erythematosus 22–24. Furthermore, it is possible that TLR-triggering promotes the pathogenesis of infections by hyperinduction of pro-inflammatory cytokines 25, 26. Consequently, cellular strategies are in place to counter-regulate TLR signaling, including naturally produced soluble TLR, intracellular regulators inhibiting components of the TLR signaling pathway, and TLR-activation induced cell death. All these mechanisms ensure that hyperactivation by TLR-triggering is diminished or hyper-responsive cells are eliminated (reviewed in 27).
In the present study, we addressed why at intermediate concentrations CpG 1668 induced BM-derived Flt3-L cultures to produce IFN-α, whereas at higher concentrations it did not 28. In contrast, other ODN such as CpG 2216 were stimulatory at any concentration above 0.1 μM. We found that high-dose CpG 1668 induced conventional DC (cDC) to secrete IL-10 that generated an anti-inflammatory environment, which abrogated the production of CpG or virus-induced IFN-α/β by Flt3-L-derived pDC.
Graded concentrations of CpG 2216 or CpG 1668 induce differential cytokine production by Flt3-L BM cultures
Previous studies indicated that pDC stimulated with intermediate but not high concentrations of CpG 1668 produced IFN-α 28. To further analyze this phenomenon, we used BM cells cultivated for 8 days in Flt3-L-enriched medium (BM-pDC). Such cultures consisted of 20–40% B220+CD11c+ pDC (data not shown). Upon stimulation of BM-pDC with graded concentrations of CpG 2216, all concentrations above 8 nM induced IFN-α secretion. In contrast, CpG 1668 was able to induce IFN-α only when used at concentrations between 40 and 1.6 nM, whereas at higher concentrations it did not (Fig. 1A). Similar results were obtained when IFN-β responses were analyzed (Fig. 1B). Thus, with respect to the potential to induce type I IFN responses, CpG 2216 showed stimulatory activities over a broad range of concentrations, whereas CpG 1668 was inhibitory or non-stimulating at high concentrations and stimulatory only at intermediate concentrations.
To investigate whether the absence of type I IFN responses of BM-pDC stimulated with high concentrations of CpG 1668 was associated with an overall impairment of the cells, production of the IL-12 subunit p40 was analyzed. Even at high concentrations both CpG ODN induced p40 (Fig. 1C) and upon treatment with high concentrations of CpG 2216 or CpG 1668 (5 μM), BM-pDC upregulated the co-stimulatory molecules CD40 and CD69 (Fig. 1D). Thus, BM-pDC treated with high-dose CpG 1668 are activated, secrete certain cytokines, but are unable to produce type I IFN.
High-dose CpG 1668 can actively block IFN-α responses induced by CpG 2216 or by MCMV
To examine whether high-dose CpG 1668 was unable to induce IFN-α responses, or whether it actively inhibited the production of IFN-α, Flt3-L cultures were co-incubated with 5 μM CpG 2216 and 5 μM CpG 1668. Interestingly, the stimulatory activity of CpG 2216 was completely extinguished upon co-incubation with CpG 1668 (Fig. 2A, left panel). Similar results were obtained when instead of in vitro differentiated Flt3-L cultures total BM or spleen cells were analyzed (Fig. 2A, middle and right panels).
To study whether CpG 1668 inhibited the uptake of CpG 2216 and thus prevented IFN-α induction, BM-pDC were pre-incubated with stimulatory CpG 2216 for 3 or 6 h and subsequently treated with high concentrations of CpG 1668. As shown in Fig. 2B, 3 h delayed treatment with CpG 1668 still abrogated CpG 2216 induced IFN-α responses completely, whereas a 6 h delayed treatment had only a moderate effect. These data and the finding that BM-pDC took up Alexa 647-conjugated CpG 2216 within 10 min (data not shown) indicated that high-dose CpG 1668 inhibited CpG 2216 induced IFN-α responses by mechanisms others than blocking the uptake of stimulatory CpG ODN. Furthermore, down regulation of the CpG ODN receptor TLR9 did not seem to play a role in the CpG-1668-mediated blockade, since TLR9 mRNA levels did not change significantly upon CpG 1668 stimulation of Flt3-L-differentiated cells (data not shown).
To investigate whether secreted or intracellular factor(s) accounted for the inhibition of IFN-α secretion, supernatant (SN) of CpG-1668-treated BM-pDC was generated by incubating the cells with high concentrations of CpG 1668 for 1 h and then washing the cells intensively before incubating them another 23 h with fresh medium (for a detailed description of the protocol, refer to the legend of Fig. 2). SN prepared in that way (SN 1668) inhibited CpG-2216-induced IFN-α production of Flt3-L-differentiated cells by more than 90% (Fig. 2C). This observation indicated that secreted factor(s) played a role in abolishing IFN-α responses. In contrast, SN of untreated BM-pDC did not affect CpG-2216-induced IFN-α responses (SN co; Fig. 2C). To exclude any carryover or efflux of CpG 1668 from cells treated with high-dose ODN, serial log 5 dilutions of the blocking SN were added to BM-pDC. Under such conditions, low concentrations of CpG 1668, if present, should be able to induce IFN-α secretion (compare with Fig. 1A). Nevertheless, the experiment showed that BM-pDC incubated with graded concentrations of SN of CpG-1668-treated cells did not produce IFN-α at any dilution tested excluding carryover or efflux effects of CpG 1668 (Supporting Information Fig. 1).
To further evaluate inhibitory properties of CpG 1668 in the context of natural TLR ligands, infection experiments were carried out. High-dose CpG 1668 was able to completely turn off IFN-α responses induced by DNA-encoded mouse cytomegalovirus (MCMV) that previously has been suggested to induce IFN-α responses by TLR9 triggering 29. In contrast, high-dose CpG 1668 reduced IFN-α responses induced by infection with RNA-encoded VSV, which has been reported to be sensed via TLR7 and probably other mechanisms but not TLR9, by approximately 40% (Fig. 2D). Interestingly, when UV-irradiated VSV was used, which induced only minor IFN-α levels, CpG 1668 was able to inhibit these IFN-α responses completely (Fig. 2D). Thus, treatment of BM-pDC with high-dose CpG 1668 was able to abrogate IFN-α responses induced by synthetic CpG ODN or virus-mediated triggering of TLR9.
CpG 1668 induces IL-10 that inhibits the CpG-2216-mediated induction of IFN-α responses
Given that IL-10 has been described as an anti-inflammatory cytokine that can modulate cytokine responses 13, we next analyzed IL-10 responses induced by graded concentrations of CpG 2216 and CpG 1668. Cells within Flt3-L cultures secreted IL-10 in response to both ODN in a dose-dependent manner. Nevertheless, CpG 1668 induced higher quantities of IL-10 than CpG 2216 (Fig. 3A). Of note, a dose-dependent IL-10 production upon CpG 1668 treatment could also be observed in vivo (Fig. 3B). This suggested that IL-10 might be a critical inhibitor of IFN-α. Indeed, exogenous addition of recombinant IL-10 inhibited the CpG-2216-mediated induction of IFN-α responses in a dose-dependent manner (Fig. 3C). Comparable results were obtained when recombinant IL-10 was used to block IFN-α induction by low-dose CpG 1668 (40 nM) (data not shown). Thus, reminiscent of high-dose CpG 1668, also exogenously added IL-10 was able to abolish CpG-2216-mediated IFN-α production. Moreover, depletion of IL-10 from the SN of high-dose CpG-1668-treated BM-pDC with an IL-10 neutralizing antibody restored CpG-2216-induced IFN-α secretion (Fig. 3D). IL-12 and TNF-α have previously been shown to modulate cytokine responses 16, 30. However, recombinant murine IL-12p70 and/or TNF-α did not significantly affect the CpG-2216-induced IFN-α secretion (data not shown). Thus, IL-10 was one main factor within the SN of high-dose CpG-1668-treated BM-pDC that inhibited IFN-α production.
As many factors have been described that can negatively regulate type I IFN responses (compare introduction), the impact of IL-10 on CpG-1668-triggered IFN-α suppression was further studied by analyzing IL-10-deficient (IL-10−/−) BM-pDC. Numbers and phenotype of cell subsets within BM-pDC differentiated from BM cells of IL-10−/− mice were comparable with those of WT mice (data not shown). Unlike the SN of high-dose CpG-1668-treated WT BM-pDC, the SN of IL-10−/− BM-pDC stimulated with high-dose CpG 1668 only marginally reduced CpG-2216-induced IFN-α responses of WT cultures (Fig. 3E). Furthermore, supplementation of the SN of high-dose CpG-1668-stimulated IL-10−/− BM-pDC with recombinant IL-10 restored the inhibitory capacity (Fig. 3F), showing that indeed IL-10 was the critical factor and that deletion of IL-10 did not have some secondary effect that affected IFN-α production.
Inhibitory and stimulatory effects of CpG ODN are both mediated via TLR9
It has been discussed that prokaryotic DNA can exhibit effects independently of TLR9 31–33. To investigate whether stimulatory and/or inhibitory effects of CpG 2216 and CpG 1668 were mediated by TLR9 triggering, TLR9-deficient (TLR9−/−) BM-pDC were generated (numbers and phenotype of cell subsets within BM-pDC differentiated from BM cells of TLR9−/− mice were comparable with those of WT mice; data not shown). Neither CpG 1668 nor CpG 2216 induced TLR9−/− BM-pDC to produce IFN-α (Fig. 4A), indicating that sensing of both high (hi) and low (lo) dose CpG 2216 and of low-dose CpG 1668 was critically dependent on TLR9. Cytokine responses induced upon VSV infection were not changed in TLR9−/− BM-pDC when compared with WT controls (Fig. 4A and data not shown). To analyze whether high-dose CpG 1668 triggering was mediated via TLR9 as well, WT BM-pDC were pre-incubated with the SN of high-dose CpG-1668-stimulated TLR9−/− BM-pDC and subsequently stimulated with CpG 2216. As shown in Fig. 4B (SN TLR9−/−), IFN-α responses were not significantly reduced, indicating that inhibitory activities of CpG 1668 were also mediated via TLR9 triggering. In addition, no IL-10 production was observed upon CpG 1668 stimulation of TLR9−/− BM-pDC (data not shown). In conclusion, TLR9 played a critical role in sensing of CpG 1668 and of CpG 2216 under all conditions tested. Thus, stimulatory and inhibitory properties of CpG ODN were mediated by the contribution of TLR9.
Upon high-dose CpG 1668 treatment cDC produce IL-10
It is well accepted that BM-pDC are a heterogeneous population consisting of CD11c+CD11b−B220+pDC, CD11c+CD11b+B220− cDC, and few B cells with a varying distribution of the respective cell subsets. FACS analyses of CpG 1668 (5 μM) treated BM-pDC revealed that within these cultures CD11b positive cells accounted for IL-10 production (Fig. 5A). Interestingly, both CD11c+CD11b+ and CD11c+B220+ cells were positive for CD24, a marker defined for the in vitro counterpart of primary CD8+ DC (34; Fig. 5B). To further specify the cell subset accounting for IL-10 production upon treatment with high-dose CpG 1668, DC subsets were FACS-sorted from BM-pDC cultures and then stimulated. To avoid effects of residual contaminating cell subsets, highly purified cell subsets showing a purity of ⩾99% were analyzed. Treatment with 5 μM CpG 1668 resulted in massive IL-10 production only by CD11c+CD11b+B220− cDC, but not by CD11c+CD11b−B220+pDC (Fig. 5C). The inability of pDC to produce huge amounts of IL-10 upon CpG stimulation is in accordance with a recent report 35. In line with these data, spleen cells MACS-depleted of B220+ cells were still able to produce IL-10 upon treatment with high-dose CpG 1668, whereas depletion of CD11b+ cells completely inhibited IL-10 production upon high-dose CpG 1668 treatment (Fig. 5D). Interestingly, GM-CSF cultivated BM-derived myeloid DC (mDC) did not produce IL-10 when stimulated with high-dose CpG 1668 (data not shown). Collectively, these data indicated that cDC but not pDC or mDC primarily accounted for IL-10 production upon high-dose CpG 1668 stimulation of BM-pDC.
In earlier studies it was shown that high concentrations of certain CpG-containing ODN do not induce IFN-α responses, whereas lower concentrations of the same ODN do 28, 36. In this study we analyzed why high concentrations of CpG 1668 did not induce Flt3-L BM cultures (BM-pDC) to produce IFN-α (Fig. 1A). Our results demonstrate that high concentrations of CpG 1668 induced IL-10 that abrogated IFN-α responses (Figs. 2 and 3), whereas at lower CpG 1668 dosages no or less IL-10 was induced (Fig. 3A). Among various different CpG ODN sequences, motifs have been classified as being either immune stimulatory or modulatory 37. Furthermore, CpG-containing ODN have been classified as A-, B-, and C-type, depending on their biological function. Whereas CpG 2216 is an A-type, CpG 1668 has been classified as CpG-B that should induce low IFN-α production by pDC and strong B-cell activation 6–9. Here, we show that depending on the concentration administered, CpG 1668 can either induce or actively inhibit IFN-α responses by BM-pDC. Thus, stimulatory and modulatory activities are associated with one and the same CpG ODN sequence and A-, B-, and C-type classification may be misleading. Interestingly, Kerkmann et al. showed that under physiological conditions CpG 2216 can spontaneously form nanoparticles via G-tetrads, which were required for the immunological activity of CpG 2216 in human pDC 38. Hence, distinct physical properties of CpG ODN used in this study might contribute to the induction of immunomodulatory and -stimulatory cytokines. Inhibitory properties of CpG 1668 affected type I IFN responses by BM-pDC, but did not generally inhibit their cellular activity. As evidenced by upregulation of co-stimulatory markers and secretion of certain cytokines such as IL-12p40, also high-dose CpG 1668 activated BM-pDC (Fig. 1C and D).
We showed that the SN of CpG-1668-treated WT BM-pDC, but not of IL-10−/− BM-pDC, was able to block CpG-2216-induced IFN-α responses by more than 90% (Fig. 3F). In experiments using an IL-10 neutralizing antibody (Fig. 3D) and by supplementation with recombinant IL-10 of the SN of IL-10-deficient BM-pDC treated with high-dose CpG 1668 (Fig. 3F) we showed that among various potential secreted inhibitors, IL-10 played the key role. However, we also observed that direct stimulation of IL-10-deficient Flt3-L-derived DC with high-dose CpG 1668 still resulted in a shut off of CpG-2216-induced IFN-α responses (Supporting Information Fig. 3A). Thus, in the absence of immunoregulatory IL-10, it became evident that intracellular factor(s) also accounted for the inhibition of IFN-α production. In line with this we observed that FACS-purified pDC, which produced only little IL-10 upon stimulation with high-dose CpG 1668 (Fig. 5C), did not secrete IFN-α (Supporting Information Fig. 3B). In this context the role of intracellular inhibitor(s) of type I IFN expression will be a matter of future investigations. A multitude of intracellular molecules has been described, which play a role in negative regulation of TLR triggered responses. These intracellular regulators are capable of affecting TLR signaling cascades at different steps and include the short form of MyD88, MyD88s, that acts as a MyD88 antagonist, interleukin-1 receptor-associated kinase M, SOCS1, and Toll-interacting protein, all of which affect signaling further downstream of TLR and MyD88 (reviewed in 27). The identity of intracellular modulator(s) that play a role in CpG-1668-induced IFN-α suppression remains to be determined.
The observation that in the absence of IL-10 intracellular factor(s) are induced by an external trigger to modulate IFN-α responses points toward a complex regulation mechanism to maintaining adequate TLR9 signaling. TLR9-mediated signaling required for CpG ODN recognition by pDC activates a specific pathway critically involving MyD88, interleukin-1 receptor-associated kinase 1/4, TNF receptor-associated factor 6 (TRAF6), and interferon regulatory factor 7 (IRF-7) 1, 39–41. However, accumulating evidence suggests that DNA can also be recognized independently of TLR9 31, 33, 42–44. Recently, Barrat et al. identified two inhibitory CpG ODN that decreased R848-induced IL-6 production independently of TLR9 45. Klinman et al. showed that TLR9 overexpression can render human kidney cells responsive to some, but not all CpG ODN 32. This suggested that in addition to TLR9, other molecules are able to sense CpG DNA, or that besides TLR9 other molecules are needed as co-receptors for certain CpG DNA motifs to be recognized. Although only low but not high concentrations of CpG 1668 were able to induce type I IFN responses by Flt3-L-derived DC, Hemmi et al. demonstrated that under both conditions IRF-7 expression was induced, suggesting that also high-dose CpG 1668 induced TLR9 signaling 28. To address the contribution of TLR9 in triggering CpG-1668-induced responses, we studied TLR9-deficient Flt3-L-derived DC that were not able to mount IFN-α responses upon low-dose CpG 1668 stimulation and that did not secrete inhibitory IL-10 when stimulated with high-dose CpG 1668 (Fig. 4 and data not shown). Thus, for sensing inhibitory and stimulatory concentrations of CpG 1668, TLR9 expression was necessary and CpG-1668-mediated effects are primarily triggered by TLR9. Interestingly, when Flt3-L-derived DC were treated with GpC 1720, a control ODN in which the CpG motif of CpG 1668 is changed into GpC, minor IFN-α production was induced, whereas in TLR9−/− Flt3-L-derived DC it was not (Supporting Information Fig. 2). Thus, GpC 1720 was sensed via TLR9 and therefore was not an appropriate control for differential TLR9 triggering by varying concentrations of CpG 1668 in this study. Future experiments will reveal whether TLR9 alone is sufficient to mediate stimulatory and inhibitory effects mediated by CpG 1668. At present, it can not be ruled out that TLR9 co-receptors are involved in sensing of high and/or low concentrations of different CpG motifs.
Among TLR family members, TLR9 is closely related to TLR7, which is a sensor of ssRNA. Both receptors share features with respect to molecular structure and function. We found that CpG 1668 was effective in abolishing TLR9-mediated IFN-α production upon stimulation of cells with the TLR9 ligands CpG 2216 or DNA-encoded MCMV (Fig. 2). VSV is sensed via TLR7 46, retinoic-acid-inducible gene-I 47, and most probably other mechanisms but not TLR9. Interestingly, a recent publication showed the ability of different immunoregulatory ODN to inhibit either TLR9, TLR7 or TLR9 and TLR7-mediated IFN-α production by human pDC 45. In the present study, VSV-induced IFN-α responses were reduced by approximately 40% by high concentrations of CpG 1668 (Fig. 2D). This suggested that high-dose CpG 1668 primarily blocked TLR9-triggered IFN-α responses and also had some impact on IFN responses triggered by other TLR.
The property of CpG 1668 to induce modulators of IFN-α at high concentrations seemed to reflect a rather common feature among various different CpG ODN. Duramad et al. found that most CpG ODN tested induced higher IFN-α responses in the presence of IL-10 neutralizing antibodies 13. Similarly, we found that in CpG 2216 stimulation assays, the highest concentration induced lower IFN-α responses than a 5-fold lower concentration tested (Fig. 1A). Furthermore, compared with the WT counterparts, IL-10-deficient BM-pDC produced significantly increased IFN-α levels upon CpG 2216 stimulation (Supporting Information Fig. 3C). Thus, it is likely that under saturating conditions the induction of modulators of IFN-α/β responses constitute a physiological mechanism to maintain appropriate cytokine responses. Our results indicate that IL-10 was one critical modulator among possible others. It is probable that different modulators can synergize to inhibit IFN-α/β responses under appropriate conditions.
Anti-inflammatory cytokines such as IL-10 can be produced by a variety of cell types, including activated T cells, monocytes/macrophages, B cells, and glial cells 48–51. As mice lacking IL-10 develop chronic enterocolitis 52, IL-10 plays a critical role in the maintenance of a local anti-inflammatory microenvironment. Furthermore, IL-10-deficient mice develop massive inflammation, necrosis, and increased serum cytokine level after subcutaneous CpG injection 53. We show that within BM-pDC cultures, cDC were the main IL-10 producers. In contrast, neither pDC nor GM-CSF-derived mDC showed significant IL-10 production upon high-dose GpG 1668 stimulation (Fig. 5 and data not shown). These findings suggest a model in which the interplay between different DC subsets initiates a pro- or an anti-inflammatory environment. Therefore, it is possible that in different tissues, pDC treated with the same stimulus respond by the secretion of differential cytokine patterns, depending on whether they have been stimulated in an anti-inflammatory or a pro-inflammatory microenvironment. This possibility constitutes a new level of regulation of cytokine responses. It will be a matter of future research to analyze how viruses, such as pox viruses, are able to make use of physiological mechanisms to shut off production of anti-viral IFN-α/β responses 54–56.
The present study reveals differential activities of CpG 1668, depending on the concentration administered. CpG-1668-mediated abrogation of type I IFN responses is critically regulated by IL-10. Mechanisms described here probably reflect a physiological system to protect the organism from hyperactivation of DC that could lead to autoimmune reactions. Data presented can have an impact on CpG-based therapeutic approaches in the context of autoimmune diseases and vaccination strategies.
Materials and methods
Mice and viruses
IL-10-deficient mice were provided by Werner Müller 52 and TLR9-deficient mice were provided by Shizuo Akira 3. All mice were bred under specific pathogen-free conditions at the Zentrale Tierhaltung of the Paul-Ehrlich-Institut. Unmutated C57BL/6 mice were purchased from Charles River. Mouse experimental work was carried out using 8- to 12-week old mice in compliance with regulations of German animal welfare.
The VSV variant T1026 used in the experiments was shown to be a particularly strong type I IFN-inducing variant of the HR strain of WT VSV 57. MCMV was provided by Hartmut Hengel (Smith strain ATCC VR-194). All viruses were used at MOI 1 for stimulation experiments. For UV irradiation of virus a UV irradiation chamber (Herolab) was used. In such a device, UV irradiation is adjusted by energy per area (in mJ/cm2); 300 mJ/cm2 irradiation was used, which took approximately 45 s.
Cell isolation and culture
BM cells were isolated by flushing femur and tibia with RPMI supplemented with 10% FBS. Upon red blood cell lysis, cells were washed and seeded at a density of 1×106 cells/mL and 2×106 cells/mL in a medium supplemented with GM-CSF (100 ng/mL; R&D Systems) or Flt3-L (100 ng/mL; R&D systems), respectively. Flt3-L-supplemented cultures were cultivated for 8 days with one medium change at day four, whereas the medium of GM-CSF-supplemented cultures was changed every 1 to 2 days, depending on the status of cultures, by replacing half of the medium with a fresh cytokine-supplemented medium. BM-derived Flt3-L cultures are referred to as BM-pDC.
Flow cytometric analysis and cell separation
Cells were stained with anti-B220-PECy5.5 or -PE mAb, anti-CD11c-APC mAb, anti-CD11b-PacBlue mAb, anti-CD40-PE or -FITC mAb, anti-CD24-PE mAb, and anti-CD69-PE or -FITC mAb (all from BD PharMingen). For intracellular IL-10 staining, cells were stimulated for 18–22 h with 5 μM CpG 1668, subsequently treated with 5 μg/ml Brefeldin A (Sigma), and fixed with 2% PFA. Upon incubation with saponin-containing buffer intracellular staining was performed using anti-IL-10-FITC mAb (eBioscience). Cell sorting was performed using FACS Aria (BD Biosciences). Cells from Flt3-L cultures were FACS-sorted for CD11c+B220+CD11b− or CD11c+B220−CD11b+ with a purity of ⩾99%. For MACS-depletion of CD11b+ or B220+ cells from Flt3-L BM-cultures the autoMACSpro was used (program “deplete S”; Miltenyi Biotech). The efficacy of MACS-depletion usually exceeded 94%.
In vitro stimulation of cells and quantification of cytokine production
For stimulation, ex vivo isolated BM cells, ex vivo isolated spleen cells, or in vitro differentiated BM-derived cultures were seeded at 1×106 cells/well in 24-well culture plates. For stimulation of FACS-sorted pDC or cDC, 2×105 cells/well were seeded in 96-well culture plates. CpG-containing ODN used were CpG 1668 (tccatgacgttcctgatgct; TIB MolBiol), CpG 2216 (ggGGGACGATCGTCgggggG; Sigma-ARK), and GpC 1720 (tccatgagcttcctgatgct; TIB MolBiol). Recombinant murine IL-10, IL-12, and IL-10 neutralizing antibody was purchased by R&D systems. After stimulation, cell-free SN was collected and analyzed with the following ELISA kits: mouse IFN-α and IFN-β (PBL Biomedical Laboratories), mouse IL-10 (R&D Systems), and mouse IL-12p40 or IL-12p70 (R&D Systems).
We thank Dorothea Kreuz and Sabine Falk for expert technical assistance. Claudia Detje for analyzing TLR9 mRNA expression levels, and Harmut Hengel for providing MCMV. This work was supported in part by the Deutsche Forschungsgemeinschaft (SFB432, B15) the EU (INVADERS, contract number QLK2-CT-2001-02103), and ZAFES (CpG-Cluster).
Conflict of interest: The authors declare no financial or commercial conflict of interest.