MyD88 Expression Is Required for Efficient Cross-P
http://www.100md.com
病菌学杂志 2005年第5期
Microbiology and Tumor Biology Center, Karolinska Institutet, Stockholm
Department of Vaccine Research, Swedish Institute for Infectious Disease Control, Solna, Sweden
ABSTRACT
While virus-infected dendritic cells (DCs) in certain instances have the capacity to activate na?ve T cells by direct priming, cross-priming by DCs via the uptake of antigens from infected cells has lately been recognized as another important pathway for the induction of antiviral immunity. During cross-priming, danger and stranger signals play important roles in modulating immune responses. Analogous to what has been shown for other microbial infections, virally infected cells may contain several pathogen-associated molecular patterns that are recognized by Toll-like receptors (TLRs). We analyzed whether the efficient presentation of antigens derived from infected cells requires the usage of MyD88, which is a common adaptor molecule used by all TLRs. For this study, we used murine DCs that were wild type or deficient in MyD88 expression and fibroblasts that were infected with an alphavirus replicon to answer this question. Our results show that when DCs are directly infected, they are able to activate antigen-specific CD8+ T cells in a MyD88-independent manner. In contrast, a strict requirement of MyD88 for cross-priming was observed when virally infected cells were used as a source of antigen in vitro and in vivo. This indicates that the effects of innate immunity stimulation via the MyD88 pathway control the efficiency of cross-presentation, but not direct presentation or DC maturation, and have important implications in the development of cytotoxic T lymphocyte responses against alphaviral replicon infections.
INTRODUCTION
The priming of T-cell responses is influenced by several factors, such as the form of antigen, the mode of its acquisition, and the type of stimuli. In general, antigens that are endogenously synthesized by antigen-presenting cells (APCs) are directly presented via the classical major histocompatibility complex (MHC) class I pathway, whereas exogenous antigens must be internalized by APCs and rerouted to this pathway in order to elicit an MHC class I-restricted response. The latter pathway is referred to as cross-presentation and is a unique property of dendritic cells (DCs) (6, 21, 22, 31, 39, 42). It requires functional macropinocytosis or phagocytosis (21, 26) and has a pivotal role in the initiation of immune responses to viruses that lack tropism for APCs (40). While direct presentation via infection or transfection with viruses or viral replicons has been demonstrated as an effective way to mature APCs and to induce an efficient CD8+-T-cell response (27, 28), cross-presentation via the uptake of apoptotic materials has also been found to be an attractive way to delivery antigens to the MHC class I pathway (3, 16, 31).
As an innate sentinel, the DC is equipped with a set of Toll-like receptors (TLRs) that recognize several pathogen-associated molecular patterns (PAMPs), and so far 10 different TLRs have been identified for humans and 11 have been identified for mice (52, 59). Some of these receptors recognize structures that are present in viral infections, for instance, double-stranded RNA (dsRNA) and mRNA are recognized by TLR3, heat shock proteins (HSPs) are recognized by TLR2 and TLR4, and ssRNA is recognized by TLR7 and TLR8 (4, 5, 19, 23, 52). Because PAMPs initiate signaling cascades that activate DCs to undergo maturation to allow for the priming of na?ve T cells (17), it was hypothesized that certain TLR ligands, in particular dsRNA, have a key role in the stimulation of immune responses to viruses (32). Indeed, it was found that both TLR9 and TLR3 contribute to the immune defense against mouse cytomegalovirus (MCMV) (51). However, it has also been suggested that TLR3 is dispensable for enhancing immune responses to viruses such as MCMV (13). Because live replicating viruses were used for these studies, it is possible that compensatory effects, e.g., those induced by long-term virus replication, overrode the TLR dependence.
Most TLRs share the common signal adaptor molecule MyD88 (52) to activate the nuclear translocation of NF-B and the maturation of DCs. We asked whether a disruption of this common signaling pathway would influence the innate and adaptive responses to viruses and virally infected cells. An inability to signal via MyD88 during stimulation with certain bacterial and parasitic TLR ligands has been reported to result in an altered adaptive immune response (29, 43, 44), but so far no study has yet shown whether this is the case for viral immunogens, including cell-associated viral antigens.
In an effort to better understand the role of TLR signaling in the generation of antiviral immunity, we compared immune responses generated during alphavirus infection in wild-type or MyD88-targeted mice. We employed a replication-defective viral system, the Semliki Forest virus (SFV) replicon (48, 54), which is a genetically tailored viral replicon consisting of a recombinant single-stranded RNA encapsidated in SFV particles. The viral RNA lacks the genes coding for the viral structural proteins, and therefore new virus particles are not formed in infected cells. Upon infection, the RNA replicon is released from these particles into the cytosol, which leads to subsequent RNA replication and synthesis of the heterologous expressed protein. The heterologous genes used for this study encode green fluorescent protein (GFP) for the tracking of virus-infected cells as well as an intracellular form of ovalbumin (OVA) that allowed us to monitor MHC class I-restricted OVA-specific T-cell responses. Bone marrow-derived DCs that had been directly infected or exposed to SFV-infected cells were studied for their ability to prime naive CD8+ T cells. Our results demonstrate that SFV-infected DCs are capable of maturing and of directly presenting the virus-encoded OVA to naive CD8+ T cells in a MyD88-independent fashion. In contrast, DCs that had been pulsed with infected OVA-expressing cells were dependent on the presence of MyD88 to efficiently cross-prime CD8+ T cells. Moreover, OVA-specific cytotoxic T-lymphocyte (CTL) responses after either the transfer of ex vivo SFV-OVA-infected cells or infection with the SFV-OVA virus also required the presence of MyD88, suggesting that MyD88-dependent antigen presentation is a major pathway for the generation of CD8+-T-cell responses to alphavirus-expressed antigens.
MATERIALS AND METHODS
Mice. Age- and sex-matched MyD88+/+ and MyD88–/– mice (1) in a C57BL/6 background and T-cell receptor transgenic OT-I mice (25) in a Rag1–/– background (a gift from Mary-Jo Wick, Gothenburg University, Gothenburg, Sweden) were kept in a pathogen-free animal facility at the Swedish Institute for Infectious Disease Control. The MyD88–/– mouse strain was genotyped by use of a PCR protocol provided by S. Akira (Osaka University).
Viruses and reagents. Recombinant SFV-OVA, which expresses the intracellular form of OVA (7), and SFV-EGFP-OVA, which expresses both enhanced GFP and OVA, were produced by use of the Two-Helper RNA system (48). The virus stocks were purified by sucrose gradient ultracentrifugation. The negative control consisted of virus particles that were inactivated by short-wave (254 nm) UV light for 60 min (UVP Inc., Upland, Calif.). The H-2 Kb-restricted CTL peptide SIINFEKL (OVA257-264) was purchased from ProImmune (Oxford, United Kingdom). Chicken egg albumin (OVA) was purchased from Sigma-Aldrich (St. Louis, Mo.).
Cells. Bone marrow-derived dendritic cells (BMDCs) (37) were maintained in RP-10 (RPMI supplemented with 10% fetal calf serum, L-glutamine, and Pen/Strep [Sigma]). All in vitro stimulations of BMDCs were performed on day 7 or 8 of culture. A CpG-containing oligonucleotide (Cybergene, Huddinge, Sweden) was used at 1 μM, and a CD40 agonist (HM40-3; BD Biosciences, San Diego, Calif.) was used at 5 μg/ml. The embryonic fibroblast cell line BLK CL.4 (ATCC), derived from the C57BL/6 mouse line, and the EL4 and P815 cell lines were maintained in RP-10. All cell lines tested negative for mycoplasma contamination. OT-I T cells were isolated from the spleens of Rag-1-deficient OT-I mice. In brief, splenocyte suspensions were treated with 25-9-3 (anti-I-Ab), GK.1.5 (anti-CD4), and J11d (anti-HSA) for 45 min on ice, washed, and then depleted by a treatment with rabbit complement for 45 min at 37°C. The purity of OT-I preparations was approximately 75%, as estimated by fluorescence-activated cell sorting (FACS) analysis of CD8+ T cells.
SFV infection of BMDCs and fibroblasts. BMDCs were incubated for 45 min at 37°C in a 5% CO2 incubator with the indicated SFV particles at a multiplicity of infection of 50 in serum-free minimum essential medium (MEM) supplemented with glutamine and Pen/Strep (Sigma), with or without 12.5% autoclaved polyethylene glycol 3000 (PEG 3000; WWR International AB, Stockholm, Sweden). After this incubation, the cells were washed with RP-10 and incubated further. FACS analysis and sorting of SFV-EGFP-OVA-expressing DCs was done at 9 h postinfection. The populations that were GFP positive and GFP negative were collected separately in RP-10 and used immediately in the OT-I assay. BLK CL.4 fibroblasts were infected with SFV-OVA or SFV-EGFP-OVA at a multiplicity of infection of 10 as described above. Free SFV particles were removed from infected cells by a brief treatment with trypsin and succinic acid (20 mM) followed by three phosphate-buffered saline (PBS) washes and then incubated for 2 h before in vivo transfer or for the indicated times for each experiment. This treatment did not affect cell viability.
DC and fibroblast coculture experiment. BMDCs were cocultured with BLK CL.4 fibroblasts that had been infected 8 or 24 h earlier at a ratio of 3:1 (DCs to fibroblasts). After overnight coculturing, the cells were subjected to FACS analysis or were tested in OT-I T-cell assays.
FACS analysis. Cells were blocked with anti-CD16/CD32 and stained with the indicated antibodies. Anti-CD11c (clone HL3)-phycoerythrin (PE), anti-MHC II (clone NIMR-4), anti-MHC I (clone AF6-88.5)-PE or -fluorescein isothiocyanate, anti-CD86 (clone GL1), anti-CD40 (clone 3/23)-biotin, and streptavidin-PE were all obtained from BD Biosciences. Anti-F480-PE (Serotec Ltd. Scandinavia, Hamar, Norway) and anti-rat immunoglobulin G-PE (Jackson Research Laboratories) were also used. Cells were fixed in PBS containing 1% paraformaldehyde and then run on a FACScan instrument. Data were analyzed with Cell Quest Pro software (BD Biosciences).
OT-I ELISPOT assays. Enzyme-linked immunospot (ELISPOT) immunoprecipitation plates (Millipore Co., Bedford, Mass.) were coated with anti-gamma interferon (anti-IFN-; MabTech, Stockholm, Sweden) or anti-interleukin-2 (anti-IL-2; BioSite, T?by, Sweden) antibodies according to the manufacturer's recommendations. After being washed, the plates were blocked for 1 h with RP-5. The indicated numbers of stimulated DCs were incubated with 3 x 104 purified OT-I T cells in the plates, as indicated, with or without 5 μg of CD40 agonist (HM40-3; BD Biosciences)/ml. The plates were then incubated for 24 h for the IL-2 assay and 48 h for the IFN- assay and were washed before the addition of a biotinylated anti-IFN- antibody (MabTech). The spots were developed by use of an avidin-peroxidase complex kit (Vector, Burlingame, Calif.) and the AEC substrate (Sigma) and were counted with an automated ELISPOT reader (Axioplan 2 Imaging; Zeiss, G?ttingen, Germany).
Immunization. Immunizations were performed on days 0 and 14, and T-cell responses were measured on day 24. Immunizations with SFV-OVA-infected fibroblasts (106 cells) or SFV-OVA particles (106 IU) were administered intraperitoneally. The control groups were either given PBS or immunized subcutaneously with 100 μg of the SIINFEKL peptide in incomplete Freund adjuvant (IFA; Sigma).
Antigen restimulation and cytokine ELISA. Spleens were mashed through a 100-μm-pore-size cell strainer (BD Biosciences). After lysis of the red blood cells in RBC lysis buffer (Sigma), the cells were resuspended in RP-5, adjusted to the required concentration, and restimulated with RP-10 alone or RP-10 supplemented with either the SIINFEKL peptide (2 μg/ml) or concanavalin A (ConA) (4 μg/ml). Supernatants collected at 48 h were analyzed by use of an enzyme-linked immunosorbent assay (ELISA) for IFN-. Cytokine ELISA detection reagents were purchased from BD Biosciences (IFN-, IL-12 p70, and tumor necrosis factor alpha [TNF-]), Endogen (IL-1?), and MBL Japan (IL-18). The detection limits for these cytokines were 16, 32, 50, 16, and 26 pg/ml, respectively.
CTLs. (i) Bulk CTL cultures. A cell suspension containing four-fifths of the total number of splenocytes was incubated in RP-5 supplemented with 2 μg of the SIINFEKL peptide/ml for 5 days. On day 5, effector cells were washed in complete medium and used in the 51Cr release assay described below.
(ii) Target cells and 51Cr release assay. Effector cells were diluted twofold in 96-well U-bottomed plates to give dilutions containing 250,000 to 15,000 effector cells per well. A total of 5,000 51Cr-labeled EL4 and P815 target cells in RP-5, with or without the SIINFEKL peptide, were then added to the effector cells and incubated at 37°C for 5 h. The supernatants were harvested, and the amounts of 51Cr released were measured. Spontaneous and total chromium releases were estimated for wells in which the target cells were incubated with RP-5 alone or with 1% Triton X-100, respectively. The percentage of specific lysis was calculated as follows: % lysis = [(sample release – spontaneous release)/(total release – spontaneous release)] x 100.
RESULTS
SFV-infected DCs undergo MyD88-independent maturation. To determine whether MyD88 expression is required for DCs to undergo maturation following exposure to a virus, we infected bone marrow-derived wild-type (WT) or MyD88 knockout (KO) DCs with recombinant SFV particles. These DC preparations consisted of >95% CD11c+ cells (Fig. 1A). The recombinant virus (SFV-EGFP-OVA) contained a double-cistronic RNA molecule that encoded both GFP and the intracellular form of OVA. We found in initial infection experiments that the percentage of infected DCs was low even at a high multiplicity of infection (Fig. 1B). The results indicated that the DCs were refractory to infection under these conditions, but they did not rule out the possibility that replication may occur following viral entry. To overcome the block to infection, we added PEG 3000, a highly hydrated polymer that favors membrane-to-membrane contacts (33), during infection. We observed that this pretreatment increased the percentage of infected cells (GFP-positive cells), from approximately 1% to 20%, as measured at 8 to 10 h postinfection (p.i.). We noted that these percentages decreased to <0.1 and 5%, respectively, at 24 h p.i. (Fig. 1B and data not shown), most probably due to the cytopathic effect caused by SFV replication, which is typically observed after SFV infection (54). We found that virus-treated DCs, irrespective of their MyD88 genotype, showed no difference in their susceptibilities to infection, as measured by the ability to express SFV-encoded GFP (Fig. 1B). Interestingly, when the cells were stained with antibodies for costimulatory molecules at 9 h p.i., we found that the GFP-positive DC population (GFP-pos) had upregulated costimulatory molecules such as CD86 and MHC class II compared to the GFP-negative (GFP-neg) population in the same culture (Fig. 2A). For these experiments, the fluorescence intensity of stained mock-treated controls (undergoing the same treatment as other groups, but with the virus omitted) was set as the baseline level (value of 1). The results suggested that the virus infection had initiated a maturation process in virally targeted DCs. This was observed irrespective of the MyD88 genotype, suggesting that SFV-EGFP-OVA-induced DC maturation is independent of MyD88 (Fig. 2A). An analysis of the whole SFV-treated DC population, which contained mostly noninfected DCs, did not demonstrate the same level of upregulation (Fig. 2B to E). This was partly due to the large proportion of noninfected DCs (approximately 80%). We chose to perform this analysis at an early time point since infection led to a gradual loss of GFP-expressing DCs due to cell death. At later time points (24 h p.i.), we did indeed note an increase in mature DCs (data not shown), suggesting that the exposure to dying infected DCs may have led to the maturation of bystander DCs.
While the control culture stimulated with CpG showed MyD88-dependent maturation (24), MyD88-independent maturation by CD40 cross-linking, which is mediated via the tumor necrosis factor receptor (TNF-R) pathway, was observed (55). The addition of inactivated virus or PEG alone had no significant effect on the maturation process (Fig. 2B to E and data not shown).
SFV-infected DCs are capable of activating antigen-specific T cells. To investigate whether the SFV-EGFP-OVA-infected DCs were capable of activating antigen-specific CD8+ T cells, we cocultured them with na?ve OT-I T cells specific for a Kb-restricted OVA peptide (SIINFEKL) (25). Bulk infected DCs demonstrated a poor capability of activating these T cells, probably because the cultures contained only minute numbers of infected DCs. However, when the DCs had been infected in the presence of PEG, a significant increase in OT-I T-cell activation was observed (Fig. 3A). We noted that virus-infected MyD88 KO DC cultures failed to show the same stimulation capacity as WT cultures. Moreover, stimulation with soluble OVA (20 μg/ml) did not lead to the activation of OT-I cells, suggesting that the result was not due to contamination of the soluble OVA protein in the virus preparations.
Since the virus-infected cultures contained a large proportion of DCs that had not been infected, we could not exclude the possibility that the OT-I activation was mediated by uninfected DCs via cross-presentation after they acquired antigenic material from dying cells. In order to determine if this was the case, we sorted the OVA-expressing DCs by FACS, with gating for GFP expression (9 h p.i.), and then immediately cultured the sorted cells with OT-I T cells. By using this enrichment, we found that the virus-targeted KO DCs showed a similar efficiency as WT DCs for the direct stimulation of OT-I T cells (Fig. 3B). Control wells stimulated with peptide-loaded DCs, with or without PEG treatment, showed that both WT and KO DCs were equally proficient at activating OT-I cells (Fig. 3C). Collectively, these results suggest that the SFV-EGFP-OVA-expressing DCs were capable of mediating direct priming of OVA-specific CD8+ T cells in a MyD88-independent manner. The results also suggested that uninfected bystander DCs may account for certain OT-I activation (Fig. 3B), which may be due to cross-presentation following the uptake of infected DCs.
Exposure to SFV-infected fibroblasts induces DC maturation. In order to examine how DCs respond to SFV-infected cells, we analyzed cocultures consisting of DCs that had been cultured overnight with SFV-EGFP-OVA-infected BLK CL.4 fibroblasts (H-2b). These fibroblasts were washed to remove any excess virus and then incubated for 8 or 24 h to allow antigen expression before DCs were added. While mock-infected fibroblasts remained healthy and adherent, the infected fibroblasts gradually became positive for annexin V and propidium iodide (PI) staining and lost their adherence capacity (Fig. 4A). No detectable amounts of IL-1?, IL-12, IL-18, or TNF- were found in the supernatants from these fibroblast cultures at any time point (data not shown). After overnight coculturing with infected fibroblasts, both WT and MyD88-defective DCs had CD86 and CD40 expression levels that were above those in cells that were cocultured with or without uninfected cells (Fig. 4B). The mean fluorescence intensity (MFI) of the latter control was set as the baseline level (value of 1). However, we observed that only the WT DCs that had been cultured with infected fibroblasts increased the levels of IL-12 production, whereas DCs with defective MyD88 expression did not (Fig. 5A).
MyD88–/– DCs are ineffective at cross-presenting OVA derived from SFV-OVA-infected cells. To assess the efficiency of DCs at cross-presenting cell-associated OVA, we cultured OT-I T cells with DCs that had been cocultured overnight with SFV-OVA-infected fibroblasts. Activated OT-I T cells were enumerated by IL-2- or IFN--specific ELISPOT assays. Our results showed that WT DCs cocultured with infected fibroblasts were fully capable of activating OT-I T cells (Fig. 5B), suggesting that these DCs had taken up and processed antigens derived from virus-infected fibroblasts and could present the processed peptide SIINFEKL under costimulatory conditions. In contrast, MyD88 KO DCs showed a poor capacity to activate OT-I cells under the same conditions. However, this capacity was restored almost fully by a treatment with a CD40 antagonist (HM40-3) during culturing with OT-I T cells (Fig. 5B). Because BLK CL.4 fibroblasts express no detectable CD40 (data not shown), they are unlikely to be subjected to CD40 cross-linking, and therefore the observed effect was most probably due to the CD40-activated DCs (33). This result confirmed the previous observation (Fig. 3A) that uninfected MyD88 KO DCs are inefficient at cross-presenting antigens derived from infected cells. Moreover, we observed that the cross-primed OT-Is produced more IL-2 than IFN-. The reason for this remained to be investigated.
MyD88 knockout mice fail to generate efficient CTL responses to SFV-OVA-infected cells. Since the in vitro results indicated that the MyD88 signaling pathway plays a role during the cross-presentation of cell-associated OVA, we next investigated whether this affected the in vivo response. Accordingly, wild-type or MyD88-deficient mice were immunized with ex vivo SFV-OVA-infected fibroblasts (106 cells/dose), and the OVA-specific MHC class I restricted T-cell response was analyzed. For comparison, we included a group that received an equal amount of SFV-OVA virus particles, as these virus vectors generally induce efficient CTL responses in vivo (36, 54). A SIINFEKL peptide-IFA emulsion was used as a control for the MyD88-independent response. We found that WT and MyD88 KO splenocytes were equally efficient at producing IFN- after stimulation with the mitogen ConA, suggesting that the KO mice had no impairment in the mitogenic IFN- response. Following stimulation with the SIINFEKL peptide, an antigen-specific IFN- response was observed for WT mice infected with the SFV-OVA virus. This was also observed for the WT group that had received SFV-OVA-infected cells (Fig. 6A), indicating that the infected cells had efficiently cross-primed OVA-specific MHC class I-restricted T cells. However, this was not observed for MyD88 knockouts that had been immunized with either infected cells or the SFV-OVA virus, despite a normal response to the ConA mitogen. These results were corroborated by a T-cell cytotoxicity assay, with a total of six independent experiments. Whereas MyD88 knockout animals failed to demonstrate efficient CTL activity, wild-type animals repeatedly demonstrated efficient CTL lysis of SIINFEKL-pulsed EL-4 target cells (Fig. 6B). This defect was not observed for MyD88 KO mice that had been immunized with peptide-IFA, an immunization schedule that circumvents TLR usage. Although the peptide-IFA emulsion elicited peptide-specific CTLs, we did not detect an IFN- response in WT or KO mice. This may have been related to the differences in assay sensitivities (IFN- ELISA versus CTL assay) and lengths of peptide restimulation (2 days versus 5 days).
DISCUSSION
In this study, we showed that MyD88-deficient DCs infected by the SFV-EGFP-OVA virus are capable of directly activating na?ve antigen-specific CD8 T cells (OT-I). In contrast, we found that efficient cross-presentation of SFV-derived OVA acquired from infected cells requires DCs that have intact MyD88 signaling. The latter observation parallels with results obtained from our in vivo experiment, in which immunization with cells infected with the SFV-OVA virus failed to elicit OVA-specific CD8+-T-cell responses in MyD88 knockout mice. This indicates that the efficient cross-priming mediated by the uptake of SFV-infected cellular material in vivo is predominantly MyD88 dependent. It also suggests that the immunogenicity of the SFV particles relies mainly on cross-presentation and cross-priming, as we observed that the SFV-OVA-elicited CTL responses also required MyD88. The fact that MyD88 knockout mice were capable of responding to peptide-IFA immunization as well as mitogen ConA stimulation suggests that they have no significant intrinsic defect in their CD8+-T-cell functions, an observation which is in line with other reports (44, 56).
Given that the virus-infected fibroblasts used were free from infectious virus particles, the in vivo T-cell priming observed for this study was unlikely to be mediated by infected professional APCs, but rather by APCs that had acquired infected cells. This was supported by the observation that SFV targeting of DCs in vitro is inefficient, even in the presence of a highly hydrated PEG polymer. Moreover, several lines of observations have shown that alphavirus replication causes a rapid cytopathic effect on infected cells that leads to apoptotic or necrotic cell death (18, 20, 58). This indicates that immunity-mediated killing is not a necessity in order to obtain cell-associated antigens and therefore differs from other live-antigen delivery systems that do not induce rapid cell death. Our results thus differ from those reported for plasmid vaccines that remained immunogenic in mice with deficiencies in MyD88 or TLR9 (49). Interestingly, it was recently reported that MyD88-deficient DCs have a defect in cross-presentation in a mycobacterial Hsp65 fusion protein model which is independent of TLR4 (41). The explanation underlying the discrepancy in these observations most likely relates to the difference in the immunogens' abilities to engage TLRs, to induce direct versus cross-priming, or to gain compensatory CD4 help (2, 24, 30, 53). In general, antigens that persist for an extended time are more likely to acquire CD4 help, a scenario that perhaps also applies to plasmid-based immunogens that can express antigens in vivo for weeks in transfected cells (57). Similarly, this may also apply to wild-type live viruses that are replication proficient, which in most cases can result in high viral loads in vivo for an extended time. Thus, these potentials may partly override TLR signaling pathways. This is a central aspect that distinguishes the SFV replicon system from other immunogens. Because of this, its immunogenicity is likely to be more vulnerable to the lack of innate stimulation, such as that provided by TLR activation.
Several viruses, including vaccinia virus, measles virus, human immunodeficiency virus, Epstein-Barr virus, and CMV, have the ability to infect DCs (28). Infection can suppress the ability of DCs to mature and can hinder the activation of T cells, as these viruses have evolved strategies to prevent virally encoded antigens from being processed and loaded onto MHC class I and II molecules, thus hampering efficient antigen presentation to virus-specific lymphocytes (15, 38, 47, 50). Our results differ from those of previous reports in the sense that SFV-targeted DCs showed no signs of down-regulating costimulatory molecules or the ability to present endogenously expressed antigens to na?ve T cells during the antigen-expressing phase. In addition, these functions did not appear to depend on the MyD88 pathway, which supports the results of a recent study demonstrating that infection by a DC-tropic RNA virus is capable of switching to conventional nonplasmacytoid DCs in a TLR-independent manner (12). In this regard, DC maturation mediated by intracellular dsRNA pattern recognizers such as protein kinase R would be an alternative way for DCs to sense and act against cytosolic viral pathogens without engaging TLR3. Interestingly, the biological role of TLR3 was recently questioned by Edelmann and colleagues (13), who found that TLR3 is not involved in generating adaptive antiviral immunity to MCMV, vesicular stomatitis virus, lymphocytic choriomeningitis virus, or reovirus. Considering that this last study used several viruses that have the capacity to infect DCs (8, 35, 46), it would be interesting to know whether or not the result was partly related to a direct viral targeting of DCs, which is one possibility for overcoming the need of TLR signaling.
In terms of cross-presentation of cell-associated antigens, an important issue that remains to be studied is the role of mRNA and viral RNA in the physiological process of cross-presentation of cell-associated antigens by APCs, in particular the CD8+ DC subset. Interestingly, this DC subset has been shown to constitutively cross-present antigens (10, 45) and was found to be the only DC subset that has high TLR3 expression (14). Other viral PAMPs, including viral ssRNA (a TLR7 ligand) and host proteins associated with tissue damage, such as the heat shock proteins, fibrinogen, fibronectin, hyaluronic acid, and heparan sulfate (TLR4 ligands), have also been suggested as potent stimulators of APCs (11, 52). Although several MyD88-independent TLR signaling pathways exist, the data shown here stress the importance of MyD88 signaling in the innate and adaptive recognition of viral antigens derived from cells infected by an RNA virus. Our results are in line with recent reports which demonstrated that cross-presentation enhanced by TLR ligands such as poly(I:C) is MyD88 dependent (9) and that a genetically modified heat shock protein (Hsp95) targeted to the cell surface can act as an endogenous signal that spontaneously activates TLR4 via a MyD88-dependent mechanism leading to immune reactions (34). Furthermore, it is known that IL-1 and IL-18 also share MyD88 adaptor usage (52). However, it is likely that these cytokines had no significant role in this study, since no detectable amount of IL-1? (the potent form of IL-1) or IL-18 was found in the supernatants from virally infected fibroblast cultures.
In conclusion, our work has presented evidence that innate signaling can play a crucial role during the elicitation of antiviral immunity and has implicated cross-priming as a main mechanism by which immunity to an alphavirus replicon is generated. Future work will need to investigate whether this also affects the efficiency of T-helper and B-cell responses upon encounters with this type of immunogen.
ACKNOWLEDGMENTS
We thank S. Akira for MyD88–/– mice, M.-J. Wick for OT-I/Rag-1 mice, J.-P. Kraehenbuhl for the OVA plasmid, and Marianne Diurson and Linda Kostic for technical assistance.
This study was supported by the Swedish Research Council and the European Union 5th Framework Programme.
M.C. and C.B. contributed equally to this work.
Present address: Virology Unit, Trinity College, Dublin, Ireland.
REFERENCES
Adachi, O., T. Kawai, K. Takeda, M. Matsumoto, H. Tsutsui, M. Sakagami, K. Nakanishi, and S. Akira. 1998. Targeted disruption of the MyD88 gene results in loss of IL-1- and IL-18-mediated function. Immunity 91:143-150.
Akbari, O., N. Panjwani, S. Garcia, R. Tascon, D. Lowrie, and B. Stockinger. 1999. DNA vaccination: transfection and activation of dendritic cells as key events for immunity. J. Exp. Med. 189:169-178.
Albert, M. L., B. Sauter, and N. Bhardwaj. 1998. Dendritic cells acquire antigen from apoptotic cells and induce class I-restricted CTLs. Nature 392:86-89.
Alexopoulou, L., A. C. Holt, R. Medzhitov, and R. A. Flavell. 2001. Recognition of double-stranded RNA and activation of NF-kappaB by Toll-like receptor 3. Nature 413:732-738.
Asea, A., M. Rehli, E. Kabingu, J. A. Boch, O. Bare, P. E. Auron, M. A. Stevenson, and S. K. Calderwood. 2002. Novel signal transduction pathway utilized by extracellular HSP70: role of Toll-like receptor (TLR) 2 and TLR4. J. Biol. Chem. 277:15028-15034.
Bevan, M. J. 1976. Cross-priming for a secondary cytotoxic response to minor H antigens with H-2 congenic cells which do not cross-react in the cytotoxic assay. J. Exp. Med. 143:1283-1288.
Boyle, J. S., C. Koniaras, and A. M. Lew. 1997. Influence of cellular location of expressed antigen on the efficacy of DNA vaccination: cytotoxic T lymphocyte and antibody responses are suboptimal when antigen is cytoplasmic after intramuscular DNA immunization. Int. Immunol. 12:1897-1906.
Dalod, M., T. Hamilton, R. Salomon, T. P. Salazar-Mather, S. C. Henry, J. D. Hamilton, and C. A. Biron. 2003. Dendritic cell responses to early murine cytomegalovirus infection: subset functional specialization and differential regulation by interferon alpha/beta. J. Exp. Med. 197:885-898.
Datta, S. K., V. Redecke, K. R. Prilliman, K. Takabayashi, M. Corr, T. Tallant, J. DiDonato, R. Dziarski, S. Akira, S. P. Schoenberger, and E. Raz. 2003. A subset of Toll-like receptor ligands induces cross-presentation by bone marrow-derived dendritic cells. J. Immunol. 170:4102-4110.
den Haan, J. M., S. M. Lehar, and M. J. Bevan. 2000. CD8(+) but not CD8(–) dendritic cells cross-prime cytotoxic T cells in vivo. J. Exp. Med. 192:1685-1696.
Diebold, S. S., T. Kaisho, H. Hemmi, S. Akira, and C. Reis e Sousa. 2004. Innate antiviral responses by means of TLR7-mediated recognition of single-stranded RNA. Science 503:1529-1531.
Diebold, S. S., M. Montoya, H. Unger, L. Alexopoulou, P. Roy, L. E. Haswell, A. Al-Shamkhani, R. Flavell, P. Borrow, and C. Reis e Sousa. 2003. Viral infection switches non-plasmacytoid dendritic cells into high interferon producers. Nature 424:324-328.
Edelmann, K. H., S. Richardson-Burns, L. Alexopoulou, K. L. Tyler, R. A. Flavell, and M. B. Oldstone. 2004. Does Toll-like receptor 3 play a biological role in virus infections? Virology 322:231-238.
Edwards, A. D., S. S. Diebold, E. M. Slack, H. Tomizawa, H. Hemmi, T. Kaisho, S. Akira, and C. Reis e Sousa. 2003. Toll-like receptor expression in murine DC subsets: lack of TLR7 expression by CD8 alpha+ DC correlates with unresponsiveness to imidazoquinolines. Eur. J. Immunol. 33:827-833.
Engelmayer, J., M. Larsson, M. Subklewe, A. Chahroudi, W. I. Cox, R. M. Steinman, and N. Bhardwaj. 1999. Vaccinia virus inhibits the maturation of human dendritic cells: a novel mechanism of immune evasion. J. Immunol. 163:6762-6768.
Feng, H., Y. Zeng, M. W. Graner, and E. Katsanis. 2002. Stressed apoptotic tumor cells stimulate dendritic cells and induce specific cytotoxic T cells. Blood 100:4108-4115.
Gallucci, S., and P. Matzinger. 2001. Danger signals: SOS to the immune system. Curr. Opin. Immunol. 13:114-119.
Glasgow, G. M., M. M. McGee, C. J. Tarbatt, D. A. Mooney, B. J. Sheahan, and G. J. Atkins. 1998. The Semliki Forest virus vector induces p53-independent apoptosis. J. Gen. Virol. 79:2405-2410.
Glotzer, J. B., M. Saltik, S. Chiocca, A. I. Michou, P. Moseley, and M. Cotten. 2000. Activation of heat-shock response by an adenovirus is essential for virus replication. Nature 407:207-211.
Grandgirard, D., E. Studer, L. Monney, T. Belser, I. Fellay, C. Borner, and M. R. Michel. 1998. Alphaviruses induce apoptosis in Bcl-2-overexpressing cells: evidence for a caspase-mediated, proteolytic inactivation of Bcl-2. EMBO J. 17:1268-1278.
Guermonprez, P., J. Valladeau, L. Zitvogel, C. Thery, and S. Amigorena. 2002. Antigen presentation and T cell stimulation by dendritic cells. Annu. Rev. Immunol. 20:621-667.
Heath, W. R., and F. R. Carbone. 2001. Cross-presentation, dendritic cells, tolerance and immunity. Annu. Rev. Immunol. 19:47-64.
Heil, F., H. Hemmi, H. Hochrein, F. Ampenberger, C. Kirschning, S. Akira, G. Lipford, H. Wagner, and S. Bauer. 2004. Species-specific recognition of single-stranded RNA via Toll-like receptor 7 and 8. Science 303:1526-1529.
Hemmi, H., O. Takeuchi, T. Kawai, T. Kaisho, S. Sato, H. Sanjo, M. Matsumoto, K. Hoshino, H. Wagner, K. Takeda, and S. Akira. 2000. A Toll-like receptor recognizes bacterial DNA. Nature 408:740-745.
Hogquist, K. A., S. C. Jameson, W. R. Heath, J. L. Howard, M. J. Bevan, and F. R. Carbone. 1999. T cell receptor antagonist peptides induce positive selection. Cell 76:17-27.
Houde, M., S. Bertholet, E. Gagnon, S. Brunet, G. Goyette, A. Laplante, M. F. Princiotta, P. Thibault, D. Sacks, and M. Desjardins. 2003. Phagosomes are competent organelles for antigen cross-presentation. Nature 425:402-406.
Humrich, J., and L. Jenne. 2003. Viral vectors for dendritic cell-based immunotherapy. Curr. Top. Microbiol. Immunol. 276:241-259.
Kaiserlian, D., and B. Dubois. 2001. Dendritic cells and viral immunity: friends or foes? Semin. Immunol. 13:303-310.
Kaisho, T., K. Hoshino, T. Iwabe, O. Takeuchi, T. Yasui, and S. Akira. 2002. Endotoxin can induce MyD88-deficient dendritic cells to support T(h)2 cell differentiation. Int. Immunol. 7:695-700.
Kumaraguru, U., R. J. Rouse, S. K. Nair, B. D. Bruce, and B. T. Rouse. 2000. Involvement of an ATP-dependent peptide chaperone in cross-presentation after DNA immunization. J. Immunol. 165:750-759.
Larsson, M., J. F. Fonteneau, and N. Bhardwaj. 2003. Cross-presentation of cell-associated antigens by dendritic cells. Curr. Top. Microbiol. Immunol. 276:261-275.
Leitner, W. W., H. Ying, and N. P. Restifo. 1999. DNA and RNA-based vaccines: principles, progress and prospects. Vaccine 18:765-777.
Lentz, B. R., and J. K. Lee. 1999. Poly(ethylene glycol) (PEG)-mediated fusion between pure lipid bilayers: a mechanism in common with viral fusion and secretory vesicle release? Mol. Membr. Biol. 16:279-296.
Liu, B., J. Dai, H. Zheng, D. Stoilova, S. Sun, and Z. Li. 2003. Cell surface expression of an endoplasmic reticulum resident heat shock protein gp96 triggers MyD88-dependent systemic autoimmune diseases. Proc. Natl. Acad. Sci. USA 100:15824-15829.
Ludewig, B., K. J. Maloy, C. Lopez-Macias, B. Odermatt, H. Hengartner, and R. M. Zinkernagel. 2000. Induction of optimal anti-viral neutralizing B cell responses by dendritic cells requires transport and release of virus particles in secondary lymphoid organs. Eur. J. Immunol. 30:185-196.
Lundstr?m, K. 2003. Alphavirus vectors for vaccine production and gene therapy. Expert Rev. Vaccines 3:447-459.
Lutz, M. B., N. Kukutsch, A. L. Ogilvie, S. Rossner, F. Koch, N. Romani, and G. Schuler. 1999. An advanced culture method for generating large quantities of highly pure dendritic cells from mouse bone marrow. J. Immunol. Methods 223:77-92.
Mathys, S., T. Schroeder, J. Ellwart, U. H. Koszinowski, M. Messerle, and U. Just. 2003. Dendritic cells under influence of mouse cytomegalovirus have a physiologic dual role: to initiate and to restrict T cell activation. J. Infect. Dis. 187:988-999.
Melief, C. J. 2003. Regulation of cytotoxic T lymphocyte responses by dendritic cells: peaceful coexistence of cross-priming and direct priming? Eur. J. Immunol. 33:2645-2654.
Norbury, C. C., and L. J. Sigal. 2003. Cross priming or direct priming: is that really the question? Curr. Opin. Immunol. 15:82-88.
Palliser, D., H. Ploegh, and M. Boes. 2004. Myeloid differentiation factor 88 is required for cross-priming in vivo. J. Immunol. 172:3415-3421.
Rodriguez, A., A. S. Regnault, M. Kleijmeer, P. Ricciardi-Castagnoli, and S. Amigorena. 1999. Selective transport of internalized antigens to the cytosol for MHC class I presentation in dendritic cells. Nat. Cell Biol. 1:362-368.
Scanga, C. A., J. Aliberti, D. Jankovic, F. Tilloy, S. Bennouna, E. Y. Denkers, R. Medzhitov, and A. Sher. 2002. Cutting edge: MyD88 is required for resistance to Toxoplasma gondii infection and regulates parasite-induced IL-12 production by dendritic cells. J. Immunol. 168:5997-6001.
Schnare, M., G. M. Barton, A. C. Holt, K. Takeda, S. Akira, and R. Medzhitov. 2001. Toll-like receptors control activation of adaptive immune responses. Nat. Immunol. 2:947-950.
Schulz, O., and C. Reis e Sousa. 2002. Cross-presentation of cell-associated antigens by CD8alpha+ dendritic cells is attributable to their ability to internalize dead cells. Immunology 107:183-189.
Sevilla, N., S. Kunz, A. Holz, H. Lewicki, D. Homann, H. Yamada, K. P. Campbell, J. C. de La Torre, and M. B. Oldstone. 2000. Immunosuppression and resultant viral persistence by specific viral targeting of dendritic cells. J. Exp. Med. 192:1249-1260.
Sevilla, N., S. Kunz, D. McGavern, and M. B. Oldstone. 2003. Infection of dendritic cells by lymphocytic choriomeningitis virus. Curr. Top. Microbiol. Immunol. 276:125-144.
Smerdou, C., and P. Liljestr?m. 1999. Two-helper RNA system for production of recombinant Semliki Forest virus particles. J. Virol. 73:1092-1098.
Spies, B., H. Hochrein, M. Vabulas, K. Huster, D. H. Busch, F. Schmitz, A. Heit, and H. Wagner. 2003. Vaccination with plasmid DNA activates dendritic cells via Toll-like receptor 9 (TLR9) but functions in TLR9-deficient mice. J. Immunol. 171:5908-5912.
Steinman, R. M., A. Granelli-Piperno, M. Pope, C. Trumpfheller, R. Ignatius, G. Arrode, P. Racz, and K. Tenner-Racz. 2003. The interaction of immunodeficiency viruses with dendritic cells. Curr. Top. Microbiol. Immunol. 176:1-30.
Tabeta, K., P. Georgel, E. Janssen, X. Du, K. Hoebe, K. Crozat, S. Mudd, L. Shamel, S. Sovath, J. Goode, L. Alexopoulou, R. A. Flavell, and B. Beutler. 2004. Toll-like receptors 9 and 3 as essential components of innate immune defense against mouse cytomegalovirus infection. Proc. Natl. Acad. Sci. USA 101:3516-3521.
Takeda, K., T. Kaisho, and S. Akira. 2003. Toll-like receptors. Annu. Rev. Immunol. 21:335-376.
Tsan, M. F., and B. Gao. 2004. Endogenous ligands of Toll-like receptors. J. Leukoc. Biol. 76:514-519.
Tubulekas, I., P. Berglund, M. Fleeton, and P. Liljestr?m. 1997. Alphavirus expression vectors and their use as recombinant vaccines: a minireview. Gene 190:191-195.
van Kooten, C., and J. Banchereau. 2000. CD40-CD40 ligand. J. Leukoc. Biol. 67:2-17.
Way, S. S., T. R. Kollmann, A. M. Hajjar, and C. B. Wilson. 2003. Protective cell-mediated immunity to Listeria monocytogenes in the absence of myeloid differentiation factor 88. J. Immunol. 171:533-537.
Wolff, J. A., R. W. Malone, P. Williams, W. Chong, G. Acsadi, A. Jani, and P. L. Felgner. 1990. Direct gene transfer into mouse muscle in vivo. Science 247:1465-1468.
Ying, H., T. Z. Zaks, R. F. Wang, K. R. Irvine, U. S. Kammula, F. M. Marincola, W. W. Leitner, and N. P. Restifo. 1999. Cancer therapy using a self-replicating RNA vaccine. Nat. Med. 5:823-827.
Zhang, D., G. Zhang, M. S. Hayden, M. B. Greenblatt, C. Bussey, R. A. Flavell, and S. Ghosh. 2004. A Toll-like receptor that prevents infection by uropathogenic bacteria. Science 303:1522-1526.(Margaret Chen, Christina )
Department of Vaccine Research, Swedish Institute for Infectious Disease Control, Solna, Sweden
ABSTRACT
While virus-infected dendritic cells (DCs) in certain instances have the capacity to activate na?ve T cells by direct priming, cross-priming by DCs via the uptake of antigens from infected cells has lately been recognized as another important pathway for the induction of antiviral immunity. During cross-priming, danger and stranger signals play important roles in modulating immune responses. Analogous to what has been shown for other microbial infections, virally infected cells may contain several pathogen-associated molecular patterns that are recognized by Toll-like receptors (TLRs). We analyzed whether the efficient presentation of antigens derived from infected cells requires the usage of MyD88, which is a common adaptor molecule used by all TLRs. For this study, we used murine DCs that were wild type or deficient in MyD88 expression and fibroblasts that were infected with an alphavirus replicon to answer this question. Our results show that when DCs are directly infected, they are able to activate antigen-specific CD8+ T cells in a MyD88-independent manner. In contrast, a strict requirement of MyD88 for cross-priming was observed when virally infected cells were used as a source of antigen in vitro and in vivo. This indicates that the effects of innate immunity stimulation via the MyD88 pathway control the efficiency of cross-presentation, but not direct presentation or DC maturation, and have important implications in the development of cytotoxic T lymphocyte responses against alphaviral replicon infections.
INTRODUCTION
The priming of T-cell responses is influenced by several factors, such as the form of antigen, the mode of its acquisition, and the type of stimuli. In general, antigens that are endogenously synthesized by antigen-presenting cells (APCs) are directly presented via the classical major histocompatibility complex (MHC) class I pathway, whereas exogenous antigens must be internalized by APCs and rerouted to this pathway in order to elicit an MHC class I-restricted response. The latter pathway is referred to as cross-presentation and is a unique property of dendritic cells (DCs) (6, 21, 22, 31, 39, 42). It requires functional macropinocytosis or phagocytosis (21, 26) and has a pivotal role in the initiation of immune responses to viruses that lack tropism for APCs (40). While direct presentation via infection or transfection with viruses or viral replicons has been demonstrated as an effective way to mature APCs and to induce an efficient CD8+-T-cell response (27, 28), cross-presentation via the uptake of apoptotic materials has also been found to be an attractive way to delivery antigens to the MHC class I pathway (3, 16, 31).
As an innate sentinel, the DC is equipped with a set of Toll-like receptors (TLRs) that recognize several pathogen-associated molecular patterns (PAMPs), and so far 10 different TLRs have been identified for humans and 11 have been identified for mice (52, 59). Some of these receptors recognize structures that are present in viral infections, for instance, double-stranded RNA (dsRNA) and mRNA are recognized by TLR3, heat shock proteins (HSPs) are recognized by TLR2 and TLR4, and ssRNA is recognized by TLR7 and TLR8 (4, 5, 19, 23, 52). Because PAMPs initiate signaling cascades that activate DCs to undergo maturation to allow for the priming of na?ve T cells (17), it was hypothesized that certain TLR ligands, in particular dsRNA, have a key role in the stimulation of immune responses to viruses (32). Indeed, it was found that both TLR9 and TLR3 contribute to the immune defense against mouse cytomegalovirus (MCMV) (51). However, it has also been suggested that TLR3 is dispensable for enhancing immune responses to viruses such as MCMV (13). Because live replicating viruses were used for these studies, it is possible that compensatory effects, e.g., those induced by long-term virus replication, overrode the TLR dependence.
Most TLRs share the common signal adaptor molecule MyD88 (52) to activate the nuclear translocation of NF-B and the maturation of DCs. We asked whether a disruption of this common signaling pathway would influence the innate and adaptive responses to viruses and virally infected cells. An inability to signal via MyD88 during stimulation with certain bacterial and parasitic TLR ligands has been reported to result in an altered adaptive immune response (29, 43, 44), but so far no study has yet shown whether this is the case for viral immunogens, including cell-associated viral antigens.
In an effort to better understand the role of TLR signaling in the generation of antiviral immunity, we compared immune responses generated during alphavirus infection in wild-type or MyD88-targeted mice. We employed a replication-defective viral system, the Semliki Forest virus (SFV) replicon (48, 54), which is a genetically tailored viral replicon consisting of a recombinant single-stranded RNA encapsidated in SFV particles. The viral RNA lacks the genes coding for the viral structural proteins, and therefore new virus particles are not formed in infected cells. Upon infection, the RNA replicon is released from these particles into the cytosol, which leads to subsequent RNA replication and synthesis of the heterologous expressed protein. The heterologous genes used for this study encode green fluorescent protein (GFP) for the tracking of virus-infected cells as well as an intracellular form of ovalbumin (OVA) that allowed us to monitor MHC class I-restricted OVA-specific T-cell responses. Bone marrow-derived DCs that had been directly infected or exposed to SFV-infected cells were studied for their ability to prime naive CD8+ T cells. Our results demonstrate that SFV-infected DCs are capable of maturing and of directly presenting the virus-encoded OVA to naive CD8+ T cells in a MyD88-independent fashion. In contrast, DCs that had been pulsed with infected OVA-expressing cells were dependent on the presence of MyD88 to efficiently cross-prime CD8+ T cells. Moreover, OVA-specific cytotoxic T-lymphocyte (CTL) responses after either the transfer of ex vivo SFV-OVA-infected cells or infection with the SFV-OVA virus also required the presence of MyD88, suggesting that MyD88-dependent antigen presentation is a major pathway for the generation of CD8+-T-cell responses to alphavirus-expressed antigens.
MATERIALS AND METHODS
Mice. Age- and sex-matched MyD88+/+ and MyD88–/– mice (1) in a C57BL/6 background and T-cell receptor transgenic OT-I mice (25) in a Rag1–/– background (a gift from Mary-Jo Wick, Gothenburg University, Gothenburg, Sweden) were kept in a pathogen-free animal facility at the Swedish Institute for Infectious Disease Control. The MyD88–/– mouse strain was genotyped by use of a PCR protocol provided by S. Akira (Osaka University).
Viruses and reagents. Recombinant SFV-OVA, which expresses the intracellular form of OVA (7), and SFV-EGFP-OVA, which expresses both enhanced GFP and OVA, were produced by use of the Two-Helper RNA system (48). The virus stocks were purified by sucrose gradient ultracentrifugation. The negative control consisted of virus particles that were inactivated by short-wave (254 nm) UV light for 60 min (UVP Inc., Upland, Calif.). The H-2 Kb-restricted CTL peptide SIINFEKL (OVA257-264) was purchased from ProImmune (Oxford, United Kingdom). Chicken egg albumin (OVA) was purchased from Sigma-Aldrich (St. Louis, Mo.).
Cells. Bone marrow-derived dendritic cells (BMDCs) (37) were maintained in RP-10 (RPMI supplemented with 10% fetal calf serum, L-glutamine, and Pen/Strep [Sigma]). All in vitro stimulations of BMDCs were performed on day 7 or 8 of culture. A CpG-containing oligonucleotide (Cybergene, Huddinge, Sweden) was used at 1 μM, and a CD40 agonist (HM40-3; BD Biosciences, San Diego, Calif.) was used at 5 μg/ml. The embryonic fibroblast cell line BLK CL.4 (ATCC), derived from the C57BL/6 mouse line, and the EL4 and P815 cell lines were maintained in RP-10. All cell lines tested negative for mycoplasma contamination. OT-I T cells were isolated from the spleens of Rag-1-deficient OT-I mice. In brief, splenocyte suspensions were treated with 25-9-3 (anti-I-Ab), GK.1.5 (anti-CD4), and J11d (anti-HSA) for 45 min on ice, washed, and then depleted by a treatment with rabbit complement for 45 min at 37°C. The purity of OT-I preparations was approximately 75%, as estimated by fluorescence-activated cell sorting (FACS) analysis of CD8+ T cells.
SFV infection of BMDCs and fibroblasts. BMDCs were incubated for 45 min at 37°C in a 5% CO2 incubator with the indicated SFV particles at a multiplicity of infection of 50 in serum-free minimum essential medium (MEM) supplemented with glutamine and Pen/Strep (Sigma), with or without 12.5% autoclaved polyethylene glycol 3000 (PEG 3000; WWR International AB, Stockholm, Sweden). After this incubation, the cells were washed with RP-10 and incubated further. FACS analysis and sorting of SFV-EGFP-OVA-expressing DCs was done at 9 h postinfection. The populations that were GFP positive and GFP negative were collected separately in RP-10 and used immediately in the OT-I assay. BLK CL.4 fibroblasts were infected with SFV-OVA or SFV-EGFP-OVA at a multiplicity of infection of 10 as described above. Free SFV particles were removed from infected cells by a brief treatment with trypsin and succinic acid (20 mM) followed by three phosphate-buffered saline (PBS) washes and then incubated for 2 h before in vivo transfer or for the indicated times for each experiment. This treatment did not affect cell viability.
DC and fibroblast coculture experiment. BMDCs were cocultured with BLK CL.4 fibroblasts that had been infected 8 or 24 h earlier at a ratio of 3:1 (DCs to fibroblasts). After overnight coculturing, the cells were subjected to FACS analysis or were tested in OT-I T-cell assays.
FACS analysis. Cells were blocked with anti-CD16/CD32 and stained with the indicated antibodies. Anti-CD11c (clone HL3)-phycoerythrin (PE), anti-MHC II (clone NIMR-4), anti-MHC I (clone AF6-88.5)-PE or -fluorescein isothiocyanate, anti-CD86 (clone GL1), anti-CD40 (clone 3/23)-biotin, and streptavidin-PE were all obtained from BD Biosciences. Anti-F480-PE (Serotec Ltd. Scandinavia, Hamar, Norway) and anti-rat immunoglobulin G-PE (Jackson Research Laboratories) were also used. Cells were fixed in PBS containing 1% paraformaldehyde and then run on a FACScan instrument. Data were analyzed with Cell Quest Pro software (BD Biosciences).
OT-I ELISPOT assays. Enzyme-linked immunospot (ELISPOT) immunoprecipitation plates (Millipore Co., Bedford, Mass.) were coated with anti-gamma interferon (anti-IFN-; MabTech, Stockholm, Sweden) or anti-interleukin-2 (anti-IL-2; BioSite, T?by, Sweden) antibodies according to the manufacturer's recommendations. After being washed, the plates were blocked for 1 h with RP-5. The indicated numbers of stimulated DCs were incubated with 3 x 104 purified OT-I T cells in the plates, as indicated, with or without 5 μg of CD40 agonist (HM40-3; BD Biosciences)/ml. The plates were then incubated for 24 h for the IL-2 assay and 48 h for the IFN- assay and were washed before the addition of a biotinylated anti-IFN- antibody (MabTech). The spots were developed by use of an avidin-peroxidase complex kit (Vector, Burlingame, Calif.) and the AEC substrate (Sigma) and were counted with an automated ELISPOT reader (Axioplan 2 Imaging; Zeiss, G?ttingen, Germany).
Immunization. Immunizations were performed on days 0 and 14, and T-cell responses were measured on day 24. Immunizations with SFV-OVA-infected fibroblasts (106 cells) or SFV-OVA particles (106 IU) were administered intraperitoneally. The control groups were either given PBS or immunized subcutaneously with 100 μg of the SIINFEKL peptide in incomplete Freund adjuvant (IFA; Sigma).
Antigen restimulation and cytokine ELISA. Spleens were mashed through a 100-μm-pore-size cell strainer (BD Biosciences). After lysis of the red blood cells in RBC lysis buffer (Sigma), the cells were resuspended in RP-5, adjusted to the required concentration, and restimulated with RP-10 alone or RP-10 supplemented with either the SIINFEKL peptide (2 μg/ml) or concanavalin A (ConA) (4 μg/ml). Supernatants collected at 48 h were analyzed by use of an enzyme-linked immunosorbent assay (ELISA) for IFN-. Cytokine ELISA detection reagents were purchased from BD Biosciences (IFN-, IL-12 p70, and tumor necrosis factor alpha [TNF-]), Endogen (IL-1?), and MBL Japan (IL-18). The detection limits for these cytokines were 16, 32, 50, 16, and 26 pg/ml, respectively.
CTLs. (i) Bulk CTL cultures. A cell suspension containing four-fifths of the total number of splenocytes was incubated in RP-5 supplemented with 2 μg of the SIINFEKL peptide/ml for 5 days. On day 5, effector cells were washed in complete medium and used in the 51Cr release assay described below.
(ii) Target cells and 51Cr release assay. Effector cells were diluted twofold in 96-well U-bottomed plates to give dilutions containing 250,000 to 15,000 effector cells per well. A total of 5,000 51Cr-labeled EL4 and P815 target cells in RP-5, with or without the SIINFEKL peptide, were then added to the effector cells and incubated at 37°C for 5 h. The supernatants were harvested, and the amounts of 51Cr released were measured. Spontaneous and total chromium releases were estimated for wells in which the target cells were incubated with RP-5 alone or with 1% Triton X-100, respectively. The percentage of specific lysis was calculated as follows: % lysis = [(sample release – spontaneous release)/(total release – spontaneous release)] x 100.
RESULTS
SFV-infected DCs undergo MyD88-independent maturation. To determine whether MyD88 expression is required for DCs to undergo maturation following exposure to a virus, we infected bone marrow-derived wild-type (WT) or MyD88 knockout (KO) DCs with recombinant SFV particles. These DC preparations consisted of >95% CD11c+ cells (Fig. 1A). The recombinant virus (SFV-EGFP-OVA) contained a double-cistronic RNA molecule that encoded both GFP and the intracellular form of OVA. We found in initial infection experiments that the percentage of infected DCs was low even at a high multiplicity of infection (Fig. 1B). The results indicated that the DCs were refractory to infection under these conditions, but they did not rule out the possibility that replication may occur following viral entry. To overcome the block to infection, we added PEG 3000, a highly hydrated polymer that favors membrane-to-membrane contacts (33), during infection. We observed that this pretreatment increased the percentage of infected cells (GFP-positive cells), from approximately 1% to 20%, as measured at 8 to 10 h postinfection (p.i.). We noted that these percentages decreased to <0.1 and 5%, respectively, at 24 h p.i. (Fig. 1B and data not shown), most probably due to the cytopathic effect caused by SFV replication, which is typically observed after SFV infection (54). We found that virus-treated DCs, irrespective of their MyD88 genotype, showed no difference in their susceptibilities to infection, as measured by the ability to express SFV-encoded GFP (Fig. 1B). Interestingly, when the cells were stained with antibodies for costimulatory molecules at 9 h p.i., we found that the GFP-positive DC population (GFP-pos) had upregulated costimulatory molecules such as CD86 and MHC class II compared to the GFP-negative (GFP-neg) population in the same culture (Fig. 2A). For these experiments, the fluorescence intensity of stained mock-treated controls (undergoing the same treatment as other groups, but with the virus omitted) was set as the baseline level (value of 1). The results suggested that the virus infection had initiated a maturation process in virally targeted DCs. This was observed irrespective of the MyD88 genotype, suggesting that SFV-EGFP-OVA-induced DC maturation is independent of MyD88 (Fig. 2A). An analysis of the whole SFV-treated DC population, which contained mostly noninfected DCs, did not demonstrate the same level of upregulation (Fig. 2B to E). This was partly due to the large proportion of noninfected DCs (approximately 80%). We chose to perform this analysis at an early time point since infection led to a gradual loss of GFP-expressing DCs due to cell death. At later time points (24 h p.i.), we did indeed note an increase in mature DCs (data not shown), suggesting that the exposure to dying infected DCs may have led to the maturation of bystander DCs.
While the control culture stimulated with CpG showed MyD88-dependent maturation (24), MyD88-independent maturation by CD40 cross-linking, which is mediated via the tumor necrosis factor receptor (TNF-R) pathway, was observed (55). The addition of inactivated virus or PEG alone had no significant effect on the maturation process (Fig. 2B to E and data not shown).
SFV-infected DCs are capable of activating antigen-specific T cells. To investigate whether the SFV-EGFP-OVA-infected DCs were capable of activating antigen-specific CD8+ T cells, we cocultured them with na?ve OT-I T cells specific for a Kb-restricted OVA peptide (SIINFEKL) (25). Bulk infected DCs demonstrated a poor capability of activating these T cells, probably because the cultures contained only minute numbers of infected DCs. However, when the DCs had been infected in the presence of PEG, a significant increase in OT-I T-cell activation was observed (Fig. 3A). We noted that virus-infected MyD88 KO DC cultures failed to show the same stimulation capacity as WT cultures. Moreover, stimulation with soluble OVA (20 μg/ml) did not lead to the activation of OT-I cells, suggesting that the result was not due to contamination of the soluble OVA protein in the virus preparations.
Since the virus-infected cultures contained a large proportion of DCs that had not been infected, we could not exclude the possibility that the OT-I activation was mediated by uninfected DCs via cross-presentation after they acquired antigenic material from dying cells. In order to determine if this was the case, we sorted the OVA-expressing DCs by FACS, with gating for GFP expression (9 h p.i.), and then immediately cultured the sorted cells with OT-I T cells. By using this enrichment, we found that the virus-targeted KO DCs showed a similar efficiency as WT DCs for the direct stimulation of OT-I T cells (Fig. 3B). Control wells stimulated with peptide-loaded DCs, with or without PEG treatment, showed that both WT and KO DCs were equally proficient at activating OT-I cells (Fig. 3C). Collectively, these results suggest that the SFV-EGFP-OVA-expressing DCs were capable of mediating direct priming of OVA-specific CD8+ T cells in a MyD88-independent manner. The results also suggested that uninfected bystander DCs may account for certain OT-I activation (Fig. 3B), which may be due to cross-presentation following the uptake of infected DCs.
Exposure to SFV-infected fibroblasts induces DC maturation. In order to examine how DCs respond to SFV-infected cells, we analyzed cocultures consisting of DCs that had been cultured overnight with SFV-EGFP-OVA-infected BLK CL.4 fibroblasts (H-2b). These fibroblasts were washed to remove any excess virus and then incubated for 8 or 24 h to allow antigen expression before DCs were added. While mock-infected fibroblasts remained healthy and adherent, the infected fibroblasts gradually became positive for annexin V and propidium iodide (PI) staining and lost their adherence capacity (Fig. 4A). No detectable amounts of IL-1?, IL-12, IL-18, or TNF- were found in the supernatants from these fibroblast cultures at any time point (data not shown). After overnight coculturing with infected fibroblasts, both WT and MyD88-defective DCs had CD86 and CD40 expression levels that were above those in cells that were cocultured with or without uninfected cells (Fig. 4B). The mean fluorescence intensity (MFI) of the latter control was set as the baseline level (value of 1). However, we observed that only the WT DCs that had been cultured with infected fibroblasts increased the levels of IL-12 production, whereas DCs with defective MyD88 expression did not (Fig. 5A).
MyD88–/– DCs are ineffective at cross-presenting OVA derived from SFV-OVA-infected cells. To assess the efficiency of DCs at cross-presenting cell-associated OVA, we cultured OT-I T cells with DCs that had been cocultured overnight with SFV-OVA-infected fibroblasts. Activated OT-I T cells were enumerated by IL-2- or IFN--specific ELISPOT assays. Our results showed that WT DCs cocultured with infected fibroblasts were fully capable of activating OT-I T cells (Fig. 5B), suggesting that these DCs had taken up and processed antigens derived from virus-infected fibroblasts and could present the processed peptide SIINFEKL under costimulatory conditions. In contrast, MyD88 KO DCs showed a poor capacity to activate OT-I cells under the same conditions. However, this capacity was restored almost fully by a treatment with a CD40 antagonist (HM40-3) during culturing with OT-I T cells (Fig. 5B). Because BLK CL.4 fibroblasts express no detectable CD40 (data not shown), they are unlikely to be subjected to CD40 cross-linking, and therefore the observed effect was most probably due to the CD40-activated DCs (33). This result confirmed the previous observation (Fig. 3A) that uninfected MyD88 KO DCs are inefficient at cross-presenting antigens derived from infected cells. Moreover, we observed that the cross-primed OT-Is produced more IL-2 than IFN-. The reason for this remained to be investigated.
MyD88 knockout mice fail to generate efficient CTL responses to SFV-OVA-infected cells. Since the in vitro results indicated that the MyD88 signaling pathway plays a role during the cross-presentation of cell-associated OVA, we next investigated whether this affected the in vivo response. Accordingly, wild-type or MyD88-deficient mice were immunized with ex vivo SFV-OVA-infected fibroblasts (106 cells/dose), and the OVA-specific MHC class I restricted T-cell response was analyzed. For comparison, we included a group that received an equal amount of SFV-OVA virus particles, as these virus vectors generally induce efficient CTL responses in vivo (36, 54). A SIINFEKL peptide-IFA emulsion was used as a control for the MyD88-independent response. We found that WT and MyD88 KO splenocytes were equally efficient at producing IFN- after stimulation with the mitogen ConA, suggesting that the KO mice had no impairment in the mitogenic IFN- response. Following stimulation with the SIINFEKL peptide, an antigen-specific IFN- response was observed for WT mice infected with the SFV-OVA virus. This was also observed for the WT group that had received SFV-OVA-infected cells (Fig. 6A), indicating that the infected cells had efficiently cross-primed OVA-specific MHC class I-restricted T cells. However, this was not observed for MyD88 knockouts that had been immunized with either infected cells or the SFV-OVA virus, despite a normal response to the ConA mitogen. These results were corroborated by a T-cell cytotoxicity assay, with a total of six independent experiments. Whereas MyD88 knockout animals failed to demonstrate efficient CTL activity, wild-type animals repeatedly demonstrated efficient CTL lysis of SIINFEKL-pulsed EL-4 target cells (Fig. 6B). This defect was not observed for MyD88 KO mice that had been immunized with peptide-IFA, an immunization schedule that circumvents TLR usage. Although the peptide-IFA emulsion elicited peptide-specific CTLs, we did not detect an IFN- response in WT or KO mice. This may have been related to the differences in assay sensitivities (IFN- ELISA versus CTL assay) and lengths of peptide restimulation (2 days versus 5 days).
DISCUSSION
In this study, we showed that MyD88-deficient DCs infected by the SFV-EGFP-OVA virus are capable of directly activating na?ve antigen-specific CD8 T cells (OT-I). In contrast, we found that efficient cross-presentation of SFV-derived OVA acquired from infected cells requires DCs that have intact MyD88 signaling. The latter observation parallels with results obtained from our in vivo experiment, in which immunization with cells infected with the SFV-OVA virus failed to elicit OVA-specific CD8+-T-cell responses in MyD88 knockout mice. This indicates that the efficient cross-priming mediated by the uptake of SFV-infected cellular material in vivo is predominantly MyD88 dependent. It also suggests that the immunogenicity of the SFV particles relies mainly on cross-presentation and cross-priming, as we observed that the SFV-OVA-elicited CTL responses also required MyD88. The fact that MyD88 knockout mice were capable of responding to peptide-IFA immunization as well as mitogen ConA stimulation suggests that they have no significant intrinsic defect in their CD8+-T-cell functions, an observation which is in line with other reports (44, 56).
Given that the virus-infected fibroblasts used were free from infectious virus particles, the in vivo T-cell priming observed for this study was unlikely to be mediated by infected professional APCs, but rather by APCs that had acquired infected cells. This was supported by the observation that SFV targeting of DCs in vitro is inefficient, even in the presence of a highly hydrated PEG polymer. Moreover, several lines of observations have shown that alphavirus replication causes a rapid cytopathic effect on infected cells that leads to apoptotic or necrotic cell death (18, 20, 58). This indicates that immunity-mediated killing is not a necessity in order to obtain cell-associated antigens and therefore differs from other live-antigen delivery systems that do not induce rapid cell death. Our results thus differ from those reported for plasmid vaccines that remained immunogenic in mice with deficiencies in MyD88 or TLR9 (49). Interestingly, it was recently reported that MyD88-deficient DCs have a defect in cross-presentation in a mycobacterial Hsp65 fusion protein model which is independent of TLR4 (41). The explanation underlying the discrepancy in these observations most likely relates to the difference in the immunogens' abilities to engage TLRs, to induce direct versus cross-priming, or to gain compensatory CD4 help (2, 24, 30, 53). In general, antigens that persist for an extended time are more likely to acquire CD4 help, a scenario that perhaps also applies to plasmid-based immunogens that can express antigens in vivo for weeks in transfected cells (57). Similarly, this may also apply to wild-type live viruses that are replication proficient, which in most cases can result in high viral loads in vivo for an extended time. Thus, these potentials may partly override TLR signaling pathways. This is a central aspect that distinguishes the SFV replicon system from other immunogens. Because of this, its immunogenicity is likely to be more vulnerable to the lack of innate stimulation, such as that provided by TLR activation.
Several viruses, including vaccinia virus, measles virus, human immunodeficiency virus, Epstein-Barr virus, and CMV, have the ability to infect DCs (28). Infection can suppress the ability of DCs to mature and can hinder the activation of T cells, as these viruses have evolved strategies to prevent virally encoded antigens from being processed and loaded onto MHC class I and II molecules, thus hampering efficient antigen presentation to virus-specific lymphocytes (15, 38, 47, 50). Our results differ from those of previous reports in the sense that SFV-targeted DCs showed no signs of down-regulating costimulatory molecules or the ability to present endogenously expressed antigens to na?ve T cells during the antigen-expressing phase. In addition, these functions did not appear to depend on the MyD88 pathway, which supports the results of a recent study demonstrating that infection by a DC-tropic RNA virus is capable of switching to conventional nonplasmacytoid DCs in a TLR-independent manner (12). In this regard, DC maturation mediated by intracellular dsRNA pattern recognizers such as protein kinase R would be an alternative way for DCs to sense and act against cytosolic viral pathogens without engaging TLR3. Interestingly, the biological role of TLR3 was recently questioned by Edelmann and colleagues (13), who found that TLR3 is not involved in generating adaptive antiviral immunity to MCMV, vesicular stomatitis virus, lymphocytic choriomeningitis virus, or reovirus. Considering that this last study used several viruses that have the capacity to infect DCs (8, 35, 46), it would be interesting to know whether or not the result was partly related to a direct viral targeting of DCs, which is one possibility for overcoming the need of TLR signaling.
In terms of cross-presentation of cell-associated antigens, an important issue that remains to be studied is the role of mRNA and viral RNA in the physiological process of cross-presentation of cell-associated antigens by APCs, in particular the CD8+ DC subset. Interestingly, this DC subset has been shown to constitutively cross-present antigens (10, 45) and was found to be the only DC subset that has high TLR3 expression (14). Other viral PAMPs, including viral ssRNA (a TLR7 ligand) and host proteins associated with tissue damage, such as the heat shock proteins, fibrinogen, fibronectin, hyaluronic acid, and heparan sulfate (TLR4 ligands), have also been suggested as potent stimulators of APCs (11, 52). Although several MyD88-independent TLR signaling pathways exist, the data shown here stress the importance of MyD88 signaling in the innate and adaptive recognition of viral antigens derived from cells infected by an RNA virus. Our results are in line with recent reports which demonstrated that cross-presentation enhanced by TLR ligands such as poly(I:C) is MyD88 dependent (9) and that a genetically modified heat shock protein (Hsp95) targeted to the cell surface can act as an endogenous signal that spontaneously activates TLR4 via a MyD88-dependent mechanism leading to immune reactions (34). Furthermore, it is known that IL-1 and IL-18 also share MyD88 adaptor usage (52). However, it is likely that these cytokines had no significant role in this study, since no detectable amount of IL-1? (the potent form of IL-1) or IL-18 was found in the supernatants from virally infected fibroblast cultures.
In conclusion, our work has presented evidence that innate signaling can play a crucial role during the elicitation of antiviral immunity and has implicated cross-priming as a main mechanism by which immunity to an alphavirus replicon is generated. Future work will need to investigate whether this also affects the efficiency of T-helper and B-cell responses upon encounters with this type of immunogen.
ACKNOWLEDGMENTS
We thank S. Akira for MyD88–/– mice, M.-J. Wick for OT-I/Rag-1 mice, J.-P. Kraehenbuhl for the OVA plasmid, and Marianne Diurson and Linda Kostic for technical assistance.
This study was supported by the Swedish Research Council and the European Union 5th Framework Programme.
M.C. and C.B. contributed equally to this work.
Present address: Virology Unit, Trinity College, Dublin, Ireland.
REFERENCES
Adachi, O., T. Kawai, K. Takeda, M. Matsumoto, H. Tsutsui, M. Sakagami, K. Nakanishi, and S. Akira. 1998. Targeted disruption of the MyD88 gene results in loss of IL-1- and IL-18-mediated function. Immunity 91:143-150.
Akbari, O., N. Panjwani, S. Garcia, R. Tascon, D. Lowrie, and B. Stockinger. 1999. DNA vaccination: transfection and activation of dendritic cells as key events for immunity. J. Exp. Med. 189:169-178.
Albert, M. L., B. Sauter, and N. Bhardwaj. 1998. Dendritic cells acquire antigen from apoptotic cells and induce class I-restricted CTLs. Nature 392:86-89.
Alexopoulou, L., A. C. Holt, R. Medzhitov, and R. A. Flavell. 2001. Recognition of double-stranded RNA and activation of NF-kappaB by Toll-like receptor 3. Nature 413:732-738.
Asea, A., M. Rehli, E. Kabingu, J. A. Boch, O. Bare, P. E. Auron, M. A. Stevenson, and S. K. Calderwood. 2002. Novel signal transduction pathway utilized by extracellular HSP70: role of Toll-like receptor (TLR) 2 and TLR4. J. Biol. Chem. 277:15028-15034.
Bevan, M. J. 1976. Cross-priming for a secondary cytotoxic response to minor H antigens with H-2 congenic cells which do not cross-react in the cytotoxic assay. J. Exp. Med. 143:1283-1288.
Boyle, J. S., C. Koniaras, and A. M. Lew. 1997. Influence of cellular location of expressed antigen on the efficacy of DNA vaccination: cytotoxic T lymphocyte and antibody responses are suboptimal when antigen is cytoplasmic after intramuscular DNA immunization. Int. Immunol. 12:1897-1906.
Dalod, M., T. Hamilton, R. Salomon, T. P. Salazar-Mather, S. C. Henry, J. D. Hamilton, and C. A. Biron. 2003. Dendritic cell responses to early murine cytomegalovirus infection: subset functional specialization and differential regulation by interferon alpha/beta. J. Exp. Med. 197:885-898.
Datta, S. K., V. Redecke, K. R. Prilliman, K. Takabayashi, M. Corr, T. Tallant, J. DiDonato, R. Dziarski, S. Akira, S. P. Schoenberger, and E. Raz. 2003. A subset of Toll-like receptor ligands induces cross-presentation by bone marrow-derived dendritic cells. J. Immunol. 170:4102-4110.
den Haan, J. M., S. M. Lehar, and M. J. Bevan. 2000. CD8(+) but not CD8(–) dendritic cells cross-prime cytotoxic T cells in vivo. J. Exp. Med. 192:1685-1696.
Diebold, S. S., T. Kaisho, H. Hemmi, S. Akira, and C. Reis e Sousa. 2004. Innate antiviral responses by means of TLR7-mediated recognition of single-stranded RNA. Science 503:1529-1531.
Diebold, S. S., M. Montoya, H. Unger, L. Alexopoulou, P. Roy, L. E. Haswell, A. Al-Shamkhani, R. Flavell, P. Borrow, and C. Reis e Sousa. 2003. Viral infection switches non-plasmacytoid dendritic cells into high interferon producers. Nature 424:324-328.
Edelmann, K. H., S. Richardson-Burns, L. Alexopoulou, K. L. Tyler, R. A. Flavell, and M. B. Oldstone. 2004. Does Toll-like receptor 3 play a biological role in virus infections? Virology 322:231-238.
Edwards, A. D., S. S. Diebold, E. M. Slack, H. Tomizawa, H. Hemmi, T. Kaisho, S. Akira, and C. Reis e Sousa. 2003. Toll-like receptor expression in murine DC subsets: lack of TLR7 expression by CD8 alpha+ DC correlates with unresponsiveness to imidazoquinolines. Eur. J. Immunol. 33:827-833.
Engelmayer, J., M. Larsson, M. Subklewe, A. Chahroudi, W. I. Cox, R. M. Steinman, and N. Bhardwaj. 1999. Vaccinia virus inhibits the maturation of human dendritic cells: a novel mechanism of immune evasion. J. Immunol. 163:6762-6768.
Feng, H., Y. Zeng, M. W. Graner, and E. Katsanis. 2002. Stressed apoptotic tumor cells stimulate dendritic cells and induce specific cytotoxic T cells. Blood 100:4108-4115.
Gallucci, S., and P. Matzinger. 2001. Danger signals: SOS to the immune system. Curr. Opin. Immunol. 13:114-119.
Glasgow, G. M., M. M. McGee, C. J. Tarbatt, D. A. Mooney, B. J. Sheahan, and G. J. Atkins. 1998. The Semliki Forest virus vector induces p53-independent apoptosis. J. Gen. Virol. 79:2405-2410.
Glotzer, J. B., M. Saltik, S. Chiocca, A. I. Michou, P. Moseley, and M. Cotten. 2000. Activation of heat-shock response by an adenovirus is essential for virus replication. Nature 407:207-211.
Grandgirard, D., E. Studer, L. Monney, T. Belser, I. Fellay, C. Borner, and M. R. Michel. 1998. Alphaviruses induce apoptosis in Bcl-2-overexpressing cells: evidence for a caspase-mediated, proteolytic inactivation of Bcl-2. EMBO J. 17:1268-1278.
Guermonprez, P., J. Valladeau, L. Zitvogel, C. Thery, and S. Amigorena. 2002. Antigen presentation and T cell stimulation by dendritic cells. Annu. Rev. Immunol. 20:621-667.
Heath, W. R., and F. R. Carbone. 2001. Cross-presentation, dendritic cells, tolerance and immunity. Annu. Rev. Immunol. 19:47-64.
Heil, F., H. Hemmi, H. Hochrein, F. Ampenberger, C. Kirschning, S. Akira, G. Lipford, H. Wagner, and S. Bauer. 2004. Species-specific recognition of single-stranded RNA via Toll-like receptor 7 and 8. Science 303:1526-1529.
Hemmi, H., O. Takeuchi, T. Kawai, T. Kaisho, S. Sato, H. Sanjo, M. Matsumoto, K. Hoshino, H. Wagner, K. Takeda, and S. Akira. 2000. A Toll-like receptor recognizes bacterial DNA. Nature 408:740-745.
Hogquist, K. A., S. C. Jameson, W. R. Heath, J. L. Howard, M. J. Bevan, and F. R. Carbone. 1999. T cell receptor antagonist peptides induce positive selection. Cell 76:17-27.
Houde, M., S. Bertholet, E. Gagnon, S. Brunet, G. Goyette, A. Laplante, M. F. Princiotta, P. Thibault, D. Sacks, and M. Desjardins. 2003. Phagosomes are competent organelles for antigen cross-presentation. Nature 425:402-406.
Humrich, J., and L. Jenne. 2003. Viral vectors for dendritic cell-based immunotherapy. Curr. Top. Microbiol. Immunol. 276:241-259.
Kaiserlian, D., and B. Dubois. 2001. Dendritic cells and viral immunity: friends or foes? Semin. Immunol. 13:303-310.
Kaisho, T., K. Hoshino, T. Iwabe, O. Takeuchi, T. Yasui, and S. Akira. 2002. Endotoxin can induce MyD88-deficient dendritic cells to support T(h)2 cell differentiation. Int. Immunol. 7:695-700.
Kumaraguru, U., R. J. Rouse, S. K. Nair, B. D. Bruce, and B. T. Rouse. 2000. Involvement of an ATP-dependent peptide chaperone in cross-presentation after DNA immunization. J. Immunol. 165:750-759.
Larsson, M., J. F. Fonteneau, and N. Bhardwaj. 2003. Cross-presentation of cell-associated antigens by dendritic cells. Curr. Top. Microbiol. Immunol. 276:261-275.
Leitner, W. W., H. Ying, and N. P. Restifo. 1999. DNA and RNA-based vaccines: principles, progress and prospects. Vaccine 18:765-777.
Lentz, B. R., and J. K. Lee. 1999. Poly(ethylene glycol) (PEG)-mediated fusion between pure lipid bilayers: a mechanism in common with viral fusion and secretory vesicle release? Mol. Membr. Biol. 16:279-296.
Liu, B., J. Dai, H. Zheng, D. Stoilova, S. Sun, and Z. Li. 2003. Cell surface expression of an endoplasmic reticulum resident heat shock protein gp96 triggers MyD88-dependent systemic autoimmune diseases. Proc. Natl. Acad. Sci. USA 100:15824-15829.
Ludewig, B., K. J. Maloy, C. Lopez-Macias, B. Odermatt, H. Hengartner, and R. M. Zinkernagel. 2000. Induction of optimal anti-viral neutralizing B cell responses by dendritic cells requires transport and release of virus particles in secondary lymphoid organs. Eur. J. Immunol. 30:185-196.
Lundstr?m, K. 2003. Alphavirus vectors for vaccine production and gene therapy. Expert Rev. Vaccines 3:447-459.
Lutz, M. B., N. Kukutsch, A. L. Ogilvie, S. Rossner, F. Koch, N. Romani, and G. Schuler. 1999. An advanced culture method for generating large quantities of highly pure dendritic cells from mouse bone marrow. J. Immunol. Methods 223:77-92.
Mathys, S., T. Schroeder, J. Ellwart, U. H. Koszinowski, M. Messerle, and U. Just. 2003. Dendritic cells under influence of mouse cytomegalovirus have a physiologic dual role: to initiate and to restrict T cell activation. J. Infect. Dis. 187:988-999.
Melief, C. J. 2003. Regulation of cytotoxic T lymphocyte responses by dendritic cells: peaceful coexistence of cross-priming and direct priming? Eur. J. Immunol. 33:2645-2654.
Norbury, C. C., and L. J. Sigal. 2003. Cross priming or direct priming: is that really the question? Curr. Opin. Immunol. 15:82-88.
Palliser, D., H. Ploegh, and M. Boes. 2004. Myeloid differentiation factor 88 is required for cross-priming in vivo. J. Immunol. 172:3415-3421.
Rodriguez, A., A. S. Regnault, M. Kleijmeer, P. Ricciardi-Castagnoli, and S. Amigorena. 1999. Selective transport of internalized antigens to the cytosol for MHC class I presentation in dendritic cells. Nat. Cell Biol. 1:362-368.
Scanga, C. A., J. Aliberti, D. Jankovic, F. Tilloy, S. Bennouna, E. Y. Denkers, R. Medzhitov, and A. Sher. 2002. Cutting edge: MyD88 is required for resistance to Toxoplasma gondii infection and regulates parasite-induced IL-12 production by dendritic cells. J. Immunol. 168:5997-6001.
Schnare, M., G. M. Barton, A. C. Holt, K. Takeda, S. Akira, and R. Medzhitov. 2001. Toll-like receptors control activation of adaptive immune responses. Nat. Immunol. 2:947-950.
Schulz, O., and C. Reis e Sousa. 2002. Cross-presentation of cell-associated antigens by CD8alpha+ dendritic cells is attributable to their ability to internalize dead cells. Immunology 107:183-189.
Sevilla, N., S. Kunz, A. Holz, H. Lewicki, D. Homann, H. Yamada, K. P. Campbell, J. C. de La Torre, and M. B. Oldstone. 2000. Immunosuppression and resultant viral persistence by specific viral targeting of dendritic cells. J. Exp. Med. 192:1249-1260.
Sevilla, N., S. Kunz, D. McGavern, and M. B. Oldstone. 2003. Infection of dendritic cells by lymphocytic choriomeningitis virus. Curr. Top. Microbiol. Immunol. 276:125-144.
Smerdou, C., and P. Liljestr?m. 1999. Two-helper RNA system for production of recombinant Semliki Forest virus particles. J. Virol. 73:1092-1098.
Spies, B., H. Hochrein, M. Vabulas, K. Huster, D. H. Busch, F. Schmitz, A. Heit, and H. Wagner. 2003. Vaccination with plasmid DNA activates dendritic cells via Toll-like receptor 9 (TLR9) but functions in TLR9-deficient mice. J. Immunol. 171:5908-5912.
Steinman, R. M., A. Granelli-Piperno, M. Pope, C. Trumpfheller, R. Ignatius, G. Arrode, P. Racz, and K. Tenner-Racz. 2003. The interaction of immunodeficiency viruses with dendritic cells. Curr. Top. Microbiol. Immunol. 176:1-30.
Tabeta, K., P. Georgel, E. Janssen, X. Du, K. Hoebe, K. Crozat, S. Mudd, L. Shamel, S. Sovath, J. Goode, L. Alexopoulou, R. A. Flavell, and B. Beutler. 2004. Toll-like receptors 9 and 3 as essential components of innate immune defense against mouse cytomegalovirus infection. Proc. Natl. Acad. Sci. USA 101:3516-3521.
Takeda, K., T. Kaisho, and S. Akira. 2003. Toll-like receptors. Annu. Rev. Immunol. 21:335-376.
Tsan, M. F., and B. Gao. 2004. Endogenous ligands of Toll-like receptors. J. Leukoc. Biol. 76:514-519.
Tubulekas, I., P. Berglund, M. Fleeton, and P. Liljestr?m. 1997. Alphavirus expression vectors and their use as recombinant vaccines: a minireview. Gene 190:191-195.
van Kooten, C., and J. Banchereau. 2000. CD40-CD40 ligand. J. Leukoc. Biol. 67:2-17.
Way, S. S., T. R. Kollmann, A. M. Hajjar, and C. B. Wilson. 2003. Protective cell-mediated immunity to Listeria monocytogenes in the absence of myeloid differentiation factor 88. J. Immunol. 171:533-537.
Wolff, J. A., R. W. Malone, P. Williams, W. Chong, G. Acsadi, A. Jani, and P. L. Felgner. 1990. Direct gene transfer into mouse muscle in vivo. Science 247:1465-1468.
Ying, H., T. Z. Zaks, R. F. Wang, K. R. Irvine, U. S. Kammula, F. M. Marincola, W. W. Leitner, and N. P. Restifo. 1999. Cancer therapy using a self-replicating RNA vaccine. Nat. Med. 5:823-827.
Zhang, D., G. Zhang, M. S. Hayden, M. B. Greenblatt, C. Bussey, R. A. Flavell, and S. Ghosh. 2004. A Toll-like receptor that prevents infection by uropathogenic bacteria. Science 303:1522-1526.(Margaret Chen, Christina )