Foot-and-mouth disease virus (FMDV) is a cytopathic virus that experimentally infects mice, inducing a thymus-independent neutralizing Ab response that rapidly clears the virus. In contrast, vaccination with UV-inactivated virus induces a typical thymus-dependent (TD) response. In this study we show that dendritic cells (DCs) are susceptible to infection with FMDV in vitro, although viral replication is abortive. Infected DCs down-regulate the expression of MHC class II and CD40 molecules and up-regulate the expression of CD11b. In addition, infected DCs exhibit morphological and functional changes toward a macrophage-like phenotype. FMDV-infected DCs fail to stimulate T cell proliferation in vitro and to boost an Ab response in vivo. Moreover, infection of DCs in vitro induces the secretion of IFN-γ and the suppressive cytokine IL-10 in cocultures of DCs and splenocytes. High quantities of these cytokines are also detected in the spleens of FMDV-infected mice, but not in the spleens of vaccinated mice. The peak secretion of IFN-γ and IL-10 is concurrent with the suppression of Con A-mediated proliferation of T cells obtained from the spleens of infected mice. Furthermore, the secretion of these cytokines correlates with the suppression of the response to OVA, a typical TD Ag. Thus, infection of DCs with FMDV induces suppression of TD responses without affecting the induction of a protective thymus-independent response. Later, T cell responses are restored, setting the stage for the development of a long-lasting protective immunity.
Foot-and-mouth disease (FMD)3 is a highly contagious disease of cloven-hoofed animals. It is the most economically important disease of livestock worldwide. FMD is caused by FMD virus (FMDV), a member of the genus Aphtovirus of the family Picornaviridae (1). Adult mice can be experimentally infected with FMDV (2). The rapid induction of neutralizing Abs is of outmost importance in the control of FMDV infection in mice (3, 4). By day 3 after infection, neutralizing Abs are detected in both euthymic and athymic mice (3), indicating that this primary Ab response is T cell independent (thymus independent (TI)). In contrast, vaccination of mice with inactivated virus requires T cell collaboration to induce an Ab response, indicating that inactivated FMDV is a thymus-dependent (TD) Ag (4, 5). The maintenance of persisting Ab titers is dependent on the presence of T cells in both infected and vaccinated mice. Although in athymic mice, Ab titers decline from the tenth day after infection, reaching nonprotective levels by day 21, euthymic mice infected with FMDV maintain high levels of neutralizing Abs for more than a year (3). Furthermore, the maintenance of the FMDV-Ab response in infected and vaccinated mice is dependent on the presence of MHC class II-restricted APCs (6). Thus, the Ab response against infectious FMDV is initiated as a TI response, but T cell collaboration is required for long-term maintenance of Ab titers in serum.
The Ab isotype profile induced in adult mice after infection with FMDV is characterized by a rapid appearance of neutralizing Abs of the IgM and IgG3 subclasses during the first week after infection. Then, as the immune response progresses, the IgG2a and IgG2b subclasses are elicited, with IgG2b the predominant isotype between 14 and 60 days after infection (7, 8). In contrast, vaccination with inactivated FMDV induces mainly IgG1 and IgG2a isotypes detected between 21 and 180 days after vaccination. The predominant isotype induced after vaccination depends on the adjuvant used for immunization. Yet the presence of IgG3, a typical isotype of TI responses, is restricted to infected mice (7). Thus, there are qualitative differences in the Ab response directed against FMDV depending on whether the virus used for immunization is infectious or inactivated.
Swine infection with FMDV results in T cell unresponsiveness during the first days after infection (9). Although the cause of the unresponsiveness is not known, it seems not to be caused by a direct effect of FMDV on T lymphocytes. Many viruses, such as poxvirus (10), measles (11, 12, 13), CMV (14), herpes simplex type 1 (15), and HIV (16), are known to target dendritic cells (DCs) and therefore impair T cell responses. As a result, an immunosuppressive state is achieved, favoring establishment of infection. These observations together with the differences in the type of immune response elicited after infection or vaccination with FMDV mentioned above led us to study the interactions of infectious and inactivated FMDV with murine DCs. These cells are important in the initiation of antiviral responses (17, 18) as well as in the modulation of the type of response elicited (19). Although the interaction between FMDV and APCs in the murine model has not been studied, it was reported that porcine macrophages (20, 21) and bovine Langerhans cells (22) can be infected with FMDV in vitro.
The data presented in this study indicate that murine DCs are susceptible to an abortive infection with FMDV. Infected DCs down-regulate the expression of Ag-presenting and costimulatory molecules and induce secretion of IL-10 and IFN-γ in cocultures of DCs and splenocytes. These events lead to suppression of TD responses. In contrast, UV-irradiated FMDV (UV-FMDV) improves the functionality of DCs, favoring the development of a typical TD response. Thus, infectious and inactivated FMDV trigger different pathways of activation of the immune system, explaining the differences reported in the early immune response in each case.
Materials and Methods
BALB/c male mice from Instituto Nacional de Tecnología Agropecuaria were used to provide DCs and for the immunization experiments. For the MLR, C57BL/6 male mice were used. Mice were used between 8 and 12 wk of age. Animal care was performed in accordance with institutional guidelines.
Generation of bone marrow-derived DCs
Bone marrow-derived DCs were obtained as previously described (23) with some minor modifications. Briefly, the epiphysis of femurs and tibiae of BALB/c mice were cut, and the marrows were flushed out with RPMI 1640 medium. RBC were lysed using 0.083% ammonium chloride. After washing, cells were suspended at a concentration of 1 × 106 cells/ml in RPMI 1640 medium (Invitrogen Life Technologies) supplemented with 10% FCS, 5.5 × 10−5 ME (Sigma-Aldrich), and 30% conditioned medium from GM-CSF-producing NIH-3T3 cells and cultured for 9 days. Every 2 days, 50% of the medium was aspirated and replaced with fresh medium containing GM-CSF. On day 9 of culture, >85% of the harvested cells expressed MHC class II, CD40, CD80, and CD11c, but no Gr-1 (data not shown).
Virus infection and LPS stimulation of DCs
FMDV serotype O1 Campos, provided by SENASA, was used throughout these studies. Infection of DCs was performed at a multiplicity of infection (moi) of 10 for 4 h at 37°C. Noninfectious, UV-FMDV was prepared by irradiation of the viral suspension with UV light. The FMDV suspension was placed into a plastic petri dish (depth, <5 mm). An 8-W UV lamp (254 nm) was positioned 10 cm from the virus suspension, and the suspension was irradiated for 7 min while being agitated every 2 min. This treatment yielded virus that was noninfectious in the FMDV-susceptible cell line BHK-21. DCs were incubated with UV-FMDV at an moi of 10 for 4 h at 37°C. Mock-infected (control) DCs were obtained by treatment with supernatant of uninfected BHK-21 cell cultures for 4 h at 37°C. After being subjected to any of these treatments, DCs were washed twice with PBS, pH 5.5 (1-min incubation), to inactivate noninternalized virus, followed by six washes with RPMI 1640 medium supplemented with 5% FCS.
Binary ethylenimine (BEI)-inactivated FMDV (BEI-FMDV) was prepared as previously described (24). All viral stocks were kept at −80°C and thawed immediately before use. In an additional set of experiments, mock-infected, UV-FMDV-loaded, or FMDV-infected DCs were stimulated for 48 h with 10 μg/ml LPS from Escherichia coli serotype O55:B5 (Sigma-Aldrich) in complete RPMI 1640 medium and then analyzed for MHC class II expression by flow cytometry.
Infection and vaccination of mice
Mice were infected or vaccinated with 105 50% tissue culture infectious doses (TCID50) of either infective or UV-inactivated FMDV O1 Campos, respectively, by the i.p. route. Vaccination with BEI-FMDV was performed by inoculation of the indicated amount of virus by the i.p. route. Mock-infected (control) mice were inoculated with supernatant of uninfected BHK-21 cell cultures. When indicated, mice were also vaccinated with either 1 μg of OVA (Sigma-Aldrich) or 10 μg of dextran 500 (Sigma-Aldrich) by the i.p. route.
Total RNA was extracted from 2 × 106 FMDV-infected and uninfected DCs and BHK-21 cells using TRIzol reagent (Invitrogen Life Technologies) according to the manufacturer’s instructions. The RT reaction contained 300 ng of total RNA and was performed using Moloney murine leukemia virus reverse transcriptase enzyme (Promega) following the manufacturer’s directions. Primers for the cDNA synthesis were FMDVRev (ACC ACT TCT GCG GGC GAG TC), which is an antisense primer for the positive strand FMDV RNA at position 3254–3273 of the FMDV genome, and primer FMDVFor1 (TTGAAGGAGGTAGGCAGCGTC), which primes the negative strand of FMDV RNA at position 3724–3744 of the genome. PCR was performed using primers FMDVFor1 and FMDVRev to amplify the positive as well as the negative strand. A GeneAmp PCR system (Applied Biosystems) was used with an initial denaturation step of 94°C for 5 min, followed by 35 cycles of 94°C for 1 min, 55°C for 1 min, and 72°C for 1.5 min and a final elongation step of 72°C for 10 min. PCR products were separated on a 1% agarose gel, stained with ethidium bromide, and visualized by a UV transilluminator.
Determination of cell viability and apoptosis
Viability and apoptosis in DCs were determined using the FITC-labeled annexin V kit (Immunotech) following the manufacturer’s instructions. Briefly, 5 × 105 cells were centrifuged and resuspended in 250 μl of binding buffer in the presence of 2 μl of FITC-annexin V. After a 10-min incubation on ice, 2.5 μl of propidium iodide was added, and cell samples were analyzed by two-color flow cytometry (FACScan flow cytometer; BD Biosciences) using CellQuest analysis software (BD Biosciences).
Quantitation of cellular apoptosis was also determined by fluorescence microscopy as previously described (25) using the fluorescent DNA-binding dyes acridine orange (100 μg/ml) to determine the percentage of cells that had undergone apoptosis and ethidium bromide (100 μg/ml) to differentiate between viable and nonviable cells. To assess the percentage of cells showing morphologic features of apoptosis, at least 200 cells were scored in each experiment.
Cells used as a positive control for apoptosis were incubated at 42°C for 10 min and then incubated for 24 h at 37°C. This procedure yields a high percentage of apoptotic cells.
To evaluate the expression of cell surface molecules, the following FITC-labeled Abs were used: anti-CD11c, anti-IAd (MHC class II), anti-CD40, anti-CD86, and anti-CD11b (BD Pharmingen). Cells were labeled as previously described (26). Briefly, single-cell suspensions (5 × 105 cells) were incubated with the indicated Abs diluted in PBS at 4°C for 30 min. Then, cells were washed twice with PBS. Analysis was performed using a FACScan flow cytometer and CellQuest software (BD Biosciences). The results are expressed as the mean fluorescence intensity (MFI).
Phagocytosis and endocytosis assays
To measure phagocytosis, cells were exposed to 10 μg/ml FITC-labeled zymosan particles for 1 h at 37°C in RPMI 1640 medium. To measure endocytosis, cells were exposed to 100 μg/ml FITC-labeled OVA for 1 h at 37°C in RPMI 1640 medium. After incubation, cells were washed three times with cold PBS, and uptake was evaluated by flow cytometry. The fluorescence background (cells incubated with FITC-OVA at 4°C for the endocytosis experiment or FITC-zymosan in the presence of cytochalasin B for the phagocytosis experiment) was subtracted. The results obtained are expressed as MFI values.
T lymphocyte proliferation assay
For the MLR, 2.5 × 105 splenocytes from naive C57BL/6 mice were cocultured in 96-well microplates with 5 × 104 DCs from BALB/c mice infected with FMDV (moi, 10), loaded with 10 moi of UV-FMDV, or mock infected. The cells were cocultured for 4 days in RPMI 1640 medium containing 10% FCS, 10 mM HEPES buffer, and 5.5 × 10−5 M ME (complete medium). On day 3 of culture, cells were pulsed for 18 h with [3H]thymidine (1 μCi/well; DuPont-New England Biolabs). Then cells were harvested using a semiautomatic cell harvester (Skatron Instruments), and the amount of [3H]thymidine incorporation was determined in a Wallac 1414 Winspectral beta scintillation counter (PerkinElmer). For the autologous Ag-specific proliferation assay, 2.5 × 105 splenocytes from the spleens of BALB/c mice previously vaccinated with two doses of 1 μg of BEI-FMDV at a 3-wk interval were cocultured with 5 × 104 DCs from BALB/c mice that were uninfected, infected with FMDV, or pulsed with UV-FMDV. Incorporation of [3H]thymidine was determined as described above. Unspecific T cell proliferation was measured by stimulation of 2.5 × 105 splenocytes with 5 μg/ml Con A (Sigma-Aldrich) for 96 h.
Cytokine concentrations were determined in the supernatants of an MLR at 48 h after the onset of the coculture or in the supernatant of DC cultures. Cytokine determinations were performed by a sandwich ELISA. Briefly, ELISA plates (Maxisorp) were coated with rat anti-mouse IL-2, IL-4, IL-10, or IFN-γ Abs (BD Pharmingen) diluted in carbonate-bicarbonate buffer (0.05 M; pH 9.6) and incubated overnight at 4°C. The plates were washed three times with PBS containing 0.05% Tween 20 (PBST) and blocked with PBS supplemented with 10% FBS for 1 h at room temperature. Culture supernatant and standards were added to the plates in duplicate and incubated for 2 h at 4°C. After washing three times as indicated, the corresponding biotinylated anti-cytokine Ab was added and incubated for 1 h at 37°C. The plates were washed and incubated with HRP-conjugated streptavidin for 1 h. After washing, 3,3′,5,5′-tetramethylbenzidine substrate was added. The absorbance at 450 nm was measured in a Multiskan EX spectrophotometer (Labsystems). Cytokine concentrations were calculated based on the ODs obtained with the standards.
Staining of adherent cells
FMDV-infected or uninfected cells were plated at a density of 1.5 × 105 cells/well of a 96-well plate and cultured at 37°C. After 48 h, nonadherent cells were removed, and the wells were washed twice with PBS. Then 0.1% crystal violet (Sigma-Aldrich) was added, and plates were incubated for 15 min at room temperature. The plates were then washed three times with water, and crystal violet was solubilized in 3% acetic acid. The level of adhesion was quantified by reading the OD in each well at 550 nm in a Multiskan EX spectrophotometer (Labsystems).
Adoptive transfer of DCs
Mice were vaccinated with 0.5 μg of BEI-FMDV. This dose of inactivated virus elicits a short-lived Ab response that starts to decay by day 30 after vaccination. At 60 days after vaccination, mice were selected according to their anti-FMDV Ab titer. Only those mice with a titer of ∼1.2 were used as recipient of DCs. A total of 5 × 105 DCs were mock infected, infected with FMDV, or pulsed with UV-FMDV for 4 h and after washing were transferred into mice by the i.p. route. The development of the secondary anti-FMDV Ab response was evaluated by ELISA 1 wk after transfer.
To measure total Ab titers against FMDV, a liquid phase ELISA test was used as previously described (27). Briefly, Immulon 2HB plates (Thermo Electron) were coated overnight at 4°C with rabbit anti-FMDV O1 Campos strain serum diluted to the optimum concentration in carbonate-bicarbonate buffer, pH 9.6. After washing the plates with PBST three times, the plates were blocked with PBST containing 1% BSA (blocking buffer) for 30 min at 37°C. Mouse serum samples were serially diluted in blocking buffer, and a fixed amount of BEI-FMDV was added. After 1-h incubation at 37°C with constant shaking, the virus-Ab mixtures were transferred to the blocked plates and incubated for 1 h at 37°C. After washing, an optimal dilution of guinea pig anti-FMDV serum diluted in PBS containing 2% normal bovine serum and 2% normal rabbit serum was added and incubated for 1 h at 37°C. Plates were washed, and peroxidase-labeled anti-guinea pig IgG (Kirkegaard & Perry Laboratories) diluted in the same buffer was added, followed by 1-h incubation at 37°C. O-phenylendiamine-H2O2 was used as the substrate for peroxidase, and absorbance at 490 nm was measured in a Multiskan EX spectrophotometer (Labsystems). Positive and negative control sera were included in each test. Ab titers were expressed as the negative logarithm of the highest dilution of serum that inhibits color development in >50% of the average value obtained in absence of serum. To measure anti-FMDV neutralizing Abs, sera were inactivated by incubation at 56°C for 30 min, diluted 1/25 in complete DMEM, and incubated with serial 10-fold dilutions of FMDV for 1 h at 37°C. The FMDV-serum mixture was transferred onto confluent BHK-21 cells and incubated for 1 h at 37°C. Then the mixtures were aspirated and replaced with fresh medium. The appearance of a cytopathic effect was recorded after 48 h of incubation at 37°C. To measure anti-OVA Abs, Immulon 2Hb plates were coated overnight at 4°C with a solution of OVA (10 μg/ml) in carbonate buffer. Blocking and subsequent steps were performed with PBST containing 0.5% gelatin (Sigma-Aldrich). Sera were added at a 1/50 dilution and incubated for 1 h at 37°C. Peroxidase-labeled anti-mouse IgG Abs (Kirkegaard & Perry Laboratories) were then added for 30 min at 37°C. O-phenylenediamine-H2O2 was used as substrate, and the absorbance was measured at 490 nm in a Multiskan EX spectrophotometer (Labsystems). To measure anti-dextran 500 Abs, Immulon 1B plates were coated overnight at 4°C with a solution of dextran 500 (50 μg/ml) in carbonate buffer. Blocking and subsequent steps were performed with PBST containing 0.5% gelatin (Sigma-Aldrich). Sera were added at a 1/50 dilution and incubated for 1 h at 37°C. Peroxidase-labeled anti-mouse IgM Abs (Kirkegaard & Perry Laboratories) were then added for 30 min at 37°C. O-phenylenediamine-H2O2 was used as a peroxidase substrate, and absorbances were measured at 490 nm in a Multiskan EX spectrophotometer (Labsystems).
Differences among mock infection, FMDV infection, and UV-FMDV vaccination were determined by one-way ANOVA, followed by post-ANOVA comparisons using the Bonferroni test. Analyses between treatments were performed using ANOVA for repeated measures with the Greenhouse and Geisser correction of the significance levels (fixed at 1%). A value of p < 0.05 was considered a significant difference.
FMDV abortively infects DCs, inducing differentiation toward a macrophage-like phenotype
DCs are the main controllers of immunity and are susceptible to infection with many viruses. Therefore, we studied the susceptibility of immature DCs to infection with FMDV in vitro. Viral replication was evaluated by RT-PCR amplification of the negative viral RNA strand, an intermediate product in FMDV replication. DCs were incubated with FMDV (moi, 10) and washed with acidic PBS to inactivate and remove noninternalized virus. Total RNA was extracted and used as a template for RT-PCR. The expected band of 490 bp was detected 5 and 24 h after infection (Fig. 1), indicating that viral RNA is replicated in DCs. Nevertheless, infectious virus could not be detected in the DC culture supernatants evaluated at 3, 6, 24, and 48 h after infection by two passages on susceptible BHK-21 cells. These results indicate that DCs are infected with FMDV, but viral replication is aborted.
FMDV is highly cytopathic in several cell lines (1). Therefore, apoptosis and viability of DCs after infection with FMDV were evaluated by annexin V staining and propidium iodide exclusion. Infection of DCs with FMDV did not decrease cell viability at either 48 or 72 h after infection (Fig. 2) compared with mock-infected or UV-FMDV-loaded DCs. These results were confirmed by fluorescence microscopy. In agreement with the results shown in Fig. 2, the percentage of apoptotic and necrotic cells detected by this technique was <5% in mock-infected, UV-FMDV-loaded, or FMDV-infected DCs (data not shown).
The immunophenotype of DCs after infection with FMDV was evaluated by flow cytometry. After 48 h of infection with FMDV, DCs exhibited a significant down-regulation of MHC class II and CD40 molecules, although the expression of CD86 was unaffected. In contrast, DCs incubated with UV-FMDV did not modify the expression of any of these molecules (Fig. 3, a and b), indicating that infectious virus is required to down-regulate the expression of MHC class II and CD40 molecules. However, FMDV-infected DCs (3 h at 37°C) stimulated for 48 h with LPS up-regulated the expression of MHC class II, CD40 and CD86 molecules in a fashion similar to mock-infected and UV-FMDV-loaded DCs (Fig. 3 c).
Changes in the immunophenotype of FMDV-infected DCs were associated with morphological changes in the cells. Infected DCs became plastic-adherent, large, elongated, macrophage-like cells (Fig. 4,a). In contrast, mock-infected DCs and UV-FMDV-loaded DCs remained mostly nonadherent, rounded cells forming small clusters as is typical of DCs. The increased adherence of FMDV-infected DCs was confirmed by staining plastic-adherent cells with crystal violet, followed by quantification of the intracellular dye in a spectrophotometer. Although FMDV-infected DCs remained attached to the plastic plate, mock-infected or UV-FMDV-loaded DCs were significantly less adherent (Fig. 4,b). Changes in cell surface markers were also indicative of differentiation of infected cells toward a macrophage-like phenotype. The macrophage marker CD11b was significantly up-regulated in FMDV-infected DCs compared with mock-infected DCs or UV-FMDV-loaded DCs (Fig. 3), whereas the DC marker CD11c was slightly down-regulated in FMDV-infected DCs. Finally, FMDV-infected DCs exhibited higher phagocytic and reduced endocytic activities, other features of macrophage-like phenotype (28). The phagocytosis of FITC-zymosan was significantly higher in FMDV-infected DCs than in mock-infected DCs (Fig. 4,c). In contrast, the endocytosis of FITC-OVA was significantly reduced in infected cells compared with mock-infected DCs (Fig. 4,d). Together, the data shown in Figs. 3 and 4 indicate that infection of DCs with FMDV induces changes in cell surface marker expression, cell morphology, and phagocytic and endocytic activities toward a macrophage-like phenotype.
FMDV-infected DCs suppress T cell proliferation
Immature DCs were either infected or loaded with inactivated FMDV in vitro to evaluate their ability to stimulate T cell proliferation in cocultures with autologous splenocytes from mice previously vaccinated with BEI-FMDV. FMDV-infected DCs did not stimulate the proliferation of FMDV-specific T cells. In contrast, UV-FMDV-loaded DCs stimulated a significant proliferative response (Fig. 5,a). The stimulatory ability of FMDV-infected DCs was also evaluated in an MLR. FMDV-infected DCs suppressed the MLR. On the contrary, UV-FMDV-loaded DCs exerted a higher stimulatory activity than mock-infected DCs, a stimulus comparable to that elicited by LPS-treated DCs (Fig. 5 a).
The series of experiments described above suggest that the immunophenotypical and morphological changes triggered by infection of DCs are involved in the suppression of T cell proliferation. To further examine the mechanisms involved in the suppression of T cell responses, the production of cytokines was analyzed in the supernatants of cocultures of DCs and splenocytes. FMDV-infected DCs induced the secretion of elevated amounts of IL-10 and IFN-γ. In contrast, UV-FMDV-loaded DCs did not induce IL-10 secretion, but induced lower amounts of IFN-γ (Fig. 5,b). Interestingly, IL-2 secretion was only detected in the cocultures conducted with UV-FMDV-loaded DCs (data not shown), in agreement with the higher proliferation observed in this experimental setting (Fig. 5 a). None of these cytokines was detected in cultures containing only FMDV-infected DCs or UV-FMDV-loaded DCs but not lymphocytes (data not shown). These data suggest the cytokines detected in the cocultures of DCs and splenocytes were not produced by DCs.
FMDV-infected DCs suppress T cell responses in vivo
Our data indicate that in vitro infection of DCs inhibits T cell proliferation. To determine whether infected DCs would suppress T cell responses in vivo, the induction of a secondary response by adoptively transferred DCs was studied. Mice were vaccinated with 1 μg of BEI-FMDV, and 60 days after vaccination, when the anti-FMDV Ab titers determined by ELISA decreased to ∼1.2, FMDV-infected DCs were transferred (Fig. 6, inset). The induction of a secondary Ab response was measured 1 wk after the transfer of DCs. Fig. 6 shows that FMDV-infected DCs did not induce a secondary response. In contrast, transfer of UV-FMDV-loaded DCs, significantly boosted an Ab response in the recipient mice (p < 0.01). These results suggest that adoptively transferred FMDV-infected DCs do not stimulate a Th secondary response in vivo.
The cytokine profile secreted by spleen cells from either FMDV-infected or UV-FMDV-vaccinated mice was then analyzed. Interestingly, and consistent with our in vitro observations, we found that the splenocytes of FMDV-infected mice evaluated 3 days after infection produced IL-10 and IFN-γ. In contrast, only IFN-γ was detected in the spleens of UV-FMDV-vaccinated mice (Fig. 7, a and b).
To study the impairment of T cell functionality after infection of mice, spleen cells were obtained at different times after infection and stimulated with the T cell mitogen Con A. There was a significant decrease in the proliferation of Con A-treated lymphocytes between 3 and 5 days after infection compared with lymphocytes from mock-infected animals. In contrast, lymphocytes from UV-FMDV-vaccinated mice exhibited a proliferative response comparable to that of mock-infected mice (Fig. 7 c). Thus, infection of mice with FMDV suppresses T cell proliferation.
Infection with FMDV impairs TD responses
To further confirm the suppression of T cell responses in vivo after infection with FMDV, we studied whether the response to OVA, a typical TD Ag, was affected in acutely infected mice. At the peak of viremia (24 h after infection), mice were immunized with OVA. Seven and 14 days after vaccination, anti-OVA IgG titers in serum were significantly lower in infected mice than in mock-infected controls (Fig. 8,a). In contrast, mice infected with FMDV and vaccinated 24 h later with dextran 500, a TI type 2 Ag, produced normal anti-dextran (Fig. 8,b) and anti-FMDV neutralizing responses (Fig. 8,c). These results indicate that although the TD response to OVA was suppressed by FMDV infection, the TI response against dextran 500 was unaffected. UV-FMDV-vaccinated mice produced an anti-OVA (Fig. 8,a) and an anti-dextran 500 (Fig. 8,b) Ab response comparable to that of mock-infected mice, indicating that vaccination with UV-FMDV does not impair TD responses. Consistent with published results (3, 4), the FMDV neutralizing Ab response in vaccinated mice developed more slowly and was of lower magnitude than that in infected mice (Fig. 8 c).
OVA-specific T cell proliferation was evaluated by stimulating splenocytes from mice at 14 days after vaccination. Lymphocytes from infected mice proliferated to a significant lesser extent upon OVA stimulation in vitro compared with lymphocytes from mock-infected or UV-FMDV-vaccinated mice (Fig. 8 d). Thus, immunization with this TD Ag at the peak of infection fails to induce a primary immune response due to the generalized T cell suppression. Together, these results suggest that the previously described TI protective neutralizing Ab response against infection with FMDV develops in a context of generalized suppression of T cell responses.
The immune response elicited after infection of mice with FMDV differs in many aspects from the immune response elicited after vaccination with inactivated virus. Although infection rapidly elicits high levels of neutralizing Abs that persist for the lifetime of the mice, vaccination induces a short-term memory (4). Moreover, the early Ab response after infection is TI, but the response to inactivated virus depends on T cell collaboration (4). In addition, vaccination with inactivated FMDV induces mainly IgG1 and IgG2a, whereas infection elicits IgM and IgG3 that are detected as early as 4 days after infection. Later in the response, IgG2b is the predominant isotype in infected mice (7, 8). In the present study we report an additional qualitative difference in the responses elicited against infectious and noninfectious FMDV; infection, but not vaccination, induces a DC-mediated suppression of T cell-dependent responses. FMDV abortively infects DCs inducing differentiation toward a macrophage-like phenotype, resulting in a failure in the development of TD responses during acute infection.
FMDV usually uses three αv integrins (αvβ1, αvβ3, and αvβ6) as cellular receptors (29, 30, 31). αvβ3 integrin is present on murine DCs (32). Moreover, preincubation of DCs with a peptide comprising aa 135–160 of viral protein 1 (VP1) of FMDV, containing the RGD integrin-binding motif, partially inhibited the uptake of FMDV (M. Ostrowski and O. J. Lopez, unpublished results). Thus, it is possible that FMDV initiates infection of DCs through interaction with one of these cellular integrins. However, other mechanisms of virus uptake by DCs cannot be ruled out. Whatever mechanism is used in the entry of infectious FMDV in DCs, RNA replication occurs, because it was demonstrated by RT-PCR amplification of the viral RNA negative strand (Fig. 1). Nevertheless, production of infectious virions is not detected. The inhibition of viral replication in DCs is a common feature of many viruses. For example, mature DCs block vaccinia (10), influenza (33), HIV (34), and vesicular stomatitis virus (17) replication. The main mechanism implicated in the abortion of infection in DCs is the induction of IFN-αβ (17, 35, 36). FMDV is highly sensitive to IFN-αβ (37). Thus, this mechanism could be responsible for the inhibition of FMDV replication in DCs.
After infection with FMDV, DCs differentiate toward a macrophage-like phenotype. This was demonstrated by morphological, immunophenotypical, and functional changes in the infected cells (Figs. 3 and 4). The conversion of DCs to a macrophage-like profile has been previously described for human monocyte-derived immature DCs upon removal of GM-CSF and IL-4 from the culture medium (38). In addition, we and others have previously reported that different stimuli, such as IL-6 (39), M-CSF (40, 41), vascular endothelial growth factor (41), IFN-γ (42), and antagonists of the angiotensin II receptor AT1 (28) switch the differentiation of monocytes from DCs toward a macrophage-like profile. As a result, these cells become poor stimulators of T cell proliferation. In this study we demonstrate that a viral infection can also trigger differentiation of DCs into a macrophage-like profile, with a subsequent decrease in their ability to stimulate T cell proliferation.
The failure of FMDV-infected DCs to stimulate T cell proliferation is associated with the production of IL-10 by cocultures of DCs and splenocytes (Fig. 5). The inhibition of T cell responses by IL-10 has been thoroughly characterized (43), and it has been reported in several viral infections. For example, coculture of T cells with immature HIV-infected DCs resulted in the secretion of IL-10 and subsequent suppression of allogeneic responses (44). The secretion of both IL-10 and IFN-γ by splenocytes cocultured with FMDV-infected DCs is consistent with the cytokine profile secreted by a subset of regulatory T cells detected in some infection models in both mice and humans (44, 45). Thus, FMDV infection might induce IL-10- and IFN-γ-secreting T cells and therefore inhibits T cell-dependent responses. Indeed, addition of the supernatant of the coculture of FMDV-infected DCs and T lymphocytes to an MLR suppressed T cell proliferation (M. Ostrowski and O. J. Lopez, unpublished results). The suppression of T cell responses by FMDV-infected DCs was also demonstrated in vivo. Adoptively transferred, FMDV-infected DCs fail to boost a secondary Ab response in vivo (Fig. 6). In addition, we demonstrate that during acute infection of mice with FMDV, Ab production to TD Ags is inhibited. Conversely, Ab production to TI Ags is not affected by infection with FMDV (Fig. 8). Finally, we show a generalized impairment of T cell functionality, as determined by Con A stimulation, between 3 and 5 days after infection (Fig. 7). This observation explains the suppression of TD responses in mice. The maximal impairment of Con A-stimulated proliferation is concurrent with the peak of secretion of IFN-γ and IL-10 by splenocytes of infected mice. These data are consistent with a previous report that demonstrates that FMDV infection of swine affects the ability of peripheral T lymphocytes to respond to Con A (9).
Based on the data contributed by several laboratories and the results shown in this study, it seems that the Ab response to infectious FMDV proceeds in two phases. In the early phase, which occurs within the first week after infection, a rapid TI production of neutralizing Abs is elicited. Splenic marginal zone B cells play an important role in TI Ab responses (46), and they might be responsible for this early anti-FMDV-neutralizing Ab response that clears the virus from the circulation. IgM and IgG3 are the predominant isotypes secreted by marginal zone B cells, and these are the isotypes of the neutralizing Abs that are detected early after infection of mice with FMDV (7). The secretion of IFN-γ and IL-10 by splenocytes stimulated by FMDV-infected DCs is also consistent with the production of IgG3 during this early stage, because these cytokines promote a switch toward this isotype (47, 48). These TI-neutralizing Abs are sufficient to clear the virus in 3 days after infection.
Six days after infection, a second phase of the anti-FMDV response develops. This late response depends on T cell collaboration (3, 4). The initial impairment of T cell functionality ceases (Fig. 7 c), and follicular B cells are activated in a TD fashion in the presence of higher local concentrations of Ags due to viral replication. The result of this T cell-dependent interaction will induce the secretion of neutralizing Abs for a long period of time. At this point, there is a switch in the isotypes of anti-FMDV Abs toward IgG2a and IgG2b (7) consistent with high secretion of IFN-γ. In contrast, the Ab response to inactivated FMDV is a typical TD response.
Infection of DCs leading to abrogation of TD responses was interpreted in the case of HIV as a viral mechanism of evasion from the immune system (44). This does not seem to be the case for FMDV infection, because despite the depression of DC and T cell functionality, the virus is readily cleared in 3 days. Although the impairment of TD responses is concurrent with the rapid production of protective TI neutralizing Abs, no causal relationship between these facts can be raised at this point. Whether the DC-mediated impairment of T cell functionality influences the production of TI neutralizing Abs requires additional investigation.
In summary, we demonstrate that FMDV infects DCs, inhibiting their ability to stimulate T cells and impairing TD responses in vitro and in vivo. In contrast, inactivated FMDV enhances the ability of DCs to stimulate T cells. These findings indicate that different mechanisms of activation of the immune system are triggered after infection or vaccination with inactivated FMDV. These observations might help to understand FMDV pathogenesis in its natural hosts and to develop more efficient vaccines against it. An explanation of the role played by other cells of the immune system, such as marginal zone B cells and regulatory T cells, during the early response to FMDV will provide additional evidence of the differences in the responses elicited after infection or vaccination.
We thank Ana M. Jar, Agustín I. Ostachuk, and Dr. Ruben O. Donis for critical revision of the manuscript, and Patricia I. Zamorano for helpful assistance.
The authors have no financial conflict of interest.
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
This work was supported by grants from INTA (Foot-and-Mouth Disease Project 522012, modules 5 and 8) and King-Chavez-Park Initiative, Michigan.
Abbreviations used in this paper: FMD, foot-and-mouth disease; BEI, binary ethylenimine; DC, dendritic cell; FMDV, FMD virus; BEI-FMDV, BEI-inactivated FMDV; MFI, mean fluorescence intensity; moi, multiplicity of infection; TCID50, 50% tissue culture infectious dose; TD, thymus dependent; TI, thymus independent; UV-FMDV, UV-irradiated FMDV; VP1, virus protein 1.