We previously described the promotion of type 1 cytokine responses by the nucleoside analogue, ribavirin, in human T cells in vitro. In this study, we examined whether type 1 cytokine polarization by ribavirin in vivo could promote contact hypersensitivity (CHS) responses to dinitrofluorobenzene, a type 1 cytokine-mediated immune response. Unexpectedly, although type 1 cytokine responses were enhanced following ribavirin treatment in vitro and in vivo, the magnitude of CHS responses in BALB/c and C57BL/6 mice was influenced more by a second ribavirin-regulated pathway. The key regulatory molecule in this pathway was IL-10. Ribavirin-mediated suppression of IL-10 in BALB/c mice was associated with increased B7-2 expression and enhanced CHS responses, whereas enhanced IL-10 levels, following ribavirin administration, led to increased B7-1 expression and impaired CHS responses in C57BL/6 mice. The effect of ribavirin on the expression of B7 molecules and on CHS responses was neutralized by IL-10 administration in BALB/c and by anti-IL-10 Ab in C57BL/6. Thus, ribavirin controlled CHS responses directly through the modulation of IL-10 expression, and in vivo outcome was dictated by the preferential expression of either B7-1, an inappropriate costimulatory molecule in CHS, or B7-2, the predominant costimulatory molecule in CHS. Replacing dinitrofluorobenzene priming with IFN-α stimulation, we showed that the ribavirin-regulated pathway could function independent of Ag priming. Altogether, these data showed that, although ribavirin treatment induced a type 1 cytokine bias in contact allergen-primed BALB/c and C57BL/6 mice, in vivo CHS responses were dependent on ribavirin-mediated regulation of both IL-10 and preferential costimulatory signaling.

Upon Ag recognition, mature CD4+ or CD8+ T cells are induced to differentiate into two distinct functional subsets, type 1 and type 2, in both murine and human systems (1, 2, 3). These subsets can be distinguished according to the array of cytokines they produce. Type 1 cells secrete IL-2, IFN-γ, and TNF-α, and are involved in T cell-mediated immune responses such as delayed-type hypersensitivity. In contrast, type 2 cells produce IL-4, IL-5, and IL-10, and provide B cell help for Ab production and regulation of IgE and IgG4 isotype switching. Type 1 responses contribute to active host defense against some intracellular microorganisms, tumors, and tissue grafts, whereas type 2 responses have been associated with a state of active tolerance in the periphery or suppression of type 1 responses (1, 2, 3). Thus, there is significant cross-regulation or antagonism between the type 1 and type 2 subsets.

The antagonistic properties of cytokines generated by these two T cell subsets can be exploited to control the predominant cytokine response. Thus, the up-regulation of type 1 cytokine responses could lead to mitigation of certain classes of immune-related diseases in which pathogenesis or persistence is associated with either a detrimental type 2 cytokine response or an insufficient type 1 cytokine response. An example of a detrimental type 2 cytokine response can be seen in certain viral diseases and allergies in which disease progression is associated with a shift in cytokine profile from type 1 to type 2 in both CD4+ T cells and CD8+ T cells (reviewed in Refs. 4, 5). Examples of insufficient type 1 cytokine responses are seen in the immunity to certain viral or tumor Ags. Enhancing type 1 cytokine expression by coexpression of type 1 cytokine genes along with viral or tumor Ag DNA sequences has been shown to enhance antiviral and antitumor immunity, respectively (6, 7). Thus, an immunomodulating agent that biases endogenous cytokine responses toward a type 1 profile would have significant therapeutic potential in treating diseases in which pathogenesis is associated with a cytokine switch from type 1 to type 2 or in which enhanced type 1 cytokine-mediated immunity would be beneficial.

Ribavirin (1-β-d-ribofuranosyl-1,2,4-triazole-3-carboxamide) is a nucleoside analogue that has demonstrated efficacy in the treatment of viral disease with (respiratory syncytial virus (8)) or without (hepatitis C (9)) directly inhibiting viral replication. Thus, its demonstrable antiviral activity may be attributed to a direct reduction in levels of circulating virus and/or the promotion of T cell-mediated immunity against viral infection. We have recently shown that ribavirin could modulate activation-induced type 1 and type 2 cytokine production in vitro in humans in isolated human T cells (10). This previous study showed that ribavirin, in the dose range 0.5–5 μM, augmented IL-2, IFN-γ, and TNF-α, and suppressed IL-4, IL-5, and IL-10 expression following human T cell activation. Therefore, these data suggested that the enhanced T cell-mediated immunity induced by ribavirin in vivo could be ascribed to the induction of a type 1 cytokine bias.

In this study, we extended those preliminary studies by assessing the functional significance of type 1 cytokine polarization by ribavirin in vivo in a mouse model of T cell-mediated immunity, contact hypersensitivity (CHS)2 to dinitrofluorobenzene (DNFB). We chose this mouse model because the initiation of cutaneous inflammatory responses by DNFB is mediated by the production of type 1 cytokines by MHC I-restricted CD8+ Tc1 cells (11) and regulated by MHC II-restricted, CD4+ Th2 cells (12). In this study, we showed that type 1 cytokine polarization does occur in DNFB-primed and unprimed lymph node cells (LNC) from BALB/c mice activated in vitro. However, ribavirin treatment, at the time of challenge, dramatically influenced in vivo CHS responses in both BALB/c and C57BL/6 mice, but in an antagonistic manner. As hypothesized, in BALB/c mice, ribavirin-mediated type 1 cytokine polarization did enhance the type 1-mediated inflammatory response to DNFB. Unexpectedly, C57BL/6 mice showed impaired CHS responses following ribavirin treatment. We found that the phenotypic outcome of CHS responses following ribavirin administration was affected by the expression of both IL-10 and the costimulatory molecules, B7-1 and B7-2. These studies show that the influence of ribavirin on type 1 cytokine polarization in vivo is biased by the type of costimulatory signal induced and by factors intrinsic to each mouse strain.

Six- to eight-week-old female BALB/c or C57BL/6 mice were purchased from Charles River Laboratories (Portage, MI) or Bantin and Kingman Universal (Fremont, CA).

mAb to the following mouse cell surface molecules and cytokines were used: B7-1 (CD80)-FITC (IG10, rat IgG2a), B7-2 (CD86)-FITC (GL1, rat IgG2a), CD3-PE (145-2C11, hamster IgG), IL-10 (JES5-2A5, rat IgG1), and rat IgG1 isotype control, all from PharMingen (San Diego, CA). Mouse rIL-10 was also obtained from PharMingen.

BALB/c or C57BL/6 mice were sensitized to contact allergen by application of 20 μl of 0.3% DNFB (Sigma, St. Louis, MO) in acetone:olive oil, in the ratio 4:1, onto the shaved abdomens 5 days before sacrifice. These primed mice and a second group of unprimed mice were sacrificed by cervical dislocation, and axillary/lateral axillary lymph nodes were removed. LNC suspensions were then prepared for individual mice in both groups. Twenty-four-well plates were coated with 150 μl of a 25 μg/ml preparation of anti-mouse αβTCR Ab (clone H57-597; PharMingen) for 90 min at 37°C, then plates were washed twice with cold PBS. A total of 2 × 106 LNC/well was added in 1 ml complete DMEM media (DMEM containing 4.5 g/L dextrose (ICN Biomedicals, Costa Mesa, CA) and supplemented with 10% FBS (HyClone, Logan, UT), 1% l-glutamine, 1% penicillin/streptomycin, 10 mM HEPES, 1× nonessential amino acids, and 50 μM 2-ME). Cells were cultured for 48 h in the presence of 0–5 μM ribavirin, then extracellular cytokine analyses were performed in cell-free supernatants.

Murine cytokine levels were determined in cell supernatants, following appropriate dilution, using ELISA kits specific for IL-2, IL-4, IL-5, IL-10, TNF-α, and IFN-γ ELISAs from Genzyme (Cambridge, MA). All ELISA results were expressed as pg/ml.

Reactivity to the contact allergen, DNFB, was determined, in BALB/c or C57BL/6 mice, as previously described (13). Briefly, mice were sensitized by application of 20 μl of 0.3% DNFB in acetone:olive oil, in the ratio 4:1, onto the shaved abdomens of naive mice. For optimal elicitation of CHS, the mice were challenged on both sides of each ear with 20 μl of 0.12% DNFB, 5 days after sensitization. Unsensitized mice were also challenged and used as controls in each experiment. After 24 h, ear thickness measurements were taken, and response to DNFB was assessed by subtracting postchallenge from prechallenge values. Where indicated, ribavirin, at a dose of 5 μg in 50 μl PBS (0.5 mg/kg), was administered by i.p. injection at the time of challenge with DNFB. This dose of ribavirin gave maximal effect in preliminary optimization studies in BALB/c mice. In some experiments, 100 μl of PBS containing either 1 μg IL-10, or 50 μg anti-mouse IL-10 Ab, or 50 μg matched isotype control Ab were injected i.p. in DNFB-primed mice 24 h before ear challenge with DNFB. Following final ear thickness measurements, mice were sacrificed by cervical dislocation, and axillary/lateral axillary lymph nodes were removed. Following isolation of total cellular RNA from isolated LNC, RT-PCR and Southern Blot analyses were performed to monitor for mouse IFN-γ, IL-2, IL-10, IFN-α, B7-1, and B7-2 mRNA levels.

Total cellular RNA was extracted using Trizol reagent (Life Technologies, Gaithersburg, MD). The cDNA synthesis reaction was performed using oligo(dT)12–18 primer and Superscript II (Life Technologies) reverse transcriptase. The PCR reactions were performed using cDNA reverse transcribed from total cellular RNA, and primers for mouse IL-2, IFN-γ, IFN-α, and β-actin were obtained from Stratagene (La Jolla, CA), and mouse IL-10 from Clontech (Palo Alto, CA). The primers for mouse B7-1 were sense 5′-TGACTTCTCTACCCCCAACA-3′ and antisense 5′-TGATGACAACGATGAC GACG-3′, and for B7-2, sense 5′-AGAAGACCCTCCTGATAGCA-3′ and antisense 5′-AAGGAAGACGGTCTGTTCAG-3′. Amplification conditions for B7-1 and B7-2 and the Stratagene mouse cytokine primers were a 5-min denaturation at 94°C and a 5-min annealing at 60°C, followed by 35 cycles of 1.5 min at 72°C, 45 s at 94°C, and 45 s at 60°C, with a final extension time of 10 min at 72°C. For IL-10, the PCR conditions were a 5-min denaturation at 94°C, followed by 35 cycles of 45 s at 94°C, 45 s at 60°C, and 2 min at 72°C, with a final extension time of 7 min at 72°C.

For each gene product, the optimum cDNA dilution was determined experimentally and was defined as the cDNA dilution that would achieve a detectable concentration that was well below saturating conditions. PCR products were separated on 2% agarose containing ethidium bromide and immobilized to Hybond N+ membrane (Amersham, Arlington Heights, IL) overnight using 0.4 M NaOH and 0.2 M NaCl. Blots were hybridized with [γ-32P]ATP-labeled oligonucleotide probes (either generated from the original primers (Stratagene) or from specific probes designed to be complementary to a central region within individual PCR products). These included IL-10 (Clontech) and the probes for mouse B7-1 and B7-2, which are 5′-CGACTCGCAACCACACCATTAAGTGTCTCA-3′ and 5′-AGGATTCGGCGCAGTAATAACAGTCGTCGT-3′, respectively. To verify that equal amounts of RNA were added to each PCR reaction within an experiment, primers for the housekeeping gene, β-actin (mouse), were used to amplify cDNA reverse transcribed from total RNA. These PCR products were hybridized using a probe generated from β-actin primers (mouse). Washed blots were then analyzed following autoradiography using a phosphor imager (Bio-Rad, Richmond, CA). Relative changes in cytokine or B7 molecule mRNA were presented as densitometric readings and normalized for any variations in input RNA by determining the densitometric ratio of mRNA of interest relative to mRNA of β-actin.

BALB/c or C57BL/6 mice were sacrificed by cervical dislocation, and spleens were removed for analysis. Isolated splenocytes from individual mice were obtained following removal of erythrocytes using ACK lysing solution containing 0.15 M ammonium chloride, 1 mM potassium bicarbonate, and 0.1 mM EDTA, pH 7.4. Splenocytes (2 × 106) were resuspended in 1 ml complete RPMI media (RPMI 1640 containing 10% FBS (HyClone), 1% l-glutamine, and 1% penicillin/streptomycin) and then treated in vitro in duplicate with 0 or 104 U of murine IFN-α (Life Technologies) in the presence of 0–5 μM ribavirin. Spleen cell cultures in 24-well plates were incubated for 24 h before FACS analysis. For RNA analyses, 5 × 106 splenocytes were resuspended in 2 ml complete RPMI media and cocultured with 2 × 104 U of murine IFN-α and either 1 μg/ml anti-IL-10 Ab or matched isotype rat IgG1 control (both from PharMingen). After 24 h, total RNA was isolated for subsequent RT-PCR analysis.

For direct staining with fluorescence-conjugated mAbs to cell surface Ags, the cells were washed twice with isotonic saline solution, pH 7.4 (Becton Dickinson, Mansfield, MA), resuspended as duplicate samples containing 0.3 × 106 cells in 50 μl isotonic saline solution. Sample aliquots were costained with a combination of either PE-CD3 and FITC-CD80 (B7-1) or PE-CD3 and FITC-CD86 (B7-2). In each assay, a separate aliquot was stained with PE/FITC-labeled isotype-matched control mAbs to assess nonspecific fluorescence. Incubations were performed at 4°C in the dark for 45 min using saturating mAb concentrations. Unincorporated label was removed by washing in PBS before the analysis with a FACScan flow cytometer (Becton Dickinson).

Ag density was determined following two-color flow cytometry analysis of 5,000–10,000 viable cells. Surface expression of specific Ag (CD80, CD86) on gated cells was determined using CellQuest software and, unless stated, expressed as the mean channel of fluorescence.

Statistical significance, where relevant, was assessed using modified one-way or two-way ANOVA or by the nonparametric Mann-Whitney test (a two-sample t test), as appropriate.

We investigated the influence of ribavirin on the cytokine pattern in vitro in BALB/c and C57BL/6 mice with or without specific (DNFB) sensitization. The levels of IL-2, IFN-γ, TNF-α, IL-4, IL-5, and IL-10 were measured following 48-h in vitro challenge with plate-bound anti-αβTCR Ab in LNC from both unprimed mice and DNFB-primed mice of both strains in the absence and presence of ribavirin, and these data are shown in Fig. 1.

FIGURE 1.

The effect of ribavirin on secreted type 1 (A) and type 2 (B) cytokine levels in LNC from unprimed BALB/c or C57BL/6 mice (○), or from BALB/c or C57BL/6 mice sensitized with DNFB (•). Extracellular levels of type 1 (IL-2, IFN-γ, and TNF-α) and type 2 (IL-4, IL-5, and IL-10) cytokines were determined in anti-mouse αβTCR-activated LNC from 10 unprimed and 10 sensitized mice of each strain following a 48-h incubation in the presence of 0–5 μM (BALB/c) or 0, 1, and 2 μM (C57BL/6) ribavirin. The effect of ribavirin on cytokine levels induced by in vitro challenge was assessed in cell-free supernatants by ELISA analyses and represented as mean cytokine concentration (pg/ml) ± SEM. The p values for the effect of ribavirin on cytokine responses in BALB/c mice were as follows: unprimed mice, IL-2, 0.04; IFN-γ, <0.001; TNF-α, <0.001; IL-4, 0.004; IL-5, 0.003; and IL-10, 0.005; and DNFB-primed mice, IL-2, 0.001; IFN-γ, 0.001; TNF-α, 0.02; IL-4, 0.006; IL-5, 0.009; and IL-10, <0.0001. The p values for C57BL/6 mice were as follows: unprimed mice, IL-2, IFN-γ and TNF-α, all <0.0001; IL-4, 0.004; and IL-10, 0.005; and DNFB-primed mice, IL-2, IFN-γ, TNF-α, IL-4, IL-5, and IL-10, all <0.0001.

FIGURE 1.

The effect of ribavirin on secreted type 1 (A) and type 2 (B) cytokine levels in LNC from unprimed BALB/c or C57BL/6 mice (○), or from BALB/c or C57BL/6 mice sensitized with DNFB (•). Extracellular levels of type 1 (IL-2, IFN-γ, and TNF-α) and type 2 (IL-4, IL-5, and IL-10) cytokines were determined in anti-mouse αβTCR-activated LNC from 10 unprimed and 10 sensitized mice of each strain following a 48-h incubation in the presence of 0–5 μM (BALB/c) or 0, 1, and 2 μM (C57BL/6) ribavirin. The effect of ribavirin on cytokine levels induced by in vitro challenge was assessed in cell-free supernatants by ELISA analyses and represented as mean cytokine concentration (pg/ml) ± SEM. The p values for the effect of ribavirin on cytokine responses in BALB/c mice were as follows: unprimed mice, IL-2, 0.04; IFN-γ, <0.001; TNF-α, <0.001; IL-4, 0.004; IL-5, 0.003; and IL-10, 0.005; and DNFB-primed mice, IL-2, 0.001; IFN-γ, 0.001; TNF-α, 0.02; IL-4, 0.006; IL-5, 0.009; and IL-10, <0.0001. The p values for C57BL/6 mice were as follows: unprimed mice, IL-2, IFN-γ and TNF-α, all <0.0001; IL-4, 0.004; and IL-10, 0.005; and DNFB-primed mice, IL-2, IFN-γ, TNF-α, IL-4, IL-5, and IL-10, all <0.0001.

Close modal

There is a notable difference between the cytokine responses from DNFB-primed BALB/c and DNFB-primed C57BL/6 mice. DNFB-primed BALB/c mice produced significantly higher amounts of IL-2 (p < 0.0001), IFN-γ (p = 0.0004), IL-10 (p = 0.003), IL-4 (p = 0.003), and IL-5 (p = 0.001). Conversely, DNFB-primed C57BL/6 mice make significantly more TNF-α (p < 0.0001) compared with BALB/c mice. Owing to the much smaller lymph nodes isolated from C57BL/6 mice, only ribavirin concentrations of 0, 1, and 2 μM were evaluated, whereas BALB/c LNC were tested at six ribavirin concentrations.

Ribavirin, at all concentrations tested, significantly enhanced activation-induced expression of the type 1 cytokines, IL-2, IFN-γ, and TNF-α. In contrast, ribavirin significantly suppressed activated levels of IL-4 in LNC from unprimed and DNFB-primed BALB/c and C57BL/6 mice, and lowered IL-5 expression in unprimed and DNFB-primed BALB/c mice and DNFB-primed C57BL/6 mice (Fig. 1,B). No IL-5 production was observed in unprimed C57BL/6 mice. Unexpectedly, the expression of IL-10 varied between mouse strains (Fig. 1 B). In BALB/c mice, IL-10 was significantly reduced following ribavirin treatment in both unprimed and DNFB-primed mice. In contrast, IL-10 expression was significantly enhanced in unprimed and DNFB-primed C57BL/6 mice following ribavirin treatment. These data show that, although ribavirin did induce a type 1 cytokine bias in both mouse strains, the polarization toward type 1 was incomplete in C57BL/6 mice. In contrast to BALB/c mice in which levels of the type 2 cytokine, IL-10, were suppressed, ribavirin elevated IL-10 levels in C57BL/6 mice.

Next we evaluated the effect of ribavirin administration in vivo on a type 1 cytokine-mediated immune response. To address this, we followed the murine CHS response to DNFB to determine whether 1) ribavirin-mediated type 1 cytokine polarization observed in both mouse strains could induce enhanced inflammatory ear responses or 2) the regulatory effects driven by contrasting IL-10 profiles could elicit opposing mouse strain-dependent in vivo effects. CHS responses were elicited and compared in DNFB (0.3%)-primed and naive mice, and in ribavirin-treated, DNFB-primed BALB/c and C57BL/6 mice following ear challenge with 0.12% DNFB. Modulation of CHS, as determined from ear swelling measurements following challenge, was calculated following subtraction of responses in nonsensitized challenged (naive) mice. CHS responses following priming and challenge in BALB/c mice were significantly greater (p < 0.0001) than those seen in DNFB-primed C57BL/6 mice (Fig. 2,A). In both strains, the untreated DNFB-primed group significantly augmented CHS responses when compared with the naive group (p < 0.0001, Fig. 2 A).

FIGURE 2.

Comparison of CHS responses and LNC cytokine mRNA expression in BALB/c (left panels) and C57BL/6 (right panels) mice following ribavirin treatment in vivo. A, CHS responses were induced in two groups of five mice of either strain by sensitization and challenge with DNFB (Primed), as described in Materials and Methods. At the time of challenge with 0.12% DNFB, one group from each strain was administered 5 μg of ribavirin (RIB) in PBS by i.p. injection. The ear swelling in all groups (including an unsensitized control group of five mice per strain (Naive)) was measured after 24 h. The data shown as mean ear swelling (mm × 10−2) ± SD are representative of five (C57BL/6) or eight (BALB/c) separate experiments with five animals in each test group. ∗, p < 0.0001 when compared with primed animals in the absence of ribavirin. B, Cytokine mRNA levels following elicitation of DNFB-mediated CHS in ribavirin-treated BALB/c or C57BL/6 mice were measured in local draining LNC. IL-2, IL-10, and IFN-γ mRNA expression, 24 h post-DNFB challenge, was determined in total RNA isolated from the LNC of individual DNFB-primed mice (Primed), unprimed mice (Naive), and DNFB-primed mice treated with ribavirin (Primed + RIB). Cytokine and β-actin mRNA levels were assessed following RT-PCR and Southern blot analysis with specific 32P-labeled DNA probes. Autoradiography data are representative of three separate experiments using three mice per treatment group. C, The relative changes in cytokine mRNA from the mouse groups in B are presented as densitometric readings for Primed (hatched bars), Naive (open bars), and Primed + RIB (filled bars) mouse groups and normalized for any variations in input RNA by determining the densitometric ratio of mRNA of interest relative to mRNA of β-actin. Data are mean ± SD from a pool of five mice per group. ∗, Primed + RIB group was significantly higher than Primed or Naive groups for IL-2 (BALB/c, p = 0.003; C57BL/6, p = 0.006), for IFN-γ (BALB/c, p = 0.002; C57BL/6, p = 0.005), and for IL-10 (C57BL/6, p = 0.002). ∗∗, Primed group was significantly higher than Primed + RIB or Naive groups for IL-10 (BALB/c, p = 0.002).

FIGURE 2.

Comparison of CHS responses and LNC cytokine mRNA expression in BALB/c (left panels) and C57BL/6 (right panels) mice following ribavirin treatment in vivo. A, CHS responses were induced in two groups of five mice of either strain by sensitization and challenge with DNFB (Primed), as described in Materials and Methods. At the time of challenge with 0.12% DNFB, one group from each strain was administered 5 μg of ribavirin (RIB) in PBS by i.p. injection. The ear swelling in all groups (including an unsensitized control group of five mice per strain (Naive)) was measured after 24 h. The data shown as mean ear swelling (mm × 10−2) ± SD are representative of five (C57BL/6) or eight (BALB/c) separate experiments with five animals in each test group. ∗, p < 0.0001 when compared with primed animals in the absence of ribavirin. B, Cytokine mRNA levels following elicitation of DNFB-mediated CHS in ribavirin-treated BALB/c or C57BL/6 mice were measured in local draining LNC. IL-2, IL-10, and IFN-γ mRNA expression, 24 h post-DNFB challenge, was determined in total RNA isolated from the LNC of individual DNFB-primed mice (Primed), unprimed mice (Naive), and DNFB-primed mice treated with ribavirin (Primed + RIB). Cytokine and β-actin mRNA levels were assessed following RT-PCR and Southern blot analysis with specific 32P-labeled DNA probes. Autoradiography data are representative of three separate experiments using three mice per treatment group. C, The relative changes in cytokine mRNA from the mouse groups in B are presented as densitometric readings for Primed (hatched bars), Naive (open bars), and Primed + RIB (filled bars) mouse groups and normalized for any variations in input RNA by determining the densitometric ratio of mRNA of interest relative to mRNA of β-actin. Data are mean ± SD from a pool of five mice per group. ∗, Primed + RIB group was significantly higher than Primed or Naive groups for IL-2 (BALB/c, p = 0.003; C57BL/6, p = 0.006), for IFN-γ (BALB/c, p = 0.002; C57BL/6, p = 0.005), and for IL-10 (C57BL/6, p = 0.002). ∗∗, Primed group was significantly higher than Primed + RIB or Naive groups for IL-10 (BALB/c, p = 0.002).

Close modal

Intraperitoneal administration of ribavirin at the time of challenge significantly enhanced CHS responses in DNFB-primed BALB/c mice (p < 0.0001, Fig. 2,A). The mean increase in ear thickness (±SD) following ribavirin treatment shown in Fig. 2 A, calculated from the mean responses of nine experiments using three to five mice per treatment group, was 9.9 mm × 10−2 ± 1.7 (an increase of 53% ± 18) 24 h postchallenge. Similarly enhanced inflammatory ear responses were observed following ribavirin treatment 48 h postchallenge (data not shown).

In marked contrast, ribavirin treatment dramatically suppressed CHS responses in C57BL/6 mice (p < 0.0001, Fig. 2,A). The mean decrease in ear thickness (±SD) following ribavirin treatment (Fig. 2 A), calculated from the mean responses of five experiments using five mice per treatment group, was 7.5 mm × 10−2 ± 2.9 (−64% ± 16) 24 h postchallenge. Similarly decreased inflammatory ear responses were observed following ribavirin treatment 48 h postchallenge (data not shown). Collectively, these data show that administration of ribavirin induced opposing in vivo CHS responses to DNFB in BALB/c and C57BL/6 mice.

In addition, mRNA expression of IL-2 and IFN-γ from LNC in ribavirin-treated, DNFB-primed BALB/c and C57BL/6 mice was significantly increased when compared with DNFB-primed mice alone. Data are shown as an autoradiograph (Fig. 2,B) or as relative changes in densitometric readings normalized to β-actin (Fig. 2,C). In contrast, LNC-derived IL-10 mRNA expression varied between mouse strains. IL-10 mRNA expression was significantly decreased in ribavirin-treated, DNFB-primed BALB/c mice when compared with DNFB-primed BALB/c mice alone, whereas IL-10 mRNA expression was significantly increased in ribavirin-treated, DNFB-primed C57BL/6 mice when compared with DNFB-primed C57BL/6 mice alone (Fig. 2, B and C). These data show that in vivo treatment with ribavirin induced a type 1 cytokine bias in both mouse strains. In BALB/c mice, this type 1 cytokine bias was complete (enhanced type 1 and suppressed type 2 cytokines), resulting in enhanced CHS responses. However, in C57BL/6 mice, the ribavirin-mediated type 1 cytokine bias was incomplete. Our data suggest that an increase in IL-10 expression in C57BL/6 LNC induced by ribavirin could override the ability of type 1 cytokine polarization to enhance CHS responses in C57BL/6 mice.

It has been previously shown that IL-10, but not IL-4, regulates CHS responses in mice (14). In addition, IL-10 can regulate expression of the costimulatory molecules, B7-1 and B7-2, on APC (15, 16, 17). We therefore determined whether a correlation existed between the effect of ribavirin on CHS responses and the expression of both B7-1 and B7-2 in LNC from BALB/c and C57BL/6 mice. Surface expression of B7-1 and B7-2 in LNC from DNFB-primed BALB/c and C57BL/6 mice was assessed by FACS following 24-h DNFB challenge and administration of ribavirin in vivo. Following ribavirin treatment, B7-2 expression was elevated in BALB/c mice, and B7-1 expression was elevated in C57BL/6 mice (Fig. 3 A). No ribavirin-mediated effects were seen on the surface expression of the B7 ligands, CD28 or CTLA-4 (data not shown).

FIGURE 3.

B7-1 and B7-2 surface expression and mRNA expression following DNFB-mediated CHS responses in ribavirin-treated BALB/c (left panels) and C57BL/6 (right panels) mice. Ribavirin (5 μg/50 μl PBS) was injected i.p. at the time of 0.12% DNFB challenge. A, Cell surface expression of B7-1 and B7-2, 24 h post-DNFB challenge, was determined in isolated LNC of individual DNFB-primed mice (dashed lines) and DNFB-primed mice treated with ribavirin (solid lines). B7-1 and B7-2 expression was also determined in unprimed mice from respective strains (dotted lines). B7-1 and B7-2 levels were assessed by staining with FITC-labeled Abs to B7-1 (CD80) and B7-2 (CD86), followed by FACS analysis of gated CD3-negative cells. The data shown are representative of three experiments using five mice per test group. B, B7-1 and B7-2 mRNA expression, 24 h post-DNFB challenge, was determined in total RNA isolated from the LNC of individual DNFB-primed mice and DNFB-primed mice treated with ribavirin. B7-1 and B7-2, IFN-α, and β-actin mRNA levels were assessed following RT-PCR and Southern blot analysis with specific 32P-labeled DNA probes. B7-1 and B7-2 mRNA expression in a mouse B cell lymphoma cell line, A20, was also determined as a control for the B7-1 and B7-2 primers. Autoradiography data are representative of two separate experiments using five mice per treatment group. C, The relative changes in B7-1, B7-2, and IFN-α mRNA from the mouse groups in B are presented as densitometric readings for Primed (hatched bars) and Primed + RIB (solid bars) mouse groups and normalized for any variations in input RNA by determining the densitometric ratio of mRNA of interest relative to mRNA of β-actin. Data are mean ± SD from a pool of five per group. ∗, BALB/c, p = 0.004; C57BL/6, p = 0.009.

FIGURE 3.

B7-1 and B7-2 surface expression and mRNA expression following DNFB-mediated CHS responses in ribavirin-treated BALB/c (left panels) and C57BL/6 (right panels) mice. Ribavirin (5 μg/50 μl PBS) was injected i.p. at the time of 0.12% DNFB challenge. A, Cell surface expression of B7-1 and B7-2, 24 h post-DNFB challenge, was determined in isolated LNC of individual DNFB-primed mice (dashed lines) and DNFB-primed mice treated with ribavirin (solid lines). B7-1 and B7-2 expression was also determined in unprimed mice from respective strains (dotted lines). B7-1 and B7-2 levels were assessed by staining with FITC-labeled Abs to B7-1 (CD80) and B7-2 (CD86), followed by FACS analysis of gated CD3-negative cells. The data shown are representative of three experiments using five mice per test group. B, B7-1 and B7-2 mRNA expression, 24 h post-DNFB challenge, was determined in total RNA isolated from the LNC of individual DNFB-primed mice and DNFB-primed mice treated with ribavirin. B7-1 and B7-2, IFN-α, and β-actin mRNA levels were assessed following RT-PCR and Southern blot analysis with specific 32P-labeled DNA probes. B7-1 and B7-2 mRNA expression in a mouse B cell lymphoma cell line, A20, was also determined as a control for the B7-1 and B7-2 primers. Autoradiography data are representative of two separate experiments using five mice per treatment group. C, The relative changes in B7-1, B7-2, and IFN-α mRNA from the mouse groups in B are presented as densitometric readings for Primed (hatched bars) and Primed + RIB (solid bars) mouse groups and normalized for any variations in input RNA by determining the densitometric ratio of mRNA of interest relative to mRNA of β-actin. Data are mean ± SD from a pool of five per group. ∗, BALB/c, p = 0.004; C57BL/6, p = 0.009.

Close modal

In addition, the expression of B7-1 and B7-2 mRNA was determined in isolated RNA from LNC of both strains of DNFB-treated mice. As seen with the surface expression data, ribavirin treatment elevated B7-2 mRNA expression in BALB/c mice, but had no effect on B7-1 expression. Data are shown as an autoradiograph (Fig. 3,B) or as relative changes in densitometric readings normalized to β-actin (Fig. 3,C). Conversely, B7-1 mRNA levels were enhanced in ribavirin-treated C57BL/6 mice, while B7-2 levels were unaffected (Fig. 3, B and C). Ribavirin had no effect on the expression of IFN-α (a known inducer of B7 molecules) mRNA in either BALB/c or C57BL/6 mice. No effect on B7-1 or B7-2 expression was observed following ribavirin treatment in LNC from unprimed mice of both strains (data not shown).

Thus, ribavirin treatment resulted in elevated B7-2 expression in DNFB-primed BALB/c mice and elevated B7-1 expression in C57BL/6 mice. These actions of ribavirin were not related to any increase in IFN-α levels (a known inducer of B7 molecules) and were independent of any effect on B7 ligands, CD28 and CTLA-4.

Collectively, these data show that the ribavirin-mediated in vivo response to DNFB in BALB/c and C57BL/6 mice was affected by 1) polarization of T cell subset cytokine expression and 2) differential regulation of IL-10 in conjunction with alterations in the balance of cell surface levels of B7-1 and B7-2.

Next we designed experiments to determine whether the effect of ribavirin on IL-10 directly regulated B7-1 and B7-2 expression in these mouse strains. First, we determined the effect of i.p. injection of 1 μg of IL-10 on CHS responses and B7-1 and B7-2 expression in both mouse strains using a protocol that previously demonstrated suppression of CHS response to DNFB in BALB/c and C3H mice (18). IL-10 administered 24 h before challenge dramatically suppressed CHS responses in both BALB/c and C57BL/6 mice (Fig. 4,A). In both mouse strains, this impaired CHS response was associated with an increase in B7-1 expression, whereas B7-2 levels remained unchanged (Fig. 4, B and C). These data show that the administration of IL-10 in vivo directly modulated B7-1 expression and induced changes to the CHS responses that were indistinguishable between both strains of mice. Administration of ribavirin alone in DNFB-primed C57BL/6 mice paralleled the effect of in vivo treatment with IL-10 alone with respect to CHS response and B7-1 mRNA expression. Coinjection of ribavirin and anti-IL-10 Ab (50 μg) reversed ribavirin-mediated CHS suppression in C57BL/6 mice and concomitantly reduced B7-1 mRNA expression to levels found in untreated DNFB-primed C57BL/6 mice (Fig. 4). These data showed that in C57BL/6 mice, ribavirin acted by inducing IL-10 and suppressing CHS response through the elevation of B7-1 expression. In contrast, adminstration of ribavirin alone in DNFB-primed BALB/c mice led to enhanced CHS responses and elevated B7-2 mRNA expression. Coinjection of ribavirin and IL-10 reversed ribavirin-mediated CHS enhancement in BALB/c mice and concomitantly reduced B7-2 expression to below levels found in untreated DNFB-primed mice while elevating B7-1 expression (Fig. 4). These data showed that in BALB/c mice, ribavirin acted by suppressing IL-10 and manifesting enhanced CHS responses through the elevation of B7-2 expression. Collectively, these findings show that ribavirin administration in vivo directly influenced the magnitude of CHS response in each mouse strain by modulating the effect of IL-10-mediated regulation of the costimulatory molecules, B7-1 and B7-2.

FIGURE 4.

The ribavirin-mediated regulation of CHS responses and expression of the costimulatory molecules, B7-1 and B7-2, in BALB/c (left panels) and C57BL/6 (right panels) mice is dependent on IL-10. A, CHS responses were induced in five groups of five mice of either strain by sensitization (day 0) and challenge (day 6) with DNFB (Primed), as described in Materials and Methods. Twenty-four hours before challenge, two Primed BALB/c groups and one Primed C57BL/6 group were administered IL-10 (1 μg) i.p., and one Primed C57BL/6 group was administered anti-IL-10 Ab (50 μg) i.p. At the time of challenge with 0.12% DNFB, two BALB/c groups, one Primed group (Primed + RIB), and one Primed group injected with IL-10 (Primed + RIB + IL-10) were administered 5 μg of ribavirin (RIB) in PBS by i. p. injection. Of the C57BL/6 mice, one Primed group (Primed + RIB) and one Primed group injected with anti-IL-10 Ab (Primed + RIB + IL-10 Ab) were administered 5 μg of ribavirin. All mouse groups, including the remaining DNFB-primed, IL-10 group from each strain (Primed + IL-10), were challenged on the ears with 0.12% DNFB on day 6. The ear swelling in all groups (including an unsensitized control group (Naive)) was measured after 24 h. The data shown as mean ear swelling (mm × 10−2) ± SD are representative of two experiments with five to six animals in each test group. ∗, p = 0.0002 for group comparisons against DNFB-primed mice alone. ∗∗, p = 0.0003 for comparison against ribavirin-treated DNFB-primed mice. B, B7-1 and B7-2 mRNA expression, 24 h post-DNFB challenge, was determined in total RNA isolated from the LNC of individual DNFB-primed mice (DNFB) and DNFB-primed mice treated with either IL-10 (DNFB + IL-10), ribavirin (DNFB + RIB), or ribavirin and IL-10 (BALB/c) (DNFB + IL-10 + RIB) or ribavirin and anti-IL-10 Ab (C57BL/6) (DNFB + anti-IL-10 Ab + RIB). B7-1 and and B7-2 mRNA levels were assessed following RT-PCR and Southern blot analysis with specific 32P-labeled DNA probes. RNA samples from an unprimed mouse challenged with DNFB (Naive) and a DNFB-primed C57BL/6 mouse injected with isotype control Ab (Ig) were also included. Autoradiography data are representative of two separate experiments using three mice per treatment group. C, The relative changes in B7-1 (hatched bars), and B7-2 (solid bars) mRNA expression from the mouse groups in B are presented as densitometric readings. Primed, Primed + RIB, Primed + IL-10, and Primed + RIB + IL-10 (BALB/c) and Primed + RIB + anti-IL-10 Ab (C57BL/6) mouse groups are normalized for any variations in input RNA by determining the densitometric ratio of mRNA of interest relative to mRNA of β-actin. Data are mean ± SD from a pool of five to six mice per group. p values for group comparisons against DNFB-primed mice alone: B7-1, 1) IL-10 alone, BALB/c, p = 0.004; C57BL/6, p = 0.009; 2) ribavirin alone, C57BL/6, p = 0.009; 3) IL-10 + ribavirin, BALB/c, p = 0.004; B7-2, 1) ribavirin alone, BALB/c, p = 0.004; 2) IL-10 + ribavirin, BALB/c, p = 0.004. p values for comparison of effect of IL-10 (BALB/c) or anti-IL-10 Ab (C57BL/6) against ribavirin-treated DNFB-primed mice: B7-1, BALB/c, p = 0.004; C57BL/6, p = 0.009; B7-2, BALB/c, p = 0.004.

FIGURE 4.

The ribavirin-mediated regulation of CHS responses and expression of the costimulatory molecules, B7-1 and B7-2, in BALB/c (left panels) and C57BL/6 (right panels) mice is dependent on IL-10. A, CHS responses were induced in five groups of five mice of either strain by sensitization (day 0) and challenge (day 6) with DNFB (Primed), as described in Materials and Methods. Twenty-four hours before challenge, two Primed BALB/c groups and one Primed C57BL/6 group were administered IL-10 (1 μg) i.p., and one Primed C57BL/6 group was administered anti-IL-10 Ab (50 μg) i.p. At the time of challenge with 0.12% DNFB, two BALB/c groups, one Primed group (Primed + RIB), and one Primed group injected with IL-10 (Primed + RIB + IL-10) were administered 5 μg of ribavirin (RIB) in PBS by i. p. injection. Of the C57BL/6 mice, one Primed group (Primed + RIB) and one Primed group injected with anti-IL-10 Ab (Primed + RIB + IL-10 Ab) were administered 5 μg of ribavirin. All mouse groups, including the remaining DNFB-primed, IL-10 group from each strain (Primed + IL-10), were challenged on the ears with 0.12% DNFB on day 6. The ear swelling in all groups (including an unsensitized control group (Naive)) was measured after 24 h. The data shown as mean ear swelling (mm × 10−2) ± SD are representative of two experiments with five to six animals in each test group. ∗, p = 0.0002 for group comparisons against DNFB-primed mice alone. ∗∗, p = 0.0003 for comparison against ribavirin-treated DNFB-primed mice. B, B7-1 and B7-2 mRNA expression, 24 h post-DNFB challenge, was determined in total RNA isolated from the LNC of individual DNFB-primed mice (DNFB) and DNFB-primed mice treated with either IL-10 (DNFB + IL-10), ribavirin (DNFB + RIB), or ribavirin and IL-10 (BALB/c) (DNFB + IL-10 + RIB) or ribavirin and anti-IL-10 Ab (C57BL/6) (DNFB + anti-IL-10 Ab + RIB). B7-1 and and B7-2 mRNA levels were assessed following RT-PCR and Southern blot analysis with specific 32P-labeled DNA probes. RNA samples from an unprimed mouse challenged with DNFB (Naive) and a DNFB-primed C57BL/6 mouse injected with isotype control Ab (Ig) were also included. Autoradiography data are representative of two separate experiments using three mice per treatment group. C, The relative changes in B7-1 (hatched bars), and B7-2 (solid bars) mRNA expression from the mouse groups in B are presented as densitometric readings. Primed, Primed + RIB, Primed + IL-10, and Primed + RIB + IL-10 (BALB/c) and Primed + RIB + anti-IL-10 Ab (C57BL/6) mouse groups are normalized for any variations in input RNA by determining the densitometric ratio of mRNA of interest relative to mRNA of β-actin. Data are mean ± SD from a pool of five to six mice per group. p values for group comparisons against DNFB-primed mice alone: B7-1, 1) IL-10 alone, BALB/c, p = 0.004; C57BL/6, p = 0.009; 2) ribavirin alone, C57BL/6, p = 0.009; 3) IL-10 + ribavirin, BALB/c, p = 0.004; B7-2, 1) ribavirin alone, BALB/c, p = 0.004; 2) IL-10 + ribavirin, BALB/c, p = 0.004. p values for comparison of effect of IL-10 (BALB/c) or anti-IL-10 Ab (C57BL/6) against ribavirin-treated DNFB-primed mice: B7-1, BALB/c, p = 0.004; C57BL/6, p = 0.009; B7-2, BALB/c, p = 0.004.

Close modal

The relative importance/necessity of Ag (DNFB) stimulation on the ribavirin-mediated effects on IL-10, B7-1, and B7-2 expression was examined. We determined the ribavirin-mediated effects on IL-10, B7-1, and B7-2 expression following treatment of unstimulated splenocytes from BALB/c and C57BL/6 mice with IFN-α, a nonantigenic inducer of IL-10 expression and B7 molecules. Following RT-PCR and analysis by autoradiography and densitometry, we showed that in IFN-α-stimulated BALB/c splenocytes, IL-10 mRNA expression was induced by IFN-α (+67%). More importantly, we showed that ribavirin inhibited IFN-α-induced IL-10 expression in a dose-dependent manner (−100%, −69%, and −2% for 5, 2, and 1 μM ribavirin, respectively) (Fig. 5 A). Conversely, in IFN-α-stimulated C57BL/6 splenocytes, IL-10 expression was enhanced by +65%, and addition of ribavirin enhanced IFN-α-induced IL-10 mRNA expression in a dose-dependent manner (+265%, +75%, and +0% for 5, 2, and 1 μM ribavirin, respectively). Thus, the effect of ribavirin on IFN-α-induced IL-10 expression paralleled the observations we have shown in DNFB-primed mice.

FIGURE 5.

The differential regulation of IL-10, B7-1, and B7-2 expression by ribavirin can occur independent of antigenic stimulation. A, Ribavirin differentially modulates IL-10 mRNA levels in IFN-α-stimulated BALB/c and C57BL/6 splenocytes. The effect of ribavirin (RIB) at 5, 2, and 1 μM on IL-10 mRNA expression was assessed in total RNA isolated from IFN-α-stimulated BALB/c and C57BL/6 splenocytes. IL-10 and reporter β-actin mRNA expression was determined following RT-PCR and Southern blot analysis with specific 32P-labeled DNA probes. Data are shown as autoradiographs and densitometric ratios and are representative of two separate experiments. ∗, Ratio = the ratio of densitometry units (following background subtraction) from IL-10 over the reporter gene, β-actin. B, The induction of the costimulatory molecules, B7-1 (left panels) and B7-2 (right panels), by IFN-α in splenocytes from BALB/c (a, c, e, and g) and C57BL/6 (b, d, f, and h) mice is influenced by endogenous IL-10. In the top panels (denoted RIB, a, b, e, and f), B7-1 and B7-2 expression were compared in splenocytes from both mouse strains without treatment (dotted line), following 24-h induction with IFN-α alone (dashed lines), or in the presence of 2 μM ribavirin in vitro (solid line). In the bottom panels (denoted anti-IL-10 Ab, c, d, g, and h), B7-1 and B7-2 expression were compared in splenocytes from both mouse strains without treatment (solid line), following 24-h induction with IFN-α alone (dashed lines), in the presence of anti-IL-10 Ab (1 μg/ml) (bold solid line) or isotype-matched rat IgG1 (1 μg/ml) (dotted line). Any positive or negative effects of ribavirin (RIB) or anti-IL-10 Ab (IL-10 Ab) on B7-1 and B7-2 levels are shown as shaded regions. B7-1 and B7-2 levels were assessed by staining with FITC-labeled Abs to B7-1 (CD80) and B7-2 (CD86), followed by FACS analysis of gated CD3-negative cells. The data shown are representative of three experiments using five mice per test group.

FIGURE 5.

The differential regulation of IL-10, B7-1, and B7-2 expression by ribavirin can occur independent of antigenic stimulation. A, Ribavirin differentially modulates IL-10 mRNA levels in IFN-α-stimulated BALB/c and C57BL/6 splenocytes. The effect of ribavirin (RIB) at 5, 2, and 1 μM on IL-10 mRNA expression was assessed in total RNA isolated from IFN-α-stimulated BALB/c and C57BL/6 splenocytes. IL-10 and reporter β-actin mRNA expression was determined following RT-PCR and Southern blot analysis with specific 32P-labeled DNA probes. Data are shown as autoradiographs and densitometric ratios and are representative of two separate experiments. ∗, Ratio = the ratio of densitometry units (following background subtraction) from IL-10 over the reporter gene, β-actin. B, The induction of the costimulatory molecules, B7-1 (left panels) and B7-2 (right panels), by IFN-α in splenocytes from BALB/c (a, c, e, and g) and C57BL/6 (b, d, f, and h) mice is influenced by endogenous IL-10. In the top panels (denoted RIB, a, b, e, and f), B7-1 and B7-2 expression were compared in splenocytes from both mouse strains without treatment (dotted line), following 24-h induction with IFN-α alone (dashed lines), or in the presence of 2 μM ribavirin in vitro (solid line). In the bottom panels (denoted anti-IL-10 Ab, c, d, g, and h), B7-1 and B7-2 expression were compared in splenocytes from both mouse strains without treatment (solid line), following 24-h induction with IFN-α alone (dashed lines), in the presence of anti-IL-10 Ab (1 μg/ml) (bold solid line) or isotype-matched rat IgG1 (1 μg/ml) (dotted line). Any positive or negative effects of ribavirin (RIB) or anti-IL-10 Ab (IL-10 Ab) on B7-1 and B7-2 levels are shown as shaded regions. B7-1 and B7-2 levels were assessed by staining with FITC-labeled Abs to B7-1 (CD80) and B7-2 (CD86), followed by FACS analysis of gated CD3-negative cells. The data shown are representative of three experiments using five mice per test group.

Close modal

Next we postulated whether the effect of ribavirin on IFN-α-induced IL-10 expression in BALB/c and C57BL/6 mice could produce the same specific B7-1 and B7-2 profile observed in ribavirin-treated, DNFB-primed mice. Thus, we determined the effect of ribavirin on surface expression of B7-1 and B7-2 in IFN-α-stimulated splenocytes from both mouse strains. In IFN-α-stimulated BALB/c splenocytes, B7-2, but not B7-1 surface expression was augmented following ribavirin treatment (Fig. 5,B, RIB, a and e). In C57BL/6 splenocytes, however, IFN-α-induced B7-1 expression was enhanced, but B7-2 levels were unaffected (Fig. 5 B, RIB, b and f). These data show that the differential effects of ribavirin on IL-10, B7-1, and B7-2 expression were intrinsic to the mouse strain, as they could occur following an antigenic (DNFB) or nonantigenic (IFN-α) stimulus.

Next we determined whether the in vitro effect of ribavirin on IL-10 directly regulated B7-1 and B7-2 expression in these mouse strains in the same manner as we have shown in vivo. Thus, we determined the effect of the addition of anti-IL-10 Ab on IFN-α-induced B7-1 and B7-2 expression in BALB/c and C57BL/6 splenocytes. Anti-IL-10 Ab (1 μg/ml) treatment enhanced IFN-α-induced B7-2 expression in BALB/c mice (Fig. 5,B, IL-10 Ab, g), but inhibited B7-1 expression in C57BL/6 mice (Fig. 5,B, IL-10 Ab, d). No effect by anti-IL-10 Ab was seen on B7-1 expression in BALB/c (Fig. 5,B, IL-10 Ab, c) or on B7-2 expression in C57BL/6 mice (Fig. 5 B, IL-10 Ab, h). These in vitro data provide further evidence of a direct role for IL-10 in the regulation of B7 molecules. In BALB/c, IL-10 acted to inhibit B7-2 expression, whereas in C57BL/6, IL-10 promoted B7-1 expression.

We had previously shown that the nucleoside analogue, ribavirin, could modulate type 1 and type 2 cytokine expression (both at the level of protein and mRNA) in human T cells (10). The goal of the present study was to determine whether type 1 cytokine polarization by ribavirin could influence a type 1 cytokine-mediated immune response in vivo, the murine CHS response to DNFB. In this study, we demonstrated that promotion of type 1 cytokine expression by ribavirin in vitro in both unprimed or DNFB-primed BALB/c and C57BL/6 mouse LNC was predictive of in vivo type 1 cytokine profiles. However, the influence of ribavirin on type 2 cytokine profiles differed between mouse strains both in vitro and in vivo. Although ribavirin suppressed IL-4 and IL-5 in DNFB-primed mice from both strains, IL-10 expression was suppressed only in BALB/c mice. This difference in ribavirin-mediated IL-10 expression manifested opposing CHS responses to DNFB in these two strains of mice. In BALB/c mice, ribavirin treatment enhanced CHS responses to DNFB. This is the first demonstration that administration of ribavirin can enhance a non-virally-induced immune response in vivo. Previous effects of ribavirin on virus-independent immune responses have always led to immunosuppression (19, 20, 21, 22). In contrast to BALB/c mice, ribavirin treatment in C57BL/6 mice impaired CHS responses to DNFB, an effect that was associated with elevated IL-10 expression. Therefore, the divergent outcomes on CHS responses between mouse strains following ribavirin treatment indicated that modulation of CHS responses by ribavirin can be influenced by factors other than type 1 cytokine polarization. However, the precise mechanism by which ribavirin regulates murine CHS responses and the importance of the role of IL-10 still needed to be clarified.

Previous studies have shown that IL-10, but not IL-4, is the natural regulator of irritant and CHS responses in skin (14). Enhanced expression of IL-10 was observed in ribavirin-treated C57BL/6 mice, suggesting that impaired CHS responses in these mice were the result of recruitment of the regulatory pathways of CHS. However, several lines of evidence suggest that IL-10 is necessary, but not sufficient to induce the suppression of CHS responses by ribavirin in C57BL/6 mice. First, during a normal CHS response in the absence of ribavirin, our data showed that upon challenge, IL-10 levels were significantly higher in DNFB-primed BALB/c than in DNFB-primed C57BL/6 mice, but the magnitude of the CHS response was greater in BALB/c than C57BL/6. Second, the previously reported suppressive effect of IL-10 on normal CHS responses occurs only upon administration of IL-10 at least 12–24 h before hapten challenge (elicitation phase), suggesting an indirect mechanism (23). Finally, the inhibitory effects of IL-10 on CHS and in psoriasis are not due to a direct effect on T cells, but rather the inhibition of accessory cell function (23, 24). Collectively, these data suggest that IL-10 influences CHS responses indirectly and possibly through interaction with APCs.

One critical interaction in the initiation of an immune response to Ag is the costimulatory signal delivered following the engagement of CD28 on T cells and either B7-1 or B7-2 on APCs. This event facilitates T cell activation following TCR recognition of processed Ag presented in association with MHC molecules by APC. The magnitude of an Ag response can be influenced dramatically by the type and strength of costimulatory signal delivered by APC following the initial T cell response to Ag. Interestingly, IL-10 is a known regulator of the costimulatory molecules, B7-1 and B7-2. In recent reports, IL-10 has been shown to up-regulate B7-1 in monocytes and suppress B7-2 expression in monocytes, Langerhans cells, and dendritic cells (15, 16, 17). Based on this evidence, we postulated that following ribavirin treatment, IL-10 could elicit its effects on CHS responses via a direct effect on the expression of costimulatory molecules, B7-1 and B7-2, on APCs. Our data showed that the amplified CHS responses in BALB/c mice following ribavirin treatment were associated with enhanced B7-2 expression in LNC. Moreover, administration of IL-10 in vivo in ribavirin-treated DNFB-primed BALB/c mice could reverse the ribavirin-mediated enhancement of inflammatory ear responses, while concomitantly elevating B7-1 and suppressing B7-2 expression in LNC. In contrast, ribavirin treatment in C57BL/6 mice results in suppressed CHS responses, and this was associated with enhanced B7-1 expression. Administration of anti-IL-10 Ab in vivo in ribavirin-treated DNFB-primed C57BL/6 mice could neutralize the IL-10-induced B7-1 expression and result in a normalized CHS response to DNFB. Collectively, these data show that the effect of ribavirin on B7-1 levels in C57BL/6 mice and on B7-2 in BALB/c by ribavirin is directly regulated by IL-10. Interestingly, impaired CHS responses and the enhancement of B7-1 expression could be induced by administration of IL-10 directly in vivo to DNFB-primed mice from either strain, demonstrating a direct effect of IL-10 on both anti-inflammatory and costimulatory functions. These data also show that the effect of ribavirin on CHS response is intrinsic to each mouse strain.

To determine possible mechanisms by which modulation of B7-1 and B7-2 expression, following ribavirin treatment, can regulate CHS responses, we needed to first assess the individual roles of B7-1 and B7-2 molecules in immune regulation. Some studies have suggested that these molecules differ in their influences on the development of type 1 and type 2 cytokines in autoimmunity (25, 26, 27). These studies gave conflicting observations, and a clear relationship between B7-1/B7-2 and Th1/Th2 development has not yet been conclusively established. However recent studies, using Abs to B7-1 and B7-2 in vivo, support the view that the functional roles of B7-1 and B7-2 may differ in various in vivo situations. Specifically, engagement of either B7-1 or B7-2 by CD28 following recognition of a particular Ag can provide either an appropriate costimulatory signal (which results in immunity) or an inappropriate costimulatory signal (which results in tolerance or ignorance). For example, B7-2 is the predominant costimulatory signal in experimental autoimmune encephalomyelitis and in Ag-induced airway hyperresponsiveness in mice, and following infection of Leishmania major in both BALB/c and C57BL/6 mice (25, 28, 29). Conversely, B7-1 signaling predominates in the pathogenesis of a murine model of insulin-dependent diabetes mellitus and is necessary in the elimination of Crypotococcus neoformans infections in mice (26, 30). Interestingly, CHS responses to DNFB have been shown to be inhibited by anti-B7-2 but not anti-B7-1 Ab administered at the elicitation (priming) stage (31, 32). This demonstrates that B7-2 is also the predominant costimulatory signal for CHS responses to DNFB. These data collectively support the view that, in addition to type 1 cytokine bias, enhancement of CHS responses by ribavirin in BALB/c mice directly results from the ability of ribavirin to suppress IL-10 and to induce APC to express greater B7-2, an appropriate costimulatory ligand for CHS response to DNFB. Conversely, ribavirin enhances IL-10 expression in C57BL/6 mice, which leads to induction on APCs of B7-1, an inappropriate or regulatory costimulatory signal. It is important to note that our studies show that the outcome of the CHS response was dependent on the enhanced expression of B7-1 or B7-2, in respective mouse strains. This is a unique observation because ribavirin treatment affected CHS responses by inducing the preferential up-regulation of a B7 molecule. In previous studies, CHS responses were modulated by the specific inhibition of engagement of a B7 molecule using anti-B7-1 and anti-B7-2 Abs (31, 32).

Although our findings suggest that up-regulation of costimulatory molecules following ribavirin treatment affects the magnitude of CHS responses, we have yet to address the issue of why the expression of IL-10, the principal mediator of these effects, is apparently regulated differently following ribavirin treatment in BALB/c and C57BL/6 mice. The factors involved could be intrinsic to the mouse strain. For example, haplotype differences (H-2d for BALB/c and H-2b for C57BL/6) permit Ag presentation by different MHC molecules or presentation of alternatively processed Ag. A difference in the strength of the Ag-MHC interaction or the type of Ag presented may lead to opposing signals, one that is anti-inflammatory and can switch on regulatory mechanisms including up-regulating IL-10 and another that is proinflammatory and suppresses IL-10. To address this, we determined the effect of ribavirin following treatment of splenocytes with a nonantigenic stimulus, IFN-α, a known inducer of IL-10 (33) and B7-1 and B7-2 expression (34, 35). IFN-α, like Ag priming with DNFB, did induce IL-10 in both mouse strains. More importantly, the effect of ribavirin on IFN-α-induced IL-10, B7-1, and B7-2 expression in splenocytes from BALB/c and C57BL/6 paralleled those observed previously with DNFB-primed LNC. Also, treatment of IFN-α-stimulated splenocytes with anti-IL-10 Ab elicited the same effect as ribavirin in BALB/c mice (enhanced B7-2 levels), but inhibited B7-1 expression in C57BL/6 mice, a reversal of the effect found following ribavirin treatment. These Ag-independent in vitro effects paralleled the findings we have shown earlier using an Ag-specific in vivo immune response.

It is noteworthy that IFN-α levels were unaffected by ribavirin and that ribavirin did not affect expression of B7 molecules in the absence of IFN-α. The synergistic effect of ribavirin on IFN-α-induced B7-1 and B7-2 expression is also noteworthy, particularly as the combination of ribavirin and IFN-α has been shown to have greater efficacy over monotherapy with either drug in the treatment of hepatitis C (36). Furthermore, they suggest that the effect of ribavirin on CHS responses is 1) intrinsic to each mouse strain, 2) can occur with or without Ag priming, and 3) does not involve modulation of IFN-α levels.

Differential regulation of IL-10 levels during CHS by ribavirin in BALB/c and C57BL/6 mice may be the result of opposing effects on certain regulatory pathways of immune suppression. Previous studies have shown that C57BL/6 are susceptible to UVB-induced immunosuppression of CHS responses to DNFB, whereas BALB/c are typical of a resistant strain (37). Evidence from studies using IL-10-deficient mice or anti-IL-10 Abs has identified IL-10 as the primary suppressor involved in the inhibition of CHS responses following UVB exposure (38). Furthermore, the susceptibility gene for UVB-induced immunosuppression has been mapped to the Uvs 1 locus in mice (39). Thus, the differential effects of ribavirin on CHS responses in BALB/c and C57BL/6 mice may be attributable to genetic differences and involve activation (C57BL/6) or suppression (BALB/c) of an IL-10-mediated regulatory pathway of CHS similar to that observed following UVB exposure. These data indicate that a complex regulatory network controlling CHS can influence the nature of the response to ribavirin. Two important observations underscore the need for future studies to focus on the identification of the precise point at which antagonism of ribavirin-mediated effects on CHS occurs: 1) The regulatory pathways that induce IL-10 and limit CHS responses in C57BL/6 mice appear to overwhelm the apparent type 1 cytokine bias induced by ribavirin, and 2) regulatory pathways such as UVB-induced immunosuppression are a complete contrast to the effects of type 1 cytokine polarization, often leading to the induction of a Th1 to Th2 cytokine switch (40) and impaired delayed-type hypersensitivity responses, CTL activity, and skin graft rejection (41).

In summary, we have shown that CHS responses to contact allergen following ribavirin treatment in BALB/c and C57BL/6 mice were dependent not only on type 1 cytokine polarization, but regulation of IL-10 expression and appropriate costimulatory signaling. These effects of ribavirin, which are the result of differential effects of ribavirin on IL-10 levels, act via the promotion of the expression of specific costimulatory molecules. The type 1 cytokine bias also elicited by ribavirin was ineffective when IL-10 expression was enhanced, and this resulted in an opposing in vivo outcome. These data presented in this work suggest that ribavirin can alter the balance between immunity and immune suppression on multiple levels. These observations should be considered in the future evaluation of the therapeutic potential of ribavirin as a type 1-inducing agent in the treatment of immune-related disorders.

We thank Drs. S. B. A. Cohen, G. Benichou, and D. Averett for critical evaluation of this manuscript, and J. Avalos for collection of blood from normal donors and for excellent technical assistance.

2

Abbreviations used in this paper: CHS, contact hypersensitivity; DNFB, dinitrofluorobenzene; LNC, lymph node cell.

1
Mosmann, T. R., H. M. Cherwinski, M. W. Bond, M. A. Giedlin, R. L. Coffmann.
1986
. Two types of murine helper T cell clone. I. Definition according to profiles of lymphokine activities and secreted proteins.
J. Immunol.
136
:
2348
2
Cherwinski, H. M., J. H. Schumacher, K. D. Brown, T. R. Mosmann.
1987
. Two types of mouse helper T cell clone: further differences in lymphokine synthesis between Th1 and Th2 clones revealed by RNA hybridization, functionally monospecific bioassays and monoclonal antibodies.
J. Exp. Med.
166
:
1229
3
Del Prete, G. F., M. De Carli, C. Mastromauro, R. Biaggiotti, D. Macchia, P. Falagiani, M. Ricci, S. Romagnani.
1991
. Purified protein derivative of Mycobacteriumtuberculosis and excretory-secretory antigen(s) of Toxocara canis expand in vitro human T cells with stable and opposite (type 1 T helper or type 2 T helper) profile of cytokine production.
J. Clin. Invest.
88
:
346
4
Mosmann, T. R., S. Sad.
1996
. The expanding universe of T cell subsets: Th1, Th2 and more.
Immunol. Today
17
:
138
5
Carter, L., R. W. Dutton.
1996
. Type 1 and Type 2: a fundamental dichotomy for all T cell subsets.
Curr. Opin. Immunol.
8
:
336
6
Chow, Y. H., B. L. Chiang, Y. L. Lee, W. K. Chi, W. C. Lin, Y. T. Chen, M. H. Tao.
1998
. Development of Th1 and Th2 populations and the nature of immune responses to hepatitis B virus DNA vaccines can be modulated by codelivery of various cytokine genes.
J. Immunol.
160
:
1320
7
Van Elsas, A., C. Aarnoudse, C. E. van der Minne, C. W. van der Spek, N. Brouwenstijn, S. Osanto, P. I. Schrier.
1997
. Transfection of IL-2 augments CTL response to human melanoma cells in vitro: immunological characterization of a melanoma vaccine.
J. Immunother.
20
:
343
8
Hall, C. B., J. T. McBride, E. E. Walsh, D. M. Bell, C. L. Gala, S. Hildreth, L. G. Ten Eyck, W. G. Hall.
1983
. Aerosolized ribavirin treatment of infants with respiratory syncytial viral infection.
N. Engl. J. Med.
308
:
1443
9
Dusheiko, G., J. Main, H. Thomas, O. Reichard, C. Lee, A. Dhillon, S. Rassam, A. Fryden, H. Reesink, M. Bassendine, et al
1996
. Ribavirin treatment for patients with chronic hepatitis C: results of a placebo-controlled study.
J. Hepatol.
25
:
591
10
Tam, R. C., B. Pai, J. Bard, C. Lim, D. R. Averett, U. T. Phan, T. Milovanovic.
1999
. Ribavirin polarizes human T cell responses towards a type 1 cytokine profile.
J. Hepatol.
30
:
376
11
Wei, L., K. M. Muller, J. H. Saurat, C. Hauser.
1993
. Lymphokine profiles in contact sensitivity induced by dinitrofluorobenzene and tolerance induced by dinitrothiocyanobenzene.
Arch. Dermatol. Res.
284
:
427
12
Xu, H., N. A. Dilulio, R. L. Fairchild.
1996
. T cell populations primed to hapten sensitization in contact hypersensitivity are distinguished by polarized patterns of cytokine production: interferon-γ (Tc1) effector CD8+ T cells and interleukin 4/interleukin 10-producing (Th2) negative regulatory CD4+ T cells.
J. Exp. Med.
183
:
1001
13
Ishii, N., K. Takahashi, H. Nakajima, S. Tanaka, P. W. Askenase.
1994
. DNFB contact sensitivity in BALB/c and C3H/He mice.
J. Invest. Dermatol.
102
:
321
14
Berg, D. J., M. W. Leach, R. Kuhn, K. Kajewsky, W. Muller, N. J. Davidson, and D. Rennick. 1995. Interleukin 10 but not interleukin 4 is a natural suppressant of cutaneous inflammatory responses J. Exp Med. 182:99.
15
Creery, W. D., F. Diaz-Mitoma, L. Filion, A. Kumar.
1996
. Differential modulation of B7-1 and B7-2 isoform expression on human monocytes by cytokines which influence the development of T helper cell phenotype.
Eur. J. Immunol.
26
:
1273
16
Buelens, C., F. Willems, A. Delvaux, G. Pierard, J. P. Delville, T. Velu, M. Goldman.
1995
. Interleukin-10 differentially regulates B7-1 (CD80) and B7-2 (CD86) expression on human peripheral blood dendritic cells.
Eur. J. Immunol.
25
:
2668
17
Kawamura, T., M. Furue.
1995
. Comparative analysis of B7-1 and B7-2 expression in Langerhans cells: differential regulation by T helper type 1 and T helper type 2 cytokines.
Eur. J. Immunol.
25
:
1913
18
Schwarz, A., S. Grabbe, H. Riemann, Y. Aragane, M. Simon, S. Manon, S. Andrade, T. A. Luger, A. Zlotnik, T. Schwarz.
1994
. In vivo effects of interleukin-10 on contact hypersensitivity and delayed-type hypersensitivity reactions.
J. Invest. Dermatol.
103
:
211
19
Peavy, D. L., W. C. Koff, D. S. Hyman, V. Knight.
1980
. Inhibition of lymphocyte proliferative responses by ribavirin.
Infect. Immun.
29
:
583
20
Powers, C. N., D. L. Peavy, V. Knight.
1982
. Selective inhibition of functional lymphocyte subpopulations by ribavirin.
Antimicrob. Agents Chemother.
22
:
108
21
Peavy, D. L., C. N. Powers, V. Knight.
1981
. Inhibition of murine plaque-forming responses in vivo by ribavirin.
J. Immunol.
126
:
861
22
Marquardt, D. L., H. E. Gruber, L. L. Walker.
1987
. Ribavirin inhibits mast cell mediator release.
J. Pharmacol. Exp. Ther.
240
:
145
23
Kondo, S., R. C. McKenzie, D. N. Sauder.
1994
. Interleukin-10 inhibits the elicitation phase of allergic contact hypersensitivity.
J. Invest. Dermatol.
103
:
811
24
Mitra, R. S., T. A. Judge, F. O. Nestle, L. A. Turka, B. J. Nickoloff.
1995
. Psoriatic skin-derived dendritic cell.
J. Immunol.
154
:
2668
25
Kuchroo, V., M. Prabhu Das, J. A. Brown, A. M. Ranger, S. S. Zamvil, R. A. Sobel, H. L. Weiner, N. Nabavi, L. H. Glimcher.
1995
. B7-1 and B7-2 costimulatory molecules differentially activate the Th1/Th2 developmental pathways: application to autoimmune disease therapy.
Cell
80
:
707
26
Lenschow, D. J., S. C. Ho, H. Sattar, L. Rhee, G. Gray, N. Nabavi, K. C. Herold, J. A. Bluestone.
1995
. Differential effects of anti-B7-1 and anti-B7-2 monoclonal antibody treatment on the development of diabetes in the non-obese diabetic mouse.
J. Exp. Med.
181
:
1145
27
Corry, D. B., S. L. Reiner, P. S. Linsley, R. M. Locksley.
1994
. Differential effects of blockade of CD28–B7 on the development of Th1 or Th2 effector cells in experimental leishmaniasis.
J. Immunol.
153
:
4142
28
Brown, J. A., R. G. Titus, N. Nabavi, L. H. Glimcher.
1996
. Blockade of CD86 ameliorates Leishmania major infection by down-regulating the Th2 response.
J. Infect. Dis.
174
:
1303
29
Tsuyuki, S., J. Tsuyuki, K. Einsle, M. Kopf, A. J. Coyle.
1997
. Costimulation through B7-2 (CD86) is required for the induction of a lung mucosal T helper cell 2 (Th2) immune response and altered airway responsiveness.
J. Exp. Med.
185
:
1671
30
Vecchiarelli, A., C. Monari, C. Retini, D. Pietrella, B. Palazzetti, L. Pitzurra, A. Casadevall.
1998
. Cryptococcus neoformans differently regulates B7-1 (CD80) and B7-2 (CD86) expression on human monocytes.
Eur. J. Immunol.
28
:
114
31
Reiser, H., E. E. Schneeberger.
1996
. Expression of B7-1 and B7-2 in hapten-induced contact sensitivity.
Eur. J. Immunol.
26
:
880
32
Nuriya, S., H. Yagita, K. Okumara, M. Azuma.
1996
. The differential role of CD86 and CD80 costimulatory molecules in the induction and the effector phases of contact hypersensitivity.
Int. Immunol.
8
:
917
33
Hermann, P., M. Rubio, T. Nakajima, G. Delespesse, M. Sarfati.
1998
. IFN-α priming of human monocytes differentially regulates Gram-positive and Gram-negative bacteria-induced IL-10 release and selectively enhances IL-12p70, CD80 and MHC I expression.
J. Immunol.
161
:
2011
34
Chakrabarti, D., B. Hultgren, T. A. Stewart.
1996
. IFN-α induces autoimmune T cells through the induction of intracellular adhesion molecule-1 and B7-2.
J. Immunol.
157
:
522
35
Tsukada, N., S. Aoki, S. Maruyuma, K. Kishi, M. Takahashi, Y. Aizawa.
1997
. The heterogeneous expression of CD80, CD86 and other adhesion molecules on leukemia and lymphoma cells and their induction by interferon.
J. Exp. Clin. Cancer Res.Please verify journal title; it is not listed in our sources.
16
:
171
36
Chemello, L., L. Cavelletto, E. Bernardinello, M. Guido, P. Pontisso, A. Alberti.
1995
. The effect of interferon α and ribavirin combination therapy in naive patients with chronic hepatitis C.
J. Hepatol.
23
:
8
37
Noonan, F. P., H. A. Hoffman.
1994
. Susceptibility to immunosuppression by ultraviolet B radiation in the mouse.
Immunogenetics
39
:
29
38
Beizzert, S., J. Hosoi, R. Kuhn, K. Rajewsky, W. Muller, R. D. Granstein.
1996
. Impaired immunosuppressive responses to ultraviolet radiation in interleukin-10-deficient mice.
J. Invest. Dermatol.
107
:
553
39
Noonan, F. P., H. A. Hoffman.
1994
. Control of UVB immunosuppression in the mouse by autosomal and sex-linked genes.
Immunogenetics
40
:
247
40
Brown, E. L., J. M. Rivas, S. E. Ullrich, C. R. Young, S. J. Norris, M. L. Kripke.
1995
. Modulation of immunity to Borrelia burgdorferi by ultraviolet irradiation: differential effect on Th1 and Th2 immune responses.
Eur. J. Immunol.
25
:
3017
41
Tamaki, K., M. Iijima.
1989
. The effect of ultraviolet B irradiation on delayed-type hypersensitivity, cytotoxic T lymphocyte activity, and skin graft rejection.
Transplantation
47
:
372