The development of hepatic fibrosis and portal hypertension is the principal cause of morbidity and mortality in schistosomiasis mansoni. Nevertheless, relatively little is known about the mechanisms that lead to excessive collagen deposition during infection with Schistosoma mansoni. In the murine model, infection leads to significant egg-induced granuloma formation, tissue eosinophilia, and hepatic fibrosis. The pathology has been linked to dominant type 2 cytokine expression, and our recent studies showed that sensitizing animals to egg Ags in combination with IL-12, before infection, led to a highly significant reduction in egg-induced immunopathology. In this study, we demonstrate that in contrast with egg/IL-12-sensitized animals that showed marked decreases in pathology, mice similarly sensitized but depleted of IFN-γ, IL-12, or TNF-α at the time of egg laying developed granulomas that were similar to the non-IL-12-treated control group. Although all three anti-cytokine-treated groups exhibited a dominant type 1 response in lymph node cells restimulated ex vivo, the expression of type 2 cytokine mRNA was markedly restored at the site of granuloma formation, which suggests that all three cytokines are required to maintain the suppressed type 2 pattern. Moreover, egg/IL-12-sensitized mice depleted of IFN-γ or IL-12 displayed a partial reduction in IFN-γ production, suggesting that multiple type 1 cytokines were required to maintain polarized type 1 responses to chronic type 2-inducing stimuli. Together, these data reveal key roles for IFN-γ, IL-12, and TNF-α in the protective effects mediated by this IL-12-based vaccine to prevent pathology.
Interleukin-12, a 70-kDa disulfide-linked heterodimer, secreted by a variety of APCs in response to bacteria, bacterial products, and intracellular parasites (1), is a key regulatory cytokine that drives cell-mediated immunity by inducing multiple lymphocyte subsets to produce IFN-γ (2). Over the past several years, it has become apparent that IL-12 is essential for promoting protective immune responses against infectious organisms including intracellular bacteria (3) and protozoan parasites (4, 5, 6, 7), as well as assisting in immune-mediated antitumor and antimetastatic activities (8). The prophylactic ability of IL-12 was additionally realized by Afonso et al. (9), who demonstrated that BALB/c mice immunized with IL-12 plus soluble leishmanial Ags developed leishmanial-specific CD4+ Th1 cells that completely rendered the normally susceptible animals resistant to Leishmania major challenge infection. Thus, all of these studies suggest that IL-12 may aid in the immunotherapeutic or immunoprophylactic treatments of clinically diverse diseases.
Schistosomiasis, a disease that currently affects 200 million people worldwide, is often effectively treated by chemotherapies. In spite of these chemotherapeutic successes, many world health agencies agree that the development of an antischistosomiasis vaccine should be aggressively pursued. This is due, in part, to the recent identification of possible praziquantel-resistant Schistosoma mansoni field isolates (10, 11). Additionally, evidence suggestive of age-dependent immunity (12) indicates that the development of an antischistosome vaccine is feasible and could lead to reduced morbidity and mortality. One current vaccine strategy is to suppress the overall pathology associated with egg deposition in the bladder, liver, or intestines. We have recently shown that IL-12 is a highly efficacious adjuvant for this “antipathology” vaccine strategy (13).
Initially, we demonstrated that IL-12 could be used prophylactically in combination with egg Ags to reduce subsequent egg-induced lung pathology (14). More importantly, a similar IL-12-based antipathology vaccine reduced both granuloma formation and fibrosis in the livers of infected animals (13). The reduction in fibrosis was accompanied by a switch in the normal Th2 response to a limited Th1 response. We observed marked increases in type 1 cytokines including IFN-γ, IL-12, and TNF-α, and corresponding decreases in type 2 cytokines such as IL-4, IL-5, and IL-13. We hypothesized that the reduction in fibrosis/granuloma formation in egg/IL-12-sensitized mice was possibly due to both the enhanced production of cytokines involved in suppressing collagen synthesis (IFN-γ) (15) and the repressed production of collagen-inducing cytokines (IL-4) (16).
In the current study, we wanted to elucidate the mechanism and describe the key components of this IL-12-based antipathology vaccine. Specifically, we examined whether the reduction in hepatic pathology induced by the egg/IL-12 sensitization protocol was dependent upon the increased expression of IFN-γ, IL-12, or TNF-α. To address their contribution to the pathology-reducing effect, egg/IL-12-sensitized mice were infected and then treated with neutralizing Abs to IFN-γ, IL-12, and TNF-α at the time of schistosome egg laying. We examined the effects on granuloma formation, recruitment of eosinophils, development of hepatic fibrosis, and the evolving cytokine response both in vitro and in vivo. The results from this study demonstrate that all three type 1 cytokines are important for the antipathology effect induced by Ag/IL-12 sensitization. In addition, these data provide evidence that type 1 responses generated against strong type 2 inducing stimuli are potentially unstable and remain highly dependent upon endogenous IFN-γ, IL-12, and TNF-α for their maintenance.
Materials and Methods
Mice, parasites, and Ag preparations
Female 42-day-old C57BL/6 mice (10 per group) were obtained from Charles River Laboratories (Raleigh, NC). All mice were housed in an National Institutes of Health American Association for the Accreditation of Laboratory Animal Care-approved animal facility. Cercariae of a Puerto Rican strain of S. mansoni (Biomedical Research Institute, Rockville, MD) were obtained from infected Biomphalaria glabrata snails (Biomedical Research Institute). Soluble egg Ag (SEA)3 and soluble worm Ag preparations (SWAP) were derived from homogenized eggs and adult parasites as previously described (13).
Immunizations and infections
The sensitization of mice to S. mansoni egg Ags in the presence of rIL-12 (generously provided by Genetics Institute, Cambridge, MA) was performed essentially as previously described (13). Briefly, S. mansoni eggs were isolated from the livers of infected mice (Biomedical Research Institute), and 5000 eggs/animal were injected i.p. on three occasions separated by 2-wk intervals. Animals were also injected i.p. with rIL-12 (0.25 μg per dose) on days 0, 1, 2, 3, and 5 after each egg exposure. Mice were infected 4 wk after the last egg/IL-12 exposure by percutaneous challenge of tail skin for 40 min in water containing between 20 and 25 cercariae. At the onset of egg production (week 5), mice were treated twice weekly with neutralizing mAbs against IFN-γ (mAb XMG1.6), TNF-α (mAb XT22.11), IL-12 (mAb C17.8.20), or isotype control β galactosidase (βGal) (mAb GL113) at 1 mg/injection. Sera were collected from mice before they were sacrificed by i.p. administration of sodium pentobarbital (18 mg/mouse, Sigma, St. Louis, MO) on week 8. Tissues were then collected for the remaining studies. Control mouse groups used in these studies were either uninfected, egg-sensitized without rIL-12, or infected without treatment. Previous studies on mice pretreated with rIL-12 alone have demonstrated no significant effects on pathologic parameters after infection (13).
Histopathology and fibrosis measurement
The collagen content of the liver, determined as hydroxyproline, was measured as described previously (17). Approximately half of the liver was fixed in Bouin-Hollande solution and histologic sections were processed and stained with Giemsa (Histo-Path of America, Clinton, MD). The diameters and eosinophil content of granulomas (30/mouse) surrounding single, mature, and viable eggs were measured by using an ocular micrometer (Leica, Columbia, MD), and the volume of each granuloma was calculated assuming a spherical shape.
Lymphocyte culture and cytokine assays
For in vitro cytokine measurements, mesenteric lymph nodes were removed aseptically at week 8 after infection, and single-cell suspensions were prepared. Mesenteric nodes were pooled from three animals per group, and cells were plated in 24-well tissue culture plates at a final concentration of 3 × 106 cells per ml in RPMI 1640 supplemented with 2 mM glutamine, 25 mM HEPES, 10% FCS, 50 μM 2-ME, penicillin, and streptomycin. Cultures were incubated at 37°C in an atmosphere of 5% CO2. Cells were stimulated with SEA at 20 μg/ml, SWAP at 50 μg/ml, or medium alone. Supernatant fluids were harvested at 72 h and assayed for cytokine activity. IFN-γ, IL-5, IL-12, and IL-10 were measured by specific two-site ELISA as previously described (13). Cytokine levels were calculated using standard curves constructed using recombinant murine cytokines.
RT-PCR detection of cytokine mRNAs
Two 25-mg portions of each liver were combined and homogenized in 1 ml RNA STAT-60 using a tissue polytron (Omni International, Waterbury, CT), and total RNA was isolated as recommended by the manufacturer. The RNA was resuspended in diethylpyrocarbonate-treated water and quantitated spectrophotometrically. A RT-PCR procedure was performed as described (18) to determine relative quantities of mRNA for IFN-γ, IL-4, IL-5, TNF-α, IL-12p40, and hypoxanthine phosphoribosyltransferase (HPRT). The primers and probes for all genes were previously published (18, 19). The PCR conditions and cycle number were strictly defined for each cytokine primer pair such that a linear relationship between input RNA and final PCR product was obtained. Positive and negative controls were included in each assay to confirm that only cDNA PCR products were detected and that none of the reagents were contaminated with cDNA or extraneous PCR products. The amplified DNA was analyzed by electrophoresis, Southern blotting, and hybridization with cytokine-specific probes. The chemiluminescent signals were quantified using a 600 ZS scanner (Microtek International, Torrance, CA). The amount of PCR product was determined by comparing the ratio of cytokine-specific signal density to that of HPRT-specific signal density for individual samples (five mice per group). Arbitrary densitometric units for individual samples were subsequently multiplied by a factor of 100 and compared with control mice (uninfected mouse liver). Amplification of HPRT served as an internal control for the amount of RNA and cDNA from each sample.
Measurement of SEA-specific Ab responses
For assesment of serum Ig, serum was collected at time of sacrifice (week 8 postchallenge). Immulon 4 (Dynatech Laboratories, Chantilly, VA) microtiter plates were coated overnight at 4°C with SEA (1 μg in 50 μl/well) diluted in PBS. Plates were blocked with 200 μl 5% nonfat dry milk/PBS for 2 h at 37°C. The blocking solution was aspirated and the wells washed six times with PBS/0.05% Tween-20 (Sigma). Individual mouse serum was serially diluted 1/100 to 1/102500 in 1% BSA/PBS, and 50 μl was added to appropriate wells. Plates were incubated at 37°C for 90 min and then washed six times with PBS/0.05% Tween-20. Fifty microliters of isotype-specific horseradish peroxidase-conjugated rabbit anti-mouse Abs in 1% BSA/PBS diluted at 1/1000 (measurement of IgG1, IgG2a, IgG2b, and total IgG/A/M, Zymed, San Francisco, CA) were added to the wells and incubated at 37°C for 2 h. Wells were again washed six times with PBS/0.05% Tween-20 and 100 μl of (2,2′-azino-di(3-ethyl-benzthiazoline sulfonate)) (ABTS:H2O2 substrate (Kirkegaard & Perry Laboratories, Gaithersburg, MD)) was added and the reactions developed in the dark at room temperature for 20 to 30 min. Absorbance at 405 nm was determined using a Vmax Kinetic Microplate Reader (Molecular Devices, Palo Alto, CA). Specific SEA-isotype titers were calculated by the product of absorbance and the reciprocal of the sera dilution from a point in the linear portion of the dilution curve.
Total serum IgE Abs were quantitated by ELISA using a protocol provided by PharMingen (San Diego, CA). Briefly, plates were coated with anti-mouse IgE capture mAb from clone R35-72 in 0.1 M NaHCO3, pH 8.2, overnight at 4°C. The secondary mAb was a biotinylated anti-mouse IgE from clone R35-92, and the streptavidin-peroxidase reagent was diluted 1/1000 in 1% BSA/PBS. A purified mouse IgE from clone IgE-2 (PharMingen) was used as the control standard.
Worm number and the number of eggs in the tissue and feces were determined as previously described (17).
Values for secreted cytokine proteins, semiquantitative RT-PCR, serum ELISA data, granuloma volume, and the percentage of eosinophils in granulomas were compared using Student’s two-tailed t test. Hepatic fibrosis was compared by analysis of covariance using the log of total liver eggs as the covariate and the log of hydroxyproline per egg. Values of p < 0.05 were regarded as significant. Two separate experiments were performed for all data presented.
Neutralization of IFN-γ, TNF-α, or IL-12 restores granuloma formation in infected egg/IL-12-sensitized mice
Our previous studies have shown that sensitizing mice to egg Ags in combination with IL-12 before infection led to increased IFN-γ, IL-12, and TNF-α expression in infected mice (13). To assess whether the reduced hepatic pathology in egg/IL-12-sensitized animals is directly influenced by these cytokines, C57BL/6 mice were presensitized i.p. with eggs/IL-12, infected, and depleted of IFN-γ, TNF-α, or IL-12 from the start of egg laying up to the time of sacrifice (Fig. 1). Hepatic pathology in the various cytokine-depleted mice was compared with three different infected control groups as well as with uninfected animals. The infected control groups included mice that were sensitized with only saline, eggs alone, or with eggs and IL-12 followed by a control mAb (anti-βGal). As expected, control mAb-treated egg/IL-12-sensitized mice developed smaller hepatic granulomas (Fig. 2,A) that contained significantly fewer eosinophils (Fig. 2,B) and displayed marked decreases in fibrosis, as assessed by hydroxyproline levels (Fig. 2,C) when compared with the saline-treated or egg-sensitized controls. Mice sensitized with eggs alone also showed reduced levels of fibrosis; however, the changes were less marked than in the egg/IL-12-sensitized group, which was consistent with previous findings (13). Interestingly, granuloma development was restored to control levels in egg/IL-12-sensitized mice by depletion of endogenous IFN-γ, IL-12, or TNF-α (Fig. 2,A). The granuloma eosinophil composition was also significantly restored in IFN-γ- or IL-12-depleted mice, while depletion of TNF-α had less of an effect (Fig. 2,B). Moreover, the level of hepatic fibrosis was also significantly increased in all three cytokine-depleted groups (Fig. 2 C). In all cases however, anti-IFN-γ mAb treatment had the largest effect.
No difference in infection intensities or egg output was observed in any infected group
Cytokine depletion had no effect on the number of eggs/worm pairs in the tissues or in the number of eggs/worm pairs passed in the feces during the 24 h before sacrifice in any of the examined groups (data not shown). Additionally, the egg- or egg/IL-12-sensitized groups did not differ from the infected untreated mice in these parasitologic criteria (data not shown). Analysis was performed by one-way ANOVA and by Student’s t test between appropriate groups.
Neutralizing endogenous IFN-γ or IL-12 decreases the type 1 cytokine response in egg/IL-12-sensitized mice but fails to significantly augment the type 2 response in the lymph nodes or spleen
To determine whether the changes in egg-induced pathology after neutralization of IFN-γ, IL-12, or TNF-α were accompanied by alterations in the type 1/type 2 cytokine profile, mesenteric lymph nodes and spleens were isolated from the animals at the time of sacrifice and analyzed in vitro for their cytokine-producing potential. Separate pools of mesenteric lymph nodes and individual spleens were processed at week 8, and the cells were restimulated in vitro with either SEA or SWAP. Seventy-two-hour culture supernatants were then assayed by ELISA for IL-5 and IL-10 as markers of a type 2 response (Fig. 3) and IFN-γ and IL-12 (Fig. 4) as indicators of a type 1 cytokine pattern. Mesenteric lymph node cultures from both saline-treated and egg-sensitized mice produced predominantly IL-5 and IL-10 in response SEA or SWAP (Figs. 3,A and 3B). Little or no IFN-γ was detected in these cultures, confirming that a highly polarized type 2 response was established in both control-infected groups (Fig. 3,B). In contrast to these data, the egg/IL-12-sensitized-control mAb-treated group (anti-βGal) displayed a highly polarized type 1 cytokine pattern (Fig. 4), which was consistent with previous findings (13). Similar data were obtained with splenocytes (data not shown). Despite displaying a marked increase in pathology after neutralization of IFN-γ, IL-12, or TNF-α, a statistically significant increase in IL-5 or IL-10 by mesenteric lymph node cells was not observed (Fig. 3). There was, however, a consistent trend in both SEA- and SWAP-restimulated cultures toward slightly elevated type 2-associated responses in anti-IFN-γ-treated mice.
Although only subtle changes in type 2 responses were detected in the cytokine-depleted groups, significant differences in the production of IFN-γ and IL-12 were observed after neutralization of endogenous IFN-γ or IL-12. While all four groups of egg/IL-12-sensitized mice showed highly polarized type 1 responses (Fig. 4), IFN-γ production by both SEA- and SWAP-stimulated cultures were significantly lower in animals that were treated in vivo with anti-IFN-γ or anti-IL-12 (Fig. 4,B). Animals treated with anti-TNF-α consistently produced the largest amount of IFN-γ when compared with the other mAb-treated groups (Fig. 4 B).
IL-12 production followed a similar pattern as IFN-γ, although the highest levels were observed in cultures that were not restimulated with parasite Ags (Fig. 4,A). In all of our studies, SEA or SWAP restimulation reduced IL-12 expression in lymphocytes obtained from infected mice. Nevertheless, similar to the results for IFN-γ, IL-12 levels were higher in the anti-βGal-treated egg/IL-12-sensitized group when compared with both infected control groups. Neutralization of TNF-α or IL-12 had little effect on the ability to produce IL-12, while mice treated with anti-IFN-γ displayed significantly reduced IL-12 levels in SEA, SWAP, and nonstimulated cultures from both the mesenteric lymph nodes (Fig. 4 A) and spleen (data not shown).
Type 2 cytokine mRNAs are restored in the livers of infected egg/IL-12-sensitized mice after depletion of IFN-γ, TNF-α, or IL-12
Although significant quantitative differences were detected among the groups, the in vitro cytokine data discussed above suggest that a prominent type 1 response is maintained in the draining lymph nodes and spleens of all egg/IL-12-sensitized mice, even after treatment with neutralizing mAbs to IFN-γ, TNF-α, or IL-12. Thus, these data suggest that the increased pathology in anti-IFN-γ-, anti-TNF-α-, and anti-IL-12-treated mice may not result from enhanced type 2 responses. The above findings might argue for a direct anti-inflammatory role for these cytokines in the granulomatous response. Nevertheless, in vitro cytokine data may not directly reflect the ongoing cytokine response at the site of granuloma formation. To more thoroughly examine the cytokine response in situ, liver biopsies from individual mice were analyzed by semiquantitative RT-PCR for changes in cytokine mRNA expression. Surprisingly, in contrast to the in vitro results, these assays revealed that a marked type 2 cytokine response is induced in the egg/IL-12-sensitized animals after treatment with anti-IFN-γ, anti-TNF-α, or anti-IL-12 mAbs (Fig. 5, A and B). In the livers of anti-IFN-γ-treated mice, type 2 cytokine mRNA levels approached that of the egg-sensitized/infected controls (positive control for type 2 cytokines). These data thus correlate with the marked changes in pathology (Fig. 2) and in vitro IFN-γ expression (Fig. 4) observed in animals depleted of IFN-γ.
In agreement with the measurement of IFN-γ protein secretion from draining lymph node cells in vitro (Fig. 4), mRNA for this type 1 cytokine was abundant in livers of mice that had been sensitized with egg/IL-12 and/or subsequently treated with anti-IFN-γ, anti-TNF-α, or anti-IL-12 Abs (Fig. 6,A). However, as reported previously, the increases in liver IFN-γ mRNA were not significantly different from the infected controls (13). Neutralization of IFN-γ or TNF-α had little effect on IFN-γ mRNA expression, although IL-12 depletion modestly down-regulated IFN-γ mRNA transcript levels when compared with the anti-βGal-treated egg/IL-12-sensitized animals. Thus, in contrast to the marked decreases in IFN-γ expression observed in Ag-stimulated lymph node cultures (Fig. 4,B), neutralization of IFN-γ did not affect the level of IFN-γ mRNA transcripts in the liver. Nevertheless, the overall dominance of type 1 vs type 2 (IFN-γ vs IL-5) associated cytokines was shifting in the granulomatous tissues. Neutralization of IFN-γ (and to a lesser extent, anti-IL-12 and anti-TNF-α treatment) restored the expression of IL-4 and IL-5 mRNA (Fig. 5), but did so without significantly modifying IFN-γ mRNA expression.
Neutralization of endogenous TNF-α increased TNF-α mRNA transcripts and partially up-regulated p40 IL-12 mRNA levels (Fig. 6, B and C). By contrast, mice depleted of IFN-γ demonstrated the greatest abrogation in TNF-α and p40 IL-12 mRNA transcript levels when compared with mice treated with anti-βGal, anti-TNF-α, or anti-IL-12 Abs (Fig. 6, B and C). Anti-IL-12 Ab treatment had little effect on p40 IL-12 or TNF-α mRNA expression in the egg/IL-12-sensitized mice (Fig. 6, B and C).
SWAP-specific IgG2a titers are decreased in the sera of infected egg/IL-12-sensitized mice after depletion of endogenous IFN-γ or IL-12
The RT-PCR data demonstrated that type 2 cytokine responses were up-regulated after treatment with anti-IFN-γ, anti-TNF-α, or anti-IL-12, at least in the livers of egg/IL-12-sensitized mice. These results likely explain the marked pathologic changes observed in the cytokine-depleted groups. To determine whether other type 2-regulated responses were similarly effected by depleting these cytokines, serum IgE and IgG2a titers were examined. For these experiments, serum was collected from individual animals at the time of sacrifice and SEA-specific IgG2a (type 1-associated) and total IgE Ab titers (type 2-associated) were measured by ELISA. As shown in Figure 7, neutralization of IFN-γ or IL-12 significantly decreased SEA-specific IgG2a Ab titers when compared with the anti-βGal-treated controls (Fig. 7 A). There were no significant changes observed in the TNF-α-depleted mice.
Total IgE levels in egg/IL-12-sensitized mice were much lower when compared with the egg-sensitized controls, regardless of the Ab treatment (Fig. 7 B). Although animals treated with anti-IFN-γ, anti-TNF-α, or anti-IL-12 all had more total IgE than animals treated with anti-βGal, only those animals treated with anti-TNF-α displayed statistically significant (p < 0.05) increases in total IgE titers when compared with the egg/IL-12-anti-βGal-treated group. Nevertheless, these levels were still much lower than either of the infected control groups. Interestingly, the egg/saline-sensitized animals developed the highest IgE titers. There were no differences in SEA-specific IgG1 or IgG2b Ab titers between any of the groups (data not shown) although all egg/IL-12-sensitized animals had the highest titers of total anti-SEA Abs (data not shown).
Pathology associated with S. mansoni infection is a result of granuloma formation surrounding parasite eggs trapped in the sinusoids of the liver, a process that leads to increased collagen synthesis, tissue eosinophilia, and severe hepatic fibrosis (20). The cytokine cascade induced upon parasite egg-deposition and granuloma formation is characterized by increased production of type 2 cytokines including IL-4, IL-5, and IL-13 (14, 21, 22). Cytokine manipulation studies (17, 23) and experiments in gene knockout mice (24, 25, 26) revealed a critical role for the type 2 cytokine response in orchestrating many aspects of egg-induced immunopathology. In recent studies, we have shown that much of the pathology associated with S. mansoni infection can be ameliorated by converting the type 2 response to a limited type 1-dominated pattern (13, 14). This was achieved by sensitizing mice to egg Ags in the presence of IL-12, a potent Th1-inducing cytokine (1) before schistosome challenge. Nevertheless, the mechanisms responsible for the reduced pathology in these mice have not been elucidated. In the current study, we revealed a critical role for TNF-α, IL-12, and IFN-γ in regulating many aspects of this IL-12-based antipathology vaccine.
In a pulmonary model of granuloma formation, IFN-γ was shown to play an anti-inflammatory role in egg-induced lesion formation (27). Neutralization of endogenous IFN-γ increased granuloma size in the lungs of i.v. egg-challenged mice and simultaneously up-regulated type 2 and decreased type 1 cytokine responses (14). Thus, maximal pulmonary granulomatous inflammation correlated with polarized type 2 cytokine responses. These data suggested that a major role of the type 1-associated IFN-γ response was to act as a down-regulator of type 2 cytokine production. IL-12 was shown to play a similar role as IFN-γ because exogenous treatment with the cytokine significantly decreased granuloma development and type 2 cytokine expression (14). It was demonstrated that sensitizing mice to egg Ags in the presence of IL-12, before egg challenge, could effectively reduce subsequent pulmonary granulomatous responses. These findings were extended to the livers of infected animals where fibrosis, as well as granuloma size, were decreased by prior egg/IL-12 sensitization (13). The livers of egg/IL-12-sensitized animals displayed marked increases in IFN-γ, IL-12 p40, and TNF-α mRNA expression when compared with nonsensitized-infected mice. The cytokine depletion experiments performed here demonstrate that elevated production of all three cytokines at the time of egg laying in infected mice is critical for the suppression of egg-induced pathology (Fig. 2). The increased pathology in all three anti-cytokine-treated groups correlated with at least a partial restoration in IL-4 and IL-5 mRNA expression at the site of granuloma formation (Fig. 5), thus likely explaining the increased fibrosis and granuloma size, as well as tissue eosinophilia (Fig. 2).
Although the local type 2/type 1 cytokine balance was altered by all three cytokine depletions, the production of type 2 cytokines was not significantly increased in the mesenteric lymph nodes (Fig. 3) or spleens (data not shown) after in vitro restimulation. Only the IFN-γ-depleted mice showed a very slight increase in IL-5 production when compared with the egg/IL-12-sensitized control group. Nevertheless, this increase was minimal when compared with the control and egg/saline-infected animals. Indeed, all three groups continued to display highly significant Ag-specific type 1 responses in the periphery (Fig. 4). Interestingly, while type 2 responses were not up-regulated in the draining lymph nodes, IFN-γ expression was significantly decreased in both IFN-γ- and IL-12-depleted mice (Fig. 4,B). A similar decrease in the type 1 to type 2 cytokine ratio was also observed at the site of granuloma formation. However, the cytokine shift in the liver was a result of increased type 2 cytokine-associated mRNA expression rather than a result of decreased IFN-γ mRNA expression. This finding suggests that IFN-γ responses may be more stable in the liver as opposed to the draining lymph nodes. The pathologic changes observed in the liver additionally suggest that an important functional change occurred as a result of this shift in the cytokine balance (Fig. 2). These data are somewhat surprising because studies with protozoan parasites have suggested that while both IFN-γ and IL-12 are required to generate Th1 responses (4, 5, 28) only IFN-γ is necessary to maintain an established Th1 response (29, 30). The results presented here indicate that IL-12, as well as IFN-γ, may be required to maintain strong polarized type 1 responses to schistosome eggs. Moreover, the in vitro ELISA data, when combined with the liver RT-PCR results, suggest that the maintenance of polarized Th1-type responses to Th2-inducing stimuli, particularly in the local milieu (Fig. 5), remains highly dependent upon endogenously produced IFN-γ, IL-12, and, to a lesser extent, TNF-α. These findings may therefore be consistent with in vitro studies that found some Th1 populations are readily converted into IL-4 producers while Th2 cells are relatively fixed in their cytokine-producing pattern (31, 32). Nevertheless, we have not formally demonstrated that the changes in cytokine expression are due solely to changes in cytokine production by CD4+ T cells. Indeed, cell types other than CD4+ T cells, such as eosinophils and basophils, clearly contribute to the overall cytokine response during schistosomiasis (33) and therefore may also be contributing to the changes reported here.
To begin to understand the connection between IFN-γ, IL-12, and TNF-α and the mechanism for limiting egg-induced pathology, we examined the regulation of these cytokines following the various Ab treatments. Not unexpectedly, the production of all three cytokines was closely linked. IFN-γ-depletion resulted in the greatest restoration in liver pathology (Fig. 2) and was associated with the most significant increase in the local type 2 cytokine response (Figs. 3 and 5). Interestingly, anti-IFN-γ-treated mice also showed marked decreases in IL-12 (Figs. 4,A and 6C) and TNF-α expression (Fig. 6,B) when compared with the anti-βGal-treated mice. These findings thus confirm that IL-12 and TNF-α are up-regulated by IFN-γ (1, 34, 35) and demonstrate that anti-IFN-γ treatment actually yields a triple cytokine depletion. Anti-IL-12-treated egg/IL-12-sensitized mice also showed reduced IFN-γ responses (Fig. 4,B), while TNF-α expression was not significantly affected (Fig. 6,B). Again, these animals displayed significant increases in granuloma size, fibrosis, and tissue eosinophilia, although the changes were less marked than in the anti-IFN-γ-treated mice (Fig. 2), which may be explained by the unaffected TNF-α response in the anti-IL-12- vs anti-IFN-γ-treated mice (Fig. 6,B). Anti-TNF-α-treated mice, while also showing increased IL-4 and IL-5 mRNA expression in the liver (Fig. 5), displayed little change in their type 1 response when compared with the anti-βGal-treated controls (Figs. 4 and 6). In fact, SEA-specific IFN-γ production was slightly increased in the anti-TNF-α-treated animals (Fig. 4,B). TNF-α mRNA expression was also increased in the livers (Fig. 6,B). The fact that IFN-γ (Fig. 4,B) and IL-12 (Fig. 4,A) expression were not affected correlates with the failure to signficantly restore the tissue eosinophilia in anti-TNF-α-treated animals (Fig. 2 B). Moreover, these results suggest that the effects of TNF-α on granuloma formation and fibrosis may be more direct and downstream from IFN-γ or IL-12 because the latter cytokine-depleted animals manifested multiple Th1-associated deficiencies while the anti-TNF-α-treated group showed only modest changes in their type 1/type 2 cytokine profile.
The results obtained from anti-TNF-α-treated mice in particular are surprising given previously published information on the role of TNF-α in schistosome egg-induced granuloma formation. Indeed numerous studies have suggested that TNF-α plays a proinflammatory rather than anti-inflammatory role in lesion development as demonstrated here. Amiri et al. (36) were the first to demonstrate a role for the cytokine in schistosomiasis by showing that TNF-α could partially restore granuloma formation in SCID mice. In related studies, Joseph and Boros (37) showed that peak liver granuloma size was reduced in infected mice when the animals were treated with neutralizing Abs to TNF-α. In addition, Adewusi et al. (38) reported increased levels of TNF-α in CBA/J mice that developed a more serious hepatosplenic-like form of the disease. In contrast to these findings, our studies in egg/IL-12-sensitized mice suggest that TNF-α is playing a host-protective role (Fig. 2). Indeed, in all of our studies examining the role of IL-12, the most dramatic reductions in either pulmonary (14, 27) or hepatic (13) pathology correlated with marked increases in TNF-α expression. These data suggest that TNF-α may play distinct roles in granuloma formation and hepatic fibrosis depending on the particular cytokine milieu in which it is expressed. Indeed, in a tuberculosis model (39), TNF-α was shown to exhibit unique activities in a pure type 1 vs a mixed type 1/type 2 dominated immune response. In type 1 responses, TNF-α behaved in a manner similar to IFN-γ, acting as a key macrophage-activating cytokine, while in mixed type 1/type 2 or type 0 responses, the cytokine induced tissue damage. Thus in schistosome-infected mice, where type 2 responses dominate (21), TNF-α appears to play a proinflammatory role (36, 37) leading to hepatic pathology, while in egg/IL-12-sensitized mice, TNF-α may contribute to the collagen-suppressing activities exhibited by IFN-γ (15). Future studies aimed at elucidating the role of this important immunoregulatory cytokine in type 1- vs type 2-dominated granulomatous responses could lead to more effective strategies for limiting hepatic pathology induced by schistosome infection.
A recent study demonstrated that IgE-deficient mice (Cε knockout) develop smaller liver granulomas when compared with WT control-infected animals (40). These data were the first to suggest a possible proinflammatory role for IgE in egg-induced granuloma formation. Our data from egg/IL-12-sensitized mice correlate with these observations because IgE levels were dramatically reduced in these animals when compared with infected, non-IL-12-treated controls (Fig. 7); granuloma size was similarly decreased in these animals as well (Fig. 1,A). Nevertheless, the anti-cytokine-treated egg/IL-12-sensitized mice developed granulomas that were nearly identical to those in both infected non-IL-12-treated control groups (Fig. 1,A) but showed little restoration of their serum IgE response (Fig. 7). In addition, infected egg/saline-sensitized mice showed almost no increase in their acute stage granulomatous response when compared with the nonsensitized controls, yet displayed significantly elevated IgE Ab titers. Together, these data suggest that vaccine-induced IgE responses play little or no role in the regulation of acute stage granuloma formation. Nevertheless, a role for B cells (41), Ab (42), and possibly Fc receptor (41, 43) signaling in the regulation of granuloma formation, particularly in the chronic stage of infection, has recently been established.
Previous studies showed that IFN-γ could suppress hepatic fibrosis in schistosome-infected mice (15). The data presented here demonstrate that the reduction in granuloma formation, collagen synthesis, and tissue eosinophilia conferred by prior egg/IL-12 sensitization relies upon the combined increased expression of multiple cytokines, including IFN-γ, IL-12, and TNF-α. Nevertheless, IFN-γ emerged as a central mediator because depletion of this cytokine alone exerted the most dramatic effects on pathology while simultaneously decreasing the expression of the other key cytokines. A more surprising finding of this study was the relative plasticity of the egg/IL-12-induced type 1 response in infected animals. Most striking was the marked increase in type 2 cytokines in the local granuloma environment after neutralization of endogenous IFN-γ, IL-12, or TNF-α. In addition, the fact that type 1 responses were significantly decreased in the periphery after depletion of IFN-γ or IL-12 suggests that vaccine-induced type 1 responses generated against potent type 2-inducing stimuli such as schistosome eggs may be continuously dependent on multiple type 1-inducing cytokines for their maintenance.
We thank Dr. Fred Lewis and Ms. Barbara Clark at the Biomedical Research Institute for providing the parasite materials used in this study and Dr. Joe Sypek for providing recombinant murine IL-12 and for many helpful discussions. We also thank Drs. Monica Chiaramonte, Matthias Hesse, Alan Sher, Dragana Jankovic, and Rhian Hayward for critically reviewing this manuscript.
This investigation received financial support from the United Nations Development Program/World Bank/World Health Organization Special Programme for Research and Training in Tropical Diseases.
Abbreviations used in this paper: SEA, soluble egg Ag; SWAP, soluble worm Ag preparation; βGal, β galactosidase; HPRT, hypoxanthine phosphoribosyltransferase.