INTRODUCTION

Top Abstract Introduction Materials and Methods Results Discussion References

In murine schistosomiasis mansoni, the CD4⁺ T-helper (Th)-cell population (12, 21, 26) initiates and maintains the granulomatous inflammatory response around disseminated worm eggs (6). Over the past several years, Th0, Th1, and Th2 subsets of lymphocytes have been implicated in the generation of granuloma formation (10, 32, 40). In the lung granuloma model, as well as in infected mice, analysis of cytokine mRNA message and expression showed an early Th0-Th1 profile, which shifted to a predominant Th2 pattern with the maturation of the granulomas (3, 11, 19, 37). Moreover, cross-regulation by Th1- and Th2-type cytokines has been demonstrated; gamma interferon (IFN-) and interleukin-12 (IL-12) were shown to downregulate the IL-4-mediated granuloma response (20, 24), whereas IL-4 (9) and IL-10 regulated IFN- and IL-2 production (14, 15, 29, 38).

Recently, a strong regulatory role has been described for IL-12 in the lung granulomatous response of naive or sensitized mice, which was mediated by IFN- (36). Thus, the generation of a strong Th1 response proved to be effective in the suppression of the florid development of the Th2-type granuloma. This impressive suppression could not be duplicated in infected mice, because repeated injections of exogenous recombinant IL-12 (rIL-12) into egg-primed and subsequently infected mice showed only moderate inhibition of liver granuloma development (33).

Based on our previous observation that established the predominant Th1-type IFN- cytokine response of the early-developing liver granulomas (19), we intended to prolong this Th1 phase by the administration of exogenous rIL-12 and to examine its suppressive effect on the development of the granulomas and the cytokine production profile. Here we show that when we found the proper time window for treatment, prolonged Th1 responsiveness achieved significant suppression of the hepatic granulomatous response and decreased collagen deposition in the tissues.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results Discussion References

Infection. Six- to 8-week-old female CBA/J^k mice were infected by subcutaneous injection of 25 cercariae of the Puerto Rican strain of Schistosoma mansoni at the base of the tail.

Cell culture. Single-cell suspensions from individual spleens and pooled isolated liver granulomas were produced as previously described (26). Cells at a concentration of 3 × 10⁶/ml were incubated in 48-well plates (0.6 ml) with complete RPMI 1640 medium (Gibco/BRL, Grand Island, N.Y.) containing 20 mM HEPES (Fisher Scientific, Pittsburgh, Pa.), 2 mM sodium pyruvate (ICN, Costa Mesa, Calif.), 50 µM 2-mercaptoethanol, 1% penicillin-streptomycin-glutamine (Sigma Chemical Co., St. Louis, Mo.), and 10% fetal calf serum (Gibco/BRL) with soluble egg antigens (SEA) (0 or 10 µg/ml). Supernatants were collected at 24 h for IL-2 and IL-4 and at 48 h for IFN-, IL-5, and IL-10 cytokine measurement.

Cytokine measurement. IL-4, IL-5, IL-10, and IFN- levels were determined by sandwich enzyme-linked immunosorbent assay with matched antibody pairs (Pharmingen, San Diego, Calif.) developed with streptavidin-alkaline phosphatase and para-nitrophenyl phosphate (pNPP) (Sigma).

ELISPOT. Multiscreen-IP plates (Millipore, Bedford, Mass.) were used according to the manufacturer's suggested protocol, employing the IFN- coating and the detecting antibodies previously cited. Serial 1:5 dilutions of spleen cells were plated in duplicate for individual spleens and incubated for 20 h with 20 µg of SEA/ml. The assay was completed with streptavidin-alkaline phosphatase and developed for 30 min with BCIP (5-bromo-4-chloro-3-indolylphosphate)-nitroblue tetrazolium alkaline phosphatase substrate tablets (Sigma). Spots were counted with a 10× microscope eyepiece.

Intracellular cytokine staining-flow cytometry. Spleen cells were incubated in 24-well plates (3 × 10⁶/ml) for 20 h in the presence of 20 µg of SEA/ml and 1 µg of brefeldin A (Sigma)/ml. The cells were harvested, washed, and stained with the Fix & Perm system (Caltag, Burlingame, Calif.) according to the manufacturer's suggested protocol, using fluorescein isothiocyanate-anti-CD4 and phycoerythrin-anti-IFN- antibodies (Caltag). Unlabeled anti-IFN- antibody and labeled isotype controls were used to determine specificity. Samples were run on a Becton Dickinson FACScan and analyzed with Lysis II software. A total of 50,000 lymphocyte-gated events were counted for each of 10 individual spleens in each group.

Antibodies. Clones C15.1, C15.6, C17.15, and C17.8, which produce anti-IL-12 MAbs, were kindly provided by G. Trinchieri, The Wistar Institute of Anatomy and Biology (39). The MAbs from clones 11b11 (IL-4), R4-6A2 (IFN-), and JES-2A5 (IL-10) (American Type Culture Collection, Rockville, Md.) were produced as ascites fluid in irradiated CBA/J^k mice and purified by thiophilic resin chromatography (Pierce, Rockford, Ill.) by the suggested protocol. Following sterilization, aliquots were stored at 70°C. Intraperitoneal administration of antibodies (500 µg of protein/mouse) was on alternate days to rIL-12 injections.

rIL-12. (i) Production. The CHO cell line DUXB11, stably transfected with both IL-12 p35 and IL-12 p40, was kindly provided by Frederick Heinzel, Geographic Medicine, Case Western Reserve University School of Medicine, on behalf of Genetics Institute, Inc., Cambridge, Mass. The cells were allowed to overgrow in high-glucose Dulbecco's modified Eagle's medium (Gibco/BRL) supplemented with 5% fetal calf serum (Gibco), 1% nonessential amino acids, 1% glutamine-penicillin-streptomycin, and 10 µM methotrexate (Sigma). The rIL-12 was isolated from the supernatant by affinity chromatography with C15.6 IL-12 antibody. The column eluate was dialyzed extensively against phosphate-buffered saline, treated with polymyxin B beads, filter sterilized, and quantitated by comparison to rIL-12 standard (R&D Systems, Minneapolis, Minn.) via enzyme-linked immunosorbent assay and bioassay.

(ii) Treatment. rIL-12 was administered at 400 ng/mouse in 0.2 ml of Hanks balanced salt solution intraperitoneally on a thrice-weekly schedule. Animals receiving continuous treatment received 200 ng/injection on a daily basis for 3 weeks.

Granuloma measurements. Livers of mice were fixed in buffered formalin and processed for histology. Granulomas were measured in hematoxylin-eosin-stained sections on coded slides by computerized morphometry (Micro-Comp image analysis program; Southern Micro Instruments, Atlanta, Ga.) Fifteen to 25 granulomas were measured per mouse. The means of granuloma areas from individual mice were pooled and analyzed. Significant differences were evaluated by Student's t test.

Hydroxyproline assay. The hydroxyproline assay was carried out based on previously cited methodology (2) but using base (27) rather than acid hydrolysis for the dissolution of tissues. The hydroxyproline concentration was calculated per 10,000 eggs (8).

	RESULTS

Top Abstract Introduction Materials and Methods Results Discussion References

Prolongation by rIL-12 and anti-type 2 cytokine MAb treatment of the Th1 phase of granuloma growth attenuates the inflammatory response. To assess the effect of rIL-12 treatment on liver granuloma development, cytokine injections were given that spanned the pre-egg deposition-full-grown granuloma spectrum of the lesion. Granuloma measurements were made at the end of each treatment period to assess the effect of the cytokine on the developing Th2 response. As Fig. 1A shows, treatment that covered the pregranuloma-to-early-granuloma sensitization period (5 to 6 weeks after infection) had no effect on lesion size. Longer treatment given between 5 and 7 weeks after infection (1.5 weeks of granuloma age, with predominant IFN- production [19]), which we regard as the Th1 phase of lesion development, achieved a significant (P < 0.05) decrease in size. Treatment given after the sensitization period, between 6 and 7 weeks after infection, was less effective but still yielded significantly less granuloma development. In contrast, cytokine administration initiated after the sensitization period and continued throughout the development of the strong Th2 phase of the granuloma (6 to 8 weeks) was ineffective. The same total dosage given in daily injections throughout the entire period (5 to 7.5 weeks) of granuloma development strongly attenuated the inflammatory response (Fig. 1B). However, such a regimen caused 60% mortality and liver morbidity (cell vacuolation, central necrosis within granulomas, and mononuclear infiltration in the liver parenchyma).

View larger version (25K): [in this window] [in a new window]   FIG. 1.   Efficacy of treatment with rIL-12 and neutralizing MAbs on granuloma development at different times after infection. At the ends of the treatment periods, the mice were sacrificed and the granulomas were measured in stained histologic sections. Open bars, control; solid bars, rIL-12 treated; hatched bars, MAb treated. The asterisks denote significant differences (P < 0.05) compared with the control. SEM, standard error of the mean. (A) 400 ng of rIL-12 was injected three times/week (w). The data were pooled from a total of nine mice in three repeat experiments. (B) rIL-12 (200 ng) was injected daily for 2.5 weeks. The data are pooled from six mice. (C) Combined treatment between 6 and 7 weeks with 400 ng of rIL-12 and 500 µg of protein of either anti-DNP MAb (control; solid bar) or anti-IFN- MAb (hatched bar) each given three times/week on alternate days. The data are pooled from six mice. (D) Combined treatment with 400 ng of rIL-12 injected three times/week between 5 and 8 weeks after infection and a cocktail of 250 µg of protein (each) of anti-IL-4 and anti-IL-10 MAbs (TH2) injected three times/week between 6 and 8 weeks after infection. The control group received anti-DNP MAb of the same isotype. The data are pooled from five mice. (E) Treatment with 500 µg of protein of anti-IL-12 MAb cocktail given three times/week. Each group consisted of six mice. Hatched bars, DNP MAb control; open bars, experimental.

SEA-specific cytokine production after various treatments. Prolonged treatment with rIL-12 that spanned the pregranuloma (5 weeks) and well-developed granuloma (7 weeks) periods suppressed Th2-type (IL-4 and IL-5) cytokine production in splenic, but not in granuloma, cells. In contrast, IFN- production in both cultures rose significantly following in vitro antigenic stimulus (Fig. 2).

View larger version (16K): [in this window] [in a new window]   FIG. 2.   Type 2 and type 1 cytokine production by SEA-stimulated splenic (Spc) and granuloma (Grc) cells of mice treated with 400 ng of rIL-12 three times/week between 5 and 7 weeks after infection. Non-antigen-stimulated cultures showed very low cytokine production. Open bars, control; solid bars, experimental. The data are representative of three repeat experiments with a total of nine mice. The error bars represent intra-assay variation.

View larger version (14K): [in this window] [in a new window]   FIG. 3.   Type 2 and type 1 cytokine production by SEA-stimulated splenic (Spc) and granuloma (Grc) cells treated daily with 200 ng of rIL-12 between 5 and 7.5 weeks after infection. The designations of the bars are as in Fig. 2. The data are representative of two repeat experiments with a total of six mice. The error bars represent intra-assay variation.

View larger version (29K): [in this window] [in a new window]   FIG. 4.   Type 2 and type 1 cytokine production by SEA-stimulated splenic (Spc) and granuloma (Grc) cells of mice given the combined rIL-12 and anti-IL-4-anti-IL-10 MAb treatments (TH2) as described in the legend to Fig. 1D. The data are representative of two repeat experiments obtained from the pooled organs of five mice in each experiment. The error bars represent intra-assay variation.

View larger version (17K): [in this window] [in a new window]   FIG. 5.   Type 2 and type 1 cytokine production by SEA-stimulated splenic (Spc) and granuloma (Grc) cells treated with a mixture of anti IL-12 MAbs as described in the legend to Fig. 1E. Open bars, control; solid bars, experimental. The data were obtained from the pooled organs of six mice. The error bars represent intra-assay variation. w, weeks.

Hepatic collagen deposition following rIL-12 treatments. Hydroxyproline measurements in hydrolyzed liver tissues showed that short-term rIL-12 treatment between 6 and 7 weeks after infection which diminished granuloma size did not significantly influence the amount of deposited collagen (Fig. 6A). In contrast, prolonged daily injections of the cytokine (5 to 7.5 weeks) not only strongly affected granuloma growth but concurrently caused diminished collagen production within the liver (Fig. 6B) (P < 0.05). In the combined treatment, control mice given rIL-12 and anti-DNP MAb injections showed no diminished collagen production. Mice treated for 2 weeks with the anti IL-4, anti-IL-10 MAb cocktail had diminished liver collagen content, whereas combined treatment with rIL-12 and the antibody cocktail resulted in a strong decrease in collagen deposition (Fig. 6C) (P < 0.05 for both).

View larger version (14K): [in this window] [in a new window]   FIG. 6.   Liver collagen content (expressed as micromoles of hydroxyproline) of mice treated with 400 ng of rIL-12 three times/week between 5 and 7 weeks (w) after infection (A), with 200 ng of rIL-12 given daily between 5 and 7.5 weeks after infection (B), or with the combined treatment of 400 ng of rIL-12 and the anti-IL-4-anti-IL-10 MAb cocktail (TH2) (see the legend to Fig. 1D) (C). DNP, control MAb. The data represent the mean values (± standard error of the mean) from the livers of four mice.

	DISCUSSION

Top Abstract Introduction Materials and Methods Results Discussion References

During the formation of the schistosome egg-induced granuloma, cytokine-mediated cross-regulation influences the dynamics of Th1-Th2-cell function (4, 24, 34, 39). The activity of the Th subset of cells not only influences granuloma size but also affects the fibrous sequel of the inflammatory response (33). Whereas the Th1-type IFN- production occurs during the incipient phase of granuloma evolution (5.5 to 6.5 weeks) and may be involved in the generation of the inflammatory response (10, 19), Th2 cells and type 2 cytokines have been shown to induce the florid phase of granuloma development (17, 25) and collagen deposition (7, 17). Thus, it is presumed that interference with Th2-cell function should have an ameliorating effect on the inflammatory response. The present study was initiated with infected mice to prolong the previously demonstrated predominant IFN- producer (Th1) phase of the granulomatous inflammation (19), thereby diminishing or abrogating the Th2-cell-mediated florid granuloma development.

The results presented indicate the feasibility and the limitations of this approach. The timing of the treatment proved to be crucial to achieving suppression. Whereas injections of rIL-12 that covered the incipient phase of granuloma evolution (corresponding to 1 to 1.5 weeks of granuloma age) achieved significant suppression in the size of liver lesions, treatment that coincided with the development of the strong Th2 response (6 to 8 weeks after infection; maximal granuloma size) was ineffective. As suggested, such treatment could induce the differentiation of Th1 cells from a reservoir of precursor Th0 cells that may be abundant during the early phase of granuloma formation (24, 36). Our flow cytometry observation showing a significant increase in IFN-⁺ CD4⁺ cells in the spleens of mice treated with rIL-12 supports this contention. With the maturation of the granuloma response, such precursors would be signaled by the rising level of IL-4 to develop into Th2 cells (16, 30). IL-4 within the mature granuloma is produced by Th2 cells and potentially by mast cells and eosinophils (28), whereas IL-10 a regulator of type 1 cytokine production (22), is derived from Th2 cells (29) and macrophages (14). Thus, it is the overwhelming Th2 cytokine environment that renders the action of diminished endogenous or exogeneously given rIL-12 ineffective. Analysis of the cytokine profile of splenic and granuloma cells of rIL-12-treated mice revealed that the treatment significantly elevated IFN- production. That such higher levels of IFN- production are involved in the mediation of granuloma suppression was demonstrated by the present experiments, where neutralization of IFN- by specific antibody given concurrently with rIL-12 injections abrogated granuloma suppression. This observation agrees with the reported suppressive role of IFN- in the evolution of granulomas in infected (20) or egg-injected naive (36) and IFN--deficient (35) mice. The extent of rIL-12-mediated regulatory effect achieved in the present experiments was impressive, whether expressed as diminution in the area or in the volume of the lesions of the treated mice. It is noteworthy that continuous injections of the cytokine from 5 until 7.5 weeks (covering the entire period of granuloma growth), while not exceeding the previous dosage regimen, achieved somewhat better results. However, they also caused high mortality (60%) and caused morbidity among the survivors in the form of vacuolated liver cells, heavy infiltration of the hepatic parenchyma by mononuclear cells, and necrotic centers in the granulomas.

It appears that downregulation of liver granuloma development is feasible by the prolongation of the Th1 phase of evolution. When this phase was disturbed by neutralization of endogenously produced IL-12, enhancement of the granulomas ensued. Whether the severe mortality observed among mice with the long-term treatment resulted from rIL-12-induced toxicity or the exacerbated Th1 inflammatory response remains to be elucidated.

Polarization of the granuloma response towards the Th1 pattern was also achieved by the combined treatment with rIL-12 and anti-IL-4 and anti-IL-10 MAbs. Whereas single treatments with rIL-12 or anti-type 2 cytokine MAbs were ineffective in impeding granuloma growth, treatment that enhanced Th1-type responses, as manifested by elevated IFN- production, and concurrently lowered the level of type 2 cytokines yielded good suppression of the granulomas without mortality. This is all the more remarkable because the mice had received treatment between 6 and 8 weeks after infection, during the increasing Th2-type developmental phase of granuloma growth. Apparently, neutralization of the high IL-4 and IL-10 levels and the concurrent administration of exogenous rIL-12 could amplify the differentiation and maturation of Th1 cells from the low numbers of the residual Th0 population. This combined treatment obviated the need for prolonged rIL-12 injections, with salutary results in diminished collagen deposition.

It is noteworthy that none of the treatment modalities or their timing with respect to granuloma growth could completely abrogate the hepatic inflammatory response. It appears that, unlike in the pulmonary model (36), complete abrogation of hepatic granuloma formation is not achievable, because even if complete polarization to the Th1 pattern is obtained a smaller, Th1-cell-mediated granuloma is expected to form. This may be the case in Stat 6-deficient infected mice that developed hepatic granulomas in the absence of type 2 cytokine production (17). Complete abrogation of the protective granuloma response is also not desirable because of the damage to the hepatic parenchyma that has been reported to be caused by egg antigens that diffuse from around unsurrounded eggs in nude or SCID mice (1, 5, 13).

Ongoing experiments show that endogenous production of IL-12 by granuloma macrophages, as well as IL-12 receptor display on granuloma CD4⁺ cells, is maximal at 6 weeks, coincident with strong Th1 activity, and drops sharply thereafter. This decline combined with the sharp rise in Th2-type cytokine production promotes the development of the Th2 phase of the mature granuloma.

Whereas a previous study reported partial inhibition of liver granuloma formation and reduced collagen production in egg- and IL-12-sensitized and subsequently infected mice (33), here we show that properly timed exogenous rIL-12 treatment alone without presensitization can prolong the early Th1 phase of granuloma development that effectively downregulates the intensity of the granulomatous inflammation and collagen deposition. On the negative side, as shown in the present study, a sustained, vigorous IFN--mediated Th1 response can cause tissue damage and mortality. These results agree with observations made in several animal models of chronic inflammatory diseases (18, 23, 31). An improved approach employed in this study prolonged the Th1 phase of the granuloma development and concurrently dampened the gradually increasing Th2 response by neutralizing antibodies. This combination resulted in reduced granulomatous inflammation and collagen deposition but no mortality. Thus, interventions that aim at the reduction of egg-induced granulomatous pathology should consider the consequences that result from the polarized Th1-Th2 host response.