Antigens by T Cells from Schistosoma japonicum-Infected Mice

Vol. 54, No. 2

INFECTION AND IMMUNITY, Nov. 1986, p. 590-592

0019-9567/86/110590-03$02.00/0 Copyright C 1986, American Society for Microbiology

Inhibition of Immediate and Arthus Responses to Schistosome Egg Antigens by T Cells from Schistosoma japonicum-Infected Mice ABRAM B. STAVITSKY1* AND G. RICHARD OLDS2 Department of Molecular Biology and Microbiology1 and Division of Geographic Medicine, Department of Medicine,2 Case Western Reserve University and University Hospitals, Cleveland, Ohio 44106 Received 30 May 1986/Accepted 12 August 1986

Schistosoma japonicum-infected mice develop hepatic granulomas, immediate hypersensitivity (IH), and delayed hypersensitivity (DH) to soluble egg antigens (SEA) released by parasite eggs trapped in liver sinusoids. All of these responses spontaneously regress after 7 to 9 weeks of infection. This study aimed to develop an in vivo system for the further dissection of cellular and humoral immune responses to SEA. C57BL/6 mice immunized subcutaneously with SEA in complete Freund adjuvant developed IH, an Arthus reaction, and DH to this antigen 5 to 9 days later. IH and the Arthus reaction, but not DH, were markedly inhibited if, 1 day before injection of SEA in complete Freund adjuvant, the mice were injected intravenously with purified T cells from the spleens of mice infected for at least 9 weeks. This in vivo model system can be used to study various aspects of cellular and humoral immune responses to SEA and their modulation. These results raise questions about the role of antibodies in the pathogenesis of granulomatous inflammation and about the mechanisms of its cellular regulation in infections with S. japonicum.

esis and modulation of granulomatous inflammation have also been studied in the lung granuloma model, in which cells or serum from infected mice are adoptively transferred into egg-sensitized mice which are subsequently reinjected with eggs (11, 12). This model is more economical than the adoptive-transfer model but does not distinguish among immediate hypersensitivity (IH), AR and DH to eliciting SEA. Therefore, a new model was developed in which mice injected with S. japonicum SEA demonstrate vigorous IH, DH, and AR to SEA. The effect on these responses of adoptive transfer of cells from S. japonicum-infected mice was then determined. Using this model, we found that adoptive transfer of T cells from infected mice into normal mice markedly or even completely inhibited subsequent IH and AR, but not DH, to SEA after sensitization with SEA. Female C57BL/6 mice obtained from Jackson Laboratory, Bar Harbor, Maine, were infected at Lowell University, Lowell, Mass., with 25 cercariae of a Philippine strain of S. japonicum (15). C57BL/6 mice were chosen because they develop extensive granulomatous inflammation (13) and spontaneously modulate this inflammation (13). SEA were prepared from the livers of CF1 mice (Charles River Breeding Laboratories, Inc., Cambridge, Mass.) infected with 50 cercariae of the same strain used to infect the C57BL/6 mice. At 8 to 12 weeks, their livers and intestines were removed, and parasite eggs were processed for antigen (5). The protein concentration of the SEA was determined by the Lowry method (10). To assay IH, AR, and DH to SEA, we injected 50 ,ug of SEA in 0.05 ml of phosphate-buffered saline into the left hind footpad and 50 p.g of bovine serum albumin (Pentex Biochemical, Kankakee, Ill.) into the other hind footpad. The thickness of each hind footpad was measured with calipers at 1, 5, 24, and 48 h. Net footpad swelling was the difference in thickness between the pad that received SEA and that which was injected with albumin. Different concentrations of SEA with or without complete Freund adjuvant (CFA) were introduced into C57BL/6 mice by various routes, including subcutaneously (s.c.) into the nape of the neck, s.c. at the base of the tail, and intraperitoneally. At different times after the injection of antigen, the

The major pathologic lesion in Schistosoma japonicum is the granulomatous inflammation which forms around eggs trapped in host tissues (20). This lesion is an immunologic response to soluble egg antigens (SEA) released from the trapped ova (12). In S. mansoni the granuloma appears to be exclusively a manifestation of a cellular immune response to SEA, and the host displays prominent delayed hypersensitivity (DH) to these antigens in vivo (3-5, 19). In contrast, in infections with S. japonicum, the pathogenesis of the granuloma appears to be more complex and may result from a combination of cellular and humoral immune responses to SEA (7, 8, 17, 18). Thus, in S. japonicum infection, granulomas contain eosinophilic abscesses which may be the result of an Arthus-type response (AR) to antigen (17). Additional evidence for a humoral component in the pathogenesis of granulomatous inflammation in this infection includes the finding of plasma cells within the granulomas (17, 18) and the continued synthesis of immunoglobulin and antibody to SEA by the granuloma throughout the course of infection (7). Although S. japonicum-infected mice develop DH to SEA (7) and there is evidence that a cell-mediated immune response to SEA contributes to this inflammatory process (6, 12), elicitation of DH appears to require much higher concentrations of SEA than in S. mansoni-infected animals (7). In both S. japonicum and S. mansoni infections spontaneous regression of granulomatous inflammation begins after 8 to 10 weeks of infection (2, 13). In both infections there is evidence that T suppressor cells can mediate this process (14), but in S. japonicum infection there is evidence that immunoglobulin Gl (IgGl) mainly contributes to modulation late in infection (13, 14). These conclusions about the mechanisms of modulation were obtained by adoptive transfer of cells or serum from more chronically infected animals into naturally infected recipient mice (14). Such experiments require 6 to 8 weeks to complete, are expensive, and do not permit the study of induction and modulation of immune responses to SEA in the absence of infection. The pathogen*

Corresponding author. 590

VOL. 54, 1986

NOTES

591

TABLE 1. IH, AR, and DH to SEA in mice after adoptive transfer of normal T cells or T cells from mice chronically infected with S. japonicum and subsequent immunization with SEAa Origin of T cells transferred,

no. of

cells (no. of mice)b

Antemal days (days)h

Interval

Expt 1 Normal, 1.5 x 107 (5) 9-wk infection, 1.5 x 107 (5)

7 7

Expt 2 Normal, 1.5 x 107 (4) 10-wk infection, 1.5 x 107 (5)

5 5

Expt 3 Normal, 2 x 107 (4) 12-wk infection, 2 x 107 (4)

0.24 ± 0.2 0.14 ± 0.1

24h

0.24 ± 0.2 0.43 ± 0.4

0.45 ± 0.3 0.45 ± 0.3

0.1

0.9 ± 0.4 0.6 t 0.4c

0.5 ± 0.3 0.7 ± 0.6

6 6

0.35 ± 0.2

0.3 ± 0.3 0.28 ± 0.2

0.26 + 0.2 0.23 ± 0.1


9 9

0.9 ± 0.3 0.48 ± 0.4c

1.1 ± 0.3 0.4 ± 0.4c

0.9 ± 0.3 0.8 ± 0.4


6 6

0.35 ± 0.2

0.3 ± 0.3 0.3 ± 0.3

0.26 ± 0.2 0.23 ± 0.1


7 7

0.23 ± 0.2 0c

0.37 ± 0.2

0.66 ± 0.15 0.56 ± 0.1

0.2 oc

Oc

Oc

Oc

a The designated number of mice were injected intravenously with the indicated number of T cells; 24 h later they were challenged in the neck with SEA-CFA. At the intervals indicated they were challenged with SEA in the left hind footpad and bovine serum albumin in the right hind footpad. A mm indicates the difference in footpad thickness between the left and right hind footpads measured at the indicated times after injection of antigen. I Normal mice versus mice infected for the indicated period. Significant difference (P < 0.05 by the Student t test) in the A mm between the footpads of mice injected with T cells from infected mice compared with that between the footpads of mice injected with T cells from normal mice.

mice were tested for IH (1 h), AR (5 h), and DH (24 to 48 h) by footpad challenge with 1, 5, 10, 25, or 50 ,ug of SEA. The combination of sensitization with 0.1 ml of a 50-50 (vol/vol) mixture of SEA (1 mg/ml) and CFA s.c. in the neck and testing with 50 ,ug of SEA per footpad yielded the most consistent results, i.e., maximal IH, AR, and DH 5 to 9 days after sensitization. The 5-h response elicited under these conditions was a typical AR with predominantly polymorphonuclear infiltrate, whereas the 24- to 48-h reaction was a typical DH response with a mononuclear (lymphocytes, macrophages) infiltrate. Therefore, in subsequent experiments sensitization was accomplished by s.c. injection into the neck of the optimal antigen concentration in CFA, and footpad testing was done with 50 ,ug of SEA per footpad. After optimal conditions were established for induction and assay of IH, AR, and DH to SEA, the possibility of modulating these responses by preinjection of T cells from S. japonicum-infected mice was investigated. T cells prepared on nylon wool columns (from normal mice or mice infected for 1, 3, 5, 9, 10, 12, or 13 weeks) were used because T cells from mice infected for 10 weeks had been shown to inhibit in vitro SEA-induced proliferation of spleen cells from acutely infected mice (16) and to suppress granulomatous inflammation in naturally infected animals (14). The concentrations of T cells used (1.5 x 107 to 4 x 108) were within the range of cell concentrations shown to be effective in suppressing granulomatous inflammation in vivo (16). Table 1 presents the results of 6 typical experiments (out of 25 experiments) in which preinjection of different concentrations of T cells from infected mice was followed 1 day later by s.c. injection of SEA-CFA with subsequent footpad testing 5 to 9 days later. Significant suppression of the 1-h response was seen in all experiments in which at least 1.5 x

107 T cells from mice infected for at least 9 weeks were injected. (T cells from mice infected for 1 or 3.5 weeks did not inhibit either the 1- or 5-h responses, and these data are not shown.) However, inhibition of the 5-h response was not observed consistently until 4 x 108 T cells were transferred; even 108 T cells did not inhibit this response in two experiments. No effect was seen on DH in any of the experiments. When the T-cell preparations were treated with anti-Thy-1.2 and rabbit complement before adoptive transfer they no longer inhibited the 1- and 5- to 6-h responses; therefore, the inhibitory cells were T cells. In all of the experiments, substantial IH, AR, and DH to SEA occurred in control mice in which equivalent numbers of normal T cells were injected. Thus adoptive transfer of sufficient numbers of T cells from mice infected for 9 to 13 weeks with S. japonicum into mice injected with SEA-CFA consistently reduced the IH to SEA 5 to 9 days later. This finding is consistent with the previous observation that by 9 weeks of infection IH to SEA spontaneously modulates in infected mice (1). It cannot, however, immediately be concluded that this IH to SEA is mediated by IgE antibody. It is possible that this immediate response is an early component of DH (T cell mediated) which resembles the IH caused by IgE antibody (1). Thus, in mice contact sensitized with dinitrofluorobenzene, an early T-cell component was first detected 1 day after sensitization which gradually declined during the ensuing week, whereas DH was first observed on day 4 after sensitization (1). We did not detect immediate footpad response to SEA during the first 4 days after injection of SEA, but in 50% of the mice we did detect DH to this antigen by the day 4; thus, in the SEA system there did not seem to be an early, pre-DH immediate component. Therefore, we concluded that the SEA-evoked immediate response was most likely IgE medi-

592

NOTES

ated. This conclusion is consistent with the fact that antiSEA IgE antibody is produced during murine S. japonicum infection (9). Striking modulation of IH and AR to SEA was achieved in this model. Therefore, this model can be used for preliminary screening of modulatory cellular or humoral immune mechanisms elicited by SEA. The further utility of this model is also suggested by the finding of differential effects of cell numbers on IH and AR; although some inhibition of AR was observed with 2 x 107 cells, 4 x 108 T cells were required for complete abrogation of AR. Failure of adoptive transfer of even 4 x 108 T cells to inhibit induction of DH to SEA is in striking contrast to inhibition of hepatic granulomatous inflammation upon transfer of as few as 2 x 107 chronic T cells into acutely infected mice (14). It is not known whether this apparent discrepancy is due to the sharp differences in the two systems; in this model, T cells are transferred before SEA presentation, whereas in natural infection, suppressor T cells develop only after 5 to 7 weeks of egg deposition and exposure to SEA (14). Taken together, these observations suggest that suppressor T cells found during chronic infection inhibit the afferent rather than the efferent limb of the cell-mediated immune response to SEA. Alternatively, the present observations suggest that chronic T cells may inhibit granulomatous inflammation because IH and AR responses to SEA also play a role in the pathogenesis of granulomatous inflammation. In addition to the previous findings of abscesses and plasma cells in the granulomas in this disease (17), more recent findings of small circumoval abscesslike granulomas in athymic mice infected with S. japonicum (6) also suggest a humoral immune component. Finally, it is possible that two populations of modulatory T cells are present in mice chronically infected with S. japonicum, one which inhibits IH to SEA and presumably accounts for the drop in IH which occurs after 7 weeks of infection, and another which suppresses granulomatous inflammation after 9 weeks of infection. It is not known whether the regulatory T cells are isotype specific in their effects, i.e., whether the same populations inhibit both the IgE and the IgG or IgM responses which presumably underlie the IH and AR, respectively. The requirement for many more T cells to inhibit AR than to inhibit IH suggests different populations, but alternative explanations of these results are possible. If IgE antibody to SEA plays a role in the pathogenesis of the granulomas, its participation must be short-lived because there is a sharp decline in IH to SEA after 7 weeks of infection (7). In contrast, AR may play a role later in the pathogenesis of this lesion since IgG antibody titers continue to rise in the serum and there is evidence of continued production of antibody to SEA by the granulomas per se for several months after infection (7). In conclusion, this new in vivo system can be used as a model to dissect differences in host responses to SEA, but it also raises important questions concerning the immunopathogenesis of disease in murine S. japonicum infection and its cellular regulation. This investigation was supported by a U.S. Public Health Service grant (AII8523) from the National Institutes of Health as part of the United States-Japan Cooperative Research Program. The excellent technical assistance of Weldon W. Harold is grate-

fully acknowledged. LITERATURE CITED 1. Askenase, P. W., C. M. Metzler, and R. K. Gershon. 1982. Localization of leukocytes in sites of delayed-type hypersensitivity and in lymph nodes: dependence on vasoactive amines. Immunology 47:239-246.

INFECT. IMMUN. 2. Boros, D. L., R. Peiley, and K. S. Warren. 1975. Spontaneous modulation of granulomatous hypersensitivity in schistosomiasis mansoni. J. Immunol. 144:1437-1441. 3. Boros, D. L., H. J. Schwartz, A. Powell, and K. S. Warren. 1973. Delayed hypersensitivity as manifested by granuloma formation, dermal reactivity, macrophage migration inhibition and lymphocyte transformation, induced and elicited in guinea pigs with soluble antigen of Schistosoma mansoni eggs. J. Immunol. 110:1118-1125. 4. Boros, D. L., R. Tomford, and K. S. Warren. 1977. Induction of granulomatous and elicitation of cutaneous sensitivity by partially purified SEA of Schistosoma mansoni. J. Immunol. 118:373-376. 5. Boros, D. L., and K. S. Warren. 1970. Delayed hypersensitivitytype granuloma formation and dermal reaction induced and elicited by a soluble factor isolated from Schistosoma mansoni. J. Exp. Med. 132:488-507. 6. Cheever, A. W., J. E. Byram, and F. von Lichtenberg. 1985. Immunopathology of Schistosoma japonicum infection in athymic mice. Parasite Immunol. 7:387-398. 7. Garb, K. S., A. B. Stavitsky, and A. A. F. Mahmoud. 1981. Dynamics of antigen- and mitogen-induced responses in murine schistosomiasis japonica: in vitro comparison between hepatic granulomas and splenic cells. J. Immunol. 127:115-120. 8. Garb, K. S., A. B. Stavitsky, G. R. Olds, J. W. Tracy, and A. A. F. Mahmoud. 1982. Immune regulation in murine schistosomiasis japonica: inhibition of in vitro antigen- and mitogeninduced cellular responses by splenocyte culture supernatants and by purified fractions from serum of chronically infected mice. J. Immunol. 129:2752-2758. 9. Little, J. V., III, C. Carter, and D. G. Colley. 1982. Serologic responses to Schistosoma japonicum: evaluation of total and parasite-specific immunoglobulins during the course of murine infection. J. Parasitol. 68:519-528. 10. Lowry, 0. H., N. J. Rosebrough, A. L. Farr, and R. J. Randall. 1951. Protein measurement with the Folin phenol reagent. J. Biol. Chem. 193:265-275. 11. Olds, G. R., and T. F. Kresina. 1985. Network interactions in Schistosoma japonicum infection. Identification and characterization of a serologically distinct immunoregulatory autoantiidiotypic antibody population. J. Clin. Invest. 76:2338-2347. 12. Olds, G. R., and A. A. F. Mahmoud. 1981. Kinetics and mechanisms of pulmonary granuloma formation around Schistosomajaponicum eggs injected into mice. Cell. Immunol. 60:251-260. 13. Olds, G. R., R. Olveda, J. W. Tracy, and A. A. F. Mahmoud. 1982. Adoptive transfer of modulation of granuloma formation and hepatosplenic disease in murine schistosomiasis japonica by serum from chronically infected animals. J. Immunol. 128:1391-1393. 14. Olds, G. R., and A. B. Stavitsky. 1986. Mechanisms of in vivo modulation of granulomatous inflammation in murine schistosomiasis japonicum. Infect. Immun. 52:513-518. 15. Peters, P. A., and K. S. Warren. 1969. A rapid method of infecting mice and other laboratory animals with Schistosoma mansoni: subcutaneous injection. J. Parasitol. 42:564. 16. Stavitsky, A. B., G. R. Olds, and L. B. Peterson. 1985. Regulation of egg antigen-induced in vitro proliferative response by splenic suppressor T cells in murine Schistosoma japonicum infection. Infect. Immun. 49:635-640. 17. Warren, K. S., D. L. Boros, L. M. Hang, and A. A. F. Mahmoud. 1975. The Schistosoma japonicum egg granuloma. Am. J. Pathol. 80:279-294. 18. Warren, K. S., and E. 0. Domingo. 1970. Granuloma formation around Schistosoma mansoni, S. haematobium and S. japonicum eggs in unsensitized and sensitized mice: size and rate of development, cellular composition, cross-sensitivity and rate of egg destruction. Am. J. Trop. Med. Hyg. 19:292-304. 19. Warren, K. S., E. 0. Domingo, and R. B. T. Cowan. 1967. Granuloma formation around schistosome eggs as a manifestation of delayed hypersensitivity. Am. J. Pathol. 51:735-756. 20. Warren, K. S., D. I. Grove, and R. P. Pelley. 1978. The Schistosoma japonicum egg granuloma. II. Cellular composition, granuloma size and immunologic concomitants. Am. J. Trop. Med. Hyg. 27:271-275.