Differential effects of glucocorticosteroids on the functions of

Transcription

1 Proc. Natl. Acad. Sci. USA Vol. 78, No. 5, pp , May 1981 Immunology Differential effects of glucocorticosteroids on the functions of helper and suppressor T lymphocytes (immune regulation/dexamethasone/macrophage factors/t cell growth factors/autoimmunity) LINDA M. BRADLEY AND ROBERT I. MISHELL* Department of Microbiology and Immunology, University of Califbrnia, Berkelev, California Communicated by Philip Levine, November 17, 1980 ABSTRACT The effects of dexamethasone (Dex) on the functions of antigen-primed helper and suppressor T cells were studied in humoral immune responses in vitro. In doses equivalent to elevated physiologic concentrations, the suppressor T cell activity was abolished. In contrast, the helper T cell function was resistant to even pharmacologic concentrations of Dex. The apparent steroid resistance of the helper T cells was found to be mediated by the products of activated macrophages. While macrophage factors protected helper T cells from steroid inhibition, they did not prevent the effects of Dex on suppressor T cells. Because bacterial cell wall and membrane components are potent inducers of the factors that mediate steroid resistance of helper T cells, the combination of physiologically elevated levels of steroids and macrophage factors during acute infections may function to facilitate the expression of host immunity. However, the persistance of these conditions, as in chronic inflammation, may also contribute to the pathogenesis of autoimmunity by perturbing the balance of immune regulation by helper and suppressor T cells. Glucocorticosteroids generally suppress the immunological functions oft lymphocytes (1, 2). However, several reports suggest that, under particular conditions, T-cell functions resist steroid inhibition (3-11). These observations may reflect inherent differences either in the susceptibility of various T cells to steroids (3-6) or in the capacity of some T cells to develop resistance to the hormones (7-11). Moreover, physiological mechanisms may exist that result in differential sensitivity of functionally distinct T cells to steroids. Such mechanisms together with elevated physiologic levels of steroids might alter immune expression by affecting the balance of helper and suppressor T cell regulatory activities. We analyzed the effects of dexamethasone (Dex) on the functions ofantigen-primed helper and suppressor T cells in the generation of secondary humoral immunity in vitro. Using cells from recently primed mice, we found that doses of steroids equivalent to high physiologic concentrations selectively inhibited suppressor T cell activity in vitro. Our previous work (10, 11) had shown that helper T cells become resistant to steroids in response to adjuvant-activated monocytes/macrophages. In this report, we present data indicating that the differential steroid effects on helper and suppressor T cell functions occur because of macrophage-mediated protection of helper cells. MATERIALS AND METHODS Animals. BDF1 and BALB/k mice, 3-4 months old, bred in our mouse colony were used in all experiments. Antigens and Immunizations. Sheep erythrocytes (SRBC) were from GIBCO (lot ). 2,4-Dinitrophenyl-conjugated The publication costs ofthis article were defrayed in part by page charge payment. This article must therefore be hereby marked "advertisement" in accordance with 18 U. S. C solely to indicate this fact keyhole limpet hemocyanin (DNP-KLH) with 8 mol of dinitrophenyl per 10,000 g of KLH was the gift of Claudia Henry (University of California, Berkeley, CA). Carrier-primed mice were immunized intravenously with 2 X 106 SRBC 3 days prior to use. Hapten-primed mice were immunized initially intraperitoneally with 100,ug of alum-precipitated DNP-KLH (12) and 2 x 109 Bordetella pertussis vaccine organisms (Commonwealth of Massachusetts, Department of Public Health) and after 8-20 weeks were boosted with 20 pug of soluble DNP-KLH; the mice were used 14 days later. Dex. The Dex (lot 54C0375, Sigma) was the gift of Harry Rubin (University of California, Berkeley, CA) and was used as described (10). X-irradiation. Cells suspended at 5 x 106 per ml in balanced salt solution (13) were irradiated with a Norelco M6150 X-ray machine and received 2700 roentgens (carrier-primed spleen cells) and 1000 roentgens [activated macrophages (MEL)] (1 roentgen = 2.6 x 10-4 coulomb/kg). T-Cell Depletion. Normal spleen cells or peritoneal cells were treated with rabbit antiserum to mouse brain (14, 15) and guinea pig complement (16). Macrophage Depletion. Mb were removed by passage over columns of Sephadex G-10 (17, 18). Activated M41. T-cell-depleted spleen cells were cultured in 60-mm Corning petri dishes at 4 x 106 per ml in 3-ml volumes of RPMI 1640 medium (GIBCO) containing 5% fetal calf serum and 50 p.m 2-mercaptoethanol as described (19). After 4 hr, lipopolysaccharide (LPS) purified from Salmonella typhimurium (20) was added at 0.1,ug/ml. After 24 hr, the cells were harvested and X-irradiated. Macrophage Factors (M4F). These factors were prepared from T-cell-depleted resident peritoneal cells cultured with LPS as described (10). T Cell Growth Factors (TCGF). Partially purified TCGF were prepared as described (21) from BALB/k spleen cells. Supernates were concentrated by membrane filtration, precipitated by (NH4)2SO4 (40-80% saturation), and passed sequentially over columns of Sephadex G-200 and G-75. Fractions that contained TCGF and negligible amounts of concanavalin A (as determined by their lack of mitogenicity inhibitable by a- methyl mannoside) were pooled. Cell Cultures. Spleen cells were cultured as described (19, 22). Carrier-primed helper or suppressor T cell activities were assessed in two ways: (i) SRBC-primed spleen cells were either untreated or irradiated as described above. Then 8 x 106 of Abbreviations: Dex, dexamethasone; SRBC, sheep erythrocytes; DNP- KLH, 2,4-dinitrophenylated keyhole limpet hemocyanin; LPS, lipopolysaccharide; HRBC, horse erythrocytes; TNP, 2,4,6-trinitrophenyl; MO, macrophages; MOF, macrophage factors; TCGF, T-cell growth factors; PFC, plaque-forming cells. * To whom reprint requests should be addressed.

2 3156 Immunology: Bradley and MishellProc. Natl. Acad. Sci - USA 78 (1981) these cells and 8 x 106 DNP-KLH-primed spleen cells were cultured in l-ml volumes (12) with or without Dex or MOF. The cultures were immunized with 3 X 1(P 2,4,6-trinitrophenylmodified SRBC (TNP-SRBC) (12, 23). After 5 days, indirect (IgG-secreting) plaque-forming cells (PFC) were enumerated. (ii) SRBC-primed spleen cells, either untreated or passed through Sephadex G-10, were cultured for 24 hr in 3-ml volumes at 1 x 107 cells per ml in 60-mm petri dishes. Experimental additions to these cultures included 1 AM Dex, activated MbD, MWF. At 24 hr the cells were harvested, and irradiated to remove T suppressor activity. After preculture, 8 X 106 carrier-primed cells were combined with 8 X 106 T-cell-depleted nonimmune spleen cells and cultured with 3 x 105 TNP-SRBC in 1-ml volumes. After 5 days, direct (IgM-secreting) PFC to TNP were enumerated Hemolytic Plaque Assays. TNP-specific PFC were determined by using a slide modification (22, 24) of the method of Jerne and Nordin (25) with horse erythrocytes (HRBC) (Colorado Serum, Denver, CO), lightly modified with TNP (26, 27). In two-step experiments the number of direct (IgM-secreting) PFC specific for TNP was determined. In experiments employing a single culture period, indirect (IgG-secreting) TNPspecific PFC were enumerated by subtracting direct PFC from the total PFC obtained with a rabbit polyvalent antiserum to mouse Ig (24). RESULTS Differential Inhibitory Effects of Dex on Helper and Suppressor T Cell Activities. To study the effects of Dex on helper and suppressor T cell activities, mice were primed with heterologous erythrocytes by an immunization regimen that elicits carrier-specific helper and suppressor T cells for in vitro antihapten responses (28). Suppressor activity in the carrier-primed population inhibits much of the secondary (IgG) humoral immunity generated by an independently primed source of hapten-specific B cells. However, helper activity in the carrierprimed population can be measured because it is radioresistant and the suppressor activity is not. The difference in the antihapten responses of cultures containing nonirradiated and irradiated carrier-primed cells provides a measure of suppressor T cell activity in the carrier-primed population. The effects of Dex on suppressor activity were determined by adding various concentrations of Dex to cultures containing mixtures of hapten-primed spleen cells and nonirradiated carrier-primed cells and measuring IgG anti-tnp responses 5 days later. These were compared to responses of controls in which the carrier cells were irradiated. Data from three such experiments are shown in Table 1. Suppressor activity in the nonirradiated carrier-primed cells inhibited more than 80% of the response ofcontrols with X-irradiated carrier-primed cells. Responses similar to: those of X-irradiated (helper) controls were obtained when Dex was added to cultures containing nonirradiated carrier-primed cells. These data suggest that Dex inhibited suppressor but not helper activity. The apparent resistance of helper activity to Dex under these experimental conditions was documented by the experiment shown in Fig. 1, where the effects of Dex on cultures containing irradiated or nonirradiated carrier-primed cells are compared. These data show that Dex selectively inhibits suppressor activity. MSI Mediate the Steroid Resistance of Helper T Cells. Our earlier studies showed that helper T cells from unimmunized mice (11) and from mice primed days previously with SRBC (10) are inhibited by steroids but protected by adjuvantactivated M() (and MFF). This suggested that the resistance to Dex of the helper activity in the carrier-primed cells seen in the experiments of Table 1 and Fig. 1 might be due to MbI ac- Table 1. by Dex Inhibition of carrier-primed suppressor T cell activity Carrier-primed PFC per. culture cells Dex, M Exp. 1 Exp. 2 Exp. 3 X-irradiated None 22,000 2,100 15,000 Nonirradiated None 2, , , , ,000 12,000 18, ,000 11,000 37, ,000 14,000 33, ,000 ND 30,000 Spleen cells obtained from BDF, mice immunized 3 days previously with SRBC were used as the carrier-primed population. X-irradiated carrier-primed cells received 2700 roentgens immediately before culture. Carrier-primed. cells were combined with hapten-primed spleen cells obtained from BDF1. mice hyperimmune to. DNP-KLH and were cultured with TNP-SRBC. Dex at the concentrations indicated was added when the cultures were established. Results are IgG anti-tnp PFC per culture, determined after 5 days. ND, not determined. tivated by the immunization. To test this idea, the carrierprimed population was, depleted of MO. by passage over columns of Sephadex G-10. The column-filtered cells were initially cultured for 24 hr with or without Dex (1,uM), and with or without adjuvant-activated MbI or MOIF. Unfiltered carrierprimed cells were also incubated with or without Dex for 24 hr. The cells were then harvested, washed, and X-irradiated. Helper activity of the cultured carrier-primed populations was assayed by measuring the humoral responses to TNP ofcultures of freshly prepared T-cell-depleted spleen cells to which the precultured cells were added. 62) $ Z4 10,000 / I o Dex, M -T I I FIG. 1. Resistance of carrier-primed helper T cell activity to Dex. Cells were prepared and cultured as described in the legend of Table 1. Dex was added at the indicated concentrations when the cultures were established. The carrier-primed cells were either X-irradiated with 2700 roentgens before culture (A) or not.x-irradiated (e). Results are IgG anti-tnp PFC per culture assayed after 5 days.

3 Immunology: Bradley and Mishell Proc. Natl. Acad. Sci. USA 78 (1981) , ,000 F a) u S. v) 1,000 - Q) Dex + Dex - Dex + Dex -Dex + Dex Medium Activated M(P MPF FIG. 2. Inhibition of helper T cell activity of MbD depleted carrierprimed spleen cells by Dex and reversal by activated MS or MFF. Spleen cells from mice primed to SRBC were passed through Sephadex G-10 columns to remove MbF. Sephadex G-10-filtered carrier-primed cells were initially cultured (1 x 107 per ml) in the presence or absence of 1 jum Dex. Control cultures ("Medium") received no further additions. Activated MOD (3 x 10' per ml) or M4F (at a final concentration of20%) were added as shown. After 24 hr, cells from individual cultures were separately harvested, X-irradiated (2700 roentgens), and combined with T-cell-depleted normal spleen cells and TNP-modified SRBC. After further culture for 5 days, the number of TNP-specific PFC was determined. Results are direct anti-tnp PFC per culture. T- cell-depleted normal cells generated 14,000 PFC per culture when combined with unfiltered, X-irradiated, carrier-primed cells preincubated without Dex; 13,000 PFC per culture when combined with unfiltered, X-irradiated, carrier-primed cells preincubated with Dex; and 160 PFC per culture when cultured alone. As shown in Fig. 2, when the carrier-primed population was depleted of MO, helper activity was inhibited 90% by pretreatment with Dex. Activated Mb or MFF blocked the steroid effects. These results show that the helper T cells recovered from Sephadex G-10 filtration of the carrier-primed population are steroid sensitive and support the hypothesis that these T cells resist Dex because ofthe protective effects ofmb activated in situ by antigen. Failure of M@F to Protect Suppressor T Cell Activity from Steroid Inhibition. The results on the right offig. 2 showed that mediators from activated Mb protect carrier-primed helper T cells from steroids and suggest that the differential effects ofdex on the functions of helper and suppressor T cells (Table 1, Fig. 1) are due to the selective capacity of helper T cells to become steroid resistant in response to MOFF. Because the concentration ofmof was unknown (and may have been marginal), these experiments did not exclude the possibility that suppressor cells could be protected by them. We examined the effects of additional MOF to determine if they could block Dex inhibition of the suppressor cells. The results of such an experiment (Fig. 3) show that doses of MDFF that were optimal in blocking steroid effects on helper activity had no effect in protecting suppressor activity. Similar results were obtainedwith higher concentrations of M4F. In addition, we found that suppressor T cell activity induced by concanavalin A (29) is also Dex sensitive and is not protected by the MOF (data not shown). Thus, suppressor T cells differ from helper 1,000 Positive - Dex + Dex - Dex + Dex control Medium MEPF FIG. 3. Inhibition of carrier-primed suppressor T cell activity by Dex is not reversed by supernates of activated Mb. Carrier- and hapten-primed cells were generated and cultured as described in the legend of Table 1. Dex at 1,uM and MFF at a final concentration of 20% were added as indicated when cultures were established. The positive control contained carrier-primed cells irradiated with 2700 roentgens. Results are IgG anti-tnp PFC per culture assayed after 5 days. T cells in failing to respond to the protective effects of activated Mb Ḟailure of TCGF to Protect Suppressor T Cell Activity from Steroid Inhibition. Although M(DF do not enable suppressor T cell activity to resist steroid inhibition (Fig. 3), Gillis et al. (30) reported that proliferating T cells are protected from Dex by TCGF. Because the function of suppressor T cells is abolished by treatments that inhibit proliferation [e.g., mitomycin C (31), and X-irradiation], their studies suggested that TCGF might prevent Dex inhibition of suppressor T cell activity. Therefore, we examined the effects of partially purified TCGF on the function ofcarrier-primed suppressor T cells on IgG anti- TNP responses, in the presence and absence of Dex. The data from a representative experiment are shown in Table 2. TCGF, in doses that support maximal growth offactor-dependent T cell Table 2. Failure of TCGF to protect carrier-primed T suppressor activity from Dex Carrier-primed PFC per culture cells Dex, M Medium TCGF X-irradiated None 16,000 27,000 Nonirradiated None 3,700 3, ,000 28, ,000 24, ,000 22,000 Carrier- and hapten-primed cells were prepared and cultured as described (Materials and Methods, Table 1 legend). Controls were cultured in medium alone. Experimental cells were cultured in medium supplemented with TCGF at concentrations in excess ofthose that fully support the growthofa TCGF-dependent cell line (Materials and Methods). Dex at the concentrations indicated was added when the cultures were established. Results are IgG anti-tnp PFC per culture determined after 5 days.

4 3158 Immunology: Bradley and Mishell lines and that prevent Dex inhibition of mitogen responses (30), had no apparent effect on the expression ofeither helper or suppressor activities in control cultures and did not alter the Dex effects on suppressor activity. DISCUSSION In this report, the effects of Dex on coexisting helper and suppressor T cell activities, expressed by murine spleen cells shortly after in vivo immunization with heterologous erythrocytes (28), were investigated in secondary anti-hapten responses in vitro. Under an experimental protocol in which the carrierprimed population expressed predominantly suppressor activity, Dex significantly increased the generation of hapten-specific humoral immunity (Table 1). When the suppressor activity was eliminated by X-irradiation, the magnitude of humoral immunity was unaltered by Dex, and anti-hapten responses comparable to those generated by steroid-treated cultures with nonirradiated carrier-primed cells were observed (Fig. 1). These results suggest that when both helper and suppressor T cells are present in the carrier-primed population Dex facilitates the expression of positive immunity by selectively inhibiting the suppressor T cells. The suppressor activity was eliminated by concentrations of Dex (10- M) that are equivalent to elevated physiologic levels of glucocorticosteroids (32). The coexistence ofhelper and suppressor T cell activities, described in a number ofreports (28, 33-35), suggests that the relative balance ofthese T cell functions regulates humoral immune expression. Because the plasma concentrations of glucocorticosteroids rise during acute infections and inflammatory reactions, the differential effects of these hormones on the activities of recently primed helper and suppressor T cells observed in this study may reflect the existence of physiologic mechanisms that promote the induction of host immunity under such circumstances. The resistance of helper activity to steroids in the carrierprimed population (Table 1, Fig. 1) is consistent with the report by Segal et al. (7) that the functions of recently activated helper T cells are unaffected by steroids in vivo. In an earlier study we found that secondary in vitro humoral responses are less susceptible to steroid inhibition when cells are obtained 3 days after in vivo priming than at 1 week (36). We subsequently showed that, although helper T cells from unimmunized mice and from mice primed days previously are suppressed by steroids in vitro, they acquire resistance in response to adjuvant-activated Mb or M4IF (10, 11). These studies suggested that shortly after carrier priming, M4b activated in vivo by encounter with antigen contributed to the steroid resistance of helper activity. The role of MO in the steroid resistance ofhelper cells in this study was shown by the susceptibility of the helper cells to Dex after depletion of Mb from the carrier-primed population (left of Fig. 2). Adjuvant-activated M(F and M1F restored the resistance of these cells (center and right of Fig. 2). In contrast, the Dex sensitivity of the suppressor activity was unaltered by M(IF (Fig. 3). These data suggest that the differential steroid effects on carrier-primed helper and suppressor T cell activities (Fig. 1) are due to the selective ability of helper T cells to respond to the factors released by activated MFD. We have shown elsewhere that the M1F do not function by inactivating steroids (10). The physicochemical characteristics of the factors (11) indicate close similarity or identity to mediators designated interleukin 1 (IL-1) (37). Recently, Gillis et al. (30) and Larsson (38) independently demonstrated that Dex inhibits the production of TCGF (interleukin 2) by mitogen-activated lymphocytes. Furthermore, they showed that TCGF abrogated Dex inhibition oft cell proliferative responses to plant lectins, leading them to postulate Proc. Natl. Acad. Sci. USA 78 (1981) that steroids exert their inhibitory effects on proliferating T cells indirectly by preventing TCGF production. Although the carrier-primed suppressor T cells in our experiments, proliferate, as shown by their sensitivity to X-irradiation (Table 1, Fig. 1), TCGF did not alter the Dex inhibition of their activity (Table 2). These results suggest that steroids inactivate suppressor T cells by mechanisms that do not involve TCGF. Because various bacterial cell wall and membrane components have been found to be effective inducers of the steroid protecting activity of Mb (11), it is probable that the synthesis and release of the factors that mediate this function are stimulated during bacterial infections. Because of the selective effects of these factors on helper T cells, the physiologic combination of elevated steroids and activated M4D may alter immune regulation by helper and suppressor T cells to facilitate positive immune expression. However, in chronic infections, when elevated levels of steroids persist, the continued stimulation of factor production could contribute to establishing autoimmunity due to the absence of regulation by suppressor T cells. This hypothesis may partially explain why potent MbF inducers such as peptidoglycan and other bacterial cell wall components that stimulate production of IL-1 also induce experimental (adjuvant) polyarthritis (39). The authors gratefully acknowledge the excellent technical assistance of Ms. Elaine Kwan. We thank Dr. Sally Fairchild for helpful discussion and for critically reviewing the manuscript. This work was supported by a grant from the Kroc Foundation and by Grants CA25056 and AI15482 from the National Institutes of Health. L. M. B. was supported by National Institutes of Health Grant CA Claman, H. N. (1972) N. Engl. J. Med. 287, Spreafico, F. & Anaclerio, A. (1977) in Immunopharmacology, eds. Hadden, J. W., Coffey, R. G. & Spreafico, F. (Plenum, New York), pp Markham, R. B., Stashak, P. W., Prescott, B., Amsbaugh, D. F. & Baker, P. J. (1978)J. Immunol. 121, Altman, A. & Cohen, I. R. (1975) J. Exp. Med. 142, Cohen, J. J., Fischback, M. & Claman, H. N. (1970)J. Immunol. 105, Moorhead, J. W. & Claman, H. N. (1972) Cell. Immunol. 5, Segal, S., Cohen, I. R. & Feldman, M. (1972) Science 175, Nachtigal, O., Zan-Bar, I. & Feldman, M. (1975) Transplant. Rev. 26, Schechter, B. & Feldman, M. (1977)J. Immunol. 119, Mishell, R. I., Shiigi, J. M., Mishell, B. B., Grabstein, K. H. & Shiigi, S. M. (1980) Immunopharmacology 2, Mishell, R. I., Bradley, L. M., Chen, Y. U., Grabstein, K. H. & Shiigi, S. M. (1979) Ann. N. Y. Acad. Sci. 332, Bradley, L. M. & Shiigi, S. M. (1980) in Selected Methods in Cellular San Francisco), pp Mishell, B. B. & Shiigi, S. M. (1980) Selected Methods in Cellular Immunology (Freeman, San Francisco), pp Golub, E. S. (1971) Cell. Immunol. 2, Shiigi, S. M. & Slomich, M. (1980) in Selected Methods in Cellular San Francisco), pp Cohen, A. & Schlesinger, M. (1970) Transplantation 10, Ly, I. A. & Mishell, R. I. (1974)J. Immunol Methods 5, Mishell, B. B., Mishell, R. I. & Shiigi, J. M. (1980) in Selected Methods in Cellular Immunology, eds. Mishell, B. B. & Shiigi, S. M. (Freeman, San Francisco), pp Mishell, B. B. & Mishell, R. I. (1980) in Selected Methods in Cellular San Francisco), pp Galanos, C., Luderitz, 0. & Westphal, 0. (1969) Eur. J. Biochem. 9, Watson, J., Aardin, L. A. & Lefkovits, I. (1979)J. Immunol. 122, Mishell, R. I. & Dutton, R. W. (1967)J. Exp. Med. 126,

5 Immunology: Bradley and Mishell Proc. Natl. Acad. Sci. USA 78 (1981) Kettman, J. & Dutton, R. W. (1970)J. Immunol. 104, Mishell, B. B., Shiigi, S. M. & Mishell, R. I. (1980) in Selected Methods in Cellular Immunology, eds. Mishell, B. B. & Shiigi, S. M. (Freeman, San Francisco), pp Jerne, N. K. & Nordin, A. A. (1963) Science 140, Rittenberg, M. B. & Pratt, K. L. (1969) Proc. Soc. Exp. Biol. Med. 132, Henry, C. (1980) in Selected Methods in Cellular Immunology, eds. Mishell, B. B. & Shiigi, S. M. (Freeman, San Francisco), pp Chan, E. L. & Henry, C. (1976)J. Immunol. 117, Dutton, R. W. (1972)J. Exp. Med. 136, Gillis, S., Crabtree, G. R. & Smith, K. A. (1979) J. Immunol. 123, Swain, S. L., Trefts, P. E., Tse, H. Y. S. & Dutton, R. W. (1977) Cold Spring Harbor Symp. Quant. Biol. 41, King, R. J. B. & Mainwaring, W. I. P. (1974) Steroid-Cell Interactions (Univ. Park Press, Baltimore), p Cantor, H., Shen, F. W. & Boyse, E. A. (1976)j. Exp. Med. 143, Tada, T. & Takemori, T. (1974) J. Exp. Med. 140, Whistler, R. J. & Stobo, J. D. (1976)J. Exp. Med. 144, Mishell, R. I., Lucas, A. & Mishell, B. B. (1977)j. Immunol. 119, Aarden, L. A. et al. (1979)J. Immunol. 123, Larsson, E. L. (1980)J. Immunol. 124, Kohashi, O., Pearson, C. M., Watanabe, Y., Kotani, S. & Koga, T. (1976)J. Immunol. 116,