Eur. J. Immunol. 1991. 21: 815-818
Enterotoxin B as a potent suppressant of Tcells
815
Short paper Avraham Ben-Nun Department of Cell Biology, m e Weizmann Institute of Science, Rehovot
Staphylococcal enterotoxin B as a potent suppressant of T cell proliferative responses in rats* The staphylococcal toxins are well known as stimulators of powerful T lymphocyte proliferative responses. Staphylococcal enterotoxin B (SEB) was found, in both humans and mice, to preferentially stimulate T cells bearing particular Vg gene products as part of their functional Tcell receptor a/& In contrast to this reported stimulatory activity we demonstrate here that SEB is a poor stimulant of T lymphocyte proliferative responses in Lewis rats. Moreover, under appropriate conditions, SEB can serve as a powerful suppressant of Lewis rat T lymphocytes, capable of abolishing their antigen-specific or mitogenstimulated proliferative responses. Suppression of mitogen-induced proliferative responses of rat T cell clones was effective in the presence of syngeneic or allogeneic accessory cells and similar to its stimulatory characteristics, SEB suppressed high proportions of rat T lymphocyte subsets.These data suggest that in certain circumstances SEB can be considered a “supersuppressogen”as well as a “superantigen”.Whether SEB will be “superantigen”or “supersuppressogen” is likely to depend on theTcR elements of the lymphocytes as well as on the MHC molecules presenting SEB.
1 Introduction The staphylococcal enterotoxins are a group of proteins secreted by Staphylococcal aureus.These proteins encoded by bacteriophages are the cause of staphylococcal food poisoning and severe shock in humans and experimental animals [l,21. Recently, it has been demonstrated that staphylococcal enterotoxins bind directly to class I1 MHC molecules [3, 41. In the presence of syngeneic and some allogeneic accessory cells expressing MHC class I1 molecules, many of these staphylococcal toxins stimulate powerful proliferative Tcell responses [ 5 ] . However, in contrast toTcell mitogens, the staphylococcal toxins are stimulatory toTcells bearing only particularVp elements as part of their TcR a l p . It has recently been demonstrated that staphylococcal enterotoxin B (SEB) preferentially stimulates mouse T cells bearing Vg3, 7, 8.1, 8.2, 8.3 and 17 gene products [5-71 and humanTcells bearingVg3,12,14,15 and 17 gene products [8]. The minor lymphocyte-stimulating determinants (MIS), as endogenous antigens, display similar characteristicsof Tcell stimulation. Mls-la preferentially stimulates T lymphocytes bearing the Vp6, -8 and 9 gene products [9-111, and M I s - ~preferentially ~ stimulates T lymphocytes bearing the Vg3 element [12, 131. SEB and MIS,which stimulate subsets of T lymphocytes on the basis of their Vp usage without strict MHC restriction, have been termed “superantigens” [13].
In contrast to the reported stimulatory characteristics of SEB, here we present studies demonstrating that SEB is not stimulatory to the majority of rat Tcell subsets and that, under certain conditions, SEB can be a highly potent suppressant of T lymphocytes.
2 Materials and methods 2.1 Animals and antigens
All rats were 2-3-month-old females, obtained from the Weizmann Institute Animal Breeding Center. The mice were from the Jackson Laboratories (Bar Harbor, ME) and they were used at the age of 2-3 months. Myelin basic protein (MBP) was prepared from guinea pig spinal cords, as previously reported [14]. SEB, OVA and Con A were purchased from Sigma Chemicals (St. Louis, MO), the PPD of Mycobacterium tuberculosis was obtained from Statens Seruminstitut (Copenhagen, Denmark) and the M. tuberculosis was from Difco (Detroit, MI).
2.2 T cell clones
The Zla, D9, A2b, LOA and PPD are CD4+ Tcell clones derived from Lewis rats.The Zla and D9 are MBP-specific * This work was supported by a Cancer Research InstituteIAlbert clones causing EAE in normal syngeneic recipients [15]. and Cherry Krassner Investigator Award. The A 2 b clone is specific to mycobacterial antigen, crossIncumbent of the Swig-Weiler Career Development Chair at the reactive with rat cartilage, and it is involved in mediating Weizmann Institute of Science. adjuvant arthritis (AA; [16]). The LOA and LPD are Correspondence:Avraham Ben-Nun, Department of Cell Biology, specific to OVA and PPD, respectively. The SR-1 [17] and BUR-1 (unpublished) clones are also CD4+ Tcell clones, The Weizmann Institute of Science, Rehovot 76100, Israel specific to MBP, that were derived from SJL/J and B1O.PL Abbreviatons: AA: Adjuvant arthritis MBP: Myelin basic pro- mice, respectively; they are also functional in causing EAE ([17] and unpublished results). tein SEB: Staphylococcal entrotoxin B
[I 88601
0 VCH Verlagsgesellschaft mbH, D-6940 Weinheim, 1991
+
0014-298019110303-0815$3.50 .25/O
816
Eur. J. Immunol. 1991. 21: 815-818
A. Ben-Nun
2.4 Proliferative responses
effective in stimulating proliferative responses by the MBP-specific T cell clones from SJL/J mice (Fig. 1B).
T cell clones (105/ml) and irradiated accessory rat thymocytes (5 x 106/ml) or mouse spleen cells (2.5 x 106/ml) were The Zla, D9 and A2b clones are T cell clones using the cultured in microtiter wells in 0.2 ml (per well) RPMI Vg8.2 gene for their functional TcR a@ (as determined by medium supplemented with L-glutamine, 2-ME, antibiotics Northern analysis with Vg8.2-specificoligonucleotide; data and 2.5% FCS.The respective specific antigen (50 pg/ml) or not shown).The rat Vg8.2 gene is highly homologous (87%) Con A (1 pg/ml) and/or SEB (at increasing concentrations) with the mouseVg8.2 (19,201. According to recent findings were added in triplicate cultures and the cultures were demonstrating that SEB preferentially stimulates T cells incubated for 48 h. For assessing the proliferative responses bearingVg8.1,8.2 and 8.3 elements [5,6], the Zla, D9 and of thymocytes or splenocytes, rat or mouse thymus cells or A2b T cells bearing the rat Vg8.2 gene were expected to spleen cells were cultured in microtiter wells in 0.2 ml proliferate in response to SEB. Because of this unresponmedium, as above, Con A (1 pg/ml) and/or SEB were siveness of Vg8+ rat Tcell clones to SEB,we investigated the added in triplicate cultures and the cultures were incubated effect of SEB on the thymic and splenicTcell populations of for 60-62 h. [3H]dThd(37 kBq/well) was added for the last normal Lewis rats. Fig. 1C shows that at the level of the 16 h. Results are expressed as mean [3H]dThd incorpora- whole cell population in spleen and thymus, SEB was a tion (cpm) of triplicate cultures; SE were < 10% of the poor stimulator of Lewis rat Tcell proliferative responses, mean. while stimulating powerful proliferative responses by mouse splenocytes or thymocytes (Fig. 1D).
3 Results In a previous study, we showed that the SEB and Mls superantigens stimulate SJL/J-derived mouse T cell clones specific to MBP and activate their pathogenicity to mediate EAE in normal syngeneic recipients (Soffer, D. and Ben-Nun, A . , submitted for publication). Subsequently we wished to investigate the effects of SEB on EAE in Lewis rats and on their h4BP-specificTcell clones which mediate the disease. However, we found (Fig. 1A) that SEB was not stimulatory for the Lewis rat MBP-specificTcel1clones Zla and D9 Tcell clones functional in mediating EAE [15, 181 or for the A2b T cell clone specific to a mycobacterial antigen which is cross-reactive with rat cartilage and involved in A A [16]. SEB was also nonstimulatory for nonautoimmune Tcells, the LOA and LPD clones specific to OVA and PPD, respectively (Fig. 1A). In contrast to its failure in stimulating the rat T lymphocytes, SEB was highly
To further investigate the effect of SEB on rat T lymphocytes, we activated Lewis rat thymocytes or splenocytes in vitro, in the presence of SEB, using Con A. As shown in Fig. 2A, the presence of SEB imposed a powerful, dosedependent suppression on the proliferative responses by Lewis rat thymocytes or splenocytes to Con A. The presence of SEB (from the same batch) in the in vifroculture of Con A activation of SJL/J thymocytes or splenocytes did not suppress their activation by Con A. On the contrary, due to its stimulatory effect on murine T cells, the SEB further supported their activation by Con A (Fig. 2B). The proliferative responses of thymocytes and splenocytes of other rat strains were also affected by SEB. However, as shown in Fig. 2C and D, the thymocytes and the splenocytes from the different rat strains displayed different levels of sensitivities to the suppression effect of SEB. Thus, although BN thymcoyte stimulation by Con A was strongly
100-
-
La-
0
(0-
m-
p!
7
x
,
20,
[SEB] lg/ml
Figure 1. SEB is non-stimulatory to Lewis rat Tcells. The proliferative responses to SEB of (A) Lewis rat CD4+ Tcell clones specific for MBP (Zla, D9), M. tuberculosis (A2b), OVA (LOA) and PPD (LPD): (B) SJL/J CD4+ Tcell clones specific for MBP (SR-1). OVA (SOD) and PPD (SPD); (C) Lewis rat thymocytes and splenocytes; ( D ) SJLN thymocytes and splenocytes.
[SEB] pg/ml
Figure 2. Suppression of the mitogenic stimulation of rat Tcells by SEB. The thymocytes (l@/well), splenocytes (5 x 10Vwell) from Lewis rat (A), SJL/J mice (B) or different strains of rats (C and D) were activated with Con A (1 pglml) in the presence of indicated concentrations of SEB. Proliferative responses were assessed by [3H]dThd uptake as mentioned in Sect. 2.4.
Enterotoxin B as a potent suppressant of Tcells
Eur. J. Immunol. 1991.21: 815-818
817
The results (Fig. 2D) showing that SEB was suppressive for Lewis, Fisher and Wistar, but not BN, spleen cells indicated that the suppression by SEB could not be attributed to nonspecific cytotoxic activity. To gain more information concerning the nature of the suppression effect by SEB, Lewis rat Tcell clones were activated with Con A in the presence of SEB and irradiated syngeneic or allogeneic thymocytes, or of xenogeneic mouse spleen cells, as 00 01 .I 1 10 tm accessory cells expressing MHC class I1 molecules. SEB was highly effective in suppressing the activation of Lewis rat T cell clones in the presence of syngeneic (Fig. 4A) as well as allogeneic rat accessory cells (Fig. 4B). However, when xenogeneic mouse spleen cells were used as accessory cells, SEB failed to suppress the proliferative responses of the Zla rat T cell clone activated by Con A (Fig. 4C). Conversely, the SEB presented by the mouse accessory cells was highly stimulatory of proliferative responses by the Lewis rat T cell clones (Fig. 4C; similar results were [=31 Clg/ml obtained with D9 and A2b clones), eliminating the possiFigure 3. Suppression by SEB of the antigen-specific and mitogenic stimulation of Lewis rat Tcell clones. Lewis rat Tcell clones bility that the highly potent suppression by SEB could be Zla (A) and D9 (B) specific to MBP, the A2b clone (C) specific to attributed to a nonspecific cytotoxic effect on the rat Tcells. M. tuberculosis and LOA clone (D) specific to OVA were cultured In addition, SEB could not suppress the Con A stimulation (l@/well) in the presence of their specific antigen (10 &well) or of the BUR-1 MBP-specific T cell clone derived from B1O.PL mice in the presence of either mouse (Fig. 4C) or with Con A (1 &well) and in the presence of SEB. rat accessory cells (Fig. 4A and B). On the contrary, SEB presented by rat accessory cells stimulated, in the absence of Con A, powerful proliferative responses by the murine suppressed, thymocytes from Wistar and Fisher rats were MBP-specific T cell clones to comparable levels as in the more sensitive to suppression by SEB. Splenocytes were presence of their respective syngeneic accessory cells less sensitive than thymocytes, and BN or Wistar spleno- (Fig. 4A-C). cytes were far less sensitive to the suppression by SEB. ~~
SEB had a powerful suppression activity also on the in vim-maintained, antigen-specific rat Tcell clones.The Zla (Fig. 3A), D9 (Fig. 3B) and A2b (Fig. 3C) Tcell clones and the nonautoimmuneTcell clone specific to OVA (Fig. 3D) were strongly suppressed when activated by their respective specific antigen or by Con A in the presence of SEB. The suppression of theTcell clones, as well as of the thymocytes and splenocytes derived from Lewis rats, was extremely potent. SEB (5 x lop2pg/ml) was sufficient to cause 50% suppression of the T cell proliferative response stimulated by the specific antigen or by aTcell mitogen (Figs. 2A and C). In contrast, the optimal stimulation of mouse T cells required 10-20 pg/ml of SEB.Thus, the suppression of rat T cell activation by SEB was at least 10 times more potent than its stimulatory activity on mouse Tcells.
*-
(A) 100-
x
+ 21. + ZlaOnA
loo -
+ BURl
*
.
4 Discussion The results presented above demonstrate that not only was SEB a poor stimulant of rat T lymphocyte proliferative responses (Fig. l ) , but it was also a highly potent suppressor of rat T lymphocytes' proliferative responses stimulated by a Tcell mitogen or by the specific antigen (Figs. 2-4). The failure of SEB to stimulate rat T lymphocyte proliferative responses could not be explained by the inability of SEB to bind to rat MHC molecules, nor by a major deletion of T lymphocyte subsets bearing the appropriate Vg elements required for recognizing SEB. These conclusions are based on the results showing that SEB associated with rat MHC is a powerful stimulant of proliferative responses
21.
(B)
(c)
+ Zla+ConA + BUR1 &
BURl+ConA
21.
+ ZlarConA
100-
BURl
& BURl+ConA
BUR.l+CmA
20-
0-• 0.0
.01
.l
1
lo
loo
0.0
.01
.I
1
lo
loo
0.0
01
.I
1
lo
loo
[SEB] W d
Figure 4. The mitogenic activation of rat T lymphocytes is suppressedby SEB associated with syngeneic or allogeneic, but not xenogeneic APC.Tcell clones derived from Lewis rats (Zla) or B1O.PL mice (BUR-1) were stimulated with SEB or were activated with Con A in the presence or absence of SEB and in the presence of APC from Lewis rat (A), BN rat (B) or BlO.PL mouse (C).
818
A. Ben-Nun
by mouse Tcell clones (Fig. 3C), but not rat Tcell clones (Fig. l), whereas SEB associated with mouse MHC is a powerful stimulant of both mouse and rat T lymphocyte proliferative responses (Fig. 4).Thus, SEB can bind the rat MHC and rat T lymphocytes do recognize SEB. However, the SEB/rat MHC complex, in contrast to the SEB/mouse MHC complex, could not stimulate large proportions of rat T lymphocyte subsets. The failure of SEB presented by syngeneic or allogeneic rat APC to stimulate rat Vp8+ T lymphocytes (Figs. 3 and 4) and the powerful stimulation of rat Vg8+ T lymphocyte proliferative responses by SEB associated with mouse APC (Fig. 4) indicated that SEB may bind to a nonpolymorphic region of the rat class I1 MHC. which is not conserved between mice and rats.
Eur. J. Immunol. 1991.21: 815-818
all Vg8+ rat T cells studied (Fig. 3). The suppression of different Vg8+ rat T cells, irrespective of their antigenic specificity and the suppression of large proportions of rat T lymphocyte subsets, suggests that SEB may well be considered a “supersuppressogen”as well as a “superantigen”. Whether SEB will be a “superantigen”or a “supersuppressogen” is likely to depend on the TcR elements of the lymphocytes as well as on the MHC presenting the SEB. 1 am grateful to Ms. Sara Yossefi for her expert technical assistance and 1 thank Prof. T E Davies, 1. R. Cohen and G. Berke for critically reading and commenting on the manuscript, and Ms. Malvine Baer for preparing the manuscript and for editorial assistance.
The mechanism by which SEB imposes a strong suppression on rat T lymphocytes responding to their specific Received September 4, 1990. antigen or to aTcell mitogen is not clear.The possibility that SEB is nonspecifically cytotoxic for rat Tcells or accessory cells was excluded by showing that SEB associated with 5 References mouse MHC is stimulatory for rat T lymphocytes and SEB 1 Bergdoll, M. S., in Riemann, H. and Bryan, F. L. (Eds.), associated with rat MHC is stimulatory to mouse T cells. Foodborne Infections and Intoxications, Academic Press, New These experiments also ruled out the possibility that Lewis York 1979, p. 443. rat cells expressing SEBMHC confer Tcell anergy due to 2 Betley, M. J. and Mekalanos, J. J., Science 1985. 229: 185. their incapability to provide co-stimulatory signals [21]. 3 Mollick, J. A . , Cook, R. G . and Rich, R. R., Science 1989.244: MHC blockade [22], as another possible explanation for 817. the highly potent suppression, was also considered, SEB 4 Hermann, T., Accolla, R. S. and MacDonald, H. R . , Eur. J. Immunol. 1989. 18: 2171. was shown to bind directly to MHC class I1 molecules with a high binding affinity 1231. Therefore, SEB as a nonstimula- 5 Janeway, C. A . , Jr., Immunol. Rev. 1989. 56: 61. 6 White, J., Herman, A . , Pullen, A . M.. Kubo, R., Kappler, J.W. tory antigen for Lewis rat T lymphocytes may simply and Marrack, P., Cell 1989. 56: 27. occupy the MHC-binding sites and block the antigen7 Callahan. J.. Herman. A . . Kaoder. J. and Marrack. F?. J. specific stimulation of Lewis rat T lymphocytes. This Immunol. 1990. 144: 2473. possibility, however, was unlikely, for three reasons: (a) 8 Kappler, J., Kotzin, B., Herron, L., Gelfand, E. W., Bigler, R. SEB did not always suppress the stimulation of Tcells for D., Boylston, A . , Carrel, S., Posnett, D. N., Choi, Y. and which it was nonstimulatory. BN splenocytes were not Marrack, €?, Science 1989. 244: 811. suppressed by SEB, even though SEB was only a poor 9 Kappler, J. W., Staerz, U., White, J. and Marrack, P., Nature 1988. 332: 35. stimulator for these lymphocytes (Fig. 2D); (b) suppression by SEB was usually not dependent on the SEB/Ag 10 MacDonald, R. H., Schneider, R., Lees, R. K., Howe, R. C., Acha-Orbea, M., Festenstein, H., Zinkernagel, R. M. and ratios. Increasing antigen or Con A concentrations did not Hengartner, M., Nature 1988. 332: 40. result in a significantly lower suppression by SEB (data not 11 Happ, M. I?, Woodland, D. L. and Palmer, E., Proc. Nad. shown); and (c) the non-MHC-restricted suppressogenic Acad. Sci. USA 1989. 86: 6293. effect of SEB indicated that SEB may bind to a nonpoly- 12 Pullen, A. P., Marrack. P. and Kappler, J.W., Nature 1988.335: morphic region of the MHC, as previously suggested for its 796. stimulatory effect [23], rather than to the highly polymorp- 13 Abe, R.,Vacchio, M. S., Fox, B. and Hodes, R. J., Nature 1988. hic region of MHC where the MHC binds the nominal 335: 827. 14 Hirschfeld, M.,Teitelbaum, D., Arnon, R. and Sela, M., FEBS antigens. , I
Interestingly, the characteristics of the rat T lymphocyte suppression by SEB resembled in several respects the stimulatory characteristics of SEB on mouse T lymphocytes. The suppression was dependent upon the interaction of T lymphocytes with accessory cells presenting SEB and the specific antigen or Tcell mitogen. Suppression by SEB was also dependent on MHC and similar to the stimulation by SEB, the suppression was not MHC restricted. Furthermore, the ability of SEB to suppress the mitogen stimulation of Lewis rat thymocytes and splenocytes indicated that SEB is a powerful suppressor affecting large proportions of Lewis rat T lymphocyte subsets. Whether SEB suppresses rat T lymphocytes on the basis of their Vp gene usage is an intriguingquestion which requires further investigation.We demonstrated here that SEB presented by rat MHC suppresses the mitogenic or antigen-specific stimulation of
Lett. 1970. 7: 317. 15 Ben-Nun, A.,Wekerle, H. and Cohen, 1. R., Nature 1981.229:
60. 16 Van Eden,W., Holoshitz, Z., Nevo, A . , Frenkel, A . , Klajman, A. and Cohen, 1. R., Proc. Nail. Acad. Sci. USA 1985. 82: 5117. 17 Ben-Nun, A. and Soffer, D., Eur. J. lmmunol. 1990. 20: 195. 18 Ben-Nun, A. and Cohen, I. R . , J. lmmunol. 1982. I28: 1450. 19 Bruns, F. R., Li, X., Shen, N . , Offner, H . , Chou, Y. K., Vandenbark, A . A . and Heber-Katz, E., J. Exp. Med. 1989. 169: 27. 20 Chluba, J., Steeg, C., Becker, A.,Wekerle, H. and Epplen, J.T.. Eur. J. lmmunol. 1989. 19: 279. 21 Schwartz, R. H., Science 1990. 218: 1349. 22 Guillet, J. G . ,Lai, M. Z., Briner,T. J., Smith, J. A. and Gefter, M. L., Nature 1986. 324: 260. 23 Marrack, P. and Kappler, J., Science 1990. 248: 705.
Enterotoxin B as a potent suppressant of Tcells
815
Short paper Avraham Ben-Nun Department of Cell Biology, m e Weizmann Institute of Science, Rehovot
Staphylococcal enterotoxin B as a potent suppressant of T cell proliferative responses in rats* The staphylococcal toxins are well known as stimulators of powerful T lymphocyte proliferative responses. Staphylococcal enterotoxin B (SEB) was found, in both humans and mice, to preferentially stimulate T cells bearing particular Vg gene products as part of their functional Tcell receptor a/& In contrast to this reported stimulatory activity we demonstrate here that SEB is a poor stimulant of T lymphocyte proliferative responses in Lewis rats. Moreover, under appropriate conditions, SEB can serve as a powerful suppressant of Lewis rat T lymphocytes, capable of abolishing their antigen-specific or mitogenstimulated proliferative responses. Suppression of mitogen-induced proliferative responses of rat T cell clones was effective in the presence of syngeneic or allogeneic accessory cells and similar to its stimulatory characteristics, SEB suppressed high proportions of rat T lymphocyte subsets.These data suggest that in certain circumstances SEB can be considered a “supersuppressogen”as well as a “superantigen”.Whether SEB will be “superantigen”or “supersuppressogen” is likely to depend on theTcR elements of the lymphocytes as well as on the MHC molecules presenting SEB.
1 Introduction The staphylococcal enterotoxins are a group of proteins secreted by Staphylococcal aureus.These proteins encoded by bacteriophages are the cause of staphylococcal food poisoning and severe shock in humans and experimental animals [l,21. Recently, it has been demonstrated that staphylococcal enterotoxins bind directly to class I1 MHC molecules [3, 41. In the presence of syngeneic and some allogeneic accessory cells expressing MHC class I1 molecules, many of these staphylococcal toxins stimulate powerful proliferative Tcell responses [ 5 ] . However, in contrast toTcell mitogens, the staphylococcal toxins are stimulatory toTcells bearing only particularVp elements as part of their TcR a l p . It has recently been demonstrated that staphylococcal enterotoxin B (SEB) preferentially stimulates mouse T cells bearing Vg3, 7, 8.1, 8.2, 8.3 and 17 gene products [5-71 and humanTcells bearingVg3,12,14,15 and 17 gene products [8]. The minor lymphocyte-stimulating determinants (MIS), as endogenous antigens, display similar characteristicsof Tcell stimulation. Mls-la preferentially stimulates T lymphocytes bearing the Vp6, -8 and 9 gene products [9-111, and M I s - ~preferentially ~ stimulates T lymphocytes bearing the Vg3 element [12, 131. SEB and MIS,which stimulate subsets of T lymphocytes on the basis of their Vp usage without strict MHC restriction, have been termed “superantigens” [13].
In contrast to the reported stimulatory characteristics of SEB, here we present studies demonstrating that SEB is not stimulatory to the majority of rat Tcell subsets and that, under certain conditions, SEB can be a highly potent suppressant of T lymphocytes.
2 Materials and methods 2.1 Animals and antigens
All rats were 2-3-month-old females, obtained from the Weizmann Institute Animal Breeding Center. The mice were from the Jackson Laboratories (Bar Harbor, ME) and they were used at the age of 2-3 months. Myelin basic protein (MBP) was prepared from guinea pig spinal cords, as previously reported [14]. SEB, OVA and Con A were purchased from Sigma Chemicals (St. Louis, MO), the PPD of Mycobacterium tuberculosis was obtained from Statens Seruminstitut (Copenhagen, Denmark) and the M. tuberculosis was from Difco (Detroit, MI).
2.2 T cell clones
The Zla, D9, A2b, LOA and PPD are CD4+ Tcell clones derived from Lewis rats.The Zla and D9 are MBP-specific * This work was supported by a Cancer Research InstituteIAlbert clones causing EAE in normal syngeneic recipients [15]. and Cherry Krassner Investigator Award. The A 2 b clone is specific to mycobacterial antigen, crossIncumbent of the Swig-Weiler Career Development Chair at the reactive with rat cartilage, and it is involved in mediating Weizmann Institute of Science. adjuvant arthritis (AA; [16]). The LOA and LPD are Correspondence:Avraham Ben-Nun, Department of Cell Biology, specific to OVA and PPD, respectively. The SR-1 [17] and BUR-1 (unpublished) clones are also CD4+ Tcell clones, The Weizmann Institute of Science, Rehovot 76100, Israel specific to MBP, that were derived from SJL/J and B1O.PL Abbreviatons: AA: Adjuvant arthritis MBP: Myelin basic pro- mice, respectively; they are also functional in causing EAE ([17] and unpublished results). tein SEB: Staphylococcal entrotoxin B
[I 88601
0 VCH Verlagsgesellschaft mbH, D-6940 Weinheim, 1991
+
0014-298019110303-0815$3.50 .25/O
816
Eur. J. Immunol. 1991. 21: 815-818
A. Ben-Nun
2.4 Proliferative responses
effective in stimulating proliferative responses by the MBP-specific T cell clones from SJL/J mice (Fig. 1B).
T cell clones (105/ml) and irradiated accessory rat thymocytes (5 x 106/ml) or mouse spleen cells (2.5 x 106/ml) were The Zla, D9 and A2b clones are T cell clones using the cultured in microtiter wells in 0.2 ml (per well) RPMI Vg8.2 gene for their functional TcR a@ (as determined by medium supplemented with L-glutamine, 2-ME, antibiotics Northern analysis with Vg8.2-specificoligonucleotide; data and 2.5% FCS.The respective specific antigen (50 pg/ml) or not shown).The rat Vg8.2 gene is highly homologous (87%) Con A (1 pg/ml) and/or SEB (at increasing concentrations) with the mouseVg8.2 (19,201. According to recent findings were added in triplicate cultures and the cultures were demonstrating that SEB preferentially stimulates T cells incubated for 48 h. For assessing the proliferative responses bearingVg8.1,8.2 and 8.3 elements [5,6], the Zla, D9 and of thymocytes or splenocytes, rat or mouse thymus cells or A2b T cells bearing the rat Vg8.2 gene were expected to spleen cells were cultured in microtiter wells in 0.2 ml proliferate in response to SEB. Because of this unresponmedium, as above, Con A (1 pg/ml) and/or SEB were siveness of Vg8+ rat Tcell clones to SEB,we investigated the added in triplicate cultures and the cultures were incubated effect of SEB on the thymic and splenicTcell populations of for 60-62 h. [3H]dThd(37 kBq/well) was added for the last normal Lewis rats. Fig. 1C shows that at the level of the 16 h. Results are expressed as mean [3H]dThd incorpora- whole cell population in spleen and thymus, SEB was a tion (cpm) of triplicate cultures; SE were < 10% of the poor stimulator of Lewis rat Tcell proliferative responses, mean. while stimulating powerful proliferative responses by mouse splenocytes or thymocytes (Fig. 1D).
3 Results In a previous study, we showed that the SEB and Mls superantigens stimulate SJL/J-derived mouse T cell clones specific to MBP and activate their pathogenicity to mediate EAE in normal syngeneic recipients (Soffer, D. and Ben-Nun, A . , submitted for publication). Subsequently we wished to investigate the effects of SEB on EAE in Lewis rats and on their h4BP-specificTcell clones which mediate the disease. However, we found (Fig. 1A) that SEB was not stimulatory for the Lewis rat MBP-specificTcel1clones Zla and D9 Tcell clones functional in mediating EAE [15, 181 or for the A2b T cell clone specific to a mycobacterial antigen which is cross-reactive with rat cartilage and involved in A A [16]. SEB was also nonstimulatory for nonautoimmune Tcells, the LOA and LPD clones specific to OVA and PPD, respectively (Fig. 1A). In contrast to its failure in stimulating the rat T lymphocytes, SEB was highly
To further investigate the effect of SEB on rat T lymphocytes, we activated Lewis rat thymocytes or splenocytes in vitro, in the presence of SEB, using Con A. As shown in Fig. 2A, the presence of SEB imposed a powerful, dosedependent suppression on the proliferative responses by Lewis rat thymocytes or splenocytes to Con A. The presence of SEB (from the same batch) in the in vifroculture of Con A activation of SJL/J thymocytes or splenocytes did not suppress their activation by Con A. On the contrary, due to its stimulatory effect on murine T cells, the SEB further supported their activation by Con A (Fig. 2B). The proliferative responses of thymocytes and splenocytes of other rat strains were also affected by SEB. However, as shown in Fig. 2C and D, the thymocytes and the splenocytes from the different rat strains displayed different levels of sensitivities to the suppression effect of SEB. Thus, although BN thymcoyte stimulation by Con A was strongly
100-
-
La-
0
(0-
m-
p!
7
x
,
20,
[SEB] lg/ml
Figure 1. SEB is non-stimulatory to Lewis rat Tcells. The proliferative responses to SEB of (A) Lewis rat CD4+ Tcell clones specific for MBP (Zla, D9), M. tuberculosis (A2b), OVA (LOA) and PPD (LPD): (B) SJL/J CD4+ Tcell clones specific for MBP (SR-1). OVA (SOD) and PPD (SPD); (C) Lewis rat thymocytes and splenocytes; ( D ) SJLN thymocytes and splenocytes.
[SEB] pg/ml
Figure 2. Suppression of the mitogenic stimulation of rat Tcells by SEB. The thymocytes (l@/well), splenocytes (5 x 10Vwell) from Lewis rat (A), SJL/J mice (B) or different strains of rats (C and D) were activated with Con A (1 pglml) in the presence of indicated concentrations of SEB. Proliferative responses were assessed by [3H]dThd uptake as mentioned in Sect. 2.4.
Enterotoxin B as a potent suppressant of Tcells
Eur. J. Immunol. 1991.21: 815-818
817
The results (Fig. 2D) showing that SEB was suppressive for Lewis, Fisher and Wistar, but not BN, spleen cells indicated that the suppression by SEB could not be attributed to nonspecific cytotoxic activity. To gain more information concerning the nature of the suppression effect by SEB, Lewis rat Tcell clones were activated with Con A in the presence of SEB and irradiated syngeneic or allogeneic thymocytes, or of xenogeneic mouse spleen cells, as 00 01 .I 1 10 tm accessory cells expressing MHC class I1 molecules. SEB was highly effective in suppressing the activation of Lewis rat T cell clones in the presence of syngeneic (Fig. 4A) as well as allogeneic rat accessory cells (Fig. 4B). However, when xenogeneic mouse spleen cells were used as accessory cells, SEB failed to suppress the proliferative responses of the Zla rat T cell clone activated by Con A (Fig. 4C). Conversely, the SEB presented by the mouse accessory cells was highly stimulatory of proliferative responses by the Lewis rat T cell clones (Fig. 4C; similar results were [=31 Clg/ml obtained with D9 and A2b clones), eliminating the possiFigure 3. Suppression by SEB of the antigen-specific and mitogenic stimulation of Lewis rat Tcell clones. Lewis rat Tcell clones bility that the highly potent suppression by SEB could be Zla (A) and D9 (B) specific to MBP, the A2b clone (C) specific to attributed to a nonspecific cytotoxic effect on the rat Tcells. M. tuberculosis and LOA clone (D) specific to OVA were cultured In addition, SEB could not suppress the Con A stimulation (l@/well) in the presence of their specific antigen (10 &well) or of the BUR-1 MBP-specific T cell clone derived from B1O.PL mice in the presence of either mouse (Fig. 4C) or with Con A (1 &well) and in the presence of SEB. rat accessory cells (Fig. 4A and B). On the contrary, SEB presented by rat accessory cells stimulated, in the absence of Con A, powerful proliferative responses by the murine suppressed, thymocytes from Wistar and Fisher rats were MBP-specific T cell clones to comparable levels as in the more sensitive to suppression by SEB. Splenocytes were presence of their respective syngeneic accessory cells less sensitive than thymocytes, and BN or Wistar spleno- (Fig. 4A-C). cytes were far less sensitive to the suppression by SEB. ~~
SEB had a powerful suppression activity also on the in vim-maintained, antigen-specific rat Tcell clones.The Zla (Fig. 3A), D9 (Fig. 3B) and A2b (Fig. 3C) Tcell clones and the nonautoimmuneTcell clone specific to OVA (Fig. 3D) were strongly suppressed when activated by their respective specific antigen or by Con A in the presence of SEB. The suppression of theTcell clones, as well as of the thymocytes and splenocytes derived from Lewis rats, was extremely potent. SEB (5 x lop2pg/ml) was sufficient to cause 50% suppression of the T cell proliferative response stimulated by the specific antigen or by aTcell mitogen (Figs. 2A and C). In contrast, the optimal stimulation of mouse T cells required 10-20 pg/ml of SEB.Thus, the suppression of rat T cell activation by SEB was at least 10 times more potent than its stimulatory activity on mouse Tcells.
*-
(A) 100-
x
+ 21. + ZlaOnA
loo -
+ BURl
*
.
4 Discussion The results presented above demonstrate that not only was SEB a poor stimulant of rat T lymphocyte proliferative responses (Fig. l ) , but it was also a highly potent suppressor of rat T lymphocytes' proliferative responses stimulated by a Tcell mitogen or by the specific antigen (Figs. 2-4). The failure of SEB to stimulate rat T lymphocyte proliferative responses could not be explained by the inability of SEB to bind to rat MHC molecules, nor by a major deletion of T lymphocyte subsets bearing the appropriate Vg elements required for recognizing SEB. These conclusions are based on the results showing that SEB associated with rat MHC is a powerful stimulant of proliferative responses
21.
(B)
(c)
+ Zla+ConA + BUR1 &
BURl+ConA
21.
+ ZlarConA
100-
BURl
& BURl+ConA
BUR.l+CmA
20-
0-• 0.0
.01
.l
1
lo
loo
0.0
.01
.I
1
lo
loo
0.0
01
.I
1
lo
loo
[SEB] W d
Figure 4. The mitogenic activation of rat T lymphocytes is suppressedby SEB associated with syngeneic or allogeneic, but not xenogeneic APC.Tcell clones derived from Lewis rats (Zla) or B1O.PL mice (BUR-1) were stimulated with SEB or were activated with Con A in the presence or absence of SEB and in the presence of APC from Lewis rat (A), BN rat (B) or BlO.PL mouse (C).
818
A. Ben-Nun
by mouse Tcell clones (Fig. 3C), but not rat Tcell clones (Fig. l), whereas SEB associated with mouse MHC is a powerful stimulant of both mouse and rat T lymphocyte proliferative responses (Fig. 4).Thus, SEB can bind the rat MHC and rat T lymphocytes do recognize SEB. However, the SEB/rat MHC complex, in contrast to the SEB/mouse MHC complex, could not stimulate large proportions of rat T lymphocyte subsets. The failure of SEB presented by syngeneic or allogeneic rat APC to stimulate rat Vp8+ T lymphocytes (Figs. 3 and 4) and the powerful stimulation of rat Vg8+ T lymphocyte proliferative responses by SEB associated with mouse APC (Fig. 4) indicated that SEB may bind to a nonpolymorphic region of the rat class I1 MHC. which is not conserved between mice and rats.
Eur. J. Immunol. 1991.21: 815-818
all Vg8+ rat T cells studied (Fig. 3). The suppression of different Vg8+ rat T cells, irrespective of their antigenic specificity and the suppression of large proportions of rat T lymphocyte subsets, suggests that SEB may well be considered a “supersuppressogen”as well as a “superantigen”. Whether SEB will be a “superantigen”or a “supersuppressogen” is likely to depend on the TcR elements of the lymphocytes as well as on the MHC presenting the SEB. 1 am grateful to Ms. Sara Yossefi for her expert technical assistance and 1 thank Prof. T E Davies, 1. R. Cohen and G. Berke for critically reading and commenting on the manuscript, and Ms. Malvine Baer for preparing the manuscript and for editorial assistance.
The mechanism by which SEB imposes a strong suppression on rat T lymphocytes responding to their specific Received September 4, 1990. antigen or to aTcell mitogen is not clear.The possibility that SEB is nonspecifically cytotoxic for rat Tcells or accessory cells was excluded by showing that SEB associated with 5 References mouse MHC is stimulatory for rat T lymphocytes and SEB 1 Bergdoll, M. S., in Riemann, H. and Bryan, F. L. (Eds.), associated with rat MHC is stimulatory to mouse T cells. Foodborne Infections and Intoxications, Academic Press, New These experiments also ruled out the possibility that Lewis York 1979, p. 443. rat cells expressing SEBMHC confer Tcell anergy due to 2 Betley, M. J. and Mekalanos, J. J., Science 1985. 229: 185. their incapability to provide co-stimulatory signals [21]. 3 Mollick, J. A . , Cook, R. G . and Rich, R. R., Science 1989.244: MHC blockade [22], as another possible explanation for 817. the highly potent suppression, was also considered, SEB 4 Hermann, T., Accolla, R. S. and MacDonald, H. R . , Eur. J. Immunol. 1989. 18: 2171. was shown to bind directly to MHC class I1 molecules with a high binding affinity 1231. Therefore, SEB as a nonstimula- 5 Janeway, C. A . , Jr., Immunol. Rev. 1989. 56: 61. 6 White, J., Herman, A . , Pullen, A . M.. Kubo, R., Kappler, J.W. tory antigen for Lewis rat T lymphocytes may simply and Marrack, P., Cell 1989. 56: 27. occupy the MHC-binding sites and block the antigen7 Callahan. J.. Herman. A . . Kaoder. J. and Marrack. F?. J. specific stimulation of Lewis rat T lymphocytes. This Immunol. 1990. 144: 2473. possibility, however, was unlikely, for three reasons: (a) 8 Kappler, J., Kotzin, B., Herron, L., Gelfand, E. W., Bigler, R. SEB did not always suppress the stimulation of Tcells for D., Boylston, A . , Carrel, S., Posnett, D. N., Choi, Y. and which it was nonstimulatory. BN splenocytes were not Marrack, €?, Science 1989. 244: 811. suppressed by SEB, even though SEB was only a poor 9 Kappler, J. W., Staerz, U., White, J. and Marrack, P., Nature 1988. 332: 35. stimulator for these lymphocytes (Fig. 2D); (b) suppression by SEB was usually not dependent on the SEB/Ag 10 MacDonald, R. H., Schneider, R., Lees, R. K., Howe, R. C., Acha-Orbea, M., Festenstein, H., Zinkernagel, R. M. and ratios. Increasing antigen or Con A concentrations did not Hengartner, M., Nature 1988. 332: 40. result in a significantly lower suppression by SEB (data not 11 Happ, M. I?, Woodland, D. L. and Palmer, E., Proc. Nad. shown); and (c) the non-MHC-restricted suppressogenic Acad. Sci. USA 1989. 86: 6293. effect of SEB indicated that SEB may bind to a nonpoly- 12 Pullen, A. P., Marrack. P. and Kappler, J.W., Nature 1988.335: morphic region of the MHC, as previously suggested for its 796. stimulatory effect [23], rather than to the highly polymorp- 13 Abe, R.,Vacchio, M. S., Fox, B. and Hodes, R. J., Nature 1988. hic region of MHC where the MHC binds the nominal 335: 827. 14 Hirschfeld, M.,Teitelbaum, D., Arnon, R. and Sela, M., FEBS antigens. , I
Interestingly, the characteristics of the rat T lymphocyte suppression by SEB resembled in several respects the stimulatory characteristics of SEB on mouse T lymphocytes. The suppression was dependent upon the interaction of T lymphocytes with accessory cells presenting SEB and the specific antigen or Tcell mitogen. Suppression by SEB was also dependent on MHC and similar to the stimulation by SEB, the suppression was not MHC restricted. Furthermore, the ability of SEB to suppress the mitogen stimulation of Lewis rat thymocytes and splenocytes indicated that SEB is a powerful suppressor affecting large proportions of Lewis rat T lymphocyte subsets. Whether SEB suppresses rat T lymphocytes on the basis of their Vp gene usage is an intriguingquestion which requires further investigation.We demonstrated here that SEB presented by rat MHC suppresses the mitogenic or antigen-specific stimulation of
Lett. 1970. 7: 317. 15 Ben-Nun, A.,Wekerle, H. and Cohen, 1. R., Nature 1981.229:
60. 16 Van Eden,W., Holoshitz, Z., Nevo, A . , Frenkel, A . , Klajman, A. and Cohen, 1. R., Proc. Nail. Acad. Sci. USA 1985. 82: 5117. 17 Ben-Nun, A. and Soffer, D., Eur. J. lmmunol. 1990. 20: 195. 18 Ben-Nun, A. and Cohen, I. R . , J. lmmunol. 1982. I28: 1450. 19 Bruns, F. R., Li, X., Shen, N . , Offner, H . , Chou, Y. K., Vandenbark, A . A . and Heber-Katz, E., J. Exp. Med. 1989. 169: 27. 20 Chluba, J., Steeg, C., Becker, A.,Wekerle, H. and Epplen, J.T.. Eur. J. lmmunol. 1989. 19: 279. 21 Schwartz, R. H., Science 1990. 218: 1349. 22 Guillet, J. G . ,Lai, M. Z., Briner,T. J., Smith, J. A. and Gefter, M. L., Nature 1986. 324: 260. 23 Marrack, P. and Kappler, J., Science 1990. 248: 705.
